repositories / deb_xorg-server.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
Add patch that contain Mali fixes.
[deb_xorg-server.git] / ChangeLog
CommitLineData
a09e091a
JB		 1commit 9182af8a09c326bb4b01d3645fb2936e93f02242
2Author: Matt Dew <marcoz@osource.org>
3Date: Sun Apr 13 20:55:40 2014 -0600
4
5 Bump version to 1.15.1
6
7commit eec04d76a39a7334de4e00ef9f0f6e44c92b3d91
8Author: Eric Anholt <eric@anholt.net>
9Date: Thu Mar 6 18:50:07 2014 -0800
10
11 glx: Clear new FBConfig attributes to 0 by default.
12
13 The visualSelectGroup wasn't getting set (since our DRI drivers don't
14 use it), and and since it's the top priority in the sort order, you
15 got random sorting of your visuals unless malloc really returned you
16 new memory. This manifested as Xephyr -glamor rendering to a
17 multisampled window on my system, which as you might guess was
18 slightly lower performance than expected.
19
20 Signed-off-by: Eric Anholt <eric@anholt.net>
21 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
22
23commit d6268c25a8d33241d817977a84f127f8ef0cb9ee
24Author: Julien Cristau <jcristau@debian.org>
25Date: Wed Mar 26 23:24:20 2014 +0100
26
27 Xephyr: restore initial window resize lost in xcb conversion
28
29 The XResizeWindow call wasn't replaced by the xcb equivalent, so we
30 were no longer setting the initial window size, only wm size hints.
31
32 Regression from commit a2b73da "Xephyr: start converting hostx.c over to
33 xcb"
34
35 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=74849
36
37 Signed-off-by: Julien Cristau <jcristau@debian.org>
38 Reported-by: Laércio de Sousa <lbsousajr@gmail.com>
39 Tested-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
41 Signed-off-by: Keith Packard <keithp@keithp.com>
42
43commit 300458fb8ad0a7957e941cd65f47d204c7886e22
44Author: Matt Dew <marcoz@osource.org>
45Date: Fri Mar 21 23:49:41 2014 -0600
46
47 Bump bersion to 1.15.0.901
48
49commit c971864565eb114e4b34f17b0d2244e8e6e1e005
50Author: Peter Hutterer <peter.hutterer@who-t.net>
51Date: Wed Feb 26 07:54:56 2014 +1000
52
53 config: search for PnPID on all parents (#75513)
54
55 The PnPID for a device may not be on the immediate parent, so search up the
56 device tree until we find one.
57
58 X.Org Bug 75513 <http://bugs.freedesktop.org/show_bug.cgi?id=75513>
59
60 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61 Tested-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
62 (cherry picked from commit 795066477ee81b5b82e490eac8bed6b656d01f17)
63
64commit 2cbbe80efc485d2d05b06efe1e2e01f81f3e06f6
65Author: Peter Hutterer <peter.hutterer@who-t.net>
66Date: Thu Feb 20 13:18:05 2014 +1000
67
68 Xi: check for invalid modifiers for XI2 passive grabs
69
70 The other values are checked correctly, but if a modifier was outside the
71 allowed range, it would go unnoticed and cause a out-of-bounds read error for
72 any mask equal or larger than 256. The DetailRec where we store the grab masks
73 is only sized to 8 * sizeof(Mask).
74
75 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
77 (cherry picked from commit 0f10cfd4b903d4db293ec47c8a9a0d8b33965803)
78
79commit b3656c0b52d57333c64f932676a237d0f3a97f36
80Author: Mark Kettenis <kettenis@openbsd.org>
81Date: Sun Dec 15 14:31:10 2013 +0100
82
83 sync: Avoid ridiculously long timeouts
84
85 On OpenBSD, passing a timeout longer than 100000000 seconds to select(2) will
86 make it fail with EINVAL. As this is original 4.4BSD behaviour it is not
87 inconceivable that other systems suffer from the same problem. And Linux,
88 though not suffering from any 4.4BSD heritage, briefly did something similar:
89
90 <https://lkml.org/lkml/2012/8/31/263>
91
92 So avoid calling AdjustWaitForDelay() instead of setting the timeout to
93 (effectively) ULONG_MAX milliseconds.
94
95 Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
96 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
97 (cherry picked from commit ddeca927498c9df3b5e62c7bf05e31e2a3aeaa52)
98
99commit b332cd20ee14049606e3656490d13a8efa6b23ee
100Merge: f41ab8c 5e0432f
101Author: Matt Dew <marcoz@osource.org>
102Date: Thu Mar 6 22:03:03 2014 -0700
103
104 Merge branch 'server-1.15-branch' of git://people.freedesktop.org/~jeremyhu/xserver into server-1.15-branch
105
106commit f41ab8c60780ea8f87354e536e5b73cb23878eb7
107Author: Peter Hutterer <peter.hutterer@who-t.net>
108Date: Fri Jan 24 18:32:54 2014 +1000
109
110 dix: prevent a driver from initializing or submitting buttons > MAX_BUTTONS
111
112 The server internally relies on arrays with a MAX_BUTTONS maximum size (which
113 is the max the core protocol can transport). Make sure a driver adheres to
114 that.
115
116 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
117 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
118 (cherry picked from commit 87ca80a7196949597113225405f3e4ee03bbee13)
119
120commit 4447d71b9a74afe91aaf4cc01eae12a44ef09306
121Author: Alan Coopersmith <alan.coopersmith@oracle.com>
122Date: Tue Jan 28 20:27:51 2014 -0800
123
124 xf86DeleteScreen: move check for NULL pScrn before first dereference
125
126 Flagged by cppcheck 1.62:
127 [hw/xfree86/common/xf86Helper.c:220] -> [hw/xfree86/common/xf86Helper.c:231]:
128 (warning) Possible null pointer dereference: pScrn - otherwise it is
129 redundant to check it against null.
130
131 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
132 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
133 Signed-off-by: Keith Packard <keithp@keithp.com>
134 (cherry picked from commit c1ac89c793614797e08d3d8e7fc9ba55be899130)
135
136commit 2ac840a14958fe74170518ee2c3a6b2dd88b20bd
137Author: Alan Coopersmith <alan.coopersmith@oracle.com>
138Date: Tue Jan 28 20:27:50 2014 -0800
139
140 On realloc failure, free font_path_string instead of leaking it
141
142 Flagged by cppcheck 1.62:
143 [dix/dixfonts.c:1792]: (error) Common realloc mistake:
144 'font_path_string' nulled but not freed upon failure
145
146 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
147 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
148 Signed-off-by: Keith Packard <keithp@keithp.com>
149 (cherry picked from commit e6733ae91b7be52930f22a87de15fa05819ef948)
150
151commit 157cc02fc13c998bba70e1652907972015e15e8e
152Author: Alan Coopersmith <alan.coopersmith@oracle.com>
153Date: Fri Jan 24 23:42:49 2014 -0800
154
155 Check for calloc() failure in add_master()
156
157 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
158 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
159 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
160 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
161 (cherry picked from commit 9fc19168e7ca6308275bf8769d1ccb982f88465b)
162
163commit f11c5938d732af717aeebbbf3b356138f3411bb7
164Author: Peter Hutterer <peter.hutterer@who-t.net>
165Date: Fri Jan 24 16:51:02 2014 +1000
166
167 Xi: fix modifier offset in XIPassiveGrab swapping function
168
169 The request is followed by mask_len 4-byte units, then followed by the actual
170 modifiers.
171
172 Also fix up the swapping test, which had the same issue.
173
174 Reported-by: Alan Coopersmith <alan.coopersmith@oracle.com>
175 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
176 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
177 (cherry picked from commit 76b3be75b62657e346731444736f7e4d200beb5b)
178
179 Conflicts:
180 test/xi2/protocol-xipassivegrabdevice.c
181
182commit 345b7ead1dd262020e10b4aeb71044d46d16e134
183Author: Peter Hutterer <peter.hutterer@who-t.net>
184Date: Mon Jan 13 17:00:23 2014 +1000
185
186 os: restrict display names to digits
187
188 We call atoi() on the server's display to get the socket but otherwise use the
189 unmodified display for log file name, xkb paths, etc. This results in
190 Xorg :banana being the equivalent of Xorg :0, except for the log files being
191 in /var/log/Xorg.banana.log. I'm not sure there's a good use-case for this
192 behaviour.
193
194 Check the display for something that looks reasonable, i.e. digits only, but
195 do allow for :0.0 (i.e. digits, followed by a period, followed by one or two
196 digits).
197
198 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
199 Reviewed-by: Keith Packard <keithp@keithp.com>
200 (cherry picked from commit 71baa466b1f6b02fe503f9a3089b7b9d61aa0f80)
201
202commit 940d6a8e9d3544696654d5e68c615ab887f81945
203Author: Carlos Garnacho <carlosg@gnome.org>
204Date: Thu Jan 2 21:33:30 2014 +0100
205
206 Xi: Ensure DeviceChanged is emitted after grabs are deactivated
207
208 When a grab on a slave device is deactivated, the master device must
209 be checked, just in case there were events from other devices while
210 the slave device was stolen away by the passive grab. This may
211 introduce misbehaviors on mismatching valuators and device features
212 later on UpdateDeviceState().
213
214 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
215 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
216 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
217 (cherry picked from commit b2d5ee2e3684951b611fd2068d57cc65fd8305a3)
218
219commit 47da6e3f47a55aeeef2c849067dd2d09fc2fe481
220Author: Peter Hutterer <peter.hutterer@who-t.net>
221Date: Fri Jan 24 18:16:54 2014 +1000
222
223 dix: fix button state check before changing a button mapping
224
225 dev->button->down is a bitmask, not a normal array. Use the helper function to
226 check, we technically allow the mapping to change after the physical button
227 has been pressed (but not yet processed yet), so only check BUTTON_PROCESSED.
228
229 From XSetPointerMapping(3):
230 "If any of the buttons to be altered are logically in the down state,
231 XSetPointerMapping returns MappingBusy, and the mapping is not changed."
232
233 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
234 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
235 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
236 (cherry picked from commit 25d10464f440b8b34594b7c988a99a830ea39a29)
237
238commit 5e0432f797d58fe1a69ef538694f65bbba38737f
239Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
240Date: Wed Jan 1 11:10:41 2014 -0800
241
242 XQuartz: Avoid passing uninitialized pointers to X11ApplicationSetWindowMenu in AppleWMSetWindowMenu
243
244 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
245 (cherry picked from commit 77df653ae3d8448be21221711851acde12c6bc1a)
246
247commit 859ced56d9ee168796c2fcf7fa6056e0c1491f3a
248Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
249Date: Wed Jan 1 11:04:07 2014 -0800
250
251 XQuartz: Check for allocated memory before using it in AppleWMSetWindowMenu
252
253 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
254 (cherry picked from commit 3bc608a361a01043b226fb9aaebf88f6fd852925)
255
256commit 9444955d16f85926c75ee8eafb0ca796dcf16439
257Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
258Date: Wed Jan 1 11:00:40 2014 -0800
259
260 XQuartz: Silence a clang static analysis warning about a memory leak
261
262 It seems the alanyzer can't comprehend dixSetPrivate().
263
264 quartz.c:119:12: warning: Potential leak of memory pointed to by 'displayInfo'
265 return quartzProcs->AddScreen(index, pScreen);
266 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
267
268 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
269 (cherry picked from commit 64327226ddfba8f0653615cd678d2d4336fb993d)
270
271commit 64b0669d1c062e7e11fb0e3b8ea7fe56113fc86f
272Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
273Date: Wed Jan 1 10:55:10 2014 -0800
274
275 XQuartz: Silence a clang static analysis warning about a possible memory leak on exit
276
277 stub.c:356:9: warning: Potential leak of memory pointed to by 'newargv'
278 asl_log(aslc, NULL, ASL_LEVEL_ERR,
279 ^~~~~~~
280 stub.c:356:9: warning: Potential leak of memory pointed to by 'newenvp'
281 asl_log(aslc, NULL, ASL_LEVEL_ERR,
282 ^~~~~~~
283 2 warnings generated.
284
285 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
286 (cherry picked from commit b2f6b3497c33a4897afae80a2cf69c596b9f81e8)
287
288commit 472e8e5d1c26573cef4bd42451230ef47791ed6f
289Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
290Date: Wed Jan 1 10:47:52 2014 -0800
291
292 XQuartz: Validate length in appledri before swapping
293
294 Avoids potential memory corruption from bad requests
295
296 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
297 (cherry picked from commit a03f096a85537d9e881cedaa6cb71aca43a97086)
298
299commit 782fe5c174def75a9e4cb188c00c2fec72821472
300Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
301Date: Wed Jan 1 10:39:56 2014 -0800
302
303 XQuartz: Validate screen in AppleDRIQueryDirectRenderingCapable requests
304
305 Return an error to the caller rather than crashing the server on
306 invalid screens.
307
308 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
309 (cherry picked from commit b3572c0d1ab7888ac26d6b2b8be6d1d19ed9af3f)
310
311commit 045122566c0532378b50c1af3ffec3254e416fe2
312Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
313Date: Sun Dec 29 12:45:23 2013 -0800
314
315 XQuartz: Simplify hook_run to quiet static analyzer
316
317 x-hook.c:96:9: warning: Called function pointer is an uninitalized pointer value
318 (*fun[i])(arg, data[i]);
319 ^~~~~~~~~~~~~~~~~~~~~~~
320 1 warning generated.
321
322 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
323 (cherry picked from commit 959e8f23af7850fcaf40d6c67f5228241a36a9ab)
324
325commit 33b2ae0f3b4a80fd962d876f7437d98fcfc27791
326Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
327Date: Sun Dec 29 12:41:18 2013 -0800
328
329 XQuartz: Mark applicationWillTerminate: noreturn
330
331 X11Controller.m:938:1: warning: method 'applicationWillTerminate:' could be declared with attribute 'noreturn'
332 [-Wmissing-noreturn,Semantic Issue]
333 {
334 ^
335
336 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
337 (cherry picked from commit f79af1941776fd6f1ec26c50603fcc35ca7d514b)
338
339commit 404f26866147be1c26585e59d9795e87e0c3e5fa
340Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
341Date: Sun Dec 29 12:36:51 2013 -0800
342
343 XQuartz: Fix get_proc_address signature
344
345 indirect.c:675:28: warning: incompatible pointer types passing 'glx_gpa_proc (*)(const char *)' to parameter of type
346 'glx_gpa_proc' (aka 'glx_func_ptr (*)(const char *)') [-Wincompatible-pointer-types,Semantic Issue]
347 __glXsetGetProcAddress(&get_proc_address);
348 ^~~~~~~~~~~~~~~~~
349 ../../../glx/glxserver.h:122:42: note: passing argument to parameter 'get_proc_address' here [Semantic Issue]
350 void __glXsetGetProcAddress(glx_gpa_proc get_proc_address);
351 ^
352
353 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
354 (cherry picked from commit ea80279e292e59a9fe9651489f03e9f2f39810d9)
355
356commit 7803c4075b824ece79af0425cee736025c41d901
357Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
358Date: Sun Dec 29 12:31:23 2013 -0800
359
360 XQuartz: Fix darwinfb.h header guard
361
362 ./darwinfb.h:28:9: warning: '_DARWIN_FB_H' is used as a header guard here, followed by #define of a different macro
363 [-Wheader-guard,Lexical or Preprocessor Issue]
364 ^~~~~~~~~~~~
365 ./darwinfb.h:29:9: note: '_DARWIN_DB_H' is defined here; did you mean '_DARWIN_FB_H'? [Lexical or Preprocessor Issue]
366 ^~~~~~~~~~~~
367 _DARWIN_FB_H
368
369 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
370 (cherry picked from commit 2e3ebec9520719a8e5c3c92390e83bcb5216f978)
371
372commit 6c055f0d78349ff08a570c5dcc92c6f7b6924993
373Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
374Date: Sun Dec 29 12:22:55 2013 -0800
375
376 XQuartz: Silence some static analyzer warnings by annotating referencing counts
377
378 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
379 (cherry picked from commit 9da6c0918f40359f28fe8889d5b7cae7efcc8377)
380
381commit 2ea973e12f5d954211e1d10085a4c74581b43aca
382Author: Keith Packard <keithp@keithp.com>
383Date: Fri Dec 27 09:50:55 2013 -0800
384
385 Bump version to 1.15.0
386
387 Signed-off-by: Keith Packard <keithp@keithp.com>
388
389commit 2eb9915eea358f941702d3dad7434197991885c5
390Author: Keith Packard <keithp@keithp.com>
391Date: Thu Dec 12 15:48:08 2013 -0800
392
393 present: Set complete notify mode to Skip as needed
394
395 Skipped present pixmap calls were not setting the mode to
396 PresentCompleteModeSkip for skipped operations.
397
398 Signed-off-by: Keith Packard <keithp@keithp.com>
399
400commit e487babd525ef1bd016ec7b283fa08cf9e6c6f4f
401Author: Keith Packard <keithp@keithp.com>
402Date: Thu Dec 12 14:52:35 2013 -0800
403
404 present: Don't abandon presents which are already queued for flip
405
406 Presents which are not marked 'queued' and are in the window present
407 list are waiting for the flip event; discarding those won't work very
408 well (it'll end up trashing displayed content for the next frame), so
409 skip over those when looking for duplicate frame presents
410
411 Signed-off-by: Keith Packard <keithp@keithp.com>
412
413commit ca3a5b2a8f9f627bbaa9883e16512a21c2a0b81d
414Author: Keith Packard <keithp@keithp.com>
415Date: Thu Dec 12 14:17:40 2013 -0800
416
417 present: Handle PresentOptionAsync for copy mode
418
419 Check for Async flag and execute immediately if set, otherwise wait
420 for the next appropriate vblank before copying.
421
422 Signed-off-by: Keith Packard <keithp@keithp.com>
423
424commit a68df147421da21528b5be2d34678383922fa352
425Author: Keith Packard <keithp@keithp.com>
426Date: Thu Dec 19 14:31:07 2013 -0800
427
428 Bump version to 1.14.99.905 (1.15 RC5)
429
430 Another week, another RC. This should be the last before 1.15 final
431
432 Signed-off-by: Keith Packard <keithp@keithp.com>
433
434commit 4b1ead9d3400acc3402c2480d7cc0527750c32f0
435Merge: 4d62646 929795d
436Author: Keith Packard <keithp@keithp.com>
437Date: Thu Dec 19 14:14:59 2013 -0800
438
439 Merge remote-tracking branch 'whot/for-keith'
440
441commit 4d62646142718024b0981eb4f1fd0131e829161f
442Merge: f4bfb14 81ba89d
443Author: Keith Packard <keithp@keithp.com>
444Date: Mon Dec 16 09:27:57 2013 -0800
445
446 Merge remote-tracking branch 'jeremyhu/master'
447
448commit f4bfb14f53a939574da1f5995f0dad949898b86a
449Author: Adam Jackson <ajax@redhat.com>
450Date: Thu Dec 12 10:57:40 2013 -0500
451
452 configure: Fix a typo near dri3/shmfence detection
453
454 " is not the mate of ].
455
456 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
457 Signed-off-by: Adam Jackson <ajax@redhat.com>
458
459commit 8248b4af31ef71ac03158871b77f20eb456dbe38
460Author: Adam Jackson <ajax@redhat.com>
461Date: Fri Dec 13 15:07:48 2013 -0500
462
463 glx: Add null pointer protection to __glGetProcAddress
464
465 This can't happen when GLX is the backing window system, but can
466 elsewhere. We may as well protect against it at a high level.
467
468 Reviewed-by: Eric Anholt <eric@anholt.net>
469 Signed-off-by: Adam Jackson <ajax@redhat.com>
470
471commit 128449dd6498a2f74c3770f89a9dae0f70e2b351
472Author: Adam Jackson <ajax@redhat.com>
473Date: Fri Dec 13 11:39:16 2013 -0500
474
475 present: Don't use the major/minor version from the protocol headers
476
477 We want to advertise the version we implement, not the version the
478 protocol headers happen to describe.
479
480 Reviewed-by: Jasper St. Pierre <<jstpierre@mecheye.net>
481 Signed-off-by: Adam Jackson <ajax@redhat.com>
482
483commit ec6087bf070131bc3a26848fb9fb840f09b0e19d
484Author: Adam Jackson <ajax@redhat.com>
485Date: Fri Dec 13 11:36:38 2013 -0500
486
487 dri3: Don't use the major/minor version from the protocol headers
488
489 We want to advertise the version we implement, not the version the
490 protocol headers happen to describe.
491
492 Reviewed-by: Jasper St. Pierre <<jstpierre@mecheye.net>
493 Signed-off-by: Adam Jackson <ajax@redhat.com>
494
495commit e6fafd3de70d315130fb6e5732cfd02a5901b788
496Author: Adam Jackson <ajax@redhat.com>
497Date: Fri Dec 13 11:28:59 2013 -0500
498
499 dri3: Guard against buggy clients
500
501 There's nothing to stop a client from sending these requests to screens
502 without DRI3 support, and if they do, we'll crash. Let's not do that.
503
504 Reviewed-by: Jasper St. Pierre <<jstpierre@mecheye.net>
505 Signed-off-by: Adam Jackson <ajax@redhat.com>
506
507commit 81ba89d6703a22178a153aa39478ba2d4bde262b
508Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
509Date: Fri Dec 13 01:39:44 2013 -0800
510
511 configure.ac: Add PRESENT_LIB to XQUARTZ_LIBS
512
513 Undefined symbols for architecture x86_64:
514 "_present_extension_init", referenced from:
515 _staticExtensions in libXquartz.a(miinitext.o)
516 "_present_register_complete_notify", referenced from:
517 ___glXregisterPresentCompleteNotify in libglx.a(glxcmds.o)
518
519 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
520
521commit d7c9235ee261b0f780320985233e00dec5e2689c
522Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
523Date: Sat Dec 7 01:36:33 2013 -0800
524
525 XQuartz: Use asl_log_descriptor to log stdout/stderr of child processes
526
527 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
528
529commit ad8111d7c971ce448905c733d65ba0cfc72bdca4
530Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
531Date: Sat Dec 7 01:14:37 2013 -0800
532
533 darwin: Don't leave stdin/stdout closed
534
535 <rdar://problem/15609419>
536
537 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
538
539commit 929795d50d788358d6269ce423f72c6cc40e334b
540Author: Peter Hutterer <peter.hutterer@who-t.net>
541Date: Tue Dec 3 10:14:51 2013 +1000
542
543 dix: fix check for grab type
544
545 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
546
547commit 23394c7fea0f5c33333198c87ecfecc9f6c6a791
548Author: Peter Hutterer <peter.hutterer@who-t.net>
549Date: Tue Dec 3 08:36:45 2013 +1000
550
551 Xi: ungrab device when releasing a passive grab without ButtonReleaseMask (#71878)
552
553 If an touch triggers an async button grab and that grab does not have the
554 ButtonReleaseMask set, the TouchEnd is never delivered, deliveries is 0 and
555 the grab is never deactivated.
556
557 If the grab is pointer async and keyboard sync, the keyboard events are stuck
558 in EnqueueEvent until some other pointer event terminates the grab.
559
560 Change this to check for the number of listeners. If we're about to deliver a
561 TouchEnd to a passive pointer grab, the number of listeners is already 1 -
562 pointer grabs always accept so other listeners were removed.
563
564 X.Org Bug 71878 <http://bugs.freedesktop.org/show_bug.cgi?id=71878>
565
566 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
567
568commit c1d30b5bd7f90e68bc38404fd0cc32578d6d3018
569Author: Ben Gamari <bgamari.foss@gmail.com>
570Date: Thu Nov 21 21:24:20 2013 -0500
571
572 Xi: Don't ActivateEarlyAccept POINTER_REGULAR listeners
573
574 Bug #71878 describes a bug resulting in the server ceasing to respond to
575 keyboard input after a touch event. The problem might be the following:
576
577 DeliverTouchBeginEvent tries to deliver an event to a listener of type
578 LISTENER_POINTER_REGULAR, taking the following if branch,
579
580 if (listener->type == LISTENER_POINTER_REGULAR ||
581 listener->type == LISTENER_POINTER_GRAB) {
582 rc = DeliverTouchEmulatedEvent(dev, ti, ev, listener, client, win,
583 grab, xi2mask);
584 if (rc == Success) {
585 listener->state = LISTENER_IS_OWNER;
586 /* async grabs cannot replay, so automatically accept this touch */
587 if (dev->deviceGrab.grab &&
588 dev->deviceGrab.fromPassiveGrab &&
589 dev->deviceGrab.grab->pointerMode == GrabModeAsync)
590 ActivateEarlyAccept(dev, ti);
591 }
592 goto out;
593 }
594
595 DeliverTouchEmulatedEvent succeeds. The deviceGrab meets all
596 three of the conditions of the inner if, enters
597 ActivateEarlyAccept which then fails due to,
598
599 BUG_RETURN(ti->listeners[0].type != LISTENER_GRAB &&
600 ti->listeners[0].type != LISTENER_POINTER_GRAB);
601
602 That is, despite listener->type == LISTENER_POINTER_REGULAR. With my
603 non-existent knowledge of XINPUT, it seems like the solution here
604 might be to only ActivateEarlyAccept when listener->type ==
605 LISTENER_POINTER_GRAB.
606
607 Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
608 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
609 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
610
611commit fe07ec19e212a68076560d243a2a935c54589068
612Author: Keith Packard <keithp@keithp.com>
613Date: Tue Dec 10 11:27:47 2013 -0800
614
615 present: recursively set window pixmaps on flip
616
617 Newly created windows inherit the pixmap of their parent, similarly,
618 reparenting a tree inherits the pixmap of the destination tree.
619
620 Making present preserve the invariant that unredirected windows always
621 have the same pixmap as their parent ensures that the above cases work
622 correctly.
623
624 v2: name the recursive function to 'set_tree_pixmap' instead of 'set_window_pixmap'
625
626 Signed-off-by: Keith Packard <keithp@keithp.com>
627 Reviewed-by: Adam Jackson <ajax@redhat.com>
628
629commit b3533d0b212b6747a8f9a01931253d6bdb648ee2
630Author: Keith Packard <keithp@keithp.com>
631Date: Tue Dec 10 10:29:11 2013 -0800
632
633 Bump version to 1.14.99.904 (1.15 RC4)
634
635 Signed-off-by: Keith Packard <keithp@keithp.com>
636
637commit 9a4d7572fe3be16bb5aded5f48d4217fdf725b1d
638Author: Jon TURNEY <jon.turney@dronecode.org.uk>
639Date: Thu Sep 27 18:59:45 2012 +0100
640
641 hw/xwin/glx: Rewrite WGL wrappers after Xserver conversion to direct GL dispatch
642
643 v1: Rewrite by Marc Haesen of the WGL wrapper function generation script to use
644 Khronos group XML.
645
646 v2: Remove -dispatchheader option, since dispatch.h doesn't exist anymore, use
647 the private glapi interface to construct the GL dispatch table for the native
648 WGL thunks.
649
650 v3:
651 Rewrite to generate shims for the OpenGL 1.2.1 (GL 1.2 + GL_ARB_imaging
652 +GL_ARB_multitexture + GL_ARB_texture_compression(?)) functions the server links
653 directly with rather than libGL.
654
655 These shims dispatch to either the mesa GL DLL, or a thunking DLL containing
656 cdecl-to-stcall wrapper functions for the native GL DLL.
657
658 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
659 Reviewed-by: Adam Jackson <ajax@redhat.com>
660
661commit cb48877a3c4cfb1ae2248000fcd3443e3092bdaa
662Author: Jon TURNEY <jon.turney@dronecode.org.uk>
663Date: Fri Dec 6 14:00:21 2013 +0000
664
665 glx: Consistently use ARB-suffixed names for ARB_multitexture functions
666
667 At the moment we have a mix of ARB and non-ARB suffixed forms for ARB_multitexture functions
668 e.g. glMultiTexCoord1fvARB and glMultiTexCoord1dv
669
670 Consistently use the ARB-suffixed form, assuming that is present in all libGL
671 which provide the OpenGL 1.2.1 ABI we expect to be able to directly link with.
672
673 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
674 Reviewed-by: Adam Jackson <ajax@redhat.com>
675
676commit 1a021f57a1e75ab63e8f87704f0394162402c4d4
677Author: Jon TURNEY <jon.turney@dronecode.org.uk>
678Date: Sun Dec 1 22:25:46 2013 +0000
679
680 ephyr: Fix compilation when ./configure'd with --enable-debug
681
682 /jhbuild/checkout/xorg/xserver/hw/kdrive/ephyr/ephyr.c: In function ‘ephyrProcessMouseMotion’:
683 /jhbuild/checkout/xorg/xserver/hw/kdrive/ephyr/ephyr.c:946:188: error: ‘ephyrCurScreen’ undeclared (first use in this function)
684 /jhbuild/checkout/xorg/xserver/hw/kdrive/ephyr/ephyr.c: In function ‘ephyrProcessButtonPress’:
685 /jhbuild/checkout/xorg/xserver/hw/kdrive/ephyr/ephyr.c:980:186: error: ‘ephyrCurScreen’ undeclared (first use in this function)
686 /jhbuild/checkout/xorg/xserver/hw/kdrive/ephyr/ephyr.c: In function ‘ephyrProcessButtonRelease’:
687 /jhbuild/checkout/xorg/xserver/hw/kdrive/ephyr/ephyr.c:1007:186: error: ‘ephyrCurScreen’ undeclared (first use in this function)
688
689 Fix ephyr compilation when ./configure'd with --enable-debug after commit
690 46cf6bf5692ef751ec9d17ae2292565d4b13f14b, some instances of ephyrCurScreen were
691 not converted to screen->pScreen->myNum.
692
693 v2: Don't use a trivial local variable which will be unused when ./configure'd
694 with --disable-debug
695
696 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
697 Reviewed-by: Adam Jackson <ajax@redhat.com>
698
699commit 875dbcef5bfcb6a94ef8456be6151e133408793f
700Author: Jon TURNEY <jon.turney@dronecode.org.uk>
701Date: Fri Nov 29 14:29:20 2013 +0000
702
703 configure.ac: Link XWin with present extension if we are building with it enabled
704
705 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
706 Reviewed-by: Adam Jackson <ajax@redhat.com>
707
708commit e61e19959d9138d5b81b1f25b7aa3e257918170d
709Author: Adam Jackson <ajax@redhat.com>
710Date: Tue Dec 3 13:45:43 2013 -0500
711
712 xquartz/glx: Convert to non-glapi dispatch
713
714 CGL doesn't have anything like glXGetProcAddress, and the old code just
715 called down to dlsym in any case. It's a little mind-warping since
716 dlopening a framework actually loads multiple dylibs, but that's just
717 how OSX rolls.
718
719 Signed-off-by: Adam Jackson <ajax@redhat.com>
720 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
721
722commit a668aa0e41bc33ff2db7f9c53b6dc321a96926a9
723Author: Adam Jackson <ajax@redhat.com>
724Date: Tue Dec 3 15:03:13 2013 -0500
725
726 drisw: Wire up GetProcAddress
727
728 Signed-off-by: Adam Jackson <ajax@redhat.com>
729 Reviewed-by: Keith Packard <keithp@keithp.com>
730
731commit 4fcdfeb7bc55d00d50d7f89788d9b929946ae3d9
732Author: Adam Jackson <ajax@redhat.com>
733Date: Tue Dec 3 15:02:57 2013 -0500
734
735 dri2: wire up GetProcAddress
736
737 Signed-off-by: Adam Jackson <ajax@redhat.com>
738 Reviewed-by: Keith Packard <keithp@keithp.com>
739
740commit 47f00b3920be9d8df8b148263d744934f3a02cd4
741Author: Adam Jackson <ajax@redhat.com>
742Date: Tue Dec 3 15:01:44 2013 -0500
743
744 glx: Untangle the prototypes around the GetProcAddress thunk
745
746 Signed-off-by: Adam Jackson <ajax@redhat.com>
747 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
748 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
749
750commit c1fd143f2847ef425ec9891d9e1cba44a1bb007d
751Author: Adam Jackson <ajax@redhat.com>
752Date: Tue Dec 3 13:11:52 2013 -0500
753
754 glx: Remove function stubs
755
756 Now that we're calling non-1.2 ABI things by function pointer this is no
757 longer needed.
758
759 Signed-off-by: Adam Jackson <ajax@redhat.com>
760 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
761 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
762
763commit 6b93e1f5ff39c6ac5b0d97d4245aeac9113eed8b
764Author: Adam Jackson <ajax@redhat.com>
765Date: Tue Dec 3 13:11:09 2013 -0500
766
767 glx: Convert non-generated function pointer thunking
768
769 Same concept as the generated code conversion.
770
771 Signed-off-by: Adam Jackson <ajax@redhat.com>
772 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
773
774commit c4567a376083eb7b142a7f003ddf8372d376ea86
775Author: Adam Jackson <ajax@redhat.com>
776Date: Tue Dec 3 12:40:24 2013 -0500
777
778 glx: Convert generated code function pointer thunking
779
780 We're meant not to call these by name due to ABI. Rather than try to
781 generate a bunch of little stub functions that do the lookup, just
782 inline it all directly into the calling function.
783
784 This does not cache results. That's fine, this is not a performance
785 path, and if we're atop WGL then we effectively have to do this every
786 time anyway because wglGetProcAddress results are context-dependent.
787
788 Signed-off-by: Adam Jackson <ajax@redhat.com>
789 Jon TURNEY <jon.turney@dronecode.org.uk>
790
791commit f69f4f417c6807405e45289b39a80f1b9a55f9e6
792Author: Keith Packard <keithp@keithp.com>
793Date: Tue Dec 10 07:09:17 2013 -0800
794
795 Depend on latest glproto (1.4.17)
796
797 Signed-off-by: Keith Packard <keithp@keithp.com>
798
799commit 576f3d36dd324bd0d16c9968f1ded5c5eead3569
800Author: Adam Jackson <ajax@redhat.com>
801Date: Mon Dec 9 13:16:01 2013 -0500
802
803 damageext: Die if we can't create the Xinerama resource type
804
805 Reviewed-by: Keith Packard <keithp@keithp.com>
806 Signed-off-by: Adam Jackson <ajax@redhat.com>
807
808commit f10f36d91db1b21c2ce5a531b4fa5f96fcbdc2f0
809Author: Adam Jackson <ajax@redhat.com>
810Date: Mon Sep 16 15:17:26 2013 -0400
811
812 damageext: Xineramify (v7)
813
814 v7: Don't bother making resources for the backing listeners. [keithp]
815
816 This is now slightly unlike how other resources are xineramified. We
817 create N+1 internal damage listeners, one that's a real resource and
818 reflects the protocol view, and then one per backend screen where the
819 report function piles onto the protocol view. The internal listeners
820 are not stored in the resource database directly, they just hang off the
821 xinerama resource. We don't wrap Subtract at the dispatch level, but we
822 do extend it for the Xinerama case to clip to the root window geometry.
823
824 As a result of the N+1 design here, the damage reports we generate are
825 not quite minimal. However they are indistinguishable from sequential
826 rendering events happening before the client hears damage, and we don't
827 need to add a post-dispatch callback just for this one extension.
828
829 Add is probably (still) somewhat broken since it will only hit screen 0,
830 but Add really only exists for DRI1's sake, and DRI1 disables itself
831 with Xinerama enabled anyway. In the absence of a use case, I'm leaving
832 it unwrapped under Xinerama; if someone wants to define how it ought to
833 work, be my guest.
834
835 Reviewed-by: Keith Packard <keithp@keithp.com>
836 Signed-off-by: Adam Jackson <ajax@redhat.com>
837
838commit 808303fe5232ba0320ae5a4310b1ed1322e85f1d
839Author: Adam Jackson <ajax@redhat.com>
840Date: Mon Nov 4 17:49:34 2013 -0500
841
842 fixes: Fix PanoramiXSetWindowShapeRegion for root windows (v2)
843
844 Root windows in Xinerama are in the coordinate space of their root window
845 pixmap, not in protocol space.
846
847 v2: Only translate for root windows, sice the window shape is
848 drawable-relative.
849
850 Reviewed-by: Keith Packard <keithp@keithp.com>
851 Signed-off-by: Adam Jackson <ajax@redhat.com>
852
853commit 5c10c7ea2129b70015e745523918d143cc29318d
854Author: Adam Jackson <ajax@redhat.com>
855Date: Mon Nov 4 17:49:33 2013 -0500
856
857 fixes: Fix PanoramiXSetPictureClipRegion for root windows (v2)
858
859 Root windows in Xinerama are in the coordinate space of their root
860 window pixmap, not in protocol space. This fixes 'xcompmgr -n' when
861 Xinerama is active.
862
863 v2: Only translate for root windows, since the clip origin is
864 drawable-relative.
865
866 Reviewed-by: Keith Packard <keithp@keithp.com>
867 Signed-off-by: Adam Jackson <ajax@redhat.com>
868
869commit 1dd839a425adc6e5a1dc377003ed86a374d81f0b
870Author: Adam Jackson <ajax@redhat.com>
871Date: Thu Nov 7 13:59:30 2013 -0500
872
873 composite: Fix COW creation for Xinerama (v2)
874
875 Say you have two 800x600 screens left/right of each other. A window
876 that's 200x200 at +700+0 in protocol coordinate space will appear to be
877 at -100+0 in the coordinate space of the right hand screen. Put another
878 way: windows are in the coordinate space of their root window pixmap.
879
880 We weren't doing this translation for the COW, so when rendering came in
881 to it you'd see the top-left chunk of the COW on all screens. Cool
882 effect and all, but wrong.
883
884 v2: Only translate when Xinerama is active [keithp]
885
886 Reviewed-by: Keith Packard <keithp@keithp.com>
887 Signed-off-by: Adam Jackson <ajax@redhat.com>
888
889commit 8dbe456abae1a32fb6c5c74fa456bc7301cccabe
890Author: Adam Jackson <ajax@redhat.com>
891Date: Tue Oct 29 10:44:07 2013 -0400
892
893 xinerama: Export the screen region
894
895 damageext wants this so it can intersect subtract requests against the
896 root window geometry.
897
898 Reviewed-by: Keith Packard <keithp@keithp.com>
899 Signed-off-by: Adam Jackson <ajax@redhat.com>
900
901commit c6d4c2a24140feee4ceae48c4c74fedcae9b68b3
902Author: Adam Jackson <ajax@redhat.com>
903Date: Fri Nov 15 15:13:31 2013 -0500
904
905 xfree86: Prefer fbdev to vesa
906
907 On UEFI machines you'd prefer fbdev to grab efifb instead of vesa trying
908 to initialize and failing in a way we can't unwind from. On BIOS
909 machines this is harmless: either there is an fbdev driver and it'll
910 probably be more capable, or there's not and vesa will kick in anyway.
911
912 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
913 Reviewed-by: Eric Anholt <eric@anholt.net>
914 Signed-off-by: Adam Jackson <ajax@redhat.com>
915
916commit 7b5d4f147fdef9edfeaa9c6565375111079efd11
917Author: Adam Jackson <ajax@redhat.com>
918Date: Tue Nov 5 10:12:28 2013 -0500
919
920 composite: Don't double-redirect if someone asks for backing store twice
921
922 v2: Belt-and-suspenders the unredirection path [vsyrjala]
923
924 Reviewed-by: Keith Packard <keithp@keithp.com>
925 Signed-off-by: Adam Jackson <ajax@redhat.com>
926
927commit a2b2c271e0ca87d3188ba2741b6db9bbbdc599f5
928Author: Adam Jackson <ajax@redhat.com>
929Date: Tue Nov 5 10:08:17 2013 -0500
930
931 composite: Automatically enable backing store support on the screen
932
933 ... unless you explicitly disabled it with -bs on the command line, or
934 with the corresponding thing in xorg.conf.
935
936 v2: Drop a bogus hunk from compChangeWindowAttributes [vsyrjala]
937 v3: s/TRUE/WhenMapped/ [jcristau]
938
939 Reviewed-by: Keith Packard <keithp@keithp.com>
940 Signed-off-by: Adam Jackson <ajax@redhat.com>
941
942commit e0cac005608a2e5618c7be59701318d684e0bb93
943Author: Adam Jackson <ajax@redhat.com>
944Date: Tue Nov 5 09:58:17 2013 -0500
945
946 bs: Set the screen's bs support level to WhenMapped
947
948 Since we're using RedirectAutomatic to do this, we don't actually
949 preserve contents when unmapped.
950
951 v2: Don't say WhenMapped if Composite didn't initialize [vsyrjala]
952
953 Reviewed-by: Julien Cristau <jcristau@debian.org>
954 Signed-off-by: Adam Jackson <ajax@redhat.com>
955
956commit b61ccd5d9d368f3fbbae27ce14ac02a3db1884c4
957Author: Adam Jackson <ajax@redhat.com>
958Date: Tue Nov 5 10:20:04 2013 -0500
959
960 smartsched: Tweak the default scheduler intervals
961
962 A default timeslice of 20ms means a pathological client can ruin up to
963 two frames per scheduler tick. And a fifth of a second is just insane.
964
965 Pick two different numbers out of the hat. A 5ms slice means you can
966 probably keep up with two or three abusive clients, and letting it burst
967 to 15ms should give you about all the timeslice you need for a
968 fullscreen game (that's doing server-side rendering for some reason).
969
970 If you're running on a system with a 10ms granularity on SIGALRM, then
971 this effectively changes the intervals to 10ms and 30ms. Which is still
972 better, just not as better.
973
974 I suspect this is about as good as we can do without actually going
975 preemptive, which is an entire other nightmare.
976
977 Reviewed-by: Julien Cristau <jcristau@debian.org>
978 Signed-off-by: Adam Jackson <ajax@redhat.com>
979
980commit 66310ea2893811614b3c3aade8992624a95a6ee7
981Author: Adam Jackson <ajax@redhat.com>
982Date: Fri Nov 8 09:22:37 2013 -0500
983
984 dri3: Disable when Xinerama is active
985
986 Pretty sure this can't work.
987
988 Reviewed-by: Julien Cristau <jcristau@debian.org>
989 Signed-off-by: Adam Jackson <ajax@redhat.com>
990
991commit 793fd5eefb0e417614d77fe1522c6548587fbd4e
992Author: Adam Jackson <ajax@redhat.com>
993Date: Fri Nov 8 09:20:29 2013 -0500
994
995 dri2: Disable when Xinerama is active
996
997 Would only work on ScreenRec 0, which means it's broken.
998
999 Reviewed-by: Julien Cristau <jcristau@debian.org>
1000 Signed-off-by: Adam Jackson <ajax@redhat.com>
1001
1002commit 2bf92108294805c5c6fef249c2f7de6ecf4975d8
1003Author: Adam Jackson <ajax@redhat.com>
1004Date: Fri Nov 8 09:06:59 2013 -0500
1005
1006 present: Disable when Xinerama is active
1007
1008 Among much else Present depends on RANDR types, and RANDR isn't properly
1009 Xinerama-aware yet anyway.
1010
1011 Reviewed-by: Julien Cristau <jcristau@debian.org>
1012 Signed-off-by: Adam Jackson <ajax@redhat.com>
1013
1014commit ac772cb187ddf7e04b8f4b3a071b90f18f4488d0
1015Author: Eric Anholt <eric@anholt.net>
1016Date: Thu Nov 14 17:40:47 2013 -0800
1017
1018 glx: Fix incorrect use of dri_interface.h version defines in driver probing.
1019
1020 If we extend __DRI_CORE or __DRI_SWRAST in dri_interface.h to allow a
1021 new version, it shouldn't make old server code retroactively require
1022 the new version from swrast drivers.
1023
1024 Notably, new Mesa defines __DRI_SWRAST version 4, but we still want to
1025 be able to probe version 1 drivers, since we don't use any features
1026 beyond version 1 of the struct.
1027
1028 Signed-off-by: Keith Packard <keithp@keithp.com>
1029 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
1030
1031commit 6e926b18ca1b182253bac435a1d53caaff7ffff6
1032Author: Eric Anholt <eric@anholt.net>
1033Date: Thu Nov 14 17:40:46 2013 -0800
1034
1035 glx: Fix incorrect use of dri_interface.h version defines in extensions.
1036
1037 Those defines are so you can compile-time check "do I have a
1038 dri_interface.h that defines this new field of the struct?" You don't
1039 want the server to claim it implements the new struct just because you
1040 installed a new copy of Mesa.
1041
1042 Signed-off-by: Keith Packard <keithp@keithp.com>
1043 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
1044
1045commit 57a8ce927332e855dd29ff30210a211a907adb25
1046Author: Jon TURNEY <jon.turney@dronecode.org.uk>
1047Date: Fri Nov 29 12:26:43 2013 +0000
1048
1049 configure.ac: Fixup for "Require libpciaccess for int10"
1050
1051 On 16/11/2013 01:00, Connor Behan wrote:
1052 > A --disable-pciaccess build will fail with an int10 module other than
1053 > stub.
1054 >
1055 > Signed-off-by: Connor Behan <connor.behan-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
1056 > ---
1057 > configure.ac | 3 +++
1058 > 1 file changed, 3 insertions(+)
1059 >
1060 > diff --git a/configure.ac b/configure.ac
1061 > index 5e621e0..a843770 100644
1062 > --- a/configure.ac
1063 > +++ b/configure.ac
1064 > @@ -1822,6 +1822,9 @@ if test "x$XORG" = xyes; then
1065 > if test "x$CONFIG_UDEV_KMS" = xyes; then
1066 > AC_MSG_ERROR([Platform device enumeration requires libpciaccess])
1067 > fi
1068 > + if test "x$INT10" != xstub; then
1069 > + AC_MSG_ERROR([Cannot build int10 without libpciaccess])
1070 > + fi
1071 > fi
1072 > AC_MSG_RESULT([$PCI])
1073 >
1074
1075 This causes my build to fail where --disable-int10-module --disable-pciaccess
1076 is the default (as INT10 still has the default value 'x86emu')
1077
1078 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
1079 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1080 Tested-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1081 Signed-off-by: Keith Packard <keithp@keithp.com>
1082
1083commit 653d33941b0808ef910aaa5f3aeab05d9c1a100b
1084Author: Keith Packard <keithp@keithp.com>
1085Date: Tue Nov 26 12:18:12 2013 -0800
1086
1087 present: Report damage when flipping
1088
1089 Limit damage to the 'update' region.
1090
1091 Signed-off-by: Keith Packard <keithp@keithp.com>
1092 Reviewed-by: Adam Jackson <ajax@redhat.com>
1093
1094commit 5cf12c9569ac3d83fe1b7a8376c15f8f0b01655e
1095Author: Keith Packard <keithp@keithp.com>
1096Date: Tue Nov 26 12:10:48 2013 -0800
1097
1098 present: Also set the root window pixmap when flipping
1099
1100 This makes sure that things like software cursors continue to work
1101 while the screen is flipped.
1102
1103 Signed-off-by: Keith Packard <keithp@keithp.com>
1104 Reviewed-by: Adam Jackson <ajax@redhat.com>
1105
1106commit 4aa77378de69efdc10bced6ba650b0ebff50c112
1107Author: Keith Packard <keithp@keithp.com>
1108Date: Tue Nov 26 12:06:57 2013 -0800
1109
1110 present: Clear target_crtc if driver lacks Present support
1111
1112 If the driver doesn't have the necessary hooks for Present, then the
1113 target_crtc needs to be set to NULL to make sure the extension uses
1114 the present_fake code.
1115
1116 Signed-off-by: Keith Packard <keithp@keithp.com>
1117 Tested-by: Fredrik Höglund <fredrik@kde.org>
1118 Reviewed-by: Adam Jackson <ajax@redhat.com>
1119
1120commit 3dd5bfe540b295bb37a2c2fd0ba4a31fb217612b
1121Author: Keith Packard <keithp@keithp.com>
1122Date: Thu Nov 21 22:48:31 2013 -0800
1123
1124 present: Send GLX_BufferSwapComplete events from present extension
1125
1126 This allows GL to support the GLX_INTEL_swap_event extension.
1127
1128 v2: Return GLX_BLIT_COMPLETE_INTEL for unknown swap types
1129
1130 Signed-off-by: Keith Packard <keithp@keithp.com>
1131 Reviewed-by: Adam Jackson <ajax@redhat.com>
1132
1133commit cde86e68fcb716f34c90f5a16eb868870f5c85e4
1134Author: Keith Packard <keithp@keithp.com>
1135Date: Mon Nov 25 23:06:08 2013 -0800
1136
1137 present: Set window pixmap to flipped pixmap
1138
1139 This makes other drawing to the window appear on the screen.
1140
1141 Note that no child windows can be affected because only full-screen
1142 windows are eligible for flipping, and so we only need to set pixmap
1143 for the window itself.
1144
1145 Signed-off-by: Keith Packard <keithp@keithp.com>
1146 Reviewed-by: Adam Jackson <ajax@redhat.com>
1147
1148commit 04e138846e128670d409798aa2e797c3c5503a47
1149Author: Keith Packard <keithp@keithp.com>
1150Date: Mon Nov 25 23:10:17 2013 -0800
1151
1152 present: Leave vblank on window list until flip complete
1153
1154 If the window is destroyed, then we've got cleanup work to do, even if
1155 the vblank has already been executed -- we need to clear the window
1156 pointer so that we don't try to deliver events to it.
1157
1158 Leaving it on the window list meant that when walking that list, we
1159 need to know whether the vblank is waiting to be executed or waiting
1160 for the flip to complete, so a new 'queued' flag was added to the
1161 vblank to distinguish between the two states.
1162
1163 Signed-off-by: Keith Packard <keithp@keithp.com>
1164 Reviewed-by: Adam Jackson <ajax@redhat.com>
1165
1166commit b121d62accb8c346b4e1b1bce99586e13712f04a
1167Author: Keith Packard <keithp@keithp.com>
1168Date: Tue Nov 26 03:07:55 2013 -0800
1169
1170 present: Add a debug output line when skipping a pending present
1171
1172 When an application provides two pixmaps for the same MSC, the
1173 previous one is skipped. This just dumps out some information at that point
1174
1175 Signed-off-by: Keith Packard <keithp@keithp.com>
1176 Reviewed-by: Adam Jackson <ajax@redhat.com>
1177
1178commit eafba23b34be31c141ddafb8380520ac9a0622ac
1179Author: Keith Packard <keithp@keithp.com>
1180Date: Thu Nov 21 22:45:18 2013 -0800
1181
1182 miext/sync: Handle libxshmfence API change
1183
1184 libxshmfence had an unfortunate 'int32_t' type for the mapped fence.
1185 That changed to exposing a 'struct shmfence' instead, which is nice
1186 and opaque and offers fine type checking across the API.
1187
1188 This patch requires the newer version of the library and uses
1189 the new interface type.
1190
1191 Signed-off-by: Keith Packard <keithp@keithp.com>
1192 Reviewed-by: Julien Cristau <jcristau@debian.org>
1193
1194commit b6d7ed4d787a652e8150532f384bfdf51760f3c2
1195Author: Keith Packard <keithp@keithp.com>
1196Date: Thu Nov 21 22:12:34 2013 -0800
1197
1198 miext: Move SyncShm FDs out of the way of clients
1199
1200 Applications may end up allocating a bunch of shmfence objects, each
1201 of which uses a file descriptor, which must be kept open lest some
1202 other client ask for a copy of it later on.
1203
1204 Lacking an API that can turn a memory mapping back into a file
1205 descriptor, about the best we can do is push the file descriptors out
1206 of the way of other X clients so that we don't run out of the ability
1207 to accept new connections.
1208
1209 This uses fcntl F_GETFD to push the FD up above MAXCLIENTS.
1210
1211 Signed-off-by: Keith Packard <keithp@keithp.com>
1212 Reviewed-by: Julien Cristau <jcristau@debian.org>
1213
1214commit cc63204926c6da83d9221c5f8c0dc8f5e2f2481d
1215Author: Keith Packard <keithp@keithp.com>
1216Date: Wed Nov 13 14:16:33 2013 +0900
1217
1218 Xext: Use SHMDIR and O_TMPFILE when creating mapping files
1219
1220 ShmCreateSegment asks for a file descriptor for a memory mapped file
1221 created by the X server. This patch uses O_TMPFILE where available,
1222 and also uses the SHMDIR directory to store the files, both for the
1223 O_TMPFILE and mkstemp cases.
1224
1225 Signed-off-by: Keith Packard <keithp@keithp.com>
1226 Reviewed-by: Julien Cristau <jcristau@debian.org>
1227
1228commit 5a969f0928b84da5cfe0777dfb542caaacc915ad
1229Author: Keith Packard <keithp@keithp.com>
1230Date: Wed Nov 13 12:17:10 2013 +0900
1231
1232 Select directory for MIT-SHM temp files at configure time
1233
1234 By default, this looks through a list of directories to find one which
1235 exists, but can be overridden with --with-shared-memory-dir=PATH
1236
1237 This patch doesn't actually do anything with this directory, just
1238 makes it available in the configuration
1239
1240 Signed-off-by: Keith Packard <keithp@keithp.com>
1241 Reviewed-by: Julien Cristau <jcristau@debian.org>
1242
1243commit d695cb7fb6bb8fd159a332d902673edbbd2237d7
1244Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1245Date: Mon Nov 25 22:19:34 2013 -0800
1246
1247 Xdmx: Initialize DMX extension even if not built with GLX support
1248
1249 dmxAddExtensions takes an argument to determine if it should just
1250 initialize the DMX extension, or both DMX & GLX, but if GLX wasn't
1251 compiled in, the entire call was #ifdef'ed out, leaving the DMX
1252 extension unavailable.
1253
1254 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1255 Reviewed-by: Julien Cristau <jcristau@debian.org>
1256 Signed-off-by: Keith Packard <keithp@keithp.com>
1257
1258commit bd70def07754357d26b86c287037eb3884631daf
1259Author: Connor Behan <connor.behan@gmail.com>
1260Date: Fri Nov 15 17:00:40 2013 -0800
1261
1262 configure.ac: Require libpciaccess for int10
1263
1264 A --disable-pciaccess build will fail with an int10 module other than
1265 stub.
1266
1267 Signed-off-by: Connor Behan <connor.behan@gmail.com>
1268 Reviewed-by: Adam Jackson <ajax@redhat.com>
1269 Signed-off-by: Keith Packard <keithp@keithp.com>
1270
1271commit a575c1dc64c10bc88bd0e963ebb8ab137f59a078
1272Author: Connor Behan <connor.behan@gmail.com>
1273Date: Fri Nov 15 17:00:20 2013 -0800
1274
1275 configure.ac: Require libpciaccess for platform bus support
1276
1277 There is currently no reason to build with --enable-config-udev-kms and
1278 --disable-pciaccess but anyone who tries this should know that the build
1279 will fail.
1280
1281 Signed-off-by: Connor Behan <connor.behan@gmail.com>
1282 Reviewed-by: Adam Jackson <ajax@redhat.com>
1283 Signed-off-by: Keith Packard <keithp@keithp.com>
1284
1285commit 361dfe74eaa2f1eddb54abdecca1d8b1b3f58704
1286Author: Connor Behan <connor.behan@gmail.com>
1287Date: Fri Nov 15 17:00:00 2013 -0800
1288
1289 configure.ac: Add whitespace near PCI configuration
1290
1291 Signed-off-by: Connor Behan <connor.behan@gmail.com>
1292 Reviewed-by: Adam Jackson <ajax@redhat.com>
1293 Signed-off-by: Keith Packard <keithp@keithp.com>
1294
1295commit 80481267662c8687e73081237913fa561e7a6561
1296Author: Keith Packard <keithp@keithp.com>
1297Date: Sat Nov 23 22:22:37 2013 -0800
1298
1299 Bump release to 1.14.99.903 (1.15 RC3)
1300
1301 Signed-off-by: Keith Packard <keithp@keithp.com>
1302
1303commit f1604002a32b7f098c2a16b4a8649c694af570c8
1304Author: Keith Packard <keithp@keithp.com>
1305Date: Mon Nov 18 22:36:17 2013 -0800
1306
1307 miext: Ensure xshmfence is only called when driver supports it
1308
1309 This provides a place for drivers to insert their own FD-based
1310 SyncFence implementations, and prevents applications from using DRI3
1311 SyncFence creation functions unless the driver has some support for
1312 them.
1313
1314 Signed-off-by: Keith Packard <keithp@keithp.com>
1315 Tested-by: Fredrik Höglund <fredrik@kde.org>
1316
1317commit 037566c57caff93fd7717f385d4b532b81f19c77
1318Author: Keith Packard <keithp@keithp.com>
1319Date: Mon Nov 18 22:33:27 2013 -0800
1320
1321 Xext: Recover from miSyncInitFenceFromFD failure without crashing
1322
1323 miSyncDestroyFence must not be called unless miSyncInitFence has been
1324 invoked, so if miSyncInitFenceFromFD fails, we must free the fence
1325 manually.
1326
1327 Signed-off-by: Keith Packard <keithp@keithp.com>
1328 Tested-by: Fredrik Höglund <fredrik@kde.org>
1329
1330commit e7000534a456fdf9cd4eaada3193846c8525f740
1331Author: Chris Wilson <chris@chris-wilson.co.uk>
1332Date: Sat Oct 5 08:47:03 2013 +0100
1333
1334 glx/glxdri2: Unwrap EnterVT/LeaveVT upon CloseScreen
1335
1336 In a similar spirit to
1337
1338 commit d75e8146c414bfd512ba5dbd4a83acb334bbe19b
1339 Author: Keith Packard <keithp@keithp.com>
1340 Date: Mon Jul 12 16:01:34 2010 -0700
1341
1342 Unwrap/rewrap EnterVT/LeaveVT completely, Fixes 28998
1343
1344 we need to unwrap our pScrn->EnterVT/LeaveVT hooks around server
1345 regeneration or else we cause an infinite recursion on the next VT
1346 switch afterwards.
1347
1348 Bugzilla: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/1235516
1349 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
1350 Reviewed-by: Keith Packard <keithp@keithp.com>
1351 Signed-off-by: Keith Packard <keithp@keithp.com>
1352
1353commit 6d5883bd7e5b765f8f0548501b825d9e56840799
1354Author: Keith Packard <keithp@keithp.com>
1355Date: Sat Nov 23 16:19:46 2013 -0800
1356
1357 xnest: Ignore GetImage() error in xnestGetImage() (v3)
1358
1359 (I found an amended version of this patch and applied the difference
1360 here - keithp)
1361
1362 v3: Don't call Xsync before restoring error handler as any errors
1363 generated by XGetImage() should be processed when this call
1364 returns as suggested by Jamey Sharp <jamey@minilop.net>
1365
1366 Signed-off-by: Egbert Eich <eich@freedesktop.org>
1367 Reviewed-by: Jamey Sharp <jamey@minilop.net>
1368
1369commit 6403cbb143c67872ca9c58e3116ae7942def0ae1
1370Author: Keith Packard <keithp@keithp.com>
1371Date: Tue Nov 19 22:13:54 2013 -0800
1372
1373 present: When unflipping, copy to flip window rather than screen
1374
1375 unflip happens after the clip lists have been updated, so instead of
1376 smashing the whole screen and drawing over other windows, just draw to
1377 the original flip window; it'll have the right clip list and so the
1378 copy will work just fine.
1379
1380 Signed-off-by: Keith Packard <keithp@keithp.com>
1381 Reviewed-by: Adam Jackson <ajax@redhat.com>
1382
1383commit 8bdd2ccc776ded3f527596b5009ef25129aa3287
1384Author: Keith Packard <keithp@keithp.com>
1385Date: Mon Nov 11 18:03:42 2013 -0800
1386
1387 present: Block for wait_fence in present_execute
1388
1389 Pend presentation until wait_fence is also triggered by having the
1390 SyncFence trigger invoke present_execute once triggered.
1391
1392 Signed-off-by: Keith Packard <keithp@keithp.com>
1393 Reviewed-by: Adam Jackson <ajax@redhat.com>
1394
1395commit e5a188cb919edee2e3a03054276bce0db02f7b62
1396Author: Keith Packard <keithp@keithp.com>
1397Date: Sat Nov 9 12:36:47 2013 -0800
1398
1399 present: Signal destroyed flip window with vblank->window == NULL
1400
1401 This eliminates dereferencing freed window pointers when there is a
1402 flip for that window in progress. The flip will complete, and then
1403 immediately get undone (as we can't stop an in-progress flip).
1404
1405 Remove the vblank->window_destroyed field as we can signal this with
1406 vblank->window == NULL instead.
1407
1408 Change check to vblank->window == NULL in:
1409
1410 present_flip_notify
1411
1412 Add check for vblank->window == NULL in:
1413 present_vblank_notify
1414 present_execute
1415
1416 present_flip_notify was also using vblank->window->drawable.pScreen,
1417 so stop doing that and use vblank->screen instead.
1418
1419 Signed-off-by: Keith Packard <keithp@keithp.com>
1420 Reviewed-by: Adam Jackson <ajax@redhat.com>
1421
1422commit a5bcc4f7b9499caf8993f0a6ef96088553399ca3
1423Author: Keith Packard <keithp@keithp.com>
1424Date: Sat Nov 9 12:33:02 2013 -0800
1425
1426 present: Ignore event_id 0 from driver vblank notify
1427
1428 We use event_id 0 to mean 'no such event'; if a driver sends us that
1429 event_id, make sure we don't accidentally match it.
1430
1431 Signed-off-by: Keith Packard <keithp@keithp.com>
1432 Reviewed-by: Adam Jackson <ajax@redhat.com>
1433
1434commit 4f3c37a1f17ffcfbbff71d217e1caad3d0148c90
1435Author: Keith Packard <keithp@keithp.com>
1436Date: Thu Nov 7 14:17:12 2013 -0800
1437
1438 present: Round fake MSC computations intead of truncating
1439
1440 If the timer fired too early, we'd sometimes mis-compute the MSC for
1441 fake vblanks. Rounding the computation to the nearest MSC fixes this nicely.
1442
1443 Signed-off-by: Keith Packard <keithp@keithp.com>
1444 Reviewed-by: Adam Jackson <ajax@redhat.com>
1445
1446commit da9997f89f14ab619f244d5b2e80a423b028c789
1447Author: Peter Hutterer <peter.hutterer@who-t.net>
1448Date: Tue Nov 19 08:07:09 2013 +1000
1449
1450 configure: allow for --enable-libunwind and --disable-libunwind
1451
1452 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1453 Reviewed-by: Keith Packard <keithp@keithp.com>
1454 Signed-off-by: Keith Packard <keithp@keithp.com>
1455
1456commit 29240e5cbf6e7f875b128fc7bfc4d56b2350835a
1457Merge: 0492deb 8ff7e32
1458Author: Keith Packard <keithp@keithp.com>
1459Date: Thu Nov 14 17:02:04 2013 +0900
1460
1461 Merge remote-tracking branch 'whot/for-keith'
1462
1463commit 0492deb8f8238b7782e5a706ec6219d88aa1091d
1464Author: Adam Jackson <ajax@redhat.com>
1465Date: Tue Oct 29 12:09:27 2013 -0400
1466
1467 mieq: Bump default queue size to 512
1468
1469 Based on some bugzilla scraping I did around November 2012. Of xserver
1470 bugs in Red Hat bugzilla with an EQ size message in the log, the
1471 distribution looked like:
1472
1473 String | Matches
1474 -------------------------------------
1475 Increasing EQ size to 512 | 460
1476 Increasing EQ size to 1024 | 52
1477 Increasing EQ size to 2048 | 6
1478 Increasing EQ size to 4096 | 0
1479
1480 Most of the "512" ones appear to be mostly harmless, some relatively
1481 expensive path in either rendering or resource destruction simply taking
1482 too long due to external pressures like paging or CPU contention. So
1483 let's raise the initial queue size, both to reduce the number of
1484 spurious abrt reports and to drop fewer events in all but the most
1485 pathological cases.
1486
1487 Signed-off-by: Adam Jackson <ajax@redhat.com>
1488 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
1489 Signed-off-by: Keith Packard <keithp@keithp.com>
1490
1491commit d1440783a7367ff0d0c47d256bbca3b3cf8a5034
1492Author: Dave Airlie <airlied@redhat.com>
1493Date: Tue Oct 29 12:09:26 2013 -0400
1494
1495 xfree86: return NULL for compat output if no outputs.
1496
1497 With outputless GPUs showing up we crash here if there are not outputs
1498 try and recover with a bit of grace.
1499
1500 Reviewed-by: Adam Jackson <ajax@redhat.com>
1501 Signed-off-by: Dave Airlie <airlied@redhat.com>
1502 Signed-off-by: Keith Packard <keithp@keithp.com>
1503
1504commit d7ee27e5e415778240919082c83a65226c6f17e6
1505Author: Dan Horák <dan@danny.cz>
1506Date: Tue Oct 29 12:09:25 2013 -0400
1507
1508 test: build the touch test only when building Xorg
1509
1510 Reviewed-by: Adam Jackson <ajax@redhat.com>
1511 Signed-off-by: Dan Horák <dan@danny.cz>
1512 Signed-off-by: Keith Packard <keithp@keithp.com>
1513
1514commit 8ff7e32c3ef7b0c13c4ab9664f651e9782d35a85
1515Author: Peter Hutterer <peter.hutterer@who-t.net>
1516Date: Wed Nov 13 17:14:11 2013 +1000
1517
1518 include: export key_is_down and friends
1519
1520 VNC needs key_is_down to check if a key is processed as down before it
1521 simulates various key releases. Make it available, because I seriously can't
1522 be bothered thinking about how to rewrite VNC to not need that.
1523
1524 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1525 Acked-by: Keith Packard <keithp@keithp.com>
1526
1527commit 0ba7fc8472f1227fd1cd51bd58048f164b8ea1ab
1528Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1529Date: Tue Nov 12 14:46:04 2013 -0800
1530
1531 Stop including inline assembly .il file for Solaris Studio builds
1532
1533 Since all the inb/outb/etc. use in the X server itself (except for
1534 xf86SlowBcopy) has been replaced by calls to libpciaccess, we no
1535 longer need to pass inline assembly files to replace the gcc inline
1536 assembly from hw/xfree86/common/compiler.h when building Xorg itself.
1537
1538 The .il files are still generated and installed in the SDK for the
1539 benefit of drivers who may use them.
1540
1541 Binary diff of before and after showed that xf86SlowBcopy was the
1542 only function changed across the Xorg binary and all modules built
1543 in the Xserver build, it just calls the outb() function now instead
1544 of having the outb instructions inlined, making it a slightly slower
1545 bcopy.
1546
1547 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1548 Reviewed-by: Keith Packard <keithp@keithp.com>
1549 Signed-off-by: Keith Packard <keithp@keithp.com>
1550
1551commit 550baf38f6096658f0bcf0ad647c4fedf93132f2
1552Author: Peter Hutterer <peter.hutterer@who-t.net>
1553Date: Fri Oct 4 10:55:52 2013 +1000
1554
1555 kdrive: fix cursor jumps on CursorOffScreen behavior
1556
1557 This patch fixes cursor jumps when there is a grab on the Xephyr window and
1558 the pointer moves outside the window.
1559
1560 So on two side-by-side 640x480 screens, a coordinate of 0/481
1561 triggers KdCursorOffscreen.
1562
1563 If the delta between two screens is 0, they share the same offset for
1564 that dimension. When searching for the new screen, the loop always rules out
1565 the current screen. So we get to the second screen, trigger the conditions
1566 where dy <= 0 and decide that this new screen is the correct one. The result
1567 is that whenever KdCursorOffScreen is called, the pointer jumps to the other
1568 screen.
1569
1570 Change to check for dy < 0 etc. so that the cursor stays on the same screen if
1571 there is no other screen at the target location.
1572
1573 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1574 Reviewed-by: Keith Packard <keithp@keithp.com>
1575
1576commit a94d945065177d73f3ee8dc0b9147264ba281136
1577Author: Peter Hutterer <peter.hutterer@who-t.net>
1578Date: Tue Oct 29 14:24:41 2013 +1000
1579
1580 kdrive: modify ephyr events to use POINTER_DESKTOP and scale them to that
1581
1582 A multi-head Xephyr instance has the pointer stuck on one screen
1583 because of bad coordinate calculation. The coordinates passed to
1584 GetPointerEvents are per-screen, so the cursor gets stuck on the left-most
1585 screen by default.
1586
1587 Adjust and mark the events as POINTER_DESKTOP, so the DIX
1588 can adjust them accordingly.
1589
1590 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1591 Reviewed-by: Keith Packard <keithp@keithp.com>
1592
1593commit d66832a3b8a8675f1e5f3656dcb1bbe95598f0ea
1594Author: Keith Packard <keithp@keithp.com>
1595Date: Thu Nov 14 13:21:54 2013 +1000
1596
1597 kdrive: handle WxH as valid geometry spec
1598
1599 If a screen size was specified as WxH, the loop returned early and kdOrigin
1600 was never advanced. Thus, screen->origin was always 0 (or whatever was given
1601 at the -origin commandline flag).
1602
1603 If a screen size was given with a bit depth (WxHxD), kdOrigin would always
1604 advance by the current screen, offsetting the next screen.
1605
1606 Signed-off-by: Keith Packard <keithp@keithp.com>
1607 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
1608 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1609
1610commit c8c5105c1d5c093675d7f571f158147f22f7572b
1611Author: Peter Hutterer <peter.hutterer@who-t.net>
1612Date: Tue Oct 29 14:33:56 2013 +1000
1613
1614 ephyr: xcb_connect returns an error, not NULL
1615
1616 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1617 Reviewed-by: Keith Packard <keithp@keithp.com>
1618
1619commit 2aa67ccef6ad8f51deb94b66defc668ddb7d226c
1620Author: Keith Packard <keithp@keithp.com>
1621Date: Thu Nov 14 10:26:20 2013 +0900
1622
1623 Update to version 1.14.99.902
1624
1625 1.15 RC2
1626
1627 Signed-off-by: Keith Packard <keithp@keithp.com>
1628
1629commit 9309d2a009a6a3b6090fc377fc7a3eded1a34dc0
1630Author: Keith Packard <keithp@keithp.com>
1631Date: Wed Nov 13 07:01:45 2013 +0900
1632
1633 xext: Fix shmint.h to not use headers outside of sdk_HEADERS
1634
1635 shmint.h is part of sdk_HEADERS, and so can't use anything not
1636 included in sdk_HEADERS.
1637
1638 busfault.h includes dix-config.h which is not. Leave the use of
1639 struct busfault in shmint.h and move the include of busfault.h to
1640 shm.c.
1641
1642 protocol-versions.h is not part of sdk_HEADERS, so instead of using
1643 that, just use XTRANS_SEND_FDS to choose whether to expose the fd
1644 passing requests directly.
1645
1646 Reported-by: Julien Cristau <jcristau@debian.org>
1647 Signed-off-by: Keith Packard <keithp@keithp.com>
1648 Tested-by: Knut Petersen <Knut_Petersen@t-online.de>
1649 Reviewed-by: Julien Cristau <jcristau@debian.org>
1650
1651 v2: also avoid using protocol-versions.h
1652
1653commit 4a251f5883b042cd902c192060a0be2b11148f2b
1654Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1655Date: Fri Nov 8 14:55:33 2013 -0800
1656
1657 xfree86: Fix build without libpciaccess
1658
1659 Regression fix from commit 04ab07ca19236d6c9a947e065fb69b0dd0d16639
1660
1661 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1662 Reviewed-by: Connor Behan <connor.behan@gmail.com>
1663
1664commit 5b02d5b7aaabf1ba8dcbdfe4525b7995b4e79f92
1665Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1666Date: Thu Nov 7 21:08:56 2013 -0800
1667
1668 Enable XTRANS_SEND_FDS on Solaris too.
1669
1670 Requires passing through the __EXTENSIONS__ and _XOPEN_SOURCE defines
1671 in order to expose the msg_control members in struct msghdr.
1672
1673 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1674 Reviewed-by: Adam Jackson <ajax@redhat.com>
1675
1676commit c4c154d18ef42dc550f2675a9ee88e07b1ca6bed
1677Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1678Date: Thu Nov 7 21:05:33 2013 -0800
1679
1680 Avoid conflicts with Solaris <sys/regset.h> defines that clash with our names
1681
1682 When building on Solaris with _XOPEN_SOURCE set to a recent XPG release,
1683 <stdlib.h> and other core headers start including <sys/regset.h>, which
1684 has a bunch of unfortunately named macros such as "CS", "ES", etc. for
1685 x86 & x64 registers which clash with existing variable & struct member
1686 names in Xorg - so #undef these so they don't interfere with our use.
1687
1688 (Yes, have filed a bug against the system headers for exposing these,
1689 but this solves the problem for building on existing releases.)
1690
1691 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1692 Reviewed-by: Adam Jackson <ajax@redhat.com>
1693
1694commit d7f9be0f8541368bf1a095ad1fbe7f38be6d3d15
1695Author: Keith Packard <keithp@keithp.com>
1696Date: Mon Nov 11 15:48:41 2013 -0800
1697
1698 Proper spelling of MAP_ANONYMOUS is MAP_ANON.
1699
1700 The former doesn't exist on BSD and the latter is available everywhere
1701 AFAIK (checked Solaris and Linux).
1702
1703 You also might want to wrap that line ;).
1704
1705 Reported-by: Mark Kettenis <mark.kettenis@xs4all.nl>
1706 Signed-off-by: Keith Packard <keithp@keithp.com>
1707
1708commit fc84166e65c35ad75f566b135dcfc305ad5a2fde
1709Author: Keith Packard <keithp@keithp.com>
1710Date: Thu Nov 7 15:18:28 2013 -0800
1711
1712 Get rid of the rest of the FD passing code when XTRANS_SEND_FDS isn't set
1713
1714 req_fds and SetReqFds in include/dixstruct.h
1715
1716 ReadFdFromClient, WriteFdToClient and the FD flushing in os/io.c
1717
1718 Signed-off-by: Keith Packard <keithp@keithp.com>
1719
1720commit a239e6faf3fce848ac0d10c48f8e817db68a493c
1721Merge: 43e5a43 f70a8bf
1722Author: Keith Packard <keithp@keithp.com>
1723Date: Mon Nov 11 15:26:12 2013 -0800
1724
1725 Merge remote-tracking branch 'jeremyhu/master'
1726
1727commit 43e5a43fa8994e50cf01dd954118f4ef2c4e7933
1728Author: Adam Jackson <ajax@redhat.com>
1729Date: Thu Nov 7 12:01:41 2013 -0500
1730
1731 present: Don't try to initialize when building without present support
1732
1733 There's a --disable-present, so it'd be nice if it worked.
1734
1735 Signed-off-by: Adam Jackson <ajax@redhat.com>
1736 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
1737 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
1738 Signed-off-by: Keith Packard <keithp@keithp.com>
1739
1740commit 0822a23e048b12c98c654e8b6af711c5f2c97141
1741Author: Keith Packard <keithp@keithp.com>
1742Date: Thu Nov 7 03:19:42 2013 -0800
1743
1744 present: Change debug output a bit to help diagnose missing vblank signals
1745
1746 Signed-off-by: Keith Packard <keithp@keithp.com>
1747 Reviewed-by: Adam Jackson <ajax@redhat.com>
1748
1749commit 20bb49ae9ba11d3dccfba191483cd682d9c9d96c
1750Author: Keith Packard <keithp@keithp.com>
1751Date: Wed Nov 6 23:56:26 2013 -0800
1752
1753 Present: Check for window/fence destroyed when idling pixmaps
1754
1755 A client destroying objects in the middle of an unflip can end up
1756 having the screen flip window or fence set to NULL in the unflip
1757 notify path. Check for these and don't try to use those objects.
1758
1759 Signed-off-by: Keith Packard <keithp@keithp.com>
1760 Reviewed-by: Adam Jackson <ajax@redhat.com>
1761
1762commit 41da295eb50fa08eaacd0ecde99f43a716fcb41a
1763Author: Keith Packard <keithp@keithp.com>
1764Date: Sun Nov 3 13:12:40 2013 -0800
1765
1766 Trap SIGBUS to handle truncated shared memory segments
1767
1768 If a client passes a section of memory via file descriptor and then
1769 subsequently truncates that file, the underlying pages will be freed
1770 and the addresses invalidated. Subsequent accesses to the page will
1771 fail with a SIGBUS error.
1772
1773 Trap that SIGBUS, figure out which segment was causing the error and
1774 then allocate new pages to fill in for that region. Mark the offending
1775 shared segment as invalid and free the resource ID so that the client
1776 will be able to tell when subsequently attempting to use the segment.
1777
1778 Signed-off-by: Keith Packard <keithp@keithp.com>
1779
1780 v2: Use MAP_FIXED to simplify the recovery logic (Mark Kettenis)
1781 v3: Also catch errors in ShmCreateSegment
1782
1783 Conflicts:
1784 include/dix-config.h.in
1785 include/xorg-config.h.in
1786
1787commit 719e880d7698d92f9b854b217ef9680aaa446f2e
1788Author: Keith Packard <keithp@keithp.com>
1789Date: Wed Nov 6 23:22:51 2013 -0800
1790
1791 Require libXtrans version 1.3.2
1792
1793 This has the FD passing support included
1794
1795 Signed-off-by: Keith Packard <keithp@keithp.com>
1796 Reviewed-by: Adam Jackson <ajax@redhat.com>
1797
1798commit bee2ec54049377e0033d49abff20d7bd069c62aa
1799Author: Keith Packard <keithp@keithp.com>
1800Date: Sun Nov 3 10:08:15 2013 -0800
1801
1802 Xext: Enable MIT-SHM FD-passing request definitions only when possible
1803
1804 Check to see if xtrans FD passing is available and use that to
1805 advertise the appropriate version of the SHM extension
1806
1807 Signed-off-by: Keith Packard <keithp@keithp.com>
1808 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1809
1810commit ea8acfe3e2f74a46c3f1e91809b4b99af18502b7
1811Author: Keith Packard <keithp@keithp.com>
1812Date: Wed Nov 6 23:06:28 2013 -0800
1813
1814 Require xextproto version 7.2.99.901
1815
1816 This includes the MIT-SHM FD passing requests
1817
1818 Signed-off-by: Keith Packard <keithp@keithp.com>
1819 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1820
1821commit 914672fefacc15386041e7c2bc3beed4faf45e9d
1822Author: Keith Packard <keithp@keithp.com>
1823Date: Thu Nov 7 14:20:26 2013 -0800
1824
1825 Remove dix-config.h config variables from xorg-config.h
1826
1827 xorg-config.h includes dix-config, so there's no need to duplicate.
1828
1829 Signed-off-by: Keith Packard <keithp@keithp.com>
1830
1831commit 7aad79c5a582ece301d950bd65f0bcb4b9956e86
1832Author: Keith Packard <keithp@keithp.com>
1833Date: Wed Nov 6 23:05:46 2013 -0800
1834
1835 Make XTrans FD passing support optional. Define only on Linux
1836
1837 Until other operating systems have a libXtrans port for FD passing,
1838 disable this on non-Linux systems.
1839
1840 Note that this define affects how libXtrans gets built into the X
1841 server, which is why it need only define the symbol
1842
1843 Signed-off-by: Keith Packard <keithp@keithp.com>
1844
1845commit f70a8bf3714d89bccaad36841ef9149e91ad3bba
1846Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1847Date: Sun Nov 10 11:40:27 2013 -0800
1848
1849 XQuartz: Don't sleep forever on exit if usleep() returns an error
1850
1851 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1852
1853commit ebcc1c214c466582d7b92826b4860256fd9c582a
1854Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1855Date: Sun Nov 10 11:26:49 2013 -0800
1856
1857 XQuartz: Update Info.plist copyright years
1858
1859 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1860
1861commit 81c123ea2dd833864f7ba217791e59acca0f7c97
1862Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1863Date: Mon Nov 4 12:13:45 2013 -0800
1864
1865 configure.ac: Move GLX section after DRI
1866
1867 The GLX section of configura.ac checks the state of DRI2, so it
1868 needs to be after DRI2=auto is resolved.
1869
1870 Also reset libgl requirement to 7.1.0 in non-dri2 case.
1871
1872 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1873 Reviewed-by: Adam Jackson <ajax@redhat.com>
1874
1875commit ab4b1fb38a61feb73d8336cc7a3399eb9d3d25be
1876Author: Andreas Schwab <schwab@suse.de>
1877Date: Thu Aug 15 12:03:42 2013 +0200
1878
1879 ARM64: Add support for aarch64
1880
1881 Signed-off-by: Andreas Schwab <schwab@suse.de>
1882 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
1883 Reviewed-by: Adam Jackson <ajax@redhat.com>
1884 Signed-off-by: Keith Packard <keithp@keithp.com>
1885
1886commit 8fcf82cbf015609e00b3feb9bc567133c7aee5cf
1887Author: Keith Packard <keithp@keithp.com>
1888Date: Tue Nov 5 18:39:11 2013 -0800
1889
1890 Use $GL_LIBS instead of -lGL for linking
1891
1892 -lGL presumes that the GL library is in the system path, while
1893 $GL_LIBS is auto-detected.
1894
1895 Signed-off-by: Keith Packard <keithp@keithp.com>
1896
1897commit f348935e7d2c84a438aba08eee2f2c4e0dc77d20
1898Author: Keith Packard <keithp@keithp.com>
1899Date: Tue Nov 5 18:37:12 2013 -0800
1900
1901 Link with xshmfence, reference miSyncShmScreenInit in sdksyms
1902
1903 This gets the server to link with xshmfence again, and also ensures
1904 that the miSyncShm code is linked into the server with the reference
1905 from sdksyms.
1906
1907 Signed-off-by: Keith Packard <keithp@keithp.com>
1908
1909commit 5f1e832694e57986c0185048a941b3af51b2f85f
1910Author: Julien Cristau <jcristau@debian.org>
1911Date: Tue Nov 5 07:08:21 2013 -0800
1912
1913 os: Actually use the computed clockid in GetTimeInMicros
1914
1915 The selection of which clock to use for this function was not actually
1916 getting used when fetching the final clock value.
1917
1918 Reported-by: Julien Cristau <jcristau@debian.org>
1919 Signed-off-by: Keith Packard <keithp@keithp.com>
1920
1921commit 903a058370645ea075ea98d380fd565efb6160c9
1922Author: Keith Packard <keithp@keithp.com>
1923Date: Mon Nov 4 19:01:26 2013 -0800
1924
1925 hw/xfree86: Link libdri3 only when DRI3 is defined
1926
1927 Don't attempt to link non-existant libraries...
1928
1929 Signed-off-by: Keith Packard <keithp@keithp.com>
1930
1931commit aaf0e29619196a283fee7ead2020a91032d84f48
1932Author: Keith Packard <keithp@keithp.com>
1933Date: Sun Nov 3 09:56:02 2013 -0800
1934
1935 Disable DRI3 and sync fence FD functions if xshmfence isn't available
1936
1937 Make sure the server can build when the xshmfence library isn't present
1938
1939 Signed-off-by: Keith Packard <keithp@keithp.com>
1940
1941commit 33c85beed521c9db140cadd8c5aa9992398ee1fe
1942Author: Keith Packard <keithp@keithp.com>
1943Date: Fri Nov 1 01:01:58 2013 -0700
1944
1945 Set version to 1.14.99.901 (1.15 RC1)
1946
1947 Signed-off-by: Keith Packard <keithp@keithp.com>
1948
1949commit 9db1d66f6d2c6857de220d8f1c08965962fe1416
1950Author: Keith Packard <keithp@keithp.com>
1951Date: Fri Nov 1 01:43:59 2013 -0700
1952
1953 Add dri3 and present directorys to distribution
1954
1955 Signed-off-by: Keith Packard <keithp@keithp.com>
1956
1957commit 977e2644b1d9e13b44debcb9a372bbc832ee32a3
1958Merge: 66c5ee0 f36f5a6
1959Author: Keith Packard <keithp@keithp.com>
1960Date: Fri Nov 1 00:34:18 2013 -0700
1961
1962 Merge remote-tracking branch 'whot/for-keith'
1963
1964commit 66c5ee0ff4690e630a69656331fd9d545c99c26c
1965Author: Egbert Eich <eich@freedesktop.org>
1966Date: Wed Aug 14 18:18:56 2013 +0200
1967
1968 config/APM: Add option to disable building of APM support on Linux
1969
1970 APM support in the Xserver was used to restore the console mode
1971 prior to a power management event. This was to ensure the mode
1972 upon suspend/resume was one that the system firmware or kernel
1973 could deal with.
1974 APM support is now largely obsolete, KMS drivers don't require a
1975 mode restoration anyhow. Therefore it should be possible to disable
1976 this feature.
1977
1978 (small modification by keithp - move test for XF86PM flag after check
1979 for APM, then move XF86PM flag to xorg-config.h.in)
1980
1981 Signed-off-by: Egbert Eich <eich@freedesktop.org>
1982 Tested-by: Keith Packard <keithp@keithp.com>
1983 Signed-off-by: Keith Packard <keithp@keithp.com>
1984
1985commit f36f5a65f639b6524191d888d5bf89e73027156c
1986Author: Peter Hutterer <peter.hutterer@who-t.net>
1987Date: Wed Oct 30 15:40:58 2013 +1000
1988
1989 sync: fix corner-case in triggering idle alarms
1990
1991 ProcessInputEvent() resets the device idle times. If idle time was higher than
1992 the lower bracket, this should trigger an event in the idle time wakeup
1993 handler.
1994
1995 If processing is slow, the idle time may advance past the lower bracket
1996 between the reset and the time the BlockHandler is called. In that case, we'd
1997 never schedule a wakeup to handle the event, causing us to randomly miss
1998 events.
1999
2000 Ran tests with a neg transition trigger on 5ms with 200 repeats of the test
2001 and it succeeded. Anything below that gets a bit tricky to make sure the
2002 server sees the same idle time as the client usleeps for.
2003
2004 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2005 Reviewed-by: Keith Packard <keithp@keithp.com>
2006
2007commit f994d99539192a10a8c83047e11991f52e0ca677
2008Author: Egbert Eich <eich@freedesktop.org>
2009Date: Wed Aug 14 18:18:39 2013 +0200
2010
2011 config/ACPI: Add option to disable building of ACPI support on Linux
2012
2013 ACPI support in the Xserver was used to restore the console mode
2014 prior to a power management event. This was to ensure the mode
2015 upon suspend/resume was one that the system firmware or kernel
2016 could deal with.
2017 The feature depended on acpid to be running. Most of this functionality
2018 is now take over by systemd, KMS drivers don't require a mode restoration
2019 anyhow. Therefore it should be possible to disable this feature under
2020 some circumstances.
2021
2022 Tested-by: Keith Packard <keithp@keithp.com>
2023 Signed-off-by: Keith Packard <keithp@keithp.com>
2024
2025commit 2aa5092b88174cb9988076ae96298217c97ad75f
2026Author: Egbert Eich <eich@freedesktop.org>
2027Date: Mon Aug 12 18:21:33 2013 +0200
2028
2029 DDX/Randr: Avoid server crash when xrandr SetConfig is called while switched away
2030
2031 A call to Xrandr SetScreenConfig (for randr 1.1) causes the Xserver to
2032 crash when xf86SetViewport() which does not check if the hardware is
2033 accessible.
2034 Wrap accesses to xf86SetViewport() with if (vtSema) { ... } to avoid that.
2035
2036 Signed-off-by: Egbert Eich <eich@freedesktop.org>
2037 Reviewed-by: Keith Packard <keithp@keithp.com>
2038 Signed-off-by: Keith Packard <keithp@keithp.com>
2039
2040commit 508e05777aba823c93a2417cc26d18c9f061ca26
2041Author: Egbert Eich <eich@freedesktop.org>
2042Date: Mon Aug 12 18:20:36 2013 +0200
2043
2044 DDX/Events: Distinguish between Input- and GeneralHandlers in xf86VTSwitch()
2045
2046 When enabling/disabling input handlers in xf86VTSwitch() we treat Input-
2047 and GeneralHandlers equally. The result is that after a VT switch the
2048 masks for EnabledDevices and AllSockets are equal and the distiction
2049 between both types is lost.
2050
2051 Signed-off-by: Egbert Eich <eich@freedesktop.org>
2052 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
2053 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
2054 Signed-off-by: Keith Packard <keithp@keithp.com>
2055
2056commit 0cb33ce34081d7d123ec565582d7d19bdb3964ad
2057Author: Egbert Eich <eich@freedesktop.org>
2058Date: Sat Oct 5 08:02:28 2013 +0200
2059
2060 DDX/modes: Add a sanity check when using screen sizes from EDID
2061
2062 EDID sometimes lies about screen sizes. Since the screen size is used
2063 by clients to determine the DPI a wrong ration will lead to terrible
2064 looking fonts.
2065 Add a sanity check for the h/v ratio cutting off at 2.4. This would
2066 still accept the cinemascope aspect ratio as valid.
2067 Also add message suggesting to add a quirk table entry.
2068
2069 Signed-off-by: Egbert Eich <eich@freedesktop.org>
2070 Reviewed-by: Keith Packard <keithp@keithp.com>
2071 Signed-off-by: Keith Packard <keithp@keithp.com>
2072
2073commit 41d4beb2616ceb3f1a1b8694733e85bae70de59a
2074Author: Egbert Eich <eich@freedesktop.org>
2075Date: Wed Aug 14 11:50:26 2013 +0200
2076
2077 DDX/DPMS: Call dixSaveScreens() also when screen is turned on
2078
2079 DMPS calls dixSaveScreens() when turned off but not when turned
2080 on. In most cases this is irrelevant as DPMS is done when a
2081 key is hit in which case dixSaveScreens() will be called to
2082 unblank anyhow. This isn't the case if we use xset (or the
2083 DPMS extension directly) to unblank.
2084 Check screenIsSaved to make sure the state needs to be changed
2085 before calling dixSaveScreens().
2086
2087 Signed-off-by: Egbert Eich <eich@freedesktop.org>
2088 Reviewed-by: Keith Packard <keithp@keithp.com>
2089 Signed-off-by: Keith Packard <keithp@keithp.com>
2090
2091commit 623c4147650d0404cfbea0f9b7df66dc7d928e00
2092Author: Søren Sandmann Pedersen <ssp@redhat.com>
2093Date: Mon Oct 21 17:11:56 2013 -0400
2094
2095 ephyr: Ensure stride of private framebuffer is multiple of 4
2096
2097 The fb layer of X can't deal with strides that are not a multiple of
2098 4, so when Xephyr allocates its own framebuffer it should make sure to
2099 align it.
2100
2101 This fixes crashes and rendering corruption when Xephyr runs in a
2102 depth that is different from the host X server and its screen size is
2103 not a multiple of 4 / depth. (This is particularly easy to trigger if
2104 you use the -resizeable option).
2105
2106 Reviewed-by: Eric Anholt <eric@anholt.net>
2107 Signed-off-by: Soren Sandmann <ssp@redhat.com>
2108 Reviewed-by: Adam Jackson <ajax@redhat.com>
2109
2110commit 97cf53cc2ad7ecfdd495133bad31d0ec7d939326
2111Author: Søren Sandmann Pedersen <ssp@redhat.com>
2112Date: Mon Oct 21 16:58:54 2013 -0400
2113
2114 ephyr: hostx_screen_init(): Fix bits_per_pixel and bytes_per_line
2115
2116 When the depth of the Xephyr server matches that of the host X server,
2117 Xephyr simply uses the buffer associated with the XImage as its
2118 framebuffer. In this case, it is correct to get the bits_per_pixel and
2119 bytes_per_line values returned from hostx_screen_init() from the XImage.
2120
2121 However, when the depth doesn't match the host, Xephyr uses a private
2122 framebuffer that is periodically copied to the XImage. In this case,
2123 the returned values of bits_per_pixel and bytes_per_line should be
2124 those of the private framebuffer, not those of the XImage.
2125
2126 Reviewed-by: Eric Anholt <eric@anholt.net>
2127 Signed-off-by: Soren Sandmann <ssp@redhat.com>
2128 Reviewed-by: Adam Jackson <ajax@redhat.com>
2129
2130commit 55246b67b755d4c1039d54971fe3f77ea60d604e
2131Author: Søren Sandmann Pedersen <ssp@redhat.com>
2132Date: Tue Oct 8 17:45:40 2013 -0400
2133
2134 xf86AddBusDeviceToConfigure(): Store device in DevToConfig[i].pVideo
2135
2136 After fc3ab84d the pVideo field in DevToConfig[i] is no longer
2137 initialized, so it's always NULL. This causes the duplicate finding
2138 algorithm in the beginning of the function to not work anymore as it
2139 is based on this field.
2140
2141 The symptom of this bug is that X -configure reports
2142
2143 Number of created screens does not match number of detected devices.
2144 Configuration failed.
2145 Server terminated with error (2). Closing log file.
2146
2147 rather than producing a working config file.
2148
2149 This patch fixes that bug by initializing the field before calling
2150 xf86PciConfigureNewDev().
2151
2152 Cc: tvignatti@gmail.com
2153 Signed-off-by: Soren Sandmann <ssp@redhat.com>
2154 Reviewed-by: Adam Jackson <ajax@redhat.com>
2155
2156commit 95bf478b78e466002c382bcde7d4d62591e9215d
2157Merge: abf5d5a bb745f2
2158Author: Keith Packard <keithp@keithp.com>
2159Date: Thu Oct 31 18:18:19 2013 -0700
2160
2161 Merge remote-tracking branch 'jeremyhu/master'
2162
2163commit abf5d5ac12437ebe156b4dd500c2acd69eea3654
2164Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
2165Date: Thu Oct 31 08:57:56 2013 -0700
2166
2167 miext/damage: Partial revert of "Only wrap into the GC ops chain if there's a listener (v3)"
2168
2169 Fixes regression from: 4dc2a76740d921c824a4d8193f39dd373475f02a
2170
2171 http://lists.x.org/archives/xorg-devel/2013-May/036241.html
2172 https://trac.macports.org/ticket/38993
2173
2174 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
2175 Reviewed-by: Adam Jackson <ajax@redhat.com>
2176 Signed-off-by: Keith Packard <keithp@keithp.com>
2177
2178commit e4636072949a76477fe7c9d54662a0b3536b1372
2179Merge: 17ed7ac 5c5c1b7
2180Author: Keith Packard <keithp@keithp.com>
2181Date: Thu Oct 31 17:05:48 2013 -0700
2182
2183 Merge remote-tracking branch 'keithp/dri3'
2184
2185commit 17ed7ac1fe7426d66f688087c09da5ff1dbb41d3
2186Author: Adam Jackson <ajax@redhat.com>
2187Date: Thu Oct 31 18:39:22 2013 -0400
2188
2189 glx: Lie about GLX_Y_INVERTED_EXT
2190
2191 Well, that was lame. The problem with reporting y inversion honestly is
2192 that libGL asks the driver _its_ opinion of Y inversion, which it just
2193 fabricates from whole cloth. So then when libGL goes to compare the
2194 driver's idea of fbconfigs with that of the server - a fairly dumb idea
2195 to begin with - nothing matches, and direct rendering fails, and
2196 sadness.
2197
2198 So until the DRI drivers are fixed we should just continue to lie about
2199 Y inversion. GLX_DONT_CARE is what libGL would make up for that
2200 attribute if we hadn't sent it, so just send that instead.
2201
2202 Signed-off-by: Adam Jackson <ajax@redhat.com>
2203 Tested-by: Keith Packard <keithp@keithp.com>
2204 Signed-off-by: Keith Packard <keithp@keithp.com>
2205
2206commit 5c5c1b77982a9af7279a90bc3c2be48adaa9c778
2207Author: Keith Packard <keithp@keithp.com>
2208Date: Thu Jul 11 16:11:57 2013 -0700
2209
2210 present: Add Present extension
2211
2212 Provides both a software implementation using timers and driver hooks
2213 to base everything on vblank intervals.
2214
2215 Signed-off-by: Keith Packard <keithp@keithp.com>
2216 Reviewed-by: Adam Jackson <ajax@redhat.com>
2217
2218commit 563138298868f62501875d3016f03469dcffaad0
2219Author: Keith Packard <keithp@keithp.com>
2220Date: Tue Apr 9 19:59:39 2013 -0700
2221
2222 dri3: Add DRI3 extension
2223
2224 Adds DRM compatible fences using futexes.
2225 Uses FD passing to get pixmaps from DRM applications.
2226
2227 Signed-off-by: Keith Packard <keithp@keithp.com>
2228 Reviewed-by: Adam Jackson <ajax@redhat.com>
2229
2230commit fdec793cdc2ef9a6ea66b311cb1068a7bd4a3be3
2231Author: Keith Packard <keithp@keithp.com>
2232Date: Thu Jan 17 13:46:55 2013 -0800
2233
2234 Add support for MIT-SHM AttachFd request
2235
2236 This passes a file descriptor from the client to the server, which is
2237 then mmap'd
2238
2239 Signed-off-by: Keith Packard <keithp@keithp.com>
2240 Reviewed-by: Adam Jackson <ajax@redhat.com>
2241
2242commit 9fd35daa3160fd36f00ed354bfcbefefa1353cce
2243Author: Keith Packard <keithp@keithp.com>
2244Date: Thu Jan 17 13:43:02 2013 -0800
2245
2246 Add interfaces to get FDs from clients over the socket
2247
2248 This adds two interfaces:
2249
2250 void SetReqFds(ClientPtr client, int req_fds)
2251
2252 Marks the number of file descriptors expected for this
2253 request. Call this before any request processing so that
2254 any un-retrieved file descriptors will be closed
2255 automatically.
2256
2257 int ReadFdFromClient(ClientPtr client)
2258
2259 Reads the next queued file descriptor from the connection. If
2260 this request is not expecting any more file descriptors, or
2261 if there are no more file descriptors available from the
2262 connection, then this will return -1.
2263
2264 Signed-off-by: Keith Packard <keithp@keithp.com>
2265 Reviewed-by: Adam Jackson <ajax@redhat.com>
2266
2267commit 264fc3abe5f18341d0cf9ddb6766e10e4154e447
2268Author: Keith Packard <keithp@keithp.com>
2269Date: Thu Oct 31 13:01:46 2013 -0700
2270
2271 misync: Don't smash custom screen sync functions
2272
2273 There was a check to avoid smashing custom functions, but the sense
2274 was backwards causing it to always smash them, and also not set them otherwise.
2275
2276 Signed-off-by: Keith Packard <keithp@keithp.com>
2277 Reviewed-by: Adam Jackson <ajax@redhat.com>
2278
2279commit 2d96948ab5c952b68875ac63844cf7d778d4bf63
2280Author: Keith Packard <keithp@keithp.com>
2281Date: Thu Jul 11 16:10:34 2013 -0700
2282
2283 os: Add GetTimeInMicros
2284
2285 64-bit higher resolution current time value.
2286
2287 Signed-off-by: Keith Packard <keithp@keithp.com>
2288 Reviewed-by: Adam Jackson <ajax@redhat.com>
2289
2290commit 0c33f47281c36726848daf513fb0483cdea57bff
2291Author: Keith Packard <keithp@keithp.com>
2292Date: Thu Jul 11 16:09:34 2013 -0700
2293
2294 Add swapll to byte swap 64-bit datatypes
2295
2296 Signed-off-by: Keith Packard <keithp@keithp.com>
2297 Reviewed-by: Adam Jackson <ajax@redhat.com>
2298
2299commit 26f013ba45b08a02bb028a461af68288a86fadb1
2300Author: Keith Packard <keithp@keithp.com>
2301Date: Thu Jul 11 16:08:41 2013 -0700
2302
2303 Add a RegionDuplicate function
2304
2305 This allocates a new region structure and copies a source region into
2306 it in a single API rather than forcing the caller to do both steps themselves.
2307
2308 Signed-off-by: Keith Packard <keithp@keithp.com>
2309 Reviewed-by: Adam Jackson <ajax@redhat.com>
2310
2311commit d25c217964eb1fe54c3a54bca4cac7f47b4b9fdf
2312Author: Keith Packard <keithp@keithp.com>
2313Date: Mon Jan 14 14:24:36 2013 -0800
2314
2315 Clean up a couple of warnings in os/
2316
2317 Signed-off-by: Keith Packard <keithp@keithp.com>
2318 Reviewed-by: Adam Jackson <ajax@redhat.com>
2319
2320commit 7710f2b927ec8e3c631f72d66d494b523377b48e
2321Author: Keith Packard <keithp@keithp.com>
2322Date: Fri Jan 18 21:43:40 2013 -0800
2323
2324 Xext: Move MIT-SHM 'ShmDesc' to shmint.h
2325
2326 This data structure is required to use shared memory objects in any
2327 extension. That includes the Xv extension, which (before this patch)
2328 duplicated the definition of this structure in its own code.
2329
2330 Signed-off-by: Keith Packard <keithp@keithp.com>
2331 Reviewed-by: Adam Jackson <ajax@redhat.com>
2332
2333commit e8961b718d82f1c081ec110d8d962f64e8406b82
2334Author: Peter Hutterer <peter.hutterer@who-t.net>
2335Date: Tue Oct 22 14:24:52 2013 +1000
2336
2337 os: use a constant for backtrace array size
2338
2339 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2340 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
2341
2342commit f12a9ed870017f35cf6d2a82b1405e843aae42ac
2343Author: Peter Hutterer <peter.hutterer@who-t.net>
2344Date: Tue Oct 29 15:24:11 2013 +1000
2345
2346 configure: remove a comment
2347
2348 94ed0ba1b5043ad9fc33b42756af447d5ab15bbd moved backtracing into the DIX, so
2349 this comment is outdated. since no-one noticed and it's easier to just grep
2350 than update file references, remove the comment.
2351
2352 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2353 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
2354
2355commit bb745f2b45768d0936f3960951e1a5cee8b03194
2356Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
2357Date: Thu Oct 31 08:50:57 2013 -0700
2358
2359 XQuartz: Bump bundle version to 2.7.6
2360
2361 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
2362
2363commit 2eedf42c229e2251035d716a96b654222155cd51
2364Author: Adam Jackson <ajax@redhat.com>
2365Date: Thu Oct 31 09:23:52 2013 -0400
2366
2367 dri: Fix GLX_Y_INVERTED_EXT fbconfig attribute
2368
2369 We're Y-inverted from GL's coordinates, so this is correct. gnome-shell
2370 doesn't seem to check this - somewhat reasonable since the only server
2371 that answered the other way around was Xglx - but kwin does, and
2372 upside-down hilarity ensues.
2373
2374 Tested-by: maelcum on #xorg-devel
2375 Signed-off-by: Adam Jackson <ajax@redhat.com>
2376 Signed-off-by: Keith Packard <keithp@keithp.com>
2377
2378commit 902ff0b3497d202b86bf9a411e17db7b694d6eaa
2379Author: Adam Jackson <ajax@redhat.com>
2380Date: Tue Oct 29 12:58:14 2013 -0400
2381
2382 xfree86: Bump video/input/extension ABIs for 1.15
2383
2384 Signed-off-by: Adam Jackson <ajax@redhat.com>
2385 Reviewed-by: Julien Cristau <jcristau@debian.org>
2386 Signed-off-by: Keith Packard <keithp@keithp.com>
2387
2388commit 899451ae595f5f51a500295c624bf7ad14aca602
2389Author: Keith Packard <keithp@keithp.com>
2390Date: Wed Oct 30 08:54:24 2013 -0700
2391
2392 kdrive/ephyr: Don't discard one-time driver structure at server reset
2393
2394 KdScreenInfo is constructed at server startup time, and not
2395 re-generated at server reset time. Freeing the 'driver' element at
2396 reset time means this information is lost, and the server crashes
2397 pretty quickly afterwards.
2398
2399 Signed-off-by: Keith Packard <keithp@keithp.com>
2400 Reviewed-by: Adam Jackson <ajax@redhat.com>
2401
2402commit 85ae44f07fa21b4a48d95d736dc8135bd4f751d0
2403Author: Michal Srb <msrb@suse.com>
2404Date: Wed Oct 30 13:33:51 2013 +0200
2405
2406 randr: deliver Output and Crtc events of attached output providers.
2407
2408 Consider all attached output providers when looking for changed outputs and
2409 crtcs.
2410
2411 Reviewed-by: Dave Airlie <airlied@redhat.com>
2412 Signed-off-by: Michal Srb <msrb@suse.com>
2413 Signed-off-by: Keith Packard <keithp@keithp.com>
2414
2415commit a9ca93dcf9a711b3d60fbad31cbd44e0c82b6f42
2416Author: Michal Srb <msrb@suse.com>
2417Date: Wed Oct 30 13:33:50 2013 +0200
2418
2419 randr: send RRResourceChangeNotify event
2420
2421 Send RRResourceChangeNotify event when provider, output or crtc was created or
2422 destroyed. I.e. when the list of resources returned by RRGetScreenResources and
2423 RRGetProviders changes.
2424
2425 Reviewed-by: Dave Airlie <airlied@redhat.com>
2426 Signed-off-by: Michal Srb <msrb@suse.com>
2427 Signed-off-by: Keith Packard <keithp@keithp.com>
2428
2429commit 4b39ea8a918fe6117bf34dc1cba67a4e56d0d1fb
2430Author: Michal Srb <msrb@suse.com>
2431Date: Wed Oct 30 13:33:49 2013 +0200
2432
2433 randr: send RRProviderChangeNotify event
2434
2435 Send RRProviderChangeNotify event when a provider becomes output source or
2436 offload sink.
2437
2438 Reviewed-by: Dave Airlie <airlied@redhat.com>
2439 Signed-off-by: Michal Srb <msrb@suse.com>
2440 Signed-off-by: Keith Packard <keithp@keithp.com>
2441
2442commit 1470c0a132806fb2ce817515ad4ddfe66eb7c98f
2443Merge: 04ab07c 9f4afe7
2444Author: Keith Packard <keithp@keithp.com>
2445Date: Tue Oct 29 21:00:13 2013 -0700
2446
2447 Merge remote-tracking branch 'whot/for-keith'
2448
2449commit 9f4afe7e835cf9c513387ae01eaf8b7a6ba4aac9
2450Author: Gaetan Nadon <memsize@videotron.ca>
2451Date: Mon Oct 28 18:32:36 2013 -0400
2452
2453 config: replace deprecated use of AC_OUTPUT with AC_CONFIG_FILES
2454
2455 Fix Automake warning: AC_OUTPUT should be used without arguments.
2456 www.gnu.org/software/autoconf/manual/autoconf.html#Configuration-Files
2457
2458 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
2459 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
2460 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2461
2462commit e16dea72d6f45dd6d7788362bb7dbe76263719b1
2463Author: Peter Hutterer <peter.hutterer@who-t.net>
2464Date: Wed Oct 23 15:16:25 2013 +1000
2465
2466 dmx: queue button events with a flags of 0
2467
2468 Setting POINTER_SCREEN with a unset valuator mask causes a jump to 0/0. Set
2469 the flags to 0 so we don't generate any motion on a button event.
2470
2471 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2472 Reviewed-by: Keith Packard <keithp@keithp.com>
2473
2474commit 04ab07ca19236d6c9a947e065fb69b0dd0d16639
2475Author: Connor Behan <connor.behan@gmail.com>
2476Date: Thu Oct 17 18:26:28 2013 -0700
2477
2478 xfree86: Find primary entity when bus types are nominally different
2479
2480 As of server 1.13, systems with DRM and Udev will have BUS_PLATFORM as
2481 their primary bus type. However, drivers not implementing a
2482 platformProbe function will still create entities of type BUS_PCI. We
2483 need to account for this when checking for the primary entity.
2484
2485 Signed-off-by: Connor Behan <connor.behan@gmail.com>
2486 Acked-by: Tormod Volden <debian.tormod@gmail.com>
2487 Reviewed-by: Adam Jackson <ajax@redhat.com>
2488 Signed-off-by: Keith Packard <keithp@keithp.com>
2489
2490commit b32a4c91ccd479638c2bd2c0143b6ea170c717d1
2491Merge: be66809 c671e93
2492Author: Keith Packard <keithp@keithp.com>
2493Date: Tue Oct 29 09:37:30 2013 -0700
2494
2495 Merge remote-tracking branch 'idr/glx-float-fbconfig'
2496
2497commit be6680967a479eedbcab2fe1718c5f981e1029c7
2498Author: Adam Jackson <ajax@redhat.com>
2499Date: Wed Jul 10 10:00:46 2013 -0400
2500
2501 glx: convert to direct GL dispatch (v2)
2502
2503 We now expect to be linked against something that provides the GL API,
2504 instead of manually grubbing about in the DRI driver's dispatch table.
2505 Since the GLX we expose calls GL functions that are meant to be looked
2506 up dynamically, also add a way to thunk through to GetProcAddress.
2507
2508 This includes a refresh of the generated sources, which requires a
2509 correspondingly new Mesa.
2510
2511 The GetProcAddress stubs are at the moment merely enough to make this
2512 link against Mesa 9.2, but should really be provided for everything not
2513 in the OpenGL 1.2 ABI.
2514
2515 v2: Explicitly hide the GetProcAddress stubs so we can't conflict with
2516 libGL symbols; fix leading tab/space issues [anholt]
2517
2518 Reviewed-by: Keith Packard <keithp@keithp.com>
2519 Signed-off-by: Adam Jackson <ajax@redhat.com>
2520
2521commit 8aacf47e1778d8b72811b025a82452b933d3c1f2
2522Author: Adam Jackson <ajax@redhat.com>
2523Date: Fri Oct 4 12:58:19 2013 -0400
2524
2525 glx: Remove DRI1 AIGLX (v2)
2526
2527 Mesa doesn't ship DRI1 drivers as of 8.0, which is about 18 months and
2528 three releases ago. The main reason to have wanted DRI1 AIGLX was to
2529 get a GLX compositor working, but DRI1's (lack of) memory management API
2530 meant that the cost of a GLX compositor was breaking direct GLX apps,
2531 which isn't a great tradeoff.
2532
2533 Of the DRI1 drivers Mesa has dropped, I believe only mga stands to lose
2534 some functionality here, since it and only it has support for
2535 NV_texture_rectangle. Since that's required for every extant GLX
2536 compositor I know of, I conclude that anybody with a savage, say, would
2537 probably not notice AIGLX going away, since they wouldn't be running a
2538 GLX compositor in the first place.
2539
2540 In the future we'd like to use GL in the server in a more natural way,
2541 as just another EGL client, including in the GLX implementation itself.
2542 Since there's no EGL implemented for DRI1 drivers, this would already
2543 doom AIGLX on DRI1 (short of entirely forking the GLX implementation,
2544 which I'm not enthusiastic about).
2545
2546 v2: Remove DRI1 from AIGLX conditionals in configure.ac [anholt]
2547
2548 Reviewed-by: Eric Anholt <eric@anholt.net>
2549 Signed-off-by: Adam Jackson <ajax@redhat.com>
2550
2551commit 7f5adf73a0f9a951a6df201532b4031d38054369
2552Author: Adam Jackson <ajax@redhat.com>
2553Date: Wed Oct 2 17:38:05 2013 -0400
2554
2555 glx: Simplify glXDestroyContext
2556
2557 We can just free the resource unconditionally here. ContextGone (which
2558 FreeResourceByType will call) already does:
2559
2560 cx->idExists = GL_FALSE;
2561 if (!cx->currentClient) {
2562 __glXFreeContext(cx);
2563 }
2564
2565 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2566 Signed-off-by: Adam Jackson <ajax@redhat.com>
2567
2568commit 53653f4889dcaea0cf67f8a4180af71eeb4b2f6b
2569Author: Adam Jackson <ajax@redhat.com>
2570Date: Mon Aug 5 15:46:11 2013 -0400
2571
2572 glx: Fix a corner case in DrawableGone
2573
2574 Losing the drawable does not change our notion of current client. Since
2575 the GL under us doesn't understand having a current context without
2576 current drawables (sigh), we do still need to loseCurrent so that we
2577 re-bind the context on the next request.
2578
2579 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=65030
2580 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2581 Signed-off-by: Adam Jackson <ajax@redhat.com>
2582
2583commit 276d8057aa7f6e36c9c874257a6383f416c5cedd
2584Author: Adam Jackson <ajax@redhat.com>
2585Date: Sat Aug 3 09:47:55 2013 -0400
2586
2587 glx: Fix memory leak in context garbage collection (v2)
2588
2589 I broke this, back in:
2590
2591 commit a48dadc98a28c969741979b70b7a639f24f4cbbd
2592 Author: Adam Jackson <ajax@redhat.com>
2593 Date: Mon Mar 21 11:59:29 2011 -0400
2594
2595 glx: Reimplement context tags
2596
2597 In that, I changed the glx client state to not explicitly track the list
2598 of current contexts for the client (since that was what we were deriving
2599 tags from). The bug was that I removed the code for same from
2600 glxClientCallback without noticing that it had the side effect of
2601 effectively de-currenting those contexts, so that ContextGone could free
2602 them. So, if you had a client exit with a context still current, the
2603 context's memory would leak. Not a huge deal for direct clients, but
2604 viciously bad for indirect, since the swrast context state at the bottom
2605 of Mesa is like 15M.
2606
2607 Fix this by promoting Bool isCurrent to ClientPtr currentClient, so that
2608 we have a back-pointer to chase when walking the list of contexts when
2609 ClientStateGone happens.
2610
2611 v2: Explicitly call __glXFreeContext on the ClientStateGone path. Our
2612 current context might be one we got from EXT_import_context and whose
2613 creating client has since died. Without the explicit call, the creating
2614 client's FreeClientResources would not free the context because it's
2615 still current, and the using client's FreeClientResources would not free
2616 the context because it's not an XID it created. This matches the logic
2617 from a48dadc.
2618
2619 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2620 Signed-off-by: Adam Jackson <ajax@redhat.com>
2621
2622commit 75b362763cca8b6139582ad85e3b58f0b2e30a1c
2623Author: Adam Jackson <ajax@redhat.com>
2624Date: Tue Oct 8 15:02:44 2013 -0400
2625
2626 glx: Only take a Pixmap reference if creating the GLXPixmap succeeded
2627
2628 Reviewed-by: Julien Cristau <jcristau@debian.org>
2629 Signed-off-by: Adam Jackson <ajax@redhat.com>
2630
2631commit 4292a39c8014f9f4920bed3001802e4a3e523c5e
2632Author: Peter Hutterer <peter.hutterer@who-t.net>
2633Date: Mon Oct 28 12:10:10 2013 +1000
2634
2635 Xi: fix logic error when calculating emulated motion events
2636
2637 gcc -Wlogical-op
2638 exevents.c: In function 'DeliverEmulatedMotionEvent':
2639 exevents.c:1480:13: warning: logical 'or' of collectively exhaustive
2640 tests is always true [-Wlogical-op]
2641
2642 The relevant snippet of exevents.c:
2643
2644 1479 if (ti->listeners[0].type != LISTENER_POINTER_REGULAR ||
2645 1480 ti->listeners[0].type != LISTENER_POINTER_GRAB)
2646 1481 return;
2647
2648 This condition was always true, causing dropped motion events.
2649
2650 Reported-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2651 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2652 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2653
2654commit c671e935e7c44d4711ba9588ee96cb9796b879e6
2655Author: Daniel Czarnowski <daniel.czarnowski@intel.com>
2656Date: Mon Oct 21 19:48:38 2013 +0200
2657
2658 glx: Enable GLX_ARB_fbconfig_float for DRI2 drivers
2659
2660 v2 (idr): Maintain sort order of extension list. Fixed rebase error
2661 (comment header for initializeExtensions was in front of
2662 __glXDRIinvalidateBuffers). Re-write commit message.
2663
2664 Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
2665 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2666 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2667
2668commit ccc8bb1153ec669f7540a174a81a8ec0e0d6dd67
2669Author: Daniel Czarnowski <daniel.czarnowski@intel.com>
2670Date: Wed Oct 23 12:53:59 2013 +0200
2671
2672 glx: Handle float config types in glxConvertConfigs
2673
2674 Replaces old use of floatMode attribute with new, extended range of
2675 values in __DRI_ATTRIB_RENDER_TYPE. Also adds new conditions, where the
2676 float modes support requires it. Enables support for not only float
2677 configs, but packed float configs as well.
2678
2679 v2 (idr): Whitespace and formatting fixes. Refactor render type
2680 vs. pbuffer checking to a separate function that includes a quote from
2681 the spec. Re-write commit message. Fix compiler warnings:
2682
2683 glxdricommon.c: In function 'glxConvertConfigs':
2684 glxdricommon.c:212:35: warning: pointer targets in passing argument 3 of 'core->getConfigAttrib' differ in signedness [-Wpointer-sign]
2685 glxdricommon.c:212:35: note: expected 'unsigned int *' but argument is of type 'int *'
2686 glxdricommon.c:230:35: warning: pointer targets in passing argument 3 of 'core->getConfigAttrib' differ in signedness [-Wpointer-sign]
2687 glxdricommon.c:230:35: note: expected 'unsigned int *' but argument is of type 'int *'
2688
2689 Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
2690 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2691 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2692
2693commit 4e5eb15b4cc170499cf87e6633c1737f0b1c514b
2694Author: Daniel Czarnowski <daniel.czarnowski@intel.com>
2695Date: Mon Oct 21 19:57:53 2013 +0200
2696
2697 glx: Correctly set render type enum
2698
2699 The type specified is bits, but the context wants an enum.
2700
2701 v2 (idr): Fix various whitespace and formatting problems. Rename
2702 convFBconfRenderTypeBits2CtxRenderType to
2703 renderTypeBitsToRenderTypeEnum. Re-write commit message.
2704
2705 Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
2706 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2707 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2708
2709commit eabb523aa45f3d8940cadac197ef1225fbfd5c56
2710Author: Daniel Czarnowski <daniel.czarnowski@intel.com>
2711Date: Mon Oct 21 19:45:35 2013 +0200
2712
2713 glx: Allow float renderType in glXCreateContextAttribsARB
2714
2715 This enables GLX_RGBA_FLOAT_TYPE_ARB and
2716 GLX_RGBA_UNSIGNED_FLOAT_TYPE_EXT as valid renderType parameters to
2717 glXCreateContextAttribsARB.
2718
2719 v2 (idr) : Use C-style comments and fix word wrapping. Re-write commit
2720 message.
2721
2722 Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
2723 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2724 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2725
2726commit 7ecfab47eb221dbb996ea6c033348b8eceaeb893
2727Author: Eric Anholt <eric@anholt.net>
2728Date: Tue Oct 22 14:22:04 2013 -0700
2729
2730 glx: Add support for the new DRI loader entrypoint.
2731
2732 This is going to be exposed (and not the old entrypoint) for some DRI
2733 drivers once the megadrivers series lands, and the plan is to
2734 eventually transition all drivers to that. Hopefully this is
2735 unobtrusive enough to merge to stable X servers so that they can be
2736 compatible with new Mesa versions.
2737
2738 v2: typo fix in the comment
2739
2740 Signed-off-by: Eric Anholt <eric@anholt.net>
2741 Reviewed-by: Adam Jackson <ajax@redhat.com>
2742 Signed-off-by: Keith Packard <keithp@keithp.com>
2743
2744commit 83bb366e71ef41165998eed2934208f84714a37e
2745Author: Strake <strake888@gmail.com>
2746Date: Thu Oct 17 23:20:20 2013 -0500
2747
2748 unbreak KdEnableInput
2749
2750 pi = ki = 0 here, so not follow them.
2751
2752 Signed-off-by: Strake <strake888@gmail.com>
2753 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
2754 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2755
2756commit 8db4121a3c7104548fd4a138ac1d1bdea5fd22d9
2757Author: Keith Packard <keithp@keithp.com>
2758Date: Fri Oct 18 17:31:01 2013 -0700
2759
2760 Update to version 1.14.99.3
2761
2762 Signed-off-by: Keith Packard <keithp@keithp.com>
2763
2764commit e3220d7bae4548bf64e5879c981ad070ac1f78f7
2765Merge: 73b2660 2523a44
2766Author: Keith Packard <keithp@keithp.com>
2767Date: Fri Oct 18 17:22:52 2013 -0700
2768
2769 Merge commit '2523a445a09a75a8baf642608d099a5e12d5437f'
2770
2771 This is merging the commits along Peter's for-keith branch, without
2772 also merging in the spurious merge at the top of that branch.
2773
2774commit f8662a133690773692dfa392ecfa3b8b56af8fa9
2775Merge: 7cf1b59 2523a44
2776Author: Peter Hutterer <peter.hutterer@who-t.net>
2777Date: Fri Oct 18 16:50:12 2013 +1000
2778
2779 Merge branch 'xsync-transition-bug-59644' into for-keith
2780
2781commit 2523a445a09a75a8baf642608d099a5e12d5437f
2782Author: Peter Hutterer <peter.hutterer@who-t.net>
2783Date: Thu Oct 17 12:02:27 2013 +1000
2784
2785 sync: split updating and triggering a counter up
2786
2787 Now that the brackets are always the nearest change points (regardless of
2788 transition) we need to update the counters whenever we check for any updates.
2789
2790 Otherwise we end up with a situation where counter->value is out of date and
2791 an alarm doesn't trigger because we're still using the value from last time
2792 something actually triggered.
2793
2794 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2795 Reviewed-by: Adam Jackson <ajax@redhat.com>
2796 Reviewed-by: Keith Packard <keithp@keithp.com>
2797
2798commit e57ec99b03b2ad840c384a97ab2766ce9da0f5cc
2799Author: Peter Hutterer <peter.hutterer@who-t.net>
2800Date: Wed Oct 16 16:31:15 2013 +1000
2801
2802 sync: always set the brackets (#59644)
2803
2804 The current code sets bracket_greater to the first trigger after the current
2805 value, and bracket_less to the last trigger before the current value.
2806
2807 For example, the idle timer with three negative and three positive transitions
2808 would set this:
2809
2810 nt1 nt2 nt3
2811 |--------|------|--|------- idle --|---|--|-----> t
2812 pt1 pt2 pt3
2813 bracket_less == nt2
2814 bracket_greater == pt2
2815
2816 This is an optimization so we can skip code paths in the block/wakeup handlers
2817 if the current value doesn't meet any of the trigger requirements. Those
2818 handlers largely do a
2819 if (bracket_less is less than current value &&
2820 bracket_greater is greater than current value)
2821 return, nothing to do
2822
2823 However, unless the bracket values are updated at the correct time, the
2824 following may happen:
2825
2826 nt
2827 |--------------|---------- idle ------|--------> t
2828 pt
2829
2830 In this case, neither bracket is set, we're past the pos transition and not
2831 yet at the neg transition. idle may now go past nt, but the brackets are not
2832 updated. If idle is then reset to 0, no alarm is triggered for nt. Likewise,
2833 idle may now go past pt and no alarm is triggered.
2834
2835 Changing an alarm or triggering an alarm will re-calculate the brackets, so
2836 this bug is somewhat random. If any other client triggers an alarm when the
2837 brackets are wrongly NULL, the recalculation will set them this bug may not
2838 appear.
2839
2840 This patch changes the behavior, so that the brackets are always the nearest
2841 positive or negative transitions to the current counter value. In the example
2842 above, nt will trigger a wakeup and a re-calculation of the brackets, so that
2843 going past it in the negative direction will then cause the proper alarm
2844 triggers.
2845
2846 Or, in Keith's words:
2847
2848 Timer currently past a positive trigger
2849 No bracket values, because no trigger in range
2850
2851 Timer moves backwards before the positive trigger
2852 Brackets not reset, even though there is now a trigger in range
2853
2854 Timer moves forward past the positive trigger
2855 Trigger doesn't fire because brackets not set
2856
2857 Setting the LT bracket in this case will cause everything to get
2858 re-evaluated when the sync value moves backwards before the trigger
2859 value.
2860
2861 X.Org Bug 59644 <http://bugs.freedesktop.org/show_bug.cgi?id=59644>
2862
2863 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2864 Reviewed-by: Adam Jackson <ajax@redhat.com>
2865 Reviewed-by: Keith Packard <keithp@keithp.com>
2866
2867commit 06b87aa528d7a739ba20101a1f83b1a428691a01
2868Author: Peter Hutterer <peter.hutterer@who-t.net>
2869Date: Wed Oct 16 10:08:46 2013 +1000
2870
2871 sync: if the idle time was reset, force alarms to trigger (#70476)
2872
2873 The time between the idle reset and the IdleTimeWakeupHandler to be called is
2874 indeterminate. Clients with an PositiveTransition or NegativeTransition alarm
2875 on a low threshold may miss an alarm.
2876
2877 Work around this by keeping a reset flag for each device. When the
2878 WakeupHandler triggers and the reset flag is set, we force a re-calculation of
2879 everything and pretend the current idle time is zero. Immediately after is the
2880 next calculation with the real idle time.
2881
2882 Relatively reproducible test case: Set up a XSyncNegativeTransition alarm for
2883 a threshold of 1 ms. May trigger, may not.
2884
2885 X.Org Bug 70476 <http://bugs.freedesktop.org/show_bug.cgi?id=70476>
2886
2887 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2888 Reviewed-by: Adam Jackson <ajax@redhat.com>
2889 Reviewed-by: Keith Packard <keithp@keithp.com>
2890
2891commit efc1035ca958f2c9d266338a308518a0834b1773
2892Author: Peter Hutterer <peter.hutterer@who-t.net>
2893Date: Wed Oct 16 09:36:01 2013 +1000
2894
2895 dix: provide accessor methods for the last device event time
2896
2897 And now that we have the accessors, localize it. No functional changes, just
2898 preparing for a future change.
2899
2900 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2901 Reviewed-by: Adam Jackson <ajax@redhat.com>
2902 Reviewed-by: Keith Packard <keithp@keithp.com>
2903
2904commit b7c9bd9cf276e92a73be57ff2ed32b47a80f13fb
2905Author: Peter Hutterer <peter.hutterer@who-t.net>
2906Date: Wed Oct 16 09:21:47 2013 +1000
2907
2908 sync: supply the counter from IdleTimeBlockHandler
2909
2910 The main idletime counter has an initialized deviceid, might as well be
2911 supplying it properly. Without this, we'd only ever check the XIAllDevices
2912 counter, so the wait would never be adjusted for the device-specific triggers.
2913
2914 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2915 Reviewed-by: Adam Jackson <ajax@redhat.com>
2916 Reviewed-by: Keith Packard <keithp@keithp.com>
2917
2918commit 2efe49c1029f959fe80879bcf50df42e8b80451d
2919Author: Peter Hutterer <peter.hutterer@who-t.net>
2920Date: Wed Oct 16 13:01:01 2013 +1000
2921
2922 sync: always call BracketValues when recalculating upper/lower brackets
2923
2924 Both ServertimeBracketValues and IdleTimeBracketValues copy the value into
2925 there SysCounter privates. Call it for a NULL set as well, so we don't end up
2926 with stale pointers and we can remove the block/wakeup handlers.
2927
2928 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2929 Reviewed-by: Adam Jackson <ajax@redhat.com>
2930 Reviewed-by: Keith Packard <keithp@keithp.com>
2931
2932commit 5c7cfd4c6978834551848e1be62af240102e39b5
2933Author: Peter Hutterer <peter.hutterer@who-t.net>
2934Date: Tue Oct 15 10:11:20 2013 +1000
2935
2936 sync: compress two if statements
2937
2938 No functional changes, just merges a > and == condition into a >= condition.
2939
2940 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2941 Reviewed-by: Adam Jackson <ajax@redhat.com>
2942 Reviewed-by: Keith Packard <keithp@keithp.com>
2943
2944commit 73b2660d7273d175d279d22f8ca0c3932a14ff1c
2945Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2946Date: Mon Sep 16 21:47:16 2013 -0700
2947
2948 Avoid use-after-free in dix/dixfonts.c: doImageText() [CVE-2013-4396]
2949
2950 Save a pointer to the passed in closure structure before copying it
2951 and overwriting the *c pointer to point to our copy instead of the
2952 original. If we hit an error, once we free(c), reset c to point to
2953 the original structure before jumping to the cleanup code that
2954 references *c.
2955
2956 Since one of the errors being checked for is whether the server was
2957 able to malloc(c->nChars * itemSize), the client can potentially pass
2958 a number of characters chosen to cause the malloc to fail and the
2959 error path to be taken, resulting in the read from freed memory.
2960
2961 Since the memory is accessed almost immediately afterwards, and the
2962 X server is mostly single threaded, the odds of the free memory having
2963 invalid contents are low with most malloc implementations when not using
2964 memory debugging features, but some allocators will definitely overwrite
2965 the memory there, leading to a likely crash.
2966
2967 Reported-by: Pedro Ribeiro <pedrib@gmail.com>
2968 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2969 Reviewed-by: Julien Cristau <jcristau@debian.org>
2970
2971commit 8afe20d4e34adcfd29bdf43a01d55335ea2c5dba
2972Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2973Date: Mon Sep 16 21:45:29 2013 -0700
2974
2975 Update GLX dependencies now that DRI & DRI2 are builtins, not modules
2976
2977 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2978
2979commit 2704bdb24a2c7bac65b90e05f1a68438b34ecf58
2980Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2981Date: Tue Sep 3 22:53:28 2013 -0700
2982
2983 DMX glxproxy: Don't allocate & copy data just to free it unused
2984
2985 Two functions in the DMX glxproxy code loop over all the backend
2986 screens, starting at the highest numbered and counting down to
2987 the lowest.
2988
2989 Previously, for each screen, the code would allocate a buffer
2990 large enough to read the reply from the backend, copy that reply
2991 into the buffer, and then if it wasn't the final screen, free it.
2992 Only the buffer from the final screen is used, to pass on to the
2993 client in the reply.
2994
2995 This modifies it to just immediately discard the responses from
2996 the screens as we loop through it, only doing the allocate & copy
2997 work for the one buffer we pass back to the client.
2998
2999 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3000 Reviewed-by: Alex Deucher <aleander.deucher@amd.com>
3001
3002commit 6c06c268adbab63ebe7490489aa030724cbdc54b
3003Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3004Date: Sun Aug 18 18:02:49 2013 -0700
3005
3006 Skip damage calls if DamageCreate fails in exa functions
3007
3008 Fixes parfait errors such as:
3009 Null pointer dereference (CWE 476): Write to null pointer pDamage
3010 at line 1833 of miext/damage/damage.c in function 'DamageRegister'.
3011 Function DamageCreate may return constant 'NULL' at line 1775,
3012 called at line 232 of exa/exa_migration_mixed.c
3013 in function 'exaPrepareAccessReg_mixed'.
3014 Constant 'NULL' passed into function DamageRegister,
3015 argument pDamage, from call at line 237.
3016 Null pointer introduced at line 1775 of miext/damage/damage.c
3017 in function 'DamageCreate'.
3018 Null pointer dereference (CWE 476): Write to null pointer pDamage
3019 at line 1833 of miext/damage/damage.c in function 'DamageRegister'.
3020 Function DamageCreate may return constant 'NULL' at line 1775,
3021 called at line 104 of exa/exa_mixed.c
3022 in function 'exaCreatePixmap_mixed'.
3023 Constant 'NULL' passed into function DamageRegister,
3024 argument pDamage, from call at line 109.
3025 Null pointer introduced at line 1775 of miext/damage/damage.c
3026 in function 'DamageCreate'.
3027
3028 Checks are similar to handling results of other calls to DamageCreate.
3029
3030 [ This bug was found by the Parfait 1.3.0 bug checking tool.
3031 http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ]
3032
3033 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3034
3035commit 7cf1b595c8c8f9776a39559d2878cf90af3f2859
3036Author: Peter Hutterer <peter.hutterer@who-t.net>
3037Date: Wed Sep 4 15:34:29 2013 +1000
3038
3039 dix: only deliver for the current grab type
3040
3041 Use the grabtype to determine which type of event to send - all other events
3042 are pointless and may result in erroneous events being delivered.
3043
3044 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3045
3046commit 78944d62ffc7fec6f75b6f514ab7a139ba9bc77b
3047Author: Peter Hutterer <peter.hutterer@who-t.net>
3048Date: Wed Sep 4 15:34:28 2013 +1000
3049
3050 dix: don't attempt to deliver an event for a different grabtype
3051
3052 For an active grab, grab->eventMask can be either the core or the XI1 mask.
3053 With the overlap of event filters, calling DeliverOneGrabbedEvent(XI1) for a
3054 ProximityOut event will trigger if the client has selected for enter events -
3055 the filter is the same for both.
3056
3057 Thus, we end up delivering a proximity event to a client not expecting one.
3058
3059 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3060
3061commit 6159811a1aaf848016dbfa7bde68df097027870c
3062Author: Peter Hutterer <peter.hutterer@who-t.net>
3063Date: Wed Sep 4 15:34:27 2013 +1000
3064
3065 include: change grabtypes to start at 1
3066
3067 Avoid erroneous detection of an unset grabtype as CORE
3068
3069 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3070
3071commit be6ea80b79b3d6eb97106ea142e01acbdf059995
3072Author: Peter Hutterer <peter.hutterer@who-t.net>
3073Date: Wed Sep 4 15:34:26 2013 +1000
3074
3075 dix: only allow button and key events to freeze a sync'd pointer
3076
3077 If a client calls XAllowEvents(SyncPointer) it expects events as normal until
3078 the next button press or release event - that freezes the device. An e.g.
3079 proximity event must thus not freeze the pointer.
3080
3081 As per the spec, only button and key events may do so, so narrow it to these
3082 cases.
3083
3084 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3085
3086commit 93a27b2dd0d7ed51120f42456a91a7c59902ebc2
3087Author: Michele Baldessari <michele@acksyn.org>
3088Date: Sun Sep 15 19:30:38 2013 +0100
3089
3090 Xephyr: restore cursor visibility
3091
3092 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69388
3093
3094 Commit c100211034ab69ce453a1644fb61c6808d7e3eda (dix: only show the cursor
3095 if a window defines one (#58398)) broke the default cursor behaviour in
3096 Xephyr (unless run with -retro). Restore the default cursor visibility
3097 so that '-retro' or '-host-cursor' are not needed to have a visible
3098 cursor.
3099
3100 Signed-off-by: Michele Baldessari <michele@acksyn.org>
3101 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3102
3103 as of ba387cf21f7d95987211f75d8024601e7d64e322 "ephyr: Use host (HW) cursors
3104 by default." this only applies if -sw-cursor is given on the cmdline.
3105
3106 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3107
3108commit 70efc799cbeec26f19831b3b14c0d4646198a077
3109Author: Michele Baldessari <michele@acksyn.org>
3110Date: Sun Sep 15 19:56:34 2013 +0100
3111
3112 Fix Xephyr compilation when DEBUG is enabled
3113
3114 When DEBUG is enabled Xephyr compilation fails:
3115 ephyrdriext.c:343:133: error: 'is_ok' undeclared (first use in this
3116 function)
3117 EPHYR_LOG("leave. is_ok:%d\n", is_ok);
3118
3119 Just reemove bogus is_ok variable.
3120
3121 Signed-off-by: Michele Baldessari <michele@acksyn.org>
3122 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3123 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3124
3125commit 5ac4bfca64bd75474e550b26bc1195a5ca245752
3126Author: Adam Jackson <ajax@redhat.com>
3127Date: Tue Sep 10 14:18:18 2013 -0400
3128
3129 input: calloc minimization for xi2mask_new
3130
3131 There's no reason to do this as (nmasks + 2) callocs, and it's a
3132 surprisingly hot path. Turns out you hit this ~once per passive grab,
3133 and you do a few bajillion passive grab changes every time you enter or
3134 leave the overview in gnome-shell. According to a callgrind of Xorg
3135 with gnome-shell-perf-tool run against it:
3136
3137 Ir before: 721437275
3138 Ir after: 454227086
3139
3140 Signed-off-by: Adam Jackson <ajax@redhat.com>
3141 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
3142 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3143
3144commit 44d502c6f1d03a9e548f0941cde0228b6fde8394
3145Author: Jon TURNEY <jon.turney@dronecode.org.uk>
3146Date: Sun Oct 6 16:45:52 2013 +0100
3147
3148 hw/xwin: Fix for "glx: Remove screen number from __GLXconfig"
3149
3150 Fix compilation after commit c3c976f54c3c282d6fa6c8360688e036bc43d210 "glx:
3151 Remove screen number from __GLXconfig"
3152
3153 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
3154 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
3155 Reviewed-by: Adam Jackson <ajax@redhat.com>
3156
3157commit 6a9bd103cb993e873cb82664b677ceed73c0bd85
3158Merge: ea3b7db 8aae28e
3159Author: Keith Packard <keithp@keithp.com>
3160Date: Mon Oct 7 17:29:38 2013 -0700
3161
3162 Merge remote-tracking branch 'jeremyhu/master'
3163
3164commit ea3b7db4b747f883ef7b8b524ee13bf8a63add14
3165Author: Emil Velikov <emil.l.velikov@gmail.com>
3166Date: Mon Oct 7 18:29:25 2013 +0100
3167
3168 configure.ac: Include missing proto declarations in SDK_REQUIRED_MODULES
3169
3170 Commits a1d41e311c21e, 7d859bd87834d & 3ed2c6e11298c made extinit.h require
3171 the XF86 Big Font, XRes & ScrnSaver proto headers, but failed to add them
3172 to the SDK_REQUIRED_MODULES so pkg-config would find them for driver builds.
3173
3174 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
3175 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
3176 Signed-off-by: Keith Packard <keithp@keithp.com>
3177
3178commit 8aae28e3cbdd49bfa8d940a4a517c94a53b94f86
3179Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3180Date: Sun Oct 6 09:20:45 2013 -0700
3181
3182 XQuartz: Dont't set screen on __GLXConfig
3183
3184 Fixes regression introduced by c3c976f54c3c282d6fa6c8360688e036bc43d210
3185
3186 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3187
3188commit c5f8eb968efa970201712105a1ac880b517e7b6d
3189Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3190Date: Sun Oct 6 09:18:15 2013 -0700
3191
3192 rootless: Use miCopyRegion instead of fbCopyRegion
3193
3194 Fixes regression introduced by e657635dbe6b92875b0e88370557c2cbab673a49
3195
3196 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3197
3198commit ccbe17b1c6da1ad9d085fc8133cdd15dc7004a4a
3199Author: Pino Toscano <toscano.pino@tiscali.it>
3200Date: Wed Aug 28 17:15:03 2013 +0200
3201
3202 os: move <arpa/inet.h> for any !win32 system
3203
3204 It is needed in IPv6 configurations (for inet_pton) also when
3205 SIOCGIFCONF is not defined.
3206
3207 Signed-off-by: Pino Toscano <toscano.pino@tiscali.it>
3208 Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
3209 Signed-off-by: Julien Cristau <jcristau@debian.org>
3210
3211commit c079b8e675dd2762c9e07fe5d8c07612299dceb7
3212Author: Pino Toscano <toscano.pino@tiscali.it>
3213Date: Wed Aug 28 17:04:48 2013 +0200
3214
3215 xfree86/hurd: include <hurd.h>
3216
3217 Needed for using get_privileged_port.
3218
3219 Signed-off-by: Pino Toscano <toscano.pino@tiscali.it>
3220 Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
3221 Signed-off-by: Julien Cristau <jcristau@debian.org>
3222
3223commit f0659a7f68a5a334879af01a6a63742eab1bf5de
3224Author: Keith Packard <keithp@keithp.com>
3225Date: Fri Oct 4 16:44:46 2013 -0700
3226
3227 Bump version to 1.14.99.2
3228
3229 Signed-off-by: Keith Packard <keithp@keithp.com>
3230
3231commit f2d149b6d10edea7233d93c7e19f71decf11dc62
3232Author: Keith Packard <keithp@keithp.com>
3233Date: Fri Oct 4 16:43:41 2013 -0700
3234
3235 kdrive/ephyr: Remove duplicate ephyrExtensions and ephyrExtensionInit decls
3236
3237 These were duplicated when GLX support was re-added on two different branches.
3238
3239 Signed-off-by: Keith Packard <keithp@keithp.com>
3240
3241commit e0a678f0596f3547b98d74d5f0438b484b059458
3242Author: Gaetan Nadon <memsize@videotron.ca>
3243Date: Sat Sep 28 15:26:38 2013 -0400
3244
3245 xfree86: add a comment as to why the logdir is created
3246
3247 Without the logdir, the xserver will write the content of the log file on the
3248 terminal stating that it cannot be written and will stop.
3249
3250 Refer to https://bugs.freedesktop.org/show_bug.cgi?id=3889
3251
3252 Reviewed-By: Matt Dew <marcoz@osource.org>
3253 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3254 Signed-off-by: Keith Packard <keithp@keithp.com>
3255
3256commit 5bdbf2dba3f7da4a593b286020eab0c7857e940c
3257Author: Gaetan Nadon <memsize@videotron.ca>
3258Date: Sat Sep 28 15:26:37 2013 -0400
3259
3260 xfree86: Use $(MKDIR_P) for better code portability
3261
3262 Still true that we should not use the lower case $(mkdir_p) version.
3263 However, remove the 2005 comment as the MKDIR_P is widely used now.
3264
3265 Reviewed-By: Matt Dew <marcoz@osource.org>
3266 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3267 Signed-off-by: Keith Packard <keithp@keithp.com>
3268
3269commit b8cfb0dc8426754b477b561bc158b8a24c143ddf
3270Author: Gaetan Nadon <memsize@videotron.ca>
3271Date: Sat Sep 28 15:26:36 2013 -0400
3272
3273 Uninstall X link and CYGWIN libXorg.exe.a in local install targets
3274
3275 It is our duty to uninstall any files and/or directories that we installed
3276 through install-data-local and install-exec-hook.
3277
3278 Currently the X symbolic link to Xorg remains on disk after running
3279 make uninstall.
3280
3281 Note the exception for logdir which is usually shared by other modules.
3282
3283 Reviewed-By: Matt Dew <marcoz@osource.org>
3284 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3285 Signed-off-by: Keith Packard <keithp@keithp.com>
3286
3287commit 35a528e49269c3df75486cc606e51a8c1c083c06
3288Author: Gaetan Nadon <memsize@videotron.ca>
3289Date: Sat Sep 28 15:26:35 2013 -0400
3290
3291 Xorg binary: use install-exec-hook rather than install-exec-local
3292
3293 The former was explicitly designed to execute additional code after the binary
3294 has been installed. The latter can be executed in any order, hence it's
3295 current dependency on install-binPROGRAMS as a workaround.
3296
3297 The CYGWIN libXorg.exe.a target is an installation target rather than
3298 a post-installation one, so it should not be done as a hook. It does not depend
3299 on the Xorg executable being installed.
3300
3301 Automake:
3302 "These hooks are run after all other install rules of the appropriate type,
3303 exec or data, have completed. So, for instance, it is possible to perform
3304 post-installation modifications using an install hook".
3305
3306 "With the -local targets, there is no particular guarantee of execution order;
3307 typically, they are run early, but with parallel make, there is no way
3308 to be sure of that".
3309
3310 Reviewed-by: Eric Anholt <eric@anholt.net>
3311 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3312 Signed-off-by: Keith Packard <keithp@keithp.com>
3313
3314commit a1d87576a30008253eb0cc92b44c9718f6665062
3315Author: Gaetan Nadon <memsize@videotron.ca>
3316Date: Sat Sep 28 15:26:34 2013 -0400
3317
3318 Use $(LN_S) provided by AC_PROG_LN_S macro to create links
3319
3320 For better code portability.
3321
3322 Reviewed-By: Matt Dew <marcoz@osource.org>
3323 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3324 Signed-off-by: Keith Packard <keithp@keithp.com>
3325
3326commit 28c8e1910797970dbf1d40794f1804a8fcc812bc
3327Author: Gaetan Nadon <memsize@videotron.ca>
3328Date: Sat Sep 28 15:26:33 2013 -0400
3329
3330 The Xorg binary is missing the extension $(EXEEXT) in the makefile
3331
3332 This is not a problem on UNIX platforms, but on CYGWIN it creates a broken
3333 link to Xorg rather than a link to Xorg.exe.
3334
3335 From the CYGWIN log on tinderbox, we can see that the executable Xorg.exe is
3336 installed correctly. We can see the command used to create the link:
3337
3338 (cd /jhbuild/install/[...]/install/bin && rm -f X && ln -s Xorg X)
3339
3340 Note that the "relink" makefile target correctly appends $(EXEEXT) to Xorg.
3341
3342 Reviewed-By: Matt Dew <marcoz@osource.org>
3343 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3344 Signed-off-by: Keith Packard <keithp@keithp.com>
3345
3346commit f3b529bf25db2700addbafc7aa4fc5b126ba9929
3347Merge: 4399bd3 abc2bfc
3348Author: Keith Packard <keithp@keithp.com>
3349Date: Fri Oct 4 14:04:48 2013 -0700
3350
3351 Merge remote-tracking branch 'anholt/ephyr-fixes'
3352
3353commit 4399bd38326b81f7b92365346504d7c6354772a9
3354Author: Gaetan Nadon <memsize@videotron.ca>
3355Date: Mon Sep 30 13:30:46 2013 -0400
3356
3357 miext/shadow: missing c2p_core.h breaks "make distcheck" target.
3358
3359 Reviewed-by: Julien Cristau <jcristau@debian.org>
3360 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3361 Signed-off-by: Keith Packard <keithp@keithp.com>
3362
3363commit 9ccb4a63982942fdce19609d2cadbd8e51708462
3364Merge: 5d2ec69 da5e201
3365Author: Keith Packard <keithp@keithp.com>
3366Date: Fri Oct 4 13:54:55 2013 -0700
3367
3368 Merge remote-tracking branch 'whot/for-keith'
3369
3370commit 5d2ec6933f6bb5116666ca1bfb07f0f20bf2a0b5
3371Merge: 7d3d4ae 6ee4d9f
3372Author: Keith Packard <keithp@keithp.com>
3373Date: Fri Oct 4 13:50:04 2013 -0700
3374
3375 Merge remote-tracking branch 'ajax/xserver-next'
3376
3377commit da5e20127a5b6e11a47eb6e61f3ea87b056cdf5f
3378Author: Gaetan Nadon <memsize@videotron.ca>
3379Date: Thu Sep 19 19:13:35 2013 -0400
3380
3381 test: add new os executable to .gitignore
3382
3383 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
3384 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3385 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3386
3387commit c73c36b537f996574628e69681833ea37dec2b6e
3388Author: Laércio de Sousa <lbsousajr@gmail.com>
3389Date: Wed Sep 18 13:42:17 2013 -0300
3390
3391 xserver: enable InputClass option "GrabDevice" by default for non-seat0 seats (#69478)
3392
3393 This patch contributes to fill the remaining gaps which make
3394 systemd-multi-seat-x wrapper still necessary in some multiseat setups.
3395
3396 This also replaces previous evdev patch that does the same thing
3397 for that particular driver.
3398
3399 When option "-seat" is passed with an argument different from "seat0",
3400 option "GrabDevice" for input devices is enabled by default
3401 (no need of enabling it in xorg.conf's "InputClass" section).
3402
3403 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69478
3404
3405 Signed-off-by: Laércio de Sousa <lbsousajr@gmail.com>
3406 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3407 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3408
3409commit 160c2db32d0b9ee5316a44179a6af4f4722e3d0a
3410Author: Peter Hutterer <peter.hutterer@who-t.net>
3411Date: Wed Sep 18 16:04:44 2013 +1000
3412
3413 dmx: provide enough space for axis mappings
3414
3415 relmap/absmap is used as a evdev-axis-to-x-axis mapping. ABS_X maps to
3416 axis 0, ABS_Y to 1, etc. skipping over non-existing axes so that the third bit
3417 set in the ABS_* range is axis 2, and so on. This requires us to actually have
3418 enough space to have all the ABS_*/REL_* range.
3419
3420 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3421 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3422
3423commit 7d3d4ae55dd6ee338439e2424ac423b1df80501b
3424Author: Keith Packard <keithp@keithp.com>
3425Date: Fri Sep 20 14:43:41 2013 -0500
3426
3427 damage: Must translate initial window damage by window offset
3428
3429 Damage is reported relative to the drawable origin, but the window
3430 borderClip is absolute. Translate the region by the window position
3431 before reporting damage to adjust.
3432
3433 Reported-by: Adam Jackson <ajax@redhat.com>
3434 Signed-off-by: Keith Packard <keithp@keithp.com>
3435 Reviewed-by: Adam Jackson <ajax@redhat.com>
3436
3437commit 8010d3a48bd0b224dcb0883e39c2351ad364d846
3438Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3439Date: Sun Sep 15 08:10:33 2013 -0700
3440
3441 XQuartz: pbproxy: Fix build with -DDEBUG
3442
3443 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3444
3445commit 39c548da0cf0522ad246630605b96751ab6d39c0
3446Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3447Date: Sun Sep 15 08:07:44 2013 -0700
3448
3449 XQuartz: Fix build with moved pseudoramiX
3450
3451 Regression from: e716baedc4d8e52a60f43ef21aba771b340d8c8b
3452
3453 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
3454 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
3455
3456commit 6ee4d9f94ae2cc27c39381d7554134d9b4807a65
3457Author: Adam Jackson <ajax@redhat.com>
3458Date: Tue Jul 9 11:19:11 2013 -0400
3459
3460 glx: Fill in some missing attributes from DoGetFBConfigs
3461
3462 Reviewed-by: Eric Anholt <eric@anholt.net>
3463 Signed-off-by: Adam Jackson <ajax@redhat.com>
3464
3465commit b257fabff00ae925bb7e70364dd1d738634087c9
3466Author: Adam Jackson <ajax@redhat.com>
3467Date: Thu Jul 4 10:31:40 2013 -0400
3468
3469 glx: Remove a dead comment
3470
3471 Reviewed-by: Eric Anholt <eric@anholt.net>
3472 Signed-off-by: Adam Jackson <ajax@redhat.com>
3473
3474commit abd086502123086ce3d1fba25c159bb022d2750f
3475Author: Adam Jackson <ajax@redhat.com>
3476Date: Mon Jul 8 17:01:13 2013 -0400
3477
3478 glx: Catch another failure case in drawable creation
3479
3480 Reviewed-by: Eric Anholt <eric@anholt.net>
3481 Signed-off-by: Adam Jackson <ajax@redhat.com>
3482
3483commit c3c976f54c3c282d6fa6c8360688e036bc43d210
3484Author: Adam Jackson <ajax@redhat.com>
3485Date: Mon Jul 8 12:28:56 2013 -0400
3486
3487 glx: Remove screen number from __GLXconfig
3488
3489 Not used. There's no real reason to match against this instead of
3490 matching against fbconfig or visual ID anyway.
3491
3492 Reviewed-by: Eric Anholt <eric@anholt.net>
3493 Signed-off-by: Adam Jackson <ajax@redhat.com>
3494
3495commit 1d1484e9bd55d1fa8316b1574754f28d5d31a076
3496Author: Adam Jackson <ajax@redhat.com>
3497Date: Mon Jul 8 12:09:13 2013 -0400
3498
3499 glx: Remove pixmapMode from __GLXconfig
3500
3501 This has never been filled in with anything meaningful afaict, and you
3502 can't get to it from the client in any event.
3503
3504 Reviewed-by: Eric Anholt <eric@anholt.net>
3505 Signed-off-by: Adam Jackson <ajax@redhat.com>
3506
3507commit 34e6e60105796ad015469c10921c5ff09acf7eb0
3508Author: Adam Jackson <ajax@redhat.com>
3509Date: Tue Jul 2 12:33:36 2013 -0400
3510
3511 glx: Remove support for NV_vertex_program and NV_fragment_program
3512
3513 Mesa doesn't implement these anymore, never really did outside of swrast
3514 anyway.
3515
3516 Reviewed-by: Eric Anholt <eric@anholt.net>
3517 Signed-off-by: Adam Jackson <ajax@redhat.com>
3518
3519commit acf14c1de766c4ea206cd7ce7f25b780a589d33c
3520Author: Adam Jackson <ajax@redhat.com>
3521Date: Tue Jul 2 10:35:19 2013 -0400
3522
3523 glx: realloc style fix in RenderLarge
3524
3525 Reviewed-by: Eric Anholt <eric@anholt.net>
3526 Signed-off-by: Adam Jackson <ajax@redhat.com>
3527
3528commit 9ebf739a6864c9ec38bf72f63ef2e3b9cd1951db
3529Author: Adam Jackson <ajax@redhat.com>
3530Date: Tue Jul 2 10:30:40 2013 -0400
3531
3532 glx: Eliminate a small malloc from QueryContext
3533
3534 No reason to have that be a failure path.
3535
3536 Reviewed-by: Eric Anholt <eric@anholt.net>
3537 Signed-off-by: Adam Jackson <ajax@redhat.com>
3538
3539commit b99f7975407f111b99d772cd28224d7dc1b34fe4
3540Author: Adam Jackson <ajax@redhat.com>
3541Date: Fri Nov 9 18:05:27 2012 -0500
3542
3543 glx: Handle failure to create the pixmap backing the pbuffer
3544
3545 We happen not to sanitize the width/height we pass to CreatePixmap here,
3546 oops. It's not exploitable, but it's certainly a crash, so let's just
3547 throw BadAlloc instead.
3548
3549 Reviewed-by: Eric Anholt <eric@anholt.net>
3550 Signed-off-by: Adam Jackson <ajax@redhat.com>
3551
3552commit 22fbfdcb31f335c22053a0e29f0e73c03cd70a76
3553Author: Adam Jackson <ajax@redhat.com>
3554Date: Fri Nov 9 17:52:12 2012 -0500
3555
3556 glx: Implement GLX_PRESERVED_CONTENTS drawable attribute
3557
3558 We back pixmaps with pbuffers so they're never actually clobbered. Say
3559 so when asked.
3560
3561 Reviewed-by: Eric Anholt <eric@anholt.net>
3562 Signed-off-by: Adam Jackson <ajax@redhat.com>
3563
3564commit 2e20b8382cdf37ae5a2802f85b8f06a370f9645d
3565Author: Adam Jackson <ajax@redhat.com>
3566Date: Fri Nov 9 17:12:49 2012 -0500
3567
3568 glx: Implement GLX_FBCONFIG_ID in GetDrawableAttributes
3569
3570 Required by GLX 1.4, section 3.3.6, "Querying Attributes".
3571
3572 Reviewed-by: Eric Anholt <eric@anholt.net>
3573 Signed-off-by: Adam Jackson <ajax@redhat.com>
3574
3575commit 0d76191baeee005982754ac18ca995742e32d8c1
3576Author: Adam Jackson <ajax@redhat.com>
3577Date: Fri Nov 9 17:10:57 2012 -0500
3578
3579 glx: Implement GLX_{WIDTH,HEIGHT} in GetDrawableAttributes
3580
3581 Required by GLX 1.4, section 3.3.6, "Querying Attributes".
3582
3583 Reviewed-by: Eric Anholt <eric@anholt.net>
3584 Signed-off-by: Adam Jackson <ajax@redhat.com>
3585
3586commit d11f13e383dda1e9cabe32793d261d46edc44418
3587Author: Adam Jackson <ajax@redhat.com>
3588Date: Fri Nov 9 17:07:20 2012 -0500
3589
3590 glx: Compute number of attributes in GetDrawableAttributes on the fly
3591
3592 This doesn't have any effect yet, but is needed to properly build the
3593 reply for pbuffers.
3594
3595 Reviewed-by: Eric Anholt <eric@anholt.net>
3596 Signed-off-by: Adam Jackson <ajax@redhat.com>
3597
3598commit 468b57324f838144d79e3a9d85a0bde853578c4b
3599Author: Adam Jackson <ajax@redhat.com>
3600Date: Fri Nov 9 16:48:16 2012 -0500
3601
3602 glx: Style fixes
3603
3604 Reviewed-by: Eric Anholt <eric@anholt.net>
3605 Signed-off-by: Adam Jackson <ajax@redhat.com>
3606
3607commit 2b181ad0952851f47f23dcb1a36471685313c13e
3608Author: Adam Jackson <ajax@redhat.com>
3609Date: Fri Nov 9 16:33:54 2012 -0500
3610
3611 glx: Pull GLX vendor string out of __GLXscreen
3612
3613 Given how we're currently implementing GLX this can't meaningfully vary
3614 per-screen.
3615
3616 Reviewed-by: Eric Anholt <eric@anholt.net>
3617 Signed-off-by: Adam Jackson <ajax@redhat.com>
3618
3619commit b9c489cf13e1ad35b7d9bea9de8d85e929374d17
3620Author: Adam Jackson <ajax@redhat.com>
3621Date: Wed Oct 31 10:05:24 2012 -0400
3622
3623 glx: Remove unused bits from the context struct
3624
3625 Reviewed-by: Eric Anholt <eric@anholt.net>
3626 Signed-off-by: Adam Jackson <ajax@redhat.com>
3627
3628commit f579cc85c29f52b575c62e6ae1e16eb87c506bd3
3629Author: Adam Jackson <ajax@redhat.com>
3630Date: Thu Jul 4 10:08:40 2013 -0400
3631
3632 glxproxy: Don't track GLCLientm{aj,in}orVersion
3633
3634 Basically just a port of 62f06b0d to glxproxy.
3635
3636 Reviewed-by: Eric Anholt <eric@anholt.net>
3637 Signed-off-by: Adam Jackson <ajax@redhat.com>
3638
3639commit e549ffb5f19cf044bc3e2d368a6f081f8b84bfaa
3640Author: Adam Jackson <ajax@redhat.com>
3641Date: Tue Oct 30 18:20:43 2012 -0400
3642
3643 dmx/glx: Remove unused __glXNop
3644
3645 Reviewed-by: Eric Anholt <eric@anholt.net>
3646 Signed-off-by: Adam Jackson <ajax@redhat.com>
3647
3648commit d962b8ed272d4a8921c0dadf16cd926484e30d10
3649Author: Adam Jackson <ajax@redhat.com>
3650Date: Thu Aug 22 16:42:23 2013 -0400
3651
3652 damageext: Style fix
3653
3654 Reviewed-by: Eric Anholt <eric@anholt.net>
3655 Signed-off-by: Adam Jackson <ajax@redhat.com>
3656
3657commit c90add0f1ce0c85aea7a60635de765848d778266
3658Author: Adam Jackson <ajax@redhat.com>
3659Date: Wed Aug 21 11:57:24 2013 -0400
3660
3661 damage: Remove "post-rendering" hooks
3662
3663 This is a revert of 974db58f5b730c3770ee461665a02dd4334d1dea. There are
3664 no consumers of this API, we must not need it.
3665
3666 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
3667 Reviewed-by: Eric Anholt <eric@anholt.net>
3668 Signed-off-by: Adam Jackson <ajax@redhat.com>
3669
3670commit 0f3f8e4295d4c05f33b93e0ffd2f6f8cd5759bf7
3671Author: Adam Jackson <ajax@redhat.com>
3672Date: Wed Aug 21 11:13:11 2013 -0400
3673
3674 damage: Don't rewrite Text ops to GlyphBlt ops
3675
3676 There's no particularly good reason to, and it breaks Xnest.
3677
3678 Bugzilla: http://bugs.freedesktop.org/2454
3679 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
3680 Reviewed-by: Eric Anholt <eric@anholt.net>
3681 Signed-off-by: Adam Jackson <ajax@redhat.com>
3682
3683commit d08966227e7d567df8d26eebc80f35f886e59a4a
3684Author: Adam Jackson <ajax@redhat.com>
3685Date: Mon Aug 26 14:03:51 2013 -0400
3686
3687 damage: Simplify DamageUnregister
3688
3689 You can only register one drawable on a given damage, so there's no
3690 reason to require the caller to specify the drawable, the damage is
3691 enough. The implementation would do something fairly horrible if you
3692 _did_ pass mismatched drawable and damage, so let's avoid the problem
3693 entirely.
3694
3695 v2: Simplify xf86RotateDestroy even more [anholt]
3696
3697 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
3698 Reviewed-by: Eric Anholt <eric@anholt.net>
3699 Signed-off-by: Adam Jackson <ajax@redhat.com>
3700
3701commit 28708a045de7d9043d20fb06b61c44a46eb5526b
3702Author: Adam Jackson <ajax@redhat.com>
3703Date: Mon Aug 26 13:52:14 2013 -0400
3704
3705 damage: Implicitly unregister on destroy
3706
3707 There's no reason not to, and it simplifies quite a few callers.
3708
3709 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
3710 Reviewed-by: Eric Anholt <eric@anholt.net>
3711 Signed-off-by: Adam Jackson <ajax@redhat.com>
3712
3713commit e657635dbe6b92875b0e88370557c2cbab673a49
3714Author: Adam Jackson <ajax@redhat.com>
3715Date: Thu Oct 18 14:24:23 2012 -0400
3716
3717 fb: Remove unused compatibility wrappers
3718
3719 Originally added in early 2009, not being used elsewhere anymore.
3720
3721 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3722 Signed-off-by: Adam Jackson <ajax@redhat.com>
3723
3724commit e4b15125bb06c97f1f34f7074fcdb597e718399c
3725Author: Adam Jackson <ajax@redhat.com>
3726Date: Thu Oct 18 14:24:22 2012 -0400
3727
3728 mipointer: Remove deprecated miPointerCurrentScreen
3729
3730 The only remaining use was in some debugging code in DMX.
3731
3732 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3733 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
3734 Signed-off-by: Adam Jackson <ajax@redhat.com>
3735
3736commit b5d66b9513a0f1e13f5c5bc9e0958f4eac9bf6c3
3737Author: Adam Jackson <ajax@redhat.com>
3738Date: Thu Oct 18 14:24:21 2012 -0400
3739
3740 xfree86: Remove deprecated unimplemented xf86MapReadSideEffects
3741
3742 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3743 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
3744 Signed-off-by: Adam Jackson <ajax@redhat.com>
3745
3746commit abbd85742a390e18497b96902a0c0d976739e3bd
3747Author: Adam Jackson <ajax@redhat.com>
3748Date: Thu Oct 18 14:24:20 2012 -0400
3749
3750 dix: FIXES is not optional
3751
3752 It's already not optional at configure time, this just makes it so at
3753 build time too.
3754
3755 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3756 Signed-off-by: Adam Jackson <ajax@redhat.com>
3757
3758commit dff81687f5eac3eac9b49f58d4654cc30add2547
3759Author: Adam Jackson <ajax@redhat.com>
3760Date: Thu Oct 18 14:24:19 2012 -0400
3761
3762 vbe: Don't try to load the ddc submodule
3763
3764 DDC is built into the server now.
3765
3766 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3767 Signed-off-by: Adam Jackson <ajax@redhat.com>
3768
3769commit bb82191bcfa8042f8d98ec8b1c6616dc31d2c5c8
3770Author: Adam Jackson <ajax@redhat.com>
3771Date: Thu Oct 18 14:24:18 2012 -0400
3772
3773 misprite: Delete some dead private storage
3774
3775 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3776 Signed-off-by: Adam Jackson <ajax@redhat.com>
3777
3778commit 3639ab4ce25dc82a3f8592b7322198f1d2aca5f9
3779Author: Adam Jackson <ajax@redhat.com>
3780Date: Tue Oct 16 14:56:25 2012 -0400
3781
3782 dri2: Don't bother with xf86LoaderCheckSymbol("DRI2Connect")
3783
3784 The DRI2 code is now built-in to the server, even for Xorg. The only
3785 thing this could protect against is trying to run a libglx built with
3786 DRI2 support against an Xorg built without it, which is firmly in
3787 "doctor it hurts when I do this" territory.
3788
3789 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3790 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
3791 Signed-off-by: Adam Jackson <ajax@redhat.com>
3792
3793commit 13b507409f9e01adebba50e7eb801b52e52692ba
3794Author: Adam Jackson <ajax@redhat.com>
3795Date: Tue Oct 16 14:56:23 2012 -0400
3796
3797 list: Avoid using X types
3798
3799 In particular, Bool. This is not an ABI break:
3800
3801 /usr/include/X11/Xdefs.h:typedef int Bool;
3802
3803 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3804 Signed-off-by: Adam Jackson <ajax@redhat.com>
3805
3806commit dbe10ef0e35659ee0e5db73f29670a3894a3e3f5
3807Author: Adam Jackson <ajax@redhat.com>
3808Date: Tue Oct 16 14:56:22 2012 -0400
3809
3810 dbe: Fold the window private private into the window private
3811
3812 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3813 Signed-off-by: Adam Jackson <ajax@redhat.com>
3814
3815commit 1489637d567b0034a916f88079b5f360e7fce99c
3816Author: Adam Jackson <ajax@redhat.com>
3817Date: Tue Oct 16 14:56:21 2012 -0400
3818
3819 dbe: Fold midbe reset into dix
3820
3821 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3822 Signed-off-by: Adam Jackson <ajax@redhat.com>
3823
3824commit 44d770750069486a8452fb2384f9b85745268e61
3825Author: Adam Jackson <ajax@redhat.com>
3826Date: Tue Oct 16 14:56:20 2012 -0400
3827
3828 dbe: Drop the unused idiom support
3829
3830 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3831 Signed-off-by: Adam Jackson <ajax@redhat.com>
3832
3833commit 8cce0cf4d378f6fa52bc2cd277737b8986ab683c
3834Author: Adam Jackson <ajax@redhat.com>
3835Date: Tue Oct 16 14:56:19 2012 -0400
3836
3837 damage: Remove the 'damage window' resource type.
3838
3839 Note that the existing code never actually creates a resource of this
3840 type, so it can't possibly do anything. This is clearly meant to track
3841 window destruction and clean up any associated Damages, but that's
3842 already handled by miext/damage's DestroyWindow wrapper.
3843
3844 Previous discussion:
3845 http://lists.freedesktop.org/archives/xorg-devel/2011-March/020847.html
3846
3847 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3848 Signed-off-by: Adam Jackson <ajax@redhat.com>
3849
3850commit e674815e0a217ac01ec2d6a8517f8466b27fb39f
3851Author: Adam Jackson <ajax@redhat.com>
3852Date: Tue Oct 16 14:56:18 2012 -0400
3853
3854 dmx: Remove ShadowFB support
3855
3856 This has been listed as deprecated ever since DMX was merged.
3857
3858 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3859 Signed-off-by: Adam Jackson <ajax@redhat.com>
3860
3861commit 19d3a7cb5db3188210830efa746987ba50d839e9
3862Author: Adam Jackson <ajax@redhat.com>
3863Date: Tue Oct 16 14:56:17 2012 -0400
3864
3865 dix: Remove a dead parameter from DoGetImage
3866
3867 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3868 Signed-off-by: Adam Jackson <ajax@redhat.com>
3869
3870commit 4dca026880da90cff9734cf5787c5b57c02e7310
3871Author: Adam Jackson <ajax@redhat.com>
3872Date: Tue Oct 16 14:56:16 2012 -0400
3873
3874 mipointer: Remove EnqueueEvent from miPointerScreenFuncRec
3875
3876 No DDX overrode this, and we never actually called through that slot
3877 anyway.
3878
3879 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3880 Signed-off-by: Adam Jackson <ajax@redhat.com>
3881
3882commit ad076dc6e82019defa422500710b3b42c6309e39
3883Author: Adam Jackson <ajax@redhat.com>
3884Date: Tue Oct 16 14:56:15 2012 -0400
3885
3886 mipointer: Flatten calls to mieqSwitchScreen
3887
3888 No DDX was overriding this.
3889
3890 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3891 Signed-off-by: Adam Jackson <ajax@redhat.com>
3892
3893commit e2b14a1d3bb27e86ff1d7aaf59af850d06263ef5
3894Author: Adam Jackson <ajax@redhat.com>
3895Date: Tue Oct 16 14:56:14 2012 -0400
3896
3897 os: Factor out some common code in input buffer handling
3898
3899 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
3900 Signed-off-by: Adam Jackson <ajax@redhat.com>
3901
3902commit 65fca558848acd907483d3c85c7cd4db5560f1b4
3903Author: Radek Doulik <rodo@novell.com>
3904Date: Tue Aug 13 08:45:47 2013 +0200
3905
3906 xnest: Ignore GetImage() error in xnestGetImage()
3907
3908 When an Xnest instance is not viewable it will crash when a client in
3909 that instance calls GetImage. This is because the Xnest server will
3910 itself receives a BadMatch error.
3911 This patch ignores the error. The application which has requested the
3912 image will receive garbage - this however is fully legal according
3913 to the specs as obscured areas will always contain garbage if there
3914 isn't some sort of backing store as discussed in
3915 https://bugs.freedesktop.org/show_bug.cgi?id=9488
3916 The applied patch is a version from Dadek Doulik.
3917
3918 v2: Call XSync() before changing error handlers as suggested by
3919 Daniel Stone <daniel@fooishbar.org>.
3920
3921 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
3922 Signed-off-by: Egbert Eich <eich@freedesktop.org>
3923
3924commit b902c8abb6d03e68c93e13881a350523b5ac900c
3925Author: Michal Srb <msrb@novell.com>
3926Date: Tue Aug 6 16:56:53 2013 +0200
3927
3928 Xnest: Implement xnestModifyPixmapHeader
3929
3930 Xnest variant of ModifyPixmapHeader that creates new Pixmap in parent X
3931 server if it's size is modified from 0x0 to anything bigger.
3932
3933 xnestCreatePixmap doesn't create pixmap in parent X server if it has
3934 dimensions 0x0. If it is later resized and accessed, Xnest will be
3935 aborted with BadDrawable error from parent X server because it will
3936 use XID 0. This happens with ScratchPixmap, for example as used from
3937 XaceCensorImage. Applications using XACE crash Xnest.
3938
3939 Reviewed-by: Adam Jackson <ajax@redhat.com>
3940
3941commit 132507eba93e4dfd466d5c30c9b818cdc6f8536e
3942Author: Peter Harris <pharris@opentext.com>
3943Date: Mon Jul 15 19:44:45 2013 -0400
3944
3945 security: Don't count RetainPermanent clients twice
3946
3947 If a RetainPermanent client is subsequently killed by a KillClient
3948 request, the reference count is decremented twice. This can cause the
3949 server to prematurely kill other clients using the same Authorization.
3950
3951 Reviewed-by: Adam Jackson <ajax@redhat.com>
3952 Signed-off-by: Peter Harris <pharris@opentext.com>
3953
3954commit 20c2a3bcb11b5baf564e2c73a477ba23f5ae2b10
3955Author: Peter Harris <pharris@opentext.com>
3956Date: Mon Jul 15 19:44:29 2013 -0400
3957
3958 mi: Avoid stack smash when drawing dashed lines
3959
3960 X.org Bug 54013 <https://bugs.freedesktop.org/show_bug.cgi?id=54013>
3961
3962 Reviewed-by: Adam Jackson <ajax@redhat.com>
3963 Signed-off-by: Peter Harris <pharris@opentext.com>
3964
3965commit 9bf46610a9d20962854016032de4567974e87957
3966Author: Chris Wilson <chris@chris-wilson.co.uk>
3967Date: Fri Jun 21 22:58:31 2013 +0100
3968
3969 os: Immediately queue initial WriteToClient
3970
3971 If we immediately put the WriteToClient() buffer into the socket's write
3972 queue, not only do we benefit from sending the response back to client
3973 earlier, but we also avoid the overhead of copying the data into our own
3974 staging buffer and causing extra work in the next select(). The write is
3975 effectively free as typically we may only send one reply per client per
3976 select() call, so the cost of the FlushClient() is the same.
3977
3978 shmget10: 26400 -> 110000
3979 getimage10: 25000 -> 108000
3980
3981 shmget500: 3160 -> 13500
3982 getimage500: 1000 -> 1010
3983
3984 The knock-on effect is that on a mostly idle composited desktop, the CPU
3985 overhead is dominated by the memmove in WriteToClient, which is in turn
3986 eliminated by this patch.
3987
3988 Reviewed-by: Adam Jackson <ajax@redhat.com>
3989 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
3990
3991commit ceffb34774d44ada83cc1994d20c8d65b46bd555
3992Author: Laurent Carlier <lordheavym@gmail.com>
3993Date: Sat Jun 8 00:43:56 2013 +0200
3994
3995 ephyr: restore GLX support
3996
3997 It was removed since version 1.13
3998
3999 Signed-off-by: Sebastien Bacher <seb128@ubuntu.com>
4000 Acked-by: Daniel Stone <daniel@fooishbar.org>
4001
4002commit f37972f4a67b49ba9649310a6778bd1bf31e0130
4003Author: Keith Packard <keithp@keithp.com>
4004Date: Wed Apr 24 14:58:27 2013 -0700
4005
4006 miext/sync: Fix multi-screen support in SyncFence initialization
4007
4008 miSyncSetup was checking to see if the screen private key had been
4009 registered, and if so, skipping the setup of the provided screen.
4010
4011 Instead, it should register the private index only once, but then
4012 initialize the screen unless it has already been initialized. This
4013 latter step allows drivers to initialize the sync private structures
4014 before the Sync extension itself is initialized.
4015
4016 Reviewed-by: Adam Jackson <ajax@redhat.com>
4017 Signed-off-by: Keith Packard <keithp@keithp.com>
4018
4019commit 1c8beedfe4343377fa3504f8ab71430cc137b22c
4020Author: Adam Jackson <ajax@redhat.com>
4021Date: Wed Aug 21 14:12:52 2013 -0400
4022
4023 randr: Fix a copypasta bug in CRTC confinement
4024
4025 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
4026 Signed-off-by: Adam Jackson <ajax@redhat.com>
4027
4028commit 47ff382d1fce25a8b097d45b79489e891f1f1228
4029Merge: 6f89ae3 3752a81
4030Author: Keith Packard <keithp@keithp.com>
4031Date: Mon Sep 9 15:13:42 2013 -0700
4032
4033 Merge remote-tracking branch 'jturney/master'
4034
4035commit 6f89ae3e64c4dfeea508813e546c10ba1da3ea8e
4036Author: Thomas Klausner <wiz@NetBSD.org>
4037Date: Wed Sep 4 20:06:07 2013 +0200
4038
4039 Fix typo in configure warning.
4040
4041 Signed-off-by: Thomas Klausner <wiz@NetBSD.org>
4042 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
4043 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4044
4045commit 47218a6e09549781fd61dbf5e0d3d5c81da64323
4046Author: Thomas Klausner <wiz@NetBSD.org>
4047Date: Wed Sep 4 20:05:51 2013 +0200
4048
4049 Fix bug in cursor handling.
4050
4051 CreateCursor (Xlib call XCreatePixmapCursor) with a non-bitmap
4052 source pixmap and a None mask is supposed to error out with BadMatch,
4053 but didn't.
4054
4055 From der Mouse <mouse@Rodents-Montreal.ORG>, changed following
4056 comments by Alan Coopersmith <alan.coopersmith@oracle.com>.
4057
4058 Signed-off-by: Thomas Klausner <wiz@NetBSD.org>
4059 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
4060 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4061
4062commit 1110b71e360195aab040d835b54540ab558638c5
4063Author: Chris Clayton <chris2553@googlemail.com>
4064Date: Wed Sep 4 15:42:04 2013 +1000
4065
4066 kdrive: fix build error on gcc 4.8 for out-of-bounds array access
4067
4068 I'm getting a error building xorg-server-1.14.1.902 with thelatest snapshot
4069 of gcc-4.8:
4070
4071 input.c:225:43: error: array subscript is above array bounds
4072 [-Werror=array-bounds]
4073
4074 This is because kdNumInputFds can become equal to KD_MAX_INPUT_FDS in
4075 KdRegisterFd(). This means that in KdUnregisterFd(), kdInputFds[j + 1] can
4076 be beyond the end of the array.
4077
4078 Signed-off-by: Chris Clayton <chris2553@googlemail.com>
4079 Reviewed-by: Keith Packard <keithp@keithp.com>
4080 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4081
4082commit abc2bfca16adcd1e5657d4ce54f4e1335d5ed53f
4083Author: Sebastien Bacher <seb128@ubuntu.com>
4084Date: Fri Aug 23 21:21:44 2013 +0200
4085
4086 kdrive: initialize GLX for xephyr
4087
4088 Like commit ac1a60e7b6f06fd075cc5bf55d6bc67206a01d29, re-add
4089 initialization of GLX after it was accidentally dropped from non-Xorg
4090 servers in 5f5bbbe543f65c48ecbb5cce80116a86ca3fbe86.
4091
4092 Signed-off-by: Sebastien Bacher <seb128@ubuntu.com>
4093 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=62346
4094 Reviewed-by: Eric Anholt <eric@anholt.net>
4095 Reviewed-by: Adam Jackson <ajax@redhat.com>
4096 Reviewed-by: Julien Cristau <jcristau@debian.org>
4097 Signed-off-by: Eric Anholt <eric@anholt.net>
4098
4099commit aa5534ec6931ffcc81cca76b98785d57de8e0e52
4100Author: Eric Anholt <eric@anholt.net>
4101Date: Fri Aug 23 11:23:00 2013 -0700
4102
4103 ephyr: Do grab/ungrab for ctrl+shift, not just shift+ctrl.
4104
4105 Given that the window title says "ctrl+shift", having pressing those
4106 keys in that order not ungrab you is fairly mean.
4107
4108 Signed-off-by: Eric Anholt <eric@anholt.net>
4109 Reviewed-by: Adam Jackson <ajax@redhat.com>
4110 Reviewed-by: Julien Cristau <jcristau@debian.org>
4111
4112commit 5cc15e1c427bacd73db7381df3474a23f7616624
4113Author: Eric Anholt <eric@anholt.net>
4114Date: Fri Aug 23 11:19:17 2013 -0700
4115
4116 ephyr: Flush the X connection when updating the window title.
4117
4118 Otherwise when you're doing the ctrl-shift mouse grab thing, you
4119 don't know what state you're in until the next rendering occurs.
4120
4121 Signed-off-by: Eric Anholt <eric@anholt.net>
4122 Reviewed-by: Adam Jackson <ajax@redhat.com>
4123 Reviewed-by: Julien Cristau <jcristau@debian.org>
4124
4125commit ba387cf21f7d95987211f75d8024601e7d64e322
4126Author: Eric Anholt <eric@anholt.net>
4127Date: Fri Aug 23 11:17:14 2013 -0700
4128
4129 ephyr: Use host (HW) cursors by default.
4130
4131 Unless you're working on the sw cursor rendering code, you surely want
4132 to have real hardware cursors.
4133
4134 Signed-off-by: Eric Anholt <eric@anholt.net>
4135 Reviewed-by: Adam Jackson <ajax@redhat.com>
4136 Reviewed-by: Julien Cristau <jcristau@debian.org>
4137
4138commit 0f5a2f13dc284dd5f0f28e2c4323b2e08d253a7d
4139Author: Eric Anholt <eric@anholt.net>
4140Date: Fri Aug 23 11:04:22 2013 -0700
4141
4142 ephyr: Garbage collect some DOA host window clipping code.
4143
4144 Introduced in 79782726617d4b5a4f9b376f21936d035fc870e1 but never used.
4145
4146 Signed-off-by: Eric Anholt <eric@anholt.net>
4147 Reviewed-by: Julien Cristau <jcristau@debian.org>
4148
4149commit 090de17d361dc9828203a2562565d6499998db46
4150Author: Eric Anholt <eric@anholt.net>
4151Date: Fri Aug 23 10:56:22 2013 -0700
4152
4153 ephyr: Garbage collect some dead XV clipping code.
4154
4155 Signed-off-by: Eric Anholt <eric@anholt.net>
4156 Reviewed-by: Julien Cristau <jcristau@debian.org>
4157
4158commit 347d0fa6961c2b3dc4efd4e7cca8ef98cd3ea06e
4159Author: Eric Anholt <eric@anholt.net>
4160Date: Fri Aug 23 10:44:23 2013 -0700
4161
4162 ephyr: Remove the helper libs for each of the optional components.
4163
4164 Note that EXTRA_DIST was always unnecessary, because automake notices
4165 conditional compile of source files and includes them.
4166
4167 Copyright header is added because git noted that this was a 61%
4168 rewrite.
4169
4170 Signed-off-by: Eric Anholt <eric@anholt.net>
4171 Reviewed-by: Julien Cristau <jcristau@debian.org>
4172
4173commit 74dff87790dbb3dcb23634f1e8cd54018e5ddc9d
4174Author: Eric Anholt <eric@anholt.net>
4175Date: Fri Aug 23 10:40:52 2013 -0700
4176
4177 ephyr: Remove the remaining bits of host/server XV split.
4178
4179 Signed-off-by: Eric Anholt <eric@anholt.net>
4180 Reviewed-by: Julien Cristau <jcristau@debian.org>
4181
4182commit 2f994dfe9d1add9ea8cb9c88ddb0d15843854788
4183Author: Eric Anholt <eric@anholt.net>
4184Date: Fri Aug 23 10:14:36 2013 -0700
4185
4186 ephyr: Remove the host/server split of the actual XV video operations.
4187
4188 I suspect there's more cleanup possible in ephyrHostXVPutImage() by
4189 sticking a bunch of the args in the port priv earlier.
4190
4191 Signed-off-by: Eric Anholt <eric@anholt.net>
4192 Reviewed-by: Julien Cristau <jcristau@debian.org>
4193
4194commit 0b7ce74315213ccd31802847e5a83e67090ded14
4195Author: Eric Anholt <eric@anholt.net>
4196Date: Fri Aug 23 10:01:23 2013 -0700
4197
4198 ephyr: Remove another host/server split for XV image formats.
4199
4200 Signed-off-by: Eric Anholt <eric@anholt.net>
4201 Reviewed-by: Julien Cristau <jcristau@debian.org>
4202
4203commit 1dcc28bcb402d7286d18b41b469062a5c176d7eb
4204Author: Eric Anholt <eric@anholt.net>
4205Date: Thu Aug 22 17:00:41 2013 -0700
4206
4207 ephyr: Remove the host/server split for video encodings.
4208
4209 Signed-off-by: Eric Anholt <eric@anholt.net>
4210 Reviewed-by: Julien Cristau <jcristau@debian.org>
4211
4212commit 28a37757e11068b4594f4128a0d64c6bc9a53ee0
4213Author: Eric Anholt <eric@anholt.net>
4214Date: Thu Aug 22 17:26:21 2013 -0700
4215
4216 ephyr: Remove the host/server split for XV formats.
4217
4218 Signed-off-by: Eric Anholt <eric@anholt.net>
4219 Reviewed-by: Julien Cristau <jcristau@debian.org>
4220
4221commit 41134d393a37dd5b971b018a3888fbfea0b9c736
4222Author: Eric Anholt <eric@anholt.net>
4223Date: Thu Aug 22 17:21:37 2013 -0700
4224
4225 ephyr: Remove the host/server split for XV attributes.
4226
4227 v2: Fix leaks of xcb replies (caught by Julien)
4228
4229 Signed-off-by: Eric Anholt <eric@anholt.net>
4230 Reviewed-by: Julien Cristau <jcristau@debian.org>
4231
4232commit 1342a93c49beb6e4924534304119d57fa587ad0f
4233Author: Eric Anholt <eric@anholt.net>
4234Date: Thu Aug 22 17:26:08 2013 -0700
4235
4236 ephyr: Remove the host/server split for XV adaptors.
4237
4238 Signed-off-by: Eric Anholt <eric@anholt.net>
4239 Reviewed-by: Julien Cristau <jcristau@debian.org>
4240
4241commit fad79d2e3e53de4f25b3d1dd7accb558464a4a46
4242Author: Eric Anholt <eric@anholt.net>
4243Date: Sun Aug 18 20:08:32 2013 +0200
4244
4245 ephyr: Refactor XV adaptor feature detection.
4246
4247 This obviously wanted a helper function beforehand, but even more so
4248 now that we have XCB.
4249
4250 Signed-off-by: Eric Anholt <eric@anholt.net>
4251 Reviewed-by: Julien Cristau <jcristau@debian.org>
4252
4253commit 7a9c311add62026972419ef0b04a55680072bfeb
4254Author: Eric Anholt <eric@anholt.net>
4255Date: Sun Aug 18 20:02:53 2013 +0200
4256
4257 ephyr: Remove some pointless indirection in the XV code.
4258
4259 Now that we have XCB on the server side, we don't need to split these
4260 functions out.
4261
4262 Signed-off-by: Eric Anholt <eric@anholt.net>
4263 Reviewed-by: Julien Cristau <jcristau@debian.org>
4264
4265commit 46cf6bf5692ef751ec9d17ae2292565d4b13f14b
4266Author: Eric Anholt <eric@anholt.net>
4267Date: Sun Aug 18 18:38:29 2013 +0200
4268
4269 ephyr: Move event processing into ephyr.c.
4270
4271 No more extra event structure to translate between hostx.c and
4272 ephyr.c!
4273
4274 Signed-off-by: Eric Anholt <eric@anholt.net>
4275 Reviewed-by: Julien Cristau <jcristau@debian.org>
4276
4277commit 847c856effde482c13cdefcdca8176cee3bf67c4
4278Author: Eric Anholt <eric@anholt.net>
4279Date: Sun Aug 18 17:28:08 2013 +0200
4280
4281 ephyr: Move the host screen info into the kdrive screen private.
4282
4283 We can include xcb bits from the same place as server headers, so
4284 there's no need to hide them any more.
4285
4286 Signed-off-by: Eric Anholt <eric@anholt.net>
4287 Reviewed-by: Julien Cristau <jcristau@debian.org>
4288
4289commit 8dadc78e233f341f1ecb8d404de03d98ab210620
4290Author: Eric Anholt <eric@anholt.net>
4291Date: Sun Aug 18 16:45:06 2013 +0200
4292
4293 ephyr: Rename and use the proper type for what was host_screen->info.
4294
4295 Now that we can include server headers in talking to host X, we don't
4296 need to hide any more.
4297
4298 Signed-off-by: Eric Anholt <eric@anholt.net>
4299 Reviewed-by: Julien Cristau <jcristau@debian.org>
4300
4301commit 18d836f6dde46bcac8b794d34f76032688c28851
4302Author: Eric Anholt <eric@anholt.net>
4303Date: Sun Aug 18 14:12:45 2013 +0200
4304
4305 ephyr: Expose a single function for detecting extensions.
4306
4307 v2: Fix trying to include xcb-dri in the non-dri-build case (Noted by
4308 Julien)
4309
4310 Signed-off-by: Eric Anholt <eric@anholt.net>
4311 Reviewed-by: Julien Cristau <jcristau@debian.org>
4312
4313commit 3aacb22d6473480c31fb08f0b2527747dbf5be57
4314Author: Eric Anholt <eric@anholt.net>
4315Date: Sun Aug 18 14:01:53 2013 +0200
4316
4317 ephyr: Drop dead ephyrHostAtomToLocal code.
4318
4319 Signed-off-by: Eric Anholt <eric@anholt.net>
4320 Reviewed-by: Julien Cristau <jcristau@debian.org>
4321
4322commit 4aaad0bfbcef018eb36839b1d034c3f60022694a
4323Author: Julien Cristau <jcristau@debian.org>
4324Date: Wed Oct 20 20:10:43 2010 +0200
4325
4326 Xephyr: we're not using Xlib anymore, no need to undef _XSERVER64
4327
4328 Drop obsolete comments about interaction between Xlib code and that
4329 macro, and stop undefining it.
4330
4331 Signed-off-by: Julien Cristau <jcristau@debian.org>
4332 Reviewed-by: Eric Anholt <eric@anholt.net>
4333
4334commit 0fc2f9f04a44599f598380d4df992bec60c0e3a1
4335Author: Julien Cristau <jcristau@debian.org>
4336Date: Sat Sep 25 16:05:07 2010 +0200
4337
4338 Xephyr: handle errors in event loop
4339
4340 Signed-off-by: Julien Cristau <jcristau@debian.org>
4341 Reviewed-by: Eric Anholt <eric@anholt.net>
4342 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4343
4344commit 3309a73c826f02c4b2e14180a3254546314ec323
4345Author: Julien Cristau <jcristau@debian.org>
4346Date: Fri Jun 25 14:20:57 2010 +0100
4347
4348 Xephyr: drop remaining Xlib dependency
4349
4350 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4351 Reviewed-by: Eric Anholt <eric@anholt.net>
4352 Signed-off-by: Julien Cristau <jcristau@debian.org>
4353
4354commit 7da2174cf93f59b7e0372a6dc9d37a510076e187
4355Author: Julien Cristau <jcristau@debian.org>
4356Date: Mon Jun 21 14:39:03 2010 +0100
4357
4358 Xephyr: move glx code to xcb
4359
4360 v2: Rebase on master, notably adding XCB for
4361 X_GLXvop_MakeCurrentReadSGI (anholt).
4362 v3: Fix use-after-free (caught by Julien).
4363
4364 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> (v1)
4365 Signed-off-by: Julien Cristau <jcristau@debian.org>
4366 Signed-off-by: Eric Anholt <eric@anholt.net>
4367
4368commit 19c2699b954e2978301c4449dfccc7ca40bf6ae6
4369Author: Julien Cristau <jcristau@debian.org>
4370Date: Wed Jun 23 18:36:00 2010 +0100
4371
4372 Xephyr: remove unused DRI1 code
4373
4374 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4375 Signed-off-by: Julien Cristau <jcristau@debian.org>
4376
4377commit 68b7db19e8d3c1ac55a811327d3058f4c8fb45ec
4378Author: Julien Cristau <jcristau@debian.org>
4379Date: Sun Jun 20 00:07:08 2010 +0100
4380
4381 Xephyr: move ephyrdri over to xcb
4382
4383 Require new xcb-xf86dri to get fixed GetDrawableInfo.
4384
4385 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4386 Signed-off-by: Julien Cristau <jcristau@debian.org>
4387
4388commit aea119cd05807ed77336f96a40f8e4607c6ce169
4389Author: Julien Cristau <jcristau@debian.org>
4390Date: Sun Jun 20 11:13:26 2010 +0100
4391
4392 Xephyr: use xcb-xv instead of libXv
4393
4394 v2: Massive, conflict-filled rebase. I think I resolved everything
4395 right, but this host xv code makes my eyes bleed. Touch-tested XV
4396 after the rebase (anholt).
4397
4398 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> (v1)
4399 Signed-off-by: Julien Cristau <jcristau@debian.org>
4400 Signed-off-by: Eric Anholt <eric@anholt.net>
4401
4402commit 3ea75a59d5fec3c4099388724942e216f2344328
4403Author: Julien Cristau <jcristau@debian.org>
4404Date: Mon Jun 21 17:20:32 2010 +0100
4405
4406 Xephyr: use xcb for event handling
4407
4408 v2: Rebase on indentation changes, squash in a simpler variant of the
4409 later event compression patch, fix server hang or segfault on
4410 window close by reimplementing the x_io_error_handler in the new
4411 XCB event loop (anholt).
4412
4413 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> (v1)
4414 Signed-off-by: Julien Cristau <jcristau@debian.org>
4415 Signed-off-by: Eric Anholt <eric@anholt.net>
4416
4417commit ff26845252889f7cc3258a0d3aa116189acc6aae
4418Author: Julien Cristau <jcristau@debian.org>
4419Date: Thu Jun 24 23:45:16 2010 +0100
4420
4421 Xephyr: some more hostx.c xcb-ification
4422
4423 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4424 Reviewed-by: Eric Anholt <eric@anholt.net>
4425 Signed-off-by: Julien Cristau <jcristau@debian.org>
4426
4427commit 9b55f76a265aab0d9943a40208a38800e5782ba5
4428Author: Julien Cristau <jcristau@debian.org>
4429Date: Thu Jun 24 23:35:15 2010 +0100
4430
4431 Xephyr: move HostX.visual to xcb
4432
4433 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4434 Reviewed-by: Eric Anholt <eric@anholt.net>
4435 Signed-off-by: Julien Cristau <jcristau@debian.org>
4436
4437commit 126312fb4a6e06024a306341e675163108426390
4438Author: Julien Cristau <jcristau@debian.org>
4439Date: Thu Jun 24 23:26:48 2010 +0100
4440
4441 Xephyr: replace XKeycodeToKeysym with xcb-keysyms
4442
4443 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4444 Reviewed-by: Eric Anholt <eric@anholt.net>
4445 Signed-off-by: Julien Cristau <jcristau@debian.org>
4446
4447commit 8543d3fdfd47d2a0423f3ef8e7429c23c5d6b053
4448Author: Julien Cristau <jcristau@debian.org>
4449Date: Thu Jun 24 23:17:32 2010 +0100
4450
4451 Xephyr: delete unused hostx_get_extension_info function
4452
4453 v2: Also remove the prototype (anholt)
4454
4455 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> (v1)
4456 Signed-off-by: Julien Cristau <jcristau@debian.org>
4457 Signed-off-by: Eric Anholt <eric@anholt.net>
4458
4459commit d72c11b9df4a98fd80cb6635d34964483e33bc41
4460Author: Julien Cristau <jcristau@debian.org>
4461Date: Mon Jun 21 14:25:11 2010 +0100
4462
4463 Xephyr: delete unused proxy code
4464
4465 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4466 Reviewed-by: Eric Anholt <eric@anholt.net>
4467 Signed-off-by: Julien Cristau <jcristau@debian.org>
4468
4469commit a71ea3f96769da157dde82a591e43264820772ac
4470Author: Julien Cristau <jcristau@debian.org>
4471Date: Sun Jun 20 23:34:54 2010 +0100
4472
4473 Xephyr: xcb-ify visual list construction
4474
4475 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4476 Reviewed-by: Eric Anholt <eric@anholt.net>
4477 Signed-off-by: Julien Cristau <jcristau@debian.org>
4478
4479commit ac2b0348c00d692ec1bd8d6c8ab99e9198e755a7
4480Author: Julien Cristau <jcristau@debian.org>
4481Date: Sun Jun 20 23:34:39 2010 +0100
4482
4483 Xephyr: xcb-ify pointer/keyboard grab
4484
4485 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4486 Reviewed-by: Eric Anholt <eric@anholt.net>
4487 Signed-off-by: Julien Cristau <jcristau@debian.org>
4488
4489commit 1f93e6b350d1b9a058009052a666a2e8796708cb
4490Author: Julien Cristau <jcristau@debian.org>
4491Date: Sun Jun 20 23:34:03 2010 +0100
4492
4493 Xephyr: no need for XDisplayKeycodes
4494
4495 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4496 Reviewed-by: Eric Anholt <eric@anholt.net>
4497 Signed-off-by: Julien Cristau <jcristau@debian.org>
4498
4499commit 73398e5bf61c30e69a40a19a1057c2263fbb91ef
4500Author: Julien Cristau <jcristau@debian.org>
4501Date: Thu Jun 24 23:00:38 2010 +0100
4502
4503 Xephyr: use xcb-shape instead of XShape*
4504
4505 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
4506 Reviewed-by: Eric Anholt <eric@anholt.net>
4507 Signed-off-by: Julien Cristau <jcristau@debian.org>
4508
4509commit a2b73da78de4e627965213d24a6c33f243a60eb6
4510Author: Julien Cristau <jcristau@debian.org>
4511Date: Sun Jun 20 00:05:40 2010 +0100
4512
4513 Xephyr: start converting hostx.c over to xcb
4514
4515 v2: Dropped the hostx_load_keymap changes, now that that function is
4516 gutted (anholt).
4517
4518 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> (v1)
4519 Signed-off-by: Julien Cristau <jcristau@debian.org>
4520 Signed-off-by: Eric Anholt <eric@anholt.net>
4521
4522commit 762606b4cdfb57c5202ae128ce4dbb8afe172d4e
4523Author: Julien Cristau <jcristau@debian.org>
4524Date: Fri Jun 25 17:59:13 2010 +0100
4525
4526 Xephyr: stop loading the host's keymap
4527
4528 This isn't used anywhere.
4529
4530 v2: Rebase to the top of the patch series (anholt)
4531
4532 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> (v1)
4533 Signed-off-by: Julien Cristau <jcristau@debian.org>
4534 Signed-off-by: Eric Anholt <eric@anholt.net>
4535
4536commit 3752a815800626ae93f64a856f5889aa6636344e
4537Author: Marc Haesen <marha@users.sourceforge.net>
4538Date: Sat Aug 10 17:38:38 2013 +0100
4539
4540 hw/xwin: Remove unnecessary and incorrect HWND casts
4541
4542 Signed-off-by: Marc Haesen <marha@users.sourceforge.net>
4543 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4544 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4545
4546commit 34932cef6cc800fbdf2f375506057ed8cd1e07f6
4547Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4548Date: Mon Apr 1 13:43:00 2013 +0100
4549
4550 hw/xwin: Fix compilation of winauth.c with -Werror=implicit-function-declaration
4551
4552 /jhbuild/checkout/xorg/xserver/hw/xwin/winauth.c: In function ‘MitGenerateCookie’:
4553 /jhbuild/checkout/xorg/xserver/hw/xwin/winauth.c:87:5: error: implicit declaration of function ‘MitAddCookie’ [-Werror=implicit-function-declaration]
4554
4555 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4556 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4557 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4558
4559commit 2a08554a0bf80b862c15676c31ecd463b616518c
4560Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4561Date: Mon Jun 17 23:54:01 2013 +0100
4562
4563 hw/xwin: Remove unused extern g_fUseUnicode from winclipboardwndproc.c
4564
4565 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk
4566 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4567 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4568
4569commit 69401f38a697c941ea58351028f4cd4b810f9188
4570Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4571Date: Mon Jun 17 18:37:26 2013 +0100
4572
4573 hw/xwin: Remove unneeded forward declaration of winProcessXEventsTimeout()
4574
4575 Remove unneeded forward declaration of winProcessXEventsTimeout(), the actual
4576 definition immediately follows.
4577
4578 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4579 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4580 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4581
4582commit 9c45a1ad44f12ac25e1c3b7a18a3bceb85a2cf38
4583Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4584Date: Mon Jun 17 16:58:47 2013 +0100
4585
4586 hw/xwin: Remove prototype for non-existent winDeinitClipboard()
4587
4588 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4589 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4590 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4591
4592commit acc46fc217352b94f77a6b5b14808ea89875a2a1
4593Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4594Date: Mon Jun 17 00:40:13 2013 +0100
4595
4596 hw/xwin: Remove unused externs from winclipboardwrappers.c
4597
4598 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4599 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4600 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4601
4602commit fcc7767f8cd6c318861b1060ef6a3936750a0dd2
4603Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4604Date: Mon Jun 10 17:31:12 2013 +0100
4605
4606 hw/xwin: Remove unused winProcQueryTreeOrig variable
4607
4608 Unused since 47c7b6d3e626497747ae2780f259a15b8e6c846f "Remove no-longer needed
4609 tricks used to prevent the clipboard client from being killed"
4610
4611 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4612 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4613 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4614
4615commit 8fa882a980abea6f05b4b59447a187aea2c793a1
4616Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4617Date: Sat Aug 10 12:41:07 2013 +0100
4618
4619 hw/xwin: Fix ARGB cursor conversion on x86_64
4620
4621 Fix erroneous use of unsigned long * for lpBits in winXCursorToHCURSOR() which
4622 leads to ARGB cursors being vertically streched on x86_64 by interleaving blank
4623 rows of pixels.
4624
4625 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4626 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4627 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4628
4629commit 4bfb2dce5eea4923eaf86eca33b96087b28235f5
4630Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4631Date: Sun Apr 7 23:59:44 2013 +0100
4632
4633 hw/xwin: Fix an issue in winSetSpansNativeGDI() identifed by -Warray-bounds
4634
4635 The BITMAPINFO local only has room for a single RBGQUAD in bmiColors, but we
4636 access two (black and white for a mono-color DIB). Fix by changing to a dynamic
4637 allocation big enough for a BITMAPINFO and a RGBQUAD.
4638
4639 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4640 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4641 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4642
4643commit 784c006adb22c6693ad30dd0b1b05369cfaa7b5e
4644Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4645Date: Sat Aug 10 11:55:23 2013 +0100
4646
4647 hw/xwin: Fix -Warray-bounds warning in winXCursorToHCURSOR()
4648
4649 Rewrite winXCursorToHCURSOR() so access to BITMAPINFO bmiColors member doesn't
4650 trigger an -Warray-bounds warning.
4651
4652 Note that the underlying storage is allocated as a BITMAPV4HEADER, so has
4653 sufficent room for the extra RGBQUADs bmiColors after the BITMAPINFO bmiHeader.
4654
4655 wincursor.c: In function 'winSetCursor':
4656 wincursor.c:293:24: error: array subscript is above array bounds [-Werror=array-bounds]
4657 wincursor.c:294:24: error: array subscript is above array bounds [-Werror=array-bounds]
4658 wincursor.c:295:24: error: array subscript is above array bounds [-Werror=array-bounds]
4659 wincursor.c:296:24: error: array subscript is above array bounds [-Werror=array-bounds]
4660 wincursor.c:297:24: error: array subscript is above array bounds [-Werror=array-bounds]
4661 wincursor.c:298:24: error: array subscript is above array bounds [-Werror=array-bounds]
4662 wincursor.c:299:24: error: array subscript is above array bounds [-Werror=array-bounds]
4663 wincursor.c:300:24: error: array subscript is above array bounds [-Werror=array-bounds]
4664
4665 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4666 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4667 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4668
4669commit 41d03c9ccce665b919b180bbb819b78c750aa325
4670Author: Marc Haesen <marha@users.sourceforge.net>
4671Date: Fri Aug 2 18:19:22 2013 +0100
4672
4673 hw/xwin: Correct size of _WINDOWSWM_NATIVE_HWND property on x86_64
4674
4675 Use the correct size of a HWND on x86_64 in XChangeProperty() and
4676 XGetWindowProperty() calls for the _WINDOWSWM_NATIVE_HWND property.
4677
4678 Signed-off-by: Marc Haesen <marha@users.sourceforge.net>
4679 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4680 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4681
4682commit 390cfec10ef7d52391f890c7593ac82a7d77d856
4683Author: Colin Harrison <colin.harrison@virgin.net>
4684Date: Fri Aug 9 16:48:12 2013 +0100
4685
4686 hw/xwin: Correct winprefs.c function signatures for x64.
4687
4688 Correct SetupRootMenu(), SetupSysMenu(), HandleCustomWM_INITMENU() and
4689 HandleCustomWM_COMMAND() function signatures which use unsigned long parameters,
4690 where just a specific HWND or HMENU handle type should have been used.
4691
4692 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
4693 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4694 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4695
4696commit e716baedc4d8e52a60f43ef21aba771b340d8c8b
4697Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4698Date: Sat Mar 31 14:42:45 2012 +0100
4699
4700 Move pseudoramiX code where it can be shared between Xwin and Xquartz
4701
4702 Move pseudoramiX code to a separate top-level directory. Link Xwin and Xquartz
4703 with libPseudoramiX
4704
4705 I'm not sure moving this to a top-level directory is appropriate, but I'm not
4706 sure where else it fits.
4707
4708 Future work: pseudoramiX can probably be consolidated with the rrxinerama code
4709 (which I think provides fake xinerama data when real XINERAMA is disabled and we
4710 only have one screen)
4711
4712 v2: fix distcheck
4713
4714 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4715 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4716 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4717
4718commit ce0217b98261dcf8b20cfb743831965dda1950ec
4719Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4720Date: Sat Aug 10 12:35:55 2013 +0100
4721
4722 hw/xwin: Fix winglobals.h for MinGW64 build
4723
4724 Include pthread.h in winglobals.h to fix build with latest MinGW64 headers
4725
4726 In file included from
4727 winmultiwindowicons.c:47:0:
4728 winglobals.h:92:1: error: unknown type name ‘pthread_mutex_t’
4729
4730 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4731 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4732 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4733
4734commit b27756dbe6a1be4633ec47a8fe92a4c217f68f93
4735Author: Jon TURNEY <jon.turney@dronecode.org.uk>
4736Date: Sun Aug 11 12:15:11 2013 +0100
4737
4738 hw/xwin: Fix inclusion of shlobj.h in MinGW64 build
4739
4740 Fix build with latest MinGW64 headers by wrapping Status type in shlobj.h as well
4741
4742 In file included from InitOutput.c:51:0:
4743 /usr/i686-w64-mingw32/sys-root/mingw/include/shlobj.h:1231:44: error: expected identifier or ‘(’ before ‘int’
4744 /usr/i686-w64-mingw32/sys-root/mingw/include/shlobj.h:1248:44: error: expected identifier or ‘(’ before ‘int’
4745
4746 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
4747 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
4748 Reviewed-by: Marc Haesen <marha@users.sourceforge.net>
4749
4750commit 94d4e29aedc69431fa9b299ca1b67947173d7a24
4751Author: Peter Hutterer <peter.hutterer@who-t.net>
4752Date: Mon Aug 26 15:51:46 2013 +1000
4753
4754 Xi: allow for XIAllowEvent requests larger than XI < 2.2 size (#68554)
4755
4756 XIAllowEvents changed length in XI 2.2 (for the touchid). A bug in libXi
4757 causes libXi to always use the new request length if the server supports
4758 2.2, regardless of the client's XIQueryVersion request.
4759
4760 The server takes the client's XIQueryVersion request into account though,
4761 resulting in a BadLength error if a 2.[0,1] client calls XIAllowEvents on a
4762 XI 2.2+ server.
4763
4764 Can't fix this in libXi, so work around this in the server.
4765
4766 X.Org Bug 68554 <http://bugs.freedesktop.org/show_bug.cgi?id=68554>
4767
4768 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4769 Reviewed-by: Adam Jackson <ajax@redhat.com>
4770
4771commit 82939e02392cbb880313fe92957091ff89ce2f2b
4772Author: Peter Hutterer <peter.hutterer@who-t.net>
4773Date: Fri Aug 16 15:30:59 2013 +1000
4774
4775 dix: check for grab type before checking XI2 mask
4776
4777 if the grab type isn't XI2, grab->xi2mask is random. That random data may
4778 have the enter/leave mask set, leading to events sent to the client that the
4779 client can't handler.
4780
4781 Source of these errors:
4782 _xgeWireToEvent: Unknown extension 131, this should never happen.
4783
4784 Simplest reproducer:
4785 Start Xephyr, press button inside window, move out. As the pointer leaves
4786 the Xephyr window, the errors appear.
4787
4788 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4789 Reviewed-by: Adam Jackson <ajax@redhat.com>
4790
4791commit 16894df1ffb6fe21bf723dd954578e8d934799ca
4792Author: Peter Hutterer <peter.hutterer@who-t.net>
4793Date: Fri Aug 16 10:55:09 2013 +1000
4794
4795 Xi: replace loop with memset
4796
4797 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4798 Reviewed-by: Adam Jackson <ajax@redhat.com>
4799
4800commit 8f2292b683f591052da2fe94fb3831a3c9a5aa61
4801Author: Peter Hutterer <peter.hutterer@who-t.net>
4802Date: Fri Aug 16 11:05:49 2013 +1000
4803
4804 include: wrap EMASKSIZE in parentheses
4805
4806 Otherwise things like EMASKSIZE * foo will yield interesting results.
4807
4808 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4809 Reviewed-by: Adam Jackson <ajax@redhat.com>
4810
4811commit d803f296c616ccf55711cf0af9e7d3eae2edda16
4812Author: Peter Hutterer <peter.hutterer@who-t.net>
4813Date: Fri Aug 23 09:04:01 2013 +1000
4814
4815 test: fix the gcc diagnostics pragma
4816
4817 pop without push restores the commandline options. The proper way is to
4818 push, then ignore, then pop.
4819
4820 And while we're at it, change the pop argument to a comment - pop ignores
4821 the argument, but be proper about it.
4822
4823 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4824
4825commit 05ea6307fa4f0bee3d3bf536396059f3906791a0
4826Author: Egbert Eich <eich@freedesktop.org>
4827Date: Fri Aug 16 19:52:16 2013 +0200
4828
4829 DIX/Xi: Pass correct client to CheckDeviceGrabAndHintWindow()
4830
4831 If we have a client which has registered for a DeviceButton grab
4832 be sure to pass this to CheckDeviceGrabAndHintWindow(). Since the
4833 order of clients is arbitrary there is no guarantee that the last
4834 client in the list is the one that belongs to this class.
4835
4836 Signed-off-by: Egbert Eich <eich@freedesktop.org>
4837 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
4838 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4839
4840commit e01a3f65d3e6d92f92ef2992b338cc9625bde082
4841Author: Eric Anholt <eric@anholt.net>
4842Date: Sat Aug 17 14:50:37 2013 +0200
4843
4844 ephyr: Fix warning about XID vs unsigned long * by changing function args
4845
4846 There's no reason to pass the data back out to the caller, since the
4847 caller was dropping it on the floor. The original data is a CARD32,
4848 so no need to mess with weird unsigned longs.
4849
4850 Signed-off-by: Eric Anholt <eric@anholt.net>
4851 Reviewed-by: Julien Cristau <jcristau@debian.org>
4852
4853commit f1cefe020bd21c07dfcc9f204e79939499abbd35
4854Author: Eric Anholt <eric@anholt.net>
4855Date: Sat Aug 17 14:38:08 2013 +0200
4856
4857 ephyr: Drop is_ok logging.
4858
4859 The only cases that lead to !is_ok are already EPHYR_LOG_ERROR, and it
4860 fixes set-but-unused compiler warnings.
4861
4862 Signed-off-by: Eric Anholt <eric@anholt.net>
4863 Reviewed-by: Julien Cristau <jcristau@debian.org>
4864
4865commit 77c7a2014e3942b55c0b78308edf02eab33f2e60
4866Author: Eric Anholt <eric@anholt.net>
4867Date: Sat Aug 17 14:36:09 2013 +0200
4868
4869 kdrive: Fix const cast warnings in arguments processing.
4870
4871 Signed-off-by: Eric Anholt <eric@anholt.net>
4872 Reviewed-by: Julien Cristau <jcristau@debian.org>
4873
4874commit 7107937e473a3557c37d719cb100b2e9b09075dd
4875Author: Eric Anholt <eric@anholt.net>
4876Date: Sat Aug 17 13:11:17 2013 +0200
4877
4878 kdrive: Fix const cast warnings in driver name handling.
4879
4880 Signed-off-by: Eric Anholt <eric@anholt.net>
4881 Reviewed-by: Julien Cristau <jcristau@debian.org>
4882
4883commit 56c405d46a86853771a0b808da4c8a9ef626960c
4884Author: Eric Anholt <eric@anholt.net>
4885Date: Sat Aug 17 13:07:23 2013 +0200
4886
4887 kdrive: Fix a few easy cast warnings.
4888
4889 Signed-off-by: Eric Anholt <eric@anholt.net>
4890 Reviewed-by: Julien Cristau <jcristau@debian.org>
4891
4892commit a9c48a07eea50468fa47ab178cdd91edb0b5727f
4893Author: Eric Anholt <eric@anholt.net>
4894Date: Sat Aug 17 13:06:23 2013 +0200
4895
4896 kdrive: Drop dead code.
4897
4898 This is unused as of the introduction of a helper in
4899 e7150db5350bc2113ff4126019b489847a4dc217
4900
4901 Signed-off-by: Eric Anholt <eric@anholt.net>
4902 Reviewed-by: Julien Cristau <jcristau@debian.org>
4903
4904commit 873ae962bdb7559c952b5955ba262285701982a8
4905Author: Eric Anholt <eric@anholt.net>
4906Date: Sat Aug 17 13:01:57 2013 +0200
4907
4908 ephyr: Fix dead code warning.
4909
4910 Signed-off-by: Eric Anholt <eric@anholt.net>
4911 Reviewed-by: Julien Cristau <jcristau@debian.org>
4912
4913commit 34af2eebe19e2f8afe35a5614d4d44fa4fd89d6b
4914Author: Eric Anholt <eric@anholt.net>
4915Date: Sat Aug 17 13:01:32 2013 +0200
4916
4917 ephyr: Fix const-cast warnings for setting window title.
4918
4919 Signed-off-by: Eric Anholt <eric@anholt.net>
4920 Reviewed-by: Julien Cristau <jcristau@debian.org>
4921
4922commit 7edf5066db48434739ac64e0576fbb4fdf0f37f9
4923Author: Eric Anholt <eric@anholt.net>
4924Date: Sat Aug 17 12:54:33 2013 +0200
4925
4926 ephyr: Fix const cast warnings in ephyrhostvideo.
4927
4928 Signed-off-by: Eric Anholt <eric@anholt.net>
4929 Reviewed-by: Julien Cristau <jcristau@debian.org>
4930
4931commit fe7463b8ce0de301c2f82b108c93963424f77219
4932Author: Julien Cristau <jcristau@debian.org>
4933Date: Sat Jul 27 12:09:07 2013 +0200
4934
4935 dix: add missing include for DeleteWindowFromAnySelections
4936
4937 Fixes build error with XACE disabled:
4938 window.c:886:5: error: implicit declaration of function 'DeleteWindowFromAnySelections' [-Werror=implicit-function-declaration]
4939 DeleteWindowFromAnySelections(pWin);
4940 ^
4941
4942 Debian bug#701372
4943
4944 Reported-by: Matthias Klose <doko@debian.org>
4945 Signed-off-by: Julien Cristau <jcristau@debian.org>
4946 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4947 Signed-off-by: Keith Packard <keithp@keithp.com>
4948
4949commit 6a49f8836a649322d38a364d9ae084be40fb2305
4950Author: Keith Packard <keithp@keithp.com>
4951Date: Sat Aug 17 12:17:36 2013 +0200
4952
4953 Add .dir-locals.el
4954
4955 Enforce X.org style with an emacs configuration file
4956
4957 Signed-off-by: Keith Packard <keithp@keithp.com>
4958
4959commit f67d022523c59a27f3bf8791aa9ca6624318b1fd
4960Merge: 5a36cdd 9680f6a
4961Author: Keith Packard <keithp@keithp.com>
4962Date: Tue Aug 6 07:12:12 2013 +0200
4963
4964 Merge commit '9680f6a12d642c2e4bfa09bf459d678214059d74'
4965
4966commit 5a36cdd91530d27627e39159a89b53f9fbb75280
4967Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4968Date: Mon Aug 5 20:46:45 2013 -0700
4969
4970 Allow disabling XFree86-DGA, DRI, VidModeExtension extensions
4971
4972 Code to recognize these in extension enable/disable options was wrapped
4973 in #ifdef XorgLoader, but that's not defined when building miinitext.c
4974 since the great module merge of 1.13. Change to an #ifdef that is defined.
4975
4976 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4977 Reviewed-by: Robert Hooker <robert.hooker@canonical.com>
4978 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
4979 Signed-off-by: Keith Packard <keithp@keithp.com>
4980
4981commit 9680f6a12d642c2e4bfa09bf459d678214059d74
4982Author: Peter Hutterer <peter.hutterer@who-t.net>
4983Date: Tue Aug 6 13:14:56 2013 +1000
4984
4985 xfree86: de-duplicate some AM_CPPFLAGS
4986
4987 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4988 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4989
4990commit bbef8e46f2fed8f823ee4b97e96c3fcee28f3b75
4991Author: Peter Hutterer <peter.hutterer@who-t.net>
4992Date: Tue Aug 6 13:08:13 2013 +1000
4993
4994 Replace INCLUDES with AM_CPPFLAGS
4995
4996 newer automake gets quite noisy about this.
4997 hw/xfree86/ddc/Makefile.am:7: warning:
4998 'INCLUDES' is the old name for 'AM_CPPFLAGS' (or '*_CPPFLAGS')
4999 and many more of these.
5000
5001 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5002 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5003
5004commit c218ba8423a73a7a643cb17789db8a1dd0901ca6
5005Author: Julien Cristau <jcristau@debian.org>
5006Date: Fri Aug 2 20:07:36 2013 +0200
5007
5008 xfree86: improve check for posix saved ids
5009
5010 Replace hardcoded SVR4 || linux || CSRG_BASED with an autoconf check and
5011 the _POSIX_SAVED_IDS macro.
5012
5013 Suggested-by: Mark Kettenis <mark.kettenis@xs4all.nl>
5014 Signed-off-by: Julien Cristau <jcristau@debian.org>
5015 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5016 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>.
5017 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5018
5019commit 930c6ff15d437dfb0f897e8cb4253abba70819cc
5020Author: Julien Cristau <jcristau@debian.org>
5021Date: Fri Aug 2 23:46:00 2013 +0200
5022
5023 test: include dix-config.h in hashtabletest.c
5024
5025 Missing _XSERVER64 define caused inconsistent sizeof(XID) between the
5026 test and hashtable code, leading to test failures on 64bit big endian
5027 archs like s390x or ppc64.
5028
5029 Signed-off-by: Julien Cristau <jcristau@debian.org>
5030 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
5031 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5032
5033commit b6e5c4669e0db391966deb397e8c975ec7f0124d
5034Author: Maarten Lankhorst <maarten.lankhorst@canonical.com>
5035Date: Tue Jul 30 15:31:24 2013 +0200
5036
5037 test/xi2: fix protocol-xiqueryversion test
5038
5039 The old code was broken and allowed setting client version >= XIVersion,
5040 this was fixed in the previous patch, but updating the value for XIVersion
5041 broke the tests, so fix the tests too.
5042
5043 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
5044 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
5045 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5046
5047commit 500e844a24962c9e70abb3d614f1973013b2de73
5048Author: Maarten Lankhorst <maarten.lankhorst@canonical.com>
5049Date: Tue Jul 30 14:45:21 2013 +0200
5050
5051 Xi: Clamp XIClient maximal version to XIVersion
5052
5053 Do not allow setting client version to an arbitrary value >= XIVersion.
5054 Fixes a test error with test/xi2/protocol-xiqueryversion.c, introduced by
5055 commit 4360514d1c "Xi: Allow clients to ask for 2.3 and then 2.2 without failing"
5056
5057 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
5058 Reviewed-by: Keith Packard <keithp@keithp.com>
5059 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5060
5061commit e1ab8f239bdc6af17ef6cd0e9c6187c4e2e2f5bc
5062Author: Dave Airlie <airlied@redhat.com>
5063Date: Wed Jul 31 11:35:44 2013 +1000
5064
5065 xfree86/man: document AutoAddGPU
5066
5067 This at least mentions AutoAddGPU and hints at when you might
5068 want to disable it.
5069
5070 Signed-off-by: Dave Airlie <airlied@redhat.com>
5071 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
5072 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5073
5074commit ff38bbe81ace85bf675bbaa0a9ca5f3b32ede449
5075Merge: bdd1e22 4360514
5076Author: Keith Packard <keithp@keithp.com>
5077Date: Thu Jul 25 22:50:27 2013 -0700
5078
5079 Merge remote-tracking branch 'whot/for-keith'
5080
5081commit 4360514d1cc8e3132f93f56172d291074e8c770f
5082Author: Keith Packard <keithp@keithp.com>
5083Date: Wed Jul 10 22:42:55 2013 -0700
5084
5085 Xi: Allow clients to ask for 2.3 and then 2.2 without failing
5086
5087 This allows different sub-systems within the same application to
5088 request different Xi versions without either getting old behaviour
5089 everywhere or simply failing with a BadValue.
5090
5091 Signed-off-by: Keith Packard <keithp@keithp.com>
5092 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5093
5094commit bdd1e22cbde9ea2324e4e1991c9e152f22f88151
5095Author: Aaron Plattner <aplattner@nvidia.com>
5096Date: Tue Apr 30 14:14:23 2013 -0700
5097
5098 xfree86: detach scanout pixmaps when detaching output GPUs
5099
5100 Commit 8f4640bdb9d3988148e09a08d2c7e3bab1d538d6 fixed a bit of a
5101 chicken-and-egg problem by detaching GPU screens when their providers
5102 are destroyed, which happens before CloseScreen is called. However,
5103 this created a new problem: the GPU screen tears down its RandR crtc
5104 objects during CloseScreen and if one of them is active, it tries to
5105 detach the scanout pixmap then. This crashes because
5106 RRCrtcDetachScanoutPixmap tries to get the master screen's screen
5107 pixmap, but crtc->pScreen->current_master is already NULL at that
5108 point.
5109
5110 It doesn't make sense for an unbound GPU screen to still be scanning
5111 out its former master screen's pixmap, so detach them first when the
5112 provider is destroyed.
5113
5114 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
5115 Reviewed-by: Dave Airlie <airlied@redhat.com>
5116 Signed-off-by: Keith Packard <keithp@keithp.com>
5117
5118commit 3d8756631070c440cefa31b35fea3d407f187810
5119Author: Peter Hutterer <peter.hutterer@who-t.net>
5120Date: Mon Jul 22 13:43:07 2013 +1000
5121
5122 dix: set the valuator mask to ensure XI 1.x events have data
5123
5124 XI 1.x only allows for first + num valuators, so if a device sends data for
5125 valuators 0 and 2+ only (i.e. valuator 1 is missing) we still need to get
5126 the data for that from somewhere.
5127 XI 1.x uses the hack of an unset valuator mask to get the right coordinates,
5128 i.e. we set the value but don't set the mask for it so XI2 events have the
5129 right mask.
5130
5131 For an absolute device in relative mode, this broke in b28a1af55cf, the
5132 value was now always 0. This wasn't visible on the cursor, only in an XI 1.x
5133 client. The GIMP e.g. sees jumps to x/0 every few events.
5134
5135 Drop the condition introduced in b28a1af55cf, data in valuators is always
5136 absolute, regardless of the mode.
5137
5138 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5139
5140commit 4fb686d6a6777950f0e0d55b848cd2af4cbad372
5141Author: Peter Hutterer <peter.hutterer@who-t.net>
5142Date: Wed Jul 24 11:50:00 2013 +1000
5143
5144 dix: check the xi2mask, not the grab type for touch listeners
5145
5146 grab->type is only non-zero for passive grabs. We're checking an active grab
5147 here, so we need to check if the touch mask is set on the grab.
5148
5149 Test case: grab the device, then start two simultaneous touches. The
5150 grabbing client won't see the second touchpoints because grab->type is 0
5151 and the second touch is not an emulating pointer.
5152
5153 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5154
5155commit 43ac0491e36cdbb716b5c9d39c97d0aba3bebd75
5156Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5157Date: Mon Apr 8 01:28:36 2013 +0100
5158
5159 hw/xwin: Fix numerous 64-bit format/type cast issues with debug printing of pointers
5160
5161 Numerous pieces of debug output cast a pointer to an int and then use a "%08x"
5162 printf format.
5163
5164 Use "%p" format for 64-bit portability.
5165
5166 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5167 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5168
5169commit e95bb97073ca77193e4c51bd1504403fbe245533
5170Author: Colin Harrison <colin.harrison@virgin.net>
5171Date: Tue Jun 25 21:34:43 2013 +0100
5172
5173 hw/xwin: Fix possible crash in winMultiWindowGetClassHint
5174
5175 Fix a possible crash in winMultiWindowGetClassHint() when an application doesn't
5176 null terminate the WM_CLASS property class name (which is an ICCCM conformance
5177 bug in the application)
5178
5179 (Reported for running the contiki cooja simulator in multiwindow mode, although
5180 it seems that many Java clients may have this problem, see [1])
5181
5182 Based on a patch by Marc Haesen.
5183
5184 v2: Avoid using strnlen() which is missing on MinGW
5185 v3: Align with Xming patch
5186
5187 [1] http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6961123
5188
5189 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
5190 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
5191 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5192
5193commit 4bc375aa2f08bd092f60d3e358d345fc9a0d3637
5194Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5195Date: Tue Jun 18 19:21:02 2013 +0100
5196
5197 hw/xwin: Remove obsolete WIN_XEVENTS_SHUTDOWN
5198
5199 Remove obsolete WIN_XEVENTS_SHUTDOWN. This event is never generated.
5200
5201 (I think the idea was to listen for WM_DELETE_WINDOW, but that's not a very
5202 useful thing to do for a hidden window.)
5203
5204 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5205 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5206
5207commit 2663203cd6994166e107e796c4724a4ed3b59370
5208Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5209Date: Tue Jun 18 18:03:05 2013 +0100
5210
5211 configure.ac: Check for python at configure time when building XWin with AIGLX
5212
5213 Check for python at configure time when building XWin with AIGLX, it's used to
5214 generate the wrapper code for native GL functions.
5215
5216 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5217 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5218
5219commit 9fe360bf2be5c959fb21835955ef550098ccbbf0
5220Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5221Date: Sun Jun 16 00:07:57 2013 +0100
5222
5223 hw/xwin: Avoid a null dereference if CreateDIBSection() fails in NetWMToWinIconAlpha()
5224
5225 Avoid a null dereference of DIB_pixels if CreateDIBSection() fails in
5226 NetWMToWinIconAlpha()
5227
5228 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5229 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5230
5231commit 71b5f56302bbd8be62f63f0dd62cbcd33aab3ac5
5232Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5233Date: Thu Feb 21 17:12:17 2013 +0000
5234
5235 hw/xwin: Handle WM_MOUSEHWHEEL
5236
5237 Handle WM_MOUSEHWHEEL tilt wheel messages, similarly to WM_MOUSEWHEEL scroll
5238 wheel messages, to generate X button 6 and 7 presses and releases.
5239
5240 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5241 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5242
5243commit cf9c777ee094d660e0c95559373fd23ee910362e
5244Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5245Date: Thu Feb 21 17:10:45 2013 +0000
5246
5247 hw/xwin: Map extra mouse buttons 1 and 2 to X buttons 8 and 9
5248
5249 Map extra mouse buttons 1 and 2 to X buttons 8 and 9, as conventional, leaving X
5250 buttons 6 and 7 for tilt wheel.
5251
5252 Also add button labels for buttons 6, 7, 8 and 9 and change btn_labels in from a
5253 dynamic allocation to a fixed one of the required size for all the labels we
5254 use.
5255
5256 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5257 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5258
5259commit c9d7b9516f4a04cb9012c6c1e9466491a468aa07
5260Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5261Date: Tue Feb 19 14:38:40 2013 +0000
5262
5263 hw/xwin: Remove unneeded WM_XBUTTON message defines
5264
5265 Remove unneeded WM_XBUTTON message defines, they have been provided by w32api
5266 for a long time now.
5267
5268 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5269 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5270
5271commit 5dccfc63f9c7056e7cd30884e1d3ccea86a2f419
5272Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5273Date: Mon Dec 17 22:38:00 2012 +0000
5274
5275 hw/xwin: Change winTranslateKey() to return it's result as it's return value
5276
5277 Change winTranslateKey() to return it's result as it's return value, and change
5278 it's uses as well.
5279
5280 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5281 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5282
5283commit 9b4cec76f1d7792d4bf64fa069ea6b64bc42da0d
5284Author: Oliver Schmidt <oschmidt-mailinglists@gmx.de>
5285Date: Mon Nov 5 15:05:32 2012 +0000
5286
5287 hw/xwin: Consider left and right modifier keys independently on gaining focus
5288
5289 Handle left and right ctrl and shift keys independently
5290
5291 Assume that all modifiers are cleared when all keys are released on focus lost,
5292 as internalKeyState doesn't record which modifier key was pressed.
5293
5294 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5295 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5296
5297commit 235149d0b4af9097e47b3af0ba56d4eb0179518c
5298Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5299Date: Mon Apr 5 14:22:39 2010 +0100
5300
5301 hw/xwin: Remove an extra '\n' from some log messages
5302
5303 Remove an extra '\n' from internal client IOError log messages
5304
5305 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5306 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5307
5308commit c7aa9f7578e721b0c9e565e7a085b6aeb30bf2ac
5309Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5310Date: Sat Feb 4 17:04:11 2012 +0000
5311
5312 hw/xwin: _NET_WM_STATE is ATOM[] not ATOM
5313
5314 _NET_WM_STATE is ATOM[] not ATOM, a list of window state hints, so check all of
5315 the atoms, not just the first one
5316
5317 See EWMH specifcation, section "Application Window Properties"
5318
5319 v2: Actually use [] on the returned atom data
5320
5321 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5322 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5323
5324commit 682ccac90b18f293520f77db5b163fcb40328e2b
5325Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5326Date: Wed Aug 31 21:35:14 2011 +0100
5327
5328 hw/xwin: Improve WM_ENDSESSION handling using separate messaging window thread
5329
5330 Currently, WM_ENDSESSION just calls GiveUp() to set the DE_TERMINATE flag. But
5331 for the X server to exit cleanly, we also need the X server dispatch loop to be
5332 unblocked so it can notice that DE_TERMINATE has been set and exit, removing
5333 it's lock file and any unix domain socket.
5334
5335 It appears that the system will terminate the process when the last UI thread in
5336 that process returns from processing WM_ENDSESSION for the last top-level
5337 window.
5338
5339 Since WM_ENDSESSION appears to sent by the system via SendMessage()
5340 (synchronously) and the wndproc is called to process it in the message thread
5341 for that window (the X server thread), we can't easily terminate the X server
5342 dispatch loop from inside the WM_ENDSESSION message processing.
5343
5344 So, create a messaging window, a hidden, top-level window, with a separate
5345 thread to catch this message, and process it by calling GiveUp() and then
5346 blocking on a mutex until the X server dispatch loop exits.
5347
5348 Also, notice when this is a shutdown cancel WM_ENDSESSION message and take no
5349 action.
5350
5351 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5352 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5353
5354commit 15febb05d77cc7e7185c942f35459838f75cfdc0
5355Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5356Date: Mon Sep 3 15:19:43 2012 +0100
5357
5358 Allow DDX to provide a main()
5359
5360 XQuartz already conditionally renames main() as dix_main() so it can provide
5361 it's own main(). This isn't the ideal way of doing this, as it prevents libdix
5362 built this way from being useful with any other DDX.
5363
5364 So instead, always name that function dix_main(), and also provide a stub main()
5365 which just calls dix_main(), which can be overriden in the DDX.
5366
5367 Add a main() to XWin (XQuartz already has one, of course).
5368
5369 It's no longer neccessary to link XWin and XQuartz with libmain.
5370
5371 v2: Remove unneeded stub main hw/xwin/InitOutput.c
5372
5373 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5374 Reviewed-by: Julien Cristau <jcristau@debian.org>
5375 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5376
5377commit d5ebe20f9ba9569351c4a41449866679fd60ba45
5378Merge: 21ea7eb 33f7e60
5379Author: Keith Packard <keithp@keithp.com>
5380Date: Mon Jul 22 20:34:41 2013 -0700
5381
5382 Merge remote-tracking branch 'alanc/master'
5383
5384commit 21ea7ebb6a1ad56f3060747af0df235ca6ff27fa
5385Author: Peter Hutterer <peter.hutterer@who-t.net>
5386Date: Fri Jul 19 08:56:38 2013 +1000
5387
5388 dix: scale y back instead of x up when pre-scaling coordinates
5389
5390 The peculiar way we handle coordinates results in relative coordinates on
5391 absolute devices being added to the last value, then that value is mapped to
5392 the screen (taking the device dimensions into account). From that mapped
5393 value we get the final coordinates, both screen and device coordinates.
5394
5395 To avoid uneven scaling on relative coordinates, they are pre-scaled by
5396 screen ratio:resolution:device ratio factor before being mapped. This
5397 ensures that a circle drawn on the device is a circle on the screen.
5398
5399 Previously, we used the ratio to scale x up. Synaptics already does its own
5400 scaling based on the resolution and that is done by scaling y down by the
5401 ratio. So we can remove the code from the driver and get approximately the
5402 same behaviour here.
5403
5404 Minor ABI bump, so we can remove this from synaptics.
5405
5406 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5407 Tested-by: Emmanuel Benisty <benisty.e@gmail.com>
5408
5409commit 33f7e60785f12770ce10558c2ca7ce1323eefc59
5410Author: Thomas Klausner <wiz@NetBSD.org>
5411Date: Fri Jul 12 08:21:19 2013 +0200
5412
5413 Fix typo in header guard.
5414
5415 Signed-off-by: Thomas Klausner <wiz@NetBSD.org>
5416 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5417 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5418
5419commit ddc52b9cbae017f04c7a232af4e8b16c9f96823d
5420Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5421Date: Fri Jul 5 22:43:17 2013 -0700
5422
5423 Xephyr: Use _XEatDataWords (from Xlib 1.6) instead of _XEatData
5424
5425 Simplifies code and reduces risk of overflow from converting length
5426 field in X replies from words to bytes.
5427
5428 One call to _XEatData is left in ephyrHostGLXGetStringFromServer where
5429 it's already been checked for overflow, and other values have been
5430 subtracted from it to reduce the size of data remaining to be eaten.
5431
5432 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5433
5434commit a3d43edf71847f4b486f971405d2b457f81b73d1
5435Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5436Date: Fri Jul 5 22:35:32 2013 -0700
5437
5438 glxproxy: Use _XEatDataWords (from Xlib 1.6) instead of _XEatData
5439
5440 Reduces risk of overflow from converting length field in X replies
5441 from words to bytes. (Also seems to be what several calls were
5442 already incorrectly passing to _XEatData.)
5443
5444 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5445
5446commit 87b0cabc145a9b5f6faffdfb544ce1c647b8ab72
5447Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5448Date: Fri Jul 5 22:32:10 2013 -0700
5449
5450 glxproxy: Use _XReadPad instead of _XEatData to clean up the slop
5451
5452 Xlib already provides a function to eat padding bytes after the
5453 data read, so use it instead of calculating it ourselves.
5454
5455 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5456
5457commit 1cb182cbdc11fc1c97507c57875f1d2453f27328
5458Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5459Date: Mon Apr 15 08:41:14 2013 -0700
5460
5461 Xephyr: integer overflow in XF86DRIGetClientDriverName()
5462
5463 clientDriverNameLength is a CARD32 and needs to be bounds checked before
5464 adding one to it to come up with the total size to allocate, to avoid
5465 integer overflow leading to underallocation and writing data from the
5466 network past the end of the allocated buffer.
5467
5468 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
5469 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5470 Reviewed-by: Julien Cristau <jcristau@debian.org>
5471
5472commit 20644e53b385c54d73242c86a7d3f981d18a3843
5473Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5474Date: Mon Apr 15 08:39:03 2013 -0700
5475
5476 Xephyr: integer overflow in XF86DRIOpenConnection()
5477
5478 busIdStringLength is a CARD32 and needs to be bounds checked before adding
5479 one to it to come up with the total size to allocate, to avoid integer
5480 overflow leading to underallocation and writing data from the network past
5481 the end of the allocated buffer.
5482
5483 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
5484 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5485 Reviewed-by: Julien Cristau <jcristau@debian.org>
5486
5487commit c37ceda76bf9ec6f5166122e864663e10f106546
5488Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5489Date: Sun Apr 14 10:50:50 2013 -0700
5490
5491 Xephyr: integer overflow in ephyrHostGLXGetStringFromServer()
5492
5493 reply.length & reply.size are CARD32s and need to be bounds checked before
5494 multiplying or adding to come up with the total size to allocate, to avoid
5495 integer overflow leading to underallocation and writing data from the
5496 network past the end of the allocated buffer.
5497
5498 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
5499 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5500
5501commit 5fbd8c45b46ab93522e417240aa770466c30b735
5502Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5503Date: Sun Apr 14 09:45:26 2013 -0700
5504
5505 Xdmx: integer overflow in GetGLXFBConfigs()
5506
5507 numFBConfigs & numAttribs are CARD32s and need to be bounds checked before
5508 multiplying by structure sizes to come up with the total size to allocate,
5509 to avoid integer overflow leading to underallocation and writing data from
5510 the network past the end of the allocated buffer.
5511
5512 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
5513 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5514
5515commit c89bcd12f4fc7233830a8dbe7863c312f657da3c
5516Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5517Date: Sun Apr 14 09:33:57 2013 -0700
5518
5519 Xdmx: integer overflow in GetGLXVisualConfigs()
5520
5521 numVisuals & numProps are both CARD32 and need to be bounds checked before
5522 multiplying by structure sizes to come up with the total size to allocate,
5523 to avoid integer overflow leading to underallocation and writing data from
5524 the network past the end of the allocated buffer.
5525
5526 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
5527 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5528
5529commit 74469895e39fa38337f59edd64c4031ab9bb51d8
5530Author: Peter Hutterer <peter.hutterer@who-t.net>
5531Date: Mon Jul 15 13:06:28 2013 +1000
5532
5533 dix: allow a ConstantDeceleration between 0 and 1 (#66134)
5534
5535 A constant deceleration of x simply means (delta * 1/x). We limited that to
5536 values >= 1.0f for obvious reasons, but can also allow values from 0-1.
5537 That means that ConstantDeceleration is actually a ConstantAcceleration, but
5538 hey, if someone needs it...
5539
5540 X.Org Bug 66134 <http://bugs.freedesktop.org/show_bug.cgi?id=66134>
5541
5542 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5543 Reviewed-by: Keith Packard <keithp@keithp.com>
5544
5545commit 0e3be0b25fcfeff386bad132526352c2e45f1932
5546Author: Peter Hutterer <peter.hutterer@who-t.net>
5547Date: Fri Jul 12 14:10:10 2013 +1000
5548
5549 dix: UpdateTouchesForGrab must only free the listener grab if it is non-NULL
5550
5551 If a client calls XIGrabDevice in response to a ButtonPress event (regular
5552 event selection), the device will have a grab, but listener->grab is NULL.
5553
5554 Check for that, to avoid logspam.
5555
5556 [ 26293.863] (EE) BUG: triggered 'if (!pGrab)'
5557 [ 26293.863] (EE) BUG: grabs.c:256 in FreeGrab()
5558 [ 26293.863] (EE)
5559 [ 26293.863] (EE) Backtrace:
5560 [ 26293.864] (EE) 0: /usr/bin/Xorg (FreeGrab+0x54) [0x45d3fc]
5561 [ 26293.864] (EE) 1: /usr/bin/Xorg (UpdateTouchesForGrab+0x135) [0x447d4e]
5562 [ 26293.864] (EE) 2: /usr/bin/Xorg (ActivatePointerGrab+0x1ba) [0x447f3d]
5563 [ 26293.864] (EE) 3: /usr/bin/Xorg (GrabDevice+0x3e6) [0x4503bc]
5564 [ 26293.864] (EE) 4: /usr/bin/Xorg (ProcXIGrabDevice+0x1f9) [0x5981b1]
5565 [ 26293.865] (EE) 5: /usr/bin/Xorg (ProcIDispatch+0x78) [0x58aa17]
5566 [ 26293.865] (EE) 6: /usr/bin/Xorg (Dispatch+0x30d) [0x43347e]
5567 [ 26293.865] (EE) 7: /usr/bin/Xorg (main+0x61d) [0x498175]
5568 [ 26293.865] (EE) 8: /lib64/libc.so.6 (__libc_start_main+0xf5) [0x3df5621b75]
5569 [ 26293.865] (EE) 9: /usr/bin/Xorg (_start+0x29) [0x423a19]
5570 [ 26293.866] (EE) 10: ? (?+0x29) [0x29]
5571 [ 26293.866] (EE)
5572
5573 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5574 Reviewed-by: Keith Packard <keithp@keithp.com>
5575
5576commit 24cbb47ca80cd7a47749ece0ff3a98657ddf3ae3
5577Author: Peter Hutterer <peter.hutterer@who-t.net>
5578Date: Fri Jul 12 12:17:47 2013 +1000
5579
5580 config: provide example configuration for multi-seat setups
5581
5582 Seats other than seat0 need custom configuration. Provide that with a
5583 default configuration file so we can share it across distros.
5584
5585 This file intentionally does not end in .conf so it won't get picked up by
5586 the server after a normal installation. gdm, or whatever starts up the
5587 servers will have to explicitly specifiy this config file.
5588
5589 This file replaces the one currently written by systemd's multi-seat-x
5590 binary:
5591 http://cgit.freedesktop.org/systemd/systemd/tree/src/login/multi-seat-x.c
5592
5593 CC: Lennart Poettering <lennart@poettering.net>
5594 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5595 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
5596
5597commit 791121e00692e12cb4c38c8d018bd72eb5081eb7
5598Author: François Tigeot <ftigeot@wolfpond.org>
5599Date: Mon Jul 15 21:54:26 2013 +0200
5600
5601 Fix mouse header include on DragonFly and FreeBSD
5602
5603 * __FreeBSD_kernel_version doesn't exist anymore
5604
5605 * The removed check was for FreeBSD versions from before September 2000
5606 which are no longer supported anyway
5607
5608 * Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=66045
5609
5610 Signed-off-by: François Tigeot <ftigeot@wolfpond.org>
5611 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5612
5613commit 59a6d3f1eb03010ab831b32b03706a1d6143c732
5614Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5615Date: Thu Nov 8 13:41:13 2012 +0000
5616
5617 os: Remove any old logfile before trying to write to it
5618
5619 If we are not backing up logfiles, remove the old logfile before trying to write
5620 a new logfile, as otherwise the operation may fail if the previous logfile was
5621 created by a different user.
5622
5623 This change is useful when:
5624 - The DDX doesn't use the logfile backup mechanism (i.e. not Xorg)
5625 - The DDX is run by a non-root user, and then by a different non-root user
5626 - The logfile directory doesn't have the restricted-deletion flag set
5627
5628 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5629 Acked-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
5630 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
5631 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5632
5633commit 8eeaa74bc241acb41f1d3ed64971e0b01e794776
5634Author: Peter Hutterer <peter.hutterer@who-t.net>
5635Date: Tue Jul 9 13:27:19 2013 +1000
5636
5637 dix: when ungrabbing an active grab, accept pointer grabs (#66720)
5638
5639 Ungrabbing a device during an active touch grab rejects the grab. Ungrabbing
5640 a device during an active pointer grab accepts the grab.
5641
5642 Rejection is not really an option for a pointer-emulated grab, if a client
5643 has a button mask on the window it would get a ButtonPress emulated after
5644 UngrabDevice. That is against the core grab behaviour.
5645
5646 X.Org Bug 66720 <http://bugs.freedesktop.org/show_bug.cgi?id=66720>
5647
5648 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5649 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
5650
5651commit a2d6932ad48f2cbfcc4c5fb8272e47106d08ac78
5652Author: Julien Cristau <jcristau@debian.org>
5653Date: Mon Jul 1 19:39:30 2013 +0200
5654
5655 xkb: don't call atoi(NULL) when parsing argv
5656
5657 If the -ardelay or -arinterval options have no argument, there's no
5658 point trying to read it.
5659
5660 See
5661 http://www.forallsecure.com/bug-reports/feb3db57fc206d8df22ca53a6907f74973876272/
5662
5663 Reported-by: Alexandre Rebert <alexandre@cmu.edu>
5664 Signed-off-by: Julien Cristau <jcristau@debian.org>
5665 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5666 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5667
5668commit 77e51d5bbb97eb5c9d9dbff9a7c44d7e53620e68
5669Author: Eric Anholt <eric@anholt.net>
5670Date: Mon Jun 17 15:51:19 2013 -0700
5671
5672 Revert "DRI2: re-allocate DRI2 drawable if pixmap serial changes"
5673
5674 This reverts commit 3209b094a3b1466b579e8020e12a4f3fa78a5f3f. After a
5675 long debug session by Paul Berry, it appears that this was the commit
5676 that has been producing sporadic failures in piglit front buffer
5677 rendering tests for the last several years.
5678
5679 GetBuffers may return fresh buffers with invalid contents at a couple
5680 reasonable times:
5681
5682 - When first asked for a non-fake-front buffer.
5683 - When the drawable size is changed, an Invalidate has been sent, and
5684 obviously the app needs to redraw the whole buffer.
5685 - After a glXSwapBuffers(), GL allows the backbuffer to be undefined,
5686 and an Invalidate was sent to tell the GL that it should grab these
5687 appropriate new buffers to avoid stalling.
5688
5689 But with the patch being reverted, GetBuffers would also return fresh
5690 invalid buffers when the drawable serial number changed, which is
5691 approximately "whenever, for any reason". The app is not expecting
5692 invalid buffer contents "whenever", nor is it valid. Because the GL
5693 usually only GetBuffers after an Invalidate is sent, and the new
5694 buffer allocation only happened during a GetBuffers, most apps saw no
5695 problems. But apps that do (fake-)frontbuffer rendering do frequently
5696 ask the server for the front buffer (since we drop the fake front
5697 allocation when we're not doing front buffer rendering), and if the
5698 drawable serial got bumped midway through a draw, the server would
5699 pointlessly ditch the front *and* backbuffer full of important
5700 drawing, resulting in bad rendering.
5701
5702 The patch was originally to fix bugzilla:
5703 https://bugs.freedesktop.org/show_bug.cgi?id=28365
5704 Specifically:
5705
5706 To reproduce, start with a large-ish display (i.e. 1680x1050 on my
5707 laptop), use the patched glxgears from bug 28252 to add the
5708 -override option. Then run glxgears -override -geometry 640x480
5709 to create a 640x480 window in the top left corner, which will work
5710 fine. Next, run xrandr -s 640x480 and watch the fireworks.
5711
5712 I've tested with an override-redirect glxgears, both with vblank sync
5713 enabled and disabled, both with gnome-shell and no window manager at
5714 all, before and after this patch. The only problem observed was that
5715 before and after the revert, sometimes when alt-tabbing to kill my
5716 gears after completing the test gnome-shell would get confused about
5717 override-redirectness of the glxgears window (according to a log
5718 message) and apparently not bother doing any further compositing.
5719
5720 Signed-off-by: Eric Anholt <eric@anholt.net>
5721 Reviewed-by: Keith Packard <keithp@keithp.com>
5722 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
5723 Tested-by: Chris Wilson <chris@chris-wilson.co.uk>
5724 Signed-off-by: Keith Packard <keithp@keithp.com>
5725
5726commit 227015d6544115258b74c31b700f9bd163540e24
5727Merge: d9005d5 80054fc
5728Author: Keith Packard <keithp@keithp.com>
5729Date: Tue Jun 18 10:53:08 2013 -0700
5730
5731 Merge remote-tracking branch 'jturney/master'
5732
5733commit 80054fcafb095aff76f6df00d348c7fd85e8ddbc
5734Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5735Date: Sat Mar 30 19:44:14 2013 +0000
5736
5737 hw/xwin: Update manifest to target all architectures, not just x86
5738
5739 Update manifest to target all architectures, not just x86
5740
5741 Also: Write explicit dependencies for Xwin.rc. XWin.exe.manifest and X.ico are
5742 included the resource compiler output, so add a dependency on them to cause it
5743 to be recompiled if they change.
5744
5745 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5746 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5747
5748commit 689e7ad9b0fccd582cd63796cc90a999f141f589
5749Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5750Date: Sun Mar 31 18:36:37 2013 +0100
5751
5752 hw/xwin: Win32 Windows properties are of pointer type HANDLE, not an integer type
5753
5754 Win32 Windows properties are of pointer type HANDLE, not an integer type, but we
5755 use the Windows property WINDOW_WID_PROP to store the X window XID.
5756
5757 Add appropriate casts to show it doesn't matter that an XID is smaller than a
5758 pointer on 64-bit build.
5759
5760 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5761 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5762
5763commit e1a0945e8d4a3858838a14e64022524efd73c905
5764Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5765Date: Sun Mar 31 18:35:35 2013 +0100
5766
5767 hw/xwin: winMultiWindowGetTransientFor() returns an XID not a WindowPtr
5768
5769 winMultiWindowGetTransientFor() accesses the WM_TRANSIENT_FOR property, which is
5770 an X window XID (which we compare with WINDOW_WID_PROP) and not a WindowPtr.
5771
5772 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5773 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5774
5775commit be201ac257687355bf0916d7c525d598ed9b1282
5776Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5777Date: Sat Mar 30 19:47:16 2013 +0000
5778
5779 hw/xwin: Change winIconIsOverride() to take a HICON parameter
5780
5781 Change winIconIsOverride() to take a HICON parameter, so some unneccessary
5782 casts, which weren't portable to 64-bit Cygwin, can be removed.
5783
5784 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5785 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5786
5787commit fc40f9a92cf5981bccc45bbc0b959cc06c5accc2
5788Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5789Date: Sat Mar 30 19:41:06 2013 +0000
5790
5791 hw/xwin: Fix uses of CreateDialogParam() for 64-bit builds
5792
5793 Fix uses of CreateDialogParam() to be more strictly correct, for 64-bit builds
5794
5795 dwInitParam parameter is LPARAM, not int
5796 Return type of DLGPROC is INT_PTR, not WINBOOL
5797
5798 Also: Reporting the value of g_hDlgDepthChange is completely uninteresting, we
5799 are just interested in GetLastError() if it is NULL
5800
5801 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5802 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5803
5804commit b542976846eb5d4c82910272af32ec8298da4177
5805Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5806Date: Sat Mar 30 19:40:31 2013 +0000
5807
5808 hw/xwin: deprecated interface cygwin_conv_to_win32_path() doesn't exist in 64-bit cygwin
5809
5810 The deprecated interface cygwin_conv_to_win32_path() doesn't exist in 64-bit
5811 cygwin, but both the ID_ABOUT_CHANGELOG control and the referenced file who's
5812 path we are converting haven't existed for a while, so just remove this unused,
5813 cygwin-specific code.
5814
5815 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5816 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5817
5818commit 5a47c6420cfe298143310938f08cc36a503e8d3d
5819Author: Jon TURNEY <jon.turney@dronecode.org.uk>
5820Date: Mon Apr 8 01:27:26 2013 +0100
5821
5822 hw/xwin: Fix a 64-bit portability issue with casting HINSTANCE result of ShellExecute() to an integer
5823
5824 Fix a 64-bit portability issue with casting HINSTANCE result of ShellExecute()
5825 to an integer: Since HINSTANCE is a pointer type, use INT_PTR rather than int.
5826
5827 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
5828 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
5829
5830commit d9005d58c0413dcc3a99150c13d05163881cf782
5831Author: Peter Hutterer <peter.hutterer@who-t.net>
5832Date: Tue Jun 4 14:39:33 2013 +1000
5833
5834 config: add a quirks file for devices needing special configuration
5835
5836 This is the fedora quirks file configuring the special handling some devices
5837 need. Rather than keeping this distro-specific add it to the tree so we can
5838 share these quirks.
5839
5840 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5841 Reviewed-by: Jamey Sharp <jamey@minilop.net>
5842
5843commit 04ce6f6f6c880281e42c9ebf21d6f980ddae79b4
5844Author: Maarten Lankhorst <maarten.lankhorst@canonical.com>
5845Date: Thu May 30 11:23:07 2013 +0200
5846
5847 dix: add missing dixFreeRegistry define
5848
5849 Fixes a build failure on debian's udeb builds.
5850
5851 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
5852 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
5853 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5854
5855commit c21344add2fc589df83b29be5831c36a372201bd
5856Author: Peter Hutterer <peter.hutterer@who-t.net>
5857Date: Mon May 27 13:46:49 2013 +1000
5858
5859 dix: remove logspam in RefCursor()
5860
5861 This shouldn't have been in the patch
5862
5863 Reported-by: Colin Harrison <colin.harrison@virgin.net>
5864 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5865 Reviewed-by: Keith Packard <keithp@keithp.com>
5866 Signed-off-by: Keith Packard <keithp@keithp.com>
5867
5868commit cb3018d8a15add48efb3335e1f79b66378166008
5869Merge: c76a1b3 6589f3b
5870Author: Keith Packard <keithp@keithp.com>
5871Date: Thu May 23 19:58:46 2013 -0600
5872
5873 Merge remote-tracking branch 'whot/unreviewed'
5874
5875commit c76a1b343d6a56aa9529e87f0eda8d61355d562b
5876Merge: 891123c 9a5ad65
5877Author: Keith Packard <keithp@keithp.com>
5878Date: Thu May 23 19:58:36 2013 -0600
5879
5880 Merge remote-tracking branch 'whot/touch-grab-race-condition-56578-v3'
5881
5882commit 891123ce9dac5e52685b9a921b33d8279ba52956
5883Merge: 7e97166 49521ed
5884Author: Keith Packard <keithp@keithp.com>
5885Date: Thu May 23 10:30:15 2013 -0600
5886
5887 Merge remote-tracking branch 'whot/for-keith'
5888
5889commit 6589f3b55e335eef6c658c8ba1fe15a062f7e31c
5890Author: Peter Hutterer <peter.hutterer@who-t.net>
5891Date: Tue May 14 08:31:19 2013 +1000
5892
5893 dix: devices must have valuators before touch is initialized
5894
5895 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5896
5897commit 756ab88d93542f0589c9bf46f40ccc57df64f0fd
5898Author: Peter Hutterer <peter.hutterer@who-t.net>
5899Date: Fri Apr 26 15:10:08 2013 +1000
5900
5901 dix: fix device scaling to use a [min,max[ range.
5902
5903 defmin/defmax are screen coords and thus use a min-inclusive, max-exclusive
5904 range. device axes ranges are inclusive, so bump the max up by one to get the
5905 scaling right.
5906
5907 This fixes off-by-one coordinate errors if the coordinate matrix is used to
5908 bind the device to a fraction of the screen. It introduces an off-by-one
5909 scaling error in the device coordinate range, but since most devices have a
5910 higher resolution than the screen (e.g. a Wacom I4 has 5080 dpi) the effect
5911 of this should be limited.
5912
5913 This error manifests when we have numScreens > 1, as the scaling from
5914 desktop size back to screen size drops one device unit.
5915
5916 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5917
5918commit 5cc2c96f824dbb28b9f8da61efc41596f8bd0561
5919Author: Peter Hutterer <peter.hutterer@who-t.net>
5920Date: Sat Mar 9 20:43:51 2013 +1000
5921
5922 dix: pre-scale x by the screen:device:resolution ratio
5923
5924 commit 61a99aff9d33728a0b67920254d2d4d79f80cf39
5925 dix: pre-scale relative events from abs devices to desktop ratio (#31636)
5926
5927 added pre-scaling of relative coordinates coming from absolute devices to
5928 undo uneven scaling based on the screen dimensions.
5929
5930 Devices have their own device width/height ratio as well (in a specific
5931 resolution) and this must be applied for relative devices as well to avoid
5932 scaling of the relative events into the device's ratio.
5933
5934 e.g. a Wacom Intuos4 6x9 is in 16:10 format with equal horiz/vert
5935 resolution (dpi). A movement by 1000/1000 coordinates is a perfect diagonal
5936 on the tablet and must be reflected as such on the screen.
5937
5938 However, we map the relative device-coordinate events to absolute screen
5939 coordinates based on the axis ranges. This results in an effective scaling
5940 of 1000/(1000 * 1.6) and thus an uneven x/y axis movement - the y
5941 axis is always faster.
5942
5943 So we need to pre-scale not only by the desktop dimenstions but also by the
5944 device width/height ratio _and_ the resolution ratio.
5945
5946 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5947
5948commit 9a5ad65330693b3273972b63d10f2907d9ab954a
5949Author: Peter Hutterer <peter.hutterer@who-t.net>
5950Date: Wed May 15 19:01:11 2013 +1000
5951
5952 Abstract cursor refcounting
5953
5954 Too many callers relied on the refcnt being handled correctly. Use a simple
5955 wrapper to handle that case.
5956
5957 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5958
5959commit 35c2e263db01b2b61354298e5e85aa3cae8ac317
5960Author: Peter Hutterer <peter.hutterer@who-t.net>
5961Date: Tue May 14 14:51:31 2013 +1000
5962
5963 dix: call UpdateDeviceState() for emulated TouchEndEvents
5964
5965 ProcessTouchEvents() calls UDS for all touch events, but if the event type
5966 was switched to TouchUpdate(pending end) UDS is a noop.
5967
5968 Daniel Drake found this can cause stuck buttons if a touch grab is
5969 activated, rejected and the touch event is passed to a regular listener.
5970 This sequence causes the TouchEnd to be changed to TouchUpdate(pending end).
5971
5972 The actual TouchEnd event is later generated by the server once it is passed
5973 to the next listener. UDS is never called for this event, thus the button
5974 remains logically down.
5975
5976 A previous patch suggested for UDS to handle TouchUpdate events [1], however
5977 this would release the button when the first TouchEvent is processed, not
5978 when the last grab has been released (as is the case for sync pointer
5979 grabs). A client may thus have the grab on the device, receive a ButtonPress
5980 but see the button logically up in an XQueryPointer request.
5981
5982 This patch adds a call to UDS to TouchEmitTouchEnd(). The device state must
5983 be updated once a TouchEnd event was sent to the last grabbing listener and
5984 the number of grabs on the touchpoint is 0.
5985
5986 [1] http://patchwork.freedesktop.org/patch/13464/
5987
5988 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5989
5990commit 8b9dc2628115dcb3f3601ad19b1ae157df21b9ee
5991Author: Peter Hutterer <peter.hutterer@who-t.net>
5992Date: Tue May 14 07:46:25 2013 +1000
5993
5994 dix: devices must have valuators before touch is initialized
5995
5996 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5997
5998commit fd5ea0237db6d725a48f76b706135df9d3246b82
5999Author: Peter Hutterer <peter.hutterer@who-t.net>
6000Date: Mon May 13 15:22:12 2013 +1000
6001
6002 Xi: fix warning - remove unused 'rc'
6003
6004 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6005
6006commit 481702101b86fff003430e952dc65fb41eb56400
6007Author: Peter Hutterer <peter.hutterer@who-t.net>
6008Date: Fri May 3 15:07:58 2013 +1000
6009
6010 dix: fix cursor refcounting
6011
6012 The cursor is referenced during CopyGrab(), thus doesn't need to be handled
6013 manually anymore. It does need to be refcounted for temp grabs though.
6014
6015 The oldGrab handling in ProcGrabPointer is a leftover from the cursor in the
6016 grab being refcounted, but the grab itself being a static struct in the
6017 DeviceIntRec. Now that all grabs are copied, this lead to a double-free of
6018 the cursor (Reproduced in Thunderbird, dragging an email twice (or more
6019 often) causes a crash).
6020
6021 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6022
6023commit 49521edaf8930b335afac2185209e26e7000820b
6024Author: Peter Hutterer <peter.hutterer@who-t.net>
6025Date: Wed May 15 08:51:51 2013 +1000
6026
6027 ephyr: add -resizeable to the man page
6028
6029 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6030 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6031
6032commit 7a4bbfd9f2ce7aad23148ba01bb3f1071da606a3
6033Author: Peter Hutterer <peter.hutterer@who-t.net>
6034Date: Wed May 15 08:46:38 2013 +1000
6035
6036 ephyr: fix typo in man page
6037
6038 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6039 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6040
6041commit 7e97166c4fc38076b6224e4dbdd632b43444149f
6042Author: Geert Uytterhoeven <geert@linux-m68k.org>
6043Date: Tue May 14 13:04:53 2013 +0200
6044
6045 Shadow: Switch the Amiga/Atari bitplane code to the canonical X.Org license
6046
6047 Also add my name to the list of authors in COPYING
6048
6049 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6050 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
6051 Signed-off-by: Keith Packard <keithp@keithp.com>
6052
6053commit 3093f78d17e48a506aab170a9089cd10e21af299
6054Author: Peter Hutterer <peter.hutterer@who-t.net>
6055Date: Fri May 3 15:02:05 2013 +1000
6056
6057 dix: free the old grab when activating a new grab
6058
6059 A client may call XIGrabDevice twice, overwriting the existing grab. Thus,
6060 make sure we free the old copy after we copied it. Free it last, to make
6061 sure our refcounts don't run to 0 and inadvertantly free something on the
6062 way.
6063
6064 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6065
6066commit 2566bdd8bc996cccde77b846819808c6239a89d2
6067Author: Peter Hutterer <peter.hutterer@who-t.net>
6068Date: Wed Apr 24 14:46:06 2013 +1000
6069
6070 Xi: check for HAS_ACCEPTED only for grab listeners
6071
6072 If we have one listener left but it's not a grab, it cannot be in
6073 LISTENER_HAS_ACCEPTED state.
6074
6075 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6076
6077commit 5b00fc52270e9cfdfe7ac1838a21defe50fc3d31
6078Author: Peter Hutterer <peter.hutterer@who-t.net>
6079Date: Wed Apr 24 14:40:31 2013 +1000
6080
6081 Move TouchListenerGone call to CloseDownClient
6082
6083 TouchListenerGone cleans up if a client disappears. Having this in
6084 FreeGrab() triggers cyclic removal of grabs, emitting wrong events. In
6085 particular, it would clean up a passive grab record while that grab is
6086 active.
6087
6088 Move it to CloseDownClient() instead, cleaning up before we go.
6089
6090 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6091
6092commit 34c9b39d9937c2e19c2dffc9748605f90d40f965
6093Author: Peter Hutterer <peter.hutterer@who-t.net>
6094Date: Wed Apr 24 12:53:52 2013 +1000
6095
6096 dix: remove all listeners when freeing a touch
6097
6098 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6099
6100commit 395124bd2782823de37e5c5b2f15dba49cff05f6
6101Author: Peter Hutterer <peter.hutterer@who-t.net>
6102Date: Wed Apr 17 20:15:35 2013 +1000
6103
6104 dix: always copy grabs, don't reference them
6105
6106 Introduced in xorg-server-1.13.99.901-2-g9ad0fdb. Storing the grab pointer
6107 in the listener turns out to be a bad idea. If the grab is not an active
6108 grab or an implicit grab, the pointer stored is the one to the grab attached
6109 on the window. This grab may be removed if the client calls UngrabButton or
6110 similar while the touch is still active, leaving a dangling pointer.
6111
6112 To avoid this, copy the grab wherever we need to reference it later. This
6113 is also what we do for pointer/keyboard grabs, where we copy the grab as
6114 soon as it becomes active.
6115
6116 Reported-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
6117 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6118
6119commit 925e35122ebad877395bcf13676e9dbeb254bdfa
6120Author: Peter Hutterer <peter.hutterer@who-t.net>
6121Date: Tue Apr 23 15:52:18 2013 +1000
6122
6123 dix: AllocGrab can copy if an argument is passed in
6124
6125 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6126
6127commit 4980bcef9973ba1f90f53028f061669ee5d2661b
6128Author: Peter Hutterer <peter.hutterer@who-t.net>
6129Date: Tue Apr 23 15:46:04 2013 +1000
6130
6131 dix: freeing a null grab is a bug, complain if doing so
6132
6133 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6134
6135commit ccfa0f2d5de557546815a5e4f59552e2af46b578
6136Author: Peter Hutterer <peter.hutterer@who-t.net>
6137Date: Tue Apr 23 15:39:32 2013 +1000
6138
6139 dix: use a temporary variable for listeners[0]
6140
6141 no functional changes
6142
6143 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6144
6145commit 5363433a5cc64e2f83859aa1c32a89e5e1ddf9e4
6146Author: Peter Hutterer <peter.hutterer@who-t.net>
6147Date: Thu Apr 18 10:32:11 2013 +1000
6148
6149 dix: drop DeviceIntRec's activeGrab struct
6150
6151 Obsolete since 4bc2761ad5ec2d0668aec639780ffb136605fbc8. This struct
6152 existed so copying a passive grab could be simply done by
6153 activeGrab = *grab
6154
6155 and thus have a copy of the GrabPtr we'd get from various sources but still
6156 be able to check device->grab for NULL.
6157
6158 Since 4bc2761 activeGrab is a pointer itself and points to the same memory
6159 as grabinfo->grab, leaving us with the potential of dangling pointers if
6160 either calls FreeGrab() and doesn't reset the other one.
6161
6162 There is no reader of activeGrab anyway, so simply removing it is
6163 sufficient.
6164
6165 Note: field is merely renamed to keep the ABI. Should be removed in the
6166 future.
6167
6168 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6169
6170commit 7dbf61817d3bd4b1fc71710677e56c5d8cfcdb4e
6171Author: Peter Hutterer <peter.hutterer@who-t.net>
6172Date: Wed Apr 17 20:14:56 2013 +1000
6173
6174 dix: use a tmp variable for the to-be-removed touch listener
6175
6176 No functional changes.
6177
6178 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6179
6180commit a71a283934406d870bcd8dc376eb1c9ce1c8bbb4
6181Author: Peter Hutterer <peter.hutterer@who-t.net>
6182Date: Wed Apr 17 20:13:34 2013 +1000
6183
6184 dix: invert a loop condition
6185
6186 Change the single if condition in the loop body to a
6187 if (!foo) continue;
6188 and re-indent the rest.
6189
6190 No functional changes.
6191
6192 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6193
6194commit 5174b1f98204beee79eba74c4cda5f2be0a60a8f
6195Author: Peter Hutterer <peter.hutterer@who-t.net>
6196Date: Thu Feb 28 11:02:40 2013 +1000
6197
6198 dix: XAllowEvents() on a touch event means accepting it
6199
6200 A sync grab is the owner once it gets events. If it doesn't replay the
6201 event it will get all events from this touch, equivalent to accepting it.
6202
6203 If the touch has ended before XAllowEvents() is called, we also now need to
6204 send the TouchEnd event and clean-up since we won't see anything more from
6205 this touch.
6206
6207 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6208
6209commit e7f79c48b0faea910dc881034c00eb807fcd6210
6210Author: Peter Hutterer <peter.hutterer@who-t.net>
6211Date: Fri Mar 1 12:52:35 2013 +1000
6212
6213 dix: move EmitTouchEnd to touch.c
6214
6215 No functional changes, this just enables it to be re-used easier.
6216
6217 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6218
6219commit 0eb9390f6048049136347d5a5834d88bfc67cc09
6220Author: Peter Hutterer <peter.hutterer@who-t.net>
6221Date: Fri Mar 1 10:41:01 2013 +1000
6222
6223 Xi: Don't emit a TouchEnd event to a frozen device
6224
6225 EmitTouchEnd calls DeliverTouchEvents directly instead of through
6226 public.processInputProc. If a device is frozen, the TouchEnd is
6227 processed while the device is waiting for a XAllowEvents and thus ends the
6228 touch point (and the grab) before the client decided what to do with it. In
6229 the case of ReplayPointer, this loses the event.
6230
6231 This is a hack, but making EmitTouchEnd use processInputProc breaks
6232 approximately everything, especially the touch point is cleaned up during
6233 ProcessTouchEvents. Working around that is a bigger hack than this.
6234
6235 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6236
6237commit a7d989d335f903ffd8b168cd2beeb82c78d30c21
6238Author: Peter Hutterer <peter.hutterer@who-t.net>
6239Date: Thu Feb 28 13:07:26 2013 +1000
6240
6241 Xi: use public.processInputProc to replay the touch history
6242
6243 If a device is frozen in results to a grab, we need to enqueue the events.
6244
6245 This makes things complicated, and hard to follow since touch events are now
6246 replayed in the history, pushed into EnqueueEvent, then replayed later
6247 during PlayReleasedEvents in response to an XAllowEvents.
6248
6249 While the device is frozen, no touch events are processed, so if there is a
6250 touch client with ownership mask _below_ the grab this will delay the
6251 delivery and potentially screw gesture recognition. However, this is the
6252 behaviour we have already anyway if the top-most client is a sync pgrab or
6253 there is a sync grab active on the device when the TouchBegin was generated.
6254
6255 (also note, such a client would only reliably work in case of ReplayPointer
6256 anyway)
6257
6258 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6259
6260commit 214d11d3fcdac51fc7afbf3770516ec14f9e13c1
6261Author: Peter Hutterer <peter.hutterer@who-t.net>
6262Date: Wed Feb 27 15:05:54 2013 +1000
6263
6264 Xi: when punting to a new owner, always create TouchEnd events
6265
6266 If a touch is pending_finish and we just punted it to the next owner, that
6267 client must receive a TouchEnd event.
6268
6269 If we just punted to the last owner and that owner not a touch grab, we need
6270 to end the touch since this is the last event to be sent, and the client
6271 cannot accept/reject this.
6272
6273 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6274
6275commit 026627fe19aad007544eccf209f0dea05e67a7a7
6276Author: Peter Hutterer <peter.hutterer@who-t.net>
6277Date: Fri Mar 1 09:15:37 2013 +1000
6278
6279 Xi: save state for early acceptance
6280
6281 Delivering an event changes the state to LISTENER_IS_OWNER and we thus lose
6282 the information of early acceptance.
6283
6284 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6285
6286commit d905348134c80f19793eefb761731b00559ddf3a
6287Author: Peter Hutterer <peter.hutterer@who-t.net>
6288Date: Thu Feb 28 15:28:46 2013 +1000
6289
6290 Xi: if a passive async grab is activated from an emulated touch, accept
6291
6292 Async grabs cannot replay events, they cannot reject, so we can do an early
6293 accept here.
6294
6295 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6296
6297commit 8b0d21044956f3810199d5e2f38ce33069e97be7
6298Author: Peter Hutterer <peter.hutterer@who-t.net>
6299Date: Thu Feb 28 13:04:36 2013 +1000
6300
6301 Xi: fix lookup in ActivateEarlyAccept
6302
6303 ActivateEarlyAccept() can only be called from a grabbing client, so we can
6304 ignore the rest. And it's easy enough to get the client from that since
6305 9ad0fdb135a1c336771aee1f6eab75a6ad874aff.
6306
6307 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6308
6309commit d08bae297f9d7651edb1923d6b0d6b14b3d674fc
6310Author: Peter Hutterer <peter.hutterer@who-t.net>
6311Date: Thu Feb 28 13:08:27 2013 +1000
6312
6313 Xi: update the core listener state if we delivered the touch event
6314
6315 If a TouchBegin is sent to a core client, that client is now the owner.
6316
6317 By the time the TouchEnd is being processed, the client cannot replay
6318 anymore, so we can assume that this is the final touch end and we can clean
6319 up the touch record.
6320
6321 Note: DeliverTouchEmulatedEvent is called for all listeners and immediately
6322 bails out if the client is not the owner and thus shouldn't yet get the
6323 event. Thus, check the return code.
6324
6325 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6326
6327commit 81554d274f04951c55ea7f2e38d0455e2025e08d
6328Author: Peter Hutterer <peter.hutterer@who-t.net>
6329Date: Mon Feb 25 11:21:07 2013 +1000
6330
6331 Xi: if we delivered a TouchEnd to a passive grab, end it
6332
6333 ef64b5ee97099618cf2e2cbbd3e471095695ae24 (which introduced the
6334 TOUCH_CLIENT_ID check) has a wrong assumption that generated touch events
6335 (TOUCH_CLIENT_ID) should not terminate passive grabs.
6336 This is untrue, a TouchEnd may be generated in response to a TouchReject
6337 higher up. If we _deliver_ an event to a client, terminate the passive grab.
6338
6339 This requires us to count the actually delivered events too (first hunk).
6340
6341 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6342
6343commit 363b6387da6e669099a2da3861c73a29808295a6
6344Author: Peter Hutterer <peter.hutterer@who-t.net>
6345Date: Wed Feb 13 11:26:11 2013 +1000
6346
6347 dix: don't prepend an activated passive grab to the listeners
6348
6349 If the device is currently grabbed as the result of a passive grab
6350 activating, do not prepend that grab to the listeners (unlike active grabs).
6351 Otherwise, a client with a passive pointer grab will prevent touch grabs
6352 from activating higher up in the window stack.
6353
6354 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6355
6356commit 9cc45c18ad1511adf3fb163dd4cefbef106edb23
6357Author: Peter Hutterer <peter.hutterer@who-t.net>
6358Date: Wed Feb 13 10:49:23 2013 +1000
6359
6360 Xi: not having an ownership mask does not mean automatic acceptance
6361
6362 If we only have a single touch-grabbing client, setting the client as owner
6363 would clean up the touch once the TouchEnd was processed. If the client then
6364 calls XIAllowTouches() it will receive a BadValue for the touch ID (since
6365 the internal record is already cleaned up).
6366
6367 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6368
6369commit 2f1aedcaed8fd99b823d451bf1fb02330c078f67
6370Author: Peter Hutterer <peter.hutterer@who-t.net>
6371Date: Thu May 9 14:30:50 2013 +1000
6372
6373 input: print warnings if drivers don't initialize properly
6374
6375 If drivers supply incorrect values don't just quietly return False, spew to
6376 the log so we can detect what's going on. All these cases are driver bugs
6377 and should be fixed immediately.
6378
6379 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6380 Reviewed-by: Dave Airlie <airlied@redhat.com>
6381
6382commit 8a88b0ab52ba375ae84463a90503db88af10e368
6383Author: Peter Hutterer <peter.hutterer@who-t.net>
6384Date: Thu May 9 14:30:49 2013 +1000
6385
6386 dix: don't overwrite proximity/focus classes
6387
6388 InitPointerClassDeviceStruct/InitKeyboardDeviceStruct allocate a
6389 proximity/focus class, respectively. If a driver calls
6390 InitFocusClassDeviceStruct or InitProximityClassDeviceStruct beforehand,
6391 the previously allocated class is overwritten, leaking the memory.
6392
6393 Neither takes a parameter other than the device, so we can simply skip
6394 initialising it if we already have one.
6395
6396 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6397 Reviewed-by: Dave Airlie <airlied@redhat.com>
6398
6399commit 34b0d07ebf4a7874fe7fd336bef5bbdd8debda1c
6400Author: Peter Hutterer <peter.hutterer@who-t.net>
6401Date: Fri May 10 09:44:29 2013 +1000
6402
6403 dix: reset the OsBuffers after killing all clients
6404
6405 ==21860== 24 bytes in 1 blocks are still reachable in loss record 85 of 397
6406 ==21860== at 0x4C2B3F8: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
6407 ==21860== by 0x61ED93: AllocateOutputBuffer (io.c:1037)
6408 ==21860== by 0x61E15A: WriteToClient (io.c:764)
6409 ==21860== by 0x457B30: ProcQueryExtension (extension.c:275)
6410 ==21860== by 0x43596B: Dispatch (dispatch.c:432)
6411 ==21860== by 0x425DAB: main (main.c:295)
6412
6413 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6414 Reviewed-by: Dave Airlie <airlied@redhat.com>
6415
6416commit 67c66606c760c263d7a4c2d1bba43ed6225a4e7c
6417Author: Robert Morell <rmorell@nvidia.com>
6418Date: Thu May 9 13:09:02 2013 -0700
6419
6420 os: Reset input buffer's 'ignoreBytes' field
6421
6422 If a client sends a request larger than maxBigRequestSize, the server is
6423 supposed to ignore it.
6424
6425 Before commit cf88363d, the server would simply disconnect the client. After
6426 that commit, it attempts to gracefully ignore the request by remembering how
6427 long the client specified the request to be, and ignoring that many bytes.
6428 However, if a client sends a BigReq header with a large size and disconnects
6429 before actually sending the rest of the specified request, the server will
6430 reuse the ConnectionInput buffer without resetting the ignoreBytes field. This
6431 makes the server ignore new X clients' requests.
6432
6433 This fixes that behavior by resetting the ignoreBytes field when putting the
6434 ConnectionInput buffer back on the FreeInputs list.
6435
6436 Signed-off-by: Robert Morell <rmorell@nvidia.com>
6437 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
6438 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6439
6440commit ddc11397a56c745b5d1fb377e9d9b1fcc73802c8
6441Author: Peter Hutterer <peter.hutterer@who-t.net>
6442Date: Mon Feb 27 18:36:39 2012 +1000
6443
6444 dix: delete all callbacks before reset
6445
6446 DeleteCallbackManager() introduced for better symmetry in the caller, they
6447 do the same thing.
6448
6449 ==20085== 24 bytes in 1 blocks are still reachable in loss record 11 of 103
6450 ==20085== at 0x4C2A4CD: malloc (vg_replace_malloc.c:236)
6451 ==20085== by 0x43A097: CreateCallbackList (dixutils.c:837)
6452 ==20085== by 0x43A1D3: AddCallback (dixutils.c:869)
6453 ==20085== by 0x4B1736: GEExtensionInit (geext.c:209)
6454 ==20085== by 0x41C8A8: InitExtensions (miinitext.c:389)
6455 ==20085== by 0x5AC918: main (main.c:208)
6456
6457 ==2042== 8 bytes in 1 blocks are still reachable in loss record 2 of 97
6458 ==2042== at 0x4C2A4CD: malloc (vg_replace_malloc.c:236)
6459 ==2042== by 0x4C2A657: realloc (vg_replace_malloc.c:525)
6460 ==2042== by 0x4802F5: XNFrealloc (utils.c:1095)
6461 ==2042== by 0x43A17A: CreateCallbackList (dixutils.c:855)
6462 ==2042== by 0x43A1EF: AddCallback (dixutils.c:870)
6463 ==2042== by 0x4B1752: GEExtensionInit (geext.c:209)
6464 ==2042== by 0x41C8A8: InitExtensions (miinitext.c:389)
6465 ==2042== by 0x5AC9E4: main (main.c:208)
6466 ==2042==
6467
6468 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6469 Reviewed-by: Keith Packard <keithp@keithp.com>
6470
6471commit 6993f8b459e25b1a7a0a03e209688a28ce6c2c56
6472Author: Peter Hutterer <peter.hutterer@who-t.net>
6473Date: Wed Apr 24 14:59:11 2013 +1000
6474
6475 Xi: free barrier code at reset time
6476
6477 ==29423== 16 bytes in 1 blocks are definitely lost in loss record 73 of 328
6478 ==29423== at 0x4A06B6F: calloc (vg_replace_malloc.c:593)
6479 ==29423== by 0x5987C0: XIBarrierInit (xibarriers.c:908)
6480 ==29423== by 0x58F370: XInputExtensionInit (extinit.c:1300)
6481 ==29423== by 0x4F33C3: InitExtensions (miinitext.c:337)
6482 ==29423== by 0x4997DB: main (main.c:208)
6483
6484 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6485 Reviewed-by: Keith Packard <keithp@keithp.com>
6486
6487commit 042c6d861f7bb7038ddcdd6b59766fd9094d0e52
6488Author: Peter Hutterer <peter.hutterer@who-t.net>
6489Date: Tue Apr 30 15:44:37 2013 +1000
6490
6491 os: Use ErrorFSigSafe from FatalError and it's friends
6492
6493 Backtrace logging etc. is already sigsafe, but the actual FatalError message
6494 in response is not yet, leading to amusing logs like this:
6495
6496 (EE) Segmentation fault at address 0x0
6497 (EE) BUG: triggered 'if (inSignalContext)'
6498 (EE) BUG: log.c:499 in LogVMessageVerb()
6499 (EE) Warning: attempting to log data in a signal unsafe manner while in
6500 signal context.
6501 Please update to check inSignalContext and/or use LogMessageVerbSigSafe() or
6502 ErrorFSigSafe().
6503 The offending log format message is:
6504
6505 Fatal server error:
6506
6507 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6508 Reviewed-by: Keith Packard <keithp@keithp.com>
6509
6510commit d9848fb4b182ca21bacf28ed7410d1a502cb000e
6511Author: Peter Hutterer <peter.hutterer@who-t.net>
6512Date: Thu Feb 14 16:24:53 2013 +1000
6513
6514 os: complain about unsupported pnprintf directives
6515
6516 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6517 Reviewed-by: Keith Packard <keithp@keithp.com>
6518
6519commit d903d17d7f006fa333265b8476063b189c20d082
6520Author: Peter Hutterer <peter.hutterer@who-t.net>
6521Date: Thu Feb 14 16:19:34 2013 +1000
6522
6523 os: support %c in pnprintf
6524
6525 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6526 Reviewed-by: Keith Packard <keithp@keithp.com>
6527
6528commit 58ef34ee6d0f68aa28f6f1a26e56f49ec85ed9bf
6529Author: Peter Hutterer <peter.hutterer@who-t.net>
6530Date: Thu Feb 14 16:13:22 2013 +1000
6531
6532 os: support %% in pnprintf
6533
6534 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6535 Reviewed-by: Keith Packard <keithp@keithp.com>
6536
6537commit 5ea21560dd071ea4ab87430000d087fd5fe1f092
6538Author: Peter Hutterer <peter.hutterer@who-t.net>
6539Date: Thu Feb 14 15:34:32 2013 +1000
6540
6541 os: support pnprintf length modifiers for integers
6542
6543 Mainly for %ld, smaller than int is propagated anyway, and %lld isn't really
6544 used.
6545
6546 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6547 Reviewed-by: Keith Packard <keithp@keithp.com>
6548
6549commit d3d4af5f9e505d444b6c82c6ea238206433fd24b
6550Author: Peter Hutterer <peter.hutterer@who-t.net>
6551Date: Mon Feb 27 18:59:45 2012 +1000
6552
6553 dix: reset the registry before quitting
6554
6555 Heaps of these:
6556 ==2042== 15,360 bytes in 120 blocks are still reachable in loss record 94 of
6557 97
6558 ==2042== at 0x4C2A4CD: malloc (vg_replace_malloc.c:236)
6559 ==2042== by 0x4C2A657: realloc (vg_replace_malloc.c:525)
6560 ==2042== by 0x45FB91: double_size (registry.c:65)
6561 ==2042== by 0x45FC97: RegisterRequestName (registry.c:85)
6562 ==2042== by 0x460095: RegisterExtensionNames (registry.c:179)
6563 ==2042== by 0x460729: dixResetRegistry (registry.c:334)
6564 ==2042== by 0x5AC992: main (main.c:201)
6565
6566 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6567 Reviewed-by: Keith Packard <keithp@keithp.com>
6568
6569commit 6f44d672aa34d343f63f0ea81ad58154a66b57ec
6570Author: Peter Hutterer <peter.hutterer@who-t.net>
6571Date: Mon Feb 27 17:17:41 2012 +1000
6572
6573 xkb: free XkbRulesUsed and XkbRulesDflt on extension cleanup
6574
6575 ==2547== 1 bytes in 1 blocks are still reachable in loss record 1 of 111
6576 ==2547== at 0x4C2A4CD: malloc (vg_replace_malloc.c:236)
6577 ==2547== by 0x64D1551: strdup (strdup.c:43)
6578 ==2547== by 0x4802FB: Xstrdup (utils.c:1113)
6579 ==2547== by 0x585B6C: XkbSetRulesUsed (xkbInit.c:219)
6580 ==2547== by 0x58700F: InitKeyboardDeviceStruct (xkbInit.c:595)
6581 ==2547== by 0x419FA3: vfbKeybdProc (InitInput.c:74)
6582 ==2547== by 0x425A3D: ActivateDevice (devices.c:540)
6583 ==2547== by 0x425F65: InitAndStartDevices (devices.c:713)
6584 ==2547== by 0x5ACA57: main (main.c:259)
6585
6586 and a few more of the above.
6587
6588 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6589 Reviewed-by: Keith Packard <keithp@keithp.com>
6590
6591commit 623981ddaeb8836f3b0939e527c0e943f9c4e974
6592Author: Peter Hutterer <peter.hutterer@who-t.net>
6593Date: Fri Feb 1 09:06:20 2013 +1000
6594
6595 If neither HAL nor udev backends are enabled, warn the user
6596
6597 If both are missing, input device hotplugging will not work out of the box.
6598 While we still have a DBus-API or the user may want to set AAD off all the
6599 time, the most likely source of this is misconfiguration (i.e. lack of the
6600 udev/hal devel packages).
6601
6602 Message printed last to make it more obvious to the user.
6603
6604 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6605 Reviewed-by: Keith Packard <keithp@keithp.com>
6606
6607commit dce40e2266200421647044ec7e856656d2ef952d
6608Author: Peter Hutterer <peter.hutterer@who-t.net>
6609Date: Tue Feb 26 12:34:21 2013 +1000
6610
6611 Xi: always return BadMatch for XTest devices ChangeDeviceControl requests
6612
6613 The only controls that still do something are DEVICE_RESOLUTION and
6614 DEVICE_ENABLE.
6615
6616 XTest devices have no resolution to change, and they cannot be disabled. So
6617 skip the lot, and prevent a crash in the DDX when it's trying to
6618 de-reference pInfo->control_proc on device with no pInfo struct.
6619
6620 Likewise, don't allow setting device mode or the valuators.
6621 XTest pointers are always relative, they don't have a mode.
6622
6623 Test cases:
6624 xts5/XI/ChangeDeviceControl (1/10)
6625 xts5/XI/SetDeviceValuators (1/6)
6626 and a few others
6627
6628 Reported-by: Knut Petersen <Knut_Petersen@t-online.de>
6629 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6630 Reviewed-by: Keith Packard <keithp@keithp.com>
6631
6632commit 23d1bc69f305edd5a6e2cfec3dfc84befda0881c
6633Author: Peter Hutterer <peter.hutterer@who-t.net>
6634Date: Thu Mar 14 11:07:57 2013 +1000
6635
6636 dix: send the current axis value in DeviceChangedEvents (#62321)
6637
6638 X.Org Bug 62321 <http://bugs.freedesktop.org/show_bug.cgi?id=62321>
6639
6640 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6641 Reviewed-by: Keith Packard <keithp@keithp.com>
6642
6643commit 27b11f5c19d67e2f6784d86cb7df86bbab3b58bb
6644Author: Peter Hutterer <peter.hutterer@who-t.net>
6645Date: Fri Feb 15 11:19:10 2013 +1000
6646
6647 Xi: fix comment - XI2 grabs aren't keysym grabs
6648
6649 Comment dates back to a pre-release version of XI2 that supported keysym
6650 grabs. That never made it into a release, it was ditched before.
6651
6652 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6653 Reviewed-by: Keith Packard <keithp@keithp.com>
6654
6655commit 2746c681639f9512e6e45fb8d0151b996b6aff7f
6656Merge: 1e6cf8e 9878e09
6657Author: Keith Packard <keithp@keithp.com>
6658Date: Mon May 6 10:52:40 2013 -0700
6659
6660 Merge remote-tracking branch 'alanc/master'
6661
6662commit 1e6cf8ec20d07b73a11116564aba71b4e4291dcd
6663Merge: dbfeaf7 5860408a
6664Author: Keith Packard <keithp@keithp.com>
6665Date: Mon May 6 10:20:21 2013 -0700
6666
6667 Merge remote-tracking branch 'jturney/unused-but-set-variable-warning-fix'
6668
6669commit dbfeaf70623a83e1a3f3255c94d52e0e04702837
6670Author: Aaron Plattner <aplattner@nvidia.com>
6671Date: Tue Apr 30 14:30:18 2013 -0700
6672
6673 xfree86: don't enable anything in xf86InitialConfiguration for GPU screens
6674
6675 There's no point in turning on outputs connected to GPU screens during initial
6676 configuration. Not only does this cause them to just display black, it also
6677 confuses clients when these screens are attached to a master screen and RandR
6678 reports that the outputs are already on.
6679
6680 Also, don't print the warning about no outputs being found on GPU screens,
6681 since that's expected.
6682
6683 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
6684 Reviewed-by: Dave Airlie <airlied@gmail.com>
6685
6686commit 5860408a19d956f8cebb7ae39cfdbfcebd5e6360
6687Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6688Date: Wed Apr 3 18:43:26 2013 +0100
6689
6690 hw/xwin: Fix unused-but-set-variable warning in ProcWindowsWMFrameGetRect()
6691
6692 /jhbuild/checkout/xorg/xserver/hw/xwin/winwindowswm.c: In function ‘ProcWindowsWMFrameGetRect’:
6693 /jhbuild/checkout/xorg/xserver/hw/xwin/winwindowswm.c:322:12: error: variable ‘ir’ set but not used [-Werror=unused-but-set-variable]
6694
6695 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6696 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6697
6698commit bcfd523c150b34e9304d174ba812160c611f270d
6699Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6700Date: Wed Apr 3 18:34:38 2013 +0100
6701
6702 hw/xwin: Fix unused-but-set-variable warning in winMWExtWMUpdateWindowDecoration()
6703
6704 /jhbuild/checkout/xorg/xserver/hw/xwin/winwin32rootlesswindow.c: In function ‘winMWExtWMUpdateWindowDecoration’:
6705 /jhbuild/checkout/xorg/xserver/hw/xwin/winwin32rootlesswindow.c:189:11: error: variable ‘dwStyle’ set but not used [-Werror=unused-but-set-variable]
6706
6707 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6708 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6709
6710commit fbf819c24dc080f166cff29bf46b0feb604c6b8c
6711Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6712Date: Wed Apr 3 16:40:35 2013 +0100
6713
6714 hw/xwin: Fix unused-but-set-variable warning in winHotKeyAltTabPrimaryDD()
6715
6716 /jhbuild/checkout/xorg/xserver/hw/xwin/winpfbdd.c: In function ‘winHotKeyAltTabPrimaryDD’:
6717 /jhbuild/checkout/xorg/xserver/hw/xwin/winpfbdd.c:518:20: error: variable ‘rcSrc’ set but not used [-Werror=unused-but-set-variable]
6718
6719 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6720 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6721
6722commit 7dae1e59ce6077f475c04a41fa00096a74114064
6723Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6724Date: Wed Apr 3 16:33:39 2013 +0100
6725
6726 hw/xwin: Fix unused-but-set-variable warning in winMinimizeWindow() when built !XWIN_MULTIWINDOWEXTWM
6727
6728 /jhbuild/checkout/xorg/xserver/hw/xwin/winmultiwindowwindow.c: In function ‘winMinimizeWindow’:
6729 /jhbuild/checkout/xorg/xserver/hw/xwin/winmultiwindowwindow.c:813:20: error: variable ‘pScreenInfo’ set but not used [-Werror=unused-but-set-variable]
6730
6731 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6732 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6733
6734commit b284666f09b77f9c2f048fee22015a353152a57f
6735Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6736Date: Wed Apr 3 16:27:45 2013 +0100
6737
6738 hw/xwin: Fix unused-but-set-variable warning in winClipboardProc()
6739
6740 /jhbuild/checkout/xorg/xserver/hw/xwin/winclipboardthread.c: In function ‘winClipboardProc’:
6741 /jhbuild/checkout/xorg/xserver/hw/xwin/winclipboardthread.c:88:25: error: variable ‘atomClipboardManager’ set but not used [-Werror=unused-but-set-variable]
6742
6743 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6744 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6745
6746commit d414a09bb7f8d4440b36cbc6f0b4ca0eaefe5cbf
6747Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6748Date: Wed Apr 3 16:25:54 2013 +0100
6749
6750 hw/xwin: Fix unused-but-set-variable warning in winWindowProc()
6751
6752 /jhbuild/checkout/xorg/xserver/hw/xwin/winwndproc.c: In function ‘winWindowProc’:
6753 /jhbuild/checkout/xorg/xserver/hw/xwin/winwndproc.c:65:22: error: variable ‘s_hInstance’ set but not used [-Werror=unused-but-set-variable]
6754
6755 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6756 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6757
6758commit 7726102671690eb28138d032c8398ac903b97606
6759Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6760Date: Wed Apr 3 16:24:08 2013 +0100
6761
6762 hw/xwin: Fix unused-but-set-variable warning in winCopyWindowNativeGDI()
6763
6764 /jhbuild/checkout/xorg/xserver/hw/xwin/winwindow.c: In function ‘winCopyWindowNativeGDI’:
6765 /jhbuild/checkout/xorg/xserver/hw/xwin/winwindow.c:131:15: error: variable ‘pwinRoot’ set but not used [-Werror=unused-but-set-variable]
6766
6767 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6768 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6769
6770commit b9a2566a6e2f56a0335cd4a089b2066c06a67455
6771Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6772Date: Wed Apr 3 16:50:38 2013 +0100
6773
6774 hw/xwin: Fix unused-but-set-variable warning in winFinishScreenInitFB()
6775
6776 /jhbuild/checkout/xorg/xserver/hw/xwin/winscrinit.c:264:11: error: variable ‘pbits’ set but not used [-Werror=unused-but-set-variable]
6777
6778 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6779 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6780
6781commit c97fbd39adbc0a8cb8537f40b12c091aee29221a
6782Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6783Date: Wed Apr 3 16:18:27 2013 +0100
6784
6785 hw/xwin: Fix unused-but-set-variable warning in winIsFakeCtrl_L()
6786
6787 /jhbuild/checkout/xorg/xserver/hw/xwin/winkeybd.c:331:17: error: variable ‘lastMessage’ set but not used [-Werror=unused-but-set-variable]
6788
6789 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6790 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6791
6792commit cdb74fe17dcb3b5a9c61951f897deaf07f86775f
6793Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6794Date: Wed Apr 3 16:16:06 2013 +0100
6795
6796 hw/xwin: Fix unused-but-set-variable warnings in windialog.c
6797
6798 /jhbuild/checkout/xorg/xserver/hw/xwin/windialogs.c: In function ‘winChangeDepthDlgProc’:
6799 /jhbuild/checkout/xorg/xserver/hw/xwin/windialogs.c:432:22: error: variable ‘s_pScreen’ set but not used [-Werror=unused-but-set-variable]
6800 /jhbuild/checkout/xorg/xserver/hw/xwin/windialogs.c: In function ‘winAboutDlgProc’:
6801 /jhbuild/checkout/xorg/xserver/hw/xwin/windialogs.c:564:22: error: variable ‘s_pScreen’ set but not used [-Werror=unused-but-set-variable]
6802
6803 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6804 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
6805
6806commit f2fd8ec3725a61abbc831f0a9ec28fa2b7020c47
6807Author: Dave Airlie <airlied@redhat.com>
6808Date: Wed Jan 9 12:52:13 2013 +1000
6809
6810 gpu: call CreateScreenResources for GPU screens
6811
6812 I didn't think we needed this before, but after doing some more
6813 work with reverse optimus it seems like it should be called.
6814
6815 Reviewed-by: Keith Packard <keithp@keithp.com>
6816 Signed-off-by: Dave Airlie <airlied@redhat.com>
6817
6818commit 8fcb9d91b69abc72ddef31b9f2e8585580c6cad2
6819Author: Dave Airlie <airlied@redhat.com>
6820Date: Wed Jan 9 12:52:08 2013 +1000
6821
6822 dix: allow pixmap dirty helper to be used for non-shared pixmaps
6823
6824 this allows the pixmap dirty helper to be used for reverse optimus,
6825 where the GPU wants to copy from the shared pixmap to its VRAM copy.
6826
6827 [airlied: slave_dst is wrong name now but pointless ABI churn at this point]
6828 Reviewed-by: Keith Packard <keithp@keithp.com>
6829 Signed-off-by: Dave Airlie <airlied@redhat.com>
6830
6831commit 16077b81c502e04d77f81f683e0c213b9fe75393
6832Author: Dave Airlie <airlied@redhat.com>
6833Date: Wed Jan 9 12:52:03 2013 +1000
6834
6835 xf86crtc: don't use scrn->display for gpu screens
6836
6837 scrn->display is a property of the main screen really, and we don't
6838 want to have the GPU screens use it for anything when picking modes
6839 or a front buffer size.
6840
6841 This fixes a bug where when you plugged a display link device, it
6842 would try and allocate a screen the same size as the current running
6843 one (3360x1050 in this case), which was too big for the device. Avoid
6844 doing this and just pick sizes based on whats plugged into this device.
6845
6846 Reviewed-by: Keith Packard <keithp@keithp.com>
6847 Signed-off-by: Dave Airlie <airlied@redhat.com>
6848
6849commit f0d0d75bfe62553dde353f89e46ff13dd863fbe8
6850Author: Dave Airlie <airlied@redhat.com>
6851Date: Wed Jan 9 12:51:55 2013 +1000
6852
6853 dix/gpu: remove asserts for output/offload from same slave
6854
6855 We should have no problem allowing output/offload from the same slave,
6856 I asserted here, but in order to implement reverse optimus this makes
6857 perfect sense. (reverse optimus is intel outputting to nvidia).
6858
6859 Reviewed-by: Keith Packard <keithp@keithp.com>
6860 Signed-off-by: Dave Airlie <airlied@redhat.com>
6861
6862commit 9d26e8eaf5a2d7c3e65670ac20254c60f665c463
6863Author: Dave Airlie <airlied@redhat.com>
6864Date: Wed Jan 9 14:26:35 2013 +1000
6865
6866 randr: report changes when we disconnect a GPU slave
6867
6868 When we disconnect an output/offload slave set the changed bits,
6869 so a later TellChanged can do something.
6870
6871 Then when we remove a GPU slave device, sent change notification
6872 to the protocol screen.
6873
6874 This allows hot unplugged USB devices to disappear in clients.
6875
6876 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
6877 Signed-off-by: Dave Airlie <airlied@redhat.com>
6878
6879commit b724324252d13ff95f62eebd12d125b194d2ccc2
6880Author: Dave Airlie <airlied@redhat.com>
6881Date: Wed Jan 9 14:25:43 2013 +1000
6882
6883 randr: only respected changed on the protocol screen
6884
6885 We don't want to know about changes on the non-protocol screen,
6886 we will fix up setchanged to make sure non-protocol screens update
6887 the protocol screens when they have a change.
6888
6889 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
6890 Signed-off-by: Dave Airlie <airlied@redhat.com>
6891
6892commit b3f70f38edebac87afe9351538365292f1aaaff3
6893Author: Dave Airlie <airlied@redhat.com>
6894Date: Wed Jan 9 14:29:47 2013 +1000
6895
6896 randr: make SetChanged modify the main protocol screen not the gpu screen
6897
6898 When SetChanged is called we now modify the main protocol screen,
6899 not the the gpu screen. Since changed stuff should work at the protocol level.
6900
6901 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
6902 Signed-off-by: Dave Airlie <airlied@redhat.com>
6903
6904commit f9c8248b8326ad01f33f31531c6b2479baf80f02
6905Author: Dave Airlie <airlied@redhat.com>
6906Date: Wed Jan 9 14:23:57 2013 +1000
6907
6908 randr: don't directly set changed bits in randr screen
6909
6910 Introduce a wrapper interface so we can fix things up for multi-gpu
6911 situations later.
6912
6913 This just introduces the API for now.
6914
6915 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
6916 Signed-off-by: Dave Airlie <airlied@redhat.com>
6917
6918commit 451ba4bd41b82acd4aec6236ba121e00cfeb311b
6919Author: Chris Wilson <chris@chris-wilson.co.uk>
6920Date: Thu Jan 10 03:26:33 2013 +0000
6921
6922 hw/xfree86: Only report SetDesiredModes() failed if at least one modeset fails
6923
6924 commit 6703a7c7cf1a349c137e247a0c8eb462ff7b07be
6925 Author: Keith Packard <keithp@keithp.com>
6926 Date: Tue Jan 8 20:24:32 2013 -0800
6927
6928 hw/xfree86: Require only one working CRTC to start the server.
6929
6930 changed the logic to try to set the mode on all connected outputs rather
6931 than abort upon the first failure. The return error code was then
6932 tweaked such that it reported success if it set a mode on any crtc.
6933 However, this confuses the headless case where we never enable any crtcs
6934 and also, importantly, never fail to set a crtc.
6935
6936 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=59190
6937
6938 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
6939 Also-written-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
6940 Reviewed-by: Keith Packard <keithp@keithp.com>
6941 Signed-off-by: Keith Packard <keithp@keithp.com>
6942
6943commit 9878e097a7de2f86eff0dcfd9fe5d83b162197ec
6944Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6945Date: Wed Apr 24 15:24:31 2013 -0700
6946
6947 Only call xf86platformVTProbe() when it's defined
6948
6949 Fixes build on non-udev systems, since XSERVER_PLATFORM_BUS is only
6950 defined in configure.ac if $CONFIG_UDEV_KMS is true.
6951
6952 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6953 Reviewed-by: Dave Airlie <airlied@redhat.com>
6954
6955commit e779402d531625ba4ed5d628d06c6b1e06b2ec7b
6956Author: Keith Packard <keithp@keithp.com>
6957Date: Thu Apr 25 09:02:31 2013 -0700
6958
6959 shadow: Define c2p_unsupported
6960
6961 Just make this call BUG_WARN(1) to indicate that something unexpected
6962 happened
6963
6964 Signed-off-by: Keith Packard <keithp@keithp.com>
6965
6966commit 2b361fbda536f0125e4b87a6d455acc58f4e8690
6967Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6968Date: Sun Feb 10 10:24:59 2013 -0800
6969
6970 sparcPromPathname2Node: free name when returning error, instead of leaking it
6971
6972 Reported with other leaks found by cppcheck in bugzilla #50281
6973 https://bugs.freedesktop.org/show_bug.cgi?id=50281
6974
6975 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6976 Reviewed-by: Keith Packard <keithp@keithp.com>
6977
6978commit 174ccd84931ece5a92a09c4a1d6a47e0958ebf7e
6979Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6980Date: Sun Feb 10 10:24:59 2013 -0800
6981
6982 xf86SbusCmapLoadPalette: Delay malloc until needed, avoiding leak on error
6983
6984 Reported with other leaks found by cppcheck in bugzilla #50281
6985 https://bugs.freedesktop.org/show_bug.cgi?id=50281
6986
6987 V2: check for malloc failure
6988
6989 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6990
6991commit 6bca0184d167388cd417d113031317990489987d
6992Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6993Date: Sun Feb 10 10:18:02 2013 -0800
6994
6995 dmxVDLRead: if we opened a file, close it instead of leaking it
6996
6997 Reported with other leaks found by cppcheck in bugzilla #50281
6998 https://bugs.freedesktop.org/show_bug.cgi?id=50281
6999
7000 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7001 Reviewed-by: Keith Packard <keithp@keithp.com>
7002
7003commit 8c5ff2e93f73b24adff741b25fc1e31b2f5dd0ac
7004Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7005Date: Sun Feb 10 10:24:59 2013 -0800
7006
7007 __glXDRIscreenProbe: free screen when DRI2Connect fails, instead of leaking it
7008
7009 Reported with other leaks found by cppcheck in bugzilla #50281
7010 https://bugs.freedesktop.org/show_bug.cgi?id=50281
7011
7012 V2: goto existing error handler, instead of replicating more of it here
7013
7014 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7015 Reviewed-by: Keith Packard <keithp@keithp.com>
7016
7017commit 9567fa9a7269765bc59ac71b931e16df261d794d
7018Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7019Date: Sun Feb 10 10:24:59 2013 -0800
7020
7021 __glXDRIscreenCreateContext: free context on failure, instead of leaking it
7022
7023 Reported with other leaks found by cppcheck in bugzilla #50281
7024 https://bugs.freedesktop.org/show_bug.cgi?id=50281
7025
7026 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7027 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
7028
7029commit 7ab98bafc9a3426fd40f8ae693430491333ba4fc
7030Merge: 53da26a 5ab2603
7031Author: Keith Packard <keithp@keithp.com>
7032Date: Wed Apr 24 14:14:45 2013 -0700
7033
7034 Merge remote-tracking branch 'geertu/master'
7035
7036commit 53da26afb76bcce70b91c5df23d25e7e85c010cf
7037Merge: 5ece86e d8d3c78
7038Author: Keith Packard <keithp@keithp.com>
7039Date: Wed Apr 24 10:27:19 2013 -0700
7040
7041 Merge remote-tracking branch 'whot/for-keith'
7042
7043commit 5ece86e921a1df457a9baab6a2357ccda4ad6957
7044Merge: ef0a726 22cab8a
7045Author: Keith Packard <keithp@keithp.com>
7046Date: Wed Apr 24 10:23:51 2013 -0700
7047
7048 Merge remote-tracking branch 'airlied/for-keithp-gpu-vt-owner'
7049
7050commit ef0a726bc8d5c415244cb4f86d664938b0127a5d
7051Author: Jeremy White <jwhite@codeweavers.com>
7052Date: Thu Mar 21 10:58:42 2013 -0500
7053
7054 Eliminate the use of xf86Rename.h
7055
7056 Signed-off-by: Jeremy White <jwhite@codeweavers.com>
7057 Reviewed-by: Keith Packard <keithp@keithp.com>
7058 Signed-off-by: Keith Packard <keithp@keithp.com>
7059
7060commit 2b539239e465faa9e8c6d96f1c6f00e593d46c18
7061Author: Keith Packard <keithp@keithp.com>
7062Date: Wed Apr 24 10:11:03 2013 -0700
7063
7064 Bump to 1.14.99.1
7065
7066 Snapshot with fix for CVE-2013-1940.
7067
7068 Signed-off-by: Keith Packard <keithp@keithp.com>
7069
7070commit d8d3c78b6ebdf63836951d2dbd2d3d621ff26da2
7071Author: Andreas Müller <schnitzeltony@googlemail.com>
7072Date: Tue Apr 16 14:30:43 2013 +0200
7073
7074 dixstruct.h: fix segfaults - char is unsigned for ARM and PowerPC architectures
7075
7076 see ARM related bug reports [1-3]
7077
7078 [1] https://github.com/archlinuxarm/PKGBUILDs/issues/446I
7079 [2] http://www.raspberrypi.org/phpBB3/viewtopic.php?t=38568&p=321673
7080 [3] http://lists.linuxtogo.org/pipermail/openembedded-core/2013-April/037805.html
7081
7082 Signed-off-by: Andreas Müller <schnitzeltony@googlemail.com>
7083 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
7084 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7085
7086commit 7d722796c678532e8c5897c673c43184da353f44
7087Author: Peter Hutterer <peter.hutterer@who-t.net>
7088Date: Wed Apr 17 19:47:42 2013 +1000
7089
7090 dix: plug memory leak in freeing TouchClass
7091
7092 ==15562== 1,800 bytes in 1 blocks are definitely lost in loss record 298 of 330
7093 ==15562== at 0x4A06B6F: calloc (vg_replace_malloc.c:593)
7094 ==15562== by 0x4312C7: InitTouchClassDeviceStruct (devices.c:1644)
7095
7096 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7097 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7098
7099commit 5ab260317ad3b2aafff31a97df21620db52eacd1
7100Author: Geert Uytterhoeven <geert@linux-m68k.org>
7101Date: Sun Mar 10 21:04:19 2013 +0100
7102
7103 Xfbdev: Wire up Amiga afb4 and afb8 support
7104
7105 Add support for Amiga-style bitplanes, with 4 or 8 bits per pixel.
7106
7107 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7108 Acked-by: Keith Packard <keithp@keithp.com>
7109
7110commit 672bc5bb38918304cf68114a1112cd48651a5e83
7111Author: Geert Uytterhoeven <geert@linux-m68k.org>
7112Date: Sun Mar 10 17:17:26 2013 +0100
7113
7114 Xfbdev: Wire up Atari iplan2p4 and iplan2p8 support
7115
7116 Add support for Atari-style interleaved bitplanes, with 2 bytes interleave
7117 and 4 or 8 bits per pixel.
7118
7119 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7120 Reviewed-by: Keith Packard <keithp@keithp.com>
7121
7122commit 95a3c7536c1a4afe97aaf955980034cc69af9c2f
7123Author: Geert Uytterhoeven <geert@linux-m68k.org>
7124Date: Sun Mar 10 17:17:26 2013 +0100
7125
7126 Xfbdev: Force shadowfb for frame buffers with non-packed pixels
7127
7128 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7129 Reviewed-by: Keith Packard <keithp@keithp.com>
7130
7131commit 0e808110df216649e05503baecd06cd5a3e50421
7132Author: Geert Uytterhoeven <geert@linux-m68k.org>
7133Date: Sun Mar 10 17:17:26 2013 +0100
7134
7135 Xfbdev: Reject unsupported frame buffer types
7136
7137 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7138 Reviewed-by: Keith Packard <keithp@keithp.com>
7139
7140commit cfd10576812c36f5844805eb95ed1f2d093d1691
7141Author: Geert Uytterhoeven <geert@linux-m68k.org>
7142Date: Sun Mar 10 21:03:44 2013 +0100
7143
7144 Shadow: Add support for Amiga afb8
7145
7146 Add support for Amiga-style bitplanes, with 8 bits per pixel.
7147
7148 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7149 Acked-by: Keith Packard <keithp@keithp.com>
7150
7151commit a1b8e7f1e6118b611ba9d332b8763ee2b44f550c
7152Author: Geert Uytterhoeven <geert@linux-m68k.org>
7153Date: Wed Mar 27 09:37:02 2013 +0100
7154
7155 Shadow: Add support for Amiga afb4
7156
7157 Add support for Amiga-style bitplanes, with 4 bits per pixel.
7158
7159 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7160 Acked-by: Keith Packard <keithp@keithp.com>
7161
7162commit d7181e567d1629c387b834da7eecdf618d14718e
7163Author: Geert Uytterhoeven <geert@linux-m68k.org>
7164Date: Sun Mar 10 17:16:17 2013 +0100
7165
7166 Shadow: Add support for Atari iplan2p8
7167
7168 Add support for Atari-style interleaved bitplanes, with 2 bytes interleave
7169 and 8 bits per pixel.
7170
7171 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7172 Acked-by: Keith Packard <keithp@keithp.com>
7173
7174commit 3f7506b0fca72b2462b85bc4f613de809ae95859
7175Author: Geert Uytterhoeven <geert@linux-m68k.org>
7176Date: Sun Mar 10 21:01:02 2013 +0100
7177
7178 Shadow: Add support for Atari iplan2p4
7179
7180 Add support for Atari-style interleaved bitplanes, with 2 bytes interleave
7181 and 4 bits per pixel.
7182
7183 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7184 Acked-by: Keith Packard <keithp@keithp.com>
7185
7186commit 87af9ab7d52578b0ea315fc4f8dfd906d19755fb
7187Author: Geert Uytterhoeven <geert@linux-m68k.org>
7188Date: Sun Mar 10 17:15:05 2013 +0100
7189
7190 Shadow: Add c2p core
7191
7192 Add Chunky-to-Planar core functionality, to be used by the Atari and Amiga
7193 (interleaved) bitplanes code.
7194
7195 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7196 Acked-by: Keith Packard <keithp@keithp.com>
7197
7198commit 4ee2566c3e420a717be36f79126a14e15edac30f
7199Author: Geert Uytterhoeven <geert@linux-m68k.org>
7200Date: Tue Mar 12 18:10:40 2013 +0100
7201
7202 Xfbdev: Treat 1 bpp pseudocolor as monochrome
7203
7204 miCreateDefColormap() only preallocates black and white pixels if
7205 depth > 1.
7206 Hence override the visual, so fbdevCreateColormap() takes care of it.
7207
7208 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7209 Reviewed-by: Keith Packard <keithp@keithp.com>
7210
7211commit 6dfb94a891ed445f47832ba1364fcbf1de018f4a
7212Author: Geert Uytterhoeven <geert@linux-m68k.org>
7213Date: Tue Mar 12 14:16:10 2013 +0100
7214
7215 Xfbdev: Add support for monochrome visuals
7216
7217 Monochrome supports StaticGray, with hardcoded black and white pixels.
7218
7219 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7220 Reviewed-by: Keith Packard <keithp@keithp.com>
7221
7222commit 1049b32166760bdc00106625e213d31a8fc60bad
7223Author: Geert Uytterhoeven <geert@linux-m68k.org>
7224Date: Tue Mar 12 14:19:25 2013 +0100
7225
7226 Xfbdev: Handle unset fix.line_length
7227
7228 Older frame buffer devices may not fill in fix.line_length, in which
7229 case it must be calculated by the application.
7230
7231 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7232
7233commit 5c509c360d97bea78ef461596ab4b7bde69020ac
7234Author: Geert Uytterhoeven <geert@linux-m68k.org>
7235Date: Fri Feb 22 13:21:28 2013 +0100
7236
7237 Xfbdev: Make char *fbdevDevicePath const
7238
7239 This fixes:
7240
7241 hw/kdrive/fbdev/fbdev.c: In function 'fbdevInitialize':
7242 hw/kdrive/fbdev/fbdev.c:41:25: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]
7243
7244 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7245 Reviewed-by: Keith Packard <keithp@keithp.com>
7246
7247commit b08afbc53c1d583c3913b92e67db44823077b112
7248Author: Geert Uytterhoeven <geert@linux-m68k.org>
7249Date: Tue Mar 12 15:20:00 2013 +0100
7250
7251 KDrive: Bail out if screen initialization failed
7252
7253 Else we may get a segmentation fault later.
7254
7255 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7256 Reviewed-by: Keith Packard <keithp@keithp.com>
7257
7258commit e7045c9dd208a1afe36526ab21ef15f8b01c8bd2
7259Author: Geert Uytterhoeven <geert@linux-m68k.org>
7260Date: Sat Feb 23 20:09:22 2013 +0100
7261
7262 test/input: Fix double-aligned test in dix_valuator_alloc() on m68k
7263
7264 On m68k, doubles are not 64-bit aligned, just like on i386 and sh.
7265
7266 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7267
7268commit 8bb0616cc2799c2c34448da6337ea8cb9438cda7
7269Author: Geert Uytterhoeven <geert@linux-m68k.org>
7270Date: Fri Feb 22 21:10:07 2013 +0100
7271
7272 miext/shadow/shpacked.c: Remove unused PickBit() define
7273
7274 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7275 Reviewed-by: Keith Packard <keithp@keithp.com>
7276
7277commit 6ca03b9161d33b1d2b55a3a1a913cf88deb2343f
7278Author: Dave Airlie <airlied@gmail.com>
7279Date: Wed Apr 10 16:09:01 2013 +1000
7280
7281 xf86: fix flush input to work with Linux evdev devices.
7282
7283 So when we VT switch back and attempt to flush the input devices,
7284 we don't succeed because evdev won't return part of an event,
7285 since we were only asking for 4 bytes, we'd only get -EINVAL back.
7286
7287 This could later cause events to be flushed that we shouldn't have
7288 gotten.
7289
7290 This is a fix for CVE-2013-1940.
7291
7292 Signed-off-by: Dave Airlie <airlied@redhat.com>
7293 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7294 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7295
7296commit ddc149beaf708e84fb5573c4322d4f859c517917
7297Author: Peter Hutterer <peter.hutterer@who-t.net>
7298Date: Mon Feb 27 18:50:46 2012 +1000
7299
7300 Stop the shouting
7301
7302 Meanwhile, here in the future lowercase letters have been invented.
7303
7304 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7305 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
7306
7307commit 27356a45b4f6bad4d0bbf356e6d976d0c8dd257b
7308Author: Peter Hutterer <peter.hutterer@who-t.net>
7309Date: Mon Feb 27 17:22:16 2012 +1000
7310
7311 dix: only allocate unused classes for master devices
7312
7313 Slave devices don't need these and the matching code in CloseDevice() has a
7314 IsMaster() condition on freeing these, causing a leak.
7315
7316 ==16111== 384 bytes in 4 blocks are definitely lost in loss record 72 of 105
7317 ==16111== at 0x4C28BB4: calloc (vg_replace_malloc.c:467)
7318 ==16111== by 0x42AEE2: AllocDevicePair (devices.c:2707)
7319 ==16111== by 0x4BAA27: AllocXTestDevice (xtest.c:617)
7320 ==16111== by 0x4BA89A: InitXTestDevices (xtest.c:570)
7321 ==16111== by 0x425F5E: InitCoreDevices (devices.c:690)
7322 ==16111== by 0x5ACB2D: main (main.c:257)
7323
7324 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7325 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
7326
7327commit 98b94c36d6b1d286bbd4cb414e54b4b95a1484b0
7328Author: Maarten Lankhorst <maarten.lankhorst@canonical.com>
7329Date: Mon Apr 15 16:53:48 2013 +0200
7330
7331 dix: copy event in TouchConvertToPointerEvent correctly
7332
7333 Fixes reading random memory read beyond the end of original event.
7334
7335 sizeof device_event: 424
7336 sizeof internal_event: 2800
7337
7338 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
7339 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7340 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7341
7342commit 4bff442ec5aa4b93a3f5c11782d4b7b9d1ae13ac
7343Author: Knut Petersen <Knut_Petersen@t-online.de>
7344Date: Tue Feb 26 07:52:59 2013 +0100
7345
7346 Never try to execute BellProcPtr NULL.
7347
7348 This prevents xts XI/XDeviceBell-2 test
7349 from segfaulting the server.
7350
7351 Signed-off-by: Knut Petersen <Knut_Petersen@t-online.de>
7352 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7353 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7354
7355commit 9003399708936481083424b4ff8f18a16b88b7b3
7356Author: Rui Matos <tiagomatos@gmail.com>
7357Date: Sat Apr 13 04:22:54 2013 +0200
7358
7359 Xi: fix swapped grab mode args for keyboard devices in XIGrabDevice
7360
7361 The protocol says that the grab_mode argument applies to the device
7362 being grabbed and paired_device_mode to the paired master
7363 device. GrabDevice() however takes in a pointer mode and a keyboard
7364 mode and so we have to swap the values according the type of device
7365 being grabbed.
7366
7367 Signed-off-by: Rui Matos <tiagomatos@gmail.com>
7368 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7369 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7370
7371commit d16284687ddeffde32a561b8c67b2e2e3d66aeaa
7372Author: Bryce Harrington <bryce@canonical.com>
7373Date: Wed Apr 10 23:05:40 2013 -0700
7374
7375 xfree86: Revert workaround for drm race condition.
7376
7377 Revert 70739e817b2d64bc020ea491f23a3574bdb6155e and mostly revert
7378 c31eac647a9ecf0fb20dc98266cadf0ba923ba14.
7379
7380 Further investigation shows the encountered race condition is between
7381 lightdm and plymouth-splash, as implemented in the Ubuntu distribution
7382 within the limitations of upstart's job coordination logic, and can (and
7383 should) be fixed within those limiations. Not in xserver itself.
7384
7385 This leaves some of the diagnostic improvements from the recent patch
7386 series, in case others run into a similar situation.
7387
7388 Signed-off-by: Bryce Harrington <bryce@canonical.com>
7389 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7390
7391commit 131f883f85b9b734e5e5652f16ba1d3b5f4de12f
7392Author: Peter Hutterer <peter.hutterer@who-t.net>
7393Date: Fri Apr 12 09:14:53 2013 +1000
7394
7395 xfree86: change a log message
7396
7397 This path is technically executed through config/udev, but having two
7398 messages in the form "config/udev: Adding drm device" makes it appear as if
7399 the udev filters are wrong and it's trying to add the same device twice. In
7400 fact, it's only one device, only added once, but a duplicate log message.
7401
7402 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7403 Reviewed-by: Dave Airlie <airlied@redhat.com>
7404
7405commit 22cab8a28a433d03a4e6ba97f9a160271d73cb52
7406Author: Dave Airlie <airlied@gmail.com>
7407Date: Wed Apr 10 16:32:15 2013 +1000
7408
7409 xf86: don't hotplug output devices while VT switched.
7410
7411 We don't want to hotplug output devices while we are VT switched,
7412 as we get races between multiple X servers on the device open, and
7413 drm device master status. This just queues device opens until we return
7414 from VT switch.
7415
7416 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7417 Signed-off-by: Dave Airlie <airlied@redhat.com>
7418
7419commit 5b359cf6135ca173d8f65cb92926332f07f91efe
7420Author: Dave Airlie <airlied@gmail.com>
7421Date: Wed Apr 10 16:32:11 2013 +1000
7422
7423 xf86: use new xf86VTOwner interface in a few places
7424
7425 This replaces some previous uses of direct xf86Screens[0] accesses.
7426
7427 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7428 Signed-off-by: Dave Airlie <airlied@gmail.com>
7429
7430commit d61ea1f64db45201c1a2b39c39293c5768d98092
7431Author: Dave Airlie <airlied@gmail.com>
7432Date: Wed Apr 10 16:32:02 2013 +1000
7433
7434 xfree86: add VT owner interface
7435
7436 This is just a simple interface to avoid accessing x86Screens[0]
7437 directly.
7438
7439 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7440 Signed-off-by: Dave Airlie <airlied@gmail.com>
7441
7442commit 7347f39f94d8cebbf73ce1a2f94d1abdaf7ff383
7443Author: Maarten Lankhorst <maarten.lankhorst@canonical.com>
7444Date: Tue Apr 9 11:19:07 2013 +0200
7445
7446 Xi: Do not handle ET_TouchOwnership in ProcessTouchEvent
7447
7448 The event struct is different, causing memory corruption on 1.13 and 1.14,
7449
7450 as can be witnessed in https://bugs.freedesktop.org/show_bug.cgi?id=56578
7451
7452 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
7453 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7454 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7455
7456commit dbba50a1280cbda9ecff6f37884b4c5756c30bab
7457Author: Keith Packard <keithp@keithp.com>
7458Date: Tue Apr 9 16:23:19 2013 -0700
7459
7460 Xi: Use correct destination when swapping barrier events
7461
7462 Write the swapped values to the destination rather than the source.
7463
7464 Signed-off-by: Keith Packard <keithp@keithp.com>
7465 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
7466 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7467
7468commit b86b3d10bb2fee1a922b8831e8bb415c339f3d99
7469Author: Peter Hutterer <peter.hutterer@who-t.net>
7470Date: Mon Mar 4 07:58:41 2013 +1000
7471
7472 dix: don't set non-exisiting flags on touch events
7473
7474 Unlike pointer/keyboard events, the flags field for ET_Touch* is a set of
7475 server-internal defines that we need to convert to XI protocol defines.
7476 Currently only two of those defines actually translate to the protocol, so
7477 make sure we don't send internal garbage down the wire.
7478
7479 No effect to current clients since they shouldn't look at undefined bits
7480 anyway.
7481
7482 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7483
7484commit ecf62755086fd65898998d5a509aee5f29a9583d
7485Author: Jeremy White <jwhite@codeweavers.com>
7486Date: Thu Mar 21 10:58:57 2013 -0500
7487
7488 Define prototypes for hw/xfree86/modes/xf86Modes.c only in xf86Modes.h.
7489
7490 This removes a large number of redundant declaration warnings.
7491
7492 Signed-off-by: Jeremy White <jwhite@codeweavers.com>
7493 Reviewed-by: Robert Morell <rmorell@nvidia.com>
7494 Signed-off-by: Keith Packard <keithp@keithp.com>
7495
7496commit e13f29984251dbbe6076264ed710c2922312eed1
7497Author: Bryce Harrington <bryce@canonical.com>
7498Date: Tue Mar 19 12:12:46 2013 -0700
7499
7500 xfree86: Be verbose if waiting on opening the drm device
7501
7502 Signed-off-by: Bryce Harrington <bryce@canonical.com>
7503 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7504 Signed-off-by: Keith Packard <keithp@keithp.com>
7505
7506commit 70739e817b2d64bc020ea491f23a3574bdb6155e
7507Author: Bryce Harrington <bryce@canonical.com>
7508Date: Tue Mar 19 12:12:45 2013 -0700
7509
7510 xfree86: Fix race condition failure opening drm.
7511
7512 If other processes have had drm open previously, xserver may attempt to
7513 open the device too early and fail, with xserver error exit "Cannot
7514 run in framebuffer mode" or Xorg.0.log messages about "setversion 1.4
7515 failed".
7516
7517 In this situation, we're receiving back -EACCES from libdrm. To address
7518 this we need to re-set ourselves as the drm master, and keep trying to
7519 set the interface until it works (or until we give up).
7520
7521 See https://bugs.launchpad.net/ubuntu/+source/libdrm/+bug/982889
7522
7523 Signed-off-by: Bryce Harrington <bryce@canonical.com>
7524 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7525 Signed-off-by: Keith Packard <keithp@keithp.com>
7526
7527commit c31eac647a9ecf0fb20dc98266cadf0ba923ba14
7528Author: Bryce Harrington <bryce@canonical.com>
7529Date: Tue Mar 19 12:12:44 2013 -0700
7530
7531 xfree86: Keep trying to set interface on drm for 2 seconds.
7532
7533 And if we've had to delay booting due to not being able to set the
7534 interface, fess up.
7535
7536 Signed-off-by: Bryce Harrington <bryce@canonical.com>
7537 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7538 Signed-off-by: Keith Packard <keithp@keithp.com>
7539
7540commit d1cc210de8c13f2db9f6f284ecc652305c28801e
7541Author: Bryce Harrington <bryce@canonical.com>
7542Date: Tue Mar 19 12:12:43 2013 -0700
7543
7544 xfree86: Provide more details on failure
7545
7546 Signed-off-by: Bryce Harrington <bryce@canonical.com>
7547 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7548 Signed-off-by: Keith Packard <keithp@keithp.com>
7549
7550commit f059d0dabc553a5f748d86de9115da00be5997d5
7551Author: Bryce Harrington <bryce@canonical.com>
7552Date: Tue Mar 19 12:12:42 2013 -0700
7553
7554 xfree86: Track error code and add label for error handling.
7555
7556 Signed-off-by: Bryce Harrington <bryce@canonical.com>
7557 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7558 Signed-off-by: Keith Packard <keithp@keithp.com>
7559
7560commit 4d7052bd7bbf49b573dc4d34ad14e7f058a0d884
7561Author: Bryce Harrington <bryce@canonical.com>
7562Date: Tue Mar 19 12:12:41 2013 -0700
7563
7564 xfree86: (Cleanup) Close fd if drm interface 1.4 could not be set.
7565
7566 Signed-off-by: Bryce Harrington <bryce@canonical.com>
7567 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
7568 Signed-off-by: Keith Packard <keithp@keithp.com>
7569
7570commit 8928f8fa0bb154ce437af703ff702016f0dcf127
7571Author: Geert Uytterhoeven <geert@linux-m68k.org>
7572Date: Mon Mar 18 21:38:10 2013 +0100
7573
7574 kdrive/fbdev: revive randr new screen size logic
7575
7576 hw/kdrive/fbdev/fbdev.c: In function 'fbdevRandRSetConfig':
7577 hw/kdrive/fbdev/fbdev.c:470:19: warning: variable 'newheight' set but not used [-Wunused-but-set-variable]
7578 hw/kdrive/fbdev/fbdev.c:470:9: warning: variable 'newwidth' set but not used [-Wunused-but-set-variable]
7579
7580 Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
7581 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7582 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7583
7584commit 808c87bff710263f5a7b3c12a7e7fec54672fecd
7585Author: Peter Hutterer <peter.hutterer@who-t.net>
7586Date: Mon Mar 25 14:51:58 2013 +1000
7587
7588 Revert "kdrive: fix "set but not used" warnings"
7589
7590 Partial revert of commit 4149ee8ec0193acbf3812c7ee2627b93b9a89997, better
7591 fix coming up.
7592
7593 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7594
7595commit ac4c2abe985ce6816a3389c41eb23462c6ac2486
7596Author: vdb@picaros.org <vdb@picaros.org>
7597Date: Thu Mar 7 16:07:31 2013 +0100
7598
7599 xserver: add monitor Option "ZoomModes" [v2]
7600
7601 Section "Monitor"
7602 Identifier "a21inch"
7603 Option "PreferredMode" "1600x1200"
7604 Option "ZoomModes" "1600x1200 1280x1024 1280x1024 640x480"
7605 EndSection
7606
7607 The option's effect is to search for and mark once each named mode in
7608 the output modes list. So the specification order is free and the zoom
7609 modes sequence follows the order of the output modes list. All marked
7610 modes are available via the Ctrl+Alt+Keypad-{Plus,Minus} key
7611 combination.
7612
7613 See also http://bugs.freedesktop.org/show_bug.cgi?id=17954.
7614
7615 This option has its use for combined monitor and television setups.
7616 It allows for easy switching between 60 Hz and 50 Hz modes even when a
7617 monitor refuses to display the input signal.
7618
7619 (Includes a few minor changes suggested by Aaron for v2)
7620
7621 Signed-off-by: Servaas Vandenberghe <vdb@picaros.org>
7622 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
7623 Signed-off-by: Keith Packard <keithp@keithp.com>
7624
7625commit 2967391c6d35f03121afa8003e0fb94b62495129
7626Author: Peter Hutterer <peter.hutterer@who-t.net>
7627Date: Thu Feb 28 10:43:05 2013 +1000
7628
7629 Xi: add a comment to make a condition a bit clearer
7630
7631 The commit message to 676447190190d8546165e21be242cf16dd69f5ae explains it,
7632 but that doesn't stop the WTF moment when reading the code.
7633
7634 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7635 Reviewed-by: Keith Packard <keithp@keithp.com>
7636
7637commit 2fdde2c40d83695438b6f5615f98bd7ae801b43d
7638Author: Peter Hutterer <peter.hutterer@who-t.net>
7639Date: Mon Feb 25 15:21:32 2013 +1000
7640
7641 dix: update coords for touch events in PlayReleasedEvents
7642
7643 Note: this is only hit for #ifdef PANORAMIX
7644
7645 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7646 Reviewed-by: Keith Packard <keithp@keithp.com>
7647
7648commit 697071ab2b7f8910e01ed74618138538291cd1e8
7649Author: Peter Hutterer <peter.hutterer@who-t.net>
7650Date: Thu Feb 28 10:30:15 2013 +1000
7651
7652 Xi: compress two if statements with the same body
7653
7654 We do the same thing here, compress them into one body.
7655
7656 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7657 Reviewed-by: Keith Packard <keithp@keithp.com>
7658
7659commit 11bead1fa205a1353e6a33c6024c7e8ace80be7c
7660Author: Peter Hutterer <peter.hutterer@who-t.net>
7661Date: Mon Feb 25 11:13:52 2013 +1000
7662
7663 dix: fix a comment
7664
7665 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7666 Reviewed-by: Keith Packard <keithp@keithp.com>
7667
7668commit fc504a44d12d537d4e07f659f1863f200a0272ad
7669Author: Peter Hutterer <peter.hutterer@who-t.net>
7670Date: Fri Mar 1 08:26:06 2013 +1000
7671
7672 Xi: use a temp variable for the new listener
7673
7674 Instead of accessing ti->listener[0] all the time.
7675
7676 No functional changes.
7677
7678 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7679 Reviewed-by: Keith Packard <keithp@keithp.com>
7680
7681commit 9978b57b8d94f061d72a67b99a02b0ba16a11429
7682Author: Peter Hutterer <peter.hutterer@who-t.net>
7683Date: Fri Mar 1 08:43:58 2013 +1000
7684
7685 Xi: return !Success from DeliverTouchEmulatedEvent if we didn't deliver
7686
7687 All callers currently ignore the new value, so this patch has no effect.
7688 Inverse call graph:
7689
7690 DeliverTouchEmulatedEvent
7691 DeliverEmulatedMotionEvent Ignores value
7692 DeliverTouchBeginEvent
7693 DeliverTouchEvent
7694 DeliverTouchEvents Ignores value
7695 DeliverTouchEndEvent
7696 DeliverTouchEvent
7697 DeliverTouchEvents Ignores value
7698
7699 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7700 Reviewed-by: Keith Packard <keithp@keithp.com>
7701
7702commit cc3d1a5a6120e721a46c67446ba68f5596055633
7703Author: Piotr Dziwinski <piotrdz@gmail.com>
7704Date: Sat Feb 23 13:14:45 2013 +0100
7705
7706 glx: fix uninitialized var in __glXDRIscreenProbe
7707
7708 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=59825
7709
7710 Signed-off-by: Piotr Dziwinski <piotrdz@gmail.com>
7711 Reviewed-by: Keith Packard <keithp@keithp.com>
7712 Signed-off-by: Keith Packard <keithp@keithp.com>
7713
7714commit c2ede8f92f0d6f15b1b2a04889cb54cd1728780b
7715Merge: 190b032 7050aae
7716Author: Keith Packard <keithp@keithp.com>
7717Date: Mon Mar 18 11:26:24 2013 -0700
7718
7719 Merge remote-tracking branch 'whot/for-keith'
7720
7721commit 190b0321510b99d4738915f540cea5c0c51e07e2
7722Author: Zack Rusin <zackr@vmware.com>
7723Date: Thu Feb 14 15:06:57 2013 -0800
7724
7725 GLX/DRI2: Do not expose INTEL_swap_event without swap control
7726
7727 Swap events depent on the implementation of ScheduleSwap. By
7728 unconditionally enabling GLX_INTEL_swap_event we're breaking
7729 the system with drivers that don't support it because the apps
7730 are forever stuck waiting for an event that will never be
7731 delivered. So lets enable the extension only if the hooks it
7732 depends on are actually there.
7733
7734 Signed-off-by: Zack Rusin <zackr@vmware.com>
7735 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
7736 Signed-off-by: Keith Packard <keithp@keithp.com>
7737
7738commit 116f020102fd6c2a603069a639b113dfa31b48b7
7739Merge: cf89aa5 0f537da
7740Author: Keith Packard <keithp@keithp.com>
7741Date: Mon Mar 18 11:18:58 2013 -0700
7742
7743 Merge remote-tracking branch 'whot/next'
7744
7745commit cf89aa53748b964f9d9eceaa12a7d6f1a076d1ee
7746Author: Tomasz Lis <tomasz.lis@intel.com>
7747Date: Mon Mar 11 10:21:28 2013 +0100
7748
7749 Full support of sRGB capable fbconfigs.
7750
7751 Changes to correctly initialize the sRGB capability attribute and
7752 transfer it between XServer and the client. Modifications include
7753 extension string, transferring visual config attribs and fbconfig
7754 attribs. Also, attribute is initialized in the modules which do not
7755 really use it (xquartz and xwin).
7756 This version advertises both ARB and EXT strings, and initializes
7757 the capability to default value of FALSE. It has corrected required
7758 GLX version and does not influence swrast. The sRGB capable attribute
7759 is attached only to those configs which do have this capability.
7760 Both ARB and EXT versions share the same GLX extension enabling bit.
7761
7762 Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
7763 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
7764 Signed-off-by: Keith Packard <keithp@keithp.com>
7765
7766commit 679ccecd8bc6d797e99f3b707f8041b727ca7a49
7767Author: Keith Packard <keithp@keithp.com>
7768Date: Mon Mar 18 10:01:20 2013 -0700
7769
7770 Bump release to 1.14.99.0
7771
7772 Get us off the release number so that we don't conflict with the
7773 stable branch.
7774
7775 Signed-off-by: Keith Packard <keithp@keithp.com>
7776
7777commit 7050aae69c2a55dfdbb5c6af7882307e90ba4275
7778Author: Robert Morell <rmorell@nvidia.com>
7779Date: Tue Mar 12 09:40:16 2013 -0700
7780
7781 list.h: Make xorg_list_init inline
7782
7783 Otherwise this file is emitted in every unit that includes it.
7784
7785 Signed-off-by: Robert Morell <rmorell@nvidia.com>
7786 Reviewed-by: Jamey Sharp <jamey@minilop.net>
7787 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7788 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7789
7790commit 31595b528645a7e4903eb81da0fc332d78407f25
7791Author: Robert Morell <rmorell@nvidia.com>
7792Date: Tue Mar 12 09:37:43 2013 -0700
7793
7794 configure.ac: Require inputproto 2.3
7795
7796 This picks up support for Xi pointer barriers in the protocol.
7797
7798 Signed-off-by: Robert Morell <rmorell@nvidia.com>
7799 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7800 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7801
7802commit 3ac2e61705432951f9e1b96b42b4214e7f748f94
7803Author: Peter Hutterer <peter.hutterer@who-t.net>
7804Date: Sat Mar 9 17:12:53 2013 +1000
7805
7806 xfixes: ifdef PanoramiXFixes* (#62015)
7807
7808 Fixes build failure with --disable-xinerama introduced by
7809 482e0cb cursor: Move pointer barrier code over to XI
7810
7811 Reason is new include order: sdksyms.sh includes xfixes.h, which previously
7812 did not include xfixesint.h.
7813
7814 As of 482e0cb xfixes.h includes xibarriers.h which includes xfixesint.h
7815
7816 X.Org Bug 62015 <http://bugs.freedesktop.org/show_bug.cgi?id=62015>
7817
7818 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7819 Tested-by: Daniel Martin <consume.noise@gmail.com>
7820
7821commit 5047810a4c20fab444b8c6eb146c55dcdb0d4219
7822Author: Aaron Plattner <aplattner@nvidia.com>
7823Date: Wed Mar 6 09:40:28 2013 -0800
7824
7825 fb: Rename wfbDestroyGlyphCache
7826
7827 Renaming this function was missed in commit
7828 9cbcb5bd6a5360a128d15b77a02d8d3351f74366, so both libfb.so and libwfb.so define
7829 functions named fbDestroyGlyphCache.
7830
7831 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
7832 Reviewed-by: Søren Sandmann <ssp@redhat.com>
7833 Signed-off-by: Keith Packard <keithp@keithp.com>
7834
7835commit 103b77c59e3638a45179bf6d7908f5c738d2d872
7836Author: Keith Packard <keithp@keithp.com>
7837Date: Tue Mar 5 22:31:17 2013 -0800
7838
7839 Version bumped to 1.14
7840
7841 Signed-off-by: Keith Packard <keithp@keithp.com>
7842
7843commit 0f537da72d414ed84e3cd14e3bb7e08565136bd7
7844Author: Andreas Wettstein <wettstein509@solnet.ch>
7845Date: Sun Mar 3 20:25:44 2013 +0100
7846
7847 xkb: Fixes to LatchMods/LatchGroup
7848
7849 The main problem this patch addresses is that if a latch is put on
7850 multi-level key with a Latch/Lock/Set, it is possible that after all
7851 keys are released, still base modifiers are set, which typically will
7852 make the keyboard unusable. To see how it happens (without the patch),
7853 assume that key AltGr sets Mod5 when pressed by itself, and latches Mod3
7854 when pressed together with Shift. Now press Shift, then AltGr and
7855 release both keys in reverse order. Mod3 is now latched, and the
7856 LatchMods filter remains active as the second filter. Now press AltGr;
7857 Mod5 base modifier gets set, and the SetMods filter will become active
7858 as the first filter. Release AltGr: First, the SetMods filter will set
7859 clearMods to Mod5, then the LatchMods filter will overwrite clearMods
7860 with Mod3. Result: the Mod5 base modifier will remain set. This
7861 example becomes practically relevant for the revised German standard
7862 layout (DIN 2137-1:2012-06).
7863
7864 Other changes implement the latch behaviour more accurately according to
7865 the specification. For example, releasing a modifier latching key can
7866 at the same time clear a locked modifier, promote another modifier that
7867 is latched to locked, and latch a third modifier. Overall, what the
7868 code does should be straightforward to compare what the XKB protocol
7869 specification demands, see the table in section 6.3.
7870
7871 Finally, releasing a key no longer cancels a latch that has not become
7872 pending yet. In my opinion, the specification is not clear; it speaks
7873 of "operating" a key, which the patch effectivly interprets as "press"
7874 rather than "press or release". From my experience, using the latter
7875 interpretation makes latches on higher levels practically unusable. In
7876 the example given above, one would have to release AltGr always before
7877 Shift to get the Mod3-Latch. The practical relevance of latches on
7878 higher levels is once more given by the revised German standard layout.
7879
7880 Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
7881 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7882 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7883
7884commit eadda231091aa6feb68207ee22f6bc3a390d4556
7885Author: Peter Hutterer <peter.hutterer@who-t.net>
7886Date: Fri Feb 15 11:19:09 2013 +1000
7887
7888 xephyr: fix "set but not used warnings"
7889
7890 ephyrvideo.c: In function 'ephyrPutVideo':
7891 ephyrvideo.c:1009:42: warning: variable 'drw_h' set but not used
7892 ephyrvideo.c:1009:31: warning: variable 'drw_w' set but not used
7893 ephyrvideo.c:1009:20: warning: variable 'drw_y' set but not used
7894 ephyrvideo.c:1009:9: warning: variable 'drw_x' set but not used
7895 ephyrvideo.c: In function 'ephyrGetVideo':
7896 ephyrvideo.c:1058:42: warning: variable 'drw_h' set but not used
7897 ephyrvideo.c:1058:31: warning: variable 'drw_w' set but not used
7898 ephyrvideo.c:1058:20: warning: variable 'drw_y' set but not used
7899 ephyrvideo.c:1058:9: warning: variable 'drw_x' set but not used
7900 ephyrvideo.c: In function 'ephyrPutStill':
7901 ephyrvideo.c:1107:42: warning: variable 'drw_h' set but not used
7902 ephyrvideo.c:1107:31: warning: variable 'drw_w' set but not used
7903 ephyrvideo.c:1107:20: warning: variable 'drw_y' set but not used
7904 ephyrvideo.c:1107:9: warning: variable 'drw_x' set but not used
7905 ephyrvideo.c: In function 'ephyrGetStill':
7906 ephyrvideo.c:1156:42: warning: variable 'drw_h' set but not used
7907 ephyrvideo.c:1156:31: warning: variable 'drw_w' set but not used
7908 ephyrvideo.c:1156:20: warning: variable 'drw_y' set but not used
7909 ephyrvideo.c:1156:9: warning: variable 'drw_x' set but not used
7910
7911 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7912 Reviewed-by: Robert Morell <rmorell@nvidia.com>
7913
7914commit 4149ee8ec0193acbf3812c7ee2627b93b9a89997
7915Author: Peter Hutterer <peter.hutterer@who-t.net>
7916Date: Fri Feb 15 11:19:08 2013 +1000
7917
7918 kdrive: fix "set but not used" warnings
7919
7920 kinput.c: In function 'KdEnqueueKeyboardEvent':
7921 kinput.c:1845:16: warning: variable 'ctrl' set but not used
7922 kinput.c:1844:17: warning: variable 'keyc' set but not used
7923
7924 kinput.c: In function 'KdEnqueuePointerEvent':
7925 kinput.c:1887:12: warning: variable 'ms' set but not used
7926
7927 kxv.c: In function 'KdXVDisable':
7928 kxv.c:1181:19: warning: variable 'ScreenPriv' set but not used
7929
7930 mouse.c: In function 'ps2SkipInit':
7931 mouse.c:444:9: warning: variable 'skipping' set but not used
7932 mouse.c: In function 'ps2Init':
7933 mouse.c:473:10: warning: variable 'waiting' set but not used
7934 mouse.c:472:9: warning: variable 'skipping' set but not used
7935
7936 fbdev.c: In function 'fbdevRandRSetConfig':
7937 fbdev.c:468:19: warning: variable 'newheight' set but not used
7938 fbdev.c:468:9: warning: variable 'newwidth' set but not used
7939
7940 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7941 Reviewed-by: Robert Morell <rmorell@nvidia.com>
7942
7943commit 44fc062f85df7288c17d2d64b73aa4957b91fd6d
7944Author: Peter Hutterer <peter.hutterer@who-t.net>
7945Date: Mon Feb 18 14:57:58 2013 +1000
7946
7947 os: document pnprintf as sigsafe snprintf
7948
7949 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7950
7951commit c100211034ab69ce453a1644fb61c6808d7e3eda
7952Author: Peter Hutterer <peter.hutterer@who-t.net>
7953Date: Tue Dec 18 14:12:40 2012 +1000
7954
7955 dix: only show the cursor if a window defines one (#58398)
7956
7957 e02f864fdf "Suppress cursor display until the first XDefineCursor() request"
7958 disabled cursor display a priori unless -retro is given.
7959
7960 On a plain server, caling XFixesHideCursor() and XFixesShowCursor() would
7961 show the default root cursor, despite no client actually defining a cursor.
7962
7963 Change the logic, disable CursorVisible by default and only enable it from
7964 the window's CWCursor logic. If no window ever defines a cursor, said cursor
7965 stays invisible.
7966
7967 X.Org Bug 58398 <http://bugs.freedesktop.org/show_bug.cgi?id=58398>
7968
7969 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7970 Tested-by: Bastien Nocera <hadess@hadess.net>
7971 Reviewed-by: Daniel Martin <consume.noise@gmail.com>
7972
7973commit 6238bd68bd71323f8b4f1808f34dabe2ae447fe3
7974Author: Aaron Plattner <aplattner@nvidia.com>
7975Date: Tue Mar 5 13:04:46 2013 -0800
7976
7977 DPMS: include GPU screens in DPMS code
7978
7979 Otherwise, displays driven by GPU screens remain on all the time.
7980
7981 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
7982 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
7983 Signed-off-by: Keith Packard <keithp@keithp.com>
7984
7985commit dd4ab8b572956f9457616869d6e383c8ed964c73
7986Author: Peter Hutterer <peter.hutterer@who-t.net>
7987Date: Tue Mar 5 10:14:29 2013 +1000
7988
7989 Xi: force dtime to 0 on the first BarrierHit
7990
7991 dtime to the previous event is 0 on the first BarrierHit event.
7992
7993 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
7994 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
7995 Signed-off-by: Keith Packard <keithp@keithp.com>
7996
7997commit 604169af8b67afc74a292cdb9070a3a1f2d7c536
7998Merge: 8f4640b 6ea59dc
7999Author: Keith Packard <keithp@keithp.com>
8000Date: Mon Mar 4 21:09:59 2013 -0800
8001
8002 Merge commit '6ea59dc2d8887102bfd8d7c838d2e7ab17645aec'
8003
8004commit 6ea59dc2d8887102bfd8d7c838d2e7ab17645aec
8005Author: Jasper St. Pierre <jstpierre@mecheye.net>
8006Date: Sun Mar 3 04:50:55 2013 -0500
8007
8008 xibarriers: Remove accidental use of the comma operator
8009
8010 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8011 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8012
8013commit bd58ebe4cf3b0ce60f87fb26a3715f774dabd349
8014Author: Daniel Martin <consume.noise@gmail.com>
8015Date: Thu Dec 20 13:50:17 2012 +0100
8016
8017 ephyr: Fix crash on 24bpp host framebuffer
8018
8019 Use bytes_per_line and bits_per_pixel from the created XImage to fix
8020 https://bugzilla.redhat.com/show_bug.cgi?id=518960
8021
8022 Signed-off-by: Daniel Martin <consume.noise@gmail.com>
8023 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8024 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8025
8026commit 8f4640bdb9d3988148e09a08d2c7e3bab1d538d6
8027Author: Dave Airlie <airlied@redhat.com>
8028Date: Wed Jan 9 12:58:28 2013 +1000
8029
8030 randr: cleanup provider properly
8031
8032 So in the cold plug server shutdown case, we reap the resources
8033 before we call CloseScreen handlers, so the config->randr_provider
8034 is a dangling pointer when the xf86CrtcCloseScreen handler is called,
8035
8036 however in the hot screen unplug case, we can't rely on automatically
8037 reaped resources, so we need to clean up the provider in the xf86CrtcCloseScreen
8038 case.
8039
8040 This patch provides a cleanup callback from the randr provider removal
8041 into the DDX so it can cleanup properly, this then gets called by the automatic
8042 code for cold plug, or if hot unplug it gets called explicitly.
8043
8044 Fixes a number of random server crashes on shutdown
8045 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=58174
8046 Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=891140
8047
8048 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8049 Signed-off-by: Dave Airlie <airlied@redhat.com>
8050
8051commit 3ec35c45ca17f5ed6fd02c50fc49ae7b8d128dcb
8052Author: Dave Airlie <airlied@redhat.com>
8053Date: Wed Jan 9 12:53:14 2013 +1000
8054
8055 xf86: actually set the compat output in the failure case
8056
8057 The previous fix for the previous fix, didn't fully work,
8058
8059 If we don't set compat_output we end up doing derferences
8060 of arrays with -1, leading to valgrind warnings.
8061
8062 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
8063 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8064 Signed-off-by: Dave Airlie <airlied@redhat.com>
8065
8066commit da8ee26023fc2868fe970471195a5f3c86fb574b
8067Author: Dave Airlie <airlied@redhat.com>
8068Date: Wed Jan 9 12:51:45 2013 +1000
8069
8070 xfree86/hotplug: cleanup properly if the screen fails to initialise
8071
8072 Due to another bug, the modesetting/udl driver would fail to init properly
8073 on hotplug, when it did the code didn't clean up properly, and on removing
8074 the device the server could crash.
8075
8076 Found in F18 testing.
8077
8078 Signed-off-by: Dave Airlie <airlied@redhat.com>
8079 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
8080
8081commit e21e183059df5975e7086850d1931edb2c1bbd06
8082Author: Marcin Slusarz <marcin.slusarz@gmail.com>
8083Date: Thu Feb 21 09:02:44 2013 +1000
8084
8085 os: use libunwind to generate backtraces
8086
8087 Libunwind generates backtraces much more reliably than glibc's "backtrace".
8088
8089 Before:
8090 0: /opt/xserver/bin/X (0x400000+0x18ce36) [0x58ce36]
8091 1: /opt/xserver/bin/X (xorg_backtrace+0x9) [0x58d119]
8092 2: /opt/xserver/bin/X (0x400000+0x190d69) [0x590d69]
8093 3: /lib64/libpthread.so.0 (0x7fb904268000+0x10a90) [0x7fb904278a90]
8094 4: /lib64/libc.so.6 (ioctl+0x7) [0x7fb902fbf987]
8095 5: /usr/lib64/libdrm.so.2 (drmIoctl+0x28) [0x7fb90405ffa8]
8096 6: /usr/lib64/libdrm.so.2 (drmCommandWrite+0x1b) [0x7fb90406235b]
8097 7: /usr/lib64/libdrm_nouveau.so.2 (nouveau_bo_wait+0x89) [0x7fb902009719]
8098 8: /opt/xserver/lib/xorg/modules/drivers/nouveau_drv.so (0x7fb90220e000+0x76f3) [0x7fb9022156f3]
8099 9: /opt/xserver/lib/xorg/modules/libexa.so (0x7fb9019c7000+0xbae0) [0x7fb9019d2ae0]
8100 10: /opt/xserver/bin/X (0x400000+0x17d2b3) [0x57d2b3]
8101 11: /opt/xserver/bin/X (0x400000+0xc9930) [0x4c9930]
8102 12: /opt/xserver/bin/X (0x400000+0x3a81a) [0x43a81a]
8103 13: /opt/xserver/bin/X (0x400000+0x3d6a1) [0x43d6a1]
8104 14: /opt/xserver/bin/X (0x400000+0x2c2ca) [0x42c2ca]
8105 15: /lib64/libc.so.6 (__libc_start_main+0xf5) [0x7fb902f019b5]
8106 16: /opt/xserver/bin/X (0x400000+0x2c60d) [0x42c60d]
8107 17: ?? [0x0]
8108
8109 After:
8110 0: /opt/xserver/bin/X (OsSigHandler+0x39) [0x590d69]
8111 1: /lib64/libpthread.so.0 (__restore_rt+0x0) [0x7fb904278a8f]
8112 2: /lib64/libc.so.6 (ioctl+0x7) [0x7fb902fbf987]
8113 3: /usr/lib64/libdrm.so.2 (drmIoctl+0x28) [0x7fb90405ffa8]
8114 4: /usr/lib64/libdrm.so.2 (drmCommandWrite+0x1b) [0x7fb90406235b]
8115 5: /usr/lib64/libdrm_nouveau.so.2 (nouveau_bo_wait+0x89) [0x7fb902009719]
8116 6: /opt/xserver/lib/xorg/modules/drivers/nouveau_drv.so (nouveau_exa_download_from_screen+0x1a3) [0x7fb9022156f3]
8117 7: /opt/xserver/lib/xorg/modules/libexa.so (exaGetImage+0x1f0) [0x7fb9019d2ae0]
8118 8: /opt/xserver/bin/X (miSpriteGetImage+0x173) [0x57d2b3]
8119 9: /opt/xserver/bin/X (compGetImage+0xb0) [0x4c9930]
8120 10: /opt/xserver/bin/X (ProcGetImage+0x55a) [0x43a81a]
8121 11: /opt/xserver/bin/X (Dispatch+0x341) [0x43d6a1]
8122 12: /opt/xserver/bin/X (main+0x3ba) [0x42c2ca]
8123 13: /lib64/libc.so.6 (__libc_start_main+0xf5) [0x7fb902f019b5]
8124 14: /opt/xserver/bin/X (_start+0x29) [0x42c60d]
8125 15: ? (?+0x29) [0x29]
8126
8127 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
8128 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8129 Tested-by: Knut Petersen <knut.petersen@t-online.de>
8130
8131commit e27b2e61632b220ddc36d0d0387581a9f4774f21
8132Author: Peter Harris <pharris@opentext.com>
8133Date: Tue Feb 19 16:08:52 2013 -0500
8134
8135 xkb: Set nIndicators in XkbGetIndicatorMap
8136
8137 Xlib doesn't use this value (it computes it from the reply length
8138 instead) which is why nobody has noticed yet. But the spec
8139 http://www.x.org/releases/X11R7.7/doc/kbproto/xkbproto.html
8140 says that it should be set.
8141
8142 Signed-off-by: Peter Harris <pharris@opentext.com>
8143 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8144 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8145
8146commit 3aac7a59dc6ef2d8bbf46ba5d37acdf6013e9450
8147Author: Daniel Martin <daniel.martin@secunet.com>
8148Date: Tue Dec 11 17:23:55 2012 +0100
8149
8150 ephyr: Add -resizeable option
8151
8152 With this option passed, ephyr windows can be resized like normal
8153 windows on the fly, without the need of an explicit parent window.
8154
8155 Signed-off-by: Daniel Martin <daniel.martin@secunet.com>
8156 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8157 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8158
8159commit f9198e278becec158b570204cf9fc1de822ac76b
8160Author: Peter Hutterer <peter.hutterer@who-t.net>
8161Date: Mon Feb 27 16:43:10 2012 +1000
8162
8163 dix: FreeAllAtoms() on reset
8164
8165 ==5712== 6 bytes in 1 blocks are still reachable in loss record 17 of 585
8166 ==5712== at 0x4A074CD: malloc (vg_replace_malloc.c:236)
8167 ==5712== by 0x3D1DE885B1: strndup (strndup.c:46)
8168 ==5712== by 0x41CB71: MakeAtom (atom.c:121)
8169 ==5712== by 0x55AE3E: XIGetKnownProperty (xiproperty.c:401)
8170 ==5712== by 0x4251C9: AddInputDevice (devices.c:312)
8171 ==5712== by 0x42AC0C: AllocDevicePair (devices.c:2657)
8172 ==5712== by 0x425E6E: InitCoreDevices (devices.c:677)
8173 ==5712== by 0x5ACA05: main (main.c:257)
8174
8175 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8176 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
8177
8178commit 88517ced1fa2e621be2d05a319b522b3111da516
8179Merge: 82425c6 73974dd
8180Author: Peter Hutterer <peter.hutterer@who-t.net>
8181Date: Fri Feb 15 11:58:52 2013 +1000
8182
8183 Merge branch 'master' of git+ssh://people.freedesktop.org/~alanc/xserver into next
8184
8185commit 82425c66e7a8c436b5be7ad44880dd4be6c1f06a
8186Author: Peter Hutterer <peter.hutterer@who-t.net>
8187Date: Thu Feb 14 10:21:47 2013 +1000
8188
8189 xfree86: remove redundant declaration of inputInfo
8190
8191 xf86Cursor.c:19:18: warning: redundant redeclaration of 'inputInfo'
8192 [-Wredundant-decls]
8193 In file included from xf86Cursor.c:18:0:
8194 ../../../include/inputstr.h:614:57: note: previous declaration of
8195 'inputInfo' was here
8196
8197 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8198 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8199 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
8200
8201commit 6133c417595a9c8fc158b0d68fd4a7a2c58fdc47
8202Author: Peter Hutterer <peter.hutterer@who-t.net>
8203Date: Thu Feb 14 10:15:33 2013 +1000
8204
8205 xkb: remove unused variable 'names'
8206
8207 xkb.c: In function '_XkbSetNamesCheck':
8208 xkb.c:3987:18: warning: variable 'names' set but not used
8209 [-Wunused-but-set-variable]
8210
8211 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8212 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8213 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
8214
8215commit 63d00c6b146d7f5fb80f20f19066088059042a5e
8216Author: Peter Hutterer <peter.hutterer@who-t.net>
8217Date: Thu Feb 14 10:13:53 2013 +1000
8218
8219 Xext: rename two shadowing variables
8220
8221 panoramiX.c: In function 'PanoramiXCreateConnectionBlock':
8222 panoramiX.c:599:10: warning: declaration of 'disableBackingStore' shadows a
8223 global declaration [-Wshadow]
8224 In file included from ../include/windowstr.h:60:0,
8225 from panoramiX.c:47:
8226 ../include/opaque.h:56:52: warning: shadowed declaration is here [-Wshadow]
8227
8228 panoramiX.c: In function 'PanoramiXConsolidate':
8229 panoramiX.c:834:19: warning: declaration of 'pScreen' shadows a previous
8230 local [-Wshadow]
8231 panoramiX.c:813:15: warning: shadowed declaration is here [-Wshadow]
8232
8233 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8234 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8235 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
8236
8237commit 8bbea9f48f43e54c55c8b60ea36bda3134a86af2
8238Author: Peter Hutterer <peter.hutterer@who-t.net>
8239Date: Thu Feb 14 10:09:53 2013 +1000
8240
8241 Xext: renaming shadowing variable
8242
8243 xvdisp.c: In function 'ProcXvStopVideo':
8244 xvdisp.c:712:11: warning: declaration of 'rc' shadows a previous local
8245 [-Wshadow]
8246 xvdisp.c:705:17: warning: shadowed declaration is here [-Wshadow]
8247
8248 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8249 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8250 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
8251
8252commit 7b79a2e4a11b5c5f0ebaa495828725e235d2b08e
8253Author: Peter Hutterer <peter.hutterer@who-t.net>
8254Date: Thu Feb 14 10:06:55 2013 +1000
8255
8256 fb: drop two unneeded shadowing variables
8257
8258 fbpict.c: In function 'fbGlyphs':
8259 fbpict.c:188:6: warning: declaration of 'x' shadows a previous local
8260 [-Wshadow]
8261 fbpict.c:111:9: warning: shadowed declaration is here [-Wshadow]
8262 fbpict.c:188:9: warning: declaration of 'y' shadows a previous local
8263 [-Wshadow]
8264 fbpict.c:111:12: warning: shadowed declaration is here [-Wshadow]
8265
8266 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8267 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8268 Reviewed-by: Soren Sandmann <ssp@redhat.com>
8269 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
8270
8271commit 858d8b19b3a5bb59e5c5f9a9e68adce2495b5e31
8272Author: Peter Hutterer <peter.hutterer@who-t.net>
8273Date: Thu Feb 14 10:04:06 2013 +1000
8274
8275 xfree86: drop unused prevSIGIO
8276
8277 Unused as of 5d309af2ed93e91c7d72f548a11052051efbb40f
8278
8279 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8280 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8281 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
8282
8283commit 592d35aef0a8089a344543cf5a425e0537c0431b
8284Author: Peter Hutterer <peter.hutterer@who-t.net>
8285Date: Thu Feb 14 10:02:02 2013 +1000
8286
8287 randr: fix "set but unused" warnings
8288
8289 rrcrtc.c: In function 'RRCrtcDetachScanoutPixmap':
8290 rrcrtc.c:366:9: warning: variable 'ret' set but not used
8291 [-Wunused-but-set-variable]
8292 rrcrtc.c: In function 'rrCheckPixmapBounding':
8293 rrcrtc.c:505:13: warning: variable 'ret' set but not used
8294 [-Wunused-but-set-variable]
8295 rrcrtc.c:445:9: warning: unused variable 'i' [-Wunused-variable]
8296
8297 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8298 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8299 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
8300
8301commit 90642948cc78834d95f7a3bddaac7ff77b68ed7e
8302Merge: 9a35d42 eda7dbf
8303Author: Keith Packard <keithp@keithp.com>
8304Date: Thu Feb 14 11:05:48 2013 -0800
8305
8306 Merge remote-tracking branch 'jeremyhu/master'
8307
8308commit 9a35d4240e2aa91ac104f0f9f86f83ff9a2d3d04
8309Author: Peter Hutterer <peter.hutterer@who-t.net>
8310Date: Thu Feb 14 16:31:13 2013 +1000
8311
8312 os: fix pnprintf OOB buffer read for unterminated length modifiers
8313
8314 Format strings with length modifiers but missing format specifier like "%0"
8315 will read one byte past the array size.
8316
8317 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8318 Reviewed-by: Keith Packard <keithp@keithp.com>
8319 Signed-off-by: Keith Packard <keithp@keithp.com>
8320
8321commit eda7dbff5a9f35fefe1e3dedacb82daadbf5945e
8322Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
8323Date: Sat Feb 9 20:53:02 2013 -0800
8324
8325 XORG_TLS: Pick the first option that works (ie: prefer __thread)
8326
8327 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
8328
8329commit 5da82f872daf25939a8fa04f1561b01f0315a623
8330Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
8331Date: Sat Feb 9 20:40:10 2013 -0800
8332
8333 os: Ensure <dix-config.h> is included in strndup.c
8334
8335 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
8336 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8337
8338commit 67eaf4b46f678bc904f47dfcab6655e170d843e0
8339Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
8340Date: Sat Feb 9 20:34:33 2013 -0800
8341
8342 XQuartz: Ensure <dix-config.h> is included in capabilities.c
8343
8344 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
8345 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8346
8347commit 955d434f4d755d00a24ae4068b9957765989d672
8348Author: Keith Packard <keithp@keithp.com>
8349Date: Wed Feb 13 21:39:37 2013 -0800
8350
8351 Update to version 1.3.99.902 (1.14 RC2)
8352
8353 Signed-off-by: Keith Packard <keithp@keithp.com>
8354
8355commit 7115f6c709898a5124b67e19c61dc01334471358
8356Author: Aaron Plattner <aplattner@nvidia.com>
8357Date: Wed Jan 23 16:58:47 2013 -0800
8358
8359 randr: unref the provider shared pixmap the appropriate number of times
8360
8361 When an RandR shared pixmap is created in rrCreateSharedPixmap, it has a refcnt
8362 of 1. Then, PixmapShareToSlave bumps the refcnt to 2. However, there's no
8363 corresponding PixmapUnshareFromSlave where the refcnt can be decreased again,
8364 and there's no convenient common place where the refcnt can be decremented when
8365 the slave pixmap is destroyed.
8366
8367 Fix this by just unreffing the pixmap twice in RRCrtcDetachScanoutPixmap.
8368
8369 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
8370 Reviewed-by: Dave Airlie <airlied@redhat.com>
8371 Signed-off-by: Keith Packard <keithp@keithp.com>
8372
8373commit da92690107d90061205340d4cdc98b73b59db9b2
8374Author: Aaron Plattner <aplattner@nvidia.com>
8375Date: Mon Feb 11 14:31:56 2013 -0800
8376
8377 xf86: use nt_list_for_each_entry_safe to walk InputHandlers in xf86Wakeup
8378
8379 This is necessary when the input handler deletes itself from the
8380 list. Bug found by Maarten Lankhorst, this patch uses the list macros
8381 instead of open-coding the fix.
8382
8383 Signed-off-by: Keith Packard <keithp@keithp.com>
8384 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8385
8386commit d0a14877872b6a33172fb8c8f335792a28b9499b
8387Author: Bryce Harrington <bryce@canonical.com>
8388Date: Fri Feb 8 15:56:02 2013 -0800
8389
8390 xfree86: Man page shouldn't say Device is mandatory anymore
8391
8392 man xorg.conf states that the 'Device' identifier is required in the
8393 'Screen' section, yet current xserver defaults properly and boots up
8394 fine without it.
8395
8396 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=20742
8397 Signed-off-by: Bryce Harrington <bryce@canonical.com>
8398 Signed-off-by: Keith Packard <keithp@keithp.com>
8399
8400commit 5e91054aa070ea09b61325b7ec479d767b02730e
8401Author: Bryce Harrington <bryce@canonical.com>
8402Date: Fri Feb 8 15:56:01 2013 -0800
8403
8404 xfree86: Use fbdev/vesa driver on Oaktrail, Medfield, CDV rather than -intel
8405
8406 Instead of defaulting to -intel for Oaktrail, Medfield, and CDV chips,
8407 default to -fbdev. For Poulsbo (only), attempt to use -psb if it's
8408 installed, and fallback to fbdev otherwise. All other Intel chips
8409 should use -intel.
8410
8411 This fixed an issue where -intel would load on these chips and cause a
8412 boot failure. Newer -intel drivers avoid the boot hang, but it's still
8413 the wrong driver to load, so why take chances.
8414
8415 The patch was originally created by Stefan Dirsch for OpenSUSE. We have
8416 included it in our stable release (Ubuntu "quantal" 12.10) since
8417 December.
8418
8419 ref: https://bugzilla.novell.com/show_bug.cgi?id=772279
8420 ref: https://bugs.launchpad.net/ubuntu/+bug/1069031
8421 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=60514
8422 Signed-off-by: Bryce Harrington <bryce@canonical.com>
8423 Signed-off-by: Keith Packard <keithp@keithp.com>
8424
8425commit c1602d1c17967bdd4db9db19b3a9c0dfca6a58aa
8426Author: Dave Airlie <airlied@gmail.com>
8427Date: Tue Feb 5 07:46:06 2013 -0800
8428
8429 randr: bump advertised RandR version to 1.4
8430
8431 Signed-off-by: Dave Airlie <airlied@redhat.com>
8432 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
8433 Tested-by: Aaron Plattner <aplattner@nvidia.com>
8434 Signed-off-by: Keith Packard <keithp@keithp.com>
8435
8436commit 9f79e93b6b3416055d08a0e8f9f16d5fd0649e36
8437Author: Peter Hutterer <peter.hutterer@who-t.net>
8438Date: Wed Jan 30 03:20:07 2013 +0000
8439
8440 Short-cut the input device cleanup process during AbortServer()
8441
8442 If we're about to abort, we're already in the signal handler and cannot call
8443 down to the default device cleanup routines (which reset, free, alloc, and
8444 do a bunch of other things).
8445
8446 Add a new DEVICE_ABORT mode to signal a driver's DeviceProc that it must
8447 reset the hardware if needed but do nothing else. An actual HW reset is only
8448 required for some drivers dealing with the HW directly.
8449
8450 This is largely backwards-compatible, hence the input ABI minor bump only.
8451
8452 Drivers we care about either return BadValue on a mode that's not
8453 DEVICE_{INIT|ON|OFF|CLOSE} or print an error and return BadValue. Exception
8454 here is vmmouse, which currently ignores it and would not reset anything.
8455 This should be fixed if the reset is required.
8456
8457 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8458 Reviewed-by: Keith Packard <keithp@keithp.com>
8459 Signed-off-by: Keith Packard <keithp@keithp.com>
8460
8461commit b58221f9da8c549d979215271359c6cd88b5568a
8462Author: Peter Hutterer <peter.hutterer@who-t.net>
8463Date: Fri Feb 8 14:52:02 2013 +1000
8464
8465 dix: support the transformation matrix for relative devices.
8466
8467 The transformation matrix we previously stored was a scaled matrix based on
8468 the axis ranges of the device. For relative movements, the scaling is not
8469 required (or desired).
8470
8471 Store two separate matrices, one as requested by the client, one as the
8472 product of [scale . matrix . inv_scale]. Depending on the type of movement,
8473 apply the respective matrix.
8474
8475 For relative movements, also drop the translation component since it doesn't
8476 really make sense to use that bit.
8477
8478 Input ABI 19
8479
8480 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8481 Reviewed-by: Keith Packard <keithp@keithp.com>
8482 Signed-off-by: Keith Packard <keithp@keithp.com>
8483
8484commit b173eb2ae3349c557db1ff9e424fa540b8289bb2
8485Author: Keith Packard <keithp@keithp.com>
8486Date: Fri Feb 8 08:57:43 2013 -0800
8487
8488 os: Round fraction in pnprintf %f format
8489
8490 Truncating the fraction part leads to a test failure where -1203.30 is
8491 printed as -1203.29. Round this to the nearest value instead by adding
8492 0.5 before converting to an integer
8493
8494 Signed-off-by: Keith Packard <keithp@keithp.com>
8495
8496commit 509b3c3dc82e7abce1900d5e1cddd90f23be5a87
8497Author: Carlos Garnacho <carlosg@gnome.org>
8498Date: Wed Feb 6 14:07:22 2013 +0100
8499
8500 dix: Set focus field on XI2 crossing events
8501
8502 Set on DeviceEnterLeaveEvent() the xXIEnterEvent->focus field
8503 similarly to how the CoreEnterLeaveEvent() function above does
8504 for core events.
8505
8506 This fixes bug https://bugzilla.gnome.org/show_bug.cgi?id=677329
8507 reported to GTK+, where focus handling on window managers with
8508 sloppy focus or no window manager present was broken due to this
8509 field being always set to FALSE.
8510
8511 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
8512 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8513 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8514
8515commit 3e4be4033aed78b2bb3a18d51f0963989efd1af3
8516Author: Peter Hutterer <peter.hutterer@who-t.net>
8517Date: Fri Jan 25 11:47:32 2013 +1000
8518
8519 dix: when shutting down slave devices, shut down xtest devices last
8520
8521 XTest devices are the first ones in the list, being initialised together
8522 with the master devices. If we disable the devices in-order and a device has
8523 a button down when being disabled, the XTest device is checked for a
8524 required button release (xkbAccessX.c's ProcessPointerEvent). This fails if
8525 the device is already NULL.
8526
8527 Instead of putting the check there, disable the devices in the reverse order
8528 they are initialised. Disable physical slaves first, then xtest devices,
8529 then the master devices.
8530
8531 Testcase: shut down server with a button still held down on a physical
8532 device
8533
8534 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8535 Reviewed-by: Keith Packard <keithp@keithp.com>
8536
8537commit 0d5bb882600ee7734af034fbea935a79d21d1e70
8538Merge: b33fcb1 61a99af
8539Author: Peter Hutterer <peter.hutterer@who-t.net>
8540Date: Fri Feb 8 14:10:52 2013 +1000
8541
8542 Merge branch 'ptraccel-fixes' into for-keith
8543
8544commit 61a99aff9d33728a0b67920254d2d4d79f80cf39
8545Author: Peter Hutterer <peter.hutterer@who-t.net>
8546Date: Fri Jan 11 14:22:07 2013 +1000
8547
8548 dix: pre-scale relative events from abs devices to desktop ratio (#31636)
8549
8550 Absolute devices may send relative events depending on the mode (synaptics
8551 by default, wacom per option). The relative events are added to the previous
8552 position, converted into device coordinates and then scaled into desktop
8553 coordinates for pointer movement.
8554
8555 Because the device range must be mapped into the desktop coordinate range,
8556 this results in uneven scaling depending dimensions, e.g. on a setup with
8557 width == 2 * height, a relative movement of 10/10 in device coordinates
8558 results in a cursor movement of 20/10 (+ acceleration)
8559
8560 Other commonly user-visible results:
8561 * the touchpad changing acceleration once an external monitor as added.
8562 * drawing a circle on a wacom tablet in relative mode gives an ellipsis in
8563 the same ratio as the desktop dimensions.
8564
8565 Solution: pre-scale the incoming relative x/y coordinates by width/height
8566 ratio of the total desktop size. Then add them to the previous
8567 coordinates and scale back with the previous mapping, which will undo the
8568 pre-scaling and give us the right movement.
8569
8570 X.Org Bug 31636 <http://bugs.freedesktop.org/show_bug.cgi?id=31636>
8571
8572 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8573 Reviewed-by: Keith Packard <keithp@keithp.com>
8574
8575commit a6ba2b79ae8ad0fdee3f208d5e030b012df48785
8576Author: Peter Hutterer <peter.hutterer@who-t.net>
8577Date: Fri Jan 11 08:53:24 2013 +1000
8578
8579 dix: unify prefix for ptraccel debugging in DebugAccelF macro
8580
8581 If we're already using our own custom macro, might as well use it properly.
8582
8583 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8584 Reviewed-by: Keith Packard <keithp@keithp.com>
8585
8586commit 0d7d79406011169ad95e23c6e937f6d15a5ec3cc
8587Author: Peter Hutterer <peter.hutterer@who-t.net>
8588Date: Fri Jan 11 08:52:08 2013 +1000
8589
8590 dix: use BUG_RETURN_VAL for an error message
8591
8592 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8593 Reviewed-by: Keith Packard <keithp@keithp.com>
8594
8595commit a0c38ea6cbad61edcfefff0e5dd6330edb706f13
8596Author: Peter Hutterer <peter.hutterer@who-t.net>
8597Date: Fri Jan 11 08:49:15 2013 +1000
8598
8599 dix: add some more info to a ptraccel debug msg
8600
8601 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8602 Reviewed-by: Keith Packard <keithp@keithp.com>
8603
8604commit 95125a7c0cbbbae40216a0497acdd863ddc645ed
8605Author: Peter Hutterer <peter.hutterer@who-t.net>
8606Date: Thu Jan 10 13:19:27 2013 +1000
8607
8608 dix: fix ptraccel debugging printfs
8609
8610 This is mostly sigsafe code, so use sigsave printf. And update some fields
8611 to double that used to be int.
8612
8613 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8614 Reviewed-by: Keith Packard <keithp@keithp.com>
8615
8616commit b33fcb149710a28fd8767b2307a97bf367de695e
8617Author: Andreas Wettstein <wettstein509@solnet.ch>
8618Date: Tue Jan 29 21:49:20 2013 +0100
8619
8620 xkb: Fix repeat behaviour of redirect and message actions
8621
8622 The redirect and the message action filter functions implicitly assumed that
8623 when they receive an event for the same keycode they were activated for, that
8624 this is the a release of the key that activated the filter. This is not true
8625 if the key autorepeats. Due to the incorrect assumption, the effective key
8626 repeat rate was effectively halved.
8627
8628 Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
8629 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8630 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8631
8632commit 8571c648a79444bcee9a0fe6e395129116372f49
8633Author: Peter Hutterer <peter.hutterer@who-t.net>
8634Date: Sat Jan 26 15:53:08 2013 +1000
8635
8636 Xext: if a root window is given in XTestFakeInput, move to that
8637
8638 For absolute events, if the client specifies a screen number offset the
8639 coordinates by that. And add a new flag so we know when _not_ to add the
8640 screen offset in GPE.
8641
8642 Without this offset and the flag, GPE would simply add the offset of the
8643 current screen if POINTER_SCREEN is set.
8644
8645 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8646 Reviewed-by: Keith Packard <keithp@keithp.com>
8647
8648commit 9fd6cb89539fde44a41ae5183c89ef9c8831c8dd
8649Author: Peter Hutterer <peter.hutterer@who-t.net>
8650Date: Tue Jan 29 15:13:44 2013 +1000
8651
8652 Xext: pass the current screen to miProcessDeviceEvent() from xtest calls
8653
8654 Not passing in a screen means we skip the screen crossing updates, so a
8655 xtest event that changes between ScreenRecs won't do so until the next
8656 physical event comes in or never, whichever comes earlier.
8657
8658 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8659 Reviewed-by: Keith Packard <keithp@keithp.com>
8660
8661commit a191dbfe850ed9c6440346f59cb0078e0e844edc
8662Author: Sybren van Elderen <sowmestno@msn.com>
8663Date: Tue Jan 29 15:43:57 2013 +1000
8664
8665 dix: when scaling from desktop coord, take the total desktop size (#51904)
8666
8667 Scaled is already in desktop coordinates, take the total width into account,
8668 not just the current screen's width.
8669
8670 Fixes Xdmx pointer position calculation.
8671
8672 X.Org Bug 51904 <http://bugs.freedesktop.org/show_bug.cgi?id=51904>
8673
8674 Signed-off-by: Sybren van Elderen <sowmestno@msn.com>
8675 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8676 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8677
8678commit 1cb19803f0f8dfd1e0fb9d189afe2262e24a0be5
8679Author: Peter Hutterer <peter.hutterer@who-t.net>
8680Date: Tue Jan 29 12:51:15 2013 +1000
8681
8682 include: fix typo in list description
8683
8684 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8685
8686commit 3d35dfcf5bad1b0a028fbecd65cb6cf6ebf12503
8687Author: Ted Felix <ted@tedfelix.com>
8688Date: Tue Jan 29 16:36:48 2013 +1000
8689
8690 xfree86: bail on misformed acpi strings (#73227)
8691
8692 If acpid sends a string in a format that we can't parse, bail out instead of
8693 potentially dereferencing a NULL-pointer.
8694
8695 X.Org Bug 73227 <http://bugs.freedesktop.org/show_bug.cgi?id=73227>
8696
8697 Signed-off-by: Ted Felix <ted@tedfelix.com>
8698 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8699 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8700
8701commit fdc451588816c4bc798d54e56316530e9066be80
8702Author: Peter Hutterer <peter.hutterer@who-t.net>
8703Date: Tue Jan 29 11:01:29 2013 +1000
8704
8705 Xi: limit valuator copy to valuator array size (#59939)
8706
8707 mask[(MAX_VALUATORS + 7)/8] is larger than data[MAX_VALUATORS], so static
8708 code checkers think we may be running OOB on the data array. Mask is
8709 initialized to 0, so this should not happen, but change it anyway to shut up
8710 code analyzer noise.
8711
8712 X.Org Bug 59939 <http://bugs.freedesktop.org/show_bug.cgi?id=59939>
8713
8714 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8715 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8716
8717commit 48bc30c5413a1be0039fa77affcbbb4fe677479f
8718Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8719Date: Tue Jan 29 10:24:32 2013 +1000
8720
8721 Xext: avoid null-pointer dereference in XTestFakeInput (#59937)
8722
8723 dv is still NULL at this point, so return firstValuator instead (which is
8724 the same value dv->firstValuator would be once initialized)
8725
8726 X.Org Bug 59937 <http://bugs.freedesktop.org/show_bug.cgi?id=59937>
8727
8728 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8729 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8730 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8731
8732commit 1058fcf57fdcb94d92e7b5f4483b347853d5f8e6
8733Author: Peter Hutterer <peter.hutterer@who-t.net>
8734Date: Sat Jan 26 14:13:33 2013 +1000
8735
8736 dmx: don't include dmx-config.h from xdmxconfig (#37502)
8737
8738 dmx-config.h is a server header which includes dix-config.h. That again
8739 defines a bunch of server-specifics, including setting the size of XID to
8740 32 bit.
8741
8742 libX11 uses unsigned long (8 bits on x86_64). XGCValues thus ends up being
8743 16 bytes smaller in xdmxconfig than in the library, causing garbage to be
8744 sent to the server.
8745
8746 X.Org Bug 37502 <http://bugs.freedesktop.org/show_bug.cgi?id=37502>
8747
8748 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8749 Reviewed-by: Keith Packard <keithp@keithp.com>
8750
8751commit 73974dd7ea9ca4d4cdd5464cb813088a6ee9770b
8752Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8753Date: Sun Jan 27 15:42:02 2013 -0800
8754
8755 Avoid memory leak in ddc resort() if find_header() fails
8756
8757 Call find_header first, returning on failure before calling malloc.
8758
8759 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8760 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8761
8762commit b1129a1f1771c9d1653cc15aae94a614f081638a
8763Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8764Date: Sun Jan 27 14:00:54 2013 -0800
8765
8766 xf86XvMCScreenInit: Avoid leak if dixRegisterPrivateKey fails
8767
8768 Found by parfait 1.1 memory analyser:
8769 Memory leak of pointer 'pAdapt' allocated with malloc((88 * num_adaptors))
8770 at line 162 of hw/xfree86/common/xf86xvmc.c in function 'xf86XvMCScreenInit'.
8771 'pAdapt' allocated at line 158 with malloc((88 * num_adaptors)).
8772
8773 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8774 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8775
8776commit 563db909bcf965b6103c1807bf9f00ede957077d
8777Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8778Date: Sun Jan 27 13:55:50 2013 -0800
8779
8780 Avoid memory leak on realloc failure in localRegisterFreeBoxCallback
8781
8782 Also avoids leaving invalid pointers in structures if realloc had to
8783 move them elsewhere to make them larger.
8784
8785 Found by parfait 1.1 code analyzer:
8786 Memory leak of pointer 'newCallbacks' allocated with realloc(((char*)offman->FreeBoxesUpdateCallback), (8 * (offman->NumCallbacks + 1)))
8787 at line 328 of hw/xfree86/common/xf86fbman.c in function 'localRegisterFreeBoxCallback'.
8788 'newCallbacks' allocated at line 320 with realloc(((char*)offman->FreeBoxesUpdateCallback), (8 * (offman->NumCallbacks + 1))).
8789 newCallbacks leaks when newCallbacks != NULL at line 327.
8790 Memory leak of pointer 'newPrivates' allocated with realloc(((char*)offman->devPrivates), (8 * (offman->NumCallbacks + 1)))
8791 at line 328 of hw/xfree86/common/xf86fbman.c in function 'localRegisterFreeBoxCallback'.
8792 'newPrivates' allocated at line 324 with realloc(((char*)offman->devPrivates), (8 * (offman->NumCallbacks + 1))).
8793 newPrivates leaks when newCallbacks == NULL at line 327.
8794
8795 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8796 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8797
8798commit 08f75d3a9661c6c32800e1b4f150626200b889d9
8799Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8800Date: Sun Jan 27 13:50:30 2013 -0800
8801
8802 Avoid NULL pointer dereference in xf86TokenToOptinfo if token not found
8803
8804 Reported by parfait 1.1 code analyzer:
8805
8806 Error: Null pointer dereference (CWE 476)
8807 Read from null pointer 'p'
8808 at line 746 of hw/xfree86/common/xf86Option.c in function 'xf86TokenToOptName'.
8809 Function 'xf86TokenToOptinfo' may return constant 'NULL' at line 721, called at line 745.
8810 Null pointer introduced at line 721 in function 'xf86TokenToOptinfo'.
8811
8812 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8813 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8814
8815commit c1c01e350834a23161b33bd34b2fa9c01d02a65b
8816Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8817Date: Sun Jan 27 13:10:08 2013 -0800
8818
8819 Make xf86ValidateModes actually copy clock range list to screen pointer
8820
8821 Our in-house parfait 1.1 code analysis tool complained that every exit
8822 path from xf86ValidateModes() in hw/xfree86/common/xf86Mode.c leaks the
8823 storeClockRanges allocation made at line 1501 with XNFalloc.
8824
8825 Investigating, it seems that this code to copy the clock range list to
8826 the clockRanges list in the screen pointer is just plain insane, and
8827 according to git, has been since we first imported it from XFree86.
8828
8829 We start at line 1495 by walking the linked list from scrp->clockRanges
8830 until we find the end. But that was just a diversion, since we've found
8831 the end and immediately forgotten it, and thus at 1499 we know that
8832 storeClockRanges is NULL, but that's not a problem since we're going to
8833 immediately overwrite that value as the first thing in the loop.
8834
8835 So we move on through this loop at 1499, which takes us through the
8836 linked list from the clockRanges variable, and for every entry in
8837 that list allocates a new structure and copies cp to it. If we've
8838 not filled in the screen's clockRanges pointer yet, we set it to
8839 the first storeClockRanges we copied from cp. Otherwise, as best
8840 I can tell, we just drop it into memory and let it leak away, as
8841 parfait warned.
8842
8843 And then we hit the loop action, which if we haven't hit the end of
8844 the cp list, advances cp to the next item in the list, and then just
8845 for the fun of it, also sets storeClockRanges to the ->next pointer it
8846 has just copied from cp as well, even though it's going to overwrite
8847 it as the very first instruction in the loop body.
8848
8849 v2: rewritten using nt_list_* macros from Xorg's list.h header
8850
8851 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8852 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8853
8854commit 89badba082c81d20fe35cb064c16e131ff288ca3
8855Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8856Date: Sun Jan 27 12:08:47 2013 -0800
8857
8858 Free keymap on error in Xephyr's hostx_load_keymap
8859
8860 Found by parfait 1.1 code analyser:
8861 Memory leak of pointer 'keymap' allocated with XGetKeyboardMapping(HostX.dpy, min_keycode, ((max_keycode - min_keycode) + 1), &host_width)
8862 at line 861 of hw/kdrive/ephyr/hostx.c in function 'hostx_load_keymap'.
8863 'keymap' allocated at line 845 with XGetKeyboardMapping(HostX.dpy, min_keycode, ((max_keycode - min_keycode) + 1), &host_width).
8864
8865 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8866 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8867
8868commit 48b94651205b175760904e448f94111d1ab85e13
8869Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8870Date: Sun Jan 27 11:20:45 2013 -0800
8871
8872 Stop leaking overlayWin in PanoramiXCompositeGetOverlayWindow error paths
8873
8874 Found by parfait 1.1 code analyzer:
8875
8876 Error: Memory leak (CWE 401)
8877 Memory leak of pointer 'overlayWin' allocated with malloc(72)
8878 at line 806 of composite/compext.c in function 'PanoramiXCompositeGetOverlayWindow'.
8879 pointer allocated at line 794 with malloc(72).
8880 <unknown> leaks when rc != 0 at line 804.
8881 at line 816 of composite/compext.c in function 'PanoramiXCompositeGetOverlayWindow'.
8882 pointer allocated at line 794 with malloc(72).
8883 <unknown> leaks when pOc == NULL at line 815.
8884 at line 825 of composite/compext.c in function 'PanoramiXCompositeGetOverlayWindow'.
8885 pointer allocated at line 794 with malloc(72).
8886 <unknown> leaks when cs->pOverlayWin == NULL at line 822
8887 and compCreateOverlayWindow(pScreen) == 0 at line 823.
8888 at line 834 of composite/compext.c in function 'PanoramiXCompositeGetOverlayWindow'.
8889 pointer allocated at line 794 with malloc(72).
8890 <unknown> leaks when rc != 0 at line 832.
8891
8892 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8893 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8894
8895commit f4a9332ad149ed15353a9c482563bdd042d0b403
8896Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8897Date: Sun Jan 27 10:06:42 2013 -0800
8898
8899 Handle failure to create counter in init_system_idle_counter
8900
8901 Check for NULL pointer (which can be returned for multiple reasons)
8902 before trying to dereference it to add privates. To avoid memory leak
8903 in error path, delay malloc of privates until we're ready to add them.
8904
8905 In case we do return NULL up through SyncInitDeviceIdleTime, handle the
8906 possibility of getting NULL passed back down to SyncRemoveDeviceIdleTime.
8907
8908 As reported by parfait 1.1:
8909 Error: Null pointer dereference (CWE 476)
8910 Read from null pointer 'idle_time_counter'
8911 at line 2764 of xserver/Xext/sync.c in function 'init_system_idle_counter'.
8912 Function 'SyncCreateSystemCounter' may return constant 'NULL' at line 952, called at line 2756.
8913 Null pointer introduced at line 952 in function 'SyncCreateSystemCounter'.
8914
8915 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8916 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8917
8918commit 7fe5e6dfa5c1e71d8b7540b28c1d508687a2fbee
8919Author: Jasper St. Pierre <jstpierre@mecheye.net>
8920Date: Wed Jan 23 13:11:55 2013 -0500
8921
8922 protocol-versions: Bump minor version of XI
8923
8924 This was accidentally excluded when we added barriers.
8925
8926 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8927 Signed-off-by: Keith Packard <keithp@keithp.com>
8928
8929commit 70b127c9f1c53bdb42f078265e67f76b464deae2
8930Author: Aaron Plattner <aplattner@nvidia.com>
8931Date: Thu Jan 10 17:01:17 2013 -0800
8932
8933 config/udev: fix "removing GPU device" format string mistake
8934
8935 udev.c: In function 'device_removed':
8936 udev.c:270:9: warning: format '%d' expects argument of type 'int', but argument 3 has type 'const char *' [-Wformat]
8937
8938 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
8939 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8940 Signed-off-by: Keith Packard <keithp@keithp.com>
8941
8942commit 605dfc6804a05ff2bda5692fec26c37344fd95cb
8943Author: Dave Airlie <airlied@gmail.com>
8944Date: Tue Jan 22 07:39:53 2013 +1000
8945
8946 xserver: fix build regression since 91ab237358c6e33da854914d3de493a9cbea7637
8947
8948 inputstr, double defines TouchListener typedef, maybe some gcc handles it,
8949 but not all.
8950
8951 fixes tinderbox
8952
8953 Reported-by: Jon TURNEY <jon.turney@dronecode.org.uk>
8954 Signed-off-by: Dave Airlie <airlied@redhat.com>
8955 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8956 Signed-off-by: Keith Packard <keithp@keithp.com>
8957
8958commit 069d8ed3eb659c48dd2b0f8b7b8c11f092fdb362
8959Merge: 591c062 d6dcde7
8960Author: Keith Packard <keithp@keithp.com>
8961Date: Sun Jan 20 15:58:38 2013 -0800
8962
8963 Merge remote-tracking branch 'jturney/xserver-next'
8964
8965commit 591c06277bb120ab9615633f2d28addbd3a2aa5f
8966Merge: 6703a7c fa6ab7d
8967Author: Keith Packard <keithp@keithp.com>
8968Date: Sun Jan 20 15:52:26 2013 -0800
8969
8970 Merge remote-tracking branch 'whot/for-keith'
8971
8972commit cde7cbe9674e8a771f9a4e646c1772a46a8230fb
8973Author: Peter Hutterer <peter.hutterer@who-t.net>
8974Date: Thu Jan 10 13:20:12 2013 +1000
8975
8976 os: add support for %f to pnprintf
8977
8978 This is the lazy man's %f support. Print the decimal part of the number,
8979 then append a decimal point, then print the first two digits of the
8980 fractional part. So %f in sigsafe printing is really %.2f.
8981
8982 No boundary checks in place here.
8983
8984 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8985 Reviewed-by: Keith Packard <keithp@keithp.com>
8986
8987commit 20def57632583aef095ca18792c7fce16d2d9004
8988Author: Peter Hutterer <peter.hutterer@who-t.net>
8989Date: Thu Jan 10 13:24:05 2013 +1000
8990
8991 os: silently ignore length modifiers in pnprintf
8992
8993 Until we have support for them, ignore any length modifiers so we don't need
8994 to update all callers.
8995
8996 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8997 Reviewed-by: Keith Packard <keithp@keithp.com>
8998
8999commit f53b2012f39085d866f267dda1442a48ace3c5a5
9000Author: Peter Hutterer <peter.hutterer@who-t.net>
9001Date: Thu Jan 17 16:19:51 2013 +1000
9002
9003 test/signal-logging: simplify tests using sprintf
9004
9005 Ever looked at your own code and thought 'WTF was I thinking?'. yeah, that.
9006
9007 Instead of passing in the expected string just use sprintf to print the
9008 number for us and compare. In the end we're just trying to emulate printf
9009 behaviour anyway.
9010
9011 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9012 Reviewed-by: Keith Packard <keithp@keithp.com>
9013
9014commit d6dcde7a03bb38c17ffc4ec5f0ca1c161e54569f
9015Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9016Date: Sat Feb 11 12:22:17 2012 +0000
9017
9018 hw/xwin: Stop assuming WS_EX_APPWINDOW style in WM_SHOWWINDOW
9019
9020 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9021 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9022
9023commit 56e94403f8f9182e05428d895a983371c7737d2a
9024Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9025Date: Wed Jan 9 20:15:01 2013 +0000
9026
9027 hw/xwin: Use ITaskBarList interface to ensure show-on-taskbar state is updated correctly
9028
9029 Use ITaskBarList interface to ensure that the taskbar notices if the window has
9030 changed it's style in a way which affects if the taskbar shows it or not.
9031
9032 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9033 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9034
9035commit c94d1cb0a49106f44714f4511720a197cc549164
9036Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9037Date: Thu Jan 10 14:35:56 2013 +0000
9038
9039 hw/xwin: Ensure full styling is applied when the window is mapped
9040
9041 Move styling update code from WM_WM_HINTS_EVENT to a function UpdateStyle(),
9042 which is also invoked from WM_WM_MAP3, so everything which needs to be done
9043 to style the window happens when it is mapped
9044
9045 (Otherwise, the appearance of the window is sensitive to the timing of the
9046 notification of the windows appearance hint properties being set relative to
9047 window creation. e.g. see [1])
9048
9049 [1] http://sourceware.org/ml/cygwin-xfree/2012-06/msg00004.html
9050
9051 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9052 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9053
9054commit ef61f8cacc84080c9156675f9ce26a27e8a90ac1
9055Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9056Date: Sat Mar 31 18:45:28 2012 +0100
9057
9058 hw/xwin: Make sure that WM_WM_HINTS_EVENT does nothing for override-redirect windows
9059
9060 Future work: It looks like this code could be rationalized quite a lot: It might
9061 make sense to pull the checking for override-redirect up out of UpdateIcon() and
9062 UpdateName() and consolidate WM_WM_MAP2 and WM_WM_MAP3
9063
9064 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9065 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9066
9067commit 3628559e594fcbdfcc14b1e8fa60aa841f184e19
9068Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9069Date: Sun Feb 5 11:25:39 2012 +0000
9070
9071 hw/xwin: Add a new WM_WM_HINTS_EVENT event to update window style
9072
9073 Add a new WM_WM_HINTS_EVENT event to update window style if any of the
9074 properties which affect window style change
9075
9076 Check PropertyNotify events for any of the window properties which we consider
9077 to decide on the window style, and update the window style by sending a
9078 WM_WM_HINTS_EVENT message to the WM.
9079
9080 This allows the styling of the window to change during it's lifetime.
9081
9082 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9083 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9084
9085commit 066ecbd11d516ea68d7ebc7470232d01c5717546
9086Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9087Date: Thu Jan 10 14:37:45 2013 +0000
9088
9089 hw/xwin: Move reshape code from winUpdateWindowPosition() to the map event handler
9090
9091 Move reshape code, which was only used when handling a map event, from
9092 winUpdateWindowPosition(), to put it explicitly in the map event handler.
9093
9094 Remove 'reshape' parameter from winUpdatePosition().
9095
9096 (Note that there's no handling of the ShapeNotify event to notice when the
9097 window shape changes, instead we hook the screen SetShape procedure and reshape
9098 the native window then)
9099
9100 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9101 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9102
9103commit 852d1fb042f4160fe023a015f1c9a34126bf911a
9104Author: Ryan Pavlik <rpavlik@iastate.edu>
9105Date: Sat Dec 1 16:58:40 2012 +0000
9106
9107 hw/xwin: Add missing include xwin-config.h to winglobals.h
9108
9109 winglobals.h checks if RELOCATE_PROJECTROOT is defined to see if a declaration
9110 of g_fLogFileChanged is needed, so must include xwin-config.h
9111
9112 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
9113 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9114 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9115
9116commit ab686ce029208abf970a4bcd1435bf8411a44de9
9117Author: Ryan Pavlik <rpavlik@iastate.edu>
9118Date: Wed Oct 26 17:03:25 2011 -0500
9119
9120 include: Add RELOCATE_PROJECTROOT to xwin-config.h header
9121
9122 RELOCATE_PROJECTROOT is AC_DEFINED in configure.ac, but currently has no effect
9123 as it doesn't appear in any AC_CONFIG_HEADER header.
9124
9125 When packaged for Windows, we do not have a unix-style filesystem tree, where
9126 file needed by the X server can be found in fixed, absolute paths under the
9127 prefix (PROJECTROOT).
9128
9129 Instead, the filesystem tree containing files needed by the X server and clients
9130 will be installed with the directory containing the X server executable as the
9131 root directory of that tree.
9132
9133 (Typically, this will be in the Program Files directory, which does not have a
9134 fixed name, as it can be moved, localized, or added to to indicate x86 or x64
9135 binaries)
9136
9137 So, RELOCATE_PROJECTROOT is used to make a native Windows build of the X server
9138 look for various files (fonts, xkb data) in locations relative to the X server
9139 rather than at absolute paths, by translating those paths at run-time.
9140
9141 Additionally the XKEYSYMDB, XERRORDB, XLOCALEDIR env vars checked by libX11 are
9142 set appropriately for clients started by the X server.
9143
9144 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
9145 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9146 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9147
9148commit fa6ab7d9b2d7fd8184f1e068360607845f5c33ab
9149Merge: adde4e6 0e1ab43
9150Author: Peter Hutterer <peter.hutterer@who-t.net>
9151Date: Fri Jan 11 14:58:17 2013 +1000
9152
9153 Merge branch 'pointer-emulation-fixes-56558-v2' into for-keith
9154
9155commit adde4e64480315dc5b47a727ee37d86f5cd8584f
9156Author: Peter Hutterer <peter.hutterer@who-t.net>
9157Date: Thu Jan 10 10:33:05 2013 +1000
9158
9159 dix: typo fix in comment
9160
9161 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9162
9163commit 05ed095dd8d6cf939b4ebd9a59d70ce32705df7c
9164Author: Benjamin Tissoires <benjamin.tissoires@gmail.com>
9165Date: Wed Jan 9 19:32:19 2013 +0100
9166
9167 dix: fix error logging occuring in signal context of GetTouchEvents
9168
9169 GetTouchEvents is usually called in a signal context.
9170 Calling ErrorF for the error messages leads to X complaining about log:
9171
9172 (EE) BUG: triggered 'if (inSignalContext)'
9173 (EE) BUG: log.c:484 in LogVMessageVerb()
9174 (EE) Warning: attempting to log data in a signal unsafe manner while in signal context.
9175 Please update to check inSignalContext and/or use LogMessageVerbSigSafe() or ErrorFSigSafe().
9176 The offending log format message is:
9177 %s: Attempted to start touch without x/y (driver bug)
9178
9179 Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
9180 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9181 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9182
9183commit f4a58469a298c226668fd8dce375bf22331c902d
9184Author: Peter Hutterer <peter.hutterer@who-t.net>
9185Date: Wed Jan 9 13:58:56 2013 +1000
9186
9187 xfree86: don't access the old input handler after freeing it
9188
9189 Introduced in 323869f3298cbbfe864af9404a8aed1bf7995d79
9190
9191 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9192 Reviewed-by: Dave Airlie <airlied@redhat.com>
9193
9194commit 205cfbd6d9824fb9a67c21b19bc8f1e66c9df4d2
9195Author: Dave Airlie <airlied@gmail.com>
9196Date: Sat Jan 5 18:35:42 2013 +1000
9197
9198 xf86: bump input ABI version to 19
9199
9200 The changes to miPointerSetPosition interface from int->double breaks
9201 the SIS driver build, so time to bump this.
9202
9203 Signed-off-by: Dave Airlie <airlied@redhat.com>
9204 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9205
9206commit ad3bc571348a7007a2960bf87ae739397c5511ee
9207Author: Peter Hutterer <peter.hutterer@who-t.net>
9208Date: Tue Jan 8 11:19:09 2013 +1000
9209
9210 xfree86: update the device state for all DGA events (#59100)
9211
9212 DGA only handles master devices but it does intercept slave device events as
9213 well (since the event handlers are per event type, not per device).
9214
9215 The DGA code must thus call into UpdateDeviceState to reset the button/key
9216 state on the slave device before it discards the remainder of the event.
9217
9218 Test case:
9219 - Passive GrabModeSync on VCP
9220 - Press button
9221 - Enable DGA after ButtonPress
9222 - AllowEvents(SyncPointer)
9223 - Release button
9224
9225 The button release is handled by DGAProcessPointerEvent but the device state
9226 is never updated, so the slave ends up with the button permanently down.
9227 And since the master's button state is the union of the slave states, the
9228 master has the button permanently down.
9229
9230 X.Org Bug 59100 <http://bugs.freedesktop.org/show_bug.cgi?id=59100>
9231
9232 Reported-by: Steven Elliott <selliott4@austin.rr.com>
9233 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9234 Reviewed-by: Dave Airlie <airlied@redhat.com>
9235
9236commit c5f2818edbec2f87383baa6c6be5c389b73ca6f9
9237Author: Peter Hutterer <peter.hutterer@who-t.net>
9238Date: Tue Jan 8 10:13:53 2013 +1000
9239
9240 xfree86: set event->detail for DGA pointer events
9241
9242 Reported-by: Steven Elliott <selliott4@austin.rr.com>
9243 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9244 Reviewed-by: Dave Airlie <airlied@redhat.com>
9245
9246commit 519d183d78e0b0eaf47a473e94f5d8611baf8463
9247Author: Peter Hutterer <peter.hutterer@who-t.net>
9248Date: Mon Jan 7 10:44:33 2013 +1000
9249
9250 Fix two typos "requires an string value"
9251
9252 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9253
9254commit 4e13dd90144dde47550aceea4db4b4329e531279
9255Author: Peter Hutterer <peter.hutterer@who-t.net>
9256Date: Wed Aug 22 10:34:07 2012 +1000
9257
9258 dix: don't filter RawEvents if the grab window is not the root window (#53897)
9259
9260 If a XI2.1+ client has a grab on a non-root window, it must still receive
9261 raw events on the root window.
9262
9263 Test case: register for XI_ButtonPress on window and XI_RawMotion on root.
9264 No raw events are received once the press activates an implicit grab on the
9265 window.
9266
9267 X.Org Bug 53897 <http://bugs.freedesktop.org/show_bug.cgi?id=53897>
9268
9269 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9270 Reviewed-by: Keith Packard <keithp@keithp.com>
9271
9272commit a2037d7080ae64ea55f7d76971716346aa3ec6d3
9273Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9274Date: Wed Nov 28 22:25:35 2012 +0000
9275
9276 hw/xwin: Fix MinGW build of winSetAppModelID.c
9277
9278 Add missing #include <pthread.h>
9279
9280 In file included from /jhbuild/checkout/xorg/xserver/hw/xwin/winSetAppUserModelID.c:31:0:
9281 /jhbuild/checkout/xorg/xserver/hw/xwin/winwindow.h:140:11: error: expected declaration specifiers or ‘...’ before ‘pthread_t’
9282 /jhbuild/checkout/xorg/xserver/hw/xwin/winwindow.h:141:11: error: expected declaration specifiers or ‘...’ before ‘pthread_t’
9283 /jhbuild/checkout/xorg/xserver/hw/xwin/winwindow.h:142:11: error: expected declaration specifiers or ‘...’ before ‘pthread_mutex_t’
9284
9285 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9286 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9287
9288commit f57100bb36eae3b4d75f3c315973405f705b8de6
9289Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9290Date: Tue Feb 23 13:38:48 2010 +0000
9291
9292 hw/xwin: Process one Windows message per wakeup, rather than all of them.
9293
9294 De-queuing Windows messages and X events happens in the same thread of
9295 execution. Draining the windows message queue can lead to the X event queue
9296 overflowing if lots of those windows messages cause X events (e.g. if a keyboard
9297 macro program has just dumped thousands of keypresses into the Windows message
9298 queue). See the mailing list thread [1] for more details.
9299
9300 Processing one Windows message per wakeup, rather than all of them gives the X
9301 server a chance to do stuff as well after each message.
9302
9303 [1] http://cygwin.com/ml/cygwin-xfree/2010-01/msg00056.html
9304
9305 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9306 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9307
9308commit e30e1ea98720acc583f34c830a1c1b7e3e88f694
9309Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9310Date: Mon Dec 17 22:38:25 2012 +0000
9311
9312 hw/xwin: Fix some comments in winkeybd.c
9313
9314 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9315 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9316
9317commit 6f4a48f8a55bc54b6d3e9d80734be05750c024de
9318Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9319Date: Thu Apr 1 15:08:26 2010 +0100
9320
9321 hw/xwin: Bring the X screen window to the front on a single left-click on the tray icon
9322
9323 Bring the X screen window to the front on a single left click on the tray icon,
9324 like the comment says we do
9325
9326 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9327 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
9328
9329commit 6703a7c7cf1a349c137e247a0c8eb462ff7b07be
9330Author: Keith Packard <keithp@keithp.com>
9331Date: Tue Jan 8 20:24:32 2013 -0800
9332
9333 hw/xfree86: Require only one working CRTC to start the server.
9334
9335 Instead of requiring every mode set to complete successfully, start up
9336 as long as at least one CRTC is working. This avoids failures when one
9337 or more CRTCs can't start due to mode setting conflicts.
9338
9339 Signed-off-by: Keith Packard <keithp@keithp.com>
9340 Reviewed-by: Dave Airlie <airlied@redhat.com>
9341
9342commit 0e1ab433f4048b3367bb2f01d16cd00502538e4d
9343Author: Peter Hutterer <peter.hutterer@who-t.net>
9344Date: Fri Jan 4 12:26:58 2013 +1000
9345
9346 dix: remove already-moved hunk
9347
9348 Should've been removed in bc1f90a615018c05994fae3e678dd2341256cd82a, but got
9349 left here due to a botched rebase.
9350
9351 Fixes stray button events sent to clients after deactivating an async
9352 pointer grab on a pointer-emulating-touch.
9353
9354 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9355 Reviewed-by: Keith Packard <keithp@keithp.com>
9356
9357commit 32a6d8a6b59c42f8d65002d7ca1cafb1957b656f
9358Author: Peter Hutterer <peter.hutterer@who-t.net>
9359Date: Thu Dec 20 16:25:43 2012 +1000
9360
9361 dix: check for the right device's xi2 mask
9362
9363 events.c: In function 'DeactivatePointerGrab':
9364 events.c:1524:51: warning: 'dev' may be used uninitialized in this function
9365 [-Wuninitialized
9366
9367 dev is unset when we get here, the device to check is "mouse".
9368 Introduced in ece8157a59751b3ed492fb2e1eb8d5f20221e195.
9369
9370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9371 Reviewed-by: Keith Packard <keithp@keithp.com>
9372
9373commit f59499b5d05fde83813709e9848152951592120d
9374Author: Peter Hutterer <peter.hutterer@who-t.net>
9375Date: Tue Oct 30 12:44:08 2012 +1000
9376
9377 dix: add resource type to touch listeners
9378
9379 Instead of guessing what resource type the listener is and what property to
9380 retrieve, store the resource type in the listener directly.
9381
9382 Breaks XIT test cases:
9383 TouchGrabTestMultipleTaps.PassiveGrabPointerEmulationMultipleTouchesFastSuccession
9384
9385 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=56557
9386
9387 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9388 Reviewed-by: Keith Packard <keithp@keithp.com>
9389 Reviewed-by: Chase Douglas <chase.douglas@ubuntu.com>
9390
9391commit 9ad0fdb135a1c336771aee1f6eab75a6ad874aff
9392Author: Keith Packard <keithp@keithp.com>
9393Date: Tue Nov 27 11:21:17 2012 -0800
9394
9395 input: Record grab pointer in TouchListener
9396
9397 This places a pointer to the grab related to a TouchListener directly
9398 in the TouchListener structure rather than hoping to find the grab
9399 later on using the resource ID.
9400
9401 Passive grabs have resource ID in the resource DB so they can be
9402 removed when a client exits, and those resource IDs get copied when
9403 activated, but implicit grabs are constructed on-the-fly and have no
9404 resource DB entry.
9405
9406 Signed-off-by: Keith Packard <keithp@keithp.com>
9407 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9408 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9409
9410commit 91ab237358c6e33da854914d3de493a9cbea7637
9411Author: Keith Packard <keithp@keithp.com>
9412Date: Tue Nov 27 11:21:16 2012 -0800
9413
9414 input: Pull TouchListener declaration to top-level
9415
9416 No reason to have a struct declared inside another struct
9417
9418 Signed-off-by: Keith Packard <keithp@keithp.com>
9419 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9420 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9421
9422commit 3578cc3c2e1b5cb8eb191e2d12ad88e1bc9e6e1e
9423Author: Andreas Wettstein <wettstein509@solnet.ch>
9424Date: Wed Dec 19 18:13:21 2012 +0100
9425
9426 xkb: Do not use base group as an array index.
9427
9428 The base group is not brought into range and, therefore, using it as an array
9429 index crashed the X server. Also, at this place, we should ignore locked
9430 groups, but not latched groups. Therefore, use sum of base and latched groups,
9431 brought into range.
9432
9433 Reproducible with:
9434 key <FK07> {
9435 type= "ONE_LEVEL",
9436 symbols[Group1]= [ NoSymbol ],
9437 actions[Group1]= [ LatchGroup(group=-1, clearLocks) ]
9438 };
9439
9440 And hitting F7 will exceed the group level and access arbitrary memory.
9441
9442 Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
9443 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9444 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9445
9446commit df746a73410b892a4d41a2934cf9cd2e8ad7ba51
9447Author: Carlos Garnacho <carlosg@gnome.org>
9448Date: Wed Dec 19 18:42:39 2012 +0100
9449
9450 render: Unwrap early on the animated cursor BlockHandler
9451
9452 The loop above the previous call may end up triggering other
9453 handlers attaching to the same function slot, so unwrapping
9454 the handler after that could leave the just attached handler
9455 in a dangling but not unset state.
9456
9457 This issue was most visible on the XO, where destroying a
9458 window with an animated cursor set and running would trigger
9459 this inconsistent state, never calling the miSpriteBlockHandler
9460 again after the animated cursor is unset.
9461
9462 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
9463 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9464 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9465
9466commit 0fbd779a82919d5dbf8776be9b57a76c0eae6b14
9467Author: Carlos Garnacho <carlosg@gnome.org>
9468Date: Wed Dec 19 18:42:38 2012 +0100
9469
9470 mi: Ensure pointer emulating touch events update the sprite
9471
9472 Different miPointerSpriteFuncRec implementations do a varying
9473 business at ultimately calling miPointerUpdateSprite(), this
9474 particularly fails when using the plain mi sprite on touch events,
9475 where the sprite is just moved/updated on cursor changes.
9476
9477 So, ensure miPointerUpdateSprite() is called generically for
9478 pointer emulating touch events as with regular motion events.
9479
9480 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
9481 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9482 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9483
9484commit 8b328d4ee3873bc0a7a34f2cb9d301827244b98c
9485Author: Aaron Plattner <aplattner@nvidia.com>
9486Date: Fri Dec 21 07:37:33 2012 -0800
9487
9488 dix: Make small bitfields that store enums unsigned
9489
9490 Commit 31bf81772e146af79b0c456aae2159eba8b0280f changed the clientState field
9491 from a signed int to a signed int 2-bit bitfield. The ClientState enum that is
9492 expected to be assigned to this field has four values: ClientStateInitial (0),
9493 ClientStateRunning (1), ClientStateRetained (2), and ClientStateGone (3).
9494 However, because this bitfield is signed, ClientStateRetained becomes -2 when
9495 assigned, and ClientStateGone becomes -1. This causes warnings:
9496
9497 test.c:54:10: error: case label value exceeds maximum value for type [-Werror]
9498 test.c:55:10: error: case label value exceeds maximum value for type [-Werror]
9499
9500 The code here is a switch statement:
9501
9502 53 switch (client->clientState) {
9503 54 case ClientStateGone:
9504 55 case ClientStateRetained:
9505 56 [...]
9506 57 break;
9507 58
9508 59 default:
9509 60 [...]
9510 61 break;
9511 62 }
9512
9513 It also causes bizarre problems like this:
9514
9515 client->clientState = ClientStateGone;
9516 assert(client->clientState == ClientStateGone); // this assert fails
9517
9518 Also change the signedness of nearby bitfields to match.
9519
9520 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
9521 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
9522 Reviewed-by: Colin Harrison <colin.harrison at virgin.net>
9523 Signed-off-by: Keith Packard <keithp@keithp.com>
9524
9525commit 8f4820be7a2e0f6e286ddc85c4b75bccdbe8a730
9526Author: Peter Hutterer <peter.hutterer@who-t.net>
9527Date: Thu Dec 20 12:44:16 2012 +1000
9528
9529 test/xi2: fix compiler warning
9530
9531 protocol-xiwarppointer.c: In function ‘ScreenSetCursorPosition’:
9532 protocol-xiwarppointer.c:71:53: warning: declaration of ‘screen’ shadows a
9533 global declaration [-Wshadow]
9534
9535 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9536
9537commit bd91b05b631f13afd1f7a9d6cbc4f0c5408b523a
9538Author: Keith Packard <keithp@keithp.com>
9539Date: Wed Dec 19 12:47:35 2012 -0800
9540
9541 Update to version 1.13.99.901 (1.14 RC1)
9542
9543 Signed-off-by: Keith Packard <keithp@keithp.com>
9544
9545commit 2a0b544f5c09f79e4ab04ea8103a5ecf59ee2e7b
9546Merge: 0eb1559 ba4bb3b
9547Author: Keith Packard <keithp@keithp.com>
9548Date: Wed Dec 19 12:26:00 2012 -0800
9549
9550 Merge remote-tracking branch 'jeremyhu/master'
9551
9552commit 0eb1559eb29d11e63c2b33e317590a88d86fe313
9553Merge: 386e4d7 ea1d76d
9554Author: Keith Packard <keithp@keithp.com>
9555Date: Wed Dec 19 12:21:31 2012 -0800
9556
9557 Merge remote-tracking branch 'yselkowitz/master'
9558
9559 I checked this patch with diff -w to check that it only affected
9560 whitespace.
9561
9562commit 386e4d76baaffe226d2d561ff936509454eb0ac2
9563Merge: 014a5c8 9ff2e83
9564Author: Keith Packard <keithp@keithp.com>
9565Date: Wed Dec 19 12:17:59 2012 -0800
9566
9567 Merge remote-tracking branch 'alanc/master'
9568
9569commit 014a5c8a9d86f2f992183bff9106354fac2c3b0e
9570Merge: f793b5f 2eefa5d
9571Author: Keith Packard <keithp@keithp.com>
9572Date: Wed Dec 19 12:09:31 2012 -0800
9573
9574 Merge remote-tracking branch 'whot/barriers'
9575
9576 Conflicts:
9577 Xi/xichangehierarchy.c
9578
9579 Small conflict with the patch from
9580
9581 Xi: don't use devices after removing them
9582
9583 Was easily resolved by hand.
9584
9585 Signed-off-by: Keith Packard <keithp@keithp.com>
9586
9587commit 9ff2e831517875f96477862f979abff394e8d551
9588Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9589Date: Tue Dec 18 00:41:08 2012 -0800
9590
9591 EnableDisableExtensionError: Use ARRAY_SIZE rather than sentinel
9592
9593 d785368e0e converted the other miinitext functions to use ARRAY_SIZE,
9594 and removed the sentinel, but missed EnableDisableExtensionError so
9595 passing an invalid extension name could cause the server to walk off
9596 the end off the list looking for a sentinel that wasn't there.
9597
9598 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9599 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9600
9601commit ba4bb3bc1b87eb57cc34d2ad1f302c9d2a15c847
9602Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
9603Date: Tue Dec 18 01:29:12 2012 -0800
9604
9605 XQuartz: Don't add the 15bit visual any more
9606
9607 Mountain Lion only supports 32bit backing stores, so don't use 15bit visuals until libXplugin adapts
9608
9609 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
9610
9611commit c298f9c42ec05486ac88fc91a84739d7c3e3ea9e
9612Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
9613Date: Tue Dec 18 01:28:02 2012 -0800
9614
9615 XQuartz: Revert some unfortunate auto-indenting mishaps with our super-nested for-loops
9616
9617 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
9618
9619commit 07a91fa6c6d535f3f05d4c3bd9c4d2b8c382c475
9620Author: Rob Clark <robdclark@gmail.com>
9621Date: Wed Dec 5 23:21:27 2012 +0000
9622
9623 hw/dmx: fix build without GLX
9624
9625 Fixes this compile break that showed up on arm recently:
9626
9627 dmxinit.c:746:26: error: 'glxSupported' undeclared (first use in this function)
9628 dmxinit.c:746:26: note: each undeclared identifier is reported only once for each function it appears in
9629
9630 Signed-off-by: Rob Clark <robdclark@gmail.com>
9631 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
9632
9633commit 5692a1e8f5befd3698134b1a5516a4dadda00115
9634Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9635Date: Thu Dec 6 16:15:50 2012 -0800
9636
9637 Support compilers with alternate spellings of typeof
9638
9639 The AC_C_TYPEOF adds a #undef typeof to its autogenerated config.h.in
9640 template, but b8ab93dfbc7f292 didn't copy that to dix-config.h.in
9641 when HAVE_TYPEOF was, so the macro could claim typeof support but not
9642 make it work, when used with compilers like Solaris Studio 12.1 which
9643 only recognize it as __typeof__.
9644
9645 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9646 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
9647
9648commit f793b5fd3eb16a2ada130367c2ffebeede69a322
9649Author: Peter Hutterer <peter.hutterer@who-t.net>
9650Date: Tue Dec 11 13:36:02 2012 +1000
9651
9652 dix: don't copy the wrong event mask when activating a passive grab
9653
9654 GrabMask is a union of core, XI1 and XI2 masks. If a XI2 grab is activated,
9655 the value is a random pointer value, using it as mask has unpredictable
9656 effects.
9657
9658 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9659 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9660 Reviewed-by: Keith Packard <keithp@keithp.com>
9661
9662commit dd3242c87a0a58cba055eb99c0c3fcf03153e4b8
9663Author: Peter Hutterer <peter.hutterer@who-t.net>
9664Date: Fri Dec 14 11:34:15 2012 +1000
9665
9666 dix: don't allow overriding a grab with a different type of grab (#58255)
9667
9668 If a client has a core grab, don't allow re-grabbing with type XI2, etc.
9669 This was the intent of the original commit
9670 xorg-server-1.5.99.1-782-g09f9a86, but ineffective.
9671
9672 X.Org Bug 58255 <http://bugs.freedesktop.org/show_bug.cgi?id=58255>
9673
9674 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9675 Reviewed-by: Keith Packard <keithp@keithp.com>
9676
9677commit 2eefa5d6e870c57ac6a5930883d8cfe3a3882a43
9678Author: Peter Hutterer <peter.hutterer@who-t.net>
9679Date: Thu Dec 13 14:42:32 2012 +1000
9680
9681 Xi: if a MD is removed, send a barrier leave event (if applicable)
9682
9683 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9684 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9685
9686commit e2423b627ef7e4d08ea0685af5e771ddd2b511ce
9687Author: Peter Hutterer <peter.hutterer@who-t.net>
9688Date: Thu Dec 13 14:04:13 2012 +1000
9689
9690 Xi: don't store the window pointer in barriers, store the window ID
9691
9692 When a client shuts down and resources are being freed, the window may have
9693 been freed already, so accessing it to get the window ID is bad. Plus, we
9694 never care about the window anyway other than for stuffing it into the
9695 event.
9696
9697 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9698 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9699
9700commit f71c2f895c9e2f9d0d42feaac2a3e1d2deb71f67
9701Author: Peter Hutterer <peter.hutterer@who-t.net>
9702Date: Wed Dec 12 15:41:32 2012 +1000
9703
9704 Xi: fix per-device barrier handling
9705
9706 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9707 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9708
9709commit 58bff17e43a80eb21b3ff6d4bb1596230e61f707
9710Author: Peter Hutterer <peter.hutterer@who-t.net>
9711Date: Thu Dec 13 11:49:54 2012 +1000
9712
9713 Xi: don't use devices after removing them
9714
9715 RemoveDevice() frees the DeviceIntPtr, we shouldn't use the pointer after
9716 that
9717
9718 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9719 Reviewed-by: Keith Packard <keithp@keithp.com>
9720
9721commit 3420a7778c7d5eaa638327f31dd460554c257bb1
9722Author: Peter Hutterer <peter.hutterer@who-t.net>
9723Date: Wed Dec 12 11:02:19 2012 +1000
9724
9725 xfree86: print message to the log when zapping the server
9726
9727 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9728 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
9729 Signed-off-by: Keith Packard <keithp@keithp.com>
9730
9731commit 6d508b81857edaed03c7ee06410434ea56d9b701
9732Merge: d982d87 bb6f351
9733Author: Keith Packard <keithp@keithp.com>
9734Date: Mon Dec 17 11:45:45 2012 -0800
9735
9736 Merge remote-tracking branch 'whot/for-keith'
9737
9738 I've looked at these patches, but I can't say I've actually
9739 reviewed them...
9740
9741commit d982d877436377597b8ed04bca1438d2edaf53ee
9742Merge: da3eaf6 6b4aa8a
9743Author: Keith Packard <keithp@keithp.com>
9744Date: Mon Dec 17 10:56:12 2012 -0800
9745
9746 Merge remote-tracking branch 'airlied/for-keithp'
9747
9748commit da3eaf6bdbd7ce3bebf2c490cbe2448b4c402dba
9749Author: Dave Airlie <airlied@redhat.com>
9750Date: Mon Dec 17 15:40:17 2012 +1000
9751
9752 glx/dri2: initialise api to avoid indirect rendering failing randomly
9753
9754 Running glxinfo under indirect rendering would randomly fail against the
9755 intel driver, as it would create a context with no attribs, and then the
9756 api value would be passed to the driver uninitialised.
9757
9758 Signed-off-by: Dave Airlie <airlied@redhat.com>
9759 Reviewed-by: Keith Packard <keithp@keithp.com>
9760 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
9761 Signed-off-by: Keith Packard <keithp@keithp.com>
9762
9763commit cc10ac8f0e07854647e1fd7cc70b7e9f8d919fd1
9764Author: Peter Hutterer <peter.hutterer@who-t.net>
9765Date: Mon Dec 10 11:00:46 2012 +1000
9766
9767 Xi: fix swapping for barrier events
9768
9769 Protocol events don't contain pointers, so it's easier to copy everything
9770 over, then swap in-place.
9771
9772 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9773 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9774
9775commit 151d44149a09dd125e25b3d94f22a609f0221548
9776Author: Peter Hutterer <peter.hutterer@who-t.net>
9777Date: Mon Dec 10 10:53:49 2012 +1000
9778
9779 Xi: swap sequence number and evtype in barrier events
9780
9781 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9782 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9783
9784commit d4065493b211e90b46f9d4178f23f347afd8043f
9785Author: Peter Hutterer <peter.hutterer@who-t.net>
9786Date: Mon Dec 10 11:02:32 2012 +1000
9787
9788 tests/xi2: at protocol conversion test for barrier events
9789
9790 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9791 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9792
9793commit 938187f2fec006daf4cc677df26d5b0b6999b54b
9794Author: Peter Hutterer <peter.hutterer@who-t.net>
9795Date: Fri Dec 7 15:55:59 2012 +1000
9796
9797 Require inputproto 2.2.99.1
9798
9799 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9800 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9801
9802commit 78376a9217058aa66e52a0399fc2be469d70ce32
9803Author: Peter Hutterer <peter.hutterer@who-t.net>
9804Date: Fri Dec 7 09:26:13 2012 +1000
9805
9806 mi: rename mipointer's internal event list
9807
9808 Avoid name shadowing warnings, change the event list to a more specific
9809 name.
9810
9811 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9812 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9813
9814commit 88a2cccc37cac4e171f9bfc020017ddfa66ae417
9815Author: Peter Hutterer <peter.hutterer@who-t.net>
9816Date: Fri Dec 7 09:11:13 2012 +1000
9817
9818 Xi: if the device is currently grabbed, flag the barrier event
9819
9820 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9821 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9822
9823commit 353aa515922e1095047161ec47a2722772218f20
9824Author: Peter Hutterer <peter.hutterer@who-t.net>
9825Date: Thu Dec 6 15:09:27 2012 +1000
9826
9827 Xi: deliver barrier events as grabbed events where necessary
9828
9829 If the grab_window is the barrier window and the client owns the grab,
9830 deliver as normal grabbed event (respecting owner_events). Otherwise,
9831 deliver as usual.
9832
9833 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9834 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9835
9836commit 3b161401700a2d916da0f81f99b39e75fdbe78df
9837Author: Peter Hutterer <peter.hutterer@who-t.net>
9838Date: Thu Dec 6 14:13:26 2012 +1000
9839
9840 dix: ignore barrier events in FixUpEventFromWindow
9841
9842 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9843 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9844
9845commit a1eeb6fbecf0bd9665a9158b27c273e83b3ab351
9846Author: Peter Hutterer <peter.hutterer@who-t.net>
9847Date: Thu Dec 6 13:25:29 2012 +1000
9848
9849 dix: handle barrier events properly when converting to core/XI 1.x
9850
9851 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9852 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9853
9854commit 1b83775f6712a7e48c49a3411a87103e8e906e67
9855Author: Peter Hutterer <peter.hutterer@who-t.net>
9856Date: Thu Dec 6 12:39:04 2012 +1000
9857
9858 dix: skip delivery if it's not the right pointer barrier client
9859
9860 Only deliver to the client that created the barrier, not to other clients.
9861
9862 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9863 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9864
9865commit c50db6faba4ee1c27b735c6e9c98a4276ba3c7ff
9866Author: Peter Hutterer <peter.hutterer@who-t.net>
9867Date: Thu Dec 6 12:58:55 2012 +1000
9868
9869 Xi: fill in barrier root x/y after clamping to RandR outputs
9870
9871 x/y for barrier events should contain the actual pointer position.
9872
9873 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9874 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9875
9876commit 21a15f9a04ec0a6c8f654eef943561e98db2475d
9877Author: Peter Hutterer <peter.hutterer@who-t.net>
9878Date: Mon Dec 3 10:41:16 2012 +1000
9879
9880 Pass the event list through to the pointer barrier code to return it
9881
9882 Instead of having the pointer barrier code enqueue events separately from
9883 GetPointerEvents, pass the event list through and let it add to it.
9884
9885 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9886 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9887
9888commit 707b4dc61f18960611409ef5ad8947be189f7296
9889Author: Jasper St. Pierre <jstpierre@mecheye.net>
9890Date: Tue Nov 27 15:31:52 2012 -0500
9891
9892 barriers: Support line and ray barriers
9893
9894 This allows clients to add barriers that extend to the edge of the
9895 screen. Clients are encouraged to use these instead of precise coordinates
9896 in these cases to help prevent pointer leaks.
9897
9898 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
9899 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9900
9901commit 0a5a0d7c242add4d79721c1b90073b603a475534
9902Author: Jasper St. Pierre <jstpierre@mecheye.net>
9903Date: Tue Nov 27 14:29:52 2012 -0500
9904
9905 barriers: Replace complex intersection test with simpler math
9906
9907 Since barriers are axis-aligned, we can do the intersection test with
9908 simple interpolation rather than line-segment intersection. This also
9909 helps us out in the future when we want the barriers to extend to be
9910 rays and lines rather than just segments.
9911
9912 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
9913 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9914
9915commit 6401317bdcf58b2717214ac65c8f47b0c384a7db
9916Author: Jasper St. Pierre <jstpierre@mecheye.net>
9917Date: Sun Dec 9 19:45:48 2012 -0500
9918
9919 barriers: Send an XI_BarrierLeave event when a barrier is destroyed
9920
9921 This ensures that we always complete an event sequence.
9922
9923 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
9924 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9925
9926commit 207e8dee00dd26ff4363edb72058b48b8a76b824
9927Author: Jasper St. Pierre <jstpierre@mecheye.net>
9928Date: Sun Dec 9 19:44:13 2012 -0500
9929
9930 barriers: Clean up code
9931
9932 Rename a variable. This is to make the diff in the next commit cleaner.
9933
9934 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
9935 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9936
9937commit 7fabecd8e3efe76caeb740d905a02ae4256e3246
9938Author: Jasper St. Pierre <jstpierre@mecheye.net>
9939Date: Wed Dec 5 19:58:05 2012 -0500
9940
9941 barriers: Send a BarrierLeave event when we leave the hitbox
9942
9943 Additionally, add flags when the pointer is released.
9944
9945 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9946 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9947
9948commit dac9e13a6c874f3b99ba4cf9085ed0e63beb0871
9949Author: Jasper St. Pierre <jstpierre@mecheye.net>
9950Date: Mon Nov 26 20:37:56 2012 -0500
9951
9952 barriers: Increment event ID on hit box leave
9953
9954 We eventually want to send a new notify event on hitbox leave,
9955 which signifies the dawn of a new barrier event ID, so it's
9956 convenient if we can put the code here.
9957
9958 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
9959 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9960
9961commit e3a734d081ebda4460e6c1eb7dcdd0f7c10ba9b4
9962Author: Jasper St. Pierre <jstpierre@mecheye.net>
9963Date: Sat Nov 24 21:42:56 2012 -0500
9964
9965 barriers: Add a couple pixels of elbow room for the hit detection
9966
9967 Pointers (and the hands that drive them) aren't very precise, and the
9968 slightest amount of nudging to either side might be enough to reset
9969 the event ID, making clients think they have an entirely new hit. Allow
9970 for a slightly bigger "hit box" before these barriers get reset.
9971
9972 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
9973 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9974
9975commit 3b2fbcfa6c75ab072926ab9659adf785b324ce28
9976Author: Jasper St. Pierre <jstpierre@mecheye.net>
9977Date: Thu Nov 8 01:17:40 2012 -0500
9978
9979 barriers: Add support for edge cases when releasing barriers
9980
9981 If we release a barrier, we want to ensure that we block all
9982 other barriers afterwards, rather than capping the limit to
9983 the two nearest barriers.
9984
9985 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
9986 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9987
9988commit a1ee26e624e6a7030ecb37a608190cb8e3d1f65b
9989Author: Peter Hutterer <peter.hutterer@who-t.net>
9990Date: Thu Dec 6 12:06:28 2012 +1000
9991
9992 barriers: Don't allow releasing the pointer on other client's barriers
9993
9994 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9995 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
9996
9997commit 2884b249294e3e8fc749bbb81e2a897314fa0818
9998Author: Peter Hutterer <peter.hutterer@who-t.net>
9999Date: Thu Dec 6 11:59:53 2012 +1000
10000
10001 barriers: Don't allow destroying other client's barriers
10002
10003 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10004 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
10005
10006commit e130a46ab45839f91b29ee75bf19a597dfe3da28
10007Author: Jasper St. Pierre <jstpierre@mecheye.net>
10008Date: Thu Nov 1 16:06:40 2012 -0400
10009
10010 Add support for XI2.3: Pointer barrier events and releases.
10011
10012 This adds support for clients that would like to get a notification
10013 every time a barrier is hit, and allows clients to temporarily release
10014 a barrier so that pointers can go through them, without having to
10015 destroy and recreate barriers.
10016
10017 Based on work by Chris Halse Rogers <chris.halse.rogers@canonical.com>
10018
10019 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
10020 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10021 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10022
10023commit 85a37ddcc2ff49adaec4075532b60e7d015f901c
10024Author: Jasper St. Pierre <jstpierre@mecheye.net>
10025Date: Sun Dec 9 22:48:18 2012 -0500
10026
10027 barriers: Reindent the constrainment hook
10028
10029 This is to make future diffs much cleaner. Best viewed with -w.
10030
10031 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
10032 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10033
10034commit 97da74c80e8b8eb03f7c1e34180cf5cd343bd608
10035Author: Jasper St. Pierre <jstpierre@mecheye.net>
10036Date: Sun Dec 9 22:19:14 2012 -0500
10037
10038 barriers: Switch to finding the nearest barrier client
10039
10040 When we add events, we eventually want to add more state to the
10041 PointerBarrierClient, so return one of these instead of the dummy
10042 public structure that's not very interesting.
10043
10044 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
10045 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10046
10047commit 7e16dd3628334a5991b6713d778a46c1ce3b8b78
10048Author: Jasper St. Pierre <jstpierre@mecheye.net>
10049Date: Sat Nov 24 19:39:47 2012 -0500
10050
10051 barriers: Switch to an explicit hook for barrier constrainment
10052
10053 Rather than riding on the ConstrainCursorHarder hook, which has
10054 several issues, move to an explicit hook, which will help us with
10055 some RANDR interaction issues.
10056
10057 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
10058 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10059
10060commit 2868a93945f043e1efd897d56543fe673e341faf
10061Author: Jasper St. Pierre <jstpierre@mecheye.net>
10062Date: Sun Dec 9 18:41:52 2012 -0500
10063
10064 barriers: Don't loop over the server to destroy a barrier
10065
10066 This is completely pointless as far as I can tell.
10067
10068 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
10069 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10070
10071commit 482e0cb35225dc3b0664f9429db19c61b0c98ef6
10072Author: Jasper St. Pierre <jstpierre@mecheye.net>
10073Date: Thu Nov 1 14:14:13 2012 -0400
10074
10075 cursor: Move pointer barrier code over to XI
10076
10077 In order to send events to specific windows associated with the barrier,
10078 we need to move the code that handles barriers to somewhere where it's
10079 easier to construct and send events. Rather than duplicating XSync with
10080 its XSyncSelectAlarm, re-use the existing XI infrastructure.
10081
10082 For now, just move a bunch of code over, rename some things, and initialize
10083 the new structures, but still consider it a separate codebase. Pointer barrier
10084 requests are still handled by XFixes, so this is a weird intermediate state.
10085 It's unknown whether we'll add explicit requests to pointer barriers inside
10086 XI.
10087
10088 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
10089 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10090
10091commit bb6f3514ca17d993c1af380e8d4480d61e5bbcae
10092Merge: f961c3a 08da994
10093Author: Peter Hutterer <peter.hutterer@who-t.net>
10094Date: Wed Dec 12 17:25:28 2012 +1000
10095
10096 Merge branch 'stack-smash-on-touchpoint' into for-keith
10097
10098commit f961c3a3b9dfbe1201da317c24797ba7f979731e
10099Merge: 36740d0 39f19b3
10100Author: Peter Hutterer <peter.hutterer@who-t.net>
10101Date: Wed Dec 12 17:25:23 2012 +1000
10102
10103 Merge branch 'touch-selection-conflict-fixes' into for-keith
10104
10105commit 39f19b3f3b8c9b714e70e339dfb0083ff629ab2a
10106Author: Peter Hutterer <peter.hutterer@who-t.net>
10107Date: Tue Nov 20 11:48:31 2012 +1000
10108
10109 Xi: fix touch event selction conflicts (#57301)
10110
10111 There are limits on which client may select for touch events on a given
10112 window, with restrictions being that no two clients can select on the same
10113 device, but narrower selections are allowed, i.e. if one client has
10114 XIAllDevices, a second client may still select for device X.
10115
10116 The current code had a dependency on which client selected first and which
10117 device, resulting in inconsistencies when selecting for events. Fix that,
10118 responding with the right errors regardless of who selected what first.
10119
10120 X.Org Bug 57301 <http://bugs.freedesktop.org/show_bug.cgi?id=57301>
10121
10122 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10123
10124commit a7c97d737ef0d14ec97869082decd049637cfe7a
10125Author: Peter Hutterer <peter.hutterer@who-t.net>
10126Date: Tue Nov 20 11:38:39 2012 +1000
10127
10128 dix: split xi2_mask_isset into a per-device function
10129
10130 For touch selection conflicts, we need to check not only if the mask is set
10131 for the device, but if it is set for only that specific device (regardless
10132 of XIAll*Devices)
10133
10134 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10135
10136commit 08da994a08bb74afae81176c56fb525d0439274b
10137Author: Peter Hutterer <peter.hutterer@who-t.net>
10138Date: Mon Nov 26 12:33:29 2012 +1000
10139
10140 dix: add FIXME, TouchRemovePointerGrab does nothing
10141
10142 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10143
10144commit 00def5144557cfe8bf535f926212a8e084dc7cf6
10145Author: Peter Hutterer <peter.hutterer@who-t.net>
10146Date: Mon Nov 26 14:55:13 2012 +1000
10147
10148 Xi: if a TouchEnd appears on a actively grabbing client, always accept
10149
10150 Once the TouchEnd appears on the device, the touch is done. If the client
10151 still has a pointer grab, accept it to avoid clients with TouchOwnership
10152 selections to wait indefinitely for the actual touch event.
10153
10154 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10155
10156commit ece8157a59751b3ed492fb2e1eb8d5f20221e195
10157Author: Peter Hutterer <peter.hutterer@who-t.net>
10158Date: Mon Nov 26 15:14:19 2012 +1000
10159
10160 dix: when deactivating pointer-only grabs, don't emulate TouchEnd events
10161
10162 A client with a pointer grab on a touch device must reject the touch when
10163 detactivating the grab while the touch is active. However, such a rejecting
10164 must not trigger a ButtonRelease event to be emulated and sent to the
10165 client.
10166 Set the grabbing listener's state to HAS_END, so we simply skip delivery to
10167 that client.
10168
10169 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10170
10171commit bc1f90a615018c05994fae3e678dd2341256cd82
10172Author: Peter Hutterer <peter.hutterer@who-t.net>
10173Date: Mon Nov 26 12:23:54 2012 +1000
10174
10175 dix: only reject active grabs on ungrab and do it before actually ungrabbing
10176
10177 An active grab ungrabbing is the same as rejecting the grab, since the
10178 client is no longer interested in those events. So reject any touch grab,
10179 but do so before actually deactivating since we're interested in the
10180 TouchEnd for the current grabbing client.
10181
10182 A passive grab otoh is _not_ like rejecting a grab, since it deactivates
10183 automatically when the touch ends.
10184
10185 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10186
10187commit 146f48c2934fc85ec095496da5c8f0102bc7f5b5
10188Author: Peter Hutterer <peter.hutterer@who-t.net>
10189Date: Thu Nov 22 13:49:34 2012 +1000
10190
10191 dix: don't call ProcessInputEvents() when accepting/rejecting touches
10192
10193 TouchListenerAcceptReject may be called during normal event processing, but
10194 ProcessInputEvents is not reentrant and calling it here smashes the event
10195 queue.
10196
10197 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10198
10199commit ead21f9426122536adfb4787ac181008ae83cd4b
10200Author: Peter Hutterer <peter.hutterer@who-t.net>
10201Date: Mon Nov 19 16:16:10 2012 +1000
10202
10203 Xi: fix typo "mechansims" → "mechanisms"
10204
10205 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10206
10207commit 6b4aa8a359822cc70666de3b18744f05dd9b5f45
10208Author: Dave Airlie <airlied@redhat.com>
10209Date: Wed Dec 12 12:56:46 2012 +1000
10210
10211 Revert "xf86: Fix non-PCI configuration-less setups"
10212
10213 This reverts commit 76d9c62eb2be2010a19bf36285012d086cdd180b.
10214
10215 This breaks multi-GPU setups here, so lets drop it for now.
10216
10217 Signed-off-by: Dave Airlie <airlied@redhat.com>
10218
10219commit 785af88ab0120036e0ce3d0139f3c560ff71e10b
10220Author: Dave Airlie <airlied@redhat.com>
10221Date: Wed Sep 26 16:16:40 2012 +1000
10222
10223 dri1: fix dri1 startup since 459c6da0f907ba33d733c7e62a116184ba2f14e5
10224
10225 This commit regresses dri1 since it moves the drmSetServerInfo from being
10226 called at module load time to extension init time. However DRIScreenInit
10227 relies on this being called before it gets control.
10228
10229 This patches moves the call into DRIScreenInit and seems to work here.
10230
10231 Signed-off-by: Dave Airlie <airlied@redhat.com>
10232
10233commit 36740d02b9ca117f1404e077367fbbbe271a17d6
10234Author: Daniel Martin <consume.noise@gmail.com>
10235Date: Fri Dec 7 19:38:55 2012 +0100
10236
10237 xfixes: Fix minor number in QueryVersion
10238
10239 Due to a typo the major version number was passed as minor version to
10240 version_compare().
10241
10242 Regression-from: ffd4874798ba54f86acac75779a15b4babeaa5f3
10243
10244 Signed-off-by: Daniel Martin <consume.noise@gmail.com>
10245 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10246 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10247
10248commit b34cffd27b9ef46e1dd477219cbb90e36d5609e3
10249Author: Peter Hutterer <peter.hutterer@who-t.net>
10250Date: Mon Dec 3 11:39:08 2012 +1000
10251
10252 include: fix comment
10253
10254 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10255
10256commit ea1d76d1b68c7cbcedcaa1f967c5027e0cc1c7a2
10257Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10258Date: Mon Nov 5 23:50:58 2012 -0600
10259
10260 Fix formatting of address operators
10261
10262 The formatter confused address operators preceded by casts with
10263 bitwise-and expressions, placing spaces on either side of both.
10264 That syntax isn't used by ordinary address operators, however,
10265 so fix them for consistency.
10266
10267 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10268 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10269
10270commit 5daa442fe15d9a87112a2def673c99a5f2506dcf
10271Author: Peter Hutterer <peter.hutterer@who-t.net>
10272Date: Tue Dec 4 15:28:14 2012 +1000
10273
10274 xkb: only post a XTest release if the XTest device has the button down
10275
10276 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10277
10278commit c4fee9d2ecd6fdd4c05b04c9f02b23f10f2938b8
10279Author: Peter Hutterer <peter.hutterer@who-t.net>
10280Date: Tue Dec 4 15:03:37 2012 +1000
10281
10282 xkb: always post XTest button up when the physical button released (#28808)
10283
10284 Regression introduced by commit 2decff6393a44b56d80d53570718f95354fde454
10285 xkb: ProcesssPointerEvent must work on the VCP if it gets the VCP
10286
10287 XTest buttons must be released when a physical button is released. This was
10288 fixed in 14327858391ebe929b806efb53ad79e789361883, but
10289 2decff6393a44b56d80d53570718f95354fde454 changed a condition that this code
10290 didn't get triggered anymore.
10291
10292 "dev" for pointer events is now always the VCP which doesn't have a xkbi
10293 struct. So move this condition out and always trigger the XTest released for
10294 button events.
10295
10296 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10297 Tested-by: Frank Roscher <Frank-Roscher@gmx.net>
10298
10299commit b51a1bd2766e7dc975ca8f1cacc3f8bd0e1a68a3
10300Merge: e9ea96d ce6b652
10301Author: Keith Packard <keithp@keithp.com>
10302Date: Fri Nov 30 09:31:46 2012 -0800
10303
10304 Merge remote-tracking branch 'whot/for-keith'
10305
10306commit e9ea96d6a88f197928c82b008953f1810de6f792
10307Author: Dave Airlie <airlied@gmail.com>
10308Date: Fri Nov 30 20:54:38 2012 +1000
10309
10310 randr: call RRProviderInit in the proper place.
10311
10312 No idea where this got lost across development cycles, but its
10313 definitely missing.
10314
10315 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=57448
10316
10317 Signed-off-by: Dave Airlie <airlied@redhat.com>
10318 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
10319 Signed-off-by: Keith Packard <keithp@keithp.com>
10320
10321commit ce6b6529290f7d42717aad3fd8a83aaaaf8962de
10322Merge: 6e18599 3b9f1c7
10323Author: Peter Hutterer <peter.hutterer@who-t.net>
10324Date: Thu Nov 29 14:49:22 2012 +1000
10325
10326 Merge branch 'high-resolution-touch-devices' into for-keith
10327
10328commit 6e18599d691036eca6ff082c17727d9ffb926984
10329Author: Sybren van Elderen <sowmestno@msn.com>
10330Date: Tue Nov 27 19:27:10 2012 +0100
10331
10332 hw/dmx: add update_desktop_dimensions() call
10333
10334 The Xdmx server did not update the desktop dimensions when computing screen
10335 origins.
10336
10337 Signed-off-by: Sybren van Elderen <sowmestno@msn.com>
10338 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10339 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10340
10341commit 59d70b30e93f8be93db9e873423353f7e4419811
10342Author: Peter Hutterer <peter.hutterer@who-t.net>
10343Date: Fri Nov 23 10:05:02 2012 +1000
10344
10345 dix: use pixman for fp1616 conversions
10346
10347 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10348 Reviewed-by: Søren Sandmann <ssp@redhat.com>>
10349
10350commit 2c4388a00ec308bc2d48ba751ff510cd5c1b2384
10351Author: Peter Hutterer <peter.hutterer@who-t.net>
10352Date: Tue Jun 26 10:30:35 2012 +1000
10353
10354 xkb: fill in keycode and event type for slow keys enablement
10355
10356 eventType is set for the type that triggered a XkbControlsNotify event.
10357 Technically, SlowKeys is triggered by a timer which doesn't have a matching
10358 core event type. So we used to use 0 here.
10359
10360 Practically, the timer is triggered by a key press + hold and cancelled when
10361 the key is released before the timeout expires. So we might as well set
10362 KeyPress (keycode) in the ControlsNotify to give clients a chance to differ
10363 between timer-triggered SlowKeys and client-triggered ones.
10364
10365 This is a chance in behaviour, though I suspect with little impact.
10366
10367 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10368 Acked-by: Daniel Stone <daniel@fooishbar.org>
10369
10370commit 2dc6d9228456cbc274a1fcea478258e5ae4dc629
10371Author: Peter Hutterer <peter.hutterer@who-t.net>
10372Date: Thu Nov 15 14:00:57 2012 +1000
10373
10374 When resetting device idle time, reset XIAll(Master)Devices too (#56649)
10375
10376 When the screen saver is forcibly deactivated, the idle time counter is
10377 reset for all devices but not for the fake XIAllDevices and
10378 XIAllMasterDevices. XScreenSaverQueryInfo uses XIAlldevices to fill the
10379 "idle" field, thus returning the wrong value.
10380
10381 Regression introduced in
10382 commit 6aef209ebc2e54f5465da505a780f7b4cc273ee0
10383 Author: Peter Hutterer <peter.hutterer@who-t.net>
10384 Date: Mon Mar 12 13:51:02 2012 +1000
10385
10386 Change lastDeviceIdleTime to be per-device
10387
10388 X.Org Bug 56649 <http://bugs.freedesktop.org/show_bug.cgi?id=56649>
10389
10390 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10391 Tested-by: Giacomo Perale <ghepeu@virgilio.it>
10392 Reviewed-by: Keith Packard <keithp@keithp.com>
10393
10394commit a51b2c3913fc8556f6bd1c76805d045fc424c4bb
10395Author: Jasper St. Pierre <jstpierre@mecheye.net>
10396Date: Tue Nov 20 14:50:41 2012 -0500
10397
10398 cursor: Revise edge cases for the pointer moving towards barriers
10399
10400 Since barriers block the invisible line between pixels, that means
10401 that we need to explicitly check the boundaries, or else we'll have
10402 a potential off-by-one error. This fixes issues when trying to move
10403 down or right across a barrier and having the pointer visibly bounce.
10404
10405 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
10406 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10407 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10408
10409commit 1712a45422a63f11b2146541279616fcfda09ec6
10410Merge: e54f71a 338bec3
10411Author: Keith Packard <keithp@keithp.com>
10412Date: Wed Nov 28 07:54:13 2012 -0800
10413
10414 Merge remote-tracking branch 'jturney/mingw-patches'
10415
10416commit 338bec30054bb5d41cb6fdb5d37439f3c8124cad
10417Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10418Date: Fri Nov 9 12:19:31 2012 -0600
10419
10420 hw/xwin: Fix for MinGW-w64 DirectDraw headers
10421
10422 mingw-w64 headers handle NONAMELESSUNION earlier than mingw.org's, so it must be
10423 defined before including any headers. It also provides a ddraw.h, so use it.
10424
10425 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10426 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10427 Reviewed-by: Keith Packard <keithp@keithp.com>
10428
10429commit fb170498abc746a850864b59db2ddcba7ee29215
10430Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10431Date: Tue Nov 13 13:22:44 2012 +0000
10432
10433 dix/dispatch.c, os/utils.c: Disable smart scheduler on WIN32
10434
10435 setitimer() and SIGALRM aren't available on WIN32, so smart scheduler
10436 code cannot be built. Provide only stubs for smart scheduler timer
10437 code, and disable smart scheduler by default.
10438
10439 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
10440 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10441 Tested-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10442 Reviewed-by: Keith Packard <keithp@keithp.com>
10443
10444commit b20d2998cd9a1320ef0e45ab6b741c7e25f646e0
10445Author: Ryan Pavlik <rpavlik@iastate.edu>
10446Date: Thu Nov 3 19:25:57 2011 -0500
10447
10448 os/osinit.c: no getpgrp() and setpgrp() on WIN32
10449
10450 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
10451 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10452 Tested-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10453 Reviewed-by: Keith Packard <keithp@keithp.com>
10454
10455commit 2c712d094a49de0b96e2e9f5bcc1043b54ad0f61
10456Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10457Date: Sun Oct 14 23:05:30 2012 -0500
10458
10459 hw/xwin: use raise() instead of kill()
10460
10461 MinGW doesn't have kill(), so use raise() instead
10462
10463 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10464 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10465 Reviewed-by: Keith Packard <keithp@keithp.com>
10466
10467commit 98d5acc1213b2ec71c2a7bd3a3b77a1b5536e963
10468Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10469Date: Fri Aug 17 13:43:29 2012 +0100
10470
10471 os/utils.c: Fix compilation of OsBlockSIGIO when SIGIO isn't defined
10472
10473 Fix compilation of OsBlockSIGIO with -Werror=return-type when SIGIO isn't
10474 defined.
10475
10476 /jhbuild/checkout/xorg/xserver/os/utils.c: In function 'OsBlockSIGIO':
10477 /jhbuild/checkout/xorg/xserver/os/utils.c:1248:1: error: control reaches end of non-void function [-Wreturn-type]
10478
10479 v2: Shuffle around to avoid writing unreachable code
10480
10481 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10482 Tested-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10483 Reviewed-by: Keith Packard <keithp@keithp.com>
10484
10485commit 8e2bac0a69173b51e17a39d70e1d154b47fd0139
10486Author: Ryan Pavlik <rpavlik@iastate.edu>
10487Date: Fri Apr 20 13:46:21 2012 +0100
10488
10489 os/osinit.c, os/utils.c: Exclude sigaction code when building for MinGW
10490
10491 MinGW doesn't have sigaction, so this patch is needed for building.
10492
10493 No attempt is made to actually install the fatal error signal handler, as MinGW
10494 will simply terminate the process rather than deliver a fatal signal.
10495
10496 Also avoid using strsignal
10497
10498 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
10499 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10500 Tested-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10501 Reviewed-by: Keith Packard <keithp@keithp.com>
10502
10503commit e54f71a2c7a1677a5c5032d1e7dc099d5204144b
10504Author: Chris Wilson <chris@chris-wilson.co.uk>
10505Date: Sat Nov 10 10:26:08 2012 +0000
10506
10507 xf86: select a fake output for headless servers
10508
10509 Following commit 37d956e3ac9513b74078882dff489f9b0a7a5a28
10510 Author: Keith Packard <keithp@keithp.com>
10511 Date: Mon Sep 10 11:14:20 2012 +1000
10512
10513 xf86: fix compat output selection for no output GPUs
10514
10515 headless servers can no longer startup as we no longer select a compat
10516 output for the fake framebuffer.
10517
10518 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=56343
10519 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
10520 Reviewed-by: Dave Airlie <airlied@redhat.com>
10521 Signed-off-by: Keith Packard <keithp@keithp.com>
10522
10523commit 76d9c62eb2be2010a19bf36285012d086cdd180b
10524Author: Thierry Reding <thierry.reding@avionic-design.de>
10525Date: Wed Oct 17 12:06:47 2012 +0200
10526
10527 xf86: Fix non-PCI configuration-less setups
10528
10529 For non-PCI video devices, such as those found on many ARM embedded
10530 systems, the X server currently requires the BusID option to specify the
10531 full path to the DRM device's sysfs node in order to properly match it
10532 against the probed platform devices.
10533
10534 In order to allow X to start up properly if either the BusID option was
10535 omitted or no configuration is present at all, the first video device is
10536 used by default.
10537
10538 Signed-off-by: Thierry Reding <thierry.reding@avionic-design.de>
10539 Reviewed-by: Dave Airlie <airlied@redhat.com>
10540 Signed-off-by: Keith Packard <keithp@keithp.com>
10541
10542commit d50a94568e06a06df682a0028579d37a7b209d15
10543Merge: 87b60ef cf05db8
10544Author: Keith Packard <keithp@keithp.com>
10545Date: Tue Nov 27 15:14:26 2012 -0800
10546
10547 Merge remote-tracking branch 'daniels/master'
10548
10549commit 87b60efb859d86fd9e96306112bc5453a549fa69
10550Author: Colin Harrison <colin.harrison@virgin.net>
10551Date: Sun Sep 2 22:39:32 2012 +0100
10552
10553 hw/xwin: Don't spam MotionNotify events when the mouse hasn't moved
10554
10555 Don't spam MotionNotify events when the mouse hasn't moved, when polling the
10556 mouse position outside any X window
10557
10558 (Test with 'xev -root' after mouse polling has started.)
10559
10560 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
10561 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10562
10563commit 886f4c389266bf759414f15a5e25e50b5935dc9d
10564Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10565Date: Mon Jun 25 15:58:21 2012 +0100
10566
10567 hw/xwin: Don't log all fbConfigs and GL/WGL extensions
10568
10569 Don't log all fbConfigs and GL/WGL extensions, unless verbose logging
10570 is requested
10571 Log the number of pixelFormats which gave rise to the fbConfigs
10572
10573 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10574 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10575
10576commit 9b10f20f472e6b0d6114fb61c88906e349e47f2b
10577Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10578Date: Fri Feb 10 13:36:46 2012 +0000
10579
10580 hw/xwin: Use pre-computed extent of damage for creating GDI region
10581
10582 When -clipupdates option is specified, use the pre-computed extent of damage,
10583 rather than computing a GDI region which combines all the individual boxes in
10584 the damage.
10585
10586 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10587 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10588
10589commit 3ab95a2a5c848e8ce4c39be243fd5e2ac6bb1c1a
10590Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10591Date: Fri Feb 24 15:32:26 2012 +0000
10592
10593 hw/xwin: Ensure we own the clipboard before checking the format of it's contents
10594
10595 Ensure we own the clipboard before checking the format of it's contents, this
10596 prevents the contents from changing underneath us.
10597
10598 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10599 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10600
10601commit a4f357c620ae9e2270316197a61fb8e1d5183a51
10602Author: Colin Harrison <colin.harrison@virgin.net>
10603Date: Mon Apr 9 15:48:08 2012 +0100
10604
10605 hw/xwin: Disable minimize button on window with skip-taskbar state
10606
10607 If a window has had its taskbar button removed, disable its minimize
10608 button to prevent it becoming lost
10609
10610 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
10611 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10612
10613commit a2983452ee1e2051ab54cded7a3a7735be020db6
10614Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10615Date: Wed Jul 8 17:48:03 2009 +0100
10616
10617 hw/xwin: Process _NET_WM_STATE_SKIP_TASKBAR hint in multiwindow mode.
10618
10619 Set WS_EX_TOOLWINDOW style to hide window from Alt-Tab switcher
10620
10621 Use ITaskBarList interface to ensure that the taskbar notices if the window has
10622 changed it's style in a way which affects if the taskbar shows it or not
10623
10624 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10625 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10626
10627commit de759cc38ccc8a7811923eb9620f01301aedc10e
10628Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10629Date: Sun Feb 5 17:57:04 2012 +0000
10630
10631 hw/xwin: Add a simple interface to the ITaskbarList COM interface
10632
10633 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10634 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10635
10636commit 19d875187a113ca7772a770767ea2ceb2ade5a43
10637Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10638Date: Wed Jan 4 17:40:14 2012 +0000
10639
10640 hw/xwin: Enable RANDR resize by default
10641
10642 Also fix erroneous use of '--resize' not '-resize', and document '-noresize' in man page
10643
10644 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10645 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10646
10647commit 444580673221bf3f89dfa3cbc518b5ed4ecabc23
10648Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10649Date: Sat Dec 31 22:11:18 2011 -0600
10650
10651 hw/xwin: Enable AIGLX by default
10652
10653 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
10654 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10655 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10656
10657commit 11bb32e561b3f1c657a99b3902f9beea786babbb
10658Author: Oliver Schmidt <oschmidt-mailinglists@gmx.de>
10659Date: Sun Jan 8 20:30:02 2012 +0000
10660
10661 hw/xwin: Restore non-latching modifier key state when an X window gains focus
10662
10663 In multiwindow mode, the state of the modifier keys was lost when a window is
10664 created (or raised) and focus moved to that window.
10665
10666 For example: In window A Ctrl + some key opens a window B, then in window B Ctrl
10667 + some other key triggers the next action. However after the opening of window B
10668 the Ctrl key has to be released and pressed again. If the user keeps the Ctrl
10669 key held down when the window B is opened, the next key press X will be
10670 interpreted as X and not as Ctrl+X.
10671
10672 Extended the function winRestoreModeKeyStates in winkeybd.c to consider not only
10673 the latching modifier keys but also the modifiers Ctrl, Shift, Alt/AltGr by
10674 using the Windows function GetAsyncKeyState.
10675
10676 A combined Ctrl+AltGr modifier state cannot be restored correctly, as Windows
10677 always fakes a Ctrl-L when AltGr is pressed.
10678
10679 Signed-off-by: Oliver Schmidt <oschmidt-mailinglists@gmx.de>
10680 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10681 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10682
10683commit 47291d0b7d7bfee74156f83badae15a0818c68ad
10684Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10685Date: Mon Nov 5 14:54:51 2012 +0000
10686
10687 hw/xwin: Give our logical xor operator a more logical name
10688
10689 Also, rather than a comment about why we need a logical operator, let's have a
10690 comment about what we are doing to the keyboard state...
10691
10692 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10693 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10694
10695commit 8aa27ae82109e4fab0ff3ed86ad1d152438a2585
10696Author: Tobias Häußler <tobias.haeussler@gmx.de>
10697Date: Mon Jul 18 14:48:23 2011 +0100
10698
10699 hw/xwin: Add correct taskbar grouping of X windows on Windows 7
10700
10701 I created a small patch for XWin that adds correct grouping of taskbar icons
10702 when 'Always combine, hide labels' is set in the taskbar properties. It uses the
10703 new taskbar APIs introduced in Windows 7 to set an application id for each
10704 window. The id is based on the X11 class hints.
10705
10706 v2: Add file to _SOURCES to fix distcheck
10707
10708 v3 : Fix compilation with mingw-w64 w32api headers
10709 Include propkey.h, propsys.h rather than defining necessary stuff ourselves
10710
10711 v4: Fix up names taskbar->propertystore, AppID->AppUserModelID, etc.
10712 Link directly with ole32 for PropVariantClear(), prototyping it if neccessary.
10713
10714 v5: Put winSetAppUserModelID()-related code in a separate file.
10715 Drop a superfluous assign to hr of ignored HRESULT of SetValue()
10716
10717 Signed-off-by: Tobias Häußler <tobias.haeussler@gmx.de>
10718 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10719 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10720
10721commit f3aef7f9561d2723da0d1438a8b276b77a1e672e
10722Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10723Date: Sun Sep 2 16:04:25 2012 +0100
10724
10725 hw/xwin: Fix function name in log message
10726
10727 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10728 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10729
10730commit 127a7068b75258d396776cc101aba0bef10a5cef
10731Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10732Date: Thu Apr 14 23:10:23 2011 +0100
10733
10734 hw/xwin: Allow XScreenSaverSuspend() to effect Windows screen saver or monitor power off
10735
10736 Reset the idle timer in the block handler if screenSaverSuspended is set.
10737
10738 This isn't quite a complete solution. We should also set the block timeout to
10739 something less than the idle timer timeout to ensure we will reset the idle
10740 timer before it times out if we are blocking.
10741
10742 The idle timer timeout is presumably the first one to expire of the screen saver
10743 activation or monitor low power or power down timeout, depending on
10744 configuration.
10745
10746 Unfortunately this is probably not straightforward to do. Whilst
10747 SystemParametersInfo(SPI_GETSCREENSAVETIMEOUT) is portable, apparently
10748 SPI_GETLOWPOWERTIMEOUT and SPI_GETPOWEROFFTIMEOUT are not supported by Windows
10749 versions 6.0 or later, and the interface for discovering equivalent value is
10750 complex.
10751
10752 This doesn't matter in the case where a media player or similar application is
10753 the one making the XScreenSaverSuspend() requests, as it will be continuously
10754 drawing, causing the X server to become unblocked often.
10755
10756 In the case where slide show presentation application or similar is the one
10757 making the XScreenSaverSuspend() request, this might be a problem.
10758
10759 Since "Powerpoint is Evil" [1], I think we'll leave it like this till someone
10760 complains :-)
10761
10762 [1] http://www.edwardtufte.com/tufte/books_pp
10763
10764 v2: conditionalize on SCREENSAVER
10765
10766 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10767 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10768
10769commit 190f165317c4a58adf38e3e32acfe7cf04850e91
10770Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10771Date: Mon Nov 5 13:02:46 2012 +0000
10772
10773 hw/xwin: Avoid logging an extra blank line if BUILDERSTRING is empty
10774
10775 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10776 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10777
10778commit ef9dbea24f5ce7e1e400463ab2a333076d98a28e
10779Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10780Date: Wed Aug 31 20:57:05 2011 +0100
10781
10782 hw/xwin: Report which drawing engines are being enabled when checking for support
10783
10784 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10785 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10786
10787commit 5824166256c8e96c91e225f4ccb2b094b8fb045b
10788Author: Jon TURNEY <jon.turney@dronecode.org.uk>
10789Date: Fri Mar 12 14:38:51 2010 +0000
10790
10791 hw/xwin: Show any fatal error message
10792
10793 Report the fatal error message in the dialog we pop up, rather than just
10794 referring the user to the logfile.
10795
10796 v2: Do this a better way since the "Pass the FatalError message to
10797 OsVendorFatalError" patch has landed, and OsVendorFatalError() now gets passed
10798 the fatal error message
10799
10800 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
10801 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
10802
10803commit cf05db8ddec948d907915e1c9569d743e8771def
10804Author: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
10805Date: Tue Oct 30 17:02:58 2012 +0100
10806
10807 Ephyr: Find the right host screen when embedded
10808
10809 When Xephyr is embedded into another application (e.g. in xoo), the
10810 input events may come from the parent Window instead of our own. So make
10811 sure we find the host screen in that case as well instead of crashing.
10812
10813 Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
10814 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
10815
10816commit ac1a60e7b6f06fd075cc5bf55d6bc67206a01d29
10817Author: Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>
10818Date: Fri Nov 16 19:51:58 2012 +0200
10819
10820 vfb: Initialize the GLX extension again.
10821
10822 This should fix a regression in the 1.13.0 release: commit
10823 5f5bbbe543f65c48ecbb5cce80116a86ca3fbe86 removed a code path used by Xvfb
10824 and made it use the default one when initializing extensions. However, this
10825 meant the GLX extension was not initialized anymore since it is not part of
10826 the `staticExtensions' array.
10827
10828 Since it is not possible to just add it to that array after commit
10829 aad428b8e21c77397c623b78706eb64b1fea77c9, adopt an approach similar to
10830 xwin's and xquartz's and initialize the extension from vfb's `InitOutput'.
10831
10832 Signed-off-by: Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>
10833 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
10834
10835commit 353b14ea77324b3f7b3ea29d6f03618fd9916d5f
10836Author: Daniel Stone <daniel@fooishbar.org>
10837Date: Mon Nov 19 15:00:22 2012 +1100
10838
10839 DMX: Add DMX and GLX extensions
10840
10841 Unfortunately this also got lost in the extmod fallout, leaving the DMX
10842 server not exposing the DMX or GLX extensions.
10843
10844 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
10845 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10846
10847commit 3556d43010f0a37bf30aa2e0ceee9a8bbcad311d
10848Author: Daniel Stone <daniel@fooishbar.org>
10849Date: Mon Nov 19 15:29:34 2012 +1100
10850
10851 Constify extensions in LoadExtension users
10852
10853 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
10854 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10855
10856commit db83830eef38a86a471378121213162da180537d
10857Author: Daniel Stone <daniel@fooishbar.org>
10858Date: Mon Nov 19 15:28:07 2012 +1100
10859
10860 Constify argument to LoadExtension
10861
10862 Since we never modify it.
10863
10864 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
10865 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10866
10867commit 3b9f1c701787965246638c1a6fd99fb2b6078114
10868Author: Yuly Novikov <ynovikov@chromium.org>
10869Date: Mon Nov 19 21:04:57 2012 -0500
10870
10871 dix: Save touchpoint last coordinates before transform. #49347
10872
10873 DDXTouchPointInfoRec.valuators used to store axis values after transform.
10874 This resulted in Coordinate Transformation Matrix
10875 being applied multiple times to the last coordinates,
10876 in the case when only pressure changes in the last touch event.
10877
10878 Changed DDXTouchPointInfoRec.valuators to store values before transform.
10879
10880 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=49347
10881
10882 Signed-off-by: Yuly Novikov <ynovikov@chromium.org>
10883 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10884 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10885
10886commit d0fd592fc7be1706084923f43fa057e50122562c
10887Author: Thomas Jaeger <ThJaeger@gmail.com>
10888Date: Sat Oct 13 23:18:50 2012 -0400
10889
10890 Simplify GetTouchEvents
10891
10892 With only one callee left, we are free to assume that
10893 !(flags & TOUCH_CLIENT_ID)
10894
10895 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
10896 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10897 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10898
10899commit fe59774c55e5d423633405e0869c22f4ce382548
10900Author: Thomas Jaeger <ThJaeger@gmail.com>
10901Date: Sat Oct 13 23:08:27 2012 -0400
10902
10903 Don't use GetTouchEvents in EmitTouchEnd
10904
10905 As before GetTouchEvents causes unwanted side effects. Add a new
10906 function GetDixTouchEnd, which generates a touch event from the touch
10907 point. We fill in the event's screen coordinates from the MD's current
10908 sprite position.
10909
10910 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
10911 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10912 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10913
10914commit cc79107a5b60d2926e16ddbee04149e8d5acc969
10915Author: Thomas Jaeger <ThJaeger@gmail.com>
10916Date: Sat Oct 13 22:51:24 2012 -0400
10917
10918 Don't use GetTouchEvents when replaying events
10919
10920 GetTouchEvents has plenty of side effects such as moving the pointer or
10921 updating the master device, which we don't want to happen when
10922 replaying. The only reason for calling it was to generate a DCCE event,
10923 but GetTouchEvents doesn't even do that right (we might need a DCCE
10924 event even when replaying a master event, or clients could interpret
10925 valuator data incorrectly).
10926
10927 This discussion is moot at the moment anyway, since DeliverTouchEvents
10928 doesn't appear to deliver DCCE events.
10929
10930 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
10931 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10932
10933 [Added call to processInputProc instead of direct call to DeliverTouchEvents]
10934
10935 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10936
10937commit 90b177e5cbb9047453af426d2abfe9c0514fe866
10938Author: Thomas Jaeger <ThJaeger@gmail.com>
10939Date: Sat Oct 13 22:43:26 2012 -0400
10940
10941 Update the MD's position when a touch event is received
10942
10943 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
10944 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10945 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10946
10947commit 6a6c3afe71ac82a93d9fd0034dd5bbdcf0eae1ea
10948Merge: 14c9e24 6f14508
10949Author: Keith Packard <keithp@keithp.com>
10950Date: Sun Nov 18 21:50:20 2012 -0800
10951
10952 Merge remote-tracking branch 'whot/for-keith'
10953
10954commit 6f145084d5b9ca4a023dfc538a69bbf30edeac4a
10955Author: Adam Jackson <ajax@redhat.com>
10956Date: Fri Nov 16 13:46:32 2012 -0500
10957
10958 linux: Prefer ioctl(KDSKBMUTE, 1) over ioctl(KDSKBMODE, K_OFF)
10959
10960 K_OFF is a slightly broken interface, since if some other process
10961 (cough, systemd) sets the console state to K_UNICODE then it undoes
10962 K_OFF, and now Alt-F2 will switch terminals instead of summoning the
10963 Gnome "run command" dialog.
10964
10965 KDSKBMUTE separates the "don't enqueue events" logic from the keymap, so
10966 doesn't have this problem. Try it first, then continue falling back to
10967 older methods.
10968
10969 Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=859485
10970 Tested-by: Josh Boyer <jwboyer@redhat.com>
10971 Signed-off-by: Adam Jackson <ajax@redhat.com>
10972 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10973 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10974
10975commit 0ff1c77b17062a3a3ff520709ef719316bc1d8bb
10976Author: Peter Hutterer <peter.hutterer@who-t.net>
10977Date: Fri Nov 16 14:21:47 2012 +1000
10978
10979 xfixes: let PointerBarrierClient->device_ids point to allocated memory
10980
10981 "Fixed" up Jasper's original patch [1] to use pointers instead of
10982 zero-length arrays but didn't actually set the pointer to anywhere.
10983
10984 [1] 9a536820f6174befb22431bab375ef1af0a1ec29
10985
10986 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
10987 Reviewed-By: Jasper St. Pierre <jstpierre@mecheye.net>
10988
10989commit aa9a9ad08b8a6e7e95de7c2bf45d93dd50f9ca87
10990Author: Daniel Martin <consume.noise@gmail.com>
10991Date: Wed Nov 7 11:31:01 2012 +0100
10992
10993 Xi: Fix modifier swapping in XIPassiveGrabDevice
10994
10995 XIPassiveGrabDevice uses a list of uint32_t as modifier sets.
10996
10997 The ModifierInfo struct represents the current modifier states and is
10998 therefor used in XIQueryPointer and various events.
10999
11000 Signed-off-by: Daniel Martin <consume.noise@gmail.com>
11001 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11002 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11003
11004commit aa9da5eae109154edf740a5c1eee4bf98e8fbd4f
11005Author: Thomas Jaeger <ThJaeger@gmail.com>
11006Date: Sat Oct 13 22:39:27 2012 -0400
11007
11008 remove init_event
11009
11010 The function is identical to init_device_event from inpututils.c with
11011 the first two arguments swapped.
11012
11013 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
11014 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11015 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11016
11017commit 710065da37b68a9ec7a0c9dc995247f39e3d01e9
11018Author: Daniel Stone <daniel@fooishbar.org>
11019Date: Mon Nov 5 15:58:25 2012 +1100
11020
11021 XKB: Remove component listing support
11022
11023 No-one uses this - not xkbcomp, not GNOME, not KDE. The preferred way
11024 to deal with component listing (which gives you RMLVO rather than
11025 KcCGST) is to use the XML files on the client side.
11026
11027 Indeed, a couple of hours after making this commit, it emerged that all
11028 *.dir files built with xkbcomp 1.1.1 (released two years ago) and later
11029 have been catastrophically broken and nearly empty. So I think that's
11030 reasonable proof that no-one uses them.
11031
11032 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11033 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11034 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11035
11036commit fd214aabf77c44ba0ca3fd438f0c3c8eb22c7211
11037Author: Peter Hutterer <peter.hutterer@who-t.net>
11038Date: Tue Nov 6 10:27:25 2012 +1000
11039
11040 input: drop FP1616 macro
11041
11042 The double_to_f1616() functions do the same thing, and they're tested.
11043
11044 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11045 Reviewed-by: Keith Packard <keithp@keithp.com>
11046
11047commit 14c9e245bae2447f0e57a3ea4daa8d2085fa8b8f
11048Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
11049Date: Sat Nov 17 11:52:10 2012 -0500
11050
11051 XQuartz: Opt-in to NSSupportsAutomaticGraphicsSwitching
11052
11053 http://xquartz.macosforge.org/trac/ticket/668
11054
11055 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
11056
11057commit 011f8458805e443ac9130865d2840a929a00cabf
11058Merge: 54ba26c b4e44b2
11059Author: Keith Packard <keithp@keithp.com>
11060Date: Mon Nov 5 17:16:07 2012 -0800
11061
11062 Merge remote-tracking branch 'whot/for-keith'
11063
11064commit b4e44b285ed0eee1d06514215a4b01d54f40094b
11065Author: Carlos Garnacho <carlosg@gnome.org>
11066Date: Wed Oct 31 19:29:45 2012 +0100
11067
11068 Xi: Set modifier mask on touch events
11069
11070 Button mask should be out-of-band with the emulated
11071 pointer events as touch devices don't truly have
11072 "buttons". Even though, it's handy to have the modifier
11073 mask from the paired keyboard on touch events.
11074
11075 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
11076 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11077 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11078
11079commit 863f32c930d71073ee5f78452b78bd459d024867
11080Author: Carlos Garnacho <carlosg@gnome.org>
11081Date: Wed Oct 31 19:32:57 2012 +0100
11082
11083 Xi: Update the device after delivering the emulated pointer event(#56558)
11084
11085 Ensure emulated pointer events contain the state that applies before the
11086 event was processed, so the device state must be updated after delivering
11087 such emulated events.
11088
11089 Co-authored-by: Peter Hutterer <peter.hutterer@who-t.net>
11090 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
11091 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11092 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11093
11094commit 9a536820f6174befb22431bab375ef1af0a1ec29
11095Author: Jasper St. Pierre <jstpierre@mecheye.net>
11096Date: Sat Nov 3 20:23:46 2012 -0400
11097
11098 cursor: Fix up implementation for per-device barriers
11099
11100 Support multiple mast devices being specified as a parameter to the
11101 barrier. This should implement all parts of the XFixes specification,
11102 minus the existing non-specification detail where specifying no devices
11103 is the same as specifying XIAllDevices.
11104
11105 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
11106 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11107 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11108
11109commit 944213bbc7099f1e0238f39973653d459efdc2c9
11110Author: Jasper St. Pierre <jstpierre@mecheye.net>
11111Date: Sat Nov 3 20:23:45 2012 -0400
11112
11113 cursor: Clean up barrier finding code a bit
11114
11115 The current code manually unrolls the loop, by finding a barrier,
11116 clamping it, and then repeaing the code. Since we want to modify
11117 both passes of the loop to support device-specific barriers, make
11118 it into an actual loop.
11119
11120 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
11121 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11122
11123commit 1536bc2d213713af45541a15b5e49c552d8d0592
11124Author: Jasper St. Pierre <jstpierre@mecheye.net>
11125Date: Sat Nov 3 20:23:44 2012 -0400
11126
11127 cursor: Clean up pointer barrier creation code a tiny bit
11128
11129 This will make it much simpler when we add more error paths to the code
11130 that constructs pointer barrier clients.
11131
11132 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
11133 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11134
11135commit 04c885de715a7c989e48fc8cf2e61db2b401de2d
11136Author: Jasper St. Pierre <jstpierre@mecheye.net>
11137Date: Sat Nov 3 20:23:43 2012 -0400
11138
11139 cursor: CreatePointerBarrier has a variable request length
11140
11141 Support this, and swap the devices in the SProc as well. Don't
11142 actually do anything with the devices just yet -- specifying any
11143 should still emit a BadImplementation.
11144
11145 Based on a patch by Peter Hutterer <peter.hutterer@who-t.net>
11146
11147 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
11148 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11149
11150commit d74b63f95fc622afe83c53580f1b55346d55fa7a
11151Author: Jasper St. Pierre <jstpierre@mecheye.net>
11152Date: Sat Nov 3 20:23:42 2012 -0400
11153
11154 cursor: Fix a minor unused variable warning
11155
11156 Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
11157 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11158 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11159
11160commit aad65415bff12c6860c19beac42e4165e598a40f
11161Author: Peter Hutterer <peter.hutterer@who-t.net>
11162Date: Thu Oct 11 13:02:27 2012 +1000
11163
11164 dix: don't allow disabling XTest devices
11165
11166 Disabling a XTest device followed by an XTest API call crashes the server.
11167 This could be fixed elsewhere but disabled devices must not send events
11168 anyway. The use-case for disabled XTest devices is somewhat limited, so
11169 simply disallow disabling the devices.
11170
11171 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11172 Reviewed-by: Keith Packard <keithp@keithp.com>
11173
11174commit 5b7384a3154a95a805b040e1910e276b52aada96
11175Author: Alexey Ten (Lynn) <alexeyten@gmail.com>
11176Date: Tue Oct 23 12:32:31 2012 +0400
11177
11178 Apply partial matches for option (#25873)
11179
11180 Rules which match star (*) and option, like one below, should be applied
11181
11182 layout[2] option = symbols
11183 * misc:typo = +typo(base)
11184
11185 This is port of patch from #19563 (https://bugs.freedesktop.org/19563)
11186 because here we have own copy of maprules.c
11187
11188 X.Org Bug 25873 <http://bugs.freedesktop.org/show_bug.cgi?id=25873>
11189
11190 Signed-off-by: Alexey Ten (Lynn) <alexeyten@gmail.com>
11191 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11192 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11193
11194commit 54ba26cb1f9c59559cc3c449abeb31b2ce23bdba
11195Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11196Date: Mon Oct 29 22:37:37 2012 -0500
11197
11198 os: Add libnettle as a choice of SHA1 implementation
11199
11200 libnettle is smaller than libgcrypt, currently being released more
11201 frequently, and has replaced the latter in gnutls-3.x (which is used
11202 by TigerVNC, so they can avoid pulling in two crypto libraries
11203 simultaneously).
11204
11205 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11206 Reviewed-by: Julien Cristau <jcristau@debian.org>
11207
11208commit 2ff56033de2b493a11d2bdf411b7057b1b3a22d7
11209Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11210Date: Tue Oct 16 02:11:59 2012 -0500
11211
11212 Xi: fix fprint format warning
11213
11214 exevents.c: In function 'ProcessTouchEvent':
11215 exevents.c:1601:20: warning: too many arguments for format
11216
11217 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11218 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11219
11220commit d631dbe9b6a1faa4cd18ab7a4a0276db02b84b77
11221Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11222Date: Sat Oct 27 21:50:01 2012 -0500
11223
11224 Xext: fix shadow warnings
11225
11226 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11227 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11228
11229commit 20cf7918ed954a932a4cc404b3d0c4ca87a6ae16
11230Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11231Date: Tue Oct 16 21:08:52 2012 -0500
11232
11233 Xext: fix redundant redeclaration warnings
11234
11235 panoramiX.c:595:13: warning: redundant redeclaration of 'CreateConnectionBlock'
11236 ../include/dix.h:167:23: note: previous declaration of 'CreateConnectionBlock' was here
11237 xres.c:193:13: warning: redundant redeclaration of 'ResExtensionInit'
11238 ../include/extinit.h:109:13: note: previous declaration of 'ResExtensionInit'
11239 xtest.c:60:12: warning: redundant redeclaration of 'DeviceValuator'
11240 ../Xi/exglobals.h:61:12: note: previous declaration of 'DeviceValuator' was here
11241
11242 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11243 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11244
11245commit 89447e1b50d2fb0e046102664045d1f9d4efc542
11246Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11247Date: Mon Oct 15 01:33:05 2012 -0500
11248
11249 Xext: fix unused variable warnings in xf86bigfont.c
11250
11251 These were rendered unused by commit 2c7c520cfe0df30f4bc3adba59d9c62582823bf8.
11252
11253 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11254 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11255
11256commit 344eea237fc07dedfd733d14f95ed0ad26bb5f81
11257Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11258Date: Tue Oct 16 02:04:34 2012 -0500
11259
11260 xkb: fix shadow warnings
11261
11262 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11263 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11264
11265commit 7f1d74e8a906210eafc637df81ded62c3adff748
11266Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11267Date: Tue Oct 16 02:06:54 2012 -0500
11268
11269 render: fix shadow warnings
11270
11271 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11272 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11273
11274commit 8e86123998d26d26f0f60dcbb9836c38e5a4c3a5
11275Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11276Date: Tue Oct 16 01:19:12 2012 -0500
11277
11278 randr: export more provider property symbols
11279
11280 These were added as part of commit 66d92afeaeed9f4a19267d95a1f81b9bf27162a5
11281 but never declared or exported. Fixes warnings:
11282
11283 rrproviderproperty.c:255:1: warning: no previous prototype for 'RRPostProviderPendingProperties'
11284 rrproviderproperty.c:327:1: warning: no previous prototype for 'RRConfigureProviderProperty'
11285
11286 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11287 Reviewed-by: Dave Airlie <airlied@redhat.com>
11288 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11289
11290commit f02e27e4fcc34413b2051e5a01edc92172fa8612
11291Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11292Date: Tue Oct 16 02:16:17 2012 -0500
11293
11294 mi: fix shadow warnings
11295
11296 mibitblt.c: In function 'miGetImage':
11297 mibitblt.c:617:20: warning: declaration of 'pt' shadows a previous local
11298 mibitblt.c:609:17: warning: shadowed declaration is here
11299 mispans.c: In function 'miFillUniqueSpanGroup':
11300 mispans.c:456:33: warning: declaration of 'i' shadows a previous local
11301 mispans.c:382:9: warning: shadowed declaration is here
11302 mispans.c:488:17: warning: declaration of 'i' shadows a previous local
11303 mispans.c:382:9: warning: shadowed declaration is here
11304
11305 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11306 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11307
11308commit 1fe30c00679bd36a6355b48b94b87564d528ff28
11309Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11310Date: Tue Oct 16 02:21:04 2012 -0500
11311
11312 fb: fix shadow warnings
11313
11314 fbblt.c: In function 'fbBlt':
11315 fbblt.c:76:16: warning: declaration of 'src' shadows a previous local
11316 fbblt.c:52:13: warning: shadowed declaration is here
11317 fbblt.c:77:16: warning: declaration of 'dst' shadows a previous local
11318 fbblt.c:52:19: warning: shadowed declaration is here
11319 fbbltone.c: In function 'fbBltPlane':
11320 fbbltone.c:742:13: warning: declaration of 'w' shadows a previous local
11321 fbbltone.c:725:9: warning: shadowed declaration is here
11322
11323 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11324 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11325
11326commit e8d45f301845f70b76407577b92363934ca4f19e
11327Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11328Date: Tue Oct 16 20:54:56 2012 -0500
11329
11330 dix: fix shadow warnings
11331
11332 dispatch.c: In function 'ProcCopyArea':
11333 dispatch.c:1608:5: warning: declaration of 'rc' shadows a previous local
11334 dispatch.c:1604:9: warning: shadowed declaration is here
11335 dispatch.c: In function 'ProcCopyPlane':
11336 dispatch.c:1647:5: warning: declaration of 'rc' shadows a previous local
11337 dispatch.c:1643:9: warning: shadowed declaration is here
11338 events.c: In function 'GetClientsForDelivery':
11339 events.c:2030:68: warning: declaration of 'clients' shadows a global declaration
11340 ../include/dix.h:124:28: warning: shadowed declaration is here
11341 events.c: In function 'DeliverEventToWindowMask':
11342 events.c:2113:19: warning: declaration of 'clients' shadows a global declaration
11343 ../include/dix.h:124:28: warning: shadowed declaration is here
11344 events.c: In function 'EventSuppressForWindow':
11345 events.c:4420:12: warning: declaration of 'free' shadows a global declaration
11346
11347 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11348 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11349
11350commit 1aa783754e21a263b0973516850656b13fd18f0d
11351Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11352Date: Tue Oct 16 14:06:30 2012 -0500
11353
11354 dix: fix redundant redeclaration warnings in dixfont
11355
11356 These functions are already declared in <X11/fonts/fontproto.h>.
11357 Redeclaring them just for _X_EXPORT causes tons of warnings throughout
11358 xserver, but they need to be declared somewhere to be picked up by
11359 sdksyms.sh. Doing so in a private header limits the warnings to
11360 sdksyms.c; fixing those as well would require changes to fontsproto.
11361
11362 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11363 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11364
11365commit 27c5966de35d4726dd9795b4828d4236851f6a88
11366Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11367Date: Mon Oct 15 01:25:17 2012 -0500
11368
11369 xfree86: os-support: fix old-style function definition warnings
11370
11371 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11372 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11373 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11374
11375commit 64b961bb21369aaea694d883f361a36bc23b19b9
11376Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11377Date: Wed Oct 24 23:57:11 2012 -0500
11378
11379 macros: clarify documentation
11380
11381 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11382 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11383
11384commit 04f980ead5e9af066143a921e0a0e98f26f28ee5
11385Merge: a194630 a4941ce
11386Author: Keith Packard <keithp@keithp.com>
11387Date: Mon Nov 5 08:45:21 2012 -0800
11388
11389 Merge remote-tracking branch 'jturney/mingw-w64-w32api'
11390
11391commit a4941ce4961f254ac31e2cef6dd0863866246bab
11392Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11393Date: Tue Aug 21 15:31:16 2012 +0100
11394
11395 hw/xwin: Wrap 'Status' when including ddraw.h
11396
11397 Status is #defined as an alias for a type in xkbsrv.h, which conflicts with it's
11398 use as a parameter name in rpcdce.h
11399
11400 This fixes compilation with MinGW-w64 w32api headers
11401
11402 (The MinGW-w64 w32api headers actually provide a ddraw.h, so this fix tries to
11403 do things in a way which will still work if our private copy of ddraw.h is
11404 eventually removed)
11405
11406 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11407 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11408
11409commit 57bbf6e2aead6cce76a9f5d43829b58868a26059
11410Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11411Date: Mon Oct 29 14:35:45 2012 +0000
11412
11413 hw/xwin: Remove pointless winFinishCreateWindowsWindowDDNL()
11414
11415 Remove pointless winFinishCreateWindowsWindowDDNL() and the useless DirectDraw
11416 surface pointers in the window privates
11417
11418 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11419 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11420
11421commit 5bc4c53d23d02d3d7297a8f00a92b6e4d5e0d6fe
11422Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11423Date: Tue Sep 4 17:28:16 2012 +0100
11424
11425 hw/xwin: Link with libdxguid rather than defining the DirectX GUIDs ourselves
11426
11427 Decruftify by linking with libdxguid rather than defining the DirectX GUIDs
11428 ourselves. This fixes compilation with MinGW-w64 w32api headers, which do not
11429 provide GUID_SECT (which only ever did anything for gcc prior to 2.95, anyhow)
11430
11431 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11432 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11433
11434commit a194630f7f7b287cb4ea4a459df0745f4c0d4c1a
11435Merge: 74a3130 d511a30
11436Author: Keith Packard <keithp@keithp.com>
11437Date: Thu Nov 1 13:38:54 2012 -0700
11438
11439 Merge remote-tracking branch 'whot/for-keith'
11440
11441commit d511a3016a79c50cb38e7504d4831a9ae128e422
11442Author: Benjamin Tissoires <benjamin.tissoires@gmail.com>
11443Date: Fri Oct 26 16:27:34 2012 +0200
11444
11445 Add missing labels for multitouch valuators
11446
11447 ABS_MT_DISTANCE exists since kernel v2.6.38,
11448 ABS_MT_TOOL_X|Y appeared in v3.6.
11449
11450 Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
11451 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11452 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11453
11454commit ced56f322ead10d1bc93fcd1f8e0ec3ae51292a3
11455Author: Carlos Garnacho <carlosg@gnome.org>
11456Date: Thu Oct 25 15:03:50 2012 +0200
11457
11458 Sync TouchListener memory allocation with population in TouchSetupListeners()
11459
11460 The allocated TouchListener array may fall short by 1 if hitting the worst case
11461 situation where there's an active grab, passive grabs on each window in the
11462 sprite trace and event selection for touch in one of the windows. This may lead
11463 to memory corruptions as the array is overflown.
11464
11465 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
11466 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11467 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11468
11469commit 74a3130426d7e1b456e45daa517769cd3e4b58c0
11470Merge: 1ca096d f32ad6d
11471Author: Keith Packard <keithp@keithp.com>
11472Date: Mon Oct 29 09:09:16 2012 -0700
11473
11474 Merge remote-tracking branch 'jturney/master'
11475
11476commit 1ca096d5e07221025c4c4110528772b7d94f15ee
11477Author: Michel Dänzer <michel.daenzer@amd.com>
11478Date: Mon Oct 29 12:57:54 2012 +0100
11479
11480 EXA: Track source/mask pixmaps more explicitly for Composite fallback regions.
11481
11482 In particular, make sure pExaScr->src/maskPix are cleared when the
11483 corresponding pictures aren't associated with drawables, i.e. solid or gradient
11484 pictures. Without this, we would in some cases associate the source/mask region
11485 with unrelated pixmaps from previous Composite fallbacks, resulting in random
11486 corruption.
11487
11488 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=47266
11489
11490 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
11491 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
11492 Signed-off-by: Keith Packard <keithp@keithp.com>
11493
11494commit f32ad6dd314a587c301b7f1a6e4ee74e5d8cc06c
11495Author: Ryan Pavlik <rpavlik@iastate.edu>
11496Date: Fri Oct 21 12:32:49 2011 -0500
11497
11498 os/utils.c: Provide only stubs for Lock/UnlockServer on WIN32
11499
11500 MinGW and MSVC lack the POSIX functions to compile the lock file code.
11501
11502 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
11503 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11504 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11505
11506commit 74735c5ebff15db8891808068fe8569d8ccde9e7
11507Author: Colin Harrison <colin.harrison@virgin.net>
11508Date: Sat Feb 18 20:37:19 2012 +0000
11509
11510 hw/xwin: fmemopen is available on cygwin but not MinGW
11511
11512 This is a follow-up to commit 0659437f5ec0e3f646373394f5f9c5461e2170f3.
11513
11514 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
11515 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11516 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11517
11518commit 748be9da20a8bf3dd99701a95eac7ee3895a28de
11519Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11520Date: Tue May 1 20:08:41 2012 +0100
11521
11522 doc: Update documentation about Windows platforms support a bit
11523
11524 Update what c-extensions says about Windows platforms support a bit:
11525 - Document that MinGW Win32 gets let off being POSIX-compilant
11526 - Document the minimum Windows version supported
11527
11528 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11529 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11530
11531commit 760be785eb49fed4652950fb55dc07b0a41d87de
11532Author: Peter Hutterer <peter.hutterer@who-t.net>
11533Date: Wed Oct 24 11:20:54 2012 +1000
11534
11535 xfree86: remove unused variable sigstate
11536
11537 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11538 Reviewed-by: Keith Packard <keithp@keithp.com>
11539
11540commit e7cd5cce740e653000fb1192b600268dcf77dde2
11541Author: Peter Hutterer <peter.hutterer@who-t.net>
11542Date: Thu Oct 18 15:11:31 2012 +1000
11543
11544 dix: fix zaphod screen scrossing (#54654)
11545
11546 POINTER_SCREEN coordinates are screen-relative. For a Zaphod setup, the
11547 coordinates after a screen crossing are already relative to the new screen's
11548 origin. Add that offset to the coordinates before re-setting.
11549
11550 regression introduced by
11551 commit bafbd99080be49a17be97d2cc758fbe623369945
11552 Author: Peter Hutterer <peter.hutterer@who-t.net>
11553 Date: Wed Aug 8 11:34:32 2012 +1000
11554
11555 dix: work around scaling issues during WarpPointer (#53037)
11556
11557 X.Org Bug 54654 <http://bugs.freedesktop.org/show_bug.cgi?id=54654>
11558
11559 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11560 Reviewed-by: Keith Packard <keithp@keithp.com>
11561
11562commit 676447190190d8546165e21be242cf16dd69f5ae
11563Author: Peter Hutterer <peter.hutterer@who-t.net>
11564Date: Wed Oct 17 14:13:29 2012 +1000
11565
11566 Xi: don't deliver TouchEnd to a client waiting for TouchBegin (#55738)
11567
11568 If a client is still waiting for the TouchBegin, don't deliver a TouchEnd
11569 event.
11570
11571 X.Org Bug 55738 <http://bugs.freedesktop.org/show_bug.cgi?id=55738>
11572
11573 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11574 Tested-by: Thomas Jaeger <thjaeger@gmail.com>
11575 Reviewed-by: Keith Packard <keithp@keithp.com>
11576
11577commit 3018f9c1e5109680dcf69b8f2d7807696a473bde
11578Author: Peter Hutterer <peter.hutterer@who-t.net>
11579Date: Fri Oct 12 13:48:27 2012 +1000
11580
11581 Xi: set xChangeDeviceControlReply.status to Success by default
11582
11583 If the status is other than Success, the code will set it to the required
11584 value.
11585
11586 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11587 Reviewed-by: Keith Packard <keithp@keithp.com>
11588
11589commit 2decff6393a44b56d80d53570718f95354fde454
11590Author: Peter Hutterer <peter.hutterer@who-t.net>
11591Date: Thu Oct 11 16:03:33 2012 +1000
11592
11593 xkb: ProcesssPointerEvent must work on the VCP if it gets the VCP
11594
11595 For button release events, the current code picks the VCK. Because that has
11596 a XKB struct, it thinks this is a PointerKeys event and proceeds to send the
11597 release event through the XTest pointer. That has no effect in normal
11598 operation as the button is never down and an attempt is silently discarded
11599 (normal event processing continues with the VCP).
11600
11601 On server shutdown, the XTest device is already removed, leading to a
11602 null-pointer derefernce when the device is checked for whether buttons are
11603 down (XkbFakeDeviceButton → button_is_down(xtest pointer)).
11604
11605 The current state has only worked by accident, the right approach here is to
11606 handle the VCP's event as such and not switch to the keyboard.
11607
11608 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11609 Reviewed-by: Keith Packard <keithp@keithp.com>
11610
11611commit cc1d8fa8a4ef2e34e9837a44945094fd77e40a0b
11612Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11613Date: Sun Oct 14 23:38:20 2012 -0500
11614
11615 Configure fixes for MinGW
11616
11617 As for Cygwin, define FD_SETSIZE for uses of select() to be correct.
11618 Link with libpthread.
11619
11620 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
11621 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11622
11623commit 53830281b4da096f9c13107d73ec9c76ff1d14cc
11624Merge: abc2ef5 9cbcb5b
11625Author: Keith Packard <keithp@keithp.com>
11626Date: Fri Oct 26 18:04:34 2012 -0700
11627
11628 Merge remote-tracking branch 'sandmann/for-keithp'
11629
11630commit abc2ef590c7cb33ebe21726f83b7347406975f95
11631Merge: 0e85e5e c5396ec
11632Author: Keith Packard <keithp@keithp.com>
11633Date: Fri Oct 26 17:11:58 2012 -0700
11634
11635 Merge remote-tracking branch 'whot/for-keith'
11636
11637commit 9cbcb5bd6a5360a128d15b77a02d8d3351f74366
11638Author: Søren Sandmann Pedersen <ssp@redhat.com>
11639Date: Wed May 30 05:19:08 2012 -0400
11640
11641 Use new pixman_glyph_cache_t API that will be in pixman 0.28.0
11642
11643 This new API allows glyphs to be cached in a data structure in pixman,
11644 and entire glyph strings to be composited in one go.
11645
11646 Also bump pixman dependency to 0.27.2.
11647
11648 Results from the cairo peformance test suite running against Xvfb with
11649 a screen size of 1680x1050@32bpp:
11650
11651 Speedups
11652 ========
11653 xlib firefox-talos-gfx 12416.63 -> 3603.93 3.45x speedup
11654 ██▌
11655 xlib xfce4-terminal-a1 1727.57 -> 1048.85: 1.65x speedup
11656 ▋
11657 xlib evolution 1370.49 -> 869.34: 1.58x speedup
11658 ▋
11659 xlib gnome-terminal-vim 1832.83 -> 1251.94: 1.46x speedup
11660 ▌
11661 xlib poppler 1519.70 -> 1204.05: 1.26x speedup
11662 ▎
11663 xlib firefox-planet-gnome 6982.55 -> 5598.16: 1.25x speedup
11664 ▎
11665 xlib ocitysmap 1142.77 -> 1071.53: 1.07x speedup
11666 ▏
11667
11668 No slowdowns were reported.
11669
11670 Results of x11perf -aa10text:
11671
11672 Before:
11673
11674 8000000 reps @ 0.0007 msec (1450000.0/sec)
11675 8000000 reps @ 0.0007 msec (1460000.0/sec)
11676 8000000 reps @ 0.0007 msec (1460000.0/sec)
11677 8000000 reps @ 0.0007 msec (1470000.0/sec)
11678 8000000 reps @ 0.0007 msec (1480000.0/sec)
11679 40000000 trep @ 0.0007 msec (1460000.0/sec)
11680
11681 After:
11682
11683 32000000 reps @ 0.0002 msec (4910000.0/sec)
11684 32000000 reps @ 0.0002 msec (4830000.0/sec)
11685 32000000 reps @ 0.0002 msec (4890000.0/sec)
11686 32000000 reps @ 0.0002 msec (4830000.0/sec)
11687 32000000 reps @ 0.0002 msec (4900000.0/sec)
11688 160000000 trep @ 0.0002 msec (4870000.0/sec)
11689
11690 Version 2: Destroy the glyph cache at server regen time
11691
11692 Acked-by: Aaron Plattner <aplattner@nvidia.com>
11693 Reviewed-by: Keith Packard <keithp@keithp.com>
11694 Signed-off-by: Soren Sandmann <ssp@redhat.com>
11695
11696commit c5396ec05a5c6cab6608ba677f703c5227b1de13
11697Author: Thierry Reding <thierry.reding@avionic-design.de>
11698Date: Wed Oct 17 12:11:49 2012 +0200
11699
11700 xf86: Fix build against recent Linux kernel
11701
11702 Recent Linux kernels reworked the linux/input.h header file, which is
11703 now part of the "user-space API". The include guard therefore has an
11704 additional additional _UAPI prefix.
11705
11706 Instead of adding another case to the #ifdef, drop any include guard
11707 checks and instead always undefine the BUS_* definitions on Linux.
11708
11709 Signed-off-by: Thierry Reding <thierry.reding@avionic-design.de>
11710 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11711 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11712
11713commit 7d89eb4978a6649be22747c3de2886781e311647
11714Author: Colin Walters <walters@verbum.org>
11715Date: Wed Jan 4 17:37:06 2012 -0500
11716
11717 autogen.sh: Honor NOCONFIGURE=1
11718
11719 (Also while we have the patient open, use exec instead of hanging
11720 around uselessly in waitpid)
11721
11722 http://people.gnome.org/~walters/docs/build-api.txt
11723
11724 Reviewed-by: Adam Jackson <ajax@redhat.com>
11725 Signed-off-by: Colin Walters <walters@verbum.org>
11726 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11727
11728commit c0a752d2864872023216005375a6a1973fadeffe
11729Author: Lionel Elie Mamane <lionel@mamane.lu>
11730Date: Tue Oct 9 03:17:49 2012 +0200
11731
11732 dix: fix Ungrab action #55785
11733
11734 UngrabAllDevices(Bool kill_client):
11735 If we are not going to kill the client (kill_clients false),
11736 we need to deactivate grabs of active clients, too.
11737 (If we are going to kill the client,
11738 no need to deactivate the grab,
11739 as this will be done as part of the client kill.)
11740
11741 Fixes: X.Org Bug 55785 <http://bugs.freedesktop.org/show_bug.cgi?id=55785>
11742
11743 Signed-off-by: Lionel Elie Mamane <lionel@mamane.lu>
11744 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11745 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11746
11747commit 3b67cd26149eb991b5f015061a818af65369e668
11748Author: Chase Douglas <chase.douglas@canonical.com>
11749Date: Thu Jun 7 14:12:06 2012 -0700
11750
11751 End physically active touches when device is disabled
11752
11753 Otherwise:
11754
11755 * We can't end the touches while device is disabled
11756 * New touches after enabling the device may erroneously be mapped to old
11757 logical touches
11758
11759 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
11760 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11761 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
11762
11763commit 0e85e5e62819c9e516d6e19836f9f5d7a5d53e87
11764Author: Colin Harrison <colin.harrison@virgin.net>
11765Date: Thu Oct 11 18:05:47 2012 +0100
11766
11767 hw/xwin: Fix no return value warning in winClipboardProc()
11768
11769 winclipboardthread.c: In function ‘winClipboardProc’:
11770 winclipboardthread.c:415:9: warning: ‘return’ with no value, in function returning non-void
11771 winclipboardthread.c:424:13: warning: ‘return’ with no value, in function returning non-void
11772
11773 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
11774 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11775
11776commit 51ed6a7d46c6cc19eb9aaf838be41eb57fd8230a
11777Author: Colin Harrison <colin.harrison@virgin.net>
11778Date: Thu Oct 11 17:57:37 2012 +0100
11779
11780 hw/xwin: Remove unused variable in winmultiwindowwm.c
11781
11782 winmultiwindowwm.c: In function ‘UpdateName’:
11783 winmultiwindowwm.c:522:14: warning: unused variable ‘pszName’ [-Wunused-variable]
11784
11785 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
11786 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11787
11788commit 5886b1dc4d2ce12a238a2e3600fdb576abc6e419
11789Author: Colin Harrison <colin.harrison@virgin.net>
11790Date: Thu Oct 11 17:56:11 2012 +0100
11791
11792 hw/xwin: Remove unused variables in winwindow.c
11793
11794 winwindow.c: In function ‘winCreateWindowNativeGDI’:
11795 winwindow.c:62:19: warning: unused variable ‘pWinPriv’ [-Wunused-variable]
11796 winwindow.c: In function ‘winDestroyWindowNativeGDI’:
11797 winwindow.c:85:19: warning: unused variable ‘pWinPriv’ [-Wunused-variable]
11798 winwindow.c: In function ‘winPositionWindowNativeGDI’:
11799 winwindow.c:108:19: warning: unused variable ‘pWinPriv’ [-Wunused-variable]
11800 winwindow.c: In function ‘winChangeWindowAttributesNativeGDI’:
11801 winwindow.c:213:19: warning: unused variable ‘pWinPriv’ [-Wunused-variable]
11802 winwindow.c: In function ‘winUnmapWindowNativeGDI’:
11803 winwindow.c:241:19: warning: unused variable ‘pWinPriv’ [-Wunused-variable]
11804 winwindow.c: In function ‘winMapWindowNativeGDI’:
11805 winwindow.c:265:19: warning: unused variable ‘pWinPriv’ [-Wunused-variable]
11806
11807 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
11808 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11809
11810commit 08b33882f34fed0c23d54ff7361efc5833868a4b
11811Author: Colin Harrison <colin.harrison@virgin.net>
11812Date: Thu Oct 11 17:52:11 2012 +0100
11813
11814 hw/xwin: Remove unused variables in winvalargs.c
11815
11816 winvalargs.c: In function ‘winValidateArgs’:
11817 winvalargs.c:62:10: warning: unused variable ‘fImplicitScreenFound’ [-Wunused-variable]
11818 winvalargs.c:60:9: warning: unused variable ‘iMaxConsecutiveScreen’ [-Wunused-variable]
11819
11820 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
11821 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11822
11823commit 7e665f1a21468118dababdbf6d8573f1ce1ebd47
11824Author: Colin Harrison <colin.harrison@virgin.net>
11825Date: Thu Oct 11 17:50:42 2012 +0100
11826
11827 hw/xwin: Remove unused variables in winwindowswm.c
11828
11829 winwindowswm.c: In function ‘ProcWindowsWMQueryVersion’:
11830 winwindowswm.c:85:9: warning: unused variable ‘n’ [-Wunused-variable]
11831 winwindowswm.c: In function ‘SProcWindowsWMQueryVersion’:
11832 winwindowswm.c:565:9: warning: unused variable ‘n’ [-Wunused-variable]
11833
11834 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
11835 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11836
11837commit e893d570aa3e9ae8df6a30ac4adaf0949a2fc733
11838Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11839Date: Thu Oct 11 19:36:13 2012 +0100
11840
11841 Use X_ATTRIBUTE_PRINTF where suggested for os.h
11842
11843 InitOutput.c: In function ‘OsVendorInit’:
11844 InitOutput.c:630:29: warning: assignment left-hand side might be a candidate for a format attribute [-Wmissing-format-attribute]
11845 winprocarg.c: In function ‘ddxProcessArgument’:
11846 winprocarg.c:231:29: warning: assignment left-hand side might be a candidate for a format attribute [-Wmissing-format-attribute]
11847
11848 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11849 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11850
11851commit e59cc83bcdca6b64c1bfb2de395d741757f615f6
11852Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11853Date: Thu Oct 11 19:35:19 2012 +0100
11854
11855 hw/xwin: Use X_ATTRIBUTE_PRINTF where suggested for winmsg.c
11856
11857 winmsg.c: In function ‘winVMsg’:
11858 winmsg.c:47:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11859 winmsg.c: In function ‘winDrvMsg’:
11860 winmsg.c:56:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11861 winmsg.c: In function ‘winMsg’:
11862 winmsg.c:66:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11863 winmsg.c: In function ‘winDrvMsgVerb’:
11864 winmsg.c:77:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11865 winmsg.c: In function ‘winMsgVerb’:
11866 winmsg.c:87:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11867 winmsg.c: In function ‘winErrorFVerb’:
11868 winmsg.c:97:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11869 winmsg.c: In function ‘winDebug’:
11870 winmsg.c:107:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11871 winmsg.c: In function ‘winTrace’:
11872 winmsg.c:117:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11873
11874 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11875 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11876
11877commit 3d6e6a394ea2d5dcb77aa706f751423315d8a7ed
11878Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11879Date: Thu Oct 11 19:32:24 2012 +0100
11880
11881 hw/xwin: Use X_ATTRIBUTE_PRINTF where suggested for winerror.c
11882
11883 Also, fix the typoed name OsVenderVErrorF (sic), so it actually gets prototyped
11884 and remove redundant declarations of it's prototype.
11885
11886 winerror.c: In function ‘OsVendorVErrorF’:
11887 winerror.c:56:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11888 winerror.c: In function ‘winMessageBoxF’:
11889 winerror.c:104:5: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wmissing-format-attribute]
11890
11891 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11892 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11893
11894commit 13fb6b36b8742a79b9768944eee6d1ad66d7e4d9
11895Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11896Date: Thu Oct 11 16:20:35 2012 +0100
11897
11898 hw/xwin: Use char strings in winClipboardUNIXtoDOS for consistency with the rest of the clipboard code
11899
11900 winclipboardxevents.c: In function ‘winClipboardFlushXEvents’:
11901 winclipboardxevents.c:575:13: warning: passing argument 1 of ‘winClipboardUNIXtoDOS’ from incompatible pointer type
11902 winclipboard.h:102:2: note: expected ‘unsigned char **’ but argument is of type ‘char **’
11903
11904 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11905 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11906
11907commit fa36a7cd488e4f7fb91beed8a87764b8e0c1a72f
11908Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11909Date: Thu Oct 11 15:51:39 2012 +0100
11910
11911 hw/xwin: Fix warning about discarding const in initializing winKBLayouts
11912
11913 winlayouts.h:46:5: warning: initialization discards qualifiers from pointer target type
11914 winlayouts.h:46:5: warning: initialization discards qualifiers from pointer target type
11915 winlayouts.h:46:5: warning: initialization discards qualifiers from pointer target type
11916 [repeated for each layout]
11917
11918 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11919 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11920
11921commit 0ea65df8a5f97a6fe5225226493aa01380080d2f
11922Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11923Date: Thu Sep 27 17:23:43 2012 +0100
11924
11925 hw/xwin: Fix some of the warnings in generated gl wrapper code
11926
11927 Fix some of the warnings in generated gl wrapper code:
11928 - glWinResolveHelper takes a const char * argument
11929 - ensure formal parameter names don't collide with reserved names or shadow
11930 global declarations (e.g. near, far, index (from string.h), remainder (from
11931 math.h), pointer (from Xdefs.h) by postpending a '_'.
11932
11933 generated_gl_wrappers.c:13:3: warning: passing argument 2 of 'glWinResolveHelper' discards qualifiers from pointer target type
11934 glwrap.c:70:1: note: expected 'char *' but argument is of type 'const char *'
11935 [repeated many times]
11936 generated_gl_wrappers.c: In function 'glVertexAttrib1dARBWrapper': generated_gl_wrappers.c:6884:47: warning: declaration of 'index' shadows a global declaration [-Wshadow]
11937 generated_gl_wrappers.c:6886:3: warning: passing argument 2 of 'glWinResolveHelper' discards qualifiers from pointer target type
11938 [repeated many times]
11939
11940 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11941 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11942
11943commit f7f0739311927ad177dece7119272481b75b73b4
11944Author: Colin Harrison <colin.harrison@virgin.net>
11945Date: Tue Oct 19 23:42:53 2010 +0100
11946
11947 hw/xwin: Warning fix in ProcWindowsWMFrameSetTitle()
11948
11949 winwindowswm.c: In function ‘ProcWindowsWMFrameSetTitle’:
11950 winwindowswm.c:514: error: pointer targets in passing argument 2 of ‘strncpy’ differ in signedness
11951
11952 If you're going to stick random casts into your code, at least use the goddammed
11953 right ones. :-)
11954
11955 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
11956 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11957
11958commit 97c9ed026a7f7efe5fca02c188bcbdbcd7c594e9
11959Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11960Date: Wed Jan 25 19:01:55 2012 +0000
11961
11962 hw/xwin: Fix shadowed local variable i in HandleCustomWM_COMMAND()
11963
11964 Fix shadowed local variable i in HandleCustomWM_COMMAND()
11965
11966 Also, fds are meant to be representable as an int
11967
11968 winprefs.c: In function ‘HandleCustomWM_COMMAND’:
11969 winprefs.c:346:23: error: declaration of ‘i’ shadows a previous local
11970 winprefs.c:322:7: error: shadowed declaration is here
11971
11972 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11973 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11974
11975commit b55d0b92fc536793a877f6073754801a7c3c2dd5
11976Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11977Date: Wed Jan 25 18:56:22 2012 +0000
11978
11979 hw/xwin: Fix shadowed local variables in winLoadCursor()
11980
11981 Move shadowed local variables x and y to the places they are used
11982
11983 wincursor.c: In function ‘winLoadCursor’:
11984 wincursor.c:212:11: error: declaration of ‘x’ shadows a previous local
11985 wincursor.c:166:7: error: shadowed declaration is here
11986 wincursor.c:212:14: error: declaration of ‘y’ shadows a previous local
11987 wincursor.c:166:10: error: shadowed declaration is here
11988 wincursor.c:228:11: error: declaration of ‘x’ shadows a previous local
11989 wincursor.c:166:7: error: shadowed declaration is here
11990 wincursor.c:228:14: error: declaration of ‘y’ shadows a previous local
11991 wincursor.c:166:10: error: shadowed declaration is here
11992
11993 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11994 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
11995
11996commit a8c9c3699e6dfde29eeec058d903740a499dcf3b
11997Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11998Date: Thu Oct 11 19:11:36 2012 +0100
11999
12000 hw/xwin/glx: Fix using Mask as a formal parameter shadows a global typedef of the same name
12001
12002 Fix using Mask as a formal parameter shadows the typedef of the same name from X.h
12003
12004 indirect.c: In function 'GetShift':
12005 indirect.c:1629:14: warning: declaration of 'Mask' shadows a global declaration [-Wshadow]
12006
12007 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12008 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12009
12010commit 47df98c7851fd9e97a54f0b4950340506d4a9244
12011Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12012Date: Wed Jan 25 19:17:16 2012 +0000
12013
12014 hw/xwin: Fix using screenInfo as a formal parameter to InitOutput() shadows a global declaration
12015
12016 Using screenInfo as a formal parameter to InitOutput() shadows a global declaration of screenInfo.
12017 Change the formal parameter name from screenInfo to pScreenInfo, as everywhere else uses.
12018
12019 InitOutput.c: In function ‘InitOutput’:
12020 InitOutput.c:891:25: error: declaration of ‘screenInfo’ shadows a global declaration
12021 ../../include/scrnintstr.h:570:19: error: shadowed declaration is here
12022
12023 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12024 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12025
12026commit 97e9c34687f7dc05ee0bdef097a19e7623fb59a3
12027Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12028Date: Wed Jan 25 19:07:49 2012 +0000
12029
12030 hw/xwin: Fix using menu as a formal parameter shadows a global variable of the same name
12031
12032 Using menu as a formal parameter shadows a global variable of the same name
12033
12034 winprefsyacc.y: In function ‘SetRootMenu’:
12035 winprefsyacc.y:286:20: error: declaration of ‘menu’ shadows a global declaration
12036 winprefsyacc.y:55:19: error: shadowed declaration is here
12037 winprefsyacc.y: In function ‘SetDefaultSysMenu’:
12038 winprefsyacc.y:293:26: error: declaration of ‘menu’ shadows a global declaration
12039 winprefsyacc.y:55:19: error: shadowed declaration is here
12040
12041 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12042 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12043
12044commit efe96a17bd741f222ccb226d306b5c1dc1f25ade
12045Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12046Date: Wed Jan 25 18:54:57 2012 +0000
12047
12048 hw/xwin: Fix using index as a formal parameter shadows index()
12049
12050 Using index as a formal parameter shadows index() from strings.h
12051
12052 winallpriv.c: In function ‘winInitCmapPrivates’:
12053 winallpriv.c:119:45: error: declaration of ‘index’ shadows a global declaration
12054
12055 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12056 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12057
12058commit 451c5d9175cdeb36fd614502ff0317f968490dfc
12059Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12060Date: Wed Jan 25 19:09:12 2012 +0000
12061
12062 hw/xwin: Fix using index as a formal parameter in winscrinit.c shadows index()
12063
12064 Using index as a local variable shadows index() from strings.h
12065
12066 winscrinit.c: In function ‘winFinishScreenInitFB’:
12067 winscrinit.c:274:28: error: declaration of ‘index’ shadows a global declaration
12068 winscrinit.c: In function ‘winFinishScreenInitNativeGDI’:
12069 winscrinit.c:623:35: error: declaration of ‘index’ shadows a global declaration
12070
12071 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12072 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12073
12074commit 21faee4b383a12bef58f300b4ad1a3bd34d6de2d
12075Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12076Date: Wed Jan 25 19:05:42 2012 +0000
12077
12078 hw/xwin: Fix using index as a local variable shadows index()
12079
12080 Using index as a local variable shadows index() from strings.h
12081
12082 winprefs.c: In function ‘LoadImageComma’:
12083 winprefs.c:574:7: error: declaration of ‘index’ shadows a global declaration
12084
12085 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12086 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12087
12088commit 2d9123fd0c89c1a72e7281e56985196bc53507bf
12089Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12090Date: Wed Jan 25 18:59:21 2012 +0000
12091
12092 hw/xwin: Fix using index as a formal parameter in winmonitors.c shadows index()
12093
12094 Fix using index as a local variable shadows global declaration of index() from strings.h
12095
12096 winmonitors.c: In function ‘QueryMonitor’:
12097 winmonitors.c:59:23: error: declaration of ‘index’ shadows a global declaration
12098
12099 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12100 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12101
12102commit 5b0435dbdaa47404629800e4e9c5a6952a7bb260
12103Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12104Date: Wed Jan 25 18:53:02 2012 +0000
12105
12106 hw/xwin: Fix using system as a local variable in winCheckMount() shadows system()
12107
12108 Using system as local variable in winCheckMount() shadows the global declaration of system() from stdlib.h
12109
12110 InitOutput.c: In function ‘winCheckMount’:
12111 InitOutput.c:296:10: error: declaration of ‘system’ shadows a global declaration
12112
12113 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12114 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12115
12116commit 17d84c743d6b52f32c4ca0341aebaff972522945
12117Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12118Date: Wed Jan 25 17:18:12 2012 +0000
12119
12120 hw/xwin: Remove obsolete redundant declarations of winPushPixels()
12121
12122 winPushPixels() and winpushpxl.c was removed in it's entirety in commit
12123 f31bd087e8a7f65cd588bd1d022bb18e72b2a60c "Death to mfb"
12124
12125 wingc.c:37:1: error: redundant redeclaration of ‘winPushPixels’
12126 win.h:1104:1: note: previous declaration of ‘winPushPixels’ was here
12127
12128 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12129 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12130
12131commit f06b468d96afd323e28645dde44170ab367da4f1
12132Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12133Date: Wed Jan 25 17:09:18 2012 +0000
12134
12135 hw/xwin: Fix redundant declarations in winmultiwindowwm.c
12136
12137 winmultiwindowwm.c:139:14: error: redundant redeclaration of ‘display’
12138 ../../include/opaque.h:52:14: note: previous declaration of ‘display’ was here
12139 winmultiwindowwm.c:140:13: error: redundant redeclaration of ‘ErrorF’
12140 ../../include/os.h:558:13: note: previous declaration of ‘ErrorF’ was here
12141
12142 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12143 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12144
12145commit 3e8269177ee01fe107830032b60a8d311d4b4b0e
12146Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12147Date: Wed Jan 25 17:08:17 2012 +0000
12148
12149 hw/xwin: Fix redundant declarations in winclipboardwrappers.c
12150
12151 Fix redundant declarations in winclipboardwrappers.c.
12152 Also remove obsolete winProcQueryTree() prototype.
12153
12154 winclipboardwrappers.c:64:12: error: redundant redeclaration of ‘g_iNumScreens’
12155 winglobals.h:33:12: note: previous declaration of ‘g_iNumScreens’ was here
12156 winclipboardwrappers.c:67:13: error: redundant redeclaration of ‘g_fXdmcpEnabled’
12157 winglobals.h:45:13: note: previous declaration of ‘g_fXdmcpEnabled’ was here
12158 winclipboardwrappers.c:75:27: error: redundant redeclaration of ‘winProcEstablishConnectionOrig’
12159 winglobals.h:66:27: note: previous declaration of ‘winProcEstablishConnectionOrig’ was here
12160 winclipboardwrappers.c:76:27: error: redundant redeclaration of ‘winProcQueryTreeOrig’
12161 winglobals.h:67:27: note: previous declaration of ‘winProcQueryTreeOrig’ was here
12162 winclipboardwrappers.c:77:27: error: redundant redeclaration of ‘winProcSetSelectionOwnerOrig’
12163 winglobals.h:68:27: note: previous declaration of ‘winProcSetSelectionOwnerOrig’ was here
12164
12165 winclipboardwrappers.c:56:5: warning: redundant redeclaration of ‘winProcEstablishConnection’ [-Wredundant-decls]
12166 winclipboardwrappers.c:52:5: note: previous declaration of ‘winProcEstablishConnection’ was here
12167 winclipboardwrappers.c:57:5: warning: redundant redeclaration of ‘winProcSetSelectionOwner’ [-Wredundant-decls]
12168 winclipboardwrappers.c:54:5: note: previous declaration of ‘winProcSetSelectionOwner’ was here
12169
12170 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12171 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12172
12173commit 0975fa0b391d7b8d7e8a127ebf7a863e53fcab7d
12174Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12175Date: Wed Jan 25 17:06:27 2012 +0000
12176
12177 hw/xwin: Fix redundant declaration in winclipboardthread.c
12178
12179 winclipboardthread.c:51:22: error: redundant redeclaration of ‘serverGeneration’
12180 ../../include/misc.h:362:22: note: previous declaration of ‘serverGeneration’ was here
12181 winclipboardthread.c: In function ‘winClipboardProc’:
12182
12183 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12184 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12185
12186commit 4a1da00c495706e0fbcec2b30aef3ad7acf5b703
12187Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12188Date: Wed Jan 25 17:05:25 2012 +0000
12189
12190 hw/xwin: Fix redundant declaration in winclipboardinit.c
12191
12192 In file included from winclipboardinit.c:35:0:
12193 winclipboard.h:81:13: error: redundant redeclaration of ‘ErrorF’
12194 ../../include/os.h:558:13: note: previous declaration of ‘ErrorF’ was here
12195
12196 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12197 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12198
12199commit 30f31b80f0105676359105c895bc195b9f3e704c
12200Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12201Date: Wed Jan 25 16:48:45 2012 +0000
12202
12203 hw/xwin: Fix various warnings in code generated by lex/yacc
12204
12205 Remove various redundant declarations.
12206
12207 Instruct lex not to generate code which isn't going to be used, thus avoiding
12208 some unused function warnings.
12209
12210 Note that bison <2.5 expects stdlib.h to define _STDLIB_H if it's been included,
12211 and avoids prototyping free/malloc in that case. Some stdlib.h (e.g. newlib) may
12212 not match precisely (e.g. defining _STDLIB_H_), so we define it after including
12213 stdlib.h to be sure, and avoid a redudant declaration warning for those symbols.
12214
12215 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12216 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12217
12218commit 4231b9541c6fd26f43b605ecc9dd76622dd786f7
12219Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12220Date: Wed Jan 25 16:38:49 2012 +0000
12221
12222 hw/xwin: Fix redundant declaration in winprefs.c
12223
12224 winprefs.c:63:14: error: redundant redeclaration of ‘display’
12225 ../../include/opaque.h:52:14: note: previous declaration of ‘display’ was here
12226
12227 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12228 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12229
12230commit 70087c55fffc3cbc52f550418e6b1230bbb1b3b3
12231Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12232Date: Wed Jan 25 16:36:40 2012 +0000
12233
12234 hw/xwin: Fix redundant declaration warnings in winmonitors.c
12235
12236 Include xwin-config.h so HAVE_STRLCPY and HAVE_STRNDUP are defined, so
12237 duplicate declarations of strlcpy(), strlcat() and strndup() aren't made
12238 by os.h
12239
12240 In file included from ../../include/misc.h:111:0,
12241 from ../../include/screenint.h:51,
12242 from ../../include/scrnintstr.h:51,
12243 from win.h:157,
12244 from winmonitors.c:31:
12245 ../../include/os.h:491:15: error: redundant redeclaration of ‘strlcpy’
12246 /usr/include/string.h:86:35: note: previous declaration of ‘strlcpy’ was here
12247 ../../include/os.h:492:15: error: redundant redeclaration of ‘strlcat’
12248 /usr/include/string.h:85:35: note: previous declaration of ‘strlcat’ was here
12249 ../../include/os.h:496:15: error: redundant redeclaration of ‘strndup’
12250
12251 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12252 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12253
12254commit 9904c8fe75c96cf84d8885aa95f2743c1b25ea37
12255Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12256Date: Wed Jan 25 16:09:59 2012 +0000
12257
12258 hw/xwin: Remove completely unused winregistry.c
12259
12260 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12261 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12262
12263commit a69429a17bf4630f6e26f61630a1c2b287202627
12264Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12265Date: Fri Sep 7 14:58:06 2012 +0100
12266
12267 Fix 'make distcheck' for hw/xwin
12268
12269 Add some missing header files to _SOURCES so they get distributed properly
12270
12271 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12272 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
12273
12274commit 1d9fd7ffb03ff6e44f3a8736a94e4558203ce607
12275Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12276Date: Thu Sep 6 13:51:38 2012 +0100
12277
12278 Fix compilation of Xorg DDX without XF86VIDMODE
12279
12280 Fix compilation of Xorg DDX without XF86VIDMODE since 6e74fdda, by putting
12281 xf86vmode.c back under the XF86VIDMODE automake conditional it was accidentally
12282 taken out of.
12283
12284 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12285 Tested-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
12286 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12287
12288commit 25741b1aa3a0fb39cb842a43ab52935673cfdefa
12289Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12290Date: Sun Sep 2 15:17:05 2012 +0100
12291
12292 hw/xwin: Only add GLX extension once.
12293
12294 Using of LoadExtension() to add GLX onto to the end of the list of extensions on
12295 every server regeneration leads to hilarious consequences (i.e. crashing) after
12296 the first regeneration :-)
12297
12298 Possibly xf86ExtensionInit() and QuartzExtensionInit() need a similar fix.
12299
12300 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12301 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
12302 Tested-by: Colin Harrison <colin.harrison@virgin.net>
12303
12304commit 7ceb85481280fecbf033f147f0f7834cad3890c6
12305Author: Jon TURNEY <jon.turney@dronecode.org.uk>
12306Date: Sun Sep 2 14:55:25 2012 +0100
12307
12308 Correct description of -displayfd option in man page.
12309
12310 A display number, not a port number, is written to the specified fd.
12311
12312 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
12313 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12314
12315commit 4b7f00346daed20c96f3e8ea13ae411858a5424b
12316Author: Peter Hutterer <peter.hutterer@who-t.net>
12317Date: Wed Oct 10 13:33:48 2012 +1000
12318
12319 dix: fix crash on shutdown if a disabled device is still grabbed (XI1 grab)
12320
12321 A disabled device doesn't have a sprite (less so a sprite->win) and triggers
12322 a NULL-pointer dereference on shutdown when all active grabs are released as
12323 part of the cleanup.
12324
12325 Fix this by checking for sprite being non-null and setting the focus window
12326 to the NullWindow if it is. The rest of the patch just attempts to make
12327 things more readable.
12328
12329 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12330 Reviewed-by: Keith Packard <keithp@keithp.com>
12331
12332commit 0c7109f3215378bb3ed6c00c71129a63ffb4e787
12333Author: Denys Vlasenko <dvlasenk@redhat.com>
12334Date: Mon Oct 8 14:18:02 2012 +1000
12335
12336 os: fix typo in OsSigHandler() error message
12337
12338 Recieved → Received
12339
12340 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
12341 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12342
12343commit 0a75bd640b3dc26b89d9e342999a7f4b7e98edbf
12344Author: Peter Hutterer <peter.hutterer@who-t.net>
12345Date: Wed Oct 3 13:12:27 2012 +1000
12346
12347 xfree86: add xf86UpdateDesktopDimensions()
12348
12349 This call is required for external drivers (specifically NVIDIA) that do
12350 not share the xfree86 infrastructure to update the desktop dimensions.
12351 Without it, the driver would update the ScreenRecs but not update the total
12352 dimensions the input code relies on for transformation.
12353
12354 This call is a thin wrapper around the already-existing internal call and
12355 should be backported to all stable series servers, with the minor ABI bump.
12356
12357 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12358 CC: Andy Ritger <aritger@nvidia.com>
12359 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
12360
12361commit 09f1e5b15b769e1122f0a8d7cae0820038992312
12362Author: Julien Cristau <jcristau@debian.org>
12363Date: Sun Oct 7 18:40:35 2012 +0200
12364
12365 Revert "kinput: allocate enough space for null character."
12366
12367 This reverts commit 531785dd746d64ef7f473a83ca73bb20e74b6fca.
12368
12369 The above commit breaks Xephyr option parsing. Andrzej writes:
12370
12371 Xephyr -retro -keybd evdev,,device=/dev/input/event2,xkbrules=evdev,xkbmodel=evdev,xkblayout=pl -mouse evdev,,device=/dev/input/event1 :3
12372
12373 results in:
12374
12375 <snip>
12376 Pointer option key (device=) of value (/dev/input/event1) not assigned!
12377 Kbd option key (device=) of value (/dev/input/event2) not assigned!
12378 Kbd option key (xkbrules=) of value (evdev) not assigned!
12379 Kbd option key (xkbmodel=) of value (evdev) not assigned!
12380 Kbd option key (xkblayout=) of value (pl) not assigned!
12381 <snip>
12382
12383 The effect of the patch is that the "key=value" pairs are parsed in such
12384 a way that the key is added an "equals" sign to it and we end up with
12385 keys like "device=" instead of "device". This in turn has effect on
12386 KdParsePointerOptions and KdParseKbdOptions: the key does not match
12387 any choice presented in the "switch" statement, and so "Pointer/Kbd
12388 option key (...) of value (...) not assigned!" happens, making all
12389 "key=value" options inaccessible to the user. Reverting the patch makes
12390 them available again.
12391
12392 Reference: http://bugs.debian.org/689246
12393 Reported-by: Andrzej Pietrasiewicz <andrzejtp2010@gmail.com>
12394 Signed-off-by: Julien Cristau <jcristau@debian.org>
12395 Cc: Dave Airlie <airlied@redhat.com>
12396 Reviewed-by: Søren Sandmann <ssp@redhat.com>
12397 Signed-off-by: Keith Packard <keithp@keithp.com>
12398
12399commit 7f9d78d8ffb1a312f9b3d69c05323f601571e162
12400Author: Keith Packard <keithp@keithp.com>
12401Date: Thu Oct 4 22:51:06 2012 -0700
12402
12403 os: Don't re-declare ConnectionInputPtr and ConnectionOutputPtr
12404
12405 They're declared in osdep.h, so don't redeclare them in io.c as
12406 well. Keeps the compiler happier.
12407
12408 Signed-off-by: Keith Packard <keithp@keithp.com>
12409 Reviewed-by: Dave Airlie <airlied@redhat.com>
12410
12411commit d5bf6f95f31037bd49b11348b500c3c13b7e0c99
12412Author: Keith Packard <keithp@keithp.com>
12413Date: Thu Oct 4 14:42:37 2012 -0700
12414
12415 Fix FlushClient to write extraBuf when provided (regression fix)
12416
12417 In commit:
12418
12419 commit 092c57ab173c8b71056f6feb3b9d04d063a46579
12420 Author: Adam Jackson <ajax@redhat.com>
12421 Date: Fri Jun 17 14:03:01 2011 -0400
12422
12423 os: Hide the Connection{In,Out}put implementation details
12424
12425 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12426 Signed-off-by: Adam Jackson <ajax@redhat.com>
12427
12428 the check for an empty output buffer was moved from one calling
12429 location into the FlushClient implementation itself. However, this
12430 neglected the possibility that additional data, in the form of
12431 'extraBuf' would be passed to FlushClient from other code paths. If the
12432 output buffer happened to be empty at that time, the extra data would
12433 never be written to the client.
12434
12435 This is fixed by checking the total data to be written, which includes
12436 both pending and extra data, instead of just the pending data.
12437
12438 Signed-off-by: Keith Packard <keithp@keithp.com>
12439 Reviewed-by: Julien Cristau <jcristau@debian.org>
12440
12441commit 8367dd9736d74eca971da345c2bf559ce5bbf649
12442Merge: 36c18bb 7998e26
12443Author: Keith Packard <keithp@keithp.com>
12444Date: Thu Oct 4 13:08:35 2012 -0700
12445
12446 Merge remote-tracking branch 'whot/for-keith'
12447
12448commit 36c18bb81be619df712778bcb99dd6f1fa38701a
12449Author: Stephan Schreiber <info@fs-driver.org>
12450Date: Wed Aug 29 20:03:58 2012 +0200
12451
12452 int10: fix pci_device_read_rom usage
12453
12454 I noticed that the build-in int10 driver always reports
12455 "Unable to retrieve all of segment 0x0C0000."
12456 even though the entire BIOS data is retrieved with success.
12457
12458 The associated code is in hw/xfree86/int10/generic.c, in the function
12459 xf86ExtendedInitInt10():
12460
12461 if (pci_device_read_rom(pInt->dev, vbiosMem) < V_BIOS_SIZE) {
12462 xf86DrvMsg(screen, X_WARNING,
12463 "Unable to retrieve all of segment 0x0C0000.\n");
12464 }
12465
12466 The function pci_device_read_rom() is from libpciaccess; its return
12467 value is not a size but an error status code: 0 means success.
12468 If pci_device_read_rom() returns 0 for success, the warning is generated.
12469
12470 The proposed patch corrects the evaluation of the return value of
12471 pci_device_read_rom() and of the supplied BIOS size.
12472
12473 Debian bug#686153
12474
12475 Signed-off-by: Julien Cristau <jcristau@debian.org>
12476 Reviewed-by: Dave Airlie <airlied@redhat.com>
12477 Signed-off-by: Keith Packard <keithp@keithp.com>
12478
12479commit 4bf3eac5fe20fb203b917a486f69514c55be595a
12480Author: Adam Jackson <ajax@redhat.com>
12481Date: Thu Sep 27 17:12:19 2012 -0400
12482
12483 configure: Stop using AM_MAINTAINER_MODE
12484
12485 All this does is make it so editing configure.ac or Makefile.am doesn't
12486 rebuild the makefiles. Which is just stupid.
12487
12488 v2: Remove --enable-maintainer-mode from autogen.sh
12489
12490 Signed-off-by: Adam Jackson <ajax@redhat.com>
12491 Reviewed-by: Eric Anholt <eric@anholt.net>
12492 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
12493 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
12494 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
12495 Signed-off-by: Keith Packard <keithp@keithp.com>
12496
12497commit 262679d436941580485a10e30938974958f7d588
12498Author: Keith Packard <keithp@keithp.com>
12499Date: Thu Oct 4 11:42:22 2012 -0700
12500
12501 Set version to 1.14 development branch (1.13.99.0)
12502
12503 Signed-off-by: Keith Packard <keithp@keithp.com>
12504
12505commit f179b13b91c8891254d31345ee04e773839e6b87
12506Merge: 4dd5989 3d2b768
12507Author: Keith Packard <keithp@keithp.com>
12508Date: Thu Oct 4 11:39:46 2012 -0700
12509
12510 Merge remote-tracking branch 'daniels/master'
12511
12512commit 7998e26159893674be69183a73a89a53f5608d58
12513Author: Jason Gerecke <killertofu@gmail.com>
12514Date: Fri Sep 28 15:03:42 2012 -0700
12515
12516 Fix additional gcc -Wwrite-strings warning in xf86 ddx
12517
12518 Commit 09e4b78f missed a case.
12519
12520 Signed-off-by: Jason Gerecke <killertofu@gmail.com>
12521 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
12522 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12523
12524commit f64254d85e731d0b4369d871a9a735b03f283ba6
12525Author: Peter Hutterer <peter.hutterer@who-t.net>
12526Date: Thu Aug 23 15:00:24 2012 +1000
12527
12528 mi: drop two useless conditions in miPointerSetPosition
12529
12530 pDev cannot be NULL here since fill_pointer_events is the only caller.
12531 And if the screen is NULL, then the device tries to send events before it is
12532 fully initialised. That certainly shouldn't happen and would be a bug
12533 elsewhere.
12534
12535 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12536 Reviewed-by: Keith Packard <keithp@keithp.com>
12537
12538commit 314776eb369ca2e438907795ae030dd743c281fc
12539Author: Daniel Stone <daniel@fooishbar.org>
12540Date: Fri Sep 7 18:30:23 2012 +0100
12541
12542 Touch: Fix duplicate TouchBegin selection with virtual devices
12543
12544 Given the following scenario:
12545 1) client A selects for TouchBegin on window W for device D
12546 2) client B selects for TouchBegin on window W for XIAllDevices
12547 3) client C selects for TouchBegin on window W with device E
12548
12549 Step 3 will fail with BadImplementation, because attempting to look up
12550 XIAllDevices or XIAllMasterDevices with dixLookupDevices doesn't work.
12551 This should succeed (or, if it was selecting for device D, fail with
12552 BadAccess as it would be a duplicate selection).
12553
12554 Fix this by performing the appropriate lookup for virtual devices.
12555
12556 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12557 Cc: Peter Hutterer <peter.hutterer@who-t.net>
12558 Cc: Chase Douglas <chase.douglas@ubuntu.com>
12559 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12560
12561commit 3e6358ee6c33979329b78fe2097a1fdf76fb69cd
12562Author: Daniel Drake <dsd@laptop.org>
12563Date: Fri Sep 7 21:48:35 2012 -0400
12564
12565 Xi: Don't check for TOUCH_END, it's never set
12566
12567 This flag is never set, so checking for it here means that we'll
12568 never release the simulated mouse button press after the user touches
12569 (and releases) the touchscreen for the first time.
12570
12571 Fixes a problem where the XO laptop touchpad became totally
12572 unusable after touching the screen for the first time (since X then
12573 behaved as if the mouse button was held down all the time).
12574
12575 Signed-off-by: Daniel Drake <dsd@laptop.org>
12576 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
12577 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12578
12579commit 9d6b8365702e4648e793fea21ad22f7174558680
12580Author: Peter Hutterer <peter.hutterer@who-t.net>
12581Date: Fri Sep 28 11:49:29 2012 +1000
12582
12583 dix: fix crash on XI 1.x grabs on disabled devices. (#54934)
12584
12585 If the device is disabled, the sprite window is NULL and dereferencing
12586 crashes the server.
12587
12588 This is only triggered for XI 1.x grabs (ProcXGrabDevice) as XI2 grabs would
12589 trigger another code path, creating a sprite for the disabled device as if
12590 detaching it (which is wrong and fixed with this patch too).
12591
12592 Grabbing a disabled device doesn't make sense as it won't send events
12593 anyway. However, the protocol specs do not prohibit it, so we need to keep
12594 it working.
12595 Luckily, oldWin is only used for focus out events, which aren't necessary
12596 given that the device is disabled.
12597
12598 X.Org Bug 54934 <http://bugs.freedesktop.org/show_bug.cgi?id=54934>
12599
12600 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12601 Reviewed-by: Chase Douglas <chase.douglas@ubuntu.com>
12602
12603commit 3d2b768efae9936c6929c2bc13c7a1acc074ecd3
12604Author: Daniel Stone <daniel@fooishbar.org>
12605Date: Fri Sep 7 18:17:46 2012 +0100
12606
12607 Touch: Fix duplicate TouchBegin selection with virtual devices
12608
12609 Given the following scenario:
12610 1) client A selects for TouchBegin on window W for device D
12611 2) client B selects for TouchBegin on window W for XIAllDevices
12612 3) client C selects for TouchBegin on window W with device E
12613
12614 Step 3 will fail with BadImplementation, because attempting to look up
12615 XIAllDevices or XIAllMasterDevices with dixLookupDevices doesn't work.
12616 This should succeed (or, if it was selecting for device D, fail with
12617 BadAccess as it would be a duplicate selection).
12618
12619 Fix this by performing the appropriate lookup for virtual devices.
12620
12621 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12622 Reviewed-by: Chase Douglas <chase.douglas@ubuntu.com>
12623
12624commit e3903a9383351b061b1a99dfc653ca50de764ec4
12625Author: Frederic Plourde <frederic.plourde@collabora.com>
12626Date: Wed Sep 12 10:48:32 2012 -0400
12627
12628 Add glXCreateNewContext support in Xephyr #54798
12629
12630 Similar to how we intercept and pass through CreateContext, also pass
12631 through newer CreateNewContext requests.
12632
12633 Fixes Clutter → Xephyr → VirtualBox.
12634
12635 Signed-off-by: Frederic Plourde <frederic.plourde@collabora.com>
12636 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12637
12638commit 11afebc92ce1a7462ff2886286504425b1c8f0ba
12639Author: Daniel Stone <daniel@fooishbar.org>
12640Date: Fri Sep 7 18:10:35 2012 +0100
12641
12642 Xephyr: GLX: Support MakeContextCurrent and MakeCurrentReadSGI
12643
12644 We need to pass these requests through to the host server in the same
12645 way we do glXMakeCurrent. Generalise the existing MakeCurrent
12646 submission into once that will send MakeCurrent, MakeContextCurrent or
12647 MakeCurrentReadSGI, depending on the request and the host server's
12648 capabilities, and add decoding/hijack support for incoming
12649 MakeContextCurrent and MakeCurrentReadSGI requests.
12650
12651 Fixes Clutter → Xephyr → VirtualBox.
12652
12653 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12654 Reviewed-by: Adam Jackson <ajax@redhat.com>
12655 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
12656
12657commit 4dd5989d15465f3f3480b521d4e36673972fa24a
12658Merge: 0b02150 506e343
12659Author: Keith Packard <keithp@keithp.com>
12660Date: Mon Sep 24 11:43:01 2012 -0700
12661
12662 Merge remote-tracking branch 'ajax/server-1.14-abi-churn'
12663
12664commit 0b02150c27e98f996e10d7489f9f67a30e4e3497
12665Author: Yufeng Shen <miletus@chromium.org>
12666Date: Mon Sep 24 14:03:31 2012 -0400
12667
12668 dix: fix scale_to_desktop for edge ABS events
12669
12670 Scale_to_desktop() converts ABS events from device coordinates
12671 to screen coordinates:
12672 [dev_X_min, dev_X_max] -> [screen_X_min, screen_X_max]
12673 [dev_Y_min, dev_Y_max] -> [screen_Y_min, screen_Y_max]
12674
12675 An edge ABS event with X = dev_X_max (e.g., generated from the
12676 edge of a touchscreen) will be converted to have screen X value
12677 = screen_X_max, which, however, will be filterd out when xserver
12678 tries to find proper Window to receive the event, because the
12679 range check for a Window to receive events is
12680 window_X_min <= event_screen_X < window_X_max
12681 Events with event_screen_X = screen_X_max will fail the test get
12682 and rejected by the Window.
12683
12684 To fix this, we change the device to screen coordinates mapping to
12685 [dev_X_min, dev_X_max] -> [screen_X_min, screen_X_max-1]
12686 [dev_Y_min, dev_Y_max] -> [screen_Y_min, screen_Y_max-1]
12687
12688 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
12689 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
12690 Signed-off-by: Yufeng Shen <miletus@chromium.org>
12691 Signed-off-by: Keith Packard <keithp@keithp.com>
12692
12693commit 7722bcbab2507d263c7685b15cccbfdd52fc3a24
12694Merge: 4dc2a76 f54987d
12695Author: Keith Packard <keithp@keithp.com>
12696Date: Sun Sep 23 16:04:13 2012 -0700
12697
12698 Merge remote-tracking branch 'jeremyhu/master'
12699
12700commit 4dc2a76740d921c824a4d8193f39dd373475f02a
12701Author: Adam Jackson <ajax@redhat.com>
12702Date: Thu Sep 20 17:57:00 2012 -0400
12703
12704 miext/damage: Only wrap into the GC ops chain if there's a listener (v3)
12705
12706 before after Operation
12707 -------- ----------------- -----------------
12708 1148346.9 1191807.5 ( 1.04) PutImage 10x10 square
12709 2091666.1 2180983.0 ( 1.04) ShmPutImage 10x10 square
12710
12711 v3: In miDamage{R,Unr}egister, bump the serial number of the affected
12712 drawable (and all children if it's a window) so subsequent drawing
12713 against the damage will trigger another ValidateGC pass and we wrap
12714 in/out correctly. Spotted by Aaron Plattner.
12715
12716 Signed-off-by: Adam Jackson <ajax@redhat.com>
12717 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
12718 Signed-off-by: Keith Packard <keithp@keithp.com>
12719
12720commit ad0156c369a40762df6b70780358ce14e5f2c9b1
12721Author: Adam Jackson <ajax@redhat.com>
12722Date: Thu Sep 20 11:16:26 2012 -0400
12723
12724 dix: Remove MapUnmapEventsEnabled and friends
12725
12726 This hack was added to suppress events generated by Composite's internal
12727 unmap/map cycle on redirection state change. Since that cycle was
12728 removed in 193ecc8b4, these can go.
12729
12730 Signed-off-by: Adam Jackson <ajax@redhat.com>
12731 Reviewed-by: Ville Syrjälä <syrjala@sci.fi>
12732 Signed-off-by: Keith Packard <keithp@keithp.com>
12733
12734commit 387b1ac33ca63d27e42d4f61cf248bafb7c43c17
12735Author: Adam Jackson <ajax@redhat.com>
12736Date: Thu Sep 20 11:16:25 2012 -0400
12737
12738 dix: Factor out DeliverUnmapNotify
12739
12740 Signed-off-by: Adam Jackson <ajax@redhat.com>
12741 Reviewed-by: Keith Packard <keithp@keithp.com>
12742 Signed-off-by: Keith Packard <keithp@keithp.com>
12743
12744commit d20cc0fca4e36ad9bdbda8950c4466b1976e70a1
12745Author: Adam Jackson <ajax@redhat.com>
12746Date: Thu Sep 20 11:16:24 2012 -0400
12747
12748 dix: Factor out DeliverMapNotify
12749
12750 Signed-off-by: Adam Jackson <ajax@redhat.com>
12751 Reviewed-by: Keith Packard <keithp@keithp.com>
12752 Signed-off-by: Keith Packard <keithp@keithp.com>
12753
12754commit 63843cb70030812bb3a311b2e8dfb0d1b0a75176
12755Author: Adam Jackson <ajax@redhat.com>
12756Date: Thu Sep 20 11:16:23 2012 -0400
12757
12758 dix: Factor out MaybeDeliverMapRequest
12759
12760 Signed-off-by: Adam Jackson <ajax@redhat.com>
12761 Reviewed-by: Keith Packard <keithp@keithp.com>
12762 Signed-off-by: Keith Packard <keithp@keithp.com>
12763
12764commit dab7a1ec7fd97e07405e0b01dc1f1ec6e5fc2e7c
12765Author: Adam Jackson <ajax@redhat.com>
12766Date: Thu Sep 20 11:16:22 2012 -0400
12767
12768 dix: Fix some indentation
12769
12770 Signed-off-by: Adam Jackson <ajax@redhat.com>
12771 Reviewed-by: Keith Packard <keithp@keithp.com>
12772 Signed-off-by: Keith Packard <keithp@keithp.com>
12773
12774commit 3d1051aecbb1955084804133cacd12c7f696833a
12775Author: Peter Hutterer <peter.hutterer@who-t.net>
12776Date: Thu Sep 20 05:56:39 2012 +1000
12777
12778 dix: set the device transformation matrix
12779
12780 The property handler is registered after setting the property, so
12781 dev->transform remains as all-zeros. That causes pixman_f_transform_invert()
12782 to fail (in transformAbsolute()) and invert remains as garbage. This
12783 may then cause a cursor jump to 0,0.
12784
12785 Since the axes are not yet initialized here and we need to allow for drivers
12786 changing the matrix, we cannot use the property handler for matrix
12787 initialization, essentially duplicating the code.
12788
12789 Triggered by the fix to (#49347) in 749a593e49adccdf1225be28a521412ec85333f4
12790
12791 https://bugzilla.redhat.com/show_bug.cgi?id=852841
12792
12793 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12794 Reviewed-by: Chase Douglas <chase.douglas@ubuntu.com>
12795 Signed-off-by: Keith Packard <keithp@keithp.com>
12796
12797commit 1cb0261ef54b7dd6a7ef84e1c3959e424706228b
12798Author: Daniel Martin <consume.noise@gmail.com>
12799Date: Thu Sep 6 00:38:26 2012 +0200
12800
12801 dix: Delete mibstore.h
12802
12803 Since Nov 2010 (commit c4c4676) the only purpose of mibstore.h was to
12804 define an empty function (miInitializeBackingStore()) for backward
12805 compatibility. Time to say goodbye.
12806
12807 Reviewed-by: Adam Jackson <ajax@redhat.com>
12808 Signed-off-by: Keith Packard <keithp@keithp.com>
12809
12810commit 08a9ed2524679ed939fa0ad466ee3faf905b8ac4
12811Author: Daniel Martin <consume.noise@gmail.com>
12812Date: Thu Sep 6 00:38:25 2012 +0200
12813
12814 dix: Remove refs to mi backing store from docs
12815
12816 Remove any reference to mibstore.h and miInitializeBackingStore() from
12817 the documentation.
12818
12819 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
12820 Reviewed-by: Adam Jackson <ajax@redhat.com>
12821 Signed-off-by: Keith Packard <keithp@keithp.com>
12822
12823commit 76d8739b1f4ed07b874ae56a492d1e0aa0fd9b04
12824Author: Daniel Martin <consume.noise@gmail.com>
12825Date: Thu Sep 6 00:38:24 2012 +0200
12826
12827 dix: Remove #includes of mibstore.h
12828
12829 Remove more backing store leftovers.
12830
12831 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
12832 Reviewed-by: Adam Jackson <ajax@redhat.com>
12833 Signed-off-by: Keith Packard <keithp@keithp.com>
12834
12835commit 056fd15deaa85d3ed936df9a3b2e0cdcfa599da8
12836Author: Daniel Martin <consume.noise@gmail.com>
12837Date: Thu Sep 6 00:38:23 2012 +0200
12838
12839 dix: Delete mibstore.c
12840
12841 The only reference to mibstore.c was removed with commit c4c4676. Remove the
12842 file itself too.
12843
12844 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
12845 Reviewed-by: Adam Jackson <ajax@redhat.com>
12846 Signed-off-by: Keith Packard <keithp@keithp.com>
12847
12848commit f54987de97720200ee94eba1c7a737d4ad8c55c8
12849Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
12850Date: Thu Sep 20 21:11:21 2012 -0700
12851
12852 XQuartz: Add some verbose logging to debug xp_lock_window being unbalanced
12853
12854 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
12855
12856commit 25d26875bc9bd6fd23ae1b5280f015abf1b033b7
12857Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
12858Date: Thu Sep 20 21:49:40 2012 -0700
12859
12860 XQuartz: Avoid a possible deadlock with DRI on OS X 10.7.5 and OS X 10.8.2
12861
12862 <rdar://problem/12338921>
12863 http://bugs.winehq.org/show_bug.cgi?id=31751
12864
12865 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
12866
12867commit 506e3437c73e5ae935ff7c056d7808fbb0c7e614
12868Author: Adam Jackson <ajax@redhat.com>
12869Date: Thu Sep 20 14:22:48 2012 -0400
12870
12871 dix: Fix types in WindowOptRec
12872
12873 No reason for these to be 64 bits on LP64.
12874
12875 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12876 Signed-off-by: Adam Jackson <ajax@redhat.com>
12877
12878commit 092c57ab173c8b71056f6feb3b9d04d063a46579
12879Author: Adam Jackson <ajax@redhat.com>
12880Date: Fri Jun 17 14:03:01 2011 -0400
12881
12882 os: Hide the Connection{In,Out}put implementation details
12883
12884 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12885 Signed-off-by: Adam Jackson <ajax@redhat.com>
12886
12887commit e2c7d70e5ddb8b17676a13ceebfbb87d14d63243
12888Author: Adam Jackson <ajax@redhat.com>
12889Date: Fri Jun 17 13:43:38 2011 -0400
12890
12891 dix: Extend initial connection handshake for forwarding proxies
12892
12893 Forwarding proxies like sshd will appear to be local, even though they
12894 aren't really. This leads to weird behaviour for extensions that truly
12895 require running under the same OS services as the client, like MIT-SHM
12896 and DRI2.
12897
12898 Add two new legal values for the initial connection's byteOrder field,
12899 'r' and 'R'. These act like 'l' and 'B' respectively, but have the side
12900 effect of forcing the client to be treated as non-local. Forwarding
12901 proxies should attempt to munge the first packet of the connection
12902 accordingly; older servers will reject connections thusly munged, so the
12903 proxy should fall back to passthrough if the munged connection attempt
12904 fails.
12905
12906 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12907 Signed-off-by: Adam Jackson <ajax@redhat.com>
12908
12909commit 31bf81772e146af79b0c456aae2159eba8b0280f
12910Author: Adam Jackson <ajax@redhat.com>
12911Date: Thu Sep 20 14:00:09 2012 -0400
12912
12913 dix: Repack ClientRec
12914
12915 Pick smaller types where possible, including bitfielding some Bools and
12916 small enums, then shuffle the result to be hole-free. 192 -> 128 bytes
12917 on LP64, 144 -> 96 bytes on ILP32.
12918
12919 Signed-off-by: Adam Jackson <ajax@redhat.com>
12920
12921commit ff8e3ad8074cd2c8bed49b39c40c2b4892118270
12922Author: Adam Jackson <ajax@redhat.com>
12923Date: Thu Sep 20 13:16:59 2012 -0400
12924
12925 dix: Pull client-is-local flag up to the ClientRec
12926
12927 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12928 Signed-off-by: Adam Jackson <ajax@redhat.com>
12929
12930commit 3f7bc222638d5d38324ecbc8c2c4e39af17d110e
12931Author: Adam Jackson <ajax@redhat.com>
12932Date: Thu Jun 16 17:40:24 2011 -0400
12933
12934 os: Repack ConnectionOutput for LP64
12935
12936 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
12937 Signed-off-by: Adam Jackson <ajax@redhat.com>
12938
12939commit d01921ec18c21f21d377b60626cc2d3418b84a7c
12940Merge: 70e5766 245e7e0
12941Author: Keith Packard <keithp@keithp.com>
12942Date: Thu Sep 20 17:26:48 2012 +0200
12943
12944 Merge remote-tracking branch 'ajax/ioperm'
12945
12946commit 5109c7f6581b3e4e23700bfd03b4daf0ba97630e
12947Author: Adam Jackson <ajax@redhat.com>
12948Date: Thu Sep 20 11:19:39 2012 -0400
12949
12950 xfree86: Bump video ABI to 14
12951
12952 Signed-off-by: Adam Jackson <ajax@redhat.com>
12953
12954commit 245e7e0361b18766583ae391a2ac1231bb1a1f84
12955Author: Adam Jackson <ajax@redhat.com>
12956Date: Tue Jun 26 14:32:31 2012 -0400
12957
12958 xfree86: Change the semantics of driverFunc(GET_REQUIRED_HW_INTERFACES)
12959
12960 This is a really awkward interface, since we're calling it well before
12961 the driver knows what device it's going to drive. Drivers with both KMS
12962 and UMS support therefore don't know whether to say they need I/O port
12963 access or not, and have to assume they do.
12964
12965 With this change we now call it only to query whether port access might
12966 be needed; we don't use that to determine whether to call a driver's
12967 probe function or not, instead we call them unconditionally. If the
12968 driver doesn't check whether port access was enabled, they might crash
12969 ungracefully. To accomodate this, we move xorgHWAccess to be explicitly
12970 intentionally exported (sigh xf86Priv.h) so that drivers can check that
12971 before they attempt port access.
12972
12973 v2: Move initial xf86EnableIO() nearer the logic that determines whether
12974 to call it, suggested by Simon Farnsworth.
12975
12976 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
12977 Reviewed-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
12978 Signed-off-by: Adam Jackson <ajax@redhat.com>
12979
12980commit d88fb00d791c2b19cf9dd244276838aba3a6b442
12981Author: Adam Jackson <ajax@redhat.com>
12982Date: Tue Jun 26 13:15:45 2012 -0400
12983
12984 linux: Make failure to iopl non-fatal
12985
12986 We load the driver list, then enable I/O, then call driver probe based
12987 on whether I/O enable succeeded. That's bad, because the loaded
12988 security policy might forbid port access. We happen to treat that as
12989 fatal for some reason, which means even drivers that don't need I/O
12990 access (like kms and fbdev) don't get the chance to run. Facepalm.
12991
12992 How about we just make that non-fatal instead, that sounds like a much
12993 better plan.
12994
12995 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
12996 Reviewed-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
12997 Signed-off-by: Adam Jackson <ajax@redhat.com>
12998
12999commit 048674a6aeb61149a1b5f6b0bc3762ddf57f38ee
13000Author: Adam Jackson <ajax@redhat.com>
13001Date: Tue Jun 26 13:12:45 2012 -0400
13002
13003 linux: Refactor xf86{En,Dis}ableIO
13004
13005 Pull platform methods into their own sections for legibility, and
13006 rewrite the ifdefs to be more concise.
13007
13008 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
13009 Reviewed-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
13010 Signed-off-by: Adam Jackson <ajax@redhat.com>
13011
13012commit 70e5766874a919039678bb2ed75f2ccea0cb4345
13013Author: Dave Airlie <airlied@redhat.com>
13014Date: Mon Aug 27 15:22:44 2012 +1000
13015
13016 xf86: fix multi-seat video device support. (v2)
13017
13018 If we are not seat 0 the following apply:
13019
13020 don't probe any bus other than platform
13021 don't probe any drivers other than platform
13022 assume the first platform device we match on the bus is the primary GPU.
13023
13024 This just adds checks in the correct places to ensure this, and
13025 with this X can now start on a secondary seat for an output device.
13026
13027 v2: fix Seat0 macros
13028 Reviewed-by: Keith Packard <keithp@keithp.com>
13029 Signed-off-by: Dave Airlie <airlied@redhat.com>
13030
13031commit 37d956e3ac9513b74078882dff489f9b0a7a5a28
13032Author: Keith Packard <keithp@keithp.com>
13033Date: Mon Sep 10 11:14:20 2012 +1000
13034
13035 xf86: fix compat output selection for no output GPUs
13036
13037 This should work properly with dynamic outputs.
13038
13039 Reviewed-by: Dave Airlie <airlied@redhat.com>
13040 Signed-off-by: Dave Airlie <airlied@redhat.com>
13041
13042commit e0b92f5f86f8ce71ab4c80e9902e51e2525f9705
13043Author: Dave Airlie <airlied@gmail.com>
13044Date: Thu Sep 13 18:44:06 2012 +1000
13045
13046 config/udev: ignore change on drm devices
13047
13048 for input devices we handle change like remove/add, but for
13049 drm devices we get change events when we hotplug outputs,
13050 so lets just ignore change at this level, and let the drivers
13051 handle it. We may in the future want to route driver udev
13052 from here instead.
13053
13054 Reported-by: Timo Aaltonen <timo.aaltonen@canonical.com>
13055 Reviewed-by: Keith Packard <keithp@keithp.com>
13056 Signed-off-by: Dave Airlie <airlied@redhat.com>
13057
13058commit 10672a5abe7d4e924844d7e41c86c24d6367a52b
13059Author: Dave Airlie <airlied@redhat.com>
13060Date: Mon Sep 10 14:48:37 2012 +1000
13061
13062 xf86/platform: scan pci after probing devices
13063
13064 This solves a race if we are trying to dynamically power off
13065 secondary GPUs. Its not the greatest fix ever but it probably
13066 as good as we can do for now.
13067
13068 The GPU probing causes the devices to be powered up, then when
13069 we scan the PCI bus we get the correct information from the kernel,
13070 rather than a bunch of 0xff due to the device being powered off.
13071
13072 drop gratuitous '&'.
13073
13074 Reviewed-by: Keith Packard <keithp@keithp.com>
13075 Signed-off-by: Dave Airlie <airlied@redhat.com>
13076
13077commit 22746df15b5f75dc85f5cf5b73932eb8a44cb89b
13078Author: Dave Airlie <airlied@redhat.com>
13079Date: Thu Sep 6 16:33:54 2012 +1000
13080
13081 dri2: invalidate drawable after sharing pixmap
13082
13083 After we share the pixmap, the backing storage may have changed,
13084 and we need to invalidate and buffers pointing at it.
13085
13086 This fixes GL compositors and prime windows lacking contents initially.
13087
13088 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
13089 Signed-off-by: Dave Airlie <airlied@redhat.com>
13090
13091commit f0bad69edd57facd6cffde8cb0863d1a735e2492
13092Author: Keith Packard <keithp@keithp.com>
13093Date: Wed Sep 5 14:45:08 2012 -0700
13094
13095 Version bumped to 1.13
13096
13097 Signed-off-by: Keith Packard <keithp@keithp.com>
13098
13099commit 856f80c8d7f22b979c72d9c70b70187df6004a03
13100Merge: 0db936a b8ab93d
13101Author: Keith Packard <keithp@keithp.com>
13102Date: Wed Sep 5 11:02:58 2012 -0700
13103
13104 Merge remote-tracking branch 'jeremyhu/master'
13105
13106commit 0db936a5b7125d9b73633733cb34e5e4b72f01e8
13107Author: Dave Airlie <airlied@redhat.com>
13108Date: Mon Sep 3 15:09:36 2012 +1000
13109
13110 xf86: call enter/leave VT for gpu screens as well
13111
13112 Otherwise we can't do fast user switch properly for multiple GPUs.
13113
13114 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
13115 Signed-off-by: Dave Airlie <airlied@redhat.com>
13116
13117commit 18deac48db0d2387242461025904eb28a0f71fac
13118Author: Dave Airlie <airlied@redhat.com>
13119Date: Mon Aug 27 15:17:21 2012 +1000
13120
13121 config/udev: respect seat for hotplugged video devices.
13122
13123 This respects the seat tag for hotplugged video devices at X start.
13124
13125 Reviewed-by: Keith Packard <keithp@keithp.com>
13126 Signed-off-by: Dave Airlie <airlied@redhat.com>
13127
13128commit ec740cffeba59216f4293fba829097654cde3b11
13129Author: Dave Airlie <airlied@redhat.com>
13130Date: Mon Aug 27 15:15:19 2012 +1000
13131
13132 config/udev: add wrapper around check if server is not seat 0
13133
13134 this is a simple clean-up that is useful to stop further propogation
13135 of this construct.
13136
13137 Reviewed-by: Keith Packard <keithp@keithp.com>
13138 Signed-off-by: Dave Airlie <airlied@redhat.com>
13139
13140commit 49ec57d5094be0dd2b67435ac7bb04ead780f36d
13141Author: Dave Airlie <airlied@redhat.com>
13142Date: Fri Aug 24 13:23:01 2012 +1000
13143
13144 dix: free default colormap before screen deletion
13145
13146 If we don't free this here, it gets freed later in the resource
13147 cleanups, however it then looks up up pmap->pScreen, which we
13148 freed already in this function. So free the default colormap
13149 when we should.
13150
13151 This fixes a bug after a couple of hotplug cycles when you try
13152 to exit the X server and it crashes.
13153
13154 Reviewed-by: Keith Packard <keithp@keithp.com>
13155 Signed-off-by: Dave Airlie <airlied@redhat.com>
13156
13157commit 20f601a0fbc01a4f8ab85e54ee45f78f7817cdb5
13158Author: Dave Airlie <airlied@redhat.com>
13159Date: Fri Aug 24 12:56:18 2012 +1000
13160
13161 xf86/crtc: don't free config->name
13162
13163 This is set by pre_init not screen init, so if we free it here
13164 and then recycle the server, we lose all the providers.
13165
13166 I think we need to wrap FreeScreen here to do this properly,
13167 will investigate for 1.14 most likely, safer to just leak this
13168 on server exit for now.
13169
13170 Reviewed-by: Keith Packard <keithp@keithp.com>
13171 Signed-off-by: Dave Airlie <airlied@redhat.com>
13172
13173commit b8ab93dfbc7f292b5bfe7e9113e1af824ccbd1a8
13174Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13175Date: Tue Aug 28 10:06:51 2012 -0700
13176
13177 list: Use offsetof() and typeof() to determine member offsets within a structure
13178
13179 Some compilers have difficulty with the previous implementation which
13180 relies on undefined behavior according to the C standard. Using
13181 offsetof() from <stddef.h> (which most likely just uses
13182 __builtin_offsetof on modern compilers) allows us to accomplish this
13183 without ambiguity.
13184
13185 This fix also requires support for typeof(). If your compiler does not
13186 support typeof(), then the old implementation will be used. If you see
13187 failures in test/list, please try a more modern compiler.
13188
13189 v2: Added fallback if typeof() is not present.
13190
13191 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13192 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
13193
13194commit 148583d62b84832369e9df39b3e527b99ca96761
13195Author: Peter Hutterer <peter.hutterer@who-t.net>
13196Date: Wed Aug 29 09:11:35 2012 +1000
13197
13198 tests: move GCC diagnostics pragma outside of function
13199
13200 This is a a gcc 4.6+ feature.
13201
13202 signal-logging.c:210: error: #pragma GCC diagnostic not allowed inside
13203 functions
13204
13205 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13206 Reviewed-by: Dave Airlie <airlied@redhat.com>
13207 Signed-off-by: Keith Packard <keithp@keithp.com>
13208
13209commit c75c947b6e9bc725821b28835f3667c4aabef9ee
13210Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13211Date: Tue Aug 28 12:43:55 2012 -0700
13212
13213 test/list: Fix test_xorg_list_del test
13214
13215 We never use child[2], so it's state is undefined.
13216
13217 This issue seems to have existed since the test was first
13218 written: 92788e677be79bd04e5ef140f4ced50ad8b1bf8e
13219
13220 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13221 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
13222
13223commit 486038fe16957b39d39842365c5409c494c9a8df
13224Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13225Date: Mon Aug 27 22:35:51 2012 -0700
13226
13227 XQuartz: Bump Info.plist version to 2.7.4
13228
13229 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13230
13231commit b90b3f6eac34616c46ab693fafbf6b73a04f407b
13232Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13233Date: Mon Aug 27 17:15:19 2012 -0700
13234
13235 test: Make os test more compliant
13236
13237 sighandler_t is not UNIX.
13238
13239 Regression from: 7f09126e068015db54c56bb982b8f91065375700
13240
13241 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13242 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
13243
13244commit a557edca6152ad09c685a00e01534c35a1f00f2e
13245Merge: 6619f5c 24ffcfc
13246Author: Keith Packard <keithp@keithp.com>
13247Date: Mon Aug 27 08:06:09 2012 -0700
13248
13249 Merge remote-tracking branch 'whot/for-keith'
13250
13251commit 24ffcfcded6b4b024958801e8a6cecad36d9a3e3
13252Author: Peter Hutterer <peter.hutterer@who-t.net>
13253Date: Mon Aug 20 10:28:26 2012 +1000
13254
13255 os: fix typo, fsync when WIN32 is _not_ defined
13256
13257 Introduced in 164b38c72fe9c69d13ea4f9c46d4ccc46566d826
13258
13259 Reported-by: Jon TURNEY <jon.turney@dronecode.org.uk>
13260 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13261 Reviewed-by: Keith Packard <keithp@keithp.com>
13262
13263commit d53e6e02a2595ced1882f5fcd34d08ea039b3b85
13264Author: Peter Hutterer <peter.hutterer@who-t.net>
13265Date: Thu Aug 16 13:54:42 2012 +1000
13266
13267 mi: don't check for core events in miPointerSetPosition (#53568)
13268
13269 As of 81cfe44b1ed0de84ad1941fe2ca74bebef3fc58d, miPointerSetPosition now
13270 returns the screen pointer of the device. This broke floating slave devices,
13271 as soon as a motion event was submitted, miPointerSetPosition returned NULL,
13272 crashing the server.
13273
13274 dev->coreEvents is only false if the device is a floating slave, in which
13275 case it has a sprite.
13276
13277 X.Org Bug 53568 <http://bugs.freedesktop.org/show_bug.cgi?id=53568>
13278
13279 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13280 Reviewed-by: Keith Packard <keithp@keithp.com>
13281
13282commit 6619f5c0e1086b57888ff7146e8ed5897b50d440
13283Author: Keith Packard <keithp@keithp.com>
13284Date: Tue Aug 21 13:46:34 2012 -0700
13285
13286 Relase 1.12.99.905
13287
13288 RC with updated ABI version numbers.
13289
13290 Signed-off-by: Keith Packard <keithp@keithp.com>
13291
13292commit bafbd99080be49a17be97d2cc758fbe623369945
13293Author: Peter Hutterer <peter.hutterer@who-t.net>
13294Date: Wed Aug 8 11:34:32 2012 +1000
13295
13296 dix: work around scaling issues during WarpPointer (#53037)
13297
13298 In WarpPointer calls, we get input in screen coordinates. They must be
13299 scaled to device coordinates, and then back to screen coordinates for screen
13300 crossing and root coordinates in events.
13301
13302 The rounding errors introduced (and clipping in core/XI 1.x events) can lead
13303 to the actual position being different to the requested input coordinates.
13304 e.g. 200 scales to 199.9999, truncated to 199 in the event.
13305
13306 Avoid this by simply overwriting the scaled screen coordinates with the
13307 input coordinates for the POINTER_SCREEN case.
13308
13309 X.Org Bug 53037 <http://bugs.freedesktop.org/show_bug.cgi?id=53037>
13310
13311 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13312 Reviewed-by: Keith Packard <keithp@keithp.com>
13313
13314commit 1ebba43052d68d874148e63c9ae38489ddfc5ec1
13315Author: Peter Hutterer <peter.hutterer@who-t.net>
13316Date: Wed Aug 15 14:49:04 2012 +1000
13317
13318 os: don't block signal-unsafe logging, merely warn about it.
13319
13320 Throw an error into the log file, but continue anyway. And after three
13321 warnings, stop complaining. Not all input drivers will be fixed in time (or
13322 ever) and our printf implementation is vastly inferior, so there is still a
13323 use-case for non-sigsafe logging.
13324
13325 This also adds more linebreaks to the message.
13326
13327 CC: Chase Douglas <chase.douglas@canonical.com>
13328 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13329 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
13330
13331commit 4912b4adb666dad96b832ab2d7caaae49808723e
13332Author: Peter Hutterer <peter.hutterer@who-t.net>
13333Date: Mon Aug 13 14:44:44 2012 +1000
13334
13335 os: add support for %d and %i to pnprintf
13336
13337 The mouse driver uses %i in some debug messages
13338
13339 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13340 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
13341
13342commit 7f8c39c8b5ef89153ecd84d16331e96d8feb18ef
13343Author: Peter Hutterer <peter.hutterer@who-t.net>
13344Date: Mon Aug 13 14:24:36 2012 +1000
13345
13346 Add FormatInt64 to convert signed integers in signal-safe manner
13347
13348 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13349 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
13350
13351commit 36c1d92ec0ef0f3927034a12d4cb79dcc22bd185
13352Author: Peter Hutterer <peter.hutterer@who-t.net>
13353Date: Mon Aug 13 12:24:39 2012 +1000
13354
13355 test: add a few tests for signal-safe logging
13356
13357 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13358 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
13359
13360commit b69536b475118a8787c1726355b504207bf83f8f
13361Author: Peter Hutterer <peter.hutterer@who-t.net>
13362Date: Mon Aug 13 11:15:54 2012 +1000
13363
13364 test: assert from signal-safe number conversion
13365
13366 Throw an assert when the conversion fails instead of just returning. Asserts
13367 are more informative.
13368
13369 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13370 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
13371
13372commit 3e091e1075dc29120e54ab2ef110b04ce7383ffd
13373Author: Aaron Plattner <aplattner@nvidia.com>
13374Date: Fri Aug 17 11:26:05 2012 -0700
13375
13376 xfree86: Bump extension ABI to 7.0
13377
13378 Commit 9d457f9c55f12106ba44c1c9db59d14f978f0ae8 added an array of
13379 DevPrivateSetRec structures in the middle of the ScreenRec, which throws off
13380 extension modules trying to call things like pScreen->DestroyPixmap.
13381
13382 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
13383 Signed-off-by: Keith Packard <keithp@keithp.com>
13384
13385commit ac616d8ed5a634c6b32775eb7b071a13c575fd97
13386Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13387Date: Thu Aug 16 19:09:42 2012 -0700
13388
13389 XQuartz: Use asl_log_descriptor for children as well
13390
13391 This change is #if'd out due to a bug in asl_log_descriptor, but
13392 it is left here as reference to be enabled in the future.
13393
13394 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13395
13396commit ad123071e54d890d36d1373f59b8a6eb4937e936
13397Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13398Date: Thu Aug 16 13:02:28 2012 -0700
13399
13400 XQuartz: Use asl_log_descriptor on Mountain Lion
13401
13402 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13403
13404commit 7c3d2e4828deb4e8ec38e8ef88d6f92b2d931033
13405Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13406Date: Thu Aug 16 19:42:54 2012 -0700
13407
13408 XQuartz: console_redirect: Properly zero-out the tail of the array on realloc()
13409
13410 We forgot to multiply by sizeof(), so it wasn't fully zeroed out.
13411
13412 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13413
13414commit a32e01802ff1c938c0afe0cc007a273b9ada8610
13415Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13416Date: Thu Aug 16 15:43:34 2012 -0700
13417
13418 XQuartz: console_redirect: Set the correct location for reading into the buffer
13419
13420 Prior to this change, it was possible that a large message would have some
13421 of its data prepended to subsequent messages due to our not incorrectly
13422 setting the location to write into the buffer.
13423
13424 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13425
13426commit ad5fe2d9614959b68bf71e23abf7e5abac9c2734
13427Merge: 16d8da5 99b94af
13428Author: Keith Packard <keithp@keithp.com>
13429Date: Wed Aug 15 13:29:17 2012 -0700
13430
13431 Merge remote-tracking branch 'jeremyhu/master'
13432
13433commit 16d8da5ca99211bd180f532d78d7c5943d38c8ad
13434Author: Paul Berry <stereotype441@gmail.com>
13435Date: Tue Jul 31 15:15:56 2012 -0700
13436
13437 glx: Skip multisampled configs when matching pre-existing X visuals.
13438
13439 In __glXScreenInit() we generate the set of GLX visuals in two steps:
13440 first we match each pre-existing X visual with a corresponding
13441 FBConfig, then we generate a new X visual to correspond to all the
13442 remaining FBConfigs.
13443
13444 The first step is used for the two default 24-bit visuals (true color
13445 and direct color) and for the 32-bit visual. If windowsystem
13446 multisampling is enabled in Mesa, we need to ensure that none of these
13447 three visuals gets matched to a multisampled config.
13448
13449 Fixes a bug with windowsystem multisampling in gnome-shell. If the X
13450 server happens to match up a multisampled FBConfig to the 32-bit
13451 visual, gnome-shell will try to use it to read pixels from
13452 alpha-blended windows (such as gnome-terminal), resulting in no window
13453 appearing on screen.
13454
13455 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
13456 Reviewed-by: Chad Versace <chad.versace@linux.intel.com>
13457 Signed-off-by: Keith Packard <keithp@keithp.com>
13458
13459commit 99b94af1ffbef7047f88e7d6c594242d172bec02
13460Author: Jeremy Huddleston <jeremyhu@apple.com>
13461Date: Wed Aug 15 10:28:16 2012 -0700
13462
13463 XQuartz: Fix build regression for GlxExtensionInit
13464
13465 quartz.c:153:6: error: use of undeclared identifier 'GlxExtensionInit'; did you mean 'GEExtensionInit'? [Semantic Issue]
13466 {GlxExtensionInit, "GLX", &noGlxExtension},
13467
13468 Regression-from: aad428b8e21c77397c623b78706eb64b1fea77c9
13469 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
13470
13471commit c0540b4c8db676d18a9a09828f1590b0b16f9ebc
13472Author: Keith Packard <keithp@keithp.com>
13473Date: Tue Aug 7 17:49:48 2012 -0700
13474
13475 Kludge -- Call RandR screen before cleaning up xf86 crtcs
13476
13477 The core RandR screen cleanup now involves cleaning up any GPU screen
13478 associations, and those call down into DDX to clean up the driver. If
13479 the pointers from the xf86 structures back to the core randr
13480 structures are set to NULL at that point, bad things happen.
13481
13482 This patch "knows" that the core RandR close screen is underneath the
13483 xf86 randr close screen function, and so makes sure it gets called
13484 first.
13485
13486 Signed-off-by: Keith Packard <keithp@keithp.com>
13487 Reviewed-by: Dave Airlie <airlied@redhat.com>
13488
13489commit 288b87e42cbb5db34cda848143655e84af8455da
13490Author: Keith Packard <keithp@keithp.com>
13491Date: Tue Aug 7 17:49:47 2012 -0700
13492
13493 Close GPU screens before core screens
13494
13495 This should make cleaning up the GPU screens easier as the core
13496 screens they are associated with will still be around.
13497
13498 Signed-off-by: Keith Packard <keithp@keithp.com>
13499 Reviewed-by: Dave Airlie <airlied@redhat.com>
13500
13501commit 19772670e3041fe1c7f5bbf32fa039a6d8245ccb
13502Author: Keith Packard <keithp@keithp.com>
13503Date: Tue Aug 7 17:49:46 2012 -0700
13504
13505 Only free Render filter names on last screen close
13506
13507 Hotplugging screens causes the render filter names to get freed while
13508 still in use; wait for the last core screen to be closed before
13509 freeing them. That only happens at server reset, when we want them to
13510 be freed.
13511
13512 Signed-off-by: Keith Packard <keithp@keithp.com>
13513 Reviewed-by: Dave Airlie <airlied@redhat.com>
13514
13515commit 863d528a9f76d0e8f122aebf19f8564a4c67a938
13516Author: Simon Schubert <2@0x2c.org>
13517Date: Sun Aug 12 09:40:16 2012 -0700
13518
13519 fb: reorder Bresenham error correction to avoid overshoot.
13520
13521 When fbBresSolid draws a line, it can happen that after the last
13522 pixel, the Bresenham error term overflows, and fbBresSolid paints
13523 another pixel before adjusting the error term.
13524
13525 However, if this happens on the last pixel (len=0), this extra pixel
13526 might overshoot the boundary, and, in rare cases, lead to a segfault.
13527
13528 Fix this issue by adjusting for the Bresenham error term before
13529 drawing the main pixel, not after.
13530
13531 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=24274
13532 Signed-off-by: Simon Schubert <2@0x2c.or>
13533 Tested-by: Mitch Davis <mjd+freedesktop.org@afork.com>
13534 Signed-off-by: Matt Turner <mattst88@gmail.com>
13535 Signed-off-by: Keith Packard <keithp@keithp.com>
13536
13537commit c22c936b35cee60609041ccb7ee52c00b0634d7b
13538Author: Keith Packard <keithp@keithp.com>
13539Date: Tue Aug 7 16:04:25 2012 -0700
13540
13541 Release 1.12.99.904
13542
13543 Closing the non-critical bug window for 1.13
13544
13545 Signed-off-by: Keith Packard <keithp@keithp.com>
13546
13547commit f94ec283efc922106e7ea1d85771950c66ee8f3b
13548Author: Keith Packard <keithp@keithp.com>
13549Date: Tue Aug 7 16:18:21 2012 -0700
13550
13551 Distribute include/glx_extinit.h
13552
13553 This changed should have been in aad428b8e21c77397c623b78706eb64b1fea77c9
13554
13555 Signed-off-by: Keith Packard <keithp@keithp.com>
13556
13557commit 9ca67de8feab59fcf7d6e3fdec8d8a7cfbb21651
13558Author: Julien Cristau <jcristau@debian.org>
13559Date: Sat Jul 14 16:07:51 2012 +0200
13560
13561 Bump glproto requirement to 1.4.16
13562
13563 Signed-off-by: Julien Cristau <jcristau@debian.org>
13564 Reviewed-by: Dave Airlie <airlied@redhat.com>
13565 Signed-off-by: Keith Packard <keithp@keithp.com>
13566
13567commit 02f94b2d446f6700f791f318c551ac835af8445a
13568Merge: 360fa77 7f09126
13569Author: Keith Packard <keithp@keithp.com>
13570Date: Mon Aug 6 16:52:12 2012 -0700
13571
13572 Merge remote-tracking branch 'whot/for-keith'
13573
13574commit 360fa7736b214cae3ccd39feb4dc87b9e7e9734c
13575Merge: 2100e72 ac09a4a
13576Author: Keith Packard <keithp@keithp.com>
13577Date: Mon Aug 6 16:42:34 2012 -0700
13578
13579 Merge remote-tracking branch 'airlied/for-keithp'
13580
13581commit 2100e72388a3dc174093d80c9723f021b4614d17
13582Merge: 4a6f42d c37c650
13583Author: Keith Packard <keithp@keithp.com>
13584Date: Mon Aug 6 16:40:54 2012 -0700
13585
13586 Merge remote-tracking branch 'alanc/master'
13587
13588commit 7f09126e068015db54c56bb982b8f91065375700
13589Author: Peter Hutterer <peter.hutterer@who-t.net>
13590Date: Fri Aug 3 15:36:34 2012 +1000
13591
13592 os: don't unconditionally unblock SIGIO in OsReleaseSignals()
13593
13594 Calling OsReleaseSignal() inside the signal handler releases SIGIO, causing
13595 the signal handler to be called again from within the handler.
13596
13597 Practical use-case: when synaptics calls TimerSet in the signal handler,
13598 this causes the signals to be released, eventually hanging the server.
13599
13600 Regression introduced in 08962951de.
13601
13602 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13603 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
13604
13605commit cb306a8f174bec9ded95191b91797f59250e6808
13606Author: Peter Hutterer <peter.hutterer@who-t.net>
13607Date: Tue Jul 31 16:09:38 2012 +1000
13608
13609 dix: make sure the mask is set for emulated scroll events (#52508)
13610
13611 If a device has smooth scrolling axes, but submits scroll button events, we
13612 convert those to motion events and update the valuators. For legacy button
13613 events, the valuator mask is likely unset though, causing
13614 add_to_scroll_valuator() to return early, leaving us with an empty mask.
13615 That again skipped the rest of the code and no events were generated.
13616
13617 Fix it by making sure that the scroll valuator in the mask is at least
13618 initialized to 0.
13619
13620 Broke evdev wheel emulation, introduced by
13621 54476b5e4461ff523e935961affabcf0de12c556.
13622
13623 X.Org Bug 52508 <http://bugs.freedesktop.org/show_bug.cgi?id=52508>
13624
13625 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13626 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
13627
13628commit 4a6f42dda00ba3b5616f8a86f0d4c9a652c7d9d4
13629Author: Adam Jackson <ajax@redhat.com>
13630Date: Mon Jul 23 16:34:28 2012 -0400
13631
13632 sync: Fix logic error from b55bf248581dc66321b24b29f199f6dc8d02db1b
13633
13634 That commit adds two hunks, and I _think_ they're backwards. It adds
13635 code to modify bracket_greater on NegativeTransition triggers, and
13636 bracket_less on PositiveTransition triggers. That breaks symmetry with
13637 the surrounding code; the code as of this commit could probably be
13638 simplified further.
13639
13640 I can't keep the sync trigger rules in my head for more than about five
13641 minutes at a time, so I'm sending this on for more eyes. RHEL 6.3's
13642 xserver is shipping with b55bf248 reverted:
13643
13644 https://bugzilla.redhat.com/show_bug.cgi?id=748704#c33
13645
13646 And there appear to be some upstream reports of the same issue:
13647
13648 https://bugzilla.gnome.org/show_bug.cgi?id=658955
13649
13650 So I'd like to get this sorted out.
13651
13652 Signed-off-by: Adam Jackson <ajax@redhat.com>
13653 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
13654 Signed-off-by: Keith Packard <keithp@keithp.com>
13655
13656commit 454d0e3a1bb14d7f2579ccb5e513cec5686160e7
13657Author: Dave Airlie <airlied@redhat.com>
13658Date: Thu Jul 26 12:56:51 2012 +1000
13659
13660 randr: fix xinerama output for output slaves
13661
13662 This fixes the xinerama geometry when output slaves are enabled.
13663
13664 Tested with xdpyinfo -ext XINERAMA before after slave added.
13665
13666 Signed-off-by: Dave Airlie <airlied@redhat.com>
13667 Signed-off-by: Keith Packard <keithp@keithp.com>
13668
13669commit ac09a4a091a43188241365152e1244fd8fd93ee7
13670Author: Dave Airlie <airlied@redhat.com>
13671Date: Fri Aug 3 16:43:22 2012 +1000
13672
13673 dri2: fix master pixmap free and reset pointer
13674
13675 These are two minor changes, one to reset the pointer to NULL,
13676 after freeing the pixmaps, one to make sure we use the right API for
13677 the master pixmap, though I doubt it'll ever really matter.
13678
13679 Signed-off-by: Dave Airlie <airlied@redhat.com>
13680 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
13681
13682commit 64623ef90db649e89345f71871affe7a9606f6ea
13683Author: Dave Airlie <airlied@redhat.com>
13684Date: Fri Aug 3 16:42:44 2012 +1000
13685
13686 dri2: free slave pixmap on app exit
13687
13688 When the drawable disappears we need to free the prime master/slave combos.
13689
13690 This fixes a leak after a prime app is run.
13691
13692 Signed-off-by: Dave Airlie <airlied@redhat.com>
13693 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
13694
13695commit 1a465fef9bc21142eecca3999f0761a3c0501a09
13696Author: Dave Airlie <airlied@redhat.com>
13697Date: Fri Aug 3 16:41:45 2012 +1000
13698
13699 pixmap: have slave pixmap take a reference on master pixmap
13700
13701 Since the free routines free the master pixmap then the slave, we should
13702 be taking a reference when we bind them together.
13703
13704 Fixes a use-after-free when resizing a primed gears.
13705
13706 Signed-off-by: Dave Airlie <airlied@redhat.com>
13707 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
13708
13709commit c37c65052f674cd58894ad0b9ec22928a62c624e
13710Author: Alan Coopersmith <alan.coopersmith@oracle.com>
13711Date: Mon Jul 16 21:12:06 2012 -0700
13712
13713 Make indentation of dix/tables.c much more consistent and readable
13714
13715 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13716 Acked-by: Daniel Stone <daniel@fooishbar.org>
13717
13718commit 9f7ef7f7f0566f6319d8328ce0a1e6d0fa866720
13719Author: Alan Coopersmith <alan.coopersmith@oracle.com>
13720Date: Mon Jul 16 20:53:00 2012 -0700
13721
13722 Fix up formatting of initializers for arrays of structs
13723
13724 The indenter seems to have gotten confused by initializing arrays of
13725 structs with the struct defined inline - for predefined structs it did
13726 a better job, so match that.
13727
13728 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13729
13730commit c7b7abfaa068042e396d19538215402cfbb4f1e4
13731Author: Alan Coopersmith <alan.coopersmith@oracle.com>
13732Date: Sat Jul 14 11:21:15 2012 -0700
13733
13734 RRModeCreate: plug memory leak of newModes if AddResource fails
13735
13736 Reported by parfait 1.0:
13737
13738 Error: Memory leak (CWE 401)
13739 Memory leak of pointer 'newModes' allocated with realloc(((char*)modes), ((num_modes + 1) * 8))
13740 at line 93 of randr/rrmode.c in function 'RRModeCreate'.
13741 pointer allocated at line 82 with realloc(((char*)modes), ((num_modes + 1) * 8)).
13742 Error: Memory leak (CWE 401)
13743 Memory leak of pointer 'newModes' allocated with malloc(8)
13744 at line 93 of randr/rrmode.c in function 'RRModeCreate'.
13745 pointer allocated at line 84 with malloc(8).
13746
13747 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13748 Reviewed-by: Keith Packard <keithp@keithp.com>
13749
13750commit 1eb7be863353be2cf3e83738253eb60c5fe49d19
13751Author: Alan Coopersmith <alan.coopersmith@oracle.com>
13752Date: Sat Jul 14 09:29:52 2012 -0700
13753
13754 rrproperty.c: free newly allocated prop in more error paths
13755
13756 Reported by parfait 1.0:
13757
13758 Error: Memory leak (CWE 401)
13759 Memory leak of pointer 'prop' allocated with RRCreateOutputProperty(property)
13760 at line 220 of randr/rrproperty.c in function 'RRChangeOutputProperty'.
13761 'prop' allocated at line 154 with RRCreateOutputProperty(property).
13762 prop leaks when pending != 0 at line 160.
13763 Error: Memory leak (CWE 401)
13764 Memory leak of pointer 'prop' allocated with RRCreateOutputProperty(property)
13765 at line 346 of randr/rrproperty.c in function 'RRConfigureOutputProperty'.
13766 'prop' allocated at line 334 with RRCreateOutputProperty(property).
13767 at line 350 of randr/rrproperty.c in function 'RRConfigureOutputProperty'.
13768 'prop' allocated at line 334 with RRCreateOutputProperty(property).
13769
13770 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13771 Reviewed-by: Keith Packard <keithp@keithp.com>
13772
13773commit 42e655de4d95cb108aec50efec6bbdb709bb13d7
13774Author: Alan Coopersmith <alan.coopersmith@oracle.com>
13775Date: Tue Jul 10 23:29:53 2012 -0700
13776
13777 rrproviderproperty.c: free newly allocated prop in more error paths
13778
13779 Reported by parfait 1.0:
13780
13781 Error: Memory leak (CWE 401)
13782 Memory leak of pointer 'prop' allocated with RRCreateProviderProperty(property)
13783 at line 221 of randr/rrproviderproperty.c in function 'RRChangeProviderProperty'.
13784 'prop' allocated at line 155 with RRCreateProviderProperty(property).
13785 prop leaks when pending != 0 at line 161.
13786
13787 Error: Memory leak (CWE 401)
13788 Memory leak of pointer 'prop' allocated with RRCreateProviderProperty(property)
13789 at line 345 of randr/rrproviderproperty.c in function 'RRConfigureProviderProperty'.
13790 'prop' allocated at line 333 with RRCreateProviderProperty(property).
13791 at line 349 of randr/rrproviderproperty.c in function 'RRConfigureProviderProperty'.
13792 'prop' allocated at line 333 with RRCreateProviderProperty(property).
13793
13794 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13795 Reviewed-by: Keith Packard <keithp@keithp.com>
13796
13797commit 94b514d5e4b376d05e106eb3853da511256e8545
13798Merge: 7d87545 7328900
13799Author: Keith Packard <keithp@keithp.com>
13800Date: Mon Aug 6 15:13:17 2012 -0700
13801
13802 Merge remote-tracking branch 'whot/for-keith'
13803
13804commit 7d87545ba7395ade507cca7bdca7052b26ed18d8
13805Merge: ad707a7 988d7ac
13806Author: Keith Packard <keithp@keithp.com>
13807Date: Mon Aug 6 15:11:13 2012 -0700
13808
13809 Merge remote-tracking branch 'jturney/master'
13810
13811commit ad707a7dcc65c30e030c7600b54f734090e56a91
13812Merge: 5a51cb8 884f51e
13813Author: Keith Packard <keithp@keithp.com>
13814Date: Mon Aug 6 15:08:01 2012 -0700
13815
13816 Merge remote-tracking branch 'jeremyhu/master'
13817
13818commit 5a51cb86f39a6809305c403beea54c3625e36259
13819Author: Keith Packard <keithp@keithp.com>
13820Date: Tue Jul 10 23:09:47 2012 -0700
13821
13822 xfree86: When xf86CrtcCloseScreen is called, the randr CRTCs are gone
13823
13824 The RandR CRTC structures are freed when their resource IDs are
13825 destroyed during server shut down, which is before the screen is
13826 closed. Calling back into RandR with stale pointers just segfaults the
13827 server.
13828
13829 Signed-off-by: Keith Packard <keithp@keithp.com>
13830 Reviewed-by: Dave Airlie <airlied@redhat.com>
13831 Tested-by: Knut Petersen <knut_petersen@t-online.de>
13832
13833commit 1bf81af4a6be1113bcc3b940ab264d5c9e0f0c5d
13834Author: Rui Matos <tiagomatos@gmail.com>
13835Date: Mon Jul 30 14:32:12 2012 -0400
13836
13837 xf86RandR12: Don't call ConstrainCursorHarder() if panning is enabled
13838
13839 Panning is at odds with CRTC cursor confinement. This disables CRTC cursor
13840 confinement as long as panning is enabled.
13841
13842 Fixes regression introduced in 56c90e29f04727c903bd0f084d23bf44eb1a0a11.
13843
13844 Reviewed-by: Adam Jackson <ajax@redhat.com>
13845 Signed-off-by: Rui Matos <tiagomatos@gmail.com>
13846 Signed-off-by: Keith Packard <keithp@keithp.com>
13847
13848commit ff56f88616aa63797384c2c484b2bd0f194df96a
13849Author: Adam Jackson <ajax@redhat.com>
13850Date: Mon Jul 30 14:32:11 2012 -0400
13851
13852 randr: Fix up yet another corner case in preferred mode selection
13853
13854 Let's say - purely for the sake of argument, mind you - that you had a
13855 server GPU with anemic memory bandwidth, and you walked up to it and
13856 plugged in a monitor that was 1920x1080 because that's what happened to
13857 be on the crash cart. Say the memory bandwidth is such that anything
13858 larger than 1280x1024 gets filtered away. Now you're in trouble,
13859 because the established timings section includes a 720x400 mode because
13860 that's what DOS 80x25 is, and that happens to just about match the
13861 physical aspect ratio.
13862
13863 Instead let's reuse the logic from the existing aspect-match path: pick
13864 the larger mode of either the physical aspect ratio or 4:3.
13865
13866 Reviewed-by: Julien Cristau <jcristau@debian.org>
13867 Signed-off-by: Adam Jackson <ajax@redhat.com>
13868 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
13869 Signed-off-by: Keith Packard <keithp@keithp.com>
13870
13871commit 531785dd746d64ef7f473a83ca73bb20e74b6fca
13872Author: Dave Airlie <airlied@redhat.com>
13873Date: Mon Jul 30 14:32:10 2012 -0400
13874
13875 kinput: allocate enough space for null character.
13876
13877 This code wasn't allocating enough space and was assigning the NULL
13878 one past the end.
13879
13880 Pointed out by coverity.
13881
13882 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
13883 Reviewed-by: Adam Jackson <ajax@redhat.com>
13884 Signed-off-by: Dave Airlie <airlied@redhat.com>
13885 Signed-off-by: Keith Packard <keithp@keithp.com>
13886
13887commit 8843aed82e7d69422e7763a35832a2be8f26723f
13888Author: Vic Lee <llyzs@163.com>
13889Date: Mon Jul 30 14:32:09 2012 -0400
13890
13891 ephyr: Resize screen automatically when parent window is resized
13892
13893 Bugzilla: https://bugs.freedesktop.org/25804
13894 Reviewed-by: Adam Jackson <ajax@redhat.com>
13895 Signed-off-by: Vic Lee <llyzs@163.com>
13896 Signed-off-by: Keith Packard <keithp@keithp.com>
13897
13898commit b46bbafae6d0a8b3f2f7853d5c1475fc223b1ed6
13899Author: Adam Jackson <ajax@redhat.com>
13900Date: Mon Jul 30 14:32:08 2012 -0400
13901
13902 ephyr: Fix up some bizarre formatting
13903
13904 Signed-off-by: Adam Jackson <ajax@redhat.com>
13905 Signed-off-by: Keith Packard <keithp@keithp.com>
13906
13907commit 7328900042b9c1312aed48753fd6054e64613e4c
13908Author: Alan Coopersmith <alan.coopersmith@oracle.com>
13909Date: Mon Jul 30 23:37:06 2012 -0700
13910
13911 XIChangeDeviceProperty: free newly allocated prop when SetProperty fails
13912
13913 Reported by parfait 1.0:
13914
13915 Error: Memory leak (CWE 401)
13916 Memory leak of pointer 'prop' allocated with XICreateDeviceProperty(property)
13917 at line 774 of Xi/xiproperty.c in function 'XIChangeDeviceProperty'.
13918 'prop' allocated at line 700 with XICreateDeviceProperty(property).
13919 prop leaks when handler != NULL at line 768
13920 and handler->SetProperty != NULL at line 769
13921 and checkonly != 0 at line 772
13922 and rc != 0 at line 772.
13923
13924 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13925 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
13926 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13927
13928commit 4dbbcdf64563cb95f83c04b2442cb7e868384264
13929Author: Daniel d'Andrada <daniel.dandrada@canonical.com>
13930Date: Thu Jul 26 17:31:57 2012 -0300
13931
13932 Do sent TouchEnd to listeners that don't own an accepted touch
13933
13934 When the owner of a touch accepts it, the other listeners must
13935 receive a TouchEnd.
13936
13937 Even though there's code implementing the logic above in
13938 ProcessTouchOwnershipEvent(), DeliverTouchEndEvent() was refusing to send
13939 those TouchEnd events in this situatuation.
13940
13941 Signed-off-by: Daniel d'Andrada <daniel.dandrada@canonical.com>
13942 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
13943 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13944
13945commit 02d91ccb0955252153206061a44340f051077624
13946Author: Peter Hutterer <peter.hutterer@who-t.net>
13947Date: Thu Jun 21 15:42:17 2012 +1000
13948
13949 test: always add DIX_LIB and OS_LIB on XORG builds
13950
13951 With --disable-xorg, We also disabled a bunch of tests because of their
13952 perceived reliance on a DDX. The cause was libtool missing some object files
13953 that never ended up in libxservertest.la. Only the xfree86 test has a true
13954 dependency on XORG.
13955
13956 DIX_LIB was pointing to dix.O (instead of libdix.la) when
13957 DTRACE_SPECIAL_OBJECTS was defined. libdix.la should be part of XSERVER_LIBS
13958 but dix.O is not a recognised libtool object, so it got skipped for
13959 libxservertest.a. Only in the XORG case would we add DIX_LIB and OS_LIB
13960 manually, thus forcing linkage with the dtrace-generated objects.
13961
13962 Fixing this by packaging up the dtrace-generated files as part of
13963 libdix.la/libos.la doesn't work for Solaris (and possible others), so simply
13964 always force linkage against the DIX_LIB/OS_LIB in the case of dtrace objects.
13965
13966 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13967 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13968 Tested-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13969
13970commit 988d7ace19a009991a4528e783d1a94c2444c66a
13971Author: Jon TURNEY <jon.turney@dronecode.org.uk>
13972Date: Thu Jul 5 09:38:44 2012 +0100
13973
13974 glx: Do not report the GLX_INTEL_swap_event extension for indirect swrast
13975
13976 Commit 84956ca4 bogusly adds GLX_INTEL_swap_event to the extensions reported
13977 by swrast.
13978
13979 "DRI2 supports this now - and already enables it explicitly - but drisw does
13980 not and should not. Otherwise toolkits like clutter will only ever SwapBuffers
13981 once and wait forever for an event that's not coming."
13982
13983 (A similar bug for direct swrast is already fixed in mesa commit 25620eb1)
13984
13985 (Note that this may be papering over the cracks somewhat, as if we do report
13986 GLX_INTEL_swap_event, some clutter apps fail with GLXBadDrawable calling
13987 GLXChangeDrawableAttributes to change the setting of GLX_BUFFER_SWAP_COMPLETE_INTEL_MASK
13988 in the GLX_EVENT_MASK, apparently after the drawable is destroyed, which suggests
13989 a bug with GLXDrawable lifetimes)
13990
13991 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
13992 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
13993
13994commit fd3d45c137bb849aa9030d732ea9277292e01d3d
13995Author: Jon TURNEY <jon.turney@dronecode.org.uk>
13996Date: Thu Jul 5 09:34:24 2012 +0100
13997
13998 glx: Don't note GLX_INTEL_swap_event as being required by GLX 1.4, it isn't.
13999
14000 Don't note GLX_INTEL_swap_event as being required by GLX 1.4, it isn't.
14001 (This data is not currently used in the server)
14002
14003 (A similar change is made in mesa commit d3f7597bc9f6d5)
14004
14005 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14006 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
14007
14008commit a8464dfa28dea78201e3e4398eb3bcb745e10087
14009Author: Colin Harrison <colin.harrison@virgin.net>
14010Date: Sun Jul 22 13:15:02 2012 +0100
14011
14012 os: Fix TMP fall-back in Win32TempDir()
14013
14014 Fix Win32TempDir() in the case where we fell back to checking the TMP
14015 environment variable. It looks like this has been wrong since forever.
14016
14017 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
14018 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14019
14020commit f6e7b82acadfca8239edc0f7e72cd0f3f9cfc2c4
14021Author: Marc Haesen <marha@users.sourceforge.net>
14022Date: Sat Jul 21 21:18:44 2012 +0100
14023
14024 hw/xwin: Fixes to pixelFormat <-> fbConfig conversion in WGL mode
14025
14026 Fix FIXME in fbConfigToPixelFormat() to correctly populate RGBA-mask shift
14027 parameters. Also request colourindex pixelFormats correctly.
14028
14029 Now that they are requested correctly, don't skip colorindex visuals when
14030 converting pixelFormats to fbConfigs.
14031
14032 Populate transparent colour information when converting pixelFormat from
14033 DescribePixelFormats() to a fbConfig.
14034
14035 Signed-off-by: Marc Haesen <marha@users.sourceforge.net>
14036 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14037 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14038
14039commit 45c432871d6a244e9e558a6a4e7c36e90764135e
14040Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14041Date: Sat Jul 21 14:13:37 2012 +0100
14042
14043 hw/xwin: Introduce winProcessXEventsTimeout() to the concept of fractions of a second
14044
14045 Oh this is terrible.
14046
14047 Currently we only compute the select timeout in whole seconds. This means if we
14048 have less than 1 second remaining, we select with a timeout of 0 (i.e. poll)
14049 which causes the task to spin, burning 100% CPU for the remaining timeout (and
14050 possibly preventing the process we are waiting for from running :S)
14051
14052 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14053 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14054
14055commit 23cd4d0174194e10721d2e465fd1a1c52f001520
14056Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14057Date: Sat Jul 21 12:33:05 2012 +0100
14058
14059 hw/xwin: Fix winUpdateWindowPosition() not to assume WS_EX_APPWINDOW style
14060
14061 Also improve it's debug output a bit
14062
14063 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14064 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14065
14066commit 527cf13135cfd279733060e0028bbfbe02be5167
14067Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14068Date: Sat Jul 21 14:09:16 2012 +0100
14069
14070 hw/xwin: Make winOverrideIcon() thread-safe for icon data access
14071
14072 winOverrideIcon() is called from the internal WM client thread.
14073
14074 Accessing server-internal data structures to get icon data or window hints is
14075 not safe, as there is no lock to ensure we do not collide with these data
14076 structures being updated in the server thread.
14077
14078 Rewrite so the internal client thread uses X client calls to obtain this data
14079 safely
14080
14081 We used to also set the icon inside the server when the window was initially
14082 created. For simplicity, we simply send a message to the internal WM to update
14083 the icon when the window is created (rather than writing different icon update
14084 code which can work in the server thread for that one case...)
14085
14086 extwm mode used to do the icon update in the server. I'm not sure that actually
14087 made much sense. Let's assume the external WM client can do it instead...
14088
14089 v2
14090 Make sure that WM_WM_ICON_EVENT does nothing for override-redirect windows
14091
14092 v3
14093 Reinstate check that native window actually has expected properties for an X
14094 window before trying to update it's icon; some auxiliary windows owned by the
14095 XWin process don't, which would cause a crash
14096
14097 v4
14098 Various fixes to pixmap icon conversion:
14099 - remove left-over malloc in winScaleXimageToWindowsIcon causing a memory leak
14100 - don't recalculate DDBitmap stride in winScaleXimageToWindowsIcon, when we already have worked it out
14101 - properly check that XGetWindowProperty(NET_WM_ICON) returned some data
14102 - don't try to retrieve WM_HINTS icon_mask if it isn't set
14103 - restore accidentally dropped calculation of effBpp, stride, maskStride of output DDBitmap
14104 - make sure imageMask is zero-initalized before we use it to mask the DDBitmap
14105
14106 v5
14107 Remove a left-over unused variable
14108
14109 v6
14110 Avoid XDestroyImage(NULL) crash if XGetImage failed for icon_pixmap
14111
14112 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14113 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14114
14115commit 2677d89823b851fea31036f51589985bb86864b7
14116Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14117Date: Sun Jul 22 17:38:55 2012 +0100
14118
14119 hw/xwin: Also update icon when _NET_WM_ICON property changes
14120
14121 _NET_WM_ICON property is also considered to decide on the window icon, so also
14122 send a WM_WM_ICON_EVENT message to the WM if the PropertyNotify event is for
14123 _NET_WM_ICON property
14124
14125 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14126 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14127
14128commit a07541f1ffc28c57a762beefb96c7bf0ac40b0a4
14129Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14130Date: Sun Jul 22 16:15:12 2012 +0100
14131
14132 hw/xwin: Rename WM_WM_HINTS_EVENT to WM_WM_ICON_EVENT
14133
14134 WM_WM_HINTS_EVENT only updates the icon, so rename it to WM_WM_ICON_EVENT
14135
14136 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14137 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14138
14139commit 2a47c91eb382d1ff0fb009a39efa7dc9c6fd5112
14140Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14141Date: Fri Jul 20 13:51:35 2012 +0100
14142
14143 hw/xwin: Refactor Xutf8TextPropertyToString() from GetWindowName() as a separate utility function
14144
14145 Simplify GetWindowName() by moving UTF-8 to wchar conversion out to it's call
14146 site. This allows us to do extra processing on the window name in future.
14147
14148 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14149 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14150
14151commit 884f51e9770c9ccd3b38a14ea393ea2c7a5ef235
14152Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
14153Date: Wed Aug 1 18:46:08 2012 -0700
14154
14155 XQuartz: Fix xp_window_bring_all_to_front linking on OS versions with older libXplugin
14156
14157 Found-by: Tinderbox
14158 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
14159
14160commit b4c2358e0a388730f1beda5d3a12e79755208ce6
14161Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
14162Date: Wed Aug 1 18:45:43 2012 -0700
14163
14164 pbproxy: Fix a warning about a bad method prototype
14165
14166 x-selection.m:1502:1: warning: method has no return type specified; defaults to 'id'
14167 [-Wmissing-method-return-type,Semantic Issue]
14168 - init
14169 ^
14170 (id)
14171 1 warning generated.
14172
14173 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
14174
14175commit f193907f1b2f8d17dac9b59b7e93ddbb0f4e31ca
14176Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14177Date: Fri Jun 22 14:56:22 2012 +0100
14178
14179 Fix pseudoramiX.c compilation without darwin.h
14180
14181 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14182 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
14183
14184commit 067931ccce592e319baffa26d4f7380d8d924537
14185Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14186Date: Mon Jul 23 12:16:36 2012 +0100
14187
14188 hw/xquartz: Various fixes for pseudoramiX.c
14189
14190 Various fixes, applied to panoramiX.c in commit 2b266eda, also need applying to pseudoramiX.c:
14191 Fix panoramiX request and reply swapping
14192 Set window and screen values in panoramix replies
14193 Prevent buffer overrun in ProcPanoramiXGetScreenSize
14194
14195 These fixes seem to be necessary in order to compile pseudoramiX.c with gcc
14196
14197 pseudoramiX.c: In function 'ProcPseudoramiXGetState':
14198 pseudoramiX.c:221:56: error: call to 'wrong_size' declared with attribute error: wrong sized variable passed to swap
14199 pseudoramiX.c: In function 'ProcPseudoramiXGetScreenCount':
14200 pseudoramiX.c:250:62: error: call to 'wrong_size' declared with attribute error: wrong sized variable passed to swap
14201 pseudoramiX.c: In function 'ProcPseudoramiXGetScreenSize':
14202 pseudoramiX.c:283:56: error: call to 'wrong_size' declared with attribute error: wrong sized variable passed to swap
14203 pseudoramiX.c:284:57: error: call to 'wrong_size' declared with attribute error: wrong sized variable passed to swap
14204
14205 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14206 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
14207
14208commit aad428b8e21c77397c623b78706eb64b1fea77c9
14209Author: Dave Airlie <airlied@redhat.com>
14210Date: Thu Jul 26 09:54:08 2012 +1000
14211
14212 glx: drop GLX_LIBS from X server and workaround sdksyms.
14213
14214 We've had reports of two copies of the GLX bits, one in the server
14215 and one in libglx.so causing problems, I didn't understand why the
14216 X server needed a copy so drop it, however then we have to fix a missing
14217 GlxExtensionInit that comes from sdksyms, so work around it by moving
14218 that one declaration into a header that sdksyms doesn't scan.
14219
14220 Thanks to Jon Turney for debugging the actual problem.
14221 (copyright header from extinit.h that seems most appropriate put on top).
14222
14223 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=52402
14224 Tested-by: Peter Hutterer <peter.hutterer@who-t.net>
14225 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
14226 Tested-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14227 Signed-off-by: Dave Airlie <airlied@redhat.com>
14228 Signed-off-by: Keith Packard <keithp@keithp.com>
14229
14230commit afa53fe7cffd430cf11f25ca818cb955a78c0c1c
14231Author: Keith Packard <keithp@keithp.com>
14232Date: Wed Jul 25 14:29:48 2012 -0700
14233
14234 Version 1.12.99.903 (1.13 RC3)
14235
14236 Signed-off-by: Keith Packard <keithp@keithp.com>
14237
14238commit 98e3f3fde35ca805d44ee38394bfd86058f9f4b8
14239Author: Adam Jackson <ajax@redhat.com>
14240Date: Thu Jul 19 15:00:05 2012 -0400
14241
14242 doc: Drop XAA references from xorg.conf man page
14243
14244 Signed-off-by: Adam Jackson <ajax@redhat.com>
14245 Reviewed-by: Dave Airlie <airlied@redhat.com>
14246 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14247 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
14248 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
14249 Acked-by: Aaron Plattner <aplattner@nvidia.com>
14250 Signed-off-by: Keith Packard <keithp@keithp.com>
14251
14252commit dea928477bf835a13b2bd2b16c0dbaaf5468130c
14253Author: Adam Jackson <ajax@redhat.com>
14254Date: Thu Jul 19 15:00:04 2012 -0400
14255
14256 xfree86: Drop some dead XAA decls from SDK headers
14257
14258 Signed-off-by: Adam Jackson <ajax@redhat.com>
14259 Reviewed-by: Dave Airlie <airlied@redhat.com>
14260 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14261 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
14262 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
14263 Acked-by: Aaron Plattner <aplattner@nvidia.com>
14264 Signed-off-by: Keith Packard <keithp@keithp.com>
14265
14266commit a44a379a2a4c198dbd6ffd6cc0db376f8fb64327
14267Author: Adam Jackson <ajax@redhat.com>
14268Date: Thu Jul 19 15:00:03 2012 -0400
14269
14270 cw: Remove
14271
14272 XAA was the only consumer.
14273
14274 Signed-off-by: Adam Jackson <ajax@redhat.com>
14275 Reviewed-by: Dave Airlie <airlied@redhat.com>
14276 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14277 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
14278 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
14279 Acked-by: Aaron Plattner <aplattner@nvidia.com>
14280 Signed-off-by: Keith Packard <keithp@keithp.com>
14281
14282commit 20b4d90f32b3ca5def5be2fdf04a91ae6c47f125
14283Merge: 83df169 fd228d3
14284Author: Keith Packard <keithp@keithp.com>
14285Date: Wed Jul 25 13:52:35 2012 -0700
14286
14287 Merge remote-tracking branch 'jeremyhu/master'
14288
14289commit fd228d318428d0d9dea52ca560a9b66660dd2d02
14290Author: Jeremy Huddleston <jeremyhu@apple.com>
14291Date: Wed Jul 25 11:40:33 2012 -0700
14292
14293 XQuartz: Call xp_window_bring_all_to_front if available in libXplugin
14294
14295 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
14296
14297commit 83df169538bddcf3a61e2c5e02a4f3e10af93b60
14298Author: Aaron Plattner <aplattner@nvidia.com>
14299Date: Fri Jul 20 00:59:20 2012 -0700
14300
14301 dix: don't use "new" as a parameter name
14302
14303 "new" is a reserved word in C++.
14304
14305 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
14306 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14307 Signed-off-by: Keith Packard <keithp@keithp.com>
14308
14309commit 3b7f313b3d74f9df876152c55d55cbb339bf2d02
14310Author: Jeremy Huddleston <jeremyhu@apple.com>
14311Date: Tue Jul 24 00:32:30 2012 -0700
14312
14313 XQuartz: Fix runtime regressions introduced by extension loading changes
14314
14315 * GLX is now loaded
14316 * PseudoramiX loading is back in miinitext. It needs to be loaded
14317 before RandR.
14318
14319 Regression-from: 27a624bee66d2797eacf90317054207dcf248028
14320 Regression-from: 5f5bbbe543f65c48ecbb5cce80116a86ca3fbe86
14321
14322 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
14323
14324commit 0d64e881a31d89d28f2ded9b8e83e93632870a16
14325Author: Jeremy Huddleston <jeremyhu@apple.com>
14326Date: Mon Jul 23 23:16:29 2012 -0700
14327
14328 XQuartz: Build fix for __GLXscreen::createContext changes
14329
14330 Regression-from: 96d74138b1c0273e88933220d99a893858b649cd
14331
14332 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
14333
14334commit 77de2994774e7be24a0bad521180628feb1027c6
14335Author: Keith Packard <keithp@keithp.com>
14336Date: Tue Jul 17 15:47:39 2012 -0700
14337
14338 Bump to version 1.12.99.902
14339
14340 Signed-off-by: Keith Packard <keithp@keithp.com>
14341
14342commit beeea70495a8d5c4afc1b1eb2cc06a24ab1a986d
14343Merge: 6e438a0 6910280
14344Author: Keith Packard <keithp@keithp.com>
14345Date: Tue Jul 17 12:17:39 2012 -0700
14346
14347 Merge remote-tracking branch 'alanc/master'
14348
14349commit 6e438a0e183f5a6c22070746c038af53d5b935ca
14350Merge: 5884e7d 75966a4
14351Author: Keith Packard <keithp@keithp.com>
14352Date: Tue Jul 17 12:12:53 2012 -0700
14353
14354 Merge remote-tracking branch 'jturney/master'
14355
14356commit 75966a4186955f3a4625e93796145036c5986d2e
14357Author: Keith Packard <keithp@keithp.com>
14358Date: Fri Jul 13 14:36:25 2012 -0700
14359
14360 xwin: Clean up os wrappers for System, Popen and Pclose on Windows
14361
14362 Popen and Pclose are never used on Windows, so don't bother to even
14363 try to define them.
14364
14365 System(s) was defined as system(s), but the two users of that
14366 function are in xkb, which carefully redefines that as
14367 Win32System. Move Win32System and Win32TempDir to os/utils.c, renaming
14368 Win32System to be just System, which simplifies the xkb code
14369
14370 Signed-off-by: Keith Packard <keithp@keithp.com>
14371 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14372
14373commit 6910280297ab610100dd8e7fbe5bb5cc08968f2f
14374Author: Alan Coopersmith <alan.coopersmith@oracle.com>
14375Date: Sat Jul 14 10:54:36 2012 -0700
14376
14377 Use C99 designated initializers in DRI2GetParam replies
14378
14379 DRI2GetParam was going through review in parallel with main batch of
14380 C99 initialization changes - sync up now that both have landed.
14381
14382 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14383 Reviewed-by: Keith Packard <keithp@keithp.com>
14384
14385commit a6c5b8d3ee053cadf685b0e4de96407ad6df4dd4
14386Author: Alan Coopersmith <alan.coopersmith@oracle.com>
14387Date: Sat Jul 14 10:50:41 2012 -0700
14388
14389 Use C99 designated initializers in RandR 1.4 extension Events
14390
14391 RandR 1.4 was going through review in parallel with main batch of
14392 C99 initialization changes - sync up now that both have landed.
14393
14394 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14395 Reviewed-by: Keith Packard <keithp@keithp.com>
14396
14397commit ada04ef0ca8a1213e865ba2c000ccd6ed6d0e371
14398Author: Alan Coopersmith <alan.coopersmith@oracle.com>
14399Date: Sat Jul 14 10:45:29 2012 -0700
14400
14401 Use C99 designated initializers in RandR 1.4 extension Replies
14402
14403 RandR 1.4 was going through review in parallel with main batch of
14404 C99 initialization changes - sync up now that both have landed.
14405
14406 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14407 Reviewed-by: Keith Packard <keithp@keithp.com>
14408
14409commit fb73f7f40fa46458990038332017d4496caa0691
14410Author: Alan Coopersmith <alan.coopersmith@oracle.com>
14411Date: Sat Jul 14 10:02:52 2012 -0700
14412
14413 Fix two more C99 initialization mistakes using members of same struct
14414
14415 Similar to 34cf559bcf99dad, use temporary variables instead of
14416 referencing members of the struct being initialized in the middle
14417 of the initialization.
14418
14419 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14420 Reviewed-by: Keith Packard <keithp@keithp.com>
14421
14422commit 5884e7dedecdd82ddbb037360cf9c85143e094b5
14423Author: Aaron Plattner <aplattner@nvidia.com>
14424Date: Mon Jul 16 17:40:31 2012 -0700
14425
14426 xf86: Re-export extension disable flags
14427
14428 These flags were unexported by commit a1d41e311c21eb6627caa0d168e070ceaf90806f,
14429 which moved the declarations around and lost the _X_EXPORT attributes in the
14430 process. Since drivers need these and it's late in the release cycle, just
14431 re-export them for now.
14432
14433 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
14434 Reviewed-by: Andy Ritger <aritger@nvidia.com>
14435 Signed-off-by: Keith Packard <keithp@keithp.com>
14436
14437commit 357ec54f46d31ff77c441c15c322152e04f89a11
14438Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14439Date: Wed Mar 14 13:58:32 2012 +0000
14440
14441 hw/xwin: Refer to xkeyboard-config man page for XKB configuration option values
14442
14443 Also polish the keyboard configuration text a bit
14444 Include man section number in references to setxkbmap
14445 Consistently refer to 'manual page' rather using both that and 'man page'
14446
14447 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14448 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14449
14450commit 5940580f8c12145c1ef4c3eee898676b48d95879
14451Author: Jörg Mensmann <joerg_ml@bitplanet.de>
14452Date: Tue Apr 17 10:48:19 2012 +0100
14453
14454 hw/xwin: Fix command line arguments for multiple monitors.
14455
14456 Moving Xwin to a certain monitor using "-screen 0 @2" would fail,
14457 printing "ddxProcessArgument - screen - Querying monitors failed".
14458
14459 This happened since commit 3ead1d810b0e157078db39712e02ea6dc85216d8,
14460 because EnumDisplayMonitor() returns FALSE if its callback function
14461 returns FALSE (which is not clearly documented), and QueryMonitor()
14462 would then also return FALSE.
14463
14464 Moving back to the old behaviour, where the return value of
14465 EnumDisplayMonitors() is ignored.
14466
14467 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14468 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14469
14470commit 18801f5e5a98d80f4811ade8c98df65175b1935a
14471Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14472Date: Fri Feb 10 09:27:16 2012 +0000
14473
14474 hw/xwin: Remove unused WindowExposuresProcPtr storage in screen private
14475
14476 We don't wrap the WindowExposures screen function
14477
14478 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14479 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14480
14481commit 8fd30859c293035fe505da76d8d7e64131bd5bc0
14482Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14483Date: Fri Feb 10 13:25:32 2012 +0000
14484
14485 hw/xwin: Remove long obsolete README for hw/xwin
14486
14487 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14488 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14489
14490commit 3568d4af4ff2490e79286d242916e81afed57b58
14491Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14492Date: Thu Feb 23 22:32:09 2012 +0000
14493
14494 hw/xwin: Remove a pointless OS version check in winInstallKeyboardHookLL()
14495
14496 Checking for OS later than NT4 SP3 (which we didn't do correctly anyhow,
14497 just checking for NT) is pointless when other code in hw/xwin assumes at
14498 least NT 5.0
14499
14500 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14501 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14502
14503commit 0ed0d3979050c4ee7bef64441d50c494d58d93b0
14504Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14505Date: Thu Jul 5 22:08:38 2012 +0100
14506
14507 hw/xwin: Remove winClipboardDetectUnicodeSupport()
14508
14509 This code for detecting if the Windows clipboard can support unicode
14510 is long obsolete.
14511
14512 All NT versions of Windows support unicode clipboard, so any version
14513 of Windows we can run on must support unicode clipboard.
14514
14515 The -nounicodeclipboard flag to disable use of unicode on the clipboard
14516 is retained.
14517
14518 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14519 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14520
14521commit c98471fad7d85fc9d9959b1c48edd90415f7bf15
14522Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14523Date: Thu Jul 5 18:00:06 2012 +0100
14524
14525 hw/xwin: Consolidate duplicated code as getHwnd() function
14526
14527 Consolidate code to find the native HWND of an X window, stored in the _WINDOWSWM_NATIVE_HWND
14528 property, duplicated in UpdateName() and PreserveWin32Stack() as getHwnd()
14529
14530 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14531 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14532
14533commit df7636dddb4981e14ee7d128c43bb08bdf6a8727
14534Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14535Date: Thu Jul 5 14:20:52 2012 +0100
14536
14537 hw/xwin: Correct function name in log message
14538
14539 Correct function name in log message winInitializeDefaultScreens -> winInitializeScreenDefaults
14540
14541 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14542 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14543
14544commit 9cdd3ed07425b222e8197b9a2f57ec1525c6fd54
14545Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14546Date: Fri Mar 16 12:56:45 2012 +0000
14547
14548 hw/xwin: Tidy up some more pointless output which is always emitted
14549
14550 It's not very useful to log lines like the below when the About.. dialog is used
14551
14552 winAboutDlgProc - WM_COMMAND - IDOK or IDCANCEL
14553 winAboutDlgProc - WM_CLOSE
14554
14555 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14556 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14557
14558commit eb5b4d8d809712053701f057a35bbf5348ac9059
14559Author: Colin Harrison <colin.harrison@virgin.net>
14560Date: Thu Jul 5 14:03:44 2012 +0100
14561
14562 hw/xwin: RANDR resize didn't occur when screen window was maximized or restored
14563
14564 In windowed mode, with option -resize, RANDR resize was missing when the screen
14565 window was maximized or restored.
14566
14567 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
14568 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14569
14570commit 7e07d61857c5d2ee6e3c36c81b51bb4fe378b5a0
14571Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14572Date: Thu Jul 5 14:01:03 2012 +0100
14573
14574 hw/xwin: Bug fixes for screen sizing when the screen window is on a non-primary monitor
14575
14576 There is a bug that when the -screen option is used to specify a monitor for
14577 the screen window to be located on, but no explicit size is specified
14578 (and the -multiplemonitors option isn't specified), the screen
14579 window size is always constrained to fit the work area of the primary
14580 monitor (rather than the work area of the specified monitor)
14581
14582 This gives incorrect results if you want a screen the same size as your
14583 non-primary monitor (e.g. by using -screen 0 @2) and your non-primary
14584 monitor is larger than your primary monitor.
14585
14586 (This can be worked around by specifying -multiplemonitors and an explicit
14587 screen size the same size as the monitor (e.g. -multiplemonitors -screen 0
14588 1600x1200@2))
14589
14590 Fix to use work area for the monitor specified for the screen, rather than the
14591 primary monitor work area (unless -multiplemonitors is used, in which case we
14592 continue to use the virtual desktop work area instead)
14593
14594 Also fix the adjustment for an autohide taskbar, so that it is only done if the
14595 taskbar is on the same monitor as the screen (or -multiplemonitors is used)
14596
14597 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14598 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14599
14600commit 090856b55b5895ce07f36ddebe6d8f7cb86c1826
14601Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14602Date: Thu Jul 5 13:19:02 2012 +0100
14603
14604 hw/xwin: Tidy up some pointless output which is always emitted
14605
14606 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14607 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14608
14609commit 9fe9c5f868886abf671b3be3964af1965dd92472
14610Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14611Date: Sun Feb 13 18:52:52 2011 +0000
14612
14613 hw/xwin: Don't bother reporting XORG_VERSION_CURRENT
14614
14615 Don't bother reporting XORG_VERSION_CURRENT, when we also
14616 report the version number broken down into it's components
14617 XORG_VERSION_MAJOR, XORG_VERSION_MINOR, etc. as well.
14618
14619 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14620 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14621
14622commit d5977e5bd2c8ee77db08d5f948f266c9661b9fb5
14623Author: Dave Airlie <airlied@gmail.com>
14624Date: Fri Jul 13 00:28:19 2012 +1000
14625
14626 xf86: include xf86platformBus.h in xf86AutoConfig.c
14627
14628 This fixes an implicit declaration,
14629 xf86AutoConfig.c:202:5: error: implicit declaration of function 'xf86PlatformMatchDriver' [-Werror=implicit-function-declaration]
14630 xf86AutoConfig.c:202:5: warning: nested extern declaration of 'xf86PlatformMatchDriver' [-Wnested-externs]
14631
14632 Signed-off-by: Dave Airlie <airlied@redhat.com>
14633 Reviewed-by: Keith Packard <keithp@keithp.com>
14634
14635commit afa3b9ccd27ff9c58f536cb3b727e85feb6a6f6a
14636Author: Bartosz Brachaczek <b.brachaczek@gmail.com>
14637Date: Fri Jul 6 02:52:40 2012 +0200
14638
14639 glx: Fix checking GL version
14640
14641 Signed-off-by: Bartosz Brachaczek <b.brachaczek@gmail.com>
14642 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
14643 Signed-off-by: Dave Airlie <airlied@gmail.com>
14644
14645commit 743ccf6fa0a9a4579d6de6d10e2e6040640772d9
14646Author: Dave Airlie <airlied@gmail.com>
14647Date: Thu Jul 12 19:09:15 2012 +1000
14648
14649 glx/dri2: don't return NULL in a function with no return.
14650
14651 Reported by gcc.
14652
14653 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
14654 Reviewed-by: Matt Turner <mattst88@gmail.com>
14655 Signed-off-by: Dave Airlie <airlied@redhat.com>
14656
14657commit d1a00c26becdc2490ab4660abf108eca09c30d23
14658Author: Dave Airlie <airlied@gmail.com>
14659Date: Thu Jul 12 19:05:41 2012 +1000
14660
14661 glx/dri2: use correct define from dri attribs.
14662
14663 Looks like idr renamed this and pushed the wrong one.
14664
14665 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
14666 Reviewed-by: Matt Turner <mattst88@gmail.com>
14667 Signed-off-by: Dave Airlie <airlied@redhat.com>
14668
14669commit c62205adcd48d497450655d0136f8ca5bd5cbae7
14670Author: Dave Airlie <airlied@gmail.com>
14671Date: Thu Jul 12 19:04:13 2012 +1000
14672
14673 glx/dri2: fix incorrect == vs =
14674
14675 glxdri2.c:486: warning: statement with no effect
14676
14677 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
14678 Reviewed-by: Matt Turner <mattst88@gmail.com>
14679 Signed-off-by: Dave Airlie <airlied@redhat.com>
14680
14681commit defca4c192bd7f4af2b273c7cb1fe8665888772a
14682Author: Dave Airlie <airlied@gmail.com>
14683Date: Thu Jul 12 18:52:34 2012 +1000
14684
14685 glx/dri2: fix typo in the notification reset for robustness.
14686
14687 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
14688 Reviewed-by: Matt Turner <mattst88@gmail.com>
14689 Signed-off-by: Dave Airlie <airlied@redhat.com>
14690
14691commit d1cae6bb804f7dfc889fd3c5d5eb116f4baf5149
14692Author: Dave Airlie <airlied@gmail.com>
14693Date: Thu Jul 12 18:51:52 2012 +1000
14694
14695 glx: pass screen into the convert function
14696
14697 when robustness is enabled, this is required.
14698
14699 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
14700 Signed-off-by: Dave Airlie <airlied@redhat.com>
14701
14702commit ea74e381ae91f45ec195a5505262e2e5a3c0e1c9
14703Author: Dave Airlie <airlied@gmail.com>
14704Date: Fri Jul 13 00:47:24 2012 +1000
14705
14706 xtest: add extinit.h for SyncExtensionInit
14707
14708 This adds the decl for SyncExtenionInit.
14709
14710 Signed-off-by: Dave Airlie <airlied@redhat.com>
14711 Tested-by: Keith Packard <keithp@keithp.com>
14712 Signed-off-by: Keith Packard <keithp@keithp.com>
14713
14714commit 746193a7f10f11cccae05c98b933aadf58fc7375
14715Merge: a2d0829 2b74949
14716Author: Keith Packard <keithp@keithp.com>
14717Date: Thu Jul 12 15:10:51 2012 -0700
14718
14719 Merge remote-tracking branch 'jturney/xwin-extmod-removal-fixes'
14720
14721commit a2d0829531249e24dbca25fc20ed30a2bb2d8ed8
14722Author: Michel Dänzer <michel.daenzer@amd.com>
14723Date: Thu Jul 12 13:16:37 2012 +0200
14724
14725 glx: Free DRI2 drawable reference to destroyed GLX drawable.
14726
14727 Otherwise the reference can lead to use after free in
14728 __glXDRIinvalidateBuffers().
14729
14730 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=50019
14731
14732 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
14733 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
14734 Signed-off-by: Keith Packard <keithp@keithp.com>
14735
14736commit 8a87acc9e57be740e3bd252567622246051a0723
14737Author: Michel Dänzer <michel.daenzer@amd.com>
14738Date: Thu Jul 12 13:16:36 2012 +0200
14739
14740 dri2: Add DRI2CreateDrawable2.
14741
14742 Same as DRI2CreateDrawable, except it can return the DRI2 specific XID of the
14743 DRI2 drawable reference to the base drawable.
14744
14745 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
14746 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
14747 Signed-off-by: Keith Packard <keithp@keithp.com>
14748
14749commit 2b74949ad02bd916c3ac502db3f28057f12a8117
14750Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14751Date: Thu Jul 12 00:36:43 2012 +0100
14752
14753 Restore GLX extension to XWin
14754
14755 - Add an XwinExtensionInit(), similar in spirit to QuartzExtensionInit()
14756 and xf86ExtensionInit()
14757 - Load the GLX extension (I'm guessing XQuartz needs this adding somewhere
14758 as well)
14759 - Also, since we now have a proper place to install the native GL provider
14760 where it will take priority over the the swrast provider, do so
14761
14762 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14763 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
14764 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14765
14766commit 656e5fea6c9ca15f932889edcc590dd01c881945
14767Merge: 8b820f2 1720cd9
14768Author: Keith Packard <keithp@keithp.com>
14769Date: Thu Jul 12 08:26:35 2012 -0700
14770
14771 Merge remote-tracking branch 'jeremyhu/master'
14772
14773commit 8b820f221ad54eb96a16c7060048abc8824371b6
14774Author: Daniel Stone <daniel@fooishbar.org>
14775Date: Wed Jul 11 17:31:53 2012 +0100
14776
14777 sdksyms: Fix build with --disable-xv
14778
14779 I hate this [redacted] script.
14780
14781 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
14782 Tested-by: Keith Packard <keithp@keithp.com>
14783 Signed-off-by: Keith Packard <keithp@keithp.com>
14784
14785commit 06ac7937fc37cdf073b77386bed47e14cff23628
14786Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14787Date: Thu Jul 12 12:37:05 2012 +0100
14788
14789 Restore extern to the declaration of __glXDRISWRastProvider
14790
14791 b86aa74 dropped the 'extern' from the declaration of __glXDRISWRastProvider
14792
14793 This turns out to be important to me, as without it, the final link only gets
14794 the tentative definition of __glXDRISWRastProvider implied by the declaration,
14795 and not the proper one from glxdriswrast.c, presumably because nothing else
14796 references anything in the object that file generates.
14797
14798 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14799 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
14800 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14801
14802commit bcbf95b1bafa6ffe724768b9309295e2fdb4b860
14803Author: Jon TURNEY <jon.turney@dronecode.org.uk>
14804Date: Thu Jul 12 00:36:10 2012 +0100
14805
14806 Revert bogus GlxPushProvider() in commit a1d41e3
14807
14808 a1d41e3 "Move extension initialisation prototypes into extinit.h"
14809 also includes a change to GlxExtensionInit to install the swrast GLX
14810 provider.
14811
14812 Since b86aa74 "GLX: Insert swrast provider from GlxExtensionInit"
14813 already does this (correctly, by installing the swrast provider
14814 at the end of the chain, rather than at the beginning), and since this
14815 would seem to have the effect of making the swrast provider the most
14816 preferred provider, I'm guessing this wasn't intended.
14817
14818 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
14819 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
14820 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
14821
14822commit 1720cd9badf21e5d6b2c6c929e609ab3b0b92fd2
14823Author: Jeremy Huddleston <jeremyhu@apple.com>
14824Date: Wed Jul 11 23:54:27 2012 -0700
14825
14826 XQuartz: Re-add prototype for DRIExtensionInit
14827
14828 Regression from: a1d41e311c21eb6627caa0d168e070ceaf90806f
14829
14830 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
14831 CC: Daniel Stone <daniel@fooishbar.org>
14832
14833commit 1a64d3cc9c319d5e9ab2365b60469fb5b83407fc
14834Author: Jeremy Huddleston <jeremyhu@apple.com>
14835Date: Wed Jul 11 23:49:59 2012 -0700
14836
14837 XQuartz: Fix build failure due to incorrect PseudoramiXExtensionInit prototype
14838
14839 pseudoramiX.c:129:1: error: conflicting types for 'PseudoramiXExtensionInit' [Semantic Issue]
14840 PseudoramiXExtensionInit(int argc, char *argv[])
14841 ^
14842 ../../include/extinit.h:175:13: note: previous declaration is here
14843 extern void PseudoramiXExtensionInit(void);
14844 ^
14845
14846 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
14847
14848commit 228f8bfe81890770f1ba109a0acca1c81a2ffee9
14849Author: Jeremy Huddleston <jeremyhu@apple.com>
14850Date: Wed Jul 11 23:49:33 2012 -0700
14851
14852 XQuartz: Silence an unused-variable warning
14853
14854 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
14855
14856commit deb08658e2a6b1647a7213a316c6f3019bcdce48
14857Author: Maarten Lankhorst <maarten.lankhorst@canonical.com>
14858Date: Wed Jul 11 16:27:46 2012 +0200
14859
14860 xfree86: Strip dangling pointers from desiredMode
14861
14862 Based on the original patch by Chris Wilson, which was a better fix than mine.
14863
14864 We stash a copy of the desiredMode on the crtc so that we can restore it
14865 after a vt switch. This copy is a simple memcpy and so also stashes a
14866 references to the pointers contained within the desiredMode. Those
14867 pointers are freed the next time the outputs are probed and mode list
14868 rebuilt, resulting in us chasing those dangling pointers on the next
14869 mode switch.
14870
14871 ==22787== Invalid read of size 1
14872 ==22787== at 0x40293C2: __GI_strlen (in
14873 /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
14874 ==22787== by 0x668F875: strdup (strdup.c:42)
14875 ==22787== by 0x5DBA00: XNFstrdup (utils.c:1124)
14876 ==22787== by 0x4D72ED: xf86DuplicateMode (xf86Modes.c:209)
14877 ==22787== by 0x4CA848: xf86CrtcSetModeTransform (xf86Crtc.c:276)
14878 ==22787== by 0x4D05B4: xf86SetDesiredModes (xf86Crtc.c:2677)
14879 ==22787== by 0xA7479D0: sna_create_screen_resources
14880 (sna_driver.c:220)
14881 ==22787== by 0x4CB914: xf86CrtcCreateScreenResources (xf86Crtc.c:725)
14882 ==22787== by 0x425498: main (main.c:216)
14883 ==22787== Address 0x72c60e0 is 0 bytes inside a block of size 9 free'd
14884 ==22787== at 0x4027AAE: free (in
14885 /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
14886 ==22787== by 0x4A547E: xf86DeleteMode (xf86Mode.c:1984)
14887 ==22787== by 0x4CD84F: xf86ProbeOutputModes (xf86Crtc.c:1578)
14888 ==22787== by 0x4DC405: xf86RandR12GetInfo12 (xf86RandR12.c:1537)
14889 ==22787== by 0x518119: RRGetInfo (rrinfo.c:202)
14890 ==22787== by 0x51D997: rrGetScreenResources (rrscreen.c:335)
14891 ==22787== by 0x51E0D0: ProcRRGetScreenResources (rrscreen.c:475)
14892 ==22787== by 0x513852: ProcRRDispatch (randr.c:493)
14893 ==22787== by 0x4346DB: Dispatch (dispatch.c:439)
14894 ==22787== by 0x4256E4: main (main.c:287)
14895
14896 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
14897 Reported-by: Zdenek Kabelac <zdenek.kabelac@gmail.com>
14898 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=36108
14899 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
14900 Signed-off-by: Keith Packard <keithp@keithp.com>
14901
14902commit 59c2c4f645b4d782599c274f4bc752de9623d308
14903Author: Daniel Stone <daniel@fooishbar.org>
14904Date: Tue Jul 10 09:10:22 2012 +0100
14905
14906 AllocDevicePair: Ensure XKB privates are initialised
14907
14908 Since we call directly into XKB and may be doing so before the extension
14909 has been initialised, make sure its privates are set up first. XTest
14910 had a hack to do this itself, but seems cleaner to just make sure we do
14911 it in AllocDevicePair.
14912
14913 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
14914 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
14915 Signed-off-by: Keith Packard <keithp@keithp.com>
14916
14917commit f937c3d50157a9a42d2f2f10590b6b139ce18a29
14918Author: Daniel Stone <daniel@fooishbar.org>
14919Date: Tue Jul 10 09:10:21 2012 +0100
14920
14921 Sync: Always initialise system counter list
14922
14923 Sync is designed to let you add system counters before the extension has
14924 been initialised, which means the system counter list may well be full
14925 of bees. Make sure it's initialised before we add to it, to avoid the
14926 risk of fatal injury.
14927
14928 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
14929 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14930 Signed-off-by: Keith Packard <keithp@keithp.com>
14931
14932commit 34cf559bcf99dad550527b5ff53f247f0e8e73ee
14933Author: Keith Packard <keithp@keithp.com>
14934Date: Tue Jul 10 15:58:48 2012 -0700
14935
14936 ProcGetPointerMapping uses rep.nElts before it is initialized
14937
14938 In:
14939
14940 commit d792ac125a0462a04a930af543cbc732f8cdab7d
14941 Author: Alan Coopersmith <alan.coopersmith@oracle.com>
14942 Date: Mon Jul 9 19:12:43 2012 -0700
14943
14944 Use C99 designated initializers in dix Replies
14945
14946 the initializer for the .length element of the xGetPointerMappingReply
14947 structure uses the value of rep.nElts, but that won't be set until
14948 after this initializer runs, so we get garbage in the length element
14949 and clients using it will generally wedge.
14950
14951 Easy to verify:
14952
14953 $ xmodmap -pp
14954
14955 Fixed by creating a local nElts variable and using that.
14956
14957 Signed-off-by: Keith Packard <keithp@keithp.com>
14958 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
14959
14960commit 44bd27cdd13b6e6c4e6fd3873361e3308ef25ffe
14961Author: Keith Packard <keithp@keithp.com>
14962Date: Tue Jul 10 01:14:13 2012 -0700
14963
14964 Version 1.12.99.901 (1.13 RC1)
14965
14966 Signed-off-by: Keith Packard <keithp@keithp.com>
14967
14968commit 6e12cb147da58bd12a158807b2cd5afac32ae70f
14969Merge: 532fbc2 0231279
14970Author: Keith Packard <keithp@keithp.com>
14971Date: Tue Jul 10 00:52:11 2012 -0700
14972
14973 Merge branch 'local-fixes'
14974
14975commit 532fbc29c900469ef4867ccb7eabb953019dcab5
14976Author: Daniel Stone <daniel@fooishbar.org>
14977Date: Tue Jul 10 02:03:38 2012 +0100
14978
14979 GE: Remove unused RT_GECLIENT
14980
14981 Remove remnants of an earlier experiment which had the GE extension
14982 handling event delivery directly. Nothing's used the resource since, so
14983 purge it.
14984
14985 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
14986 Reviewed-by: Adam Jackson <ajax@redhat.com>
14987 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
14988 Signed-off-by: Keith Packard <keithp@keithp.com>
14989
14990commit 0e70b333d4846b166bf4a0a0c5d1a80623ba08dc
14991Author: Daniel Stone <daniel@fooishbar.org>
14992Date: Tue Jul 10 02:03:37 2012 +0100
14993
14994 XFree86: os-support: Remove unused xf86MakeNewMapping
14995
14996 No drivers used this, so it got unexported, and now it's so unused it
14997 got culled during the link. Take the poor function out behind the shed
14998 and put it out of its misery.
14999
15000 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15001 Reviewed-by: Adam Jackson <ajax@redhat.com>
15002 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
15003 Signed-off-by: Keith Packard <keithp@keithp.com>
15004
15005commit 43f2393746c44e9d35be6a2a3555205f8e1b4ddf
15006Author: Daniel Stone <daniel@fooishbar.org>
15007Date: Tue Jul 10 02:03:36 2012 +0100
15008
15009 Xinerama: Reset Composite request vector on close
15010
15011 Similar (identical) to how it interacts with Render and XFixes, also
15012 call PanoramiXCompositeReset() to restore the Composite dispatch table
15013 to how it was when it started, on reset.
15014
15015 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15016 Reviewed-by: Adam Jackson <ajax@redhat.com>
15017 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
15018 Signed-off-by: Keith Packard <keithp@keithp.com>
15019
15020commit 766a403bc7ff249fdc6b9049329b633e73738329
15021Author: Daniel Stone <daniel@fooishbar.org>
15022Date: Tue Jul 10 02:03:35 2012 +0100
15023
15024 XvMC: Remove unused XvMCGeneration
15025
15026 I wonder if this even works across multiple generations.
15027
15028 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15029 Reviewed-by: Adam Jackson <ajax@redhat.com>
15030 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
15031 Signed-off-by: Keith Packard <keithp@keithp.com>
15032
15033commit 7cef23a6dd7a5d76983cf20979e845ea6ae63a10
15034Author: Daniel Stone <daniel@fooishbar.org>
15035Date: Tue Jul 10 02:03:34 2012 +0100
15036
15037 Render: Remove unused glyphDepths
15038
15039 No-one has used this since 0a71e154.
15040
15041 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15042 Reviewed-by: Adam Jackson <ajax@redhat.com>
15043 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
15044 Signed-off-by: Keith Packard <keithp@keithp.com>
15045
15046commit 65011064d7e70aed59fb716f9ed3d76d196244ce
15047Author: Daniel Stone <daniel@fooishbar.org>
15048Date: Tue Jul 10 02:03:33 2012 +0100
15049
15050 Xi: Remove dead Device{Enter,Leave}WindowMask
15051
15052 These were an unused remnant of earlier MPX work; their only users got
15053 cleared out in dc153271, but the mask declarations remained. Remove
15054 them, and move DevicePropertyNotify's mask up to be contiguous with the
15055 rest of the range.
15056
15057 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15058 Reviewed-by: Adam Jackson <ajax@redhat.com>
15059 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15060 Signed-off-by: Keith Packard <keithp@keithp.com>
15061
15062commit fadfc2947075c832cd1107d2c0866ad13fa756c4
15063Author: Daniel Stone <daniel@fooishbar.org>
15064Date: Tue Jul 10 02:03:32 2012 +0100
15065
15066 GLX: Remove unused noGlxVisualInit
15067
15068 No-one ever did anything with this variable except assign its default
15069 value to it.
15070
15071 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15072 Reviewed-by: Adam Jackson <ajax@redhat.com>
15073 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
15074 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
15075 Signed-off-by: Keith Packard <keithp@keithp.com>
15076
15077commit 06e4ba8b26fb974bdd84144b3e002f0f7589e7f7
15078Author: Daniel Stone <daniel@fooishbar.org>
15079Date: Tue Jul 10 02:03:31 2012 +0100
15080
15081 XKB: Geom: Remove unused code
15082
15083 These codepaths were never called by anyone. Shame there weren't more
15084 of them.
15085
15086 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15087 Reviewed-by: Adam Jackson <ajax@redhat.com>
15088 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15089 Signed-off-by: Keith Packard <keithp@keithp.com>
15090
15091commit e191e296e6e7861978ea4a0ae9aa7b780e52732b
15092Author: Daniel Stone <daniel@fooishbar.org>
15093Date: Mon Jul 9 20:04:15 2012 +0100
15094
15095 Remove XAA
15096
15097 Commit 0c6987df in June 2008 disabled XAA offscreen pixmaps per default,
15098 as they were broken, leaving XAA only able to accelerate operations
15099 directly on the screen pixmap and nowhere else, eliminating acceleration
15100 for basically every modern toolkit, and any composited environment.
15101
15102 So, it hasn't worked for over four years. No-one's even come close to
15103 fixing it.
15104
15105 RIP.
15106
15107 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15108 Reviewed-by: Dave Airlie <airlied@redhat.com>
15109 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
15110
15111commit 66362441867f68dac97de33744bd0ca478fb65d3
15112Author: Daniel Stone <daniel@fooishbar.org>
15113Date: Tue Jul 10 02:03:29 2012 +0100
15114
15115 Remove last renmants of XFree86LOADER define
15116
15117 Not to be confused with XFree86Loader or XorgLoader. Which are both now
15118 dead too.
15119
15120 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15121 Reviewed-by: Keith Packard <keithp@keithp.com>
15122 Signed-off-by: Keith Packard <keithp@keithp.com>
15123
15124commit a15dac550918c55b08bbe0007f068cc5096ebaa1
15125Author: Daniel Stone <daniel@fooishbar.org>
15126Date: Tue Jul 10 02:03:28 2012 +0100
15127
15128 Remove unused setupFunc from extensions
15129
15130 setupFunc was used as an early callback for half-modular extensions such
15131 as Xv, XvMC and DGA to set up hooks between the core server and the
15132 modular component. Now we've rid ourselves of that, we can also bin
15133 setupFunc.
15134
15135 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15136 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15137 Signed-off-by: Keith Packard <keithp@keithp.com>
15138
15139commit 710318e803c6d5c6b8ae170d38c8253f5d780472
15140Author: Daniel Stone <daniel@fooishbar.org>
15141Date: Tue Jul 10 02:03:27 2012 +0100
15142
15143 Replace NO_HW_ONLY_EXTS with Xorg DDX test
15144
15145 Turns out the only thing we use NO_HW_ONLY_EXTS for is to check whether
15146 or not we're building inside the Xorg DDX. Replace it with an
15147 XorgLoader test instead, and remove all its users.
15148
15149 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15150 Reviewed-by: Keith Packard <keithp@keithp.com>
15151 Signed-off-by: Keith Packard <keithp@keithp.com>
15152
15153commit 5f5bbbe543f65c48ecbb5cce80116a86ca3fbe86
15154Author: Daniel Stone <daniel@fooishbar.org>
15155Date: Tue Jul 10 02:03:26 2012 +0100
15156
15157 Unify miinitext.c
15158
15159 Rather than having a non-Xorg and an Xorg-specific path which basically
15160 just duplicated each other for no reason, we could ... just have one.
15161
15162 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15163 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15164 Signed-off-by: Keith Packard <keithp@keithp.com>
15165
15166commit 147967f67d4976a87ed07fe11b59b174907f5d43
15167Author: Daniel Stone <daniel@fooishbar.org>
15168Date: Tue Jul 10 02:03:25 2012 +0100
15169
15170 Remove Xorg-specific extensions from non-Xorg miinitext
15171
15172 miinitext.c had a completely separate codepath for non-Xorg servers,
15173 which included tests for Xorg-specific extensions such as
15174 XFree86-VidMode, which were external even to the Xorg DDX. So we can
15175 just remove them, and the associated #undefs.
15176
15177 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15178 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15179 Signed-off-by: Keith Packard <keithp@keithp.com>
15180
15181commit 27a624bee66d2797eacf90317054207dcf248028
15182Author: Daniel Stone <daniel@fooishbar.org>
15183Date: Tue Jul 10 02:03:24 2012 +0100
15184
15185 Quartz: Move PseudoramiX initialisation to DDX
15186
15187 As PseudoramiX is a DDX-specific extension, move its loading and
15188 initialisation to hw/xquartz. This creates a QuartzExtensionInit()
15189 similar in spirit to xf86ExtensionInit.
15190
15191 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15192 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
15193 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
15194 Signed-off-by: Keith Packard <keithp@keithp.com>
15195
15196commit 79d14d9fc3d92ba866864d3a5910a8c6c8e9b64f
15197Author: Daniel Stone <daniel@fooishbar.org>
15198Date: Tue Jul 10 02:03:23 2012 +0100
15199
15200 Move the remnants of loadext.c to miinitext.c
15201
15202 There was nothing XFree86-specific or loader-specific about this, aside
15203 from using xf86MsgVerb instead of ErrorF.
15204
15205 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15206 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15207 Signed-off-by: Keith Packard <keithp@keithp.com>
15208
15209commit a089af3477dd01a831b89c06f982679c60e42975
15210Author: Daniel Stone <daniel@fooishbar.org>
15211Date: Tue Jul 10 02:03:22 2012 +0100
15212
15213 Loader: Move ExtensionModule types to DIX
15214
15215 In preparation for gutting loadext.c, move the ExtensionModule struct to
15216 the DIX, and unexport ExtensionModuleList (why, why, why, why was this
15217 ever exported in the first place, tbqh).
15218
15219 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15220 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15221 Signed-off-by: Keith Packard <keithp@keithp.com>
15222
15223commit 8171108602556446f12c1cfbc37c6abf3b0195eb
15224Author: Daniel Stone <daniel@fooishbar.org>
15225Date: Tue Jul 10 02:03:21 2012 +0100
15226
15227 Loader: Remove extension initialisation sorting
15228
15229 Extensions could previously declare initialisation dependencies on other
15230 extensions, which would then get nicely sorted by the loader. We only
15231 had one user for this, GLX, which had one pointless (Composite) and one
15232 possibly useful dependency (DBE). As DBE is now a built-in, it will
15233 always be sorted by GLX, so we no longer have any users for it.
15234
15235 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15236 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15237 Signed-off-by: Keith Packard <keithp@keithp.com>
15238
15239commit d52ab85c7ef5058dcbd8a75eb84058908dc22906
15240Author: Daniel Stone <daniel@fooishbar.org>
15241Date: Tue Jul 10 02:03:20 2012 +0100
15242
15243 GLX: Remove extension init dependencies
15244
15245 GLX was the only user of extension init order dependencies, using them
15246 to depend on Composite, which has always been built-in anyway, and DBE,
15247 which is now built-in.
15248
15249 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15250 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15251 Signed-off-by: Keith Packard <keithp@keithp.com>
15252
15253commit 2fba9445a0357f67641e41ac334b5529c37774a2
15254Author: Daniel Stone <daniel@fooishbar.org>
15255Date: Tue Jul 10 02:03:19 2012 +0100
15256
15257 Add static extensions before those in modules
15258
15259 Make sure we add static extensions before anything in a module. This is
15260 more or less a no-op at the moment, but will come in handy later when
15261 extension dependency sorting is removed.
15262
15263 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15264 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15265 Signed-off-by: Keith Packard <keithp@keithp.com>
15266
15267commit 9a953e0e9dcb8a8e43cc27ffaef460268fbe1916
15268Author: Daniel Stone <daniel@fooishbar.org>
15269Date: Tue Jul 10 02:03:18 2012 +0100
15270
15271 Move DRI2 from external module to built-in
15272
15273 Instead of keeping a tiny amount of code in an external module, just man
15274 up and build it into the core server.
15275
15276 v2: Fix test/Makefile.am to only link libdri2.la if DRI2 is set
15277
15278 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15279 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15280 Signed-off-by: Keith Packard <keithp@keithp.com>
15281
15282commit b8a3267c36e2e335b888bd4f2ef2f2c477cdfdce
15283Author: Daniel Stone <daniel@fooishbar.org>
15284Date: Tue Jul 10 02:03:17 2012 +0100
15285
15286 DRI2: Remove prototype for DRI2DestroyDrawable
15287
15288 DRI2DestroyDrawable() was still being _X_EXPORTed, but hasn't existed
15289 since 1da1f33f last year.
15290
15291 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15292 Reviewed-by: Cyril Brulebois <kibi@debian.org>
15293 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15294 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
15295 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15296 Signed-off-by: Keith Packard <keithp@keithp.com>
15297
15298commit 7025a909bfe4d1f5518ec6fc5f89e2c3fb6501fd
15299Author: Daniel Stone <daniel@fooishbar.org>
15300Date: Tue Jul 10 02:03:16 2012 +0100
15301
15302 XFree86: DRI: Don't use per-target CFLAGS
15303
15304 AM_CFLAGS will suffice, given we only have one target in this directory.
15305
15306 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15307 Reviewed-by: Cyril Brulebois <kibi@debian.org>
15308 Signed-off-by: Keith Packard <keithp@keithp.com>
15309
15310commit 459c6da0f907ba33d733c7e62a116184ba2f14e5
15311Author: Daniel Stone <daniel@fooishbar.org>
15312Date: Tue Jul 10 02:03:15 2012 +0100
15313
15314 Move DRI1 from external module to built-in
15315
15316 Rather than building the tiny amount of code required for XFree86-DRI as
15317 an external module, build it in if it's enabled at configure time.
15318
15319 v2: Fix test/Makefile.am to only link libdri.la if DRI is set
15320
15321 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15322 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15323 Signed-off-by: Keith Packard <keithp@keithp.com>
15324
15325 fixup for DRI1 move
15326
15327 Signed-off-by: Keith Packard <keithp@keithp.com>
15328
15329commit b905d87ba0a9b045ea41bb1ad36a4f9f1170a1b8
15330Author: Daniel Stone <daniel@fooishbar.org>
15331Date: Tue Jul 10 02:03:14 2012 +0100
15332
15333 Xext: Only build one library
15334
15335 Now that libXextmodule.la is both empty and unused, we can just build
15336 the one libXext.la for everyone, rather than having Xorg be special and
15337 unique.
15338
15339 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15340 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15341 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15342 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15343 Signed-off-by: Keith Packard <keithp@keithp.com>
15344
15345commit a7a2f9f66ddc27df96557093f3dd266e24eb653b
15346Author: Daniel Stone <daniel@fooishbar.org>
15347Date: Tue Jul 10 02:03:13 2012 +0100
15348
15349 Remove the last remnants of extmod
15350
15351 extmod was originally a big pointless module. Now it's an empty,
15352 pointless module. This commit makes it unexist.
15353
15354 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15355 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15356 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15357 Signed-off-by: Keith Packard <keithp@keithp.com>
15358
15359commit 6e74fdda42fb4bd80fd48e1ae6471bffbb837804
15360Author: Daniel Stone <daniel@fooishbar.org>
15361Date: Wed Jun 22 00:28:18 2011 +0100
15362
15363 Move XFree86-VidMode from extmod to built-in
15364
15365 As with DGA, move VidMode from being part of extmod to a built-in part
15366 of the server, if compiled as such. This is initialised from
15367 xf86ExtensionInit rather than miinitext because it's wholly dependent on
15368 the Xorg DDX.
15369
15370 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15371 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15372 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15373
15374commit 60f53e3012be795eee741ecd4be858552c5e86c8
15375Author: Tomas Carnecky <tom@dbservice.com>
15376Date: Tue Jul 10 02:03:11 2012 +0100
15377
15378 DGA: Remove excessive module-induced indirection
15379
15380 The DGA event base used to have to be passed through a function pointer,
15381 as the code was cleaved in two with half in a module, and half in the
15382 core server. Now that's not the case, just access DGAEventBase
15383 directly.
15384
15385 v2: Deal with Alan's event initialization cleanups
15386
15387 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15388 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15389 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15390 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15391 Signed-off-by: Keith Packard <keithp@keithp.com>
15392
15393commit 7a5880bc3b47e1f90a638f056ad4d40be6931977
15394Author: Daniel Stone <daniel@fooishbar.org>
15395Date: Tue Jul 10 02:03:10 2012 +0100
15396
15397 Move DGA from extmod to built-in
15398
15399 Rather than leave DGA languishing in extmod, move it to be a built-in
15400 extension. As it's quite specific to the Xorg DDX, just move it
15401 sideways to the rest of the DGA code in hw/xfree86/common, and
15402 initialise it from xf86ExtensionInit, rather than miinitext.
15403
15404 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15405 Acked-by: Keith Packard <keithp@keithp.com>
15406 Signed-off-by: Keith Packard <keithp@keithp.com>
15407
15408commit 6fb481d1251626456914d0012d6ca5a342bfdf71
15409Author: Daniel Stone <daniel@fooishbar.org>
15410Date: Tue Jul 10 02:03:09 2012 +0100
15411
15412 Move SELinux from extmod to built-in
15413
15414 Instead of letting it languish in extmod just because we want to
15415 configure bits of it from xf86, move XSELinux to the builtin part of
15416 Xext, and do its configuration from xf86ExtensionInit.
15417
15418 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15419 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15420 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15421 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15422 Signed-off-by: Keith Packard <keithp@keithp.com>
15423
15424commit 5d92ee4081a23a6b4d1f9ba2a98985f539f99312
15425Author: Tomas Carnecky <tom@dbservice.com>
15426Date: Tue Jul 10 02:03:08 2012 +0100
15427
15428 Xv: Remove excessive module-induced indirection
15429
15430 Xv used to call XvScreenInit and co. through function pointers, as
15431 XvScreenInit may have been sitting on the other side of a module
15432 boundary from xf86XvScreenInit. Why this was so is a mystery, but make
15433 it not so any more.
15434
15435 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15436 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15437 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
15438 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15439 Signed-off-by: Keith Packard <keithp@keithp.com>
15440
15441commit 2e6c5f959153cdcea4bcf719621ebec408250f54
15442Author: Daniel Stone <daniel@fooishbar.org>
15443Date: Tue Jul 10 02:03:07 2012 +0100
15444
15445 XFree86: sdksyms: Remove unused -DXorgLoader
15446
15447 We no longer have anything in the tree that checks for XorgLoader. This
15448 was a fairly monumental hack: xvdi.h used to hide all its functions
15449 behind #ifndef XorgLoader, solely to avoid sdksyms.sh picking up its
15450 symbols, as it was previously a module rather than built-in.
15451
15452 This is no longer the case, so we can remove the define.
15453
15454 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15455 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15456 Signed-off-by: Keith Packard <keithp@keithp.com>
15457
15458commit 7a11b817e770cd2196814f1ac6264c6d87b76c6a
15459Author: Tomas Carnecky <tom@dbservice.com>
15460Date: Tue Jul 10 02:03:06 2012 +0100
15461
15462 Move Xv and XvMC from extmod to built-in
15463
15464 Always build these extensions into the core server, rather than letting
15465 them languish in extmod.
15466
15467 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15468 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15469 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15470 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15471 Signed-off-by: Keith Packard <keithp@keithp.com>
15472
15473commit 7d859bd87834dd79c7fa3792075496ece698c082
15474Author: Tomas Carnecky <tom@dbservice.com>
15475Date: Tue Jul 10 02:03:05 2012 +0100
15476
15477 Move XRes from extmod to built-in
15478
15479 Always build XRes support into the core server, rather than letting it
15480 languish in extmod.
15481
15482 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15483 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15484 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15485 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15486 Signed-off-by: Keith Packard <keithp@keithp.com>
15487
15488commit ba21fc29587e4f11320b88a70433beb1fe22cdf3
15489Author: Tomas Carnecky <tom@dbservice.com>
15490Date: Tue Jul 10 02:03:04 2012 +0100
15491
15492 Move DPMS from extmod to built-in
15493
15494 Always build DPMS support into the core server, rather than letting it
15495 languish in extmod.
15496
15497 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15498 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15499 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15500 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15501 Signed-off-by: Keith Packard <keithp@keithp.com>
15502
15503commit 3ed2c6e11298c4299042cfe4578f6cc02e441d58
15504Author: Tomas Carnecky <tom@dbservice.com>
15505Date: Tue Jul 10 02:03:03 2012 +0100
15506
15507 Move MIT-SCREEN-SAVER from extmod to built-in
15508
15509 If we've built MIT-SCREEN-SAVER support, then just build it into the
15510 main binary, rather than leaving it in extmod.
15511
15512 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15513 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15514 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15515 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15516 Signed-off-by: Keith Packard <keithp@keithp.com>
15517
15518commit b8c9ab0fea0be299d9f48ac0c4ed0474104a7556
15519Author: Tomas Carnecky <tom@dbservice.com>
15520Date: Tue Jul 10 02:03:02 2012 +0100
15521
15522 Move RECORD from external module to built-in
15523
15524 Rather than languishing in its own special module, move RECORD into the
15525 core server.
15526
15527 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15528 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15529 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15530 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15531 Signed-off-by: Keith Packard <keithp@keithp.com>
15532
15533commit bf61bf69b2b4f9e9993ca771de100079c7696bea
15534Author: Tomas Carnecky <tom@dbservice.com>
15535Date: Tue Jul 10 02:03:01 2012 +0100
15536
15537 Move DBE from an external module to built-in
15538
15539 If DBE support is compiled in the server, just man up and build it into
15540 the server, rather than having it as an external module.
15541
15542 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15543 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15544 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15545 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15546 Signed-off-by: Keith Packard <keithp@keithp.com>
15547
15548commit d785368e0e171080fea5447c87e81c93f515ff21
15549Author: Daniel Stone <daniel@fooishbar.org>
15550Date: Tue Jul 10 02:03:00 2012 +0100
15551
15552 miinitext: Use ARRAY_SIZE rather than sentinel
15553
15554 NULL sentinels are totally lame.
15555
15556 v2: numExtensionModules isn't public, so keep using the provided
15557 sentinal - Keith Packard <keithp@keithp.com>
15558
15559 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15560 Reviewed-by: Stéphane Marchesin <stephane.marchesin@gmail.com>
15561 Signed-off-by: Keith Packard <keithp@keithp.com>
15562
15563commit f48d8f58b39af0036365368d6ac35319d6d856d5
15564Author: Daniel Stone <daniel@fooishbar.org>
15565Date: Tue Jul 10 02:02:59 2012 +0100
15566
15567 extmod: Use ARRAY_SIZE rather than sentinel
15568
15569 When the array gets down to size zero (which it does in later patches),
15570 gcc complains that the index is out of bounds. Avoid this by using
15571 ARRAY_SIZE on extensionModules instead.
15572
15573 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15574 Reviewed-by: Stéphane Marchesin <stephane.marchesin@gmail.com>
15575 Signed-off-by: Keith Packard <keithp@keithp.com>
15576
15577commit d35884da2f5a250ff6dd5131dc23ac629eccdd0c
15578Author: Daniel Stone <daniel@fooishbar.org>
15579Date: Tue Jul 10 02:02:58 2012 +0100
15580
15581 Add xf86ExtensionInit for DDX extension configuration
15582
15583 xf86ExtensionInit is called after configuration file parsing, so it can
15584 perform the two parts of extension initialisation currently done by
15585 extmod: enabling and disabling of extensions through an 'omit' option,
15586 and SELinux configuration.
15587
15588 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15589 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15590 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15591 Signed-off-by: Keith Packard <keithp@keithp.com>
15592
15593commit 4170e4e1f8e879394c896c490d9f08a38ecfdefd
15594Author: Daniel Stone <daniel@fooishbar.org>
15595Date: Tue Jul 10 02:02:57 2012 +0100
15596
15597 Loader: Drop EXTERN_MODULE flag
15598
15599 EXTERN_MODULE was used to specify that we shouldn't worry about modules
15600 lacking a ModuleData object. It was also completely unused. *shrug*
15601
15602 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15603 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15604 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15605 Signed-off-by: Keith Packard <keithp@keithp.com>
15606
15607commit a1d41e311c21eb6627caa0d168e070ceaf90806f
15608Author: Daniel Stone <daniel@fooishbar.org>
15609Date: Tue Jul 10 02:02:56 2012 +0100
15610
15611 Move extension initialisation prototypes into extinit.h
15612
15613 Create extinit.h (and xf86Extensions.h, for Xorg-specific extensions) to
15614 hold all our extension initialisation prototypes, rather than
15615 duplicating them everywhere.
15616
15617 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15618 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15619 Signed-off-by: Keith Packard <keithp@keithp.com>
15620
15621commit b86aa74cafddf9bca5addfeb458a23f46ddcf132
15622Author: Daniel Stone <daniel@fooishbar.org>
15623Date: Tue Jul 10 02:02:55 2012 +0100
15624
15625 GLX: Insert swrast provider from GlxExtensionInit
15626
15627 Rather than making poor old miinitext.c do it, including making DMX
15628 have fake symbols just to keep it happy.
15629
15630 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15631 Reviewed-by: Stéphane Marchesin <stephane.marchesin@gmail.com>
15632 Signed-off-by: Keith Packard <keithp@keithp.com>
15633
15634commit 13552a5c4701097a1507a99ed8366ab1438844b6
15635Author: Daniel Stone <daniel@fooishbar.org>
15636Date: Tue Jul 10 02:02:54 2012 +0100
15637
15638 Reorder extension initialisation for non-Xorg
15639
15640 Reorder static extension initialisation in miinitext for non-Xorg
15641 servers to match Xorg's order.
15642
15643 Tested with Xephyr; checked that the extension list was identical before
15644 and after.
15645
15646 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15647 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15648 Signed-off-by: Keith Packard <keithp@keithp.com>
15649
15650commit 5079db78aea8fa42e781dd876c1ee1c31571b3ae
15651Author: Tomas Carnecky <tom@dbservice.com>
15652Date: Tue Jul 10 02:02:53 2012 +0100
15653
15654 Replace INITARGS with void
15655
15656 INITARGS was a hardcoded define to void. Since knowing the function
15657 signature for your extensions is kinda useful, just replace it with a
15658 hardcoded void, but leave the define there for API compatibility.
15659
15660 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
15661 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
15662 Reviewed-by: Cyril Brulebois <kibi@debian.org>
15663 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15664 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15665 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15666 Signed-off-by: Keith Packard <keithp@keithp.com>
15667
15668commit bddb8c6cbe52ba9923b1d36d01f5ac9391b0ec0e
15669Author: Daniel Stone <daniel@fooishbar.org>
15670Date: Tue Jul 10 02:02:52 2012 +0100
15671
15672 Xinerama: Fix ExtensionInit prototype
15673
15674 Huh, so I guess INITARGS used to be int argc, char *argv then. Either
15675 way, it's now void, so fix that ...
15676
15677 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15678 Reviewed-by: Cyril Brulebois <kibi@debian.org>
15679 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15680 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15681 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15682 Signed-off-by: Keith Packard <keithp@keithp.com>
15683
15684commit eb9661fef9f59587f898371f97a0952ac14d125f
15685Author: Daniel Stone <daniel@fooishbar.org>
15686Date: Tue Jul 10 02:02:51 2012 +0100
15687
15688 Make extension.h self-contained, remove C++ externs
15689
15690 externsion.h required bits from Xfuncproto.h and dixstruct.h, but
15691 included neither; fix that.
15692
15693 It also had _XFUNCPROTOBEGIN and _XFUNCPROTOEND wrappers, which is a bit
15694 pointless for a server-only library, as it's only needed for C++.
15695
15696 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15697 Reviewed-by: Cyril Brulebois <kibi@debian.org>
15698 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15699 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15700 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15701 Signed-off-by: Keith Packard <keithp@keithp.com>
15702
15703commit 854c1fa4a1e90288d1e2f3777fac3e1a9acb1235
15704Author: Daniel Stone <daniel@fooishbar.org>
15705Date: Tue Jul 10 02:02:50 2012 +0100
15706
15707 Add a common ARRAY_SIZE macro to dix.h
15708
15709 Does what it says on the box, replacing those from Xi/ and glx/.
15710
15711 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15712 Reviewed-by: Cyril Brulebois <kibi@debian.org>
15713 Reviewed-by: Jamey Sharp <jamey@minilop.net>
15714 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15715 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
15716 Signed-off-by: Keith Packard <keithp@keithp.com>
15717
15718commit 67953d6975ef2ee573b95e7641aaf3d72e9f8379
15719Author: Daniel Stone <daniel@fooishbar.org>
15720Date: Tue Jul 10 02:02:49 2012 +0100
15721
15722 Xorg: Link XKB DDX library after core server libs
15723
15724 libxorgxkb.a contains a number of libraries which are used by XKB action
15725 code to call back into the DDX, e.g. for VT switching, termination, grab
15726 breaking, et al. Make sure libxkb.a comes first in the link order, so
15727 it can mark XkbDDX* as used in order for the linker to not discard them.
15728
15729 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15730 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15731 Signed-off-by: Keith Packard <keithp@keithp.com>
15732
15733commit 656af2c7e73d948338683675e14cd827b46a97c4
15734Author: Daniel Stone <daniel@fooishbar.org>
15735Date: Tue Jul 10 02:02:46 2012 +0100
15736
15737 Don't make failure to -nolisten fatal
15738
15739 If failing to disable a protocol specified by -nolisten failed, we'd
15740 throw a FatalError and bomb startup entirely. From poking at xtrans, it
15741 looks like the only way we can get a failure here is because we've
15742 specified a protocol name which doesn't exist, which probably doesn't
15743 constitute a security risk.
15744
15745 And it makes it possible to start gdm even though you've built with
15746 --disable-tcp-transport.
15747
15748 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15749 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15750 Signed-off-by: Keith Packard <keithp@keithp.com>
15751
15752commit dc0c2dd92d5eb314582ea10830678753824375d9
15753Author: Daniel Stone <daniel@fooishbar.org>
15754Date: Tue Jul 10 02:02:45 2012 +0100
15755
15756 Xext: Add hashtable.h to sources to fix distcheck
15757
15758 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
15759 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15760 Signed-off-by: Keith Packard <keithp@keithp.com>
15761
15762commit ad4092cf7d59a89b1b2922440eef65be5c0c5ebd
15763Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15764Date: Mon Jul 9 19:12:44 2012 -0700
15765
15766 Replace padlength tables with inline functions from misc.h
15767
15768 Adds new function padding_for_int32() and uses existing pad_to_int32()
15769 depending on required results.
15770
15771 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15772 Reviewed-by: Keith Packard <keithp@keithp.com>
15773 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
15774 Tested-by: Daniel Stone <daniel@fooishbar.org>
15775
15776commit 2b1c1300cc23912ee1c59f8dde938dd4d7287f4a
15777Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15778Date: Mon Jul 9 19:12:44 2012 -0700
15779
15780 ephyrGLXQueryServerString: Stop making an unused copy of server_string
15781
15782 ephyrGLXQueryServerString() carefully allocated a buffer padded to the
15783 word-aligned string length for sending to the client, copied the string
15784 to it, and then forgot to use it, potentially reading a few bytes of
15785 garbage past the end of the server_string buffer.
15786
15787 Since WriteToClient already handles the necessary padding, just send
15788 it the actual length of the original server_string, and don't bother
15789 making a padded copy.
15790
15791 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15792 Reviewed-by: Keith Packard <keithp@keithp.com>
15793 Tested-by: Daniel Stone <daniel@fooishbar.org>
15794
15795commit 7a29f6878284e2d65e69fcd157aa9ec01d21b3c0
15796Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15797Date: Mon Jul 9 19:12:44 2012 -0700
15798
15799 Initialize padding bits to 0 in ErrorConnMax()
15800
15801 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15802 Reviewed-by: Keith Packard <keithp@keithp.com>
15803 Tested-by: Daniel Stone <daniel@fooishbar.org>
15804
15805commit bed610fcae41ddfe21fa9acde599b17d1d15f5d1
15806Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15807Date: Mon Jul 9 19:12:44 2012 -0700
15808
15809 Set padding bytes to 0 in WriteToClient
15810
15811 Clear them out when needed instead of leaving whatever values were
15812 present in previously sent messages.
15813
15814 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15815 Reviewed-by: Keith Packard <keithp@keithp.com>
15816 Tested-by: Daniel Stone <daniel@fooishbar.org>
15817
15818commit 1622dd8ab24d2fe7b9908110eb237774d4e03568
15819Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15820Date: Mon Jul 9 19:12:44 2012 -0700
15821
15822 Use C99 designated initializers in dix registry
15823
15824 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15825 Reviewed-by: Keith Packard <keithp@keithp.com>
15826 Tested-by: Daniel Stone <daniel@fooishbar.org>
15827
15828commit 9805cedf7b0f76d3b75f94e956c4cc2dcf0d8b64
15829Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15830Date: Mon Jul 9 19:12:44 2012 -0700
15831
15832 Use C99 designated initializers in extension Events
15833
15834 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15835 Reviewed-by: Keith Packard <keithp@keithp.com>
15836 Tested-by: Daniel Stone <daniel@fooishbar.org>
15837
15838commit 0af79b124e1317c36d1613d28755c5a8ce612e2a
15839Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15840Date: Mon Jul 9 19:12:44 2012 -0700
15841
15842 Use C99 designated initializers in dix Events
15843
15844 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15845 Reviewed-by: Keith Packard <keithp@keithp.com>
15846 Tested-by: Daniel Stone <daniel@fooishbar.org>
15847
15848commit 483266a583b0fd8a552303f318a71502ce3ff1d2
15849Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15850Date: Mon Jul 9 19:12:43 2012 -0700
15851
15852 Use C99 designated initializers in xf86 extension Replies
15853
15854 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15855 Reviewed-by: Keith Packard <keithp@keithp.com>
15856 Tested-by: Daniel Stone <daniel@fooishbar.org>
15857
15858commit 14501fd33ee89acf9fc1d908003ed62f383d8bae
15859Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15860Date: Mon Jul 9 19:12:43 2012 -0700
15861
15862 Use C99 designated initializers in Xephyr Replies
15863
15864 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15865 Reviewed-by: Keith Packard <keithp@keithp.com>
15866 Tested-by: Daniel Stone <daniel@fooishbar.org>
15867
15868commit e4e827ec36fd20b7aea608db09790f76fb87e519
15869Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15870Date: Mon Jul 9 19:12:43 2012 -0700
15871
15872 Use C99 designated initializers in dmx Replies
15873
15874 v2: fix in __glXGetVisualConfigs to not re-declare local 'reply' variable
15875
15876 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15877 Reviewed-by: Keith Packard <keithp@keithp.com>
15878 Tested-by: Daniel Stone <daniel@fooishbar.org>
15879
15880 Fixup for Use C99 designated initializeres in dmx Replies
15881
15882 Signed-off-by: Keith Packard <keithp@keithp.com>
15883
15884commit bd6f948c41865c2c9d3fba1000bf5f7458d3afc1
15885Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15886Date: Mon Jul 9 19:12:43 2012 -0700
15887
15888 Use C99 designated initializers in randr Replies
15889
15890 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15891 Reviewed-by: Keith Packard <keithp@keithp.com>
15892 Tested-by: Daniel Stone <daniel@fooishbar.org>
15893
15894commit a406bd07593edb69285cf2fd91a6af4a5d956817
15895Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15896Date: Mon Jul 9 19:12:43 2012 -0700
15897
15898 Use C99 designated initializers in xkb Replies
15899
15900 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15901 Reviewed-by: Keith Packard <keithp@keithp.com>
15902 Tested-by: Daniel Stone <daniel@fooishbar.org>
15903
15904commit 26efa09d0c7ee2f6a18547205c84be75376e688b
15905Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15906Date: Mon Jul 9 19:12:43 2012 -0700
15907
15908 Use C99 designated initializers in glx Replies
15909
15910 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15911 Reviewed-by: Keith Packard <keithp@keithp.com>
15912 Tested-by: Daniel Stone <daniel@fooishbar.org>
15913
15914commit cc5f09c86f7bea23b7546c3491b2c52ce8100a71
15915Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15916Date: Mon Jul 9 19:12:43 2012 -0700
15917
15918 Use C99 designated initializers in various extension Replies
15919
15920 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15921 Reviewed-by: Keith Packard <keithp@keithp.com>
15922 Tested-by: Daniel Stone <daniel@fooishbar.org>
15923
15924commit 2f5caeaddb3616dc9ff57d784f7feba589c536e7
15925Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15926Date: Mon Jul 9 19:12:43 2012 -0700
15927
15928 Use C99 designated initializers in Xinput Replies
15929
15930 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15931 Reviewed-by: Keith Packard <keithp@keithp.com>
15932 Tested-by: Daniel Stone <daniel@fooishbar.org>
15933
15934commit 6a721e3af5e11bae1d59ca551a02470f35d9e844
15935Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15936Date: Mon Jul 9 19:12:43 2012 -0700
15937
15938 Use C99 designated initializers in Xext Replies
15939
15940 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15941 Reviewed-by: Keith Packard <keithp@keithp.com>
15942 Tested-by: Daniel Stone <daniel@fooishbar.org>
15943
15944commit d792ac125a0462a04a930af543cbc732f8cdab7d
15945Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15946Date: Mon Jul 9 19:12:43 2012 -0700
15947
15948 Use C99 designated initializers in dix Replies
15949
15950 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15951 Reviewed-by: Keith Packard <keithp@keithp.com>
15952 Tested-by: Daniel Stone <daniel@fooishbar.org>
15953
15954commit 69fa5630b5902aaad267fc67d0da4ca93625886a
15955Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15956Date: Mon Jul 9 19:12:43 2012 -0700
15957
15958 Use C99 designated initializers in SendErrorToClient
15959
15960 Let the compiler worry about 0-filling the rest of the fields,
15961 instead of memsetting the whole struct and then going back to
15962 overwrite some of the fields.
15963
15964 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15965 Reviewed-by: Keith Packard <keithp@keithp.com>
15966 Tested-by: Daniel Stone <daniel@fooishbar.org>
15967
15968commit cdf5bcd420e5bcf4a4a24a275d3133a4e16ce41e
15969Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15970Date: Mon Jul 9 19:12:42 2012 -0700
15971
15972 Use calloc to zero fill buffers being allocated for replies & events
15973
15974 Ensures padding bytes are zero-filled
15975
15976 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15977 Reviewed-by: Keith Packard <keithp@keithp.com>
15978 Tested-by: Daniel Stone <daniel@fooishbar.org>
15979
15980commit ef0f701c9245b0373eb0c7a84dd804af4738bb29
15981Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15982Date: Mon Jul 9 19:12:42 2012 -0700
15983
15984 xf86dga2.c & xf86vmode.c: Move REQUEST_SIZE_MATCH checks before using stuff
15985
15986 Seems silly waiting to check if the client failed to send us enough bytes
15987 until after we've already tried using them.
15988
15989 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
15990 Reviewed-by: Keith Packard <keithp@keithp.com>
15991 Tested-by: Daniel Stone <daniel@fooishbar.org>
15992
15993commit 2e739a8870d91fbac7f536ffa8f6d2cee0f66cec
15994Author: Alan Coopersmith <alan.coopersmith@oracle.com>
15995Date: Mon Jul 9 19:12:42 2012 -0700
15996
15997 ProcRRListOutputProperties: skip atom walk if the list is empty
15998
15999 pAtoms is only allocated if numProps was non-zero, so move the walk
16000 through the property list to copy atoms to it inside the if (numProps)
16001
16002 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16003 Reviewed-by: Keith Packard <keithp@keithp.com>
16004 Tested-by: Daniel Stone <daniel@fooishbar.org>
16005
16006commit 15bc13c8d088e05f14c7262348e0066929c29251
16007Author: Alan Coopersmith <alan.coopersmith@oracle.com>
16008Date: Mon Jul 9 19:12:42 2012 -0700
16009
16010 ProcRRGetScreenInfo: swap configTimestamp as well
16011
16012 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16013 Reviewed-by: Keith Packard <keithp@keithp.com>
16014 Tested-by: Daniel Stone <daniel@fooishbar.org>
16015
16016commit 3a013b8816bc0bb765c99e8a2bf2ea536dc62c3b
16017Author: Jaroslav Šmíd <jardasmid@gmail.com>
16018Date: Mon Jul 9 19:12:42 2012 -0700
16019
16020 Bug 51375: Xorg doesn't set status for RRGetOutputInfo
16021
16022 https://bugs.freedesktop.org/show_bug.cgi?id=51375
16023 https://bugs.freedesktop.org/attachment.cgi?id=63397
16024
16025 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16026 Reviewed-by: Keith Packard <keithp@keithp.com>
16027 Tested-by: Daniel Stone <daniel@fooishbar.org>
16028
16029commit 5b86c072d1d586ce040d8831a05cf97ff8b17821
16030Author: Alan Coopersmith <alan.coopersmith@oracle.com>
16031Date: Mon Jul 9 19:12:42 2012 -0700
16032
16033 Use temporary variables instead of parts of reply structures
16034
16035 When passing variable pointers to functions or otherwise doing long
16036 sequences to compute values for replies, create & use some new
16037 temporary variables, to allow for simpler initialization of reply
16038 structures in the following patches.
16039
16040 Move memsets & other initializations to group with the rest of the
16041 filling in of the reply structure, now that they're not needed so
16042 early in the code path.
16043
16044 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16045 Reviewed-by: Keith Packard <keithp@keithp.com>
16046 Tested-by: Daniel Stone <daniel@fooishbar.org>
16047
16048commit c2fb1a7b2ab58d70b38ee03ab2fdeb4e7183a356
16049Author: Alan Coopersmith <alan.coopersmith@oracle.com>
16050Date: Mon Jul 9 19:12:42 2012 -0700
16051
16052 ProcQueryKeymap: rework logic around permission to copy key states
16053
16054 Always initialize to zero, and then if permission is granted, copy
16055 the current key state maps, instead of always copying and then
16056 zeroing out if permission was denied.
16057
16058 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16059 Reviewed-by: Keith Packard <keithp@keithp.com>
16060 Tested-by: Daniel Stone <daniel@fooishbar.org>
16061
16062commit dccb0858d71341e1258800b37d1bb78c0115838d
16063Author: Alan Coopersmith <alan.coopersmith@oracle.com>
16064Date: Mon Jul 9 19:12:42 2012 -0700
16065
16066 Core events: invert check for permission to copy key states
16067
16068 Always initialize to zero, and then if permission is granted, copy
16069 the current key state maps. Use memcpy instead of memmove for the
16070 copy, since we're always copying to a newly allocated event on the
16071 stack, so guaranteed not to overlap with the device map structure.
16072
16073 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16074 Reviewed-by: Keith Packard <keithp@keithp.com>
16075 Tested-by: Daniel Stone <daniel@fooishbar.org>
16076
16077commit db69212df8a0bf09140368356d2d430c54afe346
16078Author: Alan Coopersmith <alan.coopersmith@oracle.com>
16079Date: Mon Jul 9 19:12:42 2012 -0700
16080
16081 Rework reply initialization in ProcGetProperty & NullPropertyReply
16082
16083 Don't need to pass an empty reply to NullPropertyReply, let it make
16084 it's own. Move reply initialization code in remaining replies in
16085 ProcGetProperty to group with the rest of the fields. (Prepares
16086 for coming C99 designated initializer conversion.)
16087
16088 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16089 Reviewed-by: Keith Packard <keithp@keithp.com>
16090 Tested-by: Daniel Stone <daniel@fooishbar.org>
16091
16092commit 6be74a9080e30fc502421cd438cd0c73fb8eb0b0
16093Author: Alan Coopersmith <alan.coopersmith@oracle.com>
16094Date: Mon Jul 9 19:12:42 2012 -0700
16095
16096 Fix more poorly indented/wrapped comments & code
16097
16098 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16099 Reviewed-by: Keith Packard <keithp@keithp.com>
16100 Tested-by: Daniel Stone <daniel@fooishbar.org>
16101
16102commit 789d64e19a3b3d98b88bc80f677e0c37bfb5c631
16103Author: Alan Coopersmith <alan.coopersmith@oracle.com>
16104Date: Sun May 13 00:03:35 2012 -0700
16105
16106 Remove unneccesary casts from WriteToClient calls
16107
16108 Casting return to (void) was used to tell lint that you intended
16109 to ignore the return value, so it didn't warn you about it.
16110
16111 Casting the third argument to (char *) was used as the most generic
16112 pointer type in the days before compilers supported C89 (void *)
16113 (except for a couple places it's used for byte-sized pointer math).
16114
16115 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
16116 Reviewed-by: Keith Packard <keithp@keithp.com>
16117 Tested-by: Daniel Stone <daniel@fooishbar.org>
16118
16119commit 023127915e6922bc53e4c768de760d8a4f25c07c
16120Author: Keith Packard <keithp@keithp.com>
16121Date: Mon Jul 9 16:34:39 2012 -0700
16122
16123 Reliably reset signals at server init time
16124
16125 Each DDX currently calls OsReleaseSIGIO in case it was suspended when
16126 the server regen started. This causes a BUG to occur if SIGIO was
16127 *not* blocked at that time. Instead of relying on each DDX, make the
16128 OS layer reliably reset all signal state at server init time, ensuring
16129 that signals are suitably unblocked and that the various signal state
16130 counting variables are set back to zero.
16131
16132 Signed-off-by: Keith Packard <keithp@keithp.com>
16133
16134commit d6756e0298e660a0aca58d75bcf79471ce07a634
16135Author: Keith Packard <keithp@keithp.com>
16136Date: Mon Jul 9 16:33:11 2012 -0700
16137
16138 xfree86: In InitOutput, only call OsReleaseSIGIO if OsBlockSIGIO was called
16139
16140 Otherwise, OsReleaseSIGIO will complain, or perhaps something worse
16141 will happen (if SIGIO actually needs to be blocked here).
16142
16143 Signed-off-by: Keith Packard <keithp@keithp.com>
16144
16145commit ea8b04507e5464e4817791db516a20cfed2a6724
16146Author: Keith Packard <keithp@keithp.com>
16147Date: Mon Jul 9 16:30:24 2012 -0700
16148
16149 privates: Resize GPU screen-specific privates too
16150
16151 When allocating new global privates, make sure the gpu
16152 screens get their private offsets updated.
16153
16154 This only affects GPU screens that enumerate before the non-GPU
16155 screens, which generally requires that the related device be present
16156 when the system boots so that it can get an earlier DRM filename.
16157
16158 Signed-off-by: Keith Packard <keithp@keithp.com>
16159
16160commit d1c639c006526b8cab14dac582508f3f54848967
16161Author: Keith Packard <keithp@keithp.com>
16162Date: Mon Jul 9 16:29:04 2012 -0700
16163
16164 Add 'install-headers' target in the top-level Makefile
16165
16166 This target recursively locates directories with sdk headers and
16167 installs them all. Useful when you want to build a complete new X
16168 server and drivers without having to install the X server before the
16169 drivers are actually working.
16170
16171 Signed-off-by: Keith Packard <keithp@keithp.com>
16172
16173commit 329db3292223cccd4887062956622285c45a1523
16174Author: Dave Airlie <airlied@redhat.com>
16175Date: Sat Jul 7 10:10:08 2012 +0100
16176
16177 dri2/prime: allocate prime id at screen allocation time
16178
16179 Add a static mask of prime id and allocate them at screen time,
16180 if the driver supports the prime interfaces and is a gpu screen.
16181
16182 This is instead of them changing due to user controlled randr commands,
16183 as suggested by Keith.
16184
16185 Reviewed-by: Keith Packard <keithp@keithp.com>
16186 Signed-off-by: Dave Airlie <airlied@redhat.com>
16187
16188commit 30298012162de7f76e8a4c7b0362e98703f80011
16189Author: Dave Airlie <airlied@redhat.com>
16190Date: Tue Jun 26 10:20:52 2012 +0100
16191
16192 dri2: add initial prime support. (v1.2)
16193
16194 This adds the initial prime support for dri2 offload. The main thing is
16195 when we get a connection from a prime client, we stored the information
16196 and mark all drawables from that client as prime. We then create all
16197 buffers for that drawable on the prime device dri2screen.
16198
16199 Then DRI2UpdatePrime is provided which drivers can call to get a shared
16200 pixmap which they can use as the front buffer. The driver is then
16201 responsible for doing the back->front copy to the shared buffer.
16202
16203 prime requires a compositing manager be run, but it handles the case where
16204 a window get un-redirected by allocating a new pixmap and pointing the crtc
16205 at it while the client is in that state.
16206
16207 Currently prime can't handle pageflipping, so always does straight copy swap,
16208
16209 v1.1: renumber on top of master.
16210 v1.2: fix auth on top of master.
16211
16212 Reviewed-by: Keith Packard <keithp@keithp.com>
16213 Signed-off-by: Dave Airlie <airlied@redhat.com>
16214
16215commit 234022cfb3ad2a1b16ab7981ce69f9cd5ba0fbeb
16216Author: Dave Airlie <airlied@redhat.com>
16217Date: Sat Jul 7 10:20:58 2012 +0100
16218
16219 configure: bump dri2proto to 2.8
16220
16221 This is required for the prime macros.
16222
16223 Reviewed-by: Keith Packard <keithp@keithp.com>
16224 Signed-off-by: Dave Airlie <airlied@redhat.com>
16225
16226commit c41922940adbc8891575b3321fadf01ff4cb5854
16227Author: Dave Airlie <airlied@redhat.com>
16228Date: Tue Jun 26 10:00:21 2012 +0100
16229
16230 dix/randr: add a hook into screen to replace scanout pixmap
16231
16232 For DRI2 in some offload cases we need to set a new pixmap on the crtc,
16233 this hook allows dri2 to call into randr to do the necessary work to set
16234 a pixmap as the scanout pixmap for the crtc the drawable is currently on.
16235
16236 This is really only to be used for unredirected full screen apps in composited
16237 environments.
16238
16239 Reviewed-by: Keith Packard <keithp@keithp.com>
16240 Signed-off-by: Dave Airlie <airlied@redhat.com>
16241
16242commit e2fd447e76c4a1233374c9d8fa9cae54a55dad50
16243Author: Dave Airlie <airlied@redhat.com>
16244Date: Thu Jul 5 15:50:56 2012 +0100
16245
16246 xf86: add callback for offloak sink setting support.
16247
16248 This adds support for setting the offload sink to the xf86 ddx.
16249
16250 Reviewed-by: Keith Packard <keithp@keithp.com>
16251 Signed-off-by: Dave Airlie <airlied@redhat.com>
16252
16253commit 22f02120eb4092e1c283f81c8040f178693ff0a4
16254Author: Dave Airlie <airlied@redhat.com>
16255Date: Tue Jun 26 09:53:54 2012 +0100
16256
16257 xf86: store scanout pixmap in the xf86 struct as well.
16258
16259 This is so we can tell the scanout pixmap has changed between calls
16260 to the crtc set function.
16261
16262 Reviewed-by: Keith Packard <keithp@keithp.com>
16263 Signed-off-by: Dave Airlie <airlied@redhat.com>
16264
16265commit 426bc0a28edbe0e9153f692a02dd25f744ffa034
16266Author: Dave Airlie <airlied@redhat.com>
16267Date: Thu Jul 5 15:41:38 2012 +0100
16268
16269 randr: add hooks for offload sink provider protocol
16270
16271 This adds the protocol handler and associated providers handling
16272 for the offload slaves, it allows two providers to be connected as
16273 offload sink/source.
16274
16275 Reviewed-by: Keith Packard <keithp@keithp.com>
16276 Reviewed-by: Adam Jackson <ajax@redhat.com>
16277 Signed-off-by: Dave Airlie <airlied@redhat.com>
16278
16279commit 12905dfaf01088a00f4a0a78cffba03329e7b724
16280Author: Dave Airlie <airlied@redhat.com>
16281Date: Sat May 26 13:11:01 2012 +0100
16282
16283 dix/xf86: initial offload slave tracking (v1.1)
16284
16285 add the linked list and provider hooks.
16286
16287 v1.1: add another assert in the add path.
16288
16289 Reviewed-by: Keith Packard <keithp@keithp.com>
16290 Reviewed-by: Adam Jackson <ajax@redhat.com>
16291 Signed-off-by: Dave Airlie <airlied@redhat.com>
16292
16293commit a7c01da54ab43f9d29333ccbd79dfc001f9dc5e3
16294Author: Dave Airlie <airlied@redhat.com>
16295Date: Thu Jul 5 20:34:54 2012 +0100
16296
16297 xf86: make sure rotate calcs are done on the right screen boundaries
16298
16299 This fixes a segfault where this code believes we are outside the screen
16300 boundaries on a slave device, but we aren't.
16301
16302 Reviewed-by: Keith Packard <keithp@keithp.com>
16303 Reviewed-by: Adam Jackson <ajax@redhat.com>
16304 Signed-off-by: Dave Airlie <airlied@redhat.com>
16305
16306commit cc02f4ef3e407be794132d004c947c9a4e14f4c2
16307Author: Dave Airlie <airlied@redhat.com>
16308Date: Tue Jun 19 16:01:16 2012 +0100
16309
16310 xf86/cursor: fallback to sw cursor if we have slaves present.
16311
16312 Current USB devices have no hw rendered cursors, so we need the
16313 master GPU to render the cursor, so whenever we plug in a
16314 slave device, fallback to sw rendered cursors.
16315
16316 Reviewed-by: Adam Jackson <ajax@redhat.com>
16317 Signed-off-by: Dave Airlie <airlied@redhat.com>
16318
16319commit 98686512cbe1affb75222a63136854af79f1411b
16320Author: Dave Airlie <airlied@redhat.com>
16321Date: Thu Jul 5 15:50:24 2012 +0100
16322
16323 xf86: add output source setting callback (v2)
16324
16325 This adds support for the randr callback for setting the output source
16326 for a device.
16327
16328 v2: drop root clip change on detach
16329
16330 Reviewed-by: Keith Packard <keithp@keithp.com>
16331 Reviewed-by: Adam Jackson <ajax@redhat.com>
16332 Signed-off-by: Dave Airlie <airlied@redhat.com>
16333
16334commit 4c92eb00f97f7b8258de8e366226880382cc9ce9
16335Author: Dave Airlie <airlied@redhat.com>
16336Date: Thu Jul 5 15:41:22 2012 +0100
16337
16338 randr: add output source setup
16339
16340 This adds the output sources to the associated list and adds the protocol
16341 handler for the randr SetProviderOutputSource.
16342
16343 Reviewed-by: Keith Packard <keithp@keithp.com>
16344 Reviewed-by: Adam Jackson <ajax@redhat.com>
16345 Signed-off-by: Dave Airlie <airlied@redhat.com>
16346
16347commit bec4cb72c55bb6dee09c65c0844af201067a090f
16348Author: Dave Airlie <airlied@redhat.com>
16349Date: Thu May 24 15:47:22 2012 +0100
16350
16351 randr: hook up output slave to screen resources return
16352
16353 When the client asks for the screen resources list, it will now
16354 get a list of crtc/outputs for the master + all attached slaves,
16355 this will let randr configure all attached slave devices properly.
16356
16357 Keith asked I merge the two functions, but not just yet, the current
16358 multi screen code doesn't handle primary yet properly, will fix it up later.
16359
16360 Reviewed-by: Keith Packard <keithp@keithp.com>
16361 Reviewed-by: Adam Jackson <ajax@redhat.com>
16362 Signed-off-by: Dave Airlie <airlied@redhat.com>
16363
16364commit 88bc02bfaae2c15a30c606d34cf7940e6ad5ea14
16365Author: Dave Airlie <airlied@redhat.com>
16366Date: Mon Jun 18 14:11:24 2012 +0100
16367
16368 xf86/crtc: add new interface to detach slave crtcs
16369
16370 This will detach any scanout pixmaps attached to slave crtcs.
16371
16372 Reviewed-by: Keith Packard <keithp@keithp.com>
16373 Reviewed-by: Adam Jackson <ajax@redhat.com>
16374 Signed-off-by: Dave Airlie <airlied@redhat.com>
16375
16376commit 9b8217f9ef6279fff6628631d18497bed0343ef9
16377Author: Dave Airlie <airlied@redhat.com>
16378Date: Thu Jun 14 15:24:46 2012 +0100
16379
16380 dix/pixmap: track dirty pixmaps in server. (v4)
16381
16382 This adds two functions for drivers to use directly to keep a
16383 linked list of slave pixmaps to do damage tracking on and keep
16384 updated. It also adds a helper function that drivers may optionally
16385 call to do a simple copy area damage update.
16386
16387 v2: use damage.h not damagestr.h, fixes ephyr build.
16388
16389 v3: address ajax review: use slave_dst, drop unused dst member.
16390
16391 v4: check DamageCreate return, add SourceValidate comment,
16392 add a comment addressing possible optimisation possibility
16393
16394 Reviewed-by: Keith Packard <keithp@keithp.com>
16395 Reviewed-by: Adam Jackson <ajax@redhat.com>
16396 Signed-off-by: Dave Airlie <airlied@redhat.com>
16397
16398commit c5cc2a8243c5e6bf454af989b7512ec7e20fc3f2
16399Author: Dave Airlie <airlied@redhat.com>
16400Date: Tue May 29 14:41:39 2012 +0100
16401
16402 randr: check the screen size bounds against the master
16403
16404 The master contains the complete screen size bounds, so check
16405 the width/height against it.
16406
16407 Reviewed-by: Keith Packard <keithp@keithp.com>
16408 Reviewed-by: Adam Jackson <ajax@redhat.com>
16409 Signed-off-by: Dave Airlie <airlied@redhat.com>
16410
16411commit 26d848cb916c8a85f2c69d2aeb004665ba07bf00
16412Author: Dave Airlie <airlied@redhat.com>
16413Date: Tue May 29 14:35:24 2012 +0100
16414
16415 randr: fix tellchanged to work with slave scanouts
16416
16417 Tell changed need to tell only for the master pixmap,
16418 however it gets called from various places for slave screens,
16419 so convert to telling clients about changes on the master screen.
16420
16421 Reviewed-by: Keith Packard <keithp@keithp.com>
16422 Reviewed-by: Adam Jackson <ajax@redhat.com>
16423 Signed-off-by: Dave Airlie <airlied@redhat.com>
16424
16425commit 4bf0192d810e01c89a1903cc4bc5e639fc13a547
16426Author: Dave Airlie <airlied@redhat.com>
16427Date: Tue May 29 14:37:11 2012 +0100
16428
16429 randr: fixup constrain to work with slave screens.
16430
16431 Current code constrains the cursor to the crtcs on the master
16432 device, for slave outputs to work we have to include their crtcs
16433 in the constrain calculations.
16434
16435 Reviewed-by: Keith Packard <keithp@keithp.com>
16436 Reviewed-by: Adam Jackson <ajax@redhat.com>
16437 Signed-off-by: Dave Airlie <airlied@redhat.com>
16438
16439commit 2ed3f64d9d453628ab5c03b8aee006093019c933
16440Author: Dave Airlie <airlied@redhat.com>
16441Date: Wed Jun 20 12:37:26 2012 +0100
16442
16443 xf86: add initial scanout pixmap support (v2)
16444
16445 Add the simple passthrough interface for drivers to use,
16446 so they can set scanout pixmaps.
16447
16448 v2: detach scanout pixmap properly.
16449
16450 Reviewed-by: Keith Packard <keithp@keithp.com>
16451 Reviewed-by: Adam Jackson <ajax@redhat.com>
16452 Signed-off-by: Dave Airlie <airlied@redhat.com>
16453
16454commit f2da2c12042d3447b1c954e7301632d174a08b4f
16455Author: Dave Airlie <airlied@redhat.com>
16456Date: Wed Jun 20 12:37:12 2012 +0100
16457
16458 randr: add initial scanout pixmap support (v3)
16459
16460 When randr notices a crtc configuration request for a slave device,
16461 it checks if the slave allocated pixmap exists and is suitable,
16462 if not it allocates a new shared pixmap from the master, shares
16463 it to the slave, and starts the master tracking damage to it,
16464 to keep it updated from the current front pixmap.
16465
16466 If the resize means the crtc is no longer used it will destroy
16467 the slave pixmap.
16468
16469 This adds the concept of a scanout_pixmap to the randr_crtc object,
16470 and also adds a master pixmap pointer to the pixmap object, along
16471 with defining some pixmap helper functions for getting pixmap box/regions.
16472
16473 v2: split out pixmap sharing to a separate function.
16474
16475 v3: update for void *
16476
16477 Reviewed-by: Keith Packard <keithp@keithp.com>
16478 Reviewed-by: Adam Jackson <ajax@redhat.com>
16479 Signed-off-by: Dave Airlie <airlied@redhat.com>
16480
16481commit 0b0e7148925e1f835d4f4f06e402a97023531828
16482Author: Dave Airlie <airlied@redhat.com>
16483Date: Thu May 24 16:03:31 2012 +0100
16484
16485 exa: pixmap sharing infrastructure (v3)
16486
16487 This just adds exa interfaces for mixed exa so drivers can
16488 share and set shared pixmaps up correctly.
16489
16490 v2: update for passing slave screen.
16491 v3: update for void *
16492
16493 Reviewed-by: Keith Packard <keithp@keithp.com>
16494 Reviewed-by: Adam Jackson <ajax@redhat.com>
16495 Signed-off-by: Dave Airlie <airlied@redhat.com>
16496
16497commit fd6c1bf0a3efced3ad4623dbd4716230a5f2419e
16498Author: Dave Airlie <airlied@redhat.com>
16499Date: Thu May 24 16:03:19 2012 +0100
16500
16501 dix: pixmap sharing infrastructure (v3)
16502
16503 This is a hooks for pixmap sharing and tracking.
16504
16505 The pixmap sharing ones get an integer handle for the pixmap
16506 and use a handle to be the backing for a pixmap.
16507
16508 The tracker interface is to be used when a GPU needs to
16509 track pixmaps to be updated for another GPU.
16510
16511 v2: pass slave to sharing so it can use it to work out driver.
16512
16513 v3: use void * as per keithp's suggestion.
16514
16515 Reviewed-by: Keith Packard <keithp@keithp.com>
16516 Reviewed-by: Adam Jackson <ajax@redhat.com>
16517 Signed-off-by: Dave Airlie <airlied@redhat.com>
16518
16519commit 382dd45bb88fa92764bf0ecd655cb9890eb84960
16520Author: Dave Airlie <airlied@redhat.com>
16521Date: Thu May 24 15:39:05 2012 +0100
16522
16523 dix: add ability to link output slave gpus to the current gpu (v1.1)
16524
16525 Just add the interfaces to attach/detach output slaves, and
16526 a linked list to keep track of them. Hook up the randr providers
16527 list to include these slaves.
16528
16529 v1.1: add another assert to the add path.
16530
16531 Reviewed-by: Keith Packard <keithp@keithp.com>
16532 Reviewed-by: Adam Jackson <ajax@redhat.com>
16533 Signed-off-by: Dave Airlie <airlied@redhat.com>
16534
16535commit 05d2472cd242293a5fa205efff5734a3380072b4
16536Author: Dave Airlie <airlied@redhat.com>
16537Date: Thu Jun 14 15:03:02 2012 +0100
16538
16539 xf86dga: handle DGAAvailable for gpu screens. (v2)
16540
16541 v2: Split out DGAAvailable into two interfaces, one for calls from protocol
16542 decoding and one for internal usage, after discussion with ajax and keithp.
16543
16544 Reviewed-by: Keith Packard <keithp@keithp.com>
16545 Reviewed-by: Adam Jackson <ajax@redhat.com>
16546 Signed-off-by: Dave Airlie <airlied@redhat.com>
16547
16548commit 3cbc4c10b52896324fe14d2ab56bd54577c9294c
16549Author: Dave Airlie <airlied@redhat.com>
16550Date: Wed Jun 13 14:35:49 2012 +0100
16551
16552 randr: expose unattached providers to the clients.
16553
16554 This provides the unattached provider list to the clients.
16555
16556 Reviewed-by: Keith Packard <keithp@keithp.com>
16557 Reviewed-by: Adam Jackson <ajax@redhat.com>
16558 Signed-off-by: Dave Airlie <airlied@redhat.com>
16559
16560commit d0e138c3f3e58ef127dad6c6aef48b1713b5ce58
16561Author: Dave Airlie <airlied@redhat.com>
16562Date: Wed Jun 13 14:32:05 2012 +0100
16563
16564 dix: attach unbound screens to protocol screen 0 (v2)
16565
16566 This is the default attachment, unbound gpu screens get
16567 attached to the 0 protocol screen.
16568
16569 detach on hotunplug.
16570
16571 v2: detach after tearing down crtc/outputs.
16572
16573 Reviewed-by: Keith Packard <keithp@keithp.com>
16574 Reviewed-by: Adam Jackson <ajax@redhat.com>
16575 Signed-off-by: Dave Airlie <airlied@redhat.com>
16576
16577commit 4caad34c932115dc0a9fd9c2995ebb2d38b9359c
16578Author: Dave Airlie <airlied@redhat.com>
16579Date: Wed Jun 13 14:27:42 2012 +0100
16580
16581 dix: add unattached list for attaching screens to initially. (v1.1)
16582
16583 This list is meant for attaching unbound gpu screens to initially,
16584 before the client side rebinds them.
16585
16586 v1.1: add another assert in the add path.
16587
16588 Reviewed-by: Keith Packard <keithp@keithp.com>
16589 Reviewed-by: Adam Jackson <ajax@redhat.com>
16590 Signed-off-by: Dave Airlie <airlied@redhat.com>
16591
16592commit 9b5cf2ed76b00937af12304e0e8dbd60df62067e
16593Author: Dave Airlie <airlied@redhat.com>
16594Date: Wed Jun 13 14:17:06 2012 +0100
16595
16596 xfree86: add framework for provider support in ddx. (v4)
16597
16598 This adds the framework for DDX provider support.
16599
16600 v2: as per keithp's suggestion remove the xf86 provider object
16601 and just store it in the toplevel object.
16602
16603 v3: update for new protocol
16604
16605 v4: drop devPrivate, free name
16606
16607 Reviewed-by: Keith Packard <keithp@keithp.com>
16608 Reviewed-by: Adam Jackson <ajax@redhat.com>
16609 Signed-off-by: Dave Airlie <airlied@redhat.com>
16610
16611commit 66d92afeaeed9f4a19267d95a1f81b9bf27162a5
16612Author: Dave Airlie <airlied@redhat.com>
16613Date: Wed Jun 13 14:03:04 2012 +0100
16614
16615 randr: add provider object and provider property support (v6)
16616
16617 This adds the initial provider object and provider property
16618 support to the randr dix code.
16619
16620 v2: destroy provider in screen close
16621 v2.1: fix whitespace
16622
16623 v3: update for latest rev of protocol + renumber after 1.4 tearout.
16624
16625 v4: fix logic issue, thanks Samsagax on irc
16626
16627 v5: keithp's review: fix current_role, fix copyrights, fix master
16628 reporting crtc/outputs.
16629
16630 v6: port to new randr interface, drop all set role bits for now
16631
16632 v7: drop devPrivate in provider, not needed, add BadMatch returns
16633 for NULL SetProviderOffloadSink and SetProviderOutputSource, drop
16634 the old typedef.
16635
16636 Reviewed-by: Keith Packard <keithp@keithp.com>
16637 Reviewed-by: Adam Jackson <ajax@redhat.com>
16638 Signed-off-by: Dave Airlie <airlied@redhat.com>
16639
16640commit 44eae69f1df2d00e0c9f1ea8d3c4fae06bcacfbc
16641Author: Dave Airlie <airlied@redhat.com>
16642Date: Thu Jul 5 21:11:16 2012 +0100
16643
16644 configure: bump randrproto min requirements to 1.4.0
16645
16646 Needed to build latest randr stuff.
16647
16648 Reviewed-by: Adam Jackson <ajax@redhat.com>
16649 Signed-off-by: Dave Airlie <airlied@redhat.com>
16650
16651commit 4d24192bd24fb20d16d0054d77bffda883faeb79
16652Merge: 7c9d8cb 74b786f
16653Author: Keith Packard <keithp@keithp.com>
16654Date: Fri Jul 6 12:17:17 2012 -0700
16655
16656 Merge remote-tracking branch 'airlied/for-keithp-stage1'
16657
16658commit 7c9d8cbd368cac53e7aed9308292614337db3be2
16659Author: Torsten Kaiser <just.for.lkml@googlemail.com>
16660Date: Fri Jul 6 12:10:25 2012 -0700
16661
16662 xfree86: EDID Est III parsing skips some modes
16663
16664 This loop needs to count from 7 to 0, not only from 7 to 1.
16665 The current code always skips the modes {1152, 864, 75, 0}, {1280, 1024, 85, 0},
16666 {1400, 1050, 75, 0}, {1600, 1200, 70, 0} and {1920, 1200, 60, 0}.
16667
16668 Signed-off-by: Torsten Kaiser <x11@ariolc.dyndns.org>
16669 Reviewed-by: Adam Jackson <ajax@redhat.com>
16670 Signed-off-by: Keith Packard <keithp@keithp.com>
16671
16672commit 0b3abacb6430e6593c4358420ba457d8149bd32b
16673Author: Torsten Kaiser <just.for.lkml@googlemail.com>
16674Date: Fri Jul 6 12:06:13 2012 -0700
16675
16676 xfree86: EDID Est III parsing can walk off end of array
16677
16678 Using -O3 gcc notes that m could reach beyound the end of the EstIIIModes array,
16679 if the last bits of the 11s byte where set.
16680 Fix this, by extending the array to cover all possible bits from est.
16681
16682 https://bugs.freedesktop.org/show_bug.cgi?id=45623
16683
16684 Signed-off-by: Torsten Kaiser <x11@ariolc.dyndns.org>
16685 Reviewed-by: Adam Jackson <ajax@redhat.com>
16686 Signed-off-by: Keith Packard <keithp@keithp.com>
16687
16688commit 74b786f7cee5723468bba251a2155a65770b2ad4
16689Author: Dave Airlie <airlied@redhat.com>
16690Date: Tue Jun 5 17:45:03 2012 +0100
16691
16692 xfree86: add autoAddGPU option (v2)
16693
16694 This option is to stop the X server adding non-primary devices as
16695 gpu screens.
16696
16697 v2: fix per Keith's suggestion.
16698
16699 Reviewed-by: Keith Packard <keithp@keithp.com>
16700 Signed-off-by: Dave Airlie <airlied@redhat.com>
16701
16702commit ef6686480ad1de6582b9ef5d13e5d4ffb3b3918c
16703Author: Dave Airlie <airlied@redhat.com>
16704Date: Tue Jun 5 17:30:39 2012 +0100
16705
16706 xfree86: add platform bus hotplug support (v3)
16707
16708 This provides add/remove support for platform devices at xfree86 ddx level.
16709
16710 v2: cleanup properly if no driver found.
16711
16712 v3: load the modesetting driver before checking driver list.
16713
16714 Reviewed-by: Keith Packard <keithp@keithp.com>
16715 Signed-off-by: Dave Airlie <airlied@redhat.com>
16716
16717commit b27cf30995e05da11e827ec41060051673336a7b
16718Author: Dave Airlie <airlied@redhat.com>
16719Date: Tue Jun 5 17:29:56 2012 +0100
16720
16721 xserver/config: add udev/drm hotplug callbacks. (v2)
16722
16723 This adds callbacks into the ddx for udev gpu hotplug.
16724
16725 v2: fix some strncmp returns.
16726
16727 Reviewed-by: Keith Packard <keithp@keithp.com>
16728 Signed-off-by: Dave Airlie <airlied@redhat.com>
16729
16730commit 53b66c084fbf5bf47f8121632fabdc39b9fe8337
16731Author: Dave Airlie <airlied@redhat.com>
16732Date: Tue Jun 5 17:18:10 2012 +0100
16733
16734 xfree86: add DDX gpu screen support. (v3)
16735
16736 This just adds the structures and interfaces required for adding/deleteing
16737 gpu screens at the DDX level. The platform probe can pass a new flag
16738 to the driver, so they can call xf86AllocateScreen and pass back the new
16739 gpu screen flag.
16740
16741 It also calls the gpu screens preinit and screeninit routines at
16742 startup.
16743
16744 v2: fix delete screen use after free.
16745
16746 v3: split out pScrn into separate patch
16747
16748 Reviewed-by: Keith Packard <keithp@keithp.com>
16749 Signed-off-by: Dave Airlie <airlied@redhat.com>
16750
16751commit 726d467b534a30d8cbaafb858339c638c9f83195
16752Author: Dave Airlie <airlied@redhat.com>
16753Date: Mon Jun 25 11:51:26 2012 +0100
16754
16755 xf86: cleanup helper code to use a pointer. (v1.1)
16756
16757 This is in preparation for gpu screens in here, just use
16758 a pScrn pointer to point at the new screen.
16759
16760 suggested by Keith.
16761 v1.1: fix spacing as suggested by Aaron.
16762
16763 Reviewed-by: Keith Packard <keithp@keithp.com>
16764 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
16765 Signed-off-by: Dave Airlie <airlied@redhat.com>
16766
16767commit 9d179818293b466ec6f1777f0b792e1fbbeb318c
16768Author: Dave Airlie <airlied@redhat.com>
16769Date: Tue Jun 5 17:09:38 2012 +0100
16770
16771 dix: introduce gpu screens. (v5)
16772
16773 This patch introduces gpu screens into screenInfo. It adds interfaces
16774 for adding and removing gpu screens, along with adding private fixup,
16775 block handler support, and scratch pixmap init.
16776
16777 GPU screens have a myNum that is offset by GPU_SCREEN_OFFSET (256),
16778 this is used for logging etc.
16779
16780 RemoveGPUScreen isn't used until "xfree86: add platform bus hotplug support".
16781
16782 v2: no glyph pictures for GPU screens for now.
16783 v3: introduce MAXGPUSCREENS, fix return value check
16784 v4: fixup myNum when renumbering screens (ajax)
16785 v5: drop cursor privates for now.
16786
16787 Reviewed-by: Keith Packard <keithp@keithp.com>
16788 Signed-off-by: Dave Airlie <airlied@redhat.com>
16789
16790commit 3478af3374abffa0c226ee077fda1fcfc0751e74
16791Author: Dave Airlie <airlied@redhat.com>
16792Date: Tue Jun 5 17:01:21 2012 +0100
16793
16794 screen: split out screen init code. (v2)
16795
16796 This is a precursor for reusing this code to init gpu screens.
16797
16798 v2: fixup int check as per Keith's review.
16799
16800 Reviewed-by: Keith Packard <keithp@keithp.com>
16801 Signed-off-by: Dave Airlie <airlied@redhat.com>
16802
16803commit cf66471353ac5899383b573a3cfca407e90d501e
16804Author: Dave Airlie <airlied@redhat.com>
16805Date: Tue May 8 14:26:47 2012 +0100
16806
16807 xfree86: use udev to provide device enumeration for kms devices (v10)
16808
16809 On Linux in order for future hotplug work, we are required to interface
16810 to udev to detect device creation/removal. In order to try and get
16811 some earlier testing on this, this patch adds the ability to use
16812 udev for device enumeration on Linux.
16813
16814 At startup the list of drm/kms devices is probed and this info is
16815 used to load drivers.
16816
16817 A new driver probing method is introduced that passes the udev
16818 device info to the driver for probing.
16819
16820 The probing integrates with the pci probing code and will fallback
16821 to the pci probe and old school probe functions in turn.
16822
16823 The flags parameter to the probe function will be used later
16824 to provide hotplug and gpu screen flags for the driver to behave
16825 in a different way.
16826
16827 This patch changes the driver ABI, all drivers should at least
16828 be set with a NULL udev probe function after this commit.
16829
16830 v2: rename to platform bus, now with 100% less udev specific,
16831
16832 this version passes config_odev_attribs around which are an array
16833 of id/string pairs, then the udev code can attach the set of attribs
16834 it understands, the OS specific code can attach its attrib, and then
16835 the core/drivers can lookup the required attribs.
16836
16837 also add MATCH_PCI_DEVICES macro.
16838
16839 This version is mainly to address concerns raised by ajax.
16840
16841 v3: Address comments from Peter.
16842 fix whitespace that snuck in.
16843 rework to use a linked list with some core functions that
16844 xf86 wraps.
16845
16846 v4: add free list, fix struct whitespace.
16847 ajax this address most of your issues?
16848
16849 v5: drop probe ifdef, fix logic issue
16850
16851 v6: some overhaul after more testing.
16852
16853 Implement primaryBus for platform devices.
16854 document hotplug.h dev attribs - drop sysname attrib
16855 fix build with udev kms disabled
16856 make probing work like the PCI probe code,
16857 match against bus id if one exists, or primary device.
16858
16859 RFC: add new bus id support "PLAT:syspath". we probably
16860 want to match on this a bit different, or use a different
16861 property maybe. I was mainly wanting this for use with
16862 specifying usb devices in xorg.conf directly, but PLAT:path
16863 could also work I suppose.
16864
16865 v6.1: add missing noop platform function
16866
16867 v7: fix two interactions with pci probing and slot claiming, prevents
16868 pci and platform trying to load two drivers for same slot.
16869
16870 v8: test with zaphod mode on -ati driver, fixup resulting issue
16871 clean up common probe code into another function, change busid
16872 matching to allow dropping end of strings.
16873
16874 v9: fix platform probing logic so it actually works.
16875 v9.1: fix pdev init to NULL properly.
16876
16877 v10: address most of Keith's concerns.
16878
16879 v4 was thanks to Reviewed-by: Adam Jackson <ajax@redhat.com>
16880 v5 was Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
16881 Reviewed-by: Keith Packard <keithp@keithp.com>
16882 Signed-off-by: Dave Airlie <airlied@redhat.com>
16883
16884commit 2c52d776a42a28bb3e1463edf0cfe0672d0c08a9
16885Author: Dave Airlie <airlied@redhat.com>
16886Date: Tue Jun 19 16:01:08 2012 +0100
16887
16888 xf86: cursor code got mangled by indenting
16889
16890 This fixes some really ugly code that got mangled by the indenting.
16891
16892 Reviewed-by: Keith Packard <keithp@keithp.com>
16893 Signed-off-by: Dave Airlie <airlied@redhat.com>
16894
16895commit 9e4b8b74b008dd72d1da7074bd7ecae677cefd20
16896Author: Keith Packard <keithp@keithp.com>
16897Date: Thu Jul 5 11:31:55 2012 -0700
16898
16899 privates: Allow device privates to be allocated after server start
16900
16901 This will permit midispcur to allocate its privates for hotplug outputs
16902
16903 Signed-off-by: Keith Packard <keithp@keithp.com>
16904 Reviewed-by: Dave Airlie <airlied@redhat.com>
16905
16906commit 493ad833233707b0a42e5fbdd957a24fc6e52917
16907Author: Keith Packard <keithp@keithp.com>
16908Date: Thu Jul 5 11:31:29 2012 -0700
16909
16910 dix: Allocate device privates separate from devices
16911
16912 This will allow device privates to be resized.
16913
16914 Signed-off-by: Keith Packard <keithp@keithp.com>
16915 Reviewed-by: Dave Airlie <airlied@redhat.com>
16916
16917commit f38b2b628322f4d763b0c7f8387d5ab0472a7d2b
16918Author: Keith Packard <keithp@keithp.com>
16919Date: Thu Jul 5 11:15:18 2012 -0700
16920
16921 mi: Don't save midispcur rendering resources for each cursor
16922
16923 Instead, only save them for the most recently drawn cursor. This saves
16924 a bunch of storage for idle cursors.
16925
16926 Signed-off-by: Keith Packard <keithp@keithp.com>
16927 Reviewed-by: Dave Airlie <airlied@redhat.com>
16928
16929commit 7651176b0093cf3a4cff2277f6918a7d0451af41
16930Author: Keith Packard <keithp@keithp.com>
16931Date: Tue Jul 3 16:12:59 2012 -0700
16932
16933 Use screen-specific privates for exa pixmap and gc privates
16934
16935 Signed-off-by: Keith Packard <keithp@keithp.com>
16936 Reviewed-by: Dave Airlie <airlied@redhat.com>
16937
16938commit 7ca32f1c282cb6d4d7255f118a1e2e6f54e6b3fa
16939Author: Keith Packard <keithp@keithp.com>
16940Date: Tue Jul 3 16:05:42 2012 -0700
16941
16942 Use new screen-specific privates for fb window and gc privates
16943
16944 This ensures that only screens using fb will have this space allocated.
16945
16946 Signed-off-by: Keith Packard <keithp@keithp.com>
16947 Reviewed-by: Dave Airlie <airlied@redhat.com>
16948
16949commit 9d457f9c55f12106ba44c1c9db59d14f978f0ae8
16950Author: Keith Packard <keithp@keithp.com>
16951Date: Tue Jul 3 15:45:09 2012 -0700
16952
16953 Add screen-specific privates.
16954
16955 Screen-specific privates areas are only allocated for objects related
16956 to the target screen; objects allocated for other screens will not
16957 have the private space reserved. This saves memory in these objects
16958 while also allowing hot-plug screens to have additional private
16959 allocation space beyond what the core screens are using.
16960
16961 Drivers are encouraged to switch to this mechanism as it will reduce
16962 memory usage in multi-GPU environments, but it is only required for
16963 drivers which will be loaded after the server starts, like
16964 modesetting.
16965
16966 Objects providing screen-specific privates *must* be managed by the
16967 screen-specific private API when allocating or initializing privates
16968 so that the per-screen area can be initialized properly.
16969
16970 The objects which support screen-specific privates are:
16971
16972 Windows
16973 Pixmaps
16974 GCs
16975 Pictures
16976
16977 Extending this list to include Colormaps would be possible, but
16978 require slightly more work as the default colormap is created before
16979 all colormap privates are allocated during server startup, and hence
16980 gets a bunch of special treatment.
16981
16982 Of particular note, glyphs are *not* capable of supporting
16983 screen-specific privates as they are global objects, not allocated on
16984 a screen-specific basis, and so each driver must be able to see their
16985 privates within the glyph.
16986
16987 Signed-off-by: Keith Packard <keithp@keithp.com>
16988 Reviewed-by: Dave Airlie <airlied@redhat.com>
16989
16990commit ed6daa15a7dcf8dba930f67401f4c1c8ca2e6fac
16991Author: Ian Romanick <ian.d.romanick@intel.com>
16992Date: Wed Jul 4 15:21:09 2012 -0700
16993
16994 glx/dri2: Enable GLX_ARB_create_context_robustness
16995
16996 If the driver supports __DRI2_ROBUSTNESS, then enable
16997 GLX_ARB_create_cotnext_robustness as well. If robustness values are
16998 passed to glXCreateContextAttribsARB and the driver doesn't support
16999 __DRI2_ROBUSTNESS, existing drivers will already generate the correct
17000 error values (so that the correct GLX errors are generated).
17001
17002 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17003 Reviewed-by: Dave Airlie <airlied@redhat.com>
17004 Signed-off-by: Keith Packard <keithp@keithp.com>
17005
17006commit 90aa2486e394c0344aceb2a70432761665a79333
17007Author: Ian Romanick <ian.d.romanick@intel.com>
17008Date: Wed Jul 4 15:21:08 2012 -0700
17009
17010 glx: Allow attributes and flags from GLX_ARB_create_context_robustness
17011
17012 The assumtion is that if the underlying provider (e.g., glxdri2.c) doesn't
17013 support this extension, it will generate BadMatch for these flags and
17014 attributes.
17015
17016 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17017 Reviewed-by: Dave Airlie <airlied@redhat.com>
17018 Signed-off-by: Keith Packard <keithp@keithp.com>
17019
17020commit aef0b4593e165634ca7fa0745882dad12a54ebed
17021Author: Ian Romanick <ian.d.romanick@intel.com>
17022Date: Wed Jul 4 15:21:07 2012 -0700
17023
17024 glx: Track the reset notification strategy for a context
17025
17026 Also require that the reset notification for a new context and the other
17027 context in the share group match. There isn't yet any way to specify a
17028 non-default reset notification strategy, but that will come.
17029
17030 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17031 Reviewed-by: Dave Airlie <airlied@redhat.com>
17032 Signed-off-by: Keith Packard <keithp@keithp.com>
17033
17034commit f8198285cfc4704bc4d56922e5a2e99030a83778
17035Author: Ian Romanick <ian.d.romanick@intel.com>
17036Date: Wed Jul 4 15:21:06 2012 -0700
17037
17038 glx: Add extension string tracking for GLX_ARB_create_context_robustness
17039
17040 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17041 Reviewed-by: Dave Airlie <airlied@redhat.com>
17042 Signed-off-by: Keith Packard <keithp@keithp.com>
17043
17044commit 4ffbc984eac8a15925094614021de30dc8084e65
17045Author: Ian Romanick <ian.d.romanick@intel.com>
17046Date: Wed Jul 4 15:21:05 2012 -0700
17047
17048 glx: Reject creations of indirect contexts with unsupportable versions.
17049
17050 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17051 Reviewed-by: Dave Airlie <airlied@redhat.com>
17052 Signed-off-by: Keith Packard <keithp@keithp.com>
17053
17054commit 41d5120eaf2bdf308eb904cac9b5027b003fea8d
17055Author: Ian Romanick <ian.d.romanick@intel.com>
17056Date: Wed Jul 4 15:21:04 2012 -0700
17057
17058 glx: Implement GLX_EXT_create_context_es2_profile
17059
17060 This patch builds on the infrastucture put in place for
17061 GLX_ARB_create_context_profile. If GLX_CONTEXT_ES2_PROFILE_BIT_EXT is
17062 specified and the requested version is 2.0, create a context with the
17063 __DRI_API_GLES2 API.
17064
17065 This change assumes that any DRI2 driver can handle (possibly by saying "no
17066 seeing an API setting other than __DRI_API_OPENGL or __DRI_API_OPENGL_CORE.
17067 This allows enabling the extension any time GLX_ARB_create_context (and
17068 GLX_ARB_create_context_profile) is enabled.
17069
17070 v2: Clean up some comments. Note that our behavior for
17071 GLX_CONTEXT_ES2_PROFILE_BIT_EXT w/version != 2.0 matches NVIDIA's.
17072
17073 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17074 Reviewed-by: Dave Airlie <airlied@redhat.com>
17075 Signed-off-by: Keith Packard <keithp@keithp.com>
17076
17077commit cef0b808d3152a20cd98d32457dde08bd5434e7c
17078Author: Ian Romanick <ian.d.romanick@intel.com>
17079Date: Wed Jul 4 15:21:03 2012 -0700
17080
17081 glx: Implement GLX_ARB_create_context_profile
17082
17083 Most of the infrastructure was already in place. This just adds:
17084
17085 * Validate values specified with the GLX_CONTEXT_PROFILE_MASK_ARB
17086 attribute.
17087
17088 * Select a DRI2 "api" based on the setting of
17089 GLX_CONTEXT_PROFILE_MASK_ARB.
17090
17091 * Enable GLX_ARB_create_context_profile extension.
17092
17093 This change assumes that any DRI2 driver can handle (possibly by saying "no")
17094 seeing an API setting other than __DRI_API_OPENGL. This allows enabling this
17095 extension any time GLX_ARB_create_context is enabled.
17096
17097 Also, the validation code in __glXDisp_CreateContextAttribsARB is structured
17098 in a very verbose manner (using a switch-statement) to ease the addition of
17099 GLX_EXT_create_context_es2_profile.
17100
17101 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17102 Reviewed-by: Dave Airlie <airlied@redhat.com>
17103 Signed-off-by: Keith Packard <keithp@keithp.com>
17104
17105commit 09a8a169d5929665778ae33e58e345fe1192b64c
17106Author: Ian Romanick <ian.d.romanick@intel.com>
17107Date: Wed Jul 4 15:21:02 2012 -0700
17108
17109 glx: Implement GLX SetClientInfo2ARB protocol
17110
17111 The protocol is almost identical to SetClientInfoARB. The only
17112 difference is the GL versions include an extra 4 bytes for the supported
17113 profile.
17114
17115 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17116 Reviewed-by: Dave Airlie <airlied@redhat.com>
17117 Signed-off-by: Keith Packard <keithp@keithp.com>
17118
17119commit 167993254a5cbe11a1f44fad1e8ae042089c1619
17120Author: Peter Hutterer <peter.hutterer@who-t.net>
17121Date: Thu Jul 5 16:17:56 2012 +1000
17122
17123 list.h: don't crash when removing an element from a NULL list
17124
17125 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17126 Signed-off-by: Keith Packard <keithp@keithp.com>
17127
17128commit 8aa6d49d46dacc31a85023d0159c1fb67401df73
17129Author: Ian Romanick <ian.d.romanick@intel.com>
17130Date: Wed Jul 4 15:21:01 2012 -0700
17131
17132 glx: Don't request prototypes for GL extension functions
17133
17134 The server does not want GL extension prototypes. It never links with
17135 anything that could possibly provide implementations of these functions. It
17136 *is* the provide, and it does not provde these symbols. All this does is
17137 create hundreds of warnings like:
17138
17139 incude/GL/glext.h:5570:45: warning: redundant redeclaration of 'glMultTransposeMatrixd' [-Wredundant-decls]
17140 include/GL/gl.h:1940:45: note: previous declaration of 'glMultTransposeMatrixd' was here
17141
17142 I believe this was necessary back before the AIGLX days, but that was a long,
17143 long time ago. The warnings recently showed up with the addition of
17144 -Wredundant-decls to the CWARNFLAGS.
17145
17146 v2: Update the commit message, no code changes.
17147
17148 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
17149 Cc: Adam Jackson <ajax@redhat.com>
17150 Cc: Dave Airlie <airlied@redhat.com>
17151 Acked-by: Keith Packard <keithp@keithp.com>
17152 Signed-off-by: Keith Packard <keithp@keithp.com>
17153
17154commit 258abbf823f753757e4dddc13ef495f4024680db
17155Author: Michal Srb <msrb@suse.com>
17156Date: Thu Jun 28 17:17:12 2012 +0200
17157
17158 Look for ModuleData only in appropriate library
17159
17160 LoaderSymbol calls dlsym with RTLD_DEFAULT pseudo handle making it search in
17161 every loaded library. In addition glibc adds NODELETE flag to the library
17162 containing the symbol.
17163
17164 It's used in doLoadModule to locate <modulename>ModuleData symbol, the
17165 module's library gets the flag and is kept in memory even after it is
17166 unloaded.
17167
17168 This patch adds LoaderSymbolFromModule function that looks for symbol only in
17169 library specified by handle. That way the NODELETE flag isn't added.
17170
17171 This glibc behavior doesn't seem to be documented, but even if other
17172 implementations differ, there is no reason to search ModuleData symbol outside
17173 the module's library.
17174
17175 Signed-off-by: Michal Srb <msrb@suse.com>
17176 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
17177
17178 v2: Switch LoaderSymbolFromModule arguments order.
17179 Correct description.
17180 Reviewed-by: Adam Jackson <ajax@redhat.com>
17181 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17182
17183commit d84f0f823eeeecdf0498aadd3fbb1d11dabc0837
17184Merge: 12bfb4c d04dfe3
17185Author: Peter Hutterer <peter.hutterer@who-t.net>
17186Date: Wed Jul 4 21:23:48 2012 +1000
17187
17188 Merge branch 'sigio-vt-switch-issues' into for-keith
17189
17190 Conflicts:
17191 test/Makefile.am
17192
17193 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17194
17195commit 12bfb4cf1bebb66d2c2eb76b93c18a2915b865e5
17196Author: Peter Hutterer <peter.hutterer@who-t.net>
17197Date: Tue Jun 26 09:19:33 2012 +1000
17198
17199 Xi: extend PropagateMask to EMASKSIZE
17200
17201 Number of devices is 2 + MAXDEVICES, with index 0 and 1 reserved for
17202 XIAll{Master}Devices. At the current size, PropagateMask would be overrun in
17203 RecalculateDeviceDeliverableEvents().
17204
17205 Found by Coverity.
17206
17207 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17208 Reviewed-by: Dave Airlie <airlied@redhat.com>
17209
17210commit a9c09f8f8e2a97e4bfe927bc4c7f29b04ee3403a
17211Author: Peter Hutterer <peter.hutterer@who-t.net>
17212Date: Mon Jun 25 14:24:11 2012 +1000
17213
17214 dix: fix memory leak in TouchEventHistoryReplay
17215
17216 Don't leak if ti->history is NULL.
17217
17218 Found by coverity.
17219
17220 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17221 Reviewed-by: Dave Airlie <airlied@redhat.com>
17222
17223commit 55ff20eb37f00a3ad07b5acff19d4497ab513a97
17224Author: Peter Hutterer <peter.hutterer@who-t.net>
17225Date: Mon Jun 25 14:22:31 2012 +1000
17226
17227 dix: fix dereference before null check
17228
17229 Found by Coverity.
17230
17231 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17232 Reviewed-by: Dave Airlie <airlied@redhat.com>
17233
17234commit e3f47be9fbc489e3eb5832445924810b2ff300f7
17235Author: Peter Hutterer <peter.hutterer@who-t.net>
17236Date: Mon Oct 24 10:01:26 2011 +1000
17237
17238 xfree86: fix use-after-free issue in checkInput
17239
17240 *dev is the condition of the while loop we're in, reset to NULL after
17241 freeing
17242
17243 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17244 Reviewed-by: Dave Airlie <airlied@redhat.com>
17245
17246commit 252a69b592001b34714d0fe7508e1898883f9d01
17247Author: Peter Hutterer <peter.hutterer@who-t.net>
17248Date: Wed Jun 20 14:25:54 2012 +1000
17249
17250 xkb: use local variable instead of casting arg
17251
17252 No functional changes.
17253
17254 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17255
17256commit d04dfe3f754ad3a5f158057175cbd44319c1ae51
17257Author: Peter Hutterer <peter.hutterer@who-t.net>
17258Date: Fri Jun 22 15:18:53 2012 +1000
17259
17260 Drop custom sigio block/unblock functions from kdrive, ephyr and dmx
17261
17262 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17263 Reviewed-by: Keith Packard <keithp@keithp.com>
17264
17265commit 5d309af2ed93e91c7d72f548a11052051efbb40f
17266Author: Peter Hutterer <peter.hutterer@who-t.net>
17267Date: Fri Jun 22 14:56:25 2012 +1000
17268
17269 xfree86: drop ddx-specific SIGIO blocking
17270
17271 The hooks are left for this cycle, we can drop it next cycle once the
17272 drivers that need it (e.g. wacom) have been updated.
17273
17274 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17275 Reviewed-by: Keith Packard <keithp@keithp.com>
17276
17277commit ff67135bec3ecea8cebeb46b42ece0a6671e231d
17278Author: Peter Hutterer <peter.hutterer@who-t.net>
17279Date: Fri Jun 22 13:09:35 2012 +1000
17280
17281 xfree86: use OsBlockSIGIO from the ddx
17282
17283 We can ignore the "wasset" argument now since the DIX will keep proper
17284 refcounting.
17285
17286 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17287 Reviewed-by: Keith Packard <keithp@keithp.com>
17288
17289commit 6bf356ef2831baeccd7a650ed3fde0831e33c6c0
17290Author: Peter Hutterer <peter.hutterer@who-t.net>
17291Date: Fri Jun 22 13:02:40 2012 +1000
17292
17293 os: add OsBlockSIGIO and OsReleaseSIGIO
17294
17295 Let the dix be in charge of changing the sigprocmask so we only have one
17296 entity that changes it.
17297
17298 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17299 Reviewed-by: Keith Packard <keithp@keithp.com>
17300
17301commit 24525d96a3b9dba67eb75042500b2f208a2cc246
17302Merge: 4cd91bd 35e3d22
17303Author: Keith Packard <keithp@keithp.com>
17304Date: Mon Jul 2 22:35:39 2012 -0700
17305
17306 Merge branch 'sigsafe-logging-varargs'
17307
17308 This merge includes a minor fixup for '%p' arguments; must cast to
17309 uintptr_t instead of uint64_t as we use -Werror=pointer-to-int-cast
17310 which complains when doing a cast (even explicitly) from a pointer
17311 to an integer of different size.
17312
17313commit 35e3d229150395a222a0f53318daf5dbeb8f6eb6
17314Author: Peter Hutterer <peter.hutterer@who-t.net>
17315Date: Thu May 24 14:04:42 2012 +1000
17316
17317 Bump to ABI_XINPUT_VERSION 18
17318
17319 The input ABI hasn't changed, but input drivers need something to hook on if
17320 they want to log from within signal handlers and the input ABI is the
17321 simplest way of doing so.
17322
17323 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17324
17325commit 541934168dbeb17059542bb5a1da8eba7995fa05
17326Author: Peter Hutterer <peter.hutterer@who-t.net>
17327Date: Mon May 28 10:10:30 2012 +1000
17328
17329 xfree86: constify InputDriverPtr->driverName and default_options
17330
17331 Already treated as const anyway by all drivers.
17332
17333 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17334 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
17335 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17336
17337commit c66089d2206bafc01307a8327ff6089edcb4ed2d
17338Author: Peter Hutterer <peter.hutterer@who-t.net>
17339Date: Mon May 28 09:57:33 2012 +1000
17340
17341 xfree86: constify InputInfoPtr->type_name
17342
17343 This corresponds to XListInputDevice(3)'s "type" field (after being
17344 converted to an Atom). Input drivers use the XI_KEYBOARD and similar
17345 defines, even Wacom which falls out of the common defines uses constant
17346 strings here. The use-case for having this non-const is small.
17347
17348 Input ABI break technically, since we never freed this information anyway it
17349 is not a noticable change.
17350
17351 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17352 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
17353 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17354
17355commit 505c8a2b2cae0318db1148417ec850d54b38f7df
17356Author: Chase Douglas <chase.douglas@canonical.com>
17357Date: Mon Apr 9 09:41:38 2012 -0700
17358
17359 Log in OsVendorFatalError() in a signal safe manner
17360
17361 The function can be called from a fatal signal handler.
17362
17363 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17364 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17365 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17366
17367commit d51aebdbf99a9f240f7c318a70ba40e61cd43049
17368Author: Chase Douglas <chase.douglas@canonical.com>
17369Date: Mon Apr 9 08:30:50 2012 -0700
17370
17371 Log in LoaderUnload() in a signal safe manner
17372
17373 The function may be called from a fatal signal handler.
17374
17375 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17376 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17377 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17378
17379commit c3e1168778ec20beeac9979dc57e36400c70dd63
17380Author: Chase Douglas <chase.douglas@canonical.com>
17381Date: Mon Apr 9 08:28:17 2012 -0700
17382
17383 Log in UnloadModuleOrDriver() in a signal safe manner
17384
17385 The function may be called from a fatal signal handler.
17386
17387 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17388 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17389 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17390
17391commit 89e3ac07aca1def155299aff6f7a57ccafb68fd7
17392Author: Chase Douglas <chase.douglas@canonical.com>
17393Date: Mon Apr 9 08:23:32 2012 -0700
17394
17395 Log safely in fatal signal handler
17396
17397 While we probably don't need to be signal safe here since we will never
17398 return to the normal context, the logging signal context check will
17399 cause unsafe logging to be unhandled. Using signal safe logging here
17400 resolves the issue.
17401
17402 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17403 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17404 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17405
17406commit 6fd5add005d0660b591d808583d1a6c6a85f1277
17407Author: Chase Douglas <chase.douglas@canonical.com>
17408Date: Fri Apr 6 16:17:41 2012 -0700
17409
17410 Log mieq enqueue overflow in a signal safe manner
17411
17412 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17413 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17414 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17415
17416commit 7f4a69b628a6246855054a0b94d6d6dd14e8842c
17417Author: Chase Douglas <chase.douglas@canonical.com>
17418Date: Fri Apr 13 16:01:38 2012 -0700
17419
17420 Log messages in TouchBeginDDXTouch() in a signal-safe manner
17421
17422 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17423 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17424 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17425
17426commit f752226e40890643df213a62f0c96e6a0243e754
17427Author: Chase Douglas <chase.douglas@canonical.com>
17428Date: Fri Apr 6 10:21:14 2012 -0700
17429
17430 Log messages in GetTouchEvents() in a signal safe manner
17431
17432 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17433 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17434 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17435
17436commit 82d1c6b310eaa5095eed9ee4ea958261a46a78e1
17437Author: Chase Douglas <chase.douglas@canonical.com>
17438Date: Fri Apr 6 08:32:28 2012 -0700
17439
17440 Warn when attempting to log in a signal unsafe manner from signal context
17441
17442 Also, print out the offending message format. This will hopefully help
17443 developers track down unsafe logging.
17444
17445 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17446 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17447 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17448
17449commit 512bec06be6c79ca263da9de8f40430b8095b57b
17450Author: Chase Douglas <chase.douglas@canonical.com>
17451Date: Mon Apr 16 09:47:42 2012 -0700
17452
17453 Make BUG_WARN* signal safe
17454
17455 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17456 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17457 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17458
17459commit 0fa5217836cf7fd3872fccc9f3ff9ff32426c25b
17460Author: Chase Douglas <chase.douglas@canonical.com>
17461Date: Fri Apr 6 16:09:05 2012 -0700
17462
17463 Print backtrace in a signal-safe manner
17464
17465 Backtraces are often printed in signal context, such as when a segfault
17466 occurs.
17467
17468 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17469 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17470 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17471
17472 os: print offset as unsigned int, not long unsigned int
17473
17474 pnprintf() takes unsigned int for %u
17475
17476 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17477
17478commit ac20815d5235e7a8e7b331365aabf5a489fc5e34
17479Author: Chase Douglas <chase.douglas@canonical.com>
17480Date: Tue Jun 5 15:39:41 2012 +1000
17481
17482 Add ErrorFSigSafe() alternative to ErrorF()
17483
17484 ErrorF() is not signal safe. Use ErrorSigSafe() whenever an error
17485 message may be logged in signal context.
17486
17487 [whot: edited to "ErrorFSigSafe"]
17488
17489 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17490 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17491 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17492
17493commit 164b38c72fe9c69d13ea4f9c46d4ccc46566d826
17494Author: Chase Douglas <chase.douglas@canonical.com>
17495Date: Fri Apr 6 08:28:40 2012 -0700
17496
17497 Add LogMessageVerbSigSafe() for logging messages while in signal context
17498
17499 [whot: edited to use varargs, squashed commit below]
17500
17501 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17502 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17503 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17504
17505 os: fix vararg length calculation
17506
17507 Make %u and %x sizeof(unsigned int), %p sizeof(void*). This is printf
17508 behaviour and we can't guarantee that void* is uint64_t anyway.
17509
17510 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17511 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
17512
17513commit 4cd91bd4c6325b2e617ca90a725f02ac124c2372
17514Author: Alan Coopersmith <alan.coopersmith@oracle.com>
17515Date: Sat May 12 10:11:33 2012 -0700
17516
17517 Enable no-undefined symbols mode in Solaris builds if -z parent is in ld
17518
17519 The Solaris linker recently added a -z parent flag for easier checking
17520 of symbol definitions in plugins against the program that loads them.
17521 If that's present, this enables it, along with -z defs to error on
17522 undefined symbols to alert us if any modules call symbols that won't
17523 be found at runtime.
17524
17525 This builds upon, and requires, the recent Cygwin work to build Xorg.
17526 It moves a couple more modules to be after the Xorg binary in the build
17527 order so that they can find the binary to check against, much as the
17528 Cygwin changes did (these would be modules built on Solaris but not
17529 Cygwin).
17530
17531 v2: This version only sets the flags for the xorg-server build itself,
17532 and does not yet export them in xorg-server.pc to the drivers, since
17533 most of the drivers are not ready to build with -no-undefined yet.
17534
17535 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17536 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
17537
17538commit d31942e15c619e9355ef8183223f9165902a9d56
17539Author: Alan Coopersmith <alan.coopersmith@oracle.com>
17540Date: Sat May 12 10:10:49 2012 -0700
17541
17542 Add pixman to module dependencies for pixman_region_* functions
17543
17544 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17545 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
17546
17547commit 339f38ef380d750c4c7afd60331a559845e4184e
17548Author: Alan Coopersmith <alan.coopersmith@oracle.com>
17549Date: Sat May 12 10:09:41 2012 -0700
17550
17551 Build tda9885.c functions into fi1236 to avoid libtool relinking problems
17552
17553 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17554 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
17555
17556commit a00066d2916b6910429cc1c7feedafbaee0d4750
17557Author: Alan Coopersmith <alan.coopersmith@oracle.com>
17558Date: Thu Jun 28 15:15:53 2012 -0700
17559
17560 Add dixGetGlyphs to replace GetGlyphs from libXfont to simplify linking
17561
17562 No other Xfont consumer used it, and this saves us from having to link
17563 callers against libXfont for one simple function when doing
17564 -no-undefined symbols builds.
17565
17566 The function is given a new name to avoid clashing with existing libXfont
17567 binaries, but a #define is provided to preserve the API so we don't have
17568 to fix all the callers at the same time.
17569
17570 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17571 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
17572
17573commit 212b9803238d2de2e77cbe5de62d3f616ae50daf
17574Author: Aaron Plattner <aplattner@nvidia.com>
17575Date: Fri Jun 29 13:33:58 2012 -0700
17576
17577 randr: Fix REQUEST vs. REQUEST_SIZE_MATCH mismatch
17578
17579 ProcRRGetScreenSizeRange uses REQUEST(xRRGetScreenSizeRangeReq) followed by
17580 REQUEST_SIZE_MATCH(xRRGetScreenInfoReq). This happens to work out because both
17581 requests have the same size, so this is not a functional change, just a cosmetic
17582 one.
17583
17584 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
17585 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17586 Signed-off-by: Keith Packard <keithp@keithp.com>
17587
17588commit 957bf959fb577b292a3e4f6bb67740ca09e7aeb9
17589Author: Dave Airlie <airlied@redhat.com>
17590Date: Wed Jun 27 14:36:04 2012 +0100
17591
17592 dix/render: consolidate window format matching code.
17593
17594 This code existed in 3 different forms, perhaps it should be
17595 consolidated.
17596
17597 Reviewed-by: Keith Packard <keithp@keithp.com>
17598 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
17599 Signed-off-by: Dave Airlie <airlied@redhat.com>
17600 Signed-off-by: Keith Packard <keithp@keithp.com>
17601
17602commit a7b97b0fa85d695ae19d194cfa3267159d149e5d
17603Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
17604Date: Thu Jun 28 12:07:11 2012 +1000
17605
17606 dri2: Fix authentication for < v8 clients (v2).
17607
17608 The legacy logic was embarassingly wrong; AuthMagic should return errno,
17609 so returning FALSE only when AuthMagic returns nonzero is exactly wrong.
17610
17611 v2: Match drmAuthMagic by returning -EINVAL rather than EINVAL
17612 Fix trailing whitespace
17613
17614 https://bugs.freedesktop.org/show_bug.cgi?id=51400
17615 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
17616 Tested-by: Chris Wilson <chris@chris-wilson.co.uk>
17617 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
17618 Tested-by: Knut Petersen <knut_petersen@t-online.de>
17619 Signed-off-by: Keith Packard <keithp@keithp.com>
17620
17621commit 8b4f0a4fdd6c15f49458bc3a8c3135b8c389240d
17622Merge: 59294a2 d642e71
17623Author: Keith Packard <keithp@keithp.com>
17624Date: Thu Jun 28 10:38:26 2012 -0700
17625
17626 Merge remote-tracking branch 'jturney/master'
17627
17628commit 59294a2179bd5bb996693af004b470932df00ac6
17629Merge: 855003c 191b630
17630Author: Keith Packard <keithp@keithp.com>
17631Date: Thu Jun 28 10:11:44 2012 -0700
17632
17633 Merge remote-tracking branch 'whot/for-keith'
17634
17635commit 855003c333a0ead1db912695bc9705ef2b3144b4
17636Author: Keith Packard <keithp@keithp.com>
17637Date: Thu Jun 21 18:45:18 2012 -0700
17638
17639 randr: Catch two more potential unset rrScrPriv uses
17640
17641 Ricardo Salveti <ricardo.salveti@linaro.org> found one place where the
17642 randr code could use the randr screen private data without checking
17643 for null first. This happens when the X server is running with
17644 multiple screens, some of which are randr enabled and some of which
17645 are not. Applications making protocol requests to the non-randr
17646 screens can cause segfaults where the server touches the unset private
17647 structure.
17648
17649 I audited the code and found two more possible problem spots; the
17650 trick to auditing for this issue was to look for functions not taking
17651 a RandR data structure and where there was no null screen private
17652 check above them in the call graph.
17653
17654 Signed-off-by: Keith Packard <keithp@keithp.com>
17655
17656commit 32603f57ca03b6390b109960f8bb5ea53ac95ecb
17657Author: Ricardo Salveti de Araujo <ricardo.salveti@linaro.org>
17658Date: Thu Jun 21 00:55:53 2012 -0300
17659
17660 randr: first check pScrPriv before using the pointer at RRFirstOutput
17661
17662 Fix a seg fault in case pScrPriv is NULL at ProcRRGetScreenInfo,
17663 which later calls RRFirstOutput.
17664
17665 Signed-off-by: Ricardo Salveti de Araujo <ricardo.salveti@linaro.org>
17666 Reviewed-by: Keith Packard <keithp@keithp.com>
17667 Signed-off-by: Keith Packard <keithp@keithp.com>
17668
17669commit d642e7128790ec6b709df8d1a91333ab87598077
17670Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
17671Date: Thu Jun 21 20:01:49 2012 +0100
17672
17673 configure.ac: link XWIN with GLX_SYS_LIBS
17674
17675 Link XWIN with GLX_SYS_LIBS, just like all the other DDXs
17676
17677 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
17678 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
17679 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
17680
17681commit 9f1edced9abc066f0ba47672d006fe50fb206371
17682Author: Peter Hutterer <peter.hutterer@who-t.net>
17683Date: Fri Jun 15 10:00:51 2012 +1000
17684
17685 xfree86: always enable SIGIO on OsVendorInit (#50957)
17686
17687 Drivers call xf86InstallSIGIOHandler() for their fd on DEVICE_ON. That
17688 function does not actually enable the signal if it was blocked to begin
17689 with. As a result, if one vt-switches away from the server (SIGIO is
17690 blocked) and then triggers a server regeneration, the signal remains
17691 blocked and input devices are dead.
17692
17693 Avoid this by always unblocking SIGIO when we start the server.
17694
17695 X.Org Bug 50957 <http://bugs.freedesktop.org/show_bug.cgi?id=50957>
17696
17697 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17698 Reviewed-by: Keith Packard <keithp@keithp.com>
17699
17700commit 4ba340cfaa8d430c808566495f8deda0ff1b4424
17701Author: Keith Packard <keithp@keithp.com>
17702Date: Thu Jun 21 18:42:46 2012 -0700
17703
17704 randr: Clean up compiler warnings about unused and shadowing variables
17705
17706 set but not used variables
17707 shadowing a previous local
17708
17709 A hidden problem was that the VERIFY_RR_* macros define local 'rc'
17710 variables, any other local definitions for those would be shadowed and
17711 generate warnings from gcc. I've renamed the other locals 'ret'
17712 instead of 'rc'.
17713
17714 Signed-off-by: Keith Packard <keithp@keithp.com>
17715
17716commit 191b630656ba206ad78b7803c3a76f7d9567d9d8
17717Author: Peter Hutterer <peter.hutterer@who-t.net>
17718Date: Thu Jun 21 11:25:24 2012 +1000
17719
17720 include: document _XkbErrCode2 macros
17721
17722 Why sending the number of the (implementation-dependent) error statement to
17723 the client is a good idea is a bit beyond me, but at least document it so we
17724 can all share the despair.
17725
17726 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17727 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
17728
17729commit 7100118c8d3314d73bfef0a262af56686b227055
17730Author: Peter Hutterer <peter.hutterer@who-t.net>
17731Date: Mon Jun 18 11:28:52 2012 +1000
17732
17733 AC_SUBST the GLX_SYS_LIBS
17734
17735 libxservertest needs -lpthread from glxapi.c's pthread_once() call. Usually
17736 this would be pulled in by the XORG_LIBS but not when building without Xorg.
17737
17738 This commit has no visible effect on the current tree, preparation for test
17739 cleanups.
17740
17741 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17742 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17743
17744commit 014ad46f1b353a95e2c4289443ee857cfbabb3ae
17745Author: Peter Hutterer <peter.hutterer@who-t.net>
17746Date: Thu Jun 21 14:26:15 2012 +1000
17747
17748 os: print newline after printing display name
17749
17750 Much easier for scripts that try to read the display value off the file
17751 descriptor. Plus, this restores the behaviour we had for this patch in
17752 Fedora since server 1.6 (April 2009).
17753
17754 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17755 Reviewed-by: Dave Airlie <airlied@redhat.com>
17756 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
17757
17758commit e1f86a7c77ed49e001ea42ef76ab83e87dbdf126
17759Author: Sergei Trofimovich <slyfox@gentoo.org>
17760Date: Mon Jun 18 01:13:21 2012 +0300
17761
17762 test/.gitignore: add hashtabletest
17763
17764 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
17765 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17766 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17767
17768commit c9bbf8fe574591c8e191eeef20f8994781718772
17769Author: Jon TURNEY <jon.turney@dronecode.org.uk>
17770Date: Thu Jun 21 19:38:55 2012 +0100
17771
17772 hw/xwin/glx: Fix glxWinCreateContext() function pointer mismatch
17773
17774 Fix glxWinCreateContext() function signature to align with commit 96d74138
17775 "glx: Extend __GLXscreen::createContext to take attributes", which added more
17776 parameters to the screen createContext function for implementing GLX_ARB_create_context
17777
17778 indirect.c: In function 'glxWinScreenProbe':
17779 indirect.c:683:36: warning: assignment from incompatible pointer type
17780
17781 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
17782 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
17783
17784commit 6e85660c7503c7e88e24003562444dffde6c88ff
17785Author: Jon TURNEY <jon.turney@dronecode.org.uk>
17786Date: Mon Mar 5 23:07:56 2012 +0000
17787
17788 hw/xwin/glx: Update gen_gl_wrappers.py for the current glx/dispatch.h
17789
17790 Fix the code generator for the dispatch table initialization after
17791 the whitespace/coding style changes to glx/dispatch.h
17792
17793 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
17794 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
17795
17796commit cb638ed9c81f07d8603b07db395e31a41fed8685
17797Author: Jon TURNEY <jon.turney@dronecode.org.uk>
17798Date: Thu Jun 21 18:55:57 2012 +0100
17799
17800 hw/xwin/glx: Blacklist 'GDI generic' GL renderer
17801
17802 If the native GL renderer is the GDI generic renderer (as can happen
17803 if we are in safe mode, or the video driver is VGA, or we have hybrid
17804 graphics which hasn't noticed that xwin requires 3d acceleration), don't
17805 use it. It's not accelerated and we will probably get better conformance
17806 and perfomance from swrast.
17807
17808 Fix so we don't install screen function wrappers in glxWinScreenProbe
17809 unless we are succesful.
17810
17811 Also, move fbConfig dumping to after GLX version has been determined
17812 from extensions
17813
17814 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
17815 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
17816
17817commit 8c24d20933ba9c2ef0e9a81815c3be23fa896aea
17818Author: Jon TURNEY <jon.turney@dronecode.org.uk>
17819Date: Thu Jun 21 18:05:24 2012 +0100
17820
17821 hw/xwin/glx: Fix glxLogExtensions to handle a null string without crashing
17822
17823 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
17824 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
17825
17826commit c08c7c8f655d7721c1e02bfeeb965b6379f72553
17827Author: Jon TURNEY <jon.turney@dronecode.org.uk>
17828Date: Fri Jun 24 13:32:10 2011 +0100
17829
17830 hw/xwin/glx: Create a new dispatch table rather than modifying the existing one
17831
17832 Create a new dispatch table rather than modifying the existing one
17833
17834 struct _glapi_table is not a complete type after including glapi.h, so we use
17835 glapi_get_dispatch_table_size() to determine it's size (alternatively, we could
17836 include glapitable.h, to complete the type)
17837
17838 This could possibly be written to use _glapi_create_table_from_handle() instead, but
17839 that requires making all the wrapper functions exports
17840
17841 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
17842 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
17843
17844commit 3ef3ce069d52dcfa932c90ccd30854a8d9daa15a
17845Merge: a0c8716 ff541e0
17846Author: Keith Packard <keithp@keithp.com>
17847Date: Thu Jun 21 08:52:34 2012 -0700
17848
17849 Merge remote-tracking branch 'alanc/master'
17850
17851commit 704b847abfd29e9adde27127a15a963414f8bcf4
17852Author: Chase Douglas <chase.douglas@canonical.com>
17853Date: Fri Apr 6 10:13:45 2012 -0700
17854
17855 Add FormatUInt64{,Hex}() for formatting numbers in a signal safe manner
17856
17857 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17858 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17859 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17860
17861commit bc85c81687a24aea738094ff11f4448fb3b3afbb
17862Author: Chase Douglas <chase.douglas@canonical.com>
17863Date: Fri Apr 6 08:03:09 2012 -0700
17864
17865 Save log file file descriptor for signal context logging
17866
17867 None of the FILE based functions are signal safe, including fileno(), so
17868 we need to save the file descriptor for when we are in signal context.
17869
17870 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17871 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17872 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17873
17874commit d3725549f0276487fba1d419094209d18e86669f
17875Author: Chase Douglas <chase.douglas@canonical.com>
17876Date: Fri Apr 6 07:43:57 2012 -0700
17877
17878 Add global variable inSignalContext
17879
17880 This will be used for checking for proper logging when in signal
17881 context.
17882
17883 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
17884 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17885 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17886
17887commit a0c8716cdca139059e7aec8a3cb75726dd37a4a1
17888Author: Peter Hutterer <peter.hutterer@who-t.net>
17889Date: Thu Jun 21 13:45:06 2012 +1000
17890
17891 kdrive: adjust BlockHandler/WakeupHandler to new API
17892
17893 Missing API update caused Xephyr to crash on startup, garbage memory
17894 dereference when accessing timeout.
17895 Introduced in 1f0e8bd5eb1a5539689cfc4f5a6b86b530907ec5
17896
17897 kdrive.c:868:27: warning: assignment from incompatible pointer type
17898 kdrive.c:869:28: warning: assignment from incompatible pointer type
17899
17900 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17901 Signed-off-by: Keith Packard <keithp@keithp.com>
17902
17903commit ff541e0a1f84bd98966148793f35afeaf71c271b
17904Author: Alan Coopersmith <alan.coopersmith@oracle.com>
17905Date: Wed Jun 13 21:20:51 2012 -0700
17906
17907 Remove obsolete tab stop comments from hw/xfree86/parser/*.c
17908
17909 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17910 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17911
17912commit 687536b1044a3297f6b9a45f6a2d2987daa7be59
17913Author: Alan Coopersmith <alan.coopersmith@oracle.com>
17914Date: Tue Jun 12 22:38:46 2012 -0700
17915
17916 Fix some overly indented/poorly line wrapped comments in dix/events.c
17917
17918 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17919 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17920 Reviewed-by: Keith Packard <keithp@keithp.com>
17921
17922commit 93a378aad4a4a03f37ddafccfedeb9d6e2be2a6a
17923Author: Alan Coopersmith <alan.coopersmith@oracle.com>
17924Date: Tue Jun 12 22:29:51 2012 -0700
17925
17926 OtherClientGone: Remove unreachable return statement
17927
17928 Now that FatalError is marked as _X_NORETURN, the compilers know we
17929 can't get here, and the return statement added to make them happy in
17930 the past now makes them unhappy.
17931
17932 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
17933 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
17934 Reviewed-by: Keith Packard <keithp@keithp.com>
17935
17936commit 3f97284b10c250457888902debd4d793cb4544d4
17937Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
17938Date: Wed Jun 20 13:53:53 2012 +1000
17939
17940 dri2: Pass a ScreenPtr through to the driver's AuthMagic function. (v3)
17941
17942 xwayland drivers need access to their screen private data to authenticate.
17943 Now that drivers no longer have direct access to the global screen arrays,
17944 this needs to be passed in as function context.
17945
17946 v2: Don't break ABI
17947 v3: Paint the bikeshed blue; drop fd from AuthMagic2ProcPtr prototype
17948
17949 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
17950 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
17951 Signed-off-by: Keith Packard <keithp@keithp.com>
17952
17953commit dae317e7265007b38012244722e3b3a06e904ed5
17954Author: Andy Ritger <aritger@nvidia.com>
17955Date: Thu Jun 14 09:15:37 2012 -0700
17956
17957 randr: Don't recurse into mieqProcessInputEvents() from RRTellChanged().
17958
17959 Call UpdateCurrentTimeIf(), not UpdateCurrentTime(), from RRTellChanged().
17960 The latter calls ProcessInputEvents(), which can trigger a recursion
17961 into mieqProcessInputEvents(). The former omits the call to
17962 ProcessInputEvents().
17963
17964 Signed-off-by: Andy Ritger <aritger@nvidia.com>
17965 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
17966 Signed-off-by: Keith Packard <keithp@keithp.com>
17967
17968commit 54476b5e4461ff523e935961affabcf0de12c556
17969Author: Peter Hutterer <peter.hutterer@who-t.net>
17970Date: Thu Jun 7 16:52:20 2012 +1000
17971
17972 dix: if the scroll valuator reaches INT_MAX, reset to 0
17973
17974 Too much scrolling down may eventually trigger an overflow of the valuator.
17975 If this happens, reset the valuator to 0 and skip this event for button
17976 emulation. Clients will have to figure out a way to deal with this, but a
17977 scroll event from (close to) INT_MAX to 0 is a hint of that it needs to be
17978 ignored.
17979
17980 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
17981 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
17982
17983commit 8dc70acbf3d82611ac9ec1ec2a52edcc01934850
17984Merge: ffb47a1 b59adc9
17985Author: Keith Packard <keithp@keithp.com>
17986Date: Thu Jun 14 13:04:29 2012 -0700
17987
17988 Merge remote-tracking branch 'alanc/master'
17989
17990commit ffb47a123ddd1233fb4229cf23483652065c5e82
17991Merge: db9d2b8 b840ba5
17992Author: Keith Packard <keithp@keithp.com>
17993Date: Thu Jun 14 13:01:17 2012 -0700
17994
17995 Merge remote-tracking branch 'idr/GLX_ARB_create_context'
17996
17997commit db9d2b8a508ab812e5c717a41310faad81879b09
17998Author: Dave Airlie <airlied@redhat.com>
17999Date: Thu Jun 7 15:01:59 2012 +0100
18000
18001 xserver: fix build on arm tinderbox
18002
18003 Rob Clark got an ARM tinderbox up and running, and this code is built there
18004 but not here, this should fix it, though I hope that code never gets executed.
18005
18006 Signed-off-by: Dave Airlie <airlied@redhat.com>
18007 Signed-off-by: Keith Packard <keithp@keithp.com>
18008
18009commit 6d86b64dbaef5a16712fd1fdc3157f716d238877
18010Merge: 812786f 4c68f5d
18011Author: Keith Packard <keithp@keithp.com>
18012Date: Thu Jun 14 11:05:22 2012 -0700
18013
18014 Merge remote-tracking branch 'whot/for-keith'
18015
18016commit b59adc9c2427e5efac3513c20992723e23f92691
18017Author: Alan Coopersmith <alan.coopersmith@oracle.com>
18018Date: Thu Jun 7 20:58:01 2012 -0700
18019
18020 Make stub version of fbdevHWAdjustFrame match new prototype in fbdevhw.h
18021
18022 Removed flags argument to match api rework in commit 1f0e8bd5eb1a55
18023
18024 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18025 Reviewed-by: Dave Airlie <airlied@redhat.com>
18026
18027commit 4c68f5d395c66f28b56e488cb3cd12f36820357b
18028Author: Peter Hutterer <peter.hutterer@who-t.net>
18029Date: Wed May 9 09:21:28 2012 +1000
18030
18031 dix: disable all devices before shutdown
18032
18033 f3410b97cf9b48a47bee3d15d232f8a88e75f4ef introduced a regression on server
18034 shutdown. If any button or key was held on shutdown (ctrl, alt, backspace
18035 are usually still down) sending a raw event will segfault the server. The
18036 the root windows are set to NULL before calling CloseDownDevices().
18037
18038 Avoid this by disabling all devices first when shutting down. Disabled
18039 devices won't send events anymore.
18040
18041 Master keyboards must be disabled first, otherwise disabling the pointer
18042 will trigger DisableDevice(keyboard) and the keyboard is removed from the
18043 inputInfo.devices list and moved to inputInfo.off_devices. A regular loop
18044 through inputInfo.devices would thus jump to off_devices and not recover.
18045
18046 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18047 Acked-by: Chase Douglas <chase.douglas@canonical.com>
18048 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18049
18050commit 9c0e820216cd1631f75b037b7908d55ac091692c
18051Author: Peter Hutterer <peter.hutterer@who-t.net>
18052Date: Thu May 10 12:55:44 2012 +1000
18053
18054 Xi: drop forced unpairing when changing the hierarchy
18055
18056 Devices are unpaired as needed on DisableDevice now.
18057
18058 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18059 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18060
18061commit e433d1046c222f9d969c2c28a4651ff9097614f4
18062Author: Peter Hutterer <peter.hutterer@who-t.net>
18063Date: Thu May 10 12:42:59 2012 +1000
18064
18065 dix: disable non-sprite-owners first when disabling paired devices
18066
18067 If a sprite-owner is to be disabled but still paired, disable the paired
18068 device first. i.e. disabling a master pointer will disable the master
18069 keyboard first.
18070
18071 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18072 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18073
18074commit df1704365e700d3cf1d36a241bdfc479159a8df7
18075Author: Peter Hutterer <peter.hutterer@who-t.net>
18076Date: Thu May 10 15:33:15 2012 +1000
18077
18078 dix: free the sprite when disabling the device
18079
18080 Disabled devices don't need sprites (they can't send events anyway) and the
18081 device init process is currently geared to check for whether sprite is
18082 present to check if the device should be paired/attached.
18083
18084 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18085 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18086
18087commit e57d6a89027c55fef987cdc259668c48a8b4ea1b
18088Author: Peter Hutterer <peter.hutterer@who-t.net>
18089Date: Thu May 10 15:32:20 2012 +1000
18090
18091 dix: move freeing the sprite into a function
18092
18093 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18094 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18095
18096commit 46adcefb0e08515195d8e49985a4e210395700b3
18097Author: Peter Hutterer <peter.hutterer@who-t.net>
18098Date: Thu May 10 12:10:12 2012 +1000
18099
18100 dix: return early from DisableDevice if the device is already disabled
18101
18102 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18103 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18104
18105commit 076f9d3a6660fa7fef4ae9decaffd2feebdd32c4
18106Author: Peter Hutterer <peter.hutterer@who-t.net>
18107Date: Thu May 10 12:20:37 2012 +1000
18108
18109 dix: drop client argument from PairDevices - unused
18110
18111 This is a leftover from early MPX days where any keyboard could be paired
18112 with any pointer (before the device hierarchy).
18113
18114 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18115 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18116
18117commit d034605f3914e41e0d636eafff9278be1c041b2c
18118Author: Peter Hutterer <peter.hutterer@who-t.net>
18119Date: Thu May 10 14:13:20 2012 +1000
18120
18121 dix: rename shadowing loop variable in TouchResizeQueue
18122
18123 second 'i' shadows the function-wide one, rename to 'j'
18124
18125 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18126 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18127
18128commit a86f3ef511db16be5ddc723bf14e770dad23ee45
18129Author: Peter Hutterer <peter.hutterer@who-t.net>
18130Date: Thu May 10 11:55:21 2012 +1000
18131
18132 dix: change "still paired warning" to use BUG_RETURN_VAL
18133
18134 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18135 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18136
18137commit 4bcf43919a14cd2cd20af4099cd213ce25792edb
18138Author: Peter Hutterer <peter.hutterer@who-t.net>
18139Date: Thu May 10 15:32:47 2012 +1000
18140
18141 dix: use BUG_RETURN_VAL for pairing errors
18142
18143 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18144 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18145 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18146
18147commit 642569fc79a1814acca1c8f529539b054bf36907
18148Author: Peter Hutterer <peter.hutterer@who-t.net>
18149Date: Thu May 10 12:11:57 2012 +1000
18150
18151 Replace a few BUG_WARN with BUG_RETURN_VAL
18152
18153 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18154 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18155
18156commit 11ecfb8884b7c4def0863a57868a6b9fa1b63ae5
18157Author: Peter Hutterer <peter.hutterer@who-t.net>
18158Date: Thu May 10 11:48:05 2012 +1000
18159
18160 include: add BUG_RETURN_* macros
18161
18162 Helper functions to avoid things like
18163
18164 if (foo) {
18165 BUG_WARN(foo);
18166 return 1;
18167 }
18168
18169 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18170 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
18171 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18172
18173commit ff41753b1bee414b16c0f1e8d183776e87c94065
18174Author: Peter Hutterer <peter.hutterer@who-t.net>
18175Date: Wed Jun 6 14:38:49 2012 +1000
18176
18177 xkb: warn if XKB SlowKeys have been automatically enabled
18178
18179 Slow keys are enabled when the XKB AccessX features are generally enabled
18180 (ctrls->enabled_ctrls & XkbAccessXKeysMask) and either shift key is held for
18181 8 seconds. For the unsuspecting user this appears as if the keyboard
18182 suddenly stops working.
18183
18184 Print a warning to the log, so we can later tell them "told you so".
18185
18186 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18187 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
18188
18189commit f20a532ba3c7b7d74530f3906c7bacf016bb8faf
18190Author: Alan Coopersmith <alan.coopersmith@oracle.com>
18191Date: Sun May 20 00:00:59 2012 -0700
18192
18193 Fix statement not reached warning in _DMXXineramaActive
18194
18195 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18196 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
18197
18198commit 1be317c4e5d99d11a124ad4e7a90a5dcae591e20
18199Author: Alan Coopersmith <alan.coopersmith@oracle.com>
18200Date: Wed May 23 19:27:23 2012 -0700
18201
18202 Provide prototypes for Mmio functions for Solaris Studio on SPARC
18203
18204 Actual inline implementations are provided via external *.il files
18205 generated from *.S files in hw/xfree86/os-support/solaris
18206
18207 Fixes missing prototype warnings that xorg-macros has recently elevated
18208 to build-breaking errors.
18209
18210 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18211 Signed-off-by: Henry Zhao <henry.zhao@oracle.com>
18212 Reviewed-by: Dave Airlie <airlied@redhat.com>
18213
18214commit 812786f4d4306cb16f8ed57fa4a1a32bb1d13fe3
18215Author: Dave Airlie <airlied@redhat.com>
18216Date: Tue Jun 5 19:55:52 2012 +0100
18217
18218 xwin: fixup block/wakeup handlers
18219
18220 These got missed out in the api changeover.
18221
18222 Reported-by: Colin Harrison <colin.harrison@virgin.net>
18223 Signed-off-by: Dave Airlie <airlied@redhat.com>
18224 Signed-off-by: Keith Packard <keithp@keithp.com>
18225
18226commit 3cbaf621782fb5f5679acf9104571d3ae2b48b72
18227Author: Dave Airlie <airlied@redhat.com>
18228Date: Tue Jun 5 17:54:02 2012 +0100
18229
18230 xfree86: fix new warnings introduced in rebase.
18231
18232 Spotted these in tinderbox:
18233 xf86xv.c: In function 'xf86XVScreenInit':
18234 xf86xv.c:282: warning: assignment from incompatible pointer type
18235 xf86xv.c: In function 'xf86XVCloseScreen':
18236 xf86xv.c:1275: warning: assignment from incompatible pointer type
18237 xf86xv.c: In function 'xf86XVAdjustFrame':
18238 xf86xv.c:1366: warning: assignment from incompatible pointer type
18239
18240 Missed removing one flags.
18241
18242 Signed-off-by: Dave Airlie <airlied@redhat.com>
18243 Tested-by: Keith Packard <keithp@keithp.com>
18244 Signed-off-by: Keith Packard <keithp@keithp.com>
18245
18246commit 1f0e8bd5eb1a5539689cfc4f5a6b86b530907ec5
18247Author: Dave Airlie <airlied@redhat.com>
18248Date: Tue Jun 5 13:22:18 2012 +0100
18249
18250 api: rework the X server driver API to avoid global arrays.
18251
18252 This is a squash merge containing all the API changes, as
18253 well as the video ABI bump.
18254
18255 Its been squashed to make bisection easier.
18256
18257 Full patch log below:
18258
18259 commit b202738bbf0c5a1c1172767119c2c71f1e7f8070
18260 Author: Aaron Plattner <aplattner@nvidia.com>
18261 Date: Mon May 14 15:16:11 2012 -0700
18262
18263 xfree86: Bump video ABI to 13.0
18264
18265 The ABI was broken by changes to convert from screen index numbers to ScreenPtr
18266 / ScrnInfoPtr in various structures and function signatures.
18267
18268 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
18269 Reviewed-by: Dave Airlie <airlied@redhat.com>
18270 Signed-off-by: Dave Airlie <airlied@redhat.com>
18271
18272 commit 3d5f7d9f8d408bcad3f83277d255f25d3b0edbf3
18273 Author: Dave Airlie <airlied@redhat.com>
18274 Date: Thu May 24 10:56:57 2012 +0100
18275
18276 xf86: xf86ClearEntityListForScreen should take a pScrn
18277
18278 When adding GPU screens this make life easier.
18279
18280 (also fix comment, as pointed out by Alan)
18281
18282 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18283 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
18284 Signed-off-by: Dave Airlie <airlied@redhat.com>
18285
18286 commit afee8b5ab4501597ecc1ade34124d7ca227ab055
18287 Author: Dave Airlie <airlied@redhat.com>
18288 Date: Thu May 24 07:07:32 2012 +0100
18289
18290 xf86i2c: add pscrn for drivers to use
18291
18292 This just adds a pScrn pointer into the struct for the drivers to use
18293 instead of scrnIndex. Mostly scrnIndex is used for logging, but some
18294 drivers use it to lookup xf86Screens, so let them stash a pScrn instead.
18295
18296 Removing the scrnIndex is a bit more involved and I'm not sure its worth
18297 the effort. Doing i2c in the X server is legacy code as far as I'm concerned.
18298
18299 Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
18300 Signed-off-by: Dave Airlie <airlied@redhat.com>
18301
18302 commit ea5092f1f679691d187f1eee9427e6057beec56e
18303 Author: Dave Airlie <airlied@redhat.com>
18304 Date: Wed May 23 19:25:20 2012 +0100
18305
18306 dix/gc: consolidate GC object creation in one place
18307
18308 The standard GC create and scratch GC create were 90% the same really,
18309 and I have a need in the future for creating GC objects without the
18310 other bits, so wanted to avoid a third copy.
18311
18312 Reviewed-by: Keith Packard <keithp@keithp.com>
18313 Signed-off-by: Dave Airlie <airlied@redhat.com>
18314
18315 commit 3d91482ea9b4883e64e496f2768168e0ffa21ba1
18316 Author: Dave Airlie <airlied@redhat.com>
18317 Date: Wed May 23 10:24:06 2012 +0100
18318
18319 xf86: add a define to denote the new non-index interfaces are being used
18320
18321 This can be used by drivers to provide compatible APIs.
18322
18323 Reviewed-by: Keith Packard <keithp@keithp.com>
18324 Signed-off-by: Dave Airlie <airlied@redhat.com>
18325
18326 commit 37c3ae3e6cd4f3dedc72f371096d6743f8f99df3
18327 Author: Dave Airlie <airlied@redhat.com>
18328 Date: Wed May 23 15:09:12 2012 +0100
18329
18330 dix: make Create/Free scratch pixmaps take a ScreenPtr
18331
18332 While technically an API/ABI change I doubt anyone uses it,
18333 but it helps in splitting screens up.
18334
18335 Reviewed-by: Keith Packard <keithp@keithp.com>
18336 Signed-off-by: Dave Airlie <airlied@redhat.com>
18337
18338 commit 75f2062a3fe94f04764ecc7d2ff2fbbeccb9da60
18339 Author: Dave Airlie <airlied@redhat.com>
18340 Date: Wed May 23 14:57:55 2012 +0100
18341
18342 xf86/xv: remove scrnIndexfrom xf86FindXvOptions.
18343
18344 Move this interface to taking an ScrnInfoPtr.
18345
18346 Reviewed-by: Keith Packard <keithp@keithp.com>
18347 Signed-off-by: Dave Airlie <airlied@redhat.com>
18348
18349 commit f80c2374f40ea7b2ee0556e2e76cc07406f3d843
18350 Author: Dave Airlie <airlied@redhat.com>
18351 Date: Wed May 23 14:53:59 2012 +0100
18352
18353 xf86: make xf86DeleteScreen take a ScrnInfoPtr (v2)
18354
18355 stop passing indices into this function.
18356
18357 v2: drop flags argument.
18358
18359 Reviewed-by: Keith Packard <keithp@keithp.com>
18360 Signed-off-by: Dave Airlie <airlied@redhat.com>
18361
18362 commit 58824e414f35682435f15bfe6c4b656bd90b9235
18363 Author: Dave Airlie <airlied@redhat.com>
18364 Date: Wed May 23 14:48:09 2012 +0100
18365
18366 xf86: fix xf86IsScreenPrimary interface to take a pScrn (API/ABI)
18367
18368 Reviewed-by: Keith Packard <keithp@keithp.com>
18369 Signed-off-by: Dave Airlie <airlied@redhat.com>
18370
18371 commit 6b4fc1f9d391bcdf7ca288766e49bce60f4635cd
18372 Author: Dave Airlie <airlied@gmail.com>
18373 Date: Tue Apr 10 14:18:59 2012 +0100
18374
18375 xserver: convert block/wakeup handlers to passing ScreenPtr (ABI/API) (v2)
18376
18377 Instead of passing an index, pass the actual ScreenPtr. This allows
18378 more moving towards not abusing xf86Screens + screenInfo.
18379
18380 v2: drop the blockData/wakeupData args as per ajax's suggestion.,
18381 fix docs.
18382
18383 Reviewed-by: Keith Packard <keithp@keithp.com>
18384 Signed-off-by: Dave Airlie <airlied@redhat.com>
18385
18386 commit 790d003de20fb47674420a24dadd92412d78620d
18387 Author: Dave Airlie <airlied@gmail.com>
18388 Date: Wed Apr 11 09:53:14 2012 +0100
18389
18390 xf86/common: remove some more pScrn->pScreen uses
18391
18392 remove some more conversions that appeared after api cleanups.
18393
18394 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18395 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18396 Reviewed-by: Adam Jackson <ajax@redhat.com>
18397 Signed-off-by: Dave Airlie <airlied@redhat.com>
18398
18399 commit aac85e18d1dd093f2cad6bd29375e40bd7af0b8f
18400 Author: Dave Airlie <airlied@gmail.com>
18401 Date: Tue Apr 10 16:34:53 2012 +0100
18402
18403 ddc: change API to take ScrnInfoPtr (v2)
18404
18405 This removes all xf86Screens usage from ddc code,
18406 it modifies the API for some functions to avoid taking indices.
18407
18408 v2: address Alan's comments about dropping DDC2Init parameter.
18409
18410 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18411 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18412 Reviewed-by: Adam Jackson <ajax@redhat.com>
18413 Signed-off-by: Dave Airlie <airlied@redhat.com>
18414
18415 commit fe3f57b6eaf6860a33876a54f9439f69578f03a5
18416 Author: Dave Airlie <airlied@gmail.com>
18417 Date: Tue Apr 10 16:31:26 2012 +0100
18418
18419 vbe: don't use index for VBEInterpretPanelID (API)
18420
18421 Remove use of xf86screens from vbe module.
18422
18423 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18424 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18425 Reviewed-by: Adam Jackson <ajax@redhat.com>
18426 Signed-off-by: Dave Airlie <airlied@redhat.com>
18427
18428 commit abf1965f4ed91529036d3fdb470d6a3ce6f29675
18429 Author: Dave Airlie <airlied@gmail.com>
18430 Date: Tue Apr 10 16:25:11 2012 +0100
18431
18432 int10/vbe: don't use xf86Screens. (ABI) (v3)
18433
18434 Pass the ScrnInfoPtr instead of the index in the int10 struct.
18435
18436 This saves us using it to dereference xf86Screens.
18437
18438 v2: address Alan's comment to fix struct alignment.
18439
18440 v3: squash in all the int10 fixes, test the vm86 code builds,
18441 after comments by Keith.
18442
18443 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18444 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18445 Reviewed-by: Adam Jackson <ajax@redhat.com>
18446 Signed-off-by: Dave Airlie <airlied@redhat.com>
18447
18448 commit 23cca612b4fb5efc33683c7624b803b457387e3d
18449 Author: Dave Airlie <airlied@gmail.com>
18450 Date: Tue Apr 10 15:30:18 2012 +0100
18451
18452 xserver: drop index argument to ScreenInit (ABI/API) (v2)
18453
18454 This drops the index argument, its the same as pScreen->myNum,
18455 and its the last major index abuse I can find.
18456
18457 v2: address Alan's review - update docs, fix xwin/xnest/darwin
18458
18459 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18460 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18461 Reviewed-by: Adam Jackson <ajax@redhat.com>
18462 Signed-off-by: Dave Airlie <airlied@redhat.com>
18463
18464 commit 40d360e2d7e832407f3ed64e3a02c27ecc89a960
18465 Author: Dave Airlie <airlied@gmail.com>
18466 Date: Tue Apr 10 15:23:01 2012 +0100
18467
18468 xf86: migrate PointerMoved from index to ScrnInfoPtr (ABI/API)
18469
18470 This migrates PointerMoved from an index to ScrnInfoPtr.
18471
18472 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18473 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18474 Reviewed-by: Adam Jackson <ajax@redhat.com>
18475 Signed-off-by: Dave Airlie <airlied@redhat.com>
18476
18477 commit aa60a2f38679d0eeb979a9c2648c9bc771409bf9
18478 Author: Dave Airlie <airlied@gmail.com>
18479 Date: Tue Apr 10 15:20:46 2012 +0100
18480
18481 xf86: migrate PMEvent to a ScrnInfoPtr (ABI/API)
18482
18483 This migrates the PMEvent from index to ScrnInfoPtr.
18484
18485 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18486 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18487 Reviewed-by: Adam Jackson <ajax@redhat.com>
18488 Signed-off-by: Dave Airlie <airlied@redhat.com>
18489
18490 commit d3f28ef44371ed4a039ffc5dd7eb6408d1269ba2
18491 Author: Dave Airlie <airlied@gmail.com>
18492 Date: Tue Apr 10 15:18:30 2012 +0100
18493
18494 xf86: migrate SetDGAMode from index to ScrnInfoPtr (ABI/API)
18495
18496 This migrates the SetDGAMode callback from an index to ScrnInfoPtr.
18497
18498 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18499 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18500 Reviewed-by: Adam Jackson <ajax@redhat.com>
18501 Signed-off-by: Dave Airlie <airlied@redhat.com>
18502
18503 commit baf5e4818a74f2b68c3dfdcc56f54322351039a0
18504 Author: Dave Airlie <airlied@gmail.com>
18505 Date: Tue Apr 10 15:14:11 2012 +0100
18506
18507 xf86: migrate ChangeGamma from index to ScrnInfoPtr (ABI/API) (v2)
18508
18509 This migrates the ChangeGamma interface to avoid passing a index.
18510
18511 v2: fix xf86RandR12.c + xf86cmap.c call
18512
18513 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18514 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18515 Reviewed-by: Adam Jackson <ajax@redhat.com>
18516 Signed-off-by: Dave Airlie <airlied@redhat.com>
18517
18518 commit 51e5f90ada929d6b23176090badbb42fdb3fa550
18519 Author: Dave Airlie <airlied@gmail.com>
18520 Date: Tue Apr 10 15:11:09 2012 +0100
18521
18522 xf86/exa: migrate index to screen types for EnableDisableFBAccess (ABI/API)
18523
18524 The EXA interface migrates to ScreenPtr,
18525 and the xf86 interface migrated to ScrnInfoPtr.
18526
18527 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18528 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18529 Reviewed-by: Adam Jackson <ajax@redhat.com>
18530 Signed-off-by: Dave Airlie <airlied@redhat.com>
18531
18532 commit 94f1f21d17e86f96d4a54292a399160950087675
18533 Author: Dave Airlie <airlied@gmail.com>
18534 Date: Tue Apr 10 15:02:11 2012 +0100
18535
18536 xf86: migrate ValidMode callback to ScrnInfoPtr (ABI/API)
18537
18538 This migrates the ValidMode to passing a ScrnInfoPtr instead
18539 of an index.
18540
18541 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18542 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18543 Reviewed-by: Adam Jackson <ajax@redhat.com>
18544 Signed-off-by: Dave Airlie <airlied@redhat.com>
18545
18546 commit 3f8f18198fed4f39ec805b508a3482e91eea26b2
18547 Author: Dave Airlie <airlied@gmail.com>
18548 Date: Tue Apr 10 14:59:46 2012 +0100
18549
18550 xf86: migrate SwitchMode to taking ScrnInfoPtr (ABI/API) (v2)
18551
18552 This migrate the SwitchMode interface to take a ScrnInfoPtr
18553 instead of an index.
18554
18555 v2: drop flags.
18556
18557 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18558 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18559 Reviewed-by: Adam Jackson <ajax@redhat.com>
18560 Signed-off-by: Dave Airlie <airlied@redhat.com>
18561
18562 commit d06a038a5c49328ab3a8d969d24f9fcd22c63202
18563 Author: Dave Airlie <airlied@gmail.com>
18564 Date: Tue Apr 10 14:50:37 2012 +0100
18565
18566 xf86: move AdjustFrame to passing ScrnInfoPtr (ABI/API) (v2)
18567
18568 This converts AdjustFrame code paths to passing a ScrnInfoPtr
18569 instead of an integer index.
18570
18571 v2: drop flags args.
18572
18573 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18574 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18575 Reviewed-by: Adam Jackson <ajax@redhat.com>
18576 Signed-off-by: Dave Airlie <airlied@redhat.com>
18577
18578 commit 53d2f8608ffd4090d08e7d5cf2e92fb954959b90
18579 Author: Dave Airlie <airlied@gmail.com>
18580 Date: Tue Apr 10 14:41:27 2012 +0100
18581
18582 xf86: modify FreeScreen callback to take pScrn instead of index. (ABI/API) (v2)
18583
18584 Another index->pScrn conversion.
18585
18586 v2: drop flags arg.
18587
18588 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18589 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18590 Reviewed-by: Adam Jackson <ajax@redhat.com>
18591 Signed-off-by: Dave Airlie <airlied@redhat.com>
18592
18593 commit 60db37c0b247052e0f5c54b1921fe58a3609c2e3
18594 Author: Dave Airlie <airlied@gmail.com>
18595 Date: Tue Apr 10 14:35:41 2012 +0100
18596
18597 xf86: change EnterVT/LeaveVT to take a ScrnInfoPtr (ABI/API break) (v2)
18598
18599 This modifies the EnterVT/LeaveVT interfaces to take a ScrnInfoPtr
18600 instead of an index into xf86Screens. This allows dropping more
18601 public dereferences of the xf86Screens and screenInfo.
18602
18603 v2: drop flags args as suggested by Keith, fix docs.
18604
18605 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18606 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18607 Reviewed-by: Adam Jackson <ajax@redhat.com>
18608 Signed-off-by: Dave Airlie <airlied@redhat.com>
18609
18610 commit 06729dbbc804a20242e6499f446acb5d94023c3c
18611 Author: Dave Airlie <airlied@gmail.com>
18612 Date: Tue Apr 10 14:04:59 2012 +0100
18613
18614 xserver: remove index from CloseScreen (API/ABI breakage)
18615
18616 This drops the index from the CloseScreen callback,
18617 its always been useless really, since the pScreen contains it.
18618
18619 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18620 Acked-by: Aaron Plattner <aplattner@nvidia.com>
18621 Reviewed-by: Adam Jackson <ajax@redhat.com>
18622 Signed-off-by: Dave Airlie <airlied@redhat.com>
18623
18624commit 12f9aaff38ac2cf528387ce610e41fbd448698e2
18625Author: Chase Douglas <chase.douglas@canonical.com>
18626Date: Fri Apr 6 12:44:07 2012 -0700
18627
18628 test/touch: Initialize device name
18629
18630 Without this change, the test will segfault when we switch to signal-
18631 safe logging.
18632
18633 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
18634 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
18635
18636commit 3476eb38063473a7a5fcd78e2095e284118de839
18637Merge: 78f0d9c 42ae2e8
18638Author: Keith Packard <keithp@keithp.com>
18639Date: Sun Jun 3 20:24:24 2012 -0700
18640
18641 Merge remote-tracking branch 'whot/for-keith'
18642
18643commit 42ae2e8199fe875319978249963de7499607988b
18644Author: Siddhesh Poyarekar <siddhesh.poyarekar@gmail.com>
18645Date: Tue May 29 10:17:50 2012 +0530
18646
18647 xkb: Allocate size_syms correctly when width of a type increases
18648
18649 The current code seems to skip syms with width less than
18650 type->num_levels when calculating the total size for the new
18651 size_syms. This leads to less space being allocated than necessary
18652 during the next phase, which is to copy over the syms to the new
18653 location. This results in an overflow leading to a crash.
18654
18655 Signed-off-by: Siddhesh Poyarekar <siddhesh.poyarekar@gmail.com>
18656 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18657
18658commit 88c767edb01ed7efb19ffe3a453e16107b27130b
18659Author: Michal Suchanek <hramrach@gmail.com>
18660Date: Wed May 23 11:22:05 2012 +0200
18661
18662 Fix crash for motion events from devices without valuators
18663
18664 A WarpPointer request may trigger a motion event on a device without
18665 valuators. That request is ignored by GetPointerEvents but during smooth
18666 scroll emulation we dereference dev->valuators to get the number of axes.
18667
18668 Break out early if the device doesn't have valuators.
18669
18670 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
18671 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
18672 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18673
18674commit 78f0d9cdc441a575e15e5f7200ce74750d2f7ad2
18675Author: Chad Versace <chad.versace@linux.intel.com>
18676Date: Wed May 2 16:18:49 2012 -0700
18677
18678 dri2: Add DRI2GetParam request (v2)
18679
18680 Bump dri2proto dependency to 2.7.
18681 Bump DRI2INFOREC_VERSION to 7.
18682
18683 This new protocol request effectively allows clients to perform feature
18684 detection on the DDX. The request was added in DRI2 protocol 1.4.
18685
18686 If I had DRI2GetParam in June 2011, when I was implementing support in the
18687 Intel DDX and Mesa for new hardware that required a new DRI2 attachment
18688 format, then I could have avoided a week of pain caused by the necessity
18689 to write a horrid feature detection hack [1] in Mesa. In the future, when
18690 the work begins to add MSAA support to the Intel DDX, having a clean way
18691 to do feature detection will allow us to avoid revisiting and expanding
18692 that hack.
18693
18694 [1] mesa, commit aea2236a, function intel_verify_dri2_has_hiz
18695
18696 v2: If driver doesn't define ds->GetParam, dont' crash. Fall back to
18697 default behavior, per keithp.
18698
18699 Reviewed-by: Keith Packard <keithp@keithp.com>
18700 Reviewed-by: Eric Anholt <eric@anholt.net> (v1)
18701 Reviewed-by: Ian Romanick <idr@freedesktop.org> (v1)
18702 Signed-off-by: Chad Versace <chad.versace@linux.intel.com>
18703
18704commit 594b4a4f0601066d09e49a84c5af8c2eb6f07240
18705Author: Peter Hutterer <peter.hutterer@who-t.net>
18706Date: Thu May 24 13:34:11 2012 +1000
18707
18708 Xext: include dix-config.h
18709
18710 silences
18711 In file included from ../include/misc.h:111:0, from hashtable.c:2:
18712 ./include/os.h:559:1: warning: redundant redeclaration of 'strndup' [-Wredundant-decls]
18713
18714 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18715 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
18716
18717commit d348ab06aae21c153ecbc3511aeafc8ab66d8303
18718Author: Julien Cristau <jcristau@debian.org>
18719Date: Fri May 11 21:31:50 2012 +0200
18720
18721 Xvfb: destroy the screen pixmap in CloseScreen
18722
18723 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18724 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
18725
18726commit e4153c1d9138ed40de1c615525066a0f5bb599dc
18727Author: Julien Cristau <jcristau@debian.org>
18728Date: Fri May 11 21:31:49 2012 +0200
18729
18730 Xi: make stub DeleteInputDeviceRequest call RemoveDevice
18731
18732 DeleteInputDeviceRequest is called from CloseDownDevices on reset, so
18733 call RemoveDevice to avoid leaking devices in Xvfb/Xnest/Xwin.
18734
18735 Signed-off-by: Julien Cristau <jcristau@debian.org>
18736 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18737 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
18738
18739commit b840ba5f54de5b00a7700ca49e51308b8cc66f92
18740Author: Ian Romanick <ian.d.romanick@intel.com>
18741Date: Fri Dec 16 14:44:55 2011 -0800
18742
18743 glx: Implement protocol for glXCreateContextAttribsARB
18744
18745 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18746 Reviewed-by: Dave Airlie <airlied@redhat.com>
18747 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18748
18749commit 23612a63fc12be13442a63193a19f883d7a87e5d
18750Author: Ian Romanick <ian.d.romanick@intel.com>
18751Date: Fri Dec 16 14:42:51 2011 -0800
18752
18753 glx: Make several functions available outside the glxcmds.c compilation unit
18754
18755 validGlxScreen, validGlxFBConfig, validGlxContext, and
18756 __glXdirectContextCreate will soon be used by createcontext.c.
18757
18758 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18759 Reviewed-by: Dave Airlie <airlied@redhat.com>
18760 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18761
18762commit cd5689cac5ed722e9dc1a3fab9a260b798062ee6
18763Author: Ian Romanick <ian.d.romanick@intel.com>
18764Date: Sat Dec 10 15:15:18 2011 -0800
18765
18766 glx: Use one function to add a context to all global tables
18767
18768 Instead of having separate __glXAddContextToList and AddResource
18769 functions, just have one function that does both steps.
18770
18771 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18772 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
18773 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18774
18775commit c1d91ab37025d3e0df7e8f647fb21816fe0e8420
18776Author: Ian Romanick <ian.d.romanick@intel.com>
18777Date: Sat Dec 10 15:11:57 2011 -0800
18778
18779 glx: Initialize remaining context fields
18780
18781 There is no reason to assume the screen's context allocated
18782 initialized these fields, so don't.
18783
18784 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18785 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
18786 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18787
18788commit 0db76e5f771fb964e67de716a254850a38e02f19
18789Author: Ian Romanick <ian.d.romanick@intel.com>
18790Date: Sat Dec 10 15:09:57 2011 -0800
18791
18792 glx: Initialize all context fields together
18793
18794 v2: Fix whitespace error noticed by Christopher James Halse Rogers.
18795
18796 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18797 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
18798 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18799
18800commit befa76d251eb5de85edb621056eed52784722bc7
18801Author: Ian Romanick <ian.d.romanick@intel.com>
18802Date: Fri Dec 9 17:28:21 2011 -0800
18803
18804 glx: Implement GLX SetClientInfoARB protocol
18805
18806 v2: Bump glproto version to 1.4.15. This patch uses structure names
18807 that only exist in that glproto version and later. Noticed by
18808 Christopher James Halse Rogers.
18809
18810 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18811 Reviewed-by: Dave Airlie <airlied@redhat.com>
18812 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18813
18814commit 8b8cd6f6f4dbf9d058ace638221f31801da7df9f
18815Author: Ian Romanick <ian.d.romanick@intel.com>
18816Date: Mon Dec 5 13:52:01 2011 -0800
18817
18818 glx: Optionally call DRI2 createContextAttribs from __glXDRIscreenCreateContext
18819
18820 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18821 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18822
18823commit 0a9226376c9a44c3b50baad25f2b7d2b9fecb708
18824Author: Ian Romanick <ian.d.romanick@intel.com>
18825Date: Mon Dec 5 13:48:19 2011 -0800
18826
18827 glx: Add tracking for GLX_ARB_create_context and GLX_ARB_create_context_profile
18828
18829 The formatting of the extension data table was mangled by indent.
18830 This patch also undoes that and adds approripate control comments so
18831 that indent won't do it again.
18832
18833 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18834 Reviewed-by: Dave Airlie <airlied@redhat.com>
18835 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18836
18837commit 96d74138b1c0273e88933220d99a893858b649cd
18838Author: Ian Romanick <ian.d.romanick@intel.com>
18839Date: Mon Dec 5 10:56:07 2011 -0800
18840
18841 glx: Extend __GLXscreen::createContext to take attributes
18842
18843 The attributes will be used for glXCreateContextAttribsARB additions
18844 in follow-on patches.
18845
18846 v2: Add missing 'int *error' parameters noticed by Christopher James
18847 Halse Rogers.
18848
18849 v3: Remove redundant 'int err;' declaration noticed by Christopher
18850 James Halse Rogers. This was supposed to be in v2, but I missed it.
18851
18852 v4: Add comma missing from additions in v2. Ugh.
18853
18854 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18855 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
18856 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18857
18858commit 62f06b0dcde13cb87bc3d0c5ff86d8db7e3ff0ef
18859Author: Ian Romanick <ian.d.romanick@intel.com>
18860Date: Fri Dec 2 21:14:46 2011 -0800
18861
18862 glx: Don't track GLClientmajorVersion or GLClientminorVersion
18863
18864 Nothing uses these fields anywhere in the server.
18865
18866 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18867 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
18868 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18869
18870commit 724176a87afdb9a79ce6307214bc2e869530aef3
18871Author: Ian Romanick <ian.d.romanick@intel.com>
18872Date: Mon Dec 5 10:55:20 2011 -0800
18873
18874 glx: Fix mishandling of shared contexts
18875
18876 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
18877 Reviewed-by: Eric Anholt <eric@anholt.net>
18878 Reviewed-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
18879
18880commit 2d4fda4b09e67e47d3e6fc4743fc6e81bfe40f28
18881Author: Marcin Slusarz <marcin.slusarz@gmail.com>
18882Date: Mon May 21 21:39:43 2012 +0200
18883
18884 xfree86: fix mouse wheel support for DGA clients
18885
18886 xf86-input-evdev (since "smooth scrolling" support was added) can send mouse
18887 motion and wheel events in one batch, so we need to handle it properly.
18888 Otherwise mouse wheel events which come with motion events are lost
18889 and separate mouse wheel events are handled through non-DGA path.
18890
18891 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
18892 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
18893 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
18894
18895commit 9bc53d8cb04af2be3feeebb1b10774c2d599a76b
18896Author: Michal Srb <msrb@suse.com>
18897Date: Tue May 22 21:19:40 2012 -0700
18898
18899 dri2: SProcDRI2Connect - send the response.
18900
18901 The swapped implementation of DRI2Connect is always responding with empty
18902 device and driver values. However the response was only prepared and never
18903 sent (also had undefined .type member), causing e.g. glxinfo get stuck waiting
18904 for response when started remotely from machine with different endianity.
18905
18906 Signed-off-by: Michal Srb <msrb@suse.com>
18907 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
18908 Tested-by: Michel Dänzer <michel@daenzer.net>
18909 Signed-off-by: Keith Packard <keithp@keithp.com>
18910
18911commit 457fc77691b2df66f6dd3ba10ce7c1b61b9b42a3
18912Merge: 2e237c8 5a3a98f
18913Author: Keith Packard <keithp@keithp.com>
18914Date: Tue May 22 10:32:29 2012 -0700
18915
18916 Merge remote-tracking branch 'alanc/master'
18917
18918commit 2e237c838fe1780c2fcbd2306f35fd15b6747831
18919Author: Dave Airlie <airlied@gmail.com>
18920Date: Wed Apr 11 10:02:25 2012 +0100
18921
18922 xf86: reimplement XF86SCRNINFO macro using new functions.
18923
18924 This macro did lookups via privates but we can just use the ScreenToScrn
18925 conversion instead.
18926
18927 This patch drops all in-server uses, we should drop the macro later,
18928 once drivers have been converted to not use it.
18929
18930 Reviewed-by: Adam Jackson <ajax@redhat.com>
18931 Reviewed-by: Keith Packard <keithp@keithp.com>
18932 Signed-off-by: Dave Airlie <airlied@redhat.com>
18933
18934commit 8da35202f390cd8a0410814911017d99fa9dddcb
18935Author: Dave Airlie <airlied@gmail.com>
18936Date: Wed Apr 11 09:53:35 2012 +0100
18937
18938 xf86/modes: drop more pScrn->pScreen usages
18939
18940 Just simple conversion to use the lookup function.
18941
18942 Reviewed-by: Adam Jackson <ajax@redhat.com>
18943 Reviewed-by: Keith Packard <keithp@keithp.com>
18944 Signed-off-by: Dave Airlie <airlied@redhat.com>
18945
18946commit e7a9e30fd20a7c2a526783726d0a76b8a04b053a
18947Author: Dave Airlie <airlied@gmail.com>
18948Date: Wed Apr 11 09:52:37 2012 +0100
18949
18950 xaa: convert pScrn->pScreen to use accessor.
18951
18952 This uses a standard conversion function to do the conversion.
18953
18954 Reviewed-by: Adam Jackson <ajax@redhat.com>
18955 Reviewed-by: Keith Packard <keithp@keithp.com>
18956 Signed-off-by: Dave Airlie <airlied@redhat.com>
18957
18958commit 738367ac9b6179593a38074d0105de206cbed296
18959Author: Dave Airlie <airlied@gmail.com>
18960Date: Wed Apr 11 09:46:39 2012 +0100
18961
18962 xf86/modes: drop two uses of screenInfo
18963
18964 Just use new macros to access scrn->screen.
18965
18966 Reviewed-by: Adam Jackson <ajax@redhat.com>
18967 Reviewed-by: Keith Packard <keithp@keithp.com>
18968 Signed-off-by: Dave Airlie <airlied@redhat.com>
18969
18970commit cc4fe613d0b0891ea9502340801b37e64301f573
18971Author: Dave Airlie <airlied@redhat.com>
18972Date: Sat May 5 09:47:24 2012 +0100
18973
18974 render/exa: use glyph picture accessors
18975
18976 use the glyph picture accessors in the X server, render and EXA code.
18977
18978 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
18979 Reviewed-by: Keith Packard <keithp@keithp.com>
18980 Signed-off-by: Dave Airlie <airlied@redhat.com>
18981
18982commit a45e1d6cc6f2a5f31ed737294351cfa251caa030
18983Author: Dave Airlie <airlied@redhat.com>
18984Date: Sat May 5 09:38:28 2012 +0100
18985
18986 render: add GetGlyphPicture accessor.
18987
18988 This is a new API to stop the drivers directly looking up the glyph pictures
18989 in a global array. It provides a define GLYPH_HAS_GLYPH_PICTURE_ACCESSOR for
18990 drivers to work in a compat way.
18991
18992 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
18993 Reviewed-by: Keith Packard <keithp@keithp.com>
18994 Signed-off-by: Dave Airlie <airlied@redhat.com>
18995
18996commit 41151f88a62f01b00c1912822b8b430df1eb307e
18997Author: Dave Airlie <airlied@gmail.com>
18998Date: Tue Apr 10 15:47:33 2012 +0100
18999
19000 xf86: migrate to using xf86ScreenToScrn wrapper (v2)
19001
19002 migrate to new helper API.
19003
19004 This just wraps all the obvious uses of xf86Screens[pScreen->myNum],
19005 and should be fairly simple to review.
19006
19007 v2: remove commented out lines.
19008
19009 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
19010 Reviewed-by: Keith Packard <keithp@keithp.com>
19011 Signed-off-by: Dave Airlie <airlied@redhat.com>
19012
19013commit 53932b3803fa2c02949fd7d4d0e433ea58fa89f1
19014Author: Dave Airlie <airlied@gmail.com>
19015Date: Tue Apr 10 15:47:32 2012 +0100
19016
19017 xf86: add helper functions to convert to from ScrnInfoPtr/ScreenPtr (v2)
19018
19019 These are just simple functions that we should start migrating drivers
19020 to using.
19021
19022 The end goal is to remove xf86Screens and screenInfo from the ABI.
19023
19024 This includes a define XF86_HAS_SCRN_CONV that drivers can ifdef to provide
19025 their own copies. I'll probably post a generic compat.h file for drivers later.
19026
19027 v2: add asserts.
19028
19029 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
19030 Reviewed-by: Keith Packard <keithp@keithp.com>
19031 Signed-off-by: Dave Airlie <airlied@redhat.com>
19032
19033commit 39f73e813f7d404498629f6104a9003d092af28d
19034Author: Dave Airlie <airlied@redhat.com>
19035Date: Tue May 8 13:01:12 2012 +0100
19036
19037 xf86/pci: fix slot claiming counting.
19038
19039 Currently if we claim a slot for a PCI driver, we never let it go properly,
19040 this prevents the fallback probe from reusing the slot, even though it
19041 isn't claimed for that pci slot.
19042
19043 So if you set the modesetting driver to point at a specific kms device,
19044 that isn't a PCI device (i.e. USB dongle), then the modesetting driver
19045 loads, the pci probe tries to bind the config slot to the primary PCI
19046 device, however we then check the kms device bus id to discover it
19047 isn't valid. However we don't remove the claim on the slot. Next the
19048 old probe function is called and there is no slots to claim.
19049
19050 This patch fixes that and converts the pciSlotClaimed boolean into
19051 a counter, and changes the unclaim api to take a device pointer
19052 to remove from the entity.
19053
19054 Reviewed-by: Adam Jackson <ajax@redhat.com>
19055 Signed-off-by: Dave Airlie <airlied@redhat.com>
19056
19057commit 07dcc3f1a98dff2ee374a15ac5beac778d2ccc97
19058Author: Dave Airlie <airlied@redhat.com>
19059Date: Sun May 6 17:13:02 2012 +0100
19060
19061 config/udev: add pre_init stage to config and udev.
19062
19063 In order to use udev for gpu enumeration, we need to init udev earlier
19064 than input initialisations. This splits the config init stuff so that udev
19065 pre init sets up before output initialisation.
19066
19067 this is just a prepatory patch, doesn't change anything major.
19068
19069 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19070 Signed-off-by: Dave Airlie <airlied@redhat.com>
19071
19072commit a2a02882ab65133e6c0c69db1f38bc20b406236f
19073Author: Dave Airlie <airlied@redhat.com>
19074Date: Sun May 6 17:35:34 2012 +0100
19075
19076 xfree86: add modesetting driver to fallback list on Linux
19077
19078 Add the modesetting driver to the fallback list on Linux, after vesa
19079 before fbdev.
19080
19081 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
19082 Signed-off-by: Dave Airlie <airlied@redhat.com>
19083
19084commit 5a3a98fcb799c2ac8fa7494645ad9030f1cac837
19085Author: Alan Coopersmith <alan.coopersmith@oracle.com>
19086Date: Fri Apr 27 13:35:32 2012 -0700
19087
19088 Undocument Font Module loading
19089
19090 Code was deleted in commit affec10635343668839994ea2bac16c1d4524200
19091
19092 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
19093 Reviewed-by: Dave Airlie <airlied@redhat.com>
19094 Reviewed-by: Adam Jackson <ajax@redhat.com>
19095
19096commit afcb7ba24ebd9968d48a192df438e7f1c0ecb770
19097Author: Alan Coopersmith <alan.coopersmith@oracle.com>
19098Date: Fri Apr 27 13:31:56 2012 -0700
19099
19100 Undocument mandatory loadable modules
19101
19102 The code to implement was deleted when BaseModules[] was emptied by
19103 the replacement of the "pcidata" module with libpciaccess calls
19104 in commit 46f55f5dead5d70cdff30531d80a72f6be042315.
19105
19106 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
19107 Reviewed-by: Dave Airlie <airlied@redhat.com>
19108 Reviewed-by: Adam Jackson <ajax@redhat.com>
19109
19110commit c3180a74a4fc6304bcfe676e3285d707874a9fea
19111Author: Alan Coopersmith <alan.coopersmith@oracle.com>
19112Date: Fri Apr 6 21:51:37 2012 -0700
19113
19114 cvt man page should use Hz, not kHz, for vertical refresh rate
19115
19116 https://bugs.freedesktop.org/show_bug.cgi?id=48311
19117
19118 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
19119 Reviewed-by: Julien Cristau <jcristau@debian.org>
19120
19121commit 96e0ab54962ab59f03275a098ee02c1f0f86b94e
19122Author: Alan Coopersmith <alan.coopersmith@oracle.com>
19123Date: Thu Jan 26 15:44:40 2012 -0800
19124
19125 Convert sbusPaletteKey to latest DevPrivate API
19126
19127 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
19128 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
19129 Reviewed-by: Dave Airlie <airlied@redhat.com>
19130
19131commit ba883a0f3435d5da82a8134e696c4905eea70f23
19132Merge: ae1c48e acdc4f5
19133Author: Keith Packard <keithp@keithp.com>
19134Date: Thu May 17 16:49:19 2012 -0700
19135
19136 Merge remote-tracking branch 'jeremyhu/master'
19137
19138commit acdc4f54ee5784d26c251ff8c941b4360ae80e9b
19139Author: Jeremy Huddleston <jeremyhu@apple.com>
19140Date: Thu May 17 10:08:02 2012 -0700
19141
19142 XQuartz: Avoid a race in initialization of darwinPointer
19143
19144 http://xquartz.macosforge.org/trac/ticket/579
19145
19146 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19147
19148commit c495211373a118acee02a29c6eb25a332abc856c
19149Author: Jeremy Huddleston <jeremyhu@apple.com>
19150Date: Thu May 17 16:20:30 2012 -0700
19151
19152 XQuartz: Declare noPseudoramiXExtension for miinitext.c
19153
19154 Regression-from: 6699f54fe0b3f4952a75e46fb6c3a643988e65dd
19155
19156 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19157
19158commit f5fc5334116ad2b0c8930ac68c52bf0f7a2f1417
19159Author: Jeremy Huddleston <jeremyhu@apple.com>
19160Date: Mon May 7 17:07:27 2012 -0700
19161
19162 XQuartz: Provide fls implementation for Tiger
19163
19164 Regression-from: 30623d6ff7bca223f9e1a825e86bd317eef8bf16
19165 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19166
19167commit e0097ce9719f9b0aeccf7102630c0299f8acdf93
19168Author: Jeremy Huddleston <jeremyhu@apple.com>
19169Date: Mon May 7 16:31:38 2012 -0700
19170
19171 XQuartz: Tiger build fix
19172
19173 http://trac.macports.org/ticket/34364
19174
19175 Regression-from: 662d41acdde1dcb9774fbe4054e251c708acaffe
19176
19177 Reported-by: Peter Dyballa <Peter_Dyballa@Freenet.DE>
19178 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19179
19180commit 4acbaa3027d89728ed932616f6ef66d93537b101
19181Author: Jeremy Huddleston <jeremyhu@apple.com>
19182Date: Mon May 7 16:16:20 2012 -0700
19183
19184 XQuartz: Workaround an SDK bug on Leopard/x86_64
19185
19186 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19187
19188commit ae1c48ebc8a3c0b32b3d732e8a340bfe00bb9f79
19189Author: Peter Hutterer <peter.hutterer@who-t.net>
19190Date: Wed May 16 10:54:48 2012 +1000
19191
19192 Xi: fix "discards ''const' qualifier" warnings
19193
19194 extinit.c: In function 'XInputExtensionInit':
19195 extinit.c:1301:29: warning: assignment discards 'const' qualifier from
19196 pointer target type [enabled by default]
19197 extinit.c:1303:36: warning: assignment discards 'const' qualifier from
19198 pointer target type [enabled by default]
19199
19200 property.c: In function 'XIChangeDeviceProperty':
19201 xiproperty.c:757:39: warning: cast discards '__attribute__((const))'
19202 qualifier from pointer target type [-Wcast-qual]
19203
19204 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19205 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19206
19207commit e42bf98283ddc4eac2213c96e8ebcd711583e6b1
19208Author: Peter Hutterer <peter.hutterer@who-t.net>
19209Date: Wed May 16 10:47:45 2012 +1000
19210
19211 Xi: fix "shadows previous local" warnings
19212
19213 exevents.c: In function 'DeepCopyFeedbackClasses':
19214 exevents.c:272:20: warning: declaration of 'classes' shadows a previous
19215 local [-Wshadow]
19216 exevents.c:245:16: warning: shadowed declaration is here [-Wshadow]
19217 (and a few more like this)
19218
19219 exevents.c: In function 'DeliverTouchEmulatedEvent':
19220 exevents.c:1442:27: warning: declaration of 'win' shadows a parameter
19221 [-Wshadow]
19222 exevents.c:1404:55: warning: shadowed declaration is here [-Wshadow]
19223 exevents.c:1475:28: warning: declaration of 'listener' shadows a parameter
19224 [-Wshadow]
19225 exevents.c:1403:62: warning: shadowed declaration is here [-Wshadow]
19226
19227 xiselectev.c: In function 'ProcXISelectEvents':
19228 xiselectev.c:178:34: warning: declaration of 'dummy' shadows a previous
19229 local [-Wshadow]
19230 xiselectev.c:91:18: warning: shadowed declaration is here [-Wshadow]
19231
19232 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19233 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19234
19235commit ca6d25ad2d1fb4efae9d40c7e74c8494f9918625
19236Author: Peter Hutterer <peter.hutterer@who-t.net>
19237Date: Wed May 16 10:10:11 2012 +1000
19238
19239 dix: Move DeviceFocusEvent from Xi into enterleave.c
19240
19241 This is only called from the enterleave implementation, so move it and its
19242 helper functions to there. No functional changes.
19243
19244 Fixes build error introduced in 31174565ec0090b4c03c9334c82878be2455f938 if
19245 building with '-Werror=implicit-function-declaration'
19246
19247 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19248 Reviewed-by: Dave Airlie <airlied@redhat.com>
19249 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
19250
19251commit 749a593e49adccdf1225be28a521412ec85333f4
19252Author: Peter Hutterer <peter.hutterer@who-t.net>
19253Date: Wed May 9 11:30:46 2012 +1000
19254
19255 dix: undo transformation for missing valuators (#49347)
19256
19257 last.valuators contains the transformed valuators of the device. If the
19258 device submits events with x/y missing, we need to get that from
19259 last.valuators and undo the transformation to that axis.
19260
19261 X.Org Bug 49347 <http://bugs.freedesktop.org/show_bug.cgi?id=49347>
19262
19263 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19264 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19265
19266commit 03318835a5a87a8a5cb3515130b5380d565c0d91
19267Author: Peter Hutterer <peter.hutterer@who-t.net>
19268Date: Mon May 14 15:06:56 2012 +1000
19269
19270 include: don't redeclare device events in xkbsrv.h
19271
19272 ../../include/xkbsrv.h:308:51: warning: redundant redeclaration of
19273 ‘DeviceKeyPress’ [-Wredundant-decls]
19274
19275 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19276 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19277
19278commit 43754803c13c232fd5c37eb3d4b1ca2898d6a6cd
19279Author: Peter Hutterer <peter.hutterer@who-t.net>
19280Date: Mon May 14 15:07:57 2012 +1000
19281
19282 test: don't shadow parameter 'len'
19283
19284 protocol-xiquerydevice.c:226:25: warning: declaration of ‘len’ shadows a
19285 parameter [-Wshadow]
19286
19287 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19288 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19289
19290commit 1bafc9f26bef0997c98b97efa888ed5e2a54f6f0
19291Author: Peter Hutterer <peter.hutterer@who-t.net>
19292Date: Mon May 14 15:01:09 2012 +1000
19293
19294 test: fix redundant declaration of "BadDevice"
19295
19296 In file included from protocol-common.c:36:0:
19297 protocol-common.h:36:12: warning: redundant redeclaration of ‘BadDevice’
19298 [-Wredundant-decls]
19299 In file included from protocol-common.c:30:0:
19300 ../../Xi/exglobals.h:41:12: note: previous declaration of ‘BadDevice’ was
19301 here
19302
19303 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19304 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19305
19306commit 47e1d0ea43eb7a6caf774f375f11849d0d515e14
19307Author: Peter Hutterer <peter.hutterer@who-t.net>
19308Date: Mon May 14 14:56:02 2012 +1000
19309
19310 test: fix "userdata shadows global declaration" warnings
19311
19312 protocol-xiquerypointer.c:124:72: warning: declaration of
19313 ‘userdata’ shadows a global declaration [-Wshadow]
19314
19315 and similar
19316
19317 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19318 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19319
19320commit e6da635826bdda7f67bd492dda8dba8a97809213
19321Author: Peter Hutterer <peter.hutterer@who-t.net>
19322Date: Mon May 14 14:51:58 2012 +1000
19323
19324 test: fix "redundant declaration of devices" warning
19325
19326 In file included from protocol-xiwarppointer.c:41:0:
19327 protocol-common.h:91:23: warning: redundant redeclaration of ‘devices’
19328 [-Wredundant-decls]
19329 protocol-common.h:86:3: note: previous declaration of ‘devices’ was here
19330
19331 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19332 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19333
19334commit 693a31e704c0745b209e69711522417c4fa8541e
19335Author: Simon Thum <simon.thum@gmx.de>
19336Date: Mon Apr 2 18:49:53 2012 +0200
19337
19338 dix: indentation fixes for pointer acceleration
19339
19340 Signed-off-by: Simon Thum <simon.thum@gmx.de>
19341 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19342 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19343
19344commit 31174565ec0090b4c03c9334c82878be2455f938
19345Author: Michal Suchanek <hramrach@gmail.com>
19346Date: Wed Apr 25 15:03:31 2012 +0200
19347
19348 dix: Remove redundant declarations.
19349
19350 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19351 Reviewed-by: Dave Airlie <airlied@redhat.com>
19352 Signed-off-by: Dave Airlie <airlied@redhat.com>
19353
19354commit f1cec791d1def506b99b0e6c160e9e8be107bd13
19355Author: Michal Suchanek <hramrach@gmail.com>
19356Date: Tue Apr 24 20:08:02 2012 +0200
19357
19358 xfree86: Add ifdef wrapper to xvmodproc.h
19359
19360 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19361 Reviewed-by: Dave Airlie <airlied@redhat.com>
19362 Signed-off-by: Dave Airlie <airlied@redhat.com>
19363
19364commit d2f71b13946a14d9698d0355cc45b114b236e73b
19365Author: Michal Suchanek <hramrach@gmail.com>
19366Date: Tue Apr 24 18:44:48 2012 +0200
19367
19368 Xi: Remove redundant declaration.
19369
19370 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19371 Reviewed-by: Dave Airlie <airlied@redhat.com>
19372 Signed-off-by: Dave Airlie <airlied@redhat.com>
19373
19374commit a4f87f885a55373a6daf56fd540e10c5f1db5a5c
19375Author: Michal Suchanek <hramrach@gmail.com>
19376Date: Tue Apr 24 18:43:31 2012 +0200
19377
19378 kdrive: Remove redundant declarations.
19379
19380 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19381 Reviewed-by: Dave Airlie <airlied@redhat.com>
19382 Signed-off-by: Dave Airlie <airlied@redhat.com>
19383
19384commit 369edd78769fa1a174e4091d0fa0722350df5d68
19385Author: Michal Suchanek <hramrach@gmail.com>
19386Date: Tue Apr 24 18:41:07 2012 +0200
19387
19388 dix: don't duplicate DoFocusEvents declaration.
19389
19390 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19391 Reviewed-by: Dave Airlie <airlied@redhat.com>
19392 Signed-off-by: Dave Airlie <airlied@redhat.com>
19393
19394commit 36377fb0e88a6d237d98805e802ab3793582c976
19395Author: Michal Suchanek <hramrach@gmail.com>
19396Date: Tue Apr 24 18:37:30 2012 +0200
19397
19398 dix: Remove redundant declarations.
19399
19400 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19401 Reviewed-by: Dave Airlie <airlied@redhat.com>
19402 Signed-off-by: Dave Airlie <airlied@redhat.com>
19403
19404commit ea9ed83f8f0da241099d0a11c3460d46c5114dea
19405Author: Michal Suchanek <hramrach@gmail.com>
19406Date: Tue Apr 24 18:32:53 2012 +0200
19407
19408 xfree86: don't include xvmodproc.h when not needed
19409
19410 The functions are already declared in xf86xv.h
19411
19412 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19413 Reviewed-by: Dave Airlie <airlied@redhat.com>
19414 Signed-off-by: Dave Airlie <airlied@redhat.com>
19415
19416commit 1927e5ab25c9c9bf481e75ac50f4b140a96e634a
19417Author: Michal Suchanek <hramrach@gmail.com>
19418Date: Tue Apr 24 18:24:19 2012 +0200
19419
19420 xserver: Remove duplicate declaration in resource.h
19421
19422 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19423 Reviewed-by: Dave Airlie <airlied@redhat.com>
19424 Signed-off-by: Dave Airlie <airlied@redhat.com>
19425
19426commit 1339681e4fbd4f5981e6246eedb23e8a9c5d10f4
19427Author: Michal Suchanek <hramrach@gmail.com>
19428Date: Tue Apr 24 18:19:43 2012 +0200
19429
19430 miext: Remove redundant declaration.
19431
19432 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19433 Reviewed-by: Dave Airlie <airlied@redhat.com>
19434 Signed-off-by: Dave Airlie <airlied@redhat.com>
19435
19436commit 6699f54fe0b3f4952a75e46fb6c3a643988e65dd
19437Author: Michal Suchanek <hramrach@gmail.com>
19438Date: Tue Apr 24 18:18:43 2012 +0200
19439
19440 mi: Remove redundant declarations.
19441
19442 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19443 Reviewed-by: Dave Airlie <airlied@redhat.com>
19444 Signed-off-by: Dave Airlie <airlied@redhat.com>
19445
19446commit b4bf90f9fb20e0e56d2451096d4f62ecfc24a962
19447Author: Michal Suchanek <hramrach@gmail.com>
19448Date: Tue Apr 24 18:16:30 2012 +0200
19449
19450 exa: Remove redundant declaration.
19451
19452 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19453 Reviewed-by: Dave Airlie <airlied@redhat.com>
19454 Signed-off-by: Dave Airlie <airlied@redhat.com>
19455
19456commit 5595e7ce0735db6f82e1fe9df45434826019bf9e
19457Author: Michal Suchanek <hramrach@gmail.com>
19458Date: Tue Apr 24 18:15:17 2012 +0200
19459
19460 randr: Remove redundant declaration.
19461
19462 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19463 Reviewed-by: Dave Airlie <airlied@redhat.com>
19464 Signed-off-by: Dave Airlie <airlied@redhat.com>
19465
19466commit c59c9dac84301f8eb2a0e4a09a8839146aa5a069
19467Author: Michal Suchanek <hramrach@gmail.com>
19468Date: Tue Apr 24 18:13:22 2012 +0200
19469
19470 xkb: Remove redundant declarations.
19471
19472 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19473 Reviewed-by: Dave Airlie <airlied@redhat.com>
19474 Signed-off-by: Dave Airlie <airlied@redhat.com>
19475
19476commit d0a83460451df1a4c5667b2f414c35ee81b4ffa4
19477Author: Michal Suchanek <hramrach@gmail.com>
19478Date: Tue Apr 24 18:11:16 2012 +0200
19479
19480 xfixes: remove redundant declaration.
19481
19482 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19483 Reviewed-by: Dave Airlie <airlied@redhat.com>
19484 Signed-off-by: Dave Airlie <airlied@redhat.com>
19485
19486commit 3a94b338aaa48b656aae1bb3a5a9d64300c0a093
19487Merge: afc153a c91d00e
19488Author: Keith Packard <keithp@keithp.com>
19489Date: Wed May 9 21:17:16 2012 -0700
19490
19491 Merge remote-tracking branch 'whot/for-keith'
19492
19493commit afc153a5b4fc58ae70dc214f61a71b1a8c855f06
19494Author: James Cloos <cloos@jhcloos.com>
19495Date: Tue May 8 17:55:10 2012 -0400
19496
19497 Fix RANDR’s gamma_to_ramp().
19498
19499 In order to generate a 256-entry ramp in [0,65535] which covers the full
19500 range, one must mupliply eight-bit values not by 256 but rather by 257.
19501
19502 Many years back – well before the RANDR extension was written, and
19503 before xorg@fdo – a similar bug fix was made to the DIX for converting
19504 client-supplied eight-bit color values into sixteen-bit values.
19505
19506 Noticed by: Elle Stone and Graeme Gill.
19507
19508 Signed-off-by: James Cloos <cloos@jhcloos.com>
19509
19510commit e501c34d4937d5e6f19abd29f1ec7f95faa3bb55
19511Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19512Date: Tue May 1 22:03:24 2012 -0500
19513
19514 hw/xwin/glx: Use silent rules
19515
19516 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19517 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19518
19519commit 026d402fef623a0bda0bedd9c0938ba787bf9829
19520Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19521Date: Fri Mar 30 14:33:18 2012 -0500
19522
19523 xfree86: use silent rules with sdksyms generation
19524
19525 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19526 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19527
19528commit 3d98dac46c17bc6e03546b8a532bfb55f00e5084
19529Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19530Date: Sun Apr 29 20:45:17 2012 -0500
19531
19532 xfree86: respect EXEEXT in relink target
19533
19534 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19535 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19536
19537commit e6461db6dcb437cfa7d4b23008c2c12f8169ff79
19538Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19539Date: Thu Apr 26 01:49:06 2012 -0500
19540
19541 os: Add CryptoAPI as a choice of SHA1 implementation
19542
19543 Both Cygwin and MinGW can use Windows' native CryptoAPI for SHA1,
19544 saving a dependency on libgcrypt or OpenSSL. The necessary functions
19545 are in ADVAPI32.DLL, which is among the default lib flags and is
19546 already used in hw/xwin for accessing the registry.
19547
19548 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19549 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
19550 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19551 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
19552 Tested-by: Colin Harrison <colin.harrison@virgin.net>
19553
19554commit c91d00e0f330b9de604068e1bfcb0a307096434f
19555Author: Daniel Kurtz <djkurtz@chromium.org>
19556Date: Wed Apr 18 17:51:53 2012 +0800
19557
19558 os/log: refactor logging
19559
19560 It is not safe to ever use an arbitrary (possibly user supplied) string as
19561 part of the format for a *sprintf() call.
19562
19563 For example:
19564 1. Name a Bluetooth keyboard "%n%n%n%n%n%n%n%n"
19565 2. Pair it with a computer running X and try to use it
19566 3. X is not happy when trying to do the following in xf86-input-evdev:
19567 xf86IDrvMsg(pInfo, X_CONFIG, "Device: \"%s\"\n", device);
19568 because LogVHdrMessageVerb() has put the %n from the device name
19569 into a format string of the form:
19570 "evdev: %n%n%n%n%n%n%n%n: Device: \"%s\"\n"
19571
19572 Instead, build up a log message in place by appending successive formatted
19573 strings by sncprintf'ing to the end of the previous.
19574
19575 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
19576 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19577 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19578
19579commit 6ce0eac4f8a05f6d7401445cab95027709d3a479
19580Author: Daniel Kurtz <djkurtz@chromium.org>
19581Date: Wed Apr 18 17:51:52 2012 +0800
19582
19583 os/log: only write timestamp if a message is actually written to logfile
19584
19585 The current code will write a timestamps into the logFile whenever
19586 the last message ended with a '\n' - even if the verb for that timestamp
19587 is at too high a level. This timestamp will sit there with no matching
19588 message until the next call to LogVWrite with a valid verb.
19589
19590 In other words, in some cases, timestamps in the X.org.log are for some
19591 completely unrelated message that was previously ignored due to
19592 insufficient verbosity, and not for the message that appears next to it
19593 in the log file.
19594
19595 We keep the current policy which appears to be to only apply timestamps if
19596 a message is actually written to a log file. That is, no timestamps on
19597 stderr, or in the mem buffer. Therefore, the timestamp stringification
19598 is moved to the conditional where it is used.
19599
19600 Since logging uses a fixed length buffer, this patch also forces a '\n'
19601 whenever a buffer is terminated due to a too-long write request. This
19602 allows the newline detection to work even on overflow, and also cleans up
19603 the log a bit in the overflow case.
19604
19605 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
19606 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19607 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19608
19609commit 5c2e2a164d615ab06be28a663734e782614b5cc7
19610Author: Daniel Kurtz <djkurtz@chromium.org>
19611Date: Wed Apr 18 09:51:51 2012 +0000
19612
19613 os/xprintf: add Xvscnprintf and Xscnprintf
19614
19615 Normal snprintf() usually returns the number of bytes that would have been
19616 written into a buffer had the buffer been long enough.
19617
19618 The scnprintf() variants return the actual number of bytes written,
19619 excluding the trailing '\0'.
19620
19621 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
19622 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19623 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19624
19625commit c30862879d2c766519780bb7f353f35edf0daa9b
19626Author: Daniel Kurtz <djkurtz@chromium.org>
19627Date: Wed Apr 18 17:51:50 2012 +0800
19628
19629 os/log: trivial cleanups
19630
19631 * space->tab
19632 * remove comment that doesn't make any sense
19633
19634 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
19635 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19636 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19637
19638commit 97041364a6acb2b66b5cfd06757c90a006ad50e9
19639Merge: 1908272 ee542b8
19640Author: Keith Packard <keithp@keithp.com>
19641Date: Wed May 2 20:47:25 2012 -0700
19642
19643 Merge remote-tracking branch 'whot/for-keith'
19644
19645 No conflicts here
19646
19647commit ee542b85590814ee25369babce1ad14feeb137af
19648Author: Chase Douglas <chase.douglas@canonical.com>
19649Date: Tue May 1 10:21:12 2012 -0700
19650
19651 Report touch emulated buttons in XIQueryPointer for XI 2.1 and earlier
19652
19653 XInput 2.1 and earlier clients do not know about touches. We must report
19654 touch emulated button presses for these clients. For later clients, we
19655 only report true pointer button presses.
19656
19657 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
19658 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19659 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19660
19661commit 19082726cb2c69f53e9720904521e3c98f788bff
19662Author: Ryan Pavlik <rpavlik@iastate.edu>
19663Date: Fri Oct 21 12:47:38 2011 -0500
19664
19665 configure.ac: on MinGW, link with ws2_32 instead of winsock2
19666
19667 ws2_32 is the correct name for the libary (even on 64 bit Windows :-))
19668
19669 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
19670 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19671 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19672
19673commit ad9605a3e21b20fa8565dc6701610712ac6dd9e1
19674Author: Ryan Pavlik <rpavlik@iastate.edu>
19675Date: Fri Nov 4 13:26:14 2011 -0500
19676
19677 configure.ac: MinGW doesn't have setuid binaries either.
19678
19679 If the target platform isn't in a list of platforms we know don't use a
19680 setuid binary, we try to test if we can chown something to root.
19681
19682 This test possibly won't give the right answer if we are cross-compiling,
19683 which is common for the MinGW target. This patch adds MinGW to the list
19684 of platforms we know don't use a setuid binary.
19685
19686 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
19687 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19688 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19689
19690commit 2225208c2ef9857f14813376ce85d305f19263dd
19691Author: Ryan Pavlik <rpavlik@iastate.edu>
19692Date: Fri Nov 4 13:30:43 2011 -0500
19693
19694 configure.ac: auto-disable MITSHM if we lack IPC
19695
19696 The MITSHM extension uses SYSV IPC, but even if configure's test
19697 for IPC failed, MITSHM was still enabled by default, breaking
19698 MinGW builds by default.
19699
19700 Unfortunately, fixing this exposes the fact that the HAVE_SYSV_IPC
19701 test wasn't being used for anything before and so we hadn't noticed it
19702 was failing on Cygwin.
19703
19704 Change from using SHM_W|SHM_R flags (which aren't required by POSIX) to
19705 S_IRUSR|S_IWUSR flags (which are)
19706
19707 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
19708 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19709 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19710
19711commit 1e7b500a8e1d79b91a4e857a2da06194efe8cf69
19712Author: Chase Douglas <chase.douglas@canonical.com>
19713Date: Tue May 1 10:21:11 2012 -0700
19714
19715 Report logical button state in ProcXIQueryPointer
19716
19717 Physical button state is usually meaningless to an X client.
19718
19719 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
19720 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19721 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19722
19723commit 0426e6d65b6598edfbcb2fc66ee65fb08256469e
19724Author: Jon TURNEY <jon.turney@dronecode.org.uk>
19725Date: Tue Jan 3 00:14:32 2012 +0000
19726
19727 configure.ac: Make default configuration for MinGW disable unsupported extensions and DDXs
19728
19729 Same as the default configuration for Cygwin, --disable-xorg and --disable-dmx DDX by default,
19730 and force --disable-xv and other unsupported extensions
19731
19732 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
19733 Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
19734
19735commit f3410b97cf9b48a47bee3d15d232f8a88e75f4ef
19736Author: Peter Hutterer <peter.hutterer@who-t.net>
19737Date: Mon Apr 30 10:01:48 2012 +1000
19738
19739 dix: when disabling a device, release all buttons and keys
19740
19741 A suspend-induced device disable may happen before the device gets to see
19742 the button release event. On resume, the server's internal state still has
19743 some buttons pressed, causing inconsistent behaviour.
19744
19745 Force the release and the matching events to be sent to the client.
19746
19747 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19748 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19749
19750commit af88b43f9e604157b74270d609c08bdfa256a792
19751Author: Peter Hutterer <peter.hutterer@who-t.net>
19752Date: Fri Apr 27 16:31:17 2012 +1000
19753
19754 dix: don't emulate scroll events for non-existing axes (#47281)
19755
19756 Test case:
19757 - create a device with REL_HWHEEL and ABS_X and ABS_Y. evdev 2.7.0 will set
19758 that up as device with 1 relative axis
19759 - move pointer to VGA1
19760 - xrandr --output VGA1 --off
19761
19762 Warps the pointer to the new spot and calls GPE with the x/y mask bits set.
19763 When running through the loop to check for scroll event, this overruns the
19764 axes and may try to emulate scroll events based on random garbage in the
19765 memory. If that memory contained non-zero for the scroll type but near-zero
19766 for the increment field, the server would hang in an infinite loop.
19767
19768 This was the trigger for this suggested, never-merged, patch here:
19769 http://patchwork.freedesktop.org/patch/9543/
19770
19771 X.Org Bug 47281 <http://bugs.freedesktop.org/show_bug.cgi?id=47281>
19772
19773 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19774 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19775
19776commit 08962951de969b9d8c870af8b6e47303dc0decfd
19777Author: Peter Hutterer <peter.hutterer@who-t.net>
19778Date: Fri Apr 27 10:52:39 2012 +1000
19779
19780 os: make timers signal-safe
19781
19782 If TimerSet() is called from a signal handler (synaptics tap handling code)
19783 may result in list corruption if we're currently inside TimerSet().
19784
19785 See backtrace in
19786 https://bugzilla.redhat.com/show_bug.cgi?id=814869
19787
19788 Block signals for all list manipulations in the timers.
19789
19790 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19791 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
19792
19793commit d662fa2450856777b59c4b62b912395a8bfd52fd
19794Author: Michal Suchanek <hramrach@gmail.com>
19795Date: Thu Apr 26 15:11:20 2012 +0200
19796
19797 dmx: Annotate dmxlog.c with _X_ATTRIBUTE_PRINTF and _X_NORETURN
19798
19799 and fix resulting printf warning in dmxLogVisual
19800
19801 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19802 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19803 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19804
19805commit 5c361d59c5031d9b3f7f9093a52d2b1ff4d9ae5f
19806Author: Chase Douglas <chase.douglas@canonical.com>
19807Date: Fri Apr 20 11:08:15 2012 -0700
19808
19809 TouchListenerAcceptReject: Warn and return early on bad listener index
19810
19811 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
19812 Reviewed-by: Bryce Harrington <bryce@canonical.com>
19813 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19814
19815commit 88bacc49f06da5927f716869f5a32672a8297ed0
19816Author: Chase Douglas <chase.douglas@canonical.com>
19817Date: Wed Apr 4 15:29:42 2012 -0700
19818
19819 os: Add -displayfd option
19820
19821 This option specifies a file descriptor in the launching process. X
19822 will scan for an available display number and write that number back to
19823 the launching process, at the same time as SIGUSR1 generation. This
19824 means display managers don't need to guess at available display numbers.
19825 As a consequence, if X fails to start when using -displayfd, it's not
19826 because the display was in use, so there's no point in retrying the X
19827 launch on a higher display number.
19828
19829 Signed-off-by: Adam Jackson <ajax@redhat.com>
19830 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
19831 Reviewed-by: Julien Cristau <jcristau@debian.org>
19832 Tested-by: Julien Cristau <jcristau@debian.org>
19833 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
19834 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19835
19836commit 1d82ec95942b88dd01f0ac6b883368360a0b5fe6
19837Author: Michal Suchanek <hramrach@gmail.com>
19838Date: Mon Apr 23 13:52:40 2012 +0200
19839
19840 xserver: Fix out-of-tree build
19841
19842 Fixes regression caused by ccb3e78124fb05defd0c9b438746b79d84dfc3ae
19843
19844 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19845 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
19846 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19847
19848commit ea51e9b2877df60135edaf2a8f88d0f2a2b41060
19849Author: Peter Hutterer <peter.hutterer@who-t.net>
19850Date: Mon Apr 23 10:35:53 2012 +1000
19851
19852 Xi: return BadValue on XIQueryVersion if the version is less than first call
19853
19854 Clients that use plugin systems may require multiple calls to
19855 XIQueryVersion from different plugins. The current error handling requires
19856 client-side synchronisation of version numbers.
19857
19858 The first call to XIQueryVersion defines the server behaviour. Once cached,
19859 always return that version number to any clients. Unless a client requests a
19860 version lower than the first defined one, then a BadValue must be returned
19861 to be protocol-compatible.
19862
19863 Introduced in 2c23ef83b0e03e163aeeb06133538606886f4e9c
19864
19865 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
19866 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
19867
19868commit 93d6ba5b711cbd3f502d83e54c9739856d2e6f2a
19869Author: Jeremy Huddleston <jeremyhu@apple.com>
19870Date: Thu Apr 26 21:17:54 2012 -0700
19871
19872 XQuartz: Make sure we tell startx the correct server binary to use in our fallback path
19873
19874 With multiple servers installed, we can't be certain if X is Xorg or Xquartz
19875
19876 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19877
19878commit fa6dddc6ce51c1b7a43fb379fb2a19550f6c8683
19879Author: Michal Suchanek <hramrach@gmail.com>
19880Date: Wed Mar 28 18:12:39 2012 -0700
19881
19882 xfree86: workaround crash on close
19883
19884 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=41653
19885
19886 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
19887 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
19888 Tested-by: Knut Petersen <Knut_Petersen@t-online.de>
19889
19890commit adcb48a29dcb3f32ff570551de4b014775c1e147
19891Author: Jonas Maebe <jonas.maebe@elis.ugent.be>
19892Date: Sun Apr 22 20:52:47 2012 -0700
19893
19894 glapi: Correct size of allocated _glapi_table struct
19895
19896 The __glapi_gentable_set_remaining_noop() routine treats the _glapi_struct
19897 as an array of _glapi_get_dispatch_table_size() pointers, so we have to
19898 allocate _glapi_get_dispatch_table_size()*sizeof(void*) bytes rather than
19899 sizeof(struct _glapi_struct) bytes.
19900
19901 Signed-off-by: Jonas Maebe <jonas.maebe@elis.ugent.be>
19902 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
19903
19904commit e34519e525559b01a63d26639f13f0487468de28
19905Author: Jeremy Huddleston <jeremyhu@apple.com>
19906Date: Thu Apr 19 16:38:06 2012 -0700
19907
19908 XQuartz: darwinPointer now sends both absolute and relative motion
19909
19910 This should hopefully help out wine clients that were continuing to
19911 have issues after the earlier changes.
19912
19913 http://xquartz.macosforge.org/trac/ticket/548
19914
19915 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19916
19917commit 662d41acdde1dcb9774fbe4054e251c708acaffe
19918Author: Jeremy Huddleston <jeremyhu@apple.com>
19919Date: Thu Apr 19 18:48:22 2012 -0700
19920
19921 XQuartz: Add a hack to better handle clicky wheel scroll mice
19922
19923 We loose information from AppKit being in our way. Before adopting
19924 smooth scrolling, we always rounded-up the number of scroll button
19925 clicks per NSEvent. Now, the scroll value is accumulated in the
19926 dix, and clicky scroll wheels with legacy X11 clients are seeing
19927 an accumulation of error due to so many translations (button press
19928 to smooth scrolling value in AppKit, passed to the dix, and then
19929 synthesized into a button press). This attempts to make the
19930 situation better.
19931
19932 http://xquartz.macosforge.org/trac/ticket/562
19933
19934 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19935
19936commit eda85290ae47d8dd497ef9335d10268bbe3e02a0
19937Author: Jeremy Huddleston <jeremyhu@apple.com>
19938Date: Thu Apr 19 16:20:30 2012 -0700
19939
19940 XQuartz: Use screenInfo.{width,height} instead of grabbing it from the first screen
19941
19942 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19943
19944commit b99586c9086ac29df144ef75df92f8c0c7554651
19945Author: Jeremy Huddleston <jeremyhu@apple.com>
19946Date: Wed Apr 18 17:50:55 2012 -0700
19947
19948 XQuartz: Separate out tablet and mouse event delivery into separate functions
19949
19950 This should have no immediate impact aside from fake mouse buttons no longer
19951 working with tablets (where they aren't needed or desired anyways). This
19952 prepares us for future changes.
19953
19954 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19955
19956commit 30623d6ff7bca223f9e1a825e86bd317eef8bf16
19957Author: Jeremy Huddleston <jeremyhu@apple.com>
19958Date: Thu Apr 19 16:06:45 2012 -0700
19959
19960 XQuartz: Correct calculation of the size of our file descriptor array in console_redirect
19961
19962 Reported-by: Joe Rohde <joer@valvesoftware.com>
19963 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19964
19965commit b4c4c65a35640e7274ccd4228f62110147878b72
19966Author: Jeremy Huddleston <jeremyhu@apple.com>
19967Date: Wed Apr 18 01:01:44 2012 -0700
19968
19969 XQuartz: Fix a deadlock in pre-dispatch code
19970
19971 The fact that this has been in place so long makes me really wonder if
19972 anybody cares about this running in Tiger or Leopard.
19973
19974 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19975
19976commit d1ef0d4964e43a4ab777750ad214d1092fd1ffda
19977Author: Jeremy Huddleston <jeremyhu@apple.com>
19978Date: Sat Apr 7 15:26:53 2012 -0700
19979
19980 test: Fix make dist
19981
19982 It seems like make dist should be doing te right thing without this commit,
19983 but it's not in some cases. Don't ask me to explain why.
19984
19985 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19986
19987commit d97868d934fd1a00f9e9ef23a1154a70c7bc0464
19988Author: Jeremy Huddleston <jeremyhu@apple.com>
19989Date: Thu Apr 19 16:39:23 2012 -0700
19990
19991 os: Annotate OsVendorFatalError as _X_ATTRIBUTE_PRINTF
19992
19993 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
19994 Tested-By: Michal Suchanek <hramrach@gmail.com>
19995
19996commit 0ba1794fe2896395875f62e911720c7585687444
19997Author: Jeremy Huddleston <jeremyhu@apple.com>
19998Date: Thu Apr 19 16:51:46 2012 -0700
19999
20000 xres: Fix build without composite
20001
20002 Regression from: b8d0d19a6d410776b53a41e7cae90f68d4b22bb7
20003 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20004 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
20005 Tested-By: Michal Suchanek <hramrach@gmail.com>
20006
20007commit d77eb7ee49ef19c2c4c7381d56e9d0f9c3fbc890
20008Merge: 31e3c0f c7b1625
20009Author: Keith Packard <keithp@keithp.com>
20010Date: Thu Apr 19 15:48:34 2012 -0500
20011
20012 Merge remote-tracking branch 'yselkowitz/master'
20013
20014 Pull in Cygwin for XFree86
20015
20016commit 31e3c0ff7f5fb36421be8de06e5f6e675ec26ddc
20017Author: Keith Packard <keithp@keithp.com>
20018Date: Thu Apr 19 15:25:41 2012 -0500
20019
20020 Export CompositeClientWindowType
20021
20022 Make sure CompositeClientWindowType is visible for XResource v1.2
20023
20024 Signed-off-by: Keith Packard <keithp@keithp.com>
20025
20026commit e6308e32fe2b5f74133d4d238ffa512257f6327c
20027Merge: 3720aa3 51a8d8d
20028Author: Keith Packard <keithp@keithp.com>
20029Date: Thu Apr 19 10:45:07 2012 -0500
20030
20031 Merge remote-tracking branch 'whot/for-keith'
20032
20033 Touch input changes from Chase
20034
20035commit 51a8d8dd19d7496fe84b37a1f0a7a03658120539
20036Merge: ebf2148 00cf1c4
20037Author: Peter Hutterer <peter.hutterer@who-t.net>
20038Date: Thu Apr 19 17:03:54 2012 +1000
20039
20040 Merge branch 'input-fixes' of git://people.freedesktop.org/~cndougla/xserver into for-keith
20041
20042commit 00cf1c40b28417d7035c2917d048553eb720023c
20043Author: Chase Douglas <chase.douglas@canonical.com>
20044Date: Wed Apr 18 18:21:54 2012 -0700
20045
20046 Replay original touch begin event instead of generated begin event
20047
20048 The generated event does not have axes other than X and Y and has a
20049 newer timestamp. In particular, the newer timestamp may be newer than
20050 the real touch end event, which may be stuck in the syncEvents queue. If
20051 a client uses the timestamps for grabbing bad things may happen.
20052
20053 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20054 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20055
20056commit 312910b4e34215aaa50fc0c6092684d5878dc32f
20057Author: Chase Douglas <chase.douglas@canonical.com>
20058Date: Wed Apr 18 11:15:40 2012 -0700
20059
20060 Update currentTime in dispatch loop
20061
20062 A request, like input device grabs, may check a request timestamp
20063 against currentTime. It is possible for currentTime to lag a previously
20064 sent event timestamp. If the client makes a request based on such an
20065 event timestamp, the request may fail the validity check against
20066 currentTime unless we always update the time before processing the
20067 request.
20068
20069 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20070 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20071
20072commit a986f2f30cbe2a00e72ded7315c4951d7703e549
20073Author: Chase Douglas <chase.douglas@canonical.com>
20074Date: Tue Apr 17 11:40:15 2012 -0700
20075
20076 Update device state including when touch record does not exist
20077
20078 If a touch is physically active, the pointer core state should reflect
20079 that the first button is pressed. Currently, this only occurs when there
20080 are active listeners of the touch sequence. By moving the device state
20081 updating to the beginning of touch processing we ensure it is updated
20082 according to the processed physical state no matter what.
20083
20084 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20085 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20086
20087commit ec9c4295830c3de610e65aca17f4da4a7af3c4c5
20088Author: Chase Douglas <chase.douglas@canonical.com>
20089Date: Wed Apr 18 12:04:58 2012 -0700
20090
20091 Check other clients' core masks properly when adding touch listener
20092
20093 The current code checks the core event mask as though it were an XI
20094 mask. This change fixes the checks so the proper client and event masks
20095 are used.
20096
20097 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20098 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20099
20100commit e175971a6f44d94aa8306dc6b9228ccb2c8a0b4d
20101Author: Chase Douglas <chase.douglas@canonical.com>
20102Date: Tue Apr 17 11:39:10 2012 -0700
20103
20104 Ensure touch is ended when last listener is rejected
20105
20106 Currently, the touch is only logically ended if the touch has physically
20107 ended. If the touch hasn't physically ended, the touch record is never
20108 ended. If there aren't any more listeners, we don't need to keep the dix
20109 touch record around any more.
20110
20111 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20112 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20113
20114commit d0449851d1233543c3133d77e0ab7233319cdf5f
20115Author: Chase Douglas <chase.douglas@canonical.com>
20116Date: Tue Apr 17 10:49:14 2012 -0700
20117
20118 Create a new dix touch record for an emulated touch with no listeners
20119
20120 As a special case, if a still physically active pointer emulated touch
20121 has no listeners and the device is explicitly grabbed for pointer
20122 events, create a new dix touch record for the grab only.
20123
20124 This allows for clients to "hand off" grabs. For example, when dragging
20125 a window under compiz the window decorator sees the button press and
20126 then ungrabs the implicit grab. It then tells compiz to grab the device,
20127 and compiz then moves the window with the pointer motion. This is racy,
20128 but is allowed by the input protocol for pointer events when there are
20129 no other clients with a grab on the device.
20130
20131 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20132 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20133
20134commit 3d06bfe93d33cfe6150d8fb0058ee7bc8d80622b
20135Author: Chase Douglas <chase.douglas@canonical.com>
20136Date: Tue Apr 17 13:56:08 2012 -0700
20137
20138 Rename TouchEnsureSprite to TouchBuildSprite and event type checks
20139
20140 The function will be used for building a sprite for pointer emulation
20141 after an explicit device grab. This commit refactors the code so that
20142 TouchBuildSprite will function with any event type and moves the checks
20143 to the caller.
20144
20145 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20146 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20147
20148commit 2efbed23c29020f9994ab7c3155ce7386950dc7a
20149Author: Chase Douglas <chase.douglas@canonical.com>
20150Date: Tue Apr 17 08:39:38 2012 -0700
20151
20152 When activating an explicit grab, update owning listener
20153
20154 Pointer passive grabs may be changed by the grabbing client. This allows
20155 for a selecting client to change an implicit grab to an active grab,
20156 which is the mechanism used for pop-up windows like application menus.
20157
20158 We need to do the same thing with touches. If the grabbing client is the
20159 owner of a touch sequence, change the listener record to reflect the new
20160 grab. If the grabbing client is not the owner, nothing changes for the
20161 touch.
20162
20163 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20164 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20165
20166commit ef64b5ee97099618cf2e2cbbd3e471095695ae24
20167Author: Chase Douglas <chase.douglas@canonical.com>
20168Date: Mon Apr 16 15:53:51 2012 -0700
20169
20170 Don't deactivate implicit pointer grab on fake touch end event
20171
20172 Fake touch end events are generated by touch acceptance and rejection.
20173 These should not cause implicit pointer grabs to be deactivated.
20174
20175 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20176 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20177
20178commit fc518cd9f59060cc19bb90361767c0f47f0e25eb
20179Author: Chase Douglas <chase.douglas@canonical.com>
20180Date: Mon Apr 16 15:31:47 2012 -0700
20181
20182 End a pointer emulated touch event only on a "real" end event
20183
20184 Fake end events are generated by touch acceptance or rejection. These
20185 should not end the touch point.
20186
20187 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20188 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20189
20190commit 80d7d1ec6a9d61aa96e7d019dc1bee29d90cea34
20191Author: Chase Douglas <chase.douglas@canonical.com>
20192Date: Mon Apr 16 14:19:59 2012 -0700
20193
20194 On touch accept, only process end event for owner if it has seen the end
20195
20196 We still need to generate the touch ownership event to process the
20197 ending of the touch event in the case where the owner has the end
20198 already.
20199
20200 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20201 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20202
20203commit 8dfd98245d2c44a1eb4c8b7c275e6cfc10fe40f1
20204Author: Chase Douglas <chase.douglas@canonical.com>
20205Date: Wed Apr 11 11:14:51 2012 -0700
20206
20207 Fix copy/paste error from before git history in UpdateCurrentTimeIf()
20208
20209 See UpdateCurrentTime() for reference. I don't know what bug this might
20210 trigger, but it wouldn't hurt to fix this.
20211
20212 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20213 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20214
20215commit 6ca30cb33e829b4edd01822367e44ffe6f0951b0
20216Author: Chase Douglas <chase.douglas@canonical.com>
20217Date: Wed Apr 11 08:17:25 2012 -0700
20218
20219 When deactivating an explicit pointer grab, reject all grabs on touches
20220
20221 Explicit pointer grabs are placed at the head of the touch listener
20222 array for pointer emulated touches. If the grab is deactivated, we must
20223 remove it from all touches for the device.
20224
20225 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20226 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20227
20228commit cacdb9a74065ccba7d50a82e14abdf04b36c5309
20229Author: Chase Douglas <chase.douglas@canonical.com>
20230Date: Wed Apr 11 08:13:17 2012 -0700
20231
20232 Accept touch sequence for pointer listener after second event delivery
20233
20234 This is a bit of unimplemented code for touchscreen pointer emulation. A
20235 pointer grabbing client currently never accepts the touch sequence. The
20236 sequence must be accepted once any touch-derived event is irrevocably
20237 delivered to a client.
20238
20239 The first pointer event, derived from a touch begin event, may be caught
20240 in a sync grab and then replayed. This is essentially a revocable
20241 delivery of an event. Thus, we must wait till a non-begin event is
20242 delivered.
20243
20244 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20245 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20246
20247commit 447fe7a1a72513aa68145962c47894242c589cc9
20248Author: Chase Douglas <chase.douglas@canonical.com>
20249Date: Wed Apr 11 08:12:08 2012 -0700
20250
20251 Split out helper function TouchListenerAcceptReject()
20252
20253 This will be used for accepting and rejecting touches in the future.
20254
20255 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20256 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20257
20258commit 93c33403642a3de3c9d141ad7940a7b880846aad
20259Author: Chase Douglas <chase.douglas@canonical.com>
20260Date: Wed Apr 4 12:59:55 2012 -0700
20261
20262 Only set XI2 mask if pointer emulation is for XI2 client
20263
20264 The current code returns a reference to memory that may not actually be
20265 an XI2 mask. Instead, only return a value when an XI2 client has
20266 selected for events.
20267
20268 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20269 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20270 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
20271
20272commit 4c1dfd21937efc6a85fb204a73dd7d7151d54daa
20273Author: Chase Douglas <chase.douglas@canonical.com>
20274Date: Wed Apr 4 12:57:40 2012 -0700
20275
20276 Check core event mask properly for pointer emulated touch events
20277
20278 The current code checks the core event mask as though it were an XI2
20279 mask. This change fixes the checks so the proper client and event masks
20280 are used.
20281
20282 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20283 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20284 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
20285
20286commit 3720aa33ee50788dd3d4acc9bbf8dfcb72c8f5ce
20287Author: Erkki Seppälä <erkki.seppala@vincit.fi>
20288Date: Thu Apr 7 12:53:51 2011 +0300
20289
20290 Added resourceproto version dependency, >= 1.2.0 now required
20291
20292 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20293
20294commit 233eab4d05cae1fdb4129a2e9905961b78693f74
20295Author: Erkki Seppälä <erkki.seppala@vincit.fi>
20296Date: Wed Dec 8 17:09:30 2010 +0200
20297
20298 dix: add reference count of the resource to ResourceSizeRec
20299
20300 The ResourceSizeRec now contains the number of references to the
20301 resource. For example a Pixmap knows this value and it can be useful
20302 for determining the "weight" of the resource. Typically this value
20303 is 1.
20304
20305 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
20306 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20307
20308commit ccb3e78124fb05defd0c9b438746b79d84dfc3ae
20309Author: Erkki Seppälä <erkki.seppala@vincit.fi>
20310Date: Tue Dec 14 12:18:23 2010 +0200
20311
20312 Xext: add a generic hashtable implementation
20313
20314 The generic hashtable implementation adds a key-value container, that
20315 keeps the key and value inside the hashtable structure and manages
20316 their memory by itself. This data structure is best suited for
20317 fixed-length keys and values.
20318
20319 One creates a new hash table with ht_create and disposes it with
20320 ht_destroy. ht_create accepts the key and value sizes (in bytes) in
20321 addition to the hashing and comparison functions to use. When adding
20322 keys with ht_add, they will be copied into the hash and a pointer to
20323 the value will be returned: data may be put into this structure (or if
20324 the hash table is to be used as a set, one can just not put anything
20325 in).
20326
20327 The hash table comes also with one generic hashing function plus a
20328 comparison function to facilitate ease of use. It also has a custom
20329 hashing and comparison functions for hashing resource IDs with
20330 HashXID.
20331
20332 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
20333 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20334
20335commit a2ac01a8ea8508ed35aa844a589672c1165e05e4
20336Author: Erkki Seppälä <erkki.seppala@vincit.fi>
20337Date: Wed Apr 6 10:16:53 2011 +0300
20338
20339 dix: don't use a local wrapper for calling HashResourceID
20340
20341 Calls to Hash(client, id) were replaced with calls directly to
20342 HashResourceID(id, clientTable[client].hashsize) and the Hash-function
20343 was removed.
20344
20345 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20346
20347commit a0b0fb83f91bb82534a0d83fdd6c0222567b098d
20348Author: Erkki Seppälä <erkki.seppala@vincit.fi>
20349Date: Mon Dec 20 12:58:37 2010 +0200
20350
20351 dix: add hashing functions to resource.h for others to use.
20352
20353 The public hashing function HashResourceID uses the same hashing
20354 hashing algorithm as resource.c uses internally, but it provides an
20355 interface that will is usable by external modules. It provides a
20356 parameter for the number of bits for the hash, instead of finding the
20357 size from its internal hash table.
20358
20359 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20360
20361commit 3ba0decb4b55a1fd122d269e15b2b2da8ced8624
20362Author: Erkki Seppälä <erkki.seppala@vincit.fi>
20363Date: Wed Dec 8 15:30:57 2010 +0200
20364
20365 dix: add a mechanism for iterating through all subresources
20366
20367 The mechanism allows iterating even through subresources that don't
20368 have specific XID's. When such 'resources' are iterated, the XID for
20369 them will be zero. A resource type can assign an iteration function
20370 for its subresources with SetResourceTypeFindSubResFunc; by default
20371 resources are assumed not to contain subresources.
20372
20373 The purpose of this extension is to enable accurate accounting of
20374 the resources a resource consumes or uses.
20375
20376 This patch provides the subresource iteration functions for Windows
20377 and GCs.
20378
20379 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
20380 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20381
20382commit b8d0d19a6d410776b53a41e7cae90f68d4b22bb7
20383Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
20384Date: Wed Oct 27 17:25:50 2010 +0300
20385
20386 composite: Report pixmap usage of client windows to resource extension.
20387
20388 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20389 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
20390 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
20391 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
20392
20393commit e83388cc70e21e7f377ed2e417d04469e23eb706
20394Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
20395Date: Wed Oct 27 16:59:06 2010 +0300
20396
20397 render: Report pixmap usage of pictures to resource extension.
20398
20399 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20400 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
20401 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
20402 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
20403
20404commit 96864bfa951ea8bf4ab697753fc62c6a97598bc0
20405Author: Erkki Seppälä <erkki.seppala@vincit.fi>
20406Date: Mon Nov 29 12:40:56 2010 +0200
20407
20408 Implemented first part of XResource extension v1.2: X_XResQueryClientIds
20409
20410 This patch implements a part of the XResource extension v1.2 (as specified in
20411 http://patchwork.freedesktop.org/patch/2720/ ). The request implemented is
20412 X_XResQueryClientIds.
20413
20414 This patch depends on the feature introduced by
20415 1e933665bef26c74196bb7c59910e6a78bcacf0e "dix: Add facilities for
20416 client ID tracking." .
20417
20418 This latest version also adds Doxygen-formatted comments and takes a better
20419 notice of coding conventions (as in http://www.x.org/wiki/CodingStyle ).
20420
20421 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
20422
20423commit ebf214876a4885a98ded4f5525925b69005fae05
20424Author: Peter Hutterer <peter.hutterer@who-t.net>
20425Date: Wed Apr 18 15:56:37 2012 +1000
20426
20427 dix: indentation fix
20428
20429 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20430
20431commit 80fefc42f5e67e6b4a4b440d8991bee7e5f38359
20432Merge: 9779b90 12188c8
20433Author: Keith Packard <keithp@keithp.com>
20434Date: Sun Apr 15 21:05:30 2012 -0700
20435
20436 Merge remote-tracking branch 'whot/for-keith'
20437
20438commit 12188c8a8a537b38b1ca4cf8c0de5447e19c886a
20439Author: Chase Douglas <chase.douglas@canonical.com>
20440Date: Tue Apr 3 17:31:01 2012 -0700
20441
20442 Use touch state when querying pointer through core protocol
20443
20444 QueryPointer is part of the core protocol. As such, it knows nothing
20445 about touch devices. Touches are converted to button 1 press, pointer
20446 motion, and button 1 release for core clients, so we should ensure the
20447 pointer state mask has button 1 set when XQueryPointer is used.
20448
20449 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20450 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20451
20452commit 32ece7c09bf0ebc3d99b4078aacebbd44314776a
20453Author: Chase Douglas <chase.douglas@canonical.com>
20454Date: Tue Apr 10 17:12:42 2012 -0700
20455
20456 Ensure sequential touches are pointer emulated sequentially
20457
20458 Issue:
20459 * Two sequential touches (i.e. down, up, down, up)
20460 * Both are grabbed by a touch grab
20461 * Both have a second listener in the form of a pointer grab or selection
20462 * The second and first touches are rejected in that order
20463
20464 The first touch must be pointer emulated before the second touch, so the
20465 second touch must be paused until the first touch is rejected or
20466 accepted and all events are delivered to pointer clients.
20467
20468 This change ensures all pointer emulated events are emitted
20469 sequentially. It necessarily imposes a delay on further touch events
20470 when pointer grabs and selections are used, but there is no way around
20471 it.
20472
20473 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20474 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20475 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20476
20477commit 163b0f375d73c05873fb341652de3ed347337828
20478Author: Chase Douglas <chase.douglas@canonical.com>
20479Date: Tue Apr 10 17:12:41 2012 -0700
20480
20481 Update event type when delivering end event to a pointer listener
20482
20483 Just like when we deliver to a touch listener, we must convert a touch
20484 end event to an update event for further clients. This also ensures that
20485 the touch record is not deleted at the end of ProcessTouchEvent().
20486
20487 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20488 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20489 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20490
20491commit 210cd12c47d063f97915ff23292b61d09abfd73a
20492Author: Chase Douglas <chase.douglas@canonical.com>
20493Date: Tue Apr 10 17:12:40 2012 -0700
20494
20495 Don't update listener after deactivating implicit pointer grab
20496
20497 After the pointer grab is deactivated, the touch listener record is
20498 updated at the end of DeliverTouchEmulatedEvent. However, the touch
20499 record is ended when the grab is deactivated, so the update to the
20500 listener record is in an array of memory that has been freed.
20501
20502 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20503 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20504 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20505
20506commit 82a1ae0af3b136371638659c3e909880a99f721c
20507Author: Peter Hutterer <peter.hutterer@who-t.net>
20508Date: Thu Apr 12 15:54:00 2012 +1000
20509
20510 xfree86: after VT switching back, only enable previously enabled devices
20511
20512 If a device was enabled before the VT switch, re-enabled it. Otherwise leave
20513 it as is, there was probably a reason why it was disabled.
20514
20515 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20516 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
20517 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
20518
20519commit c5a45b0f7658c77725adce2b64a0fbd62f208328
20520Author: Daniel Kurtz <djkurtz@chromium.org>
20521Date: Thu Apr 12 10:11:10 2012 +1000
20522
20523 dix: don't BUG_WARN for button events from button-only device
20524
20525 Events from button-only devices still need coordinates, and they get them
20526 from scale_to_desktop(). Therefore, a dev without valuators is not a bug.
20527 However, a dev with valuators, but less than two of them still is a bug.
20528
20529 This was noticed when unplugging a "Creative Technology SB Arena Headset",
20530 which has some BTNs and some KEYs, but no REL or ABS valuators.
20531 It emits [BTN_3] = 0 on unplug, which would trigger the BUG_WARN.
20532
20533 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
20534 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
20535 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20536
20537commit 9779b904c7c0b49c74054c22c420012c40595cdc
20538Author: Keith Packard <keithp@keithp.com>
20539Date: Wed Apr 11 09:33:54 2012 -0700
20540
20541 hw/xfree86: Re-indent xf86vmode.c
20542
20543 This is the result of re-running the 'x-indent.sh' script over
20544 xf86vmode.c to clean up the disaster caused by broken syntax in the
20545 file.
20546
20547 Signed-off-by: Keith Packard <keithp@keithp.com>
20548
20549commit 592bd0ae2b60cd6f6afd3efc40f5f659b12900b4
20550Author: Keith Packard <keithp@keithp.com>
20551Date: Wed Apr 11 09:28:21 2012 -0700
20552
20553 hw/xfree86: Spurious ');' in xf86vmode.c messed up indentation badly
20554
20555 Inside the unfinished XF86VIDMODE_EVENTS #ifdef block the
20556 function definition for xf86VidModeNotifyEvent had an extra ');'
20557 before the prototype argument declarations. This was harmless for the
20558 compiler as the code never gets used, but completely messed up the
20559 file re-indentation. This patch removes the spurious characters in
20560 preparation for re-indenting the file.
20561
20562 Signed-off-by: Keith Packard <keithp@keithp.com>
20563
20564commit 72cfc1a097dc1e09d2cd9415ef7855a2cef92351
20565Author: Peter Hutterer <peter.hutterer@who-t.net>
20566Date: Wed Apr 11 09:43:23 2012 +1000
20567
20568 Xi: fix XITouchClass sourceid assignment
20569
20570 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20571 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
20572
20573commit 10cd6fdc431d2d0835c313b2f5bdda6dcea281fd
20574Author: Keith Packard <keithp@keithp.com>
20575Date: Sun Apr 8 00:16:54 2012 -0700
20576
20577 Revert "xfree86: workaround crash on close"
20578
20579 This reverts commit 55f552adb651715d2620db7248cd5b9b8187654a.
20580
20581 This appears to cause a crash at init time instead of close.
20582
20583 Reported-by: Knut Petersen <Knut_Petersen@t-online.de>
20584 Acked-by: Jeremy Huddleston <jeremyhu@apple.com>
20585 Signed-off-by: Keith Packard <keithp@keithp.com>
20586
20587commit c7b162555849ab2d4cb195d02d98a40494c5ce55
20588Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20589Date: Thu Mar 29 02:26:41 2012 -0500
20590
20591 xfree86: link modules against Xorg symbols on Cygwin
20592
20593 As a PE platform, all symbols in both EXEs and DLLs must be resolved
20594 at link time. As Xorg modules depend on symbols in the Xorg
20595 executable, we must build Xorg before its modules, creating an implib
20596 from the former which is used to link the latter. This implib must
20597 then be installed in order to build the drivers.
20598
20599 Currently only two drivers are supported on Cygwin: xf86-video-dummy
20600 (to replace Xvfb/Xfake) and xf86-video-nested (to replace Xnest/Xephyr).
20601
20602 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20603 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20604 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20605
20606commit 2dffdcd60f70f534ad730830fe97f0bb1a192e2a
20607Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20608Date: Thu Mar 29 02:23:17 2012 -0500
20609
20610 xf86Config: load DIX libraries before drivers on Cygwin
20611
20612 Cygwin doesn't have ELF rpath capabilities, so these libraries need
20613 to be loaded before the drivers (namely dummy and nested) which
20614 depend on their symbols.
20615
20616 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20617 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20618 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20619
20620commit 5dd3d2dbbafee28142552640ac6ef3eef2e6d9d3
20621Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20622Date: Thu Mar 29 02:21:01 2012 -0500
20623
20624 xf86Init: provide ddxBeforeReset ifdef DDXBEFORERESET
20625
20626 This is necessary when building Xorg and XWin simultaneously, otherwise
20627 undefined symbol errors result in sdksyms.c.
20628
20629 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20630 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20631 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20632
20633commit 96186bc7219a88e549928707b3da4b8bfc16c54d
20634Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20635Date: Thu Mar 29 02:18:41 2012 -0500
20636
20637 loader: add Cygwin support
20638
20639 Cygwin libraries use the .dll extension and "cyg" prefix in place of "lib".
20640
20641 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20642 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20643 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20644
20645commit 0ce48729d34ab610ff119303355883048b651500
20646Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20647Date: Thu Mar 29 02:17:22 2012 -0500
20648
20649 os-support: add Cygwin support
20650
20651 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20652 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20653 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20654
20655commit fd115ee114e95b4e7b96f789d3ad67e2e15555de
20656Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20657Date: Thu Mar 29 02:13:27 2012 -0500
20658
20659 xfree86: allow modules to be built without undefined symbols
20660
20661 This will be necessary to port Xorg to Cygwin, but other platforms may
20662 find this useful as well.
20663
20664 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20665 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20666 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20667
20668commit 12cd7ca83c35bc61fbda7e61535692e3ec78194c
20669Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20670Date: Thu Mar 29 02:06:36 2012 -0500
20671
20672 Cygwin/X: disable all unused or unapplicable extensions in configure
20673
20674 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20675 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20676 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20677
20678commit cb5661e86d4a428ec0ca4efc30a1a165d94d7215
20679Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20680Date: Tue Mar 27 22:16:08 2012 -0500
20681
20682 xf86Crtc: include "xf86xv.h" only if XV
20683
20684 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
20685 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
20686 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
20687 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20688
20689commit 1110facdfeb95b1ad47d03c0ca3d73933b86dbd6
20690Author: Chase Douglas <chase.douglas@canonical.com>
20691Date: Thu Mar 29 18:06:03 2012 -0700
20692
20693 Implement passive touch ungrabbing
20694
20695 Whoops. Forgot to implement this. The code currently generates an error
20696 due to the unhandled grab type.
20697
20698 X.Org Bug 48069 <https://bugs.freedesktop.org/show_bug.cgi?id=48069>
20699
20700 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
20701 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
20702 Signed-off-by: Keith Packard <keithp@keithp.com>
20703
20704commit e9d3848d7bda7f7de4c0f497aee4482d0b30ad49
20705Author: Jeremy Huddleston <jeremyhu@apple.com>
20706Date: Thu Apr 5 14:03:35 2012 -0700
20707
20708 XQuartz: Automatically start our virtual tablet devices
20709
20710 Fixes: https://bugs.launchpad.net/inkscape/+bug/972914
20711 Regression introduced by: 7790dc86384cc451ac44663737fde84dd81ad4e1
20712
20713 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20714
20715commit 6cb83b78c47b80556cb4d573524d60f0cd1a64ac
20716Author: Jeremy Huddleston <jeremyhu@apple.com>
20717Date: Fri Mar 30 14:29:48 2012 -0700
20718
20719 XQuartz: Tiger build fix
20720
20721 Fixes: https://trac.macports.org/ticket/33818
20722
20723 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20724
20725commit eca58ee4db43d85e5940919be094c38ad166d5d0
20726Author: Jeremy Huddleston <jeremyhu@apple.com>
20727Date: Wed Mar 28 15:07:51 2012 -0700
20728
20729 os: Fix regression with FatalError not calling va_start
20730
20731 Regression From: a818b305989bbcde4e585112a7ee70cbc0b14a92
20732
20733 Found-by: Colin Harrison <colin.harrison@virgin.net>
20734 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20735
20736commit 55f552adb651715d2620db7248cd5b9b8187654a
20737Author: Michal Suchanek <hramrach@gmail.com>
20738Date: Wed Mar 28 18:12:39 2012 -0700
20739
20740 xfree86: workaround crash on close
20741
20742 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=41653
20743
20744 Signed-off-by: Michal Suchanek <hramrach@gmail.com>
20745 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20746
20747commit 641a1b9363d59808d2586f9e84847ccc69701482
20748Author: Peter Hutterer <peter.hutterer@who-t.net>
20749Date: Mon Mar 26 10:34:34 2012 +1000
20750
20751 test: reset the client before checking the swapped version
20752
20753 As of 2c23ef83b0e03e163aeeb06133538606886f4e9c, the server returns BadValue
20754 for the same client with multiple versions. Avoid this by resetting the
20755 client before we issue the same request as a fake swap client.
20756
20757 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20758 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20759 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
20760
20761commit 17f33ca0e5bff45a2d03e4d3b4753a714ce25baf
20762Author: Peter Hutterer <peter.hutterer@who-t.net>
20763Date: Mon Mar 26 10:28:32 2012 +1000
20764
20765 test: init the sync extensions for tests to pass again
20766
20767 Introduced in d645edd11e7482f98c8b7e0d6c8693285c484907
20768
20769 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20770 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
20771 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
20772
20773commit 8053faa7b81520b0a46459ba6e3b98f2599d5bbf
20774Author: Peter Hutterer <peter.hutterer@who-t.net>
20775Date: Fri Mar 23 08:20:22 2012 +1000
20776
20777 Bump input ABI to 17 for per-device idlecounters
20778
20779 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
20780
20781commit 7f3997b01a8813c1d7055317eb06111aed8572c9
20782Merge: 92d50c3 31646d8
20783Author: Keith Packard <keithp@keithp.com>
20784Date: Mon Mar 26 16:41:52 2012 -0700
20785
20786 Merge remote-tracking branch 'jeremyhu/master'
20787
20788commit 92d50c38b2bfeb82e7374a1966a53235e3f8275e
20789Author: Gaetan Nadon <memsize@videotron.ca>
20790Date: Sun Feb 26 12:15:31 2012 -0500
20791
20792 man: s/__xservername__/Xorg/g - no longer required
20793
20794 Excerpt from http://lists.x.org/archives/xorg-devel/2011-March/020481.html:
20795
20796 The Xorg & xorg.conf substitutions are leftover from the transitional
20797 period where some distros were building our sources with the XFree86
20798 and XF86config names until they had time to adjust the rest of their
20799 packages/installer/config code to the new names.
20800
20801 This will fix inconsistencies and prevent the creation of new unneeded
20802 sed patterns.
20803
20804 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
20805 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
20806 Signed-off-by: Keith Packard <keithp@keithp.com>
20807
20808commit 9e880cd2e36236e466fda0adadee34a407ae3341
20809Author: Gaetan Nadon <memsize@videotron.ca>
20810Date: Sun Feb 26 12:15:30 2012 -0500
20811
20812 man: s/__xconfigfile__/xorg.conf/g - no longer required
20813
20814 Excerpt from http://lists.x.org/archives/xorg-devel/2011-March/020481.html:
20815
20816 The Xorg & xorg.conf substitutions are leftover from the transitional
20817 period where some distros were building our sources with the XFree86
20818 and XF86config names until they had time to adjust the rest of their
20819 packages/installer/config code to the new names.
20820
20821 This will fix inconsistencies and prevent the creation of new unneeded
20822 sed patterns.
20823
20824 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
20825 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
20826 Signed-off-by: Keith Packard <keithp@keithp.com>
20827
20828commit 31646d8fa9524adca1d7bc2cd2db90d47c2eb96e
20829Author: Jeremy Huddleston <jeremyhu@apple.com>
20830Date: Sun Mar 25 23:33:54 2012 -0700
20831
20832 XQuartz: Use scrolling valuators
20833
20834 This lets use send more accurate data to Xi clients and uses dix
20835 for legacy scroll buttons rather than reinventing the wheel.
20836
20837 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20838 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
20839
20840commit 7790dc86384cc451ac44663737fde84dd81ad4e1
20841Author: Jeremy Huddleston <jeremyhu@apple.com>
20842Date: Sun Mar 25 22:30:55 2012 -0700
20843
20844 XQuartz: Use AllocDevicePair to allocate our keyboard and mouse at the same time
20845
20846 Also correct isMaster to FALSE while we're here.
20847
20848 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20849 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
20850
20851commit 74fb534bf15687610367248071341fe4bfd1d8bf
20852Author: Jeremy Huddleston <jeremyhu@apple.com>
20853Date: Sun Mar 25 22:16:22 2012 -0700
20854
20855 XQuartz: Call update_desktop_dimensions() after updating our screen coordinates
20856
20857 Fixes: http://xquartz.macosforge.org/trac/ticket/555
20858 Regression from: 8bebb4b4896d8b6ba3309b5b28fce883bb9f8a96
20859
20860 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20861 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
20862
20863commit 41a46546e92b66fed323b5845dd71a6385feb418
20864Author: Jeremy Huddleston <jeremyhu@apple.com>
20865Date: Sun Mar 25 23:05:43 2012 -0700
20866
20867 XQuartz: Formatting cleanup for sendX11NSEvent
20868
20869 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
20870
20871commit 5324557c7b12dcacddac830bdb65769cb622c6cf
20872Author: Jeremy Huddleston <jeremyhu@apple.com>
20873Date: Fri Mar 23 19:58:49 2012 -0700
20874
20875 XQuartz: Source formatting cleanup
20876
20877 indent butchered Objective-C formatting. This patch was created by:
20878
20879 1) Reverting the indent changes in hw/xquartz
20880 2) Editing X11Application.m and chaning some #ifdef logic to work
20881 with uncrustify
20882 3) Hand edited some (c) notifications
20883 4) Opened all XQuartz sources in XCode and re-indented (^I)
20884 5) Ran uncrustify with this configuration
20885 (as close to the indent rules as I could get):
20886
20887 tok_split_gte=false
20888 utf8_byte=true
20889 utf8_force=true
20890 indent_cmt_with_tabs=false
20891 indent_align_string=false
20892 indent_braces=false
20893 indent_braces_no_func=false
20894 indent_braces_no_class=false
20895 indent_braces_no_struct=false
20896 indent_brace_parent=false
20897 indent_namespace=false
20898 indent_extern=false
20899 indent_class=false
20900 indent_class_colon=false
20901 indent_else_if=false
20902 indent_var_def_cont=false
20903 indent_func_call_param=false
20904 indent_func_def_param=false
20905 indent_func_proto_param=false
20906 indent_func_class_param=false
20907 indent_func_ctor_var_param=false
20908 indent_template_param=false
20909 indent_func_param_double=false
20910 indent_relative_single_line_comments=false
20911 indent_col1_comment=false
20912 indent_access_spec_body=false
20913 indent_paren_nl=false
20914 indent_comma_paren=false
20915 indent_bool_paren=false
20916 indent_first_bool_expr=false
20917 indent_square_nl=false
20918 indent_preserve_sql=false
20919 indent_align_assign=true
20920 sp_balance_nested_parens=false
20921 align_keep_tabs=false
20922 align_with_tabs=false
20923 align_on_tabstop=false
20924 align_number_left=false
20925 align_func_params=false
20926 align_same_func_call_params=false
20927 align_var_def_colon=true
20928 align_var_def_attribute=true
20929 align_var_def_inline=true
20930 align_right_cmt_mix=false
20931 align_on_operator=false
20932 align_mix_var_proto=false
20933 align_single_line_func=false
20934 align_single_line_brace=false
20935 align_nl_cont=false
20936 align_left_shift=true
20937 align_oc_decl_colon=true
20938 nl_collapse_empty_body=true
20939 nl_assign_leave_one_liners=true
20940 nl_class_leave_one_liners=true
20941 nl_enum_leave_one_liners=true
20942 nl_getset_leave_one_liners=true
20943 nl_func_leave_one_liners=true
20944 nl_if_leave_one_liners=true
20945 nl_multi_line_cond=false
20946 nl_multi_line_define=false
20947 nl_before_case=true
20948 nl_after_case=true
20949 nl_after_return=false
20950 nl_after_semicolon=true
20951 nl_after_brace_open=true
20952 nl_after_brace_open_cmt=false
20953 nl_after_vbrace_open=false
20954 nl_after_vbrace_open_empty=false
20955 nl_after_brace_close=false
20956 nl_after_vbrace_close=false
20957 nl_define_macro=false
20958 nl_squeeze_ifdef=false
20959 nl_ds_struct_enum_cmt=false
20960 nl_ds_struct_enum_close_brace=false
20961 nl_create_if_one_liner=false
20962 nl_create_for_one_liner=false
20963 nl_create_while_one_liner=false
20964 ls_for_split_full=false
20965 ls_func_split_full=false
20966 nl_after_multiline_comment=false
20967 eat_blanks_after_open_brace=false
20968 eat_blanks_before_close_brace=false
20969 mod_full_brace_if_chain=false
20970 mod_pawn_semicolon=false
20971 mod_full_paren_if_bool=false
20972 mod_remove_extra_semicolon=false
20973 mod_sort_import=false
20974 mod_sort_using=false
20975 mod_sort_include=false
20976 mod_move_case_break=false
20977 mod_remove_empty_return=false
20978 cmt_indent_multi=true
20979 cmt_c_group=false
20980 cmt_c_nl_start=false
20981 cmt_c_nl_end=false
20982 cmt_cpp_group=false
20983 cmt_cpp_nl_start=false
20984 cmt_cpp_nl_end=false
20985 cmt_cpp_to_c=false
20986 cmt_star_cont=false
20987 cmt_multi_check_last=true
20988 cmt_insert_before_preproc=false
20989 pp_indent_at_level=false
20990 pp_region_indent_code=false
20991 pp_if_indent_code=false
20992 pp_define_at_level=false
20993 indent_columns=4
20994 indent_brace=0
20995 indent_switch_case=0
20996 align_struct_init_span=2
20997 align_pp_define_gap=0
20998 align_pp_define_span=2
20999 align_oc_msg_colon_span=16
21000 nl_end_of_file_min=1
21001 nl_func_var_def_blk=0
21002 code_width=78
21003 nl_max=2
21004 newlines=auto
21005 indent_with_tabs=0
21006 sp_arith=force
21007 sp_assign=force
21008 sp_assign_default=force
21009 sp_before_assign=force
21010 sp_after_assign=force
21011 sp_enum_assign=force
21012 sp_enum_before_assign=force
21013 sp_enum_after_assign=force
21014 sp_pp_stringify=add
21015 sp_bool=force
21016 sp_compare=force
21017 sp_inside_paren=remove
21018 sp_paren_paren=remove
21019 sp_paren_brace=force
21020 sp_before_ptr_star=ignore
21021 sp_before_unnamed_ptr_star=force
21022 sp_before_byref=force
21023 sp_before_unnamed_byref=force
21024 sp_after_byref=remove
21025 sp_after_type=force
21026 sp_before_sparen=force
21027 sp_inside_sparen=remove
21028 sp_inside_sparen_close=remove
21029 sp_after_sparen=force
21030 sp_sparen_brace=force
21031 sp_special_semi=force
21032 sp_before_semi=remove
21033 sp_after_semi=force
21034 sp_after_semi_for=force
21035 sp_after_semi_for_empty=remove
21036 sp_before_square=remove
21037 sp_inside_square=remove
21038 sp_after_comma=force
21039 sp_before_comma=remove
21040 sp_paren_comma=force
21041 sp_before_ellipsis=force
21042 sp_after_class_colon=force
21043 sp_before_class_colon=force
21044 sp_before_case_colon=remove
21045 sp_after_cast=remove
21046 sp_inside_paren_cast=remove
21047 sp_sizeof_paren=remove
21048 sp_inside_braces_enum=force
21049 sp_inside_braces_struct=force
21050 sp_inside_braces=force
21051 sp_inside_braces_empty=remove
21052 sp_func_proto_paren=remove
21053 sp_func_def_paren=remove
21054 sp_inside_fparens=remove
21055 sp_inside_fparen=remove
21056 sp_square_fparen=remove
21057 sp_fparen_brace=force
21058 sp_func_call_paren=remove
21059 sp_func_call_paren_empty=remove
21060 sp_return_paren=force
21061 sp_attribute_paren=remove
21062 sp_defined_paren=remove
21063 sp_macro=force
21064 sp_macro_func=force
21065 sp_else_brace=force
21066 sp_brace_else=force
21067 sp_brace_typedef=force
21068 sp_not=remove
21069 sp_inv=remove
21070 nl_start_of_file=remove
21071 nl_end_of_file=force
21072 nl_assign_square=remove
21073 nl_after_square_assign=remove
21074 nl_fcall_brace=remove
21075 nl_enum_brace=remove
21076 nl_struct_brace=remove
21077 nl_union_brace=remove
21078 nl_if_brace=remove
21079 nl_brace_else=force
21080 nl_elseif_brace=remove
21081 nl_else_brace=remove
21082 nl_else_if=remove
21083 nl_for_brace=remove
21084 nl_do_brace=remove
21085 nl_brace_while=remove
21086 nl_switch_brace=remove
21087 nl_case_colon_brace=force
21088 nl_func_type_name=force
21089 nl_func_type_name_class=force
21090 nl_func_proto_type_name=force
21091 nl_func_paren=remove
21092 nl_func_def_paren=remove
21093 nl_func_decl_start=remove
21094 nl_func_def_start=remove
21095 nl_func_decl_args=remove
21096 nl_func_decl_end=remove
21097 nl_func_def_end=remove
21098 nl_func_decl_end_single=remove
21099 nl_func_def_end_single=remove
21100 nl_func_decl_empty=remove
21101 nl_func_def_empty=remove
21102 nl_fdef_brace=force
21103 nl_return_expr=remove
21104 nl_before_if=ignore
21105 nl_after_if=ignore
21106 nl_before_for=ignore
21107 nl_after_for=ignore
21108 nl_before_while=ignore
21109 nl_after_while=ignore
21110 nl_before_switch=ignore
21111 nl_after_switch=ignore
21112 nl_before_do=ignore
21113 nl_after_do=ignore
21114 pp_space=remove
21115
21116 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21117
21118commit 3505e1faadddeeec85a0d3f823c877ea33f86e00
21119Author: Jeremy Huddleston <jeremyhu@apple.com>
21120Date: Fri Feb 17 13:15:12 2012 -0800
21121
21122 XQuartz: Detect FatalErrors on startup to prevent tight crash loops
21123
21124 If a FatalError occurs before the server finishes launching, it will
21125 not have drained the launchd-owned DISPLAY socket, so launchd will
21126 just relaunch it. This can cause the server to crash in a tight loop
21127 which will spam the user with CrashReporter windows that claim focus on
21128 appearance.
21129
21130 This allows users stuck in this loop to "deal" with the problem without
21131 popping up a crash report every 10 seconds.
21132
21133 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21134
21135commit a818b305989bbcde4e585112a7ee70cbc0b14a92
21136Author: Jeremy Huddleston <jeremyhu@apple.com>
21137Date: Fri Feb 17 12:35:02 2012 -0800
21138
21139 os: Pass the FatalError message to OsVendorFatalError
21140
21141 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21142 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
21143
21144commit 0bb8a2566debd6cd11164df9ddca534150ec440a
21145Author: Jeremy Huddleston <jeremyhu@apple.com>
21146Date: Tue Mar 13 00:15:55 2012 -0700
21147
21148 XQuartz: Add a defaults option to disable the RENDER extension
21149
21150 RENDER has some ugly issues on XQuartz, so add an option to disable RENDER.
21151
21152 Enables workaround for: https://bugs.freedesktop.org/show_bug.cgi?id=26124
21153
21154 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21155
21156commit 5ff46d651ac0cd396be2053b79675997ed7af765
21157Author: Jeremy Huddleston <jeremyhu@apple.com>
21158Date: Sat Mar 17 00:04:27 2012 -0700
21159
21160 XQuartz: Use doubles for input valuators
21161
21162 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21163
21164commit cd84c0949a9f397d8b803d3b87c9d02b260d4795
21165Author: Jeremy Huddleston <jeremyhu@apple.com>
21166Date: Sat Mar 17 00:08:19 2012 -0700
21167
21168 XQuartz: Xi: darwinPointer is now Relative
21169
21170 There is really no real reason why this should be necessary, but wine
21171 developers are stuborn, so doing this to try to work around this wine
21172 issue:
21173
21174 http://bugs.winehq.org/show_bug.cgi?id=29732
21175
21176 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21177
21178commit 87939bf8311c0e7d168741fe57bc8fac414811d8
21179Author: Jeremy Huddleston <jeremyhu@apple.com>
21180Date: Tue Mar 13 00:25:53 2012 -0700
21181
21182 XQuartz: Move our logs into an X11 subdirectory
21183
21184 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21185
21186commit df327f98f6b8e02fadcfd313a85f906722acc572
21187Author: Jeremy Huddleston <jeremyhu@apple.com>
21188Date: Mon Mar 5 00:05:48 2012 -0800
21189
21190 test: Fix 'make dist' for configurations that build the Xorg DDX
21191
21192 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21193
21194commit a7eac500e652f30deffd9dc5e623fab701077738
21195Merge: bf876c8 d645edd
21196Author: Peter Hutterer <peter.hutterer@who-t.net>
21197Date: Thu Mar 22 13:13:07 2012 +1000
21198
21199 Merge branch 'per-device-sync-counters' into for-keith
21200
21201commit d645edd11e7482f98c8b7e0d6c8693285c484907
21202Author: Peter Hutterer <peter.hutterer@who-t.net>
21203Date: Mon Mar 12 16:36:31 2012 +1000
21204
21205 Xext: Add per-device SyncCounters
21206
21207 Previously, we only had one idle alarm that was triggered for all devices,
21208 whenever the user used any device, came back from suspend, etc.
21209
21210 Add system SyncCounters for each device (named "DEVICEIDLETIME x", with x
21211 being the device id) that trigger on that device only. This allows for
21212 enabling/disabling devices based on interaction with other devices.
21213
21214 Popular use-case: disable the touchpad when the keyboard just above the
21215 touchpad stops being idle.
21216
21217 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21218 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21219 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21220 Reviewed-by: James Jones <jajones@nvidia.com>
21221
21222commit 6aef209ebc2e54f5465da505a780f7b4cc273ee0
21223Author: Peter Hutterer <peter.hutterer@who-t.net>
21224Date: Mon Mar 12 13:51:02 2012 +1000
21225
21226 Change lastDeviceIdleTime to be per-device
21227
21228 Preparation work for per-device idle counters.
21229
21230 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21231 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21232 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21233 Reviewed-by: James Jones <jajones@nvidia.com>
21234
21235commit 20cf0ef825e3f14b0688b691691e0aeba0a4860a
21236Author: Peter Hutterer <peter.hutterer@who-t.net>
21237Date: Tue Mar 13 10:21:23 2012 +1000
21238
21239 Xext: strdup() the SystemSyncCounter name
21240
21241 Required for future dynamic names.
21242
21243 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21244 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21245 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21246 Reviewed-by: James Jones <jajones@nvidia.com>
21247
21248commit f1b28aea4156f0381ea733ad2afbdd1f34f75599
21249Author: Peter Hutterer <peter.hutterer@who-t.net>
21250Date: Tue Mar 13 09:31:09 2012 +1000
21251
21252 Xext: store the bracket values for idle counters in the private
21253
21254 And drop the three global variables, we have a reference to the counter
21255 everywhere now.
21256
21257 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21258 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21259 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21260 Reviewed-by: James Jones <jajones@nvidia.com>
21261
21262commit d9553b2bbe06fba0b209218ffed9465edd79b4d2
21263Author: Peter Hutterer <peter.hutterer@who-t.net>
21264Date: Tue Mar 13 09:29:39 2012 +1000
21265
21266 Xext: pass the counter into block/wakeup handlers
21267
21268 No functional changes, currently unused. Preparation work, we don't need a
21269 global variable if we can pass the counters around anyway.
21270
21271 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21272 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21273 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21274 Reviewed-by: James Jones <jajones@nvidia.com>
21275
21276commit 90e6dc6de1ac339212989fd8e54131b196ebb369
21277Author: Peter Hutterer <peter.hutterer@who-t.net>
21278Date: Tue Mar 13 09:28:15 2012 +1000
21279
21280 Xext: add a private field to SyncSystemCounters
21281
21282 Will be used to store counter-specific data.
21283
21284 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21285 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21286 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21287 Reviewed-by: James Jones <jajones@nvidia.com>
21288
21289commit 3ddae647c307005309daa2d5dfe4bc6acb8170ab
21290Author: Peter Hutterer <peter.hutterer@who-t.net>
21291Date: Mon Mar 12 15:31:39 2012 +1000
21292
21293 Xext: localise pIdleTimeValueLess/Greater
21294
21295 Cleanup for future features, no functional changes.
21296
21297 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21298 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21299 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21300 Reviewed-by: James Jones <jajones@nvidia.com>
21301
21302commit 43eb2f2758dfc6ca5a49afce97cc5baea8caf9f5
21303Author: Peter Hutterer <peter.hutterer@who-t.net>
21304Date: Mon Mar 12 15:27:56 2012 +1000
21305
21306 Xext: localise use of IdleTimeCounter
21307
21308 Instead of referring to the global IdleTimeCounter everywhere, only do it
21309 once and use a local variable for the rest.
21310
21311 Cleanup for future features, no functional changes.
21312
21313 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21314 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21315 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21316 Reviewed-by: James Jones <jajones@nvidia.com>
21317
21318commit 1f12f059ef994e0b9b68fbd1f1556d0285c96b8b
21319Author: Jamey Sharp <jamey@minilop.net>
21320Date: Wed Mar 14 17:22:18 2012 -0700
21321
21322 sync: Use a linked list instead of an array for SysCounterList.
21323
21324 Signed-off-by: Jamey Sharp <jamey@minilop.net>
21325 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21326 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21327
21328commit bf876c87a9099fdfa63ed599f8ed9a954dd023d9
21329Merge: 908ab3d c0b0a9b
21330Author: Peter Hutterer <peter.hutterer@who-t.net>
21331Date: Thu Mar 22 11:34:43 2012 +1000
21332
21333 Merge branch 'dtrace-input-abi' into for-keith
21334
21335commit c0b0a9bce9237b0abe150c1a7b54939affecc751
21336Author: Peter Hutterer <peter.hutterer@who-t.net>
21337Date: Mon Mar 5 14:12:52 2012 +1000
21338
21339 dix: add dtrace probes to input API
21340
21341 For driver debugging, it is helpful to know whether the driver has actually
21342 submitted an event to the server. dtrace hooks can help here.
21343
21344 Note that GetPointerEvents and friends may also be triggered by the server
21345 for other emulated devices, some care must be taken when analysing the
21346 results.
21347
21348 Additional difficulty: proximity events have a run-time assigned type, so
21349 this may make automatic detection a tad harder. If in doubt, go for any
21350 event > 64 since the only two that can have that value are ProximityIn and
21351 ProximityOut.
21352
21353 An example systemtap script is below:
21354
21355 # Compile+run with
21356 # stap -g xorg.stp /usr/bin/Xorg
21357 #
21358
21359 function print_valuators:string(nvaluators:long, mask_in:long, valuators_in:long) %{
21360 int i;
21361 unsigned char *mask = (unsigned char*)THIS->mask_in;
21362 double *valuators = (double*)THIS->valuators_in;
21363 char str[128] = {0};
21364 char *s = str;
21365
21366 #define BitIsSet(ptr, bit) (((unsigned char*)(ptr))[(bit)>>3] & (1 << ((bit) & 7)))
21367
21368 s += sprintf(s, "nval: %d ::", (int)THIS->nvaluators);
21369 for (i = 0; i < THIS->nvaluators; i++)
21370 {
21371 s += sprintf(s, " %d: ", i);
21372 if (BitIsSet(mask, i))
21373 s += sprintf(s, "%d", (int)valuators[i]);
21374 }
21375
21376 sprintf(THIS->__retvalue, "%s", str);
21377 %}
21378
21379 probe process(@1).mark("input__event")
21380 {
21381 deviceid = $arg1
21382 type = $arg2
21383 detail = $arg3
21384 flags = $arg4
21385 nvaluators = $arg5
21386
21387 str = print_valuators(nvaluators, $arg6, $arg7)
21388 printf("Event: device %d type %d detail %d flags %#x %s\n",
21389 deviceid, type, detail, flags, str);
21390 }
21391
21392 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21393 Acked-by: Jeremy Huddleston <jeremyhu@apple.com>
21394
21395commit 61cb98da1c6199964825de158d9eee7682d9c983
21396Author: Peter Hutterer <peter.hutterer@who-t.net>
21397Date: Mon Mar 12 16:03:50 2012 +1000
21398
21399 Xext: SyncCreateSystemCounter returns a SyncCounter*
21400
21401 type safety++
21402
21403 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21404 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21405 Reviewed-by: Bryce Harrington <bryce@canonical.com>
21406
21407commit 908ab3d580188533168c8cdfd2cab9dc689b4218
21408Author: Peter Hutterer <peter.hutterer@who-t.net>
21409Date: Wed Mar 21 14:05:29 2012 +1000
21410
21411 dix: set raw event values before adding up relative values (#46976)
21412
21413 Regression introduced in 4e52cc0ef48145134cd58d357fb7289e6f8bb709
21414
21415 Raw event values are values as-is from the driver, modified only be
21416 transformation or acceleration. 4e52cc caused the mask to be updated from
21417 relative to absolute coordinates which then got written into the raw events.
21418
21419 Move the raw event update into the respective branches for absolute/relative
21420 events.
21421
21422 X.Org Bug 46976 <http://bugs.freedesktop.org/show_bug.cgi?id=46976>
21423
21424 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21425 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
21426 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
21427 Tested-by: Sven Arvidsson <sa@whiz.se>
21428 Reviewed-by: Simon Thum <simon.thum@gmx.de>
21429
21430commit 45fe3085f8f45e529383025414fdd263d86dfd2b
21431Author: Peter Hutterer <peter.hutterer@who-t.net>
21432Date: Mon Mar 12 14:08:39 2012 +1000
21433
21434 Xext: remove needless /* parameter */ comments in declaration
21435
21436 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21437 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21438 Reviewed-by: Bryce Harrington <bryce@canonical.com>
21439
21440commit 9c3bd3ae652af386b6821b197d24528f20ba867d
21441Author: Peter Hutterer <peter.hutterer@who-t.net>
21442Date: Wed Mar 21 14:03:27 2012 +1000
21443
21444 dix: fix compiler warning "unused variable 'scr'"
21445
21446 getevents.c: In function 'updateSlaveDeviceCoords':
21447 getevents.c:326:15: warning: unused variable 'scr' [-Wunused-variable]
21448
21449 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21450
21451commit e21ffff4761d2b75815391c6947adcba425ab11e
21452Author: Peter Hutterer <peter.hutterer@who-t.net>
21453Date: Mon Mar 12 14:07:44 2012 +1000
21454
21455 Xext: typedef QueryValue and BracketValue prototypes
21456
21457 No functional changes, just for readability
21458
21459 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21460 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21461 Reviewed-by: Bryce Harrington <bryce@canonical.com>
21462
21463commit 14e3ea730eed344e6ed69c873f918d6f076d13dc
21464Author: Peter Hutterer <peter.hutterer@who-t.net>
21465Date: Wed Mar 14 13:48:56 2012 +1000
21466
21467 include: add an X_DEBUG message type
21468
21469 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21470 Reviewed-by: Jamey Sharp <jamey@minilop.net>
21471
21472commit 5910f2df58beaae2187438fef0b62c29a563e853
21473Author: Peter Hutterer <peter.hutterer@who-t.net>
21474Date: Mon Mar 12 16:26:29 2012 +1000
21475
21476 Xext: drop InitServertime() declaration.
21477
21478 Not implemented anywhere.
21479
21480 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21481 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21482 Reviewed-by: Bryce Harrington <bryce@canonical.com>
21483
21484commit 31df08a449cf9b6e1740e1c02257997490630d93
21485Author: Chase Douglas <chase.douglas@canonical.com>
21486Date: Wed Mar 7 16:06:27 2012 -0800
21487
21488 Use a new sprite trace for indirect touches when all touches have physically ended
21489
21490 All touches of an indirect device, such as a trackpad, are sent to the
21491 same window set. When there are no active touches, a new window set is
21492 created; otherwise, the window set of an existing touch is copied.
21493
21494 The current code checks for any logically active touches. This includes
21495 touches that have physically ended but are still logically active due to
21496 unhandled touch grabs. Instead, we want a new window set whenever there
21497 are no physically active touches.
21498
21499 This change skips over logically active but pending end touches, which
21500 are touches that have physically ended.
21501
21502 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21503 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21504 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21505
21506commit 58427e08a4a36ce9e213e4b4fe5249d5db2c764d
21507Author: Chase Douglas <chase.douglas@canonical.com>
21508Date: Wed Mar 7 16:06:26 2012 -0800
21509
21510 Xi: Fix TouchEnd to TouchUpdate change for one accepted grab
21511
21512 If there is only one listener of a touch, the listener is a grab, and is
21513 accepted before the touch has ended, the current code will not end the
21514 touch record when the touch does end.
21515
21516 This change adds a listener state for when a touch is accepted but has
21517 not yet ended. We now keep the touch record alive in this state, but end
21518 it when the touch ends.
21519
21520 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21521 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21522 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21523
21524commit e884ff8ad4df2b3272a3d8ece772906207af5142
21525Author: Peter Hutterer <peter.hutterer@who-t.net>
21526Date: Wed Mar 7 19:08:33 2012 +1000
21527
21528 xfree86: remove out-of-date comment
21529
21530 No idea what this was referring to but it goes past git history.
21531
21532 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21533
21534commit 5497ce3da442d27c2dc7796bfef6ccd670bbadc4
21535Author: Peter Hutterer <peter.hutterer@who-t.net>
21536Date: Mon Feb 20 12:09:33 2012 +1000
21537
21538 dix: IsFloating() on master devices is always false
21539
21540 There are a few subtle bugs during startup where IsFloating() returns true
21541 if the device is a master device that is not yet paired with its keyboard
21542 device.
21543
21544 Force IsFloating() to always return FALSE for master devices, that was the
21545 intent after all and any code that relies on the other behaviour should be
21546 fixed instead.
21547
21548 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21549 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
21550 Tested-by: Jon TURNEY <jon.turney@dronecode.org.uk>
21551
21552commit 2c23ef83b0e03e163aeeb06133538606886f4e9c
21553Author: Peter Hutterer <peter.hutterer@who-t.net>
21554Date: Mon Feb 27 08:01:07 2012 +1000
21555
21556 Xi: prohibit multiple XIQueryVersion requests with different versions
21557
21558 Return BadValue if major or minor differs on the second call.
21559
21560 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21561 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
21562
21563commit eb84c154ed38194c32651727b6dfe2d1bde4c599
21564Author: Peter Hutterer <peter.hutterer@who-t.net>
21565Date: Mon Feb 27 10:09:44 2012 +1000
21566
21567 dix: when rescaling from master, rescale from desktop dimensions (#46657)
21568
21569 master->last.valuators[] is in desktop dimensions, so use those as
21570 rescale axis ranges, not the screen. Otherwise, a rescale on any screen
21571 not the top-left will cause out-of-bounds coordinates which will always
21572 map to the bottom-right screen, causing the device to be stuck on that
21573 screen.
21574
21575 X.Org Bug 46657 <http://bugs.freedesktop.org/show_bug.cgi?id=46657>
21576
21577 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21578 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
21579
21580commit 6b6afd3d013e3f4910fae3520d1d786df2b0e47a
21581Author: Peter Hutterer <peter.hutterer@who-t.net>
21582Date: Thu Feb 16 15:11:40 2012 +1000
21583
21584 Xext: return BadAccess if PickPointer fails (#45796)
21585
21586 PickPointer or PickKeyboard return NULL, all MDs are currently disabled and
21587 we cannot emulate a core event. This wasn't anticipated by the protocol, so
21588 we don't really have an error code we may use here - BadAccess is simply the
21589 least bad of the possible ones.
21590
21591 And returning BadAccess beats crashing the server.
21592
21593 X.Org Bug 45796 <http://bugs.freedesktop.org/show_bug.cgi?id=45796>
21594
21595 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21596
21597commit 9e017cf0cf1f0c9d0d9c2cfeb82ea5dc0eb5905e
21598Author: Andreas Wettstein <wettstein509@solnet.ch>
21599Date: Sat Feb 25 20:48:17 2012 +0100
21600
21601 XKB: Redirect actions defunct with Gtk3 (XInput?)
21602
21603 When redirect actions are used with Gtk3, Gtk3 complained about
21604 events not holding a GdkDevice. This was caused by device IDs
21605 not being set for redirect actions.
21606
21607 More seriously, Gtk3 did not receive state changes redirect
21608 actions might specify. This was because event_set_state in
21609 dix/inpututils.c accesses the prev_state field, but the changes
21610 for the redirect action were only put into the state field.
21611
21612 Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
21613 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21614 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21615
21616commit ab3a815a75ab5695753fa37a98b0ea5293d4cb91
21617Author: Daniel Stone <daniel@fooishbar.org>
21618Date: Thu Mar 15 15:18:29 2012 +0000
21619
21620 Indentation: Change '& stuff' to '&stuff'
21621
21622 If the typedef wasn't perfect, indent would get confused and change:
21623 foo = (SomePointlessTypedef *) &stuff[1];
21624 to:
21625 foo = (SomePointlessTypedef *) & stuff[1];
21626
21627 Fix this up with a really naïve sed script, plus some hand-editing to
21628 change some false positives in XKB back.
21629
21630 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
21631
21632commit 58b1f739d73b03ff7952ca986ed8746a7307fffe
21633Author: Daniel Stone <daniel@fooishbar.org>
21634Date: Thu Mar 15 15:13:51 2012 +0000
21635
21636 Xinerama: Fix up obnoxiously-indented code
21637
21638 Comments inside of conditionals confuse indent to death. Which is fair
21639 enough really.
21640
21641 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
21642
21643commit 9838b7032ea9792bec21af424c53c07078636d21
21644Author: Keith Packard <keithp@keithp.com>
21645Date: Wed Mar 21 12:55:09 2012 -0700
21646
21647 Introduce a consistent coding style
21648
21649 This is strictly the application of the script 'x-indent-all.sh'
21650 from util/modular. Compared to the patch that Daniel posted in
21651 January, I've added a few indent flags:
21652
21653 -bap
21654 -psl
21655 -T PrivatePtr
21656 -T pmWait
21657 -T _XFUNCPROTOBEGIN
21658 -T _XFUNCPROTOEND
21659 -T _X_EXPORT
21660
21661 The typedefs were needed to make the output of sdksyms.sh match the
21662 previous output, otherwise, the code is formatted badly enough that
21663 sdksyms.sh generates incorrect output.
21664
21665 The generated code was compared with the previous version and found to
21666 be essentially identical -- "assert" line numbers and BUILD_TIME were
21667 the only differences found.
21668
21669 The comparison was done with this script:
21670
21671 dir1=$1
21672 dir2=$2
21673
21674 for dir in $dir1 $dir2; do
21675 (cd $dir && find . -name '*.o' | while read file; do
21676 dir=`dirname $file`
21677 base=`basename $file .o`
21678 dump=$dir/$base.dump
21679 objdump -d $file > $dump
21680 done)
21681 done
21682
21683 find $dir1 -name '*.dump' | while read dump; do
21684 otherdump=`echo $dump | sed "s;$dir1;$dir2;"`
21685 diff -u $dump $otherdump
21686 done
21687
21688 Signed-off-by: Keith Packard <keithp@keithp.com>
21689 Acked-by: Daniel Stone <daniel@fooishbar.org>
21690 Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
21691
21692commit 75199129c603fc8567185ac31866c9518193cb78
21693Author: Keith Packard <keithp@keithp.com>
21694Date: Wed Mar 21 12:33:19 2012 -0700
21695
21696 Handle blank betweeen type and name in sdksyms.sh
21697
21698 indent sometimes adds a blank line between the type and the name in a
21699 function declaration that includes _X_EXPORT, so handle that before
21700 the files are re-indented.
21701
21702 Signed-off-by: Keith Packard <keithp@keithp.com>
21703
21704commit a615b90cab7569fae9d123e4da1d3373c871d84b
21705Author: Keith Packard <keithp@keithp.com>
21706Date: Wed Mar 14 11:32:36 2012 -0700
21707
21708 Bump version number to 1.12.99.0
21709
21710 Now that 1.12 has branched, reset the version on master to a
21711 development number.
21712
21713 Signed-off-by: Keith Packard <keithp@keithp.com>
21714
21715commit b1be72c5ca6cb98ba64637990b142be0f1710a19
21716Author: Keith Packard <keithp@keithp.com>
21717Date: Sun Mar 4 20:26:18 2012 -0800
21718
21719 Version bumped to 1.12
21720
21721 Signed-off-by: Keith Packard <keithp@keithp.com>
21722
21723commit e08ed0b757b9b48344a301f612fabb3e39ffec78
21724Author: Gaetan Nadon <memsize@videotron.ca>
21725Date: Sun Feb 26 17:51:50 2012 -0500
21726
21727 test: add new test cases to .gitignore
21728
21729 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21730 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
21731 Signed-off-by: Keith Packard <keithp@keithp.com>
21732
21733commit 2416ee4a015068359807a10f433e8c54192c78a9
21734Author: Peter Hutterer <peter.hutterer@who-t.net>
21735Date: Wed Feb 22 15:32:56 2012 +1000
21736
21737 dix: avoid NULL-pointer dereference on button-only devices (#38313)
21738
21739 And for such devices simply take the last.valuators[] which must be valid at
21740 all times anyway. UpdateSlaveDeviceCoords takes care of that.
21741
21742 X.Org Bug 38313 <http://bugs.freedesktop.org/show_bug.cgi?id=38313>
21743
21744 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21745 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
21746
21747commit 38000e7d1f958f5944e641de3e716944a5876d41
21748Author: Jeremy Huddleston <jeremyhu@apple.com>
21749Date: Tue Jan 31 11:06:21 2012 -0800
21750
21751 Revert "dix: don't XWarpPointer through the last slave anymore (#38313)"
21752
21753 This reverts commit 2bfb802839688ecf328119c4c6979390fc60348d.
21754
21755 This commit caused a regression.
21756
21757 See: http://xquartz.macosforge.org/trac/ticket/517#comment:10
21758
21759 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
21760
21761commit cb6a32da27f09261c4d561c19d1877c750d98cc3
21762Author: Jeremy Huddleston <jeremyhu@apple.com>
21763Date: Fri Feb 17 13:44:12 2012 -0800
21764
21765 XQuartz: Short-circuit activateX:
21766
21767 This also avoids a deadlock when calling activateX: before the server
21768 thread has initialized
21769
21770 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21771
21772commit 1562c4a1ca748d2c4a8c69d57ba09ff7ca9cc453
21773Author: Jeremy Huddleston <jeremyhu@apple.com>
21774Date: Fri Feb 17 13:54:57 2012 -0800
21775
21776 XQuartz: Actually install the new locales
21777
21778 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21779
21780commit 71594746c7da32e1c7986341a0da30f241a7a2be
21781Merge: d53235a eaba06a
21782Author: Keith Packard <keithp@keithp.com>
21783Date: Wed Feb 22 18:07:20 2012 +1300
21784
21785 Merge remote-tracking branch 'whot/for-keith'
21786
21787commit eaba06a27c5520a02f08431ac1e4b0e0bdc22cd8
21788Author: Chase Douglas <chase.douglas@canonical.com>
21789Date: Mon Feb 13 16:00:48 2012 -0800
21790
21791 Keep virtual core pointer touch class around if new slave doesn't have one
21792
21793 The VCP may have active touch grabs. The touch records must be kept so
21794 these touch grabs may be accepted/rejected in the future. This means the
21795 touch class list will not represent the touch class of the attached
21796 slave device if it does not have a touch class, but we already were
21797 breaking that assumption by keeping a separate touches array for the
21798 VCP.
21799
21800 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21801 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21802 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21803
21804commit 8573b3519af138a3a12a2e77098718165f9fd8ff
21805Author: Peter Hutterer <peter.hutterer@who-t.net>
21806Date: Mon Feb 13 16:00:47 2012 -0800
21807
21808 Don't clobber virtual core pointer touches array length
21809
21810 The VCP has its own touches array, don't overwrite it when the class is
21811 copied from the SD to the master.
21812
21813 Reported-by: Chase Douglas <chase.douglas@canonical.com>
21814 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21815 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
21816
21817commit 03d32fe7a718d9016053cdb5d57f51a74ef99b59
21818Author: Chase Douglas <chase.douglas@ubuntu.com>
21819Date: Mon Feb 13 12:09:32 2012 -0800
21820
21821 Don't dereference a touch after it has been ended when punting to next owner
21822
21823 In this case, we have ended the touch because the last owner has
21824 rejected it. We need to return from the function right now so we don't
21825 attempt to dereference another touch client for early acceptance
21826 processing.
21827
21828 Signed-off-by: Chase Douglas <chase.douglas@ubuntu.com>
21829 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
21830 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21831
21832commit 1ecb7aaf2adedad1996cd26176ef5802113e3ad9
21833Author: Chase Douglas <chase.douglas@ubuntu.com>
21834Date: Sat Feb 11 07:58:38 2012 -0800
21835
21836 Focus event button state must show the logical buttons, not physical buttons
21837
21838 Similar to the fix in fcda98c48610fd507ca0b89c6006a5497d9dc1c9. This
21839 ensures we show the correct logical state of the buttons in device focus
21840 events too.
21841
21842 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21843 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21844 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21845
21846commit 6f28388187cffae9e5bc9bfc9425acff4f478b59
21847Author: Peter Hutterer <peter.hutterer@who-t.net>
21848Date: Sat Feb 11 01:29:26 2012 +1000
21849
21850 dix: reset last.scroll when resetting the valuator (#45611)
21851
21852 last.scroll remained on the last-submitted scrolling value but last.valuator
21853 was changed whenever the slave device changed. The first scrolling delta
21854 after a switch was then calculated as (last.scroll - new abs value), causing
21855 erroneous scrolling events.
21856
21857 Test case:
21858 - synaptics with a scrolling method enabled, other device with 3+ axes (e.g.
21859 wacom)
21860 - scroll on touchpad
21861 - use other device
21862 - scroll on touchpad
21863
21864 The second scroll caused erroneous button press/release events.
21865
21866 X.Org Bug 45611 <http://bugs.freedesktop.org/show_bug.cgi?id=45611>
21867
21868 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21869 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
21870
21871commit d53235af85d50774c68347720ce132daf9a5bc49
21872Author: Jeremy Huddleston <jeremyhu@apple.com>
21873Date: Mon Feb 13 23:06:07 2012 -0800
21874
21875 XQuartz: Build fix for possible conflict of BOOL type
21876
21877 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
21878
21879commit 62edd970f6f36058fcb31de4555eb7c1329cce74
21880Author: Keith Packard <keithp@keithp.com>
21881Date: Sat Feb 11 16:13:44 2012 +1300
21882
21883 Bump to version 1.11.99.903 (1.12 RC3)
21884
21885 Signed-off-by: Keith Packard <keithp@keithp.com>
21886
21887commit 42b6756463ee0476340656707f1088dc6c2fd220
21888Merge: 7674d00 ca64912
21889Author: Keith Packard <keithp@keithp.com>
21890Date: Sat Feb 11 15:36:43 2012 +1300
21891
21892 Merge remote-tracking branch 'alanc/master'
21893
21894commit 7674d00b04da5cf73cfa5c7ed1d3a9f42b59960e
21895Author: Peter Hutterer <peter.hutterer@who-t.net>
21896Date: Fri Jan 27 12:41:09 2012 +1000
21897
21898 Xi: handle new XIAllowEvents request in inputproto 2.1.99.6
21899
21900 grab_window and touchid were removed from the struct for ABI compatibility
21901 reasons, we need to pull in the new, XI 2.2-specific struct.
21902
21903 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21904 Reviewed-by: Keith Packard <keithp@keithp.com>
21905 Signed-off-by: Keith Packard <keithp@keithp.com>
21906
21907commit ceb026c6a6f1eea8b34e745f06f1ebcd652c0ba1
21908Author: Peter Hutterer <peter.hutterer@who-t.net>
21909Date: Fri Jan 27 12:41:09 2012 +1000
21910
21911 Xi: handle new XIAllowEvents request in inputproto 2.1.99.6
21912
21913 grab_window and touchid were removed from the struct for ABI compatibility
21914 reasons, we need to pull in the new, XI 2.2-specific struct.
21915
21916 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21917
21918commit b96275c4cdb164aa71f7aa9fbf88be18886d1936
21919Author: Benjamin Otte <otte@redhat.com>
21920Date: Tue Feb 7 18:01:25 2012 +1000
21921
21922 dix: fix an out-of-memory crash
21923
21924 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21925 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21926
21927commit 6241b5e4fdbdb08d30cc8787d858ac27122d2d49
21928Author: Chase Douglas <chase.douglas@canonical.com>
21929Date: Fri Feb 3 16:19:11 2012 -0800
21930
21931 Implement touch early accept
21932
21933 This doesn't really implement early accept as it should. Ideally, the
21934 server should send end events to all subsequent touch clients as soon as
21935 an early accept comes in. However, this implementation is still protocol
21936 compliant. We can always improve it later.
21937
21938 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21939 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21940 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21941
21942commit b0c54856df71f9cabf9dad176fdade960ef8c5d9
21943Author: Chase Douglas <chase.douglas@canonical.com>
21944Date: Fri Feb 3 16:19:10 2012 -0800
21945
21946 Implement early touch reject
21947
21948 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21949 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21950 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21951
21952commit 656ab879f20892975510723ce8fe78faf64aadb4
21953Author: Chase Douglas <chase.douglas@canonical.com>
21954Date: Fri Feb 3 16:19:09 2012 -0800
21955
21956 Check for proper window ID when processing touch allow requests
21957
21958 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21959 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21960 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21961
21962commit 9a260e9af83feb5c53ffd3b2da2dc3adf06240a1
21963Author: Chase Douglas <chase.douglas@canonical.com>
21964Date: Fri Feb 3 16:19:08 2012 -0800
21965
21966 Move AllowTouch to dix/touch.c, and rename to TouchAcceptReject
21967
21968 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21969 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21970 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21971
21972commit 192b2c9a2ec04522655675ddfe71de5cf974cc7b
21973Author: Chase Douglas <chase.douglas@canonical.com>
21974Date: Fri Feb 3 16:19:07 2012 -0800
21975
21976 Export TouchEventRejected as TouchRejected
21977
21978 This function is mostly correct for early reject usage. With a small
21979 change to pass the client resource explicitly and making the
21980 TouchOwnership event optional, it is usable for all rejection scenarios.
21981 This change exports it for use outside Xi/exevents.c and modifies the
21982 name accordingly.
21983
21984 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21985 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
21986 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
21987
21988commit 19073425e5df1317f73f74ce0b95ab24e0e0da14
21989Author: Chase Douglas <chase.douglas@canonical.com>
21990Date: Fri Feb 3 16:19:06 2012 -0800
21991
21992 Factor out TouchEnd generation and delivery
21993
21994 The server often needs to generate and deliver TouchEnd events for
21995 circumstances including touch grab acceptance and rejection. This change
21996 refactors the code so it can be used more easily.
21997
21998 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
21999 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22000 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22001
22002commit ab60cadc2afb9d8882144e30fde11c3ef0f999cd
22003Author: Chase Douglas <chase.douglas@canonical.com>
22004Date: Fri Feb 3 16:19:05 2012 -0800
22005
22006 Store window pointer in touch listener record
22007
22008 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
22009 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22010 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22011
22012commit f92ba92adfce0e1975feb54ab0e1165b7c2a7bea
22013Author: Peter Hutterer <peter.hutterer@who-t.net>
22014Date: Sat Feb 4 19:27:21 2012 +1000
22015
22016 Revert "dix: deduplicate callers of DeliverDeviceEvents in DeliverGrabbedEvents"
22017
22018 This call was supposed to have no functional changes but in some cases
22019 DeliverDeviceEvents() was called with a uninitialised win variable.
22020 Revert, safer than trying to sort this out otherwise.
22021
22022 This reverts commit 6eff14a789341d366b3013c5aa020e959c954651.
22023
22024 Reported-by: Mathieu Taillefumier <mathieu.taillefumier@free.fr>
22025 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22026 Reviewed-by: Keith Packard <keithp@keithp.com>
22027
22028commit ca64912c02bdff486fee420a49b11f54f8f5ba08
22029Author: Alan Coopersmith <alan.coopersmith@oracle.com>
22030Date: Thu Jan 26 16:41:25 2012 -0800
22031
22032 Namespace list api to reduce conflicts with similar system headers
22033
22034 Rename functions/macros from list_* to xorg_list_*
22035 Rename struct from struct list to struct xorg_list.
22036
22037 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
22038 Reviewed-by: Keith Packard <keithp@keithp.com>
22039 In-sed-I-trust: Peter Hutterer <peter.hutterer@who-t.net>
22040
22041commit 1541e242d1607d0db57b0cc12faca9b1c2850c67
22042Author: Alan Coopersmith <alan.coopersmith@oracle.com>
22043Date: Fri Dec 23 15:17:28 2011 -0800
22044
22045 Stop including <sys/proc.h> from xf86_OSlib.h on Solaris
22046
22047 We don't need anything from that header (which defines /proc & kernel
22048 structures for process information), and it causes some namespace conflicts.
22049
22050 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
22051 Reviewed-by: Keith Packard <keithp@keithp.com>
22052
22053commit 85cecd981191f9c3dab0fb13310d91eff643d423
22054Author: Jeremy Huddleston <jeremyhu@apple.com>
22055Date: Sat Jan 28 16:06:28 2012 -0800
22056
22057 XQuartz: Release all buttons and keys when deactivating
22058
22059 http://xquartz.macosforge.org/trac/ticket/486
22060
22061 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22062 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22063
22064commit 47b457541b33f00807fd495f5b0b24d5f143bf84
22065Author: Jeremy Huddleston <jeremyhu@apple.com>
22066Date: Sat Jan 28 23:11:44 2012 -0800
22067
22068 XQuartz: Toggle off fullscreen mode when XQuartz is hidden
22069
22070 http://xquartz.macosforge.org/trac/ticket/478
22071
22072 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22073
22074commit c2ba3f339f089c595386c4c579935c7984a2a545
22075Author: Jeremy Huddleston <jeremyhu@apple.com>
22076Date: Sun Jan 29 14:26:47 2012 -0800
22077
22078 XQuartz: Localization Updates
22079
22080 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22081
22082commit 052ca3f22eadd0aa60dd24ac7d5d76137273926f
22083Author: Keith Packard <keithp@keithp.com>
22084Date: Fri Jan 27 22:08:08 2012 -0800
22085
22086 Bump version to 1.11.99.902 (1.12 RC2)
22087
22088 Signed-off-by: Keith Packard <keithp@keithp.com>
22089
22090commit bafedb7e9bcff31e2963eeb54669b2492214fae7
22091Author: Jeremy Huddleston <jeremyhu@apple.com>
22092Date: Wed Jan 18 11:52:04 2012 -0800
22093
22094 XQuartz: Bump bundle version to 2.7.2
22095
22096 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22097
22098commit 8e78bbb2d2dc5b18f127540b63b45ba293bbdd25
22099Author: Colin Harrison <colin.harrison@virgin.net>
22100Date: Thu Jan 26 13:28:24 2012 +0000
22101
22102 hw/xwin: Fix spelling of 'Canadian' in winkeybd.h
22103
22104 The Mounties always get their typo
22105
22106 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
22107 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22108
22109commit 80c073352a276bad3722263629de3cd61df758ab
22110Author: Colin Harrison <colin.harrison@virgin.net>
22111Date: Wed Jan 4 17:16:36 2012 +0000
22112
22113 hw/xwin: Fix winEnqueueMotion() for change in miPointerSetPosition()
22114
22115 Commit 3b36fd1b49030ead44358945f62e5abe7f4609ce changed miPointerSetPosition()
22116 to take co-ordinates as doubles, not ints, so this code as it stands is now wrong
22117 (if it ever was correct in the first place :-))
22118
22119 It's unclear that we can safely promote x,y to doubles, apply miPointerSetPosition()
22120 which potentially constrains the cursor, and then convert back to ints.
22121
22122 Fortunately, this whole dance seems to be unnecessary, and we can simply remove the
22123 call to miPointerSetPosition() entirely, and just QueuePointerEvents() like any other
22124 input driver.
22125
22126 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
22127 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22128
22129commit ced9db65950e402d7ddc663225b888e8482b8c57
22130Author: Colin Harrison <colin.harrison@virgin.net>
22131Date: Wed Jan 25 01:37:16 2012 +0000
22132
22133 hw/xwin: Ignore WM_DISPLAYCHANGE messages with 0 bpp
22134
22135 Ignore WM_DISPLAYCHANGE messages which indicate bpp is changing to 0.
22136 That has no defined meaning I can find, but some graphics card drivers
22137 appear to generate it on suspend/resume or screensaver activate/deactivate.
22138
22139 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
22140 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22141
22142commit bea6fb6c7af551778cfe1c0e8412ef4ccf560808
22143Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22144Date: Sun Jan 22 19:31:51 2012 +0000
22145
22146 hw/xwin: Avoid WIN_WINDOW_PROP races during Windows window destruction
22147
22148 The WIN_WINDOW_PROP is removed during WM_DESTROY handling, so it is not neccessary to
22149 remove it in winDestroyWindowsWindow(), and doing so opens a race condition, as we may
22150 attempt to access that property in the wndproc before the WM_DESTROY has completed.
22151
22152 A specific example of that race is if a WM_KILLFOCUS occurs in the window between property
22153 removal and WM_DESTROY processing, where we will attempt to apply DeleteWindowFromAnyEvents()
22154 on an invalid (null) WindowPtr.
22155
22156 Also guard against null WindowPtr in the WM_KILLFOCUS handler
22157
22158 See http://cygwin.com/ml/cygwin-xfree/2012-01/msg00009.html
22159
22160 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22161 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22162
22163commit 9a709d5028094fcbeb7a49f47cb85e22cd772f36
22164Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22165Date: Thu Sep 29 15:05:27 2011 +0100
22166
22167 hw/xwin: Handle more motif window decoration hinting
22168
22169 Handle the MWM_DECOR_MINIMIZE, MWM_DECOR_MAXIMIZE and MWM_DECOR_MENU
22170 decoration hints in a _MOTIF_WM_HINTS window property
22171
22172 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22173 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22174
22175commit b8b0b841a0e7cfcecaf0b6a5aa67e1b4499374d9
22176Author: Oliver Schmidt <oschmidt-mailinglists@gmx.de>
22177Date: Mon Sep 5 13:32:01 2011 +0100
22178
22179 hw/xwin: Fix AltGr key sometimes firing an additional Ctrl-L key
22180
22181 I also had problems with the AltGr key. These could reliably
22182 be reproduced by holding the AltGr for some seconds (causing
22183 Windows generating auto repeat events)
22184
22185 I discovered that the mechanism in winkeybd.c function
22186 winIsFakeCtrl_L had a problem if PeekMessage cannot obtain
22187 the next Alt_R message because it is not there yet.
22188
22189 I prepared a patch that remembers the last Ctrl_L event and
22190 reacts on a later Alt_R.
22191
22192 It was also necessary to alter the order in winWindowProc() in
22193 winwndproc.c: the invocation of winIsFakeCtrl_L had to be done
22194 before discarding auto-repeated key presses, as winIsFakeCtrl_L()
22195 now has an internal state which must be updated by all key events.
22196
22197 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22198 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22199
22200commit 3d3114d55a2a323f8d49c3549a0dfdf9d4acf89d
22201Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22202Date: Thu Jun 30 14:19:01 2011 +0100
22203
22204 hw/xwin: Handle the virtual key code generated by the Fn key on IBM Lenovo laptops
22205
22206 Apparently, IBM Leonovo laptops can generate a key-press event for the Fn
22207 key, with virtual key code 0xFF and scan code extended 0x63
22208
22209 Handle this specially, rather than just passing on key code 0x63 (delete),
22210 so you don't delete what you just typed when you adjust the screen brightness,
22211 etc. :-)
22212
22213 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22214 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22215
22216commit 74af860f9a59332f6ed1ac9b3e7867fbbb3d7305
22217Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22218Date: Wed Mar 24 22:41:22 2010 +0000
22219
22220 hw/xwin: turn on -emulate3buttons if less than 3 mouse buttons are reported
22221
22222 Try to be more intelligent with default options, turn on -emulate3buttons by
22223 default if less than 3 mouse buttons are reported by Windows
22224
22225 Also, add -noemulate3buttons option so this default setting can be reversed
22226 if desired
22227
22228 Also, correctly report the number of mouse buttons windows is reporting, rather
22229 than always at least 3
22230
22231 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22232 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22233
22234commit c0f3709501a0afd3bf77e783f11d2c2e5f489d3b
22235Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22236Date: Thu Mar 24 20:19:47 2011 +0000
22237
22238 hw/xwin: In multiwindow mode, don't grab native input focus for new windows which hint they don't want it
22239
22240 In multiwindow mode, avoid grabbing the input focus for newly
22241 created windows which have InputHint FALSE
22242
22243 (this is used by e.g. glean to avoid every test window grabbing
22244 the focus)
22245
22246 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22247 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22248
22249commit 25caa8565d7d10f4c254bca5bb9efa05a77542ad
22250Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22251Date: Wed Jan 25 13:33:39 2012 +0000
22252
22253 hw/xwin: Remove some redundant OS version reporting
22254
22255 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22256 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22257
22258commit a9aca218f557c723e637287272819a7c17174e1e
22259Author: Roland Cassard <roland.cassard@gmail.com>
22260Date: Sat Oct 23 18:12:36 2010 +0100
22261
22262 hw/xwin: Don't assume we'll always have converted the clipboard selection after 2 attempts
22263
22264 Rather than knowing we have to call winProcessXEventsTimeout() for up to 2 WIN_XEVENTS_CONVERT messages, process
22265 all messages in winProcessXEventsTimeout() until either: (i) the time out expired, (ii) an error occurred, or
22266 (iii) received a WIN_XEVENTS_NOTIFY messaage indicating the data has been to put on the clipboard.
22267
22268 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22269 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22270
22271commit 75fe336b6c903133ae386f5cb8d308a0e9e2768e
22272Author: Michel Hummel <hummel.michel@gmail.com>
22273Date: Fri Aug 6 15:23:52 2010 +0100
22274
22275 hw/xwin: Mitigate a race condition in clipboard thread initialization
22276
22277 Remove the variables g_fClipboardLaunched and g_fClipboardStarted from
22278 winInitializeGlobals(), as their re-initialization is handled in the
22279 file hw/xwin/InitOutput.c.
22280
22281 Re-initializing g_fClipboardLaunched and g_fClipboardStarted during
22282 the server reset procedure can lead to the clipboard thread being
22283 launched two times and sometimes leads to a crash of the X server...
22284
22285 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22286 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22287
22288commit 95b1391fe3d3192abdfbad4140513b2112cfa02a
22289Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22290Date: Mon Apr 5 13:57:24 2010 +0100
22291
22292 hw/xwin: Chain IOError handlers to avoid longjmp across threads
22293
22294 Avoid crashes on shutdown due to the undefined behaviour of calling longjmp() on the
22295 result of setjmp() from a different thread, by chaining IOError handlers and only
22296 jumping back up to the frame for this thread
22297
22298 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22299 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22300
22301commit 0659437f5ec0e3f646373394f5f9c5461e2170f3
22302Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
22303Date: Sat Feb 20 23:40:53 2010 -0600
22304
22305 hw/xwin: Improve XWinrc loading and error recovery
22306
22307 If $HOME/.XWinrc is present but badly formed, ignore it and try
22308 system.XWinrc instead. If neither file is present or both are badly
22309 formed, provide a built-in default which gives the user the chance to
22310 load their new or fixed configuration without restarting.
22311
22312 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
22313 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22314 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22315
22316commit 47c7b6d3e626497747ae2780f259a15b8e6c846f
22317Author: Michel Hummel <hummel.michel@gmail.com>
22318Date: Sat Oct 23 18:35:57 2010 +0100
22319
22320 hw/xwin: Remove no-longer needed tricks used to prevent the clipboard client from being killed
22321
22322 Remove no-longer needed tricks used to hide the clipboard client from XDM to prevent
22323 it from being killed
22324
22325 - Delete XQuery wrapper used to hide clipboard client
22326 - Delete XDMCP mode heuristic which waits until some magic number of connections have
22327 been established before starting the clipboard
22328
22329 We still need the EstablishConnection wrapper to ensure that the clipboard client isn't
22330 the first client (causing a server restart if it disconnects)
22331
22332 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22333 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22334
22335commit c1bf3baa44fbd8af33a2b3ce045324485b85a7a7
22336Author: Michel Hummel <hummel.michel@gmail.com>
22337Date: Sun Oct 31 14:53:02 2010 +0000
22338
22339 hw/xwin: Automatically restart clipboard thread
22340
22341 Automatically restart clipboard thread on unexpected exit
22342
22343 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22344 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22345
22346commit ae981341a98f7a1ee5d6f3bc8ebde04e42042dc5
22347Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22348Date: Mon Feb 8 22:37:30 2010 +0000
22349
22350 hw/xwin: Give the X window for the clipboard integration client a name
22351
22352 This makes it a bit easier to find when staring at the output of
22353 'xwininfo -tree -root'
22354
22355 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22356 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22357
22358commit b1093241f3e28223a139e7cb5c3ae85182bdf545
22359Author: Jon TURNEY <jon.turney@dronecode.org.uk>
22360Date: Tue Oct 19 20:47:21 2010 +0100
22361
22362 Xext: Warning fix for shm.c
22363
22364 shm.c: In function 'CheckForShmSyscall':
22365 shm.c:182:5: warning: function declaration isn't a prototype [-Wstrict-prototypes]
22366
22367 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22368 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
22369 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22370
22371commit 02775efb8930291cc62fc84086c97da75b912a55
22372Author: Adam Jackson <ajax@redhat.com>
22373Date: Tue Jan 24 17:35:04 2012 -0500
22374
22375 int10: Fix unmapping of the BIOS scratch area
22376
22377 342f3eac8460fc48cfad1f1d7be939d671e6e1cd introduced a bug, 'base' is
22378 incremented before use. The old code corrected this when unmapping, so
22379 the new code should too.
22380
22381 Signed-off-by: Adam Jackson <ajax@redhat.com>
22382 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22383 Signed-off-by: Keith Packard <keithp@keithp.com>
22384
22385commit e1085a0da0b9299f48b3dc41dee5e33bf022bea5
22386Author: Chase Douglas <chase.douglas@canonical.com>
22387Date: Wed Jan 18 19:09:21 2012 -0800
22388
22389 Don't set X and Y valuators for indirect touch events
22390
22391 For expediency, it made sense to always have the X and Y axes set for
22392 direct touch device event propagation. The last X and Y values are
22393 stored internally. However, indirect device touch event propagation
22394 does not depend on the touch's X and Y values. Thus, we don't need to
22395 set the values for every indirect touch event.
22396
22397 On top of this, the previous X and Y values aren't stored for indirect
22398 touches, so without this change the axes get erroneously set to 0.
22399
22400 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
22401 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22402
22403commit 5201310559fe8708ba8278bdef77cdc1673fff71
22404Author: Daniel Stone <daniel@fooishbar.org>
22405Date: Thu Jan 19 22:40:32 2012 +1100
22406
22407 UngrabAllDevices: Don't kill clients if not told to
22408
22409 The kill_client argument to UngrabAllClients specifies if we want to
22410 kill the client holding the grab or just deactivate the grab.
22411
22412 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
22413 Reported-by: Julien Cristau <jcristau@debian.org>
22414 Reviewed-by: Cyril Brulebois <kibi@debian.org>
22415 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22416 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22417
22418commit 9b1e18f42a05a79d9ebec372a37b47442e397ca9
22419Author: Chase Douglas <chase.douglas@canonical.com>
22420Date: Wed Jan 18 18:04:14 2012 -0800
22421
22422 Only update pointer motion data for pointer emulated touch events
22423
22424 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
22425 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22426 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22427
22428commit f00e5b02f5d9928cd88961e761da91b5faff1b32
22429Author: Chase Douglas <chase.douglas@canonical.com>
22430Date: Wed Jan 18 18:04:13 2012 -0800
22431
22432 Only scale direct device touch coordinates
22433
22434 Indirect touch devices provide valuator values in pure device
22435 coordinates. They also don't need to be fixed up for screen crossings.
22436
22437 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
22438 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22439 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22440
22441commit 24dc0389da3970e83e4908039957567e55f9ea05
22442Author: Peter Hutterer <peter.hutterer@who-t.net>
22443Date: Mon Jan 16 12:53:39 2012 +1000
22444
22445 include: Fix comment typo
22446
22447 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22448
22449commit 954bb994842aa43a0f272858e65036c016b729a4
22450Author: Jeremy Huddleston <jeremyhu@apple.com>
22451Date: Mon Jan 9 00:40:10 2012 -0800
22452
22453 configure.ac: Remove MIEXT_SHADOW_LIB from XORG_LIBS
22454
22455 This is provided by dixmods/libshadow.so and is not part of the main binary.
22456
22457 This addresses a build failure on darwin due to MIEXT_SHADOW_LIB having
22458 unsatisfied dependencies (FB_LIB) in XORG_LIBS.
22459
22460 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22461
22462commit ba0f5cc1961bfcfb5e7d66ac5df19d429952631d
22463Author: Jeremy Huddleston <jeremyhu@apple.com>
22464Date: Mon Jan 9 00:14:29 2012 -0800
22465
22466 xfree86: Don't link libxorgxkb against libdix.la
22467
22468 libdix.a is already provided by XSERVER_LIBS. Including it in libxorgxkb
22469 results can result in duplicate symbols landing in the Xorg binary on some
22470 configurations (buggy glibtool on darwin).
22471
22472 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22473
22474commit 2387fb23858d645f15061bc7bcbe4654386ba116
22475Author: Jeremy Huddleston <jeremyhu@apple.com>
22476Date: Fri Jan 13 12:00:14 2012 -0800
22477
22478 sdksyms.sh: Exit on error rather than building an empty symbol table
22479
22480 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22481 Reviewed-by: Keith Packard <keithp@keithp.com>
22482
22483commit cd89482088f71ed517c2e88ed437e4752070c3f4
22484Author: Jeremy Huddleston <jeremyhu@apple.com>
22485Date: Wed Jan 11 12:17:06 2012 -0800
22486
22487 test: Fix linking issues when building unit tests without the Xorg DDX
22488
22489 This allows unit tests to build and run successfully on darwin when
22490 only the Xvfb or XQuartz DDX is built.
22491
22492 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22493 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22494 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22495
22496commit 78d39b3222a4b0b92d840137f5455d3a20dd6906
22497Author: Jeremy Huddleston <jeremyhu@apple.com>
22498Date: Sun Jan 15 02:26:43 2012 -0800
22499
22500 Revert "glx: don't leak fbconfigs"
22501
22502 This reverts commit d26fae246d7c451b4d5ffe24fdb959d4bd00b107.
22503
22504 This patch free()s memory in __glXScreenDestroy which was allocated in
22505 glxdricommon (which isn't in every DDX). That breaks abstraction and
22506 causes a crash when XQuartz quits, because it results in freeing a pointer
22507 that was never allocated.
22508
22509 The correct fix is to do this cleanup in __glXDRIscreenDestroy.
22510
22511 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22512
22513commit f9e6858d5c10be6a8439c0f18bfb2325fa0ee070
22514Author: Chase Douglas <chase.douglas@canonical.com>
22515Date: Wed Jan 11 07:38:12 2012 -0800
22516
22517 Use event time instead of CurrentTime for grab times
22518
22519 When {XI,X,}AllowEvents is called, the timestamp is compared against the
22520 grab time to ensure that the request pertains to the current grab in the
22521 server. While many clients may use CurrentTime (client-side), the
22522 timestamp of the event causing the grab is also valid.
22523
22524 This change ensures that the server's notion of the grab time is the
22525 time of the event that activated the grab rather than the time that the
22526 grab is actually activated.
22527
22528 This bug was exposed through nested touch then pointer grabs.
22529
22530 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
22531 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
22532 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22533
22534commit a6273cc85c01fc020643a68e49ca4e7a2d2ae898
22535Author: Peter Hutterer <peter.hutterer@who-t.net>
22536Date: Thu Jan 12 10:17:34 2012 +1000
22537
22538 xfree86: mention udev in the xorg.conf manpage AutoAddDevices section
22539
22540 And point out what "hotplugging" means.
22541
22542 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22543 Reviewed-by: Julien Cristau <jcristau@debian.org>
22544
22545commit a60d87ffe6d9a15fa830c8da2947c72487863c2b
22546Author: Peter Hutterer <peter.hutterer@who-t.net>
22547Date: Wed Jan 11 09:04:02 2012 +1000
22548
22549 os: prettify backtrace output
22550
22551 Changes to output:
22552 * "Backtrace:" now appears on a separate line _with_ a timestamp
22553 * A blank line is inserted after the last backtrace line
22554
22555 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22556 Reviewed-by: Julien Cristau <jcristau@debian.org>
22557
22558commit c495a839ace7fcc1f1fe414d3d3ba04f08885434
22559Author: Peter Hutterer <peter.hutterer@who-t.net>
22560Date: Wed Jan 11 09:01:05 2012 +1000
22561
22562 include: prettify BUG_WARN output
22563
22564 ErrorF output is prefixed with a timestamp, so the previous output would
22565 look like this:
22566
22567 [ 50.423] BUG: triggered 'if (dev->valuator->numAxes < 2)'
22568 BUG: getevents.c:842 in scale_to_desktop()
22569
22570 Change this to have the prefix on both lines:
22571 [ 50.423] BUG: triggered 'if (dev->valuator->numAxes < 2)'
22572 [ 50.423] BUG: getevents.c:842 in scale_to_desktop()
22573
22574 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22575 Reviewed-by: Julien Cristau <jcristau@debian.org>
22576
22577commit bbb6b8c834e0e1491ca14403b5d0840dd14380d3
22578Author: Peter Hutterer <peter.hutterer@who-t.net>
22579Date: Fri Jan 6 13:20:45 2012 +1000
22580
22581 render: don't bother with animated cursors on floating slaves (#39989)
22582
22583 X.Org Bug 39989 <http://bugs.freedesktop.org/show_bug.cgi?id=39989>
22584
22585 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22586 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
22587
22588commit 35bd77e9d0701daae87d681900d749604fc6471f
22589Author: Peter Hutterer <peter.hutterer@who-t.net>
22590Date: Mon Jan 9 15:04:10 2012 +1000
22591
22592 dix: Update pointer limits for floating devices too (#43635)
22593
22594 When the screen is restructured, the pointer limits need to be reset for
22595 floating slave devices as well, not just for master pointers. Only skip
22596 devices that don't have a cursor (attached slaves and keyboard)
22597
22598 Bug reproducer: float an absolute slave device, rotate the screen - the
22599 device is now confined to a section of the screen only.
22600
22601 X.Org Bug 43635 <http://bugs.freedesktop.org/show_bug.cgi?id=43635>
22602
22603 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22604 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
22605
22606commit 5de0c2582f9c80874b4f4a5b86d250059bfaa85b
22607Author: Keith Packard <keithp@keithp.com>
22608Date: Thu Jan 12 12:10:07 2012 -0800
22609
22610 Revert "os: Repack ConnectionOutput for LP64"
22611
22612 This reverts commit d5f724544afd2949cebfcf4f0b4510ec0c701bec.
22613
22614 ABI change pended for 1.13
22615
22616commit 11331305d4acd117b71502e45c0e86684fa9280b
22617Author: Keith Packard <keithp@keithp.com>
22618Date: Thu Jan 12 12:10:03 2012 -0800
22619
22620 Revert "dix: Repack ClientRec"
22621
22622 This reverts commit f702372822dadb1fef92cfc25086481f640147b3.
22623
22624 ABI change pended for 1.13
22625
22626commit d9eeede52f2d5ba9dd6368d988a5d2abb3b8b4e5
22627Author: Keith Packard <keithp@keithp.com>
22628Date: Thu Jan 12 12:09:59 2012 -0800
22629
22630 Revert "dix: Pull client-is-local flag up to the ClientRec"
22631
22632 This reverts commit 49d38b75c8f3276cfce33ffe6b8c4fbeb1081b96.
22633
22634 ABI change pended for 1.13
22635
22636commit 3be37375eed9eb9cfb9c42821deda4213af4057b
22637Author: Keith Packard <keithp@keithp.com>
22638Date: Thu Jan 12 12:09:55 2012 -0800
22639
22640 Revert "dix: Extend initial connection handshake for forwarding proxies"
22641
22642 This reverts commit 78fa121f4097d29458e5453c13473595df06e26e.
22643
22644 ABI change pended for 1.13
22645
22646commit 5b9f5c8a53aca03c3c73bc10bd362987621a9d72
22647Author: Keith Packard <keithp@keithp.com>
22648Date: Thu Jan 12 12:09:44 2012 -0800
22649
22650 Revert "os: Hide the Connection{In,Out}put implementation details"
22651
22652 This reverts commit 48e7a2ef574c8b38c4f8f07b45f54c8bfd02552b.
22653
22654 ABI change pended for 1.13
22655
22656commit 5867d453cebebec6b4f10409af86b603755fbf51
22657Author: Keith Packard <keithp@keithp.com>
22658Date: Thu Jan 12 12:09:34 2012 -0800
22659
22660 Revert "dix: Fix types in WindowOptRec"
22661
22662 This reverts commit a4553019a10b4e01cc06f3081db71a83338697b4.
22663
22664 ABI change pended for 1.13
22665
22666commit e722ad6c3efa57b806ca0f2dc13114bd3619a88c
22667Merge: e476af4 4e44580
22668Author: Keith Packard <keithp@keithp.com>
22669Date: Mon Jan 9 13:22:28 2012 -0800
22670
22671 Merge remote-tracking branch 'jturney/rpavlik-xwin-fixes'
22672
22673commit e476af417d83730b11054f4e5b127ab5540bb332
22674Merge: f4956fa 1460083
22675Author: Keith Packard <keithp@keithp.com>
22676Date: Mon Jan 9 13:17:37 2012 -0800
22677
22678 Merge remote-tracking branch 'whot/multitouch'
22679
22680commit f4956faab9ccf9aba6cf9603f4489f5dad19a347
22681Author: Zhigang Gong <zhigang.gong@linux.intel.com>
22682Date: Wed Jan 4 07:01:19 2012 +0000
22683
22684 mi/mibitblt: Fix an overflow bug of bit shift.
22685
22686 When depth equal to 32 and planeMask equal to 0, the overflow will
22687 occur and cause the pixmap can't be cleared. There are some test
22688 cases in XTS hit this bug, and this fix can eliminate the corresponding
22689 failures.
22690
22691 Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
22692 Reviewed-by: Adam Jackson <ajax@redhat.com>
22693 Signed-off-by: Keith Packard <keithp@keithp.com>
22694
22695commit 6d6d4cb6043905d850834946e9bfc526ed5a9ef7
22696Author: Matthieu Herrb <matthieu.herrb@laas.fr>
22697Date: Mon Jan 2 13:23:59 2012 +0000
22698
22699 Add OpenBSD support to DetermineClientCmd()
22700
22701 Uses kvm_getargv() from libkvm.
22702
22703 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
22704 Reviewed-by: Adam Jackson <ajax@redhat.com>
22705 Signed-off-by: Keith Packard <keithp@keithp.com>
22706
22707commit a55214d11916b707b7c8c65c555cc0cbb59ac503
22708Author: Adam Jackson <ajax@redhat.com>
22709Date: Tue Jan 3 06:22:24 2012 +0000
22710
22711 Always install xaa sdk headers
22712
22713 Always install XAA SDK headers so drivers still build even with
22714 --disable-xaa
22715
22716 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22717 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
22718 Signed-off-by: Keith Packard <keithp@keithp.com>
22719
22720commit dafc327f3c75205cf7e5360e9ccd71c7457f61a5
22721Author: Matthieu Herrb <matthieu.herrb@laas.fr>
22722Date: Mon Jan 2 13:11:41 2012 +0000
22723
22724 UnloadSubModule(): accept pointer value '1' and ignore it.
22725
22726 Some driver modules try to unload submodules that are now built-in.
22727
22728 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
22729 Reviewed-by: Julien Cristau <jcristau@debian.org>
22730 Reviewed-by: Adam Jackson <ajax@redhat.com>
22731 Signed-off-by: Keith Packard <keithp@keithp.com>
22732
22733commit 8db029064bcbe378061e812bb8136608e3123226
22734Author: Adam Jackson <ajax@redhat.com>
22735Date: Wed Jan 4 16:39:33 2012 +0000
22736
22737 vgahw: Fix DACDelay() macro to use the driver's vtable
22738
22739 We don't want to unconditionally use I/O routines here, since if the
22740 driver is using mmap'd VGA ports then the I/O handle won't be set up.
22741
22742 Tested-by: Jeff Chua <jeff.chua.linux@gmail.com>
22743 Signed-off-by: Adam Jackson <ajax@redhat.com>
22744 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
22745 Signed-off-by: Keith Packard <keithp@keithp.com>
22746
22747commit 63a8012947fc44ccb3d661dec88dbda14e7f3c04
22748Author: Chris Halse Rogers <christopher.halse.rogers@canonical.com>
22749Date: Thu Jan 5 01:22:40 2012 +0000
22750
22751 Revert "dix: don't return BadMatch from GetProperty (#23562)"
22752
22753 This reverts commit f04fe06ae244b851b38be824b1a80f2f8a030591.
22754
22755 dixLookupWindow no longer returns BadMatch. No other caller was checking
22756 for it, so this problem is now fixed in the utility function.
22757
22758 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
22759 Reviewed-by: Adam Jackson <ajax@redhat.com>
22760 Signed-off-by: Keith Packard <keithp@keithp.com>
22761
22762commit ef492e9797b6d4f6bbc25e86bedc24477819fde7
22763Author: Chris Halse Rogers <christopher.halse.rogers@canonical.com>
22764Date: Thu Jan 5 01:22:39 2012 +0000
22765
22766 dix: Return BadWindow rather than BadMatch from dixLookupWindow
22767
22768 dixLookupWindow uses dixLookupDrawable internally, which returns
22769 BadMatch when the XID matches a non-Window drawable. Users
22770 of dixLookupWindow don't care about this, just that it's not
22771 a valid Window.
22772
22773 This is a generalised version of the fix for X.Org Bug 23562,
22774 where GetProperty was incorrectly returning BadMatch. Auditing other
22775 window requests, all that I checked would incorrectly return BadMatch
22776 in these circumstances. An incomplete list of calls that could
22777 incorrectly return BadMatch is: ListProperties, SetSelectionOwner,
22778 {Destroy,Map,Unmap}{,Sub}Window.
22779
22780 None of the callers of dixLookupWindow, except for GetProperty, check
22781 for BadMatch
22782
22783 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
22784 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
22785 Reviewed-by: Adam Jackson <ajax@redhat.com>
22786 Signed-off-by: Keith Packard <keithp@keithp.com>
22787
22788commit ff891bbf68caefc22cabb541b6b56af086ac2280
22789Author: Arthur Taylor <art@ified.ca>
22790Date: Fri Jan 6 07:21:37 2012 +0000
22791
22792 linux: Use K_OFF VT KB mode over K_RAW if available.
22793
22794 Linux kernels since 2.6.38 (March 2011) have an VT KB mode K_OFF in
22795 which special keys (like Ctrl+C) are not interpreted and input is not
22796 buffered. Use of this mode over K_RAW removes the need for a
22797 xf86ConsoleHandler to drain the VT input buffer, removing the grief it
22798 causes when it goes wrong or is (de)initialized out-of-order. (This
22799 also saves a few needless context switches per key event.)
22800
22801 If K_OFF is not defined or not understood by the kernel, K_RAW and the
22802 previous method is used as a fall-back.
22803
22804 Reviewed-by: Adam Jackson <ajax@redhat.com>
22805 Signed-off-by: Arthur Taylor <art@ified.ca>
22806 Signed-off-by: Keith Packard <keithp@keithp.com>
22807
22808commit 0b113f7cdf5228dccd51a749a339c8669e3f20ff
22809Merge: 0b2c649 777bf90
22810Author: Keith Packard <keithp@keithp.com>
22811Date: Mon Jan 9 13:07:25 2012 -0800
22812
22813 Merge commit '777bf90abeac37087a3d0538b847742523d5acf2'
22814
22815commit 0b2c6491c5e5e1c9ba067299f3de61f5acee263b
22816Merge: 1f5587e 6b19436
22817Author: Keith Packard <keithp@keithp.com>
22818Date: Mon Jan 9 11:40:23 2012 -0800
22819
22820 Merge remote-tracking branch 'whot/for-keith'
22821
22822commit 1f5587e14406c083687203030db6a11691ac9d9f
22823Merge: a97252d 6269977
22824Author: Keith Packard <keithp@keithp.com>
22825Date: Mon Jan 9 11:37:59 2012 -0800
22826
22827 Merge remote-tracking branch 'kibi/master'
22828
22829commit a97252db24669a3e26a935024ea38b8a28d30586
22830Author: Jeremy Huddleston <jeremyhu@apple.com>
22831Date: Tue Dec 20 19:49:47 2011 -0800
22832
22833 XQuartz: GL: Buildfix for recent GLX changes
22834
22835 dispatch.h was leftover from an earlier implementation and is no longer
22836 needed, so remove it since including it causes a build failure due to
22837 conflicts between GL/gl.h and OpenGL/gl.h
22838
22839 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
22840 Reviewed-by: Jamey Sharp <jamey@minilop.net>
22841 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
22842
22843commit 6b194365364cb7a74c2b19f4a8c5e0e0404339da
22844Author: Andreas Wettstein <wettstein509@solnet.ch>
22845Date: Wed Nov 30 20:20:21 2011 +0100
22846
22847 xkb: Message actions suppress other key presses #28575
22848
22849 When a key to which a message action is mapped is held down, presses of
22850 other keys were not registered.
22851
22852 Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
22853 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
22854 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
22855
22856commit 4e44580efd4522ced644c698336d2f6ea54f3917
22857Author: Ryan Pavlik <rpavlik@iastate.edu>
22858Date: Fri Nov 4 14:29:01 2011 -0500
22859
22860 hw/xwin: Include manifest file in the dist tarball
22861
22862 Commit c02638fd added the manifest file, but didn't add it to EXTRA_DIST.
22863
22864 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22865 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22866 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22867
22868commit c763fe51b8ff18e204caab9cf97376a1b72324f0
22869Author: Ryan Pavlik <rpavlik@iastate.edu>
22870Date: Fri Nov 4 13:17:50 2011 -0500
22871
22872 hw/xwin: Fix duplicate definition of HAS_WINSOCK when building for MinGW
22873
22874 hw/xwin: Fix duplicate definition of HAS_WINSOCK when building for MinGW
22875 but still provide it if building for Win32 without autotools
22876
22877 xserver/hw/xwin/winclipboard.h:42:0: warning: "HAS_WINSOCK" redefined
22878 ../../include/xwin-config.h:11:0: note: this is the location of the previous definition
22879
22880 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22881 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22882 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22883
22884commit 3d80f202b06227f7fc03b674f5fbf809c2d1efb2
22885Author: Ryan Pavlik <rpavlik@iastate.edu>
22886Date: Fri Oct 28 09:56:19 2011 -0500
22887
22888 hw/xwin: Fix possible null ptr deref in winMWExtWMDecorateWindow()
22889
22890 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22891 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22892 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22893
22894commit c824004b4592b3d86b3514be7cab37b36642b13a
22895Author: Ryan Pavlik <rpavlik@iastate.edu>
22896Date: Fri Oct 28 09:56:11 2011 -0500
22897
22898 hw/xwin: Fix possible null ptr deref in winMWExtWMRestackFrame()
22899
22900 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22901 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22902 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22903
22904commit 3c501691a0a9fa17da4e2cc84f55010ef2a4790d
22905Author: Ryan Pavlik <rpavlik@iastate.edu>
22906Date: Fri Oct 28 09:52:34 2011 -0500
22907
22908 hw/xwin: Fix possible null ptr deref in winActivateAppPrimaryDD()
22909
22910 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22911 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22912 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22913
22914commit a492c02649de4c60ac21a7fcb6b7c730b558dca6
22915Author: Ryan Pavlik <rpavlik@iastate.edu>
22916Date: Fri Oct 28 09:49:00 2011 -0500
22917
22918 hw/xwin: Fix possible null ptr deref in winActivateAppNativeGDI()
22919
22920 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22921 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22922 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22923
22924commit b907079596bc8600d3420c189409053b0b5016f6
22925Author: Ryan Pavlik <rpavlik@iastate.edu>
22926Date: Fri Oct 28 09:46:56 2011 -0500
22927
22928 hw/xwin: Clarify an if statement mixed with ifdef in winSetEngine()
22929
22930 Use the same pattern as elsewhere so it's a bit clearer what we
22931 are checking
22932
22933 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22934 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22935 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22936
22937commit 5c35dd7be716a3464f012c92866919a3606f9beb
22938Author: Ryan Pavlik <rpavlik@iastate.edu>
22939Date: Thu Oct 27 15:49:59 2011 -0500
22940
22941 hw/xwin: Fix rrScreenSetSize function pointer mismatch
22942
22943 winrandr.c: In function ‘winRandRInit’:
22944 winrandr.c:218:31: warning: assignment from incompatible pointer type
22945
22946 Fix winRandRScreenSetSize() function signature to match RRScreenSetSizeProcPtr type,
22947 to align with commit fd9331f6 'Revert "Separate out screen size and screen pixmap
22948 sizes in RRScreenSizeSet"'
22949
22950 This is fall-out from the late revert of RANDR 1.4 in the 1.10 release cycle, it will
22951 probably need to be reverted if/when that goes back in again.
22952
22953 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22954 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22955 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22956
22957commit aa07d82908c28f4d083c0c55846a5b34f0e1ef31
22958Author: Ryan Pavlik <rpavlik@iastate.edu>
22959Date: Fri Oct 28 09:56:26 2011 -0500
22960
22961 hw/xwin: Remove an empty #if 0/#endif
22962
22963 Um... yeah
22964
22965 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22966 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22967 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22968
22969commit d459f42e64f2bafef5124a456d2ad6fa2cd92a7f
22970Author: Ryan Pavlik <rpavlik@iastate.edu>
22971Date: Fri Oct 28 09:52:37 2011 -0500
22972
22973 hw/xwin: Fix double-free in error path in winQueryRGBBitsAndMasks()
22974
22975 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22976 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22977 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22978
22979commit cc7dedd6b51a01996ddab10fe8b84e30700d1570
22980Author: Ryan Pavlik <rpavlik@iastate.edu>
22981Date: Fri Oct 28 09:46:45 2011 -0500
22982
22983 hw/xwin: Fix a memory leak in error path in winInitWM()
22984
22985 Signed-off-by: Ryan Pavlik <rpavlik@iastate.edu>
22986 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
22987 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
22988
22989commit 777bf90abeac37087a3d0538b847742523d5acf2
22990Author: Adam Jackson <ajax@redhat.com>
22991Date: Mon Jul 18 15:07:07 2011 -0400
22992
22993 xfree86: Remove the pretense of EDID v2 support
22994
22995 We don't do anything with EDID v2 blocks besides publish them on the
22996 root window. Worse, the check deleted by this patch would attempt to
22997 take a checksum of arbitrary memory if the rawData array isn't 256+
22998 bytes long (and, for the monitors mentioned, it probably is only 128).
22999
23000 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
23001 Signed-off-by: Adam Jackson <ajax@redhat.com>
23002
23003commit 8f9bdfd293ad8e45755efe8d764b4dcc2a724f51
23004Author: Adam Jackson <ajax@redhat.com>
23005Date: Tue Dec 13 21:23:12 2011 -0500
23006
23007 dix: Tune dixLookupDrawable for success
23008
23009 The vast vast vast majority of resource lookups are successful. Move some
23010 work to the error paths so we don't punish success.
23011
23012 Before:
23013 40000000 trep @ 0.0009 msec (1109091.3/sec): PutImage 10x10 square
23014 60000000 trep @ 0.0005 msec (2072652.2/sec): ShmPutImage 10x10 square
23015
23016 After:
23017 40000000 trep @ 0.0009 msec (1148346.9/sec): PutImage 10x10 square
23018 60000000 trep @ 0.0005 msec (2091666.1/sec): ShmPutImage 10x10 square
23019
23020 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
23021 Signed-off-by: Adam Jackson <ajax@redhat.com>
23022
23023commit 83a98543b58c661a22574a6f8d8f9d777c0955b8
23024Author: Adam Jackson <ajax@redhat.com>
23025Date: Tue Dec 13 20:23:40 2011 -0500
23026
23027 dix: Tune dtrace hooks around Dispatch
23028
23029 Don't call LookupMajorName if the hooks aren't active, it's quite expensive.
23030
23031 Before:
23032 40000000 trep @ 0.0009 msec (1087458.5/sec): PutImage 10x10 square
23033 60000000 trep @ 0.0005 msec (2012238.6/sec): ShmPutImage 10x10 square
23034
23035 After:
23036 40000000 trep @ 0.0009 msec (1109091.3/sec): PutImage 10x10 square
23037 60000000 trep @ 0.0005 msec (2072652.2/sec): ShmPutImage 10x10 square
23038
23039 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
23040 Signed-off-by: Adam Jackson <ajax@redhat.com>
23041
23042commit 11977fab546da35d579ebe285e3c26864007805e
23043Author: Adam Jackson <ajax@redhat.com>
23044Date: Tue Dec 13 21:00:05 2011 -0500
23045
23046 xace: ricer tuning for XaceHook
23047
23048 gcc doesn't want to hoist the check for XaceHooks[hook] != NULL above the
23049 varargs code for some reason, so do it ourselves.
23050
23051 Before:
23052 40000000 trep @ 0.0010 msec (1050420.2/sec): PutImage 10x10 square
23053 60000000 trep @ 0.0005 msec (1921147.6/sec): ShmPutImage 10x10 square
23054
23055 After:
23056 40000000 trep @ 0.0009 msec (1087458.5/sec): PutImage 10x10 square
23057 60000000 trep @ 0.0005 msec (2012238.6/sec): ShmPutImage 10x10 square
23058
23059 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
23060 Signed-off-by: Adam Jackson <ajax@redhat.com>
23061
23062commit a4553019a10b4e01cc06f3081db71a83338697b4
23063Author: Adam Jackson <ajax@redhat.com>
23064Date: Fri Jun 17 14:30:06 2011 -0400
23065
23066 dix: Fix types in WindowOptRec
23067
23068 No reason for these to be 64 bits on LP64.
23069
23070 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
23071 Signed-off-by: Adam Jackson <ajax@redhat.com>
23072
23073commit c44ef2e1ff7bad168c348da63477b4636b3054fd
23074Author: Adam Jackson <ajax@redhat.com>
23075Date: Fri Jun 17 14:17:09 2011 -0400
23076
23077 os: Minor header cleanup
23078
23079 Move some constants near their only users, and remove some
23080 getdtablesize() logic that's second-guessing configure.
23081
23082 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
23083 Signed-off-by: Adam Jackson <ajax@redhat.com>
23084
23085commit 48e7a2ef574c8b38c4f8f07b45f54c8bfd02552b
23086Author: Adam Jackson <ajax@redhat.com>
23087Date: Fri Jun 17 14:03:01 2011 -0400
23088
23089 os: Hide the Connection{In,Out}put implementation details
23090
23091 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
23092 Signed-off-by: Adam Jackson <ajax@redhat.com>
23093
23094commit 78fa121f4097d29458e5453c13473595df06e26e
23095Author: Adam Jackson <ajax@redhat.com>
23096Date: Fri Jun 17 13:43:38 2011 -0400
23097
23098 dix: Extend initial connection handshake for forwarding proxies
23099
23100 Forwarding proxies like sshd will appear to be local, even though they
23101 aren't really. This leads to weird behaviour for extensions that truly
23102 require running under the same OS services as the client, like MIT-SHM
23103 and DRI2.
23104
23105 Add two new legal values for the initial connection's byteOrder field,
23106 'r' and 'R'. These act like 'l' and 'B' respectively, but have the side
23107 effect of forcing the client to be treated as non-local. Forwarding
23108 proxies should attempt to munge the first packet of the connection
23109 accordingly; older servers will reject connections thusly munged, so the
23110 proxy should fall back to passthrough if the munged connection attempt
23111 fails.
23112
23113 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
23114 Signed-off-by: Adam Jackson <ajax@redhat.com>
23115
23116commit 49d38b75c8f3276cfce33ffe6b8c4fbeb1081b96
23117Author: Adam Jackson <ajax@redhat.com>
23118Date: Fri Jun 17 13:22:41 2011 -0400
23119
23120 dix: Pull client-is-local flag up to the ClientRec
23121
23122 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
23123 Signed-off-by: Adam Jackson <ajax@redhat.com>
23124
23125commit f702372822dadb1fef92cfc25086481f640147b3
23126Author: Adam Jackson <ajax@redhat.com>
23127Date: Fri Jun 17 12:58:25 2011 -0400
23128
23129 dix: Repack ClientRec
23130
23131 sizeof(ClientRec) ILP32 LP64
23132 before 120 184
23133 after 104 136
23134
23135 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
23136 Signed-off-by: Adam Jackson <ajax@redhat.com>
23137
23138commit d5f724544afd2949cebfcf4f0b4510ec0c701bec
23139Author: Adam Jackson <ajax@redhat.com>
23140Date: Thu Jun 16 17:40:24 2011 -0400
23141
23142 os: Repack ConnectionOutput for LP64
23143
23144 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
23145 Signed-off-by: Adam Jackson <ajax@redhat.com>
23146
23147commit ed8f3c4bd17bddf1369d050ea8e63b9451d887ce
23148Author: Peter Hutterer <peter.hutterer@who-t.net>
23149Date: Fri Jan 6 10:10:16 2012 +1000
23150
23151 Xi: change XIChangeDeviceProperty from const pointer to const void *
23152
23153 According to Daniel Kurtz, a typedef void *pointer is a atomic type. So a
23154 'const pointer' is equivalent to 'void* const' instead of the intended
23155 'const void*'.
23156
23157 This technically changes the ABI, but we don't bump it for this.
23158
23159 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23160 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23161
23162commit 146008358e9dd9b178bd5305bcc1680ca16c328d
23163Author: Chase Douglas <chase.douglas@ubuntu.com>
23164Date: Thu Jan 5 16:43:40 2012 -0800
23165
23166 Remove last listener on touch reject
23167
23168 The current code short-circuits around the block that removes the
23169 rejecting listener if it is the only listener left. It also does not
23170 delete the touchpoint record if the touch has not physically ended.
23171
23172 This change ensures the listener is removed under these circumstances.
23173
23174 Signed-off-by: Chase Douglas <chase.douglas@ubuntu.com>
23175 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
23176 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23177
23178commit e30c3c5c59abba9713cf3be532a46220aa9aeeca
23179Author: Chase Douglas <chase.douglas@canonical.com>
23180Date: Wed Jan 4 15:50:40 2012 -0800
23181
23182 Don't end touchpoint if owning client hasn't accepted/rejected
23183
23184 A touchpoint is ended when no further processing will take place for it.
23185 This includes the situation where there is only one grabbing client, and
23186 the client receives a touch end before it has accepted/rejected the
23187 touchpoint.
23188
23189 This change ensures that a delivered touch end event is converted into a
23190 touch update event under the above scenario. If the event is left as a
23191 touch end event, the touchpoint will be ended in ProcessTouchEvent().
23192
23193 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
23194 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
23195 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23196
23197commit 075a473e538efb771497b762101694b6c0c45c2b
23198Author: Carlos Garnacho <carlosg@gnome.org>
23199Date: Tue Jan 3 02:43:40 2012 +0100
23200
23201 dix: Set XITouchEmulatingPointer on events from the touch sequence emulating pointer events
23202
23203 The internal flag is kept around, merely translated to XITouchEmulatingPointer
23204 when creating the XI2 events that will be delivered to the client.
23205
23206 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
23207 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
23208 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23209 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23210
23211commit e8db10a033491059ddf2b4777c9fafed14c403e8
23212Author: Peter Hutterer <peter.hutterer@who-t.net>
23213Date: Fri Jan 6 13:49:04 2012 +1000
23214
23215 Require inputproto 2.1.99.5
23216
23217 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23218
23219commit 223ba8b46eacbc8e573bc5136a3d6677f3e39099
23220Author: Peter Hutterer <peter.hutterer@who-t.net>
23221Date: Thu Jan 5 15:55:04 2012 +1000
23222
23223 dix: fix wrong condition checking for attached slave (#44003)
23224
23225 We need to update the master if the device is not a master _and_ it is not
23226 floating.
23227
23228 X.Org Bug 44003 <http://bugs.freedesktop.org/show_bug.cgi?id=44003>
23229
23230 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23231 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23232
23233commit 3ce102c362cadcd7087bdcf48440d9498eaf77d0
23234Author: Peter Hutterer <peter.hutterer@who-t.net>
23235Date: Thu Jan 5 07:08:01 2012 -0500
23236
23237 dix: on PointerRootWin send a FocusIn to the sprite window too
23238
23239 XTS XSetDeviceFocus-7
23240
23241 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23242 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23243
23244commit a125aabda3a5cf27aa98cb61f16e49280b66f451
23245Author: Peter Hutterer <peter.hutterer@who-t.net>
23246Date: Thu Jan 5 07:02:51 2012 -0500
23247
23248 dix: send focus events to the immediate parent (#44079)
23249
23250 For a transition from windows A to B, A->parent did not receive an event.
23251 DeviceFocusOutEvents sends to windows ]from, to[, so start with the actual
23252 window, not it's parent.
23253
23254 X.Org Bug 44079 <http://bugs.freedesktop.org/show_bug.cgi?id=44079>
23255
23256 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23257 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23258 Reviewed-by: Keith Packard <keithp@keithp.com>
23259
23260commit 75953ccb9e9e5f1dbbaa44c714716a0049e22aae
23261Author: Peter Hutterer <peter.hutterer@who-t.net>
23262Date: Tue Jan 3 10:45:45 2012 +1000
23263
23264 xfree86: split warning about missing identifier or input driver
23265
23266 Check for identifier first and bail if it's missing (also remove the current
23267 identifier check after we've already bailed due to missing identifiers)
23268
23269 If a driver is missing, warn but also say that we may have added this device
23270 already. I see too many bugreports with incorrectly shortened log files.
23271
23272 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23273 Reviewed-by: Cyril Brulebois <kibi@debian.org>
23274
23275commit 6269977c91071e0ea16ca5b4b8e15fd6db0b6fcf
23276Author: Julien Cristau <jcristau@debian.org>
23277Date: Fri Dec 30 20:41:25 2011 +0100
23278
23279 os: don't ignore failure from dladdr
23280
23281 If dladdr returns 0, don't go and use the returned Dl_info, it may
23282 contain garbage.
23283
23284 X.Org bug#44315 <https://bugs.freedesktop.org/show_bug.cgi?id=44315>
23285
23286 Reported-and-tested-by: Cyril Brulebois <kibi@debian.org>
23287 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
23288 Reviewed-by: Cyril Brulebois <kibi@debian.org>
23289 Signed-off-by: Julien Cristau <jcristau@debian.org>
23290
23291commit 644efb43e0639af648dd46d9729c2cabd9511238
23292Author: Cyril Brulebois <kibi@debian.org>
23293Date: Fri Dec 30 02:34:36 2011 +0000
23294
23295 linux/ia64: Fix regression after domain I/O support code removal.
23296
23297 Side effect of aa0bfb0f133481c57762012e8e30c05ffa151423:
23298 | CCLD Xorg
23299 | sdksyms.o:(.data.rel+0x27d8): undefined reference to `outl'
23300 | collect2: ld returned 1 exit status
23301
23302 Since the linux/ia64 domain I/O support code got removed in that
23303 commit, there's no reason to keep on declaring those functions
23304 (inb, inl, inw, outb, outl, outw).
23305
23306 Bugzilla: https://bugs.freedesktop.org/43985
23307
23308 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
23309 Signed-off-by: Cyril Brulebois <kibi@debian.org>
23310
23311commit 98cde254acb9b98337ddecf64c138d38c14ec2bf
23312Author: Keith Packard <keithp@keithp.com>
23313Date: Tue Dec 27 13:42:21 2011 -0800
23314
23315 Bump version to 1.11.99.901 (1.12 RC1)
23316
23317 Signed-off-by: Keith Packard <keithp@keithp.com>
23318
23319commit 5037c9af78da6652189de7202e70e1b833395af5
23320Author: Keith Packard <keithp@keithp.com>
23321Date: Tue Dec 27 14:02:01 2011 -0800
23322
23323 glx/glapioffsets.h is no longer part of the build, remove it
23324
23325 Signed-off-by: Keith Packard <keithp@keithp.com>
23326
23327commit 8dedf9831bd80514d800f0085213296a3726dba7
23328Merge: cfc4c3d cf96183
23329Author: Keith Packard <keithp@keithp.com>
23330Date: Tue Dec 27 13:13:48 2011 -0800
23331
23332 Merge remote-tracking branch 'kibi/master'
23333
23334commit cfc4c3d7fa8bd4da4c08b2ab8e6f85435f75353a
23335Author: Alan Coopersmith <alan.coopersmith@oracle.com>
23336Date: Sat Dec 24 10:00:56 2011 -0800
23337
23338 Add Solaris support to DetermineClientCmd
23339
23340 Uses /proc/pid/psinfo to read command & partial arguments.
23341
23342 Moves cmdsize & argsize variables into non-Solaris #else clause
23343 to avoid unused variable warnings.
23344
23345 Fixes format mismatch errors when building with DEBUG defined on
23346 a 64-bit platform (where Mask is defined as CARD32).
23347
23348 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
23349 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
23350 Signed-off-by: Keith Packard <keithp@keithp.com>
23351
23352commit a28ff2cf92c2b35e02eacca21af929afabbf6b83
23353Author: Chase Douglas <chase.douglas@canonical.com>
23354Date: Thu Dec 22 12:00:37 2011 -0800
23355
23356 test/xi2: Really fix infinite loop in test_convert_XITouchOwnershipEvent
23357
23358 long i;
23359 for (i = 1; ; i <<= 1)
23360 if (i == (1 << 31))
23361 break;
23362
23363 (1 << 31) is compiled as an int, and thus is equal to -2147483648. We
23364 are trying to compare it against a long, which on 64-bit machines is
23365 2147483648. This results in an infinite loop.
23366
23367 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
23368 Signed-off-by: Keith Packard <keithp@keithp.com>
23369
23370commit e7df42ab68e30588a5e32ed543b0711821daf009
23371Author: Keith Packard <keithp@keithp.com>
23372Date: Thu Dec 22 09:35:51 2011 -0800
23373
23374 test/xi2: Fix infinite loop in test_convert_XITouchOwnershipEvent
23375
23376 The touchid test was using a loop like:
23377
23378 for(i = 1; i < 0xffffffff; i <<= 1)
23379
23380 When 'i' is a 32-bit variable, this infinite loops as it goes from
23381 0x80000000 to 0. 'i' is declared as 'long', which is 32-bit in 32-bit mode.
23382
23383 Signed-off-by: Keith Packard <keithp@keithp.com>
23384
23385commit cf96183122a85e3376e560f0360b38b5676e57a9
23386Author: Cyril Brulebois <kibi@debian.org>
23387Date: Wed Nov 2 12:41:20 2011 +0100
23388
23389 xorg.conf.man: Fix bad whatis entry.
23390
23391 Debian's QA tool “lintian” reported a bad whatis entry for the
23392 xorg.conf(.d) manpages.
23393
23394 It comes with the following pointers:
23395 For manual pages that document multiple programs, functions, files, or
23396 other things, the part before "\-" should list each separated by a
23397 comma and a space. […]
23398
23399 Refer to the lexgrog(1) manual page, the groff_man(7) manual page, and
23400 the groff_mdoc(7) manual page for details.
23401
23402 Indeed, the current situation is:
23403 $ whatis xorg.conf; whatis xorg.conf.d
23404 xorg.conf (5) - (unknown subject)
23405 xorg.conf.d (5) - (unknown subject)
23406
23407 With this patch:
23408 xorg.conf (5) - configuration files for Xorg X server
23409 xorg.conf.d (5) - configuration files for Xorg X server
23410
23411 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
23412 Signed-off-by: Cyril Brulebois <kibi@debian.org>
23413
23414commit f75bdf7fbe757f4603e39139acc3c90538a45e15
23415Author: Peter Hutterer <peter.hutterer@who-t.net>
23416Date: Thu Dec 22 10:00:15 2011 +1000
23417
23418 test: fix grab mode value tests for new XIGrabModeTouch
23419
23420 This test checks that last-valid-mode + 1 returns a BadValue. With the
23421 addition of XIGrabModeTouch, that value has changed - sync it up again.
23422
23423 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23424 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23425
23426commit e395efc25f9fd60377f32c2bf1a9153d70908236
23427Merge: 2bb282c fcda98c
23428Author: Peter Hutterer <peter.hutterer@who-t.net>
23429Date: Thu Dec 22 09:29:59 2011 +1000
23430
23431 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into multitouch
23432
23433 Conflicts:
23434 configure.ac
23435 dix/inpututils.c
23436
23437 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23438
23439commit 2bb282cd45cb8149b39d72397ef5bbcfebca11d2
23440Author: Peter Hutterer <peter.hutterer@who-t.net>
23441Date: Wed Dec 21 14:45:27 2011 +1000
23442
23443 Xi: only activate the device grab if we don't already have one
23444
23445 If the device is already grabbed, don't activate the passive grab, it screws
23446 with our event masks. Just deliver to the grabbing client instead.
23447
23448 Reported-by: Carlos Garnacho <carlosg@gnome.org>
23449 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23450 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23451
23452commit 475ef5cc8339fac3696fe654ef683f92f0b0cc2c
23453Author: Carlos Garnacho <carlosg@gnome.org>
23454Date: Wed Dec 21 01:08:40 2011 +0100
23455
23456 dix: fix xi2 mask/filter when delivering grabbed events
23457
23458 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
23459 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
23460 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23461
23462commit 4551510f752624592f9201f73db89ac1325a234a
23463Author: Peter Hutterer <peter.hutterer@who-t.net>
23464Date: Mon Dec 19 09:08:05 2011 +1000
23465
23466 test: add a basic test for ownership event values
23467
23468 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23469
23470commit 0c661dc478b4b436a3e61066d31bc510c7a08456
23471Author: Peter Hutterer <peter.hutterer@who-t.net>
23472Date: Mon Dec 19 08:31:37 2011 +1000
23473
23474 test: drop printfs from protocol-eventconvert.c
23475
23476 The test outputs are noisy enough, no need having these here too.
23477
23478 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23479
23480commit 8080d785b2c1e5791312ed0975efd1b8aea58d98
23481Author: Peter Hutterer <peter.hutterer@who-t.net>
23482Date: Wed Dec 21 11:11:48 2011 +1000
23483
23484 dix: remove requirement for client_id be the first element
23485
23486 Leftover code from an earlier version of GetTouchEvents.
23487
23488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23489 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23490
23491commit 552de314bf0c8e4de5d002bdd1c12388c2f42283
23492Author: Peter Hutterer <peter.hutterer@who-t.net>
23493Date: Mon Dec 19 14:52:41 2011 +1000
23494
23495 dix: don't retrieve the syncEvents tail on an empty list
23496
23497 An empty list points to itself but syncEvents has the list head only and is
23498 of a different format than the elements. Thus, casting it to a QdEventPtr
23499 gives us garbage.
23500
23501 Segfaults with XTS test case Xlib13/XGrabKeyboard
23502
23503 Introduced in 7af23259d88f4c28ed21140f82cc03b3724c06bb.
23504
23505 Reported-by: Aaron Plattner <aplattner@nvidia.com>
23506 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23507 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23508
23509commit 76c11e0872dedcd360cbe02cf62bb9de3b212957
23510Author: Peter Hutterer <peter.hutterer@who-t.net>
23511Date: Fri Dec 16 12:41:08 2011 +1000
23512
23513 dix: button state must show the logical buttons, not physical buttons
23514
23515 If the device is mapped 3 2 1, a click on physical button 1 sends a button 3
23516 press, but the state was set for button 1. Fix this, the state must be set
23517 for that button's logical mapping.
23518
23519 https://bugzilla.gnome.org/show_bug.cgi?id=655928
23520
23521 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23522 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23523
23524commit 8a32c39c00789c16006de61112627722ce548be1
23525Author: Peter Hutterer <peter.hutterer@who-t.net>
23526Date: Thu Dec 15 07:59:06 2011 +1000
23527
23528 mi: handle screen switching on pointer emulated touch events
23529
23530 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23531 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23532
23533commit 48547068b115b1f5e5f46a70110454a2175fb9a4
23534Author: Peter Hutterer <peter.hutterer@who-t.net>
23535Date: Thu Dec 15 07:57:47 2011 +1000
23536
23537 mi: split move to new screen logic to a helper function
23538
23539 No functional changes.
23540
23541 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23542 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23543
23544commit 3b1e2035cc4740711360c845cfcdff07f7b60558
23545Author: Daniel Stone <daniel@fooishbar.org>
23546Date: Thu Dec 15 07:52:28 2011 +1000
23547
23548 dix: Remove touch grabs if the grab disappears
23549
23550 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23551 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23552
23553commit cd3de8324e8908955a2e4be3000c8ffee8684c68
23554Author: Peter Hutterer <peter.hutterer@who-t.net>
23555Date: Thu Dec 15 07:48:49 2011 +1000
23556
23557 dix: hook up passive grabs and pointer emulated passive grabs
23558
23559 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23560 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23561
23562commit 5b169cb695bd450d7f64e3800f00c9237ee67f96
23563Author: Carlos Garnacho <carlosg@gnome.org>
23564Date: Tue Dec 13 15:41:23 2011 +0100
23565
23566 Xi: assign correct grab_mode/other_device_mode in XI2 passive grabs
23567
23568 CreateGrab() expects the keyboard mode to be stored in grab_mode, and the
23569 pointer mode in other_device_mode, so respect this in passive XI2 grabs,
23570 and switch modes if needed.
23571
23572 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
23573 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23574 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23575
23576commit 4469430b109fa2da7ba3d2fadf66eca78b7156a0
23577Author: Peter Hutterer <peter.hutterer@who-t.net>
23578Date: Thu Dec 15 07:42:12 2011 +1000
23579
23580 Xi: add the hooks for passive touch grabs
23581
23582 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
23583 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23584 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23585
23586commit bb0af002dc9ad5b464dc6793aedb6d1ff65d001d
23587Author: Peter Hutterer <peter.hutterer@who-t.net>
23588Date: Thu Dec 15 07:21:38 2011 +1000
23589
23590 Xi: handle grab accept/reject requests
23591
23592 Consists mostly of generating an ownership event and processing it.
23593
23594 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
23595 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23596 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23597
23598commit 209b3d613a7bed126c81daedbad6461e4391e9e6
23599Author: Peter Hutterer <peter.hutterer@who-t.net>
23600Date: Tue Dec 20 16:34:43 2011 -0800
23601
23602 Xi: hook up touch ownership handling
23603
23604 Ownership changes don't get processed directly when they happen, instead the
23605 DIX submits an ownership event which goes through ProcessTouchEvents and
23606 ProcessTouchOwnershipEvents.
23607 Then on the required events are generated and sent to clients.
23608
23609 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
23610 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23611 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23612 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23613
23614commit 1894468c2b9788efa70ef4fb5fc911711402f3a7
23615Author: Peter Hutterer <peter.hutterer@who-t.net>
23616Date: Tue Dec 20 16:33:33 2011 -0800
23617
23618 Xi: hook up pointer emulation for emulating touches
23619
23620 Includes a hack for implicit grab activation, because integrating this
23621 requires a larger rewrite and I'm not sleeping enough as it is.
23622 Right now, we deliver the event and check before/after if there is an
23623 implicit grab on. If one activated, then store the event in the grab and
23624 switch the listener type to a grab listener.
23625
23626 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23627 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23628 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23629
23630commit 753c4dca1e42035e29fb21458a746e484ed93ca3
23631Author: Peter Hutterer <peter.hutterer@who-t.net>
23632Date: Wed Dec 14 17:04:38 2011 +1000
23633
23634 dix: add touch to pointer event conversion helper
23635
23636 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23637 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23638
23639commit c3843f5498a849920ab68cca3557411ad8ab3454
23640Author: Peter Hutterer <peter.hutterer@who-t.net>
23641Date: Wed Dec 14 17:02:30 2011 +1000
23642
23643 dix: add real_event parameter to ActivatePassiveGrab
23644
23645 For touch events with pointer emulation, the event that triggers the grab
23646 (the pointer event) is not the same as the actual event (the touch event).
23647 For replaying, we need to store the real event then.
23648
23649 No effective changes at this point, for the current caller event and
23650 real_event are identical.
23651
23652 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23653 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23654
23655commit e389a19a730055d4550ff708da83c78099a516fd
23656Author: Peter Hutterer <peter.hutterer@who-t.net>
23657Date: Tue Dec 20 16:32:52 2011 -0800
23658
23659 Process and deliver touch events
23660
23661 Does not include pointer emulation handling.
23662 Does include partial ownership handling but not the actual processing of
23663 ownership events.
23664
23665 Note: this commit is a retroactive commit extracted from a series of ~50
23666 commits and may thus appear a bit more complicated than what you'd write out
23667 from scratch.
23668
23669 Pointer processing tree is roughly:
23670 - ProcessOtherEvents
23671 - ProcessTouchEvents
23672 - DeliverTouchEvents
23673 - DeliverTouchBeginEvent|DeliverTouchEndEvent|...
23674 - DeliverOneTouchEvent
23675
23676 Also hooks up the event history playing to the right function now.
23677
23678 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
23679 Co-authored-by: Chase Douglas <chase.douglas@canonical.com>
23680 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23681 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23682 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23683
23684commit 634b0da9a83076d0e9e0fc44dc5dc77b0c368bc1
23685Author: Peter Hutterer <peter.hutterer@who-t.net>
23686Date: Thu Dec 15 07:53:19 2011 +1000
23687
23688 dix: set core and button state based on the touch state
23689
23690 If a pointer-emulating touch caused a button to be logically down, set that
23691 state in the input events.
23692
23693 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23694 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23695
23696commit dbfd7b37a0ba21899d8ebb7e0b324301bd466c49
23697Author: Peter Hutterer <peter.hutterer@who-t.net>
23698Date: Wed Dec 14 16:33:05 2011 +1000
23699
23700 Xi: make UpdateDeviceState aware of touch events
23701
23702 Update the logical button state for pointer-emulating events. Button state
23703 must be kept separate from the ButtonClassRec to avoid clearing the button
23704 state on a touch end if there is a physical button still down.
23705
23706 And obviously don't change the button state if we're currently replaying the
23707 event history for some client.
23708
23709 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23710 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23711
23712commit 593c3e2eb3da5c5fb957b68c8025dfdbe1139639
23713Author: Peter Hutterer <peter.hutterer@who-t.net>
23714Date: Wed Dec 14 16:30:05 2011 +1000
23715
23716 dix: add TouchResourceIsOwner helper function
23717
23718 The first listener in the sequence is the owner of the touch sequence.
23719
23720 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23721 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23722
23723commit 5d0260cc986e8c1f0907870b851d47fd73bb1b7d
23724Author: Peter Hutterer <peter.hutterer@who-t.net>
23725Date: Wed Dec 14 16:24:17 2011 +1000
23726
23727 dix: add helper functions for adding/removing touch listeners
23728
23729 The DIX will call TouchSetupListeners once for a new touch. After that
23730 the listener list remains static, with listeners only dropping out when they
23731 either reject the grab or disappear.
23732
23733 Exception: if grabs activate they are prefixed to the listeners.
23734
23735 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23736 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23737
23738commit 6a68f97ce6704dd2ff2a50f91acd2da72c9995bd
23739Author: Chase Douglas <chase.douglas@canonical.com>
23740Date: Wed Dec 14 16:26:04 2011 +1000
23741
23742 dix: add TouchGetPointerEventType helper function
23743
23744 Returns the respective pointer event type for a given touch event type.
23745
23746 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23747 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23748
23749commit 8e58ce73c8b79cb86e890369d0216cdd850c52a5
23750Author: Daniel Stone <daniel@fooishbar.org>
23751Date: Wed Dec 14 16:53:04 2011 +1000
23752
23753 dix: when a window disappears, remove it from the touch sprite trace
23754
23755 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23756 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23757
23758commit c18a173cf5f53778356c5e8f25ab132956f8c27d
23759Author: Daniel Stone <daniel@fooishbar.org>
23760Date: Wed Dec 14 16:03:01 2011 +1000
23761
23762 dix: add helper functions to build up/verify the sprite trace
23763
23764 Touch events' sprite trace stays the same for the duration of the touch
23765 sequence.
23766
23767 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23768 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23769 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23770
23771commit 40475261eacabb79a824db1138407a9d0f9fe37f
23772Author: Peter Hutterer <peter.hutterer@who-t.net>
23773Date: Wed Dec 14 15:56:05 2011 +1000
23774
23775 dix: add touch event history helper functions
23776
23777 If touch client has not registered for ownership events and a grab above
23778 that client is rejected, the client needs to receive the complete event
23779 history.
23780
23781 The history currently doesn't really do fancy overflow handling. We assume
23782 that the first TOUCH_HISTORY_SIZE events are the important ones and anything
23783 after that is dropped. If that is a problem, fix the client that takes > 100
23784 event to decide whether to accept or reject.
23785
23786 Events marked with TOUCH_CLIENT_ID or TOUCH_REPLAYING must not be stored in
23787 the history, they are events created by the DIX to comply with the protocol.
23788 Any such event should already be in the history anyway.
23789
23790 A fixme in this patch: we don't have a function to actually deliver the
23791 event yet.
23792
23793 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23794 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23795
23796commit b274c472a5ad700aea5fe651d677e8cb34ffc976
23797Author: Peter Hutterer <peter.hutterer@who-t.net>
23798Date: Wed Oct 26 09:08:17 2011 +1000
23799
23800 Support XI 2.2 officially
23801
23802 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23803 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23804
23805commit d2af968cb65873780a6e61342d5d3c23b9654e2a
23806Author: Daniel Stone <daniel@fooishbar.org>
23807Date: Wed Dec 14 15:45:19 2011 +1000
23808
23809 Xi: allow selecting for touch events
23810
23811 Selecting for any of XI_TouchBegin/Update/End/Ownership requires the three
23812 bits for begin/update/end to be set.
23813
23814 Only one client at a time may select for XI_TouchBegin event
23815
23816 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23817 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23818
23819commit f3df3ad668fcd417ffb5afa3bea79a73a348bc1a
23820Author: Peter Hutterer <peter.hutterer@who-t.net>
23821Date: Wed Dec 14 16:45:04 2011 +1000
23822
23823 dix: handle DIX-submitted touch events
23824
23825 The DIX needs to submit touch events for e.g. TouchEnd after an
23826 acceptance/rejection. These have the TOUCH_CLIENT_ID flag set.
23827
23828 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23829 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23830
23831commit fcda98c48610fd507ca0b89c6006a5497d9dc1c9
23832Author: Peter Hutterer <peter.hutterer@who-t.net>
23833Date: Fri Dec 16 12:41:08 2011 +1000
23834
23835 dix: button state must show the logical buttons, not physical buttons
23836
23837 If the device is mapped 3 2 1, a click on physical button 1 sends a button 3
23838 press, but the state was set for button 1. Fix this, the state must be set
23839 for that button's logical mapping.
23840
23841 https://bugzilla.gnome.org/show_bug.cgi?id=655928
23842
23843 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23844 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23845
23846commit 47da6b6b2c2e9245422655f94770146fb8cd430b
23847Author: Peter Hutterer <peter.hutterer@who-t.net>
23848Date: Mon Dec 19 14:52:41 2011 +1000
23849
23850 dix: don't retrieve the syncEvents tail on an empty list
23851
23852 An empty list points to itself but syncEvents has the list head only and is
23853 of a different format than the elements. Thus, casting it to a QdEventPtr
23854 gives us garbage.
23855
23856 Segfaults with XTS test case Xlib13/XGrabKeyboard
23857
23858 Introduced in 7af23259d88f4c28ed21140f82cc03b3724c06bb.
23859
23860 Reported-by: Aaron Plattner <aplattner@nvidia.com>
23861 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23862 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
23863
23864commit 2d34b34ed7547b829979756ccfb2cff79368d0a4
23865Merge: e4b4d83 1deede3
23866Author: Keith Packard <keithp@keithp.com>
23867Date: Tue Dec 20 00:23:33 2011 -0800
23868
23869 Merge remote-tracking branch 'jeremyhu/master'
23870
23871commit e4b4d83fad37bb737e25f7226dbcd15b892f9528
23872Merge: 4df65d2 d26cb70
23873Author: Keith Packard <keithp@keithp.com>
23874Date: Mon Dec 19 22:36:31 2011 -0800
23875
23876 Merge remote-tracking branch 'idr/glx-fixes'
23877
23878commit 4df65d247b56c6b1b06e17a61236d6637312cdbb
23879Author: Ville Syrjälä <syrjala@sci.fi>
23880Date: Sun Dec 18 18:29:41 2011 +0200
23881
23882 dri2: Invalidate window pixmaps
23883
23884 While a redirected window is flipped, its pixmap may still be used as
23885 and EGL image and should also get invalidated. When sending invalidate
23886 events for a window, also send the events for its pixmap.
23887
23888 Signed-off-by: Ville Syrjälä <syrjala@sci.fi>
23889 Reviewed-by: Keith Packard <keithp@keithp.com>
23890 Signed-off-by: Keith Packard <keithp@keithp.com>
23891
23892commit e8fd23fad046877ca0ee08c1e147557756debc8e
23893Author: Keith Packard <keithp@keithp.com>
23894Date: Sun Dec 18 18:29:40 2011 +0200
23895
23896 dri2: Invalidate DRI2 buffers for all windows with the same pixmap on swap
23897
23898 Without this, when a compositing manager unredirects a fullscreen window which
23899 uses DRI2 and page flipping, the DRI2 buffer information for the compositing
23900 manager's output window (typically the Composite Overlay Window or root window)
23901 may become stale, resulting in all kinds of hilarity.
23902
23903 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=35452 .
23904
23905 [Original patch by Michel Dänzer <michel@daenzer.net>]
23906 [Tree walk optimized version by Keith Packard <keithp@keithp.com>]
23907
23908 Signed-off-by: Ville Syrjälä <syrjala@sci.fi>
23909 Signed-off-by: Keith Packard <keithp@keithp.com>
23910
23911commit 6f916ffec7767eeab62132eb6575043969104c81
23912Author: Michel Dänzer <daenzer@vmware.com>
23913Date: Sun Dec 18 18:29:39 2011 +0200
23914
23915 dri2: Always re-generate front buffer information when asked for it.
23916
23917 Otherwise we might keep stale cached information, e.g. after the driver
23918 performed page flipping.
23919
23920 This is part of the fix for
23921 https://bugs.freedesktop.org/show_bug.cgi?id=35452 .
23922
23923 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
23924 Reviewed-by: Ville Syrjälä <ville.syrjala@nokia.com>
23925 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
23926 Signed-off-by: Keith Packard <keithp@keithp.com>
23927
23928commit 603fcb3abf155e91fd3250f4a7aa16f289506e8e
23929Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
23930Date: Sun Dec 18 18:29:38 2011 +0200
23931
23932 dri2: Initialize needInvalidate member of DRI2Drawable.
23933
23934 If the client is not behaving correctly and swaps buffers before
23935 getting them, Valgrind will complain about uninitialized memory being
23936 used in DRI2InvalidateDrawable.
23937
23938 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
23939 Reviewed-by: Ville Syrjälä <syrjala@sci.fi>
23940 Signed-off-by: Keith Packard <keithp@keithp.com>
23941
23942commit 3aca8199405cff5097ee5813605ec78b5ae44a43
23943Author: Peter Hutterer <peter.hutterer@who-t.net>
23944Date: Fri Dec 16 09:15:25 2011 -0500
23945
23946 dmx: force -fno-strict-aliasing for xinput example
23947
23948 Compiler warning:
23949 xinput.c:272: warning: dereferencing pointer 'e' does break strict-aliasing
23950 rules
23951
23952 The code itself is the usual XInput client-side code:
23953 XEvent event;
23954 XDeviceMotionEvent *e = (XDeviceMotionEvent *)&event;
23955 XNextEvent(display, &event);
23956 printf("%d\n", e->type);
23957
23958 Since XDeviceMotionEvent is not guaranteed the same size as XEvent, clients
23959 must use pointer aliasing as above when using the XNextEvent API. Disable
23960 strict aliasing for this example.
23961
23962 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
23963 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
23964 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
23965 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
23966 Signed-off-by: Keith Packard <keithp@keithp.com>
23967
23968commit 1deede3c48d5ac91f50774c79f02afc6e036614e
23969Author: Jeremy Huddleston <jeremyhu@apple.com>
23970Date: Sat Dec 17 12:56:59 2011 -0800
23971
23972 kdrive/linux: Fix compilation with -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast
23973
23974 Reported-by: Arkadiusz Miśkiewicz <arekm@maven.pl>
23975 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
23976 Reviewed-by: Jamey Sharp <jamey@minilop.net>
23977
23978commit d0170d54c71fc761b8f1ae7a0e555dc2ca23cf9c
23979Author: Jeremy Huddleston <jeremyhu@apple.com>
23980Date: Sat Dec 17 12:25:26 2011 -0800
23981
23982 configure.ac: Make Xephyr dependency error message more informative
23983
23984 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
23985 Reviewed-by: Jamey Sharp <jamey@minilop.net>
23986
23987commit 511beddffab3c4fd43c75556132159117b6ba884
23988Author: Jeremy Huddleston <jeremyhu@apple.com>
23989Date: Fri Nov 25 09:45:25 2011 -0800
23990
23991 XQuartz: Provide in-tree implementation of strndup(3) if needed
23992
23993 This function was added to Mac OS X 10.7, so we need to provide it
23994 for earlier releases.
23995
23996 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
23997
23998commit 9ce8463a3334f76f128bb35b541aa6e67b4e4374
23999Author: Daniel Stone <daniel@fooishbar.org>
24000Date: Wed Dec 14 15:41:07 2011 +1000
24001
24002 dix: generate touchpoints from driver-submitted data
24003
24004 The touchpoints are generated, enqueued but not processed since we don't
24005 handle them in the event processing yet.
24006
24007 Co-authored-by: Peter Hutterer <peter.hutterer@who-t.net>
24008 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24009 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24010
24011commit bec51f5ec6575a1971ae5371a4b6d8563933032c
24012Author: Daniel Stone <daniel@fooishbar.org>
24013Date: Wed Dec 14 15:29:48 2011 +1000
24014
24015 dix: add GetTouchOwnership event API
24016
24017 No callers yet. This API is not to be used by drivers, it's an API for the
24018 DIX which will create ownership events mainly on touch acceptance/rejection.
24019
24020 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24021 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24022
24023commit 5c63dc6dbcbebbb19d79575a9f1ec9878e6537f1
24024Author: Peter Hutterer <peter.hutterer@who-t.net>
24025Date: Wed Dec 14 14:57:46 2011 +1000
24026
24027 dix: add DIX API to create touchpoints
24028
24029 The DIX touchpoints are the ones used for event processing.
24030
24031 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
24032 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24033 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24034
24035commit 7f8127d203394cae45c3ded0d063030d7c5fdb70
24036Author: Peter Hutterer <peter.hutterer@who-t.net>
24037Date: Wed Dec 14 14:53:52 2011 +1000
24038
24039 dix: if we run out of space for new touch events, resize the queue
24040
24041 The SIGIO handler forces us to drop the current touch and schedule the
24042 actual resize for later. Should not happen if the device sets up the
24043 TouchClassRec with the correct number of touchpoints.
24044
24045 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
24046 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24047 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24048
24049commit d26cb70272a1245a7b1db639cebeddafd8ed4c40
24050Author: Ian Romanick <ian.d.romanick@intel.com>
24051Date: Thu Dec 8 12:07:38 2011 -0800
24052
24053 glx: use dispatch.h instead of glapioffsets.h
24054
24055 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
24056 Reviewed-by: Adam Jackson <ajax@redhat.com>
24057
24058commit 66e6037c334c2bcfe6e60d24cac83a8d6087fdd5
24059Author: Ian Romanick <ian.d.romanick@intel.com>
24060Date: Mon Dec 5 15:27:08 2011 -0800
24061
24062 glx: Regenerate GL API files from Mesa
24063
24064 Adam reviewed the underlying changes in the Mesa generated scripts.
24065
24066 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
24067 Reviewed-by: Adam Jackson <ajax@redhat.com>
24068
24069commit e8485768c1f1d9d9fb7ca9e52ff61175b7bb80e1
24070Author: Ian Romanick <ian.d.romanick@intel.com>
24071Date: Tue Dec 6 12:30:17 2011 -0800
24072
24073 glx: Add stub implementations of next GLX extension functions
24074
24075 Generated code from Mesa contains dispatch for GLX_ARB_create_context
24076 and GLX_ARB_create_context_profile. Provide stub implementations of
24077 these functions so that the GLX extension will link.
24078
24079 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
24080 Reviewed-by: Adam Jackson <ajax@redhat.com>
24081
24082commit 2df539c0bc3300ea858f8bc7d52e95e67ff379b8
24083Author: Ian Romanick <ian.d.romanick@intel.com>
24084Date: Mon Dec 5 14:42:35 2011 -0800
24085
24086 glx: Only declare GlxExtensionInit in one header file
24087
24088 This silences the GCC warning:
24089
24090 In file included from singlepixswap.c:36:0:
24091 glxext.h:47:13: warning: redundant redeclaration of 'GlxExtensionInit' [-Wredundant-decls]
24092 glxserver.h:80:6: note: previous declaration of 'GlxExtensionInit' was here
24093
24094 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
24095 Reviewed-by: Adam Jackson <ajax@redhat.com>
24096
24097commit 758bc57ba5a89f765d83f0b169aa09e79a89bf89
24098Author: Peter Hutterer <peter.hutterer@who-t.net>
24099Date: Wed Dec 14 14:48:56 2011 +1000
24100
24101 dix: add helper functions to create DDX touch recs
24102
24103 DDX touch points are the ones that keep records of the driver-submitted
24104 touchpoints. They're unaffected by the grab state and terminate on a
24105 TouchEnd submitted by the driver.
24106
24107 The client ID assigned is server-global.
24108
24109 Since drivers usually submit in the SIGIO handler, we cannot allocate in the
24110 these functions.
24111
24112 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
24113 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24114 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24115
24116commit 1a133eb8b1ddbe0da7c2fbf7f6a686ec4512373e
24117Author: Peter Hutterer <peter.hutterer@who-t.net>
24118Date: Thu Dec 15 08:30:02 2011 +1000
24119
24120 xfree86: bump the input ABI for the touch changes
24121
24122 New additions to the API:
24123 - InitTouchClassDeviceStruct
24124 - xf86PostTouchEvent
24125
24126 Changes to the ABI:
24127 - DeviceIntRec now contains a TouchClassPtr
24128
24129 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24130 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24131
24132commit 3fb258ca28850c998097b55884774cb95f476f69
24133Author: Daniel Stone <daniel@fooishbar.org>
24134Date: Wed Dec 14 12:46:40 2011 +1000
24135
24136 input: add a TouchClassRec to the devices
24137
24138 These structs will be used to store touch-related data, events and
24139 information.
24140
24141 Drivers must call InitTouchClassDeviceStruct to set up a multi-touch capable
24142 device.
24143
24144 Touchpoints for the DDX and the DIX are handled separately - touchpoints
24145 submitted by the driver/DDX will be stored in the DDXTouchPointInfoRec. Once
24146 the touchpoints are processed by the DIX, new TouchPointInfoRecs are created
24147 and stored. This process is already used for pointer events with the
24148 last.valuators field.
24149
24150 Note that this patch does not actually add the generation of touch events,
24151 only the required structs.
24152
24153 TouchListeners are (future) recipients of touch or emulated pointer events.
24154 Each listener is in a state, depending which event they have already
24155 received. The type of listener defines how the listener got to be one.
24156
24157 Co-authored-by: Peter Hutterer <peter.hutterer@who-t.net>
24158 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24159 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24160
24161commit 098b837440e40bbc485368ec9658e12efd6ef581
24162Author: Daniel Stone <daniel@fooishbar.org>
24163Date: Wed Dec 14 14:41:48 2011 +1000
24164
24165 Add the touch input API stubs
24166
24167 xf86PostTouchEvent is the driver API to submit touch events to the server.
24168 This API doesn't do anything yet though but now we can at least bump the
24169 API.
24170
24171 For valuators, drivers should use the existing xf86InitValuatorAxisStruct
24172 function.
24173
24174 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24175 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24176
24177commit 93c2a1628a9f6e7480d834bf55c080997fd18911
24178Author: Peter Hutterer <peter.hutterer@who-t.net>
24179Date: Wed Dec 14 13:51:12 2011 +1000
24180
24181 Hook up the ownership events
24182
24183 This patch applies most of the protocol conversions and the internal event
24184 type for ownership events.
24185
24186 Note that ownership events are generated by the DIX only, they do not pass
24187 through the event queue.
24188
24189 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
24190 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24191 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24192
24193commit c065f7b16329e30cb976359e0a807c2adaf50123
24194Author: Keith Packard <keithp@keithp.com>
24195Date: Sat Dec 17 17:22:45 2011 -0800
24196
24197 Version 1.11.99.2 (1.12 snapshot 2)
24198
24199 Signed-off-by: Keith Packard <keithp@keithp.com>
24200
24201commit ead968a4300c0adeff89b9886e888b6d284c75cc
24202Author: Antoine Martin <antoine@nagafix.co.uk>
24203Date: Sat Dec 17 01:36:51 2011 +0700
24204
24205 xserver: check for elevated privileges not uid=0
24206
24207 This allows us to run the server as a normal user whilst still
24208 being able to use the -modulepath, -logfile and -config switches
24209 We define a xf86PrivsElevated which will do the checks and cache
24210 the result in case it is called more than once.
24211 Also renamed the paths #defines to match their new meaning.
24212 Original discussion which led to this patch can be found here:
24213 http://lists.freedesktop.org/archives/xorg-devel/2011-September/025853.html
24214
24215 Signed-off-by: Antoine Martin <antoine@nagafix.co.uk>
24216 Tested-by: Michal Suchanek <hramrach at centrum.cz>
24217 Reviewed-by: Jamey Sharp <jamey at minilop.net>
24218 Reviewed-by: Adam Jackson <ajax@redhat.com>
24219
24220commit 3390d3fc0347947796aa245d70fefecf59b7bfe4
24221Author: Peter Hutterer <peter.hutterer@who-t.net>
24222Date: Wed Dec 14 13:21:59 2011 +1000
24223
24224 Xi: process raw touch events
24225
24226 No-one can generated them yet, but if they could, we'd be processing them
24227 like there was no tomorrow.
24228
24229 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24230 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24231
24232commit 84db813b9db34975b5fe288a8a551bb98f0cc1c1
24233Author: Peter Hutterer <peter.hutterer@who-t.net>
24234Date: Wed Dec 14 12:45:28 2011 +1000
24235
24236 Hook up TouchBegin/Update/End events
24237
24238 The are the same as device events internally but require the touch ID
24239 separately from the detail.button field (the protocol uses the detail field
24240 for the touch id).
24241 For simpler integration of pointer emulation we need to set the
24242 detail.button field while keeping the touchid around.
24243
24244 Add the three new touch event types to the various places in the server
24245 where they need to be handled. The actual handling of the events is somewhat
24246 more complicated in most places.
24247
24248 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24249 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24250
24251commit 92a5862d0c120b009a688237ec7142b7c21ae272
24252Author: Peter Hutterer <peter.hutterer@who-t.net>
24253Date: Wed Dec 14 15:00:33 2011 +1000
24254
24255 include: add a bunch of flags for GetTouchEvent processing
24256
24257 TOUCH_CLIENT_ID is set if the touch was generated from a client ID instead
24258 of a DDX/driver touch ID. i.e. submitted by the dix.
24259
24260 TOUCH_END is a special flag that's required to force the touch to end.
24261 Since the protocol with grab replaying and pointer emulation is rather
24262 complex, it's quite hard to know otherwise when a touch sequence should
24263 really die.
24264
24265 The others do what it says on the imaginary box.
24266
24267 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24268 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24269
24270commit 956a97487beada8cf5f354550ff779635cc64361
24271Author: Peter Hutterer <peter.hutterer@who-t.net>
24272Date: Wed Dec 14 12:19:27 2011 +1000
24273
24274 include: RawTouchEnd is the last event now
24275
24276 Plus, use the actual definition from the protocol instead of the numeric
24277 values. Turns out not everyone knows the protocol event IDs by heart.
24278
24279 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24280 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24281
24282commit 92c6c0c1d2d6a8882ed97b0e7f2952978ee5ce82
24283Author: Peter Hutterer <peter.hutterer@who-t.net>
24284Date: Wed Dec 14 12:15:29 2011 +1000
24285
24286 tests: update for touch support
24287
24288 Touch event mask must be set for all three event types.
24289
24290 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24291 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24292 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24293
24294commit 731b1db2fcc82c0a8afd28d7f28363c1db0a20ce
24295Author: Peter Hutterer <peter.hutterer@who-t.net>
24296Date: Wed Dec 14 10:41:17 2011 +1000
24297
24298 Require inputproto 2.1.99.3
24299
24300 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24301 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24302 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24303
24304commit 4c825eae89063fdc2241f92ab4101ba682b5b4b7
24305Author: Peter Hutterer <peter.hutterer@who-t.net>
24306Date: Wed Nov 2 10:38:26 2011 +1000
24307
24308 Xi: add a FIXME
24309
24310 All the DeepCopy stuff really needs to be shared between the init calls the
24311 drivers use and this code here. Too many bugs by not keeping the two in
24312 sync.
24313
24314 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24315 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24316 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24317
24318commit 0b9d2e426defabbf6df7e872e158b080a6a5e0be
24319Author: Peter Hutterer <peter.hutterer@who-t.net>
24320Date: Wed Dec 14 13:21:03 2011 +1000
24321
24322 Xi: split ProcessOtherEvent into ProcessDeviceEvent
24323
24324 No functional changes.
24325
24326 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24327 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24328 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24329
24330commit 190a6f45ee7412870b3d95c5db6c1079a285bc59
24331Author: Peter Hutterer <peter.hutterer@who-t.net>
24332Date: Fri Dec 16 09:52:43 2011 +1000
24333
24334 dix: split positionSprite into scale_to_desktop and positionSprite
24335
24336 For future touch points, we need positionSprite to calculate the coordinates
24337 but we don't want to actually change the cursor position for non-emulating
24338 touches.
24339
24340 No functional changes at this point.
24341
24342 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24343 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24344 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24345
24346commit 2c1d0a539c0fb1860ac89a00d1fb609f1959e1b6
24347Author: Gaetan Nadon <memsize@videotron.ca>
24348Date: Fri Dec 9 07:45:02 2011 -0500
24349
24350 dmx: fix distcheck failure, missing compsize.h in Makefile.am
24351
24352 which was added in commit:
24353 dmx: Build fix for -Werror=implicit-function-declaration
24354
24355 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24356 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
24357 Signed-off-by: Keith Packard <keithp@keithp.com>
24358
24359commit ff6d9c79c1aa502f1aef81d8063425e5fc002534
24360Author: Peter Hutterer <peter.hutterer@who-t.net>
24361Date: Thu Dec 15 10:19:26 2011 +1000
24362
24363 include: fix BUG_WARN_MSG for constant messages only
24364
24365 Previous declaration required the use of a message + printf varargs. We
24366 obviously want to allow the use of just a message.
24367
24368 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24369 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24370
24371commit f7ea7a324520844beaac8be7503ac50e20da5ab1
24372Author: Peter Hutterer <peter.hutterer@who-t.net>
24373Date: Wed Dec 14 10:34:19 2011 +1000
24374
24375 configure: split the required modules up
24376
24377 We do the same thing for libraries and optional modules already, and it's
24378 much easier to read when one of them changes
24379
24380 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24381 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24382 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
24383 Signed-off-by: Keith Packard <keithp@keithp.com>
24384
24385commit 7da7aa96a0acb895e1fe57a9d3f2d093479cf2a2
24386Merge: b79de3f 7528a6b
24387Author: Keith Packard <keithp@keithp.com>
24388Date: Wed Dec 14 11:40:10 2011 -0800
24389
24390 Merge remote-tracking branch 'whot/for-keith'
24391
24392commit 4c80dc1394344dc10cffe4f757c980944312dc3d
24393Author: Peter Hutterer <peter.hutterer@who-t.net>
24394Date: Wed Dec 14 10:31:23 2011 +1000
24395
24396 configure: split the required modules up
24397
24398 We do the same thing for libraries and optional modules already, and it's
24399 much easier to read when one of them changes
24400
24401 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24402
24403commit 7528a6b88eb32098af4369a8bd9d70a808fa6f1c
24404Merge: f0de5e3 898d976
24405Author: Peter Hutterer <peter.hutterer@who-t.net>
24406Date: Tue Dec 13 14:19:54 2011 +1000
24407
24408 Merge branch 'for-whot' of git://people.freedesktop.org/~alanc/xserver into for-keith
24409
24410commit f0de5e3990b70df9899b7dd17d7ee6ff7c768183
24411Author: Peter Hutterer <peter.hutterer@who-t.net>
24412Date: Sun Dec 11 07:31:43 2011 +1000
24413
24414 dix: add a MAXEVENTS define for the number of core + extension events
24415
24416 Not including GenericEvents
24417
24418 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24419 Reviewed-by: Walter Harms <wharms@bfs.de>
24420 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24421
24422commit adf8a0d62c0643f4c44a3b6d2e3c6c9b02532915
24423Author: Peter Hutterer <peter.hutterer@who-t.net>
24424Date: Fri Dec 9 18:38:53 2011 +1000
24425
24426 dix: move event filter retrieval helpers to inpututils.c
24427
24428 No functional changes
24429
24430 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24431 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24432 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24433
24434commit a03fe819506fbcbfcb299d4919d16061603fd4ec
24435Author: Peter Hutterer <peter.hutterer@who-t.net>
24436Date: Fri Dec 9 18:02:49 2011 +1000
24437
24438 dix: switch EventIsDeliverable to take the event type only
24439
24440 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24441 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24442 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24443
24444commit 10a8c428fac0d4ff8f8a3e3487c13dd349d27ed7
24445Author: Peter Hutterer <peter.hutterer@who-t.net>
24446Date: Fri Dec 9 15:49:04 2011 +1000
24447
24448 dix: move storing last.valuators into a helper function
24449
24450 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24451 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24452
24453commit a15abf82ddb5a209dbc30c4d12e1dc23c4adf3e4
24454Author: Peter Hutterer <peter.hutterer@who-t.net>
24455Date: Thu Dec 8 15:38:44 2011 +1000
24456
24457 dix: always allocate and set the grab's sync.event
24458
24459 Allocate the memory at device creation time and always store the event, even
24460 if we're not frozen. This way we know which event triggered the grab.
24461 Since the event was never freed anyway except on device shutdown, this
24462 doesn't really change things much.
24463
24464 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24465 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24466
24467commit 218752bdc5d9323d1e6202e762573a925cf8a4eb
24468Author: Peter Hutterer <peter.hutterer@who-t.net>
24469Date: Thu Dec 8 14:27:01 2011 +1000
24470
24471 input: replace GRABTYPE_* with the InputLevel enums
24472
24473 They achieve the same thing, re-use the more generic InputLevel so we can
24474 convert to/fro easier.
24475
24476 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24477 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24478
24479commit cd56bd7b3ee34a4b10eb3a57a6e94cac7512167a
24480Author: Peter Hutterer <peter.hutterer@who-t.net>
24481Date: Thu Dec 8 14:59:06 2011 +1000
24482
24483 Add GrabIsPointerGrab and GrabIsKeyboardGrab helpers
24484
24485 No functional changes
24486
24487 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24488 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24489
24490commit 47734b2ba2e1f9246fd1a3b7059ca60b0e15acb2
24491Author: Peter Hutterer <peter.hutterer@who-t.net>
24492Date: Thu Dec 8 12:00:34 2011 +1000
24493
24494 dix: move delivery stop condition out of event mask
24495
24496 Previously, this was only called if there was a mask match, so even if we
24497 had a no-propagate flag set or a stopAt window specified, if no mask
24498 triggered on the window we would recurse up to the root window and
24499 eventually deliver.
24500 Move this, so that the stopAt and do-not-propagate mask is honoured.
24501
24502 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24503 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24504
24505commit 1d01e861b69d7ca8fe2335270b8286bc115fb6e9
24506Author: Peter Hutterer <peter.hutterer@who-t.net>
24507Date: Thu Dec 8 10:22:26 2011 +1000
24508
24509 dix: split out core state and event state setting into helper functions
24510
24511 no functional changes.
24512
24513 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24514 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24515
24516commit fb010c543696b781a83b945f72865b7d7b75bf95
24517Author: Peter Hutterer <peter.hutterer@who-t.net>
24518Date: Wed Dec 7 14:41:18 2011 +1000
24519
24520 dix: move grab matching code into a helper function
24521
24522 No functional changes.
24523
24524 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24525 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24526
24527commit d253a262c2c690357a4db7e235c48ab5dd0b77f8
24528Author: Peter Hutterer <peter.hutterer@who-t.net>
24529Date: Wed Dec 7 13:57:25 2011 +1000
24530
24531 dix: split core grab interference check into helper function
24532
24533 No functional changes.
24534
24535 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24536 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24537
24538commit 97d79dd740ed1ac4eb02b0d4135850f953716278
24539Author: Peter Hutterer <peter.hutterer@who-t.net>
24540Date: Wed Dec 7 13:54:28 2011 +1000
24541
24542 dix: compare the grab type, not the tempGrab type
24543
24544 No functional change. To get here, GrabMatchesSecond() needs to be TRUE and
24545 for that the two grab types must be identical.
24546
24547 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24548 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24549
24550commit b9d6ae42b686287d7d23c2eefe2b12da98fe6ff8
24551Author: Peter Hutterer <peter.hutterer@who-t.net>
24552Date: Wed Dec 7 15:43:31 2011 +1000
24553
24554 dix: remove event type check
24555
24556 Can't remember why this is there but we'll need to pass in XI2 events soon,
24557 so this check is obsolete.
24558
24559 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24560 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24561
24562commit 80816366aa77f1730a1b6ddabfa55a946d76d494
24563Author: Chase Douglas <chase.douglas@canonical.com>
24564Date: Fri Nov 25 13:57:03 2011 -0800
24565
24566 dix: Split ActivatePassiveGrab() from CheckPassiveGrab()
24567
24568 The changed logic means we don't require the explicit grab = NULL setting
24569 and early exit anymore. Not 100% of it, but if we see that message pop up in
24570 a log we know it's broken.
24571
24572 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
24573 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24574 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24575
24576commit 9ee62cd8ce3c3effc3663f3d56b322385ce12fdb
24577Author: Chase Douglas <chase.douglas@canonical.com>
24578Date: Thu Nov 17 17:40:24 2011 -0800
24579
24580 dix: Move grab check and activation logic to CheckPassiveGrab()
24581
24582 This is needed for future pointer emulation work.
24583
24584 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
24585 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24586 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24587
24588commit c53651dabc66aeb9882819a404a799a364f00a29
24589Author: Peter Hutterer <peter.hutterer@who-t.net>
24590Date: Tue Dec 6 12:57:38 2011 +1000
24591
24592 dix: move EventDeliveryState into a header file, we'll need it elsewhere
24593
24594 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24595 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24596
24597commit 86c3137c81e924bacd919e9d65809e07afeabbff
24598Author: Peter Hutterer <peter.hutterer@who-t.net>
24599Date: Tue Dec 6 12:04:11 2011 +1000
24600
24601 Xi: split updating button count and state into helper functions
24602
24603 Functional change: for a button mapped to 0, the motionHintWindow is not
24604 updated to the NullWindow anymore. Before it got updated unconditionally to
24605 the button mapping. I have no idea what the practical effect of this is, but
24606 I guess it's closer to the correct behaviour: pressing a button that's
24607 logically disabled now does not disrupt the motion hint delivery.
24608
24609 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24610 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24611
24612commit e0f37250ffff5dcb3bc0e8cad63439995ce01a20
24613Author: Peter Hutterer <peter.hutterer@who-t.net>
24614Date: Tue Dec 6 11:40:33 2011 +1000
24615
24616 Xi: deduplicate button motion mask setting
24617
24618 No functional changes
24619
24620 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24621 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24622
24623commit 6eff14a789341d366b3013c5aa020e959c954651
24624Author: Peter Hutterer <peter.hutterer@who-t.net>
24625Date: Mon Dec 5 18:54:30 2011 +1000
24626
24627 dix: deduplicate callers of DeliverDeviceEvents in DeliverGrabbedEvents
24628
24629 No functional change.
24630
24631 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24632 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24633
24634commit 93945b0a74aa8156a88f52b8ba77f1210042f396
24635Author: Peter Hutterer <peter.hutterer@who-t.net>
24636Date: Mon Dec 5 18:42:05 2011 +1000
24637
24638 dix: split grab event conversion and delivery into a helper function
24639
24640 Effective functional change: XI2 events are checked with XACE now.
24641
24642 DeliverOneGrabbedEvent is exported for future use by touch events.
24643
24644 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24645 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24646
24647commit c81cdb0862e2184f033b3933e6bacbe0809ef2c0
24648Author: Peter Hutterer <peter.hutterer@who-t.net>
24649Date: Mon Dec 5 18:29:27 2011 +1000
24650
24651 dix: replace conversion errors with BUG_WARN_MSG
24652
24653 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24654 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24655
24656commit 6368c2aa4613a7c7eb0e8afca8d41f1a9bc4fc4d
24657Author: Peter Hutterer <peter.hutterer@who-t.net>
24658Date: Mon Dec 5 16:46:18 2011 +1000
24659
24660 dix: deduplicate event delivery code
24661
24662 Move all the event delivery code into DeliverOneEvent, based on the
24663 InputLevel we're sending to.
24664
24665 Functional change: we now check XI2 events with XACE too.
24666
24667 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24668 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
24669
24670commit 7e2207548b3173afc9accb5ccd532c181a7b94ed
24671Author: Chase Douglas <chase.douglas@canonical.com>
24672Date: Wed Dec 7 20:32:28 2011 +1000
24673
24674 include: Add an InputLevel enum
24675
24676 Currently unused, but will be in the future.
24677
24678 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
24679 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24680 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
24681
24682commit 898d97672907a8074031d0d1038cd44bb5caf40d
24683Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24684Date: Mon Dec 12 16:49:33 2011 -0800
24685
24686 verify_internal_event: preserve constness of data pointer
24687
24688 All we're using it for is ErrorF calls, so make it a const char *
24689 to stop gcc from warning:
24690
24691 inpututils.c: In function 'verify_internal_event':
24692 inpututils.c:629:9: warning: cast discards qualifiers from pointer target type
24693
24694 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24695 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24696 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24697
24698commit 79d09647d878e66721a778979a9eb1f4bba5f8d6
24699Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24700Date: Mon Dec 12 16:49:33 2011 -0800
24701
24702 Use const cast in BitIsOn macro to avoid angering gcc
24703
24704 Fixes gcc warnings such as:
24705 inpututils.c: In function 'valuator_mask_isset':
24706 inpututils.c:498:5: warning: cast discards qualifiers from pointer target type
24707 inpututils.c: In function 'CountBits':
24708 inpututils.c:613:9: warning: cast discards qualifiers from pointer target type
24709
24710 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24711 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24712 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24713
24714commit 99dfe9b1de09cb481e95bd4c45dcf7493480798b
24715Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24716Date: Mon Dec 12 16:49:32 2011 -0800
24717
24718 Fix deconstifying cast warning in xi2_get_type
24719
24720 Since we're just comparing values in the struct, cast it to a
24721 const xGenericEvent * to clear gcc warning of:
24722
24723 events.c: In function 'xi2_get_type':
24724 events.c:193:5: warning: cast discards qualifiers from pointer target type
24725
24726 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24727 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24728 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24729
24730commit dd80156bf05c9eb4000d0981e00d0d9fb69a94f6
24731Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24732Date: Wed Dec 7 22:58:45 2011 -0800
24733
24734 Include client name if available in PrintDeviceGrabInfo
24735
24736 Also adds missing newline to first line of output.
24737
24738 Before patch:
24739
24740 [3581472.414] (II) Printing all currently active device grabs:
24741 [3581472.414] Active grab 0x1800000 (core) on device 'Virtual core pointer' (2):
24742 client pid 26174 uid 0 gid 10
24743 [3581472.415] at 3581469139 (from active grab) (device thawed, state 1)
24744 [3581472.415] core event mask 0x0
24745 [3581472.415] owner-events true, kb 1 ptr 1, confine 0, cursor 0x0
24746 [3581472.415] Active grab 0x1800000 (core) on device 'Virtual core keyboard' (3)
24747 : client pid 26174 uid 0 gid 10
24748 [3581472.415] at 3581469139 (from active grab) (device thawed, state 1)
24749 [3581472.415] core event mask 0x3
24750 [3581472.415] owner-events true, kb 1 ptr 1, confine 0, cursor 0x0
24751 [3581472.415] (II) End list of active device grabs
24752
24753 After patch:
24754 [3581736.601] (II) Printing all currently active device grabs:
24755 [3581736.601] Active grab 0x1600000 (core) on device 'Virtual core pointer' (2):
24756 [3581736.601] client pid 26741 /usr/bin/xscreensaver -nosplash
24757 [3581736.601] at 3581735000 (from active grab) (device thawed, state 1)
24758 [3581736.601] core event mask 0x0
24759 [3581736.601] owner-events true, kb 1 ptr 1, confine 0, cursor 0x0
24760 [3581736.601] Active grab 0x1600000 (core) on device 'Virtual core keyboard' (3)
24761 :
24762 [3581736.601] client pid 26741 /usr/bin/xscreensaver -nosplash
24763 [3581736.601] at 3581735000 (from active grab) (device thawed, state 1)
24764 [3581736.601] core event mask 0x3
24765 [3581736.601] owner-events true, kb 1 ptr 1, confine 0, cursor 0x0
24766 [3581736.601] (II) End list of active device grabs
24767
24768 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24769 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
24770
24771commit b79de3f42ff8412dcabf2f2a805cbcbd1c6907c3
24772Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24773Date: Mon Dec 12 16:49:35 2011 -0800
24774
24775 xf86 parser: convert Error to a varargs macro to clear gcc format warnings
24776
24777 Previously it always passed a format string with exactly one argument,
24778 using NULL when the format string needed none. Now pass the right number
24779 of arguments to clear gcc warnings of 'too many arguments for format'.
24780
24781 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24782 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24783 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24784
24785commit 7801b3dcd6501e3de93a6d2cee93b2593806e922
24786Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24787Date: Mon Dec 12 16:49:35 2011 -0800
24788
24789 Add some printf format attributes suggested by gcc
24790
24791 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24792 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24793 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24794
24795commit f68df9dfd2c7ca03c24aaa0a895b7191ecb81e85
24796Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24797Date: Mon Dec 12 16:49:34 2011 -0800
24798
24799 xf86Priv.h: Add some noreturn attributes suggested by gcc
24800
24801 Both functions call exit() at the end and have no other return path.
24802 Also correct comment/heading to reflect commit 6450f6ca7ee0 moving
24803 DoShowOptions into xf86Configure.c.
24804
24805 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24806 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24807 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24808
24809commit 3823eedf3c124d3a20360480ba349bea72de3069
24810Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24811Date: Mon Dec 12 16:49:34 2011 -0800
24812
24813 Remove duplicate declarations of KdAdd*Driver in kdrive.h
24814
24815 Clears gcc warnings in every file that includes kdrive.h:
24816 kdrive.h:507:1: warning: redundant redeclaration of 'KdAddPointerDriver'
24817 kdrive.h:225:6: note: previous declaration of 'KdAddPointerDriver' was here
24818 kdrive.h:510:1: warning: redundant redeclaration of 'KdAddKeyboardDriver'
24819 kdrive.h:284:6: note: previous declaration of 'KdAddKeyboardDriver' was here
24820
24821 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24822 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24823 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24824
24825commit 471e5373b6f87336dac3b9de9b3852f63a2b10dc
24826Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24827Date: Mon Dec 12 16:49:34 2011 -0800
24828
24829 Remove duplicate declaration of xf86ValidateModesFlags in xf86Modes.h
24830
24831 Clears gcc warning in every file that includes xf86Modes.h:
24832 xf86Modes.h:102:1: warning: redundant redeclaration of 'xf86ValidateModesFlags'
24833 xf86Modes.h:72:1: note: previous declaration of 'xf86ValidateModesFlags' was here
24834
24835 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24836 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24837 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24838
24839commit 41dd7cf673e47ab74c5065b8dbf268b2e5facb64
24840Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24841Date: Mon Dec 12 16:49:34 2011 -0800
24842
24843 _XkbFilterDeviceBtn: move variable declarations to match usage scope
24844
24845 The main body of this function is an if { } else if { } pair of blocks.
24846 Previously there was int button at the top level scope which is used
24847 only in the first block, and a redeclaration of int button inside the
24848 second block. Since there's no overlap in the code paths for the
24849 two uses of button, move the one from the outer block into the first
24850 block to help the programmer more quickly determine they are unrelated
24851 usages, and to silence the gcc warning of:
24852
24853 xkbActions.c: In function '_XkbFilterDeviceBtn':
24854 xkbActions.c:999:6: warning: declaration of 'button' shadows a previous local
24855 xkbActions.c:955:6: warning: shadowed declaration is here
24856
24857 For consistency, move DeviceIntPtr dev declarations as well that are
24858 used in the same way.
24859
24860 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24861 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24862 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24863
24864commit d8eb9b5faa2d2c9350bdaffef8749a9cac3c87b8
24865Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24866Date: Mon Dec 12 16:49:34 2011 -0800
24867
24868 XkbFindSrvLedInfo: remove extraneous name-clashing sli variable
24869
24870 Variable is already defined outside the outer if statement, and
24871 there's no need to redefine inside the if statement.
24872
24873 No point in setting sli before if (dev->kbdfeed->xkb_sli==NULL)
24874 check - if check is true, we immediately set it, if check is false,
24875 we immediately return without further reference or use of it.
24876
24877 The one thing we do with it inside the inner if statement is store
24878 an allocation in it for a brief moment before writing to the final
24879 destination, which is immediately returned to the caller.
24880
24881 In short, there's no benefit to the variable at all in this block,
24882 it just gives the optimizer more code to figure out how to omit.
24883
24884 Fixes gcc warning:
24885 xkbLEDs.c: In function 'XkbFindSrvLedInfo':
24886 xkbLEDs.c:683:19: warning: declaration of 'sli' shadows a previous local
24887 xkbLEDs.c:679:18: warning: shadowed declaration is here
24888
24889 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24890 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24891 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24892
24893commit 9edfa47bd58cd318d57d0f114714fd146be35a79
24894Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24895Date: Mon Dec 12 16:49:34 2011 -0800
24896
24897 KdParseFindNext: Constify delim argument
24898
24899 It's only used as input to strchr to find the delimiters, never
24900 written to.
24901
24902 Clears a bunch of gcc warnings of the form:
24903 kdrive.c:323:2: warning: passing argument 2 of 'KdParseFindNext' discards qualifiers from pointer target type
24904 kdrive.c:261:1: note: expected 'char *' but argument is of type 'const char *'
24905
24906 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24907 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24908 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24909
24910commit ff64ad6c74a3e14ca34bacb8866bdbd2262bddff
24911Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24912Date: Mon Dec 12 16:49:34 2011 -0800
24913
24914 Convert KdDoSwitchCmd to use asprintf instead of malloc/strcat/etc.
24915
24916 Also fix the reason argument to be const char * to clear several gcc
24917 warnings of:
24918 kdrive.c:151:2: warning: passing argument 1 of 'KdDoSwitchCmd' discards qualifiers from pointer target type
24919 kdrive.c:116:1: note: expected 'char *' but argument is of type 'const char *'
24920
24921 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24922 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24923 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24924
24925commit 5bc590bde23ce1c57015b1d9e1cc63189c37448e
24926Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24927Date: Mon Dec 12 16:49:34 2011 -0800
24928
24929 DoShowOptions: preserve constness of options list as we walk it
24930
24931 Since all we do with the option list is walk down the list printing
24932 the names, there's no need to cast away its constness.
24933
24934 Clears gcc warning:
24935 xf86Configure.c: In function 'DoShowOptions':
24936 xf86Configure.c:781:4: warning: cast discards qualifiers from pointer target type
24937
24938 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24939 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24940 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24941
24942commit 71efd868282d47a6db405a16de18ac322926962b
24943Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24944Date: Mon Dec 12 16:49:34 2011 -0800
24945
24946 x86emu: constify debug strings
24947
24948 Strings are all pointers to literal constants, just used as input
24949 to printf calls when debugging is enabled.
24950
24951 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24952 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24953 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24954
24955commit d5a5eece670dee963765eab1431c21525c16d9ee
24956Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24957Date: Mon Dec 12 16:49:34 2011 -0800
24958
24959 CompareISOLatin1Lowered: constify arguments
24960
24961 Allows callers to avoid deconstifying arguments when calling, fixing
24962 gcc warning:
24963
24964 filter.c: In function 'PictureGetFilterId':
24965 filter.c:59:2: warning: cast discards qualifiers from pointer target type
24966
24967 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24968 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24969 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24970
24971commit 83ac9502ea9f1dedf3a8002745668af16bb1f2af
24972Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24973Date: Mon Dec 12 16:49:33 2011 -0800
24974
24975 xdmcp.c: fix three small const warnings
24976
24977 xdmcp.c:63:36: warning: initialization discards qualifiers from pointer target type
24978
24979 xdmcp.c: In function 'XdmcpRegisterConnection':
24980 xdmcp.c:482:8: warning: cast discards qualifiers from pointer target type
24981 xdmcp.c:482:8: warning: cast discards qualifiers from pointer target type
24982 xdmcp.c:482:8: warning: cast discards qualifiers from pointer target type
24983
24984 xdmcp.c: In function 'get_mcast_options':
24985 xdmcp.c:1596:21: warning: initialization discards qualifiers from pointer target type
24986
24987 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
24988 Reviewed-by: Jamey Sharp <jamey@minilop.net>
24989 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
24990
24991commit 3839d1480877818f3ccc04e25237d853bab210c3
24992Author: Alan Coopersmith <alan.coopersmith@oracle.com>
24993Date: Mon Dec 12 16:49:33 2011 -0800
24994
24995 LockServer: store path to LOCKDIR literal string in a const char *
24996
24997 And instead of initializing to NULL, then resetting to LOCKDIR almost
24998 immediately (before ever using the NULL value), skip directly to setting
24999 it to LOCKDIR.
25000
25001 tmppath variable is only used as input for generating the path name
25002 via calls to strlen, sprintf, etc.
25003
25004 Fixes gcc warning of:
25005 utils.c: In function 'LockServer':
25006 utils.c:259:11: warning: assignment discards qualifiers from pointer target type
25007
25008 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25009 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25010 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25011
25012commit af4113bfeb1d155f5f037c3492e50513336fa4a7
25013Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25014Date: Mon Dec 12 16:49:33 2011 -0800
25015
25016 WriteToClient: preserve constness of buf while extracting length value
25017
25018 Fixes gcc warning:
25019 io.c: In function 'WriteToClient':
25020 io.c:826:6: warning: cast discards qualifiers from pointer target type
25021
25022 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25023 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25024 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25025
25026commit 2c9800f91559fbb12dd276cf0536631104092f67
25027Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25028Date: Mon Dec 12 16:49:33 2011 -0800
25029
25030 OsInit: store "/dev/null" in a const char *
25031
25032 It's only passed as the input side of a strcpy and as the filename to
25033 fopen, so doesn't need to be non-const. Fixes gcc warning:
25034
25035 osinit.c: In function 'OsInit':
25036 osinit.c:154:28: warning: initialization discards qualifiers from pointer target type
25037
25038 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25039 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25040 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25041
25042commit 50b1097643934c8caec9530e5eda6ed6534aaf61
25043Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25044Date: Mon Dec 12 16:49:33 2011 -0800
25045
25046 Constify the reason string throughout the authorization check framework
25047
25048 Almost all of the places the string is assigned point to a literal
25049 string constant, so use const char * for those, and const char **
25050 for function calls that return it via an argument. Fortunately
25051 the top level function, ClientAuthorized, which returns the string
25052 as its return value is called from only one place, ProcEstablishConnection.
25053
25054 ProcEstablishConnection stores either that return value or a string literal
25055 in char *reason. It only uses reason as an argument to SendConnSetup.
25056 SendConnSetup passes the reason argument to strlen & WriteToClient,
25057 both of which already have const qualifiers on their args.
25058 Thus added const to the reason variable in ProcEstablishConnection
25059 and the reason argument to SendConnSetup.
25060
25061 Fixes gcc warnings:
25062 dispatch.c: In function 'ProcEstablishConnection':
25063 dispatch.c:3711:9: warning: assignment discards qualifiers from pointer target type
25064 auth.c: In function 'CheckAuthorization':
25065 auth.c:218:14: warning: assignment discards qualifiers from pointer target type
25066 auth.c:220:20: warning: assignment discards qualifiers from pointer target type
25067 connection.c: In function 'ClientAuthorized':
25068 connection.c:683:3: warning: return discards qualifiers from pointer target type
25069 mitauth.c: In function 'MitCheckCookie':
25070 mitauth.c:88:13: warning: assignment discards qualifiers from pointer target type
25071 xdmauth.c:259:14: warning: assignment discards qualifiers from pointer target type
25072 xdmauth.c:270:14: warning: assignment discards qualifiers from pointer target type
25073 xdmauth.c:277:11: warning: assignment discards qualifiers from pointer target type
25074 xdmauth.c:293:15: warning: assignment discards qualifiers from pointer target type
25075 xdmauth.c:313:14: warning: assignment discards qualifiers from pointer target type
25076 xdmauth.c:322:11: warning: assignment discards qualifiers from pointer target type
25077 rpcauth.c: In function 'SecureRPCCheck':
25078 rpcauth.c:136:10: warning: assignment discards qualifiers from pointer target type
25079
25080 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25081 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25082 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25083
25084commit 99998196017ef38ec88459b50605aa20f628788b
25085Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25086Date: Mon Dec 12 16:49:33 2011 -0800
25087
25088 Constify string for authorization protocol names
25089
25090 gcc was warning from storing string constants in a char *name field:
25091 auth.c:64:1: warning: initialization discards qualifiers from pointer target type
25092 auth.c:72:1: warning: initialization discards qualifiers from pointer target type
25093 auth.c:81:1: warning: initialization discards qualifiers from pointer target type
25094
25095 Making the field const requires changing AuthorizationFromID to take
25096 a const char ** pointer for the name argument which it sets to point
25097 to the matching name entry.
25098
25099 Changing that argument requires changing its sole caller in the security
25100 extension to pass the address of a const char * variable to it, which it
25101 can do, since the only thing it does with the returned name is to pass
25102 it back to the RemoveAuthorization function that already expects a const
25103 char *name.
25104
25105 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25106 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25107 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25108
25109commit 01834e99e461e2a8354f3b4aef7f14c5e83fa255
25110Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25111Date: Mon Dec 12 16:49:33 2011 -0800
25112
25113 os/access.c: replace acmp & acopy macros with memcmp & memcpy calls
25114
25115 No need to cast to char * now that all supported platforms use C89-standard
25116 void * argument types, so just drop the casts from acmp & acopy macros,
25117 which clears the gcc warnings for places const pointers were cast non-const:
25118
25119 access.c: In function 'DefineSelf':
25120 access.c:786:3: warning: cast discards qualifiers from pointer target type
25121 access.c:795:6: warning: cast discards qualifiers from pointer target type
25122 access.c: In function 'NewHost':
25123 access.c:1293:9: warning: cast discards qualifiers from pointer target type
25124 access.c:1298:6: warning: cast discards qualifiers from pointer target type
25125 access.c:1309:5: warning: cast discards qualifiers from pointer target type
25126
25127 Without the casts, acmp & acopy are just a funny way to write memcmp
25128 & memmove, so drop the macros and inline the calls, taking care to
25129 swap the first two arguments to memmove since it had swapped them.
25130
25131 Since all the calls to memmove end up being to non-overlapping memory
25132 (mostly copying from an existing pointer to a newly allocated one),
25133 replace those with memcpy.
25134
25135 And finally, don't actually call memcpy to copy 0 bytes from one place
25136 to another, since that's just a waste of a perfectly good function call.
25137
25138 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25139 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25140 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25141
25142commit 020d83d361b0ab54ac962e97b12935be785f9a67
25143Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25144Date: Mon Dec 12 16:49:33 2011 -0800
25145
25146 xres.c: Preserve constness of string returned by LookupResourceName
25147
25148 MakeAtom now accepts a const char * so we don't need to cast down to
25149 char * anymore. Fixes gcc warning of:
25150
25151 xres.c: In function 'ProcXResQueryClientResources':
25152 xres.c:155:6: warning: cast discards qualifiers from pointer target type
25153
25154 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25155 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25156 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25157
25158commit 2ddae8f0bd2a9ce0cd15bf3848393af29e615acf
25159Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25160Date: Mon Dec 12 16:49:33 2011 -0800
25161
25162 constify strings in resource name registry
25163
25164 LookupResourceName already returned a const char *, so just needed
25165 to change the variable we're storing the list in to be a const char **
25166 and then add const to the name argument to RegisterResourceName
25167 (which just stores name in the array) and CreateNewResourceType
25168 (which just passes name to RegisterResourceName).
25169
25170 Clears a bunch of gcc warnings of the form:
25171 registry.c:319:5: warning: passing argument 2 of 'RegisterResourceName' discards qualifiers from pointer target type
25172 registry.c:200:1: note: expected 'char *' but argument is of type 'const char *'
25173
25174 and from all the extensions:
25175 damageext.c: In function 'DamageExtensionInit':
25176 damageext.c:490:5: warning: passing argument 2 of 'CreateNewResourceType' discards qualifiers from pointer target type
25177 ../include/resource.h:159:26: note: expected 'char *' but argument is of type 'const char *'
25178
25179 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25180 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25181 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25182
25183commit 424dbde891486ad6a6c00c61a334031ff18f5556
25184Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25185Date: Fri Dec 9 10:08:55 2011 -0800
25186
25187 CheckForEmptyMask does not need to declare int n twice
25188
25189 Just use the existing n variable again in the ARGB_CURSOR loop
25190 instead of creating another one.
25191
25192 Fixes gcc -Wshadow warning:
25193 cursor.c: In function 'CheckForEmptyMask':
25194 cursor.c:155:6: warning: declaration of 'n' shadows a previous local
25195 cursor.c:146:9: warning: shadowed declaration is here
25196
25197 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25198 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25199 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25200
25201commit b2bc38e4a553c29f49a0284333b34b4d6c8a8c12
25202Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25203Date: Sat Dec 10 10:01:18 2011 -0800
25204
25205 Even more correctly free config file names
25206
25207 If we didn't go into the if (!autoconfig) { } block, the filename,
25208 dirname, and sysdirname pointers were never initialized, but we
25209 freed them outside the block, leading to potential memory corruption.
25210
25211 Move the frees inside the block where they're initialized to avoid this.
25212
25213 To avoid similar problems, move the declarations of the variables that
25214 are only used in this block inside the block.
25215
25216 Regression introduced by commit 3d635fe84d6de53e2f74203b10e89f7851fe3fc1
25217
25218 Found by gcc warning:
25219 xf86Config.c: In function 'xf86HandleConfigFile':
25220 xf86Config.c:2303:11: warning: 'filename' may be used uninitialized in this function
25221 xf86Config.c:2303:22: warning: 'dirname' may be used uninitialized in this function
25222 xf86Config.c:2303:32: warning: 'sysdirname' may be used uninitialized in this function
25223
25224 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25225 Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
25226
25227commit 33d6e6743d86324c2078f156991b16ac4f2593fc
25228Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25229Date: Wed Dec 7 19:06:05 2011 -0800
25230
25231 xf86RegisterRootWindowProperty is confused about xnfcalloc
25232
25233 It will never return NULL, so don't try to handle a NULL condition,
25234 since that just confuses programmers and static analyzers.
25235
25236 It uses calloc, so all the allocated memory is cleared, so there's
25237 no point looping over the memory to manually initialize it NULL.
25238
25239 And just because it's annoying, it doesn't need to be the only
25240 place in this file to do if (NULL==...) instead of if (... == NULL).
25241
25242 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25243 Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
25244
25245commit e5aa00989cda9ebd18063c5e955235123ad37b88
25246Author: Peter Hutterer <peter.hutterer@who-t.net>
25247Date: Wed Dec 7 14:14:10 2011 +1000
25248
25249 Change GetXI2/XI/CoreType to just take a type argument
25250
25251 Avoids the dummy-event dance if we have an event type and need to get the
25252 matching XI2 type.
25253
25254 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25255 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25256
25257commit 6cc0e6a0afa6a87802767d484aa4e68fa935d3eb
25258Author: Peter Hutterer <peter.hutterer@who-t.net>
25259Date: Fri Oct 28 12:55:55 2011 +1000
25260
25261 include: add GetXI2MaskByte and GetXI2EventFilterMask to headers
25262
25263 This is needed for touch event processing.
25264
25265 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25266 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25267
25268commit 372a6f10dc2d74d2d179e8b92449e9b8636a99ef
25269Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25270Date: Thu Dec 8 21:52:07 2011 -0800
25271
25272 Change disable_clientpointer return type to void
25273
25274 It doesn't return anything, nor does it's caller expect it to.
25275
25276 Fixes Solaris Studio compiler error:
25277 "xichangehierarchy.c", line 214: Function has no return statement : disable_clientpointer
25278
25279 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25280 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25281 Signed-off-by: Keith Packard <keithp@keithp.com>
25282
25283commit 522f8bcc0360d6a117e929a004bc956ab92037e9
25284Merge: 6369b59 2abe83d
25285Author: Keith Packard <keithp@keithp.com>
25286Date: Thu Dec 8 20:57:26 2011 -0800
25287
25288 Merge remote-tracking branch 'whot/for-keith'
25289
25290commit 2abe83df686ed64c4f4df711ac3c1fd12131c2e4
25291Author: Peter Hutterer <peter.hutterer@who-t.net>
25292Date: Mon Dec 5 14:02:51 2011 +1000
25293
25294 include: add BUG_WARN_MSG for custom error message on bug condition
25295
25296 __BUG_WARN_MSG is a simple helper to enable call with and without varargs. I
25297 couldn't find a way to otherwise do this without getting gcc warnings.
25298
25299 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25300 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25301
25302commit 4fc797f3756611a97767f407e1af0b6a7cf2f1d9
25303Author: Peter Hutterer <peter.hutterer@who-t.net>
25304Date: Fri Nov 11 16:25:30 2011 +1000
25305
25306 xfree86: include xorg-config.h from xaalocal.h
25307
25308 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25309 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25310
25311commit 8c9589c71d47f287588314d77ddbfcc22cd04c8a
25312Author: Peter Hutterer <peter.hutterer@who-t.net>
25313Date: Mon Dec 5 11:55:58 2011 +1000
25314
25315 Xi: rename "state" to "corestate" in ProcessDeviceEvents
25316
25317 'state' is shadowed by the XKB 'state' as well (which feeds into the event
25318 too), so rename this one to clarify that this is the core event state only.
25319
25320 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25321 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25322
25323commit 631516a4aa9858874ee197444cd93d91b97a1089
25324Author: Peter Hutterer <peter.hutterer@who-t.net>
25325Date: Fri Dec 2 15:47:58 2011 +1000
25326
25327 Xi: check button mapping value _before_ assigning it
25328
25329 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25330 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25331
25332commit a1304d6cb69301899c3c8450d6bf3e68573599df
25333Author: Peter Hutterer <peter.hutterer@who-t.net>
25334Date: Mon Dec 5 11:26:30 2011 +1000
25335
25336 Xi: skip superfluous cast
25337
25338 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25339 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25340
25341commit 1ab50be938524dcd4a9e56d27e3b96a27c2db2c0
25342Author: Peter Hutterer <peter.hutterer@who-t.net>
25343Date: Wed Nov 30 09:06:06 2011 +1000
25344
25345 xfixes: don't dereference a NULL cursor
25346
25347 If the new cursor is the NULL cursor, don't dereference it and use zeros
25348 instead.
25349
25350 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25351 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
25352
25353commit d2ebbcdaf6b13d70eee704b1764ff349e1be22a0
25354Author: Peter Hutterer <peter.hutterer@who-t.net>
25355Date: Tue Nov 29 16:15:37 2011 +1000
25356
25357 Xi: when removing a device, reset ClientPointers where needed
25358
25359 if a client had the to-be-removed device as ClientPointer, reset to NULL.
25360
25361 Fixes #43165
25362
25363 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25364 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
25365
25366commit 27425f07b29e0ddaa782345c1899273ca742891e
25367Author: Peter Hutterer <peter.hutterer@who-t.net>
25368Date: Wed Nov 9 14:45:02 2011 +1000
25369
25370 dix: use BUG_WARN for input mask size issues
25371
25372 Yes, we're likely corrupting memory here but really this is unlikely to be
25373 triggered other than a real bug in the server. In which case a stacktrace is
25374 going to be more useful than any silent error handling.
25375
25376 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25377 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25378
25379commit 9b570ecbda954227c89938ee6f94b9efd192d3c6
25380Author: Peter Hutterer <peter.hutterer@who-t.net>
25381Date: Fri Dec 9 10:48:05 2011 +1000
25382
25383 xfree86: bump the input ABI
25384
25385 The last few patches broke the ABI, bump it for convenience.
25386
25387 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25388
25389commit 86bb3781b336c09e4279136ed81974de5acdba7f
25390Author: Peter Hutterer <peter.hutterer@who-t.net>
25391Date: Fri Nov 4 11:29:01 2011 +1000
25392
25393 input: swap the server over to use the XI2mask struct
25394
25395 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25396 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25397
25398commit b8b90cd1610331ff12fa3f70bf372670af7795ec
25399Author: Peter Hutterer <peter.hutterer@who-t.net>
25400Date: Thu Nov 3 13:39:59 2011 +1000
25401
25402 Add a new XI2Mask struct and a few helper functions.
25403
25404 The current XI2 mask handling is handy for copying (fixed size arrays) but a
25405 pain to deal with otherwise. Add a struct for XI2 masks and the required
25406 accessors.
25407
25408 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25409 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25410
25411commit 4bc2761ad5ec2d0668aec639780ffb136605fbc8
25412Author: Peter Hutterer <peter.hutterer@who-t.net>
25413Date: Fri Nov 4 14:16:37 2011 +1000
25414
25415 dix: switch the dev->deviceGrab.activeGrab from GrabRec to GrabPtr
25416
25417 This breaks the input ABI.
25418
25419 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25420 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25421
25422commit b601ea769f1b8a4d7f19e9d4a13541c78e865fe5
25423Author: Peter Hutterer <peter.hutterer@who-t.net>
25424Date: Fri Nov 4 10:47:27 2011 +1000
25425
25426 dix: allocate temporary grabs on the heap
25427
25428 Once grabs start having nested memory locations, we can't just use the
25429 GrabRec on the stack anymore, we need to alloc/copy/free the grabs.
25430
25431 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25432 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25433
25434commit b0e9e2e32616d09c30a02b9d0ae9db0b13e150d1
25435Author: Peter Hutterer <peter.hutterer@who-t.net>
25436Date: Fri Nov 4 10:44:31 2011 +1000
25437
25438 dix: add CopyGrab() function
25439
25440 Not really needed at this point, but will be once touch support is added.
25441 Since grabs are now expected to be allocated/freed with AllocGrab and
25442 FreeGrab, CopyGrab must increase the refcount and duplicate the modifier
25443 masks. Until the callers are switched to use FreeGrab, this introduces
25444 memleaks.
25445
25446 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25447 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25448
25449commit 347f377f3b3f8c9d230d6309ec8ae92aa86d78b7
25450Author: Peter Hutterer <peter.hutterer@who-t.net>
25451Date: Thu Nov 3 16:12:09 2011 +1000
25452
25453 dix: add AllocGrab helper function
25454
25455 Not needed since the GrabRec is a self-contained struct but will be needed
25456 for the xi2 input mask rework.
25457 FreeGrab already exists, make it available to other callers.
25458
25459 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25460 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25461
25462commit 7af23259d88f4c28ed21140f82cc03b3724c06bb
25463Author: Peter Hutterer <peter.hutterer@who-t.net>
25464Date: Mon Nov 21 11:41:12 2011 -0800
25465
25466 dix: switch the syncEvent queue to a struct list
25467
25468 No effective functionality change, just cleanup to make this code slightly
25469 more sane.
25470
25471 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25472 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25473
25474commit 6369b59668613e51131857fbaeaa393a65a73048
25475Author: Gaetan Nadon <memsize@videotron.ca>
25476Date: Thu Dec 1 16:24:32 2011 -0500
25477
25478 test: update .gitignore with xfree86 and sort alphabetically
25479
25480 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25481 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
25482 Signed-off-by: Keith Packard <keithp@keithp.com>
25483
25484commit 3ab8ee32470f7cf1223e04238bb8c5f74ed52fc3
25485Merge: 22a666f 98c4a88
25486Author: Keith Packard <keithp@keithp.com>
25487Date: Wed Dec 7 12:42:17 2011 -0800
25488
25489 Merge remote-tracking branch 'airlied/reviewed-fixes'
25490
25491commit 22a666f9952feb7248e9bb2faf777edaaac8175f
25492Merge: 3824f55 e89b032
25493Author: Keith Packard <keithp@keithp.com>
25494Date: Wed Dec 7 12:27:23 2011 -0800
25495
25496 Merge remote-tracking branch 'alanc/master'
25497
25498commit 3824f558cc2ee051da8314c7bf08b8647b44e84a
25499Author: Keith Packard <keithp@keithp.com>
25500Date: Wed Dec 7 12:13:37 2011 -0800
25501
25502 hw/xfree86: fix segfault in config parser when config dir is missing
25503
25504 Treat a scandir error from a missing (or unusable) directory return as
25505 if it simply returned no files at all, which is what we want.
25506
25507 cc: Paulo Zanoni <przanoni@gmail.com>
25508 Signed-off-by: Keith Packard <keithp@keithp.com>
25509
25510commit 98c4a888a4428789386c7c47cecc81933b5999ba
25511Author: Dave Airlie <airlied@redhat.com>
25512Date: Mon Nov 28 16:37:59 2011 +0000
25513
25514 kdrive: drop screen crossing code.
25515
25516 The only kdrive server we probably care about anymore is Xephyr,
25517 and this screen enable/disable code totally breaks it in multi-screen mode.
25518
25519 When you are in one screen the other stops updating.
25520
25521 Fixes https://bugzilla.redhat.com/show_bug.cgi?id=757457
25522
25523 Signed-off-by: Dave Airlie <airlied@redhat.com>
25524 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
25525
25526commit 8d3731a811e33e263920dd7c8ec63d02968cb56e
25527Author: Dave Airlie <airlied@redhat.com>
25528Date: Thu Oct 20 10:48:26 2011 +0100
25529
25530 hal: free tmp_val in one missing case
25531
25532 Pointed out by coverity scan.
25533
25534 Signed-off-by: Dave Airlie <airlied@redhat.com>
25535 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
25536
25537commit 41229392b790f30a0f0ef1f4ed95647c5bca4001
25538Author: Dave Airlie <airlied@redhat.com>
25539Date: Thu Oct 20 11:00:43 2011 +0100
25540
25541 xv: test correct number of requests. (v2)
25542
25543 Pointed out by coverity.
25544
25545 v2: fix swapped as well, as pointed out by Alan
25546
25547 Signed-off-by: Dave Airlie <airlied@redhat.com>
25548 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25549
25550commit 1049139499d9132a20cd6d4d156fe9da9cddb6c2
25551Author: Dave Airlie <airlied@redhat.com>
25552Date: Wed Oct 19 16:57:13 2011 +0100
25553
25554 xaa: avoid possible freed pointer reuse in epilogue
25555
25556 If the pGCPriv->flags == 2, then we try to assign the freed pGCPriv->XAAOps
25557 avoid this by clearing the flags in to be destroyed pGCPriv.
25558
25559 Reported by coverity.
25560
25561 Signed-off-by: Dave Airlie <airlied@redhat.com>
25562 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
25563
25564commit 682c09a2cedd234b005334cc01247d859dd7f26a
25565Author: Dave Airlie <airlied@redhat.com>
25566Date: Wed Oct 19 16:22:31 2011 +0100
25567
25568 Xi: avoid overrun of callback array.
25569
25570 This code had an off-by-one and would allow writing one past the end of
25571 the callbacks array.
25572
25573 Pointed out by coverity.
25574
25575 Signed-off-by: Dave Airlie <airlied@redhat.com>
25576 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
25577
25578commit b62dc4fcbcffd10de16650bee284702c8608bb60
25579Author: Dave Airlie <airlied@redhat.com>
25580Date: Wed Oct 19 16:21:26 2011 +0100
25581
25582 xext: don't free uninitialised pointer when malloc fails. (v2)
25583
25584 Initialise the pAttr->values to values so if the values allocation
25585 fails it just ends up as free(NULL).
25586
25587 Pointed out by coverity.
25588
25589 v2: use Alan's suggestion.
25590
25591 Signed-off-by: Dave Airlie <airlied@redhat.com>
25592 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25593 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
25594
25595commit 22605effd188436629a0dbc688666549473741e4
25596Author: Adam Jackson <ajax@redhat.com>
25597Date: Thu Apr 28 13:34:28 2011 +1000
25598
25599 fbdevhw: iterate over all modes that match a mode. (v3)
25600
25601 So on RHEL5 anaconda sets an xorg.conf with a fixed 800x600 mode in it,
25602 we run radeonfb and fbdev since ati won't work in userspace due to domain
25603 issues in the older codebase.
25604
25605 On certain pseries blades the built-in KVM can't accept an 800x600-43 mode,
25606 it requires the 800x600-60 mode, so we have to have the kernel radeonfb
25607 driver reject the 800x600-43 mode when it sees it. However then fbdev
25608 doesn't try any of the other 800x600 modes in the modelist, and we end up
25609 getting a default 640x480 mode we don't want.
25610
25611 This patch changes the mode validation loop to continue on with the other modes
25612 that match to find one that works.
25613
25614 v2: move code around to avoid extra loop, after comment from Jamey.
25615 v3: move loop setup back into loop as per Jeremy's review.
25616
25617 Signed-off-by: Dave Airlie <airlied@redhat.com>
25618 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25619 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
25620
25621commit fac464e310b82fadcedf790798d1016c4805640b
25622Author: Peter Hutterer <peter.hutterer@who-t.net>
25623Date: Fri Dec 2 08:52:53 2011 +1000
25624
25625 include: rename "foos" to "list_of_foos" in the struct list documentation
25626
25627 Makes things a little easier to read.
25628
25629 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25630 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25631
25632commit 7dfe8c32a96d3f96c8aaeb2802f5b122e381a1e4
25633Author: Peter Hutterer <peter.hutterer@who-t.net>
25634Date: Fri Dec 2 08:51:24 2011 +1000
25635
25636 include: update struct list documentation to use one set of structs only
25637
25638 The example at the top of the file used a struct bar and a list of struct
25639 foos. Use those two throughout instead of a different struct foo for the
25640 examples and for the API documentation.
25641
25642 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25643 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25644
25645commit 18539c89eca8f6e272ead2b631365da530065ae7
25646Author: Peter Hutterer <peter.hutterer@who-t.net>
25647Date: Fri Dec 2 08:51:04 2011 +1000
25648
25649 include: a new list element does not need initialization, state so
25650
25651 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25652 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25653
25654commit 58dc73314b6508121ca094bbcf00612fe19ed69f
25655Author: Peter Hutterer <peter.hutterer@who-t.net>
25656Date: Fri Dec 2 08:43:45 2011 +1000
25657
25658 include: point to the tests in list.c as examples
25659
25660 Even with the documentation, the list.c tests are the best examples.
25661
25662 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25663 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25664
25665commit c8e451a8e719ba432bcfa8976774c07307087809
25666Author: Peter Hutterer <peter.hutterer@who-t.net>
25667Date: Thu Dec 1 14:12:11 2011 +1000
25668
25669 include: add list_last_entry to get the tail of a list
25670
25671 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25672 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25673
25674commit 6acebf9e1298939593b942ec91ae9ec9e74faa19
25675Author: Peter Hutterer <peter.hutterer@who-t.net>
25676Date: Thu Dec 1 13:35:50 2011 +1000
25677
25678 include: add list_append()
25679
25680 The existing list_add() prepends to the list, but in some cases we need the
25681 list ordered in the way we append the elements.
25682
25683 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25684 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25685
25686commit e89b0324da89ba5c0ba64af1ef46a12b7f55f879
25687Author: Gaetan Nadon <memsize@videotron.ca>
25688Date: Sat Dec 3 21:23:51 2011 -0500
25689
25690 autoconf: the minimum required level for autoconf is 2.60
25691
25692 The toolchain requirements are documented here:
25693 http://www.x.org/wiki/ModularDevelopersGuide#Required_Tools
25694
25695 Note that autoconf features only found in versions later
25696 than 2.60 must not be used.
25697
25698 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
25699 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25700 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25701
25702commit 427cb53fddf0c517d1fd1025a87b1104735edd2a
25703Author: Gaetan Nadon <memsize@videotron.ca>
25704Date: Sat Dec 3 21:23:21 2011 -0500
25705
25706 autoconf: remove redundant AC_PROG_MAKE_SET
25707
25708 Already included during Automake initialization.
25709
25710 After the patch, no change:
25711 configure:3893: checking whether make sets $(MAKE)
25712 configure:3915: result: yes
25713
25714 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
25715 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25716 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25717
25718commit c19c55a93ab1bc6db56075ca34d6e1c3f522f622
25719Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25720Date: Fri Dec 2 00:09:07 2011 -0800
25721
25722 Fix builds of Xnest & Xephyr with Solaris Studio compilers
25723
25724 Required in order to build with Studio cc now that xorg-macros is
25725 setting -errwarn=E_FUNC_HAS_NO_RETURN_STMT since a bug in the Solaris
25726 system headers causes the noreturn attribute to not be correctly
25727 applied to the exit() prototype in <stdlib.h> when building with
25728 Studio instead of gcc.
25729
25730 Otherwise compiler exits with errors:
25731 "Display.c", line 65: Function has no return statement : x_io_error_handler
25732 "hostx.c", line 341: Function has no return statement : x_io_error_handler
25733
25734 Uses Studio-specific pragma instead of adding another exit() prototype
25735 with a noreturn attribute to avoid causing gcc to warn about having
25736 a redundant prototype for the exit() function.
25737
25738 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25739 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
25740
25741commit e4dcf580f007d6adcf5b0c639865d7aaab1a319d
25742Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25743Date: Thu Dec 1 23:45:42 2011 -0800
25744
25745 LoaderOpen returns either a valid pointer or NULL, so don't check for < 0
25746
25747 Fixes Sun cc warning that was recently elevated to error by the
25748 stricter default CFLAGS changes to xorg-macros:
25749
25750 "loadmod.c", line 914: improper pointer/integer combination: op "<"
25751
25752 Should have been changed when commit ab7f057ce9df4e905b12 changed the
25753 LoaderOpen return type from int to void *.
25754
25755 Changes log message when file is found but dlopen() fails from:
25756 (EE) LoadModule: Module dbe does not have a dbeModuleData data object.
25757 (EE) Failed to load module "dbe" (invalid module, 0)
25758 to:
25759 (EE) Failed to load module "dbe" (loader failed, 7)
25760
25761 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25762 Reviewed-by: Adam Jackson <ajax@redhat.com>
25763
25764commit 8b6a75009746db4b8e2e180210ac097e35136bcd
25765Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25766Date: Wed Nov 30 23:01:34 2011 -0800
25767
25768 Fix gcc warnings about redundant declarations of fallback functions
25769
25770 Ensure ffs, strndup, strlcat, etc. aren't defined by our headers
25771 if they're already defined in the system headers.
25772
25773 This does export the HAVE_FFS, HAVE_STRNDUP, etc. definitions to drivers,
25774 but if you built the Xserver with a libc that had those, and then build
25775 the drivers with a less capable libc, you're going to have problems anyway,
25776 and this should solve some reported problems with conflicts between our
25777 strndup definition and gcc magic for it.
25778
25779 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25780 Reviewed-by: Adam Jackson <ajax@redhat.com>
25781
25782commit d829a7c5cb42c979b58f3547136df5b05d906423
25783Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25784Date: Wed Nov 30 22:20:09 2011 -0800
25785
25786 Move to autoconf standard function name checks & defines
25787
25788 Replace multiple methods of checking for functions with AC_CHECK_FUNCS
25789 Replace multiple methods of selecting fallback funcs with AC_REPLACE_FUNCS
25790 Replace HAS_* and NEED_* #defines with autogenerated HAVE_*
25791
25792 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25793 Reviewed-by: Adam Jackson <ajax@redhat.com>
25794 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
25795 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
25796
25797commit 4be68b03128e958d2a6dc1b7feb3587329b9561b
25798Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25799Date: Wed Nov 30 19:39:51 2011 -0800
25800
25801 Remove unused function checks from configure.ac & include/*.h.in
25802
25803 The code that used getisax to check for MMX support was moved to pixman
25804 and removed from the X server by commit eb2d7fe02f9cbc.
25805
25806 The code that used HAVE_MKSTEMP was deleted by the Xprint removal in
25807 commit 1c8bd318fbaf.
25808
25809 All alloca calls were removed by the patch series end in commit 5e363500c8,
25810 and used custom X checks instead of the autoconf HAVE_ALLOCA anyway.
25811
25812 I can find no record of HAVE_GETUID, HAVE_GETEUID, HAVE_LINK, HAVE_MEMMOVE,
25813 HAVE_MEMSET, HAVE_STRCHR, HAVE_STRRCHR, HAVE_GETOPT, HAVE_GETOPT_LONG,
25814 HAVE_DOPRNT, or HAVE_VPRINTF ever being used, and the calls to those
25815 functions are not wrapped in #ifdefs.
25816 (Most of those are in our baseline requirements of C89 & Unix98 anyway.)
25817
25818 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25819 Reviewed-by: Adam Jackson <ajax@redhat.com>
25820 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
25821 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
25822
25823commit 84207def93e4489df8dca8e044ce544be8a3f845
25824Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25825Date: Mon Nov 28 20:38:58 2011 -0800
25826
25827 Disable building of tests requiring DDX functions when Xorg is not built
25828
25829 Some test cases require linking with some sort of DDX - ideally we'd
25830 have a stub ddx for testing, but for now, since we link with the Xorg
25831 ddx, disable those tests when configured with --disable-xorg
25832
25833 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=43320
25834
25835 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25836 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
25837 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
25838
25839commit fd976e4051e15ab2d01ec0bf89ff26926d4ed04e
25840Author: Gaetan Nadon <memsize@videotron.ca>
25841Date: Thu Dec 1 16:24:12 2011 -0500
25842
25843 config: conftest needs an include directive to locate X11/Xfuncproto.h
25844
25845 The compiler option -fvisibility=hidden is erroneously missing
25846 due to a faulty configuration test. The gcc command is unable to
25847 locate X11/Xfuncproto.h unless the build occurs on a
25848 system where X11 development headers are installed.
25849
25850 configure:21294: checking for symbol visibility support
25851 configure:21323: gcc -std=gnu99 -c -g -O2 -fvisibility=hidden conftest.c >&5
25852 conftest.c:144:28: fatal error: X11/Xfuncproto.h: No such file or directory
25853
25854 The solution is to add an include directive to obtain the location
25855 of X11/Xfuncproto.h which may or may not be the system installed headers.
25856
25857 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25858 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
25859 Signed-off-by: Keith Packard <keithp@keithp.com>
25860
25861commit feebf6746374aa04b12e9e3e51313a3a82c03530
25862Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25863Date: Wed Nov 23 00:30:02 2011 -0800
25864
25865 Limit the number of screens Xvfb will attempt to allocate memory for
25866
25867 Commit f9e3a2955d2ca7 removing the MAXSCREEN limit left the screen
25868 number too unlimited, and allowed any positive int for a screen number:
25869
25870 Xvfb :1 -screen 2147483647 1024x1024x8
25871
25872 Fatal server error:
25873 Not enough memory for screen 2147483647
25874
25875 Found by Parfait 0.3.7:
25876 Error: Integer overflow (CWE 190)
25877 Integer parameter of memory allocation function realloc() may overflow due to multiplication with constant value 1112
25878 at line 293 of hw/vfb/InitOutput.c in function 'ddxProcessArgument'.
25879
25880 Since the X11 connection setup only has a CARD8 for number of SCREENS,
25881 limit to 255 screens, which is also low enough to avoid overflow on the
25882 sizeof(*vfbScreens) * (screenNum + 1) calculation for realloc.
25883
25884 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25885 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25886
25887commit fb22a408c69a84f81905147de9e82cf66ffb6eb2
25888Author: Keith Packard <keithp@keithp.com>
25889Date: Tue Nov 8 10:13:15 2011 -0800
25890
25891 Save major/minor opcodes in ClientRec for RecordAReply
25892
25893 The record extension needs the major and minor opcodes in the reply
25894 hook, but the request buffer may have been freed by the time the hook
25895 is invoked. Saving the request major and minor codes as the request is
25896 executed avoids fetching from the defunct request buffer.
25897
25898 This patch also eliminates the public MinorOpcodeOfRequest function,
25899 inlining it into Dispatch. Usages of that function have been replaced
25900 with direct access to the new ClientRec field.
25901
25902 Signed-off-by: Keith Packard <keithp@keithp.com>
25903 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25904
25905commit 05f09354a30a4f5edd421220e1aa97be754c71bb
25906Merge: 3197b77 08ec4da
25907Author: Keith Packard <keithp@keithp.com>
25908Date: Thu Dec 1 14:44:52 2011 +0000
25909
25910 Merge remote-tracking branch 'jcristau/for-keith'
25911
25912commit 3197b773c81e58cd4aa412eccaee49526012ed6f
25913Merge: 2dc5ba4 dfcec1d
25914Author: Keith Packard <keithp@keithp.com>
25915Date: Thu Dec 1 14:36:58 2011 +0000
25916
25917 Merge remote-tracking branch 'whot/for-keith'
25918
25919commit 2dc5ba4a1b127e3ed286718ab0a6c20438361192
25920Author: Matt Turner <mattst88@gmail.com>
25921Date: Tue Nov 29 23:52:19 2011 -0500
25922
25923 Remove another if (E != NULL) check around free(E)
25924
25925 I wonder if there are any other patterns we haven't seen yet?
25926
25927 Reviewed-by: Jamey Sharp <jamey@minilop.net>
25928 Signed-off-by: Matt Turner <mattst88@gmail.com>
25929 Signed-off-by: Keith Packard <keithp@keithp.com>
25930
25931commit dfcec1d3f9d7bac5cde9eb034341428ee0ad3728
25932Author: Peter Hutterer <peter.hutterer@who-t.net>
25933Date: Fri Nov 11 08:55:14 2011 +1000
25934
25935 test: remove unneeded printf statements from misc.c
25936
25937 Leftover from debugging, is not really needeed in a test.
25938
25939 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25940 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25941
25942commit 9c38422fc4cf853c3104fada2a3851c79df2a66f
25943Author: Peter Hutterer <peter.hutterer@who-t.net>
25944Date: Wed Nov 9 14:37:26 2011 +1000
25945
25946 include: add BUG_WARN macro for internal bug cases.
25947
25948 There are plenty of cases that can only be triggered by a real bug in the
25949 server and doing the ErrorF dance manually everywhere is a tad painful and
25950 the error message is usually used only to find the spot in the file anyway.
25951
25952 Plus, reading BUG_WARN somewhere is a good indicator to the casual reader
25953 that this isn't intended behaviour.
25954
25955 Note that this is intentionally different to the BUG_ON behaviour on the
25956 kernel, we do not FatalError the server. It's just a warning + stacktrace.
25957 If the bug is really fatal, call FatalError.
25958
25959 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25960 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25961 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
25962
25963commit 61ef4daf6450da573b9de72ba7b200566821b916
25964Author: Peter Hutterer <peter.hutterer@who-t.net>
25965Date: Fri Nov 4 15:49:23 2011 +1000
25966
25967 Xi: add FreeInputMask function
25968
25969 Does what it says on the box, complements MakeInputMask.
25970
25971 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25972 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25973
25974commit ee9346bb31efce4036df1b8dd8f1a5dc43ae955a
25975Author: Peter Hutterer <peter.hutterer@who-t.net>
25976Date: Thu Nov 3 15:45:34 2011 +1000
25977
25978 Xi: add helper functions to alloc/free InputClientPtrs
25979
25980 Currently not needed since the InputClientRec is a self-contained struct. As
25981 part of the touch rework that won't be the case in the future and a function
25982 to allocate/free memory appropriately is required.
25983
25984 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25985 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25986
25987commit 4acf999c294868a44e559d212c6d88a4978256b2
25988Author: Peter Hutterer <peter.hutterer@who-t.net>
25989Date: Fri Nov 4 15:37:32 2011 +1000
25990
25991 dix: use a single return statement in CheckPassiveGrabsOnWindow
25992
25993 No functional change.
25994
25995 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
25996 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
25997
25998commit b371795f01c1d7fc338cfe18b8a567ed9f402846
25999Author: Peter Hutterer <peter.hutterer@who-t.net>
26000Date: Thu Nov 3 14:54:06 2011 +1000
26001
26002 dix: rename GetWindowXI2Mask to WindowXI2MaskIsset
26003
26004 And let it return a boolean value, that's all the callers need anyway.
26005
26006 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26007 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
26008
26009commit bedb8fd90de8e2db33d5e3b1d859f24bf34efc9a
26010Author: Peter Hutterer <peter.hutterer@who-t.net>
26011Date: Thu Nov 3 14:25:51 2011 +1000
26012
26013 Xi: use single return code from XIPassiveGrabDevice
26014
26015 Some failures returned status but the actual return code was "ret". Use
26016 "ret" consistently and move status to the local block is used in.
26017
26018 [the goto isn't necessary yet, but for a future patch]
26019
26020 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26021 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
26022
26023commit 2aad1a2b42b7def7812abfa2462b6bcc6382e03a
26024Author: Peter Hutterer <peter.hutterer@who-t.net>
26025Date: Wed Nov 2 14:07:19 2011 +1000
26026
26027 include: fix mask size calculation
26028
26029 Same bug as inputproto-2.0.1-9-gb1149ab, if the XI2LASTEVENT was a multiple
26030 of 8, the mask was one bit too short.
26031
26032 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26033 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26034 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
26035
26036commit b2015a2c01711646bb7ae23d684abee0cd55d4d0
26037Author: Jeremy Huddleston <jeremyhu@apple.com>
26038Date: Thu Nov 24 13:54:08 2011 -0800
26039
26040 dmx: Build fix for -Werror=implicit-function-declaration on linux
26041
26042 Fixes regression introduced by: 6e6d732bac3c21cb85f8e998908f9b393630e5f8
26043 Found-by: Tinderbox
26044
26045 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
26046 CC: Alan Coopersmith <alan.coopersmith@oracle.com>
26047 Signed-off-by: Keith Packard <keithp@keithp.com>
26048
26049commit 873a1ace3646994adf95961f48719e95dcade7a2
26050Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
26051Date: Tue Nov 1 10:57:56 2011 -0200
26052
26053 parser: free val.str after xstrtokenize
26054
26055 After we tokenize val.str, we discard it.
26056
26057 This is just one example:
26058 6 bytes in 1 blocks are definitely lost in loss record 24 of 652
26059 at 0x4C2779D: malloc (in vgpreload_memcheck-amd64-linux.so)
26060 by 0x4D744D: xf86getToken (scan.c:400)
26061 by 0x4D75F1: xf86getSubToken (scan.c:462)
26062 by 0x4DB060: xf86parseInputClassSection (InputClass.c:145)
26063 by 0x4D664C: xf86readConfigFile (read.c:184)
26064 by 0x490556: xf86HandleConfigFile (xf86Config.c:2360)
26065 by 0x49AA77: InitOutput (xf86Init.c:365)
26066 by 0x425A7A: main (main.c:204)
26067
26068 Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26069 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
26070 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
26071
26072commit d41987d77c903e00cca7bcf3e04ed07151e3bb4d
26073Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
26074Date: Tue Nov 1 10:42:36 2011 -0200
26075
26076 parser: free val.str after xf86getBoolValue
26077
26078 After we convert the value to a boolean, we discard the string.
26079
26080 This is just one example:
26081
26082 3 bytes in 1 blocks are definitely lost in loss record 5 of 657
26083 at 0x4C2779D: malloc (vgpreload_memcheck-amd64-linux.so)
26084 by 0x4D744D: xf86getToken (scan.c:400)
26085 by 0x4D75F1: xf86getSubToken (scan.c:462)
26086 by 0x4DB3E0: xf86parseInputClassSection (InputClass.c:189)
26087 by 0x4D664C: xf86readConfigFile (read.c:184)
26088 by 0x490556: xf86HandleConfigFile (xf86Config.c:2360)
26089 by 0x49AA77: InitOutput (xf86Init.c:365)
26090 by 0x425A7A: main (main.c:204)
26091
26092 Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26093 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
26094 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
26095
26096commit d5c7338b3eaea55177ade6fcba71a47ccd5547f5
26097Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
26098Date: Mon Oct 31 17:54:03 2011 -0200
26099
26100 parser: free scandir's list
26101
26102 v2: move the free()s to the function that calls scandir
26103
26104 80 bytes in 1 blocks are definitely lost in loss record 411 of 631
26105 at 0x4C2779D: malloc (vgpreload_memcheck-amd64-linux.so)
26106 by 0x4C27927: realloc (vgpreload_memcheck-amd64-linux.so)
26107 by 0x696A80D: scandir (scandir.c:108)
26108 by 0x4D8828: OpenConfigDir (scan.c:854)
26109 by 0x4D8A43: xf86openConfigDirFiles (scan.c:952)
26110 by 0x49031F: xf86HandleConfigFile (xf86Config.c:2327)
26111 by 0x49A9E3: InitOutput (xf86Init.c:365)
26112 by 0x425A7A: main (main.c:204)
26113
26114 Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26115 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
26116
26117commit 3d635fe84d6de53e2f74203b10e89f7851fe3fc1
26118Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
26119Date: Sun Oct 30 18:04:59 2011 -0200
26120
26121 Correctly free config file names
26122
26123 We call xf86penConfigDirFiles twice, so we overwrite the configDirPath
26124 variable, losing the pointer. If we move the pointer management to the
26125 upper layer (the function callers), they will be able to call these
26126 functions as many times as they want, but they'll have to free those
26127 returned values.
26128
26129 v2: don't leak inside XWin
26130
26131 4,097 bytes in 1 blocks are definitely lost in loss record 625 of 632
26132 at 0x4C2779D: malloc (in vgpreload_memcheck-amd64-linux.so)
26133 by 0x4D7899: DoSubstitution (scan.c:615)
26134 by 0x4D87B0: OpenConfigDir (scan.c:845)
26135 by 0x4D8A2D: xf86openConfigDirFiles (scan.c:955)
26136 by 0x49031F: xf86HandleConfigFile (xf86Config.c:2327)
26137 by 0x49A9BF: InitOutput (xf86Init.c:365)
26138 by 0x425A7A: main (main.c:204)
26139
26140 Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26141 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
26142
26143commit 0ae087e13192d9b498db782be5ba49ca91a81547
26144Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
26145Date: Sat Oct 29 18:33:24 2011 -0200
26146
26147 glx: don't leak driConfigs
26148
26149 For dri, dri2 and driswrast.
26150
26151 12,968 (584 direct, 12,384 indirect) bytes in 1 blocks are definitely lost in loss record 569 of 570
26152 at 0x4C2779D: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
26153 by 0x7821E3B: driConcatConfigs (utils.c:560)
26154 by 0x7827CF2: dri_fill_in_modes (dri_screen.c:224)
26155 by 0x782831E: dri_init_screen_helper (dri_screen.c:405)
26156 by 0x7826C03: drisw_init_screen (drisw.c:266)
26157 by 0x782225F: driCreateNewScreen (drisw_util.c:69)
26158 by 0x4826E2: __glXDRIscreenProbe (glxdriswrast.c:451)
26159 by 0x4812FA: GlxExtensionInit (glxext.c:327)
26160 by 0x41FB14: InitExtensions (miinitext.c:471)
26161 by 0x568622: main (main.c:208)
26162
26163 Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26164 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26165
26166commit d26fae246d7c451b4d5ffe24fdb959d4bd00b107
26167Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
26168Date: Tue Oct 25 14:56:35 2011 -0200
26169
26170 glx: don't leak fbconfigs
26171
26172 29,952 (208 direct, 29,744 indirect) bytes in 1 blocks are definitely lost in loss record 573 of 573
26173 at 0x4C2779D: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
26174 by 0x4829BC: createModeFromConfig (glxdricommon.c:131)
26175 by 0x482C09: glxConvertConfigs (glxdricommon.c:185)
26176 by 0x482788: __glXDRIscreenProbe (glxdriswrast.c:468)
26177 by 0x4812FA: GlxExtensionInit (glxext.c:327)
26178 by 0x41FB14: InitExtensions (miinitext.c:471)
26179 by 0x568636: main (main.c:208)
26180
26181 Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26182 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26183
26184commit 305a8bcb2fc6f87f8d891fcb774198b6fd118ee1
26185Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
26186Date: Thu Oct 20 17:39:29 2011 -0200
26187
26188 glx: fix memory leak when destroying screen
26189
26190 1,152 bytes in 1 blocks are definitely lost in loss record 536 of 575
26191 at 0x4C25E84: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
26192 by 0x483820: __glXScreenInit (glxscreens.c:357)
26193 by 0x48271C: __glXDRIscreenProbe (glxdriswrast.c:469)
26194 by 0x4812BE: GlxExtensionInit (glxext.c:327)
26195 by 0x41FB14: InitExtensions (miinitext.c:471)
26196 by 0x5685AE: main (main.c:208)
26197
26198 Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26199 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26200
26201commit f405dfffe7d5db81d398615a80acbeba7e014ada
26202Author: Jeremy Huddleston <jeremyhu@apple.com>
26203Date: Mon Nov 7 22:14:30 2011 +0000
26204
26205 dmx: Build fix for -Werror=implicit-function-declaration
26206
26207 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
26208 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26209 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26210
26211commit 632d205b309d0515b4ae0f9902059aa4b093a313
26212Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26213Date: Mon Nov 7 20:24:08 2011 -0800
26214
26215 Fix gcc -Wwrite-strings warnings in xf86Modes code
26216
26217 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26218 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26219
26220commit 09e4b78f7944234865f4648940453303af0c2663
26221Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26222Date: Mon Nov 7 19:19:47 2011 -0800
26223
26224 Fix gcc -Wwrite-strings warnings in xf86 ddx
26225
26226 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26227 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26228
26229commit 8e4556f560487f3d07812834d4785441d0a8323c
26230Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26231Date: Mon Nov 7 20:09:47 2011 -0800
26232
26233 FindModule: stop copying const char *dirname to char *dirpath
26234
26235 Not needed since 6cf844ab69926b split out the allocation/manipulation
26236 into the helper function, leaving FindModule just copying the pointer
26237 around, and causing gcc warnings and an unreachable call to free.
26238
26239 Also no longer need to store the combined strlen results in dirlen.
26240
26241 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26242 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26243
26244commit 05f589d464a961aea8d25632a390fb66a06cd186
26245Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26246Date: Fri Nov 4 23:21:34 2011 -0700
26247
26248 Fix gcc -Wwrite-strings warnings in various extensions
26249
26250 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26251 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26252
26253commit 232f1ddf3d060f3ce9d2ebd35f33b1294cac380e
26254Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26255Date: Thu Nov 3 21:21:19 2011 -0700
26256
26257 Fix gcc -Wwrite-strings warnings in XkbGetRulesDflts
26258
26259 Stop temporarily storing a pointer to a constant literal string
26260 in a char *, just to strdup it a few lines later.
26261
26262 Fixes gcc -Wwrite-strings warnings:
26263
26264 xkbInit.c: In function 'XkbGetRulesDflts':
26265 xkbInit.c:121:38: warning: assignment discards qualifiers from pointer target type
26266 xkbInit.c:123:23: warning: assignment discards qualifiers from pointer target type
26267 xkbInit.c:125:24: warning: assignment discards qualifiers from pointer target type
26268 xkbInit.c:127:25: warning: assignment discards qualifiers from pointer target type
26269 xkbInit.c:129:25: warning: assignment discards qualifiers from pointer target type
26270
26271 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26272 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26273
26274commit 0bc41d5f8d1a6192f57d9bf646cc46ed839b06b9
26275Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26276Date: Mon Nov 7 18:08:21 2011 -0800
26277
26278 Remove redundant redeclarations of functions in the same header file
26279
26280 Exposed by recent addition of -Wredundant-decls to default CWARNFLAGS
26281
26282 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26283 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26284
26285commit 285133a35eede03e37f41aeea6cbfd15a0800d98
26286Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26287Date: Mon Nov 7 15:44:26 2011 -0800
26288
26289 sun_agp: cast key to uintptr_t before casting to (int *)
26290
26291 Matches what linux_agp already does and prevents gcc from throwing up:
26292
26293 sun_agp.c: In function 'xf86DeallocateGARTMemory':
26294 sun_agp.c:236:40: error: cast to pointer from integer of different size
26295
26296 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26297 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26298
26299commit f8dd5efb673439e7e9f7d82e6f9cfb97afacc85a
26300Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26301Date: Thu Nov 3 21:18:56 2011 -0700
26302
26303 Mark XKB char * as const to clean up gcc -Wwrite-strings warnings
26304
26305 Cleans up around 120 warnings from this set
26306
26307 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26308 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26309
26310commit cccafabd56d0f8e5784ea0ad9fdc03224f952bfe
26311Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26312Date: Thu Nov 3 21:10:51 2011 -0700
26313
26314 Mark arguments to fopen/popen/system wrappers as const char *
26315
26316 Silencing more gcc -Wwrite-strings warnings
26317
26318 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26319 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26320
26321commit 3d0ece5e8467d823afb227b7a7f8b12b906b6ba9
26322Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26323Date: Thu Nov 3 20:44:43 2011 -0700
26324
26325 Reduce unnecessary string copying in xkbtext routines
26326
26327 Instead of using sprintf to copy a static string to a local buffer,
26328 just to pass it to TryCopyStr, pass the static string to TryCopyStr
26329 directly, as is already done in other parts of this code.
26330
26331 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26332 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26333
26334commit 05d8a7f7a785eff3292f0f0537bb3902930f1b5c
26335Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26336Date: Sat Oct 29 11:08:17 2011 -0700
26337
26338 Convert a bunch of sprintf to snprintf calls
26339
26340 This batch is the straightforward set - others are more complex and
26341 need more analysis to determine right size to pass.
26342
26343 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26344 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26345
26346commit b967bf2af264726042e2f6ffb9ca7d234f34b56b
26347Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26348Date: Mon Oct 31 23:01:35 2011 -0700
26349
26350 Remove xf86FormatPciBusNumber from API, inline the one place its used
26351
26352 Found no calls from current driver modules
26353
26354 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26355 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26356
26357commit 6450f6ca7ee070da3b6d70c2d3a6977f018ac421
26358Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26359Date: Thu Nov 3 22:13:05 2011 -0700
26360
26361 Move DoShowOptions to xf86Configure.c, delete xf86ShowOpts.c
26362
26363 Gets rid of duplicate static copy of optionTypeToString by putting
26364 both callers of that helper function in the same source file.
26365
26366 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26367 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26368 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26369
26370commit 43fa1274263f76faaca995f6e498bc3179a857b8
26371Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26372Date: Mon Oct 31 22:17:35 2011 -0700
26373
26374 Remove bad code from DoShowOptions (Xorg -showopts handler)
26375
26376 When we want to print a string, it's okay to just print it.
26377 We don't need to first allocate a buffer 2 bytes bigger than the
26378 string, copy the entire string unmodified to the buffer, print the
26379 buffer, and then leak the buffer (though we AbortDDX 8 lines later,
26380 and then just in case we survived that, call exit as well, so the
26381 leak is short lived, just oh so pointless).
26382
26383 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26384 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26385 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26386
26387commit 71a89d97332cc181becc5a5f73166e642f96c076
26388Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26389Date: Mon Oct 31 21:36:47 2011 -0700
26390
26391 Convert glx/single2.c:DoGetString() to use asprintf()
26392
26393 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26394 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26395
26396commit f3cb512dc4daaeed389bb4740e21b6e2330e01e1
26397Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26398Date: Sat Oct 29 11:19:44 2011 -0700
26399
26400 LogVMessageVerb: Fix const mismatch warning
26401
26402 "log.c", line 382: warning: assignment type mismatch:
26403 pointer to char "=" pointer to const char
26404
26405 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26406 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26407
26408commit 3d2d88029b29d6e1c53220ad275ba8ba2dedd89e
26409Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26410Date: Sat Oct 29 11:13:32 2011 -0700
26411
26412 AuthAudit: clean up string handling calls
26413
26414 The extra "out" pointer to redirect writes to the array isn't needed since
26415 the removal of LBX (commit a9ed5a87902a), and eliminating it allows more
26416 logical use of sizeof(addr) in length-checked strlcpy & snprintf calls to
26417 write to it.
26418
26419 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26420 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26421
26422commit 615f93a3d03d40924365061c6ae242240dd0ab7e
26423Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26424Date: Tue Nov 1 14:57:41 2011 -0700
26425
26426 Remove unnecessary variable rtrn in XkbKeysymText
26427
26428 Also removes even more unnecessary use of variable assignment inside
26429 function arguments.
26430
26431 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26432 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26433
26434commit 5f285a30a1d6ffba82ebe5e08a0b68352bb51556
26435Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26436Date: Tue Nov 1 16:56:18 2011 -0700
26437
26438 Make XIGetKnownProperty take a const char * argument
26439
26440 Now that MakeAtom takes const char *, so can XIGetKnownProperty.
26441 Clears 71 warnings from gcc -Wwrite-strings of the form:
26442 devices.c:145:5: warning: passing argument 1 of 'XIGetKnownProperty' discards qualifiers from pointer target type
26443 ../include/exevents.h:128:23: note: expected 'char *' but argument is of type 'const char *'
26444
26445 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26446 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
26447
26448commit e0f3633632de609e60950aef07c82df534db7888
26449Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26450Date: Fri Oct 28 21:31:46 2011 -0700
26451
26452 Convert ProcRenderQueryFilters to use memcpy instead of strncpy
26453
26454 We just got the string length with strlen, might as well use it
26455 to copy the whole string quickly instead of checking each character
26456 a second time to see if it's 0 or not.
26457
26458 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26459 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26460
26461commit 780133f9ae7fada462714b47e79d26075bbd9abe
26462Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26463Date: Fri Oct 28 21:29:50 2011 -0700
26464
26465 Convert DetermineClientCmd to use strdup instead of malloc+strncpy
26466
26467 *cmdname is initialized to NULL earlier in the function, so it's
26468 okay to overwrite it with NULL if strdup fails, don't need that
26469 extra check.
26470
26471 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26472 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26473
26474commit 03ddca6f71339fad089c56484bf35c63642ae1be
26475Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26476Date: Fri Oct 28 21:25:20 2011 -0700
26477
26478 Convert dmxSetDefaultFontPath to use strdup instead of malloc+strncpy
26479
26480 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26481 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26482
26483commit d9243777c77d9b2992fbaf7f459430283837e323
26484Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26485Date: Sat Oct 29 10:32:52 2011 -0700
26486
26487 matchDriverFromFiles: use one snprintf instead of strncpy/cat series
26488
26489 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26490 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26491
26492commit 6e6d732bac3c21cb85f8e998908f9b393630e5f8
26493Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26494Date: Fri Oct 28 21:18:46 2011 -0700
26495
26496 Convert strncpy/strncat to strlcpy/strlcat
26497
26498 As long as we're carrying around a compatibility copy in os/strl*.c,
26499 might as well use them.
26500
26501 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26502 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26503
26504commit e189dbb3e57d30eb96034d4ce9544ce7a93a371e
26505Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26506Date: Fri Oct 28 20:19:44 2011 -0700
26507
26508 Convert AllocXTestDevice to use asprintf
26509
26510 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26511 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26512
26513commit 08093c25a91c07ab8af7cece9bba738b827cfd1b
26514Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26515Date: Mon Oct 24 23:16:30 2011 -0700
26516
26517 Convert some malloc + strncpy pairs into strndup calls
26518
26519 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26520 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26521
26522commit acde97a39d35bfb03af2614c68176ad9afb71f53
26523Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26524Date: Mon Oct 24 20:28:32 2011 -0700
26525
26526 Add fallback implementation of strndup()
26527
26528 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26529 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26530 Linux test code fixed by: Keith Packard <keithp@keithp.com>
26531
26532commit ed38c2648cf7cc04c1d03f8d14375815f6cf536e
26533Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26534Date: Fri Oct 28 14:32:56 2011 -0700
26535
26536 Fix Sun compiler check that got turned around
26537
26538 Since the check is for !(compilers that support __builtin_constant_p)
26539 it needs to be !(gcc or new enough Sun cc), but was written as
26540 !(gcc or too old Sun cc).
26541
26542 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26543 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26544
26545commit 7ee7fd1f4c72b2ab6dba0413e63dd2e8b95b2112
26546Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26547Date: Mon Oct 24 20:39:24 2011 -0700
26548
26549 Remove a couple Error() instances left behind by 09dbfcb0ad7b6c8
26550
26551 Two instances found in the SIOCGIFCONF code for listing network interfaces.
26552
26553 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26554 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26555 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
26556
26557commit e47aa9475027ed6a255daefec85561c6b15789bd
26558Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26559Date: Mon Oct 24 23:40:45 2011 -0700
26560
26561 Enable memory checking during unit testing
26562
26563 If configure is generated with xorg-macros 1.16.0 or newer, and
26564 an appropriate memory checking library is found for the platform,
26565 use it when running unit tests. If not, then no harm is done.
26566
26567 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26568 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26569
26570commit 36670065bd74b870d0da7c6a69a9c0d222b21706
26571Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26572Date: Tue Oct 25 00:12:48 2011 -0700
26573
26574 Don't require ld -wrap for tests that don't need it
26575
26576 If configure is generated with xorg-macros 1.16 or later, this allows
26577 builders to --enable-unit-test and run the tests other than the xi2
26578 tests which require ld -wrap (and are still wrapped in if HAVE_LD_WRAP
26579 in tests/xi2/Makefile). If an older xorg-macros is used, the existing
26580 behaviour is preserved of requiring ld -wrap for all unit tests, but
26581 no side effects occur, so the minimum xorg-macros version is not raised.
26582
26583 If unit testing is enabled without ld -wrap, then we create a bogus
26584 "xi2-tests" script just to report that we're skipping them.
26585
26586 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26587 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26588
26589commit 0e6b88db7f2e94ccc8153a0e002d176440914e01
26590Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26591Date: Mon Aug 1 14:09:33 2011 -0700
26592
26593 Don't fallback to wsfb or fbdev on Solaris
26594
26595 We don't ship either one, so don't waste time and make confusing log
26596 entries trying to load them.
26597
26598 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26599 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26600
26601commit 7b0f53f0a5824fcd81ef505beb3e19ac7c77961f
26602Author: Alan Coopersmith <alan.coopersmith@oracle.com>
26603Date: Wed Nov 23 10:50:13 2011 -0800
26604
26605 Fix Xdmx build on Linux to work with strlcpy changes
26606
26607 Include strlcpy.c in the libdmxconfig.a library with the other functions
26608 shared among the xdmx configuration programs.
26609
26610 Also add a #include "os.h" to the scanner.l file that now calls strlcpy
26611 to include the prototype from $(top_srcdir)/include/os.h.
26612
26613 (To be squashed into
26614 http://cgit.freedesktop.org/~alanc/xserver/commit/?id=c19f0ff5223d428f8ad2ab3c563c974c96a521ba
26615 before next PULL request to avoid breaking bisection.)
26616
26617 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26618 Signed-off-by: Keith Packard <keithp@keithp.com>
26619
26620commit 922c1d817091c7eedfb701f4d9eb4324e22ccca4
26621Author: Gaetan Nadon <memsize@videotron.ca>
26622Date: Fri Nov 11 20:44:19 2011 -0500
26623
26624 docs: spell "X Server Version" consistently in titles. Add where missing.
26625
26626 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
26627 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
26628 Signed-off-by: Keith Packard <keithp@keithp.com>
26629
26630commit 5952efa903d57ac69070f1477ce46223a2a7893c
26631Author: Keith Packard <keithp@keithp.com>
26632Date: Sun Nov 20 13:07:33 2011 -0800
26633
26634 Bump version to 1.11.99.1 (1.12 development release 1)
26635
26636 Signed-off-by: Keith Packard <keithp@keithp.com>
26637
26638commit 795fbe3e6e440f023e52c452106f4c266c13d36c
26639Author: Keith Packard <keithp@keithp.com>
26640Date: Sun Nov 20 13:33:12 2011 -0800
26641
26642 Add 'optionstr.h' to include/Makefile.am
26643
26644 This makes sure it gets distributed
26645
26646 Signed-off-by: Keith Packard <keithp@keithp.com>
26647
26648commit 08ec4da6a7617637914bee1636858474213cf39d
26649Author: Julien Cristau <jcristau@debian.org>
26650Date: Sat Nov 5 13:00:07 2011 +0100
26651
26652 Fix ShmPutImage for XYBitmap
26653
26654 We can't call CopyArea in that case because the image has depth 1, which
26655 might not match the target drawable, so we might overrun the shm
26656 segment. Commit 11817a881cb93a89788105d1e575a468f2a8d27c apparently
26657 fixed a similar bug for XYPixmap, but missed the bitmap case.
26658
26659 Fixes: http://bugs.debian.org/629611
26660
26661 Thanks to Alan Curry for diagnosing this and providing a test case.
26662
26663 Reported-and-tested-by: Alan Curry <pacman@kosh.dhis.org>
26664 Reviewed-by: Peter Harris <pharris@opentext.com>
26665 Signed-off-by: Julien Cristau <jcristau@debian.org>
26666
26667commit 328074890eeb111950e984c6f618311983600b20
26668Author: dtakahashi42 <dtakahashi42@gmail.com>
26669Date: Fri Nov 18 11:30:22 2011 -0800
26670
26671 rootless: Fix a server crash when choosing a color with the gimp color wheel
26672
26673 https://trac.macports.org/ticket/30927
26674
26675 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
26676 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26677
26678commit 58864146fbdf1820d04825838691e84784ef91bc
26679Author: Ross Burton <ross@linux.intel.com>
26680Date: Wed Sep 28 11:46:02 2011 +0100
26681
26682 edid: Add quirk for Acer Aspire One 110
26683
26684 At least one revision of the AAO reports a 190x110mm maximum size but a
26685 451x113mm mode.
26686
26687 X.Org Bug 41141 <https://bugs.freedesktop.org/show_bug.cgi?id=41141>
26688
26689 Signed-off-by: Ross Burton <ross@linux.intel.com>
26690 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26691 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26692
26693commit 34b0e4eee911f8b09a3682a7f1b4c8598ef48b8d
26694Author: Chris Wilson <chris@chris-wilson.co.uk>
26695Date: Thu Aug 25 16:04:04 2011 +0100
26696
26697 dri2: Register the DRI2DrawableType after server regeneration
26698
26699 The Resource database is reset upon regeneration and so the dri2 module
26700 needs to re-register its RESTYPE for the drawable or else it will
26701 clobber the next unsuspecting user of the database. Fortunately, DRI2 is
26702 loaded late in the initialisation sequence and was last up until
26703 xf86-video-intel started using the Resource database to track
26704 outstanding swaps...
26705
26706 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
26707 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26708 Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
26709
26710commit bfa1a0dd190ed88020d60eba3bb04681c8e83a68
26711Author: Chris Wilson <chris@chris-wilson.co.uk>
26712Date: Mon Jan 24 11:17:03 2011 +0000
26713
26714 DRI2: Avoid a NULL pointer dereference
26715
26716 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=41211
26717
26718 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
26719 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26720
26721commit eeb21a133b982f71de739baf62e53c8a68f5d495
26722Author: Chris Wilson <chris@chris-wilson.co.uk>
26723Date: Mon Jan 24 11:17:03 2011 +0000
26724
26725 VidMode: prevent crash with no modes
26726
26727 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=17431
26728
26729 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
26730 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26731
26732commit c1bb8f43b9290c2b18a9f0ac59773ff8f1eb974f
26733Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
26734Date: Tue Oct 4 12:25:26 2011 +0300
26735
26736 record: Prevent out of bounds access when recording a reply.
26737
26738 Any pad bytes in replies are written to the client from a zeroed
26739 array. However, record extension tries to incorrectly access the pad
26740 bytes from the end of reply data.
26741
26742 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
26743 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
26744
26745commit 1f5baa924a5907978f564372265d9830ac72a077
26746Author: Jeremy Huddleston <jeremyhu@apple.com>
26747Date: Sun Oct 30 20:26:42 2011 -0700
26748
26749 xfree86: Deprecate the use of xf86PciInfo.h
26750
26751 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
26752 Reviewed-by: Tormod Volden <debian.tormod@gmail.com>
26753
26754commit eb3377ffb8a7baa26c9831e56ed782d48b28fa71
26755Author: Jeremy Huddleston <jeremyhu@apple.com>
26756Date: Tue Nov 1 14:59:15 2011 -0700
26757
26758 xfree86: Fix powerpc build with -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast
26759
26760 memType is a uint64_t on powerpc. Using memType only really makes
26761 sense for *physical* addresses, which can be 64-bit for 32-bit
26762 systems running on 64-bit hardware.
26763
26764 However, unmapVidMem() only deals with *virtual* addresses, which
26765 are guaranteed to fit into an uintptr_t.
26766
26767 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
26768 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
26769
26770commit a551f126cc4af7a476d9c8fd0cb309d5aa600d54
26771Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
26772Date: Fri Nov 4 23:26:25 2011 -0700
26773
26774 xfree86: Fix RandR rotation across server generations
26775
26776 245cb8e94fd1599 fixed xf86RotateDestroy() to actually run its teardown
26777 code, causing the Damage object to properly be re-allocated after a
26778 server regeneration. However the block that does that still thinks the
26779 Rotate layer BlockHandler is wrapped from the last generation, meaning
26780 the shadow pixmap is never re-allocated and the Damage object is never
26781 re-registered, causing a blank screen, and potentially a driver crash
26782 on the next teardown after the server asks it to free a 0x0 Pixmap.
26783
26784 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
26785 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
26786 Signed-off-by: Keith Packard <keithp@keithp.com>
26787
26788commit bfa2a1857a1efda7f171f10e855d200ca0dbcc1a
26789Merge: f0d50cc e7cb8f8
26790Author: Keith Packard <keithp@keithp.com>
26791Date: Mon Nov 14 09:07:06 2011 -0800
26792
26793 Merge remote-tracking branch 'whot/for-keith'
26794
26795commit e7cb8f802adf6fc22e7a930c0dfc5c1076781c96
26796Author: Peter Hutterer <peter.hutterer@who-t.net>
26797Date: Wed Nov 9 15:31:57 2011 +1000
26798
26799 Xext: remove unused variable 'n'
26800
26801 xselinux_ext.c: In function 'SELinuxSendItemsToClient':
26802 xselinux_ext.c:340:16: warning: unused variable 'n' [-Wunused-variable]
26803
26804 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26805 Reviewed-by: Julien Cristau <jcristau@debian.org>
26806
26807commit bb4aa1f263ad38c175bfda3b7e6c325260ce3f28
26808Author: Peter Hutterer <peter.hutterer@who-t.net>
26809Date: Wed Nov 9 15:31:10 2011 +1000
26810
26811 Xext: don't swap CARD8 in SProcSELinuxQueryVersion
26812
26813 xselinux_ext.c: In function 'SELinuxSendItemsToClient':
26814 xselinux_ext.c:340:16: warning: unused variable 'n' [-Wunused-variable]
26815 xselinux_ext.c: In function 'SProcSELinuxQueryVersion':
26816 xselinux_ext.c:532:62: error: call to 'wrong_size' declared with attribute
26817 error: wrong sized variable passed to swap
26818 xselinux_ext.c:533:62: error: call to 'wrong_size' declared with attribute
26819 error: wrong sized variable passed to swap
26820
26821 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26822 Reviewed-by: Julien Cristau <jcristau@debian.org>
26823
26824commit 35ec24cf245e5ef676e98a0bf7c77296a3f1ff63
26825Author: Peter Hutterer <peter.hutterer@who-t.net>
26826Date: Mon Aug 1 14:14:02 2011 +1000
26827
26828 input: replace remaining GetPairedDevice() with GetMaster()
26829
26830 Wherever it's obvious which device we need (keyboard or pointer), use
26831 GetMaster() instead of GetPairedDevice(). It is more reliable in actually
26832 getting the device type we want.
26833
26834 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26835 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26836
26837commit fc16917ad6f0d2722bfb5c5eeca955dd3bc4125a
26838Author: Peter Hutterer <peter.hutterer@who-t.net>
26839Date: Fri Apr 15 13:32:10 2011 +1000
26840
26841 include: export GetProximityEvents and QueueProximityEvents
26842
26843 This is mainly needed for consistency with GetPointerEvents and friend.
26844 No-one seems to actually need this function from outside the usual DDXs.
26845
26846 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26847 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26848
26849commit 11840595a1be9f2df7390bcc708cc176e60d3ef2
26850Author: Peter Hutterer <peter.hutterer@who-t.net>
26851Date: Thu Nov 3 08:58:58 2011 +1000
26852
26853 dix: Don't let a driver without a ProximityClassRec post events
26854
26855 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26856 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26857
26858commit 22715e465b415b3351b83b8279a4f44157f63199
26859Author: Peter Hutterer <peter.hutterer@who-t.net>
26860Date: Thu Oct 27 11:03:39 2011 +1000
26861
26862 Xi: allow passive keygrabs on the XIAll(Master)Devices fake devices
26863
26864 They don't have a KeyClassRec, but we must still allow passive grabs on
26865 them.
26866
26867 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26868 Tested-by: Bastien Nocera <hadess@hadess.net>
26869 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26870
26871commit f0d50cc6651dce3a8a3cd3fb84210aa92b139763
26872Author: Derek Buitenhuis <derek.buitenhuis@gmail.com>
26873Date: Sun Nov 6 12:20:51 2011 -0500
26874
26875 Fix vesa's VBE PanelID interpretation
26876
26877 xserver's VESA driver's VBE (Vesa BIOS Extensions) code
26878 includes a PanelID probe, which can get a monitor's native
26879 resolution. From this, using CVT formulas, it derives
26880 horizontal sync rate and a vertical refresh rate ranges.
26881
26882 It however, only derives the upper bounds of the ranges, and
26883 the lower bounds cannot de derived. By default, they are set
26884 to hardcoded constants which represent the lowest supported
26885 resolution: 640x480. The constants in vbe.c however, were
26886 not actually derived from forulas, but carried over from
26887 other code from the bad old days, and are not relevant
26888 to flat panel displays. This caused, for example, EEEPC701's
26889 panel, with a native resolution of 800x480, to end up with
26890 a upper bound of the horizontal sync rate that was lower
26891 than the hardcoded lower bound, which of course broke things.
26892
26893 These numbers have been rederived using both my own CVT tool
26894 based on xf86CVTMode(), and using the provided 'cvt' tool
26895 that comes with xserver.
26896
26897 Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
26898 Reviewed-by: Adam Jackson <ajax@redhat.com>
26899 Signed-off-by: Keith Packard <keithp@keithp.com>
26900
26901commit c643c2b7bf480f5c27ff8606bda087f8bff3b154
26902Author: Peter Hutterer <peter.hutterer@who-t.net>
26903Date: Tue Nov 1 12:17:50 2011 +1000
26904
26905 xfree86: duplicate name and driver from pInfo for NewInputDeviceRequest
26906
26907 xorg.conf devices had the name and driver set in the DDX's InputInfoPtr list
26908 but not in the option list for those devices. That information was lost when
26909 passing the options into NewInputDeviceRequest. NIDR then refused to start
26910 the devices.
26911
26912 Introduced in xorg-server-1.11.0-250-ge4cd24e
26913
26914 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
26915 Tested-by: James Cloos <cloos@jhcloos.com>
26916
26917commit 9cc44b955b27de33348d6a20bebc9704930ee18e
26918Author: Rui Matos <tiagomatos@gmail.com>
26919Date: Tue Nov 1 21:32:36 2011 +0000
26920
26921 randr: Make the RRConstrainCursorHarder logic the same as miPointerSetPosition
26922
26923 The constraining logic in RRConstrainCursorHarder allows the cursor to reach
26924 crtc positions of x = width and y = height while the constraining code in
26925 miPointerSetPosition only allows it to reach x = width - 1 and y = height - 1
26926 for the analogous screen case.
26927
26928 This patch makes the former's logic equivalent to the latter's which allows
26929 applications to benefit from Fitts's law. E.g. a maximized application
26930 adjacent to a crtc border wouldn't get pointer events if the user moved the
26931 pointer all the way until it's contained.
26932
26933 Signed-off-by: Rui Matos <tiagomatos@gmail.com>
26934 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
26935 Signed-off-by: Keith Packard <keithp@keithp.com>
26936
26937commit 548c6fe044068ffba9b5306dc6b11f2ba22782a4
26938Merge: 3881b0b ffe8ec8
26939Author: Keith Packard <keithp@keithp.com>
26940Date: Wed Nov 2 21:35:31 2011 -0700
26941
26942 Merge remote-tracking branch 'jturney/master'
26943
26944commit 3881b0bf1ce11a62473650bdf6ee01a1762f6f91
26945Author: Gaetan Nadon <memsize@videotron.ca>
26946Date: Tue Nov 1 07:17:07 2011 -0400
26947
26948 mi: remove deprecated miPointerAbsoluteCursor from design doc
26949
26950 Function was removed from the code by commit f5409aa026
26951
26952 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
26953 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
26954 Signed-off-by: Keith Packard <keithp@keithp.com>
26955
26956commit d91aa0e6601bdf3e2cfa57b3412ab14ac486edc4
26957Merge: 8df3a9c 401150d
26958Author: Keith Packard <keithp@keithp.com>
26959Date: Wed Nov 2 21:20:07 2011 -0700
26960
26961 Merge remote-tracking branch 'whot/two-screen-coordinates'
26962
26963commit 8df3a9ca5abbd56eb2013fa65250d21a8f18865b
26964Merge: 8329afa ef89548
26965Author: Keith Packard <keithp@keithp.com>
26966Date: Wed Nov 2 21:18:16 2011 -0700
26967
26968 Merge remote-tracking branch 'koba/reviewed'
26969
26970commit ffe8ec86db2655b2c83e5ae61521898140c855ac
26971Author: Colin Harrison <colin.harrison@virgin.net>
26972Date: Sat Oct 8 15:05:22 2011 +0100
26973
26974 hw/xwin: Fix a typo in ddraw.h
26975
26976 Fix a (fortunately benign) typo in ddraw.h
26977
26978 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
26979
26980commit 48fda3c52b41157313b7fd56d9f3fa3ad4166ed7
26981Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
26982Date: Thu Feb 18 20:37:00 2010 -0600
26983
26984 hw/xwin: Revert "Fix bug #5735, Serious flaw in CygwinX clipboard"
26985
26986 This commit wreaks havoc with other programs which manage the clipboard,
26987 such as MS Office Clipboard or Win32 VNC viewers:
26988
26989 http://sourceware.org/bugzilla/show_bug.cgi?id=9910
26990
26991 This reverts commit 70ddd0f39d5118db72a1a4e473cbfb502f1ed9ec.
26992
26993 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
26994 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
26995 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
26996 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
26997
26998commit 3ead1d810b0e157078db39712e02ea6dc85216d8
26999Author: Jon TURNEY <jon.turney@dronecode.org.uk>
27000Date: Sun Jul 3 19:10:52 2011 +0100
27001
27002 hw/xwin: Stop pretending we work on NT4
27003
27004 We already link directly to some functions not available in NT4, so stop
27005 pretending we will work on NT4 and link directly to EnumDisplayMonitors()
27006 and SHGetFolderPath()
27007
27008 Also remove mentions of NT4 & Win95 from error messages
27009
27010 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
27011 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
27012
27013commit f6529a05a2f087cb57a7fe84f226fef3780728d0
27014Author: Colin Harrison <colin.harrison@virgin.net>
27015Date: Sat Oct 8 14:39:07 2011 +0100
27016
27017 Xming: Always remove temporary file used when invoking xkbcomp on Win32
27018
27019 When built for native Win32, pipe() & fork() aren't available, so we
27020 use a tempoary file and system() to invoke xkbcomp
27021
27022 Ensure the temporary file is always removed. It was only being removed
27023 on most errors, not on success :S
27024
27025 Also fix a couple of warnings which occur when built with WIN32 defined
27026
27027 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
27028 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
27029
27030commit ee1985386700288af1a8b1789dcad0d7d9c97393
27031Author: Jon TURNEY <jon.turney@dronecode.org.uk>
27032Date: Sat Apr 16 23:28:45 2011 +0100
27033
27034 hw/xwin: Link directly to TrackMouseEvent()
27035
27036 TrackMouseEvent has existed in user32 since at least NT4, so
27037 don't bother with jumping through all the ancient compatibility hoops
27038 of finding if _TrackMouseEvent() exists in comctl32 so it can check
27039 if TrackMouseEvent() exists in user32 to see if it needs to emulate
27040 it...
27041
27042 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
27043 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
27044
27045commit e2e6fab1efa6895256ac0ed4d5b054a88ad18077
27046Author: Jon TURNEY <jon.turney@dronecode.org.uk>
27047Date: Tue Mar 23 04:45:37 2010 -0500
27048
27049 Cygwin/X: Make default DPI match native DPI
27050
27051 Make the default DPI match the current Windows DPI setting. If that
27052 setting can't be retrieved, change the fallback DPI value from 75 dpi
27053 to 96 dpi.
27054
27055 Mark the application as dpiAware in the manifest, which prevents
27056 dpi virtualization for high (>96) dpi values on Vista and later.
27057
27058 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
27059 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
27060 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
27061
27062commit 8329afa59dd5ea3adf7adebdb2111a9bccbb126b
27063Author: Aaron Plattner <aplattner@nvidia.com>
27064Date: Mon Oct 31 15:01:12 2011 -0700
27065
27066 dix: Reinstate GetImage window size check
27067
27068 Commit 587c3a2d1961834558193e8e14e8e381a077a253 fixed DoGetImage to
27069 check windows against their backing drawables, rather than against the
27070 screen dimensions, to prevent reading outside the bounds of redirected
27071 windows' backing pixmaps (see bug #22804). Unfortunately, while
27072 making that change I also removed the check that the rectangle is
27073 contained within the bounds of the source window, which is a violation
27074 of the specification:
27075
27076 If the drawable is a window, the window must be viewable, and it
27077 must be the case that, if there were no inferiors or overlapping
27078 windows, the specified rectangle of the window would be fully
27079 visible on the screen *and wholly contained within the outside
27080 edges of the window* (or a Match error results). Note that the
27081 borders of the window can be included and read with this request.
27082
27083 (emphasis mine)
27084
27085 Reinstate the window dimension check, to return BadMatch if the
27086 GetImage request falls outside the bounds of the window.
27087
27088 Fixes X Test Suite test XGetImage-15:
27089
27090 400|0 15 1 11:05:41|IC Start
27091 200|0 15 11:05:41|TP Start
27092 520|0 15 00005146 1 1|VSW5TESTSUITE PURPOSE 15
27093 520|0 15 00005146 1 2|Assertion XGetImage-15.(A)
27094 520|0 15 00005146 1 3|When the drawable is a window and the window is viewable
27095 520|0 15 00005146 1 4|and it is not the case that given there were no inferiors or
27096 520|0 15 00005146 1 5|overlapping windows the specified rectangle of the window
27097 520|0 15 00005146 1 6|would be fully visible on the screen and wholly contained
27098 520|0 15 00005146 1 7|within the outside edges of the window, then a BadMatch
27099 520|0 15 00005146 1 8|error occurs.
27100 520|0 15 00005146 1 9|METH: Create window which is not fully visible on the screen.
27101 520|0 15 00005146 1 10|METH: Call XMapWindow to make sure the window is viewable.
27102 520|0 15 00005146 1 11|METH: Call XGetImage with rectangle extending beyond edge of screen.
27103 520|0 15 00005146 1 12|METH: Verify XGetImage return value is null.
27104 520|0 15 00005146 1 13|METH: Verify that BadMatch error occurred.
27105 520|0 15 00005146 1 14|METH: Create window which is fully visible on the screen.
27106 520|0 15 00005146 1 15|METH: Call XMapWindow to make sure the window is viewable.
27107 520|0 15 00005146 1 16|METH: Call XGetImage with rectangle extending beyond edge of window.
27108 520|0 15 00005146 1 17|METH: Verify XGetImage return value is null.
27109 520|0 15 00005146 1 18|METH: Verify that BadMatch error occurred.
27110 520|0 15 00005146 1 19|REPORT: Got Success, Expecting BadMatch
27111 520|0 15 00005146 1 20|REPORT: Null image not returned.
27112 220|0 15 1 11:05:41|FAIL
27113
27114 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
27115 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
27116 Signed-off-by: Keith Packard <keithp@keithp.com>
27117
27118commit 132545ff576cc69ed63f5a08127151fe550de4c3
27119Merge: d0c6732 d7c44a7
27120Author: Keith Packard <keithp@keithp.com>
27121Date: Sun Oct 30 16:57:58 2011 -0700
27122
27123 Merge remote-tracking branch 'whot/for-keith'
27124
27125commit d7c44a7c9760449bef263413ad3b20f19b1dc95a
27126Author: Peter Hutterer <peter.hutterer@who-t.net>
27127Date: Mon Oct 24 12:00:32 2011 +1000
27128
27129 dix: block signals when closing all devices
27130
27131 When closing down all devices, we manually unset master for all attached
27132 devices, but the device's sprite info still points to the master's sprite
27133 info. This leaves us a window where the master is freed already but the
27134 device isn't yet. A signal during that window causes dereference of the
27135 already freed spriteInfo in mieqEnqueue's EnqueueScreen macro.
27136
27137 Simply block signals when removing all devices. It's not like we're really
27138 worrying about high-responsive input at this stage.
27139
27140 https://bugzilla.redhat.com/show_bug.cgi?id=737031
27141
27142 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27143 Reviewed-by: Julien Cristau <jcristau@debian.org>
27144
27145commit 820d9040f50a8440741b3aefbc069a3ad81e824e
27146Author: Servaas Vandenberghe <vdb@picaros.org>
27147Date: Wed Aug 31 07:06:49 2011 +0200
27148
27149 xfree86: fix potential buffer overflow
27150
27151 The patch below fixes a potential buffer overflow in xf86addComment().
27152 This occurs if curlen > 0 && eol_seen == 0 && iscomment == 0 , as
27153 follows from the code:
27154
27155 char *xf86addComment(char *cur, char *add)
27156
27157 <...>
27158
27159 len = strlen(add);
27160 endnewline = add[len - 1] == '\n';
27161 len += 1 + iscomment + (!hasnewline) + (!endnewline) + eol_seen;
27162
27163 if ((str = realloc(cur, len + curlen)) == NULL)
27164 return cur;
27165
27166 cur = str;
27167
27168 if (eol_seen || (curlen && !hasnewline))
27169 cur[curlen++] = '\n';
27170 if (!iscomment)
27171 cur[curlen++] = '#';
27172 strcpy(cur + curlen, add);
27173 if (!endnewline)
27174 strcat(cur, "\n");
27175
27176 Signed-off-by: Servaas Vandenberghe <vdb@picaros.org>
27177 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27178
27179 [whot: added buffer overflow test case]
27180
27181 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27182
27183commit 63e87b8639eb8e0b4e32e5d3a09099d31a03bbcd
27184Author: Peter Hutterer <peter.hutterer@who-t.net>
27185Date: Tue Oct 25 11:49:26 2011 +1000
27186
27187 xfree86: reduce calls to input_option_get_key/value
27188
27189 No functional changes.
27190
27191 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27192 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
27193
27194commit d0c6732a99c9a7e40752b9ba7898a01c325103fa
27195Author: Anssi Hannula <anssi.hannula@iki.fi>
27196Date: Wed Apr 20 14:34:52 2011 +0300
27197
27198 xfree86: add nouveau as the first automatic driver for NVIDIA hardware
27199
27200 Add nouveau as the first driver on linux for NVIDIA hardware when
27201 driver autoconfiguration is done, as it is more capable than nv.
27202
27203 nv is also kept in the list as it is more widely supported and because
27204 some old cards are not supported by nouveau.
27205
27206 Signed-off-by: Anssi Hannula <anssi.hannula@iki.fi>
27207 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27208 Reviewed-by: Cyril Brulebois <kibi@debian.org>
27209 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
27210 Signed-off-by: Keith Packard <keithp@keithp.com>
27211
27212commit ef895484c8a43d447d9d3ea7d4226936cc5f92cc
27213Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27214Date: Tue Oct 4 10:46:00 2011 +0600
27215
27216 bsd: alpha_video: Remove unused variables
27217
27218 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27219 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27220
27221commit fea7c7a8c070ec17ec03e4cb193834b6ca72fb72
27222Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27223Date: Mon Oct 3 23:03:31 2011 +0600
27224
27225 bsd: alpha_video: Simplify #include
27226
27227 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27228 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27229
27230commit 93a3a28f2c1e95ea7fae6cf1c2fb50b30eae9c33
27231Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27232Date: Tue Oct 4 00:45:09 2011 +0600
27233
27234 bsd: alpha_video: Function sethae() need only for FreeBSD
27235
27236 Return value sethae() is becoming void because no caller used it. Also old
27237 msb_set static checked by each caller is replaced by the p.hae static checked
27238 in sethae() when it's called.
27239
27240 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27241 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27242
27243commit 05b41e2dc60a0aefc685a33260e5d8e09580908b
27244Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27245Date: Tue Oct 4 17:20:13 2011 +0600
27246
27247 Move check definition MAP_FAILED in xf86_OSlib.h
27248
27249 Also remove odd definition MAP_FAILED.
27250
27251 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27252 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27253
27254commit af56e502f5ddf1d2d495b633734cadcb023b7c36
27255Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27256Date: Tue Oct 4 16:49:22 2011 +0600
27257
27258 Remove odd definition DEV_MEM
27259
27260 DEV_MEM defined in xf86_OSlib.h
27261
27262 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27263 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27264
27265commit 0481e9d3d1e0dbc05099184128de5cd6f13445f7
27266Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27267Date: Tue Oct 4 15:09:45 2011 +0600
27268
27269 Remove BSDi support
27270
27271 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27272 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27273
27274commit 8838a86fd325cb7f262a582d22ccdaa8a8f1a534
27275Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27276Date: Tue Oct 4 14:57:20 2011 +0600
27277
27278 Remove unused VT_SYSREQ_DEFAULT
27279
27280 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27281 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27282
27283commit 91131037a2d2e07079de5d222d8a8c95300750cc
27284Author: Alexandr Shadchin <alexandr.shadchin@gmail.com>
27285Date: Sat Oct 29 00:14:32 2011 +0600
27286
27287 Fix position _X_EXPORT
27288
27289 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27290 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27291 Signed-off-by: Keith Packard <keithp@keithp.com>
27292
27293commit 219bcec73d3554c57b87f28812ab06a66b739b19
27294Author: Alexandr Shadchin <alexandr.shadchin@gmail.com>
27295Date: Sat Oct 29 00:14:31 2011 +0600
27296
27297 bsd: Remove odd message about -sharevts
27298
27299 This is missing in commit 'xfree86: move -novtswitch & -sharevts argument
27300 handling up to common layer'
27301
27302 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
27303 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27304 Signed-off-by: Keith Packard <keithp@keithp.com>
27305
27306commit 005ab41986b0bb6a4e626aee7a7a542247f422e7
27307Author: Dave Airlie <airlied@redhat.com>
27308Date: Thu Oct 27 08:38:45 2011 +1000
27309
27310 test: fix two more failing FP3232 tests
27311
27312 And put a comment in to explain why we're testing for a frac between .3 and
27313 .6. We can't directly compare the frac since the floating/fixed point
27314 conversion loses precision.
27315
27316 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27317 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27318
27319commit 286fa9bf9b4ffb0f9dbd3f1cae46ddcb27ee9318
27320Author: Jeremy Huddleston <jeremyhu@apple.com>
27321Date: Tue Oct 25 23:53:43 2011 -0700
27322
27323 XWin: windowswm: Correct byte swapping in event notifications
27324
27325 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27326 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
27327
27328commit dc054fefc586c5b404fe0f949dbd456e5267c647
27329Author: Jeremy Huddleston <jeremyhu@apple.com>
27330Date: Tue Oct 25 23:49:33 2011 -0700
27331
27332 XQuartz: appledri: Correct byte swapping in event notifications
27333
27334 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27335
27336commit d5fee2b638fabeb8bec31a24e57d5c2a232d0893
27337Author: Jeremy Huddleston <jeremyhu@apple.com>
27338Date: Tue Oct 25 23:48:16 2011 -0700
27339
27340 XQuartz: applewm: Correct byte swapping in event notifications
27341
27342 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27343
27344commit 5701ab4a441eb113abd0851b0d71b82d12112854
27345Merge: 7d50211 a41214b
27346Author: Keith Packard <keithp@keithp.com>
27347Date: Mon Oct 24 22:09:00 2011 -0700
27348
27349 Merge remote-tracking branch 'whot/for-keith'
27350
27351commit a41214bc9a0f326c6dc129e4a6382efb8b826862
27352Author: Peter Hutterer <peter.hutterer@who-t.net>
27353Date: Tue Oct 25 12:57:07 2011 +1000
27354
27355 kdrive: check for null memory, fix OOB
27356
27357 If key/value allocation failed, don't bother adding another InputOption. And
27358 make sure the memory allocated is large enough for the trailing \0
27359
27360 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27361 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
27362
27363commit ffe20acedb3cdc4811eb52f8fc540ba6af7339fa
27364Author: Peter Hutterer <peter.hutterer@who-t.net>
27365Date: Thu Oct 20 16:42:20 2011 +1000
27366
27367 Use new FP1616/FP3232 conversion functions
27368
27369 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27370 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
27371
27372commit 7500d841f4c709ae0edc6420332096d3b9ef9fcc
27373Author: Peter Hutterer <peter.hutterer@who-t.net>
27374Date: Thu Oct 20 16:49:28 2011 +1000
27375
27376 test: fix test with new double -> fp3232 conversion functions
27377
27378 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27379 Tested-by: Dave Airlie <airlied@redhat.com>
27380 Reviewed-by: Dave Airlie <airlied@redhat.com>
27381
27382commit 4bb5d8fae4f9a70f12591315f0b267a2ea826a0c
27383Author: Peter Hutterer <peter.hutterer@who-t.net>
27384Date: Wed Oct 19 13:17:51 2011 +1000
27385
27386 Xi: send DeviceChangedEvents when the scroll valuators change value
27387
27388 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27389 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27390
27391commit 1f4af6c12fb5d4c19f4eac3df768517c9132cc88
27392Author: Peter Hutterer <peter.hutterer@who-t.net>
27393Date: Wed Oct 19 12:28:37 2011 +1000
27394
27395 dix: accept a NULL master for CreateClassesChangedEvent
27396
27397 If a floating device changes, the master is NULL but we must still create a
27398 DCE for it.
27399
27400 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27401 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27402
27403commit d77dec6971e4a0b306c8dbd5adf627908d7972cb
27404Author: Peter Hutterer <peter.hutterer@who-t.net>
27405Date: Wed Oct 19 12:11:16 2011 +1000
27406
27407 Xi: ensure the deviceid for DeviceChangedEvents is always the right one
27408
27409 If we're sending the event for a given device, make sure the deviceid is
27410 that of the device.
27411 This allows callers to use the same DCE for slave and master without having
27412 to fiddle the DCE's internal fields.
27413
27414 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27415 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27416
27417commit e9dee21fa3213bfe87b2b728a38eb41d3ba0e664
27418Author: Peter Hutterer <peter.hutterer@who-t.net>
27419Date: Tue Oct 18 17:47:59 2011 +1000
27420
27421 dix: pass the flags into the CreateClassesChangedEvent
27422
27423 No effective functional changes, prep work for future patches.
27424
27425 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27426 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27427
27428commit fcdd2587a17437b643b4592aa7f65d11c05a4cd8
27429Author: Peter Hutterer <peter.hutterer@who-t.net>
27430Date: Tue Oct 18 17:41:29 2011 +1000
27431
27432 dix: drop unused argument from XISendDeviceChangedEvent
27433
27434 Instead of device and master (and just using master), drop the master
27435 argument and let the callers pass in the device the event is to be sent for.
27436
27437 No effective functional changes.
27438
27439 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27440 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27441
27442commit 8473e441b0f832775153281bc3df5e2d4feb2b36
27443Author: Peter Hutterer <peter.hutterer@who-t.net>
27444Date: Tue Oct 18 17:11:27 2011 +1000
27445
27446 dix: add ScrollInfo to DeviceChangedEvents
27447
27448 3304bbff9b4ed63f1a47410a5320a136420ba2c6 added smooth scrolling support for
27449 pointer events and for XIQueryDevice but didn't add the matching parts to
27450 XIDeviceChangedEvents.
27451
27452 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27453 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27454
27455commit e3f6a76dd480717eae4b17ad8e2ff707de2ffe4c
27456Author: Andreas Wettstein <wettstein509@solnet.ch>
27457Date: Thu Aug 11 16:33:33 2011 +1000
27458
27459 xkb: Support noLock and noUnlock flags for LockMods
27460
27461 These flags are required by the XKB spec section 6.3.
27462
27463 Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
27464 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27465 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27466
27467commit 6f33593dc087d367d899d1cb6e6e30b282d922e2
27468Author: Peter Hutterer <peter.hutterer@who-t.net>
27469Date: Fri Oct 21 09:09:55 2011 +1000
27470
27471 xfree86 doc: replace driver "keyboard" with "kbd"
27472
27473 We've deprecated keyboard a long time ago
27474
27475 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27476 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27477
27478commit e4cd24e717ef1059804b3f6bb483810b708cd56a
27479Author: Peter Hutterer <peter.hutterer@who-t.net>
27480Date: Fri Aug 5 14:48:02 2011 +1000
27481
27482 xfree86: use NewInputDeviceRequest for xorg.conf devices too
27483
27484 Only use one init path for input devices - through NIDR.
27485
27486 This requires that inp_driver and inp_identifier from the
27487 XF86ConfInputRec are copied over into the options for NIDR to see them.
27488
27489 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27490 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27491
27492commit 1ecc427a39d41e723912492b846512fd0ad9af2d
27493Author: Peter Hutterer <peter.hutterer@who-t.net>
27494Date: Wed Aug 10 14:37:41 2011 +1000
27495
27496 dix: move #if statement to stop compiler warning
27497
27498 ptrveloc.c: In function 'QueryTrackers':
27499 ptrveloc.c:598:34: warning: variable 'used_offset' set but not used
27500 [-Wunused-but-set-variable]
27501
27502 used_offset is used, but only in the debugging code. Move the #if statement
27503 to ignore that warning.
27504
27505 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27506 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27507 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27508
27509commit f9067c1dd8ce9058eb48a20dfae52bc8cf3a1e55
27510Author: Peter Hutterer <peter.hutterer@who-t.net>
27511Date: Fri Aug 5 14:05:58 2011 +1000
27512
27513 xfree86: Fix a comment, the old function doesn't exist anymore
27514
27515 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27516 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27517 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27518
27519commit aeab26e9e1751e1e3514798fa53e9bd604b0d254
27520Author: Peter Hutterer <peter.hutterer@who-t.net>
27521Date: Wed Aug 10 15:58:34 2011 +1000
27522
27523 xfree86: use xf86AddNewOption instead of xf86addNewOption
27524
27525 The former strdups for us. If the strdup fails we miss out on the
27526 CorePointer option (default on anyway) and we're likely to fall over soon
27527 anyway, so let's pretend this is the same behaviour.
27528
27529 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27530 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27531 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27532
27533commit c39c8d34282b82d73c3c69a16cf0c2816256d85b
27534Author: Peter Hutterer <peter.hutterer@who-t.net>
27535Date: Wed Aug 10 14:38:58 2011 +1000
27536
27537 input: switch InputOption to use XF86OptionRec storage.
27538
27539 Use the same struct for both InputOption and XF86OptionRec so we don't need
27540 to convert to and fro the two in the config backends.
27541
27542 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27543 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
27544 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27545
27546commit 16ac78a53c1edeae183db8672104587b306cfe13
27547Author: Peter Hutterer <peter.hutterer@who-t.net>
27548Date: Tue Oct 25 11:40:58 2011 +1000
27549
27550 kdrive: switch to new InputOption API.
27551
27552 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27553 Reviewed-by: Keith Packard <keithp@keithp.com>
27554
27555commit 7d50211ab57a35910d79fc3f67ae89aff91fa995
27556Author: Christopher Yeleighton <giecrilj@stegny.2a.pl>
27557Date: Mon Oct 24 18:47:06 2011 -0700
27558
27559 Bug 38420: Xvfb crashes in miInitVisuals() when started with depth=2
27560
27561 https://bugs.freedesktop.org/show_bug.cgi?id=38420
27562
27563 Exit with fatal error message, not segfault.
27564
27565 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27566 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
27567 Signed-off-by: Keith Packard <keithp@keithp.com>
27568
27569commit d9d3a01ffca5e2de650d098231e16205781804c5
27570Merge: 17416e8 fb55f8f
27571Author: Keith Packard <keithp@keithp.com>
27572Date: Mon Oct 24 18:12:23 2011 -0700
27573
27574 Merge remote-tracking branch 'jeremyhu/master'
27575
27576commit 17416e88dcfcc584fe5f87580d5d2b719b3521c3
27577Author: Dave Airlie <airlied@redhat.com>
27578Date: Thu Oct 20 14:43:01 2011 +0100
27579
27580 xf86Crtc: handle no outputs with no modes harder.
27581
27582 If you started an X server with no connected outputs, we pick a default
27583 1024x768 mode, however if you then ran an xvidmode using app against that
27584 server it would segfault the server due to not finding any valid modes.
27585
27586 This was due to the no output mode set code, only adding the modes to the
27587 scrn->modes once, when something called randr 1.2 xf86SetScrnInfoModes would
27588 get called and remove all the modes and we'd end up with 0.
27589
27590 This change fixes xf86SetScrnInfoModes to always report a scrn mode of at
27591 least 1024x768, and pushes the initial configuration to just call it instead
27592 of setting up the mode itself.
27593
27594 Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=746926
27595
27596 I've seen other bugs like this on other distros so it might also actually fix them.
27597
27598 Signed-off-by: Dave Airlie <airlied@redhat.com>
27599 Reviewed-by: Keith Packard <keithp@keithp.com>
27600 Signed-off-by: Keith Packard <keithp@keithp.com>
27601
27602commit fb55f8f790f69f30b0ab18648c337bde9d41f5ff
27603Author: Jeremy Huddleston <jeremyhu@apple.com>
27604Date: Mon Oct 24 16:45:36 2011 -0700
27605
27606 XQuartz: Fix the filename of our log file
27607
27608 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27609
27610commit 1c8bda798b4480bab0f0da6f531e4770c7a1f771
27611Author: Jeremy Huddleston <jeremyhu@apple.com>
27612Date: Fri Oct 21 10:27:16 2011 -0700
27613
27614 XQuartz: appledri: Allow byte swapped requests
27615
27616 Even though it's only valid when local, it is possible for a local
27617 client and the server to not match endianness, such as when running
27618 a ppc application under Rosetta.
27619
27620 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27621
27622commit 14205ade0c750191bf43fba8bd55c65dba912cf4
27623Author: Jeremy Huddleston <jeremyhu@apple.com>
27624Date: Thu Oct 20 22:34:44 2011 -0700
27625
27626 XQuartz: appledri: Fix byte swapping in replies
27627
27628 Even though it's only valid when local, it is possible for a local
27629 client and the server to not match endianness, such as when running
27630 a ppc application under Rosetta.
27631
27632 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27633 Reviewed-by: Jamey Sharp <jamey@minilop.net>
27634
27635commit 2ba0ac202ad64eb4a555715980ba538de19c9fd7
27636Author: Jeremy Huddleston <jeremyhu@apple.com>
27637Date: Thu Oct 20 22:23:07 2011 -0700
27638
27639 XQuartz: appledri: Set the correct reply length for XAppleDRICreatePixmap
27640
27641 http://xquartz.macosforge.org/trac/ticket/508
27642
27643 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27644 Reviewed-by: Jamey Sharp <jamey@minilop.net>
27645
27646commit e4f257748b51d041275a64b8be408729ea05fbb5
27647Author: Jeremy Huddleston <jeremyhu@apple.com>
27648Date: Thu Oct 20 11:35:54 2011 -0700
27649
27650 XQuartz: appledri: Change whitespace in appledristr.h to match mesa
27651
27652 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27653 Reviewed-by: Jamey Sharp <jamey@minilop.net>
27654
27655commit 83fef4235db86343477b4ec9858c6ba35e1aa7d9
27656Author: Jeremy Huddleston <jeremyhu@apple.com>
27657Date: Sun Oct 16 02:12:38 2011 -0700
27658
27659 Xnest: Match the host's keymap
27660
27661 This was a regression.
27662
27663 Introduced by: 08363c5830bdea34012dcd954b45ccfdc79a3a7e and
27664 32db27a7f867b503c2840ca7b815e96d10be9210
27665 Masked by: 1e69fd4a60147287b31e53bfc61543fb17bb82c8
27666
27667 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27668 Reviewed-by: Jamey Sharp <jamey@minilop.net>
27669
27670commit 4597ac6fe0580323860080dae2576733acc49757
27671Author: Jeremy Huddleston <jeremyhu@apple.com>
27672Date: Mon Oct 17 23:45:32 2011 -0700
27673
27674 test: Add unit test for mieq
27675
27676 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27677 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27678
27679commit ad450d76e468cd62406c4f09e71e131f7ff4555c
27680Author: Jeremy Huddleston <jeremyhu@apple.com>
27681Date: Mon Oct 17 21:16:37 2011 -0700
27682
27683 mieq: Reserve some space in EQ for release and other special events
27684
27685 The last 64 events in the event queue will be reserved for release
27686 events in order to help return the system to a cleaner state when
27687 it comes back from a soft wedge.
27688
27689 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27690 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27691
27692commit 7d153f25be678acf500236ae422b33d17e6cb7a6
27693Author: Jeremy Huddleston <jeremyhu@apple.com>
27694Date: Sat Oct 15 22:51:30 2011 -0700
27695
27696 mieq: Provide better adaptability and diagnostics during mieq overflow
27697
27698 This patch changes from a static length event queue (512) to one that
27699 starts at 128 and grows to 4096 as it overflows, logging each time it
27700 grows.
27701
27702 This change also allows for multiple backtraces to be printed when the
27703 server is wedged rather than just one. This increased sampling should
27704 help identify the true hog in cases where one backtrace might be
27705 insufficient.
27706
27707 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27708 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27709
27710commit 09eaa1f6f19f9a85a3cb74a18ac898162db451df
27711Author: Jeremy Huddleston <jeremyhu@apple.com>
27712Date: Sat Oct 30 14:55:06 2010 -0700
27713
27714 configure.ac: Add -fno-strict-aliasing to CFLAGS
27715
27716 This should force the server to have -fno-strict-aliasing even once it
27717 is removed from the warning flags.
27718
27719 See: https://bugs.freedesktop.org/show_bug.cgi?id=31238
27720
27721 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27722 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
27723
27724commit a046d649e4c7e4d28f350382dcdd293e92a59aad
27725Author: Jeremy Huddleston <jeremyhu@apple.com>
27726Date: Wed Oct 5 15:02:52 2011 -0700
27727
27728 test: Silence some debug lines from the input unit tests
27729
27730 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27731 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27732
27733commit 8db554d2d830953de3c6d0cbf58e23d53d89009f
27734Author: Jeremy Huddleston <jeremyhu@apple.com>
27735Date: Thu Oct 20 14:28:56 2011 -0700
27736
27737 Xnest: Fix DestroyNotify handler
27738
27739 This partially reverts the unwanted changes that crept into
27740 c13a48e74ec89eafa9b529a87253a631ef02f2e2
27741
27742 Reported-by: Julien Cristau <jcristau@debian.org>
27743 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
27744 Reviewed-by: Julien Cristau <jcristau@debian.org>
27745
27746commit ff61592441916b83aeb778c74352bb5b26247f84
27747Merge: af3f64f 525d417
27748Author: Keith Packard <keithp@keithp.com>
27749Date: Wed Oct 19 19:14:32 2011 -0700
27750
27751 Merge remote-tracking branch 'aplattner/for-master'
27752
27753commit af3f64fb77c13180e513ee99d1fd9a1b624fd8ea
27754Merge: 15bbdc1 df0dd36
27755Author: Keith Packard <keithp@keithp.com>
27756Date: Wed Oct 19 17:33:07 2011 -0700
27757
27758 Merge remote-tracking branch 'hramrach/pull'
27759
27760commit 15bbdc103b34b6b374815698946e6c409421a644
27761Merge: a5266dc 323869f
27762Author: Keith Packard <keithp@keithp.com>
27763Date: Wed Oct 19 17:26:50 2011 -0700
27764
27765 Merge remote-tracking branch 'whot/for-keith'
27766
27767commit a5266dcb3a60587e1877f90c18552baf60b597a0
27768Author: Ville Syrjala <syrjala@sci.fi>
27769Date: Sun Oct 9 01:11:04 2011 +0300
27770
27771 composite: Update borderClip in compAllocPixmap()
27772
27773 Previously the parent constrained borderClip was copied over
27774 when compRedirectWindow() is called. That is insufficient eg. in
27775 case the window was already redirected, but not yet realized. So
27776 copy the borderClip over in compAllocPixmap() instead.
27777
27778 Example:
27779 Window 1 is below an automatically redirect window 2. Window 2 is
27780 unmapped and moved outside the extents of window 1. Window 2 is
27781 then mapped again, and MarkOverlappedWindows() uses the up to
27782 date borderSize of window 2 to mark windows, which leaves
27783 window 1 unmarked. Then exposures are calculated using the stale
27784 borderClip of window 2, which causes the window below window 2,
27785 to be exposed through an apparent hole in window 1.
27786
27787 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=22566
27788
27789 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
27790 Reviewed-by: Keith Packard <keithp@keithp.com>
27791 Signed-off-by: Keith Packard <keithp@keithp.com>
27792
27793commit e4787ec20b1b9d30fe00f17b60eb0898190bf77d
27794Author: Gaetan Nadon <memsize@videotron.ca>
27795Date: Mon Oct 17 15:08:14 2011 -0400
27796
27797 XWinrc.man: fix warning: tab character in unquoted macro argument #35054
27798
27799 Spaces or tabs do not affect the text output layout.
27800
27801 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
27802 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
27803 Signed-off-by: Keith Packard <keithp@keithp.com>
27804
27805commit 321873f8045139edd47b526a765f8b1c66f53cd4
27806Author: Gaetan Nadon <memsize@videotron.ca>
27807Date: Mon Oct 17 15:08:13 2011 -0400
27808
27809 xorg.conf.man: fix 382: warning: missing `)' (got `R') #35054
27810
27811 http://www.gnu.org/software/groff/manual/html_node/Man-usage.html
27812
27813 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27814 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
27815 Signed-off-by: Keith Packard <keithp@keithp.com>
27816
27817commit 3e145d3d67b472cb5d74cb8078b614160c4722b7
27818Author: Jesse Barnes <jbarnes@virtuousgeek.org>
27819Date: Thu Oct 13 12:01:10 2011 -0700
27820
27821 crtc: match full preferred modes if possible when choosing an initial config
27822
27823 It's fairly common to have multiple, identical monitors plugged in. In
27824 that case, it's preferable to run the monitor's preferred mode on each
27825 output, rather than just matching the width & height and end up with
27826 different timings or refresh rates.
27827
27828 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
27829 Reviewed-by: Keith Packard <keithp@keithp.com>
27830 Signed-off-by: Keith Packard <keithp@keithp.com>
27831
27832commit 525d4172b246e13b8122e059e3b22866e00561d9
27833Author: Chris Wilson <chris@chris-wilson.co.uk>
27834Date: Sun Sep 4 09:34:08 2011 -0700
27835
27836 render: export TriStrip and TriFan to the drivers
27837
27838 Rather than perform an intermediate copy and expand the strip and the
27839 fan into a triangle list (thereby tripling the number of edges that the
27840 driver needs to process), allow the backend to hook directly into the
27841 appropriate Composite function.
27842
27843 In order to extend the PictureScreen, without needlessly bumping the
27844 ABI, we move the existing copy implementations to mipict.c and assign
27845 those by default. To notify the ddx that the new entry points are
27846 available, we introduce PICTURE_SCREEN_VERSION.
27847
27848 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
27849 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
27850
27851commit 9b26e6bc8d2cdf5bac3025796855ccf05972358f
27852Author: Luc Verhaegen <libv@skynet.be>
27853Date: Tue Aug 23 15:19:59 2011 -0700
27854
27855 randr: stop clients from deleting immutable output properties
27856
27857 Immutable in randr means that clients are not able to alter the
27858 property itself, they are only allowed to alter the property value.
27859 This logically means that the property then should not be deleted
27860 by the client either.
27861
27862 Signed-off-by: Luc Verhaegen <libv@skynet.be>
27863 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
27864 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
27865 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
27866
27867commit c8413362049cee8c30e0a9d67f78f9ebefe8e71f
27868Merge: 64d2d1b b67581c
27869Author: Keith Packard <keithp@keithp.com>
27870Date: Tue Oct 18 07:45:24 2011 -0700
27871
27872 Merge remote-tracking branch 'herrb/master'
27873
27874commit b67581cf825940fdf52bf2e0af4330e695d724a4
27875Author: Matthieu Herrb <matthieu.herrb@laas.fr>
27876Date: Mon Oct 17 22:27:35 2011 +0200
27877
27878 Fix CVE-2011-4029: File permission change vulnerability.
27879
27880 Use fchmod() to change permissions of the lock file instead
27881 of chmod(), thus avoid the race that can be exploited to set
27882 a symbolic link to any file or directory in the system.
27883
27884 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
27885 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27886
27887commit 6ba44b91e37622ef8c146d8f2ac92d708a18ed34
27888Author: Matthieu Herrb <matthieu.herrb@laas.fr>
27889Date: Mon Oct 17 22:26:12 2011 +0200
27890
27891 Fix CVE-2011-4028: File disclosure vulnerability.
27892
27893 use O_NOFOLLOW to open the existing lock file, so symbolic links
27894 aren't followed, thus avoid revealing if it point to an existing
27895 file.
27896
27897 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
27898 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27899
27900commit df0dd36deea0c756819825113e825059ddd19243
27901Author: Michal Suchanek <hramrach@centrum.cz>
27902Date: Sat Oct 8 14:26:24 2011 +0200
27903
27904 Do not uselessly reload modules in DuplicateModule
27905
27906 The function does not initialize the module so it has no business
27907 loading it. If some user of DuplicateModule expects a module actually
27908 loaded they should use LoadModule.
27909
27910 Signed-off-by: Michal Suchanek <hramrach@centrum.cz>
27911 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27912
27913commit 24d435163eb5fbd9b73cd8ba13a9b3cdbbe8a1df
27914Author: Michal Suchanek <hramrach@centrum.cz>
27915Date: Sat Oct 8 14:19:34 2011 +0200
27916
27917 Use UnloadModuleOrDriver for UnloadSubModule.
27918
27919 Signed-off-by: Michal Suchanek <hramrach@centrum.cz>
27920 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27921
27922commit 0d4bb5442ceb8e8e4a8de6cfc4203cae469eee72
27923Author: Michal Suchanek <hramrach@centrum.cz>
27924Date: Sat Oct 8 14:13:33 2011 +0200
27925
27926 Unload submodules.
27927
27928 Signed-off-by: Michal Suchanek <hramrach@centrum.cz>
27929 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27930
27931commit b04aff76ac2eb461c71b85525a00e25efb8bf267
27932Author: Michal Suchanek <hramrach@centrum.cz>
27933Date: Thu Oct 13 17:14:53 2011 +0200
27934
27935 Document -background none option
27936
27937 Document option introduced in commit 8976e97.
27938
27939 Signed-off-by: Michal Suchanek <hramrach@centrum.cz>
27940 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
27941 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
27942
27943commit 323869f3298cbbfe864af9404a8aed1bf7995d79
27944Author: Tomáš Trnka <tomastrnka@gmx.com>
27945Date: Tue Oct 11 09:11:18 2011 +0200
27946
27947 Fix drain_console unregistration
27948
27949 Bug introduced by 9dca441670d261a9a9fb6108960ed48f3d58fb7f
27950 xfree86: add a hook to replace the new console handler.
27951
27952 console_handler was not being set, making the server eat up CPU spinning
27953 in WaitForSomething selecting consoleFd over and over again, every time
27954 trying to unregister drain_console without success due to
27955 console_handler being NULL.
27956
27957 Let's just fix the unregistration in xf86SetConsoleHandler() and use that.
27958
27959 But wait, there could be a catch: If some driver replaced the handler using
27960 xf86SetConsoleHandler(), the unregistration in xf86CloseConsole will unregister
27961 that one. I don't understand Xorg well enough to know whether this poses a
27962 problem (could mess up driver deinit somehow or something like that). As it is,
27963 xf86SetConsoleHandler() doesn't offer any way to prevent this (i.e. check which
27964 handler is currently registered).
27965
27966 I had been using it for two days on my machine that previously hit 100% CPU
27967 several times a day. That has now gone away without any new problems appearing.
27968
27969 Signed-off-by: Tomas Trnka <tomastrnka@gmx.com>
27970 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
27971 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27972
27973commit 2cb63180fa9b54f763e7e92e433943e3e73741f3
27974Author: Peter Hutterer <peter.hutterer@who-t.net>
27975Date: Wed Oct 5 12:08:31 2011 +1000
27976
27977 dix: mark motion events as emulated if we're scrolling from button press
27978
27979 The protocol requires that the emulated event is marked as such. So if a
27980 driver with smooth scrolling axis sends legacy button events, the motion
27981 event must be marked as emulated.
27982
27983 Pass the real type to emulate_scroll_button_events and create the events
27984 accordingly. For real button press or relase events, only that event must be
27985 generated since a release event will follow or a press event has already
27986 occured, respectively. (This fixes a bug where we'd get two release events
27987 for each legacy button event)
27988
27989 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
27990 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
27991
27992commit 82c60232c07f50774ccc0198950f64c9338057a5
27993Author: Peter Hutterer <peter.hutterer@who-t.net>
27994Date: Fri Sep 30 10:59:47 2011 +1000
27995
27996 dix: add valuator_mask_fetch_double()
27997
27998 Using this call simplifies callers that don't know if the mask bit is set.
27999
28000 Before:
28001 if (valuator_mask_isset(mask, valnum))
28002 value = valuator_mask_get_double(mask, valnum));
28003 else
28004 value = someothervalue;
28005
28006 Now:
28007 if (!valuator_mask_fetch_double(mask, valnum, &value))
28008 value = someothervalue;
28009
28010 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28011 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
28012 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28013
28014commit 64d2d1bef14c7ec7156bb3aae64eff69f9f7be54
28015Merge: fb84be4 679c84b
28016Author: Keith Packard <keithp@keithp.com>
28017Date: Mon Oct 17 14:36:34 2011 -0700
28018
28019 Merge remote-tracking branch 'jeremyhu/master'
28020
28021commit fb84be47db7cdaff406792c08e34670e8e0cbda9
28022Merge: fae7ed6 32b289e
28023Author: Keith Packard <keithp@keithp.com>
28024Date: Mon Oct 17 13:50:25 2011 -0700
28025
28026 Merge remote-tracking branch 'whot/next'
28027
28028commit 679c84bce9b03c17aa35fa69058e7e46267dbc7b
28029Author: Jeremy Huddleston <jeremyhu@apple.com>
28030Date: Sat Sep 24 00:01:11 2011 -0700
28031
28032 Bump ABI_VIDEODRV_VERSION to 12
28033
28034 The ABI changed in the previous series of changes, so bump the ABI version for
28035 the next release.
28036
28037 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28038 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28039
28040commit a89cdcee4ebab6bbfa9ace94e1988a94cfb67087
28041Author: Jeremy Huddleston <jeremyhu@apple.com>
28042Date: Sun Oct 9 02:59:12 2011 -0700
28043
28044 xfree86: Deprecate xf86MapVidMem and friends
28045
28046 Drivers should transition over to using libpciaccess's instead.
28047
28048 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28049
28050commit 41d903918a5721e53ed694880eab44170f9717cc
28051Author: Jeremy Huddleston <jeremyhu@apple.com>
28052Date: Wed Sep 14 13:45:18 2011 -0500
28053
28054 xorg-server.pc.in: Remove libpciaccess and pixman-1 from Requires
28055
28056 Every module building against xorg-server does not *Require* pixman nor
28057 libpciaccess. If such modules need pixman or pciaccess, they should be
28058 depending on them directly rather than inheriting a dependency from
28059 xorg-server. To do this, they should use PKG_CHECK_MODULES in configure.ac
28060 to check for pixman-1 or pciaccess and include the apporpriate _LIBS variable
28061 to the appropriate _LDFLAGS variable in Makefile.am
28062
28063 This also moves pixman-1 to Requires.private, so CPPFLAGS is set right to
28064 to satisfy include dependencies but avoid linking needlessly.
28065
28066 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28067 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
28068
28069commit 40d338e9f723c4b3d63516aa62fdcded4d8af75e
28070Author: Jeremy Huddleston <jeremyhu@apple.com>
28071Date: Wed Sep 14 10:37:38 2011 -0500
28072
28073 darwin: configure: Force some irrelevant options to off on darwin
28074
28075 This removes the need to pass the following to configure on darwin:
28076 --disable-pciaccess -disable-int10-module --disable-vbe --disable-vgahw --disable-libdrm
28077
28078 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28079 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28080
28081commit 43d730c0e4b1a00b29843c0660dcc7c60da6725d
28082Author: Jeremy Huddleston <jeremyhu@apple.com>
28083Date: Wed Sep 14 15:42:19 2011 -0500
28084
28085 xfree86: Link modules with -module
28086
28087 This makes a difference on darwin (and apparently nowhere else)
28088
28089 https://www.gnu.org/s/libtool/manual/libtool.html#Modules-for-libltdl
28090
28091 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28092 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28093 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
28094
28095commit f7edc00a2af83764910ff89514a046dc5ce2ff9e
28096Author: Jeremy Huddleston <jeremyhu@apple.com>
28097Date: Wed Sep 14 12:14:56 2011 -0500
28098
28099 xfree86: fbdevhw: Remove unused include of pciaccess.h
28100
28101 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28102 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28103
28104commit 6817050f3196d460ba03a62da23fb46ffd5a0f97
28105Author: Tiago Vignatti <tiago.vignatti@nokia.com>
28106Date: Tue Jun 1 13:05:47 2010 +0300
28107
28108 configure: wrap PCI code with macro and set it at build time
28109
28110 --disable-pciaccess, used together with --disable-module-int10, can be used to
28111 disable all pci code inside the server.
28112
28113 Note that XSERVER_LIBPCIACCESS was previously used only in the driver side and
28114 now it defines also whether the library is used inside the server. Also,
28115 XORG_BUS_PCI automake variable is introduced to track PCI code needs.
28116
28117 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
28118 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
28119 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28120 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28121
28122commit 5c12399b6c3a8def2df1bdde1d5d4d8d9e31fd84
28123Author: Tiago Vignatti <tiago.vignatti@nokia.com>
28124Date: Mon May 31 19:27:07 2010 +0300
28125
28126 configure: make PCI configuration more sane
28127
28128 No semantical changes. Just moved code around, grouping PCI related stuff in a
28129 single chunk.
28130
28131 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
28132 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
28133 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28134 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28135
28136commit a319e9e697faa0faa241e0e9baf0bc41efdcdf8f
28137Author: Tiago Vignatti <tiago.vignatti@nokia.com>
28138Date: Mon May 31 18:50:50 2010 +0300
28139
28140 configure: change PCI function checking by a meaningful version of the library
28141
28142 People that don't want VGA arbiter active can go to the library and enable the
28143 stubs there.
28144
28145 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
28146 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
28147 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28148 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28149 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28150
28151commit b3ca84430d98f5338b9b1bc817c3656efb71cff1
28152Author: Jeremy Huddleston <jeremyhu@apple.com>
28153Date: Tue Sep 13 15:32:16 2011 -0500
28154
28155 xfree86: Work around issue where ar may be told to make an archive with no contents
28156
28157 Automake:
28158
28159 "Be careful when selecting library components conditionally. Because building
28160 an empty library is not portable, you should ensure that any library
28161 always contains at least one object."
28162
28163 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28164 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
28165
28166commit e8bafb9d8c7d7a7077e163ecfac6174356989bbf
28167Author: Jeremy Huddleston <jeremyhu@apple.com>
28168Date: Tue Sep 13 15:38:45 2011 -0500
28169
28170 xfree86: Add stubs for os-support to help adding new architecture support
28171
28172 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28173 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28174 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
28175
28176commit 342f3eac8460fc48cfad1f1d7be939d671e6e1cd
28177Author: Adam Jackson <ajax@redhat.com>
28178Date: Tue May 10 19:00:02 2011 -0400
28179
28180 int10: Port internal users off xf86MapVidMem
28181
28182 This API is apparently semi-deprecated even by XFree86 standards, and
28183 there are only four drivers left using it. Let's start chopping it off.
28184
28185 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28186 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28187 Signed-off-by: Adam Jackson <ajax@redhat.com>
28188
28189commit 7757b8092474c4f4ec04f4ebd1272236416154c3
28190Author: Adam Jackson <ajax@redhat.com>
28191Date: Tue May 10 17:34:23 2011 -0400
28192
28193 pci: Remove xf86MapDomainMemory
28194
28195 This is slightly draconian, but that API is just awful. In all but
28196 one case in the callers it's used to get a map of some legacy VGA
28197 memory, and it would be cleaner for the caller to just call
28198 pci_device_map_legacy.
28199
28200 The sole exception is in the vesa driver, which uses it to avoid having
28201 to look up which device the BAR belongs to. That's similarly trivial to
28202 fix.
28203
28204 Having done that, Linux's PCI layer is now very small indeed.
28205
28206 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28207 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28208 Signed-off-by: Adam Jackson <ajax@redhat.com>
28209
28210commit 6f5041d0b8e11e761d6403f1f8f8bb2ba2a9626d
28211Author: Adam Jackson <ajax@redhat.com>
28212Date: Tue May 10 18:04:10 2011 -0400
28213
28214 int10: Use pciaccess rom fetch for !PC machines
28215
28216 ... instead of rolling our own, badly.
28217
28218 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28219 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28220 Signed-off-by: Adam Jackson <ajax@redhat.com>
28221 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28222
28223commit 492ed3e53b0ee6746bd4b597978fd866d4a89ef6
28224Author: Adam Jackson <ajax@redhat.com>
28225Date: Tue May 10 16:41:08 2011 -0400
28226
28227 pci: Deprecate the PCITAG type
28228
28229 It is kept around to help drivers through the API transition and will be
28230 removed at some point in the future.
28231
28232 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28233 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28234 Signed-off-by: Adam Jackson <ajax@redhat.com>
28235 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28236
28237commit a248fa3a33cb015942386d52b7faa8558e592a51
28238Author: Adam Jackson <ajax@redhat.com>
28239Date: Fri Jan 7 17:20:15 2011 -0500
28240
28241 xfree86: Remove unused bios_devmem.c
28242
28243 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28244 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28245 Signed-off-by: Adam Jackson <ajax@redhat.com>
28246 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28247
28248commit aa0bfb0f133481c57762012e8e30c05ffa151423
28249Author: Adam Jackson <ajax@redhat.com>
28250Date: Fri Jan 7 16:57:15 2011 -0500
28251
28252 linux: Remove ia64 domain I/O support code
28253
28254 pciaccess handles this now.
28255
28256 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28257 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28258 Signed-off-by: Adam Jackson <ajax@redhat.com>
28259 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28260
28261commit 95b6935637628ffa511f39004e90c75aa75141c4
28262Author: Adam Jackson <ajax@redhat.com>
28263Date: Fri Oct 16 16:09:56 2009 -0400
28264
28265 linux: Remove pre-2.6 PCI interface support
28266
28267 If you haven't ported 2.6 to your architecture in the intervening seven
28268 years, you can keep running older servers.
28269
28270 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28271 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28272 Signed-off-by: Adam Jackson <ajax@redhat.com>
28273 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28274
28275commit 769f583783d17ffa3398bc8529f61e0f08b35530
28276Author: Adam Jackson <ajax@redhat.com>
28277Date: Fri Oct 16 14:05:54 2009 -0400
28278
28279 bus: remove some dead struct fields
28280
28281 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28282 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28283 Signed-off-by: Adam Jackson <ajax@redhat.com>
28284 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28285
28286commit dd72b3c1f1df57215329aa5ba9e9bb9ad0a65086
28287Author: Adam Jackson <ajax@redhat.com>
28288Date: Mon Sep 20 11:42:11 2010 -0400
28289
28290 linux: Use pci_device_get_parent_bridge instead of open-coding it
28291
28292 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28293 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28294 Signed-off-by: Adam Jackson <ajax@redhat.com>
28295 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28296
28297commit 858fbbb40d7c69540cd1fb5315cebf811c6e7b3f
28298Author: Adam Jackson <ajax@redhat.com>
28299Date: Fri Sep 16 13:33:04 2011 -0400
28300
28301 pci: Port xf86MapLegacyIO to pciaccess
28302
28303 Per-domain I/O is now something drivers must manually request, and must
28304 keep track of within their own state rather than in the ScrnInfoRec.
28305 It's not really possible to split that into two steps without an
28306 additional intermediate ABI break, so don't even try. Drivers that want
28307 source compatibility should ifdef on the presence of xf86UnmapLegacyIO.
28308
28309 As a fringe benefit, domain-aware I/O is now OS-independent, relying
28310 only on support in pciaccess. Simplify OS PCI setup to reflect this.
28311
28312 The IOADDRESS type is kept around to help drivers through the API
28313 transition and will be removed at some point in the future.
28314
28315 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28316 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28317 Signed-off-by: Adam Jackson <ajax@redhat.com>
28318 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28319
28320commit 6d9efdce0d06df6b85f0681bea306c0b1e851502
28321Author: Adam Jackson <ajax@redhat.com>
28322Date: Tue Sep 20 18:12:29 2011 -0400
28323
28324 vgahw: Port to pciaccess IO space routines
28325
28326 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28327 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28328 Signed-off-by: Adam Jackson <ajax@redhat.com>
28329 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28330
28331commit 4bd6579188e718654c35f95623fd4772f9e0ef06
28332Author: Adam Jackson <ajax@redhat.com>
28333Date: Thu Sep 22 13:45:57 2011 -0400
28334
28335 vgahw: Don't default to standard (port space) access routines
28336
28337 In fact, don't default to anything; drivers must explicitly say which
28338 kind they want, and they are strongly encouraged to do MMIO if possible.
28339 This is an ABI change in that drivers that don't will crash, but drivers
28340 that are explicit will work with both old and new servers.
28341
28342 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28343 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28344 Signed-off-by: Adam Jackson <ajax@redhat.com>
28345 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28346
28347commit 30fb334d218f8c1d809f88054b9fe8f5f556bb62
28348Author: Adam Jackson <ajax@redhat.com>
28349Date: Thu Sep 22 13:44:53 2011 -0400
28350
28351 vgahw: Remove IO domain setup
28352
28353 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28354 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28355 Signed-off-by: Adam Jackson <ajax@redhat.com>
28356 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28357
28358commit 51a5558beb71bddeff9352ef4f43269a8a22317d
28359Author: Adam Jackson <ajax@redhat.com>
28360Date: Sat Sep 18 08:24:19 2010 -0400
28361
28362 int10: Port to pciaccess' legacy IO API
28363
28364 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28365 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28366 Signed-off-by: Adam Jackson <ajax@redhat.com>
28367 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28368
28369commit c0b63ff88a6dc20f86141aa1a1ed5a9384aa63aa
28370Author: Adam Jackson <ajax@redhat.com>
28371Date: Sat Sep 18 08:05:11 2010 -0400
28372
28373 xfree86: Move xf86GetClocks to vgahw
28374
28375 This is really a vga-specific hack anyway. The only modern driver that
28376 uses it is trident, but it's already loaded vgahw by the time it would
28377 call xf86GetClocks.
28378
28379 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28380 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
28381 Signed-off-by: Adam Jackson <ajax@redhat.com>
28382 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28383
28384commit 1433103a49b97e356da0f20aa65046cdb2f5ece6
28385Author: Dave Airlie <airlied@redhat.com>
28386Date: Wed Oct 12 09:59:38 2011 +0100
28387
28388 test: fix input test
28389
28390 The test was memsetting the wrong thing, this fixes make check in my tinderbox.
28391
28392 Signed-off-by: Dave Airlie <airlied@redhat.com>
28393 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28394 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28395
28396commit f9c6903d4a90b59c328f4fa05d2be9e0ce1c5189
28397Author: Jeremy Huddleston <jeremyhu@apple.com>
28398Date: Wed Oct 5 15:02:52 2011 -0700
28399
28400 dix: add utility functions for double to/fro FP1616/FP3232 conversion
28401
28402 Co-authored-by: Jeremy Huddleston <jeremyhu@apple.com>
28403 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28404 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28405 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28406 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
28407
28408commit 401150d7dcad08be7c1f07e076f810cd61e2105c
28409Author: Peter Hutterer <peter.hutterer@who-t.net>
28410Date: Tue Oct 4 11:41:17 2011 +1000
28411
28412 input: change pointer screen crossing behaviour for multiple ScreenRecs
28413
28414 miPointerSetPosition traditionally took coordinates on a per-screen basis,
28415 triggering a screen switch when these went out-of-bounds. For absolute
28416 devices, this prevented screen crossing in the negative x/y direction.
28417
28418 This patch changes the event generation patch to handle screen coordinates
28419 in a desktop range (i.e. all screens together). Screen switches are
28420 triggered when these coordinates are not on the current screen.
28421
28422 This unifies the pointer behaviour of single ScreenRec multihead and
28423 multiple ScreenRecs multihead in that the cursor by default moves about the
28424 whole screen rather than be confined to one single screen. The
28425 transformation matrix may then be used to actually confine the cursor to the
28426 screen again.
28427
28428 Note: fill_pointer_events has to deal with several different coordinate
28429 systems. Make sure you read the comment before trying to understand the code.
28430
28431 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28432
28433commit 6bd0eff40fae1e5d8fed28751851eb5b932d131b
28434Author: Peter Hutterer <peter.hutterer@who-t.net>
28435Date: Wed Oct 12 15:53:57 2011 +1000
28436
28437 dix: extend rescaleValuatorAxis to take a minimum default
28438
28439 Allow rescaling to non-zero based axis ranges as default (for when screen
28440 offsets are non-zero). Currently unused.
28441
28442 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28443
28444commit 8bebb4b4896d8b6ba3309b5b28fce883bb9f8a96
28445Author: Peter Hutterer <peter.hutterer@who-t.net>
28446Date: Wed Sep 7 14:04:10 2011 +1000
28447
28448 Store desktop dimensions in screenInfo.
28449
28450 For Zaphod mode screen crossing handling we need to know the size of all
28451 screens together (i.e. the whole desktop size). Store that in the screenInfo to
28452 have it readily available in events.
28453
28454 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28455
28456commit 9cbfa4739a51e4cc6b8094833928bf8678f63876
28457Author: Peter Hutterer <peter.hutterer@who-t.net>
28458Date: Tue Sep 6 15:55:19 2011 +1000
28459
28460 xfree86: remove xf86XInputSetScreen
28461
28462 Keeping track of which screen the pointer within the input driver is
28463 obsolete now. To bind to a screen, use the transformation matrix instead.
28464
28465 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28466 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28467 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28468
28469commit 09dbfcb0ad7b6c8bac94502f2801e82f2a2ef435
28470Author: Jeremy Huddleston <jeremyhu@apple.com>
28471Date: Tue Oct 11 17:37:44 2011 -0700
28472
28473 os: Remove Error()
28474
28475 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28476 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28477
28478commit db30615bcb3b872475e7d40eeee8cdda5b723078
28479Author: Jeremy Huddleston <jeremyhu@apple.com>
28480Date: Mon Oct 10 12:16:31 2011 -0700
28481
28482 Xephyr: Remove socket and its lock file on exit
28483
28484 https://bugs.freedesktop.org/show_bug.cgi?id=11484
28485
28486 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28487 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
28488
28489commit c13a48e74ec89eafa9b529a87253a631ef02f2e2
28490Author: Jeremy Huddleston <jeremyhu@apple.com>
28491Date: Mon Oct 10 12:09:39 2011 -0700
28492
28493 Xnest: Remove socket and its lock file on exit
28494
28495 https://bugs.freedesktop.org/show_bug.cgi?id=11484
28496
28497 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28498 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
28499
28500commit 32b289e46cc2d5ec32ff0c4ba5bbfbf602afb388
28501Author: Peter Hutterer <peter.hutterer@who-t.net>
28502Date: Mon Oct 3 13:58:01 2011 +1000
28503
28504 dix: move MD last.valuator update into fill_pointer_events
28505
28506 Don't update the MD where it's not expected, positionSprite should really
28507 just do that - position the sprite.
28508
28509 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28510 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28511
28512commit 3b36fd1b49030ead44358945f62e5abe7f4609ce
28513Author: Peter Hutterer <peter.hutterer@who-t.net>
28514Date: Mon Oct 3 13:10:53 2011 +1000
28515
28516 mi: switch miPointerSetPosition to take doubles
28517
28518 Don't switch between doubles and ints in the caller, instead take doubles in
28519 miPointerSetPosition and do the conversion there. For full feature we should
28520 change everything down from here for doubles too.
28521
28522 Functional change: previously we'd restore the remainder regardless of
28523 screen switching/confinement (despite what the comment said). Now,
28524 screen changing or cursor constraints will cause the remainder be clipped
28525 off. This should happen for cursor constraints but arguably not for screen
28526 crossing.
28527
28528 This also corrects a currently wrong comment about miPointerSetPosition's
28529 input coordinates.
28530
28531 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28532 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28533
28534commit 81cfe44b1ed0de84ad1941fe2ca74bebef3fc58d
28535Author: Peter Hutterer <peter.hutterer@who-t.net>
28536Date: Mon Oct 3 12:49:49 2011 +1000
28537
28538 mi: return the screen from miPointerSetPosition
28539
28540 miPointerSetPosition may switch screens. Always return the screen the sprite
28541 is on instead of relying on callers to call miPointerGetScreen().
28542
28543 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28544 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28545 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28546
28547commit 88dfe5366d9855e0ebf8bbff74967b793ede57d1
28548Author: Peter Hutterer <peter.hutterer@who-t.net>
28549Date: Mon Oct 3 12:37:28 2011 +1000
28550
28551 dix: drop screen argument from positionSprite
28552
28553 We can just get this in the function, no effective functional changes.
28554
28555 Also return the screen to the caller. Though we don't use it yet, we will in
28556 a follow-up patch.
28557
28558 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28559 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28560 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28561
28562commit 967bc25da221a69c8fc390253465145ce534fcb9
28563Author: Peter Hutterer <peter.hutterer@who-t.net>
28564Date: Mon Oct 3 11:42:08 2011 +1000
28565
28566 dix: move screen- to device coordinate scaling to separate function
28567
28568 No functional changes.
28569
28570 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28571 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28572 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28573
28574commit b059e06e19ac9417ceeb8be58c1c91b159291865
28575Author: Peter Hutterer <peter.hutterer@who-t.net>
28576Date: Mon Aug 29 12:36:26 2011 +1000
28577
28578 dix: don't allow keyboard devices to submit motion or button events.
28579
28580 GPE unconditionally dereferences pDev->valuator if a mask is present. This
28581 shouldn't really happen but if it does, don't crash, just ignore the events
28582 with an error.
28583
28584 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28585 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28586 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28587
28588commit bccff533184a051b614a26304ce77ad30bede5e0
28589Author: Peter Hutterer <peter.hutterer@who-t.net>
28590Date: Mon Oct 3 12:19:21 2011 +1000
28591
28592 dix: moveRelative modifies parameter in-place, say so.
28593
28594 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28595 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28596 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28597
28598commit b966362ccf0fe6fdd44f4d778d47e3677f55f11b
28599Author: Peter Hutterer <peter.hutterer@who-t.net>
28600Date: Mon Oct 3 12:18:20 2011 +1000
28601
28602 dix: rename moveAbsolute to clipAbsolute
28603
28604 Let's be honest about what it does.
28605
28606 moveRelative accumulates delta _and_ clips in some cases, so that one can
28607 keep it's name.
28608
28609 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28610 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28611 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28612
28613commit 959d18c3765e447897a8cfd358e9ee645df595d9
28614Author: Peter Hutterer <peter.hutterer@who-t.net>
28615Date: Fri Sep 30 10:50:51 2011 +1000
28616
28617 dix: fix missing verb in comment
28618
28619 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28620 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28621 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28622
28623commit 7074ec87bdf81699df172619aea7aae1ad4ec3c6
28624Author: Peter Hutterer <peter.hutterer@who-t.net>
28625Date: Fri Sep 30 10:47:00 2011 +1000
28626
28627 dix: document transformAbsolute
28628
28629 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28630 Reviewed-by: Jamey Sharp <jamey@minilop.net>
28631 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28632
28633commit f4ca19ce3ab91a9c8ad9de60f7dc95466f21f589
28634Author: Peter Hutterer <peter.hutterer@who-t.net>
28635Date: Fri Jul 29 10:56:44 2011 +1000
28636
28637 dix: copy the source ID into the RawEvent (#34240)
28638
28639 X.Org Bug 34240 <http://bugs.freedesktop.org/show_bug.cgi?id=34240>
28640
28641 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28642 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28643
28644commit 8b29addc30163bb7ba74c74351f312aac02f049a
28645Author: Jeremy Huddleston <jeremyhu@apple.com>
28646Date: Sun Oct 9 04:00:41 2011 -0700
28647
28648 loader: when creating sdksyms.c only include shmint.h if MITSHM is enabled #29109
28649
28650 https://bugs.freedesktop.org/show_bug.cgi?id=29109
28651
28652 When configured with --disable-mitshm the symbols declared in shmint.h
28653 do not exist. By guarding the include with '#ifdef MITSHM' these
28654 symbols are skipped when generating sdksyms.c with --disable-mitshm.
28655
28656 Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de>
28657 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28658
28659commit eabd5532fb589bd0550344ebd23c9765e591aadf
28660Author: Julien Cristau <jcristau@debian.org>
28661Date: Sun Oct 9 17:06:14 2011 -0700
28662
28663 xfree86: fix build with xv disabled
28664
28665 https://bugs.freedesktop.org/show_bug.cgi?id=29111
28666
28667 Signed-off-by: Julien Cristau <jcristau@debian.org>
28668 Tested-by: Cyril Brulebois <kibi@debian.org>
28669 Reviewed-by: Cyril Brulebois <kibi@debian.org>
28670
28671commit c53380be804852045722203c993379813d97f6d7
28672Author: Ville Skyttä <ville.skytta@iki.fi>
28673Date: Sun Mar 6 10:18:30 2011 +0200
28674
28675 Man page syntax and spelling fixes.
28676
28677 Signed-off-by: Ville Skyttä <ville.skytta@iki.fi>
28678 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28679
28680commit 765ef69295ddc473640c96f1b4f54e0b8bfc670e
28681Author: Max Schwarz <max.schwarz@online.de>
28682Date: Tue Oct 4 22:06:08 2011 +0200
28683
28684 dix: fix inverted handling of legacy scroll button events
28685
28686 This bug led to inverted scrolling axes with drivers that support smooth
28687 scrolling axes but send legacy button events.
28688
28689 Signed-off-by: Max Schwarz <Max@x-quadraht.de>
28690 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28691 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28692
28693commit fae7ed62ad476114bd37b566202cf6b6546c0a1f
28694Author: Jeremy Huddleston <jeremyhu@apple.com>
28695Date: Tue Oct 4 21:40:03 2011 -0700
28696
28697 XQuartz: pbproxy: Add missing AM_OBJCFLAGS
28698
28699 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
28700
28701commit e4cddf509e1729e8ff40354275b65455111ad2bd
28702Author: vdb@picaros.org <vdb@picaros.org>
28703Date: Sat Sep 17 18:55:47 2011 +0200
28704
28705 Fix a rare memory leak
28706
28707 Signed-off-by: Servaas Vandenberghe <vdb@picaros.org>
28708 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28709
28710commit 53611213396479abfdce0f7752681572e0d26d78
28711Author: Kirill Elagin <kirelagin@gmail.com>
28712Date: Tue Oct 4 23:02:20 2011 +0400
28713
28714 Fix server crash due to invalid images
28715
28716 See https://bugs.freedesktop.org/show_bug.cgi?id=39383
28717
28718 Signed-off-by: Kirill Elagin <kirelagin@gmail.com>
28719 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28720
28721commit 7a33c5b934382b5554f41ab1502514e91c9bc52e
28722Author: Alan Hourihane <alanh@vmware.com>
28723Date: Tue Oct 4 19:42:46 2011 -0700
28724
28725 dixfonts: Don't overwrite local c variable until new_closure is safely initialized.
28726
28727 Signed-off-by: Alan Hourihane <alanh@vmware.com>
28728 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
28729
28730commit 911e7368bf9c00d327e994a6f7a1d8d8f9b83c72
28731Author: Peter Hutterer <peter.hutterer@who-t.net>
28732Date: Tue Aug 30 16:58:04 2011 -0400
28733
28734 Move pointOnScreen to inpututils.c
28735
28736 We need this from other files too.
28737
28738 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28739 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28740
28741commit d0a7cd759d4741a1ae118d579c90704410cde244
28742Author: Peter Hutterer <peter.hutterer@who-t.net>
28743Date: Fri Sep 2 09:53:02 2011 +1000
28744
28745 dix: NewCurrentScreen must work on pointers where possible
28746
28747 When a screen switch is triggered by PointerKeys, the device for
28748 NewCurrentScreen is the keyboard. Submitting pointer events for this
28749 keyboard (without valuators) has no effect as GPE ignores the event.
28750
28751 Force the dequeuing through the XTest device attached to this device.
28752
28753 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28754 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28755
28756commit 535b3789be3a7b43b5d9026e2b5150521d91e32b
28757Author: Peter Hutterer <peter.hutterer@who-t.net>
28758Date: Wed Aug 31 14:15:02 2011 +1000
28759
28760 dix: warn about keyboard events with valuator masks
28761
28762 We don't actually handle the mask correctly. They're clipped and dropped
28763 into the event but that's about it. I don't think we did since 1.4, let's
28764 warn the user if this happens.
28765
28766 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28767 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28768
28769commit 9537afb13f2750d22350b7441570332ae60e4860
28770Author: Peter Hutterer <peter.hutterer@who-t.net>
28771Date: Tue Aug 30 13:37:31 2011 +1000
28772
28773 dix: fill out root_x/y for keyboard events
28774
28775 Switching screens relies on rootx/y to be set to the correct value. Note:
28776 though we technically take a mask for GetKeyboardEvents we don't actually
28777 handle it properly to move the pointer as required (and generate motion
28778 events if needed).
28779
28780 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
28781 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
28782
28783commit 6378d0233d21088b6429755627b4253859892c72
28784Merge: 6e965d8 98b2306
28785Author: Keith Packard <keithp@keithp.com>
28786Date: Mon Oct 3 13:56:06 2011 -0700
28787
28788 Merge remote-tracking branch 'herrb/master'
28789
28790commit 6e965d8a185087a55b2100b817e18f05b3ce5a00
28791Merge: 466e4b3 ac5881d
28792Author: Keith Packard <keithp@keithp.com>
28793Date: Mon Oct 3 13:47:49 2011 -0700
28794
28795 Merge remote-tracking branch 'koba/reviewed'
28796
28797commit 466e4b3f174bdd7a69a76eb06e6c3e3344b8adc2
28798Author: Gaetan Nadon <memsize@videotron.ca>
28799Date: Fri Sep 30 07:21:57 2011 -0400
28800
28801 Xserver-spec: use appropriate copyright markup
28802
28803 Use docbook copyright markup for year and holder.
28804
28805 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28806 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28807 Signed-off-by: Keith Packard <keithp@keithp.com>
28808
28809commit 62940bf38f70e3d9fb80521de40d6361548a0ea5
28810Author: Gaetan Nadon <memsize@videotron.ca>
28811Date: Fri Sep 30 07:21:56 2011 -0400
28812
28813 Xserver-spec: re-order markup to follow outpout order
28814
28815 No content change, no layout/ordering change.
28816 It is easier to find <title> at the top.
28817
28818 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
28819 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28820 Signed-off-by: Keith Packard <keithp@keithp.com>
28821
28822commit 365bc2f5b132ac76a1f827923a99a75b33b158ac
28823Author: Gaetan Nadon <memsize@videotron.ca>
28824Date: Fri Sep 30 07:21:55 2011 -0400
28825
28826 Xserver-spec: add release information
28827
28828 It has been added in all docs (X Version 11, Release 7.6)
28829
28830 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28831 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28832 Signed-off-by: Keith Packard <keithp@keithp.com>
28833
28834commit ee3e26079e120bf9914cc27871d50501b35e75df
28835Author: Gaetan Nadon <memsize@videotron.ca>
28836Date: Fri Sep 30 07:16:34 2011 -0400
28837
28838 ddxDesign: drop the url in the coporate authors list
28839
28840 It was such an eyesore once rendered in html.
28841 Now it looks like other authors.
28842
28843 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
28844 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28845 Signed-off-by: Keith Packard <keithp@keithp.com>
28846
28847commit 0ed71b48c0c7bd65cd05eda6c06b9c29b4e3f607
28848Author: Gaetan Nadon <memsize@videotron.ca>
28849Date: Fri Sep 30 07:16:33 2011 -0400
28850
28851 ddxDesign: remove server version from the document title
28852
28853 Such version information is already written in the appropriate location
28854
28855 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28856 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28857 Signed-off-by: Keith Packard <keithp@keithp.com>
28858
28859commit 83ec3bf1aa72848421660c44c4503fec655f5924
28860Author: Gaetan Nadon <memsize@videotron.ca>
28861Date: Fri Sep 30 07:16:32 2011 -0400
28862
28863 ddx-Design: add X Version and Release information
28864
28865 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
28866 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28867 Signed-off-by: Keith Packard <keithp@keithp.com>
28868
28869commit 2f09f6e3f610e585aeb5bd1023323a909d1ba20d
28870Author: Gaetan Nadon <memsize@videotron.ca>
28871Date: Thu Sep 29 07:53:16 2011 -0400
28872
28873 Xinput: use appropriate copyright markup for a multi licensed doc
28874
28875 Removed a duplicate paragraph.
28876
28877 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
28878 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28879 Signed-off-by: Keith Packard <keithp@keithp.com>
28880
28881commit 597348e969bf78dc2de4cf9bae7b76c110144266
28882Author: Gaetan Nadon <memsize@videotron.ca>
28883Date: Thu Sep 29 07:53:15 2011 -0400
28884
28885 Xinput: replace hard coded X11 version with X server version
28886
28887 Which is more useful.
28888
28889 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
28890 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
28891 Signed-off-by: Keith Packard <keithp@keithp.com>
28892
28893commit fad04dd9701aa76249a0bab31676103debb84307
28894Author: Lennart Poettering <lennart@poettering.net>
28895Date: Tue Sep 27 18:04:06 2011 +0100
28896
28897 udev: make use of udev_enumerate_add_match_tag() only when it is available
28898
28899 udev_enumerate_add_match_tag() and udev_monitor_filter_add_match_tag()
28900 are mostly optimizations, hence simply skip these calls if they are not
28901 available in the installed version of libudev.
28902
28903 This should fix the build on older versions of udev.
28904
28905 [airlied: fixes tinderbox failures on RHEL6]
28906
28907 Signed-off-by: Lennart Poettering <lennart@poettering.net>
28908 Reviewed-by: Dave Airlie <airlied@redhat.com>
28909 Tested-by: Dave Airlie <airlied@redhat.com>
28910 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28911 Signed-off-by: Keith Packard <keithp@keithp.com>
28912
28913commit 77743f877d573cdf70fcb5cb5897250f668ebf9b
28914Author: Jamey Sharp <jamey@minilop.net>
28915Date: Mon Sep 26 22:38:44 2011 -0700
28916
28917 kdrive: Fix build for opaque InputOption structure.
28918
28919 Commit 05284a03f9002b03a66ae355b34790ec02b726f0 missed fixing up
28920 kdrive's use of the old non-opaque structure.
28921
28922 Signed-off-by: Jamey Sharp <jamey@minilop.net>
28923 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
28924 Signed-off-by: Keith Packard <keithp@keithp.com>
28925
28926commit 9a55b3661b4172f9a32cd6fefb254e06edb1a88a
28927Merge: cf11ca3 84bb020
28928Author: Keith Packard <keithp@keithp.com>
28929Date: Mon Oct 3 11:49:23 2011 -0700
28930
28931 Merge remote-tracking branch 'alanc/master'
28932
28933commit cf11ca360c2ded5ca309faa9d039160947387bb8
28934Merge: 463dd87 6eae9fa
28935Author: Keith Packard <keithp@keithp.com>
28936Date: Mon Oct 3 11:44:59 2011 -0700
28937
28938 Merge remote-tracking branch 'jeremyhu/for-keith'
28939
28940commit 463dd87062edaa974cb8c7163328a53197f237e1
28941Author: Pauli Nieminen <pauli.nieminen@linux.intel.com>
28942Date: Thu Sep 22 18:38:53 2011 +0300
28943
28944 xf86/modes: Fix shadow rotation crashing when screen pixmap changes
28945
28946 Driver may change screen pixmaps after page flipping that would then
28947 make damage lose track of the root pixmap.
28948
28949 Using root window for shadow damages fixes the problem because
28950 SetWindowPixmap is implemented in shadow code.
28951
28952 Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
28953 Reviewed-by: Keith Packard <keithp@keithp.com>
28954 Signed-off-by: Keith Packard <keithp@keithp.com>
28955
28956commit f5d50b46ddeb039ab6564141f61261e94ee67637
28957Merge: 57cd32e 524e544
28958Author: Keith Packard <keithp@keithp.com>
28959Date: Mon Oct 3 11:36:28 2011 -0700
28960
28961 Merge remote-tracking branch 'whot/next'
28962
28963commit 57cd32e93425597317b4b7722859155419836e4c
28964Author: Aaron Plattner <aplattner@nvidia.com>
28965Date: Thu Aug 25 15:41:55 2011 -0700
28966
28967 xfree86/modes: Make cursor position transform a helper function
28968
28969 When the driver can handle the crtc transform in hardware, it sets
28970 crtc->driverIsPerformingTransform, which turns off both the shadow
28971 layer and the cursor's position-transforming code. However, some
28972 drivers actually do require the cursor position to still be
28973 transformed in these cases. Move the cursor position transform into a
28974 helper function that can be called by such drivers.
28975
28976 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
28977
28978commit 245cb8e94fd15990e1b7d6622added460f104dba
28979Author: Aaron Plattner <aplattner@nvidia.com>
28980Date: Thu Aug 25 10:19:48 2011 -0700
28981
28982 xfree86/modes: Let the driver handle the transform
28983
28984 If a driver can use hardware to handle the crtc transform, then
28985 there's no need for the server's shadow layer to do it. Add a crtc
28986 flag that lets the driver indicate that it is handling the transform.
28987 If it's set, consider the transformed size of the screen but don't
28988 actually enable the shadow layer. Also stop adjusting the cursor
28989 image and position.
28990
28991 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
28992
28993commit e089737fb13868bd3a72b7ac4799d502d188f03e
28994Author: Keith Packard <keithp@keithp.com>
28995Date: Mon Oct 3 11:23:16 2011 -0700
28996
28997 Add AC_LANG_SOURCE wrappers around configure.ac code fragments
28998
28999 Current autoconf versions are very unhappy when code fragments are not
29000 wrapped in AC_LANG_SOURCE macros, generating errors like:
29001
29002 configure.ac:723: warning: AC_LANG_CONFTEST: no AC_LANG_SOURCE call detected in body
29003 ../../lib/autoconf/lang.m4:194: AC_LANG_CONFTEST is expanded from...
29004 ../../lib/autoconf/general.m4:2662: _AC_LINK_IFELSE is expanded from...
29005 ../../lib/autoconf/general.m4:2679: AC_LINK_IFELSE is expanded from...
29006 ../../lib/m4sugar/m4sh.m4:606: AS_IF is expanded from...
29007 ../../lib/autoconf/general.m4:2032: AC_CACHE_VAL is expanded from...
29008 ../../lib/autoconf/general.m4:2053: AC_CACHE_CHECK is expanded from...
29009 configure.ac:723: the top level
29010
29011 Signed-off-by: Keith Packard <keithp@keithp.com>
29012
29013commit 6eae9fa28445b6cfa09061b338242df5116aafe8
29014Author: Matt Turner <mattst88@gmail.com>
29015Date: Wed Sep 28 17:27:45 2011 -0400
29016
29017 Remove incorrect & in swap_uint32
29018
29019 Caused by commit 893e86a4, and hidden by the (char *) cast.
29020
29021 Signed-off-by: Matt Turner <mattst88@gmail.com>
29022 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29023 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29024 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
29025 Tested-by: Aaron Plattner <aplattner@nvidia.com>
29026
29027commit 524e5445c0b6df5247d4aac5368470bb89ef4080
29028Author: Alan Coopersmith <alan.coopersmith@oracle.com>
29029Date: Sat Oct 1 08:43:52 2011 -0700
29030
29031 Add #include "inpututils.h" to xkbAccessX.c for init_device_event
29032
29033 Fixes Sun compiler warning:
29034 "xkbAccessX.c", line 128: warning: implicit function declaration: init_device_event
29035
29036 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29037 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29038 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29039
29040commit 98b230669f591bf5b8b8764db3b287d91857ee9e
29041Author: Matthieu Herrb <matthieu.herrb@laas.fr>
29042Date: Sun Jul 31 20:43:39 2011 +0200
29043
29044 sdksyms.sh may not be executable.
29045
29046 Use $(SHELL) to run it. Someone may want to build out of a source tree
29047 in a filesystem with the noexec mount flag set.
29048
29049 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29050 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
29051 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29052
29053commit f54852edc35a2831a4cd91906188101ea19904e1
29054Author: Matthieu Herrb <matthieu.herrb@laas.fr>
29055Date: Sun Sep 18 12:46:52 2011 +0200
29056
29057 Add a 'wscons' autoconf mechanism to configure input devices on BSD.
29058
29059 This does not really handle hotplug (it's handled inside the kernel,
29060 by the 'mux' devices), but uses the wscons console driver
29061 configuration to figure out the keyboard layout and the list of
29062 pointing devices found by the kernel.
29063
29064 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29065 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
29066
29067commit f32c827d513c44f07e1d0fbcc0c96cef18c9a4d9
29068Author: Max Schwarz <Max@x-quadraht.de>
29069Date: Sun Sep 25 20:44:26 2011 +0200
29070
29071 Input: Fix frac calculation on [Raw]DeviceEvent conversion
29072
29073 (1UL << 32) evaluates to 0 (at least here), so do the
29074 fraction calculation in two steps as in libXi. Fractions on xXIRawEvent
29075 were not multiplied at all, which also gave 0 as result.
29076
29077 Signed-off-by: Max Schwarz <Max@x-quadraht.de>
29078 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29079
29080commit 4c6bc0e76599dbe5ede2e1f48c9936a0e996b638
29081Merge: 057cc92 3304bbf
29082Author: Peter Hutterer <peter.hutterer@who-t.net>
29083Date: Fri Sep 30 09:24:56 2011 +1000
29084
29085 Merge branch 'smooth-scrolling' into next
29086
29087commit 3304bbff9b4ed63f1a47410a5320a136420ba2c6
29088Author: Peter Hutterer <peter.hutterer@who-t.net>
29089Date: Tue Feb 15 18:49:58 2011 +0000
29090
29091 Input: Add smooth-scrolling support to GetPointerEvents
29092
29093 For scroll wheel support, we used to send buttons 4/5 and 6/7 for
29094 horizontal/vertical positive/negative scroll events. For touchpads, we
29095 really want more fine-grained scroll values. GetPointerEvents now
29096 accepts both old-school scroll button presses, and new-style scroll axis
29097 events, while emitting both types of events to support both old and new
29098 clients.
29099
29100 This works with the new XIScrollClass to mark axes as scrolling axes.
29101 Drivers mark any valuators that send scroll events with SetScrollValuator.
29102 (Currently missing: the XIDeviceChangeEvent being sent when a driver changes
29103 a scroll axis at run-time. This can be added later.)
29104
29105 Note: the SCROLL_TYPE enums are intentionally different values to the XI2
29106 proto values to avoid copy/overlapping range bugs.
29107
29108 Co-authored-by: Daniel Stone <daniel@fooishbar.org>
29109 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29110 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29111 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29112
29113commit ac5881d6d0bcf4bcc66a5d6ddde4eca950db1a5b
29114Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29115Date: Thu Sep 15 23:05:24 2011 +0600
29116
29117 Remove unused vtSysreq
29118
29119 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29120 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29121 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29122
29123commit 0be1640dbbdd072f26d937de2e49f5ed58d54ef9
29124Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29125Date: Sat Aug 27 02:02:36 2011 +0600
29126
29127 bsd: ioctl KDENABIO/KDDISABIO do not matter for OpenBSD
29128
29129 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29130 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29131
29132commit 91042b98afdc6cecc746e101a88f79e56db1c0d0
29133Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29134Date: Sat Aug 27 00:47:34 2011 +0600
29135
29136 bsd: OpenBSD and NetBSD not need extra headers in PCVT_SUPPORT
29137
29138 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29139 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29140
29141commit 86eaa9bbe4ed59f89d6509d5e13211ab7f5a8038
29142Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29143Date: Sat Aug 27 01:03:28 2011 +0600
29144
29145 bsd: Some clean up
29146
29147 OpenBSD and NetBSD does not support syscons
29148
29149 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29150 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29151
29152commit dc864770d47fc1e4ec993e37a59e5aecb5b0a85e
29153Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29154Date: Fri Aug 26 22:40:19 2011 +0600
29155
29156 bsd: Remove dead code
29157
29158 Since OsInit closes stdin before the xfree86 DDX opens the
29159 console, fstat on stdin will always fail, so it's safe to delete
29160 code that attempts it.
29161
29162 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29163 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29164 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29165
29166commit cbb842666fa7ff26bbdd1e89a7f41b835a677c5e
29167Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29168Date: Fri Aug 26 21:42:41 2011 +0600
29169
29170 bsd: Variable devConsoleFd need only if defined PCCONS_SUPPORT
29171
29172 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29173 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29174 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29175
29176commit 38bd1e123d959e0a13514ec415afc8bfb5a170f2
29177Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29178Date: Fri Aug 26 18:54:01 2011 +0600
29179
29180 bsd: Replacement screenFd on consoleFd because they are equivalent
29181
29182 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29183 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29184 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29185
29186commit f015351c73bb24ce4172911c3e95d67922de42ab
29187Author: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29188Date: Fri Aug 26 18:27:02 2011 +0600
29189
29190 bsd: Remove unused macros KBD_FD
29191
29192 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
29193 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29194 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29195 Tested-by: Matthieu Herrb <matthieu.herrb@laas.fr>
29196
29197commit 057cc92ebfeebe81b7d01ff2c6aa2c066c8d1a64
29198Merge: afb1fe6 52c9b59
29199Author: Peter Hutterer <peter.hutterer@who-t.net>
29200Date: Thu Sep 29 14:18:35 2011 +1000
29201
29202 Merge branch 'raw-events' into next
29203
29204 Conflicts:
29205 configure.ac
29206 dix/events.c
29207
29208commit 52c9b59a9fed6abfeca775a7a04133cee18eac0b
29209Author: Peter Hutterer <peter.hutterer@who-t.net>
29210Date: Mon May 30 09:47:06 2011 +1000
29211
29212 dix: use 'rc' for return code in DeliverRawEvent
29213
29214 Reported-by: Walter Harms <wharms@bfs.de>
29215 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29216 Reviewed-by: Cyril Brulebois <kibi@debian.org>
29217 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29218 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29219
29220commit b6b1f1b514bf14ddf6b265b8d4551f892ded0dd2
29221Author: Peter Hutterer <peter.hutterer@who-t.net>
29222Date: Wed May 11 13:56:08 2011 +1000
29223
29224 input: deliver raw events unconditionally for XI 2.1 clients.
29225
29226 Deliver raw events regardless whether there is a grab on or not for clients
29227 supporting 2.1 or later.
29228
29229 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29230 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29231 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29232
29233commit 16244fba001826190445302f56784f5de9c59c01
29234Author: Peter Hutterer <peter.hutterer@who-t.net>
29235Date: Thu Jun 2 13:50:13 2011 +1000
29236
29237 Support (and require) XI 2.1
29238
29239 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29240 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29241 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29242
29243commit 54e05d80a122bac57920fce1704c0f57492b849c
29244Author: Peter Hutterer <peter.hutterer@who-t.net>
29245Date: Fri May 27 14:21:39 2011 +1000
29246
29247 dix: rename ProcessRawEvents to dix/events.c:DeliverRawEvent
29248
29249 No functional changes, prep work for future changes.
29250
29251 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29252 Reviewed-by: Cyril Brulebois <kibi@debian.org>
29253 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29254 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29255
29256commit c48133f133c64d356a1208f185619bfdf7e9a5f2
29257Author: Peter Hutterer <peter.hutterer@who-t.net>
29258Date: Fri May 27 14:19:45 2011 +1000
29259
29260 Xi: use temporary variable for filter.
29261
29262 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29263 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29264 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29265
29266commit ac0850e86f22191091a1eb07237cae9de49ee00d
29267Author: Peter Hutterer <peter.hutterer@who-t.net>
29268Date: Fri May 27 12:14:32 2011 +1000
29269
29270 dix: split DeliverEventToWindowMask up a bit more.
29271
29272 Move out the actual event delivery, it needs to be used from elsewhere.
29273
29274 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29275 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29276 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29277
29278commit 5ea2fb389fce235366e9fce83d20abdc8874f4e2
29279Author: Peter Hutterer <peter.hutterer@who-t.net>
29280Date: Fri May 27 11:43:50 2011 +1000
29281
29282 dix: rename DeliverEventsToClients to DeliverEventsToWindowMask
29283
29284 To avoid confusion with a future patch and it better describes what this
29285 does anyway - delivering events to all clients that have the event mask on
29286 the window.
29287
29288 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29289 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29290 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29291
29292commit 016413dae947fe4e8e918f728c87354fa4928275
29293Author: Peter Hutterer <peter.hutterer@who-t.net>
29294Date: Fri May 27 11:40:56 2011 +1000
29295
29296 dix: split client list retrieval out of DeliverEventToClients
29297
29298 No functional change, but "other" was renamed to "clients".
29299
29300 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29301 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29302 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29303
29304commit 537c5f631d1f8d526e63355b7d88cb1fb2793492
29305Author: Cyril Brulebois <kibi@debian.org>
29306Date: Sun Jun 5 03:21:18 2011 +0200
29307
29308 xkb: Fix case checks for Latin 4.
29309
29310 That one was missing _XkbKSLower:
29311 XK_kra: U+0138 LATIN SMALL LETTER KRA
29312
29313 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29314 Signed-off-by: Cyril Brulebois <kibi@debian.org>
29315
29316commit 0b56a7f0ad01d3a86cdd7a14d5e567a00a3553f6
29317Author: Cyril Brulebois <kibi@debian.org>
29318Date: Sun Jun 5 03:12:44 2011 +0200
29319
29320 xkb: Fix case checks for Latin 2.
29321
29322 Those ones were getting _XkbKSLower for no reasons:
29323 XK_ogonek: U+02DB OGONEK
29324 XK_doubleacute: U+02DD DOUBLE ACUTE ACCENT
29325
29326 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29327 Signed-off-by: Cyril Brulebois <kibi@debian.org>
29328
29329commit 512a9750c197437889ff7f26a8d2bd242e030745
29330Author: Cyril Brulebois <kibi@debian.org>
29331Date: Sun Jun 5 03:03:47 2011 +0200
29332
29333 xkb: Fix case checks for Latin 1.
29334
29335 That one was missing _XkbKSLower:
29336 XK_ssharp: U+00DF LATIN SMALL LETTER SHARP S
29337
29338 That one was getting _XkbKSLower for no reasons:
29339 XK_division: U+00F7 DIVISION SIGN
29340
29341 For reference, XK_multiply was already excluded from the _XkbKSUpper
29342 check, it's no big surprise XK_division has to be excluded from the
29343 _XkbKSLower check.
29344
29345 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29346 Signed-off-by: Cyril Brulebois <kibi@debian.org>
29347
29348commit 635a1f50bc65512ec491fe71b9dfa8cf9118e2c2
29349Author: Peter Hutterer <peter.hutterer@who-t.net>
29350Date: Fri Sep 23 12:04:39 2011 +1000
29351
29352 input: allow for max < min for relative axes on InitValuatorAxisStruct
29353
29354 Relative axes are initialized with 0, -1 but so far this never had any
29355 effect as all users of this function (for relative axes) just set it to the
29356 defaults anyway.
29357
29358 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29359 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29360
29361commit 8d1a414cca51e7f8a93470c5aa2e998f5ca1bc9a
29362Author: Peter Hutterer <peter.hutterer@who-t.net>
29363Date: Fri Sep 23 12:03:48 2011 +1000
29364
29365 input: switch InitValuatorAxisStruct to return Bool
29366
29367 Return errors instead of silently ignoring them.
29368
29369 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29370 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
29371
29372commit 5ac1f885f5c106b970835de36e13a9a7ea413df4
29373Author: Daniel Stone <daniel@fooishbar.org>
29374Date: Tue Feb 15 19:44:53 2011 +0000
29375
29376 Input: Add POINTER_EMULATED flag to GetPointerEvents
29377
29378 POINTER_EMULATED merely sets XIPointerEmulated in the generated
29379 DeviceEvent.
29380
29381 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29382 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29383
29384commit 78d8d6dd7f31cb903d5c8baf64181795736f33ed
29385Author: Daniel Stone <daniel@fooishbar.org>
29386Date: Tue Feb 15 14:23:25 2011 +0000
29387
29388 Input: Add vertical and horizontal scroll axes
29389
29390 To be used for smooth scrolling with future driver APIs, replacing
29391 Rel Vert Wheel and Rel Horiz Wheel axes, which have not been used in any
29392 open driver to date.
29393
29394 Combined with double-granularity ValuatorMasks, these axes allow for
29395 fine-grained scroll data to be sent to clients. Future commits allow
29396 drivers to post these scroll axes to
29397 QueuePointerEvents/GetPointerEvents, which take care of emulating legacy
29398 scroll button events.
29399
29400 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29401
29402commit d8e42decbad4abe13265f4c546a0c561905d018f
29403Author: Daniel Stone <daniel@fooishbar.org>
29404Date: Tue Feb 15 18:54:14 2011 +0000
29405
29406 Input: Split GetPointerEvents body into a helper function
29407
29408 For smooth-scrolling support, we want GetPointerEvents to generate
29409 multiple events, so split the body of the function out into a helper
29410 function in order to call it multiple times.
29411
29412 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29413 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29414
29415commit bc8aad2376207b5ca9c74effae67fb8183222d2e
29416Author: Daniel Stone <daniel@fooishbar.org>
29417Date: Thu Mar 3 19:48:15 2011 +0000
29418
29419 Input: Set last valuators in GetPointerEvents only
29420
29421 Previously, various pieces of code, including acceleration, used to drop
29422 the values into DeviceIntRec::last.valuators. Remove all this and only
29423 do it in GetPointerEvents after all transformation, acceleration and
29424 clipping, so we're guaranteed to always have the correct values.
29425
29426 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29427 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29428
29429commit 94c19a0a72403fc522e3d05eeb57f35e111a2562
29430Author: Daniel Stone <daniel@fooishbar.org>
29431Date: Thu Mar 3 15:46:45 2011 +0000
29432
29433 Input: Convert DeviceIntRec::last to use doubles
29434
29435 Change the last real user of a split integer/fractional co-ordinate
29436 system, DeviceIntRec's last->{valuators,remainder} to just have one set
29437 of doubles.
29438
29439 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29440 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29441
29442commit 2d9beeb2174661d0a0732403ad4e49e9ff56718d
29443Author: Daniel Stone <daniel@fooishbar.org>
29444Date: Thu Mar 3 15:36:55 2011 +0000
29445
29446 Input: Make DeviceEvent use doubles internally
29447
29448 Change the DeviceEvent InternalEvent to use doubles for its valuators,
29449 instead of data and data_frac.
29450
29451 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29452 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29453
29454commit 626f487b97c99b48bf1d8c8ef4367a686639c3e6
29455Author: Daniel Stone <daniel@fooishbar.org>
29456Date: Thu Mar 3 15:06:45 2011 +0000
29457
29458 Input: Make RawDeviceEvent use doubles internally
29459
29460 Change RawDeviceEvent to use doubles for valuators internally, rather
29461 than data(_raw) and data(_raw)_frac.
29462
29463 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29464 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29465
29466commit 3b7fb0f68dc0d40c968c2cfc9bb74b1d0fb48bc8
29467Author: Daniel Stone <daniel@fooishbar.org>
29468Date: Wed Mar 2 17:18:48 2011 +0000
29469
29470 Input: Modify mask in-place in positionSprite
29471
29472 Instead of taking pointers to x and y values to modify in
29473 positionSprite, just modify the mask (as well as dev->last) in place.
29474
29475 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29476 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29477
29478commit 629a575261c08ca67324fea4c975636a1a95dc75
29479Author: Daniel Stone <daniel@fooishbar.org>
29480Date: Wed Mar 2 17:04:57 2011 +0000
29481
29482 Input: Convert positionSprite and GetPointerEvents to double
29483
29484 Use doubles internally in both of these functions, eliminating most of
29485 the remaining int co-ordinate usage.
29486
29487 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29488 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29489
29490commit 51437995a5041a8c53c33b508b1607c78a5fa463
29491Author: Daniel Stone <daniel@fooishbar.org>
29492Date: Wed Mar 2 16:50:55 2011 +0000
29493
29494 Input: Don't call positionSprite for non-pointer devices
29495
29496 If the device doesn't have any valuators, or if it has less than two of
29497 them, don't bother calling positionSprite. Users with one-dimensional
29498 pointing devices may be upset.
29499
29500 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29501 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29502
29503commit 4c364a312daf2b743a0a60b9907f671804a1b1b6
29504Author: Daniel Stone <daniel@fooishbar.org>
29505Date: Wed Mar 2 16:49:53 2011 +0000
29506
29507 Input: Convert rescaleValuatorAxis to double
29508
29509 Instead of passing fractional pointers around everywhere, just pass
29510 doubles instead. Much easier.
29511
29512 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29513 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29514
29515commit 5680fa41ea3373651f7017898a307e97cf29b0d3
29516Author: Daniel Stone <daniel@fooishbar.org>
29517Date: Wed Mar 2 16:30:30 2011 +0000
29518
29519 Input: Remove x and y from moveAbsolute/moveRelative
29520
29521 Both these functions modify the mask and
29522 pDev->last.{valuators,remainder} in-place now, so there's no need to
29523 pass in pointers to local x and y values.
29524
29525 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29526 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29527
29528commit 0882b788da97c75e464eb352dac1d83c938a148e
29529Author: Daniel Stone <daniel@fooishbar.org>
29530Date: Wed Mar 2 16:27:31 2011 +0000
29531
29532 Input: Convert acceleration code to using ValuatorMask
29533
29534 Instead of passing a set of int* to the acceleration code, pass it a
29535 mask instead, which avoids an unfortunate loss of precision.
29536
29537 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29538 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29539 Reviewed-by: Simon Thum <simon.thum@gmx.de>
29540
29541commit 2b8f1d07bd42c9d3db3dbacfe6a1335e47236a6c
29542Author: Daniel Stone <daniel@fooishbar.org>
29543Date: Wed Mar 2 15:37:53 2011 +0000
29544
29545 Input: Widen pointer acceleration types to double
29546
29547 This widens almost all of the float-using code in ptrveloc.[ch] to
29548 doubles, other than values coming from properties which are specified to
29549 be floats by the property API.
29550
29551 Bumps input API to v14 as this changes the AccelScheme signature, as
29552 used by xf86-input-synaptics.
29553
29554 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29555
29556commit 8a4a4e1b8a33899653d68a21bb7ff0557a576338
29557Author: Daniel Stone <daniel@fooishbar.org>
29558Date: Thu Jun 9 15:26:55 2011 +0100
29559
29560 Input: Use trunc instead of lrintf in acceleration code
29561
29562 For compatibility with all the rest of the input code.
29563
29564 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29565 Reviewed-by: Simon Thum <simon.thum@gmx.de>
29566
29567commit e1df51421bc392d57f88c831fe0661871a8aa769
29568Author: Daniel Stone <daniel@fooishbar.org>
29569Date: Wed Mar 2 13:34:32 2011 +0000
29570
29571 Input: Set fractional member in set_raw_valuators
29572
29573 RawDeviceEvents have space for fractional valuator members, so might as
29574 well start using them.
29575
29576 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29577 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29578
29579commit cdf202250ef616f3b261db70da30e421ea59a630
29580Author: Daniel Stone <daniel@fooishbar.org>
29581Date: Wed Mar 2 14:50:29 2011 +0000
29582
29583 Input: Convert transformAbsolute to work on doubles
29584
29585 Change transformAbsolute to use doubles internally.
29586
29587 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29588 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29589
29590commit 3463078f9697fad0ee11837d80e88889fc6a28a4
29591Author: Daniel Stone <daniel@fooishbar.org>
29592Date: Wed Mar 2 13:29:24 2011 +0000
29593
29594 Input: Convert clipAxis, moveAbsolute and moveRelative to double
29595
29596 Change all these three to use doubles internally, though the outputs of
29597 moveAbsolute and moveRelative are still truncated to int.
29598
29599 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29600
29601commit 4e52cc0ef48145134cd58d357fb7289e6f8bb709
29602Author: Daniel Stone <daniel@fooishbar.org>
29603Date: Wed Mar 2 13:20:37 2011 +0000
29604
29605 Input: Prepare moveRelative for conversion to double
29606
29607 Shuffle some code around in moveRelative to make the conversion to
29608 double easier later.
29609
29610 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29611 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29612
29613commit 80fdf9ca33a8f50f98a4b1a3279e66b040353b80
29614Author: Daniel Stone <daniel@fooishbar.org>
29615Date: Wed Mar 2 13:19:00 2011 +0000
29616
29617 Input: Prepare moveAbsolute for conversion to double
29618
29619 Shuffle some code around to make moving to double easier later.
29620
29621 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29622 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29623
29624commit 6a6b4eb05c7822860e2362fa9b8441fc67055e27
29625Author: Daniel Stone <daniel@fooishbar.org>
29626Date: Wed Mar 2 17:22:59 2011 +0000
29627
29628 Input: Store clipped absolute axes in the mask
29629
29630 Change moveAbsolute to be more symmetric with moveRelative by storing a
29631 clipped axis value back in the mask, rather than just in
29632 dev->last.valuators.
29633
29634 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29635 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29636
29637commit 7e919ef5bfa94d51a06eefb150ab947bdbfb6885
29638Author: Daniel Stone <daniel@fooishbar.org>
29639Date: Wed Feb 23 17:28:18 2011 +0000
29640
29641 Input: Add double-precision valuator_mask API
29642
29643 Add API for valuator_mask that accepts and returns doubles, rather than
29644 ints. No double API is provided for set_range at the moment.
29645
29646 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29647 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29648
29649commit 79d4deb76d5612b75ef26b4549b7ade3a656687d
29650Author: Daniel Stone <daniel@fooishbar.org>
29651Date: Wed Feb 23 14:31:01 2011 +0000
29652
29653 Input: Convert ValuatorMask to double-precision internally
29654
29655 Switch the ValuatorMask struct to using doubles instead of ints for the
29656 actual values. Preserve the old int API, and (attempt to) round towards
29657 zero for values we return.
29658
29659 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29660 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29661
29662commit 60a766a96f9ec97adf6d0273f70ec23f313a11c6
29663Author: Daniel Stone <daniel@fooishbar.org>
29664Date: Thu Jun 9 16:02:47 2011 +0100
29665
29666 Input: Add flags to RawDeviceEvent
29667
29668 Add a flags member which will be copied wholesale into the resultant
29669 xXIRawEvent.
29670
29671 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29672
29673commit 5d62c32981dbe9630bed32567b55008b536c81c5
29674Author: Daniel Stone <daniel@fooishbar.org>
29675Date: Mon Dec 27 13:24:35 2010 +0000
29676
29677 Input: Add flags to DeviceEvent
29678
29679 Add a flags member which will be copied wholesale into the resultant
29680 xXIDeviceEvent.
29681
29682 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
29683 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29684 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
29685 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29686
29687commit 38e9e28ba2fbffee52ad9889ef6d4e94c7af3e10
29688Author: Jeremy Huddleston <jeremyhu@apple.com>
29689Date: Tue Sep 20 20:39:06 2011 -0700
29690
29691 XQuartz: Use set_front_process rather than X11ApplicationSetFrontProcess since we're already in the AppKit thread
29692
29693 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
29694
29695commit 206b30ebc608d29a91cc18665d89e887cac4dba3
29696Author: Jeremy Huddleston <jeremyhu@apple.com>
29697Date: Thu Sep 22 18:55:19 2011 -0700
29698
29699 Address regressions from e8ff555b95ba and d206d52f657c to work with other compilers
29700
29701 1) The error attribute appeared in gcc-4.3
29702 2) The return type of __builtin_constant_p is int
29703 3) Sun Studio 12.0 and later builtin support for __builtin_constant_p
29704
29705 Found by Tinderbox.
29706
29707 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
29708 Reviewed-by: Matt Turner <mattst88@gmail.com>
29709 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29710 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
29711
29712commit afb1fe695d197187a301c19863a128a65389b15c
29713Merge: 7fb4bef c7163fd
29714Author: Keith Packard <keithp@keithp.com>
29715Date: Mon Sep 26 20:24:15 2011 -0700
29716
29717 Merge remote-tracking branch 'whot/next'
29718
29719commit c7163fdd302f706a3d67f0fdf93eeb3396bb3332
29720Author: vdb@picaros.org <vdb@picaros.org>
29721Date: Wed Aug 31 07:23:01 2011 +0200
29722
29723 xfree86: .BI style: monitor section in xorg.conf man page
29724
29725 The xorg.conf manual uses the following convention in most of its
29726 sections:
29727
29728 bold = text to be copied literally to the config file,
29729 italic = a symbolic name to be substituted by a true value.
29730
29731 Some configuration keywords seem to have been changed into generic
29732 options. Prepending Option to the manual entry swapped the
29733 bold-italic logic. This patch restores the convention in the monitor
29734 section and consists of
29735
29736 -.BI "Option " "\*qPreferredMode\*q " \*qstring\*q
29737 +.BI "Option \*qPreferredMode\*q \*q" name \*q
29738
29739 modifications.
29740
29741 Plus a few minor changes (Modes → Mode) and a typo fix.
29742
29743 Signed-off-by: Servaas Vandenberghe
29744 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29745 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29746
29747commit 4b4caecb7d2c8e0b7e4fedc95fba2a728bbc25e6
29748Author: Peter Hutterer <peter.hutterer@who-t.net>
29749Date: Tue Sep 6 09:33:43 2011 +1000
29750
29751 xfree86: expose Option "TransformationMatrix"
29752
29753 Recent changes to the server change the default absolute input device
29754 behaviour on zaphods to span the whole desktop too. Since these setups
29755 usually use an xorg.conf, allow the transformation matrix to be specified in
29756 the config as well.
29757
29758 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29759 Reviewed-by: Simon Thum <simon.thum@gmx.de>
29760
29761commit cf51424a34fb2b567a867338ab44f83b5c43251c
29762Author: Peter Hutterer <peter.hutterer@who-t.net>
29763Date: Tue Sep 6 09:39:01 2011 +1000
29764
29765 xfree86: use subheader for Pointer Acceleration parts in xorg.conf(5)
29766
29767 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29768
29769commit 9125952b407090b040bffb3d752288293137cf77
29770Author: Peter Hutterer <peter.hutterer@who-t.net>
29771Date: Mon Sep 5 15:43:14 2011 +1000
29772
29773 xfree86: fix comment typo
29774
29775 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29776
29777commit 2d2dce558d24eeea0eb011ec9ebaa6c5c2273c39
29778Author: Sam Spilsbury <sam.spilsbury@canonical.com>
29779Date: Wed Sep 14 09:58:34 2011 +0800
29780
29781 Remove the SendEvent bit (0x80) before doing range checks on event type.
29782
29783 Some extension libraries may set this bit before converting the event to
29784 wire protocol and as such range checking the event will cause an invalid
29785 BadValue error to result. As the documentation suggests the the bit
29786 should be "forced on", remove it before doing range checks and continue
29787 to force it on in the server.
29788
29789 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29790 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29791
29792commit f52d5cd374563544dafe29587411f345e31bbdf8
29793Author: Carlos Garnacho <carlosg@gnome.org>
29794Date: Wed Aug 31 00:46:52 2011 +0200
29795
29796 Xi: Fix passive XI2 ungrabs on XIAll[Master]Devices
29797
29798 The corresponding DeviceIntPtr wasn't being gotten properly,
29799 resulting in BadDevice from dixLookupDevice().
29800
29801 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
29802 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
29803 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29804
29805commit 24823f1ab0f16e0815ef44340894537dc4babf6a
29806Author: Peter Hutterer <peter.hutterer@who-t.net>
29807Date: Wed Aug 3 16:01:28 2011 +1000
29808
29809 xfree86: switch options from pointer to XF86OptionPtr
29810
29811 In all cases, the pointer was simply type-cast anyway. Let's get some
29812 compile-time type safety going, how about that.
29813
29814 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29815 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
29816
29817 Squashed in:
29818 xfree86: Move definition of xf86OptionPtr into separate header file
29819
29820 The pile of spaghettis that is the xfree86 include dependencies make it
29821 rather hard to have a single typedef somewhere that's not interfering with
29822 everything else or drags in a whole bunch of other includes.
29823
29824 Move the xf86OptionRec and GenericListRec declarations into a separate
29825 header.
29826
29827 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
29828 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29829
29830commit 84bb0207f6db433a8d387d933393357da87ca23e
29831Author: Gaetan Nadon <memsize@videotron.ca>
29832Date: Fri Sep 23 20:03:03 2011 -0400
29833
29834 dix and os: gitignore dix.O and os.O
29835
29836 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29837 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
29838 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29839
29840commit 3d41939d0349569dcfd2aef9b846c739d0ba0060
29841Author: Alan Coopersmith <alan.coopersmith@oracle.com>
29842Date: Thu Sep 22 17:41:38 2011 -0700
29843
29844 Unconditionally #include <stdint.h>
29845
29846 The more recent inclusions of this file haven't been checking for
29847 HAVE_STDINT_H, so might as well make the older ones consistent.
29848
29849 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29850 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
29851 Reviewed-by: Jamey Sharp <jamey@minilop.net>
29852
29853commit 46d5ae5b551cc926331324073f2686b47e8aea07
29854Author: Gaetan Nadon <memsize@videotron.ca>
29855Date: Fri Sep 23 09:38:25 2011 -0400
29856
29857 dtrace: fix typo in title
29858
29859 The word provider should be capitalized.
29860
29861 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29862 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
29863 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29864
29865commit 6cae6224994e3252b3b7c7608f983ce73e0206fc
29866Author: Gaetan Nadon <memsize@videotron.ca>
29867Date: Fri Sep 23 09:38:24 2011 -0400
29868
29869 dtrace: use docbook copyright markup for copyright holder
29870
29871 No content change to copyright text.
29872
29873 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29874 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
29875 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29876
29877commit 58d3d46d6cc0d7a81db6bad61d158f3b9af8af71
29878Author: Alan Coopersmith <alan.coopersmith@oracle.com>
29879Date: Thu Sep 1 16:48:37 2011 -0700
29880
29881 Assign ids to more tags in Xserver-Dtrace.xml
29882
29883 Keeps xsltproc from adding random-number id tags to html output so that
29884 it's easier to see real changes vs. different random numbers.
29885
29886 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
29887 Reviewed-by: Matt Dew <marcoz@osource.org>
29888
29889commit 7fb4bef0394a5d09680985d34bce8252b61493cb
29890Merge: 98f4940 c90903b
29891Author: Keith Packard <keithp@keithp.com>
29892Date: Wed Sep 21 14:34:27 2011 -0700
29893
29894 Merge remote-tracking branch 'mattst88/for-keith'
29895
29896commit 98f4940093dc90c3366d3e4d3ddceacbaee299dc
29897Merge: b018b81 c8eacae
29898Author: Keith Packard <keithp@keithp.com>
29899Date: Wed Sep 21 14:30:19 2011 -0700
29900
29901 Merge remote-tracking branch 'alanc/master'
29902
29903commit b018b815333b256b5e06dd6251b2e539f09f89fa
29904Merge: 03d0329 0f380a5
29905Author: Keith Packard <keithp@keithp.com>
29906Date: Wed Sep 21 14:17:14 2011 -0700
29907
29908 Merge remote-tracking branch 'jamey/reviewed'
29909
29910commit c90903b4f7a826ae6826a8dd0a901c8362500e46
29911Author: Peter Harris <pharris@opentext.com>
29912Date: Fri Sep 2 18:45:16 2011 -0400
29913
29914 xkb: add missing swaps for xkbGetDeviceInfoReply
29915
29916 Caught during review of e095369bf.
29917
29918 Signed-off-by: Peter Harris <pharris@opentext.com>
29919 Reviewed-by-by: Matt Turner <mattst88@gmail.com>
29920 Signed-off-by: Matt Turner <mattst88@gmail.com>
29921
29922commit d206d52f657cb63a0f6ba8b62d788c1812b57c81
29923Author: Matt Turner <mattst88@gmail.com>
29924Date: Tue Aug 16 19:19:24 2011 -0400
29925
29926 Use __builtin_constant_p to determine if we can use lswapl
29927
29928 If the address of the swapped memory location is known at compile time,
29929 we can check its alignment at no runtime cost and use lswapl instead.
29930
29931 text data bss dec hex filename
29932 before: 1872820 52136 78040 2002996 1e9034 hw/xfree86/Xorg
29933 after: 1864396 52136 78040 1994572 1e6f4c hw/xfree86/Xorg
29934
29935 bswap instructions: 131 -> 308 (used in lswapl)
29936 rol instructions: 943 -> 1174 (used in lswaps)
29937
29938 Reviewed-by: Peter Harris <pharris@opentext.com>
29939 Signed-off-by: Matt Turner <mattst88@gmail.com>
29940
29941commit 889b700e7760ced38bcf5f3aff2d31d3d9a058d7
29942Author: Matt Turner <mattst88@gmail.com>
29943Date: Tue Aug 16 19:12:21 2011 -0400
29944
29945 Use lswap{l,s} in cpswap{l,s}
29946
29947 Should be safe since cpswap isn't used on pointers.
29948
29949 text data bss dec hex filename
29950 before: 1875588 52136 78040 2005764 1e9b04 hw/xfree86/Xorg
29951 after: 1872820 52136 78040 2002996 1e9034 hw/xfree86/Xorg
29952
29953 bswap instructions: 5 -> 131 (used in lswapl)
29954 rol instructions: 811 -> 943 (used in lswaps)
29955
29956 Reviewed-by: Peter Harris <pharris@opentext.com>
29957 Signed-off-by: Matt Turner <mattst88@gmail.com>
29958
29959commit a2f0ff5f73db204a9d61e65148b28f6acc5121df
29960Author: Matt Turner <mattst88@gmail.com>
29961Date: Tue Aug 16 19:30:20 2011 -0400
29962
29963 Make lswap{l,s} inline functions
29964
29965 text data bss dec hex filename
29966 before: 1875668 52136 78040 2005844 1e9b54 hw/xfree86/Xorg
29967 after: 1875588 52136 78040 2005764 1e9b04 hw/xfree86/Xorg
29968
29969 Reviewed-by: Peter Harris <pharris@opentext.com>
29970 Signed-off-by: Matt Turner <mattst88@gmail.com>
29971
29972commit e8ff555b95baab66cc7d060c1e7f9fdd49d3802f
29973Author: Matt Turner <mattst88@gmail.com>
29974Date: Tue Aug 16 19:07:24 2011 -0400
29975
29976 Add type checking to swap macros
29977
29978 The original macros are retained (instead of replacing them with inline
29979 functions) because of implicit type promotion. That is, an int16 passed
29980 to an inline function taking int32 would be implicitly promoted to int32
29981 without a warning.
29982
29983 Reviewed-by: Peter Harris <pharris@opentext.com>
29984 Signed-off-by: Matt Turner <mattst88@gmail.com>
29985
29986commit 893e86a49e3e381cff48a9e86dc2d9b3d5431d95
29987Author: Matt Turner <mattst88@gmail.com>
29988Date: Tue Aug 16 19:03:26 2011 -0400
29989
29990 Introduce swap_uint{16,32} functions, used in swap{l,s}
29991
29992 Reviewed-by: Peter Harris <pharris@opentext.com>
29993 Signed-off-by: Matt Turner <mattst88@gmail.com>
29994
29995commit 54770c980cd2b91a8377f975a58ed69def5cfa42
29996Author: Matt Turner <mattst88@gmail.com>
29997Date: Tue Aug 16 16:59:07 2011 -0400
29998
29999 Cast char* buffers to swap functions
30000
30001 Reviewed-by: Peter Harris <pharris@opentext.com>
30002 Signed-off-by: Matt Turner <mattst88@gmail.com>
30003
30004commit 9edcae78c46286baff42e74bfe26f6ae4d00fe01
30005Author: Matt Turner <mattst88@gmail.com>
30006Date: Wed Sep 21 17:14:16 2011 -0400
30007
30008 Use correct swap{l,s} (or none at all for CARD8)
30009
30010 Swapping the wrong size was never caught because swap{l,s} are macros.
30011
30012 It's clear in the case of Xext/xres.c, that the author believed
30013 client_major/minor to be CARD16 from looking at the code in the first
30014 hunk.
30015
30016 v2: dmx.c fixes from Keith.
30017
30018 Reviewed-by: Peter Harris <pharris@opentext.com>
30019 Signed-off-by: Matt Turner <mattst88@gmail.com>
30020
30021commit 03d032991da21bf866ca30dcbb6b7f9f99df441a
30022Author: Aaron Plattner <aplattner@nvidia.com>
30023Date: Wed Sep 14 15:25:06 2011 -0700
30024
30025 fb: Rename wfbTriangles and wfbTrapezoids
30026
30027 These symbols were not renamed when they were added to libfb:
30028
30029 # nm -D libwfb.so | grep ' fb'
30030 0000000000028d00 T fbTrapezoids
30031 0000000000028d60 T fbTriangles
30032
30033 This causes corruption and/or crashes on wfb-ful drivers like nvidia:
30034
30035 Program received signal SIGABRT, Aborted.
30036 0x00007fd67f3a0405 in *__GI_raise (sig=<optimized out>) at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
30037 64 ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
30038 in ../nptl/sysdeps/unix/sysv/linux/raise.c
30039 (gdb) bt
30040 #0 0x00007fd67f3a0405 in *__GI_raise (sig=<optimized out>) at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
30041 #1 0x00007fd67f3a3680 in *__GI_abort () at abort.c:92
30042 #2 0x00007fd67f3995b1 in *__GI___assert_fail (assertion=0x7fd679ecb804 "key->initialized", file=<optimized out>, line=116, function=0x7fd679ecbbc0 "dixGetPrivateAddr")
30043 at assert.c:81
30044 #3 0x00007fd679ec55b6 in ?? () from /usr/lib/xorg/modules/libfb.so
30045 #4 0x00007fd679eca9ef in ?? () from /usr/lib/xorg/modules/libfb.so
30046 #5 0x00007fd679ecae20 in fbTriangles () from /usr/lib/xorg/modules/libfb.so
30047 #6 0x00007fd67a58fc55 in ?? () from /usr/lib/xorg/modules/drivers/nvidia_drv.so
30048 #7 0x00000000004f38d1 in ?? ()
30049 #8 0x0000000000437ae9 in ?? ()
30050 #9 0x0000000000426eaa in ?? ()
30051 #10 0x00007fd67f38cead in __libc_start_main (main=<optimized out>, argc=<optimized out>, ubp_av=<optimized out>, init=<optimized out>, fini=<optimized out>,
30052 rtld_fini=<optimized out>, stack_end=0x7fff99860d78) at libc-start.c:228
30053 #11 0x000000000042719d in _start ()
30054
30055 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
30056 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30057 Signed-off-by: Keith Packard <keithp@keithp.com>
30058
30059commit 2c7c520cfe0df30f4bc3adba59d9c62582823bf8
30060Author: Matt Turner <mattst88@gmail.com>
30061Date: Thu Aug 4 15:35:41 2011 -0400
30062
30063 Use internal temp variable for swap macros
30064
30065 Also, fix whitespace, mainly around
30066 swaps(&rep.sequenceNumber)
30067
30068 Reviewed-by: Peter Harris <pharris@opentext.com>
30069 Signed-off-by: Matt Turner <mattst88@gmail.com>
30070
30071commit c10bad3d3e8ff1b90014770fd470f9c67263e46f
30072Author: Matt Turner <mattst88@gmail.com>
30073Date: Wed Apr 27 19:29:06 2011 -0400
30074
30075 Silence printf format warnings in helper_exec.c
30076
30077 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30078 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30079 Signed-off-by: Matt Turner <mattst88@gmail.com>
30080
30081commit 9eab5b3443a1926a29385948acc6c5e0843465ea
30082Author: Matt Turner <mattst88@gmail.com>
30083Date: Wed Apr 27 18:25:27 2011 -0400
30084
30085 Replace Fabs() macro with fabs() function
30086
30087 gcc generates better code with fabs() anyway.
30088
30089 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30090 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
30091 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30092 Signed-off-by: Matt Turner <mattst88@gmail.com>
30093
30094commit 40a47bd628f525d2d8bd3ca76554089a6e9d2a1d
30095Author: Matt Turner <mattst88@gmail.com>
30096Date: Wed Apr 27 14:19:41 2011 -0400
30097
30098 Remove unnecessary #undefs of [f]abs and old comments
30099
30100 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30101 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
30102 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30103 Signed-off-by: Matt Turner <mattst88@gmail.com>
30104
30105commit 079a7585e44d25a0a9fb6662c6382dfbe8f6cd3c
30106Author: Gaetan Nadon <memsize@videotron.ca>
30107Date: Thu Sep 15 20:12:41 2011 -0400
30108
30109 devbook.am: maintenance update from docbook.am
30110
30111 The developer docs are generated from a subset of docbook.am
30112 which is sometimes updated.
30113
30114 The one difference for xserver is the embedded css style in the HEAD element.
30115
30116 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30117 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
30118 Signed-off-by: Keith Packard <keithp@keithp.com>
30119
30120commit 82c55b5a99a1bacf2bc11bbd9d58f6f2c6c07286
30121Author: Gaetan Nadon <memsize@videotron.ca>
30122Date: Thu Sep 15 20:12:40 2011 -0400
30123
30124 docbook.am: embed css styles inside the HTML HEAD element
30125
30126 Rather than referring to the external xorg.css stylesheet, embed the content
30127 of the file in the html output produced. This is accomplished by using
30128 version 1.10 of xorg-xhtml.xsl.
30129
30130 This makes the whole html docs tree much more relocatable.
30131 In addition, it eliminates xorg.css as a runtime file which makes
30132 xorg-sgml-doctools a build time only package.
30133
30134 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30135 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
30136 Signed-off-by: Keith Packard <keithp@keithp.com>
30137
30138commit 1602444e05cc59afe32c085c0cedb85e3268cd98
30139Author: Gaetan Nadon <memsize@videotron.ca>
30140Date: Thu Sep 15 20:12:39 2011 -0400
30141
30142 docbook.am: global maintenance update - entities, images and olinking
30143
30144 Adding support in libX11 for html chunking caused a reorg of docbook.am
30145 as well as the xorg-sgml-doctools masterdb for olinking.
30146 The parameter img.src.path is added for pdf images.
30147 A searchpath to the root builddir is added for local entities, if present.
30148 This feature was initiated by xserver module and made available to all.
30149
30150 The docbook.am makefile hides all the details and is identical for
30151 all 22 modules having DocBook documentation. It is included by a thin
30152 Makefile.am which requires no docbook knowledge.
30153
30154 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30155 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
30156 Signed-off-by: Keith Packard <keithp@keithp.com>
30157
30158commit 871d65790ec2ac0fd6d5105c2d599f63a86fcaf1
30159Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
30160Date: Mon Nov 1 16:02:11 2010 +0200
30161
30162 DRI2: Allow DDX to validate swap_limit changes
30163
30164 DDX can now implement validation for swap_limit changes to prevent
30165 configurations that are not support in driver.
30166
30167 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
30168 CC: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
30169 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
30170 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
30171
30172commit b435e2aac1b3fbb97d0275de73a1e36d16f170c0
30173Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
30174Date: Mon Nov 1 16:22:00 2010 +0200
30175
30176 DRI2: Expose API to set drawable swap limit.
30177
30178 This allows ddx to set swap_limit if there is more than one back
30179 buffer for drawable. Setting swap_limit has to also check if change
30180 affects a client that is blocked.
30181
30182 This can be used to implement N-buffering in driver with minimal
30183 logic in allocation and selecting next back.
30184
30185 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
30186 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
30187 Reviewed-by: Francisco Jerez <currojerez@riseup.net>
30188 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
30189 Signed-off-by: Keith Packard <keithp@keithp.com>
30190
30191commit 86f8da0aa7612558e6563f5de0d9f9793854053f
30192Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
30193Date: Mon Nov 1 16:21:59 2010 +0200
30194
30195 DRI2: Add ReuseBufferNotify hook
30196
30197 ReuseBufferNotify hook is called whenever old buffer is reused in DRI2
30198 code.
30199
30200 Driver can use this hook to rewrite the buffer name if hardware requires
30201 shared buffers. Shared buffer might be some hardware limited resources like
30202 framebuffer that is preallocated in boot.
30203
30204 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
30205 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
30206 Signed-off-by: Keith Packard <keithp@keithp.com>
30207
30208commit 2f47433fef2cf63063e069ead9003891d135e87f
30209Author: vdb@picaros.org <vdb@picaros.org>
30210Date: Wed Sep 7 17:47:56 2011 +0200
30211
30212 print DisplayMode type bits
30213
30214 Dear,
30215
30216 A patch I posted on xorg-devel was reviewed and is ready for
30217 inclusion in xserver. Would you be willing to apply the patch so that
30218 it finds its way into the master branch ?
30219
30220 Thank you, Servaas Vandenberghe.
30221
30222 http://lists.x.org/archives/xorg-devel/2011-August/024769.html
30223 http://lists.x.org/archives/xorg-devel/2011-August/024777.html
30224
30225 This patch adds printing of the DisplayMode type bits to
30226 xf86PrintModeline(). It helps to trace the modeline origin and to
30227 understand the initial configured modeline.
30228
30229 Reviewed-by: Alex Deucher <alexander.deucher at amd.com>
30230 Signed-off-by: Servaas Vandenberghe
30231 Signed-off-by: Keith Packard <keithp@keithp.com>
30232
30233commit c92c83523ede8bc361526ac93d09d089fbbdae08
30234Merge: 4ad271d 24b2b43
30235Author: Keith Packard <keithp@keithp.com>
30236Date: Wed Sep 21 13:32:06 2011 -0700
30237
30238 Merge remote-tracking branch 'kibi/master'
30239
30240commit 4ad271d06c5aa42721c0e2e01e17e34a39825c65
30241Author: Aaron Plattner <aplattner@nvidia.com>
30242Date: Mon Aug 29 20:44:18 2011 -0700
30243
30244 xfree86: Bump extension ABI version to 6.0
30245
30246 The video driver ABI was bumped to 11.0 in commit
30247 0de7cec90738a7a5020150309866bb0e23b6f479 because of a change to the
30248 size of ATOM in commit 51f353d0a0d116af16d7d9590cadef6c56328746. This
30249 also affects extension modules, so the extension ABI version should
30250 have been bumped too.
30251
30252 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
30253 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
30254 Signed-off-by: Keith Packard <keithp@keithp.com>
30255
30256commit c8eacae4f833a0bd81907ba2b4cfaa80c08e037f
30257Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30258Date: Thu Aug 25 21:46:16 2011 -0700
30259
30260 Space & style cleanup of hw/xfree86/i2c/fi1236.c
30261
30262 Fortunately, the massive decrease in the cost of whitespace in the past
30263 decade has allowed us to be much more generous with it, and much more
30264 consistent in its application, even for code like this that clearly no
30265 one has ever tried to read.
30266
30267 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30268 Reviewed-by: Jamey Sharp <jamey@minilop.net>
30269
30270commit c46215c100e964665580211a5d5893558ca9374f
30271Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30272Date: Wed Aug 10 11:17:46 2011 -0700
30273
30274 Cross-reference cvt(1) & gtf(1) man pages
30275
30276 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30277 Reviewed-by: Julien Cristau <jcristau@debian.org>
30278
30279commit 1526f1caf396fefc5f09eaff884d5b92aac44997
30280Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30281Date: Tue May 31 19:54:33 2011 -0700
30282
30283 Remove unused ClientStateAuthenticating
30284
30285 Appears to be leftover from the Kerberos code deleted in 2007
30286 (commit dfbe32b5b828cc4e3da36a0e2e6ad641164eaa5e).
30287 Nothing left ever set clientState to ClientStateAuthenticating
30288
30289 Skipped over 1 to preserve existing enum numbering.
30290
30291 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30292 Reviewed-by: Julien Cristau <jcristau@debian.org>
30293
30294commit 0f380a5005f800572773cd4667ce43c7459cc467
30295Author: Jamey Sharp <jamey@minilop.net>
30296Date: Tue Sep 14 18:35:21 2010 -0700
30297
30298 Fix pixmap double-frees on error paths.
30299
30300 If AddResource fails, it will automatically free the object that was
30301 passed to it by calling the appropriate deleteFunc; and of course
30302 FreeResource also calls the deleteFunc. In both cases it's wrong to call
30303 the destroy hook manually.
30304
30305 Commit by Jamey Sharp and Josh Triplett.
30306
30307 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30308 Signed-off-by: Josh Triplett <josh@joshtriplett.org>
30309 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
30310 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
30311
30312commit 8f69c935f6d06ad92fd0e8d9fcb3cde86cd258f5
30313Author: Jamey Sharp <jamey@minilop.net>
30314Date: Fri Sep 17 02:27:59 2010 +0200
30315
30316 Eliminate MAXSCREENS-sized CursorScreenKey array.
30317
30318 Use new per-screen privates API instead.
30319
30320 Commit by Jamey Sharp and Josh Triplett.
30321
30322 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30323 Signed-off-by: Josh Triplett <josh@joshtriplett.org>
30324 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
30325
30326commit c96a43888ceae72bf89c66de911120c0b6b7a71a
30327Author: Jamey Sharp <jamey@minilop.net>
30328Date: Sat Jul 10 09:28:47 2010 -0700
30329
30330 Revert "Bug #6924: Restore the ABI for DrawableRec and ColormapRec to the state"
30331
30332 This reverts commit bc0c56c407117d1545e20d21f7d30eb3472d618b since we're
30333 breaking ABI anyway.
30334
30335 Conflicts:
30336
30337 ChangeLog
30338 dix/colormap.c
30339 dix/pixmap.c
30340 dix/window.c
30341 include/pixmapstr.h
30342
30343 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30344 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
30345
30346commit 13ac1ba480ee3e89163825cd2777f9a6e9dcbc9f
30347Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30348Date: Tue May 31 19:42:48 2011 -0700
30349
30350 Remove unused ClientStateCheckingSecurity & ClientStateCheckedSecurity
30351
30352 Appear to be leftovers from the XC-QUERY-SECURITY code deleted in 2007
30353 (commit 375864cb74cced40ae688078b1f7750998972535).
30354 Nothing left ever set clientState to ClientStateCheckingSecurity.
30355
30356 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30357 Reviewed-by: Julien Cristau <jcristau@debian.org>
30358
30359commit 78f946c297081d39d312e05928146c78f59fedcb
30360Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30361Date: Thu May 19 23:26:31 2011 -0700
30362
30363 sun_init.c: Implement novtswitch & sharevts for Solaris
30364
30365 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30366 Reviewed-by: John Martin <john.m.martin@oracle.com>
30367
30368commit 87cc83474d9890d14d9bc60926a97bd0d5b059af
30369Author: Jamey Sharp <jamey@minilop.net>
30370Date: Sun Sep 18 20:47:04 2011 -0500
30371
30372 Replace XmuSnprintf with snprintf.
30373
30374 Alan Coopersmith explains:
30375
30376 XmuSnprintf() can be replaced by snprintf() now. (It was a
30377 implementation X provided for it's libraries to use in the days
30378 before all platforms we cared about had snprintf in libc.)
30379
30380 Reported-by: walter harms <wharms@bfs.de>
30381 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30382 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30383
30384commit df4eeb1256c7c5e7a3f88f10da2915c7a07a8db8
30385Author: Jamey Sharp <jamey@minilop.net>
30386Date: Tue Jun 8 13:58:04 2010 -0700
30387
30388 xnest: Delete unused nClipRects GC-private field.
30389
30390 This field was never read at any time in the git history.
30391
30392 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30393 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30394
30395commit 53a4744b69ab76ecabdde5dabfb56830570c0f4e
30396Author: Jamey Sharp <jamey@minilop.net>
30397Date: Sat Sep 18 01:23:12 2010 +0200
30398
30399 XineramaInitData ignores its argument. Quit passing one.
30400
30401 Also fix up XineramaInitData's caller, XineramaReinitData.
30402
30403 Commit by Jamey Sharp and Josh Triplett.
30404
30405 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30406 Signed-off-by: Josh Triplett <josh@joshtriplett.org>
30407 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30408
30409commit b85c10a35d44ad695797d0d19757e5f62eab4b98
30410Author: Jamey Sharp <jamey@minilop.net>
30411Date: Tue May 25 16:23:31 2010 -0700
30412
30413 dmx: Fix some "no previous prototype" warnings by making functions static.
30414
30415 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30416 Reviewed-by: Matt Turner <mattst88@gmail.com>
30417
30418commit d423012e776a81a197cde86be5c6bd471a593c39
30419Author: Jamey Sharp <jamey@minilop.net>
30420Date: Fri May 21 14:21:52 2010 -0700
30421
30422 dmx: Fix a non-constant printf format string warning.
30423
30424 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30425 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30426
30427commit 7e127f12a24acb8f43055af0a9be1ec585bbed36
30428Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30429Date: Thu May 19 23:22:27 2011 -0700
30430
30431 sun_init.c: Move vt switches to a switch_to helper function like lnx_init.c
30432
30433 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30434 Reviewed-by: Julien Cristau <jcristau@debian.org>
30435
30436commit a00dc6af6e84d0bee3f69d8df89d7f68cb535d35
30437Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30438Date: Thu May 19 22:43:56 2011 -0700
30439
30440 Xorg.man: Add -novtswitch and -sharevts options
30441
30442 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30443 Reviewed-by: Julien Cristau <jcristau@debian.org>
30444
30445commit 93abda6bddf770dbfc3d4d0ada61ea84a7b7b9be
30446Author: Alan Coopersmith <alan.coopersmith@oracle.com>
30447Date: Thu May 19 22:39:09 2011 -0700
30448
30449 xfree86: move -novtswitch & -sharevts argument handling up to common layer
30450
30451 Stop duplicating in each os-support variant before it gets replicated
30452 even further.
30453
30454 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30455 Reviewed-by: Julien Cristau <jcristau@debian.org>
30456 Reviewed-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
30457
30458commit d8d99e14b1241fd8c099d69b8d0ed0817585fe8d
30459Author: Jamey Sharp <jamey@minilop.net>
30460Date: Wed Sep 14 09:52:58 2011 -0500
30461
30462 configure.ac: Remove unreachable check for VM86 headers.
30463
30464 "configure --with-int10" is not a valid configuration, and the check for
30465 sys/vm86.h and sys/io.h is not used. Delete it.
30466
30467 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30468 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30469 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
30470
30471commit 94b2eea86319934316bdb013f6f3940685a88a2f
30472Author: Jamey Sharp <jamey@minilop.net>
30473Date: Wed Sep 14 11:23:03 2011 -0500
30474
30475 int10: Delete #if 0'd implementation that's older than our git history.
30476
30477 Throughout the xserver git history, the generic portion of the int10
30478 module has always used other methods for reading the video BIOS. For
30479 some time now it's been purely libpciaccess based. This commented-out
30480 use of xf86ReadBIOS is entirely superfluous.
30481
30482 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30483 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30484
30485commit 6ad7bb69eebd5e5b4068bd9aa0b9a7168f075006
30486Author: Jamey Sharp <jamey@minilop.net>
30487Date: Tue Sep 13 17:45:19 2011 -0500
30488
30489 Remove PC98 support.
30490
30491 Gaetan Nadon wrote:
30492
30493 Alan Coopersmith wrote:
30494
30495 "I think we recently dropped PC98 support from the X server, so I'd
30496 be okay with dropping the documentation now".
30497
30498 Let's make them be right, shall we?
30499
30500 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30501 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30502 Acked-by: Gaetan Nadon <memsize@videotron.ca>
30503 Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30504
30505commit 9eb22ebf46b77c15e53017a37c7ef605521dc164
30506Author: Jamey Sharp <jamey@minilop.net>
30507Date: Tue Sep 13 16:30:06 2011 -0500
30508
30509 configure.ac: XORG_OS is not used, so delete it.
30510
30511 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30512 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30513 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
30514
30515commit 76636f30f7cf2344f4038c59bf292db19be33d53
30516Author: Jamey Sharp <jamey@minilop.net>
30517Date: Tue Sep 13 13:10:10 2011 -0500
30518
30519 configure.ac: Remove unused XORG_OS_PCI variable.
30520
30521 This is the last mention after ccfaf82367c9d057fd8314ce36b47f0a8eb696b6
30522 quit using the variable.
30523
30524 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30525 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30526 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
30527
30528 Tiago Vignatti posted an identical patch in June 2010, which I only
30529 noticed after getting the above reviews. His patch was:
30530
30531 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
30532 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
30533
30534commit 60b4bd181bbdc794c7f3547f67df916132aa111e
30535Author: Jamey Sharp <jamey@minilop.net>
30536Date: Tue Sep 13 14:31:24 2011 -0500
30537
30538 x86emu: There is no NO_INLINE implementation of unaligned access here.
30539
30540 Patch produced with:
30541 unifdef -UNO_INLINE -B
30542
30543 This change isn't relevant to the similar code in
30544 hw/xfree86/common/compiler.h, because x86emu is expected to someday move
30545 out of xserver entirely and so should not depend on any xserver headers.
30546 Also, some platforms apparently do have NO_INLINE versions of
30547 compiler.h.
30548
30549 Signed-off-by: Jamey Sharp <jamey@minilop.net>
30550 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
30551 Reviewed-by: Matt Turner <mattst88@gmail.com>
30552
30553commit 24b2b43e85c1e3f3d242ccab8793bf4ef83cfac4
30554Author: Cyril Brulebois <kibi@debian.org>
30555Date: Sun Jun 5 14:53:08 2011 +0200
30556
30557 render: Replace __inline with inline.
30558
30559 Also remove traces from the past.
30560
30561 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30562 Signed-off-by: Cyril Brulebois <kibi@debian.org>
30563
30564commit 9f01249d7f81514a38c44dff2a4a4612dbf97cae
30565Author: Cyril Brulebois <kibi@debian.org>
30566Date: Sun Jun 5 15:10:56 2011 +0200
30567
30568 configure.ac: Fix help string for mitshm.
30569
30570 Pasting from ./configure --help's output, one would get:
30571 | configure: WARNING: unrecognized options: --disable-shm
30572
30573 Fix the help string to include the previously missing “mit” bits.
30574
30575 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
30576 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30577 Signed-off-by: Cyril Brulebois <kibi@debian.org>
30578
30579commit 0caeef6146bee5fb1827ab25db191685dde9d4b4
30580Author: Keith Packard <keithp@keithp.com>
30581Date: Fri Aug 26 16:46:13 2011 -0700
30582
30583 Version bumped to 1.11
30584
30585 Signed-off-by: Keith Packard <keithp@keithp.com>
30586
30587commit 05284a03f9002b03a66ae355b34790ec02b726f0
30588Author: Peter Hutterer <peter.hutterer@who-t.net>
30589Date: Thu Aug 4 14:45:46 2011 +1000
30590
30591 input: make InputOption opaque, provide interface functions.
30592
30593 InputOptions is not switched to use struct list for a future patch to unify
30594 it with the XF86OptionRec.
30595
30596 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30597 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
30598
30599commit fcafe825751bef99f4c0b36250ca6f15f127502f
30600Author: Peter Hutterer <peter.hutterer@who-t.net>
30601Date: Mon Aug 8 15:21:46 2011 +1000
30602
30603 Add null-terminated list interface.
30604
30605 This is a set of macros to provide a struct list-alike interface for classic
30606 linked lists such as the XF86OptionRec or the DeviceIntRec. The typical
30607 format for these is to have a "struct foo *next" pointer in each struct foo
30608 and walk through those. These macros provide a few basic functions to add to,
30609 remove from and iterate through these lists.
30610
30611 While struct list is in some ways more flexible, switching legacy code to
30612 use struct list is not alway viable. These macros at least reduce the amount
30613 of open-coded lists.
30614
30615 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30616 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30617
30618commit 79ca7c0b5786a02a80a1c40ed475e928da7c82b3
30619Author: Peter Hutterer <peter.hutterer@who-t.net>
30620Date: Fri Aug 12 15:55:48 2011 +1000
30621
30622 xfree86: comment typo fix
30623
30624 in synch → in sync
30625
30626 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30627
30628commit 2bfb802839688ecf328119c4c6979390fc60348d
30629Author: Peter Hutterer <peter.hutterer@who-t.net>
30630Date: Fri Jul 29 16:33:54 2011 +1000
30631
30632 dix: don't XWarpPointer through the last slave anymore (#38313)
30633
30634 This line was introduced pre-1.6 to fix Bug 19297. The effect of warping
30635 through the VCP then was that if a device had custom valuator ranges, the
30636 warp position would be wrong. The better device for this effect is the the
30637 XTest device.
30638
30639 This fixes a server crash where the lastSlave is a pointer device without
30640 valuators (Bug 38313#0).
30641
30642 And while we're at it, make sure the Xinerama code-path does the same.
30643
30644 X.Org Bug 38313 <http://bugs.freedesktop.org/show_bug.cgi?id=38313>
30645
30646 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30647 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30648
30649commit b29ce0726d55ec41cfbce0814e21b0217ef64efe
30650Author: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
30651Date: Tue Aug 9 11:00:34 2011 +0900
30652
30653 Disable check of double-aligned in test/input.c on Renesas SH
30654
30655 Renesas SH is not aligned at size of double.
30656 When structure has double value, It is aligned in 4byte (long).
30657
30658 Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
30659 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
30660 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30661
30662commit 3be379f5076566edaf92c27df5a4d447bcf5d015
30663Author: Peter Hutterer <peter.hutterer@who-t.net>
30664Date: Thu Aug 11 10:59:07 2011 +1000
30665
30666 dix: use helper functions in EventIsDeliverable
30667
30668 Proximity events don't have an XI2 type and caused error messages in the
30669 log when trying to get the event filter. Use this opportunity to
30670 clean up the code, instead of manually setting the fields that
30671 GetEventFilter requires use EventTo(XI2|XI|Core) instead.
30672
30673 Co-Authored-by: Keith Packard <keithp@keithp.com>
30674 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30675 Reviewed-by: Keith Packard <keithp@keithp.com>
30676
30677commit 799879797505a5e891ccaec2bea73fd838c94b7a
30678Author: Peter Hutterer <peter.hutterer@who-t.net>
30679Date: Thu Aug 11 10:57:05 2011 +1000
30680
30681 dix: abstract XI2 filter mask lookup
30682
30683 Don't access the xi2mask bytes directly or calculate the offsets manually,
30684 use a few helper functions instead. XI2 masks are a bit weird in the event
30685 handling code since they slot onto the legacy code. For core/XI 1.x events,
30686 the event mask is a CARD32. That mask is used together with the event filter
30687 (also 32 bit) to determine if event delivery should be attempted.
30688 XI2 masks are of arbitrary size and their mask is simply the byte of the
30689 mask that contains the event mask. Likewise, the filter is a single byte
30690 matching that mask. Provide helper functions get these bytes and masks in
30691 the right order instead of accessing them manually.
30692
30693 EventIsDeliverable should be part of this cleanup patch but it will be
30694 gutted with the next patch.
30695
30696 Co-Authored-by: Keith Packard <keithp@keithp.com>
30697 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30698 Reviewed-by: Keith Packard <keithp@keithp.com>
30699
30700commit 8c5a4d6fbecf79f2dc4f2d836d741203b2d5e856
30701Author: Peter Hutterer <peter.hutterer@who-t.net>
30702Date: Mon Aug 1 14:12:41 2011 +1000
30703
30704 dix: don't use the pointer as modifier device in UngrabKey.
30705
30706 Modifier device is always the keyboard.
30707
30708 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30709 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30710
30711commit 3a077f246e9ac07a37c1b01c3d321e0f5ceb4153
30712Author: Peter Hutterer <peter.hutterer@who-t.net>
30713Date: Wed Aug 3 15:07:23 2011 +1000
30714
30715 input: provide a single function to init DeviceEvents to 0
30716
30717 getevents.c already had that function, but XKB was manually initializing it,
30718 causing bugs when the event structure was updated in one place but not the
30719 other.
30720
30721 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30722
30723commit 4b376ddeb4f3c9d9d279ffd2946d88edd5af4cfc
30724Author: Peter Hutterer <peter.hutterer@who-t.net>
30725Date: Mon Aug 1 14:20:53 2011 +1000
30726
30727 dix: fix compiler warnings ("foo" set but not used)
30728
30729 devices.c: In function 'AttachDevice':
30730 devices.c:2409:18: warning: variable 'oldmaster' set but not used
30731 [-Wunused-but-set-variable]
30732
30733 events.c: In function 'ConfineToShape':
30734 events.c:683:15: warning: variable 'pSprite' set but not used
30735 [-Wunused-but-set-variable]
30736
30737 events.c: In function 'ProcGrabPointer':
30738 events.c:4759:15: warning: variable 'time' set but not used
30739 [-Wunused-but-set-variable]
30740
30741 getevents.c: In function 'GetMotionHistory':
30742 getevents.c:425:9: warning: variable 'dflt' set but not used
30743 [-Wunused-but-set-variable]
30744
30745 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30746 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30747
30748commit 6fd2adc179141310e45a56ee90ef5b5f6115a1f6
30749Author: Peter Hutterer <peter.hutterer@who-t.net>
30750Date: Mon Aug 1 14:27:54 2011 +1000
30751
30752 mi: fix compiler warnings ("foo" set but not used)
30753
30754 misprite.c: In function 'miSpriteSaveUnderCursor':
30755 misprite.c:940:12: warning: variable 'y' set but not used
30756 [-Wunused-but-set-variable]
30757 misprite.c:940:9: warning: variable 'x' set but not used
30758 [-Wunused-but-set-variable]
30759
30760 mivaltree.c: In function 'miComputeClips':
30761 mivaltree.c:226:10: warning: variable 'resized' set but not used
30762 [-Wunused-but-set-variable]
30763
30764 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30765 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30766
30767commit 484cef5b29ef82402a15e155b3b8505b1e4a6830
30768Author: Peter Hutterer <peter.hutterer@who-t.net>
30769Date: Tue Aug 2 09:21:00 2011 +1000
30770
30771 Xi: silence compiler warnings (set but not used)
30772
30773 exevents.c: In function 'UpdateDeviceState':
30774 exevents.c:719:9: warning: variable 'bit' set but not used
30775 [-Wunused-but-set-variable]
30776
30777 exevents.c: In function 'ProcessOtherEvent':
30778 exevents.c:889:22: warning: variable 'v' set but not used
30779 [-Wunused-but-set-variable]
30780 exevents.c:888:17: warning: variable 'k' set but not used
30781 [-Wunused-but-set-variable]
30782
30783 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30784 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30785
30786commit 98fe735ea1d756711019c3d90ed6abd9c06abebf
30787Author: Peter Hutterer <peter.hutterer@who-t.net>
30788Date: Mon Aug 1 13:52:13 2011 +1000
30789
30790 dix: add KEYBOARD_OR_FLOAT and POINTER_OR_FLOAT to GetMaster()
30791
30792 GetMaster() currently requires an attached slave device as parameter,
30793 resuling in many calls being IsFloating(dev) ? dev : GetMaster(...);
30794
30795 Add two new parameters so GetMaster can be called unconditionally to get the
30796 right device.
30797
30798 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30799 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30800
30801commit dbbe5735d1451bb32f43bce90f0bcfeff46f9743
30802Author: Peter Hutterer <peter.hutterer@who-t.net>
30803Date: Tue Aug 2 08:54:00 2011 +1000
30804
30805 test: add a test for GetMaster() behaviour
30806
30807 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30808 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30809
30810commit b3c76b0c53ac42b70d12849da18465e8467e474c
30811Author: Peter Hutterer <peter.hutterer@who-t.net>
30812Date: Mon Aug 1 11:54:17 2011 +1000
30813
30814 mi: fix comment typo, whitespace in miPointerSetPosition
30815
30816 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30817 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30818
30819commit c9562bed0d5e26b7e3e55e26cf1ddc5086d61cc6
30820Author: Peter Hutterer <peter.hutterer@who-t.net>
30821Date: Mon Aug 1 09:52:38 2011 +1000
30822
30823 dix: rename mieqSwitchScreen argument fromDix → set_dequeue_screen, document
30824
30825 fromDIX is neither exactly true nor particularly helpful in understanding
30826 what this parameter triggers. Rename to set_dequeue_screen, because that's
30827 exactly what happens.
30828
30829 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30830 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30831
30832commit 09496996accfdaf7bc01097a25db400912004d97
30833Author: Peter Hutterer <peter.hutterer@who-t.net>
30834Date: Thu Jul 28 15:56:08 2011 +1000
30835
30836 dix: ignore devices when adding passive core grabs to list (#39545)
30837
30838 Passive core grabs are mostly device-independent. In an MPX scenario, they
30839 may change to reflect whichever master pair activated the grab last. For
30840 adding new grabs to the list, ignore the device for core grabs to return
30841 failures when trying to set the same grab combo twice on a window.
30842
30843 X.Org Bug 39545 <http://bugs.freedesktop.org/show_bug.cgi?id=39545>
30844
30845 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30846 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30847
30848commit 80c37048539daa1d257d127d66502bde45c97c85
30849Author: Peter Hutterer <peter.hutterer@who-t.net>
30850Date: Thu Jul 28 15:43:10 2011 +1000
30851
30852 dix: avoid using the VCP as modifier device
30853
30854 Core grabs may change device when they're activated to reflect the master
30855 they apply to. If the device is a keyboard, modifierDevice is erroneously
30856 set to the Virtual Core Pointer.
30857
30858 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30859 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30860
30861commit 20a61845d3c93c337bf3331a6bac30cf66c2a293
30862Author: Peter Hutterer <peter.hutterer@who-t.net>
30863Date: Tue Aug 9 11:21:06 2011 +1000
30864
30865 config: use add_option for '_source' too
30866
30867 _source was being allocated manually, with all other options added to that
30868 list through add_option. Skip the manual part, allocate the first option
30869 _source with add_option too.
30870
30871 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30872 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30873
30874commit d33652dad8838ab0a9175ca4613a3161ebc5676f
30875Author: Peter Hutterer <peter.hutterer@who-t.net>
30876Date: Tue Aug 9 11:20:31 2011 +1000
30877
30878 config: return the new InputOption from add_option.
30879
30880 Change add_option to return the new InputOption on success, or NULL
30881 failure. This way we can at least check for errors in callers.
30882
30883 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30884 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30885
30886commit e684e816acb617b4dc66a68e2b0ba8f80399170a
30887Author: Peter Hutterer <peter.hutterer@who-t.net>
30888Date: Wed Jul 20 16:21:28 2011 +1000
30889
30890 config: fix a log message
30891
30892 PRODUCT was taken from the parent, hence ppath.
30893
30894 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30895 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30896
30897commit 1357cd725143c1a35e32f15df658de111b151692
30898Author: Peter Hutterer <peter.hutterer@who-t.net>
30899Date: Fri Jul 8 16:13:45 2011 +1000
30900
30901 Revert "Attempt to add the 'mouse' driver in more situations."
30902
30903 This reverts commit 43d9edd31e31b33b9da4a50d8ab05004881c8d5a.
30904
30905 This commit was introduced in the 1.2 cycle when hotplugging was less than
30906 ideal (i.e. it didn't exist). From the commit message:
30907
30908 Always add a mouse driver instance configured to send core events, unless
30909 a core pointer already exists using either the mouse or void drivers. This
30910 handles the laptop case where the config file only specifies, say,
30911 synaptics, which causes the touchpad to work but not the pointing stick.
30912 We don't double-instantiate the mouse driver to avoid the mouse moving twice
30913 as fast, and we skip this logic when the user asked for a void core pointer
30914 since that probably means they want to run with no pointer at all.
30915
30916 To get this case above, a user would need to disable hotplugging _and_ have a
30917 xorg.conf that only references one device. This is possible, but not a use-case
30918 we should worry about too much now.
30919
30920 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30921 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30922
30923commit 159b03e13760920274b573a2bccdbf6a79f059e7
30924Author: Lennart Poettering <lennart@poettering.net>
30925Date: Mon Jul 18 21:19:23 2011 +0200
30926
30927 config: add udev/systemd multi-seat support
30928
30929 Add support for multi-seat-aware input device hotplugging. This
30930 implements the multi-seat scheme explained here:
30931
30932 http://www.freedesktop.org/wiki/Software/systemd/multiseat
30933
30934 This introduces a new X server switch "-seat" which allows configuration
30935 of the seat to enumerate hotplugging devices on. If specified the value
30936 of this parameter will also be exported as root window property
30937 Xorg_Seat.
30938
30939 To properly support input hotplugging devices need to be tagged in udev
30940 according to the seat they are on. Untagged devices are assumed to be on
30941 the default seat "seat0". If no "-seat" parameter is passed only devices
30942 on "seat0" are used. This means that the new scheme is perfectly
30943 compatible with existing setups which have no tagged input devices.
30944
30945 Note that the -seat switch takes a completely generic identifier, and
30946 that it has no effect on non-Linux systems. In fact, on other OSes a
30947 completely different identifier scheme for seats could be used but still
30948 be exposed with the Xorg_Seat and -seat.
30949
30950 I tried to follow the coding style of the surrounding code blocks if
30951 there was any one could follow.
30952
30953 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30954
30955commit 95772598b57f6054fbf88683fa0a492c77605790
30956Author: Peter Hutterer <peter.hutterer@who-t.net>
30957Date: Fri Jul 8 16:10:07 2011 +1000
30958
30959 xfree86: use xf86AllocateInput for implicit devices too
30960
30961 Slowly merging the vastly different code-paths.
30962
30963 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30964 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30965
30966commit fa8f4652819b692faaf2789cf32d7fa99fbb34aa
30967Author: Peter Hutterer <peter.hutterer@who-t.net>
30968Date: Mon Jul 4 15:02:20 2011 +1000
30969
30970 xfree86: factor out adding/removing a device from the input device array
30971
30972 No functional changes, just readability improvements. This also gets rid of
30973 the count variable. Count was just used for resizing the null-terminated
30974 list. Since we're not in a time-critical path here at all we can afford to
30975 loop the list multiple times instead of keeping an extra variable around.
30976
30977 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30978 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30979
30980commit 5b5477c05f691205064ca4d8034f8dd47ab975b7
30981Author: Peter Hutterer <peter.hutterer@who-t.net>
30982Date: Mon Jul 4 14:14:11 2011 +1000
30983
30984 xfree86: update comment for InitInput
30985
30986 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
30987 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
30988
30989commit 7354f607833c69626d8692bc5176b18ea1cf6263
30990Author: Peter Hutterer <peter.hutterer@who-t.net>
30991Date: Mon Jul 4 14:26:13 2011 +1000
30992
30993 xfree86: nest loops instead of 0x1 pointers.
30994
30995 If we find the core device, move all other device pointers forward right
30996 then and there. The break will jump out of the top loop.
30997
30998 They had a special on braces today, so I added some for readability (and
30999 fixed up tab vs space indentation.
31000
31001 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31002 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31003
31004commit 5669aa2d24dff9ab276e5f74a09f97ec77b90e75
31005Author: Peter Hutterer <peter.hutterer@who-t.net>
31006Date: Mon Jul 4 14:05:30 2011 +1000
31007
31008 xfree86: improve readability of synthesized device.
31009
31010 No functional changes.
31011
31012 The options we assign are the ones from the Pointer/Keyboard device so we
31013 might as well use those readable names instead of dev[count-1]->options.
31014
31015 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31016 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31017
31018commit 5aa826cdd1f2e768bedf23d399703a5d0b6302be
31019Author: Peter Hutterer <peter.hutterer@who-t.net>
31020Date: Mon Jul 4 13:52:11 2011 +1000
31021
31022 test: add a option duplication test
31023
31024 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31025 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31026
31027commit 4527e2b776cfcdac2b189b5439b9a3d0b6433077
31028Author: Peter Hutterer <peter.hutterer@who-t.net>
31029Date: Mon Jul 4 13:46:24 2011 +1000
31030
31031 xfree86: when implicitly choosing a core device, set the option to a value
31032
31033 Devices are core pointers/keyboards by default now anyway, but let's set the
31034 option to some value instead of just NULL.
31035
31036 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31037 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31038
31039commit 44d53728a6a533fc0a6e0a10269d1cc99e9dad32
31040Author: Peter Hutterer <peter.hutterer@who-t.net>
31041Date: Mon Jul 4 13:44:44 2011 +1000
31042
31043 xfree86: don't warn about duplicate core devices
31044
31045 It doesn't matter. All devices are core pointer devices by default now
31046 anyway.
31047
31048 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31049 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31050
31051commit 033f53c223dc12a91f00e10a69f87a4f2a7adb6b
31052Author: Daniel Kurtz <djkurtz@chromium.org>
31053Date: Mon Aug 8 15:09:48 2011 +0800
31054
31055 xf86Helper: use LogHdrMessageVerb in xf86VDrvMsgVerb
31056
31057 LogHdrMessageVerb allows passing a parameterized header to insert in a log
31058 message between MessageType and the formatted message body string.
31059
31060 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
31061 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31062 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31063
31064commit cd8ee3e5cb29b9cd6402d2fbc71463c6b04b6077
31065Author: Daniel Kurtz <djkurtz@chromium.org>
31066Date: Mon Aug 8 15:09:47 2011 +0800
31067
31068 xf86Helper: use LogHdrMessageVerb in xf86VIDrvMsgVerb
31069
31070 LogHdrMessageVerb allows passing a parameterized header to insert in a log
31071 message between MessageType and the formatted message body string.
31072
31073 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
31074 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31075 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31076
31077commit b31d104fc09a7241856ff8d226be11ec562beac3
31078Author: Daniel Kurtz <djkurtz@chromium.org>
31079Date: Mon Aug 8 15:09:46 2011 +0800
31080
31081 os/log: Add LogVHdrMessageVerb and friends
31082
31083 LogVHdrMessageVerb allows a custom header to be inserted in a log message,
31084 between the Log system's MessageType string, and a formatted variable
31085 message body. The custom header can itself be a formatted variable string.
31086
31087 These functions can be used, for example, by driver abstraction layers to
31088 format specific driver messages in a standard format, but do it in a way
31089 that is efficient, obeys the log-layers verbosity settings, and is safe
31090 to use in signal handlers (because they don't call malloc), even for
31091 types besides X_NONE.
31092
31093 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
31094 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31095 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31096
31097commit b82f934db661c59d7397ce180d08cf8f8f7118b1
31098Author: Daniel Kurtz <djkurtz@chromium.org>
31099Date: Mon Aug 8 15:09:45 2011 +0800
31100
31101 os/log: Pull LogMessageTypeVerbString out of LogVMessageVerb
31102
31103 Also, optimize how the type and format strings are combined.
31104
31105 Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
31106 Reviewed-by: Guillem Jover <guillem@hadrons.org>
31107 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31108
31109commit e32cc0b4c85c78cd8743a6e1680dcc79054b57ce
31110Author: Adam Jackson <ajax@redhat.com>
31111Date: Thu Apr 21 16:37:11 2011 -0400
31112
31113 fb: Fix memcpy abuse
31114
31115 The memcpy fast path implicitly assumes that the copy walks
31116 left-to-right. That's not something memcpy guarantees, and newer glibc
31117 on some processors will indeed break that assumption. Since we walk a
31118 line at a time, check the source and destination against the width of
31119 the blit to determine whether we can be sloppy enough to allow memcpy.
31120 (Having done this, we can remove the check for !reverse as well.)
31121
31122 On an Intel Core i7-2630QM with an NVIDIA GeForce GTX 460M running in
31123 NoAccel, the broken code and various fixes for -copywinwin{10,100,500}
31124 gives (edited to fit in 80 columns):
31125
31126 1: Disable the fastpath entirely
31127 2: Replace memcpy with memmove
31128 3: This fix
31129 4: The code before this fix
31130
31131 1 2 3 4 Operation
31132 ------ --------------- --------------- --------------- ------------
31133 258000 269000 ( 1.04) 544000 ( 2.11) 552000 ( 2.14) Copy 10x10
31134 21300 23000 ( 1.08) 43700 ( 2.05) 47100 ( 2.21) Copy 100x100
31135 960 962 ( 1.00) 1990 ( 2.09) 1990 ( 2.07) Copy 500x500
31136
31137 So it's a modest performance hit, but correctness demands it, and it's
31138 probably worth keeping the 2x speedup from having the fast path in the
31139 first place.
31140
31141 Signed-off-by: Adam Jackson <ajax@redhat.com>
31142 Signed-off-by: Keith Packard <keithp@keithp.com>
31143
31144commit ac2c307f4716ebd3e955c004ceec9f4c029401a0
31145Author: Pelle Johansson <pelle@morth.org>
31146Date: Sun Aug 14 17:44:40 2011 -0700
31147
31148 XQuartz: Initialize darwin pointer valuators
31149
31150 This fixes a regression introduced by: 633b81e8ba09cc6a1ea8b43f323874fda2cf0bde
31151
31152 http://xquartz.macosforge.org/trac/ticket/498
31153
31154 Signed-off-by: Pelle Johansson <pelle@morth.org>
31155 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
31156
31157commit 4020cab88f5cf3164fc83cf912f94f288aa5a45d
31158Author: Michel Dänzer <michel.daenzer@amd.com>
31159Date: Wed Aug 10 11:36:16 2011 +0200
31160
31161 EXA/mixed: Update sys_pitch in MPH even when there's no system memory copy.
31162
31163 Otherwise sys_pitch will be stale when a system memory copy is allocated.
31164
31165 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=38322 and a crash when
31166 unlocking the screen with xscreensaver, reported by Janne Huttunen.
31167
31168 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
31169 Tested-by: Janne Huttunen <jahuttun@gmail.com>
31170 Tested-by: Jan Kriho <Erbureth@gmail.com>
31171 Signed-off-by: Keith Packard <keithp@keithp.com>
31172
31173commit a1dec7cd6adabaf8770dc3b21b0df30ce2f0bc8d
31174Author: Julien Cristau <jcristau@debian.org>
31175Date: Thu Jul 7 19:18:03 2011 +0200
31176
31177 Xephyr/dri: register screen and window privates on init
31178
31179 Fixes assertion failure when calling dixSetPrivate
31180 Debian bug#632549 <http://bugs.debian.org/632549>
31181
31182 Reported-and-tested-by: Mohammed Sameer <msameer@foolab.org>
31183 Signed-off-by: Julien Cristau <jcristau@debian.org>
31184 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31185 Signed-off-by: Keith Packard <keithp@keithp.com>
31186
31187commit 39bc81b60ddc9e495ff3b9bb3c3a6c39e547b43a
31188Author: Keith Packard <keithp@keithp.com>
31189Date: Wed Aug 3 20:57:03 2011 -0700
31190
31191 Version bumped to 1.10.99.902 (1.11 RC2)
31192
31193 At the close of the 1.11 non-critical fixes window.
31194
31195 Signed-off-by: Keith Packard <keithp@keithp.com>
31196
31197commit 9504caf1c3243e3ab2eb7126bc2bb876a8f89918
31198Author: Ville Syrjälä <ville.syrjala@nokia.com>
31199Date: Fri May 6 18:19:34 2011 +0300
31200
31201 composite: Inhibit window background paint with manual subwindow redirection
31202
31203 The composite extension spec says that window background painting
31204 should be inhibited when the subwindow redirection mode is set to
31205 manual.
31206
31207 This eliminates the ugly flashing effect when compiz unredirects a
31208 fullscreen window.
31209
31210 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
31211 Reviewed-by: Owen Taylor <otaylor@fishsoup.net>
31212 Reviewed-by: Keith Packard <keithp@keithp.com>
31213 Signed-off-by: Keith Packard <keithp@keithp.com>
31214
31215commit f1d75f3b742231c4c09f13d7a22ed920014cb5ef
31216Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
31217Date: Thu Jul 28 15:17:59 2011 -0700
31218
31219 Revert "composite: Don't backfill non-MapWindow allocations"
31220
31221 This reverts commit db8840600e8e21356241eb87395031388d9b54d2.
31222
31223 It was an optimization for the resize case, but 193ecc8b453b22 made
31224 it so that no backfilling takes place on resize if left in.
31225
31226 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
31227 Reviewed-by: Adam Jackson <ajax@redhat.com>
31228
31229 Conflicts:
31230
31231 composite/compalloc.c
31232 Signed-off-by: Keith Packard <keithp@keithp.com>
31233
31234commit e87adcc9e024982fdad974de2876b00cf974dab8
31235Author: Julien Cristau <jcristau@debian.org>
31236Date: Sun Jul 31 21:14:59 2011 +0200
31237
31238 Xquartz: include new localization files in the tarball
31239
31240 Signed-off-by: Julien Cristau <jcristau@debian.org>
31241 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
31242
31243commit a1bb5062c72667b2f10b56925c61888acc89e3e4
31244Author: Jeremy Huddleston <jeremyhu@apple.com>
31245Date: Thu Jul 21 10:33:58 2011 -0700
31246
31247 XQuartz: xpr: Don't FatalError if xp_unlock_window fails
31248
31249 We added the FatalError in 5d1d9d9ae39fab2ee2ac085f9776f82768828dc8 but
31250 it caused a regression http://xquartz.macosforge.org/trac/ticket/482
31251
31252 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31253
31254commit c319f7b5b3ee651636cdfd165588f0dc972a22a4
31255Author: Jeremy Huddleston <jeremyhu@apple.com>
31256Date: Wed Jul 20 00:16:42 2011 -0700
31257
31258 XQuartz: Use CFSTR to avoid implicit cast warning of NSString * to CFStringRef
31259
31260 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31261
31262commit 08dfff92e8d0c2b1726634392c147f6634d1706d
31263Author: Aaron Plattner <aplattner@nvidia.com>
31264Date: Fri Jul 29 16:12:06 2011 -0700
31265
31266 randr: Compare all the bytes in RRPostPendingProperties
31267
31268 RRPostPendingProperties tries to compare the pending and current
31269 property values to decide whether they're actually changing. However,
31270 it does this using a memcmp that passes in pending_value->size as the
31271 number of bytes. This is actually the number of elements, where each
31272 element is (pending_value->format / 8) bytes long. This causes the
31273 pending value to not be propagated if the first pending_value->size
31274 bytes are the same and only the end of it is changing.
31275
31276 Fix this by computing the total number of bytes to compare in the
31277 memcmp.
31278
31279 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
31280 Reviewed-by: Keith Packard <keithp@keithp.com>
31281 Signed-off-by: Keith Packard <keithp@keithp.com>
31282
31283commit b8f61c11c94708d4f17720a55945dee4621315a4
31284Merge: 5596f10 f51e42f
31285Author: Keith Packard <keithp@keithp.com>
31286Date: Fri Jul 29 14:58:58 2011 -0700
31287
31288 Merge remote-tracking branch 'whot/for-keith'
31289
31290commit f51e42f583073bde0bc8131887cb7220636c8855
31291Author: Peter Hutterer <peter.hutterer@who-t.net>
31292Date: Wed Jul 20 13:09:05 2011 +1000
31293
31294 Terminate the log with one last message.
31295
31296 Instead of just closing the log when everything is done, put one more
31297 message in stating that we're actually terminating. Users or scripts that
31298 look at the Xorg.log will then know that a) the server has terminated
31299 properly and b) why the server terminated (to some degree, given that most
31300 real-world errors will be caused by AbortServer()).
31301
31302 Acked-by: Gaetan Nadon <memsize@videotron.ca>
31303 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31304 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
31305 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
31306 Tested-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31307 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31308 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31309
31310commit 01de08c7d2c00eef238adba6665896ea3cd7d511
31311Author: Julien Cristau <jcristau@debian.org>
31312Date: Tue Jul 26 20:40:38 2011 +0200
31313
31314 configure: set default xkb rules to evdev on Linux
31315
31316 If config/udev was enabled, this would default to base, which means that
31317 after regen the devices would get the wrong rules, and hilarity would
31318 ensue.
31319
31320 It's probably safe to default to evdev unconditionally on Linux by now.
31321
31322 Reported-by: Bastian Blank <waldi@debian.org>
31323 Signed-off-by: Julien Cristau <jcristau@debian.org>
31324 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31325 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31326
31327commit 3798dd379c1ecf325f9907128fb66d20372f6876
31328Author: Peter Hutterer <peter.hutterer@who-t.net>
31329Date: Wed Jul 20 09:00:18 2011 +1000
31330
31331 Initialize the fd to -1 for xorg.conf input devices.
31332
31333 For hotplugged devices, xf86AllocateInput does that for us but the xorg.conf
31334 path is different. Since not all drivers reset the fd during PreInit but may
31335 still call close(pInfo->fd) in all cases, this can terminate the logging
31336 early.
31337
31338 Reproducible: add a wacom driver InputDevice section with no Option Device.
31339
31340 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31341 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31342
31343commit f2a6735cfc07789cca81852b24a85578f200d83d
31344Author: Peter Hutterer <peter.hutterer@who-t.net>
31345Date: Mon Jul 4 12:34:32 2011 +1000
31346
31347 xfree86: NULL option values are technically valid, don't strdup them
31348
31349 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31350 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31351
31352commit f0d7e9db28c374a3db359bcb63a7ce79fd84b541
31353Author: Peter Hutterer <peter.hutterer@who-t.net>
31354Date: Mon Jul 4 14:14:39 2011 +1000
31355
31356 xfree86: duplicate xorg.conf device information before xf86NewInputDevice
31357
31358 xf86ConfigLayout.inputs contains the information from the xorg.conf
31359 file. Passing this into xf86NewInputDevice means the device will get
31360 cleaned up on exit and the pointers in xf86ConfigLayout.inputs are left
31361 dangling. In the second server generation, this results in a server
31362 crash.
31363
31364 Also, rename pDev to pInfo. pDev is pretty much reserved for DeviceIntPtr
31365 types.
31366
31367 Reproducible: AutoAddDevices off and xorg.conf input sections, trigger
31368 server regeneration.
31369
31370 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31371 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31372
31373commit 8ffddbcf72170e246826ee0f39f18989a29fa218
31374Author: Peter Hutterer <peter.hutterer@who-t.net>
31375Date: Mon Jun 27 16:30:28 2011 +1000
31376
31377 xfree86: Remove devices that failed to enable on startup
31378
31379 Devices that succeeded during PreInit and DEVICE_INIT but failed in
31380 DEVICE_ON would be deleted through xf86DeleteInput but not removed from the
31381 list of input devices (and not turned off). The result was a double free on
31382 server shutdown.
31383
31384 Fix this by calling RemoveDevice if EnableDevice fails.
31385
31386 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31387 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31388
31389commit 6cea28fe4b7a4a22ad270d8c71403db84a9bfb2c
31390Author: Lennart Poettering <lennart@poettering.net>
31391Date: Mon Jul 18 21:18:27 2011 +0200
31392
31393 config: don't fail if a device vanished by the time we managed to look at it
31394
31395 The nature of hotplug is that a device we enumerated might already be
31396 gone by the time we look at it, so don't assume otherwise.
31397
31398 Signed-off-by: Lennart Poettering <lennart@poettering.net>
31399 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31400 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31401
31402commit f13de9ca1b7c4dd0dd5c08037c6bd53f88ac30f5
31403Author: Lennart Poettering <lennart@poettering.net>
31404Date: Mon Jul 18 21:17:49 2011 +0200
31405
31406 config: limit the kernel subsystems we look for devices in
31407
31408 Don't enumerate/monitor all devices of the system (since that can be
31409 quite a few), but limit our search to devices from the "input"
31410 subsystem, as well as the "tty" subsystem (to cover Wacom tablets).
31411
31412 This should make X start up a bit faster and reduce the number of
31413 unnecessary wake-ups of the X server.
31414
31415 Signed-off-by: Lennart Poettering <lennart@poettering.net>
31416 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31417 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31418
31419commit 882e3c2680c339ad7aa0d664e0b0f02b8a05b11d
31420Author: Lennart Poettering <lennart@poettering.net>
31421Date: Mon Jul 18 21:17:10 2011 +0200
31422
31423 config: process udev "changed" and "add" events in the same code paths
31424
31425 udev gives no guarantee that before each "changed" event for a device
31426 there's an "add" event, or that before each "remove" is an "add", or
31427 that before each "add" there was no "add" already and so on. Users can
31428 trigger these events at any time with "udevadm trigger", and netlink is
31429 a lossy transport, hence the events can come in unexpected ordering.
31430
31431 With other words: regardless which event is generated, the X server must
31432 not choke on it and make the best of it, hence make sure that if we get
31433 an "add" event for an existing device we don't add the device a second
31434 time.
31435
31436 Signed-off-by: Lennart Poettering <lennart@poettering.net>
31437 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31438 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31439
31440commit 5596f10cce58295d8af73646352b314ecb3ddfca
31441Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31442Date: Sat Jul 9 13:21:04 2011 +0100
31443
31444 Remove hw/xwin/xlaunch
31445
31446 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31447 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31448
31449commit 73864a87aacbce85b520ccaa6e360b82c0e99716
31450Author: Jeremy Huddleston <jeremyhu@apple.com>
31451Date: Tue Jul 19 20:03:48 2011 -0700
31452
31453 XQuartz: Localization updates
31454
31455 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31456
31457commit d2463df84d3a160b2dd1e706b02985e1817784e3
31458Author: Jeremy Huddleston <jeremyhu@apple.com>
31459Date: Tue Jul 19 20:02:33 2011 -0700
31460
31461 XQuartz: Add english NIB changes for scroll_in_device_direction
31462
31463 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31464
31465commit e36e7ad6314ca5fcd3292b8022f0a618d567d72b
31466Author: Jeremy Huddleston <jeremyhu@apple.com>
31467Date: Tue Jul 19 20:01:11 2011 -0700
31468
31469 XQuartz: Add GUI controls to toggle scroll_in_device_direction
31470
31471 Also cleaned up the wording for fullscreen_menu.
31472
31473 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31474
31475commit ac43984215faf464fd48bd006cac1c6539603cea
31476Author: Jeremy Huddleston <jeremyhu@apple.com>
31477Date: Tue Jul 19 20:00:27 2011 -0700
31478
31479 XQuartz: Add a scroll_in_device_direction preference
31480
31481 This preference allows users to override the related option in Mac OS X's
31482 Mouse/Trackpad preferences. This effectively lets the user determine
31483 which "context" all of X11 fits into for context-based scrolling until
31484 such API exists within X11 itself to pass along to X11 clients.
31485
31486 This is applicable to Mav OS X 10.7+
31487
31488 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31489
31490commit ec95a9c829b8e37529aa828c05fbaabc45cffe42
31491Author: George Staplin <gstaplin@apple.com>
31492Date: Wed Mar 4 02:03:52 2009 -0700
31493
31494 XQuartz: Add diagnostic error checking to xp_destroy_surface.
31495
31496 This occurred to me in hindsight after the last commit. If the
31497 original developer had done this, we would have noticed the
31498 problem sooner.
31499 (cherry picked from commit aa0a57996f3e7d16238f69976958c2526821388b)
31500
31501commit 4fe7df265324f63025686efe9d32342e3cef40d3
31502Author: George Staplin <gstaplin@apple.com>
31503Date: Wed Mar 4 01:39:58 2009 -0700
31504
31505 XQuartz: Fix a memory leak with surfaces that a new test found.
31506
31507 xp_destroy_surface was called with a surface id of 0, due to some
31508 premature cleanup that set it to 0. This means the surfaces
31509 weren't being destroyed until the window was.
31510
31511 The code that did that was: pDRIDrawablePriv->sid = 0;
31512
31513 In long running applications this leak may or may not have been
31514 harmful. With the old libGL the surfaces weren't destroyed until
31515 the context was destroyed or a new context created. In the new
31516 libGL they are reference counted, and released much sooner, so we
31517 ran into a resource leak more noticeably with some tests.
31518
31519 Make the Apple DRI code dispatch events to the client(s) for
31520 destroyed surfaces, when a resource is destroyed. This seems to
31521 work in my tests, however this clearly wasn't working for a while,
31522 so bugs may result in the future if it enables some new (unexpected)
31523 side effects.
31524
31525 Also add a few helpful comments to aid in understanding the code
31526 in the future.
31527
31528 Tested with the test suite, Pymol, and various Mesa demos.
31529 (cherry picked from commit bede83eb19a1629396fcd5a46441f8476a8fcd1b)
31530
31531commit 0ebe45a717faa6464d3b1ab73e30570518ee4798
31532Author: Jeremy Huddleston <jeremyhu@apple.com>
31533Date: Tue Jul 19 19:42:44 2011 -0700
31534
31535 XQuartz: DRI: Dead code removal
31536
31537 Also add some comments that weren't merged in from server-1.4-apple's
31538 99babae1326485c27eb9253db83afdd6aef9e362
31539
31540 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31541
31542commit 88ad050f3c2f9f11bafb56a8cf777518795295a9
31543Author: Jeremy Huddleston <jeremyhu@apple.com>
31544Date: Mon Jul 18 22:54:21 2011 -0700
31545
31546 XQuartz: Add some sanity checking and a fallback for the bundle id.
31547
31548 This way we'll print an error and still mostly work rather than crashing if
31549 someone installs XQuartz.app incorrectly or tries running the server within
31550 the build system rather than the installed system.
31551
31552 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31553
31554commit 89bfa0c633f3096cca6f983bca1507d3f4b902a8
31555Author: Jeremy Huddleston <jeremyhu@apple.com>
31556Date: Fri Jul 15 10:39:39 2011 -0700
31557
31558 XQuartz: DarwinEQFini doesn't return Bool
31559
31560 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
31561
31562commit 82f5521a6d91ebcd2a4400f6c221ad625edc99a1
31563Author: Daniel Stone <daniel@fooishbar.org>
31564Date: Wed Jul 13 12:08:04 2011 +0100
31565
31566 XKB: Work around broken interps from old xkbcomp
31567
31568 Bugfix for broken xkbcomp: if we encounter an XFree86Private action with
31569 Any+AnyOfOrNone(All), then we skip the interp as broken. Versions of
31570 xkbcomp below 1.2.2 had a bug where they would interpret a symbol that
31571 couldn't be found in an interpret as Any. So, an
31572 XF86LogWindowTree+AnyOfOrNone(All) interp that triggered the PrWins
31573 action would make every key without an action trigger PrWins if libX11
31574 didn't yet know about the XF86LogWindowTree keysym. None too useful.
31575
31576 We only do this for XFree86 actions, as the current XKB dataset relies
31577 on Any+AnyOfOrNone(All) -> SetMods for Ctrl in particular.
31578
31579 See xkbcomp commits 2a473b906943ffd807ad81960c47530ee7ae9a60 and
31580 3caab5aa37decb7b5dc1642a0452efc3e1f5100e for more details.
31581
31582 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31583 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
31584 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31585
31586commit c177a5bcaa170d24a784540460906cc04ac5c752
31587Author: Daniel Stone <daniel@fooishbar.org>
31588Date: Wed Jul 13 12:08:03 2011 +0100
31589
31590 Fix non-Composite builds in PrintWindowTree
31591
31592 The previous patch accidentally introduced a hard dependency on
31593 Composite. Sorry, OS X.
31594
31595 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31596 Reported-by: Jeremy Huddleston <jeremyhu@apple.com>
31597 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31598
31599commit 01dbf2514aa0cc8c40a876a24e3cb0737c41da19
31600Merge: 5a801af af0fbc3
31601Author: Keith Packard <keithp@keithp.com>
31602Date: Wed Jul 13 11:44:15 2011 -0700
31603
31604 Merge remote-tracking branch 'jturney/master'
31605
31606commit 5a801af689fce9dfe84453bbc2b029cb4782de83
31607Author: Gaetan Nadon <memsize@videotron.ca>
31608Date: Tue Jul 12 19:50:05 2011 -0400
31609
31610 XWinrc: replace hard coded section number with __filemansuffix__
31611
31612 Reviewed-by: Cyril Brulebois <kibi@debian.org>
31613 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
31614 Signed-off-by: Keith Packard <keithp@keithp.com>
31615
31616commit af0fbc37e35ddaabf12a8d7abc79ea11958ec8ac
31617Author: Gaetan Nadon <memsize@videotron.ca>
31618Date: Thu Jun 23 07:45:13 2011 -0400
31619
31620 XWinrc: replace hard coded section number with __filemansuffix__
31621
31622 Reviewed-by: Cyril Brulebois <kibi@debian.org>
31623 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
31624 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31625
31626commit 42bc198775a78b328612427c4c8ed4d17e005ae9
31627Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31628Date: Mon Jan 24 18:29:41 2011 +0000
31629
31630 glx: Remove a few lingering traces of __GLXscreen.GLXVersion
31631
31632 The GLXversion member of the __GLXscreen struct
31633 is just cruft since commit ad5c0d9efa47476ed5cf75c82265c73919e468b4,
31634 when we started returning the minimum GLX version supported by all
31635 of the screens on the display, rather than the maximum version supported
31636 by the server.
31637
31638 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
31639 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31640
31641commit 005e68ba59782b60dd5c44a06c75b1fa5e8d70d8
31642Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31643Date: Thu Jan 20 16:21:04 2011 +0000
31644
31645 hw/xwin: Add -wgl option to XWin manpage
31646
31647 Document the -[no]wgl options in the XWin manpage
31648
31649 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31650 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31651
31652commit d62726019e4b7bfac8a23d0d733677b141047d66
31653Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31654Date: Tue Nov 2 11:38:53 2010 +0000
31655
31656 hw/xwin: Add items to WGL AIGLX todo
31657
31658 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31659 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31660
31661commit 276735ef5d12e79fe051996f100c5730c1e69334
31662Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31663Date: Mon Oct 11 16:22:49 2010 +0100
31664
31665 hw/xwin: Report Window XIDs in Window debug messages
31666
31667 Report Window XIDs in Window create/destroy/reparent debug messages
31668
31669 It's actually quite useful if you are trying to corrolate those events
31670 with what a client is doing...
31671
31672 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31673 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31674
31675commit b0b54338e2f8aed6e6a683ebddd2a2acbefabecd
31676Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31677Date: Mon Oct 11 16:25:03 2010 +0100
31678
31679 hw/xwin: Add a flag to track which windows have been drawn to using WGL.
31680
31681 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31682 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31683
31684commit 98c6d91a48f460fdbca93700ab637eadf946fb0e
31685Author: Colin Harrison <colin.harrison@virgin.net>
31686Date: Thu Mar 24 20:39:51 2011 +0000
31687
31688 hw/xwin/glx: Fix some warnings in generated wrapper code
31689
31690 Add a suitable cast to the generated code for glWinSetupDispatchTable()
31691 so it doesn't generate screeds of warnings
31692
31693 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31694 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31695
31696commit b67e80c84e4af54007aa1bd1e4a97a4d1b3d3765
31697Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31698Date: Mon Jan 24 18:41:05 2011 +0000
31699
31700 hw/xwin/glx: Handle failure to get any fbconfigs more gracefully.
31701
31702 Handle failure to get any useful pixel formats for GLX fbconfigs
31703 more gracefully: If we didn't get any useful pixel formats from
31704 wglGetPixelFormatAttribivARB(), fall back to using DescribePixelFormat().
31705 If that doesn't give us any useful pixel formats, fallback to software
31706 rendering.
31707
31708 This works around a problem with Intel 845G drivers, where
31709 wglGetPixelFormatAttribivARB() doesn't seem to work as we expect it to...
31710
31711 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31712 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31713
31714commit 9915b169237e3b2d817bd3d4455f3c01dd7dd2b5
31715Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31716Date: Mon Nov 1 19:53:42 2010 +0000
31717
31718 hw/xwin/glx: Don't spam log with wglwrap symbol resolution status at startup
31719
31720 ... instead just log if an attempt is made to call a wrapper for
31721 a function which didn't resolve
31722
31723 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31724 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31725
31726commit 8216316867f07819b8e8b377f2633fbf8876abf7
31727Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31728Date: Sat Oct 30 17:23:52 2010 +0100
31729
31730 hw/xwin/glx: Better handling of SetPixelFormat() failure
31731
31732 Propagate and report the failure if SetPixelFormat() fails
31733
31734 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31735 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31736
31737commit eecc8e06eb4bdfd4354144206d990cd9ccb69721
31738Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31739Date: Sat Oct 30 17:22:33 2010 +0100
31740
31741 hw/xwin/glx: Improvements to glxWinErrorMessage() reporting in WGL GLX provider
31742
31743 Request the message using languageID 0 (best effort), rather than only using language neutral messages
31744 Always report the numeric error code.
31745 Trim any trailing \r from FormatMessage() output
31746
31747 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31748 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31749
31750commit 3873be16d050ab12fc78f1e7f0c36c2ac88d0a98
31751Author: Jon TURNEY <jon.turney@dronecode.org.uk>
31752Date: Mon Oct 11 16:21:40 2010 +0100
31753
31754 hw/xwin/glx: Fix fbconfig dumper formatting for 3 digit index numbers
31755
31756 Some graphics hardware supports hundreds of pixel formats, so adjust
31757 formatting in fbconfig dumper for 3 digit index numbers
31758
31759 Also report the PFD_SUPPORT_DIRECTDRAW, PFD_DIRECT3D_ACCELERATED and
31760 PFD_SUPPORT_COMPOSITION flags introduced with aero
31761
31762 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
31763 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31764
31765commit d4096abb591353405417e53816e5c46e904e7b25
31766Merge: 61f87a7 87d4f90
31767Author: Keith Packard <keithp@keithp.com>
31768Date: Fri Jul 1 15:45:45 2011 -0700
31769
31770 Merge remote-tracking branch 'whot/for-keith'
31771
31772commit 87d4f90bfcb509471ac9e7886e14a92b33223fd7
31773Author: Peter Hutterer <peter.hutterer@who-t.net>
31774Date: Mon Jun 27 09:10:42 2011 +1000
31775
31776 input: free the EQ allocated memory on shutdown (#38634)
31777
31778 mieqFini() already does the right thing, but it needs to be called by the
31779 various DDXs and the XTest Extension.
31780
31781 X.Org Bug 38634 <http://bugs.freedesktop.org/show_bug.cgi?id=38634>
31782
31783 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31784 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
31785 Acked-by: Jeremy Huddleston <jeremyhu@apple.com>
31786 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31787
31788commit e084ac30b37c36016bb0dbb1b79e60a7799aee02
31789Author: Daniel Stone <daniel@fooishbar.org>
31790Date: Wed Mar 2 14:52:42 2011 +0000
31791
31792 Input: Reset SD remainder when copying co-ords from MD
31793
31794 In updateSlaveDeviceCoords, pDev->last.valuators was being copied from
31795 the master, but pDev->last.remainder wasn't. Make sure we copy both, to
31796 avoid minor inconsistencies.
31797
31798 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31799 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31800 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31801
31802commit 23a783278963919778d017965542e0f983814e37
31803Author: Peter Hutterer <peter.hutterer@who-t.net>
31804Date: Fri Jun 24 11:02:23 2011 +1000
31805
31806 input: add POINTER_NORAW to avoid generation of raw events (#30068)
31807
31808 RawEvents are supposed to be events coming from the driver. When warping the
31809 pointer, this should not generate a raw event.
31810
31811 X.Org Bug 30068 <http://bugs.freedesktop.org/show_bug.cgi?id=30068>
31812
31813 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31814 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31815 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31816
31817commit 9444e40d77508d545e3c0a8123f61950c22d89fa
31818Author: Daniel Stone <daniel@fooishbar.org>
31819Date: Tue Mar 1 11:11:22 2011 +0000
31820
31821 Test: Input: Only ever set up to MAX_VALUATORS valuators
31822
31823 Previously, the input tests were working up to sizeof(mask) * 8, which
31824 could be more than the arrays; the latter only being sized as
31825 MAX_VALUATORS.
31826
31827 Hypothetically, if you were switching the stored valuator values to
31828 double instead of uint32_t, and you attempted to set the 39th member of
31829 a MAX_VALUATORS-sized (36) array, you'd probably end up smashing
31830 ev->key_repeat into oblivion and then tripping the check for invalid
31831 flags because you haven't yet put XIPointerEmulated into the valid flags
31832 for XI_Motion. Probably.
31833
31834 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31835 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31836 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31837
31838commit 1e65a01cf73f1fbdfe0d63eaecea412cf8472cb4
31839Author: Daniel Stone <daniel@fooishbar.org>
31840Date: Fri Jun 17 16:28:05 2011 +0100
31841
31842 Test: Ensure libxservertest gets relinked when necessary
31843
31844 Similar to how we link Xorg, make sure that whenever any of the
31845 component libraries changes, we relink libxservertest and the tests.
31846 Not much use testing anything other than the actual source in your tree.
31847
31848 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31849 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
31850 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31851
31852commit 0f41d6c851df76d0423d9c37ad809d3440906944
31853Author: Daniel Stone <daniel@fooishbar.org>
31854Date: Fri Jun 17 13:20:19 2011 +0100
31855
31856 KDrive: Remove useless miPointerUpdateSprite call
31857
31858 miPointerUpdateSprite is already called from mieqProcessInputEvents, so
31859 calling it by hand immediately after isn't massively helpful.
31860
31861 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31862 Reviewed-by: Cyril Brulebois <kibi@debian.org>
31863 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31864
31865commit 57b767d2927e5c2e561a431be4e604df65423422
31866Author: Daniel Stone <daniel@fooishbar.org>
31867Date: Fri Jun 17 13:19:46 2011 +0100
31868
31869 DMX: Remove useless miPointerUpdateSprite call
31870
31871 miPointerUpdateSprite is already called from mieqProcessInputEvents, so
31872 calling it by hand immediately after isn't massively helpful.
31873
31874 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31875 Reviewed-by: Cyril Brulebois <kibi@debian.org>
31876 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31877
31878commit cf398755c0ca1e031514862b670956e7b9ace5ef
31879Author: Daniel Stone <daniel@fooishbar.org>
31880Date: Fri Jun 17 15:29:44 2011 +0100
31881
31882 XWin: Remove executable bit from headers
31883
31884 Headers don't really need to be mode 0755.
31885
31886 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31887 Reviewed-by: Cyril Brulebois <kibi@debian.org>
31888 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31889
31890commit 7d2543a3cb3089241982ce4f8984fd723d5312a1
31891Author: Daniel Stone <daniel@fooishbar.org>
31892Date: Wed Dec 29 12:03:01 2010 +0000
31893
31894 XKB: Add debug key actions for grabs & window tree
31895
31896 Add four new private XKB actions for debugging:
31897 * PrGrbs: print active grabs to the log file
31898 * Ungrab: ungrab all currently active grabs
31899 * ClsGrb: kill clients with active grabs
31900 * PrWins: dump the current window tree to the log file
31901
31902 To use these, you need to modify your XKB maps, e.g. the following to
31903 have Ctrl+Alt+(F9-F12) mapped to the above:
31904 - compat/xfree86:
31905 interpret XF86LogGrabInfo {
31906 action = Private(type=0x86, data="PrGrbs");
31907 };
31908 interpret XF86Ungrab {
31909 action = Private(type=0x86, data="Ungrab");
31910 }
31911 interpret XF86ClearGrab {
31912 action = Private(type=0x86, data="ClsGrb");
31913 }
31914 interpret XF86LogWindowTree {
31915 action = Private(type=0x86, data="PrWins");
31916 }
31917
31918 - symbols/pc:
31919 key <FK09> { type="CTRL+ALT", [ Return, XF86LogGrabInfo ] };
31920 key <FK10> { type="CTRL+ALT", [ Return, XF86Ungrab ] };
31921 key <FK11> { type="CTRL+ALT", [ Return, XF86ClearGrab ] };
31922 key <FK12> { type="CTRL+ALT", [ Return, XF86LogWindowTree ] };
31923
31924 At the moment, this only works if the grabbing client continues to call
31925 AllowEvents, as the server does no event processing at all when a device
31926 is frozen.
31927
31928 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31929 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31930 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31931
31932commit ddf735fd4e9658bb5610f5e911f3b4055d5acf89
31933Author: Daniel Stone <daniel@fooishbar.org>
31934Date: Tue Oct 13 19:56:57 2009 +1100
31935
31936 DIX: Make PrintWindowTree actually useful
31937
31938 Rewrite PrintWindowTree to make it actually tell you what you want to
31939 know.
31940
31941 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
31942 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
31943 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31944
31945commit b27d61e4433a2a0140ed8a6128e4427451e37fec
31946Author: Matěj Cepl <mcepl@redhat.com>
31947Date: Fri Jun 17 15:26:17 2011 +0100
31948
31949 Fix UTF-8 encoding
31950
31951 Report to find out all non-UTF-8 files created by
31952
31953 cat extensions |xargs -I XXXX find . -name \*.XXXX |while read FILE ; do
31954 if ( iconv -f utf8 -t ucs2 $FILE >/dev/null 2>/dev/null ) ; then
31955 /bin/true
31956 else
31957 echo $FILE
31958 fi
31959 done >>report
31960
31961 Signed-off-by: Matěj Cepl <mcepl@redhat.com>
31962 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31963
31964 [Daniel: git am failed for me, so I redid it. The method listed in the
31965 commit message also failed, so I just used file/grep/iconv. The
31966 results are the same though.]
31967 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31968
31969commit 2ee85d954c8f3a6affbd9b1b745594ff12e8b670
31970Author: Scott James Remnant <scott@netsplit.com>
31971Date: Tue Jun 14 16:36:07 2011 -0700
31972
31973 dix: avoid calling deleted block and wakeup handlers
31974
31975 BlockHandler and WakeupHandlers may be removed within a different
31976 BlockHandler or WakeupHandler, especially since config/udev uses
31977 these and removes devices.
31978
31979 Calling the deleted handlers and passing potentially freed data
31980 can result in the X server segfaulting after device removal, or
31981 events that result in device removal such as undocking or suspend/
31982 resume.
31983
31984 Signed-off-by: Scott James Remnant <scott@netsplit.com>
31985 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
31986 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
31987
31988commit b573cdd40ee49fae299fe2e7cbd02159ae8ae617
31989Author: Marcin Slusarz <marcin.slusarz@gmail.com>
31990Date: Tue Jun 7 21:22:15 2011 +0200
31991
31992 dri2: restore Screen->ConfigNotify on close
31993
31994 ConfigNotify is set by DRI2ScreenInit, but not restored to
31995 previous state on close. Fix it.
31996 (I'm preparing a patch for xf86-video-nouveau which detects GPU lockup
31997 after dri2 init and it needs to reinitialize dri2)
31998
31999 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
32000 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32001 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32002
32003commit 61f87a75f2bcda939a1778d39be8cfa5c886e6d8
32004Merge: 0d2153a 72d9143
32005Author: Keith Packard <keithp@keithp.com>
32006Date: Wed Jun 29 20:14:51 2011 -0700
32007
32008 Merge remote-tracking branch 'jturney/master'
32009
32010commit 0d2153a46cc72d5f2e6a9081a3cf153aa3eb7787
32011Merge: b631c39 932513e
32012Author: Keith Packard <keithp@keithp.com>
32013Date: Wed Jun 29 20:08:32 2011 -0700
32014
32015 Merge remote-tracking branch 'jbarnes/master'
32016
32017commit b631c39a21c8d74fd00b8932df342d6921cce8f4
32018Author: Gaetan Nadon <memsize@videotron.ca>
32019Date: Thu Jun 23 13:37:45 2011 -0400
32020
32021 Remove unused check for PERL program
32022
32023 Unable to find any use of the PERL Automake variable.
32024 It was used in hw/xfree86/scanpci around 2005.
32025 Should it ever be needed, use XORG_WITH_PERL macro.
32026
32027 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32028 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
32029 Signed-off-by: Keith Packard <keithp@keithp.com>
32030
32031commit d5ca33ca2dd08e8436439da926d2dedaabc268fa
32032Merge: 38d55f0 4edf49d
32033Author: Keith Packard <keithp@keithp.com>
32034Date: Wed Jun 29 18:54:33 2011 -0700
32035
32036 Merge remote-tracking branch 'kibi/master'
32037
32038commit 72d914335c9ad0e0f4a34b9f7e2901b830c499bf
32039Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
32040Date: Tue Mar 23 04:34:22 2010 -0500
32041
32042 Cygwin/X: Left-justify website link in About box
32043
32044 Left-justify website link in About box. This is a cosmetic fix to make
32045 the About box display correctly when Windows is configured with a
32046 non-default DPI value
32047
32048 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
32049 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
32050 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
32051 Tested-by: Colin Harrison <colin.harrison@virgin.net>
32052
32053commit e78628406f0428454983db1c51a8c828bf5684e5
32054Author: Jon TURNEY <jon.turney@dronecode.org.uk>
32055Date: Thu Aug 12 15:00:01 2010 +0100
32056
32057 Cygwin/X: Preserve client area size and position on Windows window style change
32058
32059 When the style changes, adjust the window size so the client area remains the same.
32060
32061 Otherwise the window size may change when sizing is reflected from Windows to X, and
32062 some windows are drawn expecting them to be exactly the requested size (e.g. the
32063 gmplayer control window)
32064
32065 Use DeferWindowPos to delay the resize to preserve client area on WM_STYLECHANGING
32066 until after the style change has actually happened in WM_STYLECHANGED
32067
32068 As a consquence of this, we need to be more careful to create windows with exactly
32069 the requested placement and client area initially, so the client area matches what
32070 the X client requested
32071
32072 Also synchronize the X windows idea of the placement of a window which Windows is
32073 allowed to place
32074
32075 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
32076 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
32077 Tested-by: Colin Harrison <colin.harrison@virgin.net>
32078
32079commit 1d7f3a0031b4ae7d8aa984799ffa578788061ce4
32080Author: Jon TURNEY <jon.turney@dronecode.org.uk>
32081Date: Sat Nov 7 19:34:12 2009 +0000
32082
32083 glx: Move GLX extension string utility functions
32084
32085 Move GLX extension string utility functions from the DRI1 convenience library
32086 into the GLX extension convenience library, so other DDX which don't have DRI
32087 can use them.
32088
32089 This is probably also needed if anyone actually tries to build an Xorg DDX
32090 with only DRI2 support...
32091
32092 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
32093 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
32094
32095commit 932513e23b5fe1793cc38eabe02df82e1cc6e4d9
32096Author: Jesse Barnes <jbarnes@virtuousgeek.org>
32097Date: Thu May 5 13:11:46 2011 -0700
32098
32099 DRI2/GLX: use new swap event types
32100
32101 Use the new event types so we can pass a valid SBC value to clients.
32102 Fix up the completion calls to use CARD32 instead of CARD64 to match
32103 the new field size.
32104
32105 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32106 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
32107 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
32108
32109commit a3cb7d097a281252cebbc3c65c67149e106482ac
32110Author: Jesse Barnes <jbarnes@virtuousgeek.org>
32111Date: Tue May 3 10:36:44 2011 -0700
32112
32113 GLX/DRI2: pass drawable correctly for indirect swap events
32114
32115 Pass the right drawable pointer as data to the swap complete function.
32116
32117 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
32118
32119commit 3e8270b869ebf788778d5d56107f6d47c7c050d0
32120Author: Adam Jackson <ajax@redhat.com>
32121Date: Tue Apr 12 15:30:25 2011 -0400
32122
32123 glx: Fix fbconfigs with no corresponding visual
32124
32125 There are, in general, more fbconfig depths than visual depths.
32126 fbconfigs need not support Window rendering, however any that do must
32127 have an associated visual ID (which we got right), and any that do not
32128 must not claim GLX_WINDOW_BIT in GLX_DRAWABLE_TYPE (which we got
32129 wrong).
32130
32131 Fixes piglit/glx-fbconfig-sanity, assuming you have a sufficiently
32132 recent piglit, as that test formerly wrongly required pixmap-capable
32133 fbconfigs to have a visual.
32134
32135 v2: Additional check for fbconfigs that didn't have GLX_WINDOW_BIT in
32136 the first place, from previous patch by Jon TURNEY; also, also clear
32137 ->visualID.
32138
32139 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
32140 Signed-off-by: Adam Jackson <ajax@redhat.com>
32141
32142commit 38d55f06cfbf4b05fee0aef8f97f90e835beaacb
32143Author: Jon TURNEY <jon.turney@dronecode.org.uk>
32144Date: Mon Jun 27 15:06:16 2011 +0100
32145
32146 Guard use of backtrace() with HAVE_BACKTRACE
32147
32148 Guard the use of backtrace() with HAVE_BACKTRACE, since we already
32149 have the autoconf machinery for setting that.
32150
32151 For the moment, assume dladdr() is available when backtrace() is
32152
32153 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
32154 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32155
32156commit bea57392c724f439987832b422941ae897953e4a
32157Author: Jeremy Huddleston <jeremyhu@apple.com>
32158Date: Sat Jun 25 11:29:53 2011 -0700
32159
32160 XQuartz: Fix Makefile dependencies
32161
32162 automake generates _DEPENDENCIES from _LIBADD, but it strips out variables.
32163 This resulted in not relinking if some components were rebuilt (like
32164 libdix, libos, etc).
32165
32166 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32167
32168commit c8b80a82d97d92d445c1c0af6dba42c9de7c56cd
32169Author: Jeremy Huddleston <jeremyhu@apple.com>
32170Date: Tue Jun 21 21:14:46 2011 -0700
32171
32172 XQuartz: Make a call to activateIgnoringOtherApps in our NSApplicationActivatedEventType handler
32173
32174 In addition, this change will not call into the X11 activation unless an X11
32175 window was active when we deactivated. We can't rely on the event and current
32176 key windows because the key window will be nil until activated, and the event
32177 will only reference the window if the window was clicked (whereas it will be
32178 nil if we activated via dock or cmd-tab).
32179
32180 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32181
32182commit 9d568450b1319e9057319ebb37e76003bcba447d
32183Author: Keith Packard <keithp@keithp.com>
32184Date: Tue Jun 28 09:33:25 2011 -0700
32185
32186 Create sdksyms.dep in the right place at configure time
32187
32188 sdksyms moved from hw/xfree86/loader to hw/xfree86, so the
32189 configure-time create of sdksyms.dep needs to reflect that
32190 change. Otherwise, make might be confused by a missing file and (more
32191 importantly to me) hw/xfree86/loader/sdksyms.dep will be left around
32192 after 'make clean' causing 'make distcheck' to fail.
32193
32194commit f968f4ace9410d827fb6b68c4e38ea9516641309
32195Merge: 4dbed26 163441f
32196Author: Keith Packard <keithp@keithp.com>
32197Date: Tue Jun 28 09:10:14 2011 -0700
32198
32199 Merge remote-tracking branch 'dbn/no-libxorg'
32200
32201commit 4dbed2625431a92e20bede7b8cef847b5d5d99ba
32202Author: Peter Hutterer <peter.hutterer@who-t.net>
32203Date: Mon Jun 27 14:45:18 2011 +1000
32204
32205 xfree86: Only log the serial bytes if debug is on.
32206
32207 Introduced in e3f296d91dfe6b827195e1d387e1a04aa73b85c3, when the ifdef DEBUG
32208 around the whole block was removed, but only two of the three ErrorF
32209 switched to DebugF.
32210
32211 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32212 Signed-off-by: Keith Packard <keithp@keithp.com>
32213
32214commit 163441fe531a77f7fa54bfa59c5ab49c78f44306
32215Author: Dan Nicholson <dbn.lists@gmail.com>
32216Date: Tue Jun 14 20:03:06 2011 -0700
32217
32218 xfree86: Remove libxorg convenience library to speed up build
32219
32220 libxorg.la served to collect all the Xorg convenience libraries into one
32221 massive archive to link into Xorg. This made things easy for symbol
32222 resolution, but it tremendously slowed down the build since each change
32223 caused libxorg.la to be rebuilt. This is an extremely slow process of
32224 extracting all the objects from the sub-libraries and recombining them.
32225
32226 Instead, the archives are linked directly into Xorg. The order of the
32227 libraries had to be tweaked a bit to make symbols resolve correctly with
32228 the lower level code moving later in the link command.
32229
32230 As a side effect, since the dtrace objects are now being linked
32231 directly into Xorg, we don't need the SPECIAL_DTRACE_OBJECTS hack to
32232 add them twice.
32233
32234 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
32235 Tested-by: Jamey Sharp <jamey@minilop.net>
32236 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32237 Tested-by: Peter Hutterer <peter.hutterer@who-t.net>
32238
32239commit 6259b30111f568ec3d1b32dc6382bce46d8e6ccc
32240Author: Dan Nicholson <dbn.lists@gmail.com>
32241Date: Tue Jun 14 19:47:02 2011 -0700
32242
32243 xfree86: Move sdksyms generation to ddx toplevel
32244
32245 The symbols in sdksyms.c cover the entire source tree. In order to make
32246 them resolve when libxorg.la goes away, move the objects from libloader
32247 to Xorg. Unfortunately, this means sdksyms needs to get built again for
32248 the test code.
32249
32250 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
32251 Tested-by: Jamey Sharp <jamey@minilop.net>
32252 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32253 Tested-by: Peter Hutterer <peter.hutterer@who-t.net>
32254
32255commit 7d5c8a12cbc295b3e33e1b60bd7330e2bc93a966
32256Author: Dan Nicholson <dbn.lists@gmail.com>
32257Date: Tue Jun 14 17:21:48 2011 -0700
32258
32259 Don't use empty source files
32260
32261 When an empty _SOURCES variable is declared, automake will recognize that
32262 only linking is needed.
32263
32264 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
32265 Tested-by: Jamey Sharp <jamey@minilop.net>
32266 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32267 Tested-by: Peter Hutterer <peter.hutterer@who-t.net>
32268
32269commit 8ac651552bb70ba36238f430adab2f7a7f24db6e
32270Merge: 945b2ff 47b6ba3
32271Author: Keith Packard <keithp@keithp.com>
32272Date: Wed Jun 22 11:03:37 2011 -0700
32273
32274 Merge remote-tracking branch 'alanc/master'
32275
32276commit 945b2ff8141c58344fc9558ee3853fad3e86134b
32277Author: Gaetan Nadon <memsize@videotron.ca>
32278Date: Wed Jun 8 15:18:12 2011 -0400
32279
32280 docbook.am: --path is also required for xsltproc
32281
32282 The XSLT processor also needs to know about the entities defined
32283 in xserver.ent. It removes error messages.
32284
32285 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32286 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
32287 Signed-off-by: Keith Packard <keithp@keithp.com>
32288
32289commit 6326b0f0746ded410672ad0270d1c8a0e9cb59f9
32290Author: Gaetan Nadon <memsize@videotron.ca>
32291Date: Wed Jun 8 15:17:59 2011 -0400
32292
32293 Update .gitignore as new tests have been added
32294
32295 Refactor to follow the pattern one Makefile, one .gitignore
32296 where needed.
32297
32298 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
32299 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
32300 Signed-off-by: Keith Packard <keithp@keithp.com>
32301
32302commit 47b6ba3204d650393a7255ce2af27b8c018bb586
32303Author: Alan Coopersmith <alan.coopersmith@oracle.com>
32304Date: Mon Jun 20 22:17:03 2011 -0700
32305
32306 Add xkeyboard-config to See Also of man pages referencing xkb
32307
32308 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32309 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
32310
32311commit cfbe471b586282aea89f7b4802c051f22301bf59
32312Author: Alan Coopersmith <alan.coopersmith@oracle.com>
32313Date: Mon Jun 20 22:06:00 2011 -0700
32314
32315 Xdmx.man: convert section references to substituted forms
32316
32317 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32318 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
32319 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
32320
32321commit c3278f3c1786e66019a39e2612761f74782a48ee
32322Author: Alan Coopersmith <alan.coopersmith@oracle.com>
32323Date: Sat Jun 11 10:56:08 2011 -0700
32324
32325 Xdmx.man: additional updates to modernize XKB information
32326
32327 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32328 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
32329 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
32330 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32331
32332commit 19e37067131175219eca7d2b11c2ee59af280575
32333Author: Alan Coopersmith <alan.coopersmith@oracle.com>
32334Date: Sat Jun 11 10:41:48 2011 -0700
32335
32336 Xdmx.man: Show actual configured XKB defaults instead of old hardcoded values
32337
32338 Passed through from configure.ac via manpages.am
32339
32340 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32341 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
32342 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
32343 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32344
32345commit acf2173b321749c42d29d858d3521a4d506c732e
32346Author: Gaetan Nadon <memsize@videotron.ca>
32347Date: Thu Jun 16 20:12:03 2011 -0400
32348
32349 Install xml versions of specs even if HAVE_XMLTO is false
32350
32351 DocBook/XML input source is also a useful output format that can be viewed
32352 with an XML viewer or editor and by some O/S help system.
32353
32354 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
32355 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32356 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
32357 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32358
32359commit 44890b3e266116ae71bfca2590e5fd9cdfbdb69b
32360Author: Gaetan Nadon <memsize@videotron.ca>
32361Date: Thu Jun 16 20:11:52 2011 -0400
32362
32363 Install target dbs alongside generated documents
32364
32365 This matches a change in xorg-sgml-docs whereby the masterdb will look for
32366 the target dbs into the same location as the generated documents.
32367
32368 The target dbs are now installed alongside the generated documents.
32369 Previously they are installed in $prefix/sgml/X11/dbs alongside masterdb which
32370 has the potential of installing outside the package prefix and cause
32371 distcheck to fail when user does not have write permission in this package.
32372
32373 Requires XORG_CHECK_SGML_DOCTOOLS(1.8) which was released 2011-06-11
32374
32375 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32376 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
32377 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32378
32379commit a42ef6c940604172e54de110b7d989cb2208f7ef
32380Author: Rob Clark <robdclark@gmail.com>
32381Date: Tue Jun 14 11:45:38 2011 -0500
32382
32383 EXA: fix typo
32384
32385 The incorrect drawable deltas were applied if dst was a redirected
32386 window. Resulting in a bogus region passed to prepare_access_reg().
32387
32388 Signed-off-by: Rob Clark <rob@ti.com>
32389 Reviewed-by: Michel Dänzer <michel@daenzer.net>
32390 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32391
32392commit 4edf49d032028aa4b9887d02b926b0c372c4451d
32393Author: Cyril Brulebois <kibi@debian.org>
32394Date: Tue May 24 18:33:18 2011 +0200
32395
32396 dmx/examples: Fix missing key_click_percent assignment.
32397
32398 Thanks to gcc's -Wunused-but-set-variable, stop ignoring the percent
32399 parameter, and add it to the XKeyboardControl structure before the
32400 XChangeKeyboardControl call.
32401
32402 This warning goes away accordingly:
32403 | CC xbell-xbell.o
32404 | xbell.c: In function ‘main’:
32405 | xbell.c:74:22: warning: variable ‘percent’ set but not used [-Wunused-but-set-variable]
32406
32407 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32408 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32409
32410commit 33dad2b74e3489c8a6b5fa574172d6cd2b6de037
32411Author: Cyril Brulebois <kibi@debian.org>
32412Date: Tue May 24 18:33:17 2011 +0200
32413
32414 hw/dmx: Stop using variables for text widgets.
32415
32416 They are unused in the sense they're not getting any callback set up, so
32417 there's no point in storing them into variables. Keep the initial
32418 alignment of the parameters to try and reduce the diff noise.
32419
32420 Those warnings go away accordingly:
32421 | CC xdmxconfig-xdmxconfig.o
32422 | xdmxconfig.c: In function ‘main’:
32423 | xdmxconfig.c:881:29: warning: variable ‘quittext’ set but not used [-Wunused-but-set-variable]
32424 | xdmxconfig.c:880:53: warning: variable ‘abouttext’ set but not used [-Wunused-but-set-variable]
32425
32426 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32427 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32428
32429commit 34a9f2a90d34a2ab7ecf56d329e75e8da569f939
32430Author: Cyril Brulebois <kibi@debian.org>
32431Date: Tue May 24 18:33:17 2011 +0200
32432
32433 xfixes: Remove unused variable.
32434
32435 The last use of pScreen in ProcXFixesSetPictureClipRegion was removed in
32436 aa7096ca6f108e399d9916639cf20c57f9776305 so remove it entirely.
32437
32438 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32439 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32440
32441commit ad0d0833d45ea24d7af1f724b9905133fe3b065f
32442Author: Cyril Brulebois <kibi@debian.org>
32443Date: Tue May 24 18:33:17 2011 +0200
32444
32445 xfixes: Mark some variables as unused.
32446
32447 Calling Unwrap() is just a way of performing an assignment while keeping a
32448 backup of the original value. In the CursorCloseScreen function, the backup
32449 value doesn't seem to be useful, but Unwrap() is used anyway (probably to stay
32450 in line with other functions). As a consequence, mark those variables as unused.
32451
32452 The following warnings go away accordingly:
32453 | CC cursor.lo
32454 | cursor.c: In function 'CursorCloseScreen':
32455 | cursor.c:186:26: warning: variable 'display_proc' set but not used [-Wunused-but-set-variable]
32456 | cursor.c:185:24: warning: variable 'close_proc' set but not used [-Wunused-but-set-variable]
32457
32458 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32459 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32460
32461commit f41ac25d932ef34ab912201ab5ce65f85f2ff171
32462Author: Cyril Brulebois <kibi@debian.org>
32463Date: Tue May 24 18:33:17 2011 +0200
32464
32465 miext: Mark some variables as unused.
32466
32467 There's no use for the values set through the various macro calls
32468 (fbGetDrawable and fbGetDrawablePixmap), so mark those variables as unused.
32469
32470 The following warnings go away accordingly:
32471 | CC shpacked.lo
32472 | shpacked.c: In function 'shadowUpdatePacked':
32473 | shpacked.c:55:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32474 | shpacked.c:55:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32475 | CC shplanar8.lo
32476 | shplanar8.c: In function 'shadowUpdatePlanar4x8':
32477 | shplanar8.c:105:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32478 | shplanar8.c:105:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32479 | CC shplanar.lo
32480 | shplanar.c: In function 'shadowUpdatePlanar4':
32481 | shplanar.c:101:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32482 | shplanar.c:101:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32483 | CC shrot16pack_180.lo
32484 | In file included from shrot16pack_180.c:31:0:
32485 | shrotpack.h: In function 'shadowUpdateRotate16_180':
32486 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32487 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32488 | CC shrot16pack_270.lo
32489 | In file included from shrot16pack_270.c:31:0:
32490 | shrotpack.h: In function 'shadowUpdateRotate16_270':
32491 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32492 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32493 | CC shrot16pack_270YX.lo
32494 | In file included from shrot16pack_270YX.c:31:0:
32495 | shrotpackYX.h: In function 'shadowUpdateRotate16_270YX':
32496 | shrotpackYX.h:72:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32497 | shrotpackYX.h:72:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32498 | CC shrot16pack_90.lo
32499 | In file included from shrot16pack_90.c:31:0:
32500 | shrotpack.h: In function 'shadowUpdateRotate16_90':
32501 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32502 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32503 | CC shrot16pack_90YX.lo
32504 | In file included from shrot16pack_90YX.c:31:0:
32505 | shrotpackYX.h: In function 'shadowUpdateRotate16_90YX':
32506 | shrotpackYX.h:72:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32507 | shrotpackYX.h:72:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32508 | CC shrot16pack.lo
32509 | In file included from shrot16pack.c:30:0:
32510 | shrotpack.h: In function 'shadowUpdateRotate16':
32511 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32512 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32513 | CC shrot32pack_180.lo
32514 | In file included from shrot32pack_180.c:31:0:
32515 | shrotpack.h: In function 'shadowUpdateRotate32_180':
32516 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32517 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32518 | CC shrot32pack_270.lo
32519 | In file included from shrot32pack_270.c:31:0:
32520 | shrotpack.h: In function 'shadowUpdateRotate32_270':
32521 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32522 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32523 | CC shrot32pack_90.lo
32524 | In file included from shrot32pack_90.c:31:0:
32525 | shrotpack.h: In function 'shadowUpdateRotate32_90':
32526 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32527 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32528 | CC shrot32pack.lo
32529 | In file included from shrot32pack.c:30:0:
32530 | shrotpack.h: In function 'shadowUpdateRotate32':
32531 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32532 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32533 | CC shrot8pack_180.lo
32534 | In file included from shrot8pack_180.c:31:0:
32535 | shrotpack.h: In function 'shadowUpdateRotate8_180':
32536 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32537 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32538 | CC shrot8pack_270.lo
32539 | In file included from shrot8pack_270.c:31:0:
32540 | shrotpack.h: In function 'shadowUpdateRotate8_270':
32541 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32542 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32543 | CC shrot8pack_90.lo
32544 | In file included from shrot8pack_90.c:31:0:
32545 | shrotpack.h: In function 'shadowUpdateRotate8_90':
32546 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32547 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32548 | CC shrot8pack.lo
32549 | In file included from shrot8pack.c:30:0:
32550 | shrotpack.h: In function 'shadowUpdateRotate8':
32551 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32552 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32553 | CC shrotate.lo
32554 | shrotate.c: In function 'shadowUpdateRotatePacked':
32555 | shrotate.c:62:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32556 | shrotate.c:62:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32557 | CC shpacked.lo
32558 | shpacked.c: In function 'shadowUpdatePacked':
32559 | shpacked.c:55:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32560 | shpacked.c:55:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32561 | CC shplanar8.lo
32562 | shplanar8.c: In function 'shadowUpdatePlanar4x8':
32563 | shplanar8.c:105:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32564 | shplanar8.c:105:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32565 | CC shplanar.lo
32566 | shplanar.c: In function 'shadowUpdatePlanar4':
32567 | shplanar.c:101:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32568 | shplanar.c:101:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32569 | CC shrot16pack_180.lo
32570 | In file included from shrot16pack_180.c:31:0:
32571 | shrotpack.h: In function 'shadowUpdateRotate16_180':
32572 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32573 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32574 | CC shrot16pack_270.lo
32575 | In file included from shrot16pack_270.c:31:0:
32576 | shrotpack.h: In function 'shadowUpdateRotate16_270':
32577 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32578 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32579 | CC shrot16pack_270YX.lo
32580 | In file included from shrot16pack_270YX.c:31:0:
32581 | shrotpackYX.h: In function 'shadowUpdateRotate16_270YX':
32582 | shrotpackYX.h:72:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32583 | shrotpackYX.h:72:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32584 | CC shrot16pack_90.lo
32585 | In file included from shrot16pack_90.c:31:0:
32586 | shrotpack.h: In function 'shadowUpdateRotate16_90':
32587 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32588 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32589 | CC shrot16pack_90YX.lo
32590 | In file included from shrot16pack_90YX.c:31:0:
32591 | shrotpackYX.h: In function 'shadowUpdateRotate16_90YX':
32592 | shrotpackYX.h:72:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32593 | shrotpackYX.h:72:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32594 | CC shrot16pack.lo
32595 | In file included from shrot16pack.c:30:0:
32596 | shrotpack.h: In function 'shadowUpdateRotate16':
32597 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32598 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32599 | CC shrot32pack_180.lo
32600 | In file included from shrot32pack_180.c:31:0:
32601 | shrotpack.h: In function 'shadowUpdateRotate32_180':
32602 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32603 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32604 | CC shrot32pack_270.lo
32605 | In file included from shrot32pack_270.c:31:0:
32606 | shrotpack.h: In function 'shadowUpdateRotate32_270':
32607 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32608 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32609 | CC shrot32pack_90.lo
32610 | In file included from shrot32pack_90.c:31:0:
32611 | shrotpack.h: In function 'shadowUpdateRotate32_90':
32612 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32613 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32614 | CC shrot32pack.lo
32615 | In file included from shrot32pack.c:30:0:
32616 | shrotpack.h: In function 'shadowUpdateRotate32':
32617 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32618 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32619 | CC shrot8pack_180.lo
32620 | In file included from shrot8pack_180.c:31:0:
32621 | shrotpack.h: In function 'shadowUpdateRotate8_180':
32622 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32623 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32624 | CC shrot8pack_270.lo
32625 | In file included from shrot8pack_270.c:31:0:
32626 | shrotpack.h: In function 'shadowUpdateRotate8_270':
32627 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32628 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32629 | CC shrot8pack_90.lo
32630 | In file included from shrot8pack_90.c:31:0:
32631 | shrotpack.h: In function 'shadowUpdateRotate8_90':
32632 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32633 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32634 | CC shrot8pack.lo
32635 | In file included from shrot8pack.c:30:0:
32636 | shrotpack.h: In function 'shadowUpdateRotate8':
32637 | shrotpack.h:109:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32638 | shrotpack.h:109:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32639 | CC shrotate.lo
32640 | shrotate.c:62:19: warning: variable 'shaYoff' set but not used [-Wunused-but-set-variable]
32641 | shrotate.c:62:10: warning: variable 'shaXoff' set but not used [-Wunused-but-set-variable]
32642
32643 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32644 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32645
32646commit fe6f489d21d77aacf0a073883625619018d764b3
32647Author: Cyril Brulebois <kibi@debian.org>
32648Date: Tue May 24 18:33:17 2011 +0200
32649
32650 fb: Mark some variables as unused.
32651
32652 There's no use for the values set through the various macro calls
32653 (fbGetDrawable and fbGetDrawablePixmap), so mark those variables as unused.
32654
32655 The following warnings go away accordingly:
32656 | CC libfb_la-fb24_32.lo
32657 | fb24_32.c: In function 'fb24_32ReformatTile':
32658 | fb24_32.c:544:19: warning: variable 'newYoff' set but not used [-Wunused-but-set-variable]
32659 | fb24_32.c:544:10: warning: variable 'newXoff' set but not used [-Wunused-but-set-variable]
32660 | fb24_32.c:543:19: warning: variable 'oldYoff' set but not used [-Wunused-but-set-variable]
32661 | fb24_32.c:543:10: warning: variable 'oldXoff' set but not used [-Wunused-but-set-variable]
32662 | CC libfb_la-fbfill.lo
32663 | fbfill.c: In function 'fbFill':
32664 | fbfill.c:72:21: warning: variable 'stipYoff' set but not used [-Wunused-but-set-variable]
32665 | fbfill.c:72:11: warning: variable 'stipXoff' set but not used [-Wunused-but-set-variable]
32666 | fbfill.c:100:21: warning: variable 'stipYoff' set but not used [-Wunused-but-set-variable]
32667 | fbfill.c:100:11: warning: variable 'stipXoff' set but not used [-Wunused-but-set-variable]
32668 | fbfill.c:142:20: warning: variable 'tileYoff' set but not used [-Wunused-but-set-variable]
32669 | fbfill.c:142:10: warning: variable 'tileXoff' set but not used [-Wunused-but-set-variable]
32670 | CC libfb_la-fbgc.lo
32671 | fbgc.c: In function 'fbPadPixmap':
32672 | fbgc.c:92:19: warning: variable 'yOff' set but not used [-Wunused-but-set-variable]
32673 | fbgc.c:92:13: warning: variable 'xOff' set but not used [-Wunused-but-set-variable]
32674 | fbgc.c: In function 'fbCanEvenStipple':
32675 | fbgc.c:166:23: warning: variable 'stipYoff' set but not used [-Wunused-but-set-variable]
32676 | fbgc.c:166:13: warning: variable 'stipXoff' set but not used [-Wunused-but-set-variable]
32677 | CC libfb_la-fbpush.lo
32678 | fbpush.c: In function 'fbPushPixels':
32679 | fbpush.c:238:20: warning: variable 'stipYoff' set but not used [-Wunused-but-set-variable]
32680 | fbpush.c:238:10: warning: variable 'stipXoff' set but not used [-Wunused-but-set-variable]
32681
32682 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32683 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32684
32685commit 8da16898d9de5da1f80b789f0b371070968285f8
32686Author: Cyril Brulebois <kibi@debian.org>
32687Date: Tue May 24 18:33:17 2011 +0200
32688
32689 fb: Simplify logic, get rid of set but unused variable.
32690
32691 wrapped is only useful is FB_ACCESS_WRAPPER is set; simplify the logic
32692 accordingly, and only set it when that's defined.
32693
32694 The following warning goes away accordingly:
32695 | CC libfb_la-fbarc.lo
32696 | fbarc.c: In function 'fbPolyArc':
32697 | fbarc.c:71:11: warning: variable 'wrapped' set but not used [-Wunused-but-set-variable]
32698
32699 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32700 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32701 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32702
32703commit 9a8bfc2c06f0a9206b64ae01a783cf5513f021a6
32704Author: Cyril Brulebois <kibi@debian.org>
32705Date: Tue May 24 18:33:16 2011 +0200
32706
32707 configure: Bump dependency on xproto.
32708
32709 Bumping this dependency means some additional macros can be used, like
32710 _X_UNUSED.
32711
32712 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32713 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32714
32715commit 9da5aa09d70e26dd4d2ef878a21317dae7bf24de
32716Author: Cyril Brulebois <kibi@debian.org>
32717Date: Tue May 24 18:33:11 2011 +0200
32718
32719 Xext: Fix buggy checks.
32720
32721 XvMCScreenKey is defined as XvMCScreenKeyRec's address, so will never be
32722 NULL. Use the check proposed by Alan Coopersmith and Jeremy Huddleston
32723 instead.
32724
32725 Those warnings go away accordingly:
32726 | CC xvmc.lo
32727 | xvmc.c: In function 'ProcXvMCListSubpictureTypes':
32728 | xvmc.c:490:22: warning: the comparison will always evaluate as 'false' for the address of 'XvMCScreenKeyRec' will never be NULL [-Waddress]
32729 | xvmc.c: In function 'XvMCExtensionInit':
32730 | xvmc.c:671:21: warning: the comparison will always evaluate as 'false' for the address of 'XvMCScreenKeyRec' will never be NULL [-Waddress]
32731 | xvmc.c: In function 'XvMCFindXvImage':
32732 | xvmc.c:749:22: warning: the comparison will always evaluate as 'false' for the address of 'XvMCScreenKeyRec' will never be NULL [-Waddress]
32733
32734 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32735 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32736
32737commit 778309fd0ef6b586468c1a3b569e3548503459ee
32738Author: Jeremy Huddleston <jeremyhu@apple.com>
32739Date: Tue Jun 14 13:23:00 2011 -0700
32740
32741 XQuartz: GLX: Fix a -Wformat-security compilation warning
32742
32743 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32744
32745commit 34e2598f0ad247071bd6a4312d9014d6e3b2305a
32746Author: Jeremy Huddleston <jeremyhu@apple.com>
32747Date: Tue Jun 14 11:18:50 2011 -0700
32748
32749 XQuartz: GLX: Use _glapi_create_table_from_handle to create out dispatch table
32750
32751 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32752
32753commit ecec578e35f91a2cbc5d07bc8d45241af7bb585f
32754Author: Jeremy Huddleston <jeremyhu@apple.com>
32755Date: Tue Jun 14 12:51:04 2011 -0700
32756
32757 glx: Add _glapi_create_table_from_handle
32758
32759 _glapi_create_table_from_handle was recently added to mesa to ease creation of
32760 a _glapi_table. This commit brings this API over to xserver for use in
32761 XQuartz (next commit).
32762
32763 The API was generated with scripts from current mesa followed by various
32764 script-foo to remove entries that are not in the server and add those which
32765 are still in the server but not in mesa (CullParameterdvEXT and
32766 CullParameterfvEXT)
32767
32768 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32769 Reviewed-by: Adam Jackson <ajax@redhat.com>
32770
32771commit c3df4df40761760b281a21f63336ca955670f7bc
32772Author: Jeremy Huddleston <jeremyhu@apple.com>
32773Date: Sun Jun 5 21:59:13 2011 -0400
32774
32775 XQuartz: Add missing include for <pthread.h>
32776
32777 BuildFix for Leopard and earlier.
32778
32779 https://trac.macports.org/ticket/29717
32780
32781 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32782
32783commit 08a7246f43c9f18079219a1c982bed558722f545
32784Author: Peter Hutterer <peter.hutterer@who-t.net>
32785Date: Fri May 27 14:21:39 2011 +1000
32786
32787 dix: rename ProcessRawEvents to dix/events.c:DeliverRawEvent
32788
32789 No functional changes, prep work for future changes.
32790
32791 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32792 Reviewed-by: Cyril Brulebois <kibi@debian.org>
32793 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32794 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32795
32796commit 0aa45c5c53e2e7ba0723f5a12c1a5f5d4baab933
32797Author: Peter Hutterer <peter.hutterer@who-t.net>
32798Date: Fri May 27 14:19:45 2011 +1000
32799
32800 Xi: use temporary variable for filter.
32801
32802 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32803 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32804 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32805
32806commit a1beae635caf81ec31a8998687df0ae92b505fda
32807Author: Peter Hutterer <peter.hutterer@who-t.net>
32808Date: Fri May 27 12:14:32 2011 +1000
32809
32810 dix: split DeliverEventToWindowMask up a bit more.
32811
32812 Move out the actual event delivery, it needs to be used from elsewhere.
32813
32814 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32815 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32816 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32817
32818commit 58e4831ed68c2fbc219c7250e9837e76a73a6432
32819Author: Peter Hutterer <peter.hutterer@who-t.net>
32820Date: Fri May 27 11:43:50 2011 +1000
32821
32822 dix: rename DeliverEventsToClients to DeliverEventsToWindowMask
32823
32824 To avoid confusion with a future patch and it better describes what this
32825 does anyway - delivering events to all clients that have the event mask on
32826 the window.
32827
32828 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32829 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32830 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32831
32832commit 37a53f15b0dfa1b25f7e3e826c7c945453398ac5
32833Author: Peter Hutterer <peter.hutterer@who-t.net>
32834Date: Fri May 27 11:40:56 2011 +1000
32835
32836 dix: split client list retrieval out of DeliverEventToClients
32837
32838 No functional change, but "other" was renamed to "clients".
32839
32840 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32841 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32842 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32843
32844commit 02d11af01211da55e9d93fe0e1851a0c6fe57472
32845Merge: bf9fd0a ab0df72
32846Author: Keith Packard <keithp@keithp.com>
32847Date: Tue Jun 7 12:31:15 2011 -0700
32848
32849 Merge remote-tracking branch 'kibi/master'
32850
32851commit bf9fd0a83eef771ee018438f05e2d9f20b43539b
32852Merge: 21eec36 0d14056
32853Author: Keith Packard <keithp@keithp.com>
32854Date: Tue Jun 7 07:45:22 2011 -0700
32855
32856 Merge remote-tracking branch 'whot/for-keith'
32857
32858commit 0d140567ba2b5bf364265dba07de519f691c67a1
32859Author: Peter Hutterer <peter.hutterer@who-t.net>
32860Date: Tue Jun 7 10:42:50 2011 +1000
32861
32862 dix: fix crashers with floating device.
32863
32864 dc57f89959e549403f8488eb9f23425bd7118b22 accidentally reversed the
32865 conditions.
32866
32867 in dix/events.c we try to detach floating devices. This leads to a
32868 NULL-dereference on GetMaster()->id.
32869
32870 in dix/getevents.c we try to get the master device for the floating slave
32871 and dereference it.
32872
32873 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
32874 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32875
32876commit 21eec367d0d9c142458f973062e17add4f693033
32877Author: Nicolas Kaiser <nikai@nikai.net>
32878Date: Mon Jun 6 16:53:41 2011 +0200
32879
32880 randr: void function cannot return value
32881
32882 Providing an argument to return in a function with void return type
32883 is not allowed by the C standard, and makes the Sun compilers unhappy.
32884
32885 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
32886 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
32887 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
32888 Signed-off-by: Keith Packard <keithp@keithp.com>
32889
32890commit ab0df72cd33f011cc48736cfbfce9bbe9c359a80
32891Author: Cyril Brulebois <kibi@debian.org>
32892Date: Sun Jun 5 03:21:18 2011 +0200
32893
32894 xkb: Fix case checks for Latin 4.
32895
32896 That one was missing _XkbKSLower:
32897 XK_kra: U+0138 LATIN SMALL LETTER KRA
32898
32899 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32900 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32901
32902commit 9eb6e34c2dd09af714400ce6a1063477d7a64776
32903Author: Cyril Brulebois <kibi@debian.org>
32904Date: Sun Jun 5 03:12:44 2011 +0200
32905
32906 xkb: Fix case checks for Latin 2.
32907
32908 Those ones were getting _XkbKSLower for no reasons:
32909 XK_ogonek: U+02DB OGONEK
32910 XK_doubleacute: U+02DD DOUBLE ACUTE ACCENT
32911
32912 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32913 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32914
32915commit 2b88189863579eb377bca617465bb76b05fbdebd
32916Author: Cyril Brulebois <kibi@debian.org>
32917Date: Sun Jun 5 03:03:47 2011 +0200
32918
32919 xkb: Fix case checks for Latin 1.
32920
32921 That one was missing _XkbKSLower:
32922 XK_ssharp: U+00DF LATIN SMALL LETTER SHARP S
32923
32924 That one was getting _XkbKSLower for no reasons:
32925 XK_division: U+00F7 DIVISION SIGN
32926
32927 For reference, XK_multiply was already excluded from the _XkbKSUpper
32928 check, it's no big surprise XK_division has to be excluded from the
32929 _XkbKSLower check.
32930
32931 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32932 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32933
32934commit feab04397de2684568dedaaaa8f299cac9f44f8b
32935Author: Jeremy Huddleston <jeremyhu@apple.com>
32936Date: Fri Jun 3 02:50:40 2011 -0400
32937
32938 XQuartz: AIGLX: Remove unnecessary includes in indirect.c
32939
32940 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32941
32942commit 3fa5d27e46ea8af55dcaaad7a99973c38bd11967
32943Author: Jeremy Huddleston <jeremyhu@apple.com>
32944Date: Thu Jun 2 20:04:04 2011 -0700
32945
32946 XQuartz: AIGLX: Setup dispatch table based on runtime capabilities rather than buildtime
32947
32948 This will allow XQuartz built on older systems to pickup capabilities on
32949 newer systems and prevent runtime failures when building on newer systems and
32950 running on older ones.
32951
32952 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32953
32954commit dd5d91fc954450e99d53862900ef4fa6320ff382
32955Author: Jeremy Huddleston <jeremyhu@apple.com>
32956Date: Fri Jun 3 02:34:28 2011 -0400
32957
32958 XQuartz: Remove explicit link against libGL
32959
32960 This was indented to force a link against OpenGL.framework's libGL, but it
32961 actually resulted in linking against mesa's libGL due to the ordering of -L
32962
32963 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
32964
32965commit 5115c2c7216fc10acb8a26deb904ad15ef18ce8a
32966Author: Cyril Brulebois <kibi@debian.org>
32967Date: Fri Jun 3 18:09:51 2011 +0200
32968
32969 Xephyr: Mention Xserver's options are also accepted.
32970
32971 Originally reported in the Debian BTS:
32972 http://bugs.debian.org/582650
32973
32974 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
32975 Signed-off-by: Jari Aalto <jari.aalto@cante.net>
32976 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32977
32978commit 408ed1576dbcf93f4bd27bd3b66917b1bba2226b
32979Author: Cyril Brulebois <kibi@debian.org>
32980Date: Fri Jun 3 18:09:48 2011 +0200
32981
32982 xkb: Fix case checks for Latin 8.
32983
32984 Spotted by -Wlogical-op:
32985 | CC xkbfmisc.lo
32986 | xkbfmisc.c: In function '_XkbKSCheckCase':
32987 | xkbfmisc.c:104:3: warning: logical 'and' of mutually exclusive tests is always false [-Wlogical-op]
32988 | xkbfmisc.c:118:3: warning: logical 'and' of mutually exclusive tests is always false [-Wlogical-op]
32989
32990 A quick look at the keysymdef.h file (from xproto) suggests the
32991 implementor chose to use interval checks to determine the case, but
32992 since lines weren't sorted by codepoints, checks were quite wrong.
32993
32994 Implement _XkbKSUpper/_XkbKSLower checks based on a grep for
32995 CAPITAL/SMALL (respectively) on the Latin 8 part of the said file.
32996
32997 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
32998 Signed-off-by: Cyril Brulebois <kibi@debian.org>
32999
33000commit d9bda34d0df576d155e1d682d5e2a382b8e1ffda
33001Author: Peter Hutterer <peter.hutterer@who-t.net>
33002Date: Wed Jun 1 10:24:43 2011 +1000
33003
33004 test: fix memset size for WindowRec (#37801)
33005
33006 X.Org Bug 37801 <http://bugs.freedesktop.org/show_bug.cgi?id=37801>
33007
33008 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33009 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
33010 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
33011 Reviewed-by: Cyril Brulebois <kibi@debian.org>
33012 Signed-off-by: Keith Packard <keithp@keithp.com>
33013
33014commit bfd8422e8877acf02155ca0bd7fbd2416b029720
33015Author: Peter Hutterer <peter.hutterer@who-t.net>
33016Date: Fri May 27 12:24:21 2011 +1000
33017
33018 Xi: use __func__ instead of function name.
33019
33020 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33021 Reviewed-by: Cyril Brulebois <kibi@debian.org>
33022
33023commit c042a267c40d35b4fb947f4bafdfeb541dd9d772
33024Author: Peter Hutterer <peter.hutterer@who-t.net>
33025Date: Fri May 27 10:10:48 2011 +1000
33026
33027 dix: use xi2_get_type instead of manual typecast
33028
33029 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33030 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
33031
33032commit b75cdb5bf76303162e947173e8ee68253a861272
33033Author: Peter Hutterer <peter.hutterer@who-t.net>
33034Date: Tue May 31 10:51:19 2011 +1000
33035
33036 dix: drop x/y back into the right valuators after transformation.
33037
33038 If the matrix is used for rotation, the coordinates affected may change.
33039 e.g. a valuator mask of (x, nil) becomes [x, lasty] and is rotated to
33040 [lasty, x]. Since the second value was unset, we would not drop x back into
33041 the mask, resulting in a loss of movement.
33042
33043 Thus, drop any value that changed after applying the matrix into the
33044 valuators. Thus, the example above becomes
33045 (x, nil) → [x, lasty] → [lasty, x] → (lasty, x)
33046
33047 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33048 Reviewed-by: Simon Thum <simon.thum@gmx.de>
33049
33050commit fe4b818700453b76d9f46749dac5d12540b91e63
33051Author: Peter Hutterer <peter.hutterer@who-t.net>
33052Date: Tue May 31 10:23:57 2011 +1000
33053
33054 dix: don't pass x/y to transformAbsolute
33055
33056 We passed in the mask, but didn't do anything with it. Move the logic to
33057 take the axes out of the valuator masks into transformAbsolute.
33058
33059 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33060 Reviewed-by: Simon Thum <simon.thum@gmx.de>
33061
33062commit 62f1bf96e1ea27e1eff137cb9333f3dd3b014dd0
33063Author: Peter Hutterer <peter.hutterer@who-t.net>
33064Date: Wed Jun 1 10:22:44 2011 +1000
33065
33066 test: fix memset size for WindowRec (#37801)
33067
33068 X.Org Bug 37801 <http://bugs.freedesktop.org/show_bug.cgi?id=37801>
33069
33070 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33071 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
33072 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
33073 Reviewed-by: Cyril Brulebois <kibi@debian.org>
33074
33075commit f94898928dfaf8df648bc3eea7d304826b7ac1b8
33076Author: Peter Hutterer <peter.hutterer@who-t.net>
33077Date: Tue May 31 14:48:33 2011 +1000
33078
33079 dix: fix an error message.
33080
33081 event type, not device type.
33082
33083 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33084
33085commit 8ef0974874f322e3ce2f3e6be4ab3c7e73a7b380
33086Author: Peter Hutterer <peter.hutterer@who-t.net>
33087Date: Thu May 19 14:50:17 2011 +1000
33088
33089 test: don't test for double alignment on i386. (#36986)
33090
33091 i386 is one of the few architectures that doesn't need double alignment.
33092
33093 X.Org Bug 36986 <http://bugs.freedesktop.org/show_bug.cgi?id=36986>
33094
33095 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33096
33097commit 48b4b778f61daaa161483b510ab520f3fcb84925
33098Author: Jeremy Huddleston <jeremyhu@apple.com>
33099Date: Wed Jun 1 21:55:06 2011 -0700
33100
33101 XQuartz: xpbproxy: Correct NSUInteger format strings
33102
33103 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33104
33105commit 971193b2750b98b28597ba194ed388e79b3489b6
33106Author: Jeremy Huddleston <jeremyhu@apple.com>
33107Date: Wed Jun 1 21:18:03 2011 -0700
33108
33109 XQuartz: ASL: Use xpbproxy subsystem for logging xpbproxy messages
33110
33111 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33112
33113commit 7dfd65705ccd3c6e2458d6892282520fe5ed8a6b
33114Author: Jeremy Huddleston <jeremyhu@apple.com>
33115Date: Wed Jun 1 21:06:53 2011 -0700
33116
33117 XQuartz: ASL: Use GLXAqua subsystem for logging GLXAqua messages
33118
33119 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33120
33121commit 19020b23efa15d72349bd833aebddcc64780a615
33122Author: Jeremy Huddleston <jeremyhu@apple.com>
33123Date: Wed Jun 1 21:06:08 2011 -0700
33124
33125 XQuartz: ASL: Use xpr subsystem for logging xpr messages
33126
33127 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33128
33129commit a66400f8dba9b7c5b65f9d26996fff677c42d866
33130Author: Jeremy Huddleston <jeremyhu@apple.com>
33131Date: Wed Jun 1 20:45:58 2011 -0700
33132
33133 XQuartz: ASL: Update logging to support differnet levels and subsystems
33134
33135 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33136
33137commit c0077b41b4c8bb08129db97a2a01d0af55d13fa4
33138Author: Jeremy Huddleston <jeremyhu@apple.com>
33139Date: Wed Jun 1 20:54:27 2011 -0700
33140
33141 XQuartz: Add a semicolon after DEBUG_LOG
33142
33143 Don't rely on the macro to provide it
33144
33145 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33146
33147commit fbc8aae8e9b5c7b175073f02e79d7ff1d1363a31
33148Author: Jeremy Huddleston <jeremyhu@apple.com>
33149Date: Sun May 29 01:20:44 2011 -0700
33150
33151 XQuartz: Move -lXplugin to LDFLAGS from LDADD
33152
33153 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33154
33155commit 38df706c6b8a339c4267cae7be0881c4362f36da
33156Author: Jeremy Huddleston <jeremyhu@apple.com>
33157Date: Tue May 31 13:01:05 2011 -0700
33158
33159 XQuartz: GLX: Create a new dispatch table rather than modifying the existing one
33160
33161 Fixes regression introduced by b0c665ac0fe6840dda581e4d0d0b76c703d62a7b
33162
33163 0 X11.bin 0x0000000100118293 __glXAquaScreenCreateContext + 684
33164 1 X11.bin 0x00000001001315b0 DoCreateContext + 163
33165 2 X11.bin 0x000000010013509f __glXDispatch + 211
33166 3 X11.bin 0x00000001000c7dad Dispatch + 785
33167 4 X11.bin 0x00000001000b97e5 dix_main + 1022
33168 5 X11.bin 0x00000001000122bc server_thread + 50
33169 6 libSystem.B.dylib 0x00007fff836554f6 _pthread_start + 331
33170 7 libSystem.B.dylib 0x00007fff836553a9 thread_start + 13
33171
33172 http://lists.apple.com/archives/X11-users/2011/May/msg00045.html
33173
33174 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33175 Reviewed-by: Adam Jackson <ajax@redhat.com>
33176
33177commit c5b72fd350bbdfd1facd0ddd5085f238c4cf252a
33178Author: Marko Macek <Marko.Macek@gmx.net>
33179Date: Sat May 21 13:30:59 2011 +0100
33180
33181 DIX: Set backgroundState correctly for root window
33182
33183 When we change the root window's background to None, and we've run with
33184 -wr or -br for a forced solid background, make sure we also change the
33185 background state to BackgroundPixel, so we don't try to lookup either
33186 pScreen->whitePixel or pScreen->blackPixel as a pixmap.
33187
33188 Signed-off-by: Marko Macek <Marko.Macek@gmx.net>
33189 Reviewed-by: Walter Harms <wharms@bfs.de>
33190 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
33191 Signed-off-by: Keith Packard <keithp@keithp.com>
33192
33193commit db228d3d07a6d831f53e4c05c878327ad4d045eb
33194Author: Keith Packard <keithp@keithp.com>
33195Date: Wed Jun 1 11:20:10 2011 -0700
33196
33197 Version bumped to 1.10.99.901 (1.11 RC1)
33198
33199 Signed-off-by: Keith Packard <keithp@keithp.com>
33200
33201commit 0643c056512d10be8db223d18d6563292d57e916
33202Merge: a2e6cfc d45f5b2
33203Author: Keith Packard <keithp@keithp.com>
33204Date: Tue May 31 23:45:07 2011 -0700
33205
33206 Merge remote-tracking branch 'ajax/xserver-next'
33207
33208commit a2e6cfc18aec1c0027b51572b03ec9f2ab074b56
33209Merge: bc04065 8d84fd2
33210Author: Keith Packard <keithp@keithp.com>
33211Date: Tue May 31 23:42:52 2011 -0700
33212
33213 Merge remote-tracking branch 'sandmann/for-keithp'
33214
33215commit d45f5b2493bc0a2882bf972849b5c9c50cd533ca
33216Author: Adam Jackson <ajax@redhat.com>
33217Date: Wed May 25 05:54:35 2011 -0400
33218
33219 fixes: Add support for pointer barriers
33220
33221 Implements pointer barriers as specified by version 5 of the XFIXES
33222 protocol. Barriers are axis-aligned, zero-width lines that block pointer
33223 movement for relative input devices. Barriers may block motion in either
33224 the positive or negative direction, or both.
33225
33226 v3:
33227 - Fix off-by-one in version_requests array
33228 - Port to non-glib test harness
33229 - Fix review notes from Søren Sandmann Pedersen, add tests to match
33230
33231 Co-authored-by: Peter Hutterer <peter.hutterer@who-t.net>
33232 Tested-by: Peter Hutterer <peter.hutterer@who-t.net>
33233 Signed-off-by: Adam Jackson <ajax@redhat.com>
33234 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33235
33236commit bc04065b5ce277f3ac3491ff221a60ef3c7605cf
33237Author: Alan Coopersmith <alan.coopersmith@oracle.com>
33238Date: Fri May 20 19:24:34 2011 -0700
33239
33240 "privates.h", line 198: warning: void function cannot return value
33241
33242 Providing an argument to return in a function with void return type
33243 is not allowed by the C standard, and makes the Sun compilers unhappy.
33244 (They actually flag it as an error, unless using a new enough version
33245 to be able to downgrade it to a warning with "-features=extensions".)
33246
33247 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
33248 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
33249 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
33250 Reviewed-by: Cyril Brulebois <kibi@debian.org>
33251
33252commit eadf5021794782fde861d471ed408675f4926b89
33253Author: Alan Coopersmith <alan.coopersmith@oracle.com>
33254Date: Mon May 2 19:48:42 2011 -0700
33255
33256 Use XORG_STRICT_OPTION from util-macros 1.14 to set -Werror flags
33257
33258 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
33259 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
33260
33261commit 9275b1fb6f82a6971c4177ddd3d5a859a8f24119
33262Author: Alan Coopersmith <alan.coopersmith@oracle.com>
33263Date: Mon May 2 19:47:44 2011 -0700
33264
33265 Use XORG_COMPILER_BRAND from util-macros 1.14 to check for SUNCC
33266
33267 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
33268 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
33269 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
33270
33271commit 4621bb270a36d35d4ab67f1d7fb47674683dfc5b
33272Author: Peter Hutterer <peter.hutterer@who-t.net>
33273Date: Wed May 18 15:00:54 2011 +1000
33274
33275 Add a property for device/product ID.
33276
33277 In some cases, knowing about the device model number and the device's vendor
33278 is important to activate product-specific settings. Since this is
33279 nonetheless driver-specific, only provide the property but don't do anything
33280 with it.
33281
33282 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33283 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
33284
33285commit f26a5b9a994a980ab309bc7f3b4295a4992d81c2
33286Author: Peter Hutterer <peter.hutterer@who-t.net>
33287Date: Mon Jan 31 14:43:01 2011 +1000
33288
33289 Xi: add device node property to known properties.
33290
33291 Since the server has little choice (or even knowledge) of the actual device
33292 node used by the driver, this property is merely provided for
33293 standardisation. It is up to the driver to set it to the appropriate value,
33294 usually a device node in the form of /dev/input/event0 or similar.
33295
33296 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33297 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
33298
33299commit 8d84fd2566f8466b6152724df7eefc73780df093
33300Author: Søren Sandmann Pedersen <ssp@redhat.com>
33301Date: Wed May 25 12:14:05 2011 -0400
33302
33303 Don't call pixman_disable_out_of_bounds_workaround() anymore
33304
33305 Pixman used to have a workaround for a bug in old X servers, and this
33306 function was used to disable that workaround in servers known to be
33307 fixed.
33308
33309 Since 0.22, which the X server depends on, the workaround doesn't
33310 exist anymore, so there is no point disabling it.
33311
33312 Reviewed-by: Cyril Brulebois <kibi at debian.org>
33313 Signed-off-by: Soren Sandmann <sandmann@cs.au.dk>
33314
33315commit b6c7b9b2f39e970cedb6bc1e073f901e28cb0fa3
33316Author: Aaron Plattner <aplattner@nvidia.com>
33317Date: Tue May 24 16:02:42 2011 -0700
33318
33319 randr: check rotated virtual size limits correctly
33320
33321 Commit d1107918d4626268803b54033a07405122278e7f introduced checks to
33322 the RandR path that cause RRSetScreenConfig requests to fail if the
33323 size is too large. Unfortunately, when RandR 1.1 rotation is enabled
33324 it compares the rotated screen dimensions to the unrotated limits,
33325 which causes 90- and 270-degree rotation to fail unless your screen
33326 happens to be square:
33327
33328 X Error of failed request: BadValue (integer parameter out of range for operation)
33329 Major opcode of failed request: 153 (RANDR)
33330 Minor opcode of failed request: 2 (RRSetScreenConfig)
33331 Value in failed request: 0x780
33332 Serial number of failed request: 14
33333 Current serial number in output stream: 14
33334
33335 Fix this by moving the check above the code that swaps the dimensions
33336 based on the rotation.
33337
33338 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
33339 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
33340 Tested-by: Robert Hooker <robert.hooker@canonical.com>
33341 Tested-by: Kent Baxley <kent.baxley@canonical.com>
33342 Signed-off-by: Keith Packard <keithp@keithp.com>
33343
33344commit 4725d6b0dbb6371af4a1b2fbef851dcfe155514a
33345Merge: 1b2d177 b5d8287
33346Author: Keith Packard <keithp@keithp.com>
33347Date: Wed May 25 09:40:51 2011 -0600
33348
33349 Merge remote-tracking branch 'whot/for-keith'
33350
33351commit 1b2d17748f0154da142e9b421d4f6a46e4e5a18c
33352Author: Alan Coopersmith <alan.coopersmith@oracle.com>
33353Date: Sun May 22 13:50:14 2011 -0700
33354
33355 fbbltone.c: Mark bitmasks as unsigned ints
33356
33357 Clears many Sun compiler warnings:
33358 "fbbltone.c", line 491: warning: integer overflow detected: op "<<"
33359 "fbbltone.c", line 491: warning: integer overflow detected: op "<<"
33360 "fbbltone.c", line 491: warning: integer overflow detected: op "<<"
33361 "fbbltone.c", line 491: warning: initializer will be sign-extended: -16777216
33362 "fbbltone.c", line 491: warning: integer overflow detected: op "<<"
33363 "fbbltone.c", line 491: warning: initializer will be sign-extended: -1
33364 "fbbltone.c", line 495: warning: integer overflow detected: op "<<"
33365 "fbbltone.c", line 495: warning: integer overflow detected: op "<<"
33366 "fbbltone.c", line 495: warning: integer overflow detected: op "<<"
33367 "fbbltone.c", line 495: warning: initializer will be sign-extended: -256
33368 "fbbltone.c", line 495: warning: integer overflow detected: op "<<"
33369 "fbbltone.c", line 495: warning: initializer will be sign-extended: -1
33370 "fbbltone.c", line 499: warning: integer overflow detected: op "<<"
33371 "fbbltone.c", line 499: warning: integer overflow detected: op "<<"
33372 "fbbltone.c", line 499: warning: integer overflow detected: op "<<"
33373 "fbbltone.c", line 499: warning: initializer will be sign-extended: -65536
33374 "fbbltone.c", line 499: warning: integer overflow detected: op "<<"
33375 "fbbltone.c", line 499: warning: initializer will be sign-extended: -1
33376
33377 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
33378 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33379 Signed-off-by: Keith Packard <keithp@keithp.com>
33380
33381commit b5d828789c092c994722a896c252212e2b3614ab
33382Author: Simon Thum <simon.thum@gmx.de>
33383Date: Thu Mar 10 21:03:40 2011 +0100
33384
33385 xserver: remove AbsoluteClass, breaking the A(P|B)I
33386
33387 This struct was unused and has been effectively removed in
33388 commit 633b81e8ba09cc6a1ea8b43f323874fda2cf0bde
33389 Refs: xorg-server-1.10.0-133-g633b81e
33390
33391 Remove the remainder, with an ABI bump to 13.0.
33392
33393 Signed-off-by: Simon Thum <simon.thum@gmx.de>
33394 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
33395 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33396
33397commit c05c8640f13708384e77579e7714ca846fa93f5c
33398Author: Oleh Nykyforchyn <oleh.nyk@gmail.com>
33399Date: Thu May 19 09:39:52 2011 +0300
33400
33401 xfree86: Allow "MatchLayout" statements in config files
33402
33403 Usage example (tested on a dual-seat PC):
33404 Section "InputClass"
33405 Identifier "keyboard-all"
33406 MatchIsKeyboard "on"
33407 MatchDevicePath "/dev/input/event*"
33408 MatchLayout "!GeForce|!Matrox"
33409 Driver "evdev"
33410 Option "XkbLayout" "us"
33411 Option "XkbOptions" "terminate:ctrl_alt_bksp"
33412 EndSection
33413
33414 It disables auto keyboard configuration for layouts "GeForce" and "Matrox".
33415 Note that "" in patterns means "no Layout sections found", e.g.
33416 MatchLayout "GeForce|"
33417 is "in layout GeForce or without explicit layout at all".
33418
33419 Signed-off-by: Oleh Nykyforchyn <oleh.nyk@gmail.com>
33420 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
33421 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
33422 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33423
33424commit 83c059f03463f2d7f41e172afe510d1ca9bba8b0
33425Author: Tomas Frydrych <tomas@sleepfive.com>
33426Date: Wed May 18 20:49:52 2011 +0100
33427
33428 Xephyr: fix pointer coordinate translation when screen is rotated
33429
33430 In the Xephyr case the position of the pointer relative toward the
33431 Xephyr window is controlled by the host server without taking into
33432 account rotation of the Xephyr screen. Consequently the pointer coords
33433 must always be translated when the fb is rotated.
33434
33435 Signed-off-by: Tomas Frydrych <tomas@sleepfive.com>
33436 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33437
33438commit b387069fafbe549f2091f364e2aa92af6cc21261
33439Author: Tomas Frydrych <tomas@sleepfive.com>
33440Date: Wed May 18 20:47:52 2011 +0100
33441
33442 Xephyr: added dummy ephyrDeviceCursorCleanup() to avoid crashing
33443
33444 The DeviceCursorCleanup in miPointerSpriteFuncRec can no longer be
33445 NULL it seems.
33446
33447 Signed-off-by: Tomas Frydrych <tomas@sleepfive.com>
33448 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
33449 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33450
33451commit 53ccc54ac71c2655276678e13f0b3d879d691489
33452Author: Colin Harrison <colin.harrison@virgin.net>
33453Date: Mon May 16 19:31:57 2011 +0100
33454
33455 Fix XWin compilation after updates for input API changes
33456
33457 Fix XWin compilation after updates for input API changes in commits
33458 e7150db5 8670c46b and 20fb07f4
33459
33460 Also remove a no longer needed InternalEvent* variable
33461
33462 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
33463 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33464
33465commit 4c4df3ac0d8162c874b85fede12ffbe42f0062c3
33466Author: Jeremy Huddleston <jeremyhu@apple.com>
33467Date: Sat May 14 16:23:59 2011 -0700
33468
33469 input: Don't implicitly define verify_internal_event
33470
33471 Fixes regression introduced by 56901998020b6f443cbaa5eb303100d979e81b22
33472
33473 mieq.c:159:5: error: implicit declaration of function 'verify_internal_event' is invalid in C99 [-Wimplicit-function-declaration,Semantic Issue]
33474 verify_internal_event(e);
33475 ^
33476 1 error generated.
33477
33478 Also includes some other warning cleanups in events.c we're there.
33479
33480 events.c:2198:24: warning: equality comparison with extraneous parentheses [-Wparentheses,Semantic Issue]
33481 else if ((type == MotionNotify))
33482 ~~~~~^~~~~~~~~~~~~~~
33483 events.c:2198:24: note: remove extraneous parentheses around the comparison to silence this warning [Semantic Issue]
33484 else if ((type == MotionNotify))
33485 ~ ^ ~
33486 events.c:2198:24: note: use '=' to turn this equality comparison into an assignment [Semantic Issue]
33487 else if ((type == MotionNotify))
33488 ^~
33489 =
33490 events.c:2487:5: error: implicit declaration of function 'verify_internal_event' is invalid in C99 [-Wimplicit-function-declaration,Semantic Issue]
33491 verify_internal_event(event);
33492 ^
33493 events.c:5909:22: warning: declaration shadows a local variable [-Wshadow,Semantic Issue]
33494 DeviceIntPtr it = inputInfo.devices;
33495 ^
33496 events.c:5893:18: note: previous declaration is here
33497 DeviceIntPtr it = inputInfo.devices;
33498 ^
33499 3 warnings and 1 error generated.
33500
33501 events.c:2836:27: warning: incompatible pointer types passing 'DeviceEvent *' (aka 'struct _DeviceEvent *') to parameter of type
33502 'const InternalEvent *' (aka 'const union _InternalEvent *')
33503 verify_internal_event(ev);
33504 ^~
33505 ../include/inpututils.h:40:56: note: passing argument to parameter 'ev' here
33506 extern void verify_internal_event(const InternalEvent *ev);
33507 ^
33508 1 warning generated.
33509
33510 Found-by: yuffie tinderbox (-Werror=implicit)
33511 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33512 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
33513 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33514
33515commit ba7d1020ac0135ebc8acda4cd57ed48b331e0133
33516Author: Jeremy Huddleston <jeremyhu@apple.com>
33517Date: Sat May 14 16:31:10 2011 -0700
33518
33519 input: Fix format string for verify_internal_event
33520
33521 inpututils.c:577:25: warning: conversion specifies type 'unsigned short' but the argument has type 'unsigned char' [-Wformat,Format String Issue]
33522 ErrorF("%02hx ", *data);
33523 ~~~~^ ~~~~~
33524 %02hhx
33525 1 warning generated.
33526
33527 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33528 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
33529 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33530
33531commit 0de7cec90738a7a5020150309866bb0e23b6f479
33532Author: Peter Hutterer <peter.hutterer@who-t.net>
33533Date: Thu May 19 15:36:29 2011 +1000
33534
33535 xfree86: bump to video ABI 11
33536
33537 We've broken the ABI with some commit and drivers built against ABI 10
33538 happily segfault now.
33539
33540 (The relevant patch is 51f353d0a0d116af16d7d9590cadef6c56328746 which
33541 changed the ATOM typedef from unsigned long to uint32_t, thanks to
33542 Cyril Brulebois <kibi@debian.org> for figuring this out)
33543
33544 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33545 Signed-off-by: Keith Packard <keithp@keithp.com>
33546
33547commit 1fb501ad1521cfedaa5cf3052d45a924ef1866cf
33548Author: Jeremy Huddleston <jeremyhu@apple.com>
33549Date: Sun May 22 10:18:36 2011 -0700
33550
33551 XQuartz: Don't crash if CG increases our display resolution
33552
33553 miPaintWindow would cause fbFill() to overwrite pScreen's pixmap which was
33554 sized for the old resolution.
33555
33556 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33557
33558commit 60af79e35ee8546a99d15a1358aac3deabfa22be
33559Author: Jeremy Huddleston <jeremyhu@apple.com>
33560Date: Sun May 22 09:32:57 2011 -0700
33561
33562 XQuartz: RandR: Don't crash if X11 is launched while there are no attached displays
33563
33564 If CG reports no displays when launching, we could crash in RandR. Instead, just
33565 provide a fake 800x600 display until we are notified about displays being attached.
33566
33567 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33568
33569commit f25ca898c54cb88c7886005fc75a53762c42710b
33570Author: Jeremy Huddleston <jeremyhu@apple.com>
33571Date: Thu May 19 14:39:22 2011 -0700
33572
33573 XQuartz: Mark functions _X_NORETURN
33574
33575 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33576
33577commit 7413886d650aef492ecbfdc4298c2d92d9af5f87
33578Author: Jeremy Huddleston <jeremyhu@apple.com>
33579Date: Thu May 19 14:35:56 2011 -0700
33580
33581 XQuartz: Silence clang warnings about shadow declarations
33582
33583 X11Application.m:1272:26: warning: declaration shadows a local variable [-Wshadow,Semantic Issue]
33584 xp_error e;
33585 ^
33586 X11Application.m:1098:36: note: previous declaration is here
33587 - (void) sendX11NSEvent:(NSEvent *)e {
33588 ^
33589 1 warning generated.
33590
33591 bundle-main.c:648:36: warning: declaration shadows a local variable [-Wshadow,Semantic Issue]
33592 int max_files, i;
33593 ^
33594 bundle-main.c:594:9: note: previous declaration is here
33595 int i;
33596 ^
33597 1 warning generated.
33598
33599 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33600
33601commit 96ac4e61f4618332d95d1fd0e4799dd82844f90f
33602Author: Jeremy Huddleston <jeremyhu@apple.com>
33603Date: Thu May 19 14:34:39 2011 -0700
33604
33605 XQuartz: Update DEBUG_LOG to report to ASL
33606
33607 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33608
33609commit 384eb45b944a4386eae74a5503423c13b5f2a659
33610Author: Jeremy Huddleston <jeremyhu@apple.com>
33611Date: Sat May 14 18:33:57 2011 -0700
33612
33613 XQuartz: RandR: Avoid over-releasing if we are unable to determine the current display mode.
33614
33615 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33616
33617commit 25191648b8db87735a99243697f73036255c1eb6
33618Author: Jeremy Huddleston <jeremyhu@apple.com>
33619Date: Sat May 14 16:11:32 2011 -0700
33620
33621 XQuartz: Don't call mieqEnqueue during server shutdown
33622
33623 Found-by: GuardMalloc
33624 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33625
33626commit fd5f630bc41de73be3b1d26bf9ce3e9ef1badc3b
33627Author: Jeremy Huddleston <jeremyhu@apple.com>
33628Date: Sat May 14 14:09:45 2011 -0700
33629
33630 XQuartz: Fix an array-index-out-of-bounds crasher
33631
33632 Found-by: GuardMalloc
33633 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33634
33635commit 0e7f61d72c4a929319e57c9b5b777e9413c23051
33636Author: Gaetan Nadon <memsize@videotron.ca>
33637Date: Sat May 14 13:30:20 2011 -0400
33638
33639 doc: use devbook.am for developers documentation
33640
33641 Relocate the docs under the doc subdir.
33642 Remove redundant xml subdir.
33643 The xmlrules set of makefiles are no longer used.
33644
33645 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33646 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33647 Signed-off-by: Keith Packard <keithp@keithp.com>
33648
33649commit 96c65be39b98cfe035be0a3988dce519549249f3
33650Author: Gaetan Nadon <memsize@videotron.ca>
33651Date: Sat May 14 13:30:19 2011 -0400
33652
33653 doc: relocate xserver.ent in the package root directory
33654
33655 This is the appropriate location for reusable configuration and/or
33656 makefile artifacts as opposed to picking one of the subdirs where
33657 it is used.
33658
33659 It shields them from future doc reorg as every subdir
33660 will refer to the root package which never changes location.
33661
33662 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33663 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33664 Signed-off-by: Keith Packard <keithp@keithp.com>
33665
33666commit dbe26634e4f12eba905540666783c9879aa3a49a
33667Author: Gaetan Nadon <memsize@videotron.ca>
33668Date: Sat May 14 13:30:18 2011 -0400
33669
33670 doc: add external doc references support to Xserver-DTrace
33671
33672 This makefile is used to build the 60+ docbooks in xorg.
33673 Dtrace is a user document and should be located under doc subdir.
33674
33675 This user document can now refer to external user/specs docs
33676 or can be referred to by such documnets.
33677
33678 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33679 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33680 Signed-off-by: Keith Packard <keithp@keithp.com>
33681
33682commit 221507e3bf6fcaa4a4c2e1bc264a1f806b4362c0
33683Author: Gaetan Nadon <memsize@videotron.ca>
33684Date: Thu Apr 28 21:16:13 2011 -0400
33685
33686 man: relocate manual pages in the man subdir outside doc
33687
33688 The convention is to have the manual pages in a man subdir
33689 which is not under a doc dir. The doc dir contains users docs.
33690 This will move man pages out of the way for upcoming DocBook patches.
33691
33692 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33693 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33694 Signed-off-by: Keith Packard <keithp@keithp.com>
33695
33696commit 9129beb507642e2414ef1f90d650572325d8c2dc
33697Author: Gaetan Nadon <memsize@videotron.ca>
33698Date: Thu Apr 28 21:16:12 2011 -0400
33699
33700 dmx: fix warning for doxygen explicit links
33701
33702 Explicit links to functions in another file are not supported.
33703
33704 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33705 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33706 Signed-off-by: Keith Packard <keithp@keithp.com>
33707
33708commit 0d8a5766a39320dedd34a215fc6655f9ac19e8bc
33709Author: Gaetan Nadon <memsize@videotron.ca>
33710Date: Thu Apr 28 21:16:11 2011 -0400
33711
33712 dmx: modernize doxygen generation.
33713
33714 The configuration and stylesheet were very old.
33715 The stylesheet is not checked-in, use the generated one.
33716 The header is not checked-in, use the generated one.
33717 Add datetime and projectname in default footer.
33718 Developer documentation is not installed and not included in tarball.
33719
33720 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33721 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33722 Signed-off-by: Keith Packard <keithp@keithp.com>
33723
33724commit 622ebb88157b0ed99355419767685dd70ce0f16c
33725Author: Gaetan Nadon <memsize@videotron.ca>
33726Date: Thu Apr 28 21:16:10 2011 -0400
33727
33728 dmx: split DocBook/XML and Doxygen makefile targets
33729
33730 Maintaining either requires full knowledge of both.
33731 It's not obvious one has to check the usage of global variables
33732 in devbook.am when maintaining doxygen target. Or vice-versa.
33733
33734 Being in their respective directory, one less thing to worry about.
33735
33736 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33737 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33738 Signed-off-by: Keith Packard <keithp@keithp.com>
33739
33740commit 595460c397a543fe512becd3c432de2a0a68d187
33741Author: Gaetan Nadon <memsize@videotron.ca>
33742Date: Thu Apr 28 21:16:09 2011 -0400
33743
33744 dmx/doc: use common makefile for developers documentation
33745
33746 The user/specs docs now have external references support.
33747 Developers doc are not installed so they do not participate.
33748 However, using a similar makefile shared amongst developers
33749 document reduces maintenance and is forward looking.
33750
33751 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33752 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33753 Signed-off-by: Keith Packard <keithp@keithp.com>
33754
33755commit 04011b0bc3b5b59efa0d981866349c4ad0868d8f
33756Author: Gaetan Nadon <memsize@videotron.ca>
33757Date: Thu Apr 28 21:16:08 2011 -0400
33758
33759 doc: use common makefile for developers documentation
33760
33761 The user/specs docs now have external references support.
33762 Developers doc are not installed so they do not participate.
33763 However, using a similar makefile shared amongst developers
33764 document reduces maintenance and is forward looking.
33765
33766 Man pages being out of here, reorg developers docs under the same roof.
33767 Drop the obsolete sgml subdir.
33768
33769 Reviewed-by Jeremy Huddleston <jeremyhu@apple.com>
33770 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
33771 Signed-off-by: Keith Packard <keithp@keithp.com>
33772
33773commit 531869448d07e00ae241120b59f3aaaa5709d59c
33774Author: Ville Syrjälä <ville.syrjala@nokia.com>
33775Date: Fri May 6 18:18:15 2011 +0300
33776
33777 dri2: Don't send so many needless invalidate events
33778
33779 Only send invalidate events for drawables if some client has requested
33780 some buffers.
33781
33782 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
33783 Reviewed-by: Michel Dänzer <michel@daenzer.net>
33784 Signed-off-by: Keith Packard <keithp@keithp.com>
33785
33786commit ba5540221f2a46133371b4ff0d527b1a0a1443b1
33787Merge: 6347a0b 728d0bf
33788Author: Keith Packard <keithp@keithp.com>
33789Date: Fri May 13 13:59:36 2011 -0700
33790
33791 Merge remote-tracking branch 'whot/for-keith'
33792
33793commit 6347a0b802812bb185ada1bf0951add306935184
33794Merge: 4d02c53 f144fb7
33795Author: Keith Packard <keithp@keithp.com>
33796Date: Fri May 13 13:54:29 2011 -0700
33797
33798 Merge remote-tracking branch 'jeremyhu/master'
33799
33800commit 4d02c5397114ac4d15e794908f0708427e258261
33801Merge: 043c175 eac37f3
33802Author: Keith Packard <keithp@keithp.com>
33803Date: Fri May 13 13:52:18 2011 -0700
33804
33805 Merge remote-tracking branch 'vsyrjala/composite_validatetree_2'
33806
33807commit f144fb771f2e0665677aa88d08778899f666a1d5
33808Author: Jeremy Huddleston <jeremyhu@apple.com>
33809Date: Fri May 13 09:10:42 2011 -0700
33810
33811 XQuartz: Don't circumvent NDEBUG
33812
33813 If someone wants to turn off asserts with NDEBUG, let them.
33814
33815 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33816
33817commit f46835a09662f71f2cb90540609c34561d8af87f
33818Author: Jeremy Huddleston <jeremyhu@apple.com>
33819Date: Wed May 11 15:23:17 2011 -0700
33820
33821 XQuartz: Redirect stdout/stderr to asl
33822
33823 In order to improve logging in XQuartz, stdout and stderr should be redirected
33824 to asl (syslog).
33825
33826 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33827
33828commit 5bc05d96f883add324bab338ffb9b9409cf45b2c
33829Author: Jeremy Huddleston <jeremyhu@apple.com>
33830Date: Thu May 12 15:45:29 2011 -0700
33831
33832 XQuartz: Add a LOGGING section to our man page
33833
33834 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33835
33836commit c18b7165f971cbf6b634ccf670eb4aa9840eb9e4
33837Author: Jeremy Huddleston <jeremyhu@apple.com>
33838Date: Thu May 12 18:29:23 2011 -0700
33839
33840 XQuartz: stub: Log directly to ASL rather than stdout/stderr
33841
33842 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33843
33844commit d694601591d2a86a437c5150015d9889e2fe16fc
33845Author: Jeremy Huddleston <jeremyhu@apple.com>
33846Date: Thu May 12 15:15:40 2011 -0700
33847
33848 Fix a typo: laucnd instead of launchd
33849
33850 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33851
33852commit 728d0bf20ed8e2612b100fca6526705fa6e1eef4
33853Author: Peter Hutterer <peter.hutterer@who-t.net>
33854Date: Wed May 11 13:35:39 2011 +1000
33855
33856 dix: replace CORE_EVENT and XI2_EVENT macros with inline functions.
33857
33858 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33859 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33860
33861commit ffd4874798ba54f86acac75779a15b4babeaa5f3
33862Author: Peter Hutterer <peter.hutterer@who-t.net>
33863Date: Wed May 11 12:20:50 2011 +1000
33864
33865 include: add version_compare helper function
33866
33867 Compare two version numbers in the major.minor form.
33868 Switch the few users of manual version switching over to the new function.
33869
33870 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33871 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33872
33873commit c4f9c3a07dbb05b81c8e2193a083102f710ebb27
33874Author: Peter Hutterer <peter.hutterer@who-t.net>
33875Date: Wed May 11 11:43:16 2011 +1000
33876
33877 dix: use a tmp variable instead of multiple rClient(other).
33878
33879 no functional changes.
33880
33881 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33882 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33883
33884commit dc45d5816dd65168645f0017394eebfc5599d698
33885Author: Peter Hutterer <peter.hutterer@who-t.net>
33886Date: Thu Apr 14 15:59:08 2011 +1000
33887
33888 Xi: split DeviceStateNotify delivery into a separate function
33889
33890 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33891 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33892
33893commit 5bcc22757e6e1f24ee2bfec65f68a5f567300532
33894Author: Peter Hutterer <peter.hutterer@who-t.net>
33895Date: Thu Apr 14 15:43:56 2011 +1000
33896
33897 dix: return deliveries from DeliverGrabbedEvent
33898
33899 This isn't currently used by any of the callers but it will likely be in the
33900 future.
33901
33902 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33903 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33904
33905commit 2054ca73060a20b5a3025e8d5ef68182149484d3
33906Author: Peter Hutterer <peter.hutterer@who-t.net>
33907Date: Thu Apr 14 15:09:39 2011 +1000
33908
33909 dix: move the grab activation condition into a if block.
33910
33911 Rather than 3 conditions with if (deliveries && ...), have one block with
33912 the three in them.
33913 No functional changes.
33914
33915 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33916 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33917
33918commit 236ed6f50675dc0303a505ac6f0418c515438fe1
33919Author: Peter Hutterer <peter.hutterer@who-t.net>
33920Date: Tue Apr 12 14:04:37 2011 +1000
33921
33922 dix: split out client delivery from DeliverEventsToWindow
33923
33924 No real functional changes, this is just for improved readability.
33925
33926 DeliverEventsToWindow used to return an int to specify the number of
33927 deliveries (or rejected deliveries if negative). The number wasn't used by
33928 any caller other than for > 0 comparison.
33929
33930 This patch also changes the return value to be -1 or 1 even in case of
33931 multiple deliveries/rejections. The comment was updated accordingly.
33932
33933 A future patch should probably use the enum EventDeliveryState for
33934 DeliverEventsToWindow.
33935
33936 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33937 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33938
33939commit 536ca28f1b0b4d8715a41b8acc5f30364c833f9b
33940Author: Peter Hutterer <peter.hutterer@who-t.net>
33941Date: Tue Apr 12 13:44:30 2011 +1000
33942
33943 dix: split out window owner event delivery from DeliverEventsToWindow
33944
33945 No functional changes, just for readability.
33946
33947 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33948 Reviewed-by: Jamey Sharp <jamey@minilop.net>
33949
33950commit a311a03135f9734aa4f7dc0f2fa0e4c91768343b
33951Author: Jamey Sharp <jamey@minilop.net>
33952Date: Wed May 11 14:08:28 2011 -0700
33953
33954 dix: split implicit grab activation into a separate function.
33955
33956 I'm not sure I like splitting the check for button-press event from the
33957 code which makes assumptions about that check. How about replacing
33958 patches 3 and 4 with this patch instead?
33959
33960 Signed-off-by: Jamey Sharp <jamey@minilop.net>
33961 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
33962 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33963
33964commit a558a66d249cd51d3d675b20cc3a76651fd87930
33965Author: Jeremy Huddleston <jeremyhu@apple.com>
33966Date: Wed May 11 17:27:52 2011 -0700
33967
33968 configure.ac: XQuartz: Fix support for the deprecated --with-launchd-id-prefix
33969
33970 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33971
33972commit c89a6f824eaf647d2b182f79fbd78f5bd1c3a27f
33973Author: Jeremy Huddleston <jeremyhu@apple.com>
33974Date: Tue May 10 23:42:46 2011 -0700
33975
33976 XQuartz: Don't call into CoreFoundation after fork() and before exec()
33977
33978 After fork()ing, we should just limit ourselves to setting up
33979 the environment, file descriptors, and exec()ing.
33980
33981 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
33982
33983commit c8674a328c68f03de6e4fad7790a595cdfc18736
33984Author: Peter Hutterer <peter.hutterer@who-t.net>
33985Date: Thu Apr 14 16:45:56 2011 +1000
33986
33987 dix: replace unneded goto with break.
33988
33989 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
33990 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
33991
33992commit 20fb07f436f7d4a0f330b2067a93a5a4829fccf5
33993Author: Peter Hutterer <peter.hutterer@who-t.net>
33994Date: Fri Apr 15 10:07:10 2011 +1000
33995
33996 input: remove DDX event list handling
33997
33998 The current approach to event posting required the DDX to request the event
33999 list (allocated by the DIX) and then pass that list into QueuePointerEvent
34000 and friends.
34001
34002 Remove this step and use the DIX event list directly. This means that
34003 QueuePointerEvent is not reentrant but it wasn't before anyway.
34004
34005 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
34006 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34007
34008commit 8670c46bdfdade64e63119d2ebbd5ef63b6fa2c3
34009Author: Peter Hutterer <peter.hutterer@who-t.net>
34010Date: Thu Apr 14 22:05:41 2011 +1000
34011
34012 input: replace EventListPtr with InternalEvent array
34013
34014 EventListPtr is a relic from pre-1.6, when we had protocol events in the
34015 event queue and thus events of varying size.
34016
34017 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
34018 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34019
34020commit e7150db5350bc2113ff4126019b489847a4dc217
34021Author: Peter Hutterer <peter.hutterer@who-t.net>
34022Date: Mon Apr 11 15:48:15 2011 +1000
34023
34024 input: Provide Queue{Button|Keyboard|Proximity}Event helpers
34025
34026 Don't require every caller to use GPE + mieqEnqueue, provide matching
34027 Queue...Event functions instead.
34028
34029 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
34030 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34031
34032commit 1b8593a6c12315b1071a4fa586151e12f46458f5
34033Author: Peter Hutterer <peter.hutterer@who-t.net>
34034Date: Mon May 9 15:13:17 2011 +1000
34035
34036 xfree86: print the device ID to the log when adding a device.
34037
34038 Sometimes the name isn't enough, it's handy to see the device ID's from the
34039 log file.
34040
34041 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
34042
34043commit 3231962db826f5efd431596a309c96e907a191d1
34044Author: Daniel Stone <daniel@fooishbar.org>
34045Date: Tue May 3 03:20:23 2011 +0100
34046
34047 XKB: Fix sense inversion for core MapNotify events
34048
34049 Due to an unfortunate sense inversion incident while switching from a
34050 if (foo) { ... } to if (!foo) continue; style in f06a9d, we punished any
34051 client who attempted to use XKB to restrict the MapNotify events they
34052 wanted by sending them exactly the events they _didn't_ want, and
34053 nothing else.
34054
34055 NewKeyboardNotifies (coming from a client setting the map with an XKB
34056 request, when switching between master devices, etc) weren't affected,
34057 but this would impact anyone using xmodmap-style core requests. Could
34058 explain a fair bit.
34059
34060 Clarified the comments while I was at it.
34061
34062 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34063 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34064
34065commit 460a377ef2d645d9ae38a8356fb248ccc47bed4a
34066Author: Daniel Stone <daniel@fooishbar.org>
34067Date: Tue May 3 03:07:50 2011 +0100
34068
34069 XKB: Send XKB events for all devices to all clients
34070
34071 We were using XIShouldNotify(client, device) as a test for whether or
34072 not to send XKB map/state/etc changed events, which limits it to only
34073 sending events for the current ClientPointer/ClientKeyboard for that
34074 client. While this makes perfect sense for core events (e.g.
34075 MappingNotify), XKB events carry a device ID, so are safe to send to all
34076 clients for all devices.
34077
34078 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34079 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34080
34081commit a79d4544fee806a25447d0147535ebc5a1cae6b9
34082Author: Daniel Stone <daniel@fooishbar.org>
34083Date: Tue May 3 03:03:06 2011 +0100
34084
34085 XKB: Send NewKeyboardNotify for dev before its master/slaves
34086
34087 When we change the keymap on a device, send the NewKeyboardNotify for
34088 that device before we copy the keymap to and notify for its attached
34089 master/slave devices.
34090
34091 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34092 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34093
34094commit c7634498d4cd42c8571805122224dc2d0e44a585
34095Author: Daniel Stone <daniel@fooishbar.org>
34096Date: Tue May 3 02:59:53 2011 +0100
34097
34098 XKB: Remove duplicate keymap-copying loop
34099
34100 Previously we had:
34101 foreach (device + slaves of device) {
34102 XkbCopyDeviceKeymap(i, device);
34103 [...]
34104 }
34105 if (device was last slave of its MD) {
34106 XkbCopyDeviceKeymap(master, device);
34107 }
34108 and now:
34109 foreach (device + slaves of device + MD if device was last slave) {
34110 XkbCopyDeviceKeymap(i, device);
34111 [...]
34112 }
34113
34114 As an extra bonus, when changing the keymap on a slave device, we now
34115 ensure the LED info on the master is kept in sync.
34116
34117 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34118 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34119
34120commit b8540d18c7d3a0f93f9e2565a410986eddedcecb
34121Author: Daniel Stone <daniel@fooishbar.org>
34122Date: Tue May 3 02:50:48 2011 +0100
34123
34124 XKB: Simplify a loop in ProcXkbGetKbdByName
34125
34126 Replace:
34127 for (stuff; things; etc) {
34128 if (misc || other) {
34129 [...]
34130 }
34131 }
34132 with:
34133 for (stuff; things; etc) {
34134 if (!misc && !other)
34135 continue;
34136 [...]
34137 }
34138
34139 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34140 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34141
34142commit 72b6639c83df74767094a5e0c2861fdc5ea03ecd
34143Author: Daniel Stone <daniel@fooishbar.org>
34144Date: Tue May 3 02:39:55 2011 +0100
34145
34146 XKB: Don't send unnecessary NewKeyboardNotifies
34147
34148 In the XKB GetKeyboardByName handler, we had the following pseudocode:
34149 if (device was last slave of its MD) {
34150 XkbCopyDeviceKeymap(master, slave);
34151 XkbSendNewKeyboardNotify(slave, &notify);
34152 }
34153
34154 Even if the SendNewKeyboardNotify line nominated the correct device,
34155 which it didn't, it's unnecessary as XkbCopyDeviceKeymap already sends a
34156 NewKeyboardNotify on the destination device.
34157
34158 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34159 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34160
34161commit 70cef8d8baf058bacaff87ef49e3851628269597
34162Author: Daniel Stone <daniel@fooishbar.org>
34163Date: Wed Jan 19 20:42:10 2011 +0000
34164
34165 Input: Simplify CheckPassiveGrabsOnWindow loop
34166
34167 Instead of a mega never-ending if branch with no else, just continue
34168 to the next iteration of the loop if the conditions aren't met - pretty
34169 much entirely reindentation.
34170
34171 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34172 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34173
34174commit 00ba884556c675b2b25e116f5ab4eb4590b6dd56
34175Author: Daniel Stone <daniel@fooishbar.org>
34176Date: Wed Jan 19 20:38:44 2011 +0000
34177
34178 Input: Make CheckPassiveGrabsOnWindow take InternalEvent
34179
34180 Previously, it only took DeviceEvents, but it would be much more useful
34181 if it took InternalEvents. Any event that activates a grab must still
34182 be a DeviceEvent, so put in a check to enforce this.
34183
34184 Change all callers to make the appropriate casts.
34185
34186 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34187 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34188
34189commit 56901998020b6f443cbaa5eb303100d979e81b22
34190Author: Peter Hutterer <peter.hutterer@who-t.net>
34191Date: Thu May 5 08:48:19 2011 +1000
34192
34193 input: change CHECKEVENT macro to verify_internal_event function
34194
34195 The macro is sufficient if called during a development cycle, but not
34196 sufficient information when triggered by a user (e.g.
34197 https://bugzilla.redhat.com/show_bug.cgi?id=688693).
34198
34199 Expand what this does to print the event content and a backtrace, so at
34200 least we know where we're coming from. Only the first 32 bytes are printed
34201 since if something goes wrong, the event we have is almost certainly an
34202 xEvent or xError, both restricted to 32 bytes.
34203
34204 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
34205 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
34206
34207commit b47d2e43eb2cb3817c995d1f7e58500fb40efa2b
34208Author: Jeremy Huddleston <jeremyhu@apple.com>
34209Date: Thu May 5 09:04:41 2011 -0700
34210
34211 XQuartz: Make a copy of args for our crash reporter vsnprintf
34212
34213 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34214
34215commit bf2059b07a97e5e579c13c2c9d49707093427dc2
34216Author: Peter Hutterer <peter.hutterer@who-t.net>
34217Date: Fri Apr 15 14:06:20 2011 +1000
34218
34219 input: Only release SD buttons for explicit floating/reattachment (#36146)
34220
34221 Grabbing an SD device temporary floats the device but we must not release
34222 the buttons. Introduced in
34223
34224 commit 9d23459415b84606ee4f38bb2d19054c432c8552
34225 Author: Peter Hutterer <peter.hutterer@who-t.net>
34226 Date: Fri Feb 25 11:08:19 2011 +1000
34227
34228 dix: release all buttons and keys before reattaching a device (#34182)
34229
34230 X.Org Bug 36146 <http://bugs.freedesktop.org/show_bug.cgi?id=36146>
34231
34232 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
34233 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34234
34235commit eac37f32b85b631d94ee3ba11fa65b9d2cb72c38
34236Author: Ville Syrjälä <ville.syrjala@nokia.com>
34237Date: Wed Apr 13 21:46:20 2011 +0300
34238
34239 composite: Recompute clipping when changing between manual and automatic redirection
34240
34241 Call compMarkWindows() when changing between manual and automatic
34242 redirection modes. Otherwise the window clipping won't be recomputed
34243 correctly.
34244
34245 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
34246
34247commit f2001b0f6dffa0a8f05def4a86ea37c4c91db724
34248Author: Ville Syrjälä <ville.syrjala@nokia.com>
34249Date: Wed Apr 13 21:45:43 2011 +0300
34250
34251 composite: Fix pWin->redirectDraw when changing between manual and automatic redirection
34252
34253 compAllowPixmap() is not called when changing between manual and
34254 automatic redirection modes. That means pWin->redirectDraw is left
34255 with an incorrect value, and miComputeClips() gets confused whether
34256 the window is supposed to be treated as transparent or not. Fix
34257 the issue by updating pWin->redirectDraw in compCheckRedirect()
34258 even when not calling compAllocPixmap().
34259
34260 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
34261
34262commit 74663e61528346aeea9c11908b6980b51dcaeb68
34263Author: Ville Syrjälä <ville.syrjala@nokia.com>
34264Date: Mon Dec 20 16:37:24 2010 +0200
34265
34266 composite: Copy the window contents back from the pixmap
34267
34268 Since extra expose events are no longer generated during window
34269 unredirection, the window contents must be preserved by the server.
34270 So copy the window contents back from the pixmap. The copy can only
34271 be done after the clips have been recomputed, so delay the copy and
34272 the pixmap destruction until ValidateTree is done. Window borders are
34273 restored by HandleExposures and thus don't need to be copied back.
34274
34275 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
34276 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
34277
34278commit 193ecc8b453b22b3e60248b9354c768dbd405598
34279Author: Ville Syrjälä <ville.syrjala@nokia.com>
34280Date: Fri Dec 17 23:46:34 2010 +0200
34281
34282 composite: Get rid of the internal UnmapWindow+MapWindow cycle
34283
34284 Eliminate the internal MapWindow+UnmapWindow cycle around window
34285 redirection changes. Instead do the work in a single pass by marking
34286 the afected windows and calling ValidateTree and HandleExposures
34287 directly. This gets rid of unnecessary expose events, and invalid
34288 ClipNotify calls during rediredction changes. Now ClipNotify will only
34289 get called with the final clip values, and expose events are only sent
34290 to areas that actually got exposed.
34291
34292 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
34293 Reviewed-by: Adam Jackson <ajax@redhat.com>
34294 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
34295
34296commit a6ae91746212203a19450ac955fbb7abffff1ed3
34297Author: Ville Syrjälä <ville.syrjala@nokia.com>
34298Date: Mon Dec 20 16:30:52 2010 +0200
34299
34300 composite: Initialize borderClip with current values
34301
34302 ValidateTree needs a valid borderClip so initialize the parent
34303 constrained border clip with the window's current borderClip
34304 in compRedirectWindow.
34305
34306 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
34307 Reviewed-by: Adam Jackson <ajax@redhat.com>
34308 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
34309
34310commit ce9eff9e4c30deb16f059ed84b436d37da45d5d5
34311Author: Ville Syrjälä <ville.syrjala@nokia.com>
34312Date: Mon Dec 20 16:33:36 2010 +0200
34313
34314 composite: Call ValidateGC after ChangeGC
34315
34316 ChangeGC changes the GC, so ValidateGC should be called after it, not
34317 before.
34318
34319 Also pass NullClient instead of serverClient to ChangeGC() since we
34320 know the changed values to be valid, and setting
34321 serverClient->errorValue seems pointless anyway.
34322
34323 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
34324 Reviewed-by: Keith Packard <keithp@keithp.com>
34325 Reviewed-by: Adam Jackson <ajax@redhat.com>
34326 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
34327
34328commit 3ac220d6cc386c9ab9995e5b83463c40330e18c2
34329Author: Jeremy Huddleston <jeremyhu@apple.com>
34330Date: Tue May 3 10:54:09 2011 -0700
34331
34332 XQuartz: prefs_copy_url and prefs_get_copy return retained objects
34333
34334 No functional change. This just annotates the return policy.
34335
34336 Found-by: clang static analyzer
34337 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34338
34339commit bac8d12555dd44b2139be0696629910cd4d8c782
34340Author: Jeremy Huddleston <jeremyhu@apple.com>
34341Date: Tue May 3 10:36:19 2011 -0700
34342
34343 XQuartz: Ensure that {CF,NS}_RETURNS{,_NOT}_RETAINED are defined
34344
34345 These will be used in subsequent patches to denote proper retain counts in XQuartz
34346
34347 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34348
34349commit 043c1758652259fd12b88ae37720fe6e93eda76b
34350Author: Alan Coopersmith <alan.coopersmith@oracle.com>
34351Date: Tue Apr 19 19:02:54 2011 -0700
34352
34353 Clean up memory better when GetVisualInfo fails in ProcDbeGetVisualInfo
34354
34355 Use calloc to initialize pScrVisInfo array so we don't have to check
34356 which ones were already initialized when freeing them all.
34357
34358 On failure, set rc if necessary, and jump to code at end that already
34359 frees all the necessary allocations and return rc.
34360
34361 Fixes parfait reported error:
34362 Error: Memory leak (CWE 401)
34363 Memory leak of pointer 'pScrVisInfo' allocated with malloc((count * 16))
34364 at line 724 of dbe/dbe.c in function 'ProcDbeGetVisualInfo'.
34365 'pScrVisInfo' allocated at line 693 with malloc((count * 16)).
34366 pScrVisInfo leaks when rc != 0 at line 710
34367 and j >= i at line 716.
34368
34369 [ This bug was found by the Parfait 0.3.7 bug checking tool.
34370 For more information see http://labs.oracle.com/projects/parfait/ ]
34371
34372 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
34373 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34374
34375commit dadb0791ebfd05cd3bb82d4addf0fbc21aad6fbb
34376Author: Alan Coopersmith <alan.coopersmith@oracle.com>
34377Date: Tue Apr 19 18:53:22 2011 -0700
34378
34379 Only free pContext once when AddResource fails in ProcRecordCreateContext
34380
34381 Since RecordDeleteContext frees its argument, don't fall through to free
34382 it again.
34383
34384 Error: Double free (CWE 415)
34385 Double free of pointer 'malloc(1072)' defined by malloc
34386 at line 1964 of record/record.c in function 'ProcRecordCreateContext'.
34387 Previously freed at line 1960 with RecordDeleteContext.
34388 'malloc(1072)' was allocated at line 1926 with malloc.
34389
34390 [ This bug was found by the Parfait 0.3.7 bug checking tool.
34391 For more information see http://labs.oracle.com/projects/parfait/ ]
34392
34393 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
34394 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34395
34396commit 73de54210446e9eca81b96ea6775ee5ea1a31d75
34397Author: Daniel Stone <daniel@fooishbar.org>
34398Date: Mon Jan 24 08:40:10 2011 +1000
34399
34400 Input: Fix event size confusion in CheckPassiveGrabsOnWindow
34401
34402 We were just storing a DeviceEvent, but allocating enough space for an
34403 InternalEvent.
34404
34405 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34406 Reviewed-by: Julien Cristau <jcristau@debian.org>
34407 Reviewed-by: Adam Jackson <ajax@redhat.com>
34408 Reviewed-by: Cyril Brulebois <kibi@debian.org>
34409
34410commit 706326491011be8cecb9b56c06f7241b7cbd425f
34411Author: Daniel Stone <daniel@fooishbar.org>
34412Date: Tue Jan 18 20:16:36 2011 +0000
34413
34414 Input: Add DeepestSpriteWin function
34415
34416 Does what it says on the box: returns the deepest child window in a
34417 given sprite's trace.
34418
34419 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
34420 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
34421 Reviewed-by: Keith Packard <keithp@keithp.com>
34422
34423commit 65b54548dce80c8e8ff5ff91fc4f0659e9b2d921
34424Author: Chase Douglas <chase.douglas@canonical.com>
34425Date: Tue Jan 18 20:08:09 2011 +0000
34426
34427 Input: Pass co-ordinates by reference to transformAbsolute
34428
34429 With the upcoming XI 2.1 touch work, the co-ordinate values will need to
34430 be passed by reference, rather than modified in-place.
34431
34432 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
34433 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
34434
34435commit a14a0c711397ff7ca0220946010300fc1b2a6e67
34436Author: Alan Coopersmith <alan.coopersmith@oracle.com>
34437Date: Fri Apr 22 22:19:39 2011 -0700
34438
34439 Move event filter initializer out of the structure itself
34440
34441 When kept in the structure, it causes the entire MAXDEVICES * 128 masks
34442 to be stored in the data segment and loaded from the file, and also leads
34443 to worries about later generations inheriting changes across server reset.
34444
34445 text data bss dec hex filename
34446 Before: 91837 20528 32 112397 1b70d .libs/events.o
34447 After: 92277 48 20512 112837 1b8c5 .libs/events.o
34448 Before: 3013384 122696 163156 3299236 3257a4 Xorg
34449 After: 3013832 102216 183636 3299684 325964 Xorg
34450
34451 File size before: 4337008 Xorg
34452 File size after: 4316568 Xorg
34453
34454 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
34455 Reviewed-by: Jamey Sharp <jamey@minilop.net>
34456 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34457 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
34458
34459commit 8d229c4cf9e5bde78373ef3dd32708817ac97152
34460Author: Alan Coopersmith <alan.coopersmith@oracle.com>
34461Date: Fri Mar 11 15:43:14 2011 -0800
34462
34463 Make xorg.conf.example rule compatible with Solaris make
34464
34465 Solaris make won't substitute $< in explicit rules, only implicit ones
34466
34467 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
34468 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
34469
34470commit c6029246ddcf5ead3884dbf2ab4f57e0c3586e93
34471Author: Alan Coopersmith <alan.coopersmith@oracle.com>
34472Date: Wed Feb 2 17:00:25 2011 -0800
34473
34474 Move Xinput server API documentation from libXi to doc/xml
34475
34476 Documentation is quite out of date but still fits better here than
34477 in the client library.
34478
34479 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
34480 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
34481
34482commit 0fc7ec6dd504aa03e1a1b18c60942c0c8c8b701b
34483Author: Jeremy Huddleston <jeremyhu@apple.com>
34484Date: Fri Apr 29 11:06:18 2011 -0700
34485
34486 XQuartz: Fix incorrect typedefs with XPLUGIN_VERSION < 4
34487
34488 Ok, this time for sure... how many brown bags can I fit over my face?
34489
34490 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34491
34492commit 29d471663e4414a3cdf154fd032d74381a921ae0
34493Author: Jeremy Huddleston <jeremyhu@apple.com>
34494Date: Sat Apr 23 18:09:11 2011 -0700
34495
34496 XQuartz: Use a rwlock instead of a mutex to protect window_hash in the pthread case
34497
34498 Concurrent reads are acceptable, so using an rwlock should be better.
34499
34500 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34501
34502commit 5cb31cd0cbf83fff5f17a475e7b0e45246b19bf3
34503Merge: 0f284f0 4d8735d
34504Author: Keith Packard <keithp@keithp.com>
34505Date: Fri Apr 29 09:59:49 2011 -0700
34506
34507 Merge remote-tracking branch 'jturney/remove-opengl-spec-download'
34508
34509commit 4d8735d388c3c6af27ef3d6a645e38e556f392f4
34510Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34511Date: Wed Apr 6 16:26:45 2011 +0100
34512
34513 hw/xwin: wglext.h should be provided by w32api, rather than downloaded
34514
34515 wglext.h should be provided by the w32api package, rather than downloaded.
34516 if it's not, do 'wget -P /usr/include/w32api/GL http://www.opengl.org/registry/api/wglext.h'
34517
34518 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34519 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34520 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
34521
34522commit a50878682cda541766c830e49ab406c78b38d903
34523Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34524Date: Wed Apr 6 16:23:45 2011 +0100
34525
34526 configure: Look for Khronos OpenGL spec files using pkg-config
34527
34528 Look for Khronos OpenGL spec files using pkg-config, rather than downloading them
34529
34530 Also add a --with-khronos-spec-dir=PATH configure option so XWin can be directed
34531 where to find these files without using the khronos-spec-files package
34532
34533 XWin with AIGLX requires OpenGL spec files in order to generate wrapper code which:
34534 (1) thunks from the glapi dispatch table which uses the default cdecl calling convention
34535 to native GL functions using the stdcall calling convention.
34536 (2) performs function address lookup for OpenGL 1.2+ functions, which are treated
34537 as extensions and so not directly linkable.
34538
34539 v2: KHRONOS_SPEC_DIR is only valid when XWIN_GLX_WINDOWS is defined. Avoid 'make dist'
34540 seeing invalid dependencies by only including rules using KHRONOS_SPEC_DIR if
34541 XWIN_GLX_WINDOWS is defined
34542
34543 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34544 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34545 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
34546
34547commit 0f284f0f42f26a242eb3153787f2e98b6b88dff8
34548Author: Jeremy Huddleston <jeremyhu@apple.com>
34549Date: Wed Apr 27 22:12:59 2011 -0700
34550
34551 XQuartz: BuildFix to build correctly with XPLUGIN_VERSION < 4
34552
34553 This fixes a regression introduced by d79cc14a51f3e8d4d2f66aad055092672cab1526
34554
34555 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34556
34557commit f6d4e75ec55ac6812f9dead42ecdffb9614578c7
34558Merge: ec9ea40 4318e6a
34559Author: Keith Packard <keithp@keithp.com>
34560Date: Wed Apr 27 12:08:51 2011 -0700
34561
34562 Merge remote-tracking branch 'jturney/master'
34563
34564commit ec9ea4017885cefe5519a4e890b0ff1a5518235a
34565Merge: c6cb70b 6f29dbf
34566Author: Keith Packard <keithp@keithp.com>
34567Date: Wed Apr 27 12:01:56 2011 -0700
34568
34569 Merge remote-tracking branch 'jeremyhu/master'
34570
34571commit 4318e6a147e78b2663c5e0ea6ba0d351a1e87f98
34572Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34573Date: Sat Mar 5 17:34:42 2011 +0000
34574
34575 Cygwin/X: Handle failure during winScreenInit()
34576
34577 Handle failure during winScreenInit() a bit more cleanly, rather than crashing
34578
34579 This avoids a crash with 'XWin -fullscreen -screen 0 @2 -screen 1 @1'
34580
34581 Also document that fullscreen may only be applied to one screen.
34582
34583 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34584 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34585 Tested-by: Colin Harrison <colin.harrison@virgin.net>
34586
34587commit ce6136f8c553bbc6d3e3affa0faa2afbf8054f44
34588Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34589Date: Tue Mar 23 20:06:33 2010 +0000
34590
34591 Cygwin/X: Make winOverrrideStyle() thread-safe
34592
34593 Make winOverrrideStyle() thread-safe
34594
34595 winOverrideStyle() is called from the internal WM client thread.
34596
34597 Accessing server-internal data structures to get window name and
34598 class is not safe, as there is no lock to ensure we do not collide
34599 with these data structures being updated in the server thread.
34600
34601 Rewrite so the internal client thread uses X client calls to
34602 obtain this data safely
34603
34604 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34605 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34606 Tested-by: Colin Harrison <colin.harrison@virgin.net>
34607
34608commit 0c603509eb7f9c83baf4e00b4558dce78f897ebf
34609Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34610Date: Fri Feb 11 13:15:40 2011 +0000
34611
34612 Cygwin/X: Cosmetic fixes to logging of result from X*TextPropertyToTextList()
34613
34614 Report XLocaleNotSupported result from X*TextPropertyToTextList()
34615 Fix formatting for unknown results reported for X*TextPropertyToTextList()
34616
34617 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34618 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34619 Tested-by: Colin Harrison <colin.harrison@virgin.net>
34620
34621commit 71550a8665d861384332d81239ca0c1586a17137
34622Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34623Date: Fri Jan 28 20:17:22 2011 +0000
34624
34625 Cygwin/X: Decorate function pointers retrieved via GetProcAddress with WINAPI
34626
34627 Decorate function pointers retrieved via GetProcAddress which are currently
34628 missing it with WINAPI, to ensure stdcall convention is used when calling them.
34629
34630 This fixes a crash currently seen when compiled -O2 and the -screen option uses
34631 a size and monitor number e.g. -screen 0 1280x1000@2
34632
34633 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34634 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34635 Tested-by: Colin Harrison <colin.harrison@virgin.net>
34636
34637commit 38a1f5c613a48ef9fd6ba043bc3028f487750d3a
34638Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34639Date: Tue Oct 12 17:12:02 2010 +0100
34640
34641 Cygwin/X: Don't make InputOnly windows visible
34642
34643 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34644 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34645 Tested-by: Colin Harrison <colin.harrison@virgin.net>
34646
34647commit bd288c3458bc1ba2cbb4c8416e5b2dfd849581e6
34648Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34649Date: Thu Jul 22 18:36:51 2010 +0100
34650
34651 Cygwin/X: Fix a GDI bitmap resource leak of window icons
34652
34653 Ensure any icon created specially for a window is destroyed when
34654 the window is destroyed
34655
34656 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34657 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34658 Tested-by: Colin Harrison <colin.harrison@virgin.net>
34659
34660commit 19e764eee0c8b74d877fb2b1d6aedc933976660e
34661Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34662Date: Wed Jul 21 18:11:13 2010 +0100
34663
34664 Cygwin/X: Internal WM workaround for Java AWT bug
34665
34666 Java applications using AWT on JRE 1.6.0 break with non-reparenting WMs AWT
34667 doesn't explicitly know about (See sun bug #6434227)
34668
34669 XDecoratedPeer.handleConfigureNotifyEvent() only processes non-synthetic
34670 ConfigureNotify events to update window location if it's identified the
34671 WM as a non-reparenting WM it knows about (compiz or lookingglass)
34672
34673 Rather than tell all sorts of lies to get XWM to recognize us as one of
34674 those, simply send a synthetic ConfigureNotify for every non-synthetic one
34675
34676 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34677 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34678 Tested-by: Colin Harrison <colin.harrison@virgin.net>
34679
34680commit c5a612fc18a7f5c542fca580494e2c66ecfc2159
34681Author: Jon TURNEY <jon.turney@dronecode.org.uk>
34682Date: Fri Apr 16 21:55:01 2010 +0100
34683
34684 configure: Let configure --enable/disable-aiglx control building of AIGLX for all DDXs
34685
34686 Let configure --enable/disable-aiglx control building of AIGLX for all DDXs. Currently
34687 we can't use --enable/disable-aiglx to control if Xwin DDX is built with AIGLX enabled,
34688 as at the moment it's forced off if we aren't building the X.Org DDX DRI or DRI2 loader
34689
34690 Rearrange things a bit, introducing a new automake conditional, AIGLX_DRI_LOADER to
34691 specifically indicate if the X.Org DDX DRI/DRI2 loader convenience library should be
34692 built, and replace the previous X.Org DDX-specific uses of the AIGLX conditional with that
34693
34694 As before, AIGLX_DRI_LOADER is only enabled if --enable-glx, --enable-aiglx and at least one
34695 of --enable-dri or --enable-dri2 are enabled
34696
34697 This allows the general conditional AIGLX to control if AIGLX is built for the XWin DDX as
34698 well
34699
34700 The C #define AIGLX set by AC_DEFINE(AIGLX) seems to be obsolete, I can't find anything
34701 which checks it
34702
34703 Updated for ajax's "glx: Make --disable-dri not disable AIGLX" patch, which allows DRI2
34704 to be enabled independently of DRI1
34705
34706 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
34707 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
34708 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
34709
34710commit 6f29dbf5e36bef5ecb08f02d367988dee0f9f1cd
34711Author: Jeremy Huddleston <jeremyhu@apple.com>
34712Date: Mon Apr 25 22:00:41 2011 -0700
34713
34714 XQuartz: Use ErrorF rather than fprintf to log errors
34715
34716 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34717
34718commit cb083b05c469352ef80e1005a29ac320f2e4e096
34719Author: Jeremy Huddleston <jeremyhu@apple.com>
34720Date: Mon Apr 25 21:25:10 2011 -0700
34721
34722 XQuartz: stub: Dead code removal
34723
34724 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34725
34726commit 72ed7551f494c61283a7ac3d7b570eac39cc9786
34727Author: Emanuele Giaquinta <emanuele.giaquinta@gmail.com>
34728Date: Mon Apr 25 10:38:17 2011 -0700
34729
34730 XQuartz: pbproxy: LP64: Fix itteration through XGetWindowProperty where sizeof(long) != 4
34731
34732 http://xquartz.macosforge.org/trac/ticket/476
34733
34734 Signed-off-by: Emanuele Giaquinta <emanuele.giaquinta@gmail.com>
34735 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
34736
34737commit 0a60192a85ba9f64b522da181c2fe8a5b93b79df
34738Author: Jeremy Huddleston <jeremyhu@apple.com>
34739Date: Sun Apr 24 22:01:48 2011 -0700
34740
34741 XQuartz: Enable logging to a file for better debugging
34742
34743 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34744
34745commit de4023f194c561b74f8fa904bea3dd5298230cd1
34746Author: Jeremy Huddleston <jeremyhu@apple.com>
34747Date: Sun Apr 24 21:35:04 2011 -0700
34748
34749 XQuartz: Rename launchd-id-prefix to bundle-id-prefix
34750
34751 It's used many other places than just for launchd.
34752
34753 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34754
34755commit e466745109416a2fcdf5d7389c80b814a0363676
34756Author: Jeremy Huddleston <jeremyhu@apple.com>
34757Date: Sun Apr 24 20:35:06 2011 -0700
34758
34759 XQuartz: Dead code removal
34760
34761 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34762
34763commit f1d867c0a1577e2c46d5dd4e262444e0cf6f8e25
34764Author: Jeremy Huddleston <jeremyhu@apple.com>
34765Date: Sat Apr 23 23:22:27 2011 -0700
34766
34767 XQuartz: Silence clang static analyzer
34768
34769 Call to 'malloc' has an allocation size of 0 bytes.
34770
34771 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34772
34773commit 913223e9ddf2cb927ecad99aeadfa730df8561ac
34774Author: Jeremy Huddleston <jeremyhu@apple.com>
34775Date: Sat Apr 23 23:15:49 2011 -0700
34776
34777 XQuartz: xpr: Dead code removal
34778
34779 Assigned value is always the same as the existing value.
34780
34781 Found by clang static analyzer
34782
34783 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34784
34785commit d0caa0e4cd255bea315757be2ff387cf57e8bdf7
34786Author: Jeremy Huddleston <jeremyhu@apple.com>
34787Date: Sat Apr 23 20:12:38 2011 -0700
34788
34789 XQuartz: Silence warnings about deprecated functionality where it is an intended fallback
34790
34791 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34792
34793commit 9244a3a24f63239ec12675f3df091583e06c8314
34794Author: Jeremy Huddleston <jeremyhu@apple.com>
34795Date: Sat Apr 23 12:55:39 2011 -0700
34796
34797 XQuartz: xpr: Use a serial queue rather than pthread mutexes for window_hash
34798
34799 Additionally removes some dead code and fixes double-locking in
34800 xprIsX11Window. xprIsX11Window doesn't need to do any locking because
34801 those resources are protected by the called functions themselves.
34802
34803 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34804 Reviewed-by: Daniel A. Steffen <dsteffen@apple.com>
34805
34806commit bac34a54f7cb84f346b1b833df9917bd1ac70223
34807Author: Jeremy Huddleston <jeremyhu@apple.com>
34808Date: Sat Apr 23 12:11:39 2011 -0700
34809
34810 XQuartz: xpr: Initialize window_hash in xprInit
34811
34812 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34813
34814commit 1596ea72d66a03d9accb534679172ca6f63f78e1
34815Author: Jeremy Huddleston <jeremyhu@apple.com>
34816Date: Sat Apr 23 11:55:49 2011 -0700
34817
34818 XQuartz: Use a lighter spinlock instead of a pthread_mutex_t in QuartzScreenSaver
34819
34820 Currently, we only end up here through a call to QuartzShowFullscreen, and
34821 this is always on the same thread. Future changes (such as further
34822 incorporating libdispatch) may allow this to change, but contention will
34823 remain minimal since the call is infrequent and it is short held.
34824
34825 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34826 Reviewed-by: Daniel A. Steffen <dsteffen@apple.com>
34827
34828commit 3e253c603bc18f06fa48b611797eb5a7c8a96fe4
34829Author: Jeremy Huddleston <jeremyhu@apple.com>
34830Date: Sat Apr 23 01:48:25 2011 -0700
34831
34832 XQuartz: Remove the threadSafety dead-ish code
34833
34834 It's been a few years now since we've needed this to debug thread
34835 boundaries, so punt it out to clean up the namespace polution.
34836
34837 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34838
34839commit a52c8078c9cc83c84a8c6eb58810f49bdb90bcc1
34840Author: Jeremy Huddleston <jeremyhu@apple.com>
34841Date: Sat Apr 23 01:39:16 2011 -0700
34842
34843 XQuartz: Use xorg_backtrace() instead of spewCallStack()
34844
34845 xorg_backtrace() has been in os for two years now, we might as well
34846 start using it.
34847
34848 Ref: 94ed0ba1b5043ad9fc33b42756af447d5ab15bbd
34849
34850 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34851
34852commit d79cc14a51f3e8d4d2f66aad055092672cab1526
34853Author: Jeremy Huddleston <jeremyhu@apple.com>
34854Date: Fri Apr 22 12:05:32 2011 -0700
34855
34856 XQuartz: Fix compilation warnings with XPLUGIN_VERSION >= 4
34857
34858 xprAppleWM.c:143: warning: initialization from incompatible pointer type
34859 xprAppleWM.c:144: warning: initialization from incompatible pointer type
34860
34861 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34862
34863commit 6128544fd58ced0ed738b7150865294d214fb4eb
34864Author: Jeremy Huddleston <jeremyhu@apple.com>
34865Date: Mon Apr 25 12:38:07 2011 -0700
34866
34867 XQuartz: Bump bundle version to 2.7.0
34868
34869 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34870
34871commit 4944de24849a109c65f1b353bc12e44e90a1211d
34872Author: Jeremy Huddleston <jeremyhu@apple.com>
34873Date: Sat Apr 23 23:25:39 2011 -0700
34874
34875 rootless: Fix a typo in RootlessGlyphs which resulted in a garbage value
34876
34877 Found by clang static analyzer
34878
34879 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34880
34881commit 2098cb03c6b64bfca7694fc2b5213edb77bc12e4
34882Author: Jeremy Huddleston <jeremyhu@apple.com>
34883Date: Sat Apr 23 23:27:16 2011 -0700
34884
34885 rootless: Dead code removal
34886
34887 Found by clang static analyzer
34888
34889 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34890
34891commit bb4d145bd25e2aee988b100ecf1105ea3b6a40b8
34892Author: Jeremy Huddleston <jeremyhu@apple.com>
34893Date: Sat Apr 23 21:24:27 2011 -0700
34894
34895 glx: Silence warnings when building with clang
34896
34897 This replaces AX_TLS (GPL3) with XORG_TLS (MIT)
34898
34899 In file included from glapi.c:46:
34900 In file included from ./glapi.h:51:
34901 ./glthread.h:237:20: error: unknown attribute 'tls_model' ignored [-Werror,-Wunknown-attributes]
34902 __attribute__((tls_model("initial-exec")));
34903 ^
34904 In file included from glapi.c:46:
34905 ./glapi.h:92:20: error: unknown attribute 'tls_model' ignored [-Werror,-Wunknown-attributes]
34906 __attribute__((tls_model("initial-exec")));
34907 ^
34908 glapi.c:82:20: error: unknown attribute 'tls_model' ignored [-Werror,-Wunknown-attributes]
34909 __attribute__((tls_model("initial-exec"))) = NULL;
34910 ^
34911 glapi.c:85:20: error: unknown attribute 'tls_model' ignored [-Werror,-Wunknown-attributes]
34912 __attribute__((tls_model("initial-exec")));
34913 ^
34914 4 errors generated.
34915
34916 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34917 Tested-by: Jamey Sharp <jamey@minilop.net>
34918 Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
34919 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
34920
34921commit 5c9eafc8e5f8575c06591d244c2cb5ea07691cb4
34922Author: Jeremy Huddleston <jeremyhu@apple.com>
34923Date: Sat Apr 23 20:55:53 2011 -0700
34924
34925 render: Silence warnings when building with clang
34926
34927 picture.c:351:37: error: implicit conversion from 'unsigned int' to 'CARD16' (aka 'unsigned short') changes value from 4294967295 to 65535
34928 [-Werror,-Wconstant-conversion]
34929 pFormats[f].direct.alphaMask = Mask(PICT_FORMAT_A(format));
34930 ^~~~~~~~~~~~~~~~~~~~~~~~~~~
34931 ...
34932 fatal error: too many errors emitted, stopping now [-ferror-limit=]
34933
34934 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34935 Suggested-by: Jamey Sharp <jamey@minilop.net>
34936 Reviewed-by: Jamey Sharp <jamey@minilop.net>
34937
34938commit c524f8bb768f886d413839bc22184098394c2559
34939Author: Jeremy Huddleston <jeremyhu@apple.com>
34940Date: Sat Apr 23 20:49:27 2011 -0700
34941
34942 os: Silence warnings when building with clang
34943
34944 access.c:1492:20: error: equality comparison with extraneous parentheses [-Werror,-Wparentheses]
34945 if ((host->family == FamilyServerInterpreted)) {
34946 ~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
34947 access.c:1492:20: note: use '=' to turn this equality comparison into an assignment
34948 if ((host->family == FamilyServerInterpreted)) {
34949 ^~
34950 =
34951 access.c:1492:20: note: remove extraneous parentheses around the comparison to silence this warning
34952 if ((host->family == FamilyServerInterpreted)) {
34953 ~ ^ ~
34954
34955 In file included from xstrans.c:8:
34956 In file included from /usr/X11/include/X11/Xtrans/transport.c:62:
34957 /usr/X11/include/X11/Xtrans/Xtranssock.c:262:5: error: implicit declaration of function 'ErrorF' is invalid in C99
34958 [-Werror,-Wimplicit-function-declaration]
34959 PRMSG (3,"SocketSelectFamily(%s)\n", family, 0, 0);
34960 ^
34961
34962 log.c:180:29: error: format string is not a string literal [-Werror,-Wformat-nonliteral]
34963 if (asprintf(&logFileName, fname, display) == -1)
34964 ^~~~~
34965 log.c:190:26: error: format string is not a string literal [-Werror,-Wformat-nonliteral]
34966 if ((asprintf(&suffix, backup, display) == -1) ||
34967 ^~~~~~
34968 log.c:382:25: error: format string is not a string literal [-Werror,-Wformat-nonliteral]
34969 LogVWrite(verb, tmpBuf, args);
34970 ^~~~~~
34971
34972 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34973 Reviewed-by: Jamey Sharp <jamey@minilop.net>
34974
34975commit ab81aa91404ca0fa6843ce8021cbd9de42255a8f
34976Author: Jeremy Huddleston <jeremyhu@apple.com>
34977Date: Sat Apr 23 20:39:25 2011 -0700
34978
34979 Xext: Silence warnings when building with clang
34980
34981 xvmain.c:1113:22: warning: equality comparison with extraneous parentheses [-Wparentheses]
34982 if ((pf->depth == pDraw->depth)
34983 ~~~~~~~~~~^~~~~~~~~~~~~~~
34984 xvmain.c:1113:22: note: use '=' to turn this equality comparison into an assignment
34985 if ((pf->depth == pDraw->depth)
34986 ^~
34987 =
34988 xvmain.c:1113:22: note: remove extraneous parentheses around the comparison to silence this warning
34989 if ((pf->depth == pDraw->depth)
34990 ~ ^ ~
34991 1 warning generated.
34992
34993 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
34994 Reviewed-by: Jamey Sharp <jamey@minilop.net>
34995
34996commit fd086f87cd1bab53c9e918cb687009e7ac8718d8
34997Author: Jeremy Huddleston <jeremyhu@apple.com>
34998Date: Sat Apr 23 20:36:53 2011 -0700
34999
35000 fb: Silence warnings when building with clang
35001
35002 fbpict.c:163:8: warning: implicit conversion from enumeration type 'PictFormatShort' (aka 'enum _PictFormatShort') to different enumeration
35003 type 'pixman_format_code_t' [-Wconversion]
35004 pict->format,
35005 ~~~~~~^~~~~~
35006
35007 fbbltone.c:486:2: warning: shift result (281474959933440) requires 49 bits to represent, but 'int' only has 32 bits [-Wshift-overflow]
35008 C2_24( 0, 0), C2_24 ( 1, 0), C2_24 ( 2, 0), C2_24 ( 3, 0),
35009 ^~~~~~~~~~~~
35010 fbbltone.c:474:6: note: instantiated from:
35011 SelMask24(b,1,r))
35012 ^
35013 fbbltone.c:429:46: note: instantiated from:
35014 ^
35015 fbbltone.c:427:18: note: instantiated from:
35016 0xffffff << Mask24Check(x,r)) : 0)
35017 ~~~~~~~~ ^
35018
35019 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35020 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35021
35022commit aad7b324aefc906f28ac9d10c64650b3445c2ae0
35023Author: Jeremy Huddleston <jeremyhu@apple.com>
35024Date: Sun Apr 24 19:57:22 2011 -0700
35025
35026 os: Add missing _X_ATTRIBUTE_PRINTF to va_list variants
35027
35028 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35029
35030commit 5423da9fb2ec0dfeff866f12cf26fbba04ea673d
35031Author: Nicolas Kaiser <nikai@nikai.net>
35032Date: Sun Apr 24 12:44:15 2011 +0200
35033
35034 xkb: remove duplicated include
35035
35036 Remove duplicated include.
35037
35038 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35039 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35040
35041commit 79b3a7f83d7cbee397eecebef1af31c048d295e8
35042Author: Nicolas Kaiser <nikai@nikai.net>
35043Date: Sun Apr 24 12:43:05 2011 +0200
35044
35045 test/xi2: remove duplicated include
35046
35047 Remove duplicated include.
35048
35049 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35050 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35051
35052commit a17c30d43e0db6c65cc73cd2be5193c02f138bf5
35053Author: Nicolas Kaiser <nikai@nikai.net>
35054Date: Sun Apr 24 12:42:07 2011 +0200
35055
35056 hw/xwin: remove duplicated includes
35057
35058 Remove duplicated includes.
35059
35060 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35061 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35062 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
35063
35064commit c270cfc30c547e28017533f490904013f73cf6f0
35065Author: Nicolas Kaiser <nikai@nikai.net>
35066Date: Sun Apr 24 12:41:11 2011 +0200
35067
35068 hw/xquartz: remove duplicated includes
35069
35070 Remove duplicated includes.
35071
35072 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35073 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35074
35075commit 50ced6cfa0aef472e713ffccfdba003829df68df
35076Author: Nicolas Kaiser <nikai@nikai.net>
35077Date: Sun Apr 24 12:40:11 2011 +0200
35078
35079 hw/xnest: remove duplicated include
35080
35081 Remove duplicated include.
35082
35083 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35084 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35085
35086commit 387f45a7077faadf232edc3f608c4a081a7fd290
35087Author: Nicolas Kaiser <nikai@nikai.net>
35088Date: Sun Apr 24 12:39:11 2011 +0200
35089
35090 hw/xfree86/modes: remove duplicated include
35091
35092 Remove duplicated include.
35093
35094 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35095 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35096
35097commit 78a9ec125dcb1546ca7e7e18aedf2b323eebb1fc
35098Author: Nicolas Kaiser <nikai@nikai.net>
35099Date: Sun Apr 24 12:38:08 2011 +0200
35100
35101 hw/xfree86/fbdevhw: remove duplicated include
35102
35103 Remove duplicated include.
35104
35105 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35106 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35107
35108commit 0320db25f917980621107b790c4fc5171cb0b6ac
35109Author: Nicolas Kaiser <nikai@nikai.net>
35110Date: Sun Apr 24 12:36:48 2011 +0200
35111
35112 hw/xfree86/dri: remove duplicated includes
35113
35114 Remove duplicated includes.
35115
35116 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35117 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35118
35119commit b56271b954d6ab8eeb320f314925b77403458b9c
35120Author: Nicolas Kaiser <nikai@nikai.net>
35121Date: Sun Apr 24 12:35:45 2011 +0200
35122
35123 hw/xfree86/ddc: remove duplicated include
35124
35125 Remove duplicated include.
35126
35127 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35128 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35129
35130commit 86c0c8b9c567f57fe7477d5302fa22dc7557ce90
35131Author: Nicolas Kaiser <nikai@nikai.net>
35132Date: Sun Apr 24 12:34:46 2011 +0200
35133
35134 dix: remove duplicated includes
35135
35136 Remove duplicated includes.
35137
35138 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35139 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35140
35141commit 9b5046d2130aead79956019faf7103b5c676fd02
35142Author: Nicolas Kaiser <nikai@nikai.net>
35143Date: Sun Apr 24 12:33:32 2011 +0200
35144
35145 Xi: remove duplicated includes
35146
35147 Remove duplicated includes.
35148
35149 Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
35150 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
35151
35152commit c6cb70be1ed7cf73bd3411b8d66ec05a9efcfeb9
35153Author: Søren Sandmann Pedersen <ssp@redhat.com>
35154Date: Mon Mar 28 13:30:52 2011 -0400
35155
35156 Fix trapezoid and triangle rendering to windows
35157
35158 For fbAdd{Traps,Triangles}() and fbRasterizeTrapezoid() this is just a
35159 matter of adding the image offsets to the trap offsets.
35160
35161 For fbShapes, the story is more complicated:
35162
35163 The recently added pixman API did not allow offsetting
35164 trapezoids. Instead, it would use x_dst and y_dst in such a way that
35165 the effect was to only offset the source image.
35166
35167 In pixman 0.21.8, this API has changed such that all the traps are
35168 conceptually rendered to an infinitely big image, and the source and
35169 destination coordinates are then aligned with (0, 0) of that
35170 image. This means offsetting dst_x and dst_y will now offset the
35171 entire drawing, which is similar to how other composite functions
35172 work.
35173
35174 This patch then changes fbComposite{Triangles,Traps} such that the
35175 source image is aligned with the shapes, and the destination
35176 coordinates offset according to drawable->{x, y}.
35177
35178 Reviewed-by: Keith Packard <keithp@keithp.com>
35179 Signed-off-by: Soren Sandmann <ssp@redhat.com>
35180
35181commit 1b96a99d8edd9016bc4a35348f9d5ddb45832f14
35182Author: Søren Sandmann Pedersen <ssp@redhat.com>
35183Date: Thu Mar 10 08:52:41 2011 -0500
35184
35185 fb: Call miCompositeSourceValidate() on the source in fbShapes()
35186
35187 Reviewed-by: Keith Packard <keithp@keithp.com>
35188 Signed-off-by: Soren Sandmann <ssp@redhat.com>
35189
35190commit 04635069554859ec67003b89f56965421cba7f52
35191Author: Søren Sandmann Pedersen <ssp@redhat.com>
35192Date: Tue Mar 29 00:11:00 2011 -0400
35193
35194 render: Remove unused TriStrip and TriFan typedefs
35195
35196 Reviewed-by: Keith Packard <keithp@keithp.com>
35197 Signed-off-by: Soren Sandmann <ssp@redhat.com>
35198
35199commit 2b0cabec620f3a2a5e431052441b092ef979bf94
35200Author: Søren Sandmann Pedersen <ssp@redhat.com>
35201Date: Tue Mar 29 00:07:44 2011 -0400
35202
35203 render: Remove unused fields in the source picture structs
35204
35205 The fields class, stopRange, colorTable and colorTableSize are not
35206 used by any current code.
35207
35208 Reviewed-by: Keith Packard <keithp@keithp.com>
35209 Signed-off-by: Soren Sandmann <ssp@redhat.com>
35210
35211commit b0d84f94393edab395d65d2b2cb983fc9fec3d36
35212Author: Søren Sandmann Pedersen <ssp@redhat.com>
35213Date: Mon Mar 28 20:59:34 2011 -0400
35214
35215 render: Delete PictureGradientColor()
35216
35217 PictureGradientColor(), INTERPOLATE_PIXEL_256() and premultiply() are
35218 not used by anything.
35219
35220 Reviewed-by: Keith Packard <keithp@keithp.com>
35221 Signed-off-by: Soren Sandmann <ssp@redhat.com>
35222
35223commit 61a92a78cd49969f74a046fa26c3199e06365814
35224Author: Søren Sandmann Pedersen <ssp@redhat.com>
35225Date: Tue Mar 29 13:06:36 2011 -0400
35226
35227 Add RegionInitBoxes(), and fix some buggy callers of RegionInit().
35228
35229 The interface to RegionInit():
35230
35231 RegionInit (RegionPtr pReg, BoxPtr rect, int size);
35232
35233 is very confusing because it doesn't take a list of boxes, it takes
35234 *one* box, but if that box is NULL, it initializes an empty region
35235 with 'size' rectangles preallocated.
35236
35237 Most callers of this function were correctly passing either NULL or
35238 just one box, but there were three confused cases, where the code
35239 seems to expect a region to be created from a list of boxes.
35240
35241 This patch adds a new function RegionInitBoxes() and fixes those
35242 instances to call that instead.
35243
35244 And yes, the pixman function to initialize a region from a list of
35245 boxes is called init_rects() because pixman is also awesome.
35246
35247 V2: Make RegionInitBoxes() return a Bool indicating whether the call
35248 succeeded, and fix the callers to check this return value.
35249
35250 Reviewed-by: Keith Packard <keithp@keithp.com>
35251 Signed-off-by: Søren Sandmann <ssp@redhat.com>
35252
35253commit c7bce22b58530239e583d91ae56312bad1630da4
35254Author: Søren Sandmann Pedersen <ssp@redhat.com>
35255Date: Mon Mar 28 11:38:11 2011 -0400
35256
35257 Track damage for fbTrapezoids() and fbTriangles().
35258
35259 These calls no longer go through the CompositePicture() hook, so
35260 damage was no longer generated for them. This patch simply damages the
35261 entire destination clip region.
35262
35263 It would be possible to generate tighter damage for certain operators
35264 such as Over and Add, where blank source pixels have no effect on the
35265 destination, but given that virtually all trapezoid rendering takes
35266 place on pixmaps, it's unlikely that anybody would actually benefit
35267 from this optimization, and the miTrapezoidBounds function did
35268 sometimes show up on profiles, probably because it does several
35269 divisions per trapezoid.
35270
35271 V2: Call DamageRegionProcessPending() - pointed out by Michel Dänzer.
35272 V3: Call DamageRegionProcessPending() *after* rendering -
35273 pointed out by Maarten Maathuis
35274
35275 Reviewed-by: Michel Dänzer <daenzer@vmware.com>
35276 Signed-off-by: Søren Sandmann <ssp@redhat.com>
35277
35278commit 918a9c99cf2ebc73acb34b95f597904b93c690d9
35279Merge: 88c4622 5f496bc
35280Author: Keith Packard <keithp@keithp.com>
35281Date: Fri Apr 22 11:20:16 2011 -0700
35282
35283 Merge remote-tracking branch 'jeremyhu/master'
35284
35285commit 88c4622b594a1725d0cee86bc82ad640d241c520
35286Author: Aaron Plattner <aplattner@nvidia.com>
35287Date: Mon Apr 18 08:23:48 2011 -0700
35288
35289 linux: Retry VT ioctls while errno == EINTR
35290
35291 When the smart scheduler is enabled, the VT ioctls (particularly
35292 VT_WAITACTIVE) can be interrupted by the smart scheduler's SIGALRMs.
35293 Previously, this caused the server to immediately continue on to
35294 ScreenInit, almost certainly causing a crash or failure because the X
35295 server that owned the VT hadn't finished cleaning up. As of commit
35296 7ee965a300c9eddcc1acacf9414cfe3e589222a8, it causes a FatalError
35297 instead.
35298
35299 Retrying the ioctl as long as it fails with errno == EINTR fixes the
35300 problem and allows server regenerations to trigger VT switches that
35301 actually succeed.
35302
35303 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
35304 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
35305 Reviewed-by: Cyril Brulebois <kibi@debian.org>
35306 Signed-off-by: Keith Packard <keithp@keithp.com>
35307
35308commit 302bdc3c9678c028bed71dbe9c5ba04997626b6b
35309Merge: 001b6b8 6f97fe5
35310Author: Keith Packard <keithp@keithp.com>
35311Date: Fri Apr 22 10:48:37 2011 -0700
35312
35313 Merge remote-tracking branch 'whot/for-keith'
35314
35315commit 5f496bc91941140ceaa3061472fda95bd812f2d2
35316Author: Jeremy Huddleston <jeremyhu@apple.com>
35317Date: Fri Apr 22 01:23:09 2011 -0700
35318
35319 XQuartz: Do translation and handoff of NSEvent to X11 in a separate serial queue
35320
35321 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35322
35323commit 72bd232b117b2867282e0ae1855d779e126f912b
35324Author: Jeremy Huddleston <jeremyhu@apple.com>
35325Date: Fri Apr 22 00:39:12 2011 -0700
35326
35327 XQuartz: Send tablet proximity events with tilt and pressure
35328
35329 <rdar://problem/6257569>
35330
35331 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35332
35333commit 034538ea9b4770025e3573bc708039cabbe1e10d
35334Author: Jeremy Huddleston <jeremyhu@apple.com>
35335Date: Thu Apr 21 16:31:58 2011 -0700
35336
35337 XQuartz: Use dispatch_async to handoff the FD
35338
35339 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35340
35341commit ca7b9e6c817681b9cec738e43cf020ac19b5e732
35342Author: Jeremy Huddleston <jeremyhu@apple.com>
35343Date: Thu Apr 21 15:51:32 2011 -0700
35344
35345 configure.ac: Add check for libdispatch when building for darwin
35346
35347 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35348
35349commit 3960115dbc83ec1eb8d9c8e90466af3fa0b32abd
35350Author: Jeremy Huddleston <jeremyhu@apple.com>
35351Date: Thu Apr 21 15:19:12 2011 -0700
35352
35353 XQuartz: Fix prototypes for thread functions
35354
35355 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35356
35357commit 7524dbd06113ec081eaa882aa54e03553ccf96aa
35358Author: Jeremy Huddleston <jeremyhu@apple.com>
35359Date: Thu Apr 21 15:11:52 2011 -0700
35360
35361 XQuartz: Make the DarwinProcessFDAdditionQueue_thread wait 3 seconds to allow xinitrc to catch up
35362
35363 Previously, we weren't always waiting the full three seconds. This should
35364 be better, but is still sub-optimal. We really want to start processing
35365 these once a WM has been started.
35366
35367 http://xquartz.macosforge.org/trac/ticket/416
35368
35369 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
35370
35371commit 6f97fe5df16f947e52875187ff4f9e25fab33ad7
35372Author: Peter Hutterer <peter.hutterer@who-t.net>
35373Date: Thu Apr 21 14:55:22 2011 +1000
35374
35375 dix: improve control flow in QueryTrackers
35376
35377 If the velocity is 0, skip the remainder.
35378 If we're not in range, skip the remainder.
35379
35380 No functional change.
35381
35382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35383 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35384
35385commit a0eda8cd36ffba2ecc79e0d7a1908c89b45e76c4
35386Author: Peter Hutterer <peter.hutterer@who-t.net>
35387Date: Thu Apr 21 14:45:09 2011 +1000
35388
35389 dix: rename a bunch of variables to be more self-explanatory
35390
35391 i → used_offset
35392 iveloc → initial_velocity
35393 res → result
35394 vdiff → velocity_diff
35395 vfac → velocity_factor
35396 tmp → tracker_velocity
35397
35398 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35399 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35400 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35401 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35402
35403commit e4901905903c5630a1092c69e42f313abd05187b
35404Author: Peter Hutterer <peter.hutterer@who-t.net>
35405Date: Wed Apr 20 15:41:45 2011 +1000
35406
35407 dix: reduce scope of tmp and mult.
35408
35409 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35410 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35411 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35412 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35413
35414commit 81c6e79f42cc263349e04c0d82b3a85d6b557885
35415Author: Peter Hutterer <peter.hutterer@who-t.net>
35416Date: Wed Apr 20 15:39:27 2011 +1000
35417
35418 dix: reduce the work done by ApplySoftening
35419
35420 We can modify fdx/fdy in-place rather than requiring dx/dy as well. And the
35421 decision to soften can be made in the caller (unless decided by the velocity
35422 state).
35423
35424 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35425 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35426 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35427 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35428
35429commit 5870d507cee8547c0993791195fe04d96b271289
35430Author: Peter Hutterer <peter.hutterer@who-t.net>
35431Date: Wed Apr 20 15:32:33 2011 +1000
35432
35433 dix: split softening and constant deceleration into two functions
35434
35435 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35436 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35437 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35438 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35439
35440commit 8ae90552f95f025ad09a17258d56603a424ca076
35441Author: Peter Hutterer <peter.hutterer@who-t.net>
35442Date: Wed Apr 20 15:30:19 2011 +1000
35443
35444 dix: use single return statement in ApplySimpleSoftening
35445
35446 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35447 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35448 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35449 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35450
35451commit c45f4cc6bd885855ecb54d9113010906d9ee58e2
35452Author: Peter Hutterer <peter.hutterer@who-t.net>
35453Date: Wed Apr 20 15:28:32 2011 +1000
35454
35455 dix: rename od, d to prev_delta, delta
35456
35457 And res to result.
35458
35459 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35460 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35461 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35462 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35463
35464commit 5ffe06bada85e98d46a071dad578d72f5b96cb51
35465Author: Peter Hutterer <peter.hutterer@who-t.net>
35466Date: Wed Apr 20 15:26:11 2011 +1000
35467
35468 dix: Don't use short as bool
35469
35470 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35471 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35472 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35473 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35474
35475commit 97c1967689d7990be86ee490690ac5a91935c875
35476Author: Peter Hutterer <peter.hutterer@who-t.net>
35477Date: Wed Apr 20 15:02:46 2011 +1000
35478
35479 dix: rename "res" to "result" for improved readability
35480
35481 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35482 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35483 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35484 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35485
35486commit 1c162ebcaba78930f34639145a8ac01b87f8e6fd
35487Author: Peter Hutterer <peter.hutterer@who-t.net>
35488Date: Wed Apr 20 14:56:44 2011 +1000
35489
35490 dix: change ProcessVelocityData2D to BOOL.
35491
35492 Don't confuse users with a return type of short, that's even less indicative
35493 that it returns 0/non-0 than "int".
35494
35495 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35496 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35497 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35498
35499commit 70b76286dbef99daabbafccd23f3ba436fe07b09
35500Author: Peter Hutterer <peter.hutterer@who-t.net>
35501Date: Wed Apr 20 16:28:21 2011 +1000
35502
35503 dix: add some more documentation to ptraccel code
35504
35505 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35506 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35507 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35508 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35509
35510commit 6b3a4cc6af240c12d05e27250b61e954eafe9f29
35511Author: Peter Hutterer <peter.hutterer@who-t.net>
35512Date: Wed Apr 20 14:44:51 2011 +1000
35513
35514 dix: don't pass the index for a tracker around, pass the tracker
35515
35516 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35517 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35518 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35519 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35520
35521commit 9e8645cacf5f1c398d7a6e837a36c66cf6cf0c9c
35522Author: Peter Hutterer <peter.hutterer@who-t.net>
35523Date: Wed Apr 20 14:23:29 2011 +1000
35524
35525 dix: CalcTracker only uses the tracker, thus only pass the tracker.
35526
35527 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35528 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35529 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35530 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35531 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35532
35533commit 62f4bef82d1252515689420f8aac198e2f29c4f1
35534Author: Peter Hutterer <peter.hutterer@who-t.net>
35535Date: Thu Apr 21 13:58:55 2011 +1000
35536
35537 dix: use single return statement in DoGetDirection
35538
35539 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35540 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35541 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35542 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35543
35544commit 5f8edbe47e3915f71a20f063a9e8b49591239600
35545Author: Peter Hutterer <peter.hutterer@who-t.net>
35546Date: Wed Apr 20 13:12:53 2011 +1000
35547
35548 dix: use single return value in GetDirection
35549
35550 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35551 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35552 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35553 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35554
35555commit afaa17812513232785c8c629294190629197dfd8
35556Author: Peter Hutterer <peter.hutterer@who-t.net>
35557Date: Wed Apr 20 13:12:35 2011 +1000
35558
35559 dix: document GetDirection
35560
35561 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35562 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35563 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35564 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35565
35566commit df90db892d16143742eb3c09e048f5b76457a62e
35567Author: Peter Hutterer <peter.hutterer@who-t.net>
35568Date: Wed Apr 20 11:49:57 2011 +1000
35569
35570 dix: document DoGetDirection's maths
35571
35572 This is the best explanation I can come up with, but it seems to hold true
35573 for my example values.
35574
35575 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35576 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35577 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35578 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35579
35580commit 7513190ec7bc08db1629b04d3f88f86ea2b1588e
35581Author: Peter Hutterer <peter.hutterer@who-t.net>
35582Date: Wed Apr 20 11:47:21 2011 +1000
35583
35584 dix: improve comment, directions flagged are 45° each
35585
35586 The two directions returned by this calculation are always the two boundary
35587 conditions. Since we don't do quadrants but octants, the flagged ones are
35588 45° each. e.g. an angle of 35° flags E and NE.
35589
35590 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35591 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35592 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35593
35594commit e804ae85fec290894af69c2e65df55e73b417287
35595Author: Peter Hutterer <peter.hutterer@who-t.net>
35596Date: Wed Apr 20 11:12:35 2011 +1000
35597
35598 dix: improve readbility of DoGetDirection.
35599
35600 Use enums for the direction bits, not hardcoded bitfield values that are
35601 added up.
35602
35603 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35604 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35605 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35606 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35607
35608commit 709b291972f8c65f4427f36b45ee78c3895d1e6e
35609Author: Peter Hutterer <peter.hutterer@who-t.net>
35610Date: Wed Apr 20 10:40:28 2011 +1000
35611
35612 dix: only use a single return where only one is needed.
35613
35614 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35615 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35616 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35617 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35618
35619commit 8c528c667da7c7fba0ba07973a06b2d4613fbf70
35620Author: Peter Hutterer <peter.hutterer@who-t.net>
35621Date: Wed Apr 20 13:14:11 2011 +1000
35622
35623 dix: fix typo in direction calculation
35624
35625 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35626 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35627 Reviewed-by: Jamey Sharp <jamey@minilop.net>
35628 Reviewed-by: Simon Thum <simon.thum@gmx.de>
35629
35630commit 47f8cba6f3ae24e5dcdc4e348bdaaaf00d25a91c
35631Author: Peter Hutterer <peter.hutterer@who-t.net>
35632Date: Mon Apr 11 14:50:16 2011 +1000
35633
35634 xfree86: removed unused "event" variable in xf86PostMotionEventM
35635
35636 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35637
35638commit c6212a3e5dc43bd3b5263b392cb75ec96cdf4e4f
35639Author: Peter Hutterer <peter.hutterer@who-t.net>
35640Date: Fri Apr 15 14:13:51 2011 +1000
35641
35642 dix: silence compiler warning
35643
35644 resource.c: In function 'AddResource':
35645 resource.c:493:3: warning: format '%lx' expects type 'long unsigned int',
35646 but argument 3 has type 'RESTYPE'
35647
35648 RESTYPE is uint32_t, not long.
35649
35650 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35651 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35652
35653commit 196d679bed2e9959d2fca89d4ee4bbc551681d0b
35654Author: Peter Hutterer <peter.hutterer@who-t.net>
35655Date: Fri Apr 15 14:51:06 2011 +1000
35656
35657 test: remove glib dependency
35658
35659 The few features from the glib test suite we used can be replaced with
35660 assert and printf. This patch is a simple replacement for these two
35661 g_assert → assert
35662 g_test_message → printf
35663
35664 g_test_init is removed and so is g_test_bug_base. g_test_run replaced with a
35665 simple return 0.
35666
35667 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35668 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
35669 Acked-by: Gaetan Nadon <memsize@videotron.ca>
35670
35671commit 071a6ac4d0c347aa7fc6efe37f4f6992524d7ef1
35672Author: Peter Hutterer <peter.hutterer@who-t.net>
35673Date: Mon Apr 11 15:22:52 2011 +1000
35674
35675 input: remove GetKeyboardValuatorEvents, this is now unnecessary.
35676
35677 GetKeyboardValuatorEvents handles NULL valuator masks already, so the
35678 GetKeyboardEvents wrapper is not needed. Rename GKVE to GKE.
35679
35680 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35681 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35682 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35683
35684commit 91a735328c59db006f0ed52b49e9c59895796c0c
35685Author: Peter Hutterer <peter.hutterer@who-t.net>
35686Date: Wed Mar 9 10:45:55 2011 +1000
35687
35688 test: add some XIPassiveGrab protocol testing.
35689
35690 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35691 Acked-by: Daniel Stone <daniel@fooishbar.org>
35692
35693commit 60b08e013dd1e971f82e5bc8708d3f120c217497
35694Author: Peter Hutterer <peter.hutterer@who-t.net>
35695Date: Wed Mar 9 10:44:48 2011 +1000
35696
35697 Xi: don't swap the status byte in the XIPassiveGrab replies
35698
35699 Reported-by: Julien Cristau <jcristau@debian.org>
35700 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35701 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35702
35703commit a3f37f3698880aec508b5ecfb88daf39360610aa
35704Author: Peter Hutterer <peter.hutterer@who-t.net>
35705Date: Wed Mar 9 10:07:19 2011 +1000
35706
35707 Xi: exit with error value if CheckGrabValues failed.
35708
35709 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35710 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35711 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35712
35713commit f4d9ff73b161d84d3fcacc5834fa714c113c7a10
35714Author: Peter Hutterer <peter.hutterer@who-t.net>
35715Date: Wed Mar 9 09:54:12 2011 +1000
35716
35717 Xi: fix reply swapping function check for XIPassiveGrabDevice
35718
35719 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35720 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35721 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35722
35723commit 2a49ffa3c94819e9f28dd939f23ee8a675258172
35724Author: Peter Hutterer <peter.hutterer@who-t.net>
35725Date: Wed Mar 9 09:53:47 2011 +1000
35726
35727 Xi: return the bad device ID if a passive grab fails with BadDevice.
35728
35729 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35730 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35731 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
35732
35733commit 315aa8444b13f827237674535eb2ad55a2066808
35734Author: Peter Hutterer <peter.hutterer@who-t.net>
35735Date: Wed Mar 23 10:32:35 2011 +1000
35736
35737 configure: actually require macros 1.13
35738
35739 Introduced in 6a5bf15fa99cf5b2358b3b3e2f29e5044aa8724a
35740
35741 Reported-by: Dave Airlie <airlied@redhat.com>
35742 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35743 Reviewed-by: Dave Airlie <airlied@redhat.com>
35744 Reviewed-by: Julien Cristau <jcristau@debian.org>
35745 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
35746
35747commit f0f0eec869d932a2d9228ff3e41e9e7cda155e68
35748Author: Peter Hutterer <peter.hutterer@who-t.net>
35749Date: Mon Apr 11 11:24:29 2011 +1000
35750
35751 Xi: silence valgrind warning. (#36120)
35752
35753 Conditional jump or move depends on uninitialised value(s)
35754 at 0x4357A1: GetEventMask (events.c:454)
35755 by 0x43B9E8: DeliverEventsToWindow (events.c:2029)
35756 by 0x4E0C59: SendEventToAllWindows (exevents.c:2125)
35757 by 0x4E8301: XISendDeviceHierarchyEvent (xichangehierarchy.c:118)
35758 by 0x426F99: DisableDevice (devices.c:507)
35759 by 0x46BF72: xf86Wakeup (xf86Events.c:457)
35760 by 0x432ABA: WakeupHandler (dixutils.c:419)
35761 by 0x45B708: WaitForSomething (WaitFor.c:235)
35762 by 0x42E8D9: Dispatch (dispatch.c:367)
35763 by 0x422DC9: main (main.c:287)
35764 Uninitialised value was created by a stack allocation
35765 at 0x4E8190: XISendDeviceHierarchyEvent (xichangehierarchy.c:61)
35766
35767 Conditional jump or move depends on uninitialised value(s)
35768 at 0x43BB78: DeliverEventsToWindow (events.c:2010)
35769 by 0x4DDEEA: FindInterestedChildren (exevents.c:2103)
35770 by 0x4DDEFF: FindInterestedChildren (exevents.c:2104)
35771 by 0x4DDEFF: FindInterestedChildren (exevents.c:2104)
35772 by 0x4DDEFF: FindInterestedChildren (exevents.c:2104)
35773 by 0x4DDEFF: FindInterestedChildren (exevents.c:2104)
35774 by 0x4E0C6F: SendEventToAllWindows (exevents.c:2127)
35775 by 0x4E8301: XISendDeviceHierarchyEvent (xichangehierarchy.c:118)
35776 by 0x426F99: DisableDevice (devices.c:507)
35777 by 0x46BF72: xf86Wakeup (xf86Events.c:457)
35778 by 0x432ABA: WakeupHandler (dixutils.c:419)
35779 by 0x45B708: WaitForSomething (WaitFor.c:235)
35780 Uninitialised value was created by a stack allocation
35781 at 0x4E8190: XISendDeviceHierarchyEvent (xichangehierarchy.c:61)
35782
35783 Set the type of dummyDev to SLAVE. The jump listed above comes from a check
35784 to IsMaster() in GetEventMask() that would then set the
35785 XIAllMasterDevices mask.
35786 Hierarchy events can only be set for XIAllDevices so the above IsMaster()
35787 check had no effect and the device type doesn't really matter anyway beyond
35788 shuting up valgrind.
35789
35790 Also initialize dummyDev to 0 to ease future debugging.
35791
35792 X.Org Bug 36120 <http://bugs.freedesktop.org/show_bug.cgi?id=36120>
35793
35794 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35795
35796commit 419a27b5219a739f2fbd50cc96a1b54c469e4a88
35797Author: Peter Hutterer <peter.hutterer@who-t.net>
35798Date: Tue Apr 12 10:17:07 2011 +1000
35799
35800 Xi: fix valuator alignment in DeepCopyDeviceClasses (#36119)
35801
35802 commit 678f5396c91b3d0c7572ed579b0a4fb62b2b4655 only fixed the
35803 initialization, not the copy. After a slave device change, the valuator
35804 were out of alignment again.
35805
35806 X.Org Bug 36119 <http://bugs.freedesktop.org/show_bug.cgi?id=36119>
35807
35808 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35809 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35810
35811commit 7762de65e153790ee9f63903964d168a6680d815
35812Author: Alexandr Shadchin <alexandr.shadchin@gmail.com>
35813Date: Sat Apr 9 17:45:10 2011 +0600
35814
35815 Simplify auto-detect mouse for WSCONS_SUPPORT
35816
35817 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
35818 Reviewed-by: Matthieu Herrbb <matthieu.herrb@laas.fr>
35819 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
35820
35821commit 001b6b8b70734db1fa2f68e45c1db8337ba9f662
35822Author: Adam Jackson <ajax@redhat.com>
35823Date: Fri Apr 15 13:01:37 2011 -0400
35824
35825 glx: Make --disable-dri not disable AIGLX
35826
35827 Either the DRI1 or DRI2 loaders are sufficient.
35828
35829 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
35830 Signed-off-by: Adam Jackson <ajax@redhat.com>
35831
35832commit f8540b9dcc1fba886be5c4ce7ea0d74952cb48c4
35833Author: Adam Jackson <ajax@redhat.com>
35834Date: Fri Apr 15 12:57:05 2011 -0400
35835
35836 glx: Use 0 rather than garbage for unknown INTEL_swap_event types
35837
35838 Otherwise the garbage you return could well be numerically identical to
35839 one of the swap type tokens, and apps which rely on us to tell the truth
35840 would be in trouble.
35841
35842 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
35843 Signed-off-by: Adam Jackson <ajax@redhat.com>
35844
35845commit dc0cf7596782087bdda0e7f9cd2f60907c45b2c4
35846Author: Adam Jackson <ajax@redhat.com>
35847Date: Tue Mar 29 10:09:46 2011 -0400
35848
35849 Revert "composite: Don't backfill non-bg-None windows"
35850
35851 This reverts commit 6dd775f57d2f94f0ddaee324aeec33b9b66ed5bc.
35852
35853 Bugzilla: https://bugs.freedesktop.org/34427
35854
35855 Acked-by: Alex Deucher <alexdeucher@gmail.com>
35856 Signed-off-by: Adam Jackson <ajax@redhat.com>
35857
35858commit 6a433b67ca15fd1ea58334e607f867554f227451
35859Author: Adam Jackson <ajax@redhat.com>
35860Date: Mon Mar 28 12:30:09 2011 -0400
35861
35862 glx: Fix lifetime tracking for pixmaps
35863
35864 GLX pixmaps take a reference on the underlying pixmap; X and GLX pixmap
35865 IDs can be destroyed in either order with no error. Only windows need
35866 to be tracked under both XIDs.
35867
35868 Fixes piglit/glx-pixmap-life.
35869
35870 Reviewed-by: Michel Dänzer <michel@daenzer.net>
35871 Signed-off-by: Adam Jackson <ajax@redhat.com>
35872
35873commit b3d2164a0361f636bfe77b51456bee9213af4f13
35874Author: Ville Syrjälä <ville.syrjala@nokia.com>
35875Date: Tue Apr 12 17:16:50 2011 +0300
35876
35877 dri2: Pass out_count by value to update_dri2_drawable_buffers()
35878
35879 update_dri2_drawable_buffers() doesn't modify out_count, so pass it
35880 by value.
35881
35882 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
35883 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
35884 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
35885
35886commit 93c833ee84a3465ec5d251e622ba26434cb532f8
35887Author: Ville Syrjälä <ville.syrjala@nokia.com>
35888Date: Tue Apr 12 17:13:28 2011 +0300
35889
35890 dri2: Handle calloc() failure
35891
35892 Don't access invalid memory if calloc() fails to allocate the buffers
35893 array.
35894
35895 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
35896 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
35897 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
35898
35899commit b2997431fd426ab318bc5dfd2cd43956d733ebec
35900Author: Jeremy Huddleston <jeremyhu@apple.com>
35901Date: Wed Apr 13 11:51:30 2011 -0700
35902
35903 Send events that were missing from RRSelectInput
35904
35905 The RANDR spec (randrproto.txt) specifies that RRSelectInput will send out
35906 events corresponding to the event mask, if there have been changes to
35907 CRTCs or outputs. Only screen events were being generated, however.
35908
35909 Fixes http://bugs.freedesktop.org/21760
35910
35911 Signed-off-by: Federico Mena Quintero <federico@novell.com>
35912 Reviewd-by: Keith Packard <keithp@keithp.com>
35913 Signed-off-by: Keith Packard <keithp@keithp.com>
35914
35915commit e409fb32b97033718f270a273f29f24c0b562b84
35916Author: Erkki Seppälä <erkki.seppala@vincit.fi>
35917Date: Tue Apr 12 12:55:56 2011 +0300
35918
35919 damage: use DamageReportDamage for the initial borderClip damage report
35920
35921 Instead of using DamageDamageRegion for reporting the first (virtual)
35922 damage in ProcDamageCreate that covers the borderClip of the drawable
35923 window, use a function DamageReportDamage directly (previously called
35924 damageReportDamage). This avoids sending all other damage listeners a
35925 full window update when a new damage object is created.
35926
35927 As this patch makes DamageReportDamage a public interface, the
35928 function has been moved into the part of the file that contains all
35929 the other public functions. The function has not been otherwise
35930 modified.
35931
35932 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
35933 Reviewed-by: Keith Packard <keithp@keithp.com>
35934 Signed-off-by: Keith Packard <keithp@keithp.com>
35935
35936commit 274dca8f2c6707121d45df8015fe7eddb129dec9
35937Author: Tiago Vignatti <tiago.vignatti@nokia.com>
35938Date: Mon Apr 4 22:31:42 2011 +0300
35939
35940 dix: don't free stranger pointers inside AllocARGBCursor
35941
35942 This seems a good convention to follow: if pointers are allocate outside a
35943 given function, then free there as well when a failure occurs.
35944
35945 AllocARGBCursor and its callers were mixing up the freeing of resources and
35946 causing a particular double free inside TileScreenSaver (srcbits and mskbits).
35947
35948 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
35949 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
35950
35951commit f603061e9482ad5caf1975ba5395b3294852d072
35952Author: Tiago Vignatti <tiago.vignatti@nokia.com>
35953Date: Mon Apr 4 21:40:06 2011 +0300
35954
35955 os: fix use after free in EstablishNewConnections
35956
35957 In the case of failure on AllocNewConnection, new_trans_conn cannot be
35958 dereferenced because it's already freed. Swapping the order of this logic fix
35959 the changes introduced in 04956b80431169e0ae713a3e6ba4cdc157ce3a66.
35960
35961 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
35962 CC: Jeremy Huddleston <jeremyhu@freedesktop.org>
35963 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35964
35965commit 82498e3c2cce6f515063ecb4b6ae9303e828da00
35966Author: Tiago Vignatti <tiago.vignatti@nokia.com>
35967Date: Mon Apr 4 20:25:32 2011 +0300
35968
35969 xfree86: xv: set pointers to NULL in xf86XVFreeAdaptor
35970
35971 As a good practice and for eventual double frees.
35972
35973 The reason of this patch is due the resilience of xf86XVInitAdaptors, where
35974 for any adaptor failure it's able to keep trying registering the following
35975 ones.
35976
35977 I discussed briefly with Pauli and Ville about a bigger refactoring of such
35978 function, doing it in a way to return instantly when a failure happens; after
35979 all that's how mostly of the other driver functions work. Instead, we just
35980 thought that xf86XVInitAdaptors is wise and cool, and eventually other driver
35981 functions should be even following the main idea of resilience.
35982
35983 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
35984 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
35985
35986commit 81414c1c836ae30628606545edbf7392d9b3d009
35987Author: Tiago Vignatti <tiago.vignatti@nokia.com>
35988Date: Thu Mar 31 23:44:03 2011 +0300
35989
35990 xfree86: xv: fix double free in xf86XVFreeAdaptor
35991
35992 When xf86XVFreeAdaptor is called more than once in xf86XVInitAdaptors (it may,
35993 but not often), the conditional being changed in this patch will always take
35994 true path and will keep freeing pAdaptor->pAttributes, thus letting the system
35995 error-prone.
35996
35997 This patch fix such problem checking for a pointer instead the number of
35998 attributes. Such pointer will be deallocated when xf86XVFreeAdaptor is called
35999 first and will not let the code re-run in the following calls. This is a bit
36000 similar how the surroundings code is already doing.
36001
36002 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36003 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
36004
36005commit 74476b700f1e499a731ba2ddbba87b12b9b5139b
36006Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36007Date: Thu Mar 31 17:46:42 2011 +0300
36008
36009 xfree86: loader: use one exit code only for readability
36010
36011 No functional changes. Spaghetti code for the win! \o/
36012
36013 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36014 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36015
36016commit 7aad7dbf409eaa16a2c80d76ba2a42dc84501919
36017Author: Peter Hutterer <peter.hutterer@who-t.net>
36018Date: Thu Apr 7 08:43:13 2011 +1000
36019
36020 Xext: use EXT_MASK macro instead of manual & 0x7f
36021
36022 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36023 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
36024
36025commit b28a1af55cf1ad2a882cc8cd73b77341dec7ff5a
36026Author: Chase Douglas <chase.douglas@canonical.com>
36027Date: Wed Apr 6 14:51:45 2011 -0400
36028
36029 Fix unset valuator handling for XI 1.x valuator events again
36030
36031 Set the valuator values for unset masked absolute valuators in the
36032 internal device event. This ensures the values will always be correct in
36033 getValuatorEvents even if the device has been removed.
36034
36035 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
36036 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36037
36038commit 3085b178626c957854385c7a88a05ec3c7eb74f3
36039Merge: c9d89ce 3c45b59
36040Author: Keith Packard <keithp@keithp.com>
36041Date: Wed Apr 6 08:34:10 2011 -0700
36042
36043 Merge remote-tracking branch 'jturney/master'
36044
36045commit c9d89cec1407550cb2568f4cef146c93607bbae6
36046Merge: a52049d bc61787
36047Author: Keith Packard <keithp@keithp.com>
36048Date: Mon Apr 4 11:57:39 2011 -0700
36049
36050 Merge remote-tracking branch 'vignatti/for-keith'
36051
36052commit bc61787a20e7683cbc4dfa45fe855da98a8c0cd0
36053Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36054Date: Tue Mar 29 18:22:50 2011 +0300
36055
36056 render: fix memory leaks in ProcRenderCompositeGlyphs
36057
36058 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36059 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36060 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36061 Reviewed-by: Soren Sandmann <ssp@redhat.com>
36062
36063commit 45b6667b651a0a26b17f64c9e99d70784045e4bf
36064Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36065Date: Tue Mar 29 17:17:40 2011 +0300
36066
36067 xkb: fix fd leak in XkbDDXListComponent
36068
36069 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36070 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36071 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36072
36073commit 0b288c8738a97cf6aa3f36aa5c05e7ac2a5cbca8
36074Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36075Date: Tue Mar 29 16:54:14 2011 +0300
36076
36077 dix: fix memory leak in ProcListExtensions
36078
36079 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36080 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36081 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36082
36083commit 90e03ab026cd8ffdc6202e6b8bae119717ee8528
36084Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36085Date: Tue Mar 29 16:48:21 2011 +0300
36086
36087 mi: fix memory leak in miZeroLine
36088
36089 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36090 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36091 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36092
36093commit daae5e5de194757f7084f9b2b24353c34b961f19
36094Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36095Date: Mon Mar 28 20:13:11 2011 +0300
36096
36097 xi: fix memory leak in AddExtensionClient
36098
36099 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36100 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36101 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36102
36103commit c1875ac25b7b730d464c98c4c151c35efd64a562
36104Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36105Date: Mon Mar 28 20:09:04 2011 +0300
36106
36107 dix: fix memory leak in AllocShared
36108
36109 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36110 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36111 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36112
36113commit 6e5020f46165862ff09211d5425c8b6a13966303
36114Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36115Date: Mon Mar 28 20:07:06 2011 +0300
36116
36117 dix: fix memory leak in AllocPseudo
36118
36119 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36120 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36121 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36122
36123commit a6c71ce5d2d2fe89e07a2ef5041c915acc3dc686
36124Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36125Date: Mon Mar 28 19:21:28 2011 +0300
36126
36127 os: fix memory and fd leaks in Popen
36128
36129 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36130 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36131 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36132
36133commit bafec9a25efa902bef2a3730dc44dc50f0e45877
36134Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36135Date: Mon Mar 28 19:19:57 2011 +0300
36136
36137 os: use DebugF for debugging
36138
36139 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36140 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36141 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36142
36143commit 719b37c33a89d6ad86c79097e5c67b9b3cea5aac
36144Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36145Date: Tue Mar 29 20:04:24 2011 +0300
36146
36147 xfree86: fix memory leak in xf86LoadModules
36148
36149 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36150 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
36151 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36152 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36153
36154commit af054db005b48160b117dfd5bec5f821ee614ea9
36155Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36156Date: Tue Mar 29 19:51:23 2011 +0300
36157
36158 xfree86: fix bad free configInputDevices
36159
36160 introduced in 93ca526892c0d22afa05cce6496198c652043a19.
36161
36162 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36163 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
36164 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36165 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36166
36167commit ac2fac24d8baa56f369f4e54908bc00c2b73a60a
36168Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36169Date: Tue Mar 29 19:49:32 2011 +0300
36170
36171 xfree86: fix memory leaks in configLayout
36172
36173 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36174 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36175 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36176
36177commit bc123319206930638ef1217922824d586d4ada6d
36178Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36179Date: Fri Mar 25 22:21:21 2011 +0200
36180
36181 xfree86: fix memory leak in xf86ConfigFbEntity
36182
36183 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36184 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36185 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36186
36187commit 2aa935bc5cc1e2d5365a97b8c5bb3d33eb5fc758
36188Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36189Date: Fri Mar 25 22:10:55 2011 +0200
36190
36191 fb: fix memory leak in fbOverlayFinishScreenInit
36192
36193 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36194 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36195 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36196
36197commit 4d87606a0d0eb63458098028c300c39c6f1bd2bf
36198Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36199Date: Fri Mar 25 22:07:44 2011 +0200
36200
36201 Xi: fix memory leak in ProcXGetSelectedExtensionEvents
36202
36203 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36204 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36205 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36206
36207commit 12e46e83733b47d2704e1509960192365102af46
36208Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36209Date: Fri Mar 25 22:07:31 2011 +0200
36210
36211 dix: fix memory leak in SetDefaultFontPath
36212
36213 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36214 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36215 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36216
36217commit 7ae46b69ba3f05f46529131e6a864904967cde3a
36218Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36219Date: Fri Mar 25 21:33:15 2011 +0200
36220
36221 mi: fix memory leak in miFillUniqueSpanGroup
36222
36223 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36224 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36225 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36226
36227commit 623bb34342641811e2151232f1ab4e2ee3d6b871
36228Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36229Date: Fri Mar 25 19:27:17 2011 +0200
36230
36231 xi: fix memory leak in ProcXIQueryDevice
36232
36233 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36234 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36235 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36236
36237commit dce2f10cf7bef9929cefadb5088d5b66df43a865
36238Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36239Date: Fri Mar 25 19:21:03 2011 +0200
36240
36241 mi: fix memory leak in miInitVisuals
36242
36243 Free the pointers inside miInitVisuals, so the callers of this function
36244 (fboverlay.c and fbscreen.c) don't need to worry with deallocation in the case
36245 of failure.
36246
36247 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36248 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36249 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36250
36251commit 8ab92cd9822510f426d179a636ef34bb0ace3bb3
36252Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36253Date: Fri Mar 25 18:56:02 2011 +0200
36254
36255 xfree86: dri2: fix memory leak and free resources properly
36256
36257 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36258 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36259 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36260
36261commit aa7096ca6f108e399d9916639cf20c57f9776305
36262Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36263Date: Fri Mar 25 18:24:34 2011 +0200
36264
36265 xorg: remove unused pointer values all over the server
36266
36267 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36268 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36269 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36270
36271commit 49051355d47849b3dd73631e3e2287e319a7f4e5
36272Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36273Date: Wed Mar 23 18:08:29 2011 +0200
36274
36275 dix: remove unused macro
36276
36277 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36278 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36279 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36280
36281commit f1a80e1268fa4d235a411a8cdf20b4f13eafcde0
36282Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36283Date: Wed Mar 23 18:06:51 2011 +0200
36284
36285 dix: remove unused debug code
36286
36287 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36288 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
36289 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36290 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36291
36292commit a52049de2f846fe984d4db5ac8d2c1826c7b2d0b
36293Merge: d044d36 266ea63
36294Author: Peter Hutterer <peter.hutterer@who-t.net>
36295Date: Mon Apr 4 09:58:53 2011 +1000
36296
36297 Merge branch 'master' of git://people.freedesktop.org/~herrb/xserver into for-keith
36298
36299commit d044d3675635f037bf0eb30e47f82460f78227d1
36300Author: Tiago Vignatti <tiago.vignatti@nokia.com>
36301Date: Thu Mar 31 16:26:06 2011 +0300
36302
36303 xfree86: loader: fix memory leaks in LoaderListDirs
36304
36305 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
36306 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36307 Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org>
36308 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36309
36310commit 9c4aae2141161e4bf69313a771db91c0acc4cc83
36311Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
36312Date: Wed Mar 30 16:47:31 2011 +0300
36313
36314 xkb: Prevent leaking of XKB geometry information on copy.
36315
36316 Currently shapes, sections and doodads may leak on copy.
36317
36318 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36319 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
36320 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
36321 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36322
36323commit 29d63ba175ff1ef1587c390b18ce61c8f1c150f3
36324Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
36325Date: Wed Mar 30 16:47:30 2011 +0300
36326
36327 xkb: Introduce helper function to handle similar reallocations.
36328
36329 This is preparation for a memory leak fix and doesn't contain any
36330 functional changes.
36331
36332 Note that two variables are generally used for reallocation and
36333 clearing of arrays: geom->sz_elems (reallocation) and geom->num_elems
36334 (clearing). The interface of XkbGeomRealloc is deliberately kept
36335 simple and it only accepts geom->sz_elems as argument, because that is
36336 needed to determine whether the array needs to be resized. When the
36337 array is cleared, we just assume that either geom->sz_elems and
36338 geom->num_elems are synchronized to be equal or that unused elements
36339 are cleared whenever geom->num_elems is set to be less than
36340 geom->sz_elems without reallocation.
36341
36342 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
36343 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
36344 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36345 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36346
36347commit f40103cee1d591387359f401a5a7c21f4105aeb4
36348Author: Chase Douglas <chase.douglas@canonical.com>
36349Date: Thu Mar 31 11:29:01 2011 -0400
36350
36351 Don't report old relative values in getValuatorEvents
36352
36353 Relative valuator values should not be reported in any future events. If
36354 a relative valuator value is not set in an internal event, set the value
36355 to 0 for XI 1.x valuator events sent over the wire.
36356
36357 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
36358 Reviewed-by: Simon Thum <simon.thum@gmx.de>
36359 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36360 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36361
36362commit 8199eac443d2c22d313cb23e39d5e607a8cc7f99
36363Author: Chase Douglas <chase.douglas@canonical.com>
36364Date: Mon Mar 28 16:04:48 2011 -0400
36365
36366 Handle non continuous valuator data in getValuatorEvents
36367
36368 This allows for masked valuators to be handled properly in XI 1.x
36369 events. Any unset valuators in the device event are set to the last
36370 known value when transmitted on the wire through XI 1.x valuator events.
36371
36372 Fixes https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/736500
36373
36374 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
36375 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36376 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36377
36378commit 266ea63bc3e052b753c5484fa08dcc4fb67c8952
36379Author: Matthieu Herrb <matthieu.herrb@laas.fr>
36380Date: Fri Apr 1 22:55:57 2011 +0200
36381
36382 Xi: fix querydevice request swapping
36383
36384 WriteReplyToClient() swaps rep.length, so it can't be used
36385 on return of WriteReplyToClient(). So save it's value for later
36386 use.
36387
36388 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
36389 Reviewed-by: Julien Cristau <jcristau@debian.org>
36390
36391commit 552c78592c5e5f52fe16e2429f77c63adf398247
36392Author: Matthieu Herrb <matthieu.herrb@laas.fr>
36393Date: Fri Apr 1 22:54:54 2011 +0200
36394
36395 Xext: fix test on extension number for the swapped case.
36396
36397 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
36398 Reviewed-by: Julien Cristau <jcristau@debian.org>
36399
36400commit a074e6b6754d6c2706b2b5de54e22cbffc5b1a17
36401Author: Matthieu Herrb <matthieu.herrb@laas.fr>
36402Date: Fri Apr 1 22:53:40 2011 +0200
36403
36404 Xi: add XI_Focus{In,Out} to swapped events.
36405
36406 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
36407 Reviewed-by: Julien Cristau <jcristau@debian.org>
36408
36409commit 81257377a2d011ce47fba5822df0f7918dea1d72
36410Author: Matthieu Herrb <matthieu.herrb@laas.fr>
36411Date: Fri Apr 1 22:51:48 2011 +0200
36412
36413 Xi: take XI2 requests into account also for the swapping case.
36414
36415 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
36416 Reviewed-by: Julien Cristau <jcristau@debian.org>
36417
36418commit f9834d312e3059073e8ad77d9f9d57cb9d96e1e5
36419Author: Dave Airlie <airlied@redhat.com>
36420Date: Thu Mar 31 15:33:46 2011 +1000
36421
36422 fb: cleanup fbChangeWindowAttributes
36423
36424 This cleans up the duplication in fbChangeWindowAttributes,
36425 and fixes a bug if the fb24_32ReformatTile ever failed,
36426 since the old code would happily dereference it in the fbEvenTile
36427 call a few lines later.
36428
36429 Signed-off-by: Dave Airlie <airlied@redhat.com>
36430 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
36431
36432commit 36c7158133660520034d645b124c2c973d2971bb
36433Author: Dave Airlie <airlied@redhat.com>
36434Date: Thu Mar 31 13:21:44 2011 +1000
36435
36436 fb: drop comments around 24-bit support.
36437
36438 Signed-off-by: Dave Airlie <airlied@redhat.com>
36439 Reviewed-by: Keith Packard <keithp@keithp.com>
36440
36441commit ac2ae2ed378cc877444bec17b353dda83b620186
36442Author: Dave Airlie <airlied@redhat.com>
36443Date: Tue Mar 29 14:24:52 2011 +1000
36444
36445 fb: add back fb defines for driver compat.
36446
36447 we could drop these really and just fixup the drivers, but
36448 since they'll build fine but fail to work this seems safer.
36449
36450 Signed-off-by: Dave Airlie <airlied@redhat.com>
36451 Reviewed-by: Keith Packard <keithp@keithp.com>
36452
36453commit be9cfb249da18d134c1d65bbb75161aff65f2942
36454Author: Dave Airlie <airlied@redhat.com>
36455Date: Tue Mar 29 14:27:41 2011 +1000
36456
36457 fb: drop defines that aren't used/necessary anymore.
36458
36459 This was generated by:
36460 cd fb
36461 coan source --replace -DFB_SCREEN_PRIVATE -DFB_24BIT -DFB_24_32BIT -DFB_SCREEN_PRIVATE -UFBNOPIXADDR -UFBNO24BIT -UFBNO24_32 *.[ch]
36462
36463 A follow up patch readds the FB_24_32BIT define for Intel UXA.
36464
36465 Signed-off-by: Dave Airlie <airlied@redhat.com>
36466 Reviewed-by: Keith Packard <keithp@keithp.com>
36467
36468commit e0a2ad51dfb7373aa602335490d9666d6101b5ea
36469Merge: 327e1d8 3d68831
36470Author: Keith Packard <keithp@keithp.com>
36471Date: Wed Mar 30 10:51:27 2011 -0700
36472
36473 Merge remote-tracking branch 'ajax/xserver-next'
36474
36475commit ac00ab77d5a00cfd198958aa1afaa4c3ccc6d7bc
36476Author: Chase Douglas <chase.douglas@canonical.com>
36477Date: Mon Mar 28 16:04:47 2011 -0400
36478
36479 Clean up getValuatorEvents using array loop logic
36480
36481 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
36482 Reviewed-by: Jamey Sharp <jamey@minilop.net>
36483 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36484 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36485
36486commit 3d688316af0e4ff3c44acfc168bd9c76c892b629
36487Author: Adam Jackson <ajax@redhat.com>
36488Date: Tue Mar 15 11:03:26 2011 -0400
36489
36490 xfree86: warning fix
36491
36492 xf86RandR12.c: In function 'xf86RandR12EnterVT':
36493 xf86RandR12.c:1769:5: warning: ISO C90 forbids mixed declarations and code
36494
36495 Reviewed-by: Keith Packard <keithp@keithp.com>
36496 Signed-off-by: Adam Jackson <ajax@redhat.com>
36497
36498commit 2762eef8c34b48be803226f48fc71f790286f968
36499Author: Adam Jackson <ajax@redhat.com>
36500Date: Tue Mar 15 11:01:50 2011 -0400
36501
36502 xfree86: warning fix
36503
36504 Pointer.c: In function 'xf86parsePointerSection':
36505 Pointer.c:192:5: warning: format '%u' expects type 'unsigned int', but argument 3 has type 'long unsigned int'
36506
36507 Reviewed-by: Keith Packard <keithp@keithp.com>
36508 Signed-off-by: Adam Jackson <ajax@redhat.com>
36509
36510commit 5275fbe2e226e1c2a46d26288390c8ef6e1a6a52
36511Author: Adam Jackson <ajax@redhat.com>
36512Date: Tue Mar 15 10:58:36 2011 -0400
36513
36514 render: warning fixes
36515
36516 picture.c: In function 'CompositeTriStrip':
36517 picture.c:1777:25: warning: unused variable 'ps'
36518 picture.c: In function 'CompositeTriFan':
36519 picture.c:1807:16: warning: unused variable 'pScreen'
36520
36521 Reviewed-by: Keith Packard <keithp@keithp.com>
36522 Signed-off-by: Adam Jackson <ajax@redhat.com>
36523
36524commit f1c2c01ac9ea7381cb4cd21ac1b2185ff28da3ec
36525Author: Adam Jackson <ajax@redhat.com>
36526Date: Thu Mar 24 16:51:54 2011 -0400
36527
36528 dix: Small refactor to resource deletion
36529
36530 Reviewed-by: Matt Turner <mattst88@gmail.com>
36531 Signed-off-by: Adam Jackson <ajax@redhat.com>
36532
36533commit 3e0b311aba54895d806174526511ef11b31432b1
36534Author: Adam Jackson <ajax@redhat.com>
36535Date: Mon Mar 21 12:12:53 2011 -0400
36536
36537 glx: minor cast cleanup
36538
36539 Reviewed-by: Eric Anholt <eric@anholt.net>
36540 Signed-off-by: Adam Jackson <ajax@redhat.com>
36541
36542commit 501506cf9df3d81556d44d074d8321fe6c171576
36543Author: Adam Jackson <ajax@redhat.com>
36544Date: Fri Mar 18 17:48:51 2011 -0400
36545
36546 glx: dead variable removal
36547
36548 Reviewed-by: Eric Anholt <eric@anholt.net>
36549 Signed-off-by: Adam Jackson <ajax@redhat.com>
36550
36551commit 10317682e2fbb95f1d4caa1bdbbb8116272c0f86
36552Author: Adam Jackson <ajax@redhat.com>
36553Date: Fri Mar 18 17:44:11 2011 -0400
36554
36555 glx: Remove some obfuscatory macros
36556
36557 Reviewed-by: Eric Anholt <eric@anholt.net>
36558 Signed-off-by: Adam Jackson <ajax@redhat.com>
36559
36560commit c327d07cc68802256c679a65d78f98f17327abda
36561Author: Adam Jackson <ajax@redhat.com>
36562Date: Fri Mar 18 17:03:09 2011 -0400
36563
36564 glx: Remove ->forceCurrent from the context vtable
36565
36566 All the implementations of makeCurrent and forceCurrent are identical,
36567 so just use makeCurrent everywhere.
36568
36569 Reviewed-by: Eric Anholt <eric@anholt.net>
36570 Signed-off-by: Adam Jackson <ajax@redhat.com>
36571
36572commit a48dadc98a28c969741979b70b7a639f24f4cbbd
36573Author: Adam Jackson <ajax@redhat.com>
36574Date: Mon Mar 21 11:59:29 2011 -0400
36575
36576 glx: Reimplement context tags
36577
36578 This would let you do a constant-time context lookup, but if that's your
36579 performance problem you have two problems. Just use the context's XID
36580 as the tag value instead.
36581
36582 In order to do this, we have to defer destroying a context until it
36583 actually goes unreferenced, as you're allowed to mention a context tag
36584 after you've (ostensibly) destroyed the context, as long as it's still
36585 your current context. Thus, change DestroyContext to merely mark the
36586 context as dead if it's a current context, and call down to actual
36587 resource destruction (and XID reclamation) in StopUsingContext.
36588
36589 Also, stop trying to delete context state from DrawableGone. This was
36590 always broken, as GLX does not say that contexts are destroyed when
36591 their drawables are destroyed. But with the above change to defer
36592 context destruction, this would trigger a server crash on client exit as
36593 we'd free the context state twice.
36594
36595 Reviewed-by: Eric Anholt <eric@anholt.net>
36596 Signed-off-by: Adam Jackson <ajax@redhat.com>
36597
36598commit 30d6947cee714385cf62a40ea6fa6d2e68388c78
36599Author: Adam Jackson <ajax@redhat.com>
36600Date: Mon Mar 7 14:53:28 2011 -0500
36601
36602 glx: Fix _glapi_add_dispatch
36603
36604 We never need to generate stubs, because those conditions can't happen
36605 in the server. Yank that code out, but keep the bookkeeping for which
36606 extension functions are registered so the DRI driver doesn't get
36607 confused.
36608
36609 As a pleasant bonus, we're now friendlier for environments like selinux
36610 that make runtime code generation difficult, and we're portable to more
36611 arches since we don't have to port the assembly stubs.
36612
36613 Fixes the following clutter conformance tests (indirect rendering,
36614 llvmpipe driver):
36615
36616 test-cogl-backface-culling
36617 test-cogl-materials
36618 test-cogl-readpixels
36619 test-cogl-texture-mipmaps
36620 test-cogl-texture-get-set-data
36621 test-cogl-viewport
36622 test-cogl-offscreen
36623
36624 Reviewed-by: Dave Airlie <airlied@redhat.com>
36625 Signed-off-by: Adam Jackson <ajax@redhat.com>
36626
36627commit 0cb44cec7b9fbd75285f7380f162cf6140a5bf3f
36628Author: Adam Jackson <ajax@redhat.com>
36629Date: Thu Mar 3 10:42:41 2011 -0500
36630
36631 glx: Flatten -DXFree86Server
36632
36633 Always defined by the makefile, so, just get rid of it.
36634
36635 Reviewed-by: Dave Airlie <airlied@redhat.com>
36636 Signed-off-by: Adam Jackson <ajax@redhat.com>
36637
36638commit ed6c13c01e017f9f40c89247ab6652895fb1ea14
36639Author: Adam Jackson <ajax@redhat.com>
36640Date: Wed Mar 2 13:37:43 2011 -0500
36641
36642 glx: Update some glapi comments
36643
36644 Reviewed-by: Dave Airlie <airlied@redhat.com>
36645 Reviewed-by: Julien Cristau <jcristau@debian.org>
36646 Signed-off-by: Adam Jackson <ajax@redhat.com>
36647
36648commit 17d9e374721d6c8ee3f7f9cdc882f80127bdb57f
36649Author: Adam Jackson <ajax@redhat.com>
36650Date: Wed Mar 2 13:21:39 2011 -0500
36651
36652 glx: Lobotomize _glapi_get_proc_address
36653
36654 This isn't a meaningful thing in the indirect glx loader, so just warn
36655 if it ever happens and move on.
36656
36657 But also, mark it PUBLIC, so if the driver does ever call it we merely
36658 warn instead of aborting because ld.so can't find the symbol.
36659
36660 Reviewed-by: Dave Airlie <airlied@redhat.com>
36661 Signed-off-by: Adam Jackson <ajax@redhat.com>
36662
36663commit b0c665ac0fe6840dda581e4d0d0b76c703d62a7b
36664Author: Adam Jackson <ajax@redhat.com>
36665Date: Wed Mar 2 12:30:22 2011 -0500
36666
36667 glx: Remove noop dispatch table
36668
36669 We can never hit this, because the indirect GLX dispatch code always
36670 forces a current context and checks that it's non-NULL before calling
36671 into the dispatch table. If it's _not_ null, then _glapi_set_context
36672 will call into the driver, which is responsible for calling
36673 _glapi_set_dispatch to make sure the dispatch table is non-NULL.
36674
36675 Also remove _glapi_set_warning_func and friends, since we can no longer
36676 call them even from dead code.
36677
36678 Reviewed-by: Dave Airlie <airlied@redhat.com>
36679 Signed-off-by: Adam Jackson <ajax@redhat.com>
36680
36681commit 327e1d88012102af6aca6c6840aa0ed3c7041a77
36682Author: Michel Dänzer <daenzer@vmware.com>
36683Date: Mon Mar 28 17:18:39 2011 +0200
36684
36685 EXA: Use dixGetPrivate(Addr) instead of dixLookupPrivate.
36686
36687 The latter calls the former, let's cut the middle man and eliminate a branch
36688 in a hot path.
36689
36690 According to Git history, ExaSetPixmapPriv was never used anywhere, just drop
36691 it.
36692
36693 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
36694 Reviewed-by: Cyril Brulebois <kibi@debian.org>
36695 Signed-off-by: Keith Packard <keithp@keithp.com>
36696
36697commit a095a6d4e8f5090907e8d3d66018636216300846
36698Merge: a22486f ef9d04f
36699Author: Keith Packard <keithp@keithp.com>
36700Date: Sun Mar 27 20:06:29 2011 -0700
36701
36702 Merge remote-tracking branch 'airlied/pwin-cleanup'
36703
36704commit a22486f848014000dc13dda470f77d4d8ea9e9f6
36705Merge: 2ef4ff4 633b81e
36706Author: Keith Packard <keithp@keithp.com>
36707Date: Sun Mar 27 18:27:10 2011 -0700
36708
36709 Merge remote-tracking branch 'whot/for-keith'
36710
36711commit 2ef4ff45ef1fcfc4967ebe3d550408769e5f6500
36712Author: Erkki Seppälä <erkki.seppala@vincit.fi>
36713Date: Fri Mar 25 10:38:23 2011 +0200
36714
36715 os/client: Prevent rare fd leak in DetermineClientPid
36716
36717 DetermineClientPid didn't close file descriptor if read on
36718 /proc/pid/cmdline failed. Adjusted the code to disregard the close
36719 return value and perform the return after that, if the read failed or
36720 returned EOF.
36721
36722 Signed-off-by: Mark Kettenis <mark.kettenis@xs4all.nl>
36723 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
36724 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
36725 Signed-off-by: Keith Packard <keithp@keithp.com>
36726
36727commit ef9d04f8ad4239ac30984769c9c8c101bb8e78bf
36728Author: Dave Airlie <airlied@redhat.com>
36729Date: Thu Mar 24 14:07:34 2011 +1000
36730
36731 dri: kill if 0 out code.
36732
36733 This refers to pWin->winSize in some #if 0 code remove it.
36734
36735 Signed-off-by: Dave Airlie <airlied@redhat.com>
36736 Reviewed-by: Cyril Brulebois <kibi@debian.org>
36737
36738commit eb9266c7176eb240a85e72beda9b6033338294f1
36739Author: Dave Airlie <airlied@redhat.com>
36740Date: Thu Mar 24 14:04:25 2011 +1000
36741
36742 consolidate SetRootClip (v2)
36743
36744 each DDX has its own copy, I've taken the darwin one,
36745 though I'm not sure why it needs the pOldClip piece that nobody
36746 else has and the commit msg is like an "Updates from magic land"
36747 type message.
36748
36749 This removes the main uses of pWin->winSize from the DDXen.
36750
36751 v2: drop old clip like ajax suggests.
36752
36753 Signed-off-by: Dave Airlie <airlied@redhat.com>
36754 Reviewed-by: Adam Jackson <ajax@redhat.com>
36755
36756commit 3c45b59e675ba926ec95842ac3d49e2091e41eb8
36757Author: Jon TURNEY <jon.turney@dronecode.org.uk>
36758Date: Fri Mar 18 15:00:14 2011 +0000
36759
36760 Fix XWin compilation after commit 769531b9
36761
36762 commit 769531b9 "Add mode field to pointer movement hooks" changes the
36763 function signature of miPointerSetPosition() to include the movement mode
36764 which resulted in the pointer position
36765
36766 Update use of miPointerSetPosition() in winEnqueueMotion() appropriately
36767
36768 (See http://tinderbox.freedesktop.org/builds/2011-03-16-0008/logs/xserver/#build)
36769
36770 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
36771 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36772
36773commit 633b81e8ba09cc6a1ea8b43f323874fda2cf0bde
36774Author: Simon Thum <simon.thum@gmx.de>
36775Date: Wed Mar 9 20:48:22 2011 +0100
36776
36777 xserver: remove AbsoluteClassRec keeping the ABI
36778
36779 This removes the struct, but keeps InitAbsoluteClassDeviceStruct as
36780 a no-op and preserves related struct layout.
36781
36782 Signed-off-by: Simon Thum <simon.thum@gmx.de>
36783 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36784 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36785
36786commit 2103d61909c7e6bd345622962df7784a19df72c5
36787Author: Simon Thum <simon.thum@gmx.de>
36788Date: Sat Mar 5 05:02:54 2011 +0100
36789
36790 simplify ChangeDeviceControl in stubs
36791
36792 Signed-off-by: Simon Thum <simon.thum@gmx.de>
36793 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36794 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36795
36796commit 118ef6f806f629a8e2ad61e4f2a43820b0fa621c
36797Author: Simon Thum <simon.thum@gmx.de>
36798Date: Sat Mar 5 05:01:22 2011 +0100
36799
36800 xf86: don't pretend to support DEVICE_ABS_* in ChangeDeviceControl
36801
36802 Signed-off-by: Simon Thum <simon.thum@gmx.de>
36803 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36804 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36805
36806commit dee83dff4b8fbf1351263beaf2d531d033b6d0c4
36807Author: Simon Thum <simon.thum@gmx.de>
36808Date: Sat Mar 5 04:59:33 2011 +0100
36809
36810 kdrive: don't pretent to support DEVICE_ABS_* in ChangeDeviceControl
36811
36812 Signed-off-by: Simon Thum <simon.thum@gmx.de>
36813 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36814 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36815
36816commit 9eaecb1bfe358a0c5f24645a29d6c377c0c4b37c
36817Author: Simon Thum <simon.thum@gmx.de>
36818Date: Sun Mar 20 16:28:22 2011 +0100
36819
36820 xquartz: simplify ChangeDeviceControl
36821
36822 Signed-off-by: Simon Thum <simon.thum@gmx.de>
36823 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36824 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36825
36826commit 33e257ab43ebf35e838b80c416762e9b743e9385
36827Author: Peter Hutterer <peter.hutterer@who-t.net>
36828Date: Tue Mar 22 11:53:00 2011 +1000
36829
36830 test: when unit tests are enabled, build them during "make"
36831
36832 Catch compiler errors that were otherwise only spotted on make check.
36833
36834 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36835 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
36836
36837commit 6243332782f2d286d49f1dec8a628b0a403665fd
36838Author: Erkki Seppälä <erkki.seppala@vincit.fi>
36839Date: Fri Mar 18 16:35:36 2011 +0200
36840
36841 config: handle device change event properly
36842
36843 wakeup_handler in udev.c wasn't dealing with udev change events.
36844 There are situations when a device can gain its input capabilities
36845 after it has been added to the system and therefore the change events
36846 must be handled as well.
36847
36848 The change is handled as a consecutive device removal and addition.
36849
36850 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
36851 Signed-off-by: Stefan Kost <Stefan.Kost@nokia.com>
36852 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36853 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36854
36855commit fad10cb38ef2433c82f8a43d3b7d64e323728060
36856Author: Peter Hutterer <peter.hutterer@who-t.net>
36857Date: Fri Mar 18 11:06:57 2011 +1000
36858
36859 xfree86: print out which driver is about to be used.
36860
36861 Makes reading the log file a lot easier for those that don't magically
36862 recognise the log spew by the individual drivers.
36863
36864 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
36865 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
36866
36867commit cb5d4b416a0fc850a1c119755a9cc3c4f0d7a545
36868Author: Jeremy Huddleston <jeremyhu@apple.com>
36869Date: Wed Mar 23 12:10:15 2011 -0700
36870
36871 XQuartz: applewm: Don't check if requested window level is < 0 because it is unsigned (-Wtautological-compare)
36872
36873 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
36874
36875commit c116e32dc2799c4abcf38d22bc773b755387c89a
36876Author: Jeremy Huddleston <jeremyhu@apple.com>
36877Date: Wed Mar 23 12:07:36 2011 -0700
36878
36879 XQuartz: Put ifdef guards around have_depth label (-Wunused-label)
36880
36881 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
36882
36883commit 1a583f7940e18a8ef28ca14e74e85e0785d39067
36884Author: Jeremy Huddleston <jeremyhu@apple.com>
36885Date: Wed Mar 23 12:05:05 2011 -0700
36886
36887 XQuartz: Properly comment extra tokens (-Wextra-tokens) after endif
36888
36889 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
36890
36891commit c55baebf4ebf1887262cc16899eb297b9f284f6e
36892Author: Jeremy Huddleston <jeremyhu@apple.com>
36893Date: Tue Mar 22 19:01:48 2011 -0700
36894
36895 GLX: Support TLS with better portability
36896
36897 AX_TLS detects when toolchains support __thread or __declspec(thread),
36898 but existing code assumed __thread.
36899
36900 This also adds a check to configure.ac to error out if TLS is requested
36901 but unsupported.
36902
36903 Found-by: Tinderbox
36904 http://tinderbox.x.org/builds/2011-03-22-0007
36905
36906 Regression-from: 82b1eaa6cad20f39dbf15573bdb3d62acbcd91f9
36907
36908 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
36909 Reviewed-by: Tom Fogal <tfogal@alumni.unh.edu>
36910 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
36911
36912commit 03f45df93469f6aef391e97007b9614e0770cc4c
36913Merge: efcb727 5fb329a
36914Author: Keith Packard <keithp@keithp.com>
36915Date: Wed Mar 23 13:38:37 2011 +0900
36916
36917 Merge remote-tracking branch 'airlied/xinerama-cleanup'
36918
36919commit efcb7275ce5de651f91ba4ff8bb227dfb68bb154
36920Author: Gaetan Nadon <memsize@videotron.ca>
36921Date: Thu Mar 17 19:26:37 2011 -0400
36922
36923 test: git ignore the list test executable
36924
36925 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36926 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
36927 Signed-off-by: Keith Packard <keithp@keithp.com>
36928
36929commit 6a5bf15fa99cf5b2358b3b3e2f29e5044aa8724a
36930Author: Gaetan Nadon <memsize@videotron.ca>
36931Date: Thu Mar 17 19:26:36 2011 -0400
36932
36933 Add generalized unit test support using util-macros.
36934
36935 A handful of modules have begun adding unit test programs.
36936 These macros will help providing a consistent interface which will
36937 help package builders and developers to manage the functionality.
36938
36939 XORG_ENABLE_UNIT_TESTS will turn on/off unit testing, regardless
36940 of how it is implemented. The default (yes/no) can be specified by each
36941 module. It can be used by itself if glib or -wrap support is not needed.
36942
36943 XORG_WITH_GLIB will probe the system for glib-2.0. A different version
36944 can be specified in each module. It will consult XORG_ENABLE_UNIT_TESTS
36945 but can be used by itself in contexts other then unit testing.
36946 The default (yes/no) can be specified by each module.
36947
36948 XORG_LD_WRAP will probe the linker for -wrap support. It will consult
36949 XORG_ENABLE_UNIT_TESTS but can be used by itself in contexts
36950 other then unit testing.
36951
36952 configure options:
36953 --enable-unit-tests Enable building unit test cases (default: auto)
36954 --with-glib Use GLib library for unit testing (default: auto)
36955
36956 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36957 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
36958 Signed-off-by: Keith Packard <keithp@keithp.com>
36959
36960commit ee583cb33423fa79beb22db20b30e10a677f9b5a
36961Author: Gaetan Nadon <memsize@videotron.ca>
36962Date: Thu Mar 17 19:26:35 2011 -0400
36963
36964 config: group document related XORG_ macros together
36965
36966 No functional changes.
36967
36968 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
36969 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
36970 Signed-off-by: Keith Packard <keithp@keithp.com>
36971
36972commit d5b16b037b8fe12ba85c68c8289b6a8cc5e3a09d
36973Author: Keith Packard <keithp@keithp.com>
36974Date: Thu Mar 17 23:48:52 2011 -0700
36975
36976 Revert "dix: Remove usage_hint from pixmaps, store it in ->drawable.class"
36977
36978 This reverts commit 1564c82417d201de5b9a5ec5e7aa4ef14c45fbad.
36979
36980 The drivers used the top bits of the usage_hint to store driver
36981 private flags (intel, radeon, nouveau).
36982
36983 With EXA we need to get at this data so if we migrate the pixmap we
36984 can create the correct type of pixmap in the driver, however this
36985 commit truncates the usage_hint into 8-bit class and loses all the
36986 good stuff.
36987
36988 Signed-off-by: Dave Airlie <airlied@gmail.com>
36989 Reviewed-by: Keith Packard <keithp@keithp.com>
36990
36991commit dc9ce695a69ca0787f58f8d160212a7a41acb703
36992Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
36993Date: Wed Mar 9 15:45:40 2011 +0200
36994
36995 xkb: Initialize pad bytes sent in replies of geometry requests.
36996
36997 Valgrind complains about uninitialized data being written to clients.
36998
36999 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37000 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37001 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37002 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37003 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37004
37005commit ee3a4951a49312a2b68025db00c71472eb84293f
37006Author: Peter Hutterer <peter.hutterer@who-t.net>
37007Date: Thu Mar 10 09:30:02 2011 +1000
37008
37009 xkb: Document XkbWriteCountedString.
37010
37011 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37012 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37013
37014commit 5c47f8beac7f87680c6f7331483b9cf94a1dbc86
37015Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
37016Date: Fri Mar 11 14:30:49 2011 +0200
37017
37018 xkb: Release XKB component names when compiling keymap.
37019
37020 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37021 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37022 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37023 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37024
37025commit 6f46ae3c69c48f389560af1491ed1729ba45866f
37026Author: Peter Hutterer <peter.hutterer@who-t.net>
37027Date: Tue Mar 15 13:20:03 2011 +1000
37028
37029 man: list the drivers that are ignored when hotplugging (#35209)
37030
37031 X.Org Bug 35209 <http://bugs.freedesktop.org/show_bug.cgi?id=35209>
37032
37033 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37034 Reviewed-by: Cyril Brulebois <kibi@debian.org>
37035 Reviewed-by: Julien Cristau <jcristau@debian.org>
37036
37037commit 642569591a73806d038f00f2f9a5c8738f764c0b
37038Author: Gaetan Nadon <memsize@videotron.ca>
37039Date: Sat Mar 12 07:29:35 2011 -0500
37040
37041 dmx/doc: remove dead code in the makefile
37042
37043 This was leftover from some older ways of building dmx/scale docbook.
37044
37045 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
37046 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
37047 Signed-off-by: Keith Packard <keithp@keithp.com>
37048
37049commit 21f70cad19474a6f3ba419e03c0df12881ed1092
37050Author: Gaetan Nadon <memsize@videotron.ca>
37051Date: Sat Mar 12 07:29:34 2011 -0500
37052
37053 man: relocate manual pages in the man subdir outside doc
37054
37055 The convention is to have the manual pages in a man subdir
37056 which is not under a doc dir. The doc dir contains users docs.
37057 This will move man pages out of the way for upcoming DocBook patches.
37058
37059 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
37060 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
37061 Signed-off-by: Keith Packard <keithp@keithp.com>
37062
37063commit d3adf2d9350bee4125107e2ea1ed0c51bb736562
37064Author: Erkki Seppälä <erkki.seppala@vincit.fi>
37065Date: Thu Mar 10 11:40:40 2011 +0200
37066
37067 xfree86/modes: Fixed memory leak in xf86InitialConfiguration
37068
37069 There were two memory leaks in the function: one was the lack of free
37070 for "enabled", the other was the full lack of releasing anything when
37071 configuration was too small. The first issue was fixed by adding the
37072 missing free, the other was addressed by replacing the duplicate
37073 memory releasing sequences with one that is gotoed into.
37074
37075 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37076 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37077 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37078 Signed-off-by: Keith Packard <keithp@keithp.com>
37079
37080commit a713832cbe494ff468a627e88b11164074bbd14e
37081Author: Jon TURNEY <jon.turney@dronecode.org.uk>
37082Date: Sun Mar 13 20:01:30 2011 +0000
37083
37084 When XDMCP -from is specified, only register the requested address
37085
37086 When XDMCP -from is specified, only register the requested address,
37087 rather than the requested address, and any others we have of different
37088 address families to the requested address.
37089
37090 e.g. if we have 4 interfaces with both IPv4 and IPv6 addresses (which
37091 are not IPv6 mapped IPV4 addresses), using -from with one of those IPv4
37092 addresses currently means only that IPv4 address, and all IPv6 addresses
37093 are used in the connection data in XDMCP REQUEST packet.
37094
37095 (See http://cygwin.com/ml/cygwin-xfree/2011-02/msg00000.html)
37096
37097 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
37098 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
37099 Signed-off-by: Keith Packard <keithp@keithp.com>
37100
37101commit 0ac4931753a5d5925fc844c8cbec08585aea57a7
37102Merge: a8146f6 d7f8011
37103Author: Keith Packard <keithp@keithp.com>
37104Date: Mon Mar 14 13:06:41 2011 -0700
37105
37106 Merge remote-tracking branch 'ajax/xserver-next'
37107
37108commit d7f8011418f9da06631f27c66c29bcb226d0dffe
37109Author: Søren Sandmann Pedersen <ssp@redhat.com>
37110Date: Tue Mar 8 10:14:28 2011 -0500
37111
37112 Remove TriStrip and TriFan from the picture screen
37113
37114 These functions no longer go through the screen vtable, so remove
37115 them and fix up the various wrappers.
37116
37117 Reviewed-by: Adam Jackson <ajax@redhat.com>
37118 Acked-by: Keith Packard <keithp@keithp.com>
37119 Signed-off-by: Soren Sandmann <ssp@redhat.com>
37120
37121commit 0eb5b0fbcf1233a93f285ff1e1609fcbd01e7c79
37122Author: Søren Sandmann Pedersen <ssp@redhat.com>
37123Date: Tue Mar 8 10:14:27 2011 -0500
37124
37125 Absorb miTriStrip() into CompositeTriStrip()
37126
37127 There is no need to virtualize this function that nobody cares about.
37128
37129 Reviewed-by: Adam Jackson <ajax@redhat.com>
37130 Acked-by: Keith Packard <keithp@keithp.com>
37131 Signed-off-by: Soren Sandmann <ssp@redhat.com>
37132
37133commit c2af0cea02bd85f4d5954c16e34b4a8fb0fe2243
37134Author: Søren Sandmann Pedersen <ssp@redhat.com>
37135Date: Tue Mar 8 10:14:26 2011 -0500
37136
37137 Absorb miTriFan() into CompositeTriFan()
37138
37139 There is no need to virtualize this function that nobody cares about.
37140
37141 Reviewed-by: Adam Jackson <ajax@redhat.com>
37142 Acked-by: Keith Packard <keithp@keithp.com>
37143 Signed-off-by: Soren Sandmann <ssp@redhat.com>
37144
37145commit a8146f6becc44bf9ad611d33bded17df07e6af21
37146Author: Erkki Seppälä <erkki.seppala@vincit.fi>
37147Date: Wed Mar 9 17:29:14 2011 +0200
37148
37149 mi/misprite: use memory management provided by dixRegisterPrivateKey
37150
37151 The record allocated by miSpriteDeviceCursorInitialize was not being
37152 released.
37153
37154 This patch makes misprite use dixRegisterPrivateKey with the record
37155 size argument, which handles the memory management
37156 issues. miSpriteDeviceCursorInitialize is restructured to initialize
37157 pCursorInfo only if miDCDeviceInitialize succeeds. The record itself
37158 is zeroed on cleanup to ensure that the assumptions in the code still
37159 hold.
37160
37161 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37162 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37163 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37164 Signed-off-by: Keith Packard <keithp@keithp.com>
37165
37166commit f985a7319ef80b9b613eeaf24581000827cb220f
37167Author: Erkki Seppälä <erkki.seppala@vincit.fi>
37168Date: Wed Mar 9 17:29:14 2011 +0200
37169
37170 mi/misprite: use memory management provided by dixRegisterPrivateKey
37171
37172 The record allocated by miSpriteDeviceCursorInitialize was not being
37173 released.
37174
37175 This patch makes misprite use dixRegisterPrivateKey with the record
37176 size argument, which handles the memory management issues.
37177 miSpriteDeviceCursorInitialize is restructured to initialize pCursorInfo
37178 only if miDCDeviceInitialize succeeds. The record itself is zeroed on
37179 cleanup to ensure that the assumptions in the code still hold.
37180
37181 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37182 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37183 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37184
37185commit 57b35adaed112520c3b3b2fbad13cf5a91cd6652
37186Author: Søren Sandmann Pedersen <ssp@redhat.com>
37187Date: Wed Mar 9 13:57:25 2011 -0500
37188
37189 Remove geometry arguments from miSourceValidate()
37190
37191 The only user of the geometry coordinates is the software sprite code,
37192 which uses them to remove the pointer whenever the window beneath is
37193 being used as a source. However, using Window pictures as a source is
37194 extremely rare (let alone *partial* windows), so there is no harm done
37195 in just validating all of the drawable.
37196
37197 Additionally, the miSourceValidate() function was buggy in at least
37198 three respects:
37199
37200 (a) It added drawable->{x,y} before calling down, which is wrong since
37201 the misprite code already adds them in its check. (Alternatively,
37202 the misprite code is wrong, but there are actual users who would
37203 notice if that code was broken).
37204
37205 (b) It didn't account for the width of the interpolation filter, so if
37206 the Picture had a bilinear or convolution filter, the edges
37207 surrounding the source area would not be validated.
37208
37209 (c) It didn't validate alpha maps.
37210
37211 Finally, computing the bounding box of the transform on every
37212 composite request was a real performance issue in pixman, so
37213 presumably it could be one here as well.
37214
37215 This patch changes miSourceValidate() to simply validate all of the
37216 underlying drawable.
37217
37218 Reviewed-by: Adam Jackson <ajax@redhat.com>
37219 Reviewed-by: Keith Packard <keithp@keithp.com>
37220 Signed-off-by: Soren Sandmann <ssp@redhat.com>
37221
37222commit 016edc17512ba966d60edede8cf947996bae0b3c
37223Author: Adam Jackson <ajax@redhat.com>
37224Date: Mon Feb 28 17:12:26 2011 -0500
37225
37226 dix: Define RESTYPE as uint32_t
37227
37228 long is needlessly long on LP64.
37229
37230 Reviewed-by: Dave Airlie <airlied@redhat.com>
37231 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37232 Signed-off-by: Adam Jackson <ajax@redhat.com>
37233
37234commit 1f2bc777f96fd41feb55a4799ece939652130ef4
37235Author: Adam Jackson <ajax@redhat.com>
37236Date: Mon Feb 28 13:11:12 2011 -0500
37237
37238 dix: Shrink PropertyRec on LP64
37239
37240 size needn't be a long. No change on ILP32 but, combined with the
37241 previous change, 56 -> 40 bytes on LP64.
37242
37243 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37244 Signed-off-by: Adam Jackson <ajax@redhat.com>
37245
37246commit 51f353d0a0d116af16d7d9590cadef6c56328746
37247Author: Adam Jackson <ajax@redhat.com>
37248Date: Mon Feb 28 13:10:20 2011 -0500
37249
37250 dix: Fix ATOM typedef
37251
37252 unsigned long is needlessly large on LP64. Use uint32_t instead.
37253
37254 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37255 Signed-off-by: Adam Jackson <ajax@redhat.com>
37256
37257commit 7ca75abbbdd2a1211e52a4f43ac4ed24d3c8ab34
37258Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
37259Date: Wed Mar 9 11:17:28 2011 +1100
37260
37261 glx: Use PATH_MAX as size of filename buffer
37262
37263 Reviewed-by: Adam Jackson <ajax@redhat.com>
37264 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
37265
37266commit d17a9fb8414becf6a8998041df68f209f9222b2b
37267Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
37268Date: Wed Mar 9 11:17:27 2011 +1100
37269
37270 Consolidate all the PATH_MAX handling into misc.h
37271
37272 Reviewed-by: Adam Jackson <ajax@redhat.com>
37273 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
37274
37275commit 021393d1b8bcc9ff2ff5deb2306360e6b0afa1c6
37276Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
37277Date: Wed Mar 9 11:15:07 2011 +1100
37278
37279 glx: Factor out glxProbeDriver function.
37280
37281 DRI, DRI2 and swrast all had near-identical driver probing logic.
37282 Pull it into glxdricommon.
37283
37284 [ajax: warning fix]
37285 Reviewed-by: Adam Jackson <ajax@redhat.com>
37286 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
37287
37288commit 56c90e29f04727c903bd0f084d23bf44eb1a0a11
37289Author: Adam Jackson <ajax@redhat.com>
37290Date: Mon Nov 15 14:29:14 2010 -0500
37291
37292 randr: Add RRConstrainCursorHarder
37293
37294 Confine cursor motion to within the bounds of a single CRTC, iff all the
37295 CRTCs within a ScreenRec are reachable from each other. If not you get
37296 the same "cursor floats within the bounding rect" behaviour you get now.
37297
37298 v3:
37299 - Incorporate review feedback from Christopher James Halse Rogers
37300 v4:
37301 - Add mode field.
37302
37303 Signed-off-by: Adam Jackson <ajax@redhat.com>
37304 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37305
37306commit 810fbfa44626bff9f443ab17c0ad27ff7ae121d7
37307Author: Adam Jackson <ajax@redhat.com>
37308Date: Wed Feb 9 17:32:16 2011 -0500
37309
37310 mi: Call pScreen->ConstrainCursorHarder from the position update path
37311
37312 v2: Cover more paths, spotted by Daniel Stone.
37313 v3: pass down the mode field for movement mode.
37314
37315 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37316 Signed-off-by: Adam Jackson <ajax@redhat.com>
37317 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37318
37319commit 769531b9ccade723a56498b0888af58d085fec9e
37320Author: Peter Hutterer <peter.hutterer@who-t.net>
37321Date: Fri Feb 18 14:19:18 2011 +1000
37322
37323 Add mode field to pointer movement hooks.
37324
37325 Preparation work for pointer barriers.
37326
37327 Reviewed-by: Adam Jackson <ajax@redhat.com>
37328 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37329
37330commit 92788e677be79bd04e5ef140f4ced50ad8b1bf8e
37331Author: Peter Hutterer <peter.hutterer@who-t.net>
37332Date: Tue Feb 22 12:32:01 2011 +1000
37333
37334 test: add some tests for basic list manipulation.
37335
37336 This has less purpose as a test but more as documentation on how to actually
37337 use the differnent list calls.
37338
37339 Reviewed-by: Adam Jackson <ajax@redhat.com>
37340 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37341
37342commit 1ad9f01c31742157934a791e6141d10520d13e8a
37343Author: Adam Jackson <ajax@redhat.com>
37344Date: Thu Mar 10 10:47:40 2011 -0500
37345
37346 glx: Add texbuffer2 support to swrast
37347
37348 Reviewed-by: Julien Cristau <jcristau@debian.org>
37349 Signed-off-by: Adam Jackson <ajax@redhat.com>
37350
37351commit 5fb329a04a18835ce864d0563f6dfeb3d3c78d69
37352Author: Dave Airlie <airlied@redhat.com>
37353Date: Wed Mar 9 15:05:26 2011 +1000
37354
37355 panoramiX: convert 1->panoramiXNumScreens loops to use macro (v2)
37356
37357 This converts all the remaining 1->num loops to the macro,
37358 this removes nearly all the panoramiXNumScreens usage in
37359 loops, and is a step to replacing it.
37360
37361 v2: move some from the other patch.
37362
37363 Signed-off-by: Dave Airlie <airlied@redhat.com>
37364 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37365 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
37366
37367commit 0bccfcc97b2300b83aa4693454c27ac87f23f221
37368Author: Dave Airlie <airlied@redhat.com>
37369Date: Wed Mar 9 15:01:20 2011 +1000
37370
37371 panoramiX: convert 0->panoramiXNumScreens loops to macro (v3)
37372
37373 This just uses the FOR_NSCREENS macro instead.
37374
37375 v2: remove some of the 1->x loops.
37376 v3: drop the 1->0 loop, will rework later.
37377
37378 Signed-off-by: Dave Airlie <airlied@redhat.com>
37379 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37380 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
37381
37382commit 5a94934487ea477947e24dcd4720b7cde77d3d2f
37383Author: Dave Airlie <airlied@redhat.com>
37384Date: Wed Mar 9 14:53:26 2011 +1000
37385
37386 panoramiX: consolidate common id assignment code.
37387
37388 This adds a new FOR_NSCREENS_FORWARD_SKIP, which skips the first
37389 element and is a common idiom throughout panoramiX code.
37390
37391 It then adds a new inline function to hide id assignment to a
37392 panoramiX resource and cleans up lots of common repeated code.
37393
37394 Signed-off-by: Dave Airlie <airlied@redhat.com>
37395 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37396 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
37397
37398commit c48c530e541e201ea1afe2512d85967659ceef84
37399Author: Dave Airlie <airlied@redhat.com>
37400Date: Wed Mar 9 15:11:42 2011 +1000
37401
37402 panoramiX: macro checking if drawable is root (v2)
37403
37404 this code appears in quite a few places, consolidate it into
37405 a macro in a header.
37406
37407 v2: align braces with macro just above it, and with
37408 lines removed
37409
37410 Signed-off-by: Dave Airlie <airlied@redhat.com>
37411 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37412
37413commit c3c0e2fdd326214e271ce144e8fc2725cbd738ef
37414Merge: a19771e 7355555
37415Author: Keith Packard <keithp@keithp.com>
37416Date: Wed Mar 9 14:25:54 2011 -0800
37417
37418 Merge remote branch 'whot/for-keith'
37419
37420commit 73555555a440855f9ae64c3367c5c7dca98c8741
37421Author: Adam Jackson <ajax@redhat.com>
37422Date: Tue Mar 8 15:33:13 2011 -0500
37423
37424 record: warning fix
37425
37426 record.c:810:9: warning: unused variable 'count'
37427
37428 Scope-shadowed by a later variable of the same name, safe to just
37429 delete.
37430
37431 Signed-off-by: Adam Jackson <ajax@redhat.com>
37432 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37433 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37434
37435commit 4d114cc5467a514faa437ce7f4c5e772e2f6a21d
37436Author: Adam Jackson <ajax@redhat.com>
37437Date: Tue Mar 8 15:33:12 2011 -0500
37438
37439 input: warning fixes
37440
37441 eventconvert.c:287:9: warning: enumeration value 'ET_Enter' not handled in switch
37442 eventconvert.c:287:9: warning: enumeration value 'ET_Leave' not handled in switch
37443 eventconvert.c:287:9: warning: enumeration value 'ET_FocusIn' not handled in switch
37444 eventconvert.c:287:9: warning: enumeration value 'ET_FocusOut' not handled in switch
37445 eventconvert.c:287:9: warning: enumeration value 'ET_DeviceChanged' not handled in switch
37446 eventconvert.c:287:9: warning: enumeration value 'ET_Hierarchy' not handled in switch
37447 eventconvert.c:287:9: warning: enumeration value 'ET_DGAEvent' not handled in switch
37448 eventconvert.c:287:9: warning: enumeration value 'ET_RawKeyPress' not handled in switch
37449 eventconvert.c:287:9: warning: enumeration value 'ET_RawKeyRelease' not handled in switch
37450 eventconvert.c:287:9: warning: enumeration value 'ET_RawButtonPress' not handled in switch
37451 eventconvert.c:287:9: warning: enumeration value 'ET_RawButtonRelease' not handled in switch
37452 eventconvert.c:287:9: warning: enumeration value 'ET_RawMotion' not handled in switch
37453 eventconvert.c:287:9: warning: enumeration value 'ET_XQuartz' not handled in switch
37454 eventconvert.c:287:9: warning: enumeration value 'ET_Internal' not handled in switch
37455
37456 From the code it appears these are can't happens, so if they ever do,
37457 BadImplementation seems entirely appropriate.
37458
37459 Signed-off-by: Adam Jackson <ajax@redhat.com>
37460 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37461 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37462
37463commit 7b5e562ea74039832116ee13db910f290f074782
37464Author: Adam Jackson <ajax@redhat.com>
37465Date: Tue Mar 8 15:33:11 2011 -0500
37466
37467 input: warning fix
37468
37469 getevents.c:770:5: warning: suggest parentheses around '&&' within '||'
37470
37471 Introduced with dc57f89959e549403f8488eb9f23425bd7118b22:
37472
37473 - if(dev->u.master && dev->valuator) {
37474 + if(dev->valuator && IsMaster(dev) || !IsFloating(dev)) {
37475
37476 So I'm assuming the two terms around the || are meant to be a unit.
37477
37478 Signed-off-by: Adam Jackson <ajax@redhat.com>
37479 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37480 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37481
37482commit 33fee13361e745e1db29e250b08622c83046d488
37483Author: Peter Hutterer <peter.hutterer@who-t.net>
37484Date: Tue Mar 8 14:41:21 2011 +1000
37485
37486 Xi: fix XI2 passive grab reply length calculation
37487
37488 If modifiers failed, the reply length was 4 bytes too short.
37489
37490 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37491 Reviewed-by: Julien Cristau <jcristau@debian.org>
37492 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37493
37494commit eb8141b6edd8b477c0ba796be71e985c35520a9b
37495Author: Julien Cristau <jcristau@debian.org>
37496Date: Mon Mar 7 18:55:19 2011 +0100
37497
37498 Xi: fix length checks for swapped clients
37499
37500 ChangeDeviceProperty and XIChangeProperty are followed by some data, so
37501 use REQUEST_AT_LEAST_SIZE instead of REQUEST_SIZE_MATCH.
37502
37503 X.Org bug#35082 <https://bugs.freedesktop.org/show_bug.cgi?id=35082>
37504
37505 Reported-by: Markus Fleschutz <markus.fleschutz@x-software.com>
37506 Signed-off-by: Julien Cristau <jcristau@debian.org>
37507 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37508 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37509
37510commit 4114533db6704324fc26f28a444415e325ace8e0
37511Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
37512Date: Fri Mar 4 17:55:33 2011 +0200
37513
37514 config: Ensure that stolen option list elements are released.
37515
37516 NewInputDeviceRequest steals the contents of option list elements but
37517 doesn't use the elements themselves for anything. Therefore the list
37518 elements need to be released always.
37519
37520 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37521 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37522 Reviewed-by: Adam Jackson <ajax@redhat.com>
37523 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37524 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37525
37526commit 8d30aff4aa708b9b885d492602ced7493a96a4df
37527Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
37528Date: Fri Mar 4 17:55:32 2011 +0200
37529
37530 dix: Release input device config info when the device disconnects.
37531
37532 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37533 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37534 Reviewed-by: Adam Jackson <ajax@redhat.com>
37535 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37536 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37537
37538commit 40e56d34538f4663426db50893c231a2b5d760dc
37539Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
37540Date: Fri Mar 4 17:55:31 2011 +0200
37541
37542 xkb: Ensure that XKB device private won't leak on device disconnect.
37543
37544 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37545 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37546 Reviewed-by: Adam Jackson <ajax@redhat.com>
37547 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37548 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37549
37550commit 18413f55089623123537c1499b02aa95ca2014d2
37551Author: Peter Hutterer <peter.hutterer@who-t.net>
37552Date: Thu Mar 3 14:15:55 2011 +1000
37553
37554 xfree86: block signals between EnableDevice and first CheckMotion()
37555
37556 Devices usually enable SIGIO processing in EnableDevice. CheckMotion
37557 initialises the pointer sprite, sends Enter/Leave events, etc. This leaves
37558 us with a small window where events may be processed without the sprite or
37559 pointer position (as seen from the protocol) is valid.
37560 Block signals during this window.
37561
37562 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37563 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37564
37565commit a19771e4337d1c4600550314bbc42a1495a023ff
37566Author: Erkki Seppälä <erkki.seppala@vincit.fi>
37567Date: Tue Mar 8 13:29:41 2011 -0500
37568
37569 xfree86/common: Remove a configScreen leak when conf_screen is NULL
37570
37571 configScreen used a dynamically allocated buffer for XF86ConfScreenRec
37572 when conf_screen argument was NULL. This pointer was never stored
37573 anywhere, nor was it released, so this patch makes the function use
37574 automatically allocated storage in that situation.
37575
37576 [ajax: minor grammar fix]
37577 Reviewed-by: Adam Jackson <ajax@redhat.com>
37578 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
37579 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
37580
37581commit b73ebe65cd071a063417ac9c38513a6742e32428
37582Author: Adam Jackson <ajax@redhat.com>
37583Date: Wed Mar 2 12:12:35 2011 -0500
37584
37585 glx: Remove unused _glapi_get_proc_name
37586
37587 Reviewed-by: Julien Cristau <jcristau@debian.org>
37588 Signed-off-by: Adam Jackson <ajax@redhat.com>
37589
37590commit 4377288abdbaac8506d8643b440c300ba03182fe
37591Author: Adam Jackson <ajax@redhat.com>
37592Date: Wed Mar 2 11:52:27 2011 -0500
37593
37594 glx: Simplify _glapi_check_multithread
37595
37596 We can guarantee that the X server does not call DRI driver services
37597 from multiple threads, so _glapi_check_multithread need never do
37598 anything special. As a result, ThreadSafe is always false, so remove it
37599 and simplify expressions where it appeared to match.
37600
37601 Reviewed-by: Julien Cristau <jcristau@debian.org>
37602 Signed-off-by: Adam Jackson <ajax@redhat.com>
37603
37604commit d5e8bb2150ce9d8a4132d61f76621a2c7a7f0acf
37605Author: Adam Jackson <ajax@redhat.com>
37606Date: Wed Mar 2 11:26:05 2011 -0500
37607
37608 glx: unifdef BEOS_THREADS
37609
37610 We don't support BeOS, so, no.
37611
37612 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
37613 Signed-off-by: Adam Jackson <ajax@redhat.com>
37614
37615commit 3692ff0b35e774a6f6cbedf634d53c4df2e53a24
37616Author: Adam Jackson <ajax@redhat.com>
37617Date: Wed Mar 2 11:22:45 2011 -0500
37618
37619 glx: unifdef USE_XTHREADS
37620
37621 xthreads is an ancient wrapper around a half dozen or so common thread
37622 APIs, including pthreads. Just use pthreads directly if you have it; if
37623 you don't, get with the times.
37624
37625 Reviewed-by: Julien Cristau <jcristau@debian.org>
37626 Signed-off-by: Adam Jackson <ajax@redhat.com>
37627
37628commit 2371b44f9eb9181ec2c081ce66befee51d18bf6e
37629Author: Adam Jackson <ajax@redhat.com>
37630Date: Wed Mar 2 11:17:07 2011 -0500
37631
37632 glx: unifdef SOLARIS_THREADS
37633
37634 Solaris 7 and older are not supported; Solaris 8 and later have (and
37635 use) pthreads.
37636
37637 Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
37638 Signed-off-by: Adam Jackson <ajax@redhat.com>
37639
37640commit a5fdd1aa06c7c8ecefeafafc4a778fc816ad41a4
37641Author: Adam Jackson <ajax@redhat.com>
37642Date: Wed Mar 2 10:20:02 2011 -0500
37643
37644 glx: Remove unused _glapi_get_proc_offset and friends
37645
37646 Reviewed-by: Julien Cristau <jcristau@debian.org>
37647 Signed-off-by: Adam Jackson <ajax@redhat.com>
37648
37649commit 2cb3dc020cbec1ba64848dac532b3bb2d84d0635
37650Author: Adam Jackson <ajax@redhat.com>
37651Date: Mon Feb 28 17:11:35 2011 -0500
37652
37653 xi: Use RESTYPE consistently
37654
37655 No functional change
37656
37657 Reviewed-by: Soren Sandmann <ssp@redhat.com>
37658 Signed-off-by: Adam Jackson <ajax@redhat.com>
37659
37660commit 2413702aa0b488f1a4f828a017d5b508c4f219f1
37661Author: Adam Jackson <ajax@redhat.com>
37662Date: Mon Feb 28 16:54:17 2011 -0500
37663
37664 xvmc: Use RESTYPE consistently
37665
37666 Also mark these static since they're not used externally.
37667
37668 Reviewed-by: Soren Sandmann <ssp@redhat.com>
37669 Signed-off-by: Adam Jackson <ajax@redhat.com>
37670
37671commit cd58924d294de2abddbd2f03c2db021894fe9759
37672Author: Adam Jackson <ajax@redhat.com>
37673Date: Mon Feb 28 16:28:26 2011 -0500
37674
37675 xinerama: Use RESTYPE consistently
37676
37677 No functional change
37678
37679 Reviewed-by: Soren Sandmann <ssp@redhat.com>
37680 Signed-off-by: Adam Jackson <ajax@redhat.com>
37681
37682commit 9a6a9971b03381b7c44257284edecdfa0135957f
37683Author: Adam Jackson <ajax@redhat.com>
37684Date: Mon Feb 28 16:25:23 2011 -0500
37685
37686 xv: Use RESTYPE where appropriate
37687
37688 No functional change.
37689
37690 Reviewed-by: Soren Sandmann <ssp@redhat.com>
37691 Signed-off-by: Adam Jackson <ajax@redhat.com>
37692
37693commit fddfd026a0e4a8190dc01fb0f38af4b2c29b63a3
37694Author: Adam Jackson <ajax@redhat.com>
37695Date: Mon Feb 28 13:01:15 2011 -0500
37696
37697 dix: Remove PIXEL typedef
37698
37699 Doesn't appear to be used anywhere.
37700
37701 Reviewed-by: Soren Sandmann <ssp@redhat.com>
37702 Signed-off-by: Adam Jackson <ajax@redhat.com>
37703
37704commit 82a8677d9175732a61df4116a396b76a7704efb4
37705Author: Adam Jackson <ajax@redhat.com>
37706Date: Thu Jan 20 22:49:32 2011 -0500
37707
37708 dix: Remove deprecated.c
37709
37710 No more internal users, this can be dropped now.
37711
37712 Reviewed-by: Soren Sandmann <ssp@redhat.com>
37713 Signed-off-by: Adam Jackson <ajax@redhat.com>
37714
37715commit 1564c82417d201de5b9a5ec5e7aa4ef14c45fbad
37716Author: Adam Jackson <ajax@redhat.com>
37717Date: Thu Jan 20 18:37:45 2011 -0500
37718
37719 dix: Remove usage_hint from pixmaps, store it in ->drawable.class
37720
37721 The class field was unused for pixmaps, and we don't have enough classes
37722 to justify a whole uint32 anyway.
37723
37724 Reviewed-by: Soren Sandmann <ssp@redhat.com>
37725 Signed-off-by: Adam Jackson <ajax@redhat.com>
37726
37727commit d8caa782009abf4dc17b945e325e83fda299a534
37728Author: Adam Jackson <ajax@redhat.com>
37729Date: Thu Feb 24 16:06:34 2011 -0500
37730
37731 vbe: Fix malloc size bug
37732
37733 v2: Slightly more obvious sizing math.
37734
37735 ==14882== Invalid write of size 2
37736 ==14882== at 0x6750267: VBEGetVBEInfo (vbe.c:400)
37737 ==14882== by 0x6142064: ??? (in /usr/lib64/xorg/modules/drivers/vesa_drv.so)
37738 ==14882== by 0x471895: InitOutput (xf86Init.c:519)
37739 ==14882== by 0x422778: main (main.c:205)
37740 ==14882== Address 0x4f32fa8 is 72 bytes inside a block of size 73 alloc'd
37741 ==14882== at 0x4A0640D: malloc (vg_replace_malloc.c:236)
37742 ==14882== by 0x675024B: VBEGetVBEInfo (vbe.c:398)
37743 ==14882== by 0x6142064: ??? (in /usr/lib64/xorg/modules/drivers/vesa_drv.so)
37744 ==14882== by 0x471895: InitOutput (xf86Init.c:519)
37745 ==14882== by 0x422778: main (main.c:205)
37746
37747 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
37748 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
37749 Signed-off-by: Adam Jackson <ajax@redhat.com>
37750
37751commit 628d16a92a7fa556fbb70bf4a4adf57ec05c190b
37752Author: Keith Packard <keithp@keithp.com>
37753Date: Sat Feb 26 10:31:50 2011 -0800
37754
37755 loader: Don't distribute sdksyms.c and make it depend on the config
37756
37757 sdksyms.c is constructed by processing header files with the C
37758 preprocessor. Its contents will vary depending on the precise
37759 configuration options, and so must depend on the config header
37760 files.
37761
37762 We have one header file which is always changed when any config option
37763 is modified called do-not-use-config.h (which may want a different
37764 name at some point), so make sdksyms.c depend on that file.
37765
37766 Also, we don't want to ship this file; it always needs to be
37767 built. So, include it in the nodist_libloader_la_SOURCES list to
37768 prevent it from being added to the tarball.
37769
37770 Signed-off-by: Keith Packard <keithp@keithp.com>
37771 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
37772
37773commit 705f047876f78bbdfb368c3643925af1d1d17f71
37774Author: Julien Cristau <jcristau@debian.org>
37775Date: Fri Mar 4 01:12:01 2011 +0100
37776
37777 configure.ac: bump version for 1.11 development
37778
37779 Signed-off-by: Julien Cristau <jcristau@debian.org>
37780 Signed-off-by: Keith Packard <keithp@keithp.com>
37781
37782commit 6c90e839d9872a37d371578c9c423e8213922044
37783Merge: 0bc95d5 edcceed
37784Author: Keith Packard <keithp@keithp.com>
37785Date: Thu Mar 3 14:41:44 2011 -0800
37786
37787 Merge remote branch 'whot/for-keith'
37788
37789commit 0bc95d5b06dcea65a1aa193ea907b50f7dd168b5
37790Merge: 3f41f4a 69a9171
37791Author: Keith Packard <keithp@keithp.com>
37792Date: Thu Mar 3 14:33:08 2011 -0800
37793
37794 Merge remote branch 'jeremyhu/master'
37795
37796commit 3f41f4adea4bbb90d4bda4dab600595b655e3ed8
37797Merge: 8e4c3ce dae24ab
37798Author: Keith Packard <keithp@keithp.com>
37799Date: Thu Mar 3 14:12:36 2011 -0800
37800
37801 Merge remote branch 'sandmann/for-keithp'
37802
37803commit 8e4c3ce55b0f186bc6ba4039e30629669b6087b7
37804Merge: f3d19c0 2965615
37805Author: Keith Packard <keithp@keithp.com>
37806Date: Thu Mar 3 13:42:07 2011 -0800
37807
37808 Merge remote branch 'rjy/clientids'
37809
37810commit 1c008e7e7865b405b8033f625333cd64ece4499e
37811Author: Simon Thum <simon.thum@gmx.de>
37812Date: Wed Feb 23 02:29:35 2011 +0100
37813
37814 dix: change all timestamps in pointer acceleration to CARD32
37815
37816 CARD32 is being returned by GetTimeInMilis(), so use it consistently.
37817
37818 Signed-off-by: Simon Thum <simon.thum@gmx.de>
37819 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37820
37821commit a4b85261859b17dba9ad8f7f1ce650133f0235d4
37822Author: Simon Thum <simon.thum@gmx.de>
37823Date: Wed Feb 23 02:29:34 2011 +0100
37824
37825 dix: update pointer acceleration code to use ValuatorMask
37826
37827 Signed-off-by: Simon Thum <simon.thum@gmx.de>
37828 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37829
37830commit 8128846e16b406c7d459ae7556f7cd09bdc24c91
37831Author: Simon Thum <simon.thum@gmx.de>
37832Date: Wed Feb 23 02:29:33 2011 +0100
37833
37834 dix: refactor predictable scheme initialization
37835
37836 This intends to clean up the predictable accel struct
37837 from purely scheme-related things like input properties,
37838 as they would be useless in other use cases such
37839 as wheel acceleration.
37840
37841 Signed-off-by: Simon Thum <simon.thum@gmx.de>
37842 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37843 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37844
37845commit 69a9171dbbafd6a7db702d48770d28fb54717545
37846Author: Jeremy Huddleston <jeremyhu@apple.com>
37847Date: Tue Mar 1 20:02:46 2011 -0800
37848
37849 XQuartz: pbproxy: Protect against possible collision between Cocoa and X11 Cursor types
37850
37851 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
37852
37853commit dae24abcd44f3bb1966faa88222f851c784b37dc
37854Author: Søren Sandmann Pedersen <ssp@redhat.com>
37855Date: Mon Feb 28 11:10:38 2011 -0500
37856
37857 Delete RegionClipSpans()
37858
37859 Nothing uses it.
37860
37861 Reviewed-by: Adam Jackson <ajax@redhat.com>
37862 Signed-off-by: Søren Sandmann Pedersen <ssp@redhat.com>
37863
37864commit edcceedbd35df576929685767d0b619659e5b020
37865Author: Adam Jackson <ajax@redhat.com>
37866Date: Mon Feb 28 12:32:53 2011 -0500
37867
37868 xext: Remove XCALIBRATE extension
37869
37870 Nobody can have been using this, it's never called from extension init.
37871 XI2 device properties should now be a functional replacement.
37872
37873 Signed-off-by: Adam Jackson <ajax@redhat.com>
37874 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37875 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37876
37877commit c2207d11f243900bc2f641e08d80da63d84e97a8
37878Merge: 0077993 579ee8f
37879Author: Peter Hutterer <peter.hutterer@who-t.net>
37880Date: Mon Feb 28 11:26:09 2011 +1000
37881
37882 Merge branch 'next' into for-keith
37883
37884 Conflicts:
37885 dix/devices.c
37886
37887 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37888
37889commit 0343aed1f082baf4bfbd730c1c3582f1410b8741
37890Author: Jeremy Huddleston <jeremyhu@apple.com>
37891Date: Sun Feb 27 16:18:27 2011 -0800
37892
37893 XQuartz: Don't use deprecated CoreGraphics API on SL and Lion
37894
37895 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
37896
37897commit de32d4dcf57c10fc100c1b33630127947a8c8460
37898Author: Jeremy Huddleston <jeremyhu@apple.com>
37899Date: Sun Feb 27 15:49:10 2011 -0800
37900
37901 Revert "rootless: Remove ROOTLESS_WORKAROUND"
37902
37903 Christof Wolf has reported a regression that seems to be caused by
37904 this change, so reverting the change in the 1.9 branch. We'll
37905 investigate a proper fix in master for 1.10.
37906
37907 This reverts commit c89f0521044083a11d538ebfeaabee6fc7fb9a03.
37908
37909 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
37910
37911commit 00779932de861bf7ac8c625802f1afed75d01b6f
37912Author: Matthieu Herrb <matthieu.herrb@laas.fr>
37913Date: Sun Feb 27 20:16:03 2011 +0100
37914
37915 Don't clobber input device options from xorg.conf
37916
37917 Since commit b8d9c5ff removed commonOptions, we now
37918 need to append the "Core{Keyboard,Pointer}" options to
37919 the existing list.
37920
37921 Fixes passing options to devices confirured in xorg.conf
37922 on systems where autoaddevices is false.
37923
37924 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
37925 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
37926 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
37927
37928commit 8cf3348e90846f5b04236479042228fb98ac8f70
37929Author: Jeremy Huddleston <jeremyhu@apple.com>
37930Date: Tue Feb 8 11:38:35 2011 -0800
37931
37932 XQuartz: RandR: Add RandR modes for the primary display in multi-monitor configs
37933
37934 We now support using RandR to set the resolution of the primary display (and
37935 place a shielding window on other displays) in multi-monitor configurations.
37936
37937 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
37938
37939commit 968652983f8e6ae6889b48e15f4098ff6ad4a15a
37940Author: Jeremy Huddleston <jeremyhu@apple.com>
37941Date: Tue Feb 8 12:26:35 2011 -0800
37942
37943 XQuartz: RandR: Provide an alert box when entering a RandR mode for the first time.
37944
37945 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
37946
37947commit 13578b852b7631f99cf1fd5e2e5469edc5aae369
37948Author: Jeremy Huddleston <jeremyhu@apple.com>
37949Date: Tue Feb 8 11:37:52 2011 -0800
37950
37951 XQuartz: RandR: Capture the display when switching modes with RandR
37952
37953 This will prevent native windows from resizing as we change resolutions.
37954
37955 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
37956
37957commit 7dff79e39564b403c3afbc5e7bacffa0df190a23
37958Author: Søren Sandmann Pedersen <ssp@redhat.com>
37959Date: Fri Feb 11 09:00:10 2011 -0500
37960
37961 Implement fbAddTriangles() in terms of pixman_add_triangles()
37962
37963 This allows the remaining triangle-to-trap conversion code to be
37964 deleted.
37965
37966 Reviewed-by: Adam Jackson <ajax@redhat.com>
37967 Signed-off-by: Søren Sandmann <ssp@redhat.com>
37968
37969commit 566f1931ee2916269e164e114bffaf2da1d039d1
37970Author: Søren Sandmann Pedersen <ssp@redhat.com>
37971Date: Tue Jan 11 10:33:57 2011 -0500
37972
37973 Move miTriangles to fb as fbTriangles()
37974
37975 The fb version simply calls the new pixman_composite_triangles(). This
37976 allows us to get rid of miCreateAlphaPicture().
37977
37978 Reviewed-by: Adam Jackson <ajax@redhat.com>
37979 Signed-off-by: Søren Sandmann <ssp@redhat.com>
37980
37981commit 788ccb9a8bcf6a4fb4054c507111eec3338fb969
37982Author: Søren Sandmann Pedersen <ssp@redhat.com>
37983Date: Tue Jan 11 09:46:46 2011 -0500
37984
37985 Move miTrapezoids() into fb as fbTrapezoids()
37986
37987 The main consumer of trapezoids, cairo, is using the Trapezoids
37988 request, which is currently implemented in the miTrapezoids()
37989 function. That function splits the request into smaller bits and calls
37990 lower level functions such as AddTrap.
37991
37992 By moving the implementation of the whole request into fb, we can
37993 instead call pixman_composite_trapezoids() to do the whole request in
37994 one step.
37995
37996 There are no callers of miTrapezoids in any of the open source
37997 drivers, although exa and uxa have their own copies of the function.
37998
37999 Reviewed-by: Adam Jackson <ajax@redhat.com>
38000 Signed-off-by: Søren Sandmann <ssp@redhat.com>
38001
38002commit 197df069a4037d6faa2723c31ffba09c95d71166
38003Author: Søren Sandmann Pedersen <ssp@redhat.com>
38004Date: Wed Feb 23 10:36:57 2011 -0500
38005
38006 Require pixman 0.21.6
38007
38008 The following patches need pixman_composite_trapezoids() and
38009 pixman_add_triangles().
38010
38011 Signed-off-by: Soren Sandmann <ssp@redhat.com>
38012
38013commit f3d19c0cf3327cceb90f7378f5d8fc0c3e327400
38014Author: Keith Packard <keithp@keithp.com>
38015Date: Fri Feb 25 21:22:23 2011 -0800
38016
38017 Version bumped to 1.10
38018
38019 Signed-off-by: Keith Packard <keithp@keithp.com>
38020
38021commit 678f5396c91b3d0c7572ed579b0a4fb62b2b4655
38022Author: Keith Packard <keithp@keithp.com>
38023Date: Fri Feb 25 21:10:21 2011 -0800
38024
38025 input: Ensure Valuator axes are aligned as needed
38026
38027 Let the compiler figure out the correct alignment for the axes data
38028 for a valuator by using a union to force double alignment of the
38029 initial ValuatorClassRec structure in the allocation.
38030
38031 Signed-off-by: Keith Packard <keithp@keithp.com>
38032 Tested-by: Julien Cristau <jcristau@debian.org>
38033 Reviewed-by: Julien Cristau <jcristau@debian.org>
38034
38035commit dc8f52e77f51b6fa8908d9611c3f7e3cfbbaf2d1
38036Author: Keith Packard <keithp@keithp.com>
38037Date: Fri Feb 25 16:38:27 2011 -0800
38038
38039 hw/dmx/doc: Add explicit dependency for all doxygen output files
38040
38041 Instead of listing one of the doxygen output files and depending on
38042 sequential execution to ensure that the other files were present
38043 before make checked for them, create explicit dependencies so that
38044 make will not check for the additional files until after doxygen has
38045 been run.
38046
38047 This allows parallel make to work correctly in this directory.
38048
38049 Signed-off-by: Keith Packard <keithp@keithp.com>
38050 Tested-by: Kristian Høgsberg <krh@bitplanet.net>
38051
38052commit fd4d9c75c265add8d6cc74afb341c9c4f2ec1073
38053Author: Keith Packard <keithp@keithp.com>
38054Date: Fri Feb 25 10:54:50 2011 -0800
38055
38056 hw/xwin: Look for gl spec files in $(srcdir) or .
38057
38058 Tarballs include the downloaded gl spec files, which will end up in
38059 $(srcdir). But, git-based builds will not have them at all and will
38060 need to download them from opengl.org. They'll land in in the build
38061 directory instead of $(srcdir), and so we need to allow them to be in
38062 either place.
38063
38064 This change checks for the files in $(srcdir), linking them to . if
38065 present. Otherwise, it downloads them from opengl.org.
38066
38067 A suggested better solution is to have Mesa install these files somewhere.
38068
38069 Signed-off-by: Keith Packard <keithp@keithp.com>
38070 Tested-by: Kristian Høgsberg <krh@bitplanet.net>
38071
38072commit 6b951dec69103ca76d6d4a61b1a2f67a52d013b9
38073Author: Dan Nicholson <dbn.lists@gmail.com>
38074Date: Sun Feb 20 09:51:15 2011 -0800
38075
38076 dmx: Construct paths in doxygen.conf to fix VPATH builds
38077
38078 The paths in doxygen.conf assumed that srcdir=builddir and broke
38079 otherwise. Use autoconf to fill in the paths to the srcdir so that the
38080 files can be found when users have a separate build directory (as with
38081 distcheck).
38082
38083 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
38084 Tested-by: Cyril Brulebois <kibi@debian.org>
38085 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
38086 Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
38087
38088commit 61ce915bf78570b7c8d53c118700ef7274bbdfb7
38089Author: Dan Nicholson <dbn.lists@gmail.com>
38090Date: Tue Feb 15 06:03:28 2011 -0800
38091
38092 xfree86: Allow sdksyms.dep to be included portably
38093
38094 Non-GNU makes don't deal with the sinclude or -include variants that
38095 allow Makefile stubs to be created and then included during the build.
38096 Instead, create an empty file at the end of configure so that the
38097 regular include statement can be included. This is how automake handles
38098 automatic source dependencies.
38099
38100 In order to trick automake into not processing the include statement, a
38101 variable is used.
38102
38103 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
38104 Tested-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38105
38106commit 780a77acce1dd369549ece802b3e2c4006058dfe
38107Author: Keith Packard <keithp@keithp.com>
38108Date: Thu Feb 24 20:45:46 2011 -0800
38109
38110 Version bumped to 1.9.99.903 (1.10 RC3)
38111
38112 Signed-off-by: Keith Packard <keithp@keithp.com>
38113
38114commit 9d23459415b84606ee4f38bb2d19054c432c8552
38115Author: Peter Hutterer <peter.hutterer@who-t.net>
38116Date: Fri Feb 25 11:08:19 2011 +1000
38117
38118 dix: release all buttons and keys before reattaching a device (#34182)
38119
38120 Testcase:
38121 xinput float <keyboard name>
38122
38123 results in the keyboard's enter key being repeated as the device is detached
38124 while the key is still physically down. To avoid this, release all keys and
38125 buttons before reattaching the device.
38126
38127 X.Org Bug 34182 <http://bugs.freedesktop.org/show_bug.cgi?id=34182>
38128
38129 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38130 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
38131
38132commit 229b055bdbc53370944401649234e8a38183f4fb
38133Merge: 6178959 b17fc99
38134Author: Keith Packard <keithp@keithp.com>
38135Date: Thu Feb 24 19:44:35 2011 -0800
38136
38137 Merge remote branch 'jeremyhu/master'
38138
38139commit 6178959e3dd6482a4317de6eb14eb19ca7329b9c
38140Author: Keith Packard <keithp@keithp.com>
38141Date: Thu Feb 24 19:42:02 2011 -0800
38142
38143 xfree86: Bump video ABI to 10.0
38144
38145 RandR 1.4 revert changed things
38146
38147 Signed-off-by: Keith Packard <keithp@keithp.com>
38148
38149commit 31704510f448706524b7b7085cc4ff0ada7bfe7e
38150Author: Michel Dänzer <daenzer@vmware.com>
38151Date: Thu Feb 24 12:17:57 2011 +0100
38152
38153 EXA/mixed: ModifyPixmapHeader pitch fixes. (bug #33929)
38154
38155 If there's a GPU copy and a non-zero devKind was passed in, set the GPU copy
38156 pitch to that instead of to a possibly bogus value derived from the new width.
38157 This is e.g. used by the radeon driver's drmmode_xf86crtc_resize hook, fixes
38158 https://bugs.freedesktop.org/show_bug.cgi?id=33929 .
38159
38160 On the other hand, the system memory copy doesn't need the pitch to be aligned
38161 beyond the PixmapBytePad of the width.
38162
38163 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
38164 Acked-by: Cyril Brulebois <kibi@debian.org>
38165 Tested-by: Cyril Brulebois <kibi@debian.org>
38166 Reported-by: Thierry Vignaud <thierry.vignaud@gmail.com>
38167 Tested-by: Thierry Vignaud <thierry.vignaud@gmail.com>
38168 Signed-off-by: Keith Packard <keithp@keithp.com>
38169
38170commit b17fc99cb9ca9ff0a3592f783a906cdcc35da748
38171Author: Jeremy Huddleston <jeremyhu@apple.com>
38172Date: Thu Feb 24 19:24:35 2011 -0800
38173
38174 XQuartz: Localization Updates
38175
38176 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
38177
38178commit 59850630fefe5e4f0430d7e2e106937fdf1e7ffb
38179Author: Jeremy Huddleston <jeremyhu@apple.com>
38180Date: Thu Feb 24 19:18:35 2011 -0800
38181
38182 XQuartz: Add LSApplicationCategoryType key to Info.plist
38183
38184 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
38185
38186commit 83861595782aaa05907f9cf7b236d50261d404d9
38187Merge: a1cc0e5 4102a00
38188Author: Keith Packard <keithp@keithp.com>
38189Date: Thu Feb 24 18:59:07 2011 -0800
38190
38191 Merge remote branch 'kibi/master'
38192
38193commit a1cc0e52b0e8ca40bc1218de553424b2bcb760ef
38194Author: Alexandr Shadchin <alexandr.shadchin@gmail.com>
38195Date: Tue Nov 30 21:17:19 2010 +0500
38196
38197 Removing unused code
38198
38199 In OpenBSD removed support PCCONS in 2002 year
38200 http://marc.info/?l=openbsd-cvs&m=102435816424294&w=2
38201
38202 Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
38203 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
38204 Signed-off-by: Keith Packard <keithp@keithp.com>
38205
38206commit 365ad68fb9f7029550505b7c276a808050cada9c
38207Merge: 0801afb 720c895
38208Author: Keith Packard <keithp@keithp.com>
38209Date: Thu Feb 24 18:49:40 2011 -0800
38210
38211 Merge remote branch 'ajax/for-keithp'
38212
38213commit 0801afbd7c2c644c672b37f8463f1a0cbadebd2e
38214Author: Erkki Seppälä <erkki.seppala@vincit.fi>
38215Date: Thu Feb 10 15:35:14 2011 +0200
38216
38217 record: avoid crash when calling RecordFlushReplyBuffer recursively
38218
38219 RecordFlushReplyBuffer can call itself recursively through
38220 WriteClient->CallCallbacks->_CallCallbacks->RecordFlushAllContexts
38221 when the recording client's buffer cannot be completely emptied in one
38222 WriteClient. When a such a recursion occurs, it will not be broken out
38223 of which results in segmentation fault when the stack is exhausted.
38224
38225 This patch adds a counter (a flag, really) that guards against this
38226 situation, to break out of the recursion.
38227
38228 One alternative to this change would be to change _CallCallbacks to
38229 check the corresponding counter before the callback loop, but that
38230 might affect existing behavior, which may be relied upon.
38231
38232 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
38233 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
38234 Signed-off-by: Keith Packard <keithp@keithp.com>
38235
38236commit 816f1f8ffafbfbf0dd31ea86f295987f84151feb
38237Author: Peter Hutterer <peter.hutterer@who-t.net>
38238Date: Tue Feb 22 12:48:04 2011 +1000
38239
38240 include: document list interface.
38241
38242 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38243 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38244
38245commit b833f9b924ddc043244c0a026547c438f1c4d4df
38246Author: Keith Packard <keithp@keithp.com>
38247Date: Wed Feb 23 11:18:35 2011 -0800
38248
38249 Revert "Replace huge argument list in xf86CrtcSetModeTransform with struct"
38250
38251 This reverts commit 8b35118c03590a7ad3786d3284bafb3f40fcb8cc.
38252
38253commit 7030c82648930cea7ef8a2ea49d8138078920d4d
38254Author: Keith Packard <keithp@keithp.com>
38255Date: Wed Feb 23 11:18:16 2011 -0800
38256
38257 Revert "Require RandR protocol version 1.4 or newer"
38258
38259 This reverts commit c8bc25fd7629df10f2825b7cc713b031ae78f223.
38260
38261commit a91d33917befd6719bb7f01007bc253a2aefb0bb
38262Author: Keith Packard <keithp@keithp.com>
38263Date: Wed Feb 23 11:18:12 2011 -0800
38264
38265 Revert "randr: Add sprite position transforms"
38266
38267 This reverts commit 66294afcab7b7a82f7dd897767e46c48a94b8ee8.
38268
38269commit 3d4ee25a1e2e0ed2463ceaabea8f5a284a6ddd60
38270Author: Keith Packard <keithp@keithp.com>
38271Date: Wed Feb 23 11:18:08 2011 -0800
38272
38273 Revert "randr: Implement RRSetCrtcConfigs"
38274
38275 This reverts commit d94a035ea9eb3167fc4f35b2d9f0d53f8807014c.
38276
38277commit 8b0e651f7d09a7f35c87a4d0a6e2053cae93b516
38278Author: Keith Packard <keithp@keithp.com>
38279Date: Wed Feb 23 11:18:05 2011 -0800
38280
38281 Revert "hw/xfree86/modes: Add optional driver API for RRSetCrtcConfigs"
38282
38283 This reverts commit 86c489c319c705f710bee3897fe27600ce15008e.
38284
38285commit 9e8c20b0d49783d6b9334d4c11c4e6e3ba273524
38286Author: Keith Packard <keithp@keithp.com>
38287Date: Wed Feb 23 11:17:42 2011 -0800
38288
38289 Revert "randr: Add per-crtc pixmaps"
38290
38291 This reverts commit 82612045e11f2b882ae132e184a9629f43f1c424.
38292
38293commit 9f47780ecdc7693f756587a758ec0141e75cb1eb
38294Author: Keith Packard <keithp@keithp.com>
38295Date: Wed Feb 23 11:17:36 2011 -0800
38296
38297 Revert "randr: Hook up the new RandR 1.4 functionality"
38298
38299 This reverts commit afb6ebf1d5829346c40fe1053c9f50afe926e6c6.
38300
38301commit 793a242c89fccdbf1a2557ba33da1da2cb4ff8c4
38302Author: Keith Packard <keithp@keithp.com>
38303Date: Wed Feb 23 11:17:33 2011 -0800
38304
38305 Revert "DIX is responsible for ref counting scanout pixmaps."
38306
38307 This reverts commit 96b4d4787bf82edd9d06eb9a6e94bc45412c7df2.
38308
38309commit 7b7cd3f121b3904ad851155f853ce87a0e00d002
38310Author: Keith Packard <keithp@keithp.com>
38311Date: Wed Feb 23 11:17:29 2011 -0800
38312
38313 Revert "Set sprite transforms from RRSetCrtcConfigs"
38314
38315 This reverts commit a88d70fb20a2bc3152b84adff4380857e6cfadf5.
38316
38317commit fd9331f6eb39a28da7de4867b2e3a460f667d514
38318Author: Keith Packard <keithp@keithp.com>
38319Date: Wed Feb 23 11:17:24 2011 -0800
38320
38321 Revert "Separate out screen size and screen pixmap sizes in RRScreenSizeSet"
38322
38323 This reverts commit 752c368421c1c824752cf467fba9318d75d2ca2c.
38324
38325commit 4d91e7a63161cdde9af53aedb3fe3e53dbe3049e
38326Author: Keith Packard <keithp@keithp.com>
38327Date: Wed Feb 23 11:17:20 2011 -0800
38328
38329 Revert "ProcRRSetCrtcConfigs uses 'configs' without being initialized"
38330
38331 This reverts commit b0f4bd61f0caf80f3be9a176f1f7a707bc6628d8.
38332
38333commit 00d0b235cfbfb162ddd58c088d2ed03e55a9c5cb
38334Author: Keith Packard <keithp@keithp.com>
38335Date: Wed Feb 23 11:16:48 2011 -0800
38336
38337 Revert "randr: handle RRSetCrtcConfigs request with zero configs"
38338
38339 This reverts commit 0d01b66df9081ef48843b3bad81c56bb2cd1ae69.
38340
38341commit 720c895db9327bbdc32215a67236303d2ceafad4
38342Author: Peter Hutterer <peter.hutterer@who-t.net>
38343Date: Wed Feb 23 13:36:20 2011 -0500
38344
38345 Add mode field to ConstrainCursorHarder
38346
38347 For Pointer Barriers, the movement mode is important and must be passed
38348 through.
38349
38350 Reviewed-by: Adam Jackson <ajax@redhat.com>
38351 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38352
38353commit a16e282200f84233041e5a565c6363a5a78be525
38354Author: Adam Jackson <ajax@redhat.com>
38355Date: Mon Feb 8 15:40:22 2010 -0500
38356
38357 xfree86: Fix the sdk headers to be multilib-safe
38358
38359 Use _LP64 (verified on gcc and sun compilers) instead of _XSERVER64 in
38360 internal header usage, and always define _XSERVER64 for modules if _LP64
38361 is defined. Prevents differing xorg-server.h between 32 and 64 bit
38362 packages.
38363
38364 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38365 Signed-off-by: Adam Jackson <ajax@redhat.com>
38366
38367commit dc498b433f36af5d2de3065e7c64cdb575385d81
38368Author: Adam Jackson <ajax@redhat.com>
38369Date: Mon Feb 8 15:23:04 2010 -0500
38370
38371 xfree86: If the driver found modes on an output, don't add more
38372
38373 Inferring modes from sync ranges is only valid if the monitor says it's
38374 valid. If the monitor says it's valid, then we'll have already added
38375 those modes during EDID block parse. If it doesn't, then we should
38376 believe it.
38377
38378 If there's no EDID for an output, but sync ranges from the config, we'll
38379 still add default modes as normal.
38380
38381 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
38382 Signed-off-by: Adam Jackson <ajax@redhat.com>
38383
38384commit 303977fbcf9c641b7e19dfde192cef585f5b455c
38385Author: Adam Jackson <ajax@redhat.com>
38386Date: Tue Feb 8 14:52:49 2011 -0500
38387
38388 glxproxy: warning fix
38389
38390 glxvendor.c: In function ‘__glXVForwardPipe0WithReply’:
38391 glxvendor.c:205:10: warning: ‘be_buf’ may be used uninitialized in this function
38392
38393 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38394 Signed-off-by: Adam Jackson <ajax@redhat.com>
38395
38396commit 405ad0610d6e2f4606768294b655ac52aee9ae58
38397Author: Adam Jackson <ajax@redhat.com>
38398Date: Tue Feb 8 14:51:47 2011 -0500
38399
38400 glxproxy: warning fix
38401
38402 glxvendor.c: In function ‘__glXVForwardAllWithReply’:
38403 glxvendor.c:284:10: warning: ‘be_buf’ may be used uninitialized in this function
38404 glxvendor.c:285:10: warning: ‘be_buf_size’ may be used uninitialized in this function
38405
38406 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38407 Signed-off-by: Adam Jackson <ajax@redhat.com>
38408
38409commit a4cd2e0da6415ec1b3b182579bebbe2a41f29d30
38410Author: Adam Jackson <ajax@redhat.com>
38411Date: Tue Feb 8 14:47:05 2011 -0500
38412
38413 glxproxy: warning fix
38414
38415 glxsingle.c: In function ‘__glXForwardPipe0WithReply’:
38416 glxsingle.c:218:10: warning: ‘be_buf’ may be used uninitialized in this function
38417
38418 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38419 Signed-off-by: Adam Jackson <ajax@redhat.com>
38420
38421commit f928caca6e60215bc9ab423acae5542dfabc9bec
38422Author: Adam Jackson <ajax@redhat.com>
38423Date: Tue Feb 8 14:45:31 2011 -0500
38424
38425 glxproxy: warning fix
38426
38427 glxsingle.c: In function ‘__glXForwardAllWithReply’:
38428 glxsingle.c:300:10: warning: ‘be_buf’ may be used uninitialized in this function
38429 glxsingle.c:301:10: warning: ‘be_buf_size’ may be used uninitialized in this function
38430
38431 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38432 Signed-off-by: Adam Jackson <ajax@redhat.com>
38433
38434commit dad2712c9328e113db4de768a12a8dafa6c177e9
38435Author: Adam Jackson <ajax@redhat.com>
38436Date: Tue Feb 8 14:44:03 2011 -0500
38437
38438 glxproxy: warning fix
38439
38440 glxsingle.c: In function ‘__glXDisp_ReadPixels’:
38441 glxsingle.c:760:11: warning: ‘buf’ may be used uninitialized in this function
38442
38443 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38444 Signed-off-by: Adam Jackson <ajax@redhat.com>
38445
38446commit 4270157bac645550e2c0afe89479c0bfe9d53447
38447Author: Adam Jackson <ajax@redhat.com>
38448Date: Tue Feb 8 14:38:23 2011 -0500
38449
38450 glxproxy: warning fix
38451
38452 render2swap.c:264:13: warning: ‘swapArray’ defined but not used
38453
38454 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38455 Signed-off-by: Adam Jackson <ajax@redhat.com>
38456
38457commit 5d0910b4fdf99e3239635a01eb2709c32d0e5bb9
38458Author: Adam Jackson <ajax@redhat.com>
38459Date: Tue Feb 8 14:32:54 2011 -0500
38460
38461 glxproxy: warning fix
38462
38463 glxcmds.c: In function ‘CreateGLXPixmap’:
38464 glxcmds.c:1663:20: warning: comparison between pointer and integer
38465 glxcmds.c:1663:38: warning: comparison between pointer and integer
38466
38467 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38468 Signed-off-by: Adam Jackson <ajax@redhat.com>
38469
38470commit e1b5d3e5e7f157ab769c40ad2efdc1281c25a03a
38471Author: Adam Jackson <ajax@redhat.com>
38472Date: Tue Feb 8 14:18:53 2011 -0500
38473
38474 glxproxy: warning fix
38475
38476 glxcmds.c: In function ‘CreateGLXPixmap’:
38477 glxcmds.c:1641:22: warning: ‘pGlxScreen’ may be used uninitialized in this function
38478
38479 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38480 Signed-off-by: Adam Jackson <ajax@redhat.com>
38481
38482commit 3b9de273a94fb68de51238c20c3182396aa41b84
38483Author: Adam Jackson <ajax@redhat.com>
38484Date: Tue Feb 8 14:07:20 2011 -0500
38485
38486 glxproxy: warning fix
38487
38488 glxcmds.c: In function ‘CreateContext.clone.6’:
38489 glxcmds.c:105:19: warning: ‘be_fbconfigId’ may be used uninitialized in this function
38490 glxcmds.c:104:14: warning: ‘be_vid’ may be used uninitialized in this function
38491
38492 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38493 Signed-off-by: Adam Jackson <ajax@redhat.com>
38494
38495commit 767b93e783a981b4fc926299a85a1a18387e693a
38496Author: Adam Jackson <ajax@redhat.com>
38497Date: Tue Feb 8 14:02:37 2011 -0500
38498
38499 glxproxy: warning fix
38500
38501 glxcmds.c: In function ‘__glXGetDrawableAttributes’:
38502 glxcmds.c:3295:8: warning: ‘screen’ may be used uninitialized in this function
38503 glxcmds.c:3298:8: warning: ‘attribs_size’ may be used uninitialized in this function
38504
38505 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38506 Signed-off-by: Adam Jackson <ajax@redhat.com>
38507
38508commit a391089186cd1063d807bf14a9651e6861b248de
38509Author: Adam Jackson <ajax@redhat.com>
38510Date: Tue Feb 8 13:55:23 2011 -0500
38511
38512 glxproxy: warning fix
38513
38514 glxcmds.c: In function ‘__glXChangeDrawableAttributes’:
38515 glxcmds.c:3464:8: warning: ‘screen’ may be used uninitialized in this function
38516
38517 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38518 Signed-off-by: Adam Jackson <ajax@redhat.com>
38519
38520commit 4102a00962f568b1d34cf72ad524e1cc705b0b1c
38521Author: Cyril Brulebois <kibi@debian.org>
38522Date: Wed Feb 23 14:44:23 2011 +0100
38523
38524 xfree86: Fix undefined reference to `XNFsprintf' on sparc.
38525
38526 Build failure on sparc:
38527 | copying selected object files to avoid basename conflicts...
38528 | CCLD Xorg
38529 | ./.libs/libxorg.a(xf86sbusBus.o): In function `xf86SbusConfigureNewDev':
38530 | […]/hw/xfree86/common/xf86sbusBus.c:712: undefined reference to `XNFsprintf'
38531 | collect2: ld returned 1 exit status
38532
38533 Fix typo introduced in:
38534 3a9bb93dd178084f4ff1abcea331ca5a62f88ce6
38535
38536 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38537 Apologized-for-by: Alan Coopersmith <alan.coopersmith@oracle.com>
38538 Signed-off-by: Cyril Brulebois <kibi@debian.org>
38539
38540commit 579ee8f5d84c3a523b7b3e3941eabb226d1d19e2
38541Merge: b636893 17265cc
38542Author: Peter Hutterer <peter.hutterer@who-t.net>
38543Date: Wed Feb 23 08:44:42 2011 +1000
38544
38545 Merge branch 'mi-cleanup' into next
38546
38547commit b636893137da1695e235e3a9354bfd9243fdddc2
38548Author: Daniel Stone <daniel@fooishbar.org>
38549Date: Tue Feb 22 13:43:28 2011 +0000
38550
38551 Input: Don't freeze unrelated devices in DeliverGrabbedEvent
38552
38553 When delivering an event to a device grabbed with SyncBoth,
38554 DeliverGrabbedEvent walks the device tree looking for associated devices
38555 to freeze them. Unfortunately, it froze all devices instead of just the
38556 paired device, and the previous fix in 4fbadc8b17237f3c would still break
38557 if the same client had a non-SyncBoth grab on another unrelated master
38558 device.
38559
38560 Fix this by completely ignoring devices that aren't our paired device.
38561
38562 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
38563 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38564 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38565
38566commit 17265ccb027e3f956bf7409106174f44621d1cb8
38567Author: Peter Hutterer <peter.hutterer@who-t.net>
38568Date: Fri Feb 18 12:28:43 2011 +1000
38569
38570 Move master/lastSlave out of the union into separate fields.
38571
38572 The removal of the double-use will cause some suble bugs as some conditions
38573 to check for the dev->u.master case were broken and also evaluated as true
38574 if lastSlave was set (instead of master).
38575
38576 Also breaks the input ABI.
38577
38578 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38579 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38580
38581commit 77113dd3eef03dd65b556b672d976817b3f4542e
38582Author: Peter Hutterer <peter.hutterer@who-t.net>
38583Date: Fri Feb 18 11:45:29 2011 +1000
38584
38585 input: Change a bunch of direct dev->u.master accesses to use GetMaster()
38586
38587 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38588 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38589
38590commit 2936635698619271a790004480a14285149277cb
38591Author: Peter Hutterer <peter.hutterer@who-t.net>
38592Date: Fri Feb 18 12:06:56 2011 +1000
38593
38594 xkb: Fix a check for MASTER_KEYBOARD
38595
38596 And copy into the master keyboard, not just the directly attached device.
38597
38598 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38599 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
38600 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38601
38602commit 0ba526deeb4d54ee1a75400291ee7a3673ed4d96
38603Author: Peter Hutterer <peter.hutterer@who-t.net>
38604Date: Fri Feb 18 12:04:46 2011 +1000
38605
38606 mi: check for MASTER_POINTER instead of manual checks.
38607
38608 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38609 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38610
38611commit 52c13896ce9a1e178d4eeed15f68020947cc20ed
38612Author: Peter Hutterer <peter.hutterer@who-t.net>
38613Date: Fri Feb 18 11:57:06 2011 +1000
38614
38615 mi: update macro to search for MASTER_POINTER
38616
38617 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38618 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38619
38620commit d270f12a3e44f4ea01f176a86783e8cd4c59ddf7
38621Author: Peter Hutterer <peter.hutterer@who-t.net>
38622Date: Fri Feb 18 11:55:51 2011 +1000
38623
38624 xfree86: update GetMaster check for the VCP.
38625
38626 Same result, but now also triggers on slave keyboards that send pointer
38627 events.
38628
38629 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38630 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38631
38632commit ce7f79efd588899f01c74f95c2f58d6dd6d816cd
38633Author: Peter Hutterer <peter.hutterer@who-t.net>
38634Date: Fri Feb 18 11:53:06 2011 +1000
38635
38636 dix: fix a master device access in change_modmap.
38637
38638 We need to check if our master keyboard is the given device since we may be
38639 a pointer with keys and thus need to change the modmap too.
38640
38641 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38642 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38643
38644commit e1ac704185dee31b427a46cd41a00ef7a28b4129
38645Author: Peter Hutterer <peter.hutterer@who-t.net>
38646Date: Fri Feb 18 11:42:09 2011 +1000
38647
38648 dix: fix up a master check in ChangeKeyboardMapping handling.
38649
38650 We don't just care about the directly attached master, we care about the
38651 master keyboard.
38652
38653 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38654 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38655
38656commit febce8cb814df46018f6ae1e6a9daea019b8ad0a
38657Author: Peter Hutterer <peter.hutterer@who-t.net>
38658Date: Fri Feb 18 11:32:28 2011 +1000
38659
38660 Xi: replace a direct master access with GetMaster()
38661
38662 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38663 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38664
38665commit df6559237a2d641b2fc38d14975beab9bae0d971
38666Author: Peter Hutterer <peter.hutterer@who-t.net>
38667Date: Fri Feb 18 11:30:10 2011 +1000
38668
38669 dix: add MASTER_ATTACHED as allowed type for GetMaster().
38670
38671 In some cases, we don't know/care whether we want the master pointer or keyboard
38672 for a device. Add a new type MASTER_ATTACHED to return the master this
38673 device is attached to.
38674
38675 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38676 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38677
38678commit dc57f89959e549403f8488eb9f23425bd7118b22
38679Author: Peter Hutterer <peter.hutterer@who-t.net>
38680Date: Fri Feb 11 12:47:37 2011 +1000
38681
38682 Switch to use IsFloating()
38683
38684 This is not a straightforward search/replacement due to a long-standing
38685 issue.
38686
38687 dev->u.master is the same field as dev->u.lastSlave. Thus, if dev is a master
38688 device, a check for dev->u.master may give us false positives and false
38689 negatives.
38690 The switch to IsFloating() spells out these cases and modifies the
38691 conditions accordingly to cover both cases.
38692
38693 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38694 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38695
38696commit e48bf3b6403dde33586ca0e421db61e402525453
38697Author: Peter Hutterer <peter.hutterer@who-t.net>
38698Date: Fri Feb 11 11:11:57 2011 +1000
38699
38700 xfree86: replace two inputInfo.pointer uses with device loops.
38701
38702 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38703 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38704
38705commit 703baece7ebd128a6742d0523d5b3ebe65126fa5
38706Author: Peter Hutterer <peter.hutterer@who-t.net>
38707Date: Fri Feb 11 12:19:26 2011 +1000
38708
38709 dix: Add IsFloating(device) wrapper.
38710
38711 Simplifies check for floating devices.
38712
38713 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38714 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38715
38716commit 77af45ebc3eda32dc534968cab8d5adfb01bd9e3
38717Author: Peter Hutterer <peter.hutterer@who-t.net>
38718Date: Fri Feb 11 12:48:30 2011 +1000
38719
38720 dix: Simplify retrieving the master device.
38721
38722 GetMaster() returns NULL for floating slaves.
38723
38724 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38725 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38726
38727commit dc32a23890776edf575bf18b3f3c079da6214340
38728Author: Peter Hutterer <peter.hutterer@who-t.net>
38729Date: Fri Feb 11 15:21:51 2011 +1000
38730
38731 Fix two incorrect checks for master devices.
38732
38733 These two were sideeffects of lastSlave being in the same field as the
38734 master. For devices generated by the master device directly, lastSlave was 0
38735 and the device would (with the old checks) be interpreted as floating.
38736
38737 Add the required checks to safeguard against master devices.
38738
38739 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38740 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38741
38742commit d63c979c7fe0f2b114b27e73ebe0a706be8840ae
38743Author: Peter Hutterer <peter.hutterer@who-t.net>
38744Date: Fri Feb 11 15:21:31 2011 +1000
38745
38746 dix: replace direct master access with GetMaster and temp. device.
38747
38748 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38749 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38750
38751commit c2a9e0a96983f22f721196513e75a53ea0b86d7e
38752Author: Peter Hutterer <peter.hutterer@who-t.net>
38753Date: Fri Feb 11 11:41:47 2011 +1000
38754
38755 mi: simplify a "check for keyboard" condition
38756
38757 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38758 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38759
38760commit 3a6160b408447ce620140849f962683a5d139cb9
38761Author: Peter Hutterer <peter.hutterer@who-t.net>
38762Date: Fri Feb 11 11:40:01 2011 +1000
38763
38764 mi: remove if 0 code.
38765
38766 Note sure why this is in if 0 but it's been that way since 2007.
38767
38768 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38769 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38770
38771commit d9987c8c45236bc9cfcaf716f4bfcac2a9d0e7e6
38772Author: Peter Hutterer <peter.hutterer@who-t.net>
38773Date: Fri Feb 11 11:22:45 2011 +1000
38774
38775 mi: Add a few comments explaining various cursor move functions.
38776
38777 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38778 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38779
38780commit 15fe86e69fc256342881112cd07565527e32435b
38781Author: Peter Hutterer <peter.hutterer@who-t.net>
38782Date: Fri Feb 11 11:01:19 2011 +1000
38783
38784 mi: remove static GenerateEvent variable.
38785
38786 Push into the respective devices. This should have no functional changes
38787 since we never warp more than one device at a time. In the glorious future
38788 with true multithreading, still the better thing to do.
38789
38790 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38791 Reviewed-by: Benjamin Tissoires <tissoire@cena.fr>
38792
38793commit 93a73993708b1345c86ec3ec06b02ed236595673
38794Author: Peter Hutterer <peter.hutterer@who-t.net>
38795Date: Mon Feb 21 15:40:40 2011 +1000
38796
38797 test: write some event → XI1 conversion tests.
38798
38799 Don't test everything, but hey, life is short and I'm trying to have one.
38800
38801 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38802 Reviewed-by: Keith Packard <keithp@keithp.com>
38803 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
38804
38805commit 4cdf1013771bc86fe2f6d9223bc4a46753bc918f
38806Author: Peter Hutterer <peter.hutterer@who-t.net>
38807Date: Mon Feb 21 15:32:57 2011 +1000
38808
38809 dix: a valuator number of 0 is valid (#34510)
38810
38811 For all but motion and proximity events, having no valuators is ok.
38812 Regression from 1.9, keyboard events are not converted to protocol events.
38813
38814 X.Org Bug 34510 <http://bugs.freedesktop.org/show_bug.cgi?id=34510>
38815
38816 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38817 Reviewed-by: Keith Packard <keithp@keithp.com>
38818 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
38819 Tested-by: Timo Aaltonen <timo.aaltonen@canonical.com>
38820
38821commit b4ef34d4664e0eaac7211f7a22a2025958aa1527
38822Author: Maarten Maathuis <madman2003@gmail.com>
38823Date: Sun Feb 20 11:59:41 2011 +0100
38824
38825 Revert "exa/mixed: Exclude frontbuffer from deferred pixmap handling."
38826
38827 This reverts commit 541b25038a5de74411a094570b407c5ae018c2ba.
38828
38829 - It turns out that the high latency was a driver problem.
38830 - catting a large amount of text turns out to look prettier when the
38831 throughput is lower, but it's not worth the loss for a minor
38832 improvement that may not even exist on someone else's computer.
38833
38834 Reviewed-by: Michel Dänzer <michel@daenzer.net>
38835 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
38836 Signed-off-by: Keith Packard <keithp@keithp.com>
38837
38838commit 443d75446bcbe7d97a23860a1e2d46c0b7e7cb26
38839Author: Keith Packard <keithp@keithp.com>
38840Date: Fri Feb 18 14:46:33 2011 -0800
38841
38842 Version bumped to 1.9.99.902 (1.10 RC2)
38843
38844 Signed-off-by: Keith Packard <keithp@keithp.com>
38845
38846commit 2c77aeb39f59f1a94cc603a2e6256a62ce785c36
38847Merge: 816d67d 402b329
38848Author: Keith Packard <keithp@keithp.com>
38849Date: Fri Feb 18 14:29:32 2011 -0800
38850
38851 Merge remote branch 'jcristau/for-keith'
38852
38853commit 816d67de2ba9bdfe652da32cb6447a3056342b98
38854Merge: a73c28f 3bbb70a
38855Author: Keith Packard <keithp@keithp.com>
38856Date: Fri Feb 18 12:20:26 2011 -0800
38857
38858 Merge remote branch 'whot/for-keith'
38859
38860commit a73c28f0bdafb1c5cb8129179188a99c0ca052e2
38861Author: Justin Dou <Justin.Dou@intel.com>
38862Date: Thu Feb 10 16:27:29 2011 -0500
38863
38864 Replace malloc with calloc to initialize the buffers[] as NULL in do_get_buffers function
38865
38866 The calling for allocate_or_reuse_buffer may fail due to some reason, e.g. out of memory.
38867 If the buffers[] were not initialized to be NULL, the following err_out may try to access an illegal memory, which will cause X crash afterward.
38868
38869 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
38870 Signed-off-by: Justin Dou <Justin.Dou@intel.com>
38871 Signed-off-by: Keith Packard <keithp@keithp.com>
38872
38873commit 0bc9b15a622377c57e617411ccd26c4b5bf8cba4
38874Merge: d1123b6 93cd4b1
38875Author: Keith Packard <keithp@keithp.com>
38876Date: Fri Feb 18 12:04:41 2011 -0800
38877
38878 Merge remote branch 'ajax/for-keithp'
38879
38880commit d1123b66ebbcf96dd816236f54befc568a5f7c68
38881Merge: 5d020c3 f6e4ace
38882Author: Keith Packard <keithp@keithp.com>
38883Date: Fri Feb 18 11:59:25 2011 -0800
38884
38885 Merge remote branch 'jturney/master'
38886
38887commit 649269d40667cfb387cb5286dd3519dd68f7dd80
38888Author: Bryce Harrington <bryce@canonical.com>
38889Date: Wed Feb 16 16:55:57 2011 -0800
38890
38891 Check for OOM condition in XISendDeviceHierarchyEvent
38892
38893 When system is out of memory, calloc can fail returning a NULL pointer.
38894 Check for this before dereferencing it, and bail out if it fails.
38895
38896 Ref.: https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/720445
38897
38898 Signed-off-by: Bryce Harrington <bryce@canonical.com>
38899 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38900 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38901
38902commit 31ddb7ef4f5bac2c13c199e74a0716e43b8dc38e
38903Author: Simon Thum <simon.thum@gmx.de>
38904Date: Wed Feb 2 00:03:44 2011 +0100
38905
38906 dix: avoid FP promotion during pointer acceleration
38907
38908 Signed-off-by: Simon Thum <simon.thum@gmx.de>
38909 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38910 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38911
38912commit 006157f203f8493bb3d18e333a3fd3d6cb10f8ea
38913Author: Simon Thum <simon.thum@gmx.de>
38914Date: Sat Sep 4 16:31:24 2010 +0200
38915
38916 dix: refactor scheme init
38917
38918 This makes it possible to init a scheme in one init call, so we
38919 get rid of the tightly coupled two-phase init used before.
38920
38921 Signed-off-by: Simon Thum <simon.thum@gmx.de>
38922 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38923 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38924
38925commit 38ffeec0c89e83afc62579dec221c325d667cc1e
38926Author: Simon Thum <simon.thum@gmx.de>
38927Date: Sun Feb 6 19:13:00 2011 +0100
38928
38929 xfree86: allow to check for options without warnings in the log
38930
38931 This allows set_percent_option in synaptics to work as described,
38932 and should generally enable to check option syntax without log spam.
38933
38934 Signed-off-by: Simon Thum <simon.thum@gmx.de>
38935 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38936 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38937
38938commit 2fbccc881bfab4bdc5b97f74a7e5bcec7fdeae20
38939Author: Simon Thum <simon.thum@gmx.de>
38940Date: Sun Feb 6 19:07:19 2011 +0100
38941
38942 fix percent options parsing
38943
38944 Signed-off-by: Simon Thum <simon.thum@gmx.de>
38945 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38946 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38947
38948commit 46b49796015762fd131bec96eff6caaa890bfd5f
38949Author: Daniel Stone <daniel@fooishbar.org>
38950Date: Tue Feb 15 11:28:02 2011 +0000
38951
38952 Input: Allow EventToCore to return multiple events
38953
38954 Some event types (notably Expose and GraphicsExpose) require multiple
38955 events, a la XI 1.x. Bring the EventToCore API in line with EventToXI's
38956 and allow it to generate multiple events.
38957
38958 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
38959 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38960
38961commit ea71495adabca52df44241c0eba39e6bceb58202
38962Author: Daniel Stone <daniel@fooishbar.org>
38963Date: Tue Feb 15 11:28:01 2011 +0000
38964
38965 Input: Reduce use of filter in DeliverEvents
38966
38967 Instead of switching on the event filter to determine delivery, use the
38968 event type instead.
38969
38970 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
38971 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38972 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38973
38974commit 8a0241656cecb6ab2fb377f02b14238d18e65fc1
38975Author: Daniel Stone <daniel@fooishbar.org>
38976Date: Tue Feb 15 11:28:00 2011 +0000
38977
38978 Input: Rename EventIsDeliverable return masks
38979
38980 Rename the return mask values for EventIsDeliverable:
38981 * CORE_MASK -> EVENT_CORE_MASK
38982 * XI_MASK -> EVENT_XI1_MASK
38983 * XI2_MASK -> EVENT_XI2_MASK
38984 * DONT_PROPAGATE_MASK -> EVENT_DONT_PROPAGATE_MASK
38985
38986 And don't undef them in dix/events.c, since they're supposed to be
38987 global.
38988
38989 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
38990 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
38991 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
38992
38993commit 737562257e8ee30b1c438e5160a61fbb26ca609e
38994Author: Daniel Stone <daniel@fooishbar.org>
38995Date: Tue Feb 15 11:27:59 2011 +0000
38996
38997 Input: Actually send Xi 1.x DeviceStateNotify events
38998
38999 When a client has selected for Xi 1.x DeviceStateNotify events, they
39000 should receive them when a DeviceFocusIn event is generated. The code
39001 to do this was there, but an incorrect test meant they were never being
39002 sent.
39003
39004 The "type" passed in is the XI2 type, the XI1 type is in event.type.
39005
39006 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
39007 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39008 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39009
39010commit da39d57a20d4281fc4099e356645fef980290030
39011Author: Daniel Stone <daniel@fooishbar.org>
39012Date: Tue Feb 15 11:27:57 2011 +0000
39013
39014 Input: Store passive-activating key in CheckDeviceGrabs
39015
39016 CheckDeviceGrabs will activate a passive grab for KeyPress and
39017 ButtonPress events. GrabInfoRec::activatingKey contains the keycode
39018 which activated the passive grab, so we can deactivate it later in
39019 ProcessOtherEvents.
39020
39021 Previously, CheckDeviceGrabs relied on its callers to set
39022 activatingKey, which not all callers were doing (I'm looking at you,
39023 ComputeFreezes). Just set it in CheckDeviceGrabs instead.
39024
39025 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
39026 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39027 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39028
39029commit 3bbb70a1a7b24d3d1375b20a13db7011cf961c86
39030Author: Peter Hutterer <peter.hutterer@who-t.net>
39031Date: Wed Feb 16 07:56:58 2011 +1000
39032
39033 xfree86: fix up an out-of-date comment.
39034
39035 InitInput simply initialises all input devices now.
39036
39037 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39038 Reviewed-by: Keith Packard <keithp@keithp.com>
39039
39040commit 402b329c3aa8ddbebaa1f593306a02d4cd6fed26
39041Author: Julien Cristau <jcristau@debian.org>
39042Date: Sun Jan 23 13:35:54 2011 +0100
39043
39044 glx: Work around wrong request lengths sent by mesa
39045
39046 mesa used to send too long requests for GLXDestroyPixmap,
39047 GLXDestroyWindow, GLXChangeDrawableAttributes, GLXGetDrawableAttributes
39048 and GLXGetFBConfigsSGIX.
39049
39050 Fixes a regression introduced in ec9c97c6bf70b523bc500bd3adf62176f1bb33a4
39051 X.Org bug#33324 <https://bugs.freedesktop.org/show_bug.cgi?id=33324>
39052
39053 Reported-by: xunx.fang@intel.com
39054 Signed-off-by: Julien Cristau <jcristau@debian.org>
39055 Reviewed-by: Adam Jackson <ajax@redhat.com>
39056
39057commit 1137c11be0f82049d28024eaf963c6f76e0d4334
39058Author: Julien Cristau <jcristau@debian.org>
39059Date: Wed Jan 26 13:06:53 2011 +0100
39060
39061 glx: fix BindTexImageEXT length check
39062
39063 The request is followed by a list of attributes.
39064
39065 X.Org bug#33449
39066
39067 Reported-and-tested-by: meng <mengmeng.meng@intel.com>
39068 Signed-off-by: Julien Cristau <jcristau@debian.org>
39069 Reviewed-by: Adam Jackson <ajax@redhat.com>
39070
39071commit a883cf1545abd89bb2cadfa659718884b56fd234
39072Author: Julien Cristau <jcristau@debian.org>
39073Date: Sun Jan 23 17:05:26 2011 +0100
39074
39075 glx: fix request length check for CreateGLXPbufferSGIX
39076
39077 The request is followed by an attribute list.
39078
39079 Signed-off-by: Julien Cristau <jcristau@debian.org>
39080 Reviewed-by: Adam Jackson <ajax@redhat.com>
39081
39082commit 0f9c6f2f822ff53b9d12ff4fa0b26cbeb7394ba5
39083Author: Alan Coopersmith <alan.coopersmith@oracle.com>
39084Date: Sun Feb 13 21:36:05 2011 -0800
39085
39086 xkb: Replace malloc(strlen) + strcpy with strdup
39087
39088 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
39089 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39090 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
39091
39092commit 682865c460945e1299f943561140f46439e2b4cb
39093Author: Alan Coopersmith <alan.coopersmith@oracle.com>
39094Date: Sun Feb 13 21:36:04 2011 -0800
39095
39096 XkbAddGeomProperty: Fix checks for malloc failure
39097
39098 Check the variable we just tried to malloc, not the string we're copying
39099 and already checked for NULL at the beginning of the function.
39100
39101 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
39102 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39103 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
39104
39105commit aac1b435664819008989ed19f73e9c89920602c5
39106Author: Alan Coopersmith <alan.coopersmith@oracle.com>
39107Date: Sun Feb 13 21:36:03 2011 -0800
39108
39109 Replace _XkbDupString with Xstrdup
39110
39111 The two functions have identical semantics, including safely returning
39112 NULL when NULL is passed in (which POSIX strdup does not guarantee).
39113
39114 Some callers could probably be adjusted to call libc strdup directly,
39115 when we know the input is non-NULL.
39116
39117 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
39118 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39119 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
39120
39121commit a4a2e814d5d0e6152307a301eda1d6fc1c555aaa
39122Author: Alan Coopersmith <alan.coopersmith@oracle.com>
39123Date: Sun Feb 13 21:36:02 2011 -0800
39124
39125 xkb: Use snprintf to measure string lengths instead of manual strlen math
39126
39127 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
39128 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39129 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
39130
39131commit c9f7b303a36ca501c6ecf1196c266ee8e8f49d2d
39132Author: Peter Hutterer <peter.hutterer@who-t.net>
39133Date: Fri Feb 11 13:50:10 2011 +1000
39134
39135 xfixes: calloc, not malloc the cursorScreenRec
39136
39137 Debugging NULL pointers is significantly easier than random memory.
39138 Plus, if new fields (such as pointer barriers) are added they may just be
39139 properly initialised.
39140
39141 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39142 Reviewed-by: Keith Packard <keithp@keithp.com>
39143
39144commit d3499556d8d83396fa2585bd00371a81e086be36
39145Author: Peter Hutterer <peter.hutterer@who-t.net>
39146Date: Thu Feb 10 15:12:14 2011 +1000
39147
39148 xkb: if the keymap failed to compile, load the default keymap instead.
39149
39150 We really need symbols, compat, keynames, vmods and types for a sensible keymap.
39151
39152 Try this in your xorg.conf.d snippets for all keyboards:
39153 Option "XkbLayout" "us"
39154 Option "XkbVariant" "nodeadkeys"
39155
39156 us(nodeadkeys) doesn't exist so xkbcomp provides everything but the symbols
39157 map. We say we want everything but don't _need_ anything, the server happily
39158 gives us a keymap with every key mapped to NoSymbol. This in turn isn't what
39159 we want after all.
39160
39161 So instead, require symbols, compat, keynames, vmods and types from the
39162 keymap and if that fails, load the default keymap instead. If that fails
39163 too, all bets are off.
39164
39165 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39166 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
39167
39168commit 47d1d2fed656c3a3b2600491078da90962c46934
39169Author: Peter Hutterer <peter.hutterer@who-t.net>
39170Date: Thu Feb 10 15:11:34 2011 +1000
39171
39172 xkb: split out keymap compilation.
39173
39174 Refactoring for simpler double-use in the next patch. No functional changes.
39175
39176 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39177 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
39178
39179commit 787ba25a8a3af52b38448a1a6f8c9704ea8b7905
39180Author: Carlos Garnacho <carlosg@gnome.org>
39181Date: Mon Feb 7 18:21:31 2011 +0100
39182
39183 Xi: make XIQueryPointer return the current modifiers/group as documented.
39184
39185 The previous XKB info was being returned instead of the current
39186 one, producing inconsistent results between the latest events
39187 and the modifiers/group returned by this call.
39188
39189 Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
39190 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
39191 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>`
39192 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39193
39194commit 85f9017393c9bb19553e9afcf554673a44a09993
39195Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39196Date: Tue Feb 8 11:10:10 2011 +0200
39197
39198 ProcXkbGetXkbByName: fix use of uninitialised bytes valgrind error.
39199
39200 ==9999== Syscall param writev(vector[...]) points to uninitialised byte(s)
39201 ==9999== at 0x4AB5154: writev (writev.c:51)
39202 ==9999== by 0x7C7C3: _XSERVTransWritev (Xtrans.c:912)
39203 ==9999== by 0x61C8B: FlushClient (io.c:924)
39204 ==9999== by 0x62423: WriteToClient (io.c:846)
39205 ==9999== by 0xCE39B: XkbSendMap (xkb.c:1408)
39206 ==9999== by 0xD247B: ProcXkbGetKbdByName (xkb.c:5814)
39207 ==9999== by 0x4AB53: Dispatch (dispatch.c:432)
39208 ==9999== by 0x205BF: main (main.c:291)
39209 ==9999== Address 0x557eb68 is 40 bytes inside a block of size 4,096 alloc'd
39210 ==9999== at 0x48334A4: calloc (vg_replace_malloc.c:467)
39211 ==9999== by 0x62567: WriteToClient (io.c:1065)
39212 ==9999== by 0x452EB: ProcEstablishConnection (dispatch.c:3685)
39213 ==9999== by 0x4AB53: Dispatch (dispatch.c:432)
39214 ==9999== by 0x205BF: main (main.c:291)
39215 ==9999== Uninitialised value was created by a stack allocation
39216 ==9999== at 0xD1910: ProcXkbGetKbdByName (xkb.c:5559)
39217
39218 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39219 Reviewed-by: Oliver McFadden <oliver.mcfadden@nokia.com>
39220 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39221
39222commit 8a34d7a8532c7ca013e67307f3baf200167abb92
39223Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39224Date: Tue Feb 8 11:10:11 2011 +0200
39225
39226 XkbSendNames: fix use of uninitialised bytes valgrind error.
39227
39228 ==537== Syscall param writev(vector[...]) points to uninitialised byte(s)
39229 ==537== at 0x4AB7154: writev (writev.c:51)
39230 ==537== by 0x8935B: _XSERVTransWritev (Xtrans.c:912)
39231 ==537== by 0x6C55F: FlushClient (io.c:924)
39232 ==537== by 0x6CCF3: WriteToClient (io.c:846)
39233 ==537== by 0xD51D3: XkbSendNames (xkb.c:3765)
39234 ==537== by 0xD8183: ProcXkbGetKbdByName (xkb.c:5825)
39235 ==537== by 0x27B7B: Dispatch (dispatch.c:432)
39236 ==537== by 0x205B7: main (main.c:291)
39237 ==537== Address 0x55899f2 is 154 bytes inside a block of size 1,896 alloc'd
39238 ==537== at 0x4834C48: malloc (vg_replace_malloc.c:236)
39239 ==537== by 0xD47AF: XkbSendNames (xkb.c:3642)
39240 ==537== by 0xD8183: ProcXkbGetKbdByName (xkb.c:5825)
39241 ==537== by 0x27B7B: Dispatch (dispatch.c:432)
39242 ==537== by 0x205B7: main (main.c:291)
39243 ==537== Uninitialised value was created by a heap allocation
39244 ==537== at 0x4834C48: malloc (vg_replace_malloc.c:236)
39245 ==537== by 0xD47AF: XkbSendNames (xkb.c:3642)
39246 ==537== by 0xD8183: ProcXkbGetKbdByName (xkb.c:5825)
39247 ==537== by 0x27B7B: Dispatch (dispatch.c:432)
39248 ==537== by 0x205B7: main (main.c:291)
39249
39250 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39251 Reviewed-by: Oliver McFadden <oliver.mcfadden@nokia.com>
39252 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39253
39254commit 87fbef9157a6f1e1318382e368d27942d7ad72ab
39255Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39256Date: Tue Feb 8 11:10:09 2011 +0200
39257
39258 ProcRRCreateMode: fix use of uninitialised bytes valgrind error.
39259
39260 ==543== Syscall param writev(vector[...]) points to uninitialised byte(s)
39261 ==543== at 0x4AB7154: writev (writev.c:51)
39262 ==543== by 0x8935B: _XSERVTransWritev (Xtrans.c:912)
39263 ==543== by 0x6C55F: FlushClient (io.c:924)
39264 ==543== by 0x6D013: FlushAllOutput (io.c:668)
39265 ==543== by 0x27A83: Dispatch (dispatch.c:453)
39266 ==543== by 0x205B7: main (main.c:291)
39267 ==543== Address 0x556dc8c is 12 bytes inside a block of size 4,096 alloc'd
39268 ==543== at 0x48334A4: calloc (vg_replace_malloc.c:467)
39269 ==543== by 0x6CE37: WriteToClient (io.c:1065)
39270 ==543== by 0x223A7: ProcEstablishConnection (dispatch.c:3685)
39271 ==543== by 0x27B7B: Dispatch (dispatch.c:432)
39272 ==543== by 0x205B7: main (main.c:291)
39273 ==543== Uninitialised value was created by a stack allocation
39274 ==543== at 0xA3350: ProcRRCreateMode (rrmode.c:289)
39275
39276 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39277 Reviewed-by: Oliver McFadden <oliver.mcfadden@nokia.com>
39278 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39279
39280commit 0ef5973860e17c5edc996c923610f7ad88b4dfbe
39281Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39282Date: Tue Feb 8 11:10:08 2011 +0200
39283
39284 ProcRRQueryVersion: fix use of uninitialised bytes valgrind error.
39285
39286 ==9999== Syscall param writev(vector[...]) points to uninitialised byte(s)
39287 ==9999== at 0x4AB5154: writev (writev.c:51)
39288 ==9999== by 0x7C7C3: _XSERVTransWritev (Xtrans.c:912)
39289 ==9999== by 0x61C8B: FlushClient (io.c:924)
39290 ==9999== by 0x62743: FlushAllOutput (io.c:668)
39291 ==9999== by 0x4AA5B: Dispatch (dispatch.c:453)
39292 ==9999== by 0x205BF: main (main.c:291)
39293 ==9999== Address 0x55711b9 is 1 bytes inside a block of size 4,096 alloc'd
39294 ==9999== at 0x48334A4: calloc (vg_replace_malloc.c:467)
39295 ==9999== by 0x62567: WriteToClient (io.c:1065)
39296 ==9999== by 0x452EB: ProcEstablishConnection (dispatch.c:3685)
39297 ==9999== by 0x4AB53: Dispatch (dispatch.c:432)
39298 ==9999== by 0x205BF: main (main.c:291)
39299 ==9999== Uninitialised value was created by a stack allocation
39300 ==9999== at 0x160E78: ProcRRQueryVersion (rrdispatch.c:37)
39301
39302 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39303 Reviewed-by: Oliver McFadden <oliver.mcfadden@nokia.com>
39304 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39305
39306commit 81daba8ce906bfbbe44cd71d0ff269ad34e2f6b5
39307Author: Peter Hutterer <peter.hutterer@who-t.net>
39308Date: Mon Jan 31 13:53:08 2011 +1000
39309
39310 Xi: constify XIChangeDeviceProperty()
39311
39312 We don't modify "value", make it official.
39313
39314 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39315 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
39316 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
39317
39318commit 93cd4b1034e3c8bed32c5acd391eadd7605e10f0
39319Author: Adam Jackson <ajax@redhat.com>
39320Date: Thu Jan 20 00:27:00 2011 -0500
39321
39322 shadow: Remove hw/xfree86/ from includes
39323
39324 Reviewed-by: Matt Turner <mattst88@gmail.com>
39325 Reviewed-by: Julien Cristau <jcristau@debian.org>
39326 Signed-off-by: Adam Jackson <ajax@redhat.com>
39327
39328commit 0c230728f3af2c4250e93a070c0851a597069447
39329Author: Adam Jackson <ajax@redhat.com>
39330Date: Thu Jan 20 00:25:57 2011 -0500
39331
39332 rootless: Remove hw/xfree86/ from includes
39333
39334 Reviewed-by: Matt Turner <mattst88@gmail.com>
39335 Reviewed-by: Julien Cristau <jcristau@debian.org>
39336 Signed-off-by: Adam Jackson <ajax@redhat.com>
39337
39338commit 505defd270e49cfbcbe6a04d1de817d305edb3c2
39339Author: Adam Jackson <ajax@redhat.com>
39340Date: Thu Jan 20 00:24:20 2011 -0500
39341
39342 damage: Remove hw/xfree86/ from includes
39343
39344 Reviewed-by: Matt Turner <mattst88@gmail.com>
39345 Reviewed-by: Julien Cristau <jcristau@debian.org>
39346 Signed-off-by: Adam Jackson <ajax@redhat.com>
39347
39348commit 59d8e3cbdba032c7354c28c86cbd155e9da6447a
39349Author: Adam Jackson <ajax@redhat.com>
39350Date: Thu Jan 20 00:23:07 2011 -0500
39351
39352 fb: Remove hw/xfree86/ from includes
39353
39354 Reviewed-by: Matt Turner <mattst88@gmail.com>
39355 Reviewed-by: Julien Cristau <jcristau@debian.org>
39356 Signed-off-by: Adam Jackson <ajax@redhat.com>
39357
39358commit 994b7b3dc97f9afa713fe636af45a78002a4366f
39359Author: Adam Jackson <ajax@redhat.com>
39360Date: Thu Jan 20 00:17:52 2011 -0500
39361
39362 xfree86: warning fix
39363
39364 Did you know that anonymous enums with function scope will not only
39365 override the enum values from global scope, but will be treated as
39366 entirely different types? C's type system just rules.
39367
39368 xf86Crtc.c: In function 'handle_detailed_monrec':
39369 xf86Crtc.c:1555:33: warning: comparison between 'enum det_monrec_source' and 'enum <anonymous>'
39370 xf86Crtc.c:1562:33: warning: comparison between 'enum det_monrec_source' and 'enum <anonymous>'
39371
39372 Reviewed-by: Matt Turner <mattst88@gmail.com>
39373 Reviewed-by: Julien Cristau <jcristau@debian.org>
39374 Signed-off-by: Adam Jackson <ajax@redhat.com>
39375
39376commit ae16c5796fdb22fb6b2f680fe5123bfd2c89a825
39377Author: Adam Jackson <ajax@redhat.com>
39378Date: Thu Jan 20 00:14:11 2011 -0500
39379
39380 fbdevhw: Remove the non-sysfs hack for mapping from PCI to fbdev
39381
39382 It's broken for devices with BARs above 4G, and the sysfs method should
39383 work everywhere anyway. As a pleasant side effect, this fixes some
39384 warnings:
39385
39386 fbdevhw.c: In function 'fbdev_open_pci':
39387 fbdevhw.c:333:4: warning: cast from pointer to integer of different size
39388 fbdevhw.c:334:4: warning: cast from pointer to integer of different size
39389 fbdevhw.c:336:4: warning: cast from pointer to integer of different size
39390 fbdevhw.c:337:4: warning: cast from pointer to integer of different size
39391
39392 Reviewed-by: Matt Turner <mattst88@gmail.com>
39393 Reviewed-by: Julien Cristau <jcristau@debian.org>
39394 Signed-off-by: Adam Jackson <ajax@redhat.com>
39395
39396commit f1b7c9cead94b520e6b96774d605d63d3f341b50
39397Author: Adam Jackson <ajax@redhat.com>
39398Date: Thu Jan 20 00:09:19 2011 -0500
39399
39400 int10: warning fix
39401
39402 helper_exec.c: In function 'pciCfg1in':
39403 helper_exec.c:507:4: warning: passing argument 2 of 'pci_device_cfg_read_u32' from incompatible pointer type
39404 /usr/include/pciaccess.h:153:5: note: expected 'uint32_t *' but argument is of type 'CARD32 *'
39405
39406 Reviewed-by: Matt Turner <mattst88@gmail.com>
39407 Reviewed-by: Julien Cristau <jcristau@debian.org>
39408 Signed-off-by: Adam Jackson <ajax@redhat.com>
39409
39410commit be186586e504197623c69637b8122fb814f59429
39411Author: Adam Jackson <ajax@redhat.com>
39412Date: Thu Jan 20 00:06:48 2011 -0500
39413
39414 xf86vidmode: warning fix
39415
39416 xf86VidMode.c: In function 'VidModeGetMonitorValue':
39417 xf86VidMode.c:637:19: warning: 'ret.i' may be used uninitialized in this function
39418
39419 Reviewed-by: Matt Turner <mattst88@gmail.com>
39420 Reviewed-by: Julien Cristau <jcristau@debian.org>
39421 Signed-off-by: Adam Jackson <ajax@redhat.com>
39422
39423commit f6e4ace9eabea1bb2a06dd86b6ffb954917a74ce
39424Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39425Date: Sun Jul 18 13:24:48 2010 +0100
39426
39427 Cygwin/X: Fix for Canadian Multilingual Standard keyboard layout handling
39428
39429 Add a keycode mapping for VK_OEM_8 as RCtrl, which is issued by Canadian
39430 Multilingual Standard layout
39431
39432 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39433 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
39434
39435commit b421f5e5122dbe125a5629969cc657bd966b4261
39436Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39437Date: Thu Apr 1 23:10:58 2010 +0100
39438
39439 Cygwin/X: Ignore MappingNotify events sent to clipboard integration client
39440
39441 Ignore MappingNotify events sent to clipboard integration client,
39442 xmodmap changes aren't of interest to it, but there is no mechanism
39443 to express that disinterest.
39444
39445 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39446 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
39447
39448commit e930aa7bbc7f04a128df8e6fd441aed5123a2eb6
39449Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39450Date: Tue Sep 14 16:02:53 2010 +0100
39451
39452 Cygwin/X: Add Turkish keyboard layouts to keyboard layout mapping table
39453
39454 0x0000041f "Turkish Q" => layout tr
39455 0x0001041f "Turkish F" => layout tr variant f
39456
39457 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39458 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
39459
39460commit ae99a6f895a3a5b6f71e2dd4874cd4bb0235aff0
39461Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39462Date: Sun Jul 18 13:42:29 2010 +0100
39463
39464 Cygwin/X: Make the keyboard layout detection logging a bit clearer
39465
39466 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39467 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
39468
39469commit 57ca09441ee5ff2a8deb95eaa17a2625a756d912
39470Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39471Date: Thu Jul 1 14:54:48 2010 +0100
39472
39473 Cygwin/X: Update mapping for Canadian keyboard layouts
39474
39475 0x00000c0c "Canadian French (legacy)" => layout ca variant fr-legacy
39476 0x00001009 "Canadian French" => layout ca variant fr
39477 0x00011009 "Canadian Multilingual Standard" => layout ca variant multix
39478
39479 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
39480 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39481 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
39482
39483commit be3e5bb50178406e46368be00860331e1b6c4093
39484Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39485Date: Wed Apr 14 21:45:29 2010 +0100
39486
39487 Cygwin/X: Better keycode debugging output
39488
39489 Replace useless #if 0/ErrorF/#endif with winDebug
39490
39491 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39492 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
39493
39494commit 81892bf6b7e3730ebd19318183734f55dbaa5d4b
39495Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39496Date: Wed Apr 14 21:34:39 2010 +0100
39497
39498 Cygwin/X: Generate X keycodes for multimedia keys
39499
39500 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39501 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
39502
39503commit 5d020c31351594c104fb46615ce41f256cd8e8b9
39504Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39505Date: Fri Jan 14 15:03:28 2011 +0200
39506
39507 xf86/xv: Only register PostValidateTree hook when there is work to do
39508
39509 If none of Xv ports were affected by window tree modifications we don't
39510 want scan the port list. To avoid useless scanning of port list
39511 PostValidateTree hook is only registered when ClipNotify was called for
39512 any port.
39513
39514 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39515 Reviewed-by: Ville Syrjälä <ville.syrjala@nokia.com>
39516
39517commit ddf32df5cc2e73b833f6e414276d938fdc38869e
39518Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39519Date: Thu Jan 13 20:50:16 2011 +0200
39520
39521 xf86/xv: Use PostValidateTree to do reput
39522
39523 ValidateTree calls first ClipNotify and later might call
39524 WindowExposures. To avoid useless double reput ClipNotify delays reput
39525 to WindowExposures or PostValidateTree.
39526
39527 PostValidatTree checks all ports if there is clip changes. On clip
39528 changes reput is done to move or scale the overlay.
39529
39530 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39531 Reviewed-by: Ville Syrjälä <ville.syrjala@nokia.com>
39532
39533commit 5678a41f6bc0ef8965bd50a88c5ef01ba7eb8b24
39534Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39535Date: Mon Jan 17 16:58:00 2011 +0200
39536
39537 xf86/xv: Fill color key on expose
39538
39539 If window gets exposed but clipboxes doesn't change drivers would avoid
39540 color key fill. This makes XResizeWindo&co to lose colorkey if
39541 background is painted.
39542
39543 To help drivers to avoid filling colorkey for each put server can
39544 provide helper function if there is exposed areas. Server can subtract
39545 exposed areas from filled region.
39546
39547 As a side effect we can avoid useless color key fills if window only
39548 moves in screen without background fills.
39549
39550 v3:
39551 * Change tracking to filled area to account for client initiated clip
39552 changes
39553 * Make overlaid XvPutImage behavior like textured XvPutImage or PutImage
39554 * Make region dynamically allocated only when required.
39555
39556 v4:
39557 * Simplify new driver interface to reduce duplicate code
39558
39559 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39560 Reviewed-by: Ville Syrjälä <ville.syrjala@nokia.com>
39561
39562commit 9b048ca420864e19862f63f4a491a4da0ff9b54f
39563Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39564Date: Mon Jan 17 15:07:28 2011 +0200
39565
39566 xf86/xv: Remove unused variable from XvPortRecPrivate
39567
39568 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39569 Reviewed-by: Ville Syrjälä <ville.syrjala@nokia.com>
39570
39571commit 4f8f803280894a8c41691e350b7a34bfd719fe05
39572Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39573Date: Thu Jan 13 20:21:09 2011 +0200
39574
39575 xf86/xv: Remove unused GC pointers
39576
39577 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39578 Reviewed-by: Ville Syrjälä <ville.syrjala@nokia.com>
39579
39580commit e012f2312b4bcdf31bb5d622ddd34571f3c12429
39581Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39582Date: Thu Jan 13 20:19:18 2011 +0200
39583
39584 xf86/xv: Remove copy paste code.
39585
39586 xf86XVFillKeyHelperDrawable can be used to implement
39587 xf86XVFillKeyHelper.
39588
39589 V2:
39590 * Remove RegionTranslate that clobbered parameter region.
39591
39592 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
39593 Reviewed-by: Ville Syrjälä <ville.syrjala@nokia.com>
39594
39595commit ea1ffd3e60bdcedbec5a6f28929f8677bf45d450
39596Merge: 12b0f7d 53602c3
39597Author: Keith Packard <keithp@keithp.com>
39598Date: Wed Feb 2 15:19:55 2011 -0800
39599
39600 Merge remote branch 'whot/for-keith'
39601
39602commit 12b0f7df2caa78d68b453aaa91248bb3577fd724
39603Author: Evan Broder <ebroder@mokafive.com>
39604Date: Tue Feb 1 16:09:50 2011 -0800
39605
39606 xfree86/modes: Be sure to only use new EDID for physical output dimensions
39607
39608 The EDID processing regards physical dimensions of 0mm x 0mm as
39609 invalid. Previously the old values for height and width would be
39610 preserved if none of the physical dimension specifications in the new
39611 EDID were considered valid.
39612
39613 This will come up in particular if first a monitor is connected to an
39614 output, and then a projector is connected. Since projectors generally
39615 report physical dimensions of 0mm x 0mm, this would result in the
39616 projector claiming to have the physical dimensions of the monitor.
39617
39618 Signed-off-by: Evan Broder <ebroder@mokafive.com>
39619 Reviewed-by: Adam Jackson <ajax@redhat.com>
39620 Signed-off-by: Keith Packard <keithp@keithp.com>
39621
39622commit 0b15033031d78836582bd1b641c3160c3803a708
39623Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
39624Date: Mon Jan 31 23:25:10 2011 +0100
39625
39626 xserver: enable TLS even if AIGLX is not enabled
39627
39628 This aligns the xorg server build with the mesa build, which is needed on
39629 systems where aiglx with dri support is not enabled. Else the following error is
39630 obtained when trying to load the software raster:
39631
39632 (EE) AIGLX error: dlopen of /usr/lib/dri/swrast_dri.so failed (/usr/lib/dri/swrast_dri.so: undefined symbol: _glapi_tls_Context)
39633 (EE) GLX: could not load software renderer
39634 (II) GLX: no usable GL providers found for screen 0
39635
39636 because mesa always enables TLS use in GLX, even if dri is not available.
39637
39638 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
39639 Signed-off-by: Julien Cristau <jcristau@debian.org>
39640 Signed-off-by: Keith Packard <keithp@keithp.com>
39641
39642commit 541b25038a5de74411a094570b407c5ae018c2ba
39643Author: Maarten Maathuis <madman2003@gmail.com>
39644Date: Sun Jan 30 14:06:16 2011 +0100
39645
39646 exa/mixed: Exclude frontbuffer from deferred pixmap handling.
39647
39648 - Apps like xterm can trigger a lot of fallback rendering.
39649 - This can lead to (annoyingly) high latencies, because you
39650 have to wait for the block handler.
39651 - You need a driver that doesn't directly access the front
39652 buffer to trigger this (NV50+ nouveau for example).
39653 - Repeatingly doing dmesg on an xterm with a bitmap font
39654 will reveal that you never see part of the text.
39655 - I have recieved at least one complaint in the past of slow
39656 terminal performance, which was related to core font
39657 rendering.
39658 - This does sacrifice some throughput, roughly 33% slower.
39659
39660 Reviewed-by: Michel Dänzer <michel@daenzer.net>
39661 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
39662 Signed-off-by: Keith Packard <keithp@keithp.com>
39663
39664commit af87f6367ef733d1a4f3cfca4eeb92bfd84c2b6f
39665Author: Maarten Maathuis <madman2003@gmail.com>
39666Date: Sun Jan 30 14:06:15 2011 +0100
39667
39668 exa: Only call driver FinishAccess hook if PrepareAccess hook succeeded.
39669
39670 Reviewed-by: Michel Dänzer <michel@daenzer.net>
39671 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
39672 Signed-off-by: Keith Packard <keithp@keithp.com>
39673
39674commit 648d4fe5172cad33c6f09e05dd3d8d7171ef1a7f
39675Author: Maarten Maathuis <madman2003@gmail.com>
39676Date: Sun Jan 30 14:06:14 2011 +0100
39677
39678 exa/driver: set pExaPixmap->use_gpu_copy to the right value
39679
39680 - Not sure if it was causing problems, but you never know.
39681
39682 Reviewed-by: Michel Dänzer <michel@daenzer.net>
39683 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
39684 Signed-off-by: Keith Packard <keithp@keithp.com>
39685
39686commit 246d40b685292fa6393ba22127152137494af2af
39687Merge: 70f2d39 8c1f75a
39688Author: Keith Packard <keithp@keithp.com>
39689Date: Wed Feb 2 14:51:33 2011 -0800
39690
39691 Merge remote branch 'jeremyhu/master'
39692
39693commit 70f2d39fbde38d0baa687f1755b6c4c686f23052
39694Merge: 3a91c2d a9d3c43
39695Author: Keith Packard <keithp@keithp.com>
39696Date: Wed Feb 2 14:43:19 2011 -0800
39697
39698 Merge remote branch 'jturney/master'
39699
39700commit 3a91c2d21c3231e2a011d95c3cfc01785430c1cf
39701Merge: ac0a00a 82b1eaa
39702Author: Keith Packard <keithp@keithp.com>
39703Date: Wed Feb 2 14:37:00 2011 -0800
39704
39705 Merge remote branch 'dbn/build-fixes'
39706
39707commit ac0a00a840a5d8d8374e5db2b849e5959550ca4e
39708Author: Adam Jackson <ajax@redhat.com>
39709Date: Thu Jan 20 23:01:03 2011 -0500
39710
39711 xselinux: Fix GetDrawableContext
39712
39713 M_DRAWABLE_PIXMAP is the lookup mask to dixLookupDrawable, and _not_ the
39714 type value in the drawable itself.
39715
39716 Signed-off-by: Adam Jackson <ajax@redhat.com>
39717 Reviewed-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
39718 Signed-off-by: Keith Packard <keithp@keithp.com>
39719
39720commit 8c1f75a1d4cb5a34b4d10f3f5068bccc4356f31a
39721Author: Jeremy Huddleston <jeremyhu@apple.com>
39722Date: Fri Jan 28 21:13:38 2011 -0800
39723
39724 XQuartz: Localization Updates
39725
39726 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
39727
39728commit 53602c36217487e0a80880e8328f89ab302d85c8
39729Author: Ville Syrjala <syrjala@sci.fi>
39730Date: Mon Jan 24 01:06:49 2011 +0200
39731
39732 xfree86/dga: Remove DGAIsDgaEvent()
39733
39734 DGAIsDgaEvent() is not used anymore.
39735
39736 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39737 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39738 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39739
39740commit 2081728b7432e949c274e0dbbabc296fa4db17f9
39741Author: Ville Syrjala <syrjala@sci.fi>
39742Date: Mon Jan 24 01:06:48 2011 +0200
39743
39744 xfree86/dga: Only send DGA events for master devices
39745
39746 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39747 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39748 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39749
39750commit 9f09b62e594321a8e45a33f4f13559827d660ed3
39751Author: Ville Syrjala <syrjala@sci.fi>
39752Date: Mon Jan 24 01:06:47 2011 +0200
39753
39754 xfree86/dga: Remove useless ifdefs
39755
39756 Move some variables to the scope where they are used.
39757
39758 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39759 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39760 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39761
39762commit 5c941c04ca68cdc2b236c524970bc23f91906af4
39763Author: Ville Syrjala <syrjala@sci.fi>
39764Date: Mon Jan 24 01:06:46 2011 +0200
39765
39766 xfree86/dga: DGAProcessKeyboardEvent() forgot ET_Internal
39767
39768 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39769 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39770 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39771
39772commit ce4e0f660beb097e90110c97d039981702f0b0bd
39773Author: Ville Syrjala <syrjala@sci.fi>
39774Date: Mon Jan 24 01:06:45 2011 +0200
39775
39776 xfree86/dga: Remove DGAMouseX and DGAMouseY
39777
39778 Previously some sort of absolute coordinates were sent out in
39779 the padding of the DGA2 Motion and Button events. DGAMouseX
39780 and DGAMouseY were used to keep track of said coordinates.
39781 libXxf86dga doesn't use that data for anything, and at least
39782 git history didn't show any past usage either. So let's just
39783 remove the last remnants of of this mess.
39784
39785 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39786 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39787 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39788
39789commit ee8faeadb60f51a36edce19e568600dab7c8b357
39790Author: Ville Syrjala <syrjala@sci.fi>
39791Date: Mon Jan 24 01:06:44 2011 +0200
39792
39793 xfree86/dga: DGA2 events are missing the dx/dy information
39794
39795 Copy dx/dy from the internal event to the DGA2 Motion/Button events.
39796 Do the same for Key events for the sake of keeping the code consistent.
39797
39798 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39799 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39800 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39801
39802commit e3fef815cf85cf59256f3edeaa21c1267898b42c
39803Author: Ville Syrjala <syrjala@sci.fi>
39804Date: Mon Jan 24 01:06:43 2011 +0200
39805
39806 xfree86/dga: Make mieq_install Bool
39807
39808 mieq_installed is used as a boolean, so why not make it such. Also
39809 it's a static variable, so the the explicit zero initialization can
39810 be removed.
39811
39812 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39813 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39814 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39815
39816commit f4cb25c2ed832bf4cfcd4d25198fe5a895430fed
39817Author: Ville Syrjala <syrjala@sci.fi>
39818Date: Mon Jan 24 01:06:42 2011 +0200
39819
39820 xfree86/dga: Remove the ET_DGAEvent handler only if it was installed
39821
39822 Remove the handler only if it was installed. Also mark it as
39823 uninstalled, otherwise it wouldn't get reinstalled after a
39824 server reset.
39825
39826 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39827 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39828 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39829
39830commit e1547a9f1f6d7991219b7711f4cacc2852dbf88c
39831Author: Ville Syrjala <syrjala@sci.fi>
39832Date: Mon Jan 24 01:06:41 2011 +0200
39833
39834 xfree86/dga: Install the DGA event handler when DGA2 is used
39835
39836 The ET_DGAEvent handler is only installed when a client
39837 requests relative events via DGA1. Do it also when a client
39838 requests DGA2 events.
39839
39840 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39841 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39842 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39843
39844commit 7b16e8633d5a95167aac0c19e43fcfe38bd4a6a4
39845Author: Ville Syrjala <syrjala@sci.fi>
39846Date: Mon Jan 24 01:06:40 2011 +0200
39847
39848 xfree86/input: Restore DGAStealKeyEvent()
39849
39850 DGA key event support was lost in commit
39851 8da0ff2d51086666d10ca7330d428e8610a4a0e3. Bring it back.
39852
39853 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39854 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39855 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39856
39857commit a2c481821997ae14aa5816895ae36db2ca2e4dec
39858Author: Ville Syrjala <syrjala@sci.fi>
39859Date: Mon Jan 24 01:06:39 2011 +0200
39860
39861 include: Fix a copy/paste error in a comment
39862
39863 Signed-off-by: Ville Syrjala <syrjala@sci.fi>
39864 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39865 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
39866
39867commit 77af0dc5217fc9c446097d6d487c8ddaf00bdd2a
39868Author: Peter Hutterer <peter.hutterer@who-t.net>
39869Date: Mon Jan 24 10:06:34 2011 +1000
39870
39871 xfree86: bump input minor ABI
39872
39873 The new valuator_mask_free() call requires a minor bump.
39874
39875 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39876
39877commit 5927e070b45fbe3fea4b394920ffd6f7d036f1ed
39878Author: Erkki Seppälä <erkki.seppala@vincit.fi>
39879Date: Fri Jan 21 15:56:22 2011 +0200
39880
39881 xkb: Cancel a key's repetition when its autorepeat is disabled.
39882
39883 When XkbChangeEnabledControls is called to disable key repetition of a
39884 certain key (or keys), currently ongoing repetition of that key was
39885 not cancelled. It was cancelled if ChangeKeyboardControl was used to
39886 disable key repetition globally.
39887
39888 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
39889 Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
39890 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
39891
39892commit a9d3c43a57cbaf42d9e96ad49cf12bbf1d275fe5
39893Author: Jon TURNEY <jon.turney@dronecode.org.uk>
39894Date: Wed Jan 19 15:57:59 2011 +0000
39895
39896 Fix compilation of xf86bigfont.c
39897
39898 xf86bigfont.c: In function 'XFree86BigfontExtensionInit':
39899 xf86bigfont.c:146: error: 'ProcXF86BigfontDispatch' undeclared (first use in this function)
39900 xf86bigfont.c:147: error: 'SProcXF86BigfontDispatch' undeclared (first use in this function)
39901
39902 It seems this has been broken since commit cbd4d5dbb70db62ba1cb79c7b904e6fa11f62d7e
39903 "delete pervasively use of DISPATCH_PROC" (2010-09-28), which is a bit worrying as
39904 that presumably indicates that no tinderbox is configuring with --enable-xf86bigfont.
39905
39906 In a similar fashion to that commit, fix by moving XFree86BigfontExtensionInit()
39907 below the definitions of the static dispatch functions it references.
39908
39909 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
39910 Reviewed-by: Julien Cristau <jcristau@debian.org>
39911
39912commit effaf9459c450f13e35bf263835d3e37d48ff29d
39913Author: Jeremy Huddleston <jeremyhu@apple.com>
39914Date: Sun Jan 23 22:35:34 2011 -0800
39915
39916 XQuartz: Use the default signal handler
39917
39918 This allows better interaction with CrashTracer
39919
39920 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
39921
39922commit 418bb57a3923311338baa8a85c0794a72e15ceeb
39923Author: Jeremy Huddleston <jeremyhu@apple.com>
39924Date: Thu Jan 13 17:12:17 2011 -0800
39925
39926 XQuartz: Call RRScreenSizeNotify after handling externally-generated XP_EVENT_DISPLAY_CHANGED
39927
39928 We get an XP_EVENT_DISPLAY_CHANGED event when our display configuration is
39929 changed. If this change was caused by hotplugging a monitor or Mac Display
39930 Preferences changes by the user, we need to call RRScreenSizeNotify in order
39931 to ensure new connections get the correct screen size.
39932
39933 http://xquartz.macosforge.org/trac/ticket/460
39934
39935 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
39936
39937commit 6f52b10b9e8068e3deafb2bbb64b62175a208c49
39938Author: Jeremy Huddleston <jeremyhu@apple.com>
39939Date: Thu Jan 13 16:08:46 2011 -0800
39940
39941 XQuartz: Bump version string to 2.6.1
39942
39943 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
39944
39945commit 1cfb96dd6151ebe84a9a26a428005db97e5d5577
39946Author: Jeremy Huddleston <jeremyhu@apple.com>
39947Date: Thu Jan 13 16:08:29 2011 -0800
39948
39949 XQuartz: Update copyright dates in bundle's plist
39950
39951 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
39952
39953commit be3be7580b6f6fd2f7fa4d4abfe5e1ab19470223
39954Merge: 57a1d9b a1d885f
39955Author: Keith Packard <keithp@keithp.com>
39956Date: Thu Jan 20 21:21:21 2011 -0800
39957
39958 Merge remote branch 'ajax/for-keithp'
39959
39960commit 57a1d9b85331f6fe19d5111f5163139572ffbf02
39961Merge: 24ce650 6423769
39962Author: Keith Packard <keithp@keithp.com>
39963Date: Thu Jan 20 21:16:24 2011 -0800
39964
39965 Merge remote branch 'whot/for-keith'
39966
39967commit 24ce650cf4f0c6fa72faecd38c53d40703e6c959
39968Merge: bbdf81a 09fd010
39969Author: Keith Packard <keithp@keithp.com>
39970Date: Thu Jan 20 21:11:53 2011 -0800
39971
39972 Merge remote branch 'jturney/jturney-framebuffer-resize-for-master'
39973
39974commit bbdf81a056be0ea645da17a642dad5eadef3a906
39975Author: Alan Coopersmith <alan.coopersmith@oracle.com>
39976Date: Tue Jan 18 19:26:20 2011 -0800
39977
39978 Add xorg.conf.d shadow man page pointing to xorg.conf man page
39979
39980 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
39981 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
39982 Tested-by: Gaetan Nadon <memsize@videotron.ca>
39983 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
39984 Signed-off-by: Keith Packard <keithp@keithp.com>
39985
39986commit a1d885fdd67503a442b348626d2eddf6d22419e8
39987Author: Adam Jackson <ajax@redhat.com>
39988Date: Wed Jun 23 13:28:55 2010 -0400
39989
39990 resource: Fix indentation
39991
39992 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
39993 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
39994 Signed-off-by: Adam Jackson <ajax@redhat.com>
39995
39996commit cb61cf5c99004ba3c76b504220c6728b5f2d2de6
39997Author: Adam Jackson <ajax@redhat.com>
39998Date: Thu Apr 15 18:36:55 2010 -0400
39999
40000 resource: Remove expectID hack
40001
40002 This is clearly meant to short-circuit the (modestly) expensive resource
40003 lookup in LegalNewID. The problem is that long-lived clients will
40004 eventually run completely through their XID space and start asking
40005 XC-MISC for IDs to reuse. Once that happens, the comparison against
40006 expectID will always be true, and we'll no longer catch XID collisions
40007 at all.
40008
40009 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40010 Signed-off-by: Adam Jackson <ajax@redhat.com>
40011
40012commit 3282e3c627f97f079e3a9af756a6b13bd9a5f227
40013Author: Adam Jackson <ajax@redhat.com>
40014Date: Thu Apr 15 08:46:28 2010 -0400
40015
40016 resource: s/NullResource/NULL/g
40017
40018 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40019 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40020 Signed-off-by: Adam Jackson <ajax@redhat.com>
40021
40022commit f953ae7d8a578d135a6faaf69d9c06eae7c85ede
40023Author: Adam Jackson <ajax@redhat.com>
40024Date: Mon Dec 14 14:38:10 2009 -0500
40025
40026 os: Reduce smart scheduler setup calls
40027
40028 We can return from WaitForSomething with no clients ready for any number
40029 of reasons. There's no reason to set up the scheduler timer when this
40030 happens.
40031
40032 Reviewed-by: Keith Packard <keithp@keithp.com>
40033 Signed-off-by: Adam Jackson <ajax@redhat.com>
40034
40035commit d127075da06239852c1cc745abfe63d0d180d984
40036Author: Adam Jackson <ajax@redhat.com>
40037Date: Thu Jan 20 19:03:33 2011 -0500
40038
40039 xdmxconfig: warning fix
40040
40041 xdmxconfig.c: In function ‘dmxConfigCanvasDraw’:
40042 xdmxconfig.c:299:23: warning: ‘maxHeight’ may be used uninitialized in this function
40043
40044 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40045 Signed-off-by: Adam Jackson <ajax@redhat.com>
40046
40047commit 7a08f9abef7219fabdab8d1d49e8d3afb042e36a
40048Author: Adam Jackson <ajax@redhat.com>
40049Date: Thu Jan 20 18:59:39 2011 -0500
40050
40051 dmx: warning fixes
40052
40053 dmxinputinit.c: In function ‘dmxBlockHandler’:
40054 dmxinputinit.c:610:44: warning: cast from pointer to integer of different size
40055 dmxinputinit.c: In function ‘dmxWakeupHandler’:
40056 dmxinputinit.c:637:41: warning: cast from pointer to integer of different size
40057 dmxinputinit.c: In function ‘dmxInputInit’:
40058 dmxinputinit.c:1041:36: warning: cast to pointer from integer of different size
40059
40060 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40061 Signed-off-by: Adam Jackson <ajax@redhat.com>
40062
40063commit ffd323b7c0212ed9b348e51cd9b36363d7c4d1f2
40064Author: Adam Jackson <ajax@redhat.com>
40065Date: Thu Jan 20 18:55:20 2011 -0500
40066
40067 dmx: warning fix
40068
40069 dmxinputinit.c: At top level:
40070 dmxinputinit.c:135:29: warning: ‘DMXCommonOth’ defined but not used
40071
40072 DMXCommonOth is actually mentioned in a #if 0 block, so delete it and
40073 the block that references it. If anyone needs it, git remembers.
40074
40075 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40076 Signed-off-by: Adam Jackson <ajax@redhat.com>
40077
40078commit c1fe0b155d0567440228aa5d9e36036f37670e3b
40079Author: Adam Jackson <ajax@redhat.com>
40080Date: Thu Jan 20 18:52:57 2011 -0500
40081
40082 dmx: warning fix
40083
40084 dmxgc.c: In function ‘dmxChangeClip’:
40085 dmxgc.c:386:5: warning: case label value exceeds maximum value for type
40086 dmxgc.c:387:5: warning: case label value exceeds maximum value for type
40087 dmxgc.c:388:5: warning: case label value exceeds maximum value for type
40088 dmxgc.c:389:5: warning: case label value exceeds maximum value for type
40089
40090 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40091 Signed-off-by: Adam Jackson <ajax@redhat.com>
40092
40093commit b02e006b2733ea457df41791f6054309e4edf7f6
40094Author: Adam Jackson <ajax@redhat.com>
40095Date: Thu Jan 20 18:46:00 2011 -0500
40096
40097 dmx: warning fixes
40098
40099 Dear gcc: I do not care about machines where sizeof(void *) <
40100 sizeof(int), and neither should you.
40101
40102 dmxextension.c: In function ‘dmxBECreateResources’:
40103 dmxextension.c:858:26: warning: cast from pointer to integer of different size
40104 dmxextension.c: In function ‘dmxBERestoreRenderPict’:
40105 dmxextension.c:1062:29: warning: cast from pointer to integer of different size
40106 dmxextension.c: In function ‘dmxBERestoreRenderGlyph’:
40107 dmxextension.c:1084:35: warning: cast from pointer to integer of different size
40108 dmxextension.c: In function ‘dmxAttachScreen’:
40109 dmxextension.c:1277:8: warning: cast to pointer from integer of different size
40110 dmxextension.c:1286:34: warning: cast to pointer from integer of different size
40111 dmxextension.c:1292:35: warning: cast to pointer from integer of different size
40112 dmxextension.c: In function ‘dmxBEDestroyResources’:
40113 dmxextension.c:1456:26: warning: cast from pointer to integer of different size
40114 dmxextension.c: In function ‘dmxDetachScreen’:
40115 dmxextension.c:1599:8: warning: cast to pointer from integer of different size
40116
40117 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40118 Signed-off-by: Adam Jackson <ajax@redhat.com>
40119
40120commit 64237697994871adfcf4905b5784e75cd7281579
40121Author: Peter Hutterer <peter.hutterer@who-t.net>
40122Date: Thu Jan 13 12:20:38 2011 +1000
40123
40124 xkb: after making changes to the xkb ctrls, copy them back into kbdfeed.
40125
40126 enabled_ctrls_changes nowhere near the usual event or config paths. So this
40127 condition always evaluated to false and the memcpy would thus never been
40128 hit. As a result, any modification to the XKB struct during
40129 XkbUpdateDescActions was not reflected in the kbdfeed ctrls.
40130 The flag that is set by XkbUpdateDescActions() if ctrls were changed are in
40131 enabled_ctrls.
40132
40133 This mainly affected keyboard repeat control as XKB uses the kbdfeed ctrls,
40134 not XKB's per_key_repeats, to determine if a key needs to be repeated. Thus,
40135 adding a "repeat= False" to the XKB map of any action did not have any
40136 effect.
40137
40138 Test case:
40139 assign Mode_switch to any key that by default repeats, e.g. the menu key.
40140
40141 key <COMP> { [ Mode_switch ] };
40142
40143 Then modify the Mode_switch action to not repeat the key.
40144
40145 interpret Mode_switch+AnyOfOrNone(all) {
40146 virtualModifier= AltGr;
40147 useModMapMods=level1;
40148 action= SetGroup(group=+1);
40149 // Add this line
40150 repeat= False;
40151 };
40152
40153 Though the flags are correctly reflected in the description loaded in the
40154 server, the change is not handed back to the kbdfeed struct and XKB will
40155 trigger softrepeats of this key.
40156
40157 This patch also adds two explanatory comments and an extra check, as this
40158 path may be hit before the CtrlProc for the kbdfeed struct is set.
40159
40160 Red Hat Bug 537708 <https://bugzilla.redhat.com/show_bug.cgi?id=537708>
40161
40162 Also fixes broken auto-repeat of the backspace key in the colemak layout
40163 (mapped to CapsLock).
40164
40165 X.Org Bug 16318 <http://bugs.freedesktop.org/show_bug.cgi?id=16318>
40166
40167 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
40168 Tested-by: Dirk Wallenstein <halsmit@t-online.de>
40169 Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
40170 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40171
40172commit f1326ed2910bd985fafdb48714b1d6f38116b083
40173Author: Peter Hutterer <peter.hutterer@who-t.net>
40174Date: Thu Jan 13 09:19:28 2011 +1000
40175
40176 xkb: Replace a few manual bitflips with SetBit & friends.
40177
40178 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
40179 Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
40180
40181commit 16eb4f2a7b67ee169785973f9e0a3773e13fdea6
40182Author: Peter Hutterer <peter.hutterer@who-t.net>
40183Date: Fri Jan 7 09:53:35 2011 +1000
40184
40185 Xi: reset remainders when warping the device.
40186
40187 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
40188 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40189 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40190 Reviewed-by: Simon Thum <simon.thum@gmx.de>
40191
40192commit 09fd010902fad56735b8069b1becb80d85bd6a35
40193Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40194Date: Wed Sep 29 22:54:22 2010 +0100
40195
40196 Cygwin/X: DirectDraw engines shouldn't try to blit if the surface wasn't allocated
40197
40198 Fix winShadowUpdateDD(|NL) so we don't try to blit to primary surface if it didn't get allocated
40199
40200 (Intel drivers, in particular, seem to like to issue a WM_DISPLAYCHANGE during a suspend/resume
40201 cycle, but not allow surface to be allocated right then)
40202
40203 Also:
40204 Use winReleasePrimarySurfaceShadowDD(|NL) in winFreeFBShadowDD(|NL) rather than open coding it
40205 Don't mess about recreating surface if we're going to resize it anyhow
40206
40207 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40208 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40209 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40210
40211commit 625ab9701fd75b879c1dafc05fa979591eea87c0
40212Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40213Date: Tue Aug 31 13:45:43 2010 +0100
40214
40215 Cygwin/X: Simplify and consolidate reporting of the bpp value we are going to use
40216
40217 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40218 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40219 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40220
40221commit 3f7339a7c5d3dcd05909b041865125f4cb6fa29e
40222Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40223Date: Tue Aug 31 10:13:01 2010 +0100
40224
40225 Cygwin/X: Deal with RANDR depth changes correctly in ShadowGDI drawing engine
40226
40227 Make ShadowGDI drawing engine only change the size of the screen
40228 pixmap/shadow framebuffer on an RANDR change, not the bpp/depth
40229 as well.
40230
40231 The server requires the screen pixmap's depth to be invariant.
40232
40233 Other drawing engines aren't quite as affected by this issue as
40234 they won't draw to the display, if it has changed colour depth,
40235 but probably still need some attention.
40236
40237 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40238 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40239 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40240
40241commit 8b22f83113fbdc09b932b5ad7e44f629fc15e3b5
40242Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40243Date: Mon Aug 30 17:17:57 2010 +0100
40244
40245 Cygwin/X: Use winUpdateFBPointer() in winshaddd.c rather than duplicating it inline
40246
40247 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40248 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40249 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40250
40251commit 8385c426f86e9955e9e570a46f75bddd3c10ca01
40252Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40253Date: Tue Aug 10 21:55:15 2010 +0100
40254
40255 Cygwin/X: Remove WIN_DIB_MAXIMUM_SIZE check
40256
40257 This is only relevant to pre-NT versions of Windows, which are all EOL.
40258
40259 Also, it's in the wrong place now as framebuffer can get resized.
40260
40261 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40262 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40263 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40264
40265commit 981ad1f364cf4fe8008c0f3592eb0f73dd14a118
40266Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40267Date: Tue Aug 10 16:24:57 2010 +0100
40268
40269 Cygwin/X: Fix a typo in command line argument validation code
40270
40271 Fortunately, these swapped constants are benign as they have the same
40272 value, 0
40273
40274 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40275 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40276 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40277
40278commit 5390b494672393506466d8afdb9b146b0e585cc0
40279Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40280Date: Tue Aug 10 13:16:13 2010 +0100
40281
40282 Cygwin/X: Don't turn off -multiplemonitors when all monitors don't have the same pixel format when using shadow GDI engine
40283
40284 Don't turn off -multiplemonitors when all monitors don't have the
40285 same pixel format and when using shadow GDI engine, just warn that
40286 performance may be degraded
40287
40288 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40289 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40290 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40291
40292commit 33106e1e807a828208b306512e78c5e3e93960d3
40293Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40294Date: Tue Mar 30 19:49:41 2010 +0100
40295
40296 Cygwin/X: Generate RANDR change on WM_DISPLAYCHANGE for rootless modes
40297
40298 When RANDR resizing is enabled, generate an internal RANDR change when
40299 WM_DISPLAYCHANGE occurs in rootless modes for screens which occupy an
40300 entire monitor or the virtual desktop.
40301
40302 Store the monitor number and use that to handle WM_DISPLAYCHANGE for a
40303 screen specified with '-screen @monitor'
40304
40305 In rooted mode, WM_DISPLAYCHANGE isn't relevant (except where display
40306 depth changes may cause problems). (A maximized screen window will get
40307 WM_SIZE to adjust it to the new monitor size)
40308
40309 In rooted fullscreen mode, WM_DISPLAYCHANGE shouldn't be seen, as we
40310 have the resolution we have selected for the fullscreen session)
40311 (Could client randr requests be handled in fullscreen to cause a change
40312 of the fullscreen resolution? )
40313
40314 Don't bother do a RANDR resize if the dimensions aren't actually changing
40315 when WM_DISPLAYCHANGE is sent (should handle WM_DISPLAYCHANGE to size 0x0
40316 that the intel driver seems to like to send)
40317
40318 Various debug output improvements
40319
40320 Also, remove the note that XWin can't handle display mode changes from
40321 the man page
40322
40323 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40324 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40325 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40326
40327commit bbc511e80b2a9365f6a1528bc1595772f83be654
40328Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40329Date: Mon Feb 15 13:42:04 2010 +0000
40330
40331 Cygwin/X: Make WM_SIZE use RandR resizing when -resize=randr
40332
40333 To avoid recursion, WM_SIZE requests shouldn't generate XRANDR requests
40334 when no change is neeeded.
40335
40336 We do the actual resize on WM_EXITSIZEMOVE, as resizing occurs in
40337 a modal loop, to avoid a backlog of resize events building up as
40338 the X server doesn't get a change to process anything until the resize
40339 is completed.
40340
40341 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40342 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40343 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40344
40345commit 85c497a8b6c488ef9ea2c6d7b49e6f9b992fb4a2
40346Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40347Date: Tue Mar 30 20:48:10 2010 +0100
40348
40349 Cygwin/X: Implement framebuffer resizing in RANDR extension
40350
40351 Implement framebuffer resizing in RANDR extension:
40352
40353 Resize the frame buffer, the screen's root window and the native window
40354 containing the root window image.
40355
40356 Correctly allow for decorations in new native window size when resizing native window
40357 to fit the new framebuffer size with AdjustWindowRectEx()
40358
40359 Update physical size info for a screen when it is changed by RANDR
40360
40361 Forbid client-requested RANDR changes in fullscreen and rootless modes
40362
40363 Only resize window on an external RandR request, to avoid recursing on
40364 a WM_SIZE requested resize.
40365
40366 Also, add prototypes for winRandRInit() and winDoRandRScreenSetSize() to header file
40367
40368 Also, update the author list and copyright for winrandr.c
40369
40370 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40371 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40372 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40373
40374commit 873abef315f5d947b864428891381bff539c5869
40375Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40376Date: Tue Mar 30 20:07:19 2010 +0100
40377
40378 Cygwin/X: Add -resize command line option
40379
40380 Add -resize command line option to configure how native window sizing frame
40381 is used.
40382
40383 In additions to the existing fixed and scrollbars modes, add a new mode to
40384 allow framebuffer to be resized using native window frame
40385
40386 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40387 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40388 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40389
40390commit 46c57788539d8a5f0246528b4f88ad4ed6d867d1
40391Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40392Date: Tue Feb 23 15:44:36 2010 +0000
40393
40394 Cygwin/X: Move QueryMonitor() out of windprocarg.c
40395
40396 Move QueryMonitor() out of windprocarg.c into a new file, winmonitors.c,
40397 as we use to use it from other places as well
40398
40399 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40400 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40401 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40402
40403commit 73b02e964787ea9fb2d139dc781c9b2495ccb031
40404Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40405Date: Fri Feb 12 12:29:30 2010 +0000
40406
40407 Cygwin/X: Remove an attempt at detecting if WM_DISPLAYCHANGE affects the X screen
40408
40409 Remove an attempt at the rather difficult optimization of detecting
40410 if WM_DISPLAYCHANGE affects any of the monitors which intersect the
40411 native window for the X screen.
40412
40413 We'll always act as if it does, which it probably usually the case.
40414
40415 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40416 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40417 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40418
40419commit a46146af5f6c2d96728c43f0df369ea00bcae017
40420Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40421Date: Thu Apr 1 21:00:43 2010 +0100
40422
40423 Cygwin/X: Rather than storing calculated physical display sizes, calculate them when needed
40424
40425 (This stored calculation was wrong if -dpi came after -screen on the command
40426 line, anyhow)
40427
40428 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40429 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40430 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40431
40432commit df518001271dbc0c49935350b755f0c0562cdb92
40433Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40434Date: Tue Oct 13 14:47:33 2009 +0100
40435
40436 Cygwin/X: Refactor framebuffer allocation/release in drawing engines
40437
40438 Refactor the drawing engines so we can explicitly allocate and release the framebuffer for a screen
40439
40440 Move the setting of dwPaddedWidth into the DDNL engine, so it is updated when the framebuffer changes size
40441
40442 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40443 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40444 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40445
40446commit 708d0b9781c6f3f0c0ae7052f45223b1c2524809
40447Author: Jon TURNEY <jon.turney@dronecode.org.uk>
40448Date: Tue Oct 13 14:48:17 2009 +0100
40449
40450 Cygwin/X: Enable building of nativegdi and primaryfb engines to avoid further rusting
40451
40452 Some trivial build fixes required
40453
40454 Also fill out all function pointers for primaryfb engine
40455
40456 Also tidy up the man page section describing drawing engines.
40457
40458 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
40459 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
40460 Tested-by: Colin Harrison <colin.harrison@virgin.net>
40461
40462commit c6aa4755ec355101a62bef86dbb090262fe806f6
40463Author: Alan Coopersmith <alan.coopersmith@oracle.com>
40464Date: Sat Jan 15 12:13:39 2011 -0800
40465
40466 xkb/ddxLoad.c doesn't need <paths.h> any more
40467
40468 Was previously used for _PATH_VARTMP, but that was removed in
40469 534fc5140b039a8c98ab715d0a6740d513b41209
40470
40471 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40472 Reviewed-by: Julien Cristau <jcristau@debian.org>
40473 Signed-off-by: Keith Packard <keithp@keithp.com>
40474
40475commit abfea17342da774aa4e3a351a73c8f2af6e6fd28
40476Merge: 3611283 e65c3f8
40477Author: Keith Packard <keithp@keithp.com>
40478Date: Tue Jan 18 15:19:34 2011 -0800
40479
40480 Merge remote branch 'ajax/for-keithp'
40481
40482commit 361128389e5cb0101cbd091ff8de77cf34608f6c
40483Merge: 65ceaad d9225b9
40484Author: Keith Packard <keithp@keithp.com>
40485Date: Tue Jan 18 15:18:08 2011 -0800
40486
40487 Merge remote branch 'jcristau/for-keith'
40488
40489commit 65ceaadbf0fb65f1962cb96d5b87fd1175f85a3b
40490Author: Gaetan Nadon <memsize@videotron.ca>
40491Date: Sun Jan 9 20:28:31 2011 -0500
40492
40493 Xquartz: rename man page directory from doc to man
40494
40495 Following the convention makes it easier to locate man pages,
40496 user's or developer's documentation and specifications.
40497
40498 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40499 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40500 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40501 Signed-off-by: Keith Packard <keithp@keithp.com>
40502
40503commit 7ed0d426f616fe9e44ac7091622a798f2043dfe4
40504Author: Gaetan Nadon <memsize@videotron.ca>
40505Date: Sun Jan 9 20:28:30 2011 -0500
40506
40507 xfree86: use sed rather than cpp to perform string substitutions
40508
40509 Makefile.am: there are only 4 string substitutions to be done in this file.
40510 SED is much simpler than the C pre processor which adds its own
40511 strings which must be substituted by sed, still.
40512
40513 xorgconf.cpp: replaced __xconfigfile__ with xorg.conf as this file name
40514 is hard coded in the xserver configuration and cannot change.
40515 Replace XCOMM with # permanently.
40516
40517 Delete cpprules.in as it isn't used anywhere else. Should one need
40518 cpprules for real cpp work, there is one in Xquartz from which the
40519 the old man pages code have been stripped.
40520
40521 Fix trailing spaces.
40522
40523 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40524 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40525 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40526 Signed-off-by: Keith Packard <keithp@keithp.com>
40527
40528commit 885a3330092c13786baa3aaabd728e71962486c4
40529Author: Gaetan Nadon <memsize@videotron.ca>
40530Date: Sun Jan 9 20:28:29 2011 -0500
40531
40532 man: refactor common code in the man pages makefiles
40533
40534 Create a manpages.am makefile snippet in the root dir.
40535 Each man page makefile includes manpages.am.
40536
40537 Now all man pages in xserver are generated the same way
40538 using the same method as all of other xorg modules.
40539 All ".man.pre" files in git are ".man" now.
40540 Links are no longer created between different file types.
40541
40542 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40543 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40544 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40545 Signed-off-by: Keith Packard <keithp@keithp.com>
40546
40547commit c86f7c2b8eedd3126066a3aff44bc8767eb5e277
40548Author: Gaetan Nadon <memsize@videotron.ca>
40549Date: Sun Jan 9 20:28:28 2011 -0500
40550
40551 Xdmx: build all man pages using XORG_MANPAGE_SECTIONS
40552
40553 Use standard directory and makefile.
40554
40555 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40556 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40557 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40558 Signed-off-by: Keith Packard <keithp@keithp.com>
40559
40560commit 9bd9d550a0d6fb4fe3daacb604658de92e69d99c
40561Author: Gaetan Nadon <memsize@videotron.ca>
40562Date: Sun Jan 9 20:28:27 2011 -0500
40563
40564 exa: build man pages using XORG_MANPAGE_SECTIONS
40565
40566 Nothing requires the use of a C preprocessor
40567 Using standard file extensions (.man) means no need for .gitignore
40568 Use standard directory and makefile
40569 Fix trailing whitespaces
40570
40571 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40572 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40573 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40574 Signed-off-by: Keith Packard <keithp@keithp.com>
40575
40576commit d4b1fd3ff66d1a3e44a85aed34d9a01eee0d37fd
40577Author: Gaetan Nadon <memsize@videotron.ca>
40578Date: Sun Jan 9 20:28:26 2011 -0500
40579
40580 fbdevhw: build man pages using XORG_MANPAGE_SECTIONS
40581
40582 Nothing requires the use of a C preprocessor
40583 Using standard file extensions (.man) means no need for .gitignore
40584 Use standard directory and makefile
40585 Fix trailing whitespaces
40586
40587 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40588 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40589 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40590 Signed-off-by: Keith Packard <keithp@keithp.com>
40591
40592commit b9276cb86b433bc740fe26c2e624d3d5399732f6
40593Author: Gaetan Nadon <memsize@videotron.ca>
40594Date: Sun Jan 9 20:27:53 2011 -0500
40595
40596 XFree86 utils: build utils man pages using XORG_MANPAGE_SECTIONS
40597
40598 Nothing requires the use of a C preprocessor
40599 Using standard file extensions (.man) means no need for .gitignore
40600 Use standard directory and makefile
40601 Fix trailing whitespaces
40602
40603 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40604 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40605 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40606 Signed-off-by: Keith Packard <keithp@keithp.com>
40607
40608commit ac5ffef60e05eb05275913b133786709881e7e7e
40609Author: Gaetan Nadon <memsize@videotron.ca>
40610Date: Sun Jan 9 20:27:52 2011 -0500
40611
40612 Xephyr: build Xephyr man pages using XORG_MANPAGE_SECTIONS
40613
40614 Nothing requires the use of a C preprocessor
40615 Using standard file extensions (.man) means no need for .gitignore
40616 Use standard directory and makefile
40617 Fix trailing whitespaces
40618
40619 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40620 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40621 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40622 Signed-off-by: Keith Packard <keithp@keithp.com>
40623
40624commit 9240cffd659b1c13d66162200c6e384ea1cdbc44
40625Author: Gaetan Nadon <memsize@videotron.ca>
40626Date: Sun Jan 9 20:27:51 2011 -0500
40627
40628 Xvfb: build Xvfb man pages using XORG_MANPAGE_SECTIONS
40629
40630 Nothing requires the use of a C preprocessor
40631 Using standard file extensions (.man) means no need for .gitignore
40632 Use standard directory and makefile
40633 Fix trailing whitespaces
40634
40635 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40636 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40637 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40638 Signed-off-by: Keith Packard <keithp@keithp.com>
40639
40640commit 3f4c71799b3164440624d09b7b631d3d3c0a29ef
40641Author: Gaetan Nadon <memsize@videotron.ca>
40642Date: Sun Jan 9 20:27:50 2011 -0500
40643
40644 XWin: build XWin man pages using XORG_MANPAGE_SECTIONS
40645
40646 Nothing requires the use of a C preprocessor
40647 Using standard file extensions (.man) means no need for .gitignore
40648 Use standard directory and makefile
40649 Now can be built easily on any platform in the man directory
40650
40651 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40652 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40653 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40654 Signed-off-by: Keith Packard <keithp@keithp.com>
40655
40656commit b8492e18a7e5322c4a492f6cf6cfeb5b0221d994
40657Author: Gaetan Nadon <memsize@videotron.ca>
40658Date: Sun Jan 9 20:27:49 2011 -0500
40659
40660 Xnest: build Xnest man pages using XORG_MANPAGE_SECTIONS
40661
40662 Nothing requires the use of a C preprocessor
40663 Using standard file extensions (.man) means no need for .gitignore
40664 Use standard directory and makefile
40665
40666 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40667 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40668 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40669 Signed-off-by: Keith Packard <keithp@keithp.com>
40670
40671commit 57ca20d26614a0787cdd0db91a2eb8cdb301f2dc
40672Author: Gaetan Nadon <memsize@videotron.ca>
40673Date: Sun Jan 9 20:27:48 2011 -0500
40674
40675 Xquartz: strip the man page code from cpprules.in
40676
40677 This is dead code which will not be used in the future.
40678
40679 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40680 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40681 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40682 Signed-off-by: Keith Packard <keithp@keithp.com>
40683
40684commit c06c46acdc2ead20a6030118fad6092b3da6709f
40685Author: Gaetan Nadon <memsize@videotron.ca>
40686Date: Sun Jan 9 20:27:47 2011 -0500
40687
40688 Xquartz: build Xquartz man pages using XORG_MANPAGE_SECTIONS
40689
40690 Nothing requires the use of a C preprocessor
40691 Using standard file extensions (.man)
40692 Use standard directory and makefile
40693
40694 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40695 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40696 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40697 Signed-off-by: Keith Packard <keithp@keithp.com>
40698
40699commit a9e85013b1045c7e02891a019bfa8042a219b7f1
40700Author: Gaetan Nadon <memsize@videotron.ca>
40701Date: Sun Jan 9 20:27:46 2011 -0500
40702
40703 man: build Xorg and xorg.conf man pages using XORG_MANPAGE_SECTIONS
40704
40705 Nothing requires the use of a C preprocessor
40706 Using standard file extensions (.man) means no need for .gitignore
40707 Use standard directory and makefile
40708
40709 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40710 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40711 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40712 Signed-off-by: Keith Packard <keithp@keithp.com>
40713
40714commit dcdd8ea82c0e2f657abdc80a22aa0c01ead07e54
40715Author: Gaetan Nadon <memsize@videotron.ca>
40716Date: Sun Jan 9 20:27:45 2011 -0500
40717
40718 Xorg.man: fix whitespace issues
40719
40720 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40721 Reviewed-by: Rémi Cardona <remi@gentoo.org>
40722 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
40723 Signed-off-by: Keith Packard <keithp@keithp.com>
40724
40725commit d9225b9602c85603ae616a7381c784f5cf5e811c
40726Author: Julien Cristau <jcristau@debian.org>
40727Date: Wed Nov 10 22:39:54 2010 +0100
40728
40729 glx: validate numAttribs field before using it
40730
40731 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
40732 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40733 Signed-off-by: Julien Cristau <jcristau@debian.org>
40734
40735commit 62319e8381ebd645ae36b25e5fc3c0e9b098387b
40736Author: Julien Cristau <jcristau@debian.org>
40737Date: Sun Aug 22 16:20:45 2010 +0100
40738
40739 glx: swap the request arrays entirely, not just half of them
40740
40741 Various glx requests include a list of pairs of attributes. We were
40742 only swapping the first half.
40743
40744 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
40745 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40746 Signed-off-by: Julien Cristau <jcristau@debian.org>
40747
40748commit 6c69235a9dfc52e4b4e47630ff4bab1a820eb543
40749Author: Julien Cristau <jcristau@debian.org>
40750Date: Sun Aug 22 00:50:05 2010 +0100
40751
40752 glx: check request length before swapping
40753
40754 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
40755 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40756 Signed-off-by: Julien Cristau <jcristau@debian.org>
40757
40758commit ec9c97c6bf70b523bc500bd3adf62176f1bb33a4
40759Author: Julien Cristau <jcristau@debian.org>
40760Date: Sat Jul 3 19:47:55 2010 +0100
40761
40762 glx: validate request lengths
40763
40764 Reviewed-by: Adam Jackson <ajax@redhat.com>
40765 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
40766 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40767 Signed-off-by: Julien Cristau <jcristau@debian.org>
40768
40769commit 3f0d3f4d97bce75c1828635c322b6560a45a037f
40770Author: Julien Cristau <jcristau@debian.org>
40771Date: Sat Jul 3 19:42:26 2010 +0100
40772
40773 glx: make sure screen is non-negative in validGlxScreen
40774
40775 Reviewed-by: Adam Jackson <ajax@redhat.com>
40776 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
40777 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40778 Signed-off-by: Julien Cristau <jcristau@debian.org>
40779
40780commit 82b1eaa6cad20f39dbf15573bdb3d62acbcd91f9
40781Author: Tom Fogal <tfogal@alumni.unh.edu>
40782Date: Wed Dec 22 14:39:48 2010 -0700
40783
40784 Autodetect TLS support instead of defaulting to disabled.
40785
40786 Uses the AX_TLS macro to figure out if the current platform
40787 supports TLS. If it does, enable TLS automagically. Still
40788 respects --enable option, regardless.
40789
40790 Signed-off-by: Tom Fogal <tfogal@alumni.unh.edu>
40791 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
40792
40793commit 93393411e39fbb0f72cf6c2f79ebde177aa86932
40794Author: Tom Fogal <tfogal@alumni.unh.edu>
40795Date: Wed Dec 22 16:03:03 2010 -0700
40796
40797 Add macro for detecting thread local storage support.
40798
40799 This adds an AX_TLS macro which attempts to identify if the
40800 underlying compiler/platform supports thread local storage (TLS).
40801 The macro comes from the autoconf macro archive.
40802
40803 Signed-off-by: Tom Fogal <tfogal@alumni.unh.edu>
40804 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
40805
40806commit 60801ff8703ec36995139a6dd7b1fc26ae8854a6
40807Author: Dan Nicholson <dbn.lists@gmail.com>
40808Date: Tue Nov 23 11:38:50 2010 -0800
40809
40810 xfree86: Convert libxf86config to static libtool library
40811
40812 In order to use libxf86config in a shared library, all the code must be
40813 compiled with -fPIC. Add proper PIC support for libxf86config by turning
40814 it into a libtool library. However, since we don't want to guarantee API
40815 or ABI stability, make sure it's only built static.
40816
40817 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
40818 Acked-by: Gaetan Nadon <memsize@videotron.ca>
40819
40820commit 8456625d64bb0013a496f3b56ea3aa20cf7a6142
40821Merge: aa30a86 ce74e75
40822Author: Peter Hutterer <peter.hutterer@who-t.net>
40823Date: Fri Jan 7 09:35:07 2011 +1000
40824
40825 Merge branch 'for-peter' of git://people.freedesktop.org/~daniels/xserver into for-keith
40826
40827commit aa30a86583c3fc1a83c46e8c7583182d750edcde
40828Author: Tomas Carnecky <tom@dbservice.com>
40829Date: Sun Aug 1 21:28:18 2010 +0200
40830
40831 NO_PANORAMIX preprocessor symbol is nowhere defined
40832
40833 Which makes it useless, so remove it.
40834
40835 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
40836 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40837 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
40838
40839commit fc6cbee772bafabf7ddd7a75043bd55bb78ad09e
40840Author: Peter Hutterer <peter.hutterer@who-t.net>
40841Date: Thu Jan 6 12:58:00 2011 +1000
40842
40843 input: add valuator_mask_free() to free a valuator mask.
40844
40845 Expecting the caller to free the mask requires us to keep it in a single
40846 memory block (which may be an issue lateron), aside from leaving the API
40847 asymetrical. Provide valuator_mask_free() to free the memory and reset the
40848 mask pointer to NULL.
40849
40850 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
40851 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
40852 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40853
40854commit 4fbadc8b17237f3c7f8c0bf56003d1139c86655e
40855Author: Daniel Stone <daniel@fooishbar.org>
40856Date: Thu Jan 6 11:32:26 2011 +1000
40857
40858 Input: Prevent MD grabs from wandering on to other MDs
40859
40860 The code to set sync.other in DeliverGrabbedEvents is supposed to reset
40861 sync.other for a paired MD to the grab under consideration, but was
40862 rather optimistic in resetting sync.other for _all_ devices.
40863
40864 This would fall apart given two sets of MDs (A paired with B, Y paired
40865 with Z), where both MDs were in FREEZE_BOTH_NEXT_EVENT due to being
40866 called with SyncBoth, where no event had yet triggered the grab. An
40867 event being processed on MD A would result in B, Y and Z all having
40868 sync.other set to A's grab, rather than just B.
40869
40870 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40871 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40872 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
40873
40874commit e65c3f8bcc86845f21ac575e2bfb4b21b67d5ebf
40875Author: Adam Jackson <ajax@redhat.com>
40876Date: Wed Jan 5 12:04:19 2011 -0500
40877
40878 dix: Add a Screen method for additional cursor confinement
40879
40880 This just reserves the slot in the ABI. Confining cursors to CRTCs will
40881 come soon.
40882
40883 v2: Just reserve the slot.
40884
40885 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
40886 Signed-off-by: Adam Jackson <ajax@redhat.com>
40887
40888commit ce74e7562dcc800306ff92b6d3c5ee885cd4763d
40889Author: Daniel Stone <daniel@fooishbar.org>
40890Date: Tue Dec 7 11:48:14 2010 +0000
40891
40892 Input: Make CheckPassiveGrabsOnWindow return grab, export
40893
40894 Change CheckPassiveGrabsOnWindow to return the GrabPtr it used (or NULL
40895 if none) rather than a boolean, and export it. Also add an additional
40896 boolean 'activate' parameter; use TRUE for existing behaviour, or FALSE
40897 to only find the grab and then return it.
40898
40899 This will be used in forthcoming touch patches to find the grabs, rather
40900 than open-coding same.
40901
40902 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40903 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40904
40905commit eb5aaf5eb819f1287ce40da79edbe07bafb2b5c0
40906Author: Daniel Stone <daniel@fooishbar.org>
40907Date: Tue Dec 28 12:15:28 2010 +0000
40908
40909 Input: Export GetEventMask
40910
40911 Make it non-static, add to headers.
40912
40913 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40914 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40915 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40916
40917commit 30c8c7a86310139e79c9698caee21b3bd51936bd
40918Author: Daniel Stone <daniel@fooishbar.org>
40919Date: Tue Dec 28 12:13:29 2010 +0000
40920
40921 Input: Handle grabs with no Xi 1.x equivalent
40922
40923 Don't try to search for an Xi 1.x grab in CheckPassiveGrabsOnWindow for
40924 events with no Xi 1.x equivalent.
40925
40926 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40927 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40928 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40929
40930commit 690476250f58616d5156317e01ba547497f4331a
40931Author: Daniel Stone <daniel@fooishbar.org>
40932Date: Mon Sep 20 15:03:05 2010 +1000
40933
40934 Input: Pass sprite instead of device to XYToWindow, make non-static
40935
40936 XYToWindow calculates the position of the cursor and updates the sprite
40937 trace, but does nothing else with the device. Pass a SpritePtr instead
40938 so we can update an alternate focus instead of hardcoding the device's
40939 sprite. Also make this function non-static, so we can use it elsewhere.
40940
40941 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40942 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40943 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40944
40945commit 07a892cd8276f8092de95f1909f2ebec6f5fdf08
40946Author: Daniel Stone <daniel@fooishbar.org>
40947Date: Mon Sep 20 15:03:04 2010 +1000
40948
40949 Input: Pass sprite instead of device to FixUpEventFromWindow
40950
40951 Since FixUpEventFromWindow only uses the sprite trace to determine the
40952 window stack, pass in a sprite instead of hardcoding the device sprite,
40953 so we can deliver to windows other than the one currently containing the
40954 sprite.
40955
40956 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40957 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40958 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40959
40960commit 10d6c3ca103ea155488cd2f29052334b5691e680
40961Author: Daniel Stone <daniel@fooishbar.org>
40962Date: Mon Dec 27 13:06:03 2010 +0000
40963
40964 Input: Make RootWindow() take a sprite instead of device
40965
40966 GetCurrentRootWindow already works for the device case, although not as
40967 an lvalue.
40968
40969 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40970 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40971 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40972
40973commit 23c37ce1fe42e3da0609a566478e947843b05023
40974Author: Daniel Stone <daniel@fooishbar.org>
40975Date: Mon Dec 27 12:48:42 2010 +0000
40976
40977 Input: Export SpriteRec for use in public API
40978
40979 Change its definition to be more in line with other structs, so we can
40980 use it from input.h.
40981
40982 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40983 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40984 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40985
40986commit 06a103ad1d6a0ef6146c97a8c81dc7de1c0a0083
40987Author: Daniel Stone <daniel@fooishbar.org>
40988Date: Tue Dec 28 16:37:48 2010 +0000
40989
40990 Test: Input: Test up to supported server, not protocol, events
40991
40992 When building against a newer inputproto, the server may support fewer
40993 than XI_LASTEVENT events. We already have XI2LASTEVENT for the highest
40994 event number supported by the server, so use that instead.
40995
40996 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
40997 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
40998 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
40999
41000commit a083efe8715e8b29c9bd7f4e7bb429a94620bfc4
41001Author: Daniel Stone <daniel@fooishbar.org>
41002Date: Tue Dec 28 16:06:29 2010 +0000
41003
41004 Test: Input: Check flags on DeviceEvent
41005
41006 Add initial validation of acceptable flags for XI2 device events, and
41007 make sure they're swapped.
41008
41009 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41010 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41011 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
41012
41013commit 2592effef5f171af3f01a2b5130d9747403140f6
41014Author: Daniel Stone <daniel@fooishbar.org>
41015Date: Tue Dec 28 13:42:06 2010 +0000
41016
41017 Test: Input: Add helper function for failing EventToCore
41018
41019 We have quite a few tests which involve checking that EventToCore fails
41020 for specific events, so refactor them into a separate function.
41021
41022 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41023 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41024 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
41025
41026commit e1aed88be92c27d76bff1743f35f9915dea9264f
41027Author: Daniel Stone <daniel@fooishbar.org>
41028Date: Tue Dec 28 17:00:29 2010 +0000
41029
41030 Input: Swap flags in DeviceEvents
41031
41032 Swap flags for different-endian clients when delivering XI2
41033 DeviceEvents.
41034
41035 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41036 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41037 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
41038
41039commit 6358a60065eef167d4e5f4afd981ff26deeba80d
41040Author: Michel Dänzer <daenzer@vmware.com>
41041Date: Thu Jan 6 09:55:41 2011 +0100
41042
41043 EXA: Pad size of system memory copy for 1x1 pixmaps (bug #32803).
41044
41045 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=32803 .
41046
41047 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
41048 Reviewed-by: Dave Airlie <airlied@redhat.com>
41049 Signed-off-by: Keith Packard <keithp@keithp.com>
41050
41051commit f3480286aeb3009623d8d4b0202eadda0049552d
41052Author: Ville Syrjälä <ville.syrjala@nokia.com>
41053Date: Wed Jan 5 20:41:09 2011 +0200
41054
41055 composite: Support updating an arbitrary subtree
41056
41057 Rename compUpdateWindow to compPaintWindowToParent and split the child
41058 walk to compPaintChildrenToWindow. Calling compPaintChildrenToWindow
41059 allows an arbitrary subtree to be updated, instead of having to update
41060 all the windows. This will be used to make sure all the descendants are
41061 copied to the parent when the parent window contents need to be accessed
41062 in IncludeInferios sub-window mode.
41063
41064 WindowRec has a new member 'damagedDescendants' that is used to keep
41065 track of which subtrees need updating. When a window is damaged,
41066 'damagedDescendants' will be set for all the ancestors, and when a
41067 subtree is updated, the tree walk can be stopped early if no damaged
41068 descendants are present.
41069
41070 CompScreenRec no longer needs the 'damaged' member since the root
41071 window's 'damagedDescendants' provides the same information.
41072
41073 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41074 Signed-off-by: Keith Packard <keithp@keithp.com>
41075
41076commit b89e6dbdfbb50e3b5bc7fcb7eccc397c467c92f8
41077Author: Ville Syrjälä <ville.syrjala@nokia.com>
41078Date: Wed Jan 5 20:41:08 2011 +0200
41079
41080 composite: Add SourceValidate wrapper
41081
41082 When SourceValidate is performed on a window with IncludeInferiors
41083 sub-window mode, force an immediate update of all the automatically
41084 redirected windows, so that the current window contents will be up
41085 to date.
41086
41087 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41088 Reviewed-by: Keith Packard <keithp@keithp.com>
41089 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41090 Signed-off-by: Keith Packard <keithp@keithp.com>
41091
41092commit 84154954db54696d4661eb8d0a6014cdbff3c91f
41093Author: Ville Syrjälä <ville.syrjala@nokia.com>
41094Date: Wed Jan 5 20:41:07 2011 +0200
41095
41096 composite: Add GetImage wrapper
41097
41098 When GetImage is performed on a window, force an immediate update of
41099 all the automatically redirected windows, so that the current window
41100 contents will be up to date.
41101
41102 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41103 Reviewed-by: Keith Packard <keithp@keithp.com>
41104 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41105 Signed-off-by: Keith Packard <keithp@keithp.com>
41106
41107commit a5dc3531e14589ac473cea482944d2d67517aabd
41108Author: Ville Syrjälä <ville.syrjala@nokia.com>
41109Date: Wed Jan 5 20:41:06 2011 +0200
41110
41111 Revert "composite: Convert compWindowUpdate to use TraverseTree"
41112
41113 TraverseTree visits the parent before the children. When performing
41114 the automatic redirection updates, the children must be visited before
41115 the parent.
41116
41117 If there are automatically redirected windows on multiple levels of the
41118 tree, updating the parents before the children would cause the parent
41119 updates to use stale data for areas covered by the children. Also
41120 updating the damaged children would re-damage the parent, which would
41121 cause additional walks over the tree.
41122
41123 In the worst case with an unbroken chain of automatically redirected
41124 subwindows, all of which are damaged, only the leaf window would be
41125 properly updated on the first round. Then it's parent would be properly
41126 updated on the second round, and so on. And on every round all of the
41127 ancestor windows would be updated as well, but with stale data.
41128 So with N damaged windows you would end up with (N^2+N)/2 updates,
41129 instead of the expected N.
41130
41131 This reverts commit 648c8871c92727d7b6b16859f27f12266a06a16e.
41132
41133 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41134 Reviewed-by: Keith Packard <keithp@keithp.com>
41135 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41136 Signed-off-by: Keith Packard <keithp@keithp.com>
41137
41138commit 0998574699502e6ab14fd8899c2e42961d4df7d0
41139Author: Ville Syrjälä <ville.syrjala@nokia.com>
41140Date: Wed Jan 5 20:41:05 2011 +0200
41141
41142 Call SourceValidate even if src == dst
41143
41144 The extra SourceValidate calls from damageCopyArea and damageCopyPlane
41145 can be removed.
41146
41147 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41148 Reviewed-by: Keith Packard <keithp@keithp.com>
41149 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41150 Signed-off-by: Keith Packard <keithp@keithp.com>
41151
41152commit e41e907b3c19908f5316346fa587ced3115478cd
41153Author: Ville Syrjälä <ville.syrjala@nokia.com>
41154Date: Wed Jan 5 20:41:04 2011 +0200
41155
41156 Add subWindowMode parameter to SourceValidate
41157
41158 Pass the subWindowMode from the GC/source Picture to SourceValidate.
41159
41160 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41161 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41162 Reviewed-by: Keith Packard <keithp@keithp.com>
41163 Signed-off-by: Keith Packard <keithp@keithp.com>
41164
41165commit 0dede200c9ac7adbe8b8c16efacc3edc1f183cd9
41166Merge: 7714357 2e78145
41167Author: Keith Packard <keithp@keithp.com>
41168Date: Wed Jan 5 08:51:46 2011 -0800
41169
41170 Merge remote branch 'vsyrjala/misc_fixes'
41171
41172commit 7714357f506782973d25e270f85140b42507ed35
41173Merge: 02e18c9 b01dd9d
41174Author: Keith Packard <keithp@keithp.com>
41175Date: Wed Jan 5 08:43:18 2011 -0800
41176
41177 Merge remote branch 'whot/for-keith'
41178
41179commit 02e18c9fb58c33d340af4573551fb9c7c59e0f43
41180Author: James Jones <jajones@nvidia.com>
41181Date: Mon Dec 20 11:05:57 2010 -0800
41182
41183 X Sync Cleanups
41184
41185 Various cleanups identified during review of the
41186 X Sync Fence Object patches.
41187
41188 -Correctly handle failure of AddResource()
41189
41190 -Don't assert when data structures are corrupt. Instead,
41191 use a new helper function to check for counter sync
41192 objects when they're expected, and warn if the type is
41193 wrong.
41194
41195 -Use the default switch label rather than reimplementing
41196 it.
41197
41198 -Re-introduce cast of result of dixAllocateObjectWithPrivate()
41199 to kill an incompatible pointer type warning.
41200
41201 -Remove comments claiming protocol updates are needed. One
41202 wasn't true and the other was addressed with a xextproto
41203 change.
41204
41205 -Return BadFence, not BadCounter from XSyncAwaitFence()
41206
41207 Signed-off-by: James Jones <jajones@nvidia.com>
41208 Reviewed-by: Keith Packard <keithp@keithp.com>
41209 Signed-off-by: Keith Packard <keithp@keithp.com>
41210
41211commit 86ca434a1ab766c1519ee7146b48dbfead843524
41212Author: Zhao Yakui <yakui.zhao@intel.com>
41213Date: Tue Jan 4 14:08:16 2011 -0500
41214
41215 edid: Fix incorrect timings for VIC61
41216
41217 Reviewed-by: Adam Jackson <ajax@redhat.com>
41218 Signed-off-by: Anssi Hannula <anssi.hannula@iki.fi>
41219 Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
41220
41221commit bb18f277156c08be028a6e12d8987fb1593e9168
41222Author: Adam Jackson <ajax@redhat.com>
41223Date: Fri Dec 10 14:24:02 2010 -0500
41224
41225 x86emu: Fix more mis-decoding of the data prefix
41226
41227 cc2c73ddcb4370a7c3ad439cda4da825156c26c9's three-cent titanium tax
41228 doesn't go too far enough. Fix the rest of the call and jmp
41229 instructions to handle the data prefix correctly.
41230
41231 Reference: Intel 64 and IA-32 Architectures Software Developer's Manual
41232 Volume 2A: Instruction Set Reference, A-M
41233
41234 http://www.intel.com/Assets/PDF/manual/253666.pdf
41235
41236 Reviewed-by: Julien Cristau <jcristau@debian.org>
41237 Signed-off-by: Adam Jackson <ajax@redhat.com>
41238
41239commit 261d0d16af797bb52d4c778e220296d7f2b28e14
41240Author: Alan Coopersmith <alan.coopersmith@oracle.com>
41241Date: Thu Dec 30 16:08:48 2010 -0800
41242
41243 Fix asprintf typo in commit d2c42b10
41244
41245 Reported-by: Jerome Carretero <cJ@zougloub.eu>
41246 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
41247 Reviewed-by: Cyril Brulebois <kibi@debian.org>
41248 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41249
41250commit c6afe64bad7a27ff8828de499d56776c113df60f
41251Author: Daniel Stone <daniel@fooishbar.org>
41252Date: Fri Dec 31 03:15:27 2010 +0000
41253
41254 XFree86: Linux: AGP: Fix void-pointer-arithmetic warning
41255
41256 Cast it to a char *, mimicking the return immediately below it.
41257
41258 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41259 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
41260
41261commit 2d67ada3c4079a11c52024a9c3d4138becca5171
41262Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41263Date: Thu Dec 30 19:19:43 2010 +0200
41264
41265 os: always check if client is local when connection is accepted
41266
41267 LocalClient is used for all DRI2 requests that makes it frequently
41268 called function. Querying if connection is local or not takes 10-15us
41269 (on ARM) depending on malloc speed.
41270
41271 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41272 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41273
41274commit 617b7d22115ccaaaa7ec69c99885054d33a3bc37
41275Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41276Date: Thu Dec 30 19:19:42 2010 +0200
41277
41278 os: Fix a memory leak
41279
41280 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41281 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
41282 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41283
41284commit 3e1455505addc7e52178fa04399aef2a8522c921
41285Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41286Date: Thu Dec 30 19:19:41 2010 +0200
41287
41288 udev: Fix removing of the wake up handler
41289
41290 RemoveBlockAndWakeupHandlers requires caller to pass same block data
41291 parameter as for RegisterBlockAndWakeupHandlers.
41292
41293 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41294 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41295 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41296
41297commit 13ac3deedb532b39f41dcd45d70f9eb4a85c1f58
41298Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41299Date: Thu Dec 30 19:19:40 2010 +0200
41300
41301 mi: Only register sprite block handler when needed
41302
41303 miSpriteBlockHandler takes about 10us in arm each time BlockHandlers are
41304 called. To eliminate that overhead from xserver side only register the
41305 BlockHandler when there actually is any cursor down.
41306
41307 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41308 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41309
41310commit 4f6e3b0f378d7306dbd8c00ef9a7df81e24e5769
41311Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41312Date: Thu Dec 30 19:19:39 2010 +0200
41313
41314 mi: Register sprite damage handler only when required
41315
41316 There is no need to have damage handler registered when there isn't any
41317 pointer in the screen. This avoids some overhead from damage handling
41318 which takes tens of microseconds on arm.
41319
41320 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41321 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41322
41323commit bf48082f30818b96bc623834be3022600371d4fc
41324Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41325Date: Thu Dec 30 19:19:38 2010 +0200
41326
41327 mi: Fix the debug message
41328
41329 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41330 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41331 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41332
41333commit 8fb43b8bf9fcbe015d4e98c7e09889184d136a1e
41334Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41335Date: Thu Dec 30 19:19:37 2010 +0200
41336
41337 mi: Fix wrapping for sprite screen
41338
41339 SCREEN_EPILOGUE should read the current function pointer from screen
41340 after calling wrapped function in case the pointer changes.
41341
41342 Passing pPriv to macros instead of asking dix each time makes sense when
41343 both macros need same pointer.
41344
41345 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41346 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41347 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41348
41349commit 4be2a6d1b33fac7ceb50ad1de249a1acd5a46ab0
41350Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41351Date: Thu Dec 30 19:19:36 2010 +0200
41352
41353 mi: Add GetSpriteScreen macro to hide dixLookupPrivate
41354
41355 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41356 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41357
41358commit ed8db09b4bff5c4a8d8056808368650e1876547b
41359Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41360Date: Thu Dec 30 19:19:35 2010 +0200
41361
41362 composite: Remove unnecessary variable.
41363
41364 We know that there is damage if the dynamic block handler is registered.
41365
41366 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41367 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41368
41369commit c038b8b28e5c436cd31bea7a3bef2c8c9d818b58
41370Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41371Date: Thu Dec 30 19:19:34 2010 +0200
41372
41373 composite: Only register the block handler when it is required
41374
41375 Even calling block handler that doesn't do much is costly in arm. It
41376 takes a few microseconds each time which adds up to relative high CPU
41377 time because it is done 500+ times per second.
41378
41379 Simple optimization is to register the block handler only when it is
41380 required.
41381
41382 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41383 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41384
41385commit 6d0e9e5d6e1b847961ab52a11aae96981a1cf1c0
41386Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41387Date: Thu Dec 30 19:19:33 2010 +0200
41388
41389 render: Enable animated cursor block handler only when needed
41390
41391 Calling BlockHandlers takes some time for each iteration in main loop
41392 which adds up quickly over multiple request. To reduce the round-trip
41393 costs to xserver BlockHandlers should be registered only when required.
41394
41395 AnimCurScreenBlockHandler is the first victim for this optimization.
41396
41397 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41398 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41399
41400commit aa8cea953dc66bcf4cb4d08f2681f9e6cb1bc8c5
41401Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41402Date: Thu Dec 30 19:19:32 2010 +0200
41403
41404 render: Don't filter 0 return from GetTimeInMillis
41405
41406 In animate cursor block handler code assumes GetTimeInMillis returns
41407 always nonzero value. This isn't true when time wraps around.
41408
41409 To prevent any problems in case GetTimeInMillis would return zero use
41410 activeDevice variable to track if we have received time.
41411
41412 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
41413 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41414
41415commit 91beeee05f88eed10ab0fd97dc625e96cb7763ba
41416Author: Daniel Stone <daniel@fooishbar.org>
41417Date: Tue Oct 13 16:27:30 2009 +1100
41418
41419 DIX: Reset window properties when freeing them
41420
41421 This enables us to reliably inspect properties when destroying windows.
41422
41423 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41424 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
41425
41426commit 36ebdd361616eedbe4919deae2a4e6f6606ae6b4
41427Author: Adam Jackson <ajax@redhat.com>
41428Date: Tue Oct 19 11:59:24 2010 -0400
41429
41430 linux: warning fix
41431
41432 lnx_agp.c: In function ‘xf86DeallocateGARTMemory’:
41433 lnx_agp.c:267: warning: cast to pointer from integer of different size
41434
41435 Signed-off-by: Adam Jackson <ajax@redhat.com>
41436 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41437
41438commit 63ccaec2c39f4b5742383472c951ee2cd35c9e14
41439Author: Adam Jackson <ajax@redhat.com>
41440Date: Tue Oct 19 11:59:23 2010 -0400
41441
41442 glx: warning fixes
41443
41444 render2.c: In function ‘__glXDisp_Map2d’:
41445 render2.c:127: warning: ‘u1’ may be used uninitialized in this function
41446 render2.c: In function ‘__glXDisp_Map1d’:
41447 render2.c:90: warning: ‘u1’ may be used uninitialized in this function
41448
41449 Remove unnecessary test, and change memcpy to memmove as all users were
41450 doing overlapping copies.
41451
41452 Signed-off-by: Adam Jackson <ajax@redhat.com>
41453 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41454
41455commit d057e265c862f90e9b6fddff89f40cb82b2a59b1
41456Author: Adam Jackson <ajax@redhat.com>
41457Date: Tue Oct 19 11:59:22 2010 -0400
41458
41459 xcmisc: warning fix
41460
41461 xcmisc.c:202: warning: no previous prototype for ‘XCMiscExtensionInit’
41462
41463 Signed-off-by: Adam Jackson <ajax@redhat.com>
41464 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41465
41466commit 67e0bcba4307de5a330e027f4504c9aed4987e5a
41467Author: Adam Jackson <ajax@redhat.com>
41468Date: Tue Oct 19 11:59:21 2010 -0400
41469
41470 dri1: warning fix
41471
41472 dri.c: In function ‘DRIScreenInit’:
41473 dri.c:434: warning: cast from pointer to integer of different size
41474
41475 Signed-off-by: Adam Jackson <ajax@redhat.com>
41476 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41477
41478commit a298d044f965e5ba91f178c6b599c1df26a958ba
41479Author: Adam Jackson <ajax@redhat.com>
41480Date: Tue Oct 19 11:59:20 2010 -0400
41481
41482 loader: warning fix
41483
41484 loadmod.c: In function ‘FreeSubdirs’:
41485 loadmod.c:377: warning: passing argument 1 of ‘free’ discards qualifiers
41486 from pointer target type
41487 /usr/include/stdlib.h:488: note: expected ‘void *’ but argument is of
41488 type ‘const char *’
41489
41490 Signed-off-by: Adam Jackson <ajax@redhat.com>
41491 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41492
41493commit ee0b1b564413327c73bb8cdbabb76667ca4a744c
41494Author: Adam Jackson <ajax@redhat.com>
41495Date: Tue Oct 19 11:59:18 2010 -0400
41496
41497 config: warning fixes
41498
41499 xf86AutoConfig.c: In function ‘FreeList’:
41500 xf86AutoConfig.c:123: warning: passing argument 1 of ‘free’ discards
41501 qualifiers from pointer target type
41502 /usr/include/stdlib.h:488: note: expected ‘void *’ but argument is of
41503 type ‘const char *’
41504
41505 Signed-off-by: Adam Jackson <ajax@redhat.com>
41506 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41507
41508commit 32c85ad4b8fe27fed0c494f69c39e3902ce57b1b
41509Author: Daniel Stone <daniel@fooishbar.org>
41510Date: Tue Dec 28 12:40:31 2010 +0000
41511
41512 GLX: DRI2: Fix mismatched-types warning
41513
41514 All the DRI extension types have a base extension type as their first
41515 member to avoid exactly these types of warning.
41516
41517 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41518 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
41519 Reviewed-by: Cyril Brulebois <kibi@debian.org>
41520
41521commit 22796cfa4805cc9551e1b3fa1d3e2e1bfae5bad1
41522Author: Daniel Stone <daniel@fooishbar.org>
41523Date: Tue Dec 28 12:29:53 2010 +0000
41524
41525 udev: Add strdups to kill const warnings
41526
41527 InputAttributes wants non-const members, and while it appears safe to
41528 cast it, just leave it be for the moment.
41529
41530 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41531 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
41532
41533commit 03f2eb1e156796afb70118d7f7f60ac61beed026
41534Author: Daniel Stone <daniel@fooishbar.org>
41535Date: Fri Dec 17 17:13:34 2010 +0000
41536
41537 Input: Set client error value for invalid mask bits
41538
41539 When we send BadValue back to the client for having invalid mask bits,
41540 at least tell them what the (first) invalid bit was.
41541
41542 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41543 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41544
41545commit 47818287983b04625661b08ae362e0966cddf3e1
41546Author: Chase Douglas <chase.douglas@canonical.com>
41547Date: Fri Dec 17 17:13:30 2010 +0000
41548
41549 Make EventIsDeliverable non-static
41550
41551 Will be used outside dix/events.c in proceeding XI 2.1 MT changes.
41552
41553 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
41554 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41555 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41556
41557commit f7d8ade3c5ac44faf7f3ea0d846e35f75b31c9ec
41558Author: Daniel Stone <daniel@fooishbar.org>
41559Date: Fri Dec 17 17:13:26 2010 +0000
41560
41561 Resources: Move rClient to resource.h
41562
41563 The definition of rClient was duplicated across three source files, so
41564 move it to resource.h.
41565
41566 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41567 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41568
41569commit 05e437c2ce230b588362f012c06ead5c6d89c006
41570Author: Chase Douglas <chase.douglas@canonical.com>
41571Date: Fri Dec 17 17:13:27 2010 +0000
41572
41573 Export all valuator_mask_* functions
41574
41575 Input drivers may use valuator masks for internal state. Having all the
41576 valuator_mask_* functions available will help.
41577
41578 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
41579 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41580
41581 Bump ABI_XINPUT_VERSION minor.
41582
41583 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41584
41585commit 2e781457d43ec4bf0d633257ac6852cde3b00541
41586Author: Ville Syrjälä <ville.syrjala@nokia.com>
41587Date: Thu Dec 16 18:22:13 2010 +0200
41588
41589 xfree86/modes: Take rotation into account when checking mode size
41590
41591 Assume that a mode can be used in either landscape or portrait
41592 orientation. I suppose the correct thing to do would be to
41593 collect all the supported rotations from the CRTCs that can be used
41594 with a specific output, but that information doesn't seem to be
41595 readily available when these checks are done. So just assume that
41596 either orientation is fine.
41597
41598 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41599 Reviewed-by: Adam Jackson <ajax@redhat.com>
41600
41601commit 0ce25fd7904c792924c3e0ee6fc21a5f1bec1a68
41602Author: Ville Syrjälä <ville.syrjala@nokia.com>
41603Date: Fri Dec 3 17:42:16 2010 +0200
41604
41605 dri2: Don't page flip when the window size doesn't match the pixmap size
41606
41607 If the drawable size doesn't match the pixmap size page flipping should
41608 not be allowed.
41609
41610 If the window is larger than the pixmap, page flipping might need to
41611 reposition the CRTC somewhere in the middle of the pixmap. I didn't
41612 spot any code that would handle that at least in the intel driver.
41613
41614 Also the root pixmap could then move to some negative screen
41615 coordinates. Not sure if all bits of code could handle that. Perhaps
41616 when composite is enabled screen_x/y would make it work, but without
41617 composite there's no way that it would work AFAICS.
41618
41619 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
41620 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
41621
41622commit efcb63d0ce43f96d0ac02b6f4a480dfd2374fc84
41623Author: Daniel Stone <daniel@fooishbar.org>
41624Date: Thu Dec 23 13:33:00 2010 +0000
41625
41626 Render: Fix 'comparing between distinct pointer types' warning
41627
41628 Add the appropriate casts so that gcc shuts up, even if it doesn't
41629 matter.
41630
41631 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41632 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41633
41634commit 469d5bf8b75038631c27edbb0f9cdf7d737fa233
41635Author: Daniel Stone <daniel@fooishbar.org>
41636Date: Fri Dec 17 16:48:45 2010 +0000
41637
41638 mi: Sync: Don't free managed screen private
41639
41640 misync allocates space for its screen private with
41641 dixRegisterPrivateKey, which means it doesn't have to free it at
41642 CloseScreen time; doing so will, in fact, result in a crash.
41643
41644 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41645 Reviewed-by: James Jones <jajones@nvidia.com>
41646 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41647
41648commit 44adb31bfece29260a9bbd9075c9212ebf00d24d
41649Author: Daniel Stone <daniel@fooishbar.org>
41650Date: Fri Aug 27 20:36:37 2010 +1000
41651
41652 GetTimeInMillis: Use CLOCK_MONOTONIC_COARSE where available
41653
41654 On some systems, using CLOCK_MONOTONIC forces a readback of HPET or some
41655 similarly expensive timer. CLOCK_MONOTONIC_COARSE can alleviate this,
41656 at the cost of negligibly-reduced resolution, so prefer that where we
41657 can.
41658
41659 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41660 Reviewed-by: Julien Cristau <jcristau@debian.org>
41661 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41662
41663commit 296561506a91742cc150a0fb6fc0df5dbe98c780
41664Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
41665Date: Wed Dec 22 16:57:17 2010 +0200
41666
41667 Xext: Use general OS functions to determine client command string in SELinux.
41668
41669 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
41670 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41671
41672commit 1e933665bef26c74196bb7c59910e6a78bcacf0e
41673Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
41674Date: Wed Dec 22 16:51:09 2010 +0200
41675
41676 os: Add facilities for client ID tracking.
41677
41678 An interface is provided for figuring out the PID and process name of
41679 a client. Make some existing functionality from SELinux and IA
41680 extensions available for general use.
41681
41682 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
41683 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41684
41685commit 671b2a1823a1c90d0b6254e2e2af8865151fff8c
41686Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
41687Date: Wed Dec 22 16:38:54 2010 +0200
41688
41689 config: Fix linking order of Xnest libraries.
41690
41691 MAIN depends on DIX and not vice versa.
41692
41693 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
41694
41695commit 2b364bf970b2ce6829af656990c33afd0d365f3c
41696Author: Rami Ylimäki <rami.ylimaki@vincit.fi>
41697Date: Tue Jun 15 14:44:38 2010 +0200
41698
41699 Revert "os: Prevent backtrace from being stopped in noreturn functions."
41700
41701 This reverts commit 579715f830fbbca9e1ecb17dc18176132f5969e7.
41702
41703 The patch is not needed anymore. I haven't encountered backtrace
41704 problems with GCC 4.3.3. Even if the problems still persisted, this
41705 commit should be removed and instead the definition of _X_NORETURN
41706 should be modified to be empty if GCC/ARM is used. However, currently
41707 it seems that ARM backtraces are OK even if _X_NORETURN is used and
41708 -mapcs-frame is not defined in CFLAGS.
41709
41710 Signed-off-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
41711 Reviewed-by: Jamey Sharp <jamey@minilop.net>
41712
41713commit b01dd9d33651999b0390bf8ed412b84f6929a13e
41714Author: Daniel Stone <daniel@fooishbar.org>
41715Date: Fri Dec 17 17:13:34 2010 +0000
41716
41717 Input: Set client error value for invalid mask bits
41718
41719 When we send BadValue back to the client for having invalid mask bits,
41720 at least tell them what the (first) invalid bit was.
41721
41722 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41723 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41724
41725commit 495e0142a77152c4cde5c88cdba4103ddc94df2c
41726Author: Chase Douglas <chase.douglas@canonical.com>
41727Date: Fri Dec 17 17:13:30 2010 +0000
41728
41729 Make EventIsDeliverable non-static
41730
41731 Will be used outside dix/events.c in proceeding XI 2.1 MT changes.
41732
41733 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
41734 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41735 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41736
41737commit 4fbe67beb56e5600c1df789c08725c45b9bef0d4
41738Author: Daniel Stone <daniel@fooishbar.org>
41739Date: Fri Dec 17 17:13:26 2010 +0000
41740
41741 Resources: Move rClient to resource.h
41742
41743 The definition of rClient was duplicated across three source files, so
41744 move it to resource.h.
41745
41746 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
41747 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41748
41749commit 871c1e17e527857cd86b5e517d9a18e5041f00fd
41750Author: Chase Douglas <chase.douglas@canonical.com>
41751Date: Fri Dec 17 17:13:27 2010 +0000
41752
41753 Export all valuator_mask_* functions
41754
41755 Input drivers may use valuator masks for internal state. Having all the
41756 valuator_mask_* functions available will help.
41757
41758 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
41759 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41760
41761 Bump ABI_XINPUT_VERSION minor.
41762
41763 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
41764
41765commit e06fa804009798ea95efa8babaabb0228dfdfe65
41766Author: Michel Dänzer <daenzer@vmware.com>
41767Date: Wed Dec 22 11:45:36 2010 +0100
41768
41769 EXA: Fix crash with fill using 1x1 tile of depth < 8 (bug #24703).
41770
41771 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24703 .
41772
41773 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
41774 Reviewed-by: Dave Airlie <airlied@redhat.com>
41775 Signed-off-by: Keith Packard <keithp@keithp.com>
41776
41777commit 56d9c736f0c450bffd43300107689eb6dd1ea70b
41778Author: Adam Jackson <ajax@redhat.com>
41779Date: Thu Dec 16 14:26:10 2010 -0500
41780
41781 xfree86: Convert some xf86BlockSIGIO to OsBlockSignals
41782
41783 The DDC1 and int10 code are blocking SIGIO to get some assurance that
41784 their usleep() calls take as long as they expect. That's a good start
41785 but you really want to be blocking more than just SIGIO, SIGALRM too at
41786 minimum.
41787
41788 At this point, except for SIGIO handler setup itself, BlockSIGIO really
41789 means "block input events".
41790
41791 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41792 Signed-off-by: Adam Jackson <ajax@redhat.com>
41793
41794commit ce13a1dbbfbeee3a3d51d0385817df50944cfdee
41795Author: Adam Jackson <ajax@redhat.com>
41796Date: Thu Dec 16 14:03:13 2010 -0500
41797
41798 xfree86: Remove xf86EnterServerState
41799
41800 Back when we had RAC this was a vaguely meaningful thing. Since then
41801 it's been a glorified (and confusing) wrapper around xf86BlockSIGIO.
41802
41803 Note that the APM and VT switch code are unusual relative to other code
41804 that cares about SIGIO state. Most callers push a SIGIO disable to
41805 create a critical section for the duration of the caller's stack frame,
41806 but those two effectively disable SIGIO after their return and re-enable
41807 on their next entry.
41808
41809 Reviewed-by: Tiago Vignatti <tigo.vignatti@nokia.com>
41810 Signed-off-by: Adam Jackson <ajax@redhat.com>
41811
41812commit a04e74cb480f98bc3dd4def2da6f6752640d9e38
41813Author: Adam Jackson <ajax@redhat.com>
41814Date: Wed Dec 15 15:16:33 2010 -0500
41815
41816 xfree86: Symmetrize xf86AccessEnter and xf86AccessLeave
41817
41818 Enter was changing server operating state, Leave wasn't. Which was
41819 wholly redundant, since all callers of Enter would immediately change
41820 the operating state to exactly what Enter had just done.
41821
41822 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41823 Signed-off-by: Adam Jackson <ajax@redhat.com>
41824
41825commit 7a9062f2f029b4f911ba56f291375fbf5a98ca73
41826Author: Adam Jackson <ajax@redhat.com>
41827Date: Tue Dec 14 15:44:16 2010 -0500
41828
41829 fonts: Fix typo in async ListFonts logic
41830
41831 This was introduced in 3ab6cd31cbdf8095b2948034fce5fb645422d8da. Mea
41832 culpa. This logic is still incorrect [1], but at least it's less
41833 incorrect.
41834
41835 [1] - https://bugzilla.redhat.com/658587
41836
41837 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41838 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41839 Signed-off-by: Adam Jackson <ajax@redhat.com>
41840
41841commit f28e48834e40c7901c2efc72f962c9724e74a531
41842Author: Adam Jackson <ajax@redhat.com>
41843Date: Thu Dec 16 13:13:54 2010 -0500
41844
41845 Remove SCO support
41846
41847 This has never been buildable in any modular server release.
41848
41849 Reviewed-by: Julien Cristau <jcristau@debian.org>
41850 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
41851 Signed-off-by: Adam Jackson <ajax@redhat.com>
41852
41853commit 040d54fb88f3970e19d0fbf911d6447f592496bd
41854Author: Adam Jackson <ajax@redhat.com>
41855Date: Wed Dec 15 14:59:13 2010 -0500
41856
41857 xfree86: Remove unused xf86inSuspend
41858
41859 What is this, I don't even.
41860
41861 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
41862 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41863 Signed-off-by: Adam Jackson <ajax@redhat.com>
41864
41865commit e7dc253452a1ba64718a08fdc070405b494f53cd
41866Merge: d110791 d5b0d58
41867Author: Keith Packard <keithp@keithp.com>
41868Date: Mon Dec 20 00:48:45 2010 -0800
41869
41870 Merge remote branch 'alanc/master'
41871
41872commit d1107918d4626268803b54033a07405122278e7f
41873Author: Tiago Vignatti <tiago.vignatti@nokia.com>
41874Date: Fri Dec 17 16:09:35 2010 +0200
41875
41876 randr: check for virtual size limits before set crtc
41877
41878 Return a error if the screen is configured to an invalid size.
41879
41880 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
41881 Reviewed-by: Adam Jackson <ajax@redhat.com>
41882 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
41883 Signed-off-by: Keith Packard <keithp@keithp.com>
41884
41885commit 4b88c7be8de4149fe3d166bf115775f9e81a1373
41886Author: Jörn Horstmann <launchpad@planetxml.de>
41887Date: Mon Dec 6 11:24:02 2010 +1100
41888
41889 Add EDID quirk for HP Compaq nc8430.
41890
41891 Like some other LPL panels, this one reports the vertical size in cm rather
41892 than mm.
41893 Patch taken from Launchpad bug #380009 <https://launchpad.net/bugs/380009>
41894
41895 X.Org Bug 28414 <https://bugs.freedesktop.org/show_bug.cgi?id=28414>
41896
41897 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
41898 Reviewed-by: Adam Jackson <ajax@redhat.com>
41899 Signed-off-by: Keith Packard <keithp@keithp.com>
41900
41901commit a2e67a6412386782cb8b644b86a5744591397d45
41902Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
41903Date: Mon Dec 6 11:24:01 2010 +1100
41904
41905 IDLETIME: Fix edge-case in IdleTimeBlockHandler
41906
41907 Ensure that if we're called exactly on the threshold of a
41908 NegativeTransition trigger that we reshedule to pick up
41909 an idle time over the threshold.
41910
41911 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
41912 Reviewed-by: Adam Jackson <ajax@redhat.com>
41913 Signed-off-by: Keith Packard <keithp@keithp.com>
41914
41915commit b55bf248581dc66321b24b29f199f6dc8d02db1b
41916Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
41917Date: Mon Dec 6 11:24:00 2010 +1100
41918
41919 Xext: Fix edge case with {Positive, Negative}Transition triggers.
41920
41921 The {Positive,Negative}Transition triggers only fire when the counter
41922 goes from strictly {below,above} the threshold. If
41923 SyncComputeBracketValues gets called exactly at this threshold we may update
41924 the bracket values so that the counter is not updated past the threshold.
41925
41926 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
41927 Reviewed-by: Adam Jackson <ajax@redhat.com>
41928 Signed-off-by: Keith Packard <keithp@keithp.com>
41929
41930commit d5b0d58573ffb90ea70273cd013102168fd2a18a
41931Author: Alan Coopersmith <alan.coopersmith@oracle.com>
41932Date: Thu Dec 16 22:29:37 2010 -0800
41933
41934 Bug 32436 - hw/xfree86/parser/Makefile.am TOP_SRCDIR should be top_srcdir
41935
41936 https://bugs.freedesktop.org/show_bug.cgi?id=32436
41937 Fix typo introduced in 2416255f7e3fd9190a9 that breaks builds when
41938 configured --enable-install-libxf86config
41939
41940 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
41941 Reviewed-by: Julien Cristau <jcristau@debian.org>
41942 Tested-by: Simon Thum <simon.thum@gmx.de>
41943
41944commit 058b889fde47b4c32534f11ed651bf1749d1dbb2
41945Author: David Barksdale <david.barksdale@adcedosolutions.com>
41946Date: Mon Dec 13 18:35:20 2010 -0800
41947
41948 Bug 21827: faulty sdksyms.c generated by the awk script in sdksyms.sh
41949
41950 https://bugs.freedesktop.org/show_bug.cgi?id=21827
41951
41952 Tested-by: Alan Coopersmith <alan.coopersmith@oracle.com>
41953 Signed-off-by: David Barksdale <david.barksdale@adcedosolutions.com>
41954 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
41955
41956commit 1324b0ca9f8a7fdaf03b374c75eb3c9df407c2f1
41957Author: Colin Harrison <colin.harrison@virgin.net>
41958Date: Fri Dec 10 00:08:24 2010 +0000
41959
41960 Typo in xserver Xvasprintf()
41961
41962 I needed this patch in the wrapper around vsnprintf() in os/xprintf.c
41963 (MinGW for Windows build) to correct various crashes.
41964
41965 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
41966
41967commit f1542f1d716723cba7c323849086585635121893
41968Merge: 9716d31 8a8fdd7
41969Author: Keith Packard <keithp@keithp.com>
41970Date: Tue Dec 14 15:04:12 2010 -0800
41971
41972 Merge remote branch 'whot/for-keith'
41973
41974commit 9716d3124799c6db0d1c782aa72c72f972d5a158
41975Merge: 03ea0b7 f641e4b
41976Author: Keith Packard <keithp@keithp.com>
41977Date: Tue Dec 14 15:01:36 2010 -0800
41978
41979 Merge remote branch 'jeremyhu/master'
41980
41981commit 03ea0b7726487b96df73936518da73d023e51334
41982Author: Gaetan Nadon <memsize@videotron.ca>
41983Date: Tue Dec 7 16:31:55 2010 -0500
41984
41985 docs: replace BUILLDOCS with existing ENABLE_DEVEL_DOCS
41986
41987 Use the util-macros AM Conditionals to control generation of developers
41988 documents. This is used throughout xorg modules.
41989
41990 The doxygen generated docs are now also managed by --enable-devel-docs.
41991
41992 Remove --enable-builddocs as this was last use for BUILDDOCS
41993
41994 *** From the RELEASE NOTES ***
41995
41996 New configure options for documentation in modules
41997 --------------------------------------------------
41998 As many more modules now contain documentation to be converted from DocBook XML to text,
41999 HTML, PostScript, and/or PDF formats, new standard options have been added to the configure
42000 macros to control the build of these in the modules.
42001 --with-xmlto=yes|no
42002 Enables or disables use of the xmlto [https://fedorahosted.org/
42003 xmlto/] command to translate DocBook XML to other formats.
42004 All DocBook XML conversions require use of this command.
42005 --with-fop=yes|no
42006 Enables or disables use of the Apache fop [http://
42007 xmlgraphics.apache.org/fop/] command to translate DocBook
42008 XML to PostScript and PDF formats.
42009 --enable-docs=yes|no
42010 Enables or disables the build and installation of all
42011 documentation except traditional man pages or those covered
42012 by the --enable-devel-docs and --enable-specs options.
42013 --enable-devel-docs=yes|no
42014 Enables or disables the build and installation of documentation
42015 for developers of the X.Org software modules.
42016 --enable-specs=yes|no
42017 Enables or disables the build and installation of the formal
42018 specification documents for protocols and APIs.
42019
42020 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
42021 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
42022 Signed-off-by: Keith Packard <keithp@keithp.com>
42023
42024commit c78c71c3a872665e426a0a4b82f17007bd8c12cd
42025Author: Gaetan Nadon <memsize@videotron.ca>
42026Date: Tue Dec 7 16:31:54 2010 -0500
42027
42028 hw/dmx/doc: DIST_SUBDIRS not required when using AM_CONDITIONAL
42029
42030 The makefile can be simplyfied as Automake handle this automatically.
42031 All directories in SUBDIRS are visited for dist purposes.
42032
42033 www.gnu.org/software/automake/manual/automake.html#SUBDIRS-vs-DIST_005fSUBDIRS
42034
42035 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
42036 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
42037 Signed-off-by: Keith Packard <keithp@keithp.com>
42038
42039commit 8a8fdd762ad89c350854943311ec4aadc50245fa
42040Author: Peter Hutterer <peter.hutterer@who-t.net>
42041Date: Mon Dec 13 11:28:30 2010 +1000
42042
42043 xfree86: always report the input options before initialising the device.
42044
42045 After collecting the driver's default options, report the list of options
42046 set for the device before calling PreInit(). This helps with debugging those
42047 cases where options are not merged correctly.
42048
42049 xf86OptionListReport reports with verbosity 5, higher than the default
42050 verbosity so this won't generate logspam in the default case.
42051
42052 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42053 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
42054 Reviewed-by: Simon Thum <simon.thum@gmx.de>
42055
42056commit 9db9e964f6ca553dcbd3b7b037745d9581eaa065
42057Author: Peter Hutterer <peter.hutterer@who-t.net>
42058Date: Fri Dec 10 13:23:13 2010 +1000
42059
42060 xfree86: swap the order to-be-merged lists in xf86CollectInputOptions.
42061
42062 Current order causes the user-configured option list to be overwritten with
42063 the default list supplied by the driver. Swap around so we overwrite the
42064 driver's default values instead.
42065
42066 This only affected options supplied by the driver such as XkbLayout in the
42067 case of evdev.
42068
42069 Reported-by: Sebastian Glita <glseba@yahoo.com>
42070 Reported-by: Simon Thum <simon.thum@gmx.de>
42071 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42072 Reviewed-by: Simon Thum <simon.thum@gmx.de>
42073 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
42074 Tested-by: Sebastian Glita <glseba@yahoo.com>
42075
42076commit 0d440a1c6e219cd39dbddd2b7e813c6431aac6ea
42077Author: Peter Hutterer <peter.hutterer@who-t.net>
42078Date: Wed Dec 8 14:52:19 2010 +1000
42079
42080 dix: allow for button-only input devices (#21457)
42081
42082 Add a few checks for the existence of a valuator class on the device to
42083 avoid null-pointer dereferences for button events from devices without a
42084 valuator class.
42085
42086 X.Org Bug 21457 <http://bugs.freedesktop.org/show_bug.cgi?id=21457>
42087
42088 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42089 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
42090 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
42091
42092commit f641e4b34aa91ecda29e546b8b975e72ce037ed0
42093Author: Jeremy Huddleston <jeremyhu@apple.com>
42094Date: Wed Dec 8 22:58:12 2010 -0800
42095
42096 XQuartz GL: Add $(GL_CFLAGS) to CFLAGS
42097
42098 This fixes a build failure I found on tinderbox.
42099
42100 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
42101
42102commit 9cf055892dd413932e54b43cc2dfea70bafd525f
42103Author: Peter Hutterer <peter.hutterer@who-t.net>
42104Date: Wed Dec 8 14:43:51 2010 +1000
42105
42106 xfree86: don't set movement flags for non-valuator events.
42107
42108 If a device doesn't send valuators, don't try to move its position.
42109
42110 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42111 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
42112
42113commit aba8133c9c5a50753c388d76407868ac69f4134b
42114Author: Peter Hutterer <peter.hutterer@who-t.net>
42115Date: Wed Dec 8 14:02:17 2010 +1000
42116
42117 dix: clear up an overly convoluted if statement.
42118
42119 No functional changes, just improves readability. This statement had things
42120 added to/removed from it for a few server releases while the input event
42121 queue was revamped. What made sense once is now mainly confusing.
42122
42123 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42124 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
42125
42126commit 883039e07b0c0ef23c7fc4ba3455b41173c7a7fc
42127Author: Jeremy Huddleston <jeremyhu@apple.com>
42128Date: Wed Dec 8 13:27:41 2010 -0800
42129
42130 XQuartz: Just NSBeep() for XBell()
42131
42132 The CoreAudio path uses deprecated API and has reported crashes that
42133 aren't worth fixing (4e8bf12b13690afa6d9fee0e339d3819ef16fb3f fixed
42134 one and introduced another). NSBeep() does the job just fine.
42135
42136 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
42137
42138commit 2c70b650b342378898064cf27e2f95c4b2e53d24
42139Author: Jon TURNEY <jon.turney@dronecode.org.uk>
42140Date: Wed Dec 8 18:54:30 2010 +0000
42141
42142 Fix Xwin build after af0f9f91
42143
42144 Fix damage to XWIN_LIBS in commit af0f9f913398d34a885c3fb4e8d40c1a7e2b3ee9
42145 "Move some sync code to miext", which broke the XWin build
42146
42147 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
42148 Reviewed-by: Colin Harrison <colin.harrison at virgin.net>
42149 Signed-off-by: Keith Packard <keithp@keithp.com>
42150
42151commit 519d243f0c8e3c80226701f71d2cfa62e42dbff7
42152Merge: 9f9c732 446482e
42153Author: Keith Packard <keithp@keithp.com>
42154Date: Tue Dec 7 13:39:58 2010 -0800
42155
42156 Merge remote branch 'alanc/master'
42157
42158commit 446482efaa3d266266e2a143492a3ec0523622bd
42159Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42160Date: Sun Nov 28 09:41:17 2010 -0800
42161
42162 Convert cvt code to use XNFasprintf()
42163
42164 Requires linking xprintf.c into standalone cvt utility
42165
42166 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42167 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
42168 Reviewed-by: Julien Cristau <jcristau@debian.org>
42169
42170commit d2c42b102707db6c1ca9dccc4de9210a43650cb1
42171Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42172Date: Sat Nov 27 23:10:46 2010 -0800
42173
42174 Replace alloc+strcpy+strcat with asprintf() & XNFasprintf() calls
42175
42176 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42177 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
42178
42179commit 3a9bb93dd178084f4ff1abcea331ca5a62f88ce6
42180Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42181Date: Sat Nov 27 22:38:27 2010 -0800
42182
42183 Convert alloc+sprintf pairs into asprintf() & XNFasprintf() calls
42184
42185 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42186 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
42187
42188commit 2416255f7e3fd9190a9e01bda57c992932de4bd9
42189Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42190Date: Sat Nov 27 22:36:44 2010 -0800
42191
42192 Convert hw/xfree86/parser code to use asprintf() calls
42193
42194 Requires linking xprintf.c into libxf86config for those who build it.
42195
42196 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42197 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
42198
42199commit 03e8bfa1d122f7dea905d48c93cfd54afd991dfd
42200Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42201Date: Sat Nov 27 20:09:04 2010 -0800
42202
42203 Convert existing Xprintf style calls to asprintf style
42204
42205 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42206 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
42207 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
42208
42209commit c95c1d338fdb62dbe3dba934b97324fa778b7fce
42210Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42211Date: Sat Nov 27 18:43:12 2010 -0800
42212
42213 Add asprintf() implementation for platforms without it
42214
42215 Provides a portable implementation of this common allocating sprintf()
42216 API found in many, but not yet all, of the platforms we support.
42217 If the platform provides vasprintf() we simply wrap it, otherwise we
42218 implement it - either way callers can use it regardless of platform.
42219
42220 Since not all platforms guarantee to NULL out the return pointer on
42221 failure, we don't either, and require callers to check the return
42222 value for -1.
42223
42224 The old Xprintf() API is deprecated, but left for compatibility for now.
42225
42226 The new API is added in a new header so that it can be used in parts of
42227 the server such as hw/xfree86/parser that don't include all the server
42228 headers.
42229
42230 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42231 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
42232
42233commit 2db6951763115a4a5dfac53d1a8f78ac33f91235
42234Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42235Date: Fri Dec 3 16:32:01 2010 -0800
42236
42237 Sun's copyrights now belong to Oracle
42238
42239 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42240 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
42241
42242commit f06e8a61d6ed8f3054878167bffee619f852aad9
42243Author: Alan Coopersmith <alan.coopersmith@oracle.com>
42244Date: Tue Nov 30 16:34:38 2010 -0800
42245
42246 Move inclusion of panoramiX headers to top of render.c
42247
42248 Fixes compiler error from Sun compilers due to _X_EXPORT declaration
42249 being included after the unlabeled version:
42250
42251 "../Xext/panoramiXsrv.h", line 29: redeclaration must have the same or more restrictive linker scoping: XRT_PICTURE
42252
42253 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42254 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
42255
42256commit 9f9c732311ef0cc198e5ec44d02412eb7e8c3f3a
42257Author: Cyril Brulebois <kibi@debian.org>
42258Date: Fri Nov 12 21:35:25 2010 +0100
42259
42260 dix: Fix logic for displaying deprecation warning only once.
42261
42262 As pointed out by Jamey Sharp (again), the logic is faulty: --warn is
42263 always going to be false. Replace it with warn-- accordingly, so that
42264 there's (at least, but also only) one warning showing up.
42265
42266 Reviewed-by: Jamey Sharp <jamey@minilop.net>
42267 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42268
42269commit a937803c1f671ef29332e5fe8c190d8b48239912
42270Author: Cyril Brulebois <kibi@debian.org>
42271Date: Fri Nov 12 21:29:26 2010 +0100
42272
42273 dix: Simplify deprecated *Lookup* wrappers around dixLookup*.
42274
42275 As pointed out by Jamey Sharp: “the result pointer is already guaranteed
42276 to be NULL if the return value is not Success”, so get rid of the
42277 variable used to catch the return value, and used in a ternary operation
42278 to decide whether to return the pointer or NULL. Always return the
42279 result pointer instead.
42280
42281 Reviewed-by: Jamey Sharp <jamey@minilop.net>
42282 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42283
42284commit 73fbc4a4a7997b3ee1c779d8f394114270bcb20d
42285Author: Cyril Brulebois <kibi@debian.org>
42286Date: Fri Nov 12 12:04:11 2010 +0100
42287
42288 Replace SecurityLookupIDByClass() with dixLookupResourceByClass().
42289
42290 This patch has been prepared with the following Coccinelle semantic patch:
42291
42292 @@
42293 expression a, b, c, d;
42294 identifier r;
42295 type t;
42296 @@
42297 -r = (t) SecurityLookupIDByClass(a, b, c, d);
42298 +dixLookupResourceByClass((pointer*) &r, b, c, a, d);
42299
42300 The only occurrence not matching directly was processed after separating
42301 declaration of pXinDraw and call to SecurityLookupIDByClass().
42302
42303 Reviewed-by: Jamey Sharp <jamey@minilop.net>
42304 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42305
42306commit 6e8b34b7048bc9e49ef52b81e2de1b518a05890e
42307Author: Cyril Brulebois <kibi@debian.org>
42308Date: Fri Nov 12 11:36:36 2010 +0100
42309
42310 Replace SecurityLookupIDByType() with dixLookupResourceByType(), take 2.
42311
42312 These occurrences are a bit harder to catch through a semantic patch,
42313 so process them “manually”.
42314
42315 Reviewed-by: Jamey Sharp <jamey@minilop.net>
42316 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42317
42318commit ac04c1377a42ab6803adf551756653fd8e1a694b
42319Author: Cyril Brulebois <kibi@debian.org>
42320Date: Fri Nov 12 11:36:31 2010 +0100
42321
42322 Replace SecurityLookupIDByType() with dixLookupResourceByType(), take 1.
42323
42324 This patch has been prepared with the following Coccinelle semantic patch:
42325
42326 @@
42327 expression a, b, c, d;
42328 identifier r;
42329 @@
42330 -r = SecurityLookupIDByType(a, b, c, d);
42331 +dixLookupResourceByType((pointer*) &r, b, c, a, d);
42332
42333 Reviewed-by: Jamey Sharp <jamey@minilop.net>
42334 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42335
42336commit 28b4c02d7463ffe5c29f803f5656c5ecfc040185
42337Author: Cyril Brulebois <kibi@debian.org>
42338Date: Thu Nov 11 18:36:45 2010 +0100
42339
42340 Enable silent rules for cat and echo.
42341
42342 One gets this:
42343 | GEN lf3-xaaBitmap.c
42344 | GEN lf3-xaaBitmap.c
42345 | GEN lf3-xaaBitmap.c
42346 | GEN lf3-xaaBitmap.c
42347 | CC lf3-xaaBitmap.lo
42348
42349 instead of this:
42350 | echo "#define LSBFIRST" > l3-xaaBitmap.c
42351 | echo "#define TRIPLE_BITS" >> l3-xaaBitmap.c
42352 | echo '#include "../../../../hw/xfree86/xaa/xaaBitmap.c"' >> l3-xaaBitmap.c
42353 | CC l3-xaaBitmap.lo
42354
42355 Occurrences found using:
42356 | git grep -E '(cat|echo)' -- '*Makefile.am'
42357
42358 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
42359 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42360
42361commit 838b9582280030a159878af566126cb0a523ec29
42362Author: Cyril Brulebois <kibi@debian.org>
42363Date: Thu Nov 11 15:40:13 2010 +0100
42364
42365 Replace LookupIDByType() with dixLookupResourceByType(), take 2.
42366
42367 These occurrences are a bit harder to catch through a semantic patch,
42368 so process them “manually”.
42369
42370 Reviewed-by: Jamey Sharp <jamey@minilop.net>
42371 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42372
42373commit ffb93533fd8067c7383aef5e0018a197af59b46c
42374Author: Cyril Brulebois <kibi@debian.org>
42375Date: Thu Nov 11 15:30:15 2010 +0100
42376
42377 Replace LookupIDByType() with dixLookupResourceByType(), take 1.
42378
42379 This patch has been prepared with the following Coccinelle semantic patch:
42380
42381 @@
42382 expression x, y;
42383 type t;
42384 identifier r;
42385 @@
42386 -r = (t) LookupIDByType(x, y);
42387 +dixLookupResourceByType((pointer*)&r, x, y, NullClient, DixUnknownAccess);
42388
42389 @@
42390 expression x, y;
42391 type t;
42392 identifier r;
42393 @@
42394 -t r = (t) LookupIDByType(x, y);
42395 +t r;
42396 +dixLookupResourceByType((pointer*)&r, x, y, NullClient, DixUnknownAccess);
42397
42398 Reviewed-by: Jamey Sharp <jamey@minilop.net>
42399 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42400
42401commit d1b45b0fd54efb952f9ff435a516c1bfd85ca186
42402Author: Cyril Brulebois <kibi@debian.org>
42403Date: Thu Nov 11 11:35:47 2010 +0100
42404
42405 Fix missing <string.h> include.
42406
42407 The following happens otherwise (with -Wall -Werror):
42408 | In file included from /usr/include/X11/Xfuncs.h:47,
42409 | from ../../include/misc.h:112,
42410 | from ../../include/screenint.h:52,
42411 | from ../../include/scrnintstr.h:52,
42412 | from ../../dix/cursor.c:58:
42413 | /usr/include/string.h:534: error: conflicting types for ‘xstrcasecmp’
42414 | ../../include/os.h:488: note: previous declaration of ‘xstrcasecmp’ was here
42415 | /usr/include/string.h:538: error: conflicting types for ‘xstrncasecmp’
42416 | ../../include/os.h:493: note: previous declaration of ‘xstrncasecmp’ was here
42417
42418 Reviewed-by: Julien Cristau <jcristau@debian.org>
42419 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42420
42421commit 780754050bc9cb1489f92a2a890ab5665e3e6358
42422Author: Cyril Brulebois <kibi@debian.org>
42423Date: Fri Nov 12 12:18:59 2010 +0100
42424
42425 Fix GLX_LIBS vs. DMX_LIBS order.
42426
42427 If glxproxy needs symbols which aren't pulled in by dmx itself, glxproxy
42428 fails to link because of undefined references.
42429
42430 Signed-off-by: Cyril Brulebois <kibi@debian.org>
42431 Reviewed-by: Julien Cristau <jcristau@debian.org>
42432 Signed-off-by: Keith Packard <keithp@keithp.com>
42433
42434commit 14983286c489ea1ec6ba4fadbeaec5c2d2ee34e1
42435Author: Keith Packard <keithp@keithp.com>
42436Date: Mon Dec 6 20:45:35 2010 -0800
42437
42438 Version bumped to 1.9.99.901 (1.10 RC1)
42439
42440 Signed-off-by: Keith Packard <keithp@keithp.com>
42441
42442commit d96d5f5443358e33e47d1a61f9bd2afd8064a8dd
42443Author: James Jones <jajones@nvidia.com>
42444Date: Sun Dec 5 19:42:41 2010 -0800
42445
42446 Bump extension ABI to 5
42447
42448 Commit 606e079cc4d9a9db3197652ca51683c36f74efb8 moved the visual
42449 field in WindowOptRec, breaking the extension module ABI.
42450
42451 Signed-off-by: James Jones <jajones@nvidia.com>
42452 Reviewed-by: Keith Packard <keithp@keithp.com>
42453 Signed-off-by: Keith Packard <keithp@keithp.com>
42454
42455commit 435361bd73b9fc733f093d81af6b839953e35176
42456Author: Keith Packard <keithp@keithp.com>
42457Date: Mon Dec 6 20:38:14 2010 -0800
42458
42459 sync: syncObject may be None in SyncInitTrigger
42460
42461 And often is, especially when called from ProcSyncCreateAlarm.
42462 Crashing in this case seems unwise.
42463
42464 Signed-off-by: Keith Packard <keithp@keithp.com>
42465
42466commit 0d01b66df9081ef48843b3bad81c56bb2cd1ae69
42467Author: Keith Packard <keithp@keithp.com>
42468Date: Sun Dec 5 23:35:28 2010 -0800
42469
42470 randr: handle RRSetCrtcConfigs request with zero configs
42471
42472 Need to actually return a reply in this case.
42473
42474 Signed-off-by: Keith Packard <keithp@keithp.com>
42475 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42476
42477commit b0f4bd61f0caf80f3be9a176f1f7a707bc6628d8
42478Author: Keith Packard <keithp@keithp.com>
42479Date: Sun Dec 5 21:53:25 2010 -0800
42480
42481 ProcRRSetCrtcConfigs uses 'configs' without being initialized
42482
42483 If the client sends invalid data for this request, the server
42484 will jump to 'sendReply' and call RRFreeCrtcConfigs, passing it the
42485 uninitialized 'configs' and 'num_configs' values.
42486
42487 Signed-off-by: Keith Packard <keithp@keithp.com>
42488 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42489
42490commit 752c368421c1c824752cf467fba9318d75d2ca2c
42491Author: Keith Packard <keithp@keithp.com>
42492Date: Sun Dec 5 20:57:47 2010 -0800
42493
42494 Separate out screen size and screen pixmap sizes in RRScreenSizeSet
42495
42496 This provides for separate sizes for the screen scanout and rendering
42497 buffer and the application-visible screen size.
42498
42499 Signed-off-by: Keith Packard <keithp@keithp.com>
42500 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42501
42502commit a88d70fb20a2bc3152b84adff4380857e6cfadf5
42503Author: Keith Packard <keithp@keithp.com>
42504Date: Sun Dec 5 20:55:46 2010 -0800
42505
42506 Set sprite transforms from RRSetCrtcConfigs
42507
42508 These were getting ignored.
42509
42510 Signed-off-by: Keith Packard <keithp@keithp.com>
42511 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42512
42513commit 96b4d4787bf82edd9d06eb9a6e94bc45412c7df2
42514Author: Keith Packard <keithp@keithp.com>
42515Date: Sun Dec 5 20:49:19 2010 -0800
42516
42517 DIX is responsible for ref counting scanout pixmaps.
42518
42519 Remove some extra ref counting inside hw/xfree86/modes
42520
42521 Signed-off-by: Keith Packard <keithp@keithp.com>
42522 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42523
42524commit afb6ebf1d5829346c40fe1053c9f50afe926e6c6
42525Author: Keith Packard <keithp@keithp.com>
42526Date: Fri Dec 3 13:04:37 2010 -0800
42527
42528 randr: Hook up the new RandR 1.4 functionality
42529
42530 This bumps the supported RandR protocol version and adds the dispatch
42531 hooks needed to call the new functions
42532
42533 Signed-off-by: Keith Packard <keithp@keithp.com>
42534 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42535
42536commit 82612045e11f2b882ae132e184a9629f43f1c424
42537Author: Keith Packard <keithp@keithp.com>
42538Date: Fri Dec 3 13:00:46 2010 -0800
42539
42540 randr: Add per-crtc pixmaps
42541
42542 This adds new driver hooks to allocate scanout pixmaps and
42543 changes the mode setting APIs to pass the new scanout pixmaps
42544 along from DIX. DIX is responsible for reference counting the pixmaps
42545 by tracking them through RRCrtcNotify.
42546
42547 Signed-off-by: Keith Packard <keithp@keithp.com>
42548 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42549
42550commit 86c489c319c705f710bee3897fe27600ce15008e
42551Author: Keith Packard <keithp@keithp.com>
42552Date: Sat Dec 4 20:12:26 2010 -0800
42553
42554 hw/xfree86/modes: Add optional driver API for RRSetCrtcConfigs
42555
42556 This provides a driver hook which can either completely replace, or
42557 just validate the parameters for, the RRSetCrtcConfigs request.
42558
42559 Signed-off-by: Keith Packard <keithp@keithp.com>
42560 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42561
42562commit d94a035ea9eb3167fc4f35b2d9f0d53f8807014c
42563Author: Keith Packard <keithp@keithp.com>
42564Date: Sat Dec 4 19:22:11 2010 -0800
42565
42566 randr: Implement RRSetCrtcConfigs
42567
42568 This provides a driver-independent implementation of the
42569 RRSetCrtcConfigs API by simply using the existing interfaces.
42570
42571 Signed-off-by: Keith Packard <keithp@keithp.com>
42572 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42573
42574commit 66294afcab7b7a82f7dd897767e46c48a94b8ee8
42575Author: Keith Packard <keithp@keithp.com>
42576Date: Fri Dec 3 22:08:06 2010 -0800
42577
42578 randr: Add sprite position transforms
42579
42580 This implements sprite position transformations. Sprite image
42581 transforms are passed all the way to the DDX layer, but the images are
42582 not yet manipulated before being passed to the drivers.
42583
42584 Signed-off-by: Keith Packard <keithp@keithp.com>
42585 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42586
42587commit c8bc25fd7629df10f2825b7cc713b031ae78f223
42588Author: Keith Packard <keithp@keithp.com>
42589Date: Mon Dec 6 19:40:30 2010 -0800
42590
42591 Require RandR protocol version 1.4 or newer
42592
42593 Signed-off-by: Keith Packard <keithp@keithp.com>
42594
42595commit 8b35118c03590a7ad3786d3284bafb3f40fcb8cc
42596Author: Keith Packard <keithp@keithp.com>
42597Date: Sun Jun 6 15:23:37 2010 -0700
42598
42599 Replace huge argument list in xf86CrtcSetModeTransform with struct
42600
42601 xf86CrtcSetModeTransform was starting to get ridiculous with 6
42602 arguments, this change has it take a single structure that contains
42603 all of those values along with a set of flags that says which have
42604 changed.
42605
42606 Signed-off-by: Keith Packard <keithp@keithp.com>
42607 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
42608
42609commit f5b8bd620f91214c0b87e9b04d57015655792352
42610Merge: 01e9fa7 8127c83
42611Author: Keith Packard <keithp@keithp.com>
42612Date: Mon Dec 6 19:51:06 2010 -0800
42613
42614 Merge remote branch 'jajones/for-keith'
42615
42616commit 01e9fa7da389fc7ab834b4234b8484514144b7f4
42617Merge: 79870db 68a1b0d
42618Author: Keith Packard <keithp@keithp.com>
42619Date: Mon Dec 6 19:45:11 2010 -0800
42620
42621 Merge remote branch 'whot/for-keith'
42622
42623commit 79870dbf72d463cf9efea1b3e61d1d5759df27b5
42624Author: Gaetan Nadon <memsize@videotron.ca>
42625Date: Sat Dec 4 16:13:29 2010 -0500
42626
42627 config: build Xserver man pages using XORG_MANPAGE_SECTIONS
42628
42629 Nothing requires the use of a C preprocessor
42630 Using standard file exentions (.man) means no need for .gitignore
42631 Use standard directory and makefile
42632 Fix man page whitespace issues
42633
42634 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42635 Reviewed-by: Matt Dew <matt@osource.org>
42636 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
42637 Signed-off-by: Keith Packard <keithp@keithp.com>
42638
42639commit c4a7a5917ab828d4a0bd825a98e4d641bcb378f1
42640Merge: 5de312a 714b68d
42641Author: Keith Packard <keithp@keithp.com>
42642Date: Mon Dec 6 19:28:07 2010 -0800
42643
42644 Merge remote branch 'jeremyhu/master'
42645
42646commit 5de312a60d8227ed670849ecf888ea878aa81430
42647Merge: 311cad3 8f42b2b
42648Author: Keith Packard <keithp@keithp.com>
42649Date: Mon Dec 6 19:22:52 2010 -0800
42650
42651 Merge remote branch 'alanc/master'
42652
42653commit 8127c83c81bf64369a8ba2999088226d14e0b128
42654Author: James Jones <jajones@nvidia.com>
42655Date: Mon Dec 6 12:14:01 2010 -0800
42656
42657 Expose Sync Fence Object protocol
42658
42659 Add the new protocol handlers for XSync 3.1 to the
42660 dispatch tables and report support for Sync protocol
42661 version 3.1.
42662
42663 Signed-off-by: James Jones <jajones@nvidia.com>
42664 Reviewed-by: Keith Packard <keithp@keithp.com>
42665
42666commit 9c0c7cc9a7adf230b5eee0b4166f9e17e07a088b
42667Author: James Jones <jajones@nvidia.com>
42668Date: Mon Dec 6 13:10:42 2010 -0800
42669
42670 Add XSyncAwaitFence() handler
42671
42672 -Add the actual ProcSyncAwaitFence() dispatch func
42673
42674 -Add support for fence sync triggers.
42675
42676 Signed-off-by: James Jones <jajones@nvidia.com>
42677 Reviewed-by: Keith Packard <keithp@keithp.com>
42678
42679commit 397dfd9f87e1cdf105d10a789a97230f4f0d204e
42680Author: James Jones <jajones@nvidia.com>
42681Date: Fri Jun 25 17:59:14 2010 -0700
42682
42683 Create/Destroy/Trigger/Reset/Query Fence Sync objs
42684
42685 Initial server side implementation of fence sync
42686 objects. Allows creation, management, and state
42687 queries of binary state objects. Currently they
42688 are not very useful as there is no way to wait for
42689 them efficiently.
42690
42691 The basic trigger operation added here triggers
42692 relative to a given X screen's rendering operations.
42693 To perform this operation, fence sync objects must
42694 be tied to a screen. As Aaron Plattner pointed out,
42695 screens are identified but a drawable in X protocol,
42696 so a drawable argument is included in
42697 XSyncCreateFence(). The screen also could have been
42698 specified as part of the trigger operation. However,
42699 it is also desireable to associate a screen with
42700 fence sync objects at creation time so that the
42701 associated screen's driver can allocate any HW-
42702 specific resources needed by the fence object up
42703 front.
42704
42705 Signed-off-by: James Jones <jajones@nvidia.com>
42706 Reviewed-by: Keith Packard <keithp@keithp.com>
42707
42708commit 53ea965d64bd9563e6bfcc81d04f283d2b85f09d
42709Author: James Jones <jajones@nvidia.com>
42710Date: Mon Nov 29 10:09:36 2010 -0800
42711
42712 Require xextproto 7.1.99
42713
42714 Subsequent changes rely on fence sync protocol
42715 in the sync extension. This protocol is only
42716 complete in xextproto version 7.1.99 and
42717 above.
42718
42719 Signed-off-by: James Jones <jajones@nvidia.com>
42720 Reviewed-by: Keith Packard <keithp@keithp.com>
42721
42722commit 1c4a0db2c6bf0320cb630b84ab87bcfd3801a53d
42723Author: James Jones <jajones@nvidia.com>
42724Date: Mon Jun 28 16:10:13 2010 -0700
42725
42726 Add fence sync driver interface
42727
42728 -Add fence sync objects
42729
42730 -Add fence sync devPrivates
42731
42732 -Add a X sync module screen private
42733
42734 -Add wrappable functions to create and destroy
42735 fence sync objects
42736
42737 -Give fence sync objects wrappable functions to
42738 trigger, test, and reset their 'triggered' value.
42739
42740 -Give fence sync objects wrappable functions to
42741 notify driver when adding/removing triggers to/
42742 from the sync object.
42743
42744 Signed-off-by: James Jones <jajones@nvidia.com>
42745 Reviewed-by: Keith Packard <keithp@keithp.com>
42746
42747commit af0f9f913398d34a885c3fb4e8d40c1a7e2b3ee9
42748Author: James Jones <jajones@nvidia.com>
42749Date: Mon Jun 28 15:39:04 2010 -0700
42750
42751 Move some sync code to miext
42752
42753 As a precursor to the fence sync object video driver
42754 and extension API, move some code from Xext to
42755 miext/sync. Most of this is just code to set up the
42756 build system to include the new directory. No
42757 functional code is added in this change.
42758
42759 Signed-off-by: James Jones <jajones@nvidia.com>
42760 Reviewed-by: Keith Packard <keithp@keithp.com>
42761
42762commit 12b65de7db6e3e8bf831914d247da269d01c5fbe
42763Author: James Jones <jajones@nvidia.com>
42764Date: Mon Dec 6 10:11:45 2010 -0800
42765
42766 Factor out generic code from ProcSyncAwait()
42767
42768 In preparation for adding more sync object types
42769 that will need Await requests of their own, factor
42770 out some setup and finalization code from
42771 ProcSyncAwait() into SyncAwaitPrologue() and
42772 SyncAwaitEpilogue()
42773
42774 Signed-off-by: James Jones <jajones@nvidia.com>
42775 Reviewed-by: Keith Packard <keithp@keithp.com>
42776
42777commit c66a410d378090f350beb398649e9d9262933785
42778Author: James Jones <jajones@nvidia.com>
42779Date: Mon Jun 28 14:59:01 2010 -0700
42780
42781 Make Await SyncTrigger functions generic
42782
42783 Update all the functions dealing with Await
42784 sync triggers handle generic sync objects
42785 instead of just counters. This will
42786 facilitate code sharing between the counter
42787 sync waits and the fence sync waits.
42788
42789 Signed-off-by: James Jones <jajones@nvidia.com>
42790 Reviewed-by: Keith Packard <keithp@keithp.com>
42791
42792commit 99daf419a3070c347c0cd29b4a9bae07eca7bd6d
42793Author: James Jones <jajones@nvidia.com>
42794Date: Mon Jun 28 14:51:54 2010 -0700
42795
42796 Create SyncObject base type.
42797
42798 SyncObject is now the base type for SyncCounter.
42799 Data to be used by all sync types is stored in
42800 the base object. SyncCounter can be safely cast
42801 to SyncObject, and a SyncObject can be cast to
42802 the correct type based on SyncObject::type.
42803
42804 Signed-off-by: James Jones <jajones@nvidia.com>
42805 Reviewed-by: Keith Packard <keithp@keithp.com>
42806
42807commit 27593eea7efcbed8de0c6e8233cbd1a1b8a50459
42808Author: James Jones <jajones@nvidia.com>
42809Date: Mon Dec 6 13:36:14 2010 -0800
42810
42811 Add and use SERVER_SYNC_*_VERSION
42812
42813 Most extensions have a version defined
42814 in the protocol headers, and also in the
42815 server's protocol-versions.h. The latter
42816 defines which version the server advertises
42817 support for. Sync wasn't included in
42818 protocol-versions.h, and was advertising
42819 support for whatever was in the protocol
42820 headers the server was built against.
42821
42822 Signed-off-by: James Jones <jajones@nvidia.com>
42823 Reviewed-by: Keith Packard <keithp@keithp.com>
42824
42825commit 68a1b0de95f71f74835c6c0f002699fcdccbb268
42826Author: Peter Hutterer <peter.hutterer@who-t.net>
42827Date: Mon Dec 6 14:58:37 2010 +1000
42828
42829 xfree86: use xf86AllocateInput() for xorg.conf devices too.
42830
42831 Single allocation point for input devices, most notably a single point to
42832 reset default values.
42833 Without this patch, the file descriptor default was -1 for hotplugged
42834 devices and 0 for config devices. Drivers that don't overwrite the default
42835 themselves would thus fail if configured in the xorg.conf.
42836
42837 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42838 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
42839 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
42840
42841commit beea2378f142556471c62290e275935af848e137
42842Author: Peter Hutterer <peter.hutterer@who-t.net>
42843Date: Mon Dec 6 14:33:43 2010 +1000
42844
42845 xfree86: don't overwrite option list (#32115)
42846
42847 Options set in the configuration file were unconditionally overwritten by
42848 the server. Merge the already existing options and the new options together
42849 instead of just overwriting ones.
42850
42851 Introduced in commit 2199842ed50b3eb40d54146827fc58cae7e873ec
42852 Author: Peter Hutterer <peter.hutterer@who-t.net>
42853 Date: Thu Sep 2 10:52:54 2010 +1000
42854
42855 xfree86: remove extraOptions field from IDevRec.
42856
42857 X.Org Bug 32115 <http://bugs.freedesktop.org/show_bug.cgi?id=32115>
42858
42859 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42860 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
42861 Tested-by: David Ronis <ronis@ronispc.chem.mcgill.ca>
42862
42863commit 8f3fa8fb0b0a75dac714fc213c034b20595898d3
42864Author: Peter Hutterer <peter.hutterer@who-t.net>
42865Date: Mon Dec 6 15:44:50 2010 +1000
42866
42867 xfree86: add option "Floating", deprecate SendCoreEvents and friends.
42868
42869 Some devices should be initialised as floating from the start (e.g.
42870 Joysticks and accelerometers benefit from this). Currently users use the
42871 "SendCoreEvents" "off" flag for this, which isn't the most appropriate
42872 naming.
42873
42874 Add an option "Floating", deprecate the others. Still parsed and handled by
42875 the server.
42876
42877 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42878 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
42879
42880commit 36b614dedf4ddc428e43ad1542d4f9314f73f60a
42881Author: Eoghan Sherry <ejsherry@gmail.com>
42882Date: Sun Nov 28 16:15:51 2010 -0500
42883
42884 Xi: Fix master button update when slave buttons are mapped. #24887
42885
42886 It is currently assumed that an event button delieved to a master device
42887 corresponds to the slave button states. However, the event button is a
42888 logical (mapped) slave button and slave button states correspond to
42889 physical (unmapped) slave buttons. This leads to incorrect update of the
42890 master button state and incorrect events devlivered to clients. Fix the
42891 situation by taking the slave button map into account when querying a
42892 slave button state.
42893
42894 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=24887
42895
42896 Signed-off-by: Eoghan Sherry <ejsherry@gmail.com>
42897 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42898
42899commit 31ab9f8860848504df18a8be9d19b817b191e0df
42900Author: Peter Hutterer <peter.hutterer@who-t.net>
42901Date: Fri Sep 3 11:54:41 2010 +1000
42902
42903 mi: handle DGA subtypes when determining the master device.
42904
42905 The subtype in the DGA event is the core type and all ET_ event types (where
42906 applicable) are identical to the core types. Thus the switch statement below
42907 will work as required and assign the right master device.
42908
42909 Fixes a crasher bug on keyboard devices with valuators. If a device sends a
42910 motion event while grabbed and a DGA client is active (but has not selected
42911 input through DGA), the valuator event is posted through the VCK and
42912 eventually results in a NULL-pointer dereference on dev->valuator.
42913
42914 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42915
42916commit 5d31c3e705dfd9f38f0fffcd07a6d8d06644735c
42917Author: Peter Hutterer <peter.hutterer@who-t.net>
42918Date: Fri Nov 26 09:37:37 2010 +1000
42919
42920 dix: don't stop processing valuators when the mode changes.
42921
42922 XI 1.x events still contain absolute coordinates anyway. By the time we get
42923 to the InternalEvent to XI event conversion, the valuators are already
42924 absolute.
42925
42926 Stopping because of a different mode on a valuator is not necessary.
42927
42928 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
42929 Reviewed-by: Adam Jackson <ajax@redhat.com>
42930 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
42931
42932commit 714b68d9e5bf624a6703f168e0f7dc980e88e8c0
42933Author: Jeremy Huddleston <jeremyhu@apple.com>
42934Date: Fri Dec 3 17:27:44 2010 -0800
42935
42936 XQuartz: RandR: Fix mode changing for multi-monitor configurations.
42937
42938 This just fixes the regression whereby we couldn't switch between the legacy
42939 fullscreen mode and rootless on multi-monitor configurations. This was
42940 happening because ref wasn't being set in these cases (since we don't ever
42941 actually change CG modes), so we failed a CFEqual. Setting the references
42942 fixes this regression and places us one step closer to more mode RandR
42943 mode switching in multi-monitor configurations.
42944
42945 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
42946
42947commit 14f00449eb81771c01fffcdaf3dd697cdf4e41de
42948Author: Jeremy Huddleston <jeremyhu@apple.com>
42949Date: Fri Dec 3 16:46:11 2010 -0800
42950
42951 XQuartz: Cleanup some compiler warnings
42952
42953 Mark __crashreporter_info__ as __attribute__((__used__))
42954
42955 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
42956
42957commit 14aec8b82a316c6df51ac2a81985ebf1990abd2b
42958Author: Jeremy Huddleston <jeremyhu@apple.com>
42959Date: Fri Dec 3 16:45:37 2010 -0800
42960
42961 XQuartz: Remove one more backing store leftover
42962
42963 See: c4c4676e6874b42c2371eee96faa2c2dbb59a704
42964
42965 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
42966
42967commit ca431371a23a2b9ad36c1d64e11ea41d5e4e5f04
42968Author: Jeremy Huddleston <jeremyhu@apple.com>
42969Date: Fri Nov 26 15:31:22 2010 -0500
42970
42971 XQuartz: Disable the Mac OS X screensaver when in full screen mode
42972
42973 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
42974
42975commit 311cad33155c64ed996418808727fc417168592e
42976Author: Nicolas Peninguy <nico@lostgeeks.org>
42977Date: Wed Nov 24 00:11:11 2010 +0100
42978
42979 Fix Xdmx and Xephyr build when DTrace support is enabled
42980
42981 This fixes the following build errors when DTrace is enabled
42982 (--with-dtrace):
42983
42984 CCLD Xdmx
42985 /usr/bin/ld: ../../os/os.O: undefined reference to symbol 'dladdr@@GLIBC_2.2.5'
42986 /usr/bin/ld: note: 'dladdr@@GLIBC_2.2.5' is defined in DSO /lib64/libdl.so.2 so try adding it to the linker command line
42987
42988 CCLD Xephyr
42989 ../../../os/os.O: In function `TimerForce':
42990 /home/nico/work/xserver/os/WaitFor.c:481: multiple definition of `TimerForce'
42991 ../../../os/os.O:/home/nico/work/xserver/os/WaitFor.c:481: first defined here
42992
42993 Signed-off-by: Nicolas Peninguy <nico@lostgeeks.org>
42994 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
42995 Signed-off-by: Keith Packard <keithp@keithp.com>
42996
42997commit 3824417335ec00a87b51f8d05e592099b8507680
42998Merge: dab064f f4f41c8
42999Author: Keith Packard <keithp@keithp.com>
43000Date: Thu Dec 2 08:39:33 2010 -0800
43001
43002 Merge remote branch 'vsyrjala/xv_reput'
43003
43004commit dab064fa5e0b1f5c67222562ad5367005832cba1
43005Author: Andrea Canciani <ranma42@gmail.com>
43006Date: Tue Nov 2 20:10:32 2010 +0100
43007
43008 render: Fix byteswapping of gradient stops
43009
43010 The function swapStops repeatedly swaps the color components as
43011 CARD16, but incorrectly steps over them as if they were CARD32.
43012
43013 This causes half of the stops not to be swapped at all and some
43014 unrelated data be swapped instead.
43015
43016 Signed-off-by: Andrea Canciani <ranma42@gmail.com>
43017 Reviewed-by: Soren Sandmann <sandmann@daimi.au.dk>
43018 Reviewed-by: Julien Cristau <jcristau@debian.org>
43019 Signed-off-by: Keith Packard <keithp@keithp.com>
43020
43021commit 279ef1ffd787dba2f0d5056849b9cb15d36aa3eb
43022Author: Cyril Brulebois <kibi@debian.org>
43023Date: Wed Dec 1 14:12:55 2010 +0100
43024
43025 Fix screen number checks.
43026
43027 screenInfo.numScreens is not a valid screen number, they go from 0 to
43028 numScreens - 1.
43029
43030 Signed-off-by: Cyril Brulebois <kibi@debian.org>
43031 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
43032 Signed-off-by: Keith Packard <keithp@keithp.com>
43033
43034commit 8684543021b9b1aa165b1bc69bc58685cb5942c1
43035Author: Julien Cristau <jcristau@debian.org>
43036Date: Tue Nov 30 17:23:10 2010 +0100
43037
43038 DGA: fix screen number check
43039
43040 screenInfo.numScreens is not a valid screen number, they go from 0 to
43041 numScreens - 1.
43042
43043 Signed-off-by: Julien Cristau <jcristau@debian.org>
43044 Reviewed-by: Keith Packard <keithp@keithp.com>
43045 Signed-off-by: Keith Packard <keithp@keithp.com>
43046
43047commit 8f42b2b69387b006bfcd373c3d023ebea9035db2
43048Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43049Date: Sat Nov 27 22:34:57 2010 -0800
43050
43051 Simplify Error() - don't allocate temporary copy of error string
43052
43053 Doesn't seem to be any reason to just not pass the error string
43054 as another argument directly to LogVWrite()
43055
43056 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43057 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
43058 Reviewed-by: Julien Cristau <jcristau@debian.org>
43059
43060commit 685286b17d30335d799a9da11914943e466ea955
43061Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43062Date: Sat Nov 27 20:43:28 2010 -0800
43063
43064 FindModuleInSubdir: Stop allocating one more byte than needed
43065
43066 15ac25627e7239629be59 removed the "/" from the sprintf strings,
43067 but failed to remove the extra byte allocated for the '/'.
43068
43069 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43070 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
43071 Reviewed-by: Julien Cristau <jcristau@debian.org>
43072
43073commit 40d5a019352fa8f12230c863e11cbb1f6258a93e
43074Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43075Date: Sat Nov 27 19:50:38 2010 -0800
43076
43077 xf86VIDrvMsgVerb: print args, not format string
43078
43079 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43080 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
43081 Reviewed-by: Julien Cristau <jcristau@debian.org>
43082
43083commit 4bbc90cd8b7e749fd8072ce7cd8dd998f4396981
43084Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43085Date: Sat Nov 27 19:06:56 2010 -0800
43086
43087 xf86AutoConfig: make copyScreen memory allocation & error handling more sane
43088
43089 No point calling the no-fail-alloc if you check for failure and your
43090 only caller checks for failure.
43091
43092 No point calling calloc to zero fill memory you're about to memcpy over.
43093
43094 In the unlikely event of a loss of memory allocation, drop your previous
43095 allocations before returning to others.
43096
43097 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43098 Reviewed-by: Julien Cristau <jcristau@debian.org>
43099 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
43100
43101commit 2c8e534c8e9334562485aeaaef374871cf14d5fe
43102Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43103Date: Sat Nov 27 23:49:55 2010 -0800
43104
43105 xf86ValidateModes: xnfalloc(strlen) + strcpy => xnfstrdup
43106
43107 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43108 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
43109 Reviewed-by: Julien Cristau <jcristau@debian.org>
43110
43111commit 29e467a1f1548a826ee2793244e3ff416aa1a0f2
43112Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43113Date: Fri Nov 26 16:53:40 2010 -0800
43114
43115 xf86OutputRename: Replace another strlen/malloc/strcpy set with strdup
43116
43117 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43118 Reviewed-by: Matt Turner <mattst88@gmail.com>
43119
43120commit d346bc3083c6d4bea59b77f634c7c5ec6c1d8cc9
43121Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43122Date: Sun Nov 28 13:48:16 2010 -0800
43123
43124 Fix compiler warnings in hw/xfree86/os-support/solaris
43125
43126 sun_init.c: In function `xf86OpenConsole':
43127 sun_init.c:99: warning: cast does not match function type
43128 sun_init.c:74: warning: unused variable `FreeVTslot'
43129 sun_init.c: In function `xf86UseMsg':
43130 sun_init.c:417: warning: old-style parameter declaration
43131
43132 sun_vid.c: In function `solUnMapVidMem':
43133 sun_vid.c:162: warning: long unsigned int format, pointer arg (arg 6)
43134 sun_vid.c: In function `xf86ReadBIOS':
43135 sun_vid.c:217: warning: long unsigned int format, pointer arg (arg 5)
43136 sun_vid.c:217: warning: long unsigned int format, int arg (arg 6)
43137
43138 sun_agp.c: In function `xf86EnableAGP':
43139 sun_agp.c:321: warning: unsigned int format, CARD32 arg (arg 4)
43140
43141 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43142 Reviewed-by: Julien Cristau <jcristau@debian.org>
43143
43144commit d75777d54c2107163305f50e8ee4306da202b95e
43145Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43146Date: Sun Nov 28 13:45:40 2010 -0800
43147
43148 Move xchomp inside #ifdef __linux__
43149
43150 static function only called from the matchDriverFromFiles function
43151 that's inside #ifdef __linux__ section
43152
43153 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43154 Reviewed-by: Matt Turner <mattst88@gmail.com>
43155
43156commit ccbba444b7b8e1ba555532a847377600bea43d03
43157Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43158Date: Sun Nov 28 10:45:17 2010 -0800
43159
43160 config: Remove AC_PROG_CC, SED & INSTALL that XORG_DEFAULT_OPTIONS provide
43161
43162 Most importantly removes AC_PROG_CC call that resets compiler flags back
43163 to C89 mode, breaking use of C99 isfinite() on Solaris in dix/devices.c.
43164
43165 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43166 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
43167
43168commit 92cde0b84cdec164cd698dc74a7512d0791d7708
43169Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43170Date: Sat Nov 27 08:15:52 2010 -0800
43171
43172 Xserver-spec: Update discussion of font library
43173
43174 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43175 Reviewed-by: Julien Cristau <jcristau@debian.org>
43176 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43177
43178commit c1e769250078cbc74d7e6e6ddc5323c4f420ab14
43179Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43180Date: Sat Nov 27 08:06:40 2010 -0800
43181
43182 Xserver-spec: Update location of log functions
43183
43184 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43185 Reviewed-by: Julien Cristau <jcristau@debian.org>
43186 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43187
43188commit deae18f3cad94959110761ef2ecd9665690e3db5
43189Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43190Date: Sat Nov 27 00:45:48 2010 -0800
43191
43192 Xserver-spec: Fix assorted typos
43193
43194 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43195 Reviewed-by: Julien Cristau <jcristau@debian.org>
43196 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43197
43198commit a410bf53798bdca43f99476a01ef27cabdf73e01
43199Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43200Date: Sat Nov 27 00:35:28 2010 -0800
43201
43202 Xserver-spec: Update lists of macros
43203
43204 LOOKUP_DRAWABLE & VERIFY_GC are no longer in dix.h, but
43205 WriteReplyToClient & WriteSwappedDataToClient are.
43206
43207 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43208 Reviewed-by: Julien Cristau <jcristau@debian.org>
43209 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43210
43211commit de518c8f378ea31345c946693d58a26a493af603
43212Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43213Date: Sat Nov 27 00:30:37 2010 -0800
43214
43215 Xserver-spec: Remove CreateCallbackList
43216
43217 The function is defined as a static, so can't be called by anyone
43218 but AddCallback.
43219
43220 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43221 Reviewed-by: Julien Cristau <jcristau@debian.org>
43222 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43223
43224commit f6c880b257a21a574cf1a47095cb39f32252802e
43225Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43226Date: Sat Nov 27 00:27:46 2010 -0800
43227
43228 Xserver-spec: Replace deprecated resource id lookup functions
43229
43230 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43231 Reviewed-by: Julien Cristau <jcristau@debian.org>
43232 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43233
43234commit 922806a5aa6eafc432d6787495b475aaa3f1790d
43235Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43236Date: Sat Nov 27 00:14:51 2010 -0800
43237
43238 Xserver-spec: Update Memory Management functions
43239
43240 Xalloc, Xrealloc, & Xfree are deprecated now
43241 ALLOCATE_LOCAL is removed due to stack overflow issues
43242
43243 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43244 Reviewed-by: Julien Cristau <jcristau@debian.org>
43245 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43246
43247commit 02449ee24b484c9fea501df5274d95a9f87cab23
43248Merge: afd6eb6 b85f906
43249Author: Keith Packard <keithp@keithp.com>
43250Date: Tue Nov 30 13:30:59 2010 -0800
43251
43252 Merge remote branch 'ajax/for-keithp'
43253
43254commit afd6eb66d5ef6a59be4c03da4588aaa2ee47b9ad
43255Merge: b169649 ffcbfa0
43256Author: Keith Packard <keithp@keithp.com>
43257Date: Tue Nov 30 13:28:13 2010 -0800
43258
43259 Merge remote branch 'aplattner/for/keith'
43260
43261commit b16964910d29c0bd039e8bb48bcf1199d709fe3e
43262Author: Ferry Huberts <ferry.huberts@pelagic.nl>
43263Date: Tue Nov 30 19:06:55 2010 +0100
43264
43265 dix: do not use bit-wise operators on the boolean result of BitIsOn
43266
43267 Performing bit-wise operations on a boolean amounts to mixing types,
43268 is confusing and basically incorrect; one should only perform
43269 logical operations on booleans.
43270
43271 Performing such operations relies on the implementation detail
43272 that a boolean is in fact an integer and that its value FALSE
43273 is implemented as zero.
43274
43275 Signed-off-by: Ferry Huberts <ferry.huberts@pelagic.nl>
43276 Reviewed-by: Matt Turner <mattst88@gmail.com>
43277 Reviewed-by: Keith Packard <keithp@keithp.com>
43278 Signed-off-by: Keith Packard <keithp@keithp.com>
43279
43280commit 5f34853d6f936e8a1a317a59972b07e3f191d1ab
43281Author: Adam Jackson <ajax@redhat.com>
43282Date: Mon Nov 15 11:25:13 2010 +1000
43283
43284 xfree86: Bump classic driver default to 1024x768
43285
43286 Signed-off-by: Adam Jackson <ajax@redhat.com>
43287 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43288 Reviewed-by: Keith Packard <keithp@keithp.com>
43289 Signed-off-by: Keith Packard <keithp@keithp.com>
43290
43291commit 903e0f6f0f0ec54151640fc459ff03dcba0522af
43292Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43293Date: Sat Nov 27 10:46:44 2010 -0800
43294
43295 Xserver-spec: Delete DBE Idioms section
43296
43297 The code has been gone for a while
43298
43299 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43300 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43301 Reviewed-by: Keith Packard <keithp@keithp.com>
43302 Signed-off-by: Keith Packard <keithp@keithp.com>
43303
43304commit f4ba75a494cf9f4f5b0a979e9ceb519f4a6fcacc
43305Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43306Date: Sat Nov 27 10:46:43 2010 -0800
43307
43308 Remove unused ReqLen & CastxReq macros
43309
43310 According to Xserver-spec, they were part of the now-deleted DBE "Idioms"
43311 code. The last callers of them were removed in commits fe616f9230b6 &
43312 3d642905477f.
43313
43314 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43315 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43316 Reviewed-by: Keith Packard <keithp@keithp.com>
43317 Signed-off-by: Keith Packard <keithp@keithp.com>
43318
43319commit b85f9063c2370d7710523bccb8bf030a0710425b
43320Author: Adam Jackson <ajax@redhat.com>
43321Date: Mon Nov 22 16:42:04 2010 -0500
43322
43323 xfree86: Remove the xf8_16bpp overlay module
43324
43325 Only one (marginal) driver was using it, and it's been fixed to just
43326 implement it directly.
43327
43328 v2: Also fix sdksyms.sh (spotted by Jesse Adkins)
43329 v3: Also fix DESIGN.xml (spotted bu Julien Cristau)
43330
43331 Reviewed-by: Julien Cristau <jcristau@debian.org>
43332 Signed-off-by: Adam Jackson <ajax@redhat.com>
43333
43334commit a861fe6a1cf56df76928d21707551602289963c1
43335Author: Adam Jackson <ajax@redhat.com>
43336Date: Tue Nov 30 13:50:42 2010 -0500
43337
43338 vbe: Refuse to believe tiny (or negative) panel sizes from PanelID
43339
43340 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
43341 Signed-off-by: Adam Jackson <ajax@redhat.com>
43342
43343commit 84a14fab8f930ef1855444ae4e9e3e14ee008328
43344Author: David Reveman <davidr@novell.com>
43345Date: Wed Jan 6 14:55:48 2010 -0500
43346
43347 composite: add panoramix support
43348
43349 Taken from:
43350 50d2d8c8969c165582d215c6e85c4be9eac02b6a
43351 dbffd0d44a33dcc84898c7a891d7ba212f65cbb8
43352 9b5b102163b4eaa1b70647354fcab4f6e461c94c
43353 75f9b98af31abf537ac6616c99f3797deb7ba017
43354 07fba8b1f77a6bca44ea6568b346a18ce9d1e61d
43355
43356 With minor style fixes, ported to dixLookupResourceByType, and ported
43357 away from client->noClientException and xalloc/xfree.
43358
43359 v2: Fix a memory leak in PanoramiXCompositeNameWindowPixmap, spotted by
43360 James Jones.
43361 v3: Fix a buglet in PanoramiXCompositeUnredirectSubwindows, spotted by
43362 Dave Airlie.
43363 v4: Fix a style issue with resource lookup noted by Jamey Sharp.
43364
43365 Reviewed-by: Dave Airlie <airlied@gmail.com>
43366 Signed-off-by: Adam Jackson <ajax@redhat.com>
43367
43368commit 6dea617e0f71a6fc7937e3a1e10474fa87488284
43369Author: David Reveman <davidr@novell.com>
43370Date: Tue Jan 5 15:35:10 2010 -0500
43371
43372 fixes: Add panoramix support
43373
43374 Taken from:
43375 583f4dde81d8d5e9101b0289946e1914ea1ee124
43376
43377 With minor style fixes and ported to dixLookupResourceByType.
43378
43379 Reviewed-by: Dave Airlie <airlied@gmail.com>
43380 Signed-off-by: Adam Jackson <ajax@redhat.com>
43381
43382commit 1b3c57f075049250c6d059412af3d6954468a5b7
43383Author: Adam Jackson <ajax@redhat.com>
43384Date: Fri Aug 20 12:25:23 2010 -0400
43385
43386 xfree86: Remove unused xf86PixmapKeyRec
43387
43388 Reviewed-by: Matt Turner <mattst88@gmail.com>
43389 Signed-off-by: Adam Jackson <ajax@redhat.com>
43390
43391commit 63b9d67d7f4f1fa108370c619e1d0d06baedbd4c
43392Author: Adam Jackson <ajax@redhat.com>
43393Date: Thu Aug 19 18:40:24 2010 -0400
43394
43395 mi: Remove unused miAllocateGCPrivateIndex
43396
43397 Reviewed-by: Matt Turner <mattst88@gmail.com>
43398 Signed-off-by: Adam Jackson <ajax@redhat.com>
43399
43400commit 87f5b843a08fd92b4b31c04e3e6d947b8bc0d160
43401Author: Adam Jackson <ajax@redhat.com>
43402Date: Fri Nov 19 16:25:51 2010 -0500
43403
43404 render: Reshuffle and cull dead field from GlyphSetRec
43405
43406 Reviewed-by: Keith Packard <keithp@keithp.com>
43407 Reviewed-by: Matt Turner <mattst88@gmail.com>
43408 Signed-off-by: Adam Jackson <ajax@redhat.com>
43409
43410commit 606e079cc4d9a9db3197652ca51683c36f74efb8
43411Author: Adam Jackson <ajax@redhat.com>
43412Date: Fri Nov 19 16:17:48 2010 -0500
43413
43414 dix: reshuffle WindowOptRec to fill a hole on LP64
43415
43416 Reviewed-by: Keith Packard <keithp@keithp.com>
43417 Signed-off-by: Adam Jackson <ajax@redhat.com>
43418
43419commit fba5c8154dbc8f8c33801a5a43d7361cd6b54ffe
43420Author: Adam Jackson <ajax@redhat.com>
43421Date: Fri Nov 19 15:21:31 2010 -0500
43422
43423 dix: Remove unused ChangeSaveUnder hooks
43424
43425 Reviewed-by: Keith Packard <keithp@keithp.com>
43426 Reviewed-by: Matt Turner <mattst88@gmail.com>
43427 Signed-off-by: Adam Jackson <ajax@redhat.com>
43428
43429commit c4c4676e6874b42c2371eee96faa2c2dbb59a704
43430Author: Adam Jackson <ajax@redhat.com>
43431Date: Fri Nov 19 15:04:09 2010 -0500
43432
43433 dix: Remove the backing store leftovers
43434
43435 Reviewed-by: Keith Packard <keithp@keithp.com>
43436 Reviewed-by: Matt Turner <mattst88@gmail.com>
43437 Signed-off-by: Adam Jackson <ajax@redhat.com>
43438
43439commit 48bc8d0dd99284f509bc6de01a8fd24a7bfc69d5
43440Author: Adam Jackson <ajax@redhat.com>
43441Date: Wed Oct 6 09:56:56 2010 -0400
43442
43443 render: repack PictureRec
43444
43445 Eliminate the unused dither field, move filter and stateChanges into the
43446 bitfield, and reorder elements to pack holes on LP64.
43447
43448 sizeof(PictureRec) ILP32 LP64
43449 before: 84 152
43450 after: 72 120
43451
43452 Reviewed-by: Dave Airlie <airlied@redhat.com>
43453 Reviewed-by: Matt Turner <mattst88@gmail.com>
43454 Signed-off-by: Adam Jackson <ajax@redhat.com>
43455
43456commit ffcbfa0063bdc7c9ad5ac724285c7b6e67044c18
43457Author: Aaron Plattner <aplattner@nvidia.com>
43458Date: Mon Nov 15 20:43:18 2010 -0800
43459
43460 xfree86: Fix rotation of 2-color non-interleaved cursor images
43461
43462 When RandR 1.2's transformation code is enabled, it rotates the cursor
43463 image so that it appears upright on a rotated screen. This code
43464 completely mangles 2-color cursors on hardware where the the mask and
43465 source images are not interleaved due to two problems:
43466
43467 1. stride is calculated as (width / 4) rather than (width / 8), so the
43468 expression (y * stride) skips two lines instead of one for every
43469 time y is incremented.
43470 2. cursor_bitpos ignores the 'mask' parameter if the hardware doesn't
43471 specify any of the HARDWARE_CURSOR_SOURCE_MASK_INTERLEAVE_* flags.
43472
43473 To fix this, refactor the code to pass the whole xf86CursorInfoPtr
43474 through to cursor_bitpos and compute the correct stride there based on
43475 the flags. If none of the SOURCE_MASK_INTERLEAVE flags are set, use
43476 the total cursor size to move the 'image' variable into the mask part
43477 of the image before computing the desired byte pointer.
43478
43479 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
43480 Reviewed-by: Robert Morell <rmorell@nvidia.com>
43481 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
43482 Tested-by: Cyril Brulebois <kibi@debian.org>
43483
43484commit 7ee965a300c9eddcc1acacf9414cfe3e589222a8
43485Author: Adam Jackson <ajax@redhat.com>
43486Date: Tue May 11 09:16:19 2010 -0400
43487
43488 linux: Refactor VT switch on {re,}generation and shutdown
43489
43490 This makes more things fatal than were fatal before, but that's correct;
43491 if you need the VT, then failing to get it on regeneration means things
43492 are about to go very very badly.
43493
43494 Reviewed-by: Julien Cristau <jcristau@debian.org>
43495 Signed-off-by: Adam Jackson <ajax@redhat.com>
43496
43497commit 0def735c6a9fab19acc4b806def6921a702f617f
43498Author: Adam Jackson <ajax@redhat.com>
43499Date: Tue May 11 08:40:33 2010 -0400
43500
43501 xfree86: Remove dead ifdefs from VT switching
43502
43503 Reviewed-by: Julien Cristau <jcristau@debian.org>
43504 Signed-off-by: Adam Jackson <ajax@redhat.com>
43505
43506commit 96a7a7d197d5f96e7aa51928155d4b8d5a115875
43507Author: Adam Jackson <ajax@redhat.com>
43508Date: Mon May 10 18:40:09 2010 -0400
43509
43510 linux: Don't muck about with tty permissions
43511
43512 This is not X's job.
43513
43514 Reviewed-by: Julien Cristau <jcristau@debian.org>
43515 Signed-off-by: Adam Jackson <ajax@redhat.com>
43516
43517commit 6ce1908ba4959aaa0a48171acb62cf6a5d5cbf18
43518Author: Adam Jackson <ajax@redhat.com>
43519Date: Wed May 12 08:50:54 2010 -0400
43520
43521 xfree86: Remove os-support/sysv
43522
43523 Nothing's using it, the SysV derivatives we support have their own
43524 custom versions.
43525
43526 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43527 Signed-off-by: Adam Jackson <ajax@redhat.com>
43528
43529commit 01159ffc4718ba48ab7b16acaffd8008f9803811
43530Author: Adam Jackson <ajax@redhat.com>
43531Date: Mon May 10 18:14:49 2010 -0400
43532
43533 linux: Remove redundant variable
43534
43535 Reviewed-by: Julien Cristau <jcristau@debian.org>
43536 Signed-off-by: Adam Jackson <ajax@redhat.com>
43537
43538commit a80780a7638f847c3be20e5e0c7fe85e83d9bdd1
43539Author: Adam Jackson <ajax@redhat.com>
43540Date: Wed Nov 17 09:03:06 2010 -0500
43541
43542 glx: Remove swap barrier and hyperpipe support
43543
43544 Never implemented in any open source driver. The implementation
43545 assumed explicit DDX driver knowledge of how the client-side driver
43546 worked, since at the time the server's GL renderer was not a DRI driver.
43547 But now, it is, so any implementation of these should be done with
43548 additional DRI driver API, like the swap control extension.
43549
43550 Reviewed-by: Julien Cristau <jcristau@debian.org>
43551 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
43552 Signed-off-by: Adam Jackson <ajax@redhat.com>
43553
43554commit f4f41c812d2ba5edba7e0e0dbaab4bbf0af834b3
43555Author: Ville Syrjälä <ville.syrjala@nokia.com>
43556Date: Fri Oct 29 17:34:56 2010 +0300
43557
43558 xfree86/xv: Provide a ModeSet hook
43559
43560 Reput all ports when the display mode or panning has been changed by
43561 RandR code. This makes the overlays appear in the correct position
43562 on the screen.
43563
43564 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43565 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43566
43567commit 640c8716e0400e35afd7e91efc826fc447ea6745
43568Author: Ville Syrjälä <ville.syrjala@nokia.com>
43569Date: Thu Oct 28 18:13:02 2010 +0300
43570
43571 xfree86/modes: Provide a ModeSet hook in ScrnInfo
43572
43573 Add a new hook ModeSet that will be called after display mode is
43574 changed, or after the display has been panned.
43575
43576 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43577 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43578
43579commit a942534ca3908418c407115c6393263c2fe05931
43580Author: Ville Syrjälä <ville.syrjala@nokia.com>
43581Date: Tue Nov 2 16:05:13 2010 +0200
43582
43583 xfree86/xv: Pass all coordinate arguments to ReputImage
43584
43585 Pass all of the src/dst coordinates to ReputImage so that drivers
43586 don't necessarily have to do double bookkeeping.
43587
43588 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43589 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43590
43591commit 0fef4e94480f2bf1157ce5f92fcb0c7dd1585371
43592Author: Ville Syrjälä <ville.syrjala@nokia.com>
43593Date: Tue Nov 2 16:23:02 2010 +0200
43594
43595 xfree86/xv: Document VIDEO_CLIP_TO_VIEWPORT incompatibility with reput
43596
43597 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43598 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43599
43600commit 5f8ec1ade8b485f48de8c72011409219afad9dd7
43601Author: Ville Syrjälä <ville.syrjala@nokia.com>
43602Date: Tue Nov 2 16:04:36 2010 +0200
43603
43604 xfree86/xv: Add some helpful comments about ReputImage
43605
43606 Document the fact that ReputImage is used for stills as well as images.
43607
43608 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43609 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43610
43611commit b4ebde23d25bef1b891902d75b2db3aad92685b7
43612Author: Ville Syrjälä <ville.syrjala@nokia.com>
43613Date: Mon Nov 1 20:27:25 2010 +0200
43614
43615 xfree86/xv: Fix ReputImage clipping
43616
43617 PutImage/PutStill respect the GC clip, however ReputImage does not.
43618 PutImage/PutStill are supposed to be oneshot operations so ReputImage
43619 should never expand the area covered by the clip, instead it should
43620 only shrink if the window clip shrinks. So commandeer clientClip
43621 into use by ReputImage and initially make it a copy of the original
43622 GC composite clip. Whenever ReputImage needs reclipping update
43623 clientClip with the newly calculated composite clip.
43624
43625 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43626 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43627
43628commit 7294236bdb29b4fa7a7bc27aff9c786c5a33c544
43629Author: Ville Syrjälä <ville.syrjala@nokia.com>
43630Date: Mon Nov 1 20:02:40 2010 +0200
43631
43632 xfree86/xv: Remove clipOrg from XvPortRecPrivate
43633
43634 clipOrg never changes except when clientClip changes, so instead of
43635 keeping copies of both originals translate clientClip by clipOrg
43636 immediately and just keep the translated clientClip.
43637
43638 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43639 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43640
43641commit 47d8bd0a999693f47fd244dc876ef7f8fcd52854
43642Author: Ville Syrjälä <ville.syrjala@nokia.com>
43643Date: Fri Oct 29 17:31:14 2010 +0300
43644
43645 xfree86/xv: Factor out the meat of xf86XVAdjustFrame for later reuse
43646
43647 No functional change.
43648
43649 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43650 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43651
43652commit 3d4d0237a3981820a9b7290f69c7172a48659a10
43653Author: Ville Syrjälä <ville.syrjala@nokia.com>
43654Date: Fri Oct 29 17:14:34 2010 +0300
43655
43656 xfree86/xv: Change the behaviour of AdjustFrame to reput everything
43657
43658 Also reput PutVideo/GetVideo ports in AdjustFrame. This makes the
43659 overlay track the screen panning instead of staying stationary in the
43660 wrong place.
43661
43662 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43663 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43664
43665commit d794be81218934ad4580211c7a0a27cb71eb5d60
43666Author: Ville Syrjälä <ville.syrjala@nokia.com>
43667Date: Fri Oct 29 18:07:32 2010 +0300
43668
43669 xfree86/xv: No need to free composite clip in AdjustFrame
43670
43671 Nothing should change in AdjustFrame that would need the composite clip
43672 to be recomputed.
43673
43674 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43675 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43676
43677commit 6051c7e940767b7120ad9c7e9188aee6a6f62b94
43678Author: Ville Syrjälä <ville.syrjala@nokia.com>
43679Date: Fri Oct 29 17:05:02 2010 +0300
43680
43681 xfree86/xv: Change the behaviour of ClipNotify to reput instead of stop
43682
43683 When ClipNotify gets called for a visible window, reput instead of
43684 stopping the port. This eliminates nasty overlay flickering that
43685 happens during clip changes.
43686
43687 If the window is invisible or if ReputImage isn't supported stop
43688 and remove the port from the window as was done before.
43689
43690 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43691 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43692
43693commit 27707ac5bdf6831630944f02f2128951a529faab
43694Author: Ville Syrjälä <ville.syrjala@nokia.com>
43695Date: Fri Oct 29 17:00:24 2010 +0300
43696
43697 xfree86/xv: Allow xf86XVReputOrStopPort() to stop PutVideo/GetVideo ports
43698
43699 Modify xf86XVReputOrStopPort() to allow stopping of all types of ports.
43700 Will be useful later.
43701
43702 No functional change.
43703
43704 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43705 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43706
43707commit 38987aae199982f17eae664f3a28ff5941d3ee73
43708Author: Ville Syrjälä <ville.syrjala@nokia.com>
43709Date: Fri Oct 29 16:51:24 2010 +0300
43710
43711 xfree86/xv: Factor out the meat of xf86XVWindowExposures for later reuse
43712
43713 No fucntional change.
43714
43715 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43716 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43717
43718commit 0178173f0579c4b9dd0cf60b9cd3e5d264f86ae4
43719Author: Ville Syrjälä <ville.syrjala@nokia.com>
43720Date: Fri Oct 29 16:39:13 2010 +0300
43721
43722 xfree86/xv: Eliminate two open coded copies of xf86XVRemovePortFromWindow
43723
43724 No functional change.
43725
43726 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43727 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43728
43729commit b02858b5af47d524001ec571737cc92f6a6f06d3
43730Author: Ville Syrjälä <ville.syrjala@nokia.com>
43731Date: Fri Oct 29 16:54:42 2010 +0300
43732
43733 xfree86/xv: Remove useless NULL check from ClipNotify
43734
43735 WinPriv->PortRec should never be NULL as WinPriv itself would be removed
43736 from the list when the port is removed from the window.
43737
43738 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
43739 Reviewed-by: Luc Verhaegen <luc.verhaegen@basyskom.de>
43740
43741commit 23e3d1f23318ce69623f91908f888a09f8b74ac2
43742Author: Peter Hutterer <peter.hutterer@who-t.net>
43743Date: Fri Nov 26 10:00:49 2010 +1000
43744
43745 dix: remove now unnecessary !! before BitIsOn()
43746
43747 The macro has been changed to do this already, no need for double
43748 not-not-ing.
43749
43750 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43751 Reviewed-by: Keith Packard <keithp@keithp.com>
43752
43753commit 42dc91e32a02b6b21ff5c45f465f3349e5822615
43754Author: Peter Hutterer <peter.hutterer@who-t.net>
43755Date: Wed Nov 24 14:20:30 2010 +1000
43756
43757 include: let BitIsOn() return a boolean value.
43758
43759 Simply returning the mask bit breaks checks like
43760 BitIsOn(mask, 0) != BitIsOn(mask, 1);
43761 as used in 048e93593e3f7a99a7d2a219e1ce2bdc9d407807.
43762
43763 The naming of this macro suggests that it should return boolean values
43764 anyway. This patch also adds a few simple tests for these macros to make
43765 sure they don't accidentally break in the future.
43766
43767 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43768 Reviewed-by: Pat Kane <pekane52@gmail.com>
43769 Reviewed-by: Julien Cristau <jcristau@debian.org>
43770 Reviewed-by: Keith Packard <keithp@keithp.com>
43771
43772commit 4e0f8f666e61390206c42ad2087477a912525bc7
43773Merge: 400ddf2 639600f
43774Author: Keith Packard <keithp@keithp.com>
43775Date: Wed Nov 24 11:47:33 2010 -0800
43776
43777 Merge remote branch 'whot/for-keith'
43778
43779commit 400ddf2f58ce673c74d4e8ee829729cd3c75f8c9
43780Merge: 7250f07 e074f74
43781Author: Keith Packard <keithp@keithp.com>
43782Date: Wed Nov 24 11:46:18 2010 -0800
43783
43784 Merge remote branch 'jeremyhu/master'
43785
43786commit 7250f078c12fd20d5ac9150f54495926e5121461
43787Author: Gaetan Nadon <memsize@videotron.ca>
43788Date: Sun Nov 21 14:27:58 2010 -0500
43789
43790 doc: refactor Makefile and xmlrules.in code for reusability
43791
43792 A different approach which requires less variables setting
43793 and internal knowledge of the reused code.
43794 Changing from "install" to "not install" is very easy now.
43795
43796 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43797 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
43798 Signed-off-by: Keith Packard <keithp@keithp.com>
43799
43800commit f33512b70cf0350a62df2ee2287a870d5fc991a8
43801Author: Gaetan Nadon <memsize@videotron.ca>
43802Date: Sun Nov 21 14:27:57 2010 -0500
43803
43804 xmlrules.in: use $(top_srcdir) rather than ../../../ [...]
43805
43806 Relative paths don't always work in distcheck when srcdir not = builddir
43807 include $(top_srcdir)/doc/xml/xmlrules.in
43808
43809 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43810 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
43811 Signed-off-by: Keith Packard <keithp@keithp.com>
43812
43813commit c25b407f22456f50eef90d9bc5e026c05415c021
43814Author: Gaetan Nadon <memsize@videotron.ca>
43815Date: Sun Nov 21 14:27:56 2010 -0500
43816
43817 xmlrules.in: specify the xserver entities depedencies on the target
43818
43819 The generated docs will rebuild when the xserver.ent file changes.
43820
43821 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43822 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
43823 Signed-off-by: Keith Packard <keithp@keithp.com>
43824
43825commit 73841074eb856e6be3b9f9e3fc2d2721443166ab
43826Author: Gaetan Nadon <memsize@videotron.ca>
43827Date: Sun Nov 21 14:27:55 2010 -0500
43828
43829 xmlrules.in: use pattern rules to enable dependencies
43830
43831 This will allow a dependency to be specified as done in libX11:
43832
43833 %.html: %.xml $(dist_spec_DATA)
43834 $(AM_V_GEN)$(XMLTO) $(XMLTO_FLAGS) xhtml-nochunks $<
43835
43836 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43837 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
43838 Signed-off-by: Keith Packard <keithp@keithp.com>
43839
43840commit 97e307dda201d6ba28a723c21a95e484921d1b08
43841Author: Gaetan Nadon <memsize@videotron.ca>
43842Date: Sun Nov 21 14:27:54 2010 -0500
43843
43844 xmlrules.in: no need to setup xmlto flags when configuring --without-xmlto
43845
43846 The AM conditional HAVE_XMLTO should wrap more statements.
43847
43848 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43849 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
43850 Signed-off-by: Keith Packard <keithp@keithp.com>
43851
43852commit 1a7b14d1183057fc69a5da446e898104b65c8347
43853Author: Gaetan Nadon <memsize@videotron.ca>
43854Date: Sun Nov 21 14:27:53 2010 -0500
43855
43856 xmlrules.in: remove unrequired "@rm -f $@" from doc targets
43857
43858 Unable to find a purpose for this, not used anywhere else
43859 but in the font module.
43860
43861 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43862 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
43863 Signed-off-by: Keith Packard <keithp@keithp.com>
43864
43865commit d3c523bd124dc1a3f4539f7f6cd4868adbfc25e1
43866Author: Gaetan Nadon <memsize@videotron.ca>
43867Date: Sun Nov 21 14:27:52 2010 -0500
43868
43869 doc: HTML file generation: use the installed copy of xorg.css
43870
43871 Currently the xorg.css file is copied in each location
43872 where a DocBook/XML file resides. This produces about
43873 70 copies in the $(docdir) install tree for all of xorg.
43874
43875 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43876 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
43877 Signed-off-by: Keith Packard <keithp@keithp.com>
43878
43879commit 504e3010e976ccc008ff48417850b594beaee2ec
43880Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43881Date: Fri Nov 19 22:09:05 2010 -0800
43882
43883 Xorg.man: Replace XDarwin reference with Xquartz
43884
43885 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
43886 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43887 Signed-off-by: Keith Packard <keithp@keithp.com>
43888
43889commit 566d09a5cd6452ee2e05e23d2205e7c3aa31f0da
43890Author: Bill Nottingham <notting@redhat.com>
43891Date: Mon Nov 15 11:25:14 2010 +1000
43892
43893 xfree86: store the screen's gamma information on init.
43894
43895 This fixes a gamma issue on vt switch observed with KDM. VT switching away
43896 and back would result in a black screen. Avoid this by storing the current
43897 gamma information on init.
43898
43899 https://bugzilla.redhat.com/show_bug.cgi?id=533217
43900
43901 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43902 Reviewed-by: Keith Packard <keithp@keithp.com>
43903 Signed-off-by: Keith Packard <keithp@keithp.com>
43904
43905commit c050aa2f5fa6a7ff8ce4f91def14ca44f799f067
43906Author: Adam Jackson <ajax@redhat.com>
43907Date: Mon Nov 15 11:25:12 2010 +1000
43908
43909 xfree86: apply gamma settings on EnterVT.
43910
43911 When entering the VT, re-apply the saved gamma settings for each screen.
43912
43913 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43914 Reviewed-by: Keith Packard <keithp@keithp.com>
43915 Signed-off-by: Keith Packard <keithp@keithp.com>
43916
43917commit e074f745a80b87d1d82fb39489d7f35163d7552a
43918Author: Jeremy Huddleston <jeremyhu@apple.com>
43919Date: Tue Nov 23 16:03:48 2010 -0800
43920
43921 XQuartz: Remove unused CloseInputDevice stub
43922
43923 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
43924
43925commit 639600fa7ebad7a4953d0b632abeff2d339a49c6
43926Author: Peter Hutterer <peter.hutterer@who-t.net>
43927Date: Tue Nov 23 13:30:05 2010 +1000
43928
43929 dix: add a fixme about a corner-case that should probably be fixed.
43930
43931 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43932 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43933
43934commit 0aca9e8424d8b978053a29c4468f800fc2e7238c
43935Author: Peter Hutterer <peter.hutterer@who-t.net>
43936Date: Tue Nov 23 13:04:19 2010 +1000
43937
43938 dix: fix typo, set the second valuator with the y-axis data.
43939
43940 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43941 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
43942 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43943
43944commit 45131bb67f9eacb3b55de4bcd3730fb9bb028e8a
43945Author: Peter Hutterer <peter.hutterer@who-t.net>
43946Date: Tue Nov 23 12:55:50 2010 +1000
43947
43948 dix: GetProximityEvents needs to check up to the last valuator
43949
43950 valuator_mask_size() returns the highest valuator set as opposed to the
43951 number of set bits (which obviously changes as we unset valuators).
43952
43953 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43954 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43955
43956commit 048e93593e3f7a99a7d2a219e1ce2bdc9d407807
43957Author: Peter Hutterer <peter.hutterer@who-t.net>
43958Date: Tue Nov 23 11:15:52 2010 +1000
43959
43960 dix: when converting to XI 1.x events, use the first_valuator's device mode
43961
43962 Don't stop processing events when a valuator's mode doesn't match the 0th
43963 valuator's mode. Instead, start with the first_valuator in the event and
43964 keep stacking them on until the first valuator with a different mode is hit.
43965
43966 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43967 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43968
43969commit 4f3e670fd28fceec384ff0d4ca7c1ff8d7a872ad
43970Author: Peter Hutterer <peter.hutterer@who-t.net>
43971Date: Tue Nov 23 10:59:52 2010 +1000
43972
43973 dix: when comparing axis modes, compare the bits, not the bytes.
43974
43975 The DeviceEvent's mode field is a set of bits for each valuator, not bytes.
43976
43977 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43978 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43979
43980commit 6f12934d4ea52f17ac7b21e92f4f929db165f83d
43981Author: Peter Hutterer <peter.hutterer@who-t.net>
43982Date: Tue Nov 23 10:58:32 2010 +1000
43983
43984 dix: replace a manual valuator check with valuator_get_mode().
43985
43986 This check was missing the OutOfProximity mask and resulted in the wrong
43987 bits being set in InternalEvents.
43988
43989 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43990 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
43991
43992commit ebe3ddaf284ad388835da9cbaa6ad61fa21ce6dd
43993Author: Simon Thum <simon.thum@gmx.de>
43994Date: Fri Nov 19 11:09:21 2010 +1000
43995
43996 dix: fix up valuators passed to acceleration code.
43997
43998 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
43999 Reviewed-by: Simon Thum <simon.thum@gmx.de>
44000 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
44001
44002commit d435e1ecb86e2fe6292b5945262956644f979bbb
44003Author: Peter Hutterer <peter.hutterer@who-t.net>
44004Date: Fri Feb 5 11:08:27 2010 +1000
44005
44006 test: reduce range of byte-padding macro tests.
44007
44008 Byte padding and conversion is interesting for the rage of 0-8 bytes, and
44009 then interesting towards the end of the valid range (INT_MAX - 7 and INT_MAX
44010 - 3).
44011
44012 Note: this changes the upper range for pad_to_int32() and bytes_to_int32()
44013 from the previous (INT_MAX - 4) to (INT_MAX - 3).
44014
44015 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44016 Reviewed-by: Julien Cristau <jcristau@debian.org>
44017
44018commit f49ee9074a66883a3c525b0d6e71589123288465
44019Author: Peter Hutterer <peter.hutterer@who-t.net>
44020Date: Fri Feb 5 10:59:52 2010 +1000
44021
44022 test: compare byte padding macros against the expected bytes.
44023
44024 We calculate the expected bytes for each value, let's use it.
44025
44026 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44027 Reviewed-by: Julien Cristau <jcristau@debian.org>
44028
44029commit b31df0439fe336a43a2355e2f1fb223d86045a05
44030Author: Peter Hutterer <peter.hutterer@who-t.net>
44031Date: Mon Nov 22 15:14:02 2010 +1000
44032
44033 xfree86: add missing linebreak in error message.
44034
44035 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44036 Reviewed-by: Magnus Kessler <Magnus.Kessler@gmx.net>
44037
44038commit 97b9bb1e2cebbb769ef347455d9c7979d19de99f
44039Author: Jeremy Huddleston <jeremyhu@apple.com>
44040Date: Mon Nov 22 20:43:14 2010 -0800
44041
44042 XQuartz: Localization Updates
44043
44044 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44045
44046commit f9fbafa1a07f951e89f208bf4928a0bf37b67cfe
44047Author: Jeremy Huddleston <jeremyhu@apple.com>
44048Date: Mon Nov 22 20:41:39 2010 -0800
44049
44050 XQuartz: Remove existing localization files on install to avoid conflicts
44051
44052 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44053
44054commit 6eed13ad5998ba80296bba2bb3f89928ee51194a
44055Author: Jeremy Huddleston <jeremyhu@apple.com>
44056Date: Mon Nov 22 20:25:33 2010 -0800
44057
44058 XQuartz dri: Don't check CoreGraphics version
44059
44060 The version check is hackish, and I highly doubt anyone using this code is on
44061 a version of OS X that predates this support (10.2 has support for it).
44062
44063 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44064
44065commit a4698754c1da1cf2ca64add5276d0c6012960504
44066Author: Jeremy Huddleston <jeremyhu@apple.com>
44067Date: Sun Nov 21 10:49:24 2010 -0800
44068
44069 XQuartz: Add pointer debugging
44070
44071 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44072
44073commit 31737fff08ec19b394837341d5e358ec401f5cd8
44074Author: Chase Douglas <chase.douglas@ubuntu.com>
44075Date: Mon Nov 15 11:42:12 2010 -0500
44076
44077 Fix transformAbsolute
44078
44079 transformAbsolute must use old values if valuator mask doesn't have new
44080 ones, and it must only set new values if there was a change.
44081
44082 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
44083 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44084 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44085
44086commit 463841f45aafd1227bac652965c9ea337dd527fa
44087Author: Chase Douglas <chase.douglas@ubuntu.com>
44088Date: Mon Nov 15 11:42:11 2010 -0500
44089
44090 Fix GPE Y axis scaling
44091
44092 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
44093 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44094 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44095
44096commit a6b9e8f1e5d5d0b3b0f121a6f677eeca7aab1950
44097Author: Adam Jackson <ajax@redhat.com>
44098Date: Wed Aug 25 11:06:38 2010 -0400
44099
44100 linux: Fix CPU usage bug in console fd flushing
44101
44102 If the vt gets a vhangup from under us, then the tty will appear ready
44103 in select(), but trying to tcflush() it will return -EIO, so we'll spin
44104 around at 100% CPU for no reason. Notice this condition and unregister
44105 the handler if it happens.
44106
44107 Signed-off-by: Adam Jackson <ajax@redhat.com>
44108 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44109 Reviewed-by: Julien Cristau <jcristau@debian.org>
44110 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44111
44112commit ecdbe817fd7eb53fd1e7485c0492355f7f002d2a
44113Author: Dan Horák <dan@danny.cz>
44114Date: Tue Jun 22 19:45:26 2010 +0200
44115
44116 XF86CONFIGDIR is undefined when built with --disable-xorg (#28672)
44117
44118 The 10-evdev.conf file gets installed as /usr/share/X11/10-evdev.conf on
44119 platforms that built the server with --disable-xorg like s390/s390x. The
44120 definition/installation should be guarded with "if XORG" because it makes
44121 sense only when built with xorg.
44122
44123 X.Org Bug 28672 <http://bugs.freedesktop.org/show_bug.cgi?id=28672>
44124
44125 Signed-off-by: Dan Horák <dan@danny.cz>
44126 Acked-by: Julien Cristau <jcristau@debian.org>
44127 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44128
44129commit 38c46ccf796462d068d95acb9022de2acfc3a1f7
44130Merge: 3051dd5 c8998af
44131Author: Keith Packard <keithp@keithp.com>
44132Date: Mon Nov 15 09:06:33 2010 +0800
44133
44134 Merge remote branch 'whot/for-keith'
44135
44136commit 3051dd5135962283fc6c7de377a7c819d8a04e7f
44137Merge: ed547a9 c597572
44138Author: Keith Packard <keithp@keithp.com>
44139Date: Mon Nov 15 08:58:02 2010 +0800
44140
44141 Merge remote branch 'jeremyhu/master'
44142
44143commit ed547a9fc8a4f5cd95ef882b656a376c163eda6d
44144Merge: 78f94f1 84e8de1
44145Author: Keith Packard <keithp@keithp.com>
44146Date: Mon Nov 15 08:51:27 2010 +0800
44147
44148 Merge remote branch 'adkins/master'
44149
44150commit 78f94f19aab66a1e5331df0ce29f36e310b4195d
44151Author: Linus Arver <linusarver@gmail.com>
44152Date: Sun Oct 17 12:26:01 2010 -0700
44153
44154 Xext: panoramiXprocs: fix typo
44155
44156 This fixes a typo introduced in commit
44157 80b5d3a3264d2c5167e5ac85a3b04af0f89cece1. The pointer pDst was changed
44158 unintentionally to pWin from a copy/paste error. This resulted in all
44159 QT-based apps and some tcl/tk ones (like fontforge) to crash X 1.9 on
44160 starting up, when Xinerama was enabled.
44161
44162 Bug report: https://bbs.archlinux.org/viewtopic.php?id=106125
44163
44164 Signed-off-by: Elie Bleton <drozofil@gmail.com>
44165 Reviewed-by: Adam Jackson <ajax@redhat.com>
44166 Reviewed-by: Matt Turner <mattst88@gmail.com>
44167 Tested-by: Linus Arver <linusarver@gmail.com>
44168
44169commit c8998af3eba49a542625912d512087e4eaaf5ede
44170Author: Cyril Brulebois <kibi@debian.org>
44171Date: Sat Nov 13 02:05:16 2010 +0100
44172
44173 xfree86: Initialize Pointer and Keyboard.
44174
44175 That helps us get rid of:
44176 | CC xf86Config.lo
44177 | xf86Config.c: In function ‘T.174’:
44178 | xf86Config.c:1228: warning: ‘Pointer$module’ may be used uninitialized in this function
44179 | xf86Config.c:1097: note: ‘Pointer$module’ was declared here
44180 | xf86Config.c:1228: warning: ‘Pointer$drv’ may be used uninitialized in this function
44181 | xf86Config.c:1097: note: ‘Pointer$drv’ was declared here
44182 | xf86Config.c:1228: warning: ‘Pointer$type_name’ may be used uninitialized in this function
44183 | xf86Config.c:1097: note: ‘Pointer$type_name’ was declared here
44184 | xf86Config.c:1228: warning: ‘Pointer$private’ may be used uninitialized in this function
44185 | xf86Config.c:1097: note: ‘Pointer$private’ was declared here
44186 | xf86Config.c:1228: warning: ‘Pointer$dev’ may be used uninitialized in this function
44187 | xf86Config.c:1097: note: ‘Pointer$dev’ was declared here
44188 | xf86Config.c:1228: warning: ‘Pointer$fd’ may be used uninitialized in this function
44189 | xf86Config.c:1097: note: ‘Pointer$fd’ was declared here
44190 | xf86Config.c:1228: warning: ‘Pointer$set_device_valuators’ may be used uninitialized in this function
44191 | xf86Config.c:1097: note: ‘Pointer$set_device_valuators’ was declared here
44192 | xf86Config.c:1228: warning: ‘Pointer$switch_mode’ may be used uninitialized in this function
44193 | xf86Config.c:1097: note: ‘Pointer$switch_mode’ was declared here
44194 | xf86Config.c:1228: warning: ‘Pointer$control_proc’ may be used uninitialized in this function
44195 | xf86Config.c:1097: note: ‘Pointer$control_proc’ was declared here
44196 | xf86Config.c:1228: warning: ‘Pointer$read_input’ may be used uninitialized in this function
44197 | xf86Config.c:1097: note: ‘Pointer$read_input’ was declared here
44198 | xf86Config.c:1228: warning: ‘Pointer$device_control’ may be used uninitialized in this function
44199 | xf86Config.c:1097: note: ‘Pointer$device_control’ was declared here
44200 | xf86Config.c:1228: warning: ‘Pointer$flags’ may be used uninitialized in this function
44201 | xf86Config.c:1097: note: ‘Pointer$flags’ was declared here
44202 | xf86Config.c:1228: warning: ‘Pointer$next’ may be used uninitialized in this function
44203 | xf86Config.c:1097: note: ‘Pointer$next’ was declared here
44204
44205 Signed-off-by: Cyril Brulebois <kibi@debian.org>
44206 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44207
44208commit 98abac809a024ac3ed9dd0514ca96367f668c05d
44209Author: Jon TURNEY <jon.turney@dronecode.org.uk>
44210Date: Thu Nov 11 23:06:13 2010 +0000
44211
44212 Fix compilation after input API change
44213
44214 Fix compilation after commit 675f4a8525d29ebad783351e17be785b2f32b2e8
44215 converts winmouse.c to the new ValuatorMask input API, but doesn't
44216 include the required header inpututil.h
44217
44218 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
44219 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44220
44221commit c5975722a95c41b8715a4612fc1c3275f8e9b59e
44222Author: Jeremy Huddleston <jeremyhu@apple.com>
44223Date: Sat Nov 13 16:04:10 2010 -0800
44224
44225 XQuartz: Fix make distcheck
44226
44227 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44228 Reviewed-by: Cyril Brulebois <kibi@debian.org>
44229
44230commit 7b33ef74738d6a557cb25aeb14693b2a340c0e12
44231Author: Jeremy Huddleston <jeremyhu@apple.com>
44232Date: Sat Nov 13 15:46:16 2010 -0800
44233
44234 dix: Fix make distcheck
44235
44236 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44237 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
44238
44239commit 965e709f2b0b17f1e59e5aeb5e7717fede51ef97
44240Author: Jeremy Huddleston <jeremyhu@apple.com>
44241Date: Thu Nov 11 17:23:33 2010 -0800
44242
44243 XQuartz: Build-fixes for XQuartz for recent input API changes
44244
44245 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44246 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44247
44248commit fa50670c32637faff6bb91f1206a0e3c17a9bada
44249Author: Peter Hutterer <peter.hutterer@who-t.net>
44250Date: Fri Nov 12 11:48:08 2010 +1000
44251
44252 xfree86: rename allowEmptyInput to forceInputDevices.
44253
44254 This is a more accurate name for the actual functionality than
44255 allowEmptyInput. Historically, allowEmptyInput has allowed the server to
44256 start with no input devices. Since 1.4 and the introduction of VCP and VCK,
44257 there are always two input devices present.
44258
44259 allowEmptyInput was changed in behaviour to essentially "ignore xorg.conf
44260 devices or not", auto-adding the built-in devices if disabled.
44261
44262 Rename to forceInputDevices, because that's essentially what it does. When
44263 disabled (i.e. when hotplugging is enabled), it disables all
44264 mouse/kbd/vmmouse devices configured in the xorg.conf file.
44265 When enabled, it forces the traditional behaviour for input devices:
44266 - use input devices configured in the server layout
44267 - if none are configured, use the first pointer and the first keyboard
44268 device in the xorg.conf
44269 - if none are configured, create the default pointer/keyboard devices.
44270
44271 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44272 Reviewed-by: Julien Cristau <jcristau@debian.org>
44273
44274commit c481dae0dc89d8d50cac23210023d75bfaf6012b
44275Author: Cyril Brulebois <kibi@debian.org>
44276Date: Thu Nov 11 23:55:55 2010 +0100
44277
44278 test: Fix missing xkbsrv.h include.
44279
44280 Otherwise, building fails with CFLAGS="-Wall -Werror" this way:
44281 | protocol-common.c: In function ‘init_simple’:
44282 | protocol-common.c:159: error: implicit declaration of function ‘XkbInitPrivates’
44283 | protocol-common.c:159: error: nested extern declaration of ‘XkbInitPrivates’
44284
44285 Signed-off-by: Cyril Brulebois <kibi@debian.org>
44286 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44287
44288commit b8114f25b266624c8f73d03c710349f98693d877
44289Author: Peter Hutterer <peter.hutterer@who-t.net>
44290Date: Fri Nov 12 08:32:05 2010 +1000
44291
44292 dix: fix typo in setting valuator modes (#21548)
44293
44294 X.Org Bug 31548 <http://bugs.freedesktop.org/show_bug.cgi?id=31548>
44295
44296 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44297 Tested-by: Cyril Brulebois <kibi@debian.org>
44298
44299commit 5cd11d2356d153840f1b429bdb8284367f8dc468
44300Author: Peter Hutterer <peter.hutterer@who-t.net>
44301Date: Wed Jun 9 17:10:57 2010 +1000
44302
44303 Xi: move property reset from extension shutdown to init.
44304
44305 If any part of the stack calls XIGetKnownProperty during device shutdown
44306 the property is re-initialized before the server generation resets, leaving
44307 the value invalid again.
44308
44309 Move the reset to the extension init which happens before input devices are
44310 initialized before the first property is requested.
44311
44312 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44313 Reviewed-by: Julien Cristau <jcristau@debian.org>
44314
44315commit da66119593b63ef82ae6fa2e034ccd0856b03a80
44316Author: Tiago Vignatti <tiago.vignatti@nokia.com>
44317Date: Wed Nov 10 21:48:04 2010 +0200
44318
44319 dix: fix root window background behaviour for protocol calls
44320
44321 Instead always paint root tiled (-retro like), protocol calls
44322 (XSetWindowBackgroundPixmap and related) should behave accordingly when None
44323 and ParentRelative is set as background pixmap.
44324
44325 It follow what the protocol states: "changing the background of a root window
44326 to None or ParentRelative restores the default background pixmap".
44327
44328 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
44329 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
44330 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
44331
44332commit 8976e9766edfb33f5cf3f6d54f09e46d29bc4ec5
44333Author: Tiago Vignatti <tiago.vignatti@nokia.com>
44334Date: Wed Nov 10 21:48:03 2010 +0200
44335
44336 dix: adds support for none root window background
44337
44338 It lets the driver notify the server whether it can draw a background when
44339 '-background none' option is used by the system platform. Use cases for that
44340 could be video drivers performing mode-setting in kernel time, before X is up,
44341 so a seamless transition would happen until X clients start to show up.
44342
44343 If the driver can copy the framebuffer cleanly then it can set the flag
44344 (canDoBGNoneRoot), otherwise the server will fallback to the normal behaviour.
44345 The system must explicit indicates willingness of doing so through
44346 '-background none'. We could do this option as default; in such case,
44347 malicious users would be able to steal the framebuffer with a bit of tricks.
44348
44349 For instance, I can see the content of my nVidia Quadro FX 580 framebuffer
44350 old X session modifying a bit nv driver:
44351
44352 xf86DPMSInit(pScreen, xf86DPMSSet, 0);
44353
44354 - /* Clear the screen */
44355 - if(pNv->xaa) {
44356 - /* Use the acceleration engine */
44357 - pNv->xaa->SetupForSolidFill(pScrn, 0, GXcopy, ~0);
44358 - pNv->xaa->SubsequentSolidFillRect(pScrn,
44359 - 0, 0, pScrn->displayWidth, pNv->offscreenHeight);
44360 - G80DmaKickoff(pNv);
44361 - } else {
44362 - /* Use a slow software clear path */
44363 - memset(pNv->mem, 0, pitch * pNv->offscreenHeight);
44364 - }
44365 + pScreen->canDoBGNoneRoot = TRUE;
44366
44367 The commit is originally based on discussions happened on xorg-devel:
44368 http://lists.freedesktop.org/archives/xorg-devel/2010-June/009755.html
44369
44370 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
44371 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
44372 Acked-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
44373 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
44374
44375commit bfe9cdd11bcb60cf33dc48136ebee028001a737e
44376Author: Tiago Vignatti <tiago.vignatti@nokia.com>
44377Date: Wed Oct 20 17:54:58 2010 -0200
44378
44379 dix: delete logo hack screen saver
44380
44381 Protocol doesn't mention about screen saver with logo being required and
44382 people are already using more intelligent ways to draw screen saver themes. So
44383 consider -logo as deprecated option, deleting its code.
44384
44385 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
44386 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44387 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
44388 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
44389
44390commit 84e8de1271bb11b5b4b9747ae4647f47333a8ab7
44391Author: Jesse Adkins <jesserayadkins@gmail.com>
44392Date: Wed Nov 10 21:37:24 2010 -0800
44393
44394 fb: Remove fbcmap.c (bug 5436)
44395
44396 This is obsolete since a240c039c47c0be22ea5e100692307b26d938747.
44397 Updated fb.h to mention that the functions come from fbcmap_mi.c now.
44398 Dropped fbcmap.c from the makefile.
44399
44400 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
44401 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
44402
44403commit 23e329b5641ef044c34ac712c9193171eaf47198
44404Author: Jesse Adkins <jesserayadkins@gmail.com>
44405Date: Wed Nov 10 21:35:58 2010 -0800
44406
44407 xfree86: parser: Remove 'CUSTOM' flag option in Monitor section.
44408
44409 Not used in the initial import, and also not documented.
44410
44411 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
44412 Reviewed-by: Adam Jackson <ajax@redhat.com>
44413
44414commit 88cb61e1e55c54982b90e2a77465faaac6a0ba89
44415Merge: 19f4383 4ed4915
44416Author: Peter Hutterer <peter.hutterer@who-t.net>
44417Date: Thu Nov 11 12:54:46 2010 +1000
44418
44419 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-api
44420
44421 Conflicts:
44422 dix/getevents.c
44423 hw/xfree86/common/xf86Xinput.h
44424
44425 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44426
44427commit 4ed4915bc0fbe9b4419ca4a0d4e43c2fee032ae9
44428Merge: 291c39d ec1bfbc
44429Author: Keith Packard <keithp@keithp.com>
44430Date: Wed Nov 10 16:58:21 2010 -0800
44431
44432 Merge remote branch 'whot/for-keith'
44433
44434commit ec1bfbc66926130e1153facc3b92ee175f1cb6b6
44435Author: Peter Hutterer <peter.hutterer@who-t.net>
44436Date: Tue Nov 9 15:27:26 2010 +1000
44437
44438 xfree86: remove user-configured AllowEmptyInput
44439
44440 An estimated 100% (rounded down to the nearest percent) of the people who
44441 have this in their configuration don't actually know what this option does.
44442 Protect the users from themselves.
44443
44444 IIRC, AEI on was useful for some time between 1.4 and 1.5 and never since.
44445
44446 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44447 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
44448
44449commit cbaa6a66e0f9e3b7e305606924ecda0147b59e96
44450Author: Peter Hutterer <peter.hutterer@who-t.net>
44451Date: Tue Nov 9 15:21:23 2010 +1000
44452
44453 config: remove mention of AllowEmptyInput
44454
44455 "You will probably want to add the following option to the ServerFlags of
44456 your xorg.conf:
44457 Option "AllowEmptyInput" "True""
44458
44459 I can't imagine why you would want to do that. My life is painful enough
44460 already.
44461
44462 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44463 Reviewed-by: Julien Cristau <jcristau@debian.org>
44464 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
44465
44466commit 291c39dfef84d1b204d7f5e63e925dee2dbdaca7
44467Author: Jeremy Huddleston <jeremyhu@apple.com>
44468Date: Tue Nov 9 13:33:55 2010 -0800
44469
44470 DOC: Only build dtrace documentation with --enable-docs
44471
44472 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
44473 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
44474 Signed-off-by: Keith Packard <keithp@keithp.com>
44475
44476commit f0b4df99486066ad0e638b3b5debc905bc8ef1c4
44477Author: Adam Jackson <ajax@redhat.com>
44478Date: Sat Sep 11 17:55:57 2010 -0400
44479
44480 shadow: Optimize shadowUpdatePacked(). (#26973)
44481
44482 Signed-off-by: Matt Turner <mattst88@gmail.com>
44483 Reviewed-by: Soren Sandmann <sandmann@daimi.au.dk>
44484 Reviewed-by: Matt Turner <mattst88@gmail.com>
44485 Signed-off-by: Keith Packard <keithp@keithp.com>
44486
44487commit c9c0f93f8ed42413a05193ee71c69fa6b41f6a03
44488Merge: 73f6de1 4132b1c
44489Author: Keith Packard <keithp@keithp.com>
44490Date: Wed Nov 10 16:11:29 2010 -0800
44491
44492 Merge remote branch 'whot/for-keith'
44493
44494commit 73f6de1ad8b5e1641102b54ff48a36ebb99ccacb
44495Merge: 333b6ed 3dc3aef
44496Author: Keith Packard <keithp@keithp.com>
44497Date: Wed Nov 10 16:02:22 2010 -0800
44498
44499 Merge remote branch 'sandmann/fbdelete'
44500
44501commit 333b6ed26e36735ef72f314a3f8c4ec565af223f
44502Author: Kristian Høgsberg <krh@bitplanet.net>
44503Date: Thu Oct 28 20:46:23 2010 -0700
44504
44505 Set DamageSetReportAfterOp to true for the damage extension
44506
44507 Change the damage extension reporter to queue up events after we chain
44508 to the wrapped functions. Damage events are typically sent out after
44509 the rendering happens anyway, since we submit batch buffers from the
44510 flush callback chain and then flush client io buffers. Compositing
44511 managers relie on this order, and there is no way we could reliably
44512 provide damage events to clients before the rendering happens anyway.
44513
44514 By queueing up the damage events before the rendering happens, there's
44515 a risk that the client io buffer may overflow and send the damage
44516 events to the client before the driver has even seen the rendering
44517 request. Reporting damage events after the rendering fixes this
44518 corner case and better corresponds with how we expect this to work.
44519
44520 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
44521 Reviewed-by: Keith Packard <keithp@keithp.com>
44522 (cherry picked from commit 8d7b7a0d71e0b89321b3341b781bc8845386def6)
44523 [anholt: re-applied to revert the revert, now that the cause of the
44524 revert is fixed]
44525 Signed-off-by: Keith Packard <keithp@keithp.com>
44526
44527commit f36153e3ef6b13a87d016caab09cc9be274b0dd5
44528Author: Eric Anholt <eric@anholt.net>
44529Date: Thu Oct 28 20:46:22 2010 -0700
44530
44531 Replace usage of DamageRegionAppend with DamageDamageRegion to fix reportAfter.
44532
44533 In all these cases, any rendering implied by this damage has already
44534 occurred, and we want to get the damage out to the client. Some of
44535 the DamageRegionAppend calls were explicitly telling damage to flush
44536 the reportAfter damage out, but not all.
44537
44538 Bug #30260. Fixes the compiz wallpaper plugin with client damage
44539 changed to reportAfter.
44540
44541 Signed-off-by: Eric Anholt <eric@anholt.net>
44542 Reviewed-by: Keith Packard <keithp@keithp.com>
44543 Signed-off-by: Keith Packard <keithp@keithp.com>
44544
44545commit c80c41767eb101e9dbd8393d8cca7764b4e248a4
44546Author: Aaron Plattner <aplattner@nvidia.com>
44547Date: Mon Oct 25 22:01:32 2010 -0700
44548
44549 os: Fix BigReq ignoring when another request is pending
44550
44551 Commit cf88363db0ebb42df7cc286b85d30d7898aea840 fixed the handling of
44552 BigReq requests that are way too large and handles the case where the
44553 read() syscall returns a short read. However, it neglected to handle
44554 the case where it returns a long read, which happens when the client
44555 has another request in the queue after the bogus large one.
44556
44557 Handle the long read case by subtracting the smaller of 'needed' and
44558 'gotnow' from oci->ignoreBytes. If needed < gotnow, simply subtract
44559 the two, leaving gotnow equal to the number of extra bytes read.
44560 Since the code immediately following the (oci->ignoreBytes > 0) block
44561 tries to handle the next request, advance oci->bufptr immediately
44562 instead of setting oci->lenLastReq and letting the next call to
44563 ReadRequestFromClient do it.
44564
44565 Fixes the XTS pChangeKeyboardMapping-3 test.
44566
44567 CASES TESTS PASS UNSUP UNTST NOTIU WARN FIP FAIL UNRES UNIN ABORT
44568 -Xproto 122 389 367 2 19 0 0 0 1 0 0 0
44569 +Xproto 122 389 368 2 19 0 0 0 0 0 0 0
44570
44571 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
44572 Reviewed-by: Adam Jackson <ajax@redhat.com>
44573 Signed-off-by: Keith Packard <keithp@keithp.com>
44574
44575commit cc2c73ddcb4370a7c3ad439cda4da825156c26c9
44576Author: Luc Verhaegen <libv@skynet.be>
44577Date: Sun Oct 24 23:57:06 2010 +0200
44578
44579 x86emu: fix jump_near_IMM to handle DATA: flag correctly.
44580
44581 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=24348
44582
44583 Before (data flag ignored -> broken):
44584 66 DATA:
44585 e944f1 JMP 1ff6
44586
44587 After (fixed):
44588 66 DATA:
44589 e944f1ffff JMP 00001ff8
44590
44591 This subtle difference in the length of decoded instruction meant
44592 that the VBE call jumped to the routine setting AX=0x14F (VBE Failed)
44593 instead of the routine that set AX=0x4F (VBE success).
44594
44595 The ability to run the same code in vm86 significantly aided the
44596 debugging of this issue. Those X.org developers who would like to drop
44597 vm86 better take special care towards _all_ vesa bugs, as those will
44598 expose further issues.
44599
44600 Patch applies easily to even xserver 1.4.2.
44601
44602 Signed-off-by: Luc Verhaegen <libv@skynet.be>
44603 Tested-by: Luc Verhaegen <libv@skynet.be>
44604 Reviewed-by: Adam Jackson <ajax@redhat.com>
44605 Signed-off-by: Keith Packard <keithp@keithp.com>
44606
44607commit be7cf14c365c8ee0d69c4335e01316bcfcba69a4
44608Author: Siddhesh Poyarekar <siddhesh.poyarekar@gmail.com>
44609Date: Tue Oct 26 09:11:53 2010 +0530
44610
44611 mi: Integer overflow for dashed lines longer than 46340. #31093
44612
44613 Lines of length greater than 46340 can be drawn with one of the
44614 coordinates being negative. However for dashed lines, miPolyBuildPoly
44615 overflows the int type when setting up edges for a section of the
44616 dashed line. This results in the dashed segments not being drawn at
44617 all.
44618
44619 Signed-off-by: Siddhesh Poyarekar <siddhesh.poyarekar@gmail.com>
44620 Reviewed-by: Keith Packard <keithp@keithp.com>
44621 Signed-off-by: Keith Packard <keithp@keithp.com>
44622
44623commit a6c64d96456d794b2f3ce40915922a3fba636b95
44624Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
44625Date: Sun Oct 24 15:18:28 2010 +0200
44626
44627 Do not trap access to timer and keyboard
44628
44629 Disable timer/keyboard trapping on GNU/Hurd for now
44630
44631 Trapping disabled for now, as some VBIOSes (mga-g450 notably) use these
44632 ports, and the int10 wrapper is not emulating them.
44633
44634 It's effectively what happens in the Linux variant too, as iopl() is used there,
44635 making the ioperm() meaningless.
44636
44637 Signed-off-by: Olaf Buddenhagen <antrik@users.sf.net>
44638 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
44639 Reviewed-by: Adam Jackson <ajax@redhat.com>
44640 Signed-off-by: Keith Packard <keithp@keithp.com>
44641
44642commit 69e8e1b0b95a325da3e3a2d76d092e7131baa9ad
44643Author: Julien Cristau <jcristau@debian.org>
44644Date: Mon May 17 19:39:54 2010 +0200
44645
44646 os: include dix-config.h, not xorg-config.h
44647
44648 os/strlc{at,py}.c were trying to include xorg-config.h, which is not
44649 available in dix.
44650
44651 Signed-off-by: Julien Cristau <jcristau@debian.org>
44652 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
44653 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44654
44655commit acba00c0c1129fe4ef7655f018cd21dbb1ee9a1f
44656Merge: 383dfe2 44e8165
44657Author: Keith Packard <keithp@keithp.com>
44658Date: Wed Nov 10 14:41:08 2010 -0800
44659
44660 Merge remote branch 'dottedmag/master'
44661
44662commit 383dfe23f1641435c0713f8a56424814a156af8e
44663Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
44664Date: Mon Oct 25 17:13:57 2010 +0300
44665
44666 DRI2: Avoid call to NULL pointer
44667
44668 DDX driver may implement schedule swap without GetMSC. In that case we
44669 can't call GetMSC in DRI2SwapBuffers.
44670
44671 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
44672 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
44673 CC: Keith Packard <keithp@keithp.com>
44674 CC: Kristian Høgsberg <krh@bitplanet.net>
44675 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
44676 Signed-off-by: Keith Packard <keithp@keithp.com>
44677
44678commit 68e4a628d65312df93cc71f3e76241584c4bbd23
44679Author: Cyril Brulebois <kibi@debian.org>
44680Date: Wed Nov 10 15:28:29 2010 +0100
44681
44682 Remove more superfluous if(p!=NULL) checks around free(p).
44683
44684 This patch has been generated by the following Coccinelle semantic patch:
44685
44686 @@
44687 expression E;
44688 @@
44689 - if (E != NULL) {
44690 - free(E);
44691 - }
44692 + free(E);
44693
44694 Signed-off-by: Cyril Brulebois <kibi@debian.org>
44695 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
44696 Reviewed-by: Matt Turner <mattst88@gmail.com>
44697 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44698 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44699
44700commit b142b0d27442191d628762604f4eb1f65263d717
44701Author: Cyril Brulebois <kibi@debian.org>
44702Date: Mon Nov 8 23:35:33 2010 +0100
44703
44704 Remove more superfluous if(p!=NULL) checks around free(p).
44705
44706 This patch has been generated by the following Coccinelle semantic patch:
44707
44708 @@
44709 expression E;
44710 @@
44711 - if (E != NULL)
44712 - free(E);
44713 + free(E);
44714
44715 Signed-off-by: Cyril Brulebois <kibi@debian.org>
44716 Reviewed-by: Matt Turner <mattst88@gmail.com>
44717 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44718 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44719
44720commit 0649ac0afdb3f8ed11c2634563d6b9df161cb9ec
44721Author: Cyril Brulebois <kibi@debian.org>
44722Date: Mon Nov 8 23:35:32 2010 +0100
44723
44724 Remove more superfluous if(p) checks around free(p).
44725
44726 This patch has been generated by the following Coccinelle semantic patch:
44727
44728 @@
44729 expression E;
44730 @@
44731 - if (E)
44732 - free(E);
44733 + free(E);
44734
44735 Signed-off-by: Cyril Brulebois <kibi@debian.org>
44736 Reviewed-by: Matt Turner <mattst88@gmail.com>
44737 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44738 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44739
44740commit 99275ad2fa99778afaefc54b62c8638afc59e755
44741Author: Cyril Brulebois <kibi@debian.org>
44742Date: Wed Nov 10 16:06:10 2010 +0100
44743
44744 Remove superfluous if(p!=NULL) checks around free(p); p=NULL;
44745
44746 This patch has been generated by the following Coccinelle semantic patch:
44747
44748 @@
44749 expression E;
44750 @@
44751 - if (E != NULL) {
44752 - free(E);
44753 (
44754 - E = NULL;
44755 |
44756 - E = 0;
44757 )
44758 - }
44759 + free(E);
44760 + E = NULL;
44761
44762 Signed-off-by: Cyril Brulebois <kibi@debian.org>
44763 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44764 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44765 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44766
44767commit 9e999d18b004b8ead9c6c5d79b4a3d4bbf0e3152
44768Author: Peter Hutterer <peter.hutterer@who-t.net>
44769Date: Fri Nov 5 12:08:11 2010 +1000
44770
44771 Xi: if XTEST device creation fails, fail the master devices.
44772
44773 When getting close to the MAXDEVICES limit, the creation of XTEST devices
44774 may fail due to device id exhaustion. In that case, fail the creation of
44775 master devices too and return an error to the client.
44776
44777 Theoretically, we could alloc the MDs without the XTEST devices but that
44778 will get interesting when a client starts sending XTEST events through those
44779 devices.
44780
44781 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44782 Reviewed-by: Julien Cristau <jcristau@debian.org>
44783
44784commit 5a455e0c80d433adc4109ebf313fd92afa194545
44785Author: Peter Hutterer <peter.hutterer@who-t.net>
44786Date: Fri Nov 5 11:49:12 2010 +1000
44787
44788 Xi: rename two variables from ptr to dev.
44789
44790 They were named ptr when everything was in one function to save one more
44791 variable. Now that the stuff is split out, "dev" makes more sense.
44792
44793 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44794 Reviewed-by: Julien Cristau <jcristau@debian.org>
44795
44796commit 9b89b91c685426c9944f7fc8890f436c18b7583c
44797Author: Peter Hutterer <peter.hutterer@who-t.net>
44798Date: Fri Nov 5 11:47:43 2010 +1000
44799
44800 Xi: split hierarchy manipulation into static functions.
44801
44802 No functional changes, just code cleanup to improve readability.
44803
44804 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44805 Reviewed-by: Julien Cristau <jcristau@debian.org>
44806
44807commit 4132b1c591bd01c29df91b88afbd144647da9ed8
44808Author: Peter Hutterer <peter.hutterer@who-t.net>
44809Date: Thu Oct 21 14:04:42 2010 +1000
44810
44811 xfree86: fix compiler warnings - unused variable i
44812
44813 ../../../../hw/xfree86/common/xf86Config.c: In function 'configDRI':
44814 ../../../../hw/xfree86/common/xf86Config.c:2213:9: warning: unused variable
44815 'i'
44816
44817 Introduced in 788bfbf18ac19923604b0d676933fdc0d78526a1
44818
44819 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44820 Reviewed-by: Matt Turner <mattst88@gmail.com>
44821
44822commit 6b982d73bb8878238ed46665b220a94a591e11a9
44823Author: Peter Hutterer <peter.hutterer@who-t.net>
44824Date: Thu Oct 21 13:53:35 2010 +1000
44825
44826 xfree86: fix two compiler warnings - unused variable ati, atimisc
44827
44828 ../../../../hw/xfree86/common/xf86Config.c: In function
44829 'fixup_video_driver_list':
44830 ../../../../hw/xfree86/common/xf86Config.c:507:19: warning: unused variable
44831 'atimisc'
44832 ../../../../hw/xfree86/common/xf86Config.c:507:12: warning: unused variable
44833 'ati'
44834
44835 Introduced in 52577ae8eeda64601db8dd425027cf8e4271b873.
44836
44837 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44838 Reviewed-by: Matt Turner <mattst88@gmail.com>
44839
44840commit 0799f0771cca335d6400a8494ba236a0d10e122b
44841Author: Peter Hutterer <peter.hutterer@who-t.net>
44842Date: Thu Nov 4 12:53:18 2010 +1000
44843
44844 xfree86: remove two unused variables from Makefile.am
44845
44846 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44847 Reviewed-by: Julien Cristau <jcristau@debian.org>
44848
44849commit a52efb096e166e325deb3d6b502671f339a4fa15
44850Merge: fe8d122 f72aadd
44851Author: Keith Packard <keithp@keithp.com>
44852Date: Wed Nov 3 06:50:45 2010 -0700
44853
44854 Merge remote branch 'sthibaul/master-iopl'
44855
44856commit 3dc3aefb88c167d365795dcd44017742c52e4967
44857Author: Søren Sandmann Pedersen <ssp@redhat.com>
44858Date: Mon Nov 1 22:23:05 2010 -0400
44859
44860 fb: Delete a bunch more left-overs
44861
44862 All of these definitions were unused since compositing moved to pixman.
44863
44864 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
44865 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44866 Signed-off-by: Søren Sandmann <ssp@redhat.com>
44867
44868commit 565b5e2b0261d1148cc2707ffcbb915414dd3e02
44869Author: Søren Sandmann Pedersen <ssp@redhat.com>
44870Date: Mon Nov 1 22:14:29 2010 -0400
44871
44872 wfb: Delete unused renamings
44873
44874 These functions haven't existed in a while.
44875
44876 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
44877 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44878 Signed-off-by: Søren Sandmann <ssp@redhat.com>
44879
44880commit 0dae479e16590e767df35985150c5acf696e5df2
44881Author: Søren Sandmann Pedersen <ssp@redhat.com>
44882Date: Mon Nov 1 22:17:08 2010 -0400
44883
44884 fb: Delete fbCompositeGeneral()
44885
44886 This function was an unused and trivial wrapper around fbComposite().
44887
44888 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
44889 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44890 Signed-off-by: Søren Sandmann <ssp@redhat.com>
44891
44892commit 74bc2d8c6bafc5373ff9c14eed12e26b5b965953
44893Author: Søren Sandmann Pedersen <ssp@redhat.com>
44894Date: Mon Nov 1 22:12:03 2010 -0400
44895
44896 render: Delete renderedge.[ch]
44897
44898 The functions in these files have not been used since trap
44899 rasterization was moved to pixman. They survived until now to preserve
44900 the server abi.
44901
44902 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
44903 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44904 Signed-off-by: Søren Sandmann <ssp@redhat.com>
44905
44906commit 6118346d64e3c2fbe1fe2f041ea773dd2a3c0438
44907Author: Søren Sandmann Pedersen <ssp@redhat.com>
44908Date: Mon Nov 1 22:04:49 2010 -0400
44909
44910 Delete unused fbWalkCompositeRegion()
44911
44912 This function has not been used since most of the compositing was
44913 moved to pixman. The only reason it has survived until now is that it
44914 was part of the server ABI.
44915
44916 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
44917 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44918 Signed-off-by: Søren Sandmann <ssp@redhat.com>
44919
44920commit fe8d122b133a43125cc961d2c3c811fa776f29c8
44921Author: Peter Hutterer <peter.hutterer@who-t.net>
44922Date: Thu Oct 28 16:13:09 2010 +1000
44923
44924 Xi: reshuffle conditions for labeling a device as IsXExtensionKeyboard (#29046)
44925
44926 From the original bug reporter Ezra Reeves:
44927
44928 "I did some more digging on this today, and I found that an HP branded
44929 wireless USB mouse has the same issue. With this mouse (as well as the
44930 logitech wireless mouse), the return from:
44931
44932 xdev = XListInputDevices(GDK_WINDOW_XDISPLAY(rootwin), &ndevices_return);
44933
44934 lists the USB device twice, but both have xdev[num].use == 3
44935 (IsXExtensionKeyboard as defined in X11/XI.h).
44936
44937 [...]
44938
44939 Swapping the order of the test in Xi/listdev.c that determines whether a
44940 device is a pointer or a keyboard properly detects my devices (OEM USB
44941 wireless mouse/kb combo) -- one as a keyboard and one as a pointer."
44942
44943 X.Org Bug 29046 <http://bugs.freedesktop.org/show_bug.cgi?id=29046>
44944
44945 Reported-by: Erik Kilfoil <ekilfoil@gmail.com>
44946 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44947 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44948
44949commit a210068c5222acbdb04db17cb8cef314735bf977
44950Author: Tiago Vignatti <tiago.vignatti@nokia.com>
44951Date: Tue Oct 26 20:50:13 2010 +0300
44952
44953 dix: advance parent window pointer when no node is found
44954
44955 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
44956 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
44957 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
44958 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44959
44960commit 19f43836d1006d253391926e880b14ed2e7a0cf8
44961Author: Peter Hutterer <peter.hutterer@who-t.net>
44962Date: Mon Oct 25 15:10:19 2010 +1000
44963
44964 Re-export xf86CollectInputOptions.
44965
44966 Some drivers, most notably the mouse driver need this and reimplementing on
44967 the driver side doesn't make sense.
44968
44969 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44970
44971commit 4613ce3f78a32f2a6f422e719150d55b035c758e
44972Author: Peter Hutterer <peter.hutterer@who-t.net>
44973Date: Fri Oct 22 15:17:50 2010 +1000
44974
44975 test: valuator_mode tests.
44976
44977 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44978 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
44979
44980commit 58554f1c6429535e7ab432c920aabca71a77edb8
44981Author: Peter Hutterer <peter.hutterer@who-t.net>
44982Date: Fri Oct 22 15:10:32 2010 +1000
44983
44984 Convert some leftover axes->mode access to valuator_get_mode()
44985
44986 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
44987 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
44988
44989commit 44e81654147065b7b907ce0d14fdc2e74a638f79
44990Author: Mikhail Gusarov <dottedmag@dottedmag.net>
44991Date: Mon Oct 25 01:30:33 2010 +0400
44992
44993 Remove now-misleading comment
44994
44995 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
44996 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
44997
44998commit 433dddcf29d257310e95e4215dceef8b0b976e2d
44999Author: Mikhail Gusarov <dottedmag@dottedmag.net>
45000Date: Mon Oct 25 01:28:38 2010 +0400
45001
45002 Replace "if(buf) realloc(buf, size) else malloc(size)" with realloc()
45003
45004 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
45005 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
45006
45007commit 04a04e162886ca9bc3977d3a7509def876487921
45008Author: Mikhail Gusarov <dottedmag@dottedmag.net>
45009Date: Mon Oct 25 01:27:46 2010 +0400
45010
45011 Remove now-misleading comment
45012
45013 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
45014 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
45015
45016commit f72aadd38b730c46c0a027dacbbb57f03068383d
45017Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
45018Date: Sat Oct 23 22:42:33 2010 +0200
45019
45020 hurd: Fix use of deprecated iopl device
45021
45022 Fix Xserver on GNU/Hurd into using the "mem" device instead of
45023 the deprecated "iopl" device.
45024
45025 Reviewed-by: Olaf Buddenhagen <antrik@users.sf.net>
45026 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
45027
45028commit dd11f734a9a73bf34357c149d2809842938d57b5
45029Author: Peter Hutterer <peter.hutterer@who-t.net>
45030Date: Fri Oct 22 16:11:24 2010 +1000
45031
45032 input: remove "mode" field from ValuatorClassRec.
45033
45034 We have per-axis mode now. For those bits that still need it (XI 1.x),
45035 assume that the first axis holds the device's mode.
45036
45037 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45038 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45039
45040commit 5cf3b654fccf09de63a3f983c85096cda78fb693
45041Author: Peter Hutterer <peter.hutterer@who-t.net>
45042Date: Fri Oct 22 15:49:40 2010 +1000
45043
45044 input: move proximity state into ProximityClassRec.
45045
45046 Previously the OutOfProximity bit in the valuator mode.
45047
45048 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45049 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45050
45051commit e909af88bf17dbadcc00d50c13a5bdb6001367be
45052Author: Peter Hutterer <peter.hutterer@who-t.net>
45053Date: Fri Oct 22 14:22:44 2010 +1000
45054
45055 dmx: clarify a valuator mode condition.
45056
45057 Relative is defined as 0, so change the condition to be more obvious.
45058
45059 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45060 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45061
45062commit b5ef88c911f697530ddc2c7302fce15ba2ead4b2
45063Author: Peter Hutterer <peter.hutterer@who-t.net>
45064Date: Fri Oct 22 14:02:17 2010 +1000
45065
45066 dix: clip absolute axes depending on their mode.
45067
45068 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45069 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45070
45071commit ea567b675f814ac41e75f1ed8ded0ac3e9d552a8
45072Author: Peter Hutterer <peter.hutterer@who-t.net>
45073Date: Fri Oct 22 14:01:11 2010 +1000
45074
45075 dix: populate motion history only if the mode matches the first axis.
45076
45077 XI1 doesn't cater for mixed mode devices, so bail out on the first valuator
45078 that has a different mode.
45079
45080 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45081 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45082
45083commit 4381b70f5aa293400735e9b4e5a792408afa25a2
45084Author: Peter Hutterer <peter.hutterer@who-t.net>
45085Date: Fri Oct 22 13:57:27 2010 +1000
45086
45087 input: add valuator_get_mode() helper.
45088
45089 Returns the mode of the specified valuator.
45090
45091 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45092 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45093
45094commit 6f6f460c2485d16d126a1ab5d939083c322dc3dd
45095Author: Peter Hutterer <peter.hutterer@who-t.net>
45096Date: Fri Oct 22 13:48:06 2010 +1000
45097
45098 dix: send proximity events if one or more axes are Absolute.
45099
45100 We only skip relative events for proximity, not absolute ones. Now with
45101 mixed mode, just unset those axes that are relative.
45102
45103 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45104 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45105
45106commit 65c0fc81eb920085e650b8c9e874c9dd26c7ec98
45107Author: Chase Douglas <chase.douglas@canonical.com>
45108Date: Fri Jul 16 09:21:19 2010 -0400
45109
45110 Add support for per-axis valuator modes (Relative/Absolute)
45111
45112 The XI2 protocol supports per-axis modes, but the server so far does
45113 not. This change adds support in the server.
45114
45115 A complication is the fact that XI1 does not support per-axis modes.
45116 The solution provided here is to set a per-device mode that defines the
45117 mode of at least the first two valuators (X and Y). Note that initializing
45118 the first two axes to a different mode than the device mode will fail.
45119
45120 For XI1 events, any axes following the first two that have the same mode
45121 will be sent to clients, up to the first axis that has a different mode.
45122 Thus, if a device has relative, then absolute, then relative mode axes,
45123 only the first block of relative axes will be sent over XI1.
45124
45125 Since the XI2 protocol supports per-axis modes, all axes are sent to the
45126 client.
45127
45128 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
45129
45130commit 290af0418f8c94355e42ad81aa7779b90e173fc9
45131Author: Chase Douglas <chase.douglas@ubuntu.com>
45132Date: Thu Oct 21 22:41:09 2010 -0400
45133
45134 test: input - set valuators mask for event to core conversion
45135
45136 Commit de8be07cc0a8163b6ef04455706fd5ca2cebe587 adds a requirement to
45137 event to core conversion that at least one of the X or Y valuators are
45138 set in the valuator mask. This commit fixes the event conversion test to
45139 be compliant.
45140
45141 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
45142 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
45143 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45144
45145commit 8990b31214bcbc08090604147287455bfde91c11
45146Author: Paulius Zaleckas <paulius.zaleckas@gmail.com>
45147Date: Mon Oct 18 00:02:01 2010 +0300
45148
45149 KDrive: Fix error handlig in tslib driver
45150
45151 If ts_open() fails and return NULL, then next call to ts_fd()
45152 segfaults because of NULL dereference. There is no need to
45153 check output of ts_fd() as ts_open() did this internally.
45154
45155 Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
45156 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
45157 Reviewed-by: Keith Packard <keithp@keithp.com>
45158 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45159
45160commit 9696c782c8cb86b06d12949899582533a2e04cfe
45161Author: Peter Hutterer <peter.hutterer@who-t.net>
45162Date: Tue Oct 19 12:52:57 2010 +1000
45163
45164 test: valuator_mask tests.
45165
45166 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45167 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45168
45169commit 2362adc2da9745e01c79ed4905aa81926355c4e2
45170Author: Peter Hutterer <peter.hutterer@who-t.net>
45171Date: Tue Oct 19 10:59:09 2010 +1000
45172
45173 dix: remove valuator_mask_copy_valuators, not needed anymore.
45174
45175 With the switch to masks internally, this isn't needed anymore.
45176
45177 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45178 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45179
45180commit 2b04a3ef739a7ba5708400999e5d81418236a0a4
45181Author: Peter Hutterer <peter.hutterer@who-t.net>
45182Date: Tue Oct 19 09:58:58 2010 +1000
45183
45184 xfree86: convert xf86Post{Motion|Key|Button|Proximity}Event to masks
45185
45186 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45187 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45188
45189commit 0418a39e71e50b88e050b0d8a758265b17c4bf0e
45190Author: Peter Hutterer <peter.hutterer@who-t.net>
45191Date: Mon Oct 18 15:55:53 2010 +1000
45192
45193 dix: get rid of the now-superfluous valuator arrays in GPE and friends.
45194
45195 The valuators are stored inside the mask, use it from there. are stored
45196 inside the mask, use it from there. are stored inside the mask, use it from
45197 there. are stored inside the mask, use it from there.
45198
45199 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45200 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45201
45202commit 675f4a8525d29ebad783351e17be785b2f32b2e8
45203Author: Peter Hutterer <peter.hutterer@who-t.net>
45204Date: Tue Oct 19 13:37:46 2010 +1000
45205
45206 Abstract valuator masks through a set of APIs.
45207
45208 This commit introduces an abstraction API for handling masked valuators. The
45209 intent is that drivers just allocate a mask, set the data and pass the mask
45210 to the server. The actual storage type of the mask is hidden from the
45211 drivers.
45212
45213 The new calls for drivers are:
45214 valuator_mask_new() /* to allocate a valuator mask */
45215 valuator_mask_zero() /* to reset a mask to zero */
45216 valuator_mask_set() /* to set a valuator value */
45217
45218 The new interface to the server is
45219 xf86PostMotionEventM()
45220 xf86PostButtonEventM()
45221 xf86PostKeyboardEventM()
45222 xf86PostProximityEventM()
45223
45224 all taking a mask instead of the valuator array.
45225
45226 The ValuatorMask is currently defined for MAX_VALUATORS fixed size due to
45227 memory allocation restrictions in SIGIO handlers.
45228
45229 For easier review, a lot of the code still uses separate valuator arrays.
45230 This will be fixed in a later patch.
45231
45232 This patch was initially written by Chase Douglas.
45233
45234 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
45235 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45236 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45237
45238commit 1a0d9324b3d9fd93e685066e0e5cea0611878c0d
45239Author: Aaron Plattner <aplattner@nvidia.com>
45240Date: Sun Oct 17 09:58:50 2010 -0700
45241
45242 Revert "Set DamageSetReportAfterOp to true for the damage extension" (#30260)
45243
45244 This commit breaks the Compiz "Wallpaper" plugin.
45245
45246 This reverts commit 8d7b7a0d71e0b89321b3341b781bc8845386def6.
45247
45248 Reviewed-by: Keith Packard <keithp@keithp.com>
45249 Signed-off-by: Keith Packard <keithp@keithp.com>
45250
45251commit d738175eaf1098e29b8afb6de8e99b5098e366a7
45252Merge: 8806a04 9872e0f
45253Author: Keith Packard <keithp@keithp.com>
45254Date: Tue Oct 19 22:50:15 2010 -0700
45255
45256 Merge remote branch 'whot/for-keith'
45257
45258commit fc48a8f9f5f66e591b3e39211d44ce68267303f8
45259Author: Chase Douglas <chase.douglas@canonical.com>
45260Date: Tue Oct 19 13:37:38 2010 +1000
45261
45262 Add CountBits() to the server.
45263
45264 Function to count the number of bits set in the given array.
45265
45266 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
45267 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45268 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45269
45270commit 9872e0f0115cd7d0543e55a1831facddf3aff90c
45271Author: Peter Hutterer <peter.hutterer@who-t.net>
45272Date: Wed Oct 13 10:31:51 2010 +1000
45273
45274 xfree86: set mask for valuators 0/1 when emulating core events (#30267)
45275
45276 EventToCore as of the commit below won't generate core motion events if the
45277 valuator mask for x/y isn't set. For DGA, we work around this check by
45278 forcibly setting the mask in the event we pass down.
45279
45280 commit de8be07cc0a8163b6ef04455706fd5ca2cebe587
45281 Author: Peter Hutterer <peter.hutterer@who-t.net>
45282 Date: Tue Aug 17 12:08:52 2010 +1000
45283
45284 dix: don't create core motion events for non-x/y valuators.
45285
45286 X.Org Bug 30267 <http://bugs.freedesktop.org/show_bug.cgi?id=30267>
45287
45288 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45289 Tested-by: Andrew Randrianasulu <randrik@mail.ru>
45290 Tested-by: Andy Furniss
45291 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
45292
45293commit 649293f6b634e6305b6737a841d6e9d0f0065d6c
45294Author: Peter Hutterer <peter.hutterer@who-t.net>
45295Date: Mon Oct 11 16:09:18 2010 +1000
45296
45297 xkb: always fill the symsPerKey array, regardless of client flags (#30527)
45298
45299 Even if a client does not modify the symbols, symsPerKey and mapWidths must
45300 be filled from the current configuration. Both arrays are then passed into
45301 other functions (pending the right flag), thus they must contain valid
45302 values regardless of the XkbKeySymsMask flag in req->present.
45303
45304 X.Org Bug 30527 <http://bugs.freedesktop.org/show_bug.cgi?id=30527>
45305
45306 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45307 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
45308
45309commit ca21a266224b6eff4fd817c2082d2e144f1ea58c
45310Author: Peter Hutterer <peter.hutterer@who-t.net>
45311Date: Mon Oct 11 15:38:38 2010 +1000
45312
45313 xkb: init mapWidth and symsPerKey arrays to 0.
45314
45315 Helps debugging greatly, random 8 or 16 bit values can sometimes look like
45316 valid values, causing much excitement on the client front.
45317
45318 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45319 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
45320
45321commit 8806a04f7f028dd9f6a7ff5c36e460d3895c7deb
45322Author: Michał Górny <mgorny@gentoo.org>
45323Date: Thu Oct 14 15:49:51 2010 +0200
45324
45325 kdrive: Fix linking with tslib, with -Wl, --as-needed. #30600
45326
45327 Change the library order in TSLIB_LIBS variable, placing ${TSLIB_LIBS}
45328 after the local static libraries. This fixes linking with
45329 -Wl,--as-needed.
45330
45331 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=30600
45332 Signed-off-by: Michał Górny <mgorny@gentoo.org>
45333 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
45334 Signed-off-by: Keith Packard <keithp@keithp.com>
45335
45336commit c7e436e9987a6fb0876c63e96f59ff4c20b4bef0
45337Author: Michał Górny <mgorny@gentoo.org>
45338Date: Thu Oct 14 15:49:50 2010 +0200
45339
45340 kdrive: Fix tslib check fallback to set TSLIB_LIBS. #30599
45341
45342 If pkg-config is unable to find tslib but the fallback check does find
45343 it, the compilation continues with tslib support enabled though
45344 TSLIB_LIBS are unset. Thus, the compilation fails with a linking error
45345 on tslib functions.
45346
45347 This patch sets TSLIB_LIBS to '-lts' whenever the tslib fallback check
45348 succeeds.
45349
45350 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=30599
45351 Signed-off-by: Michał Górny <mgorny@gentoo.org>
45352 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
45353 Signed-off-by: Keith Packard <keithp@keithp.com>
45354
45355commit 25801a2556ca9b42646023e5a95a3cbf7309e852
45356Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45357Date: Thu Oct 14 22:38:17 2010 +0100
45358
45359 Cygwin/X: Move duplicate extern variable declarations from various .c files to a new header file
45360
45361 For the global variables defined in winglobals.c, remove duplicate extern
45362 declarations from the beginning of various .c files, and move most of them
45363 into a new header file, winglobals.h
45364
45365 Leave some clipboard related variables alone for the moment, they need treating
45366 more carefully, to avoid mixing client and server type definitions.
45367
45368 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45369 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45370
45371commit 04d11abcf2ea2f36d899428e9c885b69e4f779ad
45372Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45373Date: Thu Oct 14 22:45:31 2010 +0100
45374
45375 Cygwin/X: Remove g_hwndKeyboardFocus
45376
45377 It's set but it's value is never used
45378
45379 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45380 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45381
45382commit 5e896e425fe14022b0714301531bd54151c763f0
45383Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45384Date: Thu Oct 14 22:43:51 2010 +0100
45385
45386 Cygwin/X: Make g_hmodCommonControls static
45387
45388 Make g_hmodCommonControls static, it's only used in InitOutput.c
45389
45390 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45391 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45392
45393commit 49b996faadf0fedcee25d4a8f96f4c4007c29f5f
45394Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45395Date: Fri Oct 15 13:00:46 2010 +0100
45396
45397 Cygwin/X: Make g_hmodDirectDraw static
45398
45399 Add winReleaseDDProcAddresses() for releasing the direct draw module
45400 handle, so g_hmodDirectDraw can be made static
45401
45402 Remove unneeded duplicate extern definitions of g_fpDirectDrawCreate
45403 and g_fpDirectDrawCreateClipper, and move the definition from
45404 winglobals.c to winengine.c
45405
45406 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45407 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45408
45409commit 6f12277e37d16a7de9d3f3079688c43b07921b0a
45410Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45411Date: Thu Oct 14 22:44:53 2010 +0100
45412
45413 Cygwin/X: Make g_hhookKeyboardLL static
45414
45415 Make g_hhookKeyboardLL static, it's only used in winkeyhook.c
45416
45417 Also remove unused externs
45418
45419 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45420 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45421
45422commit 9952b51f6822aeaa19c27caf85698741eb4c27d8
45423Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45424Date: Fri Oct 15 13:01:45 2010 +0100
45425
45426 Cygwin/X: Fix typos in comments
45427
45428 Fix a few typos in comments
45429
45430 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45431 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45432
45433commit 1de5db73d31411798ae93619d9e626fa32e0c4bd
45434Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45435Date: Thu Oct 14 22:46:09 2010 +0100
45436
45437 Cygwin/X: Drop several unneeded includes of winprefs.h
45438
45439 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45440 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45441
45442commit 8e723108151136f115e8e6d4b97ed4e2fa685dbf
45443Author: Colin Harrison <colin.harrison@virgin.net>
45444Date: Wed Sep 29 14:18:27 2010 +0100
45445
45446 Xming: Remove uses of register keyword
45447
45448 I'm pretty sure the compiler has a better idea how to optimize this
45449
45450 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45451 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45452
45453commit d11761c6a6dd2a9963dfd3bcb313f1b2525387fc
45454Author: Colin Harrison <colin.harrison@virgin.net>
45455Date: Fri Feb 26 13:31:18 2010 +0000
45456
45457 Xming: Casts to remove warnings on event pointers
45458
45459 winkeybd.c: In function ‘winSendKeyEvent’:
45460 winkeybd.c:489: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible pointer type
45461 winmouse.c: In function ‘winMouseButtonsSendEvent’:
45462 winmouse.c:247: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible pointer type
45463 winmouse.c: In function ‘winEnqueueMotion’:
45464 winmouse.c:380: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible pointer type
45465
45466 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45467 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45468
45469commit 10bf8345cf15c6030147a9edb7f6a093c8db8099
45470Author: Colin Harrison <colin.harrison@virgin.net>
45471Date: Mon Oct 12 13:50:00 2009 +0100
45472
45473 Xming: Warning fix in winDeviceCursorCleanup()
45474
45475 return statement with an expression in a function whose return-type is void
45476
45477 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45478 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45479
45480commit cab837a60a7aa653ccea77ff714335186ce38a4b
45481Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45482Date: Sun Jul 18 13:22:46 2010 +0100
45483
45484 Cygwin/X: Reformat the keyboard layout mapping table
45485
45486 Also remove the out-of-date listing of keyboard layout IDs and which ones
45487 have known mappings
45488
45489 XXX: layout zh_TW doesn't exist anymore
45490
45491 This patch brought to you by C-u M-x align-regexp
45492
45493 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45494 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45495
45496commit ee2867649d9485f6559f88a973eae6393971ff35
45497Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45498Date: Sun Jul 18 13:20:01 2010 +0100
45499
45500 Cygwin/X: Clarify XKB options in XWin man page
45501
45502 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45503 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45504
45505commit 1ba1b12d24d8f825babd8cf38710c7fa9a94821e
45506Author: Colin Harrison <colin.harrison@virgin.net>
45507Date: Fri Oct 15 13:46:09 2010 +0100
45508
45509 Xming: Warning fix in winApplyHints()
45510
45511 winmultiwindowwm.c: In function ‘winApplyHints’:
45512 winmultiwindowwm.c:1587: warning: ISO C90 forbids mixed declarations and code
45513
45514 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45515 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45516
45517commit fb64414548c437bf52c3dcd0613aab5079faffba
45518Author: Colin Harrison <colin.harrison@virgin.net>
45519Date: Fri Oct 15 14:19:57 2010 +0100
45520
45521 Xming: Fix warnings in ProcWindowsWMFrameSetTitle()
45522
45523 winwindowswm.c: In function ‘ProcWindowsWMFrameSetTitle’:
45524 winwindowswm.c:516: warning: pointer targets in passing argument 1 of ‘strncpy’ differ in signedness
45525 winwindowswm.c:516: warning: pointer targets in passing argument 2 of ‘strncpy’ differ in signedness
45526 winwindowswm.c:528: warning: pointer targets in passing argument 2 of ‘SetWindowTextA’ differ in signedness
45527
45528 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45529 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45530
45531commit da5955cc023ae980fdc0397f6fd541e788040c29
45532Author: Colin Harrison <colin.harrison@virgin.net>
45533Date: Fri Oct 15 14:18:53 2010 +0100
45534
45535 Xming: Fix warnings in windialogs.c
45536
45537 windialogs.c: In function ‘winDisplayExitDialog’:
45538 windialogs.c:327: warning: passing argument 3 of ‘PostMessageA’ makes integer from pointer without a cast
45539 windialogs.c: In function ‘winDisplayAboutDialog’:
45540 windialogs.c:597: warning: passing argument 3 of ‘PostMessageA’ makes integer from pointer without a cast
45541 windialogs.c: In function ‘winAboutDlgProc’:
45542 windialogs.c:697: warning: comparison between pointer and integer
45543 windialogs.c:701: warning: format ‘%d’ expects type ‘int’, but argument 2 has type ‘HINSTANCE’
45544 windialogs.c:716: warning: assignment makes integer from pointer without a cast
45545 windialogs.c:736: warning: assignment makes integer from pointer without a cast
45546 windialogs.c:756: warning: assignment makes integer from pointer without a cast
45547
45548 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45549 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45550
45551commit 6d9fb07db210bcfdcdde632fa341c4ee08d44b35
45552Author: Colin Harrison <colin.harrison@virgin.net>
45553Date: Fri Oct 15 14:16:54 2010 +0100
45554
45555 Xming: Fix warnings in winNameCompare(), winNormalizeName()
45556
45557 winconfig.c: In function ‘winNameCompare’:
45558 winconfig.c:715: warning: array subscript has type ‘char’
45559 winconfig.c:715: warning: array subscript has type ‘char’
45560 winconfig.c:716: warning: array subscript has type ‘char’
45561 winconfig.c:716: warning: array subscript has type ‘char’
45562 winconfig.c:730: warning: array subscript has type ‘char’
45563 winconfig.c:730: warning: array subscript has type ‘char’
45564 winconfig.c:731: warning: array subscript has type ‘char’
45565 winconfig.c:731: warning: array subscript has type ‘char’
45566 winconfig.c: In function ‘winNormalizeName’:
45567 winconfig.c:1092: warning: array subscript has type ‘char’
45568 winconfig.c:1093: warning: array subscript has type ‘char’
45569
45570 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45571 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45572
45573commit 20510e51001d1659000c8d8607c0de16022d1c74
45574Author: Colin Harrison <colin.harrison@virgin.net>
45575Date: Fri Oct 15 14:15:51 2010 +0100
45576
45577 Xming: Fix warnings in winClipboardFlushXEvents()
45578
45579 winclipboardxevents.c: In function ‘winClipboardFlushXEvents’:
45580 winclipboardxevents.c:225: warning: format ‘%08x’ expects type ‘unsigned int’, but argument 2 has type ‘DWORD’
45581 winclipboardxevents.c:266: warning: format ‘%08x’ expects type ‘unsigned int’, but argument 2 has type ‘DWORD’
45582 winclipboardxevents.c:619: warning: pointer targets in passing argument 1 of ‘strcat’ differ in signedness
45583 winclipboardxevents.c:659: warning: pointer targets in passing argument 1 of ‘strlen’ differ in signedness
45584 winclipboardxevents.c:669: warning: pointer targets in passing argument 3 of ‘MultiByteToWideChar’ differ in signedness
45585 winclipboardxevents.c:690: warning: pointer targets in passing argument 3 of ‘MultiByteToWideChar’ differ in signedness
45586 winclipboardxevents.c:698: warning: pointer targets in passing argument 1 of ‘strdup’ differ in signedness
45587
45588 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45589 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45590
45591commit 44c8b449d281c4d984d65f66bac96a65806b2efa
45592Author: Colin Harrison <colin.harrison@virgin.net>
45593Date: Fri Oct 15 14:13:57 2010 +0100
45594
45595 Xming: Fix warning in winClipboardErrorHandler()
45596
45597 winclipboardthread.c: In function ‘winClipboardErrorHandler’:
45598 winclipboardthread.c:444: warning: format ‘%d’ expects type ‘int’, but argument 3 has type ‘long unsigned int’
45599
45600 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45601 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45602
45603commit 1f2a57d9e46424a2e0861f55253b91d930d79083
45604Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45605Date: Fri Oct 15 14:12:22 2010 +0100
45606
45607 Cygwin/X: Turn off the bad-function-cast warning for XWin code
45608
45609 Turn off the bad-function-cast warning for the XWin code, it's near impossible
45610 to write code which uses Win32 API calls that doesn't trigger this warning
45611
45612 For example, SendMessage(WM_SETICON) returns an LRESULT, but we are supposed to
45613 know this is safe to cast this result to a HICON, which gcc considers a non-matching
45614 type.
45615
45616 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45617 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45618
45619commit 55e528b6cc6451dbb1e65baec199d4df0441e86d
45620Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45621Date: Thu Mar 25 23:39:37 2010 +0000
45622
45623 Cygwin/X: Remove execute permission from source files
45624
45625 Remove execute permission from source files
45626
45627 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45628 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45629
45630commit b2b685e2693ff5f089006f5b3882d6b0c0853862
45631Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45632Date: Thu Mar 25 23:00:48 2010 +0000
45633
45634 Cygwin/X: unifdef XFree86Server
45635
45636 Remove XFree86Server define, which was always on anyhow, and the
45637 code which was guarded by !XFree86Server
45638
45639 This completes the process of removal started in 2006 :-)
45640
45641 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45642 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45643
45644commit 9ed70f15eff6a85222b46db1fe7af25154314464
45645Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45646Date: Mon Nov 2 17:37:14 2009 +0000
45647
45648 Cygwin/X: Use normal apostrophe in XWin DDX help text
45649
45650 Use U+0027 APOSTROPHE in XWin DDX help text, rather than
45651 U+2019 RIGHT SINGLE QUOTATION MARK
45652
45653 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45654 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45655
45656commit 9f8f056a21b9480fd9fd43a22c9d6f45a2c1ef87
45657Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45658Date: Tue Mar 23 16:34:26 2010 +0000
45659
45660 Cygwin/X: Cleanup some VENDOR_STRING/VENDOR_CONTACT cruft
45661
45662 Cleanup some VENDOR_STRING/VENDOR_CONTACT cruft
45663
45664 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45665 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
45666
45667commit 5aff712a8d2eb9f965ecbb93216cc0bcdc327ae6
45668Merge: 5b98c62 e354cca
45669Author: Keith Packard <keithp@keithp.com>
45670Date: Mon Oct 18 17:16:23 2010 -0700
45671
45672 Merge remote branch 'whot/for-keith'
45673
45674commit 5b98c6267f575ec4ff498ecb651e34dbff030b83
45675Author: Adam Jackson <ajax@redhat.com>
45676Date: Mon Oct 18 16:01:40 2010 -0400
45677
45678 randr: Remove mirandr
45679
45680 This isn't used anywhere, not least because it's completely
45681 nonfunctional.
45682
45683 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
45684 Signed-off-by: Adam Jackson <ajax@redhat.com>
45685
45686commit 7bb653bedceb6180a0361ead1c612839e776ce98
45687Author: Olivier Fourdan <ofourdan@redhat.com>
45688Date: Mon Oct 18 15:59:35 2010 -0400
45689
45690 modes: improve aspect ratio match for classic drivers
45691
45692 After we infer the aspect ratio for the screen, we pick the largest
45693 mode matching that aspect ratio from the best mode pool available.
45694 We then clamp virtual size to that mode, and run the resulting mode
45695 list through the driver's ValidMode hook. In doing so we might filter
45696 away our initial guess. If this happens we shrink the default mode
45697 to the next largest mode from _any_ mode pool. This is usually wrong,
45698 and we should instead pick the next aspect-matched mode from the best
45699 available mode pool (as always, user then driver then default).
45700
45701 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
45702 Signed-off-by: Adam Jackson <ajax@redhat.com>
45703
45704commit d6d90a4fccef086e7c7934adea901b2cd9a595bf
45705Author: Adam Jackson <ajax@redhat.com>
45706Date: Mon Oct 18 15:53:43 2010 -0400
45707
45708 composite: Move the backfill comment to a more appropriate indent level
45709
45710 Reviewed-by: Keith Packard <keithp@keithp.com>
45711 Signed-off-by: Adam Jackson <ajax@redhat.com>
45712
45713commit db8840600e8e21356241eb87395031388d9b54d2
45714Author: Adam Jackson <ajax@redhat.com>
45715Date: Mon Oct 18 15:52:52 2010 -0400
45716
45717 composite: Don't backfill non-MapWindow allocations
45718
45719 Reviewed-by: Keith Packard <keithp@keithp.com>
45720 Signed-off-by: Adam Jackson <ajax@redhat.com>
45721
45722commit 6dd775f57d2f94f0ddaee324aeec33b9b66ed5bc
45723Author: Adam Jackson <ajax@redhat.com>
45724Date: Mon Oct 18 15:50:37 2010 -0400
45725
45726 composite: Don't backfill non-bg-None windows
45727
45728 If there's a defined background then backfilling is a waste of effort,
45729 since exposure processing will paint that in for us. But note that we
45730 have to backfill if any children are bg=None to preserve semantics with
45731 non-composited servers.
45732
45733 Reviewed-by: Keith Packard <keithp@keithp.com>
45734 Signed-off-by: Adam Jackson <ajax@redhat.com>
45735
45736commit d2064fbb687839c297a851a5d85f32dfbbe4a0d5
45737Author: Adam Jackson <ajax@redhat.com>
45738Date: Mon Oct 18 15:42:54 2010 -0400
45739
45740 ddc: Don't probe for DDC/CI or EEPROM
45741
45742 For whatever reason, some (broken) monitors will crash if you do this.
45743 We're not actually using this information for anything, so let's just
45744 not do it.
45745
45746 Originally reported as http://bugzilla.redhat.com/620333
45747
45748 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
45749 Signed-off-by: Adam Jackson <ajax@redhat.com>
45750
45751commit 47c91dca8d8eecb429123e8370302831bcd57938
45752Author: Adam Jackson <ajax@redhat.com>
45753Date: Mon Oct 18 15:40:28 2010 -0400
45754
45755 xfree86: Drop linux libc5 support from the SIGIO code
45756
45757 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
45758 Reviewed-by: Matt Turner <mattst88@gmail.com>
45759 Signed-off-by: Adam Jackson <ajax@redhat.com>
45760
45761commit 788bfbf18ac19923604b0d676933fdc0d78526a1
45762Author: Adam Jackson <ajax@redhat.com>
45763Date: Thu Oct 14 15:10:43 2010 -0400
45764
45765 dri1: Remove "buffers" from the config logic
45766
45767 This was only ever used from the glint driver, which has since lost its
45768 DRI support.
45769
45770 Reviewed-by: Matt Turner <mattst88@gmail.com>
45771 Signed-off-by: Adam Jackson <ajax@redhat.com>
45772
45773commit a77458486a7b9d78dcd284e2bcc96c1fda0e862a
45774Author: Adam Jackson <ajax@redhat.com>
45775Date: Thu Oct 14 14:00:53 2010 -0400
45776
45777 xfree86: Remove %M expansion from config parser
45778
45779 This was to distinguish XFree86 3.x files from XFree86 4.x files. It
45780 never really made sense to be looking for xorg.conf-4.
45781
45782 Reviewed-by: Matt Turner <mattst88@gmail.com>
45783 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
45784 Reviewed-by: Jesse Adkins <jesserayadkins@gmail.com>
45785 Signed-off-by: Adam Jackson <ajax@redhat.com>
45786
45787commit f8ec71603c796f3b272a0592cbb934c2a6e8d633
45788Author: Adam Jackson <ajax@redhat.com>
45789Date: Thu Oct 14 13:45:31 2010 -0400
45790
45791 xfree86: Remove an open-coded strtoul()
45792
45793 Reviewed-by: Matt Turner <mattst88@gmail.com>
45794 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
45795 Signed-off-by: Adam Jackson <ajax@redhat.com>
45796
45797commit 52577ae8eeda64601db8dd425027cf8e4271b873
45798Author: Adam Jackson <ajax@redhat.com>
45799Date: Thu Oct 14 13:22:00 2010 -0400
45800
45801 config: Remove atimisc workaround
45802
45803 Reviewed-by: Matt Turner <mattst88@gmail.com>
45804 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
45805 Reviewed-by: Jesse Adkins <jesserayadkins@gmail.com>
45806 Signed-off-by: Adam Jackson <ajax@redhat.com>
45807
45808commit 8bb1983a5f2dbf72366de9fb793a46e35912ab61
45809Author: Adam Jackson <ajax@redhat.com>
45810Date: Mon Aug 2 15:22:18 2010 -0400
45811
45812 config: Don't look for .o drivers anymore
45813
45814 This hasn't worked since we switched to dlloader.
45815
45816 Reviewed-by: Jesse Adkins <jesserayadkins@gmail.com>
45817 Reviewed-by: Matt Turner <mattst88@gmail.com>
45818 Signed-off-by: Adam Jackson <ajax@redhat.com>
45819
45820commit e354ccac36a8ee3a23bdc845833c16a5646cc200
45821Author: Joe Shaw <joeshaw@litl.com>
45822Date: Thu Oct 14 15:09:20 2010 -0400
45823
45824 fix a sign problem with valuator data.
45825
45826 Without this patch, any negative valuator value is wrong when returned
45827 from XQueryDeviceState(). This is a regression from at least xserver
45828 1.4.
45829
45830 Valuator data is set in dix/getevents.c:set_valuators() by copying
45831 signed int values into an unsigned int field
45832 DeviceEvent.valuators.data.
45833
45834 That data is converted into a double with an implicit cast by
45835 assignment to axisVal[i] in Xi/exevents.c:UpdateDeviceState().
45836
45837 That double is converted back to a signed int in
45838 queryst.c:ProcXQueryDeviceState(). If the original value in
45839 set_valuators() is negative, the double value will be > 2^31 and the
45840 conversion back to a signed int is undefined. (Although I
45841 consistently see the value -2^31.)
45842
45843 Fix this by changing the definition of DeviceEvent.valuators.data from
45844 uint32_t to int32_t.
45845
45846 Signed-off-by: Joe Shaw <joeshaw@litl.com>
45847 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45848 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
45849 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45850
45851commit 424b856e8e19f35c24bfc0a9fced9464d2f17c90
45852Author: Peter Hutterer <peter.hutterer@who-t.net>
45853Date: Fri Oct 15 14:35:22 2010 +1000
45854
45855 dix: update comments for GetPointerEvents and friends
45856
45857 All these now generate InternalEvents, point this out. Remove XKB/XI
45858 references, that's just confusing. This comment referred to the old-style
45859 event generation code from server 1.4 to including 1.6 but is now just
45860 confusing to newcomers.
45861
45862 Remove comment about SwitchCoreKeyboard() for the same reason.
45863
45864 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45865 Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
45866
45867commit e2ada55de3e7ec8724b98c4e4adfec741d415499
45868Author: Adam Jackson <ajax@redhat.com>
45869Date: Mon Sep 27 13:58:40 2010 -0400
45870
45871 os: Clean up various xtrans bits that we've never supported
45872
45873 Or at least, not supported since xserver 1.0.
45874
45875 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
45876 Signed-off-by: Adam Jackson <ajax@redhat.com>
45877
45878commit eaf0b6a4d83d49930d21d5191f335fcac962632e
45879Merge: 693e92d 23229c7
45880Author: Peter Hutterer <peter.hutterer@who-t.net>
45881Date: Fri Oct 15 13:46:59 2010 +1000
45882
45883 Merge branch 'master' into input-api
45884
45885 Conflicts:
45886 config/udev.c
45887 hw/xfree86/common/xf86Helper.c
45888 hw/xfree86/common/xf86Module.h
45889 hw/xfree86/common/xf86Xinput.h
45890 hw/xfree86/os-support/linux/lnx_init.c
45891
45892 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45893
45894commit 3c28a29e132d6f73c36d4b64818d112b1c6e9a40
45895Author: Peter Hutterer <peter.hutterer@who-t.net>
45896Date: Wed Oct 13 15:53:59 2010 +1000
45897
45898 xfree86: rename parameter names to xf86ScaleAxis.
45899
45900 Maybe it's just me but every time I look at it I get confused again and need
45901 to work it out from scratch. Rename the parameters to something
45902 self-explanatory, to/from and min/max.
45903
45904 No functional change.
45905
45906 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
45907 Reviewed-by: Julien Cristau <jcristau@debian.org>
45908
45909commit 23229c7ce5970516b97a1cc12e65257ff65ab9a7
45910Author: Jeremy Huddleston <jeremyhu@apple.com>
45911Date: Wed Oct 13 16:28:14 2010 -0700
45912
45913 test: Fix make distcheck when not building unit tests
45914
45915 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
45916 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
45917 Signed-off-by: Keith Packard <keithp@keithp.com>
45918
45919commit ff82f724e4ed24d73f15fa9c35a1d0c2c7d8cef7
45920Author: Colin Harrison <colin.harrison@virgin.net>
45921Date: Wed Oct 13 10:20:06 2010 -0700
45922
45923 Comparison between pointer and integer in latest dix/dispatch.c
45924
45925 Signed-off-by: Keith Packard <keithp@keithp.com>
45926
45927commit da4eebe58ec81e2297cc1773af1fb622fb392bbd
45928Author: Jon TURNEY <jon.turney@dronecode.org.uk>
45929Date: Wed Oct 13 17:34:31 2010 +0100
45930
45931 event_size is currently never assigned to in mieqProcessInputEvents().
45932
45933 event_size is never assigned to in mieqProcessInputEvents(), so realloc()
45934 is always called. This is benign, but I'm guessing not intended.
45935
45936 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45937 Reviewed-by: Keith Packard <keithp@keithp.com>
45938 Signed-off-by: Keith Packard <keithp@keithp.com>
45939
45940commit 6274dca9d984ad3c553b4901edc3151e770e6c40
45941Author: Adam Jackson <ajax@redhat.com>
45942Date: Mon Oct 11 16:37:16 2010 -0400
45943
45944 dix: optimize CallCallbacks
45945
45946 Move the basic sanity checking to an inline wrapper, which avoids the
45947 function call overhead if the callback list is empty. On an XACEful
45948 server on a 2.4GHz Core 2 Duo:
45949
45950 1 2 Operation
45951 -------- ----------------- -----------------
45952 20000000.0 25100000.0 ( 1.25) X protocol NoOperation
45953
45954 Signed-off-by: Adam Jackson <ajax@redhat.com>
45955 Reviewed-by: Keith Packard <keithp@keithp.com>
45956 Signed-off-by: Keith Packard <keithp@keithp.com>
45957
45958commit e418cd332c1f458d028df3fdf684011109d0c183
45959Author: Adam Jackson <ajax@redhat.com>
45960Date: Mon Oct 11 10:49:57 2010 -0400
45961
45962 xwin: Remove some undue paranoia about UNDRAWABLE_WINDOW
45963
45964 mi will throw away rendering to input-only windows.
45965
45966 Signed-off-by: Adam Jackson <ajax@redhat.com>
45967 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45968 Signed-off-by: Keith Packard <keithp@keithp.com>
45969
45970commit febf3e7ead8dbb355622c896ff1744b2ca52a49a
45971Author: Adam Jackson <ajax@redhat.com>
45972Date: Mon Oct 11 10:49:56 2010 -0400
45973
45974 dix: Remove the memory of the multibuffer extension
45975
45976 Drop DRAWABLE_BUFFER and related checks, mbuf was the only thing that
45977 used them and it was killed in 0ba82562.
45978
45979 Signed-off-by: Adam Jackson <ajax@redhat.com>
45980 Reviewed-by: Keith Packard <keithp@keithp.com>
45981 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
45982 Signed-off-by: Keith Packard <keithp@keithp.com>
45983
45984commit 1333e101b4f6ecca72568b7f462884f8b102a5c2
45985Author: Alan Coopersmith <alan.coopersmith@oracle.com>
45986Date: Thu Oct 7 18:17:02 2010 -0700
45987
45988 Remove leftover fbSaveAreas & fbRestoreAreas defines from wfbrename.h
45989
45990 The fb functions they try to rename were deleted in 2007 by
45991 commit ae7f71a8b3d6756161e55d998d6eec37d2695c98
45992
45993 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
45994 Reviewed-by: Matt Turner <mattst88@gmail.com>
45995 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
45996 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
45997 Signed-off-by: Keith Packard <keithp@keithp.com>
45998
45999commit 8efa3623f32939ef525926b57fd533f4c01ec4b3
46000Author: Jamey Sharp <jamey@minilop.net>
46001Date: Wed Aug 4 13:48:33 2010 -0700
46002
46003 Use GetCurrentRootWindow or equivalent instead of spriteTrace[0].
46004
46005 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46006 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
46007 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
46008
46009commit 32de0c19075b886c2fc03f6ad73d9c15df1b2309
46010Author: Jamey Sharp <jamey@minilop.net>
46011Date: Wed Aug 4 11:44:05 2010 -0700
46012
46013 Move replay-window check from ComputeFreezes to CheckDeviceGrabs.
46014
46015 This just simplifies ComputeFreezes, eliminating some duplicated code
46016 and a goto.
46017
46018 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46019 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
46020 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
46021
46022commit 1c1811ecafc144151630b3588394e2c345c53aa4
46023Author: Jamey Sharp <jamey@minilop.net>
46024Date: Tue Aug 3 19:49:59 2010 -0700
46025
46026 CheckDeviceGrabs: Delete redundant pWin->optional test.
46027
46028 CheckPassiveGrabsOnWindow returns FALSE if pWin->optional is NULL,
46029 because wPassiveGrabs uses wUseDefault, so don't bother checking at the
46030 caller.
46031
46032 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46033 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
46034 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
46035
46036commit c360b3eb1deeac51913e7b1286b2767711c541fc
46037Author: Jamey Sharp <jamey@minilop.net>
46038Date: Tue Aug 3 19:40:26 2010 -0700
46039
46040 Let CheckPassiveGrabsOnWindow callers check if device is already grabbed.
46041
46042 CheckDeviceGrabs checked all the ancestors of the window containing this
46043 device's pointer even if no new grabs could possibly apply due to the
46044 device already being grabbed.
46045
46046 ActivateFocusInGrab and ActivateEnterGrab already checked whether they
46047 should break an existing grab, and then set up an event that was
46048 completely ignored if they didn't actually break the grab.
46049
46050 In both cases, just do what we would have done eventually anyway--return
46051 FALSE from CheckPassiveGrabsOnWindow's caller--but do it sooner.
46052
46053 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46054 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
46055 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
46056
46057commit 853d7ebfa3e2d281d92890a39010ff5787a00ffd
46058Author: Adam Jackson <ajax@redhat.com>
46059Date: Tue Oct 5 11:24:28 2010 -0400
46060
46061 vbe: Fix copying the mode info block
46062
46063 Just use memcpy, seriously.
46064
46065 Reviewed-by: Rémi Cardona <remi@gentoo.org>
46066 Signed-off-by: Adam Jackson <ajax@redhat.com>
46067 Signed-off-by: Keith Packard <keithp@keithp.com>
46068
46069commit b2f9ce201cc2a14d5d6ad055b46c9317b040ec2e
46070Author: Adam Jackson <ajax@redhat.com>
46071Date: Tue Jul 27 17:34:36 2010 -0400
46072
46073 edid: Fix the HDTV sync pulse adjustment
46074
46075 Simple typo, should have been adjusting the horizontal timings
46076 consistently since we're not trying to mangle vertical at all.
46077
46078 Signed-off-by: Adam Jackson <ajax@redhat.com>
46079 Reviewed-by: Patrick E. Kane <pekane52@gmail.com>
46080 Signed-off-by: Keith Packard <keithp@keithp.com>
46081
46082commit 9a893fe52f1bfd415e9ffd9d2f5aff5c4da7a756
46083Author: Chris Wilson <chris@chris-wilson.co.uk>
46084Date: Sun Oct 3 16:56:03 2010 +0100
46085
46086 xfree86: Do not call dlclose(NULL) [regression after ab7f057]
46087
46088 During unwind following an error when attempting to a load a module, we
46089 attempt to call dlclose on a potentially NULL handle. This is a
46090 side-effect of removing the abstraction layer in ab7f057.
46091
46092 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
46093 Cc: Adam Jackson <ajax@redhat.com>
46094 Reviewed-by: Adam Jackson <ajax@redhat.com>
46095 Signed-off-by: Keith Packard <keithp@keithp.com>
46096
46097commit 533faf7f43cb4b5155a74d78d6144b702e227efa
46098Author: Keith Packard <keithp@keithp.com>
46099Date: Fri Oct 1 10:57:48 2010 -0700
46100
46101 Bump version to 1.9.99.1 (1.10 development series)
46102
46103 Make the 1.10 development series distinguishable from the 1.9 branch.
46104
46105 Signed-off-by: Keith Packard <keithp@keithp.com>
46106
46107commit a769f4c22a9cfb5ba248c924a66c31ec966bd8a0
46108Author: Adrian Bunk <bunk@stusta.de>
46109Date: Thu Sep 30 13:18:20 2010 +0300
46110
46111 remove dolt
46112
46113 With libtool 2.2.10 the difference in build time is so small
46114 (< 5% with -j8 builds) that it isn't worth having this hack.
46115
46116 Signed-off-by: Adrian Bunk <bunk@stusta.de>
46117 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
46118 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
46119 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46120 Signed-off-by: Keith Packard <keithp@keithp.com>
46121
46122commit c7e4222c9a27094ce4fc2831ac92acbb7b21fb1a
46123Author: Tobias Droste <tdroste@gmx.de>
46124Date: Wed Sep 29 22:51:48 2010 +0200
46125
46126 randr: set error numbers of resource types in RRExtenstionInit() (V2)
46127
46128 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=30367
46129
46130 Currently the ddx calls xf86RandR12Init()
46131 (-> RRScreenInit()
46132 -> RRInit()
46133 -> RRModeInit()
46134 -> RRCrtcInit()
46135 -> RROutputInit())
46136 before RRExtensionInit() is called. This causes RRErrorBase
46137 being 0 while setting resource type error values (resource types:
46138 RROutput, RRMode and RRCrtc). The fix moves the setting of error
46139 values to own functions which are called in RRExtensionInit()
46140 to get the right RRErrorBase.
46141
46142 V2: With header file
46143
46144 Signed-off-by: Tobias Droste <tdroste@gmx.de>
46145 Reviewed-by: Keith Packard <keithp@keithp.com>
46146 Signed-off-by: Keith Packard <keithp@keithp.com>
46147
46148commit 72a9c686d7ecaa68d27687b316dbba92902809b4
46149Author: Jon TURNEY <jon.turney@dronecode.org.uk>
46150Date: Wed Sep 29 17:05:12 2010 +0100
46151
46152 Cygwin/X: Fix compilation after "delete pervasively use of DISPATCH_PROC"
46153
46154 commit cbd4d5dbb70db62ba1cb79c7b904e6fa11f62d7e
46155 removes the static declarations of ProcWindowsWMDispatch and
46156 SProcWindowsWMDispatch which precede their first use in
46157 winWindowsWMExtensionInit()
46158
46159 Move winWindowsWMExtensionInit() to after the definition of those
46160 two functions to fix compilation.
46161
46162 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
46163 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46164 Signed-off-by: Keith Packard <keithp@keithp.com>
46165
46166commit 1a9022d3289ff74043112808cc8aa315fd9c40e8
46167Author: Aaron Plattner <aplattner@nvidia.com>
46168Date: Tue Sep 28 13:59:19 2010 -0700
46169
46170 Bump video driver ABI version to 9.0
46171
46172 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
46173 Reviewed-by: Keith Packard <keithp@keithp.com>
46174 Signed-off-by: Keith Packard <keithp@keithp.com>
46175
46176commit c45bea0c044ad37bedb42209f7e6ea8b587999f0
46177Author: Jeremy Huddleston <jeremyhu@apple.com>
46178Date: Tue Aug 3 12:51:51 2010 -0700
46179
46180 XQuartz: RandR: Refactor legacy mode-switching to be better integrated with RandR
46181
46182 Adds three new functions
46183 void QuartzRandRSetFakeRootless (void);
46184 void QuartzRandRSetFakeFullscreen (void);
46185 void QuartzRandRToggleFullscreen (void);
46186
46187 The first two are identical to requesting the fake modes from a RandR client
46188 The third responds to cmd-alt-a to leave fullscreen or RandR.
46189
46190 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
46191
46192commit 229323a19b06f80d9b03f487e598b933b9b31d87
46193Author: Jeremy Huddleston <jeremyhu@apple.com>
46194Date: Mon Jul 26 23:34:49 2010 -0700
46195
46196 XQuartz: RandR: Don't change the rootless preference when changing RandR mode
46197
46198 Also renames a bunch of other variables for better consistency.
46199
46200 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
46201
46202commit 721edc69c3597578d3e7f711769ff0195af6068a
46203Author: Jeremy Huddleston <jeremyhu@apple.com>
46204Date: Mon Jul 26 11:45:59 2010 -0700
46205
46206 XQuartz: RandR: Better handle switching betwen RandR modes that share CG modes
46207
46208 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
46209
46210commit 026a47e21226b64eb9a7624ada3d74c3fe5fa763
46211Author: Jeremy Huddleston <jeremyhu@apple.com>
46212Date: Mon Jul 26 01:17:12 2010 -0700
46213
46214 XQuartz: RandR: Respond better to resolution changes made outside X
46215
46216 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
46217
46218commit 59cd93f48a3a0a34781d4a94a3cd05a575e2d4ba
46219Author: Jeremy Huddleston <jeremyhu@apple.com>
46220Date: Sun Jul 25 22:25:49 2010 -0700
46221
46222 XQuartz: RandR: Use deprecated CG APIs only on Leopard and earlier
46223
46224 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
46225
46226commit 629c3f473b7415069a87ef4717ec3d701978dc48
46227Author: Jan Hauffa <hauffa@in.tum.de>
46228Date: Fri Jul 16 17:56:44 2010 +0200
46229
46230 XQuartz: RandR: Remove FAKE_RANDR code.
46231
46232 Signed-off-by: Jan Hauffa <hauffa@in.tum.de>
46233 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46234
46235commit 97b5f5306437bfd13390485fc7a58a363c261ec9
46236Author: Jan Hauffa <hauffa@in.tum.de>
46237Date: Fri Jul 16 17:54:55 2010 +0200
46238
46239 XQuartz: RandR: Toggle rootless mode on XRandR mode switch.
46240
46241 Report a fake screen mode that corresponds to the screen mode at startup of the
46242 server excluding the height of the menu bar. If a client requests this mode,
46243 rootless mode is enabled. In all other modes, the root window is shown.
46244
46245 Signed-off-by: Jan Hauffa <hauffa@in.tum.de>
46246 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46247
46248commit 2d411472c2915614bdeb8fdc15d19dae4621444d
46249Author: Jan Hauffa <hauffa@in.tum.de>
46250Date: Fri Jul 16 17:18:03 2010 +0200
46251
46252 XQuartz: RandR: Implement basic RandR functionality.
46253
46254 Querying and changing of resolution and refresh rate is supported,
46255 rotation is not implemented yet.
46256
46257 Signed-off-by: Jan Hauffa <hauffa@in.tum.de>
46258 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46259
46260commit f4926709485c0605f0b1171c84c803b68eebfc38
46261Author: Jeremy Huddleston <jeremyhu@apple.com>
46262Date: Tue Sep 28 09:48:41 2010 -0700
46263
46264 XQuartz: Move use of static dispatch procs after declaration
46265
46266 This fixes fallout from cbd4d5dbb70db62ba1cb79c7b904e6fa11f62d7e
46267
46268 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
46269
46270commit c36ca7b1862ecd86df089f4e1933379a113928ea
46271Author: Gaetan Nadon <memsize@videotron.ca>
46272Date: Sat Sep 25 10:05:10 2010 -0400
46273
46274 config: remove --with-dri-driver-path option, use dri.pc #29740
46275
46276 The trusted location of the dri driver directory can be obtained
46277 from the mesa dri.pc pkg-config file.
46278
46279 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
46280 Acked-by: Matt Turner <mattst88@gmail.com>
46281 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
46282 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
46283 Signed-off-by: Keith Packard <keithp@keithp.com>
46284
46285commit 5532687a929426c4b1c4667f4591ed362f097c9b
46286Merge: 9df4fb0 682a3ee
46287Author: Keith Packard <keithp@keithp.com>
46288Date: Tue Sep 28 08:15:07 2010 -0700
46289
46290 Merge remote branch 'ajax/less-loader'
46291
46292commit 9df4fb0adf66fb82f0c007897d79af0f54a6dad9
46293Author: Adam Jackson <ajax@redhat.com>
46294Date: Mon Sep 27 17:06:46 2010 -0400
46295
46296 xfree86: Add 18bpp support
46297
46298 Signed-off-by: Adam Jackson <ajax@redhat.com>
46299 Reviewed-by: Keith Packard <keithp@keithp.com>
46300 Signed-off-by: Keith Packard <keithp@keithp.com>
46301
46302commit 5c6a2f93ebc16a78093782b442306de23ae94e78
46303Author: Chris Wilson <chris@chris-wilson.co.uk>
46304Date: Mon Sep 27 19:39:23 2010 +0100
46305
46306 xfree86: Kill pixmapPrivate with a vengeance (v2)
46307
46308 ScrnInfo->pixmapPrivate only existed in order to catch invalid access to
46309 the framebuffer by making the backing data NULL across the VT switch.
46310 This was causing more confusion in the higher layers during mode setting
46311 without any real benefit, so remove it.
46312
46313 v2: Kill ShadowModifyPixmapHeader() as well.
46314
46315 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
46316 Cc: Julien Cristau <jcristau@debian.org>
46317 Cc: Andrew Guertin <lists@dolphinling.net>
46318 Reviewed-by: Keith Packard <keithp@keithp.com>
46319 Signed-off-by: Keith Packard <keithp@keithp.com>
46320
46321commit fdabcec57ae0fdc9910060609bb0848552c1db4d
46322Author: Chris Wilson <chris@chris-wilson.co.uk>
46323Date: Mon Sep 27 17:30:13 2010 +0100
46324
46325 modes: Beware the driver switching root pixmaps
46326
46327 Program received signal SIGSEGV, Segmentation fault.
46328 0x080d4a2d in xf86RandR12ScreenSetSize (pScreen=0x8dca3a0, width=800,
46329 height=600, mmWidth=210, mmHeight=157)
46330 at ../../../../hw/xfree86/modes/xf86RandR12.c:731
46331 731 ../../../../hw/xfree86/modes/xf86RandR12.c: No such file or directory.
46332 in ../../../../hw/xfree86/modes/xf86RandR12.c
46333 (gdb) bt full
46334 height=600, mmWidth=210, mmHeight=157)
46335 at ../../../../hw/xfree86/modes/xf86RandR12.c:731
46336 randrp = 0x8dcae68
46337 pScrn = 0x8dbeb28
46338 config = <value optimized out>
46339 pRoot = 0x8e08e30
46340 pScrnPix = 0xb6d12008
46341 ret = 1
46342 c = <value optimized out>
46343 mmWidth=210, mmHeight=157) at ../../randr/rrscreen.c:185
46344 No locals.
46345 at ../../randr/rrscreen.c:307
46346 pWin = 0x8e08e30
46347 pScreen = 0x8dca3a0
46348 i = <value optimized out>
46349 rc = 0
46350 ../../randr/randr.c:485
46351
46352 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
46353 Cc: Julien Cristau <jcristau@debian.org>
46354 Tested-by: Julien Cristau <jcristau@debian.org>
46355 Cc: Keith Packard <keithp@keithp.com>
46356 Reviewed-by: Keith Packard <keithp@keithp.com>
46357 Signed-off-by: Keith Packard <keithp@keithp.com>
46358
46359commit 2f0abd7d690c3bcce2a80c093a0880f43e56d8cb
46360Author: Kristian Høgsberg <krh@bitplanet.net>
46361Date: Thu Sep 23 09:04:11 2010 -0400
46362
46363 glx: Fix use after free in DrawableGone
46364
46365 (note that nearly identical patches were proposed by Kristian and Chris)
46366
46367 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
46368 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
46369 Reported-by: Julien Cristau <jcristau@debian.org>
46370 Reviewed-by: Adam Jackson <ajax@redhat.com>
46371 Reviewed-by: Jamey Sharp <jamey@minilop.net>
46372 Tested-by: Chris Wilson <chris@chris-wilson.co.uk>
46373 Signed-off-by: Keith Packard <keithp@keithp.com>
46374
46375commit e416965bfd08dfae5d2b7932b5118efac20546ad
46376Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46377Date: Sun Jul 18 18:56:02 2010 +0300
46378
46379 dix: remove proc tables initialization
46380
46381 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46382 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46383
46384commit cbd4d5dbb70db62ba1cb79c7b904e6fa11f62d7e
46385Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46386Date: Fri Jul 16 20:19:50 2010 +0300
46387
46388 xserver: delete pervasively use of DISPATCH_PROC
46389
46390 Some functions had to be moved around due some missing static definitions.
46391 Another minor clean up like inexistent function declarations and etc were made
46392 also.
46393
46394 Part of this patch was cooked using:
46395 sed -i -e '/static DISPATCH_PROC*.*;/d' `git ls-files`
46396
46397 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46398 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46399
46400commit 58bd317e29f4abf7f950891339d2a6a78ddf7903
46401Author: Macpaul Lin <macpaul@andestech.com>
46402Date: Tue Sep 21 15:13:34 2010 +0800
46403
46404 xfree86: nds32: add nds32 support for compiler related mmio codes
46405
46406 Add nds32 support for compiler related mmio codes.
46407 It includes byte-swap or non-swap operations.
46408
46409 Signed-off-by: Macpaul Lin <macpaul@andestech.com>
46410 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46411 Signed-off-by: Keith Packard <keithp@keithp.com>
46412
46413commit 28e6de66b42062a885ba38416387f2dbc15707fd
46414Author: Macpaul Lin <macpaul@andestech.com>
46415Date: Tue Sep 21 15:13:33 2010 +0800
46416
46417 xfree86: nds32: add nds32 support for compiler specific codes
46418
46419 Add nds32 definitions and related assembly codes to compiler header files.
46420
46421 Signed-off-by: Macpaul Lin <macpaul@andestech.com>
46422 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46423 Signed-off-by: Keith Packard <keithp@keithp.com>
46424
46425commit 2b24b2bd859a4b4b8ccaf3b7d1b529dcda2d3b94
46426Author: Macpaul Lin <macpaul@andestech.com>
46427Date: Tue Sep 21 15:13:32 2010 +0800
46428
46429 xfree86: nds32: add nds32 definition for vgaHW support.
46430
46431 Add __nds32__ definitions for vgaHW support.
46432
46433 Signed-off-by: Macpaul Lin <macpaul@andestech.com>
46434 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46435 Signed-off-by: Keith Packard <keithp@keithp.com>
46436
46437commit 22376aa7aa44464636b12ee1d61b94e52ca02b0f
46438Author: Macpaul Lin <macpaul@andestech.com>
46439Date: Tue Sep 21 15:13:31 2010 +0800
46440
46441 xfree86: nds32: add nds32 definition for support linux video related io.
46442
46443 Add __nds32__ definitions for supporing correct io method for lnx_video.c
46444
46445 Signed-off-by: Macpaul Lin <macpaul@andestech.com>
46446 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46447 Signed-off-by: Keith Packard <keithp@keithp.com>
46448
46449commit d52799f67f590b0f91b521cd2bdf5a5a8a229c96
46450Author: Macpaul Lin <macpaul@andestech.com>
46451Date: Tue Sep 21 15:13:30 2010 +0800
46452
46453 xfree86: nds32: add nds32 related definitions into include headers.
46454
46455 Add MSB/LSB related definitions into include/servermd.h
46456
46457 Signed-off-by: Macpaul Lin <macpaul@andestech.com>
46458 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46459 Signed-off-by: Keith Packard <keithp@keithp.com>
46460
46461commit 98197d931b266674557b52b4a7099c6470114e55
46462Merge: 01ad372 402942c
46463Author: Keith Packard <keithp@keithp.com>
46464Date: Thu Sep 23 15:30:37 2010 -0700
46465
46466 Merge remote branch 'jamey/for-keith'
46467
46468commit 682a3ee60867da027fe1bcda0c8587dd3db9d58e
46469Author: Adam Jackson <ajax@redhat.com>
46470Date: Sat Sep 18 07:27:13 2010 -0400
46471
46472 xfree86: Remove useless module setup functions where appropriate
46473
46474 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46475 Reviewed-by: Julien Cristau <jcristau@debian.org>
46476 Signed-off-by: Adam Jackson <ajax@redhat.com>
46477
46478commit 6130170e7e9b64c611ee942ec3455dd1a185193d
46479Author: Adam Jackson <ajax@redhat.com>
46480Date: Sat Sep 18 06:41:35 2010 -0400
46481
46482 xfree86: Remove unused refcounting from input drivers
46483
46484 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46485 Reviewed-by: Julien Cristau <jcristau@debian.org>
46486 Signed-off-by: Adam Jackson <ajax@redhat.com>
46487
46488commit ab7f057ce9df4e905b12cebc1e587b9a7f200418
46489Author: Adam Jackson <ajax@redhat.com>
46490Date: Fri Sep 17 11:19:17 2010 -0400
46491
46492 loader: Remove a silly layer of reference counting
46493
46494 libdl will refcount objects for us just fine, thanks.
46495
46496 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46497 Reviewed-by: Julien Cristau <jcristau@debian.org>
46498 Signed-off-by: Adam Jackson <ajax@redhat.com>
46499
46500commit 0438002cd238011f33dbf176c555de5e380ba77e
46501Author: Adam Jackson <ajax@redhat.com>
46502Date: Fri Sep 17 09:13:04 2010 -0400
46503
46504 loader: include cleanup
46505
46506 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46507 Reviewed-by: Julien Cristau <jcristau@debian.org>
46508 Signed-off-by: Adam Jackson <ajax@redhat.com>
46509
46510commit 2f003fe49697f8343c052dff2b3752eace814ea7
46511Author: Adam Jackson <ajax@redhat.com>
46512Date: Fri Sep 17 09:02:13 2010 -0400
46513
46514 loader: Remove the handle field from LoaderOpen
46515
46516 This was always 0 from all the callers.
46517
46518 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46519 Reviewed-by: Julien Cristau <jcristau@debian.org>
46520 Signed-off-by: Adam Jackson <ajax@redhat.com>
46521
46522commit 09929da50503b559459f5b722c11647c47857fdb
46523Author: Adam Jackson <ajax@redhat.com>
46524Date: Fri Sep 17 08:16:51 2010 -0400
46525
46526 loader: Remove unused canonical name field
46527
46528 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46529 Reviewed-by: Julien Cristau <jcristau@debian.org>
46530 Signed-off-by: Adam Jackson <ajax@redhat.com>
46531
46532commit 3a26e7f459764d4aee71b2d7e25b113b729b94ac
46533Author: Adam Jackson <ajax@redhat.com>
46534Date: Fri Sep 17 07:55:48 2010 -0400
46535
46536 loader: Remove unused module serial number
46537
46538 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46539 Reviewed-by: Julien Cristau <jcristau@debian.org>
46540 Signed-off-by: Adam Jackson <ajax@redhat.com>
46541
46542commit 2a24a013bf3f554bb03c0f5af155d23dbb27b599
46543Author: Adam Jackson <ajax@redhat.com>
46544Date: Fri Sep 17 07:32:17 2010 -0400
46545
46546 loader: Merge dlloader directly into the loader
46547
46548 This lets us drop some double-tracking of loaded modules too. If your
46549 OS is too lame to have libdl, fix that first.
46550
46551 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46552 Reviewed-by: Julien Cristau <jcristau@debian.org>
46553 Signed-off-by: Adam Jackson <ajax@redhat.com>
46554
46555commit 693e92d4049ee76d074737480f3fc1bb236f4278
46556Author: Peter Hutterer <peter.hutterer@who-t.net>
46557Date: Fri Sep 10 09:54:33 2010 +1000
46558
46559 xfree86: add xf86OptionListDuplicate()
46560
46561 Does what it says on the box.
46562
46563 Some drivers need to duplicate option lists from the original device to
46564 ensure that devices created by the driver (driver-internal hotplugging) have
46565 the same list of options as the original device.
46566
46567 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
46568 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
46569 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46570
46571commit 402942cdbc518395a2943fd226b9f3071f24d39f
46572Author: Jamey Sharp <jamey@minilop.net>
46573Date: Fri Sep 17 02:11:44 2010 +0200
46574
46575 Introduce per-object per-screen privates.
46576
46577 This replaces dixCreatePrivateKey and the only uses, which were in
46578 midispcur.
46579
46580 Commit by Jamey Sharp and Josh Triplett.
46581
46582 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46583 Signed-off-by: Josh Triplett <josh@joshtriplett.org>
46584 Reviewed-by: Keith Packard <keithp@keithp.com>
46585
46586commit e4d4d6ddd52801cf0b0b253d9ba3bdabfa9a9d8d
46587Author: Jamey Sharp <jamey@minilop.net>
46588Date: Fri Sep 17 02:18:10 2010 +0200
46589
46590 Xserver need not be compatible with old versions of xserver.
46591
46592 Delete time-traveling multiple personality disorder from the server.
46593
46594 Gaetan notes:
46595
46596 There were a couple of drivers containing an unknown version of the
46597 modes/parser code. This was done in server 1.2 time frame because it
46598 was released without mode code. It was barely or not maintained
46599 afterwards. There are currently no video drivers with a copy of the
46600 modes code.
46601
46602 Most of these ifdefs were introduced in commit
46603 a8d760f567b19268329c4682495caa591f08a854, where Aaron wrote,
46604
46605 This change uses XORG_VERSION_CURRENT < 7.0 to mean "server newer
46606 than 1.2" since XORG_VERSION current went backwards at some point.
46607
46608 Alan explains that:
46609
46610 In Xorg 1.3, when we first released an Xorg server release decoupled
46611 from the katamari release schedule. (1.0 through 1.2 were released
46612 as part of X11R7.0 through 7.2, while 1.3 came out between X11R7.2 &
46613 7.3.)
46614
46615 Commit by Jamey Sharp and Josh Triplett.
46616
46617 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46618 Signed-off-by: Josh Triplett <josh@joshtriplett.org>
46619 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
46620
46621commit 01ad3725a8d0a8958720b81d19598a08c072ccee
46622Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46623Date: Tue Jun 1 20:37:54 2010 +0300
46624
46625 xfree86: delete useless "Primary device is not PCI" message
46626
46627 The primary device being PCI or not has no effect on the server working. This
46628 message is superfluous.
46629
46630 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46631 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46632
46633commit 7360235ecca25af29bb2d99a8e46d5b259ad1b1e
46634Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46635Date: Mon May 31 18:20:57 2010 +0300
46636
46637 xfree86: bus: assign PCI_SOURCES with the actual source files
46638
46639 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46640 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46641
46642commit 18b2d2e8bd056266410226d779752e7933847b6b
46643Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46644Date: Mon May 31 18:09:50 2010 +0300
46645
46646 xfree86: bus: move macros from common PCI header to private file
46647
46648 Only int10/helper_exec.c is using them.
46649
46650 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46651 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46652
46653commit 17a23f9ba8fa95f3e15afba7de22a12fe60bacfb
46654Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46655Date: Fri Sep 10 19:18:42 2010 +0300
46656
46657 xfree86: dri: remove unused PCI macro
46658
46659 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46660 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46661
46662commit fc3ab84de7b5692f0db2b282ab0ed8e5a61d1fce
46663Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46664Date: Fri Sep 10 19:16:30 2010 +0300
46665
46666 xfree86: configure: move buses references to their own location
46667
46668 This patch makes xf86Configure.c free of PCI and SBUS code, moving to a more
46669 meaningful location.
46670
46671 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46672 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46673 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
46674
46675commit 49b817501f97d55480063c0b62544b3af75b4b7c
46676Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46677Date: Tue Jun 1 16:16:53 2010 +0300
46678
46679 xfree86: bus: remove xfree86 and few other references from OS files
46680
46681 Conceptually, os-support should have only a basic set of OS helpers which
46682 wouldn't mix with any DDX common structure (e.g. windowing structures, etc)
46683
46684 This patch removes some xfree86-only and a few other unused references from
46685 os-support/bus.
46686
46687 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46688 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46689
46690commit d9c682a5e250677883c4d375ca84cf184ee19d77
46691Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46692Date: Sun May 30 21:35:23 2010 +0300
46693
46694 xfree86: bus: remove useless automake variable
46695
46696 Leftover that was missed in f7abe05b.
46697
46698 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46699 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
46700 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46701
46702commit a7efeda8a56a540542e056a426e89cfc053a4162
46703Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46704Date: Sun May 30 19:12:31 2010 +0300
46705
46706 xfree86: use xf86PciIsolateDevice to get PCI config information
46707
46708 Make xf86IsolateDevice private on PCI common file.
46709
46710 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46711 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46712
46713commit c768cdda92696b636c10bb2df64167d5274b4b99
46714Merge: 5a72538 2079853
46715Author: Keith Packard <keithp@keithp.com>
46716Date: Mon Sep 13 16:48:33 2010 -0700
46717
46718 Merge remote branch 'jamey/reviewed'
46719
46720commit 20798536326f94ac9c97b4ec7a51329f7c912991
46721Author: Jamey Sharp <jamey@minilop.net>
46722Date: Wed Jul 14 13:14:28 2010 -0700
46723
46724 rootless: Delete dead remnants of rootless acceleration.
46725
46726 Jeremy deleted rootlessAccelInit's implementation in 2008 in commit
46727 587c010a1cd733fded4d49dc339df0634bda8be6. Delete its prototype and the
46728 remaining commented-out call to it.
46729
46730 It still makes sense for the rootless GC ops to relax the planemask, but
46731 that's independent of the size of the operation, so quit checking the
46732 thresholds there.
46733
46734 FillBytes and CompositePixels are not called anywhere, so delete
46735 everything related to both.
46736
46737 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46738 Cc: Jeremy Huddleston <jeremyhu@apple.com>
46739 Cc: Jon TURNEY <jon.turney@dronecode.org.uk>
46740 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46741
46742commit 58d1ec0fe84412c66f1ea1de15e3901c0b120164
46743Author: Jamey Sharp <jamey@minilop.net>
46744Date: Wed Jul 14 11:30:17 2010 -0700
46745
46746 rootless: ROOTLESS_GLOBAL_COORDS is always set, so unifdef it.
46747
46748 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46749 Cc: Jeremy Huddleston <jeremyhu@apple.com>
46750 Cc: Jon TURNEY <jon.turney@dronecode.org.uk>
46751 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46752
46753commit 35ce892a705f9377a36f19414e5a25dcbe44a354
46754Author: Jamey Sharp <jamey@minilop.net>
46755Date: Wed Jul 14 11:19:41 2010 -0700
46756
46757 rootless: ROOTLESS_TRACK_DAMAGE is never set, so unifdef it.
46758
46759 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46760 Cc: Jeremy Huddleston <jeremyhu@apple.com>
46761 Cc: Jon TURNEY <jon.turney@dronecode.org.uk>
46762 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46763
46764commit ccdf15689050a0f988721c91fab75861ec4d339c
46765Author: Jamey Sharp <jamey@minilop.net>
46766Date: Mon May 31 16:01:18 2010 -0700
46767
46768 miPolyGlyphBlt: Delete redundant ValidateGC.
46769
46770 The caller is required to have validated the GC for the drawable before
46771 calling any ops, including PolyGlyphBlt.
46772
46773 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46774 Reviewed-by: Keith Packard <keithp@keithp.com>
46775
46776commit 3fd963de27f3c4452c320e617ad33b91b0967737
46777Author: Jamey Sharp <jamey@minilop.net>
46778Date: Wed May 19 14:16:20 2010 -0700
46779
46780 Un-export CreateScratchGC now that nothing external uses it.
46781
46782 The server and drivers sometimes use GetScratchGC, but never
46783 CreateScratchGC.
46784
46785 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46786 Reviewed-by: Keith Packard <keithp@keithp.com>
46787
46788commit 0f48e362c5baee2bc281a4b88cfb68de24172d15
46789Author: Jamey Sharp <jamey@minilop.net>
46790Date: Wed May 19 14:03:19 2010 -0700
46791
46792 glxdriswrast: Use CreateGC, not CreateScratchGC.
46793
46794 Since this code was using CreateScratchGC and not GetScratchGC,
46795 FreeScratchGC would always call FreeGC, so just call it directly.
46796
46797 As long as the drawable provided to CreateGC has the same screen and
46798 depth as were passed to CreateScratchGC, these functions are basically
46799 identical. There are two differences:
46800
46801 - CreateGC gives you a non-null stipple. You probably don't care.
46802
46803 - CreateGC passes the gcid and client to XACE. There are several
46804 internal GCs allocated in the server, and they all pass 0 and
46805 serverClient. I expect XACE will never reject that combination.
46806
46807 The callers of createDrawable all verify that pDraw has the same screen
46808 that driScreen does. In short, this should have no behavioral change.
46809
46810 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46811 Cc: Kristian Høgsberg <krh@bitplanet.net>
46812 Reviewed-by: Keith Packard <keithp@keithp.com>
46813
46814commit af2a94e12fe4e426dfca461a0d14526193c01cef
46815Author: Jamey Sharp <jamey@minilop.net>
46816Date: Tue Jul 13 13:33:59 2010 -0700
46817
46818 VGA arbiter: No need for arbitration around CreateGC.
46819
46820 None of XAA, EXA, or UXA do any hardware access during CreateGC, so they
46821 don't need VGA arbitration. I haven't found any open source drivers that
46822 hook CreateGC, so they're safe. I'd be surprised if any driver directly
46823 hooks CreateGC and does hardware access from it and needs VGA
46824 arbitration.
46825
46826 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46827 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46828
46829commit 7a63b2db636f01694cb54726bbb04287718e286a
46830Author: Jamey Sharp <jamey@minilop.net>
46831Date: Sun Jul 11 20:36:34 2010 -0700
46832
46833 Delete Colormap->devPriv. It isn't used any more.
46834
46835 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46836 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46837
46838commit 2e8069dd36c2f5caf868937405fc03328004a85d
46839Author: Jamey Sharp <jamey@minilop.net>
46840Date: Sun Jul 11 20:35:43 2010 -0700
46841
46842 xnest: Replace use of Colormap->devPriv with standard devPrivates.
46843
46844 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46845 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46846
46847commit 34d6492436c0f4a81822769b24d8c7816d0226ad
46848Author: Jamey Sharp <jamey@minilop.net>
46849Date: Sat Jul 10 06:55:58 2010 -0700
46850
46851 xquartz: Delete compile-time-constant DRIGCRec.driOps field.
46852
46853 The driOps field of this structure is never set to anything but
46854 &driGCOps, and this structure is not visible outside this source file.
46855 Just use the constant in the one place the field was used.
46856
46857 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46858 Cc: Jeremy Huddleston <jeremyhu@apple.com>
46859 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
46860
46861commit 2051e0f371de892b2e598b30c18dae96b58d68de
46862Author: Jamey Sharp <jamey@minilop.net>
46863Date: Thu Jul 8 02:38:13 2010 -0700
46864
46865 dixfonts: Deobfuscate GC ops calls.
46866
46867 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46868 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
46869
46870commit bbbf795e6b1ac048ea17164ba1089055cfe4306d
46871Author: Jamey Sharp <jamey@minilop.net>
46872Date: Sun May 23 17:15:43 2010 -0700
46873
46874 Delete redundant GC initializations.
46875
46876 When a GC is allocated, it is zeroed, including all storage requested
46877 with dixRegisterPrivateKey. So CreateGC hooks don't need to initialize
46878 anything to zero.
46879
46880 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46881 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
46882
46883commit 3288232f3d2a0239944bd856101c2d16c713ef40
46884Author: Jamey Sharp <jamey@minilop.net>
46885Date: Sun Jun 13 18:10:38 2010 -0700
46886
46887 Delete xaaWrapper.
46888
46889 This was part of "An experimental pseudocolor emulation layer. Not fully
46890 completed, currently only works for 16bpp." Only neomagic tried to use
46891 it, and that was neutered by the removal of the fbpseudocolor portion of
46892 that emulation layer; the rest is easily removed.
46893
46894 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46895 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
46896
46897commit f856dcdc3a7b635dd94d3dac387ddec952e4880b
46898Author: Jamey Sharp <jamey@minilop.net>
46899Date: Sat Jun 12 08:42:31 2010 -0700
46900
46901 Delete unused lastWinOrg field from GCs.
46902
46903 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46904 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46905
46906commit b2184024361717358e35a56b084c4a71cbc29234
46907Author: Jamey Sharp <jamey@minilop.net>
46908Date: Sat Jun 12 16:20:00 2010 -0700
46909
46910 Delete unused devPrivate field from GCFuncs and GCOps.
46911
46912 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46913 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46914
46915commit 139aa6e74227f0fe52cd8dccf1d563664103aaba
46916Author: Jamey Sharp <jamey@minilop.net>
46917Date: Sun Jun 13 18:57:08 2010 -0700
46918
46919 damage: Delete NOTUSED block--it was never not NOTUSED.
46920
46921 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46922 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46923
46924commit 5ff9e58bef668c1279c5e253795711a1e118aaa1
46925Author: Jamey Sharp <jamey@minilop.net>
46926Date: Sat Jun 12 08:51:04 2010 -0700
46927
46928 fb: Delete unused oneRect private field.
46929
46930 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46931 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
46932
46933commit e99bbfa0a24ec038b88e81881133a676bf63580e
46934Author: Jamey Sharp <jamey@minilop.net>
46935Date: Tue May 25 13:50:11 2010 -0700
46936
46937 dmx: Delete '#undef Xmalloc' and friends.
46938
46939 Nothing in dmx uses these names any more, and it builds cleanly without
46940 trying to undef them.
46941
46942 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46943 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
46944
46945commit b5217bfdca67ef2c5323838afcfcd89eba8899d5
46946Author: Jamey Sharp <jamey@minilop.net>
46947Date: Tue May 25 13:28:32 2010 -0700
46948
46949 dmx: __glXMalloc -> malloc, etc.
46950
46951 Kristian made equivalent edits to the reference GLX implementation in
46952 2006, with commit 2d2d38d17cc2558f8a41166a4a1578bc4c663c37.
46953
46954 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46955 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
46956
46957commit a715de7f11afeda7798f0882148a94d4db2291e6
46958Author: Jamey Sharp <jamey@minilop.net>
46959Date: Tue May 25 13:10:05 2010 -0700
46960
46961 dmx: Delete unused GLX visual matching code.
46962
46963 Kristian deleted this API from the rest of the server in 2007, in commit
46964 7d74690536b64f7b8e8036507ab7790807349c50.
46965
46966 Signed-off-by: Jamey Sharp <jamey@minilop.net>
46967 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
46968
46969commit 5a725385fbb93553931ca0217e79271ab2a2c613
46970Author: Tiago Vignatti <tiago.vignatti@nokia.com>
46971Date: Mon Sep 13 12:33:53 2010 +0300
46972
46973 xfree86: vgaarb: fix decoding stub API
46974
46975 This fix a problem introduced in commit 65466652. It closes also:
46976 https://bugs.freedesktop.org/show_bug.cgi?id=30160
46977
46978 Reported-by: <xunx.fang@intel.com>
46979 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
46980 Signed-off-by: Keith Packard <keithp@keithp.com>
46981
46982commit 0a3e6b0345a96aae278dd776c00b5a7c7ccf65af
46983Author: Jeremy Huddleston <jeremyhu@apple.com>
46984Date: Sat Sep 11 14:25:21 2010 -0700
46985
46986 XQuartz: Localization updates
46987
46988 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
46989
46990commit 4e09a69c5b1f3c8d6f2f92ec24aaab3a1119940c
46991Author: Jeremy Huddleston <jeremyhu@apple.com>
46992Date: Sun Aug 15 20:07:28 2010 -0700
46993
46994 rootless: Remove ROOTLESS_WORKAROUND
46995
46996 This was already removed for XWin (20701522be803fe47e921fcf059dadf64c7f287d)
46997 with no reported side effects. XQuartz seems to be behaving ok without it as
46998 well. While this possibly brings back bug #1168, we don't have any reproduction
46999 steps for that issue, and if it crops up again, we should fix it a real way
47000 rather than this hokey workaround which doesn't even work for COMPOSITE.
47001
47002 This effectively reverts the following two changes:
47003 b2135e589baeb2ea26da50b9167feaea23bcce3c
47004 d7fef52254126aa5897a5c58faeda1f61d5b13d8
47005
47006 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
47007 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
47008
47009commit 63b7d9448abbded1cb91e82a13319d5e4313ca0c
47010Author: Tiago Vignatti <tiago.vignatti@nokia.com>
47011Date: Sun May 30 18:44:06 2010 +0300
47012
47013 xfree86: fbdevhw: remove unnecessary debug code
47014
47015 It very likely no one will want to print all functions of this file for
47016 debugging purposes. If this is the case, then a mix of ctags + cpp + gdb
47017 can do the same job.
47018
47019 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47020 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
47021
47022commit 66105d8281f8a6442cf7fee5d10e8b8a34bef952
47023Author: Tiago Vignatti <tiago.vignatti@nokia.com>
47024Date: Sun May 30 18:16:32 2010 +0300
47025
47026 xfree86: fbdevhw: remove unused cmap fields from main structure
47027
47028 It was never used since first git revision and probably no one cares for it.
47029
47030 ABI break.
47031
47032 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47033 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
47034
47035commit 317e491e067e31ea7cf8cb7301e47de0c10f119e
47036Author: Tiago Vignatti <tiago.vignatti@nokia.com>
47037Date: Mon Jul 19 15:44:39 2010 +0300
47038
47039 xext: remove unused header
47040
47041 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47042 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
47043
47044commit 5098801b3a570d0bb3386d7d210f21a1cbf3bf12
47045Author: Tiago Vignatti <tiago.vignatti@nokia.com>
47046Date: Sat Jun 12 16:57:50 2010 +0300
47047
47048 mi: delete unused code
47049
47050 pBSReg is always NULL, so the statement after the conditional will never be
47051 reached.
47052
47053 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47054 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
47055
47056commit 7d8cabd027cfc55e9921307e5b3dd7052de8496f
47057Author: Tiago Vignatti <tiago.vignatti@nokia.com>
47058Date: Wed Aug 4 16:12:59 2010 +0300
47059
47060 os: simplify smart scheduler init process
47061
47062 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47063 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
47064 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
47065
47066commit d9c18c3b9badf38d499e3f53291bd23fda3c4340
47067Author: Tiago Vignatti <tiago.vignatti@nokia.com>
47068Date: Wed Aug 4 15:54:48 2010 +0300
47069
47070 os: remove useless smart scheduler macros
47071
47072 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47073 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
47074 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
47075
47076commit a9e6080dc94633a4f497535f619a14fd8b566645
47077Author: Tiago Vignatti <tiago.vignatti@nokia.com>
47078Date: Fri Sep 10 15:01:29 2010 +0300
47079
47080 os/xfree86: remove macro checking for POSIX symbols
47081
47082 We assume already that our X implementation is POSIX compliant anyway. So
47083 remove those redundant checking.
47084
47085 SA_SIGINFO is left there.
47086
47087 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47088 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
47089 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
47090
47091commit 648c8871c92727d7b6b16859f27f12266a06a16e
47092Author: Adam Jackson <ajax@redhat.com>
47093Date: Wed Sep 8 16:08:24 2010 -0400
47094
47095 composite: Convert compWindowUpdate to use TraverseTree
47096
47097 v2: Use != RedirectDrawNone, since we aren't called for manual windows.
47098
47099 Signed-off-by: Adam Jackson <ajax@redhat.com>
47100 Reviewed-by: Keith Packard <keithp@keithp.com>
47101 Signed-off-by: Keith Packard <keithp@keithp.com>
47102
47103commit bbffb85461eb63bfb1e01e2cb9674607b9221604
47104Merge: 9a1a4cc 71972c2
47105Author: Keith Packard <keithp@keithp.com>
47106Date: Fri Sep 10 11:55:34 2010 -0700
47107
47108 Merge remote branch 'whot/for-keith'
47109
47110commit 9a1a4ccbe026f34beca5db3b2243874906961b49
47111Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
47112Date: Mon Aug 30 16:37:05 2010 -0400
47113
47114 Add screens to the PRIVATE_XSELINUX set.
47115
47116 The SELinux extension does store a security label in the screen
47117 devPrivates. Fixes crash caused by overwriting another private.
47118
47119 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
47120 Reported-by: Justin Mattock <justinmattock@gmail.com>
47121 Signed-off-by: Keith Packard <keithp@keithp.com>
47122
47123commit 5b680f2c23fb3d35c721647d0334f63297589371
47124Author: Vignatti Tiago (Nokia-MS/Helsinki) <tiago.vignatti@nokia.com>
47125Date: Mon Aug 30 21:24:30 2010 +0300
47126
47127 os: add -sigstop option for Upstart (or equivalent) startup
47128
47129 This is very similar to the RunFromSmartParent (implicit) option, except
47130 we do not send the signal to our parent process, but our own process
47131 instead, and that signal is SIGSTOP, not SIGUSR1.
47132
47133 Upstart or a similar equivalent program will detect this, realize that
47134 we are ready to accept clients now, send us SIGCONT and move our job
47135 status from SPAWNED to RUNNING.
47136
47137 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
47138 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
47139 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47140 Signed-off-by: Keith Packard <keithp@keithp.com>
47141
47142commit ca0d578d29f37b61e62556fff59732741885625b
47143Merge: cf88363 08adf41
47144Author: Keith Packard <keithp@keithp.com>
47145Date: Fri Sep 10 11:50:27 2010 -0700
47146
47147 Merge remote branch 'mattst88/master'
47148
47149commit cf88363db0ebb42df7cc286b85d30d7898aea840
47150Author: Aaron Plattner <aplattner@nvidia.com>
47151Date: Fri Aug 27 10:20:29 2010 -0700
47152
47153 os: Return BadLength instead of disconnecting BigReq clients (#4565)
47154
47155 If a client sends a big request that's too big (i.e. bigger than
47156 maxBigRequestSize << 2 bytes), the server just disconnects it. This makes the
47157 client receive SIGPIPE the next time it tries to send something.
47158
47159 The X Test Suite sends requests that are too big when the test specifies the
47160 TOO_LONG test type. When the client receives SIGPIPE, XTS marks it as
47161 UNRESOLVED, which counts as a failure.
47162
47163 Instead, remember how long the request is supposed to be and then return that
47164 size. Dispatch() checks the length and sends BadLength to the client. Then,
47165 whenever oci->ignoreBytes is nonzero, ignore the data read instead of trying to
47166 process it as a request.
47167
47168 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
47169 Reviewed-by: Keith Packard <keithp@keithp.com>
47170 Signed-off-by: Keith Packard <keithp@keithp.com>
47171
47172commit fa22f97af65efc2a147e9be0b7f288848965ff60
47173Merge: 6edbdb7 6546665
47174Author: Keith Packard <keithp@keithp.com>
47175Date: Fri Sep 10 11:47:41 2010 -0700
47176
47177 Merge remote branch 'vignatti/vgaarb-fixes'
47178
47179commit 6edbdb7f9e02ecb069c6fe6c0600fc389553ec5d
47180Author: Aaron Plattner <aplattner@nvidia.com>
47181Date: Mon Aug 23 11:37:46 2010 -0700
47182
47183 Tag sdksyms.c entries with source file and line number
47184
47185 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
47186 Reviewed-by: Adam Jackson <ajax@redhat.com>
47187 Signed-off-by: Keith Packard <keithp@keithp.com>
47188
47189commit 9ef6241c2382bfc555284a4985f6d1e37d750d6f
47190Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
47191Date: Mon Aug 23 16:04:03 2010 -0400
47192
47193 Fix property and selection devPrivate allocation.
47194
47195 Selection objects were not being allocated with privates, and both
47196 objects had a stray statement that zeroed out the devPrivates field.
47197
47198 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
47199 Reported-by: Justin Mattock <justinmattock@gmail.com>
47200 Reviewed-by: Adam Jackson <ajax@redhat.com>
47201 Signed-off-by: Keith Packard <keithp@keithp.com>
47202
47203commit 9dca441670d261a9a9fb6108960ed48f3d58fb7f
47204Author: Peter Hutterer <peter.hutterer@who-t.net>
47205Date: Mon Sep 6 11:32:38 2010 +1000
47206
47207 xfree86: add a hook to replace the new console handler.
47208
47209 This hook is only necessary for the keyboard driver to remove the race
47210 condition between drain_console() and the driver's ReadInput (Bug 29969).
47211
47212 The idea is that a driver that needs to handle events from the console
47213 calls xf86ReplaceConsoleHandler() with it's own ReadInput (or NULL) and thus
47214 removes the drain_console call. It's the driver's responsibility to restore
47215 the previous behaviour when the driver is unloaded.
47216
47217 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47218 CC: Thomas Hellstrom <thellstrom@vmware.com>
47219 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47220 Reviewed-by: Adam Jackson <ajax@redhat.com>
47221
47222commit da31ca747f8e3993b2aca5d31b7d09b9907ad1d9
47223Author: Thomas Hellstrom <thellstrom@vmware.com>
47224Date: Thu Sep 2 10:53:28 2010 +0200
47225
47226 linux: Don't lose console events on non-evdev drivers (#29969)
47227
47228 The drain_console() function will race with new keyboard events being added
47229 by the hardware causing the server to lose keyboard events if the console fd
47230 is used for input.
47231
47232 Only use the drain_console() when AllowEmptyInput is off which is the best
47233 indicator we have for whether the keyboard driver will be used. This patch
47234 will only fix the bug when hotplugging is disabled.
47235 What we really need is a way to figure out either whether we're _not_ using
47236 the keyboard driver (not predictable) or a way for the keyboard driver to
47237 disable drain_console().
47238
47239 X.Org Bug 29969 <http://bugs.freedesktop.org/show_bug.cgi?id=29969>
47240
47241 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
47242 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47243 Reviewed-by: Adam Jackson <ajax@redhat.com>
47244 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47245 (cherry picked from commit 71972c2534d490284d3d42b456c2f34b964b2894)
47246
47247commit 52df92a563980a7af0fb61248da3654bd90e8232
47248Author: Peter Hutterer <peter.hutterer@who-t.net>
47249Date: Thu Sep 2 15:08:01 2010 +1000
47250
47251 xfree86: move XI_VERIFY_VALUATORS to the source file it's used in.
47252
47253 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47254 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47255
47256commit 824e970c5cc94fc7d5fe5f78a24d0d3057d87a84
47257Author: Peter Hutterer <peter.hutterer@who-t.net>
47258Date: Thu Sep 2 15:03:39 2010 +1000
47259
47260 xfree86: remove TS_Raw and TS_Scaled defines.
47261
47262 There are no references to it other than the commit that added them. But
47263 since we're re-doing the API anyway, now is a good time to break things.
47264
47265 commit 9398d62f27ee1b287e4458fd8b011c10f7b59efd
47266 Author: Daniel Stone <daniel@fooishbar.org>
47267 Date: Wed Mar 21 00:18:24 2007 +0200
47268
47269 XFree86 input: Add backwards compatibility for motion history
47270 Add the old motion history API back, as a shim around the new mi
47271 API.
47272
47273 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47274 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47275
47276commit ac3f88e604212fc3668d623798d50aa218c13995
47277Author: Peter Hutterer <peter.hutterer@who-t.net>
47278Date: Thu Sep 2 15:01:26 2010 +1000
47279
47280 xfree86: remove {Dont}SendCoreEvents defines
47281
47282 The input drivers that use it only do so with ABI 0 and we're long past this
47283 one now. Input driver don't have a say in whether they send core events now
47284 anyway.
47285
47286 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47287 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47288
47289commit 3f264149ff226d7c0bfcfcfc2c8845fa6326d0bd
47290Author: Peter Hutterer <peter.hutterer@who-t.net>
47291Date: Thu Sep 2 14:52:30 2010 +1000
47292
47293 xfree86: don't export xf86InputDevs.
47294
47295 Use xf86FirstLocalDevice() instead (but don't get me started on the naming
47296 of that one...)
47297
47298 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47299 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47300 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47301
47302commit 8f0531bddac8e774aa25b51155445309219da0f6
47303Author: Peter Hutterer <peter.hutterer@who-t.net>
47304Date: Thu Sep 2 14:46:59 2010 +1000
47305
47306 xfree86: remove XI_PRIVATE macro, unused.
47307
47308 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47309 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47310 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47311
47312commit c08f7afdbcf16d2ba30b67e27162c5aaa1ed3f61
47313Author: Peter Hutterer <peter.hutterer@who-t.net>
47314Date: Thu Sep 2 14:45:03 2010 +1000
47315
47316 xfree86: update comment for xf86DeleteInput
47317
47318 We have a driver hook - it's UnInit.
47319
47320 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47321 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47322 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47323
47324commit b8d9c5ff0003f15720737949e65e3159c4f4576e
47325Author: Peter Hutterer <peter.hutterer@who-t.net>
47326Date: Thu Sep 2 14:30:11 2010 +1000
47327
47328 xfree86: remove IDevRec, replace with InputInfoRec.
47329
47330 This struct is superfluous, maintaining the same info as the InputInfoRec
47331 (with the exception of the driver name).
47332
47333 This is a rather large commit with the majority of changes being a rename
47334 from the fields of the IDevRec (idev, commonOptions) to the InputInfoRec
47335 (pInfo, options).
47336
47337 The actual changes affect the initialization process of the input device:
47338 In NewInputDeviceRequest, the InputInfoRec is now always allocated and just
47339 added to the internal list in xf86NewInputDevice() if the init process
47340 succeeded.
47341
47342 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47343 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47344
47345commit c5da32a6105b3cf57b5996971149c39103cd64b3
47346Author: Peter Hutterer <peter.hutterer@who-t.net>
47347Date: Thu Sep 2 14:26:20 2010 +1000
47348
47349 xfree86: plug minor memory leak.
47350
47351 When no identifier for the device was specified, the allocated IDevRec (and
47352 its associated fields) need to be freed.
47353
47354 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47355 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47356 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47357
47358commit 34ade08174bfd652cc79bf26d2ad6a10ca73d2c4
47359Author: Peter Hutterer <peter.hutterer@who-t.net>
47360Date: Thu Sep 2 14:04:50 2010 +1000
47361
47362 xfree86: Remove extraOptions parameter from xf86CollectInputOptions.
47363
47364 And unexport it, drivers don't need to call this in the new init process.
47365
47366 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47367 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47368
47369commit 2199842ed50b3eb40d54146827fc58cae7e873ec
47370Author: Peter Hutterer <peter.hutterer@who-t.net>
47371Date: Thu Sep 2 10:52:54 2010 +1000
47372
47373 xfree86: remove extraOptions field from IDevRec.
47374
47375 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47376 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47377
47378commit 5ab8ca0807f9d2f8c8de2f9a1421e59e2a8e976c
47379Author: Peter Hutterer <peter.hutterer@who-t.net>
47380Date: Thu Sep 2 10:41:27 2010 +1000
47381
47382 xfree86: remove some ifdef 0 code.
47383
47384 InputInfoRec hasn't had a free function pointer since the git import.
47385
47386 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47387 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47388 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47389
47390commit b0da1bc509446a714eb2b29cbd35d43b1e5187ed
47391Author: Peter Hutterer <peter.hutterer@who-t.net>
47392Date: Thu Sep 2 10:39:24 2010 +1000
47393
47394 xfree86: move a declaration down to the block it is used in.
47395
47396 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47397 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47398 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47399
47400commit cd7059db4a766bc9d03d7968aebeba8542e6c6c0
47401Author: Peter Hutterer <peter.hutterer@who-t.net>
47402Date: Thu Sep 2 10:36:47 2010 +1000
47403
47404 xfree86: minor comment fix. HAL → config backend
47405
47406 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47407 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47408 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47409
47410commit f3861522fe7d148fd4b0802f2f9ea5d98ab2db93
47411Author: Peter Hutterer <peter.hutterer@who-t.net>
47412Date: Thu Sep 2 11:47:30 2010 +1000
47413
47414 xfree86: skip a few NULL initialization, calloc does it for us.
47415
47416 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47417 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47418 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47419
47420commit 12170978617c6862fb6aa286b18aaacd0a4e2737
47421Author: Peter Hutterer <peter.hutterer@who-t.net>
47422Date: Wed Sep 1 15:44:44 2010 +1000
47423
47424 xfree86: remove conversion procs and close proc from InputInfoRec.
47425
47426 None of them are called by the server.
47427
47428 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47429 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47430 Reviewed-by: Fernando Carrijo <fcarrijo@freedesktop.org>
47431
47432commit 71972c2534d490284d3d42b456c2f34b964b2894
47433Author: Thomas Hellstrom <thellstrom@vmware.com>
47434Date: Thu Sep 2 10:53:28 2010 +0200
47435
47436 linux: Don't lose console events on non-evdev drivers (#29969)
47437
47438 The drain_console() function will race with new keyboard events being added
47439 by the hardware causing the server to lose keyboard events if the console fd
47440 is used for input.
47441
47442 Only use the drain_console() when AllowEmptyInput is off which is the best
47443 indicator we have for whether the keyboard driver will be used. This patch
47444 will only fix the bug when hotplugging is disabled.
47445 What we really need is a way to figure out either whether we're _not_ using
47446 the keyboard driver (not predictable) or a way for the keyboard driver to
47447 disable drain_console().
47448
47449 X.Org Bug 29969 <http://bugs.freedesktop.org/show_bug.cgi?id=29969>
47450
47451 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
47452 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47453 Reviewed-by: Adam Jackson <ajax@redhat.com>
47454 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47455
47456commit e00e2e7b68fbc932269d607ac5dc2c441d07ad9d
47457Author: Peter Hutterer <peter.hutterer@who-t.net>
47458Date: Mon Aug 23 08:35:01 2010 +1000
47459
47460 xfree86: Check for existence of button class before dereferencing it.
47461
47462 The Irxon Super Mini Bluetooth Wireless Keyboard for PC/PDA/Cell Phones
47463 keyboards have axes but not buttons. The evdev driver doesn't set up a
47464 button class for these keyboards and a motion event handled by
47465 DGAProcessPointerEvent dereferences the dev->button NULL pointer, causing a
47466 server crash.
47467
47468 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47469 Reviewed-by: Keith Packard <keithp@keithp.com>
47470
47471commit 9802cca816884e7f055f054ac1ba9194e0eba10b
47472Author: Peter Hutterer <peter.hutterer@who-t.net>
47473Date: Wed Sep 1 15:42:44 2010 +1000
47474
47475 Remove atom field from InputInfoRec.
47476
47477 This field was only used in one location where we can use a local variable.
47478
47479 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47480 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47481
47482commit 79ea9ef39971d008d199b18f34d1aef2bab6e33f
47483Author: Peter Hutterer <peter.hutterer@who-t.net>
47484Date: Tue Aug 10 15:19:20 2010 +1000
47485
47486 input: constify valuators passed in by input drivers.
47487
47488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47489 Reviewed-by: Keith Packard <keithp@keithp.com>
47490 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
47491
47492commit 86560b5d05f14bdf04d21b3457a66c0d5045db9c
47493Author: Peter Hutterer <peter.hutterer@who-t.net>
47494Date: Tue Aug 24 13:46:35 2010 +1000
47495
47496 dix: don't set time to CurrentTime in DeviceChangedEvents.
47497
47498 CurrentTime is used by clients to skip setting the time, but not by the
47499 server.
47500
47501 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47502 Reviewed-by: Keith Packard <keithp@keithp.com>
47503
47504commit ff055506f0cbb852bed17acb9f9bbf1d715a854e
47505Author: Peter Hutterer <peter.hutterer@who-t.net>
47506Date: Mon Aug 23 13:20:09 2010 +1000
47507
47508 dix: fix crash when removing devices on a buttonless MD pointer (#29669)
47509
47510 If the master does not have a button class, recalculating the number of
47511 buttons required for this master dereferences a NULL pointer. Guard against
47512 this, if the master pointer doesn't have a button class, it doesn't need to
47513 update it's number of buttons.
47514
47515 Reproducible:
47516 Two devices on the same master, device NB with axes but no buttons, device
47517 A+B with axes and button .
47518 If NB was the last one to send an event through the master when A+B is
47519 removed from the server, master->button is NULL and leads to the above
47520 NULL-pointer dereference.
47521
47522 X.Org Bug 29669 <http://bugs.freedesktop.org/show_bug.cgi?id=29669>
47523
47524 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47525 Reviewed-by: Keith Packard <keithp@keithp.com>
47526
47527commit be978c7fc97e98ef1bc7b20c6c0740d8a81b609a
47528Author: Jesse Adkins <jesserayadkins@gmail.com>
47529Date: Tue Aug 3 18:21:28 2010 -0700
47530
47531 xfree86: Purge kbdCustomKeycodes from xf86Info.
47532
47533 This was obsolete after 3eeb62e8f587732e6b433c2b9c6879eb26a3f1b4 "bug #890: completely remove deprecated keyboard driver".
47534
47535 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
47536 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47537 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47538
47539commit 6922b31a136b2b0ac185d61785969a11f84c7943
47540Author: Peter Hutterer <peter.hutterer@who-t.net>
47541Date: Wed Sep 1 14:35:40 2010 +1000
47542
47543 xfree86: remove history_size
47544
47545 For a couple of ABIs now the history size was essentially static anyway.
47546
47547 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47548
47549commit 892e39025b45109ba38c0b5d0b5f4535e092f58c
47550Author: Peter Hutterer <peter.hutterer@who-t.net>
47551Date: Wed Sep 1 14:33:52 2010 +1000
47552
47553 xfree86: remove always_core_feedback from InputInfoRec.
47554
47555 Unused
47556
47557 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47558
47559commit 5e2912bd1d2bc8ff1508e5f726e77aa5f8677cbd
47560Author: Peter Hutterer <peter.hutterer@who-t.net>
47561Date: Wed Sep 1 14:33:26 2010 +1000
47562
47563 xfree86: remove first/last from InputInfoRec.
47564
47565 Unused.
47566
47567 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47568
47569commit 974bc2322b3667db38d28063786fbc76fb303d00
47570Author: Peter Hutterer <peter.hutterer@who-t.net>
47571Date: Wed Sep 1 14:32:51 2010 +1000
47572
47573 xfree86: remove private_flags from InputInfoRec.
47574
47575 Unused field.
47576
47577 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47578
47579commit 23b361b25b7b48a820a4ef851fb89706f7ec2cd0
47580Author: Peter Hutterer <peter.hutterer@who-t.net>
47581Date: Wed Sep 1 14:14:20 2010 +1000
47582
47583 xfree86: Purge old_x/old_y from InputInfoRec.
47584
47585 Unused field.
47586
47587 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47588
47589commit 4ac3be29bc1af19f15e84ee0ea609de5c54ca0e0
47590Author: Peter Hutterer <peter.hutterer@who-t.net>
47591Date: Fri Jul 30 14:26:06 2010 +1000
47592
47593 input: Purge AddOtherInputDevices DDX hook.
47594
47595 This hook wasn't used by any DDX. Device addition and removal is handled by
47596 the config backend, so we don't need to do anything special that during the
47597 ListInputDevices request processing.
47598
47599 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47600 Reviewed-by: Adam Jackson <ajax@redhat.com>
47601 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47602
47603commit cb672a461cc6cd668ab7e61994b94e9ff46b3ef1
47604Author: Peter Hutterer <peter.hutterer@who-t.net>
47605Date: Fri Jul 30 14:21:14 2010 +1000
47606
47607 input: remove OpenInputDevice and CloseInputDevice DDX hooks.
47608
47609 In theory, these hooks were to be used for DDX-specific device enablement.
47610 None of the DDXs however did anything here. Now we call DEVICE_INIT on all
47611 devices when they are added, so the xfree86 DDX as the only one with real
47612 code didn't do anything here.
47613
47614 kdrive checked for device validity but that's already handled in
47615 ProcXOpenDevice.
47616
47617 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47618 Reviewed-by: Adam Jackson <ajax@redhat.com>
47619 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47620
47621commit a46d2bb344e822bbd9d69cb59829b85f9d8f0213
47622Author: Simon Thum <simon.thum@gmx.de>
47623Date: Sun Jul 18 12:31:26 2010 +0200
47624
47625 xfree86: Fix xf86 backend-specific input initialization
47626
47627 Instead of shoving it in rather unrelated places, move acceleration init
47628 into xf86NewInputDevice.
47629
47630 Caveat: It's not clear atm how relevant other callers of ActivateDevice
47631 (like OpenDevice) actually are.
47632
47633 Signed-off-by: Simon Thum <simon.thum@gmx.de>
47634 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47635 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47636 Reviewed-by: Adam Jackson <ajax@redhat.com>
47637 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47638
47639commit e930710ae5579752785d6b96ace4b44bf0199a6e
47640Author: Peter Hutterer <peter.hutterer@who-t.net>
47641Date: Fri Jul 30 13:59:29 2010 +1000
47642
47643 xfree86: purge superfluous includes from xf86Xinput.c
47644
47645 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47646 Reviewed-by: Adam Jackson <ajax@redhat.com>
47647 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47648
47649commit 2b7840b63da6bc00aa40e8a427d9fd719ba5e6d0
47650Author: Peter Hutterer <peter.hutterer@who-t.net>
47651Date: Fri Jul 30 13:44:20 2010 +1000
47652
47653 xfree86: remove unused DeviceAssocRec struct.
47654
47655 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47656 Reviewed-by: Adam Jackson <ajax@redhat.com>
47657 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47658
47659commit cdb5863c9f68548afa0c69e08c697c83f9a5e671
47660Author: Peter Hutterer <peter.hutterer@who-t.net>
47661Date: Fri Jul 23 14:16:38 2010 +1000
47662
47663 xfree86: remove LocalDeviceRec/Ptr definition.
47664
47665 Two names pointing to the same struct for over 7 years now. Remove the
47666 define, if drivers don't want to change they can always do the typedef
47667 themselves.
47668
47669 Rename all "LocalDevicePtr local" to "InputInfoPtr pInfo".
47670
47671 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47672 Reviewed-by: Adam Jackson <ajax@redhat.com>
47673 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47674
47675commit 8764782f6de56a9dc5e9d5a8e9fb616a8ddb2f7c
47676Author: Peter Hutterer <peter.hutterer@who-t.net>
47677Date: Mon Jun 7 11:17:10 2010 +1000
47678
47679 xfree86: add xf86IDrvMsg and friends for input driver logging.
47680
47681 Input driver messages are only standardised by convention, with the drivers
47682 prefixing the device name to most messages. This makes it rather hard to
47683 grep on "evdev" for example when looking for the evdev ouput.
47684
47685 This patch adds three new logging functions, modeled after xf86DrvMsg(), the
47686 logging function for output drivers. New functions are
47687 xf86IDrvMsg() - input driver log message in default verbosity.
47688 xf86IDrvMsgVerb() - input driver log message in specified verbosity.
47689 xf86VIDrvMsgVerb() - same as xf86IDrvMsgVerb() but takes a varargs
47690 argument.
47691
47692 Default log format is <driver name>: <device name>: <message>.
47693
47694 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47695 Reviewed-by: Adam Jackson <ajax@redhat.com>
47696 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47697
47698commit b55ac354afa0f16f71bb4f2d3d4277c887082520
47699Author: Peter Hutterer <peter.hutterer@who-t.net>
47700Date: Thu Jul 22 12:00:07 2010 +1000
47701
47702 xfree86: move xf86AllocateInput and xf86DeleteInput to xf86Xinput.c
47703
47704 Make xf86AllocateInput static in the process, this function is only called
47705 from one location.
47706
47707 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47708 Reviewed-by: Adam Jackson <ajax@redhat.com>
47709 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47710
47711commit d2a5f4166417b9d4a02a219b28470e41b6cfe012
47712Author: Peter Hutterer <peter.hutterer@who-t.net>
47713Date: Thu Jul 22 10:30:56 2010 +1000
47714
47715 xfree86: return the device from xf86ActivateDevice.
47716
47717 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47718 Reviewed-by: Adam Jackson <ajax@redhat.com>
47719 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47720
47721commit fbf35e27f3005cf009f2dd1e112260bb294342c7
47722Author: Peter Hutterer <peter.hutterer@who-t.net>
47723Date: Thu Jul 22 10:28:14 2010 +1000
47724
47725 input: set XKB extension for all new devices, not just xfree86 ones.
47726
47727 Right now, Xephyr and others don't get to use XKB on the slave devices.
47728 Which works given that no-one cares about SDs just yet but event processing
47729 is different if the ProcessInputProc isn't wrapped properly.
47730
47731 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47732 Reviewed-by: Adam Jackson <ajax@redhat.com>
47733 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47734
47735commit 11ed32b62c8793ecc1c68e1e4ba91e2eb45eef3e
47736Author: Peter Hutterer <peter.hutterer@who-t.net>
47737Date: Thu Jul 22 10:24:02 2010 +1000
47738
47739 xfree86: purge SendDragEvents support.
47740
47741 From the documentation:
47742 "This is mainly to allow a touch screen to be used with netscape and other
47743 browsers which do strange things if the mouse moves between button down and
47744 button up."
47745
47746 CLOSED - NOTOURBUG
47747
47748 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47749 Reviewed-by: Adam Jackson <ajax@redhat.com>
47750 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47751
47752commit 7defd282beb57e8880980416ed579f62d561d1ac
47753Author: Peter Hutterer <peter.hutterer@who-t.net>
47754Date: Thu Jul 22 10:16:15 2010 +1000
47755
47756 xfree86: remove XI86_CONFIGURED flag.
47757
47758 PreInit returns a status code. Let's use that instead of having it report
47759 Success in some cases but not set the XI86_CONFIGURED flag and thus signal
47760 an init failure.
47761
47762 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47763 Reviewed-by: Adam Jackson <ajax@redhat.com>
47764 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47765
47766commit 97a4acdac23b8ff33da43917c10b65d2fee5c07f
47767Author: Peter Hutterer <peter.hutterer@who-t.net>
47768Date: Thu Jul 22 10:04:53 2010 +1000
47769
47770 xfree86: purge some unused defines.
47771
47772 These defines have been write-only for a while now.
47773
47774 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47775 Reviewed-by: Adam Jackson <ajax@redhat.com>
47776 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47777
47778commit 0fb7a5c261aa0d87d6596d72b70696bffe0c0aff
47779Author: Peter Hutterer <peter.hutterer@who-t.net>
47780Date: Thu Jul 22 09:53:35 2010 +1000
47781
47782 input: Purge Register*Device() functions.
47783
47784 RegisterPointerDevice() and RegisterKeyboardDevice() were already mapped to
47785 RegisterOtherDevice() and obsolete.
47786
47787 RegisterOtherDevice() was called for all devices and the two assignments can
47788 simply be moved into AddInputDevice(). Purge RegisterOtherDevice() and
47789 pretend it never happened.
47790
47791 *lalalalala*
47792
47793 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47794 Reviewed-by: Adam Jackson <ajax@redhat.com>
47795 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47796
47797commit 67ffbcc14cbc61474520d4531599edca24965543
47798Author: Peter Hutterer <peter.hutterer@who-t.net>
47799Date: Thu Jul 22 09:24:08 2010 +1000
47800
47801 xfree86: remove superflous assignments.
47802
47803 ActivateGrab and DeactivateGrab are set in AddInputDevice() already.
47804
47805 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47806 Reviewed-by: Adam Jackson <ajax@redhat.com>
47807 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47808
47809commit 4cd54d9ed9e87074734789a9a7708c2218f87f1e
47810Author: Peter Hutterer <peter.hutterer@who-t.net>
47811Date: Thu Jul 22 09:22:05 2010 +1000
47812
47813 xfree86: make xf86ActivateDevice static.
47814
47815 No-one but the joystick driver uses it and that one should be using NIDR
47816 instead.
47817
47818 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47819 Reviewed-by: Adam Jackson <ajax@redhat.com>
47820 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47821
47822commit 31c71425ac13a7f554316356691a79175ea82a67
47823Author: Peter Hutterer <peter.hutterer@who-t.net>
47824Date: Thu Jul 22 09:09:10 2010 +1000
47825
47826 config: expose config_info as an input option.
47827
47828 config_info is the only reliable indicator we have in the server for
47829 duplicate devices (drivers can test for maj/min on fds as well). Don't set
47830 this after the device has been initialized but assume it's important enough
47831 to set during NIDR.
47832
47833 This makes the option "config_info" available to the drivers as well.
47834
47835 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47836 Reviewed-by: Adam Jackson <ajax@redhat.com>
47837 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47838
47839commit de0cc5a72deb7c477e368aa4fe9a713788d7ae4c
47840Author: Peter Hutterer <peter.hutterer@who-t.net>
47841Date: Wed Jul 21 16:00:26 2010 +1000
47842
47843 xfree86: rework driver PreInit API - XInput ABI 12
47844
47845 The main change introduced in this patch is the removal of the
47846 back-and-forth between DDX and the driver.
47847 The DDX now allocates the InputInfoRec and fills it with default values. The
47848 DDX processes common options (and module-specific default options, if
47849 appropriate) before passing the initialised struct to the driver.
47850
47851 The driver may do module-specific initializations and return Success or an
47852 error code in the case of a failure.
47853
47854 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47855 Reviewed-by: Adam Jackson <ajax@redhat.com>
47856 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47857
47858commit 7925e8945649d4af237e6c3c5593b895a461bd1e
47859Author: Chase Douglas <chase.douglas@ubuntu.com>
47860Date: Wed Sep 1 14:45:34 2010 +1000
47861
47862 Fix udev population of Bluetooth input device product IDs
47863
47864 The udev device_added function takes the vendor and model IDs of added
47865 devices and converts them into an attribute that can be matched for by
47866 an InputClass configuration using MatchUSBID. Currently, the udev
47867 mechanism works for USB devices, but fails to work properly for
47868 Bluetooth devices. The product IDs of the event node are actually the
47869 IDs of the Bluetooth receiver instead of the device.
47870
47871 This patch reads the product ID from the PRODUCT property of the parent
47872 of the added device. This tag is set correctly for both USB and
47873 Bluetooth input devices. The following devices have been tested by
47874 specifying individual InputClass sections in xorg.conf:
47875
47876 * Apple Keyboard (Bluetooth)
47877 * Apple Magic Trackpad (Bluetooth)
47878 * Apple Magic Mouse (Bluetooth)
47879 * Microsoft Bluetooth Notebook Mouse 5000 (Bluetooth)
47880 * Microsoft IntelliMouse Optical (USB)
47881 * N-Trig Touchscreen (USB)
47882 * Wacom Bamboo Touch (USB)
47883
47884 Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
47885 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47886 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47887
47888commit 05e616767e5b7e60b92d31c4042ded5892dce6d4
47889Author: Adam Tkac <atkac@redhat.com>
47890Date: Wed Aug 25 10:38:40 2010 +0200
47891
47892 Return Success from generate_modkeymap() when max_keys_per_mod is zero
47893
47894 max_keys_per_mod equal to zero is a valid situation so generate_modkeymap
47895 should not return BadAlloc in this case.
47896
47897 Signed-off-by: Adam Tkac <atkac@redhat.com>
47898 Reviewed-by: Patrick E. Kane <pekane52 at gmail.com>
47899 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47900
47901commit 4a12aecac670debd0dafb17c245fccb93eea2d60
47902Author: Jesse Adkins <jesserayadkins@gmail.com>
47903Date: Wed Aug 25 13:48:29 2010 -0700
47904
47905 xfree86: Document terminate not mapped by default (bug 25083)
47906
47907 Document that terminate is not mapped to Ctrl+Alt+Backspace by default, to help alleviate some confusion.
47908
47909 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
47910 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47911 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47912
47913commit a90052ba8697e217b0dc68057d7b9202ae8797db
47914Author: David Ge <davidqge@gmail.com>
47915Date: Thu Aug 19 00:33:57 2010 -0500
47916
47917 xkb: Fix RedirectKey didn't send any event.
47918
47919 Xorg.log shows error: Valuators reported for non-valuator device.
47920 This is caused by uninitialized valuators.mask in _XkbFilterRedirectKey(),
47921 which trigger the error in UpdateDeviceState().
47922
47923 Signed-off-by: David Ge <davidqge@gmail.com>
47924 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47925 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47926
47927commit ff109bf84401a451380eb7f3f94a6e0aa2776e3e
47928Author: Alan Coopersmith <alan.coopersmith@oracle.com>
47929Date: Thu Aug 12 00:09:02 2010 -0700
47930
47931 Use GetMaster instead of direct u.master access in core procs
47932
47933 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
47934 Acked-by: Daniel Stone <daniel@fooishbar.org>
47935 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47936 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47937
47938commit 3cc5e4422430e9ca44615f3e63feccd2e5729046
47939Author: Peter Hutterer <peter.hutterer@who-t.net>
47940Date: Fri Jul 23 14:48:32 2010 +1000
47941
47942 xfree86: fix compiler warning about implicied decl of DuplicateModule.
47943
47944 ../../../../hw/xfree86/common/xf86Xinput.c: In function ‘xf86AllocateInput’:
47945 ../../../../hw/xfree86/common/xf86Xinput.c:722: warning: implicit
47946 declaration of function ‘DuplicateModule’
47947 ../../../../hw/xfree86/common/xf86Xinput.c:722: warning: nested extern
47948 declaration of ‘DuplicateModule’
47949 ../../../../hw/xfree86/common/xf86Xinput.c:722: warning: assignment makes
47950 pointer from integer without a cast
47951
47952 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
47953 Reviewed-by: Adam Jackson <ajax@redhat.com>
47954 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
47955
47956commit b5c9953bbf4ffd11f1a70d058c6d3feb2bd1bca8
47957Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47958Date: Mon Jul 26 15:31:03 2010 +0300
47959
47960 xkb: Check if AddResource failed
47961
47962 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47963 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47964
47965commit 2e6d7174042cc8007e947b7d9fb54acc0ebe29d2
47966Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47967Date: Mon Jul 26 15:13:34 2010 +0300
47968
47969 xkb: Fix possible NULL pointer dereference
47970
47971 sli is null before allocation assigment so deference t osli has to be
47972 protected.
47973
47974 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47975 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47976
47977commit d6642de7ebdda16e0056600a86a7802bd4c393b7
47978Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47979Date: Mon Jul 26 14:50:30 2010 +0300
47980
47981 xkb: Fix possible NULL pointer dereference
47982
47983 If search for device failed sli is NULL. In that case we have to protect
47984 dereference to prevent server crash.
47985
47986 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47987 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47988
47989commit adc0697cfcfba295a15d7a307125093cbccd637f
47990Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47991Date: Thu Jul 22 15:11:27 2010 +0300
47992
47993 xkb: Fix memory leak in error path
47994
47995 map is allocated but not freed if reply length and data don't match.
47996
47997 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
47998 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
47999
48000commit 67cfb66562cd9d39f30fec6fbc38eb1eb5e5b030
48001Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48002Date: Thu Jul 22 15:05:57 2010 +0300
48003
48004 xkb: Remove redurant intialization code
48005
48006 calloc already initializes allocated memory to zero.
48007
48008 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48009 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48010
48011commit 2475ef60977f6813dec74ef0837a5915b8a48bbc
48012Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48013Date: Thu Jul 22 12:48:55 2010 +0300
48014
48015 xkb: Fix NULL pointer dereference
48016
48017 xkb->names is dereferenced in else path too.
48018
48019 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48020 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48021
48022commit 184ef0d35612d6ed0619283d376f04d9a904f47c
48023Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48024Date: Thu Jul 22 11:34:54 2010 +0300
48025
48026 xkb: Don't check for NULL before calling free
48027
48028 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48029 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48030
48031commit 1223340644744c0b38aa85f5956eb5ab7c696517
48032Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48033Date: Thu Jul 22 11:33:33 2010 +0300
48034
48035 xkb: Fix memory leak if opening file fails
48036
48037 If fopen fails pointer in buf would be overwriten with a new pointer.
48038
48039 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48040 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48041
48042commit 20cb9c923efa4edc348eba30f956a66413a8208f
48043Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48044Date: Thu Jul 22 11:13:10 2010 +0300
48045
48046 xkb: Use memcpy for copy that has known length
48047
48048 Fixes warning that strncpy is not able to append NULL to the end
48049 of destination.
48050
48051 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
48052 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48053
48054commit de8be07cc0a8163b6ef04455706fd5ca2cebe587
48055Author: Peter Hutterer <peter.hutterer@who-t.net>
48056Date: Tue Aug 17 12:08:52 2010 +1000
48057
48058 dix: don't create core motion events for non-x/y valuators.
48059
48060 Devices that send motion events with valuators other than x/y get core
48061 motion events with unchanged x/y coordinates. This confuses some
48062 applications.
48063
48064 If the DeviceEvent does not have the x/y valuators set, return BadMatch on
48065 core conversion, thus skipping the event altogether.
48066
48067 Reported-by: Bartosz Brachaczek <b.brachaczek@gmail.com>
48068 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48069 Tested-by: Bartosz Brachaczek <b.brachaczek@gmail.com>
48070
48071commit 08adf41f6315663cbac33d010214d98f3e1c8814
48072Author: Matt Turner <mattst88@gmail.com>
48073Date: Fri Aug 27 18:34:49 2010 -0400
48074
48075 Replace malloc/strlen/strcpy with strdup.
48076
48077 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
48078 Signed-off-by: Matt Turner <mattst88@gmail.com>
48079
48080commit ea239112b030588ed3cdd113643ba5f6207a5bd9
48081Author: Jesse Adkins <jesserayadkins@gmail.com>
48082Date: Wed Aug 4 09:21:33 2010 +0000
48083
48084 xfree86: Purge parsePrologueVoid.
48085
48086 This was included in the original commit, and then never used.
48087
48088 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
48089 Signed-off-by: Matt Turner <mattst88@gmail.com>
48090
48091commit b25fb9fe998209e79b6917d12d5765c5f6719a7e
48092Author: Jesse Adkins <jesserayadkins@gmail.com>
48093Date: Wed Aug 4 09:21:32 2010 +0000
48094
48095 xfree86: Removed unused messages from Configint.h
48096
48097 AUTOREPEAT_MSG, MOVED_TO_FLAGS_MSG, and XLEDS_MSG made obsolete by
48098 81913a12910e39d7ea6af8657c1c66cc6791cd65 Jul 21 2006 (remove undead files from master)
48099 UNDEFINED_DEVICE_MSG made obsolete by
48100 6033d8150be3a115b90226eaa42f237bb0cf3369 Oct 9 2007 (first pass at video driver autoloading)
48101
48102 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
48103 Signed-off-by: Matt Turner <mattst88@gmail.com>
48104
48105commit 18b62e0479f15e965611880ada6e0195367df025
48106Author: Jesse Adkins <jesserayadkins@gmail.com>
48107Date: Wed Aug 4 09:21:31 2010 +0000
48108
48109 xfree86: Fix leaks in OpenConfigFile and OpenConfigDir
48110
48111 [mattst88: fixed whitespace and a missing semicolon]
48112
48113 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
48114 Signed-off-by: Matt Turner <mattst88@gmail.com>
48115
48116commit 4f04fd595e82226f1d91226a41bb98ed3d940b37
48117Author: Jesse Adkins <jesserayadkins@gmail.com>
48118Date: Wed Aug 4 09:21:30 2010 +0000
48119
48120 xfree86: Simplify xf86Msg{,Verb}
48121
48122 Previously, the functions would call xf86VDrvMsgVerb with a screen of -1
48123 despite their comments saying they were for "non-driver messages".
48124 They now call LogVMessageVerb, which is what xf86VDrvMsgVerb does anyway
48125 when it has a screen == -1.
48126
48127 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48128 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
48129 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
48130 Signed-off-by: Matt Turner <mattst88@gmail.com>
48131
48132commit f49473abfd6034e68576b2dddd30ba8d8dd0838f
48133Author: Jesse Adkins <jesserayadkins@gmail.com>
48134Date: Wed Aug 4 09:21:29 2010 +0000
48135
48136 xfree86: Remove prototypes for non-existant functions.
48137
48138 This was obsolete from 9a0f25de7ca3c68af867b38936103d17daa92ac6 "Static cleanups, dead code deletion." (server 1.3).
48139
48140 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
48141 Signed-off-by: Matt Turner <mattst88@gmail.com>
48142
48143commit 747bf5fe80f51554205c5a50b8ed9b89065c8a54
48144Author: Jesse Adkins <jesserayadkins@gmail.com>
48145Date: Sun Aug 1 16:01:23 2010 -0700
48146
48147 xfree86: Remove comments about unable to use malloc.
48148
48149 These are leftovers from when X still used Xmalloc and friends for allocation.
48150 Now that those are gone, these comments are just confusing.
48151
48152 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
48153 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48154 Signed-off-by: Matt Turner <mattst88@gmail.com>
48155
48156commit 65466652936bbb2706be455d0a416bcf08e88f66
48157Author: Tiago Vignatti <tiago.vignatti@nokia.com>
48158Date: Wed May 12 13:03:14 2010 +0300
48159
48160 xfree86: vgaarb: fix device decoding interface to send resources type properly
48161
48162 Right now, when there is more than one vide card on the machine, we're
48163 adopting a pessimistic approach and setting all cards to decode VGA legacy
48164 address.
48165
48166 Some cards may want to skip the arbitration and the only way to do so is
48167 through pci_device_vgaarb_decodes. Therefore, send the desired kind of
48168 resource instead force the worst case.
48169
48170 Note that xf86VGAarbiterDeviceDecodes is not being used so far by any
48171 open-source driver. Even so, API break.
48172
48173 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
48174 Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
48175
48176commit 7fbf3e7cf4b261194faad747e0bf59d528df6d91
48177Author: Tiago Vignatti <tiago.vignatti@nokia.com>
48178Date: Mon May 10 21:05:14 2010 +0300
48179
48180 xfree86: vgaarb: remove useless macro
48181
48182 This was inherited from RAC and was never used there either.
48183
48184 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
48185 Reviewed-by: Jamey Sharp <jamey@minilop.net>
48186
48187commit ad698dd5bcc41fcec1e7ce1117c7ad1052710132
48188Author: Tiago Vignatti <tiago.vignatti@nokia.com>
48189Date: Mon May 10 21:03:30 2010 +0300
48190
48191 xfree86: vgaarb: remove superfluous and confusing VGAGet_GC and VGAPut_GC
48192
48193 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
48194 Reviewed-by: Jamey Sharp <jamey@minilop.net>
48195
48196commit c0aed4c99bf553bd7b8bbc79d0ed8f26d0ab3f94
48197Author: Tiago Vignatti <tiago.vignatti@nokia.com>
48198Date: Mon May 10 20:58:30 2010 +0300
48199
48200 xfree86: vgaarb: change macros by inline functions to ease debug
48201
48202 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
48203 Reviewed-by: Jamey Sharp <jamey@minilop.net>
48204
48205commit 79ee78de9de49d0cab03401662baa476a18e53b8
48206Author: Keith Packard <keithp@keithp.com>
48207Date: Fri Aug 20 17:32:27 2010 -0700
48208
48209 xserver 1.9.0
48210
48211 Signed-off-by: Keith Packard <keithp@keithp.com>
48212
48213commit 3e56efcfb63677cd8574e1e435e61d96f79ea536
48214Author: Keith Packard <keithp@keithp.com>
48215Date: Fri Aug 20 10:01:48 2010 -0700
48216
48217 fb: make isClipped always reject negative coordinates (bug 11503)
48218
48219 A window with either dimension > 32767 can be positioned such that
48220 coordinates > 32767 are visible on the screen. Attempts to draw to
48221 those pixels will generate coordinates wrapped around to negative
48222 values.
48223
48224 The optimized clipping macro, 'isClipped', in fbbits.h, computes
48225 clipping in window space rather than screen space using int16 values,
48226 and so it too has coordinates wrapped around to negative values and
48227 hence ends up accepting the wrapped drawing coordinates.
48228
48229 Two possible fixes for this problem
48230
48231 1) Detect wrapped region coordinates and clip those to 32767.
48232 2) Detect negative incoming coordinates and reject those
48233
48234 This patch takes the second approach as it is much shorter, simply
48235 detecting when either X or Y incoming coordinate is negative, which
48236 can never be 'within' any drawable.
48237
48238 Signed-off-by: Keith Packard <keithp@keithp.com>
48239 Reviewed-by: Adam Jackson <ajax@redhat.com>
48240
48241commit 951605b4660290044fb238bcf1d6d9e498567e8c
48242Author: Chris Wilson <chris@chris-wilson.co.uk>
48243Date: Fri Aug 20 13:51:04 2010 +0100
48244
48245 edid: Adjust rounding of max_clock
48246
48247 A simple hack to accommodate various EDID who have detailed modes that
48248 exceed the EDID's max pixel clock. The pixel clock is only defined in
48249 units of 10MHz and often appears as the maximum pixel code of the
48250 detailed modes, rounded to the nearest 10MHz. Adjusting the max_clock to
48251 include an extra 5MHz prevents the parser from rejecting the detailed
48252 modes.
48253
48254 The kernel uses the same fuzz and by including it in X we can use the
48255 same modes in X as for the console.
48256
48257 Fixes:
48258
48259 Bug 23833 - X uses different refresh rate to that set by kernel module
48260 https://bugs.freedesktop.org/show_bug.cgi?id=23833
48261
48262 In the future, we will want to try harder to keep the KMS modes but at
48263 the same time we need to apply the restrictions as specified by the
48264 user's configuration, and need to fill in modes for fullscreen games on
48265 fixed-mode panels.
48266
48267 Reported-and-tested-by: Fabio Pedretti <fabio.ped@libero.it>
48268 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
48269 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
48270 Signed-off-by: Keith Packard <keithp@keithp.com>
48271
48272commit 7e581780603d6b15291d032efdeeca77f969e0ba
48273Author: Chris Wilson <chris@chris-wilson.co.uk>
48274Date: Thu Jun 24 12:24:58 2010 +0100
48275
48276 glx: Prevent NULL context deref in __glXGetDrawable() (bug 29184)
48277
48278 During a SwapBuffers request, we may end up querying an unknown drawable
48279 outside of an active context, and so need to report this error prior to
48280 attempting to dereference the NULL context.
48281
48282 Also fixes:
48283
48284 [Bug 29184] glXSwapBuffers with no GLX context crashes X.
48285 https://bugs.freedesktop.org/show_bug.cgi?id=29184
48286
48287 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
48288 Cc: Kristian Høgsberg <krh@bitplanet.net>
48289 Reviewed-by: Adam Jackson <ajax@redhat.com>
48290 Signed-off-by: Keith Packard <keithp@keithp.com>
48291
48292commit 5725849a1b427cd4a72b84e57f211edb35838718
48293Author: Adam Jackson <ajax@redhat.com>
48294Date: Mon Jun 28 18:08:50 2010 -0400
48295
48296 render: Bounds check for nglyphs in ProcRenderAddGlyphs (#28801)
48297
48298 Signed-off-by: Adam Jackson <ajax@redhat.com>
48299 Reviewed-by: Julien Cristau <jcristau@debian.org>
48300 Signed-off-by: Keith Packard <keithp@keithp.com>
48301
48302commit fc091936e2bddbbab9c9a501edc5a5f08388617e
48303Author: Peter Hutterer <peter.hutterer@who-t.net>
48304Date: Mon Aug 16 14:18:45 2010 +1000
48305
48306 dix: copy the valuators passed into GPE/GKVE/GProxE.
48307
48308 GPE and friends modify the valuators array passed in. Which means any driver
48309 using e.g. xf86PostButtonEventP(..., valuators) twice to emulate a button
48310 click will provide garbage data on the second run.
48311
48312 This is currently affecting the wacom driver, xf86PostButtonEventP() with
48313 valuators is required to have input events with device-specific axis values.
48314 Passing the same valuators in twice, once with press, once with release,
48315 will see the valuators modified in the first call and garbage submitted in
48316 the next one.
48317
48318 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48319 Reviewed-by: Keith Packard <keithp@keithp.com>
48320 Signed-off-by: Keith Packard <keithp@keithp.com>
48321
48322commit 6e3e559e9fa63069a10eb834a6dab9a4cfc140ee
48323Author: Keith Packard <keithp@keithp.com>
48324Date: Sun Aug 15 20:53:20 2010 -0700
48325
48326 dix: reset pScreen->root to NULL when root window is deleted.
48327
48328 From: Dave Airlie <airlied@linux.ie>
48329
48330 We were seeing a crash in the FreeAllResources codepath,
48331 running valgrind revealed this,
48332
48333 ==12536== Invalid read of size 4
48334 ==12536== at 0x810BCAB: DeliverPropertyEvent (rrproperty.c:33)
48335 ==12536== by 0x80958A4: TraverseTree (window.c:227)
48336 ==12536== by 0x809593E: WalkTree (window.c:255)
48337 ==12536== by 0x810BC66: RRDeliverPropertyEvent (rrproperty.c:53)
48338 ==12536== by 0x810BD5D: RRDeleteProperty.clone.0 (rrproperty.c:76)
48339 ==12536== by 0x810BD98: RRDeleteAllOutputProperties (rrproperty.c:88)
48340 ==12536== by 0x810A36E: RROutputDestroyResource (rroutput.c:407)
48341 ==12536== by 0x808DF4E: FreeClientResources (resource.c:859)
48342 ==12536== by 0x808E005: FreeAllResources (resource.c:876)
48343 ==12536== by 0x8062300: main (main.c:305)
48344 ==12536== Address 0x46ba8ac is 4 bytes inside a block of size 164 free'd
48345 ==12536== at 0x40057F6: free (vg_replace_malloc.c:325)
48346 ==12536== by 0x8087F1F: _dixFreeObjectWithPrivates (privates.c:357)
48347 ==12536== by 0x809832A: DeleteWindow (window.c:926)
48348 ==12536== by 0x808DF4E: FreeClientResources (resource.c:859)
48349 ==12536== by 0x808E005: FreeAllResources (resource.c:876)
48350 ==12536== by 0x8062300: main (main.c:305)
48351
48352 Its a use after free on the root window, since we have already deleted it
48353 at this point. This patch checks if the window we are destroying is the root
48354 window and resets the pointer to NULL if it is.
48355
48356 Signed-off-by: Keith Packard <keithp@keithp.com>
48357 Reviewed-by: Dave Airlie <airlied@redhat.com>
48358 Tested-by: Dave Airlie <airlied@redhat.com>
48359
48360commit 5d1d9d9ae39fab2ee2ac085f9776f82768828dc8
48361Author: Jeremy Huddleston <jeremyhu@apple.com>
48362Date: Sun Aug 1 11:41:58 2010 -0700
48363
48364 XQuartz: xpr: Bail on errors during unlock and destroy
48365
48366 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48367
48368commit ee7fd8fc58d9fadfbb92302ddea224537f068538
48369Author: Jeremy Huddleston <jeremyhu@apple.com>
48370Date: Sun Aug 1 11:39:14 2010 -0700
48371
48372 XQuartz: UpdateScreen at the end of SetRootless
48373
48374 This will ensure that pRoot is unlocked after the miPaintWindow
48375
48376 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48377
48378commit 4fc4cab98d454afbfd0d2f48548b5b481e8e7c82
48379Author: Jeremy Huddleston <jeremyhu@apple.com>
48380Date: Thu Jul 29 14:49:10 2010 -0700
48381
48382 XQuartz: Make application switching work better for the no-spaces case
48383
48384 We still have the issue with not raising the frontmost window for the case
48385 when spaces is enabled, and the AppleSpacesSwitchOnActivate preference is
48386 disabled.
48387
48388 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48389
48390commit e5bc62a03289f956c54c4699edf47f7ff237b5be
48391Author: Jeremy Huddleston <jeremyhu@apple.com>
48392Date: Sun Jul 25 22:29:11 2010 -0700
48393
48394 XQuartz: Ignore kXquartzToggleFullscreen when rootless
48395
48396 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48397
48398commit 71af1f71c0492c365707c6b3810f94642ff39352
48399Author: Keith Packard <keithp@keithp.com>
48400Date: Thu Aug 12 23:01:59 2010 -0700
48401
48402 Bump to version 1.8.99.906 (1.9 RC6)
48403
48404 Signed-off-by: Keith Packard <keithp@keithp.com>
48405
48406commit 0af322858e86665ee43f065741318e69c2755510
48407Author: Keith Packard <keithp@keithp.com>
48408Date: Thu Aug 12 22:56:36 2010 -0700
48409
48410 Silence GCC warning about uninitialized lastSlave variable
48411
48412 Not an actual bug, but gcc can't tell that this variable cannot be
48413 used without being initialized
48414
48415 Signed-off-by: Keith Packard <keithp@keithp.com>
48416
48417commit b5cf9c5090d15a50b105470900823f2d398d4bd2
48418Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48419Date: Thu Aug 12 00:09:01 2010 -0700
48420
48421 Stop checking or calling PtrCtrlProcs
48422
48423 None of them do anything useful now that pointer acceleration is
48424 entirely handled in the server. (Does not completely nuke yet,
48425 since that would be an API/ABI break.)
48426
48427 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48428 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48429 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48430
48431commit bce12f2956f23c0ee53f7f6485dba631293a0931
48432Author: Jesse Adkins <jesserayadkins@gmail.com>
48433Date: Wed Aug 4 23:39:14 2010 -0700
48434
48435 xfree86: parser: Never use constant strings for driver names (fixes #17438)
48436
48437 When the parser sees the "keyboard" driver, it automatically (and
48438 silently) replaces it with the constant string "kbd".
48439 Everybody else uses malloc'd memory for the driver name, so input
48440 device closure assumes it can use free.
48441 Free val.str, so this crash doesn't turn into a memory leak. Whew.
48442
48443 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com>
48444 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
48445 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48446
48447commit 619ca32202cd22f2a408586cbc906b8bbaeb9358
48448Author: Peter Hutterer <peter.hutterer@who-t.net>
48449Date: Wed Jul 28 15:08:27 2010 +1000
48450
48451 Xi: reset the unused classes pointer after copying
48452
48453 After copying the unused_classes into the device, reset the original
48454 pointer. Otherwise we have two pointers pointing to the same field and both
48455 get freed on device removal.
48456
48457 Some classes already have this behaviour since 51c8fd69.
48458
48459 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48460 Reviewed-by: Keith Packard <keithp@keithp.com>
48461
48462commit 1a172f3297369a72865232c382abfc14281102a4
48463Author: Peter Hutterer <peter.hutterer@who-t.net>
48464Date: Fri Jul 23 13:24:34 2010 +1000
48465
48466 xkb: if the button isn't down, don't fake an event.
48467
48468 If the button we're about to fake isn't down (or up), don't fake a release
48469 (or press) event for it. Behaviour is the same as before, this just saves
48470 a few cycles.
48471
48472 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48473 Reviewed-by: Keith Packard <keithp@keithp.com>
48474
48475commit 651c36e95ec0ac60d3fb98966df4218712ae78c2
48476Author: Peter Hutterer <peter.hutterer@who-t.net>
48477Date: Fri Jul 23 11:46:30 2010 +1000
48478
48479 xkb: post-fix PointerKeys button events with a DeviceChangedEvent.
48480
48481 commit 14327858391ebe929b806efb53ad79e789361883
48482 xkb: release XTEST pointer buttons on physical releases. (#28808)
48483 revealed a bug with the XTEST/PointerKeys interaction.
48484
48485 Events resulting from PointerKeys are injected into the event processing
48486 stream, not appended to the event queue. The events generated for the fake
48487 button press include a DeviceChangedEvent (DCE), a raw button event and the
48488 button event itself. The DCE causes the master to switch classes to the
48489 attached XTEST pointer device.
48490
48491 Once the fake button is processed, normal event processing continues with
48492 events in the EQ. The master still contains the XTEST classes, causing some
48493 events to be dropped if e.g. the number of valuators of the event in the
48494 queue exceeds the XTEST device's number of valuators.
48495
48496 Example: the EQ contains the following events, processed one-by-one, left to
48497 right.
48498
48499 [DCE (dev)][Btn down][Btn up][Motion][Motion][...]
48500 ^ XkbFakeDeviceButton injects [DCE (XTEST)][Btn up]
48501
48502 Thus the event sequence processed looks like this:
48503
48504 [DCE (dev)][Btn down][Btn up][DCE (XTEST)][Btn up][Motion][Motion][...]
48505
48506 The first DCE causes the master to switch to the device. The button up event
48507 injects a DCE to the XTEST device, causing the following Motion events to be
48508 processed with the master still being on XTEST classes.
48509
48510 This patch post-fixes the injected event sequence with a DCE to restore the
48511 classes of the original slave device, resulting in an event sequence like
48512 this:
48513 [DCE (dev)][Btn down][Btn up][DCE (XTEST)][Btn up][DCE (dev)][Motion][Motion]
48514
48515 Note that this is a simplified description. The event sequence injected by
48516 the PointerKeys code is injected for the master device only and the matching
48517 slave device that caused the injection has already finished processing on
48518 the slave. Furthermore, the injection happens as part of the the XKB layer,
48519 before the unwrapping of the processInputProc takes us into the DIX where
48520 the DCE is actually handled.
48521
48522 Bug reproducible with a device that reports more than 2 valuators. Simply
48523 cause button releases on the device and wait for a "too many valuators"
48524 warning message.
48525
48526 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48527 Acked-by: Daniel Stone <daniel@fooishbar.org>
48528 Reviewed-by: Keith Packard <keithp@keithp.com>
48529
48530commit 6dae7f3792611aace1df0cca63bf50c50d93de43
48531Author: Chris Wilson <chris@chris-wilson.co.uk>
48532Date: Tue Aug 10 19:30:20 2010 +0100
48533
48534 xace: Invalid reference to out-of-scope data.
48535
48536 The callback data passed by reference to the hook was allocated on stack
48537 within the scope of the case statement. The compiler is free to reuse
48538 any of that stack space whilst making the function call so we may end up
48539 passing garbage into the callback.
48540
48541 References:
48542
48543 Bug 18451 - Xorg server 1.5.2 SEGV during XFixesGetCursorImage()
48544 https://bugs.freedesktop.org/show_bug.cgi?id=18451
48545
48546 v2: Drop the unrelated hunk that snuck in when ammending the commit
48547 message.
48548
48549 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
48550 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48551 Signed-off-by: Keith Packard <keithp@keithp.com>
48552
48553commit fbd02046797185715e1a120d52e410ec78fc365f
48554Author: Jon TURNEY <jon.turney@dronecode.org.uk>
48555Date: Thu Aug 5 15:19:20 2010 +0100
48556
48557 Cygwin/X: Fix glxWinCreateDrawable() for API change
48558
48559 Commit 9de0e31746d5f0d9d39d11c94ec3cbc04a9935fc changed the signature
48560 of __GLXScreen's createDrawable method.
48561
48562 Update the glxWinCreateDrawable() function in XWin's GLX provider
48563 appropriately.
48564
48565 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
48566 Tested-by: Colin Harrison <colin.harrison@virgin.net>
48567 Signed-off-by: Keith Packard <keithp@keithp.com>
48568
48569commit ff70848e623920779d20f35d47e9e1f34157de47
48570Author: Keith Packard <keithp@keithp.com>
48571Date: Fri May 21 09:01:43 2010 -0700
48572
48573 Don't let alpha maps recurse in fb. Bug 23581.
48574
48575 Recursive alpha maps (where one picture's alpha map is set to a
48576 picture with an external alpha map) would be all fine and dandy,
48577 except for the case where the client constructs a loop. Detecting this
48578 case when setting the alpha map values would be difficult as any time
48579 an alpha map is set, the server would have to check for the looping
48580 case.
48581
48582 Instead, a far simpler fix is to simply disallow recursive alpha maps
48583 in the rendering code, the Render spec is ambiguous in this area and
48584 allows us to to ignore the recursive case.
48585
48586 Signed-off-by: Keith Packard <keithp@keithp.com>
48587 Reviewed-by: Adam Jackson <ajax@redhat.com>
48588
48589commit 70a94c5b7a42adc0995bf774c44587a0778be0d0
48590Author: Jon TURNEY <jon.turney@dronecode.org.uk>
48591Date: Tue Aug 3 19:49:10 2010 +0100
48592
48593 rootless: fix uninitialized private key assert in non-rootless modes in Cygwin/X
48594
48595 IsFramedWindow() is called from miPaintWindow() if the server has
48596 been built with ROOTLESS defined, irrespective of if RootlessInit()
48597 has ever been called, or not.
48598
48599 Add a check to IsFramedWindow() to check if rootlessWindowPrivateKey
48600 has been registered (as a proxy for checking if the rootless extension
48601 has been initialized) so we don't go on to try to use that key,
48602 triggering an assert.
48603
48604 This bug exposes what appears to be a difference in opinion about
48605 the rootless extension between XQuartz and XWin. XQuartz always
48606 initializes the rootless extension, whereas XWin offers several modes
48607 of operation, and the rootless extension is only used for one of them
48608
48609 That probably means that the all code under compile time guard for
48610 ROOTLESS should be carefully checked that it doesn't also need to be
48611 under a run-time guard
48612
48613 (I've reviewed the other ROOTLESS blocks in dix/events.c and
48614 dix/window.c and they look ok -- keithp)
48615
48616 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
48617 Reviewed-by: Keith Packard <keithp@keithp.com>
48618 Signed-off-by: Keith Packard <keithp@keithp.com>
48619
48620commit 3ab6cd31cbdf8095b2948034fce5fb645422d8da
48621Author: Adam Jackson <ajax@redhat.com>
48622Date: Mon Aug 9 15:20:20 2010 -0400
48623
48624 fonts: Fix refcounting for asynchronous font operations (#3040)
48625
48626 When doing Xinerama, we'll dispatch font ops across all backend screens.
48627 If using a font server (such that some operations can sleep), we'll put
48628 the client to sleep once for each screen, but only wake up once, because
48629 we're trying to keep track of the sleep count in _each_ screen's
48630 closure.
48631
48632 Instead, just ask the core whether the client is already asleep.
48633
48634 Signed-off-by: Adam Jackson <ajax@redhat.com>
48635 Reviewed-by: Keith Packard <keithp@keithp.com>
48636 Signed-off-by: Keith Packard <keithp@keithp.com>
48637
48638commit 35c0dbe4b0a6ab790f4271325b8a0b16894daa8b
48639Author: Gaetan Nadon <memsize@videotron.ca>
48640Date: Sun Aug 8 10:03:08 2010 -0400
48641
48642 doc: add missing .gitignore for Xserver-DTrace
48643
48644 The dtrace doc was recently added in commit
48645 9c171d4aee695ab66e6db1ab92539557bd368cfa
48646
48647 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48648 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
48649
48650commit 05c4fe83fdc20b838fd59658478278dc31a55eb9
48651Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48652Date: Fri Jul 30 17:43:24 2010 -0700
48653
48654 Check HAVE_XMLTO_TEXT before trying to use xmlto to make text files
48655
48656 Reported-by: Matt Turner <mattst88@gmail.com>
48657 Tested-by: Gaetan Nadon <memsize@videotron.ca>
48658 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48659
48660commit a6fb7829ed9bf26c4c2a02c6ed075fb1b17f7b2a
48661Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48662Date: Mon Jul 26 19:55:27 2010 -0700
48663
48664 Xserver-spec: Update ChangeGC prototype, add ChangeGCXIDs
48665
48666 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48667 Reviewed-by: Jamey Sharp <jamey@minilop.net>
48668
48669commit a817271d461e2f95dd7dc62cd1c7d123ce92f555
48670Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48671Date: Thu Jul 22 23:57:57 2010 -0700
48672
48673 Update Xserver-spec for new devPrivates API
48674
48675 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48676 Reviewed-by: Patrick E. Kane <pekane52@gmail.com>
48677
48678commit 40d598a4f84091db743ceef4d60752bb910c3e56
48679Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48680Date: Thu Jul 22 23:57:02 2010 -0700
48681
48682 Correct function name in dixRegisterPrivateKey comments
48683
48684 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48685 Reviewed-by: Patrick E. Kane <pekane52@gmail.com>
48686
48687commit 8d7b7a0d71e0b89321b3341b781bc8845386def6
48688Author: Kristian Høgsberg <krh@bitplanet.net>
48689Date: Thu Jul 29 20:36:25 2010 -0400
48690
48691 Set DamageSetReportAfterOp to true for the damage extension
48692
48693 Change the damage extension reporter to queue up events after we chain
48694 to the wrapped functions. Damage events are typically sent out after
48695 the rendering happens anyway, since we submit batch buffers from the
48696 flush callback chain and then flush client io buffers. Compositing
48697 managers relie on this order, and there is no way we could reliably
48698 provide damage events to clients before the rendering happens anyway.
48699
48700 By queueing up the damage events before the rendering happens, there's
48701 a risk that the client io buffer may overflow and send the damage
48702 events to the client before the driver has even seen the rendering
48703 request. Reporting damage events after the rendering fixes this
48704 corner case and better corresponds with how we expect this to work.
48705
48706 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
48707 Reviewed-by: Keith Packard <keithp@keithp.com>
48708
48709commit c65f610e12f9df168d5639534ed3c2bd40afffc8
48710Author: Kristian Høgsberg <krh@bitplanet.net>
48711Date: Thu Jul 29 18:52:35 2010 -0400
48712
48713 Always call the flush callback chain when we flush client buffers
48714
48715 We were missing the callback in a couple of places. Drivers may use
48716 the flush callback to submit batched up rendering before events (for
48717 example, damage events) are sent out, to ensure that the rendering
48718 has been queued when the client receives the event.
48719
48720 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
48721 Reviewed-by: Keith Packard <keithp@keithp.com>
48722
48723commit 7e0575baf14ec4a89492fd2780f9ab5b9244afbd
48724Author: Matt Turner <mattst88@gmail.com>
48725Date: Tue Jul 27 23:32:36 2010 -0400
48726
48727 ddc: Fix memory leak in GetEDID_DDC1
48728
48729 Mark argument to DDC_checksum as const too.
48730
48731 Signed-off-by: Matt Turner <mattst88@gmail.com>
48732 Reviewed-by: Adam Jackson <ajax@redhat.com>
48733 Signed-off-by: Keith Packard <keithp@keithp.com>
48734
48735commit 750d4e82a0c1161292d24216bcff200cd6647611
48736Author: Jeremy Huddleston <jeremyhu@apple.com>
48737Date: Thu Jul 29 11:39:40 2010 -0700
48738
48739 XQuartz: xpbproxy: Don't take down the whole server on an IO error
48740
48741 Calls pthread_exit to prevent _XIOError from calling exit()
48742
48743 This fixes http://xquartz.macosforge.org/trac/ticket/421
48744
48745 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48746
48747commit 98f90145d786695ecbc02a667c6ffe7c619dc67e
48748Author: Jeremy Huddleston <jeremyhu@apple.com>
48749Date: Wed Jul 28 18:08:02 2010 -0700
48750
48751 XQuartz: GLX: Don't mangle __GLXDrawable's pDraw
48752
48753 We were incorrectly NULLing out pDraw in __GLXDrawable instead of ours in
48754 __GLXAquaDrawable. (we should refactor to eliminate this redundancy later)
48755
48756 This was causing http://xquartz.macosforge.org/trac/ticket/426
48757 This was benign until commit f0006aa58f6cf7552a239e169ff6e7e4fda532f4
48758 The root cause of this change was fed7ccc481ad1caaa518cafe944c2327a5d0b6c65
48759
48760 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48761
48762commit 9c171d4aee695ab66e6db1ab92539557bd368cfa
48763Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48764Date: Fri Jun 25 16:52:42 2010 -0700
48765
48766 Add documentation of the Xserver DTrace probes
48767
48768 Mostly pulled together from posts to my blog and the docs posted
48769 at http://people.freedesktop.org/~alanc/dtrace/ and converted to
48770 DocBook.
48771
48772 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48773 Reviewed-by: Patrick E. Kane <pekane52@gmail.com>
48774
48775commit 6b912b08ce65072b2401167fbf2150b99c0ca6b0
48776Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48777Date: Sat Jun 19 23:56:20 2010 -0700
48778
48779 Add name argument to CreateNewResourceType documentation
48780
48781 Reflects API change made in commit 895f40792a during Xorg 1.8 development
48782
48783 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48784 Reviewed-by: Patrick E. Kane <pekane52@gmail.com>
48785
48786commit 44c9350d72b3eda982c642feb45d6648afc626cf
48787Author: Alan Coopersmith <alan.coopersmith@oracle.com>
48788Date: Sat Jun 19 23:40:37 2010 -0700
48789
48790 Use DocBook stylesheets from xorg-sgml-doctools if they're available
48791
48792 Bumps minimum xorg-macros requirement from 1.6 to 1.10
48793
48794 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48795 Reviewed-by: Patrick E. Kane <pekane52@gmail.com>
48796
48797commit 9fbbff3c0456f1969d45cc957d3260723caf62d7
48798Merge: 95756f4 9ac8e20
48799Author: Keith Packard <keithp@keithp.com>
48800Date: Wed Jul 21 11:56:39 2010 -0700
48801
48802 Merge remote branch 'whot/for-keith'
48803
48804commit 9ac8e206ffe1016a8bc203261ade6c763a8a4f86
48805Author: Peter Hutterer <peter.hutterer@who-t.net>
48806Date: Thu Jul 15 15:46:15 2010 +1000
48807
48808 xkb: use GetMaster instead of dev->u.master.
48809
48810 Devices that are both pointers and keyboards are not affected by keyboard
48811 changes as their master device is a master pointer, not a master keyboard.
48812 Use GetMaster() instead to ensure devices that are attached to the paired
48813 master pointer device will still be update.
48814
48815 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48816 Reviewed-by: Keith Packard <keithp@keithp.com>
48817 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
48818
48819commit c54f81ba7a58faf37a612bd9a45276bb2922b5d8
48820Author: Peter Hutterer <peter.hutterer@who-t.net>
48821Date: Thu Jul 15 13:24:14 2010 +1000
48822
48823 dix: hack around enter/leave event issues for grabbed devices (#27804)
48824
48825 The current core enter/leave does not cater for device grabs during
48826 enter/leave events. If a window W contains a pointer P1 and a client grabs a
48827 pointer P2, this pointer will not generate enter/leave events inside this
48828 window.
48829
48830 Hack around this by forcing grabbed devices to always send enter/leave
48831 events.
48832
48833 X.Org Bug 27804 <http://bugs.freedesktop.org/show_bug.cgi?id=27804>
48834
48835 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48836 Reviewed-by: Keith Packard <keithp@keithp.com>
48837
48838commit 95756f410c65a6510a797e94a792b959d45cdb9e
48839Author: Jan Hauffa <hauffa@in.tum.de>
48840Date: Sat Jul 17 12:18:39 2010 -0700
48841
48842 rootless: Adjust the frame size of the native root window in RootlessResizeWindow
48843
48844 If the native root window isn't resized as well, we will likely crash the
48845 next time we draw to the root. On OS X, this can be seen by:
48846
48847 1) Put the display preferences in the menu bar and set X11's preferences so you
48848 can access the menu bar in fullscreen mode
48849 2) Set the resolution of your screen lower than normal.
48850 3) Start X11 in fullscreen mode. The root window will cover the screen as
48851 expected.
48852 4) Use the menu bar to increase the resolution of the display. The root
48853 window will now cover the old area and not the full screen, but
48854 'xwininfo -root' will report the full width.
48855 5) Run 'xsetroot -solid red', and we have the crash you mention above.
48856
48857 Leaving/entering fullscreen after #4 will fix the problem. This is because the
48858 WINREC is erased when we leave fullscreen mode and it is recreated upon
48859 re-entry:
48860
48861 RootlessUpdateRooted(FALSE)
48862 RootlessDisableRoot(screenInfo.screens[0])
48863 RootlessDestroyFrame (pRoot, winRec);
48864 RootlessUpdateRooted(TRUE)
48865 RootlessEnableRoot(screenInfo.screens[0])
48866 RootlessEnsureFrame(screenInfo.screens[0]->pRoot)
48867 creates a new WINREC...
48868
48869 Signed-off-by: Jan Hauffa <hauffa@in.tum.de>
48870 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
48871 Acked-By: Jon TURNEY <jon.turney@dronecode.org.uk>
48872 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
48873
48874commit a2c13f0d6548310e3cd115cf486d3e43edf23dcc
48875Author: Keith Packard <keithp@keithp.com>
48876Date: Wed Jul 14 12:57:29 2010 -0700
48877
48878 Bump to version 1.8.99.905 (1.9 RC5)
48879
48880 Signed-off-by: Keith Packard <keithp@keithp.com>
48881
48882commit 0fc02c0bf92f694889589e3648acc08d4684de37
48883Merge: 0540c46 9d8ec71
48884Author: Keith Packard <keithp@keithp.com>
48885Date: Tue Jul 13 15:05:36 2010 -0700
48886
48887 Merge remote branch 'jeremyhu/master'
48888
48889commit 0540c46066f938ad5611c56081cfcd8457a9b718
48890Author: Michel Dänzer <daenzer@vmware.com>
48891Date: Tue Jul 13 14:56:53 2010 +0200
48892
48893 EXA: Finish access to pixmap if it's prepared at destruction time.
48894
48895 Previously we assumed every pixmap destroyed during a software fallback was
48896 also created during a software fallback and had access prepared, but that's
48897 not always true.
48898
48899 Fixes a server abort
48900 Reported-by: 邓逸昕 <bupt.dengyixin@gmail.com>
48901
48902 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
48903 Acked-by: Maarten Maathuis <madman2003@gmail.com>
48904 Signed-off-by: Keith Packard <keithp@keithp.com>
48905
48906commit d75e8146c414bfd512ba5dbd4a83acb334bbe19b
48907Author: Keith Packard <keithp@keithp.com>
48908Date: Mon Jul 12 16:01:34 2010 -0700
48909
48910 Unwrap/rewrap EnterVT/LeaveVT completely, Fixes 28998
48911
48912 Because some EnterVT code needs to remove it self from the
48913 call chain, we need to fix all of the wrappers to correctly
48914 unwrap/rewrap during the call chain. This is a follow-on to the fix
48915 for bug 27114 in commit 68a9ee8370e6f9b38218376ac92d5130a5b0ef1e.
48916
48917 Signed-off-by: Keith Packard <keithp@keithp.com>
48918 Tested-by: Jesse Barnes <jesse.barnes@intel.com>
48919 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
48920 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
48921
48922commit 9d8ec712a67ce71ea7408f0626cda7e0fa7c3bac
48923Author: Jeremy Huddleston <jeremyhu@apple.com>
48924Date: Tue Jul 13 08:56:36 2010 -0700
48925
48926 XQuartz: Bump bundle version to 2.6.0 for xorg-server-1.9.0 and X11R7.6
48927
48928 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48929
48930commit b2b9c458a46e9a41c3c76ffe83a2b580a41d0e90
48931Author: Jan Hauffa <hauffa@in.tum.de>
48932Date: Wed Jun 16 09:25:41 2010 -0700
48933
48934 XQuartz: Remove some dead code.
48935
48936 Signed-off-by: Jan Hauffa <hauffa@in.tum.de>
48937 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
48938 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48939
48940commit 648d189548530fa23d97d1e8737f89d297f1c443
48941Author: Jeremy Huddleston <jeremyhu@apple.com>
48942Date: Tue Jul 13 08:25:27 2010 -0700
48943
48944 XQuartz: Avoid a crash when mistakenly free()ing in QuartzSetCursor on some configs
48945
48946 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
48947
48948commit 3209b094a3b1466b579e8020e12a4f3fa78a5f3f
48949Author: Jesse Barnes <jbarnes@virtuousgeek.org>
48950Date: Fri Jul 9 10:36:56 2010 -0700
48951
48952 DRI2: re-allocate DRI2 drawable if pixmap serial changes
48953
48954 If a pixmap header is modified or the drawable serial changes, some
48955 aspects of the drawable are likely to have changed so we should
48956 re-allocate the corresponding DRI2 drawable in that case. This is one
48957 way of catching when the root window pixmap changes through xrandr.
48958
48959 Fixes bug https://bugs.freedesktop.org/show_bug.cgi?id=28365.
48960
48961 Reviewed-by: Keith Packard <keithp@keithp.com>
48962 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
48963 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
48964 Signed-off-by: Keith Packard <keithp@keithp.com>
48965
48966commit 02b11509b25686ff7bd567ecb78a435701edc4c2
48967Author: Jesse Barnes <jbarnes@virtuousgeek.org>
48968Date: Fri Jul 9 10:36:12 2010 -0700
48969
48970 miModifyPixmapHeader: always update serialNumber
48971
48972 We should update the serial number even if we just change a single field.
48973
48974 Reviewed-by: Keith Packard <keithp@keithp.com>
48975 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
48976 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
48977 Signed-off-by: Keith Packard <keithp@keithp.com>
48978
48979commit 2307ab5bc9365ebbe04568edb7c7620a23689b70
48980Merge: c65280c fd4f505
48981Author: Keith Packard <keithp@keithp.com>
48982Date: Tue Jul 6 23:54:54 2010 -0400
48983
48984 Merge remote branch 'whot/for-keith'
48985
48986commit fd4f5059f08165a726071dc9f1ca877038292f6f
48987Author: Peter Hutterer <peter.hutterer@who-t.net>
48988Date: Tue Jul 6 09:19:09 2010 +1000
48989
48990 dix: purge leftover manual key down bit setting.
48991
48992 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
48993 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
48994
48995commit a1afe172559aff010e886cfc2a7a922d4a06c697
48996Author: Peter Hutterer <peter.hutterer@who-t.net>
48997Date: Tue Jul 6 09:16:42 2010 +1000
48998
48999 dix: add aux. functions for button_is_down, set_button_down, set_button_up.
49000
49001 Same as the matching key functions. Buttons, like keys, can have two states
49002 for down/up - one posted, one processed. Posted is set during event
49003 generation (usually in the signal handler). Processed is set during event
49004 processing when the event queue is emptied and events are being delivered to
49005 the client.
49006
49007 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49008 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
49009
49010commit 32473d6bf38c95b2d6d5ddbf583a1e801c6605e4
49011Author: Peter Hutterer <peter.hutterer@who-t.net>
49012Date: Tue Jul 6 08:56:12 2010 +1000
49013
49014 dix: use BitIsOn/SetBit/ClearBit macros for set_key_down helpers.
49015
49016 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49017 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
49018
49019commit c18442908080c9833dfd6bb2ff367945d1892421
49020Author: Peter Hutterer <peter.hutterer@who-t.net>
49021Date: Mon Jul 5 16:47:57 2010 +1000
49022
49023 Xi: use set_key_up/down instead of manual bit handling.
49024
49025 We have the wrappers, use them.
49026
49027 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49028
49029commit 10442ce02b5be7f82b373bee1939e2b523e291d9
49030Author: Peter Hutterer <peter.hutterer@who-t.net>
49031Date: Mon Jul 5 16:54:48 2010 +1000
49032
49033 dix: treat flags as flags, not as value in key_is_down.
49034
49035 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49036 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
49037
49038commit b46ffd25d3f211e91c67bc618ecbd58257939388
49039Author: Peter Hutterer <peter.hutterer@who-t.net>
49040Date: Fri Jul 2 15:25:14 2010 +1000
49041
49042 mi: rename miPointerMoved to miPointerMoveNoEvent.
49043
49044 Having miPointerMove and miPointerMoved is confusing, especially since both
49045 do the same thing bar the event delivery. Also, miPointerMove calls
49046 miPointerMoved which indicates some confusion in the temporal alignment of
49047 cause and effect.
49048
49049 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49050 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
49051 Reviewed-by: Keith Packard <keithp@keithp.com>
49052
49053commit dbd621705a5211540b353af81c4af83c297b74dc
49054Author: Peter Hutterer <peter.hutterer@who-t.net>
49055Date: Fri Jul 2 14:22:03 2010 +1000
49056
49057 mi: De-duplicate some code in mipointer.c
49058
49059 miPointerMoved already has the same code.
49060
49061 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49062 Reviewed-by: Keith Packard <keithp@keithp.com>
49063
49064commit c65280ce8df4836bd7424a90482e8aa00ab6f447
49065Author: Robert Hooker <sarvatt@ubuntu.com>
49066Date: Sat Jun 12 20:23:09 2010 -0400
49067
49068 Increase advertised RENDER protocol minor version to 11
49069
49070 Support for the blend mode operators was added in
49071 0ce42adbf4cff9e7f049d9fc79d588ece5936177
49072 and the requirement was bumped but when things were split off into
49073 include/protocol-versions.h it defined it to 10. render uses
49074 the lower of the client and server advertised versions so it's not
49075 using the new blend mode operators.
49076
49077 Signed-off-by: Robert Hooker <sarvatt@ubuntu.com>
49078 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
49079 Signed-off-by: Keith Packard <keithp@keithp.com>
49080
49081commit 9f0b193acdc29e491b6245390cf9f53b5222e6d3
49082Author: Keith Packard <keithp@keithp.com>
49083Date: Thu Jul 1 09:13:43 2010 -0400
49084
49085 miDbe window priv priv is pre-allocated, don't use dixSetPrivate (bug 28639)
49086
49087 miDbeInit pre-allocates space in each DBE window private private for a
49088 MiDbeWindowPrivPrivRec. miDbeAllocBackBufferName used the
49089 pre-allocated space correctly (simply fetching it instead of
49090 allocating a new piece of memory). However, it then called
49091 dixSetPrivate anyways, which isn't necessary, and (in the new
49092 dixPrivate world) causes an assert failure.
49093
49094 Signed-off-by: Keith Packard <keithp@keithp.com>
49095 Tested-by: Magnus Kessler <Magnus.Kessler@gmx.net>
49096 Reviewed-by: Magnus Kessler <Magnus.Kessler@gmx.net>
49097
49098commit a94cb400d15b8c78dc04148cbd8db8e5ec8364b5
49099Author: Keith Packard <keithp@keithp.com>
49100Date: Thu Jul 1 09:11:36 2010 -0400
49101
49102 Delete unused miDbe screen private private datatype
49103
49104 MiDbeScreenPrivPrivRec is not used in the server. Remove it, along
49105 with the MI_DBE_SCREEN_PRIV_PRIV macro that tried to use it.
49106
49107 Signed-off-by: Keith Packard <keithp@keithp.com>
49108 Reviewed-by: Magnus.Kessler <Magnus.Kessler@gmx.net>
49109
49110commit f0fcffe55f280add5e4db2f5e9198a48c6f1b015
49111Author: Alan Coopersmith <alan.coopersmith@oracle.com>
49112Date: Thu Jun 3 19:00:54 2010 -0700
49113
49114 Update the sprite immediately when moving it with MouseKeys
49115
49116 Fix for OpenSolaris bug 6949755: Mouse Keys are ununusable
49117 and possibly https://bugs.freedesktop.org/show_bug.cgi?id=24856
49118
49119 Ensures waitForUpdate is False before calling SetCursorPosition.
49120 Normally waitForUpdate is False when SilkenMouse is active, True
49121 when it's not. When it's True, the mouse cursor position on
49122 screen is not updated immediately.
49123
49124 This is more critical on Solaris, since we disabled SigIO, thus in turn
49125 disable SilkenMouse, due to the SSE2 vs. signal handler issues described in
49126 Sun bugs 6849925, 6859428, and 6879897.
49127
49128 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
49129 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
49130 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49131
49132commit 3d9079b898c432a87f9b95c1f39a85f660bf0858
49133Author: Alan Coopersmith <alan.coopersmith@oracle.com>
49134Date: Thu Jun 3 19:00:53 2010 -0700
49135
49136 Add API to update setting of waitForUpdate screen private in miPointer
49137
49138 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
49139 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
49140 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49141
49142commit 69b2b5c85ec079ef49f84722daa5f148cedc2e1b
49143Merge: b90faa7 1432785
49144Author: Keith Packard <keithp@keithp.com>
49145Date: Thu Jul 1 23:46:53 2010 -0400
49146
49147 Merge remote branch 'whot/for-keith'
49148
49149commit b90faa71567c4461b28515756ba5c1e6286dda16
49150Author: Keith Packard <keithp@keithp.com>
49151Date: Thu Jul 1 23:46:27 2010 -0400
49152
49153 Revert "xkb: merge lockedPtrButtons state from all attached SDs."
49154
49155 Preparing to merge Peter's branch.
49156
49157 This reverts commit 6052710670953b43b4fff5d101b727163fcb1187.
49158
49159commit 9fb0785449b287ba1998e08613b3c2102ec24842
49160Author: Keith Packard <keithp@keithp.com>
49161Date: Thu Jul 1 23:45:50 2010 -0400
49162
49163 Revert "Revert "dix: use the event mask of the grab for TryClientEvents.""
49164
49165 Preparing to merge Peter's branch.
49166
49167 This reverts commit 018c878e9495b21146c8f38617fdd1bf6d8cc73b.
49168
49169commit 48cac27870992f6bde2c48429ff03c0a7606d5c1
49170Author: James Jones <jajones@nvidia.com>
49171Date: Fri Jun 18 17:28:15 2010 -0700
49172
49173 Cast void* to pointer* to appease some compilers.
49174
49175 When this privates.h is included in C++ builds, the compiler
49176 complains about implicitly casting void* to void**. This small
49177 patch fixes that up.
49178
49179 Signed-off-by: James Jones <jajones@nvidia.com>
49180 Reviewed-by: Keith Packard <keithp@keithp.com>
49181 Signed-off-by: Keith Packard <keithp@keithp.com>
49182
49183commit 14327858391ebe929b806efb53ad79e789361883
49184Author: Peter Hutterer <peter.hutterer@who-t.net>
49185Date: Thu Jul 1 12:44:57 2010 +1000
49186
49187 xkb: release XTEST pointer buttons on physical releases. (#28808)
49188
49189 If a button release event is posted for the MD pointer, post a release event
49190 through the matching XTEST device. This way, a client who posts a button
49191 press through the XTEST extension cannot inadvertedly lock the button.
49192
49193 This behaviour is required for historical reasons, until server 1.7 the core
49194 pointer would release a button press on physical events, regardless of the
49195 XTEST state. Clients seem to rely on this behaviour, causing seemingly stuck
49196 grabs.
49197
49198 The merged behaviour is kept for multiple keyboard PointerKey events, if two
49199 physical keyboards hold the button down as a result of PointerKey actions,
49200 the button is not released until the last keyboard releases the button.
49201
49202 X.Org Bug 28808 <http://bugs.freedesktop.org/show_bug.cgi?id=28808>
49203
49204 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49205
49206commit a71dbc03e65cf7b0654a6eca93ce0bf6a1711ffa
49207Author: Keith Packard <keithp@keithp.com>
49208Date: Thu Jul 1 08:27:05 2010 -0400
49209
49210 Bump to version 1.8.99.904 (1.9 RC4)
49211
49212 Signed-off-by: Keith Packard <keithp@keithp.com>
49213
49214commit 5ea497fc32202ff19c2b7d8ef4eba08889108a87
49215Author: Tiago Vignatti <tiago.vignatti@nokia.com>
49216Date: Tue Jun 29 16:08:02 2010 +0300
49217
49218 xfree86: configure: remove vendor and card name matching rules
49219
49220 Although vendor and board naming are used to create the configure file, the
49221 server doesn't actually use it when fetching such file and probing devices.
49222
49223 Reported-by: Richard Barnette <jrbarnette@chromium.org>
49224 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
49225 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49226 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
49227 Tested-by: Richard Barnette <jrbarnette@chromium.org>
49228 Signed-off-by: Keith Packard <keithp@keithp.com>
49229
49230commit 90fd9e40b59e69333d250a795998f44dc3a5dc0c
49231Author: Tiago Vignatti <tiago.vignatti@nokia.com>
49232Date: Tue Jun 29 16:08:01 2010 +0300
49233
49234 xfree86: pci: remove superfluous vendor and card name logging
49235
49236 X server suffers in startup time when relying on the pciaccess's linear search
49237 to fetch vendor and video device name from PCI ID file (when existent). Such
49238 names are only used to write the log, which may be superfluous. This
49239 information often is provided by the drivers or likewise users can get the it
49240 using external tools like lspci or scanpci.
49241
49242 This patch remove the references of those functions from X start up.
49243
49244 Reported-by: Richard Barnette <jrbarnette@chromium.org>
49245 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
49246 Tested-by: Richard Barnette <jrbarnette@chromium.org>
49247 Reviewed-by: James Cloos <cloos@jhcloos.com>
49248 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49249 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
49250 Signed-off-by: Keith Packard <keithp@keithp.com>
49251
49252commit 339f62b1bfadb0ee77d67e351f4e30f5d5e9625f
49253Author: Peter Hutterer <peter.hutterer@who-t.net>
49254Date: Tue Jun 29 15:24:51 2010 +1000
49255
49256 xkb: emulate PointerKeys events only on the master device.
49257
49258 This patch replicates the behaviour for button events. Only generate a
49259 PointerKeys motion event on the master device, not on the slave device.
49260 Fixes the current issue of PointerKey motion events generating key events as
49261 well.
49262
49263 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49264
49265commit 69ac909878ef80bb74c4a9ca4150eda66debd754
49266Author: Peter Hutterer <peter.hutterer@who-t.net>
49267Date: Tue Jun 29 12:12:53 2010 +1000
49268
49269 xkb: merge lockedPtrButtons state from all attached SDs.
49270
49271 Problem:
49272 lockedPtrButtons keeps the state of the buttons locked by a PointerKeys button
49273 press. Unconditionally clearing the bits may cause stuck buttons in this
49274 sequence of events:
49275
49276 1. type Shift + NumLock to enable PointerKeys
49277 2. type 0/Ins on keypad to emulate Button 1 press
49278 → button1 press event to client
49279 3. press and release button 1 on physical mouse
49280 → button1 release event to client
49281
49282 Button 1 on the MD is now stuck and cannot be released.
49283
49284 Cause:
49285 XKB PointerKeys button events are posted through the XTEST pointer device.
49286 Once a press is generated, the XTEST device's button is down. The DIX merges
49287 the button state of all attached SDs, hence the MD will have a button down
49288 while the XTEST device has a button down.
49289
49290 PointerKey button events are only generated on the master device to avoid
49291 duplicate events (see XkbFakeDeviceButton()). If the MD has the
49292 lockedPtrButtons bit cleared by a release event on a physical device, no
49293 such event is generated when a keyboard device triggers the PointerKey
49294 ButtonRelease trigger. Since the event - if generated - is posted through
49295 the XTEST pointer device, lack of a generated ButtonRelease event on the
49296 XTEST pointer device means the button is never released, resulting in the
49297 stuck button observed above.
49298
49299 Solution:
49300 This patch merges the MD's lockedPtrButtons with the one of all attached
49301 slave devices on release events. Thus, as long as one attached keyboard has
49302 a lockedPtrButtons bit set, this bit is kept in the MD. Once a PointerKey
49303 button is released on all keyboards, the matching release event is emulated
49304 from the MD through the XTEST pointer device, thus also releasing the button
49305 in the DIX.
49306
49307 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49308
49309commit 09645864f5a52882eee51c801b3e610d683e7147
49310Author: Peter Hutterer <peter.hutterer@who-t.net>
49311Date: Tue Jun 29 13:49:27 2010 +1000
49312
49313 xkb: Mark switch case fallthrough with comment.
49314
49315 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49316
49317commit c7330ecb5d28d7a92d24feb289f7f1812ce055a4
49318Author: Peter Hutterer <peter.hutterer@who-t.net>
49319Date: Wed Jun 30 13:23:14 2010 +1000
49320
49321 dix: fix up erroneous error message.
49322
49323 (WW) Device 'device name' has 36 axes, only using first 36.
49324 does seem a bit silly.
49325
49326 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49327
49328commit dbf249ec6638f0a8dfa4c2286099845aafc8ac88
49329Author: Peter Hutterer <peter.hutterer@who-t.net>
49330Date: Tue Jun 29 10:43:51 2010 +1000
49331
49332 xkb: remove now obsolete comment.
49333
49334 Looks like nothing broke from removing the hardcoded CoreProcessPointerEvent
49335 call. Whoop. Di. Doo.
49336
49337 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49338
49339commit f028e70ca714f6956e41754f132cb9b8a1e8db63
49340Author: Keith Packard <keithp@keithp.com>
49341Date: Wed Jun 30 08:33:55 2010 -0700
49342
49343 Initialize dev privates before using any
49344
49345 Initializing the dev privates code after allocating the server client
49346 dev privates would cause the memory leak check to fire at server
49347 startup or reset.
49348
49349 Signed-off-by: Keith Packard <keithp@keithp.com>
49350 Acked-by: Daniel Stone <daniel@fooishbar.org>
49351
49352commit 433d0851cd3e61d841ff374ee0a0f052d5907029
49353Author: Julien Cristau <jcristau@debian.org>
49354Date: Fri Jun 25 11:34:44 2010 +0100
49355
49356 Xephyr: fix Xv adaptor capability tests
49357
49358 Signed-off-by: Julien Cristau <jcristau@debian.org>
49359 Reviewed-by: Keith Packard <keithp@keithp.com>
49360 Signed-off-by: Keith Packard <keithp@keithp.com>
49361
49362commit 9626eedebf620559652ffb1fefa82b5d659e57be
49363Author: Julien Cristau <jcristau@debian.org>
49364Date: Thu Jun 24 18:45:11 2010 +0100
49365
49366 configure: bail if Xephyr was requested but its dependencies are missing
49367
49368 Signed-off-by: Julien Cristau <jcristau@debian.org>
49369 Reviewed-by: Jamey Sharp <jamey@minilop.net>
49370 Signed-off-by: Keith Packard <keithp@keithp.com>
49371
49372commit e6531ae9d5bdd37775e921268366fb76056e22d5
49373Author: Julien Cristau <jcristau@debian.org>
49374Date: Thu Jun 24 18:45:10 2010 +0100
49375
49376 Xephyr: fix memory leak in XF86DRIOpenConnection
49377
49378 The allocated bus id string was not being freed.
49379
49380 Signed-off-by: Julien Cristau <jcristau@debian.org>
49381 Reviewed-by: Jamey Sharp <jamey@minilop.net>
49382 Signed-off-by: Keith Packard <keithp@keithp.com>
49383
49384commit 390a8466dd1914f4786b811ff8454f6e0c4b6b04
49385Author: Ville Syrjälä <ville.syrjala@nokia.com>
49386Date: Mon Jun 28 23:26:48 2010 +0300
49387
49388 xfree86/modes: Allow the driver to specify initial rotation
49389
49390 When the "Rotate" option isn't specified allow the driver to specify
49391 the initial rotation mode. This way the driver can choose to retain
49392 the same settings that were used by software that was used prior to
49393 starting X.
49394
49395 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
49396 Reviewed-by: Keith Packard <keithp@keithp.com>
49397 Signed-off-by: Keith Packard <keithp@keithp.com>
49398
49399commit 6052710670953b43b4fff5d101b727163fcb1187
49400Author: Keith Packard <keithp@keithp.com>
49401Date: Wed Jun 30 08:21:04 2010 -0700
49402
49403 xkb: merge lockedPtrButtons state from all attached SDs.
49404
49405 Problem:
49406 lockedPtrButtons keeps the state of the buttons locked by a PointerKeys button
49407 press. Unconditionally clearing the bits may cause stuck buttons in this
49408 sequence of events:
49409
49410 1. type Shift + NumLock to enable PointerKeys
49411 2. type 0/Ins on keypad to emulate Button 1 press
49412 → button1 press event to client
49413 3. press and release button 1 on physical mouse
49414 → button1 release event to client
49415
49416 Button 1 on the MD is now stuck and cannot be released.
49417
49418 Cause:
49419 XKB PointerKeys button events are posted through the XTEST pointer device.
49420 Once a press is generated, the XTEST device's button is down. The DIX merges
49421 the button state of all attached SDs, hence the MD will have a button down
49422 while the XTEST device has a button down.
49423
49424 PointerKey button events are only generated on the master device to avoid
49425 duplicate events (see XkbFakeDeviceButton()). If the MD has the
49426 lockedPtrButtons bit cleared by a release event on a physical device, no
49427 such event is generated when a keyboard device triggers the PointerKey
49428 ButtonRelease trigger. Since the event - if generated - is posted through
49429 the XTEST pointer device, lack of a generated ButtonRelease event on the
49430 XTEST pointer device means the button is never released, resulting in the
49431 stuck button observed above.
49432
49433 Solution:
49434 This patch merges the MD's lockedPtrButtons with the one of all attached
49435 slave devices on release events. Thus, as long as one attached keyboard has
49436 a lockedPtrButtons bit set, this bit is kept in the MD. Once a PointerKey
49437 button is released on all keyboards, the matching release event is emulated
49438 from the MD through the XTEST pointer device, thus also releasing the button
49439 in the DIX.
49440
49441 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49442 Reviewed-by: Keith Packard <keithp@keithp.com>
49443 Signed-off-by: Keith Packard <keithp@keithp.com>
49444
49445commit 28e33ae6f69f716ece5d68e63fc52557236c5f6e
49446Author: Jesse Barnes <jbarnes@virtuousgeek.org>
49447Date: Wed Jun 30 07:59:04 2010 -0700
49448
49449 OS support: fix writeable client vs IgnoreClient behavior
49450
49451 When ResetCurrentRequest is called, or IgnoreClient is called when a
49452 client has input pending, IgnoredClientsWithInput will be set. However,
49453 a subsequent IgnoreClient request will clear the client fd from that fd
49454 set, potentially causing the client to hang.
49455
49456 So add an Ignore/Attend count, and only apply the ignore logic on the
49457 first ignore and the attend logic on the last attend. This is
49458 consistent with the comments for these functions; callers must pair
49459 them.
49460
49461 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=27035.
49462
49463 Reviewed-by: Keith Packard <keithp@keithp.com>
49464 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
49465 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
49466 Signed-off-by: Keith Packard <keithp@keithp.com>
49467
49468commit 018c878e9495b21146c8f38617fdd1bf6d8cc73b
49469Author: Peter Hutterer <peter.hutterer@who-t.net>
49470Date: Fri Jun 25 09:48:10 2010 +1000
49471
49472 Revert "dix: use the event mask of the grab for TryClientEvents."
49473
49474 Behaviour of earlier X servers was to deliver the ButtonPress event
49475 unconditionally, regardless of the actual event mask being set. Thus, a
49476 GrabButton event will always deliver the button press event, a GrabKey
49477 always the key press event, etc. Same goes for XI and XI2.
49478
49479 Reproducible with a simple client requesting a button grab in the form of:
49480 XGrabButton(dpy, AnyButton, AnyModifier, win, True, ButtonReleaseMask,
49481 GrabModeAsync, GrabModeAsync, None, None);
49482
49483 On servers before MPX/XI2, the client will receive a button press and
49484 release event. On current servers, the client receives only the release.
49485 Clients that expect the press event to be delivered unconditionally may
49486 leave the user with a stuck grab.
49487
49488 XTS test results for XGrabButton are identical with and without this patch.
49489
49490 This reverts commit 48585bd1e3e98db0f3df1ecc68022510216e00cc.
49491
49492 Conflicts:
49493
49494 dix/events.c
49495
49496 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49497 Reviewed-by: Keith Packard <keithp@keithp.com>
49498 Signed-off-by: Keith Packard <keithp@keithp.com>
49499
49500commit 1884db430a5680e37e94726dff46686e2218d525
49501Author: Peter Hutterer <peter.hutterer@who-t.net>
49502Date: Thu Jun 24 12:52:53 2010 +1000
49503
49504 Revert "dix: use the event mask of the grab for TryClientEvents."
49505
49506 Behaviour of earlier X servers was to deliver the ButtonPress event
49507 unconditionally, regardless of the actual event mask being set. This is
49508 documented in the protocol:
49509 "This request establishes a passive grab. In the future, the pointer is
49510 actively grabbed as described in GrabPointer, the last-pointer-grab time is
49511 set to the time at which the button was pressed (as transmitted in the
49512 ButtonPress event), and the ButtonPress event is reported if all of the
49513 following conditions are true:
49514 <list of conditions, event mask is not one of them>"
49515
49516 Thus, a GrabButton event will always deliver the button press event, a
49517 GrabKey always the key press event, etc. Same goes for XI and XI2.
49518
49519 Reproducible with a simple client requesting a button grab in the form of:
49520 XGrabButton(dpy, AnyButton, AnyModifier, win, True, ButtonReleaseMask,
49521 GrabModeAsync, GrabModeAsync, None, None);
49522
49523 On servers before MPX/XI2, the client will receive a button press and
49524 release event. On current servers, the client receives only the release.
49525 Clients that expect the press event to be delivered unconditionally.
49526
49527 XTS Xlib13 XGrabButton 5/39 now passes.
49528
49529 This reverts commit 48585bd1e3e98db0f3df1ecc68022510216e00cc.
49530 Effectively reverts commit 1c612acca8568fcdf9761d23f112adaf4d496f1b as well,
49531 the code introduced with 1c612 is not needed anymore.
49532
49533 Conflicts:
49534
49535 dix/events.c
49536
49537 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49538 Acked-by: Daniel Stone <daniel@fooishbar.org>
49539 Reviewed-by: Keith Packard <keithp@keithp.com>
49540
49541commit 3b3c77b87070ddcdbb2acb114a81628485e7a129
49542Author: Tiago Vignatti <tiago.vignatti@Nokia.com>
49543Date: Wed Jun 23 15:18:04 2010 +0300
49544
49545 configure: don't check xfont always for each server
49546
49547 It's already defined inside REQUIRED_LIBS and all DDX are getting it already.
49548 No semantical changes.
49549
49550 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
49551 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49552 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
49553 Signed-off-by: Keith Packard <keithp@keithp.com>
49554
49555commit 75536ee80595c79bba95a1fb6844126ee08486d4
49556Author: Tiago Vignatti <tiago.vignatti@nokia.com>
49557Date: Wed Jun 23 15:18:03 2010 +0300
49558
49559 dix: use one single function to register fpe fonts
49560
49561 X server doesn't need to understand fpe internals, so use
49562 register_fpe_functions from libXfont.
49563
49564 It's required to get new version of libXfont, therefore adjust it to be passed
49565 to autoconf.
49566
49567 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
49568 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49569 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
49570 Signed-off-by: Keith Packard <keithp@keithp.com>
49571
49572commit a68a101a730df7f73421555e381a50b074607c5e
49573Author: Keith Packard <keithp@keithp.com>
49574Date: Tue Jun 22 12:13:26 2010 -0700
49575
49576 Bump to 1.8.99.903 -- 1.9 RC3
49577
49578 Signed-off-by: Keith Packard <keithp@keithp.com>
49579
49580commit 2c0159d21788b16607e6ebe4571b57fe78545c2a
49581Author: Kristian Høgsberg <krh@bitplanet.net>
49582Date: Mon Jun 14 09:25:22 2010 -0400
49583
49584 list.h: Fix list_for_each_entry_safe()
49585
49586 Can't use next as a macro argument since we're accessing the .next field
49587 of struct list.
49588
49589 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
49590 Reviewed-by: Keith Packard <keithp@keithp.com>
49591 Signed-off-by: Keith Packard <keithp@keithp.com>
49592
49593commit e2918b67395a48397bf7db545584cc2d4f4e7ccf
49594Author: Keith Packard <keithp@keithp.com>
49595Date: Tue Jun 22 12:11:20 2010 -0700
49596
49597 Revert "mi: Remove unused overlay support"
49598
49599 This reverts commit 9120e58556014e514eef9dfe43cdea4b74dc66ae.
49600
49601 Whoops, please revert this patch -- overlay is in use in nvidia drivers,
49602 and it's too late in release cycle to remove it.
49603
49604 I feel really sorry that I kept this patch in my tree. I will submit
49605 another one, removing only XAA overlay hooks which are not used at all.
49606
49607commit 7673451ade46ebacfdfc02a5190149210198fdd8
49608Author: Keith Packard <keithp@keithp.com>
49609Date: Tue Jun 22 11:49:30 2010 -0700
49610
49611 Bump to 1.8.99.902 -- 1.9 RC2
49612
49613 Signed-off-by: Keith Packard <keithp@keithp.com>
49614
49615commit 8b65f2edb407521e0ac0e207e5d5358986bd46da
49616Author: Rami Ylimäki <ext-rami.ylimaki@nokia.com>
49617Date: Tue Jun 22 14:57:20 2010 +0300
49618
49619 record: Prevent a crash on recording client disconnect.
49620
49621 Execute the following steps to reproduce the issue.
49622
49623 1. Run at least two recording clients simultaneously.
49624 $ cnee --record --request-range 1-127 &
49625 $ cnee --record --request-range 1-127 &
49626 2. Kill the recording clients.
49627 $ killall cnee
49628 3. Give X server something to do so that the clients are closed.
49629 $ xinput list
49630 $ xinput list
49631
49632 As a result RecordUninstallHooks accesses NullClient, because
49633 RecordAClientStateChange doesn't clean the recording clients up
49634 properly.
49635
49636 Fix RecordUninstallHooks to fail locally on an assertion instead of
49637 much later in privates code, if NullClient is still accessed because
49638 of some other bug. Fix RecordAClientStateChange to iterate through all
49639 contexts so that modifications of the iterated array during iteration
49640 don't cause contexts to be skipped.
49641
49642 Signed-off-by: Rami Ylimäki <ext-rami.ylimaki@nokia.com>
49643 Reviewed-by: Keith Packard <keithp@keithp.com>
49644 Signed-off-by: Keith Packard <keithp@keithp.com>
49645
49646commit 660f6ab5494a728c3ca7ba00c305e9ff06c8ecb2
49647Author: Simon Farnsworth <simon.farnsworth@onelan.com>
49648Date: Tue Jun 22 10:13:30 2010 +0100
49649
49650 Don't crash when asked if a client that has disconnected was local
49651
49652 ProcDRI2Dispatch uses LocalClient to determine if it's safe to respond
49653 to a client that has made DRI2 requests which aren't sensible for
49654 remote clients (anything but version). When the client has disappeared
49655 mid-request stream (e.g. as a result of a kill -9, or a client-side
49656 bug), LocalClient causes the X server to follow suit, as
49657 ((OsCommPtr)client->osPrivate)->trans_conn is NULL at this point.
49658
49659 The simple and obvious fix is to just return "not local" when
49660 trans_conn is NULL, which fixes the crash I was seeing; however Keith
49661 Packard pointed out that just checking trans_conn isn't enough;
49662 quoting Keith:
49663
49664 "This looks almost right to me -- I reviewed the os code to see when
49665 _XSERVTransClose is called (which is what frees the trans_conn data) and
49666 found that every place which called that immediately set trans_conn to
49667 NULL, except for the call in CloseDownFileDescriptor which is only
49668 called from CloseDownConnection and which is immediately followed by
49669 freeing the OsCommRec and setting client->osPrivate to NULL. So, I'd
49670 suggest checking client->osPrivate in addition to the above check."
49671
49672 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.com>
49673 Reviewed-by: Keith Packard <keithp@keithp.com>
49674 Signed-off-by: Keith Packard <keithp@keithp.com>
49675
49676commit 82d41ada993d8cbdcdfea878d1a5b031afe4e593
49677Author: Dave Airlie <airlied@redhat.com>
49678Date: Mon Jun 21 16:33:55 2010 +1000
49679
49680 composite: fix freeing of old pixmap until after move/resize/cbw (bug 28345)
49681
49682 The fixes for the composite reallocation failure, were freeing the oldpixmap
49683 straight after reallocating the new one, however this led to some wierd
49684 effects in e16 compositing, and I think in a few other places. This patch
49685 moves the freeing of the pixmap into the post wrapped stage. I'm not sure if
49686 we are actually better off breaking ABI and adding another callback from the
49687 ConfigureWindow to composite to make sure the old pixmap is always freed,
49688 but this should be satisfactory as we should always hit one of the freeing
49689 paths or else its a bug in the DIX.
49690
49691 bug: https://bugs.freedesktop.org/show_bug.cgi?id=28435
49692
49693 Reported-by: Andrew Randrianasulu <randrik@mail.ru>
49694 Signed-off-by: Dave Airlie <airlied@redhat.com>
49695 Reviewed-by: Keith Packard <keithp@keithp.com>
49696 Signed-off-by: Keith Packard <keithp@keithp.com>
49697
49698commit c6bc52cb6663138d1273447cb7661fa6e958f539
49699Author: Dave Airlie <airlied@redhat.com>
49700Date: Mon Jun 21 14:08:50 2010 +1000
49701
49702 dri2: avoid crash with old dri drivers.
49703
49704 I built latest server and copied over a mesa 7.8 dri2 driver to use with it,
49705 and it crashed in here starting compiz. Looks like we need to validate the
49706 flush extension is registered before calling invalidate.
49707
49708 Signed-off-by: Dave Airlie <airlied@redhat.com>
49709 Reviewed-by: Michel Dänzer <michel@daenzer.net>
49710 Reviewed-by: Keith Packard <keithp@keithp.com>
49711 Signed-off-by: Keith Packard <keithp@keithp.com>
49712
49713commit 80d1a548d6ce73c2ff097536c1bc7044bf74965d
49714Author: Dave Airlie <airlied@redhat.com>
49715Date: Mon Jun 21 10:05:08 2010 +1000
49716
49717 rotation: fix cursor and overlap of one pixel.
49718
49719 Commit 77c7a64e8885696665556c9fbcb3cffb552e367a was introduced to fix
49720 a cursor off by one on Intel hw, however it also move the whole crtc
49721 into an off by one position and you could see gnom-eshell overlapping.
49722
49723 This commit reverts that and instead fixes the cursor hotspot
49724 translation to work like pixman does. We add 0.5 to the cursor vector
49725 before translating, and floor the value afterwards.
49726
49727 Thanks to Soeren (ssp) for pointing out where the real problem was
49728 after explaning how pixman translates points.
49729
49730 Signed-off-by: Dave Airlie <airlied@redhat.com>
49731 Reviewed-by: Keith Packard <keithp@keithp.com>
49732 Signed-off-by: Keith Packard <keithp@keithp.com>
49733
49734commit 32c706c4ffd7433dbfc79dba8785b1510d2f053f
49735Author: Gaetan Nadon <memsize@videotron.ca>
49736Date: Sun Jun 20 20:37:26 2010 -0400
49737
49738 config: declare xserver private dependencies in xorg-server.pc
49739
49740 Any module (drivers) depending on xserver also depends on some of the
49741 server private dependencies. Any driver including xf86.h depends on
49742 xext, kbproto, inputproto and randr.
49743
49744 These dependencies are in separate packages, so anything can happen,
49745 removal, wrong version, etc... and the driver fails during compilation.
49746 Having the private dependencies declared will ensure all packages the
49747 server depends on are present and at the correct version.
49748
49749 Currently each module attempts to check for server dependencies with
49750 various degrees of accuracy. With this patch, the driver will only need
49751 to check for its own explicit dependencies.
49752
49753 Now that xproto is included in Requires.private it is removed from Requires.
49754 All the cflags from both Requires and Requires.private are returned
49755 to caller to pkg-config.
49756
49757 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
49758 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
49759 Signed-off-by: Keith Packard <keithp@keithp.com>
49760
49761commit 37734ba79ca3a9bad1c1e29d89710c8f9c299210
49762Author: Colin Harrison <colin.harrison@virgin.net>
49763Date: Wed Jun 16 18:07:04 2010 +0100
49764
49765 XWin: Fixes for devPrivates API change
49766
49767 Fix some typos in devPrivates API changes in XWin code.
49768 Move allocation of private keys as it's no longer valid to do them during OsVendorInit()
49769
49770 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
49771 Reviewed-by: Jamey Sharp <jamey@minilop.net>
49772 Signed-off-by: Keith Packard <keithp@keithp.com>
49773
49774commit f7dae972aac28c5792566e997c40254e787d246a
49775Author: Ville Syrjälä <ville.syrjala@nokia.com>
49776Date: Tue Jun 15 23:25:34 2010 +0300
49777
49778 xv: Don't send port notify when SetPortAttribute fails
49779
49780 Currently a port notify event is sent even if SetPortAttribute
49781 fails. Furthermore the value field in the event will contain the
49782 value that was specified in the failed request. So any client
49783 interested in the actual value of the attribute will have to
49784 double check the current value with GetPortAttribute after
49785 receiving a port notify event.
49786
49787 Fix the problem by sending port notifications only when
49788 SetPortAttribute succeeds.
49789
49790 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
49791 Reviewed-by: Keith Packard <keithp@keithp.com>
49792 Signed-off-by: Keith Packard <keithp@keithp.com>
49793
49794commit c20166739603f86888f28ca2c65f5cf474d0d8c9
49795Author: Mikhail Gusarov <dottedmag@dottedmag.net>
49796Date: Sat Jun 12 01:05:07 2010 +0700
49797
49798 os: Remove unused dev_tty_from_init variable
49799
49800 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49801 Reviewed-by: Jamey Sharp <jamey@minilop.net>
49802
49803commit 2e15f1903dac8473285ad2937c079a08cf7c5d31
49804Author: Mikhail Gusarov <dottedmag@dottedmag.net>
49805Date: Sat Jun 12 00:53:38 2010 +0700
49806
49807 os: Remove unused pread/pwrite/lockit functions
49808
49809 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49810 Reviewed-by: Jamey Sharp <jamey@minilop.net>
49811
49812commit 9120e58556014e514eef9dfe43cdea4b74dc66ae
49813Author: Mikhail Gusarov <dottedmag@dottedmag.net>
49814Date: Sat Jun 12 00:30:27 2010 +0700
49815
49816 mi: Remove unused overlay support
49817
49818 The only reference to it in server and drivers is in XAA overlay code which
49819 would segfault as no miInitOverlay is called ever. No segfaults were observed
49820 "in wild", so XAA overlay is probably also unused.
49821
49822 XAA code is modified to act as if miOverlayCopyUnderlay always returned false,
49823 because XAACopyWindow8_32 could only set doUnderlay to true if it's called from
49824 miOverlayMoveWindow or miOverlayResizeWindow, which can only be called if
49825 miInitOverlay has hooked those functions, and no driver (on fd.o) or server code
49826 calls that.
49827
49828 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49829 Reviewed-by: Jamey Sharp <jamey@minilop.net>
49830
49831commit 91b5aadbdfd9d05ca1ffdeb443e602ecdba1e04d
49832Author: Mikhail Gusarov <dottedmag@dottedmag.net>
49833Date: Fri Jun 11 20:20:00 2010 +0700
49834
49835 mi: do not use X11/extensions/shape.h header from libXext
49836
49837 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49838 Tested-by: Dan Nicholson <dbn.lists@gmail.com>
49839
49840commit b8615d592700b7be319c04cc0563fdeb5a266534
49841Author: Alan Coopersmith <alan.coopersmith@oracle.com>
49842Date: Sat Jun 12 08:19:16 2010 -0700
49843
49844 Don't coredump on "X -showopts" (bug 25874)
49845
49846 Don't try walking the xf86ConfigLayout.screens table if it's empty
49847 https://bugs.freedesktop.org/show_bug.cgi?id=25874
49848
49849 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
49850 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
49851 Signed-off-by: Keith Packard <keithp@keithp.com>
49852
49853commit 32fd57509c3e06f63ac6b0ad607767fd2b688e5a
49854Author: Rami Ylimäki <ext-rami.ylimaki@nokia.com>
49855Date: Fri Jun 18 12:40:58 2010 +0300
49856
49857 record: Register client private to prevent crash.
49858
49859 Running the following command causes an assertion to fail:
49860 cnee --record --request-range 1-127
49861
49862 3 in dixGetPrivateAddr at ../../include/privates.h:122
49863 4 in dixGetPrivate at ../../include/privates.h:137
49864 5 in dixLookupPrivate at ../../include/privates.h:167
49865 6 in RecordInstallHooks at ../../record/record.c:911
49866 7 in ProcRecordEnableContext at ../../record/record.c:2377
49867 8 in ProcRecordDispatch at ../../record/record.c:2571
49868 9 in Dispatch at ../../dix/dispatch.c:432
49869 10 in main at ../../dix/main.c:289
49870
49871 Signed-off-by: Rami Ylimäki <ext-rami.ylimaki@nokia.com>
49872 Reviewed-by: Oliver McFadden <oliver.mcfadden@nokia.com>
49873 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
49874 Signed-off-by: Keith Packard <keithp@keithp.com>
49875
49876commit 75beadd766fed7b12a76e59e57c244e297c2d2cb
49877Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
49878Date: Sun Jun 13 18:05:26 2010 +0200
49879
49880 DRI2/xserver: Don't hang in glXSwapBuffers if drawable moves between crtc's (bug 28383)
49881
49882 Detect if a drawable has been moved from an original crtc to a new crtc
49883 with a lower current vblank count than the original crtc inbetween
49884 glXSwapBuffers() calls. Reinitialize drawable's last_swap_target
49885 before scheduling next swap if such a move has taken place.
49886
49887 last_swap_target defines the baseline for scheduling the next swap.
49888 If a movement between crtc's is not taken into account, the swap may
49889 schedule for a vblank count on the new crtc far in the future, resulting
49890 in a apparent "hang" of the drawable for a long time.
49891
49892 Fixes Bugzilla bug #28383.
49893
49894 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
49895 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
49896 Signed-off-by: Keith Packard <keithp@keithp.com>
49897
49898commit 868bf3de478fac0080d525137fc728c551c78ea6
49899Author: Peter Hutterer <peter.hutterer@who-t.net>
49900Date: Tue Jun 15 13:22:20 2010 +1000
49901
49902 dix: the default axis mode is Relative, not "0".
49903
49904 Yay for readability.
49905
49906 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49907 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
49908
49909commit db0d3d4e573c93a8af43499682ed788c7bd0d70d
49910Author: Xiaoyang Yu (Max) <max.a.yu@intel.com>
49911Date: Fri Jun 4 17:17:53 2010 +0800
49912
49913 Re-enabled Xnest fix for focus in + modifier bug.
49914
49915 * See https://bugs.freedesktop.org/show_bug.cgi?id=3030
49916
49917 Signed-off-by: Xiaoyang Yu (Max) <max.a.yu@intel.com>
49918 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
49919
49920 This takes the xnest way of working around this (see
49921 5904ef2ccd6056b187ca76f104c21e2d686bfc1d "xnest: restore
49922 xnestUpdateModifierState") and copies it to Xephyr.
49923
49924 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
49925
49926commit 505724c5f3e4fc1475fa54f20d08c0c06fb32443
49927Author: Keith Packard <keithp@keithp.com>
49928Date: Mon Jun 7 16:28:34 2010 -0700
49929
49930 In DRI2Connect, check to see if DRI2 has been initialized (bug 28424)
49931
49932 This prevents DRI2GetScreen from being invoked with an uninitialized
49933 private key which would cause an assert failure.
49934
49935 Signed-off-by: Keith Packard <keithp@keithp.com>
49936 Tested-by: David Ronis <David.Ronis@McGill.CA>
49937
49938commit 68a9ee8370e6f9b38218376ac92d5130a5b0ef1e
49939Author: Keith Packard <keithp@keithp.com>
49940Date: Fri May 7 22:56:04 2010 -0700
49941
49942 Clean up RandR12 bits on screen close (bug 27114)
49943
49944 When resetting the server, pScrn->EnterVT must be unwrapped or the
49945 next server generation will end up wrapping the wrapper and causing an
49946 infinite recursion on EnterVT.
49947
49948 Signed-off-by: Keith Packard <keithp@keithp.com>
49949 Tested-by: Michael Stapelberg <michael+freedesktop@stapelberg.de>
49950
49951commit d5ab717768524c8552b81607aaeffb447ab268ee
49952Author: Keith Packard <keithp@keithp.com>
49953Date: Thu Jun 10 19:21:18 2010 -0700
49954
49955 Bump to 1.8.99.901 -- 1.9 RC1
49956
49957 Signed-off-by: Keith Packard <keithp@keithp.com>
49958
49959commit a41d6e9bffbe56cfa1c3b84388a3d9f5a982f1a9
49960Merge: 7e8f100 f4190fe
49961Author: Keith Packard <keithp@keithp.com>
49962Date: Fri Jun 11 10:08:13 2010 -0700
49963
49964 Merge remote branch 'dottedmag/for-keithp'
49965
49966commit 7e8f1001217326cc451974bacf25275420c4bb4e
49967Author: Éric Piel <E.A.B.Piel@tudelft.nl>
49968Date: Fri Jun 11 09:16:32 2010 -0700
49969
49970 exa: fix ExaCheckCopyNtoN for exa_classic when source = dest
49971
49972 In case you want to copy a region with source = dest, you have the same pixmap
49973 as source and dest.
49974
49975 At the end of exaPixmapIsOffscreen_classic() the devPrivate.ptr is reset to
49976 NULL (look at the sources).
49977
49978 Now this is what happens in ExaCheckCopyNtoN:
49979
49980 exaPrepareAccess( pDst );
49981 Calls IsOffscreen()
49982 sets devPrivate.ptr to NULL
49983 sets up devPrivate.ptr to real pointer
49984 Everything OK
49985 exaPrepareAccess( pSrc );
49986 Calls IsOffscreen()
49987 sets devPrivate.ptr to NULL
49988 BAILS OUT CAUSE OF NESTED OPERATION SINCE DST EQUALS SRC
49989
49990 We end up with devPrivate.ptr as NULL, and that is clearly wrong.
49991
49992 In particular this fixes a segfault when using the psb driver (bug 28077)
49993
49994 Signed-off-by: Éric Piel <eric.piel@tremplin-utc.net>
49995 Reviewed-by: Michel Dänzer <michel@daenzer.net>
49996 Signed-off-by: Keith Packard <keithp@keithp.com>
49997
49998commit f4190feb25ecc3d8278decc8bf28a5ef0e568942
49999Author: Matt Turner <mattst88@gmail.com>
50000Date: Thu Jun 10 04:08:54 2010 +0700
50001
50002 Remove more superfluous if(p) checks around free(p)
50003
50004 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50005 Signed-off-by: Matt Turner <mattst88@gmail.com>
50006 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50007
50008commit 89bd05106e5823fc5cfca9abf082729f2444363b
50009Author: Matt Turner <mattst88@gmail.com>
50010Date: Fri Jun 4 21:09:35 2010 -0400
50011
50012 record: move free() to after last use of pContext
50013
50014 No functional change, since free doesn't change the value of the pointer
50015 passed to it, but it makes this code less confusing.
50016
50017 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50018 Signed-off-by: Matt Turner <mattst88@gmail.com>
50019
50020commit a54a766dfb39fb3df671045878ac706215d83cef
50021Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50022Date: Mon Jun 7 03:19:03 2010 +0700
50023
50024 xkb: replace xstrdup with strdup in Win32System
50025
50026 The only caller of Win32System is XkbDDXCompileKeymapByNames. Add allocation
50027 check there to avoid passing NULL pointers to various functions down the code.
50028
50029 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50030 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50031
50032commit 6e7417c342b0624e3f3c5686bb43026786423692
50033Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50034Date: Fri Jun 4 17:05:35 2010 +0700
50035
50036 config: Replace xstrdup with strdup in add_option()
50037
50038 All callers of add_option pass string literal as "key" argument
50039 except one, where non-NULL condition is guarded by if().
50040
50041 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50042 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50043
50044commit e4570f5db5157f4233454c938733a2a0d6a1cb8f
50045Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50046Date: Fri Jun 4 17:01:42 2010 +0700
50047
50048 xfree86: Get rid of xstrdup when argument is definitely non-NULL
50049
50050 ditto for XFree86
50051
50052 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50053 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50054
50055commit 416d228481d71204cf9bfad3ab4773abc4757f79
50056Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50057Date: Fri Jun 4 17:01:25 2010 +0700
50058
50059 xquartz: Get rid of xstrdup when argument is definitely non-NULL
50060
50061 ditto for XQuartz
50062
50063 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50064 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50065
50066commit eea286f2b9e4b1acf2b1c9800f5a4bc7c3fa968d
50067Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50068Date: Fri Jun 4 17:01:05 2010 +0700
50069
50070 xwin: Get rid of xstrdup when argument is definitely non-NULL
50071
50072 ditto for XWin
50073
50074 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50075 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50076
50077commit dd45b7d74696cd4fe9545852640a3c2e66a808fd
50078Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50079Date: Fri Jun 4 17:00:46 2010 +0700
50080
50081 kdrive: Get rid of xstrdup when argument is definitely non-NULL
50082
50083 ditto for Kdrive
50084
50085 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50086 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50087
50088commit 16158366954d945db6263f6de505fc02ee03c6cd
50089Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50090Date: Fri Jun 4 17:00:15 2010 +0700
50091
50092 dmx: Get rid of xstrdup when argument is definitely non-NULL
50093
50094 ditto for DMX
50095
50096 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50097 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50098
50099commit 6592db6bb526f0c43b4c7b55859c629709e039b4
50100Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50101Date: Fri Jun 4 16:58:58 2010 +0700
50102
50103 Get rid of xstrdup when argument is definitely non-NULL
50104
50105 Replace xstrdup with strdup when either constant string is
50106 being duplicated or argument is guarded by conditionals and
50107 obviously can't be NULL
50108
50109 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50110 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50111
50112commit 620ca54aaa0b363fcf68cec1bd6c37e68c988352
50113Merge: fd69234 8ff9b50
50114Author: Keith Packard <keithp@keithp.com>
50115Date: Thu Jun 10 19:18:53 2010 -0700
50116
50117 Merge remote branch 'alanc/master'
50118
50119commit fd69234db004e98c663f5c6c4360d2cecaa8a4df
50120Author: Keith Packard <keithp@keithp.com>
50121Date: Sun Jun 6 15:21:28 2010 -0700
50122
50123 Ignore RandR timestamps harder
50124
50125 Checking timestamps in post 1.1 randr requests was never a good idea,
50126 let's ignore them and just make the configuration changes.
50127
50128 Signed-off-by: Keith Packard <keithp@keithp.com>
50129 Acked-by: Chase Douglas <chase.douglas@canonical.com>
50130
50131commit a8ec9eca850f2a7ad4c5cf31c1c011c120688496
50132Author: Keith Packard <keithp@keithp.com>
50133Date: Mon Jun 7 13:39:11 2010 -0700
50134
50135 Fix a couple more possible errors with input-only windows
50136
50137 Using type == DRAWABLE_WINDOW to differentiate between pixmaps and
50138 windows isn't sufficient as input-only windows will end up in the
50139 pixmap case. This patch changes a few more code paths to use
50140 WindowDrawable instead.
50141
50142 Signed-off-by: Keith Packard <keithp@keithp.com>
50143 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50144
50145commit 8ff9b502cfce3828f7855ffba7949d6ebee34031
50146Author: Alan Coopersmith <alan.coopersmith@oracle.com>
50147Date: Fri Jun 4 08:44:02 2010 -0700
50148
50149 Solaris: avoid memory leak if AGPIOC_INFO ioctl fails
50150
50151 Move malloc after ioctl, so we don't have to worry about free'ing the
50152 memory if the ioctl fails.
50153
50154 [ This bug was found by the Parfait bug checking tool.
50155 For more information see http://research.sun.com/projects/parfait ]
50156
50157 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50158 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50159
50160commit c5eb5d69e5183860185a05cfcce16af635cab9aa
50161Author: Alan Coopersmith <alan.coopersmith@oracle.com>
50162Date: Thu Jun 3 19:28:43 2010 -0700
50163
50164 Record some additional library dependencies in xf86 modules
50165
50166 Helps with symbol resolution when building with -z defs
50167
50168 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50169 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
50170
50171commit 1c08a37e0eb4746e8974eb7a70ca4b7b84712963
50172Author: Sam Lau <sam.lau@oracle.com>
50173Date: Thu Jun 3 19:17:14 2010 -0700
50174
50175 SecurityResource should not segfault when client owning resource has exited
50176
50177 Fixes OpenSolaris bug 6949754:
50178 Xorg crashes when the magnifier is enabled at gdm login greeter window.
50179 http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6949754
50180
50181 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50182 Reviewed-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
50183
50184commit 07a093add0b7e40c4d9b9b59273e3ff9e14a88a7
50185Merge: 84190d2 dc61448
50186Author: Keith Packard <keithp@keithp.com>
50187Date: Thu Jun 10 18:39:10 2010 -0700
50188
50189 Merge remote branch 'whot/for-keith'
50190
50191commit dc614484f93b67e8b62dbb1bb2fd247fe5a4c850
50192Author: Peter Hutterer <peter.hutterer@who-t.net>
50193Date: Thu Jun 10 12:21:36 2010 +1000
50194
50195 Xi: don't copy the modifier key count when copying device classes (#25480)
50196
50197 The modifier key count is maintained by the XKB layer and
50198 increased/decreased for all modifiers that set state.
50199
50200 Test case, MD/SD modifier key count in comment:
50201 1. keyboard 1: press and hold Shift_L # SD:1 MD:1
50202 2. keyboard 2: press and release Shift_L # SD:1,0 MD:1,0
50203 <class copy happens> # SD:1 MD:1
50204 3. keyboard 1: release Shift_L # SD:0 MD:1
50205 4. keyboard 1: press and release Shift_L # SD:1,0 MD:2,1
50206
50207 The modifier is now logically down on the MD but not on keyboard 1 or
50208 keyboard 2.
50209
50210 XKB is layered in before the DIX, it increases/decreases the modifier key
50211 count accordingly. In the above example, during (2), the MD gets the key
50212 release and thus clears the modifier bit. (3) doesn't forward the release to
50213 the MD because it is already cleared. The copy of modifierKeysDown when the
50214 lastSlave changes however increases the counter for the held key. On (4),
50215 the press and release are both forwarded to the MD, causing a offset by 1
50216 and thus do not clear the logical modifier state.
50217
50218 X.Org Bug 25480 <http://bugs.freedesktop.org/show_bug.cgi?id=25480>
50219
50220 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50221 Acked-by: Daniel Stone <daniel@fooishbar.org>
50222
50223commit 7805e45284264b4cd286eece02e85426896e9f7b
50224Author: Peter Hutterer <peter.hutterer@who-t.net>
50225Date: Thu Jun 10 14:31:48 2010 +1000
50226
50227 config: remove redundant DBUS API define, require dbus-1 >= 1.0
50228
50229 It's still being pulled in by the HAL CFLAGS but the requirement to define
50230 this was dropped from DBus pre 1.0 (November 2006).
50231 This means we require dbus 1.0 now.
50232
50233 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50234 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
50235 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
50236
50237commit 66b21b2f455a1dfbc92f7caa571dcff3f3765808
50238Author: Dan Nicholson <dbn.lists@gmail.com>
50239Date: Mon Jun 7 20:39:58 2010 -0700
50240
50241 xfree86: Match devices based on current driver setting
50242
50243 Often we want to apply a driver specific option to a set of devices and
50244 don't care how the driver was selected for that device. The MatchDriver
50245 entry can be used to match the current driver string:
50246
50247 MatchDriver "evdev|mouse"
50248 Option "Emulate3Buttons" "yes"
50249
50250 The driver string is a case sensitive match.
50251
50252 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50253 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50254 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50255
50256commit a71bdff47d4cc80da6ceeb548db1dcc8e8b59702
50257Author: Dan Nicholson <dbn.lists@gmail.com>
50258Date: Mon Jun 7 20:39:57 2010 -0700
50259
50260 xfree86: Allow multiple InputClass Match* entries for && matching
50261
50262 Currently when there multiple InputClass entries of the same type, only
50263 the last entry is used and the previous ones are ignored. Instead,
50264 multiple entries are used to create multiple matching conditions.
50265
50266 For instance, an InputClass with
50267
50268 MatchProduct "foo"
50269 MatchProduct "bar"
50270
50271 will require that the device's product name contain both foo and bar.
50272 This provides a complement to the || style matching when an entry is
50273 split using the "|" token.
50274
50275 The xorg.conf man page has added an example to hopefully clarify the two
50276 types of compound matches.
50277
50278 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50279 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50280 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50281
50282commit 842625928632ae767b0689fcdb5103c1779a4c91
50283Author: Dan Nicholson <dbn.lists@gmail.com>
50284Date: Mon Jun 7 20:39:56 2010 -0700
50285
50286 config: Script to convert HAL fdi settings to InputClass sections
50287
50288 In the new world of udev and InputClass, x11_* settings from HAL fdi
50289 files will not be honored. This script converts those settings into
50290 valid InputClass sections that can be dropped into xorg.conf.d.
50291
50292 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50293 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50294
50295commit 87a1507da7e7788232d74285ef377b67b70e0fa4
50296Author: Dan Nicholson <dbn.lists@gmail.com>
50297Date: Thu Jun 10 06:15:41 2010 -0700
50298
50299 xfree86: Match devices based on USB ID
50300
50301 Sometimes the vendor and product names aren't specific enough to target
50302 a USB device, so expose the numeric codes in the ID. A MatchUSBID entry
50303 has been added that supports shell pattern matching when fnmatch(3) is
50304 available. For example:
50305
50306 MatchUSBID "046d:*"
50307
50308 The IDs are stored in lowercase hex separated by a ':' like "lsusb" or
50309 "lspci -n".
50310
50311 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50312 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50313 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50314
50315commit 645679c1523eee7028f3244cee57936b93326a2a
50316Author: Dan Nicholson <dbn.lists@gmail.com>
50317Date: Thu Jun 10 06:11:10 2010 -0700
50318
50319 xfree86: Match devices based on PnP ID
50320
50321 Serial input devices lack properties such as product or vendor name. This
50322 makes matching InputClass sections difficult. Add a MatchPnPID entry to
50323 test against the PnP ID of the device. The entry supports a shell pattern
50324 match on platforms that support fnmatch(3). For example:
50325
50326 MatchPnPID "WACf*"
50327
50328 A match type for non-path pattern matching, match_pattern, has been added.
50329 The difference between this and match_path_pattern is the FNM_PATHNAME
50330 flag in fnmatch(3).
50331
50332 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50333 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50334 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50335
50336commit 84190d20954ef7888e4d96832c2a4b3225b4dfa2
50337Merge: 4172aa1 5e1ef1f
50338Author: Keith Packard <keithp@keithp.com>
50339Date: Thu Jun 10 13:36:35 2010 -0700
50340
50341 Merge remote branch 'jeremyhu/master'
50342
50343commit 4172aa137c1b9b6f2a25c320d847af1f5ac56fba
50344Author: Keith Packard <keithp@keithp.com>
50345Date: Thu Jun 10 07:15:49 2010 -0700
50346
50347 dri2: Only deal with output windows and pixmaps.
50348
50349 This reverts commit fdb081b430ddffb495aa5b05bcc4cf10882ff4b2
50350 "dri2: Deal with input-only windows by using WindowDrawable()"
50351 and replaces it as follows:
50352
50353 Reject the creation of a DRI2 drawable for UNDRAWABLE_WINDOW (input-only
50354 windows) and DRAWABLE_BUFFER (whatever those are) drawables and only look up
50355 privates for the supported drawable types.
50356
50357 The rest of the the code can continue pretending there's only output windows
50358 and pixmaps, which are the only kinds of drawables relevant for DRI2.
50359
50360 Fixes server crash with GLX compositing managers such as compiz or kwin, due
50361 to looking up a window private for a pixmap and getting a bogus pointer.
50362
50363 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
50364 Reviewed-by: Keith Packard <keithp@keithp.com>
50365 Signed-off-by: Keith Packard <keithp@keithp.com>
50366
50367commit 353e32d3712f3a883a796ba562ec9fb5a8354837
50368Author: Jamey Sharp <jamey@minilop.net>
50369Date: Sun Jun 6 06:42:11 2010 -0700
50370
50371 glxdriswrast bugfix: configure swapgc, not gc.
50372
50373 I got this wrong in e2929db7b737413cf93fbebdf4d15abdfebff05c.
50374
50375 Signed-off-by: Jamey Sharp <jamey@minilop.net>
50376 Cc: Kristian Høgsberg <krh@bitplanet.net>
50377 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
50378 Signed-off-by: Keith Packard <keithp@keithp.com>
50379
50380commit b3548612c7943011f79a910f9a59bb975984d8a6
50381Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
50382Date: Sat Jun 5 13:09:28 2010 +0200
50383
50384 DRI2: Don't return junk reply instead of blocking in glXWaitForSbcOML()
50385
50386 DRI2WaitSBC() didn't block if requested targetSBC wasn't yet reached.
50387
50388 Instead it returned a xreply with uninitialized junk return values, then
50389 blocked the connection until targetSBC was reached.
50390
50391 Therefore the client didn't block, but continued with bogus return
50392 values from glXWaitForSbcOML.
50393
50394 This patch fixes the problem by implementing DRI2WaitSBC similar
50395 to the clean and proven DRI2WaitMSC implementation.
50396
50397 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
50398 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
50399 Signed-off-by: Keith Packard <keithp@keithp.com>
50400
50401commit 7287ef9e6cf953066e4a092cca9d0e4a279172bf
50402Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50403Date: Sun Jun 6 21:18:40 2010 +0700
50404
50405 Remove unnecessary parentheses around return values in functions
50406
50407 This patch was generated by the following Perl code:
50408
50409 perl -i -pe 's/([^_])return\s*\(\s*([^(]+?)\s*\)s*;(\s+(\n))?/$1return $2;$4/g;'
50410
50411 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50412 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50413 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
50414 Signed-off-by: Keith Packard <keithp@keithp.com>
50415
50416commit d1b4beecbc16448282dcc825dd5c354e96e48eca
50417Author: Dan Nicholson <dbn.lists@gmail.com>
50418Date: Mon Jun 7 20:39:53 2010 -0700
50419
50420 xfree86: Add MatchOS InputClass entry for operating system matching
50421
50422 Allow InputClass sections to match against the running operating system
50423 to narrow the application of rules. An example where this could be used
50424 is to specify that the default input driver on Linux is evdev while it's
50425 mouse/kbd everywhere else.
50426
50427 The operating system name is the same as `uname -s`, and matching is
50428 case-insensitive.
50429
50430 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50431 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50432 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50433
50434commit 9b30fa9f8fedb7ddb5672f93ed1a154d13578c47
50435Author: Dan Nicholson <dbn.lists@gmail.com>
50436Date: Mon Jun 7 20:39:52 2010 -0700
50437
50438 xfree86: Refactor InputClass matching code
50439
50440 InputClassMatches was starting to get a little hairy with all the loops
50441 over the tokenized match strings. This adds code, but makes it easier to
50442 read and add new matches.
50443
50444 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50445 Reviewed-by: Jamey Sharp <jamey at minilop.net>
50446 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50447 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50448
50449commit d09bc8f8e5bb92899f7c90eab6b0907920c7d643
50450Author: Dan Nicholson <dbn.lists@gmail.com>
50451Date: Mon Jun 7 20:39:51 2010 -0700
50452
50453 xfree86: Constify InputClass functions
50454
50455 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50456 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50457 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50458
50459commit 61c35d8ac4b6423caf2dcecb4ff25baaaaa14b1c
50460Author: Dan Nicholson <dbn.lists@gmail.com>
50461Date: Mon Jun 7 20:39:50 2010 -0700
50462
50463 config/hal: Missing newline in log message
50464
50465 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50466 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50467 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50468
50469commit 697c5b010d7f6df0ec9cae6352b125c4cbdbaad3
50470Author: Dan Nicholson <dbn.lists@gmail.com>
50471Date: Mon Jun 7 20:39:49 2010 -0700
50472
50473 config/udev: Add verbose info when fetching udev properties
50474
50475 Give the user a chance to see why their input devices are being ignored,
50476 even if they have to start the server with -logverbose.
50477
50478 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
50479 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50480 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50481
50482commit 94547b2d1448232ae316c228d65482cc0b646ac2
50483Author: Dan Nicholson <dbn.lists@gmail.com>
50484Date: Mon Jun 7 20:39:48 2010 -0700
50485
50486 config/hal: don't leak the input.tags property
50487
50488 Signed-off-by: Julien Cristau <jcristau@debian.org>
50489 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
50490 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50491 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
50492
50493commit 5e1ef1f4dde4615498c1c5e7a4e422c3b3432fe8
50494Author: Jeremy Huddleston <jeremyhu@apple.com>
50495Date: Tue Jun 8 13:01:10 2010 -0700
50496
50497 XQuartz: Add more sanity checking around string manipulation for xmodmap
50498
50499 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
50500
50501commit 0be81f9a163f2838448caee41b2dda274d73391a
50502Author: Jeremy Huddleston <jeremyhu@apple.com>
50503Date: Sun Jun 6 16:31:19 2010 -0700
50504
50505 XQuartz: Fix build regression from faeebead7bfcc78535757ca7acc1faf7554c03b7
50506
50507 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
50508
50509commit b3a7b229e1e1f212bdd185af5443311091824005
50510Author: Keith Packard <keithp@keithp.com>
50511Date: Mon Jun 7 21:49:43 2010 -0700
50512
50513 Make DRI_SCREEN_PRIV just return NULL when DRI isn't initialized
50514
50515 The code relied upon the previous devPrivate implementation which
50516 offered this convenience.
50517
50518 Signed-off-by: Keith Packard <keithp@keithp.com>
50519
50520commit f7af00e9f0e0e1d854b0e882378c032518ab71ca
50521Author: Ben Skeggs <bskeggs@redhat.com>
50522Date: Mon Jun 7 10:21:18 2010 +1000
50523
50524 randr: prevent an unnecessary screen resize with multiple displays
50525
50526 crtc->{x,y} is always 0 when xf86DefaultScreenLimits() is called, so we
50527 calculate too small an area for the initial framebuffer and force a resize
50528 to happen.
50529
50530 This commit fixes the code to use desired{X,Y} instead, which contains the
50531 initial output positions.
50532
50533 Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
50534 Reviewed-by: Dave Airlie <airlied@redhat.com>
50535 Reviewed-by: Keith Packard <keithp@keithp.com>
50536 Signed-off-by: Keith Packard <keithp@keithp.com>
50537
50538commit 8e97e5f9425639ad0a084150d0b232cad417595d
50539Author: Keith Packard <keithp@keithp.com>
50540Date: Mon Jun 7 14:23:47 2010 -0700
50541
50542 If XTest is always required, then eliminate the XTest devPrivate
50543
50544 The internals of XTest are used by Xi and Xkb, and both Xi and Xkb are
50545 always required, so it makes little sense to have XTest place data in
50546 a devPrivate, especially a devPrivate which is only available when the
50547 XTest extension is enabled.
50548
50549 Signed-off-by: Keith Packard <keithp@keithp.com>
50550 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
50551
50552commit fdb081b430ddffb495aa5b05bcc4cf10882ff4b2
50553Author: Keith Packard <keithp@keithp.com>
50554Date: Mon Jun 7 00:54:18 2010 -0700
50555
50556 dri2: Deal with input-only windows by using WindowDrawable()
50557
50558 Input only windows aren't DRAWABLE_WINDOW, but casting them to a
50559 PixmapPtr is a bit harsh, and unlikely to get the appropriate privates
50560 structure. use WindowDrawable instead which checks for both
50561 input-output and input-only windows.
50562
50563 Signed-off-by: Keith Packard <keithp@keithp.com>
50564 Tested-by: Eric Anholt <eric@anholt.net>
50565
50566commit bc26665661565918af484ccd17caad951010df60
50567Author: Keith Packard <keithp@keithp.com>
50568Date: Sun Jun 6 20:45:53 2010 -0700
50569
50570 Initialize private keys in test suite
50571
50572 Make sure all of the private keys used by the test code are
50573 initialized before being used.
50574
50575 Signed-off-by: Keith Packard <keithp@keithp.com>
50576 Tested-by: Robert Hooker <sarvatt@ubuntu.com>
50577
50578commit f03be727d647183a2176355ad0ac9a6735067be9
50579Author: Keith Packard <keithp@keithp.com>
50580Date: Sun Jun 6 16:04:42 2010 -0700
50581
50582 Place glyph privates at correct location within the allocate storage
50583
50584 A glyph allocation consists of :
50585
50586 GlyphRec
50587 numScreens * PicturePtr
50588 glyph privates
50589
50590 Tell the dix private bits to start past the picture pointers.
50591
50592 Signed-off-by: Keith Packard <keithp@keithp.com>
50593 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
50594
50595commit 67b824a81bb8254827023de351a4a094031f0e94
50596Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50597Date: Sun Jun 6 20:37:07 2010 +0700
50598
50599 randr: Make deletion of output properties more robust
50600
50601 Previously there was two branches of code with small discrepancies between them
50602 (especially prop->valid_values field was not free(3)ed). Extract the common
50603 routine and fix double-free prop->valid_values in RRDestroyOutputProperty by
50604 the way.
50605
50606 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50607 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50608
50609commit 0a4d8cbdcd7b25313fb29ccdb498044af95f7de6
50610Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50611Date: Sun Jun 6 15:28:30 2010 +0700
50612
50613 Remove more superfluous if(p) checks around free(p)
50614
50615 This patch has been generated by the following Coccinelle semantic patch:
50616
50617 @@
50618 expression E;
50619 @@
50620
50621 -if(E) { free(E); }
50622 +free(E);
50623
50624 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50625 Reviewed-by: Julien Cristau <jcristau@debian.org>
50626 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
50627 Reviewed-by: Matt Turner <mattst88@gmail.com>
50628
50629commit 5a0fc0ad21d562612676ef88ef2d533b2391810a
50630Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50631Date: Fri Jun 4 15:35:31 2010 +0700
50632
50633 Replace deprecated bzero with memset
50634
50635 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50636 Reviewed-by: Marcin Baczyński <marbacz@gmail.com>
50637 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
50638 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50639 Reviewed-by: Matt Turner <mattst88@gmail.com>
50640
50641commit f9810ba914877b379cb36f1b9755f7923ceca14c
50642Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50643Date: Fri Jun 4 14:26:31 2010 +0700
50644
50645 xkb: Remove superfluous if(x) x = realloc(x, sz); else x = malloc(sz); logic
50646
50647 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50648 Reviewed-by: Marcin Baczyński <marbacz@gmail.com>
50649 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50650 Reviewed-by: Matt Turner <mattst88@gmail.com>
50651 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
50652
50653commit 77ea20895ca1d163e14c6b767d8c287c94ba3114
50654Author: Mikhail Gusarov <dottedmag@dottedmag.net>
50655Date: Fri Jun 4 14:22:06 2010 +0700
50656
50657 xkb: Remove superfluous if(p) checks around free(p)
50658
50659 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50660 Reviewed-by: Julien Cristau <jcristau@debian.org>
50661 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50662 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
50663
50664commit 5d4e2c594059ffb536c8e506c2623320d3c6a787
50665Merge: eacc421 793dd39
50666Author: Keith Packard <keithp@keithp.com>
50667Date: Sat Jun 5 22:20:28 2010 -0700
50668
50669 Merge remote branch 'jamey/for-keith'
50670
50671commit 793dd399468ff3ba22ffa9c62fa54fc87261fbe7
50672Author: Jamey Sharp <jamey@minilop.net>
50673Date: Tue May 25 15:08:38 2010 -0700
50674
50675 dmx: Split glx scalar swap variables from array swap variables.
50676
50677 This makes hw/dmx/glxProxy/unpack.h more closely resemble glx/unpack.h,
50678 and fixes the "unused variable 'swapEnd'" and "unused variable 'swapPC'"
50679 warnings.
50680
50681 Signed-off-by: Jamey Sharp <jamey@minilop.net>
50682 Reviewed-by: Keith Packard <keithp@keithp.com>
50683
50684commit eacc42132ed0f1e8f4a4d65a4b6637ab06f7e2aa
50685Author: Kim Woelders <kim@woelders.dk>
50686Date: Fri Jun 4 20:20:47 2010 +0200
50687
50688 dix: Remove test for obsolete NO_XINERAMA_PORT macro.
50689
50690 Signed-off-by: Kim Woelders <kim@woelders.dk>
50691 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50692 Signed-off-by: Keith Packard <keithp@keithp.com>
50693
50694commit a0fe6987b55b5326157b50e169b8d6b9ae26a1ae
50695Author: Jamey Sharp <jamey@minilop.net>
50696Date: Sat Jun 5 20:55:29 2010 -0700
50697
50698 Clean up after removal of screen parameters from region macros.
50699
50700 Signed-off-by: Jamey Sharp <jamey@minilop.net>
50701 Reviewed-by: Keith Packard <keithp@keithp.com>
50702
50703commit 424b593c7872a703d3d0c942b7e8acc0770eb023
50704Author: Keith Packard <keithp@keithp.com>
50705Date: Sat Jun 5 21:34:51 2010 -0700
50706
50707 Damage pixmap private key was being mis-registered as a picture key
50708
50709 This would cause all kinds of fun, in particular Xnest would crash at startup.
50710
50711 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50712 Signed-off-by: Keith Packard <keithp@keithp.com>
50713
50714commit 9c121ae45ae1d2ebd892c45e1631716ef5e7d649
50715Author: Keith Packard <keithp@keithp.com>
50716Date: Sat Jun 5 21:34:27 2010 -0700
50717
50718 Register Xnest colormap screen private key
50719
50720 This key is used to store the currently installed colormap and must be
50721 registered before we can use it.
50722
50723 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50724 Signed-off-by: Keith Packard <keithp@keithp.com>
50725
50726commit df534d05208436e9ca8c9b42acff3fbe240e3ec4
50727Author: Keith Packard <keithp@keithp.com>
50728Date: Sat Jun 5 21:24:54 2010 -0700
50729
50730 Make GetPictureScreenIfSet check if Render is running
50731
50732 This macro originally checked to see if the Render screen private
50733 index had been allocated. When the privates were changed the first
50734 time, there wasn't any need to check as dixLookupPrivate would simply
50735 return NULL in that case. Now that we care, check to see if the key
50736 has been initialized before asking for the value.
50737
50738 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50739 Signed-off-by: Keith Packard <keithp@keithp.com>
50740
50741commit e7fc8b32e41e10c057d2787fcc377296be67f2e9
50742Author: Keith Packard <keithp@keithp.com>
50743Date: Sat Jun 5 21:08:01 2010 -0700
50744
50745 Move the shadow screen private key initialization to shadowSetup
50746
50747 Some users of the shadow code don't call shadowInit, just shadowSetup
50748 and so make sure the key is initialized there.
50749
50750 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50751 Signed-off-by: Keith Packard <keithp@keithp.com>
50752
50753commit 6bd5f0d75bca727c4686b20eee166c8cae472ba2
50754Author: Keith Packard <keithp@keithp.com>
50755Date: Sat Jun 5 21:07:15 2010 -0700
50756
50757 Fix exa_priv.h declarations of privates
50758
50759 exa_priv.h exposes the privates to the rest of the exa code, and yet
50760 the declarations of the privates wasn't fixed when the exa.c versions were.
50761
50762 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50763 Signed-off-by: Keith Packard <keithp@keithp.com>
50764
50765commit 6b306f43384e5c2143197e746a5a39c4ebb2583c
50766Author: Keith Packard <keithp@keithp.com>
50767Date: Sat Jun 5 20:58:04 2010 -0700
50768
50769 kdrive: Xv code uses shared screen private instead of kdrive-specific private
50770
50771 When doing the devPrivate API conversion, I missed this as there was
50772 also a local private key which wasn't ever used.
50773
50774 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50775 Signed-off-by: Keith Packard <keithp@keithp.com>
50776
50777commit ab07e2b8ededaa2193fc199a8c09623d84032280
50778Author: Keith Packard <keithp@keithp.com>
50779Date: Fri Apr 30 19:38:48 2010 -0700
50780
50781 Allocate per-screen device/cursor-bits private keys in midispcur
50782
50783 midispcur was abusing the CursorScreenKey to index the cursor_bits
50784 privates, it also had a MAXSCREENS array of keys to index device
50785 privates. Switch both of these to the new dixCreatePrivateKey API and
50786 store a pointer to that in the screen private.
50787
50788 Signed-off-by: Keith Packard <keithp@keithp.com>
50789 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50790
50791commit 34db537907c6cb2635dbefdce7dcfcae90f7c902
50792Author: Keith Packard <keithp@keithp.com>
50793Date: Fri Apr 30 19:36:33 2010 -0700
50794
50795 Add dixCreatePrivateKey API
50796
50797 Keys need to persist through server reset so that the private system
50798 can be cleaned up in dixResetPrivates. In particular, this means that
50799 keys cannot live in objects freed at reset time. This API provides
50800 suitable object lifetime by having the privates code free the key in
50801 the reset path.
50802
50803 Signed-off-by: Keith Packard <keithp@keithp.com>
50804 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50805
50806commit 495fc3eb2d6c98bde82ae1278f89fcf131fd9bf8
50807Author: Keith Packard <keithp@keithp.com>
50808Date: Fri Apr 30 19:38:38 2010 -0700
50809
50810 Change devPrivates implementation.
50811
50812 Each key now declares which object type it belongs to, this permits
50813 keys for different types to share the same offset within the allocated
50814 privates. As a special case for XSELinux, a key may be allocated as
50815 PRIVATE_XSELINUX which will allow it to continue to be used across the
50816 relevant object types.
50817
50818 Signed-off-by: Keith Packard <keithp@keithp.com>
50819 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50820
50821commit faeebead7bfcc78535757ca7acc1faf7554c03b7
50822Author: Keith Packard <keithp@keithp.com>
50823Date: Mon Apr 26 17:22:21 2010 -0700
50824
50825 Change the devPrivates API to require dixRegisterPrivateKey
50826
50827 This patch only changes the API, not the implementation of the
50828 devPrivates infrastructure. This will permit a new devPrivates
50829 implementation to be layed into the server without requiring
50830 simultaneous changes in every devPrivates user.
50831
50832 Signed-off-by: Keith Packard <keithp@keithp.com>
50833 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com>
50834
50835commit c865a24401f06bcf1347d8b41f736a066ab25693
50836Author: Keith Packard <keithp@keithp.com>
50837Date: Fri Apr 30 22:05:27 2010 -0700
50838
50839 Create separate private key for midispcur cursor bits
50840
50841 The CursorScreenKey array is supposed to be used by the DDX for cursor
50842 private data, but midispcur was abusing it to hold cursor bits private
50843 information. Create a separate set of privates for the dispcur cursor
50844 bits information.
50845
50846 This also renames the device private index and macros to better
50847 reflect their usage:
50848
50849 miDCSpriteKey -> miDCDeviceKey
50850 MIDCBUFFER -> miGetDCDevice
50851
50852 Signed-off-by: Keith Packard <keithp@keithp.com>
50853 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50854
50855commit 431781a921251d54782f0a4f194bbef1fabd1380
50856Author: Keith Packard <keithp@keithp.com>
50857Date: Sat May 15 14:52:39 2010 -0700
50858
50859 Remove dixRegisterPrivateOffset; hard-code devPrivates offsets instead
50860
50861 For predefined resource types, the offset of the devPrivates field was
50862 already kept in a constant table. The only non-predefined type needing
50863 this treatment was dbeDrawableResType, which is just a magic alias for
50864 RT_PIXMAP.
50865
50866 This patch special-cases looking up RC_DRAWABLE offsets and uses the
50867 table directly for everything else.
50868
50869 Signed-off-by: Keith Packard <keithp@keithp.com>
50870 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50871
50872commit 7ef612de784daaed09ba13f4615c10714614033f
50873Author: Keith Packard <keithp@keithp.com>
50874Date: Fri May 21 12:35:29 2010 -0700
50875
50876 Add REGION_ macros for source compatibility with existing drivers.
50877
50878 This makes the region code changes source compatible with existing
50879 code, although none of them are used within the server source itself.
50880
50881 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50882 Signed-off-by: Keith Packard <keithp@keithp.com>
50883
50884commit 965cade6ee7c226f850e806cb2e910793d1536f9
50885Author: Keith Packard <keithp@keithp.com>
50886Date: Fri May 21 12:18:22 2010 -0700
50887
50888 Change region implementation from macros to inline functions.
50889
50890 This makes all of the previous macros into inline functions and also
50891 turns all of the direct calls to pixman region code into inline
50892 functions as well.
50893
50894 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50895 Signed-off-by: Keith Packard <keithp@keithp.com>
50896
50897commit 2dc138922b7588515d5f2447e4b9dcdc0bef15e0
50898Author: Keith Packard <keithp@keithp.com>
50899Date: Fri May 21 15:05:48 2010 -0700
50900
50901 Rename region macros to eliminate screen argument
50902
50903 This is a combination of a huge mechanical patch and a few small
50904 fixups required to finish the job. They were reviewed separately, but
50905 because the server does not build without both pieces, I've merged
50906 them together at this time.
50907
50908 The mechanical changes were performed by running the included
50909 'fix-region' script over the whole tree:
50910
50911 $ git ls-files | grep -v '^fix-' | xargs ./fix-region
50912
50913 And then, the white space errors in the resulting patch were fixed
50914 using the provided fix-patch-whitespace script.
50915
50916 $ sh ./fix-patch-whitespace
50917
50918 Thanks to Jamey Sharp for the mighty fine sed-generating sed script.
50919
50920 The hand-done changes involve removing functions from dix/region.c
50921 that duplicate inline functions in include/regionstr.h, along with
50922 their declarations in regionstr.h, mi.h and mispans.h.
50923
50924 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50925 Signed-off-by: Keith Packard <keithp@keithp.com>
50926
50927commit d17e726e89ef644310de77b960b715c2d11088da
50928Author: Keith Packard <keithp@keithp.com>
50929Date: Sat May 22 00:09:47 2010 -0700
50930
50931 Move mi/miregion.c to dix/region.c
50932
50933 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50934 Signed-off-by: Keith Packard <keithp@keithp.com>
50935
50936commit 9b6f5f549dfe4c20e3cb3c2ef7095ab47abaca3b
50937Author: Keith Packard <keithp@keithp.com>
50938Date: Fri May 21 11:14:59 2010 -0700
50939
50940 Change region implementation names to eliminate the 'mi' prefix
50941
50942 This prepares the file to be moved from mi to dix. This patch
50943 was done mechanically with the included scripts 'fix-miregion' run over
50944 the entire X server and 'fix-miregion-private' run over
50945 include/regionstr.h and mi/miregion.c.
50946
50947 Reviewed-by: Jamey Sharp <jamey@minilop.net>
50948 Signed-off-by: Keith Packard <keithp@keithp.com>
50949
50950commit 1768e51976ae8949d999309bd2384ccad72deecd
50951Author: Jeremy Huddleston <jeremyhu@apple.com>
50952Date: Sat Jun 5 15:58:21 2010 -0700
50953
50954 rootless: Fix SetShape regression
50955
50956 This fixes a regression in miext/rootless from 643cb6e87c10ab554c03ada81930001a8ebcc909
50957
50958 Found-by: tinderbox
50959 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
50960 Reviewed-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
50961 Signed-off-by: Keith Packard <keithp@keithp.com>
50962
50963commit 6b4af3b7925978cd79f717761f1b6f33bd8dfbaf
50964Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
50965Date: Fri Jun 4 16:27:28 2010 +0300
50966
50967 configure: Check for libsha1.pc
50968
50969 xserver fails to detect libsha1. Problem is that configure checks for
50970 sha1.pc when libsha1 provides libsha1.pc.
50971
50972 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
50973 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
50974 Signed-off-by: Keith Packard <keithp@keithp.com>
50975
50976commit d163266692235261e1e0cfa6b900f54e6ec63de5
50977Author: Alan Coopersmith <alan.coopersmith@oracle.com>
50978Date: Thu Jun 3 18:59:04 2010 -0700
50979
50980 Stop searching for XF86Config files
50981
50982 xorg.conf has been used since the X11R6.7 release in April 2004.
50983 6 years has been a generous transition period for users to
50984 "mv XF86Config xorg.conf" and for distros to update their
50985 configuration tools and packages.
50986
50987 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
50988 Reviewed-by: James Cloos <cloos@jhcloos.com>
50989 Signed-off-by: Keith Packard <keithp@keithp.com>
50990
50991commit 01c75522b6ed15e979347acd38e234e2cb9006c3
50992Author: Alan Coopersmith <alan.coopersmith@oracle.com>
50993Date: Thu Jun 3 18:59:03 2010 -0700
50994
50995 Stop searching for XF86Config files
50996
50997 xorg.conf has been used since the X11R6.7 release in April 2004.
50998 6 years has been a generous transition period for users to
50999 "mv XF86Config xorg.conf" and for distros to update their
51000 configuration tools and packages.
51001
51002 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51003 Reviewed-by: James Cloos <cloos@jhcloos.com>
51004 Signed-off-by: Keith Packard <keithp@keithp.com>
51005
51006commit 7f19a7a6e90a4fd7b7ec0256974f62e575218541
51007Author: Peter Hutterer <peter.hutterer@who-t.net>
51008Date: Fri Jun 4 14:15:04 2010 +1000
51009
51010 xkb: fix invalid memory writes in _XkbCopyGeom.
51011
51012 Classic strlen/strcpy mistake of
51013 foo = malloc(strlen(bar));
51014 strcpy(foo, bar);
51015
51016 Testcase: valgrind Xephyr :1
51017
51018 ==8591== Invalid write of size 1
51019 ==8591== at 0x4A0638F: strcpy (mc_replace_strmem.c:311)
51020 ==8591== by 0x605593: _XkbCopyGeom (xkbUtils.c:1994)
51021 ==8591== by 0x605973: XkbCopyKeymap (xkbUtils.c:2118)
51022 ==8591== by 0x6122B3: InitKeyboardDeviceStruct (xkbInit.c:560)
51023 ==8591== by 0x4472E2: CoreKeyboardProc (devices.c:577)
51024 ==8591== by 0x447162: ActivateDevice (devices.c:530)
51025 ==8591== by 0x4475D6: InitCoreDevices (devices.c:672)
51026 ==8591== by 0x4449EE: main (main.c:254)
51027 ==8591== Address 0x6f96505 is 0 bytes after a block of size 53 alloc'd
51028 ==8591== at 0x4A0515D: malloc (vg_replace_malloc.c:195)
51029 ==8591== by 0x6054B7: _XkbCopyGeom (xkbUtils.c:1980)
51030 ==8591== by 0x605973: XkbCopyKeymap (xkbUtils.c:2118)
51031 ==8591== by 0x6122B3: InitKeyboardDeviceStruct (xkbInit.c:560)
51032 ==8591== by 0x4472E2: CoreKeyboardProc (devices.c:577)
51033 ==8591== by 0x447162: ActivateDevice (devices.c:530)
51034 ==8591== by 0x4475D6: InitCoreDevices (devices.c:672)
51035 ==8591== by 0x4449EE: main (main.c:254)
51036
51037 Reported-by: Dave Airlie <airlied@redhat.com>
51038 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51039 Reviewed-by-and-apologised-for: Daniel Stone <daniel@fooishbar.org>
51040 Signed-off-by: Keith Packard <keithp@keithp.com>
51041
51042commit 959a1eaf1c15a691141f1b0dc53757fe9b6e9b13
51043Author: Dave Airlie <airlied@redhat.com>
51044Date: Fri Jun 4 11:09:46 2010 +1000
51045
51046 composite: use config notify hook to do pixmap resize.
51047
51048 Since reallocating the backing pixmap can fail, we need to try and do
51049 it before any other side effects of reconfiguring the window happen.
51050
51051 This changes the ConfigNotify hook to return status, and moves the
51052 composite window reconfiguration wrappers to ConfigNotify. They all
51053 basically did the same thing, so we can drop the MoveWindow,
51054 ResizeWindow, ChangeBorderWidth wrappers, and allow ConfigNotify to do
51055 all the work. If reallocation fails we fail before we send any
51056 confiureNotify events, or enter the area we can't recover from.
51057
51058 The only place we now enforce 32k limits are in EXA/UXA/fb, so drivers
51059 that don't use this should probably deal with it in their pixmap
51060 allocate if they don't already.
51061
51062 This also breaks ABI, so we need an alternate fix for older servers,
51063 working on the X server makes me realise why I'm a kernel hacker.
51064
51065 Signed-off-by: Dave Airlie <airlied@redhat.com>
51066 Reviewed-by: Keith Packard <keithp@keithp.com>
51067 Signed-off-by: Keith Packard <keithp@keithp.com>
51068
51069commit 0f12e86e600522768f5f64eafc1230526e700ab7
51070Merge: aa7c09f f114f54
51071Author: Keith Packard <keithp@keithp.com>
51072Date: Thu Jun 3 15:09:32 2010 -0700
51073
51074 Merge remote branch 'jamey/for-keith'
51075
51076commit f114f54986aca1add4c8808a05d9692e289547e7
51077Author: Jamey Sharp <jamey@minilop.net>
51078Date: Fri Apr 23 19:33:49 2010 -0700
51079
51080 Accumulate graphics exposures incrementally in PanoramiXCopyArea/Plane.
51081
51082 This fuses two loops in each function, eliminating an intermediate
51083 MAXSCREENS-sized array from each.
51084
51085 Aside from being more efficient, I believe this is equivalent to the
51086 previous implementation, since
51087 - each per-screen GC has the graphicsExposures flag set the same way,
51088 and
51089 - the REGION_* macros ignore their screen argument.
51090
51091 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51092 Reviewed-by: Keith Packard <keithp@keithp.com>
51093 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51094 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
51095
51096commit a7c7ebe4b3a0df1a4d1ffe3c690cfb21470f103d
51097Author: Jamey Sharp <jamey@minilop.net>
51098Date: Sat May 22 21:13:48 2010 -0700
51099
51100 Delete XineramaScreenRegions cache.
51101
51102 Every screen region consists of a single rectangle, so initializing a
51103 stack-allocated region for each screen on-demand does no heap allocation
51104 and is fast.
51105
51106 This eliminates a MAXSCREENS-sized array.
51107
51108 The REGION_UNINIT calls are no-ops since no boxes are actually allocated
51109 for a single-rectangle region, but it seemed wiser to include them.
51110
51111 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51112 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51113 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
51114
51115commit a0456da339fa9b3c3e6fead060338ee7348d6460
51116Author: Jamey Sharp <jamey@minilop.net>
51117Date: Sun May 23 12:57:52 2010 -0700
51118
51119 XineramaSetCursorPosition: use screen bounds directly, not POINT_IN_REGION.
51120
51121 This hides a MAXSCREENS-sized array as an implementation detail of
51122 panoramiX.c rather than an exported global.
51123
51124 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51125 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51126 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
51127
51128commit a83cff9f4d622b069c96a68b4e87a669bf1f6446
51129Author: Jamey Sharp <jamey@minilop.net>
51130Date: Sun May 23 10:11:47 2010 -0700
51131
51132 Move each screen's x/y origin into ScreenRec.
51133
51134 Many references to the dixScreenOrigins array already had the
51135 corresponding screen pointer handy, which meant they usually looked like
51136 "dixScreenOrigins[pScreen->myNum]". Adding a field to ScreenRec instead
51137 of keeping this information in a parallel array simplifies those
51138 expressions, and eliminates a MAXSCREENS-sized array.
51139
51140 Since dix declared the dixScreenOrigins array, I figure allocating a
51141 screen private for these values is overkill.
51142
51143 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51144 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51145 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
51146
51147commit 217ccaa5a341018457f468a774c035c0df47d918
51148Author: Jamey Sharp <jamey@minilop.net>
51149Date: Sun May 23 11:05:01 2010 -0700
51150
51151 Delete panoramiXdataPtr: it's redundant.
51152
51153 This eliminates a dynamically-allocated MAXSCREENS-sized array.
51154
51155 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51156 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51157 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
51158
51159commit e7fae9ecc42ab5e73b89117722dbf4117d928f9a
51160Author: Jamey Sharp <jamey@minilop.net>
51161Date: Sat May 22 00:26:28 2010 -0700
51162
51163 Move each screen's root-window pointer into ScreenRec.
51164
51165 Many references to the WindowTable array already had the corresponding
51166 screen pointer handy, which meant they usually looked like
51167 "WindowTable[pScreen->myNum]". Adding a field to ScreenRec instead of
51168 keeping this information in a parallel array simplifies those
51169 expressions, and eliminates a MAXSCREENS-sized array.
51170
51171 Since dix uses this data, a screen private entry isn't appropriate.
51172
51173 xf86-video-dummy currently uses WindowTable, so it needs to be updated
51174 to reflect this change.
51175
51176 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51177 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51178 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
51179
51180commit 80b5d3a3264d2c5167e5ac85a3b04af0f89cece1
51181Author: Jamey Sharp <jamey@minilop.net>
51182Date: Fri May 21 23:13:51 2010 -0700
51183
51184 Move each screen's screensaver data into ScreenRec.
51185
51186 Most references to the savedScreenInfo array already had the
51187 corresponding screen pointer handy, which meant they usually looked like
51188 "savedScreenInfo[pScreen->myNum]". Adding a field to ScreenRec instead
51189 of keeping this information in a parallel array simplifies those
51190 expressions, and eliminates a MAXSCREENS-sized array.
51191
51192 Since dix uses this data, a screen private entry isn't appropriate.
51193
51194 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51195 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51196 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
51197
51198commit 5030540d6cd2aa9bbd4ecb7f4487616f9ef0140a
51199Author: Jamey Sharp <jamey@minilop.net>
51200Date: Thu May 27 11:50:52 2010 -0700
51201
51202 midispcur: Add comment explaining why pRootPicture must not be freed.
51203
51204 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51205 Cc: Peter Hutterer <peter.hutterer@who-t.net>
51206 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
51207
51208commit f308cbea8c9fa5c5c6705bc301178b167de450d5
51209Author: Jamey Sharp <jamey@minilop.net>
51210Date: Mon May 24 18:41:54 2010 -0700
51211
51212 mi: Delete unused flicker-free MoveCursor code.
51213
51214 It's been commented-out for three and a half years and nobody seems to
51215 be missing it enough to resurrect it.
51216
51217 Besides deleting code that is untested and therefore buggy, this saves a
51218 little memory for each pointer device on each screen.
51219
51220 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51221 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51222
51223commit d1bad1b59f24ab4dcf2d1ce1c5d4f42b5fa40bc1
51224Author: Jamey Sharp <jamey@minilop.net>
51225Date: Tue May 25 14:25:07 2010 -0700
51226
51227 dmx: Fix XSetExtensionErrorHandler calls to agree on constness.
51228
51229 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51230 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51231
51232commit aa7c09f7d04d59852b850b2b6993f4cbdc7ec940
51233Author: Alan Coopersmith <alan.coopersmith@oracle.com>
51234Date: Wed Jun 2 14:31:56 2010 -0700
51235
51236 Install fbdevhw module in normal module directory instead of linux/
51237
51238 Since we no longer support OS-independent custom elfloader modules,
51239 we don't need to put the OS-dependent modules into os-specific subdirs
51240 any more.
51241
51242 We do however still need to install the stubs version of this module
51243 on non-Linux platforms, since a number of drivers link to functions
51244 in it, even when built on non-Linux platforms.
51245
51246 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51247 Reviewed-by: Michel Dänzer <michel@daenzer.net>
51248 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51249 Signed-off-by: Keith Packard <keithp@keithp.com>
51250
51251commit b13d7a8e09c8e3eb5003420c3c0484431e5a052b
51252Merge: f0ab726 dcceb90
51253Author: Keith Packard <keithp@keithp.com>
51254Date: Thu Jun 3 07:01:26 2010 -0700
51255
51256 Merge remote branch 'vignatti/for-keith'
51257
51258commit dcceb90b1dccbbf5793f6ba1ed857013b71dafbc
51259Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51260Date: Wed May 19 11:45:46 2010 +0200
51261
51262 xext: delete wrong source reference from fontcache
51263
51264 This should go away in 0b45ba48.
51265
51266 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51267 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
51268
51269commit c043de3261b6e14f2446b2cd9d62dde5860d4720
51270Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51271Date: Mon May 3 12:37:25 2010 +0200
51272
51273 configure: sha1: check libsha1 using pkg-config instead
51274
51275 Previously the code was using AC_CHECK_LIB, guaranteeing whether the library
51276 is correct by tracking sha1_begin function. This paranoic checking is not
51277 necessary given there's only one libsha1 in the market, which surely contains
51278 such function.
51279
51280 Moreover, this patch now improves a bit the sha1 implementation checking
51281 behavior using pkg-config to find the right flags that needs to link against.
51282
51283 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51284 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
51285
51286commit f0ab726d8966cab4e50154c216d577db79328d77
51287Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
51288Date: Wed Apr 21 11:07:47 2010 +0300
51289
51290 DRI2: Use single error path in initialization
51291
51292 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
51293 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
51294
51295commit 6eef70dc56bcc1e3047e4e488bcd6ae62c8ffac6
51296Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51297Date: Fri Jan 22 18:34:36 2010 +0200
51298
51299 DRI2: Allow building without libdrm
51300
51301 Some drivers use DRI protocol but implement their own kernel rendering
51302 manager. For these drivers, libdrm becomes useless. --disable-libdrm
51303 configure parameter can be used to disable libdrm support in dri2.
51304
51305 To provide ABI/API compatibility for libdrm based drivers, libdrm call
51306 is wrapped in ifdef.
51307
51308 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51309 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
51310 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
51311
51312commit cdcb575664d3d60b662c542e782de83a047165c9
51313Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51314Date: Fri Jan 22 17:58:17 2010 +0200
51315
51316 DRI2: add AuthMagic hook for driver side support
51317
51318 With this new hook, drmAuthMagic becomes useless and should be deprecated.
51319 You might want to implement AuthMagic on driver side instead.
51320
51321 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51322 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
51323 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
51324
51325commit 643cb6e87c10ab554c03ada81930001a8ebcc909
51326Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
51327Date: Thu May 27 09:11:50 2010 -0700
51328
51329 Only deal with input code when changing the input shape.
51330
51331 Propagate the shape kind all the way to SetShape to avoid performing non-input
51332 operations such as revalidating the tree and generating exposures when only
51333 changing a window's input shape.
51334
51335 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
51336 Acked-by: Aaron Plattner<aplattner@nvidia.com>
51337 Reviewed-by: Daniel Stone<daniel@fooishbar.org>
51338 Signed-off-by: Keith Packard <keithp@keithp.com>
51339
51340commit d90f2cd98a97e6534792d3867b3fba70d850b706
51341Author: Chris Wilson <chris@chris-wilson.co.uk>
51342Date: Thu May 27 13:26:23 2010 +0100
51343
51344 xfree86: Unbreak autoconfig following 0abf065e38c4
51345
51346 The move of the PCI device id probing into a separate file neglected to
51347 return the number of found devices, and so the PCI devices were being
51348 overwritten by the default entries for vesa and fbdev.
51349
51350 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
51351 Cc: Tiago Vignatti <tiago.vignatti@nokia.com>
51352 Cc: Alex Deucher <alexdeucher@gmail.com>
51353 Reviewed-by: Julien Cristau <jcristau@debian.org>
51354 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51355 Signed-off-by: Keith Packard <keithp@keithp.com>
51356
51357commit 1304b8b27cb12c803c4f51f04cb6f9d508b82c69
51358Author: Aaron Plattner <aplattner@nvidia.com>
51359Date: Fri May 28 09:35:54 2010 -0700
51360
51361 Fix pixmap validation in miDbePositionWindow.
51362
51363 miDbePositionWindow allocates two pixmaps: a front buffer, and a back buffer.
51364 If the buffers are supposed to be initialized, it validates a GC against the
51365 front buffer, then uses it to fill and/or copy both the front buffer *and* the
51366 back buffer, without revalidating. If the acceleration architecture needs
51367 different GC funcs for the two pixmaps -- for example if allocation of the front
51368 buffer exhausted video memory -- then this can cause crashes because the GC is
51369 not validated for the back buffer pixmap.
51370
51371 Fix this by performing the rendering for the front buffer first, then
51372 revalidating against the back buffer before performing the back buffer
51373 rendering.
51374
51375 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
51376 Reviewed-by: Jamey Sharp <jamey@minilop.net>
51377 Signed-off-by: Keith Packard <keithp@keithp.com>
51378
51379commit 91a6359caf24d94343ff76f43ea7b7fc3223203d
51380Author: Dave Airlie <airlied@redhat.com>
51381Date: Thu Jun 3 12:16:40 2010 +1000
51382
51383 composite: initialise pOldPixmap to NullPixmap at alloc time.
51384
51385 We just never initialised the malloced value.
51386
51387 Signed-off-by: Dave Airlie <airlied@redhat.com>
51388 Reviewed-by: Keith Packard <keithp@keithp.com>
51389 Signed-off-by: Keith Packard <keithp@keithp.com>
51390
51391commit 968a79dcf5e17ac3963953ef56b8f94dbd75323b
51392Author: Nicolas George <nicolas.george@normalesup.org>
51393Date: Wed Jun 2 13:40:51 2010 +0200
51394
51395 Change keyboard controls on slave keyboards (#27926)
51396
51397 Makes the use of IsMaster in ProcChangeKeyboardControl consistent with other
51398 similar loops.
51399
51400 Signed-off-by: Nicolas George <nicolas.george@normalesup.org>
51401 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51402 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51403
51404commit 108b766c31b57fb1955d34d85673a235c7f743aa
51405Author: Peter Hutterer <peter.hutterer@who-t.net>
51406Date: Thu May 20 11:04:53 2010 +1000
51407
51408 xfree86: initialize InputAttributes to NULL in the autoconfig code.
51409
51410 Reported-by: Jamey Sharp <jamey@minilop.net>
51411 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51412 Reviewed-by: Jamey Sharp <jamey@minilop.net>
51413
51414commit 151659f9636088fd70bc5586de97bf43ee706180
51415Author: Peter Hutterer <peter.hutterer@who-t.net>
51416Date: Mon May 24 16:24:03 2010 +1000
51417
51418 test: fix up InputAttributes helper function test.
51419
51420 Just some extra clarification as pointed out by Dan Nicholson,
51421 and that memcpy should have been a memcmp.
51422
51423 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51424 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
51425
51426commit 6cccf0131c8464d8838cae2200730873d7dd9e45
51427Author: Peter Korsgaard <peter.korsgaard@barco.com>
51428Date: Tue May 25 11:03:28 2010 +0200
51429
51430 dix: add 3x3 transformation matrix xinput property for multi-head handling
51431
51432 For absolute input devices (E.G. touchscreens) in multi-head setups,
51433 we need a way to bind the device to an randr output. This adds the
51434 infrastructure to the server to allow us to do so.
51435
51436 positionSprite() scales input coordinates to the dimensions of the shared
51437 (total) screen frame buffer, so to restrict motion to an output we need to
51438 scale/rotate/translate device coordinates to a subset of the frame buffer
51439 before passing them on to positionSprite.
51440
51441 This is done here using a 3x3 transformation matrix, which is applied to
51442 the device coordinates using homogeneous coordinates, E.G.:
51443
51444 [ c0 c1 c2 ] [ x ]
51445 [ c3 c4 c5 ] * [ y ]
51446 [ c6 c7 c8 ] [ 1 ]
51447
51448 Notice: As input devices have varying input ranges, the coordinates are
51449 first scaled to the [0..1] range for generality, and afterwards scaled
51450 back up.
51451
51452 E.G. for a dual head setup (using same resolution) next to each other, you
51453 would want to scale the X coordinates of the touchscreen connected to the
51454 both heads by 50%, and translate (offset) the coordinates of the rightmost
51455 head by 50%, or in matrix form:
51456
51457 left: right:
51458 [ 0.5 0 0 ] [ 0.5 0 0.5 ]
51459 [ 0 1 0 ] [ 0 1 0 ]
51460 [ 0 0 1 ] [ 0 0 0 ]
51461
51462 Which can be done using xinput:
51463
51464 xinput set-prop <left> --type=float "Coordinate Transformation Matrix" \
51465 0.5 0 0 0 1 0 0 0 1
51466
51467 xinput set-prop <right> --type=float "Coordinate Transformation Matrix" \
51468 0.5 0 0.5 0 1 0 0 0 1
51469
51470 Likewise more complication setups involving more heads, rotation or
51471 different resolution can be handled.
51472
51473 Signed-off-by: Peter Korsgaard <peter.korsgaard@barco.com>
51474 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51475
51476commit e4582d9e5c8649347742a13eae68cf27005296fc
51477Author: Peter Hutterer <peter.hutterer@who-t.net>
51478Date: Thu May 27 10:42:54 2010 -0400
51479
51480 xfree86: bump ABI_XINPUT_VERSION to 11.
51481
51482 commit c2d0b3b437b7ce6ce975f2ead4d8bb8295ef0ddc
51483 "xfree86: store the InputAttributes in the input device."
51484 introduced the new API. Bump the input version so drivers can handle this
51485 appropriately.
51486
51487 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51488 Signed-off-by: Keith Packard <keithp@keithp.com>
51489
51490commit cd0ef0b6a2f9c6bfeda98684569da922792d3feb
51491Author: Jamey Sharp <jamey@minilop.net>
51492Date: Tue May 25 14:30:22 2010 -0700
51493
51494 dmx: Delete unused ChangeKeyboardDevice/ChangePointerDevice.
51495
51496 Daniel Stone deleted the API for these in 2006, in commit
51497 96e32805d12fc36f0fa0926dbfb0dd8a5cadb739.
51498
51499 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51500 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51501
51502commit 70454864917f534483fa049e067b5646df61c88a
51503Author: Jamey Sharp <jamey@minilop.net>
51504Date: Fri May 14 12:31:44 2010 -0700
51505
51506 Ignore build products in the server's test suite.
51507
51508 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51509 Cc: Peter Hutterer <peter.hutterer@who-t.net>
51510 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51511
51512commit 8bb88379d95356384e233225bfda6abfd71389b4
51513Author: Jamey Sharp <jamey@minilop.net>
51514Date: Tue May 25 15:52:18 2010 -0700
51515
51516 dmx: Remove dead __glXDispSwap_DrawArraysEXT definition.
51517
51518 Eric Anholt made the corresponding fix in glx/render2swap.c in commit
51519 49d38ab2328c409b2a98465b52677af057121513.
51520
51521 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51522 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51523
51524commit c97a3ade5a8718b063292c73a31aac2196fe2452
51525Author: Jamey Sharp <jamey@minilop.net>
51526Date: Tue May 25 15:17:30 2010 -0700
51527
51528 dmx: Delete unused local variables.
51529
51530 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51531 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51532
51533commit 3c3d099fc10c63d746f90c08043d45180d6a7226
51534Author: Jamey Sharp <jamey@minilop.net>
51535Date: Tue May 25 13:44:32 2010 -0700
51536
51537 dmx: Xfree -> free
51538
51539 Mikhail fixed the corresponding Xallocs, but missed these uses of Xfree
51540 in commit 3f3ff971ecff9936cebafc813af9193b97bba89c.
51541
51542 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51543 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51544
51545commit 01ca6400d949a873c0bb5de7e0dbf138e8a7949b
51546Author: Jamey Sharp <jamey@minilop.net>
51547Date: Tue May 25 14:02:34 2010 -0700
51548
51549 Xephyr: Fix Xcalloc deprecation warnings.
51550
51551 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51552 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51553
51554commit 34df659687ad3936ab03d19ce8fdb21f385c2e8a
51555Author: Jamey Sharp <jamey@minilop.net>
51556Date: Tue May 25 11:51:32 2010 -0700
51557
51558 xfree86/int10/helper_mem: Fix log message.
51559
51560 Three years ago in commit f62beb6f3609e8b6e61325ac89017590811bbd07 ajax
51561 deleted the code that could have set this format string to anything
51562 else, so just use the format string literal. This makes GCC happy since
51563 it can check the argument types, which, by the way, weren't correct
51564 since this format string doesn't need any arguments.
51565
51566 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51567 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51568 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51569
51570commit 0a98d9e6cec7c611a3c56f97d4ddc0c546975c55
51571Author: Jamey Sharp <jamey@minilop.net>
51572Date: Tue May 25 10:30:58 2010 -0700
51573
51574 xf86bigfont: Silence compiler warning by initializing pDesc.
51575
51576 Access to pDesc was always guarded by (nCharInfos > 0), so the code
51577 wasn't actually buggy, but this makes it clear that it's correct.
51578
51579 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51580 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51581
51582commit 4e9d3e413245e7c01c2ff3c0f130b8967754131b
51583Author: Keith Packard <keithp@keithp.com>
51584Date: Wed May 26 07:54:35 2010 -0700
51585
51586 Revert "Add a "flags" field to DeleteInputDeviceRequest."
51587
51588 Peter wants to get a larger patch sequence put together and I didn't
51589 read past the commit message to see the 'don't take this patch
51590 please'.
51591
51592 This reverts commit 531ff40301975519af7b20109c17d296312d3f2b.
51593
51594commit 531ff40301975519af7b20109c17d296312d3f2b
51595Author: Peter Hutterer <peter.hutterer@who-t.net>
51596Date: Thu May 20 18:07:12 2010 +1000
51597
51598 Add a "flags" field to DeleteInputDeviceRequest.
51599
51600 Some input drivers need to implement an internal hotplugging scheme for
51601 dependent devices to provide multiple X devices off one kernel device file.
51602 Such dependent devices can be added with NewInputDeviceRequest() but they are
51603 not removed when the config backend calls DeleteInputDeviceRequest(),
51604 leaving the original device to clean up.
51605
51606 Example of the wacom driver:
51607
51608 config/udev calls NewInputDeviceRequest("stylus")
51609
51610 wacom PreInit calls
51611 NewInputDeviceRequest("eraser")
51612 NewInputDeviceRequest("pad")
51613 NewInputDeviceRequest("cursor")
51614 PreInit finishes.
51615
51616 When the device is removed, the config backend only calls
51617 DeleteInputDeviceRequest for "stylus". The driver needs to call
51618 DeleteInputDeviceRequest for the dependent devices eraser, pad and cursor to
51619 clean up properly.
51620 However, when the server terminates, DeleteInputDeviceRequest is called for
51621 all devices - the driver must not remove the dependent devices to avoid
51622 double-frees. There is no method for the driver to detect why a device is
51623 being removed, leading to elaborate guesswork and some amount of wishful
51624 thinking.
51625
51626 Though the input driver's UnInit already supports flags, they are unused.
51627 This patch uses the flags to supply information where the
51628 DeleteInputDeviceRequest request originates from, allowing a driver to
51629 selectively call DeleteInputDeviceRequest when necessary.
51630
51631 Also bumps XINPUT ABI.
51632
51633 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51634 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
51635 Signed-off-by: Keith Packard <keithp@keithp.com>
51636
51637commit c2d0b3b437b7ce6ce975f2ead4d8bb8295ef0ddc
51638Author: Peter Hutterer <peter.hutterer@who-t.net>
51639Date: Tue May 25 17:15:32 2010 +1000
51640
51641 xfree86: store the InputAttributes in the input device.
51642
51643 InputAttributes largely decide which configuration values get merged from
51644 the xorg.conf.d snippets. While they are available in the config backend,
51645 they are not available for any other callers of NewInputDeviceRequest().
51646
51647 Drivers implementing driver-side hotplugging do not have access to these
51648 attributes and cannot have xorg.conf.d snippets specific to dependent
51649 devices. For example, the following case cannot work right now:
51650
51651 Section "InputClass"
51652 MatchProduct "Wacom"
51653 Option "PressCurve" "0 0 100 100"
51654 ...
51655 EndSection
51656
51657 Section "InputClass"
51658 MatchProduct "Wacom"
51659 MatchProduct "eraser"
51660 Option "PressCurve" "10 10 50 50"
51661 ...
51662 EndSection
51663
51664 The second section is not triggered, as the wacom driver cannot supply the
51665 InputAttributes to NewInputDeviceRequest().
51666
51667 Add the attributes to the IDevRec and merge them into the InputInfoRec to
51668 make them accessible in the driver. This changes the ABI for input drivers.
51669
51670 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51671 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
51672 Signed-off-by: Keith Packard <keithp@keithp.com>
51673
51674commit 86303a338ad563d0b986a3c052104301c310c4ac
51675Author: Peter Hutterer <peter.hutterer@who-t.net>
51676Date: Tue May 25 17:12:34 2010 +1000
51677
51678 dix: add helper functions to duplicate and free InputAttributes.
51679
51680 No special memory handling is used to give drivers the maximum flexibility
51681 with the data. Drivers should be able to call realloc on the product string
51682 if needed and perform similar operations.
51683
51684 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51685 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
51686 Signed-off-by: Keith Packard <keithp@keithp.com>
51687
51688commit bf78e11839f8278020b604672ff7c3d194232be9
51689Author: Jamey Sharp <jamey@minilop.net>
51690Date: Mon May 24 09:15:32 2010 -0700
51691
51692 midispcur: Construct Picture objects lazily in case Render is disabled.
51693
51694 Reverts part of the effects of 518f3b189b6c8aa28b62837d14309fd06163ccbb,
51695 "mi: don't thrash resources when displaying the software cursor across
51696 screens". The per-screen cache is preserved, and the GCs are still
51697 allocated eagerly, but now it doesn't construct pRootPicture until
51698 somebody attempts to draw an ARGB cursor.
51699
51700 I noticed crashes in Xnest, which doesn't support the RENDER extension,
51701 but I suspect other DDXes that support disabling that extension would
51702 have had issues as well.
51703
51704 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51705 Reviewed-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
51706
51707commit bf9b64eefbadbf299d3b841c1890bfcaef5aab1b
51708Author: Jamey Sharp <jamey@minilop.net>
51709Date: Mon May 24 09:14:27 2010 -0700
51710
51711 xnest: Don't ignore errors from DeviceCursorInitialize.
51712
51713 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51714 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
51715
51716commit b11465a6ecdc2b8373e6fc8af427edc4602bcaa1
51717Merge: 7c085ae 0abf065
51718Author: Keith Packard <keithp@keithp.com>
51719Date: Mon May 24 22:18:31 2010 -0700
51720
51721 Merge remote branch 'vignatti/bus-cleanup-take2'
51722
51723commit 7c085aebfedeb621a6fbeb3f09f4fcc640452044
51724Author: Jamey Sharp <jamey@minilop.net>
51725Date: Mon May 24 11:45:42 2010 -0700
51726
51727 xfixes: Remove the invisible cursor sprite, using NullCursor instead.
51728
51729 Oliver McFadden reports that the invisible cursor sprite caused damage
51730 events and thus unnecessary redrawing, so removing it improves
51731 performance when using software cursor sprites, especially on those
51732 devices where you do not want a visible cursor: touchscreen tablets,
51733 embedded devices, etc.
51734
51735 For the xfree86 DDX, if hardware cursors are used, the driver is
51736 required to provide a HideCursor function, which will be called instead
51737 of trying to set a null cursor. I think software cursors are already
51738 safe. The other DDXes also look safe.
51739
51740 As far as I can tell, there's no reason to realize a null cursor. I
51741 think everything that handles null cursors doesn't rely on any setup in
51742 RealizeCursor, and treats them as empty cursors.
51743
51744 Xnest assumes that if a cursor is created, it will be realized before it
51745 is freed, which didn't happen if the invisible cursor was never used in
51746 a server generation. So this fixes a segfault in Xnest as well.
51747
51748 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51749 Cc: Oliver McFadden <oliver.mcfadden@nokia.com>
51750 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
51751 Tested-by: Oliver McFadden <oliver.mcfadden@nokia.com>
51752 Signed-off-by: Keith Packard <keithp@keithp.com>
51753
51754commit 0abf065e38c4c48917054eb3e39e12bd20e6e8ec
51755Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51756Date: Thu May 20 19:24:47 2010 +0300
51757
51758 xfree86: move all pci code from auto configurator into a meaningful location
51759
51760 This patch creates the private xf86PciMatchDriver hook, which goes inside pci
51761 code to match the drivers found in the system.
51762
51763 Now there's no direct references to PCI inside xf86AutoConfig.c anymore.
51764
51765 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51766 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51767
51768commit 0ceac6f64f5ad9bc2ac4b19be2dd245ffba78b05
51769Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51770Date: Thu May 20 18:23:24 2010 +0300
51771
51772 xfree86: no need to check for the configuration case when matching devices
51773
51774 xf86MatchDevice will never be called in configuration time.
51775
51776 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51777 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51778
51779commit b0eef1101815b08da2d375af7bec2be3136cf713
51780Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51781Date: Thu May 20 18:19:14 2010 +0300
51782
51783 xfree86: organize and group all pci related stuff inside xf86.h
51784
51785 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51786 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51787
51788commit 47df5a489ea69a68e753367423bfbe8830521f4e
51789Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51790Date: Thu May 20 17:39:59 2010 +0300
51791
51792 xfree86: remove BUS_ISA type given we don't support anymore
51793
51794 Should go together within commit df14682a.
51795
51796 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51797 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51798
51799commit 9d000a5509e0a356ee33ad08e4dc967a2ef8cbe0
51800Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51801Date: Thu May 20 16:52:37 2010 +0300
51802
51803 xfree86: remove all kind of bus and PCI dependency from the common helper file
51804
51805 Move all PCI procedures from xf86Helper.c to a more meaningful place (namely
51806 xf86pciBus.c). xf86Helper.c is free of PCI code now.
51807
51808 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51809 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51810
51811commit 610009cf39984f2f5e818a221b626f7af10bd90e
51812Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51813Date: Tue May 18 20:54:28 2010 +0300
51814
51815 xfree86: bus: remove useless field from EntityRec
51816
51817 RAC is the champion of remaining trash for sure!
51818
51819 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51820 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51821 Reviewed-by: Keith Packard <keithp@keithp.com>
51822
51823commit 95889a569feabd3f803afb92950d067bd76e7397
51824Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51825Date: Tue May 18 20:48:35 2010 +0300
51826
51827 xfree86: bus: delete useless xf86FindPrimaryDevice
51828
51829 This function had a wrong name and was just logging the primary device. No one
51830 cares about it honestly.
51831
51832 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51833 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51834
51835commit f6865909485d7d282c8131a2ae306143ad51ba2b
51836Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51837Date: Tue May 18 19:39:31 2010 +0300
51838
51839 xfree86: bus: remove unused headers
51840
51841 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51842 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51843
51844commit 111b1fff1ca69724a8350c4375089c312329c378
51845Author: Tiago Vignatti <tiago.vignatti@nokia.com>
51846Date: Tue May 18 19:35:10 2010 +0300
51847
51848 xfree86: bus: remove unused pci macros
51849
51850 Should be gone in commits 3c03d9f1 and a9d7d659a respectively.
51851
51852 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
51853 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51854
51855commit 2ffffb4daf6161e6a22d81442ecf6209acc9e975
51856Merge: b5e0f6d d530608
51857Author: Keith Packard <keithp@keithp.com>
51858Date: Sun May 23 23:22:08 2010 -0700
51859
51860 Merge remote branch 'alanc/docs'
51861
51862commit b5e0f6d8f45c5b24eb50b305c66fa80c783ef488
51863Author: Peter Hutterer <peter.hutterer@who-t.net>
51864Date: Mon May 24 12:25:15 2010 +1000
51865
51866 xfree86: fix multiple InputAttributes tag matching.
51867
51868 attr->tags is an array of strings (null-terminated). When matching, match
51869 against each string instead of each [i,end] substring in the first tag.
51870
51871 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
51872 Reviewed-by: Jamey Sharp <jamey@minilop.net>
51873 Signed-off-by: Keith Packard <keithp@keithp.com>
51874
51875commit d5306084b57583c670c56ce9e7d3c78cca7aa07b
51876Author: Brice Goglin <bgoglin@debian.org>
51877Date: Sat May 22 20:29:02 2010 +0200
51878
51879 Remove obsolete reference to README.DRI in xorg.conf.man
51880
51881 Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org>
51882 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51883
51884commit 89dd7d2b6fe18f32c2f9336c3d8cd9f35490e5ae
51885Author: Alan Coopersmith <alan.coopersmith@oracle.com>
51886Date: Sat May 22 10:33:04 2010 -0700
51887
51888 Remove ancient documentation of IBM RapidAccess keyboard hack
51889
51890 Keyboard setup belongs in drivers, not in a document no one sees
51891 because we don't even install it.
51892
51893 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51894 Reviewed-by: Julien Cristau <jcristau@debian.org>
51895 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
51896
51897commit 480fcdf0335abae0ee544b022c7985211eee462f
51898Author: Alan Coopersmith <alan.coopersmith@oracle.com>
51899Date: Sat May 22 10:29:47 2010 -0700
51900
51901 Remove completely out-of-date README.DRI
51902
51903 The license only allows distribution of verbatim copies, so we can't
51904 update it, even to correct the incorrect address to send updates to.
51905
51906 The Mesa & DRI web pages are much better sources of current information
51907 anyway.
51908
51909 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51910 Reviewed-by: Julien Cristau <jcristau@debian.org>
51911
51912commit 1805c74d9b0ed46802481bcd6beea5feeb60212c
51913Author: Alan Coopersmith <alan.coopersmith@oracle.com>
51914Date: Sat May 22 10:29:00 2010 -0700
51915
51916 Add RandR 1.2 README.modes doc to EXTRA_DIST
51917
51918 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51919 Reviewed-by: Julien Cristau <jcristau@debian.org>
51920
51921commit 5939e39a641773a36c22104e1184143678dca7a2
51922Author: Dave Airlie <airlied@redhat.com>
51923Date: Mon May 24 09:31:51 2010 +1000
51924
51925 xf86: allow for no outputs connected at startup operation.
51926
51927 When nothing is connected at startup and we canGrow, allow the server to start with a 1024x768 framebuffer, and when the drivers send hotplug events this will expand to the correct size dynamically.
51928
51929 Reviewed-by: Keith Packard <keithp@keithp.com>
51930 Signed-off-by: Dave Airlie <airlied@redhat.com>
51931 Signed-off-by: Keith Packard <keithp@keithp.com>
51932
51933commit b9f48d60bc0c839bd323c582231e8e7e2b810af6
51934Author: Jamey Sharp <jamey@minilop.net>
51935Date: Wed May 19 10:44:33 2010 -0700
51936
51937 Device init: Don't crash when CreateGC fails.
51938
51939 ActivateDevice was ignoring errors from DeviceCursorInitialize, so
51940 cursor-related calls failed later. Jeremy Huddleston saw that crash in
51941 miPointerConstrainCursor, while with Xvfb I saw it in
51942 miSpriteRealizeCursor.
51943
51944 miDCDeviceCleanup frees any non-NULL GCs. miDCDeviceInitialize calls
51945 Cleanup on any failure, but if it failed early then some of the pointers
51946 in the miDCBufferPtr were garbage. Switch from malloc to calloc to
51947 ensure everything's initialized safely first.
51948
51949 With these two fixes, if CreateGC fails then the server gracefully fails
51950 in FatalError instead of segfaulting.
51951
51952 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51953 Cc: Peter Hutterer <peter.hutterer@who-t.net>
51954 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
51955 Signed-off-by: Keith Packard <keithp@keithp.com>
51956
51957commit c51534961e1bb4c68ff881758862d2f85f572ce7
51958Author: Alan Coopersmith <alan.coopersmith@oracle.com>
51959Date: Sat May 22 11:11:48 2010 -0700
51960
51961 xorg-server.pc Requires: xproto >= 7.0.17
51962
51963 Since the headers we export in the SDK use the _X_NORETURN #define
51964 introduced in xproto 7.0.17, drivers built with the SDK will also
51965 need to have at least that version of xproto installed.
51966
51967 I considered exporting the version from configure.ac, but decided not
51968 to since the minimum required to build the X server may not always
51969 be the same as the minimum required to build drivers (for instance,
51970 if the X server used a new #define in its .c files or internal headers,
51971 but didn't export it in any of the SDK headers).
51972
51973 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
51974 Reviewed-by: Julien Cristau <jcristau@debian.org>
51975 Signed-off-by: Keith Packard <keithp@keithp.com>
51976
51977commit e90f20eed394de2200aba42bd61fd31c1067ac33
51978Author: Jamey Sharp <jamey@minilop.net>
51979Date: Fri May 21 22:16:50 2010 -0700
51980
51981 Declare functions that unconditionally call FatalError as _X_NORETURN.
51982
51983 For AtomError, this should fix a clang warning; in the other cases it's
51984 just good documentation.
51985
51986 Signed-off-by: Jamey Sharp <jamey@minilop.net>
51987 Cc: Jeremy Huddleston <jeremyhu@apple.com>
51988 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
51989 Signed-off-by: Keith Packard <keithp@keithp.com>
51990
51991commit 9d31d5b639c2a6ab8b1a80a0e2f0ee2c4f8cddd3
51992Author: Adam Jackson <ajax@redhat.com>
51993Date: Mon May 10 11:48:31 2010 -0400
51994
51995 vfb: Remove dead variable and header file
51996
51997 Signed-off-by: Adam Jackson <ajax@redhat.com>
51998 Reviewed-by: Jamey Sharp <jamey@minilop.net>
51999 Signed-off-by: Keith Packard <keithp@keithp.com>
52000
52001commit d8cbcbc01d1355d55c1c0af64f7f7f61629f6c61
52002Author: Jamey Sharp <jamey@minilop.net>
52003Date: Fri May 21 11:12:26 2010 -0700
52004
52005 doPolyText: forget about FontChange's XID after looking up pFont.
52006
52007 As of e2929db7b737413cf93fbebdf4d15abdfebff05c, doPolyText uses pFont
52008 consistently rather than looking it up again from the saved XID.
52009
52010 clang noticed that "oldfid = fid" could run when fid hadn't been
52011 initialized yet.
52012
52013 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52014 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
52015 Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
52016 Signed-off-by: Keith Packard <keithp@keithp.com>
52017
52018commit 989db930d739483759087b13b8d9a043299feafb
52019Author: Aaron Zang <Aaron.Zang@Sun.COM>
52020Date: Thu May 20 17:56:28 2010 -0700
52021
52022 Solaris: Use VT_SET_CONSUSER ioctl to set Console User rights profile
52023
52024 When Xorg is started on display :0, this ioctl is called to grant the
52025 user the rights traditionally associated with /dev/console (before VT
52026 support was added), such as access to local peripheral devices.
52027
52028 Also adds a Solaris-specific -C flag to force starting on /dev/console
52029 instead of /dev/vt*, allowing programs like xterm -C to access the
52030 console device.
52031
52032 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52033 Reviewed-by: Adam Jackson <ajax@redhat.com>
52034 Signed-off-by: Keith Packard <keithp@keithp.com>
52035
52036commit 98553e52a14b97a03aef8dc4fc0300b3f4c2c4b5
52037Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52038Date: Thu May 20 17:56:27 2010 -0700
52039
52040 vbe.h: Use __attribute__((packed)) on Sun cc 5.9 & later as well as gcc
52041
52042 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52043 Reviewed-by: Adam Jackson <ajax@redhat.com>
52044 Signed-off-by: Keith Packard <keithp@keithp.com>
52045
52046commit 7b09335a46f9428141811230c69eef7968531359
52047Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52048Date: Thu May 20 17:56:26 2010 -0700
52049
52050 Xserver(1) man page updates
52051
52052 - Note that -br is now default.
52053 - Move -bs after -br for alphabetical ordering.
52054 - Remove -config option that's been hidden in "ignore" section,
52055 since ajax removed the -config code a couple years back.
52056 - Add -nocursor option.
52057 - Add xinput & xrandr to list of runtime server control programs
52058 - Replace XDarwin with Xquartz in list of Xservers
52059
52060 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52061 Reviewed-by: Adam Jackson <ajax@redhat.com>
52062 Signed-off-by: Keith Packard <keithp@keithp.com>
52063
52064commit cf4f3d051858aadedd6e333bb317a1daa3987ad7
52065Author: Peter Hutterer <peter.hutterer@who-t.net>
52066Date: Fri May 21 13:30:25 2010 +1000
52067
52068 dix: remove obsolete comment.from EnableDevice.
52069
52070 The code this comment was referring to was removed in
52071 8b5086250aa5dae8de8b763408ff480d7beac819 "Eliminate bogus event resizing."
52072
52073 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
52074 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
52075 Signed-off-by: Keith Packard <keithp@keithp.com>
52076
52077commit 5754e66044571d549c295b7c9e02ce3348dbe3c7
52078Author: Jamey Sharp <jamey@minilop.net>
52079Date: Thu May 20 10:46:43 2010 -0700
52080
52081 Replace screen->rgf scratch GC flags with a bit in each GC.
52082
52083 This eliminates a poorly-named, poorly-documented field from the
52084 ScreenRec, using a previously-unused flag bit in each GC instead.
52085
52086 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52087 Cc: Keith Packard <keithp@keithp.com>
52088 Reviewed-by: Keith Packard <keithp@keithp.com>
52089 Signed-off-by: Keith Packard <keithp@keithp.com>
52090
52091commit 5a7275d78a2f1c20ed5bb7b228cf370c4ada22c9
52092Author: Keith Packard <keithp@keithp.com>
52093Date: Thu May 20 13:57:02 2010 -0700
52094
52095 Find windows GL wrapper files in $(srcdir) instead of .
52096
52097 This fixes 'make distcheck' for me.
52098
52099 Signed-off-by: Keith Packard <keithp@keithp.com>
52100 Reviewed-by: Julien Cristau <jcristau@debian.org>
52101
52102commit 1197a87b5638037cc48084fba4eec9d3b599e900
52103Author: Keith Packard <keithp@keithp.com>
52104Date: Thu May 20 13:40:24 2010 -0700
52105
52106 winpriv.h was moved from hw/xwin to hw/xwin/glx in November
52107
52108 hw/xwin/Makefile.am was referencing this instead of hw/xwin/glx/Makefile.am
52109
52110 Signed-off-by: Keith Packard <keithp@keithp.com>
52111 Reviewed-by: Julien Cristau <jcristau@debian.org>
52112
52113commit 1cad520f3f1d9d520422e3e4d5d8f4ea85a212a4
52114Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52115Date: Mon May 17 20:35:10 2010 -0700
52116
52117 XFree86 Design doc: Convert LinuxDoc ``quotes'' to DocBook <quote> tags
52118
52119 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52120 Tested-by: Gaetan Nadon <memsize@videotron.ca>
52121
52122commit 8367913be3305be4f03d095615d3885107094427
52123Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52124Date: Fri May 14 22:05:50 2010 -0700
52125
52126 XFree86 Design doc: Explain this version covers the current Xorg release
52127
52128 Update the title & preface to explain that while this was originally
52129 the XFree86 4.0 design, we've changed a lot since forking.
52130
52131 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52132 Tested-by: Gaetan Nadon <memsize@videotron.ca>
52133
52134commit 78fe4be4c43db1f07a30e8a6c9a17f81c1082d76
52135Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52136Date: Mon May 17 20:27:10 2010 -0700
52137
52138 Show Xserver release/version date in DIX & DDX docs
52139
52140 Uses a fake absolute path to the entity definition files so that
52141 the xmlto --searchpath will work for finding the actual path
52142
52143 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52144 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
52145 Tested-by: Gaetan Nadon <memsize@videotron.ca>
52146
52147commit 2bdc73bcf41ff0c10e70ecc15fc8b8333af6db65
52148Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52149Date: Fri May 14 15:33:46 2010 -0700
52150
52151 Move Xserver-spec.xml from xorg-docs to doc/xml
52152
52153 Allows keeping it in sync with the sources it documents, and to be
52154 released with them
52155
52156 Requires the previous patch to convert the Xserver tree from
52157 LinuxDoc to DocBook for SGML/XML documents.
52158
52159 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52160 Reviewed-by: Rémi Cardona <remi@gentoo.org>
52161 Tested-by: Gaetan Nadon <memsize@videotron.ca>
52162
52163commit fc6ebe1e1d3057378f61f992549a98e67a04dc6c
52164Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52165Date: Fri May 14 14:56:09 2010 -0700
52166
52167 Convert LinuxDoc documents to DocBook/XML
52168
52169 Only the markup/formatting is changed - the contents should still
52170 be wildly out of date for now.
52171
52172 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52173 Reviewed-by: Rémi Cardona <remi@gentoo.org>
52174 Tested-by: Gaetan Nadon <memsize@videotron.ca>
52175
52176commit ebd745ced89b2a2d9f6b4dcbd9f5f7e7f3b35451
52177Author: Simon Farnsworth <simon.farnsworth@onelan.com>
52178Date: Thu May 20 09:58:38 2010 +0100
52179
52180 Add documentation for the new DefaultModes option
52181
52182 This patch adds documentation for the DefaultModes monitor option
52183 added in the previous patch.
52184
52185 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.com>
52186 Reviewed-by: Adam Jackson <ajax@redhat.com>
52187 Signed-off-by: Keith Packard <keithp@keithp.com>
52188
52189commit cfc629148891be317ba60f2495abcda485780191
52190Author: Simon Farnsworth <simon.farnsworth@onelan.com>
52191Date: Thu May 20 09:58:37 2010 +0100
52192
52193 Add configuration option to disable default modes on an output
52194
52195 Setting 'Option "DefaultModes" "No"' in an output will guarantee that
52196 X will omit the default modes, just picking up modes from the
52197 configuration file and modes provided by the output.
52198
52199 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.com>
52200 Reviewed-by: Adam Jackson <ajax@redhat.com>
52201 Signed-off-by: Keith Packard <keithp@keithp.com>
52202
52203commit a1c2bdb6ab7a4f374844f80e83620b155991e817
52204Author: Michel Dänzer <daenzer@vmware.com>
52205Date: Thu May 20 10:13:06 2010 +0200
52206
52207 EXA: Wrap Glyphs even without Composite acceleration.
52208
52209 In order to avoid migration ping-pong when accumulating glyphs in a mask
52210 picture.
52211
52212 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
52213 Reviewed-by: Adam Jackson <ajax@redhat.com>
52214 Signed-off-by: Keith Packard <keithp@keithp.com>
52215
52216commit 00581d4a42e2a0e7652834973645a4a9ed59cf93
52217Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52218Date: Fri May 14 19:31:56 2010 +0300
52219
52220 configure: force new xproto version dependency
52221
52222 commit bca85e2e127a8a23e3a2debcfeb3ae07cd3c66ac introduced it.
52223
52224 Use the version that includes _X_NORETURN
52225
52226 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52227 Reviewed-by: Julien Cristau <jcristau@debian.org>
52228 Signed-off-by: Keith Packard <keithp@keithp.com>
52229
52230commit 103507af0c0ce7d7482a67163249864af36a2374
52231Merge: 8bd8d81 99fcf65
52232Author: Keith Packard <keithp@keithp.com>
52233Date: Wed May 19 22:27:53 2010 -0700
52234
52235 Merge remote branch 'vignatti/bus-cleanup'
52236
52237commit 8bd8d81dc473bf72ea108b1896a55e32defa02e2
52238Merge: c620a1c 66d5ecc
52239Author: Keith Packard <keithp@keithp.com>
52240Date: Wed May 19 22:27:20 2010 -0700
52241
52242 Merge remote branch 'vignatti/animcursor-state-fix'
52243
52244commit c620a1c0a4ca564e778cce019dd30046c6969956
52245Author: Peter Hutterer <peter.hutterer@who-t.net>
52246Date: Thu May 20 10:26:56 2010 +1000
52247
52248 xfree86: fix typo in optionTypeToSting.
52249
52250 Because we don't want anyone to get hurt.
52251
52252 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
52253 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
52254 Reviewed-by: Matt Turner <mattst88@gmail.com>
52255 Signed-off-by: Keith Packard <keithp@keithp.com>
52256
52257commit 8b6c1809c09f832051327d86e1a25dc0ec5cc878
52258Merge: d88ba77 c38552d
52259Author: Keith Packard <keithp@keithp.com>
52260Date: Wed May 19 12:58:02 2010 -0700
52261
52262 Merge remote branch 'jamey/for-keith'
52263
52264commit c38552d115e3bc71ad6179a8ad0d68778e943793
52265Author: Jamey Sharp <jamey@minilop.net>
52266Date: Sat Apr 24 23:56:36 2010 -0700
52267
52268 Add typed resource-lookup errors for non-core resource types.
52269
52270 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52271 Reviewed-by: Adam Jackson <ajax@redhat.com>
52272
52273commit e291c561821ae86b7dd74269d5cd29bc31703962
52274Author: Jamey Sharp <jamey@minilop.net>
52275Date: Sat Apr 24 23:26:40 2010 -0700
52276
52277 Return an appropriately-typed error from dixLookupResourceByType.
52278
52279 Rather than always returning BadValue, associate an error status like
52280 BadWindow with a resource type like RT_WINDOW, and return the
52281 appropriate one for the requested type.
52282
52283 This patch only touches the core protocol resource types. Others still
52284 return BadValue and need to be mapped appropriately.
52285
52286 dixLookupResourceByType can now return BadImplementation, if the caller
52287 asked for a resource type that has not been allocated in the server.
52288
52289 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52290 Reviewed-by: Adam Jackson <ajax@redhat.com>
52291
52292commit 90e612dcbe370da095d317fac62c80ac2447fa0b
52293Author: Jamey Sharp <jamey@minilop.net>
52294Date: Sat May 15 15:51:32 2010 -0700
52295
52296 Use WriteEventsToClient rather than TryClientEvents where possible.
52297
52298 If filter is NoEventMask (aka CantBeFiltered), grab is null, and the
52299 first event is not in the set of "critical events", then TryClientEvents
52300 simply calls WriteEventsToClient. In that case, it returns 0 for fake or
52301 dead clients, and 1 otherwise. Inline for this special case.
52302
52303 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52304 Reviewed-by: Julien Cristau <jcristau@debian.org>
52305 Reviewed-by: Adam Jackson <ajax@redhat.com>
52306
52307commit 8033fb6c9792820a82fbdff6a14ff8a7a141ba74
52308Author: Jamey Sharp <jamey@minilop.net>
52309Date: Sat May 15 13:12:44 2010 -0700
52310
52311 Set event sequence number in WriteEventsToClient instead of at callers.
52312
52313 TryClientEvents already did this; this commit just moves the assignment
52314 one level down so that no event source has to worry about sequence
52315 numbers.
52316
52317 ...No event source, that is, except XKB, which inexplicably calls
52318 WriteToClient directly for several events.
52319
52320 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52321 Reviewed-by: Julien Cristau <jcristau@debian.org>
52322 Reviewed-by: Adam Jackson <ajax@redhat.com>
52323
52324commit 4b9600a4167427a8fe88bca9b738c9a99cac9469
52325Author: Jamey Sharp <jamey@minilop.net>
52326Date: Sat May 15 12:31:34 2010 -0700
52327
52328 Make WriteEventsToClient/WriteToClient no-op on fake or dead clients.
52329
52330 This matches the test in TryClientEvents, and is a superset of tests
52331 done by the callers of these functions. The consequence of forgetting
52332 these tests is a server crash, so they're always desirable. In my
52333 opinion, it's better to not require the callers to remember to do these
52334 checks.
52335
52336 For callers that don't do very much work before calling WriteToClient or
52337 WriteEventsToClient, I've removed the redundant checks.
52338
52339 hw/xquartz/xpr/appledri.c has an interesting case: While its check for
52340 "client == NULL" appears redundant with the test in WriteEventsToClient,
52341 it dereferences client to get the sequence number.
52342
52343 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=27497
52344 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52345 Cc: Chris Wilson <chris@chris-wilson.co.uk>
52346 Reviewed-by: Julien Cristau <jcristau@debian.org>
52347 Reviewed-by: Adam Jackson <ajax@redhat.com>
52348
52349commit ee9cd2df4a5af66c6b7073563785b8aef61fcdb8
52350Author: Jamey Sharp <jamey@minilop.net>
52351Date: Sat May 15 13:37:34 2010 -0700
52352
52353 xv: TryClientEvents already checks client and sets sequenceNumber.
52354
52355 So don't bother doing those steps again.
52356
52357 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52358 Reviewed-by: Julien Cristau <jcristau@debian.org>
52359 Reviewed-by: Adam Jackson <ajax@redhat.com>
52360
52361commit e1e7deefb59cfdd82c9de46a7ca241d42436b2f8
52362Author: Jamey Sharp <jamey@minilop.net>
52363Date: Tue May 18 16:49:21 2010 -0700
52364
52365 mi/misprite.c: Delete unused private lookups.
52366
52367 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52368 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
52369
52370commit 936b09f2a3e31afe41050c00736bbb6b812c7003
52371Author: Jamey Sharp <jamey@minilop.net>
52372Date: Tue May 18 16:48:08 2010 -0700
52373
52374 Replace miSpriteCursorFuncRec with direct calls to midispcur.c.
52375
52376 Nobody wraps the mi software-cursor sprite rendering implementations.
52377
52378 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52379 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
52380
52381commit 66d5ecc5fd36e21715531a39ac83dc2b6b988a7e
52382Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52383Date: Tue May 18 13:53:29 2010 +0300
52384
52385 render: set anim cursor state for pointer enabled devices only
52386
52387 The structure containing the state of animated cursor was amended within
52388 SpriteInfoRec, removing all previously privates logic to keep such state.
52389
52390 API change: It was removed MAXDEVICES dependency \o/
52391
52392 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52393 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
52394 Reviewed-by: Keith Packard <keithp@keithp.com>
52395
52396commit 99fcf655bdc96f7e5b2249b05da51ba4fc9190c7
52397Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52398Date: Sun May 2 22:34:44 2010 +0300
52399
52400 xfree86: remove PCI dependency from InitOutput
52401
52402 All functions that touch PCI and BUS were moved to their own files, organizing
52403 the mess inside the InitOutput. Now, inside InitOutput, mostly accesses to
52404 buses are coordinated by the new xf86BusConfig.
52405
52406 Two PCI probe functions just changed the name and a procedure to receive the
52407 isolate devices parameters was created also, named xf86PciIsolateDevice.
52408
52409 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52410 Reviewed-by: Adam Jackson <ajax@redhat.com>
52411
52412commit 0dd299864a133d478e4635bd1cd305b67fc3964d
52413Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52414Date: Sun May 2 21:33:26 2010 +0300
52415
52416 xfree86: remove xf86EnableAccess
52417
52418 It's RAC remnant. This was substituted by xf86VGAarbiter{Lock, Unlock}
52419 mechanism.
52420
52421 It's an API break, but the few drivers using it were covered already with
52422 macros to avoid problems.
52423
52424 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52425 Reviewed-by: Adam Jackson <ajax@redhat.com>
52426
52427commit f28515b5f774b83e0481acbcdda2f682738079b9
52428Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52429Date: Sun May 2 21:24:46 2010 +0300
52430
52431 xfree86: bus: remove SetSIGIOForState and simplify the code
52432
52433 Also removed some dumb debug messages.
52434
52435 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52436 Reviewed-by: Adam Jackson <ajax@redhat.com>
52437
52438commit 211ca67e4d8d00b20a74a78304a61d4dbea62706
52439Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52440Date: Sun May 2 21:16:23 2010 +0300
52441
52442 xfree86: bus: fb drivers might want to use vga arbitration either
52443
52444 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52445 Reviewed-by: Adam Jackson <ajax@redhat.com>
52446
52447commit 2464eb7d56407422e033ab3b94634ba802a119d5
52448Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52449Date: Sun May 2 21:14:53 2010 +0300
52450
52451 xfree86: bus: simplify entity related hooks
52452
52453 Remove some out dated commentaries either.
52454
52455 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52456 Reviewed-by: Adam Jackson <ajax@redhat.com>
52457
52458commit 971768e6884f20981652da09a1617c51a760fb86
52459Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52460Date: Fri May 7 15:25:48 2010 +0300
52461
52462 xfree86: bus: enable declaration of sparc function as its code usage
52463
52464 The code in xf86sbusBus.c seems too OS-specific to be usable on OpenBSD.
52465
52466 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52467 Reviewed-by: Adam Jackson <ajax@redhat.com>
52468
52469commit a70ec0d136bbb9efb6f4e683d7ebcea00375083b
52470Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52471Date: Sun May 2 20:48:01 2010 +0300
52472
52473 xfree86: bus: rework xf86PostProbe logic and remove useless log info
52474
52475 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52476 Reviewed-by: Adam Jackson <ajax@redhat.com>
52477
52478commit 2f7d630a1ee446711288af69711d57ea3054b594
52479Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52480Date: Sun May 2 20:29:11 2010 +0300
52481
52482 xfree86: bus: reuse already assigned variable when fb driver claimed
52483
52484 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52485 Reviewed-by: Adam Jackson <ajax@redhat.com>
52486
52487commit 964f29bb80c1bb05508a27969c3ac34cbf072ee9
52488Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52489Date: Sun May 2 20:14:40 2010 +0300
52490
52491 xfree86: bus: fix Enter/Leave accesses behaviour
52492
52493 Okay, seems we're not using extensively such hooks in fact. But fix the
52494 expected behaviour at least.
52495
52496 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52497 Reviewed-by: Adam Jackson <ajax@redhat.com>
52498
52499commit b37462b4da7c389ec7bbcfd800e5f88bded79330
52500Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52501Date: Sun May 2 19:53:37 2010 +0300
52502
52503 xfree86: remove unused xf86AccessInit()
52504
52505 The function was only initializing the boolean xf86ResAccessEnter, which
52506 couldn't get any other value in the life of the server.
52507
52508 The only possible, though suspicious, code was in xf86AccessLeave(), which
52509 could be triggered if AbortDDX is called before xf86AccessInit(). Even so,
52510 such change is safety because no driver would have configured any entity leave
52511 procedure at this point.
52512
52513 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52514 Reviewed-by: Adam Jackson <ajax@redhat.com>
52515
52516commit d88ba7721d2d3b58cdc664fd4c23a3c5e2a5f909
52517Author: Peter Hutterer <peter.hutterer@who-t.net>
52518Date: Tue May 18 11:12:49 2010 +1000
52519
52520 xfree86: Add option parsing for percent options.
52521
52522 In some cases, an option of "50%" would be preferable over fixed value
52523 configuration - especially if the actual values are autoprobed.
52524 Add a new set of functions to parse percent values from configurations.
52525
52526 The percent value parsing differs slightly - if the option is not to marked
52527 as used (e.g. xf86CheckPercentOption()), no warning is emitted to the log
52528 file if the value is not a percent value. This allows double-options (either
52529 as % or as absolute number) without warnings.
52530
52531 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
52532 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
52533 Signed-off-by: Keith Packard <keithp@keithp.com>
52534
52535commit 673eb707ce6737284c4886265ba149c5587a74e2
52536Author: Jamey Sharp <jamey@minilop.net>
52537Date: Sat May 15 12:06:25 2010 -0700
52538
52539 SyncSendAlarmNotifyEvents: check the correct client's clientGone flag.
52540
52541 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52542 Reviewed-by: Keith Packard <keithp@keithp.com>
52543 Signed-off-by: Keith Packard <keithp@keithp.com>
52544
52545commit c3ba199aa643e59abd30170c53fbb4e03077c377
52546Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52547Date: Tue Apr 27 16:27:55 2010 +0300
52548
52549 render: remove unused animcursor fields from private rec
52550
52551 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52552 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
52553
52554commit 9db63e4766bde07ebc375c4390bcc5c854b46ac3
52555Author: Tiago Vignatti <tiago.vignatti@nokia.com>
52556Date: Tue Apr 27 16:18:50 2010 +0300
52557
52558 render: removed unused macro from animcursor
52559
52560 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
52561 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
52562
52563commit cba9942f7716b12c329616a9994c5bce857617fa
52564Author: Kristian Høgsberg <krh@bitplanet.net>
52565Date: Mon May 17 17:40:30 2010 -0400
52566
52567 dri2: Fix NULL check in allocation error handling
52568
52569 1461b31e40a3e44f6327199d23cf2ab70d8f5176 tests to see if the allocation
52570 of buffer[i] failed after i has been incremented, causing it to look
52571 at the wrong pointer.
52572
52573 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
52574 Tested-by: Eric Anholt <eric@anholt.net>
52575 Signed-off-by: Keith Packard <keithp@keithp.com>
52576
52577commit 27cec0a0501369a68031c048e691bda4c7095f72
52578Author: Julien Cristau <jcristau@debian.org>
52579Date: Mon May 17 19:23:44 2010 +0200
52580
52581 Fix build without XACE
52582
52583 dix/colormap.c and dix/gc.c now dereference a ClientPtr, so they need to
52584 include dixstruct.h. Regression introduced by commit
52585 11c69880c7c48ef9e755c4e09fadef7a629d7bc7.
52586
52587 Reported-by: Robert Hooker <sarvatt@ubuntu.com>
52588 Signed-off-by: Julien Cristau <jcristau@debian.org>
52589 Reviewed-by: Jamey Sharp <jamey@minilop.net>
52590 Signed-off-by: Keith Packard <keithp@keithp.com>
52591
52592commit 5a387cf6d248e3cb3337a938200ed5a09fc1d8cb
52593Author: Julien Cristau <jcristau@debian.org>
52594Date: Sun May 16 21:29:02 2010 +0200
52595
52596 Only link Xorg with libconfig.la
52597
52598 Other DDXs don't use input hotplugging since config_init was moved to
52599 the DDX in commit d33adcdf03c69407d151e732fa0cf9947151eb19, so there's
52600 no need to link this in.
52601
52602 Signed-off-by: Julien Cristau <jcristau@debian.org>
52603 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
52604 Signed-off-by: Keith Packard <keithp@keithp.com>
52605
52606commit 37f0b45dbe4ea52bc127720207c7157b2b803ac0
52607Author: Jeremy Huddleston <jeremyhu@apple.com>
52608Date: Sun May 16 10:14:02 2010 -0700
52609
52610 XQuartz: Fix a build failure on Tiger
52611
52612 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
52613
52614commit ecfeabec8d0dcfe286fb893047f1fe1a7ea9f8f5
52615Author: Jeremy Huddleston <jeremyhu@apple.com>
52616Date: Sun May 16 10:03:13 2010 -0700
52617
52618 XQuartz: Don't use deltaXY for determining pointer location on scroll events
52619
52620 <rdar://problem/7989690>
52621
52622 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
52623 Reviewed-by: Edward Moy <emoy@apple.com>
52624
52625commit a911292c85f7069d2caabcb677ed716a04227526
52626Author: Jeremy Huddleston <jeremyhu@apple.com>
52627Date: Sat May 15 10:53:09 2010 -0700
52628
52629 XQuartz: Don't trust deltaXY for middle mouse clicks.
52630
52631 The middle mouse clicks return erroneous values after returning from
52632 Fast User Switching.
52633
52634 <rdar://problem/7979468>
52635 http://xquartz.macosforge.org/trac/ticket/389
52636
52637 Signed-off-by: Martin Otte <otte@duke.edu>
52638 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
52639 Reviewed-by: Edward Moy <emoy@apple.com>
52640 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
52641
52642commit 345eb171264325d73ea2c50ba8c692cf589c2a9b
52643Author: Keith Packard <keithp@keithp.com>
52644Date: Sat May 1 01:41:19 2010 -0700
52645
52646 Remove dbe screen private privates.
52647
52648 Each DBE Screen private structure could have nested privates. Oddly,
52649 no code ever used them.
52650
52651 Signed-off-by: Keith Packard <keithp@keithp.com>
52652 Reviewed-by: Jamey Sharp <jamey@minilop.net>
52653
52654commit 87ea5760f86eb60840e6e2c10012915952df5377
52655Author: Keith Packard <keithp@keithp.com>
52656Date: Mon May 10 11:46:25 2010 -0700
52657
52658 Check pixmap allocation return value when creating glyphs
52659
52660 The driver may decide that the pixmap is too large or something and
52661 fail to allocate a pixmap; not checking would lead to a segfault.
52662
52663 Signed-off-by: Keith Packard <keithp@keithp.com>
52664 Reviewed-by: Jamey Sharp <jamey@minilop.net>
52665
52666commit 03f7e92a6cbfe58db08ce840283f9809695c70e7
52667Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52668Date: Mon May 10 13:44:30 2010 -0700
52669
52670 Update list of supported platforms in configure.ac error message
52671
52672 Adds DragonFly BSD, OpenSolaris, & GNU Hurd.
52673 Drops MacOS X, since this is in the section specific to the Xorg/XFree86 DDX.
52674 (Matches the OS patterns the configure script checks for.)
52675
52676 Also uses m4 macros to fix the spacing/formatting of the resulting message.
52677
52678 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52679 Acked-by: Gaetan Nadon <memsize@videotron.ca>
52680
52681commit e42a29d269fadc11e065c63ee369e3165196f2d7
52682Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52683Date: Thu Apr 29 18:45:34 2010 -0700
52684
52685 Fix compiler issues with getifaddrs() call on OpenSolaris
52686
52687 OpenSolaris recently added support for the getifaddrs() API.
52688
52689 Building with that uncovered two compiler issues (one warning, one error)
52690 in the code that was now being built for the first time in our builds:
52691
52692 "access.c", line 768: warning: argument #1 is incompatible with prototype:
52693 prototype: pointer to struct sockaddr {unsigned short sa_family, array[14] of char sa_data} : "access.c", line 213
52694 argument : pointer to struct sockaddr_storage {unsigned short ss_family, array[6] of char _ss_pad1, double _ss_align, array[240] of char _ss_pad2}
52695
52696 "access.c", line 838: assignment type mismatch:
52697 struct sockaddr {unsigned short sa_family, array[14] of char sa_data} "=" struct sockaddr_storage {unsigned short ss_family, array[6] of char _ss_pad1, double _ss_align, array[240] of char _ss_pad2}
52698
52699 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52700 Reviewed-by: Jamey Sharp <jamey@minilop.net>
52701
52702commit c6613cfc26e76b296e620640bf33e756eb4ae178
52703Author: Jamey Sharp <jamey@minilop.net>
52704Date: Fri May 14 10:36:13 2010 -0700
52705
52706 test/xi2: Initialize predefined atoms before XInputExtensionInit.
52707
52708 XInputExtensionInit calls MakeAtom, which doesn't work without the atoms
52709 table initialized.
52710
52711 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52712 Cc: Peter Hutterer <peter.hutterer@who-t.net>
52713 Tested-by: Keith Packard <keithp@keithp.com>
52714 Signed-off-by: Keith Packard <keithp@keithp.com>
52715
52716commit 1461b31e40a3e44f6327199d23cf2ab70d8f5176
52717Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
52718Date: Fri May 14 16:25:43 2010 +0300
52719
52720 dri2: Fix xserver crash if dri2 buffer allocation fails.
52721
52722 If driver fails to allocate memory for dri2 buffer server would crash
52723 in send_buffers_reply.
52724
52725 Solution is to handle the allocation failure and return BadAlloc to
52726 client.
52727
52728 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
52729 Reviewed-by: Michel Dänzer <michel@daenzer.net>
52730 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
52731 Signed-off-by: Keith Packard <keithp@keithp.com>
52732
52733commit f87002cb7d8729b1da4cbefe7653f4cfd3a2d956
52734Merge: a4fbc0f 890f536
52735Author: Keith Packard <keithp@keithp.com>
52736Date: Fri May 14 11:46:29 2010 -0700
52737
52738 Merge remote branch 'jamey/for-keith'
52739
52740commit 890f536f8d66f08e7d6a5685659eef4001a8757d
52741Author: Jamey Sharp <jamey@minilop.net>
52742Date: Fri May 14 00:18:14 2010 -0700
52743
52744 Bugfix for "Pre-validate ChangeGC XIDs": off-by-one in loop index.
52745
52746 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52747 Tested-by: Colin Harrison <colin.harrison@virgin.net>
52748
52749commit a4fbc0fedafda6df98f647173a2deccb86ef0479
52750Author: Oliver McFadden <oliver.mcfadden@nokia.com>
52751Date: Fri May 14 07:49:31 2010 +0300
52752
52753 scrnintstr.h: removed unused PaintWindow function pointers.
52754
52755 Please note this patch breaks the ABI.
52756
52757 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
52758 Reviewed-by: Keith Packard <keithp@keithp.com>
52759
52760commit 71296885b0bb946e7ab986169f5b872952c221a0
52761Author: Oliver McFadden <oliver.mcfadden@nokia.com>
52762Date: Thu May 13 07:12:43 2010 +0300
52763
52764 mi: removed unused PointerNonInterestBox functions.
52765
52766 Please note this patch breaks the ABI.
52767
52768 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
52769 Reviewed-by: Keith Packard <keithp@keithp.com>
52770
52771commit 795432d4a92ed5b9fa4d9163e73c8e4fe4c74534
52772Merge: f2a0c32 92ed75a
52773Author: Keith Packard <keithp@keithp.com>
52774Date: Thu May 13 18:22:49 2010 -0700
52775
52776 Merge remote branch 'jamey/cleanups'
52777
52778commit 92ed75ac59e2d3af149cddb962efd05fc8487750
52779Author: Jamey Sharp <jamey@minilop.net>
52780Date: Mon May 10 20:22:05 2010 -0700
52781
52782 Eliminate boilerplate around client->noClientException.
52783
52784 Just let Dispatch() check for a noClientException, rather than making
52785 every single dispatch procedure take care of it.
52786
52787 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52788 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
52789
52790commit 11c69880c7c48ef9e755c4e09fadef7a629d7bc7
52791Author: Jamey Sharp <jamey@minilop.net>
52792Date: Sat May 8 22:16:32 2010 -0700
52793
52794 Quit using clientErrorValue in dix/colormap.c.
52795
52796 And that's it! No more clientErrorValue kludge.
52797
52798 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52799 Reviewed-by: Keith Packard <keithp@keithp.com>
52800
52801commit a3d948ddbb54b9e831e67f22d5031922a3c44107
52802Author: Jamey Sharp <jamey@minilop.net>
52803Date: Thu May 6 12:35:52 2010 -0700
52804
52805 clientErrorValue is never used outside dix. Stop importing it.
52806
52807 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52808 Reviewed-by: Keith Packard <keithp@keithp.com>
52809
52810commit 653e4878c4cc03613172a93ad4800e1eacc98f17
52811Author: Jamey Sharp <jamey@minilop.net>
52812Date: Sat May 8 19:18:11 2010 -0700
52813
52814 Quit using clientErrorValue in dix/gc.c.
52815
52816 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52817 Reviewed-by: Keith Packard <keithp@keithp.com>
52818
52819commit 5193f25ea33eed31d6a75cdc1a86427c23d8033c
52820Author: Jamey Sharp <jamey@minilop.net>
52821Date: Sat May 8 19:08:47 2010 -0700
52822
52823 Define GCAllBits as the union of all valid CreateGC masks.
52824
52825 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52826 Reviewed-by: Keith Packard <keithp@keithp.com>
52827
52828commit 6a84cd943430cfc9df55c83aef6a7f8dea6dbb94
52829Author: Jamey Sharp <jamey@minilop.net>
52830Date: Fri May 7 19:38:05 2010 -0700
52831
52832 Replace dixChangeGC with calls directly to the right variant.
52833
52834 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52835 Reviewed-by: Keith Packard <keithp@keithp.com>
52836
52837commit 2d7eb4a19b773d0406c0c2e018a7da97f3565fd5
52838Author: Jamey Sharp <jamey@minilop.net>
52839Date: Fri May 7 18:11:36 2010 -0700
52840
52841 Pre-validate ChangeGC XIDs.
52842
52843 In order to execute a wire-level ChangeGC request, we need to look up
52844 the resources named by any XIDs in the value-list. Various places in the
52845 server already have pointers to the resources they want to set into the
52846 GC, though, so over time the interface has evolved to accept either XIDs
52847 or pointers, with several different function call signatures used in
52848 different eras.
52849
52850 This patch makes the existing code require pointers to resources rather
52851 than XIDs, and adds a simple wrapper that looks up any XIDs. The old
52852 dixChangeGC API is preserved by delegating to whichever implementation
52853 is appropriate.
52854
52855 This affects error-handling: If any of the XIDs are invalid, then the GC
52856 is unchanged, and its ChangeGC callback is not invoked. This change is
52857 allowed by the protocol spec, which says, "The order in which components
52858 are verified and altered is server-dependent. If an error is generated,
52859 a subset of the components may have been altered."
52860
52861 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52862 Reviewed-by: Keith Packard <keithp@keithp.com>
52863
52864commit e2929db7b737413cf93fbebdf4d15abdfebff05c
52865Author: Jamey Sharp <jamey@minilop.net>
52866Date: Sat May 8 14:10:51 2010 -0700
52867
52868 dixChangeGC callers: Use ChangeGCVal instead of XID almost everywhere.
52869
52870 The exceptions are ProcChangeGC and CreateGC.
52871
52872 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52873 Reviewed-by: Keith Packard <keithp@keithp.com>
52874
52875commit 95728ca09d45afc84c8d1828c09c6b6725f1a58d
52876Author: Jamey Sharp <jamey@minilop.net>
52877Date: Sat May 8 09:25:34 2010 -0700
52878
52879 Don't statically allocate the ChangeGC parameter array.
52880
52881 Because that's insane.
52882
52883 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52884 Reviewed-by: Keith Packard <keithp@keithp.com>
52885
52886commit bff8525f8483304d5f93e83e36b47209381da721
52887Author: Jamey Sharp <jamey@minilop.net>
52888Date: Sat May 8 12:06:50 2010 -0700
52889
52890 Simplify miFillPolyHelper and miLineArc.
52891
52892 Both functions compute a set of spans and either fill them immediately
52893 or accumulate them into a caller-provided buffer.
52894
52895 Computing the spans used only the miTranslate and lineWidth fields of
52896 pGC, and neither could have been changed by the initial
52897 ChangeGC/ValidateGC pair, so it's safe to compute the spans first.
52898
52899 Then both functions consume the spans the same way, so factor that into
52900 a new fillSpans function.
52901
52902 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52903 Reviewed-by: Keith Packard <keithp@keithp.com>
52904
52905commit 83f7ec97279405958aed86c6a57704a460c3bfba
52906Author: Jamey Sharp <jamey@minilop.net>
52907Date: Sat May 8 13:38:00 2010 -0700
52908
52909 miwideline: Factor out span buffer allocation.
52910
52911 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52912 Reviewed-by: Keith Packard <keithp@keithp.com>
52913 Reviewed-by: Adam Jackson <ajax@redhat.com>
52914
52915commit f2a0c324e37c9fa3eb9087adbf963addb7f21d88
52916Author: Keith Packard <keithp@keithp.com>
52917Date: Thu May 13 10:33:39 2010 -0700
52918
52919 Remove devPrivates init and delete callback lists.
52920
52921 XSELinux was the only consumer of these interfaces and it no longer
52922 needs them.
52923
52924 Signed-off-by: Keith Packard <keithp@keithp.com>
52925 Reviewed-by: Jamey Sharp <jamey@minilop.net>
52926
52927commit c9e7ca4404803fe44d4684e0bb2ca2ee10fd4bb3
52928Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
52929Date: Wed May 12 20:25:30 2010 -0400
52930
52931 xselinux: Remove use of devPrivates init/free callbacks.
52932
52933 Commit eb9210097efea81065c301e5b6b4da7a566deb4a removed the
52934 sidget/sidput calls which were the major reason for using the
52935 callbacks. The remaining operations can be skipped or worked
52936 around.
52937
52938 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
52939 Reviewed-by: Keith Packard <keithp@keithp.com>
52940 Signed-off-by: Keith Packard <keithp@keithp.com>
52941
52942commit bca85e2e127a8a23e3a2debcfeb3ae07cd3c66ac
52943Author: Jeremy Huddleston <jeremyhu@apple.com>
52944Date: Thu May 13 08:32:00 2010 -0700
52945
52946 Use _X_ATTRIBUTE_PRINTF _X_DEPRECATED _X_NORETURN
52947
52948 Use the values from xproto rather than duplicating the effort
52949
52950 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
52951 Reviewed-by: Keith Packard <keithp@keithp.com>
52952 Signed-off-by: Keith Packard <keithp@keithp.com>
52953
52954commit 421606a8ef447d10c2ee0986f20e752056a47675
52955Author: Kristian Høgsberg <krh@bitplanet.net>
52956Date: Tue May 11 10:52:18 2010 -0400
52957
52958 dri2: Send out event when auxillary buffers are invalidated
52959
52960 This lets the DRI2 clients rely on the server to notify them when they
52961 need to get new buffers. Without this, OpenGL clients poll the server
52962 in glViewport() which can be a performance problems and also isn't
52963 completely correct behaviour.
52964
52965 We bump the DRI2 protocol minor to indicate the availability of the
52966 event, which the DRI2 clients can use to avoid polling. This speeds up
52967 various piglit and oglc test cases as well as real applications.
52968
52969 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
52970 Reviewed-by: Adam Jackson <ajax@redhat.com>
52971 Signed-off-by: Keith Packard <keithp@keithp.com>
52972
52973commit f281db9a5e400c822e03a19937247baa20ecc213
52974Author: Alan Coopersmith <alan.coopersmith@oracle.com>
52975Date: Wed May 12 23:57:58 2010 -0700
52976
52977 Remove ResNoAvoid definition, missed in the RAC removal
52978
52979 ResNoAvoid is #defined to ResBios, but ResBios was removed
52980 from xf86str.h in 4b42448a2388d40f257774fbffdccaea87bd0347
52981
52982 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
52983 Reviewed-by: Julien Cristau <jcristau@debian.org>
52984 Signed-off-by: Keith Packard <keithp@keithp.com>
52985
52986commit 315041762313598aad90df84226e2d2def4a0fc9
52987Author: Jamey Sharp <jamey@minilop.net>
52988Date: Sat May 8 18:55:55 2010 -0700
52989
52990 CreateGC: Pass current client on to ChangeGC.
52991
52992 Otherwise we can't check that the XIDs this GC is being initialized with
52993 are accessible to this client.
52994
52995 Signed-off-by: Jamey Sharp <jamey@minilop.net>
52996 Cc: Eamon Walsh <ewalsh@tycho.nsa.gov>
52997 Reviewed-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
52998
52999commit 04bad1b8a118a83064316d6e4327234aee19361e
53000Author: Jamey Sharp <jamey@minilop.net>
53001Date: Thu May 6 21:56:10 2010 -0700
53002
53003 Kill ChangeGC in favor of dixChangeGC.
53004
53005 This doesn't change any behavior, but it isn't clear whether NullClient
53006 is correct in all cases. As ajax says,
53007
53008 > For most of these changes, I think it's correct to use NullClient,
53009 > since they are server-initiated changes and should not fail for (eg)
53010 > xace reasons. ... At any rate, you're certainly not changing any
53011 > semantics by leaving them all as NullClient, so this patch can't be
53012 > more wrong than before.
53013
53014 The call in CreateGC is particularly questionable.
53015
53016 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53017 Reviewed-by: Adam Jackson <ajax@redhat.com>
53018
53019commit 65cedf39054cf3a1e695e84ac228cce9e8d48097
53020Author: Jamey Sharp <jamey@minilop.net>
53021Date: Thu May 6 21:39:43 2010 -0700
53022
53023 Kill DoChangeGC in favor of dixChangeGC.
53024
53025 This doesn't change any behavior, but it isn't clear whether NullClient
53026 is correct in all cases. As ajax says,
53027
53028 > For most of these changes, I think it's correct to use NullClient,
53029 > since they are server-initiated changes and should not fail for (eg)
53030 > xace reasons. ... At any rate, you're certainly not changing any
53031 > semantics by leaving them all as NullClient, so this patch can't be
53032 > more wrong than before.
53033
53034 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53035 Reviewed-by: Keith Packard <keithp@keithp.com>
53036
53037commit afcbbd6dfea51c5b0adca0d720edc02ba0c2dc16
53038Author: Jamey Sharp <jamey@minilop.net>
53039Date: Sat May 8 15:19:03 2010 -0700
53040
53041 doPolyText needs UseAccess to each font, not ReadAccess.
53042
53043 In commit 42d6112ec21949a336ee8b34469f2695273ee2d6, Eamon changed
53044 dixChangeGC to require DixUseAccess on any GCFont XID. I think
53045 doPolyText needs to require the same level of access. Otherwise
53046 dixChangeGC could fail when it does the same lookup, which doPolyText
53047 doesn't check for.
53048
53049 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53050 Cc: Eamon Walsh <ewalsh@tycho.nsa.gov>
53051 Reviewed-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
53052
53053commit 59857ee5da5f1f3f4900292581b9586477513211
53054Merge: 21ceae9 432cbbe
53055Author: Keith Packard <keithp@keithp.com>
53056Date: Wed May 12 16:48:08 2010 -0700
53057
53058 Merge remote branch 'dottedmag/for-keithp'
53059
53060commit 432cbbec194e47bf2a117c9302146e786c8a4ee1
53061Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53062Date: Thu May 13 03:51:00 2010 +0700
53063
53064 Misc coding style cleanup
53065
53066 Use a[b] instead of *(a+b), fix whitespace.
53067
53068 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53069 Reviewed-by: Keith Packard <keithp@keithp.com>
53070
53071commit 816b79dd061e9839cec94a4986a7820b70ca8a7f
53072Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53073Date: Thu May 13 03:45:21 2010 +0700
53074
53075 Remove useless casts
53076
53077 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53078 Reviewed-by: Keith Packard <keithp@keithp.com>
53079
53080commit 63a647abd51f44226cbd16aa04ebc57d07463c6d
53081Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53082Date: Thu May 13 03:44:12 2010 +0700
53083
53084 Fix code style: extra whitespace before ()
53085
53086 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53087 Reviewed-by: Keith Packard <keithp@keithp.com>
53088
53089commit 28211c443c693a1ca3db5740d0128274a3eef723
53090Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53091Date: Thu May 13 03:43:04 2010 +0700
53092
53093 Fix warning: it's safe to pass atom strings > XA_LAST_PREDEFINED to free(3)
53094
53095 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53096 Reviewed-by: Keith Packard <keithp@keithp.com>
53097
53098commit 8b5326aa98eba201dd78aea3dd7114e1a084489b
53099Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53100Date: Wed May 12 20:27:02 2010 +0000
53101
53102 Mark OsAbort as noreturn function to make gcc happier.
53103
53104 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53105 Reviewed-by: Matt Turner <mattst88@gmail.com>
53106
53107commit 868e372a73b377705217e0379bc6e00f36c4d8e5
53108Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53109Date: Thu May 13 01:59:06 2010 +0700
53110
53111 Introduce X_NORETURN macro defined as __attribute__((noreturn)) for gcc
53112
53113 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53114 Reviewed-by: Matt Turner <mattst88@gmail.com>
53115
53116commit 5a8e2f2745ae1f74501cd3f42614a1ed2cf974f2
53117Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53118Date: Wed May 12 18:54:51 2010 +0000
53119
53120 Do not jump through the hoops to deallocate xkbbasedirflag variable
53121
53122 Fixes gcc warning as well.
53123
53124 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53125 Reviewed-by: Jamey Sharp <jamey@minilop.net>
53126 Reviewed-by: Matt Turner <mattst88@gmail.com>
53127
53128commit ff2b4cf8329b1678adafcda02e5d47a072550d47
53129Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53130Date: Thu May 13 01:51:37 2010 +0700
53131
53132 Turn sprintf argument into literaral string, shutting up gcc warning
53133
53134 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53135 Reviewed-by: Matt Turner <mattst88@gmail.com>
53136
53137commit f62ba192c285b1e49bf299f03fc0b763680afaaf
53138Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53139Date: Thu May 13 01:47:26 2010 +0700
53140
53141 Do not use deprecated Xalloc function
53142
53143 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53144 Reviewed-by: Matt Turner <mattst88@gmail.com>
53145
53146commit 21ceae9002c6364deb3d074cf2da7d3864cf6879
53147Author: Jamey Sharp <jamey@minilop.net>
53148Date: Tue May 11 10:24:00 2010 -0700
53149
53150 SetFontPath: set client->errorValue on failure.
53151
53152 Previously the callers were only setting errorValue on Success, when
53153 it's ignored, and leaving it alone on failure, when it's sent to the
53154 client.
53155
53156 Since SetFontPath takes the ClientPtr, let it set client->errorValue
53157 instead of letting the callers continue to get it wrong.
53158
53159 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53160 Reviewed-by: Julien Cristau <jcristau@debian.org>
53161 Signed-off-by: Keith Packard <keithp@keithp.com>
53162
53163commit 801162919d1c625d950a5d105ae4b3487ebc30a7
53164Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53165Date: Thu May 6 01:55:36 2010 +0700
53166
53167 Removed outdated comments obsoleted by documentation in headers
53168
53169 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53170 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53171
53172commit 4d55c8c4ffcb5d6c6500b462a41ace8e2359151d
53173Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53174Date: Thu May 6 01:54:13 2010 +0700
53175
53176 Use lowercase variant of XNFalloc and Xstrdup
53177
53178 Using one variant of function/macro makes it easier to fix the code
53179 later.
53180
53181 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53182 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53183
53184commit 3f3ff971ecff9936cebafc813af9193b97bba89c
53185Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53186Date: Thu May 6 01:44:06 2010 +0700
53187
53188 Replace X-allocation functions with their C89 counterparts
53189
53190 The only remaining X-functions used in server are XNF*, the rest is converted to
53191 plain alloc/calloc/realloc/free/strdup.
53192
53193 X* functions are still exported from server and x* macros are still defined in
53194 header file, so both ABI and API are not affected by this change.
53195
53196 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53197 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53198
53199commit 96c7ab27c383ec767f62a7a11e5fd76f86363fbc
53200Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53201Date: Thu May 6 00:28:25 2010 +0700
53202
53203 Deprecate allocation functions which are plain wrappers for C stdlib
53204
53205 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53206 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53207
53208commit a5dba0f5caefce236ebd5f71948e5a659bea58e4
53209Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53210Date: Thu May 6 00:27:47 2010 +0700
53211
53212 Document allocation functions, noting deviations from C library
53213
53214 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53215 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53216
53217commit e983848ab44b0769f97f6207f1aa8b4f127be6a9
53218Author: Mikhail Gusarov <dottedmag@dottedmag.net>
53219Date: Thu May 6 00:16:24 2010 +0700
53220
53221 Clean {X,XNF}{alloc,calloc,realloc,free,strdup} from pre-C89 baggage
53222
53223 C89 guarantees alignment of pointers returned from malloc/calloc/realloc, so
53224 stop fiddling with alignment manually and just pass the arguments to library
53225 functions.
53226
53227 Also convert silent error when negative size is passed into function into
53228 warning in log file.
53229
53230 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
53231 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53232
53233commit 4f0006c2203abe0b7660c5068d6afe236f9bd2a4
53234Author: Kristian Høgsberg <krh@bitplanet.net>
53235Date: Mon May 10 08:36:02 2010 -0400
53236
53237 dri2: Copy front to fake front in swapbuffer
53238
53239 If we don't have a fake front, nothing will happen. The fix was extracted
53240 from a bigger patch from Francisco Jerez.
53241
53242 https://bugs.freedesktop.org/show_bug.cgi?id=27305
53243 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
53244 Reviewed-by: Pauli Nieminen <suokkos@gmail.com>
53245 Signed-off-by: Keith Packard <keithp@keithp.com>
53246
53247commit ba1f1f9d9e48226f1ef6b3222a8d92fc969d6560
53248Author: Jeremy Huddleston <jeremyhu@apple.com>
53249Date: Tue May 11 09:49:34 2010 -0700
53250
53251 XQuartz GL: Fix __glXAquaScreenCreateDrawable prototype
53252
53253 Fixes regression introduced in 9de0e31746d5f0d9d39d11c94ec3cbc04a9935fc
53254
53255 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
53256 Reviewed-by: Keith Packard <keithp@keithp.com>
53257 Signed-off-by: Keith Packard <keithp@keithp.com>
53258
53259commit 286935c6f4965a6b09de1fb9e579a00afa14082a
53260Author: Keith Packard <keithp@keithp.com>
53261Date: Tue May 11 09:08:40 2010 -0700
53262
53263 Remember drawable ID used to find damage target for use in events. Bug 5730.
53264
53265 Using Composite, window pixmaps are given names in the client resource
53266 namespace and yet may not have any XID recorded in the drawable
53267 structure. As such, we need to remember the XID used to lookup the
53268 pixmap in the resource database so that we can report the correct XID
53269 back to the client in damage events.
53270
53271 Signed-off-by: Keith Packard <keithp@keithp.com>
53272 Reviewed-by: Adam Jackson <ajax@redhat.com>
53273
53274commit 3df22b293c1f1d27e7ce16662744c578d35b5aba
53275Merge: fa6c701 9de0e31
53276Author: Keith Packard <keithp@keithp.com>
53277Date: Tue May 11 08:38:44 2010 -0700
53278
53279 Merge remote branch 'krh/dri2-resource-fixes'
53280
53281commit fa6c7012572093a82c9389682977efff85590719
53282Author: Cyril Brulebois <kibi@debian.org>
53283Date: Mon Mar 1 02:11:36 2010 +0100
53284
53285 Fix null pointer dereference in xf86_reload_cursors().
53286
53287 Upon resume, X may try to dereference a null pointer, which has been
53288 reported in Debian bug #507916 (http://bugs.debian.org/507916).
53289
53290 Jim Paris came up with a patch which solves the problem for him. Here's
53291 a (hopefully) fixed version of his patch (without the typo).
53292
53293 Cc: Jim Paris <jim@jtan.com>
53294 Signed-off-by: Cyril Brulebois <kibi@debian.org>
53295 Reviewed-By: Matthias Hopf <mhopf@suse.de>
53296 Signed-off-by: Keith Packard <keithp@keithp.com>
53297
53298commit 9de0e31746d5f0d9d39d11c94ec3cbc04a9935fc
53299Author: Kristian Høgsberg <krh@bitplanet.net>
53300Date: Thu Apr 29 16:36:10 2010 -0400
53301
53302 dri2: Take an XID for tracking the DRI2 drawable
53303
53304 Some pixmaps (window pixmaps and scratch pixmaps) don't have the
53305 drawable->id set and thus DRI2 gets confused when using that field
53306 for looking up the DRI2 drawable. Go back to using privates for getting
53307 at the DRI2 drawable from a DrawablePtr. We need to keep the resource
53308 tracking in place so we can remove the DRI2 drawable when the X resource
53309 it was created for goes away. Additionally, we also now track the DRI2
53310 drawable using a client XID so we can reclaim the DRI2 drawable even if
53311 the client goes before the drawable and doesn't destroy the DRI2 drawable.
53312
53313 Tested-by: Owen W. Taylor <otaylor@fishsoup.net>
53314 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
53315
53316commit 32381363cd8f43aeb741bad70bcf96a287dac0c9
53317Author: Kristian Høgsberg <krh@bitplanet.net>
53318Date: Sat May 1 13:15:00 2010 -0400
53319
53320 list.h: Add list_for_each_entry_safe()
53321
53322 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
53323 Reviewed-by: Adam Jackson <ajax@redhat.com>
53324
53325commit 4a8a615d01b9ed18c272414bd11dc2fc661727e5
53326Author: Kristian Høgsberg <krh@bitplanet.net>
53327Date: Sat May 1 13:13:54 2010 -0400
53328
53329 glxdri2: Hard-code the extension version we need
53330
53331 If we use the #define'd version from dri_interface.h, the server will
53332 require at least that version of the extension. If we're compiling against
53333 a dri_interface.h with a newer version we don't really require, glxdri2
53334 will require a too high version of the extension.
53335
53336 The right approach is to just hard-code the version we need instead of
53337 using the #defines.
53338
53339 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
53340 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
53341 Reviewed-by: Adam Jackson <ajax@redhat.com>
53342
53343commit 6d7ba5e0fcb5d1bce6bb213dec009f3a0f802d26
53344Author: Kristian Høgsberg <krh@bitplanet.net>
53345Date: Sat May 1 13:07:46 2010 -0400
53346
53347 dix: Update element count in FreeResource*()
53348
53349 FreeResource() keeps clientTable[cid].elements up to date with the
53350 number of resources allocated to the client. The other free
53351 resource functions (FreeResourceByType(),
53352 FreeClientNeverRetainResources() and FreeClientResources()) don't
53353 maintain this invariant.
53354
53355 Typically, the only consequence is that the element count is too high
53356 and we end up allocating the hash table bigger than necessary. However,
53357 FreeResource() also relies on the element count to restart the search if
53358 the list of resources has been changed during a resource destruction
53359 callback. Since FreeResourceByType() doesn't update the count, if we call
53360 that from a resource destruction callback from FreeResource(), the
53361 loop isn't restarted and we end up following an invalid next pointer.
53362
53363 Furthermore, LookupClientResourceComplex() and
53364 FreeClientNeverRetainResources() don't use the element count to detect
53365 if a callback deleted a resource and may end up following an invalid
53366 next pointer if the resource system is called into recursively.
53367
53368 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
53369 Reviewed-by: Keith Packard <keithp@keithp.com>
53370
53371commit 10de9e8ee37265a35ceeceb2007d711da70d4f2d
53372Author: Peter Hutterer <peter.hutterer@who-t.net>
53373Date: Fri Apr 16 16:35:22 2010 +1000
53374
53375 xfree86: dga needs to use the master keyboard state (#27573)
53376
53377 GetPairedDevice() may not always return the keyboard, resulting in a
53378 null-pointer dereference when accessing the XKB state.
53379 For floating devices, the GetMaster() returns the device itself.
53380
53381 X.Org Bug 27573 <http://bugs.freedesktop.org/show_bug.cgi?id=27573>
53382
53383 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
53384 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
53385 Tested-by: Ben Hutchings <ben@decadent.org.uk>
53386
53387commit 758b8614477b53dc3de2b884fec5ccaf8a736432
53388Author: David James <davidjames@google.com>
53389Date: Mon May 10 14:00:49 2010 -0700
53390
53391 Fix wrong bracket values when startOver = FALSE.
53392
53393 Currently, SyncComputeBracketValues reuses old values of bracket_greater
53394 and bracket_less when startOver = FALSE. This can result in incorrect bracket
53395 values. To fix this issue, the startOver parameter is removed, and we do not
53396 reuse old values of bracket_greater and bracket_less.
53397
53398 X.Org Bug 27023 <http://bugs.freedesktop.org/show_bug.cgi?id=27023>
53399
53400 Signed-off-by: David James <davidjames@google.com>
53401 Reviewed-by: Adam Jackson <ajax@redhat.com>
53402 Signed-off-by: Keith Packard <keithp@keithp.com>
53403
53404commit 95074538cce93abc1739a4d9dd78b0500fb8fec9
53405Merge: c079977 2160ff5
53406Author: Keith Packard <keithp@keithp.com>
53407Date: Mon May 10 13:49:33 2010 -0700
53408
53409 Merge remote branch 'jamey/for-keith'
53410
53411commit 2160ff5240d7e11bdc6cec8859fd1f10395c8edd
53412Author: Jamey Sharp <jamey@minilop.net>
53413Date: Fri May 7 20:00:40 2010 -0700
53414
53415 Simplify error-handling in dixChangeGC.
53416
53417 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53418 Reviewed-by: Keith Packard <keithp@keithp.com>
53419
53420commit 4080cd42fde8b4c693eab7557581af2622479109
53421Author: Jamey Sharp <jamey@minilop.net>
53422Date: Sat May 8 09:41:46 2010 -0700
53423
53424 miPolyPoint bugfix: Check memory allocation before changing the GC.
53425
53426 miPolyPoint ought to leave the GC unchanged even if it fails. ajax says:
53427
53428 > We have a new winner for the oldest-bug competition! It's actually
53429 > been like that since X11R1:
53430 >
53431 > -rw-r--r--. 1 ajax ajax 2817 1987-09-12 01:20 ddx/mi/mipolypnt.c
53432
53433 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53434 Reviewed-by: Adam Jackson <ajax@redhat.com>
53435
53436commit ccdaecc5471cc2d56b1702eeed4280bb4d58c515
53437Author: Jamey Sharp <jamey@minilop.net>
53438Date: Sat May 8 12:26:47 2010 -0700
53439
53440 Move AppendSpanGroup from miwideline.h to miwideline.c.
53441
53442 Also make it stop being a macro.
53443
53444 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53445 Reviewed-by: Keith Packard <keithp@keithp.com>
53446
53447commit 4272d6722c0f65bdfafc95e91cc2fdb8f6bd072e
53448Author: Jamey Sharp <jamey@minilop.net>
53449Date: Sat May 8 12:18:37 2010 -0700
53450
53451 Expand macros that are used only in miFillPolyHelper.
53452
53453 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53454 Reviewed-by: Keith Packard <keithp@keithp.com>
53455
53456commit c0799779dffeef8cef228bdc57aa6c8b323e7485
53457Author: Keith Packard <keithp@keithp.com>
53458Date: Mon Apr 26 17:22:12 2010 -0700
53459
53460 dmxBERestoreRenderGlyph was passing wrong value to free
53461
53462 Signed-off-by: Keith Packard <keithp@keithp.com>
53463 Reviewed-by: Jamey Sharp <jamey@minilop.net>
53464 Reviewed-by: Matt Turner <mattst88@gmail.com>
53465
53466commit 946f664b6ab421f69df3dc3213b14b085424c9e4
53467Author: Adam Jackson <ajax@redhat.com>
53468Date: Thu May 6 17:58:29 2010 -0400
53469
53470 mi: Delete wide ellipse arc cache
53471
53472 This does make wide ellipses slower, by a factor of twoish:
53473
53474 946000.0 521000.0 ( 0.55) 10-pixel wide ellipse
53475 98300.0 49900.0 ( 0.51) 100-pixel wide ellipse
53476 17700.0 9310.0 ( 0.53) 500-pixel wide ellipse
53477 16900.0 7980.0 ( 0.47) 100-pixel wide dashed ellipse
53478 16100.0 5370.0 ( 0.33) 100-pixel wide double-dashed ellipse
53479
53480 But no one cares, and it's a modest size win:
53481
53482 text data bss dec hex filename
53483 1773824 69552 59288 1902664 1d0848 hw/vfb/Xvfb.before
53484 1773112 69552 58648 1901312 1d0300 hw/vfb/Xvfb
53485
53486 Signed-off-by: Adam Jackson <ajax@redhat.com>
53487 Reviewed-by: Keith Packard <keithp@keithp.com>
53488 Signed-off-by: Keith Packard <keithp@keithp.com>
53489
53490commit 580b99bb8af4ea093953c2b885bd4272aaf02baf
53491Author: Jamey Sharp <jamey@minilop.net>
53492Date: Thu May 6 11:00:39 2010 -0700
53493
53494 Return BadPicture, not BadPixmap, if alpha-map is invalid.
53495
53496 I believe this is what was intended.
53497
53498 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53499 Reviewed-by: Adam Jackson <ajax@redhat.com>
53500 Signed-off-by: Keith Packard <keithp@keithp.com>
53501
53502commit c677fc611b124c44adfcfb95718798105f26357b
53503Author: Jamey Sharp <jamey@minilop.net>
53504Date: Thu May 6 11:00:38 2010 -0700
53505
53506 VERIFY_PICTURE always returns BadPicture. Don't bother specifying.
53507
53508 Same goes for VERIFY_ALPHA, VERIFY_XIN_PICTURE, and VERIFY_XIN_ALPHA.
53509
53510 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53511 Reviewed-by: Julien Cristau <jcristau@debian.org>
53512 Reviewed-by: Adam Jackson <ajax@redhat.com>
53513 Signed-off-by: Keith Packard <keithp@keithp.com>
53514
53515commit 35761d5f811406bc0b6a68c1b02bdb699142745c
53516Author: Jamey Sharp <jamey@minilop.net>
53517Date: Thu May 6 11:00:37 2010 -0700
53518
53519 Introduce dixLookupFontable for "FONT or GC" parameters.
53520
53521 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53522 Reviewed-by: Julien Cristau <jcristau@debian.org>
53523 Reviewed-by: Adam Jackson <ajax@redhat.com>
53524 Signed-off-by: Keith Packard <keithp@keithp.com>
53525
53526commit 2eab697adba4b1858a530750e9a35fba79a7bf26
53527Author: Alan Coopersmith <alan.coopersmith@oracle.com>
53528Date: Fri May 7 09:32:34 2010 -0700
53529
53530 Constify function prototypes in auth & xdmcp code
53531
53532 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
53533 Tested-by: Gaetan Nadon <memsize@videotron.ca>
53534 Signed-off-by: Keith Packard <keithp@keithp.com>
53535
53536commit e2e2747f40c4b9e23dba38ea656d4c082e83a794
53537Author: Matthias Hopf <mhopf@suse.de>
53538Date: Fri May 7 18:46:30 2010 +0200
53539
53540 Fix mod() definition.
53541
53542 Presumably no implications, especially security-wise.
53543
53544 Signed-off-by: Matthias Hopf <mhopf@suse.de>
53545 Reviewed-by: Matt Turner <mattst88@gmail.com>
53546 Signed-off-by: Keith Packard <keithp@keithp.com>
53547
53548commit 626f97688a89dbc5da80390d822d21b1a737de0e
53549Author: Adam Tkac <atkac@redhat.com>
53550Date: Fri May 7 17:02:03 2010 +1000
53551
53552 Export XkbCopyDeviceKeymap from Xorg, it is needed by VNC.
53553
53554 Signed-off-by: Adam Tkac <atkac@redhat.com>
53555 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53556 Signed-off-by: Keith Packard <keithp@keithp.com>
53557
53558commit 21ed660f30a3f96c787ab00a16499e0fb034b2ad
53559Author: Chris Humbert <freedesktop@mahadri.com>
53560Date: Fri May 7 17:02:43 2010 +1000
53561
53562 dix: make DeviceEvent coordinates signed for Xinerama. #24986
53563
53564 With Xinerama enabled, event coordinates are relative to Screen 0, so
53565 they can be negative. The new DeviceEvent's coordinates are of type
53566 uint16_t, making screens above and to the left of Screen 0 unusable.
53567
53568 X.Org Bug 24986 <https://bugs.freedesktop.org/show_bug.cgi?id=24986>
53569
53570 Signed-off-by: Chris Humbert <freedesktop@mahadri.com>
53571 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53572 Signed-off-by: Keith Packard <keithp@keithp.com>
53573
53574commit 9802839d35aaf788790f1d0e8300db4693a70096
53575Author: Peter Hutterer <peter.hutterer@who-t.net>
53576Date: Fri May 7 16:22:12 2010 +1000
53577
53578 Xi: reset the known properties at the end of the server generation.
53579
53580 Properties allocated through XIGetKnownProperty() aren't reset on the second
53581 server generation but keep the old value. As a result, wrong Atoms are
53582 supplied to the driver, resulting in potential data corruption or weird
53583 error message.
53584
53585 Reproducible by running "xlsatom | grep FLOAT" twice on a plain X server.
53586 The second X server generation won't have the FLOAT atom defined anymore,
53587 despite the users of this atom not noticing any errors.
53588
53589 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
53590 Reviewed-by: Julien Cristau <jcristau@debian.org>
53591 Signed-off-by: Keith Packard <keithp@keithp.com>
53592
53593commit 72758287f79a4f1aa8fa388f20947042e3e14693
53594Author: Julien Cristau <jcristau@debian.org>
53595Date: Tue May 4 15:15:02 2010 +0200
53596
53597 XQuartz: add new localization files to EXTRA_DIST
53598
53599 commit 206531f75cd41c034e89fdfbc75ab0910682eef8 added localization files
53600 for ar, add them to the Makefile.
53601
53602 Signed-off-by: Julien Cristau <jcristau@debian.org>
53603 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
53604
53605commit a3f5d30ba61e70b3de8b48754dea32715ba93c40
53606Author: Keith Packard <keithp@keithp.com>
53607Date: Fri Apr 30 16:10:32 2010 -0700
53608
53609 Make sure XFixes invisible cursor gets freed on server reset
53610
53611 This uses the same hack that dix uses for the rootCursor -- allocate
53612 a resource ID for the invisible cursor so that it gets freed at reset
53613 time. This also allows us to unconditionally create it during
53614 extension initialization; necessary as the privates layout may well be
53615 different on subsequent generations.
53616
53617 Reviewed-by: Adam Jackson <ajax@redhat.com>
53618 Signed-off-by: Keith Packard <keithp@keithp.com>
53619
53620commit cdeb2c23f80ced961fe205125547e16905885534
53621Author: Keith Packard <keithp@keithp.com>
53622Date: Fri Apr 30 12:18:09 2010 -0700
53623
53624 Fix cursor ref counting mistakes with sprites and xf86Cursor.c
53625
53626 A few cursor value assignments weren't getting correctly ref counted,
53627 causing leaks of cursor objects.
53628
53629 Reviewed-by: Adam Jackson <ajax@redhat.com>
53630 Signed-off-by: Keith Packard <keithp@keithp.com>
53631
53632commit a80b1f888f2f7674a715d512ab950eeadee20448
53633Merge: 4971099 0efd7b8
53634Author: Keith Packard <keithp@keithp.com>
53635Date: Sun May 2 16:35:17 2010 -0700
53636
53637 Merge remote branch 'jeremyhu/master'
53638
53639commit 0efd7b8d82a48793341e27d4c224986e0971687c
53640Author: Jeremy Huddleston <jeremyhu@apple.com>
53641Date: Tue Apr 27 13:14:47 2010 -0700
53642
53643 XQuartz GLX: Change around includes for better compatability with different OS versions
53644
53645 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
53646
53647commit 7b506fdc840aebed6b5acb91437a2cb620b5bddc
53648Author: Jeremy Huddleston <jeremyhu@apple.com>
53649Date: Wed Apr 21 08:38:53 2010 -0700
53650
53651 rootless: Add some sanity checking to miPaintWindow
53652
53653 This avoids painting the root window when it isn't actually drawable.
53654
53655 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
53656
53657commit 4d3789c9b3fbe6aad32fcacd964353b612640f27
53658Author: Jeremy Huddleston <jeremyhu@apple.com>
53659Date: Sun Apr 18 01:22:27 2010 -0700
53660
53661 rootless: Remove an unneeded comment
53662
53663 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
53664
53665commit 048d15192af39ba5111882d18a957b6ff4e34cad
53666Author: Oliver McFadden <oliver.mcfadden@nokia.com>
53667Date: Thu Apr 22 10:29:44 2010 +0300
53668
53669 x-list.c: null-returning function malloc() was called without checking
53670
53671 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
53672 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
53673
53674commit 4971099860e8e6d7817ada92e2636cb435fd002d
53675Author: Tiago Vignatti <tiago.vignatti@nokia.com>
53676Date: Sun May 2 19:14:39 2010 +0300
53677
53678 dix: wrap variables with #ifdef when panoramix is not used
53679
53680 this shut up some warnings.
53681
53682 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53683 Reviewed-by: Jamey Sharp <jamey@minilop.net>
53684 Signed-off-by: Keith Packard <keithp@keithp.com>
53685
53686commit 40858960c02a7694b0e1b72e2a986805580ff6f1
53687Author: Keith Packard <keithp@keithp.com>
53688Date: Thu Apr 29 12:24:51 2010 -0700
53689
53690 Delete loop with no effect from GlyphUninit
53691
53692 Signed-off-by: Keith Packard <keithp@keithp.com>
53693 Reviewed-by: Jamey Sharp <jamey@minilop.net>
53694
53695commit 65e961fcc1da0f1e6c1c477701b2eeb36955ed24
53696Author: Keith Packard <keithp@keithp.com>
53697Date: Thu Apr 29 12:24:49 2010 -0700
53698
53699 Replace some input devPrivates with regular struct fields
53700
53701 In the process, fixes a memory leak in CloseDevice, and an unchecked
53702 memory allocation in InitializePredictableAccelerationProperties.
53703
53704 Signed-off-by: Keith Packard <keithp@keithp.com>
53705 Reviewed-by: Jamey Sharp <jamey@minilop.net>
53706
53707commit 998e982b775073da59412dce8ce609ea2ad7967c
53708Author: Keith Packard <keithp@keithp.com>
53709Date: Thu Apr 29 12:24:48 2010 -0700
53710
53711 Move callback definitions to their own file
53712
53713 Needed as the privates stuff uses the callbacks before they would have
53714 been defined in dix.h
53715
53716 Signed-off-by: Keith Packard <keithp@keithp.com>
53717 Reviewed-by: Jamey Sharp <jamey@minilop.net>
53718
53719commit 7dde50c649089b03650d6497079a690feea65e22
53720Author: Keith Packard <keithp@keithp.com>
53721Date: Thu Apr 29 12:24:47 2010 -0700
53722
53723 Move X_DEPRECATED to misc.h so it can be used outside resource.h
53724
53725 Signed-off-by: Keith Packard <keithp@keithp.com>
53726 Reviewed-by: Jamey Sharp <jamey@minilop.net>
53727
53728commit 5623908aeef70e5083f3b49986c7547ed044fedd
53729Author: Aaron Plattner <aplattner@nvidia.com>
53730Date: Wed Apr 28 12:37:08 2010 -0700
53731
53732 dix: Improve documentation of the DIX private data functions.
53733
53734 The functions exported by the devPrivates code were poorly documented. I tried
53735 to spruce it up a little.
53736
53737 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
53738 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53739 Signed-off-by: Keith Packard <keithp@keithp.com>
53740
53741commit 986d46144b183a36b4e98aed95eca0c55a8b4251
53742Merge: a974c8e b5b8f91
53743Author: Keith Packard <keithp@keithp.com>
53744Date: Fri Apr 30 12:40:53 2010 -0700
53745
53746 Merge remote branch 'jamey/for-keith'
53747
53748 Conflicts:
53749 hw/xfree86/common/xf86xv.c
53750
53751commit a974c8e7cba40c8d7d1b91e07de8c414627b71a2
53752Merge: 6581bc4 02e8622
53753Author: Keith Packard <keithp@keithp.com>
53754Date: Fri Apr 30 12:33:00 2010 -0700
53755
53756 Merge remote branch 'whot/for-keith'
53757
53758commit 6581bc4591746c906d97f8b868f946c47bc6d756
53759Merge: 0e91e19 d073e51
53760Author: Keith Packard <keithp@keithp.com>
53761Date: Fri Apr 30 12:29:52 2010 -0700
53762
53763 Merge remote branch 'yselkowitz/master'
53764
53765commit 0e91e19f783d21198fc2e210203e0f10040a9cd0
53766Merge: 81a081c edbc56c
53767Author: Keith Packard <keithp@keithp.com>
53768Date: Fri Apr 30 12:27:51 2010 -0700
53769
53770 Merge remote branch 'vignatti/for-keith'
53771
53772commit 02e86221b851e5423a95782aa7e297ea051022ca
53773Author: Peter Hutterer <peter.hutterer@who-t.net>
53774Date: Thu Apr 29 09:48:37 2010 +1000
53775
53776 xfree86: a missing input driver is not an error.
53777
53778 We call NIDR on all devices that make it through the config backend.
53779 Including some that have no driver assigned to them (/dev/input/mouse0 for
53780 example). Those ones then simply get ignored by NIDR, but this should not be
53781 noted as an error in the log file.
53782
53783 X_INFO is sufficient, and it may just prevent some bugreports.
53784
53785 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
53786 Acked-by: Julien Cristau <jcristau@debian.org>
53787 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
53788
53789commit 81a081c1f0cb55df94cb10495aa7ad71cd5a9afb
53790Author: Jon TURNEY <jon.turney@dronecode.org.uk>
53791Date: Sat Nov 7 19:28:18 2009 +0000
53792
53793 Cygwin/X: AIGLX using native WGL
53794
53795 A rewrite of the XWin DDX AIGLX code to actually make it do something useful again
53796
53797 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
53798 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
53799
53800commit 262c2c9031ea60cbe738ed80ddb7faa1b888d137
53801Author: Rami Ylimäki <ext-rami.ylimaki@nokia.com>
53802Date: Tue Apr 27 14:04:47 2010 +0300
53803
53804 test: Fix linking of wrapped functions in XI2 tests.
53805
53806 Running "make check" will lead to build problems in
53807 scratchbox. Building the first test that wraps dixLookupWindow fails
53808 because symbol __real_dixLookupWindow can't be resolved. Defining
53809 wrapping options as linker options instead of compiler options makes
53810 everything build nicely in scratchbox.
53811
53812 Signed-off-by: Rami Ylimäki <ext-rami.ylimaki@nokia.com>
53813 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
53814 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
53815
53816commit ffaae7c0c69a51a53a76146c79f3630ae197a443
53817Author: Adam Jackson <ajax@redhat.com>
53818Date: Mon Apr 26 20:11:31 2010 -0700
53819
53820 Remove mibank support
53821
53822 Banked framebuffers are so 1990. As of 7.4 the only drivers remaining
53823 that used this were chips, neomagic, trident, and vesa. vesa only used
53824 it when not using shadowfb, which is broadly undesirable anyway, and no
53825 longer uses it at all as of 2.3.0. neomagic never used it by default,
53826 and support for it is gone in git master. The other two effectively
53827 only ever used it for ISA chips; since ISA support is now gone from
53828 the server, they have been modified to only compile mibank support when
53829 ISA support is available.
53830
53831 Signed-off-by: Adam Jackson <ajax@redhat.com>
53832 Reviewed-by: Keith Packard <keithp@keithp.com>
53833 Signed-off-by: Keith Packard <keithp@keithp.com>
53834
53835commit b5b8f91b82d7b150c926dd3fecee6c3aafff6e39
53836Author: Jamey Sharp <jamey@minilop.net>
53837Date: Thu Apr 22 21:35:17 2010 -0700
53838
53839 xfree86: use screen privates for Xv offscreen images.
53840
53841 This replaces a globally-allocated array that depended on MAXSCREENS.
53842
53843 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53844 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53845 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
53846
53847commit a1c2acfe798c57e5be7e5f6c111a6ce91400487a
53848Author: Jamey Sharp <jamey@minilop.net>
53849Date: Wed Apr 21 18:05:45 2010 -0700
53850
53851 xfree86: use screen privates for exclusive DGA clients.
53852
53853 Most DGA requests allow at most one client to be using DGA on each
53854 screen. Instead of keeping track of the current client in a
53855 MAXSCREEN-sized array, track it in a per-screen private.
53856
53857 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53858 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53859 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
53860
53861commit f9e3a2955d2ca73604c68fc9d51405581b832edb
53862Author: Jamey Sharp <jamey@minilop.net>
53863Date: Mon Apr 26 18:23:27 2010 -0700
53864
53865 Make Xvfb independent of MAXSCREENS.
53866
53867 If a -screen option specifies a screen number higher than any previously
53868 specified, reallocate the vfb-private array of screen-info structs.
53869
53870 If built with a DIX that still has a MAXSCREENS limit, asking for too
53871 many screens won't be detected until InitOutput calls AddScreen.
53872
53873 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53874 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53875
53876commit 20e84b0b44e8b3b40a3ecab5b2e64a27de247b16
53877Author: Jamey Sharp <jamey@minilop.net>
53878Date: Mon Apr 26 18:09:23 2010 -0700
53879
53880 Xvfb: Simplify screen option processing.
53881
53882 Inspired by Jon Turney's "Xwin: Simplify screen option processing"
53883 patch, which does something similar for the Xwin server.
53884
53885 Besides making the code more readable, this eliminates most of Xvfb's
53886 references to MAXSCREENS.
53887
53888 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53889 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53890
53891commit eeb84547556b943af2acff207e034823205c7dfe
53892Author: Jamey Sharp <jamey@minilop.net>
53893Date: Mon Apr 26 18:04:25 2010 -0700
53894
53895 Delete redundant scrnum field from Xvfb private screen-info struct.
53896
53897 The screen number can be inferred from the position in the vfbScreens
53898 array, and it was only used in two places, so it was hardly important.
53899
53900 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53901 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53902
53903commit da0217891904bc48d5f0b7ea5c62c8ea0e9b95f9
53904Author: Jamey Sharp <jamey@minilop.net>
53905Date: Wed Apr 21 22:26:28 2010 -0700
53906
53907 Track screens' installed colormaps as screen privates.
53908
53909 Several DDXes allow each screen to have at most one (or in some cases,
53910 exactly one) installed colormap. These all use the same pattern: Declare
53911 a global-lifetime array of MAXSCREENS ColormapPtrs, and index it by
53912 screen number. This patch converts most of those to use screen privates
53913 instead.
53914
53915 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53916 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
53917
53918commit 22c4300ee25a20e1f815e46225bf0de9cfd6748f
53919Author: Jamey Sharp <jamey@minilop.net>
53920Date: Wed Oct 7 09:00:02 2009 -0700
53921
53922 Simplify XineramifyXv.
53923
53924 The structure of the adaptor-matching algorithm was obscured by trying
53925 to use "continue" and "break" as the only control-flow primitives. It's
53926 a lot more clear if you add "return" to that set.
53927
53928 Signed-off-by: Jamey Sharp <jamey@minilop.net>
53929 Reviewed-by: Adam Jackson <ajax@redhat.com>
53930
53931commit af170a4ab76d33254dd4e54a33bb410fa1624739
53932Author: Peter Hutterer <peter.hutterer@who-t.net>
53933Date: Wed Apr 21 15:51:20 2010 +1000
53934
53935 Xext: only update the sprite on pointer events.
53936
53937 A call to miPointerUpdateSprite for the XTEST keyboard may result in a
53938 NULL pointer dereference in miDCPutUpCursor() when the save buffer is NULL.
53939
53940 XTS test case: Xlib 11 KeymapNotify
53941
53942 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
53943
53944commit bac1c5f1be4588b2b1eb646ee98a5442e1b767d5
53945Author: Dirk Wallenstein <halsmit@t-online.de>
53946Date: Sat Apr 17 21:36:23 2010 +0200
53947
53948 xkb: Fix omissions in geometry initialization #27679
53949
53950 _XkbCopyGeom did not copy all of the data from the source geometry. This
53951 resulted in failures when trying to obtain the keymap from a server
53952 where the default geometry has not been replaced by a custom
53953 configuration.
53954
53955 Signed-off-by: Dirk Wallenstein <halsmit@t-online.de>
53956 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
53957 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
53958
53959commit 3f63db89191701139d45332ca80b9d6eb327c773
53960Merge: 41bdb6c 54e51de
53961Author: Keith Packard <keithp@keithp.com>
53962Date: Mon Apr 26 20:18:13 2010 -0700
53963
53964 Merge remote branch 'jamey/for-keith'
53965
53966commit 41bdb6c003cca3ef0ff88d9c7de318115bab1ba2
53967Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
53968Date: Wed Apr 21 18:11:05 2010 -0700
53969
53970 xf86: Don't crash when switching modes through RandR without owning the VT.
53971
53972 While VT-switched, FB access is disabled and should remain so. Trying to switch
53973 modes in that state would re-enable it, potentially causing crashes if trying
53974 to access it before the driver has recovered from the mode switch.
53975
53976 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
53977 Reviewed-by: Adam Jackson <ajax@redhat.com>
53978 Signed-off-by: Keith Packard <keithp@keithp.com>
53979
53980commit 2a763c9023b8bc978b32eaa56d2c625b6f2badee
53981Author: Adam Jackson <ajax@redhat.com>
53982Date: Mon Feb 15 16:54:35 2010 -0500
53983
53984 dix: Default DPMS timeout values to match screensaver values
53985
53986 These have the same default, but if you specify something different with
53987 -s on the command line, only the screensaver time is changed. As DPMS
53988 is usually what's desired, change it to match.
53989
53990 Signed-off-by: Adam Jackson <ajax@redhat.com>
53991 Reviewed-by: Keith Packard <keithp@keithp.com>
53992 Signed-off-by: Keith Packard <keithp@keithp.com>
53993
53994commit 518f3b189b6c8aa28b62837d14309fd06163ccbb
53995Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
53996Date: Wed Apr 21 16:46:17 2010 -0700
53997
53998 mi: don't thrash resources when displaying the software cursor across screens
53999
54000 This changes the DC layer to maintain a persistent set of GCs/pixmaps/pictures
54001 for each pScreen instead of failing to thrash between them when changing
54002 screens.
54003
54004 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
54005 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54006 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54007 Tested-by: Peter Hutterer <peter.hutterer@who-t.net>
54008
54009commit 54e51de8cd950b3f5b30e72cf1d604f54f619f0e
54010Author: Jamey Sharp <jamey@minilop.net>
54011Date: Sat Apr 24 23:07:47 2010 -0700
54012
54013 ProcGrabButton: remove redundant error check.
54014
54015 If dixLookupResourceByType did not return Success, it will have set the
54016 pointer to NULL, so the second if will always be true.
54017
54018 Signed-off-by: Jamey Sharp <jamey@minilop.net>
54019 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
54020
54021commit 7e573b5ccd2c759f74ebb84d3a5a23da6b88ec6d
54022Author: Jamey Sharp <jamey@minilop.net>
54023Date: Mon Apr 26 15:05:06 2010 -0700
54024
54025 Kill unused/useless macros in panoramiX.h: BREAK_IF, IF_RETURN, FORCE_ROOT
54026
54027 This header doesn't seem to be used outside the server tree, so I don't
54028 think anything was depending on these except what I've fixed in this
54029 patch.
54030
54031 FORCE_ROOT is not used anywhere, and never has been in all the history
54032 that's in git. ajax removed its FORCE_WIN and FORCE_CMAP companions
54033 three years ago, in 04b87d6dfae02e4ecdb5216d12c6cdafd1e8c2b4.
54034
54035 BREAK_IF and IF_RETURN expanded to exactly the keywords in their names,
54036 which is silly. They were rarely used and served only to make callers
54037 inconsistent with the idioms used elsewhere in the server.
54038
54039 Signed-off-by: Jamey Sharp <jamey@minilop.net>
54040 Reviewed-by: Matt Turner <mattst88@gmail.com>
54041
54042commit e372cc58e02391c3e8e82ac03f73fbe8b1f7d49c
54043Author: Jamey Sharp <jamey@minilop.net>
54044Date: Wed Oct 7 11:06:34 2009 -0700
54045
54046 Boolean-or short-circuits, so (a || (!a && ...)) is redundant. Simplify.
54047
54048 Signed-off-by: Jamey Sharp <jamey@minilop.net>
54049 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54050 Reviewed-by: Matt Turner <mattst88@gmail.com>
54051
54052commit 25d1c6cd37cc4079adf7c4bad8d10678fb08b409
54053Author: Keith Packard <keithp@keithp.com>
54054Date: Mon Apr 26 11:04:38 2010 -0700
54055
54056 Exa screen private has SavedSourceValidate, not SourceValidate
54057
54058 Need to check the right field to see if there is a wrapped
54059 SourceValidate function.
54060
54061 Signed-off-by: Keith Packard <keithp@keithp.com>
54062
54063commit d073e51beeed2e702fa5c7534b1345360bc302f2
54064Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54065Date: Mon Apr 26 11:39:19 2010 -0500
54066
54067 Clarify help strings for disabled features in configure
54068
54069 Features which are disabled by default use the --enable-* syntax to show
54070 how they need to be passed to override the default.
54071
54072 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54073 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
54074 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54075 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
54076
54077commit 57409af1267f48457f93134922f2450518182e09
54078Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54079Date: Wed Apr 14 13:18:28 2010 -0500
54080
54081 Catch when requested SHA1 implementation is missing
54082
54083 The other SHA1 implementation detections already error out if
54084 specifically requested but were not found.
54085
54086 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54087 Reviewed-by: Julien Cristau <jcristau@debian.org>
54088
54089commit 0d8fe2fe5c37659147cbecac793183c62fc6818a
54090Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54091Date: Tue Apr 13 13:32:51 2010 -0500
54092
54093 Prefer libsha1 over libgcrypt
54094
54095 When no SHA1 implementation is specified, we should first prefer
54096 system-builtin solutions (libc/libmd/CommonCrypto), then smaller
54097 implementations over the larger ones. libsha1 is much smaller than
54098 libgcrypt, so it should be first.
54099
54100 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54101 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
54102 Acked-by: Julien Cristau <jcristau@debian.org>
54103
54104commit 65a2c09549d2372c23b155addff428177708c910
54105Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54106Date: Thu Apr 8 21:55:45 2010 -0500
54107
54108 Revert "Disable Record by default."
54109
54110 Record was broken during the pre-1.6 development cycle and was not fixed
54111 until 1.7.6. Now that it is fixed, re-enable it by default.
54112
54113 This reverts commit 3eaecdd66e791e0f3d86b23ce10be057ca44c044.
54114
54115 Conflicts:
54116
54117 configure.ac
54118
54119 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54120 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54121
54122commit 6e0d847c273ab89323883a05f806f449180e3669
54123Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54124Date: Thu Mar 25 03:37:09 2010 -0500
54125
54126 xfree86: Fix reference to SGML entities
54127
54128 XORG_CHECK_LINUXDOC sets the SGML search path to the parent of X11/defs.ent.
54129
54130 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54131 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
54132 Tested-by: Gaetan Nadon <memsize@videotron.ca>
54133
54134commit 04b7d529c09af2c842cf020d5b26080e1ffd88d6
54135Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54136Date: Thu Mar 25 03:36:25 2010 -0500
54137
54138 xfree86: Ignore linuxdoc generated docs
54139
54140 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54141 Acked-by: Gaetan Nadon <memsize@videotron.ca>
54142
54143commit 39e5108c78e0802833a8e9ea7f29d042738625ba
54144Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54145Date: Thu Mar 25 03:35:36 2010 -0500
54146
54147 dmx: use silent rules in doc
54148
54149 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54150 Acked-by: Gaetan Nadon <memsize@videotron.ca>
54151
54152commit 3ba2ce5d1054c99354686aeb373247f1b9debf79
54153Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54154Date: Thu Mar 25 03:33:55 2010 -0500
54155
54156 dmx: Ignore linuxdoc generated docs
54157
54158 dmx.txt and scaled.txt are generated from SGML, so they probably never
54159 should have been in version control in the first place.
54160
54161 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54162 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
54163 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
54164
54165commit 40972576b606237d5d42abc13d846163e264a4ac
54166Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54167Date: Tue Mar 23 04:47:27 2010 -0500
54168
54169 Use AC_PROG_SED and respect its result
54170
54171 AC_PROG_SED sets SED as the path to a fully-functional 'sed' (which may
54172 also be called 'gsed' if GNU sed is installed alongside a proprietary
54173 version).
54174
54175 This is a follow up to commit 9be4157391edf0c5fc4ee36adfb1eb1c3bdb8e3b.
54176
54177 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
54178 Reviewed-by: Rémi Cardona <remi@gentoo.org>
54179
54180commit b557571dbff3d9a615e03cad904d46b8af9153a0
54181Author: Jerome Glisse <jglisse@redhat.com>
54182Date: Mon Apr 26 16:01:48 2010 +0200
54183
54184 exa: avoid calling non existent SourceValidate callback
54185
54186 In unaccel path we were unconditionaly calling the SourceValidate
54187 callback but in some case it could be NULL. Check if we have a
54188 valid callback before calling it.
54189
54190 Signed-off-by: Jerome Glisse <jglisse@redhat.com>
54191 Acked-by: Michel Dänzer <michel@daenzer.net>
54192 Signed-off-by: Keith Packard <keithp@keithp.com>
54193
54194commit 0e6e75b04eaca311962390533acc03d0ac5ece83
54195Author: Keith Packard <keithp@keithp.com>
54196Date: Mon Apr 26 09:14:22 2010 -0700
54197
54198 Revert "exa: avoid calling non existent SourceValidate callback"
54199
54200 This reverts commit daa6f9308fa57e41210f78d92dbdb3f1c8eae6ad.
54201
54202 This one checked pScreen->SourceValidate instead of
54203 pExaScr->SourceValidate which would cause a segfault when
54204 pExaScr->SourceValidate was NULL.
54205
54206commit daa6f9308fa57e41210f78d92dbdb3f1c8eae6ad
54207Author: Jerome Glisse <jglisse@redhat.com>
54208Date: Mon Apr 26 15:55:26 2010 +0200
54209
54210 exa: avoid calling non existent SourceValidate callback
54211
54212 In unaccel path we were unconditionaly calling the SourceValidate
54213 callback but in some case it could be NULL. Check if we have a
54214 valid callback before calling it.
54215
54216 Signed-off-by: Jerome Glisse <jglisse@redhat.com>
54217 Acked-by: Michel Dänzer <michel@daenzer.net>
54218 Signed-off-by: Keith Packard <keithp@keithp.com>
54219
54220commit edbc56c088462844f6445f5e6d7cbcfc58a207dc
54221Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54222Date: Mon Apr 26 14:49:25 2010 +0300
54223
54224 include: remove couple of unused structures fields and bump ABI
54225
54226 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54227
54228commit 4f8dc1109dcdfa973466a038c16c07da5f6c16b4
54229Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54230Date: Thu Apr 22 15:43:11 2010 +0300
54231
54232 dix and others: remove unused arraySize field from ScreenInfo
54233
54234 Bizarre. This seems to never be used before. I left the field in ScreenInfo,
54235 with another name. So, stop looking at it.
54236
54237 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54238 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
54239
54240commit 64fd39f2f0ffbbdc2d5929047d260263d0962141
54241Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54242Date: Thu Apr 22 15:48:48 2010 +0300
54243
54244 xfree86: no need to assign numScreens again
54245
54246 numScreens is always being assigned to 0 in dix for any server generation.
54247
54248 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54249 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54250
54251commit a7d398e545a4be5491248d5ccb303aa03ee1594f
54252Author: Jon TURNEY <jon.turney@dronecode.org.uk>
54253Date: Fri Apr 16 18:13:50 2010 +0100
54254
54255 Xwin: make screens structures run-time adjustable
54256
54257 Change g_ScreenInfo, an array of winScreenInfo elements, from a
54258 static array of MAXSCREENS elements, to a dynamically allocated one
54259
54260 Fix up the validation that -screen option screen numbers are
54261 contiguous from zero (which possibly didn't work correctly before
54262 anyhow)
54263
54264 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
54265 Reviewed-by: Jamey Sharp<jamey@minilop.net>
54266 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54267
54268commit d8454ae488cfc073cd6010c9a08d53855a0c2612
54269Author: Jon TURNEY <jon.turney@dronecode.org.uk>
54270Date: Mon Apr 12 20:18:13 2010 +0100
54271
54272 Xwin: Simplify screen option processing
54273
54274 Use an instance of the XWin DDX-specific screen info structure to hold
54275 the current default values, to simplify greatly the code for applying
54276 options to all screens and remove all those loops over MAXSCREENS screens
54277 in the command line option processing
54278
54279 Use g_iNumScreens for tracking the current initialized screen count
54280
54281 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
54282 Reviewed-by: Jamey Sharp <jamey@minilop.net>
54283 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54284
54285commit b61870595ba4df06006d24ed8c07cfe781bce1b7
54286Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54287Date: Thu Apr 22 20:20:48 2010 +0300
54288
54289 xfree86: track screens' installed colormaps as screen privates
54290
54291 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54292 Reviewed-by: Jamey Sharp <jamey@minilop.net>
54293
54294commit a94f5455c71363d5047668ee093901b9e1645cf1
54295Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54296Date: Thu Apr 22 20:20:43 2010 +0300
54297
54298 mi: track screens' installed colormaps as screen privates
54299
54300 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54301 Reviewed-by: Jamey Sharp <jamey@minilop.net>
54302
54303commit a2c716eaf6b3a4ce75382394636a0a890b5dcfe0
54304Author: Jamey Sharp <jamey@minilop.net>
54305Date: Thu Apr 22 07:34:04 2010 +0200
54306
54307 fb: track screens' installed colormaps as screen privates.
54308
54309 Several DDXes allow each screen to have at most one (or in some cases,
54310 exactly one) installed colormap. These all use the same pattern: Declare
54311 a global-lifetime array of MAXSCREENS ColormapPtrs, and index it by
54312 screen number. This patch converts most of those to use screen privates
54313 instead.
54314
54315 Signed-off-by: Jamey Sharp <jamey@minilop.net>
54316 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54317
54318commit e055bef055b6c726e9f3ef91a83585d13c80651d
54319Author: Jamey Sharp <jamey@minilop.net>
54320Date: Thu Apr 22 06:08:29 2010 +0200
54321
54322 xfree86: use screen privates for exclusive DGA clients
54323
54324 Most DGA requests allow at most one client to be using DGA on each screen.
54325 Instead of keeping track of the current client in a MAXSCREEN-sized array,
54326 track it in a per-screen private.
54327
54328 Signed-off-by: Jamey Sharp <jamey@minilop.net>
54329 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54330
54331commit 7c9733d063acee04735a0e75d62e508fe4ea1163
54332Author: Jamey Sharp <jamey@minilop.net>
54333Date: Fri Apr 23 06:41:00 2010 +0200
54334
54335 xfree86: use screen privates for Xv offscreen images.
54336
54337 This replaces a globally-allocated array that depended on MAXSCREENS.
54338
54339 Signed-off-by: Jamey Sharp <jamey@minilop.net>
54340 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54341
54342commit 35d208125fa0ec78e8d694adc52886b977e789cf
54343Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54344Date: Wed Apr 21 18:28:13 2010 +0300
54345
54346 xfree86: fix not reached code in fi1236 driver from i2c
54347
54348 This issue was introduced in the first dump of the code in 2004. I haven't
54349 check what's the correct fix for it so I simply kept the behaviour of someone
54350 calling this and removed the unreachable code.
54351
54352 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54353 Reviewed-by: Keith Packard <keithp@keithp.com>
54354 Signed-off-by: Keith Packard <keithp@keithp.com>
54355
54356commit 7ac6a6b7d0dfc5e021270b2898accd3425aa008a
54357Merge: 0782894 f7ef6fd
54358Author: Keith Packard <keithp@keithp.com>
54359Date: Wed Apr 21 16:07:38 2010 -0700
54360
54361 Merge remote branch 'vignatti/for-keith'
54362
54363 Conflicts:
54364 exa/exa.c
54365
54366 Signed-off-by: Keith Packard <keithp@keithp.com>
54367
54368commit 0782894b5702adcf6f4a90861793b717f3856fa5
54369Author: Keith Packard <keithp@keithp.com>
54370Date: Wed Apr 21 16:00:28 2010 -0700
54371
54372 Xvfb: Usage message typo fix.
54373
54374 Signed-off-by: Keith Packard <keithp@keithp.com>
54375
54376commit ee3412b8702072c3a0b006bd20dd3bc7071d721c
54377Author: Keith Packard <keithp@keithp.com>
54378Date: Wed Apr 21 15:58:23 2010 -0700
54379
54380 Restore some unused win32 debug code accidentally removed with unifdef
54381
54382 unifdef found some code covered by
54383
54384 which it decided to remove. This patch simply restores that in case
54385 someone wants it back.
54386
54387 Signed-off-by: Keith Packard <keithp@keithp.com>
54388
54389commit e3d2a7a613366c26e5316cf582d9a8c6c6692b0a
54390Author: Peter Hutterer <peter.hutterer@who-t.net>
54391Date: Wed Apr 21 11:47:24 2010 +1000
54392
54393 Revert "mi: don't thrash resources when displaying the software cursor across screens"
54394
54395 This commit leads to a segfault on the very first XTS test case.
54396
54397 Backtrace:
54398 0: /opt/xorg/bin/Xorg (xorg_backtrace+0x3b) [0x80a33db]
54399 1: /opt/xorg/bin/Xorg (0x8048000+0x62a75) [0x80aaa75]
54400 2: (vdso) (__kernel_rt_sigreturn+0x0) [0x5d140c]
54401 3: /lib/libc.so.6 (0x9bb000+0x73579) [0xa2e579]
54402 4: /lib/libc.so.6 (realloc+0xe0) [0xa2e830]
54403 5: /opt/xorg/bin/Xorg (Xrealloc+0x33) [0x80a3f33]
54404 6: /opt/xorg/bin/Xorg (0x8048000+0x1ab79) [0x8062b79]
54405 7: /opt/xorg/bin/Xorg (0x8048000+0x1ac4e) [0x8062c4e]
54406 8: /opt/xorg/bin/Xorg (RegisterExtensionNames+0x2ce) [0x8062fbe]
54407 9: /opt/xorg/bin/Xorg (AddExtension+0x19a) [0x807bd7a]
54408 10: /opt/xorg//lib/xorg/modules/extensions/libextmod.so (0x728000+0x1169a)
54409 [0x73969a]
54410 11: /opt/xorg/bin/Xorg (InitExtensions+0x85) [0x80c0eb5]
54411 12: /opt/xorg/bin/Xorg (0x8048000+0x1a51d) [0x806251d]
54412 13: /lib/libc.so.6 (__libc_start_main+0xe6) [0x9d1bb6]
54413 14: /opt/xorg/bin/Xorg (0x8048000+0x1a2a1) [0x80622a1]
54414 Segmentation fault at address 0x10b2d5f8
54415
54416 valgrind output:
54417 ==5069== Invalid read of size 4
54418 ==5069== at 0x80F928D: FreePicture (picture.c:1531)
54419 ==5069== by 0x818DDEF: miDCDeviceCleanup (midispcur.c:867)
54420 ==5069== by 0x81B97F0: miSpriteDeviceCursorCleanup (misprite.c:968)
54421 ==5069== by 0x80995FA: miPointerDeviceCleanup (mipointer.c:292)
54422 ==5069== by 0x807973E: CloseDevice (devices.c:840)
54423 ==5069== by 0x80799B6: CloseDownDevices (devices.c:933)
54424 ==5069== by 0x8062705: main (main.c:309)
54425 ==5069== Address 0x4cce844 is 12 bytes inside a block of size 84 free'd
54426 ==5069== at 0x40057F6: free (vg_replace_malloc.c:325)
54427 ==5069== by 0x80A3DE0: Xfree (utils.c:1154)
54428 ==5069== by 0x80F9332: FreePicture (picture.c:1576)
54429 ==5069== by 0x80FBB4B: PictureDestroyWindow (picture.c:69)
54430 ==5069== by 0x810B1A3: damageDestroyWindow (damage.c:1840)
54431 ==5069== by 0x80864F1: FreeWindowResources (window.c:846)
54432 ==5069== by 0x8086812: DeleteWindow (window.c:925)
54433 ==5069== by 0x806B53E: FreeClientResources (resource.c:806)
54434 ==5069== by 0x806B60F: FreeAllResources (resource.c:823)
54435 ==5069== by 0x80626E4: main (main.c:299)
54436 ==5069==
54437 ==5069== Invalid write of size 4
54438 ==5069== at 0x80F9295: FreePicture (picture.c:1531)
54439 ==5069== by 0x818DDEF: miDCDeviceCleanup (midispcur.c:867)
54440 ==5069== by 0x81B97F0: miSpriteDeviceCursorCleanup (misprite.c:968)
54441 ==5069== by 0x80995FA: miPointerDeviceCleanup (mipointer.c:292)
54442 ==5069== by 0x807973E: CloseDevice (devices.c:840)
54443 ==5069== by 0x80799B6: CloseDownDevices (devices.c:933)
54444 ==5069== by 0x8062705: main (main.c:309)
54445 ==5069== Address 0x4cce844 is 12 bytes inside a block of size 84 free'd
54446 ==5069== at 0x40057F6: free (vg_replace_malloc.c:325)
54447 ==5069== by 0x80A3DE0: Xfree (utils.c:1154)
54448 ==5069== by 0x80F9332: FreePicture (picture.c:1576)
54449 ==5069== by 0x80FBB4B: PictureDestroyWindow (picture.c:69)
54450 ==5069== by 0x810B1A3: damageDestroyWindow (damage.c:1840)
54451 ==5069== by 0x80864F1: FreeWindowResources (window.c:846)
54452 ==5069== by 0x8086812: DeleteWindow (window.c:925)
54453 ==5069== by 0x806B53E: FreeClientResources (resource.c:806)
54454 ==5069== by 0x806B60F: FreeAllResources (resource.c:823)
54455 ==5069== by 0x80626E4: main (main.c:299)
54456
54457 XTS test case: Xproto pAllocColor
54458
54459 This reverts commit 00b8b7ad61b6f818271fb4d1e383113170309d72.
54460
54461 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54462 Signed-off-by: Keith Packard <keithp@keithp.com>
54463
54464commit f7ef6fd9a1a1f48603df2796fe7c904c25a81077
54465Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54466Date: Tue Apr 20 15:56:58 2010 +0300
54467
54468 xfree86: fix not reached code in tty code
54469
54470 CLEARRTS_SUPPORT cannot be triggered at all. Notice that mouse driver manual
54471 page states the support for it though.
54472
54473 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54474 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54475
54476commit b36eeb713a349961a6be8a6dd183ded6d27f9974
54477Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54478Date: Tue Apr 20 16:26:02 2010 +0300
54479
54480 xkb: check for NULL pointer before dereferences it in XkbWriteXKBSymbols
54481
54482 move srv assignment to before it's being used. Also, check for xkb being nil.
54483
54484 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54485 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54486
54487commit 471f350c99b2e7c5a9045eef5ebff65cd2ddb7c5
54488Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54489Date: Mon Apr 19 20:34:30 2010 +0300
54490
54491 exa: don't need to check for NULL pointer if we already assumed it has a value
54492
54493 the alternative would be to check ps in the beginning of the function.
54494
54495 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54496 Acked-by: Michel Dänzer <michel@daenzer.net>
54497
54498commit 96784f4fcb6c2fb82f6d1abbd28ea1e189e4e6f5
54499Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54500Date: Mon Apr 19 20:46:12 2010 +0300
54501
54502 xkb: check for NULL pointer before dereferences it in XkbAddClientResource
54503
54504 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54505 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54506 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54507
54508commit 057c147541bde6f0cbe22ca069b43a97ddc95baf
54509Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54510Date: Fri Apr 16 18:01:41 2010 +0300
54511
54512 Xi: check for NULL pointer before dereferences it in ListButtonInfo
54513
54514 Both dev and dev->button are already used before their checking were being
54515 performed. So check on the beginning.
54516
54517 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54518 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54519 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54520
54521commit 7f457351d2a09013cd57fcb2f95c0d6f56bfaccb
54522Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54523Date: Mon Apr 19 20:18:51 2010 +0300
54524
54525 xfree86: check for NULL pointer before dereferences it in parser code
54526
54527 Seems to be harmless. Meh.
54528
54529 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54530 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54531 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54532
54533commit d948dcd9555c81ccb378054383ef8da464202bdf
54534Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54535Date: Fri Apr 16 17:08:09 2010 +0300
54536
54537 Xi: fix not reached code in XSendExtensionEvent
54538
54539 Error was introduced in 31a7994a. I.e., broken since 2007. I guess nobody uses
54540 XSendExtensionEvent.
54541
54542 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54543 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54544
54545commit f491b0aa5b86a5242fc4e04218202bbb52d0af39
54546Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54547Date: Fri Apr 16 16:47:33 2010 +0300
54548
54549 xfree86: fix not reached code in parser
54550
54551 ...because Error is a macro that returns NULL.
54552
54553 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54554 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54555
54556commit 099946a3ac94ff5e575e9edcecd26cf9f346b241
54557Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54558Date: Fri Apr 16 18:35:55 2010 +0300
54559
54560 mi: check for NULL pointer before dereferences it in miPointerSetPosition
54561
54562 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54563 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54564 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54565
54566commit 49835eec0c996ad95a01f0fe340336b6b60e51aa
54567Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54568Date: Fri Apr 16 17:50:43 2010 +0300
54569
54570 exa: check for NULL pointer before dereferences it
54571
54572 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54573 Acked-by: Michel Dänzer <michel@daenzer.net>
54574 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54575
54576commit 996c115deb558e8e9490ea773528b8e96650df53
54577Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54578Date: Thu Mar 25 18:17:54 2010 +0200
54579
54580 configure: remove unused builtin font macro from autoconf file
54581
54582 This should be removed together with 49b93df8.
54583
54584 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54585 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54586
54587commit 0ba82562eeba8bf3bcd00b6e3ff28ce5b2c8df3c
54588Author: Tiago Vignatti <tiago.vignatti@nokia.com>
54589Date: Tue Mar 30 19:10:36 2010 +0300
54590
54591 Death to Multibuffer extension
54592
54593 The rationale behind is because no sane application will use this when we have
54594 modern APIs such DRI2. Besides, as a fact, xfree86 server has already
54595 deprecated this extension in 1998:
54596
54597 http://www.xfree86.org/3.3.6/isc7.html
54598
54599 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
54600 Reviewed-by: Keith Packard <keithp@keithp.com>
54601
54602commit 28b7b2b8d02d975480080865f0dddebcaa2f7968
54603Author: Keith Packard <keithp@keithp.com>
54604Date: Mon Apr 19 09:26:10 2010 -0700
54605
54606 unifdef -B -DRENDER to always include RENDER code
54607
54608 This patch was created with:
54609
54610 git ls-files '*.[ch]' | while read f; do unifdef -B -DRENDER -o $f $f; done
54611
54612 Signed-off-by: Keith Packard <keithp@keithp.com>
54613
54614commit b3ab978df861c08298f57529e3db980489055c35
54615Merge: a92b2c2 0ad022a
54616Author: Keith Packard <keithp@keithp.com>
54617Date: Sun Apr 18 22:01:40 2010 -0700
54618
54619 Merge remote branch 'whot/for-keith'
54620
54621commit 0ad022a729bafa56cc7b5d241f567444a34514ad
54622Author: Peter Hutterer <peter.hutterer@who-t.net>
54623Date: Wed Apr 14 10:54:29 2010 +1000
54624
54625 xkb: rename XkbFakeDeviceButton and XkbFakeDeviceMotion, move into xkbActions.c
54626
54627 The name XkbDDXFakeDeviceButton and XkbDDXFakeDeviceMotion is somewhat
54628 misleading, there's no DDX involved in the game at all anymore.
54629
54630 This removes XkbFakeDeviceMotion and XkbFakeDeviceButton from the API where
54631 it arguably shouldn't have been in the first place.
54632
54633 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54634 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
54635 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54636
54637commit da4e2e382828d7ba460766709368ec6214b286dd
54638Author: Peter Hutterer <peter.hutterer@who-t.net>
54639Date: Wed Apr 14 09:48:53 2010 +1000
54640
54641 xkb: purge unneeded includes from ddxDevBtn.c
54642
54643 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54644 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
54645 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
54646
54647commit a92b2c2c8dd1e86ee852168146f01bdf72bfe2d0
54648Author: Kristian Høgsberg <krh@bitplanet.net>
54649Date: Fri Apr 16 05:55:35 2010 -0400
54650
54651 glx: Drop DestroyWindow hook
54652
54653 Now that glx doesn't call DRI2DestroyDrawable anymore, we don't need to
54654 force a specific resource destruction order in the DestroyWindow hook.
54655
54656 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
54657 Reviewed-by: Michel Dänzer <michel@daenzer.net>
54658
54659 https://bugs.freedesktop.org/show_bug.cgi?id=26394
54660 Signed-off-by: Keith Packard <keithp@keithp.com>
54661
54662commit 1da1f33f2dd5b437dd56cd9f5d6782de4ad5a1bc
54663Author: Kristian Høgsberg <krh@bitplanet.net>
54664Date: Fri Apr 16 05:55:34 2010 -0400
54665
54666 DRI2: Track DRI2 drawables as resources, not privates
54667
54668 The main motivation here is to have the resource system clean up the
54669 DRI2 drawable automatically so glx doesn't have to. Right now, the
54670 glx drawable resource must be destroyed before the X drawable, so that
54671 calling DRI2DestroyDrawable doesn't crash. By making the DRI2
54672 drawable a resource, GLX doesn't have to worry about that and the
54673 resource destruction order becomes irrelevant.
54674
54675 https://bugs.freedesktop.org/show_bug.cgi?id=26394
54676
54677 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
54678 Signed-off-by: Keith Packard <keithp@keithp.com>
54679
54680commit 22da7aa9d743deee198aaf6df5d370a446db9763
54681Author: Kristian Høgsberg <krh@bitplanet.net>
54682Date: Fri Apr 16 05:55:33 2010 -0400
54683
54684 glx: Let the resource system destroy pixmaps
54685
54686 GLX pbuffers are implemented using a pixmap allocated by the server.
54687 With the change to DRI2 to track DRI2 drawables as resources, we need to make
54688 sure that every drawable we create a DRI2 drawable for has an XID. By
54689 using the XID of the pbuffer, the resource system will automatically
54690 reclaim the hidden pixmap and the DRI2 drawable when the pbuffer is
54691 destroyed or the client exits.
54692
54693 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
54694 Signed-off-by: Keith Packard <keithp@keithp.com>
54695
54696commit f0006aa58f6cf7552a239e169ff6e7e4fda532f4
54697Author: Kristian Høgsberg <krh@bitplanet.net>
54698Date: Fri Apr 16 05:55:32 2010 -0400
54699
54700 glx: Track GLX 1.3 style GLX drawables under their X drawable ID as well
54701
54702 This ensures that the DrawableGone callback gets called as necessary
54703 when the X drawable goes away. Otherwise, using a GLX drawable
54704 (say, glXSwapBuffers) in indirect mode after the X drawable has been
54705 destroyed will crash the server.
54706
54707 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
54708 Reviewed-by: Michel Dänzer <michel@daenzer.net>
54709 Signed-off-by: Keith Packard <keithp@keithp.com>
54710
54711commit f4106c02318fcc4b534224df5b95a58aff555fb4
54712Author: Peter Hutterer <peter.hutterer@who-t.net>
54713Date: Tue Apr 13 14:44:59 2010 +1000
54714
54715 xkb: use GPE for XKB fake motion events.
54716
54717 Section 4.6.1 of the XKB spec says that "the initial event always moves the
54718 cursor the distance specified in the action [...]", so skip the
54719 POINTER_ACCELERATE flag for GPE, it would cause double-acceleration.
54720
54721 Potential regression - GPE expects the coordinates to be either relative or
54722 both. XKB in theory allows for x to be relative and y to be absolute (or
54723 vice versa). Let's pretend that scenario has no users.
54724
54725 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54726 Reviewed-by: Simon Thum <simon.thum@gmx.de>
54727 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
54728
54729commit 6c42c8c356be305dc7f3f92ad8d58675da8c2f07
54730Author: Peter Hutterer <peter.hutterer@who-t.net>
54731Date: Wed Apr 14 10:51:41 2010 +1000
54732
54733 xkb: Guard against SIGIO updates during PointerKeys.
54734
54735 In theory, an event coming in during GPE could reset our lastSlave, leading
54736 to rather interesting events lateron.
54737
54738 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54739 Reviewed-by: Simon Thum <simon.thum@gmx.de>
54740 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
54741
54742commit 108457dff816569453a2d88cd72595fa7eb02479
54743Author: Peter Hutterer <peter.hutterer@who-t.net>
54744Date: Tue Apr 13 14:41:07 2010 +1000
54745
54746 xkb: Post PointerKeys through the XTEST device.
54747
54748 Posting an event through a master device may cause pointer jumps once
54749 lastSlave == master, caused by double scaling. To avoid this, post the fake
54750 event generated by XKB through the XTEST device instead.
54751
54752 Fedora bug #560356 <https://bugzilla.redhat.com/560356>
54753 Tested-by: Andrew McNabb
54754
54755 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54756 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
54757
54758commit a780e5b3638a0ff81301fc68aca15b47ba0befb7
54759Author: Benjamin Tissoires <tissoire@cena.fr>
54760Date: Wed Apr 14 17:27:51 2010 +0200
54761
54762 xf86ScaleAxis: support for high resolution devices
54763
54764 High resolution devices was generating integer overflow.
54765 For instance the wacom Cintiq 21UX has an axis value up to
54766 87000. Thus the term (dSx * (Cx - Rxlow)) is greater than
54767 MAX_INT32.
54768
54769 Using 64bits integer avoids such problem.
54770
54771 Signed-off-by: Philippe Ribet <ribet@cena.fr>
54772 Signed-off-by: Benjamin Tissoires <tissoire@cena.fr>
54773 Reviewed-by: Keith Packard <keithp@keithp.com>
54774 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54775
54776commit e424d5812300e82de375d83dc0b490a76d865016
54777Merge: 5b0faf3 0e7703f
54778Author: Keith Packard <keithp@keithp.com>
54779Date: Thu Apr 15 15:01:34 2010 -0700
54780
54781 Merge remote branch 'whot/for-keith'
54782
54783commit 0e7703f9b1927328954a2fc87aac6be244819329
54784Author: Julien Cristau <jcristau@debian.org>
54785Date: Tue Apr 13 22:28:36 2010 +0200
54786
54787 vfb: add a name and type to the pointer and keyboard
54788
54789 Fixes a crash in XIQueryDevice which calls strlen on a NULL pointer.
54790
54791 #0 strlen () at ../sysdeps/x86_64/strlen.S:31
54792 #1 0x00000000004c16ed in SizeDeviceInfo (dev=0x969bd0)
54793 at ../../Xi/xiquerydevice.c:204
54794 #2 0x00000000004c1a01 in ProcXIQueryDevice (client=0xa57510)
54795 at ../../Xi/xiquerydevice.c:98
54796
54797 Debian bug#575905 <http://bugs.debian.org/575905>
54798
54799 Reported-by: "Bernhard R. Link" <brlink@debian.org>
54800 Signed-off-by: Julien Cristau <jcristau@debian.org>
54801 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54802 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54803
54804commit 2f29b163bbdebe384c24f781bb97b446959e8f4c
54805Author: Adam Tkac <atkac@redhat.com>
54806Date: Mon Mar 22 14:52:29 2010 +0100
54807
54808 dix: Export AllocDevicePair GetPointerEvents, GetKeyboardEvents and generate_modkeymap functions from Xorg.
54809
54810 Those functions are used by TigerVNC libvnc.so module which doesn't
54811 use standard XInput infrastructure but uses same functions like,
54812 for example, XTest devices.
54813
54814 Signed-off-by: Adam Tkac <atkac@redhat.com>
54815 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54816 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54817
54818commit 241b53b77750b5eea6759e79b23be4ff270a3d1f
54819Author: Peter Hutterer <peter.hutterer@who-t.net>
54820Date: Tue Mar 23 14:35:56 2010 +1000
54821
54822 mi: remove log-spamming bogus error message (#26843)
54823
54824 miSpriteRealizeCursor is called whenever a device is set floating and it's
54825 fine to do so, no need for an error message.
54826 Same goes for the other miSprite messages.
54827
54828 X.Org Bug 26843 <http://bugs.freedesktop.org/show_bug.cgi?id=26843>
54829
54830 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54831
54832commit 00b8b7ad61b6f818271fb4d1e383113170309d72
54833Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
54834Date: Wed Apr 7 13:52:47 2010 -0700
54835
54836 mi: don't thrash resources when displaying the software cursor across screens
54837
54838 This changes the DC layer to maintain a persistent set of GCs/pixmaps/pictures
54839 for each pScreen instead of failing to thrash between them when changing
54840 screens.
54841
54842 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
54843 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54844 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54845
54846commit 5f31e2196179f8db3170d65a17d8ad40da1acb0d
54847Author: Tim Yamin <plasm@roo.me.uk>
54848Date: Mon Mar 8 12:45:15 2010 +1000
54849
54850 dix: fix cursor screen check for xinerama setups.
54851
54852 The de-duplication of CheckPhysLimits 942eae6868b8b0f343b6a added a
54853 condition that is invalid for a Xinerama setup. pScreen is invalid for the
54854 Xinerama case, so comparing it to anything is a bad idea.
54855
54856 Signed-off-by: Tim Yamin <plasm@roo.me.uk>
54857 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54858 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54859
54860commit 5b0faf355465c9f21ad96e0ed266fbdbc29efb5b
54861Author: Jeremy Huddleston <jeremyhu@apple.com>
54862Date: Thu Apr 15 10:55:53 2010 -0700
54863
54864 XQuartz GLX: Don't let garbage enter our pixel request
54865
54866 https://bugs.freedesktop.org/show_bug.cgi?id=27654
54867
54868 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
54869
54870commit fe7778e58e099d353689755ed2f5aa440569ebe3
54871Author: Jeremy Huddleston <jeremyhu@apple.com>
54872Date: Wed Apr 14 09:06:04 2010 -0700
54873
54874 XQuartz: Fix possible NULL dereference in ListenOnOpenFD
54875
54876 <rdar://problem/7862319>
54877
54878 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
54879 Reviewed-by: Marc Majka <majka@apple.com>
54880
54881commit 206531f75cd41c034e89fdfbc75ab0910682eef8
54882Author: Jeremy Huddleston <jeremyhu@apple.com>
54883Date: Tue Apr 13 14:33:49 2010 -0700
54884
54885 XQuartz: Localization update
54886
54887 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
54888
54889commit 9ddbb03fa56aa73c3f417d8ee6433e45b94445b3
54890Author: Peter Hutterer <peter.hutterer@who-t.net>
54891Date: Wed Apr 14 17:43:22 2010 +1000
54892
54893 dix: Fix crash in DeliverGrabbedEvents.
54894
54895 If both devices are synchronously grabbed, first with a GrabPointer, then
54896 with a GrabKeyboard (GrabModeSync on both), sync.other of each device points
54897 to the grab of the respective other device.
54898
54899 If the keyboard is then thawed through a AllowSome request, the VCK's
54900 sync.other is reset to NULL. Subsequently, an event on the VCP would crash
54901 the server when dereferencing sync.other on the VCP.
54902
54903 The check's purpose is to compare if the other device is grabbed by the same
54904 client, which should be checked by accessing (dev->deviceGrab->grab->resource).
54905 A check of the server-1.3 sources confirms that.
54906
54907 XTS test case: Xlib13 XAllowEvents 20.
54908
54909 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
54910 Reviewed-by: Keith Packard <keithp@keithp.com>
54911 Signed-off-by: Keith Packard <keithp@keithp.com>
54912
54913commit 7b6517526631a65891b806bca30be8f49955d0a8
54914Author: Michel Dänzer <daenzer@vmware.com>
54915Date: Wed Apr 14 19:28:37 2010 +0200
54916
54917 EXA: Check sys_ptr isn't NULL before passing it to the UploadToScreen hook.
54918
54919 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=27510 .
54920
54921 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
54922 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
54923 Signed-off-by: Keith Packard <keithp@keithp.com>
54924
54925commit fc5d76740851725e3788c68e14474a012a205892
54926Author: Julien Cristau <jcristau@debian.org>
54927Date: Tue Apr 6 11:51:53 2010 +0200
54928
54929 XGE: don't register an extension event
54930
54931 The GenericEvent is a core event, we never send an extension event, so
54932 don't reserve an id for one.
54933
54934 Signed-off-by: Julien Cristau <jcristau@debian.org>
54935 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
54936 Signed-off-by: Keith Packard <keithp@keithp.com>
54937
54938commit 0b676acd47e0d6a675df374b42ddb200deab5d06
54939Author: Julien Cristau <jcristau@debian.org>
54940Date: Tue Apr 13 11:45:41 2010 +0200
54941
54942 Bump version in configure.ac
54943
54944 Signed-off-by: Keith Packard <keithp@keithp.com>
54945
54946commit 1760d2bef9f5b248cb2332f6ebf0220eb02bab42
54947Author: Kristian Høgsberg <krh@bitplanet.net>
54948Date: Fri Apr 9 10:58:48 2010 -0400
54949
54950 glx: Set the pbuffer bit for dri2 fbconfigs
54951
54952 They've been implemented for a while, but we never advertised them. All we
54953 need to do is set the GLX_PBUFFER_BIT in the drawable type fbconfig
54954 field when we're using DRI2.
54955
54956 https://bugs.freedesktop.org/show_bug.cgi?id=26581
54957
54958 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
54959 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
54960 Signed-off-by: Keith Packard <keithp@keithp.com>
54961
54962commit ab60975fe28cb135b4183c57b7f50d3615921c1f
54963Merge: adbbc66 59edde7
54964Author: Keith Packard <keithp@keithp.com>
54965Date: Mon Apr 12 21:40:41 2010 -0700
54966
54967 Merge remote branch 'jeremyhu/master'
54968
54969commit 59edde7c28db2d3174a5813a0af334e4ead2082f
54970Author: Jeremy Huddleston <jeremyhu@apple.com>
54971Date: Fri Apr 9 16:19:43 2010 -0700
54972
54973 XQuartz: Add a GUI preference for the Alt / Mode_switch toggle
54974
54975 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
54976 Reviewed-by: Kevin Van Vechten <kvv@apple.com>
54977
54978commit 5600f7f001529b3afdee95546aec212a70d6a5b2
54979Author: Jeremy Huddleston <jeremyhu@apple.com>
54980Date: Fri Apr 9 13:33:06 2010 -0700
54981
54982 XQuartz: Customize the NSDefaults id in the man file.
54983
54984 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
54985 Reviewed-by: Kevin Van Vechten <kvv@apple.com>
54986
54987commit 840d12c7a6ac6a974da185045215ce944a61fab6
54988Author: Jeremy Huddleston <jeremyhu@apple.com>
54989Date: Fri Apr 9 13:29:34 2010 -0700
54990
54991 XQuartz: Add a defaults option to toggle Alt / Mode_switch
54992
54993 See option_sends_alt in Xquartz(1)
54994
54995 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
54996 Reviewed-by: Kevin Van Vechten <kvv@apple.com>
54997
54998commit d6f160510a50d4c4eaa48c9c69a5ddda0d50052c
54999Author: Jeremy Huddleston <jeremyhu@apple.com>
55000Date: Mon Apr 5 16:54:22 2010 -0700
55001
55002 XQuartz: Blacklist some oddball legacy Mac keycodes that break wine
55003
55004 http://xquartz.macosforge.org/trac/ticket/295
55005
55006 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55007 Reviewed-by: Kevin Van Vechten <kvv@apple.com>
55008
55009commit adbbc661080ba4efdb764c154d40c4d2fe718e16
55010Author: Dan Nicholson <dbn.lists@gmail.com>
55011Date: Tue Feb 16 08:11:41 2010 -0800
55012
55013 xfree86: Fix priority ordering for ignoring input classes
55014
55015 Commit 8736d112afb0dd61dfdaadd6378eafd200b2ef5f changed the priority
55016 ordering of the InputClass option merging to be "last match wins". This
55017 fixes the handling of Option "Ignore" to follow that logic.
55018
55019 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
55020 Reviewed-by: Keith Packard <keithp@keithp.com>
55021
55022commit d7c98c1c81ae272f66edb05fde20f4c616604add
55023Merge: 82cf3a4 03ccbd2
55024Author: Keith Packard <keithp@keithp.com>
55025Date: Wed Apr 7 22:25:51 2010 -0700
55026
55027 Merge remote branch 'whot/for-keith'
55028
55029commit 03ccbd2579948d11e5f16f88cdf68a55f57c9d26
55030Author: Peter Hutterer <peter.hutterer@who-t.net>
55031Date: Thu Apr 8 13:47:26 2010 +1000
55032
55033 xfree86: remove dead input drivers from xorg.conf man page.
55034
55035 These drivers have been deactivated for over a year now, let's not refer
55036 potential users to them.
55037
55038 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55039
55040commit d4dd3d0085f8eaca8c989aaeace1a1a24c5ebf49
55041Author: Dan Nicholson <dbn.lists@gmail.com>
55042Date: Sat Apr 3 09:33:51 2010 -0700
55043
55044 Move 10-evdev.conf to system config dir $datadir/X11/xorg.conf.d
55045
55046 Since the server searches in a vendor specific path now, we can install
55047 the evdev catchall there without disturbing local administration files.
55048
55049 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
55050 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55051
55052commit 95f01bdfee8241371675f0089170fa6b2908d815
55053Author: Dan Nicholson <dbn.lists@gmail.com>
55054Date: Sat Apr 3 10:08:25 2010 -0700
55055
55056 xfree86: Search for a system xorg.conf.d
55057
55058 In addition to the conf files found in /etc/X11 or $sysconfdir/X11 used
55059 for local administration, we also reserve a system directory for vendor
55060 and package usage. The simple search path is:
55061
55062 /usr/share/X11/xorg.conf.d
55063 $datadir/X11/xorg.conf.d
55064
55065 Files from these directories will have the lowest config priority. The
55066 directory $datadir/X11/xorg.conf.d is exported from xorg-server.pc in
55067 the variable "sysconfigdir". Packages should install their .conf files
55068 to the directory specified by:
55069
55070 `pkg-config --variable=sysconfigdir xorg-server`
55071
55072 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
55073 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55074
55075commit a1bae63dc634a5a79861eab128c63506081f7ed9
55076Author: Dan Nicholson <dbn.lists@gmail.com>
55077Date: Sat Apr 3 09:33:49 2010 -0700
55078
55079 xfree86: Set a saner search path for xorg.conf.d
55080
55081 There's no reason to carry all the oddities from xorg.conf like appended
55082 hostname to the search path for xorg.conf.d. This changes it to something
55083 very simple:
55084
55085 /etc/X11/<cmdline>
55086 $sysconfdir/X11/<cmdline>
55087 /etc/X11/xorg.conf.d
55088 $sysconfdir/X11/xorg.conf.d
55089
55090 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
55091 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55092
55093commit 2ac33888a9a5766aaf88bc92a8bd1d489873524c
55094Author: Dan Nicholson <dbn.lists@gmail.com>
55095Date: Sat Apr 3 09:33:48 2010 -0700
55096
55097 xfree86: Document how -configdir affects the xorg.conf.d search path
55098
55099 Explain the "safe" path dance for -configdir, too.
55100
55101 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
55102 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55103
55104commit 2460e921d18fd264e6f6374be0908f4100442650
55105Author: Dan Nicholson <dbn.lists@gmail.com>
55106Date: Sat Apr 3 09:33:47 2010 -0700
55107
55108 xfree86: Allow adding sysconfdir and datadir to config search paths
55109
55110 We could just use $projectroot/etc and $projectroot/share, but the user
55111 might have other plans for them.
55112
55113 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
55114 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55115
55116commit c8a608cb6ce8f9c86258c1ab49084f691fa9cc51
55117Author: Peter Hutterer <peter.hutterer@who-t.net>
55118Date: Wed Apr 7 10:04:38 2010 +1000
55119
55120 config: only match sane devices in 10-evdev.conf
55121
55122 Having a generic catchall also adds devices like accelerometers. These
55123 devices make X unusable, hence restrict matching to "known sane" devices
55124 like pointers, touchpads, keyboards, tablets and touchscreens.
55125
55126 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55127 Acked-by: Julien Cristau <jcristau@debian.org>
55128 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
55129 Acked-by: James Cloos <cloos@jhcloos.com>
55130
55131commit 6b09f66d8c6ebcee70382b5cca1ba82b68f20afa
55132Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
55133Date: Fri Apr 2 12:48:21 2010 -0700
55134
55135 Don't keep a pointer to a possibly freed cursor when changing screens, preventing a crash in xf86CursorEnableDisableFBAccess() trying to restore it.
55136
55137 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
55138 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55139
55140commit 82cf3a4ae01811917f7903d6f62ba9b7132adf7e
55141Author: Alan Coopersmith <alan.coopersmith@oracle.com>
55142Date: Tue Apr 6 15:52:18 2010 -0700
55143
55144 Convert x86emu fixed size int typedefs to use stdint types
55145
55146 Fixes x86emu builds when using non-gnu compilers now that u64 is required
55147
55148 Before this fix, the u64 type would not be defined, causing
55149 x86emu/sys.c to fail to build:
55150 "sys.c", line 102: syntax error before or at: ldq_u
55151 "sys.c", line 102: syntax error before or at: *
55152
55153 Since Keith requested using <stdint.h>, converted all the x86emu
55154 typedefs to use the stdint types.
55155
55156 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
55157 Acked-by: Matt Turner <mattst88@gmail.com>
55158 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
55159 Signed-off-by: Keith Packard <keithp@keithp.com>
55160
55161commit a9fe7cfa77ccee64d68732dc3f37d35cbfc27a65
55162Author: Paulo Ricardo Zanoni <pzanoni@mandriva.com>
55163Date: Thu Mar 11 14:28:18 2010 -0300
55164
55165 dix: be more verbose when we run out of opcodes
55166
55167 If we run out of opcodes, nothing is print on the log, making the
55168 problem hard to debug. In the current Xserver, if you enable some
55169 extensions like multibuffer (+2 events) and use nvidia binary driver (+5
55170 events) you can run out of opcode numbers.
55171
55172 Signed-off-by: Paulo Ricardo Zanoni <pzanoni@mandriva.com>
55173 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
55174 Reviewed-by: Julien Cristau <jcristau@debian.org>
55175 Signed-off-by: Keith Packard <keithp@keithp.com>
55176
55177commit 6150595bdbacde13bb4f9aba81b15435089bee94
55178Author: Gaetan Nadon <memsize@videotron.ca>
55179Date: Mon Apr 5 19:12:22 2010 -0400
55180
55181 doc: specify 1.6.1 as the minimum version for doxygen.
55182
55183 Older versions generate filenames that are different from
55184 the ones listed in the Makefile.
55185
55186 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
55187 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
55188 Signed-off-by: Keith Packard <keithp@keithp.com>
55189
55190commit cbda58c963af5bb9c4643e9017dd4eec589ceff3
55191Merge: a7698a6 c983f40
55192Author: Keith Packard <keithp@keithp.com>
55193Date: Wed Apr 7 15:51:38 2010 -0700
55194
55195 Merge remote branch 'yselkowitz/master'
55196
55197commit a7698a677682a32960b885c22fdba2add70f4658
55198Merge: 495cec7 165a4a9
55199Author: Keith Packard <keithp@keithp.com>
55200Date: Tue Apr 6 12:36:15 2010 -0700
55201
55202 Merge remote branch 'jbarnes/master'
55203
55204commit c983f40fe4f30311f31a9df699cdc888709251d0
55205Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55206Date: Wed Feb 24 15:17:42 2010 -0600
55207
55208 Cygwin/X: Add configure option for WindowsWM
55209
55210 WindowsWM support is still experimental, and uses the Rootless extension
55211 which currently breaks the simultaneous build of the other DDXs (see
55212 commit b3415187e92960cbff784108b5a3a8d130dc34c5). So we disable it by
55213 default for now; once the latter issue is fixed we can make this 'auto'.
55214
55215 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55216
55217commit 20701522be803fe47e921fcf059dadf64c7f287d
55218Author: Jon TURNEY <jon.turney@dronecode.org.uk>
55219Date: Thu Feb 18 14:19:21 2010 -0600
55220
55221 Don't enable ROOTLESS_WORKAROUND, it breaks composite
55222
55223 This possibly brings back whatever the bug is in
55224 http://bugs.freedesktop.org/show_bug.cgi?id=1168 for
55225 -rootless mode, but since we don't have reproduction
55226 steps for that, I can't test that...
55227
55228 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
55229 Tested-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55230 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55231
55232commit 95c9947f6244d37592958b70760bd1ac885b625a
55233Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55234Date: Wed Mar 24 00:08:23 2010 -0500
55235
55236 Cygwin/X: Disable unsupported extensions in configure
55237
55238 Several extensions are not supported by XWin, some of which are enabled
55239 by default in configure. We forcefully disable these early on so that
55240 configure will succeed without arguments and without the corresponding
55241 proto installed.
55242
55243 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55244
55245commit 24c9d48cd6999f1e90f9f0e2bbfebd8e95d314d8
55246Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55247Date: Mon Mar 22 12:47:52 2010 -0500
55248
55249 Disable setuid configure test on Cygwin
55250
55251 Only Xorg is installed setuid, so there is no need to run this configure
55252 test on Cygwin.
55253
55254 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55255
55256commit 495cec794dad95ed0c79048f3c410ad23e7d5ea4
55257Author: Keith Packard <keithp@keithp.com>
55258Date: Fri Apr 2 00:14:23 2010 -0700
55259
55260 xserver 1.8.0
55261
55262 Signed-off-by: Keith Packard <keithp@keithp.com>
55263
55264commit b9ad452ec92a7dcbed680acb3f3b8ec29fa660df
55265Author: Tiago Vignatti <tiago.vignatti@nokia.com>
55266Date: Thu Apr 1 18:57:50 2010 +0300
55267
55268 xfree86: die gracefully in the vga arbiter if AddScreen fails
55269
55270 vga arbiter will be locked in one device while AbortDDX will call LeaveVT
55271 routines from the other device. Fail!
55272
55273 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
55274 Reviewed-by: Keith Packard <keithp@keithp.com>
55275 Signed-off-by: Keith Packard <keithp@keithp.com>
55276
55277commit e75f202e7768b7cc0eb2ad0e0f3f51ea81cd7275
55278Author: Peter Hutterer <peter.hutterer@who-t.net>
55279Date: Wed Mar 24 11:08:55 2010 +1000
55280
55281 configure: enable udev backend as "auto"
55282
55283 Due to the checks in configure, this means it gets priority over HAL if
55284 libudev is found.
55285
55286 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55287 Signed-off-by: Keith Packard <keithp@keithp.com>
55288
55289commit 24952b786625fef808cb26d539dd5109d623c053
55290Author: Keith Packard <keithp@keithp.com>
55291Date: Thu Apr 1 23:44:24 2010 -0700
55292
55293 Install 10-evdev.conf in $(prefix)/etc/X11/xorg.conf.d under udev
55294
55295 udev needs some xorg.conf file to tell it to load a suitable input
55296 driver, 10-evdev.conf is as simple as they come, mapping all evdev
55297 devices to the evdev driver.
55298
55299 Signed-off-by: Keith Packard <keithp@keithp.com>
55300
55301commit 02328f190f893cb4ad0f3af7a2ecd39ed77ae5ad
55302Author: Jeremy Huddleston <jeremyhu@apple.com>
55303Date: Thu Apr 1 18:21:51 2010 -0700
55304
55305 Bump bundle version to 2.5.1
55306
55307 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55308
55309commit 0e342add7768df55a1917e27f9b85feddd444e82
55310Author: Jeremy Huddleston <jeremyhu@apple.com>
55311Date: Thu Apr 1 15:30:46 2010 -0700
55312
55313 darwin: Correct inline assembly for ___crashreporter_info__
55314
55315 It was missing an underscore.
55316
55317 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55318
55319commit 67b814d9b2baea6beccfb1625a1e3f0b2ba7218b
55320Author: Ruediger Oertel <ro@suse.de>
55321Date: Mon Mar 29 00:23:40 2010 +0200
55322
55323 Remove now obsolete function chooseVideoDriver
55324
55325 Signed-off-by: Keith Packard <keithp@keithp.com>
55326
55327commit 1dd5fbc5a4d860af1c102b33e04dedd8c72617a1
55328Author: Ruediger Oertel <ro@suse.de>
55329Date: Fri Mar 26 09:33:20 2010 -0700
55330
55331 xfree86: Handle driver autoconfiguration when .conf files exist
55332
55333 When doing driver autoconfiguration with some parts of the config file
55334 present but no driver set (e.g. only input configuration) fix the case
55335 that we may have multiple drivers to try.
55336
55337 Create a screen section for each driver and let them be tried in a row.
55338
55339 Signed-off-by: Ruediger Oertel <ro@suse.de>
55340 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
55341 Tested-by: Timo Aaltonen <timo.aaltonen@aalto.fi>
55342 Signed-off-by: Keith Packard <keithp@keithp.com>
55343
55344commit 57cb1a854a1b6d91d214b3fa250a7df2ed20d433
55345Merge: 4b3d67a 28a5f14
55346Author: Keith Packard <keithp@keithp.com>
55347Date: Mon Mar 29 11:46:00 2010 -0700
55348
55349 Merge remote branch 'jeremyhu/master'
55350
55351commit 4b3d67a7647f696957727948f9757a261e15d14e
55352Merge: 579715f aa91508
55353Author: Keith Packard <keithp@keithp.com>
55354Date: Mon Mar 29 11:44:07 2010 -0700
55355
55356 Merge remote branch 'whot/for-keith'
55357
55358commit 165a4a9c7de0fcc6ef6a6421736b412ccb35965e
55359Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55360Date: Tue Mar 23 09:47:08 2010 -0700
55361
55362 GLX/DRI2: expose swap control extensions if DDX support is present
55363
55364 Export DDX swap control status from the DRI2 module and check for it in
55365 GLX when initializing extensions.
55366
55367 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55368 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55369
55370commit 5933b0abc6a76aaea84aa534df89900cd795c888
55371Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55372Date: Mon Mar 8 15:10:47 2010 -0800
55373
55374 DRI2: prevent swap wakes from waking MSC waiters
55375
55376 If a few swaps were queued leading to a throttle related block on the
55377 client, and then the client submitted an MSC wait, one of the previous
55378 swap wakeups could have caused the MSC wait to complete early. Add a
55379 flag for this to prevent a swap wake from prematurely waking an MSC
55380 waiter.
55381
55382 Reported-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55383 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55384 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55385
55386commit b00d435ddf2e9817e33bfd5f7e9b905442dc23c7
55387Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55388Date: Mon Mar 8 12:41:25 2010 -0800
55389
55390 DRI2: handle swapsPending better
55391
55392 Avoid a potential swapsPending underflow by incrementing it before
55393 ScheduleSwap, which may complete it immediately. And be sure to
55394 decrement it again in case the schedule failed.
55395
55396 Reported-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55397 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55398 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55399
55400commit 0294ff2a5cadddc8fcc77ba9a851f979f0b91fc3
55401Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55402Date: Mon Mar 8 12:39:54 2010 -0800
55403
55404 DRI2: throttle swaps at submission time too
55405
55406 We need to throttle swaps here in addition to when the context is made
55407 current to avoid causing problems with clients that just swap.
55408 Throttling here also ensures our swaps get ordered as long as we block
55409 the client occasionally.
55410
55411 Reported-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55412 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55413 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55414
55415commit db1c7cb604167baf49e61be4c09ccf7b592c4af3
55416Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55417Date: Mon Mar 8 12:38:37 2010 -0800
55418
55419 DRI2: advertise lowest supported DRI2 protocol version
55420
55421 Update our supported DRI2 protocol version as each driver does
55422 DRI2ScreenInit, since depending on available kernel features, each DDX
55423 may support different callbacks and therefore protocol.
55424
55425 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55426 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55427
55428commit 87ca6320f26eb3129e3c19056e1d8fa5c1784723
55429Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55430Date: Fri Mar 5 09:49:03 2010 -0800
55431
55432 DRI2: handle swap_interval of 0 correctly
55433
55434 A 0 swap interval means that swaps shouldn't be sync'd to vblank, so
55435 just complete the swap immediately in that case.
55436
55437 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55438 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55439
55440commit 8476d99231cb725c090305d60f1c1c889d25c8dc
55441Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55442Date: Fri Mar 5 09:15:24 2010 -0800
55443
55444 DRI2: drawable lifetime fixes
55445
55446 Handle drawable destruction and lifetime correctly.
55447
55448 Check whether the drawable priv is valid in DRI2SwapInterval(),
55449 DRI2WaitSBC() and DRI2WaitMSC(); it may have gone away, so be sure to
55450 check it before using it.
55451
55452 If more than 1 outstanding swap is queued, we may complete several after
55453 an app has exited. If we free it after the first one completes and the
55454 refcount reaches 0, we'll crash the server on subsequent completions.
55455 So delay freeing until all swaps complete and remove the error message
55456 as this is a normal occurence. To do this properly, we must also avoid
55457 destroying drawables in DRI2DestroyDrawable() if a swap or wait event is
55458 pending.
55459
55460 And finally, make sure we free drawables in DRI2WaitMSCComplete() if
55461 necessary (i.e. if the refcount has reached 0 and this MSC was the last
55462 pending event on the object).
55463
55464 Reported-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55465 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55466 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55467
55468commit b180e43977710b56ccfd6780f204ddcc952987a1
55469Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55470Date: Thu Mar 4 10:31:59 2010 -0800
55471
55472 DRI2: fix swapbuffers handling of SBC and target MSC
55473
55474 Returns expected SBC after completion of swap to caller, as required by
55475 OML_sync_control spec, instead of the last_swap_target value.
55476
55477 Passes target_msc, divisor, remainder, correctly for
55478 glXSwapBuffersMscOML() call, while retaining old behaviour for simple
55479 glXSwapBuffers() call.
55480
55481 An OML swap can have a 0 target_msc, which just means it needs to
55482 satisfy the divisor/remainder equation. Pass this down to the driver as
55483 needed so we can support it.
55484
55485 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55486 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55487
55488commit 751e8c09d34df4b41e8d8384a3ec1bf5cb8ca028
55489Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55490Date: Sun Feb 21 05:26:00 2010 +0100
55491
55492 DRI2WaitSbc(): Fixes for correct semantic of glXWaitForSbcOML()
55493
55494 Added implementation for case target_sbc == 0. In that case, the
55495 function shall schedule a wait until all pending swaps for the drawable
55496 have completed.
55497
55498 Fix for non-blocking case. Old implementation returned random,
55499 uninitialized values for (ust,msc,sbc) if it returned immediately
55500 without scheduling a wait due to sbc >= target_sbc.
55501
55502 Now if function doesn't schedule a wait, but returns immediately,
55503 it returns the (ust,msc,sbc) of the most recently completed swap,
55504 i.e., the UST and MSC corresponding to the time when the returned
55505 current SBC was reached.
55506
55507 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55508
55509commit 0de4974b90b10fa6a447cdf980b4a114c6c9e5a8
55510Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55511Date: Sun Feb 21 05:25:59 2010 +0100
55512
55513 DRI2: Fix glitches in DRI2SwapComplete() and DRI2WakeupClient()
55514
55515 DRI2SwapComplete(): Increment pPriv->swap_count++; before calling
55516 into callback for INTEL_swap_events extension, so the swap event
55517 contains the current SBC after swap completion instead of the
55518 previous one.
55519
55520 DRI2WakeupClient: Check for pPriv->target_sbc <= pPriv->swap_count,
55521 had wrong comparison pPriv->target_sbc >= pPriv->swap_count for
55522 unblocking of clients of DRI2WaitSBC().
55523
55524 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55525
55526commit 4c8ec49826a46eb3b36c69d2ad3f82320c179c38
55527Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55528Date: Thu Mar 4 09:54:15 2010 -0800
55529
55530 DRI2: make target_sbc signed
55531
55532 We need to track invalid targets as well as 0 targets, so just make it
55533 signed so our comparisons work like they should.
55534
55535 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55536 Reported-by: Kristian Høgsberg <krh@bitplanet.net>
55537 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55538
55539commit c4d54816f2ee4883d8f9bcf4595474fb58c95146
55540Author: Jesse Barnes <jbarnes@virtuousgeek.org>
55541Date: Thu Mar 4 09:19:13 2010 -0800
55542
55543 DRI2: fixup handling of last_swap_target
55544
55545 We need to initialize the swap target, which is passed to the driver to
55546 schedule events. Rather than using -1 to indicate that the field is
55547 uninitialized, just make sure we initialize it at drawable creation
55548 time.
55549
55550 Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
55551 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
55552
55553commit 28a5f14b4089dccb8045cc4fdc923542a73dd22d
55554Author: Jeremy Huddleston <jeremyhu@apple.com>
55555Date: Sun Mar 28 13:46:23 2010 -0700
55556
55557 XQuartz: Re-query dixScreenOrigins as the value could've changed.
55558
55559 Fix a regression in 9c9c3a85b094a3c7b2763a572715d710325091aa
55560
55561 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55562 Acked-by: Adam Jackson <ajax@redhat.com>
55563
55564commit 436d0bb9cca122bfdde32902b683d2499f61e6fc
55565Author: Jeremy Huddleston <jeremyhu@apple.com>
55566Date: Sat Mar 27 17:35:10 2010 -0700
55567
55568 darwin: Generate crash reports on FatalError()
55569
55570 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55571 Acked-by: Adam Jackson <ajax@redhat.com>
55572
55573commit 73b3b67aac9f3938a96cb8822b9c270bd82ded5c
55574Author: Jeremy Huddleston <jeremyhu@apple.com>
55575Date: Mon Mar 22 09:30:51 2010 -0700
55576
55577 GLX: Remove a redundant initialization
55578
55579 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55580 Acked-by: Adam Jackson <ajax@redhat.com>
55581
55582commit 339207be6f184cc783076fc7e2cc12f92f57f2ba
55583Author: Jeremy Huddleston <jeremyhu@apple.com>
55584Date: Thu Mar 25 22:15:58 2010 -0700
55585
55586 XQuartz: Workaround weird key data reported on some layouts
55587
55588 This should make 'Unicode Hex Input' work as an input layout.
55589
55590 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55591 Acked-by: Adam Jackson <ajax@redhat.com>
55592
55593commit aa91508356f243edc3b11795b1481edcfe0d39c2
55594Author: Dan Nicholson <dbn.lists@gmail.com>
55595Date: Wed Mar 24 23:18:35 2010 -0700
55596
55597 config/udev: Prefer product name from attribute rather than uevent
55598
55599 The input device product name for evdev devices in the kernel uevent has
55600 embedded quotes that aren't expected here. Use the sysfs name attribute
55601 instead, which does not suffer this problem. The uevent name will be
55602 used as a fallback if no name attribute is found.
55603
55604 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
55605 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55606
55607commit d00c3298a698ddddc8967215cef7dd941fa15804
55608Author: Fernando Carrijo <fcarrijo@yahoo.com.br>
55609Date: Wed Mar 24 20:11:34 2010 -0300
55610
55611 Cleanup some comments in SpriteRec
55612
55613 Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
55614 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55615
55616commit 70bd02f2ea3a91c0340df97ca3b8a4f9400f6f46
55617Author: Peter Hutterer <peter.hutterer@who-t.net>
55618Date: Tue Mar 23 12:13:56 2010 +1000
55619
55620 xfree86: merge driver from the input class into the options.
55621
55622 A driver that is assigned by an input class is only present as idev->driver.
55623 The driver itself has no access to this information once PreInit is called.
55624 For devices that rely on chain-hotplugging (wacom), this means that for the
55625 second device the driver information is lost and the second device cannot be
55626 initialized through NewInputDeviceRequest. Although this could be worked
55627 around by hardcoding the driver name in the wacom driver, having the
55628 assigned driver in the options seems like the better solution.
55629
55630 This issue only manifests itself with the udev backend. With HAL, the driver
55631 is assigned by HAL and the option is duplicated in config/hal.c.
55632
55633 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55634 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
55635
55636commit 579715f830fbbca9e1ecb17dc18176132f5969e7
55637Author: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
55638Date: Wed Mar 17 12:16:57 2010 +0200
55639
55640 os: Prevent backtrace from being stopped in noreturn functions.
55641
55642 There are two noreturn functions in the X server: FatalError and
55643 AbortServer. Having any of those two functions in the middle of a call
55644 stack will prevent unwinding the program properly and stops the
55645 backtrace at those functions in gdb.
55646
55647 The file containing FatalError and AbortServer, os/log.c, has to be
55648 compiled with the -mapcs-frame option on ARM to get proper
55649 backtraces. Automake imposes its own restrictions on compiling
55650 individual source files with different options. The recommended way to
55651 do this is to put os/log.c into a convenience library and add this
55652 library inside os/libos.la. See the documentation of GNU Automake
55653 manual, version 1.11.1, section 27.8 Per-Object Flags Emulation, for
55654 details.
55655
55656 Signed-off-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
55657 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
55658 Signed-off-by: Keith Packard <keithp@keithp.com>
55659
55660commit e086b99c1f5ce351b578de7cd9f616bc79d6cf64
55661Author: Mikhail Gusarov <dottedmag@dottedmag.net>
55662Date: Tue Mar 23 01:03:53 2010 +0600
55663
55664 kdrive: Bump evdev maxKeycode
55665
55666 There are keycodes > 193 in evdev, e.g. KEY_WIMAX which is 246 .
55667
55668 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
55669 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
55670 Acked-by: Adam Jackson <ajax@nwnk.net>
55671 Signed-off-by: Keith Packard <keithp@keithp.com>
55672
55673commit 185185eeb44a277c324be0f58a4b4a469b56b69b
55674Author: Peter Harris <pharris@opentext.com>
55675Date: Tue Mar 23 12:08:19 2010 -0400
55676
55677 Fix crash when all glyphs of a given depth are freed, but not all glyphsets
55678
55679 This is how the crash can be triggered with only two clients on the system:
55680 Client A: (already running)
55681 Client B: Connect
55682 Client B: CreateGlyphSet(depthN)
55683 Client A: Disconnect
55684 Server: free globalGlyphs(depthN)
55685 Client B: AddGlyphs(depthN)
55686 Server: SEGV
55687
55688 This crash was introduced with the FindGlyphsByHash function
55689 in 516b96387b0e57b524a37a96da22dbeeeb041712. Before that revision,
55690 ResizeGlyphSet was always called before FindGlyphRef, which would
55691 re-create globalGlyphs(depthN) if necessary.
55692
55693 X.Org Bug 20718 <http://bugs.freedesktop.org/show_bug.cgi?id=20718>
55694
55695 Reviewed-by: Adam Jackson <ajax@redhat.com>
55696 Signed-off-by: Peter Harris <pharris@opentext.com>
55697 Signed-off-by: Keith Packard <keithp@keithp.com>
55698
55699commit 8311cd5f89ca6781842bb24671b8122cdf8be148
55700Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
55701Date: Tue Mar 23 10:37:36 2010 -0700
55702
55703 XKB: Fix garbage initialization
55704
55705 XkbEnableDisableControls set extra garbage bits on the xkbControlsNotify
55706 changedControls mask because it was uninitialized on the stack.
55707
55708 Found by clang
55709
55710 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55711 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55712
55713commit 0ecfab89d1754b0b7e32b3964b40663d1810cd62
55714Author: Peter Hutterer <peter.hutterer@who-t.net>
55715Date: Mon Mar 22 14:04:09 2010 +1000
55716
55717 xfree86: remove if 1 from the dawn of time.
55718
55719 7+ years for an if 1 should be enough to just admit that there's no other
55720 option.
55721
55722 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55723 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
55724 Reviewed-by: Corbin Simpson <MostAwesomeDude@gmail.com>
55725
55726commit 6389405b56d2221d8a7b80c4bb3bd3b4e1a29bab
55727Author: Peter Hutterer <peter.hutterer@who-t.net>
55728Date: Mon Mar 22 09:20:49 2010 +1000
55729
55730 configure: Always define XINPUT.
55731
55732 This define was removed in e251c9e75afdc "Remove all traces of #ifdef XINPUT
55733 and the matching bits from the configure.ac". Drivers that support multiple
55734 server versions and still check for ifdef XINPUT now always build without
55735 XINPUT support if they're calling XORG_DRIVER_CHECK_EXT(XINPUT, inputproto).
55736
55737 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
55738 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
55739
55740commit 0c2fde5c8ad6e94b4ed1588aa93256a2b64f74d9
55741Author: Tomas Carnecky <tom@dbservice.com>
55742Date: Mon Mar 22 11:20:15 2010 -0700
55743
55744 Fix typos in the swap functions
55745
55746 This should fix bug #3539.
55747
55748 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
55749 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
55750 Reviewed-by: Keith Packard <keithp@keithp.com>
55751 Signed-off-by: Keith Packard <keithp@keithp.com>
55752
55753commit 3083c5d0c4386cdd7083b7a83ac72fdad2f1e61e
55754Author: Michel Dänzer <daenzer@vmware.com>
55755Date: Mon Mar 22 18:01:17 2010 +0100
55756
55757 Xext: Fix cursor reference counting hazard.
55758
55759 Make sure the reference count of the new cursor is increased before the old
55760 one is decreased, otherwise bad things will happen if they're one and the
55761 same and the reference count is 1 initially. Not sure this can actually happen
55762 here, but better safe than sorry.
55763
55764 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
55765 Reviewed-by: Roland Scheidegger <sroland@vmware.com>
55766 Signed-off-by: Keith Packard <keithp@keithp.com>
55767
55768commit b29220dc765cb6f878c5466e00e4bd21f3bd803d
55769Merge: c3da766 ed31d50
55770Author: Keith Packard <keithp@keithp.com>
55771Date: Sun Mar 21 23:01:58 2010 -0700
55772
55773 Merge remote branch 'jeremyhu/master'
55774
55775commit c3da76643aeebdd24047e3e14a860507eac37ec3
55776Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55777Date: Thu Mar 18 23:19:20 2010 -0500
55778
55779 Cygwin/X: Make X -> XWin symlink during install
55780
55781 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55782 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
55783
55784commit abf4e0b7e349b0238b2b97a16b8a308b5026af31
55785Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55786Date: Mon Mar 22 00:44:28 2010 -0500
55787
55788 New header for XF86Bigfont server functions
55789
55790 Xext/xf86bigfont.c contains three non-static functions which are called
55791 elsewhere in the server. This creates a new header containing these
55792 declarations in order to fix several warnings:
55793
55794 xf86bigfont.c:285: warning: no previous prototype for `XF86BigfontFreeFontShm'
55795 dixfonts.c:502: warning: implicit declaration of function `XF86BigfontFreeFontS$
55796 dixfonts.c:502: warning: nested extern declaration of `XF86BigfontFreeFontShm'
55797 log.c:436: warning: implicit declaration of function `XF86BigfontCleanup'
55798 log.c:436: warning: nested extern declaration of `XF86BigfontCleanup'
55799
55800 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55801 Reviewed-by: Julien Cristau <jcristau@debian.org>
55802
55803commit 57a049ea89b008c0b60316c3b6e6ff5c8fbd4cad
55804Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55805Date: Thu Mar 18 04:46:20 2010 -0500
55806
55807 Xext: fix old-style function definitions in xf86bigfont.c
55808
55809 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55810 Reviewed-by: Julien Cristau <jcristau@debian.org>
55811
55812commit 822b9f9a3e822df8848995eda246f83864d1366f
55813Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55814Date: Sun Feb 21 01:24:44 2010 -0600
55815
55816 mi: remove deprecated #include <X11/extensions/xf86bigfstr.h> in miinitext.c
55817
55818 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55819 Reviewed-by: Julien Cristau <jcristau@debian.org>
55820
55821commit 946b49ebcbf47e030a6e00ce1a699f3b055f638a
55822Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55823Date: Wed Mar 17 17:37:05 2010 -0500
55824
55825 Catch errors in recursive relink targets
55826
55827 If make relink fails in a subdirectory, we need to catch the error
55828 otherwise make will continue iterating the 'for' loop.
55829
55830 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55831 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
55832
55833commit 0820a6e2fb005b5f856dda25e1644f34fc994129
55834Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55835Date: Wed Mar 17 13:40:41 2010 -0500
55836
55837 Fix .man.N targets for AM_SILENT_RULES
55838
55839 Add $(AM_V_GEN) for sed-based rules so they appear as expected with
55840 automake silent rules, and $(AM_V_at) to completely hide cp/ln/rm
55841 commands which are not prone to fail.
55842
55843 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55844 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
55845 Reviewed-by: Julien Cristau <jcristau@debian.org>
55846
55847commit 9be4157391edf0c5fc4ee36adfb1eb1c3bdb8e3b
55848Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55849Date: Wed Mar 17 13:33:39 2010 -0500
55850
55851 Respect value of SED from configure
55852
55853 We now use libtool, which calls AC_PROG_SED and sets SED as the path to
55854 a fully-functional 'sed' (which may also be called 'gsed' if GNU sed is
55855 installed alongside a proprietary version). Therefore we should respect
55856 the value of SED so we are sure to use the correct one.
55857
55858 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55859 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
55860 Reviewed-by: Julien Cristau <jcristau@debian.org>
55861
55862commit 5e00f464c546debf9164f24c6d648623f7ee5946
55863Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55864Date: Wed Mar 17 14:56:24 2010 -0500
55865
55866 Fix relink targets for silent rules
55867
55868 Add $(AM_V_at) to all relink make targets to silence them when automake
55869 silent rules are in use.
55870
55871 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55872 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
55873 Reviewed-by: Julien Cristau <jcristau@debian.org>
55874
55875commit fa7e062962d17862032bc600bfcffd486addb063
55876Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55877Date: Wed Mar 17 13:52:38 2010 -0500
55878
55879 Use EXEEXT in relink rules for portable DDXs
55880
55881 On Cygwin and MinGW, executables use the .exe suffix. Autoconf and
55882 automake set EXEEXT on these platforms, and leave it empty on others
55883 where no suffix is used. $(EXEEXT) must be appended to executable names
55884 in custom rules for portability:
55885
55886 http://www.gnu.org/software/automake/manual/html_node/EXEEXT.html
55887
55888 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55889 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
55890 Reviewed-by: Julien Cristau <jcristau@debian.org>
55891
55892commit 037869d7b876f43158df88d63788b93a5d6187c1
55893Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55894Date: Wed Mar 17 13:48:49 2010 -0500
55895
55896 kdrive: Use $(MAKE) in relink rules
55897
55898 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55899 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
55900 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
55901 Reviewed-by: Julien Cristau <jcristau@debian.org>
55902
55903commit 8e5de45513577c63a33833931f2afd0cc59d42a7
55904Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55905Date: Wed Feb 24 02:56:59 2010 -0600
55906
55907 Use libtool -export-dynamic flag for portability
55908
55909 The linker flag required for exporting symbols in executables varies
55910 by platform. libtool handles this with a single -export-dynamic
55911 flag (not to be confused with the similarly-named ELF linker flag)
55912 which tells it to use the correct platform-specific flag at link time.
55913
55914 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55915 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
55916 Reviewed-by: Keith Packard <keithp@keithp.com>
55917
55918commit 2a11ffa977d2267d8b2f27b76490a98e9c73b8c7
55919Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55920Date: Fri Feb 19 14:31:28 2010 -0600
55921
55922 Rename xdmx client to dmxinfo
55923
55924 The DMX Xdmx server and xdmx client cannot both be installed on
55925 case-insensitive file systems. The client is undocumented and
55926 so renaming it is the best option.
55927
55928 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55929 Reviewed-by: Keith Packard <keithp@keithp.com>
55930
55931commit fbb4903eaa23174d21ab8c640d9de1fe3eb17539
55932Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55933Date: Sun Feb 21 02:08:08 2010 -0600
55934
55935 Cygwin/X: Fix windres rule for automake silent rules
55936
55937 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55938 Reviewed-by: Rémi Cardona <rem@gentoo.org>
55939 Reviewed-by: Keith Packard <keithp@keithp.com>
55940
55941commit 8db894690988f2bac9c0acb96fd39811d2b42f47
55942Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55943Date: Fri Feb 19 00:05:28 2010 -0600
55944
55945 Cygwin/X: Fix make dist after 11252ed82e1f361b99e86521ac9314f868bd1a3a
55946
55947 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
55948 Acked-by: Adam Jackson <ajax@redhat.com>
55949 Reviewed-by: Keith Packard <keithp@keithp.com>
55950
55951commit ed31d50b5f7f25e4db986711699704e615b8afcb
55952Author: Jeremy Huddleston <jeremyhu@apple.com>
55953Date: Fri Mar 19 17:24:27 2010 -0700
55954
55955 XQuartz: Constrain the pointer to the updated display bounds on display reconfigure.
55956
55957 http://xquartz.macosforge.org/trac/ticket/346
55958
55959commit 927480be12ff61deb4e8120d2f90ddef3a1a6beb
55960Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
55961Date: Sat Mar 20 11:03:28 2010 -0700
55962
55963 XQuartz: pbproxy: Make standalone xpbproxy respect the launchd prefix
55964
55965 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
55966
55967commit d16bc8a3cdbe0e8b9cba836c7aa2ff7c0458049e
55968Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
55969Date: Sat Mar 20 03:28:57 2010 -0700
55970
55971 XQuartz: xpbproxy: Cleanup xpbproxy threading
55972
55973 Confine xpbproxy to a single thread
55974 Runs inside its own CFRunLoop
55975
55976 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
55977
55978commit 9c9c3a85b094a3c7b2763a572715d710325091aa
55979Author: Jeremy Huddleston <jeremyhu@apple.com>
55980Date: Fri Mar 19 17:23:12 2010 -0700
55981
55982 XQuartz: Minor cleanup
55983
55984 Move RandRInit to where it will need to be (not yet implemented)
55985
55986 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
55987
55988commit bb75d0df8b5238bfe8b011bb5737fae2a3584290
55989Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
55990Date: Fri Mar 19 15:26:50 2010 -0700
55991
55992 XQuartz: GLX: Fix Availability for Tiger ppc workaround
55993
55994 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
55995
55996commit eac7cdabecafb7c505795207182ab2578d672c06
55997Author: Jeremy Huddleston <jeremyhu@apple.com>
55998Date: Tue Mar 16 11:29:08 2010 -0700
55999
56000 Revert "XQuartz: Explicitly pass a bellProc to make XBell() work again."
56001
56002 I'm not quite sure why this was necessary, but DDXRingBell is being called
56003 from CoreKeyboardBell, so we don't need a separate bellProc which would
56004 result in multiple rings.
56005
56006 This reverts commit 9071b0d69748cfa7ecca17b4cb0e431bbb0ef2a4.
56007
56008 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
56009
56010commit 13c007f9224ad871193e40466e64df1477ed26b0
56011Author: Keith Packard <keithp@keithp.com>
56012Date: Sun Mar 21 17:23:46 2010 -0700
56013
56014 Bump to 1.7.99.902 -- 1.8 RC2
56015
56016commit 235fa5030428084368e5be57fca695647b7b79c4
56017Merge: 1c612ac fa5103a
56018Author: Keith Packard <keithp@keithp.com>
56019Date: Sun Mar 21 15:38:40 2010 -0700
56020
56021 Merge commit 'fa5103a02bd509e4a102afdad2ab26cb22210367'
56022
56023commit 1c612acca8568fcdf9761d23f112adaf4d496f1b
56024Author: Peter Hutterer <peter.hutterer@who-t.net>
56025Date: Wed Mar 17 14:32:38 2010 +1000
56026
56027 dix: if owner-events is true for passive grabs, add the window mask (#25400)
56028
56029 A client requesting a GrabModeSync button grab, owner-events true, with only
56030 the ButtonRelease mask set would never receive the press event even if the
56031 grab window had the ButtonPress mask set.
56032
56033 The protocol requires that if owner-events is true, then the delivery mask
56034 is the combination of the grab mask + the window event mask.
56035
56036 X.Org Bug 25400 <http://bugs.freedesktop.org/show_bug.cgi?id=25400>
56037
56038 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56039 Tested-by: Jim Ramsay <i.am@jimramsay.com>
56040 Signed-off-by: Keith Packard <keithp@keithp.com>
56041
56042commit 95ca39b981624df5338a16b506289014c431ae82
56043Merge: e7ff956 bf18191
56044Author: Keith Packard <keithp@keithp.com>
56045Date: Sun Mar 21 15:24:06 2010 -0700
56046
56047 Merge remote branch 'jturney/master'
56048
56049commit e7ff956638780dc1c6f160a1fd041b92cebff9ac
56050Author: Oliver McFadden <oliver.mcfadden@nokia.com>
56051Date: Mon Mar 15 23:46:36 2010 +0200
56052
56053 common: xf86Configure: alloc_strlen: Allocated memory does not have space for the terminating NUL of the string
56054
56055 buffer_alloc: Called allocating function "realloc" which allocated memory dictated by parameter "len + strlen(displaySize_string)"
56056 alloc_strlen: Allocated memory does not have space for the terminating NUL of the string
56057 var_assign: Assigned "ptr->mon_comment" to storage allocated by "realloc(ptr->mon_comment, len + strlen(displaySize_string))"
56058
56059 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
56060 Reviewed-by: Matt Turner <mattst88@gmail.com>
56061 Signed-off-by: Keith Packard <keithp@keithp.com>
56062
56063commit aeae6637b52b042cf8b07c6943abcd5d31968d49
56064Author: Oliver McFadden <oliver.mcfadden@nokia.com>
56065Date: Mon Mar 15 23:46:34 2010 +0200
56066
56067 fb: fbFinishScreenInit: leaked_storage: Variable "(visuals|depths)" goes out of scope
56068
56069 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
56070 Reviewed-by: Matt Turner <mattst88@gmail.com>
56071 Signed-off-by: Keith Packard <keithp@keithp.com>
56072
56073commit d9ee31ab05efbd2d598cc93a41a8834f84d30803
56074Author: Oliver McFadden <oliver.mcfadden@nokia.com>
56075Date: Mon Mar 15 23:46:33 2010 +0200
56076
56077 exa: exaFinishAccess: Overrun of static array "pExaScr->access" of size 6 at position 6 with index variable "i"
56078
56079 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
56080 Reviewed-by: Keith Packard <keithp@keithp.com>
56081 Signed-off-by: Keith Packard <keithp@keithp.com>
56082
56083commit bf181915e103b0659c5a0bc3b2f6bccb18c9ec36
56084Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
56085Date: Mon Feb 1 17:11:51 2010 +0000
56086
56087 Cygwin/X: Allow the default log location to be configurable
56088
56089 Allow the default log location to be configurable (e.g. /var/log),
56090 and use separate logs for each display instance (e.g. XWin.0.log).
56091
56092 Make the type of g_pszLogFile const char*, per os/log.c:LogInit().
56093
56094 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
56095 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
56096
56097commit cca4952d750779a4c58a11fe08a53336042930bb
56098Author: Colin Harrison <colin.harrison@virgin.net>
56099Date: Fri Feb 26 14:40:30 2010 +0000
56100
56101 Xming: Warning fixes
56102
56103 Fix warnings due to prototypes not specifying function arguments
56104 Fix warning with RegQueryValueEx()
56105 Tidy up an include
56106
56107 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
56108 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
56109
56110commit b8cf4153f84404b2a1b62a247d5d36d24bf481da
56111Author: Colin Harrison <colin.harrison@virgin.net>
56112Date: Fri Feb 26 14:41:00 2010 +0000
56113
56114 Xming: Dead code removal
56115
56116 Remove some dead code, mostly code made obsolete by mandatory XKB
56117
56118 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
56119 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
56120
56121commit 9b18f7ac0ba9aadb8c86bc2717cbdff75e951b69
56122Author: Jon TURNEY <jon.turney@dronecode.org.uk>
56123Date: Fri Feb 26 14:22:57 2010 +0000
56124
56125 Cygwin/X: Tidy up some cosmetic issues in log strings
56126
56127 Tidy up some cosmetic issues in log strings:
56128 - Add missing '\n'
56129 - Fix some strings starting with '\n'
56130 - Remove '\f' from some log strings
56131
56132 These all just look daft in a log with timestamps.
56133
56134 Also clarify log message about screen origin coordinates
56135
56136 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
56137 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
56138
56139commit 178e830378c3514b9c53a5b1c6d2d5f930c3779b
56140Author: Jon TURNEY <jon.turney@dronecode.org.uk>
56141Date: Mon Feb 22 19:13:30 2010 +0000
56142
56143 Cygwin/X: Fix thinko in mount option checking
56144
56145 Fix a thinko in mount option checking.
56146 Use symbolic names for values assigned to binary flag for clarity.
56147
56148 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
56149 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
56150
56151commit fa5103a02bd509e4a102afdad2ab26cb22210367
56152Author: Francisco Jerez <currojerez@riseup.net>
56153Date: Wed Feb 24 23:18:01 2010 +0100
56154
56155 dri2: No need to blit from front on DRI2GetBuffers if they're just being reused.
56156
56157 It can be quite an expensive operation, so we're better off not doing
56158 it unless it's totally required.
56159
56160 Signed-off-by: Francisco Jerez <currojerez@riseup.net>
56161 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
56162 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
56163
56164commit e38e01081af42825bb7b44d18aa35845318f8556
56165Author: Francisco Jerez <currojerez@riseup.net>
56166Date: Wed Feb 24 23:18:00 2010 +0100
56167
56168 Import linked list helpers from the intel DDX.
56169
56170 Borrowed from i830.h, except for list_for_each_entry().
56171
56172 Signed-off-by: Francisco Jerez <currojerez@riseup.net>
56173 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
56174 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
56175
56176commit b8a3be5f34016b745e38cd53825a4e398c6127bc
56177Author: Francisco Jerez <currojerez@riseup.net>
56178Date: Wed Feb 24 23:17:59 2010 +0100
56179
56180 Add a ConfigNotify hook.
56181
56182 Executed from the ConfigureWindow request, right before sending
56183 ConfigureNotify to the clients.
56184
56185 This commit breaks the ScreenRec ABI.
56186
56187 Signed-off-by: Francisco Jerez <currojerez@riseup.net>
56188 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
56189 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
56190
56191commit 67a8c659f25218904bae64aac6e98e326c90330b
56192Author: Roland Scheidegger <sroland@vmware.com>
56193Date: Fri Mar 12 16:53:30 2010 +0100
56194
56195 hw/xfree86: move reference counting out of the UseHWCursor[ARGB] functions
56196
56197 The problem is that the xf86_use_hw_cursor(_argb) functions may get this
56198 correctly now, some drivers will replace these generic versions with their
56199 own functions. It is pretty insane to expect them to do reference counting
56200 of the cursor (as an example, look at driver/xf86-video-vmware to see how
56201 that looks like as a workaround). There are even places in xserver itself
56202 which replace these two functions.
56203 The segfaults if no reference counting is done are caused because the
56204 reference count of the cursor reached zero, hence the cursor was freed,
56205 however xf86CursorEnableDisableFBAccess() brought it back to life from
56206 the dead (from the SavedCursor).
56207 This patch hence adds reference counting in xf86CursorSetCursor. As per Michel
56208 Daenzer's suggestion, also free the cursor upon xf86CursorCloseScreen.
56209 In theory with this it should be possible to remove the reference
56210 counting in the UseHwCursor functions I think, though it should also be
56211 safe to keep them.
56212
56213 Signed-off-by: Roland Scheidegger <sroland@vmware.com>
56214 Reviewed-by: Michel Dänzer <michel@daenzer.net>
56215 Signed-off-by: Keith Packard <keithp@keithp.com>
56216
56217commit 75efb46a14fe45ffe73faff637b1fa6d017e1e52
56218Author: Roland Scheidegger <sroland@vmware.com>
56219Date: Fri Mar 12 16:52:56 2010 +0100
56220
56221 hw/xfree86: fix refcounting in xf86_use_hw_cursor
56222
56223 This is the same fix as was done in
56224 fcdc1d78cca3b8bb6b77d53eda7e21d649df6943 for xf86_use_hw_cursor_argb.
56225
56226 Signed-off-by: Roland Scheidegger <sroland@vmware.com>
56227 Reviewed-by: Michel Dänzer <michel@daenzer.net>
56228 Signed-off-by: Keith Packard <keithp@keithp.com>
56229
56230commit df9b6f16b27398545cd4cff8a56dd59a3813351d
56231Merge: 5f169f5 5172253
56232Author: Keith Packard <keithp@keithp.com>
56233Date: Mon Mar 15 08:26:58 2010 -0700
56234
56235 Merge remote branch 'jeremyhu/master'
56236
56237commit 5f169f54936c9868ad0f3778cb95c1f35eef41ea
56238Author: Gaetan Nadon <memsize@videotron.ca>
56239Date: Fri Mar 5 10:35:54 2010 -0500
56240
56241 XQuartz: remove undefined XSERVER_CFLAGS variable
56242
56243 This is a variable local to configure.ac which is not AC_SUBST()
56244 It is undefined in any generated Makefile.
56245
56246 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
56247 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
56248 Signed-off-by: Keith Packard <keithp@keithp.com>
56249
56250commit 5172253bae3b9867118c6717434e73c173acd5e9
56251Author: Jeremy Huddleston <jeremyhu@apple.com>
56252Date: Wed Feb 10 15:52:14 2010 -0800
56253
56254 XQuartz: Use an empty xkb keymap by default
56255
56256 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
56257
56258commit b117bc7a441bec8f61610fb384d747112f73d236
56259Author: Jeremy Huddleston <jeremyhu@apple.com>
56260Date: Thu Mar 11 00:06:01 2010 -0800
56261
56262 XQuartz: Include os.h for OsAbort()
56263
56264 Fixes regression from 5b9a52be7e975e59e0bbc6b43539ecaff96b2ecd
56265
56266 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
56267
56268commit b7a16117c6d87a9d33a5f682b592b4507f2c065e
56269Author: Jeremy Huddleston <jeremyhu@apple.com>
56270Date: Thu Mar 11 00:03:08 2010 -0800
56271
56272 XQuartz: GLX: Fix prototype for swapBuffers
56273
56274 This was a regression introduced by 04a54f69a8085ab3fe11a8713bd8b6b16ed1db27
56275
56276 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
56277
56278commit 178da6534fe7ab4b99fb87925c04e6f963c88583
56279Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56280Date: Mon Mar 8 16:33:37 2010 -0500
56281
56282 Xext: Link to external libraries when necessary.
56283
56284 Although the DDX should be linked to the necessary libraries, we may
56285 also need to pull them in directly to the module to ensure the symbols
56286 are resolved at runtime. Should fix this bug with XSELINUX:
56287
56288 /usr/bin/X: symbol lookup error:
56289 /usr/lib64/xorg/modules/extensions/libextmod.so: undefined symbol:
56290 is_selinux_enabled
56291
56292 -v2: use _LIBADD instead of _LIBS; remove SELINUX_LIBS from
56293 XSERVER_SYS_LIBS as it should only be needed in extmod.
56294
56295 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
56296 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56297 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56298
56299commit a2ea8c2f2cc53607d57d60f41c879380ea70cd02
56300Author: Daniel Stone <daniel@fooishbar.org>
56301Date: Thu Mar 11 14:19:04 2010 +0200
56302
56303 Record: Avoid duplicates from replaying frozen events
56304
56305 Reintroduce a check which used to be there in the old
56306 ProcessKeyboardEvent/ProcessPointerEvent codepath, which avoids us
56307 recording events subject to a grab twice: once when it's first processed
56308 in EnqueueEvent, and then again when it's thawed and being replayed.
56309
56310 This required a tiny amount of code motion to expose syncEvents.
56311
56312 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
56313 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
56314 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56315
56316commit f2eacb4646beb25d055de22868f93e6b24f229b6
56317Author: Peter Hutterer <peter.hutterer@who-t.net>
56318Date: Wed Mar 10 09:39:05 2010 +1000
56319
56320 Revert "dix: Use DeliverGrabbedEvent for implicit passive grabs (#25400)"
56321
56322 Several users have pointed out that this commit introduces regressions, most
56323 notably perhaps fluxbox which essentially stops working after a few clicks.
56324
56325 This reverts commit cf72b5437d2d620521279077a29c5df6d0fbb576.
56326
56327 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56328
56329commit df9f3273041c6c3e0da2d2254e8c156cd582e296
56330Author: Peter Hutterer <peter.hutterer@who-t.net>
56331Date: Wed Mar 10 09:24:32 2010 +1000
56332
56333 xfree86: fix xf86Config.c build error in --enable-debug mode. (#26971)
56334
56335 xf86Config.c: In function 'configInputDevices':
56336 xf86Config.c:1514: error: request for member 'lay_identifier' in something
56337 not a structure or union
56338 make[5]: *** [xf86Config.lo] Error 1
56339
56340 Introduced with e1165632bdfbd720889ed1adf5f7ab338032c0ee.
56341
56342 X.Org Bug 26971 <http://bugs.freedesktop.org/show_bug.cgi?id=26971>
56343
56344 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56345 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
56346
56347commit 116068103282924ef1476231d13d54719a797252
56348Author: Peter Hutterer <peter.hutterer@who-t.net>
56349Date: Tue Mar 9 09:35:23 2010 +1000
56350
56351 xfree86: don't warn about nonexisting core pointer/keyboard in config.
56352
56353 In the vast majority of cases there is no xorg.conf that specifies a core
56354 pointer/keyboard. Skip this warning, since we'll get another notification
56355 about how the server relies on the config backend for input devices anyway.
56356
56357 Leave the warning in for the error case (AEI off).
56358
56359 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56360 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
56361 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
56362
56363commit c39c8ece9bd3ce025549f4bac7794f5a62266f4a
56364Author: Peter Hutterer <peter.hutterer@who-t.net>
56365Date: Wed Feb 10 15:06:52 2010 +1000
56366
56367 os: remove INTERNAL_MALLOC define.
56368
56369 There doesn't seem to be anything that defines it and given that the
56370 counterpart (the X internal malloc) was removed in
56371 01cfba75229f4b9bf1e4fe80814931acdacde14c it's unlikely to work anyway.
56372
56373 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56374
56375commit 9f462ff9082634719e64d2d8d4dbd09ec7deaf2e
56376Author: Peter Hutterer <peter.hutterer@who-t.net>
56377Date: Fri Feb 26 14:02:09 2010 +1000
56378
56379 dix: Clip only into axis ranges if we're in absolute mode. (#26543)
56380
56381 An absolute device in relative mode may provide valuators outside of the
56382 axis range. Clipping back into the range prevents screen crossings in a
56383 multi-screen (Xinerama) setup as the required screen edge for crossing is
56384 never met: miPointerSetPosition crosses the screen conditional to the X
56385 coordinate being equal to the screen width or _less than_ 0. While the
56386 former can be met when clipping into the coordinate range and scaling, the
56387 latter cannot, resulting in a mouse pointer that gets stuck on the rightmost
56388 screen.
56389
56390 This patch only applies axis clipping for valuators in mode Absolute. If
56391 relative, we allow the values to get above/below the axis ranges. Doesn't
56392 matter, miPointerSetPosition will reset the values to the allowed range even
56393 if no screen was crossed.
56394 This leads to interesting values provided to clients, the valuator range of
56395 the device resets once a screen is crossed and essentially reflects
56396 the position of the cursor on the screen - scaled into the valuator range.
56397 The values themselves are valid given the range though.
56398
56399 In theory, the XI1 specs require that a relative device has a min/max range
56400 of 0/0. This doesn't really go well with devices that actually can switch
56401 mode between relative and absolute since they would have to reset their axis
56402 range when switching. If multiple XI clients are in use, we have no method
56403 of notifying them about the changes, so other clients may continue to use
56404 the wrong axis ranges (note: XI1 wasn't really designed to have multiple
56405 clients use a device). Expecting all relative devices to have this min/max
56406 of 0 is unrealistic at this point.
56407
56408 So pick what is possibly the lesser of all evils, pass the beer and despair.
56409
56410 X.Org Bug 26543 <http://bugs.freedesktop.org/show_bug.cgi?id=26543>
56411
56412 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56413 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
56414
56415commit bbae92795c7eab062e6722c42fa7915e0cee5d69
56416Author: Matt Turner <mattst88@gmail.com>
56417Date: Mon Feb 15 20:08:09 2010 -0500
56418
56419 Replace assembly with generic unaligned access code
56420
56421 Removes Alpha assembly, and probably works around unaligned accesses on
56422 other sensitive platforms.
56423
56424 Signed-off-by: Matt Turner <mattst88@gmail.com>
56425 Acked-by: Adam Jackson <ajax@redhat.com>
56426 Compiled-by: Tiago Vignatti <tiago.vignatti@nokia.com>
56427 Signed-off-by: Keith Packard <keithp@keithp.com>
56428
56429commit 1c2abec479cfbac9aeadc121af0825d368317658
56430Author: Peter Hutterer <peter.hutterer@who-t.net>
56431Date: Fri Feb 26 11:44:17 2010 +1000
56432
56433 configure: move SELINUX_LIBS to XSERVER_SYS_LIBS
56434
56435 All ddxs require linking against selinux if enabled.
56436
56437 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56438 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
56439 Reviewed-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56440 Signed-off-by: Keith Packard <keithp@keithp.com>
56441
56442commit cab99b947c66a42dbf8a906bdb02e23aa7cab519
56443Author: Peter Hutterer <peter.hutterer@who-t.net>
56444Date: Fri Feb 26 11:41:38 2010 +1000
56445
56446 configure: move libselinux requirement up to the common section.
56447
56448 Let's have all version-specific requirements in one block.
56449
56450 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56451 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
56452 Reviewed-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56453 Signed-off-by: Keith Packard <keithp@keithp.com>
56454
56455commit 9e7b6e565f6e69f20002bf03ed05fe8b63f86924
56456Author: Peter Hutterer <peter.hutterer@who-t.net>
56457Date: Fri Feb 26 10:14:00 2010 +1000
56458
56459 dmx: fix OsAbort()-related build error in examples/xinput.c
56460
56461 5b9a52be7e975e59e0bbc6b43539ecaff96b2ecd changed the server to use OsAbort()
56462 instead of abort(). xinput in dmx is a client program though and fails to
56463 link if it tries to use OsAbort(). Switch it back to using abort().
56464
56465 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56466 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
56467 Reviewed-by: Julien Cristau <jcristau@debian.org>
56468 Reviewed-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
56469 Signed-off-by: Keith Packard <keithp@keithp.com>
56470
56471commit cf72b5437d2d620521279077a29c5df6d0fbb576
56472Author: Peter Hutterer <peter.hutterer@who-t.net>
56473Date: Thu Feb 25 12:49:21 2010 +1000
56474
56475 dix: Use DeliverGrabbedEvent for implicit passive grabs (#25400)
56476
56477 A client requesting a GrabModeSync button grab, owner-events true, with only
56478 the ButtonRelease mask set would never receive the press event even if the
56479 grab window had the ButtonPress mask set.
56480
56481 The protocol requires that if owner-events is true, then the delivery mask
56482 is the combination of the grab mask + the window event mask.
56483
56484 DeliverGrabbedEvents does this already for us, checking first the delivery
56485 based on owner_events and then based on the grab mask. AFAICT, the device
56486 cannot enter the states FREEZE_BOTH_NEXT_EVENT or FREEZE_NEXT_EVENT that
56487 would be handled by DGE in any possible path here.
56488
56489 Bonus point - CheckPassiveGrabsOnWindows suddenly becomes a lot lesss
56490 complicated.
56491
56492 X.Org Bug 25400 <http://bugs.freedesktop.org/show_bug.cgi?id=25400>
56493
56494 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56495 Reviewed-by: Keith Packard <keithp@keithp.com>
56496 Signed-off-by: Keith Packard <keithp@keithp.com>
56497
56498commit de86a3a3448f0a55c1cd99aee9ea80070a589877
56499Author: Keith Packard <keithp@keithp.com>
56500Date: Thu Feb 25 11:37:05 2010 -0800
56501
56502 Allow for missing or disabled compat_output
56503
56504 When the compat output is missing (I don't think this is actually
56505 possible), or is disabled (and hence has no crtc), we would like to
56506 avoid dereferencing NULL pointers. This patch creates inline functions
56507 to extract the current compat output, crtc or associated RandR crtc
56508 structure, carefully checking for NULL pointers everywhere.
56509
56510 Reviewed-by: Adam Jackson <ajax@redhat.com>
56511 Signed-off-by: Keith Packard <keithp@keithp.com>
56512
56513commit fbbadca7e88391e81ab0f470290f5eec36aa9ce7
56514Author: Keith Packard <keithp@keithp.com>
56515Date: Thu Feb 25 11:35:03 2010 -0800
56516
56517 Share enum definition for det_monrec_parameter sync_source
56518
56519 There were two separate enum definitions, one inside
56520 det_monrec_parameter struct and one for a local variable (which was then
56521 stored inside the struct). Sharing a single definition makes the
56522 code more obviously correct while making the compiler happier.
56523
56524 Reviewed-by: Adam Jackson <ajax@redhat.com>
56525 Signed-off-by: Keith Packard <keithp@keithp.com>
56526
56527commit 4b55b2cf8a52c39b53bae11cd1bc7314481d4c86
56528Author: Robert Bragg <bob@o-hand.com>
56529Date: Wed Jan 27 20:38:38 2010 +0000
56530
56531 DRI2: initialize event->drawable in DRI2SwapEvent
56532
56533 We weren't initialising the drawable in the event structure so the
56534 client side DRI2WireToEvent used for translating the event into a GLX
56535 event wouldn't be able to lookup up the corresponding GLXDrawable before
56536 passing the event on.
56537
56538 Signed-off-by: Robert Bragg <robert@linux.intel.com>
56539 Reviewed-by: Eric Anholt <eric@anholt.net>
56540 Signed-off-by: Keith Packard <keithp@keithp.com>
56541
56542commit 780c95caf9888fa4548dfe4c1c78a7e7ce99a9ed
56543Merge: 018b177 758f697
56544Author: Keith Packard <keithp@keithp.com>
56545Date: Wed Feb 24 09:59:19 2010 -0800
56546
56547 Merge remote branch 'whot/for-keith'
56548
56549commit 018b177591c9fade6d065e31858cc6e054d33eff
56550Author: Kristian Høgsberg <krh@bitplanet.net>
56551Date: Wed Feb 24 10:37:26 2010 -0500
56552
56553 glx: Compile fix to let server compile with new and old mesa
56554
56555 We broke the __DRI2_FLUSH API since it was never released, but since it's
56556 taking a little longer than expected to get the X server side of the changes
56557 ready, fix things up so it compiles.
56558
56559 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
56560 Reviewed-by: Keith Packard <keithp@keithp.com>
56561 Signed-off-by: Keith Packard <keithp@keithp.com>
56562
56563commit 758f6971750ed507e64eee817d720a77181439f2
56564Author: Peter Hutterer <peter.hutterer@who-t.net>
56565Date: Mon Feb 15 16:32:16 2010 +1000
56566
56567 dix: try to ring the bell even if the current device doesn't have one. (#24503)
56568
56569 Evdev devices do not have the bell proc set, but XTEST devices do. By
56570 exiting early, the bell only rings if the last keyboard used was the XTEST
56571 keyboard and hence the bell proc is still set on the master but not if an
56572 evdev keyboard was used last.
56573
56574 The better approach here is to try to ring the bell on all devices attached
56575 to this master device in case one or more actually do produce an audible
56576 sound. That's also XKB's behaviour if XkbUseCoreKbd is specified as device
56577 identifier.
56578
56579 X.Org Bug 24503 <http://bugs.freedesktop.org/show_bug.cgi?id=24503>
56580
56581 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56582 Reviewed-by: Adam Jackson <ajax@redhat.com>
56583
56584commit c0d941913199ed251fceab62fcf65437f6f9bd31
56585Author: Peter Hutterer <peter.hutterer@who-t.net>
56586Date: Thu Feb 18 16:00:15 2010 +1000
56587
56588 dix: remove now-erroneous comment about frozen slave devices.
56589
56590 A direct grab on a slave device through XI2 detaches it, regardless of
56591 whether the grab is sync or async. So this comment doesn't apply to XI2
56592 anyway.
56593
56594 For XI1, aside from your life being miserable already, it doesn't matter as
56595 XI1 does not have a concept of attachment. You can freeze a device and if
56596 you don't freeze _all_ other devices at the same time, the master device can
56597 still happily send events to the client.
56598
56599 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56600 Acked-by: Adam Jackson <ajax@redhat.com>
56601
56602commit db4f676f25c6d8e58263d5151942be730592d444
56603Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56604Date: Wed Feb 3 15:40:47 2010 -0500
56605
56606 xselinux: Bump extension minor version.
56607
56608 Changes introduced in this version:
56609 - 3 window-related requests now handle pixmaps also.
56610
56611 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56612 Reviewed-by: Keith Packard <keithp@keithp.com>
56613
56614commit a6cbcf6e782e65f3d4d14202ed0b7a8f65b33c9d
56615Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56616Date: Wed Feb 3 15:38:57 2010 -0500
56617
56618 xselinux: Rename window-related requests that now support pixmaps.
56619
56620 Renamed requests:
56621 SetWindowCreateContext -> SetDrawableCreateContext
56622 GetWindowCreateContext -> GetDrawableCreateContext
56623 GetWindowContext -> GetDrawableContext
56624
56625 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56626 Reviewed-by: Keith Packard <keithp@keithp.com>
56627
56628commit ab68c707fd02613d9359120ba6402639dd8bd89e
56629Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56630Date: Wed Feb 3 15:08:40 2010 -0500
56631
56632 xselinux: Allow GetWindowContext to be used for pixmaps as well.
56633
56634 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56635 Reviewed-by: Keith Packard <keithp@keithp.com>
56636
56637commit eb9210097efea81065c301e5b6b4da7a566deb4a
56638Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56639Date: Wed Jan 6 12:52:51 2010 -0500
56640
56641 xselinux: Remove reference counting calls for SID objects.
56642
56643 Starting with libselinux 2.0.86, SID objects are no longer
56644 reference counted and the sidput() and sidget() calls are no-ops.
56645
56646 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56647 Reviewed-by: Keith Packard <keithp@keithp.com>
56648
56649commit 0c4b75a08922416050bb73d2e279ce938d3823e7
56650Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56651Date: Wed Oct 21 21:59:01 2009 -0400
56652
56653 Don't print a failure message when XACE denies an input event delivery.
56654
56655 A denial is normal and the behavior should be to drop the event.
56656 Having the log message creates excessive log spam.
56657
56658 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56659 Reviewed-by: Keith Packard <keithp@keithp.com>
56660
56661commit e687e11b1248494a45b9a4e4c321675526808703
56662Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56663Date: Wed Oct 21 21:46:50 2009 -0400
56664
56665 Revert "Remove some debug messages that trigger on XACE event delivery failure."
56666
56667 The log messages still need to be there for non-XACE failures.
56668
56669 This reverts commit 4be354c4c2da5168b302601b91bd80cfaca7e193.
56670
56671 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56672 Reviewed-by: Keith Packard <keithp@keithp.com>
56673
56674commit 2ecb9f7d86117f212d7845741a3e155553954a85
56675Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56676Date: Wed Oct 21 19:28:26 2009 -0400
56677
56678 libselinux now has a pkgconfig file. Use it.
56679
56680 Also remove HAVE_NETLINK_AVC_ACQUIRE_FD tests, because we now
56681 require a version of libselinux that has it.
56682
56683 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56684 Reviewed-by: Keith Packard <keithp@keithp.com>
56685
56686commit 9fc6b59d46382e949526dc10c3f5bfddbe2acb92
56687Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
56688Date: Thu Oct 15 17:32:21 2009 -0400
56689
56690 xselinux: Allow SetWindowCreateContext to be used for pixmaps as well.
56691
56692 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
56693 Reviewed-by: Keith Packard <keithp@keithp.com>
56694
56695commit 93ff1bc5884cdfca1132af7fbc06c48e06f218c8
56696Author: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
56697Date: Mon Feb 15 14:18:00 2010 +0200
56698
56699 Always enable outputs that have been forced on in the configuration file
56700
56701 If the user has gone to the effort of manually enabling an output in
56702 the configuration file assume that they know what they're doing.
56703
56704 X.org Bug 14611 <http://bugs.freedesktop.org/show_bug.cgi?id=14611>
56705
56706 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
56707 Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
56708 Signed-off-by: Keith Packard <keithp@keithp.com>
56709
56710commit ca9c2472d74c7107ccc117e2c9c723c168d2eccf
56711Merge: e40ba57 02d1116
56712Author: Keith Packard <keithp@keithp.com>
56713Date: Mon Feb 22 13:26:06 2010 -0800
56714
56715 Merge remote branch 'whot/for-keith'
56716
56717commit e40ba5798c367bec584f6437b23d2c5f801fd013
56718Author: Peter Hutterer <peter.hutterer@who-t.net>
56719Date: Thu Feb 18 15:49:09 2010 +1000
56720
56721 Xi: assume BadMode for non-BadMatch errors returned from SetDeviceMode.
56722
56723 The XI protocol spec only allows for two errors on the SetDeviceMode
56724 requests: BadMatch or BadMode. BadMode however is a dynamically assigned
56725 extension error and the driver doesn't have access to the actual error
56726 number. Hence, if a SetDeviceMode driver returns an error other than
56727 BadMatch, assume BadMode.
56728
56729 The two exceptions are BadAlloc and BadImplementations, pass these on to the
56730 client (any request is allowed to return either of those).
56731
56732 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56733 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
56734 Signed-off-by: Keith Packard <keithp@keithp.com>
56735
56736commit 02d1116e7eb84068fa32477f1640d29b0a81b638
56737Author: Peter Hutterer <peter.hutterer@who-t.net>
56738Date: Mon Jan 11 16:35:36 2010 +1000
56739
56740 Xi: assume BadMode for non-BadMatch errors returned from SetDeviceMode.
56741
56742 The XI protocol spec only allows for two errors on the SetDeviceMode
56743 requests: BadMatch or BadMode. BadMode however is a dynamically assigned
56744 extension error and the driver doesn't have access to the actual error
56745 number. Hence, if a SetDeviceMode driver returns an error other than
56746 BadMatch, assume BadMode.
56747
56748 The two exceptions are BadAlloc and BadImplementations, pass these on to the
56749 client (any request is allowed to return either of those).
56750
56751 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56752 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
56753
56754commit 122fc0e7a0712df4ec2bd9ca6773f90bcd2095cf
56755Author: Chris Dekter <cdekter@gmail.com>
56756Date: Thu Feb 11 16:34:09 2010 +1000
56757
56758 Re-enable RECORD extension.
56759
56760 RECORD was disabled during the switch to internal events. This patch
56761 modifies the record callback to work with internal events instead of
56762 xEvents. The InternalEvents are converted to core/Xi events as needed.
56763
56764 Since record is a loadable extension, the EventTo* calls must be externed.
56765
56766 Signed-off-by: Chris Dekter <cdekter@gmail.com>
56767 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
56768 Reviewed-by: Adam Jackson <ajax@redhat.com>
56769 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56770
56771commit 03ef00ae79687046dc9f9987857062a16ff2c2e0
56772Author: Michel Dänzer <daenzer@vmware.com>
56773Date: Wed Feb 17 11:40:44 2010 +0100
56774
56775 EXA/mixed: Clean up exaPrepareAccessReg_mixed() a little.
56776
56777 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
56778 Acked-by: Maarten Maathuis <madman2003@gmail.com>
56779 Signed-off-by: Keith Packard <keithp@keithp.com>
56780
56781commit 9623daad7c864d46b0dfcbb6e689820b96cc8115
56782Author: Maarten Maathuis <madman2003@gmail.com>
56783Date: Wed Feb 17 11:40:43 2010 +0100
56784
56785 exa/mixed: fix gnome-panel corruption
56786
56787 - A mapped pixmap can't be used for acceleration, any decent memory manager
56788 will refuse this.
56789 - Source pixmaps migrated with a bounding region are incomplete (from the
56790 gpu point of view), so do the upload unconditionally, instead of just for
56791 deferred destination pixmaps.
56792 - Fixes fd.o bug #26076.
56793
56794 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
56795 Acked-by: Michel Dänzer <michel@daenzer.net>
56796 Signed-off-by: Keith Packard <keithp@keithp.com>
56797
56798commit 5b9a52be7e975e59e0bbc6b43539ecaff96b2ecd
56799Author: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
56800Date: Thu Jan 28 11:08:34 2010 +0200
56801
56802 os: Prevent core dump from being truncated.
56803
56804 The problem fixed by this patch can be reproduced on Linux with the
56805 following steps.
56806 - Access NULL pointer intentionally in ProcessOtherEvent on key press.
56807 - Instead of saving core dump to a file, write it into a pipe.
56808 echo "|/usr/sbin/my-core-dumper" > /proc/sys/kernel/core_pattern
56809 - Dump the core by pressing a key.
56810
56811 While the core is being dumped into the pipe, the smart schedule timer
56812 will cause a pending SIGALRM. Linux kernel stops writing data to the
56813 pipe when there are pending signals. This causes the core dump to be
56814 truncated. On my system I'm expecting a 6 MB dump but the size will be
56815 60 kB instead. The problem is solved if we block the SIGALRM caused by
56816 expired smart schedule timer.
56817
56818 I haven't been able to reproduce this problem in the following cases.
56819 - Save core dump to a file instead of a pipe.
56820 - kill -SEGV `pidof Xorg`
56821 - Press a key to dump core while gdb is attached to Xorg.
56822 - Give option -dumbSched to Xorg.
56823
56824 Also note that the fix works only when NoTrapSignals has the default
56825 value FALSE. The problem can still be reproduced if error signals
56826 aren't trapped. In addition to pending SIGALRM, there is a similar
56827 problem with pending SIGIO from the keyboard driver during core dump.
56828
56829 Signed-off-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
56830 Reviewed-by: Keith Packard <keithp@keithp.com>
56831 Signed-off-by: Keith Packard <keithp@keithp.com>
56832
56833commit ca364ca82a760d8e5347a6f9f79636c9a5e4e03f
56834Author: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
56835Date: Thu Jan 28 11:08:33 2010 +0200
56836
56837 os: Introduce OsAbort for proper core dumps.
56838
56839 Signed-off-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
56840 Reviewed-by: Keith Packard <keithp@keithp.com>
56841 Signed-off-by: Keith Packard <keithp@keithp.com>
56842
56843commit 031f92bf9ab15226df410012a0d1c9c390efc36d
56844Author: Oliver McFadden <oliver.mcfadden@nokia.com>
56845Date: Wed Feb 3 10:05:20 2010 +0200
56846
56847 parser: corrected xf86getBoolValue to use case insensitive compare
56848
56849 commit c6e8637e29e0ca11dfb35c02da7ca6002ac8c597 introduced this
56850 regression; it can cause existing config files to be parsed incorrectly.
56851
56852 Acked-by: Julien Cristau <jcristau@debian.org>
56853 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
56854 Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
56855 Signed-off-by: Keith Packard <keithp@keithp.com>
56856
56857commit be96fb2f02c13a6ee8aba40f7d4c3f9141f06cea
56858Author: Alan Coopersmith <alan.coopersmith@sun.com>
56859Date: Mon Feb 15 17:42:11 2010 -0800
56860
56861 Solaris xf86OSRingBell() off-by-one error in filling iov[] array
56862
56863 When generating sound buffers for /dev/audio bells, insert waveform
56864 for beep *or* silence, but not both, so we don't write one entry past
56865 the end of the iov buffer when the final bit of soundwave ends up in
56866 the final entry allocated in the iov array.
56867
56868 Fixes OpenSolaris bug 6894890:
56869 http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6894890
56870
56871 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
56872 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
56873 Acked-by: Adam Jackson <ajax@redhat.com>
56874 Signed-off-by: Keith Packard <keithp@keithp.com>
56875
56876commit 2d40f22d1ec970a84b23aa42b1feca4feedeb4bb
56877Author: Matt Turner <mattst88@gmail.com>
56878Date: Mon Feb 15 20:05:24 2010 -0500
56879
56880 Use C-style comments in x86emu
56881
56882 Signed-off-by: Matt Turner <mattst88@gmail.com>
56883 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
56884 Signed-off-by: Keith Packard <keithp@keithp.com>
56885
56886commit 4e8bf12b13690afa6d9fee0e339d3819ef16fb3f
56887Author: Jeremy Huddleston <jeremyhu@apple.com>
56888Date: Mon Feb 15 16:46:21 2010 -0800
56889
56890 XQuartz: Fix a possible buffer overrun in quartzAudio
56891
56892 Also dropped deprecated API while there
56893
56894 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
56895
56896commit 4127e8e5be5c366eb752bbb8d7f0ff5b519e641b
56897Author: Jeremy Huddleston <jeremyhu@apple.com>
56898Date: Wed Feb 10 17:56:30 2010 -0800
56899
56900 XQuartz: clang static analysis fixes
56901
56902 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
56903
56904commit c686c5cbdc5f35744407af5f9aabbd18cc17714b
56905Author: Jeremy Huddleston <jeremyhu@apple.com>
56906Date: Mon Feb 15 15:03:01 2010 -0800
56907
56908 XQuartz: Fix linking (CloseInput())
56909
56910 Fixes regresison from d33adcdf03c69407d151e732fa0cf9947151eb19
56911
56912 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
56913
56914commit 84905007702da2c05a4f7446b3fc5ff52be49655
56915Author: Thomas Jaeger <ThJaeger@gmail.com>
56916Date: Mon Jan 4 15:00:49 2010 -0500
56917
56918 udev: Don't filter subsystem "input"
56919
56920 This allows serial wacom devices to work, whose subsystem is "tty".
56921
56922 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
56923 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
56924 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56925
56926commit 8736d112afb0dd61dfdaadd6378eafd200b2ef5f
56927Author: Dan Nicholson <dbn.lists@gmail.com>
56928Date: Sat Feb 13 10:40:40 2010 -0800
56929
56930 xfree86: Reorder InputClass option priorities
56931
56932 Currently the config and InputClasses are merged together so that the
56933 options from the config backend have the highest priority. This is bad
56934 since it means options such as a default XKB layout set by the backend
56935 cannot be changed by the user.
56936
56937 This patch changes order of precedence to be:
56938
56939 1. xorg.conf
56940 2. xorg.conf.d (later files have higher priority)
56941 3. config backend
56942
56943 In order to allow this ordering, the config parsing has been changed to
56944 read the xorg.conf.d files before xorg.conf. This has the consequence
56945 that the core device picking which looks for the first InputDevice may
56946 not find it in xorg.conf.
56947
56948 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
56949 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
56950 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56951
56952commit 48f7298657f91843db36566b8d66d6c4c18dbd4c
56953Author: Peter Hutterer <peter.hutterer@who-t.net>
56954Date: Wed Jan 13 15:03:45 2010 +1000
56955
56956 Xi: reset the sli pointers after copying device classes. (#25640)
56957
56958 If the indicator flags have the XkbSLI_IsDefault bit set, the indicator map
56959 and names aren't their own bit of memory but rather point into the
56960 device->key->xkbInfo->desc structure. XkbCopySrvLedInfo knows about this and
56961 leaves the pointers alone.
56962
56963 When copying the classes from the slave to the master, these pointers are
56964 copied and still point to the dev->key class of the slave device. If the
56965 slave device is removed, the memory becomes invalid and a call to modify
56966 this data (e.g. XkbSetIndicators) may cause a deadlock.
56967
56968 The copying of dev->key relies on dev->kbdfeed to be already set up. Hence
56969 the pointers need to be reset once _both_ kbdfeed and key have been copied
56970 into the master device.
56971
56972 X.Org Bug 25640 <http://bugs.freedesktop.org/show_bug.cgi?id=25640>
56973 Fedora Bug 540584 <https://bugzilla.redhat.com/show_bug.cgi?id=540584>
56974
56975 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56976 Reviewed-by: Keith Packard <keithp@keithp.com>
56977
56978commit d33adcdf03c69407d151e732fa0cf9947151eb19
56979Author: Peter Hutterer <peter.hutterer@who-t.net>
56980Date: Fri Jan 15 15:22:02 2010 +1000
56981
56982 dix: move config_init into the DDX.
56983
56984 The only DDX currently using hotplugging is the xfree86 one and it looks
56985 like it'll stay that way for a bit. Move the initialization to the DDX,
56986 since Xephyr, Xnest, and friends don't need HAL or udev notifications.
56987
56988 Add CloseInput (counterpart to InitInput) to be able to clean up the config
56989 initialization from the DDX as well.
56990
56991 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
56992 Reviewed-by: Julien Cristau <jcristau@debian.org>
56993 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
56994
56995commit 001ce71dc11287dc94cc2fbc5d35677c046e6c04
56996Author: Julien Cristau <jcristau@debian.org>
56997Date: Fri Feb 12 23:34:57 2010 +0100
56998
56999 dix: restore lastDeviceEventTime update in dixSaveScreens
57000
57001 This was removed in 6b5978dcf1f7ac3ecc2f22df06f7000f360e2066 (Do not
57002 reset lastDeviceEventTime when we do dixSaveScreens), but caused a
57003 regression for XResetScreenSaver. Add the lastDeviceEventTime update
57004 back, but restrict it to that case.
57005
57006 X.Org bug#25855 <http://bugs.freedesktop.org/25855>
57007
57008 Reported-by: Lubos Lunak <l.lunak@suse.cz>
57009 Tested-by: Lubos Lunak <l.lunak@suse.cz>
57010 Signed-off-by: Julien Cristau <jcristau@debian.org>
57011 Signed-off-by: Keith Packard <keithp@keithp.com>
57012
57013commit 97b03037f4d99fcebc7603011f41c3aff9871ce2
57014Author: Peter Harris <pharris@opentext.com>
57015Date: Fri Feb 12 15:36:30 2010 -0500
57016
57017 Don't double-swap the RandR PropertyNotify event
57018
57019 The event is already swapped in randr.c/SRROutputPropertyNotifyEvent, so
57020 it should not be swapped here.
57021
57022 X.Org Bugzilla #26511: http://bugs.freedesktop.org/show_bug.cgi?id=26511
57023
57024 Tested-by: Leonardo Chiquitto <leonardo@ngdn.org>
57025 Acked-by: Adam Jackson <ajax at redhat.com>
57026 Reviewed-by: Julien Cristau <jcristau at debian.org>
57027 Signed-off-by: Peter Harris <pharris@opentext.com>
57028 Signed-off-by: Keith Packard <keithp@keithp.com>
57029
57030commit 0b73f98cbd1803b82c2faf1a40f1a0037673d712
57031Author: Soeren Sandmann <sandmann@daimi.au.dk>
57032Date: Fri Feb 12 20:04:53 2010 +0100
57033
57034 xfree86: Add qxl driver to the autoconfig logic
57035
57036 The qxl driver is for the QXL virtualized graphics device.
57037
57038 Signed-off-by: Søren Sandmann Pedersen <ssp@redhat.com>
57039 Reviewed-by: Adam Jackson <ajax@redhat.com>
57040 Signed-off-by: Keith Packard <keithp@keithp.com>
57041
57042commit 9c5bb550a2234e4bee2a144417d74c6bdf160e89
57043Author: Keith Packard <keithp@keithp.com>
57044Date: Fri Feb 12 13:46:03 2010 -0800
57045
57046 Bump to 1.7.99.901 -- 1.8 RC1
57047
57048commit 57ffeb3f2b3313dcef92a396f1f55fdbc064b2c5
57049Merge: c6d9bc0 c76248f
57050Author: Keith Packard <keithp@keithp.com>
57051Date: Fri Feb 12 13:36:32 2010 -0800
57052
57053 Merge remote branch 'jturney/master'
57054
57055commit c6d9bc092c84ad5c68083a126aa7577baa42cef7
57056Author: Peter Hutterer <peter.hutterer@who-t.net>
57057Date: Wed Feb 10 15:36:50 2010 +1000
57058
57059 Add tag matching to input attributes.
57060
57061 Tags may be a list of comma-separated strings that match against a MatchTag
57062 InputClass section. If any of the tags specified for a device match against
57063 the MatchTag of the section, this match is evaluated true and passed on to
57064 the next match condition.
57065
57066 Tags are specified as "input.tags" (hal) or "ID_INPUT.tags" (udev), the
57067 value of the tags is case-sensitive and require an exact match (not a
57068 substring match).
57069
57070 i.e. "quirk" will not match "QUIRK", "need_quirk" or "quirk_needed".
57071
57072 Example configuration:
57073 udev:
57074 ENV{ID_INPUT.tags}="foo,bar"
57075
57076 hal:
57077 <merge key="input.tags" type="string">foo,bar</merge>
57078
57079 xorg.conf:
57080 Section "InputClass"
57081 Identifier "foobar quirks"
57082 MatchTag "foo|foobar"
57083 Option "Foobar" "on"
57084 EndSection
57085
57086 Where the xorg.conf section matches against any device with the tag "foo"
57087 or tag "foobar" set.
57088
57089 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57090 Tested-by: Dan Nicholson <dbn.lists@gmail.com>
57091 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
57092
57093commit 3ac43df5d4a25d6e0058b327fa05a1c1436b4794
57094Author: Peter Hutterer <peter.hutterer@who-t.net>
57095Date: Wed Feb 10 15:36:49 2010 +1000
57096
57097 Add xstrtokenize to the dix.
57098
57099 Move tokenize out of the parser, make it a dix util function instead.
57100 Splitting a string into multiple substrings is useful by other places, so
57101 let's use it across the line. Future users include config/hal, config/udev
57102 and of course the parser.
57103
57104 Example usage:
57105 char **substrings = xstrtokenize(my_string, "\n");
57106
57107 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57108 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
57109
57110commit 27d1b86d1b858f931b4cb1b6ddf857c76d92a6d9
57111Author: Dan Nicholson <dbn.lists@gmail.com>
57112Date: Wed Feb 10 15:36:48 2010 +1000
57113
57114 xfree86: Set fnmatch pathname flag for InputClass device matching
57115
57116 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
57117 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57118 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57119
57120commit 9b369f71273fb117c982e6ce16cd4462f206d365
57121Author: Dan Nicholson <dbn.lists@gmail.com>
57122Date: Wed Feb 10 15:36:47 2010 +1000
57123
57124 xfree86: Allow multiple arguments to InputClass matches
57125
57126 In order to keep the number of InputClass sections manageable, allow
57127 matches to contain multiple arguments. The arguments will be separated
57128 by the '|' character. This allows a policy to apply to multiple types of
57129 devices. For example:
57130
57131 Section "InputClass"
57132 Identifier "Inverted Mice"
57133 MatchProduct "Crazy Mouse|Silly Mouse"
57134 Option "InvertX" "yes"
57135 EndSection
57136
57137 This applies to the MatchProduct, MatchVendor and MatchDevicePath
57138 entries. Currently there is no way to escape characters, so names or
57139 patterns cannot contain '|'.
57140
57141 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
57142 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57143 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57144
57145commit a378e361a5de89f0be8b68ebc3e854f56cefe666
57146Author: Dan Nicholson <dbn.lists@gmail.com>
57147Date: Wed Feb 10 15:36:46 2010 +1000
57148
57149 xfree86: Use "Ignore" option in InputClass to skip devices
57150
57151 Sometimes it is desirable to skip adding specific input devices to the
57152 server. The "Ignore" option is used similarly to Monitor sections so
57153 that matched devices will not be added. BadIDChoice is returned to the
57154 config backend so that it will clean up all resources.
57155
57156 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
57157 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57158 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57159
57160commit 8b1a685f00ae76be864cc188943a0874f48b8d64
57161Author: Dan Nicholson <dbn.lists@gmail.com>
57162Date: Wed Feb 10 15:36:45 2010 +1000
57163
57164 xfree86: Handle config files ending without newline
57165
57166 The config parser expects to find a newline at the end of each line, so
57167 files ending without one would confuse it. A newline is inserted at the
57168 end of the buffer in these situations. Additionally, switching to the
57169 next config file is moved to the higher level to allow parsing of the
57170 last line of the previous file to complete before shifting the index and
57171 resetting the line number.
57172
57173 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
57174 Tested-by: Stephan Raue<stephan.raue@gmx.net>
57175 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57176 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57177
57178commit 67bc278a511ca6ec42e1f8d2d5897c0109e94e2c
57179Author: Dan Nicholson <dbn.lists@gmail.com>
57180Date: Wed Feb 10 15:36:44 2010 +1000
57181
57182 xfree86: Make InputClass docs and comments match reality
57183
57184 Drivers and options specified in InputClass sections work on a "first
57185 match wins" strategy. Let's be consistent when documenting it.
57186
57187 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
57188 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57189 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57190
57191commit a34812b09000db2ff2a1dc6182602839123edd4e
57192Author: Benjamin Tissoires <tissoire@cena.fr>
57193Date: Mon Feb 8 13:50:43 2010 +0100
57194
57195 Add labels for multitouch valuators
57196
57197 Thoses definitions have been included in the kernel but the X server is not updated accordingly.
57198 Without these definitions, the multitouch axes are not correctly labelled.
57199
57200 Signed-off-by: Benjamin Tissoires <tissoire@cena.fr>
57201 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57202 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57203
57204commit c76248fda99c38aef0ccf0ed6b58fbe95f6fe497
57205Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57206Date: Tue Feb 2 20:53:44 2010 +0000
57207
57208 Cygwin/X: Avoid a collision between DEBUG and a token name
57209
57210 Rename a token to avoid a collision between DEBUG defined via AC_DEFINE
57211 if --enable-debug is configured, and the token for the 'debug' instruction
57212 in the XWin preferences file
57213
57214 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57215 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57216
57217commit b7dbbd3cd4d28cd4939706b3bf6394ba8ecafaff
57218Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57219Date: Sun Nov 22 23:35:09 2009 +0000
57220
57221 Cygwin/X: Avoid cursor size log spam
57222
57223 Fedora 12 likes to use a 39x26 animated wait cursor. Avoid spamming
57224 the log with warnings that each frame can't be completely contained
57225 in the 32x32 native cursor
57226
57227 Also reformat log message so it doesn't contain a '\n\t'. I mean,
57228 it's not like we might want to grep the log or something...
57229
57230 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57231 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57232
57233commit c9cbbd5d1cfa58a2d9f08e25534ea8439284322d
57234Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57235Date: Sat Nov 28 17:51:33 2009 +0000
57236
57237 Repair '-nolock'
57238
57239 commit 446fe9eecddd1337f9d5164dd7c301e1ba3dfe32 removes the AC_DEFINE for
57240 SERVER_LOCK and conditional compilation checking it, making it always on
57241 everywhere, except in os/utils.c where code is left under SERVER_LOCK, which
57242 now never gets built, making the '-nolock' option non-functional...
57243
57244 This seems to have been broken since Xserver 1.7.0, but this option is
57245 actually of some slight use on cygwin, as if /tmp resides on a FAT filesystem
57246 (yes, I know...), hard links aren't supported.
57247
57248 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57249 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57250
57251commit 7f54ccafadf99c0a1a3e788734199b306b4fa51d
57252Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57253Date: Fri Jan 16 13:12:12 2009 +0000
57254
57255 Cygwin/X: Make transient windows resizable again
57256
57257 Reverts the change which makes parented windows non-resizeable
57258
57259 Because this was trying to do something which we should be doing, as an alternative we
57260 try checking WM_NORMAL_HINTS for windows which shouldn't be resizable
57261
57262 If a window has a maximum size specified, no maximize box
57263 If a window has a fixed size (max size == min size, per EWMH 1.3 Implementation Notes), no resizing frame
57264
57265 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57266 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57267
57268commit 654d2e372dc2978ce379ab9f02137333ec224f0c
57269Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57270Date: Tue Nov 10 13:24:40 2009 +0000
57271
57272 Cygwin/X: Discourage other WMs in multiwindow mode
57273
57274 Tidying up of other WM detection code when ading SWT/Motif
57275 fix in commit 71519a572fe15b85c0eb2b02636c9e871f2c858f
57276 was rather over-agressive and now allows other WMs to think
57277 they can start when the internal WM is running.
57278
57279 Revert to the behaviour of selecting ButtonPressMask events
57280 as well on the root window, so other WMs will be dissuaded
57281 from starting
57282
57283 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57284 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57285
57286commit 26857b1c2003797b02e258247f63064aa1e37c10
57287Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57288Date: Sat Oct 31 17:19:17 2009 +0000
57289
57290 Cygwin/X: Update XWin man page
57291
57292 Restructure to group similar options and offer some commentary on those groups
57293 Review option descriptions, clarify and bring up-to-date
57294
57295 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57296 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57297
57298commit 6f1836bfbd80e88d4c57a32757d0a5b398504c35
57299Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57300Date: Mon Nov 2 17:37:14 2009 +0000
57301
57302 Cygwin/X: Update DDX specific -help text
57303
57304 Alphabetize options
57305 Review option descriptions, clarify and bring up to date
57306 Add missing option descriptions
57307
57308 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57309 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57310
57311commit 22982b9e95a2339d5ba60d66263e42a0331ee41f
57312Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57313Date: Thu Oct 29 18:50:23 2009 +0000
57314
57315 Cygwin/X: XSupportsLocale() failure is non-critical
57316
57317 Treat XSupportsLocale() returning false as non-critical to internal client
57318 theads startup, and issue a warning, not an error
57319
57320 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57321 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57322
57323commit f60b7712b3451649f138b158ee282be89a66b9ef
57324Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
57325Date: Mon Feb 1 17:13:00 2010 +0000
57326
57327 Cygwin/X: Report BUILDERSTRING with version information
57328
57329 Report BUILDERSTRING with version information
57330
57331 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57332 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57333
57334commit ed2c2a68e1e6b409c198a52d99d300eb6d517e89
57335Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
57336Date: Mon Feb 1 17:12:57 2010 +0000
57337
57338 Define FD_SETSIZE on Cygwin regardless of XWin DDX
57339
57340 All DDXs segfault on Cygwin unless -DFD_SETSIZE=256 is set, so make sure
57341 we do so whether or not we are building XWin.
57342
57343 FD_SETSIZE must be at least XFD_SETSIZE for uses of select() to be correct.
57344 The Cygwin default is only 64, so it must be increased to 256
57345
57346 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
57347 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57348 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57349
57350commit 2f2f3da080629d410dd99e281c382b54f0dbbf5d
57351Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57352Date: Fri Jun 19 21:14:47 2009 +0100
57353
57354 Cygwin/X: Copy the state of the Windows keyboard device to the Virtual Core Keyboard at startup.
57355
57356 Otherwise, this happens lazily after the first keypress, which can lead
57357 to applications which are started from a shell window and inspect the
57358 keyboard state before a character is typed getting the wrong idea about
57359 the desired keymap (e.g. xemacs shows this behaviour)
57360
57361 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57362 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57363
57364commit 7a440e5b7a416e582b6c3cc4c33822854ce73aed
57365Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57366Date: Sun Oct 4 15:55:40 2009 +0100
57367
57368 Cygwin/X: Tidy up system.Xwinrc
57369
57370 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57371 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57372
57373commit 11252ed82e1f361b99e86521ac9314f868bd1a3a
57374Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
57375Date: Mon Feb 1 17:11:54 2010 +0000
57376
57377 Cygwin/X: Look for system.Xwinrc in SYSCONFDIR/X11
57378
57379 Look for system.Xwinrc in SYSCONFDIR/X11 (usually /etc/X11)
57380 Rename sample system.Xwinrc file not to have a X11R6 path in it's name
57381 Add makefile install rule for system.XWinrc
57382
57383 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57384 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57385
57386commit 34269a90ea2087f883f5dc8805894fc4998e4b81
57387Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
57388Date: Tue Feb 2 16:21:44 2010 +0000
57389
57390 Cygwin/X: Update resources file and About dialog
57391
57392 Use the configured vendor web address in the About dialog
57393
57394 Update resources file: rework About dialog, use 'MS Shell Dlg 2'
57395 logical font for all dialogs, add ellipsis to exit option in tray
57396 menu as it (may) trigger another dialog.
57397
57398 Get short vendor name from xwin-config.h, like other configuration
57399 data presented in the About dialog box, rather than creating the
57400 PROJECT_NAME define
57401
57402 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57403 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57404
57405commit 881812438b430d2856d0494707e028e5f30d6e9a
57406Author: Colin Harrison <colin.harrison@virgin.net>
57407Date: Tue Feb 2 16:22:39 2010 +0000
57408
57409 Xming: Use 8 point font for URL in About dialog
57410
57411 Use an 8 point font for URL in About dialog, to match the rest
57412 of the text in that dialog
57413
57414 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57415
57416commit c02638fd68440513b6046315547c554a910bd7e2
57417Author: Colin Harrison <colin.harrison@virgin.net>
57418Date: Tue Feb 2 16:08:45 2010 +0000
57419
57420 Xming: Add a manifest file
57421
57422 Use manifests to enable XP style controls
57423 (only effective for XP and later and when themes are enabled).
57424 The addition of manifests shouldn't cause compatibility problems with older Windows versions.
57425
57426 Manifest must have execute permissions, otherwise attempts to
57427 execute XWin.exe in the same directory will fail...
57428
57429 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57430
57431commit fd8a32baba77aba8d124658a19f4f6eda79e49c6
57432Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57433Date: Fri Jun 19 21:14:39 2009 +0100
57434
57435 Cygwin/X: Window placement refinement for multiwindow mode
57436
57437 Window placement refinement for multiwindow mode, ensure a window actually
57438 ends up somewhere visible if it tries to create itself offscreen (which
57439 can happen for e.g. if it has a stored position from a different sized
57440 display)
57441
57442 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57443 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57444
57445commit 7eb840fd42bc62d88fb4fcf6600546c9ff6e56be
57446Author: Jon TURNEY <jon.turney@dronecode.org.uk>
57447Date: Tue Feb 2 15:42:12 2010 +0000
57448
57449 Cygwin/X: Respect the system's ownership of the clipping region
57450
57451 Respect the system's ownership of the clipping region used for shaped windows
57452
57453 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57454 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
57455
57456commit bad41bdfd4a12f0c92fd221bffd9f3c5e40d0fe1
57457Author: Colin Harrison <colin.harrison@virgin.net>
57458Date: Mon Feb 1 17:11:22 2010 +0000
57459
57460 Xming: Always update the Windows title Unicode (wide-character) in -multiwindow mode.
57461
57462 Apply the Windows title wide-character in -multiwindow mode. Windows should now
57463 display correct client X Window titles for locales with wide characters.
57464
57465 Copyright (C) Colin Harrison 2005-2008
57466 http://www.straightrunning.com/XmingNotes/
57467 http://sourceforge.net/projects/xming/
57468
57469 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
57470
57471commit 901c4ebf620579c5bb951a8c003430d967ecf2ff
57472Author: Julien Cristau <jcristau@debian.org>
57473Date: Mon Jan 11 00:31:27 2010 +0000
57474
57475 main: move config_init() after InitInput()
57476
57477 With the udev backend, config_init() calls NewInputDeviceRequest(),
57478 which enables devices. They can then start sending events, even though
57479 the event queue is only initialized later in InitInput(). Oops.
57480
57481 Debian bug#564256 <http://bugs.debian.org/564256>
57482
57483 Reported-by: Sedat Dilek <sedat.dilek@googlemail.com>
57484 Signed-off-by: Julien Cristau <jcristau@debian.org>
57485 Tested-by: Sedat Dilek <sedat.dilek@googlemail.com>
57486 Reviewed-by: Adam Jackson <ajax@redhat.com>
57487 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57488
57489commit 7f648777680b9dea31c4adb657e205fc6880d98b
57490Author: Dan Nicholson <dbn.lists@gmail.com>
57491Date: Thu Jan 14 16:42:16 2010 +1000
57492
57493 Don't use AC_CHECK_FILE for fontpath checks when cross compiling
57494
57495 AC_CHECK_FILE chokes when cross compiling, so instead we set the default
57496 to the standard FONTROOTDIR directories in that case.
57497
57498 Signed-off-by: Dan Nicholson<dbn.lists@gmail.com>
57499 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57500 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57501
57502commit db687f718f760ba254ab51994769db101dc9ca3a
57503Author: Peter Hutterer <peter.hutterer@who-t.net>
57504Date: Fri Jan 15 10:07:38 2010 +1000
57505
57506 xkb: sed True -> TRUE and False -> FALSE
57507
57508 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57509 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
57510 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
57511
57512commit 0ea2b0bd02f8683998e8b9ebc2b96d606ce45f45
57513Author: Peter Hutterer <peter.hutterer@who-t.net>
57514Date: Fri Dec 18 20:07:30 2009 +1000
57515
57516 xkb: Add XKM file format description.
57517
57518 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57519 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
57520 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
57521
57522commit 08b22c7faf97217ea4d497eec6624fc3dd916d9b
57523Author: Oldřich Jedlička <oldium.pro@seznam.cz>
57524Date: Sun Jan 17 17:59:03 2010 +0100
57525
57526 Allow driver to call DeleteInputDeviceRequest during UnInit
57527
57528 When the input driver (like xf86-input-wacom) removes it's devices
57529 during a call to UnInit, the CloseDownDevices() cannot handle it. The
57530 "next" variable can become a pointer to freed memory.
57531
57532 The patch introduces order-independent device freeing mechanism by
57533 remembering the already freed device ids. The devices can reorder any
57534 time during freeing. No device will be double-freed - if the removing
57535 failed for any reason; some implementations of DeleteInputDeviceRequest
57536 don't free the devices already.
57537
57538 Signed-off-by: Oldřich Jedlička <oldium.pro@seznam.cz>
57539 Reviewed-by: Simon Thum <simon.thum@gmx.de>
57540 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57541 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57542
57543commit b91cec26de367f75388e620772747b08eee34294
57544Author: Horst Wente <horst.wente@acm.org>
57545Date: Wed Dec 30 19:35:20 2009 +0100
57546
57547 xkb: make ctrl+alt+keypad + / ctrl+alt+keypad - work again (#25743)
57548
57549 Video mode switching via keypad keys did not work
57550
57551 Signed-off-by: Horst Wente <horst.wente@acm.org>
57552 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57553 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57554
57555commit 993e78d6c433f65bd8a87890a6bb6da480b2cdc4
57556Author: Oldřich Jedlička <oldium.pro@seznam.cz>
57557Date: Tue Jan 12 19:15:06 2010 +0100
57558
57559 Fix typo in updateSlaveDeviceCoords
57560
57561 The index [0] for the second valuator looks bogus; fix it.
57562
57563 Signed-off-by: Oldřich Jedlička <oldium.pro@seznam.cz>
57564 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57565 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57566
57567commit 52456c602c3cdd7d5eac677889a18fad37dfb7ae
57568Author: Jeremy Huddleston <jeremyhu@apple.com>
57569Date: Sat Jan 30 14:49:02 2010 -0800
57570
57571 XQuartz: Attatch a stub display when CoreGraphics reports no displays.
57572
57573 This is half of the required changes to address the "stuck mouse pointer"
57574 bug that occurs when X11 launches while the displays are asleep. The
57575 remainder of the fix is part of libXplugin which needs to be updated to
57576 send XP_EVENT_DISPLAY_CHANGED when the display wakes up. If you don't
57577 have a recent enough libXplugin (expected in 2.5.0_beta2 or later), you
57578 can cause this event to be sent by changing your display resolution (or
57579 you could just start X11.app with your screens awake).
57580
57581 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
57582
57583commit 15f4bb6dc64313ff100aa5444a94c60922a498df
57584Author: Jeremy Huddleston <jeremyhu@apple.com>
57585Date: Wed Jan 27 17:12:12 2010 -0800
57586
57587 XQuartz: Dead code removal
57588
57589 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
57590
57591commit 7ba5de6e9e1e8efda9f79af601fb6269bae841ce
57592Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
57593Date: Sat Jan 30 10:55:38 2010 -0800
57594
57595 XQuartz: Add some .gitignore magic
57596
57597 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
57598
57599commit 4d575b0559817258f7a0ce6c4d2d0f9e7e5bba63
57600Author: Robert Morell <rmorell@nvidia.com>
57601Date: Fri Jan 29 19:07:03 2010 -0800
57602
57603 RENDER: Fix gradient and solid fill pictures with Xinerama, and misc cleanup
57604
57605 If these aren't wrapped, then procs that are wrapped (such as
57606 RenderChangePicture) will fail in Xinerama when they see the resource
57607 type of a picture created through one of these interfaces is PictureType
57608 and not XRT_PICTURE like those allocated via RenderCreatePicture.
57609
57610 Signed-off-by: Robert Morell <rmorell@nvidia.com>
57611 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
57612 Signed-off-by: Keith Packard <keithp@keithp.com>
57613
57614commit 8d6b1838330625b758ebc55bb0df1b425c5cefaf
57615Author: Robert Morell <rmorell@nvidia.com>
57616Date: Thu Jan 21 12:38:47 2010 -0800
57617
57618 Render: Fix request size verification
57619
57620 RenderSetPictureClipRectangles and the Xinerama version of
57621 RenderChangePicture were using the wrong structure types for request
57622 size verification.
57623
57624 Signed-off-by: Robert Morell <rmorell@nvidia.com>
57625 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
57626 Signed-off-by: Keith Packard <keithp@keithp.com>
57627
57628commit 5e81078cf56aabbf6551d96d312c7840ba9370ae
57629Author: Tiago Vignatti <tiago.vignatti@nokia.com>
57630Date: Mon Jan 18 14:31:12 2010 +0200
57631
57632 xfree86: vgaarb: remove useless debug
57633
57634 This is RAC's remnant. Any sane person would use a more wise method of
57635 debugging instead.
57636
57637 X.Org Bug 26074 <http://bugs.freedesktop.org/show_bug.cgi?id=26074>
57638
57639 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
57640 Reviewed-by: Keith Packard <keithp@keithp.com>
57641 Signed-off-by: Keith Packard <keithp@keithp.com>
57642
57643commit e38df06772965cd5099646f1e7032b7c944d16ed
57644Author: Aaron Plattner <aplattner@nvidia.com>
57645Date: Thu Jan 28 14:33:25 2010 -0800
57646
57647 DRI2: Bump the dri2 module version to 1.2.0 to reflect recent API changes.
57648
57649 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
57650 Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
57651 Signed-off-by: Keith Packard <keithp@keithp.com>
57652
57653commit 60b6477f6dc005a1b052be8c5e5a549550a4aa88
57654Author: Tiago Vignatti <tiago.vignatti@nokia.com>
57655Date: Thu Jan 14 17:18:28 2010 +0200
57656
57657 dix/configure: remove null root cursor option
57658
57659 The default behavior of the server is to start with an invisible root cursor.
57660 Be such cursor invisible or inexistent (null), in the end it doesn't matter -
57661 for the user. The content on screen will be the same. Besides, there's no
57662 difference, in terms of performance, whether such cursor is invisible or
57663 simply null. The paths that both take inside the server are roughly the same.
57664
57665 Therefore create a null root cursor becomes irrelevant.
57666
57667 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
57668 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
57669 Acked-by: Daniel Stone <daniel@fooishbar.org>
57670 Signed-off-by: Keith Packard <keithp@keithp.com>
57671
57672commit b4baab90c0d98bef98d485682d4a69a327a380d6
57673Author: Peter Hutterer <peter.hutterer@who-t.net>
57674Date: Fri Jan 8 15:18:41 2010 +1000
57675
57676 dix: EventToCore needs to copy the root window too.
57677
57678 This value isn't actually set for normal events but it saves us some work
57679 for the record extension support.
57680
57681 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57682 Reviewed-by: Keith Packard <keithp@keithp.com>
57683 Signed-off-by: Keith Packard <keithp@keithp.com>
57684
57685commit 27fc98ecfed99fefd5c14388b4c6242f87281895
57686Author: Keith Packard <keithp@keithp.com>
57687Date: Wed Jan 27 14:37:32 2010 -0800
57688
57689 Revert "dix: EventToCore needs to copy the root window too."
57690
57691 This reverts commit 0f2d297dedeff8bd227df4c498cc668b0e902344.
57692
57693 Peter provided an updated patch which fixes the tests.
57694
57695commit 711e26466ae04ae93ff4c48d377d83d68a6320e9
57696Author: Jesse Barnes <jbarnes@virtuousgeek.org>
57697Date: Mon Jan 25 09:21:51 2010 -0800
57698
57699 DRI2: handle drawable destruction properly at DRI2SwapComplete time
57700
57701 Simon reported an issue with kwin that turned out to be a general problem. If
57702 a drawable goes away before its swap completes, we'll try to free it up.
57703 However, we free it improperly, which causes a server crash in
57704 DRI2DestroyDrawable. Fix that up by splitting the free code out and calling
57705 it from DRI2SwapComplete.
57706
57707 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
57708 Reviewed-by: Michel Dänzer <michel@daenzer.net>
57709 Signed-off-by: Keith Packard <keithp@keithp.com>
57710
57711commit b68f0204a2e4fa9d8884cbdd84b6a5df21d6b36e
57712Author: Jesse Barnes <jbarnes@virtuousgeek.org>
57713Date: Wed Jan 27 09:10:41 2010 -0800
57714
57715 xserver: require libdri 7.8.0 to build
57716
57717 We depend on new DRI interfaces now, so require them in configure.ac.
57718
57719 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
57720 Signed-off-by: Keith Packard <keithp@keithp.com>
57721
57722commit 003829072853546abd973266fe9b24d803f4f5cb
57723Author: Alan Coopersmith <alan.coopersmith@sun.com>
57724Date: Tue Jan 26 22:25:04 2010 -0800
57725
57726 Avoid segfaults in XF86VidMode GammaRamp functions if randr_crtc is NULL
57727
57728 Fixes crash when xscreensaver tries to use GammaRamp calls to fade out
57729 http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6915712
57730
57731 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
57732 Reviewed-by: Keith Packard <keithp@keithp.com>
57733 Signed-off-by: Keith Packard <keithp@keithp.com>
57734
57735commit 19d03d4f49e08442f58cf02240e3e6bab04633d2
57736Author: Gaetan Nadon <memsize@videotron.ca>
57737Date: Tue Jan 26 20:15:49 2010 -0500
57738
57739 doc: finish the removal of SecurityPolicy file man pages
57740
57741 The variable was unassigned but all the code was left in.
57742
57743 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
57744 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
57745 Signed-off-by: Keith Packard <keithp@keithp.com>
57746
57747commit 7962c8f78964d460c76f76dda2795b971a8c2a94
57748Author: Gaetan Nadon <memsize@videotron.ca>
57749Date: Tue Jan 26 20:02:13 2010 -0500
57750
57751 man: add missing __datadir__ for Xserver XWin man pages
57752
57753 The man pages display __datadir__/fonts/X11/... because __datadir__
57754 is missing in cpprules.in. Problem was introduced in commit:
57755 b54bc14ce0ae38c4863794bc3096ca86cdb23908
57756 when replacing __projectroot__ in the previous font path.
57757
57758 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
57759 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
57760 Signed-off-by: Keith Packard <keithp@keithp.com>
57761
57762commit d2322b6309bf15a45002b42e7e6ba3d6b5bfa932
57763Author: Kok, Auke <auke-jan.h.kok@intel.com>
57764Date: Wed Jan 27 11:34:45 2010 -0800
57765
57766 xserver: Add timestamps to logfile output.
57767
57768 Add timestamps in seconds derived from clock_monotonic to the log
57769 file.
57770
57771 Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
57772 Reviewed-by: Keith Packard <keithp@keithp.com>
57773 Signed-off-by: Keith Packard <keithp@keithp.com>
57774
57775commit 0b21a0416b4cb2c32da5e3fda05a0682eb97d56e
57776Merge: 1e6fd65 837bd2b
57777Author: Keith Packard <keithp@keithp.com>
57778Date: Wed Jan 27 14:14:15 2010 -0800
57779
57780 Merge remote branch 'alanc/master'
57781
57782commit 1e6fd65d0b95260253828678131885a4ec21c594
57783Author: Michael Vogt <mvo@ubuntu.com>
57784Date: Mon Jan 25 18:41:20 2010 +0100
57785
57786 xfree86/modes: only call gamma_set if its non NULL
57787
57788 I ran accross a crash with xf86-video-nv-2.1.15 [1] and xserver
57789 1.7.3.901. It looks like the problem is that gamma_set is called even
57790 if that is NULL.
57791
57792 [1] https://launchpad.net/bugs/494627
57793
57794 Reviewed-By: Matthias Hopf <mhopf@suse.de>
57795 Signed-off-by: Julien Cristau <jcristau@debian.org>
57796 Signed-off-by: Keith Packard <keithp@keithp.com>
57797
57798commit 0688dca044f966abc3da667f6d4e79e7cf47f996
57799Author: Eric Anholt <eric@anholt.net>
57800Date: Thu Jan 21 10:31:04 2010 -0800
57801
57802 dri2: Fix order of operations issue in __glXdriSwapEvent test.
57803
57804 Clients would have received swap events regardless of asking for it.
57805
57806 Signed-off-by: Eric Anholt <eric@anholt.net>
57807 Reviewed-by: Keith Packard <keithp@keithp.com>
57808 Signed-off-by: Keith Packard <keithp@keithp.com>
57809
57810commit a6bd5d2e482a5aa84acb3d4932e2a166d8670ef1
57811Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
57812Date: Wed Jan 27 14:03:03 2010 -0800
57813
57814 Fix source pictures getting random transforms after 2d6a8f668342a5190cdf43b5.
57815
57816 *xoff and *yoff were uninitialized for source-only pictures.x
57817
57818 Signed-off-by: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
57819 Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
57820 Signed-off-by: Keith Packard <keithp@keithp.com>
57821
57822commit c3395158678aaab9dca5fc6a812cbe715ddc0e1a
57823Author: Gaetan Nadon <memsize@videotron.ca>
57824Date: Fri Jan 15 14:13:18 2010 -0500
57825
57826 packaging: provide a default README file #24206
57827
57828 All modules should have a README file.
57829 The bottom URL section is found in all X.Org README files.
57830
57831 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57832 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
57833 Signed-off-by: Keith Packard <keithp@keithp.com>
57834
57835commit f311f2d047120fb816897444d2101465ff5189db
57836Author: Aaron Plattner <aplattner@nvidia.com>
57837Date: Wed Jan 13 11:35:52 2010 -0800
57838
57839 DRI2: Allow multiple driver names.
57840
57841 Each driver type (e.g. DRI2DriverDRI or DRI2DriverVDPAU) can have a name in the
57842 driverNames array in DRI2InfoRec. DRI2Connect returns the name for the driver
57843 specified by driverType. Also print names of supported drivers in
57844 DRI2ScreenInit.
57845
57846 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
57847 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
57848 Signed-off-by: Keith Packard <keithp@keithp.com>
57849
57850commit f57bc0ede8e018c7e264b917927c42a018cd1d5a
57851Author: Christian Zander <chzander@nvidia.com>
57852Date: Mon Jan 11 12:29:07 2010 -0800
57853
57854 x86emu: Respect the LEA 67h address size prefix.
57855
57856 Signed-off-by: Christian Zander <chzander@nvidia.com>
57857 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
57858 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com>
57859 Signed-off-by: Keith Packard <keithp@keithp.com>
57860
57861commit 837bd2bbc02b893f96861b48c1f02b7b8e7e3e48
57862Author: Alan Coopersmith <alan.coopersmith@sun.com>
57863Date: Fri Oct 16 22:32:15 2009 -0700
57864
57865 Remove unbalanced ( from failure to move log error
57866
57867 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
57868 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
57869
57870commit 138d4c1670ebab435bf00627c97098a3a54b81a6
57871Author: Alan Coopersmith <alan.coopersmith@sun.com>
57872Date: Sat Jan 16 21:03:00 2010 -0800
57873
57874 glx: Sun compilers now support some gcc __attribute__ values
57875
57876 Sun cc 5.9 and later (__SUNPRO_C >= 0x590) support __attribute__
57877 calls for aligned, always_inline, noinline, pure, const, and malloc.
57878
57879 This commit consists of the related updates to files that were
57880 regenerated by gl_XML.py in mesa after adding the __SUNPRO_C checks to it
57881
57882 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
57883 Signed-off-by: Brian Paul <brianp@vmware.com>
57884
57885commit 79e7b0b875634d0f9e1a95232a4e38adf617bc14
57886Author: Alan Coopersmith <alan.coopersmith@sun.com>
57887Date: Sat Jan 16 10:45:32 2010 -0800
57888
57889 Only enable kdrive input drivers on Linux by default
57890
57891 Fixes build on non-Linux platforms by restoring the defaults to
57892 where they were before commit 6c2b3a4247d10a50699ffa6abb643c5e959eefa8,
57893 to only enable the Linux kbd, mouse & evdev drivers when building
57894 on Linux platforms.
57895
57896 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
57897 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
57898 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
57899
57900commit 39ab474197bdad7d8e9ef496df2d61cbea39d370
57901Author: Alan Coopersmith <alan.coopersmith@sun.com>
57902Date: Tue Dec 15 19:07:38 2009 -0800
57903
57904 Move OS-specific VT key handler code from common to os-support
57905
57906 Adds new function xf86Activate to the OS-specific *VTsw*.c files
57907 and calls it from xf86ProcessActionEvent
57908
57909 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
57910 Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (GNU/Linux)
57911
57912commit 15ca3312c069526b7f2207de9dfb9b9e851caf95
57913Author: Aaron Zang <Aaron.Zang@Sun.COM>
57914Date: Mon Dec 14 17:55:46 2009 -0800
57915
57916 Solaris: Avoid switching to inactive VT's
57917
57918 Fix for OpenSolaris bug 6876992: "[vconsole] Ctrl+Alt+F12 switchs to blank
57919 console screen with hotkeys property turned-off"
57920 http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6876992
57921
57922 Xorg needs to do sanity test for the VT it is commanded to switch to.
57923 If the VT is not opened by any process, discard the switching request.
57924
57925 The changes also contain the fix for some flaws discovered when
57926 getting the new gdm to run.
57927
57928 Signed-off-by: Aaron Zang <Aaron.Zang@Sun.COM>
57929 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
57930
57931commit 2984c18eb994696927a7f3b94d86fd47907334a0
57932Author: Gaetan Nadon <memsize@videotron.ca>
57933Date: Thu Jan 21 14:21:07 2010 -0500
57934
57935 config: replace custom code with reusable macro XORG_WITH_DOXYGEN
57936
57937 XORG_WITH_DOXYGEN provides additional functions like a configure
57938 option which allow platform builders to control the usage of
57939 the doxygen program.
57940
57941 This is a requirement from platforms that do not have such doc tool.
57942 A platform with a back level doxygen may use --without-doxygen
57943 to get the rest of the documentation built.
57944
57945 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
57946 Signed-off-by: Keith Packard <keithp@keithp.com>
57947
57948commit d9c20ee4191de7276a08288adffc24dff48aff8f
57949Author: Tiago Vignatti <tiago.vignatti@nokia.com>
57950Date: Fri Jan 15 19:34:29 2010 +0200
57951
57952 dix: move cursor realize code to its own function
57953
57954 The semantic remains, only code was moved: reuse chunk of code to realize
57955 cursor on both AllocARGBCursor and AllocGlyphCursor.
57956
57957 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
57958 Reviewed-by: Keith Packard <keithp@keithp.com>
57959 Signed-off-by: Keith Packard <keithp@keithp.com>
57960
57961commit f818f222362c93bcc859d997d96d8f3b447729f4
57962Author: Peter Hutterer <peter.hutterer@who-t.net>
57963Date: Fri Jan 15 15:32:19 2010 +1000
57964
57965 xkb: unexport xkbDevicePrivateKey and xkbUnwrapProc.
57966
57967 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57968 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
57969 Signed-off-by: Keith Packard <keithp@keithp.com>
57970
57971commit 6850ea8fb95417db9ce3a70fc17f90d6fdea1389
57972Author: Peter Hutterer <peter.hutterer@who-t.net>
57973Date: Fri Jan 15 15:32:16 2010 +1000
57974
57975 xfree86: replace True/False with TRUE/FALSE.
57976
57977 xf86Xinput.c relied on xkbsrv.h's definition of True/False which seems odd
57978 at first and weird on second glance.
57979
57980 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57981 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
57982 Signed-off-by: Keith Packard <keithp@keithp.com>
57983
57984commit 64b1372c15c3ede2696346e153bca9bd89016a20
57985Author: Peter Hutterer <peter.hutterer@who-t.net>
57986Date: Fri Jan 15 15:32:15 2010 +1000
57987
57988 xkb: remove unused _XkbIsPressEvent and _XkbIsReleaseEvent defines
57989
57990 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
57991 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
57992 Signed-off-by: Keith Packard <keithp@keithp.com>
57993
57994commit c8bba14a390fe3fa16027e6b2433a314b3ea00c3
57995Author: Peter Hutterer <peter.hutterer@who-t.net>
57996Date: Fri Jan 15 15:32:13 2010 +1000
57997
57998 xkb: remove XkbAtomGetString, replace with NameForAtom.
57999
58000 XKB really XKBdoes not XKBneed its own XKBdefines for XKBeverything.
58001
58002 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58003 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
58004 Signed-off-by: Keith Packard <keithp@keithp.com>
58005
58006commit f37799c9712afb7769f369162b904d9ea2dbd608
58007Author: Peter Hutterer <peter.hutterer@who-t.net>
58008Date: Fri Jan 15 15:32:12 2010 +1000
58009
58010 xkb: remove IsKeypadKey define, only used in two places.
58011
58012 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58013 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
58014 Signed-off-by: Keith Packard <keithp@keithp.com>
58015
58016commit c8076f317ee5044ed3cc21b097f8b1741af8e0f1
58017Author: Peter Hutterer <peter.hutterer@who-t.net>
58018Date: Fri Jan 15 15:32:11 2010 +1000
58019
58020 xkb: remove XConvertCase.
58021
58022 Since it's typedef'd to XkbConvertCase anyway and the headers are now split
58023 from the client headers, simply get rid of it altogether.
58024
58025 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58026 Reviewed-by: Keith Packard <keithp@keithp.com>
58027 Signed-off-by: Keith Packard <keithp@keithp.com>
58028
58029commit d627dd9d1ef436b01f0581e40a6736ffc6b5d2c8
58030Author: Peter Hutterer <peter.hutterer@who-t.net>
58031Date: Fri Jan 15 15:32:10 2010 +1000
58032
58033 xkb: remove _XkbClearElems, a memset will do.
58034
58035 Bonus point - it's easier to understand what's actually being done with the
58036 memory.
58037
58038 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58039 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
58040 Signed-off-by: Keith Packard <keithp@keithp.com>
58041
58042commit ea1de3fcdc40d2060bb6d13775005eadd624e767
58043Author: Peter Hutterer <peter.hutterer@who-t.net>
58044Date: Fri Jan 15 15:32:09 2010 +1000
58045
58046 xkb: remove _XkbTyped*alloc
58047
58048 Please no extension-specific macros for memory allocation.
58049
58050 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58051 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
58052 Signed-off-by: Keith Packard <keithp@keithp.com>
58053
58054commit 0f2d297dedeff8bd227df4c498cc668b0e902344
58055Author: Peter Hutterer <peter.hutterer@who-t.net>
58056Date: Tue Jan 5 16:31:37 2010 +1000
58057
58058 dix: EventToCore needs to copy the root window too.
58059
58060 This value isn't actually set for normal events but it saves us some work
58061 for the record extension support.
58062
58063 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58064 Reviewed-by: Keith Packard <keithp@keithp.com>
58065 Signed-off-by: Keith Packard <keithp@keithp.com>
58066
58067commit 6f6a99abc12ddee82898fdabfb50c17e90e094b9
58068Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
58069Date: Wed Jan 13 14:39:54 2010 -0800
58070
58071 XQuartz: Don't FatalError in x_hook_run if the list is empty
58072
58073 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
58074
58075commit 6bde306f7f6b9bbabeaa8bb910ea549be906cd8b
58076Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
58077Date: Wed Jan 13 11:20:29 2010 -0800
58078
58079 XQuartz: Update copyright in bundle for 2010
58080
58081 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
58082
58083commit 6008cc116493cb2825ad0bda0b407b7aefabb3f4
58084Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
58085Date: Mon Jan 11 18:02:55 2010 -0800
58086
58087 XQuartz: Setup the modifier map in the quartz thread
58088
58089 This avoids possible doing it twice which could result in incorrect
58090 keycodes for alt due to our loss of information about its side.
58091
58092 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
58093
58094commit 44f9c3d16c9c9b3362306a9ba26ee52e7baafeca
58095Merge: 032f978 84956ca
58096Author: Keith Packard <keithp@keithp.com>
58097Date: Wed Jan 13 10:19:21 2010 -0800
58098
58099 Merge remote branch 'jbarnes/master'
58100
58101commit 84956ca43b087600d9db297cffd62e960c516d9e
58102Author: Jesse Barnes <jbarnes@virtuousgeek.org>
58103Date: Mon Jan 11 14:56:24 2010 -0500
58104
58105 GLX/DRI2: add INTEL_swap_event support
58106
58107 This allows clients to easily check for swap completion status in their
58108 main loop.
58109
58110 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
58111 Reviewed-by: Adam Jackson <ajax@nwnk.net>
58112 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
58113
58114commit 04a54f69a8085ab3fe11a8713bd8b6b16ed1db27
58115Author: Jesse Barnes <jbarnes@virtuousgeek.org>
58116Date: Thu Jul 16 09:01:17 2009 -0400
58117
58118 DRI2: add support for new DRI2 protocol requests
58119
58120 Support the new DRI2 2.2 protocol requests: DRI2SwapBuffers, DRI2GetMSC,
58121 DRI2WaitMSC, DRI2WaitSBC and DRI2SwapInterval.
58122
58123 These requests allow the server to support the SGI_video_sync,
58124 SGI_swap_interval, and OML_sync_control GLX extensions if DDX support is
58125 present. The new DDX APIs are documented in dri2.h.
58126
58127 Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
58128 Reviewed-by: Adam Jackson <ajax@nwnk.net>
58129 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
58130
58131commit 032f97808c65771a07bac748212cf6457a5d1660
58132Author: Simon Thum <simon.thum@gmx.de>
58133Date: Wed Jan 6 17:32:24 2010 +0100
58134
58135 xfree86: init pointer feedback controls from options
58136
58137 With InputClass support, it makes more sense to cover all
58138 aspects of acceleration in options. Previously, one could only set the
58139 default on the command line.
58140
58141 Signed-off-by: Simon Thum <simon.thum@gmx.de>
58142 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58143 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58144
58145commit 0722c287a4c8a6cdedca9756192547bfcf77ade5
58146Author: Simon Thum <simon.thum@gmx.de>
58147Date: Wed Jan 6 19:43:59 2010 +0100
58148
58149 xfree86: document pointer acceleration in xorg.conf.man
58150
58151 Signed-off-by: Simon Thum <simon.thum@gmx.de>
58152 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58153 Reviewed-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
58154 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58155
58156commit 1aca2d757534879eb55faaee421cb972a9546712
58157Author: Simon Thum <simon.thum@gmx.de>
58158Date: Wed Jan 6 19:13:36 2010 +0100
58159
58160 whitespace fixes
58161
58162 Signed-off-by: Simon Thum <simon.thum@gmx.de>
58163 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58164 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58165
58166commit 14039b5a7982fbf8130501bb00766176a4e9bccb
58167Author: Simon Thum <simon.thum@gmx.de>
58168Date: Wed Jan 6 18:16:20 2010 +0100
58169
58170 doc: actually document SendDragEvents
58171
58172 Signed-off-by: Simon Thum <simon.thum@gmx.de>
58173 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58174 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58175
58176commit 14e4e4a294e648e0bdcb70c34748e1b81c5bb64f
58177Author: Simon Thum <simon.thum@gmx.de>
58178Date: Wed Jan 6 18:13:27 2010 +0100
58179
58180 os: state effect of -a and -t options more precisely
58181
58182 Signed-off-by: Simon Thum <simon.thum@gmx.de>
58183 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58184 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58185
58186commit d727ddb91a549d23f6bf323e91db4b8916c1b7a0
58187Author: Gaetan Nadon <memsize@videotron.ca>
58188Date: Sat Jan 9 16:28:32 2010 -0500
58189
58190 kdrive: klinux.h breaks make dist
58191
58192 This file is no longer part of the source code
58193 and must be removed from distribution.
58194
58195 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
58196 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58197 Signed-off-by: Keith Packard <keithp@keithp.com>
58198
58199commit 6313d2da6c6910827d68cf31fe00b46a34c5bfc7
58200Author: Gaetan Nadon <memsize@videotron.ca>
58201Date: Fri Jan 8 19:04:25 2010 -0500
58202
58203 configure: use backticks rather than $() for commands
58204
58205 This patch to xserver configure.ac is to increase code portability to
58206 non POSIX system by using backticks rather than $() for command
58207 substitution for BUILD_DATE and BUILD_TIME.
58208
58209 Reviewed-by: Rémi Cardona <remi@gentoo.org>
58210 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
58211 Signed-off-by: Keith Packard <keithp@keithp.com>
58212
58213commit 74ab27f64a747ee45810f99cdc6e29b82cf8035e
58214Author: Tiago Vignatti <tiago.vignatti@nokia.com>
58215Date: Mon Dec 28 18:00:28 2009 +0200
58216
58217 Revert "Make sys.c use unaligned access functions provided in compiler."
58218
58219 This reverts commit da923d0bc15e99a8ed1986bd6f5df37f7af7284b.
58220
58221 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
58222 Signed-off-by: Keith Packard <keithp@keithp.com>
58223
58224commit 6a2a57832709798d99d19ff4c587e22f3b3c150d
58225Author: Gaetan Nadon <memsize@videotron.ca>
58226Date: Thu Jan 7 15:23:34 2010 -0500
58227
58228 macros: use PKG_CONFIG variable rather than executable name
58229
58230 User can defined alternate location for pkg-config.
58231 Once option in place, all instances of pkg-config must be converted.
58232
58233 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
58234 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
58235 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
58236 Signed-off-by: Keith Packard <keithp@keithp.com>
58237
58238commit 1b3132f5e981b3398e08263742332cc966bebc07
58239Author: Gaetan Nadon <memsize@videotron.ca>
58240Date: Thu Jan 7 14:12:34 2010 -0500
58241
58242 xwin: ignore intermediate .man file
58243
58244 Only detectable on a Cygwin build
58245
58246 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net
58247 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
58248 Signed-off-by: Keith Packard <keithp@keithp.com>
58249
58250commit 8ab4749aaefb3727b9fc58bb37b50a2d07eb1531
58251Merge: 8d53d84 9437504
58252Author: Keith Packard <keithp@keithp.com>
58253Date: Thu Jan 7 10:32:21 2010 -0800
58254
58255 Merge remote branch 'dottedmag/for-keithp'
58256
58257commit 8d53d84485fdce8ea9686e6f300a69f7ddebd467
58258Author: Adam Tkac <atkac@redhat.com>
58259Date: Thu Jan 7 15:34:52 2010 +0100
58260
58261 Do not define members of include/eventstr.h:EventType enum conditionally.
58262
58263 Main problem is that EventType enumeration members can be different in
58264 module and in server, which obviously causes problems.
58265
58266 Signed-off-by: Adam Tkac <atkac@redhat.com>
58267 Reviewed-by: Adam Jackson <ajax@redhat.com>
58268 Signed-off-by: Keith Packard <keithp@keithp.com>
58269
58270commit 9437504b21dfdb509ab8b3a00f64403e1e4026dd
58271Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58272Date: Mon Jan 4 05:42:02 2010 +0600
58273
58274 kdrive: Remove unused kdNoopOps external variable declaration
58275
58276 Reviewed-by: Adam Jackson <ajax@redhat.com>
58277 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58278 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58279
58280commit aa6e0936be15488455e6a4236806a58496da2979
58281Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58282Date: Mon Jan 4 05:30:22 2010 +0600
58283
58284 kdrive: Remove unused kmap.c
58285
58286 Reviewed-by: Adam Jackson <ajax@redhat.com>
58287 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58288 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58289
58290commit f78ac5c6e273f575af29746e08e7c182ff7b93b3
58291Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58292Date: Mon Jan 4 02:36:30 2010 +0600
58293
58294 kdrive: Make internal functions static
58295
58296 Reviewed-by: Adam Jackson <ajax@redhat.com>
58297 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58298 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58299
58300commit 37a8ddc0dbf3fb7f059b689038979bace206dccf
58301Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58302Date: Mon Jan 4 02:28:37 2010 +0600
58303
58304 kdrive: Remove unused 'k' variable
58305
58306 Reviewed-by: Adam Jackson <ajax@redhat.com>
58307 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58308 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58309
58310commit af1f1a05e1aa9ec921f1288818a66766c301f8b0
58311Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58312Date: Mon Jan 4 01:50:04 2010 +0600
58313
58314 kdrive: Remove bus mouse driver
58315
58316 Bus mice aren't used anymore, do not keep dead code around.
58317
58318 Reviewed-by: Adam Jackson <ajax@redhat.com>
58319 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58320 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58321
58322commit a3cc3af186b55ac80abd6859ffe491f4ae936596
58323Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58324Date: Sat Jan 2 23:31:28 2010 +0600
58325
58326 kdrive: Adjust kdrive usage message as shadow overlay support has been dropped
58327
58328 Reviewed-by: Adam Jackson <ajax@redhat.com>
58329 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58330 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58331
58332commit c40b9f959175057fac726fcdfff9641ae0681cf6
58333Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58334Date: Mon Jan 4 14:07:37 2010 +0600
58335
58336 kdrive: Remove unused VxWorks* variable declarations
58337
58338 Reviewed-by: Adam Jackson <ajax@redhat.com>
58339 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58340 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58341
58342commit 4e0d580d5ac00d8d87c8e0080a02868c87a31caa
58343Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58344Date: Sat Jan 2 15:04:03 2010 +0600
58345
58346 kdrive: Remove unused KdCardInfo::lastMarker field
58347
58348 Reviewed-by: Adam Jackson <ajax@redhat.com>
58349 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58350
58351commit dbaab9759ba31edf63cf73eda86ba931d88e8cce
58352Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58353Date: Sat Jan 2 16:18:29 2010 +0600
58354
58355 kdrive: Extract common part of fbdevPutColors and fbdevEnable
58356
58357 Put framebuffer colormap updating code in separate function
58358 for brevity.
58359
58360 Reviewed-by: Adam Jackson <ajax@redhat.com>
58361 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58362
58363commit 7ee14154b230b931d294288e1291c4fc617b1354
58364Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58365Date: Sat Jan 2 15:42:53 2010 +0600
58366
58367 kdrive: Remove unused overlay fb support
58368
58369 Xfbdev, Xephyr and Xfake all use only one framebuffer, so simplify
58370 implementation by removing overlay support.
58371
58372 Reviewed-by: Adam Jackson <ajax@redhat.com>
58373 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58374
58375commit 730f7d1c4fae8ca582a4a9998a4dc4b325f98896
58376Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58377Date: Sat Jan 2 15:06:18 2010 +0600
58378
58379 kdrive: Remove unused KD_MAX_CARD_ADDRESS macro
58380
58381 Reviewed-by: Adam Jackson <ajax@redhat.com>
58382 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58383
58384commit 5337ddcfd9dc2a5b823da39df7adaad6692214f0
58385Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58386Date: Sat Jan 2 14:59:07 2010 +0600
58387
58388 kdrive: Move Xephyr-specific fields out of KdScreenInfo
58389
58390 memory_base, memory_size, off_screen_base fields in
58391 KdScreenInfo are used only by fake EXA in Xephyr. Move
58392 them into Xephyr, cleanup Xfake and Xfbdev.
58393
58394 Reviewed-by: Adam Jackson <ajax@redhat.com>
58395 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58396
58397commit 62883b499b72f678ab055de4d0370227ae441370
58398Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58399Date: Sat Jan 2 13:17:21 2010 +0600
58400
58401 kdrive: Remove unused KdCardAttr from KdCardInfo
58402
58403 Card attrs are unused in all current kdrive servers, so
58404 remove it completely to avoid allocating and passing dummy
58405 values to KdCardInfoAdd.
58406
58407 Reviewed-by: Adam Jackson <ajax@redhat.com>
58408 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58409
58410commit 6c2b3a4247d10a50699ffa6abb643c5e959eefa8
58411Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58412Date: Mon Jan 4 14:07:04 2010 +0600
58413
58414 kdrive: Add option to compile out input drivers
58415
58416 Add --without-kdrive-{kbd,mouse,evdev} configure options disabling
58417 Linux keyboard driver, Linux mouse drivers (ps2, bus,ms),
58418 and Linux evdev driver.
58419
58420 Build all drivers by default as before.
58421
58422 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
58423 Reviewed-by: Adam Jackson <ajax@redhat.com>
58424 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58425
58426commit 72ba717b1d639e94fa9ab56ca026131edc020a30
58427Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58428Date: Sat Jan 2 15:03:24 2010 +0600
58429
58430 kdrive: Remove unused KdCardInfo::needSync field
58431
58432 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58433 Reviewed-by: Adam Jackson <ajax@redhat.com>
58434 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58435
58436commit d21b41cdb5e89eb428f36b19e965f0ab49e051d9
58437Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58438Date: Sat Jan 2 13:21:25 2010 +0600
58439
58440 kdrive: Remove .gitignore left from SDL server
58441
58442 SDL server is gone, no need to keep its .gitignore anymore.
58443
58444 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
58445 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58446 Reviewed-by: Adam Jackson <ajax@redhat.com>
58447 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58448
58449commit 6f265d55a61f9be323583b8acacae783be72bda9
58450Author: Peter Hutterer <peter.hutterer@who-t.net>
58451Date: Tue Jan 5 13:15:25 2010 +1000
58452
58453 dix: don't update the slave coordinates from the VCK.
58454
58455 A keyboard event from a device with both valuators and keys will be posted
58456 through the VCK. In this case, do not update the slave device coordinates
58457 from the VCK - they're always 0/0. Leave them as-is, for the next pointer
58458 event will continue where it left.
58459
58460 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58461 Reviewed-by: Keith Packard <keithp@keithp.com>
58462 Signed-off-by: Keith Packard <keithp@keithp.com>
58463
58464commit 326429badfc76885e4652ddc72860810c0e8d102
58465Author: Adam Jackson <ajax@redhat.com>
58466Date: Tue Nov 10 14:56:36 2009 -0500
58467
58468 modes: Remove the ClockRanges type
58469
58470 ba2d39dd5428cb5922b797a1d4ea45b859412b40 introduced warnings:
58471
58472 xf86Mode.c: In function ‘xf86CheckModeForDriver’:
58473 xf86Mode.c:986: warning: passing argument 1 of ‘modeInClockRange’ from incompatible pointer type
58474 xf86Mode.c:253: note: expected ‘ClockRangePtr’ but argument is of type ‘ClockRangesPtr’
58475 xf86Mode.c:1002: warning: passing argument 1 of ‘modeInClockRange’ from incompatible pointer type
58476 xf86Mode.c:253: note: expected ‘ClockRangePtr’ but argument is of type ‘ClockRangesPtr’
58477
58478 Because I foolishly didn't notice that we had types with nearly
58479 identical members named ClockRange and ClockRanges. The latter
58480 contained an extra 'strategy' member at the end, which claimed to be
58481 needed by the vidmode extension. Of course, this was a lie: the only time
58482 we'd use it was in mode validation, for drivers using LOOKUP_CLKDIV2 with
58483 non-programmable clocks. The only driver using LOOKUP_CLKDIV2 is
58484 rendition, which has a programmable clock. The only driver using the
58485 ClockRanges type was smi, which did not use it for its 'strategy' member,
58486 so has been fixed to use ClockRange instead.
58487
58488 Signed-off-by: Adam Jackson <ajax@redhat.com>
58489 Reviewed-by: Keith Packard <keithp@keithp.com>
58490 Signed-off-by: Keith Packard <keithp@keithp.com>
58491
58492commit 13c8bd3fde3b0831921e59f84936022a16379d63
58493Author: Alan Coopersmith <alan.coopersmith@sun.com>
58494Date: Mon Jan 4 18:21:54 2010 -0800
58495
58496 CloseDevice: call XkbRemoveResourceClient before freeing key class struct
58497
58498 XkbRemoveResourceClient() returns immediately if dev->key is NULL.
58499 CloseDevice calls XkbRemoveResourceClient until it removes all resources.
58500
58501 If we free dev->key and NULL it before XkbRemoveResourceClient, then
58502 infinite loop ensues, and the server appears to hang on exit or crash.
58503
58504 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
58505 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58506 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
58507 Signed-off-by: Keith Packard <keithp@keithp.com>
58508
58509commit e707612e9de4e5e1d91dc1a8152810912d7e18bb
58510Author: Michel Dänzer <daenzer@vmware.com>
58511Date: Mon Jan 4 09:07:48 2010 +0100
58512
58513 EXA: Restore migration call in exaDoPutImage().
58514
58515 Turns out this is still necessary if the driver PrepareAccess hook succeeds.
58516
58517 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58518 Tested-by: Maarten Maathuis <madman2003@gmail.com>
58519 Tested-by: Andrew Chant <andrew.chant+debian@gmail.com>
58520 Signed-off-by: Keith Packard <keithp@keithp.com>
58521
58522commit e10072b7c74a6f5e24f6dfa37e73688d1f3425e3
58523Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58524Date: Sat Jan 2 05:43:28 2010 +0600
58525
58526 Add Xephyr.man to .gitignore
58527
58528 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58529 Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
58530 Signed-off-by: Keith Packard <keithp@keithp.com>
58531
58532commit fb26cb793c6eba189050662c566f7ea0559bd486
58533Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58534Date: Sat Jan 2 05:43:27 2010 +0600
58535
58536 Remove unused pShadow field from drivers' private structures
58537
58538 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58539 Signed-off-by: Keith Packard <keithp@keithp.com>
58540
58541commit 2c85d72fc7bd013af327321ad5d626dc73c3caf0
58542Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58543Date: Sat Jan 2 05:43:26 2010 +0600
58544
58545 Do not check xfree argument for NULL
58546
58547 xfree itself checks for NULL, and even this is not necessary
58548 as passing NULL to free(3) is safe.
58549
58550 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58551 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58552 Signed-off-by: Keith Packard <keithp@keithp.com>
58553
58554commit 15bc35c6e44c65812de5c3e85ec8bbf78b032ded
58555Author: Mikhail Gusarov <dottedmag@dottedmag.net>
58556Date: Sat Jan 2 05:43:25 2010 +0600
58557
58558 Remove trailing whitespace
58559
58560 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
58561 Signed-off-by: Keith Packard <keithp@keithp.com>
58562
58563commit 1763550d0181ac1c775b9ddf490114eff2fbe67e
58564Author: Simon Thum <simon.thum@gmx.de>
58565Date: Fri Jan 1 19:58:05 2010 +0100
58566
58567 dix: add smooth limited pointer acceleration profile
58568
58569 This profile is inspired by the accel code removed from the wacom driver.
58570 It ascends from zero to acceleration, maxing out at threshold. This means you
58571 can control the slope using threshold, which wasn't possible in wacom.
58572 For sanity's sake, threshold should grow with acceleration.
58573
58574 Works best with adaptive deceleration, since otherwise it only generates
58575 acceleration above 1, causing seldom pixel skips.
58576
58577 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58578 Signed-off-by: Keith Packard <keithp@keithp.com>
58579
58580commit 435f27667f84269768efecde34de4af2b2d43376
58581Author: Julien Cristau <jcristau@debian.org>
58582Date: Sun Sep 27 14:45:47 2009 +0200
58583
58584 config: add libudev input-hotplug backend
58585
58586 Add a backend using libudev for input hotplug, and disable the hal and
58587 dbus backends if this one is enabled.
58588
58589 XKB configuration happens using xkb{rules,model,layout,variant,options}
58590 properties (case-insensitive) on the device. We fill in InputAttributes
58591 to allow configuration through InputClass in Xorg.
58592
58593 Requires udev 148 for the input_id helper and ID_INPUT* properties.
58594
58595 Signed-off-by: Julien Cristau <jcristau@debian.org>
58596 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
58597 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58598
58599commit b8b12e41c453c3bf94b11e7a18934d3b6e1869bf
58600Author: Peter Hutterer <peter.hutterer@who-t.net>
58601Date: Wed Dec 23 16:05:16 2009 +1000
58602
58603 xfree86: move sanity checks below option and input classes merges.
58604
58605 While the identifier is likely set before the input classes are merged, the
58606 driver may not be. Hence don't check for a driver before we've completed
58607 configuration for this device.
58608
58609 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58610 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
58611 Signed-off-by: Julien Cristau <jcristau@debian.org>
58612
58613commit 9fad8f06fb89ac2ae05bea0fa24cab3df7677297
58614Merge: 871bbe1 42e8c92
58615Author: Keith Packard <keithp@keithp.com>
58616Date: Wed Dec 30 09:28:19 2009 -0800
58617
58618 Merge remote branch 'dbn/inputclass'
58619
58620commit 871bbe1d87fa3c7ebd075e1d1eec33e45b08493d
58621Merge: db2c6f7 e116563
58622Author: Keith Packard <keithp@keithp.com>
58623Date: Wed Dec 30 09:16:45 2009 -0800
58624
58625 Merge remote branch 'dbn/xorg.conf.d'
58626
58627commit db2c6f7c91289b5d49978974093a1002b3b53a56
58628Author: Michel Dänzer <daenzer@vmware.com>
58629Date: Tue Dec 29 15:48:44 2009 +0100
58630
58631 EXA: Don't use UTS/DFS directly for Put/GetImage when there's a system copy.
58632
58633 We want to save the result in the system memory copy, in case we'll need it
58634 again for subsequent software fallbacks.
58635
58636 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58637 Acked-By: Maarten Maathuis <madman2003@gmail.com>
58638 Signed-off-by: Keith Packard <keithp@keithp.com>
58639
58640commit f28ca728e90ccbb901e8e7215a842525d8e786d3
58641Author: Thomas Hellstrom <thellstrom@vmware.com>
58642Date: Tue Dec 29 12:51:30 2009 +0100
58643
58644 EXA: Restrict the regions that need to be migrated for composite fallback for src / mask pictures.
58645
58646 [ Michel: Minor fixups to address compiler warnings ]
58647
58648 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
58649 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58650 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58651 Signed-off-by: Keith Packard <keithp@keithp.com>
58652
58653commit 0c1f43c0f3e888172f11f62a2f208af70e0183cc
58654Author: Michel Dänzer <daenzer@vmware.com>
58655Date: Tue Dec 29 12:51:29 2009 +0100
58656
58657 EXA/mixed: Handle results of software fallbacks in DamageReport hook.
58658
58659 This is more elegant and probably also slightly more correct than doing it
58660 at FinishAccess time.
58661
58662 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58663 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58664 Signed-off-by: Keith Packard <keithp@keithp.com>
58665
58666commit 08bf26c28fc8147b64f2fe8345b9581e0101571c
58667Author: Michel Dänzer <daenzer@vmware.com>
58668Date: Tue Dec 29 12:51:28 2009 +0100
58669
58670 EXA: Fix migration avoidance for 1x1 pixmaps.
58671
58672 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58673 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58674 Signed-off-by: Keith Packard <keithp@keithp.com>
58675
58676commit 0bd8f0cd7f3823ee17ae8c88fd3e004bfff9982d
58677Author: Michel Dänzer <daenzer@vmware.com>
58678Date: Tue Dec 29 12:51:27 2009 +0100
58679
58680 EXA/classic: Fix crash with migration heuristic "smart".
58681
58682 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58683 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58684 Signed-off-by: Keith Packard <keithp@keithp.com>
58685
58686commit 40453bf7718f1c3d672b87e9d84991032cbef859
58687Author: Thomas Hellstrom <thellstrom@vmware.com>
58688Date: Tue Dec 29 09:34:44 2009 +0100
58689
58690 EXA: Limit src prepareAccess regions for a number of unaccelerated operations.
58691
58692 When we can trivially calculate the affected source regions,
58693 do that before calling region bounded prepareAccess.
58694
58695 [ Michel: Minor fixups to address compiler warnings ]
58696
58697 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
58698 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58699 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58700 Signed-off-by: Keith Packard <keithp@keithp.com>
58701
58702commit aa9ce8ab343d51a67924757c5a43c9572248bb24
58703Author: Michel Dänzer <daenzer@vmware.com>
58704Date: Tue Dec 29 09:34:43 2009 +0100
58705
58706 EXA: Use relevant source region to minimize migration on CopyWindow fallbacks.
58707
58708 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58709 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58710 Signed-off-by: Keith Packard <keithp@keithp.com>
58711
58712commit 39cc110caa1f6481a7213ae39f82669333ec1645
58713Author: Thomas Hellstrom <thellstrom@vmware.com>
58714Date: Tue Dec 29 09:34:42 2009 +0100
58715
58716 EXA: Fix bugs in exaGetImage / ExaCheckGetImage migration.
58717
58718 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
58719 Acked-by: Michel Dänzer <michel@daenzer.net>
58720 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58721 Signed-off-by: Keith Packard <keithp@keithp.com>
58722
58723commit d42f3960514f1c410863ac6c09c5638b5ddeae20
58724Author: Michel Dänzer <daenzer@vmware.com>
58725Date: Mon Dec 28 16:48:18 2009 +0100
58726
58727 EXA: Allow optimized migration to be enabled with mixed pixmaps.
58728
58729 This was always the intention, I only recently realized it wasn't the case
58730 yet...
58731
58732 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58733 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58734 Signed-off-by: Keith Packard <keithp@keithp.com>
58735
58736commit 661630f9099f254e54c8da513aaed12f6532cc35
58737Author: Michel Dänzer <daenzer@vmware.com>
58738Date: Mon Dec 28 16:48:17 2009 +0100
58739
58740 EXA/mixed: Don't clear deferred status of pixmaps if migration is limited.
58741
58742 * With optimized migration, only the pending damage region is synchronized for
58743 destination pixmaps.
58744 * Migration of source pixmaps can be limited to a bounding region.
58745
58746 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
58747 Acked-by: Maarten Maathuis <madman2003@gmail.com>
58748 Signed-off-by: Keith Packard <keithp@keithp.com>
58749
58750commit 42e8c9224e6c54655c45f87999d37d0d67b3f7f5
58751Author: Dan Nicholson <dbn.lists@gmail.com>
58752Date: Fri Oct 2 06:29:28 2009 -0700
58753
58754 xfree86: Introduce InputClass configuration
58755
58756 Currently Xorg uses hal's fdi files to decide what configuration options
58757 are applied to automatically added input devices. This is sub-optimal
58758 since it requires users to use a new and different configuration store
58759 than xorg.conf.
58760
58761 The InputClass section attempts to provide a system similar to hal where
58762 configuration can be applied to all devices with certain attributes. For
58763 now, devices can be matched to:
58764
58765 * A substring of the product name via a MatchProduct entry
58766 * A substring of the vendir name via a MatchVendor entry
58767 * A pathname pattern of the device file via a MatchDevicePath entry
58768 * A device type via boolean entries for MatchIsKeyboard, MatchIsPointer,
58769 MatchIsJoystick, MatchIsTablet, MatchIsTouchpad and MatchIsTouchscreen
58770
58771 See the INPUTCLASS section in xorg.conf(5) for more details.
58772
58773 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
58774 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
58775
58776commit 0711598dd3e8366217676f462f1af7d0899656d9
58777Author: Dan Nicholson <dbn.lists@gmail.com>
58778Date: Wed Sep 2 06:47:13 2009 -0700
58779
58780 config: Introduce InputAttributes in NewInputDeviceRequest
58781
58782 In order to give NewInputDeviceRequest more information, a new
58783 InputAttributes type is introduced. Currently, this collects the product
58784 and vendor name, device path, and sets booleans for attributes such as
58785 having keys and/or a pointer. Only the HAL backend fills in the
58786 attributes, though.
58787
58788 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
58789 Reviewed-by: Peter Hutterer <peter.hutterer at who-t.net>
58790
58791commit c6e8637e29e0ca11dfb35c02da7ca6002ac8c597
58792Author: Dan Nicholson <dbn.lists@gmail.com>
58793Date: Fri Oct 2 06:28:03 2009 -0700
58794
58795 xfree86: Support non-Option boolean entries in configuration
58796
58797 Refactored code into the parser to allow the freeform boolean types used
58798 in Option entries to be used in other configuration entries. This isn't
58799 as powerful as allowing "No" to precede the option names, but it atleast
58800 gives a common handling of "yes", "no", etc.
58801
58802 A type xf86TriState has been added to support an optional boolean. This
58803 allows the boolean sense of the value to be kept while providing a means
58804 to signal that it is unset.
58805
58806 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
58807 Reviewed-by: Peter Hutterer <peter.hutterer at who-t.net>
58808
58809commit e1165632bdfbd720889ed1adf5f7ab338032c0ee
58810Author: Peter Hutterer <peter.hutterer@who-t.net>
58811Date: Mon Dec 21 15:56:35 2009 +1000
58812
58813 xfree86: Add Option AutoServerLayout for input devices.
58814
58815 Any input device with this option will be automatically added to whichever
58816 server layout is selected at startup. This removes the need to reference a
58817 device from the ServerLayout section. The two following configuration are
58818 identical:
58819
58820 CONFIG 1:
58821 Section "ServerLayout"
58822 InputDevice "foo"
58823 EndSection
58824
58825 Section "InputDevice"
58826 Identifier "foo"
58827 ...
58828 EndSection
58829
58830 CONFIG 2:
58831 Section "InputDevice"
58832 Identifier "foo"
58833 Option "AutoServerLayout" "on"
58834 ...
58835 EndSection
58836
58837 The selection of the server layout affects both explicitly specified
58838 layouts and the implicit layout.
58839
58840 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58841 Reviewed-by: Keith Packard <keithp at keithp.com>
58842
58843commit 592b20c517461d32daf44a940386ffcc11c434f8
58844Author: Dan Nicholson <dbn.lists@gmail.com>
58845Date: Mon Dec 21 01:13:44 2009 -0800
58846
58847 xfree86: Allow config directory to be specified on command line
58848
58849 Add a new command line parameter, -configdir, to specify the config
58850 directory to be used. Rules are the same as -config for root vs. user
58851 privileges.
58852
58853 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
58854 Reviewed-by: Peter Hutterer <peter.hutterer at who-t.net>
58855
58856commit efa5269f23c2237eb5368bf5245ffbbf35714153
58857Author: Dan Nicholson <dbn.lists@gmail.com>
58858Date: Mon Dec 21 00:04:16 2009 -0800
58859
58860 xfree86: Use xorg.conf.d directory for multiple config files
58861
58862 Currently there is a single file, xorg.conf, for configuring the server.
58863 This works fine most of the time, but it becomes a problem when packages
58864 or system services need to adjust the configuration. Instead, allow
58865 multiple configuration files to live in a directory. Typically this will
58866 be /etc/X11/xorg.conf.d.
58867
58868 Files with a suffix of .conf will be read and added to the server
58869 configuration after xorg.conf. The server won't fall back to using the
58870 auto configuration unless there is no config file and there are no files
58871 in the config directory.
58872
58873 Right now this uses a simpler search template than the config file
58874 search path by not using the command line or environment variable
58875 parameters. The matching code was refactored a bit to make this more
58876 coherent. Any DDX wanting to read the config files will need to call
58877 xf86initConfigFiles before opening/reading them. This is to allow
58878 xf86openConfigFile without xf86openConfigDirFiles and vice-versa.
58879
58880 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
58881 Reviewed-by: Peter Hutterer <peter.hutterer at who-t.net>
58882
58883commit f1e869aca176e9113f9d234b0ea59b4cac295cc4
58884Author: Dan Nicholson <dbn.lists@gmail.com>
58885Date: Mon Dec 21 00:03:11 2009 -0800
58886
58887 xfree86: Unexport configuration file symbols
58888
58889 These functions should not be used outside of DDXs, so no need to put
58890 them in the ABI.
58891
58892 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
58893 Reviewed-by: Peter Hutterer <peter.hutterer at who-t.net>
58894
58895commit 90e6d93cf9bfafd63d7849dc16ce194d6f9c9d5f
58896Author: Peter Hutterer <peter.hutterer@who-t.net>
58897Date: Wed Dec 23 12:54:14 2009 +1000
58898
58899 test/xi2: fail if xi2 class type is garbage. (#25492)
58900
58901 If the keycode range exceeds the allowable length, memory gets overwritten.
58902 Catch this case by making sure that only allowed class types are
58903 present.
58904
58905 X.Org Bug 25492 <http://bugs.freedesktop.org/show_bug.cgi?id=25492>
58906
58907 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58908 Reviewed-by: Dave Airlie <airlied@redhat.com>
58909 Signed-off-by: Keith Packard <keithp@keithp.com>
58910
58911commit 72286eae9f4a2f2ca6c46919ff8aa1011429e03a
58912Author: Peter Hutterer <peter.hutterer@who-t.net>
58913Date: Tue Dec 22 15:42:59 2009 +1000
58914
58915 xfree86: remove LAYOUT_DEBUG section.
58916
58917 I don't think this one has been in use since 2003.
58918
58919 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58920 Reviewed-by: Dave Airlie <airlied@redhat.com>
58921 Signed-off-by: Keith Packard <keithp@keithp.com>
58922
58923commit 38b5afb0b5b46b61effa6f0e5a377b343e357e7e
58924Author: Peter Hutterer <peter.hutterer@who-t.net>
58925Date: Tue Dec 22 15:40:40 2009 +1000
58926
58927 xfree86: remove some 'enable this later' and if 0 ifdefs
58928
58929 2003 called, they want their ifdefs back.
58930
58931 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58932 Reviewed-by: Dave Airlie <airlied@redhat.com>
58933 Signed-off-by: Keith Packard <keithp@keithp.com>
58934
58935commit 7dc78db79edd15b971110acaf16bbe7f69e40bdb
58936Merge: 81a623f b44c9be
58937Author: Keith Packard <keithp@keithp.com>
58938Date: Tue Dec 22 21:25:18 2009 -0800
58939
58940 Merge remote branch 'whot/master'
58941
58942commit b44c9be244cee286835855483a69c69e80b095c0
58943Author: Julien Cristau <jcristau@debian.org>
58944Date: Tue Dec 22 17:14:09 2009 +0100
58945
58946 test/xi2: fix maximum max_keycode (bug#25492)
58947
58948 The number of keycodes needs to be lower than 0xFFFD so that the length
58949 field of xXIKeyInfo doesn't overflow.
58950
58951 Signed-off-by: Julien Cristau <jcristau@debian.org>
58952 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58953
58954commit 190610e0c62170a27ab3e40c6c6210a583ae1ad4
58955Author: Peter Hutterer <peter.hutterer@who-t.net>
58956Date: Mon Dec 21 11:08:26 2009 +1000
58957
58958 xfree86: remove HistorySize from the xorg.conf man page.
58959
58960 This option isn't parsed by anything anymore.
58961
58962 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58963 Reviewed-by: Keith Packard <keithp@keithp.com>
58964
58965commit 094c6b9f97a9f92e5a0ef3cf5be24c09ed4d6063
58966Author: Peter Hutterer <peter.hutterer@who-t.net>
58967Date: Mon Dec 21 10:42:32 2009 +1000
58968
58969 xfree86: reword InputDevice man sections, deprecate CorePointer/CoreKeyboard
58970
58971 Reshuffle and reword - InputDevice sections are only necessary if
58972 hotplugging is disabled. Put more emphasis on hotplugging and less on HAL
58973 since we'll switch backends eventually.
58974
58975 CorePointer, CoreKeyboard, and AlwaysCore should be listed as deprecated
58976 since they don't do what they used to since 1.4. These days, only
58977 SendCoreEvents matters and it's enabled for any driver calling
58978 xf86ProcessCommonOptions (== every driver).
58979 It only controls the startup behavior too, so document this.
58980
58981 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
58982 Acked-by: Keith Packard <keithp@keithp.com>
58983 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
58984
58985commit 81a623f036fe56d53c4e3bdafb3eaf945502525f
58986Author: Keith Packard <keithp@keithp.com>
58987Date: Sat Dec 19 20:11:41 2009 -0800
58988
58989 Bump to 1.7.99.3 (unreleased)
58990
58991 Signed-off-by: Keith Packard <keithp@keithp.com>
58992
58993commit 3c30c5b6d321f34736c442c9cd982308d9b8b93a
58994Author: Keith Packard <keithp@keithp.com>
58995Date: Fri Dec 18 22:58:59 2009 -0800
58996
58997 Set release date for 1.7.99.2
58998
58999 Signed-off-by: Keith Packard <keithp@keithp.com>
59000
59001commit 895f40792a14d8b88923bf3b428d31ae3bb31e46
59002Author: Alan Coopersmith <alan.coopersmith@sun.com>
59003Date: Wed Dec 2 17:43:01 2009 -0800
59004
59005 Add type name argument to CreateNewResourceType
59006
59007 Convert all calls of CreateNewResourceType to pass name argument
59008
59009 Breaks DIX ABI.
59010
59011 ABI versions bumped:
59012
59013 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59014 Reviewed-by: Keith Packard <keithp@keithp.com>
59015 Signed-off-by: Keith Packard <keithp@keithp.com>
59016
59017commit a11c58fa0c5809f918b36b84be53385cb2d8ea59
59018Author: Alan Coopersmith <alan.coopersmith@sun.com>
59019Date: Fri Dec 11 00:58:25 2009 -0800
59020
59021 Ensure all resource types created have names registered
59022
59023 Calls RegisterResourceName to record the type name for
59024 use by X-Resource, XACE/SELinux/XTsol, and DTrace.
59025
59026 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59027 Reviewed-by: Keith Packard <keithp@keithp.com>
59028
59029commit eb750f8b5e14751d4c40b50499baec5d2ba79db9
59030Author: Alan Coopersmith <alan.coopersmith@sun.com>
59031Date: Fri Dec 11 00:46:22 2009 -0800
59032
59033 Check for failures from CreateNewResourceType
59034
59035 Make sure to check return value before setting bitmask flags.
59036 For most calls, just fails to init the extension. Since Xinput
59037 already calls FatalError() on initialization failure, so does
59038 failure to allocate Xinput's resource type.
59039
59040 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59041 Reviewed-by: Keith Packard <keithp@keithp.com>
59042
59043commit 1df4bd6011e110dcf0649b15bfffd4ab9e6961d6
59044Author: Alan Coopersmith <alan.coopersmith@sun.com>
59045Date: Thu Dec 17 18:24:40 2009 -0800
59046
59047 Change default xkb model from pc104 to pc105
59048
59049 PC105 is a more useful default for non-American keyboard users,
59050 not harmful for American PC101/PC104 keyboard users.
59051
59052 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59053 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
59054 Signed-off-by: Keith Packard <keithp@keithp.com>
59055
59056commit 1d2e189cdc1d65c9ca2aa0a950780dc18e6c50f1
59057Author: Alan Coopersmith <alan.coopersmith@sun.com>
59058Date: Thu Dec 17 18:24:39 2009 -0800
59059
59060 Change sysconfdir brackets to avoid some shells trying to run it as a command
59061
59062 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59063 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
59064 Signed-off-by: Keith Packard <keithp@keithp.com>
59065
59066commit 48749cc21b074ee8e68c3854bd3977ec7c408225
59067Author: Alan Coopersmith <alan.coopersmith@sun.com>
59068Date: Thu Dec 17 18:24:38 2009 -0800
59069
59070 Add platform compatibility defines for Sun Studio compilers
59071
59072 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59073 Signed-off-by: Keith Packard <keithp@keithp.com>
59074
59075commit b63912ed4c69fedd1bea92274d6cae0429a79677
59076Author: Alan Coopersmith <alan.coopersmith@sun.com>
59077Date: Thu Dec 17 18:24:37 2009 -0800
59078
59079 Convert checks for PC98 support from platform #ifdefs to configure flag
59080
59081 Default remains the same - on for most OS'es on i386 (except Solaris),
59082 off for everyone else. Can be manually toggled via --enable-pc98 or
59083 --disable-pc98.
59084
59085 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59086 Signed-off-by: Keith Packard <keithp@keithp.com>
59087
59088commit 801bc8075aee664bd4c6b6ff842ab737c143a1de
59089Author: Peter Hutterer <peter.hutterer@who-t.net>
59090Date: Mon Dec 14 08:43:19 2009 +1000
59091
59092 xfree86: update man page for special keys handling.
59093
59094 SpecialKeyHandling was removed from the kbd driver with version 1.4.0. Since
59095 this is the only version that will build against server 1.7+ it's not
59096 reasonable to mention it in the man page. Reword, point to XKB instead and
59097 make clear that some key combinations _may_ not be available in any given
59098 config.
59099
59100 Reported-by: Derek Fawcus
59101 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59102 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
59103
59104commit 753310837cd3812882d8de67f063bb61813db675
59105Author: Peter Hutterer <peter.hutterer@who-t.net>
59106Date: Mon Dec 14 08:41:18 2009 +1000
59107
59108 xfree86: DontZap has been disabled for a while now, say so in the man page.
59109
59110 1.7 always shipped with DontZap disabled, it's just the default keymaps that
59111 may not include the symbol to trigger it.
59112
59113 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59114 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
59115
59116commit 0cb638dc6822e54567a1731ea1cf588475a226e9
59117Merge: 6a6a041 fbdf493
59118Author: Keith Packard <keithp@keithp.com>
59119Date: Wed Dec 16 21:58:44 2009 -0800
59120
59121 Merge remote branch 'alanc/master'
59122
59123commit 6a6a041c2d8d32f6355db77cd59cff371038c683
59124Merge: 8127465 d503195
59125Author: Keith Packard <keithp@keithp.com>
59126Date: Wed Dec 16 21:57:32 2009 -0800
59127
59128 Merge remote branch 'jeremyhu/master'
59129
59130commit 8127465f44fee99181f37a5e55dafc90bfafe3d8
59131Author: Ville Syrjälä <ville.syrjala@nokia.com>
59132Date: Mon Dec 14 19:19:01 2009 +0200
59133
59134 xf86xv: Fix off-by-one in viewport clipping
59135
59136 Most of the Xv Put/Get operations have an off by one error in the
59137 viewport clipping.
59138
59139 Apparently PutImage was fixed at some point but the same code was
59140 already copy-pasted all over the place, and so the other operations
59141 still suffer from the bug.
59142
59143 Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
59144 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
59145 Signed-off-by: Keith Packard <keithp@keithp.com>
59146
59147commit faca1bc582e374d32ee9d63d10e072fbef4940a3
59148Author: Peter Hutterer <peter.hutterer@who-t.net>
59149Date: Mon Dec 14 11:05:20 2009 +1000
59150
59151 xfree86: belately init RandR12 if xinerama fails. (#24627)
59152
59153 On Fri, Dec 11, 2009 at 10:19:01AM -0800, Keith Packard wrote:
59154 > On Wed, 9 Dec 2009 11:55:14 +1000, Peter Hutterer <peter.hutterer@who-t.net> wrote:
59155 > > On Tue, Dec 08, 2009 at 05:24:06PM -0800, Aaron Plattner wrote:
59156 > > > On Tue, Dec 08, 2009 at 03:52:27PM -0800, Peter Hutterer wrote:
59157 > > > > Xorg +xinerama crashes immediately due to whacky dependency between Xinerama
59158 > > > > and RandR12. The latter doesn't initialize if Xinerama is enabled, but if
59159 > > > > only one screen is found, Xinerama is disabled again and RandR12 tries to
59160 > > > > access data it never initialized.
59161 >
59162 > I'd sure like to have RandR get enabled when xinerama doesn't; is there
59163 > an easy way of making that happen here? Perhaps having the RandR12 code
59164 > disable Xinerama when only one screen is found? Or some other kludge?
59165
59166 you know the dependency better than I do so any hints are apreciated.
59167 afaict, the screenInfo.numScreens (the check used by Xinerama) isn't
59168 necessarily initialized at this point so we can't use the same check.
59169 The following seems to work though:
59170
59171 From 670b3ebdb7312a6433a8f093d0820785db2aea20 Mon Sep 17 00:00:00 2001
59172 From: Peter Hutterer <peter.hutterer@who-t.net>
59173 Date: Mon, 14 Dec 2009 11:00:58 +1000
59174 Subject: [PATCH] xfree86: if only one screen was found, disable Xinerama (#24627)
59175
59176 Xorg +xinerama crashes immediately due to whacky dependency between Xinerama
59177 and RandR12. The latter doesn't initialize if Xinerama is enabled, but if
59178 only one screen is found, Xinerama is disabled again and RandR12 tries to
59179 access data it never initialized.
59180
59181 Dependency chain is:
59182 - ProcessCommandLine sets noPanoramiXExtension to FALSE
59183 - xf86RandR12Init() is a noop
59184 - PanoramiXExtensionInit sets noPanoramiXExtension to TRUE
59185 - xf86RandR12CreateScreenResources tries to use the devPrivates key it never
59186 initialized.
59187
59188 This hack checks if there's only one screen at the time RandR12 is
59189 initialized. If so, we expect Xinerama to fail anyhow so we disable it
59190 ourselves and proceed as planned.
59191
59192 X.Org Bug 24627 <http://bugs.freedesktop.org/show_bug.cgi?id=24627>
59193
59194 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59195 Signed-off-by: Keith Packard <keithp@keithp.com>
59196
59197commit 5f898ddbaa7df01d2479e40238d8c7954fc4b67a
59198Author: Tiago Vignatti <tiago.vignatti@nokia.com>
59199Date: Fri Dec 11 17:04:14 2009 +0200
59200
59201 xfree86: fix -quiet option behaviour
59202
59203 Previously it was trying to set the same value as the default one. Sigh.
59204
59205 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
59206 Acked-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
59207 Signed-off-by: Keith Packard <keithp@keithp.com>
59208
59209commit 7284e198619cdacb38d0ad715e932acff86c8367
59210Author: Tiago Vignatti <tiago.vignatti@nokia.com>
59211Date: Fri Dec 11 17:04:13 2009 +0200
59212
59213 os: print log markers only if log level is >= 0
59214
59215 FWIW default log verbosity is 0, so this will affect only if one start the
59216 server with a different -verbose argument.
59217
59218 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
59219 Acked-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
59220 Signed-off-by: Keith Packard <keithp@keithp.com>
59221
59222commit f57240a0a8c05e0a2589d0127f8888db140d6974
59223Author: Tiago Vignatti <tiago.vignatti@nokia.com>
59224Date: Fri Dec 11 17:04:12 2009 +0200
59225
59226 xfree86: spam output but with verbose level checked instead
59227
59228 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
59229 Acked-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
59230 Signed-off-by: Keith Packard <keithp@keithp.com>
59231
59232commit fbdf493a3e7e48039a0826531e11aa33b04bf391
59233Author: Alan Coopersmith <alan.coopersmith@sun.com>
59234Date: Mon Dec 14 18:46:00 2009 -0800
59235
59236 Add freetype & type1 to the LoadModules ignore list
59237
59238 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59239 Acked-by: Adam Jackson <ajax@redhat.com>
59240
59241commit c9726bbe3148c7a8adb4dfde483741545d59c3dc
59242Author: Alan Coopersmith <alan.coopersmith@sun.com>
59243Date: Tue Dec 15 15:54:22 2009 -0800
59244
59245 Update Sun license notices to current X.Org standard form
59246
59247 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59248 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
59249
59250commit d50319550458f8127298cf8672b47f914b74ab1f
59251Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
59252Date: Fri Dec 11 16:42:43 2009 -0800
59253
59254 Xfake: Nuke -Wl,-undefined=InitExtensions from LDFLAGS
59255
59256 This is not portable, and accodring to Paulo should not be required any more.
59257
59258 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
59259
59260commit ed7d2527190ec56aed38db19d9e958abe88441b9
59261Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
59262Date: Fri Dec 11 11:53:28 2009 -0800
59263
59264 dtrace: Add Xserver-dtrace.h to CLEANFILES
59265
59266 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
59267 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
59268
59269commit 44f102ed64e2552a0e04714c7574f74ca5ee66e2
59270Author: Alan Coopersmith <alan.coopersmith@sun.com>
59271Date: Mon Nov 23 22:24:44 2009 -0800
59272
59273 COPYING: Update license preamble
59274
59275 Update the comment about "a modification is currently under discussion"
59276 since there hasn't been any such discussion for some time and this is
59277 the consensus agreed upon preferred format.
59278
59279 Also throw in a pointer to the OSI discussion page about the MIT license.
59280
59281 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59282 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
59283 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
59284
59285commit 00a7252fe3e519e3e4d62cdd8746833ca59bce77
59286Author: Alan Coopersmith <alan.coopersmith@sun.com>
59287Date: Fri Nov 20 13:38:30 2009 -0800
59288
59289 Fix dtrace object builds of libos to link with SHA1_LIBS
59290
59291 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59292 Acked-by: Adam Jackson <ajax@redhat.com>
59293
59294commit dd2c83d03ecedce8b546d583e436c12f92510754
59295Author: Alan Coopersmith <alan.coopersmith@sun.com>
59296Date: Wed Nov 11 14:42:20 2009 -0800
59297
59298 Fix builds with --with-int10=stub
59299
59300 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
59301 Acked-by: Adam Jackson <ajax@redhat.com>
59302
59303commit 77221c9155a3c96cb63b210a048db1ea71a5e359
59304Author: Adam Jackson <ajax@redhat.com>
59305Date: Thu Oct 29 14:01:29 2009 -0400
59306
59307 glx: swrast can do GLX 1.4 too
59308
59309 Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
59310 Signed-off-by: Adam Jackson <ajax@redhat.com>
59311 Signed-off-by: Keith Packard <keithp@keithp.com>
59312
59313commit a72c65e9176c51de95db2fdbf4c5d946a4911695
59314Author: Keith Packard <keithp@keithp.com>
59315Date: Wed Dec 2 08:37:20 2009 -0800
59316
59317 fb: Adjust transform or composite coordinates for pixman operations
59318
59319 Windows (or even pixmaps, in some cases) may not sit at the origin of
59320 the containing pixmap, so any coordinates relative to the drawable
59321 must be adjusted. For destinations and untransformed sources, the
59322 operation coordinates are adjusted. For transformed sources, the
59323 transform matrix is adjusted.
59324
59325 Signed-off-by: Keith Packard <keithp@keithp.com>
59326 Acked-by: Soeren Sandmann <sandmann@daimi.au.dk>
59327
59328commit bd567061c8b84b268d9bbb01bc4d8981feefb862
59329Author: Keith Packard <keithp@keithp.com>
59330Date: Wed Dec 2 15:51:22 2009 -0800
59331
59332 Split fbGetDrawable into fbGetDrawablePixmap and fbGetPixmapBitsData
59333
59334 These two sub-macros each perform half of the original macro work and
59335 the old macro is now implemented in terms of the new ones. This makes
59336 way for new code which wants to know the underlying pixmap for a
59337 window instead of just getting a pointer to the bits.
59338
59339 Signed-off-by: Keith Packard <keithp@keithp.com>
59340 Acked-by: Soeren Sandmann <sandmann@daimi.au.dk>
59341
59342commit 61335052972a78d67c0ba74f902273b34c63a198
59343Author: Keith Packard <keithp@keithp.com>
59344Date: Tue Dec 1 11:09:32 2009 -0800
59345
59346 Revert "Fix clipping when windows are used as sources"
59347
59348 This reverts commit e9aa61e9f0d663d5b34a397b943b4d1df44e873d.
59349
59350 Conflicts:
59351
59352 fb/fbpict.c
59353
59354 Signed-off-by: Keith Packard <keithp@keithp.com>
59355 Acked-by: Soeren Sandmann <sandmann@daimi.au.dk>
59356
59357commit 071b3c1810d9f2602173acc8578caac20e0b771e
59358Author: Keith Packard <keithp@keithp.com>
59359Date: Tue Dec 1 11:05:04 2009 -0800
59360
59361 Revert "Use IncludeInferiors when copying windows before compositing."
59362
59363 This reverts commit 7c7f0c2c6a04f7044d5ce69e97a615735e5831f1.
59364
59365 Signed-off-by: Keith Packard <keithp@keithp.com>
59366 Acked-by: Soeren Sandmann <sandmann@daimi.au.dk>
59367
59368commit 8e640d6b131d2865a9725d8997023865b0ef3d69
59369Author: Keith Packard <keithp@keithp.com>
59370Date: Tue Dec 1 11:04:49 2009 -0800
59371
59372 Revert "Reserve space for two GC values in copy_drawable()."
59373
59374 This reverts commit 08df24555cb432eb0d90a3f63275e9485e777c4c.
59375
59376 Signed-off-by: Keith Packard <keithp@keithp.com>
59377 Acked-by: Soeren Sandmann <sandmann@daimi.au.dk>
59378
59379commit 0816c5482c25251da8af5978cc348e1c547bf731
59380Author: Keith Packard <keithp@keithp.com>
59381Date: Tue Dec 1 11:04:06 2009 -0800
59382
59383 Revert "fb: Don't crash if copy_drawable() returns NULL."
59384
59385 This reverts commit 66a9616d645f5a23225251d197e00b94c79274f6.
59386
59387 Signed-off-by: Keith Packard <keithp@keithp.com>
59388 Acked-by: Soeren Sandmann <sandmann@daimi.au.dk>
59389
59390commit d3e054ac07dae12a82ad764b0622fddbef4b9ec5
59391Merge: 12fb318 97a6454
59392Author: Keith Packard <keithp@keithp.com>
59393Date: Fri Dec 11 09:22:22 2009 -0800
59394
59395 Merge remote branch 'jeremyhu/master'
59396
59397commit 12fb31815db9de9c01f2d4155a2b74531777c0bf
59398Author: Peter Hutterer <peter.hutterer@who-t.net>
59399Date: Mon Dec 7 12:58:57 2009 +1000
59400
59401 xkb: don't assign garbage value to led_return.
59402
59403 As the comment for the function states, led_return is undefined if map is
59404 NULL. We might as well skip writing to it then.
59405
59406 Found by clang.
59407
59408 Reported-by: Tomas Carnecky <tom@dbservice.com>
59409 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59410 Reviewed-by: Jamey Sharp <jamey@minilop.net>
59411
59412commit ea14a418214ba4c58910f5650d8b25b3fc3a7099
59413Author: Peter Hutterer <peter.hutterer@who-t.net>
59414Date: Thu Dec 3 11:21:31 2009 +1000
59415
59416 Xi: reset device properties to NULL after deleting them. (#25374)
59417
59418 Commit 0e6cee853d8e5bef3274e632ef034d37f14674a9 introduced cleanup code to
59419 remove the accel properties when switching schemes. The same code is
59420 triggered by the default closedown code but only after unconditionally
59421 removing all device properties (as part of the cleanup). The properties,
59422 although deleted never got reset to NULL.
59423
59424 X.Org Bug 25374 <http://bugs.freedesktop.org/show_bug.cgi?id=25374>
59425
59426 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59427 Tested-by: Andy Furniss <lists@andyfurniss.entadsl.com>
59428 Reviewed-by: Dave Airlie <airlied@redhat.com>
59429
59430commit 9ac6a048efaec51f158bbc1a4a3f448cbe5658ad
59431Author: Peter Hutterer <peter.hutterer@who-t.net>
59432Date: Wed Dec 9 11:43:34 2009 +1000
59433
59434 xfree86: remove unused variable in configureDDMonitorSection.
59435
59436 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59437 Reviewed-by: Dave Airlie <airlied@redhat.com>
59438
59439commit 3fad969a90ccab604359ec4de05e745bb4b909e1
59440Author: Julien Cristau <jcristau@debian.org>
59441Date: Sun Dec 6 20:27:06 2009 +0100
59442
59443 Move config_init() after CreateWellKnownSockets() and InitCoreDevices()
59444
59445 config_init() can now add devices directly instead of scheduling a
59446 timer.
59447
59448 Signed-off-by: Julien Cristau <jcristau@debian.org>
59449 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
59450 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59451
59452commit 886140787852c660978a9bb3abc7cc2702cbaed1
59453Author: Tomas Carnecky <tom@dbservice.com>
59454Date: Mon Dec 7 05:54:57 2009 +0100
59455
59456 Fix possible NULL dereference in XkbFlushLedEvents()
59457
59458 Through some code paths it is possible that NULL is being passed in the
59459 'ed' parameter to XkbFlushLedEvents(). Make sure we don't pass it along
59460 to bzero().
59461
59462 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
59463 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
59464 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59465
59466commit 92d9cb7e1340b259294ba317ea3a7d18487498b5
59467Author: Tomas Carnecky <tom@dbservice.com>
59468Date: Sun Dec 6 23:52:54 2009 +0100
59469
59470 XkbWriteCountedString(): return early if str is NULL
59471
59472 This avoids NULL from being passed to memcpy() later in the code. While
59473 that wasn't an issue before - that value being NULL implied 'size == 0'
59474 so memcpy() wouldn't try to dereference it - it made the code harder
59475 to read and also confused clang.
59476
59477 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
59478 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
59479 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59480
59481commit df2635d4644d1fa2772ff58d8d34cdf788ece412
59482Author: Michel Dänzer <daenzer@vmware.com>
59483Date: Thu Dec 10 00:04:53 2009 +0100
59484
59485 EXA/mixed: Don't consider devKind for detecting dimension change.
59486
59487 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=25536 .
59488
59489 There should be no reason for the pitch to actually change when none of the
59490 other dimensions do, and this could lead to incorrectly freeing the system
59491 memory copy when allocating a GPU copy, resulting in loss of valid pixmap
59492 contents.
59493
59494 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
59495 Acked-by: Maarten Maathuis <madman2003@gmail.com>
59496 Signed-off-by: Keith Packard <keithp@keithp.com>
59497
59498commit 97a6454ea57587db490873fee8ff0f899882972d
59499Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
59500Date: Wed Dec 9 18:27:31 2009 -0800
59501
59502 XQuartz: pbproxy: Fix building of standalone xpbproxy executable
59503
59504 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
59505
59506commit 1b8f90aa8b1017c74b254db2f398131b857323d4
59507Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
59508Date: Tue Dec 8 15:50:28 2009 -0800
59509
59510 Miscellaneous compilation warning fixes
59511
59512 main.c:134: warning: no previous prototype for 'dix_main'
59513 rootlessScreen.c: In function 'RootlessMarkOverlappedWindows':
59514 rootlessScreen.c:434: warning: function declaration isn't a prototype
59515 backtrace.c:51: warning: format '%lx' expects type 'long unsigned int', but argument 5 has type 'int'
59516 backtrace.c:54: warning: format '%lx' expects type 'long unsigned int', but argument 5 has type 'int'
59517 set.c: In function 'RecordSetMemoryRequirements':
59518 set.c:413: warning: old-style function definition
59519 set.c: In function 'RecordCreateSet':
59520 set.c:425: warning: old-style function definition
59521 stub.c: In function ‘main’:
59522 stub.c:236: warning: ISO C90 forbids mixed declarations and code
59523
59524 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
59525
59526commit 955b9f23a34cc79a5cd9676b45b3df4ffcc7302b
59527Author: Michel Dänzer <daenzer@vmware.com>
59528Date: Tue Dec 8 20:25:02 2009 +0100
59529
59530 EXA: ModifyPixmapHeader_mixed fixes.
59531
59532 * Better detection of dimension changes.
59533 * Make sure to re-create the system memory copy when the pixmap dimensions
59534 change (e.g. the screen pixmap on screen resize).
59535 * Clear the valid regions.
59536
59537 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
59538 Acked-by: Maarten Maathuis <madman2003@gmail.com>
59539 Signed-off-by: Keith Packard <keithp@keithp.com>
59540
59541commit c1503861cf75654d4f7b22e6f7f6487c47a0a395
59542Author: Michel Dänzer <daenzer@vmware.com>
59543Date: Tue Dec 8 20:25:01 2009 +0100
59544
59545 EXA: ExaDoPrepareAccess return value fixes.
59546
59547 Only return TRUE if the GPU copy is being accessed, and preserve the return
59548 value on repeated / nested calls for the same pixmap.
59549
59550 exaPrepareAccessReg_mixed could get inconsistent return values e.g. when the
59551 same pixmap is both the destination and source of an operation, potentially
59552 resulting in a crash.
59553
59554 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
59555 Acked-by: Maarten Maathuis <madman2003@gmail.com>
59556 Signed-off-by: Keith Packard <keithp@keithp.com>
59557
59558commit fd867387335b6175d76bbe93118bbe5e1e45ce88
59559Author: Maarten Maathuis <madman2003@gmail.com>
59560Date: Sat Dec 5 03:18:31 2009 +0100
59561
59562 exa/mixed: pixmaps that succeed prepare access have no need for a cpu copy
59563
59564 - When they have a gpu copy ofcource.
59565 - Use the presence of a cpu copy as a hint to fall back instead of UTS'ing in
59566 exaHWCopyNtoN.
59567
59568 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
59569 Acked-by: Michel Dänzer <michel@daenzer.net>
59570 Signed-off-by: Keith Packard <keithp@keithp.com>
59571
59572commit bb7acfbcfbc37869c2215c26791c6175a5a6c526
59573Author: Michel Dänzer <daenzer@vmware.com>
59574Date: Fri Dec 4 08:37:14 2009 +0100
59575
59576 EXA: Use correct coordinate system for calculating Composite fallback region.
59577
59578 Fixes incorrectly skipped rendering of some Composite operations to windows.
59579
59580 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
59581 Reviewed-by: Keith Packard <keithp@keithp.com>
59582 Signed-off-by: Keith Packard <keithp@keithp.com>
59583
59584commit 0f4ef7123d3e6e09e04dc55e8edb47aecf017648
59585Author: Maarten Maathuis <madman2003@gmail.com>
59586Date: Thu Dec 3 19:28:13 2009 +0100
59587
59588 exa/mixed: setting devKind before exaCopyDirty* is not needed
59589
59590 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
59591 Acked-by: Michel Dänzer <michel@daenzer.net>
59592 Signed-off-by: Keith Packard <keithp@keithp.com>
59593
59594commit 8ea415d417b3ef6b8a288d10da76ff4bc334e08b
59595Author: Maarten Maathuis <madman2003@gmail.com>
59596Date: Wed Dec 2 20:24:02 2009 +0100
59597
59598 Revert "exa: a few small pitch related changes"
59599
59600 This reverts commit 99d88ef69d5f7dbf99ca605eceb92f42230a89f4.
59601
59602 - Some pixmaps under classic have a sys_pitch which is 0, no idea why. This is
59603 causing rendering corruption.
59604
59605 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
59606 Acked-by: Michel Dänzer <michel@daenzer.net>
59607 Signed-off-by: Keith Packard <keithp@keithp.com>
59608
59609commit 98c8b752254a27ab1aaf881b36bfda0f74929d0a
59610Merge: 0e084d8 91c1bd7
59611Author: Keith Packard <keithp@keithp.com>
59612Date: Wed Dec 2 15:28:07 2009 -0800
59613
59614 Merge remote branch 'whot/master'
59615
59616commit 91c1bd78f7240c92702828f8e5a6b6ce944b9e36
59617Author: Gaetan Nadon <memsize@videotron.ca>
59618Date: Sat Nov 28 21:32:47 2009 -0500
59619
59620 configure.ac: error while checking for XDMXCONFIG_DEP
59621
59622 Introduced in commit 9998105a387e0294054502331a56e1e020cd93e4
59623 The replacement third parameters to PKG_CHECK_MODULES([DMXMODULES]
59624 was not quoted.
59625
59626 Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
59627 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
59628 Tested-by: Julien Cristau <jcristau@debian.org>
59629 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59630
59631commit 761ae22f880bd79550ccf93d321b8a28b3755956
59632Author: Peter Hutterer <peter.hutterer@who-t.net>
59633Date: Tue Dec 1 14:38:19 2009 +1000
59634
59635 xfree86: tell users to disable AutoAddDevices, not AllowEmptyInput.
59636
59637 Technically, disabling AEI is the right suggestion. AEI off forces the
59638 server to init the built-in defaults for input devices (or pick the first
59639 one from the config file). At the same time, hotplugging is still available
59640 with AEI off.
59641
59642 Unfortunatly, in the vast majority of cases users want to simply disable
59643 hotplugging or have a working server while the local HAL configuration is
59644 broken or missing. Disabling AEI will lead to duplicate events, triple
59645 keystrokes, etc. once the configuration works again.
59646 It's not actually required to remove AEI once hotplugging works again,
59647 though it will in many cases lead to a setup that appears broken.
59648
59649 Asking users to disable AutoAddDevices instead means those users disable
59650 hotplugging, can then fix the HAL setup and they _must_ remove the config
59651 line again to test if hotplugging works again. Which doesn't leave them with
59652 a broken config once everything is working nice and dandy. Less bugreports,
59653 everybody wins.
59654
59655 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59656 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
59657 Acked-by: Daniel Stone <daniel@fooishbar.org>
59658 Acked-by: Rémi Cardona <remi@gentoo.org>
59659 Acked-by: James Cloos <cloos@jhcloos.com>
59660
59661commit b584c224a888c9e7f92d7e49021f74232a727c7f
59662Author: Peter Hutterer <peter.hutterer@who-t.net>
59663Date: Fri Oct 30 12:11:41 2009 +1000
59664
59665 Set the source and deviceid for key repeat events (#24785)
59666
59667 X.Org Bug 24785 <http://bugs.freedesktop.org/show_bug.cgi?id=24785>
59668
59669 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59670 Acked-by: Adam Jackson <ajax@redhat.com>
59671
59672commit 66bb8c6fbdfc0fc0d971aac4ec6f949bb9288c1b
59673Author: Peter Hutterer <peter.hutterer@who-t.net>
59674Date: Fri Nov 27 16:20:13 2009 +1000
59675
59676 dix: remove core devices when shutting down. (#25028)
59677
59678 NewInputDeviceRequest (and RemoveDevice) have checks in place to not allow
59679 removal of the VCP/VCK. When shutting down, they need to be cleaned up
59680 nonetheless to free the memory associated.
59681
59682 X.Org Bug 25028 <http://bugs.freedesktop.org/show_bug.cgi?id=25028>
59683
59684 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59685 Reviewed-by: Keith Packard <keithp@keithp.com>
59686
59687commit 83d90b90bcb71c89750f92a177361e53dd261414
59688Author: Peter Hutterer <peter.hutterer@who-t.net>
59689Date: Fri Nov 27 16:08:44 2009 +1000
59690
59691 dix: remove some obsolete comment.
59692
59693 The "counterpart to biggest hack" included checking for the motion history
59694 function - which is unified in 1.7. Hence the check (which is already
59695 removed) would evaluate to true anyway, and this comment isn't needed.
59696
59697 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59698 Reviewed-by: Keith Packard <keithp@keithp.com>
59699
59700commit c20c8897272427cb3f755a3e28e80a9ad46f08a1
59701Author: Peter Hutterer <peter.hutterer@who-t.net>
59702Date: Fri Nov 27 16:01:53 2009 +1000
59703
59704 dix: fix memory leak, free event list on shutdown. (#25028)
59705
59706 X.Org Bug 25028 <http://bugs.freedesktop.org/show_bug.cgi?id=25028>
59707
59708 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59709 Reviewed-by: Keith Packard <keithp@keithp.com>
59710
59711commit 0e084d8c71e697a5cf5d5d7c749455ae14bd6eb7
59712Author: Luc Verhaegen <libv@skynet.be>
59713Date: Wed Nov 11 15:59:27 2009 +0100
59714
59715 Xv: Fix AdjustFrame when driver implements ReputImage.
59716
59717 Should probably also be applied to stabler xserver branches too.
59718
59719 Luc Verhaegen.
59720
59721 From a22bc20721bad506d8fa9772b1258568cbffe7d2 Mon Sep 17 00:00:00 2001
59722 From: Luc Verhaegen <libv@skynet.be>
59723 Date: Wed, 11 Nov 2009 15:52:39 +0100
59724 Subject: [PATCH] Xv: Fix AdjustFrame when driver implements ReputImage.
59725
59726 Finally fixes fd.o #4653, filed more than 4 years ago.
59727
59728 Patch can be happily applied to all modular Xorg versions.
59729
59730 Signed-off-by: Luc Verhaegen <libv@skynet.be>
59731 Reviewed-by: Keith Packard <keithp@keithp.com>
59732 Signed-off-by: Keith Packard <keithp@keithp.com>
59733
59734commit f4fc3406720410e37a2bce1b782cba0f0b734e42
59735Author: Jeremy Huddleston <jeremyhu@apple.com>
59736Date: Mon Nov 30 11:03:59 2009 -0800
59737
59738 XQuartz: Drop calls to alloca
59739
59740 This makes us more consistent with the rest of the codebase, using xalloc/xfree
59741
59742 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
59743 Signed-off-by: Keith Packard <keithp@keithp.com>
59744
59745commit 66a9616d645f5a23225251d197e00b94c79274f6
59746Author: Michel Dänzer <daenzer@vmware.com>
59747Date: Mon Nov 30 13:17:53 2009 +0100
59748
59749 fb: Don't crash if copy_drawable() returns NULL.
59750
59751 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24634 .
59752
59753 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
59754 Signed-off-by: Keith Packard <keithp@keithp.com>
59755
59756commit 8754db77d8169e5ea506a963cebee1a651bcf094
59757Author: Michel Dänzer <daenzer@vmware.com>
59758Date: Mon Nov 30 13:17:52 2009 +0100
59759
59760 EXA: Don't defragment offscreen memory at allocation time.
59761
59762 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24300 .
59763
59764 Offscreen memory allocation can occur from various places, and apparently
59765 doing defragmentation from at least some of them can confuse some driver
59766 acceleration code.
59767
59768 There's still the regular background defragmentation in the WakeupHandler,
59769 which should manage to keep fragmentation at a reasonable level.
59770
59771 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
59772 Signed-off-by: Keith Packard <keithp@keithp.com>
59773
59774commit 0e555a1033e5deed1db8582ca075455a6a2d4228
59775Author: Michel Dänzer <daenzer@vmware.com>
59776Date: Mon Nov 30 13:17:51 2009 +0100
59777
59778 Revert "EXA: Accumulate arbitrary number of glyphs without flushing."
59779
59780 This reverts commit c11678cc189551f2a01eaa7a63969c16950739b4.
59781
59782 Not sure what I was thinking, turns out alloca() of a size derived from client
59783 input is a bad idea.
59784
59785 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
59786 Signed-off-by: Keith Packard <keithp@keithp.com>
59787
59788commit 99d88ef69d5f7dbf99ca605eceb92f42230a89f4
59789Author: Maarten Maathuis <madman2003@gmail.com>
59790Date: Sat Nov 28 10:34:45 2009 +0100
59791
59792 exa: a few small pitch related changes
59793
59794 - Setting pitch before exaCopyDirty* is not needed.
59795
59796 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
59797 Acked-by: Michel Dänzer <michel@daenzer.net>
59798 Signed-off-by: Keith Packard <keithp@keithp.com>
59799
59800commit a54c23fe647cb4d610d871094193ae5959606008
59801Author: Maarten Maathuis <madman2003@gmail.com>
59802Date: Sat Nov 28 10:34:44 2009 +0100
59803
59804 exa: s/PixmapIsOffscreen/PixmapHasGpuCopy and s/pExaPixmap->offscreen/pExaPixmap->use_gpu_copy
59805
59806 - Fixup some variable names as well.
59807
59808 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
59809 Acked-by: Michel Dänzer <michel@daenzer.net>
59810 Signed-off-by: Keith Packard <keithp@keithp.com>
59811
59812commit 342f3689d17256c92cbfee079d24501d27aa1153
59813Author: Maarten Maathuis <madman2003@gmail.com>
59814Date: Sat Nov 28 10:34:43 2009 +0100
59815
59816 exa: handle pixmap create/destroy in lower layers
59817
59818 - Pixmaps that are created during a fallback are automatically prepared access.
59819 - During the fallback accelerated ops are blocked to prevent new/scratch gc's
59820 from triggering accelerated ops on mapped pixmaps.
59821 - A few cases of incorrect wrapping (on the top level pointer instead of
59822 between damage and (w)fb) have been removed.
59823
59824 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
59825 Acked-by: Michel Dänzer <michel@daenzer.net>
59826 Signed-off-by: Keith Packard <keithp@keithp.com>
59827
59828commit b54bc14ce0ae38c4863794bc3096ca86cdb23908
59829Author: Ingmar Vanhassel <ingmar@exherbo.org>
59830Date: Wed Nov 25 20:31:48 2009 +0100
59831
59832 Update man-pages for new default font paths
59833
59834 Signed-off-by: Ingmar Vanhassel <ingmar@exherbo.org>
59835 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
59836 Signed-off-by: Keith Packard <keithp@keithp.com>
59837
59838commit c0367028ac02b3ccfbc245863dc6b7295a4014d6
59839Author: Ingmar Vanhassel <ingmar@exherbo.org>
59840Date: Wed Nov 25 20:31:47 2009 +0100
59841
59842 Update COMPILEDDEFAULTFONTPATH to match the new default font path
59843
59844 Signed-off-by: Ingmar Vanhassel <ingmar@exherbo.org>
59845 Reviewed-by: Alan Coopersmith <alan.coopersmith@sun.com>
59846 Signed-off-by: Keith Packard <keithp@keithp.com>
59847
59848commit 0e6cee853d8e5bef3274e632ef034d37f14674a9
59849Author: Peter Hutterer <peter.hutterer@who-t.net>
59850Date: Wed Nov 18 14:39:25 2009 +1000
59851
59852 dix: clean up accel old scheme data when switching schemes.
59853
59854 InitValuatorClassDeviceStruct always initializes with the default profile.
59855 The default profile allocs data and adds a few properties which become
59856 obsolete if the profile is changed lateron by the driver.
59857
59858 The property handlers are stored in the device's devPrivates and cleaned up.
59859 Ideally, the property handler ID's could be stored somewhere more obvious,
59860 but that seems to require breaking the ABI.
59861
59862 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59863 Acked-by: Simon Thum <simon.thum@gmx.de>
59864
59865commit 1b127ab8429616adf9ec31ba4d8bdd9af6e104a9
59866Author: Peter Hutterer <peter.hutterer@who-t.net>
59867Date: Tue Nov 24 15:31:48 2009 +1000
59868
59869 Xi: when deleting all properties, reset property handler to NULL.
59870
59871 Trying to unregister property handlers during the device closure process
59872 leads to invalid memory accesses.
59873
59874 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59875 Reviewed-by: Keith Packard <keithp@keithp.com>
59876
59877commit 8806375ed72a3cd465fe0a49ead079a334accd6b
59878Author: Peter Hutterer <peter.hutterer@who-t.net>
59879Date: Tue Nov 17 09:25:47 2009 +1000
59880
59881 Move xdmxconfig modules into DMX conditionals (#25102)
59882
59883 xdmxconfig requires additional modules not checked for if Xdmx build is set
59884 to auto (the default). This may lead to build errors if the Xdmx modules are
59885 installed, but not the extra ones required for xdmxconfig.
59886
59887 X.Org Bug 25102 <http://bugs.freedesktop.org/show_bug.cgi?id=25102>
59888
59889 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59890 Reviewed-by: Remi Cardona <remi@gentoo.org>
59891 Acked-by: Julien Cristau <jcristau@debian.org>
59892
59893commit a2adda7ab3defd953cf0f48f5372efb037786a9e
59894Author: Peter Hutterer <peter.hutterer@who-t.net>
59895Date: Thu Nov 19 16:35:55 2009 +1000
59896
59897 Revert "Move xdmxconfig modules into DMX conditionals (#25102)"
59898
59899 New package dependency unnecessarily links in a few libraries that Xdmx
59900 doesn't need. This can be fixed more elegantly.
59901
59902 This reverts commit 0ef15ca9d2d9c78c79a2771c550563bc6931b365.
59903
59904 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59905
59906commit eb967ca36cfe0409972ac987a74d498cd6f6bafb
59907Author: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
59908Date: Tue Nov 17 17:13:43 2009 +0200
59909
59910 Xext: Fix a memory leak on GE client disconnect.
59911
59912 Add a call to dixRequestPrivate to inform dixFreePrivates that memory
59913 allocated in GEClientCallback should be released when client
59914 disconnects. Otherwise there is a leak of sizeof(GEClientInfoRec) for
59915 every client connect/disconnect.
59916
59917 Also remove the explicit allocation and let GEGetClient /
59918 dixLookupPrivate do it. This makes GEClientCallback similar to the
59919 other extension callbacks.
59920
59921 Signed-off-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
59922 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59923
59924commit 55f60e192318132157d3dfdd6732f141bd9dc352
59925Author: Maarten Maathuis <madman2003@gmail.com>
59926Date: Mon Nov 23 22:17:43 2009 +0100
59927
59928 Revert "exa/mixed: be more thorough about setting fb_pitch when needed"
59929
59930 This reverts commit d4fc245115eb2cb323e06a82f9dd52518d9b6a16.
59931
59932 - This is causing crashes/problems for some.
59933
59934 Acked-by: Michel Dänzer <michel@daenzer.net>
59935 Signed-off-by: Keith Packard <keithp@keithp.com>
59936
59937commit 4677b5a80025b50ba2a3e953fd487a549586ae9f
59938Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
59939Date: Mon Nov 23 16:33:00 2009 -0800
59940
59941 XQuartz: Allow better compatability with older versions of xinit
59942
59943 If we are id="org.x" and the launchd socket is ":0", we will claim
59944 the socket to match the old behavior before we prefixed the
59945 socket name with our id.
59946
59947 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
59948
59949commit ec16357edaba23694b12fd234ffa12fd58a73dfa
59950Author: Tiago Vignatti <tiago.vignatti@nokia.com>
59951Date: Mon Nov 23 17:00:06 2009 +0200
59952
59953 configure: remove unused kdrive Xvesa config variable
59954
59955 Xvesa was gone in commit 6d21fbf0 and this should be there as well.
59956
59957 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
59958 Signed-off-by: Keith Packard <keithp@keithp.com>
59959
59960commit a30e739a144912a68adcaa9f426d600c6ecbd529
59961Author: Peter Hutterer <peter.hutterer@who-t.net>
59962Date: Mon Nov 23 13:07:56 2009 +1000
59963
59964 Xi: don't crash when deleting invalid device properties.
59965
59966 Deleting a property that was not set on a device leads to a null-pointer
59967 reference. The protocol allows deleting those properties - it has to be a
59968 noop.
59969
59970 Reproducible:
59971 xinput --set-prop "My device" --type=int --format=8 "my property" 1
59972 xinput --delete-prop "My other device" "my property"
59973
59974 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
59975 Signed-off-by: Keith Packard <keithp@keithp.com>
59976
59977commit 19f7c15e2008dab3c46ba3e14dfa353d01c74f72
59978Author: Zhao Yakui <yakui.zhao@intel.com>
59979Date: Fri Nov 20 14:43:35 2009 +0800
59980
59981 xfree86: Edid quirk for Philips LCD LP154W01
59982
59983 v1->v2: Make one condition case for one quirk instead of merging them
59984 together. This is based on the Keithp's suggestion.
59985
59986 Move the EDID quirk for Philips LCD LP154W01 as the panel reports the vertical
59987 size in cm.
59988
59989 https://bugs.freedesktop.org/show_bug.cgi?id=24482
59990
59991 Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
59992 Signed-off-by: Keith Packard <keithp@keithp.com>
59993
59994commit b8b8db98353760738eead612fe846c2ce1c174fa
59995Author: Michel Dänzer <daenzer@vmware.com>
59996Date: Thu Nov 19 10:46:30 2009 +0100
59997
59998 EXA: Don't use UploadToScreen for CopyNtoN with mixed pixmaps.
59999
60000 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
60001
60002commit d4fc245115eb2cb323e06a82f9dd52518d9b6a16
60003Author: Maarten Maathuis <madman2003@gmail.com>
60004Date: Wed Nov 18 21:23:09 2009 +0100
60005
60006 exa/mixed: be more thorough about setting fb_pitch when needed
60007
60008 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
60009 Acked-by: Michel Dänzer <michel@daenzer.net>
60010
60011commit 647b79f87a9891225678dc6fc2fbda3bdef8fa9d
60012Author: Maarten Maathuis <madman2003@gmail.com>
60013Date: Wed Nov 18 21:23:08 2009 +0100
60014
60015 exa/mixed: avoid copying back pixmap data when no migration took place
60016
60017 - When the driver handles the prepare access no copying is needed.
60018 - Delayed pixmap creation should be fine, because it's handled by the
60019 first prepare access, but the exaPixmapIsOffscreen check in finish access
60020 will return FALSE without a driver pixmap.
60021
60022 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
60023 Acked-by: Michel Dänzer <michel@daenzer.net>
60024
60025commit 30be7ceaf228497ac1ff0a1123c1b35e3aa1fc73
60026Author: Julien Cristau <jcristau@debian.org>
60027Date: Sat Nov 14 18:39:00 2009 +0100
60028
60029 xfree86: set a sane umask before opening the log
60030
60031 Xorg creates its log file following the umask of the user running
60032 startx, which may result in a world-writable log. Set umask to 022 to
60033 prevent this.
60034
60035 Debian bug#555308 <http://bugs.debian.org/555308>
60036 See also http://thread.gmane.org/gmane.comp.security.oss.general/2299
60037
60038 Signed-off-by: Julien Cristau <jcristau@debian.org>
60039 Reviewed-by: Adam Jackson <ajax@redhat.com>
60040 Signed-off-by: Keith Packard <keithp@keithp.com>
60041
60042commit fb95090730360b6b7f5429c40937e8fc9dfe5c14
60043Author: Matt Turner <mattst88@gmail.com>
60044Date: Mon Nov 9 03:33:45 2009 +0000
60045
60046 Use glibc's in/out routines
60047
60048 Let's let glibc do the right thing for dense/sparse selection.
60049
60050 The _alpha_iobase code has been unused since the switch to libpciaccess. It
60051 really should have been killed by fba700f1f6a8976.
60052
60053 Signed-off-by: Matt Turner <mattst88@gmail.com>
60054 Tested-by: Michael Cree <mcree@orcon.net.nz>
60055 Signed-off-by: Keith Packard <keithp@keithp.com>
60056
60057commit 0ef15ca9d2d9c78c79a2771c550563bc6931b365
60058Author: Peter Hutterer <peter.hutterer@who-t.net>
60059Date: Tue Nov 17 11:48:35 2009 +1000
60060
60061 Move xdmxconfig modules into DMX conditionals (#25102)
60062
60063 xdmxconfig requires additional modules not checked for if Xdmx build is set
60064 to auto (the default). This may lead to build errors if the Xdmx modules are
60065 installed, but not the extra ones required for xdmxconfig.
60066
60067 X.Org Bug 25102 <http://bugs.freedesktop.org/show_bug.cgi?id=25102>
60068
60069 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
60070 Tested-by: Adrian Bunk <bunk@stusta.de>
60071 Signed-off-by: Keith Packard <keithp@keithp.com>
60072
60073commit df95be4d3d9a744720e37a16fd89d9569bf6e188
60074Author: Paulo Ricardo Zanoni <pzanoni@mandriva.com>
60075Date: Thu Nov 12 11:27:34 2009 -0200
60076
60077 configure: change 'sigio-default' to 'use-sigio-by-default'
60078
60079 This was discussed in this thread:
60080 http://lists.x.org/archives/xorg-devel/2009-September/002025.html
60081 The patch sent fixed the help string but not the configure option.
60082
60083 Signed-off-by: Keith Packard <keithp@keithp.com>
60084
60085commit 801f0e59800ea4a3f6c85b22dea6f2c400bda703
60086Author: Gaetan Nadon <memsize@videotron.ca>
60087Date: Wed Oct 28 14:09:11 2009 -0400
60088
60089 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206
60090
60091 Add missing INSTALL file. Use standard GNU file on building tarball
60092 README may have been updated
60093 Remove AUTHORS file as it is empty and no content available yet.
60094 Remove NEWS file as it is empty and no content available yet.
60095
60096 Reviewed-By: Jeremy Huddleston <jeremyhu@apple.com>
60097 Signed-off-by: Keith Packard <keithp@keithp.com>
60098
60099commit c93f516e850c2ef2b47cc2e282f9081ae7165ac7
60100Author: Gaetan Nadon <memsize@videotron.ca>
60101Date: Tue Oct 27 15:07:26 2009 -0400
60102
60103 Deploy the new XORG_DEFAULT_OPTIONS #24242
60104
60105 This macro aggregate a number of existing macros that sets commmon
60106 X.Org components configuration options. It shields the configuration file from
60107 future changes.
60108
60109 Reviewed-By: Jeremy Huddleston <jeremyhu@apple.com>
60110 Signed-off-by: Keith Packard <keithp@keithp.com>
60111
60112commit 0f38159cdd154a29caab0342aee41378e9921929
60113Author: Gaetan Nadon <memsize@videotron.ca>
60114Date: Tue Nov 10 09:12:27 2009 -0500
60115
60116 Makefile.am: ChangeLog not required: EXTRA_DIST or *CLEANFILES #24432
60117
60118 ChangeLog filename is known to Automake and requires no further
60119 coding in the makefile.
60120
60121 Reviewed-By: Jeremy Huddleston <jeremyhu@apple.com>
60122 Signed-off-by: Keith Packard <keithp@keithp.com>
60123
60124commit c739beb439bb813e4c82b3216daee7e89aa96f2f
60125Author: Gaetan Nadon <memsize@videotron.ca>
60126Date: Tue Nov 10 08:56:21 2009 -0500
60127
60128 .gitignore: use common defaults with custom section #24239
60129
60130 Using common defaults will reduce errors and maintenance.
60131 Only the very small or inexistent custom section need periodic maintenance
60132 when the structure of the component changes. Do not edit defaults.
60133
60134 Reviewed-By: Jeremy Huddleston <jeremyhu@apple.com>
60135 Signed-off-by: Keith Packard <keithp@keithp.com>
60136
60137commit 9071b0d69748cfa7ecca17b4cb0e431bbb0ef2a4
60138Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60139Date: Mon Nov 9 13:09:55 2009 -0800
60140
60141 XQuartz: Explicitly pass a bellProc to make XBell() work again.
60142
60143 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60144
60145commit e87a03fd531ce7974877688680d3bf9b85c2d835
60146Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60147Date: Tue Oct 27 17:16:57 2009 -0700
60148
60149 XQuartz: pbproxy: Wait for the server to finish starting up, so display is valid.
60150
60151 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60152
60153commit 15fc56addcc906592af7c0f4c0a5ac906fa5c389
60154Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60155Date: Sun Nov 8 20:25:42 2009 -0800
60156
60157 XQuartz: Buildfix for Leopard and older
60158
60159 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60160
60161commit b5aa2e0a5fe233dc883084a5026013472e85bca4
60162Author: Chase Douglas <chasedouglas@gmail.com>
60163Date: Mon Nov 9 22:54:39 2009 -0500
60164
60165 Move FD_CLR above pInfo->read_input
60166
60167 The event fd may be invalidated by the pInfo->read_input call. If it is
60168 invalidated, the subsequent FD_CLR call will segfault. Thus, the FD_CLR
60169 call must precede the pInfo->read_input call.
60170
60171 Signed-off-by: Chase Douglas <chasedouglas@gmail.com>
60172 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
60173 Signed-off-by: Keith Packard <keithp@keithp.com>
60174
60175commit 982f6648fd29d085265bf6035c1bf4d1b2499316
60176Author: Peter Hutterer <peter.hutterer@who-t.net>
60177Date: Tue Nov 10 09:50:50 2009 +1000
60178
60179 dix: increase default number of buttons to 10.
60180
60181 Currently the XTEST device is limited to the same number of buttons the core
60182 device has. This breaks if a user has a mouse with more than 3 buttons
60183 connected and is using a core client to fake button 8+ presses.
60184
60185 Rather than expecting all clients to fix themselves, just increase the
60186 default number of buttons to 10, which is somewhat a compromise. Ideally,
60187 the XTEST devices should adjust themselves to the highest number of buttons
60188 available on the slave devices (like the master pointers already do), but
60189 that's a taks for another day.
60190
60191 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
60192 Signed-off-by: Keith Packard <keithp@keithp.com>
60193
60194commit ec5417b965616e60b51466fe9fd2b44cb5abb349
60195Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60196Date: Sun Jul 12 23:58:32 2009 +0100
60197
60198 Cygwin/X: Ensure WM_STATE atom exists in multiwindow mode
60199
60200 Workaround a bug in iiimxcf (assuming the WM_STATE atom exists),
60201 which can cause many Solaris clients to simply fail with a BadAtom
60202 error
60203
60204 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60205 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60206
60207commit aa860552fd7e2888258a7b48b8c3bd4af527dc6c
60208Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
60209Date: Sun Nov 1 18:19:08 2009 +0000
60210
60211 Cygwin/X: Enable clipboard integration by default
60212
60213 Enable clipboard integration by default, can be turned off with -noclipboard.
60214 We still accept -clipboard for backwards compatibility. If both are passed,
60215 the last one is accepted (just as other arguments are handled).
60216
60217 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60218
60219commit 87b00ced3d308a9168828c0e38ecffa0640621a0
60220Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60221Date: Sun Nov 1 18:18:48 2009 +0000
60222
60223 Cygwin/X: Setup screen layout in Xinerama mode
60224
60225 Setup screen layout according to the screen window native window
60226 positions in Xinerama mode
60227
60228 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60229 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60230
60231commit 0fe2e8c64e7a5a9140c77b88b266d7ae58b0bca2
60232Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
60233Date: Sun Nov 1 18:18:31 2009 +0000
60234
60235 Cygwin/X: Allow composite to be enabled for Xwin
60236
60237 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60238 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60239
60240commit 26a69bbd9b6f415c443d6d6f8f450329f348edc3
60241Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
60242Date: Sun Nov 1 18:19:04 2009 +0000
60243
60244 Cygwin/X: Mount options have changed in cygwin-1.7
60245
60246 Mount options have changed in cygwin-1.7
60247
60248 Also fix a typo in the warning issued if /tmp is a textmode mount
60249
60250 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60251 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60252
60253commit 35901ece6a49e5f8e245364c27723da2f9009a1d
60254Author: Paul Loewenstein <paul.loewenstein@gmail.com>
60255Date: Sun Nov 1 18:18:45 2009 +0000
60256
60257 Cygwin/X: Handle fake keypresses generated by speech recognizers
60258
60259 Apparently, fake keypresses generated by speech recognizers may not bother
60260 with a scan code, so look up what scan code corresponds to the virtual key
60261 code if this occurs.
60262
60263 Patch by Paul Loewenstein <paul.loewenstein@gmail.com>
60264
60265 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60266 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60267
60268commit 0866322b574b1f3695467535ed5fa8f9e629ad1d
60269Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60270Date: Mon Jul 13 13:46:45 2009 +0100
60271
60272 Cygwin/X: Always use an authorization cookie for internal clients
60273
60274 Don't conditionalize use of an authorization cookie for internal client
60275 threads on XCSECURITY, always use one (this avoids certain problems
60276 with XDMCP setups where the XDMCP host removes localhost from the access
60277 list etc.)
60278
60279 Conditionalize the use of a XCSECURITY authorization descriptor on XCSECURITY
60280
60281 Consolidate the various places where the authorization cookie is set
60282 for internal threads into a new function, winSetAuthorization()
60283
60284 Use authorization cookie for multiwindow WM X message thread as well
60285
60286 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60287
60288commit 72f81f4e449defb0654e9bdb2c9ec014475a3977
60289Author: Colin Harrison <colin.harrison@virgin.net>
60290Date: Fri Nov 6 17:46:58 2009 +0000
60291
60292 Xming: Make -auth option work in with -multiwindow
60293
60294 Use an internally generated cookie for authentication of the internal
60295 window manager client when using the -auth option in -multiwindow mode.
60296
60297 Copyright (C) Colin Harrison 2005-2008
60298 http://www.straightrunning.com/XmingNotes/
60299 http://sourceforge.net/projects/xming/
60300
60301 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60302
60303commit 6dbf8f27c885ea3a761548183fb00ba3df2df553
60304Author: Colin Harrison <colin.harrison@virgin.net>
60305Date: Tue Jul 21 01:15:25 2009 +0100
60306
60307 Xming: Tidy up code for initial native window positioning
60308
60309 Tidy up code for initial native window positioning and avoid a
60310 duplicate call to winMultiWindowGetTransientFor()
60311
60312 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60313
60314commit 28eb61fc04811bb9bbb523e0a252933313bb16ce
60315Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60316Date: Sun Nov 1 18:19:14 2009 +0000
60317
60318 Avoid a null dereference if IFF_BROADCAST is set but there is no broadcast address
60319
60320 It seems that the getifaddrs() function can return interfaces with
60321 IFF_BROADCAST & IFF_UP set, but no broadcast address (at least
60322 under Cygwin 1.7, this seems to happen for v6 mapped v4 addresses)
60323
60324 Avoid a null dereference if this ever happens
60325
60326 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60327 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60328
60329commit 091cbbaed7aafab6f9839a40fe977ab067d1ace1
60330Author: Colin Harrison <colin.harrison@virgin.net>
60331Date: Mon Oct 12 13:40:11 2009 +0100
60332
60333 Xming: Fix UT8String and CompoundText clipboard text sharing with windows clipboard
60334
60335 XConvertSelection() in libX11 always returns 1, so there is no point in
60336 testing it incorrectly against Success. This is possibly a bug in
60337 XConvertSelection()
60338
60339 This should fix UTF8String and CompoundText selection via the clipboard.
60340
60341 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60342
60343commit 9657eae5d65eff289089afc46dfb629758ebbdf7
60344Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60345Date: Sun Nov 1 18:18:51 2009 +0000
60346
60347 Cygwin/X: Clearly diagnose a timeout while waiting for SelectionNotify event
60348
60349 Clearly diagnose a timeout while waiting for SelectionNotify event
60350 in the clipboard integration internal client.
60351 (which seems to be behind some of the reported failures)
60352
60353 Turn useless #if 0/ErrorF()/#endif into useful winDebug()
60354
60355 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60356 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60357
60358commit 062f49a8e0d1afb4dd32e9451a47ab2792639e7f
60359Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60360Date: Wed Nov 4 17:36:54 2009 +0000
60361
60362 Cygwin/X: Add a workaround for a SWT/Motif bug to internal window manager
60363
60364 SWT/Motif expects all top-level windows to get reparented, and waits until they
60365 do. So workaround that in our internal WM by forcing a reparent event to
60366 occur, even though we don't actually need to reparent the window to
60367 frame it (as the frame is a native window, not an X window)
60368
60369 http://sourceware.org/bugzilla/show_bug.cgi?id=9848
60370 https://bugs.eclipse.org/bugs/show_bug.cgi?id=36806
60371
60372 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60373 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60374
60375commit 71519a572fe15b85c0eb2b02636c9e871f2c858f
60376Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60377Date: Wed Nov 4 17:34:40 2009 +0000
60378
60379 Cygwin/X: Fix typo in g_fAnotherWMRunning and tidy up WM detection code
60380
60381 Tidy up code for detecting another WM is already running
60382 Fix typo g_fAnotherWMRunnig -> g_fAnotherWMRunning
60383 Remove some unused event mask macros
60384
60385 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60386 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
60387
60388commit f3fad371cce0f3836514ad5b29e59fa1ca0627a7
60389Author: Colin Harrison <colin.harrison@virgin.net>
60390Date: Wed Nov 4 15:33:53 2009 +0000
60391
60392 Xming: Add FORCEEXIT option to configuration file
60393
60394 Add a new option to configuration file: FORCEEXIT, like SILENTEXIT
60395 but ignores the client count. Unsaved client work may be lost with
60396 this option but it is useful if you want no dialogs.
60397
60398 Add description of this new keyword to XWinrc man page
60399
60400 Also fix grammar of the exit confirmation dialog warning to be correct
60401 when there is only one(1) client connected.
60402
60403 Also rearrange yacc tokens to one per line to make future merges
60404 easier
60405
60406 Also amend default system.XWinrc so that SILENTEXIT is on by default
60407
60408 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60409
60410commit 019a601de6a8ae083852c2d384c814dcc7aeeb72
60411Author: Colin Harrison <colin.harrison@virgin.net>
60412Date: Wed Nov 4 15:16:03 2009 +0000
60413
60414 Xming: Add mouse motion events with flag POINTER_SCREEN in winEnqueueMotion()
60415
60416 Add mouse motion events with flag POINTER_SCREEN in winEnqueueMotion(), as
60417 they are in screen coordindates and may need to be scaled to the axis range
60418 appropriately
60419
60420 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60421
60422commit 7af1240b57daa4ec55512276a9fda499d923eaa0
60423Author: Colin Harrison <colin.harrison@virgin.net>
60424Date: Wed Nov 4 14:20:17 2009 +0000
60425
60426 Xming: Replace all the uses of deprecated functions in hw/xwin with current ones
60427
60428 Replace uses of LookupIDByType() and SecurityLookupIDByType()
60429 with dixLookupResourceByType()
60430
60431 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60432
60433commit 2a38f7c0dbc890e6408eee143d77719f265d583d
60434Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60435Date: Fri Jun 19 21:12:55 2009 +0100
60436
60437 Cygwin/X: Remove a couple of extraneous '\n' in logged version info
60438
60439 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60440
60441commit 56658fc06cb773ed20ef1b23a3fe918c0d3b36bb
60442Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
60443Date: Thu Oct 15 22:40:02 2009 -0500
60444
60445 Set default font path with fontutil 1.1 and fontpath.d
60446
60447 Look for $sysconfdir/X11/fontpath.d and, if it exists, set it to the
60448 default font path as a catalogue: entry. Based on app/xfs commit
60449 c66a46e35ae40a23ad9acee838ab42300eddbd67.
60450
60451 Otherwise, use the font-util 1.1 macros to determine the system
60452 fontrootdir and standard subdirectories, or let the user configure it
60453 and the subdirectories themselves.
60454
60455 This adds a build-time (or at least an autoconf/autogen time) dependency
60456 on font-util, and changes the --with-fontdir argument to --with-fontrootdir.
60457
60458 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
60459 Reviewed-by: Alan Coopersmith <alan.coopersmith-xsfywfwIY+M@public.gmane.org>
60460 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60461
60462commit c7d1319872e0ae0b7d945e9af8c4027ee5aa2055
60463Author: Adam Jackson <ajax@redhat.com>
60464Date: Mon Nov 9 11:12:22 2009 -0500
60465
60466 randr: Fill in errorValue when verifying outputs/crtcs/modes
60467
60468 Signed-off-by: Adam Jackson <ajax@redhat.com>
60469 Signed-off-by: Keith Packard <keithp@keithp.com>
60470
60471commit d85ab6b6483d6ca7d9d22298d05ed1fa3076b042
60472Author: Dave Airlie <airlied@redhat.com>
60473Date: Tue Oct 13 10:15:00 2009 +1000
60474
60475 loader: actually stat something that has some chance of existing. (v2)
60476
60477 FindModuleInSubdir seems to expect a / at the end of the subdir its
60478 finding for, so we add the / early, the stat will fail if its
60479 not a subdir, I'm leaving the S_ISDIR in just in case there is another
60480 reason it could return 0. This does look a bit silly in strace
60481 but it seems to work fine.
60482
60483 I have a very intermittent issue where drivers loses its / that
60484 I've been seeing on/off for a while, this may or may not fix it.
60485
60486 Signed-off-by: Dave Airlie <airlied@redhat.com>
60487 Reviewed-by: Julien Cristau <jcristau@debian.org>
60488 Signed-off-by: Keith Packard <keithp@keithp.com>
60489
60490commit a5e59230de3abafda9cd9d571cea192897155a14
60491Author: Adam Jackson <ajax@redhat.com>
60492Date: Thu Nov 5 13:44:30 2009 -0800
60493
60494 randr: Turn on ModeDebug during server setup
60495
60496 Signed-off-by: Adam Jackson <ajax@redhat.com>
60497 Signed-off-by: Keith Packard <keithp@keithp.com>
60498
60499commit c61e77c77ca309f3676370992f41867807ed0c67
60500Author: Matt Turner <mattst88@gmail.com>
60501Date: Thu Nov 5 22:40:03 2009 -0800
60502
60503 Remove lnx_font.c and lnx.h
60504
60505 I couldn't find any version of the X xserver that ever used lnx_font.c
60506 so let's delete it. I tried contacting its author, Egbert, multiple
60507 times on IRC and email [*] but never got any response. It also hasn't
60508 been seriously touched since January 2005.
60509
60510 [*] http://lists.x.org/archives/xorg-devel/2009-October/002855.html
60511
60512 Signed-off-by: Matt Turner <mattst88@gmail.com>
60513 Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
60514 Signed-off-by: Keith Packard <keithp@keithp.com>
60515
60516commit 15ffe9f51b122494a2b292b3ab1f199d3e81600c
60517Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60518Date: Thu Nov 5 18:28:28 2009 -0800
60519
60520 configure.ac: Notify user about which SHA1 implementation is being used
60521
60522 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60523
60524commit 6b109919f6e1593b27b0760bb56a65b43fb86ea4
60525Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60526Date: Thu Nov 5 18:13:07 2009 -0800
60527
60528 SHA1: Add support for Common Crypto
60529
60530 libSystem on darwin can handle SHA1 computation without needing to pull in
60531 OpenSSL. See CC_crypto(3)
60532
60533 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60534
60535commit 840a68dc5e3b4d285894f86df2a8c41fca5a4bec
60536Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60537Date: Wed Nov 4 18:39:16 2009 -0800
60538
60539 XQuartz: Cleanup X11Controller.m compilation warnings.
60540
60541 Declare X11Controller as implementing NSTableViewDataSource.
60542 Use selectRowIndexes:byExtendingSelection instead of selectRow:byExtendingSelection
60543
60544 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60545
60546commit 0e3ad44c3b05cbe7b1d8f5ce7949fec3c7ae2e7c
60547Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60548Date: Wed Nov 4 18:15:08 2009 -0800
60549
60550 XQuartz: Use dixLookupResourceByType instead of LookupIDByType
60551
60552 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60553
60554commit 069fc6ce0a0e7f2c418e11941568ffcc52b6b331
60555Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60556Date: Wed Nov 4 16:18:06 2009 -0800
60557
60558 XQuartz: Don't weed out duplicates in generated keymap
60559
60560 There seems to be an issue in the 1.5+ server where shift-space does not
60561 produce a space when 'keycode 57 = space' but it does when 'keycode 57 = space
60562 space'
60563
60564 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60565
60566commit 6d6e8fb27f00f0c3128cef624b39a60aa754fdc8
60567Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60568Date: Wed Nov 4 13:34:20 2009 -0800
60569
60570 XQuartz: Controller thread launches clients
60571
60572 This avoids a memory leak due to no active auto-release pool on the server thread.
60573
60574 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60575
60576commit f7b375bd141d0cf1e3add5443a5838dd8f554ef6
60577Author: Alan Coopersmith <alan.coopersmith@sun.com>
60578Date: Thu Nov 5 11:42:34 2009 -0800
60579
60580 Enable XF86PM on all Solaris platforms, not just x86/x64
60581
60582 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
60583 Signed-off-by: Keith Packard <keithp@keithp.com>
60584
60585commit f77262513ea17401092479e17ad20fc0eb91ffb5
60586Author: Alan Coopersmith <alan.coopersmith@sun.com>
60587Date: Thu Nov 5 11:42:33 2009 -0800
60588
60589 Use $(MAKE) instead of "make" to build Solaris inline assembly
60590
60591 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
60592 Signed-off-by: Keith Packard <keithp@keithp.com>
60593
60594commit 964040764387d89ef64324cfbee31872ee6ce41c
60595Author: Tiago Vignatti <tiago.vignatti@nokia.com>
60596Date: Wed Nov 4 15:35:37 2009 +0200
60597
60598 EXA: Preserve pPixmap->devPrivate.ptr in exaPixmapIsOffscreen_driver
60599
60600 It crash the server when the drawable is 32 bit and the framebuffer is 16.
60601
60602 This is pretty much a copy-past from commit 8e873185.
60603
60604 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
60605 Signed-off-by: Keith Packard <keithp@keithp.com>
60606
60607commit 7442f3355ab8f0bb2b1a270da18c65d8d315d4dd
60608Merge: 0573042 7897b6c
60609Author: Keith Packard <keithp@keithp.com>
60610Date: Wed Nov 4 08:55:20 2009 -0800
60611
60612 Merge remote branch 'jeremyhu/master'
60613
60614commit 0573042cddb6f9942e408687a16c6842e62a8bfa
60615Author: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
60616Date: Wed Nov 4 15:59:30 2009 +0200
60617
60618 dix: Fixes a memory leak when a cursor resource is released.
60619
60620 Just open and close a client that creates cursors in order to
60621 reproduce. In the problem case bits->refcnt is -1 and therefore
60622 bits->devPrivates is never released.
60623
60624 Signed-off-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
60625 Signed-off-by: Keith Packard <keithp@keithp.com>
60626
60627commit 7897b6c2d41bccb72c19418674c3526ecce29515
60628Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60629Date: Tue Oct 27 18:00:48 2009 -0700
60630
60631 dix: Properly detect if the other device is frozen
60632
60633 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60634
60635commit 5e79976c13c5b94b12392b493846ca26be11750b
60636Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
60637Date: Tue Nov 3 16:35:27 2009 -0800
60638
60639 XQuartz: Run xmodmap after programatically updating the keymap.
60640
60641 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
60642 Signed-off-by: Martin Otte <otte@duke.edu>
60643
60644commit 412c56ef3332d09efbd861e41c3e985f44729729
60645Author: Maarten Maathuis <madman2003@gmail.com>
60646Date: Sun Nov 1 22:08:40 2009 +0100
60647
60648 exa: remove some outdated comment
60649
60650 - This comment is still in exa_driver.c and there it makes sense.
60651
60652 Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
60653 Acked-by: Michel Dänzer <michel@daenzer.net>
60654 Signed-off-by: Keith Packard <keithp@keithp.com>
60655
60656commit 48b8c076a7b7dbbddfc472e8c6e5251a0b174cbc
60657Author: Matthieu Herrb <matthieu@laas.fr>
60658Date: Mon Nov 2 11:11:24 2009 +0100
60659
60660 add libc as a choice for SHA1 implementation
60661
60662 On Sun, Nov 01, 2009 at 02:54:13PM -0800, Keith Packard wrote:
60663 > Excerpts from Matthieu Herrb's message of Sun Nov 01 09:34:35 -0800 2009:
60664 >
60665 > > +AC_CHECK_FUNCS([SHA1Init], [HAVE_LIBC_SHA1=yes])
60666 >
60667 > I'd suggest AC_CHECK_FUNC instead; as far as I can tell, AC_CHECK_FUNCS
60668 > will also define HAVE_SHA1INIT. Also, can you use HAVE_LIBC_SHA1
60669 > consistently rather than having two separate names (HAVE_LIBC_SHA1 and
60670 > HAVE_SHA1_IN_LIBC)? Yes, I know one is a preprocessor symbol and the
60671 > other is a cpp shell variable, but I think that will work anyway.
60672 >
60673 New version taking you comments into account.
60674
60675 From: Matthieu Herrb <matthieu.herrb@laas.fr>
60676 Date: Sun, 1 Nov 2009 18:19:27 +0100
60677 Subject: [PATCH] Add a probe for SHA1 functions in libc in *BSD.
60678
60679 The interface is the same as the one in libmd.
60680
60681 Signed-off-by: Keith Packard <keithp@keithp.com>
60682
60683commit da923d0bc15e99a8ed1986bd6f5df37f7af7284b
60684Author: Matt Turner <mattst88@gmail.com>
60685Date: Sun Nov 1 14:30:48 2009 -0500
60686
60687 Make sys.c use unaligned access functions provided in compiler.
60688
60689 Favorite deleted line was definitely
60690 /* to cope with broken egcs-1.1.2 :-(((( */
60691
60692 Signed-off-by: Matt Turner <mattst88@gmail.com>
60693 Signed-off-by: Keith Packard <keithp@keithp.com>
60694
60695commit d30637339963950910e5f5fb755b8465ac7dddb4
60696Author: Mikhail Gusarov <dottedmag@dottedmag.net>
60697Date: Thu Oct 29 11:46:22 2009 -0700
60698
60699 Supply all code using dl*() with DLOPEN_LIBS
60700
60701 Previously DLOPEN_LIBS was managed in top-level configure.ac.
60702 Instead bundle it with the code using dl*() functions to
60703 avoid breakages in uncommon configurations.
60704
60705 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
60706 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
60707 Signed-off-by: Keith Packard <keithp@keithp.com>
60708
60709commit ec98d7fc78efefcf9fc61492529157c0d289c3f2
60710Author: Adam Jackson <ajax@redhat.com>
60711Date: Mon Sep 21 10:23:16 2009 -0400
60712
60713 EDID: Extend the HDTV hack to handle "1368x769"
60714
60715 Hate televisions so much.
60716
60717 Signed-off-by: Adam Jackson <ajax@redhat.com>
60718
60719commit 7c0803f555782dbf451b7c79112d7deae02e5c9f
60720Author: Adam Jackson <ajax@redhat.com>
60721Date: Wed Oct 28 15:44:37 2009 -0400
60722
60723 modes: Fix duplicate detection, and do it more consistently
60724
60725 Signed-off-by: Adam Jackson <ajax@redhat.com>
60726
60727commit ba2d39dd5428cb5922b797a1d4ea45b859412b40
60728Author: Adam Jackson <ajax@redhat.com>
60729Date: Mon Oct 26 15:10:30 2009 -0400
60730
60731 modes: De-duplicate a clock range check.
60732
60733 Signed-off-by: Adam Jackson <ajax@redhat.com>
60734
60735commit 25236d19e6ef07fcb2c71569f1b7b0c12810834a
60736Author: Adam Jackson <ajax@redhat.com>
60737Date: Mon Oct 26 14:49:57 2009 -0400
60738
60739 EDID: Fix interlaced detailed timings to be frame size, not field size
60740
60741 Signed-off-by: Adam Jackson <ajax@redhat.com>
60742
60743commit fb86433d897c116315cc7994390d11ac2f577511
60744Author: Adam Jackson <ajax@redhat.com>
60745Date: Mon Oct 26 14:04:23 2009 -0400
60746
60747 modes: Decorate interlaced mode names with a trailing 'i'
60748
60749 Signed-off-by: Adam Jackson <ajax@redhat.com>
60750
60751commit fc2ec95664d55f45f77f1ebb039a7c17a1fcdaa3
60752Author: Ma Ling <ling.ma@intel.com>
60753Date: Wed Feb 18 17:41:26 2009 +0800
60754
60755 EDID: CEA extension support
60756
60757 Reviewed-by: Adam Jackson <ajax@redhat.com>
60758
60759commit fab74d1081270fb8f1d231e6e10d10aa33e164da
60760Author: Jamey Sharp <jamey@minilop.net>
60761Date: Wed Oct 28 17:48:19 2009 -0700
60762
60763 Suppress certain GCC warnings in auto-generated code.
60764
60765 - Don't warn for references to deprecated functions in xorg_symbols.
60766 - Ignore functions generated by gl_apitemp.py that are never used.
60767
60768 Signed-off-by: Jamey Sharp <jamey@minilop.net>
60769 Signed-off-by: Keith Packard <keithp@keithp.com>
60770
60771commit 239435875d6a92ed31731b500a992a3af0943594
60772Author: Jamey Sharp <jamey@minilop.net>
60773Date: Wed Oct 28 17:48:18 2009 -0700
60774
60775 Don't cast double to int: use default conversions or explicitly round.
60776
60777 GCC warns about casting a double return value to int.
60778
60779 Signed-off-by: Jamey Sharp <jamey@minilop.net>
60780 Signed-off-by: Keith Packard <keithp@keithp.com>
60781
60782commit e8c48fd8f7aab54327b0091cd17c60235ae27168
60783Author: Jamey Sharp <jamey@minilop.net>
60784Date: Wed Oct 28 17:48:17 2009 -0700
60785
60786 Suppress GCC warnings like "the address of `u1' will always evaluate as `true'".
60787
60788 Signed-off-by: Jamey Sharp <jamey@minilop.net>
60789 Signed-off-by: Keith Packard <keithp@keithp.com>
60790
60791commit 662594aeff9d1767316f08600949c73ac5060d18
60792Author: Mikhail Gusarov <dottedmag@dottedmag.net>
60793Date: Thu Oct 29 00:40:48 2009 +0600
60794
60795 kdrive: Grab evdev mouse/keyboard devices when X server is active
60796
60797 Input events are directed to both vt and input devices by default.
60798 Unless input devices are grabbed, keyboard events fill it vt buffers
60799 and cause spontaneous wakeups in kernel tty layer when buffers are full.
60800
60801 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
60802 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
60803 Signed-off-by: Keith Packard <keithp@keithp.com>
60804
60805commit 83d520d86c888f2c8a4abb18b7a8858d568aa18f
60806Author: Jamey Sharp <jamey@minilop.net>
60807Date: Wed Oct 28 16:45:44 2009 -0700
60808
60809 Add video driver flag to indicate that console access is not needed.
60810
60811 Existing video drivers will get the console enabled by default.
60812
60813 Signed-off-by: Jamey Sharp <jamey@minilop.net>
60814 Signed-off-by: Keith Packard <keithp@keithp.com>
60815
60816commit 25979c46b467847ccb54f5c86a1be6b9c303c99a
60817Author: Jamey Sharp <jamey@minilop.net>
60818Date: Wed Oct 28 16:35:28 2009 -0700
60819
60820 Alloc/free drawables array for each ProcPanoramiXShmGetImage call.
60821
60822 Updates my previous patch, b422b532f3dcab54c53f61a66f2ad76059d1874a.
60823 keithp recommended against allocating the drawables array globally, but my
60824 updated patch with that fixed isn't the patch that landed.
60825
60826 Signed-off-by: Jamey Sharp <jamey@minilop.net>
60827 Signed-off-by: Keith Packard <keithp@keithp.com>
60828
60829commit 757c11630d5999ad13dcac79191429badc92a3a6
60830Author: Tormod Volden <lists.tormod@gmail.com>
60831Date: Wed Oct 28 23:10:43 2009 +0100
60832
60833 xfree86: Fix description of DDC_QUIRK_DETAILED_USE_MAXIMUM_SIZE
60834
60835 The message ending up in the log is misleading as to what the quirk
60836 actually does: It ignores the sizes in the detailed timings and
60837 replaces them with the display "Max Image Size".
60838
60839 Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
60840 Signed-off-by: Keith Packard <keithp@keithp.com>
60841
60842commit ccf4a69db747b000aee09072aac0a2891bde139a
60843Author: Mikhail Gusarov <dottedmag@dottedmag.net>
60844Date: Thu Oct 29 01:54:00 2009 +0600
60845
60846 os: Add libsha1 as a choice of SHA1 implementation
60847
60848 There are small systems which don't need OpenSSL or gcrypt.
60849 Add libsha1 (http://github.com/dottedmag/libsha1) as an alternative
60850 small SHA1 implementation.
60851
60852 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
60853 Signed-off-by: Keith Packard <keithp@keithp.com>
60854
60855commit 8613e4b0eb04150b1e377871f02b164be5d001e9
60856Author: Mikhail Gusarov <dottedmag@dottedmag.net>
60857Date: Wed Oct 28 11:44:27 2009 -0700
60858
60859 Add missing DLOPEN_LIBS to kdrive compilation flags
60860
60861 Xfbdev failed to build due to dladdr being used by xorg_backtrace.
60862 Explicitly add DLOPEN_LIBS to KDRIVE_LIBS as there does not
60863 seem to be a better place for it.
60864
60865 Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
60866 Signed-off-by: Keith Packard <keithp@keithp.com>
60867
60868commit deb72fc61464250af8185dab2da8ee09f13c55d8
60869Merge: 55f4c80 a60e676
60870Author: Keith Packard <keithp@keithp.com>
60871Date: Wed Oct 28 10:54:13 2009 -0700
60872
60873 Merge remote branch 'jcristau/sha1'
60874
60875commit 55f4c80a4c891b355a99e6a05978ca945397c5cc
60876Author: Hans Nieser <hnsr@xs4all.nl>
60877Date: Wed Oct 28 06:59:40 2009 +0100
60878
60879 Xinput: allow non-integer values again for Constant- and AdaptiveDeceleration
60880
60881 This was initially fixed by commit 3932a848572f4eaf8b7f1d91d9b74aeafab069a2
60882 but then (presumably not intentionally) undone by commit
60883 1d54479cb3c8b4f75b7564f8b5e1c5da940b20f4 .
60884
60885 Signed-off-by: Hans Nieser <hnsr@xs4all.nl>
60886 Signed-off-by: Keith Packard <keithp@keithp.com>
60887
60888commit 50a5c32430a5267f2a05656d2417f9a8a44d8b97
60889Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60890Date: Fri Oct 23 19:32:42 2009 +0100
60891
60892 dmx: Correctly compute DMXGetScreenAttributes reply length
60893
60894 Correctly allow for excess length of DMXGetScreenAttributes reply
60895 over standard 32 byte reply in addition to the displayName string
60896 when computing the length of reply
60897
60898 http://bugs.freedesktop.org/show_bug.cgi?id=24685
60899
60900 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
60901 Reviewed-by: Julien Cristau <jcristau@debian.org>
60902 Signed-off-by: Keith Packard <keithp@keithp.com>
60903
60904commit 9a2f6135bfb0f12ec28f304c97917d2f7c64db05
60905Author: Keith Packard <keithp@keithp.com>
60906Date: Fri Oct 23 10:04:57 2009 +0900
60907
60908 DRI2: Report the correct extension minor version
60909
60910 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
60911 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
60912 Signed-off-by: Keith Packard <keithp@keithp.com>
60913
60914commit d886008c96cd16f735f54dace070cb00b23c6f44
60915Author: Marcin Baczyński <marbacz@gmail.com>
60916Date: Thu Oct 22 14:43:25 2009 +0200
60917
60918 Kill compilation warnings.
60919
60920 Signed-off-by: Marcin Baczyński <marbacz@gmail.com>
60921 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
60922 Signed-off-by: Keith Packard <keithp@keithp.com>
60923
60924commit 15b30fde179cba3877182cd51b0f870ef29ffaee
60925Author: Adam Jackson <ajax@redhat.com>
60926Date: Thu Oct 22 17:29:01 2009 -0400
60927
60928 dix: Fix up colormap fixup.
60929
60930 FindClientResourcesByType() will walk all colormaps on all screens; we
60931 only want to fix up the current screen. Otherwise, screens > 0 will
60932 have the visual pointers for their colormaps pointing off into space.
60933
60934 Signed-off-by: Adam Jackson <ajax@redhat.com>
60935
60936commit 3785475a78636eb6547ef9e46be9e009c7cf7800
60937Author: Keith Packard <keithp@keithp.com>
60938Date: Fri Oct 23 09:03:39 2009 +0900
60939
60940 Bump to 1.7.99.2 (unreleased)
60941
60942commit 909df9beb3ddd02632f36ae682537280a6a8e5b4
60943Author: Jon TURNEY <jon.turney@dronecode.org.uk>
60944Date: Tue Oct 20 12:57:05 2009 +0100
60945
60946 Resolve an inconsistency between libX11 and Xserver over GetModifierMapping
60947
60948 libX11 ModMap.c believes that GetModifierMapping can never return an error
60949
60950 Xserver devices.c believes that GetModifierMapping can return an error if
60951 the ModMap couldn't be generated
60952
60953 According to the protocol document I have, libX11 is right, so adjust the
60954 server to send back an empty modmap if one couldn't be made...
60955
60956 http://bugs.freedesktop.org/show_bug.cgi?id=24621
60957
60958 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
60959 Signed-off-by: Keith Packard <keithp@keithp.com>
60960
60961commit 9c48862ac1ac119b6cfb7e376533f53af6a857f4
60962Author: Keith Packard <keithp@keithp.com>
60963Date: Thu Oct 22 01:12:03 2009 +0900
60964
60965 Bump version to 1.7.99.1
60966
60967 This is a development snapshot for 1.8.
60968
60969 Signed-off-by: Keith Packard <keithp@keithp.com>
60970
60971commit 26f4d8a2c12ff06c4beb06842cd8d9c0d9ddd4aa
60972Author: Keith Packard <keithp@keithp.com>
60973Date: Thu Oct 22 01:04:42 2009 +0900
60974
60975 Make sure dmx docs are built for distribution.
60976
60977 Ok, dmx docs are driving me slightly nuts. We probably shouldn't
60978 include the built versions in the tarball, but we do, so this is an
60979 attempt to make that work by having both the 'all' and 'dist' targets
60980 depends on the doxygen output.
60981
60982 Signed-off-by: Keith Packard <keithp@keithp.com>
60983
60984commit 4114854893ff57a1be525ec099a4b33e5f9963b1
60985Author: Keith Packard <keithp@keithp.com>
60986Date: Thu Oct 22 00:17:37 2009 +0900
60987
60988 Don't attempt to build hw/kdrive/sdl/Makefile anymore
60989
60990 With sdl removed, there's no way to build the sdl Makefile.
60991
60992 Signed-off-by: Keith Packard <keithp@keithp.com>
60993
60994commit 4d333c5121818754356853724333eadec2dcd18c
60995Author: Keith Packard <keithp@keithp.com>
60996Date: Thu Oct 22 00:16:12 2009 +0900
60997
60998 Always check for doxygen so that distcheck gets dmx docs
60999
61000 make distcheck wants the built dmx documentation so that users don't
61001 have to install doxygen. This means that even if dmx isn't built, the
61002 docs need to be so that the tarball can include them.
61003
61004 Signed-off-by: Keith Packard <keithp@keithp.com>
61005
61006commit 7e92bac5f769aca99bd20e21fe2811f0480b647b
61007Author: Keith Packard <keithp@keithp.com>
61008Date: Thu Oct 22 00:00:35 2009 +0900
61009
61010 Make sure HAVE_DOXYGEN is defined when not building dmx
61011
61012 The DMX docs are build using doxygen if present, so configure.ac
61013 checks to see if that is available. However, when not building dmx
61014 (the default), this conditional must still be defined to make automake
61015 happy.
61016
61017 Signed-off-by: Keith Packard <keithp@keithp.com>
61018
61019commit 1228e2d052f0bb98175c55c194340773b5fedb40
61020Merge: 08e7f62 52bc6d9
61021Author: Keith Packard <keithp@keithp.com>
61022Date: Wed Oct 21 22:46:53 2009 +0900
61023
61024 Merge remote branch 'whot/master'
61025
61026commit 08e7f62faf72540cb3a6f1023024c145f7fa1a23
61027Author: Keith Packard <keithp@keithp.com>
61028Date: Wed Oct 21 16:46:55 2009 +0900
61029
61030 Fix 'distcheck' to use host xkb files but install to build dir
61031
61032 'make distcheck' needs to read xkb files and write out compiled
61033 versions as a part of the 'make check' phase. This patch passes
61034 suitable options to the configure stage of the distcheck process to
61035 read xkb files from the system location and write them to the
61036 distcheck _inst directory.
61037
61038 Signed-off-by: Keith Packard <keithp@keithp.com>
61039
61040commit 52bc6d944946e66ea2cc685feaeea40bb496ea83
61041Author: Peter Hutterer <peter.hutterer@who-t.net>
61042Date: Fri Oct 16 11:03:09 2009 +1000
61043
61044 kdrive: Purge Xsdl
61045
61046 From the original Xsdl commit:
61047 "sdl x server so that we can x-on-x the fb stuff for ease of debugging. if
61048 anyone uses this in production, a big scary monster will eat them.
61049 hrm, perhaps i should make it have a --i-know-what-i'm-doing
61050 param that it doens't start without, heh"
61051
61052 That should be reason enough to not spend time maintaing it. Also, no more
61053 elephants.
61054
61055 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61056
61057commit f713f447a2110718dfc091380699362d76f0cd6c
61058Author: Lee Leahu <freedesktop-bugs@dyweni.com>
61059Date: Mon Oct 19 15:43:59 2009 -0500
61060
61061 dmxDestroyWindow() - must call the X's native DetroyWindow()
61062
61063 Don't really know why this section was disabled, but without it,
61064 certain pPicture resources do not get free'd until later in the
61065 FreeClientResources() process after the screen has been free'd -
61066 resulting in seg fault.
61067
61068 With this patch, all resources normally free'd using vanilla X are
61069 now also being freed correctly by Xdmx.
61070
61071 https://bugs.freedesktop.org/show_bug.cgi?id=24576
61072
61073 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61074
61075commit 664a8e37fd83141974b772980f680b94e48b4f87
61076Author: Lee Leahu <freedesktop-bugs@dyweni.com>
61077Date: Sat Oct 17 00:45:44 2009 -0500
61078
61079 dmx: when setting up device axis, use the correct counter number
61080
61081 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61082
61083commit 010d5e28655d619693632938e9e4325eef938295
61084Author: Peter Hutterer <peter.hutterer@who-t.net>
61085Date: Fri Oct 16 10:01:13 2009 +1000
61086
61087 dmx: remove doxygen-generated files.
61088
61089 These can be recreated by simply running 'doxygen doxygen.conf' in
61090 hw/dmx/doc. Some of the files do not exist anymore, these have been removed.
61091 Some other files have a different naming scheme.
61092 Doxygen warnings about missing links fixed, two warnings remain:
61093
61094 /home/whot/xorg/xserver/hw/dmx/dmxwindow.c:142: Warning: explicit link
61095 request to 'dmxConfigureRootWindow' could not be resolved
61096 /home/whot/xorg/xserver/hw/dmx/dmxwindow.c:119: Warning: explicit link
61097 request to 'dmxConfigureScreenWindow()' could not be resolved
61098
61099 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61100
61101commit ff3e171568c8b604fc5eacd5f7e9b7ee9417f172
61102Author: Peter Hutterer <peter.hutterer@who-t.net>
61103Date: Fri Oct 16 10:26:21 2009 +1000
61104
61105 xkb: don't conditionally include xkb-config.h.
61106
61107 If HAVE_XKB_CONFIG_H is ever undefined, we fail to build anyway.
61108
61109 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61110
61111commit 0b9dbd4bf2d1061514bebd28648538a5fc0b60e6
61112Author: Peter Hutterer <peter.hutterer@who-t.net>
61113Date: Fri Oct 16 08:24:37 2009 +1000
61114
61115 Use the default XKB settings for dmx and kdrive.
61116
61117 Drop the dmx-specific defines, there's no reason to have separate ones
61118 considering they're about as hardcoded as the default rules anyway.
61119
61120 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61121
61122commit e08d8a2b043e5be821c5e8b2dbf37b09102b5d01
61123Author: Peter Hutterer <peter.hutterer@who-t.net>
61124Date: Thu Sep 24 11:25:28 2009 +1000
61125
61126 kdrive: silence tslib compiler warnings
61127
61128 tslib.c: In function 'TslibInit':
61129 tslib.c:157: warning: unused variable 'tsDev'
61130 tslib.c:156: warning: unused variable 'inputent'
61131 tslib.c:155: warning: unused variable 'inputdir'
61132 tslib.c:154: warning: unused variable 'i'
61133 tslib.c:154: warning: unused variable 'fd'
61134
61135 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61136
61137commit 522ca8179ad2b2b935993f08c9382cd5ad1d8b2f
61138Author: Mikhail Gusarov <dottedmag@dottedmag.net>
61139Date: Tue Oct 20 04:18:01 2009 +0700
61140
61141 Remove unused LinuxFindPci and LinuxGetPciCfg
61142
61143 These two functions are not referenced from inside xserver.
61144 Remove now-empty klinux.h too.
61145
61146 Signed-off-by: Keith Packard <keithp@keithp.com>
61147
61148commit 63f4bf39170eb2262617ef2dc95fd6d337b9dad5
61149Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61150Date: Mon Oct 19 13:07:26 2009 +1000
61151
61152 Fix make dist after 78c87bdad1feab91c2a39e01513b4b0826665f06
61153
61154 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61155 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61156 Signed-off-by: Keith Packard <keithp@keithp.com>
61157
61158commit 2bd71a6db241924bd17ac9dd4aa38e7e563ee5c1
61159Author: Alan Coopersmith <alan.coopersmith@sun.com>
61160Date: Fri Oct 16 22:29:26 2009 -0700
61161
61162 Fix segfault when -extension XKEYBOARD is passed on the command line
61163
61164 Users should be told they can't disable XKB or XInput via error messages,
61165 not core dumps.
61166
61167 Reported by T`2 on #xorg irc
61168
61169 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61170 Signed-off-by: Keith Packard <keithp@keithp.com>
61171
61172commit a32b2420d85f076282721afe005e85fff7d9837e
61173Author: Peter Hutterer <peter.hutterer@who-t.net>
61174Date: Thu Oct 15 13:50:36 2009 +1000
61175
61176 kdrive: fix Xfake build by removing the old keysym stuff
61177
61178 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61179 Acked-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61180 Signed-off-by: Keith Packard <keithp@keithp.com>
61181
61182commit 7c31dd5db8b43c7796bf97a07e08213af5afd2ae
61183Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61184Date: Thu Oct 15 04:24:25 2009 -0500
61185
61186 Remove CopyISOLatin1Lowered
61187
61188 This function was moved verbatim into libXfont-1.4, and it is not used
61189 by the server or any drivers. Exporting it in both places leads to
61190 multiple definition linking errors on Cygwin, where we need to use a
61191 static libXfont due to poor weak-symbol handling.
61192
61193 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61194 Signed-off-by: Keith Packard <keithp@keithp.com>
61195
61196commit 9bc4e88d84daf0f4faf0599b575675e74c75f4b8
61197Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61198Date: Thu Oct 15 01:53:43 2009 -0500
61199
61200 Define ddxBeforeReset stubs in platform-neutral DDXs
61201
61202 XWin uses ddxBeforeReset, which is called in DIX. Other DDXs need to
61203 define these in order to avoid an undefined symbol error at link time
61204 when building alongside XWin. Xnest and Xvfb already provide empty stubs;
61205 this does the same for Xdmx and the platform-neutral KDrive servers.
61206
61207 Also add a prototype to avoid a warning in all DDXs.
61208
61209 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61210 Signed-off-by: Keith Packard <keithp@keithp.com>
61211
61212commit a2f27b97900f335cd5f6a3e5bf8fa1d0ec9eb9ac
61213Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61214Date: Thu Oct 15 01:53:42 2009 -0500
61215
61216 Remove duplicates from Xfake_LDADD
61217
61218 KDRIVE_LIBS already contains the libs in XSERVER_LIBS, so linking against
61219 both leads to multiple-definition errors when linking on Cygwin.
61220
61221 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61222 Signed-off-by: Keith Packard <keithp@keithp.com>
61223
61224commit ed4c6bc79a797830895ad6c3601318b2d1c4bdcb
61225Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61226Date: Thu Oct 15 01:53:41 2009 -0500
61227
61228 Fix XWIN_LIBS definition
61229
61230 Move XWIN_LIBS from XWIN=auto conditional to XWIN=yes conditional,
61231 otherwise the build breaks if an explicit --enable-xwin is passed to
61232 configure.
61233
61234 Also remove obsolete library values from the definition.
61235
61236 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61237 Signed-off-by: Keith Packard <keithp@keithp.com>
61238
61239commit bcd0c02cfd0707b5f8b1ca91dfbe4412df2c643e
61240Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61241Date: Thu Oct 15 01:53:40 2009 -0500
61242
61243 Install SDK pkg-config and aclocal files only if XORG
61244
61245 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61246 Signed-off-by: Keith Packard <keithp@keithp.com>
61247
61248commit 264ce9e8360374b3a43442c8bdea08abde705446
61249Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61250Date: Wed Oct 14 20:33:53 2009 -0400
61251
61252 xselinux: Use the now-exported IsPointerDevice() instead of a copy.
61253
61254 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61255
61256commit 0ae1632be045bfbb288bb57190c830f94247460f
61257Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61258Date: Wed Oct 14 20:32:28 2009 -0400
61259
61260 dix: Export IsPointerDevice() and IsKeyboardDevice().
61261
61262 Makes the functions available to extmod for extensions to call.
61263
61264 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61265
61266commit 6c2ae5fec552366e11ad64a27626eb5dec4becf0
61267Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61268Date: Tue Sep 22 13:31:49 2009 -0700
61269
61270 xselinux: switch from x_device to separate x_pointer and x_keyboard classes.
61271
61272 This will allow separate controls over pointer and keyboard without having
61273 to relabel the devices to separate types.
61274
61275 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61276
61277commit 4be354c4c2da5168b302601b91bd80cfaca7e193
61278Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61279Date: Tue Sep 22 13:18:44 2009 -0700
61280
61281 Remove some debug messages that trigger on XACE event delivery failure.
61282
61283 It is normal for XACE to deny an event delivery, so these log messages
61284 shouldn't trigger when that happens. Just drop them for now.
61285
61286 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61287
61288commit c4ffce4dc84a0a9d134a59b7e7765c99ed767e53
61289Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61290Date: Tue Sep 22 13:13:03 2009 -0700
61291
61292 xace: Relax permissions on XkbGetState from Read to Getattr.
61293
61294 This request is used to get the current keyboard group and is called from
61295 GTK. It does not return an actual keymap (aside from modifiers) so it
61296 should be safe to relax the permission on it. However it does return
61297 button state information which should be controlled through a separate
61298 pointer Read check.
61299
61300 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61301
61302commit 0ff28319906eeb3f236acd72201c416ce01f2c6e
61303Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61304Date: Tue Sep 15 19:41:04 2009 -0400
61305
61306 xselinux: Stop special-casing QueryPointer access checks.
61307
61308 XACE has been changed to not return BadAccess on device read failures.
61309 Thus, no need for this workaround code.
61310
61311 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61312
61313commit 8502c06e19a4c00bf1311f54f9a365ee9e026e97
61314Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61315Date: Tue Sep 15 19:29:34 2009 -0400
61316
61317 xace: Fake return values on denials in input polling requests.
61318
61319 Instead of returning BadAccess when "read" permission is denied
61320 on a device, falsify the device state (buttons down, keys pressed).
61321 This is nicer to applications, but may still have undesired side
61322 effects. The long-term solution is not to use these requests in
61323 event-driven code!
61324
61325 Requests affected: QueryPointer, QueryKeymap, XiQueryDevice.
61326
61327 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61328
61329commit 0493935691e925ae137af7636fa15befa76c8b45
61330Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61331Date: Mon Aug 31 21:20:53 2009 -0400
61332
61333 xselinux: Factor out some dynamic array code into common helpers.
61334
61335 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61336
61337commit ae8891ba0b63bfe6941a324e201d9ab7c645c0f3
61338Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
61339Date: Mon Aug 31 18:46:23 2009 -0400
61340
61341 xselinux: refactor extension code into smaller files.
61342
61343 New files:
61344 xselinux_ext.c: Extension init and request handlers.
61345 xselinux_hooks.c: XACE hook functions and other callbacks.
61346 xselinux_label.c: Object security-labeling code.
61347 xselinuxint.h: Shared internal functions.
61348
61349 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61350
61351commit a60e676f1fd243c78859440b87652f523d3f2ec1
61352Author: Julien Cristau <jcristau@debian.org>
61353Date: Wed Oct 14 23:51:22 2009 +0200
61354
61355 Add libgcrypt as an option for SHA1
61356
61357 Signed-off-by: Julien Cristau <jcristau@debian.org>
61358 Reviewed-by: Rémi Cardona <remi@gentoo.org>
61359
61360commit d2a6a395435919aff8943285f9cbfe6569a9728f
61361Author: Julien Cristau <jcristau@debian.org>
61362Date: Wed Oct 14 23:30:55 2009 +0200
61363
61364 configure: add --with-sha1={libmd,libcrypto} option
61365
61366 Signed-off-by: Julien Cristau <jcristau@debian.org>
61367 Reviewed-by: Rémi Cardona <remi@gentoo.org>
61368
61369commit 55516094947dd78ad2734bb784a2fb109b64c990
61370Author: Julien Cristau <jcristau@debian.org>
61371Date: Wed Oct 14 23:20:44 2009 +0200
61372
61373 Move SHA1 computation from render/glyph.c to os/
61374
61375 Signed-off-by: Julien Cristau <jcristau@debian.org>
61376 Reviewed-by: Rémi Cardona <remi@gentoo.org>
61377
61378commit 22b38f513c93c2f2aea5909878c3c9acae7d35c1
61379Author: Jon TURNEY <jon.turney@dronecode.org.uk>
61380Date: Sat Oct 3 12:44:50 2009 +0100
61381
61382 Cygwin/X: Avoid a potential null pointer dereference before input initialization
61383
61384 Avoid a null pointer dereference if WM_MOUSEMOVE occurred before
61385 the input device had been initialized (a timing sensitive bug
61386 occassionally seen during initialization)
61387
61388 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
61389 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
61390
61391commit 1b0dfd8dee639870725d3bd9b70c3bd589d09e5a
61392Author: Jon TURNEY <jon.turney@dronecode.org.uk>
61393Date: Tue Jul 21 02:23:21 2009 +0100
61394
61395 Cygwin/X: Make -logverbose affect the verbosity of logging to the log file
61396
61397 Make -logverbose affect the verbosity of logging to the log file, not just the
61398 verbosity of logging to the console
61399
61400 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
61401 Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
61402
61403commit 83d120b90482d356be730f63aead0f8e44a4e846
61404Author: Colin Harrison <colin.harrison@virgin.net>
61405Date: Thu Oct 1 14:47:22 2009 +0100
61406
61407 Xming: Fix various 'ISO C90 forbids mixed declarations and code' warnings
61408
61409 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
61410
61411commit 78c87bdad1feab91c2a39e01513b4b0826665f06
61412Author: Colin Harrison <colin.harrison@virgin.net>
61413Date: Fri Sep 25 17:49:59 2009 +0100
61414
61415 Xming: Remove unused X-boxed icon resource
61416
61417 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
61418
61419commit fdf7f7e95c852734c620f58df408a324902172cf
61420Author: Colin Harrison <colin.harrison@virgin.net>
61421Date: Fri Sep 25 17:46:40 2009 +0100
61422
61423 Xming: update .rc file
61424
61425 Remove the obsolete DISCARDABLE flag in the .rc file
61426 Replace the obsolete DIALOG resource with DIALOGEX in the .rc file
61427
61428 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
61429
61430commit 5ddfd9d0898629627e8ab10e65ea2e082be3af37
61431Author: Jon TURNEY <jon.turney@dronecode.org.uk>
61432Date: Sat Oct 3 13:07:23 2009 +0100
61433
61434 Cygwin/X: Don't build rootless extension code unless we need it
61435
61436 Only try to build rootless extension code if we really need it (i.e. mwextwm mode is enabled)
61437
61438 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
61439
61440commit 25344ba7f7845654364d62bf15322b3b79465bd9
61441Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61442Date: Tue Oct 13 20:14:58 2009 -0500
61443
61444 dolt: add Cygwin to supported platforms
61445
61446 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
61447 Signed-off-by: Keith Packard <keithp@keithp.com>
61448
61449commit 6e158003e80534ce007290f75c89d698aec1d00b
61450Merge: 2b14e14 5e762f0
61451Author: Keith Packard <keithp@keithp.com>
61452Date: Tue Oct 13 18:40:42 2009 -0700
61453
61454 Merge remote branch 'mattst88/master'
61455
61456 * mattst88/master:
61457 [alpha] assume we have __NR_pciconfig_iobase
61458 [alpha] don't return from void functions
61459 Fix undefined symbols on alpha
61460 Fix breakage on alpha caused by c7680befe5ae
61461 Revert "alpha: kill xf86SlowBCopyToBus and xf86SlowBCopyFromBus"
61462
61463commit 5e762f0e2f203b5121a3de3b9af3c8981b31a77f
61464Author: Matt Turner <mattst88@gmail.com>
61465Date: Tue Oct 13 20:42:14 2009 -0400
61466
61467 [alpha] assume we have __NR_pciconfig_iobase
61468
61469 The code path if we didn't have support has been broken since before we
61470 switched to git.
61471
61472 The pciconfig_iobase syscall has been supported since 2000.
61473
61474 Signed-off-by: Matt Turner <mattst88@gmail.com>
61475
61476commit 46785c04bca16f495af3ed8d685aee939a1a8f39
61477Author: Matt Turner <mattst88@gmail.com>
61478Date: Tue Oct 13 20:40:59 2009 -0400
61479
61480 [alpha] don't return from void functions
61481
61482 Signed-off-by: Matt Turner <mattst88@gmail.com>
61483
61484commit 2b14e142039193ea854a02706662204f4f5f6db8
61485Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
61486Date: Tue Oct 13 12:54:56 2009 -0700
61487
61488 Rootless: src drawable window can now be NULL
61489
61490 Fix a possible crash when pSrc->pDrawable is NULL.
61491
61492 Signed-off-by: Colin Harrison <colin.harrison@virgin.net>
61493 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
61494 Signed-off-by: Keith Packard <keithp@keithp.com>
61495
61496commit 5b91dfac6fbdf35288a4558638d4923e230ab8d3
61497Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
61498Date: Tue Oct 13 12:54:11 2009 -0700
61499
61500 Rootless: Abstract some of the Xplugin specific stuff which has crept into rootlessWindow.c
61501
61502 The rootless extension now directly calls some Xplugin functions, and relies
61503 on types defined in Xplugin.h, which isn't very abstracted :-)
61504
61505 This patch is a start at abstracting some of the Xplugin specific stuff which
61506 has crept into rootlessWindow.c. This has been done in a pretty mindless fashion,
61507 without much thought as to if the additions to the generic rootless interface are
61508 the correct ones
61509
61510 There is some confusion as to if RootlesscolormapCallback() returns a Bool or
61511 xp_error_enum value (not so abstact), but I have no way of checking, of finding
61512 out if Xplugin actually checks the result :-)
61513
61514 Based on patches from Colin Harrison, Jon Turney and Yaakov Selkowitz
61515
61516 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
61517 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
61518 Signed-off-by: Keith Packard <keithp@keithp.com>
61519
61520commit 72f5874434c0c015b671c492c1318f35f1793668
61521Author: Peter Hutterer <peter.hutterer@who-t.net>
61522Date: Tue Oct 13 13:15:05 2009 +1000
61523
61524 dix: extend IsPointerDevice check to valuator-only devices.
61525
61526 A device with valuators but no keys is definitely a pointer device and needs
61527 to be attached to the VCP. Otherwise, the class copying happens on the VCK
61528 and the VCP isn't updated with the events that are to be sent through it.
61529 This addresses the trigger for #24441, not the actual issue.
61530 Jury is still out on valuator+key devices.
61531
61532 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61533 Signed-off-by: Keith Packard <keithp@keithp.com>
61534
61535commit 1088073b11ed488c0df45af3867b900ef93c6fe1
61536Author: Michel Dänzer <daenzer@vmware.com>
61537Date: Fri Oct 9 11:31:44 2009 +0200
61538
61539 EXA: Fix exaTryDriverSolidFill() for solid source pictures.
61540
61541 Solid pictures have a NULL pFormat field, but their format is always
61542 PICT_a8r8g8b8.
61543
61544 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
61545 Signed-off-by: Keith Packard <keithp@keithp.com>
61546
61547commit 55305cf8db7787883bc80b7348eb626e609626f8
61548Author: Ben Skeggs <bskeggs@redhat.com>
61549Date: Fri Oct 9 16:08:15 2009 -0700
61550
61551 EXA: fix exaGetRGBAFromPixel to not loop forever on PICT_a8 picture
61552
61553 Easily reproducible by running "rendercheck -t fill".
61554
61555 It should be safe to just test against rbits for all colour components
61556 as we should always have values for r/g/bbits for PICT_FORMAT_COLOR
61557 formats.
61558
61559 Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
61560 Signed-off-by: Michel Dänzer <daenzer@vmware
61561 Signed-off-by: Keith Packard <keithp@keithp.com>
61562
61563commit b375be9285c40467578fac2f1360c29a75306ab4
61564Author: Michel Dänzer <daenzer@vmware.com>
61565Date: Fri Oct 9 11:31:46 2009 +0200
61566
61567 composite: Revert changes from adding support for BGRA picture formats.
61568
61569 They were aimed towards a since abandoned approach for making radeon KMS work
61570 on big endian machines, and Aaron Plattner pointed out that they break the
61571 Composite extension when the X server runs in 16bpp.
61572
61573 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
61574 Tested-by: Aaron Plattner <aplattner@nvidia.com>
61575 Signed-off-by: Keith Packard <keithp@keithp.com>
61576
61577commit c0a1bb511a4629bf5683d8a710dc4a1c577a5d44
61578Author: Alan Coopersmith <alan.coopersmith@sun.com>
61579Date: Wed Oct 7 22:10:55 2009 -0700
61580
61581 Fix make warning: overriding commands for target `dix.O'
61582
61583 Not only does automake generate unnecessary rules for dix.O on platforms
61584 for which SPECIAL_DTRACE_OBJECTS is false, it generates duplicate sets
61585 when "if SPECIAL_DTRACE_OBJECTS" is nested inside "if XSERVER_DTRACE"
61586
61587 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61588 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
61589 Signed-off-by: Keith Packard <keithp@keithp.com>
61590
61591commit b0dd6be2c8703f7062d45ac9fd646550c7d54e3b
61592Author: Jamey Sharp <jamey@minilop.net>
61593Date: Thu Oct 8 13:38:44 2009 +1100
61594
61595 Cast small-int values through intptr_t when passed as pointers
61596
61597 On 64-bit systems, int and pointers don't have the same size, so GCC gives
61598 warnings about casts between int and pointer types. However, in the cases
61599 covered by this patch, it's always a value that fits in int being stored
61600 temporarily as a pointer and then converted back later, which is safe.
61601 Casting through the pointer-sized integer type intptr_t convinces the
61602 compiler that this is OK.
61603
61604 Signed-off-by: Jamey Sharp <jamey@minilop.net>
61605 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
61606
61607commit b680a89262efcfef4644adb4a61ae42ea0db0c93
61608Author: Jamey Sharp <jamey@minilop.net>
61609Date: Thu Oct 8 13:38:01 2009 +1100
61610
61611 Fix GCC warnings in xorg_backtrace
61612
61613 Signed-off-by: Jamey Sharp <jamey@minilop.net>
61614 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
61615
61616commit 9bf2ff4faf730913de3073f346646a8727be41d4
61617Author: Jamey Sharp <jamey@minilop.net>
61618Date: Thu Oct 8 13:36:44 2009 +1100
61619
61620 Fix "possibly uninitialized" warnings in glx
61621
61622 In both functions, "answer" was uninitialized if "compsize" was 0, but in
61623 that case __GLX_SEND_VOID_ARRAY(compsize) results in a call to
61624 WriteToClient for 0 bytes, which returns immediately without examining the
61625 "answer" argument. So initializing to a null pointer is as good as
61626 anything else.
61627
61628 Signed-off-by: Jamey Sharp <jamey@minilop.net>
61629 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
61630
61631commit b422b532f3dcab54c53f61a66f2ad76059d1874a
61632Author: Jamey Sharp <jamey@minilop.net>
61633Date: Thu Oct 8 13:29:27 2009 +1100
61634
61635 Remove static MAXSCREENS limit from Xext/shm.c
61636
61637 Dynamically allocate per-screen data in the SHM extension, instead of
61638 having a single static-sized array.
61639
61640 Signed-off-by: Jamey Sharp <jamey@minilop.net>
61641 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
61642
61643commit 4df3e8c8053d9d4b3043f339a73d0de97020d884
61644Author: Martin Ettl <ettl.martin@gmx.de>
61645Date: Thu Oct 8 13:27:30 2009 +1100
61646
61647 Resource leakage: 0 is a valid file descriptor
61648
61649 When testing if an fd is valid, the required construct is >= 0, not > 0.
61650
61651 [Daniel: Fixed up the Linux MTRR case as well.]
61652
61653 Signed-off-by: Martin Ettl <ettl.martin@gmx.de>
61654 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
61655
61656commit 4151a13c80f3afa43f88afcf19a7aeb16dace93a
61657Author: Francisco Jerez <currojerez@riseup.net>
61658Date: Mon Oct 5 02:39:03 2009 +0200
61659
61660 dix: Fix a double free in dixFreePrivates.
61661
61662 It can be reproduced when the server is regenerated and for some
61663 reason the private keys are reassigned in a different order: a
61664 manually allocated private may get an index formerly used by a
61665 preallocated private. In that case it will first be manually freed and
61666 then again by dixFreePrivates, as items[i].size was never zeroed
61667 out. Do it in dixResetPrivates.
61668
61669 Signed-off-by: Francisco Jerez <currojerez@riseup.net>
61670 Acked-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
61671 Signed-off-by: Keith Packard <keithp@keithp.com>
61672
61673commit 34eddbbb73bb16395dba0818247909c1b4bee4c2
61674Author: Michael Cree <mcree@orcon.net.nz>
61675Date: Wed Oct 7 18:33:29 2009 -0400
61676
61677 Fix undefined symbols on alpha
61678
61679 Signed-off-by: Matt Turner <mattst88@gmail.com>
61680
61681commit 44efcdde501d54ca9c3e33ab5b1f699956fc3f9f
61682Author: Jamey Sharp <jamey@minilop.net>
61683Date: Wed Oct 7 09:07:03 2009 -0700
61684
61685 Fix overlay detection when matching Xv adaptors across screens.
61686
61687 Signed-off-by: Jamey Sharp <jamey@minilop.net>
61688 Acked-by: Adam Jackson <ajax@redhat.com>
61689 Signed-off-by: Keith Packard <keithp@keithp.com>
61690
61691commit 68304215e25876ee639015969b4f07e1c9c515e0
61692Merge: 6676f49 315aaef
61693Author: Keith Packard <keithp@keithp.com>
61694Date: Tue Oct 6 22:41:42 2009 -0700
61695
61696 Merge remote branch 'alanc/master'
61697
61698 Conflicts:
61699 configure.ac
61700
61701 Signed-off-by: Keith Packard <keithp@keithp.com>
61702
61703commit 315aaef55750a863c08a16ad9120ffb76f9b48e0
61704Author: Alan Coopersmith <alan.coopersmith@sun.com>
61705Date: Fri Oct 2 21:54:53 2009 -0700
61706
61707 Use $(AM_V_GEN) to silence more commands when AM_SILENT_RULES is active
61708
61709 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61710 Reviewed-by: Eric Anholt <eric@anholt.net>
61711 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
61712 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
61713
61714commit 2e294380b64b6362173d0590c4292f290e50af50
61715Author: Alan Coopersmith <alan.coopersmith@sun.com>
61716Date: Fri Oct 2 16:51:24 2009 -0700
61717
61718 Migrate to xorg macros 1.3 & XORG_DEFAULT_OPTIONS
61719
61720 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61721 Reviewed-by: Eric Anholt <eric@anholt.net>
61722 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
61723 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
61724
61725commit 4b0911565d18b1aefca564315d0f1b9b4a17aa65
61726Author: Alan Coopersmith <alan.coopersmith@sun.com>
61727Date: Mon Oct 5 16:52:19 2009 -0700
61728
61729 Add platform tests for Dtrace linker magic
61730
61731 Replaces special handling for Xquartz DDX and scales better to handling
61732 the multiple platforms that now have some level of Dtrace support available.
61733
61734 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61735 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
61736
61737commit 8c51b886400c5962b31ff565771be1b01a3ca8fb
61738Author: Julien Cristau <jcristau@debian.org>
61739Date: Tue Oct 6 16:08:42 2009 +0200
61740
61741 configure: quote argument to m4_pattern_forbid
61742
61743 Without this, configure spits out
61744 ../configure: line 15460: ac_fn_c_check_member: command not found
61745
61746 Also anchor the pattern to make it stricter.
61747
61748 Signed-off-by: Julien Cristau <jcristau@debian.org>
61749 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
61750 Signed-off-by: Keith Packard <keithp@keithp.com>
61751
61752commit 6ee61578ddb9f880ef12bbe9c3671d7dd53f2809
61753Author: Alan Coopersmith <alan.coopersmith@sun.com>
61754Date: Mon Oct 5 17:27:27 2009 -0700
61755
61756 Remove shave now that automake-1.11 has AM_SILENT_RULES
61757
61758 Revert "Add shave so that we can see the steaming piles of warnings generated."
61759 This reverts commit 181cc08c8908a119fc403f970dea8cc98d3e0b9b.
61760
61761 shave was left in the xserver 1.7 release/branch to allow developers &
61762 distros time to transition to automake-1.11. xserver 1.8 series will
61763 require automake-1.11 for silencing build noise.
61764
61765 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61766 Signed-off-by: Keith Packard <keithp@keithp.com>
61767
61768commit d5d960f57834f16164457dab2e80290717dbf2cd
61769Author: Kim Woelders <kim@woelders.dk>
61770Date: Fri Oct 2 19:31:15 2009 +0200
61771
61772 render: Fix clip region translation in miClipPictureSrc().
61773
61774 Signed-off-by: Kim Woelders <kim@woelders.dk>
61775 Reviewed-by: Soren Sandmann Pedersen <sandmann@redhat.com>
61776 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61777
61778commit d2118c8ca94fc7bcb8e0de547e63cf022f53a8a9
61779Author: Peter Hutterer <peter.hutterer@who-t.net>
61780Date: Mon Oct 5 15:31:54 2009 +1000
61781
61782 xfree86: remove log-spamming DebugF
61783
61784 All input drivers use xf86PostKeyEventP indirectly now and have been since
61785 it exists. I guess that qualifies it as tested - no need to spam the logs.
61786
61787 Reported-by: Felix Wenk
61788 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61789 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
61790
61791commit bf116057c7efd43c247c93f2f5d733a5db857cc9
61792Author: Peter Hutterer <peter.hutterer@who-t.net>
61793Date: Mon Oct 5 10:44:43 2009 +1000
61794
61795 Require libXtst >= 1.0.99.2 and libdmx >= 1.0.99.1
61796
61797 libXtst requirement is already implicit since we require xextproto 7.1 and
61798 that doesn't go well with pre 1.0.99.2 versions of libXtst. Nonetheless,
61799 list it explicitly.
61800
61801 Since d0440275108920f5cb5d630f55fc9a3320c496d3 we require dmxproto 2.2.99.1.
61802 Complementing that is libdmx 1.0.99.1 with the dmxext.h header file.
61803
61804 Reported-by: Mark Rosenstand
61805 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61806
61807commit 9625f6d328d6f516520930227b218979309938bc
61808Author: Matt Turner <mattst88@gmail.com>
61809Date: Tue Oct 6 20:58:30 2009 -0400
61810
61811 Fix breakage on alpha caused by c7680befe5ae
61812
61813 Pinpointed by by Michael Cree.
61814
61815 Commit c7680befe5ae removed Jensen support, but at the same time broke
61816 support for dense memory systems.
61817
61818 Signed-off-by: Matt Turner <mattst88@gmail.com>
61819
61820commit aa07957373fd7cbe67458a001e4afd6a7f1ea37f
61821Author: Peter Hutterer <peter.hutterer@who-t.net>
61822Date: Mon Sep 28 12:20:03 2009 +1000
61823
61824 Revert "alpha: kill xf86SlowBCopyToBus and xf86SlowBCopyFromBus"
61825
61826 The vesa driver still uses slowbcopy_frombus and slowbcopy_tobus.
61827
61828 This reverts commit 5ef53a94ce4e48e11de26290cd677266308640c8.
61829
61830 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61831
61832commit 6676f49e34f8db7ef5767a6dbf6c2a4d2087f79c
61833Author: Julien Cristau <jcristau@debian.org>
61834Date: Tue Oct 6 16:08:42 2009 +0200
61835
61836 configure: quote argument to m4_pattern_forbid
61837
61838 Without this, configure spits out
61839 ../configure: line 15460: ac_fn_c_check_member: command not found
61840
61841 Also anchor the pattern to make it stricter.
61842
61843 Signed-off-by: Julien Cristau <jcristau@debian.org>
61844 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
61845 Signed-off-by: Keith Packard <keithp@keithp.com>
61846
61847commit a0363bf7b135d816e73849c91fa53d98fd6072d1
61848Author: Alan Coopersmith <alan.coopersmith@sun.com>
61849Date: Mon Oct 5 17:27:27 2009 -0700
61850
61851 Remove shave now that automake-1.11 has AM_SILENT_RULES
61852
61853 Revert "Add shave so that we can see the steaming piles of warnings generated."
61854 This reverts commit 181cc08c8908a119fc403f970dea8cc98d3e0b9b.
61855
61856 shave was left in the xserver 1.7 release/branch to allow developers &
61857 distros time to transition to automake-1.11. xserver 1.8 series will
61858 require automake-1.11 for silencing build noise.
61859
61860 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61861 Signed-off-by: Keith Packard <keithp@keithp.com>
61862
61863commit cbc886a3513079c084fb7ce47b87c3e5bba19f3f
61864Author: Kim Woelders <kim@woelders.dk>
61865Date: Fri Oct 2 19:31:15 2009 +0200
61866
61867 render: Fix clip region translation in miClipPictureSrc().
61868
61869 Signed-off-by: Kim Woelders <kim@woelders.dk>
61870 Reviewed-by: Soren Sandmann Pedersen <sandmann@redhat.com>
61871 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61872
61873commit bd7430a32e63df8cd60352764744076448ee623f
61874Author: Peter Hutterer <peter.hutterer@who-t.net>
61875Date: Mon Oct 5 15:31:54 2009 +1000
61876
61877 xfree86: remove log-spamming DebugF
61878
61879 All input drivers use xf86PostKeyEventP indirectly now and have been since
61880 it exists. I guess that qualifies it as tested - no need to spam the logs.
61881
61882 Reported-by: Felix Wenk
61883 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61884 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
61885
61886commit 1a2dde3335864abfbbc133dbc709f564272dd540
61887Author: Alan Coopersmith <alan.coopersmith@sun.com>
61888Date: Fri Oct 2 20:44:19 2009 -0700
61889
61890 Set XQUARTZ to no on non-Darwin OS'es
61891
61892 Fixes build with dtrace probes on Solaris after efacd7bfd08ffc0725de6f639c6afbf3b2f6c9fe
61893
61894 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
61895 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
61896
61897commit c629e0fc50dfaffaa40a56709da7f035c289fed2
61898Author: Peter Hutterer <peter.hutterer@who-t.net>
61899Date: Mon Oct 5 10:44:43 2009 +1000
61900
61901 Require libXtst >= 1.0.99.2 and libdmx >= 1.0.99.1
61902
61903 libXtst requirement is already implicit since we require xextproto 7.1 and
61904 that doesn't go well with pre 1.0.99.2 versions of libXtst. Nonetheless,
61905 list it explicitly.
61906
61907 Since d0440275108920f5cb5d630f55fc9a3320c496d3 we require dmxproto 2.2.99.1.
61908 Complementing that is libdmx 1.0.99.1 with the dmxext.h header file.
61909
61910 Reported-by: Mark Rosenstand
61911 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61912
61913commit 24e640e1fa6cfb3917ba90dcdabf1fadddff20aa
61914Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
61915Date: Fri Oct 2 19:20:24 2009 -0700
61916
61917 XQuartz: Fix a possible minor memory leak
61918
61919commit c4886fbabc1d8b4054654b227fcad83f58e8e798
61920Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
61921Date: Fri Oct 2 18:58:35 2009 -0700
61922
61923 XQuartz: Send mouse location with scroll events.
61924
61925 This fixes the problem where (0, 0) was sent as the mouse location with scroll
61926 button events causing the event to not reach the client.
61927
61928commit 4c6bfa2c09ae2b0cffdf9211a6dfbcaefe0366b5
61929Author: Ian Romanick <ian.d.romanick@intel.com>
61930Date: Thu Oct 1 10:39:19 2009 -0700
61931
61932 GLX: More clearly document the GLX protocol version handling
61933
61934 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
61935 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
61936
61937commit 45f447dafded5adfe11b7df3325c2d8f6ae0639b
61938Author: Peter Hutterer <peter.hutterer@who-t.net>
61939Date: Thu Oct 1 21:19:11 2009 +1000
61940
61941 dix: force a minimum of 0 for screen coordinates.
61942
61943 Currently the root coordinates may fall into ]-1..0] if the subpixel
61944 remainder is less than 0. Screen coordinates mustn't go below 0, so use
61945 miPointerSetPosition to cap off the remainder if the coordinates are below
61946 0.
61947
61948 This is cheating a bit, a more comprehensive solution to deal with subpixels
61949 correctly when crossing screens is needed. For now, this'll do.
61950
61951 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61952 Acked-by: Simon Thum <simon.thum@gmx.de>
61953
61954commit 64fe5784b49347e1fd27b0c463be5c16557594c9
61955Author: Peter Hutterer <peter.hutterer@who-t.net>
61956Date: Fri Oct 2 09:34:24 2009 +1000
61957
61958 configure: if xnest was requested but modules weren't found, fail.
61959
61960 Tested-by: Xavier Chantry <shiningxc@gmail.com>
61961 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61962 Acked-by: Julien Cristau <jcristau@debian.org>
61963
61964commit 3ebb82d61c2b56e8f7145443a552a4e913bbfc80
61965Author: Dave Airlie <airlied@redhat.com>
61966Date: Thu Oct 1 15:25:55 2009 +1000
61967
61968 rotate: drop unwrapping inside block handler.
61969
61970 Keith has shown half the block handlers wrappers are wrong, also
61971 dynamic wrapping/unwrapping from what I can see will happen after
61972 the drivers, so its really accidental ABI, that we can't change
61973 now without modifing drivers. So be safe for 1.7.
61974
61975 Signed-off-by: Dave Airlie <airlied@redhat.com>
61976 Declared-as-sane-by: Keith Packard <keithp@keithp.com>
61977 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
61978
61979commit cb54cf1b3e8c4109541cfb698542c00f2473e731
61980Author: Dave Airlie <airlied@linux.ie>
61981Date: Wed Sep 30 11:33:16 2009 +1000
61982
61983 glx: fixup deref of null pointer when glx screen init fails.
61984
61985 I think this is what the original author wanted.
61986
61987 Signed-off-by: Dave Airlie <airlied@redhat.com>
61988 Acked-by: Ian Romanick <ian.d.romanick@intel.com>
61989
61990commit 3d7cf468df96c0130a862f0d93fec990b9110c2f
61991Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
61992Date: Thu Oct 1 11:58:18 2009 -0700
61993
61994 XQuartz: Update version strings to be X11R7.5 and the bundle 2.5.0
61995
61996 Is fink really _still_ relying on this X11R7.x version string?
61997
61998commit 85b831f701d085e514afcf8a5ce9b11f8a57984e
61999Author: Julien Cristau <jcristau@debian.org>
62000Date: Thu Oct 1 17:27:11 2009 +0200
62001
62002 xfree86: fix xorg.conf manpage formatting error
62003
62004commit 9bc7cbf9c02656982c2525836b5498993f708e02
62005Author: Nirbheek Chauhan <nirbheek@gentoo.org>
62006Date: Tue Sep 29 22:35:06 2009 +0530
62007
62008 xkb: check permissions on XKM_OUTPUT_DIR
62009
62010 Checking just for root is insufficient since that does not guarantee write/read
62011 permissions in XKM_OUTPUT_DIR (for example with sandbox).
62012
62013 Check if we can write a file, as well as read it later. Otherwise, invoke the
62014 fallback to /tmp
62015
62016 Signed-off-by: Nirbheek Chauhan <nirbheek@gentoo.org>
62017 Signed-off-by: Rémi Cardona <remi@gentoo.org>
62018 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62019
62020commit 622fc98fd08aba98369e6933c3ab8c9ff85385d5
62021Author: Peter Hutterer <peter.hutterer@who-t.net>
62022Date: Thu Oct 1 10:03:42 2009 +1000
62023
62024 render: Fix crash in RenderAddGlyphs (#23645)
62025
62026 This patch fixes two bugs:
62027 size is calculated as glyph height * padded_width. If the client submits
62028 garbage, this may get above INT_MAX, resulting in a negative size if size is
62029 unsigned. The sanity checks don't trigger for negative sizes and the server
62030 goes and writes into random memory locations.
62031
62032 If the client submits glyphs with a width or height 0, the destination
62033 pixmap is NULL, causing a null-pointer dereference. Since there's nothing to
62034 composite if the width/height is 0, we might as well skip the whole thing
62035 anyway.
62036
62037 Tested with Xvfb, Xephyr and Xorg.
62038
62039 X.Org Bug 23645 <http://bugs.freedesktop.org/show_bug.cgi?id=23645>
62040
62041 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62042 Reviewed-by: Keith Packard <keithp@keithp.com>
62043
62044commit 758ab55d2defc78d0169fd61a7036eb9f889e9e7
62045Author: Peter Hutterer <peter.hutterer@who-t.net>
62046Date: Thu Oct 1 15:22:19 2009 +1000
62047
62048 render: set the glyph picture to NULL by default.
62049
62050 In a follow-up patch we may have glyphs with a NULL picture. To cope with
62051 that, always set the pictures for glyphs to NULL at creation time and cope
62052 with cleaning up such glyphs. Also, since compositing a NULL source doesn't
62053 do a lot anyway, skip trying to do so.
62054
62055 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62056 Reviewed-by: Keith Packard <keithp@keithp.com>
62057
62058commit b5fcc5553eb784c9f4826936e839079c0cdee55a
62059Author: Michel Dänzer <daenzer@vmware.com>
62060Date: Thu Oct 1 15:17:11 2009 +1000
62061
62062 exa: avoid infinite loops if UTS sw fallbacks.
62063
62064 The upload in finish access can cause an infinite loop if
62065 UTS returns FALSE in here.
62066
62067 Fixes fd.o bug #24246.
62068
62069 Signed-off-by: Dave Airlie <airlied@redhat.com>
62070 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62071
62072commit db98b26ee145f70e732e2cf4a6ac3de77fdf4adc
62073Author: Keith Packard <keithp@keithp.com>
62074Date: Wed Sep 30 11:40:19 2009 -0700
62075
62076 Re-fix DGA removal.
62077
62078 Removing DGA ended up breaking any drivers calling into the old
62079 xf86DiDGAInit function as it tried to see if DGA was already enabled
62080 and ended up crashing if the VT wasn't completely initialized. Oops.
62081
62082 Also, if the driver initializes DGA itself, have the DiDGA
62083 initialization overwrite that information as the DiDGA code will call
62084 ReInit on mode detect.
62085
62086 Signed-off-by: Keith Packard <keithp@keithp.com>
62087 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62088
62089commit 873467adad479be02cd9cc6b43685919f5612d91
62090Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62091Date: Wed Sep 30 00:22:02 2009 -0700
62092
62093 XQuartz: Set the proper bitmap for key repeats...
62094
62095 XkbSetRepeatKeys lies and doesn't do what it says it will...
62096
62097commit 7bef78e199a4e4f7916be506807513efcd8fd4cb
62098Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62099Date: Tue Sep 29 23:40:47 2009 -0700
62100
62101 xkb: Use XkbPerKeyBitArraySize instead of hardcoded value of 32
62102
62103commit 4970666827e65424ee17ccf6341ff84aac974383
62104Author: Kim Woelders <kim@woelders.dk>
62105Date: Tue Sep 29 20:31:45 2009 +0200
62106
62107 dix: Fix potential memory corruption in doListFontsWithInfo.
62108
62109 Signed-off-by: Kim Woelders <kim@woelders.dk>
62110 Reviewed-by: Keith Packard <keithp@keithp.com>
62111 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62112
62113commit 6ffda5aae75272fabdc27d6f693ae827be119e95
62114Author: Dave Airlie <airlied@redhat.com>
62115Date: Tue Sep 29 11:49:09 2009 +1000
62116
62117 dix/glx/composite: consolidate visual resize in one place.
62118
62119 The previous code was copied and in both cases incorrectly fixed
62120 up the colormaps after resizing the visuals, this patch consolidates
62121 the visual resize + colormaps fixups in one place. This version
62122 also consolidates the vid allocation for the DepthPtr inside the
62123 function.
62124
62125 I'm not 100% sure colormap.[ch] is the correct place for this but
62126 visuals are mostly created in fb and I know thats not the place to
62127 be resizing them.
62128
62129 Fixes fd.o bug #19470.
62130
62131 Signed-off-by: Dave Airlie <airlied@redhat.com>
62132 Reviewed-by: Keith Packard <keithp@keithp.com>
62133 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62134
62135commit ad5c0d9efa47476ed5cf75c82265c73919e468b4
62136Author: Ian Romanick <ian.d.romanick@intel.com>
62137Date: Tue Sep 29 16:43:43 2009 -0700
62138
62139 GLX: Enable GLX 1.4 on DRI2
62140
62141 Return the minimum GLX version supported by all screens. Assume that
62142 DRI2 screens have all the required features for GLX 1.4. Assume that
62143 everyone else can only support GLX 1.2.
62144
62145 Reviewed-by: Kristian Høgsberg <krh@redhat.com>
62146 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
62147
62148commit f11a356bcef1bc0a6440325019d5967b745a42dd
62149Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62150Date: Mon Sep 28 17:05:29 2009 -0700
62151
62152 XQuartz: Cleaned up keymap setting for easier maintenance
62153 (cherry picked from commit b9dfed9e88389cbd29406a20d38ee4297638649b)
62154
62155commit f3223c71cfc638e695981e527517d48ea00d124d
62156Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62157Date: Mon Sep 28 23:01:59 2009 -0700
62158
62159 XQuartz: Remove the redundant xquartz_resetenv_display
62160 unsetenv(DISPLAY) takes care of this for us anyway
62161 (cherry picked from commit d2263645d839c9edeedea0835d26f1f41b37f70e)
62162
62163commit 9b98b883227ed23d5470e8de689afeec4a0fd742
62164Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62165Date: Mon Sep 28 17:47:31 2009 -0700
62166
62167 XQuartz: Query the BundleIdentifier from the bundle in X11.bin rather than using the configure option.
62168
62169 This lets X11.bin drop into any .app ... the Info.plist and Xquartz binary need to have it hardcoded still.
62170 (cherry picked from commit 9ad16b8e50b13eb6d0cd20386d07aa8d7320f671)
62171
62172commit 11817a881cb93a89788105d1e575a468f2a8d27c
62173Author: Michel Dänzer <daenzer@vmware.com>
62174Date: Tue Sep 29 08:56:59 2009 +0200
62175
62176 Fix ShmPutImage non-ZPixmap case.
62177
62178 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=23298 .
62179
62180commit 19be992d9dc542b61fa3f4fd32a09071c9e64880
62181Author: Peter Hutterer <peter.hutterer@who-t.net>
62182Date: Mon Sep 28 14:18:45 2009 +1000
62183
62184 ephyr: if -parent is given, check for a trailing -screen. (#24144)
62185
62186 If -parent is given, don't open up a new window if -screen is given as well.
62187 The commandline option -screen allows to set the depth of the embedded
62188 Xephry instance, even though width and height are autoscaled on -parent.
62189
62190 This patch checks for a -screen parameter after -parent and - if one is
62191 found - delays initializing the screen. The parent window id is stored
62192 temporarily but re-set after a -screen argument.
62193 The following command is thus valid:
62194
62195 Xephyr -parent 1234 -screen 640x480@8 -screen 1024x768
62196
62197 It embeds the first 8-bit screen into window 1234 and opens up a new window
62198 for the second screen. Multiple parent arguments are possible, the screens
62199 are embedded in-order.
62200
62201 X.Org Bug 24144 <http://bugs.freedesktop.org/show_bug.cgi?id=24144>
62202
62203 Tested-by: Vic Lee
62204 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62205
62206commit efacd7bfd08ffc0725de6f639c6afbf3b2f6c9fe
62207Author: Ben Byer <bbyer@freedesktop.org>
62208Date: Sun Sep 27 10:35:53 2009 -0700
62209
62210 Add (ok, fix) support for DTrace under OS X
62211 (cherry picked from commit 8428a57184f542941d2c8c90e97d18e111a69dd2)
62212
62213commit b3415187e92960cbff784108b5a3a8d130dc34c5
62214Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62215Date: Sun Sep 27 23:09:51 2009 -0700
62216
62217 Rootless: Correct border rendering on parent-relative windows
62218
62219 Resurected code from the punted RootlessPaintBackground/Border and added it conditionally to miPaintWindow
62220 (cherry picked from commit cf2e3312cff3f341e9edba8c321a4ca7ffd8748e)
62221
62222commit 6df00917cab5c1096070625385fd76ee6c52e0f1
62223Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62224Date: Sun Sep 27 23:00:59 2009 -0700
62225
62226 XQuartz: Fix QuartzSetCursor to match the expected prototype.
62227 (cherry picked from commit dadab5a2279a19dcf709402d7f22f0cd48670db0)
62228
62229commit 67a51cd9ef60b4c9f25fdde84f9eb352936c8c67
62230Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62231Date: Sun Sep 27 22:57:25 2009 -0700
62232
62233 XQuartz: Fix a bunch of compilation warnings about style
62234 (cherry picked from commit 54000bdcbca52a2de31f7c1a1147de6d8e9dbbb8)
62235
62236commit 96780eaf32636c94a9cf33f22eb9f01d984ff754
62237Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62238Date: Sun Sep 27 22:29:49 2009 -0700
62239
62240 XQuartz: Nuke TSM
62241
62242 It's deprecated in SnowLeopard. Ben and I both have no idea what it is for. It says something about unicode input, but urxvt seems fine taking in unicode, so /shrug... bye.
62243 (cherry picked from commit 29cb904e4de2411a9b6dbe68694954788f0525f7)
62244
62245commit 15e15816a2f011d0aeeaff9e394d30a147c973ce
62246Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62247Date: Sun Sep 27 22:26:22 2009 -0700
62248
62249 XQuartz: Fix inverse map from mode_switch to alt
62250 (cherry picked from commit de6cee11e1c335a0e5f708e7641e81d3cfe52529)
62251
62252commit 558d803b2966c2e44345a80e635e091dc3e49f02
62253Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62254Date: Sun Sep 27 18:25:57 2009 -0700
62255
62256 XQuartz: Force a keymap resync on the first keypress to workaround XKB mucking with our keymap.
62257
62258 We need to find a better way to work with XKB on this.
62259 (cherry picked from commit ceaa5c779ceed3de5ea53727649613be3133b24e)
62260
62261commit 226b1033b48807fd6871ba626e20ef1411904939
62262Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62263Date: Sun Sep 27 17:08:12 2009 -0700
62264
62265 XQuartz: Transition from xEvent based mieq to InternalEvent
62266 (cherry picked from commit a3dbde2de87ee4f577748a8c447501a3ea462559)
62267
62268commit 0c2731596f27f2cdf5000ba41de37e7eb86ad6f9
62269Author: Peter Hutterer <peter.hutterer@who-t.net>
62270Date: Mon Sep 28 15:07:48 2009 +1000
62271
62272 Put tests for zero-sized strings in quotes (#24060)
62273
62274 X.Org Bug 24060 <http://bugs.freedesktop.org/show_bug.cgi?id=24060>
62275
62276 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62277
62278commit 83023ffd09a84ff48e6b99f57ebad101a00478db
62279Author: Peter Hutterer <peter.hutterer@who-t.net>
62280Date: Mon Sep 28 13:59:25 2009 +1000
62281
62282 xfree86: use the DDC size if either width or height of DisplaySize is bogus.
62283
62284 If either width or height of DisplaySize is invalid, assume that the
62285 configuration is invalid and use the DDC-reported values instead.
62286
62287 See Comment 9, Bug 9758.
62288 http://bugs.freedesktop.org/show_bug.cgi?id=9758#c9
62289
62290 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62291 Acked-by: Dave Airlie <airlied@redhat.com>
62292
62293commit f772014c435f56db56520ca13ffa39431684f122
62294Author: Peter Hutterer <peter.hutterer@who-t.net>
62295Date: Wed Sep 23 10:53:51 2009 +1000
62296
62297 render: Plug a memory leak in AddGlyph. (#23286)
62298
62299 AddGlyph was missing the FreePicture() call that DeleteGlyph used, resulting
62300 in a memory leak when more than one Glyph was added in a RenderAddGlyphs
62301 request.
62302
62303 Since the code in AddGlyph and DeleteGlyph is identical, move into a static
62304 function to avoid such mistakes in the future.
62305
62306 X.Org Bug 23286 <http://bugs.freedesktop.org/show_bug.cgi?id=23286>
62307
62308commit 3db28f92b0c810b452506abbed299a204c90ba0b
62309Author: Tilman Sauerbeck <tilman@code-monkey.de>
62310Date: Thu Sep 24 12:57:02 2009 +0200
62311
62312 configure: make XNEST default to auto.
62313
62314 When we're checking whether to build Xnest, we're comparing the
62315 variable to auto but before it never was assigned that.
62316
62317 Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
62318
62319 [Xnest was enabled to yes to increase build exposure and catch compiler
62320 errors early. The requirements to Xnest are quite low and I expect most
62321 developers have them, so Xnext will be enabled on most boxes. Anyone
62322 missing those requires probably doesn't want to build Xnest anyway.]
62323 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62324
62325commit 78ad6ca9a97440b74019c00a28144ea7d1e03431
62326Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
62327Date: Sun Sep 27 17:30:23 2009 +0200
62328
62329 xfree86: Hurd fix
62330
62331 I hadn't paid attention that the parameters order had changed, here is a
62332 trivial patch, please apply.
62333
62334 Signed-off-by: Julien Cristau <jcristau@debian.org>
62335
62336commit fb7938315bed9d4cb5641ee9acebf78a13bc109e
62337Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62338Date: Sun Sep 27 09:48:43 2009 -0700
62339
62340 Xi: Make CopyKeyClass X_HIDDEN to avoid ugly ifdef-fu
62341 (cherry picked from commit 6d436e17a9ae7f4ce8537f3fabc052d4f07ca75f)
62342
62343commit a2cd21177859eb45320a94c96dde7d9058ab7650
62344Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62345Date: Sat Sep 26 23:41:45 2009 -0700
62346
62347 XQuartz: Stop checking version numbers of the bundle because CFBundleGetVersionNumber is gimpish
62348 (cherry picked from commit 9f5bdd89608ec12012592ff395b82e954fbb4da8)
62349
62350commit fbfbf93405222322571e15e78dea57c8f2755eae
62351Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62352Date: Sat Sep 26 22:52:39 2009 -0700
62353
62354 XQuartz: Add pressure/tilt property labels
62355 (cherry picked from commit 84ea67130ef6b4086042aad6036ce66f93ea3e56)
62356
62357commit 65f14840898dd7e9d61d8c966f3ff11ae3e82d4f
62358Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62359Date: Sat Sep 26 22:48:49 2009 -0700
62360
62361 XQuartz: Fix a brain-o array indexing problem
62362
62363 /bop Peter
62364 (cherry picked from commit 494a6b046a258ad83dc98eb92b7c3d8f1d2626bb)
62365
62366commit 7159381881e5da33a49e8735811b93cbfe4f681e
62367Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62368Date: Sat Sep 26 22:17:09 2009 -0700
62369
62370 XQuartz: Nuke duplicate locks that make painful headaches
62371 (cherry picked from commit 1dd56322bd1722f2427fb2d833c5608248b60cf0)
62372
62373commit fba8c702ba6aa8fac27682e0838fd744e197d54f
62374Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62375Date: Sat Sep 26 22:13:49 2009 -0700
62376
62377 Xi: CopyKeyClass is not static for XQuartz
62378 (cherry picked from commit 85d6402354cdf143c6490f2725744c2f08b5605b)
62379
62380commit dc1e1bebff8b9d1eb6196dfd4b2f52f1e4200efa
62381Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62382Date: Fri Aug 7 00:02:21 2009 -0700
62383
62384 XQuartz: Use internal xshm header for new xextproto
62385 (cherry picked from commit 1755239330c0bdac820d88a3e06ff391d65f29be)
62386 (cherry picked from commit 697be460d0e555e2c75eed6889293650e02d423c)
62387
62388commit 73ae547d5e687ef10dea45801fc627e10ac4b659
62389Author: Michel Dänzer <daenzer@vmware.com>
62390Date: Sun Sep 27 02:08:10 2009 +0200
62391
62392 EXA: Fix mixed pixmaps crash with missing / failing UploadToScreen hook.
62393
62394 For the recent mixed pixmaps changes, I failed to realize (or hit in my
62395 testing) a problem which can occur if the driver doesn't provide an
62396 UploadToScreen hook or provides one which can fail: There can be a crash
62397 in exaMemcpyBox() because exaCopyDirtyToFb() passes pExaPixmap->fb_ptr to
62398 exaCopyDirty(), but that's normally NULL with driver allocated pixmaps.
62399
62400 The solution is to make exaCopyDirty*() no longer rely on pExaPixmap->fb_ptr
62401 but use pPixmap->devPrivate.ptr after PrepareAccess instead.
62402
62403 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24167 .
62404
62405commit c9ec2bab2f258798fd6e6676698c732f09571a60
62406Author: Kevin E Martin <kem@redhat.com>
62407Date: Sat Sep 26 13:10:54 2009 +1000
62408
62409 dmx: undefine MITSHM, move undefs to miinitext.c.
62410
62411 This patch undefines MITSHM for dmx - we don't support the required
62412 screen->ModifyPixmapHeaders. All undefines are moved from dmx-config to
62413 miinitext.c, where they belong.
62414
62415 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62416
62417commit fc9d733bab3ff0e4e51b19c73b66196dca563a70
62418Author: Kevin E Martin <kem@redhat.com>
62419Date: Sat Sep 26 13:09:52 2009 +1000
62420
62421 dmx: reshuffle linker order to avoid errors when MITSHM is undefined.
62422
62423 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62424
62425commit e7c2598f565e8252dd66ee3e6212b310856476cb
62426Author: Peter Hutterer <peter.hutterer@who-t.net>
62427Date: Fri Sep 25 14:53:33 2009 +1000
62428
62429 dmx: core events are always in screen coordinates when passed to GPE.
62430
62431 This fixes input in dmx, the pointer appears at the right positions to the
62432 clients now.
62433
62434 Also mark the spot where we pass in the button state as valuator to GPE
62435 with a FIXME. (??)
62436
62437 Tested-by: Kevin Martin
62438 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62439
62440commit 43a2eb794f19a2ba56d653f465fc5f6b2ff0d3d3
62441Author: Peter Hutterer <peter.hutterer@who-t.net>
62442Date: Wed Sep 23 21:49:11 2009 +1000
62443
62444 configure: Unify all library defines that require a specific version.
62445
62446 This patch moves all libraries that require a specific version into a single
62447 location instead or duplicating them across the configure.ac file.
62448 Libraries that do not require specific versions are left where they are.
62449
62450 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62451
62452commit 9bd08c690fc687c4d69bb70536f3079a9184476d
62453Author: Peter Hutterer <peter.hutterer@who-t.net>
62454Date: Thu Sep 24 14:07:26 2009 +1000
62455
62456 Xi: update axisVals with the right subpixel data.
62457
62458 Subpixel data in data_frac is stored as FP32.32, hence we need to get that
62459 down again before adding it to the current value.
62460
62461 Reported-by: Thomas Jaeger
62462 Tested-by: Thomas Jaeger
62463 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62464
62465commit abb8108fb44cc9b08fe4ae7d805a0a22564ca6bb
62466Author: Michel Dänzer <daenzer@vmware.com>
62467Date: Sat Sep 26 02:05:24 2009 +0200
62468
62469 EXA: Only calculate cache position once for each glyph.
62470
62471commit c11678cc189551f2a01eaa7a63969c16950739b4
62472Author: Michel Dänzer <daenzer@vmware.com>
62473Date: Sat Sep 26 02:05:06 2009 +0200
62474
62475 EXA: Accumulate arbitrary number of glyphs without flushing.
62476
62477commit 0369eeeb6bf8a808fa2df503fc8b8df81e6e07b8
62478Author: Michel Dänzer <daenzer@vmware.com>
62479Date: Sat Sep 26 02:04:48 2009 +0200
62480
62481 EXA: Try to minimize UploadToScreen calls for mixed pixmaps.
62482
62483 If there are several software fallbacks affecting the system memory copy of the
62484 same pixmap, only copy the results back to the driver pixmap when it's used for
62485 acceleration again, or in the BlockHandler, whichever happens first.
62486
62487commit 1818cbd70fc1f2e1487b4c678e67e28f1265c0ef
62488Author: Michel Dänzer <daenzer@vmware.com>
62489Date: Sat Sep 26 01:59:39 2009 +0200
62490
62491 EXA: Extend mixed pixmaps scheme to allow driver PrepareAccess hook to fail.
62492
62493 If the PrepareAccess hook fails, use the DownloadFromScreen hook to retrieve
62494 driver pixmap contents to a system RAM copy, perform software rendering on that
62495 and copy the results back using the UploadToScreen hook. Use the classic
62496 migration logic to minimize transfers (which as a bonus allows slightly
62497 cleaning up some of the existing mixed pixmap code).
62498
62499 This enables things that weren't possible before with driver-allocated pixmap
62500 storage: If some (or all) GPU pixmap storage can't be mapped directly by the
62501 CPU, this can be handled between the PrepareAccess and
62502 DownloadFrom/UploadToScreen hooks, e.g.:
62503
62504 * Radeon KMS on big endian machines can fail PrepareAccess if the pixmap
62505 requires byte-swapping and swap bytes in DownloadFrom/UploadToScreen.
62506 * Environments where GPU and CPU don't have a shared address space at all.
62507 Here the driver PrepareAccess hook will always fail and leave all transfers
62508 between GPU / CPU storage to the Download/From/UploadToScreen hooks.
62509
62510 Drivers which can handle all pixmaps in the PrepareAccess hook should notice
62511 little if any difference.
62512
62513commit e23bffc41b007f1bc2b8f5cd4ac54213062c95cc
62514Author: Alan Coopersmith <alan.coopersmith@sun.com>
62515Date: Tue Sep 15 19:17:48 2009 -0700
62516
62517 Fix build of unit tests when dtrace probes are enabled
62518
62519 ar loses the dtrace probe magic when building static libraries, so we
62520 have to link with the .O files in order to resolve the dtrace probe symbols.
62521
62522 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
62523
62524commit 9fa73be9fa543a686ea35c861084f5af37d44caa
62525Author: Peter Hutterer <peter.hutterer@who-t.net>
62526Date: Tue Sep 22 20:34:54 2009 +1000
62527
62528 Require libXext >= 1.0.99.4
62529
62530 Reported-by: Tilman Sauerbeck
62531 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62532
62533commit a9c274df5c37cb4ece6449e934342d8ff8e61705
62534Author: Peter Hutterer <peter.hutterer@who-t.net>
62535Date: Wed Sep 23 11:43:27 2009 +1000
62536
62537 kdrive: plug two memory leaks when freeing the KdKeyboard/Pointer.
62538
62539 xkbRules, xkbModel and xkbLayout are strdup'd in KdNewKeyboard, need to be
62540 freed.
62541
62542 The ephyr driver strdups the name on top of the already allocated
62543 kdrive-assigned name. Memory must be freed beforehand.
62544
62545 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62546
62547commit fd913136732ff14a0484ca28f60ac1fbf49be81d
62548Author: Peter Hutterer <peter.hutterer@who-t.net>
62549Date: Wed Sep 23 11:44:12 2009 +1000
62550
62551 dix: plug memory leak in DeviceEnterLeaveEvents.
62552
62553 'event' must be freed before exiting.
62554
62555 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62556
62557commit 6ee796e9bb4e46782b50a69c7b4fa5b49576f139
62558Author: Peter Hutterer <peter.hutterer@who-t.net>
62559Date: Thu Sep 24 14:05:52 2009 +1000
62560
62561 Xi: fix length calculation for ValuatorState in QueryDeviceState reply.
62562
62563 The length field needs to include the bytes required for the valuators
62564 (INT32) as well.
62565
62566 The reply length has the right value and since the valuator state is always
62567 last, clients didn't notice the wrong offset.
62568
62569 Tested-by: Thomas Jaeger
62570 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62571
62572commit 3b5bbb149d4c932d9624336f5cbe9fe71c87bea3
62573Author: Peter Hutterer <peter.hutterer@who-t.net>
62574Date: Wed Sep 23 12:32:44 2009 +1000
62575
62576 configure: fix up tslib check once again.
62577
62578 This patch addresses two issues:
62579 The check for HAVE_TSLIB = xauto can never be true, the check has been
62580 corrected to TSLIB = xauto.
62581
62582 Pre-pkgconfig versions of tslib fail to be found, this patch restores the
62583 additional AC_CHECK_LIB. However, the pgk-config check must happen before
62584 AC_CHECK_LIB, as AC_CHECK_LIB does not seem to honour the LD_LIBRARY_PATH.
62585 Thus, if tslib is installed outside of the default paths, AC_CHECK_LIB
62586 fails.
62587
62588 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62589 Acked--by: Daniel Stone <daniel@fooishbar.org>
62590
62591commit 5402f18d9c3f7ba19cc05b3a814e3a9e94c8d551
62592Author: Thomas Jaeger <thjaeger@gmail.com>
62593Date: Tue Sep 22 20:16:21 2009 -0400
62594
62595 dix: report XI1 axis values correctly if first_valuator != 0
62596
62597 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
62598 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62599
62600commit 33bf9cb69dccbb6889b2f83e6db61f40dc644e17
62601Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
62602Date: Wed Sep 23 17:41:28 2009 -0700
62603
62604 XQuartz: GLX capabilities: Allow 16bit accumulation buffers
62605
62606 http://xquartz.macosforge.org/trac/ticket/308
62607 (cherry picked from commit e9e63a2118b76b6c31c4081fec08a99e4d796e22)
62608
62609commit ce1fe8ddb4a4dbe6cfd909e5b1b73b459d742bec
62610Author: Michel Dänzer <daenzer@vmware.com>
62611Date: Wed Sep 23 13:10:05 2009 +0200
62612
62613 render: Don't add b8g8r8x8 format for depth 24.
62614
62615 The components are required to be packed in the bottom of the pixel, so this
62616 format can't fit in depth 24.
62617
62618 Also fix up a comment for the addition of BGRA formats.
62619
62620commit 096f21bb7a1217443d8a03529b1a2938518eb24f
62621Author: Michel Dänzer <daenzer@vmware.com>
62622Date: Wed Sep 23 08:24:06 2009 +0200
62623
62624 EXA: Fix some issues pointed out by clang.
62625
62626 Remove dead variables, fix use of uninitialized values, that kind of thing.
62627
62628commit 824a09d856a5f750694e11d2fd2faaa3de705eaa
62629Author: Simon Thum <simon.thum@gmx.de>
62630Date: Mon Sep 21 15:23:27 2009 +0200
62631
62632 dix: move bounds check before access
62633
62634 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62635
62636commit 9edb9e9b4dde6f73dc5241d078425a7a70699ec9
62637Author: Peter Hutterer <peter.hutterer@who-t.net>
62638Date: Tue Sep 22 12:56:17 2009 +1000
62639
62640 Bump to 1.7.99.1
62641
62642 X Server 1.7 has branched off.
62643
62644 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62645
62646commit 66ece3bfb19642bf0f1ef73c3caa8a99c3dfb5a9
62647Author: Kevin E Martin <kem@redhat.com>
62648Date: Tue Sep 22 11:42:24 2009 +1000
62649
62650 dmx: only free the default pixmaps that we actually allocated.
62651
62652 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62653
62654commit 20fb8c2e2fccb28c1b89e6e2a5c685e1287baedb
62655Author: Kevin E Martin <kem@redhat.com>
62656Date: Tue Sep 22 11:41:33 2009 +1000
62657
62658 dmx: disable Get/SetWindowPixmap, we don't support composite/redirected windows
62659
62660 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62661
62662commit fcdc1d78cca3b8bb6b77d53eda7e21d649df6943
62663Author: Keith Packard <keithp@keithp.com>
62664Date: Mon Sep 21 19:16:59 2009 +0200
62665
62666 Fix sporadic segfault on resume due to accidentally freeing cursor.
62667
62668 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62669
62670commit 91e1fe5863a180b2d70a6943e83b98bbd4c9ebce
62671Author: Matthias Hopf <mhopf@suse.de>
62672Date: Mon Sep 21 19:13:31 2009 +0200
62673
62674 Revert fe31f9c + 977953b to fix issue for good.
62675
62676 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62677
62678commit 90aa0e4a49cdd637178a771365e1e8dab8bceb87
62679Author: Peter Hutterer <peter.hutterer@who-t.net>
62680Date: Fri Sep 18 10:03:02 2009 +1000
62681
62682 input: don't use typecasts to access members of InternalEvent.
62683
62684 To avoid confusion, the member names are now postfixed with _event.
62685
62686 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62687
62688commit 693d4fdb1c7b7c789a812790122454e718602449
62689Author: Peter Hutterer <peter.hutterer@who-t.net>
62690Date: Tue Sep 22 11:13:58 2009 +1000
62691
62692 Xext: switch mbuf.c to dixLookupResourceByType
62693
62694 Resolves a linker error caused by LookupIDByType.
62695
62696 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62697
62698commit 493d2f9c2fc56dd3b4c03451c07cd25e3012142e
62699Author: Peter Hutterer <peter.hutterer@who-t.net>
62700Date: Tue Sep 22 11:05:07 2009 +1000
62701
62702 Xext: remove DisplayImageBuffers from mbuf.c
62703
62704 Not referenced by anything.
62705
62706 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62707
62708commit fff40b3353c7025c1d1e3dd14b5c78f7c9f5de7c
62709Author: Peter Hutterer <peter.hutterer@who-t.net>
62710Date: Tue Sep 22 10:53:47 2009 +1000
62711
62712 Xext: fix up multibuffer compiler errors.
62713
62714 Triggered by the xextproto 7.1 change, fixed by moving the matching
62715 declarations from the header file to here.
62716
62717 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62718
62719commit 4c8f834da6c8c84849313453fe223a8165c6afc0
62720Author: Peter Hutterer <peter.hutterer@who-t.net>
62721Date: Mon Sep 21 14:52:31 2009 +1000
62722
62723 configure: fix up check for tslib.
62724
62725 Reduce the tslib-check to the pkg-config check only instead of the previous
62726 library symbol check followd by a pkg-config check.
62727
62728 This patch also reduces the required version of tslib back down to
62729 tslib-0.0. Unfortunately, the 1.0 tarball available through
62730 http://tslib.berlios.de/ still announces itself as 0.0.2.
62731
62732 Reported-by: Werner Landgraf
62733 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62734
62735commit 977953bf14858d17c22208c848854ddbe7e86527
62736Author: Matthias Hopf <mhopf@suse.de>
62737Date: Mon Sep 21 15:44:00 2009 +0200
62738
62739 Less intrusive workaround for sporadic segfault on resume.
62740
62741 Hopefully fixes fdo #24010 (memleak).
62742
62743commit 55747d256d759850141e4a9c4dec965616a31dc8
62744Author: Peter Hutterer <peter.hutterer@who-t.net>
62745Date: Fri Sep 18 16:27:54 2009 +1000
62746
62747 input: define server-supported protocol versions in one single file.
62748
62749 include/protocol-versions.h specifies each extension version as supported by
62750 the server and sent back on the wire to the client.
62751
62752 This fixes up several issues with the server potentially reporting a higher
62753 version of the protocol if recompiled against a newer version of the
62754 protocol.
62755
62756 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62757 Acked-by: Rémi Cardona <remi@gentoo.org>
62758 Acked-by: Julien Cristau <jcristau@debian.org>
62759
62760commit 0b7c6c728c2e2d8433a188315cc591308a89cd85
62761Author: Keith Packard <keithp@keithp.com>
62762Date: Fri Sep 18 21:12:17 2009 -0700
62763
62764 xfree86/modes: Remove all framebuffer support from DGA
62765
62766 This removes all rendering and mapping code from xf86DiDGA, leaving
62767 just mode setting and raw input device access. The mapping code didn't
62768 have the offset within /dev/mem for the frame buffer and the pixmap
62769 support assumed that the framebuffer was never reallocated.
62770
62771 Signed-off-by: Keith Packard <keithp@keithp.com>
62772 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62773
62774commit d3a3c904e5e3b5e8bb3c51ca6a563880ff589bf3
62775Author: Peter Hutterer <peter.hutterer@who-t.net>
62776Date: Mon Sep 21 09:41:46 2009 +1000
62777
62778 configure: resourceproto is optional.
62779
62780 Although a --disable-xres flag is provided, resourceproto was listed in
62781 REQUIRED_MODULES since 2005 (5b218617fa8ba52bf65aef35da39e06c662495e6).
62782 Remove, make conditional on the xres flag only.
62783
62784 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62785
62786commit b9ae1b91f316a98eb3bbefe1fc61f3e0faa861e0
62787Author: Peter Hutterer <peter.hutterer@who-t.net>
62788Date: Mon Sep 21 09:22:53 2009 +1000
62789
62790 dmx: silence 'implicit declaration of function ‘XFixesSetPictureClipRegion’
62791
62792 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62793
62794commit aa71e268f55ac39351814ab6840196c4f58bad39
62795Author: Peter Hutterer <peter.hutterer@who-t.net>
62796Date: Fri Sep 18 11:35:47 2009 +1000
62797
62798 Xext: silence 'warning: no previous prototype' for sync, bigreq and xcmisc.
62799
62800 xcmisc and bigreq don't have their own header so just declare it here to
62801 shut up the compiler.
62802
62803 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62804
62805commit 00dfe69aab786ea650b41dfefc031f0e0fc72b0c
62806Author: Peter Hutterer <peter.hutterer@who-t.net>
62807Date: Thu Sep 17 14:48:51 2009 +1000
62808
62809 randr: reply with the server or client version, whichever is lower.
62810
62811 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62812
62813commit 8b75d0f9d347d7c3aad1e53809e0a2bed03b92fc
62814Author: Peter Hutterer <peter.hutterer@who-t.net>
62815Date: Thu Sep 17 14:17:54 2009 +1000
62816
62817 render: reply with the server or client version, whichever is lower.
62818
62819 Protocol requires that the lower of [server version, client version] is
62820 returned to the client.
62821
62822 The other part of the issue discussed in reply to [1] remains.
62823 [1] http://lists.freedesktop.org/archives/xorg-devel/2009-September/001990.html
62824
62825 Reported-by: Julien Cristau
62826
62827 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62828
62829commit 20ccc66708bcd55372bff03fa0d78a7846e8c82d
62830Author: Peter Hutterer <peter.hutterer@who-t.net>
62831Date: Sun Sep 20 21:12:31 2009 +1000
62832
62833 xfree86: silence warning: ‘ramplen’ may be used uninitialized
62834
62835 'ramplen' is only set and used if stuff->size > 0 but the compiler doesn't
62836 pick that up.
62837
62838 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62839
62840commit 2df10a49867635c341b1d1a9faadf4dbae043a0f
62841Author: Keith Packard <keithp@keithp.com>
62842Date: Thu Sep 17 18:14:38 2009 -0700
62843
62844 RRDestroyOutputProperty: Free randr property valid values
62845
62846 These were leaked when the property was destroyed.
62847
62848 Signed-off-by: Keith Packard <keithp@keithp.com>
62849 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62850
62851commit d3a84906c00b8fe4a4e33fa89bff3ed66ceafdec
62852Author: Keith Packard <keithp@keithp.com>
62853Date: Thu Sep 17 18:14:37 2009 -0700
62854
62855 Free randr crtc and output pointer arrays
62856
62857 All of the crts and outputs were freed, but not the arrays full of
62858 pointers to them.
62859
62860 Signed-off-by: Keith Packard <keithp@keithp.com>
62861 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62862
62863commit 6086a6065666fb9ded5d99919ddfa51c66b76246
62864Author: Keith Packard <keithp@keithp.com>
62865Date: Thu Sep 17 18:14:36 2009 -0700
62866
62867 xf8CrtcSetModeTransform: free adjusted_mode name too
62868
62869 The adjusted mode was freed, but any name allocated for that was leaked.
62870
62871 Signed-off-by: Keith Packard <keithp@keithp.com>
62872 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62873
62874commit 088107827347cee05a6e5317bcebdbc282b9a3bc
62875Author: Keith Packard <keithp@keithp.com>
62876Date: Thu Sep 17 18:14:35 2009 -0700
62877
62878 LoaderSortExtensions: free graph nodes
62879
62880 Signed-off-by: Keith Packard <keithp@keithp.com>
62881 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62882
62883commit d5959d37cd36895898786e4ea5bf72c480b79832
62884Author: Keith Packard <keithp@keithp.com>
62885Date: Thu Sep 17 18:14:34 2009 -0700
62886
62887 Free libpciaccess iterator in xf86PciProbe
62888
62889 Signed-off-by: Keith Packard <keithp@keithp.com>
62890 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62891
62892commit ded35b7bf7a5ad27de152476c657a56a4f232800
62893Author: Keith Packard <keithp@keithp.com>
62894Date: Thu Sep 17 18:14:33 2009 -0700
62895
62896 xf86Xinput.c: get DIX event queue pointer once at InitInput time
62897
62898 The DIX event queue is allocated before InitInput is called, so fetch
62899 the pointer there and not randomly at other times. This avoids failing
62900 to fetch the pointer sometimes during server regen and then smashing
62901 memory through the stale pointer from the previous server generation.
62902
62903 Signed-off-by: Keith Packard <keithp@keithp.com>
62904 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62905
62906commit 54f15a414130f89832b5c2604b0b44e7e4c6709f
62907Author: Keith Packard <keithp@keithp.com>
62908Date: Thu Sep 17 18:14:32 2009 -0700
62909
62910 probe_devices_from_device_sections: don't leak list of devices
62911
62912 xf86MatchDevice returned malloc'd storage containing the list of
62913 devices to look at; make sure that gets freed.
62914
62915 Signed-off-by: Keith Packard <keithp@keithp.com>
62916 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62917
62918commit 8b5086250aa5dae8de8b763408ff480d7beac819
62919Author: Keith Packard <keithp@keithp.com>
62920Date: Thu Sep 17 18:14:31 2009 -0700
62921
62922 Eliminate bogus event resizing.
62923
62924 Now that all event queues hold internal events only, they never need
62925 to be resized. Resizing them led to memory corruption as they would
62926 get sized for an appropriate xEvent, not an internal event.
62927
62928 Signed-off-by: Keith Packard <keithp@keithp.com>
62929 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62930
62931commit c7987660fd665f32406b14db7078a3f5f7b76e9c
62932Author: Peter Hutterer <peter.hutterer@who-t.net>
62933Date: Thu Sep 17 15:33:19 2009 +1000
62934
62935 xfree86: fix up wrong use of OptionRec for AutoAddDevices andAutoEnableDevices
62936
62937 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62938
62939commit 971d4eac1274db78efa546ea50ca5e01a80bbd81
62940Author: Thomas Jaeger <ThJaeger@gmail.com>
62941Date: Sat Sep 19 15:48:30 2009 -0400
62942
62943 Xext: include security protocol header instead of client header
62944
62945 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
62946 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62947
62948commit a62664bc07b35aedb6b11b984c94341deb4486ca
62949Author: Thomas Jaeger <ThJaeger@gmail.com>
62950Date: Sat Sep 19 16:01:52 2009 -0400
62951
62952 Bump fixesproto requirement to 4.1
62953
62954 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
62955 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62956
62957commit 2a890d85e923e1da0331b02e58e95f98b11f5de0
62958Author: Thomas Jaeger <ThJaeger@gmail.com>
62959Date: Sun Sep 20 01:31:59 2009 -0400
62960
62961 fixes: Make sure invisible cursor is inialized with zero src/mask
62962
62963 This fixes a regression introduced in 9040dab76182d1a019ca7fef7b29733d2c199e61.
62964
62965 Signed-off-by: Thomas Jaeger <ThJaeger@gmail.com>
62966 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62967
62968commit 2bb3f0c87119e73157141431d7c00ab104d4b34b
62969Author: Peter Hutterer <peter.hutterer@who-t.net>
62970Date: Fri Sep 18 10:27:14 2009 +1000
62971
62972 xfree86: remove now unused variable
62973
62974 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
62975
62976commit 1888d7d3e3ae81b5cb23e3a0852317362c8e215a
62977Author: Carl Worth <cworth@cworth.org>
62978Date: Fri Sep 18 10:39:19 2009 -0700
62979
62980 configure: Fix broken quoting for initialization of REQUIRED_MODULES
62981
62982 A recent change switched to C-style automatic string concatenation,
62983 (presumably for better readability and maintainability), but this
62984 doesn't actually work for the shell. So we revert to the previously
62985 working initialization.
62986
62987 This commit reverts one hunk of c947605eda082fbca0821efc0cc14fec8f958fd0
62988
62989commit 4982bbc3b49b1b61e63a8461ea803e41318c8187
62990Author: Tiago Vignatti <tiago.vignatti@nokia.com>
62991Date: Fri Sep 18 12:36:11 2009 +0300
62992
62993 xfree86: vgaarb: missing wrapping macro
62994
62995 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
62996
62997commit af1af50de621bc451e6413fa5602ae5db2dc9809
62998Author: Martin-Éric Racine <q-funk@iki.fi>
62999Date: Fri Sep 18 09:33:37 2009 +0300
63000
63001 xserver: added more Geode oddities to xf86AutoConfig.c
63002
63003commit ec0ad408ef8d5a2cf7a2bd65d13de3ef3147308c
63004Author: Peter Hutterer <peter.hutterer@who-t.net>
63005Date: Thu Sep 17 16:01:29 2009 +1000
63006
63007 xfree86: use SendDevicePresenceEvents instead of manual event handling.
63008
63009 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63010
63011commit d0440275108920f5cb5d630f55fc9a3320c496d3
63012Author: Peter Hutterer <peter.hutterer@who-t.net>
63013Date: Wed Sep 16 10:58:52 2009 +1000
63014
63015 Require dmxproto 2.2.99.1
63016
63017 This requirement was introduced in code with commit
63018 e85775aa64d1e9322c82090a26256af5e1fede0a.
63019
63020 Reported-by: Marvin Schmidt
63021 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63022
63023commit c947605eda082fbca0821efc0cc14fec8f958fd0
63024Author: Peter Hutterer <peter.hutterer@who-t.net>
63025Date: Tue Sep 15 15:05:27 2009 +1000
63026
63027 configure.ac: move all protocol requirements to a single location.
63028
63029 Protocol requirements were spread throughout configure.ac, leaving us with a
63030 few locations where one DDX asked for a different version of the package
63031 than another.
63032 For each protocol, define <NAME>PROTO="[<name>proto >= version]" and
63033 assemble REQUIRED_MODULES based on these.
63034
63035 Future changes to protocol versions need to only be done in a single place.
63036
63037 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63038
63039commit 4d6b20c25af5f590f19530b6c007e3648a8037c3
63040Author: Shunichi Fuji <palglowr@gmail.com>
63041Date: Wed Sep 16 12:02:00 2009 +1000
63042
63043 dri: use noPanoramiXExtension directly.
63044
63045 xf86LoaderCheckSymbol() is never useful if using externed variable directly.
63046 noPanoramiXExtension will be just used through dlopen() like other extension modules.
63047
63048 Signed-off-by: Shunichi Fuji <palglowr@gmail.com>
63049 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63050
63051commit 3853314d13b503684c9922715323f8033febe5de
63052Author: Alan Coopersmith <alan.coopersmith@sun.com>
63053Date: Thu Sep 17 08:11:27 2009 -0700
63054
63055 Fix typo in Xephyr man page
63056
63057 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63058
63059commit 2edf967b00963559552d061bb98b15243d978279
63060Author: Dave Airlie <airlied@redhat.com>
63061Date: Thu Sep 17 15:24:01 2009 +1000
63062
63063 xserver: SIGIO option handling was incorrect.
63064
63065 reported by a user on #radeon when DRI1 broke.
63066
63067 Signed-off-by: Dave Airlie <airlied@redhat.com>
63068
63069commit 715953bf5c00b4605dd6ab45c92140c0e81174ee
63070Author: Alan Coopersmith <alan.coopersmith@sun.com>
63071Date: Tue Sep 15 21:13:00 2009 -0700
63072
63073 Check for -wrap support in the linker
63074
63075 Allows unit tests to be built with non-gnu-linkers that
63076 also have -wrap support
63077
63078 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63079 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63080
63081commit e2c64551808988657069006f74c6780973ec0557
63082Author: Alan Coopersmith <alan.coopersmith@sun.com>
63083Date: Tue Sep 15 21:13:48 2009 -0700
63084
63085 Add configuration option for use of SIGIO handlers for input events
63086
63087 Boolean option to enable/disable SIGIO handlers is set by the first
63088 of these found:
63089 - UseSIGIO option is set in xorg.conf ServerFlags
63090 - Default set at build time by ./configure --enable-use-sigio-by-default
63091 - Platform default value: Solaris = no, all others = yes
63092
63093 This matches the current settings on all platforms except Solaris.
63094 This reverts Solaris (for now) to the settings used in Xorg 1.6, before
63095 SIGIO support for Solaris was added, due to some system level bugs that
63096 won't be resolved in time for Xorg 1.7 release, but allows us to enable
63097 when those are resolved (or when we need to test if they're resolved).
63098 See http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6879897
63099
63100 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63101 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63102
63103commit 3bdf36476179257561b9409c3f49a77640fbb9f9
63104Author: Matthias Hopf <mhopf@suse.de>
63105Date: Wed Sep 16 16:03:46 2009 +0200
63106
63107 Work around sporadic segfault on resume with intel/KMS due to cursor->bits == NULL.
63108
63109 Apparently SavedCursor is sometime tried to be set while already being set.
63110
63111commit fe31f9c646bc26163497f29c3b833c090df0b3f9
63112Author: Alan Coopersmith <alan.coopersmith@sun.com>
63113Date: Mon Sep 14 19:51:23 2009 -0700
63114
63115 Change xf86dristr.h includes to use xf86driproto.h instead
63116
63117 Clears warnings about obsolete headers, but raises minimum
63118 required version of xf86driproto to 2.1.0
63119
63120 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63121 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63122
63123commit 159e03204093991ccd4133133c50c0cae88c828f
63124Author: Peter Hutterer <peter.hutterer@who-t.net>
63125Date: Tue Sep 15 11:39:09 2009 +1000
63126
63127 dmx: don't include panoramiXsrv.h if building w/o Xinerama
63128
63129 Reported-by: Marvin Schmidt
63130 Tested-by: Marvin Schmidt
63131 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63132
63133commit f56cbe1ef24415d0142b9a7d0ab0a031069ccb52
63134Author: Rémi Cardona <remi@gentoo.org>
63135Date: Mon Sep 14 17:09:59 2009 +0200
63136
63137 dix: append "built-ins" to the font path in SetDefaultFontPath
63138
63139 49b93df8a3002db7196aa3fc1fd8dca1c12a55d6 made the hard dependency on
63140 a "fixed" font go away but only Xorg could use the built-ins fonts by
63141 default.
63142
63143 With this commit, all DDXs get "built-ins" appended to their FontPath, not
63144 just Xorg.
63145
63146 Tested with Xorg, Xvfb and Xnest.
63147
63148 Signed-off-by: Rémi Cardona <remi@gentoo.org>
63149 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
63150 Tested-by: Jon TURNEY <jon.turney@dronecode.org.uk>
63151 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63152
63153commit 139368f7ae192b592e24d013e8ca5ce4175effe1
63154Author: Alan Coopersmith <alan.coopersmith@sun.com>
63155Date: Tue Sep 15 17:53:50 2009 -0700
63156
63157 Fix num_masks/length overflow test for XiSelectEvents
63158
63159 Have to set windowid to a valid value first, since that check
63160 appears earlier in the code than the masks/length check.
63161
63162 Also have to have data[] set large enough so that reading mask
63163 data for 0xFFFF masks doesn't overflow past the end of the array
63164 into uninitialized data.
63165
63166 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63167 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63168
63169commit 13decf5efe76e735a003227c951e58a41e36a8b5
63170Author: Alan Coopersmith <alan.coopersmith@sun.com>
63171Date: Tue Sep 15 13:54:51 2009 -0700
63172
63173 test_convert_XIRawEvent: Don't write more data than the struct holds
63174
63175 sizeof(in.valuators.mask) * 8 == 40, but the valuators arrays only
63176 holds 36 entries, so the test was smashing the stack when 36 < i < 40
63177 (leading to core dumps on Solaris x86, since the return address was
63178 overwritten with an invalid pointer).
63179
63180 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63181 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63182
63183commit 38a61e9a7cf616e26a79dba815bf67737317d1eb
63184Author: Alan Coopersmith <alan.coopersmith@sun.com>
63185Date: Tue Sep 15 13:54:50 2009 -0700
63186
63187 Remove stray semi-colon in protocol-common.c
63188
63189 Caused Sun compilers to issue warning:
63190 "protocol-common.c", line 141: warning: syntax error: empty declaration
63191
63192 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63193 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63194
63195commit e320736c457ea7b2f19a4c7c2dfc0c461d0dd405
63196Author: Alan Coopersmith <alan.coopersmith@sun.com>
63197Date: Mon Sep 14 19:51:22 2009 -0700
63198
63199 Fix module path in xorg.conf man page
63200
63201 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63202 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63203
63204commit 49b98d092ec3d032e657eeff62c22fce243b1e8b
63205Author: Tiago Vignatti <tiago.vignatti@nokia.com>
63206Date: Thu Sep 10 17:52:33 2009 +0300
63207
63208 xfree86: vgaarb: rework wrapping function
63209
63210 No semantical changes.
63211
63212 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
63213 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63214
63215commit 422cd7a674e6831f40be14ee853bd946cb62eadc
63216Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
63217Date: Mon Sep 14 13:52:53 2009 -0700
63218
63219 XQuartz: pbproxy: Remove debugging XBell()
63220 (cherry picked from commit 1fd7c1fd4740393a32e261af33e64c19be48ca9a)
63221
63222commit 663807ce0e65b51537540bb95fbf3c18988c04c7
63223Author: Peter Hutterer <peter.hutterer@who-t.net>
63224Date: Mon Sep 14 16:44:52 2009 +1000
63225
63226 xserver 1.6.99.901 (xserver 1.7 RC1)
63227
63228 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63229
63230commit f450be999268d118b384c04ea070c87ad48c4c59
63231Author: Soeren Sandmann <sandmann@daimi.au.dk>
63232Date: Fri Sep 11 04:03:06 2009 +0200
63233
63234 Add xoff and yoff to drawable->x and drawable->y in create_bits_picture().
63235
63236 These are 0, except when the drawable is a redirected window.
63237
63238 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63239
63240commit 2b00afec4b5b51851aee4cb0fba7baff3e10ef3f
63241Author: Alan Coopersmith <alan.coopersmith@sun.com>
63242Date: Mon Sep 14 15:59:41 2009 +1000
63243
63244 Update xorg.conf man page & sample for changes in Xorg 1.7
63245
63246 AllowMouseOpenFail description changed to reflect actual behaviour
63247 and point to AllowEmptyInput for previously described behaviour.
63248
63249 Update default DPMS mode timeouts to match new defaults set
63250 in April 2009 by commit d52fddefaeb43f15b677eefbea4a288a9948373b
63251
63252 Update autoloaded module list to match ModuleDefaults in xf86Config.c
63253 Update module subdir list to match stdSubdirs in loadmod.c
63254
63255 Add xorg.conf options that were added to the code:
63256 - XkbDir option added in February 2009
63257 by commit 76f18b94bd2719a8199334742d021c4d0806187d
63258 - DRI2 option added in April 2008
63259 by 35982bc109d424c464551ab22ec90af69908c884
63260
63261 Remove xorg.conf options that were removed from the code:
63262 - XkbDisable option was removed in January 2009
63263 by commit 40877c6680863bd6a8475f2bb0c54df55bcf0b0e
63264 - PciProbe/Config options were removed in August 2008
63265 by commit fdf7c747a8e1bc59dfb31b7f90dd5eab2687315b
63266 - EstimateSizesAggressively was removed in August 2008
63267 by commit cd1e8f26147919227e7624ac4c6b313d972a4d35
63268 - loadable font modules were removed in July 2008
63269 by commit affec10635343668839994ea2bac16c1d4524200
63270 - ModInDev options were removed in December 2008
63271 by commit 6de6ffff35ac03d49fa61de195d4a0605e0ef8bf
63272
63273 (Also strips some trailing whitespaces to make git happier.)
63274
63275 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63276 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63277
63278commit b6f5c54fc61588c638b987bde914b0e5ff9b655e
63279Author: Alan Coopersmith <alan.coopersmith@sun.com>
63280Date: Mon Sep 14 15:49:01 2009 +1000
63281
63282 xfree86: Change default font path to match configure.ac's --default-font-path.
63283
63284 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63285 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63286
63287commit 23dc4678e9333ba0ee1a5374521905fa3f4f309c
63288Author: Peter Hutterer <peter.hutterer@who-t.net>
63289Date: Mon Sep 14 14:14:27 2009 +1000
63290
63291 dmx: use top_builddir, not top_srcdir to get libxfixes.la
63292
63293 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63294 Tested-by: Byeong-ryeol Kim <brofkims@gmail.com>
63295
63296commit 4650e6ebe6e3ea48a026ace60b25daa165580467
63297Author: Peter Hutterer <peter.hutterer@who-t.net>
63298Date: Mon Sep 14 12:00:18 2009 +1000
63299
63300 xkb: drop key presses for already repeating keys. (#23889)
63301
63302 The event sequence for continuously pressed keys with the keyboard driver is
63303 PRESS - PRESS - PRESS - ... - RELEASE.
63304 The first press sets the repeatKey to the keycode and the matching timer.
63305 The second press (on the same keycode) can be silently dropped instead of
63306 overwriting the timer again.
63307
63308 X.Org Bug 23889 <http://bugs.freedesktop.org/show_bug.cgi?id=23889>
63309
63310 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63311 Acked-by: Daniel Stone <daniel@fooishbar.org>
63312
63313commit c626a5578aa16c17ef68c1b15ed53828080d9d90
63314Author: Peter Hutterer <peter.hutterer@who-t.net>
63315Date: Sun Sep 13 17:35:51 2009 +1000
63316
63317 xfree86: fix VT_WAITACTIVE control flow (#11477)
63318
63319 Move misplaced } to get the flow of
63320
63321 if (!ShareVTs) {
63322 VT_ACTIVATE
63323 VT_WAITACTIVE
63324 }
63325
63326 X.Org Bug 11477 <http://bugs.freedesktop.org/show_bug.cgi?id=11477>
63327
63328 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63329 Acked-by: Daniel Stone <daniel@fooishbar.org>
63330
63331commit d3c66d6b69c04466b8388586090dc48a0aad7684
63332Author: Peter Hutterer <peter.hutterer@who-t.net>
63333Date: Mon Sep 14 08:03:17 2009 +1000
63334
63335 dix: FindChildForEvent needs init child to None.
63336
63337 silences compiler warning:
63338 events.c: In function 'FixUpEventFromWindow':
63339 events.c:2262: warning: 'child' may be used uninitialized in this function
63340
63341 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63342
63343commit e85775aa64d1e9322c82090a26256af5e1fede0a
63344Author: Kevin E Martin <kem@redhat.com>
63345Date: Mon Sep 14 08:01:56 2009 +1000
63346
63347 Include <X11/extensions/dmx.h> and remove _DMX_SERVER_ define.
63348
63349 Fixes build with newer dmx header files that have been split between dmx.h
63350 and dmxext.h. _DMX_SERVER_ is no longer needed with the split headers.
63351
63352 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63353
63354commit 232d43510e0f9f780a9f8eb857366e03acf2fcdc
63355Author: Kevin E Martin <kem@redhat.com>
63356Date: Mon Sep 14 08:00:42 2009 +1000
63357
63358 dmx: Make description match default option.
63359
63360 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63361
63362commit 6389a190faa7c06bdedfd1cc3acb280ebb5cc04a
63363Author: Tiago Vignatti <tiago.vignatti@nokia.com>
63364Date: Sat Sep 12 21:47:51 2009 +0300
63365
63366 shave: use CC tag variable on libtool
63367
63368 This fix scratchbox compilation.
63369
63370 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
63371 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63372
63373commit dfb106c846c880a06fa5caada92f45576eab7b00
63374Author: Peter Hutterer <peter.hutterer@who-t.net>
63375Date: Tue Sep 8 18:09:12 2009 +1000
63376
63377 configure: re-enable dmx build (default=auto)
63378
63379 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63380
63381commit 72607960c136894c4533a54cef7a9be64246f1a3
63382Author: Peter Hutterer <peter.hutterer@who-t.net>
63383Date: Wed Sep 9 11:02:22 2009 +1000
63384
63385 dmx: Remove some dead code.
63386
63387 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63388
63389commit 88187ad889ab707fcae3f64890cdb291b567ec23
63390Author: Peter Hutterer <peter.hutterer@who-t.net>
63391Date: Wed Sep 9 11:02:18 2009 +1000
63392
63393 dmx: purge DMX EQ leftovers
63394
63395 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63396
63397commit f3aaa46f27f802233e54b3c70ba827e0fcca2605
63398Author: Peter Hutterer <peter.hutterer@who-t.net>
63399Date: Wed Sep 9 09:51:45 2009 +1000
63400
63401 dmx: formatting fix to improve readability
63402
63403 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63404
63405commit 7a588c833e87dd07f167b9885d5ec43dde271ebe
63406Author: Peter Hutterer <peter.hutterer@who-t.net>
63407Date: Tue Sep 8 18:14:43 2009 +1000
63408
63409 dmx: Default rules are now base, not xfree86
63410
63411 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63412
63413commit 33243ae971db90477a29dfb0e51cc00c95e561bc
63414Author: Peter Hutterer <peter.hutterer@who-t.net>
63415Date: Tue Aug 4 17:04:08 2009 +1000
63416
63417 dmx: include xfixes lib, reshuffle include order to resolve linker errors.
63418
63419 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63420
63421commit 2d46678762b12c9746fc8f89821c15705680ba35
63422Author: Peter Hutterer <peter.hutterer@who-t.net>
63423Date: Tue Sep 8 17:46:06 2009 +1000
63424
63425 dmx: remove dmx-internal event queue.
63426
63427 The EQ is in the DIX now.
63428
63429 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63430
63431commit 545f11139d600febf1c5fd46b11bea4352b82c32
63432Author: Peter Hutterer <peter.hutterer@who-t.net>
63433Date: Tue Sep 8 17:14:53 2009 +1000
63434
63435 dmx: compiler warning fix (mixed declarations + code)
63436
63437 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63438
63439commit 7d703af74ced36968f4480507a7eb54d249b5edc
63440Author: Peter Hutterer <peter.hutterer@who-t.net>
63441Date: Tue Sep 8 17:13:38 2009 +1000
63442
63443 dmx: get the state from XKB for dmxCheckSpecialKeys
63444
63445 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63446
63447commit 6e52f9233176986508356f6eee4fce86ba740e71
63448Author: Peter Hutterer <peter.hutterer@who-t.net>
63449Date: Tue Sep 8 17:10:22 2009 +1000
63450
63451 dmx: fix dmxKeySymToKeyCode to work with mandatory XKB.
63452
63453 Approach taken is inefficient, it converts the xkb symbol table to a core
63454 symbol table first and then extracts the keycode from there.
63455 Consider this a todo for a rainy afternoon when the beer fridge demands
63456 emptying.
63457
63458 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63459
63460commit 61ae0a5ca5de24590f444c405445124404ba9753
63461Author: Peter Hutterer <peter.hutterer@who-t.net>
63462Date: Tue Sep 8 17:05:12 2009 +1000
63463
63464 dmx: fix up dmxKeyCodeToKeySym for XKB-only.
63465
63466 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63467
63468commit 9caba3e2abceda0a1ff4ad4bf05708f0166c2218
63469Author: Peter Hutterer <peter.hutterer@who-t.net>
63470Date: Tue Aug 4 15:12:25 2009 +1000
63471
63472 dmx: fix two calls to InitKeyboardDeviceStruct.
63473
63474 Takes RMLVO now instead of the keysyms directly.
63475
63476 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63477
63478commit 29b9a10dad1ca780dc90f4f1f63fd11fd7d1221b
63479Author: Peter Hutterer <peter.hutterer@who-t.net>
63480Date: Tue Aug 4 15:04:45 2009 +1000
63481
63482 dmx: XkbComponentNamesRec doesn't have a keymap field anymore.
63483
63484 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63485
63486commit a6f3f6a63ae76afbe01eb9049a63609476d2ed4d
63487Author: Peter Hutterer <peter.hutterer@who-t.net>
63488Date: Tue Sep 8 16:43:04 2009 +1000
63489
63490 dmx: switch lnx-input and usb-input over to xkb-only.
63491
63492 kbd*Convert are nearly identical, it just asks for merging. Not today
63493 though.
63494
63495 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63496
63497commit 5d81aee5f75e80ae9d3ac7cb32d293ade62d6c5c
63498Author: Peter Hutterer <peter.hutterer@who-t.net>
63499Date: Tue Aug 4 15:24:02 2009 +1000
63500
63501 dmx: include xkbsrv.h to silence compiler.
63502
63503 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63504
63505commit 4da59f478686fa7e80a3837bf9fa61672c13c50b
63506Author: Peter Hutterer <peter.hutterer@who-t.net>
63507Date: Tue Sep 8 16:30:36 2009 +1000
63508
63509 xkb: split effectiveGroup calculation into separate utility function.
63510
63511 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63512
63513commit 8fb3fa28a5a1b36cdaad38055a607400828b9e1c
63514Author: Alan Coopersmith <alan.coopersmith@sun.com>
63515Date: Fri Sep 11 11:21:26 2009 -0700
63516
63517 compiler.h: include <string.h> before using memmove()
63518
63519 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
63520
63521commit d908ee6e549fd8ff653e2c58e9637df347540ef9
63522Author: Tiago Vignatti <tiago.vignatti@nokia.com>
63523Date: Thu Sep 10 20:56:18 2009 +0300
63524
63525 Revert "Make sys.c use compiler.h unaligned access functions"
63526
63527 This reverts commit 1b30545c04a51bfa3ff95a26d64962907a62ff15.
63528
63529 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
63530 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63531
63532commit 281ebb453d2f5254e057a24e580dafa9abf82305
63533Author: Tiago Vignatti <tiago.vignatti@nokia.com>
63534Date: Thu Sep 10 17:52:32 2009 +0300
63535
63536 xfree86: vgaarb: close and cleanup vga arbiter
63537
63538 So far there are no apparently issues on not closing the fd. But let's do the
63539 right job here.
63540
63541 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
63542 Acked-by: Dave Airlie <airlied@redhat.com>
63543 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63544
63545commit e62549b8d61cd739adbbd04487d96fe706da1370
63546Author: Peter Hutterer <peter.hutterer@who-t.net>
63547Date: Wed Sep 9 16:03:38 2009 +1000
63548
63549 Xext: don't try to initialize XTEST device properties if they failed.
63550
63551 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63552
63553commit 744cdc897746dca86f5ab9e57be982621c2f0986
63554Author: Peter Hutterer <peter.hutterer@who-t.net>
63555Date: Wed Sep 9 15:05:15 2009 +1000
63556
63557 dix: Remove two _X_EXPORT defines from the function definition.
63558
63559 These two are defined _X_EXPORT in their declaration anyway.
63560
63561 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63562
63563commit 9871f9be4e5b184b151ce93458b2578cc8c7b962
63564Author: Martin-Éric Racine <q-funk@iki.fi>
63565Date: Wed Sep 9 13:43:45 2009 +0300
63566
63567 Use lowercase PCI vendor ID for NSC.
63568
63569commit f4350c66b493d63fa06be87caa958d7033232ea4
63570Author: Daniel Stone <daniel@fooishbar.org>
63571Date: Wed Sep 9 15:48:33 2009 +1000
63572
63573 fbdevhw: Test for graphics:fb%d as well as graphics/fb%d
63574
63575 Apparently the kernel can't decide on an API to expose to userspace, so
63576 let's just try both in the hope that one will work.
63577
63578 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
63579 Acked-by: Michel Dänzer <michel@daenzer.net>
63580
63581commit 72f6be540840b46aa4b158b98fac11a6f90dc9d1
63582Author: Peter Hutterer <peter.hutterer@who-t.net>
63583Date: Fri Sep 4 11:13:55 2009 +1000
63584
63585 xfree86: remove now unused xf86CrtcShadowClear
63586
63587 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63588
63589commit 1f24638985ce053b696ca3359cc50b5bc26ccb03
63590Author: Peter Hutterer <peter.hutterer@who-t.net>
63591Date: Wed Sep 9 09:28:46 2009 +1000
63592
63593 kdrive: output meaningful error message for HAL devices.
63594
63595 kdrive ignores all devices from hal as they don't have the 'type' option
63596 set. Instead of "Unrecognised device identifier!" print out "Ignoring
63597 device from HAL." to indicate that the errors surrounding the device don't
63598 really matter.
63599
63600 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63601 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
63602
63603commit ae812bb79668e59cda1f851fb1fabe9db9ea705f
63604Author: Peter Hutterer <peter.hutterer@who-t.net>
63605Date: Tue Sep 8 13:53:26 2009 +1000
63606
63607 Xnest: set the default pointer accel to the host's values (#10013)
63608
63609 The defaultPointerControl holds compile-time defaults for pointer
63610 acceleration. If an Xnest instance is started, it resets the hosts pointer
63611 accel values to these built-in defaults instead of using the host values.
63612
63613 This patch queries the host for the values before initializing the device,
63614 thus leaving the host values untouched.
63615
63616 X.Org Bug 10013 <http://bugs.freedesktop.org/show_bug.cgi?id=10013>
63617
63618 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63619
63620commit 97e3f4316b1c917f93d324f05d3a4551c1d051a3
63621Author: Peter Hutterer <peter.hutterer@who-t.net>
63622Date: Tue Sep 8 09:55:22 2009 +1000
63623
63624 dix: auto-float SD's with SendCoreEvents "false"
63625
63626 AlwaysCore and SendCoreEvents specify whether a device is to send core
63627 events. A device that has either disabled is not supposed to send core
63628 events.
63629
63630 With MPX/XI2, a device that is attached automatically sends core events when
63631 the event is routed through the master device. Floating a slave device
63632 disables core events by breaking the route.
63633
63634 This patch automatically floats devices that have coreEvents disabled in the
63635 xorg.conf/HAL. This replicates the behaviour of a SendCoreEvents "false"
63636 device in server 1.6 and earlier.
63637
63638 The devices may still be reattached to a master at runtime.
63639
63640 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63641
63642commit 447bd4ce90113037a489ce55d9878523fe5294c7
63643Author: Peter Hutterer <peter.hutterer@who-t.net>
63644Date: Mon Sep 7 16:02:50 2009 +1000
63645
63646 kdrive: fix sdl build by removing old keySymsRec references.
63647
63648 These are set in kinput now through InitKeyboardDeviceStruct.
63649
63650 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63651
63652commit d8aadfa5af85ce6613289efe119e592aa687ab03
63653Author: Peter Hutterer <peter.hutterer@who-t.net>
63654Date: Mon Sep 7 15:04:33 2009 +1000
63655
63656 dix: remove unused and half-broken code to restore original classes.
63657
63658 In theory, the MD should change back to its old, original classes when the
63659 last SD is detached. Thanks to the XTEST devices, we'll always have an SD
63660 attached until the MD is removed. So let's not worry about that and do
63661 nothing instead of having some code that's essentially untested.
63662
63663 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63664
63665commit e9a507acca01234de189f0f64da63bac32dc13fd
63666Author: Peter Hutterer <peter.hutterer@who-t.net>
63667Date: Mon Sep 7 15:59:37 2009 +1000
63668
63669 kdrive: remove xkb member from KdKeyboardInfo
63670
63671 Wasn't being used by anyone.
63672
63673 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63674
63675commit cafa7c5f9017efdc33c348b6a55d80d44df0fe17
63676Author: Martin-Éric Racine <q-funk@iki.fi>
63677Date: Tue Sep 8 23:52:52 2009 +0300
63678
63679 Added comment with the PCI device ID of Geode SCx200 series.
63680
63681commit d3ad1804a5216487b5837a80f3e53b81212dcf84
63682Author: Michel Dänzer <daenzer@vmware.com>
63683Date: Tue Sep 8 15:25:19 2009 +0200
63684
63685 EXA: Only take special code path for 1x1 fill for pixmaps.
63686
63687 It doesn't make sense for windows.
63688
63689 Also double-check that the data pointer is valid.
63690
63691 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=23461 (not sure that could
63692 happen on master even without this, but just in case)
63693
63694commit a9d30f6a038166052a4684308db205bd4e60799e
63695Author: Peter Hutterer <peter.hutterer@who-t.net>
63696Date: Tue Sep 8 17:43:24 2009 +1000
63697
63698 dix: GetKeyboardValuatorEvents doesn't recurse anymore - fix comment.
63699
63700 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63701
63702commit 1bdc9ec617d357b076c9e69296018bc212d91c7d
63703Author: Eamon Walsh <efw@eamonwalsh.com>
63704Date: Fri Sep 4 21:49:29 2009 -0400
63705
63706 Fix "warning: left-hand operand of comma expression has no effect"
63707
63708 Signed-off-by: Eamon Walsh <efw@eamonwalsh.com>
63709 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63710
63711commit 9005ec4ce19d704d5055232d2a36acee3247be67
63712Author: Peter Hutterer <peter.hutterer@who-t.net>
63713Date: Fri Sep 4 11:38:03 2009 +1000
63714
63715 dix: remove dixLookupResource - we don't have any users left.
63716
63717 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63718 Acked-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
63719 Acked-by: Aaron Plattner <aplattner@nvidia.com>
63720
63721commit 1e0caf9a2ca983f171976822d9c04bacc14e9af7
63722Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
63723Date: Sun Sep 6 14:02:53 2009 -0700
63724
63725 XQuartz: launchd: Fallback on :0 if prefix:0 gives an error for the socket name
63726 (cherry picked from commit 647c871dc9f2d0adc172b401cde89ffbdfcc4d7a)
63727
63728commit 081f72390a25e2244561e238f772f98289a6a581
63729Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
63730Date: Sat Sep 5 14:04:53 2009 -0700
63731
63732 XQuartz: Fix "warning: function declaration isn’t a prototype" noise
63733 (cherry picked from commit 9a77905a975e562daa4230739937bbb0b4caf087)
63734
63735commit 7ece2cb220e4fe2c9a8c4c62361522a45adf22c1
63736Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
63737Date: Sat Sep 5 14:03:19 2009 -0700
63738
63739 XQuartz: pbproxy: 64bit fixes: Properly process an array of Atoms
63740 (cherry picked from commit 1b659cda1af02762a31cc7875e457b08c8dc68b5)
63741
63742commit 1747120043cc5b5d201b7efd06b75ef08b032922
63743Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
63744Date: Sat Sep 5 02:51:09 2009 -0700
63745
63746 XQuartz: Fix a strcpy/strcmp typo
63747 (cherry picked from commit 701c24da3ef76a05ccb2d7e84ccfa1b7c5d38e15)
63748
63749commit be37f339a9ed90b9b5e44aeb396ca73347078027
63750Author: Michel Dänzer <daenzer@vmware.com>
63751Date: Fri Sep 4 23:11:13 2009 +0200
63752
63753 EXA: Stop hiding solid or gradient pictures from the driver.
63754
63755 Add support for solid pictures in exaTryDriverSolidFill(), but otherwise just
63756 pass solid and gradient pictures to the driver Composite hook.
63757
63758 While we're at it, clean up the logic to detect Composite operations which are
63759 effectively solid fills or copies. This should also fix some false negatives
63760 and positives.
63761
63762commit 1f61bbeffdf2ad5a2c54121c75f3b049baf01708
63763Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
63764Date: Fri Sep 4 00:38:58 2009 -0700
63765
63766 XQuartz: Check the DISPLAY environment variable to see if the socket at startup is ours. If not, ignore it and fork/exec startx.
63767 (cherry picked from commit 3d5e10cce360a04bf917227615fb9b825675124f)
63768
63769commit 0a361d0e5f896351b98dc9765b4a33c79efa6f02
63770Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
63771Date: Thu Sep 3 19:40:11 2009 -0700
63772
63773 XQuartz: Use --with-launchd-id-prefix for consistency with xinit
63774
63775 Also actually honor LAUNCHD_ID_PREFIX and APPLE_APPLICATION_NAME
63776 (cherry picked from commit 990038ab006b2f5e03dcef385514ba4e4584bd25)
63777
63778commit e6e83d81ee7366779d600cbb128034e248815339
63779Author: Peter Hutterer <peter.hutterer@who-t.net>
63780Date: Fri Sep 4 14:51:08 2009 +1000
63781
63782 xserver 1.6.99.900
63783
63784 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63785
63786commit bc0841b6e87cdd752b44b17ccb021c831a3e02a5
63787Author: Peter Hutterer <peter.hutterer@who-t.net>
63788Date: Fri Sep 4 15:58:36 2009 +1000
63789
63790 shave: hack around the configure ld check for make distcheck.
63791
63792 configure calls $(CC) -print-prog-name=ld to get the binary for the linker.
63793 The shave script prints " CC \nld" which can't be parsed by configure,
63794 resulting in a fatal error.
63795
63796 This patch special-cases -print-prog-name and passes it though directly
63797 without any additional output.
63798
63799 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63800
63801commit 58d9a4c6356c98c32be2826410df027ece70d44d
63802Author: Peter Hutterer <peter.hutterer@who-t.net>
63803Date: Fri Sep 4 15:06:45 2009 +1000
63804
63805 xfree86: fix make distcheck after removal of Domain.note and RAC.Notes.
63806
63807 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63808
63809commit 8af2793a90e8e644a6a598adce3f752c15b3e37a
63810Author: Peter Hutterer <peter.hutterer@who-t.net>
63811Date: Thu Sep 3 11:30:48 2009 +1000
63812
63813 xkb: xkbGetKbdByName on the lastSlave needs to change the master (#21859)
63814
63815 If the layout is changed on a master's lastSlave, the master needs to change
63816 layout immediately. Otherwise, the master stays on the same layout until the
63817 lastSlave changes - which may not happen if only a single keyboard is
63818 available.
63819
63820 X.Org Bug 21859 <http://bugs.freedesktop.org/show_bug.cgi?id=21859>
63821
63822 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63823
63824commit 219ae5b628f9ffe2be0876d5911cc04e52bef5f8
63825Author: Peter Hutterer <peter.hutterer@who-t.net>
63826Date: Fri Aug 21 16:38:53 2009 +1000
63827
63828 dix: hide the sequenceNumber key repeat hack behind two functions.
63829
63830 Just in case we figure out a better way to do it in the future.
63831
63832 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63833
63834commit bfb219f532f3c78ba905424365ee7c5f7b5f21a2
63835Author: Daniel Stone <daniel@fooishbar.org>
63836Date: Fri Aug 21 16:03:36 2009 +1000
63837
63838 input: allow for detectable autorepeat.
63839
63840 For core and XI1 events, store the key_repeat flag in the sequence number
63841 until TryClientEvents. The sequenceNumber is unset until TryClientEvents.
63842
63843 [Also thrown in, some random indentation changes. Thanks]
63844
63845 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63846
63847commit 3d988e01e41e98fc5160f825a250522ba274d09f
63848Author: Daniel Stone <daniel@fooishbar.org>
63849Date: Fri Aug 21 15:15:41 2009 +1000
63850
63851 dix: improve code flow in TryClientEvents, better debugging messages.
63852
63853 Instead of a massive if (blah && blah), return early where possible.
63854
63855 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63856
63857commit 6fb01c8286c16968fd07e4a5b78da89cd8768d79
63858Author: Peter Hutterer <peter.hutterer@who-t.net>
63859Date: Fri Sep 4 10:57:33 2009 +1000
63860
63861 require xf86vidmodeproto 2.2.99.1 or newer.
63862
63863commit 2122fedb2c058dc27f9df0e856d2193fabe725e2
63864Author: Peter Hutterer <peter.hutterer@who-t.net>
63865Date: Fri Sep 4 11:37:58 2009 +1000
63866
63867 os: silence bigreqsproto compiler warning
63868
63869 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63870
63871commit 59e731ef66095723e74a16b2e5548aed627089ff
63872Author: Peter Hutterer <peter.hutterer@who-t.net>
63873Date: Fri Sep 4 11:06:37 2009 +1000
63874
63875 xfree86: silence some xf86dgaproto compiler warnings
63876
63877 We already require xf86dgaproto > 2.0.99 since 6fffcd582
63878
63879 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63880
63881commit ead57a13531aa14033839112fa66e33e142f67da
63882Author: Peter Hutterer <peter.hutterer@who-t.net>
63883Date: Fri Sep 4 11:03:55 2009 +1000
63884
63885 xfree86: silence 'unused variable' compiler warning
63886
63887 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63888
63889commit 21f04f008e6b1ab66a63de3de668b1f261a6928a
63890Author: Peter Hutterer <peter.hutterer@who-t.net>
63891Date: Fri Sep 4 11:01:09 2009 +1000
63892
63893 os: don't redefine GNU_SOURCE
63894
63895 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63896
63897commit 09f0ee1324b046f2830bdd260f6e2a67ee8f2973
63898Author: Martin-Éric Racine <q-funk@iki.fi>
63899Date: Fri Sep 4 04:17:09 2009 +0300
63900
63901 Formatted recent Geode PCI ID changes to fit a 80-column coding style.
63902
63903commit a17b50b130fe01c058a74f2696a8b97c756acb49
63904Author: Martin-Éric Racine <q-funk@iki.fi>
63905Date: Fri Sep 4 03:51:01 2009 +0300
63906
63907 Further explained what the Geode PCI ID is all about as comments.
63908
63909commit afd800dae4f7491baee08509094a30160bb9c849
63910Author: Martin-Éric Racine <q-funk@iki.fi>
63911Date: Fri Sep 4 03:41:08 2009 +0300
63912
63913 Removed unnecessary curly braces for Geode LX PCI ID.
63914
63915commit 842b7a47d583a7729099de3031ae20977dddf774
63916Author: Martin-Éric Racine <q-funk@iki.fi>
63917Date: Fri Sep 4 03:25:32 2009 +0300
63918
63919 Updated the PCI Vendor and Device IDs for all Geode variants.
63920
63921 In practice, some of the native drivers for older Geode products
63922 have become deprecated due to lack of e.g. libpciaccess upgrade,
63923 but that's OK, since most distributions don't ship them anymore.
63924 In that case, we'll let X server fall back to good old VESA.
63925
63926commit 8e873185f4cbfb2a36e1f43fe7da47fd9fd5aeea
63927Author: Michel Dänzer <daenzer@vmware.com>
63928Date: Thu Sep 3 19:11:02 2009 +0200
63929
63930 EXA: Preserve pPixmap->devPrivate.ptr in exaPixmapIsOffscreen_mixed.
63931
63932 Otherwise we may incorrectly clobber it to NULL on repeated PrepareAccess
63933 calls.
63934
63935commit 120286aef59dabdb7c9fa762e08457e5cc8ec3a6
63936Author: Michel Dänzer <daenzer@vmware.com>
63937Date: Thu Sep 3 08:05:59 2009 +0200
63938
63939 glx: Add screen DestroyWindow wrapper to destroy the GLX drawable.
63940
63941 Fixes crashes exitting MacSlow's rgba-glx demo.
63942
63943commit f04fe06ae244b851b38be824b1a80f2f8a030591
63944Author: Peter Hutterer <peter.hutterer@who-t.net>
63945Date: Thu Sep 3 14:17:14 2009 +1000
63946
63947 dix: don't return BadMatch from GetProperty (#23562)
63948
63949 dixLookupWindow may return BadMatch if the window in question isn't actually
63950 a window. In this case, GetProperty needs to return BadWindow - not
63951 BadMatch.
63952
63953 X.Org Bug 23562 <http://bugs.freedesktop.org/show_bug.cgi?id=23562>
63954
63955 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63956
63957commit 84eb4c66a4a09c360cef260fb2f35dfb6d8a93c6
63958Author: Peter Hutterer <peter.hutterer@who-t.net>
63959Date: Thu Sep 3 10:00:42 2009 +1000
63960
63961 test: add protocol testing for XIWarpPointer.
63962
63963 TODO: some way to check src_x/y coordinates would be good.
63964
63965 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63966
63967commit 1b7858e8469aea6d2031039ba41d7191a4b80f28
63968Author: Peter Hutterer <peter.hutterer@who-t.net>
63969Date: Thu Sep 3 10:58:00 2009 +1000
63970
63971 test: expose the default screen to tests, some cleanup work.
63972
63973 Provide common #define for invalid window IDs.
63974 Init the sprite's hotPhys, provide a common #define for the initial sprite
63975 position.
63976
63977 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63978
63979commit 61a6e1f074d9ff75d61446b946aab6c04019c287
63980Author: Peter Hutterer <peter.hutterer@who-t.net>
63981Date: Thu Sep 3 10:00:00 2009 +1000
63982
63983 Xi: return BadDevice for master kbds and attached slaves in XIWarpPointer
63984
63985 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63986
63987commit 8939ad2b2aa0385f072d3e1169eaf99289ed737a
63988Author: Peter Hutterer <peter.hutterer@who-t.net>
63989Date: Thu Sep 3 09:44:11 2009 +1000
63990
63991 Xi: return error values to client from XIWarpPointer.
63992
63993 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
63994
63995commit d481720c57f558c47273bf89854fc69254319772
63996Author: Peter Hutterer <peter.hutterer@who-t.net>
63997Date: Wed Sep 2 15:28:28 2009 +1000
63998
63999 Xi: standardise XI2 headers.
64000
64001 Adding missing dix-config.h include, adding one missing header guard.
64002
64003 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64004
64005commit 377a5f655ca88cd836da24ab42361df9f3ff51a2
64006Author: Peter Hutterer <peter.hutterer@who-t.net>
64007Date: Wed Sep 2 11:16:15 2009 +1000
64008
64009 Xi: if XISetEventMask fails, return this to the client.
64010
64011 The only failure point can be a BadAlloc.
64012
64013 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64014
64015commit 993ca5acbdccef9e43cd6580e7cc22cfda1a2578
64016Author: Peter Hutterer <peter.hutterer@who-t.net>
64017Date: Wed Sep 2 10:53:40 2009 +1000
64018
64019 dix: when unsetting a cursor, update the sprite immediately (#23608)
64020
64021 Removing the device cursor while the cursor was within the window did not
64022 update the visible sprite until the next enter/leave event.
64023
64024 X.Org Bug 23608 <http://bugs.freedesktop.org/show_bug.cgi?id=23608>
64025
64026 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64027
64028commit b41db8fe460fc0956922b79261752d1bb104684f
64029Author: Peter Hutterer <peter.hutterer@who-t.net>
64030Date: Wed Sep 2 08:57:40 2009 +1000
64031
64032 Xi: fix broken swap code in XISelectEvents request processing.
64033
64034 The pointer advanced 12 bytes too short. Rather unfortunate if both the code
64035 and the test have the same bug.
64036
64037 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64038
64039commit 727de7c90de4198222e0dc58d7503b4a1672f642
64040Author: Peter Hutterer <peter.hutterer@who-t.net>
64041Date: Tue Sep 1 16:33:56 2009 +1000
64042
64043 Xi: Unify checking for invalid bits in grab masks.
64044
64045 Bits above XI2LASTEVENT are invalid and cause in BadValues. These checks
64046 must be performed anywhere where a mask_len parameter is given.
64047
64048 This patch also adds the missing checks to grab masks.
64049
64050 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64051
64052commit 59a6d7d478903a8bc9c5d4cc8b2e62e2ad102dba
64053Author: Peter Hutterer <peter.hutterer@who-t.net>
64054Date: Tue Sep 1 16:30:52 2009 +1000
64055
64056 Xi: don't overrun memory for grab masks.
64057
64058 A grab mask provided in the request may be larger than the one used in the
64059 server. Cut down to size before memcopying.
64060
64061 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64062
64063commit 83db2b126e8623824e1303e74070375994984599
64064Author: Peter Hutterer <peter.hutterer@who-t.net>
64065Date: Tue Sep 1 15:58:02 2009 +1000
64066
64067 Xi: don't try to set oversized or non-existing masks.
64068
64069 Fixes crash if the first XISelectEvents has a zero sized event mask.
64070 Fixes crash if the mask provided is larger than others->xi2mask[].
64071
64072 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64073
64074commit 6fe235d3635aaf995dfa198c60ded1a43fe9dd45
64075Author: Peter Hutterer <peter.hutterer@who-t.net>
64076Date: Tue Sep 1 15:46:34 2009 +1000
64077
64078 test: fix request length calculation, add length tests for XISelectEvents
64079
64080 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64081
64082commit 7ab1b50c545ae1a62384bedee60168e227bb2a94
64083Author: Peter Hutterer <peter.hutterer@who-t.net>
64084Date: Tue Sep 1 15:27:29 2009 +1000
64085
64086 test: fix build after changing libxf86config.la
64087
64088 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64089
64090commit 58c298acc1045927e0d90be73b8dbc8837252589
64091Author: Peter Hutterer <peter.hutterer@who-t.net>
64092Date: Tue Sep 1 15:16:17 2009 +1000
64093
64094 Xi: extra length checking for requests providing masks.
64095
64096 masks can be of arbitrary length. If the client did not initialize mask_len,
64097 some sort of boundary check is needed to avoid running over memory.
64098
64099 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64100
64101commit 0e4dd3b2d28d3dbbfc152d6f5030901ec063a7ae
64102Author: Peter Hutterer <peter.hutterer@who-t.net>
64103Date: Mon Aug 31 09:50:36 2009 +1000
64104
64105 Un-export ApplyPointerMapping. This is an in-server function.
64106
64107 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64108
64109commit a470e8426ee290046a927a17fc969235c4b1e2e7
64110Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64111Date: Wed Sep 2 17:56:52 2009 -0400
64112
64113 Correct outdated e-mail address in "Author" statements.
64114
64115 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64116
64117commit 17545ed80ee3bb6a058a2748fc7bfb0fc5139fc1
64118Author: Adam Jackson <ajax@redhat.com>
64119Date: Wed Sep 2 14:40:07 2009 -0400
64120
64121 randr: Fix crtcs using set_mode_major()
64122
64123 We'd never mark the crtc as active, meaning (among other things) gamma
64124 upload wouldn't work.
64125
64126commit 291408980f33b1e541c89d958535e6fad55fdac9
64127Author: Dave Airlie <airlied@linux.ie>
64128Date: Wed Sep 2 20:14:30 2009 +1000
64129
64130 vgaarb: protect fini as well just in case
64131
64132commit 8aeadbff0840efc0c9a073e60d6fec0dba313060
64133Author: Dave Airlie <airlied@linux.ie>
64134Date: Wed Sep 2 19:07:40 2009 +1000
64135
64136 vgaarb: if arb init fails, make sure locking doesn't occur.
64137
64138 pointed out by Martin Jansa on irc.
64139
64140 Signed-off-by: Dave Airlie <airlied@redhat.com>
64141
64142commit 0e46066efea4e385b2874d4a62a786957e572fe5
64143Author: Jerome Glisse <jglisse@redhat.com>
64144Date: Tue Sep 1 13:59:47 2009 +0200
64145
64146 Xext: fix a typo for bigreqsproto.h header file
64147
64148 Signed-off-by: Jerome Glisse <jglisse@redhat.com>
64149
64150commit c61d8ce1507318476b75b81bf871da19cb1bf747
64151Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
64152Date: Mon Aug 31 17:57:50 2009 -0700
64153
64154 Xext: Require newer versions of xcmiscproto, bigreqsproto, and xf86bigfontproto for new *proto.h header file names
64155
64156 bigreqsproto >= 1.1.0
64157 xcmiscproto >= 1.2.0
64158 xf86bigfontproto >= 1.2.0
64159
64160commit 1c37be80982151a5acd74fe4b960695f604f22ca
64161Author: Michael Witrant <mike@lepton.fr>
64162Date: Tue Sep 1 10:46:50 2009 +1000
64163
64164 XFree86: Linux: Fix 100% CPU usage with ShareVTs and kbd
64165
64166 Leave consoleFd open over the course of the server, even though any use
64167 of it in this context is likely to be disastrous.
64168
64169 Signed-off-by: Michael Witrant <mike@lepton.fr>
64170 Reviewed-by: Daniel Stone <daniel@fooishbar.org>
64171
64172commit 65183dc3153232943dff3cfbaacab141e7ed0e9c
64173Author: Daniel Stone <daniel@fooishbar.org>
64174Date: Sun Jul 5 19:53:55 2009 +0300
64175
64176 Input: Mark Xi input events as critical
64177
64178 Note that the Xi events are critical and should thus cause a flush to
64179 the client when an input event is pending.
64180
64181 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
64182
64183commit a4e614d301cfa2577de1ee6a15abf023590586d1
64184Author: Daniel Stone <daniel@fooishbar.org>
64185Date: Sat Apr 25 16:46:10 2009 +1000
64186
64187 Xi: Fix harmless ButtonPress/ButtonRelease confusion
64188
64189 A missing break meant that ButtonPress would fall through into
64190 ButtonRelease, but luckily it appears to have been completely harmless.
64191
64192 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
64193
64194commit 5ef53a94ce4e48e11de26290cd677266308640c8
64195Author: Matt Turner <mattst88@gmail.com>
64196Date: Mon Aug 31 20:23:26 2009 -0400
64197
64198 alpha: kill xf86SlowBCopyToBus and xf86SlowBCopyFromBus
64199
64200 xf86SlowBCopyToBus and xf86SlowBCopyFromBus cause segfaults on my
64201 system.
64202
64203 Also remove associated slowbcopy_tobus/slowbcopy_frombus macros.
64204
64205 Signed-off-by: Matt Turner <mattst88@gmail.com>
64206
64207commit a8e634b86d47353ef354717904fa66ebe35e1303
64208Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
64209Date: Mon Aug 31 15:53:58 2009 -0700
64210
64211 XQuartz: GLX: Drawable does not contain resize anymore.
64212 (cherry picked from commit 1e642f22fbf029bb3917091cb5fa2f78b4a92f62)
64213
64214commit 1b30545c04a51bfa3ff95a26d64962907a62ff15
64215Author: Matt Turner <mattst88@gmail.com>
64216Date: Mon Aug 31 16:16:02 2009 -0400
64217
64218 Make sys.c use compiler.h unaligned access functions
64219
64220 Signed-off-by: Matt Turner <mattst88@gmail.com>
64221
64222commit 9522ab9a98e98bc394703038e34b1c12aa39a179
64223Author: Matt Turner <mattst88@gmail.com>
64224Date: Mon Aug 31 16:12:29 2009 -0400
64225
64226 Remove unused Delay.c
64227
64228 Signed-off-by: Matt Turner <mattst88@gmail.com>
64229
64230commit 4fac13ff8848bc38654b5567170fbc0d7b2c96d6
64231Author: Matt Turner <mattst88@gmail.com>
64232Date: Mon Aug 31 16:11:59 2009 -0400
64233
64234 Remove unused BUSmemcpy.c
64235
64236 BUSmemcpy.c provides xf86BusToMem and xf86MemToBus, which are are memcpy
64237 wrappers written to avoid glibc's memcpy on Alpha. glibc'c memcpy on
64238 Alpha has improved much since this was written, so it's no longer
64239 needed. Neither function is used inside the xserver, and no module on
64240 my machine uses either as well.
64241
64242 Signed-off-by: Matt Turner <mattst88@gmail.com>
64243
64244commit b65146fc8418b53b7e43dd34cc20000369475671
64245Author: Matt Turner <mattst88@gmail.com>
64246Date: Mon Aug 31 16:11:10 2009 -0400
64247
64248 Remove unused ia64_flush_cache function
64249
64250 Signed-off-by: Matt Turner <mattst88@gmail.com>
64251
64252commit f201705e29bdc3309dd611b461ec67740cb706ef
64253Author: Matt Turner <mattst88@gmail.com>
64254Date: Mon Aug 31 16:09:33 2009 -0400
64255
64256 Simplify unaligned access code
64257
64258 All architectures should be able to use the same unaligned access code,
64259 regardless of whether they need special unaligned access instructions.
64260 Let's let gcc do the heavy lifting.
64261
64262 In the case that we're not using a gcc-compatible compiler, use memmove.
64263
64264 The xserver already requires pixman, so include pixman.h for its uint*_t
64265 types.
64266
64267 Signed-off-by: Matt Turner <mattst88@gmail.com>
64268
64269commit 3718beb69cc5b4cdb79dd97242fce9231a26a39a
64270Author: Matt Turner <mattst88@gmail.com>
64271Date: Mon Aug 31 16:08:52 2009 -0400
64272
64273 Clean up unaligned access functions on alpha
64274
64275 Checks for __GNUC__ are superfluous since the only other compiler for
64276 the platform is Compaq C, and it doesn't support GCC style inline
64277 assembly.
64278
64279 Signed-off-by: Matt Turner <mattst88@gmail.com>
64280
64281commit e1fdd7671346ce374a2b77fe94b8c1a25c0520da
64282Author: Matt Turner <mattst88@gmail.com>
64283Date: Mon Aug 31 16:08:20 2009 -0400
64284
64285 Move unaligned access functions to common section
64286
64287 Signed-off-by: Matt Turner <mattst88@gmail.com>
64288
64289commit 628b863bcc6a2a14fd3c0e6c2be4394833cc314c
64290Author: Matt Turner <mattst88@gmail.com>
64291Date: Mon Aug 31 15:53:41 2009 -0400
64292
64293 Remove unused arm_flush_cache function
64294
64295 Signed-off-by: Matt Turner <mattst88@gmail.com>
64296
64297commit a544139196ac7f5bc89bc642c6c5c0ea1b0bb60e
64298Author: Matt Turner <mattst88@gmail.com>
64299Date: Mon Aug 31 15:53:19 2009 -0400
64300
64301 Remove unused ppc_flush_icache function
64302
64303 Signed-off-by: Matt Turner <mattst88@gmail.com>
64304
64305commit 52aa0495218dc7821a23b045b2c74b2a66e34616
64306Author: Matt Turner <mattst88@gmail.com>
64307Date: Mon Aug 31 15:52:39 2009 -0400
64308
64309 Replace 8 nops with proper sync instruction on mips
64310
64311 Cc: Ralf Baechle <ralf@linux-mips.org>
64312 Acked-by: David Daney <ddaney@caviumnetworks.com>
64313 Signed-off-by: Matt Turner <mattst88@gmail.com>
64314
64315commit 563fa1c5d762173cd6fb78f9e33cb960ef3153bb
64316Author: Matt Turner <mattst88@gmail.com>
64317Date: Mon Aug 31 15:52:10 2009 -0400
64318
64319 Use sfence and mfence instructions on amd64
64320
64321 Signed-off-by: Matt Turner <mattst88@gmail.com>
64322
64323commit 430c4af0d6e2cad03ca5f49ec6e87f3717d0d9e2
64324Author: Matt Turner <mattst88@gmail.com>
64325Date: Mon Aug 31 15:51:41 2009 -0400
64326
64327 Add x86 barrier macros
64328
64329 Signed-off-by: Matt Turner <mattst88@gmail.com>
64330
64331commit 0d87f77a3ff4b22729220b7edcb98c6f06d0787c
64332Author: Matt Turner <mattst88@gmail.com>
64333Date: Mon Aug 31 15:51:05 2009 -0400
64334
64335 define barrier macros as nops if not otherwise defined
64336
64337 Signed-off-by: Matt Turner <mattst88@gmail.com>
64338
64339commit 1b43ee13b30d9e3f828e94cdf686d103ce1cac07
64340Author: Matt Turner <mattst88@gmail.com>
64341Date: Mon Aug 31 15:53:58 2009 -0400
64342
64343 Move memory barrier macros into common section
64344
64345 alphabetize by architecture also.
64346
64347 Signed-off-by: Matt Turner <mattst88@gmail.com>
64348
64349commit 295e598d9a83ff7162a093810ca9f546d00c0968
64350Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64351Date: Mon Aug 31 20:04:51 2009 +0300
64352
64353 xfree86: remove _more_ RAC junk
64354
64355 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64356
64357commit 38627b2eaf64aa8fe694fc3e1f5e4fee0c9c11c3
64358Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64359Date: Mon Aug 31 19:36:02 2009 +0300
64360
64361 xfree86: shut up vgaarb warnings when server doesn't support it
64362
64363 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64364
64365commit 79182538b43ecd35bf2b47eccfb2dd3bbcdc12eb
64366Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64367Date: Mon Aug 31 19:25:13 2009 +0300
64368
64369 xfree86: removal of some dead code due VGA arbiter's inclusion
64370
64371 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64372
64373commit 6f3cc08ec01609c7efe424ac3e30fe4ff9391007
64374Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
64375Date: Thu Aug 27 16:21:11 2009 -0700
64376
64377 XQuartz: GL: Unset GL_EXT_gpu_program_parameters for Tiger/ppc
64378
64379 See http://trac.macports.org/ticket/20638
64380 (cherry picked from commit 0f3a89d306838b3c75a73cd1e9e2928737222b70)
64381
64382commit fdb29ebeed143fa05630966b847b05399a446ddc
64383Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
64384Date: Thu Aug 27 11:04:00 2009 -0700
64385
64386 XQuartz: GL: Explicitly set GL_EXT symbols to 0 if they are not in OpenGL.framework to prevent X11's glext.h from setting them to 1.
64387 (cherry picked from commit 7fe37137d826d1b698e87a5b35050dd02f0a4d4b)
64388
64389commit 3be80bd98357cf362aa9a004d1292e5167d83a17
64390Author: Peter Hutterer <peter.hutterer@who-t.net>
64391Date: Fri Aug 28 16:20:59 2009 +1000
64392
64393 Xi: silence compiler warning "type may be used uninitialized"
64394
64395 GrabKey and GrabButton are only called from XI/XI2 code. Set type to -1,
64396 just in case.
64397
64398 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64399
64400commit 24716b9254fa7d609792596723a192bb044a7d3f
64401Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64402Date: Fri Aug 28 22:52:48 2009 -0400
64403
64404 xace: fix up access modes in dixLookupDrawable calls from dri2.
64405
64406 Referencing a screen through a drawable only requires GetAttr access.
64407 Treat dri2 drawables as child windows (Add/Remove access).
64408 Treat getting buffers as intent to read/write the drawable.
64409
64410 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64411
64412commit 0f3a64c3a5bce6cf3f00fe42cc7d1d3e69822945
64413Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64414Date: Thu Aug 27 15:11:12 2009 -0400
64415
64416 Add DRI2 requests to protocol.txt
64417
64418 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64419
64420commit 3d17c4f6e1538986fe7b1f17614d76e28f2d1e11
64421Author: Adam Jackson <ajax@redhat.com>
64422Date: Fri Aug 28 12:12:52 2009 -0400
64423
64424 EDID: Vendor detailed blocks aren't worth X_WARNING about
64425
64426commit 2c1a845521a31f08938b0f8d6e38835ce127c94a
64427Author: Adam Jackson <ajax@redhat.com>
64428Date: Fri Aug 28 12:09:57 2009 -0400
64429
64430 EDID: Print 1152x864 in established timings, not x870
64431
64432 The spec says x870, but we actually use x864 because that's a real DMT
64433 mode and x870 isn't. This might or might not be wrong, but we should at
64434 least tell the truth.
64435
64436commit 2a806d7fa372e28d039761c9b4087cf812e8e46b
64437Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64438Date: Fri Aug 28 16:05:04 2009 +0300
64439
64440 os: remove unused -cursor option
64441
64442 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64443
64444commit 9040dab76182d1a019ca7fef7b29733d2c199e61
64445Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64446Date: Fri Aug 28 15:50:50 2009 +0300
64447
64448 xfixes: minor clean ups on createInvisibleCursor()
64449
64450 - unused return value
64451 - no reason to declare static variable given the function is just called
64452 once
64453 - no reason to declare different type and cast it after.
64454
64455 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64456
64457commit 80ed8096f9b8e974b556a6c9f3b600fb71b994c3
64458Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64459Date: Fri Aug 28 15:33:19 2009 +0300
64460
64461 render: delete unused headers declaration
64462
64463 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64464
64465commit f959b1e5485b93700c7da449a98182f5ce067ca2
64466Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64467Date: Fri Aug 28 15:30:21 2009 +0300
64468
64469 render: AnimCurInit and AnimCursorCreate shouldn't be _X_EXPORT
64470
64471 Pointed by Peter Hutterer on xorg-devel ml.
64472
64473 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64474
64475commit e454f106dc65ecfacc154a1fa0810935022a8fee
64476Author: Dave Airlie <airlied@redhat.com>
64477Date: Fri Aug 28 11:39:42 2009 +1000
64478
64479 xf86 ddx: add vga arbiter support.
64480
64481 This adds support for using the libpciaccess interface for
64482 vga arbitration support on top of a kernel which supports it.
64483
64484 Currently patches are queued for kernel 2.6.32 in jbarnes
64485 pci tree, and shipping in Fedora kernel.
64486
64487 Co-authors:
64488 Tiago Vignatti <tiago.vignatti@nokia.com>
64489
64490 Signed-off-by: Dave Airlie <airlied@redhat.com>
64491
64492commit 6fffcd5825454a7fe58ffbcfb219f007cf38e731
64493Author: Peter Hutterer <peter.hutterer@who-t.net>
64494Date: Thu Aug 27 13:43:15 2009 +1000
64495
64496 xfree86: require xf86dgaproto 2.0.99.1.
64497
64498 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64499
64500commit 2fba2eac0b4c8d07bdf7bea20ef75ff579621728
64501Author: Peter Hutterer <peter.hutterer@who-t.net>
64502Date: Mon Aug 24 11:23:23 2009 +1000
64503
64504 test: add a few tests for xtest device initialization.
64505
64506 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64507
64508commit 903c3db1d1685bd855dceed9e7b92890743663e1
64509Author: Peter Hutterer <peter.hutterer@who-t.net>
64510Date: Mon Aug 24 09:48:00 2009 +1000
64511
64512 Xext: rename Xtst* to XTest*
64513
64514 This patch corrects a misnaming of XTest-related functions.
64515
64516 The extension itself announces itself as XTEST. Xtst is the library name
64517 itself, but all library functions are prefixed by XTest. Same with the
64518 naming in the server.
64519
64520 - Rename all *Xtst* functions to *XTest* for consistency with the library
64521 and in-server API.
64522 - Rename the "Xtst device" property to "XTEST device" for consistency with
64523 the extension naming.
64524 - Rename the device naming to "<master device name> XTEST device". The
64525 default xtest devices become "Virtual core XTEST pointer" and "Virtual
64526 core XTEST keyboard".
64527
64528 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64529
64530commit 8bfd23e144e51401e3756de9260a4811fcc59e91
64531Author: Peter Hutterer <peter.hutterer@who-t.net>
64532Date: Mon Aug 24 09:38:15 2009 +1000
64533
64534 input: move XTest device initialization into Xext/xtest.c
64535
64536 XTest devices are non-optional but nonetheless specific to the XTEST
64537 extension.
64538
64539 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64540
64541commit a95f80fa914678d360f6dfd2b58926193df6ea4a
64542Author: Peter Hutterer <peter.hutterer@who-t.net>
64543Date: Mon Aug 24 09:26:56 2009 +1000
64544
64545 dix: use IsXtstDevice instead of the direct key lookup.
64546
64547 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64548
64549commit 16b7ebd7d876034edfe8f74562bd06e747879d79
64550Author: Peter Hutterer <peter.hutterer@who-t.net>
64551Date: Mon Aug 24 14:34:23 2009 +1000
64552
64553 xfree86: Remove xf86GetMotionEvents from public API.
64554
64555 This function was used as the default motion event queue API until
64556 including XINPUT_ABI 2 (server 1.5).
64557
64558 This API was broken with 1883485 in May 2008 (wrong casting of parameters)
64559 and isn't in use by input drivers past ABI 3.
64560
64561 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64562 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64563
64564commit af76bddd7a47092e2f0f85b2547c1cc74eb5cebc
64565Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
64566Date: Sun Aug 9 16:13:32 2009 -0700
64567
64568 XQuartz: Check NSINTEGER_DEFINED to make sure the NSInteger and NSUInteger types are defined.
64569 (cherry picked from commit 36e02a8649c0a545084cce5fb2c5717277b9273c)
64570
64571commit 07c935b72fb14cb12f1cb5afbf8cc81f97086bac
64572Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
64573Date: Tue Aug 25 20:11:38 2009 -0700
64574
64575 XQuartz: Use applewmproto 1.4 updated headers.
64576 (cherry picked from commit 203df06c0eb2dcd5adfa788a1ba9569650c15f9c)
64577
64578commit 0f613007cd43545eda40825bd74ada3e6e5f36f6
64579Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
64580Date: Wed Aug 5 17:43:29 2009 -0700
64581
64582 CheckWindowOptionalNeed: Ensure w->optional is set to avoid SEGFAULT
64583
64584 If CheckWindowOptionalNeed is called consecutively (and the first removes optional), then
64585 we will SEGFAULT. This can happen in XQuartz because ReparentWindow will call
64586 pScreen->ReparentWindow which can DeleteProperty which will CheckWindowOptionalNeed... then
64587 ReparentWindow will call it again later indiscriminantly.
64588 (cherry picked from commit b608c864ccc59628bd33c033393121b61580460e)
64589
64590commit e7dd1efef408effe52d0bd3d3aa0b5d4ee10ed90
64591Author: Keith Packard <keithp@keithp.com>
64592Date: Tue Aug 25 18:07:00 2009 -0700
64593
64594 Ensure that rotation updates happen frequently
64595
64596 The smart scheduler is designed to minimize scheduler overhead by
64597 increasing the interval between WaitForSomething calls when a single
64598 client is running. However, the software rotation code depends on
64599 its BlockHandler being invoked for screen updates; the long delays
64600 caused by the smart scheduler optimizations means that screen updates
64601 can be delayed a long time as well.
64602
64603 The change is simple -- prevent the smart scheduler from increasing
64604 the scheduling interval while any screen is using software rotation.
64605
64606 Signed-off-by: Keith Packard <keithp@keithp.com>
64607
64608commit 1740cda7a37abc7d0a169ab4555b446adaa62211
64609Author: Keith Packard <keithp@keithp.com>
64610Date: Tue Aug 25 16:58:07 2009 -0700
64611
64612 Perform rotation redisplay before calling driver block handler (which may flush rendering)
64613
64614 The rotation block handler uses regular driver rendering functions to
64615 repaint the screen, if those functions queue commands in the driver,
64616 it's important that the driver block handler be invoked after the
64617 rotated image is drawn.
64618
64619 Signed-off-by: Keith Packard <keithp@keithp.com>
64620
64621commit 4aab05e3b3231f1ec9795a66a075d17a722634a7
64622Author: Keith Packard <keithp@keithp.com>
64623Date: Tue Aug 25 16:54:16 2009 -0700
64624
64625 xf86_reload_cursors: fix cursor position to eliminate jumping after mode set
64626
64627 xf86_reload_cursors restores the cursor to the correct position, but
64628 that must adjust for cursor hot spot and frame before calling down to
64629 the hardware function, otherwise the cursor jumps to the wrong
64630 position until it is repositioned by the user.
64631
64632 Signed-off-by: Keith Packard <keithp@keithp.com>
64633
64634commit 0f9ffc887ca1471e98df746253d9300e03e46a15
64635Author: Peter Hutterer <peter.hutterer@who-t.net>
64636Date: Mon Aug 24 15:10:52 2009 +1000
64637
64638 Xi: fix XIWarpPointer up for FP3232 as input coordinates.
64639
64640 requires inputproto 1.9.99.902
64641
64642 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64643
64644commit 5e96945cf54136afdb80cc17f67611251d59205d
64645Author: Peter Hutterer <peter.hutterer@who-t.net>
64646Date: Fri Aug 21 12:15:54 2009 +1000
64647
64648 Xi: fix swapping for XIWarpPointer and XIChangeCursor requests.
64649
64650 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64651
64652commit 990d204ef8074665763ad129d0b015b7ab3ae9ef
64653Author: Peter Hutterer <peter.hutterer@who-t.net>
64654Date: Mon Aug 24 09:22:16 2009 +1000
64655
64656 Xext: remove un-used extern of DeviceMotionNotify.
64657
64658 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64659
64660commit 80f18a73267bd3090e459de609a7006a21e6c649
64661Author: Peter Hutterer <peter.hutterer@who-t.net>
64662Date: Mon Aug 24 09:21:03 2009 +1000
64663
64664 input: move CorePointer/KeyboardProc declarations into header.
64665
64666 The extern declaration in xichangehierarchy.c was broken anyway.
64667 This fixes a crash on creating a new master device.
64668
64669 Reported-by: Maxim Levitsky
64670 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64671
64672commit b96823713129f641ee99306e3cadba8dfe21afa2
64673Author: Peter Hutterer <peter.hutterer@who-t.net>
64674Date: Mon Aug 24 09:06:29 2009 +1000
64675
64676 mi: include shmint.h if needed, silence compiler warning.
64677
64678 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64679
64680commit 1b3859a49a3861517c9b34f6e8750f7ae0ff2b5d
64681Author: Peter Hutterer <peter.hutterer@who-t.net>
64682Date: Fri Aug 21 12:29:39 2009 +1000
64683
64684 Xi: add swapping hook for XIGetFocus reply.
64685
64686 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64687
64688commit fe430d7d3b0a85fbea9e93171423688f76dc9e1e
64689Author: Peter Hutterer <peter.hutterer@who-t.net>
64690Date: Fri Aug 21 12:09:32 2009 +1000
64691
64692 test: add XIQueryPointer protocol tests.
64693
64694 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64695
64696commit 285c88018ba4b3819500027c9b6a05b8c27ff2ee
64697Author: Peter Hutterer <peter.hutterer@who-t.net>
64698Date: Fri Aug 21 11:38:42 2009 +1000
64699
64700 Xi: correct length field for XIQueryPointer reply.
64701
64702 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64703
64704commit e4ea91a02d6fac9dad844150c5d2fbe7dc9629b1
64705Author: Peter Hutterer <peter.hutterer@who-t.net>
64706Date: Fri Aug 21 12:12:05 2009 +1000
64707
64708 Xi: allow XIQueryPointer requests for master pointers and floating slaves.
64709
64710 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64711
64712commit 8e396f2b806496cdcac233cc731b0322735020c3
64713Author: Peter Hutterer <peter.hutterer@who-t.net>
64714Date: Fri Aug 21 12:11:19 2009 +1000
64715
64716 Xi: fix swapping for XIQueryPointer request/reply handling.
64717
64718 buttons_size is necessary as WriteToClient swaps the buttons_len field,
64719 resulting in the wrong number of bytes being written later.
64720
64721 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64722
64723commit f1c856ef9073d9d02bc84ce1472883c8441a9758
64724Author: Peter Hutterer <peter.hutterer@who-t.net>
64725Date: Fri Aug 21 10:42:15 2009 +1000
64726
64727 test: fake initialization of a sprite trace and window locations
64728
64729 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64730
64731commit 58fd28f6b6bebab7f31c54e696f96d64ebb50f5a
64732Author: Peter Hutterer <peter.hutterer@who-t.net>
64733Date: Fri Aug 21 09:44:37 2009 +1000
64734
64735 Xi: return deviceid as error value in XIQueryPointer.
64736
64737 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64738
64739commit b41dd4328bfb8a0981d299bb60f986a2bc55fd24
64740Author: Peter Hutterer <peter.hutterer@who-t.net>
64741Date: Fri Aug 21 09:43:46 2009 +1000
64742
64743 record: include recordproto.h, require recordproto 1.13.99.1
64744
64745 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64746
64747commit 65354e5a698a5b527db09afc431110afba0e14b2
64748Author: Peter Hutterer <peter.hutterer@who-t.net>
64749Date: Thu Aug 20 10:52:52 2009 +1000
64750
64751 Xi: return BadDevice for changing device cursors on non-master pointers.
64752
64753 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64754
64755commit ef75301854d41c6b11dd2742fa9d17a5de8a4cfa
64756Author: Peter Hutterer <peter.hutterer@who-t.net>
64757Date: Thu Aug 20 10:52:24 2009 +1000
64758
64759 dix: FixUpEventFromWindow mustn't scramble non-device events.
64760
64761 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64762
64763commit c73cd3b265c301b8a54ffe484d6c696f2abefb46
64764Author: Dan Nicholson <dbn.lists@gmail.com>
64765Date: Sun Aug 16 18:07:49 2009 -0700
64766
64767 xfree86: Link libselinux with Xorg system libraries
64768
64769 Put the reference to -lselinux in XORG_SYS_LIBS so it isn't added as a
64770 make dependency for libxorg.la. Otherwise, make goes looking for a file
64771 -lselinux in the current directory, which it obviously won't find.
64772
64773 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
64774 Tested-by: Justin P. Mattock <justinmattock@gmail.com>
64775
64776commit 9d8fc33cae2fee7777edd20658dbf6a3ffb3d557
64777Author: Adam Jackson <ajax@redhat.com>
64778Date: Thu Aug 20 16:56:04 2009 -0400
64779
64780 composite: Move screen init before extension registration.
64781
64782 Otherwise, you'd still advertise the extension even if no screens
64783 actually supported it, and the first Composite protocol request would
64784 probably crash.
64785
64786commit ae68708ebe24c4cdf8c5265e206f6913d9298f37
64787Author: Adam Jackson <ajax@redhat.com>
64788Date: Thu Aug 20 15:45:57 2009 -0400
64789
64790 xfree86: dump /proc/cmdline in the log on Linux
64791
64792commit 0b131a5cd91cea54240777c66a9cd385029e8cb2
64793Author: Adam Jackson <ajax@redhat.com>
64794Date: Thu Aug 20 15:28:57 2009 -0400
64795
64796 linux: Yet more malloc() avoidance for backtrace()
64797
64798 Turns out, there's an initializer at the top of backtrace() that (on
64799 some arches) calls dlopen(). dlopen(), unsurprisingly, calls malloc().
64800 So, call backtrace() early in signal handler setup so we can later
64801 safely call it from the signal handler itself.
64802
64803commit 792dee3854cbede64e17fdc4736831edad295706
64804Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64805Date: Wed Aug 19 18:20:54 2009 +0300
64806
64807 xfree86: remove bus state change notification callback
64808
64809 No one is using bus notifications now. We hope that the kernel take care of
64810 this properly.
64811
64812 For other not-so-urgent-notifications (ACPI wakeups, etc) we can just register
64813 a handler on server's scheduler (using xf86AddGeneralHandler). And for
64814 external applications, the "trend" is to use HAL to kick notifications. So
64815 we're already provided of enough notification schemes.
64816
64817 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64818 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
64819
64820commit eb031d4013d36f6aef4aba45840762ae8635cc13
64821Author: Adam Jackson <ajax@redhat.com>
64822Date: Wed Aug 19 15:42:34 2009 -0400
64823
64824 linux: hand-roll a backtrace printer instead of using backtrace_symbols
64825
64826 Why? Because backtrace_symbols calls malloc, which you can't do from a
64827 signal handler. Face? Palm.
64828
64829commit 75b9383d8a4c113ab3c6cfc1d5efcb5d9982a1bf
64830Author: Jesse Adkins <jesse_adkins7@yahoo.com>
64831Date: Tue Aug 18 21:13:40 2009 -0700
64832
64833 xserver doesn't stop all connections to localhost
64834
64835 X.Org Bugzilla #23329: http://bugs.freedesktop.org/show_bug.cgi?id=23329
64836 Patch #28648: http://bugs.freedesktop.org/attachment.cgi?id=28648
64837
64838 I noticed in xserver at os/access.c that xorg tries to stop connections
64839 to localhost by checking against the address 127.0.0.1. However, RFC
64840 3330 defines the localhost network as 127.0.0.0/8. This means that any
64841 IPv4 address that starts with 127 is just another name for localhost.
64842
64843commit 2d34eace13c2016048c627c4e96c3b2399901078
64844Author: Alan Coopersmith <alan.coopersmith@sun.com>
64845Date: Tue Aug 18 20:11:23 2009 -0700
64846
64847 Sun bug 6872917: Xorg not querying /dev/fb when no xorg.conf exists
64848
64849 commit 48ee5558333bd324463b6994735cabb23de262ec (OpenSolaris VT support)
64850 broke the autoconfiguration code in xf86AutoConfig.c that uses the
64851 Solaris-specific VIS_GETIDENTIFIER ioctl on a frame buffer device like
64852 /dev/fb by changing xf86Info.consoleFd from /dev/fb to a /dev/vt/*
64853 device.
64854
64855 This fixes it by reworking the code to split the console device
64856 (/dev/vt/*, the vtXX CLI option) from the frame buffer device
64857 (/dev/fb, -dev option) to allow both VT and autoconfig to work.
64858
64859 It also fixes the console device to use /dev/fb when VT's are not
64860 supported instead of throwing a Fatal Error because it can't open
64861 /dev/vt/0.
64862
64863 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
64864
64865commit d225230148b3cdab3b32fce3abee26b818d4bab0
64866Author: Jesse Barnes <jbarnes@virtuousgeek.org>
64867Date: Tue Aug 18 12:15:32 2009 -0700
64868
64869 xfree86/linux: don't change VT perms unless we're running as root
64870
64871 In non-setuid root installations, we shouldn't try to adjust VT/tty
64872 ownership. It will fail, and shouldn't be necessary anyway (since
64873 startup scripts or PAM should be handling perms for us in that case).
64874
64875 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
64876
64877commit 41884b8f4dfe4d931c64a6046adfefcc2b7646c2
64878Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64879Date: Tue Aug 18 15:47:56 2009 +0300
64880
64881 xfree86: remove unused code (xf86Configure.c)
64882
64883 The xorg.conf generator was not assigning correctly the primary device
64884 ("bootable") as screen zero. So just skip this kind of routines for now.
64885
64886 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64887
64888commit 097376266f256e3ae37a16b343ed1e3441a8f003
64889Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64890Date: Tue Aug 18 13:13:57 2009 +0300
64891
64892 xfree86: remove unaffected code (xf86Configure.c)
64893
64894 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64895
64896commit dd8960a9f37fbef65cb99e058c3772edfda7b6db
64897Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64898Date: Tue Aug 18 12:53:47 2009 +0300
64899
64900 xfree86: remove stupid macro (xf86Configure.c)
64901
64902 It was confusing more instead helping.
64903
64904 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64905
64906commit 4ec7667bb17839bd4ec11091c75ecfff317fc96c
64907Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64908Date: Tue Aug 18 12:33:35 2009 +0300
64909
64910 xfree86: remove wrong commentary
64911
64912 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64913
64914commit 2e8c2b8ae6a092871838b7cceaaf4a7590bce1d3
64915Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64916Date: Tue Aug 18 12:28:15 2009 +0300
64917
64918 xfree86: header clean up (xf86Configure.c)
64919
64920 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64921
64922commit 2df414a33321ae772af1e7e52c36aec222bb3c8f
64923Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64924Date: Tue Aug 18 12:14:27 2009 +0300
64925
64926 xfree86: reorganize xf86Configure.c
64927
64928 No semantical changes. Only code moved around.
64929
64930 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64931
64932commit 113e66a1ffe2a43c4d1c92a7d78ca43b4db58822
64933Author: Tiago Vignatti <tiago.vignatti@nokia.com>
64934Date: Sun Aug 16 21:02:11 2009 +0300
64935
64936 xfree86: remove unused PCI header
64937
64938 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
64939
64940commit 6b5978dcf1f7ac3ecc2f22df06f7000f360e2066
64941Author: Richard Hughes <richard@hughsie.com>
64942Date: Mon Aug 17 09:15:32 2009 +0100
64943
64944 Do not reset lastDeviceEventTime when we do dixSaveScreens
64945
64946 When we turn off DPMS with DPMSModeOff and do dixSaveScreens, don't reset the
64947 event time else session clients using IDLETIME will be reset.
64948
64949 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
64950
64951commit 4098ad72d00e65d142fb9fe8a1194be35338508b
64952Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64953Date: Sun Aug 16 00:06:10 2009 -0400
64954
64955 xace: fix access mode in dixLookupWindow within several RandR calls.
64956
64957 Referencing a screen using a window only requires GetAttr access.
64958
64959 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64960
64961commit d4b8f7602b5e266a0ebd3b1ba23724362cc7de3a
64962Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64963Date: Mon Aug 10 18:39:04 2009 -0400
64964
64965 xace: fix access mode in dixLookupWindow call within ProcUngrabKey.
64966
64967 Referencing a window (as grab-window) only requires GetAttr access.
64968
64969 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64970
64971commit 6d2f4e487869f10de4a62365b4d6de036c752ab8
64972Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64973Date: Mon Aug 10 18:36:53 2009 -0400
64974
64975 Add XI2 requests to protocol.txt
64976
64977 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64978
64979commit 73975ef3a39ce522c6206ca800ed175fbf851dcf
64980Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64981Date: Mon Aug 10 18:34:07 2009 -0400
64982
64983 xselinux: Allow per-client device create contexts.
64984
64985 The previous behavior was to set the serverClient's value which was used globally.
64986 This is in support of XI2, where clients can create device pairs directly.
64987
64988 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64989
64990commit 66e32d252cffcd4fe7d505f1c211253f23c5002c
64991Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
64992Date: Mon Aug 10 18:33:05 2009 -0400
64993
64994 xselinux: Add more new device permissions for XI2.
64995
64996 Reflects the ability of clients to create/destroy device objects.
64997
64998 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
64999
65000commit 1f4ea22a20fdd30412a259a87eee133155e1163d
65001Author: Rémi Cardona <remi@gentoo.org>
65002Date: Mon Aug 17 11:33:24 2009 +0200
65003
65004 configure: fix help string after Xnest default build was changed
65005
65006 Signed-off-by: Rémi Cardona <remi@gentoo.org>
65007
65008commit 7e37e7c50e4fdcb53296a99d90af3d90081ce9bb
65009Author: Peter Hutterer <peter.hutterer@who-t.net>
65010Date: Tue Aug 11 14:39:50 2009 +1000
65011
65012 test: add protocol tests for DeviceChangedEvents
65013
65014 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65015
65016commit 2851f04cb2f6e5c30267f733d867c86d4e69a485
65017Author: Peter Hutterer <peter.hutterer@who-t.net>
65018Date: Tue Aug 11 16:27:40 2009 +1000
65019
65020 dix: rework DeviceChangedEvents a bit.
65021
65022 DCEs are now processed when sent throught the master device, not when sent
65023 through the slave device. This includes a removal of some un-used (or partly
65024 used) fields in the DCE itself to something more self-explanatory.
65025
65026 TODO: if a device has events queued and its attachment is changed, the DCE
65027 is silently dropped now. Instead, it should be generated as soon as the
65028 first event after the attachment is sent.
65029
65030 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65031
65032commit c1d901d723c3bee523736eacc15b44a7dff484fe
65033Author: Richard Hughes <richard@hughsie.com>
65034Date: Fri Aug 14 11:44:35 2009 +0100
65035
65036 Don't reset the lastDeviceEventTime when doing DPMS actions
65037
65038 When we change the DPMS mode, don't play games with the last event time as
65039 this breaks applications using IDLETIME to turn the backlight off after a
65040 preset time.
65041
65042 This patch fixes gnome-power-manager and xfce-power-manager
65043
65044 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65045
65046commit dca4de72d375c4b9bcdd25b151e291a77a5b06cc
65047Author: Peter Hutterer <peter.hutterer@who-t.net>
65048Date: Sat Aug 15 20:17:20 2009 +1000
65049
65050 dix: fix potential use of unused variable 'mask'.
65051
65052 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65053
65054commit 978b65bce14598f2d42ca0177ea58fef71fc12c5
65055Author: Peter Hutterer <peter.hutterer@who-t.net>
65056Date: Fri Aug 14 15:05:42 2009 +1000
65057
65058 include: XInputExtensionInit doesn't need to be exported.
65059
65060 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65061
65062commit 3989dc1d34f116f30915632cc5286937392e180a
65063Author: Peter Hutterer <peter.hutterer@who-t.net>
65064Date: Tue Aug 11 16:20:20 2009 +1000
65065
65066 dix: use the XI2 defines for class types.
65067
65068 Doesn't matter really, they have the same values anyway.
65069
65070 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65071
65072commit fe045820f1fb33991e8bff5c6e192097caa85727
65073Author: Peter Hutterer <peter.hutterer@who-t.net>
65074Date: Mon Aug 10 15:34:00 2009 +1000
65075
65076 input: move DeviceChangedEvent conversion into eventconvert.c
65077
65078 The version in eventconvert.c was half broken and for some reason we ended
65079 up with a second version in exevents.c (which works). Move it over to where
65080 it belongs and call EventToXI2 instad of having a custom function for it.
65081
65082 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65083
65084commit 33eb6f70816921abc5da62e434f40e78d672274e
65085Author: Peter Hutterer <peter.hutterer@who-t.net>
65086Date: Fri Aug 14 13:49:58 2009 +1000
65087
65088 include: Unexport most symbols from exevents.h.
65089
65090 And shuffle them around so that the part used by drivers is up the top and
65091 commented.
65092
65093 Also, woo, the sdksyms script doesn't like declarations with return type and
65094 function name on the same line...
65095
65096 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65097
65098commit 1373c2ea877b562076a4a3c97e8e4f999c977217
65099Author: Dave Airlie <airlied@redhat.com>
65100Date: Sat Aug 15 12:14:26 2009 +1000
65101
65102 exa: clarify createpixmap2 new pitch return
65103
65104commit 48ee5558333bd324463b6994735cabb23de262ec
65105Author: Aaron Zang <Aaron.Zang@Sun.COM>
65106Date: Mon Aug 3 23:21:39 2009 -0700
65107
65108 Add new VT support for OpenSolaris & future Solaris releases
65109
65110 Signed-off-by: Aaron Zang <Aaron.Zang@Sun.COM>
65111 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
65112
65113commit 613e76ff9055d8ac2b1af1130668180646a9e14c
65114Author: Alan Coopersmith <alan.coopersmith@sun.com>
65115Date: Mon Aug 3 21:38:51 2009 -0700
65116
65117 Remove support for Solaris x86 releases older than Solaris 8
65118
65119 If you want to run a pre-1999 kernel, you'll need a pre-2009 X server
65120
65121 [Some pre-Solaris 8 VT support is left by this patch to allow reuse by
65122 the new Solaris VT support that follows in the next patch.]
65123
65124 Signed-off-by: Aaron Zang <Aaron.Zang@Sun.COM>
65125 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
65126
65127commit 6c292d17053eb2a7e7054e51210f423dbc0cb7e8
65128Author: Peter Hutterer <peter.hutterer@who-t.net>
65129Date: Fri Aug 14 11:48:54 2009 +1000
65130
65131 dix: update the sprite trace for all masters && floating slaves (#23257)
65132
65133 When the windows are restructured, CheckMotion needs to be called for all
65134 masters and floating slaves to update the spriteTrace.
65135
65136 X.Org Bug 23257 <http://bugs.freedesktop.org/show_bug.cgi?id=23257>
65137
65138 Tested-by: Thomas Jaeger
65139 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65140
65141commit 49046088f10cceaea7da97401d742d3fb59371f5
65142Author: Peter Hutterer <peter.hutterer@who-t.net>
65143Date: Fri Aug 14 09:48:45 2009 +1000
65144
65145 config: don't shutdown the libhal ctx if it failed to initialize (#23213)
65146
65147 Regression introduced by b1c3dc6ae226db178420e3b5f297b94afc87c94c.
65148 Shutting down the libhal_ctx if the init failed may cause an abort.
65149 This can happen if hald is not yet running at server startup.
65150
65151 X.Org Bug 23213 <http://bugs.freedesktop.org/show_bug.cgi?id=23213>
65152
65153 Tested-by: Stefan Dirsch
65154 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65155
65156commit 1545a120df6dffb5b84fe96c5a992357520b7c8d
65157Author: Dave Airlie <airlied@redhat.com>
65158Date: Tue Aug 11 15:00:36 2009 +1000
65159
65160 exa: fix CreatePixmap2 to be useful for tiling.
65161
65162 This adds a pitch return so that the driver can align the pitch to any
65163 value it wishes and not just the one it gave to EXA at startup.
65164
65165commit db568f9eabf3450d8a023597ff007df355b13ea8
65166Author: Peter Hutterer <peter.hutterer@who-t.net>
65167Date: Thu Aug 13 16:37:00 2009 +1000
65168
65169 Xext: fix up wrong conditions for negative sync transitions.
65170
65171 If the counter had a value higher than the trigger value for a negative
65172 transition, the trigger value did not get set.
65173
65174 The correct sequence of checks is:
65175 if (positive transition)
65176 if (counter value < trigger value)
65177 set up trigger
65178 if (negative transition)
65179 if (counter value > trigger value)
65180 set up trigger
65181
65182 Red Hat Bug 501601 <https://bugzilla.redhat.com/show_bug.cgi?id=501601>
65183
65184 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65185
65186commit a3e50b05747cab92090ae7d7f4475cd61d3fcadf
65187Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
65188Date: Mon Jul 27 20:58:44 2009 +0200
65189
65190 Add 4 missing 10bpc picture formats to the server format list.
65191
65192 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
65193
65194commit 36e24a6d93bd5aced4e566b80bf2d03555fab9ca
65195Author: Peter Hutterer <peter.hutterer@who-t.net>
65196Date: Thu Aug 13 10:50:26 2009 +1000
65197
65198 Xext: add missing return code check to ProcSyncDestroyAlarm
65199
65200 Introduced with 57aff88c7d0761e590806d07bee1c9410680c89f.
65201
65202 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65203
65204commit ce69a06aff934b2dcded8606cab079ac6465007c
65205Author: Peter Hutterer <peter.hutterer@who-t.net>
65206Date: Tue Aug 11 16:16:54 2009 +1000
65207
65208 Xi: fix up broken DeviceChangedEvent swapping code
65209
65210commit b44d34d5fd0d8aaacb89121e8b4afba04f1dcc80
65211Author: Peter Hutterer <peter.hutterer@who-t.net>
65212Date: Mon Aug 10 15:32:59 2009 +1000
65213
65214 include: correct a copy/paste error in a comment.
65215
65216 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65217
65218commit c477ac8a151a28ba23f5c87e8f7affd6e82cfaa9
65219Author: Peter Hutterer <peter.hutterer@who-t.net>
65220Date: Mon Aug 10 15:15:02 2009 +1000
65221
65222 dix: re-name internal eventToClassesChanged to eventToDeviceChanged
65223
65224 DeviceClassesChangedEvents (where this name comes from) have been replaced
65225 with DeviceChangedEvents.
65226
65227 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65228
65229commit af38f6790c4c8ba1e70f5c5ba3530ded85d6e372
65230Author: Peter Hutterer <peter.hutterer@who-t.net>
65231Date: Mon Aug 10 14:08:59 2009 +1000
65232
65233 test: add event conversion tests for XIDeviceEvents
65234
65235commit 3f2e4b9867b6877ee7be32b151fcaf221ef0812f
65236Author: Peter Hutterer <peter.hutterer@who-t.net>
65237Date: Mon Aug 10 14:07:54 2009 +1000
65238
65239 Xi: add event swapping for XIRawEvents.
65240
65241 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65242
65243commit f3b2f9fb734ecfff6db9ae85b0d247856ede8112
65244Author: Peter Hutterer <peter.hutterer@who-t.net>
65245Date: Mon Aug 10 13:12:40 2009 +1000
65246
65247 Xi: fix event swapping for XIDeviceEvents.
65248
65249 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65250
65251commit e46f02fa2de79261221b42ab73f9daa2ce8ac650
65252Author: Peter Hutterer <peter.hutterer@who-t.net>
65253Date: Mon Aug 10 09:50:16 2009 +1000
65254
65255 Xext: allocate a separate event list for XTest events (#23100)
65256
65257 XTest event processing may be interrupted by a SIGIO. If Xtest uses the same
65258 event list as the rest of the server, this list may be overwritten
65259 in-flight.
65260
65261 X.Org Bug 23100 <http://bugs.freedesktop.org/show_bug.cgi?id=23100>
65262
65263 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65264
65265commit 8483b08152104dc619d9e44f877d2687d866355e
65266Author: Peter Hutterer <peter.hutterer@who-t.net>
65267Date: Fri Aug 7 16:11:37 2009 +1000
65268
65269 test: add focus and enter conversion testing.
65270
65271 Doesn't actually convert anything, but verify the expected behaviour.
65272
65273commit 59dc59a72ffd4cbc4df207bc688c92bb4863e8a9
65274Author: Peter Hutterer <peter.hutterer@who-t.net>
65275Date: Fri Aug 7 11:23:33 2009 +1000
65276
65277 test: add XI2 eventconversion test for raw events.
65278
65279 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65280
65281commit 10d7948e0360860e1e9633dca39f646d492e73bf
65282Author: Peter Hutterer <peter.hutterer@who-t.net>
65283Date: Fri Aug 7 10:17:14 2009 +1000
65284
65285 test: fix build error introduced by new AllocDevicePair API
65286
65287 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65288
65289commit 3f161a0aac39fbdeef393a17269486b8dace4672
65290Author: Peter Hutterer <peter.hutterer@who-t.net>
65291Date: Mon Aug 10 14:14:35 2009 +1000
65292
65293 Xi: un-statify XI2EventSwap, it is needed for tests.
65294
65295 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65296
65297commit 55e1ea08d03d89ecc0f2db7652a4d15567204696
65298Author: Peter Hutterer <peter.hutterer@who-t.net>
65299Date: Mon Aug 3 16:46:11 2009 +1000
65300
65301 Xi: get device changed data from DeviceChangedEvents, not the device (#23100)
65302
65303 If a new device posts an event while the DCE is in the queue, getting the
65304 data from the device may result in invalid memory access.
65305
65306 X.Org Bug 23100 <http://bugs.freedesktop.org/show_bug.cgi?id=23100>
65307
65308commit 8ce4fde76fdf747fd3e00836c9c9c542e516ae9a
65309Author: Matt Turner <mattst88@gmail.com>
65310Date: Wed Aug 12 14:45:13 2009 -0400
65311
65312 Fix build on Alpha
65313
65314commit a400dbb38f93030d51afe806b4b20d5ef501c855
65315Author: Colin Harrison <colin.harrison@virgin.net>
65316Date: Wed Aug 12 17:48:36 2009 +0100
65317
65318 Xming: Add NET_WM_ICON to native icon conversion
65319
65320 fd.o bugzilla #4491
65321 originally from a patch by Joe Krahn <jkrahn@nc.rr.com>
65322
65323 Convert a NET_WM_ICON to a native icon by converting to a native
65324 bitmap and then using CreateIconIndirect()
65325
65326 Don't use icon alpha on Windows 2000 or if display isn't 32-bit, convert
65327 alpha channel to a 1-bit transparency mask using a threshold value
65328
65329 Fix warning in winScaleXBitmapToWindows() about signedness of *iconData
65330
65331 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
65332
65333commit 211511f150ce189681aff176c311fa8f312517e9
65334Author: Soren Sandmann Pedersen <ssp@redhat.com>
65335Date: Tue Aug 11 14:40:42 2009 -0400
65336
65337 Disable the out-of-bounds workaround in pixman.
65338
65339 This workaround was necessary for older X servers, but now fb should
65340 be using correct coordinates.
65341
65342 Also bump pixman requirement to 0.15.20.
65343
65344commit 93d9646c713336e03c135204c061f561d3654e23
65345Author: Tiago Vignatti <tiago.vignatti@nokia.com>
65346Date: Sat Aug 8 20:18:10 2009 +0300
65347
65348 mi: fix indentation
65349
65350 mieq.c looks indented-wise now. Let's see how long it will take to someone
65351 mess it again.
65352
65353 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
65354
65355commit f9a2fff2248d7254958857677cabfea914ed4853
65356Author: Tiago Vignatti <tiago.vignatti@nokia.com>
65357Date: Wed Aug 5 21:02:29 2009 +0300
65358
65359 mi: fix cursor warping screens
65360
65361 The server was processing ET_RawMotion type when the cursor was wrapping to
65362 another screen and getting wrong valuator values. This fix such issue
65363 considering only ET_Motion, ET_KeyPress, ET_KeyRelease, ET_ButtonPress and
65364 ET_ButtonRelease types when the cursor detects a new screen, keeping the
65365 "normal" processing of device events.
65366
65367 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
65368 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65369
65370commit d573cc46d3106824902ab4f926617bd9959af57c
65371Author: Maarten Maathuis <madman2003@gmail.com>
65372Date: Sat Aug 8 11:32:05 2009 +0200
65373
65374 exa: more safety
65375
65376commit 8b652435cd42929e2d187b353b3b20e798569356
65377Author: Maarten Maathuis <madman2003@gmail.com>
65378Date: Sat Aug 8 10:35:01 2009 +0200
65379
65380 exa: minor cleanup
65381
65382commit a73f95aa93634cf7ba2c7c9274a046b2df2adca7
65383Author: Maarten Maathuis <madman2003@gmail.com>
65384Date: Sat Aug 8 02:31:23 2009 +0200
65385
65386 exa: Fix the broken upload fallback for "mixed"
65387
65388 - Replace it with something wfb friendly while i'm at it.
65389
65390commit c029678bf7883b91f3c8095e764abbee246c9bb6
65391Author: Julien Cristau <jcristau@debian.org>
65392Date: Fri Aug 7 20:42:30 2009 +0200
65393
65394 configure.ac: drop dependency on fontenc
65395
65396 Nothing in the server uses this anymore.
65397
65398commit 3943df7f76c1b4930fb7370d9d145cee96dd562b
65399Author: Maarten Maathuis <madman2003@gmail.com>
65400Date: Fri Aug 7 20:36:03 2009 +0200
65401
65402 exa: one can never be too careful
65403
65404commit 1548e8ae5d28ed0eb6057a1a19cfc84a78ef34dc
65405Author: Maarten Maathuis <madman2003@gmail.com>
65406Date: Fri Aug 7 20:04:53 2009 +0200
65407
65408 exa: fix a potential loophole in "mixed"
65409
65410 - Always free sys_ptr before setting the pixmap to pinned.
65411
65412commit e94c7c42ce9d6a194b76ba7e8b3904b180f93a1b
65413Author: Dave Airlie <airlied@redhat.com>
65414Date: Fri Aug 7 13:36:52 2009 +1000
65415
65416 parser: make libxf86config_internal.la not installed.
65417
65418commit 5fb188b547e0ef4e2f2c59b66a3183b004e4a1e9
65419Author: Dave Airlie <airlied@redhat.com>
65420Date: Fri Aug 7 12:05:51 2009 +1000
65421
65422 ddx: fix xf86Config.a generation
65423
65424 We were generating a shared library, but this lib is foobar, the parser
65425 requires some symbols from the X server or from the program its being linked
65426 into. If the program its being linked into (say a python .so) has symbol
65427 visibility enabled then it will fail to dynamic link, also if this .so has
65428 symbol visiblity enabled it will fail to dynamic link.
65429
65430 Screw it go back to a .a file really unless someone cleans it up properly.
65431
65432 Signed-off-by: Dave Airlie <airlied@redhat.com>
65433
65434commit 3047bd067464efb9857960d3fa6324b947faa970
65435Author: Maarten Maathuis <madman2003@gmail.com>
65436Date: Wed Aug 5 18:39:47 2009 +0200
65437
65438 exa: delay malloc for "mixed"
65439
65440commit e8ac2ed5dc4c2ac0a5e1e1f371f94c15b1c729dd
65441Author: Maarten Maathuis <madman2003@gmail.com>
65442Date: Wed Aug 5 16:12:16 2009 +0200
65443
65444 exa: implement exaMoveInPixmap for "mixed"
65445
65446 - This can be used to force creation of driver pixmap.
65447 - Not for 1 or 4 bpp.
65448 - Driver can still fail (driver) pixmap creation.
65449
65450commit 9d2a7128d3e66b8c076a714d69f84bcad49391b9
65451Author: Maarten Maathuis <madman2003@gmail.com>
65452Date: Sun Aug 2 02:35:46 2009 +0200
65453
65454 exa: Use damage to optimise away useless copies.
65455
65456commit 03ecb164f2592c954aa408bf121e0c67b604d854
65457Author: Maarten Maathuis <madman2003@gmail.com>
65458Date: Sat Aug 1 19:19:19 2009 +0200
65459
65460 exa: A simple 3rd backend implementation.
65461
65462 - Based on driver pixmaps with some changes (completely transparent to driver).
65463 - It helps with the problem of known software fallbacks, such as trapezoids.
65464 - exaDoMigration is now called for all cases that provide a do_migration hook.
65465 - exa_migration.c is renamed to exa_migration_classic.c
65466
65467commit ac7ac913fd98ea359c05c89968ab53a3223615b4
65468Author: Maarten Maathuis <madman2003@gmail.com>
65469Date: Wed Jul 22 21:35:41 2009 +0200
65470
65471 exa: Split out some classic and driver allocated pixmap code into seperate files
65472
65473 - Create a few seperate functions and a few private function pointers.
65474 - Replace a few if conditions with a check for pExaPix->pDamage instead.
65475 - This is in preperation of a third scheme that lies somewhere in between.
65476 - Code clarity would have suffered (i started working on it and didn't like the mess).
65477
65478commit 3abbd327f4a732408119de1f8e9ecba4812772a5
65479Author: Peter Hutterer <peter.hutterer@who-t.net>
65480Date: Wed Aug 5 15:47:46 2009 +1000
65481
65482 dix: detach SD during XI2 grabs only.
65483
65484 XI1 grabs on slave devices leave the device attached - just like in earlier
65485 versions of XI.
65486
65487 Tested-by: Thomas Jaeger
65488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65489
65490commit c29aa7da220661532b05972cacd3dbaff29408b5
65491Author: Peter Hutterer <peter.hutterer@who-t.net>
65492Date: Wed Aug 5 15:33:07 2009 +1000
65493
65494 Revert "dix: Remove temporary detachment of slave devices."
65495
65496 This reverts commit 0c0ef42292f4c910c73b308cd75d77637312da53.
65497
65498 Tested-by: Thomas Jaeger
65499 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65500
65501commit 931160fcf378120e58849d801dbc62bed6e65a4b
65502Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65503Date: Tue Aug 4 02:53:17 2009 -0700
65504
65505 XQuartz: AIGLX: Provide empty __glXAquaDrawableResize to avoid crashing in DoMakeCurrent
65506 (cherry picked from commit 60a1d2c2764f2f02c0751940a264588717afce79)
65507
65508commit 0fe639a314de397516cf00d8100da8086abdd97a
65509Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65510Date: Sat Aug 1 22:53:45 2009 -0700
65511
65512 XQuartz: Only save lastpt on mouse/tablet events
65513 (cherry picked from commit 552be074e5be6492df3e290e8b7d9daff1a2cb34)
65514
65515commit 491c2d74c12d7f9bc9f31b427ece04e049ebff5a
65516Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65517Date: Sat Aug 1 12:43:14 2009 -0700
65518
65519 XQuartz: Use mouseLocation rather than locationInWindow when setting lastpt
65520
65521 I don't understand the *why* ... I just see that it works better this way for games like Quake2 through wine. It *should* be better the other way, but somehow it's not.
65522
65523 I guess this will go in my list of puzzles to unravel.
65524 (cherry picked from commit 65ae2d00e1a53f97f2ff9522406ab69d50bf3199)
65525
65526commit df6c01352471c766e9f71ceac03cab8c3911faa1
65527Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65528Date: Sat Aug 1 02:32:52 2009 -0700
65529
65530 XQuartz: Don't use location delta for tablets since NSEvent does not give a precise delta.
65531 (cherry picked from commit 6c5bf756a7f5389cdfe2e43a339d7c31a3e522e9)
65532
65533commit e360104880e6e2e666aa05dfd56e2ef3880f38ef
65534Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65535Date: Wed Aug 5 16:17:01 2009 -0700
65536
65537 XQuartz: Purge redundant QuartzBell
65538 (cherry picked from commit de14a63d20095e1537fd74352850c734d900031d)
65539
65540commit 8c0085c715effdc450d78eec14bc32e6214c78af
65541Author: Tiago Vignatti <tiago.vignatti@nokia.com>
65542Date: Wed Aug 5 22:50:44 2009 +0300
65543
65544 xfree86: remove more RAC junky
65545
65546 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
65547
65548commit 444723273ae82fc3f6707ed1a461aaaa8fdae39c
65549Author: Tiago Vignatti <tiago.vignatti@nokia.com>
65550Date: Wed Aug 5 22:48:02 2009 +0300
65551
65552 xfree86: reorganize pci code
65553
65554 - xf86PciVideoInfo doesn't need to be global
65555 - remove unused macros
65556
65557 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
65558
65559commit 20169414e1afd5d1d02cb1b57866b1c158b2fc6c
65560Author: Tiago Vignatti <tiago.vignatti@nokia.com>
65561Date: Wed Aug 5 22:44:19 2009 +0300
65562
65563 xfree86: remove unused functions
65564
65565 RAC trash.
65566
65567 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
65568
65569commit 95b678e6dc41f2524ada4eb11289687fafce7588
65570Author: Alan Coopersmith <alan.coopersmith@sun.com>
65571Date: Mon Aug 3 23:49:56 2009 -0700
65572
65573 Correct modifier map built when ProcSetModifierMapping is called
65574
65575 Fixes xmodmap changes to modifiers to stop corrupting modifier maps
65576
65577 Previous code had two bugs:
65578 - the code to increment mod was after the code to continue if no
65579 modifier was set, so mod wouldn't be incremented for modifiers
65580 with no keys mapped to them (such as if you called
65581 xmodmap -e 'clear Lock')
65582 - the value it set in the modifier map was the raw modifier number,
65583 not the bitmask value for that modifier
65584
65585 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
65586 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65587
65588commit 4ca305956e5ea6f606b22ef62aa462186a7b95f0
65589Author: Peter Hutterer <peter.hutterer@who-t.net>
65590Date: Wed Aug 5 09:40:56 2009 +1000
65591
65592 Re-enable Xnest by default.
65593
65594 Xnest was disabled in 82fc102568b3d6b0daeb6c5f5b3a1310a7f14fcd due do build
65595 errors. These errors have since been fixed.
65596
65597 Re-enable Xnest by default to increase coverage by default builds.
65598
65599 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65600
65601commit 5904ef2ccd6056b187ca76f104c21e2d686bfc1d
65602Author: Peter Hutterer <peter.hutterer@who-t.net>
65603Date: Wed Aug 5 10:40:20 2009 +1000
65604
65605 xnest: restore xnestUpdateModifierState
65606
65607 The meat of xnestUpdateModifierState was ifdef'd out in
65608 6ef46c40e62def4841a4cff4e0b443516a2ed782. This resulted in stuck modifiers
65609 when a modifier key release event wasn't sent to Xnest (e.g. Alt-Tab away).
65610
65611 See X.Org Bug 3664 <https://bugs.freedesktop.org/show_bug.cgi?id=3664> for
65612 the original bug report.
65613
65614 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65615
65616commit 44afc7577ad3339f4555438388f6a4d8e220c991
65617Author: Peter Hutterer <peter.hutterer@who-t.net>
65618Date: Wed Aug 5 09:51:05 2009 +1000
65619
65620 xnest: include exevents.h for XIGetKnownProperty.
65621
65622 Silences compiler warning.
65623 Pointer.c: In function ‘xnestPointerProc’:
65624 Pointer.c:64: warning: implicit declaration of function ‘XIGetKnownProperty’
65625 Pointer.c:64: warning: nested extern declaration of ‘XIGetKnownProperty’
65626
65627 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65628
65629commit 32ce5c83a504d2cafd2b19ed628b40032ef347b3
65630Author: Peter Hutterer <peter.hutterer@who-t.net>
65631Date: Wed Aug 5 09:39:47 2009 +1000
65632
65633 xnest: use AllocDevicePair for Xnest device initalization
65634
65635 All Xnest needs is a single pointer+keyboard pair. AllocDevicePair sets them
65636 up nicely with the name assigned etc.
65637
65638 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65639
65640commit 26b83ad4a29dc180b336a19d9e97589814e93e37
65641Author: Peter Hutterer <peter.hutterer@who-t.net>
65642Date: Wed Aug 5 09:32:50 2009 +1000
65643
65644 dix: require PointerProc and KeyboardProc to be passed into AllocDevicePair.
65645
65646 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65647
65648commit 6a500fdd4d717947b86f6d6844ebf9a8603eb8e2
65649Author: Peter Hutterer <peter.hutterer@who-t.net>
65650Date: Tue Aug 4 15:29:55 2009 +1000
65651
65652 dmx: include exglobals.h for DeviceKeyPress and friends.
65653
65654 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65655
65656commit 591c2f8758ef36a9bbf1625f9c207a1804f58ef4
65657Author: Peter Hutterer <peter.hutterer@who-t.net>
65658Date: Tue Aug 4 15:30:52 2009 +1000
65659
65660 dmx: typecast to InternalEvent* before mieqEnqueue.
65661
65662 Silences a few compiler warnings.
65663
65664 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65665
65666commit 15b425b5a6630d2ac95b490f4f4dbb9240ac9c5d
65667Author: Peter Hutterer <peter.hutterer@who-t.net>
65668Date: Tue Aug 4 15:31:42 2009 +1000
65669
65670 dmx: move 'state' around to silence compiler warning.
65671
65672 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65673
65674commit e10f802c8260e034fd6c835d3f2622d9cd5085f8
65675Author: Peter Hutterer <peter.hutterer@who-t.net>
65676Date: Tue Aug 4 15:40:17 2009 +1000
65677
65678 dmx: include inputstr.h for inputInfo.pointer
65679
65680 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65681
65682commit 9ffc3f280b3907a672f99101548a5eb52027b37f
65683Author: Peter Hutterer <peter.hutterer@who-t.net>
65684Date: Tue Aug 4 15:40:31 2009 +1000
65685
65686 dmx: don't call InitKbdFeedbackClassDeviceStruct - doesn't exist anymore.
65687
65688 InitKeyboardDeviceStruct does the same task now.
65689
65690 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65691
65692commit 63155cf985ee15bf6aad95066b076ab680cf5a31
65693Author: Peter Hutterer <peter.hutterer@who-t.net>
65694Date: Tue Aug 4 15:28:30 2009 +1000
65695
65696 dmx: pass button/axis labels into the init functions (currently unset)
65697
65698 Currently only None labels are passed in, in the future these labels should
65699 be whatever the respective buttions/axes are.
65700
65701 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65702
65703commit af85973a0a80fb0ba82330ae3ec897f65d618003
65704Author: Peter Hutterer <peter.hutterer@who-t.net>
65705Date: Tue Aug 4 15:25:06 2009 +1000
65706
65707 dmx: remove now-useless defines.
65708
65709 XI is mandatory so we can expect both to be defined.
65710
65711commit 654d7a02600bbafd890f555a081aa0a2fcb5e253
65712Author: Peter Hutterer <peter.hutterer@who-t.net>
65713Date: Tue Aug 4 15:23:01 2009 +1000
65714
65715 Revert "dmx: claim we support XI 2."
65716
65717 XQueryInputVersion doesn't exist anymore and for now XI 1 support is good
65718 enough.
65719
65720 This reverts commit 8da8a0fec4b1b9d9208635dedb2f449dc99e0004.
65721
65722commit e7c950c6273ff93f25c091c93d86da6e332a2277
65723Author: Peter Hutterer <peter.hutterer@who-t.net>
65724Date: Tue Aug 4 15:15:24 2009 +1000
65725
65726 dmx: fix a call to XkbSetRulesDflts, takes XkbRMLVOSet now.
65727
65728commit 2cde9208ff756e33d162e2324f4b99540230d743
65729Author: Peter Hutterer <peter.hutterer@who-t.net>
65730Date: Tue Aug 4 14:58:25 2009 +1000
65731
65732 dmx: Enable/DisableDevice take a boolean variable now.
65733
65734 We want to send events here, so pass in TRUE.
65735
65736 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65737
65738commit 1e210d6d10ec63d22247e5801890e024826fc861
65739Author: Peter Hutterer <peter.hutterer@who-t.net>
65740Date: Tue Aug 4 14:41:46 2009 +1000
65741
65742 xkb: remove now-unused XkbGetKeysym.
65743
65744 XkbGetKeysyms was only used by the now-removed Keysym grabs.
65745
65746 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65747
65748commit a6ce6c70cff5108f0751b662b8e52c83daab0722
65749Author: Michel Dänzer <daenzer@vmware.com>
65750Date: Tue Aug 4 23:23:21 2009 +0200
65751
65752 EXA: Simplify exaGetPixmapFirstPixel using GetImage.
65753
65754commit 842373104d08d47efc863cecbe30431d3faebef1
65755Author: Michel Dänzer <daenzer@vmware.com>
65756Date: Tue Aug 4 23:23:21 2009 +0200
65757
65758 Add support for RENDER BGRA formats.
65759
65760commit 0bf7eaf3052ce24066b0a7c14860b4762fb81364
65761Author: Alan Coopersmith <alan.coopersmith@sun.com>
65762Date: Mon Aug 3 23:14:50 2009 -0700
65763
65764 Ansify function arguments in VTsw_noop.c
65765
65766 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
65767
65768commit 736f2d64725c6df8413e627bd40ce7ecb011acc7
65769Author: Paul Bender <pebender@gmail.com>
65770Date: Tue Jun 30 15:50:46 2009 -0800
65771
65772 Bug 16832: XDMCP related build error when --disable-xdmcp is used
65773
65774 X.Org Bugzilla #16832: http://bugs.freedesktop.org/show_bug.cgi?id=16832
65775 Patch #27279: http://bugs.freedesktop.org/attachment.cgi?id=27279
65776
65777commit b159a98f7fbdb3f8b337cd1bfd2ff27312856619
65778Author: Lee Leahu <freedesktop-bugs@dyweni.com>
65779Date: Thu Oct 23 21:03:57 2008 -0500
65780
65781 seg fault when initializing DMX screens
65782
65783 X.Org Bugzilla #18086: http://bugs.freedesktop.org/show_bug.cgi?id=18086
65784 Patch #19837: http://bugs.freedesktop.org/attachment.cgi?id=19837
65785
65786commit 5da9b255a64bec7dbf5ddb392d54dac9be5b43c0
65787Author: Alan Coopersmith <alan.coopersmith@sun.com>
65788Date: Mon Aug 3 19:53:27 2009 -0700
65789
65790 Fix blddir != srcdir builds of solaris-*.il files
65791
65792 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=17509
65793
65794 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
65795
65796commit 1e69fd4a60147287b31e53bfc61543fb17bb82c8
65797Author: Peter Hutterer <peter.hutterer@who-t.net>
65798Date: Mon Aug 3 12:03:26 2009 +1000
65799
65800 xnest: fix up parameters to InitKeyboardDeviceStruct.
65801
65802 IKDS takes a DeviceIntPtr as first argument, and an RMVLO struct as second.
65803 The keysyms stuff is long gone now.
65804
65805 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65806
65807commit 008658049cbeea35a9f76f98037aa2f4173f3573
65808Author: Peter Hutterer <peter.hutterer@who-t.net>
65809Date: Mon Aug 3 12:00:45 2009 +1000
65810
65811 xnest: silence compiler warnings by typecasing properly.
65812
65813 Events.c: In function ‘xnestQueueKeyEvent’:
65814 Events.c:112: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible
65815 pointer type
65816 ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of
65817 type ‘struct xEvent *’
65818 Events.c: In function ‘xnestCollectEvents’:
65819 Events.c:141: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible
65820 pointer type
65821 ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of
65822 type ‘struct xEvent *’
65823 Events.c:150: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible
65824 pointer type
65825 ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of
65826 type ‘struct xEvent *’
65827 Events.c:160: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible
65828 pointer type
65829 ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of
65830 type ‘struct xEvent *’
65831 Events.c:193: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible
65832 pointer type
65833 ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of
65834 type ‘struct xEvent *’
65835
65836 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65837
65838commit 664ac92d8bbe956dd6fd80fac5dc3161028803b2
65839Author: Peter Hutterer <peter.hutterer@who-t.net>
65840Date: Fri Jul 31 14:38:35 2009 +1000
65841
65842 xfixes: backup the DisplayCursor/CloseScreen proc before restoring it (#23034)
65843
65844 The screen's DisplayCursor func is wrapped as
65845 AnimCurDisplayCursor -> CursorDisplayCursor -> miPointerDisplayCursor.
65846
65847 Calling CursorDisplayCursor while an animated cursor was currently displayed
65848 would remove AnimCurDisplayCursor from the wrap stack. Thus, the next call
65849 to ChangeToCursor wouldn't update the animated cursor state. The block
65850 handler for animated cursors would then continuously overwrite the actual
65851 cursor, leaving an animated cursor everywhere on the screen.
65852
65853 X.Org Bug 23034 <http://bugs.freedesktop.org/show_bug.cgi?id=23034>
65854
65855commit f48dfcc1b772a09428e328c72124ea0d46a73416
65856Author: Peter Hutterer <peter.hutterer@who-t.net>
65857Date: Fri Jul 31 10:44:28 2009 +1000
65858
65859 xfixes: allocate CurrentCursor for all devices.
65860
65861 XFixes requires cursor notifies to clients when the cursor changes. This
65862 should work on the ClientPointer and then on all master pointers. Hence
65863 change CurrentCursor to a MAXDEVICES array.
65864
65865 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65866
65867commit d3e5629fac224d53a972df5e2a87db02534f9b17
65868Author: Peter Hutterer <peter.hutterer@who-t.net>
65869Date: Thu Jul 30 16:36:56 2009 +1000
65870
65871 dix: call SetFocusOut and LeaveWindow when disabling a device.
65872
65873 PointerWindows[x] would be set after removing a master pointer. Destroying
65874 this window then crashed the server.
65875
65876 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65877
65878commit 97ed946cfccec5979c47b5fabf1ad56cc7d33ef4
65879Author: Peter Hutterer <peter.hutterer@who-t.net>
65880Date: Thu Jul 30 16:36:30 2009 +1000
65881
65882 dix: SetFocusOut and LeaveWindow don't need parameters other than dev.
65883
65884 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65885
65886commit d040940efe041b57e6323921c380ceb2bb43f41e
65887Author: Peter Hutterer <peter.hutterer@who-t.net>
65888Date: Thu Jul 30 16:16:51 2009 +1000
65889
65890 xfixes: FixesHideCursor should work on all master pointers.
65891
65892 Presumably, a client calling HideCursor doesn't want any cursor displayed.
65893
65894 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65895
65896commit b83c682e7ea4e10757eca6363de28281108591a8
65897Author: Oliver McFadden <oliver.mcfadden@nokia.com>
65898Date: Mon Aug 3 02:46:57 2009 +0300
65899
65900 xf86Xinput: xf86PostButtonEventP must set POINTER_ACCELERATE for relative events.
65901
65902 Thanks to Peter Hutterer for spotting this bug.
65903
65904commit ae65daab080c6269f0eadc6be1a48bf27866c542
65905Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65906Date: Fri Jul 31 15:32:07 2009 -0700
65907
65908 XQuartz: Unify how we set our bitmasks for visuals
65909 (cherry picked from commit c230b52c27ce50ac6c27011ec8e88f1b263b25f9)
65910
65911commit 4a53fed41d711f270c7d8a7fbeaafb4fb619e879
65912Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65913Date: Fri Jul 31 14:19:50 2009 -0700
65914
65915 XQuartz: Dead code removal for StaticColor visual
65916 (cherry picked from commit e457a44e87950207f6c16bf82bf5af11c777ecf1)
65917
65918commit da0a2747803dbecb1308181176a784e14cf02809
65919Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65920Date: Fri Jul 31 14:14:21 2009 -0700
65921
65922 XQuartz: Cleanup the bitmask setting for GLX visuals.
65923 (cherry picked from commit d32c3df258e748958ef997c675dc4fae118c0d7b)
65924
65925commit 70ac671af2f092471022590cb7a19ef9155c51b4
65926Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65927Date: Thu Jul 30 01:30:52 2009 -0700
65928
65929 XQuartz: no DirectColor
65930 (cherry picked from commit df2fbc410f2c484612f65a6539a6cb069ef4a468)
65931
65932commit c0c72a866a237d3c2d9e69e1c69181ef5446e3f8
65933Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
65934Date: Thu Jul 30 00:51:55 2009 -0700
65935
65936 XQuartz: Define DDXRingBell
65937 (cherry picked from commit 113347381289497cb2a79994d0ef5f427ae63ac5)
65938
65939commit 949811313989809f5b58424af04b9fd8e1d0bec4
65940Author: Peter Hutterer <peter.hutterer@who-t.net>
65941Date: Thu Jul 30 21:59:40 2009 +1000
65942
65943 test: fix build by including eventstr.h
65944
65945 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65946
65947commit 0565f4ed4519962bed40a0bbcf0b409471f4de40
65948Author: Peter Hutterer <peter.hutterer@who-t.net>
65949Date: Wed Jul 29 16:46:45 2009 +1000
65950
65951 Xi: set the sourceid for focus devices to the device id.
65952
65953 Unlike Enter/Leave events generated by a device pushing the pointer around,
65954 a device doesn't change focus all by itself. It's a result of a
65955 SetInputFocus call, a window becoming unviewable or a grab activating. As
65956 such, the sourceid for focus events is always the deviceid itself.
65957
65958 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65959
65960commit de4dd5848cab90b0f8b8243ca0b49985ef047124
65961Author: Peter Hutterer <peter.hutterer@who-t.net>
65962Date: Wed Jul 29 16:45:34 2009 +1000
65963
65964 include: DeviceFocusEvent is not to be exported.
65965
65966 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65967
65968commit 46ac9f92416f3cb99b5d84a9d200237dc33a3bb7
65969Author: Peter Hutterer <peter.hutterer@who-t.net>
65970Date: Wed Jul 29 16:34:28 2009 +1000
65971
65972 dix: pass the sourceid around for enter/leave events.
65973
65974 The sourceid for enter/leave events as a result of pointer motion is the ID
65975 of the slave device. The sourceid for those as a result of a grab activating
65976 is the device itself.
65977
65978 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65979
65980commit c299b2228fb63c192b72851c90e14ceaceb67bbc
65981Author: Peter Hutterer <peter.hutterer@who-t.net>
65982Date: Wed Jul 29 16:00:19 2009 +1000
65983
65984 dix: remove obsolete comment, parameter described doesn't exist.
65985
65986 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65987
65988commit 5085ac09a50721d87196bd9f2607dc76200ca399
65989Author: Peter Hutterer <peter.hutterer@who-t.net>
65990Date: Wed Jul 29 13:45:32 2009 +1000
65991
65992 input: switch internal event types to enums.
65993
65994 Use enum EventType instead of ints. This requires a load of default
65995 cases in various switch statements to silence compiler warnings.
65996
65997 Reported-by: Aaron Plattner
65998 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
65999
66000commit 1ae8332d643299a3ee9a9f45a8e25b8c87c751e1
66001Author: Peter Hutterer <peter.hutterer@who-t.net>
66002Date: Wed Jul 29 13:39:38 2009 +1000
66003
66004 include: fix enum EventType declaration.
66005
66006 Having EventType after the enum declares a variable. silly me.
66007
66008 Reported-by: Aaron Plattner
66009 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66010
66011commit 994f7a1c814a89e90f710dac5bf6b2445fb64712
66012Author: Peter Hutterer <peter.hutterer@who-t.net>
66013Date: Wed Jul 29 12:11:13 2009 +1000
66014
66015 record: silence some compiler warnings.
66016
66017 warning: passing argument 4 of ‘RecordAProtocolElement’ discards qualifiers
66018 from pointer target type
66019 note: expected ‘pointer’ but argument is of type ‘const void *’
66020
66021 record.c:2745: warning: passing argument 1 of ‘SwapConnSetupInfo’ from
66022 incompatible pointer type
66023 ../include/swaprep.h:243: note: expected ‘char *’ but argument is of type
66024 ‘struct xConnSetup *’
66025
66026 record.c:2745: warning: passing argument 1 of ‘SwapConnSetupInfo’ from
66027 incompatible pointer type
66028 ../include/swaprep.h:243: note: expected ‘char *’ but argument is of type
66029 ‘struct xConnSetup *’
66030
66031 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66032
66033commit a863d636293cd7361639c1a8cf9c4f7f15da1e1d
66034Author: Peter Hutterer <peter.hutterer@who-t.net>
66035Date: Wed Jul 29 12:09:34 2009 +1000
66036
66037 Xi: remove FIXME and obsolete include.
66038
66039 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66040
66041commit fac49df08f173f091cbb77feaf373d7d465358af
66042Author: Peter Hutterer <peter.hutterer@who-t.net>
66043Date: Wed Jul 29 12:07:22 2009 +1000
66044
66045 Xi: remove obsolete comment.
66046
66047 XI1 only uses 7 bits for deviceids, bit 8 is used for the MORE_EVENTS flag
66048 on the wire (when DeviceValuator events are required).
66049
66050 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66051
66052commit 845e65f08059e8f4bfd37356e99b48bba9416c0c
66053Author: Peter Hutterer <peter.hutterer@who-t.net>
66054Date: Wed Jul 29 11:54:14 2009 +1000
66055
66056 xkb: move XkbFilterEvents to xkbsrv.h
66057
66058 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66059
66060commit 8da0ff2d51086666d10ca7330d428e8610a4a0e3
66061Author: Oliver McFadden <oliver.mcfadden@nokia.com>
66062Date: Tue Jul 28 08:20:37 2009 +0300
66063
66064 xf86Xinput: Add the xf86Post(Proximity|Button|Key)EventP helper functions.
66065
66066 xf86PostKeyboardEvent also makes use of xf86PostKeyEventP to avoid code
66067 duplication, and the valuator verification has been split into the
66068 XI_VERIFY_VALUATORS macro.
66069
66070 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66071
66072commit a148d407429c7d13136b3fcafd2d279c5438df73
66073Author: Peter Hutterer <peter.hutterer@who-t.net>
66074Date: Tue Jul 28 16:53:51 2009 +1000
66075
66076 xkb: restore XKB PtrBtn actions.
66077
66078 Ifdef'd out since the switch to internal events. PtrBtn actions now work
66079 again. Instead of generating the event directly, GPE generates the event and
66080 it is then posted through the usual event processing routines
66081 (mieqProcessDeviceEvent).
66082
66083 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66084
66085commit 3d3b8babd1a5407082f1a40875ed69f62ba2153f
66086Author: Peter Hutterer <peter.hutterer@who-t.net>
66087Date: Tue Jul 28 16:51:14 2009 +1000
66088
66089 mi: update master event copying to InternalEvents.
66090
66091 This is long overdue. The device events are InternalEvents now (and only one
66092 at a time), diminishing the need for an EventList for the master event.
66093 Furthermore, don't make masterEvent a static since this will interfere if
66094 mieqProcessDeviceEvent is called from somewhere else (e.g. XKB actions).
66095
66096 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66097
66098commit f85619b14d130ec54d42cabfaee15e55ced0c665
66099Author: Peter Hutterer <peter.hutterer@who-t.net>
66100Date: Tue Jul 28 16:48:56 2009 +1000
66101
66102 dix: update GetMaximumEventsNum() to real value (3).
66103
66104 GPE and friends now use internal events so they may generate up to 3 events.
66105 One (optional) DeviceChanged event and one raw event plus a device event.
66106
66107 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66108
66109commit 0217d0370c0b0bce66a9c09092eda8e820274e2e
66110Author: Peter Hutterer <peter.hutterer@who-t.net>
66111Date: Tue Jul 28 14:54:30 2009 +1000
66112
66113 record: ifdef out RecordExtensionInit and print a warning to the log.
66114
66115 The RECORD extension is currently broken. By ifdef'ing out the content of
66116 RecordExtensionInit the extension isn't added to the server's internal list
66117 and it does not get advertised to the client. Clients can thus fail
66118 gracefully with a "extension not supported" instead of waiting forever for
66119 events that never arrive.
66120
66121 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66122
66123commit 4e9b2938cd8637a5d3b0a4c9f69d6ee75faab3a0
66124Author: Peter Hutterer <peter.hutterer@who-t.net>
66125Date: Mon Jul 27 16:54:33 2009 +1000
66126
66127 include: untangle events.h from the SDK headers.
66128
66129 InternalEvents shouldn't be used anywhere outside the X server itself. Split
66130 up into events.h for opaque typedefs for the events needed by various
66131 headers and eventstr.h for the actual struct definitions.
66132
66133 eventstr.h must only be included by code that requires internal events and
66134 is not part of the SDK.
66135
66136 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66137
66138commit fad5f96c01811af7490a071719e4017c019a8d9b
66139Author: Jon TURNEY <jon.turney@dronecode.org.uk>
66140Date: Wed Jul 29 13:50:48 2009 +0100
66141
66142 Cygwin/X: Only try to build rootless extension if multiwindow extwm mode is being built
66143
66144 Rootless extension still needs a bit more work to build successfully for Cygwin/X
66145
66146commit de2ae521abde445daaf025a07aa01563ca5ddd41
66147Author: Colin Harrison <colin.harrison@virgin.net>
66148Date: Sun Jul 26 20:48:19 2009 +0100
66149
66150 Xming: Use RegisterClassEx() instead of superseded RegisterClass()
66151
66152 RegisterClass is supserseded by RegisterClassEx, so change to using that everywhere
66153
66154 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
66155
66156commit 764ce6ee683db342264bbca4df6379eb6093fb85
66157Author: Colin Harrison <colin.harrison@virgin.net>
66158Date: Tue Jul 28 16:16:57 2009 +0100
66159
66160 Xming: Use GetClassLongPtr() instead of superseded GetClassLong()
66161
66162 GetClassLong() is superseded by GetClassLongPtr(), so change to using that
66163
66164commit a85523dc50f392a33a1c00302a0946828bc9249d
66165Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66166Date: Tue Jul 28 22:52:33 2009 -0700
66167
66168 XQuartz: Avoid a possible spinlock in applicationWillTerminate
66169 (cherry picked from commit f430cda0fdcc1a8fc5f4795743b40f09ff0bd869)
66170
66171commit 1e49c8d340ee8d8b6f90abcc2e2e9c390cbcd622
66172Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66173Date: Tue Jul 28 13:40:15 2009 -0700
66174
66175 XQuartz: Avoid namespace collission for BOOL in Sparkle
66176 (cherry picked from commit 227c6e01a641b72158201553273299283cdb5599)
66177
66178commit 442967c90dd9d8483a56bdc9237c49e33d619126
66179Author: Alan Coopersmith <alan.coopersmith@sun.com>
66180Date: Tue Jul 28 15:02:37 2009 -0700
66181
66182 Remove hardcoded gcc -Wall option from configure.ac
66183
66184 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
66185
66186commit 52e9ef5664a697a31102e8761eaa03cff01d14d8
66187Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66188Date: Tue Jul 28 19:04:59 2009 +0300
66189
66190 xfree86: remove some RAC junk
66191
66192 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66193
66194commit d57361bb9ac08bec470f76ca8ca602d60c339502
66195Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66196Date: Tue Jul 28 18:55:32 2009 +0300
66197
66198 doc: remove outdated PCI/RAC/Domain notes
66199
66200 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66201
66202commit 7b3d05ebd5a55f88098f4a763d1fa7ca110bb780
66203Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66204Date: Tue Jul 28 18:45:16 2009 +0300
66205
66206 xfree86: remove pci debug macros
66207
66208 there's no effect when their are enabled.
66209
66210 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66211
66212commit 6089d9cfde533c6a7aec7dfc08aec9c8f2f18e27
66213Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66214Date: Tue Jul 28 18:29:35 2009 +0300
66215
66216 xfree86: remove RAC/resource doc
66217
66218 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66219
66220commit c553161e17ebc577ecb91ec6c81f0bdd0ae85e13
66221Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66222Date: Fri Jul 24 13:47:31 2009 +0300
66223
66224 xfree86: delete devices probe code (-probe and -probeonly options)
66225
66226 Inside a windowing system, it's not the place to probe for devices. Goodbye
66227 -probe and -probeonly.
66228
66229 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66230
66231commit c09779f95a9772c0556760222dfc570dbaf8a28e
66232Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66233Date: Fri Jul 24 13:37:31 2009 +0300
66234
66235 xfree86: delete stupid video driver dump (-modalias option)
66236
66237 Such stupid and ugly way to dump PCI information! Oh boy... Anyway, this
66238 doesn't belong to the X server at all. Go away!
66239
66240 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66241
66242commit 0524420622df6cbdb3872917906f7b2a6ec02958
66243Author: Colin Harrison <colin.harrison@virgin.net>
66244Date: Mon Jul 27 15:49:41 2009 +0100
66245
66246 Cygwin/X: use GWLP_WNDPROC, GWLP_USERDATA with Get/SetWindowLongPtr
66247
66248 Missed from commit b3751454cbe02ee952bab213e8c3684d429c41b3
66249
66250 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
66251
66252commit b1c3dc6ae226db178420e3b5f297b94afc87c94c
66253Author: Rémi Cardona <remi@gentoo.org>
66254Date: Mon Jul 27 12:07:51 2009 +0200
66255
66256 config: add HAL error checks
66257
66258 This patch simplifies error handling in the HAL code and fixes a
66259 segfault if libhal_find_device_by_capability() failed.
66260
66261 Fixes http://bugs.gentoo.org/278760
66262
66263 Based on a patch by Martin von Gagern <Martin.vGagern@gmx.net>
66264
66265 Signed-off-by: Rémi Cardona <remi@gentoo.org>
66266 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
66267
66268commit 8898203b0d0e9fa03453b2bcd9b88843cccc3230
66269Author: Dave Airlie <airlied@redhat.com>
66270Date: Tue Jul 28 19:35:37 2009 +1000
66271
66272 sbus: fixup for rac removal
66273
66274commit 4b42448a2388d40f257774fbffdccaea87bd0347
66275Author: Dave Airlie <airlied@redhat.com>
66276Date: Tue Jul 28 14:47:42 2009 +1000
66277
66278 xserver: remove RAC/resource handling code.
66279
66280 This changes the ABI, but since the video ABI is at 6 already
66281 it should be fine.
66282
66283 driver changes are in the pipeline after this.
66284
66285 Signed-off-by: Dave Airlie <airlied@redhat.com>
66286
66287commit 0a168401c401727e49a12cae43a6a387b1f2928d
66288Author: Dave Airlie <airlied@redhat.com>
66289Date: Tue Jul 28 14:19:24 2009 +1000
66290
66291 pci: add support for pci is boot vga call.
66292
66293 This allows us to ask the kernel for the boot VGA device
66294 instead of nasty guessing.
66295
66296commit b3751454cbe02ee952bab213e8c3684d429c41b3
66297Author: Colin Harrison <colin.harrison@virgin.net>
66298Date: Tue Jul 21 16:07:56 2009 +0100
66299
66300 Cygwin/X: Update Get/SetWindowLong() to Get/SetWindowLongPtr() everywhere
66301
66302 Get/SetWindowLong() is superseded by Get/SetWindowLongPtr(), so change to
66303 using that everywhere it remains
66304
66305 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
66306
66307commit c81595e23b48368dafc054f023c1dd16bbad9494
66308Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66309Date: Sun Jul 26 02:32:17 2009 -0700
66310
66311 XQuartz: Overhaul setting up visuals
66312
66313 The main change is cleanup of the visualConfigs and setting up alpha correctly there to match the visuals being added earlier (so the default visual has a corresponding GLX visual)
66314 (cherry picked from commit 7351db5c8746be30a4047469ee9b50bc19e62a89)
66315
66316commit 7f28c555b80dda2ed4f518efdb79733647dfea80
66317Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66318Date: Sat Jul 25 20:19:05 2009 -0700
66319
66320 XQuartz: Use CopyKeyClass to copy the keymap to the virtual core keyboard.
66321 (cherry picked from commit 9a801d1716f005c30be076fcc9ac8dbb3e74d989)
66322
66323commit 1031ac3a7306e7a82169c79c64607696c826c47f
66324Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66325Date: Sat Jul 25 20:18:38 2009 -0700
66326
66327 Revert "XQuartz: Copy the keyboard map to the core keyboard"
66328
66329 This reverts commit 795de791cf18c658421d701af645718493eac51e.
66330
66331commit 48703083a1cf308306f254691d7c2ecda09b3812
66332Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66333Date: Sat Jul 25 19:34:17 2009 -0700
66334
66335 XQuartz: Use pDev=NULL for DarwinSendDDXEvent
66336
66337 These events aren't really related to physical input devices anyways, so it doesn't make sense to use the pointer.
66338 (cherry picked from commit bfe0b9cfa7af4a48dba849cab1eb152c409b4e08)
66339
66340commit b8e0f740829d0c81324aeb59222fc8e3d22493cc
66341Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66342Date: Sat Jul 25 20:17:25 2009 -0700
66343
66344 mieq: Protect from pDev=NULL in mieqEnqueue and mieqProcessInputEvents
66345 (cherry picked from commit bf60ffb49700da367f7f88983b042a88fef7219b)
66346
66347commit dab8de036808e363a8ec9be826cff7fbcd92b953
66348Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66349Date: Sat Jul 25 15:21:12 2009 -0700
66350
66351 XQuartz: Bump the reported version to X11R7.4
66352 (cherry picked from commit c83f701aa75c75433b8745f5d79bca3a7516df91)
66353
66354commit 9e74bb97ab6cb83dd99ec36bd3842197263b8a8c
66355Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66356Date: Sat Jul 25 10:39:26 2009 -0700
66357
66358 XQuartz: xpr: Added missing include for RootlessHideAllWindows
66359 (cherry picked from commit 75e104730810acbb11bf8503bac24ea25243b2ac)
66360
66361commit 09b024db636f5a306c9cec3806fffde9acd3c408
66362Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66363Date: Sat Jul 25 15:22:01 2009 -0700
66364
66365 XQuartz: Use the master device in DarwinSendDDXEvent to avoid duplicate events.
66366 (cherry picked from commit 6fa62192af937aa9656f64b516050bc099231c7a)
66367
66368commit 795de791cf18c658421d701af645718493eac51e
66369Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66370Date: Sat Jul 25 15:25:06 2009 -0700
66371
66372 XQuartz: Copy the keyboard map to the core keyboard
66373
66374 This still doesn't handle the modifier map... gotta figure out what to do now that SwitchCoreKeyboard is gone
66375 (cherry picked from commit 427e1aab41dabb54354bfd30f2baae98ac8202c0)
66376
66377commit af8047f7816b39be7015e8cbbe085c57bb354592
66378Author: Aaron Plattner <aplattner@nvidia.com>
66379Date: Fri Jul 17 15:10:46 2009 -0700
66380
66381 Damage: Add devPrivates to DamageRec
66382
66383 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
66384
66385commit c80ed4945ef6dd50473fab871211b71b3d20dc3c
66386Author: Aaron Plattner <aplattner@nvidia.com>
66387Date: Fri Jul 17 15:10:32 2009 -0700
66388
66389 Damage: Add wrappable hooks for damage create, destroy, register, & unregister.
66390
66391 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
66392
66393commit fb46474feb69d52d3dfdd61452ced5cfc38f6651
66394Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66395Date: Fri Jul 24 09:16:37 2009 -0700
66396
66397 XQuartz: Rever the "Set can_quit to true during a Sparkle-initiated relaunch." change
66398
66399 This was in place to work around the issue that was correctly solved with the
66400 previous commit (changing the Windows menu behavior). Reverting this change
66401 no longer causes crashing, so it's safe to show the dialog now.
66402 (cherry picked from commit 023cef31bbb2ab80a241098d82fcfd35ada75fc0)
66403
66404commit 0d73893a5795d5c1d28fd0287f2f1965f2ff1bda
66405Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66406Date: Thu Jul 23 20:03:29 2009 -0700
66407
66408 XQuartz: Change handling of Windows menu to workaround a bug triggered by mixing Sparkle and X11 windows
66409
66410 <rdar://problem/7088335> NSApplication releases the separator in the Windows menu even though it's an IBOutlet
66411 (cherry picked from commit 27ac5135267be9cb221329ae68461117dd43a4bf)
66412
66413commit 171409cecbd848e5fec6334fc61a20e882e80fa9
66414Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66415Date: Tue Jul 21 00:22:12 2009 -0700
66416
66417 XQuartz: Set can_quit to true during a Sparkle-initiated relaunch.
66418 (cherry picked from commit b2e9a77111d4572402d8ca95e3368db97ba7d598)
66419
66420commit 7ff84d350b44fa40669c1d0d48a715a0bf056ece
66421Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66422Date: Mon Jul 20 22:38:25 2009 -0700
66423
66424 XQuartz: Added a "Check for X11 Updates..." menu item.
66425 (cherry picked from commit 305144bfa4b59791123c44e869fab93a084792d6)
66426
66427commit 1e1dbd1e462f571dad2f9684fcf4cd8ae17eedd5
66428Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66429Date: Mon Jun 29 11:46:17 2009 +0300
66430
66431 xfree86: "Staticize" functions in xf86AutoConfig.c
66432
66433 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66434
66435commit 7c6b5458de9bc7f6cd972a36b56888aaa3d201ee
66436Author: Aaron Plattner <aplattner@nvidia.com>
66437Date: Thu Jul 23 15:16:44 2009 -0700
66438
66439 Fix dist.
66440
66441 * Makefile.am: Include the test/ subdirectory in the dist tarball.
66442
66443 * include/Makefile.am: Move events.h to sdk_HEADERS and eventconvert.h to
66444 EXTRA_DIST so they're included in the tarball. events.h shouldn't be
66445 included in the SDK either, but for now it's needed by input.h.
66446
66447commit 20daa145c437c3ba67970146f6182849f87a1b43
66448Author: Peter Hutterer <peter.hutterer@who-t.net>
66449Date: Thu Jul 23 08:33:00 2009 +1000
66450
66451 mi: fix build error caused by missing xtest.h
66452
66453 xtest.h was renamed to xtestconst.h in xextproto.
66454 Requires xextproto 7.0.99.3
66455
66456 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66457
66458commit 587c3a2d1961834558193e8e14e8e381a077a253
66459Author: Aaron Plattner <aplattner@nvidia.com>
66460Date: Wed Jul 22 10:55:46 2009 -0700
66461
66462 Bug #22804: Reject out of bounds XGetImage requests
66463
66464 The XGetImage man page states:
66465
66466 If the drawable is a window, the window must be viewable, and it must be the
66467 case that if there were no inferiors or overlapping windows, the specified
66468 rectangle of the window would be fully visible on the screen and wholly
66469 contained within the outside edges of the window, or a BadMatch error
66470 results. Note that the borders of the window can be included and read with
66471 this request.
66472
66473 However, the server was only checking the requested region against the screen
66474 bounds, allowing XGetImage requests to read pixels outside the bounds of a
66475 window's ancestors. Normally, this would just read other pixels from the
66476 screen, but if one of the ancestor windows is redirected, the window's backing
66477 pixmap may be smaller than the window itself.
66478
66479 This change checks the region against the window's bounding drawable, which is
66480 either the screen pixmap, a redirected window's backing pixmap, or the root
66481 window for servers that don't support GetWindowPixmap.
66482
66483 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
66484 Reviewed-by: Keith Packard <keithp@keithp.com>
66485
66486commit ecd618957ebf01cb4137f98efec3faed35f8a9f5
66487Author: Peter Hutterer <peter.hutterer@who-t.net>
66488Date: Wed Jul 22 15:19:34 2009 +1000
66489
66490 dix: fix null-pointer dereference on activating enter/focus grabs.
66491
66492 EventToXI2 returns a NULL event for enter/focus events since these events
66493 aren't yet wrapped into internal events. This is a quickfix only, it works
66494 thanks to the alignment of internal and XI2 event types.
66495
66496 Eventually, enter/leave events should be wrapped into internal events.
66497
66498 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66499
66500commit 48585bd1e3e98db0f3df1ecc68022510216e00cc
66501Author: Peter Hutterer <peter.hutterer@who-t.net>
66502Date: Wed Jul 22 12:09:24 2009 +1000
66503
66504 dix: use the event mask of the grab for TryClientEvents.
66505
66506 A client that grabs for button/key events may not have the
66507 ButtonPress/KeyPress mask set and should not receive an event in that case.
66508
66509 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66510
66511commit 4dc91b3e54503a1be555bae5b18f3e52f58be307
66512Author: Pierre Willenbrock <pierre@pirsoft.de>
66513Date: Tue Jul 21 17:21:28 2009 +0200
66514
66515 Check if new space was actually allocated before freeing.
66516
66517 There will be no new space allocated, if mode != PropModeReplace and
66518 len == 0, or if mode is not one of the handled modes.
66519 This fixes freeing data that is still in use, leading to double frees and
66520 other memory corruption.
66521
66522 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66523
66524commit 9a1bfa5664a80f03cedeb89b9f8a86115a08e7af
66525Author: Peter Hutterer <peter.hutterer@who-t.net>
66526Date: Mon Jul 20 16:39:16 2009 +1000
66527
66528 input: remove XI2 keysym grabs, use keycode grabs instead.
66529
66530 Keysym grabs are tricky in the details, keycode grabs are known to work. So
66531 for now, provide keycode grabs only.
66532
66533 Requires inputproto 1.9.99.15.
66534
66535 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66536
66537commit afc3e3b5955ea4a49308399820cc4c499f4312da
66538Author: Jon TURNEY <jon.turney@dronecode.org.uk>
66539Date: Tue Jul 21 16:14:21 2009 +0100
66540
66541 Cygwin/X: winInitMultiWindowClass() should be static
66542
66543 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
66544
66545commit 9dc8106ba095474fea1474622b0f0162c8fd5aa4
66546Author: Michel Dänzer <daenzer@vmware.com>
66547Date: Tue Jul 21 14:34:47 2009 +0200
66548
66549 randr12: Add compatibility for XF86VidMode gamma ramps.
66550
66551 Fixes screensaver fadeout effects.
66552
66553 Also make all RandR 1.2 compatibility code for XF86VidMode operate only on the
66554 CRTC associated with the compatibility output, not all CRTCs at once.
66555
66556commit 268e227ba06c027f5c56b1aaee5dcc6a2034403f
66557Author: Michel Dänzer <daenzer@vmware.com>
66558Date: Tue Jul 21 14:34:13 2009 +0200
66559
66560 EXA: Make Prepare/FinishAccess tracking resilient to repeated / nested calls.
66561
66562 Use reference counting and do nothing unless the reference count transitions
66563 to/from 0.
66564
66565 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=22822 .
66566
66567 As a bonus, this avoids calling the driver Prepare/FinishAccess hooks more than
66568 once per pixmap and operation.
66569
66570 Also update the Doxygen documentation for the PrepareAccess driver hook to
66571 better match current reality.
66572
66573commit de7a14ca92f99ff03c8ad204aab5be8203c86a72
66574Author: Michel Dänzer <daenzer@vmware.com>
66575Date: Tue Jul 21 12:55:27 2009 +0200
66576
66577 EXA: Fix up some issues introduced by 00fe4a297744c81b40f0243fb56ad848a9be6a2b.
66578
66579 * Check all pixmaps involved for damage records, fixes visual corruption due to
66580 the screen pixmap never having one.
66581 * Fix an array size and remove a now superfluous assignment.
66582
66583commit be4dd35ffecbf49aff13aa9f604a44c9b665ae92
66584Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66585Date: Sat Jul 18 15:51:29 2009 -0700
66586
66587 XQuartz: Initial support for automatic updates through Sparkle
66588 (cherry picked from commit c45f1be36426bceeef9af67c26351114f14f5277)
66589
66590commit 2415e2dce918efd49d5e6d71f705255a234a866b
66591Author: Michel Dänzer <daenzer@vmware.com>
66592Date: Mon Jul 20 02:08:31 2009 +0200
66593
66594 EXA: Bail earlier from exaDoPutImage if the driver has no UploadToScreen hook.
66595
66596 Also remove dead code associated with access_prepared local variable.
66597
66598commit 00fe4a297744c81b40f0243fb56ad848a9be6a2b
66599Author: Michel Dänzer <daenzer@vmware.com>
66600Date: Mon Jul 20 02:04:40 2009 +0200
66601
66602 EXA: Completely eliminate exaDoMigration calls for drivers that manage pixmaps.
66603
66604commit 7b9915b11ed9eedd0698b4563328504d686ac4ec
66605Author: Gaetan Nadon <memsize@videotron.ca>
66606Date: Sat Jul 18 20:52:20 2009 -0400
66607
66608 dix: xserver "make dist" fails due to eventconvert.h (#27825)
66609
66610 This header file is not in the /dix dir, but in the /include dir.
66611 The makefile aborted and the xserver distribution files were not created.
66612 The fix is to remove this header file from the libdix_la_SOURCES
66613 in the dix/Makefile.am.
66614
66615 X.Org Bug 27825 <http://bugs.freedesktop.org/show_bug.cgi?id=27825>
66616
66617 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66618
66619commit a98acb290737ac6b0776f2ea8128a4613c0f778b
66620Author: Adam Jackson <ajax@redhat.com>
66621Date: Wed Jul 15 13:56:19 2009 -0400
66622
66623 s/xf86EnableOutputs/xf86CollectEnabledOutputs/ for clarity
66624
66625commit 8fab7f72f2cc4ac5ca415c95ccbd05d3084f10c4
66626Author: Adam Jackson <ajax@redhat.com>
66627Date: Wed Jul 15 14:09:08 2009 -0400
66628
66629 randr: Add Option "Primary" to Monitor sections
66630
66631commit 053bb92145045e06f8d72c3bddb75806432faa1c
66632Author: Adam Jackson <ajax@redhat.com>
66633Date: Thu Jul 16 14:29:05 2009 -0400
66634
66635 xfree86: Remove xf86SetPriority
66636
66637 GetClocks is pretty doomed to begin with, this really isn't going to
66638 make it worse.
66639
66640commit 3637945a45d55385335833197b18703379892884
66641Author: Adam Jackson <ajax@redhat.com>
66642Date: Thu Jul 16 14:10:45 2009 -0400
66643
66644 loader: Remove a useless check.
66645
66646commit ef2bb08c76f0d1eed0c8b3e4937648f0b6ae01e5
66647Author: Adam Jackson <ajax@redhat.com>
66648Date: Thu Jul 16 13:53:00 2009 -0400
66649
66650 loader: remove dead LoaderCheckUnresolved
66651
66652commit 06a2784232b5d0e2440ed365d272abdfa1f5d35f
66653Author: Adam Jackson <ajax@redhat.com>
66654Date: Thu Jul 16 13:40:48 2009 -0400
66655
66656 loader: Remove useless TestFree() macro
66657
66658commit 087745b809b9a834751e5c719b9d87967194369b
66659Author: Adam Jackson <ajax@redhat.com>
66660Date: Thu Jul 16 10:51:53 2009 -0400
66661
66662 fbdevhw: Remove pointless OS check, this never gets built on non-Linux
66663
66664commit 892c93553aa0b03aeb3f4c27d952cc4bb7120aff
66665Author: Adam Jackson <ajax@redhat.com>
66666Date: Thu Jul 16 10:51:27 2009 -0400
66667
66668 fbdevhw: Remove a #if 0
66669
66670commit e5712f2926abf33b146a4ccfcf6e89c914dad37b
66671Author: Adam Jackson <ajax@redhat.com>
66672Date: Thu Jul 16 10:49:22 2009 -0400
66673
66674 fbdevhw: simplify some #if 1
66675
66676commit 8868bb4131be25340bf65ec61b998d353965bab6
66677Author: Adam Jackson <ajax@redhat.com>
66678Date: Thu Jul 16 10:44:24 2009 -0400
66679
66680 randr: Un-duplicate the reduced blanking check.
66681
66682commit 453ee39bc6a6b46e6cda11512cfcd431ba32a4b7
66683Author: Adam Jackson <ajax@redhat.com>
66684Date: Thu Jul 16 10:39:37 2009 -0400
66685
66686 xfree86: Fix some misleading comments
66687
66688commit 2c57a7aa07fdf52be315ecb498341776268c1a10
66689Author: Adam Jackson <ajax@redhat.com>
66690Date: Thu Jul 16 10:32:53 2009 -0400
66691
66692 xfree86: Remove some #if 0
66693
66694 This code all lives in xf86Modes.c now anyway
66695
66696commit 1a8c89683ff40cdd1d33da89ab47e7bd240b9bb7
66697Author: Adam Jackson <ajax@redhat.com>
66698Date: Thu Jul 16 10:15:41 2009 -0400
66699
66700 xfree86: Remove TargetRefresh option
66701
66702 This was a vestige from the days before we'd make the mode list from the
66703 EDID data, and from CRT technology when you could reasonably assume that
66704 higher refresh rates were better. Also it did not function as advertised,
66705 acting as a high-pass filter instead of a band-pass.
66706
66707commit b4ee3bf700e04ae1b3c4ec021373424ed6e5338f
66708Author: Adam Jackson <ajax@redhat.com>
66709Date: Fri Jul 17 14:57:50 2009 -0400
66710
66711 dbe: Fix indentation
66712
66713commit 0bb9a7e1650180a24246d14493a8168487cf8914
66714Author: Adam Jackson <ajax@redhat.com>
66715Date: Tue Jun 9 11:49:41 2009 -0400
66716
66717 displayid: Implement mode decoding.
66718
66719commit 2f1a9c5baa367818bf017bdb72f20a2f6fa7ac21
66720Author: Adam Jackson <ajax@redhat.com>
66721Date: Tue Jun 9 10:13:47 2009 -0400
66722
66723 ddc: s/xf86DDCMonitorSet/xf86EdidMonitorSet/
66724
66725 Since we need a second path for DisplayID.
66726
66727commit d0cb4f5a91932e901d10cac5f2a4ba12bb8a0e6f
66728Author: Adam Jackson <ajax@redhat.com>
66729Date: Tue Jun 9 10:10:18 2009 -0400
66730
66731 ddc: Refactor root window property code
66732
66733commit 14103b781201bc36896cbe9112a2e0d991fb785d
66734Author: Adam Jackson <ajax@redhat.com>
66735Date: Tue Jun 9 10:05:01 2009 -0400
66736
66737 ddc: Don't try to publish a root window property for DisplayID
66738
66739commit c302a5ff49146bff24df196fc36ed38745d42911
66740Author: Adam Jackson <ajax@redhat.com>
66741Date: Mon Jun 8 17:42:10 2009 -0400
66742
66743 ddc: Skeleton for xf86DoDisplayID()
66744
66745commit 0b36f68efb1171fcdbe53e93064394f5609b7fb5
66746Author: Adam Jackson <ajax@redhat.com>
66747Date: Mon Jun 8 15:36:15 2009 -0400
66748
66749 ddc: mv xf86DDC.c ddc.c
66750
66751commit 8eb82168fc5c7ea958a4f63676738510647dd203
66752Author: Adam Jackson <ajax@redhat.com>
66753Date: Mon Jun 8 15:34:42 2009 -0400
66754
66755 ddc: Give DisplayID a place to hang its hat
66756
66757commit ace0fe09aee48d57cd0079260cd8d20d041e8eb6
66758Author: Adam Jackson <ajax@redhat.com>
66759Date: Mon Jun 8 14:45:47 2009 -0400
66760
66761 ddc: Yet more code motion
66762
66763commit 3a350688bbe1257feaf1502b4009f5f701d2640e
66764Author: Adam Jackson <ajax@redhat.com>
66765Date: Mon Jun 8 14:37:38 2009 -0400
66766
66767 ddc: Code motion to eliminate forward decls
66768
66769commit 8797831f82637b0e65a08e1262d1ec57c075cc12
66770Author: Adam Jackson <ajax@redhat.com>
66771Date: Mon Jun 8 14:30:53 2009 -0400
66772
66773 ddc: Update a comment.
66774
66775commit bb6fa39eb3aa044ffc18632288fd59909ff06344
66776Author: Adam Jackson <ajax@redhat.com>
66777Date: Thu Jul 16 17:43:31 2009 -0400
66778
66779 dbe: Adapt to new headers
66780
66781commit 9965174f4f09a7de00a9569607e96226208bc2ce
66782Author: Pierre Willenbrock <pierre@pirsoft.de>
66783Date: Thu Jul 16 12:12:48 2009 +0200
66784
66785 Check dev->u.master if there is a custom event handler, too
66786
66787 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66788
66789commit c806162c133603a99d9cd844bb04485bb663707d
66790Author: Pierre Willenbrock <pierre@pirsoft.de>
66791Date: Tue Jul 14 13:42:19 2009 +0200
66792
66793 Always update u.lastSlave
66794
66795 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66796
66797commit 4db8950b1dfc3150ee2e9f1f975e9ecb4eabb1f2
66798Author: Peter Hutterer <peter.hutterer@who-t.net>
66799Date: Fri Jul 17 10:40:59 2009 +1000
66800
66801 Xext: include securproto.h instead of securstr.h
66802
66803 Reported-by: Byeong-ryeol Kim
66804 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66805
66806commit b5f6da1e9b37f52484948185dcf17950657cd65b
66807Author: Oliver McFadden <oliver.mcfadden@nokia.com>
66808Date: Fri Jul 17 06:34:00 2009 +0300
66809
66810 Revert "Coverity Prevent: RESOURCE_LEAK in AccelSetProfileProperty:"
66811
66812 This reverts commit 7333dc2969f60af0abcfb28e7182a5fff9918223.
66813
66814 False positive.
66815
66816commit f517fca31d8c341ad36f3de4863adb0bc5206176
66817Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66818Date: Thu Jul 16 17:28:56 2009 -0700
66819
66820 XQuartz: Localization updates
66821 (cherry picked from commit 4b797fc1edf2bd963410a3133e3d2182ccfda4c3)
66822
66823commit c8a35165147a9b2f747b25da80d0fdb4279a8331
66824Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66825Date: Thu Jul 16 17:20:16 2009 -0700
66826
66827 XQuartz: Cleanup getGlCapabilities to avoid hardcoding the number of displays
66828 (cherry picked from commit 12f7365f1f58d648217b16f09c2152fa47dcd7a1)
66829
66830commit e7bc9ff816c1848c700a376908a1411f1e20d29f
66831Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66832Date: Thu Jul 16 16:59:47 2009 -0700
66833
66834 XQuartz: Allow more than 3 OSX displays
66835 (cherry picked from commit 45045eb3968069a8d74ce9188890a5537dc60fec)
66836
66837commit 7333dc2969f60af0abcfb28e7182a5fff9918223
66838Author: Oliver McFadden <oliver.mcfadden@nokia.com>
66839Date: Thu Jul 16 18:57:49 2009 +0300
66840
66841 Coverity Prevent: RESOURCE_LEAK in AccelSetProfileProperty:
66842
66843 Event alloc_arg: Called allocation function "XIPropToInt" on "ptr" [details]
66844 167 rc = XIPropToInt(val, &nelem, &ptr);
66845
66846 Event leaked_storage: Variable "ptr" goes out of scope
66847
66848commit 66eabbebaf1b40fed7670b7c05337ed6226dd81e
66849Author: Oliver McFadden <oliver.mcfadden@nokia.com>
66850Date: Thu Jul 16 18:50:22 2009 +0300
66851
66852 Coverity Prevent: RESOURCE_LEAK in xf86CrtcSetInitialGamma
66853
66854commit 464e8ad733fa6afee028607d6e7d4663b4c273cc
66855Author: Oliver McFadden <oliver.mcfadden@nokia.com>
66856Date: Thu Jul 16 17:51:08 2009 +0300
66857
66858 Coverity Prevent: NEGATIVE_RETURNS in fbdev_open_pci:
66859
66860 Event var_tested_neg: Variable "fd" tested NEGATIVE
66861 At conditional (1): "fd != -1" taking false path
66862 335 if (fd != -1) {
66863
66864 Event negative_returns: Tracked variable "fd" was passed to a negative sink.
66865 347 close(fd);
66866
66867commit 53ae6b63387e04324b23b6f8311cc22a154c1fb8
66868Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
66869Date: Wed Jul 15 23:18:35 2009 -0700
66870
66871 XQuartz: Cleanup keymap locking, fix a possible synchro bug
66872 (cherry picked from commit 33e7437a4984ee7c1b04b87d23dee7d4739d5f12)
66873
66874commit 6a90c7b93724a2d26eae70b5806ca06c91e7df4c
66875Author: Peter Hutterer <peter.hutterer@who-t.net>
66876Date: Thu Jul 16 09:28:42 2009 +1000
66877
66878 xkb: cosmetic fix, use TRUE instead of True.
66879
66880 Rest of InitKeyboardDeviceStruct uses TRUE and FALSE.
66881
66882 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66883
66884commit 52088d3c2d53a91b2690f9cc402f839d77769e37
66885Author: Éric Piel <eric.piel@trempin-utc.net>
66886Date: Sun Jun 21 18:11:28 2009 +0200
66887
66888 xserver: remove unused code in clipValuators
66889
66890 The axes variables was never used, remove it.
66891
66892 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66893
66894commit d3f6b43a240eb763025b3cbf546cb7ae502c94fa
66895Author: Peter Hutterer <peter.hutterer@who-t.net>
66896Date: Wed Jul 15 16:51:05 2009 +1000
66897
66898 Update to xextproto 7.0.99.1.
66899
66900 xextproto had Xlib client headers moved into libXext.
66901 Protocol header files are named fooproto.h, header files with constants
66902 foo.h or fooconst.h where foo.h was already in use for client-side headers.
66903
66904commit 693babbf12cc7969c3e211c4037c7af0d41c13e9
66905Author: Peter Hutterer <peter.hutterer@who-t.net>
66906Date: Wed Jul 15 12:30:02 2009 +1000
66907
66908 xkb: Remove XKMformat.h include from xkbsrv.h into the files that need it.
66909
66910 xkbsrv.h is used by drivers, they don't need the XKM format and shouldn't
66911 require it.
66912
66913 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66914
66915commit 35ff5cd26eb7564fefebf238e30b8d43cbb4dc25
66916Author: Peter Hutterer <peter.hutterer@who-t.net>
66917Date: Wed Jul 15 10:29:46 2009 +1000
66918
66919 dix: fix wrong raw valuator copy
66920
66921 internal events keep valuator data at the index for the valuator, not like
66922 the wire events that start with first_valuator.
66923
66924 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66925
66926commit 975bf60a82f863e3bdc36fc3f8201b48a4d1e6f7
66927Author: Peter Hutterer <peter.hutterer@who-t.net>
66928Date: Wed Jul 15 10:29:19 2009 +1000
66929
66930 dix: use sizeof(FP3232) instead of 2 * sizeof(int32_t).
66931
66932 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66933
66934commit d040af7fa3c7314917414a28d723bdda3c4289c3
66935Author: Peter Hutterer <peter.hutterer@who-t.net>
66936Date: Mon Jul 13 15:09:38 2009 +1000
66937
66938 Update to type-specific raw events - require inputproto 1.9.99.14.
66939
66940 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66941
66942commit 81b64f668541217c1c9518ac4a7e3fdf0a6e4002
66943Author: Fredrik Höglund <fredrik@kde.org>
66944Date: Wed Jul 15 00:38:01 2009 +0200
66945
66946 Require renderproto >= 0.11 and pixman-1 >= 0.15.14 for the new blend modes.
66947
66948commit 0ce42adbf4cff9e7f049d9fc79d588ece5936177
66949Author: Fredrik Höglund <fredrik@kde.org>
66950Date: Wed Jul 15 00:37:05 2009 +0200
66951
66952 Render: Add support for the PDF blend mode operators.
66953
66954commit 686e4867302a741f3029c4105b997d0f0ac7c13c
66955Author: Tiago Vignatti <tiago.vignatti@nokia.com>
66956Date: Tue Jul 14 17:11:13 2009 +0300
66957
66958 xfree86: remove a bunch of unused pci headers
66959
66960 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
66961
66962commit cc575a3ba4a52265e410b325c2291fe900a54f33
66963Author: Peter Hutterer <peter.hutterer@who-t.net>
66964Date: Fri Jun 12 11:38:46 2009 +1000
66965
66966 test: add tests for ProcXIGetClientPointer.
66967
66968 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66969
66970commit 92b0065a19dcde50494d2528e19a4ee76723b3da
66971Author: Peter Hutterer <peter.hutterer@who-t.net>
66972Date: Fri Jun 12 11:38:00 2009 +1000
66973
66974 test: add ProcXISetClientPointer tests.
66975
66976 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66977
66978commit 019bc9d70c5b3beb69ca20b6bca3551e00415992
66979Author: Peter Hutterer <peter.hutterer@who-t.net>
66980Date: Fri Jun 12 11:37:17 2009 +1000
66981
66982 test: add ProcXIGetSelectedEvents tests.
66983
66984 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66985
66986commit 8579d39c12b065d8e82fe75af4b254c8ad4f4ab8
66987Author: Peter Hutterer <peter.hutterer@who-t.net>
66988Date: Fri Jun 12 11:36:22 2009 +1000
66989
66990 test: add ProcXISelectEvents tests.
66991
66992 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
66993
66994commit 60b7a7990003a22be5cc26ebe4d8d2d0803b97de
66995Author: Peter Hutterer <peter.hutterer@who-t.net>
66996Date: Fri Jun 12 11:33:22 2009 +1000
66997
66998 test: add ProcXIQueryDevice tests.
66999
67000 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67001
67002commit 441c91be05912e2665cc1d3ee72fb70dcd32e7a2
67003Author: Peter Hutterer <peter.hutterer@who-t.net>
67004Date: Fri Jun 12 11:32:15 2009 +1000
67005
67006 test: Add ProcXIQueryVersion tests.
67007
67008 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67009
67010commit 1d635ae65b217e124b3bd5a8844e73f447eb1b79
67011Author: Peter Hutterer <peter.hutterer@who-t.net>
67012Date: Fri Jun 12 11:30:53 2009 +1000
67013
67014 test: add xi2 directory, prepare for protocol testing.
67015
67016 These two files provide a couple of common defines, functions and variables
67017 that will be used in a number of protocol tests.
67018
67019 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67020
67021commit 4b93413b2a490a74de2e27b37268332b609ac872
67022Author: Peter Hutterer <peter.hutterer@who-t.net>
67023Date: Tue Jul 14 11:20:01 2009 +1000
67024
67025 Add test subdir to base Makefile.am
67026
67027 If unittests are enabled, make will build those as well - spotting potential
67028 build errors in the tests faster. Furthermore, this allows for the tests to
67029 be run from the top-level directory.
67030 This patch removes the "run make check to build the test suite" message
67031 since that'd pop up after every build now.
67032
67033 If unittests are disabled, this change has no effect.
67034
67035 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67036
67037commit e2226f3cd7d1bef598f9657b756a171b02f1a299
67038Author: Peter Hutterer <peter.hutterer@who-t.net>
67039Date: Mon Jul 6 14:57:04 2009 +1000
67040
67041 dbe: switch to byte counting functions
67042
67043 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67044
67045commit 3e1a054423c22f0e35f25127dde1ea8263892480
67046Author: Peter Hutterer <peter.hutterer@who-t.net>
67047Date: Mon Jul 6 14:24:45 2009 +1000
67048
67049 record: switch to byte counting functions
67050
67051 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67052
67053commit 089c460058c4e6814d98d40179eb4c731f37fa80
67054Author: Peter Hutterer <peter.hutterer@who-t.net>
67055Date: Mon Jul 6 14:22:40 2009 +1000
67056
67057 xkb: switch to byte counting functions
67058
67059 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67060
67061commit 0dc11da57562a7bbb6830ab657a2b818c596a537
67062Author: Peter Hutterer <peter.hutterer@who-t.net>
67063Date: Mon Jul 6 12:41:29 2009 +1000
67064
67065 ephyr: switch to byte counting functions
67066
67067 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67068
67069commit 34bfaa9d9ecd90cfe8413bc275179fdcc193eab3
67070Author: Peter Hutterer <peter.hutterer@who-t.net>
67071Date: Mon Jul 6 12:29:26 2009 +1000
67072
67073 xfixes: switch to byte counting functions
67074
67075 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67076
67077commit e216527107fda470b92b7e526f3db22465962a43
67078Author: Peter Hutterer <peter.hutterer@who-t.net>
67079Date: Mon Jul 6 12:27:23 2009 +1000
67080
67081 render: switch to byte counting functions
67082
67083 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67084
67085commit 1d6b71b8d0dc24355e84391ba413170a03ccdbf4
67086Author: Peter Hutterer <peter.hutterer@who-t.net>
67087Date: Mon Jul 6 12:25:39 2009 +1000
67088
67089 os: switch to byte counting functions
67090
67091 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67092
67093commit 86b239ff9c4d01685c357ca2b1ef761d167e3224
67094Author: Peter Hutterer <peter.hutterer@who-t.net>
67095Date: Mon Jul 6 12:24:27 2009 +1000
67096
67097 randr: switch to byte counting functions
67098
67099 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67100
67101commit 7b9e84e320a6f6449fe7bc58a8d6a094ae37b86c
67102Author: Peter Hutterer <peter.hutterer@who-t.net>
67103Date: Mon Jul 6 12:20:21 2009 +1000
67104
67105 Xext: switch to byte counting functions
67106
67107 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67108
67109commit aa19d355125a10b1a385c8f134d68e79d3d609c7
67110Author: Peter Hutterer <peter.hutterer@who-t.net>
67111Date: Fri Jul 10 14:25:22 2009 +1000
67112
67113 xwin: switch to byte counting functions
67114
67115 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67116
67117commit 0b4e6af857bcc5513e4c19912a54656d4696e56d
67118Author: Peter Hutterer <peter.hutterer@who-t.net>
67119Date: Mon Jul 6 12:12:57 2009 +1000
67120
67121 xquartz: switch to byte counting functions
67122
67123 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67124 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
67125
67126commit ad508c93c239a5ba8381000c031e96caf2769265
67127Author: Peter Hutterer <peter.hutterer@who-t.net>
67128Date: Mon Jul 6 12:12:20 2009 +1000
67129
67130 xfree86: switch to byte counting functions
67131
67132 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67133
67134commit c20304226b3ca2f8d0a4f4866480b0d71913941c
67135Author: Peter Hutterer <peter.hutterer@who-t.net>
67136Date: Mon Jul 6 12:07:00 2009 +1000
67137
67138 glx: switch to byte counting functions
67139
67140 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67141
67142commit 87ce93c9973067255b4197d82772f83cd4ea5d27
67143Author: Peter Hutterer <peter.hutterer@who-t.net>
67144Date: Mon Jul 6 12:06:47 2009 +1000
67145
67146 dmx: switch to byte counting functions
67147
67148 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67149
67150commit dc82e11e509ecf586d77c3e7c1325d515509ce51
67151Author: Peter Hutterer <peter.hutterer@who-t.net>
67152Date: Mon Jul 6 12:03:34 2009 +1000
67153
67154 exa: switch to byte counting functions.
67155
67156 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67157 Acked-by: Michel Dänzer <michel@daenzer.net>
67158
67159commit 2d35ea8d957a955e1200ba2b14424bddfe1f4148
67160Author: Peter Hutterer <peter.hutterer@who-t.net>
67161Date: Fri Jul 3 16:50:03 2009 +1000
67162
67163 dix: switch to byte-counting functions.
67164
67165 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67166
67167commit 7dd415aa6a3959f15276741db168ba264948ecfe
67168Author: Peter Hutterer <peter.hutterer@who-t.net>
67169Date: Fri Jul 3 13:57:14 2009 +1000
67170
67171 Xi: use byte-counting macros instead of manual calculation.
67172
67173 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67174
67175commit 912402fd71144bcee255141efe12a78abad39240
67176Author: Peter Hutterer <peter.hutterer@who-t.net>
67177Date: Mon Jun 29 13:09:57 2009 +1000
67178
67179 include: introduce byte counting functions.
67180
67181 This patch adds the following three functions:
67182 bits_to_bytes(bits) - the number of bytes needed to hold 'bits'
67183 bytes_to_int32(bytes) - the number of 4-byte units to hold 'bytes'
67184 pad_to_int32(bytes) - the closest multiple of 4 equal to or larger than
67185 'bytes'.
67186
67187 All three operations are common in protocol processing and currently the
67188 server has ((foo + 7)/8 + 3)/4 operations all over the place. A common set
67189 of functions reduce the error rate of these (albeit simple) calculations and
67190 improve readability of the code.
67191
67192 The functions do not check for overflow.
67193
67194 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67195
67196commit 2c535b6f13ffbf2c4ac59834dae39bb8e172c003
67197Author: Peter Hutterer <peter.hutterer@who-t.net>
67198Date: Tue Jul 14 08:58:23 2009 +1000
67199
67200 dix: don't send presence events for attaching/detaching slave devices.
67201
67202 The code that didn't list attached slave devices for XI1 clients doesn't
67203 exist anymore, so there's no need for these presence events.
67204
67205 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67206
67207commit 0c0ef42292f4c910c73b308cd75d77637312da53
67208Author: Peter Hutterer <peter.hutterer@who-t.net>
67209Date: Sun Jul 12 21:43:06 2009 +1000
67210
67211 dix: Remove temporary detachment of slave devices.
67212
67213 Previously, an active grab on an attached slave device would send the device
67214 floating for the duration of the grab. This breaks existing XI applications
67215 (e.g. the GIMP) since they grab all devices automatically - resulting in the
67216 loss of control over the VCP.
67217
67218 The behaviour of extended input devices during a grab in relation to the
67219 core pointer is not specified in the XI protocol specification.
67220 The removal of the temporary detachment restores the behaviour of extended
67221 input devices as present in currently released servers - even if a device is
67222 grabbed, an event from this device will result in an event from the core
67223 pointer.
67224
67225 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67226
67227commit 9f1570c8f4f549cdd2fbae1234011290fcc73e18
67228Author: Peter Hutterer <peter.hutterer@who-t.net>
67229Date: Fri Jun 26 09:12:47 2009 +1000
67230
67231 input: include effective modifiers in device events.
67232
67233 Require inputproto 1.9.99.13
67234
67235 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67236
67237commit 78aedbe609e309446e71835ed15e4557d3ccf723
67238Author: Alan Coopersmith <alan.coopersmith@sun.com>
67239Date: Fri Jul 10 17:35:57 2009 -0700
67240
67241 Fix int10 module build after typo in commit 2638e9899
67242
67243 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
67244
67245commit 35758544813f156eaac28844e693b2a28f6de316
67246Author: Michel Dänzer <daenzer@vmware.com>
67247Date: Fri Jul 10 20:06:02 2009 +0200
67248
67249 EXA: Only pass CT_YXBANDED to RECTS_TO_REGION() if that is really true.
67250
67251 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=22642 .
67252
67253commit 2638e9899e57d9b25a2cade3247083a6e101dee2
67254Author: Tiago Vignatti <tiago.vignatti@nokia.com>
67255Date: Fri Jul 10 16:09:08 2009 +0300
67256
67257 configure: introduce --{enable,disable}-int10-module
67258
67259 Not all chipsets need to rely on the int10 scheme to do its daily work.
67260
67261 Well, the ideal would be to remove all int10 module from the server. I'll try
67262 to provide some patches "soon" for this. Something like:
67263 http://cgit.freedesktop.org/~vignatti/libx86/
67264
67265 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
67266 Acked-by: Oliver McFadden <oliver.mcfadden@nokia.com>
67267
67268commit 42d95a38535f382bfb2ddfaef93839ab9bbf584f
67269Author: Tiago Vignatti <tiago.vignatti@nokia.com>
67270Date: Fri Jul 10 15:54:07 2009 +0300
67271
67272 configure: introduce --{enable,disable}-vbe
67273
67274 Not all drivers need this kind of access as well.
67275
67276 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
67277 Acked-by: Oliver McFadden <oliver.mcfadden@nokia.com>
67278
67279commit 53d64930513fecaa417bb5a922966b45c9ff8679
67280Author: Tiago Vignatti <tiago.vignatti@nokia.com>
67281Date: Fri Jul 10 15:38:51 2009 +0300
67282
67283 configure: introduce --{enable,disable}-vgahw
67284
67285 Not all drivers need this kind of access.
67286
67287 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
67288 Acked-by: Oliver McFadden <oliver.mcfadden@nokia.com>
67289
67290commit aafcae79d17c1f802bc880d2142af7171fed75d8
67291Author: Tiago Vignatti <tiago.vignatti@nokia.com>
67292Date: Fri Jul 10 14:56:31 2009 +0300
67293
67294 xfree86: remove unused vbe header file
67295
67296 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
67297 Acked-by: Oliver McFadden <oliver.mcfadden@nokia.com>
67298
67299commit 3711d68f657c77b947cc4670cc4eac4f62de3af8
67300Author: Peter Hutterer <peter.hutterer@who-t.net>
67301Date: Fri Jul 10 11:10:30 2009 +1000
67302
67303 Revert "XKB: Sanitise * actions" commits (#19602)
67304
67305 Reverts the following four patches:
67306
67307 feb757f384382c7782ceac55 "XKB: Sanitise vmods for redirected keys"
67308 b5f49382fe48f0a762d9a15f "XKB: Sanitise ctrls action"
67309 1bd7fd195d85681e722161f8 "XKB: Sanitise pointer actions"
67310 61c508fa78aa08ea2666fde9 "XKB: Sanitise vmods in actions"
67311
67312 Strictly speaking, the structs used in the server are not part of the client
67313 ABI. Practically, they are as we copy from the wire straight into the
67314 structs. Changing the struct sizes breaks various wire/server conversions.
67315
67316 Even when the structs have the same size, some internal magic causes
67317 conversions to fail. Visible by diffing the output files of:
67318 setxkbmap -layout de; xkbcomp -xkb :0 busted.xkb
67319 setxkbmap -layout de -print | xkbcomp -xkb - correct.xkb
67320
67321 Interestingly enough, busted.xkb is the working one although the output is
67322 incorrect. Revert the four offending patches until the exact cause of this
67323 breakage can be determined.
67324
67325 This patch restores functionality to Level3 modifiers.
67326
67327 X.Org Bug 19602 <http://bugs.freedesktop.org/show_bug.cgi?id=19602>
67328
67329 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67330
67331commit 08df24555cb432eb0d90a3f63275e9485e777c4c
67332Author: Soren Sandmann Pedersen <ssp@redhat.com>
67333Date: Thu Jul 9 17:23:04 2009 -0400
67334
67335 Reserve space for two GC values in copy_drawable().
67336
67337 Pointed out by Pierre Willenbrock.
67338
67339commit 3c53f3241319d96affad2f4539f7661e02e80521
67340Author: Adam Jackson <ajax@redhat.com>
67341Date: Thu Jul 9 17:21:07 2009 -0400
67342
67343 Since font modules are dead, don't mention them in xorg-server.pc
67344
67345commit e812103382a7bc3f7ba8b2f44ab97960dec265a9
67346Author: Adam Jackson <ajax@redhat.com>
67347Date: Thu Jul 9 10:58:17 2009 -0400
67348
67349 config: drop i810.
67350
67351 The driver assuredly no longer builds against 1.7. Please don't feel
67352 compelled to fix it.
67353
67354commit 55ac155d42268009eaecbc183329a8744d6ae48c
67355Author: Alan Coopersmith <alan.coopersmith@sun.com>
67356Date: Tue Jul 7 16:59:39 2009 -0700
67357
67358 Xephyr & Xserver man page fixes
67359
67360 Xephyr(1): Fix quote formatting, add missing ' to contraction
67361 Xserver(1): Add Xephyr(1) & startx(1) to SEE ALSO section
67362
67363 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
67364
67365commit 36dc66ae79b6e1cbe0a23587e32aebb68365b0d0
67366Author: Michael Lorenz <macallan@netbsd.org>
67367Date: Tue Jul 7 20:21:39 2009 -0400
67368
67369 The way XaceHook() mixes struct initializers and va_arg() is not portable and
67370 gives bogus data on sparc and probably others leading to a crash.
67371
67372 Fix:
67373 Don't use initializers, instead set each member directly to enforce order.
67374
67375 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67376
67377commit 48a9d65b88f56d1f8ab3bf824a4fe48c2f68725f
67378Author: Jon TURNEY <jon.turney@dronecode.org.uk>
67379Date: Sun Jul 5 16:06:32 2009 +0100
67380
67381 Cygwin/X: Window positioning improvements for multiwindow mode
67382
67383 A few tweaks to winUpdateWindowPosition():
67384
67385 * Don't allow window decoration to disappear off to top or left edge of the
67386 display as a result of adjustments to add decoration for the window style
67387
67388 * Honour the position requested by window geometry for the client area (so
67389 windows which save their position don't get moved by the decoration width
67390 every time they are created) (unless we need to bump it away from top or left edges)
67391
67392 * Fix an issue with initial window placement being offscreen on multimonitor
67393 systems when some monitors have negative coordinates (are to the left of or
67394 above the primary monitor)
67395
67396 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
67397
67398commit 5aec72745232dd61e60cfbf3acc4628d4fcd0315
67399Author: Keith Packard <keithp@keithp.com>
67400Date: Mon Jul 6 11:53:02 2009 -0700
67401
67402 Remove old DRI2 buffer alloc/free interfaces
67403
67404 These old interfaces are no longer supported by the server, removing them
67405 requires bumping the video driver ABI. Note that this is not guaranteed to
67406 be the last change in ABI version 6.
67407
67408 Signed-off-by: Keith Packard <keithp@keithp.com>
67409 Reviewed-by: Eric Anholt <eric@anholt.net>
67410
67411commit 2e2c5b216cc1c7a9bc26bd2c68226aaed5fc52ca
67412Author: Keith Packard <keithp@keithp.com>
67413Date: Wed Jul 1 14:01:57 2009 -0700
67414
67415 dri2: Preserve compatibility with 1.6 DRI2 API/ABI
67416
67417 The old DRI2 buffer allocation API wasn't great, but there's no reason to
67418 make the server stop working with those drivers. This patch has the
67419 X server adapting to the API provided by the driver, using the new API where
67420 available and falling back to the old API as necessary. A warning will be
67421 placed in the log file when the old API is in use.
67422
67423 Signed-off-by: Keith Packard <keithp@keithp.com>
67424
67425commit 7c7f0c2c6a04f7044d5ce69e97a615735e5831f1
67426Author: Søren Sandmann Pedersen <sandmann@redhat.com>
67427Date: Fri Jul 3 14:36:54 2009 -0400
67428
67429 Use IncludeInferiors when copying windows before compositing.
67430
67431 Part of bug 22484.
67432
67433commit 43ee8d2ead862f84a4526a472519663ef27a8d6a
67434Author: Matthias Hopf <mhopf@suse.de>
67435Date: Wed Jun 24 18:26:23 2009 +0200
67436
67437 Unclaim PCI slot if driver probing fails.
67438
67439 Otherwise no subsequent driver will be able to claim this pci slot.
67440 Example for this: fbdev tries to claim, but framebuffer device is not
67441 available. Later on VESA cannot claim the device.
67442
67443commit 73abdc94c3cceadeda26a9b6bd3cdfecf0df8db2
67444Author: RALOVICH, Kristóf <tade60@freemail.hu>
67445Date: Sun Jul 5 16:42:54 2009 +0200
67446
67447 glx: damage is only used with DRI
67448
67449 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
67450
67451commit 7208a0f032d487bd3bc9809800f4557f1da1e2db
67452Author: RALOVICH, Kristóf <tade60@freemail.hu>
67453Date: Mon Jun 29 15:18:56 2009 +0200
67454
67455 glx: remove Xgl leftover
67456
67457 GlxSetRenderTables was only used by the long gone Xgl.
67458
67459 Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
67460
67461commit 89cf81cd85919e3dbb5adff5e6c6056c7990b60f
67462Author: Peter Hutterer <peter.hutterer@who-t.net>
67463Date: Fri Jul 3 10:29:40 2009 +1000
67464
67465 Xext: return BadValue for XTestFakeInput on unsupported capabilities.
67466
67467 Calling XTestFakeDevice*Event on a device that doesn't allow the matching
67468 event returns BadValue.
67469
67470 Reported-by: Florian Echtler
67471 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67472
67473commit b773b4e8e0087993406d1d8c2df895db4d6301ff
67474Author: Peter Hutterer <peter.hutterer@who-t.net>
67475Date: Mon Jul 6 11:04:47 2009 +1000
67476
67477 s/MAX_DEVICES/MAXDEVICES/ updates.
67478
67479 The number of input devices is MAXDEVICES, not MAX_DEVICES (f781a752e6)
67480 Two comments updated to refer to MAXDEVICES.
67481
67482 MAX_FUNCS in sigio.c was set to 16 if MAX_DEVICES was undefined. If more
67483 than 15 physical input devices were present, this could result in a
67484 failure to install the SIGIO handler for any device above 15.
67485
67486 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67487
67488commit 183c075d2f9d5f6effa1ce7ab135fb0c2e46085a
67489Author: Peter Hutterer <peter.hutterer@who-t.net>
67490Date: Fri Jul 3 13:26:48 2009 +1000
67491
67492 dix: always init the full button map to default values (#22594)
67493
67494 Master devices must have the standard button map applied for all buttons to
67495 ensure buttons larger than 7 (the default for MDs) are mapped appropriately.
67496
67497 We can't copy the button map from SDs to MDs since that breaks the chained
67498 button mapping. However, by ensuring all buttons (even non-existing ones)
67499 are mapped, devices that send such buttons continue to work.
67500
67501 X.Org Bug 22594 <http://bugs.freedesktop.org/show_bug.cgi?id=22594>
67502
67503 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67504
67505commit 744bb559826ede37a77e9000b6c620eaa6a3c837
67506Author: Peter Hutterer <peter.hutterer@who-t.net>
67507Date: Fri Jul 3 13:58:38 2009 +1000
67508
67509 Xext: remove unused variable 'it'.
67510
67511 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67512
67513commit fd4eed69e40cf090aa16e79a8596eb1442461829
67514Author: Eric Anholt <eric@anholt.net>
67515Date: Thu Jul 2 15:24:54 2009 -0700
67516
67517 dri2: Enable GLX_SGI_make_current_read when the DRI driver supports it.
67518
67519 This matches idr's 82f150d73cc9a7d8eaf7241a51b03af05bcec159 for DRI1.
67520
67521commit d6b8205e699c0c62af76c4a9cbff1402337927b3
67522Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
67523Date: Fri Jul 3 19:25:33 2009 -0700
67524
67525 XQuartz: ProcAppleWMAttachTransient to play nice with the new Dock in SL
67526 (cherry picked from commit ddc0242d8f291d0f961ba9aa8c883e39861ce2ad)
67527
67528commit a49ae50370ec94e08e7dec3c742d33e20e79ef36
67529Author: Colin Harrison <colin.harrison@virgin.net>
67530Date: Mon Jun 29 22:55:52 2009 +0100
67531
67532 Xming: Apply window style hints in -multiwindow mode
67533
67534 Remove frames from 'dock' windows and make them topmost in -multiwindow mode.
67535 Remove frames from windows with MOTIF_WM_HINTS of no decorations in -multiwindow mode.
67536 Apply some _NET_WM_STATE hints in -multiwindow mode.
67537 Apply window styles overrides from .rc file
67538 Correctly update region of shaped windows when applying styles
67539
67540 Copyright (C) Colin Harrison 2005-2009
67541 http://www.straightrunning.com/XmingNotes/
67542 http://sourceforge.net/projects/xming/
67543
67544commit 17e67c407d130c325d3899c18d68b8eef6a88bea
67545Author: Joe Krahn <jkrahn@nc.rr.com>
67546Date: Wed Jun 24 23:32:09 2009 +0100
67547
67548 Cygwin/X: Change to a single native window class for all X windows
67549
67550 from fd.o Bugzilla #4491:
67551
67552 There is no point in having one class for every window, aside from trying to
67553 set custom icons via the class, which we no longer do, so converted to using
67554 a single class for all client windows.
67555
67556 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
67557
67558commit 2c69deb92e11542f615df0f24fdc03e3b4415475
67559Author: Rémi Cardona <remi@gentoo.org>
67560Date: Fri Jul 3 10:51:50 2009 +0200
67561
67562 configure: libXinerama isn't needed anymore
67563
67564 since libXinerama commit 90d4d23bf2e94721149ddc0a80093b10a82e8845 and
67565 xineramaproto commit 21477147613c28c968b5e1eb9d8aea7017dd399d, the
67566 server no longer needs libXinerama.
67567
67568 Signed-off-by: Rémi Cardona <remi@gentoo.org>
67569
67570commit 97e29ffb5bf04a9d8280f1fa32ceced148503492
67571Author: Peter Hutterer <peter.hutterer@who-t.net>
67572Date: Fri Jul 3 09:52:04 2009 +1000
67573
67574 xfree86: fix wrong IsMaster() check causing crashes.
67575
67576 Crashes caused by dereferencing NULL if the path was executed for a floating
67577 slave device.
67578
67579 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67580
67581commit 50a2a8dc76645d8736f7d712f0ef05f23089407e
67582Author: Peter Hutterer <peter.hutterer@who-t.net>
67583Date: Fri Jul 3 09:18:57 2009 +1000
67584
67585 Fix IsXtstDevice - returns false positives since 0814f511d5.
67586
67587 Missing check for the value of 'mid' returned false positives if master was
67588 NULL.
67589
67590 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67591
67592commit 3525d140567e0ad5f0184e4b37893c47239e1628
67593Author: Felix Kuehling <felix.kuehling@amd.com>
67594Date: Tue Jun 30 20:07:30 2009 -0400
67595
67596 Export symbols needed by the RandR implementation in fglrx
67597
67598 GIT change
67599 http://cgit.freedesktop.org/xorg/xserver/commit/?id=45c8bd0fe54273039fdaa1eeeafb81b5774f2c75
67600 changed the default symbol visibility of the Xserver. As a result 2 symbols
67601 that are needed by the RandR 1.2/1.3 implementation in the fglrx driver are no
67602 longer visible:
67603
67604 xf86configptr
67605 xf86CursorScreenKey
67606
67607 We would like to get these two symbols _X_EXPORT'ed before Xserver 1.7 is
67608 released. Otherwise it will be problematic for fglrx to support RandR 1.3 on
67609 Xserver 1.7.
67610
67611 In the future, we may want to sync our RandR implementation to later versions
67612 of the RandR implementation in hw/xfree86/modes. Therefore it would be nice if
67613 all symbols used by the Xserver RandR implementation were _X_EXPORT'ed in the
67614 future.
67615
67616commit e13605ea40cfc671314a0086c75b917564298b55
67617Author: Peter Hutterer <peter.hutterer@who-t.net>
67618Date: Tue Jun 23 14:09:22 2009 +1000
67619
67620 dix: introduce "Xtst Device" label property.
67621
67622 Xtst devices get this property assigned automatically so they can be
67623 detected easily by a client.
67624 The property is read-only.
67625
67626 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67627
67628commit 0814f511d56a89c7b1868b17eba7a89f990b9ab1
67629Author: Peter Hutterer <peter.hutterer@who-t.net>
67630Date: Tue Jun 23 10:50:52 2009 +1000
67631
67632 input: store the master device's ID in the devPrivate for XTest devices.
67633
67634 Rather than storing a simple boolean in the devPrivate for XTest devices,
67635 store the actual master device's id (since it is constant for the life of
67636 the device anyway).
67637
67638 Callers should use GetXtstDevice now instead of digging around in the
67639 devPrivates themselves.
67640
67641 This patch allows for a cleanup in the creation of new master devices since
67642 GetMaster and GetXtstDevice spare the need for loops, IsPointer checks and
67643 similar.
67644
67645 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67646 Acked-by: Benjamin Close <Benjamin.Close@clearchain.com>
67647
67648commit 1bcc0d3c244ce7d9f5cbab626aa5fd5784b41a1c
67649Author: Peter Hutterer <peter.hutterer@who-t.net>
67650Date: Mon Jun 22 16:55:12 2009 +1000
67651
67652 input: abstract Xtst device lookup
67653
67654 The callers should need to use the dev privates key to look up xtest
67655 devices.
67656
67657 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67658 Acked-by: Benjamin Close <Benjamin.Close@clearchain.com>
67659
67660commit 34424fab9abd7a4ca11036be25414129980db0e0
67661Author: Peter Hutterer <peter.hutterer@who-t.net>
67662Date: Tue Jun 23 11:50:29 2009 +1000
67663
67664 dix: set the client's error value to the bad deviceid in check_butmap_change.
67665
67666 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67667
67668commit da04e8f1354fa9d253443489dc002b16f94d6fab
67669Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
67670Date: Tue Jun 30 17:52:42 2009 -0400
67671
67672 dix/property.c: use memcpy where appropriate.
67673
67674 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67675
67676commit 65d74d93145d22b68bad5728a7ebe38dc662cb21
67677Author: Jon TURNEY <jon.turney@dronecode.org.uk>
67678Date: Wed May 20 21:32:54 2009 +0100
67679
67680 Cygwin/X: Fix multiwindow extwm mode to build again
67681
67682 Build and link with rootless extension
67683 Update Xwin code for removal of RootlessAccelInit()
67684 Fix Xwin code which now has a collision with the type name EventType
67685
67686 Based on patches from Colin Harrison, Jon Turney and Yaakov Selkowitz
67687
67688 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
67689
67690commit 638ca9a7a2363757dc5b5d456e10d34f6f158885
67691Author: Jon TURNEY <jon.turney@dronecode.org.uk>
67692Date: Mon May 18 18:14:41 2009 +0100
67693
67694 Cygwin/X: Avoid a visual glitch on window move in rootless modes
67695
67696 Handle and ignore WM_ERASEBKGND since we repaint the entire invalidated region anyhow
67697 (this avoids a white flickering on window resize)
67698
67699 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
67700
67701commit b718b2e0880cf2b969675da98d5ef8a4a01ca5d6
67702Author: Jon TURNEY <jon.turney@dronecode.org.uk>
67703Date: Mon May 18 18:14:09 2009 +0100
67704
67705 Cygwin/X: Allow pointer warping to work in rootless modes
67706
67707 Mouse pointer warping in multiwindow/rootless mode was never implemented,
67708 due to concerns that moving the mouse pointer without asking might be rude
67709
67710 This patch allows X applications to move the mouse pointer in rootless modes,
67711 Let's hope they don't abuse this privilege ;-)
67712
67713 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
67714
67715commit 85614946ba3d5a233eece612afc7f09572a347c2
67716Author: Jon TURNEY <jon.turney@dronecode.org.uk>
67717Date: Mon May 18 18:13:08 2009 +0100
67718
67719 Cygwin/X: Improve mouse tracking for moving/resizing undecorated windows
67720
67721 In -multiwindow mode, tell Windows we wish to capture the mouse when a button
67722 is down. This causes Windows to continue to send movement events for the mouse
67723 even if the mouse pointer moves outside the window frame.
67724
67725 This helps greatly with undecorated windows which have regions you can grab
67726 to move (e.g. gmplayer, xine control panels) or resize (e.g. Songbird) the
67727 window, as it means the window continues to receive the mouse motion even if the
67728 mouse pointer moves out of the window (which presumably happens if we don't
67729 manage to update the window fast enough to track the mouse pointer)
67730
67731 Consolidate the multiple instances of the code to start the mouse position
67732 polling timer into a new function winStartMousePolling(), and use that to
67733 restart the polling timer when we release the mouse.
67734
67735 Also, start the timer on WM_SHOW, so that xeyes will track the mouse position
67736 when it is first shown, even if the mouse doesn't enter it's window
67737 (You probably need focus-stealing turned off to see this problem)
67738
67739 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
67740
67741commit f351c10a9774cc0ea2cbb58f00f07ece7f7c6e73
67742Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
67743Date: Mon Jan 5 16:17:30 2009 +0000
67744
67745 Cygwin/X: Man page fixes
67746
67747 Correct path names in man pages, using cpprules.in
67748 Install XWinrc man page into section 5
67749
67750 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
67751
67752commit 5c1afac5eea1d8327c74342d12d082b75f0cebde
67753Author: Adam Jackson <ajax@redhat.com>
67754Date: Mon Jun 29 11:16:16 2009 -0400
67755
67756 EDID: Fix timing class names to match the spec
67757
67758commit bf0f3b8f2bc830c8bd6f8f9410b89394b8d96257
67759Author: Peter Hutterer <peter.hutterer@who-t.net>
67760Date: Mon Jun 29 13:31:02 2009 +1000
67761
67762 test: fix build error introduced by XINPUT_ABI 7
67763
67764 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67765
67766commit cbeb6a73c44143674a700d36a7e9804d41003a4f
67767Author: Thomas Jaeger <ThJaeger@gmail.com>
67768Date: Sat Jun 20 20:17:41 2009 -0400
67769
67770 dix: report subpixel coordinates for high-resolution devices
67771
67772 Acked-by: Simon Thum <simon.thum@gmx.de>
67773 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67774
67775commit 069b4c5f0a38649de73d1e0f70ee81bf862399dd
67776Author: Simon Thum <simon.thum@gmx.de>
67777Date: Sat Jun 20 18:57:22 2009 +0200
67778
67779 dix: make part of ptrveloc.h internal
67780
67781 Though this is a SDK header, some functions are intended solely
67782 for the server.
67783
67784 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67785
67786commit 2830e8493757e0da1253fe5ab280562b84730e77
67787Author: Simon Thum <simon.thum@gmx.de>
67788Date: Wed Jun 24 11:33:19 2009 +0200
67789
67790 dix: rename pDev->dev, pVel->vel for consistency
67791
67792 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67793
67794commit 373e8c960d00e2b8c2250dd3f66859b081e14854
67795Author: Simon Thum <simon.thum@gmx.de>
67796Date: Wed Jun 24 11:16:24 2009 +0200
67797
67798 dix: improve pointer acceleration API
67799
67800 This makes the ptr accel api actually sensible from a driver
67801 perspective, since it avoids superfluous device lookups.
67802 Also, makes independent accel contexts possible.
67803
67804 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67805
67806commit b8050bb6deebdb1ee60731f63884ffca575c09ce
67807Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
67808Date: Sat Jun 27 15:28:34 2009 -0700
67809
67810 XQuartz: Don't leave zombied processes at startup
67811 (cherry picked from commit 40c1406830588fa85d880e9f4e9ca570db1db306)
67812
67813commit aaff92c8c22a47804a21010d023ef76d82e7ec7e
67814Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
67815Date: Sat Jun 27 14:40:03 2009 -0700
67816
67817 XQuartz: 64bit fix for screen origin in AppleWM
67818 (cherry picked from commit 60a757d2802a5c34acd91ca9a052937b5a169ede)
67819
67820commit cabcc1a7ad697dde915794ddcea6a300f66b0a28
67821Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
67822Date: Sat Jun 27 10:40:59 2009 -0700
67823
67824 XQuartz: More localization updates
67825 (cherry picked from commit 5925c1f48ad05bf94195b986c1fdefc52a20ae42)
67826
67827commit 31166c2ae0ce898c96995a8b16b58b127dc85a2f
67828Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
67829Date: Fri Jun 26 16:51:22 2009 -0400
67830
67831 xace: add a new hook for checking property content after it has been set.
67832
67833 Allows security modules to enforce what property contents can be set by
67834 clients. Uses the new DixPostAccess bit to distinguish between the
67835 existing call made during the lookup (with the old property data) and
67836 this new call. Note that this only applies to writes, prepends, or
67837 appends to existing properties; for new properties the existing
67838 DixCreateAccess hook call may be used since it includes the new data.
67839
67840 Refer to the XACE-Spec document in xorg-docs, section "Property Access."
67841
67842 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67843
67844commit 51105de9b0d865c4b5e5a7d9ab23c89d808d1cfa
67845Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
67846Date: Fri Jun 26 16:50:12 2009 -0400
67847
67848 xselinux: ignore property hook calls with the new Post access mode bit set.
67849
67850 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67851
67852commit 10812204b3415c969bcebd3215e84d758a0b4dd8
67853Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
67854Date: Fri Jun 26 16:48:24 2009 -0400
67855
67856 dix: add a new DixAccess bit, "DixPostAccess".
67857
67858 This will be used for follow-up checks after a client has written something,
67859 for security modules that enforce a set of valid values a client can set.
67860
67861 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67862
67863commit 9480725af0275b72b24b9e1cb3b68d5b97288d01
67864Author: Rémi Cardona <remi@gentoo.org>
67865Date: Wed Jun 24 10:12:50 2009 +0200
67866
67867 xfree86: DIST_SUBDIRS should not contain variables
67868
67869 This patch reverts a change done in commit
67870 800d4ceb946c93724661ebfd28afb658ca63b6b1 to add conditional XAA support
67871
67872 Signed-off-by: Rémi Cardona <remi@gentoo.org>
67873
67874commit 84662e40c3d4141ebb298a1ad714f75056a4ab74
67875Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
67876Date: Thu Jun 18 23:19:21 2009 -0400
67877
67878 Xi: check for GetAttr permission when listing or querying devices.
67879
67880 If the check fails, leave the device off the returned list of info
67881 structures. Under XI2, this may cause inconsistent views of the device
67882 topology after a change (for example, devices disappearing from view,
67883 or showing as attached to a master that cannot be seen). More work is
67884 needed to deal with topology changes and device relabeling.
67885
67886 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67887
67888commit 00bc8d34c68dab6c818cd1c7e03e9992d1d0cbfc
67889Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
67890Date: Thu Jun 18 21:41:17 2009 -0400
67891
67892 Xi: check for Use permission on the device in SetClientPointer().
67893
67894 Presumably, some intelligent, XI2-aware management app will be calling
67895 XISetClientPointer on behalf of other clients; this check makes sure
67896 the target client has permission on the device.
67897
67898 Requires changing the prototype to return status code instead of Bool.
67899
67900 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67901 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
67902
67903commit 119b96667778391436998c76a68bf64e746c9e08
67904Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
67905Date: Thu Jun 18 21:02:28 2009 -0400
67906
67907 Xi: fix up access modes for calls to dixLookupDevice().
67908
67909 New access modes are being passed to the device access hook for XI2:
67910 DixCreateAccess for creating a new master device;
67911 DixAdd/RemoveAccess for attaching/removing slave devices to a master; and
67912 DixListProp/GetProp/SetPropAccess for device properties.
67913
67914 Refer to the XACE-Spec document in xorg-docs, section "Device Access."
67915
67916 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
67917
67918commit 07c36e4fdcd93df3d33bdab6cca4780ebc9c1f54
67919Author: Dave Airlie <airlied@redhat.com>
67920Date: Wed Jun 10 13:33:47 2009 +1000
67921
67922 dix/resource: fix use after free in resource code with DRI
67923
67924 LookupClientResourceComplex is used by DRI1 code to find and free a DRI
67925 drawable in a callback, however when the DRI code returns this->value
67926 is now pointing at freed memory. It seemed easiest to store the value
67927 to a temporary and return it afterwards.
67928
67929 Another option might be a new FreeClientResourceComplex or one that
67930 also returns the id, so we can free it using an alternative means.
67931
67932 found using valgrind.
67933
67934 amended along ajax's suggestions
67935
67936commit 184deb9bc325eb7aa7eb7b7d4f98aa917f0269cb
67937Author: Dave Airlie <airlied@redhat.com>
67938Date: Wed Jun 24 10:40:05 2009 +1000
67939
67940 GLX: make function static.
67941
67942 This function isn't called from anywhere else and I don't think it shuold be.
67943
67944commit 9d85b56078ec05da1369ca22930d8eb214c389db
67945Author: Dave Airlie <airlied@redhat.com>
67946Date: Wed Jun 24 10:38:49 2009 +1000
67947
67948 GLX: note the implicit flushes with ReadPixels in indirect contexts.
67949
67950 This just notes the flush has occured when readpixels returns, and
67951 fixes the glean test.
67952
67953commit e341512bfa40dd98853a20596dc65dcac4dcaa37
67954Author: Thomas Jaeger <ThJaeger@gmail.com>
67955Date: Sat Jun 20 21:37:59 2009 -0400
67956
67957 dix: update a comment
67958
67959 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67960
67961commit 5cbd4d3d6e54d202ecdbb527b57aaefeb8435600
67962Author: Thomas Jaeger <ThJaeger@gmail.com>
67963Date: Sat Jun 20 20:17:04 2009 -0400
67964
67965 dix: do away with an instance of temporary in-place modification
67966
67967 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67968
67969commit 94cdc1ef0a72802573a11ba292e2f5cc8474aa8e
67970Author: Thomas Jaeger <ThJaeger@gmail.com>
67971Date: Mon Jun 22 13:00:37 2009 -0400
67972
67973 dix: deal with first_valuator > 0 correctly if POINTER_SCREEN is set
67974
67975 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
67976
67977commit 21cbb4c9de44e2629dc0fa6f647ce2d139f2cef0
67978Author: Oliver McFadden <oliver.mcfadden@nokia.com>
67979Date: Wed Jun 24 00:24:44 2009 +0300
67980
67981 xorg-server.h.in: Export the X Access Control Extension (XACE), too.
67982
67983 Acked-by: Daniel Stone <daniel@fooishbar.org>
67984
67985commit 800d4ceb946c93724661ebfd28afb658ca63b6b1
67986Author: Tiago Vignatti <tiago.vignatti@nokia.com>
67987Date: Wed Jun 24 00:13:32 2009 +0300
67988
67989 configure: Provide the --enable/disable-xaa option.
67990
67991 Acked-by: Daniel Stone <daniel@fooishbar.org>
67992
67993commit ce3d539ac6ecab3e343cef7c4dc112bc82ea3b02
67994Author: Oliver McFadden <oliver.mcfadden@nokia.com>
67995Date: Wed Jun 24 00:08:51 2009 +0300
67996
67997 xf86Config: Avoid attempting to load non-compiled modules.
67998
67999 Acked-by: Daniel Stone <daniel@fooishbar.org>
68000
68001commit 33e25143361d63ea776dfafeeb6d4ffb2f6fd8d4
68002Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
68003Date: Fri May 15 22:42:30 2009 +0100
68004
68005 Cygwin/X: add hebrew to autodetected keyboard layouts
68006
68007 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
68008
68009commit b079945c39e2a72220b46953352e24c3a74ef39a
68010Author: Jon TURNEY <jon.turney@dronecode.org.uk>
68011Date: Fri Feb 20 15:21:35 2009 +0000
68012
68013 Cygwin/X: more warnings fixes
68014
68015 More warning fixes, mainly removing casts on function return values
68016
68017 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
68018
68019commit e79db6a97d02c8a256a4a7e145ea0b48b5a084ab
68020Author: Jon TURNEY <jon.turney@dronecode.org.uk>
68021Date: Wed Feb 11 23:00:58 2009 +0000
68022
68023 Cygwin/X: Remove an obsolete mention of xf86Config
68024
68025 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
68026
68027commit 3020b1d43e34fca08cd51f7c7c8ed51497d49ef3
68028Author: Michel Dänzer <daenzer@vmware.com>
68029Date: Tue Jun 23 16:45:40 2009 +0200
68030
68031 glx: Clean up more thoroughly if the drawable of a current context goes away.
68032
68033 Fixes crash when restarting compiz, due to cl->currentContexts[x] being stale.
68034
68035commit df597709d71f47b8516e27c6fb1bfffd59de5e48
68036Author: Michel Dänzer <daenzer@vmware.com>
68037Date: Tue Jun 23 16:45:39 2009 +0200
68038
68039 dri2: Don't crash if pPriv is NULL.
68040
68041commit 048697ccfa31cf7f7a29afa90a2f702d43efb7d4
68042Author: Ben Skeggs <bskeggs@redhat.com>
68043Date: Wed May 27 15:24:34 2009 +1000
68044
68045 quirk: use first detailed timing as preferred for PEA prod 9003 (rh#492359)
68046
68047 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68048
68049commit 07154db4a8b96467785fd6be93d20379acacffa8
68050Author: Ben Gamari <bgamari.foss@gmail.com>
68051Date: Mon Jun 22 08:59:45 2009 -0400
68052
68053 Xi: Remove redundant and incorrect butmap range check
68054
68055 Maps are CARD8s, therefore checking for values above 255 is completely
68056 unnecessary. Moreover, 0 is a valid value for maps, so the check wasn't
68057 even correct to begin with. This fixes bug #22392, which was uncovered
68058 by commit 280b7f92d729ec910ffa3d18dce7bbc215be7a3c.
68059
68060 Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
68061 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68062
68063commit 96706c24bd57fbd9b11e5bd5e38d05d81b90aebe
68064Author: Peter Hutterer <peter.hutterer@who-t.net>
68065Date: Thu Jun 18 15:40:08 2009 +1000
68066
68067 dix: fix wrong indices in set_valuator.
68068
68069 Reported-by: Thomas Jaeger
68070 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68071
68072commit ae20e748cd3a656173e1f50109bfd4af0712bb87
68073Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
68074Date: Fri Jun 19 21:16:10 2009 -0700
68075
68076 XQuartz: More localization updates
68077 (cherry picked from commit 7fbe974246f54535c545861a57c043d80e127ee7)
68078
68079commit 30df49f54945e75f033a0eb6445c26d37eb33c5e
68080Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
68081Date: Thu Jun 18 23:15:47 2009 -0400
68082
68083 Xi: fix 2 memory leaks.
68084
68085 In ProcXIQueryDevice() and XISendDeviceHierarchyEvent().
68086
68087 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
68088
68089commit eb35402d0a5290e8a73d7d1e92f173294c364cc2
68090Author: Adam Jackson <ajax@redhat.com>
68091Date: Fri Jun 19 12:42:07 2009 -0400
68092
68093 pci: Dump vendor/devices ids in the printed device list
68094
68095commit c733660428c0a7c1d11f7bd21e23e1bb934d352e
68096Author: Søren Sandmann Pedersen <sandmann@redhat.com>
68097Date: Fri Jun 19 08:37:18 2009 -0400
68098
68099 Use pixman_version_string() instead of PIXMAN_VERSION_STRING
68100
68101 Pointed out by Julien Cristau.
68102
68103commit aef6b904ebf0d7de6259058606c7c04ea177bda3
68104Author: Dave Airlie <airlied@panoply-rh.(none)>
68105Date: Thu Mar 13 16:16:46 2008 +1000
68106
68107 fbdev: make entity fail if PCI claimed already.
68108
68109 bad kitty fbdev, been shipping this in Fedora for a while now
68110
68111commit 3efb23a6c40dc3583d083d25ada3853ecc56000d
68112Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
68113Date: Thu Jun 18 19:35:22 2009 -0400
68114
68115 xselinux: Add new device permissions for XI2.
68116
68117 Refects the ability of clients to add/remove devices and device properties.
68118
68119 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
68120
68121commit 1e060c3d8b13d352a58fc65980cb9a3c6cb5718f
68122Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
68123Date: Thu Jun 18 18:48:24 2009 -0400
68124
68125 xselinux: Move the security class mapping to the header file.
68126
68127 Take the mapping of DixAccess bits to Flask permissions, move it
68128 into the header file, break up the extremely long lines, and
68129 annotate the permission names with the bit being referenced.
68130
68131 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
68132
68133commit 75c51c67b340548286efd41a53882e2acaf74ab5
68134Author: Alan Coopersmith <alan.coopersmith@sun.com>
68135Date: Thu Jun 18 09:49:12 2009 -0700
68136
68137 Clarify use of and need for mffs vs. ffs
68138
68139 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
68140
68141commit 128cd03eecacc6d5c5903d59a11966dcf3697bf1
68142Author: Søren Sandmann Pedersen <sandmann@redhat.com>
68143Date: Sat Jun 13 10:55:04 2009 -0400
68144
68145 Fix miComputeCompositeRegion() to follow new clip rules.
68146
68147 Ignore the hierarchy clip, and always apply any client clip after
68148 transformation and repeating.
68149
68150commit e9aa61e9f0d663d5b34a397b943b4d1df44e873d
68151Author: Søren Sandmann Pedersen <sandmann@redhat.com>
68152Date: Sat Jun 13 10:28:21 2009 -0400
68153
68154 Fix clipping when windows are used as sources
68155
68156 The new clipping rules:
68157
68158 - client clips happen after transformation
68159 - pixels unavailable due to the hierarchy are undefined
68160
68161 The first one is implemented in pixman; the second one is realized by
68162 making a copy of window sources (to prevent out-of-bounds access).
68163
68164commit d9b5e77a0e48a16c53653b56bc61a0b8dc4122a1
68165Author: Søren Sandmann Pedersen <sandmann@redhat.com>
68166Date: Tue Jun 9 14:36:21 2009 -0400
68167
68168 Print the current version of pixman.
68169
68170commit 1e9907499c27321a2aa5dc8a75a375b7a82c999a
68171Author: Peter Hutterer <peter.hutterer@who-t.net>
68172Date: Thu Jun 18 15:14:00 2009 +1000
68173
68174 record: use dixLookupResourceByClass instead of LookupIDByClass.
68175
68176 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68177
68178commit 66089e9129a821cfb1983d3d35f41b975a52de5e
68179Author: Peter Hutterer <peter.hutterer@who-t.net>
68180Date: Wed Mar 25 12:55:42 2009 +1000
68181
68182 xfree86: fix SWCursor check in xf86CursorSetCursor.
68183
68184 Wrong check for inputInfo.pointer resulted in a SW cursor being rendered when
68185 the pointer left the screen (in a Xinerama setup).
68186 We must call the sprite rendering function if
68187 - SW cursors are enabled, or
68188 - The current device is not the VCP and not attached to the VCP.
68189
68190 Reported-by: Gordon Yuan <GordonYuan@viatech.com.cn>
68191 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68192
68193commit afa680e495622f521cae80563511c0d284f57551
68194Author: David Miller <davem@davemloft.net>
68195Date: Thu Jun 11 05:15:05 2009 -0700
68196
68197 mi: ignore DGA events in ChangeDeviceID
68198
68199 DGA events don't have a deviceid, so they don't need changing.
68200
68201 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68202
68203commit cd8abd17abcc8ae9da6704d03b97a9f134f36aab
68204Author: Jon TURNEY <jon.turney@dronecode.org.uk>
68205Date: Wed Jun 17 14:17:07 2009 +0100
68206
68207 Cygwin/X: Fix permuted args to InitPointerDeviceStruct()
68208
68209 Fix a minor error in commit a30fef9956b296f59ea18a9ee38d0abafeb15a4e,
68210 new btn_labels argument to InitPointerDeviceStruct() wasn't added in
68211 the right place
68212
68213commit 14581afb474552716c02ca15220ca7050123c375
68214Author: Benjamin Close <Benjamin.Close@clearchain.com>
68215Date: Thu Feb 26 17:32:10 2009 +1030
68216
68217 xfree86: correctly define barriers for FreeBSD amd64
68218
68219 Previously when compiling on freebsd amd64 we'd end up at xi86
68220 block (line 1315) which would define mem_barrier and write_mem_barrier
68221 to be NOP's. Instead they should be valid, as per the linux amd64 setup.
68222
68223 This stops the hangs experienced by many when using the nv driver
68224 which would hang due to out of order dma requests as noticed in
68225 http://bugs.freedesktop.org/show_bug.cgi?id=3168
68226
68227 Signed-off-by: Benjamin Close <Benjamin.Close@clearchain.com>
68228
68229commit e92dcb6ce07aa3cfb53e8bad5701481c106c4094
68230Author: Peter Hutterer <peter.hutterer@who-t.net>
68231Date: Thu Jun 18 14:40:20 2009 +1000
68232
68233 input: unify button numbers on master devices.
68234
68235 Master devices provide the union of all attached slave devices' buttons,
68236 i.e. the number of buttons on the master device is always the number of
68237 buttons of the slave device with the highest number of buttons. When slaves
68238 are attached or detached, the master device adjusts the button number to
68239 reflect the new buttons.
68240
68241 On a slave switch, this slave's button labels are copied into the master (up
68242 to slave->num_buttons). The remaining button labels (if any) stay as they
68243 are. Thus, if any of the higher buttons is still pressed, it reflects the
68244 label of the last pressed device that provided this button.
68245
68246 If two devices press the same button and it is differently labelled the last
68247 pressed one will be reflected in the master device.
68248
68249 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68250
68251commit 280b7f92d729ec910ffa3d18dce7bbc215be7a3c
68252Author: Peter Hutterer <peter.hutterer@who-t.net>
68253Date: Wed Jun 17 22:38:24 2009 +1000
68254
68255 dix: reduce MDs and xtest pointers to 7 buttons by default.
68256
68257 MD's will soon be the union of all devices anyway. XTest pointers are only
68258 for the core protocol XTest stuff, so 7 buttons (lmr + 4 wheel buttons)
68259 should do.
68260
68261 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68262
68263commit 6c7d992735eebbd7a20247926e7725896348b865
68264Author: Peter Hutterer <peter.hutterer@who-t.net>
68265Date: Wed Jun 17 09:11:52 2009 +1000
68266
68267 Require inputproto 1.9.99.12
68268
68269 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68270
68271commit 09cef7573938e5c08007e578e1b638bc5e1796a8
68272Author: Peter Hutterer <peter.hutterer@who-t.net>
68273Date: Wed Jun 17 09:04:08 2009 +1000
68274
68275 Xi: valuator/button labels are called labels now, not 'names'
68276
68277 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68278
68279commit bc2ff5365030ad8bc11efde430b1064080dd7098
68280Author: Peter Hutterer <peter.hutterer@who-t.net>
68281Date: Tue Jun 16 19:36:57 2009 +1000
68282
68283 Xi: copy the valuator mode from SD to MD.
68284
68285 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68286
68287commit 87d1f44bad608507e3995e17eb84fa0a0119796c
68288Author: Peter Hutterer <peter.hutterer@who-t.net>
68289Date: Tue Jun 16 14:18:45 2009 +1000
68290
68291 Xi: copy the button and axes labels into the XIQueryDevice reply.
68292
68293 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68294
68295commit a30fef9956b296f59ea18a9ee38d0abafeb15a4e
68296Author: Peter Hutterer <peter.hutterer@who-t.net>
68297Date: Tue Jun 16 16:38:11 2009 +1000
68298
68299 input: Add labels to buttons and valuators - ABI_XINPUT_VERSION 7
68300
68301 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68302
68303commit 17f9723f488d0470e3879c6b0dfdba61544cdd7b
68304Author: Peter Hutterer <peter.hutterer@who-t.net>
68305Date: Wed Jun 17 08:37:44 2009 +1000
68306
68307 input: bump to ints for deviceids - XI2 requires 16-bit deviceids.
68308
68309 Note: ABI break, but ABI_XINPUT_VERSION has NOT been bumped. Recompile input
68310 drivers.
68311
68312 Revert "Xi: return BadImplementation for deviceids 256 and above"
68313 This reverts commit 2b459f44f3edaea137df9a28bc7adfeb1b9f1df7.
68314
68315 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68316
68317commit cbeb73e2055f6c013c8fe6325851f2631170137d
68318Author: Peter Hutterer <peter.hutterer@who-t.net>
68319Date: Tue Jun 16 10:51:52 2009 +1000
68320
68321 Xi: return current valuator values in XIQueryDevice.
68322
68323 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68324
68325commit 80837dbefd9d5e96ab5c1f4b4c2d5c66ce17ce67
68326Author: Peter Hutterer <peter.hutterer@who-t.net>
68327Date: Tue Jun 16 10:51:17 2009 +1000
68328
68329 input: change axisVal from uint to double.
68330
68331 With subpixel support, uint just doesn't cut it.
68332
68333 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68334
68335commit 25b6fc4a42f7698e6ae0b16becb316bfd7835d05
68336Author: Peter Hutterer <peter.hutterer@who-t.net>
68337Date: Tue Jun 16 10:48:48 2009 +1000
68338
68339 Xi: last_valuator is used like an index, so range it accordingly.
68340
68341 The previous code would always skip the last valuator due to a wrong
68342 upper boundary in the loop. last_valuator is the index of the last set
68343 valuator - which also means it must be initialized to -1, not 0.
68344
68345 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68346
68347commit 3fc762ea9474cf92f6b7ef60dd0bb68f59385272
68348Author: Peter Hutterer <peter.hutterer@who-t.net>
68349Date: Tue Jun 16 10:48:03 2009 +1000
68350
68351 Xi: Motion events update the device state too.
68352
68353 Without this line the device's axis values don't get updated properly for
68354 pure motion events.
68355
68356 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68357
68358commit ad6347afead500fefb34131841f4beb5b03ad4ec
68359Author: Peter Hutterer <peter.hutterer@who-t.net>
68360Date: Tue Jun 16 16:35:40 2009 +1000
68361
68362 Xi: don't use a constant number for class sizes - use sizeof instead
68363
68364 The protocol is still changing, and having to debug crap like this is
68365 annoying.
68366
68367 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68368
68369commit d5a15e65b7d2fdda14fb719e3ebf076cce896386
68370Author: Peter Hutterer <peter.hutterer@who-t.net>
68371Date: Sun Jun 14 09:15:42 2009 +1000
68372
68373 Xi: include button state in XIButtonInfo.
68374
68375 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68376
68377commit 01241b4247a7ad0974d48412cc0d88e1a09a3c49
68378Author: Peter Hutterer <peter.hutterer@who-t.net>
68379Date: Sun Jun 7 19:43:11 2009 +1000
68380
68381 Xi: Add support for sourceid in the device classes.
68382
68383commit d230742ea820a21a3f1ed0c58b5e6d8680b2f2aa
68384Author: Peter Hutterer <peter.hutterer@who-t.net>
68385Date: Tue Jun 16 12:13:16 2009 +1000
68386
68387 Xi: namespace XI2 files.
68388
68389 Some files (notably those merged with MPX before XI2 came along) didn't use
68390 a 'xi' prefix. This patch changes all of them to meaningful names.
68391
68392 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68393
68394commit 4ddb002b6847e8a88e6f13ae0453a35ee79946ae
68395Author: Peter Hutterer <peter.hutterer@who-t.net>
68396Date: Mon Jun 15 15:14:47 2009 +1000
68397
68398 xfree86: fix two compiler warnings in xf86vmode.c
68399
68400 xf86vmode.c:1578: warning: pointer targets in passing argument 1 of
68401 ‘SwapShorts’ differ in signedness
68402 ../../../../include/misc.h:231: note: expected ‘short int *’ but argument is
68403 of type ‘CARD16 *’
68404
68405 xf86vmode.c:1543: warning: unused variable ‘i’
68406
68407 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68408
68409commit 9d039ef5047af2dedff0c583dda11032f71bed01
68410Author: Peter Hutterer <peter.hutterer@who-t.net>
68411Date: Mon Jun 15 15:11:55 2009 +1000
68412
68413 xfree86: silence a few int10 compiler warnings.
68414
68415 'format ‘%#x’ expects type ‘unsigned int’, but argument 2 has type ‘CARD32’'
68416 and similar.
68417
68418 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68419
68420commit fb882b3da189847bc98b646faddac9e8b42aa435
68421Author: Adam Jackson <ajax@redhat.com>
68422Date: Tue Jun 16 14:58:10 2009 -0400
68423
68424 EDID: Carp about 1.4 monitors with no preferred refresh rate
68425
68426 You would think, having finally tightened down the spec, that
68427 monitor vendors would bother to implement what the spec actually
68428 mandates. You would be wrong.
68429
68430commit b2d2af76b43d62c9be07fd736652f904632fd663
68431Author: Adam Jackson <ajax@redhat.com>
68432Date: Mon Jun 15 10:10:26 2009 -0400
68433
68434 randr: fix typo in swapped dispatch
68435
68436commit 1d20b9021e8aecb942619f64017cc2fc31f02d29
68437Author: Peter Hutterer <peter.hutterer@who-t.net>
68438Date: Sun Jun 14 18:21:43 2009 +1000
68439
68440 Xi: fix wrong bit->byte conversion in ProcXIQueryPointer
68441
68442 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68443
68444commit 9afc3241c157e5c19454f913c7105b3fdfa97ec4
68445Author: Peter Hutterer <peter.hutterer@who-t.net>
68446Date: Sun Jun 14 18:24:22 2009 +1000
68447
68448 Xi: remove un-used IsOn macro.
68449
68450 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68451
68452commit b67ff1d6e0d01170c6261593ca36d0f51d917393
68453Author: Peter Hutterer <peter.hutterer@who-t.net>
68454Date: Fri Jun 12 16:45:36 2009 +1000
68455
68456 Xi: XISelectEvents/XIGetSelectedEvents use 'win' instead of 'window' now.
68457
68458 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68459
68460commit b2bf67b61c564a4b92a429ca9ad455403161f33a
68461Author: Benjamin Defnet <benjamin.r.defnet@intel.com>
68462Date: Wed Jun 10 21:35:48 2009 -0700
68463
68464 randr: fix operation order so that rotation+transform works
68465
68466 The matrix multiply to combine rotation and projective transforms was being
68467 done in the wrong order.
68468
68469 Signed-off-by: Keith Packard <keithp@keithp.com>
68470
68471commit 3627215e97734cc367fff273a3358c2a95a53481
68472Author: Peter Hutterer <peter.hutterer@who-t.net>
68473Date: Fri Jun 5 16:38:46 2009 +1000
68474
68475 Add check for GNU ld in the test suite.
68476
68477 The GNU linker supports a -wrap option to wrap function calls at link-time.
68478 This allows for easy overriding of functions in the X server with stubs in
68479 the test suite. This functionality is only supported on the GNU linker and
68480 will be used extensively in the tests. Disable the tests if GNU ld is not
68481 available.
68482
68483commit 5e0ca6fabd12401de6bd6039484be2079e28851e
68484Author: Peter Hutterer <peter.hutterer@who-t.net>
68485Date: Thu Jun 11 16:26:23 2009 +1000
68486
68487 input: remove un-used "setter" argument from SetClientPointer.
68488
68489 It's obsolete, not likely to come back, let's drop it.
68490
68491 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68492
68493commit e6a18762ef113296c6a09833be70cb4b45aa3940
68494Author: Peter Hutterer <peter.hutterer@who-t.net>
68495Date: Thu Jun 11 15:40:56 2009 +1000
68496
68497 Xi: fix XISetClientPointer return values.
68498
68499 If SetClientPointer fails, the only reason may be that the device is not a
68500 pointer or that the device is an SD. Return BadDevice instead of BadAccess.
68501 (BadAccess is a leftover from the early times of the ClientPointer
68502 implementation when only one client was allowed to set it).
68503
68504 If the window parameter doesn't name a valid window or client, return
68505 BadWindow.
68506
68507 Finally, allow both master keyboards and master pointers as deviceid.
68508
68509 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68510
68511commit ae7dab2a136d15b976b956f68feec53886951dd6
68512Author: Peter Hutterer <peter.hutterer@who-t.net>
68513Date: Thu Jun 11 15:40:38 2009 +1000
68514
68515 Xi: Fix XISetClientPointer swapping.
68516
68517 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68518
68519commit 98e8ec8deb09db1c56e06afed267e6f517042fa3
68520Author: Peter Hutterer <peter.hutterer@who-t.net>
68521Date: Fri Jun 12 10:54:55 2009 +1000
68522
68523 Xi: sanitize ProcXIGetClientPointer.
68524
68525 This was quite old code and can be streamlined a bit. The new code is
68526 essentially the same as in ProcXISetClientPointer.
68527
68528 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68529
68530commit 96ea82fdac7c28feb2748cd4ff7faa7c00dbbdcd
68531Author: Peter Hutterer <peter.hutterer@who-t.net>
68532Date: Fri Jun 12 10:54:25 2009 +1000
68533
68534 Xi: fix reply swapping in XIGetClientPointer.
68535
68536 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68537
68538commit 76f25086a9ab6f0c27d9e2cd745f8463c6b3df5f
68539Author: Adam Jackson <ajax@redhat.com>
68540Date: Thu Jun 11 16:34:06 2009 -0400
68541
68542 vfb: Re-enable 30bpp support
68543
68544commit 0de58c88aba7ddd69b04f24ab5b2967c359aa69e
68545Author: Dave Airlie <airlied@redhat.com>
68546Date: Thu Jun 11 14:21:53 2009 +1000
68547
68548 xfree86: move didLock assignment down to where the function pointer is valid.
68549
68550 crtc->funcs->lock is NULL, so it's no use calling it here. Move it down so
68551 it's actually defined before we use it.
68552
68553 Introduced with 6f59a8160042ea145514fdcb410f17f33fd437c2.
68554
68555 Tested-by: Peter Hutterer <peter.hutterer@who-t.net>
68556
68557commit a85d210b34506cb39e7f08d81d4586fe4448b47f
68558Author: Peter Hutterer <peter.hutterer@who-t.net>
68559Date: Thu Jun 11 12:32:39 2009 +1000
68560
68561 Xi: store mask_len before swapping in ProcXIGetSelectedEvents.
68562
68563 Swapping the mask_len and then advancing the pointer by the swapped length
68564 is just a bad idea.
68565
68566 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68567
68568commit 5043f42f36b49e7702d0c496dd4fe157788d44ee
68569Author: Peter Hutterer <peter.hutterer@who-t.net>
68570Date: Thu Jun 11 12:31:40 2009 +1000
68571
68572 Xi: correct return buffer size for XIGetSelectedEvents.
68573
68574 The maximum number of bytes is calculated by the mask len, and the mask len
68575 is always in 4-byte units. XI2MASKSIZE however is in bytes.
68576
68577 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68578
68579commit fcf0c0b8f332cbde03c1a479376b7913197962eb
68580Author: Peter Hutterer <peter.hutterer@who-t.net>
68581Date: Thu Jun 11 13:23:40 2009 +1000
68582
68583 Xi: XISetEventMask needs to clear the mask if len is 0
68584
68585 zero-length masks are supposed to clear the device's mask.
68586 ProcXISelectEvents passes these masks through directly, so we need to clear
68587 the bits here if such a mask is supplied.
68588
68589 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68590
68591commit 22b4ac44e4fdbe8f36f2b570cdbdaf237062e33e
68592Author: Peter Hutterer <peter.hutterer@who-t.net>
68593Date: Thu Jun 11 11:08:07 2009 +1000
68594
68595 Xi: XIGetSelectedEvents mustn't returned masks from non-existing devices.
68596
68597 Or devices the client doesn't have XACE permissions for.
68598
68599 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68600
68601commit 7868956b0fb50a9d77506f01413612e429f540a3
68602Author: Peter Hutterer <peter.hutterer@who-t.net>
68603Date: Wed Jun 10 17:09:39 2009 +1000
68604
68605 Xi: ProcXIGetSelectedEvents must use WriteToClient for swapped data.
68606
68607 The data is already swapped before, so we just post it to the client as-is,
68608 without attempting to swap it again.
68609
68610 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68611
68612commit aa2babf11c30be4d289b58212dc330b84efb4053
68613Author: Peter Hutterer <peter.hutterer@who-t.net>
68614Date: Wed Jun 10 14:41:11 2009 +1000
68615
68616 input: remove dependency on XI2 protocol for XI_LASTEVENT.
68617
68618 inputstr includes XI2proto.h for the sole purpose of XI_LASTEVENT.
68619 However, using XI_LASTEVENT in the server is prone to errors, if the server
68620 is recompiled against a newer version of the protocol it would bump this
68621 variable and associates bits, including potential ABI.
68622
68623 This patch defines an XI2LASTEVENT for use in the server and removes the
68624 XI2proto.h require. XI2LASTEVENT is the current value of XI_LASTEVENT.
68625
68626 This patch is required by components that require access to inputInfo
68627 (currently xf86-video-geode and xf86-video-cirrus) but should not have a
68628 require for the XI2 protocol.
68629
68630 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68631
68632commit 77cc816da4a2777110182ed01c22c0f5e6ac3b65
68633Author: Peter Hutterer <peter.hutterer@who-t.net>
68634Date: Wed Jun 10 13:54:42 2009 +1000
68635
68636 Xi: rename ProcXISelectEvent to ProcXISelectEvents.
68637
68638 The request name has the plural, so let's do it here too. Purely cosmetic
68639 change.
68640
68641 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68642
68643commit 9563feeeb5309de1971c4e04419fed6fd31c9cea
68644Author: Peter Hutterer <peter.hutterer@who-t.net>
68645Date: Tue Jun 9 16:57:09 2009 +1000
68646
68647 Xi: start checking for invalid mask bits _after_ LASTEVENT.
68648
68649 Two issues that combined to false positives and false negatives.
68650 - The checking for invalid bits must be performed when there are enough bits
68651 that an event outside of LASTEVENT may be selected.
68652 - The first invalid bit is LASTEVENT + 1, not LASTEVENT.
68653
68654 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68655
68656commit e2fbaebb87e18198143a4b8e6be7e650563cd819
68657Author: Peter Hutterer <peter.hutterer@who-t.net>
68658Date: Tue Jun 9 14:38:25 2009 +1000
68659
68660 Xi: XISelectEvents returns BadValue for num_masks == 0.
68661
68662 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68663
68664commit cfeb65ac45b194dead5e8b9b26192df2619d6f9c
68665Author: Peter Hutterer <peter.hutterer@who-t.net>
68666Date: Tue Jun 9 14:37:42 2009 +1000
68667
68668 Xi: XISelectEvents needs to be at least size 3, not exactly.
68669
68670 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68671
68672commit f3c26034eca98924d93a004aaa1f61ffe5d344ca
68673Author: Peter Hutterer <peter.hutterer@who-t.net>
68674Date: Tue Jun 9 13:04:56 2009 +1000
68675
68676 Xi: XIQueryDevice should use XIFooClass instead of the old FooClass defines.
68677
68678 This is merely a cosmetic change, the actual values are the same anyway.
68679
68680 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68681
68682commit 482cc72aa790bd2f23c7ebe12c5549b47c737115
68683Author: Peter Hutterer <peter.hutterer@who-t.net>
68684Date: Tue Jun 9 13:03:38 2009 +1000
68685
68686 Xi: get the class length before swapping.
68687
68688 Advancing by the already-swapped length lets our pointers point into
68689 nirvana.
68690
68691 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68692
68693commit 9974249980894f74f3ead466655da87958a43670
68694Author: Peter Hutterer <peter.hutterer@who-t.net>
68695Date: Tue Jun 9 13:02:52 2009 +1000
68696
68697 Xi: Swapping 32 bit keycodes requires swapl, not swaps.
68698
68699 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68700
68701commit 918923e285f4e269a257bb5be4d3c8a50174aad0
68702Author: Dave Airlie <airlied@redhat.com>
68703Date: Wed Jun 10 15:13:45 2009 +1000
68704
68705 glx: fix open-coded linked list removal function
68706
68707 OMG stab stab stab, YALL.
68708
68709 removal function was made of crack, actually truncated the list from
68710 the one after the find point.
68711
68712 However fixing this makes glean makecurrent fail with a GLX error.
68713
68714commit 3ea747c0dbbec0db6761d66d4f6c680d2e9ddeaf
68715Author: Daniel Stone <daniel@fooishbar.org>
68716Date: Sat Apr 25 16:53:18 2009 +1000
68717
68718 KDrive: Warning fixes
68719
68720 xEvent vs. InternalEvent confusion still reigns though.
68721
68722 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
68723
68724commit 754be1e2ec9be2486bf45000d7244d217556de07
68725Author: Daniel Stone <daniel@fooishbar.org>
68726Date: Sat Apr 25 16:51:48 2009 +1000
68727
68728 KDrive: Xephyr: DRI: Warning fixes
68729
68730 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
68731
68732commit f534e6bea17746db952feb563ffea7320846b49d
68733Author: Daniel Stone <daniel@fooishbar.org>
68734Date: Sat Apr 25 15:38:54 2009 +1000
68735
68736 OS: Fix compile warnings
68737
68738 It's a marvel the sigaction() ever actually worked.
68739
68740 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
68741
68742commit 305ab237f666936cd812c464bf43f86f6079838e
68743Author: Daniel Stone <daniel@fooishbar.org>
68744Date: Sat Apr 25 22:11:19 2009 +1000
68745
68746 KDrive: Warning fixes and cleanups
68747
68748 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
68749
68750commit 4c8812b544ce5f319a7ac59a131c626c8509ef99
68751Author: Adam Jackson <ajax@redhat.com>
68752Date: Wed Jun 10 12:46:19 2009 -0400
68753
68754 kdrive: undef PSEUDO8
68755
68756 Whatever that was, it no longer is.
68757
68758commit 973ef5a7086aedb88a4888e73630c96ae68078e4
68759Author: Adam Jackson <ajax@redhat.com>
68760Date: Wed Jun 10 12:44:45 2009 -0400
68761
68762 kdrive: Remove a lie about PCMCIA support.
68763
68764commit e3c65cf1df9bbfb126f07b4b8e1254a855fb70c7
68765Author: Adam Jackson <ajax@redhat.com>
68766Date: Wed Jun 10 12:05:09 2009 -0400
68767
68768 xephyr: Add -title option.
68769
68770commit 803522300344265f11ea9f869ee6747730413bae
68771Author: Colin Harrison <colin.harrison@virgin.net>
68772Date: Wed Jun 10 13:32:54 2009 +0100
68773
68774 Xming: Simplify logic for ownership release of native clipboard
68775
68776 Make the logic simpler for the ownership release of the Windows
68777 clipboard in winclipboardwrappers.c - We've already marked a selection
68778 as unowned if it is owned by our clipboard window
68779
68780 Copyright (C) Colin Harrison 2005-2008
68781 http://www.straightrunning.com/XmingNotes/
68782 http://sourceforge.net/projects/xming/
68783
68784 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
68785
68786commit c7d3965bee0edb72468a0135aaa8a081348c2b23
68787Author: Jon TURNEY <jon.turney@dronecode.org.uk>
68788Date: Wed Jun 10 13:09:44 2009 +0100
68789
68790 Cygwin/X: Add a needed inputstr.h
68791
68792 Add a needed inputstr.h to fix build after commit 6d4ffcc9e066bc7c49dcff1a2ff0111801c7286d
68793
68794 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
68795
68796commit 6d4ffcc9e066bc7c49dcff1a2ff0111801c7286d
68797Author: Dave Airlie <airlied@redhat.com>
68798Date: Wed Jun 10 11:15:24 2009 +1000
68799
68800 input: move inputstr.h to where its needed.
68801
68802 This stops inputstr.h being needed to be included by output drivers.
68803
68804 Signed-off-by: Dave Airlie <airlied@redhat.com>
68805 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68806
68807commit 6f59a8160042ea145514fdcb410f17f33fd437c2
68808Author: Benjamin Defnet <benjamin.r.defnet@intel.com>
68809Date: Mon Jun 8 21:45:42 2009 -0700
68810
68811 hw/xf86/modes: Set crtc mode/rotation/transform before calling set_mode_major
68812
68813 This moves code out of each implementation of set_mode_major and back into
68814 the X server. The real feature here is that the transform is now available
68815 in the crtc for use by either xf86CrtcRotate or whatever the driver wants to
68816 do. Without this change, the transform was lost for drivers providing the
68817 set_mode_major interface.
68818
68819 Note that users of this API will want to stop smashing the transformPresent
68820 field, and could also stop setting mode/x/y/rotation for new enough X servers,
68821 but there's no reason to make that change as it will break things when
68822 running against older X servers.
68823
68824 Signed-off-by: Keith Packard <keithp@keithp.com>
68825 Acked-by: Daniel Stone <daniel@fooishbar.org>
68826
68827commit e244a5991e2cc55f5aa2f6e5255f1dabf56f0235
68828Author: Federico Mena Quintero <federico@novell.com>
68829Date: Tue Jun 9 13:28:28 2009 -0700
68830
68831 dix/randr: Add missing fields to SRR*NotifyEvent()
68832
68833 Also, remove redundant field swaps and make others match the order in which
68834 they are declared in the xRR*NotifyEvent structs.
68835
68836 Signed-off-by: Federico Mena Quintero <federico@novell.com>
68837 Signed-off-by: Keith Packard <keithp@keithp.com>
68838
68839commit 746e7b22e1bdfbf2363040367209b4c46b1c2fec
68840Author: Adam Jackson <ajax@redhat.com>
68841Date: Tue Jun 9 16:20:11 2009 -0400
68842
68843 cvt: Allow multiple-of-60Hz refresh rates for reduced blanking.
68844
68845 Doing so generates the same timings as given in the DMT spec for
68846 120Hz RB, so we should be set there. Other rates might be legal
68847 too but why push our luck.
68848
68849commit 2a035600e603117a4dbe668cad50bae2a5609094
68850Author: Peter Hutterer <peter.hutterer@who-t.net>
68851Date: Tue Jun 9 10:14:18 2009 +1000
68852
68853 require inputproto 1.9.99.11
68854
68855commit c5bebca46f63bcd7279f3c73ed1d50e0c1b6051f
68856Author: Peter Hutterer <peter.hutterer@who-t.net>
68857Date: Mon Jun 8 17:54:05 2009 +1000
68858
68859 Xi: hierarchy events have a num_info now instead of num_devices.
68860
68861 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68862
68863commit 810b74dbbcc1db3e68e0f95ff5223ddb2032e46d
68864Author: Peter Hutterer <peter.hutterer@who-t.net>
68865Date: Mon Jun 8 13:41:22 2009 +1000
68866
68867 Xi: change from XICreateMaster to XIAddMaster for consistency.
68868
68869 add/remove is used for slave devices and hierarchy flags.
68870
68871 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68872
68873commit 0e66a443a09176b8a875195d5473b953c07aa1d7
68874Author: Peter Hutterer <peter.hutterer@who-t.net>
68875Date: Mon Jun 8 12:24:24 2009 +1000
68876
68877 Xi: return BadValue for XI_HierarchyChangd mask on devices.
68878
68879 This mask may only be selected for XIAllDevices.
68880
68881 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68882
68883commit 69a9545d1f8110841538410818df19fd960412c5
68884Author: Keith Packard <keithp@keithp.com>
68885Date: Thu May 28 14:43:27 2009 -0700
68886
68887 Make RANDR 'set' timestamps follow client specified time. Bug 21987.
68888
68889 The lastSetTime value which indicates when the configuration within the
68890 server was last changed was not getting set in the appropriate RandR
68891 requests.
68892
68893 Signed-off-by: Keith Packard <keithp@keithp.com>
68894
68895commit fa18c569ed26d2fdd442af6b39723cb7e30d18df
68896Author: Ben Skeggs <bskeggs@redhat.com>
68897Date: Tue Jun 9 09:14:50 2009 +1000
68898
68899 exa: driver pixmaps enabled if either CreatePixmap or CreatePixmap2 present
68900
68901commit 9390b7a133b0e17bc4d303effb7a99ff7017b086
68902Author: Peter Hutterer <peter.hutterer@who-t.net>
68903Date: Sun Jun 7 19:26:45 2009 +1000
68904
68905 dix: protect against missing ButtonClasses in GetPointerMapping.
68906
68907 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68908
68909commit 21ee53060b4dcb5c12074cf6682ff0067021ff60
68910Author: Simon Thum <simon.thum@gmx.de>
68911Date: Fri Jun 5 19:14:48 2009 +0200
68912
68913 config: fix build after XI2 API changes to RemoveDevice.
68914
68915 This patch fixes the build with --enable-config-dbus is enabled
68916 (default disabled).
68917
68918 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68919
68920commit 07c59411817c10fbcbf1bc9efa6a0818b6e0493a
68921Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
68922Date: Sat Jun 6 22:17:44 2009 -0700
68923
68924 GLX: Make sure the types match for ALIAS in indirect_reqsize.c
68925 (cherry picked from commit 31a20a573b867fb7f02da15425e3b92eccb82dec)
68926
68927commit 077556ec39ca7e0d14486005cf71fcdc70a350cf
68928Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
68929Date: Sat Jun 6 21:55:51 2009 -0700
68930
68931 XQuartz: AIGLX dispatch table cleanup
68932
68933 This should fix some issues when building on different versions of OSX. We only use extensions that OpenGL.framework's glext.h header tells us are supported.
68934 (cherry picked from commit e10c53421f1e780573c8adcd8ea89d6ffa28a819)
68935
68936commit fc24a97a8d022e986082b65b5133701e7273805c
68937Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
68938Date: Tue Jun 2 19:12:41 2009 -0700
68939
68940 Rootless: Use serverClient instead of NullClient
68941
68942 This seems to stop the crash when moving windows.
68943 (cherry picked from commit fcd857c8e295ac7a2b0e58d49dc71251ed9d1266)
68944
68945commit 2429daa7e6b35a4a8032e1c3736e417a5a9baa07
68946Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
68947Date: Tue Jun 2 10:19:10 2009 -0700
68948
68949 XQuartz: Default to forcing the dialog box before quitting.
68950
68951 This will show the quit dialog for users of other WMs.
68952 (cherry picked from commit 639ee0c913f892776fee40bcfb856e0d7c2c7db5)
68953 (cherry picked from commit 09432fa5b1cc8e9a30fb8aa4f8dfcd08c4a91e30)
68954
68955commit 6d75606a9e7f6ebc4d79a5261a48f5915c71cbcd
68956Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
68957Date: Mon Jun 1 22:58:40 2009 -0700
68958
68959 Rootless: Despite its name, we still want to allow rootless to mess with the root window
68960
68961 This fixes the bug whereby we weren't drawing the contents of the root window since RootlessPaintWindowBackground got nuked.
68962 (cherry picked from commit bb0e208787739fbed2a0c8d1af6b6504319fdf88)
68963
68964commit 7348bf4935c8887bca33fe340ac2548f551bee3d
68965Author: Peter Hutterer <peter.hutterer@who-t.net>
68966Date: Sun Jun 7 09:25:51 2009 +1000
68967
68968 dix: set the generic event type for implicit XI2 grabs.
68969
68970 event->type is always GenericEvent for XI2 events. Instead, XI_ButtonPress
68971 (the generic event's evtype must be stored.
68972
68973 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68974
68975commit bb1c131b785ca30cf40041db3513b7669beebd34
68976Author: Simon Thum <simon.thum@gmx.de>
68977Date: Thu May 28 16:43:42 2009 +0200
68978
68979 dix: suppress pointer acceleration on xtest devices
68980
68981 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68982
68983commit 707b1241681c33401d5e0eb6edeace33f5970928
68984Author: Simon Thum <simon.thum@gmx.de>
68985Date: Sat Mar 21 18:19:19 2009 +0100
68986
68987 dix: prefer lroundf() over roundf() in axis scaling
68988
68989 it's unclear whether there actually is a problem, but in a very similar
68990 case there is (bug#21456). Also, integer addition is generally faster.
68991
68992 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
68993
68994commit ef0e435052b2fda598cd8670c8e892aea685a102
68995Author: Jon TURNEY <jon.turney@dronecode.org.uk>
68996Date: Fri Jun 5 19:29:02 2009 +0100
68997
68998 Cygwin/X: Remove obsolete ChangeLog files
68999
69000 Remove obsolete ChangeLog files, this information is preserved in
69001 the automatically generated top-level ChangeLog
69002
69003commit 752508577aa7f0a1d80c8b7f4d3fe2b098e5e5fa
69004Author: Jon TURNEY <jon.turney@dronecode.org.uk>
69005Date: Tue Jun 2 21:19:46 2009 +0100
69006
69007 Cygwin/X: Tidy up a bodge to avoid collision between X header and native Win32 API definitions of ATOM
69008
69009 Remove a bodge to avoid collision between X header and the native
69010 Win32 API definitions of ATOM, use X11/Xwindows.h which exists to
69011 address these issues
69012
69013 Consequently, include misc.h in winclipboardthread.c, to provide the
69014 max() macro, as previously we were using the the one from windows.h
69015
69016 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69017
69018commit 7f3456a4080d1bf21010170e126128cb653bd2ac
69019Author: Peter Hutterer <peter.hutterer@who-t.net>
69020Date: Fri Jun 5 16:44:46 2009 +1000
69021
69022 Revert "test: check from INT_MIN to INT_MAX for core type conversion"
69023
69024 This patch requires extra special casing to check if the linker supports the
69025 -wrap option. Patches to do so will follow, in the meantime, revert this
69026 commit. It shouldn't have been pushed in the first place anyway.
69027
69028 This reverts commit d979f443946011158b6a183582728a6899c33b85.
69029
69030 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69031
69032commit d770d57f1701c7c8272b4ca3714f64c46c47adf1
69033Author: Peter Hutterer <peter.hutterer@who-t.net>
69034Date: Fri Jun 5 11:38:22 2009 +1000
69035
69036 Xi: fix XIQueryVersion major/minor swapping.
69037
69038 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69039
69040commit 17d62306b602ec61e8c5b44a79c9e1edc6b63a21
69041Author: Peter Hutterer <peter.hutterer@who-t.net>
69042Date: Fri Jun 5 11:31:39 2009 +1000
69043
69044 Xi: add XIQueryVersion reply swapping hook.
69045
69046 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69047
69048commit 2b459f44f3edaea137df9a28bc7adfeb1b9f1df7
69049Author: Peter Hutterer <peter.hutterer@who-t.net>
69050Date: Fri Jun 5 10:15:04 2009 +1000
69051
69052 Xi: return BadImplementation for deviceids 256 and above
69053
69054 The protocol allows for 16 bit device ids, but the implementation doesn't
69055 yet. We need to break the input ABI once more to shift the DeviceIntRec's
69056 CARD8 to a CARD16, along with some changes in the privates.
69057
69058 Once that is done, revert this patch.
69059
69060 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69061
69062commit c77f42049f77417c68bbce3dd2e995d916fc5af5
69063Author: Peter Hutterer <peter.hutterer@who-t.net>
69064Date: Fri Jun 5 09:29:04 2009 +1000
69065
69066 Xi: fix flawed logic in XIQueryVersion return value.
69067
69068 The server returned the minimum of major/minor each instead of the lower of
69069 the combined major.minor version. As a result, a client reporting
69070 3.0 and a server supporting 2.7, the return value would be 2.0 (the minimum
69071 of both).
69072
69073 Reported-by: Rémi Cardona
69074 Reported-by: Simon Thum
69075 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69076
69077commit b6c16fc7dad5a4ffcaaab647c0e3fabed372efd5
69078Author: Dave Airlie <airlied@redhat.com>
69079Date: Fri Jun 5 11:57:39 2009 +1000
69080
69081 crtc/gamma: check xf86_config is valid before using it.
69082
69083 If you have multiple cards, some that support randr 1.2 and some that don't
69084 you can get a null dereference in here.
69085
69086 Signed-off-by: Dave Airlie <airlied@redhat.com>
69087
69088commit da682abc78c9a5b2fde09679061fe08a3a1c665b
69089Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
69090Date: Thu Jun 4 17:33:25 2009 -0400
69091
69092 xace: fix a bad send access hook call.
69093
69094commit 993daf06497c85bb6a1e70592df380503d721dfb
69095Merge: 2bda50e e1e8c7d
69096Author: Peter Hutterer <peter.hutterer@who-t.net>
69097Date: Thu Jun 4 15:11:36 2009 +1000
69098
69099 Merge branch 'master' into xi2
69100
69101commit 2bda50ee14b7873f28081bf0f1bfb7b79a40de43
69102Author: Peter Hutterer <peter.hutterer@who-t.net>
69103Date: Thu Jun 4 14:51:32 2009 +1000
69104
69105 Xi: memset the device name padding to 0.
69106
69107 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69108
69109commit 09266d6cd1bb6e09fe8bcb875c63fb4011c05b55
69110Author: Peter Hutterer <peter.hutterer@who-t.net>
69111Date: Thu Jun 4 13:36:29 2009 +1000
69112
69113 Xi: XIQueryVersion requires major_version of 2 or higher.
69114
69115 Otherwise, a BadValue error is sent.
69116
69117 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69118
69119commit 6ea76c550486bd17eb31f43757030fb44bdf1eab
69120Author: Peter Hutterer <peter.hutterer@who-t.net>
69121Date: Thu Jun 4 12:55:59 2009 +1000
69122
69123 Xi: reply with the XI2 version that is supported by both client and server.
69124
69125 XIQueryVersion must return the client's version if the client's version is
69126 lower than or equal to the server's version, or the server's version
69127 otherwise.
69128
69129 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69130
69131commit d979f443946011158b6a183582728a6899c33b85
69132Author: Peter Hutterer <peter.hutterer@who-t.net>
69133Date: Thu Jun 4 11:40:14 2009 +1000
69134
69135 test: check from INT_MIN to INT_MAX for core type conversion
69136
69137 ErrorF is link-wrapped to reduce the run-time of the test.
69138
69139 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69140
69141commit d8471bc7a19cd248fa3e0845496c86cd9251c013
69142Author: Peter Hutterer <peter.hutterer@who-t.net>
69143Date: Thu Jun 4 11:32:15 2009 +1000
69144
69145 test: adjust event_to_core_conversion test for new ProximityIn/Out behaviour
69146
69147 Both types now return BadMatch.
69148
69149 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69150
69151commit e1e8c7ddd7fdbfd674361364295fb1cbd5f28b45
69152Author: Adam Jackson <ajax@redhat.com>
69153Date: Wed Jun 3 13:37:32 2009 -0400
69154
69155 s/MIN/min/, s/MAX/max/ (#2968)
69156
69157commit 6de67d3206f0a4e307070714564c19efc84da2ec
69158Merge: af4e658 7f027d9
69159Author: Peter Hutterer <peter.hutterer@who-t.net>
69160Date: Wed Jun 3 16:11:33 2009 +1000
69161
69162 Merge branch 'master' into xi2
69163
69164commit 7f027d9dc0146e229802aeac342ea2dbab63264a
69165Author: Olivier Blin <blino@mandriva.com>
69166Date: Thu May 28 20:57:36 2009 +0200
69167
69168 kdrive: add protocol mouse option
69169
69170 kdrive probes a lot of PS/2 protocols for the mouse device, which
69171 makes the mouse unusable for some seconds after X startup.
69172 This new "protocol" option allows forcing the mouse protocol.
69173 It can be used this way:
69174 Xfbdev -mouse mouse,,protocol=ps/2 -keybd keyboard
69175
69176 Signed-off-by: Olivier Blin <blino@mandriva.com>
69177 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69178
69179commit af4e658401ecbd2f2cc515b4c5abe5aee361cd64
69180Author: Peter Hutterer <peter.hutterer@who-t.net>
69181Date: Wed Jun 3 12:30:49 2009 +1000
69182
69183 mi: only process master if the SD is still attached.
69184
69185 SDs may be detached during event processing (e.g. if a passive grab
69186 activates). In this case, the event must not be processed through the master
69187 device.
69188
69189 Reported-by: Thomas Jaeger
69190 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69191
69192commit eb3fe72cbcd10c3953d713e07c321e695194a328
69193Author: Peter Hutterer <peter.hutterer@who-t.net>
69194Date: Wed Jun 3 12:19:14 2009 +1000
69195
69196 dix: set the device state for XI valuator events.
69197
69198 Valuator events need to include the device's state, while other device
69199 events need to include the state of the core devices.
69200
69201 Reported-by: Thomas Jaeger
69202 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69203
69204commit 08fd2ae652883393ecff6f3d5177e63408d58472
69205Author: Peter Hutterer <peter.hutterer@who-t.net>
69206Date: Wed Jun 3 12:18:17 2009 +1000
69207
69208 dix: use GetMaster in PickPointer and PickKeyboard.
69209
69210 GetMaster is more reliable than GetPairedDevice, it always returns the
69211 keyboard/pointer if desired, even if the wrong device was passed in.
69212
69213 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69214
69215commit a1d1dd06f8f91c1d1802f7581a4ba7568b85e69c
69216Author: Peter Hutterer <peter.hutterer@who-t.net>
69217Date: Wed Jun 3 12:15:32 2009 +1000
69218
69219 Xi: fix wrong grabtype from ProcXGrabDevice.
69220
69221 86077f0058ce88ee9b3df5d1ab854eeca43 switched from a boolean to a grabtype
69222 enum. ProcXGrabDevice didn't switch with it. PickPointer during an XI grab
69223 on a slave device would thus return a wrong (or NULL) device and crash the
69224 server.
69225
69226 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69227
69228commit a371a47205fc58a587e1aecaa4a031184481c228
69229Author: Peter Hutterer <peter.hutterer@who-t.net>
69230Date: Wed Jun 3 11:47:16 2009 +1000
69231
69232 dix: EventToCore must return BadMatch for proximity events.
69233
69234 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69235
69236commit 2bd5ea80652358565346071bed829070ddf5de85
69237Author: Peter Hutterer <peter.hutterer@who-t.net>
69238Date: Wed Jun 3 10:45:28 2009 +1000
69239
69240 dix: rework event conversion before delivery.
69241
69242 Let EventToXI, EventToCore and EventToXI2 return BadMatch if there's no
69243 matching event for this protocol spec.
69244 Adjust the delivery paths to cope with BadMatch errors (and clean them up on
69245 the way).
69246
69247 As a side-effect, this fixes server crashes on proximity events for a
69248 grabbed device.
69249
69250 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69251
69252commit 20d9f11abfd6ce3630d9640ff7383c7e27196956
69253Author: Adam Jackson <ajax@redhat.com>
69254Date: Tue Jun 2 13:49:09 2009 -0400
69255
69256 Typo fix
69257
69258commit ab5d1ae82ebd467d0a5b39821b286097865ba0c7
69259Author: Adam Jackson <ajax@redhat.com>
69260Date: Tue Jun 2 13:39:55 2009 -0400
69261
69262 Remove some libXfont leftovers
69263
69264commit 75795637c7160f1579dbe81c2d7600e85b1d141f
69265Author: Matthias Hopf <mhopf@suse.de>
69266Date: Tue Jun 2 18:41:36 2009 +0200
69267
69268 Revert "randr: Setting gamma: inverse logic looks more sane"
69269
69270 This reverts commit 79138eec1b49cbaca6a16f2bdd8579b5828aeb28.
69271
69272commit fc8cfc3a055d8af4ac809799c71f3db0d5246433
69273Author: Peter Hutterer <peter.hutterer@who-t.net>
69274Date: Tue Jun 2 16:53:41 2009 +1000
69275
69276 dix: ensure EventIsDeliverable has inputMasks set at all times.
69277
69278 For proximity events, the XI2 type is 0 and inputMasks never got set in the
69279 preceding condition. As a result, proximity events got never delivered.
69280
69281 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69282
69283commit ef9f851057a0ae214be02cb919f1e6634cead20b
69284Author: Peter Hutterer <peter.hutterer@who-t.net>
69285Date: Tue Jun 2 16:38:32 2009 +1000
69286
69287 dix: protect against floating slaves in DeviceFocusEvent.
69288
69289 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69290
69291commit b40289c8766c5e7543b7288009b4d965f57c04ce
69292Author: Peter Hutterer <peter.hutterer@who-t.net>
69293Date: Tue Jun 2 16:22:58 2009 +1000
69294
69295 xkb: allow pointer events to pass through for floating SDs without a key class.
69296
69297 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69298
69299commit 38df8351b36195070232b42a6d3369a9a08df8df
69300Author: Peter Hutterer <peter.hutterer@who-t.net>
69301Date: Mon Jun 1 17:09:03 2009 +1000
69302
69303 test: add grab matching tests.
69304
69305commit a66686a83edd6093ed540414dda2b8700e4696fe
69306Author: Peter Hutterer <peter.hutterer@who-t.net>
69307Date: Mon Jun 1 15:37:53 2009 +1000
69308
69309 input: add support for XIAllDevices and XIAllMasterDevices passive grabs.
69310
69311 These grabs are suported through two fake devices inputInfo.all_devices and
69312 inputInfo.all_master_devices. These devices are not part of the device list
69313 and are only initialised for their device id, nothing else.
69314
69315 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69316
69317commit f00cf7675164bb984ef310412f9e09582813adb1
69318Author: Peter Hutterer <peter.hutterer@who-t.net>
69319Date: Tue Jun 2 10:23:21 2009 +1000
69320
69321 dix: get the current MD keyboard for the grab modifier state.
69322
69323 A passive XI2 grab always uses the paired master device as a modifier
69324 device. After issuing a passive grab, the slave may be reattached to a
69325 different master and hence the modifier device may change.
69326
69327commit 1cd7ed75d903b5b0ea4e1070d2c26f4f931bdec7
69328Author: Peter Hutterer <peter.hutterer@who-t.net>
69329Date: Mon Jun 1 17:14:06 2009 +1000
69330
69331 dix: float slave devices during passive grabs.
69332
69333 Exception - implicit passive grabs.
69334
69335 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69336
69337commit f56ad22e8fabdca6d9fd1e408a539696e30599fc
69338Author: Peter Hutterer <peter.hutterer@who-t.net>
69339Date: Tue Jun 2 15:07:53 2009 +1000
69340
69341 dix: rework temporary slave detachment.
69342
69343 Rework addresses two issues:
69344 - storing the master device's pointer is a bad idea, we need to store the ID
69345 of the device in case it disappears during the grab.
69346 - restoring the old master did not actually reattach the device. Fixed now.
69347
69348commit b3e4810a283d369e1a2c6c3288177fe79fd1e9d6
69349Author: Peter Hutterer <peter.hutterer@who-t.net>
69350Date: Mon Jun 1 16:22:51 2009 +1000
69351
69352 dix: move grab device assignment into GRABTYPE_CORE block.
69353
69354 grab->type is the device type and XI2 types overlap with core events (being
69355 less than GenericEvent). Thus, for passive grabs the grab device would be
69356 overwritten with whatever device was activating it.
69357
69358 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69359
69360commit 18ae48670859d05788776e164c65985e3b35a723
69361Author: Peter Hutterer <peter.hutterer@who-t.net>
69362Date: Mon Jun 1 16:06:52 2009 +1000
69363
69364 dix: don't try to activate a passive grab on top of an active one.
69365
69366 If we already have an active grab on the device, don't attempt to activate
69367 another one.
69368
69369 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69370
69371commit ff1d6244eba8367a8d50601ca393e951c60c94b6
69372Author: Peter Hutterer <peter.hutterer@who-t.net>
69373Date: Mon Jun 1 12:14:52 2009 +1000
69374
69375 input: add support for XIAnyModifier in passive XI2 grabs.
69376
69377commit eb97d125fb6adeff0b23c2460108e0587ba253e2
69378Author: Adam Jackson <ajax@redhat.com>
69379Date: Mon Jun 1 15:50:50 2009 -0400
69380
69381 doc: Remove mention of Kerberos replay cache from Xserver.1
69382
69383 The krb5 auth code is gone, so this is no longer relevant.
69384
69385commit 80ea32ad4f1440d068e18db65830f45498703b09
69386Author: Peter Hutterer <peter.hutterer@who-t.net>
69387Date: Mon Jun 1 12:06:55 2009 +1000
69388
69389 dix: switch grab checking from unsigned shorts to unsigned ints
69390
69391 XIAnyModifier is outside of the unsigned short range.
69392
69393 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69394
69395commit 080b0331b3d1de1bab5ecdc9df175b287616e358
69396Author: Peter Hutterer <peter.hutterer@who-t.net>
69397Date: Mon Jun 1 13:33:29 2009 +1000
69398
69399 input: Add grabtype to GrabParameters.
69400
69401 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69402
69403commit bee02a154145c0e4ed97980f7857dc27a96b1be8
69404Author: Peter Hutterer <peter.hutterer@who-t.net>
69405Date: Mon Jun 1 10:49:22 2009 +1000
69406
69407 dix: ensure implicit passive grabs have the xi2mask zeroed.
69408
69409 This fixes XI2 enter/leave events being sent to non-xi2 clients.
69410
69411 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69412
69413commit cc71dfb4f5cf6113053951070dec9556e941aa24
69414Author: Peter Hutterer <peter.hutterer@who-t.net>
69415Date: Mon Jun 1 13:19:30 2009 +1000
69416
69417 test: Add GetSelectedEvents request to struct size tests.
69418
69419 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69420
69421commit de5c5b1db083a9c1d23e1461b54ca646c5a5c445
69422Author: Peter Hutterer <peter.hutterer@who-t.net>
69423Date: Mon Jun 1 13:21:36 2009 +1000
69424
69425 test: fix compile error introduced with the removal of isMaster.
69426
69427 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69428
69429commit 45530d16097459a756696e255ab4e72d6e51fbc4
69430Author: Adam Jackson <ajax@redhat.com>
69431Date: Fri May 29 18:07:48 2009 -0400
69432
69433 vfb: Fix depth setup.
69434
69435 Initialize the depth corresponding to the root window before the
69436 pixmap-only depths. Otherwise you end up with the root window depth in
69437 the depth list twice, which is mildly confusing for clients and
69438 catastrophically confusing for PanoramiXConsolidate().
69439
69440commit 5c8540d8cf9ebaafa7572cb2879d057cd3d15190
69441Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
69442Date: Fri May 29 14:49:05 2009 -0700
69443
69444 GLX: Purge glxint.h usage
69445
69446 Change (__GLXvisualConfig *) to (void *) in ABI compat stubs.
69447 (cherry picked from commit b4adec886c65e49cfbd30d1e3a41b613f75ed3f5)
69448
69449commit da2aaee24a77e1a9e5c623af4a78ca6acc94b787
69450Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
69451Date: Fri May 29 14:27:46 2009 -0700
69452
69453 GLX: Purge some glxint.h usage in glxcmds.c
69454
69455 This is related to d0b249f1c5df81f3941d638f9625fe1e16c31807 in the 1.4 branch and was ok'd by krh
69456 (cherry picked from commit e587436cf2e1237813c924699ff0662aea3034e1)
69457
69458commit f2151f1b0dcbfec23488a29a7ea887fbfd53b760
69459Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
69460Date: Fri May 29 14:09:41 2009 -0700
69461
69462 XQuartz: Fix alpha to be 1 (screenshots bug, etc)
69463
69464 This is related to: bc964ff1e3f3d8ab590eab40d767c6241e9480d4
69465 XQuartz: Stab at fixing the alpha 0/1 bug (screenshots, etc) by pulling in some old code that got gutted from rootless.
69466
69467 which was on the 1.4 branch and implemented in fbPaintWindow. Now that fbPaintWindow is gone, this is now in miPaintWindow().
69468 (cherry picked from commit 032173f693f75385a40e144ffe2b7b5378bb6f2e)
69469
69470commit 7c158ffa766d32f5e8ec459693da5ef6ba0892e9
69471Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
69472Date: Fri May 29 14:02:18 2009 -0700
69473
69474 Rootless: Use miPaintWindow since PaintWindowBackground is dead
69475 (cherry picked from commit 4975c087257b52189a5c21d93d121f183e60e4c8)
69476
69477commit ec9e9c8bcabda3d8b4ea07037fb247f7408dde8b
69478Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
69479Date: Thu May 28 15:30:20 2009 -0700
69480
69481 XQuartz: English localization update for HIG compliance
69482 (cherry picked from commit 86b319d7cbef39d75b54c9d321e5dbf097b552be)
69483
69484commit 0f47203a9a690498c18b1c04e2ef5deed23a0e3e
69485Author: Peter Hutterer <peter.hutterer@who-t.net>
69486Date: Fri May 29 14:18:37 2009 +1000
69487
69488 dix: Set the implicit passive grab flag for keyboards too.
69489
69490 Extension devices have ActivateKeyboardGrab as their grab activation
69491 function, hence we need to ensure the implicit passive grab flag is set
69492 accordingly in the grab for further event delivery.
69493
69494 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69495
69496commit 75c73ddbfc9236bb80a3709d467674c4a1018fa3
69497Author: Peter Hutterer <peter.hutterer@who-t.net>
69498Date: Fri May 29 14:01:21 2009 +1000
69499
69500 dix: fix broken XI event delivery during grabs.
69501
69502 If an implicit passive grab is active, the XI event mask is in
69503 grab->deviceMask. Otherwise, for explicit grabs, the XI event mask is in
69504 grab->eventMask.
69505
69506 Reported-by: Thomas Jaeger
69507 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69508
69509commit 29ec14625111a8ae600d245ccbf905f059074d7c
69510Author: Peter Hutterer <peter.hutterer@who-t.net>
69511Date: Fri May 29 13:26:22 2009 +1000
69512
69513 Xi: passively grabbing slave devices must be possible.
69514
69515 There's use-cases where this is useful, so take the check out preventing
69516 that.
69517
69518 Reported-by: Thomas Jaeger
69519 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69520
69521commit 235cf5713a8fe535c439628cce9ea5b42432acb1
69522Author: Peter Hutterer <peter.hutterer@who-t.net>
69523Date: Fri May 29 12:09:06 2009 +1000
69524
69525 dix: init inputMasks to NULL to avoid random value dereference.
69526
69527 If the XI2 type is 0 (e.g. proximity events), inputMasks is never set and
69528 may not be NULL.
69529
69530 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69531
69532commit 68d86adbe8d84099812c37a04098f2ab5ccc45a4
69533Author: Peter Hutterer <peter.hutterer@who-t.net>
69534Date: Fri May 29 12:03:07 2009 +1000
69535
69536 dix: GetProximityEvents doesn't need extra events for valuators.
69537
69538 With internal events, we only have one event for all the data, no need to
69539 calculate for extra events.
69540
69541 Reported-by: Thomas Jaeger
69542 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69543
69544commit 71a83d68ab07536c0778d13cd7da974c40e77457
69545Author: Adam Jackson <ajax@redhat.com>
69546Date: Thu May 28 15:32:03 2009 -0400
69547
69548 xfree86: Remove loader symbol list ABI stubs
69549
69550 These have been nops since 2006 and functionally void since 7.0.
69551
69552commit d7aef3f663f4b5d861799f8615dcd301d5ce2906
69553Merge: a25f248 e08c6a0
69554Author: Peter Hutterer <peter.hutterer@who-t.net>
69555Date: Thu May 28 17:20:58 2009 +1000
69556
69557 Merge branch 'master' into xi2
69558
69559 Conflicts:
69560 Xext/geext.c
69561 Xi/chdevcur.c
69562 Xi/extgrbdev.c
69563 Xi/xiproperty.c
69564 configure.ac
69565 dix/ptrveloc.c
69566 hw/xfree86/common/xf86Config.c
69567 mi/mipointer.h
69568 test/input.c
69569 xkb/xkb.c
69570
69571commit a25f248fc3bd0375d91ca8a44320200d445ecfbb
69572Author: Peter Hutterer <peter.hutterer@who-t.net>
69573Date: Wed May 27 15:47:12 2009 +1000
69574
69575 Xi: Send Enter or Leave events with XIPassive(Un)grabNotify
69576
69577 If a passive enter or focus in grab activates, send additional enter or
69578 focus events with mode XIPassiveGrabNotify to the grabbing client.
69579 Likewise, if the grab deactivates, send additional leave or focus out
69580 events.
69581
69582 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69583
69584commit 2107becb0ce2ffda001be65728c53563496d8d50
69585Author: Adam Jackson <ajax@redhat.com>
69586Date: Wed May 27 17:13:19 2009 -0400
69587
69588 xfree86: Remove unused (and useless) xf86{Add,Delete}ModuleInfo
69589
69590commit e08c6a0752772745f35f7afcf6c2b1c927b91cc0
69591Author: Federico Mena Quintero <federico@novell.com>
69592Date: Mon May 4 19:24:47 2009 -0500
69593
69594 randr: bug #21554 - re-probe outputs when coming back from laptop unsuspend
69595
69596 Make xf86 RANDR wrap the EnterVT call chain, so it can re-probe the
69597 outputs when a laptop comes back from suspend/unsuspend (actually, any
69598 time that we enter our VT again). The X server should then send RR*
69599 events to clients, so they can cope with a monitor that was unplugged
69600 while the laptop was suspended.
69601
69602 Signed-off-by: Federico Mena Quintero <federico@novell.com>
69603
69604commit b12010e10f38951358b48ff1076c026f943b7cc3
69605Merge: 6e69272 f1441b8
69606Author: Søren Sandmann Pedersen <sandmann@redhat.com>
69607Date: Wed May 27 15:56:30 2009 -0400
69608
69609 Merge branch 'master' of git+ssh://sandmann@git.freedesktop.org/git/xorg/xserver
69610
69611commit 6e6927247359cc0db47501fb9d2dbb7856aa5985
69612Author: Søren Sandmann Pedersen <sandmann@redhat.com>
69613Date: Wed May 27 15:51:19 2009 -0400
69614
69615 Fix alpha map computation in miComputeCompositeRegion()
69616
69617 According to the RENDER spec, the origin of the alpha map is
69618 interpreted relative to the origin of the drawable of the image, not
69619 the origin of the drawable of the alpha map.
69620
69621 This commit fixes that and adds an alpha-test.c test program.
69622
69623 The only use of alpha maps I have been able to find is in Qt and they
69624 don't use a non-zero alpha origin.
69625
69626commit f1441b83538e6ba8a1557c4f5ea50c53a1837087
69627Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
69628Date: Wed May 27 10:06:10 2009 -0700
69629
69630 XQuartz: More localization updates
69631 (cherry picked from commit b107390c9f30571902556b4d053e62d345a06254)
69632
69633commit 996357e905c1082479bb238110b93bc170b8cb84
69634Author: Joe Krahn <jkrahn@nc.rr.com>
69635Date: Tue Jan 20 17:22:37 2009 +0000
69636
69637 Cygwin/X: Update icons directly, rather than modifying the window's class
69638
69639 Update icons directly to windows rather than modifying
69640 the window's class. Respect custom icons overriden via
69641 the configuration file.
69642
69643 fd.o bugzilla #4491
69644
69645 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69646
69647commit 9955252564922e7e8702dfb6eab52c75605a4a22
69648Author: Jon TURNEY <jon.turney@dronecode.org.uk>
69649Date: Mon Jan 5 17:55:58 2009 +0000
69650
69651 Cygwin/X: Remove unused TimeSinceLastInputEvent()
69652
69653 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69654
69655commit 966c8fccf16e77c8f459b4b175718323f2ec5d58
69656Author: Colin Harrison <colin.harrison@virgin.net>
69657Date: Mon Jan 5 18:04:42 2009 +0000
69658
69659 Xming: Fix some log spam
69660
69661 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69662
69663commit d190a27a33ab8c198568baf184fad14df57e24ad
69664Author: Colin Harrison <colin.harrison@virgin.net>
69665Date: Thu Jan 29 11:00:49 2009 +0000
69666
69667 Xming: Various tidy ups in winClipboardFlushXEvents()
69668
69669 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69670
69671commit fbe9ed27b1c2f932a74aa6ced134ecec021a3fc1
69672Author: Colin Harrison <colin.harrison@virgin.net>
69673Date: Thu Jan 29 10:55:38 2009 +0000
69674
69675 Xming: Avoid log spam if a windows application won't provide CF_UNICODETEXT clipboard format
69676
69677 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69678
69679commit 71ba9856a4f01aa7a42f1178c8da98a2e5ac23ae
69680Author: Colin Harrison <colin.harrison@virgin.net>
69681Date: Tue Feb 3 15:52:11 2009 +0000
69682
69683 Xming: Cache atom lookups in clipboard integration code
69684
69685 Cache the CLIPBOARD atom lookups in winClipboardWindowProc()
69686 Cache atom lookups in winClipboardFlushXEvents()
69687 Recache on server regeneration
69688
69689 Copyright (C) Colin Harrison 2005-2008
69690 http://www.straightrunning.com/XmingNotes/
69691 http://sourceforge.net/projects/xming/
69692
69693 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69694
69695commit 4ec110327bdc4f2617cd6116539812d7ef96b24e
69696Author: Colin Harrison <colin.harrison@virgin.net>
69697Date: Sun Jan 4 18:29:49 2009 +0000
69698
69699 Xming: Prevent the mouse wheel from stalling when another window is minimized.
69700
69701 Fix internal WM to prevent the mouse wheel from stalling when another window is minimized.
69702
69703 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69704
69705commit 888e6961a4ee76d05d212cfb946f089caafb3f69
69706Author: Jon TURNEY <jon.turney@dronecode.org.uk>
69707Date: Wed Jan 7 18:29:16 2009 +0000
69708
69709 Cygwin/X: Correctly allow for the native window frame width in ValidateSizing()
69710
69711 Fix internal WM to correctly calculate the native window border when validating window sizing
69712
69713 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69714
69715commit a72865868f03b675f86990476fcee601822894b3
69716Author: Joe Krahn <jkrahn@nc.rr.com>
69717Date: Tue Jan 20 17:09:43 2009 +0000
69718
69719 Cygwin/X: Consolidate dialog initialization in winInitDialog()
69720
69721 Changed windialogs.c to set icons via window properties rather than class
69722 properties, and use LoadImage() for small icons, because LoadIcon() can only open
69723 large icons. Since this code is redundant across the dialogs, I put it in the
69724 winCenterDialog function, along with a few other redundant instructions, and
69725 renamed in winInitDialog().
69726
69727 Also, don't bogusly put our dialogs at the center of the virtual desktop if we
69728 are on a multimonitor system (this causes the dialog to end up split across two
69729 monitors in a dual-monitor side-by-side setup)
69730
69731 Corrections to use HWND_TOPMOST instead of HWND_TOP and not to use SWP_NOZORDER
69732 from Colin Harrison
69733
69734 fd.o bugzilla #4491
69735
69736 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69737
69738commit d053c3b35f730fdaddd22f841dbd8ed6c9a07ba0
69739Author: Colin Harrison <colin.harrison@virgin.net>
69740Date: Sun Nov 2 20:46:40 2008 +0000
69741
69742 Xming: Slightly adjust the formatting of the logged command line
69743
69744 Copyright (C) Colin Harrison 2005-2008
69745 http://www.straightrunning.com/XmingNotes/
69746 http://sourceforge.net/projects/xming/
69747
69748 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
69749
69750commit a7e23a79c1fc429aedbf9b6c0e78b1c8d7e02238
69751Author: Peter Hutterer <peter.hutterer@who-t.net>
69752Date: Tue May 26 14:42:25 2009 +1000
69753
69754 Xi: Add support for Enter and FocusIn grabs.
69755
69756 Enter grabs are checked for in CheckMotion(), each time the sprite window
69757 changes the current grab is deactivated (if applicable) and the new grab is
69758 activated (if applicable). Exception - if the grab is on a parent window of
69759 the current window since we keep the grab across descendants.
69760
69761 Since CheckMotion() may change the grab status of a device, we mustn't get
69762 "dev->deviceGrab.grab" in ProcessOtherEvents until after CheckMotion().
69763 FocusIn grabs are checked in much the same manner.
69764
69765 The event delivery for grabs replaces the NotifyNormal on window change with
69766 a NotifyGrab on window change. Note that this happens before the grab
69767 activates, so the EnterNotify(NotifyGrab) is still delivered to the window,
69768 not to the grabbing client. This is in line with the core protocol semantics
69769 for NotifyGrab events.
69770
69771 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69772
69773commit ec2fe9660dbc0c16cdaca33b3b878011857e0fe2
69774Author: Peter Hutterer <peter.hutterer@who-t.net>
69775Date: Wed May 27 14:18:51 2009 +1000
69776
69777 Require inputproto 1.9.99.10
69778
69779commit 0ef6ba9f3b9e68307bfe623e12dc36d44440e97f
69780Author: Peter Hutterer <peter.hutterer@who-t.net>
69781Date: Wed May 27 11:15:54 2009 +1000
69782
69783 Xi: set the passive grab's resource mask on ungrabbing.
69784
69785 Not having the resource mask set means we never match an existing grab,
69786 hence we never actually ungrab.
69787
69788 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69789
69790commit 1f61d6647f25add487c5cae9739643a6e33ef9f7
69791Author: Peter Hutterer <peter.hutterer@who-t.net>
69792Date: Wed May 27 11:15:23 2009 +1000
69793
69794 Xi: use GetMaster(MASTER_KEYBOARD) to get the modifier device for XI2 pgrabs
69795
69796 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69797
69798commit bedfd561fa534b65baea45c4a05b1752ef172c45
69799Author: Peter Hutterer <peter.hutterer@who-t.net>
69800Date: Tue May 26 15:28:33 2009 +1000
69801
69802 dix: fix wrong cast of eventMask into CreateGrab.
69803
69804commit 023f959e6812e63f9d38b078158275bc13d238ae
69805Author: Peter Hutterer <peter.hutterer@who-t.net>
69806Date: Tue May 26 14:56:40 2009 +1000
69807
69808 Xi: fix an inputInfo.keyboard usage.
69809
69810 In the case of a RevertToFollowKeyboard, the master device should be used
69811 (since this is the closest equivalent to the VCK as before). Only if the
69812 master keyboard is the same as the device, revert to the VCK itself.
69813
69814commit b3463fd6a89d50a7562c357aeb5314ce0fd6493d
69815Author: Peter Hutterer <peter.hutterer@who-t.net>
69816Date: Tue May 26 13:58:04 2009 +1000
69817
69818 dix: convert window coordinates to FP1616 before calculating event_x/y
69819
69820 root_x/y are already in FP1616 at this point.
69821
69822 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69823
69824commit 10b5e981016132b4c07ad9ce68646d0a385e9c24
69825Author: Peter Hutterer <peter.hutterer@who-t.net>
69826Date: Tue May 26 11:12:08 2009 +1000
69827
69828 dix: take grabs into account for XI_Enter/Leave events.
69829
69830commit 1815defdb2f19e79f4ec0354d188ea763ff1f15e
69831Author: Peter Hutterer <peter.hutterer@who-t.net>
69832Date: Tue May 26 10:08:46 2009 +1000
69833
69834 Xi: only deactivate passive grabs if the event type matches.
69835
69836 This didn't use to be a problem when devices could only be pointers or
69837 keyboards, not both. Nowadays, slave devices may have both buttons and
69838 keyboards, and in this case we don't want to deactivate a passive keyboard
69839 grab when a button release is detected.
69840
69841commit 9dc3810a034f067096d71b33b0b47f1f52092fe6
69842Author: Peter Hutterer <peter.hutterer@who-t.net>
69843Date: Wed May 27 13:53:46 2009 +1000
69844
69845 dix: set the right grab event type for implicit passive grabs.
69846
69847 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69848
69849commit 6583477035234e23ead2fad9db7a07e5862447a4
69850Author: Nicolai Hähnle <nhaehnle@gmail.com>
69851Date: Sat May 23 13:35:24 2009 +0200
69852
69853 Remove reference to non-existing requestLog and requestLogIndex
69854
69855 These fields were removed in 252ec504817e05b185e4896a2d899e9c00b8aeef.
69856
69857 Signed-off-by: Nicolai Haehnle <nhaehnle@gmail.com>
69858 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69859
69860commit 53e821ab4a41cbc85932b9ef150311bc2e57c108
69861Author: Peter Hutterer <peter.hutterer@who-t.net>
69862Date: Mon May 25 12:20:37 2009 +1000
69863
69864 Xi: add request processing for XIGetSelectedEvents.
69865
69866 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69867
69868commit b377994336c978875dd786f4a4a9ef2e7b20e32c
69869Author: Peter Hutterer <peter.hutterer@who-t.net>
69870Date: Mon May 25 13:55:17 2009 +1000
69871
69872 Xi: advance by the right number of bytes when trawling XI2 event masks.
69873
69874 The wire layout is [struct xXIEventMask][mask bytes]. So the pointer needs
69875 to not only be advanced by the mask bytes, but also by the size of the
69876 struct.
69877
69878 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69879
69880commit 12e725d08b4cf7dbb7f09b9ec09fa1b621156ea9
69881Author: Julien Cristau <jcristau@debian.org>
69882Date: Fri May 22 09:54:38 2009 +0200
69883
69884 randr: fix server crash in RRGetScreenInfo
69885
69886 We don't return rates to randr < 1.1 clients, so don't allocate space
69887 for them. This fixes a FatalError due to not all allocated space being
69888 used.
69889
69890 X.Org bug#21861 <http://bugs.freedesktop.org/show_bug.cgi?id=21861>
69891
69892 Reported-by: Guillaume Quintin <coincoin169g@gmail.com>
69893 Signed-off-by: Julien Cristau <jcristau@debian.org>
69894
69895commit c70511931ca1441878daed50e710fa514f6ab077
69896Author: Matthieu Herrb <matthieu.herrb@laas.fr>
69897Date: Sun May 24 10:32:38 2009 +0200
69898
69899 fix typo in cabff9007 which led to an unintialized memory read and a crash.
69900
69901commit 99e22b86c5f1a3653f3caaf01368a777d2b208d0
69902Author: Adam Jackson <ajax@redhat.com>
69903Date: Fri May 22 12:01:55 2009 -0400
69904
69905 EDID: Add modes from Established Timings III descriptor to mode pool
69906
69907 EDID 1.4, section 3.10.3.9
69908
69909commit b395da91c531d633ec47bb5a51e361d321663a3a
69910Author: Peter Hutterer <peter.hutterer@who-t.net>
69911Date: Fri May 22 15:24:46 2009 +1000
69912
69913 Xi: silence two compiler warnings
69914
69915 chdevcur.c:97: warning: ‘SecurityLookupIDByType’ is deprecated (declared at
69916 ../include/resource.h:269)
69917
69918 xiproperty.c:200: warning: passing argument 2 of ‘GetEventFilter’ from
69919 incompatible pointer type
69920
69921 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69922
69923commit 4a36db60774640ea92a2133a3b08fbf1a48d2cb6
69924Author: Peter Hutterer <peter.hutterer@who-t.net>
69925Date: Fri May 22 15:16:43 2009 +1000
69926
69927 os: fix compiler warning "too few arguments to format"
69928
69929 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69930
69931commit da0d3baf71b34657cc235d6c6b37d548541f9449
69932Author: Peter Hutterer <peter.hutterer@who-t.net>
69933Date: Thu May 21 14:11:27 2009 +1000
69934
69935 dix: protect event generation against single-valuator devices.
69936
69937 If we have a single-axis device and it sends events it should not access
69938 non-existant memory.
69939
69940commit b25e29e8014927815d4fdc9c416bfeb2af0ecba6
69941Author: Peter Hutterer <peter.hutterer@who-t.net>
69942Date: Thu May 21 12:07:03 2009 +1000
69943
69944 mi: use GetMaster() from MIPOINTER and MISPRITE.
69945
69946 Both may in some cases be called for a SD attached to a master keyboard. In
69947 this case, we need to get the right master device (i.e. the pointer).
69948
69949 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69950
69951commit c9df51b070377ca33e48644dbc842b6e3dfb2975
69952Author: Peter Hutterer <peter.hutterer@who-t.net>
69953Date: Thu May 21 11:16:37 2009 +1000
69954
69955 input: allow for master pointers to not have a button class.
69956
69957 There's devices (e.g. some barcode readers) that have axes but no buttons.
69958 When such a device sends a motion event, the valuator and button class is
69959 copied into the master pointer (i.e. removing the button class).
69960 So we need a couple of extra sanity checks for the button class to exist.
69961
69962 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69963
69964commit ebe45e1a72a81ad22413e5d0514869b3e45a69b7
69965Author: Peter Hutterer <peter.hutterer@who-t.net>
69966Date: Wed May 20 10:55:03 2009 +1000
69967
69968 input: introduce partial class copying depending on the event.
69969
69970 Copying all classes into the master device has drawbacks for hybrid devices
69971 (devices that are both mice and keyboards). If such a device posts an event,
69972 it's key classes are moved into the VCP. The key event itself is unaffected
69973 by keyboard grabs and the like.
69974
69975 Partial class copying copies depending on the event and copies the classes
69976 into the right master device (i.e. the VCK for key events, the VCP for
69977 pointer events).
69978
69979 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69980
69981commit bc63c8a4570c989f19a036965854bceb9800ce19
69982Author: Peter Hutterer <peter.hutterer@who-t.net>
69983Date: Wed May 20 12:33:49 2009 +1000
69984
69985 dix: introduce GetMaster()
69986
69987 For hybrid devices (keys + buttons/axes) the attached master device is
69988 generally the wrong one. One shouldn't post a button event through a
69989 keyboard and vice versa.
69990
69991 GetMaster(dev) returns the right master device for the given type needed.
69992 This may be the MD paired with this device's MD.
69993
69994 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
69995
69996commit d79318f269d959d566ec66239b4c985afd61b259
69997Author: Peter Hutterer <peter.hutterer@who-t.net>
69998Date: Thu May 21 08:29:14 2009 +1000
69999
70000 dix: Add a deviceid to the DeviceChangedEvent.
70001
70002 ChangeDeviceId would actually overwrite the flags field if deviceid wasn't
70003 present. Aside from the event of course not telling which device generated
70004 it in the first place.
70005
70006 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70007
70008commit eeffb83fea2d87070beafd0c692fb0ad54d8335b
70009Author: Peter Hutterer <peter.hutterer@who-t.net>
70010Date: Mon May 18 20:03:07 2009 +1000
70011
70012 ABI_XINPUT_VERSION 6
70013
70014commit 438a4eafa5d14ab676827f7a5d58782c02db4cdf
70015Author: Peter Hutterer <peter.hutterer@who-t.net>
70016Date: Wed May 20 17:02:50 2009 +1000
70017
70018 input: remove nested union from InternalEvent.
70019
70020 There's no need for internal events to be a struct with a single nested
70021 union, we might as well make the union itself the InternalEvent.
70022
70023 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70024
70025commit b12d302df8283186ce87882c29b2b0294adb2770
70026Author: Peter Hutterer <peter.hutterer@who-t.net>
70027Date: Wed May 20 15:28:16 2009 +1000
70028
70029 Input: rename DeviceIntRec->isMaster to ->type.
70030
70031 isMaster is not enough as long as we differ between master pointers and
70032 keyboard. With flexible device classes, the usual checks for whether a
70033 master device is a pointer (currently check for ->button, ->valuators or
70034 ->key) do not work as an SD may post an event through a master and mess this
70035 check up.
70036
70037 Example, a device with valuators but no buttons would remove the button
70038 class from the VCP and thus result in the
70039 IsPointerDevice(inputInfo.pointer) == FALSE.
70040
70041 This will become worse in the future when new device classes are introduced
70042 that aren't provided in the current system (e.g. a switch class).
70043
70044 This patch replaces isMaster with "type", one of SLAVE, MASTER_POINTER and
70045 MASTER_KEYBOARD. All checks for dev->isMaster are replaced with an
70046 IsMaster(dev).
70047
70048commit add2defac7e59dcb355e5ba12b6de60861741d57
70049Author: Peter Hutterer <peter.hutterer@who-t.net>
70050Date: Wed May 20 16:23:38 2009 +1000
70051
70052 Split the signal-handler's lastSlave out into a separate variable.
70053
70054 dev->u.lastSlave was not signal safe since it was accessed by the DIX and
70055 during signal handling.
70056 Replaced with:
70057 'dev->last.slave' for the signal handler's lastSlave (used to generate
70058 DeviceChangedEvents), .
70059 'dev->u.lastSlave' for the DIX lastSlave (currently only used in
70060 change_modmap)
70061
70062 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70063
70064commit 525aa17f804d37d1cfcbbf6b8e6cddb45e999b20
70065Author: Tomas Janousek <tomi@nomi.cz>
70066Date: Wed May 20 15:03:01 2009 +0200
70067
70068 Bug #6428, #16458, #21464: Fix crash due to uninitialized VModMap fields.
70069
70070 In ProcXkbGetKbdByName, mrep.firstVModMapKey, .nVModMapKeys and
70071 .totalVModMapKeys were not initialized, contained random values and caused
70072 accesses to unallocated and later modified memory, causing
70073 XkbSizeVirtualModMap and XkbWriteVirtualModMap to see different number of
70074 nonzero values, resulting in writes past the end of an array in XkbSendMap.
70075
70076 This patch initializes those values sensibly and reverts commits 5c0a2088 and
70077 6dd4fc46, which have been plain non-sense.
70078
70079 Signed-off-by: Tomas Janousek <tomi@nomi.cz>
70080 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70081
70082commit 7db55a0806c82bd4143c8bf1b8eb2b62e456ad9a
70083Author: Peter Hutterer <peter.hutterer@who-t.net>
70084Date: Wed May 20 14:38:25 2009 +1000
70085
70086 dix: remove superfluous loop in change_modmap.
70087
70088 A device can only be attached to a single master device. So instead of
70089 looping and searching for the master device, we can just use dev->u.master
70090 directly.
70091
70092 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70093
70094commit cabff9007a4abad026b450a4aab155c7bcd94326
70095Author: Peter Hutterer <peter.hutterer@who-t.net>
70096Date: Fri May 15 08:33:07 2009 +1000
70097
70098 xfree86: treat other drivers as mouse drivers in the config.
70099
70100 Historically, if no input device was referenced in the ServerLayout,
70101 the server would pick the first "mouse" device found in the xorg.conf.
70102 This patch gives evdev, synaptics, vmmouse and void the same status. If
70103 there is a section in the config file using this driver - use it as the core
70104 pointer.
70105
70106 Device selection is in driver-order, not in config-order. If a "mouse"
70107 device is listed after a "synaptics" device, the "mouse" device gets
70108 preference. This replicates the original behaviour.
70109
70110 This code only takes effect if AllowEmptyInput is off and there is no core
70111 pointer in the server layout.
70112
70113 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70114
70115commit a8bd1e1f96d8e5380972a7dce8d9940cd912aa09
70116Author: Søren Sandmann Pedersen <ssp@l3000.localdomain>
70117Date: Thu May 21 18:21:28 2009 -0400
70118
70119 Fix alpha map computation in miComputeCompositeRegion()
70120
70121 According to the RENDER spec, the origin of the alpha map is
70122 interpreted relative to the origin of the drawable of the image, not
70123 the origin of drawable of the alpha map.
70124
70125 The only use of alpha maps I have been able to find is in Qt and they
70126 don't use a non-zero alpha origin.
70127
70128commit a2c5ee36b21c2ee5c0468f1b251e74c1412dbecb
70129Author: Adam Jackson <ajax@redhat.com>
70130Date: Thu May 21 10:20:48 2009 -0400
70131
70132 EDID: Be more cautious about finding vendor blocks.
70133
70134 Many old monitors zero-fill the detailed descriptors, so check for that
70135 to avoid a useless warning like:
70136
70137 (WW) RADEON(0): Unknown vendor-specific block 0
70138
70139commit 73c7398aafa050950ad614483fb0d5bdd2da97ca
70140Author: Peter Hutterer <peter.hutterer@who-t.net>
70141Date: Wed May 20 10:40:50 2009 +1000
70142
70143 dix: 'namespace' HAS_OLD_SLAVE and HAS_NEW_SLAVE.
70144
70145 We need more flags for this in the near future, so let's namespace them now.
70146
70147commit 5b5e3fa2771383a85afff679be34df19d3a4e290
70148Author: Peter Hutterer <peter.hutterer@who-t.net>
70149Date: Fri May 15 08:33:07 2009 +1000
70150
70151 xfree86: treat other drivers as mouse drivers in the config.
70152
70153 Historically, if no input device was referenced in the ServerLayout,
70154 the server would pick the first "mouse" device found in the xorg.conf.
70155 This patch gives evdev, synaptics, vmmouse and void the same status. If
70156 there is a section in the config file using this driver - use it as the core
70157 pointer.
70158
70159 Device selection is in driver-order, not in config-order. If a "mouse"
70160 device is listed after a "synaptics" device, the "mouse" device gets
70161 preference. This replicates the original behaviour.
70162
70163 This code only takes effect if AllowEmptyInput is off and there is no core
70164 pointer in the server layout.
70165
70166 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70167
70168commit 1cce55cc0333a026474100cf2f784e220fd96473
70169Author: Peter Hutterer <peter.hutterer@who-t.net>
70170Date: Wed May 20 14:07:03 2009 +1000
70171
70172 input: rename device->type to device->xinput_type.
70173
70174 This type is only used in XI to give a hint of what type this device may be.
70175 Call it xinput_type for clarity.
70176
70177 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70178
70179commit 81b3b0cce088866dc3cda099d7c8d6655849fd43
70180Author: Tomas Janousek <tomi@nomi.cz>
70181Date: Wed May 20 15:03:01 2009 +0200
70182
70183 Bug #6428, #16458, #21464: Fix crash due to uninitialized VModMap fields.
70184
70185 In ProcXkbGetKbdByName, mrep.firstVModMapKey, .nVModMapKeys and
70186 .totalVModMapKeys were not initialized, contained random values and caused
70187 accesses to unallocated and later modified memory, causing
70188 XkbSizeVirtualModMap and XkbWriteVirtualModMap to see different number of
70189 nonzero values, resulting in writes past the end of an array in XkbSendMap.
70190
70191 This patch initializes those values sensibly and reverts commits 5c0a2088 and
70192 6dd4fc46, which have been plain non-sense.
70193
70194 Signed-off-by: Tomas Janousek <tomi@nomi.cz>
70195 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70196
70197commit 79138eec1b49cbaca6a16f2bdd8579b5828aeb28
70198Author: Matthias Hopf <mhopf@suse.de>
70199Date: Tue Mar 10 16:03:26 2009 +0100
70200
70201 randr: Setting gamma: inverse logic looks more sane
70202
70203commit c2785ae7eb6197bbfc75e92e99fffbb8ad8064da
70204Author: Peter Hutterer <peter.hutterer@who-t.net>
70205Date: Wed May 20 14:29:13 2009 +1000
70206
70207 dix: refuse events from disabled devices.
70208
70209 If the device is disabled ("off"), it must not send events to a client.
70210 The driver shouldn't send events in that case anyway, but just to make sure
70211 we simply drop events coming while the device is disabled.
70212
70213 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70214
70215commit 47a89b1cba67a43ddc225f7e1b5d2b3b217c8d93
70216Author: Peter Hutterer <peter.hutterer@who-t.net>
70217Date: Wed May 20 13:59:22 2009 +1000
70218
70219 Xi: remove DeviceIsPointerType
70220
70221 This approach is broken anyway. DIPT only checked for the XInput type
70222 "MOUSE" and the only user of this is xf86ActivateDevice when it sets the
70223 Activate/DeactivateGrab functions.
70224 Since synaptics and wacom set their own types, evdev only sets MOUSE for,
70225 well, mice half the devices didn't have this set correctly anyway.
70226
70227 Instead, ActivatePointerGrab should be merged together with
70228 ActivateKeyboardGrab.
70229
70230 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70231
70232commit 7b891e450f005688e41e61016e6d3e394c67f6be
70233Author: Peter Hutterer <peter.hutterer@who-t.net>
70234Date: Wed May 20 14:23:16 2009 +1000
70235
70236 dix: Remove obsolete comment
70237
70238 CorePointerProc doesn't back up the device classes anymore.
70239
70240commit 42719ce5c715ac3a0c74c3c7df872d35e0449de9
70241Author: Peter Hutterer <peter.hutterer@who-t.net>
70242Date: Wed May 20 14:38:25 2009 +1000
70243
70244 dix: remove superfluous loop in change_modmap.
70245
70246 A device can only be attached to a single master device. So instead of
70247 looping and searching for the master device, we can just use dev->u.master
70248 directly.
70249
70250 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70251
70252commit 22a33a7250f879b8f3eb84176a1dc57027d8cfe4
70253Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
70254Date: Tue May 19 22:45:26 2009 -0400
70255
70256 glx: Register names for the GLX resource types.
70257
70258 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
70259
70260commit 77e90261d5f91e1170c6e02deb0e3739c24fcfd6
70261Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
70262Date: Tue May 19 19:36:03 2009 -0400
70263
70264 dix/events.c: Use wClient where appropriate.
70265
70266 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
70267
70268commit 3cea176d5abcb0f14eefbdcbe17fed0847524dd4
70269Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
70270Date: Tue May 19 19:30:33 2009 -0400
70271
70272 xace: Fix a bad device access hook call.
70273
70274 Add a proper access mode, and reverse the logic of the return value.
70275 Zero ("Success") is returned on success from the hook calls.
70276
70277 Signed-off-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
70278
70279commit 4addfcd633484b7f53b5df1b92bfca8bfc1dc48d
70280Author: Michel Dänzer <daenzer@vmware.com>
70281Date: Mon May 18 17:53:35 2009 +0200
70282
70283 EXA: Allocate from the end of free offscreen memory rather than from the start.
70284
70285 This way we don't always need to scan over previously allocated areas when
70286 looking for an available one, and there might be less fragmentation.
70287
70288 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
70289
70290commit 510cbd43cd4e34bd459e8f74ab2855714b4ca95d
70291Author: Michel Dänzer <daenzer@vmware.com>
70292Date: Mon May 18 17:48:57 2009 +0200
70293
70294 EXA: Defragment offscreen memory.
70295
70296 At most once per second, under the following circumstances:
70297
70298 * We can't satisfy an offscreen memory allocation, but there seems to be enough
70299 offscreen memory available in total.
70300
70301 or
70302
70303 * The server has been idle for at least 100ms, and there is more than one
70304 available offscreen area.
70305
70306 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
70307
70308commit 8331bde0adeccefb275c4d707e7b2cb1d95b1581
70309Author: Jon TURNEY <jon.turney@dronecode.org.uk>
70310Date: Fri May 15 21:36:28 2009 +0100
70311
70312 Cygwin/X: link with libmain.a to provide main()
70313
70314 Commit 987579c930bda803427a28cb82773c389f5110d6 moves main.c into
70315 into libmain.a to be linked with separately, and updates the various
70316 DDXs to link with libmain.a, except Xwin, which also needs this change.
70317
70318 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
70319
70320commit b89dcfbfbd38f8d19f1d4f24f81820b1048572c8
70321Author: Peter Hutterer <peter.hutterer@who-t.net>
70322Date: Mon May 18 16:07:08 2009 +1000
70323
70324 Xi: fix length field in XIQueryPointer.
70325
70326 Buttons append 4-byte units to the end of the reply, they need to be
70327 included too.
70328
70329 Reported-by: Mark Dokter
70330 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70331
70332commit 33fcaaaea5ea30be39156ecfdbcca891b47ab465
70333Author: Peter Hutterer <peter.hutterer@who-t.net>
70334Date: Mon May 18 16:05:35 2009 +1000
70335
70336 Fix missing parentheses in FP1616 macro.
70337
70338 Missing parens led to interesting results if an expression instead of a
70339 constant was passed in (ProcXIQueryPointer for example).
70340
70341commit 6258bb1d6bb78ea4ba1d33470bc74e9b711b25f2
70342Author: Peter Hutterer <peter.hutterer@who-t.net>
70343Date: Fri May 15 20:24:45 2009 +1000
70344
70345 require inputproto 1.9.99.9
70346
70347commit e7e94ff5fd3205a9ee01fe9342d9f3a426b0a068
70348Author: Peter Hutterer <peter.hutterer@who-t.net>
70349Date: Thu May 14 21:45:35 2009 +1000
70350
70351 Xi: set the right length for the XIQueryPointer reply.
70352
70353commit 00ab04d73c612ac85e024105015bdb7c79e8874b
70354Author: Peter Hutterer <peter.hutterer@who-t.net>
70355Date: Thu May 14 16:24:35 2009 +1000
70356
70357 Include full modifier + button state in XIQueryPointer.
70358
70359commit de78aab0ea3a2d8c61222493f25911258dea65a2
70360Author: Peter Hutterer <peter.hutterer@who-t.net>
70361Date: Thu May 14 15:53:00 2009 +1000
70362
70363 Xi: take the paired device as modifier device only for MD pointers.
70364
70365commit 0e089f9ceb0edec532078c6e1df7f642f7220fb5
70366Author: Peter Hutterer <peter.hutterer@who-t.net>
70367Date: Thu May 14 15:29:20 2009 +1000
70368
70369 Xi: check cursor and grab_window before attempting the passive grab.
70370
70371 Both values need to fail the request, so we need to check them before
70372 looping through the modifier masks. Otherwise, a wrong grab_window will
70373 show up in each failed modifier return but not actually cause an error.
70374
70375commit 5c3e7289801c958e2c70e2629d3ebee122449be8
70376Author: Peter Hutterer <peter.hutterer@who-t.net>
70377Date: Wed May 13 16:54:01 2009 +1000
70378
70379 dix: count must be 1 if we have an XI2 match.
70380
70381 This fixes events that activated a grab not getting delivered to the client.
70382
70383 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70384
70385commit e083a0d0692baa5ad29002a38293b36dd3b4c06f
70386Author: Peter Hutterer <peter.hutterer@who-t.net>
70387Date: Wed May 13 15:23:42 2009 +1000
70388
70389 Xi: set the correct length for XIPassiveGrab replies.
70390
70391commit 8ff1bff8f462b15e05fc298704e1e40cb244587f
70392Author: Peter Hutterer <peter.hutterer@who-t.net>
70393Date: Tue May 12 21:33:54 2009 +1000
70394
70395 Update to new XI2 names
70396
70397commit bbf266cca3de726e3eb932d66dd5b8981ef40fa8
70398Author: Peter Hutterer <peter.hutterer@who-t.net>
70399Date: Tue May 12 21:24:07 2009 +1000
70400
70401 mi: un-deprecate miPointerWarpCursor
70402
70403 This function was deprecated in ef68273f5bdb27a492ec0b69548ec4fbede46c08
70404 because it didn't take a device argument. The device argument was added in
70405 1c7568b8a1417257fa67c7fca69aa253099b9461 though, so the deprecation is
70406 obsolete.
70407
70408 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70409
70410commit 273890924b8ed6f8b7949c0322c8258b9e6f8630
70411Author: Peter Hutterer <peter.hutterer@who-t.net>
70412Date: Tue May 12 16:09:35 2009 +1000
70413
70414 input: reduce the number of superfluous hierarchy events
70415
70416 There's only two reasons for hierarchy events:
70417 - device is added, removed, etc. In this case we want to send the event as
70418 it happens.
70419 - devices are added in a XIChangeDeviceHierarchy request. In this case we
70420 only want one event cumulating all changes.
70421
70422commit 033a2b12fcd02fa9a2c2f20a352bec0a43074512
70423Author: Peter Hutterer <peter.hutterer@who-t.net>
70424Date: Tue May 12 14:50:57 2009 +1000
70425
70426 Xi: set per-device hierarchy changed flags.
70427
70428 Rather than have one field per hierarchy change, XI2 has two fields - one
70429 generic one and one per-device that include the device-specific flags.
70430 This requires some funky handling for removed devices, but oh well.
70431
70432commit 8fb51feae222ff4f4aa1c440b6973ce7383bbc09
70433Author: Peter Hutterer <peter.hutterer@who-t.net>
70434Date: Mon May 11 15:45:46 2009 +1000
70435
70436 xkb: if kbd init failed, NULL out the pointers after freeing them (#21278)
70437
70438 Reproducible:
70439 Configure a server that uses the keyboard driver with an invalid ruleset,
70440 e.g. (Option "XkbRules" "foobar"). Ensure that Option "AllowEmptyInput" is
70441 "off" in the ServerFlags or ServerLayout section. Start the server.
70442 After failing to init the keymap, the server will try to clean up after the
70443 device, double-freeing some xkb structs that have not been reset properly.
70444
70445 X.Org Bug 21278 <http://bugs.freedesktop.org/show_bug.cgi?id=21278>
70446
70447 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70448
70449commit 0cea199eda772a5ddd5ce72e4823a00ec14405d4
70450Author: Peter Hutterer <peter.hutterer@who-t.net>
70451Date: Mon May 11 12:53:56 2009 +1000
70452
70453 kdrive: set Activate/Deactivate grab for input devices (#21591)
70454
70455 X.Org Bug 21591 <http://bugs.freedesktop.org/show_bug.cgi?id=21591>
70456
70457commit fdce58ca822f4e43e03e84076190f52d7104e01f
70458Author: Peter Hutterer <peter.hutterer@who-t.net>
70459Date: Mon May 11 12:51:40 2009 +1000
70460
70461 dix: ensure Activate/DeactivateGrab has a valid value.
70462
70463 Xephyr doesn't manually set Activate/DeactivateGrab for new devices,
70464 resulting in a NULL-pointer dereference later when a grab is activated.
70465 Avoid the segfault by ensuring that the pointer is always valid.
70466
70467 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70468
70469commit e3f296d91dfe6b827195e1d387e1a04aa73b85c3
70470Author: Peter Hutterer <peter.hutterer@who-t.net>
70471Date: Sat May 9 10:33:20 2009 +1000
70472
70473 xfree86: Remove superfluous ifdef DEBUG checks.
70474
70475 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70476
70477commit 2f6253376df45e508284ff7766b186cfc7fb56d6
70478Author: Peter Hutterer <peter.hutterer@who-t.net>
70479Date: Sat May 9 10:33:20 2009 +1000
70480
70481 xfree86: Remove superfluous ifdef DEBUG checks.
70482
70483 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70484
70485commit b1b5ec45c1cb650ccb8c659218f9481379c777d9
70486Author: Peter Hutterer <peter.hutterer@who-t.net>
70487Date: Mon May 11 12:53:56 2009 +1000
70488
70489 kdrive: set Activate/Deactivate grab for input devices (#21591)
70490
70491 X.Org Bug 21591 <http://bugs.freedesktop.org/show_bug.cgi?id=21591>
70492
70493commit 6f9e22049862ee9ac7f604411d005d8bb1b2dd1c
70494Author: Peter Hutterer <peter.hutterer@who-t.net>
70495Date: Mon May 11 12:51:40 2009 +1000
70496
70497 dix: ensure Activate/DeactivateGrab has a valid value.
70498
70499 Xephyr doesn't manually set Activate/DeactivateGrab for new devices,
70500 resulting in a NULL-pointer dereference later when a grab is activated.
70501 Avoid the segfault by ensuring that the pointer is always valid.
70502
70503 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70504
70505commit c1d7deca9281ba1df0df1abcea1fe7f841b10ff9
70506Author: Peter Hutterer <peter.hutterer@who-t.net>
70507Date: Tue May 12 21:24:07 2009 +1000
70508
70509 mi: un-deprecate miPointerWarpCursor
70510
70511 This function was deprecated in ef68273f5bdb27a492ec0b69548ec4fbede46c08
70512 because it didn't take a device argument. The device argument was added in
70513 1c7568b8a1417257fa67c7fca69aa253099b9461 though, so the deprecation is
70514 obsolete.
70515
70516 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70517
70518commit 8609a4e883e917d1652fbe810108de99bc89713f
70519Author: Alan Coopersmith <alan.coopersmith@sun.com>
70520Date: Fri May 15 11:22:32 2009 -0700
70521
70522 Correct some Sun license notices to Sun's standard X11 license format
70523
70524 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
70525
70526commit 94c5093c8bc803d0b14429b5a4d07ae46da865cc
70527Author: David Marx <David.Marx@Sun.COM>
70528Date: Wed May 13 17:53:30 2009 -0700
70529
70530 Solaris: Make sure non-inline versions of asm routines end with ret statements
70531
70532 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
70533
70534commit 2be5eecb4b0f30b9c55d12521415edbb88c1f8fd
70535Author: Julien Cristau <jcristau@debian.org>
70536Date: Fri May 15 17:29:32 2009 +0200
70537
70538 xfree86: add edid quirk for Samsung Syncmaster 2333HD
70539
70540 It reports vertical size in cm in the detailed mode.
70541 X.Org bug#21750 <http://bugs.freedesktop.org/show_bug.cgi?id=21750>
70542
70543 Reported-by: Peter Poklop <Peter.Poklop@gmx.at>
70544 Signed-off-by: Julien Cristau <jcristau@debian.org>
70545
70546commit 7c8327f0a75087a85864256a9cea80dd4b86def5
70547Author: Michel Dänzer <daenzer@vmware.com>
70548Date: Fri May 15 15:48:37 2009 +0200
70549
70550 EXA: Always damage glyph cache pixmap manually after uploading a glyph.
70551
70552 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
70553
70554commit 850675d4de4373e5df95507dbf2cd9affaaf54bc
70555Author: Michel Dänzer <daenzer@vmware.com>
70556Date: Fri May 15 15:48:37 2009 +0200
70557
70558 EXA: Take GC client clip type into account for migration.
70559
70560 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=18950 .
70561
70562 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
70563
70564commit c1bbac2708bbccbec01cecba39cdf2b8160c1dc3
70565Author: Matt Turner <mattst88@gmail.com>
70566Date: Thu May 14 16:00:41 2009 -0400
70567
70568 Fix mem_barrier() on Alpha
70569
70570 Some clean up (da086901c6579e41b28b1418fa12ebb82744fc87) went a bit
70571 too far and removed the needed mem_barrier macro on Alpha.
70572
70573 This re-adds it.
70574
70575commit 4cc33190ef1a4207c0f8fc74201cdf2568a239c2
70576Author: Julien Cristau <jcristau@debian.org>
70577Date: Thu May 14 17:01:57 2009 +0200
70578
70579 xfree86: Add two __FreeBSD_kernel__ checks
70580
70581 Debian bug#525475 <http://bugs.debian.org/525475>
70582
70583 Reported-by: Petr Salinger <Petr.Salinger@seznam.cz>
70584 Signed-off-by: Julien Cristau <jcristau@debian.org>
70585
70586commit ed9aecf8517bf74e943e3e5dd3ee407524e8f122
70587Author: Adam Jackson <ajax@redhat.com>
70588Date: Thu May 14 06:55:21 2009 -0400
70589
70590 COPYING: SGI FreeB 1.1 -> SGI FreeB 2.0
70591
70592commit c66cf7b64c7bcb31c81b9fde27cc40c1440753b1
70593Author: Adam Jackson <ajax@redhat.com>
70594Date: Thu May 14 06:53:08 2009 -0400
70595
70596 Update several of my and/or Red Hat's licenses to standard form.
70597
70598commit 2075d4bf9e53b8baef0b919da6c44771220cd4a5
70599Author: Michel Dänzer <daenzer@vmware.com>
70600Date: Thu May 14 11:46:41 2009 +0200
70601
70602 glx: If a destroyed window is bound to the current context, make it not current.
70603
70604 Avoids subsequent crashes due to stale pointers to the DrawableRec, see
70605 https://bugs.freedesktop.org/show_bug.cgi?id=21132#c15 and previous comments.
70606
70607 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
70608
70609commit 2c1190f888515292de01e60fe74657c34b99fd9e
70610Author: Michel Dänzer <daenzer@vmware.com>
70611Date: Thu May 14 11:39:16 2009 +0200
70612
70613 randr12: Initialize and keep track of updates to VidMode extension gamma value.
70614
70615 This way clients querying the gamma value via the VidMode extension at least
70616 get the last value set via the same, rather than always something bogus.
70617
70618 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
70619
70620commit fc3ce861cdab8606610726ce7c53f57d950c2407
70621Author: Michel Dänzer <daenzer@vmware.com>
70622Date: Thu May 14 11:35:25 2009 +0200
70623
70624 randr12: Fix calculation of gamma ramp values.
70625
70626 The reciprocal gamma value was missed in the first copy and this mistake was
70627 propagated to the second one.
70628
70629 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
70630
70631commit f3c61377d0fb1f09a18833995556fc9ffd11e097
70632Author: Alan Coopersmith <alan.coopersmith@sun.com>
70633Date: Tue May 12 18:27:09 2009 -0700
70634
70635 COPYING file updates for git master changes since 1.6 branch
70636
70637 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
70638
70639commit b746a00cffca5c553b607a8e9c1074294a23b443
70640Author: Alan Coopersmith <alan.coopersmith@sun.com>
70641Date: Tue May 12 16:11:01 2009 -0700
70642
70643 Resync COPYING file with notices in code base as of xorg-server-1.6.1
70644
70645 Remove notices for code no longer in tree (Xprint, Xgl, kdrive, cfb, etc.)
70646 Add/update notices for new/changed code in tree
70647
70648 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
70649
70650commit f250eea2e90fc50bec5214c2f41132b95edc2c46
70651Author: Jerome Glisse <glisse@freedesktop.org>
70652Date: Mon May 11 22:52:46 2009 +0200
70653
70654 DRI2: update DRI2 private drawable width & height according to X drawable
70655
70656commit 0952d12717031e9dda9e48123bb922d0f4e81834
70657Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
70658Date: Mon May 11 15:27:46 2009 -0400
70659
70660 xselinux: Relax ownership restriction on SetSelectionUseContext.
70661
70662 Instead, clients should keep track of the selection instances they use.
70663
70664commit ac13145dbcb284293582435409d8a90f276785c5
70665Author: Peter Hutterer <peter.hutterer@who-t.net>
70666Date: Mon May 11 15:45:46 2009 +1000
70667
70668 xkb: if kbd init failed, NULL out the pointers after freeing them (#21278)
70669
70670 Reproducible:
70671 Configure a server that uses the keyboard driver with an invalid ruleset,
70672 e.g. (Option "XkbRules" "foobar"). Ensure that Option "AllowEmptyInput" is
70673 "off" in the ServerFlags or ServerLayout section. Start the server.
70674 After failing to init the keymap, the server will try to clean up after the
70675 device, double-freeing some xkb structs that have not been reset properly.
70676
70677 X.Org Bug 21278 <http://bugs.freedesktop.org/show_bug.cgi?id=21278>
70678
70679 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70680
70681commit ebfd6688d1927288155221e7a78fbca9f9293952
70682Author: Søren Sandmann Pedersen <sandmann@redhat.com>
70683Date: Sun May 10 23:27:56 2009 -0400
70684
70685 Make compositing with transformed windows work again.
70686
70687 The coordinate translation was broken in pretty much every way
70688 imaginable.
70689
70690commit c2b668e6fff11c52f2a3f0dc3f4d1fc2c5bcf3e6
70691Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
70692Date: Tue May 5 20:27:53 2009 -0700
70693
70694 XQuartz: pbproxy standalone: added missing variable declarations.
70695 (cherry picked from commit 2d947d17a3ef5fb8cef959bff0eb08b8bb5b6548)
70696
70697commit 9c5b761c8c86cca56f4b4af272f42ba89c9c3ba5
70698Author: Peter Hutterer <peter.hutterer@who-t.net>
70699Date: Sat May 9 16:19:13 2009 +1000
70700
70701 Revert "xkb: write the _XKB_RF_RULES_PROP to each device."
70702
70703 This commit shouldn't have been pushed, we're still sorting out the API we
70704 want to use.
70705
70706 This reverts commit 876910a951053f0bd31e30098de3da15a1c1f5d6.
70707
70708 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70709
70710commit 1e816065e5ec3b9394dc1fa5815457a664e15fd9
70711Author: Alan Coopersmith <alan.coopersmith@sun.com>
70712Date: Fri May 8 21:31:01 2009 -0700
70713
70714 Don't printf NULL pointers on HAL connection error
70715
70716 Fixes Solaris bug 6801386 Xorg core dumps on startup if hald not running
70717 http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6801386
70718
70719 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
70720
70721commit b680bda34da130ce408783f04214771471e41e8d
70722Author: Alan Coopersmith <alan.coopersmith@sun.com>
70723Date: Thu Apr 30 18:49:06 2009 -0700
70724
70725 Fix a couple off-by-one array boundary checks.
70726
70727 Error: Write outside array bounds at Xext/geext.c:406
70728 in function 'GEWindowSetMask' [Symbolic analysis]
70729 In array dereference of cli->nextSib[extension] with index 'extension'
70730 Array size is 128 elements (of 4 bytes each), index <= 128
70731
70732 Error: Buffer overflow at dix/events.c:592
70733 in function 'SetMaskForEvent' [Symbolic analysis]
70734 In array dereference of filters[deviceid] with index 'deviceid'
70735 Array size is 20 elements (of 512 bytes each), index >= 0 and index <= 20
70736
70737 Error: Read buffer overflow at hw/xfree86/loader/loader.c:226
70738 in function 'LoaderOpen' [Symbolic analysis]
70739 In array dereference of refCount[new_handle] with index 'new_handle'
70740 Array size is 256 elements (of 4 bytes each), index >= 1 and index <= 256
70741
70742 These bugs were found using the Parfait source code analysis tool.
70743 For more information see http://research.sun.com/projects/parfait
70744
70745 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
70746 Signed-off-by: Adam Jackson <ajax@redhat.com>
70747 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
70748
70749commit 00bc043fa0398a1d14d46b87da2ff3031a9535dc
70750Author: Peter Hutterer <peter.hutterer@who-t.net>
70751Date: Fri May 8 10:10:25 2009 +1000
70752
70753 dix: export subpixel precision in XI2 events for root/event coordinates.
70754
70755 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70756
70757commit bae070914fc27db122e6131ae4838559c4a72f65
70758Author: Peter Hutterer <peter.hutterer@who-t.net>
70759Date: Thu May 7 16:52:31 2009 +1000
70760
70761 input: update to inputproto 1.9.99.8 XI2 defines.
70762
70763 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70764
70765commit 83f32d3972b8bfb0a87069dfb3fcd64b6b7c6424
70766Author: Peter Hutterer <peter.hutterer@who-t.net>
70767Date: Thu May 7 10:05:29 2009 +1000
70768
70769 Xi: Add XI2 property requests.
70770
70771commit 9935bec6e860cba9a3cc5baadd372ddb89d72ef0
70772Author: Peter Hutterer <peter.hutterer@who-t.net>
70773Date: Thu May 7 09:43:56 2009 +1000
70774
70775 Xi: split some code out of the XI 1.5 property request processing.
70776
70777 This is in preparation for the XI2 property requests that can re-use much of
70778 this code.
70779
70780commit f6f1e417063d2d61d65731a7e3ebca3ec2bdaacc
70781Author: Peter Hutterer <peter.hutterer@who-t.net>
70782Date: Wed May 6 22:25:28 2009 +1000
70783
70784 Xi: fix copy/paste error causing sizeof against wrong struct.
70785
70786 This wrong check may cause BadLength to be returned to the client even if the
70787 length is correct.
70788
70789 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70790
70791commit fc0013d744a345199f013ba5b6ef0e44201d0e68
70792Author: Peter Hutterer <peter.hutterer@who-t.net>
70793Date: Wed May 6 16:55:32 2009 +1000
70794
70795 Xi: don't double-swap the XListDeviceProperties reply.
70796
70797 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70798
70799commit 409012061b7d361a172b20455d4aaa5e47527c42
70800Author: Peter Hutterer <peter.hutterer@who-t.net>
70801Date: Wed May 6 16:42:46 2009 +1000
70802
70803 Xi: add missing break in XI event swapping function
70804
70805 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70806
70807commit c3c64978c4a231a3a8c18211d0716df875c75efa
70808Author: Peter Hutterer <peter.hutterer@who-t.net>
70809Date: Wed May 6 15:05:06 2009 +1000
70810
70811 Xi: fix a couple of wrong dixLookupDevice permission tags.
70812
70813 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70814
70815commit 0d947aa8e87c5d92b702c60190c8bc5d32c9ba9c
70816Author: Peter Hutterer <peter.hutterer@who-t.net>
70817Date: Wed May 6 12:07:07 2009 +1000
70818
70819 xfree86: fix xf86PostMotionEventP type checking
70820
70821 We only put internal events into the queue now, so let's check for ET_Motion
70822 rather than the MotionNotify.
70823
70824 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70825
70826commit 9457ce448988dacea9a3a1c14b6955767118ff31
70827Author: Simon Thum <simon.thum@gmx.de>
70828Date: Thu Apr 30 13:01:17 2009 +0200
70829
70830 dix: remove superfluous includes from ptrveloc.c
70831
70832 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70833
70834commit ee30e1b102d779403eeda97e007b81e85f006a37
70835Author: Simon Thum <simon.thum@gmx.de>
70836Date: Thu Apr 30 12:58:48 2009 +0200
70837
70838 dix: fix warning in pointer acceleration
70839
70840 newer gcc's warn against how this cast is done (though it eludes me why),
70841 and lrintf() is also faster especially on insane processors like the P4
70842 (http://www.mega-nerd.com/FPcast).
70843
70844 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70845
70846commit 5cf70183812541b33a6e83c7e1e3bc6198730cbe
70847Author: Peter Hutterer <peter.hutterer@who-t.net>
70848Date: Tue May 5 21:00:31 2009 +1000
70849
70850 xkb: remove _XkbAlloc, _XkbCalloc, _XkbRealloc and _XkbFree
70851
70852 We all agree that wrapping is fun, but seriously. One of these days someone
70853 will get hurt.
70854
70855 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70856
70857commit 0e31d3906deaee5d9ada66e538b9e93574a4d610
70858Author: Peter Hutterer <peter.hutterer@who-t.net>
70859Date: Tue May 5 21:07:07 2009 +1000
70860
70861 xkb: remove some now-useless XFUNCPROTOBEGIN
70862
70863 We bring them back if we start rewriting the server in C++, promise.
70864
70865 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70866
70867commit 8b583ca2b21155359c6255f406c96599b277c762
70868Author: Peter Hutterer <peter.hutterer@who-t.net>
70869Date: Wed May 6 22:25:28 2009 +1000
70870
70871 Xi: fix copy/paste error causing sizeof against wrong struct.
70872
70873 This wrong check may cause BadLength to be returned to the client even if the
70874 length is correct.
70875
70876 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70877
70878commit 7ab5e9b97c300bba793a23fa13506b0c77c50ddf
70879Author: Peter Hutterer <peter.hutterer@who-t.net>
70880Date: Wed May 6 12:07:07 2009 +1000
70881
70882 xfree86: fix xf86PostMotionEventP type checking
70883
70884 We only put internal events into the queue now, so let's check for ET_Motion
70885 rather than the MotionNotify.
70886
70887 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70888
70889commit 50cc8adafca4ba3838d468278d6eb8a4692d2488
70890Author: Peter Hutterer <peter.hutterer@who-t.net>
70891Date: Wed May 6 16:55:32 2009 +1000
70892
70893 Xi: don't double-swap the XListDeviceProperties reply.
70894
70895 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70896
70897commit 3fc6fcfb267888d65a22ed7c3eda175a2530a3d5
70898Author: Simon Thum <simon.thum@gmx.de>
70899Date: Wed May 6 10:39:16 2009 +0200
70900
70901 dix: add 'none' pointer acceleration profile with number -1
70902
70903 This is a shorthand for disabling acceleration, while retaining the
70904 possiblity to use constant deceleration. If constant deceleration is
70905 also unused, it will optimize motion processing.
70906
70907 Other possiblities to deactivate acceleration were quite hidden,
70908 and didn't always work as expected. E.g. xset m 1 1 would retain
70909 adaptive deceleration, while xset m 1 0 would not (in the default
70910 profile).
70911
70912 Also removes the 'reserved' profile; it was unused and it's trivial
70913 to add new ones anyway.
70914
70915 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70916
70917commit 9d1597cbefea6a5e7959f2099c46f1a284def7ad
70918Author: Simon Thum <simon.thum@gmx.de>
70919Date: Thu Apr 30 13:01:17 2009 +0200
70920
70921 dix: remove superfluous includes from ptrveloc.c
70922
70923 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70924
70925commit 35fce4e5c7fa34f98b3e4010c6cb09ce38a9205c
70926Author: Simon Thum <simon.thum@gmx.de>
70927Date: Thu Apr 30 12:58:48 2009 +0200
70928
70929 dix: fix warning in pointer acceleration
70930
70931 newer gcc's warn against how this cast is done (though it eludes me why),
70932 and lrintf() is also faster especially on insane processors like the P4
70933 (http://www.mega-nerd.com/FPcast).
70934
70935 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70936
70937commit d220d6907d1d5138d1528c48b739e77f65616225
70938Author: Peter Hutterer <peter.hutterer@who-t.net>
70939Date: Mon May 4 17:38:02 2009 +1000
70940
70941 Xi: add GrabButton and GrabKeysym code.
70942
70943 We don't do keycode grabs in XI2, they're pointless.
70944
70945commit 1b593ced171d02f1d00034f0f733060706bb6d41
70946Author: Peter Hutterer <peter.hutterer@who-t.net>
70947Date: Tue May 5 16:57:46 2009 +1000
70948
70949 Xi: access the prev_state to get group/modifier state for key events.
70950
70951 Key events may change the modifier state, so we need to get the prev_state for
70952 those (i.e. without the changes by the event already applied).
70953
70954 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70955
70956commit 35a4b8e7f4526a92d44cb16a783f21030cd1f6df
70957Author: Peter Hutterer <peter.hutterer@who-t.net>
70958Date: Tue May 5 16:51:59 2009 +1000
70959
70960 xkb: remove oldState from XkbHandleActions.
70961
70962 I really don't know what the purpose of this variable is or was, aside from
70963 potentially clobbering up our key state since there's a path where it may be
70964 used uninitialised.
70965
70966 Also, this means that xkbi->prev_state is now accessible from the DIX with
70967 meaningful data.
70968
70969 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70970
70971commit d523fbe428e4513d85402caa83e9349256b680cc
70972Author: Peter Hutterer <peter.hutterer@who-t.net>
70973Date: Mon May 4 17:00:57 2009 +1000
70974
70975 test: add test for xi2 struct sizes.
70976
70977commit 1b1b20d6e3e696e4437a9ef56128cde70a485f66
70978Author: Peter Hutterer <peter.hutterer@who-t.net>
70979Date: Tue May 5 13:47:55 2009 +1000
70980
70981 Change glib require for tests to auto.
70982
70983 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70984
70985commit 749cae82204f347248c815b1068ffcae9825fe70
70986Author: Peter Hutterer <peter.hutterer@who-t.net>
70987Date: Thu Apr 30 14:58:00 2009 +1000
70988
70989 mi: don't crash if we're trying to update the pointer for a keyboard.
70990
70991 If a device doesn't have a pointer, just return.
70992
70993 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
70994
70995commit 7ecedb0f2ee5b53513205d1f0aa381451893b9c5
70996Author: Peter Hutterer <peter.hutterer@who-t.net>
70997Date: Fri May 1 14:35:12 2009 +1000
70998
70999 include: up the number of max. input devices to 40.
71000
71001 With the Xtest virtual slave devices we have 4 devices for each MD
71002 pointer/keyboard pair, plus the AllDevices and AllMasterDevices reserved
71003 deviceids. It's quite easy to hit the current limit.
71004
71005 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71006
71007commit a38d33f6ac765bc5a3296b325e2c9b4a6a10ebf1
71008Author: Peter Hutterer <peter.hutterer@who-t.net>
71009Date: Fri May 1 15:24:02 2009 +1000
71010
71011 Xext: shut up compiler warnings in xtest.c
71012
71013 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71014
71015commit 14d5471981151b27678de826daa078977217477a
71016Author: Peter Hutterer <peter.hutterer@who-t.net>
71017Date: Fri May 1 15:22:37 2009 +1000
71018
71019 Xext: return BadDevice from XTest if we don't have keys/buttons/valuators.
71020
71021 BadDevice is an XI error, but this cannot happen for core XTest fake input
71022 anyway since the device will be the matching virtual XTest slave device.
71023
71024 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71025
71026commit f1bed0959e803784af142a0f45177ee3b675d78a
71027Author: Peter Hutterer <peter.hutterer@who-t.net>
71028Date: Fri May 1 15:15:04 2009 +1000
71029
71030 Xext: fix core Xtest button presses, don't call PickPointer.
71031
71032 We already did the device selection before, so dev should be the XTest virtual
71033 pointer.
71034
71035 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71036
71037commit 94648bb797d94b025746c60679c584e5be2fae28
71038Author: Jesse Barnes <jbarnes@virtuousgeek.org>
71039Date: Mon May 4 15:38:22 2009 -0700
71040
71041 Don't prepare outputs & crtcs if set_mode_major is present
71042
71043 A driver with this hook will take care of preparing the outputs & crtcs,
71044 so calling the prepare functions will just cause unnecessary flicker.
71045
71046 Fixes bug #21077
71047
71048commit 7d85169c7ab1d05c21fdbb2877ffd79f344dbb6b
71049Author: Michel Dänzer <daenzer@vmware.com>
71050Date: Mon May 4 10:14:43 2009 +0200
71051
71052 Fix typo in ProcRenderCreateAnimCursor.
71053
71054commit 59358de4b68a01452bf68593beb83605087ebc8f
71055Author: Peter Hutterer <peter.hutterer@who-t.net>
71056Date: Mon May 4 17:30:19 2009 +1000
71057
71058 input: ensure various ProcUngrabKey/Buttons have the right grabtype set.
71059
71060commit 834ea071b619506e0bef5bdbf0e3c59831cf6e99
71061Author: Peter Hutterer <peter.hutterer@who-t.net>
71062Date: Mon May 4 15:58:15 2009 +1000
71063
71064 dix: if grab types differ, two grabs cannot be the same.
71065
71066commit 63a3c727b35d1ffd8a850a5177bd6623cd78b179
71067Author: Peter Hutterer <peter.hutterer@who-t.net>
71068Date: Fri May 1 15:24:02 2009 +1000
71069
71070 Xext: shut up compiler warnings in xtest.c
71071
71072 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71073
71074commit 226dd90597ef29998ac5db15c5457aad3fe9729e
71075Author: Peter Hutterer <peter.hutterer@who-t.net>
71076Date: Fri May 1 15:22:37 2009 +1000
71077
71078 Xext: return BadDevice from XTest if we don't have keys/buttons/valuators.
71079
71080 BadDevice is an XI error, but this cannot happen for core XTest fake input
71081 anyway since the device will be the matching virtual XTest slave device.
71082
71083 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71084
71085commit 0fdff0a47a57a0096974ee83e07fefbad03075dd
71086Author: Peter Hutterer <peter.hutterer@who-t.net>
71087Date: Fri May 1 15:15:04 2009 +1000
71088
71089 Xext: fix core Xtest button presses, don't call PickPointer.
71090
71091 We already did the device selection before, so dev should be the XTest virtual
71092 pointer.
71093
71094 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71095
71096commit f0124ed93cdcee8d22a6690e639a3f60b9a63c54
71097Author: Peter Hutterer <peter.hutterer@who-t.net>
71098Date: Fri May 1 14:35:12 2009 +1000
71099
71100 include: up the number of max. input devices to 40.
71101
71102 With the Xtest virtual slave devices we have 4 devices for each MD
71103 pointer/keyboard pair, plus the AllDevices and AllMasterDevices reserved
71104 deviceids. It's quite easy to hit the current limit.
71105
71106 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71107
71108commit 08cd846000286eb5e07cc8d8cb313aff2d6e7faf
71109Author: Peter Hutterer <peter.hutterer@who-t.net>
71110Date: Fri May 1 11:06:06 2009 +1000
71111
71112 dix: only free the old cursor if the grab was successful.
71113
71114 Testcase: start thunderbird and move a message around, crashes after two or
71115 three moves.
71116
71117 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71118
71119commit 557dbadf3be273255e8fdb12d9321f4e88bf2b65
71120Author: Alan Coopersmith <alan.coopersmith@sun.com>
71121Date: Sat May 2 00:26:18 2009 -0700
71122
71123 XkbSetNamedIndicator should ignore SD's without LED's
71124
71125 When ProcXkbSetNamedIndicator is called on a core device, and we
71126 walk the slaves to set the LED's on each of them, ignore any slaves
71127 that do not have either a KbdFeedbackCtrl or LedCtrl structure.
71128
71129 (This is much more critical in xserver-1.5-branch, where we walk
71130 *all* devices, not just the slaves of the specified master, and
71131 thus return failure when setting an LED on the Core Keyboard and
71132 hit a xf86-input-mouse device with no LED's to set.)
71133
71134 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
71135 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
71136
71137commit 0d9d3f3e361f769822caedccf4c2a58cc9930ecc
71138Author: Ian Romanick <ian.d.romanick@intel.com>
71139Date: Mon Apr 27 15:11:10 2009 -0700
71140
71141 DRI2: Force allocation of real-front buffer for non-windows as well
71142
71143 For redirected rendering we end up with pixmaps (which the app thinks are
71144 windows) that are double buffered.
71145
71146 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
71147 Tested-by: Pierre Willenbrock <pierre@pirsoft.de>
71148
71149commit 3d30789a05a730a03faa6058c73a5eda36ef3779
71150Author: Peter Hutterer <peter.hutterer@who-t.net>
71151Date: Thu Apr 30 14:58:00 2009 +1000
71152
71153 mi: don't crash if we're trying to update the pointer for a keyboard.
71154
71155 If a device doesn't have a pointer, just return.
71156
71157 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71158
71159commit 876910a951053f0bd31e30098de3da15a1c1f5d6
71160Author: Peter Hutterer <peter.hutterer@who-t.net>
71161Date: Thu Apr 30 14:46:45 2009 +1000
71162
71163 xkb: write the _XKB_RF_RULES_PROP to each device.
71164
71165 We only have one root window and writing the rules used to the same property
71166 for each device is quite pointless if you don't have the same RMLVO on all
71167 devices. So let's be sensible and write the same property to the device too,
71168 so at least we know which device got loaded with which RMLVO.
71169
71170commit 44044adc92e6bb6537f48c84727523d5cb23528c
71171Author: Peter Hutterer <peter.hutterer@who-t.net>
71172Date: Thu Apr 30 11:39:50 2009 +1000
71173
71174 xfree86: print a message if NIDR fails due to AutoAddDevices off.
71175
71176 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71177
71178commit 43dbbc30327f826068bc7cd2cdaf870ed61b4522
71179Author: Peter Hutterer <peter.hutterer@who-t.net>
71180Date: Tue Apr 28 21:01:36 2009 +1000
71181
71182 Xi: fix typo in ProcXIGrabDevice
71183
71184 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71185
71186commit 6a618929a0c06ba0d6dac13d7e644cd9658d98ed
71187Author: Peter Hutterer <peter.hutterer@who-t.net>
71188Date: Tue Apr 28 16:49:45 2009 +1000
71189
71190 input: reshuffle CreateGrab and friends to take a GrabParameters param.
71191
71192 This is cleaning up work in preparation for XI2 passive grabs.
71193
71194commit e8e26f700c9c70d3f1bb53bdb71d1100f5c43a69
71195Author: Peter Hutterer <peter.hutterer@who-t.net>
71196Date: Sun Apr 26 21:43:25 2009 +1000
71197
71198 Xi: split some grab parameter checking out of GrabButton and GrabKey.
71199
71200commit 57aff88c7d0761e590806d07bee1c9410680c89f
71201Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
71202Date: Wed Apr 29 01:04:37 2009 -0400
71203
71204 Fix most remaining deprecated resource lookups.
71205
71206 Callsites updated to use dixLookupResourceBy{Type,Class}.
71207 TODO: Audit access modes to make sure they reflect the usage.
71208
71209commit 1abe0ee3da5e1268c7315f841d31337ea6524cf0
71210Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
71211Date: Tue Apr 28 23:51:40 2009 -0400
71212
71213 kdrive: fix Xvfb build with separate libmain.
71214
71215commit 737b49199a05299486064e6e762cf2a2f6f95be6
71216Author: Peter Hutterer <peter.hutterer@who-t.net>
71217Date: Thu Apr 9 15:40:24 2009 +1000
71218
71219 xfree86: restore default off for DontZap
71220
71221 Zapping is triggered by xkb these days, so note in the man page that it's the
71222 Terminate_Server action. Since it's XKB, personal preferences towards or
71223 against zapping should be achieved through xkb rulesets.
71224 If Terminate_Server is not in the xkb actions, then we can't zap anyway and we
71225 don't need a default of DontZap "on".
71226
71227 This patch restores the old meaning of DontZap - disallow zapping altogether,
71228 regardless of XKB's current keymap.
71229 Ideally, this patch should be accompanied by b0f64bdab00db652e in
71230 xkeyboard-config.
71231
71232 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71233
71234commit 98a1fa994158feca8bc80a10296a7675465c0e56
71235Author: Peter Hutterer <peter.hutterer@who-t.net>
71236Date: Tue Apr 21 22:18:23 2009 +1000
71237
71238 test: add InternalEvent to core event conversion tests.
71239
71240 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71241
71242commit 990e8362a76fadc173e890d9052a2e5274b6eac5
71243Author: Peter Hutterer <peter.hutterer@who-t.net>
71244Date: Thu Apr 16 11:41:58 2009 +1000
71245
71246 test: add a simple test to verify device axis intialization.
71247
71248 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71249
71250commit 4eac0df060b897470fb01cfd3b70f500da757c74
71251Author: Peter Hutterer <peter.hutterer@who-t.net>
71252Date: Wed Apr 15 17:21:08 2009 +1000
71253
71254 Add a test-suite for in-server unit-testing.
71255
71256 This patch adds a test/ directory that contains the setup for a unit-testing
71257 suite designed for in-server unit-testing. All functions available to the X
71258 server are available to the test binaries through static linking.
71259
71260 This test suite uses the glib testing framework.
71261 Do not use glib calls outside of the test/ directory.
71262
71263 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71264
71265commit c2c515ead38d9a6c9eae0b83aa7a82208f177b7e
71266Author: Peter Hutterer <peter.hutterer@who-t.net>
71267Date: Wed Apr 22 13:26:40 2009 +1000
71268
71269 dix: remove all but main() from main.c
71270
71271 All other functions are pushed into where they seemed to fit.
71272 main.c is now linked separately into libmain.a and linked in by the various
71273 DDXs.
71274
71275 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71276
71277commit dc2767d1c5db60385867c76ba2de507fe0cb8a90
71278Author: Peter Hutterer <peter.hutterer@who-t.net>
71279Date: Tue Apr 21 22:18:23 2009 +1000
71280
71281 test: add InternalEvent to core event conversion tests.
71282
71283 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71284
71285commit 1127ca097cb75450bcccfc5f5d82e435de2fb5b7
71286Author: Peter Hutterer <peter.hutterer@who-t.net>
71287Date: Thu Apr 16 11:41:58 2009 +1000
71288
71289 test: add a simple test to verify device axis intialization.
71290
71291 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71292
71293commit 4124c465a85713fe44843a16c5e2b13ece17e9d2
71294Author: Peter Hutterer <peter.hutterer@who-t.net>
71295Date: Wed Apr 15 17:21:08 2009 +1000
71296
71297 Add a test-suite for in-server unit-testing.
71298
71299 This patch adds a test/ directory that contains the setup for a unit-testing
71300 suite designed for in-server unit-testing. All functions available to the X
71301 server are available to the test binaries through static linking.
71302
71303 This test suite uses the glib testing framework.
71304 Do not use glib calls outside of the test/ directory.
71305
71306 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71307
71308commit 987579c930bda803427a28cb82773c389f5110d6
71309Author: Peter Hutterer <peter.hutterer@who-t.net>
71310Date: Wed Apr 22 13:26:40 2009 +1000
71311
71312 dix: remove all but main() from main.c
71313
71314 All other functions are pushed into where they seemed to fit.
71315 main.c is now linked separately into libmain.a and linked in by the various
71316 DDXs.
71317
71318 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71319
71320commit 2a2a7fa6856b2e1954e904577535118f20cd9133
71321Author: Alan Coopersmith <alan.coopersmith@sun.com>
71322Date: Mon Apr 27 20:23:53 2009 -0700
71323
71324 Solaris: use <sys/agpgart.h> instead of stale copy in Xorg sources
71325
71326 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
71327
71328commit 707af5f8c5fae3b6c4c5125a91917667bb0ab908
71329Author: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
71330Date: Thu Apr 23 16:15:01 2009 +0100
71331
71332 Make the cursor completely optional
71333
71334 For embedded use, it's convenient to be able to disable the cursor
71335 completely, without having to audit and fix up all your third-party
71336 code (e.g. Mozilla Firefox).
71337
71338 Add -nocursor and -cursor server options to enable and disable the
71339 cursor. The default is still -cursor, but embedded users can run the
71340 server with -nocursor to hide the cursor regardless of what
71341 application developers do.
71342
71343 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71344
71345commit 40a8f2f4088aa353de733a11327b628ef13a52e6
71346Author: Pierre Willenbrock <pierre@pirsoft.de>
71347Date: Sun Apr 19 21:15:22 2009 +0200
71348
71349 Fix obvious copypasta
71350
71351 Reviewed-by: Ian Romanick <idr@freedesktop.org>
71352
71353commit a8cf63fd18b0ddadbadf7b47d620159854920050
71354Author: Aaron Plattner <aplattner@nvidia.com>
71355Date: Thu Apr 23 14:57:04 2009 -0700
71356
71357 Export CompositeRegisterAlternateVisuals.
71358
71359 Drivers need this if they add visuals that require implicit redirection. Add a
71360 new SDK header, compositeext.h, to contain the prototype for this function.
71361
71362commit 48573e7ea13e279593249036e6d30f1d7507882f
71363Author: Peter Hutterer <peter.hutterer@who-t.net>
71364Date: Sun Apr 26 21:32:37 2009 +1000
71365
71366 Xi: swap stuff->cursor in SProcXIGrabDevice
71367
71368commit ff6c7764c2909e4126403b7211faa6c58556b341
71369Author: Ian Romanick <ian.d.romanick@intel.com>
71370Date: Fri Apr 24 12:49:19 2009 -0700
71371
71372 DRI2: Implement protocol for DRI2GetBuffersWithFormat
71373
71374 This change implements the protocol for DRI2GetBuffersWithFormat, but
71375 the bulk of the differences are the changes to the extension / driver
71376 interface to make this function work. The old CreateBuffers and
71377 DeleteBuffers routines are replaced with CreateBuffer and DeleteBuffer
71378 (both singular).
71379
71380 This allows drivers to allocate buffers for a drawable one at a time.
71381 As a result, 3D drivers can now allocate the (fake) front-buffer for a
71382 window only when it is needed. Since 3D drivers only ask for the
71383 front-buffer on demand, the real front-buffer is always created. This
71384 allows CopyRegion impelemenations of SwapBuffers to continue working.
71385 As with previous version of this code, if the client asks for the
71386 front-buffer for a window, we instead give it the fake front-buffer.
71387
71388 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
71389 Reviewed-by: Kristian Høgsberg <krh@redhat.com>
71390
71391commit 28ddfc88d8d547941c7f4713db527a3c2f9ec35a
71392Author: Ian Romanick <ian.d.romanick@intel.com>
71393Date: Fri Apr 24 12:09:21 2009 -0700
71394
71395 DRI2: Add interface for drivers to query DRI2 extension version
71396
71397 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
71398
71399commit a1c98f7fa9aa0cd5075ef2054c5d2be4d1e740f0
71400Author: Joel Bosveld <Joel.Bosveld@gmail.com>
71401Date: Fri Apr 24 10:12:26 2009 -0700
71402
71403 Correct fcntl(F_SETOWN) error check in non-Solaris sigio code
71404
71405 commit 64b7f96dca accidentally inverted the comparison, could
71406 result in crashes on some systems later on.
71407
71408 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
71409
71410commit 64b7f96dca23d2b33e0b8a8785d628d9a2f50e7b
71411Author: Alan Coopersmith <alan.coopersmith@sun.com>
71412Date: Fri Apr 24 00:14:12 2009 -0700
71413
71414 Add SIGIO/SIGPOLL support for Solaris
71415
71416 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
71417
71418commit 0c467b1f990f156a34d5ef117bd825df6b633114
71419Author: Alan Coopersmith <alan.coopersmith@sun.com>
71420Date: Fri Apr 24 00:05:28 2009 -0700
71421
71422 Add casts to generated entries in sdksyms.c to silence type mismatch warnings
71423
71424 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
71425
71426commit 7d0f7518c2235a9dc783029971259ddaada2db20
71427Author: Alan Coopersmith <alan.coopersmith@sun.com>
71428Date: Thu Apr 23 08:42:38 2009 -0700
71429
71430 Fix byte swapping of XF86VidMode{Get,Set}GammaRamp
71431
71432 Fixes OpenSolaris Bug 8315:
71433 Xorg segfaults when screensaver fades in cross-endian xdmcp session
71434 <http://defect.opensolaris.org/bz/show_bug.cgi?id=8315>
71435
71436 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
71437
71438commit 057fc9a4f89282e440b5e11458f1dd8655879dd8
71439Merge: 95628b7 932d6bc
71440Author: Peter Hutterer <peter.hutterer@who-t.net>
71441Date: Fri Apr 24 16:15:47 2009 +1000
71442
71443 Merge branch 'master' into xi2
71444
71445 Conflicts:
71446 Xi/chdevhier.c
71447 include/input.h
71448
71449commit 932d6bcbb68194c5bdfeb336f700dc8b31529223
71450Author: Peter Hutterer <peter.hutterer@who-t.net>
71451Date: Fri Apr 24 15:28:45 2009 +1000
71452
71453 mi: remove superfluous check.
71454
71455 mieqProcessInputEvents doesn't process events from MDs anymore, so we don't
71456 need to check for pDev->isMaster.
71457
71458 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71459
71460commit fab563bf8f6b63906ce9d5a568c467425843265b
71461Author: Benjamin Close <Benjamin.Close@clearchain.com>
71462Date: Fri Mar 27 16:44:15 2009 +1030
71463
71464 input: propagate XTst events through virtual slave devices.
71465
71466 A XTest virtual slave device pair (kbd/ptr) exists for every master
71467 device pair. This is so XTest events are correctly propogated via slave
71468 devices up to Master devices and the classes are correctly changed along
71469 the way. We add the XTest slave device pair to the Virtual Core pointer
71470 and provide a simple way of creating the devices.
71471
71472 A XTest Slave Device is identified by the XTstDevicePrivateKey property
71473 being set in the devices devProperties
71474
71475 XI events are still propagated through the matching device, in the hope the
71476 client knows what it is doing.
71477
71478 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71479
71480commit ccd28ca2ad40aecf951ab058d89062828e3e5d8a
71481Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
71482Date: Wed Apr 22 19:23:17 2009 -0700
71483
71484 XQuartz: Localization update
71485 (cherry picked from commit 9435fc5e20e25ce66ec85bde033daff51f39f69c)
71486
71487commit a362d0f4fb79150f7b8e0eb3d1dc1cdee2f98d95
71488Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
71489Date: Sat Apr 11 19:06:45 2009 -0700
71490
71491 XQuartz: Make sure the bgMouseLocationUpdated state is consistent
71492 (cherry picked from commit 8fa8a8e1db60c8ac60d10f1f15517471d2291f77)
71493
71494commit 8dc8812c226222fc64f8e489f9f123687a95d3d1
71495Author: Dave Airlie <airlied@redhat.com>
71496Date: Thu Apr 23 12:13:35 2009 +1000
71497
71498 exa: add missing exa.h header include
71499
71500commit 3ec6a121e110ba3cae964ee201d4d5bf818b6dff
71501Author: Dave Airlie <airlied@redhat.com>
71502Date: Thu Apr 23 12:13:05 2009 +1000
71503
71504 exa: missed exa.h change
71505
71506commit 02ae85c4c93bbd0013c063ce0014b365a8ba061f
71507Author: Dave Airlie <airlied@redhat.com>
71508Date: Thu Apr 23 12:04:17 2009 +1000
71509
71510 exa: add CreatePixmap2 hook for driver pixmaps.
71511
71512 This adds a revised pixmap hook for driver pixmaps, which is
71513 required to support tiling on various chips.
71514
71515 Signed-off-by: Dave Airlie <airlied@redhat.com>
71516
71517commit 9d684ba0bcab53841a17ebd521507b6df8bc4b45
71518Author: Dave Airlie <airlied@redhat.com>
71519Date: Thu Apr 16 10:28:36 2009 +1000
71520
71521 exa: avoid offscreen pixmap swapped out flag for driver pixmaps
71522
71523commit 0dfb97f15f591f85e079f5829c77d0c328d00464
71524Author: Julien Cristau <jcristau@debian.org>
71525Date: Wed Apr 22 20:01:49 2009 +0200
71526
71527 Bug#21324: Add quirk for Iiyama Vision Master 450
71528
71529 Reported-by: Jeremy Henty <onepoint@starurchin.org>
71530 Signed-off-by: Julien Cristau <jcristau@debian.org>
71531
71532commit 44227ef1b77467c76147b9bf79bdd0e6305a522a
71533Author: Ian Romanick <ian.d.romanick@intel.com>
71534Date: Mon Apr 20 18:20:52 2009 -0700
71535
71536 DRI2: Send the version the code actually supports
71537
71538 This prevents building an older server with a new dri2proto.h from
71539 resulting in a DRI2 extension module that lies about the version it
71540 supports.
71541
71542 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
71543
71544commit d1e916d29be8b470cbc8cadcf6e83991fdbc5a9f
71545Author: Ian Romanick <ian.d.romanick@intel.com>
71546Date: Thu Apr 16 12:10:34 2009 -0700
71547
71548 DRI2: Add missing front-buffer flush callback.
71549
71550 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
71551
71552commit 826a5bff0136b2b4d55a9e6e6bc3a7a64da9031e
71553Author: Benjamin Close <benjamin.close@clearchain.com>
71554Date: Thu Apr 16 15:33:30 2009 +1000
71555
71556 dix: Change AllocMaster into AllocDevicePair, allow creation of SDs too.
71557
71558 Allocating a slave device is essentially the same as allocating a master device.
71559 Hence we rename AllocMaster to AllocDevicePair and provided the ability to
71560 indicate if a master or slave device pair is required.
71561
71562 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71563
71564commit e2e5932bda3f473629d4be6f3ca4dcab18993eb6
71565Author: Peter Hutterer <peter.hutterer@who-t.net>
71566Date: Thu Apr 16 11:06:52 2009 +1000
71567
71568 dix: don't allow more than MAX_VALUATORS on one device.
71569
71570 Some keyboards (?) advertise more than MAX_VALUATORS axes. Parts of the
71571 internal event delivery relies on not having more than MAX_VALUATOR axes, so
71572 let's cap it down.
71573 If there's real devices that require more than the current 36, I'm sure we can
71574 bump this up.
71575
71576 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71577
71578commit 063833f3a6d9f8f657e3de309c8d6d5c3d606513
71579Author: Julien Cristau <jcristau@debian.org>
71580Date: Wed Dec 24 14:23:37 2008 +0100
71581
71582 Add XI 1.5 event and requests to protocol.txt
71583
71584commit 4f86ee61a4abf7a29e565d095aa08abd0ca9dc66
71585Author: Julien Cristau <jcristau@debian.org>
71586Date: Wed Dec 24 14:17:27 2008 +0100
71587
71588 Add RandR 1.3 requests to protocol.txt
71589
71590commit 6559f02ef89cb0ee365110d6a928ebd29632ad53
71591Author: Colin Harrison <colin.harrison@virgin.net>
71592Date: Fri Apr 17 15:14:45 2009 +0100
71593
71594 xkb: set bell_func in InitKeyboardDeviceStruct.
71595
71596 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71597
71598commit 95628b797e2b326ae15406df4f736e25f5d2f12f
71599Author: Peter Hutterer <peter.hutterer@who-t.net>
71600Date: Mon Apr 20 14:04:54 2009 +1000
71601
71602 Xi: use the XI2 defines for AsyncPair, SyncDevice, etc, not the XI ones.
71603
71604 And don't allow some values on SDs either.
71605
71606commit 335c63fcd6a53c1d3dffd6e00c0acd400c61e3d3
71607Author: Werner LEMBERG <wl@gnu.org>
71608Date: Fri Apr 17 15:22:57 2009 +0200
71609
71610 Add newline to some LogMessage strings.
71611
71612 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71613
71614commit d5ad14c8ed4d8360e1df8cd0bacf6a7c9c31df91
71615Merge: 54716fd 0e0642e
71616Author: Peter Hutterer <peter.hutterer@who-t.net>
71617Date: Sun Apr 19 22:28:22 2009 +1000
71618
71619 Merge branch 'master' into xi2
71620
71621commit 54716fd3dbc251db9d251d1d0435942efaa63259
71622Author: Peter Hutterer <peter.hutterer@who-t.net>
71623Date: Sun Apr 19 22:12:11 2009 +1000
71624
71625 Convert to using int32_t fixed point values on the wire.
71626
71627 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71628
71629commit 129ac9a9145323e3f126590b491e718f976f80ce
71630Author: Peter Hutterer <peter.hutterer@who-t.net>
71631Date: Sun Apr 19 21:31:56 2009 +1000
71632
71633 dix: set root_x/y for device events.
71634
71635 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71636
71637commit eeb1e4cd254c047539933c3b886037e67ef30762
71638Author: Peter Hutterer <peter.hutterer@who-t.net>
71639Date: Wed Apr 15 22:17:15 2009 +1000
71640
71641 Xi: add support for XIAllowEvents.
71642
71643 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71644
71645commit 74d0fc3aee05bc4f505274d81dd44c0d2b4368ff
71646Author: Peter Hutterer <peter.hutterer@who-t.net>
71647Date: Thu Apr 16 11:06:52 2009 +1000
71648
71649 dix: don't allow more than MAX_VALUATORS on one device.
71650
71651 Some keyboards (?) advertise more than MAX_VALUATORS axes. Parts of the
71652 internal event delivery relies on not having more than MAX_VALUATOR axes, so
71653 let's cap it down.
71654 If there's real devices that require more than the current 36, I'm sure we can
71655 bump this up.
71656
71657 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71658
71659commit edb70caf21fa5da3c336c47c62218a3f05c635b8
71660Author: Peter Hutterer <peter.hutterer@who-t.net>
71661Date: Wed Apr 15 22:16:49 2009 +1000
71662
71663 dix: remove un-used parameter "core" from AllowSome
71664
71665 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71666
71667commit bb5418d4901017c657031181d3839f58b387a2a3
71668Author: Peter Hutterer <peter.hutterer@who-t.net>
71669Date: Sun Apr 12 22:22:21 2009 +1000
71670
71671 Xi: Add support for XI2 active grabs and ungrabs.
71672
71673 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71674
71675commit c11ef87931f920ba782ba4e9b47d9c31ad7c1cf7
71676Author: Peter Hutterer <peter.hutterer@who-t.net>
71677Date: Sun Apr 12 22:16:14 2009 +1000
71678
71679 Xi: take XI2 requests into account when checking opcodes.
71680
71681 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71682
71683commit c94ea5bc055e4efc323e84b7a8266e8b8a4af48e
71684Author: Peter Hutterer <peter.hutterer@who-t.net>
71685Date: Sun Apr 12 17:38:28 2009 +1000
71686
71687 input: use a GrabMask union in GrabDevice to allow for XI2 masks.
71688
71689 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71690
71691commit 6bb4b5b93701535402f65ea828348ed7747c7dbf
71692Author: Peter Hutterer <peter.hutterer@who-t.net>
71693Date: Sun Apr 12 16:20:10 2009 +1000
71694
71695 Xi: add XI2 grab protocol request handling.
71696
71697 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71698
71699commit 09f9a86077f0058ce88ee9b3df5d1ab854eeca43
71700Author: Peter Hutterer <peter.hutterer@who-t.net>
71701Date: Sun Apr 12 16:19:45 2009 +1000
71702
71703 input: replace GrabRec's coreGrab field with grabtype.
71704
71705 Don't allow grabs of different types to override each other.
71706
71707 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71708
71709commit 7fbe1b7d63c94b344e8ff9d09c0bb162ca104909
71710Author: Peter Hutterer <peter.hutterer@who-t.net>
71711Date: Sat Apr 11 13:10:31 2009 +1000
71712
71713 Xi: remove the GetExtensionVersion hack.
71714
71715 Squashing data into a request that's defined to be padding is probably not the
71716 best idea.
71717
71718 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71719
71720commit 0e0642ee9466d3268476d0084a83a9d93a4aa555
71721Author: Peter Hutterer <peter.hutterer@who-t.net>
71722Date: Thu Apr 16 16:17:07 2009 +1000
71723
71724 os: don't malloc memory in LogVMessageVerb.
71725
71726 LogVWrite is limited to a buffer size of 1024, so we don't loose anything here
71727 by truncating. This way we can use LogVMessageVerb (and xf86Msg and friends)
71728 during signal handlers with the normal message types.
71729
71730 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71731 Acked-by: Alan Coopersmith <alan.coopersmith@sun.com>
71732
71733commit 62d2fb68638e9f2aa3c1d72027619c4d38f5b812
71734Author: Peter Hutterer <peter.hutterer@who-t.net>
71735Date: Thu Apr 16 17:06:33 2009 +1000
71736
71737 xkb: Add XkbFreeRMLVOSet helper function.
71738
71739 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71740 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
71741
71742commit 73e1055b1275470dcc2d53b15898c866d9721d3e
71743Author: Peter Hutterer <peter.hutterer@who-t.net>
71744Date: Sat Apr 11 11:19:29 2009 +1000
71745
71746 Xi: fix a typo in a #ifdef
71747
71748 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71749
71750commit cbbc7fed994448852961a987a8b0eadf718b24bf
71751Author: Simon Thum <simon.thum@gmx.de>
71752Date: Wed Apr 1 10:53:10 2009 +0200
71753
71754 dix: correctly utilize tracker buffer and protect from timer overruns
71755
71756 two small related fixes hard to split up
71757
71758 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71759
71760commit ca4918ac2ff003142aefbd8ac7ec2bc398ccc6eb
71761Author: Peter Hutterer <peter.hutterer@who-t.net>
71762Date: Thu Apr 9 19:31:04 2009 +1000
71763
71764 Xi: remove fail from SProcXISelectEvent.
71765
71766 Leftover from rebasing once too often, I guess.
71767
71768 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71769
71770commit 463676f016a105a790cbb386d20eb5b188c04652
71771Author: Peter Hutterer <peter.hutterer@who-t.net>
71772Date: Thu Apr 9 19:11:59 2009 +1000
71773
71774 Xi: remove un-used variable pXIClient
71775
71776 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71777
71778commit 32b13ec4cd42842a133aa6428f49638b0448efc9
71779Author: Peter Hutterer <peter.hutterer@who-t.net>
71780Date: Thu Apr 9 19:11:27 2009 +1000
71781
71782 Xi: silence compiler warnings about "wrong" event types.
71783
71784 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71785
71786commit f2e957c5342aefc8329e9fe282b7072f767fcd28
71787Author: Peter Hutterer <peter.hutterer@who-t.net>
71788Date: Thu Apr 9 19:07:36 2009 +1000
71789
71790 mi: add prototype for CopyGetMasterEvent.
71791
71792 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71793
71794commit d80798ba2ae018086aaf230fcd8bc71d3bf06b11
71795Author: Peter Hutterer <peter.hutterer@who-t.net>
71796Date: Thu Apr 9 20:27:00 2009 +1000
71797
71798 mi: fix wrong (*EnqueueEvent) declaration in miPointerScreenFuncRec.
71799
71800 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71801
71802commit 4dedb900f4fbd3bdc9538b1e82cab2044438a770
71803Author: Peter Hutterer <peter.hutterer@who-t.net>
71804Date: Thu Apr 9 08:42:06 2009 +1000
71805
71806 xfree86: shut up compiler warnings - typecast to InternalEvent
71807
71808 Reported-by: Eric Anholt <eric@anholt.net>
71809 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71810
71811commit 4fa7fcf610d2ab91c1158f8034bbc0531060a86f
71812Author: Peter Hutterer <peter.hutterer@who-t.net>
71813Date: Thu Apr 9 08:40:20 2009 +1000
71814
71815 mi: fix compiler warning - explicitly typecast to InternalEvent.
71816
71817 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71818
71819commit 8bf2f562678107c346e8664f18c10f315c2768f5
71820Author: Peter Hutterer <peter.hutterer@who-t.net>
71821Date: Thu Apr 9 08:14:39 2009 +1000
71822
71823 dix: fix dev/keybd variable mixup.
71824
71825 Reported-by: Eric Anhold <eric@anholt.net>
71826 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71827
71828commit 012cb1956a69f6cf2a878c5331e792fb12c73e9d
71829Author: Peter Hutterer <peter.hutterer@who-t.net>
71830Date: Thu Apr 9 08:06:32 2009 +1000
71831
71832 xfree86: fix use of uninitialized variable in DGAProcessPointerEvent.
71833
71834 Reported-by: Eric Anholt <eric@anholt.net>
71835 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71836
71837commit 73b01a9aac5fde46ccd90c08a070e9d2496092b9
71838Author: Simon Thum <simon.thum@gmx.de>
71839Date: Wed Apr 8 14:35:01 2009 +0200
71840
71841 dix: fix pointer accelerations remainder handling
71842
71843 This didn't really work as intended, but did amazingly well thanks
71844 to roundf() hiding the defect. Cheers!
71845
71846 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71847
71848commit 4318075140cc287871d3c3b9f777289ea4ffa23a
71849Author: Peter Hutterer <peter.hutterer@who-t.net>
71850Date: Tue Apr 7 21:51:02 2009 +1000
71851
71852 dix: store subpixel precision and send it down the wire to the client.
71853
71854 For the valuator data, not yet for root x/y and event x/y.
71855
71856 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71857
71858commit 9212948461c5e053abb61fff73ced4a00f138544
71859Author: Simon Thum <simon.thum@gmx.de>
71860Date: Sat Mar 21 18:19:19 2009 +0100
71861
71862 dix: allow relative motion buffer to accumulate in a natural way
71863
71864 Since with XI2 we pass sub-pixel motion, anything else is broken.
71865
71866 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71867
71868commit ad76656f8869e2065f0c4e66cfbeef0b42c61769
71869Author: Adam Jackson <ajax@redhat.com>
71870Date: Fri Apr 17 17:46:58 2009 -0400
71871
71872 randr: Accept gamma set requests from XF86VidMode clients too
71873
71874commit d9bf52b4abd29a3c206cd1e765b680659ddac1c6
71875Author: David Jander <david.jander@protonic.nl>
71876Date: Fri Apr 17 01:34:18 2009 -0400
71877
71878 [kdrive] Fix rotation of pointer
71879
71880 Rotation matrix for pointer coordinates was incomplete and pointers with
71881 absolute coordinates did not work correctly in xserver (kdrive) when the
71882 sceen was rotated other than by 0 degrees.
71883
71884 Signed-off-by: David Jander <david.jander@protonic.nl>
71885 Signed-off-by: James Cloos <cloos@jhcloos.com>
71886
71887commit 4559d2ace6ac55fe361f572ded0769cdd1f3b545
71888Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
71889Date: Thu Apr 16 22:48:11 2009 -0400
71890
71891 security: Grant untrusted windows remove access on all windows.
71892
71893 This allows untrusted clients to destroy their own windows when they
71894 have been reparented by a trusted window manager.
71895
71896commit 3481b32ab971c41cb972f6819ae049f3e9f7033b
71897Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
71898Date: Thu Apr 16 22:39:54 2009 -0400
71899
71900 security: Fix a crash caused by wrong ordering of format arguments.
71901
71902commit 6045506be0cebca4ebbe943ae77f020aafa703d4
71903Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
71904Date: Thu Apr 16 22:33:12 2009 -0400
71905
71906 security: Revert behavior of extension access for compatibility.
71907
71908 Previously, three extensions were defined as "trusted" by the extension:
71909 BIG-REQUESTS, XC-MISC, and XPrint. No other extensions were permitted
71910 to be used by untrusted clients.
71911
71912 In commit 8b5d21cc1d1f4e9d20e5d5eca44cb1e60a419763 this was changed for
71913 some reason. Return to the old, compatible behavior.
71914
71915commit 56a5955c8cd87137248edb2cbc65d384376d72ad
71916Author: Peter Hutterer <peter.hutterer@who-t.net>
71917Date: Tue Apr 14 17:05:04 2009 +1000
71918
71919 xkb: strdup the values returned by XkbGetRulesDflts
71920
71921 XkbGetRulesDftls may get a copy of what will later be freed when passed into
71922 XkbSetRulesDftls.
71923
71924 On the second run of XkbGet/SetRulesDflts:
71925 XkbGetRulesDflts(rmlvo)
71926 rmlvo->rules = current-rules
71927
71928 XkbSetRulesDflts(rmlvo)
71929 free(current-rules)
71930 current-rules = strdup(rmlvo->rules)
71931
71932 Leaving us with garbage in current-rules.
71933
71934 This patch requires callers of XkbGetRulesDflts to free the associated memory.
71935
71936 See also
71937 http://lists.freedesktop.org/archives/xorg-devel/2009-February/000305.html
71938
71939 Reported-by: Benjamin Close <Benjamin.Close@clearchain.com>
71940 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71941 Acked-by: Benjamin Close <Benjamin.Close@clearchain.com>
71942 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
71943
71944commit b406886bbffadaa52864a99f2a0520999eadc15d
71945Author: Peter Hutterer <peter.hutterer@who-t.net>
71946Date: Tue Apr 14 16:57:29 2009 +1000
71947
71948 input: allow NULL as XkbRMVLOSet in InitKeyboardDeviceStruct.
71949
71950 Virtually all callers use
71951 XkbGetRulesDefault(&rmlvo);
71952 InitKeyboardDeviceStruct(..., rmlvo);
71953
71954 Let's save them the trouble and accept NULL as a hint to take the
71955 default RMLVO.
71956
71957 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71958 Acked-by: Benjamin Close <Benjamin.Close@clearchain.com>
71959 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
71960
71961commit 4e4e263bc073bf452f19c932b937c4881ae71f64
71962Author: Peter Hutterer <peter.hutterer@who-t.net>
71963Date: Wed Apr 15 22:16:49 2009 +1000
71964
71965 dix: remove un-used parameter "core" from AllowSome
71966
71967 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71968
71969commit dd6e8a14ec1c8f4ed9c51ca2764261e6e48d13b3
71970Author: Peter Hutterer <peter.hutterer@who-t.net>
71971Date: Sat Apr 11 11:19:29 2009 +1000
71972
71973 Xi: fix a typo in a #ifdef
71974
71975 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
71976
71977commit de1e43181bd670877b994db221ad8a04b5d63324
71978Author: Ian Romanick <ian.d.romanick@intel.com>
71979Date: Wed Apr 15 11:13:48 2009 -0700
71980
71981 DRI2: Don't leave empty entries in private->buffers
71982
71983 This should fix bug #21130.
71984
71985 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
71986
71987commit b3e3154cce47add97f5561088036ce3b9e7dc937
71988Author: Robert Noland <rnoland@2hip.net>
71989Date: Wed Apr 15 12:06:19 2009 -0500
71990
71991 One = is more than adequate here. Make is sh safe.
71992
71993commit 74d27c8b5bac7c8d2ed02ba86e09bf09924ce05c
71994Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
71995Date: Wed Apr 15 15:44:17 2009 +0200
71996
71997 Fix build on hurd-i386
71998
71999 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=523970
72000
72001commit faf7dfa099f5b42a703313fbd1bf8afdad07a179
72002Author: Dave Airlie <airlied@linux.ie>
72003Date: Wed Apr 15 18:26:06 2009 +1000
72004
72005 randr12: looking up these bits if randr isn't initialised is bad.
72006
72007 When xinerama is enabled we don't get randr protocol, but the
72008 driver might still want randr internals
72009
72010commit efa31092d6703397121a0ada4f7205a8ecad3d3d
72011Author: Ander Conselvan de Oliveira <ander@mandriva.com.br>
72012Date: Mon Apr 6 16:01:20 2009 -0300
72013
72014 xfree86: Remove device from inputInfo.devices if ActivateDevice failed.
72015
72016 After the call to xf86ActivateDevice, the new device will be added to
72017 inputInfo.devices. However, if the subsequent call to ActivateDevice
72018 fails, the correponding InputInfoRec for the device is deleted but an
72019 entry still remains in inputInfo.devices. This might lead to a server
72020 crash later on (on InitAndStartDevices for instance) when the device
72021 control proc would be called for an invalid device.
72022
72023 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72024
72025commit d79bad0aa70403ead8ec87bac8463a6e2005802c
72026Author: Peter Hutterer <peter.hutterer@who-t.net>
72027Date: Mon Apr 13 17:49:00 2009 +1000
72028
72029 xfree86: don't synthesise a mouse section if synaptics devices are found.
72030
72031 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72032
72033commit 4a27618565f3151ab17f0ca9ecbde12fa7ba13c0
72034Author: Kristian Høgsberg <krh@redhat.com>
72035Date: Tue Apr 14 16:21:04 2009 -0400
72036
72037 composite: Fix resource lookups
72038
72039commit e7785e8af3e34f9d1089d8499d16802984ab9823
72040Author: Kristian Høgsberg <krh@redhat.com>
72041Date: Tue Apr 14 14:24:31 2009 -0400
72042
72043 xfixes: Fix a couple of resource lookups
72044
72045commit 0eb19f9437b7d8c19592e49eedb028771d300d80
72046Author: Adam Jackson <ajax@redhat.com>
72047Date: Tue Apr 14 10:54:25 2009 -0400
72048
72049 xdmcp: Don't crash on X -query with more than 255 IP addresses. (#20675)
72050
72051 You could be more clever than this, but the wire protocol says this
72052 really is an array of not more than 255 ARRAY8, so it's not just a
72053 matter of changing the types.
72054
72055commit f0543ae4ec0fcb5d696e7b2983653bd779f1eddc
72056Author: Eric Anholt <eric@anholt.net>
72057Date: Wed Apr 8 14:17:41 2009 -0700
72058
72059 Fix unused var warning from pci cleanups.
72060
72061commit 181cc08c8908a119fc403f970dea8cc98d3e0b9b
72062Author: Eric Anholt <eric@anholt.net>
72063Date: Wed Apr 8 14:17:40 2009 -0700
72064
72065 Add shave so that we can see the steaming piles of warnings generated.
72066
72067 The old style output can be reenabled for build system debugging using
72068 "make V=1", or --disable-shave at configure time.
72069
72070commit 4474c200a102feda72f9572a96cb588009aa0147
72071Author: Eric Anholt <eric@anholt.net>
72072Date: Wed Apr 8 14:17:39 2009 -0700
72073
72074 Move VENDOR_* defines from AC_SUBST to a header to avoid angering shave.
72075
72076 This is more sane anyway, as it ensures a rebuild when changing them.
72077
72078commit e72c85547b405fbd0117dc1236e5ca5a2126063c
72079Author: Eric Anholt <eric@anholt.net>
72080Date: Wed Apr 8 14:17:38 2009 -0700
72081
72082 Move contributed m4 to a subdir so we can more easily update contributions.
72083
72084commit bef1cfb2395df47458159a0b7ae27b9db15025ef
72085Author: Peter Hutterer <peter.hutterer@who-t.net>
72086Date: Tue Apr 14 16:54:42 2009 +1000
72087
72088 xnest: remove unused variable 'names'.
72089
72090 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72091
72092commit 7b6400a1b8d2f228fcbedf17c30a7e3924e4dd2a
72093Author: Kristian Høgsberg <krh@redhat.com>
72094Date: Thu Apr 9 13:16:37 2009 -0400
72095
72096 glx: Fix drawable private leak on destroy
72097
72098 When a drawable goes away, we don't destroy the GLX drawable in full,
72099 since it may be current for a context. This means that when the drawable
72100 is destroyed in full later, the backend doesn't get a chance to
72101 destroy resources associated with the drawable (the DRI2Drawable).
72102
72103 With this patch, we destroy the GLX drawable in full when it goes away
72104 and then track down all contexts that reference it and NULL their
72105 pointers.
72106
72107commit 140463a197fb93d0a4bfad924efc35b860e8cc54
72108Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72109Date: Sat Apr 11 16:53:59 2009 -0700
72110
72111 XQuartz: Use correct values for ProximityIn and ProximityOut
72112
72113 This was the other underlying cause of teh 100% CPU tablet issue.
72114 (cherry picked from commit a9cecf34c23fbcd59b56b380c51d31a9fabc3eb7)
72115
72116commit 80a7bb2605f9b439d6221c0495a629a39177a018
72117Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72118Date: Sat Apr 11 17:26:07 2009 -0700
72119
72120 XQuartz: Solve the tablet 100% CPU bug
72121
72122 This happened because we put a byte in the fd to wake up dispatch, but we never actually enqueued anything in mieq because the num_events was 0.
72123 (cherry picked from commit c21ca7558d2faf93c61f5feaafd7c878e9e21942)
72124
72125commit 6f8f7c78f1b722bc70a0ea8f6340116a1e09e858
72126Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72127Date: Sat Apr 11 16:13:42 2009 -0700
72128
72129 XQuartz: Re-enable background window checking
72130
72131 since that code was not the culprit for the wacom tablet, background 100% CPU bug
72132 (cherry picked from commit fc1dc5d71b2a488a8a94d953dd8e67353161a590)
72133
72134commit 8522a759c9e78478bb399b91d3a0af2d23ea6766
72135Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72136Date: Sat Apr 11 16:12:23 2009 -0700
72137
72138 XQuartz: Only set MotionNotify on activation if it is updated.
72139 (cherry picked from commit ae8077a251ef27381a755d57ff974767bda16148)
72140
72141commit 5e55becddccc376ab7338789562ad6d6cd750de9
72142Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72143Date: Sat Apr 11 14:23:41 2009 -0700
72144
72145 XQuartz: Make sure the Fn doesn't trigger unneccessary calls to DarwinUpdateModKeys()
72146 (cherry picked from commit 70a18558c6b0a02b633fd8974f002cdf3cdc713e)
72147
72148commit bd1d9179094657865b0606ed0ac835a8b8df3be1
72149Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72150Date: Sat Apr 11 13:53:38 2009 -0700
72151
72152 XQuartz: Re-enable Fn as an option for 3button mouse simulation.
72153
72154 Patch from Martin Otte <otte@duke.edu>
72155 (cherry picked from commit b5ec3be6b5449b5d575bc1472fdd1c9cb15cb8be)
72156
72157commit 1b5758bef0840c6614244e321790231b3c9477c9
72158Author: Maarten Maathuis <madman2003@gmail.com>
72159Date: Thu Apr 9 15:45:57 2009 +0200
72160
72161 exa: implement UTS based upload through CopyArea
72162
72163 - Some image viewers (eog, gqview) trigger the CopyArea path of Xext/shm.c
72164 - I'm not aware of any code path that wouldn't like UTS and trigger this code.
72165 - miDoCopy should handle src coordinate clipping.
72166 - Overlapping blits are obviously not an issue (both would have to be offscreen or not).
72167
72168commit 567cf67959b30432ae30f4851ec17b3a375ab838
72169Author: Ian Romanick <ian.d.romanick@intel.com>
72170Date: Thu Apr 9 14:38:24 2009 -0700
72171
72172 DRI2: Synchronize the contents of the real and fake front-buffers
72173
72174 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
72175
72176commit f1a995d1496d73741731e32f475097c44a8da972
72177Author: Ian Romanick <ian.d.romanick@intel.com>
72178Date: Thu Apr 9 14:31:01 2009 -0700
72179
72180 DRI2: Do not send the real front buffer of a window to the client
72181
72182 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
72183
72184commit aa2928325fe51d94a636dde9c090e8f54a311a12
72185Author: Ian Romanick <ian.d.romanick@intel.com>
72186Date: Wed Apr 8 15:44:34 2009 -0700
72187
72188 DRI2: Add fake front-buffer to request list for windows
72189
72190 If a front-buffer is requested for a window, add the fake front-buffer
72191 to the list of requested buffers.
72192
72193 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
72194
72195commit 808fd2c67f303cb721769375b11ce8b90ffc1909
72196Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72197Date: Thu Apr 9 20:00:37 2009 -0700
72198
72199 XQuartz: xprSetWindowLevel updated to store the level requested by the WM
72200 (cherry picked from commit c28c2ddc3a8f3c5b9beec396953bb3ac9ee4714b)
72201
72202commit dfb0d7aefbbdfc0db966e3a84d52f638135d9138
72203Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72204Date: Thu Apr 9 18:51:22 2009 -0700
72205
72206 XQuartz: Update window levels when changing rootless state
72207 (cherry picked from commit 1359ded5bfc14a80fb998b01a54ecacb96c4ff88)
72208
72209commit d5ef88d7543ed787093099ab18db766c446b47d1
72210Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72211Date: Thu Apr 9 18:27:22 2009 -0700
72212
72213 XQuartz: Fix window levels for rooted mode to allow showing the menu bar.
72214 (cherry picked from commit 80759a4186bf0335edc85aecea2faf11fe09f491)
72215
72216commit 15146b863759640e7a73fab2301fb28ef4dead84
72217Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72218Date: Thu Apr 9 17:47:41 2009 -0700
72219
72220 XQuartz: Properly set the window level for the root window
72221 (cherry picked from commit bdf9286d1cbfeaaf8eaf03d28091e91ee587ee25)
72222
72223commit 8d5dcfe2154f217bd8fde5509d78f3383add8725
72224Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72225Date: Thu Apr 9 04:36:26 2009 -0700
72226
72227 XQuartz: Properly set the menu bar and hotkey state when changing rootless mode.
72228
72229 Currently no code path exhibits the broken behavior since we only toggle into rootless if we don't have the root.
72230 (cherry picked from commit 970f100ca3c5fc0662ae7658d49d118fbd9de943)
72231
72232commit 82d7cf5cdcbc5e451a87f8c2f64cd5d2e3627d54
72233Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72234Date: Thu Apr 9 03:55:13 2009 -0700
72235
72236 XQuartz: In rooted mode, make sure we start in the hidden state.
72237 (cherry picked from commit 5ecc497f71c2133f773f6c56ad76cb778862ddd6)
72238
72239commit e86f4e93020d56385418850a9eebae8076dcb9ac
72240Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72241Date: Mon Apr 6 21:34:14 2009 -0700
72242
72243 XQuartz: Send MotionNotify before button presses when X11 is in the background
72244 (cherry picked from commit c80d0ec18ef5b842447d31360406d0b5b9424222)
72245
72246commit 5ccfad8df099e8ebc4bf2dd53c3db1460903b135
72247Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72248Date: Mon Apr 6 19:21:46 2009 -0700
72249
72250 XQuartz: Revert most of the previous override redirect patch
72251
72252 The changes actually caused all windows to move to the current space. Instead, we're going with a fix entirely within Xplugin that depends on quartz-wm being the window-manager for now.
72253 (cherry picked from commit 997b6f3142c622541bb5bac98652abae75d1101d)
72254
72255commit ff7f019bbcbc52618cc478db7baed57aa5b7c3d3
72256Author: Simon Thum <simon.thum@gmx.de>
72257Date: Wed Apr 1 10:53:10 2009 +0200
72258
72259 dix: correctly utilize tracker buffer and protect from timer overruns
72260
72261 two small related fixes hard to split up
72262
72263 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72264
72265commit 06aebecb19dd9d90d73b742a09b6068b862f1d05
72266Author: Simon Thum <simon.thum@gmx.de>
72267Date: Wed Apr 8 14:35:01 2009 +0200
72268
72269 dix: fix pointer accelerations remainder handling
72270
72271 This didn't really work as intended, but did amazingly well thanks
72272 to roundf() hiding the defect. Cheers!
72273
72274 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72275
72276commit 98f4179156391752e6688339487458ad7828abf4
72277Author: Alan Coopersmith <alan.coopersmith@sun.com>
72278Date: Thu Mar 26 23:04:24 2009 -0700
72279
72280 Use RTLD_DI_SETSIGNAL to catch runtime dynamic loader errors and clean up
72281
72282 Based on fix for Sun bug 6813925: Xorg needs to catch ld.so.1 failure
72283 so it can close down devices cleanly
72284 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6813925>
72285
72286 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
72287
72288commit a0b6a363dca8ce0dc6f4eb79333e48496153cd67
72289Author: Alan Coopersmith <alan.coopersmith@sun.com>
72290Date: Mon Mar 9 13:22:57 2009 -0700
72291
72292 Lift fatal signal handlers from DDX'es up to a common DIX implementation
72293
72294 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
72295
72296commit fcc19e673e3ef33d64916dd933853f8aa667c4d7
72297Author: Peter Hutterer <peter.hutterer@who-t.net>
72298Date: Thu Apr 9 20:27:00 2009 +1000
72299
72300 mi: fix wrong (*EnqueueEvent) declaration in miPointerScreenFuncRec.
72301
72302 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72303
72304commit 4fee979d0632751d3d54d2115e84e9654edf0622
72305Author: Peter Hutterer <peter.hutterer@who-t.net>
72306Date: Thu Apr 9 19:11:27 2009 +1000
72307
72308 Xi: silence compiler warnings about "wrong" event types.
72309
72310 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72311
72312commit 6b467bf879eeb77d167ef321e6dda97ca9d7010a
72313Author: Peter Hutterer <peter.hutterer@who-t.net>
72314Date: Thu Apr 9 08:42:06 2009 +1000
72315
72316 xfree86: shut up compiler warnings - typecast to InternalEvent
72317
72318 Reported-by: Eric Anholt <eric@anholt.net>
72319 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72320
72321commit 8a2a184da78a3e9cbeae8290431f40d5ec7f3636
72322Author: Peter Hutterer <peter.hutterer@who-t.net>
72323Date: Thu Apr 9 08:06:32 2009 +1000
72324
72325 xfree86: fix use of uninitialized variable in DGAProcessPointerEvent.
72326
72327 Reported-by: Eric Anholt <eric@anholt.net>
72328 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72329
72330commit 011cee3103c146c8096b7098a27993f99e07a824
72331Author: Peter Hutterer <peter.hutterer@who-t.net>
72332Date: Thu Apr 9 08:14:39 2009 +1000
72333
72334 dix: fix dev/keybd variable mixup.
72335
72336 Reported-by: Eric Anhold <eric@anholt.net>
72337 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72338
72339commit 630a6e9d14ffbf036fa72f580c72c0172d7c20bd
72340Author: Peter Hutterer <peter.hutterer@who-t.net>
72341Date: Thu Apr 9 08:40:20 2009 +1000
72342
72343 mi: fix compiler warning - explicitly typecast to InternalEvent.
72344
72345 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72346
72347commit e3bb7dea06b9ab0e30d801bf6c3a59f94290aaed
72348Author: Peter Hutterer <peter.hutterer@who-t.net>
72349Date: Thu Apr 9 19:07:36 2009 +1000
72350
72351 mi: add prototype for CopyGetMasterEvent.
72352
72353 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72354
72355commit 346e71525fc545c6ca4ad79425722282d1544459
72356Author: Michel Dänzer <daenzer@vmware.com>
72357Date: Thu Apr 9 09:36:41 2009 +0200
72358
72359 EXA: If the driver can't composite to an a8 mask, try an argb mask for glyphs.
72360
72361 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
72362
72363commit 682d7b55699cacbb2dbcd84a5e816bf6e2d2f02a
72364Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
72365Date: Thu Apr 9 02:48:04 2009 -0400
72366
72367 xselinux: Don't BadAlloc in List* requests if there are no items to list.
72368
72369commit 3a0ee199dcec39596756a995996eac388acf6315
72370Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
72371Date: Thu Apr 9 02:26:24 2009 -0400
72372
72373 config: fix crash caused by strdup(NULL)
72374
72375commit 7b3982eb6518da33ab01c2fbf7ceb45b89f841df
72376Author: Michel Dänzer <daenzer@vmware.com>
72377Date: Thu Apr 9 08:21:09 2009 +0200
72378
72379 glx: Test the error value, not its address...
72380
72381commit 03aebed519986c4dd03e02b3b3d4af1f64595ca7
72382Author: Ian Romanick <ian.d.romanick@intel.com>
72383Date: Wed Apr 8 14:54:30 2009 -0700
72384
72385 Use a #define instead of a magic number
72386
72387 The number of buffers is likely to change in the future, so having
72388 this as a define is the right way to go.
72389
72390 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
72391
72392commit 0d9f3ca7eabd4c514808114d30627f682c8bd030
72393Author: Ian Romanick <ian.d.romanick@intel.com>
72394Date: Wed Apr 8 14:53:46 2009 -0700
72395
72396 Allow GLX sources to build against Mesa 7.4 sources
72397
72398 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
72399
72400commit e8b324102f6e21ae2b8292a6f50d016dd6254dd6
72401Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
72402Date: Wed Apr 8 15:10:16 2009 -0400
72403
72404 xselinux: Don't require incoming context strings to be null-terminated.
72405
72406commit df27b870a8db7a5153b18a556fe77efa590f9eee
72407Author: Kristian Høgsberg <krh@redhat.com>
72408Date: Tue Apr 7 16:28:08 2009 -0400
72409
72410 Convert remaining GLX LookupIDByType() calls
72411
72412commit 92562747a0fdbef1dbedf734cb55dd6a9e1d2994
72413Author: Kristian Høgsberg <krh@redhat.com>
72414Date: Tue Apr 7 13:58:53 2009 -0400
72415
72416 Add validGlxDrawable() and use dixLookupResourceByType().
72417
72418 Fixes deprecation warnings, and fixes a couple of GLX error codes
72419 for failing drawable lookups.
72420
72421commit f70cfc8f90091ef0f5ed0a5b2e023e7fd6369b36
72422Author: Kristian Høgsberg <krh@redhat.com>
72423Date: Thu Apr 2 17:30:19 2009 -0400
72424
72425 Don't stomp on dixLookupDrawable() return value in DoCreateGLXPixmap().
72426
72427commit 30d81ad72e870cc37754bd8c8aadf605450ec16e
72428Author: Kristian Høgsberg <krh@redhat.com>
72429Date: Thu Apr 2 17:24:12 2009 -0400
72430
72431 Make GLX context lookup use dixLookupResourceByType()
72432
72433commit 66539cc05d0b017b9feb4a038499907810140623
72434Author: Alan Coopersmith <alan.coopersmith@sun.com>
72435Date: Thu Mar 26 22:22:32 2009 -0700
72436
72437 Don't leak default font path when appending built-ins
72438
72439 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
72440
72441commit 6c3b633299f12051fcf37fb8439f358de876cf03
72442Merge: adf21db 0cfd481
72443Author: Peter Hutterer <peter.hutterer@who-t.net>
72444Date: Tue Apr 7 19:36:27 2009 +1000
72445
72446 Merge branch 'master' into xi2
72447
72448commit 0cfd48130842685828e0662a27272a3a65facd23
72449Author: Peter Hutterer <peter.hutterer@who-t.net>
72450Date: Mon Mar 23 11:41:16 2009 +1000
72451
72452 dix: Dont change the keyboard mapping on non-keyboard devices.
72453
72454 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72455
72456commit 69e73e5ce0e3971eceed71216d2d7e7d9cd57cc2
72457Author: Adam Jackson <ajax@redhat.com>
72458Date: Mon Apr 6 17:56:49 2009 -0400
72459
72460 config: Remove useless xf86conf{{m,c,re}alloc},free} macros
72461
72462commit 485946120a6522db9298111752de9be3d675e0b3
72463Author: Adam Jackson <ajax@redhat.com>
72464Date: Mon Apr 6 17:51:35 2009 -0400
72465
72466 config: s/xf86configStrdup/strdup/
72467
72468commit 5f769b73441cf29c05a05c728ab81bdde1fc00b6
72469Author: David Jander <david.jander@protonic.nl>
72470Date: Mon Apr 6 13:01:08 2009 -0400
72471
72472 kdrive: Fix segfault in tslib support
72473
72474commit 843166b033dc0544b9f3cf3c91fc3ae3650bda14
72475Author: Adam Jackson <ajax@redhat.com>
72476Date: Mon Apr 6 11:05:17 2009 -0400
72477
72478 os: signal handlers return void.
72479
72480commit aa6fbc2a3679896181610aee7ce2844d4322a5ce
72481Author: Adam Jackson <ajax@redhat.com>
72482Date: Mon Apr 6 10:58:47 2009 -0400
72483
72484 Remove some OS/2 leftovers.
72485
72486commit 2c03a63ff6855272d587b4972a67f64115a1ce4c
72487Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72488Date: Sun Apr 5 03:20:40 2009 -0700
72489
72490 XQuartz: Use updated Xplugin API to send overide-redirect windows to the current space when they're ordered in
72491
72492 This fixes the annoying "using a menu moves me to another space" bug
72493 (cherry picked from commit 9e6dab89e08a26d764ba1aeaeb804c3d25c667da)
72494
72495commit 67621f3b39c387b3f7bfc65781051aea51c0b83f
72496Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72497Date: Sun Apr 5 00:06:57 2009 -0700
72498
72499 XQuartz: Send a MotionNotify event for the mouse cursor when activating X11.app
72500
72501 Otherwise if X11.app was activated with a mouse click, the location of the even is the last location of the cursor before X11 was deactivated
72502 (cherry picked from commit c7457d7b31ddd2ddfd04dd6294bed4532664e1ab)
72503
72504commit e1338d27141125af61634597c5dea880692c4017
72505Author: Julien Cristau <jcristau@debian.org>
72506Date: Sat Apr 4 23:43:28 2009 +0200
72507
72508 xfree86: Remove unused DEFAULT_UNRESOLVED and DEFAULT_BEST_REFRESH macros
72509
72510commit e25d55ad546905c4921f0a26385bdb9f22038429
72511Author: Adam Jackson <ajax@redhat.com>
72512Date: Fri Apr 3 22:38:00 2009 -0400
72513
72514 Revert accidental Makefile change from previous commit
72515
72516commit 472811dc4ba741ab09db184216878700f4adf745
72517Author: Adam Jackson <ajax@redhat.com>
72518Date: Fri Apr 3 22:27:51 2009 -0400
72519
72520 DPMS: Re-export the various DPMS variables.
72521
72522 The drivers might not need them but extmod does. Should move it to
72523 builtin though.
72524
72525commit cbb165ab88cb0810268001e84d87671440baf837
72526Author: Adam Jackson <ajax@redhat.com>
72527Date: Fri Apr 3 18:34:45 2009 -0400
72528
72529 os: Remove the useless -x option
72530
72531commit 6574ab092fb80b7731e06ede653b6af92fa7684e
72532Author: Adam Jackson <ajax@redhat.com>
72533Date: Fri Apr 3 18:03:43 2009 -0400
72534
72535 DPMS: Remove the defaultDPMS* variables
72536
72537commit c1cf36ee29e41901ecd382f0258110ebb4a0124e
72538Author: Adam Jackson <ajax@redhat.com>
72539Date: Fri Apr 3 17:51:55 2009 -0400
72540
72541 DPMS: Simplify command line parsing
72542
72543commit 2d0a4fff7985f58c21023cc4757090ee67067593
72544Author: Adam Jackson <ajax@redhat.com>
72545Date: Fri Apr 3 13:44:37 2009 -0400
72546
72547 DPMS: Remove unused DPMSGet()
72548
72549commit d52fddefaeb43f15b677eefbea4a288a9948373b
72550Author: Adam Jackson <ajax@redhat.com>
72551Date: Fri Apr 3 13:33:03 2009 -0400
72552
72553 DPMS: Align the default timeouts with the default screensaver timeout.
72554
72555 On a typical LCD, a black screensaver is actually worse for power
72556 consumption than a normal screen, because it takes more energy to turn
72557 the crystals opaque. Also, the intermediate DPMS states are essentially
72558 useless and most monitors alias them to the 'off' state, so we may as
72559 well do the same.
72560
72561 As a pleasant side effect, this brings the default DPMS timeouts in line
72562 with the EnergyStar Program Requirements for Computers:
72563
72564 http://www.energystar.gov/index.cfm?c=revisions.computer_spec
72565
72566 which state that products must be "shipped with the display's Sleep mode
72567 set to activate within 15 minutes of user inactivity".
72568
72569commit 38576dc3ae8ddec2baedd2f7e64abdfc56bedf93
72570Author: Adam Jackson <ajax@redhat.com>
72571Date: Fri Apr 3 11:55:41 2009 -0400
72572
72573 DPMS: Simplify some macro silliness.
72574
72575commit 1c01127a214731a21c4784249a429c74f17bd424
72576Author: Adam Jackson <ajax@redhat.com>
72577Date: Fri Apr 3 11:39:18 2009 -0400
72578
72579 DPMS: Unexport all the various setup variables.
72580
72581 The drivers don't need to know any of this.
72582
72583commit a99b0ab918998c9a1b9f8692591b8cb59a12eb9e
72584Author: Adam Jackson <ajax@redhat.com>
72585Date: Fri Apr 3 11:32:00 2009 -0400
72586
72587 DPMS: Fix cast abuse
72588
72589commit 4da347adb1fcc39e8f4fb2cadd98e887d7ac877a
72590Author: Adam Jackson <ajax@redhat.com>
72591Date: Fri Apr 3 11:18:21 2009 -0400
72592
72593 DPMS: Code motion.
72594
72595commit 1731882341c191b4ffd78e0c1fd9297c636e0401
72596Author: Havoc Pennington <hp@pobox.com>
72597Date: Fri Mar 27 12:30:37 2009 -0400
72598
72599 Set bg pixmap of composite overlay window to None (#20912)
72600
72601 Otherwise it's impossible to get the COW without a white
72602 flash on the screen, because it's on top, mapped immediately,
72603 and unaffected by composite redirection. This makes
72604 initial login ugly when it doesn't need to be.
72605
72606commit d2690375dfd994817f004cda133ca2a492c0b956
72607Author: Adam Jackson <ajax@redhat.com>
72608Date: Fri Apr 3 11:06:50 2009 -0400
72609
72610 kdrive: s/KdSaveString/strdup/g
72611
72612 This isn't even funny anymore.
72613
72614commit 91b697efdefba125348dbcaf584ee51a7f8c9bf6
72615Author: Kristian Høgsberg <krh@redhat.com>
72616Date: Wed Apr 1 17:42:33 2009 -0400
72617
72618 Support setTexBuffer2 in AIGLX.
72619
72620 Fixes broken GLX_tfp, specifically, lets compositors ignore un-defined
72621 alpha channel for pixmaps.
72622
72623commit 1c101d75d4855b2698e3fc8d2dd662f20585812f
72624Author: Alan Coopersmith <alan.coopersmith@sun.com>
72625Date: Thu Mar 26 22:25:08 2009 -0700
72626
72627 Don't leak canonical module name and patterns if module is built-in
72628
72629 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
72630 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
72631
72632commit 9180081ec32b6c655d193de4c89b59d222a72579
72633Author: Jeremy Huddleston <jeremy@yuffie.local>
72634Date: Wed Apr 1 13:45:02 2009 -0700
72635
72636 XQuartz: Still send mouse events while X11 is in the background if we have test extensions enabled
72637 (cherry picked from commit f393fb686cfbedb11a1539d4eaf5c491af8b3a64)
72638
72639commit 029307e79a14cad5bde8f1f4a5d0a2f740271f06
72640Author: Jeremy Huddleston <jeremy@yuffie.local>
72641Date: Wed Apr 1 13:07:15 2009 -0700
72642
72643 XQuartz: Return BadRequest when SendPSN isn't implemented rather than success
72644 (cherry picked from commit 49871e99179d5176a1ec516c8f449e2dc219faea)
72645
72646commit 0603fbc1d3b2db5c25faabbfb5e04914110b4d0e
72647Author: Jeremy Huddleston <jeremy@yuffie.local>
72648Date: Wed Apr 1 12:46:51 2009 -0700
72649
72650 XQuartz: Pad xAppleDRINotifyEvent to 32bytes to match sizeof(xEvent)
72651 (cherry picked from commit 276ee3bb087c772f57a987519aa4d94afeceadaa)
72652
72653commit b1dab580bdfb4acfe3feddeda6e760098ec4922a
72654Author: Tormod Volden <debian.tormod@gmail.com>
72655Date: Wed Apr 1 19:32:21 2009 +0200
72656
72657 xfree86: edid quirk for Philips LCD LP154W01-TLAJ
72658
72659 This panel reports its vertical size in cm.
72660
72661 X.Org bug#21000 <http://bugs.freedesktop.org/show_bug.cgi?id=21000>
72662
72663 Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
72664 Signed-off-by: Julien Cristau <jcristau@debian.org>
72665
72666commit 7d94414cf4b16501ce4c4291846498f393f150a4
72667Author: Adam Jackson <ajax@redhat.com>
72668Date: Wed Apr 1 09:42:06 2009 -0400
72669
72670 input: Remove xf86ReloadInputDevs hack
72671
72672 We have input hotplug now, no need to fake it.
72673
72674commit 3311ef6ec93df4325d198715e0fce73af19f0781
72675Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72676Date: Tue Mar 31 18:11:19 2009 -0700
72677
72678 XQuartz: Don't report mouse events while X11 is not the foreground application
72679 (cherry picked from commit 5d1dd35096334b59564d77caef7a97bd58845c5d)
72680
72681commit 89887fadc97f6fd5059886396a6d7cf61de1982f
72682Author: Adam Jackson <ajax@redhat.com>
72683Date: Tue Mar 31 15:09:57 2009 -0400
72684
72685 APM: Don't carp if not available.
72686
72687commit 2e145989d10ef454581f8aa60358ed3782fd6442
72688Author: Adam Jackson <ajax@redhat.com>
72689Date: Mon Mar 30 17:22:28 2009 -0400
72690
72691 dix: Simplify InitClient()
72692
72693commit 252ec504817e05b185e4896a2d899e9c00b8aeef
72694Author: Adam Jackson <ajax@redhat.com>
72695Date: Mon Mar 30 15:18:30 2009 -0400
72696
72697 Document which bits of ClientRec are currently unused
72698
72699commit 0b6e14d46feaf8c46fd3768ee5353b844391b79b
72700Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
72701Date: Mon Mar 30 20:37:25 2009 -0700
72702
72703 XQuartz: Fix mouse tracking for quake, et. al. in wine
72704
72705 Patch courtesy of Codeweavers
72706
72707 Fix mouse movement tracking. For a non-window-related mouse-move event,
72708 calculate the new position by adding the event's delta-x and delta-y values
72709 to the previous mouse position. Do not rely on the current mouse position
72710 because it may have been changed by a XWarpPointer call.
72711 (cherry picked from commit 7a67935b05a475215b9bdbb959e4f7e15f32416f)
72712
72713commit c74220b853cd99d03e33cc62d32431ca3a879483
72714Author: Adam Jackson <ajax@redhat.com>
72715Date: Mon Mar 30 14:56:15 2009 -0400
72716
72717 APM: Use general handlers, not input handlers.
72718
72719 Otherwise APM events get treated as input events, which messes up idle
72720 time accounting and screensavers and such. Not, we hope, that anyone
72721 is using APM anymore.
72722
72723commit 283a081572d8db787c77d09e5ba6bcadebf4f7fe
72724Author: Adam Jackson <ajax@redhat.com>
72725Date: Fri Mar 27 15:56:15 2009 -0400
72726
72727 selinux: Only activate if policy says to be an object manager
72728
72729commit 8e7facfe3013abda12a0c39ad8b4d025618077a7
72730Author: Adam Jackson <ajax@redhat.com>
72731Date: Fri Mar 27 15:48:17 2009 -0400
72732
72733 config: Fall back to vesa for Intel Poulsbo.
72734
72735commit 17239c8e3163fef0159835d43c0dde20e78192a1
72736Author: Adam Jackson <ajax@redhat.com>
72737Date: Fri Mar 27 12:44:20 2009 -0400
72738
72739 misprite: Remove private header, fold into misprite.c
72740
72741commit bb89fc2771d2e54d7e0cd0371db68ddb754a64e5
72742Author: Adam Jackson <ajax@redhat.com>
72743Date: Fri Mar 27 12:36:42 2009 -0400
72744
72745 misprite: Fix software cursor on multiple ScreenRecs
72746
72747commit 2c000f4980d30015e6da95de3c77c0fdda0090e4
72748Author: Adam Jackson <ajax@redhat.com>
72749Date: Fri Mar 27 11:54:52 2009 -0400
72750
72751 misprite: Do window check first for Get{Image,Spans} too
72752
72753commit 8beced6c02610444105f3d0b8ee3a49e1072096d
72754Author: Adam Jackson <ajax@redhat.com>
72755Date: Fri Mar 27 11:43:33 2009 -0400
72756
72757 misprite: RECT_IN_REGION -> miRectIn
72758
72759commit a232116b9ecd41ce55f8eeee9dfe2b67d76479d6
72760Author: Adam Jackson <ajax@redhat.com>
72761Date: Fri Mar 27 11:41:40 2009 -0400
72762
72763 misprite: Fix a typo
72764
72765commit 1f0eaf886d88f3935e43985d2c889b374f412e43
72766Author: Adam Jackson <ajax@redhat.com>
72767Date: Fri Mar 27 11:29:40 2009 -0400
72768
72769 misprite: Fix cast abuse
72770
72771commit e3c1096273446bb7eb993b179ceb989617d149cc
72772Author: Peter Hutterer <peter.hutterer@who-t.net>
72773Date: Mon Mar 23 13:20:22 2009 +1000
72774
72775 dix: build_modmap_from_modkeymap needs to bounds-check its argument.
72776
72777 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72778 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
72779
72780commit 45baef0b77f6f0e5e6230eca045dfcbb1e48c647
72781Author: Peter Hutterer <peter.hutterer@who-t.net>
72782Date: Mon Mar 23 13:32:32 2009 +1000
72783
72784 dix: return BadValue as error in SetModifierMapping.
72785
72786 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72787 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
72788
72789commit 603db34337a61754e0c5f71525011d10eab78411
72790Author: Peter Hutterer <peter.hutterer@who-t.net>
72791Date: Wed Mar 25 15:51:43 2009 +1000
72792
72793 Xext: set POINTER_SCREEN flag in XTestFakeInput if necessary. (RH #490984)
72794
72795 The POINTER_SCREEN flag must be set explicitly for XTest core events to avoid
72796 out-of-range events when the lastSlave was an SD with an explicit axis range.
72797 Device events sent through XTest don't need this flag, they are expected to be
72798 in the valuator range of the device anyway.
72799
72800 Red Hat Bug 490984 <https://bugzilla.redhat.com/show_bug.cgi?id=490984>
72801
72802 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72803
72804commit 4ab93f05664890ff6738ef1089bcd956ce3f06e9
72805Author: Adam Jackson <ajax@redhat.com>
72806Date: Mon Mar 23 15:19:47 2009 -0400
72807
72808 misprite: In SourceValidate, check that it's a Window first
72809
72810 It almost never is. However, you have _lots_ of input devices now, and
72811 walking them all on every Composite operation is not the cheapest thing
72812 in the world.
72813
72814commit d698e62690e0a52fa537f1cd792cab7a5b1bafdf
72815Author: Maarten Maathuis <madman2003@gmail.com>
72816Date: Thu Mar 19 21:35:25 2009 +0100
72817
72818 xkb: plug a memory leak in XkbCopySrvLedInfo (#20756)
72819
72820 X.Org Bug 20756 <http://bugs.freedesktop.org/show_bug.cgi?id=20756>
72821
72822 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72823
72824commit e26f79335bb7a9ad34fe9b23c283a214202f89ed
72825Author: Peter Hutterer <peter.hutterer@who-t.net>
72826Date: Tue Mar 17 16:22:00 2009 +1000
72827
72828 xkb: put a few extra checks in against non-keyboards
72829
72830commit 844bb629fe786ecf536039df3f5f4b3b214c9d40
72831Author: Peter Hutterer <peter.hutterer@who-t.net>
72832Date: Thu Mar 19 15:44:34 2009 +1000
72833
72834 dix: remove a truly useless a = (foo) ? bar : bar; statement.
72835
72836 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72837
72838commit 62183a06e5f72a18abc3d58c4bf5658a20ece217
72839Author: Tomas Carnecky <tom@dbservice.com>
72840Date: Fri Mar 20 16:19:33 2009 +0100
72841
72842 Remove two unused defines in C files
72843
72844 These two defines were defined in C files but not used anywhere:
72845
72846 dix/window.c #define DeviceEventMasks (KeyPressMask | [...]
72847 os/connection.c #define MAXFD 500
72848
72849 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
72850 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72851
72852commit 4e0d7cc506b661b5e5cedf8be6e17aa4c8075aa8
72853Author: Tomas Carnecky <tom@dbservice.com>
72854Date: Fri Mar 20 16:19:34 2009 +0100
72855
72856 FID, whatever that was, isn't anymore
72857
72858 No traces of FID in the xserver nor in the modules listed in
72859 util/modular/xorg.modules
72860
72861 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
72862 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72863
72864commit adf21dba7617542f08309415e315d4b2699c10e0
72865Author: Peter Hutterer <peter.hutterer@who-t.net>
72866Date: Thu Mar 19 20:40:43 2009 +1000
72867
72868 include: un-export a bunch of server-only functions.
72869
72870 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72871
72872commit 111ef10375204a6981a5be5c045860854c4da4e6
72873Author: Peter Hutterer <peter.hutterer@who-t.net>
72874Date: Thu Mar 19 09:33:02 2009 +1000
72875
72876 dix: move ProcGrabPointer guts into GrabDevice.
72877
72878 Yes, this means we have even more arguments to GrabDevice. But it beats having
72879 a copy of most but not all of GrabDevice in ProcGrabPointer.
72880 Also, reshuffle the order of parameters, the CARD* status is a return value
72881 and should be last.
72882
72883 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72884
72885commit dc153271b6458cacd63a4bc0208594011c6460b0
72886Author: Peter Hutterer <peter.hutterer@who-t.net>
72887Date: Wed Mar 18 15:13:00 2009 +1000
72888
72889 Xi: purge old device enter/leave masks.
72890
72891commit 15a969c0fcd4f19f3c560a7037c1a6a58bce51f0
72892Author: Peter Hutterer <peter.hutterer@who-t.net>
72893Date: Wed Mar 18 10:21:59 2009 +1000
72894
72895 dix: remove now obsolete mskidx parameter from DeliverEventsToWindow.
72896
72897 mskidx would always be dev->id anyway, so if we're already passing in the
72898 device, mskidx is superfluous.
72899
72900commit f5409aa026b2cb501170867d04c6e220f3fe0a44
72901Author: Peter Hutterer <peter.hutterer@who-t.net>
72902Date: Fri Mar 13 15:51:50 2009 +1000
72903
72904 mi: remove deprecated miPointerAbsoluteCursor
72905
72906 Functions that don't specify the device are so last year.
72907
72908commit 46145a9312eaf5dcd0e8b6788ed8219e086f790e
72909Author: Peter Hutterer <peter.hutterer@who-t.net>
72910Date: Fri Mar 13 15:12:55 2009 +1000
72911
72912 dix: remove coreMods field from GrabRec.
72913
72914 Nobody uses it anyway, and it's taking up a whole bit!
72915
72916commit 40e2a61e37109cd9c2ab57954ee40dcf4a062e31
72917Author: Peter Hutterer <peter.hutterer@who-t.net>
72918Date: Thu Mar 12 16:53:56 2009 +1000
72919
72920 Add Get/SetDeviceFocus handling.
72921
72922commit d9b7343eac1157490442cc0fb3b5b6d7d6e9a705
72923Author: Peter Hutterer <peter.hutterer@who-t.net>
72924Date: Thu Mar 12 12:56:18 2009 +1000
72925
72926 dix: even if we don't get a XI1 event, continue processing.
72927
72928commit e11dc10f01603b181e224c90d1dd4abe940f2ce7
72929Author: Peter Hutterer <peter.hutterer@who-t.net>
72930Date: Thu Mar 12 10:21:00 2009 +1000
72931
72932 dix: Send HierarchyEvents when devices are added/removed/enabled/disabled.
72933
72934commit 8634e1bcbae9317fc22132f7c9bde1a4a881b9a7
72935Author: Peter Hutterer <peter.hutterer@who-t.net>
72936Date: Thu Mar 12 10:42:05 2009 +1000
72937
72938 dix: send presence events when floating and unfloating SDs
72939
72940commit c9483a53bea1f721d2902ac11467f67317e7de14
72941Author: Peter Hutterer <peter.hutterer@who-t.net>
72942Date: Wed Mar 11 16:20:22 2009 +1000
72943
72944 include: add a few prototypes to silence compiler warnings.
72945
72946commit a4b61cfc78d81de02a71fa6ad8a44a06616f6794
72947Author: Peter Hutterer <peter.hutterer@who-t.net>
72948Date: Wed Mar 11 16:08:24 2009 +1000
72949
72950 dix: un-static FixUpEventFromWindow and deal with focus events too.
72951
72952commit 4cc6a96d7171e567a9bd5a10f552bb953077aafb
72953Author: Peter Hutterer <peter.hutterer@who-t.net>
72954Date: Tue Mar 10 16:08:14 2009 +1000
72955
72956 input: add support for RawDeviceEvents.
72957
72958commit a668d91e28d5a3042a8ce0d087474883b046869a
72959Author: Peter Hutterer <peter.hutterer@who-t.net>
72960Date: Mon Mar 9 16:36:07 2009 +1000
72961
72962 dix: store the xi2mask on grabs and pass it around as needed.
72963
72964 This enables passive and implicit passive grabs for XI2 events, except that we
72965 don't have the protocol spec yet to request them.
72966
72967commit 32f338263ff7de1a2e76d570c98f5be979c18d4e
72968Author: Peter Hutterer <peter.hutterer@who-t.net>
72969Date: Sun Mar 8 21:32:31 2009 +1000
72970
72971 Xi: Deliver XI2 HierarchyEvents when the hierarchy changes.
72972
72973 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
72974
72975commit 97e89a59572a4be6757510a317c142ec1d82e8f8
72976Author: Peter Hutterer <peter.hutterer@who-t.net>
72977Date: Fri Mar 6 23:22:00 2009 +1000
72978
72979 Allow XI2 event selection for AllDevices and AllMasterDevices.
72980
72981commit 808a158182b0a3b985385899252c190947b4dbbd
72982Author: Peter Hutterer <peter.hutterer@who-t.net>
72983Date: Fri Mar 6 14:45:26 2009 +1000
72984
72985 dix: Enable XI2 delivery for events and focused events.
72986
72987 No support for grabbed events yet.
72988
72989commit 87ff1159b497c66554a4036ae619f92d5631f00f
72990Author: Peter Hutterer <peter.hutterer@who-t.net>
72991Date: Wed Mar 11 09:01:23 2009 +1000
72992
72993 dix: deliver device enter/leave events.
72994
72995commit 04ed0bcb2530866f7248b412974ecd15f0fbf6ac
72996Author: Peter Hutterer <peter.hutterer@who-t.net>
72997Date: Mon Feb 23 17:07:55 2009 +1000
72998
72999 xkb: remove Device/Enter leave handling - XI2 enter/leave don't have compat state.
73000
73001commit 3f37923a727d5fc4b1880a53dac1145884ba7140
73002Author: Peter Hutterer <peter.hutterer@who-t.net>
73003Date: Mon Feb 23 16:12:25 2009 +1000
73004
73005 Xi: send XI2 focus events.
73006
73007commit daf7dd3bfecc9029bdd10c7a4ea5d2c875a89c41
73008Author: Peter Hutterer <peter.hutterer@who-t.net>
73009Date: Wed Mar 11 09:01:12 2009 +1000
73010
73011 dix: Add GetWindowXI2Mask helper
73012
73013commit 8b6a370058ad5a20e0a0e49ec9443daf03775de8
73014Author: Peter Hutterer <peter.hutterer@who-t.net>
73015Date: Mon Feb 23 15:58:07 2009 +1000
73016
73017 Add XI2 masks and XISelectEvent() request handling.
73018
73019 XI2 event masks are simply stored in the OtherEventMasks as a separate field.
73020 This replaces the XiSelectEvent code.
73021
73022commit 38bba0c1b75b84e8bbdfa7975cf701a9414a3afd
73023Author: Peter Hutterer <peter.hutterer@who-t.net>
73024Date: Fri Feb 20 16:07:33 2009 +1000
73025
73026 Xi: Change ChangeMasterDeviceClasses to new XI2 events.
73027
73028 Split ChangeMasterDeviceClasses into an extra XISendDeviceChangedEvent that
73029 assembles the XI2 wire event for the DeviceChanged event. Re-use this when
73030 detaching the last SD.
73031
73032 Not quite perfect yet, we still copy the device classes from the slave now
73033 rather than from the data we had when the event occured. But it's a start.
73034
73035 (We can now unexport SizeDeviceInfo and CopySwapDevices, not needed anymore)
73036
73037 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73038
73039commit 181e41511d5772962b8bc998a1940d2a9ba62d5e
73040Author: Peter Hutterer <peter.hutterer@who-t.net>
73041Date: Mon Feb 23 16:14:51 2009 +1000
73042
73043 dix: fix up device enter/leave for XI_Enter/XI_Leave.
73044
73045commit 0befeb36c175b14eac8d5416851f1556af90dd86
73046Author: Peter Hutterer <peter.hutterer@who-t.net>
73047Date: Tue Feb 24 15:27:07 2009 +1000
73048
73049 dix: Add device info to DeviceChangedEvent, and fill in CCCE.
73050
73051 We need to fill the info here, as the device may change until we get a chance
73052 to process it.
73053
73054 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73055
73056commit b2ba77bac4fc3edab39372dfdc529e745ae74f77
73057Author: Peter Hutterer <peter.hutterer@who-t.net>
73058Date: Thu Mar 5 14:23:47 2009 +1000
73059
73060 dix: add EventToXI2 and GetXI2Type.
73061
73062 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73063
73064commit dfa2e8e966b699f9d292628119d5a9cbd2dbdf5a
73065Author: Peter Hutterer <peter.hutterer@who-t.net>
73066Date: Mon Feb 23 15:12:32 2009 +1000
73067
73068 Xi: add XIQueryDevice request handling.
73069
73070commit 560c58b53e4e60ebd26e0c66dd00399c438bd619
73071Author: Peter Hutterer <peter.hutterer@who-t.net>
73072Date: Fri Feb 20 15:17:52 2009 +1000
73073
73074 Xi: name-space and fix XI2 requests
73075
73076 All XI2 requests have XI prefixes.
73077 Requests affected:
73078 ChangeDeviceCursor
73079 QueryDevicePointer
73080 WarpDevicePointer
73081 SetClientPointer
73082 GetClientPointer
73083 XIChangeDeviceHierarchy.
73084
73085commit 903449615b1a1188955c0287a552aec8ac5b3fbb
73086Author: Peter Hutterer <peter.hutterer@who-t.net>
73087Date: Fri Feb 20 15:03:35 2009 +1000
73088
73089 Xi: add XIQueryVersion request handling.
73090
73091commit 77d51b94bd264be5a8625a2bdd98f2bfb2aa33d3
73092Author: Peter Hutterer <peter.hutterer@who-t.net>
73093Date: Sun Mar 8 11:47:26 2009 +1000
73094
73095 dix: FixUpEventForWindow needs to handle XI2 events.
73096
73097commit 95ed2ab715fa881d36d508adf4ae5dadd92ae7b2
73098Author: Peter Hutterer <peter.hutterer@who-t.net>
73099Date: Thu Mar 12 11:46:07 2009 +1000
73100
73101 Add XI2_EVENT check.
73102
73103commit 92e0a324b5c738166ed4046cb00fe85ab8aea5a0
73104Author: Peter Hutterer <peter.hutterer@who-t.net>
73105Date: Thu Mar 12 14:15:02 2009 +1000
73106
73107 require inputproto 1.9.99.7
73108
73109commit 906fe9547957917ac463d646935473ceca10a72b
73110Author: Peter Hutterer <peter.hutterer@who-t.net>
73111Date: Fri Mar 20 09:36:09 2009 +1000
73112
73113 Xi: don't allow OpenDevice on any MD.
73114
73115 This restores the original behaviour of XI 1. Any slave device may be opened,
73116 but none of the MDs.
73117
73118 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73119
73120commit c7eb27f3927ba9a36f9904938996f346cca33021
73121Author: Peter Hutterer <peter.hutterer@who-t.net>
73122Date: Wed Mar 4 18:45:09 2009 +1000
73123
73124 Xi: always only list VCP/VCK and the SD's in XListInputDevices.
73125
73126 If a client wants other MDs, it should use XI2's XQueryDevice() instead.
73127
73128 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73129
73130commit 801f14d4b742e506201b619d472087a7e82bd018
73131Author: Peter Hutterer <peter.hutterer@who-t.net>
73132Date: Tue Mar 3 17:04:08 2009 +1000
73133
73134 Xi: don't send attachment info down with ListInputDevices.
73135
73136commit 66dd8879b3527f30309fb747cad4a161d0709d91
73137Author: Peter Hutterer <peter.hutterer@who-t.net>
73138Date: Fri Feb 20 15:01:05 2009 +1000
73139
73140 Xi: purge ExtendedGrabDevice request handling.
73141
73142commit 43e48900fccd245e98d6d697e4cdbb6fa866a0dd
73143Author: Peter Hutterer <peter.hutterer@who-t.net>
73144Date: Fri Mar 6 22:27:54 2009 +1000
73145
73146 dix: Force deviced ids of 2 and higher for actual devices.
73147
73148 0 is now reserved for the "AllDevices" virtual device.
73149 1 is now reserved for the "AllMasterDevices" virtual device.
73150
73151 This also means that wherever we passed in (mskidx = 0), we now need to pass
73152 in the deviceid.
73153
73154commit 70896f6127e2c12c13479a027fdd6e67853d1f8b
73155Author: Peter Hutterer <peter.hutterer@who-t.net>
73156Date: Fri Feb 13 16:42:02 2009 +1000
73157
73158 dix: abstract event filters through GetEventFilters.
73159
73160 Don't let everyone acces the filters[] array directly. This is necessary once
73161 we start dealing with GenericEvents, where the filters are a bit more
73162 complicated.
73163
73164 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73165
73166commit 8a8763f1502c9ddd1961f96f2cc7c73fbc0b49b5
73167Author: Peter Hutterer <peter.hutterer@who-t.net>
73168Date: Thu Feb 12 16:07:15 2009 +1000
73169
73170 dix: restructure DeliverEventsToWindow, a little bit anyway.
73171
73172 Desperate attempt of making the code easy to read and understand. A failed
73173 desparate attempt, possibly.
73174
73175commit 891cfc802028c3e008ee6c2d3e2438a31f57f39f
73176Author: Peter Hutterer <peter.hutterer@who-t.net>
73177Date: Thu Feb 12 10:12:45 2009 +1000
73178
73179 dix: clean up DeliverDeviceEvents.
73180
73181 Move mask checking into external function and streamline the code.
73182
73183 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73184
73185commit 6b6b660c3633af75a94302725137549df0dbe446
73186Author: Peter Hutterer <peter.hutterer@who-t.net>
73187Date: Tue Mar 10 10:15:45 2009 +1000
73188
73189 Xext: remove ev_fill from GEExtensions.
73190
73191commit 445daa62e720fe58fa2d6c3cdc26663671b2bc48
73192Author: Peter Hutterer <peter.hutterer@who-t.net>
73193Date: Thu Jan 15 17:16:22 2009 +1000
73194
73195 Xext: purge XGE event masks.
73196
73197 The masks were originally designed to generically handle event masks for
73198 extensions. Since all that is in-server anyway, it's much better writing
73199 custom event masks for those extensions that need it and not providing a
73200 unified mechanism.
73201 XI2 needs more than the current implementation, which is already too complex
73202 for most other extensions. good riddance.
73203
73204 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73205
73206commit 6fe4c28bf4f3a9f7c341b9403213470e9ded5cb3
73207Author: Peter Hutterer <peter.hutterer@who-t.net>
73208Date: Mon Feb 16 16:34:05 2009 +1000
73209
73210 Add agressive event type checking.
73211
73212 Best to FatalError if a wrong event comes in. At least that forces me to fix
73213 it really quickly.
73214
73215 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73216
73217commit 836864b65794dc0954a01245e418e714cead8125
73218Author: Peter Hutterer <peter.hutterer@who-t.net>
73219Date: Fri Mar 13 12:43:17 2009 +1000
73220
73221 xkb: don't overrun the map index when accessing symbols.
73222
73223 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73224
73225commit 497a12533905b98f388775b6ba49adf21017cc75
73226Author: Peter Hutterer <peter.hutterer@who-t.net>
73227Date: Fri Mar 13 11:52:33 2009 +1000
73228
73229 dix: remove duplicate PickKeyboard() command.
73230
73231 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73232
73233commit 7d4df0ee4414779ad5e519e7bd297c5dfec409a3
73234Author: Peter Hutterer <peter.hutterer@who-t.net>
73235Date: Thu Mar 19 20:24:06 2009 +1000
73236
73237 dix: fix device sync state when calling SyncBoth during AllowEvents.
73238
73239 This did access the wrong device's sync state, potentially freezing or not
73240 thawing the actual device that was supposed to be thawed.
73241
73242 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73243
73244commit f1c7b95d83948160a0d5796ef6c16fc0d1bf5c5d
73245Author: Peter Hutterer <peter.hutterer@who-t.net>
73246Date: Fri Mar 20 13:55:00 2009 +1000
73247
73248 dix: do percentage check before device check in ProcBell
73249
73250 This is just for correctness. The server should return BadValue for anything
73251 not in [-100, 100].
73252
73253 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73254
73255commit d60391d8ca9918d2089c23c0baef5c91177325f0
73256Author: Peter Hutterer <peter.hutterer@who-t.net>
73257Date: Fri Mar 20 14:01:30 2009 +1000
73258
73259 dix: fix uncredible fail in PostSyntheticMotion.
73260
73261 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73262
73263commit 1a71862d333282e2d251ff0036866cec22bcce85
73264Author: Simon Thum <simon.thum@gmx.de>
73265Date: Sat Feb 28 22:17:47 2009 +0100
73266
73267 dix/xfree86: simplified velocity approximation algorithm
73268
73269 Replace multi-stage filtering with simple linear velocity,
73270 tracked several instances backwards. A heuristic ensures
73271 only approximately linear motion is considered, so velocity
73272 remains valid in any case. Numerical stability is much
73273 better, and nothing changes to people who didn't tune the
73274 advanced features of the previous algorithm.
73275
73276 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73277
73278commit 5ae129baef85b47590c02e4cf61b23904d8f7aa9
73279Author: Peter Hutterer <peter.hutterer@who-t.net>
73280Date: Tue Mar 17 15:47:57 2009 +1000
73281
73282 Xi: check for existence of the button class before accessing it
73283
73284 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73285
73286commit e8094d8f3f69f45850af031efd79a3928e584638
73287Author: Peter Hutterer <peter.hutterer@who-t.net>
73288Date: Thu Mar 19 10:09:04 2009 +1000
73289
73290 dix: ProcSendEvent shouldn't use inputInfo.keyboard directly.
73291
73292 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73293
73294commit 12aeddf5ad41902a180f8108623f356642b3e911
73295Author: Maarten Maathuis <madman2003@gmail.com>
73296Date: Thu Mar 19 19:07:59 2009 +0100
73297
73298 exa: Accept scratch pixmaps with offscreen memory as such.
73299
73300commit a635bb5357648779815ec6729d55f6b478493213
73301Author: Maarten Maathuis <madman2003@gmail.com>
73302Date: Tue Mar 17 19:34:45 2009 +0100
73303
73304 exa: round of fb_pitch to the next byte
73305
73306commit ec328783593339e33772301a8c5fea35fd63c0a2
73307Author: Maarten Maathuis <madman2003@gmail.com>
73308Date: Tue Mar 17 10:45:21 2009 +0100
73309
73310 exa: check if the pixmap is pinned for unsupported AUX indices.
73311
73312commit 3992dd38caf33b343affd8d732c94880d1099dcf
73313Author: Adam Jackson <ajax@redhat.com>
73314Date: Mon Mar 16 13:24:48 2009 -0400
73315
73316 selinux: Add support for avc_acquire_netlink_fd()
73317
73318 Requires libselinux 2.0.79 or newer. Without this, libselinux will
73319 check for policy updates on the netlink socket on basically every policy
73320 lookup. Statistically speaking, they never happen, and the check
73321 translates to at least one more syscall on basically every operation.
73322
73323 Instead, take control of the fd from the library, and check it in
73324 WakeupHandler if it polls readable.
73325
73326commit be6dc9023b5fb3995a9ce56d607627e247918aef
73327Author: Peter Hutterer <peter.hutterer@who-t.net>
73328Date: Fri Mar 13 16:29:36 2009 +1000
73329
73330 xfree86: remove a superfluous assignment.
73331
73332 If dev is NULL, we don't have a VCP and that means we have other issues
73333 anyway.
73334
73335commit bfab422dcdbb9c009b2a91fe0dba288fac9bc859
73336Author: Peter Hutterer <peter.hutterer@who-t.net>
73337Date: Fri Mar 13 16:27:49 2009 +1000
73338
73339 xfixes: useless (void) typecast removal
73340
73341 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73342
73343commit 4eeaee1e5a05ea973cbeb653abf99f2a352edc70
73344Author: Peter Hutterer <peter.hutterer@who-t.net>
73345Date: Fri Mar 13 16:25:12 2009 +1000
73346
73347 xkb: xkbi has a pointer to the device - use this instead of inputInfo.pointer.
73348
73349commit 3f801ba62ac493d4aeb664a5ae638e3fb0c9a262
73350Author: Peter Hutterer <peter.hutterer@who-t.net>
73351Date: Fri Mar 13 16:23:50 2009 +1000
73352
73353 xkb: the VCP can post device events, don't stop xkb filtering on it.
73354
73355commit 23686e5680d565c36e0b6320ce62c38c7830d06b
73356Author: Peter Hutterer <peter.hutterer@who-t.net>
73357Date: Fri Mar 13 16:20:56 2009 +1000
73358
73359 xkb: Fix a mis-use of inputInfo.keyboard.
73360
73361commit c97c6c3de337f247406eaffd5845a847630ee6c1
73362Author: Peter Hutterer <peter.hutterer@who-t.net>
73363Date: Fri Mar 13 16:17:20 2009 +1000
73364
73365 xkb: fix a couple of device checks when looping through all devices.
73366
73367 Generally, we want to apply stuff to the device and to all attached slave
73368 devices.
73369
73370commit d21a546e9604ce2d3c914900f6e6ff2df9f5ca8b
73371Author: Peter Hutterer <peter.hutterer@who-t.net>
73372Date: Fri Mar 13 16:02:22 2009 +1000
73373
73374 Xi: VCP and VCK may register for ext. events, so delete them if necessary.
73375
73376commit 6ec49098f7ef259b43870139511006627b308bb6
73377Author: Peter Hutterer <peter.hutterer@who-t.net>
73378Date: Fri Mar 13 15:57:27 2009 +1000
73379
73380 render: replace a wrong inputInfo.pointer with pDev
73381
73382 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73383
73384commit 0b2b330f6fbf8633ae682a17fbd95f6b5db2c112
73385Author: Peter Hutterer <peter.hutterer@who-t.net>
73386Date: Fri Mar 13 15:54:27 2009 +1000
73387
73388 mi: remove a bunch of useless inputInfo.pointer assignments.
73389
73390 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73391
73392commit 31301861c1dd87a2f55c5c4300ac7c8f54154bf8
73393Author: Peter Hutterer <peter.hutterer@who-t.net>
73394Date: Fri Mar 13 15:49:22 2009 +1000
73395
73396 dix: remove wrong InputInfo.pointer usage - should be pDev instead.
73397
73398commit 639658b2eab480a5bfb943dff51e4bdfc1e16e3b
73399Author: Peter Hutterer <peter.hutterer@who-t.net>
73400Date: Fri Mar 13 15:34:29 2009 +1000
73401
73402 dix: use GetPairedDevice in ProcQueryPointer rather than inputInfo.keyboard.
73403
73404commit f01ee198ff0cbd7ca418217a84248d1c6131a0c6
73405Author: Peter Hutterer <peter.hutterer@who-t.net>
73406Date: Fri Mar 13 15:22:23 2009 +1000
73407
73408 dix: don't use inputInfo.keyboard to get the focus window in ActivateKbdGrab
73409
73410 I'm not sure if that's the right solution, but the other one is wronger.
73411
73412commit de415743f4c1878df2de79f84f5fe087a88d5241
73413Author: Peter Hutterer <peter.hutterer@who-t.net>
73414Date: Fri Mar 13 15:19:23 2009 +1000
73415
73416 dix: fix XACE checks in ProcWarpPointer
73417
73418commit 18430616cfd5d0bc5a3efe61fcc2982b1726b0de
73419Author: Peter Hutterer <peter.hutterer@who-t.net>
73420Date: Fri Mar 13 14:59:18 2009 +1000
73421
73422 dix: use PickPointer() and PickKeyboard instead of inputInfo.pointer/keyboard.
73423
73424 Affected calls:
73425 ProcChangeKeyboardControl
73426 ProcBell
73427 ProcUngrabKey
73428 ProcUngrabButton
73429
73430commit e31727158cda7729283233cfdc9bc5bcb59a52c1
73431Author: Peter Hutterer <peter.hutterer@who-t.net>
73432Date: Fri Mar 13 14:51:55 2009 +1000
73433
73434 dix: fix ProcChangePointerControl's wrong inputInfo.pointer usage.
73435
73436 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73437
73438commit fc5e8dc4a24c2b5bed4e6b5d00670c4401ec92d9
73439Author: Adam Jackson <ajax@redhat.com>
73440Date: Fri Mar 13 14:00:37 2009 -0400
73441
73442 Unexport xf86SetPriority
73443
73444commit 89d7b88f32f9227fa72a043675367b94c4c9283c
73445Author: Maarten Maathuis <madman2003@gmail.com>
73446Date: Fri Mar 13 00:40:43 2009 +0100
73447
73448 exa: allow exaModifyPixmapHeader to set sys_ptr for EXA_HANDLES_PIXMAPS
73449
73450 - exaModifyPixmapHeader would save sys_ptr if needed, but it would be NULL'ed afterwards.
73451 - This is needed to support pixmaps that are not offscreen.
73452
73453commit 544cd9e7b50cd9905afc52404de1a5a2bcff91b5
73454Author: Ander Conselvan de Oliveira <ander@mandriva.com.br>
73455Date: Thu Mar 5 12:57:28 2009 -0500
73456
73457 Honour Option "DPMS" "off" on xorg.conf
73458
73459 Signed-off-by: James Cloos <cloos@jhcloos.com>
73460
73461commit 7abb72c5c711a26a27a82604bd2398bcb6df0a99
73462Author: Simon Thum <simon.thum@gmx.de>
73463Date: Wed Feb 25 10:55:36 2009 +0100
73464
73465 xfree86: remove obsolete code
73466
73467 remove a few lines which redo part of the pointer acceleration
73468 init. Properties is the way to go for them.
73469
73470 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73471
73472commit 97cb3cc28e4a5694d17f156276efd84555a96d18
73473Author: Peter Hutterer <peter.hutterer@who-t.net>
73474Date: Thu Mar 12 10:17:16 2009 +1000
73475
73476 dix: Merge DevicePresence notify events generation into a single function.
73477
73478 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73479
73480commit 7f1ba804a1c1aef725e20a04e5f6ef74478ecf44
73481Author: Peter Hutterer <peter.hutterer@who-t.net>
73482Date: Tue Mar 10 11:55:04 2009 +1000
73483
73484 dix: s/numEvents/num_events/ in GetKeyboardValuatorEvents
73485
73486 GPE and GProxE use the same.
73487
73488 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73489
73490commit 067f26f7ba4c8cf37af7ddf0eb204b7e1d06096d
73491Author: Peter Hutterer <peter.hutterer@who-t.net>
73492Date: Fri Mar 6 19:42:27 2009 +1000
73493
73494 Xi: set the modifier + group state during POE.
73495
73496 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73497
73498commit da724e8f1d1867c36c9550a153c58829939800fc
73499Author: Peter Hutterer <peter.hutterer@who-t.net>
73500Date: Fri Mar 6 15:50:05 2009 +1000
73501
73502 Xi: Set the button state on the event during POE.
73503
73504 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73505
73506commit b368dd1720c708aaec492b4a67fb7e4f87569c11
73507Author: Peter Hutterer <peter.hutterer@who-t.net>
73508Date: Thu Mar 5 15:22:41 2009 +1000
73509
73510 dix: remove a few pointless (void) casts of return values.
73511
73512 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73513
73514commit cadf65a6e190a8952ad3cc216dc9ea55241de91a
73515Author: Matthias Hopf <mhopf@suse.de>
73516Date: Tue Mar 10 15:49:49 2009 +0100
73517
73518 randr: Nuke broken set_origin shortcut
73519
73520 Shortcut is impossible to implement this way, because we don't know for sure
73521 whether the crtc of an output has changed or not.
73522
73523commit 6544490700051b3b5e88ac1890d71b35634c9100
73524Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
73525Date: Mon Mar 9 17:28:40 2009 -0400
73526
73527 Correct access mode in call to dixLookupWindow() within RRSelectInput.
73528
73529 Reported by Alan Coopersmith.
73530
73531commit f8dd80d13bb5313a11b38b280f8ad3e22f0a6300
73532Author: Keith Packard <keithp@keithp.com>
73533Date: Wed Aug 6 15:26:24 2008 -0700
73534
73535 Replace dixLookupResource by dixLookupResourceBy{Type,Class}
73536
73537 dixLookupResource attempted to automatically detect whether the caller
73538 wanted a lookup by-type or by-class, unfortunately, it guessed wrong for
73539 RT_NONE. Instead of trying to make the guess better, this patch just reverts
73540 the unification and creates separate functions for each operation.
73541
73542commit 0d9a42dc0380d1583889b6b6521bd5a2451735d4
73543Author: Peter Hutterer <peter.hutterer@who-t.net>
73544Date: Mon Mar 9 08:59:01 2009 +1000
73545
73546 Xi: don't crash on a NULL property name, just return None.
73547
73548 strcmp doesn't like NULL-pointers.
73549
73550 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73551
73552commit 432a95032eb10f82fb59caa29e52d3f56dfb363c
73553Author: Peter Hutterer <peter.hutterer@who-t.net>
73554Date: Sun Mar 8 21:34:02 2009 +1000
73555
73556 Remove useless (void) typecasts.
73557
73558 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73559
73560commit f508446a2bcd887df9cf5db786b15fc3cbc435a0
73561Author: Peter Åstrand <astrand@cendio.se>
73562Date: Fri Mar 6 19:41:35 2009 -0800
73563
73564 Composite: Uninitialized borderClipX/borderClipY
73565
73566 Valgrind warned me about two other uninitialized variables, which are not
73567 padding. Valgrind output at the end. I'm a bit unsure of what these should
73568 be initialized to, is zero fine?
73569
73570commit d1338a94805cc774fe0a5d00c2225a3ee9673a9f
73571Author: Winfried Grünewald <winfried.gruenewald.ext@eizo-dt.de>
73572Date: Fri Mar 6 12:59:35 2009 -0500
73573
73574 [hw/xfree86] Fix StaticGray cmap.
73575
73576 Fix this bug report:
73577
73578 ,----< from http://bugzilla.freedesktop.org/show_bug.cgi?id=20504 >
73579 | Using the Visual StaticGray (8 bit depth) is missing one gray level.
73580 | The gray level of index zero and index one are the same and all
73581 | other levels are shifted by one. The max level (255) cannot be used.
73582 `----
73583
73584 Signed-off-by: James Cloos <cloos@jhcloos.com>
73585
73586commit eba3bab71ff26d58bb0e49e4a9726fc21ff07258
73587Author: Alan Coopersmith <alan.coopersmith@sun.com>
73588Date: Thu Mar 5 15:09:28 2009 -0800
73589
73590 Remove #ifdef macII code left over from ancient A/UX 3.0 support
73591
73592 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
73593
73594commit daa9727a02701cfac7b92f3e61d752be6fcc005e
73595Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
73596Date: Wed Mar 4 11:33:00 2009 -0800
73597
73598 XQuartz: Implement SendPSN for AppleWM
73599 (cherry picked from commit 779e9fdc2de9bfebc5f0a4c9d5f42e389fdd8593)
73600
73601commit ce6e1771be5f2c21af6f72a9705795df26210413
73602Author: Maarten Maathuis <madman2003@gmail.com>
73603Date: Mon Mar 2 17:05:28 2009 +0100
73604
73605 exa: fix a serious issue in exaChangeWindowAttributes (and some more related things)
73606
73607 - fbChangeWindowAttributes can create pixmaps (and access them) without use preparing access.
73608 - Also handle the destroyed pixmaps by finishing them first.
73609 - Switch to DEST indices again in exaCreatePixmapWithPrepare, because they are obviously being rendered to.
73610 - Also avoid calling FinishAccess on pixmaps that are destroyed (and their memory potentially invalid).
73611
73612commit 7fb68e8b31d259325ce311ad67befc43a933f009
73613Author: Maarten Maathuis <madman2003@gmail.com>
73614Date: Sat Feb 28 22:29:42 2009 +0100
73615
73616 exa: remove a few pExaPixmap checks.
73617
73618 - And make some fatal for a debug build.
73619
73620commit bd2f35ef0d4b8c56eb205b8694b4fe5d2fa2c049
73621Author: Maarten Maathuis <madman2003@gmail.com>
73622Date: Sat Feb 28 22:15:04 2009 +0100
73623
73624 exa: fix unwrapping of ModifyPixmapHeader upon CloseScreen.
73625
73626 - Cleanup wrapping too.
73627
73628commit da8ea41a542787691ea1120e5c8c7dc3182cbea5
73629Author: Maarten Maathuis <madman2003@gmail.com>
73630Date: Sat Feb 28 21:59:09 2009 +0100
73631
73632 exa: increase/rework safety checks in Prepare/FinishAccess.
73633
73634commit 3ea3d505e8128ab3e878edcef697fd5656b0c917
73635Author: Maarten Maathuis <madman2003@gmail.com>
73636Date: Sun Mar 1 01:15:28 2009 +0100
73637
73638 exa: avoid a potential Prepare/FinishAccess inbalance.
73639
73640commit 10334cf7e64e2e633cd507f2aa216027e840c8e5
73641Author: Maarten Maathuis <madman2003@gmail.com>
73642Date: Sat Feb 28 21:46:00 2009 +0100
73643
73644 exa: simplify exaPixmapIsOffscreen
73645
73646 - This includes properly handling the framebuffer.
73647
73648commit 57bed76e218471b60d340b18c346d492f37798c2
73649Author: Tomas Carnecky <tom@dbservice.com>
73650Date: Wed Mar 4 10:29:34 2009 -0500
73651
73652 Remove dead EXTENSION_PROC_ARGS define
73653
73654 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
73655
73656commit a6fb9f4ccfc69dc090b152e29142e3159f812166
73657Author: Tomas Carnecky <tom@dbservice.com>
73658Date: Wed Mar 4 10:27:30 2009 -0500
73659
73660 Remove dead XTest code from extmod
73661
73662 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
73663
73664commit c4081fee0995dae691665a00ea9bfdb8b8043b8c
73665Author: Topi Kanerva <topi.kanerva@nokia.com>
73666Date: Wed Mar 4 10:23:10 2009 -0500
73667
73668 Make DGA optional.
73669
73670commit c7ebb4bef152b47880abb6ea4819416e89e6522c
73671Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
73672Date: Tue Mar 3 14:02:36 2009 -0500
73673
73674 Fix 2 const warnings.
73675
73676commit c7cf926d25523d5d0d8e5c83671c901660070d71
73677Author: Eric Paris <eparis@redhat.com>
73678Date: Tue Mar 3 13:15:39 2009 -0500
73679
73680 This patch changes all places in the X code to use _raw functions. The
73681 X server should never see, translate, or deal with a munged context.
73682 Display managers which show contexts to the user should take care of
73683 translating these to human readable form.
73684
73685commit d84a2f855da9fd293e984084168c1e27f5a5115c
73686Author: Peter Hutterer <peter.hutterer@who-t.net>
73687Date: Sun Mar 1 16:59:14 2009 +1000
73688
73689 xfree86: Add linebreak to two debug statements.
73690
73691 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73692
73693commit e0ed9f16d0ab871f3209837f54055446503c4108
73694Author: Jason Vas Dias <jason.vas.dias@gmail.com>
73695Date: Tue Mar 3 13:24:54 2009 +1000
73696
73697 Fix build with --enable-debug.
73698
73699 [amended by Peter Hutterer]
73700 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73701
73702commit 6093d3eb1d7d1e9cbacc2c2f4d376352f787dd53
73703Author: Peter Hutterer <peter.hutterer@who-t.net>
73704Date: Thu Feb 26 15:38:45 2009 +1000
73705
73706 dix: fix two compiler warnings (old-style function definition).
73707
73708 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73709
73710commit b874a5c0aafc766891c9e567af48b891ea14be4f
73711Author: Peter Hutterer <peter.hutterer@who-t.net>
73712Date: Thu Feb 26 14:40:22 2009 +1000
73713
73714 mi: only print the "EQ overflowing" error once.
73715
73716 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73717
73718commit a706dd8771f87cf0a73578755552124d162baa5d
73719Author: Peter Hutterer <peter.hutterer@who-t.net>
73720Date: Thu Feb 26 10:04:14 2009 +1000
73721
73722 include: add a range of button labels.
73723
73724 Mostly the same buttons as defined by linux/input.h, with five exceptions:
73725 "Button Unknown" for a button that cannot be labelled.
73726 "Button Wheel Up", "Button Wheel Down" for buttons 4/5, traditionally the
73727 wheel buttons.
73728 "Button Horiz Wheel Up", "Button Horiz Wheel Down" for buttons 6/7,
73729 traditionally the horiz. wheel buttons.
73730
73731 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73732 Acked-by: Daniel Stone <daniel@fooishbar.org>
73733
73734commit 8a6ed44a8b2fc5f14729dc54fec17607ced03859
73735Author: Adam Jackson <ajax@redhat.com>
73736Date: Mon Mar 2 14:26:23 2009 -0500
73737
73738 randr: Fix thinko in xf86TargetPreferred
73739
73740 The only-one-output case would only work right if that also happened to
73741 be the zeroth output. Oops.
73742
73743commit 7bfad37f7706da5a76174c0da78dce4d6c6a585b
73744Author: Maarten Maathuis <madman2003@gmail.com>
73745Date: Fri Feb 27 22:26:52 2009 +0100
73746
73747 exa: fixup aux indices and ensure that the indices are used as they should be.
73748
73749 - In a previous patch i forgot to add a FALSE somewhere it seems.
73750 - Rename AUX indices so the driver (think of driver managed pixmaps) can do optimisations based upon them.
73751 - Fix one abuse of DEST index now that we have the AUX indices (same reason as above).
73752
73753commit 8c6b4a827624aa70aea94b449fee6d5df8504105
73754Author: Adam Jackson <ajax@redhat.com>
73755Date: Fri Feb 27 12:42:43 2009 -0500
73756
73757 DDC: Redo extended device probe slightly.
73758
73759 We'll now only mention the E-EDID segment register if the device is
73760 actually E-EDID-capable. While we're here, check for DDC/CI and
73761 standard EEPROM support too.
73762
73763commit b030f858f2f1ce1fd27a73ebf7f9ec5db541a668
73764Author: Adam Jackson <ajax@redhat.com>
73765Date: Sat Feb 21 19:56:20 2009 -0500
73766
73767 selinux: Don't bother relabeling resources that are being destroyed
73768
73769 Makes window destroy about 40x faster in Xvfb.
73770
73771commit f028b14876dc536b575d4b6e1df7f37ee525acec
73772Author: Maarten Maathuis <madman2003@gmail.com>
73773Date: Fri Feb 27 13:06:28 2009 +0100
73774
73775 exa: whitespace
73776
73777commit ed00515ec54b294f304b8ae2857e3073f97860d2
73778Author: Maarten Maathuis <madman2003@gmail.com>
73779Date: Thu Feb 26 22:18:03 2009 +0100
73780
73781 exa: fixup exaAssertNotDirty.
73782
73783 - Do the right thing based on prepare access.
73784
73785commit 2e88b6004f09dbcb888abf05d4d5554231b2bd1a
73786Author: Maarten Maathuis <madman2003@gmail.com>
73787Date: Thu Feb 26 21:14:23 2009 +0100
73788
73789 exa: minor glyphs cleanup.
73790
73791 - This should fix subtle coordinate bugs and make the code a bit cleaner to read.
73792
73793commit 265d20068af5434489752b6dba0bf0065b3cc3ec
73794Author: Michel Dänzer <daenzer@vmware.com>
73795Date: Fri Feb 27 16:41:39 2009 +0100
73796
73797 EXA: Fix check for whether the glyph we're evicting from the cache is in use.
73798
73799 Since commit f07f18231a921d3ae9dd9b75881c9e58e9e2e235 ('EXA: Allow using
73800 exaCompositeRects also when we can't use a mask in exaGlyphs.') we were
73801 checking the wrong set of coordinates in the buffer where glyphs to be rendered
73802 are accumulated when no mask is used in exaGlyphs.
73803
73804 This fixes occasional glyph corruption which can be corrected with redraws, in
73805 particular with Qt4.
73806
73807 Thanks to Maarten Maathuis for asking the right question: 'where do we protect
73808 against evicting glyphs that are still needed?'
73809
73810 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
73811
73812commit 9d5141f7bcfed36417873b47f850332506af83ac
73813Author: Michel Dänzer <daenzer@vmware.com>
73814Date: Fri Feb 27 16:37:28 2009 +0100
73815
73816 EXA: No longer use the driver UploadToScratch hook.
73817
73818 See http://bugs.freedesktop.org/show_bug.cgi?id=20213 .
73819
73820 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
73821
73822commit 4cfb36f6ad2df01215028fec48d99239a0e4496b
73823Author: Michel Dänzer <daenzer@vmware.com>
73824Date: Fri Feb 27 16:37:28 2009 +0100
73825
73826 EXA: Handle separate alpha maps properly in Composite fallback, take two.
73827
73828 Preserve the EXA ABI by introducing a new driver flag EXA_SUPPORTS_PREPARE_AUX.
73829 If the driver doesn't set this flag, we have to assume any Prepare/FinishAccess
73830 driver hooks can't handle the EXA_PREPARE_AUX* indices, so we move out such
73831 pixmaps at PrepareAccess time.
73832
73833 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=18710 .
73834
73835 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
73836
73837commit 4bf707f01822abe99286909fd561da7e7a4211d6
73838Author: Michel Dänzer <daenzer@vmware.com>
73839Date: Fri Feb 27 16:37:27 2009 +0100
73840
73841 Revert "Fix up Xephyr build for recent EXA changes."
73842
73843 This reverts commit 4e8d98b61e1f763c187e7994c683b543cca1a33c.
73844
73845commit de022f8e634baf5a7acd186934b370647df58355
73846Author: Michel Dänzer <daenzer@vmware.com>
73847Date: Fri Feb 27 16:37:27 2009 +0100
73848
73849 Revert "EXA: Handle separate alpha maps properly in Composite fallback."
73850
73851 This reverts commit 170cf1270dff38d3cce7f5ba5b940d1c0d70eff5.
73852
73853 Conflicts:
73854
73855 exa/exa_render.c
73856
73857commit d0dd649035fc3698c5b436f9d9d248116aa106a3
73858Author: Jon TURNEY <jon.turney@dronecode.org.uk>
73859Date: Thu Feb 12 16:03:16 2009 +0000
73860
73861 Remove references to rgb.txt from files section of Xserver and Xorg man pages
73862
73863 The references to this file in the server code were removed in commit dda10c9066a660b647384179f82e1da8e063264f
73864
73865 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
73866
73867commit b0ad9e1ced9619f37acf77764c395c57b86cf463
73868Author: Jon TURNEY <jon.turney@dronecode.org.uk>
73869Date: Thu Feb 12 16:02:17 2009 +0000
73870
73871 Remove long-gone '-co' option from Xserver man page
73872
73873 The '-co' option was removed in commit dda10c9066a660b647384179f82e1da8e063264f
73874
73875 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
73876
73877commit 7907b37d96dc81f333351a6fedd6a9f4e31ae66a
73878Author: Jeremy Huddleston <jeremy@yuffie.local>
73879Date: Thu Feb 26 21:22:48 2009 -0800
73880
73881 XQuartz: Don't erase the mode_switch+keysym if it's the same as shift+keysym
73882 (cherry picked from commit 3e2427e6e9675dd655c8ad328b82e12191b064f2)
73883
73884commit f07f18231a921d3ae9dd9b75881c9e58e9e2e235
73885Author: Michel Dänzer <daenzer@vmware.com>
73886Date: Thu Feb 26 10:35:44 2009 +0100
73887
73888 EXA: Allow using exaCompositeRects also when we can't use a mask in exaGlyphs.
73889
73890 This should give the full benefits of the glyph cache even when we can't use a
73891 mask.
73892
73893 This also means we no longer need to scan the glyphs to see if they overlap,
73894 we can just use a mask or not as the client asks.
73895
73896 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
73897
73898commit b26c136ee9bf7015c583136af53d0c9e9da67ace
73899Author: Michel Dänzer <daenzer@vmware.com>
73900Date: Thu Feb 26 10:35:44 2009 +0100
73901
73902 EXA: Stop tracking damage for pixmaps subject to ModifyPixmapHeader.
73903
73904 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
73905
73906commit 812e8d681a49b33d8748983843799ef692b8b148
73907Author: Keith Packard <keithp@keithp.com>
73908Date: Wed Feb 25 11:08:01 2009 -0800
73909
73910 Oops. 62fc98c had crtc offsets applied backwards (fix from server-1.6-branch)
73911
73912 Crtc offsets are in screen space, not crtc space, and hence should be
73913 applied relative to the screen->crtc transform, not the crtc->transform.
73914
73915 This fix was found in the 'cherry pick' of 62fc98c on server-1.6-branch,
73916 clearly some new definition of 'cherry pick' that I am unaware of.
73917
73918 Signed-off-by: Keith Packard <keithp@keithp.com>
73919
73920commit c090f5514d28e1602a6ebbe7c909e98a0e3374d7
73921Author: Keith Packard <keithp@keithp.com>
73922Date: Wed Feb 25 10:14:36 2009 -0800
73923
73924 Pre-clip panning coordinates to keep crtc within panning region
73925
73926 There is a separate panning region check, but that doesn't work under
73927 transformation, so just pre-clip the mouse coordinates when computing the
73928 panning offsets. This leaves the case where panning constants are changing
73929 unresolved.
73930
73931 Signed-off-by: Keith Packard <keithp@keithp.com>
73932
73933commit 77c7a64e8885696665556c9fbcb3cffb552e367a
73934Author: Keith Packard <keithp@keithp.com>
73935Date: Wed Feb 25 10:12:23 2009 -0800
73936
73937 RandR rotations and reflections offset by one pixel
73938
73939 The matrix computation for rotation and reflection resulted in dropping a
73940 row or column of pixels as the offsets used in the matrix computations used
73941 width and height rather than width-1 and height-1.
73942
73943 Signed-off-by: Keith Packard <keithp@keithp.com>
73944
73945commit 4e8d98b61e1f763c187e7994c683b543cca1a33c
73946Author: Michel Dänzer <daenzer@vmware.com>
73947Date: Wed Feb 25 10:26:04 2009 +0100
73948
73949 Fix up Xephyr build for recent EXA changes.
73950
73951 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
73952
73953commit d6976e1436c035b8e4a6d1e10971b083b1b4184b
73954Author: Jeremy Huddleston <jeremy@yuffie.local>
73955Date: Tue Feb 24 23:12:57 2009 -0800
73956
73957 XQuartz: Re-enable support for capslock
73958 (cherry picked from commit 4901b8147e593d26d7a31a9b73a201254b948916)
73959
73960commit 4b63f88b0ad5f6614fee5b7af4f97ce4581ab935
73961Author: Jeremy Huddleston <jeremy@yuffie.local>
73962Date: Tue Feb 24 18:54:16 2009 -0800
73963
73964 XQuartz: Comment out the background pointer interaction that seems to be causing CPU spinning on some configurations
73965 (cherry picked from commit 1d479fa4f3bca9cbd81808a6c87582422fd2a257)
73966
73967commit 5672e2d66444be74125c71b81373b82f8e1b7dd2
73968Author: Peter Hutterer <peter.hutterer@who-t.net>
73969Date: Fri Feb 20 14:55:51 2009 +1000
73970
73971 Xi: remove AllExtensionVersions.
73972
73973 Keeping an AllExtensionVersions array to save all versions of
73974 all extension is rather pointless if only one extension uses it.
73975 Rename to XIVersion, reduce to a single struct.
73976
73977 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73978
73979commit 59bc615ed2e60c595887dcc11fe04935b49bcaff
73980Author: Peter Hutterer <peter.hutterer@who-t.net>
73981Date: Tue Feb 24 15:26:20 2009 +1000
73982
73983 dix: set the valuator mode in set_valuators
73984
73985 We only have per-device mode for now anyway.
73986
73987 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73988
73989commit cffa3c499a0495e132557bd6b08a8c0b858cdf3b
73990Author: Peter Hutterer <peter.hutterer@who-t.net>
73991Date: Tue Feb 24 15:25:49 2009 +1000
73992
73993 dix: remove unused variable 'tmp'
73994
73995 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
73996
73997commit 8d6dd83a38020080808705a156d32ec2a135633c
73998Author: Peter Hutterer <peter.hutterer@who-t.net>
73999Date: Fri Feb 20 13:13:58 2009 +1000
74000
74001 include: fix indentation for lastSlave/master.
74002
74003 Yeah!
74004
74005 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74006
74007commit b3412adca8b2ba3fff348bceb4f01b9e968c0afc
74008Author: Peter Hutterer <peter.hutterer@who-t.net>
74009Date: Sun Feb 22 20:53:46 2009 +1000
74010
74011 Xi: don't need to set the XKB settings for new core devices.
74012
74013 This is done by the XKB code these days anyway, so we might as well ignore it
74014 and keep using the stanard stuff.
74015
74016 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74017
74018commit a180c2367974abe2ccdc88bb5c6f138c7df4d2e9
74019Author: Peter Hutterer <peter.hutterer@who-t.net>
74020Date: Mon Feb 23 16:38:41 2009 +1000
74021
74022 dix: remove ConvertBackToXI.
74023
74024 This was a temporary solution for the internal-events rework and is not needed
74025 anymore.
74026
74027 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74028
74029commit 26cf709e712573548990d165e71137f28271a8bb
74030Author: Peter Hutterer <peter.hutterer@who-t.net>
74031Date: Mon Feb 23 22:39:08 2009 +1000
74032
74033 Doxygenify events.h and eventconvert.c
74034
74035 Should have done that before pushing, but oh well.
74036
74037 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74038
74039commit 170cf1270dff38d3cce7f5ba5b940d1c0d70eff5
74040Author: Michel Dänzer <daenzer@vmware.com>
74041Date: Tue Feb 24 09:22:09 2009 +0100
74042
74043 EXA: Handle separate alpha maps properly in Composite fallback.
74044
74045 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=18710 .
74046
74047 As this can't work without new EXA_PREPARE_AUX* indices, this requires a major
74048 version bump, so we can also drop the UploadToScratch driver hook and
74049 ExaOffscreenSwap*(). So this also fixes
74050 http://bugs.freedesktop.org/show_bug.cgi?id=20213 .
74051
74052 Moreover, introduce EXA_DRIVER_KNOWN_MAJOR to break compilation of drivers
74053 which may not be able to handle EXA_PREPARE_AUX*, giving instructions how to
74054 make them build again in the #error message.
74055
74056 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
74057
74058commit d3b355875ac46104a174966e1974e6af99e40fd6
74059Author: Peter Hutterer <peter.hutterer@who-t.net>
74060Date: Tue Feb 24 11:14:14 2009 +1000
74061
74062 dix: fix wrong condition for setting valuators on the event.
74063
74064 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74065
74066commit ef320bdd5ec3419abba77041d3a4d96a3ff87563
74067Author: Michel Dänzer <daenzer@vmware.com>
74068Date: Mon Feb 23 16:49:51 2009 +0100
74069
74070 DRI1: Make DRICreateDrawable return TRUE for pixmaps.
74071
74072 GLX_EXT_texture_from_pixmap was broken since commit
74073 a26c77ff432d2e85a2665fc36fca25143460c476 ('glx: fix retval checks when failures
74074 occur for drawable creation.')
74075
74076 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
74077
74078commit 1df6716281579e2937743d840ab1079343c503ac
74079Author: Adam Tkac <vonsch@gmail.com>
74080Date: Mon Feb 23 09:28:26 2009 +1000
74081
74082 xfree86: don't try to UnInit virtual devices in DIDR. (#20087)
74083
74084 DeleteInputDeviceRequest function doesn't handle "virtual" devices well.
74085 TightVNC libvnc.so module to X (which makes bare Xorg VNC capable) uses such
74086 kind of devices.
74087
74088 Bare Xvnc (it is something like Xvfb) simply uses AddInputDevice &
74089 RegisterDevice functions. Xvnc uses DeleteInputDeviceRequest from Xi/stubs.c
74090 so everything works fine (now I see that DeleteInputDeviceRequest in
74091 Xi/stubs.c should call RemoveDevice function, shouldn't it? :) )
74092
74093 Situation is quite different when you use libvnc.so module. It uses same
74094 schema as Xvnc, so it simply calls AddInputDevice & RegisterDevice. Thus
74095 device is created correctly. When server is terminated it calls
74096 DeleteInputDeviceRequest (now from hw/xfree86/common/xf86Xinput.c) for each
74097 device. Here is the difference - Xvnc calls DeleteInputDeviceRequest from
74098 Xi/stubs.c as I wrote above. Thus Xorg gets sigsegv because "VNC" devices
74099 don't have real input driver.
74100
74101 X.Org Bug 20087 <http://bugs.freedesktop.org/show_bug.cgi?id=20087>
74102
74103 [This isn't really a fix (libVNC should behave correctly) but not crashing the
74104 server sounds like an improvement.]
74105 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74106
74107commit a50c40be1f28467bcef7dc71fab54ebdddacffdf
74108Author: Peter Hutterer <peter.hutterer@who-t.net>
74109Date: Sat Feb 21 08:09:40 2009 +1000
74110
74111 dix: don't apply button mappings to SDs, or the device's MD. (#20122)
74112
74113 Only ever change the button map on the device we actually care about, not the
74114 attached SDs, not the current MD of the device.
74115
74116 X.Org Bug 20122 <http://bugs.freedesktop.org/show_bug.cgi?id=20122>
74117
74118 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74119 Acked-by: Daniel Stone <daniel@fooishbar.org>
74120
74121commit 36583a49965c0bb40a84284939b1539b3cb9fc9c
74122Author: Peter Hutterer <peter.hutterer@who-t.net>
74123Date: Tue Feb 10 12:45:49 2009 +1000
74124
74125 mi: split EQ popping and event processing into two functions.
74126
74127 mieqProcessInputEvents() - pop an event off the EQ and pass it to
74128 mieqProcessDeviceEvent() - process the event according to the MD/SD hierarchy.
74129
74130 This way, we can use mieqPDE() from Xtest, xkb, and others to post an event.
74131
74132 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74133
74134commit 47f136ed6fd80310f715a2555501d1b271dd084c
74135Author: Peter Hutterer <peter.hutterer@who-t.net>
74136Date: Fri Feb 6 12:08:43 2009 +1000
74137
74138 mi: change custom handlers to internal events
74139
74140 This should re-enable DGA, but XQuartz needs to be changed to internal events
74141 too now.
74142
74143 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74144
74145commit bdc262701a37a0c12ead810d63fa99a26cbb82ec
74146Author: Peter Hutterer <peter.hutterer@who-t.net>
74147Date: Mon Feb 2 22:39:37 2009 +1000
74148
74149 dix: remove un-used getValuatorEvents and countValuatorEvents from getevents.c
74150
74151 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74152
74153commit 763848d3ab84b4e33a0254120c44b5a1877c819d
74154Author: Peter Hutterer <peter.hutterer@who-t.net>
74155Date: Mon Feb 2 17:20:17 2009 +1000
74156
74157 Input: change processing API to InternalEvents.
74158
74159 Don't pass xEvent* and count through to processing, pass a single
74160 InternalEvent.
74161
74162 Custom handlers are disabled for the time being. And for extra fun,
74163 XKB's pointer motion emulation is disabled. But stick an error in there so
74164 that we get reminded should we forget about it.
74165
74166 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74167
74168commit 0b4066c116e07918a13dc6b4159df7ac9eb92b4b
74169Author: Peter Hutterer <peter.hutterer@who-t.net>
74170Date: Mon Feb 2 17:18:16 2009 +1000
74171
74172 xkb: _XkbFilterRedirectKey needs to pass InternalEvents down.
74173
74174 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74175
74176commit 085d50360863ccc8280cd3eccea2bcb4f3dd9a14
74177Author: Peter Hutterer <peter.hutterer@who-t.net>
74178Date: Mon Feb 2 16:56:52 2009 +1000
74179
74180 dix: Fix PostSyntheticMotion to use a DeviceEvent for posting.
74181
74182 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74183
74184commit 155986a93dc78d6aa060ca3038d5fafa3d8753c2
74185Author: Peter Hutterer <peter.hutterer@who-t.net>
74186Date: Mon Feb 2 16:06:15 2009 +1000
74187
74188 dix: CheckPassiveGrabsOnWindow moved to internal events.
74189
74190 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74191
74192commit 1c38abd4b136301008ba77d4e68aea34508f4c22
74193Author: Peter Hutterer <peter.hutterer@who-t.net>
74194Date: Mon Feb 2 14:54:34 2009 +1000
74195
74196 dix: switch DeliverGrabbedEvent to internal events.
74197
74198 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74199
74200commit fb858774b80eb75c2f8e81fe893bbbdd37065fbd
74201Author: Peter Hutterer <peter.hutterer@who-t.net>
74202Date: Mon Feb 2 14:44:13 2009 +1000
74203
74204 dix: switch DeliverDeviceEvents to internal events.
74205
74206 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74207
74208commit 3aa9404fcd161e94a80b057a77ef47afe428a56d
74209Author: Peter Hutterer <peter.hutterer@who-t.net>
74210Date: Mon Feb 2 14:00:41 2009 +1000
74211
74212 dix: switch DeliverFocusedEvent to internal events.
74213
74214 And because of xfree() macro hilarity, rename "pointer" to "ptr". Oh, how we
74215 laughed.
74216
74217 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74218
74219commit 8c873e7f514844d1056f2b20e653f1dd75f4c327
74220Author: Peter Hutterer <peter.hutterer@who-t.net>
74221Date: Sun Feb 1 20:41:16 2009 +1000
74222
74223 dix: Switch DeliverGrabbedEvents to use internal events.
74224
74225 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74226
74227commit a3718536989fa7d3358e0b2d859c25fde0a2d93e
74228Author: Peter Hutterer <peter.hutterer@who-t.net>
74229Date: Sun Feb 1 20:18:33 2009 +1000
74230
74231 dix: Deliver{Grabbed|Focused|Device}Events API changed to InternalEvents.
74232
74233 With the API change, we can now purge the XI conversion from POE.
74234
74235 Note: this commit breaks DGA even more.
74236
74237 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74238
74239commit 8f94ec6f788565474931cc7d5e3d4672f0f31670
74240Author: Peter Hutterer <peter.hutterer@who-t.net>
74241Date: Sun Feb 1 09:58:15 2009 +1000
74242
74243 dix: convert passive grabs to use internal events.
74244
74245 deviceGrab.sync.event is now an internal event, and CheckDeviceGrabs and
74246 friends is changed over.
74247
74248 Note that this currently breaks some frozen grabs. See towards the end of
74249 ComputeFreezes().
74250
74251 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74252
74253commit daa3245c479b19d445a070b5b76ee005915b5335
74254Author: Peter Hutterer <peter.hutterer@who-t.net>
74255Date: Sat Jan 31 08:46:43 2009 +1000
74256
74257 dix: fix EnqueueEvent to work with internal events.
74258
74259 Note that we're only partially switched to internal events. The event in the
74260 devices' event queue (dev->deviceGrab.sync.event) is still an XI event. The
74261 events in syncEvents are InternalEvents only now.
74262 This also implies fixing CheckVirtualMotion to work with internal events.
74263
74264 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74265
74266commit 75595ba4aa9c3823ffe3b3388ce088929824816f
74267Author: Peter Hutterer <peter.hutterer@who-t.net>
74268Date: Fri Jan 30 15:14:46 2009 +1000
74269
74270 Xi: make ProcessOtherEvents more InternalEvent aware.
74271
74272 Get rid of the deviceValuator processing and a few other things, but still
74273 drop back into XI before checking device grabs or doing anything else.
74274
74275 NoticeEventTime now needs to take InternalEvents, and while we're at it,
74276 change NoticeTime from a macro to a function.
74277
74278 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74279
74280commit 3a02e538dbdb3cd482e01baeaf2aba2ddb7731df
74281Author: Peter Hutterer <peter.hutterer@who-t.net>
74282Date: Fri Jan 30 14:59:22 2009 +1000
74283
74284 dix: update CheckMotion to deal with DeviceEvents.
74285
74286 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74287
74288commit 8829d966a6bacb05d322b60531c59366b58f4514
74289Author: Peter Hutterer <peter.hutterer@who-t.net>
74290Date: Fri Jan 30 13:33:55 2009 +1000
74291
74292 Xi: support InternalEvents in UpdateDeviceState, parts of POE and EnqueueEvent
74293
74294 Note that this breaks DGA. Life is tough.
74295
74296 EnqueueEvent is a somewhat half-baked solution, we immediately drop back into
74297 XI and store them. But it should in theory work.
74298
74299 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74300
74301 Don't let the dcce be random data.
74302
74303commit 007e93c869325cafcc29d975b356dbb8e7cd2ac1
74304Author: Peter Hutterer <peter.hutterer@who-t.net>
74305Date: Fri Jan 30 14:35:22 2009 +1000
74306
74307 xkb: Switch the xkb event processing path over to InternalEvents.
74308
74309 Before dropping down into the DIX, convert back into XI events. This is a
74310 temporary solution only, until the DIX is capable of handling InternalEvents
74311 anyway.
74312
74313 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74314
74315commit 64ea6078105f73d1b727619fc123920bc7e4a06c
74316Author: Peter Hutterer <peter.hutterer@who-t.net>
74317Date: Mon Feb 2 14:15:52 2009 +1000
74318
74319 dix: change eventconvert to always return an array of xEvents
74320
74321 Just alloc the memory on demand rather than doing things with EventListPtrs
74322 etc.
74323
74324 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74325
74326commit 4026c63e4eb16481bafc1a41ad67cd2556728d40
74327Author: Peter Hutterer <peter.hutterer@who-t.net>
74328Date: Fri Jan 30 14:34:02 2009 +1000
74329
74330 mi: switch the EQ to contain InternalEvents only.
74331
74332 This gets rid of the nevents parameter, InternalEvents are always a single
74333 item per event. Also remove the special DeviceValuator handling in both
74334 enqueueing and dequeueing.
74335
74336 Custom callback handlers are now broken until fixed.
74337
74338 For bisectability, we copy the InternalEvent back into the XI required during
74339 POE and friends. Consider this a temporary solution.
74340
74341 Note: Because of misc linker bonghits, Xvfb won't link in this revision.
74342
74343 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74344
74345commit 5a827593f9517fd5593751dd8bd90c611de06c5d
74346Author: Peter Hutterer <peter.hutterer@who-t.net>
74347Date: Thu Jan 29 15:26:57 2009 +1000
74348
74349 dix: switch event generation to InternalEvents.
74350
74351 GPE, GKVE, GProxE generate InternalEvents now.
74352 DeviceClassesChangedEvents generates an InternalEvent now, but incomplete! We
74353 need to tack on the information about the new SD in the ClassesChanged events.
74354
74355 Note: To make the progress bisectable, we drop back into XI events at the end of the
74356 Get*Events functions. So the rest of the server still uses XI events.
74357
74358 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74359
74360commit 00b03683d0e5cda40fa23b9fe6a83d7227f86f5d
74361Author: Peter Hutterer <peter.hutterer@who-t.net>
74362Date: Fri Jan 30 14:31:08 2009 +1000
74363
74364 dix: Add temporary conversion function ConvertBackToXI.
74365
74366 Until the InternalEvents are used throughout the server, we can use this one
74367 to drop us back into XI la-la land where every event is the wire format.
74368
74369 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74370
74371commit 269d4d9f2e86fde8c6a28ef0293fb0e44b577886
74372Author: Peter Hutterer <peter.hutterer@who-t.net>
74373Date: Sun Feb 1 09:57:27 2009 +1000
74374
74375 dix: add GetCoreType and GetXIType.
74376
74377 Convert from an InternalEvent type to the matching core/XI type. Currently
74378 only for a few events, those we actually need in the server.
74379
74380 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74381
74382commit 656491921e17b2371057041f4551ad6165067551
74383Author: Peter Hutterer <peter.hutterer@who-t.net>
74384Date: Wed Jan 28 15:27:38 2009 +1000
74385
74386 dix: add InternalEvent -> core/xi event conversion routines.
74387
74388 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74389
74390commit e7867d12541ef9683d5d7fc766e918c13a742981
74391Author: Peter Hutterer <peter.hutterer@who-t.net>
74392Date: Fri Jan 23 15:37:23 2009 +1100
74393
74394 include: add XInternalEvent.
74395
74396 This is the event we want to feed into the EQ and process on the way through.
74397 Only applies for input events for now.
74398
74399 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74400
74401commit 10dcf3ed9c543a6811340567e586ec0e476fcf61
74402Author: Jon TURNEY <jon.turney@dronecode.org.uk>
74403Date: Fri Feb 20 15:22:35 2009 +0000
74404
74405 Cygwin/X: Fix keyboard layout mapping for Latin American keyboards
74406
74407 Fix keyboard layout mapping table entry for Windows keyboard layout 0x80a
74408 (Latin American), layout should (now) be 'latam', not 'la' which is Laos
74409
74410 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
74411
74412commit 9531b96470a2fd812eb3fa0d1e10419331842496
74413Author: Jeremy Huddleston <jeremy@yuffie.local>
74414Date: Sun Feb 22 00:41:52 2009 -0800
74415
74416 XQuartz: DefineInitialRootWindow is gone
74417
74418commit c7c50157b5137a147e69abb70b43a28fac4a9ffa
74419Author: Jeremy Huddleston <jeremy@yuffie.local>
74420Date: Sun Feb 22 00:29:48 2009 -0800
74421
74422 XQuartz: quartzKeyboard compile fixes for recent XKB Changes
74423
74424 Seems we're less divorced from XKB than we'd like for now... hopefully not having any rules installed will still let us override things the way we want to...
74425
74426commit 61abf3189f3f97ec5228d45c035dbfd0c249f844
74427Author: George Staplin <gstaplin@apple.com>
74428Date: Sat Feb 21 22:33:10 2009 -0700
74429
74430 XQuartz: GL: Change from xalloc to xcalloc to potentially avoid
74431 uninitialized data in the __GLXAquaContext.
74432 (cherry picked from commit d3120241f9f48d21f9a6ecfb848434a4a0270855)
74433
74434commit 5587f9b771f5b5427a81d4d657f403667d20f310
74435Author: George Staplin <gstaplin@apple.com>
74436Date: Sat Feb 21 22:22:52 2009 -0700
74437
74438 XQuartz: GL: Fix a bug with an uninitialized GLX data structure.
74439
74440 This was causing a crash randomly, due to random memory contents.
74441 Use xcalloc to prevent this in the future, due to future changes or mistakes.
74442
74443 Set the drawableType to include GLX_PIXMAP_BIT and GLX_PBUFFER_BIT.
74444 The new libGL supports these.
74445
74446 Set the max Pbuffer width/height, based on the results of a test program.
74447 We may someday want to revisit this depending on what users need, so that
74448 we create a CGLContextObj, make it current, and call glGetIntegerv to
74449 gather the information at runtime.
74450 (cherry picked from commit c7e338330943e0d03a99328c740540d03f018d20)
74451
74452commit ab61033700b5383a7a15370dd054eaa80e72e811
74453Author: Alan Coopersmith <alan.coopersmith@sun.com>
74454Date: Thu Feb 19 13:57:28 2009 -0800
74455
74456 Add Extensions section to xorg.conf man page
74457
74458 Extensions section was added in X11R6.8.0 and documented in the release notes:
74459 http://www.x.org/archive/X11R6.8.0/doc/RELNOTES2.html#3
74460 but never made it into the man page.
74461
74462 Also fix a bonus typo.
74463
74464 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
74465
74466commit 372977354c3a09ca77ca13ba0aec00a61046f60d
74467Author: George Staplin <gstaplin@apple.com>
74468Date: Wed Feb 18 21:43:19 2009 -0700
74469
74470 XQuartz: Fix the new ProcAppleDRIDestroyPixmap code REQUEST_SIZE_MATCH.
74471
74472 It had a copy and paste mistake that I didn't notice. :/
74473 It was using the CreatePixmapReq.
74474
74475 Also add a missing B16 to the end of the length for the DestroyPixmapReq struct.
74476
74477 Now the AppleDRIDestroyPixmap request seem to work.
74478 (cherry picked from commit 295fe25bd8fa2d141291a9d9b6ef7b75fcccb4dd)
74479
74480commit 225853d51d1fb610261ab0c295b1b5a96ce177d5
74481Author: Dan Nicholson <dbn.lists@gmail.com>
74482Date: Thu Feb 19 06:45:05 2009 -0800
74483
74484 xkb: Use cached XKB keymap when rules haven't changed
74485
74486 Rather than compiling a new keymap every time InitKeyboardDeviceStruct
74487 is called, cache the previous keymap and reuse it if the rules have not
74488 changed.
74489
74490 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
74491 Acked-by: Daniel Stone <daniel@fooishbar.org>
74492 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74493
74494commit 64e595d12e05c4df56b0230cc57924b9beb274d3
74495Author: Peter Hutterer <peter.hutterer@who-t.net>
74496Date: Thu Feb 19 09:18:53 2009 +1000
74497
74498 dix: check if keybd->key is valid before getting the keyboard state.
74499
74500 For floating SDs, GetPairedDevice() returns the SD itself, hence keybd->key
74501 may be NULL.
74502
74503 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74504
74505commit e0a451eb7cc812f11db3614b730ba0f07d6ae18c
74506Author: Alan Coopersmith <alan.coopersmith@sun.com>
74507Date: Tue Feb 17 19:37:22 2009 -0800
74508
74509 Obsolete InputDevices keyword in xorg.conf Files section
74510
74511 Was only used to provide a list of input devices that XF86-Misc could use,
74512 now that XF86-Misc is gone, was parsed and logged, then completely ignored.
74513
74514 (Depends on previous patch that introduces OBSOLETE_TOKEN in parser to
74515 make obsolete keywords like InputDevices & RgbPath be non-fatal errors.)
74516
74517 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
74518 Acked-by: Adam Jackson <ajax@redhat.com>
74519
74520commit d2cf562bbad553d7f09b70202134f5b6ada0114e
74521Author: Alan Coopersmith <alan.coopersmith@sun.com>
74522Date: Tue Feb 17 18:48:52 2009 -0800
74523
74524 Make RgbPath keyword in xorg.conf a non-fatal error
74525
74526 Xorg shouldn't refuse to run just because the user has an xorg.conf that
74527 had the previously-used RgbPath keyword in it.
74528
74529 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
74530 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
74531
74532commit 344b1f2b25d627bdf7d802c5831b6a72d22ffe34
74533Author: Jeremy Huddleston <jeremy@yuffie.local>
74534Date: Tue Feb 17 23:25:14 2009 -0800
74535
74536 configure.ac: Support version strings like W.X.Y.Z-XXXXXX
74537 (cherry picked from commit 43967514cd57ad836d7fb85c8c9e58ada07e0232)
74538
74539commit 67faf41b3f3e0cd66e35e3c176a9adb37ff1a759
74540Author: Jeremy Huddleston <jeremy@yuffie.local>
74541Date: Tue Feb 17 23:24:25 2009 -0800
74542
74543 XQuartz: Don't need GlxSetVisualConfig any more
74544
74545 Fixes build failure resulting from 516f8e2cad1311a09764e2633644188d1e3c31bb
74546 (cherry picked from commit 066b17028a35956a089815716e38571f305469c5)
74547
74548commit b349a764e98f0d8f221190157ffa0904b91beca5
74549Author: Eric Anholt <eric@anholt.net>
74550Date: Tue Feb 17 09:59:17 2009 -0800
74551
74552 xinerama: Put the proto version in the code instead using proto headers.
74553
74554 Proto headers updating resulting in the server advertising new versions is
74555 broken. This should be applied to every extension.
74556
74557 This fixes the build against slightly-older xineramaproto.
74558
74559commit 5394b7e66224d20888dd4020f5cb8ca930720fb4
74560Author: Keith Packard <keithp@keithp.com>
74561Date: Fri Feb 6 06:34:35 2009 +0100
74562
74563 Eliminate the shadow clear on transform change
74564
74565 When the crtc transformation changes, the entire crtc must be repainted.
74566 This was being done by clearing the shadow and then painting the rectangle
74567 containing the screen image; the clear being required as the screen image
74568 may not fill the crtc. When changing the transform rapidly, this leads to
74569 flashing. Eliminate the clear by painting the entire crtc instead of just
74570 the screen rectangle.
74571
74572 Signed-off-by: Keith Packard <keithp@keithp.com>
74573
74574commit 76f18b94bd2719a8199334742d021c4d0806187d
74575Author: Keith Packard <keithp@keithp.com>
74576Date: Mon Feb 9 13:09:11 2009 +0100
74577
74578 Add XkbDir to Files config file section
74579
74580 The XKB base directory was not configuable through the config file.
74581
74582 Signed-off-by: Keith Packard <keithp@keithp.com>
74583
74584commit 123093996507c4d3b6dc457240ce00f8ac42f410
74585Author: Adam Jackson <ajax@redhat.com>
74586Date: Tue Feb 17 12:59:16 2009 -0500
74587
74588 RANDR: Fail softly on GetPanning if the screen can't do it.
74589
74590 Just return a zeroed-out reply in that case. This is unambiguous, and
74591 distinguishes "you didn't name a CRTC" from "you named a CRTC that can't
74592 do panning".
74593
74594commit b2ceea3635ec05dca9d4aa2f823b96ae9fce7fe8
74595Author: Maarten Maathuis <madman2003@gmail.com>
74596Date: Tue Feb 17 19:40:59 2009 +0100
74597
74598 Revert "exa: the extent of the valid region is probably much larger than that of the pending damage."
74599
74600 This reverts commit 97c1cbc70216366e92b9371de608ce94e60aa874.
74601
74602 - Sorry for the thinko, pending damage is often not fragmentated.
74603 - Should the dst region become fragmentated, you actually want to copy more to unfragmentate it.
74604
74605commit 3175646b10c602d17d5dd37bdace7c1c7ee92b3d
74606Author: Maarten Maathuis <madman2003@gmail.com>
74607Date: Tue Feb 17 19:06:35 2009 +0100
74608
74609 exa: exaPixmapDirty should use official damage functions.
74610
74611 - Otherwise other users will not be notified of damage.
74612
74613commit 97c1cbc70216366e92b9371de608ce94e60aa874
74614Author: Maarten Maathuis <madman2003@gmail.com>
74615Date: Tue Feb 17 19:01:44 2009 +0100
74616
74617 exa: the extent of the valid region is probably much larger than that of the pending damage.
74618
74619 - Since the default has been changed from nothing to everything.
74620
74621commit 4039603413f9f46d7f725463a70b4a51838e0049
74622Author: Ian Romanick <ian.d.romanick@intel.com>
74623Date: Tue Feb 17 08:27:32 2009 -0800
74624
74625 glx: Inialize best_score before calculating visual scores
74626
74627 This bug was pointed out by Peter Åstrand.
74628
74629 Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
74630
74631commit bac13e54b164ceff7b921a2d17c4a198b42aeccf
74632Author: George Staplin <gstaplin@apple.com>
74633Date: Mon Feb 16 17:33:19 2009 -0700
74634
74635 XQuartz: Add driWrap.h that I missed in the last commit.
74636
74637 This should complete the GLXPixmap commits for now.
74638 (cherry picked from commit 7e99b55651bf7f6275bb227d02f203fb885a67a2)
74639
74640commit 95052caa2c8a79981cdc7936888a8a47e09415e0
74641Author: George Staplin <gstaplin@apple.com>
74642Date: Mon Feb 16 17:22:18 2009 -0700
74643
74644 XQuartz: Add support for GLXPixmaps to the AppleDRI.
74645
74646 This involved wrapping some GCOps to get the proper behavior
74647 when using X11 raster ops mixed with OpenGL (see driWrap.c).
74648
74649 This extends the AppleDRI protocol with create and destroy pixmap
74650 functions.
74651
74652 The dri.c code has been extended quite a bit to enable this, and
74653 to initialize the wrapping of CreateGC for GCOps.
74654
74655 This has been tested with tests/glxpixmap and proven to work with
74656 the new libGL. Existing applications seem to work fine too. Redraws
74657 all appear to be correct.
74658
74659 There may be some bugs lurking that I haven't found yet. I plan
74660 to drive them out by extending the libGL test suite.
74661 (cherry picked from commit 630518766b01022c49fe3a9e7d501808f71b06e2)
74662
74663commit 970725d23e83285d41e3f7cab678a825d9cc25cf
74664Author: Adam Jackson <ajax@redhat.com>
74665Date: Mon Feb 16 18:44:00 2009 -0500
74666
74667 kdrive: Nuke dead AGP and VGA code. (#19921)
74668
74669commit a932744d983845f81791045cf0dfb6a359790158
74670Author: Chris Ball <cjb@laptop.org>
74671Date: Mon Feb 16 18:27:26 2009 -0500
74672
74673 Build fix, remove export of pciBusAddrToHostAddr
74674
74675 a9d7d659.. (PCI: Remove pciBusAddrToHostAddr and associated nonsense)
74676 removes pciBusAddrToHostAddr(), but not its prototype, resulting in:
74677
74678 ./.libs/libxorg.a(sdksyms.o):(.data.rel+0xe64): undefined reference to
74679 `pciBusAddrToHostAddr'
74680
74681 Signed-off-by: Chris Ball <cjb@laptop.org>
74682
74683commit ab331aa76f0bc6a85a519b2deaafa4ce8a070c53
74684Author: Julien Cristau <jcristau@debian.org>
74685Date: Mon Feb 16 21:56:03 2009 +0100
74686
74687 xfree86: ModeDebug is a boolean, not a string
74688
74689 Signed-off-by: Julien Cristau <jcristau@debian.org>
74690
74691commit f5bf1fdaf36163d5c2f1b9b51df96326ebbb0e9c
74692Author: Peter Hutterer <peter.hutterer@who-t.net>
74693Date: Fri Feb 13 09:56:22 2009 +1000
74694
74695 xkb: Fix wrong colour reference in XKB geometry copying. #20081
74696
74697 base_color and label_color need to reference the color in the destination, not
74698 in the source.
74699
74700 X.Org Bug 20081 <http://bugs.freedesktop.org/show_bug.cgi?id=20081>
74701
74702 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74703 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
74704
74705commit a9d7d659a05e4b6b25e6acd2ce59926911aac0f7
74706Author: Adam Jackson <ajax@redhat.com>
74707Date: Fri Feb 13 18:36:27 2009 -0500
74708
74709 PCI: Remove pciBusAddrToHostAddr and associated nonsense
74710
74711 This was all a glorified no-op. We rely on pciaccess to create device
74712 maps anyway, so we should have no reason to care about what the host
74713 address is.
74714
74715 Acked-by: Ian Romanick <ian.d.romanick at intel.com>
74716 Signed-off-by: Adam Jackson <ajax@redhat.com>
74717
74718commit be8c3d4022faa7d169778ea10d1f11e8f90e1c0c
74719Author: Adam Jackson <ajax@redhat.com>
74720Date: Fri Feb 13 17:54:33 2009 -0500
74721
74722 Linux: Remove dead USE_DEV_FB #ifdefs
74723
74724commit 51ae4d6bbfd95cac8a9cdefcd6f22fbe21a6cb9d
74725Author: Ian Romanick <ian.d.romanick@intel.com>
74726Date: Mon Feb 16 12:08:28 2009 -0800
74727
74728 glx: Add comments around some extension string weirdness
74729
74730commit caa5310e163b040689383f6d25ccab5ef1b2e215
74731Author: Maarten Maathuis <madman2003@gmail.com>
74732Date: Mon Feb 16 20:41:14 2009 +0100
74733
74734 exa: reintroduce src rect optimisation, with a slightly higher threshold.
74735
74736 - I got some feedback on gtkperf line test regression.
74737 - The increased threshold should ensure the destination optimisation is dominant.
74738
74739commit 46eeaf82e228df1c5971a24dc815566516e19be9
74740Author: Maarten Maathuis <madman2003@gmail.com>
74741Date: Mon Feb 16 17:17:14 2009 +0100
74742
74743 exa: fix performance regression from 736b6fbd2c941b6276066cd1503523edebe7bf3d
74744
74745 - The src optimisation is more aggressive and possibly harmful in light of the new initial state of pixmaps.
74746 - There is now actually a performance improvement by almost always keeping the number of rects low.
74747
74748commit 6198373ff2a8e36113bb1dcaebe975530ff01b86
74749Author: David Nusinow <dnusinow@debian.org>
74750Date: Sun Feb 15 18:36:01 2009 -0500
74751
74752 xorg.conf (5) refer to mousedrv (4). Debian #394058
74753
74754 Signed-off-by: Julien Cristau <jcristau@debian.org>
74755
74756commit 2a8b8077d8f6001eb57deba60e1009fc99c28668
74757Author: Alan Hourihane <alanh@vmware.com>
74758Date: Mon Feb 16 11:39:34 2009 +0000
74759
74760 dri2: support glXWaitGL & glXWaitX by copying fake front to front and
74761 vice-versa.
74762
74763commit ddb8d8945d1f44d16adc366b6612eef20ae813f7
74764Author: Peter Åstrand <astrand@cendio.se>
74765Date: Fri Feb 13 10:23:28 2009 +0100
74766
74767 xserver: Avoid sending uninitialized padding data over the network
74768
74769 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74770
74771commit b735a4b4951b607e614682836f24d5fd86c1f7fb
74772Author: Peter Hutterer <peter.hutterer@who-t.net>
74773Date: Thu Feb 12 14:29:23 2009 +1000
74774
74775 dix: Remove lastEventMask variable, substitute with a #define.
74776
74777 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74778
74779commit 772e0f9159ca3ab8fb4e03ed6feee3cc93697724
74780Author: Peter Hutterer <peter.hutterer@who-t.net>
74781Date: Thu Feb 12 13:38:34 2009 +1000
74782
74783 dix: Don't set core events in SetMaskForEvent.
74784
74785 Rather, modify the two callers to call separately for the two different.
74786 events. Unexport SetMaskForEvent too.
74787 And while we're at it, get rid of the MotionFilter macro, because it's one
74788 half confusing and one half pointless.
74789
74790 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74791
74792commit 58f3127919ba5fcb3bb467b6913c28ee1127c82b
74793Author: Peter Hutterer <peter.hutterer@who-t.net>
74794Date: Thu Feb 12 08:37:49 2009 +1000
74795
74796 Xi: remove dynamic mask assignment for XI events.
74797
74798 They end up being the same anyway on startup, so let's not have a dynamic mask
74799 assignment mechanism and instead just hardcode them already.
74800 Also unexport SelectForWindow and remove the valid_masks parameter. We can
74801 check that before calling, since there's only one caller anyway.
74802
74803 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74804
74805commit 7057a9a97b9316b1de954d05451e04223fdd9fd8
74806Author: Peter Hutterer <peter.hutterer@who-t.net>
74807Date: Thu Feb 12 22:14:48 2009 +1000
74808
74809 Remove two more define XKB leftovers.
74810
74811 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74812
74813commit 736b6fbd2c941b6276066cd1503523edebe7bf3d
74814Author: Michel Dänzer <daenzer@vmware.com>
74815Date: Sun Feb 15 17:44:51 2009 +0100
74816
74817 EXA: Try to prevent the valid regions from growing too many rects.
74818
74819 This helps contain region management overhead in some pathological cases, see
74820 e.g. http://bugs.freedesktop.org/show_bug.cgi?id=16647 .
74821
74822 Signed-off-by: Michel Dänzer <daenzer@vmware.com>
74823
74824commit 45e14f515337159e600a47ebf349692751552af8
74825Author: Chris Ball <cjb@laptop.org>
74826Date: Fri Feb 13 20:32:02 2009 -0500
74827
74828 Build fix for xf86EdidModes.c
74829
74830 Commit 20ac314.. forgets to declare hsize/vsize/refresh.
74831
74832commit ae68be312619ad3bcb2d550a6267d38cce3284f5
74833Author: Adam Jackson <ajax@redhat.com>
74834Date: Fri Feb 13 17:25:32 2009 -0500
74835
74836 Input: Don't re-block SIGIO in xf86SigioReadInput()
74837
74838 It's already blocked on the way in, because we asked for it to be
74839 blocked at sigaction() time.
74840
74841commit c18f454e4a9cab03e5d55b6d761a4a735c943b6b
74842Author: Aaron Plattner <aplattner@nvidia.com>
74843Date: Thu Feb 12 18:27:48 2009 -0800
74844
74845 Query the DDX in ProcRRGetScreenInfo.
74846
74847 This is the RandR 1.1 version of GetScreenResources and needs to re-query the
74848 DDX to see if the mode pool changed.
74849
74850 Fixes Launchpad bug #325115.
74851 Signed-off-by: Adam Jackson <ajax@redhat.com>
74852 (cherry picked from commit 660c2a7d4c6cb52fd0992e9d2008a24805dc8c02)
74853
74854commit 20ac3140ce16af460992b83aa8aeff4afe28c2ee
74855Author: Adam Jackson <ajax@redhat.com>
74856Date: Mon Feb 9 13:36:05 2009 -0500
74857
74858 EDID: Hack for 1366x768 in standard timing descriptors
74859
74860 All you get for standard timing descriptors is horizontal size in
74861 multiples of 8 pixels (which means you can't say 1366) and height in
74862 terms of aspect ratio (which means you can't say 768). You'd like to
74863 just fuzzy-match this by walking the DMT list for sufficiently close
74864 modes, but you can't because DMT is useless and only defines a 1360x768
74865 mode, because it's _also_ specified in terms of character cells despite
74866 providing pixel exact timings. Neither can you use CVT or GTF to
74867 generate the timings, because they _also_ believe that modes have to be
74868 a multiple of 8 pixels.
74869
74870 You'd also hope you could find a timing definition for this in CEA, but
74871 you can't because CEA only defines transmission formats that actually
74872 exist. So there's 480p, 720p, and 1080p, but no 768p. And why would
74873 there be, after all, the encoded signal is never 768p so obviously no
74874 one would ever make a display in that format.
74875
74876 So instead, make a CVT mode since that's likely to be handled well by
74877 just about everything, smash the horizontal active down by 2, and shift
74878 the sync pulse by 1. Underscanning the hard way.
74879
74880 Pass the suicide.
74881
74882commit bcafdfbed6f1e1f901c2459f60f94a0da506bd90
74883Author: Adam Jackson <ajax@redhat.com>
74884Date: Fri Feb 13 14:06:07 2009 -0500
74885
74886 RANDR: Validate entire mode list for interlace and doublescan
74887
74888 Otherwise drivers have to refuse interlace twice: once in the output
74889 config, and once in ->valid_mode() to catch output and config modes.
74890 If you can't do interlaced modes, asking nicely for it in the config
74891 isn't going to suddenly make it work.
74892
74893commit a26c77ff432d2e85a2665fc36fca25143460c476
74894Author: Alan Hourihane <alanh@vmware.com>
74895Date: Fri Feb 13 13:45:08 2009 +0000
74896
74897 glx: fix retval checks when failures occur for drawable creation.
74898
74899commit a38ca0063c91ef221f20bf0c4ac0dc84ce07d557
74900Author: Benjamin Close <Benjamin.Close@clearchain.com>
74901Date: Fri Feb 13 14:44:53 2009 +1030
74902
74903 xext: Use proto header rather than the Xext include file, this prevents userspace being pulled in causing issues
74904
74905 Signed-off-by: Benjamin Close <Benjamin.Close@clearchain.com>
74906 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
74907 Acked-by: Daniel Stone <daniel@fooishbar.org>
74908
74909commit d62c085f36892c327785e85571bd2c7c70513ac6
74910Author: Benjamin Close <Benjamin.Close@clearchain.com>
74911Date: Fri Feb 13 14:44:09 2009 +1030
74912
74913 dix: Remove includes which are not actually used directly
74914
74915 Signed-off-by: Benjamin Close <Benjamin.Close@clearchain.com>
74916 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
74917 Acked-by: Daniel Stone <daniel@fooishbar.org>
74918
74919commit 340f1576afcdaf883d185da356e5d6282aa65e19
74920Author: Peter Hutterer <peter.hutterer@who-t.net>
74921Date: Thu Feb 12 10:44:27 2009 +1000
74922
74923 dix: remove DefineInitialRootWindow()
74924
74925 Obsolete.
74926
74927 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74928
74929commit c178c3f814f7666f43cbf47d24c7543b289f9b1a
74930Author: Peter Hutterer <peter.hutterer@who-t.net>
74931Date: Thu Feb 12 15:29:38 2009 +1000
74932
74933 dix: doxygen-ify enterleave.c
74934
74935 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74936
74937commit cc696a2b7928ec497b6a0df3602fc70dec83a629
74938Author: Peter Hutterer <peter.hutterer@who-t.net>
74939Date: Thu Feb 12 15:05:57 2009 +1000
74940
74941 Doxygen-ify bits of events.c
74942
74943commit 8364bf7374aa5f6d991700f0c02921dc6c638c9f
74944Author: Peter Hutterer <peter.hutterer@who-t.net>
74945Date: Wed Feb 11 16:06:36 2009 +1000
74946
74947 Document the event masks.
74948
74949 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74950
74951commit ed9d58c3c25ee1b3dedbc4c116823c263ccf164d
74952Author: Simon Thum <simon.thum@gmx.de>
74953Date: Sun Feb 8 17:21:09 2009 +0100
74954
74955 dix: refactor pointer acceleration
74956
74957 The algorithm is split in a 2D-specific and a general part.
74958 This potentially allows to accelerate more than just screen motion.
74959 A state machine is intoduced to make code more explicit and readable.
74960 It also improves handling of 'phase 1' mickeys when axial correction
74961 kicks in (corner case).
74962
74963 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
74964
74965commit ac470dfb4fadaa0b28b6f8b57f4f13a20842b897
74966Author: Alan Coopersmith <alan.coopersmith@sun.com>
74967Date: Thu Feb 12 18:34:22 2009 -0800
74968
74969 Check for and report errors writing xorg.conf.new from Xorg -configure
74970
74971 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
74972
74973commit bd713794ceaa1b2890522554562103c0a2d50f04
74974Author: Alan Coopersmith <alan.coopersmith@sun.com>
74975Date: Thu Feb 12 18:23:59 2009 -0800
74976
74977 Correct error message if specified config file is not found
74978
74979 By making the "Unable to open config file" header a warning, it was
74980 not appearing with the filename when a config file was specified and
74981 not found. Now we make it an error message again, but only issue
74982 the error if a filename was specified - if none was specified, then
74983 we don't even issue a warning, just the "Using autoconfig" info message.
74984
74985 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
74986
74987commit d229ba7068f0adf1bc8b30b7ccf2fce2c7e8b97b
74988Author: George Staplin <gstaplin@apple.com>
74989Date: Fri Feb 6 12:55:09 2009 -0700
74990
74991 XQuartz: xpr: The dri.c code for pixmaps was wrong in several ways. They weren't
74992 being exported correctly by Xplugin.
74993
74994 This should fix a bug with the surface for a window, when an export fails.
74995 Before the export could fail and leave behind an invalid (freed) pointer in the dix privates.
74996
74997 I have an idea of how to fix the GLXPixmaps now without using CGLSetOffScreen.
74998
74999 This work is a step towards that. The Xplugin will need a small patch to fix an
75000 issue that this change brought forth.
75001 (cherry picked from commit 58c4116c47543b5e30c2232e7bee8efc0b9be176)
75002
75003commit b17d6bed97eddd0773d0845f93f842444b7de0ae
75004Author: Jeremy Huddleston <jeremy@yuffie.local>
75005Date: Thu Feb 12 15:33:47 2009 -0800
75006
75007 XQuartz: Fix copyright info in Info.plist to be prettier when viewed in Finder
75008 (cherry picked from commit 8b607c9c507a68a6abda63831d227d0f8947028c)
75009
75010commit d1ad43a8bc3a93a9cddcb4b0be1d722993bf5dd6
75011Author: Ian Romanick <ian.d.romanick@intel.com>
75012Date: Thu Feb 12 14:05:10 2009 -0800
75013
75014 Enable pbuffers
75015
75016 If a driver does not actually support pbuffer rendering, it can just
75017 not enable any pbuffer fbconfigs.
75018
75019commit 0065896e1ae6d419a42b318b66d6e4d5f17a40ca
75020Author: Colin Harrison <colin.harrison@virgin.net>
75021Date: Sun Nov 2 20:46:37 2008 +0000
75022
75023 Xming: Update keyboard layouts mapping table
75024
75025 An update to the keyboard layout mapping table which updates some no longer valid keyboard layout names
75026 (e.g. de_CH -> ch -variant de) and adds a few more reported layouts
75027
75028 Copyright (C) Colin Harrison 2005-2008
75029 http://www.straightrunning.com/XmingNotes/
75030 http://sourceforge.net/projects/xming/
75031
75032 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75033
75034commit d7c27bfe553235e55eb48799129ff9f8e7e0d4f6
75035Author: Colin Harrison <colin.harrison@virgin.net>
75036Date: Tue Feb 3 15:57:29 2009 +0000
75037
75038 Xming: Check clipboard selection ownership after taking it
75039
75040 When you take ownership of a selection you should always check you got
75041 it (X11 ICCCM gospel).
75042
75043 Copyright (C) Colin Harrison 2005-2008
75044 http://www.straightrunning.com/XmingNotes/
75045 http://sourceforge.net/projects/xming/
75046
75047 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75048
75049commit cb6315072ff23740bdc5fc2f95dd960601df5664
75050Author: Colin Harrison <colin.harrison@virgin.net>
75051Date: Tue Feb 3 15:54:31 2009 +0000
75052
75053 Xming: Fix a memory leak in clipboard integration code
75054
75055 Free the memory allocated for ReturnData in FlushXEvents().
75056
75057 Copyright (C) Colin Harrison 2005-2008
75058 http://www.straightrunning.com/XmingNotes/
75059 http://sourceforge.net/projects/xming/
75060
75061 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75062
75063commit d54d24dd5f25188b951b82b1c95eb9d36b5be970
75064Author: Colin Harrison <colin.harrison@virgin.net>
75065Date: Tue Feb 3 15:53:43 2009 +0000
75066
75067 Xming: Fix prototype for winProcessXEventsTimeout() in clipboard integration code
75068
75069 Function 'winProcessXEventsTimeout()' is declared 'static Bool' and
75070 defined 'static int' (correct) in 'winclipboardwndproc.c'. This has been
75071 wrong since dot, but luckily was of no significance
75072
75073 Copyright (C) Colin Harrison 2005-2008
75074 http://www.straightrunning.com/XmingNotes/
75075 http://sourceforge.net/projects/xming/
75076
75077 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75078
75079commit d8bae80305057acfb8e82341f68090eac0dccb90
75080Author: Colin Harrison <colin.harrison@virgin.net>
75081Date: Tue Feb 3 15:52:57 2009 +0000
75082
75083 Xming: Improve clipboard integration startup in XDMCP mode
75084
75085 Improve clipboard integration startup in XDMCP mode
75086 Stop the "Clipboard not yet started, aborting" message from spamming the log.
75087
75088 Copyright (C) Colin Harrison 2005-2008
75089 http://www.straightrunning.com/XmingNotes/
75090 http://sourceforge.net/projects/xming/
75091
75092 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75093
75094commit 7cb1622ce44754d9dcd6f405fb8403bd79c3e7e4
75095Author: Colin Harrison <colin.harrison@virgin.net>
75096Date: Tue Feb 3 15:51:02 2009 +0000
75097
75098 Xming: Select events correctly in clipboard integration code
75099
75100 The clipboard should only select PropertyChange events and do this on
75101 the correct window, the messaging window for the clipboard not the root
75102 window!
75103
75104 Copyright (C) Colin Harrison 2005-2008
75105 http://www.straightrunning.com/XmingNotes/
75106 http://sourceforge.net/projects/xming/
75107
75108 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75109
75110commit b997b3849d902894dd5b38673b73a23a47edf502
75111Author: Colin Harrison <colin.harrison@virgin.net>
75112Date: Tue Feb 3 15:48:04 2009 +0000
75113
75114 Xming: Add Colin Harrison to copyright and authors list for clipboard integration files
75115
75116 Add Colin Harrison to copyright and authors list
75117 Adjust license text for multiple authors
75118
75119 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75120
75121commit 5ff289834d7445cc6f60dbb1cf18423e7236311a
75122Author: Jon TURNEY <jon.turney@dronecode.org.uk>
75123Date: Mon Jan 19 17:42:54 2009 +0000
75124
75125 Cygwin/X: Place prototype for winSelectIcons() in a header file
75126
75127 Fixes warning as prototype is now in scope for the definition
75128
75129 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75130
75131commit ab2c98b86f1c3968581dfdad8b67432082feec78
75132Author: Jon TURNEY <jon.turney@dronecode.org.uk>
75133Date: Mon Jan 19 17:36:58 2009 +0000
75134
75135 Cygwin/X: Place prototypes for winCheckKeyPressed() and winFixShiftKeys() in a header file
75136
75137 Fixes warnings as they are now in scope at the definition
75138
75139 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75140
75141commit cc7a4c7c6e051fceacb32997740d83fc0655414e
75142Author: Jon TURNEY <jon.turney@dronecode.org.uk>
75143Date: Mon Jan 19 17:29:28 2009 +0000
75144
75145 cygwin/X: Remove an unused variable to fix a warning
75146
75147 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75148
75149commit 7c2fb098b206c0fb97c18cce8f88387cd0c0e82a
75150Author: Jon TURNEY <jon.turney@dronecode.org.uk>
75151Date: Mon Jan 19 17:22:59 2009 +0000
75152
75153 Cygwin/X: Fix several prototypes to return HICON
75154
75155 Fix prototypes of winOverrideIcon(), winTaskbarIcon() and winOverrideDefaultIcon() to return HICON
75156 Also use HICON type in WINPREFS stucture
75157 Remove various casts these changes make unnecessary
75158
75159 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75160
75161commit 19b3e44ce51cc2f74ba388cb2cd3a26467848e9a
75162Author: Jon TURNEY <jon.turney@dronecode.org.uk>
75163Date: Mon Jan 19 17:03:23 2009 +0000
75164
75165 Cygwin/X: Fix -Wold-style-definition warnings
75166
75167 Fix old-style definition warnings caused by definitions with empty
75168 parameter lists "()", which should be "(void)" to indicate the function
75169 takes no parameters, rather than unspecified parameters
75170
75171 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75172
75173commit 734b23e5982e171031077a2d5d6b5dc2a12e1a70
75174Author: Maarten Maathuis <madman2003@gmail.com>
75175Date: Wed Feb 11 14:31:17 2009 +0100
75176
75177 fb: add fbDoCopy and fbCopyRegion compatibility wrappers.
75178
75179commit 619c4d60eb3a8a51a8300ccf8ab5b98a688e6e12
75180Author: Eric Anholt <eric@anholt.net>
75181Date: Sun Feb 8 12:37:50 2009 +0100
75182
75183 glx: Replace broken GLX visual setup with a fixed "all" mode.
75184
75185 With trying to match depths so that you didn't end up with a depth 24
75186 fbconfig for the 32-bit composite visual, I broke the alpha bits on the depth
75187 24 X visual, which angered other applications. But in fixing that, the
75188 pickFBconfigs code for "minimal" also could end up breaking GLX visuals if
75189 the same FBconfig was chosen for more than one X visual.
75190 We have no reason to not expose as many visuals as possible, but the old
75191 "all" mode didn't match any existing X visuals to GLX visuals, so normal
75192 GL apps didn't work at all.
75193
75194 Instead, replace it with a simple combination of the two modes: Create GLX
75195 visuals by picking unique FBconfigs with as many features as possible for
75196 each X visual in order. Then, for all remaining FBconfigs that are
75197 appropriate for display, add a corresponding X and GLX visual.
75198
75199 This gets all applications (even ones that aren't smart enough to do FBconfigs)
75200 get all the options to get the visual configuration they want. The only
75201 potential downside is that the composite ARGB visual is unique and gets a
75202 nearly full-featured GLX visual (except that the root visual might have taken
75203 the tastiest FBconfig), which means that a dumb compositing manager could
75204 waste resources. Write compositing managers using FBconfigs instead, please.
75205
75206commit 7bbf05d9aa75b536a5321abf51f37852b8f109ed
75207Author: Shunichi Fuji <palglowr@gmail.com>
75208Date: Tue Feb 10 14:44:18 2009 +0900
75209
75210 dix: build fix if PANORAMIX is not defined.
75211
75212 Signed-off-by: Shunichi Fuji <palglowr@gmail.com>
75213 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75214
75215commit 4844bff58f296b2851be4e6b955c3a68d02437a9
75216Author: Peter Hutterer <peter.hutterer@who-t.net>
75217Date: Wed Feb 4 11:50:18 2009 +1000
75218
75219 config: if we can't connect to HAL, listen for a startup notification.
75220
75221 If HAL isn't available when we try to connect, the registered NameOwnerChanged
75222 signal handler waits until HAL is available. Once we connected to HAL, we
75223 unregister the signal handler again.
75224 This allows HAL to be started in parallel or after the server has started.
75225
75226 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75227
75228commit 5b2bbffc12cd7d29e4200943f8736bd6336ae4aa
75229Author: Peter Hutterer <peter.hutterer@who-t.net>
75230Date: Fri Jan 30 16:46:27 2009 +1000
75231
75232 dix: deduplicate SyntheticMotion.
75233
75234 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75235
75236commit 34820f9c1304891bebb4bff6a492288aec4fc0d1
75237Author: Peter Hutterer <peter.hutterer@who-t.net>
75238Date: Fri Jan 30 16:37:19 2009 +1000
75239
75240 dix: remove XineramaChangeToCursor.
75241
75242 It's nearly identical to ChangeToCursor now anyway.
75243
75244 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75245
75246commit 942eae6868b8b0f343b6aa921ddf77e8bb70798a
75247Author: Peter Hutterer <peter.hutterer@who-t.net>
75248Date: Fri Jan 30 16:25:22 2009 +1000
75249
75250 dix: Remove XineramaCheckPhysLimits.
75251
75252 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75253
75254commit bf044f6d10ad9f2baca91a2f04e82dc1af2657cd
75255Author: Peter Hutterer <peter.hutterer@who-t.net>
75256Date: Fri Jan 30 16:14:54 2009 +1000
75257
75258 dix: reduce redunancy in XineramaConfineCursorToWindow.
75259
75260 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75261
75262commit cdce5699f085e717ee1a2ed8390746dfcb9ddadf
75263Author: Peter Hutterer <peter.hutterer@who-t.net>
75264Date: Fri Jan 30 16:09:23 2009 +1000
75265
75266 dix: Get rid of XineramaCheckVirtualMotion.
75267
75268 Yes, this is an ugly piece mess of #ifdefs, but it beats having two nearly
75269 identical functions.
75270
75271 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75272
75273commit 1cb5afdf1ef53e72e0eef28ff6197bf5e86ef758
75274Author: Maarten Maathuis <madman2003@gmail.com>
75275Date: Sun Feb 8 18:59:17 2009 +0100
75276
75277 dix: always NULL pGC->tile.pixmap in Create{Scratch}GC.
75278
75279commit 6fd8fe9cb993f9b733cf7d878bc324acd09aa48d
75280Author: Maarten Maathuis <madman2003@gmail.com>
75281Date: Sun Feb 8 18:43:15 2009 +0100
75282
75283 exa: fix exaValidateGC.
75284
75285 - use DEST in the createPixmap wrapper, because stipple already takes MASK (in case someone uses swappers).
75286 - Anticipate some of the less common situations when fbValidateDrawable will access tile related pixmaps.
75287
75288commit 30daba6903378dd47f386c0012a0667fd2368f3e
75289Author: Maarten Maathuis <madman2003@gmail.com>
75290Date: Sun Feb 8 02:08:13 2009 +0100
75291
75292 exa: All fallbacks should have a GC, remove some code.
75293
75294 - I did some testing with full fallbacks forced by the driver.
75295 - I ran rendercheck, expedite and the (full) x11perf test suite.
75296 - Thanks to ajax for pointing out this should be unneeded.
75297
75298commit d591c7d1b87c61ee8add2fce689e35cf097eed4f
75299Author: Maarten Maathuis <madman2003@gmail.com>
75300Date: Sat Feb 7 20:47:54 2009 +0100
75301
75302 mi: kill a few warnings
75303
75304commit 00226d0b589595cdd45c75e7e28237334a8883b1
75305Author: Maarten Maathuis <madman2003@gmail.com>
75306Date: Fri Feb 6 20:56:49 2009 +0100
75307
75308 mi: clip exposures to pGC->clientClip.
75309
75310commit ffa72793e25f19a388a1369708f472921f721284
75311Author: Maarten Maathuis <madman2003@gmail.com>
75312Date: Fri Feb 6 20:47:01 2009 +0100
75313
75314 exa: Calling exaMarkSync after UTS is the drivers responsibility.
75315
75316commit da086901c6579e41b28b1418fa12ebb82744fc87
75317Author: Matt Turner <mattst88@gmail.com>
75318Date: Fri Feb 6 18:31:12 2009 -0500
75319
75320 Remove unused barrier macros on alpha.
75321
75322commit 68665d78e72c4c564073d66a0f59ac8d3f5fd756
75323Author: Maarten Maathuis <madman2003@gmail.com>
75324Date: Thu Feb 5 17:07:31 2009 +0100
75325
75326 exa: don't use fbCopyNtoN
75327
75328commit 2e76958d304a3c4080d62f32449724eeb9b95d93
75329Author: Maarten Maathuis <madman2003@gmail.com>
75330Date: Wed Feb 4 19:48:03 2009 +0100
75331
75332 fb: move some code to mi
75333
75334commit f06bde3d4bf9a0f245dfa6c65cd98f8e0c7dfdfc
75335Author: Maarten Maathuis <madman2003@gmail.com>
75336Date: Sat Jan 31 19:57:07 2009 +0100
75337
75338 exa: create ExaCheckGetImage
75339
75340commit 258fc4b106dfd7b583999f97caf4d4a7a11569b6
75341Author: Maarten Maathuis <madman2003@gmail.com>
75342Date: Sat Jan 31 19:40:48 2009 +0100
75343
75344 exa: wrap the remainder of exa_unaccel.c
75345
75346commit 6fabf249494bf38b69e3c9c6c9718ee8b12fc6e1
75347Author: Maarten Maathuis <madman2003@gmail.com>
75348Date: Sat Jan 31 19:25:20 2009 +0100
75349
75350 exa: use proper wrapping in exa.c
75351
75352commit 015c99a4ad572f5a1b68054efb070e1ffee1fed0
75353Author: Maarten Maathuis <madman2003@gmail.com>
75354Date: Sat Jan 31 18:53:52 2009 +0100
75355
75356 exa: properly wrap GC functions
75357
75358commit 5e6a06fe691db63ccf2dc7161a1f6c71c09741b7
75359Author: Maarten Maathuis <madman2003@gmail.com>
75360Date: Sat Jan 31 16:30:31 2009 +0100
75361
75362 exa: add GC private
75363
75364commit 5cc67ae94c066dcac78072ad8a819c3b602d8bab
75365Author: Maarten Maathuis <madman2003@gmail.com>
75366Date: Fri Jan 30 19:32:33 2009 +0100
75367
75368 exa: kill of exaImageGlyphBlt
75369
75370 - It serves no obvious purpose, yet it directly accesses many fb
75371 symbols.
75372
75373commit 6483834200fab90164c27c72d4710014edeb4479
75374Author: Maarten Maathuis <madman2003@gmail.com>
75375Date: Fri Jan 30 18:40:46 2009 +0100
75376
75377 exa: Remove one of the many calls directly into the fb layer.
75378
75379commit f112d6bf59964f317ab236221f7bce7fe2603ab8
75380Author: Michel Dänzer <daenzer@vmware.com>
75381Date: Fri Feb 6 18:30:43 2009 +0100
75382
75383 EXA: Try harder to keep current pixmap copy up to date in exaMigrateToward*.
75384
75385 This is probably required, but apparently not sufficient, for making migration
75386 heuristics other than "always" work correctly again. Not that I really care
75387 about them...
75388
75389commit 3948b523893d3d44b6a088340c4252e969613769
75390Author: Michel Dänzer <daenzer@vmware.com>
75391Date: Fri Feb 6 11:37:53 2009 +0100
75392
75393 EXA: Guard empty pending region warning by DEBUG_MIGRATE.
75394
75395 It isn't very useful yet while the damage layer calls us for empty operations,
75396 mostly confuses users.
75397
75398commit 9a1d07ecb74b7c3267a6910af66ada917a525110
75399Author: Tomas Carnecky <tom@dbservice.com>
75400Date: Wed Feb 4 20:20:24 2009 +0100
75401
75402 Fix "warning: cast from pointer to integer of different size"
75403
75404 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75405 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75406
75407commit 0e15697b53c9448ce9911aa6499b2ea0bda92af6
75408Author: Peter Hutterer <peter.hutterer@who-t.net>
75409Date: Wed Feb 4 10:11:13 2009 +1000
75410
75411 dix: die if we can't activate or init the VCP/VCK.
75412
75413 If we have a busted xkb setup, the XKB initialization on the core devices
75414 fails and leaves us with dev->key->xkbInfo == NULL. This in turn causes
75415 segfaults lateron.
75416
75417 Return BadValue when the XKB configuration for a master device failed, and if
75418 that happens for the VCP/VCK, die semi-gracefully.
75419 The VCP init can only fail on OOM.
75420
75421 Reported by Aaron Plattner.
75422
75423 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75424 Acked-by: Daniel Stone <daniel@fooishbar.org>
75425 Acked-by: Dan Nicholson <dbn.lists@gmail.com>
75426 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
75427
75428commit db836715198a445dc3b32c3df0610f700930483f
75429Author: Adam Jackson <ajax@redhat.com>
75430Date: Wed Feb 4 23:27:32 2009 -0500
75431
75432 Revert "Bus: Remove xf86SetAccessFuncs() and related machinery"
75433
75434 This reverts commit 2fe79990e31ac684637ca3c47198ee0cead10abe.
75435
75436 Reminder: nvidia uses this, because other parts of RAC are broken in
75437 worse ways, and it needs to disable that.
75438
75439commit 2fe79990e31ac684637ca3c47198ee0cead10abe
75440Author: Adam Jackson <ajax@redhat.com>
75441Date: Mon Jan 26 07:11:49 2009 -0500
75442
75443 Bus: Remove xf86SetAccessFuncs() and related machinery
75444
75445 Only mga was using this, and even then only behind an ifdef that no one
75446 ever built. Not a great idea in the first place.
75447
75448commit 9ff6180161267e71ae750498da0ec4889cd5c260
75449Author: Adam Jackson <ajax@redhat.com>
75450Date: Mon Jan 26 06:23:09 2009 -0500
75451
75452 Bus: Delete some obfuscatory macros
75453
75454commit 8eca37cb8934e79f02291460eb7b49a5d528f391
75455Author: Adam Jackson <ajax@redhat.com>
75456Date: Wed Feb 4 22:48:52 2009 -0500
75457
75458 Bus: Remove unused xf86ExtractTypeFromList
75459
75460commit b013111e98d50f5962e9b275eb28d969d5c9576e
75461Author: Adam Jackson <ajax@redhat.com>
75462Date: Mon Jan 26 03:14:18 2009 -0500
75463
75464 Bus: make the resource list static.
75465
75466commit bde028dd8d856f8d47c39e3c6d3731c34c398c74
75467Author: Adam Jackson <ajax@redhat.com>
75468Date: Mon Jan 26 22:58:58 2009 -0500
75469
75470 loader: Remove icache flushes on alpha.
75471
75472 If libdl doesn't already get this right for you, you have two problems.
75473
75474commit 910a1e88d23e0b7e4ca4679ab50a883a918f2590
75475Author: Adam Jackson <ajax@redhat.com>
75476Date: Mon Jan 26 07:54:07 2009 -0500
75477
75478 loader: Remove a vestige of archive support
75479
75480commit 366f23c6eb504fc23112f121769bcb719948474f
75481Author: Adam Jackson <ajax@redhat.com>
75482Date: Mon Jan 26 07:53:05 2009 -0500
75483
75484 loader: Remove useless call to LoaderGetOS
75485
75486 We know what OS you're running on anyway, it's printed in the uname
75487 string at the top of the log.
75488
75489commit 05e415a5a3f376e28955be3bf976390e8631e25f
75490Author: Adam Jackson <ajax@redhat.com>
75491Date: Mon Jan 26 07:52:07 2009 -0500
75492
75493 loader: Simplify loader magic
75494
75495commit bf1ca06a1693cbe5415a152936939ce3b088a022
75496Author: Adam Jackson <ajax@redhat.com>
75497Date: Mon Jan 26 07:51:35 2009 -0500
75498
75499 loader: Make a comment slightly less stupid
75500
75501commit 499908aeca9a31692c787070e23bfaa92d1b32a4
75502Author: Adam Jackson <ajax@redhat.com>
75503Date: Mon Jan 26 07:50:00 2009 -0500
75504
75505 loader: code motion.
75506
75507commit 28b6b1519c4c526662302fac4ed9b7f540bafd4a
75508Author: Adam Jackson <ajax@redhat.com>
75509Date: Mon Jan 26 07:47:46 2009 -0500
75510
75511 loader: Simplify handle allocation and refcounting.
75512
75513commit 0bad0552bfd4d10529ab51d618af8d08c5aeea71
75514Author: Adam Jackson <ajax@redhat.com>
75515Date: Wed Feb 4 17:20:24 2009 -0500
75516
75517 XKB: Remove -kb and +kb from -help text and man page.
75518
75519commit 7968823cbc02615e1080c0d3f34dcebe14ea1771
75520Author: Keith Packard <keithp@keithp.com>
75521Date: Tue Feb 3 20:18:37 2009 -0800
75522
75523 Handle the combination of panning and crtc transforms
75524
75525 This patch gets the shadow scanout buffer repainted on panning area changes.
75526 It does not, however, track the mouse correctly.
75527
75528 Signed-off-by: Keith Packard <keithp@keithp.com>
75529
75530commit 763df9eec79a867978efc5138d8635a46c1a0d17
75531Author: Keith Packard <keithp@keithp.com>
75532Date: Tue Feb 3 20:15:41 2009 -0800
75533
75534 Damage re-used shadow scanout buffer using new transforms.
75535
75536 When the shadow scanout buffer can be re-used, the underlying framebuffer
75537 area must be damaged so that the scanout will be repainted. This patch
75538 delays the addition of that damaged area until after the transform in the
75539 crtc has been updated, otherwise the old transform would have been used and
75540 the wrong area repainted.
75541
75542 Signed-off-by: Keith Packard <keithp@keithp.com>
75543
75544commit 62fc98cb88e4e8b636f343453fc1168a87c58972
75545Author: Keith Packard <keithp@keithp.com>
75546Date: Tue Feb 3 20:12:42 2009 -0800
75547
75548 Handle matrix computation overflow in RRTransformCompute
75549
75550 If the computation of the composite fixed-point transform for RandR
75551 overflows at any point, take the resulting floating point transform and
75552 scale that back to fit in a fixed point matrix. This ensures that a matrix
75553 will always be available, although perhaps at reduced precision. Someday we
75554 should add floating point matrices to Render.
75555
75556 Signed-off-by: Keith Packard <keithp@keithp.com>
75557
75558commit 7a3ddef3bc27982f3558183c74fc8e365cf2e7fd
75559Author: Tomas Carnecky <tom@dbservice.com>
75560Date: Wed Feb 4 00:07:02 2009 +0100
75561
75562 fbdevHWGetRec() doesn't return the private data
75563
75564 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75565 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75566
75567commit e04d9ab999978995b06f4b702293b96f69b0e8c3
75568Author: Tomas Carnecky <tom@dbservice.com>
75569Date: Wed Feb 4 00:08:21 2009 +0100
75570
75571 DBUS_API_SUBJECT_TO_CHANGE is already defined on the command line
75572
75573 dbus-core.c:30:1: warning: "DBUS_API_SUBJECT_TO_CHANGE" redefined
75574 <command-line>: warning: this is the location of the previous definition
75575
75576 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75577 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75578
75579commit 9f60d127b5371f2853dcc53903c6ff8c7670ad1b
75580Author: Tomas Carnecky <tom@dbservice.com>
75581Date: Wed Feb 4 00:07:40 2009 +0100
75582
75583 Fix "warning: braces around scalar initializer"
75584
75585 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75586 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75587
75588commit a88995dfb88719ae4a03483dc571dbbe83565285
75589Author: Tomas Carnecky <tom@dbservice.com>
75590Date: Wed Feb 4 00:03:58 2009 +0100
75591
75592 Make gcc happy: correct third argument of CopyGetMasterEvent()
75593
75594 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75595 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75596
75597commit dd3f4e829286e62f4c5c07ee93f9f576876acbf9
75598Author: Tomas Carnecky <tom@dbservice.com>
75599Date: Wed Feb 4 00:02:24 2009 +0100
75600
75601 Fix "warning: cast to pointer from integer of different size"
75602
75603 Add parenthesis around the whole expression.
75604
75605 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75606 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75607
75608commit 731a907ac21a753f306b53706f689469cefeb953
75609Author: Tomas Carnecky <tom@dbservice.com>
75610Date: Tue Feb 3 23:58:25 2009 +0100
75611
75612 Fix "warning: XXX defined but not used"
75613
75614 generic.c:80: warning: ‘read_legacy_video_BIOS’ defined but not used
75615
75616 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75617 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75618
75619commit bc57efffe654c493ceb62555cccb7926f4a646f1
75620Author: Tomas Carnecky <tom@dbservice.com>
75621Date: Tue Feb 3 23:54:53 2009 +0100
75622
75623 Fix "warning: unused variable XXX"
75624
75625 events.c:4614: warning: unused variable ‘kbd’
75626 xkbUtils.c:361: warning: unused variable ‘maxKeysPerMod’
75627 xf86Events.c:409: warning: unused variable ‘ke’
75628 generic.c:131: warning: unused variable ‘cs’
75629 generic.c:130: warning: unused variable ‘size’
75630 xf86RandR12.c:591: warning: unused variable ‘crtc’
75631
75632 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75633 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75634
75635commit 111fdef74dfc7e288cef675b171b673bf64fe056
75636Author: Tomas Carnecky <tom@dbservice.com>
75637Date: Tue Feb 3 21:18:53 2009 +0100
75638
75639 ISO C90: Move declaration to the beginning of the function
75640
75641 devices.c: In function ‘DoChangeKeyboardControl’:
75642 devices.c:1768: warning: ISO C90 forbids mixed declarations and code
75643
75644 Signed-off-by: Tomas Carnecky <tom@dbservice.com>
75645 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75646
75647commit 9fe9b6e4ef669b192ee349e3290db5d2aeea273c
75648Author: Peter Hutterer <peter.hutterer@who-t.net>
75649Date: Mon Feb 2 10:20:13 2009 +1000
75650
75651 mi: don't call UpdateSpriteForScreen if we have Xinerama enabled. #18668
75652
75653 In Xinerama all windows hang off the first root window. Crossing the screens
75654 must not reset the spriteTrace, otherwise picking fails and events are sent to
75655 the root window.
75656
75657 X.Org Bug 18668 <http://bugs.freedesktop.org/show_bug.cgi?id=18668>
75658
75659 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75660
75661commit f80332a04d0714a74f16b2552cccf3457e6494af
75662Author: Peter Hutterer <peter.hutterer@who-t.net>
75663Date: Mon Feb 2 14:28:16 2009 +1000
75664
75665 dix: for core events, pass the core filter into DeliverEventsToWindow
75666
75667 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75668
75669commit 01879e583b0ce0d8fddc37e36d9b69fe4f1d1e72
75670Author: Aaron Plattner <aplattner@nvidia.com>
75671Date: Tue Feb 3 13:49:19 2009 -0800
75672
75673 Don't enable XAA offscreen pixmaps with Option "XaaOffscreenPixmaps" "no".
75674
75675 xf86IsOptionSet is rarely the right function to use for boolean options because
75676 it returns TRUE whenever the option is present, even if its value is "no",
75677 "off", "0", etc.
75678
75679commit 5623c27700b7b23a8dbbd8c8f45e5d4fa0c667e3
75680Author: Alan Coopersmith <alan.coopersmith@sun.com>
75681Date: Mon Feb 2 19:25:14 2009 -0800
75682
75683 Constify atom name strings
75684
75685 Changes MakeAtom to take a const char * and NameForAtom to return them,
75686 since many callers pass pointers to constant strings stored in read-only
75687 ELF sections. Updates in-tree callers as necessary to clear const
75688 mismatch warnings introduced by this change.
75689
75690 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
75691 Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
75692
75693commit 6869efae74381e5305b2d6517bf286e3ef7fdcb7
75694Author: Jeremy Huddleston <jeremy@yuffie.local>
75695Date: Tue Feb 3 09:58:51 2009 -0800
75696
75697 XQuartz: Accept bundle version 2.1.6 to work with our startup method
75698 (cherry picked from commit 7f2966557374cb5ea22bc7b73d5ca04891eb5e78)
75699
75700commit 43a1c91a3dac4f0c8cea5de1b9785e349e09978a
75701Author: Peter Hutterer <peter.hutterer@who-t.net>
75702Date: Sun Feb 1 20:44:27 2009 +1000
75703
75704 Xext: fix typo in GEEventFill macro
75705
75706 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75707
75708commit 412e09e1e37b74421abd4dd9606536d2785da440
75709Author: Peter Hutterer <peter.hutterer@who-t.net>
75710Date: Thu Jan 15 11:03:03 2009 +1000
75711
75712 dix: Use GenericEvent instead of LASTEvent to check for core events.
75713
75714 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75715
75716commit 9f3c5d46f81887b2b413b3e6cc9df8282f06522c
75717Author: Peter Hutterer <peter.hutterer@who-t.net>
75718Date: Mon Feb 2 12:55:39 2009 +1000
75719
75720 Xi: initialize the device before applying button map changes.
75721
75722 This part got lost in bc909f71367a0.
75723 Reported by Tomas Carnecky.
75724
75725 Acked-by: Daniel Stone <daniel@fooishbar.org>
75726 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75727
75728commit 6bf6a4c47e94780bf0eef47702f21a505c336556
75729Author: Simon Thum <simon.thum@gmx.de>
75730Date: Fri Jan 23 12:06:16 2009 +0100
75731
75732 Xi: create well-known atoms on demand, rather than preinit them
75733
75734 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75735
75736commit 32adf8d34fb7db9b78859241f2670349b75c3f7e
75737Author: Peter Hutterer <peter.hutterer@redhat.com>
75738Date: Wed Jan 28 20:52:45 2009 +1000
75739
75740 Xi: If the MD doesn't have a key/kbdfeed class, init the keyboard.
75741
75742 This fixes crashes on mouse/keyboard combos where the mouse is both a keyboard
75743 and a mouse. Upon copying the SD's key info into the VCP, the NULL xkbInfo of
75744 the VCP would crash the server.
75745
75746 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75747
75748commit caade11b8465a743b2ecc890a4d180277e9ca716
75749Author: Peter Hutterer <peter.hutterer@redhat.com>
75750Date: Wed Jan 28 20:44:47 2009 +1000
75751
75752 Xi: remove now obsolete oldXkbInfo.
75753
75754 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
75755
75756commit 5e0967f5fc2e0fcf2a55128eb03931cf6f5709e9
75757Author: Peter Hutterer <peter.hutterer@who-t.net>
75758Date: Fri Dec 12 10:49:53 2008 +1000
75759
75760 dix: add SetBit(arr, bit) and ClearBit(arr, bit) to include/inputstr.h
75761
75762 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75763
75764commit faeb18eb7eb20f15528f9c8004d843146ebc2a50
75765Author: Peter Hutterer <peter.hutterer@who-t.net>
75766Date: Tue Jan 27 15:40:31 2009 +1000
75767
75768 Xext: rename shape's EventType to ShapeEventType to avoid name clashing.
75769
75770 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75771
75772commit cfd3443fe81685e91a53063dee58e24a5684dc29
75773Author: Peter Hutterer <peter.hutterer@who-t.net>
75774Date: Wed Jan 28 13:34:10 2009 +1000
75775
75776 Xext: rename saver's EventType to SaverEventType.
75777
75778 Avoid namespace clashing with the internal events.
75779
75780 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75781
75782commit 8460425740c72e58bcb897557f033e2836251160
75783Author: Peter Hutterer <peter.hutterer@who-t.net>
75784Date: Wed Dec 10 12:31:59 2008 +0000
75785
75786 render: rename SetBit to RenderSetBit.
75787
75788 Avoiding namespace collision with the SetBit macro soon to be used in the
75789 input code.
75790
75791 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75792
75793commit 940a7aeebc9b042b0208ee475ef162c485ac2274
75794Author: Peter Hutterer <peter.hutterer@who-t.net>
75795Date: Tue Jan 27 15:52:49 2009 +1000
75796
75797 include: remove now-unused sempahore macros.
75798
75799 Obsolete with the new enter/leave model.
75800
75801 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
75802
75803commit f5c7f8a3ab5723b6c1c3cd21b90a87fd7f37cb7f
75804Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
75805Date: Mon Feb 2 12:30:57 2009 -0800
75806
75807 XQuartz: GLX: Fix accidental exclusion of glBlitFramebufferEXT on Leopard
75808 (cherry picked from commit 8620579483789ae9ac1b68dbefe32b70011386de)
75809
75810commit 5100d829a4d71ce4a9fbc2b81694a1fb90066ccf
75811Author: Eric Anholt <eric@anholt.net>
75812Date: Mon Feb 2 10:13:46 2009 -0800
75813
75814 glx: Don't match fbconfigs to visuals with mismatched channel masks.
75815
75816 This fixes at least one known bug, where the depth 32 visual would end up
75817 with a depth 24 fbconfig attached, angering compiz.
75818
75819commit 4475ae036c3d5a06a6ce88a359ce2a9a8ae9b4f1
75820Author: Jon TURNEY <jon.turney@dronecode.org.uk>
75821Date: Mon Jan 26 13:15:07 2009 +0000
75822
75823 Cygwin/X: Fix compilation for mandatory XKB
75824
75825 Fix a stray '}'
75826 Update to use RMLVO interface
75827 Remove g_winInfo.xkb.disable, can never be set since noXkbExtension has been removed
75828 Change to retrieve modifier key state using XkbStateFieldFromRec() from Colin Harrison
75829 Update to use XKB defaults from xkb-config.h
75830
75831 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
75832
75833commit e2b4d3d4a1994347979508fbb483cd55c2232502
75834Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
75835Date: Sun Feb 1 15:08:02 2009 -0800
75836
75837 XQuartz: GLX: OpenGL.framework on Tiger doesn't have glBlitFramebufferEXT
75838 (cherry picked from commit e01662cadcaa4052e3a5aa82cbaed00a2d0220e0)
75839
75840commit 66a6fbbfa3a1fd0d91fe61cf2ab299d0e3df46b9
75841Author: Jim Huang <jserv@kaffe.org>
75842Date: Sun Feb 1 09:19:20 2009 -0800
75843
75844 kdrive: Move a close() to the right place so we don't close(-1) normally.
75845
75846 Bug #18343
75847
75848commit 0055fe66d5f73742cafab868ccdb7a6f36ea1dd5
75849Author: Manuel Bouyer <bouyer@netbsd.org>
75850Date: Sun Feb 1 09:14:19 2009 -0800
75851
75852 netbsd: Force the use of ev56 instructions for register access on ev56.
75853
75854 This avoids 32-bit access which might affect other registers. The linux code
75855 uses gcc flags to get this to happen, but this seems like more of a sure thing.
75856
75857commit 1375c1b07df114d864a0a06786bc75a3a4f65177
75858Author: Manuel Bouyer <bouyer@netbsd.org>
75859Date: Sun Feb 1 09:12:21 2009 -0800
75860
75861 netbsd: Fix alpha ev6 support.
75862
75863 Bug #19721
75864
75865commit ad8ffbe7cb50387e6a9cc4a24fea0d9c13b7bfc8
75866Author: John Hein <jhein@timing.com>
75867Date: Fri Jan 30 21:30:35 2009 -0800
75868
75869 Actually require bigfontproto when it's enabled. Bug #13710.
75870
75871commit 29b3b88dc744f4919c6709747ddb7baac47486c5
75872Author: Pierre Willenbrock <pierre@pirsoft.de>
75873Date: Fri Jan 30 21:16:48 2009 -0800
75874
75875 Prevent double unref of glxdrawables
75876
75877 Found by valgrind. Bug #18917.
75878
75879commit 60bcdd687040db76490851d4b459284ce37020e0
75880Author: Alan Coopersmith <alan.coopersmith@sun.com>
75881Date: Fri Jan 30 21:44:08 2009 -0800
75882
75883 x11-input.fdi: Add options needed to handle adding USB devices on Solaris
75884
75885commit 4de5705545792d77c9990bbec3e263a517757a2a
75886Author: Keith Packard <keithp@keithp.com>
75887Date: Fri Jan 30 20:17:10 2009 -0800
75888
75889 RandR crtcs not fetched correctly when primary output is set.
75890
75891 Primary outputs may not have a CRTC.
75892 Loops fetching CRTCs respecting primary output were broken.
75893
75894 Signed-off-by: Keith Packard <keithp@keithp.com>
75895
75896commit f716e3f3445d443cbc6507d27f806e9ad387120a
75897Author: Eric Anholt <eric@anholt.net>
75898Date: Fri Jan 30 20:10:21 2009 -0800
75899
75900 modes: Protect xf86_crtc_supports_gamma() from non-RandR 1.2 drivers.
75901
75902commit 317f2b4a9fe4b606975711bc332166a82db5087d
75903Author: Eric Anholt <eric@anholt.net>
75904Date: Fri Jan 30 19:06:17 2009 -0800
75905
75906 randr: Avoid re-querying the configuration on everything but GetScreenResources.
75907
75908 The new path should only re-query on the other requests when we haven't
75909 gathered the information from the DDX yet (such as with a non-RandR 1.2 DDX).
75910
75911 Bug #19037.
75912
75913commit c1f2be1f3fd0c80cb4c85b98140b98aa9311242b
75914Author: Keith Packard <keithp@keithp.com>
75915Date: Fri Jan 30 19:14:36 2009 -0800
75916
75917 RandR SetOutputPrimary should work with or without a crtc connected
75918
75919 The test was inverted from the protocol spec, and besides, the test is bogus
75920 anyways.
75921
75922 Signed-off-by: Keith Packard <keithp@keithp.com>
75923
75924commit dd098501d800571e71d06ffc936635a6c9d44e05
75925Author: Eric Anholt <eric@anholt.net>
75926Date: Fri Jan 30 16:17:58 2009 -0800
75927
75928 Move the apple fat binary hacks back to a header file, and make it apple-only.
75929
75930commit 6e49fdd2c839b6244c23ce95c3ae76a1a265b986
75931Author: Keith Packard <keithp@keithp.com>
75932Date: Fri Jan 30 15:43:04 2009 -0800
75933
75934 Make RandR CRTC info report panning area instead of just crtc area
75935
75936 This makes the RandR info consistent with the Xinerama info.
75937
75938 Signed-off-by: Keith Packard <keithp@keithp.com>
75939
75940commit ea309e47457156b60aadbf113f04e5b6851029c8
75941Author: Keith Packard <keithp@keithp.com>
75942Date: Fri Jan 30 14:19:41 2009 -0800
75943
75944 Make crtc_notify wrap/unwrap code do nothing unless mode code is inuse
75945
75946 Drivers not using the new hw/xfree86/modes code would crash in DRI due to
75947 that code trying to monitor CRTC changes.
75948
75949 Signed-off-by: Keith Packard <keithp@keithp.com>
75950
75951commit fff00df94d7ebd18a8e24537ec96073717375a3f
75952Author: Keith Packard <keithp@keithp.com>
75953Date: Fri Jan 30 14:16:12 2009 -0800
75954
75955 Ignore EDID-supplied monitor physical sizes for core screen size
75956
75957 We report the EDID values in RandR, and we let people configure whatever
75958 they like for the screen in xorg.conf. Reporting the EDID values in the core
75959 means applications get inconsistent font sizes in the default configuration.
75960
75961 Signed-off-by: Keith Packard <keithp@keithp.com>
75962
75963commit 4e7bc2d7a9879297aff0924febeca9bd3b6c336d
75964Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
75965Date: Fri Jan 30 03:56:39 2009 -0200
75966
75967 Work around inclusion of <X11/extensions/panoramiXext.h>
75968
75969 The X Server build only needs the macros PANORAMIX_MAJOR_VERSION
75970 and PANORAMIX_MINOR_VERSION from that header.
75971 Addition of extra prototypes to <X11/extensions/panoramiXext.h>
75972 caused a X Server build failure.
75973
75974commit 15ac25627e7239629be59adedda5d8a9deb330ef
75975Author: Matt Turner <mattst88@gmail.com>
75976Date: Thu Jan 29 14:40:27 2009 -0500
75977
75978 Remove extra / from Loading module line
75979
75980 In the X log, upon module load, it prints a line similar to the following.
75981
75982 > (II) Loading /usr/lib64/xorg/modules/extensions//libdbe.so
75983
75984 The attached patch removes the extra / before the module name.
75985
75986 Code already exists in hw/xfree86/loader/loadmod.c:InitPathList to add a
75987 trailing slash if needed, removing the one added by sprintf is harmless.
75988
75989 Signed-off-by: James Cloos <cloos@jhcloos.com>
75990
75991commit 3903e23153a0d47704050944fa83a7b31bc2eebc
75992Author: Maarten Maathuis <madman2003@gmail.com>
75993Date: Thu Jan 29 12:42:57 2009 +0100
75994
75995 XKB: ifdef XKB is dead, yet it was still present in a few places.
75996
75997commit 756a2c848305aa06bfe636314e28ad7f97c1168a
75998Author: Pierre Willenbrock <pierre@pirsoft.de>
75999Date: Wed Jan 28 22:18:50 2009 +0100
76000
76001 Fix duplicate code, off-by one in space calculation, not initialized members
76002
76003commit b55db937fee5ddf34518a54f352c0ee9b8cc2c2c
76004Author: Søren Sandmann Pedersen <sandmann@redhat.com>
76005Date: Wed Jan 28 17:33:25 2009 -0500
76006
76007 [shadow] Store the closure in the buffer
76008
76009commit 82fc102568b3d6b0daeb6c5f5b3a1310a7f14fcd
76010Author: Peter Hutterer <peter.hutterer@redhat.com>
76011Date: Wed Jan 28 07:51:32 2009 +1000
76012
76013 Disable Xnest by default.
76014
76015 Xneest doesn't build right now and is in the process of being removed. So
76016 let's not build it by default.
76017
76018commit f7585d020593345e7a61b1fe1a517c38da8f6d79
76019Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
76020Date: Tue Jan 27 03:13:28 2009 -0200
76021
76022 Convert kdrive libraries to libtool convenience libraries.
76023
76024 Also correct a link failure due to unresolved symbols. This
76025 is arguably a libtool/ranlib/ld bug, that "may" be corrected
76026 by linking all convenience libraries in a single one. But in
76027 this case, it was preferred to just add a linker option to
76028 Xfake_LDFLAGS, to force linkage of all libraries.
76029
76030 This corrects #19725.
76031
76032commit b43a4e2d38adb9ff16204218f24df93cfd820d1b
76033Author: Magnus Kessler <Magnus.Kessler@gmx.net>
76034Date: Tue Jan 27 09:30:37 2009 +1000
76035
76036 Xvfb: add missing include for new xkbsrv.h
76037
76038 Commit 08363c5830bdea34012dcd954b45ccfdc79a3a7e added call to
76039 XkbGetRulesDflts defined in xkbsrv.h
76040
76041 Signed-off-by: Magnus Kessler <Magnus.Kessler@gmx.net>
76042 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76043
76044commit a7e6424d992151c504a75bb807fa34c57ad52cad
76045Author: Peter Hutterer <peter.hutterer@who-t.net>
76046Date: Fri Jan 23 17:11:57 2009 +1100
76047
76048 dix: remove obsolete comment. Event lists should not be allocated by the DDX
76049
76050commit b137a0cdb1d517a807d081fbde8e8dc0539fb3ca
76051Author: George Staplin <gstaplin@apple.com>
76052Date: Mon Jan 26 03:02:38 2009 -0700
76053
76054 XQuartz: GL: indirect.c changes to fix the build with newer OpenGL frameworks.
76055 (cherry picked from commit 143224405ba74929c702a95de52b56df140b0d1b)
76056
76057commit 34b896e50fa154ad890a458c60932964421e739f
76058Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
76059Date: Fri Jan 16 19:26:07 2009 -0800
76060
76061 XQuartz: SnowLeopard: Help system book name changed in 10.6
76062 (cherry picked from commit b57cb05c69acbedb00a97234099ea104309aa2cb)
76063
76064commit c7d62d74bbd6e0067f511fd910b9e633576befb7
76065Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
76066Date: Sun Jan 25 19:19:01 2009 -0200
76067
76068 Explicitly add libxorg_la_LIBADD to libxorg_la_DEPENDENCIES
76069
76070 Otherwise, for example, when hacking config/*.c, it is required to
76071 run make clean on that directory, to ensure the proper libconfig.a
76072 will be linked in the generated Xorg binary.
76073
76074commit 0eec7348875035792097f606487a558422774e2c
76075Author: Magnus Kessler <Magnus.Kessler@gmx.net>
76076Date: Fri Jan 23 12:11:00 2009 -0200
76077
76078 .gitignore: ignore sdksyms.dep
76079
76080 Signed-off-by: Magnus Kessler <Magnus.Kessler@gmx.net>
76081
76082commit 88efe9c76f85423d952ff38e0cfd8b690439902a
76083Author: Magnus Kessler <Magnus.Kessler@gmx.net>
76084Date: Fri Jan 23 11:25:00 2009 -0200
76085
76086 Xvfb: Remove unused function GetLK201Mappings in InitInput.c
76087
76088 Silence a gcc warning. After commit 08363c5830bdea34012dcd954b45ccfdc79a3a7e
76089 GetLK201Mappings is no longer needed.
76090
76091 Signed-off-by: Magnus Kessler <Magnus.Kessler@gmx.net>
76092
76093commit 6ef46c40e62def4841a4cff4e0b443516a2ed782
76094Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
76095Date: Sun Jan 25 19:01:10 2009 -0200
76096
76097 Update xnest keyboard code to match xephyr/kdrive.
76098
76099 Xnest was not updated in the last batch of xkb changes. This
76100 patch is basically cut&paste from hw/kdrive/src/kinput.c and
76101 hw/kdrive/ephyr/ephyr.c, and appears to generate a Xnest as
76102 functional as before the xkb changes.
76103
76104commit 759348f13ffa02fb1804238189ffce1c92d33e2a
76105Author: Adam Jackson <ajax@redhat.com>
76106Date: Thu Jan 22 22:09:11 2009 -0500
76107
76108 Xephyr: Add some rudimentary Xnest command line compat
76109
76110commit 01ffaf6834fd6a3b1ddbd33d0f0b8d086be0737e
76111Author: Peter Hutterer <peter.hutterer@who-t.net>
76112Date: Mon Jan 19 21:11:05 2009 +1100
76113
76114 Xi: define a range of axis labels.
76115
76116 This is copied from linux/input.h, presumably that's the ones at least the
76117 Linux kernel can give us for any device.
76118
76119 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76120
76121commit 132b464d734b077038e19b21e46d3a6258f4b998
76122Author: Adam Jackson <ajax@redhat.com>
76123Date: Thu Jan 22 02:11:16 2009 -0500
76124
76125 Remove a bunch of useless casts.
76126
76127 We've had void * for twenty years now people let's try to act like we
76128 know how it works.
76129
76130commit 0fb4390526bb829ab17ff4635d41a3012f63c1b2
76131Author: Adam Jackson <ajax@redhat.com>
76132Date: Thu Jan 22 01:13:17 2009 -0500
76133
76134 Remove xorgcfg 'Options' list.
76135
76136commit b5ece46b76c9182a077141505cc0e788c0fca3ea
76137Author: Daniel Stone <daniel@fooishbar.org>
76138Date: Sat Jan 17 20:07:40 2009 +0200
76139
76140 Input: Clarify valuator FatalError messages
76141
76142 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76143 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76144
76145commit ddb46c1421a98190313b79621c0f5ae43b97d3f6
76146Author: Daniel Stone <daniel@fooishbar.org>
76147Date: Sat Jan 17 20:03:10 2009 +0200
76148
76149 Input: Use previous state in valuator events
76150
76151 Make valuator event state match other events by using the device state
76152 from before processing the event, not after. Also, we already check the
76153 number of valuators in UpdateDeviceState, so no need to do it again.
76154
76155 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76156 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76157
76158commit 18e51911f5748f9a27a18cd6bb421c899f5b048f
76159Author: Daniel Stone <daniel@fooishbar.org>
76160Date: Sat Jan 17 13:44:47 2009 +0200
76161
76162 Input: Add postdown to ButtonClassRec
76163
76164 This mirrors that in KeyClassRec: the state of the buttons as posted to
76165 GetPointerEvents, rather than the state of the buttons as processed by
76166 ProcessOtherEvent and friends.
76167
76168 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76169 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76170
76171commit c05f3bc935bef4d98216d38c0e98bd57ced382df
76172Author: Daniel Stone <daniel@fooishbar.org>
76173Date: Sat Jan 17 19:35:15 2009 +0200
76174
76175 Input: Don't allow DDX to generate repeat events
76176
76177 We always use soft-repeat at the moment; XKB posts a release/press sequence,
76178 which admittedly needs cleaning up, but that's for another day.
76179
76180 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76181 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76182
76183commit 6aef4e96affcc26b9415c6bc9c9bebb5af05a1a7
76184Author: Daniel Stone <daniel@fooishbar.org>
76185Date: Sat Jan 17 13:42:46 2009 +0200
76186
76187 XKB: Fix logic error
76188
76189 Use logical or instead of bitwise or.
76190
76191 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76192 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76193
76194commit 0ec9b1069aeeaf66fd104c09f08202f7ebccd170
76195Author: Daniel Stone <daniel@fooishbar.org>
76196Date: Sat Jan 10 14:44:28 2009 +0200
76197
76198 XKB: Remove unused XkbProcessOtherEvent
76199
76200 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76201 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76202
76203commit 7af53799ccefe27b996bdc6073075b7f12353d3a
76204Author: Daniel Stone <daniel@fooishbar.org>
76205Date: Sun Jan 4 16:39:18 2009 +1100
76206
76207 Input: Remove unused CoreProcess{Keyboard,Pointer}Event
76208
76209 Everything goes through XKB's Process{Keyboard,Pointer}Event on its way
76210 through to ProcessOtherEvent now, so get rid of the old, useless functions.
76211
76212 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76213 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76214
76215commit 7c4c00649cae855e141ec9a3667bfe399e3156d0
76216Author: Daniel Stone <daniel@fooishbar.org>
76217Date: Thu Jan 1 05:25:52 2009 +1100
76218
76219 XKB: Remove unused DDX functions
76220
76221 They were complete no-ops anyway.
76222
76223 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76224 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76225
76226commit 4fa3872dc2bcfd6d1fc88d0a94c7071683eea899
76227Author: Daniel Stone <daniel@fooishbar.org>
76228Date: Sat Oct 18 20:26:00 2008 +0100
76229
76230 Input: Remove core keysyms from KeyClassRec
76231
76232 Instead of always keeping two copies of the keymap, only generate the
76233 core keymap from the XKB keymap when we really need to, and use the XKB
76234 keymap as the canonical keymap.
76235
76236 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76237 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76238
76239commit bc909f71367a02297e725bb5769c2bcadab22395
76240Author: Daniel Stone <daniel@fooishbar.org>
76241Date: Tue Aug 12 00:35:56 2008 +0300
76242
76243 Input: Centralise pointer map changing
76244
76245 Replace both core and Xi functions with one function that validates the
76246 proposed map, and sends out both kinds of notification.
76247
76248 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76249 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76250
76251commit f06a9d2e05e13466c115fc706966a90b1fb0518e
76252Author: Daniel Stone <daniel@fooishbar.org>
76253Date: Sat Oct 18 20:59:30 2008 +0100
76254
76255 Input: Clean up keymap change notifications
76256
76257 Keyboard map notifications are always generated from within XKB code,
76258 which also takes care of copying the keysyms, etc. If you need to
76259 mangle the keymap yourself, generate a new core keymap/modmap, and pass
76260 it to XkbApplyMappingChange.
76261
76262 SendMappingNotify is renamed to SendPointerMappingNotify (and ditto its
76263 Device variants), which still only _sends_ the notifications, as opposed
76264 to also doing the copying a la XkbApplyMappingChange.
76265
76266 Also have the modmap change code traverse the device hierachy, rather
76267 than just going off the core keyboard.
76268
76269 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76270
76271commit 1d1a0f67eee330a286fbdef17e967ce8ea201548
76272Author: Daniel Stone <daniel@fooishbar.org>
76273Date: Tue Oct 28 16:27:37 2008 +1100
76274
76275 Xi: Introduce XIShouldNotify
76276
76277 XIShouldNotify just lets you know if you should send an event for a
76278 keymap change (or similar) concerning a given device to a given client;
76279 at the moment, this is only for devices which are sending events to that
76280 client.
76281
76282 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76283 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76284
76285commit b5242789edb726bb539e4d42fa35bfdaca0025c9
76286Author: Daniel Stone <daniel@fooishbar.org>
76287Date: Sat Aug 9 17:02:02 2008 +0300
76288
76289 XKB: Simplify keymap writing a bit
76290
76291 We don't need no temporary variable.
76292
76293 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76294
76295commit feb757f384382c7782ceac55f99d54c7caadbd9d
76296Author: Daniel Stone <daniel@fooishbar.org>
76297Date: Fri Apr 25 19:54:29 2008 +0300
76298
76299 XKB: Sanitise vmods for redirected keys
76300
76301 Turn two unsigned chars into one unsigned int for both vmods and the
76302 vmod mask. As a bonus, remove broken unused accessor macro for setting
76303 the vmods.
76304
76305 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76306 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76307
76308commit b5f49382fe48f0a762d9a15fb10a7d7e1183fc8d
76309Author: Daniel Stone <daniel@fooishbar.org>
76310Date: Fri Apr 25 19:52:11 2008 +0300
76311
76312 XKB: Sanitise ctrls action
76313
76314 Turn four unsigned chars into one unsigned long.
76315
76316 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76317 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76318
76319commit 1bd7fd195d85681e722161f8c636a28f11b40abb
76320Author: Daniel Stone <daniel@fooishbar.org>
76321Date: Fri Apr 25 19:50:12 2008 +0300
76322
76323 XKB: Sanitise pointer actions
76324
76325 Turn two unsigned chars into one int.
76326
76327 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76328 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76329
76330commit 61c508fa78aa08ea2666fde950fbafad95d65056
76331Author: Daniel Stone <daniel@fooishbar.org>
76332Date: Fri Apr 25 19:47:16 2008 +0300
76333
76334 XKB: Sanitise vmods in actions
76335
76336 Turn vmods from two unsigned chars into one int.
76337
76338 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76339 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76340
76341commit 4f3078df2acf07364a17f738bc5c32f9a5962723
76342Author: Daniel Stone <daniel@fooishbar.org>
76343Date: Sat Aug 9 16:50:22 2008 +0300
76344
76345 XKB: Add a hell of a lot more FIXMEs
76346
76347 For some reason, we insist on having daft internal representations that
76348 make no sense, that always have to be converted to be used. We should
76349 really sort this one out.
76350
76351 Also, comment the hojillion members of XkbStateRec.
76352
76353 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76354
76355commit edeb033f29ae2d52e86c543ccbf22fb7ae1fdb7f
76356Author: Daniel Stone <daniel@fooishbar.org>
76357Date: Wed Oct 15 18:07:52 2008 +0100
76358
76359 XKB: Explicitly decode action data
76360
76361 Rather than requiring a one-to-one correspondence between XKM and struct
76362 formats in action data, explicitly fill the action data, so we can break
76363 API.
76364
76365 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76366 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76367
76368commit 699824a35791735b4b4378adf8648e3aaa599ace
76369Author: Daniel Stone <daniel@fooishbar.org>
76370Date: Fri Jul 18 16:41:29 2008 +0300
76371
76372 XKB: Remove descriptions from maprules
76373
76374 We don't use them, as they're not up to the task. We'll get a better
76375 solution someday, promise.
76376
76377 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76378 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76379
76380commit 1ad80678d88525159e55bb814205fd4eaca060a3
76381Author: Daniel Stone <daniel@fooishbar.org>
76382Date: Fri Jul 18 16:29:35 2008 +0300
76383
76384 XKB: Remove support for setting combined keymaps
76385
76386 We don't do full keymaps anymore. Deal.
76387
76388 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76389 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76390
76391commit 133e0bd6f1cc147a2ddbbe1519d10300e7dbe669
76392Author: Daniel Stone <daniel@fooishbar.org>
76393Date: Fri Jul 18 16:27:57 2008 +0300
76394
76395 XKB: Remove 'extra' functionality from rules parsing
76396
76397 When we find something weird in the rules, don't stash it as an extra
76398 freeform component, just state that the rules file is likely broken and
76399 move on with our lives.
76400
76401 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76402 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76403
76404commit cc5c6d628aa0dc87c2cc9d063972e09ad747a596
76405Author: Daniel Stone <daniel@fooishbar.org>
76406Date: Fri Apr 25 18:45:37 2008 +0300
76407
76408 XKB: Remove unsupported Xi operation flags
76409
76410 We support every XKB operation on Xi devices, so always report that we
76411 support everything, and that nothing is ever unsupported.
76412
76413 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76414 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76415
76416commit 32db27a7f867b503c2840ca7b815e96d10be9210
76417Author: Daniel Stone <daniel@fooishbar.org>
76418Date: Thu Apr 17 12:42:01 2008 -0700
76419
76420 Input: Remove modifierMap from core
76421
76422 We already have modmap (in the exact same format!) in XKB, so just use
76423 that all the time, instead of duplicating the information.
76424
76425 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76426 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76427
76428commit f062e90a95f9b7ae5458ef2100615e8ace9b66a7
76429Author: Daniel Stone <daniel@fooishbar.org>
76430Date: Wed Apr 16 19:15:30 2008 +0300
76431
76432 Input: Remove modifierKeyMap
76433
76434 Since modifierKeyMap is generated from modifierMap, just remove it, and
76435 only generate it when we need to send the modifier map to the client.
76436
76437 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76438 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76439
76440commit cf6a2fc2bd94b392cfea120444d5e032d26f1d37
76441Author: Daniel Stone <daniel@fooishbar.org>
76442Date: Sun Mar 30 17:25:39 2008 +0300
76443
76444 Input: Ignore modifiers in core input processing
76445
76446 Modifiers get cleared by the XKB code when we drop down into core input
76447 processing, so just delete the dead code path to simplify things a bit.
76448
76449 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76450 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76451
76452commit 6727ee94086d238f9f99379280d4e4827f388cce
76453Author: Daniel Stone <daniel@fooishbar.org>
76454Date: Sat Aug 9 16:45:45 2008 +0300
76455
76456 Input: Remove state from KeyClassRec
76457
76458 We already have state fully stored within XKB, so instead of duplicating it,
76459 just generate the values to send to clients when required.
76460
76461 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76462 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76463
76464commit 08363c5830bdea34012dcd954b45ccfdc79a3a7e
76465Author: Daniel Stone <daniel@fooishbar.org>
76466Date: Fri Dec 26 18:38:25 2008 +1100
76467
76468 Input: Overhaul keyboard initialisation process
76469
76470 XkbInitKeyboardDeviceStruct is now the only valid keyboard
76471 initialisation: all the details are hidden behind here. This now makes
76472 it impossible to supply a core keymap at startup.
76473
76474 If dev->key is valid, dev->key->xkbInfo->desc is also valid.
76475
76476 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76477 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76478
76479commit 40877c6680863bd6a8475f2bb0c54df55bcf0b0e
76480Author: Daniel Stone <daniel@fooishbar.org>
76481Date: Tue Mar 18 10:05:57 2008 +0200
76482
76483 XKB: Make XKB mandatory
76484
76485 No more #ifdef XKB, because you can't disable the build, and no more
76486 noXkbExtension either.
76487
76488 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76489 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76490
76491commit f9da417163b6b2d6234d2542c1f375e33db7159a
76492Author: Daniel Stone <daniel@fooishbar.org>
76493Date: Fri Mar 28 19:26:42 2008 +0200
76494
76495 Xephyr: Disable state inheritance hack
76496
76497 Prepare for the impending removal of the state field by disabling this hack
76498 for a while: it's hell of nasty and I'm amazed it ever really worked.
76499
76500 Basically, on focus out, it should do as current DDXes do and fake releases
76501 for all keys (not just mangle the core state) that are currently down;
76502 buttons too. When focus comes back in, we already have a KeymapNotify that
76503 lets us know what's currently down, so we can use this to fake the
76504 appropriate keypresses, and send it through the event routing layer.
76505
76506 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76507
76508commit 5c281446d2a8c38511d45baee6f0d25a640f12e1
76509Author: Daniel Stone <daniel@fooishbar.org>
76510Date: Fri Mar 28 11:56:34 2008 +0200
76511
76512 XKB: Remove lock actions disabling
76513
76514 For some reason, XKB allows clients to set a global (!) flag that simply
76515 turns lock keys into state no-ops. Ignore this flag.
76516
76517 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76518
76519commit 2f7cb7306709d5266688e05a066701d309323035
76520Author: Daniel Stone <daniel@fooishbar.org>
76521Date: Fri Dec 26 18:36:30 2008 +1100
76522
76523 XKB: Add XkbRMLVOSet
76524
76525 XkbRMLVOSet is just a set of strings for rules, model, layout, variant
76526 and options; use that in preference to XkbRF_VarDefsRec, which is a
76527 hideously complicated monster that somehow managed to not include the
76528 actual rules.
76529
76530 While we're at it, clean up xkbrules.h so it doesn't require xkbstr.h.
76531
76532 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76533
76534commit 3208eff2d7b856a4ac0d650498ccb36e7e62e034
76535Author: Alan Coopersmith <alan.coopersmith@sun.com>
76536Date: Wed Jan 21 14:45:06 2009 -0800
76537
76538 Update xorg-server.h.in & Xnest to use new XKB default #defines
76539
76540commit cdf216aa17b776130a91fa6792c321620e810a18
76541Author: Alan Coopersmith <alan.coopersmith@sun.com>
76542Date: Tue Jan 20 16:50:43 2009 -0800
76543
76544 Fix XKB default configure options to use the right variables
76545
76546commit 0d3c3dad21857ef502fe0d0235620a1228248e13
76547Author: Adam Jackson <ajax@redhat.com>
76548Date: Tue Jan 20 23:46:48 2009 -0500
76549
76550 Don't include fontmod.h
76551
76552commit 4f004ab9a83e209943760e1faa39f06c8eeb27af
76553Author: Benjamin Close <Benjamin.Close@clearchain.com>
76554Date: Thu Jan 22 01:36:31 2009 +1030
76555
76556 XNest: Fix the build caused by __XKBDEFRULES__ -> XKB_DFLT_RULES in 23862ede59a9ce11a06ec5151bde460fb836c603
76557
76558commit e45a8a498d1ee8e40fc334805c38c05cb04b49d0
76559Author: Adam Jackson <ajax@redhat.com>
76560Date: Tue Jan 20 22:07:07 2009 -0500
76561
76562 Remove CreateUnclippedWinSize from window.h
76563
76564 This is utterly wrong, but then, so is sdksyms.sh
76565
76566commit 20d2117eb82fb7ce91afd4f01fe1b9d4b1bf0459
76567Author: Adam Jackson <ajax@redhat.com>
76568Date: Tue Jan 20 00:17:42 2009 -0500
76569
76570 Move CreateUnclippedWinSize to mbuf
76571
76572 It's the only user, so.
76573
76574commit e1611d8d55ade35bf8c2e1da7bc2dbf05b011301
76575Author: Daniel Stone <daniel@fooishbar.org>
76576Date: Thu Nov 27 13:34:05 2008 +1100
76577
76578 XKB: Be more verbose about XkbCopyKeymap failure in debug mode
76579
76580 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76581 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76582
76583commit 5ee504cae5f17d16ee7599beaa4b1944a9a8aee6
76584Author: Daniel Stone <daniel@fooishbar.org>
76585Date: Thu Nov 27 13:31:18 2008 +1100
76586
76587 XKB: Trying to copy to the same keymap is not fatal
76588
76589 Humour the user if they run XkbCopyKeymap(foo, foo).
76590
76591 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76592 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76593
76594commit d929ba6886c6f121b101c0bf0f7c63ef9d851a3e
76595Author: Daniel Stone <daniel@fooishbar.org>
76596Date: Fri Dec 26 18:32:41 2008 +1100
76597
76598 XFree86: Input: Fail NIDR if we can't enable devices
76599
76600 If we can't enable a device, bail out of NewInputDeviceRequest rather than
76601 blithely continuing. Also, be more verbose when initialization failed. Also,
76602 be more verbose when initialization failed. Also, be more verbose when
76603 initialization failed. Also, be more verbose when initialization failed.
76604
76605 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76606 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76607
76608commit 534669b376a6a703fbc97269d279a5418cf60c98
76609Author: Daniel Stone <daniel@fooishbar.org>
76610Date: Sun Jan 4 16:24:44 2009 +1100
76611
76612 XKB: Remove unnecessary prototype
76613
76614 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76615 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76616
76617commit 27ea1a7e4e7eca394e052791c64bc6be515e075e
76618Author: Daniel Stone <daniel@fooishbar.org>
76619Date: Sun Jan 4 16:40:46 2009 +1100
76620
76621 XKB: Only Xi events are processed
76622
76623 Core events aren't run through these functions, so don't bother testing
76624 for them.
76625
76626 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76627 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76628
76629commit 2762cafc321166ac48f80146720709b1011e6894
76630Author: Daniel Stone <daniel@fooishbar.org>
76631Date: Thu Nov 27 13:35:42 2008 +1100
76632
76633 XkbCopyKeymap: inputInfo.keyboard is not a special case
76634
76635 The device-walking code is still depressing, though.
76636
76637 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76638 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76639
76640commit 07c3bb922b8c1ae8fe28198f6c451430b836f21e
76641Author: Daniel Stone <daniel@fooishbar.org>
76642Date: Sun Jan 4 14:12:19 2009 +1100
76643
76644 XKB: Move XkbCopyKeymap definition to xkbsrv.h
76645
76646 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76647 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76648
76649commit 4f35a34f3cd3df3c8af129936d1808dcc063d8a7
76650Author: Peter Hutterer <peter.hutterer@redhat.com>
76651Date: Thu Aug 14 17:26:02 2008 +0930
76652
76653 dix: Remove traces of CoreDevicePrivateKey
76654
76655 Obsolete with MPX.
76656
76657 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
76658 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76659
76660commit 58a6b5b9eff25b1b9c6de239d6f5a952a31dfd15
76661Author: Peter Hutterer <peter.hutterer@redhat.com>
76662Date: Thu Aug 14 17:21:28 2008 +0930
76663
76664 Xi: don't care about CoreDevicePrivateKey when copying keys
76665
76666 If we get here, we must copy.
76667
76668 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
76669 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76670
76671commit 23862ede59a9ce11a06ec5151bde460fb836c603
76672Author: Daniel Stone <daniel@fooishbar.org>
76673Date: Sat Jan 10 14:36:16 2009 +0200
76674
76675 XKB: Allow build-time configuration of XKB defaults
76676
76677 Instead of hardcoding base/pc105/us, allow users to change the defaults at
76678 ./configure time. Change the default model to be evdev on Linux.
76679
76680 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
76681
76682commit 57a5eb2583bdf1069861fe2184de9739c0d0c7c1
76683Author: Colin Harrison <colin.harrison@virgin.net>
76684Date: Sun Jan 4 17:40:54 2009 +0000
76685
76686 Xming: Update .rc file to tidy About and Exit dialogs
76687
76688 Tidy up About and Exit dialogs, add keyboard accelerators
76689
76690 Copyright (C) Colin Harrison 2005-2008
76691 http://www.straightrunning.com/XmingNotes/
76692 http://sourceforge.net/projects/xming/
76693
76694 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76695
76696commit df3f701ef2d41ff4be4839f473ae71d33777bea3
76697Author: Colin Harrison <colin.harrison@virgin.net>
76698Date: Sun Nov 2 20:17:23 2008 +0000
76699
76700 Xming: Ensure we dont try to print a null value when displaying xkb configuration
76701
76702 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76703
76704commit c612f0df45d388c98108e1e0c80afdfae05e338e
76705Author: Colin Harrison <colin.harrison@virgin.net>
76706Date: Sun Nov 2 20:46:41 2008 +0000
76707
76708 Xming: Add styles keyword and attributes.
76709
76710 Add handling for style keyword and atttributes in .XWinrc
76711 Update man page to document these additions
76712
76713 Copyright (C) Colin Harrison 2005-2008
76714 http://www.straightrunning.com/XmingNotes/
76715 http://sourceforge.net/projects/xming/
76716
76717 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76718
76719commit 847a91ad2e8c9b0992c09f0ec38466885d6d25d4
76720Author: Colin Harrison <colin.harrison@virgin.net>
76721Date: Sun Jan 4 19:57:56 2009 +0000
76722
76723 Xming: Notify X when the keyboard focus is lost to a pure Windows window in -multiwindow mode.
76724
76725 Fix internal WM to notify X when the keyboard focus is lost to a pure Windows window in -multiwindow mode.
76726
76727 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76728
76729commit 6c3e7e4d0568ecb7d10b814593bbbd817b7d2aab
76730Author: Colin Harrison <colin.harrison@virgin.net>
76731Date: Sun Jan 4 18:28:34 2009 +0000
76732
76733 Xming: Only allow WM_MOUSEWHEEL messages to act on the client area of a focused window.
76734
76735 Fix internal WM so it only allows WM_MOUSEWHEEL messages to act on the client area of a focused window.
76736
76737 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76738
76739commit d84e0e387764236c7e97509baf4606d96cfc19d4
76740Author: Colin Harrison <colin.harrison@virgin.net>
76741Date: Sun Jan 4 18:20:44 2009 +0000
76742
76743 Xming: Correctly parent XA_WM_TRANSIENT_FOR windows in -multiwindow mode when a windows window is created.
76744
76745 Fix internal WM to correctly parent XA_WM_TRANSIENT_FOR windows in -multiwindow mode when a windows window is created,
76746 and to de-iconize parent windows when a child window acquires focus.
76747
76748 XXX: Perhaps we should also shuffle parent(s) forward through Z-order when a child acquires focus?
76749
76750 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76751
76752commit 8d82d806ee38d45ec7cdbf8b1f671ec1e4437567
76753Author: Colin Harrison <colin.harrison@virgin.net>
76754Date: Sun Nov 2 20:46:38 2008 +0000
76755
76756 Xming: Correct the way display and screen number is reported in window titles and tooltips
76757
76758 Correct the tooltip text for the toolbar X icon to be strictly correct, 'display-number:screen' should be ':display-number.screen'.
76759 Also for the default window title.
76760 Adjust the style of the Windows title in XDMCP mode from 'Xming - hostname' to 'hostname:display-number.screen'.
76761
76762 Copyright (C) Colin Harrison 2005-2008
76763 http://www.straightrunning.com/XmingNotes/
76764 http://sourceforge.net/projects/xming/
76765
76766 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76767
76768commit 656f92e247f2f257105daf1e0e6998e1b3d10eba
76769Author: Jon TURNEY <jon.turney@dronecode.org.uk>
76770Date: Sun Nov 2 20:17:02 2008 +0000
76771
76772 Cygwin/X: should also use GetTickCount(), just like Xming
76773
76774 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76775
76776commit 7f781e780ee2b261c5cb688a8e6c5dc1a53bdb6f
76777Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
76778Date: Wed Dec 17 19:13:30 2008 +0000
76779
76780 Cygwin/X: Cygwin doesn't have RTLD_LOCAL
76781
76782 RTLD_LOCAL is not defined on Cygwin
76783
76784 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76785
76786commit 5630ff809676ab6b4833757060e48e201e8768a5
76787Author: Dan Nicholson <dbn.lists@gmail.com>
76788Date: Mon Dec 22 13:53:43 2008 -0800
76789
76790 Ignore generated dmx manpages
76791
76792 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
76793
76794commit 64f924fbf513b5f91b3f2bcf5767f6fe927d3f7a
76795Author: Dan Nicholson <dbn.lists@gmail.com>
76796Date: Mon Dec 22 13:43:30 2008 -0800
76797
76798 Kill off --with-mesa-source completely
76799
76800 There were a few spots left in the source that were using the
76801 --with-mesa-source defined headers or the now removed $(top_srcdir)/GL
76802 directory. These aren't needed anymore as all the necessary source for
76803 GLX is in $(top_srcdir)/glx.
76804
76805 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
76806
76807commit e1c8ee215797fa8c6a3a5421bb6f99b58064248b
76808Author: Dan Nicholson <dbn.lists@gmail.com>
76809Date: Sun Dec 14 14:18:58 2008 -0800
76810
76811 Don't reuse PKG_CHECK_MODULES identifiers
76812
76813 Using GL for the PKG_CHECK_MODULES identifier multiple times means only
76814 the first call will actually be used. Later calls will be skipped due to
76815 GL_CFLAGS and GL_LIBS already being set. This changes DRI to using a
76816 different identifier and DMX to just reusing GL_CFLAGS.
76817
76818 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
76819
76820commit ca46c01869768a8661a9d8a71493ed5f0760a8ab
76821Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
76822Date: Fri Jan 16 13:58:20 2009 -0800
76823
76824 XQuartz: mieq: Wait for the server to finish initializing before letting other threads mieqEnqueue
76825
76826 Avoid possible race condition whereby one thread might call mieqEnqueue before InitAndStartDevices finishes
76827 (cherry picked from commit 94e417ac87a98cd5c6bf2d7c495d702748398931)
76828
76829commit b33905234025f005819c7e2acd653a3a0ecfeb82
76830Author: Peter Hutterer <peter.hutterer@who-t.net>
76831Date: Fri Jan 9 14:32:14 2009 +1000
76832
76833 xfree86: always force RAW mode under linux.
76834
76835 The previous check for AEI on left us with the possibility that AEI is forced
76836 off in the config, but devices are added through evdev nonetheless. A keyboard
76837 added this way can CTRL+C the server. Even when we use kbd, we can set the
76838 mode to RAW, so it's safer alround to to so.
76839
76840 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76841
76842commit 646add4eb47c01c045ef64405510a08ecfb5bfd6
76843Author: Peter Hutterer <peter.hutterer@who-t.net>
76844Date: Fri Jan 16 16:04:07 2009 +1000
76845
76846 dix: remove XineramaCheckMotion, merge into CheckMotion.
76847
76848 For 5 lines difference, we don't need to duplicate the whole function. really.
76849
76850commit e76f904abe530df9ddaa09510088c545298b0b91
76851Author: Peter Hutterer <peter.hutterer@who-t.net>
76852Date: Fri Jan 16 15:53:50 2009 +1000
76853
76854 dix: remove "notyet" conditional compilation.
76855
76856 If it's "notyet" for 3 years, then "yet" won't arive anytime soon.
76857
76858commit 5c41b3f9c8b72f0115019237f812d74610b82c24
76859Author: George Staplin <gstaplin@apple.com>
76860Date: Mon Jan 12 19:30:43 2009 -0700
76861
76862 XQuartz: xpr: Cleanup some of the code and possibly fix part of the GLX Pixmap problem.
76863
76864 Split DRICreateSurface into 3 functions.
76865
76866 Make CreateSurfaceForPixmap use xp_configure_surface. I suspect this is partly why
76867 GLXPixmaps never worked.
76868
76869 It will require some more work and thoughts for pbuffers, unless we fake those with
76870 pixmaps and surfaces.
76871 (cherry picked from commit 9cf264e67744262b9f45079e6cd752eb3e3b0e08)
76872
76873commit c5d52d4c37ebec92cb668b0653bf3de2bdfc5782
76874Author: George Staplin <gstaplin@apple.com>
76875Date: Mon Jan 12 14:14:44 2009 -0700
76876
76877 XQuartz: GL: Make indirect.c build and work in the 1.6 branch.
76878
76879 Reorder some header files and provide some types earlier on.
76880
76881 Remove the static __GLXextensionInfo __glDDXExtensionInfo; that isn't
76882 used in 1.4 or 1.5, and seems to have been removed from 1.6. Remove
76883 the data structures associated with that too.
76884
76885 Remove __glXAquaDrawableResize. The GLX structure doesn't use
76886 it anymore, and the Apple version did nothing useful before.
76887
76888 __glXAquaDrawableSwapBuffers(): base no longer contains a drawGlxc
76889 member. Now provide the the Apple/Aqua context in the __GLXAquaDrawable
76890 struct. Add the context member to the __GLXAquaDrawable struct.
76891
76892 Remove the fallback configs. They aren't used in 1.4 or 1.5 either.
76893
76894 Remove init_visuals(). It's not used in 1.4 or 1.5 either.
76895
76896 In the drawable constructor initialize the state properly, including
76897 the new context member.
76898
76899 Remove glAquaResetExtension() -- it's not used anymore.
76900
76901 This has been tested remotely and proven to work with glxgears,
76902 fire, various texture programs I wrote, and various Mesa demos.
76903 (cherry picked from commit d514152195452ae11ec7769c76489651620ad380)
76904
76905commit 1c504dff8a84804888f18ee9b184f0c4584185d7
76906Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
76907Date: Thu Jan 15 21:03:27 2009 -0800
76908
76909 XQuartz: Fix builddir != srcdir issues and undef _XSERVER64 where appropriate on fat binary compilation
76910
76911commit 717a961528ec69a6e630d536e15568670e0b398a
76912Author: Thomas Jaeger <ThJaeger@gmail.com>
76913Date: Sat Dec 20 16:17:02 2008 +0100
76914
76915 Don't release grabs unless all buttons are up
76916
76917 Previously, only buttons <= 5 would count here, but the core protocol
76918 allows for 255 buttons.
76919
76920 http://lists.freedesktop.org/archives/xorg/2009-January/042092.html
76921
76922 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76923
76924commit f7f85f696570541e2dd43462675de9e6ee46f545
76925Author: Thomas Jaeger <ThJaeger@gmail.com>
76926Date: Mon Dec 22 00:55:09 2008 +0100
76927
76928 Count the number of logically down buttons in buttonsDown
76929
76930 This fixes the following bug. Assuming your window manager grabs
76931 Alt+Button1 to move windows, map Button3 to 0 via XSetPointerMapping,
76932 then press the physical button 3 (this shouldn't have any effect), press
76933 Alt and then button 1. The press event is delivered to the application
76934 instead of firing the grab.
76935
76936 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
76937
76938commit 6f065a26db62aaf1d7aab9f9a3d0e6cd09d1c1dd
76939Author: Colin Harrison <colin.harrison@virgin.net>
76940Date: Thu Jan 15 21:42:26 2009 +0000
76941
76942 Xming: Reduce MOUSE_POLLING_INTERVAL
76943
76944 Reduce MOUSE_POLLING_INTERVAL from 500ms to 50ms
76945
76946 In -multiwindow mode, this determines how quickly X windows can track the mouse
76947 when it is outside any X window... Move the cursor around an xeyes, and you will
76948 see what this timing effects.
76949
76950commit c745db1674c3cb55249c9eb6e74939b74c42409c
76951Author: Jon TURNEY <jon.turney@dronecode.org.uk>
76952Date: Mon Jan 5 13:52:45 2009 +0000
76953
76954 GLX: Avoid a crash when we have an uninitialized GL context
76955
76956 If the GL dispatch table pointer points to glapi_noop_table,
76957 (due to some kind of GL initialization failure), DoGetString(GL_VERSION)
76958 (for example as invoked by glxinfo) will crash as it tries to
76959 do atof() on the null pointer returned by the noop dispatch function
76960
76961 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
76962
76963commit e30032d0bbbf0bf71e1b53a4ac388f3bd6f68e53
76964Author: Peter Hutterer <peter.hutterer@redhat.com>
76965Date: Mon Dec 1 21:14:01 2008 +1000
76966
76967 Xext: clean up XGE macros.
76968
76969 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
76970
76971commit 1ae529bef50a8ccc4a8a264ed120aeae544b21e7
76972Author: Peter Hutterer <peter.hutterer@who-t.net>
76973Date: Wed Jan 14 14:56:10 2009 +1000
76974
76975 dix: replace manual check for event types with IsPointerEvent()
76976
76977commit 1d54479cb3c8b4f75b7564f8b5e1c5da940b20f4
76978Author: Simon Thum <simon.thum@gmx.de>
76979Date: Tue Nov 25 14:12:02 2008 +1000
76980
76981 xfree86: init properties when ptraccel configuration is found.
76982
76983 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
76984 Signed-off-by: Simon Thum <simon.thum@gmx.de>
76985
76986commit 19275ea8e9dd93d5e61906943706dfe49003dd9c
76987Author: Simon Thum <simon.thum@gmx.de>
76988Date: Wed Nov 19 16:01:21 2008 +1000
76989
76990 dix: add property support for pointer acceleration.
76991
76992 Note: properties don't need to be cleaned up, the DIX does it for us anyway.
76993 Data that is stored in properties is cleaned up by the property system.
76994 Handlers, etc. don't need to be unregistered while cleaning up, as they get
76995 deleted when the device is removed anyway.
76996
76997 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
76998 Signed-off-by: Simon Thum <simon.thum@gmx.de>
76999
77000commit d645721170b1196e5064b397cfbffd1da8c79bb1
77001Author: Peter Hutterer <peter.hutterer@who-t.net>
77002Date: Thu Jan 8 11:53:30 2009 +1000
77003
77004 mi: ensure chained button mappings from SD -> MD (#19282)
77005
77006 After copying the master event, flip the detail field to the mapped button of
77007 the SD, not the physical button. This way if the SD has a mapping 1:3 and the
77008 MD has a mapping of 3:4, a press on button 1 on the SD results in a core event
77009 on button 4.
77010
77011 X.Org Bug 19282 <http://bugs.freedesktop.org/show_bug.cgi?id=19282>
77012
77013 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77014
77015commit 639f289dcdbe00a516820f573c01a8339e120ed4
77016Author: Michel Dänzer <daenzer@vmware.com>
77017Date: Tue Jan 13 13:05:32 2009 +0100
77018
77019 EXA: Declare glyph cache picture as component-alpha when necessary.
77020
77021 Without this, rendering component-alpha glyphs may break without a mask.
77022
77023 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=19233 .
77024
77025commit f4e00b783d3f824058a5f140e6462142688f5116
77026Author: Michel Dänzer <daenzer@vmware.com>
77027Date: Tue Jan 13 11:03:27 2009 +0100
77028
77029 Fix build with --enable-visibility --disable-xdmcp.
77030
77031commit d36adf52a2b2711d22b11105f7bd907d4493fb9b
77032Author: Peter Hutterer <peter.hutterer@who-t.net>
77033Date: Mon Jan 12 15:29:36 2009 +1000
77034
77035 dix: fix WarpPointer calls for devices with custom valuator ranges (#19297)
77036
77037 If the MD's lastSlave was a devices with custom axes ranges, then a
77038 WarpPointer would position the cursor at the wrong location. A WarpPointer
77039 request provides screen coordinates and these coordinates were scaled to the
77040 device range before warping.
77041
77042 This patch consists of two parts:
77043 1) in the WarpPointer handling, get the lastSlave and post the event through
77044 this device.
77045 2) assume that WarpPointer coordinates are always in screen coordinates and
77046 scale them to device coordinates in GPE before continuing. Note that this
77047 breaks device-coordinate based XWarpDevicePointer calls (for which the spec
77048 isn't nailed down yet anyway) until a better solution is found.
77049
77050 X.Org Bug 19297 <http://bugs.freedesktop.org/show_bug.cgi?id=19297>
77051
77052 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77053
77054commit 488d45295105daf10ccd17ca93ae6a6f4d0104f1
77055Author: Peter Hutterer <peter.hutterer@who-t.net>
77056Date: Mon Jan 12 16:16:24 2009 +1000
77057
77058 dix: EnqueueEvent and PlayReleasedEvent need to handle DeviceMotionNotifies
77059
77060 No MotionNotify events in the processing anymore, so let's have them treat DMN
77061 instead.
77062
77063 Reported by Thomas Jaeger.
77064
77065commit 1e470645983ad18fa38c0885d674bc9d1dc1e560
77066Author: Julien Cristau <jcristau@debian.org>
77067Date: Tue Jan 13 00:12:56 2009 +0100
77068
77069 require randrproto 1.2.99.3
77070
77071commit a48c81dcdf569a3f634ac23e08d2491354de6a36
77072Author: Peter Hutterer <peter.hutterer@who-t.net>
77073Date: Fri Dec 5 16:24:57 2008 +1000
77074
77075 Xi: add XATOM_FLOAT to server-defined properties.
77076
77077 This property is used to denote type float for input properties. Such
77078 properties can be accessed easily through the XIPropToFloat() function.
77079
77080 Code originally written by Simon Thum.
77081
77082 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
77083
77084commit 669f6810af9a89187d6149841925fe765f3988ff
77085Author: Peter Hutterer <peter.hutterer@redhat.com>
77086Date: Wed Nov 19 15:50:57 2008 +1000
77087
77088 Xi: add XIPropToInt() auxiliary function.
77089
77090 Converts an XIPropertyValuePtr to an integer, provided that type and format is
77091 right.
77092
77093 Code originally written by Simon Thum.
77094
77095 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
77096
77097commit 56efbc0986e782da45addb05ece9f456d41d7a90
77098Author: Peter Hutterer <peter.hutterer@who-t.net>
77099Date: Fri Jan 9 13:46:20 2009 +1000
77100
77101 dix: drop x/y back into last.valuators before updating the history (#19285)
77102
77103 positionSprite needs to scale to screen coordinates and in the process of
77104 doing so alters dev->last.valuators[0:1]. Drop the real coordinates back after
77105 finishing and before updating the motion history. This way, we don't push the
77106 screen coordinates into the motion history.
77107
77108 X.Org Bug 19285 <http://bugs.freedesktop.org/show_bug.cgi?id=19285>
77109
77110commit 67710ad1d172cc4d53e28e5cc7dab2efb57a9a85
77111Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77112Date: Sun Jan 11 16:34:36 2009 -0800
77113
77114 XQuartz: Don't need explicit Activate/EnableDevice in InitInput
77115 (cherry picked from commit b1d29784410b3b93037e5636f336ba608d8ad6e3)
77116
77117commit dde5c591be74139ec6f2a90a3fdc7904d7342554
77118Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77119Date: Sun Jan 11 16:10:45 2009 -0800
77120
77121 mieq: Avoid possible race condition whereby one thread might call mieqEnqueue before InitAndStartDevices finishes
77122
77123 This is more of a hack around the problem. This is something that will need to be addressed in a
77124 more structured manner with the multi threaded input efforts.
77125 (cherry picked from commit 3c596c061e75848cfa76dd9259c23a3f3a67444c)
77126
77127commit 0720816aea20971ee94b5df3faa9cb9f16e93f82
77128Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77129Date: Sun Jan 11 14:49:39 2009 -0800
77130
77131 XQuartz: Only call DarwinUpdateModKeys when needed
77132
77133 Previously, we were calling it on almost every itteration through sendX11Event
77134 (cherry picked from commit 6461729647ff4441d80811e73f0c0d2f108f2700)
77135
77136commit c8bf582db8111f71b9250a602a9d5aa7bd828136
77137Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77138Date: Thu Oct 23 11:41:22 2008 -0700
77139
77140 XQuartz: Comment explaining the 0x10 flag for [e data2]
77141 (cherry picked from commit 3c695280641c1205b97d3bb9f1d5e15a19cfa45f)
77142
77143commit 5339c22a8ca8bce6cd2b51e97c904292e7c40e74
77144Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77145Date: Sun Jan 11 01:56:45 2009 -0800
77146
77147 XQuartz: Add locking to make mieq thread safe on OSX
77148 (cherry picked from commit 7a8d2266861e74176b5310b83652a9c10a170494)
77149
77150commit 1beff89810cd193e45d437846929b3938cee6d96
77151Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77152Date: Sun Jan 11 01:55:26 2009 -0800
77153
77154 XQuartz: misc 1.6 updates (still --disable-glx)
77155 rlAccel is not longer compatable, and it's not worth fixing
77156 Don't override DeviceCursorInitialize with a noop
77157 Don't do a SwitchCoreKeyboard (which wasn't even needed in the first place)
77158 (cherry picked from commit c137f681680e1d04b1513a8be68aeda4d1c56fd5)
77159
77160commit 2d917cfe56af1e74205f80ddbaddfb5c6ce82f0a
77161Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77162Date: Sat Jan 10 01:39:08 2009 -0800
77163
77164 Apple: Don't use DRI2
77165 (cherry picked from commit a1d35cee5907a76977ee43c49cb55c8f411c9794)
77166
77167commit 7a8abb9c8554d7967d912dcd427804f65906dc5c
77168Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77169Date: Tue Jan 6 15:32:44 2009 -0800
77170
77171 XQuartz: Bundle version 2.4.0
77172 (cherry picked from commit 3a3ccf5354e4275abe9d01f41a92602df5d690dc)
77173
77174commit 63ff7d236f67e159943312e1a551fcf5bb3f463a
77175Author: Julien Cristau <jcristau@debian.org>
77176Date: Sun Jan 11 08:45:24 2009 +0100
77177
77178 os: don't mix declarations and code
77179
77180commit f43a3c61ab1eab8aecdc0e4bdee9f3bae2cab234
77181Author: Julien Cristau <jcristau@debian.org>
77182Date: Sun Jan 11 08:44:53 2009 +0100
77183
77184 os: backtrace() returns int, not size_t
77185
77186commit 5321ae76cffc5fcfe4bebcfb10e08eecaf640503
77187Author: Julien Cristau <jcristau@debian.org>
77188Date: Sun Jan 11 08:44:29 2009 +0100
77189
77190 os: ANSI cleanups
77191
77192commit 47438a2161a92fec08081d2de28966cf8d31654a
77193Author: Julien Cristau <jcristau@debian.org>
77194Date: Sun Jan 11 08:43:48 2009 +0100
77195
77196 randr: RRSetPrimaryOutput can be static
77197
77198commit 7f82114b1dc738f98b9f22f6e61d6d5191383b67
77199Author: Julien Cristau <jcristau@debian.org>
77200Date: Sun Jan 11 08:43:35 2009 +0100
77201
77202 xkb: ANSI cleanup
77203
77204commit 46f3c1b937002f9a860878b359dcf5756a82dcf2
77205Author: Julien Cristau <jcristau@debian.org>
77206Date: Sun Jan 11 08:42:57 2009 +0100
77207
77208 mi: don't mix declarations and code
77209
77210commit 5aa9483d3ee306a79a9235610ce6b2754d697585
77211Author: Julien Cristau <jcristau@debian.org>
77212Date: Sun Jan 11 08:42:28 2009 +0100
77213
77214 mi: ANSI cleanups
77215
77216commit 359ed32c9b013ed88e14c99aa7fd7a617a85e2c5
77217Author: Julien Cristau <jcristau@debian.org>
77218Date: Sun Jan 11 08:40:58 2009 +0100
77219
77220 xnest: only define ddxBeforeReset if needed
77221
77222commit 9651cd8d7ac26635b6fb1826806ac9f4e2bddb63
77223Author: Julien Cristau <jcristau@debian.org>
77224Date: Sun Jan 11 08:40:47 2009 +0100
77225
77226 xnest: ANSI cleanups
77227
77228commit 31439195c51f183f5a8989a8d4f95b3e6326eda0
77229Author: Julien Cristau <jcristau@debian.org>
77230Date: Sun Jan 11 08:37:06 2009 +0100
77231
77232 xfree86: linuxPci.c needs a prototype for xf86AccResFromOS
77233
77234commit a00dbd0015845e08ef2af9480cc3833819cf86f5
77235Author: Julien Cristau <jcristau@debian.org>
77236Date: Sun Jan 11 08:36:16 2009 +0100
77237
77238 xfree86: use %zu to format size_t arguments
77239
77240commit 17c8cb50990a9ad20bbd3f8ce14c149dac9d72ad
77241Author: Julien Cristau <jcristau@debian.org>
77242Date: Sun Jan 11 08:34:56 2009 +0100
77243
77244 xfree86: don't mix declarations and code
77245
77246commit aec4c0caca41a06832536ba82ba5bcdccf4a0aea
77247Author: Julien Cristau <jcristau@debian.org>
77248Date: Sun Jan 11 08:33:18 2009 +0100
77249
77250 xfree86: ANSI cleanups
77251
77252commit dba3b1944d04232f88bb310277639f2c1adcbe08
77253Author: Julien Cristau <jcristau@debian.org>
77254Date: Sun Jan 11 08:28:23 2009 +0100
77255
77256 Xvfb: ANSI cleanups
77257
77258commit a2a760e33c17e211650b1f53c05a89c013916cf3
77259Author: Julien Cristau <jcristau@debian.org>
77260Date: Sun Jan 11 08:25:39 2009 +0100
77261
77262 dix: don't mix declarations and code
77263
77264commit 466dddbb83aa37ae93fb32976d51ecaa947c3616
77265Author: Julien Cristau <jcristau@debian.org>
77266Date: Sun Jan 11 08:25:17 2009 +0100
77267
77268 dix: fix cast from pointer to integer
77269
77270commit 54332e4732577dc49e440d9c9f8dd7d791f03f8e
77271Author: Julien Cristau <jcristau@debian.org>
77272Date: Sun Jan 11 08:20:06 2009 +0100
77273
77274 dix: ANSI cleanups
77275
77276commit ef9902652b8a1c716ea3851565d6e2a98bd30eb0
77277Author: Julien Cristau <jcristau@debian.org>
77278Date: Sun Jan 11 08:18:56 2009 +0100
77279
77280 config: ANSI cleanups
77281
77282commit e64473b14e091e4b75e6ac2bf9a35e9aaa907258
77283Author: Julien Cristau <jcristau@debian.org>
77284Date: Sun Jan 11 08:17:41 2009 +0100
77285
77286 Xi: fix missing declaration of XkbSetRulesDflts
77287
77288commit 6e20a4ae731b7bc72266e5809faa0ecfde2bdfe4
77289Author: Julien Cristau <jcristau@debian.org>
77290Date: Sun Jan 11 08:16:55 2009 +0100
77291
77292 xsync: make SyncAlarmCounterDestroyed static
77293
77294commit 292311b421e53ead53c7df67d735a3d6d3e5b515
77295Author: Julien Cristau <jcristau@debian.org>
77296Date: Sun Jan 11 08:16:12 2009 +0100
77297
77298 Xext: ANSI cleanups
77299
77300commit 7a05c8b1e70680ddd3b3e09ad448788f8d70a428
77301Author: Julien Cristau <jcristau@debian.org>
77302Date: Tue Dec 23 13:08:49 2008 +0100
77303
77304 xfree86/linux: fix log flood on acpid open error
77305
77306 When we can't open the acpid socket, warn once, not once every second
77307
77308commit 0d8184de591c75794083c8a8ca3fb8459e9394bc
77309Author: George Staplin <gstaplin@apple.com>
77310Date: Fri Jan 9 16:08:26 2009 -0700
77311
77312 XQuartz: GL: Set the __GLXconfig renderType to GLX_RGBA_BIT.
77313 (cherry picked from commit 3c14546f58f8a138fe67c9cacc3bd0b7fa90c29a)
77314
77315commit 9c3975d63e129ad7230cae585f08a962d45fb861
77316Author: George Staplin <gstaplin@apple.com>
77317Date: Fri Jan 9 15:53:39 2009 -0700
77318
77319 XQuartz: GL: Make many more fbconfigs and visuals available for the 1.5 branch.
77320
77321 Use a __GLXconfig linked list struct to store the configurations for the
77322 fbconfigs and visuals in a pGlxScreen.
77323
77324 Also, remove the __GLXvisualConfig/GlxSetVisualConfigs code that isn't used
77325 anymore. There is more code we can remove later, but I want to do that in
77326 separate commits.
77327 (cherry picked from commit 94162b0f8a25267aca280d25e216cc6bde47da6e)
77328
77329commit a68373994f4957a73f4eab52c90ebcfb96f285f8
77330Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77331Date: Wed Jan 7 18:27:21 2009 -0800
77332
77333 XQuartz: Update padding in appledristr. This shouldn't change how the struct ends up in memory, but should make it more obvious for human eyes
77334 (cherry picked from commit f7b6c1907c326a1befd8b5c8a1d311d551e8670f)
77335
77336commit 5f3188228eb988bd8f08b62c84f98a8ff66ee283
77337Author: Xiang, Haihao <haihao.xiang@intel.com>
77338Date: Tue Jan 6 10:03:19 2009 +0800
77339
77340 avoid a potential endless loop.
77341
77342 Previously it is possible that creating rotation data, then cleaning
77343 up and creating again so that pScreen->BlockHandler and
77344 xf86_config->BlockHandler all point to xf86RotateBlockHandler.
77345 See bug #19343.
77346
77347commit b2756a71a432f7cf7c870a48676c98625512558d
77348Author: Thomas Jaeger <ThJaeger@gmail.com>
77349Date: Fri Jan 9 02:02:38 2009 -0500
77350
77351 Xext: Send out correct events in ProcXTestFakeInput
77352
77353 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77354
77355commit 9bf77569cc8f9b0cc25c99487672d375d7c32f99
77356Author: Peter Hutterer <peter.hutterer@who-t.net>
77357Date: Thu Jan 8 16:32:24 2009 +1000
77358
77359 Xi: call CheckMotion for floating SDs too.
77360
77361 CheckMotion updates the sprite position so we need to call it for all devices
77362 that have a sprite - including floating SDs (which have an invisible sprite).
77363
77364commit 83eb863e3cfcd62549b836f3195107d8908b36d0
77365Author: Peter Hutterer <peter.hutterer@who-t.net>
77366Date: Fri Jan 9 11:39:02 2009 +1000
77367
77368 Xext: don't accept DeviceValuator if the dev doesn't have valuators (in xtest)
77369
77370 Reported by Chris Ball.
77371
77372commit a3f9e887d9ff5fdee6c35ccc0f7b324723a83a46
77373Author: Peter Hutterer <peter.hutterer@who-t.net>
77374Date: Fri Jan 9 11:40:18 2009 +1000
77375
77376 dix: don't accept Button 0 presses in GPE.
77377
77378commit 0c6987df3b9b3a37d201d740d8248c326449835e
77379Author: Adam Jackson <ajax@redhat.com>
77380Date: Tue Jun 17 16:10:51 2008 -0400
77381
77382 XAA: Disable offscreen pixmaps by default.
77383
77384 Say Option "XaaOffscreenPixmaps" to turn them back on.
77385
77386 Apropos of bugs #13795 and #15098. But this still isn't correct.
77387
77388commit eb2d7b3d700952ba88c77deacf687b251300e660
77389Author: Peter Hutterer <peter.hutterer@who-t.net>
77390Date: Fri Dec 19 21:05:47 2008 +1000
77391
77392 dix: move focus handling into enterleave.c.
77393
77394 This commit moves the focus handling from events.c into enterleave.c and
77395 implements a model similar to the core enter/leave model.
77396 For a full description of the model, see:
77397 http://lists.freedesktop.org/archives/xorg/2008-December/041740.html
77398
77399 This commit also gets rid of the focusinout array in the WindowRec, ditching
77400 it in favour of a local array that keeps the current focus window for each
77401 device.
77402
77403 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77404
77405commit 673eb23aac578dcdc04e2a99d1fa5c2987eb58b8
77406Author: Peter Hutterer <peter.hutterer@who-t.net>
77407Date: Fri Dec 19 21:05:09 2008 +1000
77408
77409 dix: add a few auxiliary functions for the updated focus model.
77410
77411 SetFocusIn and SetFocusOut, including the static array to keep all focus
77412 windows.
77413
77414 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77415
77416commit 38b28dcadd0990cb43f50db4300eebb8f044db96
77417Author: Peter Hutterer <peter.hutterer@who-t.net>
77418Date: Thu Dec 18 15:55:11 2008 +1000
77419
77420 dix: reduce FirstPointerChild complexity
77421
77422 Instead of keeping a flag on each window for the devices that are in this
77423 window, keep a local array that holds the current pointer window for each
77424 device. Benefit: searching for the first descendant of a pointer is a simple
77425 run through the array.
77426
77427 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77428
77429commit 72ad4a85cc0ffe60a90011d65ef718d5852beae4
77430Author: Peter Hutterer <peter.hutterer@who-t.net>
77431Date: Thu Dec 18 15:43:18 2008 +1000
77432
77433 dix: remove now unused "exclude" parameter from FirstPointerChild
77434
77435 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77436
77437commit 0aa4460c3b4f9bb17d5412d76fa8c4c501132429
77438Author: Peter Hutterer <peter.hutterer@who-t.net>
77439Date: Wed Dec 17 16:14:41 2008 +1000
77440
77441 dix: re-implement enter/leave model.
77442
77443 The old model was implemented based on a misunderstanding of NotifyVirtual and
77444 NotifyNonlinearVirtual events. It became complicated and was broken in some
77445 places [1]. This patch wipes this model completely.
77446
77447 A much simplified implementation is provided instead. Rather than a top-down
77448 approach ("we have a tree of windows, which ones need to get which event")
77449 this one uses a step-by-step approach. For each window W between A and B
77450 determine the pointer window P as perceived by this window and determine the
77451 event type based on this information. This is in-line with the model described
77452 by Owen Taylor [2].
77453
77454 [1] http://lists.freedesktop.org/archives/xorg/2008-December/041559.html
77455 [2] http://lists.freedesktop.org/archives/xorg/2008-August/037606.html
77456
77457commit f8ecc2f08fc375b14d60a74e2fdd7830bfdcd74f
77458Author: Christian Beier <beier@informatik.hu-berlin.de>
77459Date: Tue Jan 6 11:31:03 2009 +1000
77460
77461 Xi: fix typo in WarpDevicePointer handling.
77462
77463 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77464
77465commit 3d549438c29004d78032ecc50ab45ca0e3f49623
77466Author: Thomas Jaeger <thjaeger@gmail.com>
77467Date: Mon Jan 5 01:26:42 2009 -0500
77468
77469 Don't alter device button maps in DoSetPointerMapping
77470
77471 Currently, if a device map differs from the core pointer map, then the
77472 request may return MappingBusy, even though all the affected core
77473 buttons are in the up state.
77474
77475 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77476
77477commit b72c6b083baeadfd7b18b6025df054be502d6e28
77478Author: Peter Hutterer <peter.hutterer@who-t.net>
77479Date: Tue Jan 6 09:01:48 2009 +1000
77480
77481 dix: remove inputInfo.keyboard reference in QueryPointer handling.
77482
77483 PickKeyboard() is to be used instead.
77484
77485commit 30d2cfcd3851870178d62e5067211aa36f87fbd2
77486Author: Kim Woelders <kim@woelders.dk>
77487Date: Mon Jan 5 19:35:18 2009 +0100
77488
77489 dix: Fix handling of do_not_propagate_mask window attribute.
77490
77491 This was broken in 32aa252e988be8cbfd4f7e373fb7b7736ef1f5f2.
77492
77493 Signed-off-by: Kim Woelders <kim@woelders.dk>
77494 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77495
77496commit 515ce3e4ba42605a1ee9979e8bb5acd3cf6470a3
77497Author: Peter Hutterer <peter.hutterer@who-t.net>
77498Date: Mon Jan 5 15:48:45 2009 +1000
77499
77500 xkb: fix typo - missing negation when checking button state.
77501
77502 Introduced with a85f0d6b98237d8a196de624207acf1983a1859a.
77503
77504 Reported by Thomas Jaeger.
77505
77506 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77507
77508commit 9a313fb03153eb764aab3e9cb9988b6603541637
77509Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
77510Date: Wed Jan 7 23:07:01 2009 -0200
77511
77512 Don't call free, use xfree macro.
77513
77514 There are several other direct calls to free, check
77515 % egrep '\<free\(' `find . -name \*.c`
77516 but they are free'ing memory from explicit malloc calls.
77517 This one was not intended, and corrected, so that it would
77518 both, follow the conventions everywhere (and work on some
77519 libc that doesn't like free(0)), and make it easier to use
77520 malloc wrappers.
77521
77522commit 49b93df8a3002db7196aa3fc1fd8dca1c12a55d6
77523Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
77524Date: Wed Jan 7 19:37:03 2009 -0200
77525
77526 Default to use standard bitmap fonts, with builtins as fallback
77527
77528 The builtin-fonts configure option was removed, as it at best should
77529 have been a runtime option. Instead, now it always register all "font
77530 path element" backends, and adds built-ins fonts at the end of the
77531 default font path.
77532 This should be a more reasonable solution, to "correct" the most
77533 common Xorg FAQ (could not open default font 'fixed'), and also don't
77534 break by default applications that use only the standard/historical
77535 X Font rendering.
77536
77537commit 4f004231871c830e16c4525cfaab183fad0aee86
77538Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77539Date: Wed Jan 7 10:55:08 2009 -0800
77540
77541 XQuartz: Fat binary buildfixes
77542
77543 Let Xarch.h set X_BYTE_ORDER for us.
77544 Determine _XSERVER64 at build time.
77545 (cherry picked from commit f54736c2b0b537c8322ecb2db8e9cee5eea68165)
77546
77547commit fbf44d85bd59bfa629193d6494b16812a7343cab
77548Author: Jon TURNEY <jon.turney@dronecode.org.uk>
77549Date: Mon Jan 5 23:18:11 2009 +0000
77550
77551 Cygwin/X: Tidy up an unused #define
77552
77553 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
77554
77555commit 1badeb78b0b974a381e563e3d4f4cf37debd4266
77556Author: Rémi Cardona <remi@gentoo.org>
77557Date: Wed Jan 7 09:31:17 2009 -0500
77558
77559 Include Xmd.h from edid.h
77560
77561commit fc14d52297f8f408693ed867ab649ea5b210b0b3
77562Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77563Date: Sun Nov 16 15:35:17 2008 -0800
77564
77565 XQuartz: Use AvailabilityMacros.h instead of availability.h for Tiger
77566 (cherry picked from commit 91f3c3435716195e6d606c8e2f6b0de52eb5972d)
77567
77568commit 4c326c4d342bede0de57e71c1431f56188f812ce
77569Author: Keith Packard <keithp@keithp.com>
77570Date: Tue Jan 6 08:36:39 2009 -0800
77571
77572 Make cvt complain about invalid arguments more often.
77573
77574 cvt doesn't correctly parse 'cvt -vr 1920 1080'. This patch doesn't fix
77575 that, but it does at least generate an error message now.
77576
77577 Signed-off-by: Keith Packard <keithp@keithp.com>
77578
77579commit 1da4a41b0881b605c7cfa4c4bf637efa132c58fc
77580Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
77581Date: Tue Jan 6 05:01:32 2009 -0200
77582
77583 Add sdksyms.sh to EXTRA_DIST to correct make distcheck
77584
77585 Thanks to <vehemens at verizon.net> for noticing the problem.
77586
77587commit 200230535fa80c04faaa562bbbd8272713e2763b
77588Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
77589Date: Mon Jan 5 16:49:57 2009 -0200
77590
77591 Update sdk headers to export new symbols.
77592
77593 All symbols in installed sdk headers should be explicitly tagged
77594 as exported symbols. Otherwise, to ensure it is not a mistake, one
77595 could write it as something like:
77596 extern /* NOEXPORT */ type name ...;
77597 but the proper procedure really should be to use a non sdk header
77598 (or a "noinst_" one).
77599 This patch also removes prototypes to some functions that existed
77600 only temporarily.
77601
77602commit af6152499654a7146250d4ad0df89f611345434e
77603Author: Jon TURNEY <jon.turney@dronecode.org.uk>
77604Date: Wed Dec 17 14:15:34 2008 +0000
77605
77606 Cygwin/X: Add *.exe pattern to .gitignore
77607
77608 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
77609
77610commit 8428b92fcb478b041f778c0e8c66fdeb96c807f3
77611Author: Colin Harrison <colin.harrison@virgin.net>
77612Date: Sun Jan 4 20:12:16 2009 +0000
77613
77614 Xming: Fix a couple of warnings
77615
77616 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
77617
77618commit f6f97419bd2db651384ecaec0b296840810463a4
77619Author: Jon TURNEY <jon.turney@dronecode.org.uk>
77620Date: Mon Jan 5 18:35:52 2009 +0000
77621
77622 Cygwin/X: Add -static back to linker flags
77623
77624 Add -static back to linker flags to avoid screwing things up for
77625 people who don't know that they need to avoid dynamically linking
77626 with libXfont, as weak symbol handling requires static linking to
77627 work correctly on Cygwin...
77628
77629commit 2ec0ac08addb938a3d5aa5b9888ce7c79a0daada
77630Author: Jon TURNEY <jon.turney@dronecode.org.uk>
77631Date: Wed Dec 17 19:18:38 2008 +0000
77632
77633 Cygwin/X: Fix some remaining build issues
77634
77635 Fixes for a few remaining build issues, omitted from commmit 6e85a8304b50c0bfd16a400a05d257f0168fb99b in error
77636
77637 Provide VENDOR_RELEASE in XORG_VERSION_CURRENT for XWin
77638 Remove -static from linker flags for XWin
77639
77640 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
77641
77642commit e351d10773742d1570e1111976b7c8b6adda5ae6
77643Author: Daniel Stone <daniel@fooishbar.org>
77644Date: Mon Jan 5 17:41:38 2009 +0200
77645
77646 RandR: Only export Xinerama symbols when building Xinerama support
77647
77648 Otherwise compilation fails due to sdksyms.
77649
77650 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
77651
77652commit 3387cbb7c13636eb62b22640d28df7011c2268f0
77653Author: Daniel Stone <daniel@fooishbar.org>
77654Date: Mon Jan 5 11:25:57 2009 +0200
77655
77656 XFree86: Xv: Add ClipNotify helper
77657
77658 Add a ClipNotify helper that lets the driver know about changes in the
77659 clipping of an Xv backing drawable.
77660
77661 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
77662
77663commit 55c4ec0a1ef78481b4d82153c19c20fbf88f1624
77664Author: Daniel Stone <daniel@fooishbar.org>
77665Date: Mon Jan 5 11:24:24 2009 +0200
77666
77667 XFree86: fbdevhw: Add helper function to get fd
77668
77669 Most devices need to do custom weird ioctls, so let them know what the
77670 fd is.
77671
77672 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
77673
77674commit 7da4e901e078e5597b8dbcb7ee39b949edf7ef03
77675Author: Daniel Stone <daniel@fooishbar.org>
77676Date: Mon Jan 5 11:23:39 2009 +0200
77677
77678 XFree86: Fix build with DGA disabled
77679
77680 It's optional, so we might as well work when it's disabled.
77681
77682 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
77683
77684commit 13ce7b98d48261b061da5768a6405bdbe9caa1fb
77685Author: Dan Nicholson <dbn.lists@gmail.com>
77686Date: Mon Dec 22 14:31:03 2008 -0800
77687
77688 dmx: Require newer libXi for XQueryInputVersion
77689
77690 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
77691 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77692
77693commit 27011254c4de4e573a0851bf46892fb488db6522
77694Author: Peter Hutterer <peter.hutterer@who-t.net>
77695Date: Tue Dec 23 09:59:28 2008 +1000
77696
77697 xfree86: If an input device failed to activate, return immediately.
77698
77699 Devices are only activated once - right after they've been added to the
77700 server. If a device failes activation, it's dead. There's no reason to
77701 continue. Return the error code from ActivateDevice() without setting up
77702 sprite information or even sending a event to the client.
77703
77704 Then - in the DDX - just remove the device again.
77705
77706 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77707
77708commit e1a3a1a0d85c9971aea65c2228b5fd4dbf3bf57a
77709Author: Peter Hutterer <peter.hutterer@who-t.net>
77710Date: Tue Dec 23 10:00:34 2008 +1000
77711
77712 xfree86: don't call CheckMotion if a device hasn't been enabled. #19176
77713
77714 X.Org Bug <http://bugs.freedesktop.org/show_bug.cgi?id=19176>
77715
77716commit af820259a06a6e8dbd299fe22347aa83b0c985b0
77717Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77718Date: Fri Jan 2 11:06:12 2009 -0800
77719
77720 XQuartz: Copyright Update... happy new year
77721 (cherry picked from commit 9f1c26ae8d9ccf81cba0ab946326968a701f8c36)
77722
77723commit d8c9ba37abd0d63ea2c5394af434b46ec3b647fd
77724Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77725Date: Fri Jan 2 10:58:40 2009 -0800
77726
77727 XQuartz: pbproxy: Push dpy init and CFRunLoop hook setup into the pbproxy thread to avoid possible deadlock
77728 (cherry picked from commit 49e59d32b88e4fad070f230b5efaa261b47f78db)
77729
77730commit 2f361186f93a791e446c4a9471bb0c3b9b818d98
77731Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77732Date: Wed Dec 31 12:40:02 2008 -0800
77733
77734 XQuartz: Don't use NX_SECONDARYFNMASK, NX_NUMERICPADMASK, NX_HELPMASK
77735
77736 We don't have keycodes for them, so don't try to use them
77737 (cherry picked from commit cb912aca3a2834c4ad8e386c8a0d05c1bb31b0e7)
77738
77739commit a2abaa9fd3a5ad713c1f946e9d7f598825ad3a84
77740Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77741Date: Wed Dec 31 11:57:49 2008 -0800
77742
77743 XQuartz: Make sure to reset the saved key state when deactivating X11.app
77744 (cherry picked from commit 3eef78eb321f4f7dbca5a10c80666c621e28a1e0)
77745
77746commit 3986b683adc4f6f967cc5f643e973d89764234c7
77747Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77748Date: Wed Dec 31 12:44:43 2008 -0800
77749
77750 XQuartz: Honor system key repeat rate
77751 (cherry picked from commit 4303c9be39a86f5a21de108f72b90a989435905e)
77752
77753commit 1a69d56afda082298edc76271d2cb71af00526ce
77754Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77755Date: Mon Dec 29 19:54:12 2008 -0800
77756
77757 XQuartz: Better avoid stuck keys on context switches
77758 (cherry picked from commit e9963f1a4f4f12f253eae9d4f01694b6cabe35ad)
77759
77760commit f464b5a9164b24f74adea198bcabe9affaf5552c
77761Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77762Date: Mon Dec 29 19:38:31 2008 -0800
77763
77764 XQuartz: Workaround OSX VNC server bug for modifier key state
77765
77766 A better approach which ensures we have a L modifier key down if we are told neither are down and atleast one is down... =/
77767 (cherry picked from commit bc13dda345f716bf4de9bfe4e1d85969263b60c2)
77768
77769commit 3ead9aa2602cab9c883aa7bcbd2a8d82c5c88f5a
77770Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77771Date: Mon Dec 29 00:30:31 2008 -0800
77772
77773 XQuartz: Updated man page fullscreen_hotkeys fullscreen_menu
77774 (cherry picked from commit 00f3a2e33c8e91da5f855821313a04ea97445656)
77775
77776commit 48dbaf173a82693fd72953983da9fc556cd1c6ed
77777Author: Daniel Stone <daniel@fooishbar.org>
77778Date: Tue Dec 30 12:17:14 2008 +1100
77779
77780 XKB: Also copy keyboard feedback when copying the keymap
77781
77782 When updating the XKB keymap, make sure the keyboard feedback is also
77783 copied, to preserve autorepeat settings etc.
77784
77785 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
77786
77787commit cf85a32b5f80694f3c2f434e91cf77edb33a7c17
77788Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77789Date: Sun Dec 28 15:57:59 2008 -0800
77790
77791 XQuartz: pbproxy: We explicitly need libX11 for pbproxy
77792 (cherry picked from commit 32824120da0749b8369c2592f851bb1030dea9f1)
77793
77794commit 997e8c321bb11067591785ba38103a3a5029301d
77795Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77796Date: Sun Dec 28 01:53:38 2008 -0800
77797
77798 XQuartz: Try harder to get the user's login environment
77799 (cherry picked from commit 6bedaddd78aa04bd303df434a4c49bb87bd7deaa)
77800
77801commit ae8dc47fd910ec4b9edd64bf3b7ba33e212c5942
77802Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77803Date: Wed Dec 24 00:39:06 2008 -0800
77804
77805 XQuartz: Reposition windows when we enter fullscreen to ensure our root window
77806 (cherry picked from commit 7617d3659b5481ef85aecc1f936e7dd2662bdfbd)
77807
77808commit 895186e026881196ac9383aa165e123055d9cf09
77809Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77810Date: Tue Dec 23 20:33:27 2008 -0800
77811
77812 XQuartz: Re-enable rlAccel
77813
77814 It was incorrectly blamed for some crashes a few months ago, and it should be
77815 safe to use until we get an EXA driver to replace it.
77816 (cherry picked from commit 8121f30bd29591fc74fbc680fbbf20210271fa58)
77817
77818commit dc4498ec5cb7f25f710bdec8793746da2d2324ed
77819Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77820Date: Sun Dec 28 22:16:52 2008 -0800
77821
77822 Updated .gitignore
77823 (cherry picked from commit 06964c322e4da81f3146022049a2de13f6b2892d)
77824
77825commit 3534a5e5d9c5af85149c799f324257f89507fa23
77826Author: Maarten Maathuis <madman2003@gmail.com>
77827Date: Fri Dec 26 16:38:27 2008 +0100
77828
77829 exa: Allow drivers to set non-NULL devPrivate.ptr for !offscreen pixmaps.
77830
77831commit 027b440d4f9f0cdd46addff46fd2d5c44cd5c847
77832Author: Maarten Maathuis <madman2003@gmail.com>
77833Date: Fri Dec 26 15:51:55 2008 +0100
77834
77835 exa: preparing as source and finishing access as mask is a bad idea
77836
77837commit 89bbd07e9ddd688df81405151f9c6284fa3367a8
77838Author: Alan Coopersmith <alan.coopersmith@sun.com>
77839Date: Tue Dec 23 21:40:21 2008 -0800
77840
77841 Remove obsolete XF86_DATE & xf86Date.h
77842
77843commit 8dc3a7d6817b7371bbfece1108ef856788256853
77844Author: Alan Coopersmith <alan.coopersmith@sun.com>
77845Date: Tue Dec 23 14:59:46 2008 -0800
77846
77847 Remove unused CLOG_DATE (leftover from pre-git ChangeLog)
77848
77849commit fe05a46f191ebd36251dda7a9703d90c25efec11
77850Author: Alan Coopersmith <alan.coopersmith@sun.com>
77851Date: Tue Dec 23 14:55:26 2008 -0800
77852
77853 Remove powerpc load/store functions from export list on other platforms
77854
77855commit a1b201bcb01e702f8f0a5889a3d49317503b9475
77856Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77857Date: Tue Dec 23 12:39:38 2008 -0800
77858
77859 XQuartz: Use depth=24 instead of FatalError if we can't figure out our depth
77860 (cherry picked from commit 80b65c5b78d125c4ad3620b87b121c9e666299c3)
77861
77862commit 86dc660588a615baefb1799d78a501c95a931d77
77863Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
77864Date: Tue Dec 23 18:07:54 2008 -0200
77865
77866 Improve sdksyms.c automatic generation (Fix #19245).
77867
77868 Since it is already parsing cpp output, create a dependency file
77869 in the same process. This will cause sdksyms.c to be regenerated
77870 whenever a sdk header is modified.
77871 This also uses the gmake 'sinclude' directive (don't fail if
77872 included file doesn't exist). This should not cause any problems
77873 given that gmake only constructs are used in several other Makefiles.
77874
77875commit 477c32708718f4470ac2e786b2446d7a44e45b9c
77876Author: Peter Hutterer <peter.hutterer@who-t.net>
77877Date: Fri Dec 19 08:56:35 2008 +1000
77878
77879 dix: don't disable uninitialized devices.
77880
77881 If a device hasn't been initialized, it doesn't have a cursor yet. So don't
77882 set the cursor to the NullCursor, and don't try to DisableDevice either.
77883
77884 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77885
77886commit 25aac8b5793caa92116b766d5a9e12dbecdb0290
77887Author: Peter Hutterer <peter.hutterer@redhat.com>
77888Date: Tue Dec 2 16:16:58 2008 +1000
77889
77890 xkb: don't treat groups with different no of symbols as identical.
77891
77892 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
77893
77894commit f141c1b4bb9482f188783dd6f161d3c7960a3329
77895Author: Peter Hutterer <peter.hutterer@redhat.com>
77896Date: Tue Dec 2 14:54:50 2008 +1000
77897
77898 xkb: explicitly check for group replication in the core representation.
77899
77900 Single-group keys may get replicated amongst all groups. Check explicitly for
77901 this case and squash it down to one group.
77902
77903 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
77904
77905commit 70a977c021e107c4fabe46ec2f619be9fb55abeb
77906Author: Peter Hutterer <peter.hutterer@redhat.com>
77907Date: Tue Dec 2 15:36:51 2008 +1000
77908
77909 xkb: don't replicate past the number of groups we have.
77910
77911 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
77912
77913commit a157575eeeb523cd43197c5caeb00cb3d56f9988
77914Author: Peter Hutterer <peter.hutterer@redhat.com>
77915Date: Tue Dec 2 13:31:58 2008 +1000
77916
77917 xkb: ensure enough symbols for core Group1 replication.
77918
77919 A single-group key on a multi-group keyboard has to be replicated across all
77920 three groups (see Section 12.4 of the XKB protocol spec). Ensure that there's
77921 enough symbols available to actually do that.
77922
77923 e.g. a key ABCD on a 3 group keyboard needs to be replicated as ABABCDCDABCD,
77924 hence requiring space for 12 symbols, even if maxSymsPerKey is less than that.
77925
77926 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
77927
77928commit b5736d237a21d5e65d839c4d213dd3bda5a11e9d
77929Author: Maarten Maathuis <madman2003@gmail.com>
77930Date: Mon Dec 22 19:53:14 2008 +0100
77931
77932 randr/xfree86: Fix initial gamma computation.
77933
77934 - The previous version overflowed sometimes.
77935
77936commit ba9dc353a60e9a6bba64e854b23271011bff7015
77937Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77938Date: Sun Dec 21 21:03:59 2008 -0800
77939
77940 XQuartz: Don't use keycode 0 to determine !swallow since our most common key to swallow is actual keycode 0 (a)
77941 (cherry picked from commit 33f43a7f03023bfbab25a957cb81fc25b4afa4ca)
77942
77943commit f0cdccd1e22198f4f5e1aa0d0617441b79ffbc44
77944Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77945Date: Sun Dec 21 20:40:36 2008 -0800
77946
77947 XQuartz: Update our "screens" when we toggle rootless rather than when we toggle fullscreen
77948
77949 This old behavior was used as a workaround for the menubar behavior in the older server,
77950 but we handle it better now and need to update our screens when we toggle the rootless
77951 state instead.
77952 (cherry picked from commit 508aa95bc2cd3fdc3dff448ec090919bf807d153)
77953
77954commit 8f04a5237384fa97ff960b4607a79b688ea03fc1
77955Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77956Date: Sun Dec 21 14:54:11 2008 -0800
77957
77958 XQuartz: Run applications via '/bin/sh -c ...' to support users who expect shell parsing
77959 (cherry picked from commit 67455e716e3ecffd528930479192785958d37988)
77960
77961commit 029d255a654eca6401c374d145f63bbb923554b5
77962Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77963Date: Sun Dec 21 14:43:12 2008 -0800
77964
77965 XQuartz: pbproxy: Release display notification lock when not needed to avoid deadlock
77966 (cherry picked from commit 22f664ab95a0cae981e9cefad6f075d051583ca5)
77967
77968commit b1f166f2981595aea6f8e2a336d6d25436f8a07d
77969Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77970Date: Sun Dec 21 14:42:26 2008 -0800
77971
77972 XQuartz: update quoting in case X11.app is moved to a directory with a space.
77973 (cherry picked from commit cc805dc799efa37c8dcefa3db04d87e9b835ffbd)
77974
77975commit 2f47bda645eb43c60b71315e8d265190acce481d
77976Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77977Date: Thu Dec 18 09:14:16 2008 -0800
77978
77979 XQuartz: Changed X11.sh to allow use of a ~/.x11run as requested by users of alternate shells
77980 (cherry picked from commit b62ed1f8eaf041a946bb591165bb18ee481dedbf)
77981
77982commit 83edcc08de4c815c5cbe815f521b3b728441cad6
77983Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77984Date: Wed Dec 17 15:09:35 2008 -0800
77985
77986 XQuartz: Get rid of white rectangle bug
77987 (cherry picked from commit 3269959033ed0c675a3a906666454df34086896a)
77988
77989commit 1335a300b9cd03208a583acfdb7f7153eaef4ea6
77990Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77991Date: Wed Dec 17 14:43:02 2008 -0800
77992
77993 rootless: Make expose_1 static
77994 (cherry picked from commit 60c8d2697036a125ca5381df8e2eaedabad4d242)
77995
77996commit 22b2811e6aa0bdbfa6a0682ac862063e52ec0c62
77997Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
77998Date: Wed Dec 17 13:54:53 2008 -0800
77999
78000 XQuartz: Name the startup shell script X11 for better compatability
78001 (cherry picked from commit f84613d6fcbbb5f769ee34cff0900353a13693cf)
78002
78003commit 82fba7697fbbb6f66cc22424ba99755b99ff482e
78004Author: Maarten Maathuis <madman2003@gmail.com>
78005Date: Mon Dec 22 00:33:55 2008 +0100
78006
78007 xfree86/randr: Avoid a crash when xf86CrtcConfigPrivateIndex is -1.
78008
78009commit 1567b7243f4799808ab93fbd962df14ce3af2d49
78010Author: Maarten Maathuis <madman2003@gmail.com>
78011Date: Sun Dec 21 00:08:56 2008 +0100
78012
78013 exa: A more correct fix.
78014
78015commit 2db7b66863ae6055c3ce13c88b36d620de8a4d75
78016Author: Maarten Maathuis <madman2003@gmail.com>
78017Date: Fri Dec 19 23:12:37 2008 +0100
78018
78019 exa: a few cleanups
78020
78021 - Some warnings silenced.
78022 - Some whitespace cleanup.
78023
78024commit aedd2f566df585db7a1614f302cc8d3feda54275
78025Author: Maarten Maathuis <madman2003@gmail.com>
78026Date: Fri Dec 19 19:10:23 2008 +0100
78027
78028 randr/xfree86: Fix a one off error in the panning calculations.
78029
78030 - Example: mode 1280x1024, panned area 1281x1024
78031 panned_area.x2 = 1281
78032 mode.width = 1280
78033 If you substract 1280 from 1281, then that leaves you with one.
78034 Which is the one pixel that you need to move to actually see the last pixel collumn.
78035 Substracting 1 from this will consistently prevent you from seeing the right and bottom edge.
78036
78037commit 332d65ec7a6e94d75efe95d53742f137835274de
78038Author: Maarten Maathuis <madman2003@gmail.com>
78039Date: Fri Dec 19 18:59:27 2008 +0100
78040
78041 randr: Consider panned crtc's when calculating xinerama screen sizes.
78042
78043 - This will allow window managers and applications to actually use the panned area.
78044
78045commit f2d1de6c2a3e6e50db6d9714c293eb26324fbd3d
78046Author: Adam Jackson <ajax@redhat.com>
78047Date: Fri Dec 19 10:27:37 2008 -0500
78048
78049 xv: remove useless XVCALL macro
78050
78051commit 5a072c55350f4b73d911ea6a2aeddad844924834
78052Author: Adam Jackson <ajax@redhat.com>
78053Date: Fri Dec 19 09:51:52 2008 -0500
78054
78055 dmx: Fix calloc macro confusion.
78056
78057commit 8c488ac3b3990cd203baed7f2127b9bed8aab534
78058Author: Alan Coopersmith <alan.coopersmith@sun.com>
78059Date: Thu Dec 18 19:47:49 2008 -0800
78060
78061 Fix linking of Xorg with dtrace probes on Solaris
78062
78063commit 396433d0da721951e6e7abc02dc3b2f682495154
78064Author: Alan Coopersmith <alan.coopersmith@sun.com>
78065Date: Thu Dec 18 18:49:47 2008 -0800
78066
78067 Fix sdksyms.sh to work with Solaris/Sun compiler builds
78068
78069 - Pass $(CPP) & $(AWK) settings from configure to sdksyms.sh
78070 - Only reset sdk variable (tracks if header is part of sdk) if
78071 a filename is included on the cpp # <line-no> <filename> line,
78072 since Sun compilers omit filename when it is unchanged from
78073 previous line.
78074
78075commit aea6f19f25e13768b1d09fac4991d6a5e6c2cdac
78076Author: Peter Hutterer <peter.hutterer@who-t.net>
78077Date: Tue Dec 16 17:10:03 2008 +1000
78078
78079 xfree86: don't restore the TTY mode if we didn't initialize it ourselves
78080
78081 Restoring it unconditionally means we restore to whatever tty_mode has as
78082 default value (i.e. 0). K_RAW happens to be 0x00, so we always restore to raw
78083 mode if allowEmptyInput is off.
78084
78085 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78086 Signed-off-by: Adam Jackson <ajax@redhat.com>
78087 Signed-off-by: Julien Cristau <jcristau@debian.org>
78088
78089commit 5d065a889074558634216eebd4bba35a60db9a09
78090Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
78091Date: Thu Dec 18 13:59:36 2008 -0500
78092
78093 xselinux: Use xace Xtrans wrappers instead of the now-inaccessible wrapees.
78094
78095commit f87e66486c3610c56888915b66ae5ab0af03da8b
78096Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
78097Date: Thu Dec 18 13:58:35 2008 -0500
78098
78099 xace: Export wrappers around two Xtrans functions used by modules.
78100
78101 Don't know a better way to do this, since Xtrans isn't a library that
78102 can be linked into modules.
78103
78104commit 777408914dbf395e9969e3f34720286e6f740369
78105Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
78106Date: Thu Dec 18 12:35:57 2008 -0500
78107
78108 Add xace headers to the SDK when enabled and export the XaceHooks symbol
78109 to modules.
78110
78111commit 9e4ef3cfe5ee178c266edd64a8edd767827aa0d7
78112Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78113Date: Thu Dec 18 15:00:25 2008 -0200
78114
78115 Use regex pattern understood by all known awk variants.
78116
78117 Ubuntu uses mawk by default, but it doesn't understand posix character
78118 classes (which are locale dependent, and this patch uses only valid C
78119 identifiers).
78120 Also make sure awk runs with LC_ALL=C to match the regex patterns.
78121
78122commit 3d3c234b434a3443a00c3be9f32c698bcced111c
78123Author: Maarten Maathuis <madman2003@gmail.com>
78124Date: Thu Dec 18 16:26:38 2008 +0100
78125
78126 randr: add some notes about the stuff that changed in driver ABI 2
78127
78128commit 1b244feb4cf11d1cdba79fbead50035965535552
78129Author: Maarten Maathuis <madman2003@gmail.com>
78130Date: Thu Dec 18 16:11:21 2008 +0100
78131
78132 randr: check for malloc failure
78133
78134commit 21a2df9f359584e634dc12d8602401d54defd7db
78135Author: Colin Harrison <colin.harrison@virgin.net>
78136Date: Sun Nov 2 20:46:44 2008 +0000
78137
78138 Xming,Cygwin/X: Fix crashes when using the DirectDraw '-refresh rate-in-Hz' option in -fullscreen (#11128)
78139
78140 fd.o bugzilla #11128
78141
78142 Fix crashes when using the '-refresh rate-in-Hz' option in -fullscreen mode and using a DirectDraw engine.
78143 On failure (typically 0x80004001 DDERR_UNSUPPORTED), use the default refresh rate for the driver, instead of the
78144 entered value.
78145
78146 Also fix some nearby error reporting
78147
78148 Copyright (C) Colin Harrison 2005-2008
78149 http://www.straightrunning.com/XmingNotes/
78150 http://sourceforge.net/projects/xming/
78151
78152 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78153
78154commit b0ed1567c1782041cf3d14ec6dd4d622915c7d0c
78155Author: Colin Harrison <colin.harrison@virgin.net>
78156Date: Wed Dec 17 17:16:02 2008 +0000
78157
78158 Xming, Cygwin/X: Tidy up warnings, headers and unused variables (#11132)
78159
78160 Tidy up various warnings, include headers and unused variables, etc.
78161
78162 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78163
78164commit 15c4a6e088039e14b4f2387ff204cb5448916b39
78165Author: Colin Harrison <colin.harrison@virgin.net>
78166Date: Wed Dec 17 16:44:07 2008 +0000
78167
78168 Xming, Cygwin/X: Fix crash in function winMultiWindowGetTransientFor() (#11147)
78169
78170 The function winMultiWindowGetTransientFor() crashes, due to a bogus pointer
78171 dereference, but this fact was previously hidden by the function not being called!
78172
78173 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78174
78175commit 08a3d6928c87032998b6113d67f69b79b09eecee
78176Author: Peter Hutterer <peter.hutterer@who-t.net>
78177Date: Tue Dec 16 14:19:12 2008 +1000
78178
78179 dix: Allocate the space for the DCCE when the last SD is removed.
78180
78181 This still doesn't fix reset the MD's classes (a TODO that's been here for
78182 ages), but at least we don't segfault anymore when detaching the last SD.
78183
78184 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78185
78186commit 689be0cb4b259840208c17f974066a942196793a
78187Author: Peter Hutterer <peter.hutterer@who-t.net>
78188Date: Tue Dec 16 14:19:38 2008 +1000
78189
78190 dix: set the correct length in CreateClassesChangedEvent.
78191
78192 Fallout from aeff14d5f208d02f211b8b2a1a98999624e8c2cf. Yes, we don't malloc
78193 anymore because we are inside a SIGIO and the memory is already there anyway.
78194 But we still need to set the event length correctly, otherwise
78195 mieqEnqueue/mieqProcessInputEvent don't know how much memory to copy.
78196
78197 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78198
78199commit 1001403301001fcc84ce6833938590956df40d2a
78200Author: Maarten Maathuis <madman2003@gmail.com>
78201Date: Thu Dec 18 00:19:06 2008 +0100
78202
78203 Forgot to fix a comment.
78204
78205commit 9fa15bef59881bdcf087889f16ab3c8d953da8f1
78206Author: Maarten Maathuis <madman2003@gmail.com>
78207Date: Wed Dec 17 23:42:30 2008 +0100
78208
78209 randr: some improvements, fixes and crtc abi bump
78210
78211 - Add active field to crtc.
78212 - Set gamma (only) whenever a crtc becomes active.
78213 - Check for xf86_config being NULL.
78214 - Increase crtc abi to 3.
78215 - A few other fixes.
78216
78217commit bf65523ab0b39774f07a7ae478ff3f5653fad469
78218Author: Joe Krahn <jkrahn@nc.rr.com>
78219Date: Sun Nov 2 20:17:03 2008 +0000
78220
78221 Cygwin/X: Fix for mis-aligned icon data creates bad background masks (#4491)
78222
78223 fd.o bugzilla #4491
78224
78225 Transparent icon backgrounds appear with black stripes as mask data is incorrectly aligned.
78226 Modified the icon DDB bitmap data alignment to 16 bits
78227
78228 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78229
78230commit 91f73b79b7ae64e5b846d1efeb470bb61a913720
78231Author: Maarten Maathuis <madman2003@gmail.com>
78232Date: Wed Dec 17 16:56:26 2008 +0100
78233
78234 randr: Improve per-crtc gamma support.
78235
78236 - The Gamma values from the monitor section are now used during initial config.
78237 - The old colormap system is disabled when gamma set hook is available.
78238 - Gamma values are now persistent for the lifetime of the xserver.
78239 - This requires no driver changes and should be driver ABI compatible.
78240
78241commit 1556815d34cecb4b4b62d2a4ce813b1435a937ec
78242Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78243Date: Sun Nov 2 18:27:18 2008 +0000
78244
78245 Cygwin/X: Initialize native HWND atom when built !XWIN_MULTIWINDOWEXTWM
78246
78247 Fix the native HWND atom identifier to be initialized correctly, when built without external WM support
78248
78249 This is neccessary for native window naming to work correctly in multiwindow internal WM mode.
78250
78251 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78252
78253commit 92082a0a00f05ecdc90814a8104828853c9c8f29
78254Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78255Date: Sun Nov 2 18:27:15 2008 +0000
78256
78257 Cygwin/X: update to use standard DPMS stubs
78258
78259 Bring Cygwin/X up to date with DPMS API changes by using the standard stubs
78260
78261 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78262
78263commit 2850f26bb2999ef3fb2a10adb80fe358463aecb5
78264Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78265Date: Sun Nov 2 18:27:14 2008 +0000
78266
78267 Cygwin/X: update for changes to shadow framebuffer
78268
78269 Bring Cygwin/X up to date with the change to damage-based shadow framebuffer.
78270
78271 Register the screen bitmap as the shadow framebuffer after screen resources
78272 are created.
78273
78274 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78275
78276commit 35d76028304f910edeaf642376df71b8134ec02c
78277Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78278Date: Sun Nov 2 18:27:13 2008 +0000
78279
78280 Cygwin/X: enqueue a pointer motion event on mouse movement
78281
78282 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78283
78284commit 816e0d243df068bc06d328874b5655d4530fe16a
78285Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78286Date: Sun Nov 2 18:27:12 2008 +0000
78287
78288 Cygwin/X: update for changes in mieq API
78289
78290 Bring Cygwin/X up to date with mieq event API changes
78291
78292 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78293
78294commit bca7483c4e95ad450fb09048a3b85bb3056a0518
78295Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78296Date: Sun Nov 2 18:27:11 2008 +0000
78297
78298 Cygwin/X: update for MPX device changes
78299
78300 Bring Cygwin/X up to date with MPX device API changes
78301
78302 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78303
78304commit 829a8a970dd17623ef3cb5022d34190744f74741
78305Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78306Date: Sun Nov 2 18:27:10 2008 +0000
78307
78308 Cygwin/X: update for MPX cursor API changes
78309
78310 Bring Cygwin/X up to date with cursor API changes
78311
78312 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78313
78314commit d61e902aab92c262e6c8ee9cd70aec4493cf6cae
78315Author: Francis Giraldeau <francis.giraldeau@revolutionlinux.com>
78316Date: Wed Dec 17 09:49:39 2008 +0200
78317
78318 Don't log audit messages when -audit 0 specified
78319
78320 make the auditTrailLevel check more general and don't log when not
78321 using DTRACE and -audit 0 is set.
78322
78323commit 295a3fa721ee2f2226891de308c0c4b0ee26fe72
78324Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78325Date: Wed Dec 17 02:04:12 2008 -0200
78326
78327 Ensure symbols required by swrast_dri.so are visible.
78328
78329commit 6d8ea5104cf97dbf64612f58fc06f94f869ed5ec
78330Author: Colin Guthrie <cguthrie@mandriva.org>
78331Date: Tue Dec 16 17:44:27 2008 -0500
78332
78333 Fix compilation with -Werror=format-security
78334
78335commit 1bb5a919f4bf38ac96c73077021b0cdc82965f31
78336Author: Adam Jackson <ajax@redhat.com>
78337Date: Tue Dec 16 13:46:59 2008 -0500
78338
78339 config: Add vboxvideo and tga to the magic driver list
78340
78341 Also sort the list while I'm here.
78342
78343commit 8511a964f81b3b06a526f0fca4232afb2152f405
78344Author: Matthias Hopf <mhopf@suse.de>
78345Date: Tue Dec 16 18:21:20 2008 +0100
78346
78347 randr: Update EDID_ATOM_NAME to reflect RandR 1.3 name change
78348
78349commit a058192d970f7417b06334aa666c17e9891795c6
78350Author: Owen W. Taylor <otaylor@redhat.com>
78351Date: Wed Oct 22 16:12:53 2008 -0400
78352
78353 Xephyr: Fix crash with control-shift and GLX (#18185)
78354
78355 When we are looking up the screen for an event, we need to take
78356 into account the fact that the event may have been delivered to the
78357 "peer window" that we create when implementing GLX. Since we only
78358 ever create one such window per screen, just add a single peer_win
78359 field to EphyrHostScreen.
78360
78361commit 6e85a8304b50c0bfd16a400a05d257f0168fb99b
78362Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78363Date: Sun Nov 2 18:27:09 2008 +0000
78364
78365 Cygwin/X: build machinery fixes
78366
78367 The Cygwin/X build has been broken for a long time, probably since modular X
78368 This patch fixes up configure/makefile and some general build issues (#18568)
78369
78370 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
78371
78372commit 7aa29b9d66c3cd0f8af4fafbe92efd0c0556d225
78373Author: Robert Noland <rnoland@2hip.net>
78374Date: Tue Dec 16 00:07:42 2008 -0500
78375
78376 Support -sharevts on FreeBSD
78377
78378 Bugzilla #11179
78379
78380 Submitted-by: Nathan Whitehorn
78381
78382 Signed-off-by: Robert Noland <rnoland@2hip.net>
78383
78384commit 063eb6743cd0d98dd52d1a9559b804381ee5144d
78385Author: Keith Packard <keithp@keithp.com>
78386Date: Mon Dec 15 19:53:45 2008 -0800
78387
78388 Patch brown-bag fix for bug 19017 (scrn->virtualX/virtualY 0 before PreInit)
78389
78390 When a driver uses a crtc during device detection, the scrn has not yet been
78391 configured and virtualX/virtualY are still zero. This caused the X server
78392 to try and allocate a shadow frame buffer, which couldn't work.
78393
78394 Detect this by checking for zero virtualX/virtualY values.
78395
78396 Signed-off-by: Keith Packard <keithp@keithp.com>
78397
78398commit fde2f961035609cfff8761ef1e0f23e6a9761be2
78399Author: Keith Packard <keithp@keithp.com>
78400Date: Mon Dec 15 11:39:24 2008 -0800
78401
78402 Use scrn->virtualX/virtualY in xf86CrtcFitsScreen. Fix bug 19017.
78403
78404 pScreen->width/height are not initialized when doing initial mode setting,
78405 which makes this function incorrectly fail. Using scrn->virtualX should work
78406 in all cases though.
78407
78408 Bug 19017 reports a crash in xf86CrtcSetModeTransform when doing a modeset
78409 for output probing, long before the screen array is initialized; that was
78410 caused by a work-around to set pScreen->width/height so that xf86CrtcFitsScreen
78411 could find the right values.
78412
78413 Signed-off-by: Keith Packard <keithp@keithp.com>
78414
78415commit 29a5b0596b396d3e4a8a014cacd3e3ef77467ab7
78416Author: Keith Packard <keithp@keithp.com>
78417Date: Mon Dec 15 11:38:00 2008 -0800
78418
78419 When disabling SIGIO tracking, use SIG_IGN instead of SIG_DFL.
78420
78421 This avoids a race condition for drivers which mis-order the fd close and
78422 signal disable.
78423
78424 Signed-off-by: Keith Packard <keithp@keithp.com>
78425
78426commit 1ba4cbb15919759aadd71960c5c057af9ba94fe3
78427Author: Keith Packard <keithp@keithp.com>
78428Date: Mon Dec 15 11:36:43 2008 -0800
78429
78430 Clean up rotation data when crtc is turned off
78431
78432 The shadow frame buffer and other data used for rotation need to be freed
78433 when the crtc is disabled, not just when rotation is disabled.
78434
78435 Signed-off-by: Keith Packard <keithp@keithp.com>
78436
78437commit 2a61397d17339113b9e37995b06ca543589814ce
78438Author: Cooper Yuan <Cooper.Yuan@amd.com>
78439Date: Mon Dec 15 10:29:34 2008 -0500
78440
78441 Fix typo in xf86PickCrtcs()
78442
78443commit 3db60add6671f766f9360e7d03492dfd51eba225
78444Author: Jon TURNEY <jon.turney@dronecode.org.uk>
78445Date: Mon Dec 15 12:42:10 2008 +0000
78446
78447 Cygwin/X: Update the native icon to one based on the X.Org logo
78448
78449 Update the native icon for the X server to one created from the X.Org logo by Colin Harrison.
78450
78451commit 450739efa904a0fce5024372da7236bb7b84b67a
78452Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78453Date: Sat Dec 13 00:45:27 2008 -0800
78454
78455 XQuartz: Updated menu item ordering for better HIG compliance
78456 (cherry picked from commit 4c9bb241ce5fb856fab20ae96fa89bd2b71ef242)
78457
78458commit aee27be2785d8b09ac10f5be040b4d573190de0e
78459Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78460Date: Wed Dec 10 23:04:30 2008 -0800
78461
78462 XQuartz: Tiger fix, don't call Xplugin code in the Appkit thread if Xplugin isn't threadsafe.
78463 (cherry picked from commit 748d9e5bd756513d42c4046f3b31e1fdc55bccb6)
78464
78465commit a939368ab8140d48c1da4ba0bb229d13b221189c
78466Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78467Date: Mon Dec 15 03:18:13 2008 -0800
78468
78469 mi: Reuse memory in mieqProcessInputEvents rather than making excessive calls to calloc()
78470
78471 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78472 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
78473
78474commit f1c9b5ab230cbb4124d8d476ae4886d05022adcb
78475Author: Ian Romanick <ian.d.romanick@intel.com>
78476Date: Sun Dec 14 18:58:33 2008 -0800
78477
78478 GLX: Changes resulting from changes to Mesa generator scripts / data
78479
78480 Several recent Mesa commits (listed below) make modifications to the
78481 protocol generator data and scripts. This commit represents the
78482 changes to the generated files resulting from the previous changes.
78483
78484 - 0f73302d24f4201813da2939742c5bcb6964b3b1
78485 GLX: Fix protocol for glTexSubImage#D
78486
78487 - 1709ab01ef24279c782e420568e9257b4b92b224
78488 Return 0 as the request size when the pixels parameter is NULL
78489
78490 - 63cca2ba10ce7dcc8481cfa4be3872dfc269dded
78491 GLX: Include glapi.h before glapitable.h
78492
78493 This is the server-side part of the fix for bugzilla #11003.
78494
78495commit 6383bc93b2ae76361a58653bc7e291df0d6ddfec
78496Author: Peter Hutterer <peter.hutterer@who-t.net>
78497Date: Sat Dec 13 15:44:52 2008 +1000
78498
78499 dix: Init DeviceEnterLeave event's type before FixUpEventFromWindow (#19064)
78500
78501 X.Org Bug 19064 <http://bugs.freedesktop.org/show_bug.cgi?id=19064>
78502
78503commit 49d38ab2328c409b2a98465b52677af057121513
78504Author: Eric Anholt <eric@anholt.net>
78505Date: Tue Dec 2 15:09:57 2008 -0800
78506
78507 Warning fix: Remove dead glXDisp{,Swap}_DrawArraysEXT definitions.
78508
78509 Signed-off-by: Eric Anholt <eric@anholt.net>
78510 Acked-by: Ian Romanick <ian.d.romanick@intel.com>
78511
78512commit 2c5bfffc832e3818bbf0c0a96522865ce1ef2653
78513Author: Eric Anholt <eric@anholt.net>
78514Date: Tue Dec 2 14:59:04 2008 -0800
78515
78516 Warning fix (GL likes to call strings GLubyte * instead of char *).
78517
78518 Signed-off-by: Eric Anholt <eric@anholt.net>
78519 Acked-by: Ian Romanick <ian.d.romanick@intel.com>
78520
78521commit 7be6520d94df874c6bbd46d06a1830a12d0967f2
78522Author: Stuart Bennett <sb476@cam.ac.uk>
78523Date: Fri Dec 12 00:08:44 2008 +0000
78524
78525 dolt: allow older versions of bash to compile the xserver (#19031)
78526
78527 Signed-off-by: James Cloos <cloos@jhcloos.com>
78528
78529commit aeff14d5f208d02f211b8b2a1a98999624e8c2cf
78530Author: Peter Hutterer <peter.hutterer@redhat.com>
78531Date: Tue Nov 25 20:23:50 2008 +1000
78532
78533 dix: don't alloc in ChangeMasterDeviceClasses.
78534
78535 We mustn't realloc as we are inside a signal handler. With
78536 SetMinimumEventSize, this code should never be hit anyway, as the event list
78537 should have the required memory before this code is hit.
78538
78539 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78540
78541commit fb2a8d0e59a3d187255538f6add22ec67551507a
78542Author: Peter Hutterer <peter.hutterer@who-t.net>
78543Date: Mon Dec 8 15:58:15 2008 +1000
78544
78545 Xi: silence compiler warning
78546
78547 Don't mix declarations and statements.
78548
78549commit d961abb59f0a8bee4bd11b1540aa43ff83cddbb8
78550Author: Peter Hutterer <peter.hutterer@who-t.net>
78551Date: Thu Dec 11 15:54:15 2008 +1000
78552
78553 dix: fix compiler warning (mixing declarations + statements)
78554
78555 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78556
78557commit cb95642dc8edebb2935dd471f8b339cb98aa8481
78558Author: Peter Hutterer <peter.hutterer@redhat.com>
78559Date: Fri Nov 28 22:28:32 2008 +1000
78560
78561 Remove #define NEED_EVENTS and NEED_REPLIES
78562
78563 A grep on xorg/* revealed there's no consumer of this define.
78564
78565 Quote Alan Coopersmith:
78566 "The consumer was in past versions of the headers now located
78567 in proto/x11proto - for instance, in X11R6.0's xc/include/Xproto.h,
78568 all the event definitions were only available if NEED_EVENTS were
78569 defined, and all the reply definitions required NEED_REPLIES.
78570
78571 Looks like Xproto.h dropped them by X11R6.3, which didn't have
78572 the #ifdef's anymore, so these are truly ancient now."
78573
78574 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78575 Signed-off-by: Adam Jackson <ajax@redhat.com>
78576
78577commit 920b7622be4a68ebb5d8432f55c5362fbe964fce
78578Author: Simon Thum <simon.thum@gmx.de>
78579Date: Tue Nov 25 14:28:26 2008 +1000
78580
78581 xfree86: dump the ptraccel filter setup to the log on init.
78582
78583 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78584
78585commit fbb57a2c54b141ccbb50833143b0b7cb2c4cb903
78586Author: Simon Thum <simon.thum@gmx.de>
78587Date: Tue Nov 25 13:43:28 2008 +1000
78588
78589 dix: ptraccel - Add GetAccelerationProfile()
78590
78591 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78592
78593commit beb749c87f383a62172a9ffa36373438e7a12d8e
78594Author: Adam Jackson <ajax@redhat.com>
78595Date: Thu Dec 11 17:04:37 2008 -0500
78596
78597 Fix syncsrv.h guard define
78598
78599commit 9fba808b4f43f5dc69d85a48fa40cb83dd4d9f9d
78600Author: Colin Harrison <colin.harrison@virgin.net>
78601Date: Thu Dec 11 13:22:53 2008 -0500
78602
78603 xsync: Prototype fix.
78604
78605commit dce887ff53222652c397bc96a1dd3566f5847745
78606Author: Adam Jackson <ajax@redhat.com>
78607Date: Thu Dec 11 13:16:02 2008 -0500
78608
78609 xsync: build fix
78610
78611 argh protocol header disaster
78612
78613commit d41b8960be8ad316504ef5657c0abfe6d76d3b5e
78614Author: Matthias Hopf <mhopf@suse.de>
78615Date: Thu Dec 11 18:48:02 2008 +0100
78616
78617 randr: Fix initial panning border copy
78618
78619 Thanks to Julien Cristau for finding this.
78620
78621commit fd77ce9f884f5ac4d36736f3a99ba86101f133ea
78622Author: Matthias Hopf <mhopf@suse.de>
78623Date: Thu Dec 11 17:51:20 2008 +0100
78624
78625 randr: Oops, miscalculated panning rectangle's coordinates
78626
78627commit f8e52f1b6d7c59d007de99a1c9c69c053d4f3cbe
78628Author: Matthias Hopf <mhopf@suse.de>
78629Date: Thu Dec 11 16:48:40 2008 +0100
78630
78631 randr: Update initial screen size if panning information is present
78632
78633commit 24d6b7d1c5fc5d07c2eb06a9fc4406e393d59b17
78634Author: Matthias Hopf <mhopf@suse.de>
78635Date: Thu Dec 11 16:48:11 2008 +0100
78636
78637 randr: Fix error message for bad panning config
78638
78639commit fdbbe65a7e777b7777bfae5a161efb89d4fb9a8d
78640Author: Matthias Hopf <mhopf@suse.de>
78641Date: Thu Dec 11 14:56:51 2008 +0100
78642
78643 randr: Add monitor option "Panning" for initial panning configuration
78644
78645commit 27261a950d91c352eac25a3036656c3e4f81fb12
78646Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78647Date: Thu Dec 11 14:43:04 2008 -0200
78648
78649 Modify sdksyms.sh to receive $top_srcdir as first argument.
78650
78651 If the basename of header file processed by cpp matches $top_srcdir,
78652 check for extern symbols in the output, and add to the xorg_symbols
78653 vector.
78654 Possibly a better solution then using this script would be to somehow
78655 tell the linker to not drop any symbols from the binary being generated.
78656
78657commit 1f4fb0225b278d1cf4145aebeb0bdd23dc8f62d5
78658Author: Adam Jackson <ajax@redhat.com>
78659Date: Wed Dec 10 16:13:20 2008 -0500
78660
78661 xsync: Fix wakeup storm in idletime counter.
78662
78663 Wakeup scheduling only considered the threshold values, and not whether
78664 the trigger was edge or level.
78665
78666 See also:
78667 https://bugzilla.redhat.com/show_bug.cgi?id=474586
78668 http://svn.gnome.org/viewvc/gnome-screensaver/trunk/src/test-idle-ext.c?view=markup
78669
78670commit 1a99110f0c221b79045ea26d61c4a1ec1e0d7341
78671Author: Adam Jackson <ajax@redhat.com>
78672Date: Wed Dec 10 14:18:59 2008 -0500
78673
78674 xsync: ANSI cleanups
78675
78676commit e0d8f6a8084a3fe6c549c1dee11e4502e316c382
78677Author: Adam Jackson <ajax@redhat.com>
78678Date: Wed Dec 10 13:32:04 2008 -0500
78679
78680 xsync: Use a local header for server API definitions
78681
78682commit 1208a1dbcaecbc218a77bf51068f9c0c768e16c5
78683Author: Adam Jackson <ajax@redhat.com>
78684Date: Wed Dec 10 11:47:35 2008 -0500
78685
78686 xsync: remove cast abuse.
78687
78688commit 119d9c46e841f5fa35610f557e6fa1ec58587c24
78689Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78690Date: Wed Dec 10 01:56:01 2008 -0800
78691
78692 XQuartz: Fix path to executable
78693 (cherry picked from commit 7e9d3698e070a0c63dd2556651373c3aca58e2fe)
78694
78695commit c46b564c47fde3474e948aa1b188a975836cba47
78696Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78697Date: Wed Dec 10 00:00:29 2008 -0800
78698
78699 XQuartz: fixed make dist
78700 (cherry picked from commit 2d52367ab92429e58596d1a1ed3ef52a0a38a7aa)
78701
78702commit 23156a21d8e2f1be7d19d69cb91dae1b885ef671
78703Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78704Date: Tue Dec 9 23:48:18 2008 -0800
78705
78706 XQuartz: Make debugging output for invalid depths a bit more detailed
78707 (cherry picked from commit 609fb166b7062c76f0561df12ffe893811f6ac8f)
78708
78709commit 76351d2faf9ceb79aaa00b374c203b8b279c58f3
78710Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78711Date: Tue Dec 9 23:47:32 2008 -0800
78712
78713 XQuartz: Avoid using login /bin/sh blech. Just use a bash script to start the app, so it will inherit the right environment
78714 (cherry picked from commit f4b7ad9cc6b0c99fc7ee8516c4bf858ece938148)
78715
78716commit e5ce6e198f4c245f4dc840c5e9e90a7ef80fb6e9
78717Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78718Date: Mon Dec 8 21:59:49 2008 -0800
78719
78720 XQuartz: unset DISPLAY if we didn't get a launchd socket handoff
78721 (cherry picked from commit b959727f38733009c6381cc8ca06b5984257bdac)
78722
78723commit ecaaf0462d6c70cabfcaf6868a526e5415505e2c
78724Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78725Date: Mon Dec 8 21:00:11 2008 -0800
78726
78727 XQuartz: unsetenv(DISPLAY) if we're not org.x.X11
78728 Also some prefix related fixes.
78729 (cherry picked from commit fd459e96581a883e30323c840b71004aa0747169)
78730
78731commit ea94995982d1cc585768d52fec75cae018457fa5
78732Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
78733Date: Sat Dec 6 12:01:28 2008 -0800
78734
78735 XQuartz: Removed some debug spew
78736 (cherry picked from commit 370dcf11b360b87aa24cace35eb615419057fda0)
78737
78738commit 86c64ddf21763972aa7fc8c5770259123c9907b3
78739Author: Adam Jackson <ajax@redhat.com>
78740Date: Mon Dec 8 17:42:47 2008 -0500
78741
78742 randr: clear primaryOutput when the output is deleted
78743
78744commit fe65f400ed16cb39db8c9518b9446f590c34db1a
78745Author: Adam Jackson <ajax@redhat.com>
78746Date: Mon Dec 8 17:40:07 2008 -0500
78747
78748 randr: use primary output for RRFirstOutput()
78749
78750commit cdcb516e561e2f65eb2fa523ca001c57674d5caf
78751Author: Adam Jackson <ajax@redhat.com>
78752Date: Mon Dec 8 17:37:17 2008 -0500
78753
78754 randr: Mangle GetScreenResources sort order based on primary output
78755
78756commit d7b316e82bc7051f8829b4f4a640f50ae91c2db9
78757Author: Adam Jackson <ajax@redhat.com>
78758Date: Mon Dec 8 17:28:55 2008 -0500
78759
78760 randr: Mangle compat Xinerama reply based on primary output
78761
78762commit 9d58d2a319059989ccdfa758f586149ccdc16df6
78763Author: Adam Jackson <ajax@redhat.com>
78764Date: Tue Dec 9 10:51:37 2008 -0500
78765
78766 randr: Add [GS]etOutputPrimary
78767
78768commit d281866b74f7067f2704c278fe9720eafc0ee5ef
78769Author: Peter Hutterer <peter.hutterer@redhat.com>
78770Date: Mon Dec 1 21:20:48 2008 +1000
78771
78772 mi: Clean up CopyGetMasterEvent, re-use the memory.
78773
78774 Alloc an EventList once and then re-use instead of allocing a new event each
78775 time we need a master event.
78776 There's a trick included: because all the event processing handlers only take
78777 an xEvent, init a size 1 EventList and squash the events into this one.
78778
78779 Events that have count > 1 must be squished into an xEvent array anyway before
78780 passing into the event handlers, so we don't lose anything here.
78781
78782 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78783 Acked-by: Daniel Stone <daniel@fooishbar.org>
78784
78785commit 0b4fef6337d88ae8ef05b8b73941350a9007565c
78786Author: Peter Hutterer <peter.hutterer@who-t.net>
78787Date: Wed Dec 10 11:35:09 2008 +1000
78788
78789 dix: move MAX_VALUATOR_EVENTS into include/input.h
78790
78791 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78792
78793commit b36c398b11321a908cfe217108b26a32ffc1d850
78794Author: Peter Hutterer <peter.hutterer@who-t.net>
78795Date: Mon Dec 8 16:41:20 2008 +1000
78796
78797 dix: purge dead device-based window access code.
78798
78799 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78800
78801commit aa71ac1dd4697738752c73fe4c106b665ea3ec25
78802Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78803Date: Wed Dec 10 00:32:34 2008 -0200
78804
78805 Add dependency tracking to sdksyms.c and export composite wrapper.
78806
78807 Thanks to David Miller for noticing a make problem with sdksyms.c
78808 not being regenerated when sdksyms.sh is updated. This is not yet the
78809 best solution; automake generates dependency for sdksyms.o, but the
78810 build really should also regenerate sdksyms.c when sdksyms.o needs to
78811 be regenerated.
78812 Export the symbols in miext/cw/cw.h. These symbols are in libxaa, and
78813 at least the nvidia driver uses them. Maybe cw.h should be installed
78814 in the sdk.
78815
78816commit 090f63123975220819d531f569df6e5787607ec6
78817Author: Peter Hutterer <peter.hutterer@who-t.net>
78818Date: Mon Dec 8 11:53:20 2008 +1000
78819
78820 mi: always update the sprite for master devices.
78821
78822 Follow-up to 9ce995373e4a. This re-enables cursor rendering if the MD is
78823 controlled through software (e.g. synergy).
78824
78825 Reported by John Tapsell: "I use Xorg with no mouse attached, but use synergy
78826 to control the mouse. The commit means that I no longer have a visible mouse
78827 cursor. The mouse cursor is still 'there' in terms that I can click buttons
78828 etc with it, but it's just not visible."
78829
78830 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78831
78832commit 7c8720c1433d2c3b85bbf4b811cc54c2df4c0080
78833Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78834Date: Mon Dec 8 22:11:50 2008 -0200
78835
78836 Correct wrong symbol reference on sparc.
78837
78838 The awk script was incorrectly referencing the struct name, and
78839 not the struct variable.
78840 Also added some comments to sdksyms.sh, for the reason it generates
78841 the "symbol table" and add a message to the generated file, telling
78842 is was automatically generated.
78843
78844commit 58a27d2932164e43c0db42b1286ec2f95250b420
78845Author: Adam Jackson <ajax@redhat.com>
78846Date: Mon Dec 8 16:28:00 2008 -0500
78847
78848 Default to x86emu even on i386 linux
78849
78850 vm86 mode is a bad idea anyway, and using the emulator everywhere
78851 means we get a consistent set of bugs.
78852
78853commit bbf811514d3cdf84790bad5b852942a4e636902b
78854Author: Sascha Hlusiak <saschahlusiak@arcor.de>
78855Date: Mon Dec 8 12:24:39 2008 +0100
78856
78857 ddxCtrls.c: XkbDDXUsesSoftRepeat always returns 1 now
78858
78859 We'd like to do soft repeat in the server for all keys. Remove obscure check, that'd
78860 prevent the server from autorepeating when delay is set to exactly 660ms and rate is
78861 set to exactly 25 (interval=40).
78862
78863 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
78864
78865commit 8c1dd40a04693f09f4fcea6e7f905af34c7589de
78866Author: Benjamin Close <Benjamin.Close@clearchain.com>
78867Date: Mon Dec 8 14:49:38 2008 +1030
78868
78869 Don't use gnu specific extensions to awk when builing symbols
78870
78871 Traditional posix awk doesn't know about \W and whilst we check that
78872 awk exists in configure.ac we don't check which awk we are using.
78873 This corrects symbol generation for posix only awk.
78874
78875commit 39db182b637041255ed6dac739ff77c8e4e07c30
78876Author: Peter Hutterer <peter.hutterer@who-t.net>
78877Date: Mon Dec 8 12:12:39 2008 +1000
78878
78879 xfree86: init EQ before trying to initialise the devices (#18890)
78880
78881 The kbd driver may send events during device initialisation, and these events
78882 need the EQ set up already.
78883
78884 X.Org Bug 18890 <http://bugs.freedesktop.org/show_bug.cgi?id=18890>
78885
78886 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78887
78888commit 78a62d7713c708d067d8824ec41b0a0225c1997f
78889Author: Tom Jaeger <ThJaeger@gmail.com>
78890Date: Mon Dec 8 11:38:12 2008 +1000
78891
78892 Xi: XIGetDevice needs to ignore the MORE_EVENTS flag.
78893
78894 Reported in X.Org Bug 18882, Comment 5.
78895 <http://bugs.freedesktop.org/show_bug.cgi?id=18882>
78896
78897 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
78898
78899commit ee1a6c28418a6dad6c89f79a994f27bfbaa77368
78900Author: Peter Hutterer <peter.hutterer@redhat.com>
78901Date: Fri Dec 5 10:12:57 2008 +1000
78902
78903 dix: fix calculation of valuator events.
78904
78905 Follow-up to 4971315296cb. countValuatorEvents was copied from GKVE where it
78906 was obviously broken but nobody noticed. GPE had the correct version, but that
78907 one got lost during de-duplication. Restoring the correct calculation - if we
78908 have 6 valuators, we want 1 valuator event, not 2.
78909
78910 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78911
78912commit 13de7511b17b57a28668e1a60b196ccfe61dbcbe
78913Author: Peter Hutterer <peter.hutterer@redhat.com>
78914Date: Thu Dec 4 10:30:02 2008 +1000
78915
78916 xfree86: Only use the evdev ruleset on linux.
78917
78918 As suggested by Julien Cristau
78919
78920 This is an follow-up to
78921 commit 9c5dd7337fa93fb1650cc017e523b939dcbf482a
78922 Author: Peter Hutterer <peter.hutterer@redhat.com>
78923 Date: Wed Dec 3 14:24:25 2008 +1000
78924
78925 Let the DDX decide on the XkbRulesDefaults.
78926
78927 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78928 Acked-by: Julien Cristau <jcristau@debian.org>
78929 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
78930
78931commit 8e3279134987a45f2a89c963ef2d33bc3d3c8179
78932Author: Peter Hutterer <peter.hutterer@redhat.com>
78933Date: Thu Dec 4 09:47:25 2008 +1000
78934
78935 xfree86: fix compiler warning (use of uninitialized variable)
78936
78937 drv and idev are only set for SDs, but are only dereferenced for SDs too, so
78938 initializing them to NULL is safe.
78939
78940 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
78941
78942commit 22d4ddcc3d63b7fbf2a23f5162075e4ee06db781
78943Merge: d2dad38 c3bf15b
78944Author: Matthieu Herrb <matthieu@crux.(none)>
78945Date: Sun Dec 7 23:56:15 2008 +0100
78946
78947 Merge branch 'master' of git://anongit.freedesktop.org/git/xorg/xserver
78948
78949commit d2dad384f5725a36e593fd75f03d712f53b20620
78950Author: Matthieu Herrb <matthieu@crux.(none)>
78951Date: Sun Dec 7 23:52:26 2008 +0100
78952
78953 Fix build in separate build directory.
78954
78955commit c3bf15ba85fd3ab2e1fe809428b75bf34db25cc8
78956Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78957Date: Sun Dec 7 18:52:23 2008 -0200
78958
78959 Require macros 1.2.0 or newer for XORG_CHANGELOG and XORG_CWARNFLAGS.
78960
78961commit 5a8068b390b7d30eb526e954443d8e43a199b971
78962Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78963Date: Sun Dec 7 18:31:32 2008 -0200
78964
78965 Convert libx86emu.a to a "libtool convenience library".
78966
78967commit ed4a17274015ecd8040ae85bd5cd9d1dbcc9460a
78968Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78969Date: Sun Dec 7 14:59:25 2008 -0200
78970
78971 Export some symbols from libxf86config when installing it.
78972
78973 These are private symbols, but used by the X Server.
78974 The newly exported symbols were not added to the sdk headers.
78975 Optionally, libxf86config could be compiled without hidden symbols
78976 when being installed.
78977 Thanks to Maarten Maathuis for noticing the problem.
78978
78979commit bf4c29ab48a166eb158cf4be7c597982d65ef214
78980Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78981Date: Sun Dec 7 05:12:41 2008 -0200
78982
78983 Correct xf86acpiDisableFlag symbol.
78984
78985 It is declared as
78986 <hash>ifdef HAVE_ACPI
78987 Bool xf86acpiDisableFlag = FALSE;
78988 <hash>endif
78989 in hw/xfree86/common/xf86Globals.c
78990 but not protected by the ifdef in the sdk header xf86Priv.h, what
78991 caused a build failure in the tinderbox, due to the address of the
78992 symbol being taken (to ensure it is available) in sdksyms.c.
78993
78994commit b1dac41fb3853ca8182048ea57b88b6e84ecceb3
78995Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
78996Date: Sun Dec 7 02:22:19 2008 -0200
78997
78998 Use libtool convenience libraries and better "symbol" table.
78999
79000 All .a libraries were converted to .la, and instead of linking the
79001 Xorg binary with a mix of .a and .la, and adding some libraries more
79002 then once in the command line, etc, now it generates a single libxorg.la
79003 from all the required convenience libraries, and links with a dummy
79004 xorg.c (that should usually be the file with the main function...).
79005 This removes the requirement of some things like libosandcommon and
79006 libinit, that existed to circumvent problems when linking multiple
79007 .a and .la in the final Xorg binary.
79008
79009 The "symbol table" is now generated dynamically, by a shell script,
79010 with an embedded gawk parser that parses cpp output. The new file
79011 sdksyms.sh is generated by hand by analyzing all Makefile.am's and
79012 making it create a sdksyms.c file, that includes all sdk headers that
79013 will add symbols for the Xorg binary. Module headers aren't read, and
79014 a in 2 files it was required to add a "<hash>ifndef XorgLoader" around
79015 declarations shared between the Xorg binary and libextmod. A few
79016 other changes were added to other sdk headers, like preventing
79017 multiple inclusion, or including other headers to satisfy dependencies.
79018
79019 This should be a lot more portable, and better (hopefully properly)
79020 using libtool to generate convenience libraries.
79021
79022commit ccd2c668c13863ed704f86b29fc6b3ee628e56a2
79023Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79024Date: Sat Dec 6 14:11:52 2008 -0800
79025
79026 XQuartz: darwinPointer reports the actual pixel position now rather than a relative position
79027 (cherry picked from commit a41e7f75decd340d064fdc0d2c4fe6c88d7dbc82)
79028
79029commit 7ebd0c7e8d42a13079957d9bacf5cb30d7855a59
79030Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79031Date: Fri Dec 5 12:42:31 2008 -0800
79032
79033 XQuartz: Corrected name/command labels in the customization widget
79034 (cherry picked from commit a689c23f17eb445c36b97eb617ef4b8bd157985f)
79035
79036commit 027ff97a1354ab4c83fecb615f6bc2a6b739b871
79037Author: Alan Coopersmith <alan.coopersmith@sun.com>
79038Date: Fri Dec 5 12:00:08 2008 -0800
79039
79040 Make sure _X_EXPORT is defined in edid.h
79041
79042 Fixes build error in xf86-video-intel utilities
79043
79044commit 466b0fca9ba5d5e7fb36e47a6d1bd60218f51c75
79045Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79046Date: Fri Dec 5 17:01:09 2008 -0200
79047
79048 Add back a simplified version of the loader static address tables.
79049
79050 If not taking the symbol addresses, linkage will break badly, as not
79051 all symbols will be present, and it also requires changing library order,
79052 and/or making some changes like the "libosandcommon".
79053
79054 This table should be modified to be generated automatically, as
79055 it is required to "fool" the compiler/loader into adding all required
79056 symbols to the X Server.
79057
79058commit 44bef8b850c5a78a3d3eca5f0d92b71bdd0a87e2
79059Author: Matthias Hopf <mhopf@suse.de>
79060Date: Fri Dec 5 15:37:15 2008 +0100
79061
79062 randr: Update SProcRandrVector for panning
79063
79064commit 44c8c3cf7de589fb8e987f4ab931294eaf0b405f
79065Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79066Date: Fri Dec 5 04:26:57 2008 -0200
79067
79068 Remove static symbol address tables in hw/xfree86/loader/*sym.{c,h}
79069
79070 Those tables were once used to decide what symbols are visible to
79071 modules, but they were outdated. The only real usage was that, since
79072 it was taking the address of symbols, linkage should fail if the
79073 symbols were not available.
79074 Now the proper way to make symbols available to modules should
79075 be to use the _X_EXPORT macro, or not compile with hidden symbols,
79076 so that all symbols would be available.
79077 All symbols in the tables were revised to ensure they are exported,
79078 and only symbols that were not exported are ClientSleepUntil() and
79079 DuplicateModule(), that were not in the sdk for quite some time
79080 already, and should not have any users outside of the X Server
79081 (and/or builtin modules).
79082
79083commit 091a50d1260c70055aba1fa6a2d1b1a36de1114e
79084Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79085Date: Thu Dec 4 18:23:25 2008 -0200
79086
79087 Move _X_EXPORT attribute to header file.
79088
79089 The attribute should be set on header files to make it easier to
79090 manage what symbols are available to modules.
79091 _X_EXPORT should be used in sources only for special cases, like
79092 symbols that must be visible by non video/input driver/modules but
79093 should not be "advertised" in the sdk.
79094
79095commit e5ab9e66628cde081757cf2a1013a78e927a622e
79096Author: Matthias Hopf <mhopf@suse.de>
79097Date: Thu Dec 4 18:13:40 2008 +0100
79098
79099 randr: Allow panning to be disabled per axis
79100
79101commit bad118ace6c5bae5a5ed8a35129c90c38f1c1932
79102Author: Matthias Hopf <mhopf@suse.de>
79103Date: Thu Dec 4 16:55:14 2008 +0100
79104
79105 randr: Rework panning area verification
79106
79107commit 219c26ce0c65625d55cfd943ec66fe94a1a0ddfd
79108Author: Matthias Hopf <mhopf@suse.de>
79109Date: Thu Dec 4 16:28:40 2008 +0100
79110
79111 randr: Don't change panning parameters if verification fails.
79112
79113commit 18a8bac1a1567b6215928f96870554ea63f39aab
79114Author: Matthias Hopf <mhopf@suse.de>
79115Date: Thu Dec 4 16:30:38 2008 +0100
79116
79117 randr: Rename pan() to set_origin(), and xf86CrtcPan() to xf86CrtcSetOrigin()
79118
79119commit 825b2c2f4a59ac4852f90bbbddf18ab832297fdd
79120Author: Matthias Hopf <mhopf@suse.de>
79121Date: Thu Dec 4 16:11:21 2008 +0100
79122
79123 randr: Nuke config-timestamp for panning
79124
79125commit eeeb98d1df59baaaec954b6318d788a37e388d11
79126Author: Matthias Hopf <mhopf@suse.de>
79127Date: Fri Nov 28 17:51:20 2008 +0100
79128
79129 randr: Protocol bits for panning support
79130
79131commit b929d721efdb17bcc94b9984c4f34d0df3d267d5
79132Author: Matthias Hopf <mhopf@suse.de>
79133Date: Fri Nov 28 17:49:31 2008 +0100
79134
79135 randr: Panning support
79136
79137commit 834cbc16f3eb539704faade7bff347b161ce69d9
79138Author: Matthias Hopf <mhopf@suse.de>
79139Date: Fri Nov 28 17:39:23 2008 +0100
79140
79141 randr: Crtc interface update for panning support.
79142
79143commit a475eb9feec75e9ce1e316da0f1679acd7dd3aa8
79144Author: Matthias Hopf <mhopf@suse.de>
79145Date: Fri Nov 28 17:38:52 2008 +0100
79146
79147 randr: Weird enough, crtc->version was never set upon creation. Fix that.
79148
79149commit 9db84fac0cc767e23986223d22bc085cde0cc86f
79150Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79151Date: Tue Dec 2 14:54:10 2008 -0800
79152
79153 Removed unised option from configure.ac for launchd
79154 (cherry picked from commit 87e4de0a9a8af6e5e0cbba74e585761f1160a6ab)
79155
79156commit fa4f2527a5002711fe47c02b14097032fd845dc4
79157Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79158Date: Tue Dec 2 14:53:21 2008 -0800
79159
79160 XQuartz: pbproxy: Simplify linking
79161 (cherry picked from commit 909cc5c4dca0f63e90505575bbd454b46a4670cc)
79162
79163commit 126d8a4c8a72b039903023dce0da64b251080e1d
79164Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79165Date: Thu Dec 4 02:43:29 2008 -0200
79166
79167 Update .gitignore.
79168
79169 Add files left by make distcheck and remove files that no longer
79170 are generated.
79171
79172commit 9826b95c198e74f89680cc247b9bd9dd884d72be
79173Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79174Date: Thu Dec 4 02:05:05 2008 -0200
79175
79176 Correct make distcheck by removing dolt files.
79177
79178commit 5dbe70dd52fd8daf8d0797951cc5a758d2c3b44c
79179Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79180Date: Thu Dec 4 01:57:46 2008 -0200
79181
79182 Remove dummylib.
79183
79184 Only cvt required it, and only used XNF{,c}alloc
79185 via xnf{,c}alloc macros.
79186 Based on patch by Eric Anholt.
79187
79188commit f841d4e3cccbde02e91c948f5ffb9e32c8c3b3cc
79189Author: Julien Cristau <jcristau@debian.org>
79190Date: Wed Dec 3 23:02:03 2008 +0100
79191
79192 Move RELEASE_DATE closer to AC_INIT so it's more likely to be updated
79193
79194commit 75b02dd0be2ef64c0f8e3138c90b5c5e48b17913
79195Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79196Date: Wed Dec 3 14:46:30 2008 -0200
79197
79198 Include <X11/Xfuncproto.h> if _X_EXPORT is not defined.
79199
79200commit dd128ddcdcbe254a9cdd973590f6a979a7f0427e
79201Author: Timo Aaltonen <tjaalton@cc.hut.fi>
79202Date: Wed Dec 3 18:40:29 2008 +0200
79203
79204 If AEI is on, disable 'vmmouse' in addition to 'kbd' and 'mouse'.
79205
79206 Signed-off-by: Timo Aaltonen <tjaalton@cc.hut.fi>
79207 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79208
79209commit 0b5ecabfb803cd820338fb0364521fe39b05578b
79210Author: Julien Cristau <jcristau@debian.org>
79211Date: Wed Nov 26 21:19:55 2008 +0100
79212
79213 randr: add swapped dispatch for RR[GS]etCrtcTransform
79214
79215 Fix a memory leak in ProcRRGetCrtcTransform() while I'm at it.
79216
79217 Signed-off-by: Julien Cristau <jcristau@debian.org>
79218 Cc: Keith Packard <keithp@keithp.com>
79219
79220commit 110a71d11ab7a1a55a6a24d792457fdef0b0746d
79221Author: Kristian Høgsberg <krh@redhat.com>
79222Date: Wed Dec 3 11:22:38 2008 -0500
79223
79224 Test for DRI2 extension in dri_internal.h and only enable AIGLX DRI2 if found.
79225
79226commit fd2d40b7ec5d685dac55453eb1f2da672dc83126
79227Author: Peter Hutterer <peter.hutterer@redhat.com>
79228Date: Wed Dec 3 11:31:48 2008 +1000
79229
79230 Xi: change XIUnRegisterPropertyHandler to XIUnregisterPropertyHandler
79231
79232 CamelCase can be taken too far, and AFAICT there's no consumers of that
79233 function yet anyway.
79234
79235 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79236
79237commit 49f77fff1495c0a2050fb18f9b1fc627839bbfc2
79238Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79239Date: Wed Dec 3 05:43:34 2008 -0200
79240
79241 Rework symbol visibility for easier maintenance
79242
79243 Save in a few special cases, _X_EXPORT should not be used in C source
79244 files. Instead, it should be used in headers, and the proper C source
79245 include that header. Some special cases are symbols that need to be
79246 shared between modules, but not expected to be used by external drivers,
79247 and symbols that are accessible via LoaderSymbol/dlopen.
79248
79249 This patch also adds conditionally some new sdk header files, depending
79250 on extensions enabled. These files were added to match pattern for
79251 other extensions/modules, that is, have the headers "deciding" symbol
79252 visibility in the sdk. These headers are:
79253 o Xext/panoramiXsrv.h, Xext/panoramiX.h
79254 o fbpict.h (unconditionally)
79255 o vidmodeproc.h
79256 o mioverlay.h (unconditionally, used only by xaa)
79257 o xfixes.h (unconditionally, symbols required by dri2)
79258
79259 LoaderSymbol and similar functions now don't have different prototypes,
79260 in loaderProcs.h and xf86Module.h, so that both headers can be included,
79261 without the need of defining IN_LOADER.
79262
79263 xf86NewInputDevice() device prototype readded to xf86Xinput.h, but
79264 not exported (and with a comment about it).
79265
79266commit 0b8f8b24f718820a72ebdc52423c2e6a44e848c5
79267Author: Stuart Bennett <sb476@cam.ac.uk>
79268Date: Tue Dec 2 22:52:53 2008 -0800
79269
79270 xf86Cursors: xf86_reload_cursors shouldn't unconditionally show hwcursor (#14820)
79271
79272 Also, no need to call ShowCursor when SetCursorPosition already does it
79273 Based on a previous patch by Maarten Maathuis
79274
79275 Signed-off-by: Keith Packard <keithp@keithp.com>
79276
79277commit 9c5dd7337fa93fb1650cc017e523b939dcbf482a
79278Author: Peter Hutterer <peter.hutterer@redhat.com>
79279Date: Wed Dec 3 14:24:25 2008 +1000
79280
79281 Let the DDX decide on the XkbRulesDefaults.
79282
79283 Rather than assuming rules in the CoreKeyboardProc, init the default rules in
79284 InitCoreDevices, then re-use them later.
79285
79286 In the xfree86 DDX, set the rules to "base" or "evdev", depending on whether
79287 we'll load kbd or evdev.
79288
79289 If we create a new MD, use pc105,us as default and re-use the rules file used
79290 previously.
79291
79292 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79293
79294commit 463e02e7de5da3e582a3a049110a476713c7210e
79295Author: Peter Hutterer <peter.hutterer@redhat.com>
79296Date: Wed Dec 3 15:06:37 2008 +1000
79297
79298 xkb: Allow NULL as rulesFile in XkbSetRulesDflts.
79299
79300 If no rules file is given, simply re-use the previous one. If no RF is given
79301 the first time this function is called, use the built-in default.
79302 This includes fixing the built-in default to something that actually exists.
79303
79304 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79305
79306commit e670fd889607fa712876218882cd4a9b46937661
79307Author: Peter Hutterer <peter.hutterer@redhat.com>
79308Date: Wed Dec 3 11:55:13 2008 +1000
79309
79310 dix: fix GetMaximumEventsNum(), may return a DCCE event too.
79311
79312 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79313
79314commit 85d84c7cf2d368d1803dc2cdd03d6c9df8cc0430
79315Author: Eric Anholt <eric@anholt.net>
79316Date: Tue Dec 2 17:33:43 2008 -0800
79317
79318 Fix GLX after 180bad84774493d48f2793a6281d825560944863.
79319
79320 Sigh.
79321
79322commit 4a61ad427c634bcc38a31ce0f14ff5c2d3f706c7
79323Author: Alan Coopersmith <alan.coopersmith@sun.com>
79324Date: Tue Dec 2 14:35:45 2008 -0800
79325
79326 Update sample xorg.conf file
79327
79328commit c1db925d10fd37077bed90612ed95c3fd20cd2e2
79329Author: Alan Coopersmith <alan.coopersmith@sun.com>
79330Date: Tue Dec 2 14:32:48 2008 -0800
79331
79332 Add atKeynames.h to libdmxinput_a_SOURCES so it's included in tarballs
79333
79334 Reported by geearu in #xorg-devel
79335
79336commit 8561514574b3540c729bcc3acca9c943adcdc778
79337Author: Alan Coopersmith <alan.coopersmith@sun.com>
79338Date: Tue Dec 2 13:32:34 2008 -0800
79339
79340 Don't need to check uid/euid for every commandline argument
79341
79342 Check uid/euid only when handling the arguments that are restricted
79343 to root/non-setuid users
79344
79345commit 6de6ffff35ac03d49fa61de195d4a0605e0ef8bf
79346Author: Alan Coopersmith <alan.coopersmith@sun.com>
79347Date: Tue Dec 2 13:19:08 2008 -0800
79348
79349 Remove unused config flags from FlagValues & FlagOptions
79350
79351 FLAG_DISABLEMODINDEV, FLAG_MODINDEVALLOWNONLOCAL, & FLAG_HANDLE_SPECIAL_KEYS
79352
79353commit bb072019fa8dd292a50ef433d05caeefd1304a73
79354Author: Peter Hutterer <peter.hutterer@redhat.com>
79355Date: Mon Dec 1 16:57:06 2008 +1000
79356
79357 xfree86: don't render SW cursors for devices attached to VCP (#16805)
79358
79359 When leaving 3D games such as quake3 or sauerbraten, a cursor may stay on the
79360 screen. This is caused by one run of SW rendering for the SD, even though the
79361 SD was attached to the VCP and thus has HW rendering capabilities.
79362
79363 Check for the SD's attachment (like in all other functions) before deciding on
79364 SW or HW rendering.
79365
79366 X.Org Bug 16805 <http://bugs.freedesktop.org/show_bug.cgi?id=16805>
79367
79368 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79369
79370commit 3a690598cf18c4cdc6aadd10a1ecf0772cacd34b
79371Author: Alan Coopersmith <alan.coopersmith@sun.com>
79372Date: Tue Dec 2 12:59:57 2008 -0800
79373
79374 Remove unused HandleSpecialKeys config option
79375
79376 Was only used by "keyboard" driver - the last place the value was checked
79377 was in xf86PostKbdEvent, which was removed in commit
79378 60ea7b51fe2b8a19a08b63db48504971a7a50ee6
79379
79380commit 1dfed222e93f4684c2a450944a9a0ea9e085c43f
79381Author: Bernhard Rosenkraenzer <bero@arklinux.org>
79382Date: Tue Dec 2 09:01:56 2008 +1000
79383
79384 Xext: fix MultiBuffer compilation error with TryClientEvents. (#18835)
79385
79386 X.Org Bug 18835 <http://bugs.freedesktop.org/show_bug.cgi?id=18835>
79387
79388 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79389
79390commit 9ce995373e4aa4b1bf51b4adafa2324f781ec1cd
79391Author: Peter Hutterer <peter.hutterer@redhat.com>
79392Date: Mon Dec 1 21:32:39 2008 +1000
79393
79394 mi: UpdateSprite only if the device is attached.
79395
79396commit d507f60689f4e14383b0d24e63afc8cf836360d5
79397Author: Peter Hutterer <peter.hutterer@redhat.com>
79398Date: Wed Nov 26 14:15:04 2008 +1000
79399
79400 xfree86: don't FatalError on "too many input devices".
79401
79402 Just ignore devices after MAXDEVICES has been reached, but warn the user that
79403 the devices are ignored.
79404
79405 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79406
79407commit 9bf761468fa3baf4c5fa40efa717b4b581d920f9
79408Author: Peter Hutterer <peter.hutterer@redhat.com>
79409Date: Fri Nov 28 11:44:58 2008 +1000
79410
79411 Xi: fix xi_filters size.
79412
79413commit 260e48c252c95fd1b1ba9ad7478791f6b9a67e2d
79414Author: Peter Hutterer <peter.hutterer@redhat.com>
79415Date: Fri Nov 28 11:29:06 2008 +1000
79416
79417 dix: remove confusing (and wrong) comment. VCP is not the only pointer.
79418
79419 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79420
79421commit 5d02e580d7c4144ea6f0984240fc913e05fec877
79422Author: Peter Hutterer <peter.hutterer@redhat.com>
79423Date: Fri Nov 28 11:46:15 2008 +1000
79424
79425 dix: use UpdateFromMaster in GetProximityEvents.
79426
79427 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79428
79429commit 95fc59a199f99bf167fbb09297a9bb0e33e31869
79430Author: Peter Hutterer <peter.hutterer@redhat.com>
79431Date: Fri Nov 28 14:55:15 2008 +1000
79432
79433 xkb: Extra sanity checks to prevent dev->key == NULL dereferencing.
79434
79435commit a425abf0eaa61ee5ccb2f32a1af24edd190f7889
79436Author: Peter Hutterer <peter.hutterer@redhat.com>
79437Date: Wed Nov 26 12:20:00 2008 +1000
79438
79439 xkb: don't attempt to filter events for devices without key classes.
79440
79441 Reported by Magnus Kessler.
79442
79443 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79444
79445commit a85f0d6b98237d8a196de624207acf1983a1859a
79446Author: Peter Hutterer <peter.hutterer@redhat.com>
79447Date: Wed Nov 26 11:15:05 2008 +1000
79448
79449 Xi: fix use of button->down - bitflags instead of int arrays.
79450
79451 The device's button down state array was changed to use DOWN_LENGTH and thus
79452 bitflags for each button in cfcb3da7.
79453
79454 Update the DBSN events to copy this bit-wise state.
79455 Update xkb and Xi to check for the bit flag instead of the array value.
79456
79457 Reported by ajax.
79458
79459 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79460
79461commit 180bad84774493d48f2793a6281d825560944863
79462Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79463Date: Tue Dec 2 02:50:45 2008 -0200
79464
79465 Add visibility flags to XSERVER_CFLAGS.
79466
79467 This is done to actually change DIX_CFLAGS, as not all "modules" use
79468 XORG_CFLAGS.
79469 Also export the symbols that are required by other modules after
79470 the change.
79471
79472commit 8c560422b44e012053612754430d2b87dc44ed59
79473Author: Alan Coopersmith <alan.coopersmith@sun.com>
79474Date: Mon Dec 1 20:02:34 2008 -0800
79475
79476 More man page updates for 1.6 release for Xorg, xorg.conf & exa man pages.
79477
79478 - Remove remaining references to XFree86-Misc options AllowNonLocalModInDev
79479 and DisableModInDev.
79480 - Remove remaining references to grab-breaking keys & associated options.
79481 - Update description of Ctrl-Alt-Backspace to new -retro/DontZap defaults.
79482 - Add description of new options -modalias and -showopts.
79483 - Update list of modules loaded by default.
79484 - Update input driver references from keyboard to evdev & kbd.
79485 - Update list of driver man pages to match xf86-*-* drivers with man pages.
79486 - Add See Also section to exa man page.
79487
79488 and various formatting/typo/etc. fixes.
79489
79490 The Xorg/xorg.conf sections on input device selection could use further
79491 updates to better match the current state of HAL-enabled configuration.
79492
79493commit 6a1850b8c677e2a2993f6f6b731ee3d35aa55d09
79494Author: Alan Coopersmith <alan.coopersmith@sun.com>
79495Date: Mon Dec 1 19:19:04 2008 -0800
79496
79497 Correct warning for unknown GlxVisuals option in conf file
79498
79499commit afeacb8e74b2a1e366e5ca9f0e58805d8d96c457
79500Author: Kristian Høgsberg <krh@redhat.com>
79501Date: Mon Dec 1 20:53:17 2008 -0500
79502
79503 Bump dri2proto requirement to 1.99.3.
79504
79505commit ad01e86b5c7c528adec8a1f95ecaa294f58a8922
79506Author: Kristian Høgsberg <krh@redhat.com>
79507Date: Mon Dec 1 20:52:41 2008 -0500
79508
79509 Drop unused DRI2 vblank infrastructure.
79510
79511 For this first iteration of DRI2 we don't have any vsync functionality
79512 in place yet, so back out the support in DRI2 and the protocol for now.
79513
79514commit 24c562f04b41d219c34f5fa3f963564accf329f2
79515Author: Alan Coopersmith <alan.coopersmith@sun.com>
79516Date: Mon Dec 1 16:28:45 2008 -0800
79517
79518 Update See Also lists in Xorg & xorg.conf man pages
79519
79520 Remove xorgconfig & xorgcfg
79521 Update driver lists:
79522 - Remove vga
79523 - i810->intel
79524 - via->openchrome
79525
79526commit bd147e6f4b03f344c967c88fd433877b14fe1912
79527Author: Alan Coopersmith <alan.coopersmith@sun.com>
79528Date: Mon Dec 1 16:08:55 2008 -0800
79529
79530 Remove hack to only load font modules if magic strings appear in font path
79531
79532commit 5f3f14179edf48aad518f6f707bfdc37c27267c6
79533Author: Jay Cotton <jay.cotton@sun.com>
79534Date: Mon Dec 1 16:06:09 2008 -0800
79535
79536 Sun bug 6618220: Xorg server core dump in xf86RandRModeRefresh(NULL)
79537
79538 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6618220>
79539
79540commit 09df7cc5ad7b72d8a23c3e22fc718aad8c16f4d3
79541Author: Kristian Høgsberg <krh@redhat.com>
79542Date: Mon Dec 1 12:41:10 2008 -0500
79543
79544 Avoid dereferencing NULL pScreen in xf86CrtcSetModeTransform().
79545
79546 We can get there during PreInit as we set a mode for load detection.
79547 At that time there's no pScreen anywhere, so just skip the optimization
79548 then.
79549
79550commit b0d371ab0a6efd4956c3677faa20b2ac15c33765
79551Author: Adam Jackson <ajax@redhat.com>
79552Date: Mon Dec 1 11:36:06 2008 -0500
79553
79554 randr: Don't send output property events on server exit
79555
79556 If the Window resource type is already gone, there's no point in trying
79557 to send events, all it can do is access already-freed memory.
79558
79559 Relevant thread:
79560
79561 http://lists.freedesktop.org/archives/xorg/2008-November/040443.html
79562
79563commit 16b11cd03d8c5def07f0e598f237f71a37883a46
79564Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79565Date: Sun Nov 30 02:59:34 2008 -0200
79566
79567 Correct static symbol XkmReadTOC and first pass on compile warning fixes.
79568
79569 The warnings corrected were only the ones that should correct
79570 real problems. The most common one is 64 bit integers as
79571 "printf %l" arguments.
79572 Note that there is a patch related to this at:
79573 http://bugs.freedesktop.org/show_bug.cgi?id=18204
79574
79575commit fb22d4d928bc5d1a893494a059359da2ed2ac639
79576Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79577Date: Sun Nov 30 01:33:20 2008 -0200
79578
79579 Remove declarations of symbols that are never defined.
79580
79581 These symbols were removed from the X Server, or never declared.
79582 One symbol that may need special attention is XkbBuildCoreState(),
79583 that doesn't have a prototype anywhere, but is called from
79584 xkb/xkbEvents.c:XkbFilterEvents(), and also used by the macros
79585 XkbStateFieldFromRec() and XkbGrabStateFromRec() defined in
79586 include/xkbstr.h.
79587 fb/wfbrename.h also may need some cleanup, as it makes several
79588 "renames" of non existing symbols.
79589
79590commit d6cbd4511e35a89a0353f11834c6fdb8d4d2189f
79591Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79592Date: Sat Nov 29 23:56:06 2008 -0200
79593
79594 Export symbols defined in the sdk.
79595
79596 This is the biggest "visibility" patch. Instead of doing a "export"
79597 symbol on demand, export everything in the sdk, so that if some module
79598 fails due to an unresolved symbol, it is because it is using a symbol
79599 not in the sdk.
79600
79601 Most exported symbols shouldn't really be made visible, neither
79602 advertised in the sdk, as they are only used by a single shared object.
79603
79604 Symbols in the sdk (or referenced in sdk macros), but not defined
79605 anywhere include:
79606 XkbBuildCoreState()
79607 XkbInitialMap
79608 XkbXIUnsupported
79609 XkbCheckActionVMods()
79610 XkbSendCompatNotify()
79611 XkbDDXFakePointerButton()
79612 XkbDDXApplyConfig()
79613 _XkbStrCaseCmp()
79614 _XkbErrMessages[]
79615 _XkbErrCode
79616 _XkbErrLocation
79617 _XkbErrData
79618 XkbAccessXDetailText()
79619 XkbNKNDetailMaskText()
79620 XkbLookupGroupAndLevel()
79621 XkbInitAtoms()
79622 XkbGetOrderedDrawables()
79623 XkbFreeOrderedDrawables()
79624 XkbConvertXkbComponents()
79625 XkbWriteXKBSemantics()
79626 XkbWriteXKBLayout()
79627 XkbWriteXKBKeymap()
79628 XkbWriteXKBFile()
79629 XkbWriteCFile()
79630 XkbWriteXKMFile()
79631 XkbWriteToServer()
79632 XkbMergeFile()
79633 XkmFindTOCEntry()
79634 XkmReadFileSection()
79635 XkmReadFileSectionName()
79636 InitExtInput()
79637 xf86CheckButton()
79638 xf86SwitchCoreDevice()
79639 RamDacSetGamma()
79640 RamDacRestoreDACValues()
79641 xf86Bpp
79642 xf86ConfigPix24
79643 xf86MouseCflags[]
79644 xf86SupportedMouseTypes[]
79645 xf86NumMouseTypes
79646 xf86ChangeBusIndex()
79647 xf86EntityEnter()
79648 xf86EntityLeave()
79649 xf86WrapperInit()
79650 xf86RingBell()
79651 xf86findOptionBoolean()
79652 xf86debugListOptions()
79653 LoadSubModuleLocal()
79654 LoaderSymbolLocal()
79655 getInt10Rec()
79656 xf86CurrentScreen
79657 xf86ReallocatePciResources()
79658 xf86NewSerialNumber()
79659 xf86RandRSetInitialMode()
79660 fbCompositeSolidMask_nx1xn
79661 fbCompositeSolidMask_nx8888x0565C
79662 fbCompositeSolidMask_nx8888x8888C
79663 fbCompositeSolidMask_nx8x0565
79664 fbCompositeSolidMask_nx8x0888
79665 fbCompositeSolidMask_nx8x8888
79666 fbCompositeSrc_0565x0565
79667 fbCompositeSrc_8888x0565
79668 fbCompositeSrc_8888x0888
79669 fbCompositeSrc_8888x8888
79670 fbCompositeSrcAdd_1000x1000
79671 fbCompositeSrcAdd_8000x8000
79672 fbCompositeSrcAdd_8888x8888
79673 fbGeneration
79674 fbIn
79675 fbOver
79676 fbOver24
79677 fbOverlayGeneration
79678 fbRasterizeEdges
79679 fbRestoreAreas
79680 fbSaveAreas
79681 composeFunctions
79682 VBEBuildVbeModeList()
79683 VBECalcVbeModeIndex()
79684 TIramdac3030CalculateMNPForClock()
79685 shadowBufPtr
79686 shadowFindBuf()
79687 miRRGetScreenInfo()
79688 RRSetScreenConfig()
79689 RRModePruneUnused()
79690 PixmanImageFromPicture()
79691 extern int miPointerGetMotionEvents()
79692 miClipPicture()
79693 miRasterizeTriangle()
79694 fbPush1toN()
79695 fbInitializeBackingStore()
79696 ddxBeforeReset()
79697 SetupSprite()
79698 InitSprite()
79699 DGADeliverEvent()
79700
79701 SPECIAL CASES
79702 o defined as _X_INTERNAL
79703 xf86NewInputDevice()
79704 o defined as static
79705 fbGCPrivateKey
79706 fbOverlayScreenPrivateKey
79707 fbScreenPrivateKey
79708 fbWinPrivateKey
79709 o defined in libXfont.so, but declared in xorg/dixfont.h
79710 GetGlyphs()
79711 QueryGlyphExtents()
79712 QueryTextExtents()
79713 ParseGlyphCachingMode()
79714 InitGlyphCaching()
79715 SetGlyphCachingMode()
79716
79717commit ffb484f7ef84099019b196ef97bfb2355eb6d52a
79718Author: Maarten Maathuis <madman2003@gmail.com>
79719Date: Sat Nov 29 14:30:55 2008 +0100
79720
79721 randr: Avoid needlessly creating a shadow framebuffer.
79722
79723 - pScreen->width and height were zero, so it didn't "fit" the screen.
79724
79725commit 01eaebdc98f30fdf543af6337cdf012d3ff16f09
79726Author: George Staplin <gstaplin@apple.com>
79727Date: Fri Nov 28 13:57:45 2008 -0700
79728
79729 XQuartz: GL: Make various changes to makeFormat, so that it works better. Now glxgears looks normal with the old libGL.
79730
79731 Add various GLX extensions to the list of supported extensions.
79732
79733 Reformat the oddly formatted code in some areas.
79734
79735 Use xalloc and xfree instead of malloc and free.
79736 (cherry picked from commit 755f53eb40c4329d8c82a31cb910c31fdd4ea12e)
79737
79738commit 3d527b91b4bfa31e78d5fc7a1447a4026876f14e
79739Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79740Date: Fri Nov 28 12:59:09 2008 -0800
79741
79742 XQuartz: Added option to enable/disable test extensions
79743
79744 defaults write org.x.X11 enable_test_extensions -boolean true
79745 (cherry picked from commit fd4710aff3723d5f3422cf6a6530172eafc556d9)
79746 (cherry picked from commit 635019ad18db921fec99256294debd8571074abf)
79747
79748commit 71d3df7cc37403b5842227a035d9b995fd920e0b
79749Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79750Date: Fri Nov 28 11:18:14 2008 -0800
79751
79752 XQuartz: Avoid some warning messages being spewed to system.log by AppKit
79753 (cherry picked from commit 780eff230ce41ad785e54d61a82c731269ae0446)
79754
79755commit a9796c7bc43223df44f12a7be08e361aea963ec1
79756Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79757Date: Fri Nov 28 01:45:37 2008 -0800
79758
79759 XQuarz: Setup our PATH and PWD earlier, so our initial client benefits from it as well...
79760 (cherry picked from commit cbae2b447357b3fed6ff19414c60dd3792045600)
79761
79762commit ef4179f43a84d90f867b95ee5974ad26884253b5
79763Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79764Date: Wed Nov 26 12:15:58 2008 -0800
79765
79766 XQuartz: Fix dead-acute on Greek keyboards
79767 (cherry picked from commit 807f2ec35c5f95b56b3e2b5eac51aec300fe5cb6)
79768
79769commit 800f5c681bed300d8caf99935e8f80ea6c1993ba
79770Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79771Date: Wed Nov 26 12:07:52 2008 -0800
79772
79773 XQuartz: Fix Czech keyboard dead-acute
79774 (cherry picked from commit 771df2786bc60389489f0967e705c7c95bdda085)
79775
79776commit ac57bb36d56a7a4d41add8d5a206ff37544a1819
79777Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
79778Date: Tue Nov 25 20:14:37 2008 -0800
79779
79780 XQuartz: Fix an uninitialized keyboard_type on Tiger
79781 (cherry picked from commit 27b1a5eb3482052253ebdce1a09aedf05ac1b099)
79782
79783commit 45c8bd0fe54273039fdaa1eeeafb81b5774f2c75
79784Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79785Date: Fri Nov 28 02:01:57 2008 -0200
79786
79787 Enable compiling the X Server and modules with hidden symbols by default.
79788
79789 Note that it checks if support for symbol visibility is available by
79790 the compiler, not if it is functional. It may have problems on non x86
79791 architectures.
79792 To disable the feature, just pass the option "--disable-visibility"
79793 to the X Server configure script.
79794 Unless using an alternate build schema, drivers built from a git
79795 checkout will use the same "visibility" related CFLAGS used to compile
79796 the X Server.
79797
79798commit 31285d063ec4623cb0764437d6d57e244f20460d
79799Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79800Date: Fri Nov 28 01:55:11 2008 -0200
79801
79802 Make visible symbols required by xorg modules.
79803
79804 This patch exports all symbols required by the compilable
79805 (in a x86 linux computer) xorg/driver/* modules.
79806 Still missing symbols worth mentioning are:
79807
79808 sunleo
79809 miFindMaxBand no longer available
79810
79811 intel (uxa/uxa-accel.c)
79812 fbShmPutImage no longer available (and should have been static)
79813
79814 mga
79815 MGAGetClientPointer (should come from matrox's libhal)
79816
79817 This is not a definitive "visibility" patch, as all it does is to
79818 export missing symbols, but the modules that current don't compile,
79819 may require more symbols once fixed, and third party drivers should
79820 also require more symbols exported.
79821 A "definitive" patch should export symbols defined in the sdk.
79822
79823commit 6f18ea7e656b79e58f66f51067334414767b413f
79824Author: Julien Cristau <jcristau@debian.org>
79825Date: Wed Nov 26 19:07:04 2008 +0100
79826
79827 Typo fix
79828
79829commit efefc03f9b26738b69abb570c1af2167cc99c4d8
79830Author: Julien Cristau <jcristau@debian.org>
79831Date: Sun Oct 26 13:19:48 2008 +0100
79832
79833 Add missing include
79834
79835 ../../../../hw/xfree86/xaa/xaaTables.c:9:14: warning: symbol 'byte_expand3' was not declared. Should it be static?
79836 ../../../../hw/xfree86/xaa/xaaTables.c:53:14: warning: symbol 'byte_reversed_expand3' was not declared. Should it be static?
79837
79838commit 74bc792e1814849b8eee9e448e36c3568b821e6c
79839Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79840Date: Thu Nov 27 05:34:14 2008 -0200
79841
79842 Export symbols required by the vesa and fbdev drivers.
79843
79844 This is a minimal set of patches, to have an usable X Server when
79845 compiling it with symbols hidden by default.
79846
79847commit 74e2669103b59d51b24ddc48f378b1d04a3effd4
79848Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79849Date: Thu Nov 27 04:33:44 2008 -0200
79850
79851 Export symbols also defined in libXfont.
79852
79853 libXfont has stubs for these symbols, so, when compiling with hidden
79854 symbols by default, these symbols must be visible in the X Server, or
79855 the stubs in libXfont will be used.
79856
79857commit 87a7fb7438b70d07ebf240c530b8548a01021d48
79858Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
79859Date: Thu Nov 27 00:12:59 2008 -0200
79860
79861 Rework code using return value of LoaderSymbol as a function pointer.
79862
79863 The patch removes all macros in the format
79864 define xf86_sym ((type (*)(argument-list))LoaderSymbol("sym"))
79865 creates a new macro in the format
79866 define xf86_sym sym
79867 and ensures "sym" is a "visible" symbol.
79868 The patch doesn't add or remove features, and is source and binary
79869 compatible with previous shared objects (with the difference that it
79870 requires the dlloader).
79871 These symbols are a special case, as, due to the fact that LoaderSymbol
79872 was being used to reference them, they are not easily found by "automated"
79873 tools that check for missing symbols. And now it also have the benefit
79874 that the compiler/loader "knows what is going on".
79875
79876commit fc708394318ed92c612e2938b335c08c1ffebb28
79877Author: Keith Packard <keithp@keithp.com>
79878Date: Wed Nov 26 15:49:02 2008 -0800
79879
79880 Add server support for RRGetScreenResourcesCurrent
79881
79882 This depends on randrproto 1.2.99.1
79883
79884 Signed-off-by: Keith Packard <keithp@keithp.com>
79885
79886commit f710ce269c020a39f58b2bfbd0fe5192a3279c72
79887Author: Alan Hourihane <alanh@tungstengraphics.com>
79888Date: Wed Nov 26 20:02:44 2008 +0000
79889
79890 bump master to 1.6.99.1 (now the 1.6 branch is created)
79891
79892commit f3edc1fb0210149f35eab4e413700b5c4ac48214
79893Author: Keith Packard <keithp@keithp.com>
79894Date: Tue Nov 25 23:15:35 2008 -0800
79895
79896 New version of dolt
79897
79898commit 4715f079b9c61362755c95c1ebf89c97cc6fff2b
79899Merge: 9ffc671 ed597f1
79900Author: Keith Packard <keithp@keithp.com>
79901Date: Tue Nov 25 22:19:08 2008 -0800
79902
79903 Merge commit 'origin/master'
79904
79905commit 9ffc6719390df8fdd0a5295a7a7a0eaea792be45
79906Author: Keith Packard <keithp@keithp.com>
79907Date: Mon Nov 24 13:08:48 2008 -0800
79908
79909 Move matrix operations from X server to pixman 0.13.2
79910
79911 pixman 0.13.2 now holds all of the matrix operations. This leaves
79912 the protocol conversion routines and some ABI stubs in place
79913
79914 Signed-off-by: Keith Packard <keithp@keithp.com>
79915
79916commit ed597f19fdc3017dde6d1452b5cdf8ddcd69a5b1
79917Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
79918Date: Tue Nov 25 19:36:31 2008 -0500
79919
79920 xselinux: use "raw context" variants of getpeercon() and getcon().
79921
79922commit ec1d08442f69353cb0e73ac4eaf0346ebb975594
79923Author: Peter Hutterer <peter.hutterer@who-t.net>
79924Date: Fri Nov 21 15:13:00 2008 +1000
79925
79926 dix: Enable core devices in InitCoreDevices already.
79927
79928 Updated patch, see
79929 http://lists.freedesktop.org/archives/xorg/2008-November/040540.html
79930
79931 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79932 Signed-off-by: Adam Jackson <ajax@redhat.com>
79933
79934commit 2b45602e828a07a0817691b2838cd34ffee531bd
79935Author: Peter Hutterer <peter.hutterer@redhat.com>
79936Date: Wed Nov 26 10:42:52 2008 +1000
79937
79938 Revert "dix: Enable core devices in InitCoreDevices already."
79939
79940 I merged the wrong patch. See correct patch at:
79941 http://lists.freedesktop.org/archives/xorg/2008-November/040540.html
79942
79943 Not activating the device before attempting to enable it would leave the
79944 sprite unset, crashing the server when enabling the real devices.
79945
79946 This reverts commit e078901a4eca02bd3e7a80d9462dafbca939a187.
79947
79948 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
79949
79950commit 416685c295353b5816689994c7c58ae7db3e878d
79951Author: Jeremy Uejio <jeremy.uejio@sun.com>
79952Date: Tue Nov 25 16:26:44 2008 -0800
79953
79954 Refix Sun bug #6685465: Xephyr uses wrong or bad colortable in 8-bit mode
79955
79956 <http://bugs.opensolaris.org/view_bug.do?bug_id=6685465>
79957
79958 This is a refix of the previous fix for CR 6685465. In the first fix
79959 I was shifting the colors to match the mask by the bits_per_rgb amount
79960 in the visual structure. That field has nothing to do with the # of
79961 bits to shift by. I should just instead shift the bits to match the mask.
79962
79963commit d5f9a131a2d5bd33f82fdd4e809880b0ff792b45
79964Author: Alan Coopersmith <alan.coopersmith@sun.com>
79965Date: Tue Nov 25 15:46:39 2008 -0800
79966
79967 Fix const-mismatch warnings for DisplayModePtr's
79968
79969 Includes fixes for:
79970 "xf86Config.c", line 2434: warning: argument #1 is incompatible with prototype:
79971 prototype: pointer to struct _DisplayModeRec: "xf86.h", line 351
79972 argument : pointer to const struct _DisplayModeRec
79973
79974 "xf86EdidModes.c", line 312: warning: argument #1 is incompatible with prototype:
79975 prototype: pointer to struct _DisplayModeRec: "../../../hw/xfree86/common/xf86.h", line 351
79976 argument : pointer to const struct _DisplayModeRec
79977
79978 "xf86EdidModes.c", line 438: warning: assignment type mismatch:
79979 pointer to struct _DisplayModeRec "=" pointer to const struct _DisplayModeRec
79980
79981 "xf86Modes.c", line 701: warning: assignment type mismatch:
79982 pointer to struct _DisplayModeRec "=" pointer to const struct _DisplayModeRec
79983
79984commit d5ad296869c38ab30136b5a293a0125b76aad994
79985Author: Alan Coopersmith <alan.coopersmith@sun.com>
79986Date: Tue Nov 25 14:12:26 2008 -0800
79987
79988 Remove duplication from code paths in XkbDDXCompileKeymapByNames
79989
79990commit 2538fc0d893a150e978355d281750f0a989728a7
79991Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
79992Date: Tue Nov 25 18:18:46 2008 -0500
79993
79994 xselinux: don't pass a NULL key string to selabel_lookup().
79995
79996commit c8472a74441838e16d0d3414db1fa7fe996868a9
79997Author: Michael Vogt <michael.vogt@ubuntu.com>
79998Date: Tue Nov 25 19:35:17 2008 +0200
79999
80000 Do not send VisibilityNotify events when MapUnmapEvents are disabled
80001
80002 This prevents a protocol visible side-effect (XVisibilityEvent) on
80003 XCompositeRedirectWindow() followed by a XCompositeUnredirectWindow().
80004
80005 The problem shows up in gnome-screensaver with compiz and "unredirect
80006 fullscreen windows" enable. A VisibilityNotify event is generated (first
80007 with obscured and than with unobscured) when the window swithces from
80008 redirected to unredirected.
80009
80010 https://bugs.freedesktop.org/show_bug.cgi?id=18133
80011 http://launchpad.net/bugs/278112
80012
80013commit 88297558aada44bc714ad57adbeed3740aaadee5
80014Author: Adam Jackson <ajax@redhat.com>
80015Date: Tue Nov 25 14:20:58 2008 -0500
80016
80017 Dead code cleanup over DBE.
80018
80019 No DDXes have explicit DBE support anymore, so the init registration
80020 table never got used. Just nuke it all.
80021
80022commit e5eaea599ab16428c69912b6b3427ebe46707d7c
80023Author: Adam Jackson <ajax@redhat.com>
80024Date: Tue Nov 25 14:05:51 2008 -0500
80025
80026 Warning fix
80027
80028 xf86info.c:11: warning: initialization makes integer from pointer
80029 without a cast
80030
80031commit 37072500f7bcf39e0d6aa2ceb5d1f2aeeab0b26b
80032Author: Adam Jackson <ajax@redhat.com>
80033Date: Tue Nov 25 14:04:44 2008 -0500
80034
80035 Warning fix
80036
80037 helper_exec.c: In function ‘port_rep_inb’:
80038 helper_exec.c:219: warning: implicit declaration of function
80039 ‘DEBUG_IO_TRACE’
80040 helper_exec.c:219: warning: nested extern declaration of
80041 ‘DEBUG_IO_TRACE’
80042
80043commit d96bffce2dcf209e76be9b36ca1ede7e0c976d77
80044Author: Adam Jackson <ajax@redhat.com>
80045Date: Tue Nov 25 13:59:00 2008 -0500
80046
80047 Warning fix
80048
80049 Init.c:139: warning: no previous prototype for ‘ddxBeforeReset’
80050
80051 Just declare the prototype always, seriously.
80052
80053commit 09ea671cbff605fd2b2af71619e7db5002108bf8
80054Author: Adam Jackson <ajax@redhat.com>
80055Date: Tue Nov 25 13:40:47 2008 -0500
80056
80057 Warning fix
80058
80059 lnx_bell.c:37: warning: no previous prototype for ‘xf86OSRingBell’
80060
80061commit a9853c7d337b3b1ad49793e9b4b90e313b6fa536
80062Author: Adam Jackson <ajax@redhat.com>
80063Date: Tue Nov 25 00:53:55 2008 -0500
80064
80065 Warning fix
80066
80067 Pixmap.c: In function ‘xnestPixmapToRegion’:
80068 Pixmap.c:93: warning: ‘Box.x1’ may be used uninitialized in this
80069 function
80070
80071commit 81eafe9f93a272b06aa9f9235ec5676b9aa3ee3e
80072Author: Adam Jackson <ajax@redhat.com>
80073Date: Tue Nov 25 00:45:07 2008 -0500
80074
80075 Warning fix.
80076
80077 vbe.c: In function ‘VBEReadPanelID’:
80078 vbe.c:1145: warning: return from incompatible pointer type
80079
80080commit 09bfb25e031772611a2f0902d4ba77b587e4bdb2
80081Author: Adam Jackson <ajax@redhat.com>
80082Date: Tue Nov 25 00:43:36 2008 -0500
80083
80084 Remove unused XAAAvailableOptions
80085
80086commit 8b9253f6383df3fefe38bde43a5f892b158a77c4
80087Author: Adam Jackson <ajax@redhat.com>
80088Date: Tue Nov 25 00:38:47 2008 -0500
80089
80090 Code motion: subsume xf86DoProbe.c into xf86Init.c
80091
80092commit fbabb1c5c243cfd8c954dec4c060dff1a0b81015
80093Author: Adam Jackson <ajax@redhat.com>
80094Date: Tue Nov 25 00:34:28 2008 -0500
80095
80096 Warning fix.
80097
80098 In file included from l3-xaaStipple.c:4:
80099 ./xaaStipple.c:35: warning: no previous prototype for
80100 ‘XAAGetStippleScanlineFunc3LSBFirst’
80101
80102 etc
80103
80104commit cbaca6ec666d7349c4680b8affc13b5c9cae1fa5
80105Author: Adam Jackson <ajax@redhat.com>
80106Date: Tue Nov 25 00:14:24 2008 -0500
80107
80108 Remove dead FreeModuleDesc.
80109
80110commit eb474adf98229a43bbe17ab98ff084371cb9fa09
80111Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80112Date: Tue Nov 25 11:25:58 2008 -0800
80113
80114 XQuartz: Simplify the xquartz_resetenv_display path
80115 (cherry picked from commit d2e0624dd30eb234bb25595ceedfa51d48ca1724)
80116
80117commit 94df1ab7f09a64f57c1e1453e3640462e984619c
80118Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80119Date: Tue Nov 25 01:30:03 2008 -0800
80120
80121 XQuartz: Force X11Controller to reset a broken DISPLAY envvar.
80122 (cherry picked from commit f1a52b5b5ac31702497937efe3ac578be9a6c54f)
80123
80124commit 0b314c50a2a0ca1afbdc06663c3b719b05ebb851
80125Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80126Date: Tue Nov 25 01:13:35 2008 -0800
80127
80128 XQuartz: Removed hardcoded org.x.X11 from MachIPC as well
80129 (cherry picked from commit b4add7826d485600a13eba6a9c7be533f2c02d51)
80130
80131commit 065d2afb0ca34f89806e0936c51cd27805bc5123
80132Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80133Date: Tue Nov 25 00:51:01 2008 -0800
80134
80135 XQuartz: Add fallback for xpbproxy's display for Tiger or no-launchd-Leopard
80136 (cherry picked from commit 7a8c6665949d7804a97ef2539a74ec4aa682e1cc)
80137
80138commit 13df49dca28cf680a4d104630cd675de25d3e944
80139Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80140Date: Tue Nov 25 00:39:52 2008 -0800
80141
80142 XQuartz: Use the environment to pass the bundle's prefs domain on to xinit/quartz-wm for Tiger or no-launchd-LEOPARD
80143 (cherry picked from commit fbf4b0d33fa5dc618c3191a4e823232dfa33cd95)
80144
80145commit 40187f782beae4ae751824ef511c9f56a80357c7
80146Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80147Date: Tue Nov 25 00:20:57 2008 -0800
80148
80149 XQuartz: Dead code removal
80150 (cherry picked from commit eeb323612e0adbea37befed31bbaa1d295728385)
80151
80152commit b55cad4569e34e3c10e9a327e20b91ea87d9dd98
80153Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80154Date: Tue Nov 25 00:15:53 2008 -0800
80155
80156 XQuartz: Don't hardcode values of org.x.X11 for the preferences domain
80157 (cherry picked from commit 3a500d9247cf34686ec17b4a88c34d51ecd38ecd)
80158
80159commit d508a3dcca2f160021aced872715e1ded23cef97
80160Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80161Date: Mon Nov 24 23:33:54 2008 -0800
80162
80163 XQuartz: More dead code removal
80164 (cherry picked from commit dcb0f6a2e62823a671051874d14a33ce59505892)
80165
80166commit 41fbdf72f2154a3fca8cf484a611501e3c174fbe
80167Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80168Date: Mon Nov 24 12:33:20 2008 -0800
80169
80170 XQuartz: Disable some error spew on Tiger (where it wouldn't be an error)
80171 (cherry picked from commit 73ec6d3dfe0086d352f4eca25f1df5ae1884bb18)
80172
80173commit 09c3f6e04c273ffafcb547c252137fb17c8ce016
80174Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80175Date: Sat Nov 22 20:23:46 2008 -0800
80176
80177 XQuartz: Dead code removal
80178 (cherry picked from commit 46c077d9b4a883fc809c32077ce40f33a70d268b)
80179
80180commit ab12c7516207908f3e063a78904d68e2db14208e
80181Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80182Date: Sat Nov 22 15:47:14 2008 -0800
80183
80184 XQuartz: More Tiger cleanup: bootstrap_strerror
80185 (cherry picked from commit 37f535aff3e9a7a02711daa98152cdff97745622)
80186
80187commit 208f091bfc657e9ee57b988f035d3aac7e9e173a
80188Author: Adam Jackson <ajax@redhat.com>
80189Date: Sun Oct 12 00:07:56 2008 -0400
80190
80191 Remove xf86IODelay
80192
80193commit 47478aa10adb20b6e48335ac8c4ef31f3fdfe68c
80194Author: Adam Jackson <ajax@redhat.com>
80195Date: Sat Oct 11 23:59:24 2008 -0400
80196
80197 Remove xf86{En,Dis}ableInterrupts entirely
80198
80199commit c7680befe5aebd0f4277d11ff3984d8a7deb9d5b
80200Author: Adam Jackson <ajax@redhat.com>
80201Date: Sat Oct 11 23:35:24 2008 -0400
80202
80203 Remove the remnants of Jensen support
80204
80205 As being an EISA-only machine, and as ISA support is gone now...
80206
80207commit df14682a31b92751091571ed82f6095f55f19cca
80208Author: Adam Jackson <ajax@redhat.com>
80209Date: Sat Oct 11 22:48:51 2008 -0400
80210
80211 Bus: Remove ISA support.
80212
80213 No, really. PCI is old enough to drive now. If you want this, get the kernel
80214 to expose a framebuffer device.
80215
80216commit a8bcab2d3b224e4d4d5b6a097ea530beee920213
80217Author: Adam Jackson <ajax@redhat.com>
80218Date: Sat Oct 11 22:14:23 2008 -0400
80219
80220 Bus: Remove yet more unused overlap processing.
80221
80222commit 994b7c034fc20d76651cf7f6a285526d9aff8770
80223Author: Adam Jackson <ajax@redhat.com>
80224Date: Sat Oct 11 22:11:12 2008 -0400
80225
80226 Bus: Don't pretend to care about IRQs, DMA, or PCI config space
80227
80228commit eb5ae45127fa9f08f0badec7e21f8c26c9c7c969
80229Author: Adam Jackson <ajax@redhat.com>
80230Date: Sat Oct 11 21:44:16 2008 -0400
80231
80232 Bus: Simplify a failure case (that pretty much never happens)
80233
80234commit ee8b5cf94c2a9f8526a3bb5011ebb10f3246a4d9
80235Author: Adam Jackson <ajax@redhat.com>
80236Date: Sat Oct 11 21:41:47 2008 -0400
80237
80238 Bus: Trust the kernel when registering driver resources
80239
80240 ... everywhere, not just (linux && (ia64 || alpha)).
80241
80242commit 8397df89456558e3c85b05e0acfccb9f6af6b695
80243Author: Adam Jackson <ajax@redhat.com>
80244Date: Sat Oct 11 21:36:14 2008 -0400
80245
80246 Remove unused MIN macro
80247
80248commit c251c0baae59714a6ac83b69cd106c08baa3613e
80249Author: Adam Jackson <ajax@redhat.com>
80250Date: Sat Oct 11 21:34:27 2008 -0400
80251
80252 Bus: remove special handling for init-only resources
80253
80254 This isn't used by any driver, nor has it ever been as far as I can
80255 tell.
80256
80257commit 3e5281af17841cf50d0e52a728b12c6ab56e61df
80258Author: Adam Jackson <ajax@redhat.com>
80259Date: Sat Oct 11 21:16:45 2008 -0400
80260
80261 PCI: Unexport xf86scanpci
80262
80263commit 6b198daa46f2f609aff7900761cf82cc2fb4e0b4
80264Author: Adam Jackson <ajax@redhat.com>
80265Date: Sat Oct 11 20:51:39 2008 -0400
80266
80267 Bus: remove useless isaConvertRange2Host
80268
80269commit a96db74c2a95bb1dce132cf47ea720ae939dfad7
80270Author: Adam Jackson <ajax@redhat.com>
80271Date: Thu Oct 9 00:43:26 2008 -0400
80272
80273 Bus: Don't try to find an ISA bus just for fun.
80274
80275commit b21311a99d58997cd1fc68726d0848242e9c34fc
80276Author: Adam Jackson <ajax@redhat.com>
80277Date: Thu Oct 9 00:34:42 2008 -0400
80278
80279 Bus: Remove unused RemoveOverlaps
80280
80281commit 41be6b3f0dc0baa1c6ae8d2b41a6be73ca0e7268
80282Author: Adam Jackson <ajax@redhat.com>
80283Date: Thu Oct 9 00:33:28 2008 -0400
80284
80285 Bus: Remove the notion of estimated resources.
80286
80287commit 095ba1435501776c8c8a34e767b89f89e5dc949a
80288Author: Adam Jackson <ajax@redhat.com>
80289Date: Thu Oct 9 00:27:33 2008 -0400
80290
80291 Bus: remove the "reducer"
80292
80293 This code effectively didn't do anything anymore.
80294
80295commit 4457e31710af90f9ac295bb686c841e9473fb767
80296Author: Adam Jackson <ajax@redhat.com>
80297Date: Thu Oct 9 00:14:54 2008 -0400
80298
80299 PCI: Remove unused ia64 platform code.
80300
80301commit 86cfe0ee236bfd3613e5f9ba589211db42d009eb
80302Author: Adam Jackson <ajax@redhat.com>
80303Date: Wed Oct 8 23:45:40 2008 -0400
80304
80305 PCI: Simplify OS PCI function registration a bit.
80306
80307commit 2d427b9cb1594f8f2f66b463033fff5b459962fd
80308Author: Adam Jackson <ajax@redhat.com>
80309Date: Wed Oct 8 23:38:23 2008 -0400
80310
80311 PCI: Remove config mechanism details.
80312
80313 pciaccess handles this for us now, no need to remember PC arcana.
80314
80315commit 5bb86bafd6fda296011cbcd5d15a85a6d770ae29
80316Author: Adam Jackson <ajax@redhat.com>
80317Date: Wed Oct 8 23:34:41 2008 -0400
80318
80319 PCI: Remove non-pciaccess path for x86.
80320
80321commit a67360e79fa7e17c3d907771694009c57c1cd195
80322Author: Adam Jackson <ajax@redhat.com>
80323Date: Wed Oct 8 23:31:38 2008 -0400
80324
80325 PCI: Always build domain support.
80326
80327commit 51e105ccc3d0ac8c0fe74efd029ffbddb80b140e
80328Author: Adam Jackson <ajax@redhat.com>
80329Date: Wed Oct 8 23:12:31 2008 -0400
80330
80331 Remove xf86GetResourcesImplicitly
80332
80333commit 599a0f3f1e3ae92676e3648471576c0001cfd9ae
80334Author: Alan Coopersmith <alan.coopersmith@sun.com>
80335Date: Mon Nov 24 20:37:58 2008 -0800
80336
80337 Fix typo in 5bf2c88d2317230b95b2904cb975167d03ee13a2
80338
80339 Amazing how these things hide until you see the diff come back from
80340 the commit list.
80341
80342commit 5bf2c88d2317230b95b2904cb975167d03ee13a2
80343Author: Alan Coopersmith <alan.coopersmith@sun.com>
80344Date: Mon Nov 24 20:34:46 2008 -0800
80345
80346 Simplify filename generation code for Xorg -configure
80347
80348commit 75784e1e53ad78e21518696dd9d297bc08c17d54
80349Author: Alan Coopersmith <alan.coopersmith@sun.com>
80350Date: Mon Nov 24 20:32:20 2008 -0800
80351
80352 Solaris: Make KDSETMODE failure non fatal, and retry it on interrupts
80353
80354commit 3e6da1636093d7dc98baac40544c0b0fb7fd8aec
80355Author: Peter Hutterer <peter.hutterer@redhat.com>
80356Date: Fri Nov 14 15:55:57 2008 +1000
80357
80358 include: remove ENTER_LEAVE_SEMAPHORE macros.
80359
80360commit b292a7a2d7e259177e1cc37346c2bee27a018630
80361Author: Peter Hutterer <peter.hutterer@redhat.com>
80362Date: Fri Nov 14 15:44:29 2008 +1000
80363
80364 dix: updated enter/leave core event model.
80365
80366 As proposed by Owen Taylor [1], the enter-leave event model needs to adjust
80367 the events sent to each window depending on the presence of pointers in a
80368 window, or in a subwindow.
80369
80370 The new model can be summarised as:
80371 - if the pointer moves into or out of a window that has a pointer in a child
80372 window, the events are modified to appear as if the pointer was moved out of
80373 or into this child window.
80374 - if the pointer moves into or out of a window that has a pointer in a parent
80375 window, the events are modified to appear as if the pointer was moved out of
80376 or into this parent window.
80377
80378 Note that this model requires CoreEnterLeaveEvent and DeviceEnterLeaveEvent to
80379 be split and treated separately.
80380
80381 [1] http://lists.freedesktop.org/archives/xorg/2008-August/037606.html
80382
80383 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80384
80385commit 724f83b87bb16472d4c328e35d2a477384b29f84
80386Author: Peter Hutterer <peter.hutterer@redhat.com>
80387Date: Fri Nov 14 15:29:01 2008 +1000
80388
80389 dix: add FirstPointerChild, FirstPointerAncestor auxiliary functions.
80390
80391 FirstPointerChild: Return the first child that has a pointer within its
80392 boundaries.
80393
80394 FirstPointerAncestor: return the first ancestor with a child within its
80395 boundaries.
80396
80397 These are required for the updated enter/leave model.
80398
80399 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80400
80401commit 7d3e595f93dcd3d334e766a9dea602c05affdbaf
80402Author: Peter Hutterer <peter.hutterer@redhat.com>
80403Date: Fri Nov 14 15:27:19 2008 +1000
80404
80405 dix: Add EnterWindow, LeaveWindow, HasPointer auxiliary functions.
80406
80407 These replace the ENTER_LEAVE_SEMAPHORE_* macros. Unused currently.
80408
80409 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80410
80411commit 6bdc963cdabb4a2e77de7f00a1d062aa2b873f9b
80412Author: Peter Hutterer <peter.hutterer@redhat.com>
80413Date: Fri Nov 14 15:37:35 2008 +1000
80414
80415 dix: split enter/leave event handling into core and device handling.
80416
80417 Device events always need to be delivered, core events only in some cases.
80418 Let's keep them completely separate so we can adjust core event delivery.
80419
80420 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80421
80422commit 5e48f5e2dd2dec7cfd1fa40b61e25123dfca515e
80423Author: Peter Hutterer <peter.hutterer@redhat.com>
80424Date: Fri Nov 14 15:41:59 2008 +1000
80425
80426 dix: remove unused EnterLeaveSemaphoresIsset.
80427
80428 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80429
80430commit 868fd503b7aeab31dba72046b59061008d8b7501
80431Author: Peter Hutterer <peter.hutterer@redhat.com>
80432Date: Fri Nov 14 15:18:56 2008 +1000
80433
80434 dix: move Enter-Leave related functions into new enterleave.c
80435
80436 Preparation for the new core enter/leave model.
80437
80438 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80439
80440commit c4b9ab6bf56139fdd8c7c584a6f523c6766cddd6
80441Author: Keith Packard <keithp@keithp.com>
80442Date: Tue Apr 29 08:29:42 2008 -0700
80443
80444 Handle transform failure when computing shadow damage area.
80445
80446 PictureTransformBounds can fail, when this happens, damage the entire screen
80447 so that the shadow gets repainted correctly.
80448
80449commit 9c7679240ad90367693a462e288308b3fdc08f26
80450Author: Keith Packard <keithp@keithp.com>
80451Date: Sun Mar 30 19:05:07 2008 -0700
80452
80453 [randr] don't try to compute crtc transform when no mode is set.
80454
80455 Dereferencing the NULL mode pointer would cause a crash. As these transform
80456 matrices won't be used while the CRTC is disabled, just leave their values
80457 alone.
80458
80459commit 315b6d0a425d0257f226ee7032aca3ca31665e59
80460Author: Keith Packard <keithp@keithp.com>
80461Date: Thu Mar 27 11:46:39 2008 -0700
80462
80463 Update RandR global transform when driver notifies of transform change.
80464
80465 Need to compute and save the global transform when the driver changes it.
80466
80467commit 5d9282fde919c936816a85c2a9c5734dc9d57cc2
80468Author: Keith Packard <keithp@keithp.com>
80469Date: Thu Mar 27 11:45:53 2008 -0700
80470
80471 Export rrtransform.h as needed by randrstr.h
80472
80473 When the transform management was moved from randrstr.h, the associated
80474 header file became necessary to build drivers. Include it as a part of the
80475 sdk headers.
80476
80477commit 7236fd9dd4e9f5c4cb5bafa5c2a18cff12a0444c
80478Author: Keith Packard <keithp@keithp.com>
80479Date: Fri Mar 21 03:16:09 2008 -0700
80480
80481 Be careful about copying transforms around; they have allocated memory.
80482
80483commit 7e69e364d2ef146d8ec3651d04bdd6d641017449
80484Author: Keith Packard <keithp@keithp.com>
80485Date: Fri Mar 21 03:15:00 2008 -0700
80486
80487 Fix rotated/reflected cursor positions.
80488
80489 Doing projective transforms required repositioning the cursor using the
80490 hotspot, but that requires relocating the upper left corner in terms of said
80491 hotspot.
80492
80493commit 93179c214fc6ed88f72955a11c69ae0a47316d8c
80494Author: Keith Packard <keithp@keithp.com>
80495Date: Fri Mar 21 03:14:47 2008 -0700
80496
80497 rrtransform needs randrstr to get RANDR_INTERFACE defines
80498
80499commit 7c61db66a470a8306e346ed8bf8934f014dada42
80500Author: Keith Packard <keithp@keithp.com>
80501Date: Fri Mar 21 02:39:49 2008 -0700
80502
80503 Create rrtransform.[ch]. Add RRTransform argument to RRCrtcNotify.
80504
80505 Instead of using a separate function to notify DIX about transform changes,
80506 add the transform to RRCrtcNotify so that the whole Crtc state changes
80507 atomically.
80508
80509commit fa6a1df209bd74da1d545982cca437afc2198cc1
80510Author: Keith Packard <keithp@keithp.com>
80511Date: Fri Mar 21 02:35:28 2008 -0700
80512
80513 Avoid overflow in PictureTransformPoint. Fix PictureTransformIsIdentity.
80514
80515 PictureTransformPoint computes homogeneous coordinates internally, but fails
80516 to handle intermediate values larger than 16.16. Use 64 bit intermediate
80517 values while computing the final result at 16.16 and only complain if that
80518 result is too large.
80519
80520 PictureTransformIsIdentity was completely wrong -- it was not checking for
80521 identity transforms at all.
80522
80523commit 6fe9c15731be5f5afabacb3aa3ed71b840f4238b
80524Author: Keith Packard <keithp@keithp.com>
80525Date: Wed Mar 19 12:15:39 2008 -0700
80526
80527 Allow drivers to set crtc transforms.
80528
80529 Track curent transform down in the mode setting code so that it may be set
80530 separately from RandR.
80531
80532commit 49db14e4ac26070ed86088419483888dda18b603
80533Author: Keith Packard <keithp@keithp.com>
80534Date: Wed Mar 19 00:46:35 2008 -0700
80535
80536 Handle RandR transform matrices in floating point.
80537
80538 RandR matrix computations lose too much precision in fixed point;
80539 computations using the inverted matrix can be as much as 10 pixels off.
80540 Convert them to double precision values and pass those around. These API
80541 changes are fairly heavyweight; the official Render interface remains fixed
80542 point, so the fixed point matrix comes along for the ride everywhere.
80543
80544commit 6f734aecaec4f5c6152c3ebca197ef65bb4523da
80545Author: Keith Packard <keithp@keithp.com>
80546Date: Tue Mar 18 16:06:06 2008 -0700
80547
80548 Eliminate inverse matrix from randr transform protocol
80549
80550 It is easier, and potentially more precise, to compute the inverse in the
80551 server where everything can eventually be kept in floating point form.
80552
80553commit 197aa784694992f6bcfd194932309e440334c237
80554Author: Keith Packard <keithp@keithp.com>
80555Date: Tue Mar 18 15:35:44 2008 -0700
80556
80557 Report whether transforms are support from driver through extension to client
80558
80559 Add APIs to xf86RandR12 support and randr extension to record whether the
80560 driver supports transforms, report that value in the RRGetCrtcTransform
80561 reply.
80562
80563commit e86c34663ef97e946a1129450105efa89a123af6
80564Author: Keith Packard <keithp@keithp.com>
80565Date: Tue Mar 18 15:25:31 2008 -0700
80566
80567 Clear shadow pixmaps before using them.
80568
80569 This eliminates some ugly flashing, as well as clearing the borders when the
80570 shadow will not be completely painted.
80571
80572commit 97ab0c6eff870b52c0383b63a78cec49059b2545
80573Author: Keith Packard <keithp@keithp.com>
80574Date: Tue Mar 18 15:15:40 2008 -0700
80575
80576 When converting from double to fixed, round carefully.
80577
80578 This reduces the matrix representation error after inverting a
80579 transformation matrix (although it doesn't eliminate it entirely).
80580
80581 Perhaps we should extend Render to include 64-bit floating point transforms...
80582
80583commit 6d3a9e40a4b9ec455af11cce31e4aa616c93db32
80584Author: Keith Packard <keithp@keithp.com>
80585Date: Mon Mar 17 23:04:49 2008 -0700
80586
80587 Compute matrix inversion instead of using wire version in RRCrtcTransformSet
80588
80589 It doesn't make sense to have the client invert this matrix when the server
80590 can do so reasonably efficiently. This avoids weird fixed point rounding
80591 errors when testing the transform against its inverse. Now to fix the
80592 protocol.
80593
80594commit 160252d94f04acc95f0a4e0f884ff565a5aa0744
80595Author: Keith Packard <keithp@keithp.com>
80596Date: Mon Mar 17 23:03:56 2008 -0700
80597
80598 Add matrix inversion function (uses doubles)
80599
80600 The obvious matrix inversion function, coded using doubles to avoid fiddling
80601 with fixed point precision adventures.
80602
80603commit 8fd82c88e3f2060fda4ba30376900ece77668c63
80604Author: Keith Packard <keithp@keithp.com>
80605Date: Mon Mar 17 16:14:43 2008 -0700
80606
80607 Adjust transformed cursor position to account for hotspot
80608
80609commit eb222e64128034df8361d5a82d4f4aa1318923ce
80610Author: Keith Packard <keithp@keithp.com>
80611Date: Mon Mar 17 16:14:15 2008 -0700
80612
80613 Actually use filter kernel size to expand transform redisplay box
80614
80615commit 3fdb963f6e8287edeb4c5bc7bbadbc02eb8bb910
80616Author: Keith Packard <keithp@keithp.com>
80617Date: Mon Mar 17 16:13:25 2008 -0700
80618
80619 Correct bilinear filter kernel size (should be 2x2)
80620
80621commit ee6326a6b89b2d223f6e1eaf02aac1fe8aebfaf6
80622Author: Keith Packard <keithp@keithp.com>
80623Date: Mon Mar 17 15:22:06 2008 -0700
80624
80625 Handle filter widths in xf86Rotate
80626
80627commit ddc8466137d229fdc100cc403f492c7d61a1ba89
80628Author: Keith Packard <keithp@keithp.com>
80629Date: Mon Mar 17 15:21:26 2008 -0700
80630
80631 Initialize and cleanup new filter fields in xf86Crtc.
80632
80633commit 16c093afd4f6b30d889cacdc994b0024f9bd83d5
80634Author: Keith Packard <keithp@keithp.com>
80635Date: Mon Mar 17 15:20:52 2008 -0700
80636
80637 Pass filter kernel size through transforms
80638
80639commit 1df02d7ddd44f84bcaefd6583af1a9141cb3c78b
80640Author: Keith Packard <keithp@keithp.com>
80641Date: Mon Mar 17 15:19:17 2008 -0700
80642
80643 Add kernel size to Render filters.
80644
80645 This width/height value lets filter users know how far the filter spreads
80646 into the source image.
80647
80648commit 40f3dff6b350fe0be55ebf7dbca88a0ef8f6380f
80649Author: Keith Packard <keithp@keithp.com>
80650Date: Mon Mar 17 13:57:47 2008 -0700
80651
80652 Use transform when computing scanout size of modes
80653
80654 Report transformed crtc sizes through RandR and Xinerama. Test screen size
80655 against transformed mode sizes when configuring the Crtc.
80656
80657commit f50349e1930e620cacdf27a6f8585fcb9cb5199a
80658Author: Keith Packard <keithp@keithp.com>
80659Date: Sat Mar 15 00:36:45 2008 -0700
80660
80661 [RANDR] Support filters in CRTC transforms.
80662
80663 Create new RRTransform datatype to hold all of the transform related
80664 information, use that in lots of places to pass filters around.
80665
80666commit acda790e430b2a18c7c35379f6e538f3d01ff221
80667Author: Keith Packard <keithp@keithp.com>
80668Date: Fri Mar 14 13:46:30 2008 -0700
80669
80670 [render] Split out filter finding from filter setting.
80671
80672 To prepare for RandR using filters in transforms, split out
80673 code paths so that the RandR code can validate the filter name and
80674 parameters during the transform set operation so that use of the filter
80675 later will not have unreportable errors.
80676
80677commit e3d6f279d5c305dfb81ca109fbfb665870712d2f
80678Author: Keith Packard <keithp@keithp.com>
80679Date: Thu Mar 13 21:31:12 2008 -0700
80680
80681 Wire up RandR CRTC transform protocol, bump server to RandR 1.3
80682
80683 This involved removing a pile of matrix code from the DDX,
80684 as well as moving a bit of transform logic from DDX to DIX.
80685
80686commit ff9d1cd843a9b0aba69a3d788b21d5f6af702590
80687Author: Keith Packard <keithp@keithp.com>
80688Date: Thu Mar 13 21:30:18 2008 -0700
80689
80690 Add funcs to convert between protocol and pixman matrices
80691
80692commit c2f254da694731ea573aa8bbc2707c083743f2d4
80693Author: Keith Packard <keithp@keithp.com>
80694Date: Thu Mar 13 21:29:19 2008 -0700
80695
80696 Randr now depends on Render for matrices
80697
80698commit f547650328287545a7a4d96df8d6a6c606dd95a9
80699Author: Keith Packard <keithp@keithp.com>
80700Date: Thu Mar 13 14:50:13 2008 -0700
80701
80702 Export a bunch of matrix operations from render.
80703
80704 The render extension uses many matrix operations internally, this change
80705 exposes those functions to other parts of the server, drivers and
80706 extensions. The change is motivated by the 'transform' additions to the
80707 RandR extension but will likely be useful elsewhere.
80708
80709commit e063162e80e51ed4368874b3af7ba690ea280d9e
80710Author: Keith Packard <keithp@keithp.com>
80711Date: Thu Mar 13 14:26:01 2008 -0700
80712
80713 Add projective transforms to RandR DIX/DDX API.
80714
80715 New RRCrtcGetTransform function in DIX that DDX can use to get the pending
80716 transform. The DDX code should be complete; the DIX code is just a stub at
80717 this point.
80718
80719commit 9c7ac47b871a71e42d2f6933749ca462f1a65b40
80720Author: Keith Packard <keithp@keithp.com>
80721Date: Tue Nov 4 00:10:08 2008 -0800
80722
80723 Notify DRI when crtc regions change
80724
80725 Drivers that care about crtc positions on the screen to ensure that vblank
80726 works correctly need to be notified when crtcs are changed.
80727
80728 Provide a hook in the mode setting code that is invoked whenever any
80729 configuration is done to the screen.
80730
80731 Use this new hook in the DRI code so that DRI clients are notified and
80732 receive updated information.
80733
80734 Signed-off-by: Keith Packard <keithp@keithp.com>
80735
80736commit 554592cd70543f87cd8bee5ff47cc8281511e041
80737Author: Peter Hutterer <peter.hutterer@redhat.com>
80738Date: Mon Nov 24 11:50:38 2008 +1000
80739
80740 DGA: silence compiler warning.
80741
80742 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80743
80744commit 0dbcbd35f63db14734199a0beb2a91b6937f3838
80745Author: Peter Hutterer <peter.hutterer@redhat.com>
80746Date: Fri Nov 21 12:02:02 2008 +1000
80747
80748 xfree86: Split the working code of NIDR into new xf86NewInputDevice.
80749
80750 The xfree86 server previously hat NewInputDeviceRequest and InitInput, and
80751 both basically did the same thing. Reduce NIDR to parameter checking and use
80752 xf86NewInputDevice from both InitInput and NIDR to actually create the device.
80753
80754 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80755 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
80756 Signed-off-by: Adam Jackson <ajax@redhat.com>
80757
80758commit e078901a4eca02bd3e7a80d9462dafbca939a187
80759Author: Peter Hutterer <peter.hutterer@redhat.com>
80760Date: Fri Nov 21 12:39:55 2008 +1000
80761
80762 dix: Enable core devices in InitCoreDevices already.
80763
80764 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80765 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
80766 Signed-off-by: Adam Jackson <ajax@redhat.com>
80767
80768commit d939f2482e71ad20bac28b7aa4b2e8c8196e9d65
80769Author: Peter Hutterer <peter.hutterer@redhat.com>
80770Date: Fri Nov 21 12:21:45 2008 +1000
80771
80772 dix: fix false comment.
80773
80774 VCP and VCK are always present, this comment is a leftover from earlier MPX
80775 days.
80776
80777 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80778
80779commit 387563b77743d92be83420d982eaf57f5459a883
80780Merge: 85f650c f6e01fa
80781Author: Matthieu Herrb <matthieu.herrb@laas.fr>
80782Date: Sun Nov 23 13:15:46 2008 +0100
80783
80784 Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/xserver
80785
80786commit 85f650c9b76ab2e244f01908a7941c320635becd
80787Author: Matthieu Herrb <matthieu.herrb@laas.fr>
80788Date: Sun Nov 23 13:14:27 2008 +0100
80789
80790 missing action in if statement.
80791
80792commit f6e01fa1b87ea190ea5ad723ce46893784ea1de4
80793Author: Stefan Dirsch <sndirsch@suse.de>
80794Date: Sun Nov 23 11:16:03 2008 +0100
80795
80796 Added '-showopts' option to print available driver options (#5564).
80797
80798commit 8964b8d0ec2b7b3b6bf540cd647b14a20e8f64a5
80799Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80800Date: Sat Nov 22 14:23:23 2008 -0800
80801
80802 XQuartz: More 1.6 server API updates
80803
80804commit ad0f232165fe1a25ca4fb6da817da02b6ce31779
80805Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80806Date: Sat Nov 22 14:04:28 2008 -0800
80807
80808 XQuartz: Fixed --disable-glx
80809
80810commit 8e2287c220694953e972cd5119c2b0cd256b7a30
80811Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80812Date: Sat Nov 22 13:57:45 2008 -0800
80813
80814 XQuartz: Updated some code to use newer server API
80815
80816commit 6eb33bc0cb2e62339d323e1f1894015d7e3142f3
80817Author: George Staplin <gstaplin@apple.com>
80818Date: Sat Nov 22 11:37:08 2008 -0700
80819
80820 XQuartz: GL: Remove the inclusion of glcontextmodes.h.
80821
80822 Add some commentary about future directions needed for the GLX drawable
80823 creation and destruction code.
80824
80825 Match xalloc with xfree.
80826
80827 I made some minor formatting improvements.
80828 (cherry picked from commit b772d64fce31d16b498c621096e39d5203994d6e)
80829
80830commit 429b4b20d5708d608fd55f91dd5bcd4ac0b51a12
80831Author: George Staplin <gstaplin@apple.com>
80832Date: Sat Nov 22 10:57:58 2008 -0700
80833
80834 XQuartz: GL: Add a branch to prevent a NULL DrawablePtr structure access.
80835
80836 In attach() check for pDraw being NULL, and also print an ErrorF message,
80837 because we eventually want to track down why this is occuring.
80838
80839 It's unclear how this occurs, but as I noted in the 1.4 branch, I believe that
80840 the DrawablePtr/struct _Drawable -> id is the member being accessed that causes
80841 KERN_PROTECTION_FAILURE at 0x0000000000000004
80842
80843 This passes my tests using: env LIBGL_ALWAYS_INDIRECT=1 ./sometest.
80844
80845 I fixed a warning: caused by initializing the screen->base.visuals with the
80846 configs. It is a ** not a *. It seems that some other part of GLX will
80847 initialize this for us.
80848 (cherry picked from commit 17f6a261fca6d5856069dce28bb4838261afc6bc)
80849
80850commit bc0c7075e29b1719409cf7de8a4ab9ae1315aa67
80851Author: Matthieu Herrb <matthieu.herrb@laas.fr>
80852Date: Sat Nov 22 19:10:09 2008 +0100
80853
80854 Remove the old libusbhid copy from os-support/bsd.
80855
80856 This has been replaced by in-kernel support for USB mices several
80857 years ago and hasn't been enabled in build since X.Org 6.8.
80858
80859commit 76fcfc480133726112049e5bd9c3082d46825918
80860Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80861Date: Fri Nov 21 20:25:55 2008 -0800
80862
80863 XQuartz: GLX: sync up indirect.c to match George's updates in the xorg-server-1.5-apple branch
80864
80865commit 1a717779b67b9c1cdf734eef64135ac38829f726
80866Author: George Staplin <gstaplin@apple.com>
80867Date: Fri Nov 21 17:28:11 2008 -0700
80868
80869 XQuartz: GL: remove glcontextmodes.* from the Makefile.am. It has been replaced.
80870
80871 The new replacement is __GLXconfig.
80872 (cherry picked from commit 3bed9b65c807a1aeb662b8042826cbb54280181d)
80873
80874commit dacad9ae0bdaedff9403fbe84a5c5a4b4eef3542
80875Author: Alan Coopersmith <alan.coopersmith@sun.com>
80876Date: Fri Nov 21 17:16:22 2008 -0800
80877
80878 Xephyr man page typo fixes.
80879
80880 Also trailing white space removal to clear git warnings.
80881
80882commit ca64b5949d229a60a3c62b6d29d51873b7694bd6
80883Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80884Date: Fri Nov 21 11:20:31 2008 -0800
80885
80886 XQuartz: Don't use LS to find X11.app on Tiger.
80887 (cherry picked from commit e62107e55261ef252a2a24dd26a60e5dd295d560)
80888
80889commit 701f8e3a94d2b659ebcfffd09acd1d66f9e797b8
80890Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80891Date: Fri Nov 21 11:08:00 2008 -0800
80892
80893 XQuartz: Update applewm deps
80894 (cherry picked from commit 0ad91c59be8759a9c9e9d4f639056d8c689a3bc5)
80895
80896commit 393325f908f8bd6648e466183f3683a2d80b264e
80897Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80898Date: Fri Nov 21 10:57:31 2008 -0800
80899
80900 XQuartz: spewCallStack is noop on Tiger now...
80901 (cherry picked from commit 56c469a68b92c6cf003802f418ea00bd596a41c3)
80902
80903commit 5d47a5d6526b7e52387647a15a580dfcafd1733f
80904Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80905Date: Fri Nov 21 10:54:55 2008 -0800
80906
80907 XQuartz: pbproxy: Added some typedefs for types not available on Tiger
80908 (cherry picked from commit 0947aa7911f1de44bfe16e505a757b659c5ab2a8)
80909
80910commit b262788401715787c68cea4ca79f4b270307afee
80911Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80912Date: Fri Nov 21 10:53:39 2008 -0800
80913
80914 XQuartz: More Tiger ifdefs
80915 (cherry picked from commit 803509072f2c039e5cd555c4ac14d672f0e698c5)
80916
80917commit 510744444329170f702d31d43997a6bbc1591b44
80918Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80919Date: Fri Nov 21 10:44:11 2008 -0800
80920
80921 XQuartz: Resurrect the old audio path for Tiger
80922 (cherry picked from commit a61a8d9047e8765faf3892f4f2148c8553fd192c)
80923
80924commit d7ee76f9dd84da05b59591a971b96bf990136767
80925Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
80926Date: Wed Nov 19 11:42:59 2008 -0800
80927
80928 XQuartz: pbproxy: Push the pbproxy Xevent processing into its own thread
80929 and just have the AppKit thread wake it up.
80930 (cherry picked from commit 799715b8f3327c8da59ab45706e85af2d2c438e4)
80931
80932commit a9e20306fbe3262602f21b876a52a1ef38cdf20a
80933Author: Egbert Eich <eich@ovid.suse.de>
80934Date: Fri Nov 21 18:50:01 2008 +0100
80935
80936 int10: Do an mprotect(..,PROT_EXEC) on shmat()ed memory ranges.
80937
80938 When the linux kernel sets the NX bit vm86 segfaults when it tries to execute
80939 code in memory that is not marked EXEC. Such code gets called whenever
80940 we return from a VBIOS call to signal the calling program that the call
80941 is actually finished and that we are not trapping for other reasons (like
80942 IO accesses).
80943 Use mprotect(2) to set these memory ranges PROT_EXEC.
80944
80945commit a54153e669fd293a47f0077bf25505dd545ddce2
80946Author: Rémi Cardona <remi@gentoo.org>
80947Date: Fri Nov 21 16:39:00 2008 +0100
80948
80949 xfree86: don't reset Auto(Add|Enable)Devices, use defaults from xf86Globals
80950
80951 Without this, commit ace38fafb062372dcd3d56378b5b8f86525c6241 is useless
80952 when HAL support is disabled.
80953
80954 Signed-off-by: Julien Cristau <jcristau@debian.org>
80955
80956commit 6a1b2fed7e465a9c652e8ee250b8eab440945862
80957Author: Matthieu Herrb <matthieu.herrb@laas.fr>
80958Date: Sun Sep 28 12:18:16 2008 +0200
80959
80960 Add a man page for Xephyr, with information from the README file.
80961
80962commit 10d472bf9d61bef32d0a383b2e3783e7a1d7621f
80963Author: Matthieu Herrb <matthieu.herrb@laas.fr>
80964Date: Mon May 19 05:15:30 2008 +0200
80965
80966 Support for BSD apm
80967 XXX check amd64 pci init
80968
80969commit f4e725f248870a2de10449cc8f53de32e7840d45
80970Author: Peter Hutterer <peter.hutterer@redhat.com>
80971Date: Thu Nov 20 13:51:15 2008 +1000
80972
80973 dix: memset DeviceVelocityPtr to zero.
80974
80975 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80976 Signed-off-by: Adam Jackson <ajax@redhat.com>
80977
80978commit 7f818776b7a19bd6100596d327ecaa69be317bc6
80979Author: Simon Thum <simon.thum@gmx.de>
80980Date: Wed Nov 19 15:05:50 2008 +1000
80981
80982 dix: fix typos in comments, one formatting fix.
80983
80984 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
80985
80986commit 037ec5a3c496b07f885a128828ef804b3aa8eee2
80987Author: Peter Hutterer <peter.hutterer@redhat.com>
80988Date: Wed Nov 19 15:12:04 2008 +1000
80989
80990 gitignore: ignore .patch files
80991
80992commit ace38fafb062372dcd3d56378b5b8f86525c6241
80993Author: Peter Hutterer <peter.hutterer@redhat.com>
80994Date: Tue Nov 18 08:19:45 2008 +1000
80995
80996 xfree86: without CONFIG_HAL, Auto{Add|Enable}Devices and AEI is false.
80997
80998 There's little chance that we'll get the input devices at runtime without HAL,
80999 we might as well force the server to add mouse/kbd devices automatically -
81000 just like in the olden days.
81001
81002 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81003
81004commit 1cd894173ee7fa3d1a78d330c12453ea53e7b908
81005Author: Alan Coopersmith <alan.coopersmith@sun.com>
81006Date: Wed Nov 19 13:44:26 2008 -0800
81007
81008 Always use server-<display>.xkm to avoid races when multiple servers start
81009
81010 Previously each server starting ran xkbcomp with the output set to
81011 <keymapname>.xkm, read it, then deleted it - which led to races if
81012 two servers were starting at the same time with the same keymap.
81013
81014 Sun bug #6773816 Xorg uses the same xkm output file for compiled keymap file
81015 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6773816>
81016
81017commit 1e4412abc499b9938bbac8a3ecd82f7f12d448d9
81018Author: Rémi Cardona <remi.cardona@lri.fr>
81019Date: Mon Nov 17 09:56:49 2008 +0100
81020
81021 drop unused HAVE_LIBDRM_2_2
81022
81023 Signed-off-by: Julien Cristau <jcristau@debian.org>
81024
81025commit c4ec71b405fcb5132d5089e3f8dd7b00c235265a
81026Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81027Date: Sun Nov 16 17:57:46 2008 -0800
81028
81029 XQuartz: pbproxy: nuke RCS Id, update header license/copyright
81030 (cherry picked from commit 477c6968a88429bb9c10222a5836bdc936d10ab1)
81031
81032commit 42f330cf9ef44aea7428a67d282dfb1a60196e55
81033Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81034Date: Sun Nov 16 10:42:03 2008 -0800
81035
81036 XQuartz: pbproxy: Remove thread-main.m
81037
81038 it got accidently added and isn't used.
81039 (cherry picked from commit cb6ffce68a2d10022fa92725d2d1837da4b3b89f)
81040
81041commit 19aac4274bbac55ce944acfaf3274a4403fe392e
81042Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81043Date: Wed Nov 19 10:55:29 2008 -0800
81044
81045 mi: Push screen-saver/DPMS handling to after the EQ pop operation.
81046
81047 This way we on't need to hold the mutex during the dixSaveScreens() call.
81048
81049 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
81050 Signed-off-by: Tiago Vignatti <vignatti@c3sl.ufpr.br>
81051 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81052
81053commit 56d1793c435abbaababab031860f1160f09fff06
81054Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81055Date: Wed Nov 19 10:54:57 2008 -0800
81056
81057 mi: Only increment tail (push) when the event data is actually in the queue
81058
81059 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
81060 Signed-off-by: Tiago Vignatti <vignatti@c3sl.ufpr.br>
81061 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81062
81063commit 570835b9a89187a36e08defbccc56051bb349b52
81064Author: Alan Coopersmith <alan.coopersmith@sun.com>
81065Date: Wed Nov 19 10:19:20 2008 -0800
81066
81067 Use bsd_pci stub routines on Solaris too, since we use pciaccess for real work
81068
81069 Depends on commit 70e18a3b6b9d52169bca650f6cd4ef5d8ab40d95 to get
81070 definition of xf86InitVidMem on Solaris.
81071
81072commit 3932a848572f4eaf8b7f1d91d9b74aeafab069a2
81073Author: Keith Packard <keithp@keithp.com>
81074Date: Tue Nov 18 14:21:01 2008 -0500
81075
81076 Xinput: use floats for ConstantDeceleration and AdaptiveDeceleration
81077
81078 These values need not be constrained to integer values.
81079
81080 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81081 Signed-off-by: Adam Jackson <ajax@redhat.com>
81082
81083commit 08cd361234ed0410f67342f46ae01120c4fe3331
81084Author: Dave Airlie <airlied@redhat.com>
81085Date: Mon Nov 17 10:28:48 2008 +1000
81086
81087 exa: avoid doing prepare/done without intervening copies in exaFillRegionTiled
81088
81089 This does a precursor check to make sure the copies are required before
81090 entering the prepare/done code.
81091
81092commit 8f8a9c19ad58768b07461a3f4bccea98f7c4f958
81093Author: Dave Airlie <airlied@redhat.com>
81094Date: Mon Nov 17 10:24:39 2008 +1000
81095
81096 EXA: avoid copy operations if no boxes in use
81097
81098 Simple fix for now, I'm sure damage shouldn't be calling us with nbox = 0.
81099
81100commit 03f7a66a1ae3f522a49bcc783066d53d69c97262
81101Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81102Date: Sat Nov 15 19:45:49 2008 -0800
81103
81104 XQuartz: Don't use TIS on Tiger...
81105 (cherry picked from commit be8ac84c15e7e76c581b9e75cda74139394aa4c7)
81106
81107commit e8112784e8ef80a96ed54eca7e148b424001a38f
81108Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81109Date: Sat Nov 15 19:52:34 2008 -0800
81110
81111 XQuartz: Integrated xpbproxy into the server.
81112
81113 Added a configure option called --enable-standalone-xpbproxy which is useful for deveoping xpbproxy.
81114 The 'active' switch in preferences just disables the in-server xpbproxy (not this standalone).
81115 (cherry picked from commit 42944936326ef8732f622db9f75b79a92980550d)
81116
81117commit 24928b85dc996c8d0a052ddf383df25718ee5305
81118Author: Michel Dänzer <michel@tungstengraphics.com>
81119Date: Sat Nov 15 13:28:53 2008 +0100
81120
81121 DRI2: Don't crash in DRI2ClipNotify when DRI2DrawablePtr is NULL.
81122
81123commit e5271c374715dd4e9a724ab3e5f77108d4ce8c23
81124Author: Michel Dänzer <michel@tungstengraphics.com>
81125Date: Sat Nov 15 13:27:42 2008 +0100
81126
81127 Set libdrm flags correctly even when only DRI2 is enabled.
81128
81129commit bc4f804cd86497655f6765b0954e64dc959939e3
81130Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81131Date: Fri Nov 14 15:13:22 2008 -0800
81132
81133 XQuartz: Actually add glcontextmodes.[hc]
81134 (cherry picked from commit 763195d338345e3637d112279bdbe002b6e7791a)
81135
81136commit aedafbea0e1c7bc780ebe7ae878f62525769aca8
81137Author: Ben Byer <bbyer@apple.com>
81138Date: Mon Mar 31 20:21:24 2008 -0700
81139
81140 fix compilation on systems that don't have
81141 XP_EVENT_SPACE_CHANGED defined
81142 (cherry picked from commit 6e833e41e2f3a84f7e5164b918737b3975517c12)
81143
81144commit b0789365719f48c0b477b4703f2072f932246b76
81145Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81146Date: Fri Nov 14 14:57:09 2008 -0800
81147
81148 XQuartz: Initial work on readding GLX to the 1.5 server
81149 (cherry picked from commit a186cab6ae82340770c13b37ff5df731ce2a0ceb)
81150
81151commit 1d7049f8fd3c0798250b9213149d3ce0b37ab77f
81152Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81153Date: Thu Nov 13 12:45:35 2008 -0800
81154
81155 Xquartz: Force sending mouse clicks to AppKit if we're over the menu bar
81156 (cherry picked from commit 7dff93ec1ac92aeaf3b70d5cfe787fa4a28c0dba)
81157
81158commit 3c124832642f1ec3228a57bea3d1eda68a188ff3
81159Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81160Date: Wed Aug 20 10:37:17 2008 -0700
81161
81162 XQuartz: Made X11Application.m a little more tidy.
81163 (cherry picked from commit 3520386261b838196a8918e8bee16bdccbc9781d)
81164
81165commit e917806fc556c2054b2835c8f85770b554c78092
81166Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81167Date: Fri Nov 14 10:56:21 2008 -0800
81168
81169 os/connection: TRANS_NOXAUTH bit comparison brainfart fix.
81170
81171 https://bugs.freedesktop.org/show_bug.cgi?id=18524
81172
81173commit 17bccff4ce905f46c074fd8ee7c6567b1371a0a1
81174Author: Fernando Carrijo <fcarrijo@yahoo.com.br>
81175Date: Fri Nov 14 12:36:20 2008 -0500
81176
81177 [dix] Remove a duplicate statement.
81178
81179 Signed-off-by: James Cloos <cloos@jhcloos.com>
81180
81181commit ba4e08244ed3923eecf26842dfc1df17c696e053
81182Merge: 81fd17f 5bad5d2
81183Author: Alex Deucher <alexdeucher@gmail.com>
81184Date: Thu Nov 13 15:04:18 2008 -0500
81185
81186 Merge branch 'master' of git+ssh://agd5f@git.freedesktop.org/git/xorg/./xserver
81187
81188commit 81fd17f5f49cdd2c10d0bf3b7ddeb8b5953886a5
81189Author: Alex Deucher <alexdeucher@gmail.com>
81190Date: Thu Nov 13 15:00:30 2008 -0500
81191
81192 Only add default modes if EDID supports continuous-frequency
81193
81194 When an EDID is present, only add the default mode pool if the
81195 continuous-frequency bit is set in the EDID. Should fix bugs
81196 like 18512.
81197
81198commit c232f3d673fb00d7fceb8e82741349d64e5ac0ad
81199Author: Adam Jackson <ajax@nwnk.net>
81200Date: Thu Nov 13 14:58:21 2008 -0500
81201
81202 xf86TargetExact should try harder if there's only one monitor attached.
81203
81204 If there's no preferred mode, but only one monitor, pick the
81205 biggest mode for its aspect ratio, assuming one exists.
81206
81207commit 5bad5d2abea2f51cbf53dc8dd016e90ccb086220
81208Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81209Date: Thu Nov 13 11:32:40 2008 -0800
81210
81211 XQuartz: Fixed XBell() when not using system alert.
81212 (cherry picked from commit aa0e9ab1c7a226cdcb3c5e62be159355a290faf8)
81213
81214commit 0fca8588a238c91580b305c3671567e910150a02
81215Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81216Date: Thu Nov 13 11:10:15 2008 -0800
81217
81218 XQuartz: Added more debugging to handoff spew
81219 (cherry picked from commit acefa7e3c8706ffedc052effd50b36ce10e72c22)
81220
81221commit 47d38218706564d1437dc3ff9d1bba3511db063c
81222Author: George Staplin <gstaplin@apple.com>
81223Date: Wed Nov 12 17:17:25 2008 -0700
81224
81225 XQuartz: GL: Handle the alpha differently when the alpha is equal to GLCAPS_COLOR_BUF_INVALID_VALUE.
81226
81227 This prevents visuals with odd sizes. The machine I use didn't have
81228 this problem, but it shows up on some others.
81229 (cherry picked from commit ed181382ddeb77019577d39b9c06b1cd839e18e4)
81230
81231commit 6b501d7cc55f1d06966cdf3caf36bcfc6cf18939
81232Author: Alex Deucher <alexdeucher@gmail.com>
81233Date: Thu Nov 13 13:57:30 2008 -0500
81234
81235 Add missing newline when printing edid info
81236
81237commit 33edd32702d963802e64d382124eabb9d0352fa8
81238Author: George Staplin <gstaplin@apple.com>
81239Date: Sun Nov 9 18:35:10 2008 -0700
81240
81241 XQuartz: Disable 2 calls to miSetVisualTypesAndMasks that seem to be unnecessary, and possibly wrong (with regard to masks and bits per RGB).
81242
81243 Use the settings queried from the system in xprScreen.c, rather than those 2 calls.
81244
81245 The 2 calls increased the total number of visuals a great deal (when using GLXEXT),
81246 and not all of the visuals were usable with GLX. Some of the visuals aren't usable
81247 with GLX still, such as DirectColor, but that seems to be acceptable based on my
81248 understanding of the manual that states "a subset of visuals are made available
81249 for OpenGL rendering."
81250 (cherry picked from commit 373b8a5f32509722e06d8776109e6c3f06645ee0)
81251
81252commit 1523476afa6657c20eba112e3ee051d45e721164
81253Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81254Date: Sat Nov 8 23:32:40 2008 -0800
81255
81256 XQuartz: CFBundleShortVersionString needs to conform to X.Y.Z as well... sigh
81257 (cherry picked from commit 54d20f6d33c347bcb37a0a8287c2d92fed2b9092)
81258
81259commit 1f301a0051d65296189248b2529616d9b33276b3
81260Author: George Staplin <gstaplin@apple.com>
81261Date: Thu Nov 13 01:27:33 2008 -0700
81262
81263 XQuartz: Restore the logic of AppleWMDoReorderWindow to the same as the 1.4 branch.
81264
81265 It was returning inverted values in comparison to the 1.4 branch. This resulted in
81266 the windows not drawing due to a deep path of: RootlessReorderWindow ->
81267 SCREENREC(pScreen)->imp->DoReorderWindow(winRec) - > xprDoReorderWindow ->
81268 AppleWMDoReorderWindow.
81269 (cherry picked from commit d1d398db762fe7d5720f9b5fafa2bbe8372fceac)
81270
81271commit f5841e96487234df5ead5f5c0fb3c587c418cb46
81272Author: Peter Hutterer <peter.hutterer@redhat.com>
81273Date: Tue Nov 11 22:50:35 2008 +1000
81274
81275 dix: don't store enter/leave and focus semaphores in a devPrivate.
81276
81277 We need them for each window, every time a window is allocated. Storing them
81278 in a devPrivate is the wrong thing to do.
81279
81280 This also removes the unused ENTER_LEAVE_SEMAPHORE_ISSET macro.
81281
81282 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81283 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
81284
81285commit 26f701704b4e536cd91bd8a9f7d2194793471998
81286Author: Peter Hutterer <peter.hutterer@redhat.com>
81287Date: Wed Nov 12 16:53:29 2008 +1000
81288
81289 dix: remove duplicate line in EnterLeaveEvent.
81290
81291 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81292
81293commit 8ff62ea5ba1bc331c886840b2b1e101cb4f2ebed
81294Author: Kristian Høgsberg <krh@redhat.com>
81295Date: Wed Nov 12 15:19:09 2008 -0500
81296
81297 CopySubBuffer expects GL style coordinates.
81298
81299commit d91243beb63008fff2ce6affa7db091786e8c49a
81300Author: Peter Hutterer <peter.hutterer@redhat.com>
81301Date: Fri Nov 7 08:39:38 2008 +1030
81302
81303 dix: formatting fix.
81304
81305commit dde09076c5c0c36a4a9321cd23a2c7922c4c755b
81306Author: Peter Hutterer <peter.hutterer@redhat.com>
81307Date: Wed Nov 5 12:52:03 2008 +1030
81308
81309 dix: remove superfluous variable devgrab
81310
81311 grab == devgrab anyway, this is a leftover from the time when we had two
81312 different grabs per device (core and XI grab).
81313
81314 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81315
81316commit fc3e8bfb82323477b54181c6d546c90f53f1ebd8
81317Author: Peter Hutterer <peter.hutterer@redhat.com>
81318Date: Tue Nov 4 15:27:30 2008 +1030
81319
81320 mi: clean up mieqProcessInputEvents, copy all events before processing.
81321
81322 Copy the EventRec's information into local variables before processing them,
81323 this should make it safer for upcoming threading and also makes it easier to
81324 read.
81325
81326 Simplify the event allocation code from the abyss it was before.
81327
81328 This also fixes a potential bug where a custom handler could scramble the
81329 event before the same -now scrambled- event was then passed through the
81330 master's custom event handler.
81331
81332 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81333
81334commit 31005efb1bb6be5a892ff905f754109dc3fa8fe6
81335Author: Fernando Carrijo <fcarrijo@yahoo.com.br>
81336Date: Mon Nov 10 13:23:45 2008 -0500
81337
81338 input: Remove useless conditional.
81339
81340commit 24383056d0513cc588cee5cee78f1cba2c0fbe89
81341Author: James Cloos <cloos@jhcloos.com>
81342Date: Sat Nov 8 13:04:20 2008 -0500
81343
81344 Don’t recommend fixing HAL if HAL support is disabled.
81345
81346 As reported in http://bugs.freedesktop.org/show_bug.cgi?id=18438
81347 the server suggests reconfiguring HAL if AllowEmptyInput is enabled
81348 and no input devices are known.
81349
81350 Instead of that notice, if HAL is disabled at configure time,
81351 AllowEmptyInput is enabled in the config and no input devices are
81352 found report those facts and recommend disabling AllowEmptyInput.
81353
81354commit b3c7e62664457a3802ebc23785446051624a8c2f
81355Author: James Cloos <cloos@jhcloos.com>
81356Date: Sat Nov 8 12:18:13 2008 -0500
81357
81358 Remove some null statements.
81359
81360 Remove several doubled statement-terminal semicolons.
81361
81362 Reported by Fernando Carrijo.
81363
81364commit 21c116219cd5c6845a0955f2d88fdb5fab5c17cf
81365Author: Michel Dänzer <michel@tungstengraphics.com>
81366Date: Sat Nov 8 01:25:14 2008 +0100
81367
81368 Fix typos which caused exaCompositeRects() to use an incorrect damage region.
81369
81370 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=18065 .
81371
81372 Also, it can bail if either width or height of the bounding box is <= 0, not
81373 only if both of them are.
81374
81375commit 59f9fb4b8c031df69b3592a26b77e744ff4a556e
81376Author: Luc Verhaegen <libv@skynet.be>
81377Date: Fri Nov 7 19:11:11 2008 +0100
81378
81379 XAA PixmapOps: Sync before accessing unwrapped callbacks.
81380
81381 When using any XAAPixmapOps, we call into unknown but freshly
81382 unwrapped callbacks (like fb ones). Unlike the XAA*Fallback calls,
81383 we did so without syncing first, exposing us to all kinds of
81384 synchronisation issues.
81385
81386 I believe that the rendering errors appeared now because *PaintWindow
81387 vanished (e4d11e58), and we just use miPaintWindow instead. This
81388 takes a less direct route to the hw and ends up at
81389 PolyFillRectPixmap, which very often left drawing artifacts.
81390
81391 We now sync accordingly, and no longer get the rendering artifacts i
81392 was methodically reproducing on radeonhd, radeon, unichrome...
81393
81394 Also, in order to allow driver authors to remove extensive syncing
81395 or flushing to hide this issue, create XAA_VERSION_ defines, put
81396 them in xaa.h and bump the patchlevel.
81397
81398 (novell bug #435791)
81399
81400commit 991c88b7542164194be73573e7644164416ea90c
81401Author: Julien Cristau <jcristau@debian.org>
81402Date: Fri Nov 7 17:36:38 2008 +0100
81403
81404 xfree86: xf86SetDepthBpp needs to respect the driver's depth24flags
81405
81406 When setting the depth to 24, leave bpp unset so the logic to pick
81407 a supported value is used instead of ignoring the driver's preference
81408 and forcing 32 bpp.
81409
81410commit d3d6be4948fa19947fd3b03e6694247109cc0ffb
81411Author: Julien Cristau <jcristau@debian.org>
81412Date: Fri Nov 7 18:36:00 2008 +0100
81413
81414 mi: Fix infinite loop on regen when swrast_dri.so is missing
81415
81416 The swrast DRI provider gets pushed on the glx provider stack at every
81417 server generation, so the stack turns into a circular list on regen.
81418
81419 X.Org bug#18388 <https://bugs.freedesktop.org/show_bug.cgi?id=18388>
81420
81421commit f59352881f315a634f60c9aac885b2764b28b167
81422Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81423Date: Thu Nov 6 14:41:55 2008 -0800
81424
81425 XQuartz: Make LS parse our version properly... it didn't like our CFBundleVersion
81426 (cherry picked from commit 980df1b4ff7d4dab9dc1462dc086542265b1fb1b)
81427
81428commit b4ca9dc2393ace2415163094b916f0d17ecba9ee
81429Author: Alan Coopersmith <alan.coopersmith@sun.com>
81430Date: Wed Nov 5 21:58:04 2008 -0800
81431
81432 When HAL returns a NULL property, print "(null)" instead of a NULL pointer
81433
81434 They've promised to fix Solaris printf soon to check for NULL pointers
81435 instead of segfaulting, but that won't help people on existing releases.
81436
81437commit 13d06f5aaf6120c902a323649615c1ce3d1b5359
81438Author: George Peter Staplin <gps@Georges-Workstation.local>
81439Date: Tue Nov 4 12:53:12 2008 -0700
81440
81441 XQuartz: GL: Add capability detection for depth buffers, and multisampling.
81442
81443 GL/capabilities.c: Add handleDepthModes(), and extend
81444 handleRendererDescription() for the various depth and multisampling flags.
81445
81446 Add initialization of the new config options to initConfig().
81447
81448 GL/capabilities.h: Add depth and multisample config members.
81449
81450 GL/visualConfigs.c: Add depth and multisampling support to the visual config
81451 setup.
81452 (cherry picked from commit f527381eea6a8ae6cd791475b2060d21fcf8efb2)
81453
81454commit 34bb06b292dc8b07d8602941aab3e69a73811314
81455Author: George Peter Staplin <gps@Georges-Workstation.local>
81456Date: Mon Nov 3 19:52:48 2008 -0700
81457
81458 XQuartz: GL: Work around problems with really deep visuals.
81459
81460 GL/capabilities.c: #if 0 (for now) any capabilities above 8 bits per channel,
81461 because they introduce drawing problems.
81462
81463 GL/indirect.c: Comment out some visual setup code that shouldn't be running, and
81464 actually seemed to cause some problems. The current visualConfigs.c code seems
81465 to do a reasonable job of setting up visuals for XQuartz.
81466
81467 GL/visualConfigs.c: Make use of the proper visual .class. Eliminate depth 0.
81468 It seems we really just want 24 for now, and 0 I think was a flaw in the original
81469 code.
81470 (cherry picked from commit 1e5f63f15e13a40a6e69a1505934d10d6990b6a2)
81471
81472commit 39e82a9b1250b23768136f3c397377a97f8b908c
81473Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81474Date: Sun Nov 2 15:52:22 2008 -0800
81475
81476 XQuartz: Replace deprecated API usage thanks to Robery Murphy
81477 (cherry picked from commit 71dd052412400362793f2f0c7c02bf4f4309738c)
81478
81479commit b4db46320a437ba8894cb8cfd4ee3abdcc894b01
81480Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81481Date: Fri Oct 31 20:18:27 2008 -0700
81482
81483 XQuartz: version string updated for 2.3.2_beta3
81484 (cherry picked from commit 81bb4128570b44d779126a8dffcd3c5620747383)
81485
81486commit c9051b684b524549eab6d5b88ee3e195a6f6fbe8
81487Author: Alan Coopersmith <alan.coopersmith@sun.com>
81488Date: Wed Nov 5 18:25:57 2008 -0800
81489
81490 Use OsSignal in Popen/Pclose to avoid SysV signal() stupidity
81491
81492commit d63ea510138c8b6de66184c78cda39ed9981fc1f
81493Author: Alan Coopersmith <alan.coopersmith@sun.com>
81494Date: Wed Nov 5 14:52:29 2008 -0800
81495
81496 Non-Linux OS'es should default to kbd driver, not now-dead keyboard driver
81497
81498commit 446d9443cea31e493d05c939d0128a8116788468
81499Author: Adam Jackson <ajax@redhat.com>
81500Date: Wed Nov 5 11:51:06 2008 -0500
81501
81502 linux: Drain the console fd of data when using evdev for keyboards
81503
81504 Works around a silly bug in the kernel that causes wakeup storms after
81505 too many keypresses. Should fix the kernel bug too, but this at least
81506 keeps the idle wakeup count below 1000/sec.
81507
81508commit 6d21fbf00648307208146aca0837ec63ea490659
81509Author: Adam Jackson <ajax@redhat.com>
81510Date: Wed Nov 5 11:39:46 2008 -0500
81511
81512 kdrive: Bye bye Xvesa
81513
81514commit 9a874a71a791c6110fd57b8a5c083f777a446d0f
81515Author: Egbert Eich <eich@freedesktop.org>
81516Date: Tue Nov 4 21:31:27 2008 +0100
81517
81518 DDX/RandR: When positioning outputs assume width/height = 0 if no modes exist.
81519
81520 When a user specifies the position of an output for which no modes exist
81521 (for whatever reason) assume that the width and height of this output
81522 is 0. The result will be the same as if this output isn't taken into
81523 consideration at all and thus should be sane. It will prevent a segfault
81524 when trying to determine the width and height of a non-existent mode.
81525
81526commit a4d62bbf215894bad8e19d99f7330c637d3d49e3
81527Author: Michel Dänzer <michel@tungstengraphics.com>
81528Date: Tue Nov 4 11:27:53 2008 +0100
81529
81530 AIGLX: Reinstate call to driver texOffsetFinish hook.
81531
81532 It was accidentally lost when factoring out __glXDRIdoReleaseTexImage, so this
81533 is a regression fix and should probably be backported to server-1.5-branch.
81534
81535commit 27f1ad466cd9662d638f4a8e59beafc32a8b2a8f
81536Author: Michel Dänzer <michel@tungstengraphics.com>
81537Date: Tue Nov 4 11:26:03 2008 +0100
81538
81539 AIGLX: Don't truncate offset returned by driver texOffsetStart hook on 32 bit.
81540
81541commit de1a8b68eb10c39a0d4a197107d68d7cc009edb2
81542Author: Peter Hutterer <peter.hutterer@redhat.com>
81543Date: Mon Nov 3 11:30:48 2008 +1030
81544
81545 xkb: extract the correct device in XkbFilterEvents.
81546
81547 If the event is an XI event, we need to work on the correct device, not on
81548 the VCK.
81549
81550 Adds XIGetDevice(event) function to extract the device from an event.
81551
81552commit d97239956667f8181f30271759573b75bf455fbb
81553Author: Peter Hutterer <peter.hutterer@redhat.com>
81554Date: Fri Oct 31 16:40:41 2008 +1030
81555
81556 Purge device-based WindowAccess code.
81557
81558 Really, this was a bad idea. It's not security, the UI features that would
81559 have been cool (e.g. clicking through windows) aren't implemented anyway, and
81560 there's nothing you can't achieve just by using plain XI anyway.
81561
81562 Requires inputproto 1.9.99.6.
81563
81564commit cbc6f983959595aa21c9dd72fac6a7070a650ef7
81565Author: Peter Hutterer <peter.hutterer@redhat.com>
81566Date: Fri Oct 31 14:36:48 2008 +1030
81567
81568 xkb: when faking mouse button events, fake them on the correct devices.
81569
81570 When MouseKeys are activated, keyboard devices may generate fake mouse button
81571 events through XKB. Let's get then running through the appropriate paths, i.e.
81572 as XI events on the correct device.
81573
81574 To make matters more fun, ProcessOtherEvents drops events if the DIX device
81575 state cannot be updated accordingly, i.e. all button events from keyboard
81576 devices.
81577 Hence we need to get the paired MD for the device in XkbDDXFakeDeviceButton,
81578 and post the event through the paired MD (usually the VCP).
81579
81580 Removes now-unused ddxFakeBtn.c.
81581
81582 Note: this patch only half-arsedly fixed button events, motion events are a
81583 more complicated matter.
81584
81585commit 4971315296cbf868dd738c1c0c1c504fcfe1b619
81586Author: Peter Hutterer <peter.hutterer@redhat.com>
81587Date: Mon Nov 3 17:47:36 2008 +1030
81588
81589 dix: clean up GetPointerEvents and GetKeyboardValuatorEvents.
81590
81591 Split into several functions, remove some stale comments.
81592
81593 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81594
81595commit 6bb0e0a53656db6168a053fb51b242a8640c1461
81596Author: Peter Hutterer <peter.hutterer@redhat.com>
81597Date: Fri Oct 24 16:22:09 2008 +1030
81598
81599 xnest: I'm sure xfree works fine, even if we're 64 bit.
81600
81601 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81602
81603commit 67b7e06c4fa4e8dafb64d9f3eb976836d0cc5889
81604Author: Daniel Stone <daniel@fooishbar.org>
81605Date: Fri Apr 25 19:39:48 2008 +0300
81606
81607 XKB: Remove pointless macros
81608
81609 These weren't even being used, which isn't overly surprising, given that
81610 they were already in the struct.
81611
81612 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81613
81614commit d9ff20e323ba9fb309629ac21946e85709625589
81615Author: Daniel Stone <daniel@fooishbar.org>
81616Date: Thu Oct 23 16:48:15 2008 +1030
81617
81618 DIX: Remove unnecessary prototype
81619
81620 We don't call XkbCopyKeymap from getevents.c anymore, so axe it.
81621
81622 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81623
81624commit 064ee458c7183eb91ef2e0f262b312ba1e3ad6e5
81625Author: Daniel Stone <daniel@fooishbar.org>
81626Date: Sat Aug 9 15:54:56 2008 +0300
81627
81628 XKB: Tiny cleanups to _XkbLookupAnyDevice
81629
81630 A couple of coding style cleanups, a warning fix via removing a
81631 now-unused label, and also put an else so we don't spuriously trip a
81632 condition that should admittedly never occur anyway.
81633
81634 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81635
81636commit 97c9e6a7139e07a0761be05034fae42fcd65ba36
81637Author: Daniel Stone <daniel@fooishbar.org>
81638Date: Fri Mar 28 17:10:28 2008 +0200
81639
81640 XKB: Fix thinko, causing warning (erroneously fixed in 5544c51447)
81641
81642 newTypes is a local variable which always has an address. newTypesIn,
81643 on the other hand, might be sus.
81644
81645 See also 5544c51447f551dfc6df64438873a7ce64743976.
81646
81647 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81648
81649commit caf1e6a1c92a2e70de9159f93d6f5b48c202dd81
81650Author: Daniel Stone <daniel@fooishbar.org>
81651Date: Tue Mar 18 10:05:00 2008 +0200
81652
81653 XKB: Move XKBrules.h to the server too
81654
81655 Get our own local copy of this, so we don't have to define XKB_IN_SERVER
81656 anymore.
81657
81658 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81659
81660commit f4036f6ace5f770f0fe6a6e3dc3749051a81325a
81661Author: Daniel Stone <daniel@fooishbar.org>
81662Date: Wed Oct 22 22:40:44 2008 +1030
81663
81664 Remove XEvIE
81665
81666 It's unmaintained and has been broken for quite a while; MPX finally
81667 smashed it completely.
81668
81669 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81670
81671commit 8c3753a0b711b5708b61f64ab4010ca01608d679
81672Author: Daniel Stone <daniel@fooishbar.org>
81673Date: Wed Oct 22 22:29:54 2008 +1030
81674
81675 A bit of whitespace cleanup in include directives.
81676
81677 It's #include <foo>, not # include <foo>
81678
81679 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81680
81681commit 889a8dc02bbd3926eaf16dce8f844dc58558e982
81682Author: Daniel Stone <daniel@fooishbar.org>
81683Date: Wed Oct 22 22:29:54 2008 +1030
81684
81685 Xi: XINPUT has been mandatory for ages
81686
81687 Ditch all #ifdef XINPUT, since it's always built.
81688
81689 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81690
81691commit 4cf2f4172abc4bf848c1726e6dac535713e08f0b
81692Author: Adam Jackson <ajax@redhat.com>
81693Date: Mon Nov 3 12:50:28 2008 -0500
81694
81695 Add xf86MonitorIsHDMI() helper.
81696
81697commit a7951a4dad902edea76a5cd68f833f0e48ae804b
81698Author: Michel Dänzer <michel@tungstengraphics.com>
81699Date: Mon Nov 3 10:00:54 2008 +0100
81700
81701 AIGLX: Allow 2D driver to prevent zero-copy texturing of a pixmap.
81702
81703 The driver can return ~0ULL to achieve this, e.g. if the pixmap doesn't fit
81704 into offscreen storage or if its pixel format isn't supported by the 3D engine
81705 for texturing.
81706
81707 See http://bugs.freedesktop.org/show_bug.cgi?id=17723 or
81708 http://bugs.freedesktop.org/show_bug.cgi?id=12385 .
81709
81710commit f8af66ddb21b4fb1ae5b4f7f852418ca7b0e7aec
81711Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81712Date: Fri Oct 31 20:03:34 2008 -0700
81713
81714 XQuartz: Add option to allow access to the menu in fullscreen mode
81715 (cherry picked from commit 637a9f4bd1ff2b55c870a08ead4940df0f9818e5)
81716
81717commit c5086badf0387d4d5af10882de90a1faa180550f
81718Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81719Date: Fri Oct 31 18:58:57 2008 -0700
81720
81721 XQuartz: LP64 related casting fixes from Bob Murphy
81722 (cherry picked from commit ea71710aaa7166ab510abe70f2dc47942de0ead7)
81723
81724commit 20239a4deebccc9f86586ef6b8ef61311a6fa6a0
81725Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81726Date: Fri Oct 31 18:58:25 2008 -0700
81727
81728 Rootless: LP64 related casting fixes from Bob Murphy
81729 (cherry picked from commit 40441c4e31167932721a79d7ca572caaa1d36e72)
81730
81731commit ecd55bc859eb5d7c639d5a6d531d9df524124247
81732Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81733Date: Fri Oct 31 15:10:34 2008 -0700
81734
81735 XQuartz: displayScreenBounds uses quartzEnableRootless rather than !quartzHasRoot
81736 (cherry picked from commit c883a78ef0c2e7e29437881db85d3815a89ab874)
81737
81738commit 4e762b8aa37b588e6cffcdd4a0152090190005e2
81739Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81740Date: Fri Oct 31 14:39:54 2008 -0700
81741
81742 Rootless: Force initialization of static variable before use.
81743 (cherry picked from commit cece29317269385aa7a0d3047e6f90e6a17d01eb)
81744
81745commit 9f49f5172d10070ebd4a9e72ac3c5227c6f35853
81746Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81747Date: Fri Oct 31 13:00:01 2008 -0700
81748
81749 XQuartz: Silence pointer debug spam
81750 (cherry picked from commit 9ca099e928653b5a0fb745186e17bfcc3d74ccc0)
81751
81752commit 93298c7f8f73bdb9892da0e41d2c2a67719dc736
81753Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81754Date: Fri Oct 31 12:55:46 2008 -0700
81755
81756 XQuartz: Don't map the menu bar to the screen area.
81757 (cherry picked from commit 1115917ada127a1cd6a8b61efe7fd134e2ea85fd)
81758
81759commit 31907986a6d4dc726d8ce0071f5884ce6cfea33b
81760Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81761Date: Fri Oct 31 12:37:32 2008 -0700
81762
81763 XQuartz: Fixed Xplugin thread safety issues, so rea-enable background window check
81764 (cherry picked from commit 5fc73b1141eca8b678e54412ca33afa31a76d088)
81765
81766commit 227d782a1bddfa1393cddd0c73c0288e18645c47
81767Author: George Peter Staplin <gps@Georges-Workstation.local>
81768Date: Thu Oct 30 13:44:55 2008 -0600
81769
81770 XQuartz: GL: Mark visuals that are slower than others, due to lack of acceleration.
81771
81772 This is used in libGL to determine a preferred visual. glxinfo -v will also
81773 report "visualCaveat=Slow" for such visuals.
81774 (cherry picked from commit d5c17f585a240be9f9135975738e62951cde3353)
81775
81776commit 5287602a4161cd036e3125cda09ba750190a0b47
81777Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81778Date: Wed Oct 29 17:20:11 2008 -0700
81779
81780 XQuartz: if 0 out a block to avoid thread-unsafe lockups
81781 (cherry picked from commit 60d7359a67900ad4601dbc73e07bcd8c803100a5)
81782
81783commit 93d7cd31c527afbf26f4e039afb6adbb6a60fab7
81784Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81785Date: Wed Oct 29 17:18:05 2008 -0700
81786
81787 XQuartz: Purge the old XQUARTZ_EXPORTS_LAUNCHD_FD option
81788 (cherry picked from commit 490756c7fbe0792c4bd6ae434e7a48e0beffa5c7)
81789
81790commit 0a076b6a7f3c97ff712412eee32e013d3bfd7364
81791Author: Peter Hutterer <peter.hutterer@redhat.com>
81792Date: Fri Oct 31 18:32:26 2008 +1030
81793
81794 xkb: remove unused label "out", clean up program flow.
81795
81796commit 971aef64416058665bc903fee39fea1619cd3727
81797Author: Peter Hutterer <peter.hutterer@redhat.com>
81798Date: Fri Oct 31 17:05:40 2008 +1030
81799
81800 dix: FreeDeviceClass() and friends can be static.
81801
81802commit 3c3a47d0f5ce7fb9548a3eefee9e72488d3ac279
81803Author: Peter Hutterer <peter.hutterer@redhat.com>
81804Date: Fri Oct 31 17:00:21 2008 +1030
81805
81806 include: remove unused declaration for RegisterPairingClient().
81807
81808 Matching functions got removed in 004876355b43fb4d3c2bc5653a7dc1bfd3f985ee.
81809
81810commit 7c5dec9f716962a6e67dba99418061cebe10e1d2
81811Author: Peter Hutterer <peter.hutterer@redhat.com>
81812Date: Fri Oct 31 16:56:49 2008 +1030
81813
81814 dix: remove unused GuessFreePointerDevice().
81815
81816 Not used since the MD/SD hierarchy was introduced many moons ago.
81817
81818commit 12599af285a610fb429245798330f2317c75ac2c
81819Author: Peter Hutterer <peter.hutterer@redhat.com>
81820Date: Fri Oct 31 16:50:28 2008 +1030
81821
81822 dix: NextFreePointerDevice() can be static.
81823
81824commit 5d0a945a030cd7626046e9f6c8d5c6bd170649d9
81825Author: Peter Hutterer <peter.hutterer@redhat.com>
81826Date: Fri Oct 31 16:29:42 2008 +1030
81827
81828 dix: SwitchCorePointer is long gone, bury it.
81829
81830commit eb014021c2b3604fc67daebd6860b5dc959fabb6
81831Author: Peter Hutterer <peter.hutterer@redhat.com>
81832Date: Fri Oct 31 16:27:58 2008 +1030
81833
81834 dix: PairDevices() can be static, remove from input.
81835
81836 PairDevices() is only ever called in EnableDevice().
81837
81838commit b6b26560d65d9468c89dd046396d9aeb14e1b227
81839Author: Peter Hutterer <peter.hutterer@redhat.com>
81840Date: Fri Oct 31 16:23:34 2008 +1030
81841
81842 Move EXTENSION_BASE and EXTENSION_EVENT_BASE to misc.h.
81843
81844commit f781a752e642cfe7d08f841b332c3ae507315598
81845Author: Peter Hutterer <peter.hutterer@redhat.com>
81846Date: Fri Oct 31 16:16:18 2008 +1030
81847
81848 Move MAX_DEVICES to misc.h, rename to MAXDEVICES for consistency.
81849
81850commit 245d1c162c7b13c98a9a28dc7ad441366d7e8006
81851Author: Peter Hutterer <peter.hutterer@redhat.com>
81852Date: Fri Oct 31 11:50:47 2008 +1030
81853
81854 xkb: ProcXkbSetCompatMap should do dry-runs, then normal runs.
81855
81856 Was doing only dry-runs, which kinda explains why changing the compat map
81857 didn't really have any effect.
81858 Fallout from e8c2a3d7c996cb41c4c44ba67acae5ff9438fc06.
81859
81860 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81861
81862commit 70e18a3b6b9d52169bca650f6cd4ef5d8ab40d95
81863Author: Alan Coopersmith <alan.coopersmith@sun.com>
81864Date: Thu Oct 30 20:38:34 2008 -0700
81865
81866 Refactor Solaris mapping code to work with standard shared/vidmem.c
81867
81868 Also merge sun_bios.c into sun_vid.c and upstream Solaris patch to
81869 keep aperture device open, to allow mappings to occur after X server
81870 has given up uid 0.
81871
81872commit 2809a56b4bc7f09f132c70dc39c6695495351c4d
81873Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
81874Date: Thu Oct 30 16:05:40 2008 -0400
81875
81876 xace: Fix an incorrect call to the RECEIVE_ACCESS hook.
81877
81878commit 0f2fd0577fe2740a32d3a2749e2c3582f891417c
81879Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
81880Date: Thu Oct 30 15:35:33 2008 -0400
81881
81882 xselinux: send more specific message types to libaudit.
81883
81884commit 4ce19b4477057a724b548e342d4c6da2b6721824
81885Author: Peter Hutterer <peter.hutterer@redhat.com>
81886Date: Thu Oct 30 16:02:13 2008 +1030
81887
81888 xfree86: fix xf86ScaleAxis once again.
81889
81890 Maybe one day I stop doing stupid patches like
81891 a3a7c12fcf8e4ac1418f9ea53f76091f309a721b.
81892
81893 So, if X < low, reset to low, and _not_ to high.
81894 If X > high, reset to high, and _not_ to low.
81895
81896commit f12d7ad1647d12ccf92291ebc0607defe6a634a1
81897Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81898Date: Tue Oct 28 12:38:05 2008 -0700
81899
81900 XQuartz: Pass along SIGINT and SIGTERM from the stub to X11.app
81901 (cherry picked from commit 183ca5d68b4f34e248749f304ce140de11bd451b)
81902
81903commit 273df54685768f74ed29926e88a63dc9f86744f1
81904Author: George Peter Staplin <gps@Georges-Workstation.local>
81905Date: Mon Oct 27 22:07:00 2008 -0600
81906
81907 XQuartz: pbproxy: Fix a bug that prevented nedit CLIPBOARD from being proxied.
81908
81909 It seems that we were calling copy_completed: too soon. Now we return instead
81910 of falling through to the free_propdata, and copy_completed:.
81911 (cherry picked from commit 5797d5b662817a3369a74a673f4a7c995cde75cf)
81912
81913commit 313d87341aff2396645d2c315086d487c0073d27
81914Author: George Peter Staplin <gps@Georges-Workstation.local>
81915Date: Mon Oct 27 20:44:02 2008 -0600
81916
81917 XQuartz: pbproxy: Work around bugs in some apps that aren't ICCCM compliant.
81918
81919 Eterm ends up setting the type of the property to TARGETS, instead of ATOM.
81920 That's why it wasn't working.
81921 (cherry picked from commit 9be5998ede7427f14dd9597e117b3d6b427ba4e5)
81922
81923commit 9007d3beea2e2d8b271f696d433cb30e32cda401
81924Author: George Peter Staplin <gps@Georges-Workstation.local>
81925Date: Mon Oct 27 16:34:24 2008 -0600
81926
81927 XQuartz: pbproxy: Add code to handle PICT conversion to PNG and JPEG.
81928
81929 This may work, unfortunately I don't have test apps that fail.
81930
81931 The way it works is by using an NSImage class initWithPasteboard:
81932 method, which we then get the TIFFRepresentation of, and convert
81933 to PNG or JPEG.
81934
81935 The TIFFRepresentation uses NSTIFFCompressionNone; which should be
81936 lossless.
81937 (cherry picked from commit 8d048cfa956f4a0860250cc836a6748912b37ad8)
81938
81939commit b742da0b719198a42a086049771792af84f470ca
81940Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81941Date: Sun Oct 26 13:17:49 2008 -0700
81942
81943 XQuartz: pbproxy: fixed blocking
81944 (cherry picked from commit bb330a1b1c480e89727fb75ff5aeb71f4afc5ee3)
81945
81946commit c885110aedc19e551de22a526397d77013c3dfcb
81947Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
81948Date: Sun Oct 26 10:22:12 2008 -0700
81949
81950 Xquartz: pbproxy: Some debugging about unhandled mime types... this will spam up quick, but will be useful for a while...
81951 (cherry picked from commit 3647215310f5a52126661b9ec473dcfeed25b40f)
81952
81953commit e7c26964eef6732c9603b26b7396f95443c5ba8a
81954Author: George Peter Staplin <gps@Georges-Workstation.local>
81955Date: Fri Oct 24 19:12:05 2008 -0600
81956
81957 XQuartz: GL: Fix several error handling mistakes in handleRenderDescriptions, that were leftover before the final code evolved.
81958 (cherry picked from commit fe35a3639081dc78ac6cc4123cde37770f6a39fd)
81959
81960commit 21ea9d82aab8ff0e0d1904d188f96f9cfb2b051d
81961Author: George Peter Staplin <gps@Georges-Workstation.local>
81962Date: Fri Oct 24 18:11:46 2008 -0600
81963
81964 XQuartz: GL: Add feature detection of color sizes, and accum sizes.
81965
81966 This gives us a *lot* more visuals, and they should be more accurate.
81967 Some of the visuals may have been wrong before. This may enable more
81968 OpenGL programs to behave correctly.
81969
81970 I also suspect libGL needs a little work to handle some of these correctly.
81971 (cherry picked from commit 98a084ce9c455260ecdb5f8eecb67f33dbb53b69)
81972
81973commit c264826da96ad1859dd112b17eb8aa9e5278478f
81974Author: Peter Hutterer <peter.hutterer@redhat.com>
81975Date: Thu Oct 16 11:22:29 2008 +1030
81976
81977 xfree86: If AEI is on, disable "kbd" and "mouse" devices.
81978
81979 This consists of two parts:
81980 In the implicit server layout, ignore those drivers when looking for a core
81981 device.
81982
81983 And after finishing the server layout, run through the list of devices and
81984 remove any that use mouse or kbd.
81985
81986 AEI is mutually exclusive with the kbd and mouse drivers, so pick either - or.
81987
81988 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
81989
81990commit 6c451859552e1fc78f6589617482f9ff96d7ed8a
81991Author: Peter Hutterer <peter.hutterer@redhat.com>
81992Date: Mon Oct 20 12:19:55 2008 +1030
81993
81994 config: don't add duplicate devices through HAL.
81995
81996 If HAL is restarted, the device list is sent to the server again, leading
81997 first to duplicate devices (and thus duplicate events), and later to a
81998 FatalError "Too many input devices."
81999
82000 dev->config_info contains the UDI for the device. If the UDI of a new devices
82001 is equal to one we already have in the device list, just ignore it.
82002
82003 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
82004
82005commit 102c4dac7c521941f52652152b1660cd7f559d56
82006Author: Julien Cristau <jcristau@debian.org>
82007Date: Sun Oct 26 13:17:31 2008 +0100
82008
82009 x86emu: fix sparse warnings
82010
82011 ops2.c:113:18: warning: Using plain integer as NULL pointer
82012 ops2.c:207:18: warning: Using plain integer as NULL pointer
82013
82014commit ffaaa1a198a77eb6800c08d4613ee1cc0b068ba0
82015Author: Julien Cristau <jcristau@debian.org>
82016Date: Sun Oct 26 13:13:21 2008 +0100
82017
82018 xfree86: fix compiler warnings in DoModalias()
82019
82020 The precedence of == is higher than that of &, so that code was
82021 probably buggy.
82022
82023 xf86Init.c: In function 'DoModalias':
82024 xf86Init.c:300: warning: suggest parentheses around comparison in operand of &
82025 xf86Init.c:304: warning: suggest parentheses around comparison in operand of &
82026 xf86Init.c:308: warning: suggest parentheses around comparison in operand of &
82027
82028commit 66fd05acd6118102aaaebc230d5c03e0758c6089
82029Author: Julien Cristau <jcristau@debian.org>
82030Date: Sun Oct 26 13:11:41 2008 +0100
82031
82032 xfree86: Silence compiler warnings about function prototypes
82033
82034 xf86Init.c:136: warning: function declaration isn't a prototype
82035 xf86Init.c:243: warning: function declaration isn't a prototype
82036 xf86Init.c:249: warning: function declaration isn't a prototype
82037
82038commit 4b96ba8ddf36f84d4b8c0f2964535bd90477a742
82039Author: Julien Cristau <jcristau@debian.org>
82040Date: Sun Oct 26 11:56:35 2008 +0100
82041
82042 Nuke unused miBankNewSerialNumber
82043
82044commit e85d5d707d51d6a1558dfcf0ebd50183b84be39d
82045Author: Julien Cristau <jcristau@debian.org>
82046Date: Sun Oct 26 11:55:50 2008 +0100
82047
82048 Ansify declaration of FontToXError()
82049
82050commit 30a7a06a9e91a91b8517f67897831c1407713e57
82051Author: Julien Cristau <jcristau@debian.org>
82052Date: Sun Oct 26 11:38:44 2008 +0100
82053
82054 Nuke prototype for nonexistent function
82055
82056 miDoInitVisuals was removed in commit
82057 94825ad3c72a1c3f6a61199c302841f63241cf87, but its prototype was left.
82058
82059commit 5244690a5e0285b6243ee1f28ae6c8287a32b73d
82060Author: Julien Cristau <jcristau@debian.org>
82061Date: Sun Oct 26 11:37:11 2008 +0100
82062
82063 Add prototypes for strlcpy/strlcat
82064
82065commit eadf90aa65d1ea63ddcd74e60e66a79ad567c7e8
82066Author: Julien Cristau <jcristau@debian.org>
82067Date: Sun Oct 26 11:20:33 2008 +0100
82068
82069 xaa: Fix compiler warning (implicit declaration)
82070
82071 xaaInit.c: In function 'XAAInit':
82072 xaaInit.c:201: warning: implicit declaration of function 'miInitializeCompositeWrapper'
82073 xaaInit.c:201: warning: nested extern declaration of 'miInitializeCompositeWrapper'
82074
82075commit aee514cfb5e5a59b7770699765ccdad8fe42f904
82076Author: Julien Cristau <jcristau@debian.org>
82077Date: Sun Oct 26 11:16:05 2008 +0100
82078
82079 xfree86: Fix compiler warnings (implicit declarations)
82080
82081 Add missing includes to fix the following warnings:
82082 xf86DGA.c: In function 'DGAProcessKeyboardEvent':
82083 xf86DGA.c:1050: warning: implicit declaration of function 'UpdateDeviceState'
82084 xf86DGA.c:1050: warning: nested extern declaration of 'UpdateDeviceState'
82085 xf86Xinput.c: In function 'xf86ActivateDevice':
82086 xf86Xinput.c:303: warning: implicit declaration of function 'AssignTypeAndName'
82087 xf86Xinput.c:303: warning: nested extern declaration of 'AssignTypeAndName'
82088 xf86Xinput.c:311: warning: implicit declaration of function 'DeviceIsPointerType'
82089 xf86Xinput.c:311: warning: nested extern declaration of 'DeviceIsPointerType'
82090 xf86Xinput.c:324: warning: implicit declaration of function 'XkbSetExtension'
82091 xf86Xinput.c:324: warning: nested extern declaration of 'XkbSetExtension'
82092
82093commit add946daed531e29935593fb2e7ee70286185baf
82094Author: Julien Cristau <jcristau@debian.org>
82095Date: Sun Oct 26 11:14:40 2008 +0100
82096
82097 Nuke unused variables
82098
82099commit f2bda61c03f64133ef68dd1557b26ae7dc5beb0e
82100Author: Julien Cristau <jcristau@debian.org>
82101Date: Sun Oct 26 11:05:13 2008 +0100
82102
82103 Xext: Fix compiler warnings
82104
82105 Make GEClientGone static and include registry.h, to fix the following
82106 warnings:
82107 geext.c:225: warning: no previous prototype for 'GEClientGone'
82108 geext.c: In function 'GEExtensionInit':
82109 geext.c:280: warning: implicit declaration of function 'RegisterResourceName'
82110 geext.c:280: warning: nested extern declaration of 'RegisterResourceName'
82111
82112commit 24e863b0eb6ff11010a14cfd252a39df87a09d0e
82113Author: Bernhard R. Link <brlink@debian.org>
82114Date: Sun Oct 26 00:19:34 2008 +0200
82115
82116 Bug#16501: autodetection of sbus graphic cards
82117
82118 Add automatic detection of the graphic driver to load for sbus devices.
82119 This allows xorg to work on those devices without a "Device" section.
82120
82121 Debian bug#483942.
82122
82123 Signed-off-by: Julien Cristau <jcristau@debian.org>
82124
82125commit fd97ff1bdd30c358525068a4c833f681e00719a2
82126Author: Julien Cristau <jcristau@debian.org>
82127Date: Sat Oct 25 22:54:15 2008 +0200
82128
82129 Bug#5851: Document the "Log" option in xorg.conf(5)
82130
82131commit 80b6768536fa439bfc5ce948bc8845a6f4af857b
82132Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82133Date: Thu Oct 23 17:32:32 2008 -0700
82134
82135 XQuartz: Do check_xinitrc() later to avoid possible deadlock.
82136 (cherry picked from commit af95174a088dc787d124ec548b1005f5b67c9e8d)
82137
82138commit d837d4e2af35a83c7932e4d0422097ebd1f6ba2c
82139Author: George Peter Staplin <gps@Georges-Workstation.local>
82140Date: Thu Oct 23 17:56:46 2008 -0600
82141
82142 XQuartz: pbproxy: pbproxy.h had invalid macro usage with 2 #else statements, so it wasn't building. Fix the build.
82143 (cherry picked from commit 6bcbb9ba85d95d7df63811a482abc39c4d595366)
82144
82145commit 6a2693b2f9cfdec6f30e2ca4c4176f986f97249e
82146Author: George Peter Staplin <gps@Georges-Workstation.local>
82147Date: Thu Oct 23 17:40:23 2008 -0600
82148
82149 XQuartz: Remove the enable_stereo default. We now use feature detection to enable stereo visuals.
82150
82151 We need to revisit the GL/indirect.c changes. I think indirect.c should
82152 be using the same feature detection and GLX visual config code. The
82153 indirect changes will require testing...
82154 (cherry picked from commit 49b7a7c6f20ca85d4d624444eb83db2f91344591)
82155
82156commit b015a021b28623b7b734833dc9b8269c41db4e8a
82157Author: Alan Coopersmith <alan.coopersmith@sun.com>
82158Date: Tue Oct 21 20:30:35 2008 -0700
82159
82160 Allow non-gcc compilers to build xf86Xinput.c
82161
82162commit b7cc6ca22a4ff7a94abee6962a07fdd0b563d26a
82163Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82164Date: Thu Oct 23 16:33:11 2008 -0700
82165
82166 miext/rootless: Silence some warnings by using proper C prototypes.
82167 (cherry picked from commit 050e46e09efcc7e6f090fa4a749e9076c972b760)
82168
82169commit b397cc727787a854417b11b482cce4807fc8ab58
82170Author: George Peter Staplin <gps@Georges-Workstation.local>
82171Date: Thu Oct 23 16:19:58 2008 -0600
82172
82173 miext/rootless: Fix the usage of DeleteProperty so that this builds again.
82174 (cherry picked from commit 3de250e28a9a86bcae4464a15c2025805877ce1e)
82175
82176commit 28d4f71193d2be26d1dcae724067539ac9db1123
82177Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82178Date: Thu Oct 23 15:01:03 2008 -0700
82179
82180 XQuartz: pbproxy: Fix compilation on case-sensitive file systems...
82181 (cherry picked from commit 6622efca216663c99d112c7a226bde691d1f3215)
82182
82183commit 40f9b6bf0dc62098680235a958e0733770902cca
82184Author: Adam Jackson <ajax@redhat.com>
82185Date: Thu Oct 23 17:33:52 2008 -0400
82186
82187 Argh, macros.
82188
82189commit 8b3c278fe7502f9343355f09213837622b0d9ca0
82190Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82191Date: Thu Oct 23 11:39:04 2008 -0700
82192
82193 Rootless: Properly typed expose_1
82194 (cherry picked from commit 78e874fe17174ac8f61586573edaf7f16ece7b21)
82195
82196commit 3e999d35c26b0ffb894c257a3057d44b949744bd
82197Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82198Date: Thu Oct 23 11:37:13 2008 -0700
82199
82200 XQuartz: Honor the Spaces preference for "When switching to an application, switch to a space with open windows for the application"
82201 (cherry picked from commit bf561a06b2bf60395d26026ac7627d0cb254bc07)
82202
82203commit 75067d12e4f4daa080dfb0209a9abf9c5654271f
82204Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82205Date: Tue Oct 21 19:36:48 2008 -0700
82206
82207 XQuartz: Added missing semicolons to Localizable.strings
82208 (cherry picked from commit e9fe3f36d9529f00daeefa1379cdd6f01a88f410)
82209
82210commit a003932dccc438c4500d383982cee2376e808bd5
82211Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82212Date: Tue Oct 21 11:23:42 2008 -0700
82213
82214 XQuartz: Default dpi to 96 instead of 75
82215 (cherry picked from commit 08a5a333d7fea68146cd9cf50682b811378ab3ce)
82216
82217commit 2d764e21f278ea26bae135db6dae07b94651951c
82218Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82219Date: Tue Oct 21 11:19:06 2008 -0700
82220
82221 XQuartz: pbproxy: SIGHUP reloads preferences.
82222 (cherry picked from commit cbc0935ef74ebd76ed3db16d6ab86043eeed42a3)
82223
82224commit 0b695262176e96f508921905ee7f9cc086c7daf0
82225Author: Adam Jackson <ajax@redhat.com>
82226Date: Thu Oct 23 15:01:58 2008 -0400
82227
82228 Change some #define foo() -> static void foo()
82229
82230commit aab73d719a8de6b830e47c7f47860b7990ea8d03
82231Author: Adam Jackson <ajax@redhat.com>
82232Date: Thu Oct 23 13:20:59 2008 -0400
82233
82234 Fix previous cherry pick for API changes
82235
82236commit 8767fc8d47276c7489ae50dd556a446a462776ea
82237Author: Keith Packard <keithp@keithp.com>
82238Date: Fri Apr 4 12:11:14 2008 -0700
82239
82240 Wrap AddTraps in exa and damage.
82241
82242 This fine (and unused) function wasn't ever wrapped which made it not work
82243 under exa.
82244
82245 (cherry picked from commit 06e7e1d0486e8c516a9b3219a2c86026f88825fc)
82246
82247commit 4c926dbac65ceb699b91bb0d3fd0005a5854572d
82248Author: Peter Hutterer <peter.hutterer@redhat.com>
82249Date: Thu Oct 23 22:07:37 2008 +1030
82250
82251 dix: fix two more endian issues, correct initial "enabled" value. #18111
82252
82253 EnableDevice and DisableDevice both change the property too.
82254
82255 And enabled must be set to FALSE in AddInputDevice, the device is not enabled
82256 yet.
82257
82258 X.Org Bug 18111 <https://bugs.freedesktop.org/show_bug.cgi?id=18111>
82259
82260commit 084ae9e38812f19751a086eea7ce555761ae5a17
82261Author: Jerome Glisse <glisse@freedesktop.org>
82262Date: Thu Oct 23 11:28:53 2008 +0200
82263
82264 dri2: fail at context creation if driver fail to create it's context
82265
82266commit ff60e592f003d8bc45789a1a14e86be22f5242b2
82267Author: Peter Hutterer <peter.hutterer@redhat.com>
82268Date: Thu Oct 23 09:19:41 2008 +1030
82269
82270 dix: don't allow VCP/VCK be disabled through properties.
82271
82272commit 98f01c2abe4771d76febf8fe70111b2bddfab776
82273Author: Peter Hutterer <peter.hutterer@redhat.com>
82274Date: Thu Oct 23 09:13:41 2008 +1030
82275
82276 dix: fix endianess issue in AddInputDevice. #18111
82277
82278 dev->enabled is a Bool. Bool is two bytes.
82279 BOOL on the other hand is a protocol type and always 1 byte. So copy the value
82280 into the one-byte type before passing it into XIChangeDeviceProperty.
82281
82282 Found by Michel Dänzer.
82283
82284 X.Org Bug 18111 <http://bugs.freedesktop.org/show_bug.cgi?id=18111>
82285
82286commit f6cbe0326c57e3c9897f17baef78199b06537c44
82287Author: Peter Hutterer <peter.hutterer@redhat.com>
82288Date: Wed Oct 22 18:16:12 2008 +1030
82289
82290 dix: init remaining three xkb fields to zero in a new client.
82291
82292commit 0b56b44addc323a00eb7cd86240cb0dd4275bcf8
82293Author: Peter Hutterer <peter.hutterer@redhat.com>
82294Date: Mon Oct 20 17:06:40 2008 +1030
82295
82296 xfree86: AllowEmptyInput is true by default - update the xf86Info defaults.
82297
82298 Also set AutoAddDevices and AutoEnableDevices to their defaults.
82299
82300 And in doing so, switch the rest of the defaults over to named intializers.
82301
82302 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
82303
82304commit d72cd753b99fae147ef4c189700fc697f1ea7fb0
82305Author: Matthieu Herrb <matthieu.herrb@laas.fr>
82306Date: Tue Oct 21 22:32:57 2008 +0200
82307
82308 Close well known connections in ServerAbort()
82309
82310commit 81e197b2a5457bb9f7ed72f82c2d33fd8dbfc202
82311Author: Adam Jackson <ajax@redhat.com>
82312Date: Tue Oct 21 16:00:32 2008 -0400
82313
82314 Bug #18159: Spell "anisotropic" correctly
82315
82316commit 3891dd892449fcdb7a514e3c5e7e763ba7e74003
82317Author: Adam Jackson <ajax@redhat.com>
82318Date: Tue Oct 21 13:10:44 2008 -0400
82319
82320 exa: Add exaDrawableIsOffscreen() to the driver API.
82321
82322commit c837d182c4674c01296bb7efd7d834d9dfd55f83
82323Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82324Date: Tue Oct 21 09:11:59 2008 -0700
82325
82326 XQuartz: Only send mouse events while in the background if we're over a visible window.
82327
82328 This "breaks" xeyes but gets rid of annoying tooltips.
82329 (cherry picked from commit 10090a132a0110fe24c8c886d9f53e6e947f8038)
82330
82331commit 2a9613980f93959eedf9396735785d4c50546e2d
82332Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82333Date: Tue Oct 21 09:07:08 2008 -0700
82334
82335 XQuartz: More fullscreen fixes
82336 Start in rooted mode when enabled
82337 Make the hotkey cmd-alt-a behave
82338 (cherry picked from commit 22a54e46176c9cbdfaaa59ef33a6ae6a9c4969ef)
82339
82340commit e4ea1494debda4d3da03864c2640921fe0430694
82341Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82342Date: Mon Oct 20 12:23:04 2008 -0700
82343
82344 XQuartz: Cleaned up keyboard init and map reload.
82345 (cherry picked from commit f78c9fc06cac2cc6ddfd6e9ba435dd26a57d1f51)
82346
82347commit ebb2e1449cfb382a806e04c52d42240670717069
82348Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82349Date: Mon Oct 20 09:46:34 2008 -0700
82350
82351 XQuartz: Fixed follow-keyboard-layout
82352 ≈
82353 (cherry picked from commit a9f9fbf512c9fd5a773fce402182486edc71d5ab)
82354
82355commit f67cd31d2cf1b8ad850bd21dbed5a975d77b4c54
82356Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82357Date: Tue Oct 21 09:29:37 2008 -0700
82358
82359 XQuartz: More progress towards fullscreen...
82360 (cherry picked from commit 628cae86b4ad02c49d20df1e1afdcb7abcabb649)
82361 (cherry picked from commit b47b7ab6d6d0a7f6cbf678a5755124d6b7516d3e)
82362
82363commit d434ed400986653771bbd60c224a2275bd6147ef
82364Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82365Date: Fri Oct 17 09:25:26 2008 -0700
82366
82367 XQuartz: Fixed missing key-equiv for Hide-Others.
82368 (cherry picked from commit 2c4b0f9615b646712609ebef3c0851c9a22e781b)
82369
82370commit 3695f62d072e7864fdcee385c8b5ca40f2018101
82371Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82372Date: Thu Oct 16 22:36:00 2008 -0700
82373
82374 XQuartz: Require applewmproto-1.1.1
82375 (cherry picked from commit ed235e3406e603a8a92c4b72f9f9018bae72ad65)
82376
82377commit f935c11398b93860461672e88814026d75b71689
82378Author: George Peter Staplin <gps@Georges-Workstation.local>
82379Date: Thu Oct 16 13:09:21 2008 -0600
82380
82381 XQuartz: pbproxy: Fix 2 bugs with the preferences.
82382
82383 Add dump_prefs() for instrospection.
82384
82385 Make prefs_get_bool a little more clear.
82386
82387 In get_property don't assume that the Mac realloc(NULL, 0); returns
82388 non-NULL. We shouldn't depend on implementation-defined behavior too
82389 much.
82390
82391 Add more commentary in various spots.
82392
82393 Add TARGETS in the TARGETS response. I don't know why some X11 apps
82394 do this, but it could be something that some weird toolkit/app tests
82395 for.
82396
82397 In reload_preferences add preferences synchronization, so that xpbproxy
82398 reloads the preferences, and picks up any changes.
82399
82400 Fix a typo/misspelling of the sync_clipboard_to_pasteboard preference.
82401 (cherry picked from commit 165cbbb90c793d9335ea76f0274b1f60ad295903)
82402
82403commit 4d978af5c71430f0043b1272f8dc6344b8f03f12
82404Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82405Date: Wed Oct 15 21:26:58 2008 -0700
82406
82407 XQuartz: Fixed window levels.
82408 (cherry picked from commit ef4d37c73f31048dd74efee2c63fa08b92608f87)
82409
82410commit 58eacf0b7beb8c6f1971570ede0bfd3976a84665
82411Author: Rémi Cardona <remi@gentoo.org>
82412Date: Tue Oct 21 10:07:24 2008 -0400
82413
82414 Force LC_ALL=C when building xf86DefModeSet.c
82415
82416 Fixes build in not-quite-latin locales.
82417
82418commit 2188582e5ea90edb432a2f421d0a267439ba08f9
82419Author: Michel Dänzer <michel@tungstengraphics.com>
82420Date: Mon Oct 20 09:55:24 2008 -0400
82421
82422 EXA: Avoid excessive syncing in PutImage
82423
82424commit 8d0cb507ae568c8da92f10ff1188d797bcaa3d1a
82425Author: Peter Hutterer <peter.hutterer@redhat.com>
82426Date: Fri Oct 17 09:59:29 2008 +1030
82427
82428 dmx: avoid hilarious #define Xcalloc xcalloc game.
82429
82430commit 8a23707b866c11807e01056db97afec56e9b19c8
82431Author: Peter Hutterer <peter.hutterer@redhat.com>
82432Date: Fri Oct 17 09:15:39 2008 +1030
82433
82434 dmx: change remainder of manual EQ allocation to GetEventList().
82435
82436commit 8de26770a41ec87c46eed2eddfde6f867d71fe1f
82437Author: Pierre Willenbrock <pierre@pirsoft.de>
82438Date: Thu Oct 16 14:28:14 2008 -0400
82439
82440 RANDR: Fix output property event delivery.
82441
82442commit d936a4235c9625bd41569cef3452dd086284e0d7
82443Author: Peter Hutterer <peter.hutterer@redhat.com>
82444Date: Sun Oct 12 21:58:30 2008 +1030
82445
82446 xfree86: if AllowEmptyInput is true, enable RAW mode on the console.
82447
82448 Usually, the console is set to RAW in the kbd driver. If we hotplug all input
82449 devices (i.e. the evdev driver for keyboards) and the console is left as-is.
82450 As a result, the evdev driver must put an EVIOCGRAB on the device to avoid
82451 characters leaking onto the console. This again breaks many things, amongst
82452 them lirc, in-kernel mouse button emulation and HAL.
82453
82454 This patch sets the console to RAW if AllowEmptyInput is on.
82455
82456 Use-cases:
82457 1. AEI is off
82458 1.1. Only kbd driver is used - behaviour as-is.
82459 1.2. kbd and evdev driver is used: if evdev does not grab the device,
82460 duplicate events are generated.
82461 2. AEI is on
82462 2.1. Only evdev driver is used - behaviour as-is, but evdev does not need
82463 to grab the device anymore.
82464 2.2. evdev and kbd are used: duplicate key events are generated if evdev
82465 does not grab the device.
82466
82467 1.2 is a marginal use-case that can be fixed by adding a "grab" option to the
82468 evdev driver (update of xorg.conf is needed).
82469
82470 2.2 is an issue. If we have no ServerLayout section, AEI is on, but devices
82471 specified in the xorg.conf are still added [1], resulting in duplicate events.
82472 This is a common configuration and needs sorting out.
82473
82474 [1] 2eaed4a10fe5bf727579bca4ab8d4a47c8763a7d
82475
82476 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
82477 Signed-off-by: Adam Jackson <ajax@redhat.com>
82478
82479commit 84ef8ed6fbefd8d6c0aaa3c862879f9804299bd8
82480Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82481Date: Wed Oct 15 14:11:18 2008 -0700
82482
82483 XQuartz: implemented primary-on-grab and fixed clipboard-to-pasteboard
82484 (cherry picked from commit bcb83eea729a01026d99d1cfc2b77385b5b275fd)
82485
82486commit 0195d318465d5a6a9039091bcb252202471df5a4
82487Author: Luc Verhaegen <libv@skynet.be>
82488Date: Wed Oct 15 21:46:27 2008 +0200
82489
82490 GLX: fix build when dri2 _is_ available.
82491
82492 Caught by Julien Cristau.
82493
82494commit 398cdf959afadfa964aece13cc9d89b2dd589930
82495Author: Luc Verhaegen <libv@skynet.be>
82496Date: Wed Oct 15 18:24:10 2008 +0200
82497
82498 GLX: fix build when dri2 is not available.
82499
82500commit 2e60ebbd254290e19492b97c960d9678513c94bf
82501Author: Daniel Stone <daniel@fooishbar.org>
82502Date: Wed Oct 15 14:10:51 2008 +0100
82503
82504 XFree86: Simplify DontZap/-retro interaction
82505
82506 How hard could it possibly be, right?
82507
82508commit 66a87517bd80b21e107df9d57968d81a92f91fd5
82509Author: Alan Hourihane <alanh@tungstengraphics.com>
82510Date: Wed Oct 15 12:00:16 2008 +0100
82511
82512 re-enable DRI2
82513
82514commit ced6690284fd334f225bbc94685c114ad3ab96d8
82515Author: Kristian Høgsberg <krh@redhat.com>
82516Date: Tue Oct 7 13:49:28 2008 -0400
82517
82518 dri2: Update to latest protocol draft.
82519
82520 Mainly rename SwapBuffers to CopyRegion, which adds the xfixes region
82521 argument and the bitmask argument to let us extend it in the future.
82522
82523commit 87a016ae00feac3fbaa7e7a518076a3852d49554
82524Author: Maarten Maathuis <madman2003@gmail.com>
82525Date: Wed Oct 15 00:02:13 2008 +0200
82526
82527 exa: restore {x,y}{Src,Dst} to their original values when !done
82528
82529commit b677bff97b62c3931df853bae25ab485892ae173
82530Author: Adam Jackson <ajax@redhat.com>
82531Date: Tue Oct 14 14:30:19 2008 -0400
82532
82533 int10: Degrade the checksum warning to X_INFO.
82534
82535 I've seen about one case in three years where this has actually been
82536 correlated with the real cause of failure, and we've trained people to
82537 freak out about X_WARNING, so let's be less alarmist.
82538
82539commit bd9c6b3a4d726a3f83ac6d8cf7211eddbc28f25a
82540Author: Adam Jackson <ajax@redhat.com>
82541Date: Tue Oct 14 14:04:01 2008 -0400
82542
82543 EDID: Catch even more cases of encoding aspect as size.
82544
82545 Very cute, Samsung, not only do you claim to be 16cm by 9cm in the
82546 global size record, you also claim to be 160mm by 90mm in the detailed
82547 timings. Grrr.
82548
82549commit ad677238bc96a8578113bbe76d605d7a87aca44c
82550Author: Adam Jackson <ajax@redhat.com>
82551Date: Tue Oct 14 13:00:50 2008 -0400
82552
82553 Add backtrace definitions to dix-config.h.in
82554
82555commit 8e368cf5b964f1d29fda0a463f9510457619b14d
82556Author: Nathaniel McCallum <nathaniel@natemccallum.com>
82557Date: Tue Oct 14 09:50:29 2008 -0400
82558
82559 Xorg: add -modalias option
82560
82561 This scans the installed video drivers and prints a Linux-style modalias
82562 listing of the devices each driver claims to support.
82563
82564commit 3eb52de7f28b0050582f9ac4c28bc894d3f06f4b
82565Author: Adam Jackson <ajax@redhat.com>
82566Date: Tue Oct 14 09:45:19 2008 -0400
82567
82568 kdrive: Hardcode a different wrong value for number of mouse buttons
82569
82570 This should really come from the input driver directly, but, kdrive.
82571
82572commit f3f6ea89aa9e0ffe9e37bc059e5e6bf75be4ee9f
82573Author: Peter Hutterer <peter.hutterer@redhat.com>
82574Date: Wed Oct 8 14:12:21 2008 +1030
82575
82576 Xi: check all handlers before applying property changes.
82577
82578 The current code exposes to inconsistent updates, i.e. if handler N succeeds
82579 but handler N+1 fails in setting the property, an error is returned to the
82580 client although parts of the server now behave as if the property change
82581 succeeded.
82582
82583 This patch adds a "checkonly" parameter to the SetProperty handler. The
82584 handlers are then called twice, once with checkonly set to TRUE.
82585 On the checkonly run, handlers _MUST_ return error codes if the property
82586 cannot be applied. Handlers are not permitted to actually apply the changes.
82587 On the second run, handlers are permitted to apply property changes.
82588 Errors codes returned on the second run are ignored.
82589
82590commit ad67e3f063aa79247270f29e989bbfe5f62c9ed7
82591Author: Peter Hutterer <peter.hutterer@redhat.com>
82592Date: Sun Oct 12 10:19:36 2008 +1030
82593
82594 dix: remove duplicate code in ReleaseActiveGrabs
82595
82596 Spotted by Colin Harrison.
82597
82598commit 3fc4f40b6c6cb416c9dc4bdb35c91b4f32c03ccc
82599Author: Aaron Plattner <aplattner@nvidia.com>
82600Date: Sun Oct 12 16:08:26 2008 -0700
82601
82602 Restore xf86{Enable, Disable}GeneralHandler.
82603
82604 These were useful as part of the generic handler ABI, and are used by the NVIDIA
82605 driver.
82606
82607 This reverts part of commit 50081d2dfb79878cb931a15c265f0d60698dfd39.
82608
82609commit 2217d22a76cdb2460f9683a6bf74c7248612889d
82610Author: Aaron Plattner <aplattner@nvidia.com>
82611Date: Sun Oct 12 16:07:24 2008 -0700
82612
82613 Revert "xfree86: xf86{Enable, Disable}InputHandler can be static."
82614
82615 These were potentially useful as part of the input handler ABI, even if nobody
82616 currently uses them.
82617
82618 This reverts commit 278c11f01fbc6d6bd91c5a7127928c9ef5d29fca.
82619
82620commit 5b336585a4cdf11d20831a9536ad581e959ea7f1
82621Author: Maarten Maathuis <madman2003@gmail.com>
82622Date: Fri Oct 10 23:23:02 2008 +0200
82623
82624 dri: don't set the dixPrivate key to NULL, as this is a staticly set variable.
82625
82626 - This breaks on the 2nd server generation.
82627 - No other subsystem seems to NULL their key.
82628 - This should fix bug 17982.
82629
82630commit b736f477f5324f79af30fc0f941ba0714a34ccda
82631Author: Adam Jackson <ajax@redhat.com>
82632Date: Fri Oct 10 16:33:24 2008 -0400
82633
82634 mieq: Backtrace when the queue overflows.
82635
82636 Since we're probably stuck down in a driver somewhere, let's at least
82637 try to point out where. This will need to be rethought when the input
82638 thread work lands though.
82639
82640commit 94ed0ba1b5043ad9fc33b42756af447d5ab15bbd
82641Author: Adam Jackson <ajax@redhat.com>
82642Date: Fri Oct 10 15:53:48 2008 -0400
82643
82644 Move xorg_backtrace() up to the OS level so we can call it from DIX.
82645
82646commit f650c1aeb06e957626bb0a74eddbb133f59dbecb
82647Author: Adam Jackson <ajax@redhat.com>
82648Date: Fri Oct 10 14:07:28 2008 -0400
82649
82650 Revert "Don't leak EDID_block"
82651
82652 This reverts commit 05cdef18ca668ffc86b71033f71cfbd262cdd994.
82653
82654 http://lists.freedesktop.org/archives/xorg/2008-October/039305.html
82655
82656commit 0660dd9d7009147c395b9ea904539f76f55b9a7f
82657Author: Adam Jackson <ajax@redhat.com>
82658Date: Fri Oct 10 13:41:50 2008 -0400
82659
82660 EDID: Catch monitors that encode aspect ratio for physical size.
82661
82662 This is not legal in either EDID 1.3 or 1.4, but hey, when did a little
82663 thing like legality stop anyone.
82664
82665commit 956a4ba40b28251ab2d88b5d5b0796e01afd2bdd
82666Author: Maarten Maathuis <madman2003@gmail.com>
82667Date: Fri Oct 10 00:28:34 2008 +0200
82668
82669 pixmap: Add some notes to devKind and devPrivate.
82670
82671commit 4808bdec45775342eb9a6352b41e4919e1a69279
82672Author: Peter Hutterer <peter.hutterer@redhat.com>
82673Date: Thu Oct 9 16:23:24 2008 +1030
82674
82675 Xi: don't memcpy the KeyClassRec from SD to MD. #16167
82676
82677 Most of its component get copied during CopyKeyClass anyway.
82678 The ones that aren't:
82679 postdown - never changed for virtual devices anyway.
82680 down - shouldn't change that without sending events.
82681
82682 memcpy'ing the struct also copied mapWidth, which means we didn't realloc
82683 during SetKeySymsMap lateron, overwriting the memory assigned to us.
82684
82685 X.Org Bug 16167 <http://bugs.freedesktop.org/show_bug.cgi?id=16167>
82686
82687commit 18cdd733ad3dd36fa0a23dfbf18c7d99b86c0276
82688Author: Peter Hutterer <peter.hutterer@redhat.com>
82689Date: Thu Oct 9 17:26:00 2008 +1030
82690
82691 kdrive: fix dixflags so they're always defined before enqueuing events #17734
82692
82693 If absolute events were posted, dixflags got set conditionally on whether the
82694 valuators are different from the last posted set of values.
82695 If dixflags are undefined however, the DIX interprets them as relative
82696 valuators. Fix this by making sure defining dixflags is always defined.
82697
82698 X.Org Bug 17724 <http://bugs.freedesktop.org/show_bug.cgi?id=17734>
82699
82700commit d58f2c30c64e8b2d8179dac06e0a54be06099da6
82701Author: Peter Hutterer <peter.hutterer@redhat.com>
82702Date: Thu Oct 9 16:50:54 2008 +1030
82703
82704 kdrive: lastx/y needs to be set to the event coordinates #17728
82705
82706 X.Org Bug 17728 <http://bugs.freedesktop.org/show_bug.cgi?id=17728>
82707
82708commit decec14219bcd992ec426e202ff3c8681b520b74
82709Author: Peter Hutterer <peter.hutterer@redhat.com>
82710Date: Thu Oct 9 15:10:21 2008 +1030
82711
82712 dix: silence "unused variable" compiler warning.
82713
82714commit 0b4b683f3eeec43536e73be302a4c396fe4d9894
82715Author: Peter Hutterer <peter.hutterer@redhat.com>
82716Date: Tue Oct 7 18:44:42 2008 +1030
82717
82718 xfree86: export NewInputDeviceRequest/DeleteInputDeviceRequest
82719
82720 NIDR should be used to create a new SD from e.g. within a driver.
82721 DIDR should be used to remove a device from the server.
82722
82723 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
82724
82725commit 5e213fe1022bf96747834a3ac227929b23136ee6
82726Author: Peter Hutterer <peter.hutterer@redhat.com>
82727Date: Thu Oct 9 11:29:11 2008 +1030
82728
82729 xfree86: cache dev->isMaster before freeing the device.
82730
82731commit 05cdef18ca668ffc86b71033f71cfbd262cdd994
82732Author: Zhenyu Wang <zhenyu.z.wang@intel.com>
82733Date: Fri Oct 10 13:45:51 2008 +0800
82734
82735 Don't leak EDID_block
82736
82737commit 2d9da7a5f384d5f38b2be79b1ea0df5a3deb52d1
82738Author: Luc Verhaegen <libv@skynet.be>
82739Date: Thu Oct 9 22:22:53 2008 +0200
82740
82741 DGA: Track client state even when using old style DGA.
82742
82743 This fixes the issue that a badly killed DGA will keep on hogging
82744 mode/framebuffer/mouse/keyboard.
82745
82746commit 322335d5b5b6f155f56fe3c1cbe372f13dc20932
82747Author: Luc Verhaegen <libv@skynet.be>
82748Date: Thu Oct 9 22:21:05 2008 +0200
82749
82750 DGA: Mash together xf86dga.c and xf86dga2.c.
82751
82752 This in preparation for an upcoming client state tracking fix.
82753
82754commit d13c3cbd43bc5e6b459c7df822292cf02ad2c9c4
82755Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82756Date: Wed Oct 8 18:39:41 2008 -0700
82757
82758 XQuartz: Some motion made towards supporting fullscreen.
82759 (cherry picked from commit 99be3d68b64059caada739a373e5e01844c776e0)
82760
82761commit dc166bf6423f4243ea05f17340ff9ff99ecad74b
82762Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82763Date: Tue Oct 7 11:39:08 2008 -0700
82764
82765 XQuartz: Fixed EXTRA_DIST for new GL files.
82766 (cherry picked from commit a11c9052030b44045f037050a5825256b539ed50)
82767
82768commit 803327929445256e159c10894e6305fbaa389982
82769Author: George Staplin <georgestaplin@george-staplins-mac-pro.local>
82770Date: Tue Oct 7 12:12:38 2008 -0600
82771
82772 XQuartz: GL: setVisualConfigs() copied the behavior of indirect.c which is wrong, with regard to the handling of xcalloc failure.
82773
82774 Use ErrorF for an error message after an xcalloc failure, and return instead of falling through to GlxSetVisualConfigs, or abort()ing at the test branch.
82775 (cherry picked from commit 1056700971fd5c034396ed6dbea15e092f0c6332)
82776
82777commit f1fec6f323b27a0165325d8fcf6af7ae163bde14
82778Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82779Date: Wed Oct 8 18:45:04 2008 -0700
82780
82781 XQuartz: Call setVisualConfigs in InitOutput, and only when GLXEXT is defined.
82782 (cherry picked from commit 97173d4eda142c73bb975cc05225b791778f85af)
82783
82784commit 52accec7dbc613a521e40449b8bb5f0c92ddb105
82785Author: George Staplin <georgestaplin@george-staplins-mac-pro.local>
82786Date: Tue Oct 7 11:31:48 2008 -0600
82787
82788 XQuartz: GL: Add the proper license information to the new files, and add more commentary to setVisualConfigs().
82789 (cherry picked from commit 190a3d5e5de9915931928fb8e6da88bb9644cda4)
82790
82791commit 63abdf61151cc90f769ed1708ee19b6c434c27c9
82792Author: George Staplin <georgestaplin@george-staplins-mac-pro.local>
82793Date: Mon Oct 6 18:05:12 2008 -0600
82794
82795 XQuartz: GL: Add the new C code to the Makefile.am.
82796
82797 Commit the darwin.c changes I missed in the last commit, for calling
82798 setVisualConfigs().
82799 (cherry picked from commit eb3c014e1710bf0b93bda10ddb9b795cd150d02d)
82800
82801commit a8f5d422c9c0a39f55e80bbd180439b6ec3a805c
82802Author: George Staplin <georgestaplin@george-staplins-mac-pro.local>
82803Date: Mon Oct 6 18:01:23 2008 -0600
82804
82805 XQuartz: GL: Provide code for getting the capabilities of the underlying system's CGL.
82806
82807 Add a setVisualConfigs that is called before the fbScreenInit, to setup the __GLXvisualConfigs.
82808 (cherry picked from commit fc86f9e4482043eca76d9d7a96e166be1aabf674)
82809
82810commit 2998e48be343ab2a11d6d328fc961ab5b8eb9292
82811Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82812Date: Sat Oct 4 21:25:50 2008 -0700
82813
82814 Xquartz: Fix parallel builds
82815 (cherry picked from commit 1fe7298a3b9a2ed82c612292f9e547cf78f99ae5)
82816
82817commit 7aa535708f1f33173bf05abc17b75e0841ffc385
82818Author: Adam Jackson <ajax@redhat.com>
82819Date: Wed Oct 8 16:25:59 2008 -0400
82820
82821 Fix more xcalloc stupidity.
82822
82823 Did I mention the hate? Pretty sure I mentioned the hate.
82824
82825commit 4fe80aa14a24df6d0611c47fc5654c3c4f56fdb1
82826Author: Adam Jackson <ajax@redhat.com>
82827Date: Wed Oct 8 16:22:35 2008 -0400
82828
82829 I hate you xlib. Please die in a fire.
82830
82831commit 2c5332872f808e7a646a63ae7ec63a2b80f83ee7
82832Author: Adam Jackson <ajax@redhat.com>
82833Date: Wed Oct 8 13:39:49 2008 -0400
82834
82835 s/kdDontZap/kdAllowZap/
82836
82837commit 1feb69eb63e6739ff5db255ad529e84adf941a10
82838Author: Luc Verhaegen <libv@skynet.be>
82839Date: Wed Oct 8 14:55:29 2008 +0200
82840
82841 DGA: Fix ProcXF86DGASetViewPort for missing support in driver.
82842
82843 Fixes a segfault when trying to activate a DGA mode without checking
82844 whether DGA modesetting is at all possible.
82845
82846commit d1bb5e3b47ab71aa2f18f69daaf2a4c5ce06d800
82847Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
82848Date: Wed Oct 8 03:19:45 2008 -0300
82849
82850 mi: functions here are carrying 'time' without need.
82851
82852commit 4964a9ca6c3f26376469b6d8217815a47231f0a9
82853Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
82854Date: Wed Oct 8 03:06:32 2008 -0300
82855
82856 mi: remove deprecated and unused functions.
82857
82858 miPointerDeltaCursor, miPointerGetMotionBufferSize and
82859 miPointerSetNewScreen.
82860
82861commit adf3e296e2b00164fb22e89ad4f44acbd2e3deae
82862Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
82863Date: Wed Oct 8 02:12:49 2008 -0300
82864
82865 Remove unused declarations.
82866
82867 AvailableClientInput (LBX debris) and xf86XinputFinalizeInit.
82868
82869commit bca9fa2efe97d2816a2e9412cdff08a0145c8f96
82870Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
82871Date: Wed Oct 8 01:38:27 2008 -0300
82872
82873 Everyone knows that the loader is present. Silence it.
82874
82875commit 56c615368c5a8e7acb0398434c2c68578626aa38
82876Author: Zhenyu Wang <zhenyu.z.wang@intel.com>
82877Date: Wed Oct 8 13:33:55 2008 +0800
82878
82879 Check nextEnabledOutput()'s return in bestModeForAspect()
82880
82881 In case no enabled outputs, we will reference wrong index of
82882 output array.
82883
82884commit 278c11f01fbc6d6bd91c5a7127928c9ef5d29fca
82885Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
82886Date: Wed Oct 8 01:18:49 2008 -0300
82887
82888 xfree86: xf86{Enable, Disable}InputHandler can be static.
82889
82890commit 50081d2dfb79878cb931a15c265f0d60698dfd39
82891Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
82892Date: Wed Oct 8 01:11:38 2008 -0300
82893
82894 xfree86: remove xf86{Enable, Disable}GeneralHandler and others unused.
82895
82896commit 54294e15ecf4a8ec0fd9fa8b6ed1a1b2d912ec7e
82897Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
82898Date: Wed Oct 8 00:43:22 2008 -0300
82899
82900 xfree86: fix wrong function name message errors.
82901
82902commit 57dbb90f198116dcaf17c8111a7f80644d7ebe7f
82903Author: Adam Jackson <ajax@redhat.com>
82904Date: Tue Oct 7 20:36:45 2008 -0400
82905
82906 Document zap/-retro interaction more.
82907
82908commit 617933e8c259ad12388c0dd13421846a873975e3
82909Author: Daniel Stone <daniel@fooishbar.org>
82910Date: Wed Oct 8 02:45:59 2008 +0300
82911
82912 XFree86: Fix DontZap interaction with -retro
82913
82914 -retro will now _not_ override (No)DontZap, if it's set in the config
82915 file.
82916
82917commit d80bae2237e555025465d4d761a5cc537cc2bcdd
82918Author: Daniel Stone <daniel@fooishbar.org>
82919Date: Wed Oct 8 02:43:27 2008 +0300
82920
82921 Update docs to reflect new DontZap default
82922
82923commit 42e7b64bf90604e0099702f5447163be80a2efb6
82924Author: Adam Jackson <ajax@redhat.com>
82925Date: Tue Oct 7 20:32:02 2008 -0400
82926
82927 -retro enables zapping
82928
82929commit 9d135ac10a7374c7ccda705f1eeb02cc53076c34
82930Author: Daniel Stone <daniel@fooishbar.org>
82931Date: Wed Oct 8 02:30:05 2008 +0300
82932
82933 Input: Make DontZap the default
82934
82935 If you need to bail out the server, use Ctrl-Alt-Fx, or enable zapping
82936 if it bothers you that much. If Ctrl-Alt-Fx is broken, nag me until
82937 it's permanently fixed.
82938
82939commit b574365fca954a999ebaeb7203d1e0ea6d99b3ea
82940Author: Aaron Plattner <aplattner@nvidia.com>
82941Date: Tue Oct 7 13:09:11 2008 -0700
82942
82943 Revert "Unexport (and unimplement) xf86SetAccessFuncs" for now.
82944
82945 The nvidia driver currently uses these hooks to work around problems where RAC
82946 will disable access to the hardware at unexpected times. This change restores
82947 these hooks until we can come up with a better API for working around RAC.
82948
82949 This reverts commit c1df4fbede8058c15ce3a5759a7758fecafbb9e7.
82950
82951commit fe85ec34ec2b58188633dac4464a1e9e9977652b
82952Author: Aaron Plattner <aplattner@nvidia.com>
82953Date: Tue Oct 7 13:08:38 2008 -0700
82954
82955 Revert "Remove unused server state change callbackery" for now.
82956
82957 The nvidia driver currently uses these callbacks to work around problems where
82958 RAC will disable access to the hardware at unexpected times. This change
82959 restores these hooks until we can come up with a better API for working around
82960 RAC.
82961
82962 This reverts commit d7c0ba2e9eae7044ef4c31d9c3cbb9a71ee0f9f9.
82963
82964 Conflicts:
82965
82966 hw/xfree86/loader/xf86sym.c
82967
82968commit 94919480d8bb66e1807b4fe87b8f326ef6e012c6
82969Author: Adam Jackson <ajax@redhat.com>
82970Date: Tue Oct 7 13:41:25 2008 -0400
82971
82972 int10: Fix a nasty memory leak.
82973
82974commit a57b2f172c1291f22f7ba2780c1b2f55e353c3e9
82975Author: Adam Jackson <ajax@redhat.com>
82976Date: Tue Oct 7 13:39:10 2008 -0400
82977
82978 int10: Don't warn when scanning for devices we don't have.
82979
82980 Some BIOSes (hi XGI!) will attempt to enumerate the PCI bus by asking
82981 for the config space of every possible device number. This despite
82982 perfectly functional BIOS methods to enumerate the bus exactly.
82983
82984commit a65e36a873cd1ba9896cd0f9a3e94dd933666005
82985Author: Adam Jackson <ajax@redhat.com>
82986Date: Tue Oct 7 13:38:12 2008 -0400
82987
82988 int10: Remove useless check.
82989
82990 If you have more than one PCI device with the same d/b/d/f, you're
82991 already in trouble.
82992
82993commit 6dbfca571f6d214dc456710a60ebbbbd204cdbbd
82994Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
82995Date: Tue Oct 7 10:07:03 2008 -0700
82996
82997 Depend on xtrans >= 1.2.2 for TRANS_NOXAUTH
82998
82999commit 30193fbff9d9284fbbd81fe5f7b6c4d102daf1d1
83000Author: Adam Jackson <ajax@redhat.com>
83001Date: Tue Oct 7 10:36:12 2008 -0400
83002
83003 Reinstate cursorScreenDevPriv
83004
83005commit e7cd5d84ed1fb3637dce6e9e6ef58c8611818c4e
83006Author: Dave Airlie <airlied@redhat.com>
83007Date: Tue Oct 7 16:09:17 2008 +1000
83008
83009 fixup another calloc build issue
83010
83011commit 60cfaa45a532bd3cfcfb4bd57b2f657c0a8b6a6b
83012Author: Adam Jackson <ajax@redhat.com>
83013Date: Mon Oct 6 18:47:06 2008 -0400
83014
83015 Build fix.
83016
83017commit c6d6d3e87a472b641d883a6c000f96e5bdf48532
83018Author: Adam Jackson <ajax@redhat.com>
83019Date: Mon Oct 6 16:18:21 2008 -0400
83020
83021 Loader: Unexport a handful of consumerless symbols from dixsym.
83022
83023commit 8a5b89e8e184f4cbf33c6dee0b06e61d8f857576
83024Author: Adam Jackson <ajax@redhat.com>
83025Date: Mon Oct 6 15:36:51 2008 -0400
83026
83027 xalloc+memset(0) -> xcalloc
83028
83029commit 0b7b89fbac0b3865b2cf51295c68a5f4c7523f28
83030Author: Adam Jackson <ajax@redhat.com>
83031Date: Mon Oct 6 14:19:30 2008 -0400
83032
83033 xalloc+bzero -> xcalloc
83034
83035commit 9187f6ad9ec7ba9569a93d92561aac17eaa83491
83036Author: Keith Packard <keithp@keithp.com>
83037Date: Mon Oct 6 12:21:20 2008 -0400
83038
83039 RANDR: Delivery output property events.
83040
83041commit 717c7492a0f6ba3fb3eabda33515881eef314155
83042Author: Dave Airlie <airlied@linux.ie>
83043Date: Tue Oct 7 02:38:44 2008 +1000
83044
83045 exa: don't call composite routines with no buffer.
83046
83047 We can get a case with gnome-terminal + links, where we get two arrays
83048 of glyphs all with 0 width and 0 heights in them. If this happens
83049 we manage to get to this case without any buffer setup and segfault.
83050
83051commit aacac3edac7c6b82a0a0e3bc500f563d187e5df3
83052Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83053Date: Sat Oct 4 21:07:31 2008 -0700
83054
83055 XQuartz: Bumped bundle version to 2.3.2 for beta1
83056 (cherry picked from commit 2021c4c4e726a6b8f27f19f16ed3d86e37dbe448)
83057
83058commit ed3134918dcbbd58a5ef8359a77e8024c98396e6
83059Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83060Date: Sat Oct 4 21:06:58 2008 -0700
83061
83062 XQuartz: pbcopy: Re-enabled some code I commented out for testing
83063 (cherry picked from commit 9b5d11409264080e394c8f18e06237122de80014)
83064
83065commit b3a971d8b5df5ab4af0be49105ef32f67d010901
83066Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83067Date: Sat Oct 4 18:54:15 2008 -0700
83068
83069 XQuartz: xpbproxy: Support some of the preference toggles from X11.app, cleaned up CLIPBOARD_MANAGER atom management.
83070 (cherry picked from commit f7673bb4de3c1f71eb390a3279eed3589efc3df4)
83071
83072commit fc35c7a2e2888cb767413f5e85a79c7f422a6284
83073Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83074Date: Sat Oct 4 21:16:54 2008 -0700
83075
83076 XQuartz: GLX - enable_stereo defaults option should work now
83077 (cherry picked from commit 3a6898f48756fdb2898292679301ef3dc109b4f6)
83078
83079commit 7818aba1d1e74f4ffe68b7d6806622b94ae8a230
83080Author: Roland Scheidegger <sroland@tungstengraphics.com>
83081Date: Sat Oct 4 15:50:35 2008 +0200
83082
83083 dri2: fix compilation (remove no longer existing include file)
83084
83085commit a29af6465883796a6acf04b72ce812a0632b2186
83086Author: Maarten Maathuis <madman2003@gmail.com>
83087Date: Fri Oct 3 23:54:10 2008 +0200
83088
83089 exa: remove "direct" case for exa{Trapezoids,Triangles}
83090
83091 - By adding a small hack to the xserver i was able to easily test the performance of the normally rare direct case (using cairo).
83092 - It turned out to be 70% slower for me (large test on an otherwise idle computer), which seems enough of a reason to remove it.
83093 - AddTraps could also use a 2nd look, but since noone is using that it's a bit hard and less useful to test.
83094
83095commit 416870d8c9e7e383b87c8513c2296a0cb66e7ef4
83096Author: Maarten Maathuis <madman2003@gmail.com>
83097Date: Fri Oct 3 22:59:40 2008 +0200
83098
83099 exa: remove some excessive whitespace
83100
83101commit b72a217ad59cc527327d7b1fc81898677b977d8f
83102Author: Adam Jackson <ajax@redhat.com>
83103Date: Fri Oct 3 18:30:37 2008 -0400
83104
83105 Dangling use of miDestroyGCOps
83106
83107commit f5bbc54a93f53cd4ba124712159bb44c93282a8e
83108Author: Adam Jackson <ajax@redhat.com>
83109Date: Fri Oct 3 18:18:02 2008 -0400
83110
83111 Fix compile for miSetScreenPixmap prototype.
83112
83113commit 9dbfcd89214241626ac2704d1ffffff1cc0c67ae
83114Author: Adam Jackson <ajax@redhat.com>
83115Date: Fri Oct 3 17:40:30 2008 -0400
83116
83117 Delete unused miModifyBanking()
83118
83119commit 94825ad3c72a1c3f6a61199c302841f63241cf87
83120Author: Adam Jackson <ajax@redhat.com>
83121Date: Fri Oct 3 17:39:34 2008 -0400
83122
83123 Remove unused wrapping for visual initialization.
83124
83125commit 4791f8e2ba9a0e318b7f13c83618ece036dd53ed
83126Author: Adam Jackson <ajax@redhat.com>
83127Date: Fri Oct 3 17:23:58 2008 -0400
83128
83129 staticize mi{G,S}etScreenPixmap
83130
83131commit 14d98b4a2fdea4a4f5cc9b751151672cb392fba3
83132Author: Adam Jackson <ajax@redhat.com>
83133Date: Fri Oct 3 17:21:16 2008 -0400
83134
83135 miFindMaxBand is dead code
83136
83137commit 48b47928133bc462947104bd0f4204f00b993862
83138Author: Adam Jackson <ajax@redhat.com>
83139Date: Fri Oct 3 17:16:28 2008 -0400
83140
83141 mi{Create,Destroy}GCOps are unused, nuke 'em.
83142
83143commit bd0e87c74789bd5447ec731aa2d2d6e05dd54abc
83144Author: Adam Jackson <ajax@redhat.com>
83145Date: Fri Oct 3 17:13:10 2008 -0400
83146
83147 Remove a bunch of unused API from the mi export list.
83148
83149commit ee5e07bc22f0c3af7a412b6bc430eae6a1dafe16
83150Author: Adam Jackson <ajax@redhat.com>
83151Date: Fri Oct 3 16:58:01 2008 -0400
83152
83153 loader: sort the mi symbol list
83154
83155commit 77d0b0ecfa3c67ff8d11e773a7f083721cf4c6d7
83156Author: Adam Jackson <ajax@redhat.com>
83157Date: Fri Oct 3 16:53:08 2008 -0400
83158
83159 Remove unused pciNumBuses
83160
83161commit 0ce61e21d6d7dcca0090e319bbcdb678570f2c3f
83162Author: Adam Jackson <ajax@redhat.com>
83163Date: Fri Oct 3 16:05:19 2008 -0400
83164
83165 Remove the Must_have_memory hack.
83166
83167 Also remove an astonishing amount of misunderstanding of how casts work.
83168
83169commit e6b1c1fada19268af559d89375989973729fdb19
83170Author: Adam Jackson <ajax@redhat.com>
83171Date: Fri Oct 3 15:47:50 2008 -0400
83172
83173 const cleanup
83174
83175commit fe616f9230b672edb89fca7516e750dc1b20c690
83176Author: Adam Jackson <ajax@redhat.com>
83177Date: Fri Oct 3 15:42:58 2008 -0400
83178
83179 Drop the never-called SkippedRequestCallback list.
83180
83181commit 31136b9467c2a17bbed4a51d6a6c41e3290bd236
83182Author: Adam Jackson <ajax@redhat.com>
83183Date: Fri Oct 3 15:33:47 2008 -0400
83184
83185 Unexport MakeClientGrab{Imp,P}ervious
83186
83187 xtrap used to use these, but it's gone now. xtest still does but it's
83188 linked statically.
83189
83190commit eaf3fdf97014c27aececa8505c2713e7c9a537fc
83191Author: Adam Jackson <ajax@redhat.com>
83192Date: Fri Oct 3 15:24:35 2008 -0400
83193
83194 Unexport xf86NoSharedResources
83195
83196commit d7d9edb506bfea1c10516b6c4eb72dee3559863c
83197Author: Adam Jackson <ajax@redhat.com>
83198Date: Fri Oct 3 15:17:44 2008 -0400
83199
83200 Unexport xf86DeallocateResourcesForEntity
83201
83202commit 49673b43584928c22d699d1968ed6813b272aefe
83203Author: Adam Jackson <ajax@redhat.com>
83204Date: Fri Oct 3 15:13:39 2008 -0400
83205
83206 Unexport xf86SetEntityFuncs
83207
83208commit 949ef6b79a36d6f6bbc192c20ef0f13cc95d4f28
83209Author: Adam Jackson <ajax@redhat.com>
83210Date: Fri Oct 3 15:10:26 2008 -0400
83211
83212 Unexport xf86ChkConflict
83213
83214 Also remove the unused internal ChkConflict()
83215
83216commit d7c0ba2e9eae7044ef4c31d9c3cbb9a71ee0f9f9
83217Author: Adam Jackson <ajax@redhat.com>
83218Date: Fri Oct 3 15:06:42 2008 -0400
83219
83220 Remove unused server state change callbackery.
83221
83222commit 9ada146a16bcf220b2dd2936fb3a0349a31e99d3
83223Author: Adam Jackson <ajax@redhat.com>
83224Date: Fri Oct 3 14:40:18 2008 -0400
83225
83226 xf86RegisterResources: Always print any failed registrations.
83227
83228commit a5d3fa90f3de763a5dcbb0ff4987562449819323
83229Author: Adam Jackson <ajax@redhat.com>
83230Date: Fri Oct 3 14:38:48 2008 -0400
83231
83232 Typo fix.
83233
83234commit cc3335858508ffbc52123eb5bb0c34f42800c8c4
83235Author: Adam Jackson <ajax@redhat.com>
83236Date: Fri Oct 3 14:37:59 2008 -0400
83237
83238 Remove the old implementation of RemoveOverlaps
83239
83240commit 8fbf1185f5d6a1836b3e4265de6f605d82478e0b
83241Author: Adam Jackson <ajax@redhat.com>
83242Date: Fri Oct 3 14:32:32 2008 -0400
83243
83244 Internalise some resource list API.
83245
83246commit 0ca4f3cee7e42a393cdd32d3a772a190ae68abc5
83247Author: Adam Jackson <ajax@redhat.com>
83248Date: Fri Oct 3 13:29:55 2008 -0400
83249
83250 Unexport xf86EnterServerState
83251
83252commit c1df4fbede8058c15ce3a5759a7758fecafbb9e7
83253Author: Adam Jackson <ajax@redhat.com>
83254Date: Fri Oct 3 13:08:43 2008 -0400
83255
83256 Unexport (and unimplement) xf86SetAccessFuncs.
83257
83258 Because, no.
83259
83260commit 44269e604eb60ebaa6189fd4409d39a4ec6b9f4b
83261Author: Adam Jackson <ajax@redhat.com>
83262Date: Fri Oct 3 13:02:50 2008 -0400
83263
83264 RAC: NO_SEPARATE_*_FROM_* are useless.
83265
83266 If anyone can come up with an example of a bus where:
83267
83268 - both i/o and memory resources are addressable
83269 - access to them can be controlled
83270 - but they can't be controlled independently
83271
83272 then by all means, reinstate this logic.
83273
83274commit 51fcb58f45824c7cde88b8b4ccbd6bdcf760280e
83275Author: Adam Jackson <ajax@redhat.com>
83276Date: Fri Oct 3 12:39:49 2008 -0400
83277
83278 RAC: AccessEnable() and AccessDisable() are mandatory.
83279
83280commit 9c5fc7762e9ca09847965e67e247d95dfc197266
83281Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83282Date: Fri Oct 3 11:24:34 2008 -0700
83283
83284 Partially reverts commit 505c216d030a8eb6e9fa757ecebed9597a850c88.
83285
83286 I don't know how the hell git-cherrypick messed that up to delete fbwindow.c
83287
83288commit 7bb73a9513710feaebc127998950e9f472bfcc0d
83289Author: George Peter Staplin <gps@Georges-Workstation.local>
83290Date: Tue Sep 30 23:53:12 2008 -0600
83291
83292 XQuartz: pbproxy: Possibly fix a memory leak by using an [NSApp run] loop,
83293 instead of calling CFRunLoopRun() directly. The leak wasn't reproducible on
83294 this machine, but someone was able to produce a leak trace with Instruments
83295 that indicates it was leaking in the CFRunLoopRun() path.
83296
83297 x-input.m: dequeue and ignore events when pbproxy_active is false.
83298
83299 x-selection.h: add an is_active method that is used by x-input.m to ignore
83300 events.
83301
83302 x-selection.m: Handle nearly every preference, except for primary_on_grab,
83303 which I don't really understand yet.
83304 (cherry picked from commit 4d51ad851e64da83cbdfb0a4a22428418a7bcf75)
83305
83306commit 2a5ce41f0371ad5df52586b8d4072578b6206321
83307Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83308Date: Tue Sep 30 13:54:08 2008 -0700
83309
83310 XQuartz: Use "pointer" and "pen" for device names to please GDK.
83311 (cherry picked from commit 00ca0f4d839abf47e9573a1552473e039cf787e6)
83312
83313commit d3f597a986dcbc7e4164f350300dbd2e438949d8
83314Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83315Date: Tue Sep 30 11:10:58 2008 -0700
83316
83317 XQuartz: Workaround for initial pressure/tilt being sent as 0 with motion during the proximity event
83318 (cherry picked from commit 9a91d770a6411dd876187e64a8bda1f0745169ae)
83319
83320commit 1657dfb843c0127ea31bfd0a8ef81d0c394cbc7b
83321Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83322Date: Tue Sep 30 08:46:08 2008 -0700
83323
83324 XQuartz: Using absolute ranges for pointer location to increase resolution and better support tablets.
83325 (cherry picked from commit d79ccb45f68b65d65718b5b77efe2fae6eeda762)
83326
83327commit 88bb8c6f485a7aaf40f684d484a33cfedfa74a70
83328Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83329Date: Mon Sep 29 22:33:02 2008 -0700
83330
83331 XQuartz: Removed resolved comment about localization.
83332 (cherry picked from commit 8ed5faf058ac7b0782a9cc13a2c58b80168358d2)
83333
83334commit b1d0b196617643619d24ed520b0a1161e26b66c6
83335Author: George Peter Staplin <gps@Georges-Workstation.local>
83336Date: Tue Sep 30 13:04:35 2008 -0600
83337
83338 XQuartz: pbproxy: Make the signal handler safer by using _exit, and only exit.
83339
83340 Remove some unnecesssary headers.
83341
83342 Remove some dead code that was never called or used in pbproxy.
83343
83344 Make use of an NSAutoreleasePool in x_init. It could potentially cause a leak
83345 on a startup without this.
83346
83347 Start adding reload_preferences to the x_selection class, as well as event
83348 handling for that.
83349 (cherry picked from commit 602e8ba8f7ee196696bc9e3cea6ecdf3200dcf5c)
83350
83351commit ecd260143cf9f012f033ba029ad04cb49567c89a
83352Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83353Date: Mon Sep 29 22:32:24 2008 -0700
83354
83355 XQuartz: Added some localizable strings that were missing.
83356 (cherry picked from commit 2c8205ea8b709c5859412b466e83aec7a94acdcf)
83357
83358commit aae878b89081ffba16386e8d4987469313049bea
83359Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83360Date: Mon Sep 29 22:26:20 2008 -0700
83361
83362 XQuartz: Actually send the pointer event from the pen even if it happes to actually have 0 pressure and 0 tilt.
83363 (cherry picked from commit f41583761955cb9c92c43239bfaa8ae0e5d95e33)
83364
83365commit b7ad86ff8a9115147c0cf02af9e80ec464129392
83366Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83367Date: Sat Sep 27 15:30:10 2008 -0700
83368
83369 Xquartz: xpbproxy: Split out app-specific stuff to app-main.m in prep for making this into a thread.
83370 (cherry picked from commit c2012138a520560f8a2160518ea73fced410c3b7)
83371
83372commit 5446adebfb085dd05471729bf5c55bf820b1cbe3
83373Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83374Date: Fri Oct 3 11:12:49 2008 -0700
83375
83376 XQuartz: Send AppleWMReloadPreferences where appropriate.
83377 (cherry picked from commit 490cbe9888e2c1080495b003c429fdb1659444e1)
83378
83379commit 6020c3fbfe06530cb2b90178d17a7d63beae028a
83380Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83381Date: Fri Oct 3 11:09:26 2008 -0700
83382
83383 XQuartz: Added a --enable-integrated-xpbproxy configure option for building xpbproxy as an app or as a thread.
83384 (cherry picked from commit 8edc5fb38c922f28659d2f823148339a8907c4d9)
83385 (cherry picked from commit 88033a66a5549870fd053795b019d4c22950425b)
83386
83387commit 505c216d030a8eb6e9fa757ecebed9597a850c88
83388Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83389Date: Fri Oct 3 11:11:37 2008 -0700
83390
83391 Added dep on applewmproto 1.1.0
83392 (cherry picked from commit b350b066740a2e72585153e35478538cf1bca9ef)
83393
83394commit 1c1a6e965da412e767ad8bc15a5362116d64c6b0
83395Author: Adam Jackson <ajax@redhat.com>
83396Date: Mon Sep 22 22:09:19 2008 -0400
83397
83398 Remove an utterly, utterly ancient changelog.
83399
83400 Wow.
83401
83402commit 9c8a2be2c73abf06245b1eb7f05f93e104dcfe42
83403Author: Peter Hutterer <peter.hutterer@redhat.com>
83404Date: Thu Oct 2 08:55:14 2008 +0930
83405
83406 dix: fix axis scaling.
83407
83408 For two axes [a, b] and [x, y] (inclusive), the formula to scale point P(ab)
83409 to (x,y) is:
83410 (P - a)/(b - a) * (y - x) + x
83411
83412 And the whole end result rounded of course to get the integer we need.
83413
83414commit 93ef72fa26b27cf90bf5c64ec19ac295a113aaae
83415Author: Peter Hutterer <peter.hutterer@redhat.com>
83416Date: Sun Sep 28 21:47:21 2008 +0930
83417
83418 dix: don't lose subpixel precision when scaling.
83419
83420 Only scale x/y back from screen range if we have crossed screen, otherwise
83421 leave it as it is.
83422
83423commit 5ef5e89f0e4b9b8820093e072430e295efce32fa
83424Author: Simon Thum <simon.thum@gmx.de>
83425Date: Sun Sep 28 22:32:44 2008 +0930
83426
83427 dix: fix non-initialization of pointer accel in master devices.
83428
83429 Init MDs always with PtrAccelNoOp.
83430
83431 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
83432
83433commit 036d424827726b0c804ae1532109253270744207
83434Author: Peter Hutterer <peter.hutterer@redhat.com>
83435Date: Fri Sep 26 17:01:12 2008 +0930
83436
83437 xfree86: don't let DGA steal key events outside of [9,255].
83438
83439commit 56a24cf894a142b18c7f18e294074c32031f3d14
83440Author: Adam Jackson <ajax@redhat.com>
83441Date: Thu Oct 2 17:56:31 2008 -0400
83442
83443 xorg.conf.man: We don't look in root's ~ anymore, don't say we do.
83444
83445commit 7921c4a0bfc843c772dcb2fc185d4bcab1316330
83446Author: Adam Jackson <ajax@redhat.com>
83447Date: Thu Oct 2 17:08:48 2008 -0400
83448
83449 Remove some stale IPv6 debugging
83450
83451commit f836e1f11b7bf3449c3e469cc7799c4ed28d2356
83452Author: Adam Jackson <ajax@redhat.com>
83453Date: Thu Oct 2 17:03:54 2008 -0400
83454
83455 Unifdef ISC
83456
83457commit b1a31734731a768d586b4cb18317d4a5c5e20fa8
83458Author: Adam Jackson <ajax@redhat.com>
83459Date: Thu Oct 2 17:03:21 2008 -0400
83460
83461 Fix bad preprocessor line
83462
83463commit 00f01f39454001846869496d9124f62c01e5691a
83464Author: Adam Jackson <ajax@redhat.com>
83465Date: Thu Oct 2 15:48:18 2008 -0400
83466
83467 Don't inherit XAUTHORITY from the environment.
83468
83469 It's not especially obvious, and unpleasantly overloaded for the Xnest
83470 case. Typically this gives you a server that looks for its auth data in
83471 the authority file you were using for the running X session, which
83472 generally doesn't have an entry for the display you just started.
83473
83474 All the major dm's, and startx, pass -auth explicitly, so this shouldn't
83475 cause too much upheaval.
83476
83477commit eb8be3e90a9c90a428696026d1e3b2152d7eefb4
83478Author: George Peter Staplin <gps@Georges-Workstation.local>
83479Date: Thu Sep 25 14:16:20 2008 -0600
83480
83481 XQuartz: pbproxy: Fix another STRING bug.
83482
83483 Fix the usage of the NSString cStringUsingEncoding: - it doesn't NUL
83484 terminate the string, which lead to a bus error. So, we use
83485 lengthOfBytesUsingEncoding: to get the length in bytes instead of
83486 strlen().
83487 (cherry picked from commit 6333d619e747c3b6bd3ba7557e35c0e5f6daa40f)
83488
83489commit edfa3f8233fde490cb4255f21473684e96d595a3
83490Author: George Peter Staplin <gps@Georges-Workstation.local>
83491Date: Thu Sep 25 12:29:57 2008 -0600
83492
83493 XQuartz: pbproxy: Fix a bug that rxvt brought out in STRING copying. Now I can copy and paste to/from rxvt correctly.
83494 (cherry picked from commit 82a4dc5f0f31a7911ee1cce5d1b162077befa811)
83495
83496commit 12a59c44cb68843a60fc43257930d1cbeb971b7a
83497Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83498Date: Wed Sep 24 22:51:03 2008 -0700
83499
83500 XQuartz: Fixed threading issue with TSM.
83501 (cherry picked from commit 93ab4e0071670bb80bfa1170dd97ed9d6d51c67a)
83502
83503commit ee87c9b02401a7a08b396884ba412a503b078bbd
83504Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83505Date: Wed Sep 24 18:57:36 2008 -0700
83506
83507 XQuartz: pbproxy: Removed stray SPI include.
83508 (cherry picked from commit 2fb2bc77b804d7b63002c1423cb0bfc658eeee45)
83509
83510commit 591cc57627a9137159fb5a13287c1d61d8d2fc8c
83511Author: George Peter Staplin <gps@Georges-Workstation.local>
83512Date: Tue Sep 23 13:59:25 2008 -0600
83513
83514 XQuartz: pbproxy: Fix 3 paths that could leak memory.
83515
83516 2 of the paths leaked, when INCR transfers were done. Now we
83517 are leak free according to the leaks program for all transfers
83518 I have tried so far.
83519 (cherry picked from commit aa98db576bc02f0765cb35d0282a894ba3778213)
83520
83521commit 586fbc91df731150ff5e39170efe33f3cfac28a2
83522Author: George Peter Staplin <gps@Georges-Workstation.local>
83523Date: Tue Sep 23 12:39:32 2008 -0600
83524
83525 XQuartz: pbproxy: Fix NSObject memory leaks by properly using the
83526 NSAutoreleasePool. Now the usage is consistent. In x_input_run()
83527 we create a pool, and release it after processing the XEvents.
83528
83529 Add some getpid() output to main for debugging. It needs a bit more
83530 testing before the next release.
83531
83532 Don't retain the NSPasteboard as the old code did. That may have
83533 contributed to the leak, and it made it so that we needed the
83534 NSAutoreleasePool created in main().
83535
83536 Remove the _known_types, and _pasteboard instance variables from
83537 the x_selection class. They aren't needed anymore.
83538
83539 The leaks program now indicates 0 leaks after some usage. I want
83540 to test further, but this seems much better, and my memory usage
83541 graph indicates it's not growing.
83542 (cherry picked from commit b245d84a72ee3929546cd11a6eba3c60fb4a4d95)
83543
83544commit 238999cfc99dbf306184f0e846465a8707bcc9d2
83545Author: George Peter Staplin <gps@Georges-Workstation.local>
83546Date: Mon Sep 22 17:36:21 2008 -0600
83547
83548 XQuartz: pbproxy: Add a comment to pbproxy.h about how the DB()
83549 macro causes a leak (according to the leaks program).
83550
83551 Attempt to fix several other leaks with release method calls.
83552 For some reason the process still grows more than it should...
83553
83554 I will need to use some better methods than leaks, and malloc_history
83555 I suspect. Whatever is leaking, it's hard to find. I need to isolate
83556 the cases more.
83557
83558 Add a missing image/jpeg branch.
83559
83560 Remove read_prop_32 - it's not used.
83561 (cherry picked from commit 63a680354dcb545fef935ac97596dd35ceaed960)
83562
83563commit 106eb37e0214bbd864a59119b491abfc54f32629
83564Author: Peter Hutterer <peter.hutterer@redhat.com>
83565Date: Fri Sep 26 13:05:10 2008 +0930
83566
83567 Require inputproto 1.9.99.5
83568
83569 Pull and rebuild your input drivers, while you're at it.
83570
83571commit fbd09486c6e6c64db28d81caa5ef04821e49f2eb
83572Author: Peter Hutterer <peter.hutterer@redhat.com>
83573Date: Tue Sep 23 16:55:04 2008 +0930
83574
83575 Push server-known properties into xserver-properties.h.
83576
83577commit a2d83b9dc8387ec7e70689db1371cf6500b2f68e
83578Author: Peter Hutterer <peter.hutterer@redhat.com>
83579Date: Thu Sep 18 18:21:03 2008 +0930
83580
83581 Xi: add "deletable" flag to properties, add DeleteProperty handler.
83582
83583 A property can only be deleted if any of the following is true:
83584 - if a property is deletable and all handlers return Success.
83585 - if a property is non-deleteable and the all handlers return Success AND the
83586 delete request does not come from a client (i.e. driver or the server).
83587
83588 A client can never delete a non-deletable property.
83589
83590commit 22e90472681ebc7bdd28f82bbf2753a5b259e931
83591Author: Peter Hutterer <peter.hutterer@redhat.com>
83592Date: Thu Sep 18 16:12:02 2008 +0930
83593
83594 Xi: allow Set/GetProperties to return a status, and honour this status code.
83595
83596 If a property handler now bails out, return the error code to the caller. This
83597 allows to be slightly more specific with the errors.
83598
83599commit 1e24e7b9df3d02350c7ea18e9379e87fe4d00026
83600Author: Peter Hutterer <peter.hutterer@redhat.com>
83601Date: Thu Sep 18 15:58:46 2008 +0930
83602
83603 Xi: remove configure/query device property calls.
83604
83605 This removes all the meta-information about device properties (pending,
83606 fromClient, range, valid_values, immutable).
83607
83608commit feaa5fa6712c8c6f4ca97766e2ac0338253cf3b8
83609Author: Peter Hutterer <peter.hutterer@redhat.com>
83610Date: Thu Sep 25 11:03:22 2008 +0930
83611
83612 xfixes: realize the invisible cursor before displaying it.
83613
83614 AllocARGBCursor realizes the cursor but can only do so if we have devices
83615 there already. If we don't - then the cursor needs to be realized elsewhere.
83616 This is usually done in InitializeSprite, but since xfixes just hard-swaps the
83617 (realized) cursor to the InvisibleCursor, we need to manually realize it
83618 before trying to display it.
83619
83620commit e58be0f3425fb3e946a222077672c3c01308f887
83621Author: Peter Hutterer <peter.hutterer@redhat.com>
83622Date: Tue Sep 23 10:01:30 2008 +0930
83623
83624 config: print error code if NIDR fails.
83625
83626commit 30c3c13f1030268aaa6a3598d538fafd0592d77a
83627Author: Peter Hutterer <peter.hutterer@redhat.com>
83628Date: Mon Sep 22 11:10:46 2008 +0930
83629
83630 xkb: squash canonical types into explicit ones on core reconstruction.
83631
83632 If we update key types from core, and groups 2 - n have a canonical type but
83633 the same symbols as the explicit type of group 1, assume that it was a core
83634 sym duplication according to Section 12.4 of the XKB Protocol Spec.
83635 Ignore the canonical types and pretend there's only one group for the key -
83636 with the explicit key type.
83637
83638 The protocol spec does not cover this case, so we have to guess here.
83639
83640commit ae986d1c73d2f720bd0309d8c33328d14e8eed25
83641Author: Peter Hutterer <peter.hutterer@redhat.com>
83642Date: Fri Sep 19 18:27:24 2008 +0930
83643
83644 xkb: fix core keyboard map generation. #14373
83645
83646 According to Section 12.4 of the XKB Protocol Spec, if a key only has a single
83647 group but the keyboard has multiple groups defined, the core description of
83648 the key is a duplication of the single group across all symbols. i.e.
83649 G1L1 G1L2 G1L1 G1L2 G1L3 G1L4 G1L3 G1L4
83650
83651 The previous code generated G1L1 G1L2 G1L3 G1L4 G1L3 G1L4, leading to
83652 "invented" groups when the process is reversed.
83653
83654 Note that this creates wrong key types on reconstruction from core to xkb,
83655 i.e. any single-group key with a key type that is not one of the canonical
83656 four (Sec 12.2.3), will get the assigned type on group 1, and a canonical type
83657 for the other gruops.
83658
83659 X.Org Bug 14373 <http://bugs.freedesktop.org/show_bug.cgi?id=14373>
83660
83661commit 5bf3f0fd4e00f96cfebd14b58580c3c8733626fb
83662Author: Dave Airlie <airlied@linux.ie>
83663Date: Thu Sep 25 15:40:19 2008 +1000
83664
83665 exa: make sure pixmap devPrivate.ptr is NULL at create time
83666
83667commit 82f150d73cc9a7d8eaf7241a51b03af05bcec159
83668Author: Ian Romanick <ian.d.romanick@intel.com>
83669Date: Wed Sep 24 16:56:45 2008 -0700
83670
83671 GLX: Make sure GLX_SGI_make_current_read is enabled when possible
83672
83673commit 8d9defe8c2a685709318c1d43379443df3d2322a
83674Author: Adam Jackson <ajax@redhat.com>
83675Date: Wed Sep 24 10:11:00 2008 -0400
83676
83677 Fix un-suppressing the initial cursor.
83678
83679 The intended behaviour was "show as soon as someone calls
83680 XDefineCursor". What you actually got was, uh, slightly less well
83681 defined, since the screen's ChangeWindowAttributes hook would run after
83682 DIX handled the cursor change. Oops.
83683
83684 The trivial way to turn the cursor on is:
83685 % xsetroot -cursor_name gumby
83686
83687 Refer to /usr/include/X11/cursorfont.h for cursor names.
83688
83689 Thanks to anholt for catching this.
83690
83691commit a39377cbcbd3091095efbeab25bec18ae520147e
83692Author: Keith Packard <keithp@keithp.com>
83693Date: Tue Sep 23 09:22:07 2008 -0700
83694
83695 Revert "Render: Use built-in SHA1 library"
83696
83697 This reverts commit d3bd31fddff7894f89ba80a3cdddff49aff08db8.
83698
83699 X.org should not be providing a custom SHA1 implementation.
83700
83701commit 10a9bac0257b381367cedff395b40425d584bf59
83702Author: Keith Packard <keithp@keithp.com>
83703Date: Tue Sep 23 09:21:37 2008 -0700
83704
83705 Revert "Update COPYING for SHA1 code"
83706
83707 This reverts commit 64387c8573ca8b3909667e32d7a9fba9567b21d3.
83708
83709 X.org should not be providing a custom SHA1 implementation.
83710
83711commit 64387c8573ca8b3909667e32d7a9fba9567b21d3
83712Author: Daniel Stone <daniel@fooishbar.org>
83713Date: Tue Sep 23 19:05:14 2008 +0300
83714
83715 Update COPYING for SHA1 code
83716
83717commit d3bd31fddff7894f89ba80a3cdddff49aff08db8
83718Author: John Tapsell <johnflux@gmail.com>
83719Date: Tue Sep 23 18:56:32 2008 +0300
83720
83721 Render: Use built-in SHA1 library
83722
83723 Getting an external library for SHA1 is a mess, so just use our own,
83724 regrettably. Public domain implementation.
83725
83726commit 991ed2949ba682684952fe323f8f0f51ebdca98e
83727Author: Adam Jackson <ajax@redhat.com>
83728Date: Tue Sep 23 10:22:22 2008 -0400
83729
83730 Rename SaveSetMap() to SaveSetShouldMap().
83731
83732 Avoids preprocessor collision with xfixeswire.h
83733
83734commit ca9fae00795a114bca4397c32b543d6326a4c547
83735Author: Owen Taylor <otaylor@redhat.com>
83736Date: Mon Sep 22 12:42:41 2008 -0700
83737
83738 Change 'remap' to 'map' in saveset functions/macros
83739
83740 Now that the code has been fixed so that Unmap means unmap and not "don't
83741 remap", 'remap' was confusing to have in the function names/parameters, so
83742 change it to simple 'map'.
83743
83744 Signed-off-by: Keith Packard <keithp@keithp.com>
83745
83746commit ad14239a358cf65e5702ec7d054aa1db4f1cdd68
83747Author: Adam Jackson <ajax@redhat.com>
83748Date: Mon Sep 22 11:26:49 2008 -0400
83749
83750 Upgrade GLX Public License 1.0 to FreeB 2.0.
83751
83752 According to the press release:
83753
83754 Previous SGI contributions to the free and open source community
83755 are now available under the new license. These contributions
83756 include the SGI® OpenGL® Sample Implementation, the GLX™ API and
83757 other GLX extensions.
83758
83759 [...]
83760
83761 "SGI has been one of the most ardent commercial supporters of free
83762 and open source software, so it was important to us that we continue
83763 to support the free software development community by releasing our
83764 earlier OpenGL-related contributions under this new license," said
83765 Steve Neuner, director of Linux, SGI. "This license ensures that all
83766 existing user communities will benefit, and their work can proceed
83767 unimpeded. Both Mesa and the X.org Project can continue to utilize
83768 this code in free software distributions of GNU/Linux. Now more than
83769 ever, software previously released by SGI under earlier GLX and SGI
83770 Free Software License B is free."
83771
83772 "The GLX API" is here read to include the original GLX source release
83773 from:
83774
83775 http://www.sgi.com/products/software/opensource/glx/download.html
83776 ftp://ftp.sgi.com/sgi/opengl/glx/glx1_2.1.tgz
83777
83778 which includes glxext.c as included in XFree86, from which our copies
83779 in glx/ and hw/dmx/glxProxy/ are derived.
83780
83781commit 722c51960dd4948b8a64f8eff115fa69da597ef9
83782Author: Owen W. Taylor <otaylor@redhat.com>
83783Date: Mon Sep 22 11:04:26 2008 -0400
83784
83785 Don't remap windows for SaveSetUnmap
83786
83787 Since ReparentWindow() does a unmap/map pair for windows that are already
83788 mapped, for saveset windows with SaveSetUnmap, we must unmap the window
83789 before calling ReparentWindow() to avoid the generation of MapRequest
83790 events, and so forth.
83791
83792commit 0bb317a78b96fddcdac319c9706b3a12f931ea44
83793Author: Adam Jackson <ajax@redhat.com>
83794Date: Mon Sep 22 10:51:51 2008 -0400
83795
83796 Default to -br. Add -retro option for the nostalgic.
83797
83798 -retro also reverts to the classic cursor display behavior, meaning,
83799 the cursor will be visible before anyone calls XDefineCursor().
83800
83801commit ab17f88506a2dde32c50ff72b03743cd9c384ecc
83802Author: Julien Cristau <jcristau@debian.org>
83803Date: Mon Sep 22 01:59:31 2008 +0200
83804
83805 Don't use our version of ffs() if the system has one
83806
83807commit af0d5e2ef7b1b3f42e3522568c5f56850823c102
83808Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83809Date: Sat Sep 20 19:41:46 2008 -0700
83810
83811 XQuartz: 64bit crash fix ... don't pass pointers through xe.
83812 (cherry picked from commit 5f1265b82edead8f15f2affc79c2782519502fa2)
83813
83814commit 852a0b0dde34e7179394d96d09122b99bf7fe904
83815Author: George Peter Staplin <gps@Georges-Workstation.local>
83816Date: Sat Sep 20 17:49:16 2008 -0600
83817
83818 XQuartz: pbproxy: Add an X error handler that returns 0.
83819
83820 Move the struct atom_list into the x-selection class, so that it's
83821 no longer a global variable named atoms. This may ease pthread
83822 integration and reduce the chances of symbols conflicting.
83823 (cherry picked from commit c1403c713ca80104ae3736bd2ed1eb6ffa5192b6)
83824
83825commit 7fa6fc5ad0b12bc52a1c22906709fbb003782d11
83826Author: George Peter Staplin <gps@Georges-Workstation.local>
83827Date: Fri Sep 19 21:28:46 2008 -0600
83828
83829 XQuartz: pbproxy: Fix a bug that occured when a PICT format was available.
83830 We may need another branch to convert a PICT to a PNG or JPEG. For now
83831 TIFF works well in all of the test image copying apps when converted to
83832 PNG or JPEG with an NSBitmapImageRep class.
83833 (cherry picked from commit adf339d8f948fc1e308dbcae38fcfce504b5b0ab)
83834
83835commit f67490ceb5b9ddf25e734cc331705103599f3ed8
83836Author: George Peter Staplin <gps@Georges-Workstation.local>
83837Date: Fri Sep 19 14:29:37 2008 -0600
83838
83839 XQuartz: pbproxy: Make the proxy handle copy request races, and PRIMARY
83840 INCR transfers mixed with a CLIPBOARD change.
83841
83842 Fix a bug with some UTF-8 transfers. I can only guess that my UTF-8
83843 tests before were inadequate. I can now copy/paste any characters between
83844 uxterm and Textedit.app.
83845 (cherry picked from commit 7ca1532e8e6e2c733c9c508eee0de73d09ecb947)
83846
83847commit b666ffd78155db228174041e9d7460b8a07dde0f
83848Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83849Date: Fri Sep 19 13:15:07 2008 -0700
83850
83851 XQuartz: "Disable" the text when syncing is disabled.
83852 (cherry picked from commit 7713d7ab66e74b212a6c0d634a3ad82373e178db)
83853
83854commit dba290ec231422740f453b69b187f80df3f50e4b
83855Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83856Date: Sun Sep 21 17:40:35 2008 -0700
83857
83858 XQuartz: Added a preferences UI for pbproxy options.
83859 (cherry picked from commit d76cca5e1eae86450aa236e49a2024881d0bac01)
83860
83861commit a928cae57b47625c9e3d136840d3e5d17fd3c51e
83862Author: George Peter Staplin <gps@Georges-Workstation.local>
83863Date: Fri Sep 19 00:18:44 2008 -0600
83864
83865 XQuartz: pbproxy: Add a missing include of Xutil.h. Another header
83866 seems to have done this before, but correctness matters.
83867 (cherry picked from commit 1b4987e779d97b90669bac2405a4672085677f7c)
83868
83869commit 749bdf19b2a50201fddd04207e31122470f435e3
83870Author: George Peter Staplin <gps@Georges-Workstation.local>
83871Date: Fri Sep 19 00:02:48 2008 -0600
83872
83873 XQuartz: pbproxy: Add COMPOUND_TEXT handling. Do misc. cleanups with testing
83874 to verify that the behavior didn't change.
83875
83876 main.m: XInternAtom compound_text, and atom_pair.
83877
83878 pbproxy.h: Add compound_text and atom_pair to the struct atom_list.
83879
83880 x-selection.m: Add an #include of Xutil.h. Refactor the reply struct
83881 initialization to be done in a common place. Add send_reply: to simplify
83882 the code a bit more. Add send_compound_text: which handles the
83883 COMPOUND_TEXT type. Add the beginnings of a send_multiple:. Change
83884 handle_image:extension: to handle_image:. The extension: message isn't
83885 needed anymore.
83886 (cherry picked from commit 1e9460abdf5bafe46215966bbef3e796cb1c33e0)
83887
83888commit 00bfbee59fe3b0c8d1a55d1851206857ca563ece
83889Author: George Peter Staplin <gps@Georges-Workstation.local>
83890Date: Thu Sep 18 13:10:41 2008 -0600
83891
83892 XQuartz: pbproxy: pbproxy now responds to selection request events.
83893
83894 We can now copy and paste images to and from X11. Text copying and
83895 pasting works as well.
83896
83897 The NSPasteboard can contain TIFF or PICT images, and pbproxy will
83898 translate to an image/png or image/jpeg request, and list those in the
83899 TARGETS.
83900
83901 I added a description of the basic design at the top of x-selection.m.
83902
83903 I removed the request_data x_selection class struct. It's not needed.
83904 (cherry picked from commit 4a8daf884694c9c420c45d427f1f84e608e7e48f)
83905
83906commit 811e9f5e9e9c07f4b8fb1e9a4d9df91a2e359221
83907Author: George Peter Staplin <gps@Georges-Workstation.local>
83908Date: Wed Sep 17 11:59:36 2008 -0600
83909
83910 XQuartz: pbproxy: Use an NSBitmapImageRep to convert an "image/png" selection
83911 to a TIFFRepresentation for use with NSPasteboard.
83912
83913 This has been tested with the Gimp and works with some minor quartz-wm changes.
83914 The Finder clipboard shows the image updates after an Edit -> Copy.
83915 (cherry picked from commit 12912adaeea759d30f96d8ae51a84fd1659ea0ac)
83916
83917commit d8d555bac77509248d1145e928a2edf500b326b9
83918Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83919Date: Wed Sep 17 10:11:00 2008 -0700
83920
83921 XQuartz: Don't enable 8bit visuals in the TrueColor server, since it's not working yet
83922 (cherry picked from commit 1317c8ff94c4bd32617a5398e929f667bd87da58)
83923
83924commit 23ec8261b6a746cb84e1328284fdc5587bca6faf
83925Author: George Peter Staplin <gps@Georges-Workstation.local>
83926Date: Tue Sep 16 21:09:22 2008 -0600
83927
83928 XQuartz: pbproxy: The greedy CLIPBOARD handling now works for text.
83929
83930 This change adds some [self own_clipboard] calls in the necessary places to get the proper greedy behavior.
83931
83932 UTF8_STRING and STRING properties seem to work well now with the test cases (PRIMARY, and CLIPBOARD). I can copy from several different X apps, and have the behavior be correct when pasting. I also verified that quartz-wm isn't doing the copying, by disabling the quartz-wm paths.
83933 (cherry picked from commit 934669f732c28f07b9d934d8f8f0b63ccfebd900)
83934
83935commit 7c2eb3d41af33d1c52500111d0d34f9aab8a4542
83936Author: George Peter Staplin <gps@Georges-Workstation.local>
83937Date: Tue Sep 16 15:21:18 2008 -0600
83938
83939 XQuartz: pbproxy: return the TARGETS list as a 32-bit list to fix a bug that was in the original.
83940
83941 Add TRACE() calls where appropriate to try to figure out why we are losing CLIPBOARD at times, after transferring PRIMARY to the NSPasteboard.
83942
83943 Use the new pbproxy_clipboard_to_pasteboard where appropriate.
83944 (cherry picked from commit 40190675a6c1889cca3574e5d1a9c16ab74315a7)
83945
83946commit 81c836902dc5b0b83cd95262d48cbc4c81ff3ae2
83947Author: George Peter Staplin <gps@Georges-Workstation.local>
83948Date: Mon Sep 15 13:35:46 2008 -0600
83949
83950 Rename reclaim_clipboard to claim_clipboard.
83951
83952 Convert the puts usage to use DB().
83953
83954 Add the initial handle_image method.
83955
83956 Check for nil in the NSString instantiation in various places.
83957
83958 Add some commentary to enhance the clarity of why I did some things.
83959 (cherry picked from commit 37361567b65241eab64e8b30cd9729d0e71a86d2)
83960
83961commit 85e23affea7bf9aa7615b0292e67f170266d85f8
83962Author: George Peter Staplin <gps@Georges-Workstation.local>
83963Date: Mon Sep 15 13:31:19 2008 -0600
83964
83965 Add an image_jpeg Atom.
83966 (cherry picked from commit ea9dc27977c5eab666f2aa2d914e4e28d36758c7)
83967
83968commit 6b42f40ff9b9ff3e74ccbb2ce5cfad65016934de
83969Author: George Peter Staplin <gps@Georges-Workstation.local>
83970Date: Mon Sep 15 13:28:50 2008 -0600
83971
83972 Rename reclaim_clipboard to claim_clipboard.
83973 (cherry picked from commit d333a8e2b5514b8b76a78c6a13a3f5149ea8de27)
83974
83975commit 2ab56981b1a64138d6c1f9ecfb75655477b3e575
83976Author: George Peter Staplin <gps@Georges-Workstation.local>
83977Date: Mon Sep 15 13:27:08 2008 -0600
83978
83979 Add image/jpeg type, and rename reclaim_clipboard to claim_clipboard.
83980 (cherry picked from commit 32b175ef62b68a971784e51e937e358cb10e20ac)
83981
83982commit cbd32645f7d1054f79ad137a16766da1076c610e
83983Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83984Date: Mon Sep 15 09:33:55 2008 -0700
83985
83986 XQuartz: pbproxy: Added global variables to customize how pbproxy behaves
83987 (cherry picked from commit 21a2d0b8d03faf1c66ec0c5c11fbd2ab24547727)
83988
83989commit 5b397642cdc29920245ebe64cc85243cac893e0e
83990Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83991Date: Sun Sep 14 09:23:17 2008 -0700
83992
83993 XQuartz: pbproxy: First round of updates to pbproxy from George.
83994 (cherry picked from commit 5c8b956f8f3f17e8d577d97cb66424954be72684)
83995
83996commit dac9e91870f9c787eac97c9d7d0607979c57eb5b
83997Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
83998Date: Sun Sep 21 17:32:24 2008 -0700
83999
84000 mieq: Made custom event handlers safer for nevents >1
84001
84002commit 8c46505d7d91e0644b19cccc4b342fceb6f86cab
84003Author: Kim Woelders <kim@woelders.dk>
84004Date: Mon Sep 22 08:37:29 2008 +0930
84005
84006 xkb: fix use of uninitialized variable.
84007
84008 And some cosmetic changes to use stuff->change consistently.
84009
84010 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
84011
84012commit 2b266eda6e23d16116f8a8e258192df353970279
84013Author: Peter Harris <peter.harris@hummingbird.com>
84014Date: Thu Dec 20 15:58:01 2007 -0500
84015
84016 Fix panoramiX request and reply swapping
84017
84018 Fix panoramiX request and reply swapping
84019 Set window and screen values in panoramix replies
84020 Prevent buffer overrun in ProcPanoramiXGetScreenSize
84021
84022commit 03ab8f11d25b0ca39a3b37b5350bca4c1028768e
84023Author: Keith Packard <keithp@keithp.com>
84024Date: Thu Sep 18 12:17:05 2008 -0700
84025
84026 When resizing a window with redirected descendents, don't expose them
84027
84028 Bit/window gravity computations need to recompute exposures to manage the
84029 bits which are saved by gravity during the resize computation. That's easy
84030 for non-redirected windows where the bits are all within the parent's
84031 pixmap. For redirected windows, we don't need to deal with this at all, so
84032 just skip the whole re-computation adventure.
84033
84034 Signed-off-by: Keith Packard <keithp@keithp.com>
84035
84036commit d26083fe02658e7312a8da6a5b11652f9fe57bc8
84037Author: Keith Packard <keithp@keithp.com>
84038Date: Thu Sep 18 12:14:58 2008 -0700
84039
84040 Switching from Automatic to Manual redirect needs to unmap/remap
84041
84042 When changing a window from automatic redirection to manual redirection, the
84043 parent clip list needs to be recomputed; the easy way to get that computed
84044 right is to unmap/map the window, just as when redirecting the window the
84045 first time.
84046
84047 Thanks to Owen Taylor for helping diagnose this.
84048
84049 Signed-off-by: Keith Packard <keithp@keithp.com>
84050
84051commit 8bfd88ca61a1747714781a090d1c9dd8f3ad5dfc
84052Author: Adam Jackson <ajax@redhat.com>
84053Date: Fri Sep 19 12:13:43 2008 -0400
84054
84055 Revert accidental edid changes from previous commit.
84056
84057commit 69b79c1a6648f0d04a1b37a728b2e3d76233530f
84058Author: Adam Jackson <ajax@redhat.com>
84059Date: Fri Sep 19 12:02:28 2008 -0400
84060
84061 Update to SGI FreeB 2.0.
84062
84063 Under the terms of version 1.1, "once Covered Code has been published
84064 under a particular version of the License, Recipient may, for the
84065 duration of the License, continue to use it under the terms of that
84066 version, or choose to use such Covered Code under the terms of any
84067 subsequent version published by SGI."
84068
84069 FreeB 2.0 license refers to "dates of first publication". They are here
84070 taken to be 1991-2000, as noted in the original license text:
84071
84072 ** Original Code. The Original Code is: OpenGL Sample Implementation,
84073 ** Version 1.2.1, released January 26, 2000, developed by Silicon Graphics,
84074 ** Inc. The Original Code is Copyright (c) 1991-2000 Silicon Graphics, Inc.
84075 ** Copyright in any portions created by third parties is as indicated
84076 ** elsewhere herein. All Rights Reserved.
84077
84078 Official FreeB 2.0 text:
84079
84080 http://oss.sgi.com/projects/FreeB/SGIFreeSWLicB.2.0.pdf
84081
84082 As always, this code has not been tested for conformance with the OpenGL
84083 specification. OpenGL conformance testing is available from
84084 http://khronos.org/ and is required for use of the OpenGL logo in
84085 product advertising and promotion.
84086
84087commit cc20112a65d3f641ce0261c86a541f94fae5215c
84088Author: Peter Hutterer <peter.hutterer@redhat.com>
84089Date: Fri Sep 19 08:44:04 2008 +0930
84090
84091 Xi: don't include client-side headers.
84092
84093 Spotted by Mikhail Gusarov.
84094
84095commit e76b5b5e731e2d6c61ff6e6e9b2f42b20ac473d5
84096Author: Peter Hutterer <peter.hutterer@redhat.com>
84097Date: Thu Sep 18 14:51:15 2008 +0930
84098
84099 dix: calloc, not malloc, ValuatorClassRec.
84100
84101 For master devices, the ptraccel code could segfault on free since we'd be
84102 dereferencing random memory. Callocing the valuatorClassRec is the easy fix.
84103
84104commit f789408f7dbfe57a0111e185ac83d1c70f262ba8
84105Author: Peter Hutterer <peter.hutterer@redhat.com>
84106Date: Mon Aug 25 14:52:35 2008 +0930
84107
84108 Xi: remove broken SDeviceEnterNotifyEvent, replace with ...Leave...
84109
84110 The event format is the same for both (bar the type), so one is enough.
84111
84112commit 97b9374a8af5a4cc52f9074feb4ce40cf0b90796
84113Author: Aaron Plattner <aplattner@nvidia.com>
84114Date: Wed Sep 17 15:48:56 2008 -0700
84115
84116 Conditionalize Composite-based backing store on pScreen->backingStoreSupport.
84117 (cherry picked from commit 37876602957924c7cff759a800eddd574ee2385a)
84118
84119commit b4762c0245ed2966606171cf27f40aa745fdc76e
84120Author: Julien Cristau <jcristau@debian.org>
84121Date: Tue Sep 16 17:13:42 2008 +0200
84122
84123 exa: disable shared pixmaps
84124
84125 They got re-enabled in ee7c684f21d220d5e046bab31ae617a7d64d60f6
84126 ("Reimplement ShmPutImage.")
84127
84128commit 52bbfc5bfa64fd09ef625f35a2fb4a0518003825
84129Author: Mikhail Gusarov <dottedmag@dottedmag.net>
84130Date: Mon Sep 15 10:43:46 2008 -0400
84131
84132 scrnsaver is optional at build time.
84133
84134commit b6ab114212c0e4c3346ceb5b207f14c526ab81e7
84135Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
84136Date: Fri Sep 12 19:11:53 2008 -0400
84137
84138 Array-index based devPrivates implementation.
84139
84140 Note: DevPrivateKey is now pointer-to-int, which means
84141 each key now needs to point to some global storage of
84142 size at least sizeof(int).
84143
84144commit 999f3362d50a753c1801d565766219808efdb1a5
84145Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84146Date: Fri Sep 5 17:32:16 2008 -0700
84147
84148 XQuartz: Dead code removal
84149 (cherry picked from commit 43184cd379c4fb7254391b8a362016cbf89b5529)
84150
84151commit 6548a55ebdf43231a3c48debf54aa5aa2f1aa956
84152Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84153Date: Fri Sep 5 17:05:03 2008 -0700
84154
84155 XQuartz: 256 color support
84156 (cherry picked from commit 8dd6d5c825d457f26b41b79d02d57ed4a5ecf1f5)
84157
84158commit bad7cd14c2021b14971b3f707f927803a053003e
84159Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84160Date: Fri Sep 5 10:46:36 2008 -0700
84161
84162 XQuartz: Dead code removal
84163 (cherry picked from commit bf10fb0b1f776e72db7c76db11f764e26f9d62c4)
84164
84165commit 1119fe136f8731f26fc6f50b92f5ddf78f3f83be
84166Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84167Date: Fri Sep 5 10:46:20 2008 -0700
84168
84169 XQuartz: Changed around fd handoff model to be more robust
84170 (cherry picked from commit 4fe605c2bc62d50e0b5764d9edda245af227630b)
84171
84172commit 183233b27beb441742a53e440c3389b4ea125b8a
84173Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84174Date: Fri Sep 12 15:25:44 2008 -0700
84175
84176 XQuartz: Don't warp the pointer on server start for XQuartz
84177 (cherry picked from commit c0da576d4921c246a9ac747921b48ab9e718347f)
84178
84179commit 7a46dd30d332bc1d42f82b0391e806d3b067ebbd
84180Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84181Date: Thu Sep 4 09:18:40 2008 -0700
84182
84183 XQuartz: Dead code removal
84184 (cherry picked from commit 2e45344870f5d17181df6407da3448991036d783)
84185
84186commit 1c70e53f2560cb877c8b149034f296232de67a4e
84187Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84188Date: Thu Sep 4 08:55:22 2008 -0700
84189
84190 XQuartz: Fixed proximity logic errors for tablets.
84191 (cherry picked from commit d942849f248c368b92ca73e145e8a5bc339112d7)
84192
84193commit c661c6ae7a9cd981f914484aa4c3007922280def
84194Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84195Date: Thu Sep 4 08:07:32 2008 -0700
84196
84197 XQuartz: Remove decls of crashtracer strings that we don't use anymore in here
84198 (cherry picked from commit 4a653c6bfd270f2960a8c7e726e7628cfc3c9051)
84199
84200commit 54162aca404a7de2e560cc2aea01445e14742efe
84201Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84202Date: Thu Sep 4 05:18:31 2008 -0700
84203
84204 XQuartz: Fixed a crash in RootlessNativeWindowMoved
84205 (cherry picked from commit a16048cec08e173ce42a78a77704f5fdfcb4480f)
84206
84207commit ce781cac1b52003da6c025cd29e2c2ad97dbb324
84208Author: Eric Anholt <eric@anholt.net>
84209Date: Thu Sep 11 16:24:20 2008 -0700
84210
84211 Fix distcheck.
84212
84213commit 31c62495f1de6e9ba41e1f6d7fa263eeb849129b
84214Author: Keith Packard <keithp@keithp.com>
84215Date: Wed Sep 10 13:11:09 2008 +0800
84216
84217 Drop a reference to user mode after create
84218
84219 User mode has no customer when create until assigned
84220 to some output.
84221
84222commit 079625570d51e41569b73b2fd9237eb8f967f408
84223Author: Aaron Plattner <aplattner@nvidia.com>
84224Date: Mon Sep 8 08:50:52 2008 -0700
84225
84226 Bump ABI major versions for the TryClientExceptions change from commit 883811c.
84227
84228commit 4e004c68283828660f2259a4a4120a84667dcafa
84229Author: Zhenyu Wang <zhenyu.z.wang@intel.com>
84230Date: Mon Sep 8 09:16:59 2008 +0800
84231
84232 Revert "randr: fix user mode create initial reference count"
84233
84234 This reverts commit 7c5ca85a9e6d49ab572831b3e0c08bb4cafb395e.
84235
84236commit 40a9482f8c5b3eac86c221ccce78510e850f4d2b
84237Author: Jesse Barnes <jbarnes@virtuousgeek.org>
84238Date: Fri Sep 5 06:09:51 2008 -0700
84239
84240 Add cscope files to .gitignore
84241
84242commit bf333c2f9833a178887e7bdd7fc338f1e09c387f
84243Author: John Tapsell <john.tapsell@john-desktop.(none)>
84244Date: Wed Sep 3 12:20:17 2008 +0100
84245
84246 fbdevhw: Remove pixclock check.
84247
84248 The check can fail because the output from FBIOGET_VSCREENINFO is used to set
84249 Clock in fbdev2xfree_timing(). Then in fbdevHWSetMode(), xfree2fbdev_timing()
84250 is called which sets the pixclock based on Clock. The resulting circle results
84251 in slight rounding errors, causing the comparision check in fbdev_modes_equal
84252 to fail.
84253
84254commit 47bf269d5922a4998c33908a53ad1affa8de96c0
84255Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84256Date: Wed Sep 3 12:54:56 2008 -0700
84257
84258 mieq: Fix my broken cherry pick.
84259
84260commit 1669308be6d4cc6265e5fa8ef47fbf2cbef6f093
84261Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84262Date: Wed Sep 3 09:23:06 2008 -0700
84263
84264 XQuartz: Added pasteboard proxy code stripped out of quartz-wm.
84265 (cherry picked from commit 1f842c71c35db031a24de646429834d6054adf1d)
84266 (cherry picked from commit 144746223d8dc3f6ff2a0e805aa27a6e57df558a)
84267
84268commit 46168b3b404164c1f2e7876031e70fb77192fe75
84269Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84270Date: Wed Sep 3 09:11:25 2008 -0700
84271
84272 XQuartz: Updated Tablet code to include ranges for valuators and added cursor and erasor
84273 (cherry picked from commit 6bca78760951cb5cb57ea66b7631a2dc230dc27a)
84274 (cherry picked from commit fafcafd6de39cf6b5967793f03b30b49db36fec5)
84275
84276commit 7f5318131ddf86bed5797c0d9641f6ff34f34be0
84277Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84278Date: Wed Sep 3 03:54:42 2008 -0700
84279
84280 XQuartz: Removed "Done" button from Applications Customization. Now properly save on exit.
84281 (cherry picked from commit 2f1e163d5db6063e54fcaa045ea67c29ce3e46aa)
84282 (cherry picked from commit 7542c00b96461c3474b8c23545171c4b6ddfd29d)
84283
84284commit 9e70a3cf58c205948f6a415e203d825e9b660d9d
84285Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84286Date: Wed Sep 3 09:33:39 2008 -0700
84287
84288 mieqProcessInputEvents: Check custom event handlers first.
84289
84290commit 700e8bf33f9b1c3921de912b8110fadf5e0626da
84291Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84292Date: Wed Sep 3 02:41:20 2008 -0700
84293
84294 XQuartz: Updated man page
84295 (cherry picked from commit 6932d0d19ea75c64ed38b82a246b68c6bbfb8cec)
84296 (cherry picked from commit 8e3da36b6b0a261975808bc67b5caca733a1a23d)
84297
84298commit 7430bb9093ea8cf9589d745742c7353e53a144f8
84299Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84300Date: Wed Sep 3 02:37:09 2008 -0700
84301
84302 XQuartz: Removed dead code (unused option definitions)
84303 (cherry picked from commit 3d74f77490105c0704c0b10dc7a6eb859bf6e985)
84304 (cherry picked from commit c1e5ea9ba058f6a13f564165a3d6862335729630)
84305
84306commit b3836e5bdd7e2e04e563246d90f385e1b4741baa
84307Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84308Date: Wed Sep 3 02:36:39 2008 -0700
84309
84310 XQuartz: Cleaned up our sleeping before adding the launchd socket.
84311 (cherry picked from commit ebfe49dfca6e04d66232bad760f522629397cc35)
84312 (cherry picked from commit 2cf74a76d54d795cfc3c0380a526b9177327782d)
84313
84314commit 7c5ca85a9e6d49ab572831b3e0c08bb4cafb395e
84315Author: Zhenyu Wang <zhenyu.z.wang@intel.com>
84316Date: Wed Sep 3 09:41:00 2008 +0800
84317
84318 randr: fix user mode create initial reference count
84319
84320 Don't need extra reference count adding when creating user mode.
84321 This fixes user mode destroy, otherwise we get BadAccess error.
84322
84323commit 4e1f9f51a9eea115ce2a073824e9c5cef4e2f364
84324Author: Daniel Drake <dsd@laptop.org>
84325Date: Tue Sep 2 16:20:40 2008 -0400
84326
84327 Don't try to destroy a NULL pixmap.
84328
84329commit 613ce0955032fb032de0a3940752828d314f057a
84330Author: Adam Jackson <ajax@redhat.com>
84331Date: Tue Sep 2 15:43:15 2008 -0400
84332
84333 Fix strict-aliasing silliness in XAA.
84334
84335commit b5cdcfa55c399e83d51242e93d4f25d8bc4fec1f
84336Author: Julien Cristau <jcristau@debian.org>
84337Date: Mon Sep 1 19:45:30 2008 +0200
84338
84339 Xevie: swap replies as necessary
84340
84341commit d3ae193f4ac87530f2745f8cb5e7b70dd516881e
84342Author: Thorvald Natvig <slicer@users.sourceforge.net>
84343Date: Mon Sep 1 19:36:56 2008 +0200
84344
84345 Xevie: always initialize rep.length (bug#17394)
84346
84347 The XEvIE extension doesn't clear the rep.length field for any reply but
84348 the version check. Hence, if there is junk data in it and that is sent
84349 to the client, it hangs.
84350
84351 X.Org bug#17394 (http://bugs.freedesktop.org/show_bug.cgi?id=17394)
84352
84353commit 0466b8811ce8ebc1d993f177cbb7a22e997640f1
84354Author: Dave Airlie <airlied@redhat.com>
84355Date: Mon Sep 1 09:57:25 2008 +1000
84356
84357 exa: drop cw.h inclusion
84358
84359 this doesn't seem to be needed, at least I can't see any warning without it.
84360
84361 I'd like to build EXA into a driver and cw.h isn't exported
84362
84363commit 881bd15f445bb62c3fdea863c968c95217408b6e
84364Author: Maarten Maathuis <madman2003@gmail.com>
84365Date: Sun Aug 31 18:07:22 2008 +0200
84366
84367 xf86Crtc: Set mode to NULL, so nothing bad happens if crtc == NULL.
84368
84369 - Fix is courtesy of math_b.
84370
84371commit f4e9a1a98f97155de62908309e87c10487125926
84372Author: Maarten Maathuis <madman2003@gmail.com>
84373Date: Sun Aug 31 17:45:08 2008 +0200
84374
84375 damage: choose less ambiguous function names
84376
84377commit 0cd6709db5febbe7fbd073e429421fcd81041e08
84378Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84379Date: Sat Aug 30 16:13:58 2008 -0700
84380
84381 XQuartz: Fixed editing of Applications menu
84382 (cherry picked from commit da647f1add4c1ae7854c0fbfbf972666ce5d12c8)
84383
84384commit c4d290fc54542ccba7c3a46bc75b0c3322d0ca48
84385Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84386Date: Sat Aug 30 14:19:47 2008 -0700
84387
84388 XQuartz: Fixed missing symbol in quartzKeyboard debugging
84389 (cherry picked from commit 8ad55e484f0ba9073ce07a03de660cc546a7b668)
84390
84391commit 5d7d959b1143979c3445639c9d108e02da0c1475
84392Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84393Date: Sat Aug 30 14:18:07 2008 -0700
84394
84395 XQuartz: Don't even build with GLX support since swrast now depends on libdrm
84396 (cherry picked from commit 3739525d57dff2b3b034b5768f357b1c25577909)
84397
84398commit 45def7f0a342f4733cafb06adbfed14932f0e01b
84399Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84400Date: Fri Aug 29 22:16:03 2008 -0700
84401
84402 Removed dead glcontextmodes
84403 (cherry picked from commit c6d0ac7471209d835e8d9a25adcbda893ddfa1cb)
84404
84405commit d01c5ca7935a8340a3cd68c325da6dfec005c952
84406Author: Matthieu Herrb <matthieu.herrb@laas.fr>
84407Date: Sat Aug 30 15:35:32 2008 +0200
84408
84409 Xserver.man: Typo (the the).
84410
84411commit 229e60db8f95232afc8cdcb7cd0572d117c84b90
84412Author: Thomas Bodzar <Thomas.Bodzar@quest.com>
84413Date: Sat Aug 30 15:34:29 2008 +0200
84414
84415 Xorg.man: typo (the the).
84416
84417commit 1885582444fb06da73dea1f118a777bea26e1dc1
84418Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84419Date: Fri Aug 29 21:18:26 2008 -0700
84420
84421 XQuartz: Placate automake which doesn't like escaped newlines even in comments...
84422 (cherry picked from commit 73db1170129a31ad2d55a7e83ac6dfc8030e47f1)
84423
84424commit 8abcc12fdd5bd9edc0288a1ab8f75e65ffcb8c71
84425Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84426Date: Fri Aug 29 20:55:25 2008 -0700
84427
84428 XQuartz: Added debugging output to the crash log to help track down two crashes, since people don't often report their system.log spew.
84429 (cherry picked from commit aaf0f71db197526b6b866cc1b39fbdfe051879ef)
84430
84431commit aee19c7b5542aa147b89b1ae8340e9931d81729f
84432Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84433Date: Fri Aug 29 03:54:55 2008 -0700
84434
84435 XQuartz: xp_is_symbolic_hotkey_event catches the input menu, but not our mainMenu, so we still need to do that the old way.
84436 (cherry picked from commit 25eccf12c89d73b8bce2c9be14841efb230acedc)
84437
84438commit 974db58f5b730c3770ee461665a02dd4334d1dea
84439Author: Maarten Maathuis <madman2003@gmail.com>
84440Date: Sat Aug 30 00:37:11 2008 +0200
84441
84442 damage: initial attempt at a damage marker mechanism
84443 - This should allow drivers to recieve post submission events for X<->opengl synchronisation.
84444 - Lacking a testcase, i'm open to suggestion how to do it better.
84445 - The idea is:
84446 - driver recieves event
84447 - driver creates personal identification and inserts marker into X fifo.
84448 - when something wants to use an X pixmap, it checks if something is pending.
84449 - If so, it synchronizes the 2nd fifo using the initial identification.
84450 - Driver is not required to use interrupt based systems (price too high).
84451 - Lower latency is ofcource better.
84452 - If this is somehow unusable for you, then come up with improvements.
84453 - For that reason i wouldn't consider the api fixed for the moment.
84454
84455commit 454cb0802eec3c2c2cdbcc17971bced868462b83
84456Author: Maarten Maathuis <madman2003@gmail.com>
84457Date: Fri Aug 29 22:28:02 2008 +0200
84458
84459 damage: DamageReportRawRegion should set pDamage->damage
84460
84461 - I found no evidence in the protocol, that it should be differently from all the other modes.
84462 - It seems to have been like this from day 1.
84463 - If anyone has evidence to the contrary, please enlighten me.
84464
84465commit ae6ca434104405302f30a58bde8738d9579d9dc9
84466Author: Maarten Maathuis <madman2003@gmail.com>
84467Date: Fri Aug 29 22:21:54 2008 +0200
84468
84469 damage: internal functions start with a non-capital letter
84470
84471commit 1861250cd7e84b05e8298b74e3c7e97da72ddfba
84472Author: Maarten Maathuis <madman2003@gmail.com>
84473Date: Fri Aug 29 22:15:23 2008 +0200
84474
84475 {damage,exa}: sanitise damage
84476
84477 - Redo damage naming for more consistency.
84478 - Call post submission functions only where appropriate.
84479 - EXA can now live without it's odd damage workarounds.
84480
84481commit 5af77d43fe812e127d5d335527fa940ab9d95f38
84482Author: Kristian Høgsberg <krh@redhat.com>
84483Date: Mon Aug 11 16:59:17 2008 -0400
84484
84485 DRI2: Drop sarea use, implement server side swap buffers.
84486
84487commit 60ad8d5d05485339e89d7f1f9f1ded75de7c7ea1
84488Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
84489Date: Thu Aug 28 23:45:17 2008 -0400
84490
84491 Attempt getpeercon() on remote sockets as well as local ones.
84492
84493commit ebea78cdba0ff14a397239ee1936bd254c181e1b
84494Author: Tomas Carnecky <tom@dbservice.com>
84495Date: Thu Aug 28 18:05:40 2008 -0400
84496
84497 Prepare for array-index based devPrivates.
84498
84499 TODO: static indices can be made just an int; some indices
84500 can be combined.
84501
84502commit ec7907f8fa04dcff2649cc4846975844314f737e
84503Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
84504Date: Wed Aug 27 19:27:13 2008 -0400
84505
84506 Add an array of integers for use as per-screen cursor private keys.
84507
84508 Replaces the use of the screen pointer itself as the key, which was
84509 nice but won't work now that an array index is being stored.
84510
84511commit 86898491497a43814caf42013651086e62fe6162
84512Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
84513Date: Wed Aug 27 19:17:15 2008 -0400
84514
84515 Remove unused GetGlyphPrivatesForScreen.
84516
84517commit 90b178cc7feda1f9c4995b98364739e71233fff3
84518Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
84519Date: Thu Aug 28 19:36:22 2008 -0400
84520
84521 Don't need to request space for Xnest pixmap private.
84522
84523commit 835b532d6f51d08998d7ab49fcb41db8266487ce
84524Author: Adam Jackson <ajax@redhat.com>
84525Date: Thu Aug 28 14:49:35 2008 -0400
84526
84527 Build fix.
84528
84529commit 040212ee60af9d043c4c3f98bba5120c3b8e33f7
84530Author: Adam Jackson <ajax@redhat.com>
84531Date: Thu Aug 28 13:49:35 2008 -0400
84532
84533 MIT-SHM pixmaps, if they exist, are ZPixmap.
84534
84535commit ee7c684f21d220d5e046bab31ae617a7d64d60f6
84536Author: Adam Jackson <ajax@redhat.com>
84537Date: Wed Aug 27 16:05:47 2008 -0400
84538
84539 Reimplement ShmPutImage.
84540
84541 There's no reason to not just dispatch this straight into the GC. As a
84542 bonus, if you do so, damage wraps correctly, and thus swcursor works.
84543 The side effect is it's no longer possible to override ShmPutImage with
84544 ShmRegisterFuncs().
84545
84546 Also remove the (broken) damage tracking for same from EXA, since it didn't
84547 work right, and is now superfluous.
84548
84549commit 2db1afbf2e56d8743c701d81a5797001ce9e5c52
84550Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84551Date: Tue Aug 26 23:09:20 2008 -0700
84552
84553 XQuartz: Always use TIS for the keyboard layout seed since KB* aren't thread safe.
84554 (cherry picked from commit c8244177b0dbcb28d2f5509e6f2a3e8057889790)
84555
84556commit 319405fa4a530804c19fb7d0fcece5c912e41c9d
84557Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84558Date: Tue Aug 26 21:11:35 2008 -0700
84559
84560 XQuartz: Fixed needed xplugin version. ><
84561 (cherry picked from commit 01b3a99dff79e0d2b316e02658c19fa79d9144ae)
84562
84563commit 3350770bd6d90fefa1133ac738c1d4eae2d568a6
84564Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84565Date: Tue Aug 26 19:13:56 2008 -0700
84566
84567 XQuartz: Use new Xplugin API to determine if a keypress event corresponds to an OS-X hotkey that we should dispatch to Appkit.
84568 (cherry picked from commit e7658e745f8a3eaf5cec9f54a8a1b7e5a7708e09)
84569
84570commit 7930ea9116f232a3762f0085e5ca65e06d1142fc
84571Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84572Date: Mon Aug 25 22:10:46 2008 -0700
84573
84574 XQuartz: Made a note to come back to the xinitrc race condition in the 1.5 branch when we move up to it.
84575 (cherry picked from commit 54a882dcbdccef3f3eb4e6398e13c88185a8986c)
84576
84577commit dc1171df5baa0c4aa366ff2b07af5b8eb44dae12
84578Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84579Date: Fri Aug 22 01:12:37 2008 -0700
84580
84581 XQuartz: Fixed bogus args to debug ErrorF
84582 (cherry picked from commit f0351c2b80f30ae31f041798b84139141fc3d5a5)
84583
84584commit ac936525afe46bad329387232f8d48e77f6a4175
84585Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84586Date: Fri Aug 22 01:11:33 2008 -0700
84587
84588 XQuartz: Cleaned up mouse event translation a bit more.
84589 (cherry picked from commit c286f2a718fdaf6ad10249b59abb20731da8d904)
84590
84591commit 1b42f550001c11c93d5a2144df98a5c687c010c2
84592Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84593Date: Fri Aug 22 00:45:12 2008 -0700
84594
84595 Xquartz: Added missing ev_type for NSTabletPoint event.
84596 (cherry picked from commit 07548f13947a212dcc020d52fe0ad679255121e2)
84597
84598commit 13a89f19208cf998808ff2896b7711310768f94c
84599Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84600Date: Wed Aug 20 10:21:23 2008 -0700
84601
84602 XQuartz: Added more explanation to debug keylayout spew to cut down on report-spam.
84603 (cherry picked from commit fcdc9f8b5abd724d75f53d48fdadbed9859e9e79)
84604
84605commit dd1e54d6eed0bce289517b3cb95e6b1a187ca6e0
84606Author: Maarten Maathuis <madman2003@gmail.com>
84607Date: Tue Aug 26 20:02:55 2008 +0200
84608
84609 exa: fix thinko from 988725f32e082aee9392a71464125157a83d1e67
84610 - the drawable of the pixmap is not the same as the original drawable (possibly a DRAWABLE_WINDOW)
84611
84612commit 7c14fdbacfcd2f4d56a346e6c72e44e4ba9909c1
84613Author: Maarten Maathuis <madman2003@gmail.com>
84614Date: Tue Aug 26 17:21:43 2008 +0200
84615
84616 exa: some minor cleanup
84617 - Fix compile warning
84618 - Order exa.h by source file that exports the function.
84619 - Move the function i created earlier to private headers.
84620
84621commit de79edbd9f67762950eaac2dc79668035239897c
84622Author: Maarten Maathuis <madman2003@gmail.com>
84623Date: Tue Aug 26 17:03:12 2008 +0200
84624
84625 exa: report damage manually for exa{Trapezoids,Triangles} when needed
84626 - Plus a micro cleanup of unused variables.
84627
84628commit 988725f32e082aee9392a71464125157a83d1e67
84629Author: Maarten Maathuis <madman2003@gmail.com>
84630Date: Tue Aug 26 16:54:29 2008 +0200
84631
84632 exa: move destination damage for internal calls to a special function
84633 - This should improve clarity for someone who isn't familiar with the code.
84634
84635commit ce193476808f54d946351458361c62132d81b62f
84636Author: Maarten Maathuis <madman2003@gmail.com>
84637Date: Tue Aug 26 13:18:58 2008 +0200
84638
84639 exa_glyphs: remove useless offset
84640
84641commit fd94651fc3a6f49ec153ffb823b86c1d125bb298
84642Author: Kristian Høgsberg <krh@redhat.com>
84643Date: Tue Aug 26 10:58:35 2008 -0400
84644
84645 Fix driGetConfigAttribIndex unaligned access to GLboolean.
84646
84647 We don't actually send the float mode so just drop it. Drop a couple of
84648 other unused or redundant fields from GLXconfig.
84649
84650commit 41bd8d5dfe27d5c91fe7e4cdba1475a2a0741b1c
84651Author: Adam Jackson <ajax@redhat.com>
84652Date: Mon Aug 25 11:27:26 2008 -0400
84653
84654 EDID: Build fix.
84655
84656commit 76ed409acd772f2c041239345c6dc64cbef0e5b2
84657Author: Adam Jackson <ajax@redhat.com>
84658Date: Mon Aug 25 10:44:45 2008 -0400
84659
84660 EDID: Publish the whole block on the root window if we've got it.
84661
84662commit 5724f7fb5bea6fa1a354c64c0972c53d70e2f27b
84663Author: Adam Jackson <ajax@redhat.com>
84664Date: Mon Aug 25 10:39:36 2008 -0400
84665
84666 EDID: Publish the whole block in the RANDR property if we've got it.
84667
84668commit 668f89eba3e8f9da7843f5cb350f8dc1e5d7efbe
84669Author: Alan Coopersmith <alan.coopersmith@sun.com>
84670Date: Fri Aug 22 10:32:18 2008 -0700
84671
84672 Remove unnecessary #ifdef __SOL8__'s
84673
84674commit cc78d977cac74fcfb7c9b27e7109a1e369018dd8
84675Author: Alan Coopersmith <alan.coopersmith@sun.com>
84676Date: Fri Aug 22 10:14:11 2008 -0700
84677
84678 Convert Solaris #ifdef's for <sys/kd.h> to AC_CHECK_HEADERS check
84679
84680 Upcoming virtual terminal support changes in Solaris kernel will provide
84681 <sys/kd.h> on SPARC too, so this gets us ready for them.
84682
84683commit 95b466e457542bfe08b0ed2bbb5db8d28a961cfd
84684Author: Peter Hutterer <peter.hutterer@redhat.com>
84685Date: Thu Aug 21 17:24:40 2008 +0930
84686
84687 Xi: don't include .c files.
84688
84689commit c696da75c7326c5e1f1cd48292c0519ddc22e11b
84690Author: Peter Hutterer <peter.hutterer@redhat.com>
84691Date: Thu Aug 21 17:23:11 2008 +0930
84692
84693 Xi: swap devices property replies.
84694
84695commit 01264f17925005969c3b71ca945fc1014bcd8c8e
84696Author: Julien Cristau <jcristau@debian.org>
84697Date: Thu Aug 21 01:32:03 2008 +0200
84698
84699 Add swapped dispatch for randr 1.2 requests
84700
84701commit e02f864fdf19a5ab1682336be343c57fdb69ef43
84702Author: Adam Jackson <ajax@redhat.com>
84703Date: Wed Aug 20 13:24:03 2008 -0400
84704
84705 Suppress cursor display until the first XDefineCursor() request.
84706
84707 Yes, this means the server will start without showing a cursor. Pretty
84708 much any application that wants to interact with the mouse will define
84709 cursors, so this essentially just delays showing it until gdm (or
84710 whatever) loads.
84711
84712commit 64ef7ed072007b1d0b4de5ff1e5eababa418c794
84713Author: Adam Jackson <ajax@redhat.com>
84714Date: Wed Aug 20 13:14:03 2008 -0400
84715
84716 Centralize declaration of ConnectionInfo.
84717
84718commit f1f44940f10b0209946ec0f08104a372d7d945a0
84719Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84720Date: Wed Aug 20 09:50:54 2008 -0700
84721
84722 XQuartz: Support windowItemModMask=0 or -1 to disable window shortcuts.
84723 (cherry picked from commit 81187364e512606a6aebb95ab2967d9d420f57f2)
84724 (cherry picked from commit f2ec79e4a159dec6481691e4dd615db01770dd7a)
84725
84726commit 6c5612c2e1df1e89388ffff97251f5328558f8e8
84727Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84728Date: Wed Aug 20 09:21:44 2008 -0700
84729
84730 XQuartz: Added window_item_modifiers defaults item (and option to localization) to change the modifier keys used for the windows menu.
84731 (cherry picked from commit e4110861d307a55d9032cb83cf024ec1f294e8d7)
84732 (cherry picked from commit fcfc05482a88696cb713c7a2ddfcf935ae8378a6)
84733
84734commit bdc277c9b3dc6b3b95f74b20a49ce163c5fae980
84735Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84736Date: Wed Aug 20 09:58:48 2008 -0700
84737
84738 XQuartz: Just cleaned up formatting of event processing code... no "meat" changed...
84739 (cherry picked from commit 745bc8ab387d6794f47d8b9dca33b4c81f6dd39c)
84740 (cherry picked from commit a28a2be52478a1557a363140f7bd70ececf144dc)
84741
84742commit ed42108920cc695b96ed57bf9dcfea1470e74669
84743Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84744Date: Tue Aug 19 02:02:11 2008 -0700
84745
84746 XQuartz: Added appkit_modifiers defaults option which users can set to 'rcommand' to get access to the input menu with right command
84747 (cherry picked from commit 02af74d7a5a7225f408915254c40856159dc7f19)
84748 (cherry picked from commit 2da32894dff8340f6ca7c980277fca7ec835a193)
84749
84750commit be06961312e2111778bfd97b2c8a927acc61f43d
84751Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84752Date: Wed Aug 20 09:56:06 2008 -0700
84753
84754 XQuartz: More input fixes
84755 stuck-modifier fixes (capslock)
84756 3button-emulation now doesn't send the modifier key with the click
84757 Added other options to fake_button2 and fake_button3 defaults options:
84758 ({l,r}{control,alt,command,shift})
84759 (cherry picked from commit 8fb6a1cf44c35a20dfb0c05c9824f94898f88da7)
84760 (cherry picked from commit ae9c1b3cfb3874b4d1251681c24bda91c398bcab)
84761
84762commit 436b659091dd17bb3e316377ba31f5d2a5dfa12e
84763Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
84764Date: Thu Aug 14 09:34:12 2008 -0700
84765
84766 XQuartz: Updated bundle version to 2.3.1
84767 (cherry picked from commit eedecba0b882bb07931e8d9168589f5be7a08a69)
84768 (cherry picked from commit 7d9d864461858b25c5eb5858fdf4b53771795f11)
84769
84770commit bc4cef1d0ae7cbb8f5fe16d576db45abd4330370
84771Author: Maarten Maathuis <madman2003@gmail.com>
84772Date: Mon Aug 18 16:28:33 2008 +0200
84773
84774 exa: remove useless cache{X,Y}off from UploadToScreen() arguments.
84775
84776commit 38bdc34d555f956a1ff1e1bb8eef95e3ba2bac64
84777Author: Adam Jackson <ajax@redhat.com>
84778Date: Tue Aug 19 15:12:39 2008 -0400
84779
84780 Remove redundant check.
84781
84782commit 47b31233c26f710be70ffcca17e5402d03d5733f
84783Author: Adam Jackson <ajax@redhat.com>
84784Date: Tue Aug 19 15:10:12 2008 -0400
84785
84786 Simplify driver setup.
84787
84788 No point warning about missing driver hooks, that just means the person
84789 who gave you the driver is inept. Might as well just crash. Also,
84790 just name anonymous screens as screen%d instead of failing after the 36th
84791 screen. Bonus points if you can figure out what the failure mode would
84792 be on the 36th screen, and what the effective screen limit was.
84793
84794commit b99fc65b53769400821a2bd8f691ece35a5b8847
84795Author: Adam Jackson <ajax@redhat.com>
84796Date: Tue Aug 19 15:06:53 2008 -0400
84797
84798 Simplify calling xf86EnableIO() a bit.
84799
84800commit b1f2c7a89ace1f6e5750783f3c43c4c5a6d24374
84801Author: Adam Jackson <ajax@redhat.com>
84802Date: Tue Aug 19 15:06:11 2008 -0400
84803
84804 The ->Identify hook is not mandatory.
84805
84806 Best comment ever though. Mad props.
84807
84808commit af5125968a738ec8f496d04eb65acb7c8f812115
84809Author: Adam Jackson <ajax@redhat.com>
84810Date: Tue Aug 19 14:47:10 2008 -0400
84811
84812 Refactor misprite a bit.
84813
84814 No variable declarations in header files, please.
84815
84816commit 41b68e0dea9305d66bca2fc4ad96db01f5342c6d
84817Author: Adam Jackson <ajax@redhat.com>
84818Date: Tue Aug 19 11:02:31 2008 -0400
84819
84820 Remove unused -co option.
84821
84822 What did this even do anyway?
84823
84824commit 1f416fba994ed7a7e072a9f0a86b515855ea3bac
84825Author: Adam Jackson <ajax@redhat.com>
84826Date: Tue Aug 19 10:56:45 2008 -0400
84827
84828 Remove unused -bestRefresh option.
84829
84830commit 5a72c45d42abc7227c6cf3d14fd7043ea7527c54
84831Author: Adam Jackson <ajax@redhat.com>
84832Date: Tue Aug 19 10:54:11 2008 -0400
84833
84834 Remove unused -showunresolved option
84835
84836commit 9f08ffc557b52e2e8cd54fb692b66700c83d61c6
84837Author: Adam Jackson <ajax@redhat.com>
84838Date: Tue Aug 19 10:29:23 2008 -0400
84839
84840 Remove unused xf86Info.pMouse
84841
84842commit f227fbf74f0c619ecf3275cdb4c10b1a4b3a8cfc
84843Author: Michel Dänzer <michel@tungstengraphics.com>
84844Date: Tue Aug 19 11:36:12 2008 +0200
84845
84846 EXA: Inline Prepare/FinishAccessWindow into only caller, ChangeWindowAttributes.
84847
84848 Also check the requested mask in addition to the GC state before doing work.
84849
84850commit 825b3fe11d1b813bf8d5b24a880ed04b78ae1acf
84851Author: Michel Dänzer <michel@tungstengraphics.com>
84852Date: Tue Aug 19 11:22:40 2008 +0200
84853
84854 EXA: Don't use exaGlyphs if the driver doesn't provide a PrepareComposite hook.
84855
84856 It's buggy without Composite acceleration (leading to cropped glyphs) and not
84857 really useful in that case anyway. The bug probably still needs to be found and
84858 fixed for drivers that provide a PrepareComposite hook but can't accelerate
84859 text rendering though.
84860
84861commit 75e495a4cd823b6fa42a8d167ecc9f4723320dea
84862Author: Adam Jackson <ajax@redhat.com>
84863Date: Mon Aug 18 17:58:08 2008 -0400
84864
84865 Remove unused xf86Info.mouseLocal
84866
84867commit b3dfd7e3e4deef21b40c885f8e7eae34b55888b2
84868Author: Adam Jackson <ajax@redhat.com>
84869Date: Mon Aug 18 17:56:18 2008 -0400
84870
84871 Remove unused xf86Info.sharedMonitor
84872
84873commit 5e43cd28692bc05cac80f38b47104a26c0524385
84874Author: Adam Jackson <ajax@redhat.com>
84875Date: Mon Aug 18 17:46:42 2008 -0400
84876
84877 Remove the remainder of grab deactivation and closedown.
84878
84879commit cd1e8f26147919227e7624ac4c6b313d972a4d35
84880Author: Adam Jackson <ajax@redhat.com>
84881Date: Mon Aug 18 17:40:46 2008 -0400
84882
84883 Remove unused EstimateSizesAggressively option.
84884
84885commit faaf0046155abbd15415d5a6b62ead4f58935c28
84886Author: Mathieu Bérard <mathieu.berard@crans.org>
84887Date: Mon Aug 18 17:23:26 2008 -0400
84888
84889 Remove xaaTEGlyphBlt.S
84890
84891 Should have been removed in 593144dddd977f53bcd1a115f9544eeece46df4c but
84892 ajax was asleep at the wheel.
84893
84894commit 562462e78dd6a5e2ed5b36aa22be52f31d55485d
84895Author: Mathieu Bérard <mathieu.berard@crans.org>
84896Date: Mon Aug 18 17:05:24 2008 -0400
84897
84898 Remove dead {bsd,lnx}Resource.c
84899
84900commit e3e47b35e369bf4abb0fe15865acca0b34600b1f
84901Author: Adam Jackson <ajax@redhat.com>
84902Date: Mon Aug 18 17:03:22 2008 -0400
84903
84904 Remove MAX_PCI_{BUSES,DOMAINS} and some associated #if 0.
84905
84906commit 593144dddd977f53bcd1a115f9544eeece46df4c
84907Author: Mathieu Bérard <mathieu.berard@crans.org>
84908Date: Mon Aug 18 16:46:36 2008 -0400
84909
84910 Remove some unused assembly code and assyntax.h
84911
84912commit 2d3e478384ef8af2a760f39792e35e4a6174c967
84913Author: Adam Jackson <ajax@redhat.com>
84914Date: Mon Aug 18 15:09:44 2008 -0400
84915
84916 int10: Fix int1A for %ax == 0xB101.
84917
84918 pciNumBuses was pretty much always 1 post-pciaccess. That ain't good.
84919
84920commit 539717fe1f08a000d1eef9f345d24cb49e3663b5
84921Author: Adam Jackson <ajax@redhat.com>
84922Date: Mon Aug 18 14:50:33 2008 -0400
84923
84924 Remove (unused) pciMaxBusNum.
84925
84926commit 3c03d9f1efbbacec6e8be58da99bf0977a8e0fec
84927Author: Adam Jackson <ajax@redhat.com>
84928Date: Mon Aug 18 14:37:42 2008 -0400
84929
84930 Remove sparcPci.c
84931
84932 There is no way this code can have been building for anyone since pciaccess
84933 was merged. BSD and Linux were already using OS code on sparc, the only
84934 people who could want this are Solaris, who should be using pciaccess
84935 anyway.
84936
84937commit 9a39b6cfbda306eb594e87ce828afa01cd7aa01e
84938Author: Adam Jackson <ajax@redhat.com>
84939Date: Mon Aug 18 14:26:01 2008 -0400
84940
84941 Remove XF86SCANPCI_WRAPPER.
84942
84943commit 470b05a610e641094d46742393684acd786d78a8
84944Author: Adam Jackson <ajax@redhat.com>
84945Date: Mon Aug 18 14:03:36 2008 -0400
84946
84947 Remove vestigial MAX_PCI_DEVICES.
84948
84949commit 244a635fcdc9e0a7212d51b26d74f49d8e1b071f
84950Author: Fredrik Höglund <fredrik@kde.org>
84951Date: Mon Aug 18 19:27:34 2008 +0200
84952
84953 Fix the tile offset in miPaintWindow for ParentRelative windows.
84954
84955commit fdf7c747a8e1bc59dfb31b7f90dd5eab2687315b
84956Author: Adam Jackson <ajax@redhat.com>
84957Date: Mon Aug 18 13:13:03 2008 -0400
84958
84959 Remove PciProbeType and associated weirdness.
84960
84961 This code was effectively only used in ix86Pci.c to select PCI config
84962 access type. Nobody should be using that path anymore, in the glorious
84963 pciaccess world; kernel services should get it right for you.
84964
84965commit 95bb6f53624a3e6f4d62a2f789982c5544d2fc70
84966Author: Mathieu Bérard <mathieu.berard@crans.org>
84967Date: Mon Aug 18 11:47:19 2008 -0400
84968
84969 Remove ancient unused inline hack.
84970
84971commit e1ae8db625b5e8c298a557592ef23656b3da4886
84972Author: Adam Jackson <ajax@redhat.com>
84973Date: Mon Aug 18 09:53:48 2008 -0400
84974
84975 Drop dead PowerPC PCI code.
84976
84977 Also add a warning for anyone still using legacy PCI code.
84978
84979commit 99583b43a9a202d047ff417d47485e4c0e0c9670
84980Author: Jie Luo <clotho67@gmail.com>
84981Date: Sun Aug 17 23:13:22 2008 +0200
84982
84983 glx: avoid possible NULLptr deref, fix #16884
84984
84985commit 421b7e8f12083e9518fa7deda968a2f73c9a0006
84986Author: Maarten Maathuis <madman2003@gmail.com>
84987Date: Sun Aug 17 19:57:02 2008 +0200
84988
84989 exa: fix assert logic thinko from 361a9eb953aaa38f8ebc057185de29e50f9eef26
84990
84991 - I guess failing PrepareAccess is rare, since this a 3 year old bug.
84992
84993commit 1e62e773273539352bc21b2da5262678dd3fc7df
84994Author: Peter Hutterer <peter.hutterer@redhat.com>
84995Date: Fri Aug 15 15:19:44 2008 +0930
84996
84997 Require inputproto 1.9.99.4
84998
84999commit 609c04b62bdb0e86f33860069033c7142eecf210
85000Author: Peter Hutterer <peter.hutterer@redhat.com>
85001Date: Fri Aug 15 14:51:59 2008 +0930
85002
85003 dix: stick a warning in about loss of subpixel precision.
85004
85005commit de1573172ad2c2e98c101727930a468213295783
85006Author: Peter Hutterer <peter.hutterer@redhat.com>
85007Date: Wed Aug 13 13:11:44 2008 +0930
85008
85009 Backport device properties to XI 1.5 instead of XI 2.0
85010
85011commit 8b9ed7dec2954d4890ddcc7c874fc0832b16495a
85012Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85013Date: Thu Aug 14 09:08:14 2008 -0700
85014
85015 XQuartz: Fixed a stuck modifier key bug.
85016 (cherry picked from commit eeb6e5a9e98dcf045ec230f160d5992080dceba6)
85017
85018commit 1770c85374883229ec5e9685229b2333e62705e6
85019Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85020Date: Thu Aug 14 07:32:37 2008 -0700
85021
85022 XQuartz: Made 3-button mouse simulation a little more consistent.
85023 (cherry picked from commit d207b037d2ae213369e5627a17d8831c9bc16ad8)
85024
85025commit f51f77d25e69b51fa3fd557ffdb5573b61759706
85026Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85027Date: Wed Aug 13 19:09:05 2008 -0700
85028
85029 XQuartz: Reverted "Control" text to be accurately "Command" in the input prefs.
85030 (cherry picked from commit b287c481e1cbe77ff84d10a708505f148a0c8434)
85031
85032commit 04956b80431169e0ae713a3e6ba4cdc157ce3a66
85033Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85034Date: Sat Aug 9 10:07:32 2008 -0700
85035
85036 Transport/Apple: Add TRANS_NOXAUTH to incoming connections when the listener is flagged as such
85037 (cherry picked from commit ba4a7924261070ad0aff7211b7a1c9581fad4646)
85038
85039commit e6813e8de65eee854bbffe6ab0f8ba158f43b10b
85040Author: Peter Hutterer <peter.hutterer@redhat.com>
85041Date: Wed Aug 13 14:24:45 2008 +0930
85042
85043 Xi: byte-swap device property requests.
85044
85045commit 0ce687634fa5e0d955e896a0d0d5f584a4cfd1a4
85046Author: Adam Jackson <ajax@redhat.com>
85047Date: Wed Aug 13 17:28:08 2008 -0400
85048
85049 Eviscerate save-under support.
85050
85051 Use a compositing manager already. Plus I really wanted to use 'eviscerate'
85052 in a commit message.
85053
85054commit f456f32cd97684052cab4712e7f4c827132c554b
85055Author: Adam Jackson <ajax@redhat.com>
85056Date: Wed Aug 13 16:21:28 2008 -0400
85057
85058 Build fix.
85059
85060commit c1e9b7dc6f4600fe6c301aecd7d67461cb884df0
85061Author: Adam Jackson <ajax@redhat.com>
85062Date: Wed Aug 13 14:13:55 2008 -0400
85063
85064 Sledgehammer off the DRI2 build system.
85065
85066commit 544bfd06663ff71bc5275459cf74a3557530544f
85067Author: Adam Jackson <ajax@redhat.com>
85068Date: Wed Aug 13 11:09:54 2008 -0400
85069
85070 Remove xf86Version.h and related API.
85071
85072 We haven't meaningfully been API-compatible with xf86 modules in ages,
85073 let's stop pretending.
85074
85075commit 139c3ab7b754dc425a09d5f7b6d2a8fb2b88138a
85076Author: Peter Hutterer <peter.hutterer@redhat.com>
85077Date: Tue Aug 12 17:49:49 2008 +0930
85078
85079 Xi: return the state from the correct keyboard in QueryDevicePointer.
85080
85081commit 998375f4154b000f75b2bafd4e276c0237d24dc7
85082Author: Peter Hutterer <peter.hutterer@redhat.com>
85083Date: Tue Aug 12 17:49:16 2008 +0930
85084
85085 dix: remove superfluous check.
85086
85087commit 9f9268821b13038556fbc029df54ab0e9b2aa77f
85088Author: Mathieu Bérard <mathieu.berard@crans.org>
85089Date: Mon Aug 11 13:52:38 2008 -0400
85090
85091 The smart scheduler is not optional.
85092
85093commit 2e2ce817ce404a5e000c9750fa96f656fed370b8
85094Author: Matthieu Herrb <matthieu.herrb@laas.fr>
85095Date: Sun Aug 10 23:07:46 2008 +0200
85096
85097 Move strcasecmp(), strcasencmp() and strcasestr() prototypes to os.h
85098
85099 And make sure os.h is included in files that use it.
85100
85101commit 02efa78ce2da470b252289ff2af598d06bc84ece
85102Author: Matthieu Herrb <matthieu.herrb@laas.fr>
85103Date: Sun Aug 10 18:29:35 2008 +0200
85104
85105 add libXinerama to the list of dependencies.
85106
85107 Now that panoramiXext.h is installed bt libXinerama, it becomes required
85108 to build Xext.
85109
85110commit be2210b69defa225d8bc4bb90d94bfd740ef8b1f
85111Author: Matthieu Herrb <matthieu.herrb@laas.fr>
85112Date: Sat Aug 9 23:43:50 2008 +0200
85113
85114 remove OpenBSD/amd64 specific mtrr API.
85115
85116 The old code never worked anyways and was removed from OpenBSD.
85117 OpenBSD/amd64 4.4 and later support mtrr with the same API as
85118 OpenBSD/i386.
85119
85120commit 6e33e6f355f7f04e77a165eb67b1414724c1fba3
85121Author: Matthieu Herrb <matthieu.herrb@laas.fr>
85122Date: Sat Aug 9 23:43:03 2008 +0200
85123
85124 Move string comparaison functions to from dix/ to os/.
85125
85126commit 5968634996c08656a0c5e2fa35705cf7afac87e6
85127Author: Dave Miller <davem@davemloft.net>
85128Date: Sat Aug 9 16:45:59 2008 +1000
85129
85130 glx: fix crash in driGetConfigAttribIndex
85131
85132 Don't access GLboolean via int pointers
85133
85134commit 4212599c922373a224d2235c74672a3a3aa8e0b1
85135Author: Michel Dänzer <michel@tungstengraphics.com>
85136Date: Fri Aug 8 12:17:58 2008 +0200
85137
85138 EXA: Make sure damage tracking code is inactive if the driver manages pixmaps.
85139
85140 It was always supposed to be like that... It was only recently pointed out (in
85141 a rather convoluted way) that it was not in fact the case.
85142
85143commit 073116cc44859e96374cde46325df8540621d5ee
85144Author: Michel Dänzer <michel@tungstengraphics.com>
85145Date: Fri Aug 8 12:15:27 2008 +0200
85146
85147 Remove commented out lines that make automake 1.9 puke.
85148
85149commit 666781cd9b62207e45eebd3eb55a930f3a345f72
85150Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85151Date: Fri Aug 8 01:41:28 2008 -0700
85152
85153 Removed rogue #include
85154 (cherry picked from commit af238e99da45a4dd8d8cbb6564bb8d0a3d79d590)
85155
85156commit 4a0947bf06deacc3ae716730cfbbbeda1311bae3
85157Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85158Date: Fri Aug 8 01:53:36 2008 -0700
85159
85160 XQuartz: Disable the Xquartz AIGLX for now since it doesn't even compile any more...
85161 (cherry picked from commit 4545ba91e9d6ad62b6cafde3c73a672d3ded91a2)
85162
85163commit 5accc9b3ce1af539150d59b1800b66783fd3e947
85164Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85165Date: Fri Aug 8 00:19:41 2008 -0700
85166
85167 XQuartz: Re-added deprecated code fallback failsafe for keyboard layout on Leopard with some debugging spew.
85168 (cherry picked from commit 5854e712e9ebc210d2f8de6f5d4fb650944f314a)
85169
85170commit 465a9bb96e8a90e53e655b3245e22910201786d7
85171Author: Peter Hutterer <peter.hutterer@redhat.com>
85172Date: Fri Aug 8 17:37:06 2008 +0930
85173
85174 dix: don't try to create events if we don't have a screen. #16898
85175
85176 A NULL screen may happen during server shutdown, when the output has been shut
85177 down but the devices still generate events.
85178
85179 X.Org Bug 16898 <http://bugs.freedesktop.org/show_bug.cgi?id=16898>
85180
85181commit d684f5760f40b682a8b879641300abe689a263c5
85182Author: Peter Hutterer <peter.hutterer@redhat.com>
85183Date: Fri Aug 8 14:18:01 2008 +0930
85184
85185 xkb: actually initialise sli before using it.
85186
85187commit e00cd54c1ac57ea6e7767e90592200f343d9d08a
85188Author: Simon Thum <simon.thum@gmx.de>
85189Date: Thu Aug 7 11:06:21 2008 +0200
85190
85191 dix: remove misleading comment in ptrvelo.c
85192
85193 Signed-off-by: Peter Hutterer <peter.hutterer@redhat.com>
85194
85195commit 6ca34549b89e73e3e0ec3cb7585686b880534cef
85196Author: Peter Hutterer <peter.hutterer@who-t.net>
85197Date: Thu Aug 7 16:27:57 2008 +0930
85198
85199 dix: remove obsolete FIXME comment.
85200
85201 Thanks to Simon Thum for pointing this out.
85202
85203commit 2c3645581ee3f180d34e32c8016d5e4e1af4dca4
85204Author: Peter Hutterer <peter.hutterer@who-t.net>
85205Date: Thu Aug 7 11:55:10 2008 +0930
85206
85207 dix: SetModifierMapping should only apply to the ClientPointer and it's SDs.
85208
85209commit e1b286d495b760cf67ddf936eca11da50fba847a
85210Author: Simon Thum <simon.thum@gmx.de>
85211Date: Wed Aug 6 09:42:32 2008 +0200
85212
85213 dix: simplified debugging helper for pointer acceleration
85214
85215 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85216
85217commit d95136e1510a48cc3ad066d880d56061fc6a42ee
85218Author: Simon Thum <simon.thum@gmx.de>
85219Date: Thu Jul 31 00:39:03 2008 +0200
85220
85221 xf86: don't replicate dix defaults for pointer acceleration
85222
85223 the defaults from InitVelocityData() or hypothetic driver-side changes
85224 are now respected, not overridden.
85225
85226 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85227
85228commit 4f333d53510b82db57cfac1bfea22422cbb7be79
85229Author: Simon Thum <simon.thum@gmx.de>
85230Date: Thu Jul 31 00:33:38 2008 +0200
85231
85232 dix: simplify velocity tracking filters
85233
85234 prefer fp-mul over fp-div and remove rather pointless check
85235
85236 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85237
85238commit bf084a0769eee36ff799e5e5f2d1e875c1ebcc51
85239Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85240Date: Wed Aug 6 13:36:43 2008 -0700
85241
85242 glcontextmodes.[hc] were not added with the removal of the meas symlinks patch. Copied from mesa head 2008.08.06.
85243 (cherry picked from commit 409e1dd1e9524b5c1a1ae58a759da77e587e3780)
85244
85245commit e72b1d21e44712e90595b1c31e67db6786d0d1bd
85246Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85247Date: Wed Aug 6 10:23:31 2008 -0700
85248
85249 XQuartz: Now properly disable xauth checking on launchd socket and mostly fix the xinitrc / launchd race condition
85250
85251commit d1b96cdea96f5321e0d254abde81cdaab2eca979
85252Author: Aaron Plattner <aplattner@nvidia.com>
85253Date: Wed Aug 6 10:37:32 2008 -0700
85254
85255 Add an xf86PrintBacktrace wrapper around xorg_backtrace to aid driver debugging.
85256
85257commit 26d31ad1c7f4c550d73419ecf76912d844186b30
85258Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85259Date: Tue Aug 5 15:14:08 2008 -0700
85260
85261 XQuartz: Added code and made comments more helpful for debugging first-client-auth bug.
85262 (cherry picked from commit a8f0d32216e321b8ae6da182be9b1ea792f6e004)
85263
85264commit 0ef3243fb59fa993a7a07a08debbd4329254c265
85265Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85266Date: Tue Aug 5 18:45:54 2008 -0700
85267
85268 XQuartz: More fd handoff cleanup.
85269 (cherry picked from commit 199d2dcb708c9f9d6b9ede149ea0fe9fba6c007c)
85270
85271commit e7dd3dd8f311d3872e0c4c9c2ca22c00f9d11f59
85272Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85273Date: Tue Aug 5 14:33:03 2008 -0700
85274
85275 XQuartz: Fixed some errors / typos in the preferences pane.
85276 (cherry picked from commit e1e0c398bbb50394b164394e1f1870016489c25b)
85277
85278commit 6836f9aac801c3add0ae4aece05b32e85a0c5f85
85279Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85280Date: Tue Aug 5 14:24:25 2008 -0700
85281
85282 XQUartz: UI Cleanup. Removed done/cancel buttons from Applications->Customize
85283 (cherry picked from commit d3157ca45c1f88edd040d721c1e944b9d2c090aa)
85284
85285commit c06e27b2f6fd9f7b9f827623a48876a225264132
85286Author: Peter Hutterer <peter.hutterer@who-t.net>
85287Date: Fri Aug 1 16:42:15 2008 +0930
85288
85289 xkb: ProcXkbSetDeviceInfo should work on all attached SDs.
85290
85291 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85292 call. This way, we keep the SDs in sync with the MD as long as core clients
85293 control the MDs.
85294
85295commit d9ca9819e975e0f6832a320f8be5958e5d942f85
85296Author: Peter Hutterer <peter.hutterer@who-t.net>
85297Date: Fri Aug 1 16:41:40 2008 +0930
85298
85299 xkb: ProcXkbSetGeometry should work on all attached SDs.
85300
85301 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85302 call. This way, we keep the SDs in sync with the MD as long as core clients
85303 control the MDs.
85304
85305commit 5ba87c3327786dd7c6e8e265a19c858e8faae8fd
85306Author: Peter Hutterer <peter.hutterer@who-t.net>
85307Date: Fri Aug 1 16:41:08 2008 +0930
85308
85309 xkb: ProcXkbSetNames should work on all attached SDs.
85310
85311 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85312 call. This way, we keep the SDs in sync with the MD as long as core clients
85313 control the MDs.
85314
85315commit 7e45c80204e06562d4475741caea65bc8758f3c7
85316Author: Peter Hutterer <peter.hutterer@who-t.net>
85317Date: Fri Aug 1 16:40:25 2008 +0930
85318
85319 xkb: ProcXkbSetNamedIndicator should work on all attached SDs.
85320
85321 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85322 call. This way, we keep the SDs in sync with the MD as long as core clients
85323 control the MDs.
85324
85325commit a609dbed7cf854a6ee9d33c7cf45615db9da1fb9
85326Author: Peter Hutterer <peter.hutterer@who-t.net>
85327Date: Fri Aug 1 16:38:59 2008 +0930
85328
85329 xkb: ProcXkbSetIndicatorMap should work on all attached SDs.
85330
85331 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85332 call. This way, we keep the SDs in sync with the MD as long as core clients
85333 control the MDs.
85334
85335commit e8c2a3d7c996cb41c4c44ba67acae5ff9438fc06
85336Author: Peter Hutterer <peter.hutterer@who-t.net>
85337Date: Fri Aug 1 16:37:39 2008 +0930
85338
85339 xkb: ProcXkbSetCompatMap should work on all attached SDs.
85340
85341 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85342 call. This way, we keep the SDs in sync with the MD as long as core clients
85343 control the MDs.
85344
85345commit 3c7740aa8f21c1fda4190a1bf5d6ce089db5e35b
85346Author: Peter Hutterer <peter.hutterer@who-t.net>
85347Date: Fri Aug 1 16:36:00 2008 +0930
85348
85349 xkb: ProcXkbSetMap should work on all attached SDs.
85350
85351 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85352 call. This way, we keep the SDs in sync with the MD as long as core clients
85353 control the MDs.
85354
85355commit 31afd51dd49c0d0db2465fbc987044fab8b89f22
85356Author: Peter Hutterer <peter.hutterer@who-t.net>
85357Date: Fri Aug 1 16:32:37 2008 +0930
85358
85359 xkb: ProcXkbBell should work on all attached SDs.
85360
85361 If called with XkbUseCoreKbd, run through all attached SDs and replicate the
85362 call. This way, we keep the SDs in sync with the MD as long as core clients
85363 control the MDs.
85364
85365commit de4936d7482f820728efeef338a2041c7a9186d2
85366Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85367Date: Mon Aug 4 20:11:45 2008 -0700
85368
85369 XQuartz: Fixed proper processing of tablet button presses
85370 (cherry picked from commit ca0babafa4e9cf1b67ec460655bffe569ac9c3e9)
85371
85372commit b70a8ba3dc52b277194ab267a101a3fd15062685
85373Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85374Date: Tue Jul 15 17:48:29 2008 -0700
85375
85376 XQuartz: Nuke Sparkle.
85377 (cherry picked from commit 4991f54a8d84a9a8df89a99dbfc09391195578f8)
85378
85379commit 30851efdd4313506e92aeb6d3e5099dea99ab99c
85380Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
85381Date: Tue Aug 5 12:54:51 2008 -0700
85382
85383 XQuartz: Fixed first-client-can't-connect bug
85384 Readded the old exec() server startup path for regression testing.
85385 Don't use the dynamic fd addition code since it's not quite working correctly.
85386 (cherry picked from commit 08f3fe153edc5ab4ca010e8ce82d5c3fc0ddb72c)
85387
85388commit 277a74bcbb7d0a93c4f2e1de11daabd8c5f93ee8
85389Author: Tomas Carnecky <tom@dbservice.com>
85390Date: Mon Aug 4 23:26:00 2008 +0300
85391
85392 Redefine clients as a fixed array
85393
85394 This removes yet another xalloc() each server generation. Also, I
85395 couldn't find the corresponding xfree() so I guess that used to be a
85396 memory leak there.
85397
85398commit 5532d63488ec45953ff7f925cfb4f87adb3b04a0
85399Author: Tomas Carnecky <tom@dbservice.com>
85400Date: Mon Aug 4 23:06:08 2008 +0300
85401
85402 Redefine WindowTable as a fixed array
85403
85404 Instead of xalloc'ing it every server generation. The array is always
85405 the same size (MAXSCREENS), anyway.
85406
85407commit e882ee7056f370e0619d137b4ec3973ecb4e3479
85408Author: Daniel Stone <daniel@fooishbar.org>
85409Date: Tue Jul 29 17:52:11 2008 +0300
85410
85411 EXA: Remove unnecessary #includes
85412
85413 There wasn't actually any font code here, so no problem.
85414
85415commit 6c1bb64c6f2efd81856ca00591df37916713b142
85416Author: Peter Hutterer <peter.hutterer@who-t.net>
85417Date: Mon Aug 4 15:19:45 2008 +0930
85418
85419 xkb: break up XkbCopyKeymap into bite-sized chunks.
85420
85421commit 3c6a9c531f673b7a0cb9ca01860b4dbe79686363
85422Author: Peter Hutterer <peter.hutterer@who-t.net>
85423Date: Fri Aug 1 15:52:07 2008 +0930
85424
85425 config: protect against potential out-of-bounds indexing.
85426
85427commit 92c51b183c2ff06361dad7f918daed6577ba4935
85428Author: Peter Hutterer <peter.hutterer@who-t.net>
85429Date: Fri Aug 1 14:24:54 2008 +0930
85430
85431 config: support type strlist for XkbOptions property.
85432
85433 For backwards compatibility with server 1.4.
85434
85435commit 35b14519b4a3158592a089170ec039bbc219603e
85436Author: Peter Hutterer <peter.hutterer@who-t.net>
85437Date: Tue Jul 29 12:59:57 2008 +0930
85438
85439 config: add parsing for input.x11_options.XkbOptions. #16874
85440
85441 X.Org Bug 16874 <http://bugs.freedesktop.org/show_bug.cgi?id=16784>
85442
85443commit d762c08aebe3b7e8c88e2e7a6fcf66057a21b403
85444Author: Simon Thum <simon.thum@gmx.de>
85445Date: Mon Jul 28 14:07:48 2008 +0200
85446
85447 dix: export driver-side functions for acceleration
85448
85449 also add additional safety for accel driver api
85450
85451 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85452
85453commit 18e9fd69fe01298d825b46415b9c6bd86c75dfe5
85454Author: Simon Thum <simon.thum@gmx.de>
85455Date: Tue Jul 29 10:07:43 2008 +0200
85456
85457 dix: use average of pointer accel profile
85458
85459 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85460
85461commit 25882af6d3359e5ae42c927c555f5b257ba5665c
85462Author: Ivaylo Boyadzhiev <iboyadzhiev@nvidia.com>
85463Date: Sun Aug 3 18:55:12 2008 -0700
85464
85465 Fix a longstanding XAA CopyPlane bug.
85466
85467 TmpBitPlane is a plane mask, not a plane index.
85468
85469 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
85470
85471commit d980913f3145cdc34baab27ff818c9631c4c8571
85472Author: Adam Jackson <ajax@redhat.com>
85473Date: Fri Aug 1 16:15:04 2008 -0400
85474
85475 Erk. Actually check all the BARs, not just the first.
85476
85477commit 82d51e6df2bf677bdf24376092bcaa79b534f6e5
85478Author: Adam Jackson <ajax@redhat.com>
85479Date: Fri Aug 1 13:17:55 2008 -0400
85480
85481 Silence the "No matching Device section" warning in some harmless cases.
85482
85483 If the device doesn't have any BARs then it's just a stub for some
85484 lame operating systems that need one PCI device per output for
85485 multihead. No point in warning about it.
85486
85487commit 9643e8d3482a35d355a243db7aa397ad47f29be0
85488Author: Adam Jackson <ajax@redhat.com>
85489Date: Fri Aug 1 11:35:47 2008 -0400
85490
85491 Handle XGI cards in autoconfig.
85492
85493 It's all a bit wonky since both sis(4) and xgi(4) claim to support the
85494 Volari Z7 and V5/8 (0x0020 and 0x0040), so let's side with xgi(4), why
85495 not. Note that the V3 (not V3XT) identifies itself as a trident chip.
85496
85497commit 63bdd4c27d47323b3282bf0b6eaecae91c79b45c
85498Author: Adam Jackson <ajax@redhat.com>
85499Date: Thu Jul 31 17:31:36 2008 -0400
85500
85501 Unifdef XIDLE.
85502
85503 Seriously how was this still here.
85504
85505commit 8d214bc26f9b7ab6a5c54d7749cd4b6811cb0b96
85506Author: Adam Jackson <ajax@redhat.com>
85507Date: Thu Jul 31 15:46:52 2008 -0400
85508
85509 Document more of the OS and library assumptions.
85510
85511commit 2198e237b2ed85857c671eee2cd04dfc032befee
85512Author: Tomas Carnecky <tom@dbservice.com>
85513Date: Thu Jul 31 13:46:06 2008 -0400
85514
85515 Cosmetic cleanup to Xvfb/Xdmx configure check output.
85516
85517commit 148175fb8b365dcf00b13539b3f03ce33f3df707
85518Author: Michel Dänzer <michel@tungstengraphics.com>
85519Date: Thu Jul 31 16:54:33 2008 +0200
85520
85521 EXA: Do still return FALSE if the driver PrepareCopy hook failed...
85522
85523 Thanks to Stuart Bennett for pointing out the problem on IRC.
85524
85525commit 68fd6604a15b653d1fa244633eaaee9608a03ab3
85526Author: Tomas Carnecky <tom@dbservice.com>
85527Date: Thu Jul 31 10:03:56 2008 -0400
85528
85529 OutputPropertyNotifyMask is a valid thing to select for.
85530
85531commit 64ebeeb5265a4c425b9397fdc86a6d81521a856e
85532Author: Michel Dänzer <michel@tungstengraphics.com>
85533Date: Thu Jul 31 10:58:52 2008 +0200
85534
85535 EXA: Fix exponential growth logic for GXcopy tiled fills.
85536
85537 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=16908 .
85538
85539commit 8405c25d9ddbfddb6b155a436f07ccad689e53bd
85540Author: Michel Dänzer <michel@tungstengraphics.com>
85541Date: Thu Jul 31 10:55:44 2008 +0200
85542
85543 EXA: Simplify exaFillRegionTiled() control flow.
85544
85545 Also only call REGION_TRANSLATE() when necessary.
85546
85547commit b37b1e66996f8335dafc97b12d25aaec452b931e
85548Author: Aaron Plattner <aplattner@nvidia.com>
85549Date: Wed Jul 30 18:48:27 2008 -0700
85550
85551 Make shmint.h part of the SDK.
85552
85553 This includes ShmRegisterFuncs, ShmSetPixmapFormat, fbShmPutImage, and
85554 ShmRegisterFbFuncs. Note that fbShmPutImage was already exported.
85555
85556commit a3afa6f2fb80489f7b6a88d12def09281d32ed94
85557Author: Michel Dänzer <michel@tungstengraphics.com>
85558Date: Wed Jul 30 18:30:37 2008 +0200
85559
85560 EXA: Optimize GXcopy tiled fills.
85561
85562commit 37087bc10630ee7740df1369b3e56a44fd2ad2b0
85563Author: Michel Dänzer <michel@tungstengraphics.com>
85564Date: Wed Jul 30 18:27:33 2008 +0200
85565
85566 EXA: Replace open coded CopyArea logic with GC op call.
85567
85568commit bb2cacd33eb85f39b9e7fa554545611957df677d
85569Author: Xavier Bestel <xavier.bestel@free.fr>
85570Date: Tue Jul 29 13:06:10 2008 -0400
85571
85572 Cosmetic fix to EDID decoding.
85573
85574commit a70754a9fdb019d85fbfca1898699f3c6f609fdd
85575Author: Alan Hourihane <alanh@tungstengraphics.com>
85576Date: Tue Jul 29 14:16:25 2008 +0100
85577
85578 require inputproto 1.9.99.3
85579
85580commit 591ef3c047ab3597fef9d687205e99c254ff2040
85581Author: Peter Hutterer <peter.hutterer@who-t.net>
85582Date: Tue Jul 29 10:00:01 2008 +0930
85583
85584 Xi: ChangeDeviceControl presence events should set the appropriate devchange.
85585
85586 Requires inputproto 1.4.4 or higher.
85587
85588commit 25dd5ce0770c5ff91a79c12223fd4af52759900b
85589Author: Kristian Høgsberg <krh@redhat.com>
85590Date: Mon Jul 28 12:33:58 2008 -0400
85591
85592 Add atKeynames.h in hw/dmx/input and drop evil cross-ddx #include.
85593
85594commit e39a16aa58456aaecee54edc64f189c958db0902
85595Author: Julien Cristau <jcristau@debian.org>
85596Date: Sat Jul 26 16:35:19 2008 +0200
85597
85598 xfree86: don't output a spurious newline to the log when loading a module
85599
85600commit 2eaed4a10fe5bf727579bca4ab8d4a47c8763a7d
85601Author: Julien Cristau <jcristau@debian.org>
85602Date: Sat Jul 26 15:35:42 2008 +0200
85603
85604 xfree86: use xorg.conf input devices if there is no ServerLayout
85605
85606 If xorg.conf has no ServerLayout section, use the first mouse and
85607 keyboard sections as core devices, even with AllowEmptyInput.
85608
85609commit 805f28e96ceb20bc53792ae3cf17f9c26564ae0e
85610Author: Simon Thum <simon.thum@gmx.de>
85611Date: Wed Jul 23 12:10:52 2008 +0200
85612
85613 some trivial changes regarding C compat, indentation, etc.
85614
85615 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85616
85617commit 3209bd21d00b8673d321f70afb65720588ddacb3
85618Author: Simon Thum <simon.thum@gmx.de>
85619Date: Wed Jul 23 12:06:34 2008 +0200
85620
85621 xfree86: perform pointer accel scheme selection by name, not number
85622
85623 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85624
85625commit a17cb29f5acdfcdeac929d8c6be3600d44d038ef
85626Author: Simon Thum <simon.thum@gmx.de>
85627Date: Wed Jul 23 12:03:17 2008 +0200
85628
85629 dix: preparation change: make runtime exchange of filter chain safe
85630
85631 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85632
85633commit 1a9f9ac50f2b0db735789905cc29572a50c8ae4c
85634Author: Simon Thum <simon.thum@gmx.de>
85635Date: Wed Jul 23 11:49:36 2008 +0200
85636
85637 dix: optimize precision in device velocity estimation
85638
85639 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85640
85641commit c184b91d9aa72031c2bac9f379f56633957ded30
85642Author: Simon Thum <simon.thum@gmx.de>
85643Date: Wed Jul 23 11:33:25 2008 +0200
85644
85645 dix: improve the driver interface to predictable pointer acceleration
85646
85647 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85648
85649commit 87aa5298576ed335ac31347e14fb30430288157a
85650Author: Simon Thum <simon.thum@gmx.de>
85651Date: Wed Jul 23 11:28:09 2008 +0200
85652
85653 dix: introduce defines for accel profile numbers
85654
85655 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85656
85657commit 4e32e6fb38d19c9993de86188e4f7e7916a028e2
85658Author: Simon Thum <simon.thum@gmx.de>
85659Date: Wed Jul 23 11:10:22 2008 +0200
85660
85661 dix: rename classic accel _scheme_ to lightweight to avoid confusion with
85662 classic accel _profile_
85663
85664 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85665
85666commit d40183c665d458ac0a6e7952fbe986776a17fda7
85667Author: Simon Thum <simon.thum@gmx.de>
85668Date: Wed Jul 23 09:28:06 2008 +0200
85669
85670 dix: add legal statements to ptrvelo.{c|h}
85671
85672 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
85673
85674commit ec10eccd56be8b947cd63cae0687b8319857fe60
85675Author: Michel Dänzer <michel@tungstengraphics.com>
85676Date: Mon Jul 28 09:33:04 2008 +0200
85677
85678 GLX: Unreference drawables bound to the old context, not the new one.
85679
85680 Apart from the obvious reference counting issue, this fixes
85681 http://bugs.freedesktop.org/show_bug.cgi?id=16867 .
85682
85683commit 6ab8d6010adfd5ad6f1e1094a26c84f0aff934b1
85684Author: Michel Dänzer <michel@tungstengraphics.com>
85685Date: Mon Jul 28 09:32:59 2008 +0200
85686
85687 AIGLX/DRI1: Pay more attention to return value from DRIGetDrawableInfo().
85688
85689 Could have crashed otherwise if the num(Back)ClipRects variables referenced by
85690 the caller weren't pre-initialized to 0.
85691
85692commit b3e981c9d4ff31263a43b47f83cf8db4c2b5aeff
85693Author: Michel Dänzer <michel@tungstengraphics.com>
85694Date: Mon Jul 28 09:32:54 2008 +0200
85695
85696 Fix Makefile.am crack.
85697
85698 Looks like an artifact from early modularization.
85699
85700commit 3575d9584edf35ec1720bc3755b6576a56613685
85701Author: Peter Hutterer <peter.hutterer@who-t.net>
85702Date: Fri Jul 25 16:46:44 2008 +0930
85703
85704 config: note that HAL options must be strings.
85705
85706 Only strings are parsed by the server, all others are ignored. Doesn't matter,
85707 specifying int options as strings works fine anyway.
85708
85709commit 54651ff5ec54f4c621e060b19f31d77d519ef158
85710Author: Peter Hutterer <peter.hutterer@who-t.net>
85711Date: Thu Jul 24 17:41:48 2008 +0930
85712
85713 xkb: remove superfluous inputInfo.keyboard treatment.
85714
85715 Really not necessary, we can just walk the list and spare us the special
85716 treatment of the VCK.
85717
85718commit 70bd8261223366dae9dfdbb999691328d85f093c
85719Author: Peter Hutterer <peter.hutterer@who-t.net>
85720Date: Thu Jul 24 17:38:16 2008 +0930
85721
85722 xkb: use PickPointer/PickKeyboard in _XkbLookupAnyDevice.
85723
85724commit 47160edec7f0d9129576d83f1593a5549879a893
85725Author: Peter Hutterer <peter.hutterer@who-t.net>
85726Date: Thu Jul 24 12:12:45 2008 +0930
85727
85728 xfree86: warn some more about potential missing input devices.
85729
85730 Put out a warning if xorg.conf has InputDevice sections, but these aren't
85731 referenced in the used ServerLayout. This is only performed if AllowEmptyInput
85732 is enabled.
85733 The reason behind this is that the server used to auto-add the first
85734 mouse/keyboard sections if none where referenced. Now, with HAL and AEI
85735 enabled by default, setups that relied on this auto-adding break and are left
85736 without input devices. The least we can do is warn them.
85737
85738commit f30b0823dbfc5902e54b337b5b6b570ebf216584
85739Author: Peter Hutterer <peter.hutterer@who-t.net>
85740Date: Thu Jul 24 09:56:00 2008 +0930
85741
85742 xfree86: if AllowEmptyInput is on, warn the user that we rely on HAL now.
85743
85744commit ad4cd2e241691427689591f7769a1184c8c1c7f5
85745Author: Peter Hutterer <peter.hutterer@who-t.net>
85746Date: Wed Jul 23 17:40:28 2008 +0930
85747
85748 xkb: don't send core events on SlowKeys.
85749
85750 Core events don't happen until later in the DIX, so pump device events down
85751 instead. This makes modifiers work again when SlowKeys is enabled.
85752
85753commit 7d9dece74fc2bf130ceb8818ced5d9e3ac526900
85754Author: Peter Hutterer <peter.hutterer@who-t.net>
85755Date: Mon Jul 21 22:12:45 2008 +0930
85756
85757 dix: fix up enter/leave system once again.
85758
85759 Two corrections
85760 1. the "detail" field has NotifyVirtual, etc., not the "mode" field. This was
85761 a clear bug.
85762 2. don't set/unset the flags for NotifyGrab or NotifyUngrab. Clients are
85763 expected to deal with multiple enter/leave events per window if the mode is
85764 not NotifyNormal.
85765
85766 Testable with TCL menu boxes (such as used in gitk):
85767 tk_optionMenu .menu globVar Val1 Val2 Val3 ValJunk
85768 pack .menu
85769
85770 Thanks to Michel Dänzer for pointing this out.
85771
85772commit 2ce434f54bc0d6050ef115e310df62c035e84bf2
85773Author: Kristian Høgsberg <krh@redhat.com>
85774Date: Thu Jul 24 18:44:16 2008 -0400
85775
85776 Clean up unused definitions from glx headers.
85777
85778commit ae67508392261ae47858692668a0c192ef4a9d7b
85779Author: Mathieu Bérard <mathieu.berard@crans.org>
85780Date: Fri Jul 25 10:05:30 2008 -0400
85781
85782 Fix AddExtension now that CloseDownProc can be NULL.
85783
85784commit eff25430b4a391409e39337962ff7697165d23c7
85785Author: Daniel Drake <dsd@laptop.org>
85786Date: Thu Jul 24 21:06:34 2008 -0500
85787
85788 Don't abort if swrast library is not present
85789
85790 GLX is enabled by default, but the current swrast behaviour causes X
85791 to abort with fatal error if the swrast dri library dlopen fails.
85792
85793 Handle the case where the swrast library is not present, and do not
85794 register the GLX extension unless at least one screen has a usable
85795 GL provider.
85796
85797commit 62b5690415786c4a6e8bac464aa8db578beb74f5
85798Author: Adam Jackson <ajax@redhat.com>
85799Date: Thu Jul 24 16:13:10 2008 -0400
85800
85801 Remove some more MEMBUG garbage.
85802
85803commit 036822584b26854e2f486f1bea84d8d19d9324de
85804Author: Kristian Høgsberg <krh@redhat.com>
85805Date: Thu Jul 24 16:03:58 2008 -0400
85806
85807 Drop xf86CommonSpecialKey() and atKeynames.h
85808
85809 These are AT scancode specific, should really be part of xf86-input-keyboard.
85810 Remove from server, move to xf86-input-keyboard.
85811
85812commit f3ff2386016ced7e677817e9761a535f0f385813
85813Author: Adam Jackson <ajax@redhat.com>
85814Date: Thu Jul 24 15:28:07 2008 -0400
85815
85816 Sync is built-in and mandatory.
85817
85818commit 3fcf4d3eb89fecaa2be7b5ac4933b693c9c3d97e
85819Author: Adam Jackson <ajax@redhat.com>
85820Date: Thu Jul 24 15:09:05 2008 -0400
85821
85822 XC-MISC is built-in and mandatory.
85823
85824commit 9757106bba8f7bea99c5211817fc6b5fde4e6f66
85825Author: Adam Jackson <ajax@redhat.com>
85826Date: Thu Jul 24 14:59:14 2008 -0400
85827
85828 Remove all empty extension reset hooks, replace with NULL.
85829
85830commit 6bcde69585fcc8f8dbfe81c115649f19274922fa
85831Author: Adam Jackson <ajax@redhat.com>
85832Date: Thu Jul 24 09:13:26 2008 -0400
85833
85834 Allow extension closedown hook to be null.
85835
85836commit f65c50c4e902c7b5619fb7bbee8462434d809ec7
85837Author: Adam Jackson <ajax@redhat.com>
85838Date: Thu Jul 24 09:07:42 2008 -0400
85839
85840 Bigreqs are built-in and mandatory.
85841
85842commit d6228cb22aa89b90834d80d98b91862c1fc01b54
85843Author: Adam Jackson <ajax@redhat.com>
85844Date: Thu Jul 24 09:00:22 2008 -0400
85845
85846 Shape extension is built-in and mandatory.
85847
85848commit 990fc643ae90c034187707e7de414d80640ec6da
85849Author: Kristian Høgsberg <krh@redhat.com>
85850Date: Thu Jul 24 15:02:40 2008 -0400
85851
85852 Ugh, remove trailing backslash.
85853
85854commit 806e8cf3c01f955411445a7095f0957ea9a9b22a
85855Author: Kristian Høgsberg <krh@redhat.com>
85856Date: Thu Jul 24 14:25:24 2008 -0400
85857
85858 Remove SPARC muldiv code. Was used by ELF loader, which is no more.
85859
85860commit dff1a609bb4c0171e2abb92b54c16e739aec9ca2
85861Author: Kristian Høgsberg <krh@redhat.com>
85862Date: Mon Jul 21 18:29:08 2008 -0400
85863
85864 Drop the glx resize hook and stop chaining PositionWindow.
85865
85866commit 24dddcd0ef845f4120f8588dc63ec754338ffac8
85867Author: Kristian Høgsberg <krh@redhat.com>
85868Date: Mon Jul 21 18:16:38 2008 -0400
85869
85870 Drop unnecessary linked list of contexts from GLXDrawable.
85871
85872commit 5c1e254cc85e9ad409b0217780545c29f62d5feb
85873Author: Daniel Stone <daniel@fooishbar.org>
85874Date: Thu Jul 24 03:01:45 2008 +0300
85875
85876 Remove xorgcfg
85877
85878 Us shipping a GUI configuration utility (especially as part of the
85879 server!) was pretty pointless. There was pretty much nothing it could
85880 configure which wasn't already runtime adjustable: if you could get a
85881 server up with functioning input and output, there wasn't much xorgcfg
85882 could do for you.
85883
85884 Au revoir.
85885
85886commit b74927c3844bc2650d95f604fe782d95ade067f1
85887Author: Daniel Stone <daniel@fooishbar.org>
85888Date: Thu Jul 24 02:42:13 2008 +0300
85889
85890 Remove ioport
85891
85892 I don't know why we would possibly be shipping this.
85893
85894commit 8c0518379089d230060e9ff672ba5eba34198325
85895Author: Daniel Stone <daniel@fooishbar.org>
85896Date: Thu Jul 24 02:40:02 2008 +0300
85897
85898 Remove kbd_mode
85899
85900 We only built this on BSD and Solaris, and if such a tool is generally
85901 useful, ship it with the OS.
85902
85903commit 0d785bd635d135fcd67b4c9c88f5c8217e9b9240
85904Author: Adam Jackson <ajax@redhat.com>
85905Date: Wed Jul 23 13:43:09 2008 -0400
85906
85907 Unifdef hpux.
85908
85909commit 5035741fd4cb4918957064a8c9ded9c7c31b4e22
85910Author: Adam Jackson <ajax@redhat.com>
85911Date: Wed Jul 23 13:38:38 2008 -0400
85912
85913 Unifdef __osf__
85914
85915commit bd8bd2c700e6ccac53b47c45766c0a3a08a59344
85916Author: Adam Jackson <ajax@redhat.com>
85917Date: Wed Jul 23 13:36:35 2008 -0400
85918
85919 Unifdef AIX.
85920
85921commit 5f5eed7f790db2f0ae0cd0f5c5ee5a312079c0b0
85922Author: Adam Jackson <ajax@redhat.com>
85923Date: Wed Jul 23 13:27:24 2008 -0400
85924
85925 Unifdef DGUX.
85926
85927commit 856db05b58b71c5ff22af727aba435a8c356abfa
85928Author: Adam Jackson <ajax@redhat.com>
85929Date: Wed Jul 23 13:25:37 2008 -0400
85930
85931 Unifdef sgi.
85932
85933commit ab6557fcd4bb4fd580bb1727dc8764f7dfb2cb30
85934Author: James Cloos <cloos@jhcloos.com>
85935Date: Wed Jul 23 13:17:33 2008 -0400
85936
85937 Optimize commit 331cc3f0799a54910a99484264f76569beeee55a
85938
85939 Reorder to minimize comparisons
85940
85941commit b699364b1c807d29666633523f5b5a608a07a8ee
85942Author: Adam Jackson <ajax@redhat.com>
85943Date: Wed Jul 23 11:43:59 2008 -0400
85944
85945 Unifdef VXWORKS.
85946
85947commit 2e3c43d306616f8accdeaffec9c108ba88324805
85948Author: Adam Jackson <ajax@redhat.com>
85949Date: Wed Jul 23 11:35:06 2008 -0400
85950
85951 Assume __GNUC__ >= 3.
85952
85953 gcc 3.0 was seven years ago. Get with the times.
85954
85955commit 0c1c9d75089aac0cf673820db83801d3cab5bfc7
85956Author: Adam Jackson <ajax@redhat.com>
85957Date: Wed Jul 23 09:11:40 2008 -0400
85958
85959 Remove some broken debugging scaffolding.
85960
85961commit 0ff5bc404842f7264a8e77f9e2a0bd8d0b051356
85962Author: Adam Jackson <ajax@redhat.com>
85963Date: Wed Jul 23 09:09:11 2008 -0400
85964
85965 Unifdef QNX.
85966
85967 Again, hasn't worked since at least 7.0.
85968
85969commit 331cc3f0799a54910a99484264f76569beeee55a
85970Author: James Cloos <cloos@jhcloos.com>
85971Date: Wed Jul 23 00:01:43 2008 -0400
85972
85973 Fix LookupColor
85974
85975 Using strncasecmp(3) with the lenght of the user-supplied colour name
85976 will result in a false positive when the db key starts out with the
85977 same string.
85978
85979 Eg, blue will also match BlueViolet (aka blue violet).
85980
85981 Since the shorter strings occur first in the database, avoid such
85982 errors by treating a 0 result from strncasecmp(3) as a positive result
85983 when the key’s length is longer than the supplied string’s.
85984
85985commit bc3c03a3f3c091026310f0e8d55321cec570a0c5
85986Author: Pierre-Loup A. Griffais <pgriffais@nvidia.com>
85987Date: Tue Jul 22 17:34:37 2008 -0700
85988
85989 Don't return BadAlloc when trying to set a PictureFilter with no parameters when a filter with parameters was previously set.
85990
85991 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
85992
85993commit 92fdd01d8e46dd4b41338754e1a34e9d3c387d4a
85994Author: Dodji Seketeli <dodji@seketeli.org>
85995Date: Wed Jul 23 01:07:29 2008 +0200
85996
85997 [Xephyr] various X86_64 build cleanups
85998
85999commit 5de1867fbb0a336ff3fdc92cbf734849f6111b1b
86000Author: Dodji Seketeli <dodji@seketeli.org>
86001Date: Wed Jul 23 01:00:26 2008 +0200
86002
86003 [Xephyr] Fix #15839
86004
86005 Make sure the _XSERVER64 macro is not defined in Xlib client code.
86006 That macro is meant to be define only on pure server code, when necessary.
86007
86008commit d692e1a63e4718d3b6d486582c19f30259b813a7
86009Author: Adam Jackson <ajax@redhat.com>
86010Date: Tue Jul 22 10:21:22 2008 -0400
86011
86012 Build fix.
86013
86014commit d8937ce58137ad675f1ead8f2345eaed0ca63c8a
86015Author: Daniel Stone <daniel@fooishbar.org>
86016Date: Tue Jul 22 16:52:50 2008 +0300
86017
86018 autogen.sh: Pass --force to autoreconf
86019
86020 Ensure we get all files installed when running autoreconf by passing
86021 --force to overwrite previously-generated ones, which fixes running it
86022 from tarballs.
86023
86024 Also revert Peter's local changes to pass arguments to configure, all of
86025 which are the default now anyway, AIUI.
86026
86027commit 880625eef5d8b168df3e42836fa1b763c51a91b5
86028Author: Peter Hutterer <peter.hutterer@who-t.net>
86029Date: Tue Jul 22 14:34:28 2008 +0930
86030
86031 xfree86: plug memory leak, free driver's private data when deleting the device.
86032
86033commit 67d7821ae783d3f123b6ba7203abf847374a1e36
86034Author: Keith Packard <keithp@keithp.com>
86035Date: Mon Jul 21 11:48:24 2008 -0700
86036
86037 dix: reset potential lastSlaves when disabling an SD
86038
86039 Unplug a mouse, then warp the pointer and the warp pointer code will try
86040 to update the position of the last slave device associated with the
86041 master. That pointer will be stale and the X server will crash.
86042
86043 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
86044
86045commit 8c8c4fdf34bfc9d54ebea99fb0af14cad167b4a0
86046Author: Adam Jackson <ajax@redhat.com>
86047Date: Mon Jul 21 16:39:43 2008 -0400
86048
86049 EDID: Various reduced blanking fixes.
86050
86051 - Use a single common function to compute reducedness.
86052 - Call it from both the old-school and new-school mode validation paths.
86053 - Define monitor reduced-blanking support in accord with EDID 1.4.
86054 - Attempt to filter RB DMT modes away from the "standard" EDID pool if
86055 the monitor doesn't claim RB support.
86056
86057commit e8cd77e14d3fa40e5cf1174acaf925362b2e0a11
86058Author: Adam Jackson <ajax@redhat.com>
86059Date: Mon Jul 21 16:15:03 2008 -0400
86060
86061 EDID: For standard timing blocks, prefer DMT timings if they exist.
86062
86063commit 6ba70091e2325f534a37ef185fea568d2c43edec
86064Author: Adam Jackson <ajax@redhat.com>
86065Date: Mon Jul 21 15:40:34 2008 -0400
86066
86067 EDID: Correct DDCEstablishedModes to conform to DMT.
86068
86069commit facb255fa9267e343cbc91f841f1b64e5dc99e98
86070Author: Kristian Høgsberg <krh@redhat.com>
86071Date: Mon Jul 21 16:05:53 2008 -0400
86072
86073 Need to unref pixmaps backing pbuffers too.
86074
86075commit d5ae85b5b722821499d5796cf0973ecb6ec125f1
86076Author: Kristian Høgsberg <krh@redhat.com>
86077Date: Mon Jul 21 15:28:50 2008 -0400
86078
86079 Fix embarrasing GLXPixmap leak.
86080
86081commit c74ddc87c995c73109827717a49f14846c7c4024
86082Author: Adam Jackson <ajax@redhat.com>
86083Date: Mon Jul 21 15:10:20 2008 -0400
86084
86085 EDID: Add quirk to clamp max pixel clock to single DVI link speed.
86086
86087 On some panels you end up with all of:
86088 - No range descriptor
86089 - No description of physical connectivity
86090 - Native panel size mode in standard timings list
86091
86092 In principle you're supposed to use the timings for that mode from the DMT
86093 spec, but in practice the DMT spec has timings for both 1920x1200 normal
86094 and 1920x1200RB, and the standard timing field gives you no way to
86095 distinguish. And, of course, the non-RB timings don't fit in a single
86096 DVI link.
86097
86098commit 55803473adb0a0975fea81035402bf6b4ec0e30b
86099Author: Alan Hourihane <alanh@tungstengraphics.com>
86100Date: Mon Jul 21 10:32:57 2008 +0100
86101
86102 Bump inputproto to 1.9.99.2 for XI_PROP_ENABLED
86103
86104commit 69de40ee45a6e046be79b735cd1540f63b87aee9
86105Author: Peter Hutterer <peter.hutterer@who-t.net>
86106Date: Mon Jul 21 17:05:04 2008 +0930
86107
86108 dix: shift the duplicate button mapping check to ProcSetPointerMapping.
86109
86110 XI 2 allows two buttons to have the same button code.
86111
86112commit 0dbfe0ebc69c307c0626ba824de15d03de1251d4
86113Author: Daniel Stone <daniel@fooishbar.org>
86114Date: Mon Jul 21 03:50:15 2008 +0300
86115
86116 Revert "Try nvidia before nv under the assumption that people who install it usually want to *use* it, too."
86117
86118 Per mailing list discussion.
86119
86120 This reverts commit 66fb253082ea42179180303393e48846208987fa.
86121
86122commit 6b5206e7cb8e5279816b48f014d47d3f03f16972
86123Author: Daniel Stone <daniel@fooishbar.org>
86124Date: Fri Jul 18 15:38:10 2008 +0300
86125
86126 dix: Remove insane BC hacks
86127
86128 If your DDX needs a 1bpp mode and doesn't set it up, your DDX is
86129 incompetent.
86130
86131commit 66fb253082ea42179180303393e48846208987fa
86132Author: Aaron Plattner <aplattner@nvidia.com>
86133Date: Tue Jul 8 02:39:00 2008 -0700
86134
86135 Try nvidia before nv under the assumption that people who install it usually want to *use* it, too.
86136
86137commit 376620460ec50cb98a1e3746629b7f73c2e926d7
86138Author: Julien Cristau <jcristau@debian.org>
86139Date: Sun Jul 20 16:53:01 2008 +0200
86140
86141 Drop some more configure-generated files from the tarball
86142
86143commit 35346a42271dfe65fe0a86dc522701028bf6daf6
86144Author: Julien Cristau <jcristau@debian.org>
86145Date: Sun Jul 20 16:40:59 2008 +0200
86146
86147 Don't put xf86Build.h in the tarball (bug#9277)
86148
86149 This file is generated by configure, we don't need to ship a stale
86150 version in the tarball.
86151 X.Org bug#9277 <https://bugs.freedesktop.org/show_bug.cgi?id=9277>
86152
86153commit 43c6d5a6f83dfdf8596c8d0faf8ac1a64f73a8c1
86154Author: Julien Cristau <jcristau@debian.org>
86155Date: Sun Jul 20 16:30:24 2008 +0200
86156
86157 distcheck fixes
86158
86159 Still seems to fail because hw/xquartz has too long filenames
86160
86161commit 92b0a04ffd630a89705fc73d587182a62c34cd84
86162Author: James Cloos <cloos@jhcloos.com>
86163Date: Sat Jul 19 18:21:37 2008 -0400
86164
86165 Fix compile failure
86166
86167 This fixes a compile failure by ensuring that mouse-cfg.c
86168 can see the MTYPE_ constants.
86169
86170commit 0b45ba488d8a3e901b83e445d17e7e9a67047b6c
86171Author: Adam Jackson <ajax@redhat.com>
86172Date: Fri Jul 18 10:54:10 2008 -0400
86173
86174 Drop the FONTCACHE extension.
86175
86176 This appears to have been dead code since X-TT was merged.
86177
86178commit b85392e66b4cbfcc95897ac342f19ac8ab04d4cc
86179Author: Søren Sandmann <sandmann@redhat.com>
86180Date: Thu Jul 17 23:10:11 2008 -0400
86181
86182 Give priority to clients that are receiving expose and configure events
86183
86184commit 3b687ffe1649449b3d182f5e7690274c6c96916a
86185Author: Alan Coopersmith <alan.coopersmith@sun.com>
86186Date: Thu Jul 17 18:16:59 2008 -0700
86187
86188 Make xstrcasestr prototype return value match the implementation
86189
86190commit f7bfa4ae971b43c3ed81cc84e6d1e071c413f04c
86191Author: Daniel Stone <daniel@fooishbar.org>
86192Date: Thu Jul 17 23:03:57 2008 +0300
86193
86194 configure.ac: Remove non-existant directory
86195
86196 Oops.
86197
86198commit 16e40ecf43ea7df349e54f757db0223764d752ee
86199Author: Daniel Stone <daniel@fooishbar.org>
86200Date: Thu Jul 17 22:37:31 2008 +0300
86201
86202 Remove dead code, useless #defines, et al
86203
86204commit b0a99fb02bb7290266d7dddee4de0b9959fb3c52
86205Author: Daniel Stone <daniel@fooishbar.org>
86206Date: Thu Jul 17 21:42:17 2008 +0300
86207
86208 configure.ac: Fix release date
86209
86210 Instead of putting something that's always wrong, at least just put
86211 unreleased, which will be more obviously wrong if it's, well, in a
86212 release.
86213
86214commit 711720650cc192022f0d91f5cf94292d48dbc891
86215Author: Daniel Stone <daniel@fooishbar.org>
86216Date: Thu Jul 17 21:39:46 2008 +0300
86217
86218 Everyone has urandom
86219
86220 If you don't have urandom, please just add a fallback to /dev/random,
86221 rather than building our own random generator.
86222
86223commit 446fe9eecddd1337f9d5164dd7c301e1ba3dfe32
86224Author: Daniel Stone <daniel@fooishbar.org>
86225Date: Thu Jul 17 21:37:50 2008 +0300
86226
86227 Dead code removal
86228
86229 Remove a whole bunch of code that was never built, be it entire files or
86230 just dead ifdefs.
86231
86232commit 0564b5454ac101d9e1218767bbbc2c2d9f3e0696
86233Author: Daniel Stone <daniel@fooishbar.org>
86234Date: Thu Jul 17 20:46:27 2008 +0300
86235
86236 XFree86: Restore AllowMouseOpenFail usage text
86237
86238 This was incorrectly stuck behind the XFree86-Misc define, so got
86239 deleted with the rest of misc: restore it unconditionally, as it should
86240 be.
86241
86242commit 6c7c0ffeb71610ee00fc92b32fd2b1cc70383e38
86243Author: Daniel Stone <daniel@fooishbar.org>
86244Date: Thu Jul 17 20:45:01 2008 +0300
86245
86246 KDrive: Delete rafts of unused code
86247
86248 Most of this stuff was gone with the last real hardware drivers that
86249 used it. Au revoir.
86250
86251commit ad7a7ae290b05f007062161418c929d9cebeef78
86252Author: Julien Cristau <jcristau@debian.org>
86253Date: Thu Jul 17 18:51:05 2008 +0200
86254
86255 XF86BigFont is disabled by default
86256
86257 3c19ec47b434d4ca84db58363cc053cc0b6aa413 did that,
86258 but pretended to make it 'auto'.
86259
86260commit 058100fd19ed33a3100d18a9ca56fdcffac361b2
86261Author: Alan Coopersmith <alan.coopersmith@sun.com>
86262Date: Thu Jul 17 09:35:30 2008 -0700
86263
86264 Remove more Lynx leftovers
86265
86266 A couple #if defined(Lynx) && defined(sun) had become just if defined(sun),
86267 resulting in wrong settings for Solaris builds, so they're now just deleted.
86268
86269commit 47833eef35b20a2eb0a74782edbb0877fd491c4c
86270Author: Mathieu Bérard <mathieu.berard@crans.org>
86271Date: Thu Jul 17 11:59:24 2008 -0400
86272
86273 Drop a bunch of #ifdef Lynx.
86274
86275commit af2138bfce34a28312b26106eaf9561d96712ee0
86276Author: Adam Jackson <ajax@redhat.com>
86277Date: Thu Jul 17 10:39:29 2008 -0400
86278
86279 Nuke some ancient code for commandline-challenged OSes.
86280
86281 This appears to have been dead code even in 6.7.
86282
86283commit b9a0cee026361189dce63b87c738bd1d17e02830
86284Author: Adam Jackson <ajax@redhat.com>
86285Date: Thu Jul 17 10:32:35 2008 -0400
86286
86287 Drop hw/xfree86/os-support/lynxos/
86288
86289 This has never worked in any modular release, clearly no one cares.
86290
86291commit d87f170bd99f7908eb14272ac42c8963d089f54f
86292Author: Adam Jackson <ajax@redhat.com>
86293Date: Thu Jul 17 10:29:52 2008 -0400
86294
86295 Remove XF86Misc code from xorgcfg.
86296
86297commit e3ff8501749209ee2dd324d35ae16e37510e0574
86298Author: Jie Luo <clotho67@gmail.com>
86299Date: Thu Jul 17 10:27:24 2008 -0400
86300
86301 Remove duplicate load of freetype font renderer.
86302
86303 libXfont will do this for us in FontFileRegisterFpeFunctions.
86304
86305commit 8441e26266701776f1416bec864ae2949355da76
86306Author: Jie Luo <clotho67@gmail.com>
86307Date: Thu Jul 17 10:23:37 2008 -0400
86308
86309 Remove freetype from the list of autoloaded modules.
86310
86311commit 5d11ae7a69e3a7611c667be4628a15a4b514f72c
86312Author: Jie Luo <clotho67@gmail.com>
86313Date: Thu Jul 17 10:22:29 2008 -0400
86314
86315 Fix color lookup.
86316
86317 OsStrCaseCmp returned inverted comparison sense from normal strcasecmp.
86318
86319commit 478d3918d5b23fba1a7e3aaea766dfa785e5a0a6
86320Author: Daniel Stone <daniel@fooishbar.org>
86321Date: Thu Jul 17 03:29:26 2008 +0300
86322
86323 XFree86: Remove remnants of XFree86-Misc
86324
86325 HandleMessage: dear god, no.
86326
86327commit 95000e34df9ba066905430eb0eb39a616567183e
86328Author: Adam Jackson <ajax@redhat.com>
86329Date: Wed Jul 16 16:37:14 2008 -0400
86330
86331 Build fixes
86332
86333commit affec10635343668839994ea2bac16c1d4524200
86334Author: Adam Jackson <ajax@redhat.com>
86335Date: Wed Jul 16 14:44:33 2008 -0400
86336
86337 Remove loadable font renderer support.
86338
86339commit acc91054316208844742495dea0b80a860ef02bb
86340Author: Adam Jackson <ajax@redhat.com>
86341Date: Wed Jul 16 14:24:47 2008 -0400
86342
86343 Remove font module support from xorgcfg.
86344
86345commit a82e6efb7b9b2ab9a1597b002f375c5ee105e7f5
86346Author: Adam Jackson <ajax@redhat.com>
86347Date: Wed Jul 16 14:19:04 2008 -0400
86348
86349 Remove the numVideoScreens xprintism.
86350
86351 This was to account for cases where you had video and print screens in
86352 the same server. Lunacy. Leave the slot in ScreenInfo, but rename it,
86353 and stop looking at it.
86354
86355commit 1a573e402ec112913a404f092b5b97d8d9210f94
86356Author: Adam Jackson <ajax@redhat.com>
86357Date: Wed Jul 16 13:41:53 2008 -0400
86358
86359 Update the Allow*Grabs documentation for xf86misc removal.
86360
86361 The API to turn the grab-break keys back off is gone now, so don't
86362 say it exists.
86363
86364commit 55644d2f1cddf1777587f213ed581b6087d9fdad
86365Author: Daniel Stone <daniel@fooishbar.org>
86366Date: Wed Jul 16 14:23:34 2008 +0300
86367
86368 Make --enable-debug usable again
86369
86370 Enable region debugging causes assertion failures (should really look
86371 into that, though), and I don't need four messages every time I move my
86372 pointer.
86373
86374commit 22e64108ec63ba77779891f8df237913ef9ca731
86375Author: Daniel Stone <daniel@fooishbar.org>
86376Date: Wed Jul 16 06:25:26 2008 +0300
86377
86378 XFree86: Remove XFree86-Misc extension
86379
86380 Its last remaining purpose in life has been destroyed by input
86381 properties. Au revoir: it's been fun, by which I mean awful.
86382
86383commit b8dd07f855c555af56cbf0f69df799f424da2cca
86384Author: Daniel Stone <daniel@fooishbar.org>
86385Date: Wed Jul 16 03:00:25 2008 +0300
86386
86387 HAL: Remove grotesque open-coded strcasestr
86388
86389 Not only was this pretty ugly, but it didn't even work on systems
86390 without strcasestr anyway, due to the define not being in dix-config.h.
86391 Lack of strcasestr is handled transparently with the version from
86392 FreeBSD now anyway, so, huzzah.
86393
86394commit ad87c72edcc0d1f56658e0c4e73af335c8d5a516
86395Author: Daniel Stone <daniel@fooishbar.org>
86396Date: Wed Jul 16 02:59:51 2008 +0300
86397
86398 DIX: Add strcasestr from FreeBSD
86399
86400 Add strcasestr for use on systems which don't have it.
86401
86402commit c3c901cf44cf16bb33c4176494361b429099a372
86403Author: Daniel Stone <daniel@fooishbar.org>
86404Date: Wed Jul 16 02:54:41 2008 +0300
86405
86406 configure.ac: Fix SHA1 handling
86407
86408 Fix and marginally simplify the SHA1 handling. First, we allow people
86409 to override it. Secondly, we try for libmd. Then, we try for OpenSSL
86410 with pkg-config. In a last, desperate move, we try libcrypto on its
86411 own. This allows the server to, y'know, _link_ when using OpenSSL,
86412 instead of failing because we only have -lcrypto, and not -lssl.
86413
86414commit 69b57dc651e12a0d9a5a4295b185c62d5c0df63f
86415Author: Daniel Stone <daniel@fooishbar.org>
86416Date: Wed Jul 16 02:03:36 2008 +0300
86417
86418 dix: Actually build str(n)casecmp if we don't have it
86419
86420 Remember to add stuff to dix-config.h when you add new AC_DEFINES,
86421 people ...
86422
86423commit dcf6293030126509d7d6c61d131222037d5ed7db
86424Author: Daniel Stone <daniel@fooishbar.org>
86425Date: Wed Jul 16 01:57:00 2008 +0300
86426
86427 strcasecmp: Actually use the right license
86428
86429 Forgot to update the license when I committed the FreeBSD version, so it
86430 still had an old SGI license. Sorry. Sorry.
86431
86432commit 441f084bfe87a6ea1c94ec63f82888b8b3d81d89
86433Author: Daniel Stone <daniel@fooishbar.org>
86434Date: Wed Jul 2 19:52:58 2008 +0300
86435
86436 config: Don't attempt to use D-Bus when not strictly necessary
86437
86438 If we have D-Bus but have explicitly disabled it, don't build it.
86439
86440commit 35c89f3f5b8fa222e37b799d5bb01595e8f30d0c
86441Author: Daniel Stone <daniel@fooishbar.org>
86442Date: Wed Jul 16 01:43:58 2008 +0300
86443
86444 XFree86: Remove mysticism from Makefile.am
86445
86446 The variables were always the same, so just shove them in with the rest
86447 of the plebs.
86448
86449commit b89a59248a4a0ff06b9a0ddee45881efc6063063
86450Author: Daniel Stone <daniel@fooishbar.org>
86451Date: Fri Jun 27 12:20:56 2008 +0300
86452
86453 XFree86: Delete OSMouse code
86454
86455 This should be moved into the mouse driver, if anything.
86456
86457commit d0de5ea96d084fc5da87d8f323ddfc08fe9c03ba
86458Author: Daniel Stone <daniel@fooishbar.org>
86459Date: Wed Jun 25 17:56:28 2008 +0300
86460
86461 XFree86: Remove useless debugging code
86462
86463 Also remove documentation which told you how to use a non-module-aware
86464 GDB, albeit only with old, non-shared, modules.
86465
86466commit 4b1273c9c2da113f634be80caa28e81df3beae98
86467Author: Daniel Stone <daniel@fooishbar.org>
86468Date: Wed Jun 25 17:51:12 2008 +0300
86469
86470 XFree86: Delete empty file & function
86471
86472 xf86InitXkb() has been empty for as long as I can remember.
86473
86474commit 3a54f3f48fa1c0d60604c3ee767c569b5ec23430
86475Author: Daniel Stone <daniel@fooishbar.org>
86476Date: Thu Jun 12 01:05:04 2008 +0300
86477
86478 XFree86: Clean up init a tiny bit (no code changes)
86479
86480 Reshuffle and delete.
86481
86482commit ddcefb50dda9e398647d1c84c7153127ed26a4d2
86483Author: Daniel Stone <daniel@fooishbar.org>
86484Date: Wed Jun 11 15:14:26 2008 +0300
86485
86486 XFree86: Remove trailing whitespace
86487
86488commit e6f35f28fb3526b911101bde4aa761de8b055aef
86489Author: Daniel Stone <daniel@fooishbar.org>
86490Date: Wed Jun 11 15:13:21 2008 +0300
86491
86492 OS/KDrive/XFree86: Sanitise colour initialisation
86493
86494 OsInitColors always just returned TRUE, so just remove calls to it and
86495 insane special-case logic. Remove unused kcolor.c implementation, and
86496 merge oscolor.h into oscolor.c since it was the only user. Remove
86497 open-coded strncasecmp in oscolor.c.
86498
86499 Since we no longer need to call OsInitColors after reading the config
86500 file, just call PostConfigInit() from one place, and move PM handling to
86501 one place so we can install the signal handlers earlier.
86502
86503commit 11f9e3520249a603b95e64503ee759998ff17feb
86504Author: Daniel Stone <daniel@fooishbar.org>
86505Date: Wed Jun 11 15:09:46 2008 +0300
86506
86507 DIX: Add strncasecmp from FreeBSD, make strcasecmp args const
86508
86509 Add strncasecmp (as we're now using it) in case someone doesn't have it,
86510 and also change strncasecmp args to be const, in accordance with
86511 everything else.
86512
86513commit ae38151ddda9984effca5bb7c582540061201dce
86514Author: Daniel Stone <daniel@fooishbar.org>
86515Date: Mon May 19 08:35:05 2008 +0300
86516
86517 configure.ac: Add GLX_SYS_LIBS for Xvfb and Xnest
86518
86519 This fixes the linking for these two when using DRI2, which requires
86520 DLOPEN_LIBS.
86521
86522commit ccec16e519c143f4c0017a37e93ebc52ac472e4e
86523Author: Adam Jackson <ajax@redhat.com>
86524Date: Tue Jul 15 11:24:08 2008 -0400
86525
86526 Bug #7300: Move xf86RAC to hw/xfree86/common
86527
86528commit 9111944b292355f7478b4ae75bead8dc25edbbcb
86529Author: Julien Cristau <jcristau@debian.org>
86530Date: Tue Jul 15 10:36:38 2008 -0400
86531
86532 Bug #16674: Make sure RANDR reports refresh as 0 if pixel clock is 0.
86533
86534commit 3f3f66de669caf3606eec8c5f8d405a283cbf12d
86535Author: Brad Smith <brad@comstyle.com>
86536Date: Mon Jul 14 22:20:41 2008 +0200
86537
86538 Fix the configure tests for swap16/__swap16 macros on OpenBSD.
86539
86540 BSD systems need <sys/types.h> included before <sys/endian.h> for
86541 macros to work.
86542 <https://bugs.freedesktop.org/show_bug.cgi?id=16531>
86543
86544commit 11ee0ae9390a608a232ff94abcc0cbcf9ed7b70a
86545Author: Peter Hutterer <peter.hutterer@who-t.net>
86546Date: Mon Jul 14 10:20:11 2008 +0930
86547
86548 xfree86: append, not prepent, new input devices to xf86InputDevs.
86549
86550 If devices are prepended to the list, their wake-up order on resume is not the
86551 same as the original initialisation order. Hot-plugged devices, originally
86552 inited last, are re-enabled before the xorg.conf devices and in some cases may
86553 steal the device files. Result: we have different devices before and after
86554 suspend/resume.
86555
86556 RedHat Bug 439386 <https://bugzilla.redhat.com/show_bug.cgi?id=439386>
86557
86558commit 5bcc45e07e8726a5442567472dd29cfb5c901f2d
86559Author: Peter Hutterer <peter.hutterer@who-t.net>
86560Date: Sun Jul 13 18:40:53 2008 +0930
86561
86562 Xi: expose Enable/DisableDevice through XI_PROP_ENABLED property.
86563
86564commit c9eb0e870c87d291311491452adf7f91a911e24b
86565Author: Simon Thum <simon.thum@gmx.de>
86566Date: Thu Jul 10 22:33:39 2008 +0930
86567
86568 Add support for multiple pointer acceleration schemes. #8583
86569
86570 Available acceleration schemes:
86571 - xorg classic scheme.
86572 - the new "Predictable" polynomial accel scheme.
86573
86574 X.Org Bug 8583 <http://bugs.freedesktop.org/show_bug.cgi?id=8583>
86575
86576 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
86577
86578commit e7abe1676a6a4e4249504b8c9660cbad70569199
86579Author: Peter Hutterer <peter.hutterer@who-t.net>
86580Date: Sun Jul 13 18:41:53 2008 +0930
86581
86582 Xi: protect against NULL handlers, don't try to dereference.
86583
86584commit 2bbb12c355308d10bf123911044fbdf6ae7fb59c
86585Author: Peter Hutterer <peter.hutterer@who-t.net>
86586Date: Sun Jul 13 20:23:14 2008 +0930
86587
86588 Xi: remove ChangeDeviceControl for CoreCtl.
86589
86590 If you want to set a device to core, attach it to a master device.
86591
86592commit 18ff17756c988b6c21b76bf898e45685649d07bb
86593Author: Peter Hutterer <peter.hutterer@who-t.net>
86594Date: Thu Jul 10 19:23:14 2008 +0930
86595
86596 Xi: GetDeviceProperty reply includes deviceid.
86597
86598commit c3267106fb599213555829cb76df7848c4ebe23b
86599Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
86600Date: Wed Jul 9 10:02:41 2008 -0700
86601
86602 XQuartz: Use CFEqual to compare keyboards
86603 (cherry picked from commit 5538e43b9ae7d06d2f48842b065810ce74286eb6)
86604
86605commit 90dd2de845ae12153296f6f1bff0c87f79c57854
86606Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
86607Date: Mon Jul 7 10:55:58 2008 -0700
86608
86609 XQuartz: Some fd handoff cleanup.
86610 (cherry picked from commit 9c20a4804d97e67a988f00f49866997209cce518)
86611
86612commit 26d8030c3836816de8c12b2cb9d67315e5c887eb
86613Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
86614Date: Fri Jul 4 19:23:21 2008 -0700
86615
86616 XQuartz: Remove deprecated keyboard code.
86617 (cherry picked from commit 69cfc1a21e12bb38a6130dea2e5f20f1e6a3ee7c)
86618
86619commit 5cfcbd54d98cc77ee02a3a099ebbad9af511a0ee
86620Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
86621Date: Wed Jul 2 00:47:08 2008 -0700
86622
86623 Set machine dependent defaults for ppc64
86624 (cherry picked from commit 0733ef2e8abda99cfd62966e73017949e9cd507f)
86625
86626commit e69b9f9ca45c0c6bfb93ea9143737116bf1f2453
86627Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
86628Date: Wed Jun 25 11:51:27 2008 -0700
86629
86630 XQuartz: Set noPanoramixExtension earlier to avoid a possible race.
86631 (cherry picked from commit 49668e8a88137e9f258eae970826883b88b7d8ba)
86632
86633commit 8d2e2e1d856efec4459de2a20af642dc1ec9b8a5
86634Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
86635Date: Wed Jun 11 11:44:45 2008 -0700
86636
86637 Xquartz: Removed include directive for removed header
86638 (cherry picked from commit e65a36d57f338410c5a5b02cb5ae1214a81d072d)
86639
86640commit ebd70c82fdbe483d149ac9bdb5a64cc6e0c10e0a
86641Author: Aaron Plattner <aplattner@nvidia.com>
86642Date: Thu Jul 10 14:38:34 2008 -0700
86643
86644 Add LoaderShouldIgnoreABI to allow drivers that roll their own ABI checks (i.e. nvidia) to perform the check before ScreenInit.
86645
86646commit c42427f63f0e0495e56909ee99ad1e3c4321c39b
86647Author: Ander Conselvan de Oliveira <ander@mandriva.com.br>
86648Date: Wed Jul 9 10:50:21 2008 -0300
86649
86650 Fix incorrect test regarding keyboard map.
86651
86652 In the map stored in each keyboard device, the first line refers to
86653 minimum keycode, i.e., the 0th line refers to keycode 8. When not
86654 using XKB the wrong test caused some keys to be interpreted as
86655 locks ('m' for instance). The had to be pressed twice to generate
86656 both KeyPress and KeyRelease events.
86657
86658 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
86659
86660commit e4054e5cb3d919dfc83021be3f2b9036d0e5a2be
86661Author: Peter Hutterer <peter.hutterer@who-t.net>
86662Date: Thu Jul 10 11:03:31 2008 +0930
86663
86664 Xi: pack the property stuff into a struct, make handlers a linked list.
86665
86666 We may need more than one handler to deal with a property (e.g. one in the
86667 driver, one in the DIX), so get the handlers into a linked list and call them
86668 one-by-one. This is of course slightly less entertaining than the hilarious
86669 WRAP/UNWRAP game we play in other parts of the server.
86670
86671 XIRegisterPropertyHandler/XIUnregisterPropertyHandler are the interface
86672 drivers/the DIX should use to attach themselves to the device.
86673
86674 XIDeleteAllDeviceProperties destroys everything, including the handlers.
86675
86676commit 2039c6ea43244339659323d05d96c14e41996c20
86677Author: Peter Hutterer <peter.hutterer@who-t.net>
86678Date: Mon Jul 7 22:10:17 2008 +0930
86679
86680 Xi: add support for input device properties.
86681
86682 Basically just copied from randr properties, with minor changes only.
86683 Each device supports arbitrary properties that can be modified by clients.
86684 Modifications to the properties are passed to the driver (if applicable) and
86685 can then affect the configuration of the device.
86686
86687 Note that device properties are limited to a specific device. A property set
86688 on a slave device does not migrate to the master.
86689
86690commit 4ab01fe5db7fb330b1ec463aa49b77859527e597
86691Author: Peter Hutterer <peter.hutterer@who-t.net>
86692Date: Thu Jul 10 16:27:36 2008 +0930
86693
86694 Revert "Xi: add support for input device properties."
86695
86696 Note to self: don't mix up branches with half-finished cherrypicks.
86697
86698 This reverts commit 666838fcc8b71fdeae160844160187f345cbf4a6.
86699
86700commit 666838fcc8b71fdeae160844160187f345cbf4a6
86701Author: Peter Hutterer <peter.hutterer@who-t.net>
86702Date: Mon Jul 7 22:10:17 2008 +0930
86703
86704 Xi: add support for input device properties.
86705
86706 Basically just copied from randr properties, with minor changes only.
86707 Each device supports arbitrary properties that can be modified by clients.
86708 Modifications to the properties are passed to the driver (if applicable) and
86709 can then affect the configuration of the device.
86710
86711 Note that device properties are limited to a specific device. A property set
86712 on a slave device does not migrate to the master.
86713
86714commit acce27093571497a0626cee1cdb61ddf751dbc40
86715Author: Peter Hutterer <peter.hutterer@who-t.net>
86716Date: Tue Jul 8 16:02:13 2008 +0930
86717
86718 Xext: store the GenericMasks in the resource system.
86719
86720 This fixes a severe issue - when the client died the event mask didn't get
86721 unregistered and a future event would dereference dangling pointers. By
86722 storing the event masks in the resource system we can free them when the
86723 client dies.
86724
86725commit db86b8839f286e0e2efb5638b8ab2fe608707655
86726Author: Peter Hutterer <peter.hutterer@who-t.net>
86727Date: Tue Jul 8 15:12:45 2008 +0930
86728
86729 dix: free GenericMask allocations when destroying window.
86730
86731commit 08e2c625f620688d35f3b443ffb878e8e991ff57
86732Author: Alan Coopersmith <alan.coopersmith@sun.com>
86733Date: Wed Jul 9 20:07:59 2008 -0700
86734
86735 Remove README.font, since it's just an old text copy of xorg-docs/sgml/fonts
86736
86737commit 5e847c1d4fc30a0d263a861a76982660f11998cd
86738Author: Alan Coopersmith <alan.coopersmith@sun.com>
86739Date: Mon Jul 7 17:08:01 2008 -0700
86740
86741 Improved driver selection when autoconfiguring driver without xorg.conf
86742
86743 - Allow returning multiple drivers to try for a given PCI id (for instance,
86744 try "geode" then "amd" for AMD Geode hardware)
86745 - On Solaris, use VIS_GETIDENTIFIER ioctl as well as PCI id to choose drivers
86746 - Use wsfb instead of fbdev as a fallback on non-Linux SPARC platforms
86747
86748commit 235fea0e1a2101fb76a54e6505db321674530c60
86749Author: Alan Coopersmith <alan.coopersmith@sun.com>
86750Date: Mon Jul 7 15:21:59 2008 -0700
86751
86752 Update bsd & solaris descriptions in README.OS-lib
86753
86754commit d8af9d9ab1cd98c07fdf42490dcc0cab3c655b89
86755Author: Julien Cristau <jcristau@debian.org>
86756Date: Mon Jul 7 13:14:49 2008 +0200
86757
86758 Export xkbfile.h in the SDK
86759
86760 It's needed by xkbsrv.h
86761
86762commit d78bebb20a00e8519788c75c90b467a5750c78be
86763Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
86764Date: Fri Jul 4 04:04:19 2008 -0300
86765
86766 Remove unused function and variables.
86767
86768commit a8c4727f01949da0ae266369f0d12a96c2bf0da5
86769Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
86770Date: Fri Jul 4 04:03:46 2008 -0300
86771
86772 Fix typo in the last commit.
86773
86774commit 3164288b121990a5990747c0f92d4af13b1c929a
86775Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
86776Date: Fri Jul 4 03:55:59 2008 -0300
86777
86778 Remove deprecated function miPointerPosition().
86779
86780commit d1031a8972490ebf0ef6dc2762036a7a7172466b
86781Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
86782Date: Fri Jul 4 03:46:21 2008 -0300
86783
86784 Update README.OS-lib to reflect reality.
86785
86786commit f72500c4a09cce5dbbd16639b1951a4bf943e4ac
86787Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
86788Date: Fri Jul 4 03:45:18 2008 -0300
86789
86790 Nuke libc wrapper remaining stuffs.
86791
86792commit 3fd5fab5663fbdce2fc0b5377155aedf9a70b6c2
86793Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
86794Date: Fri Jul 4 02:51:42 2008 -0300
86795
86796 Remove inexistent function headers.
86797
86798commit 0b9ef835a0fe900c121b84e43989591e58ab1126
86799Author: Dave Airlie <airlied@redhat.com>
86800Date: Fri Jul 4 07:29:32 2008 +1000
86801
86802 modes: fix initial xorg.conf mode selection.
86803
86804 This was all kinds of broken, we ignored user preferred modes for multiple
86805 monitors and also for side-by-side configurations.
86806
86807commit e317943c125d7fdbe62c750324640eb30c88fa15
86808Author: Peter Hutterer <peter.hutterer@who-t.net>
86809Date: Wed Jul 2 12:23:03 2008 +0930
86810
86811 xfree86: AllowEmptyInput is now enabled by default if hotplugging is enabled.
86812
86813 Remove AEI check from configImpliedLayout as the setting isn't actually parsed
86814 at this point anyway (written by Sasha Hlusiak).
86815
86816 Resurrect checkInput() and check for devices there if AEI is false (this also
86817 creates the default devices if required).
86818
86819 Set AllowEmptyInput to enabled by default if hotplugging is enabled.
86820
86821commit fe5cf7cb00fd926cefff933adbdbceea7353f4c2
86822Author: Peter Hutterer <peter.hutterer@who-t.net>
86823Date: Tue Jul 1 13:18:54 2008 +0930
86824
86825 xfree86: handle missing Screen lines in the ServerLayout #16301
86826
86827 If no Screen is specified in the ServerLayout section, either take the first
86828 one from the config file or autogenerate a default screen.
86829
86830 X.Org Bug 16301 <http://bugs.freedesktop.org/show_bug.cgi?id=16301>
86831
86832commit 6674b87a7dca7d6ffd9dd9af888c5256b13d7877
86833Author: Peter Hutterer <peter@cs.unisa.edu.au>
86834Date: Sun Jun 29 20:05:27 2008 +0930
86835
86836 xfree86: move declaration of configured_device to start of function.
86837
86838 Last I checked C doesn't allow declarations halfway down a block.
86839
86840commit 1e96782011f0b4ba6ed1d3d3df118964a58d15fd
86841Author: Dave Airlie <airlied@linux.ie>
86842Date: Thu Jul 3 20:00:22 2008 +1000
86843
86844 dri2: bump libdrm numbers as 2.3.1 is released without ttm interface
86845
86846commit 1771edcb44b564f83f509748b4e4cd5b7586e2a8
86847Author: Dave Airlie <airlied@redhat.com>
86848Date: Thu Jul 3 18:57:09 2008 +1000
86849
86850 modes: check the crtc is valid before using its desired modes.
86851
86852 this fixes a crash I was getting on radeon rotate when gnome is running.
86853 I'm sure g-s-d was doing something bad, but really not crashing ftw.
86854
86855commit 689292e58cc5242cbfeaa29fcf65b6dcd9dfabb9
86856Author: Adam Jackson <ajax@redhat.com>
86857Date: Wed Jul 2 11:25:50 2008 -0400
86858
86859 Fix GLX in Xvfb and kdrive.
86860
86861commit 9a73bca859be12721ef62304b1422c455d4aec49
86862Author: Chris Ball <cjb@laptop.org>
86863Date: Wed Jul 2 09:46:06 2008 -0400
86864
86865 Remove KdOffscreenMarkUsed() as all its callers are gone.
86866
86867commit b84a27fd9fdcb579437cfa38f10cb12cf55dc7c3
86868Author: Ian Romanick <idr@us.ibm.com>
86869Date: Wed Jul 2 06:25:03 2008 -0700
86870
86871 VBO: Regenerate files based on recent changes to gl_API.xml
86872
86873 Since GL_ARB_vertex_buffer_object protocol isn't supported yet, these
86874 changes are innocuous.
86875
86876commit 703a9645f3d547144ea62dd81c41f5356a5f87d8
86877Author: Adam Jackson <ajax@redhat.com>
86878Date: Tue Jul 1 17:50:48 2008 -0400
86879
86880 Auf wiedersehen, KAA.
86881
86882commit c8216aede6c4ac41976947521d884fa010913204
86883Author: Jeremy Uejio <jeremy.uejio@sun.com>
86884Date: Tue Jul 1 13:37:12 2008 -0700
86885
86886 Sun bug #6685465: Xephyr uses wrong or bad colortable in 8-bit mode
86887
86888 <http://bugs.opensolaris.org/view_bug.do?bug_id=6685465>
86889
86890 This bug is caused by Xephyr not handling the RGB byte order correctly
86891 of the server where Xephyr is displaying on. The previous code just
86892 assumed that the order was RGB and did not take into account that
86893 Xservers may use different order (such as BGR).
86894
86895 The fix is to add a function to calculate the byte order and bits
86896 to shift based on the visual mask and the visual bits_per_rgb (which
86897 is usually 8, but could be server dependent). Since the shifts won't
86898 change once the display connection has been made, I can cache these
86899 values so that Xephyr doesn't have to keep recalculating them everytime
86900 it tries to translate the Xephyr colormap entries for Xephyr clients to
86901 the actual server colormap entries (i.e. calling the function
86902 hostx_set_cmap_entry() repeatedly for every colormap entry).
86903
86904commit 2e8daee05343b2853b677acf4554def0ceeada00
86905Author: Aaron Plattner <aplattner@nvidia.com>
86906Date: Mon Jun 30 17:29:35 2008 -0700
86907
86908 Work around the DIX losing mmWidth/mmHeight for RandR 1.1 DDXen.
86909
86910 RandR 1.1 has a physical size for each mode. It used to be that the DIX would
86911 remember these modes and pass them back up to the DDX when changing the screen
86912 configuration. The DDX uses RR_GET_MODE_MM to query the driver for the physical
86913 dimensions of the screen, allowing it to preserve the DPI.
86914
86915 With RandR 1.2, the physical dimensions are stored as part of the output, rather
86916 than per mode. The DIX only uses the sizes passed in from the DDX to select the
86917 mode pool for the "default" output, and forgets the physical sizes. Then, when
86918 reconfiguring the screen, it makes up a new RRScreenSizeRec using the dimensions
86919 from the output, screwing up the DPI.
86920
86921 This change works around this problem by ignoring the DIX and querying the real
86922 size from the driver.
86923
86924commit e2bbf2d248a2d2a2db4adfe9d1bda94cf184d8bd
86925Author: Adam Jackson <ajax@redhat.com>
86926Date: Tue Jul 1 12:01:56 2008 -0400
86927
86928 Don't try to load DRI2 if built without it.
86929
86930commit 66d7c9de82b5e733e4995f079a96efe612fca881
86931Author: Adam Jackson <ajax@redhat.com>
86932Date: Tue Jul 1 10:55:54 2008 -0400
86933
86934 Declare xf86InterpretEEDID() properly.
86935
86936commit e4335e876d254e446a965259e845ad955da5b5c2
86937Author: root <root@samich.home.cworth.org>
86938Date: Mon Jun 30 15:29:04 2008 -0700
86939
86940 Revert "XInput ABI is 2.1, not 3.1."
86941
86942 This reverts commit 76576c87b0de106e006d1cfa59d42d3a483e966c.
86943 which was an incorrect revert of previous ABI bumps. Those
86944 responsible for the accidental ABI bumps in both directions
86945 have all been sacked.
86946
86947 This allows xf86-input-mouse to build again, for example.
86948
86949commit 6447f80c17077f924b2d58e07b38d32429edcb9b
86950Author: Carl Worth <cworth@samich.home.cworth.org>
86951Date: Mon Jun 30 13:16:49 2008 -0700
86952
86953 Use AC_CHECK_LIB before PKG_CHECK_MODULES for openssl
86954
86955 Previously, the code was using PKG_CHECK_EXISTS before PKG_CHECK_MODULES,
86956 (to cater to OpenBSD systems that include openssl by default but without
86957 a .pc file). But this meant that systems that didn't have openssl installed
86958 at all would not get any error message at configure time.
86959
86960 Now, if the SHA1_Init function is found in -lcrypto without any additional
86961 flags, then that's used. Otherwise, pkg-config is used to find the right
86962 flags to link against libcrypto. And if that fails, a nice error message
86963 is now generated.
86964
86965commit 76576c87b0de106e006d1cfa59d42d3a483e966c
86966Author: Adam Jackson <ajax@redhat.com>
86967Date: Mon Jun 30 16:10:37 2008 -0400
86968
86969 XInput ABI is 2.1, not 3.1.
86970
86971commit bed12dd3df0129b7f28d0a42d470266bd3ae6b38
86972Author: Adam Jackson <ajax@redhat.com>
86973Date: Sat Jun 21 16:07:54 2008 -0400
86974
86975 Do EEDID in RANDR 1.2.
86976
86977commit 4fc18b5fc31aa287e807ee520dc9117296e3d306
86978Author: Adam Jackson <ajax@redhat.com>
86979Date: Sat Jun 21 16:07:36 2008 -0400
86980
86981 Hexdump the whole EDID block if extensions present.
86982
86983commit b4fbc31e109f1efe78613597f9a91d5363523493
86984Author: Adam Jackson <ajax@redhat.com>
86985Date: Sat Jun 21 16:00:05 2008 -0400
86986
86987 Actually fetch all blocks of EEDID if asked to.
86988
86989commit 211e2bdcc677122145998829a7f26610c7c567e6
86990Author: Adam Jackson <ajax@redhat.com>
86991Date: Mon Jun 30 10:29:17 2008 -0400
86992
86993 Bump ABI minor numbers for the devPrivate ABI functions.
86994
86995commit d6b8d9eaffaf3f976db330bc35da3d30eb656bac
86996Author: Dan Nicholson <dbn.lists@gmail.com>
86997Date: Fri Jun 27 17:16:35 2008 -0700
86998
86999 DRI: Ensure we have the DRI headers from Mesa
87000
87001 Change the DRI configure check to look for the dri pkg-config file. This
87002 prevents people who built an Xlib libGL from bombing later in the build.
87003
87004commit 7ddfb2f1587da3ec1cd27ae270e42700adf4f560
87005Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87006Date: Thu Jun 26 21:23:31 2008 -0400
87007
87008 Fix a leak of window devPrivates in window.c/CrushTree().
87009
87010commit 15e4b6c57484b6afb790c7dc1db9f529ba2219cf
87011Author: Peter Hutterer <peter@cs.unisa.edu.au>
87012Date: Tue Jun 24 13:46:40 2008 +0930
87013
87014 input: for non-device events (e.g. DevicePresence) use MAX_DEVICES as id.
87015
87016 Using id = 0 only worked pre-MPX since XInput didn't allow XOpenDevice for the
87017 core devices (0 and 1). Now we can now legally register for events so we may
87018 overwrite our device-independent classes with the ones selected for the VCP.
87019
87020 So, increase the EMASKSIZE to MAX_DEVICES + 1 and use MAX_DEVICES as the ID
87021 when we don't have a device.
87022
87023commit f0c56ffd980db19bea271aeebf118bd931f6f40d
87024Author: Peter Hutterer <peter@cs.unisa.edu.au>
87025Date: Tue Jun 24 12:58:49 2008 +0930
87026
87027 dix: don't dereference dev->key if it doesn't exist.
87028
87029 This may happen if we check for a passive grab on a floating pointer device.
87030
87031commit 01f0b9c9784c55c94104bafa6dd545b373c64104
87032Author: Alan Coopersmith <alan.coopersmith@sun.com>
87033Date: Wed Jun 25 19:23:20 2008 -0700
87034
87035 xf86GetOptValFreq wants a pointer to double, but maxPixClock is an int
87036
87037commit 7ce6dcef118139d0ccfc136b9a07f5bf0c078d70
87038Author: Alan Coopersmith <alan.coopersmith@sun.com>
87039Date: Wed Jun 25 15:06:49 2008 -0700
87040
87041 Don't log null device name in XkbDDXLoadKeymapByNames
87042
87043commit cbc20d92de92aad5ca240310a9156ccf97c24a01
87044Author: Adam Jackson <ajax@redhat.com>
87045Date: Wed Jun 25 09:52:36 2008 -0400
87046
87047 XTrap delenda est.
87048
87049commit a81b5ce452766bbc4b2ac0849711284ba8591b46
87050Author: Peter Hutterer <peter@cs.unisa.edu.au>
87051Date: Wed Jun 25 13:23:09 2008 +0930
87052
87053 dix: free memory associated with the EventList. #16507
87054
87055 Un-deprecates CloseDownEvents.
87056
87057 X.Org Bug 16507 <http://bugs.freedesktop.org/show_bug.cgi?id=16507>
87058
87059commit 3a2beb2f867fd47dc5a7723dcac345ea0b0ecf3d
87060Author: Peter Hutterer <peter@cs.unisa.edu.au>
87061Date: Wed Jun 25 13:12:41 2008 +0930
87062
87063 dix: InitEventList does not need to be _X_EXPORT.
87064
87065commit 385943e0e97463ce4681a9b6a4a40d7e3c91e51e
87066Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87067Date: Tue Jun 24 23:13:32 2008 -0300
87068
87069 Set --enable-builtin-fonts as default.
87070
87071 Some reasons to embed fonts by default:
87072 1. X server doesn't pick a good default font path so it's easiest just
87073 to built in the core fonts and let new X hackers more happy. Developers
87074 and distro guys are wise enough to just set --disable-builtin-fonts
87075 when they want.
87076
87077 2. Seems that this is by far the most popular FAQ
87078 (http://www.x.org/wiki/FAQErrorMessages).
87079
87080 3. No one gave a good argument to not do this:
87081 http://lists.freedesktop.org/archives/xorg/2008-May/035479.html
87082
87083commit f3b29efc4d7bb346cdd3fd79e780f1022c8f46e8
87084Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87085Date: Tue Jun 24 23:07:49 2008 -0300
87086
87087 Remove NOFONTSERVERACCESS macro
87088
87089 dix: this is semantically different but it goes through the same path in
87090 the code.
87091
87092commit 2ff7eb4995b663cf12a01124745de600f38d0bfc
87093Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87094Date: Tue Jun 24 22:09:34 2008 -0300
87095
87096 xfree86: kill INHERIT_LOCK_STATE macro.
87097
87098 I don't know how this survived so long. Google didn't show
87099 anything usable related with this.
87100
87101commit a3ec22627355fc08730ad7e90022e374763d333f
87102Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87103Date: Tue Jun 24 20:32:24 2008 -0400
87104
87105 Fix a leak in the code that parses the protocol names.
87106
87107 Also added some comments.
87108 Reported by Ben Gamari (bug #16492).
87109
87110commit a4cb25f8c816adc3bdf6a28d5beb558e8be40121
87111Author: Ben Gamari <bgamari@gmail.com>
87112Date: Tue Jun 24 12:08:35 2008 +0930
87113
87114 dix: free FocusPrivates when deleting a window. #16493
87115
87116 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
87117
87118 X.Org Bug 16493 <http://bugs.freedesktop.org/show_bug.cgi?id=16493>
87119
87120commit 32a7bbbba6fec0535a8faa1d6d2ecebe786bcc81
87121Author: Peter Hutterer <peter@cs.unisa.edu.au>
87122Date: Tue Jun 24 09:28:05 2008 +0930
87123
87124 xfree86: fix up leftover allocation of xf86Events.
87125
87126 Thanks to Eamon Walsh for spotting this.
87127
87128commit bcc88a1d46c7ba469c4e2861875f84f8c77d4b15
87129Author: Peter Hutterer <peter@cs.unisa.edu.au>
87130Date: Sun Jun 22 22:34:30 2008 +0930
87131
87132 xkb: remove unused variable is_core.
87133
87134commit 9ed9d98dec58d1f7468d02c4b951811b8e1ca7c1
87135Author: Julien Cristau <jcristau@debian.org>
87136Date: Tue Jun 24 22:36:32 2008 +0200
87137
87138 x86emu: fix build, fallout from 9719354
87139
87140commit e9b4f3cd856e2401675452c387b4155246239f91
87141Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87142Date: Mon Jun 23 16:18:20 2008 -0400
87143
87144 Fix "warning: suggest parentheses around && within ||".
87145
87146commit 8910dde96ec4318c9c44fb9717b6b52fad56cd1d
87147Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87148Date: Mon Jun 23 14:51:38 2008 -0400
87149
87150 Fix "warning: assignment discards qualifiers from pointer target type".
87151
87152commit 95d4ede538fbb68049ba3efa0acb0e9712e5cb01
87153Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87154Date: Fri Jun 20 23:49:32 2008 -0400
87155
87156 Fix "warning: passing argument 1 of `modeIsPresent' from incompatible pointer type".
87157
87158commit fa1545eeda2a7feebc006840a44d390177634231
87159Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87160Date: Fri Jun 20 23:47:22 2008 -0400
87161
87162 Fix "warning: passing argument 1 of `xf86addNewLoadDirective' from incompatible pointer type".
87163
87164commit 2725ba1fca45e5ba241695e24333d8efab1f862b
87165Author: Adam Jackson <ajax@redhat.com>
87166Date: Tue Jun 24 14:41:32 2008 -0400
87167
87168 Also remove xorgconfig from configure.ac.
87169
87170commit 9719354ae04a41e9834ac9ba8fd9e895092ba4f1
87171Author: Adam Jackson <ajax@redhat.com>
87172Date: Tue Jun 24 14:37:06 2008 -0400
87173
87174 Check for __amd64__, not __x86_64__.
87175
87176 Spiritual revert of 1fa4de80fcfc697b5e5879cc351fb3e9dbf6acbe. Intel's C
87177 compiler claims to be gcc-compatible; if they're not defining the same
87178 macros as gcc then that's their bug, not ours. Even if we were to do
87179 this aliasing we should do it once and for all in servermd.h.
87180
87181commit 249c892784ca5e8c75863dd82097ca2bedec4723
87182Author: Roland Bär <roland@verifysoft.de>
87183Date: Tue Jun 24 14:13:02 2008 -0400
87184
87185 Bug #11857: Avoid running off the end of a MAXCLIENTS-sized array.
87186
87187commit 9e5b3deafb97ec1f83e6bfe067bc68df7385bc6a
87188Author: Adam Jackson <ajax@redhat.com>
87189Date: Tue Jun 24 14:08:04 2008 -0400
87190
87191 Bug #11842: Fix emulation of int1A PCI BIOS services.
87192
87193 Use only %di to name the PCI register to read/write, rather than %edi.
87194 DOS is only expecting the base PCI config space anyway, and the BIOS
87195 might be using the high bits of %edi.
87196
87197commit d34430414ac0e77eec61ab0ac9ef427b236eb639
87198Author: Adam Jackson <ajax@redhat.com>
87199Date: Tue Jun 24 13:32:55 2008 -0400
87200
87201 Remove xorgconfig. Use X -configure instead.
87202
87203commit 574c62e305e5380bb1b0f4bfec4f7cc3e3f03ada
87204Author: Owen Taylor <otaylor@fishsoup.net>
87205Date: Tue Jun 24 12:40:47 2008 -0400
87206
87207 Bug #15369: Don't reduce Over to Src for transform + RepeatNone.
87208
87209 Only do no-src-alpha optimizations for a RepeatNone source if we can
87210 easily probe that we won't sample outside the edges of the source.
87211
87212commit 100afae578e59e31b65a6264c281ba7deea5ac39
87213Author: Adam Jackson <ajax@redhat.com>
87214Date: Tue Jun 24 10:52:31 2008 -0400
87215
87216 Bug #15586: (Correctly) refuse to redirect the root window.
87217
87218commit 4c4e06af7950df509fa02099788be66cf37a4d01
87219Author: Paul Bender <pebender@gmail.com>
87220Date: Tue Jun 24 10:44:47 2008 -0400
87221
87222 Bug #15665: Fix building without Composite.
87223
87224commit 59510a76436c739fd9421f01cc4210607d9e7f57
87225Author: Adam Jackson <ajax@redhat.com>
87226Date: Tue Jun 24 10:40:36 2008 -0400
87227
87228 Remove some default modes that really shouldn't be in the default set.
87229
87230commit b622f6dc5d8d70d17be0374a85a9b16b43a62a01
87231Author: Julien Cristau <jcristau@debian.org>
87232Date: Mon Jun 23 23:45:57 2008 +0200
87233
87234 kdrive: fix building with --disable-xv (bug#16488)
87235
87236commit 3b587826924c60f6736dc5faf584b964fde1b4da
87237Author: Jesse Ruffin <jesse@eloquentpeasant.net>
87238Date: Mon Jun 23 14:35:20 2008 -0400
87239
87240 Bug #16302: Fix GLX drawable refcounting in DMX.
87241
87242commit c8d066a15142678041c1d82ccf530dcdb2ea74ca
87243Author: Adam Jackson <ajax@redhat.com>
87244Date: Mon Jun 23 13:59:48 2008 -0400
87245
87246 Implement bswap in x86emu.
87247
87248 Yes, this is a 486+ instruction and thus not strictly legal in vm86
87249 mode, but enough BIOSes use it (looking at you VIA) that we might as
87250 well implement it.
87251
87252commit b55fbca4f0705aeff1c69d3ef851c5ff5af6ed94
87253Author: Adam Jackson <ajax@redhat.com>
87254Date: Fri Jun 20 13:27:32 2008 -0400
87255
87256 Bug #12414: Create full-fledged pixmaps in fb24_32ReformatTile().
87257
87258 ... instead of creating pixmaps that only fb knows about, which will
87259 have no devPrivates for any other subsystem and thus cause havoc if
87260 (when) they leak out.
87261
87262commit 155fcecb320dbcfb899d6bcd78048ce0993bf28b
87263Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87264Date: Mon May 19 08:09:33 2008 +0200
87265
87266 Various messages clean-ups.
87267
87268commit 41a65e110409a290288180e8e1517f053f877586
87269Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87270Date: Mon May 19 06:12:04 2008 +0200
87271
87272 Fix ipc check on OpenBSD
87273
87274commit ce4d4b3c1b0e25f64d6f781b09528b2f3ae33207
87275Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87276Date: Mon May 19 05:19:01 2008 +0200
87277
87278 build fix
87279
87280commit 19c7e9da55646f1f6e05c28cd71865cd8d84e1ff
87281Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87282Date: Mon May 19 02:24:17 2008 +0200
87283
87284 preserve errno around the SIGIO handler
87285
87286commit 0a81d476388c51f8aa723400dffc375e112ea97a
87287Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87288Date: Mon May 19 02:20:44 2008 +0200
87289
87290 Remove RCS Id.
87291
87292commit c0d7ad9c88172d753b932069743ae65f75de3aa9
87293Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87294Date: Thu Jun 19 16:51:18 2008 +0200
87295
87296 Let code build with gcc 2.95
87297
87298commit 219dd2d44a9da40427ad48b0f09b981fe0551f4b
87299Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87300Date: Sun May 18 13:53:32 2008 +0200
87301
87302 sprintf() -> snprintf()
87303
87304commit cfcb3da75e807dec225cc7ea469e04d2db10bf73
87305Author: Keith Packard <keithp@keithp.com>
87306Date: Wed Jun 18 11:37:00 2008 -0700
87307
87308 Make button down state a bitmask. Master buttons track union of slave buttons
87309
87310 Mixing usage where some parts of the code treated this field as a bitmask
87311 and other parts as an array of card8 was wrong, and as the wire protocol
87312 wanted bitmasks, it was less invasive to switch the newer counting code use
87313 booleans.
87314
87315 Master devices track slave buttons by waiting for all slave buttons to be
87316 released before delivering the release event to the client.
87317
87318 This also removes the state merging code in DeepCopyDeviceClasses -- that
87319 code was changing master device state without delivering any events,
87320 violating protocol invariants. The result will be that existing slave
87321 button state which does not match the master will not be visible through the
87322 master device. Fixing this would require that we synthesize events in this
87323 function, which seems like a bad idea. Note that keyboards have the same
87324 issue.
87325
87326 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
87327 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
87328
87329commit 3cc5ae6a4f725483612c00fc8bcc2c61607f66a8
87330Author: Peter Hutterer <peter@cs.unisa.edu.au>
87331Date: Wed Jun 18 16:14:42 2008 +0930
87332
87333 dix: don't attempt to send Enter/Leave events if we have a keyboard.
87334
87335 Dereferencing into dev->valuator could crash the server, although it looks
87336 like I could only reproduce this by having a keyboard send an event after it
87337 was created and the WM was still replaying. Or so.
87338
87339commit 528b4e36ade482df99747081688ae52cfaeb28eb
87340Author: Alan Hourihane <alanh@tungstengraphics.com>
87341Date: Wed Jun 18 22:34:02 2008 +0100
87342
87343 Set driverPriv immediately on CreatePixmap.
87344
87345 If it's NULL anyway, we bail, if not, it lets
87346 ModifyPixmapHeader know about the private.
87347
87348commit 068fa63496c7c363706f0d356b5132bc591c04c6
87349Author: Michel Dänzer <michel@tungstengraphics.com>
87350Date: Wed Jun 18 09:23:09 2008 +0930
87351
87352 dix: fix a crash when the client has a different byteorder.
87353
87354 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
87355
87356commit d21155a3e9b51df946766926bc6155c8972c4439
87357Author: Peter Hutterer <peter@cs.unisa.edu.au>
87358Date: Sun Jun 15 20:00:41 2008 +0930
87359
87360 input: fix up usage of button->down, used to be a bitmask, is now an array.
87361
87362 device->button->down used to be a 32-byte bitmask with one bit for each
87363 button. This has changed into a 256-byte array, with one byte assigned for
87364 each button. Some of the callers were still using this array as a bitmask
87365 however, this is fixed with this patch.
87366
87367 Thanks to Keith Packard for pointing this out. See also:
87368 http://lists.freedesktop.org/archives/xorg/2008-June/036202.html
87369
87370commit 2b9c829bdebd16910bdf48b9d64862e3d34f5b7f
87371Author: Peter Hutterer <peter@cs.unisa.edu.au>
87372Date: Wed Jun 18 09:55:53 2008 +0930
87373
87374 Xi: don't drop the grab status in XExtendedGrabDevice().
87375
87376commit 4f3ed54dc9ea7780ff03378b13de0fddfdec2e15
87377Author: Peter Hutterer <peter@cs.unisa.edu.au>
87378Date: Tue Jun 17 10:43:36 2008 +0930
87379
87380 mi: remove trailing whitespaces from misprite.c
87381
87382commit 79dd600942bbac3c6b531f284b42c7b2c822da90
87383Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87384Date: Tue Jun 17 19:09:44 2008 -0400
87385
87386 SELinux: Add an extension alias under the OS-agnostic "Flask" name.
87387
87388commit 656d3d7623c6b83024e9cdc60d1257f4d87aa268
87389Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87390Date: Tue Jun 17 17:55:11 2008 -0400
87391
87392 SELinux: Bring server-side name registry up to date.
87393
87394commit f3490d3eba94c7d9e760d6d21991ab6677196c2b
87395Author: Ian Romanick <idr@us.ibm.com>
87396Date: Tue Jun 17 10:35:24 2008 -0400
87397
87398 Bug #15169: Make the server build again on Alpha.
87399
87400 Still won't work until the kernel makes the resource files actually exist.
87401
87402commit 8c9234a163eceda2abc0a2523e0f5587ea399935
87403Author: Keith Packard <keithp@keithp.com>
87404Date: Sun Jun 15 10:01:28 2008 +0930
87405
87406 Xi: avoid double button mapping.
87407
87408 Button events were mapped once in GetPointerEvents and then again in
87409 UpdateDeviceState. While it might make sense to just fix up UpdateDeviceState,
87410 it turns out to be better to leave the raw button number in the event because
87411 DGA reports raw device events without button translation, and so when it calls
87412 UpdateDeviceState, the button down counts get scrambled and buttons get stuck
87413 down.
87414
87415 See also:
87416 http://lists.freedesktop.org/archives/xorg/2008-June/036201.html
87417
87418 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
87419
87420commit d894e86aa5bea81147c4a7ad8630bcfb04156792
87421Author: Peter Hutterer <peter@cs.unisa.edu.au>
87422Date: Sun Jun 15 10:55:08 2008 +0930
87423
87424 randr: remove superfluous check against uninitialised variable.
87425
87426 Thanks to Eamon Walsh for pointing this out.
87427
87428commit 5544c51447f551dfc6df64438873a7ce64743976
87429Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87430Date: Fri Jun 13 22:49:47 2008 -0400
87431
87432 Fix "warning: the address of ‘newTypes’ will always evaluate as ‘true’".
87433
87434commit 2391c409a2840d61fed93832650c0d6c82ebebdf
87435Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87436Date: Fri Jun 13 22:48:17 2008 -0400
87437
87438 Fix "warning: unused variable `s'".
87439
87440commit 61dafe384a88fc397d7cd55cb23f4fdc3fb0645c
87441Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87442Date: Fri Jun 13 22:47:10 2008 -0400
87443
87444 Fix "warning: unused variable `tsize'".
87445
87446commit 37a64b02df02d42a19d4b777930dc5e457d97d63
87447Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87448Date: Fri Jun 13 22:41:58 2008 -0400
87449
87450 Fix "warning: ‘rc’ may be used uninitialized in this function".
87451
87452commit e90fcd8294fe763c97610d39cab50f2836f87dca
87453Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87454Date: Fri Jun 13 22:26:43 2008 -0400
87455
87456 Fix "warning: no previous prototype for ‘miRROutputGetProperty’".
87457
87458commit 0b00bbca5df140edbfaffab627bd82165f1b1420
87459Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87460Date: Fri Jun 13 22:22:22 2008 -0400
87461
87462 Fix "Warning: no previous prototype for `SecureRPCInit'".
87463
87464commit 1c2f038d69605b0ed1067e47dc75ca6f216991f5
87465Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87466Date: Fri Jun 13 21:59:09 2008 -0400
87467
87468 Fix multiple warnings in os/xdmauth.c.
87469
87470commit 384ebe02b354bf1ad0a2ac25e4b5290f2e795ea1
87471Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87472Date: Fri Jun 13 21:49:01 2008 -0400
87473
87474 Fix "warning: label ‘done’ defined but not used".
87475
87476commit 720e44f9730b9e275579483c2a34b16bba77e146
87477Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87478Date: Fri Jun 13 21:42:37 2008 -0400
87479
87480 Fix "warning: no previous prototype for ‘XevieExtensionInit’".
87481
87482commit 9d871e19c9dee5d3ee7f123249e8890417d83093
87483Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87484Date: Fri Jun 13 21:35:29 2008 -0400
87485
87486 Fix "warning: unused variable ‘pDev’, `pPointer'".
87487
87488commit 60f64af075e8ac3e63586452f2f3d338a05785f0
87489Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87490Date: Fri Jun 13 21:23:31 2008 -0400
87491
87492 Fix "warning: passing argument 1 of 'pixman_fill' from incompatible pointer type".
87493
87494commit c448460fdc23262f2b711a32599090c85c70447a
87495Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87496Date: Fri Jun 13 21:05:26 2008 -0400
87497
87498 Fix "warning: pointer targets in assignment differ in signedness".
87499
87500commit 17ed6242e73c5f733601b15732ab30c23acf14d2
87501Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87502Date: Fri Jun 13 21:03:17 2008 -0400
87503
87504 Fix "warning: assignment from incompatible pointer type".
87505
87506commit 02a8b118cd07d98f12aab8325d8443125703ba7d
87507Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87508Date: Fri Jun 13 21:01:50 2008 -0400
87509
87510 Fix "warning: implicit declaration of function ‘miPointerGetScreen’".
87511
87512commit 2d7ba09dc4b5eff5dba8d7867f285111574b1737
87513Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
87514Date: Fri Jun 13 16:39:40 2008 -0400
87515
87516 Make devPrivates lookup functions ABI instead of static inlines.
87517
87518 This is required to preserve compatibility across changes to the
87519 internal representation of the privates list.
87520
87521commit 9e0e558f263a132babf8c91d31230831653f19c1
87522Author: Adam Jackson <ajax@redhat.com>
87523Date: Fri Jun 13 11:17:49 2008 -0400
87524
87525 Fix unterminated endif.
87526
87527commit 6bbfadb9094bb6437a6825b8293d13a32e041aff
87528Author: Adam Jackson <ajax@redhat.com>
87529Date: Fri Jun 13 11:00:50 2008 -0400
87530
87531 ... and stay dead.
87532
87533commit f31bd087e8a7f65cd588bd1d022bb18e72b2a60c
87534Author: Adam Jackson <ajax@redhat.com>
87535Date: Fri Jun 13 10:35:07 2008 -0400
87536
87537 Death to mfb.
87538
87539 Anyone still interested in 1 or 4 bpp framebuffers, talk to fb.
87540
87541commit af86c3b7ce705c1711c8cf8bf05a7a02b8afccf6
87542Author: George Sapountzis <gsap7@yahoo.gr>
87543Date: Sat May 24 21:16:34 2008 +0300
87544
87545 clean .gitignore
87546
87547commit 3108980f5e78c129b23ff32f72564cf9a746c693
87548Author: George Sapountzis <gsap7@yahoo.gr>
87549Date: Fri Jun 13 15:58:56 2008 +0300
87550
87551 glx: drop unused cleargc
87552
87553commit 23b55a61f89f69454a3b0e3413b1f07d5fdf43aa
87554Author: Michel Dänzer <michel@tungstengraphics.com>
87555Date: Fri Jun 13 11:13:56 2008 +0200
87556
87557 AIGLX/DRI1: Switch to server context for calling pScreen->GetImage.
87558
87559 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=16292 .
87560
87561commit d15b3790307053587df8daed1936ff6923881b63
87562Author: Adam Jackson <ajax@redhat.com>
87563Date: Thu Jun 12 16:00:50 2008 -0400
87564
87565 Remove xgl.
87566
87567 This code hasn't been updated with anything even resembling what anyone is
87568 shipping in nearly thirty months. It hasn't built out of the box since
87569 7.1. Most of its features over AIGLX are accomplished with DRI2 and
87570 friends.
87571
87572commit c37a93b4850866c713e3cbf39a08bbc9645e24d4
87573Author: Dave Airlie <airlied@redhat.com>
87574Date: Thu Jun 12 14:54:56 2008 +1000
87575
87576 modes: make aspect choosing work on single output case.
87577
87578 In the single output enabled case we never enter the loop and test
87579 never gets set and so we fail to match a good mode.
87580
87581 This was causing my 2560x1600 to end up at 2048x1536.
87582
87583commit 23e71ef71a178505494d4b410f9314acfff81524
87584Author: Dave Airlie <airlied@redhat.com>
87585Date: Thu Jun 12 09:04:24 2008 +1000
87586
87587 dbe: fix DoS reported by iDefense.
87588
87589 This isn't a security problem just a user could DoS themselves for fun or profit.
87590
87591commit 390b155135a451fa728b73c8107aea91dfd56ed1
87592Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87593Date: Wed Jun 11 18:21:33 2008 -0300
87594
87595 Welcome to hotplug era.
87596
87597 `rm -rf xorg.conf` does not imply an error message. Maybe a warning.
87598
87599 checkInput() is defined but not used.
87600
87601commit 6e018f6a30dab1259b8fdc1ae4c84056b2acf362
87602Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87603Date: Wed Jun 11 18:10:55 2008 -0300
87604
87605 No need for inputPending field in xf86Info.
87606
87607commit 881e551de92e5664f92ef1961092e35cd7a77b15
87608Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87609Date: Wed Jun 11 18:03:33 2008 -0300
87610
87611 CopyGetMasterEvent() can be static.
87612
87613commit 656949cbe24648a5b58f7df6ed49b25af0b47601
87614Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87615Date: Wed Jun 11 17:57:19 2008 -0300
87616
87617 Remove more unused things.
87618
87619 In function ‘miSpriteCloseScreen’:
87620 warning: unused variable ‘pDev’
87621
87622commit 84db8837d0faaac008be61df9f1261cd3028e923
87623Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
87624Date: Wed Jun 11 17:47:00 2008 -0300
87625
87626 Remove unused xf86DeviceCursorUndisplay() function.
87627
87628commit dba5455f06c439c7e7d399f30dc7067e98e6e50b
87629Author: Kristian Høgsberg <krh@redhat.com>
87630Date: Wed Jun 11 11:41:34 2008 -0400
87631
87632 Un-K&R shm extension.
87633
87634commit 9171206db349a0c6fda719746be0b15049d57aaa
87635Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87636Date: Tue Jun 10 12:23:03 2008 -0600
87637
87638 CVE-2008-2362 - RENDER Extension memory corruption
87639
87640 Integer overflows can occur in the code validating the parameters for
87641 the SProcRenderCreateLinearGradient, SProcRenderCreateRadialGradient
87642 and SProcRenderCreateConicalGradient functions, leading to memory
87643 corruption by swapping bytes outside of the intended request
87644 parameters.
87645
87646commit 5257a0f83d5f3d80d0cd44dd76d047bac3869592
87647Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87648Date: Tue Jun 10 12:22:30 2008 -0600
87649
87650 CVE-2008-2361 - RENDER Extension crash
87651
87652 An integer overflow may occur in the computation of the size of the
87653 glyph to be allocated by the ProcRenderCreateCursor() function which
87654 will cause less memory to be allocated than expected, leading later to
87655 dereferencing un-mapped memory, causing a crash of the X server.
87656
87657commit c5f69b297b1227cb802394fa90efdbe1de607f3c
87658Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87659Date: Tue Jun 10 12:21:26 2008 -0600
87660
87661 CVE-2008-2360 - RENDER Extension heap buffer overflow
87662
87663 An integer overflow may occur in the computation of the size of the
87664 glyph to be allocated by the AllocateGlyph() function which will cause
87665 less memory to be allocated than expected, leading to later heap
87666 overflow.
87667
87668commit 063f18ef6d7bf834225ddfd3527e58c078628f5f
87669Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87670Date: Tue Jun 10 12:20:43 2008 -0600
87671
87672 CVE-2008-1379 - MIT-SHM arbitrary memory read
87673
87674 An integer overflow in the validation of the parameters of the
87675 ShmPutImage() request makes it possible to trigger the copy of
87676 arbitrary server memory to a pixmap that can subsequently be read by
87677 the client, to read arbitrary parts of the X server memory space.
87678
87679commit 95d162c4389857d960da9b0158345c1714e91f31
87680Author: Matthieu Herrb <matthieu.herrb@laas.fr>
87681Date: Tue Jun 10 12:20:00 2008 -0600
87682
87683 CVE-2008-1377 - RECORD and Security extensions memory corruption
87684
87685 Lack of validation of the parameters of the
87686 SProcSecurityGenerateAuthorization SProcRecordCreateContext
87687 functions makes it possible for a specially crafted request to trigger
87688 the swapping of bytes outside the parameter of these requests, causing
87689 memory corruption.
87690
87691commit 656d5d98855eb608ec6581f8c574f343a216ea32
87692Author: Peter Hutterer <peter@cs.unisa.edu.au>
87693Date: Wed Jun 11 17:32:59 2008 +0930
87694
87695 kdrive: fix wrong use of &, should be | instead.
87696
87697 Fallout from 26e7e69ab893d1f2b35213250ada40ec90944f62
87698
87699commit 6528eb885dea0326aacdd026d44d82ab07f4a4fd
87700Author: Peter Hutterer <peter@cs.unisa.edu.au>
87701Date: Wed Jun 11 15:01:36 2008 +0930
87702
87703 Xext: init firstValuator to zero, otherwise core XTest events may get lost.
87704
87705commit 6ab311c92d036dc63ee92ab4bb231f5f61ebe69f
87706Author: Peter Hutterer <peter@cs.unisa.edu.au>
87707Date: Mon Jun 9 11:45:57 2008 +0930
87708
87709 dix: Call RealizeCursor during InitializeSprite.
87710
87711 RealizeCursor should be called when the cursor is allocated. However, when the
87712 root cursor is allocated, no devices exist yet, and thus RealizeCursor is
87713 never called. This may lead to segfaults lateron in DDXes like Xnest that
87714 actually need to do something for each cursor, and lateron rely on that
87715 DDX-specific data for each cursor has been initialized.
87716
87717commit e083b5a07507d53de0d1d365ef1565346284c997
87718Author: Peter Hutterer <peter@cs.unisa.edu.au>
87719Date: Mon Jun 9 11:23:53 2008 +0930
87720
87721 xnest: wrap the xnest cursor sprite funcs around the mi funcs.
87722
87723 Modelled after the xfree86 code. Call miDCInitialize to init the SW rendering
87724 engine, then take the pointers, store it in a xnest-local variable, and put
87725 the xnest-specific sprite funcs in place. In the xnest sprite funcs, call
87726 through to the mi sprite funcs after doing xnest-specific stuff.
87727
87728commit ea6a02c048e8589e99ee3b459f3612eb3d041558
87729Author: Peter Hutterer <peter@cs.unisa.edu.au>
87730Date: Mon Jun 9 11:22:33 2008 +0930
87731
87732 mi: protect mipointrst.h against multiple includes.
87733
87734commit a3a7c12fcf8e4ac1418f9ea53f76091f309a721b
87735Author: Peter Hutterer <peter@cs.unisa.edu.au>
87736Date: Sun Jun 8 23:38:45 2008 +0930
87737
87738 xfree86: Fix up xf86ScaleAxis
87739
87740 Some driver still call it, so we might as well work correctly. Always
87741 resetting X to the Sxhigh is sub-optimal.
87742
87743commit f8431a62d57ab3118a0b409bce977db8261e2930
87744Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87745Date: Tue Jun 10 02:15:25 2008 -0700
87746
87747 XQuartz: Under the new startup model, we no longer need to do the foreground/background dance.
87748 (cherry picked from commit 4505bae5d7341e1241be50c25cb3d1b941701de4)
87749
87750commit c180a52332385b98635068d63a8fe0bba4835be7
87751Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87752Date: Tue Jun 10 10:47:37 2008 -0700
87753
87754 XQuartz: Removed code path for old startup
87755 (cherry picked from commit a9ee6b0d00fab01a78408a85e6542e88c19fda7c)
87756
87757commit 49751fee3b82ebc4917bfb168ec78aad7874f1f1
87758Author: Roland Scheidegger <sroland@tungstengraphics.com>
87759Date: Tue Jun 10 15:40:48 2008 +0200
87760
87761 glx: copy msaa visual capabilities
87762
87763commit 5170c169db78cad3fe47dee4e438b48e67f0c7d1
87764Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87765Date: Tue Jun 10 01:39:15 2008 -0700
87766
87767 XQuartz: Fixed the "laggy" startup under the new model. This was caused by xinit blocking expecting SIGUSR1 and our stub never sending it.
87768 (cherry picked from commit ee92aced10d0743c4658e53b58b5d9f5a094a415)
87769
87770commit 01ace5f3fd56bc789ad7448059fcf907ef19a1c8
87771Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87772Date: Wed Jun 4 18:28:20 2008 -0700
87773
87774 XQuartz: Updated icon with more rsolutions and made it more dark-background friendly
87775 (cherry picked from commit d8cf5623faab952a0f1196c8fe10baa09d1fc089)
87776
87777commit 607b0d09ea003f87cfb3331e59e13495a483832a
87778Author: Aaron Plattner <aplattner@nvidia.com>
87779Date: Mon Jun 9 09:52:04 2008 -0700
87780
87781 CreateColormap returns Success on success, not TRUE.
87782
87783 Fixes a problem where enabling color index overlays disables the RENDER
87784 extension.
87785
87786commit 2854abd39d4deb79c9ee9bcc931fad969392cfc5
87787Author: Peter Hutterer <peter@cs.unisa.edu.au>
87788Date: Sun Jun 8 23:20:11 2008 +0930
87789
87790 xnest: fix up sprite funcs to stop compiler warnings.
87791
87792commit c68ca0c7b2c72df0501d26bf7f640029ff2d9568
87793Author: Peter Hutterer <peter@cs.unisa.edu.au>
87794Date: Sun Jun 8 23:14:20 2008 +0930
87795
87796 xnest: switch to using EventList rather than xEvents.
87797
87798commit 34429d16d319ef2709c22a5c22f45d0e82c78b20
87799Author: Peter Hutterer <peter@cs.unisa.edu.au>
87800Date: Sun Jun 8 23:08:34 2008 +0930
87801
87802 xnest: call miDCInitialize rather than miPointerInitialize.
87803
87804 This unfortunately gives us a lots of artefacts, so cursor rendering cannot be
87805 assumed complete. But it's better than crashing.
87806
87807commit a7d4bec884c24c0e1792fd7d6677cdd0a69978e3
87808Author: Peter Hutterer <peter@cs.unisa.edu.au>
87809Date: Sun Jun 8 23:03:20 2008 +0930
87810
87811 mi: remove leftover old license comment.
87812
87813commit 585125685b5f09245021c8ee15f57950fe6cdd20
87814Author: Peter Hutterer <peter@cs.unisa.edu.au>
87815Date: Sun Jun 8 22:43:47 2008 +0930
87816
87817 ephyr: some whitespace changes.
87818
87819commit 782394fe53220deafed06ad282b5f9ce42964c42
87820Author: Julien Cristau <jcristau@debian.org>
87821Date: Fri Jun 6 12:10:35 2008 +0200
87822
87823 mi: Fix typo in comments about deprecated functions
87824
87825commit 6b96281100f2118fe9d99536c33d48298a5bce7b
87826Author: Michel Dänzer <michel@tungstengraphics.com>
87827Date: Fri Jun 6 11:01:03 2008 +0200
87828
87829 EXA: Fix exaGetPixmapFirstPixel() crash if the driver has a CreatePixmap hook.
87830
87831 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=16243
87832
87833commit d25ffcfbfcfe8e059f2889873c9dbe76a28ad00c
87834Author: Peter Hutterer <peter@cs.unisa.edu.au>
87835Date: Thu Jun 5 15:29:35 2008 +0930
87836
87837 mi: protect against possible NULL-pointer dereference.
87838
87839commit 874dcdb3bdb9320aae252d3c7927d412685059d5
87840Author: Peter Hutterer <peter@cs.unisa.edu.au>
87841Date: Thu Jun 5 15:38:48 2008 +0930
87842
87843 dmx: remove a ifndef XINPUT, XINPUT is always defined now.
87844
87845commit 52752911ea8470dd8805f5e6c3febb9e1e1cc4b3
87846Author: Peter Hutterer <peter@cs.unisa.edu.au>
87847Date: Wed Jun 4 17:04:16 2008 +0930
87848
87849 dmx: don't free event list after use.
87850
87851commit 5c5e58124581b845a6bfa7a2991a4e0ee7f7ebdb
87852Author: Peter Hutterer <peter@cs.unisa.edu.au>
87853Date: Wed Jun 4 16:58:31 2008 +0930
87854
87855 dmx: add an XFlush.
87856
87857 Beats me why we need it, but without it we get segfaults lateron.
87858
87859commit 246c10441bbf62646b77993b55233d5ceb93a81f
87860Author: Peter Hutterer <peter@cs.unisa.edu.au>
87861Date: Wed Jun 4 16:37:25 2008 +0930
87862
87863 dmx: fix false memory allocation.
87864
87865 beNumVisuals and the number of GLX Visuals can be significantly different.
87866
87867commit 17cd26225749a1daf7460b44e52f43082399b265
87868Author: Peter Hutterer <peter@cs.unisa.edu.au>
87869Date: Wed Jun 4 16:05:36 2008 +0930
87870
87871 dmx: don't dereference a nullpointer.
87872
87873commit 75eb635e3543fcb731331d01f50e62b696967667
87874Author: Peter Hutterer <peter@cs.unisa.edu.au>
87875Date: Tue Jun 3 22:25:33 2008 +0930
87876
87877 dmx: add stubs for DeviceCursorInitialize, DeviceCursorCleanup
87878
87879commit 3ff2f3a00ef3c484d1d53569b6cbc1f25dc6bd35
87880Author: Peter Hutterer <peter@cs.unisa.edu.au>
87881Date: Tue Jun 3 22:17:14 2008 +0930
87882
87883 dmx: for now, don't acknowledge SDs.
87884
87885commit 5bcd9e8953dfb5f13e90baeebea1f6eac797644b
87886Author: Peter Hutterer <peter@cs.unisa.edu.au>
87887Date: Tue Jun 3 22:01:43 2008 +0930
87888
87889 dmx: learn about the existence of IsXExtensionPointer and IsXExtensionKeyboard
87890
87891commit 8da8a0fec4b1b9d9208635dedb2f449dc99e0004
87892Author: Peter Hutterer <peter@cs.unisa.edu.au>
87893Date: Tue Jun 3 21:59:33 2008 +0930
87894
87895 dmx: claim we support XI 2.
87896
87897 We don't really, yet, but at least we get the full device list this way.
87898
87899commit 6f1d5147cb394b7c83c4c1447c78890fa2b40a45
87900Author: Peter Hutterer <peter@cs.unisa.edu.au>
87901Date: Tue Jun 3 21:36:03 2008 +0930
87902
87903 dmx: fix a segfault caused by GC devPrivates never being initalised.
87904
87905commit d10ba4591a599a317093ffcbdd7cece11ab6de26
87906Author: Peter Hutterer <peter@cs.unisa.edu.au>
87907Date: Tue Jun 3 21:28:26 2008 +0930
87908
87909 dmx: some more build fixes.
87910
87911commit 8d4d0b47a07a298a20ffae9fefe96c8c7ca9dccc
87912Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
87913Date: Thu Jun 5 09:19:16 2008 +0930
87914
87915 gl: include assert.h if we're compiling with DEBUG.
87916
87917 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
87918
87919commit 8644aa47177bf20f3aa85181032ce75c08270fc5
87920Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
87921Date: Thu Jun 5 08:55:09 2008 +0930
87922
87923 mi: minor build fix when compiling with debug enabled.
87924
87925 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
87926
87927commit fbf4b5f16ac23359872361c8c95a617fe8c92292
87928Author: Peter Hutterer <peter@cs.unisa.edu.au>
87929Date: Wed Jun 4 11:10:46 2008 +0930
87930
87931 dix: set dst->mapWidth when allocating a new map.
87932
87933commit ff3adf3e564d94fea18e48f966de40a7ded1279e
87934Author: Peter Hutterer <peter@cs.unisa.edu.au>
87935Date: Wed Jun 4 11:10:11 2008 +0930
87936
87937 xkb: reset xkb_cached_map on CloseDownDevices.
87938
87939 Could lead to some invalid pointers in the second server generation.
87940
87941commit 40855d8000b9b9f4e6c6b4317542378ecf516345
87942Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87943Date: Wed Jun 4 12:19:28 2008 -0700
87944
87945 XQuartz: Removed async debugging sleep
87946 (cherry picked from commit 7812a8bdf9fab651ea5c07b852b2999547ec628d)
87947
87948commit ee86b751192b690973ee2a1446a406bc721ce8eb
87949Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87950Date: Wed Jun 4 12:01:19 2008 -0700
87951
87952 XQuartz: use a condition variable to signal when darwinEvents is ready rather than polling
87953 (cherry picked from commit ff1c443cadf11d12a7d939e51194f6105153870e)
87954
87955commit 38da26cd36957a45b2a47ef124282f7d863a9fd3
87956Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87957Date: Wed Jun 4 11:35:24 2008 -0700
87958
87959 XQuartz: Don't forget to destroy the mutex and cond after we're done with them
87960 (cherry picked from commit c3558bb8cd889e5b957190e9f5d23afad1e17b72)
87961
87962commit ea40fcf4344c6e918d3be34354528940b05a70f3
87963Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87964Date: Wed Jun 4 11:18:52 2008 -0700
87965
87966 XQuartz: Fork for trigger
87967 (cherry picked from commit dd0f8a0f59593d7831fe09a2a086fcd57c84910e)
87968
87969commit 2393dae6ffff7db8281a7977039450e8754cc19f
87970Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
87971Date: Tue Jun 3 00:06:26 2008 -0700
87972
87973 XQuartz: Switched over to new startup path for testing. Cleaned it up a bit. Server still crashes when using the icon to launch =/
87974 (cherry picked from commit 7f840e9dc180421eaa9b0ea3ab993fdd5b2466e5)
87975
87976commit 45b661c67aca1a728f1793dd94d701488f1f196c
87977Author: Peter Hutterer <peter@cs.unisa.edu.au>
87978Date: Wed Jun 4 13:39:39 2008 +0930
87979
87980 dix: Fix build with --disable-xinerama #16204
87981
87982 X.Org Bug 16204 <http://bugs.freedesktop.org/show_bug.cgi?id=16204>
87983
87984commit 26e7e69ab893d1f2b35213250ada40ec90944f62
87985Author: Peter Hutterer <peter@cs.unisa.edu.au>
87986Date: Mon Jun 2 11:04:41 2008 +0930
87987
87988 kdrive: don't post motion event if there was no motion. #16179
87989
87990 Based on the patch by Tomas Janousek.
87991
87992 X.Org Bug 16179 <http://bugs.freedesktop.org/show_bug.cgi?id=16179>
87993
87994commit ac1db454497ff1839aa7f7ee0649ede112824ed9
87995Author: Peter Hutterer <peter@cs.unisa.edu.au>
87996Date: Mon Jun 2 10:40:10 2008 +0930
87997
87998 xfree86: suspend signals while removing a device (corrected version).
87999
88000 Block/Release is now symmetrical.
88001
88002commit 95ecaa411aefbcd8e2e00814132cb07b4be4e5bf
88003Author: Peter Hutterer <peter@cs.unisa.edu.au>
88004Date: Mon Jun 2 10:38:45 2008 +0930
88005
88006 Revert "xfree86: suspend signals while removing a device."
88007
88008 Left the signals blocked hanging after removing a master device.
88009
88010 This reverts commit 74372fd0049676787904f1f9f80f836b97f5f60e.
88011
88012commit 5ba07cb54803b8c63e25e8ca1f1787860f684612
88013Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
88014Date: Sun Jun 1 18:31:52 2008 -0300
88015
88016 kdrive: changes ephyr API according mpx merge.
88017
88018 This fix host-cursor in Xephyr server.
88019
88020 Signed-off-by: Tiago Vignatti <vignatti@c3sl.ufpr.br>
88021 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88022
88023commit 5a3d06b8f42473cea3741dc722a775deaa2b73f6
88024Author: Peter Hutterer <peter@cs.unisa.edu.au>
88025Date: Sun Jun 1 15:20:42 2008 +0930
88026
88027 xkb: delete default rules when devices are closed.
88028
88029 We only have one set of default rules options in xkb. When the second keyboard
88030 is brought up with Xkb options specified, these new options overwrite the old.
88031 In future server generations, the rules used for the VCK are a mixture of the
88032 default ones and ones previously specified for other keyboards. Simply
88033 resetting the xkb default rules to NULL avoids this issue.
88034
88035 Reproducable by setting XkbLayout "de" and XkbVariant "nodeadkeys". In the
88036 second server generation, the VCK has "us(nodeadkeys)". This again produces a
88037 SIGABRT when the first key is hit.
88038
88039 I could not figure out why the SIGABRT happens. This patch is avoiding the
88040 issue rather than fixing it.
88041
88042commit 5ebe76f13fb6e96cebe5a386024a73eceeb828ff
88043Author: Peter Hutterer <peter@cs.unisa.edu.au>
88044Date: Sat May 31 22:13:03 2008 +0930
88045
88046 config: init dev to NULL to shut up a valgrind warning.
88047
88048commit a7ae9cbd78193fdafcafd5d7af1f6fabbb434628
88049Author: Peter Hutterer <peter@cs.unisa.edu.au>
88050Date: Sat May 31 20:44:58 2008 +0930
88051
88052 dix: don't read valuators past numAxes.
88053
88054commit f79ad5674ed41c3626e8055cd2bac0135de928c6
88055Author: Peter Hutterer <peter@cs.unisa.edu.au>
88056Date: Sat May 31 18:46:35 2008 +0930
88057
88058 kdrive: fix up callers to KdQueueEvent, need to pass xEvent, not EventListPtr
88059
88060commit 58d12c99abc4d06cb8b43386d4bcd126f0bedf83
88061Author: Peter Hutterer <peter@cs.unisa.edu.au>
88062Date: Sat May 31 18:45:11 2008 +0930
88063
88064 mi: Remove #ifdef XINPUT, kdrive otherwise won't get events.
88065
88066commit 942086fc965fc9a24be2db4f537f0ff54db51083
88067Author: Peter Hutterer <peter@cs.unisa.edu.au>
88068Date: Sat May 31 18:17:49 2008 +0930
88069
88070 dix: ProcQueryPointer doesn't need special grab handling, PickPointer does it
88071
88072commit 6cecae0e8651b1fa58b3cd0e0a636db46f56f6a6
88073Author: Peter Hutterer <peter@cs.unisa.edu.au>
88074Date: Sat May 31 13:12:53 2008 +0930
88075
88076 dix: Remove superfluous comment.
88077
88078 This code can handle devices with < 2 valuators now.
88079
88080commit 105d28652d1fb80dd8ce8511e2605dccc8812e99
88081Author: Peter Hutterer <peter@cs.unisa.edu.au>
88082Date: Fri May 30 15:20:35 2008 +0930
88083
88084 Xext: use GPE/GKE from XTestFakeInput #16145
88085
88086 This commit fixes two problems:
88087 1) XTFI used to assemble the event itself, then passed it to the device. It's
88088 much easier to just pass the variables into GPE/GKE and let the DIX do the
88089 rest.
88090
88091 2) XTFI would pass the VCP/VCK as default device to event processing. As a
88092 result, updating LEDs would be updated on the VCK, not on the actual keyboard.
88093 Instead, we now pass the events through the last-used SD, thus toggling the
88094 LEDs on the last keyboard that sent through this MD.
88095
88096 Also some cleanup in XTFI to merge validity checks a bit closer together
88097 rather than having several different sections.
88098
88099 This breaks XTestFakeMotion with Xinerama though.
88100
88101 X.Org Bug 16145 <http://bugs.freedesktop.org/show_bug.cgi?id=16145>
88102
88103commit 341a61b608873c77b1a17fe7d145ed3ad716e5a3
88104Author: Peter Hutterer <peter@cs.unisa.edu.au>
88105Date: Sat May 31 12:01:47 2008 +0930
88106
88107 mi: reduce noisyness. DebugF is enough if device ID can't be changed.
88108
88109 Also - printing out the event type is actually helpful.
88110
88111commit 62c1a32976d571f3ced1812b8b96ed5a8e658b4a
88112Author: Peter Hutterer <peter@cs.unisa.edu.au>
88113Date: Sat May 31 10:24:37 2008 +0930
88114
88115 dix: null out dummyDev, otherwise Xephyr dereferences random pointers.
88116
88117 SendEventToAllWindow eventually causes a IsInterferingGrab(), which attempts
88118 to dereference dev->deviceGrab.grab.
88119
88120commit 150c2f55a508ed24b230f68e30ec140c0901d9ae
88121Author: Julien Cristau <jcristau@debian.org>
88122Date: Fri May 23 03:09:04 2008 +0200
88123
88124 Add xkbstr.h and xkbsrv.h to sdk_HEADERS
88125
88126commit f9bcecbe51e4cf0cb2d82ad953f8c6741ed5558b
88127Author: Peter Hutterer <peter@cs.unisa.edu.au>
88128Date: Thu May 29 17:42:49 2008 +0930
88129
88130 Xext: only free events in case of core motion events.
88131
88132commit 8cf7755a85f5c1e5da7805c12b7511581a368782
88133Author: Peter Hutterer <peter@cs.unisa.edu.au>
88134Date: Wed May 28 18:05:25 2008 +0930
88135
88136 Require x11proto 7.0.13, xextproto 7.0.4 and inputproto 1.9.99.1
88137
88138commit 9b544fa40062b342ff3aa75f8f21d0bd4cec80ab
88139Author: Peter Hutterer <peter@cs.unisa.edu.au>
88140Date: Wed May 28 12:27:07 2008 +0930
88141
88142 mi: ansify.
88143
88144commit 64677f6a167a5bee99e5dfb599a3c3a6fb5dffe7
88145Author: Peter Hutterer <peter@cs.unisa.edu.au>
88146Date: Tue May 27 21:01:26 2008 +0930
88147
88148 xfree86: fix offset error in EDID parsing. #15964
88149
88150 X.Org Bug 15964 <http://bugs.freedesktop.org/show_bug.cgi?id=15964>
88151
88152commit c8af7ce35a900ac9b898f51c1b95dabad3ba1d76
88153Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
88154Date: Fri May 23 13:50:39 2008 -0300
88155
88156 Correct a NULL pointer deference
88157
88158 The problem happens if Monitor/Card combo doesn't provide EDID info,
88159 and the XFree86-VidModeExtension extension is used.
88160
88161 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88162
88163commit 3a1278907ea8059546c7b9ab0c02714832c693e3
88164Author: Peter Hutterer <peter@cs.unisa.edu.au>
88165Date: Mon May 26 18:01:07 2008 +0930
88166
88167 xfree86: Bump XINPUT ABI to 3.0.
88168
88169commit 0050165a67bb462e0bf644a11644ad9d587c62bb
88170Author: Simon Thum <simon.thum@gmx.de>
88171Date: Mon May 26 17:56:08 2008 +0930
88172
88173 input: change dxremaind, dyremaind to floats for better acceleration #9156
88174
88175 X.Org Bug 9156 <http://bugs.freedesktop.org/show_bug.cgi?id=9156>
88176
88177 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88178
88179commit 26c1958c322be3ac4dfec9ba2c41c5202bd03709
88180Merge: 5fe5778 2958610
88181Author: Peter Hutterer <peter@cs.unisa.edu.au>
88182Date: Mon May 26 17:37:31 2008 +0930
88183
88184 Merge branch 'master' into mpx
88185
88186commit 5fe57787faa2ea4f2f57b25455d200bf924206d6
88187Author: Peter Hutterer <peter@cs.unisa.edu.au>
88188Date: Mon May 26 17:23:34 2008 +0930
88189
88190 dix: store the absolute values in the motion history.
88191
88192 The core protocol requires absolute values and it's a bit hard to get them if
88193 we only have relative ones in the history. Switch the motion history to
88194 absolute, and if we really need the relative values, we can probably generated
88195 them from the abs. ones in the future.
88196
88197commit 1883485edd7eb90c0b76bca41c71e26ae2c0b91a
88198Author: Peter Hutterer <peter@cs.unisa.edu.au>
88199Date: Mon May 26 16:35:23 2008 +0930
88200
88201 If core motion history is required, scale back to screen coords and INT16.
88202
88203commit 0b88510069bdf6d020d73b424f62c2923c0db4a4
88204Author: Peter Hutterer <peter@cs.unisa.edu.au>
88205Date: Mon May 26 13:53:45 2008 +0930
88206
88207 Ensure the motion history is merged for master devices.
88208
88209 Add each event to the master's MH as well as to the SDs. In the MD, store
88210 min/max and the actual value. When retrieving the MH, rescale all coordinates
88211 to the current coordinate range and only post those valuators that are
88212 currently active on the device.
88213
88214commit d22c25bda450f6f1dfa634f0f72b32c4bec429cd
88215Author: Peter Hutterer <peter@cs.unisa.edu.au>
88216Date: Mon May 26 09:26:18 2008 +0930
88217
88218 dix: Allocate MD's motion history to be large enough for potential valuators
88219
88220 Since we can't predict how many valuators may be in a future SD attached to an
88221 MD, we need to preallocate a history buffer that is large enough to keep
88222 MAX_VALUATORS coordinates per event.
88223
88224 In addition, the history buffer needs to memorize the coordinate ranges at the
88225 time, thus requiring MDs to store (min_val, max_val, current_val, time)
88226 instead of (current_val, time) for each motion history entry.
88227
88228 This commit only fixes the allocation.
88229
88230commit 0877de13ac6ddfb55108aa3456d47f970c6c442c
88231Author: Peter Hutterer <peter@cs.unisa.edu.au>
88232Date: Sun May 25 22:49:54 2008 +0930
88233
88234 Remove GetMotionProc from ValuatorClassRec.
88235
88236 With the MD/SD device hierarchy we need control over the generation of the
88237 motion history as well as the conversion later before posting it to the
88238 client. So let's not let the drivers change it.
88239
88240 No x.org driver currently uses it anyway, linuxwacom doesn't either so dumping
88241 it seems safe enough.
88242
88243commit 00b4339168c10dd4ce026deb8e04bfb63dfd11dc
88244Author: Peter Hutterer <peter@cs.unisa.edu.au>
88245Date: Sun May 25 22:34:11 2008 +0930
88246
88247 dmx: remove some #if 0'd code.
88248
88249commit 096117cf023dc92dcadc91677931fc6ad1feb4a0
88250Author: Peter Hutterer <peter@cs.unisa.edu.au>
88251Date: Sun May 25 10:38:33 2008 +0930
88252
88253 Xi: reset motion history when swapping device classes.
88254
88255 This isn't really the correct solution, but it'll have to do until I figured
88256 out how to transfer the history over correctly.
88257
88258commit 29586101dc11d498b212510f8dedbfeca7f8c859
88259Author: Michel Dänzer <michel@tungstengraphics.com>
88260Date: Sat May 24 20:01:41 2008 +0200
88261
88262 EXA: Only record damage generated by rendering operations.
88263
88264 Recording damage from other operations (e.g. creating a client damage record)
88265 may confuse the migration code resulting in corruption.
88266
88267 Option "EXAOptimizeMigration" appears safe now, so enable it by default. Also
88268 remove it from the manpage, as it should only be necessary on request in the
88269 course of bug report diagnostics anymore.
88270
88271commit f6d61d3d86971d6a202b46ff2fab8c8799a4d057
88272Author: Michel Dänzer <michel@tungstengraphics.com>
88273Date: Sat May 24 20:01:36 2008 +0200
88274
88275 EXA: Don't migrate twice in exaImageGlyphBlt.
88276
88277 exaPrepareAccess already handles migration.
88278
88279commit d73304398255e0c3b03a497a8d4a2f8d900eef44
88280Author: Michel Dänzer <michel@tungstengraphics.com>
88281Date: Sat May 24 20:01:31 2008 +0200
88282
88283 EXA: Don't suggest exaCopyDirty be inlined.
88284
88285 Leave the decision to the compiler toolchain.
88286
88287commit 3baf3b42e079b4e7b61c1e20df305db0724d21f8
88288Author: Michel Dänzer <michel@tungstengraphics.com>
88289Date: Sat May 24 20:01:21 2008 +0200
88290
88291 EXA: Specify region used for source pixmap migration in exaCopyNtoN.
88292
88293 Avoids excessive migration overhead in some pathological cases. See
88294 http://bugs.freedesktop.org/show_bug.cgi?id=15845 .
88295
88296commit ba557e0263eee0b75847c596f31c71492611b22c
88297Author: Peter Hutterer <peter@cs.unisa.edu.au>
88298Date: Sat May 24 17:23:10 2008 +0930
88299
88300 Xi: remove deprecated comment.
88301
88302 Axis clipping wouldn't be necessary here, it's already performed in GPE.
88303
88304commit e6cbb1e11e5da1a8b9001853c25f4e5a052e7110
88305Author: Julien Cristau <jcristau@debian.org>
88306Date: Fri May 23 23:00:40 2008 +0200
88307
88308 xfree86: fix build on GNU/kFreeBSD
88309
88310 GNU/kFreeBSD defines __FreeBSD_kernel__, but not __FreeBSD__.
88311 Unify preprocessor conditionals between variable declaration and use.
88312 Debian bug #482550.
88313
88314commit 6c72961d8fa1ab1543f1b3e2cc7d34ff6d254bf8
88315Author: George Sapountzis <gsap7@yahoo.gr>
88316Date: Fri May 23 22:39:35 2008 +0300
88317
88318 glx: fix memory corruption with r5g6b5
88319
88320 should cherry-pick to xserver-1.5
88321
88322commit 21a8052fdca2115668dfc747b0b58db437b4eb6d
88323Author: George Sapountzis <gsap7@yahoo.gr>
88324Date: Fri May 23 22:39:09 2008 +0300
88325
88326 glx: fix forgotten swrast -> SWRast
88327
88328commit 877e6c35ff1f0f110627590ac0f12fddf47de506
88329Author: George Sapountzis <gsap7@yahoo.gr>
88330Date: Fri May 23 22:38:36 2008 +0300
88331
88332 glx: missing swrast is fatal
88333
88334commit 1345c93ad4be875951256dae87bc4fd9a1b5e08a
88335Author: George Sapountzis <gsap7@yahoo.gr>
88336Date: Fri May 23 22:38:11 2008 +0300
88337
88338 glx: drop stray glcore.h include
88339
88340commit ae0504d34d245f15b2f098a63ad1a9e1c4daef35
88341Author: George Sapountzis <gsap7@yahoo.gr>
88342Date: Fri May 23 22:40:03 2008 +0300
88343
88344 kdrive: need not export symbols for glx anymore
88345
88346commit 7082929ef7db20b507dce5a8cec043e787d0b55e
88347Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88348Date: Fri May 23 02:27:58 2008 -0700
88349
88350 XQuartz: A 2 second delay for launchd socket connections should be sufficient.
88351 (cherry picked from commit 2bb4251b3c6b30dbf1a556e1b51e6f03f02d2529)
88352
88353commit 5a388b27a094c71b4214aa744b2ba0fdda7fd607
88354Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88355Date: Fri May 23 02:10:14 2008 -0700
88356
88357 XQuartz: Add the launchd fd to AllSockets as well
88358 (cherry picked from commit 83f72529394be5871671d73b6ef4f8bc83708f8a)
88359
88360commit 77b688a73025238ed9fe286a2bae8f78da65fce3
88361Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88362Date: Fri May 23 02:32:28 2008 -0700
88363
88364 XQuartz: Don't need launchd checking in configure.ac anymore since its functionality has been removed from libxtrans and into the mach startup code.
88365 (cherry picked from commit 1e1f4fb717d2f3a3227be2a034d28d6a4f29b5d1)
88366
88367commit 225dddbaeded4d64fcc104da538449e5c4dedd2d
88368Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88369Date: Fri May 23 01:39:02 2008 -0700
88370
88371 XQuartz: Move the launchd display grabbing into mach_startup
88372 (cherry picked from commit c3866c98d23020d2151977ee1177b6054d05832e)
88373
88374commit 3a22190061c7ebefee85564b0def5e1112c07ae8
88375Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88376Date: Mon May 19 03:13:09 2008 -0700
88377
88378 XQuartz: Fixed a few issues with fd passing... still not working =(
88379 (cherry picked from commit 7dd351271522b475d8017e4bd1618f12817ee2fa)
88380
88381commit 3ae0f4fc832462444d03024c5fbb867f19d5e960
88382Author: Peter Hutterer <peter@cs.unisa.edu.au>
88383Date: Fri May 23 15:02:43 2008 +0930
88384
88385 dix: update valuators > 2 as well.
88386
88387 master->last.valuator[x] for x > 2 is undefined. For all other devices, it's
88388 the respective device's last valuators.
88389
88390 If the lastSlave did not have a valuator that is to be updated now, it is
88391 reset to 0.
88392
88393commit 0d1b41ace38bc918e2cc2a0a6ad107c76346fd00
88394Author: Peter Hutterer <peter@cs.unisa.edu.au>
88395Date: Fri May 23 13:32:33 2008 +0930
88396
88397 Xi: update device valuators based on the event.
88398
88399 Event is always absolute, update the device's valuators (always absolute too),
88400 and then change the deviceValuator event to reflect the device's reporting
88401 mode.
88402
88403commit 7eec1c23a34591064ea64499acb24f22ada08dfa
88404Author: Peter Hutterer <peter@cs.unisa.edu.au>
88405Date: Fri May 23 13:23:02 2008 +0930
88406
88407 dix: don't allow relative data in deviceValuators.
88408
88409 In GPE, we don't care about the device mode. Let's put the absolute values
88410 into the deviceValuator event and worry about relative valuators on the other
88411 side of the EQ.
88412
88413commit 75551a2ef949c2cfb53fdc4b97f3964a5e2f45f5
88414Author: Peter Hutterer <peter@cs.unisa.edu.au>
88415Date: Fri May 23 12:55:50 2008 +0930
88416
88417 Revert "dix: Correct values in the device axisVal are calculated in GPE"
88418
88419 We can't rely on GPE to update device->valuators->axisVal. If a SIGIO occurs
88420 during event processing this may have incoherent results.
88421
88422 This reverts commit f6645ddbf754c80e9a8b1672519534a887622270.
88423
88424commit fe59b1a62b2c7f467d20132db4913317bbf308eb
88425Author: Peter Hutterer <peter@cs.unisa.edu.au>
88426Date: Fri May 23 12:51:52 2008 +0930
88427
88428 dix: more device scaling insanity.
88429
88430 Assuming master->last.valuators is in screen coords, SD's are always in device
88431 coordinates.
88432
88433 1. If an event comes in, scale masters->last to the device, drop into device's
88434 last->valuators.
88435 2. Apply motion from the actual event
88436 3. Scale back to screen coords, check if we may need to cross screens
88437 4. Drop screen coords into master->last
88438 5. Rescale to device coords, drop into deviceValuator event and SD->last
88439 6. Drop screen coords into ev->root_x/y
88440
88441 Whoopee...
88442
88443commit 6c9e9f8a40e20fb1761440acd2755f5fd31f4d44
88444Author: Peter Hutterer <peter@cs.unisa.edu.au>
88445Date: Fri May 23 11:51:53 2008 +0930
88446
88447 input: instead of lastx/y, use a last.valuators[] array on the device.
88448
88449 During GetPointerEvents (and others), we need to access the last coordinates
88450 posted for this device from the driver (not as posted to the client!). Lastx/y
88451 is ok if we only have two axes, but with more complex devices we also need to
88452 transition between all other axes.
88453
88454 ABI break, recompile your input drivers.
88455
88456commit fb146cbb0f28e4e480e5d16d61476ac46b5d00ce
88457Author: Peter Hutterer <peter@cs.unisa.edu.au>
88458Date: Fri May 23 12:00:13 2008 +0930
88459
88460 dix: coreEvents doesn't have meaning here, use master instead.
88461
88462 We mustn't clip x/y if we are attached, otherwise we can't change screens.
88463
88464commit fc1cc0adcb91fdfa4c547bf379a85f9558c959ff
88465Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
88466Date: Fri May 23 00:33:18 2008 +0200
88467
88468 dix: Cleanup of GetPointerEvents
88469
88470 Changed all the checks for x&y valuator so the more complex
88471 calculation is only made once.
88472 Added TODOs for valuator/axis 2 and above for future correct
88473 handling of relative reporting of these.
88474
88475 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88476
88477commit a0241d5380bb5d8b10865f8ea81a9a011de4aaf1
88478Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
88479Date: Fri May 23 00:36:11 2008 +0200
88480
88481 dix: Correct clipAxis so it can handle devices with value ranges properly
88482
88483 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88484
88485commit f6645ddbf754c80e9a8b1672519534a887622270
88486Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
88487Date: Thu May 22 23:33:45 2008 +0200
88488
88489 dix: Correct values in the device axisVal are calculated in GPE
88490
88491 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88492
88493commit 7f85acdf70c67c567de688439e25081be5a7d5df
88494Author: Peter Hutterer <peter@cs.unisa.edu.au>
88495Date: Thu May 22 23:27:15 2008 +0930
88496
88497 dix: fill valuators with the correct values depending on the device mode (GPE)
88498
88499 valuators[] is passed from the DDX. Depending on the device mode, update it
88500 with either absolute values or relative values. The deviceValuator event sent
88501 to the client will then contain the respective values.
88502
88503commit 1a3f351c50cba66f71a73239318174b09fd9b63b
88504Author: Peter Hutterer <peter@cs.unisa.edu.au>
88505Date: Thu May 22 23:25:20 2008 +0930
88506
88507 dix: when floating a device, create a new cursor sprite.
88508
88509 This is essentially necessary to allow calls to miPointerGetSprite etc. to
88510 work for floating slave devices.
88511
88512commit e0fbe404a436aef24624a3a15e8405a9ca38aadb
88513Author: Peter Hutterer <peter@cs.unisa.edu.au>
88514Date: Thu May 22 23:24:29 2008 +0930
88515
88516 mi: handle sprite even for floating slave devices.
88517
88518 We still don't render it, but we accept all the other calls to update it's
88519 internal state.
88520
88521commit 0f15875a271889ae3cc4997ad15f787ea28b3a08
88522Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
88523Date: Sat May 17 19:24:00 2008 +0200
88524
88525 Make all conversion handling in GPE.
88526
88527 This isn't quite finished yet, but at least it gives us the ability to use a
88528 tablet as a normal mouse - with all the scaling in place.
88529
88530 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88531
88532commit 74372fd0049676787904f1f9f80f836b97f5f60e
88533Author: Peter Hutterer <peter@cs.unisa.edu.au>
88534Date: Thu May 22 18:02:10 2008 +0930
88535
88536 xfree86: suspend signals while removing a device.
88537
88538 Getting a keyboard event while halfway through freeing memory can be
88539 unpleasant.
88540
88541commit 30e9a33f7d1972dcf1c29c7455b2cea4e5857913
88542Author: Peter Hutterer <peter@cs.unisa.edu.au>
88543Date: Thu May 22 16:00:21 2008 +0930
88544
88545 xkb: fix crash caused by uninitialised variable.
88546
88547commit e837244947fbb18192e9056c070ce439ef24d1a3
88548Author: Dan Nicholson <dbn.lists@gmail.com>
88549Date: Wed May 21 16:14:55 2008 -0700
88550
88551 Actually enable building GLX without Mesa source
88552
88553commit ea4ec9e9983e25d94a3edf8a77ed2ad1db193284
88554Author: Dave Airlie <airlied@redhat.com>
88555Date: Tue May 20 13:59:09 2008 +1000
88556
88557 int10: add pci_device_enable support on Linux
88558
88559commit 4230b92744d54528f48ddc3b096a265b488053c9
88560Author: Kristian Høgsberg <krh@redhat.com>
88561Date: Wed May 21 17:58:15 2008 -0400
88562
88563 Clean up remaining afb references.
88564
88565commit 20ea99c655140e101f2d20cfab78fb22765fec62
88566Author: Adam Jackson <ajax@redhat.com>
88567Date: Wed May 21 17:03:25 2008 -0400
88568
88569 Death to afb.
88570
88571 Planar framebuffer people, please talk to wfb for all your 1988 hardware
88572 needs.
88573
88574commit c76fddeea28b4f88db97a2a3487d7a57cd9d578c
88575Author: Dan Nicholson <dbn.lists@gmail.com>
88576Date: Wed May 21 10:51:15 2008 -0700
88577
88578 Fix up paths for GL/glx -> glx renaming
88579
88580commit c3eb5b80d8cc7ca6a425ccf359718d1ef3d5af55
88581Author: Kristian Høgsberg <krh@redhat.com>
88582Date: Wed May 21 13:33:36 2008 -0400
88583
88584 Move GL/glx on level up now that it's the only thing left under GL.
88585
88586commit a8a9b40e554dd0c184866978889a39519afc466a
88587Author: Kristian Høgsberg <krh@redhat.com>
88588Date: Wed May 21 12:34:07 2008 -0400
88589
88590 Move the windows GL files to the xwin DDX.
88591
88592commit e5aad4bc087036489e2af576628aa9b612054a6e
88593Author: Kristian Høgsberg <krh@redhat.com>
88594Date: Wed May 21 11:44:12 2008 -0400
88595
88596 Get out of the mesa symlinking business.
88597
88598 This copies over the files generated from mesa/src/mesa/glapi. There's
88599 a corresponding mesa commit that makes it easy to generate the glapi files
88600 straight into the xserver tree when the XML definitions change.
88601
88602 The only few files that are copied from mesa but aren't generated are
88603 glapi.[ch] and glthread.[ch]. Everything in there is technically DRI
88604 driver API and the whole setup is still a bit fragile, but it's not a new
88605 problem.
88606
88607 The --with-mesa-source configure option is still around since other
88608 parts of the server (XGL and DMX - grep for MESA_SOURCE) need that,
88609 but for common case of building with GLX and AIGLX support, that
88610 option is no longer needed.
88611
88612commit f558c7d4d1d87affe1ce4aec2e62ff64545f2bc8
88613Author: Kristian Høgsberg <krh@redhat.com>
88614Date: Wed May 21 10:18:16 2008 -0400
88615
88616 Use LD_EXPORT_SYMBOLS_FLAG when linking Xnest.
88617
88618commit 77ff8fd5072fa57158463bfd84656389eeb7b503
88619Author: Kristian Høgsberg <krh@redhat.com>
88620Date: Wed May 21 10:17:53 2008 -0400
88621
88622 Drop glcore GLX provider.
88623
88624 Obsoleted by the DRI swrast driver.
88625
88626commit 61eaef22359ae2abcee7dcd73ee8610a83776fc7
88627Author: Kristian Høgsberg <krh@redhat.com>
88628Date: Tue May 20 16:56:27 2008 -0400
88629
88630 Stop symlinking glcontextmodes.[ch] from mesa.
88631
88632commit 88f43e5d06fb5543bc40ca060b2c5d8e2673afad
88633Author: George Sapountzis <gsap7@yahoo.gr>
88634Date: Wed May 21 18:13:47 2008 +0300
88635
88636 glx: drop drm.h, sareaHandle from swrast loader
88637
88638commit 255142b61eb73843b16c2afdb95aa4d1f27e4f0d
88639Author: George Sapountzis <gsap7@yahoo.gr>
88640Date: Thu Jan 3 17:53:56 2008 +0200
88641
88642 drop xprint remnants: InitGlobals
88643
88644commit 641ce9c706ce3cbf726fb3fc9623161bb96ba127
88645Author: George Sapountzis <gsap7@yahoo.gr>
88646Date: Thu Jan 3 17:28:04 2008 +0200
88647
88648 drop xprint remnants: xpstubs
88649
88650commit 5c5a1eaf68eb11053fd5b99d24eb8024b6d5b218
88651Author: Peter Hutterer <peter@cs.unisa.edu.au>
88652Date: Wed May 21 21:51:27 2008 +0930
88653
88654 Replace UniSA with full Uni name in Author affiliation.
88655
88656 Big boss says UniSA isn't unique enough. Who am I to argue?
88657
88658commit 7509fb498c02fe1cebe4139612b8871ec877c130
88659Merge: 2ae3bed dfb7de6
88660Author: Peter Hutterer <peter@cs.unisa.edu.au>
88661Date: Wed May 21 16:25:35 2008 +0930
88662
88663 Merge branch 'hwcursor' into mpx
88664
88665commit 2ae3bed33731d800bb28a968e3a1c8066d179740
88666Author: Peter Hutterer <peter@cs.unisa.edu.au>
88667Date: Wed May 21 15:51:35 2008 +0930
88668
88669 mi: shut up compiler warning (return w/o a value)
88670
88671commit 8b3d26f5b6caff1766669deb0e2100d2dee3f185
88672Author: Aaron Plattner <aplattner@nvidia.com>
88673Date: Tue May 20 12:39:28 2008 -0700
88674
88675 Short-circuit PanoramiXTranslateVisualID after verifying that the visual actually exists first.
88676
88677 This allows using PanoramiXTranslateVisualID to test whether a given visual made
88678 it through PanoramiXVisualConsolidate.
88679
88680commit 6ff6465931c397f72db27a4fd0ca0dcac0609537
88681Author: George Sapountzis <gsap7@yahoo.gr>
88682Date: Tue May 20 16:36:22 2008 -0400
88683
88684 Add loader for the swrast dri driver.
88685
88686commit 4056595fc77c20052e226b402e38a2a914dca123
88687Author: Peter Hutterer <peter@cs.unisa.edu.au>
88688Date: Tue May 20 18:57:31 2008 +0930
88689
88690 xkb: remove superfluous checks in if statement.
88691
88692commit 925e895b869e461a9e7f135891463c56ee633cd6
88693Author: Alan Coopersmith <alan.coopersmith@sun.com>
88694Date: Mon May 19 18:43:29 2008 -0700
88695
88696 Restore return type on xf86SetScrnInfoModes
88697
88698 Seems to have been accidentally lost by commit 76943fec860315f3c93539e59a59080b8a7b3e75
88699
88700commit 7a550cefd9417c22a4397ea4d103ddc347ab1a0f
88701Author: Peter Hutterer <peter@cs.unisa.edu.au>
88702Date: Tue May 20 11:07:39 2008 +0930
88703
88704 dix: shut up two compiler warnings.
88705
88706commit f9edecd1b53a2e234def9cbd954a47c4bda2bebc
88707Author: Peter Hutterer <peter@cs.unisa.edu.au>
88708Date: Tue May 20 11:07:17 2008 +0930
88709
88710 mi: sync declaration and definition of mieqResizeEvents.
88711
88712commit e73a03d589b47ac9131f3932da7f78d2fef7bfb5
88713Author: Peter Hutterer <peter@cs.unisa.edu.au>
88714Date: Tue May 20 10:54:32 2008 +0930
88715
88716 fb: shut up two compiler warnings.
88717
88718commit 99d28c3ef37aeffe7d8ec41e45a650ba248c6958
88719Merge: fc7e256 0178b6a
88720Author: Peter Hutterer <peter@cs.unisa.edu.au>
88721Date: Tue May 20 10:20:14 2008 +0930
88722
88723 Merge branch 'master' into mpx
88724
88725 Conflicts:
88726
88727 Xext/xprint.c (removed in master)
88728 config/hal.c
88729 dix/main.c
88730 hw/kdrive/ati/ati_cursor.c (removed in master)
88731 hw/kdrive/i810/i810_cursor.c (removed in master)
88732 hw/xprint/ddxInit.c (removed in master)
88733 xkb/ddxLoad.c
88734
88735commit 0178b6a4abed0df3e90ba393709ed566105e7c2c
88736Author: Adam Jackson <ajax@redhat.com>
88737Date: Mon May 19 11:22:19 2008 -0400
88738
88739 Fix reduced-blanking mode filtering in RANDR 1.2.
88740
88741 If the monitor isn't reduced-blanking (either through EDID logic, or
88742 config file setting), then remove RB modes from the default pool. Any
88743 RB modes from the driver and config file pools will stick around though;
88744 you asked for them, you got them.
88745
88746commit bb687465d0c5aef3cc1c865e6ea67c01a8b417fb
88747Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88748Date: Mon May 19 02:45:47 2008 -0700
88749
88750 XQuartz: First stab at SCM_RIGHTS passing the $DISPLAY launchd fd from the stub to server
88751 (cherry picked from commit cccee9cfc29f85cca635df3b8dd54199b45c4df8)
88752
88753commit 4b69d22bcb79210225379da2cefb0cd8d5ffa10e
88754Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88755Date: Sun May 18 23:51:06 2008 -0700
88756
88757 XQuartz: Fixed dropped code in the !XKB blocks, disable XKB support until we figure out a solution for xkeyboard-config
88758 (cherry picked from commit 2a72309c061f7060480d150791019ce232481462)
88759
88760commit e997df8cd3c4222ea9fc3f0ebda728c4d1f15df5
88761Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88762Date: Sun May 18 23:31:41 2008 -0700
88763
88764 XQuartz: Cleaned up ListenOnOpenFD...
88765 (cherry picked from commit 6fb587d3d5fbbaee9e46cdce24d03e5d1c66d58a)
88766
88767commit 738672858d9399ecbc170500c15f90bf657502d2
88768Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88769Date: Sun May 18 23:02:12 2008 -0700
88770
88771 New XQuartz icon thanks to Simone Karin Lehmann
88772 (cherry picked from commit c27e0c8f3c5108192362735db8a80bf3a9a1cc3a)
88773
88774commit 4be2104c801340075e030d06d24ceb4d8a6d6549
88775Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88776Date: Thu May 15 12:26:32 2008 -0700
88777
88778 XQuartz: More work towards Mach-IPC startup... started working on FD passing
88779 (cherry picked from commit 27fbfeacfa16d620801ac3492b25c1e50aec8588)
88780
88781commit 82c4075d4ba3bea03341c80b754b0f9d2f62a826
88782Author: Julien Cristau <jcristau@debian.org>
88783Date: Mon May 19 03:15:11 2008 +0200
88784
88785 xfree86: fix initial output positions with Above or LeftOf and rotation
88786
88787 The fix in fa19e84714aa84a2f2e817e363d6440349d0b619 was incomplete,
88788 as it was still using the wrong output's initial rotation to compute
88789 the position.
88790
88791commit 7dca84f3ee7265119fb81d598d7d2f7363e25f1f
88792Author: Dave Airlie <airlied@redhat.com>
88793Date: Mon May 19 11:16:08 2008 +1000
88794
88795 cursor: don't dereference NULL pointer is devPrivates not yet set.
88796
88797 This fixes a bug on server recycle with ati zaphod.
88798
88799commit 5af5db5033582a84b616d5c50f4288adb0210459
88800Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88801Date: Sat May 17 14:51:09 2008 -0700
88802
88803 XQuartz: Ok, pass XQUARTZ_USE_XKB since it breaks worse without it... but we have issues when we have the keyboard configs installed... need to figure out what to do there...
88804 (cherry picked from commit 301262b07024ad960f22d99a1267fe137f5c3fce)
88805
88806commit 2408303d79297385063cae557195bd5fd3698478
88807Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88808Date: Sat May 17 14:56:53 2008 -0700
88809
88810 XQuartz: Added functionality to add a file descriptor to the connection list after the server is already running.
88811 (cherry picked from commit 543c2cd68d1ffef65d4644b860faad7191c6b9da)
88812
88813commit 01612fe612aa27262fc3c8167f52e0376941f1ef
88814Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88815Date: Thu May 15 07:08:21 2008 -0700
88816
88817 XQuartz: Disable xkb since it doesn't work after getting xkeyboard-config installed
88818
88819 Need to setup configs for the quartz keyboard
88820 (cherry picked from commit c28fecc621b1803a4d4536afbc724d141de9e6ee)
88821
88822commit 1d09deaa6e5a75420ea4f24f6b5533f0ebc3ba28
88823Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88824Date: Thu May 15 05:49:43 2008 -0700
88825
88826 XQuartz: Made DarwinSendDDXEvent a little more robust to context switching diring server init.
88827 (cherry picked from commit 5626b0949b1a6bdd2fa3ec2ef53a688084349b87)
88828
88829commit adc62f62b9f08aa6633718989ab61b61e3c1e8e2
88830Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
88831Date: Wed May 14 01:33:43 2008 -0700
88832
88833 Xquartz: Xquartz stub now can send Mach IPC to start the server.
88834 (cherry picked from commit 55a3bca8ea42f8677897f94bf0a429bdc08c6696)
88835
88836commit 93c082a060bfa8d2a204c0c57ec3507a8edea087
88837Author: Brad Smith <brad@comstyle.com>
88838Date: Sat May 17 13:14:31 2008 +0200
88839
88840 OpenBSD doesn't define __x86_64__, simplify tests.
88841
88842 Seeing as this code seems to be specific to OpenBSD I don't think
88843 __x86_64__ should have been added there at all. It appears to have
88844 been added wherever __amd64__ existed before which is wrong. I
88845 think that part of the commit should be reverted but also all four of
88846 the checks should be __OpenBSD__ && __amd64__ instead of two one
88847 direction and two flipped.
88848
88849commit 895d4e274d2b4638f0e6838dfee0e543311dba34
88850Author: Brad Smith <brad@comstyle.com>
88851Date: Sat May 17 13:05:58 2008 +0200
88852
88853 BSD's also have /dev/urandom.
88854
88855commit aad1c37b0951eae216ac323c5d8bfc6fbcf096bd
88856Author: Adam Jackson <ajax@redhat.com>
88857Date: Fri May 16 10:52:41 2008 -0400
88858
88859 RANDR 1.1 compat: remove senseless comparison against the virtual size.
88860
88861commit 14726b776d6cebb7d864b6ffa7554e1ce5637d5c
88862Author: Adam Jackson <ajax@redhat.com>
88863Date: Fri May 16 10:51:32 2008 -0400
88864
88865 xf86SetDesiredModes(): Skip disabled CRTCs first thing.
88866
88867commit 459f34b089aca4f4eee9752600c3a9e4f4e343ab
88868Author: Adam Jackson <ajax@redhat.com>
88869Date: Fri May 16 10:48:00 2008 -0400
88870
88871 Fix initial mode selection even harder.
88872
88873 The first guess used to be "is the preferred mode for one output the
88874 preferred mode on all outputs". Instead, do "find the largest mode that's
88875 preferred for at least one output and available on all outputs".
88876
88877commit 96111c154713600dd534dd82104ac18b91466202
88878Author: Adam Jackson <ajax@redhat.com>
88879Date: Fri May 16 10:31:58 2008 -0400
88880
88881 Redo RANDR compatibility output selection.
88882
88883 Old logic was just the first one that happened to have an associated
88884 CRTC. The new logic tries to find one that's definitely connected, has
88885 probed modes, and has the largest candidate mode.
88886
88887commit a4bbe1c8bca08f3df5ff7e50444af6aef7ec8b25
88888Author: Adam Jackson <ajax@redhat.com>
88889Date: Fri May 16 10:25:12 2008 -0400
88890
88891 Re-add sync range inference from legacy setup to RANDR 1.2.
88892
88893commit f52f6c5c7efc281f9ac204fbaa4f71383df7463d
88894Author: Adam Jackson <ajax@redhat.com>
88895Date: Mon May 5 14:37:07 2008 -0400
88896
88897 Fix hal shutdown crash.
88898
88899 Removing the device invalidates its ->next pointer. Copy it aside before
88900 destroying the device.
88901
88902commit 99e4531f7c3fafce065dfd21f65bad1cf190c3f1
88903Author: Adam Jackson <ajax@redhat.com>
88904Date: Fri May 16 08:16:01 2008 -0400
88905
88906 The great kdrive purge.
88907
88908 Most of these drivers didn't work. ati was the only one that even came
88909 close. The igs, ipaq, itsy, pcmcia, savage, sis530, trident, trio, ts300,
88910 and vxworks directories have never built since modularisation, so clearly
88911 no one can miss them.
88912
88913commit fc7e2566cc076c0d979f74871bc436df43401058
88914Author: Peter Hutterer <peter@cs.unisa.edu.au>
88915Date: Fri May 16 16:59:34 2008 +0930
88916
88917 Xext: fix typo in condition.
88918
88919 Fall-out from dc3aba8a559d4304844ee1cc306c577a63b82762.
88920 We must free the event if it is NOT an extension event.
88921
88922commit 6fb76acc4524d6af4bfb6bc6d862c1ee2bbb8baa
88923Author: Peter Hutterer <peter@cs.unisa.edu.au>
88924Date: Fri May 16 19:22:32 2008 +0930
88925
88926 mi: remove miPointerUpdate() -> replaced by miPointerUpdateSprite().
88927
88928 In the xnest and vfb DDX we don't need the call anyway, it's performed by
88929 mieqProcessInputEvent.
88930
88931commit dfb7de6571345061d1cd88ab915064fc376529fd
88932Author: Peter Hutterer <peter@cs.unisa.edu.au>
88933Date: Fri May 16 20:56:04 2008 +0930
88934
88935 mi: don't call from miPointerMoved into the spriteFuncs for anything but VCP.
88936
88937 Calling ->MoveCursor for anything but the HW-rendered VCP causes the
88938 SW-rendered cursor routines to be started, including mallocs, etc. Since
88939 miPointerMoved is called during SIGIO, this is a bad idea.
88940
88941commit f611719edce2ae51dca8e53947a3938b53c20b00
88942Author: Peter Hutterer <peter@cs.unisa.edu.au>
88943Date: Fri May 16 19:49:29 2008 +0930
88944
88945 xfree86: always render first cursor in HW, all others in SW.
88946
88947commit 48ba58feacedf9781db81c82adbb37197d9a5cd1
88948Author: Peter Hutterer <peter@cs.unisa.edu.au>
88949Date: Fri May 16 19:22:32 2008 +0930
88950
88951 mi: remove miPointerUpdate() -> replaced by miPointerUpdateSprite().
88952
88953 In the xnest and vfb DDX we don't need the call anyway, it's performed by
88954 mieqProcessInputEvent.
88955
88956commit 622d7c1d899a6146773a2ebd1d632a805f24025e
88957Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
88958Date: Wed May 14 14:49:09 2008 -0300
88959
88960 Restructure and organize the code.
88961
88962 It was removed and simplified some conditionals. We don't need test for
88963 pDev->isMaster inside xf86CursorSetCursor() because only MD enters there.
88964
88965 In the last chunk, ScreenPriv fields were being assigned without need, so
88966 that code was wrapped inside the conditional to avoid it.
88967
88968 I also tried to make the identation more sane in some parts that I touched.
88969
88970 Signed-off-by: Tiago Vignatti <vignatti@c3sl.ufpr.br>
88971
88972 Minor modification, part of the original patch led to cursors not being
88973 updated properly when controlled through XTest.
88974
88975 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88976
88977commit 5af90025fee6a92e7d94642978feed21f902d0cb
88978Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
88979Date: Wed May 14 02:45:49 2008 -0300
88980
88981 Remove redundancy.
88982
88983 The only function that cat set SWCursor before xf86DeviceCursorInitialize()
88984 is xf86InitCursor() when VCP and is created.
88985
88986 Signed-off-by: Tiago Vignatti <vignatti@c3sl.ufpr.br>
88987 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
88988
88989commit 9fc94edfe3df4c4a84ad70714c0a4ef8bbf57fc9
88990Author: Peter Hutterer <peter@cs.unisa.edu.au>
88991Date: Fri May 16 16:59:34 2008 +0930
88992
88993 Xext: fix typo in condition.
88994
88995 Fall-out from dc3aba8a559d4304844ee1cc306c577a63b82762.
88996 We must free the event if it is NOT an extension event.
88997
88998commit 5127942f80983b2e053dddd5c5747d0c3d2f9d6d
88999Author: Brian Rogers <brian_rogers@comcast.net>
89000Date: Fri May 16 10:08:04 2008 +0930
89001
89002 DGA: pass nevents to UpdateDeviceState. #15936
89003
89004 Missing parameter caused event processing to go nuts when checking valuators.
89005
89006 X.Org Bug 15936 <http://bugs.freedesktop.org/show_bug.cgi?id=15936>
89007
89008 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
89009
89010commit 28378d26b4bae377ef1212f6a51cda9b5529f1b5
89011Author: Peter Hutterer <peter@cs.unisa.edu.au>
89012Date: Thu May 15 11:09:15 2008 +0930
89013
89014 Xi: assemble button/modifier state before updating the device. #15934
89015
89016 The state field of the event must specify the state of the devices before the
89017 event occured. With the code as it was, the state would also include the
89018 event (e.g. state from a button press event would show the button as pressed)
89019 Gathering the state before updating the device should fix this.
89020
89021 X.Org Bug 15934 <http://bugs.freedesktop.org/show_bug.cgi?id=15934>
89022
89023commit 7cabf81c8638739a15a1be6baa3fc569f38e7589
89024Author: Eric Anholt <eric@anholt.net>
89025Date: Thu May 15 13:24:27 2008 -0700
89026
89027 Add necessary include for COMPILEDDISPLAYCLASS.
89028
89029 Fixes build after 07f69023b0e75d98a4be8f0d1c3dd9391b8149a0.
89030
89031commit 07f69023b0e75d98a4be8f0d1c3dd9391b8149a0
89032Author: Adam Jackson <ajax@redhat.com>
89033Date: Wed May 14 16:07:51 2008 -0400
89034
89035 Move defaultDisplayClass to xdmcp-only visibility.
89036
89037commit 166177e36bfc7bd1df03663ad89500cff056d563
89038Author: Adam Jackson <ajax@redhat.com>
89039Date: Wed May 14 16:02:05 2008 -0400
89040
89041 Simplify default font path setup.
89042
89043commit 2efe1abb6fc786a4f86464978ae0d23fc7347a98
89044Author: Adam Jackson <ajax@redhat.com>
89045Date: Wed May 14 15:53:11 2008 -0400
89046
89047 Remove global argc/argv variables.
89048
89049 Was only used in the logging code anymore, and uselessly so there.
89050
89051commit 6c27b911222cdee9a057de0be7e8b2eff8b1ff2a
89052Author: Adam Jackson <ajax@redhat.com>
89053Date: Wed May 14 15:44:04 2008 -0400
89054
89055 Restore compression of duplicate log file entries.
89056
89057 This undoes the workaround for bug #964, which was an Xprintism.
89058
89059commit 937e5aae33d3b5112b5d10d605e25f57b48caa3f
89060Author: Peter Hutterer <peter@cs.unisa.edu.au>
89061Date: Wed May 14 22:30:15 2008 +0930
89062
89063 dix: don't undisplay cursor if we don't own a sprite.
89064
89065 Sometimes we didn't have a cursor when coming back from suspend. Reason was
89066 that the suspend caused the server to lose the device that was attached to the
89067 VCP, and a RemoveDevice() would then set the cursor to NULL.
89068 Solution: only set the cursor to NULL if we actually own the sprite.
89069
89070commit cbb4e80eb7cc8dc2aff5e5268b49cdb7b3c3d5fa
89071Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89072Date: Wed May 14 01:13:15 2008 -0700
89073
89074 XQuartz: More work on the Mach-IPC startup path
89075 (cherry picked from commit 49cd0b185fd6c99b07357a74734b6a4023faca84)
89076
89077commit 3b57c59bb08c9a3211f4ae57d9e2fb569d61bf2f
89078Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89079Date: Tue May 13 10:40:20 2008 -0700
89080
89081 Xquartz: More work on the new Mach startup
89082 (cherry picked from commit 6237acf75d3310d7d4f262556b677557c2907284)
89083
89084commit e435acc84cb9477455ad005cee658630cbd363a2
89085Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89086Date: Mon May 12 17:57:07 2008 -0700
89087
89088 Added missing to EXTRA_DIST
89089 (cherry picked from commit e39613f4633ed992bc276b70833a703560e528f9)
89090
89091commit f7d6d20ad64b235700185784d317417a94d1814b
89092Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89093Date: Mon May 12 13:36:35 2008 -0700
89094
89095 XQuartz: Cleaned up the about box.
89096 (cherry picked from commit 0279a5970694937e949ba533330ea48961c4edba)
89097
89098commit b37e1f1f5ccc5a48df665449b0e31c4d25cc323c
89099Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89100Date: Mon May 12 11:34:06 2008 -0700
89101
89102 Don't need the fink-friendly printf in the DDX anymore.
89103 (cherry picked from commit fe2279440450c795d67ba5a2234b0797d0bfe39c)
89104
89105commit 68b4f250eef441a3d75e3b9b2665a51d3a1538d6
89106Author: Peter Hutterer <peter@cs.unisa.edu.au>
89107Date: Wed May 14 16:18:48 2008 +0930
89108
89109 dmx: fix build.
89110
89111 Just because it builds, doesn't mean it works.
89112
89113commit 00db0f35acc00cb771ab1c23f35b0db1c25d81bb
89114Author: Peter Hutterer <peter@cs.unisa.edu.au>
89115Date: Wed May 14 15:29:28 2008 +0930
89116
89117 dix: InitAndStartDevices doesn't need a window parameter.
89118
89119 Leftover from the old pointer-keyboard pairing code, obsolete and gone now.
89120
89121commit da728f2127aaa5c1ed5d22bb9f3d0c8ff19e4933
89122Author: Peter Hutterer <peter@cs.unisa.edu.au>
89123Date: Wed May 14 15:17:05 2008 +0930
89124
89125 dix: don't use serverClient in AddInputDevice, use the one supplied instead.
89126
89127 We pass in the client that wants to create the device anyway, lets use the
89128 parameter instead of hardcoding the serverClient.
89129
89130 Wow. I hope this is merge detritus, otherwise it'd be a sign that I didn't
89131 have enough coffee that day.
89132
89133commit a6659291c798f417a76d9aa7944694ff27c01fa7
89134Author: Peter Hutterer <peter@cs.unisa.edu.au>
89135Date: Wed May 14 09:42:50 2008 +0930
89136
89137 mi: protect against NULL-pointer dereference.
89138
89139 Required by the recent patch to use a NullCursor instead of UndisplayCursor().
89140
89141commit 7cdc19b29d93bf15cecfd6b69e269fab2501bca0
89142Author: Alan Coopersmith <alan.coopersmith@sun.com>
89143Date: Tue May 13 16:39:30 2008 -0700
89144
89145 When XKB fails to open rules file, log the file name, not the NULL file pointer
89146
89147commit dc3aba8a559d4304844ee1cc306c577a63b82762
89148Author: Peter Hutterer <peter@cs.unisa.edu.au>
89149Date: Tue May 13 14:37:26 2008 +0930
89150
89151 Xext: emulate XI events from core events when faking input.
89152
89153 We need XI events for event processing, so lets get rid of the pure core
89154 events ASAP.
89155
89156commit 8b3802d32041547fd6ed5393a56281c5c83d260e
89157Author: Peter Hutterer <peter@cs.unisa.edu.au>
89158Date: Tue May 13 13:47:52 2008 +0930
89159
89160 Xext: core button events have rootX/Y unset too.
89161
89162commit ed65e8b4f02a6da7f1c5d85984a9ccf6a94d0181
89163Author: Alan Coopersmith <alan.coopersmith@sun.com>
89164Date: Mon May 12 18:49:34 2008 -0700
89165
89166 Check for strcasestr and workaround it on systems without it
89167
89168commit 8234af6c6a3ade13f7720743bde79957c7df3f6c
89169Author: Peter Hutterer <peter@cs.unisa.edu.au>
89170Date: Tue May 13 11:17:39 2008 +0930
89171
89172 xfree86: remove Ubuntu-ism.
89173
89174 Additional #include was required due to broken libc headers or something like
89175 that. Remove it, there's more to the world than ubuntu alone.
89176
89177commit 59b8d29b3a91c65787dfadb5610342a62e85c48c
89178Author: Peter Hutterer <peter@cs.unisa.edu.au>
89179Date: Mon May 12 21:22:54 2008 +0930
89180
89181 Xi: Fix up ProcIVector, got out of sync with the protocol.
89182
89183 When the opcode squash happened in the protocol, the processing vector got out
89184 of sync for a few requests. As a result, client and server would interpret
89185 requests differently, leading to a couple of BadLength problems.
89186
89187commit 1fc1a2897e9185838c29d9ffff07c39d2fdf174f
89188Author: Peter Hutterer <peter@cs.unisa.edu.au>
89189Date: Mon May 12 20:14:05 2008 +0930
89190
89191 Remove UndisplayCursor API.
89192
89193 We can achieve the same thing by simply displaying a NullCursor, there's no
89194 need for a separate API.
89195
89196commit 913989d643595030e67e7c0a758a16ae0da62ed9
89197Author: Peter Hutterer <peter@cs.unisa.edu.au>
89198Date: Mon May 12 20:05:43 2008 +0930
89199
89200 mi: don't try to display a cursor for devices w/o cursors.
89201
89202commit 248a1df63430717550adb0e79068d8f9fcfacb0b
89203Author: Peter Hutterer <peter@cs.unisa.edu.au>
89204Date: Mon May 12 20:05:21 2008 +0930
89205
89206 render: don't dereference cursor if cursor is NULL.
89207
89208commit c50b5d978981b13cdb22a9ad41c1b64f90cebe51
89209Author: Aaron Plattner <aplattner@nvidia.com>
89210Date: Sun May 4 13:45:27 2008 -0700
89211
89212 Bug #14692: Allow drivers to have a say in Xinerama visual consolidation.
89213
89214 Create a new exported global variable, XineramaVisualsEqualPtr. Use this
89215 pointer to decide whether two visuals are equal during visual consolidation.
89216 This pointer can be wrapped, which allows drivers and extensions to control
89217 which visuals are consolidated. A wrapper can reject the visuals without
89218 calling down, but must call down and return that result if it deems the visuals
89219 equal. This ensures that all layers agree that the visuals are equal.
89220
89221 Pass the screen of the other visual into the VisualsEqual callchain.
89222
89223 Don't free PanoramiXVisuals since we need it for PanoramiXTranslateVisualID.
89224
89225 Don't skip the first visual on the other screen in PanoramiXMaybeAddVisual.
89226
89227 Skip the loop in PanoramiXTranslateVisualID if screen is 0.
89228
89229commit 86678e7cc2b021851ff508433fa160170f500c51
89230Author: Adam Jackson <ajax@redhat.com>
89231Date: Mon May 12 14:53:37 2008 -0400
89232
89233 Fix fbdevhw initialization for PCI drivers.
89234
89235 graphics/fb%d, not graphics:fb%d. Thanks sysfs. Thysfs.
89236
89237commit 1c8bd318fbaf65890ef16fe26c76dd5e6f14dfde
89238Author: Adam Jackson <ajax@redhat.com>
89239Date: Mon May 12 14:15:11 2008 -0400
89240
89241 X n'est pas une print API.
89242
89243commit f9fae16456c30479b0cb9317e57200af36795785
89244Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89245Date: Mon May 12 10:36:44 2008 -0700
89246
89247 XQuartz: Added some version checking protection so we don't trigger an infinite exec loop with new /usr/X11/bin/Xquartz and older X11.app
89248 (cherry picked from commit 78032815aeb10c22ff45b49702e9c9df82ab471c)
89249
89250commit ac4e33a9cd0ca2f0ec76181d11d5b90b82690c05
89251Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89252Date: Mon May 12 09:27:27 2008 -0700
89253
89254 XQuartz: More startup work... listen if we're the actual server
89255 (cherry picked from commit 3b0afb47c3d8ad922cb2315ed8034f4d77d4a249)
89256
89257commit a7800d9355377ac02833613c2ffc2423beec9970
89258Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89259Date: Fri May 9 16:44:31 2008 -0700
89260
89261 XQuartz: Starting to work on the new Mach IPC startup stuff for better launchd, ApplicationServices, and Dock support
89262 (cherry picked from commit 9b67fca9b7d3050d3d5582a5210270db7eb2ed05)
89263
89264commit 04211c3532ca078420e3745a5eac3d9de120bc32
89265Author: James Cloos <cloos@jhcloos.com>
89266Date: Mon May 12 03:03:13 2008 -0400
89267
89268 Prevent the -wm command line option from causing a SEGV
89269
89270 The -wm (when mapped) option for the BackingStore support has been
89271 causing the server to dereference a NULL pointer.
89272
89273 This has probably been the case since backing store has been
89274 implemented on top of Composite.
89275
89276 It looks like (some of?) Composite didn’t expect its WIndowPtr
89277 argument to be the root window.
89278
89279 In Composite’s compCheckRedirect() function we now avoid calling
89280 compAllocPixmap() and compFreePixmap() when the pWin pointer’s
89281 parent member is NULL, as is it the case with a server’s root window.
89282
89283 This addresses:
89284
89285 https://bugs.freedesktop.org/show_bug.cgi?id=15878
89286
89287commit 7e768c08f7809b8dba4db1931e63314e2b6e1cfa
89288Author: James Cloos <cloos@jhcloos.com>
89289Date: Mon May 12 02:53:59 2008 -0400
89290
89291 Clean up whitespace
89292
89293commit 3b6735528efc6d69ab7a7cd63dd09c87db7ef115
89294Author: Dave Airlie <airlied@redhat.com>
89295Date: Mon May 12 16:36:42 2008 +1000
89296
89297 pci: don't do this pci stuff when we don't have hw access
89298
89299commit 9dfb525f6c91acab5d1a65765a046bf9ee2aa082
89300Author: Julien Cristau <jcristau@debian.org>
89301Date: Sun May 11 23:17:27 2008 +0200
89302
89303 kdrive: allow disabling Composite
89304
89305 KdInitOutput() used to enable Composite when it was disabled by default,
89306 but now this hack prevents ``-extension Composite'' from working.
89307 Remove it, as Composite is enabled by default anyway.
89308
89309commit 1a01e96c6d15ef17a8b5ab1afa361fb12476a25e
89310Author: Alan Coopersmith <alan.coopersmith@sun.com>
89311Date: Fri May 9 15:38:44 2008 -0700
89312
89313 Return a valid X error when stuck in font alias loop
89314
89315 Part of fix for Sun bug 4258475
89316 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4258475>
89317
89318commit 7b3066d9b9099135d9c49e0682161d5568fc535b
89319Merge: 2a3d142 315f089
89320Author: Drew Parsons <drew@emerall.com>
89321Date: Sat May 10 00:01:15 2008 +1000
89322
89323 Merge branch 'master' of git://anongit.freedesktop.org/git/xorg/xserver into upstream-experimental
89324
89325commit 2a3d1421e0cc18822ae8f478fcc272e16a9e9340
89326Author: Drew Parsons <dparsons@debian.org>
89327Date: Fri May 9 23:20:11 2008 +1000
89328
89329 Disable D-BUS from Xprint.
89330
89331 Use dummy config functions to replace those from config/config.c, and
89332 therefore do not link Xprt with $CONFIG_LIB.
89333
89334 Works around an endlessly spinning loop in dix/dispatch.c::Dispatch()
89335 (WaitForSomething() not waiting) when built with dbus, which was
89336 causing Xprt to use 95% cpu.
89337
89338commit 315f089056da67d4c463ed002eb2b74e38493b49
89339Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89340Date: Thu May 8 19:46:03 2008 -0700
89341
89342 XQuartz: Reorganized some of the build system in prep for the Mach IPC startup work.
89343 (cherry picked from commit 2232c91d5c277673929eab2abb5e0495c00877cb)
89344
89345commit 28ac79450c69219dc501e072c6e5028e7136380d
89346Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89347Date: Thu May 8 19:47:40 2008 -0700
89348
89349 Updated .gitignore for new Xquartz layout
89350 (cherry picked from commit cd4d2355e227549a3410485a130549dd91ccdcfe)
89351
89352commit a07c5ad172b343ef26d2b41ff25f143950441c23
89353Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89354Date: Thu May 8 16:57:42 2008 -0700
89355
89356 XQuartz: Set bundle version to 2.3.0
89357 (cherry picked from commit 8a0524b30e1e860f3ae35741c116fc8da28aef79)
89358
89359commit 05f23ed3ea6ee0f052aee41b6573325fe0063fd8
89360Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89361Date: Tue May 6 16:12:41 2008 -0700
89362
89363 XQuartz: Fixed some issue in our bundle creation
89364 (cherry picked from commit 330ffad5477e32c5ab9ed338bc628bd5ae9f4c98)
89365
89366commit 90b963c0da2b33bdc21483f1a089b95c7e717333
89367Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89368Date: Tue May 6 16:07:33 2008 -0700
89369
89370 Set CSRG_BASED on OSX
89371 (cherry picked from commit ff085deba18682caa2f93d61a75b38db87d747b1)
89372
89373commit ff013b0da4e6d33b2b69ce1212e9bd62050574e1
89374Author: Peter Hutterer <peter@cs.unisa.edu.au>
89375Date: Thu May 8 16:58:31 2008 +0930
89376
89377 config: override xkb_{r,m,l,v} with Xkb{r,m,l,v} if the latter is set.
89378
89379 The HAL spec says that input.xkb.{rmlv}* can be sent, but if the user
89380 specifies a X-specific {rmlv}, then this is overridden through the use of
89381 input.x11_options.Xkb{RMLV}.
89382 However, the way how the server parses options--by ignoring capitalisation,
89383 underscores and spaces--the HAL and the x11_options would override each other.
89384
89385 So we simply filter the options, letting Xkb{RMLV} override xkb_{rmlv} and
89386 only actually add them to the device after parsing _all_ options.
89387
89388 * rmlv ... rules, model, layout, variant
89389
89390 See Bug 13037 <http://bugs.freedesktop.org/show_bug.cgi?id=13037>
89391 (cherry picked from commit fc35d1e3be201e3821413bb2eeb8d43e1e56ba17)
89392
89393commit 901978ebe0f446532255701cd536e246e805a55b
89394Author: Peter Hutterer <peter@cs.unisa.edu.au>
89395Date: Thu May 8 14:05:56 2008 +0930
89396
89397 config: remove trailing whitespaces.
89398
89399 It makes my vim look ugly. Put "let c_space_errors=1" into your .vimrc.
89400 (cherry picked from commit 1f54c05cf8a6b82e5fc6362f7f8e8fdc2444b9e8)
89401
89402commit 1c54c148895225e4ab3c781fe57d09e5f64353aa
89403Author: Daniel Stone <daniel@fooishbar.org>
89404Date: Fri May 9 00:26:16 2008 +0300
89405
89406 Revert "GL: Make errors non-fatal"
89407
89408 Turns out this just caused segfaults further down the line. Oops.
89409
89410 This reverts commit 268d61e00cf4bc52c05f19eda7ab4f6accce12c8.
89411
89412commit f17ba5d5849c92603f453195aca384844ca76d74
89413Author: Adam Jackson <ajax@redhat.com>
89414Date: Thu May 8 16:04:24 2008 -0400
89415
89416 Bug #13104: Remove broken XAA a1 glyph fast path.
89417
89418commit ddaecfa13cefee7c66b39b606c8640c6544d4943
89419Merge: 9c2e087 cf20df3
89420Author: Adam Jackson <ajax@redhat.com>
89421Date: Thu May 8 14:33:58 2008 -0400
89422
89423 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
89424
89425commit fc35d1e3be201e3821413bb2eeb8d43e1e56ba17
89426Author: Peter Hutterer <peter@cs.unisa.edu.au>
89427Date: Thu May 8 16:58:31 2008 +0930
89428
89429 config: override xkb_{r,m,l,v} with Xkb{r,m,l,v} if the latter is set.
89430
89431 The HAL spec says that input.xkb.{rmlv}* can be sent, but if the user
89432 specifies a X-specific {rmlv}, then this is overridden through the use of
89433 input.x11_options.Xkb{RMLV}.
89434 However, the way how the server parses options--by ignoring capitalisation,
89435 underscores and spaces--the HAL and the x11_options would override each other.
89436
89437 So we simply filter the options, letting Xkb{RMLV} override xkb_{rmlv} and
89438 only actually add them to the device after parsing _all_ options.
89439
89440 * rmlv ... rules, model, layout, variant
89441
89442 See Bug 13037 <http://bugs.freedesktop.org/show_bug.cgi?id=13037>
89443
89444commit 1f54c05cf8a6b82e5fc6362f7f8e8fdc2444b9e8
89445Author: Peter Hutterer <peter@cs.unisa.edu.au>
89446Date: Thu May 8 14:05:56 2008 +0930
89447
89448 config: remove trailing whitespaces.
89449
89450 It makes my vim look ugly. Put "let c_space_errors=1" into your .vimrc.
89451
89452commit 1a314a1ef06222977e7ccb94331ce31c17534b2a
89453Author: Peter Hutterer <peter@cs.unisa.edu.au>
89454Date: Thu May 8 13:16:48 2008 +0930
89455
89456 Revert "config: deprecate (and ignore) the use of input.xkb.whatever options."
89457
89458 daniels:
89459 "Hrm, I'd prefer to have input.xkb.{m,l,v,o} be the primary keys, and
89460 have input.x11_options be a backup for that, rather than the former
89461 being deprecated, for the reasons I listed earlier ..."
89462
89463 see http://bugs.freedesktop.org/show_bug.cgi?id=13037#c51
89464
89465 This reverts commit 26188875de1ccc84ac60bfb251e3ec43de2b9e22.
89466
89467commit cf20df39cc78203d17b99223908af388ecbf7d0e
89468Author: Daniel Stone <daniel@fooishbar.org>
89469Date: Wed May 7 22:24:19 2008 +0300
89470
89471 XKB: Actually explain keymap failures
89472
89473 When something went wrong building a keymap, try to explain to the user
89474 what it actually was, instead of the dreaded 'Failed to load XKB keymap'
89475 catch-all.
89476
89477commit 641a5f955b7b3ae04eeb6bc45fb30b0b531898e4
89478Author: Daniel Stone <daniel@fooishbar.org>
89479Date: Wed May 7 22:29:04 2008 +0300
89480
89481 Build: Ensure xf86DefModeSet.c ends in an empty line
89482
89483 This shuts up a warning.
89484
89485commit 268d61e00cf4bc52c05f19eda7ab4f6accce12c8
89486Author: Daniel Stone <daniel@fooishbar.org>
89487Date: Wed May 7 22:28:45 2008 +0300
89488
89489 GL: Make errors non-fatal
89490
89491 GLX, there's more to the world than just you. If you fail to load the
89492 software renderer, don't bring the entire server down.
89493
89494 The error path probably needs better testing on this one, but it seems
89495 mostly okay to me.
89496
89497commit 9c2e0871cfbe54e73eec1f790a7e383d08555055
89498Author: Adam Jackson <ajax@redhat.com>
89499Date: Wed May 7 13:21:26 2008 -0400
89500
89501 Bug #13104: XAA: Adapt to glyph storage changes.
89502
89503 Glyph bits are now stored in a proper pixmap, not just hanging off the
89504 end of a GlyphRec.
89505
89506commit 26188875de1ccc84ac60bfb251e3ec43de2b9e22
89507Author: Peter Hutterer <peter@cs.unisa.edu.au>
89508Date: Wed May 7 13:13:57 2008 +0930
89509
89510 config: deprecate (and ignore) the use of input.xkb.whatever options.
89511
89512 These options are still sent by some HAL implementations (e.g. HAL on FC8),
89513 and may overwrite the options set in the x11-input.fdi file.
89514 For a more detailed description of why see Bug #13037, comment 42.
89515
89516 X.Org Bug 13037 <http://bugs.freedesktop.org/show_bug.cgi?id=13037#c42>
89517
89518commit b6a0c6d4864f73a18beb841b16e9be56f2fcd77e
89519Author: Alan Coopersmith <alan.coopersmith@sun.com>
89520Date: Tue May 6 17:06:34 2008 -0700
89521
89522 Allow using libmd instead of libcrypto for SHA1 hashing in render/glyph.c
89523
89524 Builders can force one or the other by passing SHA1_LIB & SHA1_CFLAGS
89525 to configure
89526
89527commit 718652eaf9221e0eeec2c971dd7baa97f827451b
89528Author: Adam Jackson <ajax@redhat.com>
89529Date: Tue May 6 17:52:37 2008 -0400
89530
89531 Bug #13104: Don't let XAA glyph pixmaps anywhere near video memory.
89532
89533 Since glyphs are stored in pixmaps now, they can make their way into VRAM,
89534 which invalidates a bunch of fast-path assumptions in the XAA code. Thus
89535 you end up doing color-expands or WriteBitmap from la-la land and your
89536 aliased glyphs go all funny.
89537
89538 Since XAA isn't ever growing the ability to do sane glyph accel, just force
89539 glyph pixmaps into host memory by catching them at CreatePixmap time.
89540
89541commit a85d3ac87cc354093bb1e88697c44254e7721bb9
89542Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89543Date: Tue May 6 02:59:13 2008 -0700
89544
89545 XQuartz: Fixed typo
89546 (cherry picked from commit 56b7988d2662caa4d31094695b414080e4470ed4)
89547
89548commit 077ced6384abad78253e857091e78f3685965b9d
89549Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89550Date: Tue May 6 02:47:03 2008 -0700
89551
89552 XQuartz: Added uncommitted files
89553 (cherry picked from commit e414ec462cfc63f8eb7f504f526f5a2c73f51e69)
89554
89555commit 80e502c5d1f7e9221c6ae40716d6402fd28d8806
89556Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89557Date: Tue May 6 02:28:36 2008 -0700
89558
89559 Fixed up dist
89560 (cherry picked from commit f225222ba2bf4f03425107f258d60b73c88efaec)
89561
89562commit 7295e544332b0fa929f651304f9d4aca3db4a33e
89563Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89564Date: Tue May 6 00:06:19 2008 -0700
89565
89566 XQuartz: Move server bits into bundle and setup stub in /usr/X11/bin/Xquartz in prep for startup rewrite
89567 (cherry picked from commit 453a982e6382cff06ea27abba225440b07068f50)
89568
89569commit d41d677ab4118e73140ea2392e0d48eb361af1cf
89570Author: Ben Byer <bbyer@bushing.apple.com>
89571Date: Mon May 5 20:55:08 2008 -0700
89572
89573 Xquartz: experimental embedding of Sparkle.framework
89574 (cherry picked from commit b7a1a640cef8c69442859cbf89034ad362a19684)
89575
89576commit 9e0d73fd8a43647b648ec5b2f0bcaae30c03259c
89577Author: Peter Hutterer <peter@cs.unisa.edu.au>
89578Date: Tue May 6 17:52:15 2008 +0930
89579
89580 xfree86: call SetCursor for the VCP when switching between SW/HW rendering.
89581
89582 We need a manual call to SetCursor when we switch from SW to HW rendering and
89583 the other way round. This way we display the new cursor after removing the old
89584 one.
89585 In addition, we only update the internal state for the VCP's sprite. This way,
89586 when we switch back to HW rendering the state is up-to-date and wasn't
89587 overwritten with the other sprite's state.
89588
89589 The second part is a hack. It would be better to keep a state for each sprite,
89590 but then again we don't have hardware that can render multiple cursors so we
89591 might as well do with the hack.
89592
89593commit e4ebfed1785597b48b68e1bbdde3e5c4061b749f
89594Author: Dave Airlie <airlied@linux.ie>
89595Date: Tue May 6 16:46:37 2008 +1000
89596
89597 xf86edid: fix typo in debug output
89598
89599commit b9c1a57e7a98dea63cd362f714411547e728a85a
89600Author: Peter Hutterer <peter@cs.unisa.edu.au>
89601Date: Tue May 6 14:51:23 2008 +0930
89602
89603 xfree86: switch between to SW cursors if more than 1 sprite is available.
89604
89605 Switches back to HW cursors when sprites other than the VCP are removed.
89606 The current state requires the cursor to change shape once before it updates
89607 to SW / HW rendering (whatever is appropriate), e.g. by moving into a
89608 different window. Until this is done, the cursor is invisible.
89609
89610commit 3df88c17e315b5ae580096e4bc88920d1f452e83
89611Author: Peter Hutterer <peter@cs.unisa.edu.au>
89612Date: Tue May 6 11:50:20 2008 +0930
89613
89614 xfree86: re-enable hardware cursor (for a single cursor)
89615
89616 This commit enables HW rendering for cursors again, but only for a single
89617 cursor. Other cursors can be created, however they will not be visible.
89618
89619commit 6acc2acd0db2826add7c47e94e4061d169a41f88
89620Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89621Date: Mon May 5 15:32:26 2008 -0700
89622
89623 Rootless: mi doesn't let us resize root, so we need to do it ourselves...
89624 (cherry picked from commit c1ec36e28cff857664090cc8792db1ae93b783fa)
89625
89626commit efa65a0317e12c9ad34fa00fe90bf5eae9fa2670
89627Author: Keith Packard <keithp@keithp.com>
89628Date: Sun May 4 21:52:58 2008 -0700
89629
89630 Rework composite overlay window code to fix several resource management bugs.
89631
89632 The composite overlay window code had several misunderstandings of the
89633 workings of the X server, in particular error handling paths would often
89634 double-free objects. Clean all of this up by using resource destruction as
89635 the sole mechanism for freeing resource-based objects.
89636
89637commit 6c1accce87c9bd640c1b4bbc49bae7d44b1cc97b
89638Author: Keith Packard <keithp@keithp.com>
89639Date: Sun May 4 21:51:08 2008 -0700
89640
89641 Empty the borderClip of manual redirect windows. (bug 15823)
89642
89643 Thanks to Owen Taylor for root-causing this one.
89644
89645 If a TreatAsTransparent window has any area in the borderClip, that will be
89646 added to the totalClip region for use by other windows. That's wrong.
89647 Instead, simply empty the borderClip for TreatAsTransparent windows right up
89648 front.
89649
89650commit d500eeb9458336780d77baf8b7db96e1f7ff4f0d
89651Author: Aaron Plattner <aplattner@nvidia.com>
89652Date: Sun May 4 10:26:25 2008 -0700
89653
89654 Xephyr: Handle depth 30 and reject higher depths rather than crashing.
89655
89656commit f4a68f3701889950d3b98842d021f357f3913fea
89657Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89658Date: Fri May 2 17:35:57 2008 -0700
89659
89660 XQuartz: Fix mouse input offsets earlier since GetPointerEvents does not like negative (x,y) values
89661 (cherry picked from commit 8d9eab3a2ec5955cc2698fdcb1fa6ed12b2aadb7)
89662
89663commit 96fa7da3b87e2f9187ec13dad259beefdc17cd94
89664Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89665Date: Fri May 2 16:53:45 2008 -0700
89666
89667 XQuartz: Avoid a possible crash at startup due to unfavorable context switching.
89668 (cherry picked from commit ff10c37bdd09656cf2f7ee9577f5552caa1ffdb8)
89669
89670commit 6f63724b28c8b3fd85314fb95d1e4f363b610e19
89671Author: Peter Hutterer <peter@cs.unisa.edu.au>
89672Date: Fri May 2 14:56:58 2008 +0930
89673
89674 xfree86: don't try to repaint the cursor before cursors exist.
89675
89676commit 334456952930cb3e55c02fcdada2d0c074cd0520
89677Merge: 8e56fd9 3b8d534
89678Author: Peter Hutterer <peter@cs.unisa.edu.au>
89679Date: Fri May 2 11:00:14 2008 +0930
89680
89681 Merge branch 'master' into mpx
89682
89683 Conflicts:
89684
89685 Xi/exevents.c
89686 dix/devices.c
89687
89688commit a338007be6fd4302af012109d959f43ac7eab2b1
89689Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89690Date: Thu May 1 18:02:59 2008 -0700
89691
89692 XQuartz: Adjust the screen origin offset properly for multimonitor setups
89693 (cherry picked from commit f2020b9836bacd0593ac0b4c8541e32714ab02a9)
89694
89695commit 8e56fd9728d63a7a48ef44503425c6e25c7c9ffb
89696Author: Peter Hutterer <peter@cs.unisa.edu.au>
89697Date: Fri May 2 10:22:06 2008 +0930
89698
89699 dix: continue to check for invalid valuator ranges.
89700
89701 In an ideal world, the drivers would init all axes with NO_AXIS_LIMIT. In the
89702 real world, this is an ABI break, so let's just leave the old check in.
89703
89704commit 3b8d53452cd6c74d32d7759964a7cd9ee775f161
89705Author: Michel Dänzer <michel@tungstengraphics.com>
89706Date: Thu May 1 15:10:52 2008 +0200
89707
89708 EXA: Fall back in CompositeRects if the driver can't accelerate Composite.
89709
89710 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=15780 .
89711
89712commit 01c61f3d972fc2f4e5bb536dd00d8b6bbeb0fb3d
89713Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
89714Date: Thu May 1 01:02:39 2008 -0300
89715
89716 xorgcfg: Fix a crash if xorg.conf doesn't have a Files section.
89717
89718 This patch only creates a Files section if required, so if no entries are
89719 added, an empty Files section will not be created.
89720
89721 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
89722
89723commit 6899ff81e8f668382e3e2f9afdd9cdf51d0e0f7d
89724Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
89725Date: Wed Apr 30 16:38:42 2008 -0300
89726
89727 Fix an off by one read error in drmSIGIOHandler().
89728
89729 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
89730
89731commit ec0419b50534006a272278b79ee51538193661cb
89732Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89733Date: Wed Apr 30 17:30:57 2008 -0700
89734
89735 XQuartz: Moved some relevant stuff from darwin.h to darwinEvents.h
89736 (cherry picked from commit 1fcf74a436f2e19cceb3f285b89859025d94c040)
89737
89738commit 28d1e21f1523d475cb757d3cfa33737ed62c534d
89739Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89740Date: Wed Apr 30 17:30:32 2008 -0700
89741
89742 Dix: Cleaned up sanity checking in Get{Pointer,Keyboard}Events
89743
89744 XQuartz was crashing because the Appkit thread was trying to GetXXXEvents while the Xserver thread was exiting.
89745 This adds some more sanity checks and avoids that crash
89746 (cherry picked from commit 34ec4bd6acb834c0e3f9a5042a0cc3f52c6f3a68)
89747
89748commit ce36ae526d88d20ff67cd6cb429fb06f48d231f6
89749Author: Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
89750Date: Wed Apr 30 11:58:32 2008 +0300
89751
89752 Add M32R architecture support (bug #10020)
89753
89754 Still needs autotools support, so this won't actually _build_: it's just
89755 a starting point.
89756
89757commit a0e6a7d4f507c5c0a0b11adb10394af58a0a6e07
89758Author: Peter Hutterer <peter@cs.unisa.edu.au>
89759Date: Wed Apr 30 17:57:48 2008 +0930
89760
89761 dix: resize EQ to minimum size to avoid reallocs during SIGIO.
89762
89763 When a new device is added, calculate the event size needed if a DCCE event is
89764 sent and set the EQ's event size to this minimum. This avoids reallocs when a
89765 event is sent (which may happen during a SIGIO).
89766
89767commit b71b51c982706501b6229532ce342752207426bb
89768Author: Ben Byer <bbyer@bushing.apple.com>
89769Date: Tue Apr 29 23:37:58 2008 -0700
89770
89771 xquartz: created darwinTablet, to represent tablet events
89772 (cherry picked from commit 50641bce730ee03738fa0c4beb8125b34fda556f)
89773
89774commit 519581b0bab5cc6dac397877da485745f18c2120
89775Author: Ben Byer <bbyer@bushing.apple.com>
89776Date: Tue Apr 29 23:00:09 2008 -0700
89777
89778 xquartz: remove unused params from DarwinEQInit
89779 (cherry picked from commit 299a056737168ca1faefd675dce6d6ab13176be9)
89780
89781commit d70487a4c09cfeb90d996ab40a23a74b2c15be6f
89782Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89783Date: Tue Apr 29 23:59:26 2008 -0700
89784
89785 XQuartz: Fix to tablet-event handling code; we now scale
89786 more conservatively (to match Linux's Wacom driver) and
89787 we now receive all tablet-related events.
89788 (cherry picked from commit 588683cecca2cfc65a28de035cd6ee3d64ff59d2)
89789
89790commit fea39c94bcc8b635d37530d8a27ee92acab484fe
89791Author: Peter Hutterer <peter@cs.unisa.edu.au>
89792Date: Wed Apr 30 13:17:14 2008 +0930
89793
89794 dix: fix typo from last commit.
89795
89796commit ffaccc2dc91f4ca4ea10da010206a0a7d2b5540c
89797Author: Peter Hutterer <peter@cs.unisa.edu.au>
89798Date: Wed Apr 30 11:45:19 2008 +0930
89799
89800 input: replace -1 as default axis limit with NO_AXIS_LIMIT define.
89801
89802 This allows easier refacturing of the coordinate limit handling. Grepping for
89803 -1 is boring.
89804
89805commit 00acb40f2bc5bb4a1977b9b08db75630677ff787
89806Author: Peter Hutterer <peter@cs.unisa.edu.au>
89807Date: Wed Apr 30 11:29:03 2008 +0930
89808
89809 dix: fix typo in clipAxis.
89810
89811 Check needs to be (min_axis < max_axis), not (min_axis < min_axis)
89812
89813commit f65a1a62f9eac2f6620d5d469dbf6c79b2eadab7
89814Author: Hong Liu <hong.liu@intel.com>
89815Date: Tue Apr 29 09:00:11 2008 -0700
89816
89817 Keep rotation wrappers in place while transforms are in use
89818
89819 LeaveVT/EnterVT cycles will free/realloc shadow frame buffers. Because of
89820 this, the presense/absence of that data is insufficient to know whether
89821 the screen function wrappers are necessary. Instead, the 'transform_in_use'
89822 flag should be used.
89823
89824 This patch also adds 'xf86RotateFreeShadow' for drivers to use at LeaveVT
89825 time to free the rotation data; it will be reallocated on EnterVT.
89826
89827commit 18b33dd4ff46f63bad70b493b1d0b0758c961ada
89828Author: Peter Hutterer <peter@cs.unisa.edu.au>
89829Date: Tue Apr 29 15:24:55 2008 +0930
89830
89831 Xi: stop excessive use of _X_EXPORT.
89832
89833commit 2b179c32ac40f5258d95ae16da96a79fa8eea696
89834Author: Peter Hutterer <peter@cs.unisa.edu.au>
89835Date: Tue Apr 29 15:15:16 2008 +0930
89836
89837 include: add declaration for CopyGetMasterEvent to shut up compiler warning.
89838
89839commit ff36753755360aaa16ab8604a0ab78123884b57d
89840Author: Peter Hutterer <peter@cs.unisa.edu.au>
89841Date: Tue Apr 29 14:53:41 2008 +0930
89842
89843 mi: call the custom event handlers for both MD and SD.
89844
89845commit ec2fca7e6f7ce8fdf33d959b7adeaae935ec4b37
89846Author: Peter Hutterer <peter@cs.unisa.edu.au>
89847Date: Tue Apr 29 14:40:44 2008 +0930
89848
89849 xfree86: DGA needs to call into Xi.
89850
89851 Two steps: first use UpdateDeviceState() instead of replicating the device
89852 changing code. And emulate XI events instead of core events.
89853
89854commit 8208b842b7dd3b30bafdd74147bdfa63bc00cc40
89855Author: Peter Hutterer <peter@cs.unisa.edu.au>
89856Date: Tue Apr 29 14:19:42 2008 +0930
89857
89858 Xi: add IEventBase global variable. Stores event base for X Input events.
89859
89860commit 46340b740640b76b52b8e69bc67f9201c7e974c3
89861Author: Peter Hutterer <peter@cs.unisa.edu.au>
89862Date: Tue Apr 29 14:17:48 2008 +0930
89863
89864 Xi: change UpdateDeviceState to be _X_EXPORT.
89865
89866commit 3a922c5d07509703a3eeda2bbec6f332e6b0ffac
89867Author: Peter Hutterer <peter@cs.unisa.edu.au>
89868Date: Mon Apr 28 17:32:54 2008 +0930
89869
89870 Xi: apply ValuatorEvent data to the device's valuators.
89871
89872 After UpdateDeviceState, the device has the current position in absolute
89873 coordinates, the event has the correct valuator data to be delivered to the
89874 client.
89875
89876commit f2a20294fe26f4c9d245d24d065331fefd8f830d
89877Author: Peter Hutterer <peter@cs.unisa.edu.au>
89878Date: Mon Apr 28 16:56:22 2008 +0930
89879
89880 dix: store all valuators as they come from the device in the valuator events.
89881
89882 The event's sequenceNumber is mis-used to determine whether the values are
89883 relative or absolute.
89884
89885commit d5ab89f4f1acbe2614036e8934122185ac0f81ee
89886Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89887Date: Mon Apr 28 15:52:24 2008 -0700
89888
89889 XQuartz: Unset CFProcessPath... blech
89890 (cherry picked from commit ce4fbfbc75c62a092214d140c7550279aebe69ef)
89891
89892commit 0c2312b21b6700c1425baccaf9c26150b4f0fd6a
89893Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89894Date: Mon Apr 28 15:28:21 2008 -0700
89895
89896 XQuartz: xprIsX11Window can be called from the Appkit thread (see X11Application.m)
89897 (cherry picked from commit 22cf72437601c07b8a6c744b4f2f1f4cd6713e60)
89898
89899commit c3dc4bdbb018c0606d561e2dfb9a36e9297fa312
89900Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89901Date: Mon Apr 28 15:42:02 2008 -0700
89902
89903 XQuartz: Disabled DPMS extension
89904 (cherry picked from commit 1448fed9b6d484f471b9b2982c76ca921a9273b9)
89905
89906commit 07382a70c7ac9807dfb31821a4763bea2309bde6
89907Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89908Date: Mon Apr 28 12:36:54 2008 -0700
89909
89910 XQuartz: Added thread debugging to xprFrame.c
89911 (cherry picked from commit 41542502b321c697271c1752525b600872b6df96)
89912
89913commit 8349732a6720652bfbad7874a952be73a0e8e77b
89914Author: Michel Dänzer <michel@tungstengraphics.com>
89915Date: Mon Apr 28 21:09:35 2008 +0200
89916
89917 EXA: Try to accelerate non-antialiased text via the glyph cache as well.
89918
89919 Treat 1 bit glyphs and masks as PICT_a8 in the glyph cache. We're not able to
89920 accelerate them otherwise.
89921
89922commit a65d530040bb561ba88c5d8c71633a7c0bf11e89
89923Author: Michel Dänzer <michel@tungstengraphics.com>
89924Date: Mon Apr 28 21:03:12 2008 +0200
89925
89926 EXA: Accumulate glyphs whenever possible, for full benefits of the glyph cache.
89927
89928commit e7eaac59c424a205dd106fc7d70734ff4b390f28
89929Author: Michel Dänzer <michel@tungstengraphics.com>
89930Date: Mon Apr 28 21:00:55 2008 +0200
89931
89932 EXA: Glyph cache upload tweaks.
89933
89934 Track damage after using UploadToScreen directly.
89935
89936 Don't waste any effort on empty glyphs.
89937
89938commit cc08c06665ffe29ad44d023d75d0f86e5338875d
89939Author: Owen Taylor <otaylor@huygens.home.fishsoup.net>
89940Date: Mon Apr 28 21:00:55 2008 +0200
89941
89942 EXA: Use UploadToScreen() for uploads to glyph cache
89943
89944 When possible, use UploadToScreen() rather than CompositePicture()
89945 to upload glyphs onto the glyph cache pixmap. This avoids allocating
89946 offscreen memory for each glyph making management of offscreen
89947 areas much more efficient.
89948
89949commit 13fd2256300b61d88b840952d838f834523f5dd7
89950Author: Owen Taylor <otaylor@huygens.home.fishsoup.net>
89951Date: Mon Apr 28 21:00:55 2008 +0200
89952
89953 EXA: Clean up debug messages
89954
89955commit fcb5949928f1c27f67f40c094c3c673786574422
89956Author: Owen Taylor <otaylor@huygens.home.fishsoup.net>
89957Date: Mon Apr 28 21:00:54 2008 +0200
89958
89959 EXA: Fix overlapping glyphs in glyph cache
89960
89961 Allocate each cache at a different vertical position in the
89962 per-format pixmap. Fix width/height confusion when choosing
89963 the cache for a glyph.
89964
89965commit 40eb14c9482457969e0bde97c49edad536285e02
89966Author: Owen Taylor <otaylor@huygens.home.fishsoup.net>
89967Date: Mon Apr 28 21:00:54 2008 +0200
89968
89969 EXA: Add exaCompositeRects()
89970
89971 Add a function to composite multiple independent rectangles
89972 from the same source to the same destination in a single
89973 operation: this is useful for building a glyph mask.
89974
89975commit 54184110f6f3e5d7276d5431e739a4fcf0c3523e
89976Author: Owen Taylor <otaylor@huygens.home.fishsoup.net>
89977Date: Mon Apr 28 21:00:54 2008 +0200
89978
89979 EXA: Use a single large glyph cache pixmap
89980
89981 Add back exaGlyphs(); the new version copies the glyph images
89982 onto a single large glyph pixmap and draws from their to the
89983 destination surface. This reduces the management of small
89984 offscreen areas and will allow us to avoid texture unit setup
89985 between each glyph.
89986
89987commit e9734306088e12f2cd68bf347ecf8415be4f0268
89988Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89989Date: Mon Apr 28 11:47:49 2008 -0700
89990
89991 XQuartz: More startup / threading house cleaning.
89992 (cherry picked from commit 72653c24c00dfba64ce35a3d400598bcd77defc1)
89993
89994commit 4b46fc931e61bec0abd6a86062e46dd7a408e745
89995Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
89996Date: Mon Apr 28 10:50:51 2008 -0700
89997
89998 XQuartz: Updated Localizable.strings
89999 (cherry picked from commit d8d9c866b90fb24c93bd6e25fa90f8f2bf58ad34)
90000
90001commit 4017ebe5bfa7a261cd1135801756c44e9fe93ca1
90002Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90003Date: Sun Apr 27 00:01:14 2008 -0700
90004
90005 XQuartz: Added missing Xquartz.man.pre to EXTRA_DIST
90006 (cherry picked from commit 03e707987f7f32e47dd0355c6d16bfb9169a379b)
90007
90008commit 53dba5381fdd8f644e16aaa0ecb05df4dc615b23
90009Author: Peter Hutterer <peter@cs.unisa.edu.au>
90010Date: Mon Apr 28 11:37:52 2008 +0930
90011
90012 dix: if alloc of a master keyboard fails, remove the master pointer.
90013
90014commit 1fab51edfc82e1ef60dfa29fd5d93478066a3998
90015Author: Peter Hutterer <peter@cs.unisa.edu.au>
90016Date: Mon Apr 28 11:37:21 2008 +0930
90017
90018 mi: guard against NULL-pointer dereference.
90019
90020commit b5004722a208479a4bc762ff428bf4cbeb430d53
90021Author: Peter Hutterer <peter@cs.unisa.edu.au>
90022Date: Mon Apr 28 10:26:01 2008 +0930
90023
90024 Xi: Only return VCP, VCK and floating SDs to Xi 1.x clients.
90025
90026 This is better than the approach implemented with
90027 8973a3f7983240407dd6da59b3643f40e6a3d83a which disabled XI altogether for 1.x.
90028 Instead, return a device list that resembles a traditional XI setup on pre XI
90029 2.0 servers. If the client tries to open a device other than a floating SD,
90030 return a BadDevice error.
90031
90032commit 0bd28315afc3b7223f8ff9f17597db09500a9388
90033Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90034Date: Sat Apr 26 23:00:42 2008 -0700
90035
90036 Xquartz: Added missing to EXTRA_DIST
90037 (cherry picked from commit f4b963256feb03e9e6b5521bdefeb390e9a49688)
90038
90039commit ef1c52053755fa14b4ca98b22c506f73f5f4a4b7
90040Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90041Date: Sat Apr 26 19:21:05 2008 -0700
90042
90043 XQuartz: Cleaned up startup and thread creation a tad.
90044 (cherry picked from commit c861fe00e112b21ee0156d09a6cd5281642a1dcc)
90045
90046commit b114d4e861885cc5b49cd81b33ad825461811b3e
90047Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90048Date: Sat Apr 26 17:04:49 2008 -0700
90049
90050 XQuartz: Don't subtract the titlebar off of the pointer_y
90051 (cherry picked from commit 00a9567acce2b27a649cbebd4790e3043688a7d8)
90052
90053commit 684b5d8382bf0bc6cd55774226ee362e81c0adbf
90054Author: Peter Hutterer <peter@cs.unisa.edu.au>
90055Date: Sat Apr 26 19:55:59 2008 +0930
90056
90057 Xi: whoops, stray ! caused a bit of memory mess.
90058
90059commit 87071b604c4b7c3c79ab784e2c09ff691e2bc5ad
90060Author: Peter Hutterer <peter@cs.unisa.edu.au>
90061Date: Sat Apr 26 19:37:57 2008 +0930
90062
90063 Revert "Xi: disable XI for clients < XI 2.0"
90064
90065 Turns out some programs don't like this change. gnome-settings-daemon crashes
90066 hard if you tell it that XI doesn't exist. So, tell them we have XI, but leave
90067 the other change (the one that pretends no devices are available).
90068
90069 This reverts commit 8973a3f7983240407dd6da59b3643f40e6a3d83a.
90070
90071commit 7447a30fb27ed50a20a85b5a2de9afe7dea8cfa5
90072Author: Peter Hutterer <peter@cs.unisa.edu.au>
90073Date: Sat Apr 26 19:03:13 2008 +0930
90074
90075 Xi: if a pre-XI2 client tries to list the devices, pretend we don't have any.
90076
90077 XI 1.x isn't supported anymore, so let's pretend we don't have any devices.
90078 This stops clients from opening them and thus stops interference.
90079
90080commit 8973a3f7983240407dd6da59b3643f40e6a3d83a
90081Author: Peter Hutterer <peter@cs.unisa.edu.au>
90082Date: Sat Apr 26 17:48:52 2008 +0930
90083
90084 Xi: disable XI for clients < XI 2.0
90085
90086 Sorry. With the huge changes in the device handling I honestly don't know how
90087 to support XI and XI2 alongside. So let's just pretend XI doesn't exist if a
90088 client doesn't request it supporting XI2.
90089
90090commit b304b0a65cc57127cdea103f2c5114e4ea79af41
90091Author: Peter Hutterer <peter@cs.unisa.edu.au>
90092Date: Sat Apr 26 17:38:55 2008 +0930
90093
90094 Xi: add versioning support.
90095
90096 Remember the version the client sent to us, so we can adjust our replies
90097 accordingly. This requires the client to use the {major|minor}Version fields
90098 in the GetExtensionVersion request. However, they were padding before, so we
90099 must assume they are garbage if nbytes is non-zero. If nbytes is zero, the
90100 client is probably a new client and we can handle it correctly.
90101
90102commit b9ca7896356f79ee27be5d5aa62052f6984282b0
90103Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90104Date: Fri Apr 25 17:54:36 2008 -0700
90105
90106 XQuartz: Added some pseudoramiX debug traces
90107 (cherry picked from commit 5bee1585a399eab0a7b6fc80ad476d81b5d227d3)
90108
90109commit e251c9e75afdc5e32f2bc9801712272358934266
90110Author: Peter Hutterer <peter@cs.unisa.edu.au>
90111Date: Fri Apr 25 16:49:35 2008 +0930
90112
90113 Remove all traces of #ifdef XINPUT and the matching bits from the configure.ac
90114
90115commit 1d0438de176551aaeff549664f00b8bd83d465af
90116Author: Peter Hutterer <peter@cs.unisa.edu.au>
90117Date: Fri Apr 25 16:28:58 2008 +0930
90118
90119 Xi: remove RemoveOtherCoreGrabs()
90120
90121 PickPointer() returns grabbed devices, so we can't get a double grab anyway.
90122
90123commit be5ff2b8d8a392eb8611e1fcd6da4752d68a7f9d
90124Author: Peter Hutterer <peter@cs.unisa.edu.au>
90125Date: Fri Apr 25 16:25:13 2008 +0930
90126
90127 Xi: remove some superflous code
90128
90129commit 0209e46249b61974a6e3ed54a51cc36dfaf0064f
90130Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90131Date: Thu Apr 24 23:48:11 2008 -0700
90132
90133 XQuartz: More multi-monitor work... reverted Ben's workaround (worked for side-by-side only) and added more debugging.
90134 (cherry picked from commit 515b8b855ac5d2d5aef881053f73b2ad07a6dd2e)
90135
90136commit b65dbd350b8518bb90bed9cdc64dd900bc75d20c
90137Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90138Date: Thu Apr 24 21:14:04 2008 -0700
90139
90140 XQuartz: Cleaned up multi-monitor support.
90141 (cherry picked from commit c05abf0a19b0ef0fc4ace9400a095ce2521456bc)
90142 (cherry picked from commit 9112f290434c246d3e797551aaaf3a89d2006b23)
90143
90144commit b093bf3ec37367172be3b44a04f0f43890ba7d7d
90145Author: Peter Hutterer <peter@cs.unisa.edu.au>
90146Date: Fri Apr 25 15:15:31 2008 +0930
90147
90148 dix: don't search for grabbed devices, PickPointer does it for us now.
90149
90150 Follow-up to 93ee33830778b9ec85ad81496572677a30022b09.
90151
90152commit 6198bad07edc51ff4a264a9361c5e9b6c74647a6
90153Author: Peter Hutterer <peter@cs.unisa.edu.au>
90154Date: Fri Apr 25 15:02:40 2008 +0930
90155
90156 dix: remove misleading (stale) comment
90157
90158commit 93ee33830778b9ec85ad81496572677a30022b09
90159Author: Peter Hutterer <peter@cs.unisa.edu.au>
90160Date: Fri Apr 25 15:02:13 2008 +0930
90161
90162 dix: let PickPointer() return devices that are core-grabbed by the client.
90163
90164 If the client currently holds a core grab on any device, this device has
90165 priority, even over the ClientPointer.
90166
90167commit f5ac98747de921d48f36d4f3f66dbe73e26a6760
90168Author: Peter Hutterer <peter@cs.unisa.edu.au>
90169Date: Fri Apr 25 14:50:24 2008 +0930
90170
90171 Standardise copyright headers for all new files created as part of MPX.
90172
90173commit 3fc67b4205851dcffcc431a07a885828549bd9cf
90174Author: Peter Hutterer <peter@cs.unisa.edu.au>
90175Date: Fri Apr 25 11:37:32 2008 +0930
90176
90177 Xi: remove leftover chpkpair.h file.
90178
90179 Used to contain declarations for ChangePointerKeyboardPairing request
90180 handling.
90181
90182commit 89add4ee986e38b833bb58750b3e5a664efb6f4a
90183Author: Peter Hutterer <peter@cs.unisa.edu.au>
90184Date: Fri Apr 25 11:19:12 2008 +0930
90185
90186 dix: remove obsolete InitWindowPrivates().
90187
90188commit dcdc66fcfc7957aa7875fbf74fa3a55989a9a187
90189Author: Peter Hutterer <peter@cs.unisa.edu.au>
90190Date: Fri Apr 25 11:14:32 2008 +0930
90191
90192 dix: add mi.h include to stop compiler warning
90193
90194commit 97552413d213337e4bff8c89b06d58a09d722b05
90195Author: Peter Hutterer <peter@cs.unisa.edu.au>
90196Date: Fri Apr 25 11:13:05 2008 +0930
90197
90198 dix: remove "unused variable" compiler warning.
90199
90200commit d9c38e84cc492b931a2238757d438f562946e5bc
90201Author: Peter Hutterer <peter@cs.unisa.edu.au>
90202Date: Fri Apr 25 11:04:02 2008 +0930
90203
90204 dix: remove pairingClient definition.
90205
90206 This variable was used originally to determine which client is allowed to
90207 change the pointer-keyboard pairing. For now, we just let anyone change it and
90208 see how that works out.
90209
90210commit c14b858aeced81ff43723644f2de0f5c43f55755
90211Author: Peter Hutterer <peter@cs.unisa.edu.au>
90212Date: Fri Apr 25 11:02:12 2008 +0930
90213
90214 Xi: fix two compiler warnings
90215
90216 unused variable in exevents.c
90217 implicit declaration in warpdevp.c
90218
90219commit 6015b7a81252cd1729e6f4045f9b0c885c95b183
90220Author: Peter Hutterer <peter@cs.unisa.edu.au>
90221Date: Fri Apr 25 10:41:30 2008 +0930
90222
90223 Xi: handle requests through a dispatch handler.
90224
90225 Same principle as e.g. in the RandR extension, rather than having a load of
90226 if (type == XI_whatever)
90227 use an array of function pointers.
90228
90229commit cb8cb87bc6decf75832c3724687000ecb658226e
90230Author: Peter Hutterer <peter@cs.unisa.edu.au>
90231Date: Fri Apr 25 10:09:50 2008 +0930
90232
90233 Xi: remove unused #define DISPATCH
90234
90235commit 5869e3f5551e7e2d18e06a2f220b77d8c4161db1
90236Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90237Date: Thu Apr 24 19:39:27 2008 -0700
90238
90239 darwin gots /dev/urandom, too yo!
90240 (cherry picked from commit bf0144f38034bc59f108bb2c5270ff37fbe70e10)
90241
90242commit b1ae7c79db78cc3b789701f81328669f9f8ed80f
90243Author: Alan Coopersmith <alan.coopersmith@sun.com>
90244Date: Thu Apr 24 19:02:43 2008 -0700
90245
90246 Bug 14247: If pkg-config can't find openssl.pc, just link with -lcrypto
90247
90248 X.Org bugzilla #14247 <http://bugs.freedesktop.org/show_bug.cgi?id=14247>
90249
90250commit 1daae4574c19a122baf73b6f9457b0d383b1bc66
90251Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90252Date: Thu Apr 24 18:22:05 2008 -0700
90253
90254 Added missing HAVE_LAUNCHD to dix-config.h.in
90255 (cherry picked from commit 1b4b73cefbc2f3e3f6d0cb84ea202f84fb60abb5)
90256
90257commit 63853e4ad211945ed25541223207a78f83766ced
90258Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90259Date: Thu Apr 24 18:21:42 2008 -0700
90260
90261 XQuartz: Compile in missing glx source that we still need
90262 (cherry picked from commit c611335dee267e41dcd1733a6bb5206b102f804e)
90263
90264commit ae982a27e4059fecd4048d245e6aa02f8dcc97d0
90265Author: Dave Airlie <airlied@redhat.com>
90266Date: Thu Apr 24 16:37:29 2008 +1000
90267
90268 glx: test for valid read and draw privates before using them
90269
90270 this should fix a bug where f-spot exiting blows up the X server
90271
90272commit f377141912594f87144d6d7f7fdd279a101d8e6c
90273Author: Peter Hutterer <peter@cs.unisa.edu.au>
90274Date: Thu Apr 24 13:30:28 2008 +0930
90275
90276 Xi: don't attempt to send to a NULL window.
90277
90278 Only applicable when the server comes down/restarts. In this case,
90279 WindowTable[i] may be NULL. Let's not try to send an event then.
90280
90281commit aec485f2dcc87b340759d67b60e7dee7931aaec5
90282Author: Peter Hutterer <peter@cs.unisa.edu.au>
90283Date: Thu Apr 24 13:29:46 2008 +0930
90284
90285 dix: NULL out WindowTable after freeing all the windows.
90286
90287 CloseDownDevices() tries to send PresenceNotify events. If the windows are
90288 already freed, then we are accessing dangling pointers.
90289
90290commit 9ab4e2fd8eaa87dbd16835affb1aa54dcb1a619e
90291Author: Peter Hutterer <peter@cs.unisa.edu.au>
90292Date: Wed Apr 23 11:38:08 2008 +0930
90293
90294 xfree86: don't free the config-file related information in DIDR. #15645
90295
90296 In DeleteInputDeviceRequest, leave the conf_idev (which is shared with
90297 xf86ConfigLayout.input) alone for devices that were specified in the
90298 ServerLayout section of the config file. This way, in the next server
90299 generation we are left with what was the original config and can thus re-init
90300 the devices.
90301
90302 This is an addon to 6d22a9615a0e6ab3d00b0bcb22ff001b6ece02ae, an attempt to
90303 fix Bug 14418.
90304
90305 X.Org Bug 15645 <https://bugs.freedesktop.org/show_bug.cgi?id=15645>
90306 X.Org Bug 14418 <https://bugs.freedesktop.org/show_bug.cgi?id=15645>
90307
90308commit 7dc40c8eca90ec1bfab84b6f54418b64c0e62d63
90309Author: Peter Hutterer <peter@cs.unisa.edu.au>
90310Date: Thu Apr 24 12:47:29 2008 +0930
90311
90312 xfree86: Don't free master devices.
90313
90314 The previous check works in the master-branch, but doesn't work with MPX. We
90315 actually copy the SD's information into the MDs public.devicePrivate, so we
90316 need to explicitly check whether a device is a MD before freeing the module.
90317
90318commit cc13f87cd8ac54223fdb0b3d3c043dc9b9be5a42
90319Author: Peter Hutterer <peter@cs.unisa.edu.au>
90320Date: Thu Apr 24 11:43:47 2008 +0930
90321
90322 Xext: XGE: change function definition to have return value on separate line.
90323
90324 This seems to be the common style in most parts of the server.
90325
90326commit d0890c40b9a9c7965c08608e6950c078a29aac16
90327Author: Peter Hutterer <peter@cs.unisa.edu.au>
90328Date: Thu Apr 24 11:32:36 2008 +0930
90329
90330 Xext: add a few lines of comments to XGE.
90331
90332commit 745b90cde1007383ec8c887f02439a34ab427f31
90333Author: Peter Hutterer <peter@cs.unisa.edu.au>
90334Date: Thu Apr 24 11:27:22 2008 +0930
90335
90336 Xext: re-do the XGE versioning handling.
90337
90338 Basically the same approach RandR takes. Remember which one the client
90339 requested, send back the one the server supports. Also divide XGE server
90340 version (now defined in geext.c) and the client's version (still in the
90341 protocol definition).
90342
90343commit 5f3e5b3462bb02e828c70d0e1890b5a83d399d42
90344Author: Peter Hutterer <peter@cs.unisa.edu.au>
90345Date: Thu Apr 24 11:03:23 2008 +0930
90346
90347 Xext: remove trailing whitespaces in geext.{c|h}
90348
90349commit d09c520b322ba5c5f4d6b630a7c0c62e56732f82
90350Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90351Date: Wed Apr 23 10:22:34 2008 -0700
90352
90353 Moved Apple GL bits into our DDX subtree
90354 (cherry picked from commit c9356f32892978faecb30f7b7af7488820ce37ff)
90355
90356commit b05b416c622063b84747702a54ffd9a802d6fc11
90357Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90358Date: Wed Apr 23 09:59:48 2008 -0700
90359
90360 XQuartz: No need to include indirect.c a second time
90361 (cherry picked from commit 41ed532525da85d1bd2c20af5ffa28ac6d1f4996)
90362
90363commit 565492c70a280d7e749500c5c0073bdcaf175c86
90364Author: Ben Byer <bbyer@apple.com>
90365Date: Wed Apr 23 06:00:57 2008 -0700
90366
90367 xquartz: build fixes for GLX/DRI2 fallout (note: not guaranteed)
90368 (cherry picked from commit 85a5796b667461bf1fafc68c07b3a704cd8efd5b)
90369
90370commit 6f6505db51763d902142a8161e71cad44f52560c
90371Author: Ben Byer <bbyer@apple.com>
90372Date: Wed Apr 23 03:43:51 2008 -0700
90373
90374 glx / xquartz: we still need glxcontentmodes.[ch] kthx
90375 (cherry picked from commit 27545fd37f8c703b72f7cfc329778abc1ceabfa0)
90376
90377commit 4e3a8af5751ff88c0cadd612821b0ef349d250c5
90378Author: Ben Byer <bbyer@apple.com>
90379Date: Wed Apr 23 03:39:41 2008 -0700
90380
90381 xquartz: clean up linker line for main binary
90382 (cherry picked from commit 821d7400f2ff917497b2ee58ceef2b69c2d47ec7)
90383
90384commit 3685b171da51f853bfdb312d36622dfa03ee75be
90385Author: Ben Byer <bbyer@apple.com>
90386Date: Tue Apr 22 23:53:23 2008 -0700
90387
90388 xquartz: remove vestigial aglGlx.c
90389 (cherry picked from commit 40bd041e4255f26fcbdf0831e68619ae0f46ab39)
90390
90391commit f6e22d69af6bc8f63c3a46535a09e217696a679f
90392Author: Hans de Goede <j.w.r.degoede@hhs.nl>
90393Date: Wed Apr 23 12:28:30 2008 -0400
90394
90395 Prefer glxvisuals with stencil buffer for default visuals
90396
90397 The first fbconfig which has a depthbuffer > 0 and doublebuf is choosen
90398 when associating fbconfigs with the visuals, indepenent of stencil bits.
90399 This happens to work ok on intel as there all fbconfigs with a
90400 depthbuffer > 0 also have stencil bits.
90401
90402 This patch fixes this by first trying to get a fbconfig for default X visuals
90403 with both stencilbuf, depthbuf and doublebuffering, and if that fails fallback
90404 to trying to get one with only a depthbuf and doublebuffering.
90405
90406commit 00effad583713e882c3f2518bcd3da51bf4db716
90407Author: George Sapountzis <gsap7@yahoo.gr>
90408Date: Wed Apr 23 17:46:30 2008 +0300
90409
90410 xephyr: XEPHYR_DRI is identical to XF86DRI
90411
90412commit 6a9f7f28ec455d2879ca1a315ce77c48af49e7f4
90413Author: George Sapountzis <gsap7@yahoo.gr>
90414Date: Wed Apr 23 17:46:15 2008 +0300
90415
90416 xephyr: sync with mesa
90417
90418commit 7ffc68c5cc7108c9e56af25f9f9b4398f5b027df
90419Author: George Sapountzis <gsap7@yahoo.gr>
90420Date: Wed Apr 23 17:45:54 2008 +0300
90421
90422 xephyr: libGL is client lib
90423
90424commit a27c244d72603e27766ff3900ab28fe094f7397e
90425Author: George Sapountzis <gsap7@yahoo.gr>
90426Date: Wed Apr 23 17:45:37 2008 +0300
90427
90428 xephyr: Makefile cleanup 2
90429
90430commit d4d19b2f22002b1579438cb9cf08dd580c7ec005
90431Author: George Sapountzis <gsap7@yahoo.gr>
90432Date: Wed Apr 23 17:45:19 2008 +0300
90433
90434 xephyr: Makefile cleanup 1
90435
90436commit cdc4571b580a8f4fd279404215bff0fb9a5b4816
90437Author: George Sapountzis <gsap7@yahoo.gr>
90438Date: Sun Apr 20 17:06:26 2008 +0300
90439
90440 drop TOGCUP remnants
90441
90442commit 6cd9287aed3c128d9c10b9b042ba7b864143d522
90443Author: George Sapountzis <gsap7@yahoo.gr>
90444Date: Sun Apr 20 16:58:51 2008 +0300
90445
90446 drop EVI remnants
90447
90448commit a88c6b66465c039bf92a54a496516fee4d900784
90449Author: George Sapountzis <gsap7@yahoo.gr>
90450Date: Fri Apr 18 17:47:16 2008 +0300
90451
90452 glcore: drop GLcore (files)
90453
90454commit dd7a53f8f21e41e1ab43f7e684ba586abc97ef7e
90455Author: George Sapountzis <gsap7@yahoo.gr>
90456Date: Fri Apr 18 17:47:01 2008 +0300
90457
90458 glcore: drop GLcore (build system)
90459
90460commit 567d389d47dee233a973b101e04ce41c47a68f34
90461Author: George Sapountzis <gsap7@yahoo.gr>
90462Date: Fri Apr 18 17:46:18 2008 +0300
90463
90464 glcore: build from mesa, dlopen from xorg
90465
90466 * The GLcore interface is disposable
90467 * GLcore is installed in DRI_DRIVER_INSTALL_DIR which is overloaded for
90468 GLX_PROVIDER_INSTALL_DIR
90469
90470commit efb723e166e5fa89e90c7b400fb4c7979b1f50c0
90471Author: George Sapountzis <gsap7@yahoo.gr>
90472Date: Fri Apr 18 17:45:53 2008 +0300
90473
90474 glcore: prepare for dynamic loading
90475
90476 glcore gets linked with -ldl, -lpthread for s3tc and glapi
90477 xserver needs
90478 DLOPEN_LIBS - to dlopen the glcore dso
90479 LD_EXPORT_SYMBOLS_FLAG - to export symbols for glcore to use
90480
90481 the ld flag is added to kdrive only when GLX is enabled, the net overhead for
90482 Xephyr is ~155KB, could be reduced with --dynamic-list.
90483
90484commit fbad87f2ae9f97fcb43546b0fa35f1100415dfec
90485Author: Julien Cristau <jcristau@debian.org>
90486Date: Tue Apr 22 23:50:11 2008 +0200
90487
90488 autoconfig: don't call closedir() when opendir() failed
90489
90490 If opendir() fails, return from matchDriverFromFiles() immediately.
90491 Ubuntu bug 217647.
90492
90493commit 76381092e8f650ec7d1f058fa4c8a7348893f775
90494Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90495Date: Wed Apr 23 02:22:14 2008 -0700
90496
90497 XQuartz: Make sure QuartzAudioInit() gets run.
90498 (cherry picked from commit bb3d034675b70e22e78df5554cab0ec2a3d913d0)
90499 (cherry picked from commit 824b31c7f8144a67a320442abd3d854e99d2bfe2)
90500
90501commit 275cdc1c74b7e43ecd931d312469fecc8d998ed1
90502Author: Peter Hutterer <peter@cs.unisa.edu.au>
90503Date: Wed Apr 23 16:16:15 2008 +0930
90504
90505 xfree86: pass correct value to mieqEnqueue (merge detritus I guess)
90506
90507commit c3659cb414ef05da8fa09009b2b82a3deeeb4f3a
90508Author: Peter Hutterer <peter@cs.unisa.edu.au>
90509Date: Tue Apr 22 21:54:02 2008 +0930
90510
90511 dix: always send FocusIn events if mode is NotifyUngrab.
90512
90513 In the case of a NotifyUngrab, the flag for the device may already be set but
90514 we still need to send the FocusIn event.
90515
90516commit 00815b3e5223e822f306db45cd4884a22ac9f7ed
90517Author: Julien Cristau <jcristau@debian.org>
90518Date: Mon Apr 21 14:34:39 2008 +0200
90519
90520 Don't set DRI2=yes if we can't find dri2proto or libdrm
90521
90522commit e77f65768efbf05cdf363a2f41f036f74eaa45de
90523Author: Ross Burton <ross@burtonini.com>
90524Date: Tue Apr 22 18:07:46 2008 +0100
90525
90526 Reuse the existing framebuffer mode in kdrive/fbdev
90527
90528 When starting up kdrive/fbdev, if the current framebuffer mode is sensible use
90529 that unless told otherwise.
90530
90531 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
90532
90533commit c4fd1121531b0cba1a3e90fa747871d784365c7e
90534Author: Ross Burton <ross@burtonini.com>
90535Date: Tue Apr 22 18:05:53 2008 +0100
90536
90537 Add mediumraw support to the linux kdrive driver
90538
90539 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
90540
90541commit 455383db95618a05ebdbeae78423e08065f0e14e
90542Author: Ross Burton <ross@burtonini.com>
90543Date: Tue Apr 22 18:04:31 2008 +0100
90544
90545 Enable the epson kdrive driver
90546
90547 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
90548
90549commit 82b37d35af13a9f402755e167493ab256b664b9c
90550Author: Ross Burton <ross@burtonini.com>
90551Date: Tue Apr 22 16:38:14 2008 +0100
90552
90553 Add support for AVR32
90554
90555commit 20a90bef8b4993f06cf76ad05e3d4c974e1614f6
90556Author: Ross Burton <ross@burtonini.com>
90557Date: Tue Apr 22 16:37:02 2008 +0100
90558
90559 Add a default touchscreen path
90560
90561 When enabling the touchscreen, open /dev/input/touchscreen0 if no path was
90562 specified.
90563
90564commit b44e89f4683ffcfd75eaf39f1f37d7461db44689
90565Author: Ross Burton <ross@burtonini.com>
90566Date: Tue Apr 22 16:28:39 2008 +0100
90567
90568 Update to tslib 1.0
90569
90570 Update the pkgconfig checks to tslib 1.0, and fix the LIBS definitions.
90571
90572commit 3848422d2354b7a5302fda92b05b0d728190e050
90573Author: Ross Burton <ross@burtonini.com>
90574Date: Tue Apr 22 16:25:23 2008 +0100
90575
90576 Fix build when XKB is disabled.
90577
90578commit 744d0cfda74f8283801cc2d6c5eda48402455bc3
90579Author: Ben Byer <bbyer@apple.com>
90580Date: Mon Apr 21 22:06:38 2008 -0700
90581
90582 add missing DARWIN_GLX_LIBS in configure.ac
90583 (cherry picked from commit a033c0b3dbb3b963261faa39f0236457cb00ff44)
90584
90585commit 8190ef87547b704848231bde10b1cdffc6442790
90586Merge: 179a082 2ddbfd3
90587Author: Peter Hutterer <peter@cs.unisa.edu.au>
90588Date: Tue Apr 22 18:04:05 2008 +0930
90589
90590 Merge branch 'master' into mpx
90591
90592 Conflicts:
90593
90594 Xext/EVI.c
90595 Xext/appgroup.c
90596 Xext/cup.c
90597 Xext/mitmisc.c
90598 Xext/sampleEVI.c
90599 dix/window.c
90600
90601commit 179a082c26f9e562492ee2e59e7f44f949f39f9c
90602Author: Peter Hutterer <peter@cs.unisa.edu.au>
90603Date: Fri Apr 18 21:13:10 2008 +0930
90604
90605 Xext: route event through master if required (XTestFakeInput)
90606
90607commit fb784d99c521823339bf00b70b9824f735d88875
90608Author: Peter Hutterer <peter@cs.unisa.edu.au>
90609Date: Fri Apr 18 20:57:42 2008 +0930
90610
90611 mi: fix typo in comment.
90612
90613commit 8209fdbc7c7f2dd068b30ff184b5fbf00db78686
90614Author: Peter Hutterer <peter@cs.unisa.edu.au>
90615Date: Fri Apr 18 20:53:57 2008 +0930
90616
90617 Xext: Update sprite on fake input only for MDs and floating SDs.
90618
90619commit 2ddbfd345786aa39b6ccaed82a1ca5c145284ee3
90620Author: Peter Hutterer <peter@cs.unisa.edu.au>
90621Date: Tue Apr 22 16:31:44 2008 +0930
90622
90623 xnest: re-enable XKB support. #10015
90624
90625 XKB was disabled in 08928afb0500d46b0caa0a1d1244dee2ed80e6a0, with the comment
90626 "Disable XKB, as we can't yet use it". Seems like "yet" is over, running GNOME
90627 and changing XKB settings seems to work in Xnest now.
90628
90629 X.Org Bug 10015 <https://bugs.freedesktop.org/show_bug.cgi?id=10015>
90630
90631commit a8b8700c7345b89953c8b63cb5c347a95e6ab988
90632Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90633Date: Mon Apr 21 23:44:37 2008 -0700
90634
90635 XQuartz: Fixed cmd-tab to bring all windows forward.
90636 (cherry picked from commit e48e2ce931228f4dfa36c39b8ec8c72a58025b1b)
90637
90638commit a368ab757edf36ed7bbda023673d28883ce11231
90639Author: Dave Airlie <airlied@linux.ie>
90640Date: Tue Apr 22 15:13:57 2008 +1000
90641
90642 GLcore: make googleearth not crash the server on sw-rendering.
90643
90644 I don't think this is the 100% correct answer as I get log spam saying
90645 (EE) DoSwapInterval: cx = 0x98b8998, GLX screen = 0x96dd780
90646 (EE) AIGLX: cx->pGlxScreen->swapInterval == NULL
90647
90648 but thats better than X exiting in my book.
90649
90650commit 449723510a1f9d024e23d9eb33795cac27f9443e
90651Author: Egbert Eich <eich@pdx.freedesktop.org>
90652Date: Tue Apr 22 13:30:03 2008 +0930
90653
90654 xkb: use the correct device instead of an uninitialised "dev". #15614
90655
90656 X.Org Bug 15614 <http://bugs.freedesktop.org/show_bug.cgi?id=15614>
90657
90658 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
90659
90660commit ba87c25321c3378fd1ad0c55dcb0af0a6e82a540
90661Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90662Date: Mon Apr 21 20:03:08 2008 -0700
90663
90664 Make rootless use dixLookupClient rather than deprecated LookupClient.
90665 (cherry picked from commit 582b5b01f9697b66489ea906a2ecb8bfc5915571)
90666
90667commit 8822110d7d6b684f373fc883aeb7cab9734e9ddb
90668Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90669Date: Mon Apr 21 20:08:05 2008 -0700
90670
90671 Fixed dixLookupClient to work with client=NullClient as it did in the 1.3 branch
90672 (cherry picked from commit e41ccc64702f856e5e09dfa652fe73c14b8a0225)
90673 (cherry picked from commit ce5a5f93990647de85e535734ee6bb430ad591cb)
90674
90675commit 5ffb6a2fe8db5871eaf26b8535af1588c43f33d3
90676Author: Ben Byer <bbyer@apple.com>
90677Date: Mon Apr 21 19:55:54 2008 -0700
90678
90679 Nuke a call to deprecated LookupClient, and hopefully prevent a
90680 null-pointer dereference, too!
90681 (cherry picked from commit 3d28e9f953709914e18807bc74c241333671cb30)
90682
90683commit 590688131d89595bdc78ca562ee88df86c9012a6
90684Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90685Date: Sun Apr 20 22:18:45 2008 -0700
90686
90687 XQuartz: Fixed quit dialog to be more conforming with HIG.
90688 (cherry picked from commit 14c6b837bb03bd0956f90882f550847f13d0ca09)
90689
90690commit d20b3ac22d960fa44632cc4a14be079daa2d5a33
90691Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90692Date: Sun Apr 20 22:18:19 2008 -0700
90693
90694 .gitignore: Added Xquartz stuff
90695 (cherry picked from commit 0a9a3bec2de8d1f442493e13cf9f039902a4928f)
90696
90697commit 570b0dca261920c9b01b4eb11fe9b9987b1c636a
90698Author: Ben Byer <bbyer@apple.com>
90699Date: Mon Apr 21 17:52:10 2008 -0700
90700
90701 Fix for pointer-offset issue when using a multi-display environment on X11.app.
90702 (cherry picked from commit 9a7e14286ced55c5e2a4512e2629e03836443009)
90703
90704commit 3f081b4de55e1378728a24d069bf06575ffca2d8
90705Author: Tilman Sauerbeck <tilman@code-monkey.de>
90706Date: Mon Apr 21 11:22:07 2008 +0200
90707
90708 EXA: Set pixmap->accel_blocked on the screen pixmap, too.
90709
90710commit 26c1801a27b81fdd988d5bd210ba0e76ecc274ae
90711Author: Tilman Sauerbeck <tilman@code-monkey.de>
90712Date: Mon Apr 21 11:03:27 2008 +0200
90713
90714 EXA: Update sys_pitch/fb_pitch in exaModifyPixmapHeader.
90715
90716 exaModifyPixmapHeader now also only evaluates arguments that have a
90717 meaningful value.
90718
90719commit 4fa89fbe18c929e0d36305ab47e7e17841309ffd
90720Author: Jordan Crouse <jordan.crouse@amd.com>
90721Date: Mon Apr 21 12:09:00 2008 -0600
90722
90723 xf86: Change AutoConfig driver for PCI ID 022:2091 to 'geode'
90724
90725commit 6c95fae1e9d6b0eb64bc78eced05a6e9f5faf02e
90726Author: Michel Dänzer <michel@tungstengraphics.com>
90727Date: Mon Apr 21 10:45:11 2008 +0200
90728
90729 EXA: Offscreen memory eviction improvements.
90730
90731 * Make sure available areas are considered to have no eviction cost. This seems
90732 to help for https://bugs.freedesktop.org/show_bug.cgi?id=15513 but I'm afraid
90733 that may just be coincidence.
90734 * Only calculate eviction cost of each area once for each eviction pass.
90735 Safeguard against potential (though unlikely) division by zero.
90736 * Cosmetic enhancements: Name eviction cost related variables 'cost' instead of
90737 'score' to emphasize that smaller values are better, update Doxygen file
90738 comment to the way eviction works now.
90739
90740commit 40c6be1408a1f0b236fdb28af27ae18aea0d578f
90741Author: Julien Cristau <jcristau@debian.org>
90742Date: Sun Apr 20 20:36:44 2008 +0200
90743
90744 Minor xorg.conf manpage fixups
90745
90746 Use __libmansuffix__ instead of __oslibmansuffix__ which isn't getting
90747 replaced, and rewrap some text to get __xservername__ replaced in the
90748 description of Option "Accel" (cpp doesn't like the preceding quote).
90749
90750commit 4bcfed2f9cf5dbf682d3bc98873ba97c4efdff44
90751Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90752Date: Sun Apr 20 08:35:42 2008 -0700
90753
90754 Revert "Removed XWin DDX"
90755
90756 This reverts commit 6550078b0925f754e3eec3bbce94dbfe5de8c419.
90757
90758 Doctor, I'm starting to get a pulse...
90759
90760commit 14396fdebac1868df17559220ed7aaa34c34251e
90761Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90762Date: Sat Apr 19 23:31:05 2008 -0700
90763
90764 XQuartz: Cleanup turning off COMPOSITE
90765 (cherry picked from commit 8f920fca6f9149185649d52569d33bf81b6c6857)
90766
90767commit d13828797fe22856b07e08a55d2b9375902194bf
90768Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90769Date: Fri Apr 18 20:38:31 2008 -0700
90770
90771 XQuartz: Cleaned up some casting to get rid of compiler warnings
90772 (cherry picked from commit 6f1c85b96550adf0bc34efb6ca649b87bcc1b18c)
90773
90774commit 6550078b0925f754e3eec3bbce94dbfe5de8c419
90775Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90776Date: Sat Apr 19 09:29:46 2008 -0700
90777
90778 Removed XWin DDX
90779
90780commit 587c010a1cd733fded4d49dc339df0634bda8be6
90781Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90782Date: Sat Apr 19 09:27:21 2008 -0700
90783
90784 Rootless: Kill off rlAccel
90785
90786commit d3d00d92586c3e1cbc88087c930b65c8b3832fcc
90787Author: Tilman Sauerbeck <tilman@code-monkey.de>
90788Date: Sat Apr 19 12:54:40 2008 +0200
90789
90790 Removed fbpseudocolor.h from sdk_HEADERS.
90791
90792commit 5bdc4198795ffd011bb07cffe3817e4cded87f60
90793Author: Adam Jackson <ajax@redhat.com>
90794Date: Sat Apr 19 04:06:19 2008 -0400
90795
90796 Remove fbpseudocolor
90797
90798 "An experimental pseudocolor emulation layer. Not fully completed,
90799 currently only works for 16bpp." That was almost four years ago.
90800 It still doesn't work, only one driver even attempts to use it, it
90801 contains an ad-hoc implementation of damage, and should really be
90802 done up in Composite now anyway.
90803
90804commit 60ff56050b64183cb6e58f54223c8a3ddc2e704b
90805Author: Daniel Stone <daniel@fooishbar.org>
90806Date: Fri Apr 18 22:17:53 2008 -0700
90807
90808 Revert "Optimize dixLookupPrivate for repeated lookups of the same private."
90809
90810 The patch was wildly unsafe for SIGIO, and made everything full of
90811 crashy crashy fail.
90812
90813 This reverts commit 9b30cc524867a0ad3d0d2227e167f4284830ab4e.
90814
90815commit ed9dabb47c467dbf49836b631d5d6bda4b0d98b0
90816Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90817Date: Fri Apr 18 20:30:43 2008 -0700
90818
90819 Last of the spam... I promise...
90820 (cherry picked from commit 45ebee4f729b148a75e925a4863b4eb850c88f8e)
90821
90822commit 49f2bb4681fdee9e45f952ef0ac9c34a090117de
90823Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90824Date: Fri Apr 18 20:25:38 2008 -0700
90825
90826 XQuartz: More sanitization of the namespace
90827 (cherry picked from commit bc50d41f9d1aec04f0de0478cbd5036f1fe9b81e)
90828
90829commit c2f0d020b5d7950267aa3df391a7a72b9ae5883b
90830Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90831Date: Fri Apr 18 20:10:57 2008 -0700
90832
90833 XQuartz: Removed unneccessary include
90834 (cherry picked from commit 45ff59e69eddfcceafced31cf6e73e381d0f6914)
90835
90836commit 5183fea6d38de4bcf657e9c2a983dfd81a2a223f
90837Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90838Date: Fri Apr 18 20:06:17 2008 -0700
90839
90840 XQuartz: Handled sanitization of namespace better
90841 (cherry picked from commit 8cb23d672177da919257c885804cecd18cf9af88)
90842
90843commit edd3fb784bad893550ee270e0a09f22f99783cf5
90844Author: Ben Byer <bbyer@bbyer.local>
90845Date: Fri Apr 18 17:17:01 2008 -0700
90846
90847 random flailing
90848 (cherry picked from commit 7fb9b2dc615a3bd1a3c087438af7a8b88265cfaa)
90849
90850commit a7503615a6893749d512f75d37646273f31b9dbf
90851Author: Adam Jackson <ajax@redhat.com>
90852Date: Fri Apr 18 19:56:41 2008 -0400
90853
90854 Death to TOG-CUP.
90855
90856 If you still care about 8bpp visuals that much, fix Composite to provide
90857 synthetic visuals.
90858
90859commit 4da9ec16e9725ebb9817b49e33ea1035b6aff09a
90860Author: Adam Jackson <ajax@redhat.com>
90861Date: Fri Apr 18 19:54:09 2008 -0400
90862
90863 Remove appgroup mentions from configure.ac
90864
90865commit 25827fde68d3bb02a2b7e05fae53a1d97edf1f76
90866Author: Eric Anholt <eric@anholt.net>
90867Date: Fri Apr 18 15:32:04 2008 -0700
90868
90869 Nuke the MIT-SUNDRY-NONSTANDARD extension.
90870
90871 This extension provided bug-compatibility with pre-X11R6, but has been
90872 stubbed out in our server since 2006 to return BadRequest when you actually
90873 asked for it.
90874
90875commit 13adef8a17d8815f4db2aaac30ae04438e125343
90876Author: Adam Jackson <ajax@redhat.com>
90877Date: Fri Apr 18 19:01:06 2008 -0400
90878
90879 Finish deleting EVI
90880
90881commit eafaf40fb3368ca7e4cf48336fdb7a6c9f536bfa
90882Author: Adam Jackson <ajax@redhat.com>
90883Date: Fri Apr 18 18:50:05 2008 -0400
90884
90885 Death to APPGROUP.
90886
90887commit f6617b4127125516583f321c961d70f762f728be
90888Author: Adam Jackson <ajax@redhat.com>
90889Date: Fri Apr 18 18:28:01 2008 -0400
90890
90891 Death to Extended Visual Information.
90892
90893commit 3b93631e59ca4d312d318eac4015e0a79ad6351f
90894Author: Peter Hutterer <peter@cs.unisa.edu.au>
90895Date: Fri Apr 18 11:31:54 2008 +0930
90896
90897 dix: remove coreOnly check. Core pointer must generate XI events now.
90898
90899 This flag was only used when an event is generated by Warp[Device]Pointer.
90900 Since the VCP now happily generates core events, this flag is obsolete.
90901
90902commit e0eaf8e5e3fa7a11c087851dff93f50f6907c4a5
90903Author: Peter Hutterer <peter@cs.unisa.edu.au>
90904Date: Fri Apr 18 10:57:47 2008 +0930
90905
90906 Xext: Let XTestFakeInput update the sprite for XI events.
90907
90908 Since XI devices can have their own sprite now, we need to update the sprite
90909 coordinates too when processing an XI event.
90910 Note: This doesn't deal with the device hierarchy correctly yet.
90911
90912commit cb0168b4ac5c59cdce6f0a6d89ddd9f30d93b5f3
90913Author: Peter Hutterer <peter@cs.unisa.edu.au>
90914Date: Fri Apr 18 09:48:06 2008 +0930
90915
90916 Xext: xtest.c: death to tabs+spaces indendation, remove #define XINPUT
90917
90918commit eebdf69e9a52e071e0467a1f99de135df4d1eabc
90919Author: Peter Hutterer <peter@cs.unisa.edu.au>
90920Date: Fri Apr 18 09:22:37 2008 +0930
90921
90922 xkb: mixing up src and dst in a memcpy is suboptimal.
90923
90924commit aa6687322de85a2d1025a3ae851fb290a089b2d3
90925Author: Peter Hutterer <peter@cs.unisa.edu.au>
90926Date: Tue Apr 15 23:03:06 2008 +0930
90927
90928 Xi: when copying button classes, copy xkb_acts as well.
90929
90930 This should be the last piece in the quest for the class copy.
90931
90932commit c14f5dc237a31b13d98ae2d0d6143bd91083cf13
90933Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90934Date: Thu Apr 17 20:21:45 2008 -0700
90935
90936 XQuartz: Forgot to commit xprEvent.[hc] ...
90937 (cherry picked from commit 70e543baf2508d636f01b2b7e8cb05172195b68c)
90938
90939commit 22bb7608a025a4ec0f442637810b20e2cb0b0820
90940Author: Ben Byer <bbyer@apple.com>
90941Date: Thu Apr 17 17:04:08 2008 -0700
90942
90943 Added XKB support for Xquartz
90944 (cherry picked from commit 56dc1215202746590dbe8758411f47e8876e1317)
90945
90946commit 652479dba38470273313dc46f17e3bcb1bc5e383
90947Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90948Date: Thu Apr 17 15:49:13 2008 -0700
90949
90950 XQuartz: Moved some rootless-specific cruft into xpr
90951 (cherry picked from commit 31625cc03b58317120c2ac7877e227e2322e1de8)
90952
90953commit 2a1ba20af98c0e9a6a7f1a50d32058dcc9759c21
90954Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90955Date: Thu Apr 17 15:23:00 2008 -0700
90956
90957 XQuartz: Use a mutex to ensure we only have one thread calling mieqEnqueue at a time.
90958 (cherry picked from commit 7b087c965bce9f440ab5233d6383aa4a7de969b8)
90959
90960commit 55f80d754525398378de1ef28aa562bd29ee750f
90961Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90962Date: Thu Apr 17 14:21:31 2008 -0700
90963
90964 XQuartz: A little more debugging output from threadSafety
90965 (cherry picked from commit f6fbdbf838ab77c3a4635f0b2356b1bbb060ff5b)
90966
90967commit 0d61f6fca1efeb4f68488e323d1c0508b9b7a711
90968Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90969Date: Thu Apr 17 13:17:58 2008 -0700
90970
90971 XQuartz: Fixed some missing prototypes
90972 (cherry picked from commit 95056afc562cfe58b116f5c36e4624018e79ff4a)
90973
90974commit dbd4c031565d269fef90af23386ff045ec78688c
90975Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90976Date: Thu Apr 17 13:12:56 2008 -0700
90977
90978 XQuartz: Added framework for asserting which thread we're in.
90979 (cherry picked from commit 00beb982510e7a82d77e1f1d43e77c84d7bf74c2)
90980
90981commit fa0645b452cbebd1800a63f1c95cb77fef4ab211
90982Author: Ben Byer <bbyer@apple.com>
90983Date: Thu Apr 17 12:27:12 2008 -0700
90984
90985 removed Xquartz debugging code that leaked into master. Our Bad.
90986
90987commit a3d40f0549f6c6f49fffc286bcdaad758fa92367
90988Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90989Date: Thu Apr 17 11:56:48 2008 -0700
90990
90991 XQuartz: Include version info for CrashReporter
90992 (cherry picked from commit b4992755c3e29086c5939683c38fa8fd7d2e6754)
90993
90994commit 6d11712c2a35b243c19eea3b26622d18c2446dbe
90995Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
90996Date: Thu Apr 17 11:06:54 2008 -0700
90997
90998 XQuartz: Use strerror(errno)... cause I like text more than grepping header files
90999 (cherry picked from commit 1b4c37d8f9b517fbec5b94ed4e4a5e86a31472a5)
91000
91001commit dcf4f917cc9488de72711255bbb030d9aa8f8bfb
91002Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91003Date: Thu Apr 17 11:03:31 2008 -0700
91004
91005 merged darwinKeyboard.[ch] into quartzKeyboard
91006 (cherry picked from commit 57bb07320908b74facea0a97822bb19ed6f960a9)
91007
91008commit 582397cd024c68df65ab9dececd6d2c40a5a261b
91009Author: Ben Byer <bbyer@apple.com>
91010Date: Thu Apr 17 02:31:53 2008 -0700
91011
91012 oops, missed a spot
91013 (cherry picked from commit 19872a6aeb8ee9cb0e33e4b4ffd794c9dbefe0cf)
91014
91015commit 0bd1c369cce05d5a4da5e3fd7033aea8c68460ec
91016Author: Ben Byer <bbyer@apple.com>
91017Date: Thu Apr 17 02:30:36 2008 -0700
91018
91019 formatting cleanup
91020 (cherry picked from commit 769acd29348abf9e5b0bebfca6ae695d345f3077)
91021
91022commit 700e14c22616b209867e4ea4d1811e53ca996164
91023Author: Ben Byer <bbyer@apple.com>
91024Date: Thu Apr 17 02:21:33 2008 -0700
91025
91026 delete debugging spew
91027 (cherry picked from commit f04f3af86a91d0cafbc86a0d71aeb0599d685f07)
91028
91029commit a440eebf2541ae0bb06bf65281b5facff2f04e00
91030Author: Ben Byer <bbyer@apple.com>
91031Date: Thu Apr 17 02:21:11 2008 -0700
91032
91033 add support for horizontal scrolling (buttons 6 and 7)
91034 (cherry picked from commit f525a4a432ebd0545ad1dd0a7ad84ad3e47e8b61)
91035
91036commit 612e901ef6aa3edc54b39e55e8040cda0e5ab7b6
91037Author: Ben Byer <bbyer@apple.com>
91038Date: Thu Apr 17 01:32:56 2008 -0700
91039
91040 enable keyboard map debugging -- it's going to x11-debug.txt, anyway ... so no harm
91041 (cherry picked from commit ab662c736e0654e2b4347091f0d9e87f26034216)
91042
91043commit 5bdfbfbedcbd9ff61cbb0b678cbf7ce7889a5826
91044Author: Ben Byer <bbyer@apple.com>
91045Date: Thu Apr 17 01:29:46 2008 -0700
91046
91047 darwinKeyboard: refactor slightly so that we're not cutting and pasting code from dix, kthx
91048 (cherry picked from commit a8a090b853e811b9843a5732572cbbe542224f32)
91049
91050commit 58e42683c9e998f6b8a55d5653b9caec7b6acf96
91051Author: Ben Byer <bbyer@apple.com>
91052Date: Thu Apr 17 00:19:56 2008 -0700
91053
91054 hack to Xquartz to prevent xmodmap from wiping out our valid modmap, per daniels
91055 (cherry picked from commit cab54466a61281cfafc12825017c23d720cd75f4)
91056
91057commit cd3470a0cffbd6b8cec7c44227b33307c9e227ae
91058Author: Ben Byer <bbyer@apple.com>
91059Date: Wed Apr 16 22:48:54 2008 -0700
91060
91061 kludge: miEqEnqueue wants a device, even if we're passing custom messages,
91062 so give it one
91063 (cherry picked from commit a494ff04b2a14470eaf5a23c7cf6dbdea182c6d1)
91064
91065commit 966ae1781f3ca563e15a9a1b8cab6fab94e07fe9
91066Author: Drew Parsons <dparsons@debian.org>
91067Date: Mon Mar 10 22:54:49 2008 +1100
91068
91069 Create dix/libXpdix.la for Xprint-specific build of libdix.la
91070
91071 (cherry picked from commit 4e2c6dbabdbbaaca213fd08edd422de15d0900cc)
91072
91073 required because of commit 7c0709a736c0f3aa011de67dd2c2962585ab146e,
91074 which made requestingClient in dix specific to Xprint only.
91075 Add to XPRINT_LIBS in hw/xprint/Makefile.am in front of
91076 $(XSERVER_LIBS) to override definitions in libdix.la for standard xservers.
91077
91078 Follows 571206832d454771e3c638c7515767958365c19c (providing -DXPRINT
91079 to xprint subdirs).
91080
91081 Note it may be possible to restructure the code so that
91082 requestingClient is stored elsewhere than in dix. See discussions
91083 following http://lists.freedesktop.org/archives/xorg/2008-March/033844.html
91084 If this is done it may be possible to revert this commit (if not 571206...).
91085
91086commit 571206832d454771e3c638c7515767958365c19c
91087Author: Drew Parsons <dparsons@debian.org>
91088Date: Mon Mar 10 13:48:05 2008 +1100
91089
91090 Define XPRINT in XPRINT_CFLAGS (configure.ac)
91091
91092 -DXPRINT had only been set for Xprt in hw/xprint/Makefile.am
91093 After commit 7c0709a736c0f3aa011de67dd2c2962585ab146e it is also
91094 required for ps/PsArea.c and PsFonts.c to ensure ‘requestingClient’ is
91095 defined, so make it a global Xprint definition in configure.ac.
91096 (cherry picked from commit 28a6719fd486d9a9cecad0b057d9ea7c59c66055)
91097
91098commit 9b30cc524867a0ad3d0d2227e167f4284830ab4e
91099Author: Michel Dänzer <michel@tungstengraphics.com>
91100Date: Thu Apr 17 16:10:10 2008 +0200
91101
91102 Optimize dixLookupPrivate for repeated lookups of the same private.
91103
91104 This gives me a 20% speedup for EXA text rendering, though I still seem to burn
91105 quite a lot of cycles in here...
91106
91107commit 886af8f3849a0fcfc6b63a9695107ce26d7a6955
91108Author: Tilman Sauerbeck <tilman@code-monkey.de>
91109Date: Wed Apr 16 16:20:19 2008 +0200
91110
91111 EXA: Avoid some fallbacks in exaCopyNtoN.
91112
91113 In some cases we can still do the copying in hardware even if the
91114 dimensions of the pixmaps are out of range. This is true when the boxes
91115 that we're to copy are all in the card's range.
91116
91117commit dc10f0a0e243b7ba38d02a4e2c43027563aead7c
91118Author: Julien Cristau <jcristau@debian.org>
91119Date: Thu Apr 17 11:13:47 2008 +0200
91120
91121 Fix composite on !darwin
91122
91123 2ffdb0eb641ab6949783b4eb574f77e7486ac929 changes the default value of
91124 COMPOSITE to 'auto', but doesn't set it back to 'yes' as appropriate.
91125
91126commit 8716d081fdf61ddf956c30aff7697c70507911fd
91127Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91128Date: Thu Apr 3 16:29:43 2008 -0700
91129
91130 XQuartz: Don't enable rootless accelerated functionality... crashy...
91131 (cherry picked from commit cdb4c291d8c10c3a9ea59d8e79275a30d2ea82b4)
91132
91133commit 2ffdb0eb641ab6949783b4eb574f77e7486ac929
91134Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91135Date: Wed Apr 16 21:54:00 2008 -0700
91136
91137 XQuartz: Don't use composite.
91138 (cherry picked from commit 6d3d344b5b95b6dc4166556d03cfd8c9576dc3f0)
91139
91140commit 757a1bf3a3d72e17eeb362f825124c4ba40cc080
91141Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91142Date: Wed Apr 16 21:48:52 2008 -0700
91143
91144 Xquartz: Don't need to link against rlAccel since we don't use it
91145 (cherry picked from commit 180ec128adef11a9a90cea1189dc31ac5de8359f)
91146
91147commit ab8c6a3c5acb2a3bf288f1d6339b09a125bbb930
91148Author: Alan Coopersmith <alan.coopersmith@sun.com>
91149Date: Mon Apr 14 19:12:00 2008 -0700
91150
91151 Update ac_define_dir macro in acinclude.m4 to 2008-04-12 version
91152
91153commit b907258ebe62642af088f6e2970a45a68cf4be19
91154Author: Eric Anholt <eric@anholt.net>
91155Date: Wed Apr 16 12:07:51 2008 -0700
91156
91157 Update dolt from upstream, fixing fallback to libtool.
91158
91159commit e1e189f8538f2b77ae0cf0d846d3899061e4c4b7
91160Author: Aaron Plattner <aplattner@nvidia.com>
91161Date: Tue Apr 15 14:49:51 2008 -0700
91162
91163 Include pciaccess in the xorg-server.pc Requires line.
91164
91165 This pulls in the include path for pciaccess.h, which is needed by, among other
91166 things, xf86.h.
91167
91168commit 9e7ced94a5e3a14762fe934aa69d91f0831cf5ca
91169Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91170Date: Tue Apr 15 12:06:07 2008 -0700
91171
91172 XQuartz: Removed a call to RootlessReorderWindow from the Carbon thread
91173 (cherry picked from commit cb27d5ca8230707b276763c0ec20e586203144c9)
91174
91175commit 51c8fd69ec9292f5e18cdc7f60e1716fbd6ae61a
91176Author: Peter Hutterer <peter@cs.unisa.edu.au>
91177Date: Tue Apr 15 15:09:40 2008 +0930
91178
91179 dix: free the unused device classes when closing a device.
91180
91181 This also requires to NULL-ify all pointers while we're actually using them,
91182 otherwise we'd try to free them twice.
91183
91184commit 48d33ab9b672b3b3ca308000cdbd573d1e368ff9
91185Author: Peter Hutterer <peter@cs.unisa.edu.au>
91186Date: Tue Apr 15 14:29:53 2008 +0930
91187
91188 dix: float attached devices _before_ disabling the master.
91189
91190 It also helps if we're actually providing the correct argument to
91191 AttachDevice...
91192
91193commit 4cf9c5909d926ec322ed1c7df47f95bd872bb607
91194Author: Peter Hutterer <peter@cs.unisa.edu.au>
91195Date: Tue Apr 15 13:56:11 2008 +0930
91196
91197 Xi: fix up button count.
91198
91199 Some leftover code from the previously used alloc/free device classes left us
91200 with a incorrect button count. So a button release didn't come through if
91201 a different pointer was moved after the button press.
91202
91203commit 35982bc109d424c464551ab22ec90af69908c884
91204Author: Kristian Høgsberg <krh@redhat.com>
91205Date: Fri Apr 11 11:09:13 2008 -0400
91206
91207 Make DRI2 a serverlayout/serverflags option.
91208
91209 Add xf86DRI2Enabled() to export the value of the setting.
91210
91211commit f133d85778462134f366389bde7673bff7845fa8
91212Author: Tilman Sauerbeck <tilman@code-monkey.de>
91213Date: Mon Apr 14 11:43:51 2008 +0200
91214
91215 EXA: Update pixmaps' accel_blocked field in ModifyPixmapHeader.
91216
91217commit bb8868540f017b121d698da45e552ffb55a57cea
91218Author: Tilman Sauerbeck <tilman@code-monkey.de>
91219Date: Mon Apr 14 09:58:49 2008 +0200
91220
91221 EXA: Teach exaCompositeFallbackPictDesc() about x8r8g8b8.
91222
91223commit 1a9d7205cd5640eb65f019336097d86301942ea7
91224Merge: 90f491c 6866e84
91225Author: Peter Hutterer <peter@cs.unisa.edu.au>
91226Date: Mon Apr 14 16:25:58 2008 +0930
91227
91228 Merge whot@wombat:~/potoroo/xserver into mpx
91229
91230commit 3e12c5bb67f3049156475d5cbf4e899aaded76bb
91231Author: Dave Airlie <airlied@redhat.com>
91232Date: Mon Apr 14 11:45:12 2008 +1000
91233
91234 glx: silly nitpick...
91235
91236 even though i and j are the same, we use i to derefence visuals everywhere else
91237
91238commit 97565c0f394f16d042c614695c8b7b4ac354f2a3
91239Author: Dave Airlie <airlied@redhat.com>
91240Date: Mon Apr 14 11:40:38 2008 +1000
91241
91242 glcore: make visualRating GLX_NONE - note GLX_NONE is not == 0
91243
91244 Finally glxinfo returns the set of 3 visuals and glxgears works again for me on
91245 sw rendering
91246
91247commit c61087c82784633e522bd9392172b43656bdf45e
91248Author: Dave Airlie <airlied@redhat.com>
91249Date: Mon Apr 14 10:47:28 2008 +1000
91250
91251 glcore: zero fbconfigs before filling them in.
91252
91253 I'm not sure this the complete proper solution, perhaps it should explicitly fill
91254 in ever field.
91255
91256 This at least makes glxinfo on glcore return sensible information, it doesn't make
91257 gears work yet though.
91258
91259commit 6866e84e3c607d00d88eab2249c2619d6707c1a4
91260Author: Peter Hutterer <peter@cs.unisa.edu.au>
91261Date: Sun Apr 13 19:57:51 2008 +0930
91262
91263 Xi: store feedback classes in devProviates system as well.
91264
91265 This is a follow-up to cb48d880856fd196ab8e8de5eb1f14944a1b4fff.
91266
91267commit cb48d880856fd196ab8e8de5eb1f14944a1b4fff
91268Author: Peter Hutterer <peter@cs.unisa.edu.au>
91269Date: Sun Apr 13 19:48:28 2008 +0930
91270
91271 Xi: store unused classes in devPrivates.
91272
91273 Rather than freeing/allocing classes each time the device capabilities need to
91274 swap, store them in the devPrivates system.
91275 When a class is unused, it is pushed into the devPrivates, and later recovered
91276 when needed again. This saves us a lot of memory allocations/frees, admittedly
91277 on the cost of some memory.
91278
91279commit fde3c836628b6cdec3e5d107d6b1b99bc8b86912
91280Author: Peter Hutterer <peter@cs.unisa.edu.au>
91281Date: Sun Apr 13 17:08:51 2008 +0930
91282
91283 Xi: copy the KeySyms.map over from the source.
91284
91285commit 3c4c9938f31755c5a59995fdcfa138c99db76bbf
91286Author: Peter Hutterer <peter@cs.unisa.edu.au>
91287Date: Sun Apr 13 16:52:14 2008 +0930
91288
91289 Xi: Fix pointer handling in KeyClassRec copy.
91290
91291 We don't free the class anymore, so just store the previous pointers, do the
91292 memcpy from the SD and then restore the pointers.
91293 Plugs a memleak too, before xkbInfo was never freed.
91294
91295commit 755f9e5d7898056cf3bead69ce25a10e23995582
91296Author: Peter Hutterer <peter@cs.unisa.edu.au>
91297Date: Sun Apr 13 16:49:25 2008 +0930
91298
91299 dix: Ignore focus for passive grabs if the event is a pointer event.
91300
91301commit 6faf5b97b92953c331d6540ceb18fd0a77197fea
91302Author: Peter Hutterer <peter@cs.unisa.edu.au>
91303Date: Sun Apr 13 15:42:33 2008 +0930
91304
91305 Xi: fix up modifierKeyMap copying.
91306
91307 Setting it to NULL isn't correct either. The correct behaviour is to realloc
91308 it to the size necessary (or newly alloc it/free it). Otherwise we have a
91309 memleak.
91310
91311commit 3106ba1116e3b9d893f66a93e4a91cc61e23226a
91312Author: Peter Hutterer <peter@cs.unisa.edu.au>
91313Date: Sun Apr 13 11:48:06 2008 +0930
91314
91315 xkb: two fixes to avoid server crashes.
91316
91317 - map can be NULL in some cases, so don't try to dereference it.
91318 - don't default to inputInfo.keyboard
91319
91320 This is firefighting, I presume something in the class copy may have gone
91321 wrong to get a NULL map in the first instance?
91322
91323commit 415c6df0da1197d487456b4c48e2e28e7ded8b8e
91324Author: Peter Hutterer <peter@cs.unisa.edu.au>
91325Date: Sun Apr 13 11:46:44 2008 +0930
91326
91327 Xi: copy feedback classes first, in some cases xkb relies on kbdfeed.
91328
91329 XkbInitIndicatorMap (in XkbInitDevice) calls XkbFindSrvLedInfo. This accesses
91330 the devices kbdfeed struct, which is all nice and dandy if it is NULL. When
91331 copying the device classes however, kbdfeed may not be NULL and thus
91332 XkbFindSrvLedInfo goes on its merry way to do whatever it does.
91333
91334 By copying kbdfeed first, we avoid XkbFSLI to reference the "old" kbdfeed
91335 struct of the previous SD.
91336
91337commit 961f6660902163e99727c2dcc1a039f32b083859
91338Author: Peter Hutterer <peter@cs.unisa.edu.au>
91339Date: Sun Apr 13 09:31:16 2008 +0930
91340
91341 Xi: modifierKeyMap needs to be set to NULL when copying classes.
91342
91343 Otherwise we have a double reference to the same memory area.
91344
91345commit bf6679cba40a936d46008c886d204ed521a4971a
91346Author: Peter Hutterer <peter@cs.unisa.edu.au>
91347Date: Sun Apr 13 08:28:07 2008 +0930
91348
91349 Xi: copy the XkbSrvLedInfo too when copying device classes.
91350
91351commit 4219e94c2f7d431be433eceddfe79760a1ee31a1
91352Author: Peter Hutterer <peter@cs.unisa.edu.au>
91353Date: Sun Apr 13 08:27:31 2008 +0930
91354
91355 xkb: Add XkbCopySrvLedInfo, deep-copies a XkbSrvLedInfoRec.
91356
91357commit 1fa4de80fcfc697b5e5879cc351fb3e9dbf6acbe
91358Author: Matt Turner <mattst88@gmail.com>
91359Date: Sat Apr 12 20:39:18 2008 +0200
91360
91361 Check for __x86_64__ when we check for __amd64__
91362
91363 It seems Intel C Compiler neglects to define __amd64__, __amd64, or
91364 amd64, but *does* define __x86_64__.
91365
91366commit b1f3f42840ec01db417345a0740b59ad5e4471cb
91367Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91368Date: Fri Apr 11 17:49:51 2008 -0700
91369
91370 Xquartz: Added applicationShouldHandleReopen:hasVisibleWindows to handle dock icon clicking
91371 (cherry picked from commit 55d9973b053f25bb95b26e00351dc5531caf5b04)
91372
91373commit eabcfce0a68d504d11be9479f09e66f574dd2f21
91374Author: Adam Jackson <ajax@redhat.com>
91375Date: Fri Apr 11 09:51:26 2008 -0400
91376
91377 Stop building mfb/afb/xf1bpp by default.
91378
91379commit 0dab6fa3582b70ccd0f01459902415c28dbc81ff
91380Author: Adam Jackson <ajax@redhat.com>
91381Date: Fri Apr 11 09:47:51 2008 -0400
91382
91383 So long, and thanks for all the cfb.
91384
91385commit 059b4876e6350aa1110648788cdfbb3f45b4d66d
91386Author: Adam Jackson <ajax@redhat.com>
91387Date: Fri Apr 11 09:46:48 2008 -0400
91388
91389 Add doltcompile to .gitignore
91390
91391commit 6d22a9615a0e6ab3d00b0bcb22ff001b6ece02ae
91392Author: Peter Hutterer <peter@cs.unisa.edu.au>
91393Date: Fri Feb 22 11:01:51 2008 +1030
91394
91395 dix: Call DeleteInputDeviceRequest from CloseDownDevices (#14418)
91396
91397 The DDX (xfree86 anyway) maintains its own device list in addition to the one
91398 in the DIX. CloseDevice will only remove it from the DIX, not the DDX. If the
91399 server then restarts (last client disconnects), the DDX devices are still
91400 there, will be re-initialised, then the hal devices come in and are added too.
91401 This repeats until we run out of device ids.
91402
91403 This also requires us to strdup() the default pointer/keyboard in
91404 checkCoreInputDevices.
91405
91406 X.Org Bug 14418 <http://bugs.freedesktop.org/show_bug.cgi?id=14418>
91407
91408commit 90f491cf8eb869f27c4278b26c1bb84432b12d63
91409Merge: cbe01b3 b4380d8
91410Author: Peter Hutterer <peter@cs.unisa.edu.au>
91411Date: Fri Apr 11 08:29:52 2008 +0930
91412
91413 Merge whot@wombat:~/potoroo/xserver into mpx
91414
91415commit 3c337e18b933881e22b0d03312511f1d23a8640b
91416Author: Tilman Sauerbeck <tilman@code-monkey.de>
91417Date: Thu Apr 10 21:36:19 2008 +0200
91418
91419 Fixed configure.ac for autoconf 2.62.
91420
91421commit 13dcde6bf994fae09c67c3edce9de42df61ef043
91422Author: Eric Anholt <eric@anholt.net>
91423Date: Thu Apr 10 11:08:49 2008 -0700
91424
91425 Fix dolt to error out on compile error, and not print errors on race to mkdir.
91426
91427 Both of these changes have been submitted upstream.
91428
91429commit a7e3ad1c6b455bda7c4abb352a20845d1d4574a0
91430Author: Eric Anholt <eric@anholt.net>
91431Date: Thu Apr 10 10:33:11 2008 -0700
91432
91433 Remove NDEBUG (assert() disable) define from the X Server.
91434
91435 A few pieces of code were abusing this define for other purposes, which are
91436 converted to #ifndef DEBUG instead. There should be no ABI consequences
91437 to this change.
91438
91439 The rationale is that having the define in xorg-server.h also disables
91440 assert() drivers, which is unexpected, and also difficult to avoid since
91441 xorg-server.h is included in their config.h, and you can't put a #undef in
91442 config.h. As for removing it from the server instead of moving it to an
91443 internal header, we probably shouldn't have unnecessary assert()s in
91444 critical server paths anyway, and if we do we could #define NDEBUG in the
91445 specific cases needed.
91446
91447commit b4380d8030927c940ddaea83c4cf24e0b9eb7b96
91448Author: Peter Hutterer <peter@cs.unisa.edu.au>
91449Date: Thu Apr 10 19:25:43 2008 +0930
91450
91451 dix: don't free MDs classes on init.
91452
91453 The device classes aren't deleted anymore on a class change, so there's no
91454 need to store the MD's original classes. We should however restore the MD to
91455 sane defaults when disconnecting the last device, consider this as TODO item.
91456
91457commit 04dff74ffdf727015e3721aae4ea13acc498cd1c
91458Author: Peter Hutterer <peter@cs.unisa.edu.au>
91459Date: Thu Apr 10 19:22:59 2008 +0930
91460
91461 dix: Rework Enter/Leave semaphore system.
91462
91463 Instead of a simple counter, use bits to keep track of which device is where
91464 etc. When device enters a window (or sets focus), the bit matching the device
91465 is set, when it leaves again, it is unset. If there are 0 bits set, then
91466 Leave/Enter/Focus events may be sent to the client.
91467
91468 Same theory as before, but this should get around the insanity with
91469 Grab/Ungrab special cases. Those cases are basically untested though.
91470
91471commit a88386ee277d136caaaeec305f8753f23f9b6274
91472Author: Peter Hutterer <peter@cs.unisa.edu.au>
91473Date: Thu Apr 10 14:36:57 2008 +0930
91474
91475 Xi: only DeliverFocusedEvents if the event is not a pointer event.
91476
91477 A pointer device may have a focus class, but even if so, pointer events must
91478 be delivered to the sprite window, not the focus window.
91479
91480commit 48249425275cc90242497aee9968e5f1ffc86698
91481Author: Peter Hutterer <peter@cs.unisa.edu.au>
91482Date: Thu Apr 10 14:36:10 2008 +0930
91483
91484 Xi: dont copy FocusClassRec if the master already has one.
91485
91486 Blindly copying will override the focus setting of the master. If there's XI
91487 applications running, they may set the SD focus, while leaving the
91488 MD's focus as it was. In this case, after a class swap we still want to get
91489 the MD's events to the same window as before.
91490
91491commit bce6091c6b04ff2db704ae4f161179d21dcbec59
91492Author: Peter Hutterer <peter@cs.unisa.edu.au>
91493Date: Thu Apr 10 09:59:45 2008 +0930
91494
91495 dix: Extend IsKeyboardDevice() to not include pointer devices.
91496
91497 If a pointer devices has key classes as well, don't register it as a keyboard
91498 device. Let's see how much that change will break.
91499
91500commit cc7dab2d04da4ca164eeec1a3296df1706585466
91501Author: Peter Hutterer <peter@cs.unisa.edu.au>
91502Date: Thu Apr 10 09:58:50 2008 +0930
91503
91504 dix: Dont deliver grabbed pointer events to a focus window.
91505
91506 If an pointer event is being processed during a device grab, don't deliver it
91507 to the focus window, even if the device has a focus class. Reason being that
91508 some pointers may have a focus class, thus killing drag-and-drop.
91509
91510commit df2545b98d888924209cb889a68737c15f1aa209
91511Author: Peter Hutterer <peter@cs.unisa.edu.au>
91512Date: Thu Apr 10 08:50:43 2008 +0930
91513
91514 xfree86: Sanity check before retrieving the paired device.
91515
91516 Some pointer devices send key events [1], blindly getting the paired device
91517 crashes the server. So let's check if the device is a pointer before we try to
91518 get the paired device.
91519
91520 [1] The MS Wireless Optical Desktop 2000's multimedia keys are sent through
91521 the pointer device, not through the keyboard device.
91522
91523commit 5a4c6621aaf4e886f2c3b633e837ba359fedf921
91524Author: Peter Hutterer <peter@cs.unisa.edu.au>
91525Date: Thu Apr 10 08:29:05 2008 +0930
91526
91527 Xi: some extra checks for validity of kbd and mouse.
91528
91529 Floating SDs are paired with themselves, so the paired device may not be a
91530 proper keyboard or mouse. Put some extra checks in to avoid dereferencing a
91531 nullpointer later.
91532
91533commit 8e0a6529303a52acc10905dd47c72a0d60979676
91534Author: Peter Hutterer <peter@cs.unisa.edu.au>
91535Date: Thu Apr 10 08:25:36 2008 +0930
91536
91537 dix: When floating, set sprite to NULL before calling InitializeSprite.
91538
91539 InitializeSprite won't create a new one if it already exists, with the result
91540 of overwriting the master's sprite. This master sprite is then assigned to the
91541 floating slave, and freed when the slave is reattached later.
91542 Setting the sprite to NULL forces InitializeSprite to alloc a new one, and
91543 this one can be freed without further repercussions.
91544
91545commit e7211eb0b3d10323dab681bcb18580405ea18ab2
91546Author: Peter Hutterer <peter@cs.unisa.edu.au>
91547Date: Thu Apr 10 08:08:54 2008 +0930
91548
91549 Xi: When attaching, check for ptr -> ptr and keybd -> keybd.
91550
91551 Some pointer devices have key classes (e.g. MS Optical Desktop 2000). The
91552 previous test was performed as Error if ptr -> keybd or keybd -> ptr. This
91553 doesnt work with such devices. New test is Succeed if ptr->ptr or
91554 keybd->keybd.
91555
91556commit 7909ebe7f163716520f843fae11ac7bdeffcb57c
91557Author: Eric Anholt <eric@anholt.net>
91558Date: Wed Apr 9 10:43:25 2008 -0700
91559
91560 dolt: add FreeBSD support (this and ppc have been submitted upstream).
91561
91562commit b4842d8dc3b1619033c5c123c8adc6e164098dc3
91563Author: Michel Dänzer <michel@tungstengraphics.com>
91564Date: Wed Apr 9 16:17:35 2008 +0200
91565
91566 dolt works on powerpc Linux.
91567
91568commit 0d1746995d91b55e40f233f0c38b56bafe896d38
91569Author: Michel Dänzer <michel@tungstengraphics.com>
91570Date: Wed Apr 9 13:37:59 2008 +0200
91571
91572 Fix off-by-one error in ProcXResQueryClients().
91573
91574 Fixes memory corruption reported at
91575 http://bugs.freedesktop.org/show_bug.cgi?id=14004 .
91576
91577commit 6d031cbdefd8072b61645955f01b470a3e6858c1
91578Author: Daniel Stone <daniel@fooishbar.org>
91579Date: Wed Apr 9 14:36:26 2008 +0300
91580
91581 configure.ac: Do the dolt
91582
91583 Use dolt instead of libtool whereever practical. See:
91584 http://lists.debian.org/debian-devel/2008/04/msg00286.html
91585
91586commit b19027fbaea4c3a146926e862983e0e3411fff3d
91587Author: Dave Airlie <airlied@linux.ie>
91588Date: Wed Apr 9 14:27:58 2008 +1000
91589
91590 quirk: add quirk for ACR 640x350 default mode is wrong
91591
91592 RH #440186
91593
91594commit 1f8188656a075dc7b1bb27a0795b5bd43610bbc8
91595Author: Ben Byer <bbyer@apple.com>
91596Date: Tue Apr 8 20:37:25 2008 -0700
91597
91598 add missing dix-config include
91599 (cherry picked from commit 126e9bc8c480b403dedc44c1e8c4fe1476340ed9)
91600
91601commit cbe01b3083eb65c9d4259b1071683febebf11600
91602Merge: 5ffbcfe 389dae7
91603Author: Peter Hutterer <peter@cs.unisa.edu.au>
91604Date: Wed Apr 9 10:46:46 2008 +0930
91605
91606 Merge whot@wombat:~/potoroo/xserver into mpx
91607
91608commit 3f51f493b6daf2464e6c2ba5a924219b88a9e57e
91609Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91610Date: Tue Apr 8 17:02:56 2008 -0700
91611
91612 XQuartz: Fix issue where clicking on an X11 window might send that event to an X11 window in another space.
91613 (cherry picked from commit df21312c8b0e9ef0c809bfc57cdf64f27db0d8a7)
91614 (cherry picked from commit 2d4194a8d124e7a9c7cd1b83635ba6957aa4ae1c)
91615
91616commit 389dae73cc0f3693f49807fd2de146c454ba9783
91617Author: Peter Hutterer <peter@cs.unisa.edu.au>
91618Date: Wed Apr 9 08:26:00 2008 +0930
91619
91620 Xi: If device "to" has a class but "from" doesn't, free the class in "to".
91621
91622commit 60c38d248c1a89392c2c6695c3802f4b54e4c00b
91623Author: Peter Hutterer <peter@cs.unisa.edu.au>
91624Date: Wed Apr 9 07:46:53 2008 +0930
91625
91626 Xi: plug memory leak, free previous motion history before allocating new.
91627
91628commit ea05cf0813b2b7c8cd2151cb935820753ae7997a
91629Author: Peter Hutterer <peter@cs.unisa.edu.au>
91630Date: Tue Apr 8 22:09:11 2008 +0930
91631
91632 Xi: check if source has a key class before copying.
91633
91634commit 08073862f8c4e1219b6459708ffd28e2bc35885f
91635Author: Hasso Tepper <hasso@estpak.ee>
91636Date: Tue Apr 8 13:00:38 2008 +0300
91637
91638 configure.ac: Fix monotonic test harder
91639
91640 This was only introduced in a later version of POSIX, so define that
91641 version to get it from more conformant systems.
91642
91643commit 5ffbcfec3d37d3b627a78acfa00dbafc5948df82
91644Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
91645Date: Tue Apr 8 14:42:00 2008 +0930
91646
91647 dix: Ensure Proximity events don't kill the server.
91648
91649 Add Prox events to the if-clauses with the other events
91650 that are usually sent from the input devices.
91651 Ensure that the event deliverers won't try to deliver
91652 events of type '0' (some extended events doesn't have
91653 an equivalent core-type)
91654
91655 Small modification by Peter Hutterer.
91656
91657 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
91658
91659commit 32e4a88ae613c7200d84d5621344b418b656346b
91660Author: Peter Hutterer <whot@potoroo.wearablelab.ml.unisa.edu.au>
91661Date: Tue Apr 8 08:42:58 2008 +0930
91662
91663 xkb: don't overwrite CtrlProc in the second run of XkbFinishDeviceInit.
91664
91665 XkbFinishDeviceInit is called once when the device is initialised, but also
91666 when a class copy causes the key class of a device to change. In this case,
91667 overwriting the CtrlProc of the KeybdFeedbackClass with XkbDDXKeybdCtrlProc
91668 sets up a nice recursive loop of XkbDDXKeybdCtrlProc calling itself until the
91669 cows come home.
91670
91671commit 6271df6953bea462be7e9e01744e5dd46841e867
91672Author: Peter Hutterer <whot@potoroo.wearablelab.ml.unisa.edu.au>
91673Date: Tue Apr 8 08:42:58 2008 +0930
91674
91675 xkb: don't overwrite CtrlProc in the second run of XkbFinishDeviceInit.
91676
91677 XkbFinishDeviceInit is called once when the device is initialised, but also
91678 when a class copy causes the key class of a device to change. In this case, overwriting the CtrlProc of the KeybdFeedbackClass with XkbDDXKeybdCtrlProc sets up a nice recursive loop of XkbDDXKeybdCtrlProc calling itself until the cows come home.
91679
91680commit 726dcd9e4ebfb09c0685450dca6e9fae7e773814
91681Author: Adam Tkac <atkac@redhat.com>
91682Date: Mon Apr 7 10:20:02 2008 -0400
91683
91684 Fix Xvfb input when building against current X sources.
91685
91686commit 0f87b41a432a6472a15ec0c9dee997e3bddbd0f2
91687Author: Hasso Tepper <hasso@estpak.ee>
91688Date: Mon Apr 7 14:09:04 2008 +0300
91689
91690 configure.ac: DragonFly BSD support
91691
91692 Add support for DragonFly BSD, which is just the same as FreeBSD for all
91693 of these cases.
91694
91695commit 6b1a27023e48b661c4bb3b61181ac57608d8e448
91696Author: Owen W. Taylor <otaylor@fishsoup.net>
91697Date: Thu Apr 3 14:50:05 2008 -0400
91698
91699 EXA: Fix TS origin computation when implementing RenderComposite with tiling.
91700
91701commit fd06e8f8c1d82a9d91931e8532bee0fd9c9ca9ab
91702Merge: b46a009 6c0cfe3
91703Author: Peter Hutterer <whot@potoroo.wearablelab.ml.unisa.edu.au>
91704Date: Mon Apr 7 07:56:41 2008 +0930
91705
91706 Merge branch 'master' into dcdc_rework
91707
91708 Conflicts:
91709
91710 Xext/xevie.c
91711 dix/dispatch.c
91712
91713commit 539bf3c2836727e7560c64144071b086f8ea32fe
91714Merge: 3ab33e7 6c0cfe3
91715Author: Sascha Hlusiak <saschahlusiak@arcor.de>
91716Date: Mon Apr 7 00:10:16 2008 +0200
91717
91718 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
91719
91720commit 3ab33e7cd46c25dfc461b2a1b13e138225a94524
91721Author: Sascha Hlusiak <saschahlusiak@arcor.de>
91722Date: Sun Apr 6 23:23:47 2008 +0200
91723
91724 Remove stale code
91725
91726 The jstk code for Joysticks is not used by any module, was never actually compiled and uses an API
91727 that is deprecated these days.
91728
91729 No reason to keep it.
91730
91731commit b46a00918691cbd5ca80b6d3acae7614f93e073b
91732Author: Peter Hutterer <whot@potoroo.wearablelab.ml.unisa.edu.au>
91733Date: Sun Apr 6 09:02:57 2008 +0930
91734
91735 dix: sprite may be NULL, don't dereference it then.
91736
91737 In some rare cases (e.g. when the init fails) a device's sprite is NULL,
91738 dereferencing it to xfree the spriteTrace is a bad idea then.
91739
91740commit 638a50552e3e2190eac9721deb72e7365bdd52e4
91741Author: Peter Hutterer <whot@potoroo.wearablelab.ml.unisa.edu.au>
91742Date: Sun Apr 6 08:36:21 2008 +0930
91743
91744 dix: remove debug error message about XI->core type conversion.
91745
91746commit 8f38feb3e464986dc523dabd3447ba13263a3a4a
91747Author: Peter Hutterer <whot@potoroo.wearablelab.ml.unisa.edu.au>
91748Date: Sat Apr 5 20:37:09 2008 +1030
91749
91750 Xi: add comments for DeepCopyDeviceClasses, ChangeMasterDeviceClasses.
91751
91752commit 6c0cfe3d43b177c4cfaf7e228f32c655f9a98459
91753Author: Julien Cristau <jcristau@debian.org>
91754Date: Fri Apr 4 19:01:40 2008 +0200
91755
91756 Fix the clock_gettime check for glibc-based non-Linux systems
91757
91758 We need to define _POSIX_C_SOURCE on glibc, not just Linux, so add a new
91759 test for the __GLIBC__ macro.
91760
91761commit cc7c045bae01d90d8f1b750080ba48a96e983c68
91762Author: Adam Jackson <ajax@redhat.com>
91763Date: Fri Apr 4 12:58:12 2008 -0400
91764
91765 Fix PCI config space cycles from int10 emulator.
91766
91767 The top bit of 0xCF8 is an enable bit, not part of the domain. Sending
91768 cycles to domain 128 instead of domain 0 is rarely the right thing to do.
91769
91770commit d1de3dda8efe501d4192c8a99c34ab4265316c32
91771Author: Eric Anholt <eric@anholt.net>
91772Date: Mon Mar 17 14:22:39 2008 -0700
91773
91774 Fix clock_gettime presence detect on FreeBSD.
91775
91776 For non-Linux, _POSIX_C_SOURCE and friends restrict symbols defined rather
91777 than enabling defines of symbols. Additionally, CLOCK_MONOTONIC was
91778 apparently added to the standard around 2000 anyway, not 1993.
91779
91780commit ec17900f52bbd25d07566834756e5c7e832e0463
91781Author: Kristian Høgsberg <krh@redhat.com>
91782Date: Fri Apr 4 10:46:45 2008 -0400
91783
91784 Convert __DRIconfigs after we've made sure createNewScreen succeeded.
91785
91786commit 16a8ce75585ea360c39e0ffce4f7bb26a359b754
91787Author: Kristian Høgsberg <krh@redhat.com>
91788Date: Thu Apr 3 16:44:32 2008 -0400
91789
91790 Only autoload RECORD if it was enabled.
91791
91792commit d0395a753079f291a78d9ab86810b5f84f237491
91793Author: Peter Hutterer <peter@cs.unisa.edu.au>
91794Date: Fri Apr 4 15:05:26 2008 +1030
91795
91796 Xi: realloc to->valuator instead of allocing it blindly.
91797
91798commit 502689847b86be5619da7134646d55a1ac322a2c
91799Author: Peter Hutterer <peter@cs.unisa.edu.au>
91800Date: Fri Apr 4 15:01:53 2008 +1030
91801
91802 Xi: ALLOC_COPY_CLASS_IF should only alloc if to->field doesn't exist.
91803
91804commit 035b1b6995e670ce5593e8aceb08f9ec812c70ea
91805Author: Peter Hutterer <peter@cs.unisa.edu.au>
91806Date: Fri Apr 4 15:00:06 2008 +1030
91807
91808 Xi: remove duplicate copying of kbdfeed and ptrfeed.
91809
91810commit 52fc7aed18930fc9e6bcfdd5dbc28ae559978288
91811Author: Peter Hutterer <peter@cs.unisa.edu.au>
91812Date: Fri Apr 4 14:47:23 2008 +1030
91813
91814 Xi: only alloc memory for feedback classes if it doesn't already exist.
91815
91816commit 4eb87c8693b57d9354832c76417797394656333e
91817Author: Peter Hutterer <peter@cs.unisa.edu.au>
91818Date: Fri Apr 4 14:20:18 2008 +1030
91819
91820 Xi: don't free device classes before copying them.
91821
91822 First commit in a series to come.
91823
91824commit f0915fb3c4a9712200882440a64d11dc595a02bb
91825Author: Dave Airlie <airlied@redhat.com>
91826Date: Fri Apr 4 09:29:51 2008 +1000
91827
91828 quirk: add quirk for ACER EDID
91829
91830commit 2e42b67b82db0f9128dd00e339b9dfdd9fe6d667
91831Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91832Date: Wed Apr 2 18:05:34 2008 -0700
91833
91834 XQuartz: Change reporting of space change to debug log rather than stderr
91835 (cherry picked from commit ed15556a9fc4ebdb88f42961052fc8456082165f)
91836
91837commit c737d04c758e03e32f692a31ed2a665ccbafa931
91838Author: Ben Byer <bbyer@apple.com>
91839Date: Tue Apr 1 00:40:46 2008 -0700
91840
91841 The AppKit thread should not be calling directly into the X server
91842 functions to change state when the keyboard is reloaded; instead,
91843 pass it as an event.
91844 (cherry picked from commit 7e653f806ff5508aace059312156f319a9ed4479)
91845
91846commit 15b0084f1ab23042190d8beeb3f088b92dee5a10
91847Author: Ben Byer <bbyer@apple.com>
91848Date: Mon Mar 31 23:31:25 2008 -0700
91849
91850 formatting cleanup for X11Application.m (no code changes)
91851 (cherry picked from commit eb083d3f68f459d90417558da1ed00729b749950)
91852
91853commit e9e2d88436597875f102085d216dc0a8fce1450a
91854Author: Ben Byer <bbyer@apple.com>
91855Date: Mon Mar 31 22:55:24 2008 -0700
91856
91857 moved and renamed QuartzMessageServerThread to
91858 DarwinSendDDXEvent to make more clear what it actually does.
91859 (cherry picked from commit bee2b377efc930e25017636e5112093a3a6549c7)
91860
91861commit c1be4e3379d8780dff20390939b657ca0973995a
91862Author: Ben Byer <bbyer@apple.com>
91863Date: Mon Mar 31 21:04:37 2008 -0700
91864
91865 shovelling code around ...
91866 (cherry picked from commit 2143182ba49195bbb2e9163ea6872fd68e7a4a85)
91867
91868commit 985c631b2e1f113039e6e620f030505435fd9815
91869Author: Ben Byer <bbyer@apple.com>
91870Date: Mon Mar 31 20:18:58 2008 -0700
91871
91872 just a bit of juggling headers around -- we're preparing
91873 to call our Xquartz-specific event handlers directly
91874 as mieqHandlers
91875 (cherry picked from commit 4aedba5aa727e22316e8ca456f7218bea9ee0313)
91876
91877commit 89f1d880e83e32b72d35c4dbd6795defa6efa847
91878Author: Ben Byer <bbyer@apple.com>
91879Date: Mon Mar 31 19:47:28 2008 -0700
91880
91881 nuke DarwinEventQueue
91882 (cherry picked from commit 1e0ec02202eeaffae480048b91bf02140ee29f8a)
91883
91884commit 8944b77ec0c18476a25ba3179bcc45b338be22b8
91885Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91886Date: Wed Apr 2 17:47:42 2008 -0700
91887
91888 continue with gutting darwinEvents.c
91889 (cherry picked from commit c34fce7051b996633291dddc061b696ff737f3fb)
91890
91891commit aa6d12e93e8661da841192ef7c3aa7c6a7731c7f
91892Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91893Date: Wed Apr 2 17:46:59 2008 -0700
91894
91895 turns out we weren't actually using these files. oops
91896 (cherry picked from commit bfec44d7b4baf0ad0aae55c8209bc60ac93c5b58)
91897
91898commit 6c5962e44730395f81cdb333322c9ad5242c32d4
91899Author: Ben Byer <bbyer@apple.com>
91900Date: Mon Mar 31 18:15:18 2008 -0700
91901
91902 remove vestigal DarwinEQPointerPost etc
91903 (cherry picked from commit a25704c423598d596fd7f2ed4290d4b860bd5d5f)
91904
91905commit c6f0d5d1e51326e5110d27918d834eb0096df7db
91906Author: Ben Byer <bbyer@apple.com>
91907Date: Mon Mar 31 17:48:09 2008 -0700
91908
91909 gut darwinEQEnqueue, and make it just call mieqEnqueue (for the moment)
91910 (cherry picked from commit a9e081a60ca227c0d96d4613075d97d6b762366a)
91911
91912commit 5b6c273eaa53d7b554d69c2b4865988068e73a26
91913Author: Ben Byer <bbyer@apple.com>
91914Date: Mon Mar 31 17:08:45 2008 -0700
91915
91916 add prototype for DarwinEventHandler
91917 (cherry picked from commit 9a03ae33c4f9de830f15eabf3b994882ead7c000)
91918
91919commit 3713be8b470b1ac0fcb4f1e4c6b79c526b2196db
91920Author: Ben Byer <bbyer@apple.com>
91921Date: Mon Mar 31 16:30:16 2008 -0700
91922
91923 add logging of current thread ID to DEBUG_LOG macro
91924 (cherry picked from commit 5848510cc5a8091b30230ab920d904ca6b159480)
91925
91926commit 1400a51ae70d8e498d9ae3975f58ba7c1768ca6f
91927Author: Ben Byer <bbyer@apple.com>
91928Date: Mon Mar 31 16:24:01 2008 -0700
91929
91930 Begin to move all of our Xquartz DDX-specific event handlers
91931 to miEQ, in preparation to remove the DDX-specific code entirely.
91932 (cherry picked from commit 3f4447b95f73a82b3aa0f7b0d1640aba5fb0d1bc)
91933
91934commit 8746daa6732d9837f66d925f2fd74818ecbf8ba2
91935Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
91936Date: Wed Apr 2 15:01:33 2008 -0700
91937
91938 XQuartz: Fixed missing close-paren in preference pane text.
91939 (cherry picked from commit ea37e151dc6032d2a1a33cef809f2a7d507aae35)
91940
91941commit b5a0a865c3045cc08c33388320d4ec3ab7065efb
91942Author: Kristian Høgsberg <krh@redhat.com>
91943Date: Wed Apr 2 19:21:41 2008 -0400
91944
91945 Pick up dri2proto from the standard proto header include path.
91946
91947commit 8cde0af3c57f0375ba8ba77af9fdf74b79d9496d
91948Author: Kristian Høgsberg <krh@redhat.com>
91949Date: Wed Apr 2 19:06:40 2008 -0400
91950
91951 Send the GLX_EXT_texture_from_pixmap attributes to the client.
91952
91953commit 7c20f65fea3dd3170cde89d7113d85f377671bfb
91954Author: Kristian Høgsberg <krh@redhat.com>
91955Date: Wed Apr 2 18:00:06 2008 -0400
91956
91957 Add @XORG_CFLAGS@ to satisfy xf86* includes.
91958
91959 Pointed out by Hasso Tepper.
91960
91961commit b13ab156894074fb38cc812738bc7aeeebd9614d
91962Author: Kristian Høgsberg <krh@redhat.com>
91963Date: Wed Apr 2 12:38:36 2008 -0400
91964
91965 dri2: Unbreak glcore visual setup.
91966
91967commit b31de6a59044f91f8230aa581c9ca8540289c168
91968Author: Xiang, Haihao <haihao.xiang@intel.com>
91969Date: Wed Apr 2 16:29:30 2008 +1000
91970
91971 dri2: fix crasher if DRI2Connect fails
91972
91973commit ebc56aca8bdfec1918cac3c8380895dfddea48ce
91974Author: Hong Liu <hong.liu@intel.com>
91975Date: Wed Apr 2 10:43:19 2008 +0800
91976
91977 Bug #15160: quirk Proview AY765C
91978
91979 prefer first detailed timing
91980
91981commit 9500033b9ecdfaf5a56a4355ffc94d74cb17ca17
91982Author: Goneri Le Bouder <goneri@rulezlan.org>
91983Date: Tue Apr 1 20:19:40 2008 +0200
91984
91985 xfree86: don't crash in AutoConfig if the primary device is not pci
91986
91987 Only call matchDriverFromFiles() if we found a pci device.
91988 Debian bug#472823 (http://bugs.debian.org/472823).
91989
91990commit 37b1258f0a288a79ce6a3eef3559e17a67c4dd96
91991Author: Thomas Jaeger <thjaeger@gmail.com>
91992Date: Tue Apr 1 15:27:06 2008 +0300
91993
91994 XKB: Fix processInputProc wrapping
91995
91996 If input processing is frozen, only wrap realInputProc: don't smash
91997 processInputProc as well. When input processing is thawed, pIP will be
91998 rewrapped correctly.
91999
92000 This supersedes the previous workaround in 50e80c9.
92001
92002 Signed-off-by: Daniel Stone <daniel@fooishbar.org>
92003
92004commit a4d034941100c6ca3b7cc4e59952c2745b9306cc
92005Author: Ben Byer <bbyer@bushing.apple.com>
92006Date: Fri Mar 28 20:47:44 2008 -0700
92007
92008 Add code to track 5 valuators for pointing device, in preparation
92009 for supporting tablet input in Xquartz.
92010 (cherry picked from commit 22c8849ea819eb70a14b2e06330b11b22aa63ebc)
92011
92012commit 6648867d8bd1e86458d2ade77a3ee4567c3d6a97
92013Author: Ben Byer <bbyer@bushing.apple.com>
92014Date: Fri Mar 28 18:27:02 2008 -0700
92015
92016 add debug statements so we can see if/when our Xinput stubs are getting called.
92017 (cherry picked from commit 6e160bbe15dd2c2b8685847c06831cb6aebc6f74)
92018
92019commit 19ff23ab0e72a27d05ed4470f75a0934d6f6c1d1
92020Author: Ben Byer <bbyer@bushing.apple.com>
92021Date: Fri Mar 28 18:25:03 2008 -0700
92022
92023 Remove calls to InitValuatorAxisStruct -- these are now handled in dix by
92024 InitValuatorDeviceClass.
92025 Add InitProximityClassDeviceStruct call to prepare for tablet support.
92026 (cherry picked from commit 1bd980a5b114f5320360943214f8f9f23b29c1e3)
92027
92028commit 9f56fc580646a519875b5a1452738d8c6e1fa860
92029Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92030Date: Mon Mar 31 17:34:07 2008 -0400
92031
92032 XSELinux: Add a request to get a client's context from a resource ID.
92033
92034commit c40e0b51f0d9ef5e1f30f233d7db1e6db9d6681b
92035Author: Kristian Høgsberg <krh@redhat.com>
92036Date: Wed Mar 26 19:28:09 2008 -0400
92037
92038 Implement DRI2 direct rendering and update AIGLX to DRI interface changes.
92039
92040 Get rid of glcontextmodes.[ch] from build, rename __GlcontextModes to
92041 __GLXcontext. Drop all #includes of glcontextmodes.h and glcore.h.
92042 Drop the DRI context modes extension.
92043
92044 Add protocol code to DRI2 module and load DRI2 extension by default.
92045
92046commit 8074676d2df8d577b443e3fa5e22d7c71c944bd1
92047Author: Fredrik Höglund <fredrik@kde.org>
92048Date: Mon Mar 31 21:24:59 2008 +0200
92049
92050 EXA: Optimize the eviction scanning loop in exaOffscreenAlloc.
92051
92052 Reduce the cost of the inner loop, by keeping a set of pointers to the
92053 first and the last areas in the series, subtracting the cost of the first
92054 area from the score, and adding the cost of the last area while walking
92055 the list. This commit also moves the scanning loop from exaOffscreenAlloc
92056 into a separate function.
92057
92058 Idea by Michel Dänzer.
92059
92060commit 93d876891dbba41b920a9a29a5de77f647f43928
92061Author: Fredrik Höglund <fredrik@kde.org>
92062Date: Mon Mar 31 21:15:50 2008 +0200
92063
92064 EXA: Improve the algorithm used for tracking offscreen pixmap use.
92065
92066 Replace the current score keeping algorithm with a rolling counter that's
92067 incremented in ExaOffscreenMarkUsed, with the previous value being stored
92068 in the area. exaOffscreenAlloc uses the difference between the counter
92069 value and the value in the area when deciding which area to evict.
92070 It now also takes the size of the areas into account, and favors evicting
92071 smaller areas.
92072
92073 The credit for these ideas goes to Michel Dänzer.
92074
92075commit 7034484f0887ea0f8ab956515f2d9301ea5842ce
92076Author: Daniel Stone <daniel@fooishbar.org>
92077Date: Fri Mar 28 10:23:36 2008 +0200
92078
92079 gitignore: Add two more bits
92080
92081 Get slightly better at ignoring vim swap files, and let people keep
92082 local changes if they want to.
92083
92084commit b8ea9f2a25aad88aee77a68f8e20ac07276f0dab
92085Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
92086Date: Fri Mar 28 17:00:08 2008 -0700
92087
92088 XQuartz: Fixed names of enums to be more consistent
92089 (cherry picked from commit c309a08806daf5d716a860c709e51eacad2c745a)
92090
92091commit bd28839eb866a9e6dc3ff80f13a67611da7eaf64
92092Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
92093Date: Fri Mar 28 17:02:02 2008 -0700
92094
92095 =?utf-8?q?XQuartz:=20Another=20Expos=C3=A9=20fix:=20F9=20doesn't=20raise=20all=20windows
92096
92097 =20(cherry=20picked=20from=20commit=20b9cffa20debae73737c674bf75ab65db1bd74556)?=
92098 MIME-Version: 1.0
92099 Content-Type: text/plain; charset=utf-8
92100 Content-Transfer-Encoding: 8bit
92101
92102commit 933ffcdf7d2eaaf7caabfe6f861f04bcb149918f
92103Author: Adam Jackson <ajax@redhat.com>
92104Date: Fri Mar 28 17:53:55 2008 -0400
92105
92106 Compile fix.
92107
92108commit 9e8451d869a4032ddb7de6a62920a3a7b0b1acc1
92109Author: Adam Jackson <ajax@redhat.com>
92110Date: Mon Dec 3 17:38:53 2007 -0500
92111
92112 Add E-EDID segment selection.
92113
92114commit 0b4aef4d6df7a5525d381de035fbbf78c5fffeef
92115Author: Adam Jackson <ajax@redhat.com>
92116Date: Mon Dec 3 16:00:00 2007 -0500
92117
92118 Refactor DDC2 code to allow for proper segmented addressing.
92119
92120commit 88ece11d6c45c6f4b94f7fb2da64a46e879d7c27
92121Author: Adam Jackson <ajax@redhat.com>
92122Date: Mon Dec 3 15:47:39 2007 -0500
92123
92124 Start E-EDID support in the DDC module.
92125
92126 Since there's no way to safely know how many blocks xf86DoEDID_DDC2 would
92127 return, add a new xf86DoEEDID entrypoint to do that, and implement the
92128 one in terms of the other.
92129
92130commit b5f98fcea2024c67e598947782913982072cf4fb
92131Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92132Date: Fri Mar 28 14:01:34 2008 -0400
92133
92134 XSELinux: Add xorg.conf option for permissive/enforcing/disabled.
92135 Patch by Joe Nall.
92136
92137 The option goes in the "extmod" subsection.
92138 TODO: Make it easier for extension modules to handle their own options.
92139
92140commit 415e49b940bba2d08870db410ebb47d2add5d836
92141Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
92142Date: Tue Mar 25 23:01:02 2008 -0700
92143
92144 XQuartz: Sanity-check the removal of the 256-color option
92145 Fixes the 'one-time-preferences' bug in 2.2.0_rc1
92146 (cherry picked from commit 38cbd13490fc21724e8eef0ec7e1a20a9cc2e39d)
92147
92148commit c1d37efe9aae5f2895b3437418f7e4bb2eb3400f
92149Author: Ben Byer <bbyer@bushing.apple.com>
92150Date: Tue Mar 25 19:25:00 2008 -0700
92151
92152 xquartz: copy in new stubs from Xi/stubs.c to replace our old
92153 Xinput stubs. Hey, it's a start.
92154 (cherry picked from commit 566412b4aece24ae6af8c7c835986b685aa456a2)
92155
92156commit c4a616a741e15865ce0ff98781c6f1dca4d62887
92157Author: Ben Byer <bbyer@bushing.apple.com>
92158Date: Mon Mar 24 22:43:10 2008 -0700
92159
92160 prevent "fake mouse clicks" from generating spurious extra events
92161 (cherry picked from commit bd85a24969427e41389688663ead2d4dd41c9999)
92162
92163commit c747030a49dd289e873e2b686cd129d840e55468
92164Author: Dave Airlie <airlied@redhat.com>
92165Date: Thu Mar 27 15:18:39 2008 +1000
92166
92167 quirk: fix LPL monitors properly.
92168
92169 no point having a h cm fix when we really want to copy the sizes from the
92170 other place.
92171
92172 RH BZ 435216
92173
92174commit 9df915b84d45d39aed7557c98883b20a66da2e96
92175Merge: 47eb658 333e712
92176Author: Sascha Hlusiak <saschahlusiak@arcor.de>
92177Date: Tue Mar 25 17:50:50 2008 +0100
92178
92179 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
92180
92181commit 333e7123dc484888d79c0f5aa3977bd72f1eb341
92182Author: Adam Jackson <ajax@redhat.com>
92183Date: Tue Mar 25 12:48:22 2008 -0400
92184
92185 Fix that last commit.
92186
92187 I can apply patches, really.
92188
92189commit 47eb658e802775021e3efec109f95431cca188ca
92190Author: Sascha Hlusiak <saschahlusiak@arcor.de>
92191Date: Tue Mar 25 17:37:25 2008 +0100
92192
92193 Support to pass arbitrary options via HAL hotplugging
92194
92195 Parse "input.x11_options" and pass every key/name pair to the driver.
92196 Remove check for input.capabilities, because that's part of the fdi files.
92197
92198 Thanks to Dustin Spicuzza <dustin@virtualroadside.com> for the patch.
92199
92200commit e7a364425547103a98acabfc67d16e1ae0c2967f
92201Author: Sascha Hlusiak <saschahlusiak@arcor.de>
92202Date: Tue Mar 25 12:32:33 2008 -0400
92203
92204 Fix getValuatorEvents to compute number of valuators correctly.
92205
92206commit f028e245a7932362656701c08fcfbfa8e8949077
92207Author: David Nusinow <dnusinow@debian.org>
92208Date: Thu Feb 28 19:45:21 2008 -0500
92209
92210 Bug #10016: Implement WM_CLASS hints in Xephyr.
92211
92212commit 862ff9ac92037e13629329eb6ba50ff6bd2c5f71
92213Author: Adam Jackson <ajax@redhat.com>
92214Date: Mon Mar 24 13:37:42 2008 -0400
92215
92216 Bug #11510: Fix build without RECORD.
92217
92218commit 87bfd3bd96c714a1c252d42408b5a1a4ff9dab06
92219Author: Adam Jackson <ajax@redhat.com>
92220Date: Mon Mar 24 13:33:38 2008 -0400
92221
92222 Bug #11508: Fix build without XV.
92223
92224commit 536f2ff5382aaaace3b55481e15366bb15d87801
92225Author: Adam Jackson <ajax@redhat.com>
92226Date: Mon Mar 24 12:22:19 2008 -0400
92227
92228 Bug #13962: Re-arm the DPMS timer when re-enabling DPMS.
92229
92230commit 4217ba0cf0c9bbea3774760e836ab372acf3237c
92231Author: Julien Cristau <jcristau@debian.org>
92232Date: Sat Mar 22 17:31:08 2008 +0100
92233
92234 xaa: use xf86ReturnOptValBool instead of xf86IsOptionSet
92235
92236 The latter doesn't return the option's value, just whether it's present
92237 in the configuration.
92238
92239commit 6b9d2bb1f7f87acbf275027af9c2982e91e5faed
92240Author: Julien Cristau <jcristau@debian.org>
92241Date: Sat Mar 22 17:28:48 2008 +0100
92242
92243 exa: use xf86ReturnOptValBool instead of xf86IsOptionSet
92244
92245 The latter doesn't give you the option's value, it just tells you if
92246 it's present in the configuration. So using Option "EXANoComposite" "false"
92247 disabled composite acceleration.
92248
92249commit 63859473965f911515bc6e8d87b32a65ec41eb73
92250Author: Jeremy Huddleston <jeremy@tifa.local>
92251Date: Fri Mar 21 19:31:31 2008 -0700
92252
92253 XQuartz: Disable 256 color option and fullscreen option
92254 (cherry picked from commit 7c1964338a33558d3f25e369dfca99e3ef9d10f9)
92255
92256commit 93daa3a3bf1a981757024847882ce92b6bdaae83
92257Author: Jeremy Huddleston <jeremy@tifa.local>
92258Date: Fri Mar 21 19:11:59 2008 -0700
92259
92260 Xquartz: Added separate preference tab for quartz-wm settings
92261 Added FFM and "Focus on new window" options
92262 (cherry picked from commit 6841d078b7cb0b0db3db948d26b4d5ec7747deb8)
92263
92264commit c49e11268322712c211f29d51d664d3f8a59b00b
92265Author: Jeremy Huddleston <jeremy@tifa.local>
92266Date: Fri Mar 21 18:07:38 2008 -0700
92267
92268 XQuartz: Initial framework for dealing with spaces on OS-X
92269 (cherry picked from commit 9831324998f9d1f05ff944c58c5bf60dcae17355)
92270
92271commit 4c76607b699431183ee7e88fa7818cb7644a5a02
92272Author: Jeremy Huddleston <jeremy@tifa.local>
92273Date: Mon Mar 17 23:57:41 2008 -0700
92274
92275 Rootless: Removed safeAlphaXXXX() in favor of using fb/pixman
92276 (cherry picked from commit f03202ad15457c98be7ca78cc59bac88cf5f1966)
92277
92278commit 87c64cfd6901083da5a9375d0bde65691d374b5b
92279Author: Jeremy Huddleston <jeremy@tifa.local>
92280Date: Fri Mar 14 17:31:54 2008 -0700
92281
92282 =?utf-8?q?Apple:=20Xserver=20half=20of=20the=20Expos=C3=A9=20bug-fix=20(requires=20updated=20libXplugin=20from=20Apple...=20coming=20with=202.2)
92283
92284 =20(cherry=20picked=20from=20commit=2037be23e8c1d8e5c7a1157e9d66ef3f30a4c472c5)?=
92285 MIME-Version: 1.0
92286 Content-Type: text/plain; charset=utf-8
92287 Content-Transfer-Encoding: 8bit
92288
92289commit 3bbd77ff98478153afe3251de9ba11d757218213
92290Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92291Date: Thu Mar 20 20:03:02 2008 -0400
92292
92293 XSELinux: Do a check for whether background "None" is allowed.
92294
92295commit e323bb426ce8a072d119cb2720b773241259c137
92296Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92297Date: Thu Mar 20 19:42:09 2008 -0400
92298
92299 XSELinux: Correctly handle some permission bits that are used more than once.
92300
92301commit da973e962d09854b571320dee7dd9569060bc39e
92302Author: Michel Dänzer <michel@tungstengraphics.com>
92303Date: Thu Mar 20 09:18:29 2008 -0400
92304
92305 Fix RandR 1.2 driver interface conversion of two colour cursors to ARGB
92306
92307 This patch (and not setting HARDWARE_CURSOR_BIT_ORDER_MSBFIRST on big endian
92308 platforms) fixes it for me with the radeon driver and doesn't break intel.
92309
92310 Correct patch this time :)
92311
92312commit f8c1eb29e28818895d744c4e1d6897353d51790b
92313Author: Alex Deucher <alex@cube.(none)>
92314Date: Thu Mar 20 09:14:41 2008 -0400
92315
92316 Revert "Fix RandR 1.2 driver interface conversion of two colour cursors to ARGB"
92317
92318 This reverts commit 267352579612155adfd4743432d6569b2cdeebde.
92319
92320 Pushed the wrong patch.
92321
92322commit 267352579612155adfd4743432d6569b2cdeebde
92323Author: Michel Dänzer <michel@tungstengraphics.com>
92324Date: Wed Mar 19 19:12:37 2008 -0400
92325
92326 Fix RandR 1.2 driver interface conversion of two colour cursors to ARGB
92327
92328 See bug 11796
92329
92330commit 060a99444ee25a684b0ab9b4819bf8e855aea2d8
92331Author: Alan Coopersmith <alan.coopersmith@sun.com>
92332Date: Wed Mar 19 16:04:16 2008 -0700
92333
92334 Make Xevie private symbol names less generic
92335
92336 Makes it easier to figure out what you're seeing in the stack trace
92337 instead of wondering where in the server "ProcSend" is.
92338
92339commit f37046984d7839faefa4d716624e4a85ddde9634
92340Author: Alan Coopersmith <alan.coopersmith@sun.com>
92341Date: Wed Mar 19 14:06:53 2008 -0700
92342
92343 Xevie cleanups, byteswapping fixes & request length check fixes
92344
92345commit edad0a9dfebcce5c54b2f9c32bd9d45549e20c51
92346Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92347Date: Tue Mar 18 17:51:21 2008 -0400
92348
92349 Apply __glXDisp_GetVisualConfigs message patch
92350 From http://bugs.freedesktop.org/show_bug.cgi?id=13863
92351
92352 Problem was that the glxcmds.c __glXDisp_GetVisualConfigs
92353 function left garbage in the tail end of the message used for extensions.
92354
92355commit cdadd2ff9bade318caac5c1d9bcdc8a001347da9
92356Author: Dodji Seketeli <dodji@seketeli.org>
92357Date: Tue Mar 18 14:00:15 2008 +0100
92358
92359 [Xephyr/DRI] correctly route motion events targeted at GL drawable
92360
92361commit afd7428690d87097117ab20335658f6d59d60103
92362Author: Jesse Barnes <jbarnes@jbarnes-t61.(none)>
92363Date: Mon Mar 17 14:55:44 2008 -0700
92364
92365 Cleanup logic in xf86PrepareOutputs
92366
92367 Should have done this in the first place. Since we're checking for the absence
92368 of the get_crtc callback in the first place, we'll short circuit the later call
92369 and disable the output, so the ugly "continue" block is unnecesary.
92370
92371commit ba85caacb565b9aa0aeace52a362350304b0566d
92372Author: Jesse Barnes <jbarnes@jbarnes-t61.(none)>
92373Date: Mon Mar 17 14:13:09 2008 -0700
92374
92375 Make xf86SetDesiredModes aware of current output configuration
92376
92377 By adding a new output callback, ->get_crtc, xf86SetDesiredModes is able to
92378 avoid turning off outputs & CRTCs if the current output<->CRTC mappings are the
92379 same as the desired configuration. This helps avoid flickering displays at
92380 startup time, which speeds things up a little and looks better.
92381
92382commit bee2ddf35f75086cee951142098637f2c756b96b
92383Author: Jesse Barnes <jbarnes@jbarnes-t61.(none)>
92384Date: Mon Mar 17 08:33:01 2008 -0700
92385
92386 Fail CRTC configuration if !vtSema
92387
92388 Unless we check for vtSema before calling into the CRTC and output callbacks,
92389 we may end up trying to access video memory that no longer exists, leading to a
92390 crash. So if we don't have vtSema, return FALSE to the caller, indicating that
92391 we didn't do anything.
92392
92393 Fixes #14444.
92394
92395commit db248ffb840a0c113b6eb508a0fa1e74e752474d
92396Author: Matthieu Herrb <matthieu.herrb@laas.fr>
92397Date: Sun Mar 16 18:46:11 2008 +0100
92398
92399 test for the presence of pci_system_init_dev_mem() before calling it.
92400
92401 This avoids creating a dependency on -current libpciaccess for
92402 BSD systems other than OpenBSD (which don't otherwise need it).
92403
92404commit aa231f28d56402d7daea6cbd3002fbf760f79497
92405Author: Donnie Berkholz <dberkholz@gentoo.org>
92406Date: Fri Mar 14 18:41:25 2008 -0700
92407
92408 Xephyr: Build fix: Port across XF86dri.c changes from Mesa.
92409
92410commit a955c3b587b22b8bf20cb6bedbbec4ad5fcb32ac
92411Author: Donnie Berkholz <dberkholz@gentoo.org>
92412Date: Fri Mar 14 18:41:07 2008 -0700
92413
92414 Xephyr: Distribute ephyrdriext.h in tarballs.
92415
92416commit 090b26db767d296e7a3452da83b136d1caa0ed01
92417Author: Daniel Stone <daniel@fooishbar.org>
92418Date: Fri Mar 14 21:58:27 2008 +0200
92419
92420 XkbCopyKeymap: Fix broken indentation
92421
92422 An astute observer will note that the entirety of XkbCopyKeymap is indented
92423 with spaces, and no tabs whatsoever, and not commit changes which break the
92424 otherwise consistent indentation.
92425 A non-astute observer will note the breakage when the commit mail comes
92426 through with clearly broken indentation.
92427 A polite, non-astute, observer will then fix it.
92428
92429 C'est la vie.
92430
92431commit 88bec0915e3867f8dbf859a3dfbb771d0d07092d
92432Author: Daniel Stone <daniel@fooishbar.org>
92433Date: Fri Mar 14 21:54:13 2008 +0200
92434
92435 mi: More meaningful assert crashes
92436
92437 When we fail an assert in miregion.c (which happens every now and then,
92438 though I haven't yet checked up why), at least generate a segfault, so
92439 we'll get a backtrace.
92440
92441commit 57d48d94b8947c571925e6fd4c9bf041fbd1b2ac
92442Author: Adam Jackson <ajax@redhat.com>
92443Date: Fri Mar 14 14:37:42 2008 -0400
92444
92445 Fix a stray use of ALLOCATE_LOCAL.
92446
92447commit 824853772241acf64bc37ac8b85254194741ae13
92448Author: Adam Jackson <ajax@redhat.com>
92449Date: Fri Mar 14 14:24:21 2008 -0400
92450
92451 RANDR 1.2: Fix initial mode aspect ratio match in a corner case.
92452
92453 Actually more like in the mainline case, where the ideal mode happens to
92454 be the very first aspect match on the first monitor. But let's not
92455 split hairs.
92456
92457commit 1b9878ffcfc0c0dbc4a6e674827fe508ba77db4b
92458Author: Bart Trojanowski <bart@jukie.net>
92459Date: Thu Mar 13 17:42:16 2008 -0400
92460
92461 Bug #14332: Fix PCI access cycles from x86emu.
92462
92463 The address written to 0xcf8 contains the PCI slot address to send the
92464 config cycle to. However, we would ignore that and always send the
92465 cycle to the device whose BIOS we were running. This breaks some
92466 integrated graphics platforms that have explicit knowledge about the
92467 system's host bridge, for example.
92468
92469commit f7abe05b3306ed9a6f2cf5e3e45ed524d725d029
92470Author: Doug Chapman <doug.chapman@hp.com>
92471Date: Thu Mar 13 17:40:34 2008 -0400
92472
92473 Bug #14091: Fix build (and runtime) on ia64.
92474
92475commit 5d7437c29e686a081b20823450d78c4c2f4e0aec
92476Author: Adam Jackson <ajax@redhat.com>
92477Date: Thu Mar 13 17:37:12 2008 -0400
92478
92479 RANDR 1.2: Fix the RANDR 1.1 screen size estimation to approach reality.
92480
92481 While the ScreenRec's notion of size in millimeters would get updates,
92482 the RANDR 1.1 notion wouldn't, so your screen would appear to be square
92483 and probably at some ludicrous DPI.
92484
92485commit 61c3f63a75d8b0cc47ffed4a0e30147fab2ae8f4
92486Author: Adam Jackson <ajax@redhat.com>
92487Date: Thu Mar 13 17:34:54 2008 -0400
92488
92489 RANDR 1.2: Don't report a square resolution to RANDR 1.1 clients.
92490
92491 It can't possibly do anything useful, and older versions of Gnome (and
92492 proably others) get very confused by it. So do the drivers, for that
92493 matter.
92494
92495commit 06c0372c3a1b45005eb6d50406f77f4e93f1de1e
92496Author: Mark Kettenis <mark.kettenis@xs4all.nl>
92497Date: Wed Mar 12 21:45:37 2008 +0100
92498
92499 OpenBSD support for libpciaccess.
92500
92501 xserver and libpciaccess both need to open /dev/xf86, which can only
92502 be opened once. I implemented pci_system_init_dev_mem() like Ian
92503 suggested. This requires some minor changes to the BSD-specific
92504 os-support code. Since pci_system_init_dev_mem() is a no-op on
92505 FreeBSD this should be no problem.
92506
92507commit 2036851125226065891f13583ade3ce559e7bd37
92508Author: Matthias Hopf <mhopf@suse.de>
92509Date: Mon Mar 10 19:29:07 2008 +0100
92510
92511 Return randr interface version in xf86CrtcScreenInit()
92512
92513 Necessary to allow drivers to be run-time backwards compatible when using the
92514 modes/ functions w/o providing their own copy.
92515
92516commit c7536f4b87e089a7e7c43026b189922fec70c565
92517Author: Kristian Høgsberg <krh@redhat.com>
92518Date: Tue Mar 11 13:11:04 2008 -0400
92519
92520 Silence REGION_INIT() warning.
92521
92522 Evaluating the address of a BoxRec as a boolean gives this warning:
92523
92524 i830_driver.c:2317: warning: the address of 'ScreenBox' will always
92525 evaluate as 'true'
92526
92527 which is pretty annoying. This patch compares the address to NULL to
92528 avoid the pointer->bool conversion and gets rid of the warning. Seems
92529 like a lame hack, but the warning is worse.
92530
92531commit cc05255191413b3f376edbc600122ff085f45f7b
92532Author: Kristian Høgsberg <krh@redhat.com>
92533Date: Tue Mar 11 00:51:43 2008 -0400
92534
92535 Make WriteToClient take a const void * like any decent IO write function.
92536
92537 Enough with the casting. Doesn't break API or even ABI, but does make
92538 a lot of silly casts superfluos.
92539
92540commit bc504ffbba3dec2e3467bab8ba1ac25db6dd317e
92541Author: Kristian Høgsberg <krh@redhat.com>
92542Date: Tue Mar 11 00:35:31 2008 -0400
92543
92544 DRI2: Add DRI2AuthConnection().
92545
92546 DRI2 uses the same authentication scheme as XF86DRI, so implement this
92547 entry point so DRI2 protocol code can access it.
92548
92549commit b2657ec5981122e7cc0bda0d8802aec63cde9014
92550Author: Keith Packard <keithp@keithp.com>
92551Date: Wed Feb 6 16:51:57 2008 -0800
92552
92553 XkbCopyKeymap was mangling doodads and overlays
92554
92555commit ab9b0b36ac8ac72fc48c0abd91a83de49a18313c
92556Author: Adam Jackson <ajax@redhat.com>
92557Date: Mon Mar 10 13:40:00 2008 -0400
92558
92559 Add the "amd" driver to magic driver selection.
92560
92561commit 0f6aaf636b7ac4c98467284ff7baf1b83e0b72e7
92562Author: Adam Jackson <ajax@redhat.com>
92563Date: Mon Mar 10 09:14:20 2008 -0400
92564
92565 Bug #14927: Fix the math for xf86NumDefaultModes.
92566
92567commit 01c2e01f2aee580438b74bfb9da8f584f3878e6b
92568Author: Kristian Høgsberg <krh@sasori.boston.redhat.com>
92569Date: Sun Mar 9 21:40:27 2008 -0400
92570
92571 GLX: Track changes to DRI_TEX_BUFFER extension.
92572
92573 We now just pass in the __DRIdrawable.
92574
92575commit acedc03367e9e69f03b4838f0f0e8d8a8e872b9b
92576Author: Kristian Høgsberg <krh@sasori.boston.redhat.com>
92577Date: Sun Mar 9 21:39:19 2008 -0400
92578
92579 DRI2: Return event buffer head index in DRI2CreateDrawable.
92580
92581 And pass it to the DRI driver in AIGLX.
92582
92583commit 911f0c147699d3b8d97491be2ef6d2e4a6682a0b
92584Author: Julien Cristau <jcristau@debian.org>
92585Date: Fri Mar 7 23:24:06 2008 +0100
92586
92587 Programs in hw/dmx/examples/ want client-side xkb headers.
92588
92589commit 9abaad115cb6245b12b2adb3552ace99b634ab4a
92590Author: Kristian Høgsberg <krh@redhat.com>
92591Date: Fri Mar 7 14:12:28 2008 -0500
92592
92593 Fix DRI2 texture target for GLX_EXT_texture_from_pixmap.
92594
92595 Thanks to Dennis Kasprzyk for pointing it out and for reminding me to
92596 commit it.
92597
92598commit 3bf7ff70323d533a3a05c0f3e22393e63beada99
92599Author: Adam Jackson <ajax@redhat.com>
92600Date: Fri Mar 7 09:25:06 2008 -0500
92601
92602 Size xf86DefaultModes explicitly.
92603
92604 i.e., don't check for the end of the list by ->name == NULL, since that
92605 won't work now. Fix the consumers of xf86DefaultModes to use the new
92606 explicit size as well.
92607
92608commit 3fcb6445dc08f42488287162e3b7009d405e9c5b
92609Author: Adam Jackson <ajax@redhat.com>
92610Date: Fri Mar 7 08:29:49 2008 -0500
92611
92612 Fix segfault when a monitor exists but has no modes.
92613
92614 Thanks to Zhenyu Wang for finding this.
92615
92616commit a8d760f567b19268329c4682495caa591f08a854
92617Author: Aaron Plattner <aplattner@nvidia.com>
92618Date: Sun Mar 2 20:13:11 2008 -0800
92619
92620 Get modes code building with old X servers again.
92621
92622 This change uses XORG_VERSION_CURRENT < 7.0 to mean "server newer than 1.2"
92623 since XORG_VERSION current went backwards at some point.
92624
92625commit ca616b902b2c5d0f046c7a042c11f045479e373a
92626Author: Jesse Barnes <jbarnes@jbarnes-t61.(none)>
92627Date: Thu Mar 6 13:47:44 2008 -0800
92628
92629 Allow RandR get output property to call into drivers
92630
92631 In order to report accurate values to users of the RandR property interface,
92632 it's sometimes necessary to ask the driver to update the value (for example
92633 when backlight brightness changes without the server's knowledge, due to hotkey
92634 events or direct sysfs banging).
92635
92636 This patch wires up the core server code with a new xf86CrtcFuncs callback,
92637 get_property, to allow for this.
92638
92639 The new code is available under the RANDR_13_INTERFACE define, which in turn
92640 depends on the RANDR_12_INTERFACE code.
92641
92642commit 34b69e3bc0e6462eb60029fbcb4f5479494007a2
92643Author: Adam Jackson <ajax@redhat.com>
92644Date: Wed Mar 5 23:56:49 2008 -0500
92645
92646 Fix distcheck.
92647 (cherry picked from commit 2a47accff8dccded4dfe031f9366c028ba927824)
92648
92649commit 8e562fe26ff77c03be0855afb08e43f3ed0bd4b9
92650Author: Peter Hutterer <peter@cs.unisa.edu.au>
92651Date: Wed Mar 5 15:11:41 2008 +1030
92652
92653 Xi: remove false memset from DeepCopyDeviceClasses.
92654
92655 It's wrong, needs to be different, but I haven't had enough coffee to figure
92656 out what it needs to do yet.
92657
92658commit d4101140f4e569f18554cf0cbf43138d08bd1277
92659Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92660Date: Tue Mar 4 22:39:41 2008 -0500
92661
92662 xselinux: Implement polyinstantiation support and related protocol.
92663
92664commit cf984dcc156958d4f8d98110e7add150628ce97e
92665Author: Alan Coopersmith <alan.coopersmith@sun.com>
92666Date: Tue Mar 4 16:03:04 2008 -0800
92667
92668 Make sure SIOGLIFCONF buffer is properly aligned for socket structures
92669
92670commit 27e7dacbf7ef17712be31ff90f98ee3a5c5cf909
92671Author: Adam Jackson <ajax@redhat.com>
92672Date: Tue Mar 4 11:38:34 2008 -0500
92673
92674 Make xf86InitialConfiguration slightly smarter.
92675
92676 Old heuristic was to find the first monitor that expressed a preference,
92677 then attempt to get all other monitors to agree. This doesn't work
92678 particularly well when the two sets of modes don't precisely intersect,
92679 you get overlapping-but-not-identical output geometry and things go wrong.
92680
92681 New heuristic is:
92682 - Exact user preference, if given
92683 - Exact output preference, if the same for all outputs
92684 - Best (largest) mode of modes common to all outputs:
92685 - with the same aspect ratio as all outputs (may be NULL)
92686 - with 4:3 aspect ratio
92687 - Then the old heuristic to try to get something lit
92688
92689 Note that it is simply not doable to have a reliable initial output guess if
92690 you insist on trying to clone all outputs together. It's far too easy to
92691 end up with displays that simply don't have modes in common. We need to
92692 switch to right-of placement someday, once we're not limited to CRTC size
92693 limits and we have working multi-GPU in RANDR.
92694
92695commit 95df04b744c6a3498a9a9e2ea9bb03ee780e60f8
92696Author: Adam Jackson <ajax@redhat.com>
92697Date: Tue Mar 4 10:59:24 2008 -0500
92698
92699 Remove all mention of the vga driver from the config logic.
92700
92701commit 7b4f3616f75a541b819d99c28bb6cb73761b6066
92702Author: Peter Hutterer <peter@cs.unisa.edu.au>
92703Date: Tue Mar 4 18:29:15 2008 +1030
92704
92705 Xi: add deviceid to QueryDevicePointer request.
92706
92707 Why not, we had the padding anyway.
92708 Suggested by Jim Gettys on the X.Org mailing list.
92709
92710commit 056a2ce02ce85013e89055ee44a7aa3eabedac09
92711Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92712Date: Tue Mar 4 02:44:48 2008 -0500
92713
92714 XACE: Check the return value of the selection create hook call.
92715
92716commit 4f2cd0ed96d3b10c78774c721c2ffbfb0556dddd
92717Merge: 23ae68a 453661a
92718Author: Peter Hutterer <peter@cs.unisa.edu.au>
92719Date: Tue Mar 4 18:11:10 2008 +1030
92720
92721 Merge branch 'master' into mpx
92722
92723 This merge reverts Magnus' device coorindate scaling changes. MPX core event
92724 generation is very different, so we can't scale in GetPointerEvents.
92725
92726 Conflicts:
92727
92728 Xi/opendev.c
92729 dix/devices.c
92730 dix/dixfonts.c
92731 dix/getevents.c
92732 dix/resource.c
92733 dix/window.c
92734 hw/xfree86/common/xf86Xinput.c
92735 mi/mipointer.c
92736 xkb/ddxBeep.c
92737 xkb/ddxCtrls.c
92738 xkb/ddxKeyClick.c
92739 xkb/ddxList.c
92740 xkb/ddxLoad.c
92741 xkb/xkb.c
92742 xkb/xkbAccessX.c
92743 xkb/xkbEvents.c
92744 xkb/xkbInit.c
92745 xkb/xkbPrKeyEv.c
92746 xkb/xkbUtils.c
92747
92748commit 72f2197545e734cd0aa785d05a57b2fc0351a763
92749Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92750Date: Tue Mar 4 02:02:54 2008 -0500
92751
92752 dix: Convert selection list to a linked list.
92753
92754 Fixes a bug where pointers were being invalidated after a realloc.
92755
92756commit 0bd0f90d7c7928052197da7119177e5a1c9eee2c
92757Author: Daniel Stone <daniel@fooishbar.org>
92758Date: Tue Mar 4 03:47:36 2008 +0200
92759
92760 XKB: Fix initial map setting on startup
92761
92762 Due to an unwitting sense inversion when eliminating XkbFileInfo, we were
92763 setting the complete wrong keymap on startup (non-XKB map if we had an XKB
92764 map available, or the XKB map if we didn't have any available). Invert the
92765 sense properly, and add two small bits that also went missing in that commit.
92766
92767commit 613852ce6a821ce6f6382fc14629f517776a3701
92768Author: James Cloos <cloos@jhcloos.com>
92769Date: Mon Mar 3 16:10:04 2008 -0500
92770
92771 Fix some documentation typos
92772
92773commit 3f23139137e024e09d207be05a61968100cf53e8
92774Author: Jesse Barnes <jesse.barnes@intel.com>
92775Date: Mon Mar 3 13:05:12 2008 -0800
92776
92777 Add cscope files to .gitignore
92778
92779commit 708f07753ff22ade54e9ee8885e4198fff363b87
92780Author: Adam Jackson <ajax@redhat.com>
92781Date: Mon Mar 3 15:49:48 2008 -0500
92782
92783 RANDR 1.2: Inherit PreferredMode from the global configuration, if any.
92784
92785 If you don't do this, then Modes "800x600" in the Display subsection will
92786 be dutifully ignored and the driver will start at whatever resolution it
92787 feels like.
92788
92789commit 605e6764dfd3e9cb917b9cfcd92fe89857c1a1c9
92790Author: Adam Jackson <ajax@redhat.com>
92791Date: Mon Mar 3 15:45:17 2008 -0500
92792
92793 Fix Motif menu drawing in Xnest.
92794
92795 See also Red Hat bug #229350, OpenSolaris bug #6366490.
92796
92797commit 3b73d62791d925c465ec855f96981d151dd3c179
92798Author: Adam Jackson <ajax@redhat.com>
92799Date: Mon Mar 3 15:43:22 2008 -0500
92800
92801 xf86DDCMonitorSet: Honor the DisplaySize from the config file.
92802
92803 We honor sync ranges and pixel clock settings from the config here, no
92804 reason to ignore DisplaySize.
92805
92806commit 7c16b68ab879f5b4b1aedfc6b2aadbe56193dd19
92807Author: Adam Jackson <ajax@redhat.com>
92808Date: Mon Mar 3 15:09:11 2008 -0500
92809
92810 1.5 has branched, start 1.5.99.x.
92811
92812commit f7ab2d3821e6bccc943f088e308fd58395a186d2
92813Author: David Nusinow <dnusinow@debian.org>
92814Date: Sun Mar 2 18:36:25 2008 -0500
92815
92816 Add missing file from previous commit.
92817
92818commit b5ce0e1d0b861dc5521fcd9db6287ed6da817726
92819Author: David Nusinow <dnusinow@debian.org>
92820Date: Sun Mar 2 17:12:02 2008 -0500
92821
92822 Bug #13860: Ensure that the DRI mode is in octal format.
92823
92824commit c934366424b0d20e013c84e6b94b226b20e7baa2
92825Author: Matthieu Herrb <matthieu.herrb@laas.fr>
92826Date: Sun Mar 2 19:27:53 2008 +0100
92827
92828 use UTILS_SYS_LIBS to pass SYS_LIBS to utils/ioports correctly
92829
92830commit 2bb9c1f36f685044b837f42076dec2ea7d22d034
92831Author: David Nusinow <dnusinow@debian.org>
92832Date: Sat Mar 1 18:44:58 2008 -0500
92833
92834 bug #10008: Make Xvfb.1 document the correct default depth
92835
92836commit b0b9c811cda3e35a8f6d0813483f750602c55ff6
92837Author: George Sapountzis <gsap7@yahoo.gr>
92838Date: Sat Mar 1 20:24:50 2008 +0200
92839
92840 fix typo
92841
92842commit 3d642905477f4b1ec3223f1fbe0d0d37e959ec81
92843Author: George Sapountzis <gsap7@yahoo.gr>
92844Date: Sat Mar 1 16:18:18 2008 +0200
92845
92846 clean some "unused" warnings
92847
92848commit e7a6f79754816976d92857d55840262cccff80a6
92849Author: George Sapountzis <gsap7@yahoo.gr>
92850Date: Sat Mar 1 16:16:29 2008 +0200
92851
92852 glcore: split mesa and X in build system
92853
92854commit cdd46aa3cd2e720558186cdbe48d871ab385fcdd
92855Author: George Sapountzis <gsap7@yahoo.gr>
92856Date: Sat Mar 1 15:57:57 2008 +0200
92857
92858 configure: minor cleanup
92859
92860 - dmx
92861
92862 - darwin: remove from xorg options
92863
92864 - xephyr: libxv is client lib
92865
92866commit 8af2c39bcc4ddc4693d5a2597c9622fa17b6c272
92867Author: Maarten Maathuis <madman2003@gmail.com>
92868Date: Sat Mar 1 16:54:01 2008 +0100
92869
92870 Fix big mistake in commit fd41f46ac62033a724bd1f4612f19448a21c1224.
92871
92872 - When a mode is deleted, the name pointer is also free()'ed.
92873 - This leaves other modes with an invalid pointer.
92874
92875commit ef60632e200853680282016e32a7a9fb01882852
92876Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92877Date: Fri Feb 29 18:00:27 2008 -0500
92878
92879 dix: Modify callers of property and selection API to use new interfaces.
92880
92881commit cc76ea6e3ac6a405f0c198c4e62be40aa8d2b546
92882Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92883Date: Fri Feb 29 17:55:31 2008 -0500
92884
92885 XACE: Add generic support for property and selection polyinstantiation.
92886
92887commit 34bf308a9e66f1a2f48630a15b1802afad50ec24
92888Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92889Date: Fri Feb 29 18:00:23 2008 -0500
92890
92891 dix: Refactoring of selection code to allow for polyinstantiation.
92892 Introduces dixLookupSelection() API.
92893 Removes NumCurrentSelections from API.
92894
92895commit d5715f7beaad6816db27b01b67d7a3c69164d106
92896Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
92897Date: Fri Feb 29 16:16:12 2008 -0500
92898
92899 dix: Refactoring of property code to allow for polyinstantiation.
92900 Introduces dixLookupProperty() API.
92901
92902commit c0e1959f285d7a7df66f42d55912a5a595decd0f
92903Author: Adam Jackson <ajax@redhat.com>
92904Date: Fri Feb 29 16:45:11 2008 -0500
92905
92906 On second thought, revert that, it'll make large pixmaps painfully slow.
92907
92908 Need to just fix the callers.
92909
92910commit 4a44fe7c8678360d0549cf0e0d63870f3623b1db
92911Author: Chris Wilson <chris@chris-wilson.co.uk>
92912Date: Fri Feb 29 16:43:14 2008 -0500
92913
92914 Bug #10465: Use calloc() for allocating PixmapRec's.
92915
92916commit 5d5fcc7198ca54fa9dc24fe974763eff9fddabee
92917Author: Chris Wilson <chris@chris-wilson.co.uk>
92918Date: Fri Feb 29 16:42:04 2008 -0500
92919
92920 Bug #10464: Set pixel value to 0 before FindColor()
92921
92922commit 8f0a4282f0ac33625eda9466e3db0bcef64e403a
92923Author: Chris Wilson <chris@chris-wilson.co.uk>
92924Date: Fri Feb 29 16:39:29 2008 -0500
92925
92926 Bug #10463: Always initialize reference pixel before AllocColor()
92927
92928commit b7eb92774a58639aff3f26bb28a3dcff910c3fb6
92929Author: Adam Jackson <ajax@redhat.com>
92930Date: Sat Jan 12 01:22:05 2008 -0500
92931
92932 EDID 1.4: Print additional CVT support data in the log.
92933
92934commit 31014d88aff8dc8a502cf0f26e4cde141e1a92f5
92935Author: Adam Jackson <ajax@redhat.com>
92936Date: Sat Jan 12 01:03:44 2008 -0500
92937
92938 EDID 1.4: Decode additional CVT support information.
92939
92940 Table 3.28: Display Range Limits & CVT Support Definition
92941
92942commit d1c48955f80692a32ab6adcee1384e3d298f471a
92943Author: Adam Jackson <ajax@redhat.com>
92944Date: Sat Jan 12 00:30:58 2008 -0500
92945
92946 Fix CVT abuse in DDCModesFromStandardTiming.
92947
92948 CVT is enough different from GTF that it should not be used on monitors
92949 that aren't expecting it. This brings us closer to what the spec says
92950 the correct behaviour is.
92951
92952commit 26c2e95fa5bf30726356cf4bdd0fea32a771a179
92953Author: Adam Jackson <ajax@redhat.com>
92954Date: Sat Jan 12 00:09:34 2008 -0500
92955
92956 Nuke a duplicate SYMFUNC(xf86CVTMode)
92957
92958commit ca5625b911e65fdfd410247b3eff57fedcfc1f79
92959Author: Adam Jackson <ajax@redhat.com>
92960Date: Sat Jan 12 00:08:00 2008 -0500
92961
92962 Add xf86GTFMode().
92963
92964 This should probably be shared like xf86CVTMode().
92965
92966commit 4cb4817c1072e1d31333db47d95f71d08bf0d1dc
92967Author: Adam Jackson <ajax@redhat.com>
92968Date: Fri Jan 11 23:38:48 2008 -0500
92969
92970 Remove the duplicate copy of xf86cvt.c
92971
92972commit e65e51a99b17a0510782775f010e9820ca567fcb
92973Author: Adam Jackson <ajax@redhat.com>
92974Date: Fri Jan 11 23:19:20 2008 -0500
92975
92976 Constify the built-in mode tables.
92977
92978commit 6828d8fc2b464e0755f46e3fbdeb07be0c38b620
92979Author: Adam Jackson <ajax@redhat.com>
92980Date: Fri Jan 11 22:57:42 2008 -0500
92981
92982 Clean up DisplayModeRec handling in many places.
92983
92984 Use xf86DuplicateMode() instead of rolling our own, and change
92985 malloc+memset to calloc.
92986
92987commit 85617b56e5e00e7b8c7d8ce5b49af289056921a7
92988Author: Adam Jackson <ajax@redhat.com>
92989Date: Fri Jan 11 22:55:37 2008 -0500
92990
92991 Remove some #if 0 guarding code duplicated in xf86Modes.c
92992
92993commit fd41f46ac62033a724bd1f4612f19448a21c1224
92994Author: Adam Jackson <ajax@redhat.com>
92995Date: Fri Jan 11 22:38:21 2008 -0500
92996
92997 Allow xf86DuplicateMode() to work correctly on read-only modes.
92998
92999 Before this it was meaningless to try to mark DisplayModeRec tables
93000 const, since the mode name would be emitted as a pointer to an
93001 anonymous string constant, and therefore would have to be fixed up by
93002 ld.so and so couldn't live in .rodata. With this change the standard
93003 mode lists can live in .rodata, and modes duplicated from them will
93004 have their names filled in on the fly.
93005
93006commit 8ac19d16a030ec416e30d3650cf43e024ada167f
93007Author: Adam Jackson <ajax@redhat.com>
93008Date: Fri Jan 11 21:58:21 2008 -0500
93009
93010 Add several comments documenting our EDID failures.
93011
93012commit 13bfa5937d43392f686b76a99ea6331e3dce5987
93013Author: Kristian Høgsberg <krh@redhat.com>
93014Date: Fri Feb 29 15:10:36 2008 -0500
93015
93016 GLX: Adjust to changes in DRI driver interface.
93017
93018commit d04ea267a4a51c16088d9ef429681a1edde536b1
93019Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93020Date: Thu Feb 28 21:53:16 2008 -0500
93021
93022 xselinux: Don't require device "read" permission for XQueryPointer.
93023
93024 These keyboard and pointer state polling calls are a real problem.
93025
93026commit 3fb17a3e647e926688c91a49a9b5b97f37dbc367
93027Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93028Date: Thu Feb 28 21:52:57 2008 -0500
93029
93030 xselinux: Log messages to both libaudit and Xorg.0.log.
93031
93032commit 4d91b1d5e422c5c460b1b7050baa9487a59b8aa8
93033Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93034Date: Thu Feb 28 21:52:32 2008 -0500
93035
93036 XACE: Adjust the location of the COMPOSITE creation hook.
93037 Avoids incrementing the refcnt if the hook fails.
93038
93039commit 5675ae1f72145e9b719c613023da525731b42461
93040Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93041Date: Thu Feb 28 18:12:52 2008 -0500
93042
93043 XACE: Call the creation hook to properly label COMPOSITE window pixmaps.
93044
93045commit 27bcf40cda1d7c52b189cc76528f1f51cbe1d5eb
93046Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93047Date: Thu Feb 28 16:43:43 2008 -0500
93048
93049 XACE: Fix instances of DixUnknownAccess at hook callsites.
93050
93051commit 453661a9e193a511cf5e54e6d330454163817316
93052Author: Peter Hutterer <peter@cs.unisa.edu.au>
93053Date: Wed Feb 27 20:43:50 2008 +1030
93054
93055 Xi: swap the control attached to a ChangeDeviceControl request. Bug #14170
93056
93057 Just swapping the request's data isn't enough, we need to swap the actual
93058 control as well.
93059
93060 X.Org Bug 14170 <http://bugs.freedesktop.org/show_bug.cgi?id=14170>
93061
93062commit aebd9dc252449747416b23c740a550d914275399
93063Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
93064Date: Thu Feb 28 01:22:31 2008 -0300
93065
93066 Oops, there's one more parenthesis.
93067
93068commit f616735f17a681e3add866bf199540327c322490
93069Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93070Date: Wed Feb 27 22:46:14 2008 -0500
93071
93072 xselinux: Prefix a few remaining error messages with "SELinux".
93073
93074commit e40cc5305bec656108077ab13fcc8e6e82b3707a
93075Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93076Date: Wed Feb 27 22:29:15 2008 -0500
93077
93078 xselinux: Don't throw BadAccess if DixUnknownAccess is passed in to a hook.
93079
93080 The avc will still appear, however, so that the callsite can be fixed.
93081
93082commit 3b1df47bd400be9dca34b5e5d1ac2b117f8cc4ed
93083Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93084Date: Wed Feb 27 22:28:54 2008 -0500
93085
93086 XACE: Require "manage" permission for XKBSetNames.
93087
93088commit 08afc70513e5496cc5cd8b76c8658c4292119e4b
93089Author: Dave Airlie <airlied@redhat.com>
93090Date: Thu Feb 28 10:45:41 2008 +1000
93091
93092 quirks: another LPL panel with the cm/mm wrong
93093
93094commit f6d51797a523ecc2d4a8f18b2681160fcd5d1d55
93095Author: Maarten Maathuis <madman2003@gmail.com>
93096Date: Wed Feb 27 23:44:39 2008 +0100
93097
93098 EXA: Let the driver decide what repeat/extend types are supported.
93099
93100commit 41aea6194bd29ab34cc166b3fd90eee64299ddf8
93101Author: Alan Hourihane <alanh@tungstengraphics.com>
93102Date: Wed Feb 27 17:06:27 2008 +0000
93103
93104 More checking for failed contexts/drawables
93105
93106commit 43e46a654fa7cf69f0c8bcb7f583008d96a98686
93107Author: Alan Hourihane <alanh@tungstengraphics.com>
93108Date: Wed Feb 27 16:58:21 2008 +0000
93109
93110 Fix context sharing between direct/indirect contexts
93111
93112commit d74b0327e8355546e6b41e8ce684f461d65fa9dc
93113Author: Dodji Seketeli <dodji@openedhand.com>
93114Date: Wed Feb 27 09:48:10 2008 +0100
93115
93116 [Xephyr/GL] unbreak the build
93117
93118 Don't touch Xephyr DRI stuff when not compiling in a DRI environment.
93119
93120commit 2257e20900460d85254734b595238e7ad5ee55c8
93121Author: Peter Hutterer <peter@cs.unisa.edu.au>
93122Date: Wed Feb 27 17:15:28 2008 +1030
93123
93124 dix: set dev->key to NULL after freeing it in CloseDevice. (Bug #12830)
93125
93126 Otherwise XkbRemoveResourceClient may try to dereference it lateron.
93127
93128 X.Org Bug 12830 <http://bugs.freedesktop.org/show_bug.cgi?id=12830>
93129
93130commit 3f0681fb0b2d0744c2ef892883ae5359b43a4a9a
93131Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93132Date: Tue Feb 26 23:14:29 2008 -0500
93133
93134 xselinux: Stub out selection protocol requests.
93135
93136commit 4632ea22580c31d44b0786321668d9e78f02900e
93137Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93138Date: Tue Feb 26 22:00:52 2008 -0500
93139
93140 xselinux: Rip out the selection code in advance of polyinstantiation support.
93141
93142 This resolves an issue where BadWindow errors were being thrown.
93143
93144commit c8e979b3b800e4c58be454daa0213285d4ee6510
93145Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93146Date: Tue Feb 26 21:34:22 2008 -0500
93147
93148 security: Register resource names in the server-side name registry.
93149
93150commit d12b7b6632fb4cf41d2e28c7792eaa503f25404a
93151Author: Dave Airlie <airlied@redhat.com>
93152Date: Wed Feb 27 07:08:00 2008 +1000
93153
93154 xf86Crtc: add higher level modesetting entry point.
93155
93156 For kernel modesetting work we need a bigger stick to beat the modesetting path
93157
93158commit dcc077c753137f37aa58231f1df3c4adb92b2c4b
93159Author: Michel Dänzer <michel@tungstengraphics.com>
93160Date: Tue Feb 26 12:13:06 2008 +0100
93161
93162 AIGLX: Switch to server context for calling DamageDamageRegion().
93163
93164 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=14518 .
93165
93166commit 23ae68a4c74a2ec90b4130c37b0d0aec3f4082ce
93167Author: Peter Hutterer <peter@cs.unisa.edu.au>
93168Date: Tue Feb 26 15:12:36 2008 +1030
93169
93170 dix: before copying the classes the first time, set the VCK's classes to NULL.
93171
93172 XkbFinishDeviceInit does the following:
93173 xkbi->kbdProc= pXDev->kbdfeed->CtrlProc;
93174 pXDev->kbdfeed->CtrlProc= XkbDDXKeybdCtrlProc;
93175
93176 If we directly copy the device classes for the VCK, pXDev->kbdfeed->CtrlProc
93177 at the time of copying is still XbkDDXKeybdCtrlProc. So at some point
93178 XkbDDXKeybdCtrlProc is called, and calls itself, and calls itself, and...
93179
93180 Setting the device's classes to NULL seems to fix things. The memory isn't
93181 lost, it gets stored into the devPrivates and freed at device closing time.
93182
93183commit ce9fb2f8c4610016e49321018fc9b24729380afc
93184Author: Peter Hutterer <peter@cs.unisa.edu.au>
93185Date: Tue Feb 26 13:19:54 2008 +1030
93186
93187 Xi: MDs return the paired device's ID in attached field of ListInputDevices.
93188
93189commit c46f7b62d2bf9f35937cfef98325ed904895396a
93190Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
93191Date: Mon Feb 25 18:14:08 2008 -0300
93192
93193 Fine, we don't need pciInit() anymore. Nuke, nuke, nuke...
93194
93195commit a9050d54249ed9f54c6fe59143b846c9c7548f59
93196Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
93197Date: Mon Feb 25 18:10:18 2008 -0300
93198
93199 Jesus, pciInit() was called 32 times in my machine without need!
93200
93201 xf86scanpci() will always call pciInit() before any other function, so
93202 there's no need to guarantee it after.
93203
93204commit 81272f7ec9a3465e1d102c2ce627a45f92268857
93205Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
93206Date: Mon Feb 25 17:39:33 2008 -0300
93207
93208 Some doc clean up:
93209 clean up legacy things in os-support/bus/Pci.c.
93210
93211 Put InitOutput() comment to its right place.
93212
93213commit 7c582dd6de27d2f4fedf73319d2dea2bfb240efa
93214Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
93215Date: Mon Feb 25 17:31:37 2008 -0300
93216
93217 remove lnxPciInit() unused function.
93218
93219commit 9727db88d57089be6483104de435626cdbad883a
93220Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
93221Date: Mon Feb 25 17:08:07 2008 -0300
93222
93223 No more "-scanpci" given that we already have it in libpciaccess.
93224
93225 (Yeah, lets nuke dead code!)
93226
93227commit f19f7b8e53ed6609fc1fdd272de5521417946209
93228Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
93229Date: Mon Feb 25 17:07:07 2008 -0300
93230
93231 Clean some garbage caused when pciaccess was merged:
93232
93233 FindPCIVideoInfo() function isn't need anymore.
93234
93235 xf86scanpci() is being called only once so we don't need permanent
93236 (static) variables there.
93237
93238 restorePciState() is not used for now (until we find why multiple
93239 cards aren't working).
93240
93241commit d61f481a4455dd2a94674d2b7b26429cf9dcece3
93242Author: Peter Hutterer <peter@cs.unisa.edu.au>
93243Date: Mon Feb 25 17:52:45 2008 +1030
93244
93245 Xi: remove GetPairedPointer handling.
93246
93247 obsolete, ListInputDevices provides this information now.
93248
93249commit 27b21278939a06ab6a22c9902f202eae6c5fae72
93250Author: Peter Hutterer <peter@cs.unisa.edu.au>
93251Date: Mon Feb 25 17:00:42 2008 +1030
93252
93253 Xi: remove ungrab handling of ExtendedUngrabDevice request.
93254
93255 This can be done by UngrabDevice, no need for separate codepaths.
93256
93257commit 4b50e71bf127c8e0f289e3b76c786f0398effe65
93258Author: Adam Jackson <ajax@redhat.com>
93259Date: Sun Feb 24 20:25:13 2008 -0500
93260
93261 Bug #13736: Fix %bx in VBEGetPixelClock to match spec.
93262
93263 Reported by Yannick Henault.
93264
93265commit 6dc369028d3ca741de57ad78febf2f5f82e0696e
93266Author: Jeremy Huddleston <jeremy@tifa.local>
93267Date: Sat Feb 23 00:01:02 2008 -0800
93268
93269 XQuartz: Quit now properly warns the user
93270 (cherry picked from commit ed3d7b3959c2a0cb63e37210455bcc6cf195b807)
93271
93272commit e6a4cde16dc99ea02ac93da1d1b9517b1073d159
93273Author: Adam Jackson <ajax@redhat.com>
93274Date: Fri Feb 22 18:36:29 2008 -0500
93275
93276 Use the client-side XKB headers for the config utilities
93277
93278commit 347db49ebe4596db16455ea8a1a608cfa826c5c7
93279Author: Adam Jackson <ajax@redhat.com>
93280Date: Fri Feb 22 16:05:33 2008 -0500
93281
93282 s/via/openchrome/ in the autoconfig logic.
93283
93284 Xorg's via driver is effectively dead anyway.
93285
93286commit 69f782676797744815ff76b8a11b11178066f501
93287Author: Adam Jackson <ajax@redhat.com>
93288Date: Fri Feb 22 16:04:35 2008 -0500
93289
93290 Match Xephyr DRI definitions to the ones in xf86dri.h
93291
93292commit 060a66b6e2feddba43ed207b6fcf2cf1f7fe39fd
93293Author: Adam Jackson <ajax@redhat.com>
93294Date: Mon Dec 24 15:55:58 2007 -0500
93295
93296 Normalize swapped dispatch for Fixes{ChangeSaveSet,SelectSelectionInput}
93297
93298commit 24bebdded44a9e184455b4fee7800257fee81efb
93299Author: Adam Jackson <ajax@redhat.com>
93300Date: Mon Dec 24 15:07:49 2007 -0500
93301
93302 fbFillRegionTiled() is now dead code.
93303
93304 Only ever called from the old PaintWindow* screen hooks, but those are
93305 gone now. As a pleasant side effect, fb loses its #ifdef PANORAMIX.
93306
93307commit ee21aba6be0078949204e315ddfffd99de60c2f1
93308Author: Adam Jackson <ajax@redhat.com>
93309Date: Mon Dec 24 13:13:19 2007 -0500
93310
93311 Fix Xinerama's consolidated visual handling.
93312
93313 Formerly the code claimed it could only handle up to 256 visuals, which
93314 was true. Also true, but not explicitly stated, was that it could only
93315 handle visuals with VID < 256. If you have enough screens, and subsystems
93316 that add lots of visuals, you can easily run off the end. (Made worse
93317 because we allocate visual IDs from the same pool as XIDs.) If your app
93318 then chooses a visual > 256, then the Xinerama code would throw BadMatch
93319 on CreateColormap and your app wouldn't start.
93320
93321 With this change, PanoramiXVisualTable is gone. Other subsystems that
93322 were using it as a translation table between each screen's visuals now
93323 use a PanoramiXTranslateVisual() helper.
93324
93325commit a4202b898f07dd733590ae5adb21e48775369781
93326Author: Daniel Stone <daniel@fooishbar.org>
93327Date: Fri Feb 22 18:22:58 2008 +0100
93328
93329 XKB: Actually use the keymap we compile at startup
93330
93331 During XkbInitKeyboardDevice, we compiled a keymap and promptly threw it away;
93332 brief inspection revealed the embarassingly simple problem. Sorry.
93333
93334commit c14fd2a5cb3f45d5c4502e09f55f5e3732c5e698
93335Author: Dodji Seketeli <dodji@seketeli.org>
93336Date: Thu Feb 21 15:33:02 2008 +0100
93337
93338 [Xephyr/GL] properly route expose event on GL drawables
93339
93340 When an expose event happens on an host GL window paired with an
93341 internal drawable, route that expose event to the clients listening
93342 to the expose event on the internal drawable.
93343
93344commit 437c78ef9ff1177e04b3d6781b5805d89b2ab81a
93345Author: Dodji Seketeli <dodji@seketeli.org>
93346Date: Thu Feb 21 15:29:27 2008 +0100
93347
93348 [Xephyr/GL] don't crash when the host returns a NULL server string
93349
93350commit f343265a289724c81017f089c024a7618267c4e3
93351Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93352Date: Fri Feb 15 19:53:45 2008 -0500
93353
93354 XACE: Make the default window background state configurable per-window.
93355
93356 To recap: the original XC-SECURITY extension disallowed background "None" if
93357 the window was untrusted. XACE 1.0 preserved this check as a hook function.
93358 XACE pre-2.0 removed the hook and first abolished background "None entirely,
93359 then restored it as a global on/off switch in response to Bug #13683.
93360 Now it's back to being per-window, via a flag instead of a hook function.
93361
93362commit 7c2f0a8befb310707ea923dbcdfde84521e52c88
93363Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93364Date: Fri Feb 15 19:17:40 2008 -0500
93365
93366 Remove COMPOSITE ifdefs around WindowRec bitfield as it has no ABI effect.
93367
93368commit 5cb9e15562a32c1f102d94d5e15d5fd298baff3f
93369Author: Eric Anholt <eric@anholt.net>
93370Date: Wed Feb 20 10:36:06 2008 -0800
93371
93372 EXA: Fix Render acceleration in copy and tiling cases.
93373
93374 Code shuffling in a634c9b03494ba80aeec28be19662ac96657cc23 broke this by
93375 leaving pSrcPixmap = NULL.
93376
93377commit f14a62f823e257f92745bbcde11838f2ddd32ac8
93378Author: Peter Hutterer <peter@cs.unisa.edu.au>
93379Date: Tue Feb 19 21:44:10 2008 +1030
93380
93381 dix: set evlen to the size of the reallocated memory.
93382
93383 What a good idea this is... I'm very proud of myself.
93384
93385commit 67a78e84a81571cedaf7fd214a21ce1bbdc4fb3b
93386Author: Peter Hutterer <peter@cs.unisa.edu.au>
93387Date: Tue Feb 19 21:31:50 2008 +1030
93388
93389 Revert "xfree86: plug a memory leak in xf86LoadModules."
93390
93391 This reverts commit 3abce3ea2b1f43bd111664d4a57e5ccd282acab0 and
93392 6cbaf15e6109ba77819c4070f5b46c78237ec460.
93393
93394 The memory returned to xf86LoadModule was allocated in doLoadModule, which
93395 calls the respective module's PreInit. As it turns out, input and output
93396 drivers store a pointer to the module elswhere, so freeing it in
93397 xf86LoadModule is a bad idea.
93398
93399 For further reference: hw/xfree86/common/xf86Helper.c
93400 Input drivers: xf86InputDriverList[blah]->module = module;
93401 Output drivers: xf86DriverList[blah]->module = module;
93402
93403 Unloading the module would not look pretty then.
93404
93405commit 5f5ec5db35e82ddd9659763875e5d6c63cf1b691
93406Author: Jeremy Huddleston <jeremy@tifa.local>
93407Date: Tue Feb 19 00:00:11 2008 -0800
93408
93409 XQuartz: unsetenv("DISPLAY") before startx if X11.app can't connect to it.
93410 Also fix casting to silence warning.
93411 (cherry picked from commit a5cbf78471ec6e6ad672dc00118fc7edbd6ddc7c)
93412
93413commit 750d70267679ddee10590c80ec621d890bd3d4a7
93414Author: Peter Hutterer <peter@cs.unisa.edu.au>
93415Date: Sun Feb 3 10:25:15 2008 +1030
93416
93417 dix: Ensure enough memory for ClassesChangedEvent for a new device.
93418
93419 Before we enable the device through the driver, we size it up and make sure
93420 that the events in the event list contain enough bytes for a possible
93421 ClassesChangedEvent lateron.
93422
93423commit 74628d639719815c1beff4cac84662fa41c55925
93424Author: Peter Hutterer <peter@cs.unisa.edu.au>
93425Date: Sun Feb 3 10:15:40 2008 +1030
93426
93427 dix: change GetEventList to return length of list and set parameter in place.
93428
93429 Changing DDXs to use new prototype too.
93430
93431commit 77dba004a9aaf35f183f61ff6875a491a52aa030
93432Author: Peter Hutterer <peter@cs.unisa.edu.au>
93433Date: Sun Feb 3 10:10:46 2008 +1030
93434
93435 dix: add InputEventListLen and SetMinimumEventSize
93436
93437 The latter is used to increase the amount of allocated memory for the event
93438 list. This will be needed for ClassesChangedEvents that can be of more or less
93439 arbitrary size (larger than 32 anyway).
93440
93441commit 6cbaf15e6109ba77819c4070f5b46c78237ec460
93442Author: Peter Hutterer <peter@cs.unisa.edu.au>
93443Date: Tue Feb 19 12:02:22 2008 +1030
93444
93445 xfree86: guard against NULL-pointer dereferences in xf86LoadModules.
93446
93447 Should have been part of 3abce3ea2b1f43bd111664d4a57e5ccd282acab0, but I
93448 forgot to git-update-index before I committed.
93449
93450 Thanks to Bill Crawford for pointing it out.
93451
93452commit 3fe64d8d271aea0863bf01b0376f3eceec0c90b5
93453Author: Peter Hutterer <peter@cs.unisa.edu.au>
93454Date: Sun Feb 3 09:56:19 2008 +1030
93455
93456 Move input event list initialisation and storage from DDX to DIX.
93457
93458 Rather than letting the DDX allocate the events, allocate them once in the DIX
93459 and just pass it around when needed.
93460
93461 DDX should call GetEventList() to obtain this list and then pass it into
93462 Get{Pointer|Keyboard}Events.
93463
93464commit 09a8fc5c7a79ca22fc23224bb544f2e709681f3f
93465Author: Peter Hutterer <peter@cs.unisa.edu.au>
93466Date: Sun Feb 3 07:47:18 2008 +1030
93467
93468 Xi: make SizeDeviceInfo public and re-use from CreateClassesChangedEvent.
93469
93470commit 3abce3ea2b1f43bd111664d4a57e5ccd282acab0
93471Author: Arjan van de Ven <arjan@infradead.org>
93472Date: Mon Feb 18 18:13:10 2008 +1030
93473
93474 xfree86: plug a memory leak in xf86LoadModules.
93475
93476 LoadModule() returns the only reference to a fresh piece of memory (a
93477 ModuleDescPtr). Sadly, xf86LoadModules dropped the return value on the floor
93478 leaking memory for each module it loaded.
93479
93480 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
93481
93482commit 6dc71f6b2c7ff49adb504426b4cd206e4745e1e3
93483Author: Arjan van de Ven <arjan@infradead.org>
93484Date: Mon Feb 18 17:52:37 2008 +1030
93485
93486 xfree86: plug memory leak in InitPathList.
93487
93488 All the failure paths were very diligent in freeing the "fullpath" temporary
93489 string, but the success case was not. All the content only got strdup()d, so
93490 it's not live memory anymore.
93491
93492 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
93493
93494commit 1bec6ad8977cefa49cc297a310f5eb0b7cd0b8bc
93495Author: Arjan van de Ven <arjan@infradead.org>
93496Date: Mon Feb 18 17:46:04 2008 +1030
93497
93498 xfree86: plug memory leak in xf86LogInit()
93499
93500 xf86LogInit allocates a piece of memory, stores it in lf. LogInit() will then
93501 effectively strdup it, but lf is never freed again.
93502
93503 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
93504
93505commit 6d9d18bf7708eb9e640ef732b5dc4c99aa6d5feb
93506Author: Peter Hutterer <peter@cs.unisa.edu.au>
93507Date: Mon Feb 18 17:25:15 2008 +1030
93508
93509 dix: add support for WindowAccessAllowAll in device access checking.
93510
93511 AllowAll is the last check before the parent window is checked. This allows
93512 to override a DenyAll in the parent window in a simpler way than before (the
93513 previous method required all devices to be in the permit list).
93514
93515commit 088067c891a78670d9509f48f56bf3ff9c16a30d
93516Author: Peter Hutterer <peter@cs.unisa.edu.au>
93517Date: Mon Feb 18 16:46:07 2008 +1030
93518
93519 Xi: even if ChangeDeviceHierarchy fails, we may need to send an event.
93520
93521 Changes are committed instantly, so if at least one change was successful, we
93522 must send an event to the client, even if subsequent ones fail.
93523
93524commit 660739c6bc84cb74f43a277052ce163fae654417
93525Author: Benjamin Close <Benjamin.Close@clearchain.com>
93526Date: Fri Feb 15 13:36:34 2008 +1030
93527
93528 dix: Handle the case where a device cursor was removed correctly
93529
93530 In the case that the device cursor was the first in the list of cursors
93531 the window knew about, unsetting the cursor could lead to a segfault
93532 due to pPrev being NULL. Instead catch the condition and correctly remove
93533 the node from the list. Since there is no cursor now set on the device,
93534 we simply return success as the parent windows cursor will propogate
93535 down later.
93536
93537 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
93538
93539commit 24db28230120ecc2b65b25164b6e7b407970f9e2
93540Author: Peter Hutterer <peter@cs.unisa.edu.au>
93541Date: Mon Feb 18 16:16:19 2008 +1030
93542
93543 dix: set num_events to 1 by default. (no RawDeviceEvents anymore)
93544
93545 This should have been part of acb412d5399f865b5048fdd683147cea0f19f0c1.
93546
93547commit 45d00433e7bfbab476dd02ec92a9fbee40af5dad
93548Author: Dan Nicholson <dbn.lists@gmail.com>
93549Date: Tue Jan 15 19:43:16 2008 -0800
93550
93551 glx: Use glapi sources from the mesa tree
93552
93553 With recent mesa HEAD, the glapi sources used only in the xserver glx
93554 module are carried in the mesa tree. Previously, these were generated
93555 separately and committed to the xserver tree.
93556
93557 The build is changed to symlink these files like the other mesa sources.
93558 This reduces the chance for mismatches between mesa's glX API and the
93559 xserver's glX API.
93560
93561commit fbd776894658e7afb2c55dc8582b2a3efab78a3c
93562Author: Daniel Stone <daniel@fooishbar.org>
93563Date: Sat Feb 16 20:01:18 2008 +0200
93564
93565 XKB: Ditch XkbFileInfo
93566
93567 Sorry about the megacommit, but this touches on a lot of stuff.
93568
93569 Get rid of XkbFileInfo, which was pretty seriously redundant, and move the
93570 only useful thing it had (defined) into XkbDescRec. defined will be removed
93571 pretty soon anyway. Is the compat map pointer non-NULL? Then you have a
93572 compat map, congratulations! Anyhow, I digress.
93573
93574 All functions that took an XkbFileInfoPtr now take an XkbDescPtr, _except_
93575 XkmReadFile, which returns an XkbDescPtr *, because people want to deal in
93576 XkbDescPtrs, not XkbDescRecs.
93577
93578commit e5f002eddef1abe324033a3155f01d048536a48d
93579Author: Daniel Stone <daniel@fooishbar.org>
93580Date: Sat Feb 16 20:00:47 2008 +0200
93581
93582 XkbProcessOtherEvent: Don't depend on now-removed header
93583
93584 We don't do XKBsrv.h anymore.
93585
93586commit 2d256f098ae05033ad76672d5ebdb9dfa7e6b995
93587Author: Daniel Stone <daniel@fooishbar.org>
93588Date: Sat Feb 16 19:57:37 2008 +0200
93589
93590 XKB: Always set size correctly in XkbCopyKeymap's geometry routines
93591
93592 We were forgetting to set the sizes for sections and rows and a couple of
93593 other misc bits in XkbCopyKeymap's geometry. Sort that out, and add a
93594 couple of clarifying comments along the way.
93595
93596commit ab79110a84b2d299ecae0605fa535edbebd99565
93597Author: Daniel Stone <daniel@fooishbar.org>
93598Date: Sun Feb 3 23:48:57 2008 +1100
93599
93600 XKB: Remove support for pre-built keymaps
93601
93602 Don't load prebuilt keymaps anymore.
93603
93604commit 1332343910a1880854bae21a790cff198855cd24
93605Author: Daniel Stone <daniel@fooishbar.org>
93606Date: Sun Feb 3 23:43:18 2008 +1100
93607
93608 XKB: Remove usage of client-side types
93609
93610 Since we're no longer sharing with Xlib, don't pass Displays and XPointers
93611 everywhere.
93612
93613commit 534fc5140b039a8c98ab715d0a6740d513b41209
93614Author: Daniel Stone <daniel@fooishbar.org>
93615Date: Sun Feb 3 23:30:22 2008 +1100
93616
93617 XKB: Remove a bunch of mad ifdefs
93618
93619 We have SEEK_SET and size_t, seriously. Also use DebugF instead of
93620 ifdef DEBUG, and ditch a couple of random bits that were never used.
93621
93622commit 0f12a448dcbbbf1f40aa98d09c9d25ee511c5bbf
93623Author: Daniel Stone <daniel@fooishbar.org>
93624Date: Sun Feb 3 23:15:39 2008 +1100
93625
93626 XKB: Deprecate XKBSRV_NEED_FILE_FUNCS
93627
93628 There's no point in having the function definitions be conditional, so
93629 whatever.
93630
93631commit 68bd7ac1930b5cffb6657b8d5f5bf8ae58eae8d9
93632Author: Daniel Stone <daniel@fooishbar.org>
93633Date: Sun Feb 3 23:12:15 2008 +1100
93634
93635 XKB: Move headers into the server tree
93636
93637 We need to start breaking the XKB API to enforce sanity, so drag whichever
93638 headers we need to do so into the server tree, as the client API is set in
93639 stone, being part of Xlib.
93640
93641commit 37867626e314e74031378a8a4ff06f69b899a3b2
93642Author: Daniel Stone <daniel@fooishbar.org>
93643Date: Fri Feb 1 14:41:04 2008 +1100
93644
93645 main(): Remove uncredible failure
93646
93647 NO, NO, NO.
93648
93649 NO.
93650
93651 The only way this could possibly be worse is if you were viewing it in
93652 Comic Sans.
93653
93654commit e4eb7e5842f0f56f07e9cf3b16249c172d0a975d
93655Author: Daniel Stone <daniel@fooishbar.org>
93656Date: Wed Jan 30 23:24:14 2008 +1100
93657
93658 XKB: Delete xkberrs.c
93659
93660 Get rid of the XKB errors code to save a bunch of space.
93661
93662commit a48cc88ea2674c28b69b8d738b168cbafcf4001f
93663Author: Maarten Maathuis <madman2003@gmail.com>
93664Date: Sun Feb 17 18:47:28 2008 +0100
93665
93666 Fix rotation for multi-monitor situation.
93667
93668 - The (x,y)-coordinates of the crtc were not being passed as xFixed values, which made it an obscure bug to find.
93669 - Fix bug #13787.
93670
93671commit 70c0592a97c7dc9db0576d32b3bdbe4766520509
93672Author: Maarten Maathuis <madman2003@gmail.com>
93673Date: Sun Feb 17 11:21:01 2008 +0100
93674
93675 Resize composite overlay window when the root window changes.
93676
93677 - This allows some compositing managers to work, even after randr12 has changed the root window size.
93678 - Thanks to ajax for figuring out the best place to put this.
93679 - Example:
93680 - xf86RandR12SetMode() calls EnableDisableFBAccess().
93681 - That calls xf86SetRootClip() which in turn calls ResizeChildrenWinSize().
93682 - The final step is the call to PositionWindow().
93683
93684commit b95059c20746a71e60ef152bb627e1d5c2210d75
93685Author: Jeremy Huddleston <jeremy@tifa-2.local>
93686Date: Sat Feb 16 01:33:13 2008 -0800
93687
93688 Added Xquartz.plist to EXTRA_DIST
93689 (cherry picked from commit 70f9495e0c1f1459507064b673fe57b1c90d3c2c)
93690
93691commit d103820bb8635c63b34b85b45cad95ed9c152d90
93692Author: Jeremy Huddleston <jeremy@tifa-2.local>
93693Date: Sat Feb 16 00:35:50 2008 -0800
93694
93695 Fixed unescaped newline in EXTRA_DIST
93696 (cherry picked from commit 567c172c4d400fdfe69e7b096a3877fce5c2de9f)
93697
93698commit a7d936c4ac8e3d7227ecbfe0ddc6cc257b450458
93699Author: Jeremy Huddleston <jeremy@tifa.local>
93700Date: Mon Feb 11 16:09:03 2008 -0800
93701
93702 XQuartz: Fixed EXTRA_DIST to include localization
93703 (cherry picked from commit 1b338c2a9330c85490a7c24d52adf24b124b70e6)
93704
93705commit 0ffb6a3ad010e80fe8f973fc228d549f9dd3effd
93706Author: Kristian Høgsberg <krh@redhat.com>
93707Date: Thu Feb 14 22:20:56 2008 -0500
93708
93709 GLX: Implement support for TTM BO based TFP when available.
93710
93711commit 879515b1399f87a47010532af70f34b9b09e2a9b
93712Author: Kristian Høgsberg <krh@sasori.boston.redhat.com>
93713Date: Mon Feb 4 13:13:35 2008 -0500
93714
93715 Add GLX provider for DRI2.
93716
93717commit b71dc54352bc56a889823040ec19c1d8e118a1f3
93718Author: Kristian Høgsberg <krh@sasori.boston.redhat.com>
93719Date: Mon Feb 4 11:58:24 2008 -0500
93720
93721 Add DRI2 module.
93722
93723commit 005e31d3de04e7003f84a94d30f2b75a9636266e
93724Author: Michel Dänzer <michel@tungstengraphics.com>
93725Date: Fri Feb 15 00:01:32 2008 +0000
93726
93727 AIGLX: Refactor code common between __glXDRI{drawableDestroy,releaseTexImage}.
93728
93729commit cd78f0d0fc08e4e2339ed09dad1a12802de7729c
93730Author: Pierre Willenbrock <pierre@pirsoft.de>
93731Date: Thu Feb 7 21:28:28 2008 +0100
93732
93733 AIGLX: Fix crash after client using GLX_EXT_texture_from_pixmap died.
93734
93735commit 3e0353c78571890f849a1db47b0540aacc6793bc
93736Author: Adam Jackson <ajax@redhat.com>
93737Date: Fri Feb 15 06:42:48 2008 +1100
93738
93739 Bah, macros are hard.
93740
93741commit c38feeb1492d9a47379b2e4d77dbadd8c421d17f
93742Author: Maarten Maathuis <madman2003@gmail.com>
93743Date: Thu Feb 14 18:52:47 2008 +0100
93744
93745 Fix typo in commit b8b7cdf6df3d338c50db670ce4cfd245f9fa8844.
93746
93747commit b8b7cdf6df3d338c50db670ce4cfd245f9fa8844
93748Author: Adam Jackson <ajax@redhat.com>
93749Date: Fri Feb 15 01:31:46 2008 +1100
93750
93751 Silence useless debug spew from XFree86-Misc extension.
93752
93753commit 4573cb2ce4b859744118d9a33de3599f033cdd2b
93754Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93755Date: Wed Feb 13 20:16:56 2008 -0500
93756
93757 security: Revise set of permissions granted to untrusted clients.
93758
93759 Bug #14480: untrusted access broken on 7.3.
93760
93761commit e99aadbc2635e87543fc9980d8156c3ede83544f
93762Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93763Date: Wed Feb 13 19:46:29 2008 -0500
93764
93765 xselinux: Add use to permission map for devices.
93766
93767commit ae43d835bdaef96c3c73d7cee5b105f07340833d
93768Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93769Date: Wed Feb 13 19:39:49 2008 -0500
93770
93771 XACE: Change access modes for some device-related requests.
93772
93773 Opening a device is not really "reading" it.
93774 Requests that globally configure a device should require "manage" access.
93775
93776commit 921c298c4cdd2c879403ebdacbef48129058cad4
93777Author: Peter Hutterer <peter@cs.unisa.edu.au>
93778Date: Thu Feb 14 09:48:19 2008 +1030
93779
93780 Xi: QueryDevicePointer doesn't have a shared field anymore.
93781
93782commit b885588756450e2fa25fdd191e300e1c5dd37ec3
93783Author: Peter Hutterer <peter@cs.unisa.edu.au>
93784Date: Thu Feb 14 09:47:31 2008 +1030
93785
93786 Xi: remove FakeDeviceData calls. And one leftover from GrabAccessControl.
93787
93788 FakeDeviceData needs more thoughtful integration.
93789
93790commit c99a9a97290c51bf2843f42d5683888cea8a3ff6
93791Author: Peter Hutterer <peter@cs.unisa.edu.au>
93792Date: Thu Feb 14 09:43:49 2008 +1030
93793
93794 Xi: remove raw device events.
93795
93796 Need to be better refined.
93797
93798commit b35a9efa16d9e14378a95c11fa39171b78710996
93799Author: Peter Hutterer <peter@cs.unisa.edu.au>
93800Date: Thu Feb 14 09:41:49 2008 +1030
93801
93802 Xi: Remove files for GrabAccessControl.
93803
93804commit b81c330f70c4a4c8f37216aa4856c7f8db93a375
93805Author: Peter Hutterer <peter@cs.unisa.edu.au>
93806Date: Thu Feb 14 09:32:54 2008 +1030
93807
93808 mi: remove RawDeviceEvent's ID changing.
93809
93810commit acb412d5399f865b5048fdd683147cea0f19f0c1
93811Author: Peter Hutterer <peter@cs.unisa.edu.au>
93812Date: Thu Feb 14 09:27:51 2008 +1030
93813
93814 dix: Don't generate RawDeviceEvents anymore.
93815
93816 These events have been removed from the protocol pending some more refinement
93817 and more thoughtful integration.
93818
93819commit ba69904148acf755bec8fbda2eb869144f0ef7d4
93820Author: Adam Jackson <ajax@redhat.com>
93821Date: Sun Feb 3 22:49:23 2008 +1100
93822
93823 Simplify dlloader a bit more.
93824
93825commit f750ce53ac450824d0c792d11eafdf311cf8abc0
93826Author: Adam Jackson <ajax@redhat.com>
93827Date: Fri Feb 1 17:33:48 2008 +1100
93828
93829 Remove some braindamage from ModuleDesc.
93830
93831 Yay dead code elimination.
93832
93833commit e91ff09568c5579818b6641e88e95c6fe122cbe7
93834Author: Adam Jackson <ajax@redhat.com>
93835Date: Fri Feb 1 17:21:34 2008 +1100
93836
93837 Friends don't let friends call xf86AddModuleInfo.
93838
93839 That code only existed to leak memory. It can go now, thanks.
93840
93841commit 0bdd20a0454c94f902fd4874855125bf7510fcf5
93842Author: Adam Jackson <ajax@redhat.com>
93843Date: Fri Feb 1 16:15:58 2008 +1100
93844
93845 Eradicate the VTInit code.
93846
93847 "This option should rarely be used." Never sounds like a better idea.
93848
93849commit 24089b06243101b1bff4f2fd79fcbfd6a93992d5
93850Author: Adam Jackson <ajax@redhat.com>
93851Date: Fri Feb 1 16:11:13 2008 +1100
93852
93853 Don't bother warning people about the keyboard driver rename, just do it.
93854
93855commit 2aaf6ac134fa9f61984b1c03929c7b596c971cc8
93856Author: Adam Jackson <ajax@redhat.com>
93857Date: Fri Feb 1 16:10:23 2008 +1100
93858
93859 Don't even attempt to parse -bpp in xfree86 DDX option parsing.
93860
93861 This hasn't been valid since 1999. Get with the times, man.
93862
93863commit ef77e4c4419703b5a802f3eee92476a43f788a86
93864Author: Adam Jackson <ajax@redhat.com>
93865Date: Fri Feb 1 16:03:01 2008 +1100
93866
93867 Remove useless commentary from environment and argument processing.
93868
93869commit f3b3b37ec6197f8884417fbc26630d3a28f2e319
93870Author: Adam Jackson <ajax@redhat.com>
93871Date: Fri Feb 1 15:36:11 2008 +1100
93872
93873 Use strerror instead of errno values in user strings.
93874
93875commit b740b865e4c156a40adc6b63fcf215156fc9151a
93876Author: Adam Jackson <ajax@redhat.com>
93877Date: Fri Feb 1 15:16:01 2008 +1100
93878
93879 Silence an error message in ConfigureWindow that never happens.
93880
93881commit 89d3249c3e7611b22414bc84b10ca60aab4b9a77
93882Author: Adam Jackson <ajax@redhat.com>
93883Date: Fri Feb 1 15:11:27 2008 +1100
93884
93885 Silence FreeResource()
93886
93887commit 238b816469355159eea98600ca900e3baa8fa313
93888Author: Adam Jackson <ajax@redhat.com>
93889Date: Fri Feb 1 15:08:03 2008 +1100
93890
93891 X.Org Group -> X.Org Foundation
93892
93893commit 9113fa3de36e84bbae2727cace82b1cf0d709a86
93894Author: Adam Jackson <ajax@redhat.com>
93895Date: Fri Feb 1 15:03:57 2008 +1100
93896
93897 Silence the harmless FreeFontPath error message.
93898
93899commit 7c0709a736c0f3aa011de67dd2c2962585ab146e
93900Author: Adam Jackson <ajax@redhat.com>
93901Date: Fri Feb 1 14:59:58 2008 +1100
93902
93903 requestingClient is an xprintism, hide it for other servers.
93904
93905commit 62cfe8863823c0994f20555cb35ee3bacb9e5225
93906Author: Adam Jackson <ajax@redhat.com>
93907Date: Fri Feb 1 14:57:25 2008 +1100
93908
93909 Redact all mention of PanoramiX from user strings.
93910
93911commit eb8dc11d19dd0b5354de408578ab0cfb865df672
93912Author: Adam Jackson <ajax@redhat.com>
93913Date: Thu Jan 31 21:45:14 2008 +1100
93914
93915 Remove some MAX* #defines that never get used.
93916
93917commit 2ce35f6d45c3e1761d33b786520ff5ba56a3c518
93918Author: Adam Jackson <ajax@redhat.com>
93919Date: Thu Jan 31 21:39:48 2008 +1100
93920
93921 Simplify critical output flushing.
93922
93923commit 3eaecdd66e791e0f3d86b23ce10be057ca44c044
93924Author: Adam Jackson <ajax@redhat.com>
93925Date: Thu Jan 31 21:36:14 2008 +1100
93926
93927 Disable Record by default.
93928
93929commit aa5216e89797b600f382c04e3eaa657e808a5c3e
93930Author: Adam Jackson <ajax@redhat.com>
93931Date: Thu Jan 31 12:05:08 2008 +1100
93932
93933 Rip out useless indirection in the callback list management.
93934
93935commit 0d492b2166c4026b9078ffd86d89a31ebe590be4
93936Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93937Date: Tue Feb 12 19:59:10 2008 -0500
93938
93939 XACE: Move the selection access hook to its own function.
93940
93941commit 4be69e41c07b784f75918141eead70415c0ec4a4
93942Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
93943Date: Mon Feb 11 16:03:25 2008 -0500
93944
93945 Revert "Bug #8937: Extension setup functions not called on server resets"
93946
93947 This reverts commit 5e946dd853a4ebc2722ae023429ce5797de3d7a6.
93948
93949 The devPrivates rework makes this workaround unnecessary.
93950
93951commit 8a244c2bc04a4184535e42d1f2f8c9e3e55678b4
93952Author: Jeremy Huddleston <jeremy@yuffie.local>
93953Date: Mon Feb 11 12:09:41 2008 -0800
93954
93955 XQuartz: Fixed layout of Applications->Customize menu
93956
93957 Now resizing it won't produce weir overlaps of the widgets. Thanks to
93958 Pelle Johansson for his help showing me how to do this.
93959 (cherry picked from commit ef3498e92d13c82633fdbe8120396bfbe1e7489a)
93960
93961commit 3570ca1cb76f084272f7d8bd22c4cfdf03201a77
93962Author: Jeremy Huddleston <jeremy@tifa.local>
93963Date: Sun Feb 10 18:27:52 2008 -0800
93964
93965 XQuartz: Added option for setting quartz-wm click-through preference
93966 (cherry picked from commit bf54c267cba97b2b3b9a621da0575776a388b2cb)
93967
93968commit 8004e160fa8cc75a3f1b7385fee64e5864b3b50a
93969Author: Jeremy Huddleston <jeremy@tifa.local>
93970Date: Sun Feb 10 16:34:40 2008 -0800
93971
93972 XQuartz: Converted interface to nib 3.x format
93973 (cherry picked from commit 656aaab95773bd70fc3504b68bf7e1d292891d93)
93974
93975commit 0b0a09797302ac2171db5df20fc5110aafc8efbb
93976Author: Peter Hutterer <peter@cs.unisa.edu.au>
93977Date: Thu Feb 7 15:48:04 2008 +1030
93978
93979 xkb: when copying the keymap, make sure the structs default to 0/NULL.
93980
93981 It actually does help if a pointer is NULL rather than pointing to nirvana
93982 when you're trying to free it lateron. Who would have thought?
93983 (cherry picked from commit 7a97ca667405a42d008265c3a870210cc1da97dd)
93984
93985commit 7018f280406eb0ef899a4046de274cfdd582881b
93986Author: Peter Hutterer <peter@cs.unisa.edu.au>
93987Date: Thu Feb 7 15:48:04 2008 +1030
93988
93989 xkb: when copying the keymap, make sure the structs default to 0/NULL.
93990
93991 It actually does help if a pointer is NULL rather than pointing to nirvana
93992 when you're trying to free it lateron. Who would have thought?
93993
93994commit 66f8001b61d12eaf4905ac71ccbb3f304914d00d
93995Author: Bart Trojanowski <bart@symbio-technologies.com>
93996Date: Thu Feb 7 21:26:54 2008 -0500
93997
93998 X86EMU: handle CPUID instruction
93999
94000 After trying to switch from X to VT (or just quit) the video-amd driver
94001 attempts to issue INT 10/0 to go to mode 3 (VGA). The emulator, running
94002 the BIOS code, would then spit out:
94003
94004 c000:0282: A2 ILLEGAL EXTENDED X86 OPCODE!
94005
94006 The opcode was 0F A2, or CPUID; it was not implemented in the emulator.
94007 This simple patch, against 1.3.0.0, handles the CPUID instruction in one of
94008 two ways:
94009 1) if ran on __i386__ or __x86_64__ then it calls the CPUID instruction
94010 directly.
94011 2) if ran elsewhere it returns a canned 486dx4 set of values for
94012 function 1.
94013
94014 This fix allows the video-amd driver to switch back to console mode,
94015 with the GSW BIOS.
94016
94017 Thanks to Symbio Technologies for funding my work, and ThinCan for
94018 providing hardware :)
94019
94020 Signed-off-by: Bart Trojanowski <bart@jukie.net>
94021 Acked-by: Eric Anholt <eric@anholt.net>
94022
94023commit de16a8c53046764dbdf26a87acc5c984ef00d818
94024Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94025Date: Thu Feb 7 20:14:16 2008 -0500
94026
94027 XACE: Correct some protocol error values in the colormap routines.
94028
94029commit 31934132a490b1b8ae73010c44e0b23217d8dab2
94030Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94031Date: Thu Feb 7 16:32:06 2008 -0500
94032
94033 xselinux: Use the device name in debugging output.
94034
94035commit 6dcb7d732bfeadc214228d68c5a13eef30248eb1
94036Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94037Date: Thu Feb 7 16:00:52 2008 -0500
94038
94039 xselinux: Split devPrivate state into subject and object records.
94040
94041commit 2259b144f0fd4855085a656111a0c64246733e78
94042Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94043Date: Thu Feb 7 14:35:02 2008 -0500
94044
94045 xselinux: Add getattr and setattr to the permission map for properties.
94046
94047commit 9d5edebe9664f4ac35b82830e000a72d5ca9cbe4
94048Author: Peter Hutterer <peter@cs.unisa.edu.au>
94049Date: Thu Feb 7 21:22:50 2008 +1030
94050
94051 Xi: finish XKB initialisation before copying the keymap.
94052
94053commit 96eafa3d4f05997dac8be6a8d66a92410511824c
94054Author: Peter Hutterer <peter@cs.unisa.edu.au>
94055Date: Tue Feb 5 16:01:56 2008 +1030
94056
94057 xkb: when copying sections, make sure num_rows is set too.
94058 (cherry picked from commit 41991fb991313202e8e6b513fe928ba14f8fcb87)
94059
94060commit 5c30327275509576b7848a5f842e7a1bffabe980
94061Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94062Date: Tue Feb 5 21:06:05 2008 -0500
94063
94064 XACE: Push the dix "structure" includes down to the security modules.
94065
94066commit bb1a577a6822f781f1e38d2434a13914e74f89aa
94067Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94068Date: Tue Feb 5 20:07:08 2008 -0500
94069
94070 XACE: Move the property access hook to its own function.
94071
94072commit 019ad5acd20e34dc2aa3b89cc426138db5164c48
94073Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94074Date: Tue Feb 5 15:44:41 2008 -0500
94075
94076 XFixes: squash a pointer/integer size mismatch warning.
94077
94078commit d3c36fe721edc55636438bc3e0e7a6c03f62784e
94079Author: liuhong <liuhong@devlinux-hong.sh.intel.com>
94080Date: Tue Feb 5 10:54:58 2008 +0800
94081
94082 validate mode clock for probed modes
94083
94084 Some modes claimed in monitor EDID data may not be supported by
94085 the monitor. So also validating the max clock for probed modes.
94086
94087commit 4b5b6e7baab58072a983d2ec136965f404c3a74a
94088Author: liuhong <liuhong@devlinux-hong.sh.intel.com>
94089Date: Tue Feb 5 10:54:10 2008 +0800
94090
94091 fix max clock unit
94092
94093 max clock from EDID data is in MHz, while we need KHz to validate modes.
94094
94095commit a56ef7aaa4b6ac13c8181f68fc7dad3ca89e6973
94096Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
94097Date: Sat Feb 2 23:04:46 2008 +0100
94098
94099 dix: Move motion history update until after screen crossing and clipping
94100
94101 Cross screen and clip the coordinates before updating the motion history
94102 so that it will have the same contents as the events that are reported.
94103
94104commit a0284d577aabea8406b72dd63773e341430ebe56
94105Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
94106Date: Sat Feb 2 23:03:51 2008 +0100
94107
94108 dix: Skip call to clipAxis for relative core-events
94109
94110 Relative events that generates both core and extention
94111 events will have its axis cliped and screen changed by
94112 miPointerSetPosition when the events are processed. For
94113 absolute and non core-generating relative events the
94114 axis must be clipped if we shouldn't end up completely
94115 outside the defined ranges (if any).
94116
94117commit d9e23c4ff1607a62164b34717ef9afd352ce2b94
94118Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
94119Date: Sat Feb 2 22:57:32 2008 +0100
94120
94121 Bug # 10324: dix: Add scaling of X and Y on the reported pointer-events
94122
94123 Restore the rescaling code for x and y axis when generating
94124 motion events.
94125
94126commit f04c0838699f1a733735838e74cfbb1677b15dc4
94127Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
94128Date: Sat Feb 2 22:45:31 2008 +0100
94129
94130 Bug # 10324: dix: Allow arbitrary value ranges in GetPointerEvents
94131
94132 Don't use a possitive value as a marker for if a max-value
94133 is defined on the valuators. Use the existence of a valid
94134 value range instead. This will also make it possible to
94135 define arbitrary start and end-values for min and max as
94136 long as min < max.
94137
94138commit 12e532403210c15a25200ef448bfe9701735ab20
94139Author: Magnus Vigerlöf <Magnus.Vigerlof@ipbo.se>
94140Date: Sat Feb 2 22:44:31 2008 +0100
94141
94142 dix: Always add valuator information if present
94143
94144 Send valuator information for all event types, not only for
94145 MotionEvents and absolute button events.
94146
94147commit 41991fb991313202e8e6b513fe928ba14f8fcb87
94148Author: Peter Hutterer <peter@cs.unisa.edu.au>
94149Date: Tue Feb 5 16:01:56 2008 +1030
94150
94151 xkb: when copying sections, make sure num_rows is set too.
94152
94153commit 521a7f26e088029707fb9a2bb80c9ddc734a3f8b
94154Author: Peter Hutterer <peter@cs.unisa.edu.au>
94155Date: Tue Feb 5 18:34:31 2008 +1030
94156
94157 mi: Only UpdateSpriteForScreen if we actually changed the screen. (Bug #12650)
94158
94159 X.Org Bug 12650 <http://bugs.freedesktop.org/show_bug.cgi?id=12650>
94160
94161commit 10617dc0fb7166ccd5b2e92fa708390c2d7d0d27
94162Author: Peter Hutterer <peter@cs.unisa.edu.au>
94163Date: Fri Feb 1 14:24:04 2008 +1030
94164
94165 xfree86: stick two more checks in for num_valuators < MAX_VALUATORS
94166
94167commit c68f063be639f39c2facbb496e8455e8e3771b41
94168Author: Matthieu Herrb <matthieu.herrb@laas.fr>
94169Date: Fri Feb 1 14:13:29 2008 +1030
94170
94171 xfree86: don't call xalloc from signal handlers when posting events.
94172
94173 Reviewed-by: Peter Hutterer <peter@cs.unisa.edu.au>
94174
94175commit e5e531ce96264518f278f572a4d0bfcd345aa618
94176Merge: 6c38682 1692dcf
94177Author: Peter Hutterer <peter@cs.unisa.edu.au>
94178Date: Thu Jan 31 16:36:17 2008 +1030
94179
94180 Merge branch 'master' into mpx
94181
94182 Conflicts:
94183
94184 dix/events.c
94185
94186commit 6c38682084e24b18bac0ea75729b1183aeb01600
94187Author: Benjamin Tissoires <tissoire@cena.fr>
94188Date: Thu Jan 31 15:51:32 2008 +1030
94189
94190 kdrive: AddInputClient takes a ClientPtr.
94191
94192commit 1692dcf197470d074f69d5af1608cb2ff1d08872
94193Author: Peter Hutterer <peter@cs.unisa.edu.au>
94194Date: Wed Jan 30 13:04:58 2008 +1030
94195
94196 dix: print out event type if a bogus pointer event occurs.
94197
94198commit d954f9c80348de294602d931d387e5cd1ef4b9a5
94199Author: Peter Hutterer <peter@cs.unisa.edu.au>
94200Date: Wed Jan 30 10:39:54 2008 +1030
94201
94202 xkb: don't update LEDs if they don't exist. (Bug #13961)
94203
94204 In some weird cases we call this function when there is no SrvLedInfo on the
94205 device. And it turns out null-pointer dereferences are bad.
94206
94207 X.Org Bug 13961 <http://bugs.freedesktop.org/show_bug.cgi?id=13961>
94208
94209commit df325be394e1f75c396b2768f81373f2989aef7c
94210Author: Coleman Kane <cokane@cokane.org>
94211Date: Tue Jan 29 09:47:00 2008 -0800
94212
94213 Bug 13101: xorg-server has a typo in hw/xfree86/os-support/bsd/i386_video.c
94214
94215 X.Org Bug #13101 <https://bugs.freedesktop.org/show_bug.cgi?id=13101>
94216 Patch #12360 <https://bugs.freedesktop.org/attachment.cgi?id=12360>
94217
94218commit 94f412cb7e954fe872fed979057cbdfbef953c6f
94219Author: Julien Goodwin <julien@studio442.com.au>
94220Date: Sun Jan 27 12:30:16 2008 +1100
94221
94222 Loader: Fix verbosity confusion
94223
94224 'Loading foo' is verbosity 3, whereas 'already built-in' is verbosity 0.
94225 This means that gdm's log would just be full of bare 'module already
94226 built-in' messages.
94227
94228commit 442838fcb3bf07ac57553ae5600d9e6c59a559bb
94229Author: Julien Goodwin <julien@studio442.com.au>
94230Date: Sun Jan 27 12:27:26 2008 +1100
94231
94232 xorg.conf.man: Fix monitor/output confusion in monitor positioning
94233
94234 On the Intel driver at least, LeftOf/RightOf/Above/Below in xorg.conf refers
94235 to output names, not monitor names. Fix nomenclature in xorg.conf.man.
94236
94237commit 5b8641a5fdc112c19e78ca2954878712e328d403
94238Author: Peter Hutterer <peter@cs.unisa.edu.au>
94239Date: Tue Jan 29 10:01:37 2008 +1030
94240
94241 xfree86: fix AlwaysCore handling. (Bug #14256)
94242
94243 Assume AlwaysCore being set by default, just like the other options.
94244
94245 X.Org Bug 14256 <http://bugs.freedesktop.org/show_bug.cgi?id=14256>
94246
94247commit 975ab11799c819a81da1dfe83505194410dbcb95
94248Author: Peter Hutterer <peter@cs.unisa.edu.au>
94249Date: Sat Jan 26 09:39:54 2008 +1030
94250
94251 config: don't reset connection info on disconnect.
94252
94253 If dbus is restarted, we try to connect again and this is difficult if the
94254 busname and/or busobject is not set.
94255 (cherry picked from commit 210eeef495770c1883c842ff003c28ce25f279d4)
94256
94257commit 7dde5a694a06efed0a9186f05d33f5be6f5dba71
94258Author: Peter Hutterer <peter@cs.unisa.edu.au>
94259Date: Fri Jan 25 13:54:47 2008 +1030
94260
94261 config: check connection != NULL before getting dbus' dispatch status.
94262
94263 (cherry picked from commit d23266522390a4ef7203ae7c062b2b920e45f9d7)
94264
94265commit f0ba7707161b8866e6fde32d6f25be6afcdecb48
94266Author: Peter Hutterer <peter@cs.unisa.edu.au>
94267Date: Fri Jan 25 13:45:22 2008 +1030
94268
94269 config: only shutdown libhal if the connection is valid.
94270
94271 Thanks to libdbus' extensive use of assert we won't just get an error, it'll
94272 bring the whole server down for us.
94273 (cherry picked from commit fb07fab2c07e7b0834724541dc47bfba02ba8574)
94274
94275commit 2cb0ebec2b85d96289c23c17cfdcdf97ef6877d2
94276Author: Peter Hutterer <peter@cs.unisa.edu.au>
94277Date: Fri Jan 25 11:48:13 2008 +1030
94278
94279 config: add a debug message, fix a whitespace error.
94280
94281 (cherry picked from commit 7732898aaa70e076000f6e6aa9420482896ed996)
94282
94283commit 4fc2d3cef8d7868b025aa14af7ed4b730e8f2c49
94284Author: Jeremy Huddleston <jeremy@tifa.local>
94285Date: Mon Jan 28 12:18:43 2008 -0800
94286
94287 Rootless: RootlessEnsureFrame: Added check for !IsRoot
94288 This was causing an issue with Apple-DRI and was reported here:
94289 http://trac.macosforge.org/projects/xquartz/ticket/51
94290 (cherry picked from commit 116800279d2ec783c63f43d3902627edde6a4cff)
94291
94292commit 16b0614c8c441eb1692bde9e58fadf54a9c6f00f
94293Merge: 1a88aed e915a26
94294Author: David Nusinow <dnusinow@debian.org>
94295Date: Sat Jan 26 16:13:25 2008 -0500
94296
94297 Merge commit 'upstream/master'
94298
94299commit 0ac175597712edfeae676f536746a4d26d625e30
94300Merge: 210eeef e915a26
94301Author: Peter Hutterer <peter@cs.unisa.edu.au>
94302Date: Sat Jan 26 13:55:07 2008 +1030
94303
94304 Merge branch 'master' into mpx
94305
94306 Conflicts:
94307
94308 Xext/sampleEVI.c
94309
94310commit e915a2639752bc0ea9e6e192e020cc2031c08063
94311Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94312Date: Fri Jan 25 19:22:19 2008 -0500
94313
94314 xselinux: Move the extension to extmod instead of being built-in.
94315
94316commit 210eeef495770c1883c842ff003c28ce25f279d4
94317Author: Peter Hutterer <peter@cs.unisa.edu.au>
94318Date: Sat Jan 26 09:39:54 2008 +1030
94319
94320 config: don't reset connection info on disconnect.
94321
94322 If dbus is restarted, we try to connect again and this is difficult if the
94323 busname and/or busobject is not set.
94324
94325commit f6a78ee143e3a3ad69538adf2b9675d724468ffa
94326Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94327Date: Fri Jan 25 18:04:10 2008 -0500
94328
94329 XACE: Remove the extension code entirely, XACE is completely static now.
94330
94331commit 7724c30a751c653ca3e2e8a6752af27bc37de3f0
94332Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94333Date: Fri Jan 25 17:28:17 2008 -0500
94334
94335 XACE: Stop using fake requestVectors in favor of a simple hook call.
94336
94337commit f82329b0811469ddae5c44dcfffa38185c11a67c
94338Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94339Date: Fri Jan 25 16:20:46 2008 -0500
94340
94341 XACE: Don't need to actually register a protocol extension.
94342
94343commit d23266522390a4ef7203ae7c062b2b920e45f9d7
94344Author: Peter Hutterer <peter@cs.unisa.edu.au>
94345Date: Fri Jan 25 13:54:47 2008 +1030
94346
94347 config: check connection != NULL before getting dbus' dispatch status.
94348
94349commit fb07fab2c07e7b0834724541dc47bfba02ba8574
94350Author: Peter Hutterer <peter@cs.unisa.edu.au>
94351Date: Fri Jan 25 13:45:22 2008 +1030
94352
94353 config: only shutdown libhal if the connection is valid.
94354
94355 Thanks to libdbus' extensive use of assert we won't just get an error, it'll
94356 bring the whole server down for us.
94357
94358commit 7732898aaa70e076000f6e6aa9420482896ed996
94359Author: Peter Hutterer <peter@cs.unisa.edu.au>
94360Date: Fri Jan 25 11:48:13 2008 +1030
94361
94362 config: add a debug message, fix a whitespace error.
94363
94364commit 46794d0c9665f07913980830d038c88d00407612
94365Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94366Date: Thu Jan 24 19:49:13 2008 -0500
94367
94368 xselinux: Rename SelectionManager to more generic SecurityManager.
94369
94370commit 466f6552190d1b83d66ebfd71e07921ed6210ee2
94371Author: Peter Hutterer <peter@cs.unisa.edu.au>
94372Date: Wed Jan 23 16:11:31 2008 +1030
94373
94374 Xi: if deviceid is > 127 assume GE mask is for all devices (XiSelectEvent).
94375
94376commit 379f057646588fa3211d7ba0ec8d77d2fc0ad246
94377Author: Peter Hutterer <peter@cs.unisa.edu.au>
94378Date: Wed Jan 23 16:08:37 2008 +1030
94379
94380 Xext: add GEDeviceMaskIsSet (GE mask checking per device)
94381
94382 If a mask was set for the NULL device, then we pretend the mask is set for all
94383 devices.
94384
94385commit 91dc8c43d2343e494c59023484883f4683bdc639
94386Author: Peter Hutterer <peter@cs.unisa.edu.au>
94387Date: Wed Jan 23 15:45:02 2008 +1030
94388
94389 Xi: Change window access right check to RecieveAccess in XiSelectEvent.
94390
94391commit 94a6a65ae7414c8247aa1741ec5c7561e6f4bbe0
94392Author: Peter Hutterer <peter@cs.unisa.edu.au>
94393Date: Wed Jan 23 15:43:48 2008 +1030
94394
94395 Xext: allow per-device GE masks.
94396
94397 Instead of a single mask per client we allow the storage of a mask per client
94398 per device.
94399
94400commit 6ffeecabb7f3f3173864e0f0af21a99bdc5b5044
94401Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94402Date: Thu Jan 24 18:11:49 2008 -0500
94403
94404 xselinux: Use a privileged bit in the state instead of passing an index
94405 to the permission checking function.
94406
94407commit 7ba8e97cbabfef4d614a6a38314830ec0f925471
94408Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94409Date: Thu Jan 24 19:09:58 2008 -0500
94410
94411 xselinux: Implement "get context" protocol requests.
94412
94413commit f0bf9a5231d4f612ac916355118484d055715f32
94414Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94415Date: Thu Jan 24 19:02:35 2008 -0500
94416
94417 xselinux: Whitespace fixups.
94418
94419commit 734e115871ce98badb8800383c423493802ae3d3
94420Author: Hong Liu <hong.liu@intel.com>
94421Date: Wed Jan 23 21:04:32 2008 +0800
94422
94423 Bug #12439: add a quirk to use +hsync +vsync for the probed detailed mode.
94424
94425 Samsung 205BW quirk is somehow reworked.
94426
94427commit cc22b05ea06e08568d0f0abdaccf67bd32662e94
94428Author: Adam Jackson <ajax@redhat.com>
94429Date: Tue Jan 22 18:57:11 2008 -0500
94430
94431 There is no such thing as /dev/cpu/mtrr.
94432
94433commit 1a88aed5c82c7c131e3d473ef7b8766a418fdf1b
94434Author: David Nusinow <dnusinow@debian.org>
94435Date: Mon Jan 21 21:16:13 2008 -0500
94436
94437 Add tags/TAGS to .gitignore for ctags usage
94438
94439commit be6c17fcf9efebc0bbcc3d9a25f8c5a2450c2161
94440Author: Matthias Hopf <mhopf@suse.de>
94441Date: Mon Jan 21 16:13:21 2008 +0100
94442
94443 CVE-2007-6429: Always test for size+offset wrapping.
94444
94445commit ba315ae5ddd00763147a4b0cdcacba7691793dfc
94446Author: Peter Hutterer <peter@cs.unisa.edu.au>
94447Date: Mon Jan 21 23:43:53 2008 +1030
94448
94449 Xi: add XACE hooks for device creation (ChangeDeviceHierarchy)
94450
94451 AddInputDevice checks for permissions already, so all we do is modify a few
94452 callers to let AID sort it out.
94453
94454commit 94a21d757ce58254accbd5dd3a86810aadeec9f0
94455Author: Michel Dänzer <michel@tungstengraphics.com>
94456Date: Sat Jan 19 13:17:45 2008 +0100
94457
94458 AIGLX: Fix GLX_EXT_texture_from_pixmap fallback with EXA.
94459
94460 Use pScreen->GetImage to obtain the pixmap contents instead of dereferencing
94461 pPixmap->devPrivate.ptr directly.
94462
94463commit e9fa7c1c88a8130a48f772c92b186b8b777986b5
94464Author: Adam Jackson <ajax@redhat.com>
94465Date: Fri Jan 18 14:41:20 2008 -0500
94466
94467 CVE-2007-6429: Don't spuriously reject <8bpp shm pixmaps.
94468
94469 Move size validation after depth validation, and only validate size if
94470 the bpp of the pixmap format is > 8. If bpp < 8 then we're already
94471 protected from overflow by the width and height checks.
94472
94473commit 6492d513c0022d70f838d1b2561442c7a41e401f
94474Author: Peter Hutterer <peter@cs.unisa.edu.au>
94475Date: Fri Jan 18 21:44:22 2008 +1030
94476
94477 dix: if a client has a core grab, return the grab device on QueryPointer.
94478
94479 The correct thing would be to return the ClientPointer. However, if the client
94480 for some reason has a core grab on another device (e.g. as result of a passive
94481 grab), return the coordinates of the grabbed device instead.
94482
94483 This makes the use of nautilus a bit saner.
94484
94485commit cd2ad4c2ed58b49cee085174e77c242048159321
94486Author: Peter Hutterer <peter@cs.unisa.edu.au>
94487Date: Fri Jan 18 20:11:55 2008 +1030
94488
94489 dix: remove some superfluous lines in ProcChangeActivePointerGrab.
94490
94491commit 8a925d201efac52ef663ea3ff46acc5046178666
94492Author: Peter Hutterer <peter@cs.unisa.edu.au>
94493Date: Fri Jan 18 20:06:29 2008 +1030
94494
94495 dix: free the genericMasks when ungrabbing.
94496
94497 This finally plugs a memory hole created by grabs registering for generic
94498 events.
94499
94500commit 8949d81a3c79eef351d7c018c76692dcca74b63e
94501Author: Peter Hutterer <peter@cs.unisa.edu.au>
94502Date: Fri Jan 18 19:45:54 2008 +1030
94503
94504 include: mark CoreProcessPointerEvent, CoreProcessKeyboardEvent as deprecated.
94505
94506 ProcessOtherEvent does all the work now.
94507
94508commit d323fd64206ddb4221daa7cf7504daf62c5e7d90
94509Author: Peter Hutterer <peter@cs.unisa.edu.au>
94510Date: Fri Jan 18 19:17:42 2008 +1030
94511
94512 dix: remove inputInfo.pointer refernce in EventSelectForWindow
94513
94514 If the window being changed is set as the motion hint window for any device,
94515 the device's motion hint window is set to NULL. Which is kinda what the old
94516 code did, except that it did it with only the VCP.
94517
94518commit 883811c2bed9c6fcb706effbf83ddc97518b6f0c
94519Author: Peter Hutterer <peter@cs.unisa.edu.au>
94520Date: Fri Jan 18 18:48:46 2008 +1030
94521
94522 dix: remove inputInfo.pointer reference in TryClientEvents.
94523
94524 Unfortunately, this requires a change in the TCE API, and thus a change in all
94525 callers. Tough luck.
94526
94527commit 6e17184486c4309ec8b47c7cfd11fef60bb360ac
94528Author: Peter Hutterer <peter@cs.unisa.edu.au>
94529Date: Fri Jan 18 17:52:38 2008 +1030
94530
94531 Xi: Check for DestroyAccess when trying to delete a master device.
94532
94533commit 23f3f0e27dc90b7b3a375f2a5dd094e6f53552b5
94534Author: Jeremy Huddleston <jeremy@yuffie.local>
94535Date: Sun Jan 13 14:00:25 2008 -0800
94536
94537 XQuartz: Moved SetFrontProcess haco to set_front_process
94538 So it is done by the other thread...
94539 (cherry picked from commit 7429379eb1001ee3dc769daa8fe6b3aef1b9cc8a)
94540
94541commit 8e133d96740d010a4fd969a8188e6e71fb2cafe2
94542Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
94543Date: Thu Jan 17 15:29:06 2008 +0100
94544
94545 Fix for CVE-2008-0006 - PCF Font parser buffer overflow.
94546
94547commit 6de61f82728df22ea01f9659df6581b87f33f11d
94548Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
94549Date: Thu Jan 17 15:28:42 2008 +0100
94550
94551 Fix for CVE-2007-6429 - MIT-SHM and EVI extensions integer overflows.
94552
94553commit 7dc1717ff0f96b99271a912b8948dfce5164d5ad
94554Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
94555Date: Thu Jan 17 15:28:03 2008 +0100
94556
94557 Fix for CVE-2007-6428 - TOG-cup extension memory corruption.
94558
94559commit dd5e0f5cd5f3a87fee86d99c073ffa7cf89b0a27
94560Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
94561Date: Thu Jan 17 15:27:34 2008 +0100
94562
94563 Fix for CVE-2007-6427 - Xinput extension memory corruption.
94564
94565commit bbde5b62a137ba726a747b838d81e92d72c1b42b
94566Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
94567Date: Thu Jan 17 15:26:41 2008 +0100
94568
94569 Fix for CVE-2007-5760 - XFree86 Misc extension out of bounds array index
94570
94571commit e85130c85f727466fc27be1cfa46c88b257499fb
94572Author: Daniel Stone <daniel@fooishbar.org>
94573Date: Sat Jan 5 10:47:39 2008 +0200
94574
94575 Xephyr: One-time keyboard leak fix
94576
94577 Don't leak the originally-allocated keysym map.
94578
94579commit 0137b0394a248f694448a7d97c9a1a3efcf24e81
94580Author: Daniel Stone <daniel@fooishbar.org>
94581Date: Sat Jan 5 10:43:53 2008 +0200
94582
94583 XKB: XkbCopyKeymap: Don't leak all the sections
94584
94585 Previously, we'd just keep num_sections at 0, which would break the
94586 geometry and lead us to leak sections. Don't do that.
94587
94588commit b99a43dfe97c1813e1c61f298b1c83c5d5ca88a2
94589Author: Daniel Stone <daniel@fooishbar.org>
94590Date: Sat Jan 5 10:38:16 2008 +0200
94591
94592 OS: IO: Zero out client buffers
94593
94594 For alignment reasons, we can write out uninitialised bytes, so allocate
94595 the whole thing with xcalloc.
94596
94597commit a6a7fadbb03ee99312dfb15ac478ab3c414c1c0b
94598Author: Kristian Høgsberg <krh@redhat.com>
94599Date: Wed Jan 16 20:24:11 2008 -0500
94600
94601 Don't break grab and focus state for a window when redirecting it.
94602
94603 Composite uses an unmap/map cycle to trigger backing pixmap allocation
94604 and cliprect recomputation when a window is redirected or unredirected.
94605 To avoid protocol visible side effects, map and unmap events are
94606 disabled temporarily. However, when a window is unmapped it is also
94607 removed from grabs and loses focus, but these state changes are not
94608 disabled.
94609
94610 This change supresses the unmap side effects during the composite
94611 unmap/map cycle and fixes this bug:
94612
94613 http://bugzilla.gnome.org/show_bug.cgi?id=488264
94614
94615 where compiz would cause gnome-screensaver to lose its grab when
94616 compiz unredirects the fullscreen lock window.
94617
94618commit e46f6ddeccd082b2d507a1e8b57ea30e6b0a2c83
94619Author: Michel Dänzer <michel@tungstengraphics.com>
94620Date: Wed Jan 16 14:24:22 2008 +0100
94621
94622 Yet another Xv extension byte swapping fix.
94623
94624commit b0bf4308acb706abc87c51658c2251fa86231c35
94625Author: Peter Hutterer <peter@cs.unisa.edu.au>
94626Date: Tue Jan 15 19:15:09 2008 +1030
94627
94628 dix: print out an error when core type can't be converted.
94629
94630 Helps a bit in tracking down bugs.
94631
94632commit 2a988ed75b7634d7cb6e83bb7aa89bc8768d3a58
94633Author: Peter Hutterer <peter@cs.unisa.edu.au>
94634Date: Tue Jan 15 19:14:46 2008 +1030
94635
94636 xkb: don't do core key repeats, XI only is the path to light.
94637
94638commit 4fd0885978be286a68b689824f1d910f929a52b0
94639Author: Peter Hutterer <peter@cs.unisa.edu.au>
94640Date: Tue Jan 15 18:58:42 2008 +1030
94641
94642 Xi: remove some leftovers from ProcessOtherEvents.
94643
94644 Core handling not needed anymore here.
94645
94646commit 32aa252e988be8cbfd4f7e373fb7b7736ef1f5f2
94647Author: Peter Hutterer <peter@cs.unisa.edu.au>
94648Date: Tue Jan 15 15:22:39 2008 +1030
94649
94650 dix: Process an input event as a single event, instead of two separate ones.
94651
94652 This is a significant shift in how input events are perceived. The common
94653 approach was to treat a core event as a different entity than the XI event.
94654 This could result in the XI event being delivered to a different client than
94655 the core event. This doesn't work nicely if they come from the same device.
94656
94657 Instead, we treat an input event as a single event, that is delivered through
94658 two separate APIs. So when delivering an event, we first try the XI event,
94659 then the core event. If the window want's neither, we go to the parent and
94660 repeat. Once either core or XI has been delivered, the processing stops.
94661
94662 Important: Different to the previous method, if a client registers for core
94663 button events, the parent window will not get XI events. This should only
94664 cause problems when you're mixing core and XI events, so don't do that!
94665
94666 Generic events don't fit into this yet, they cause me headaches.
94667
94668commit 7a0d16ef0a103bcb25fa8a20322685f017aaf5a3
94669Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
94670Date: Tue Jan 15 03:27:16 2008 -0200
94671
94672 Removed some warnings.
94673
94674commit 315d6a2b1d2a3de308e98d548afe780c59a784fc
94675Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
94676Date: Tue Jan 15 02:59:56 2008 -0200
94677
94678 Fix Xephyr compilation without GLX.
94679
94680commit 0969a9f7497e10794a6534321c10a0e1ac680ad7
94681Author: Peter Hutterer <peter@cs.unisa.edu.au>
94682Date: Tue Jan 15 11:31:12 2008 +1030
94683
94684 dix: Emulate core events within CheckDeviceGrabs and ComputeFreezes.
94685
94686 This should restore the correct passive grab processing. When checking for
94687 passive grabs, the core event is emulated and we check first for XI grabs on
94688 the window, then for core grabs. Regardless of which event activates the grab,
94689 the XI event is stored in the device's EQ.
94690
94691 When replaying the event, we take the XI event and replay it on the next
94692 window, again including the emulation of the core event.
94693
94694commit 1f83f40525acd3aff8f50b3c519bc1f307ff1e19
94695Author: Dave Airlie <airlied@linux.ie>
94696Date: Tue Jan 15 10:20:50 2008 +1000
94697
94698 xf86Cursors: fix memset for non-square cursors
94699
94700commit e6ea3147bfb686798dac381eb8900f9f18beb88e
94701Author: Bernardo Innocenti <bernie@codewiz.org>
94702Date: Sun Jan 13 19:50:37 2008 -0500
94703
94704 exa: make the prototype for exaGetPixmapFirstPixel() public
94705
94706 This fixes a warning in amd_drv which is using it.
94707
94708 Signed-off-by: Bernardo Innocenti <bernie@codewiz.org>
94709
94710commit a83a0c5a144da67dab96a857b849a5692b73245d
94711Author: Peter Hutterer <peter@cs.unisa.edu.au>
94712Date: Sun Jan 13 16:14:29 2008 +1030
94713
94714 dix: remove obsolete comment.
94715
94716 Pairings don't exist anymore and the documented issue is a non-issue now.
94717
94718commit 180a5aba4de3104fed8bc4e7d42a1e3a51575318
94719Author: Jeremy Huddleston <jeremy@yuffie.local>
94720Date: Sat Jan 12 21:24:34 2008 -0800
94721
94722 XQuartz: Fixed copy-paste error with login_shell commit
94723 (cherry picked from commit 6deec3acc6f8010b5b53a1e55a0a2c4080ba69d2)
94724
94725commit 6fd4a5e2e4d0be0ba0773df831687e11e1262c72
94726Author: Jeremy Huddleston <jeremy@yuffie.local>
94727Date: Sat Jan 12 11:56:00 2008 -0800
94728
94729 XQuartz: Corrected copyright X.org Project -> X.org Foundation
94730 (cherry picked from commit f21631444816fc12b8a534c2cf79e6ac6c2af7c9)
94731
94732commit f72255639c065d795f7767683e851b1b5b2d9480
94733Author: Jeremy Huddleston <jeremy@yuffie.local>
94734Date: Sat Jan 12 11:35:48 2008 -0800
94735
94736 XQuartz: added 'login_shell' option to defaults
94737 so the user can choose something other than /bin/sh
94738 (cherry picked from commit b549cf18cebd3435d70f62855239484974c455a1)
94739
94740commit ec24a6b5aa732ec6999a27889d9a33cf80123886
94741Author: Jeremy Huddleston <jeremy@yuffie.local>
94742Date: Sun Jan 6 18:29:54 2008 -0800
94743
94744 XQuartz: Fixed switching into XQuartz via expose.
94745 (cherry picked from commit 627ed60ce5d7499761028edf379ebd95250d3e04)
94746
94747commit 6844bd2e63490870bab3c469eec6030354ef2865
94748Author: Alan Coopersmith <alan.coopersmith@sun.com>
94749Date: Wed Jan 9 19:52:00 2008 -0800
94750
94751 More Xv extension byte swapping fixes
94752
94753commit 38bf01bd1c925614a6e67a38aa3cefc7b8fe3bca
94754Merge: 4e85c7c 0883e83
94755Author: Peter Hutterer <peter@cs.unisa.edu.au>
94756Date: Thu Jan 10 13:38:46 2008 +1030
94757
94758 Merge branch 'master' into mpx
94759
94760commit 0883e838e25227f0af84d2a90979175724166d16
94761Author: Peter Harris <peter.harris@hummingbird.com>
94762Date: Wed Jan 9 14:52:33 2008 -0500
94763
94764 xf86misc.c: Avoid use of swapped values
94765
94766commit 59a63d72a1407a8aaf9878eeff7ee7a66f65a42b
94767Author: Alan Coopersmith <alan.coopersmith@sun.com>
94768Date: Wed Jan 9 18:42:58 2008 -0800
94769
94770 Fix reference to old versions of XFree86 to not say "Xorg 4.2 or earlier"
94771
94772commit 4e85c7c322faf14c14e4229fa294b8e3d3a4d304
94773Author: Peter Hutterer <peter@cs.unisa.edu.au>
94774Date: Wed Jan 9 17:36:39 2008 +1030
94775
94776 Xi: keep a counter of buttons down to avoid duplicate press/release events.
94777
94778 If two devices are attached to the same master device, pressing button 1 on
94779 each of them leads to two button presses from the same device. Some apps
94780 really don't like that.
94781
94782 So we just put a counter in place and only send the first press and the last
94783 release.
94784
94785commit e070c2cbef0d6fbbafce8d417e8b29231c2fdc50
94786Author: Peter Hutterer <peter@cs.unisa.edu.au>
94787Date: Fri Dec 7 20:18:49 2007 +1030
94788
94789 mi: change infamous "Tossed event ..." error for something more explanatory.
94790
94791 Few whitespace errors fixed.
94792 Two ErrorF's prefixed with [mi].
94793 (cherry picked from commit 117458d2db49efd3f04432ff45871b44c7d4ad57)
94794
94795commit c31aead0fe9fe424120ce221cd9f76cb6d29a5b5
94796Author: Richard Purdie <rpurdie@openedhand.com>
94797Date: Tue Jan 8 12:12:06 2008 +0100
94798
94799 [Kdrive] make XCalibrate 'orientation aware'
94800
94801commit 30375cd6d1439a3390b41714fe116aecc94743ca
94802Author: David Nusinow <dnusinow@debian.org>
94803Date: Mon Jan 7 20:57:30 2008 -0500
94804
94805 Don't log DPMS enabling as being from xorg.conf if it's not
94806
94807commit 981bb9f1e33e4564d1b59c00c808cc43a2e9497b
94808Author: Peter Hutterer <peter@cs.unisa.edu.au>
94809Date: Tue Dec 18 13:57:07 2007 +1030
94810
94811 dix: set the correct number of valuators in valuator events.
94812
94813 (first_valuator + num_valuators) must never be larger than the number of axes,
94814 otherwise DIX freaks out. And from looking at libXI, anything larger than 6 is
94815 wrong too.
94816 (cherry picked from commit 9f6ae61ad12cc2813d04405458e1ca5aed8a539e)
94817
94818commit 20ace6321ac464d821c67a82c7023f74ae038176
94819Author: Peter Hutterer <peter@cs.unisa.edu.au>
94820Date: Mon Jan 7 20:08:56 2008 +1030
94821
94822 dix: don't try to confine cursor on screen restructure if there is no cursor.
94823
94824 Stops segfaulting when using xrandr. Yay.
94825
94826commit 57a491e6d3ebec464c0c861b02c9ddcb1cc62c3c
94827Author: Peter Hutterer <peter@cs.unisa.edu.au>
94828Date: Mon Jan 7 20:06:32 2008 +1030
94829
94830 dix: init GrabPrivateKey to correct value.
94831
94832 Merge detritus from last pull.
94833
94834commit 260505e3c5a18044e97d31ea3bcc0955e46335c8
94835Author: David Nusinow <dnusinow@debian.org>
94836Date: Sun Jan 6 16:37:13 2008 -0500
94837
94838 Log enabling of DPMS even when it's not set in xorg.conf
94839
94840commit 7e7622165940934e56ae96ae785a8f88eec1a5cf
94841Author: Julien Cristau <jcristau@debian.org>
94842Date: Sun Jan 6 18:23:09 2008 +0100
94843
94844 Fix the name of the XFree86-Misc extension in the xorg.conf manpage.
94845
94846commit 59df687835c68eda147de47edfe9bc415c0efb4f
94847Author: Julien Cristau <jcristau@debian.org>
94848Date: Sun Jan 6 16:57:45 2008 +0100
94849
94850 Document the AllowEmptyInput, AutoAddDevices and AutoEnableDevices flags
94851
94852 Add documentation for the new AllowEmptyInput, AutoAddDevices and
94853 AutoEnableDevices server flags in the xorg.conf manpage.
94854
94855commit 7d226d6a251cb90765be2b50a1973986c5b7605b
94856Author: Jeremy Huddleston <jeremy@yuffie.local>
94857Date: Sat Jan 5 03:14:07 2008 -0800
94858
94859 XQuartz: Cleanup for strict-prototyping
94860 Also fixed DarwinEQEnqueue to match changes to the callback
94861 And also use dpmsstubs.c rather than copying the code into darwin.c
94862 (cherry picked from commit 4c5c30a4beb7a427b00b18097f548876ad3c11d7)
94863
94864commit 11967dce11cd953d123d53bb3389aa257c5158e8
94865Author: Jeremy Huddleston <jeremy@yuffie.local>
94866Date: Fri Jan 4 22:54:26 2008 -0800
94867
94868 XQuartz: Fixed copyright in About box for 2008. Happy New Year!
94869
94870commit aca75f389e2b08096c3cacec03b12a58075cf12c
94871Author: Jeremy Huddleston <jeremy@yuffie.local>
94872Date: Fri Jan 4 12:23:09 2008 -0800
94873
94874 XQuartz: Flush the debug log for easier debugging
94875 Also cleaned up formatting in xpr's eventHandler
94876 (cherry picked from commit 16861d6d4239c7f3918332ef07752f1e211afb23)
94877
94878commit 57468a696e1259c1e1c185fc60230e1d195defb7
94879Author: Alan Hourihane <alanh@tungstengraphics.com>
94880Date: Fri Jan 4 12:37:55 2008 +0000
94881
94882 Fix Line drawing with CapNotLast set in PolySegment.
94883
94884commit 39cb782f28be4efb2621fd8c614f2367eb834412
94885Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94886Date: Thu Jan 3 23:16:06 2008 -0500
94887
94888 XACE: DeleteCallbackList zeroes out its argument so don't do it twice.
94889
94890commit 7f376f23db463a65176de632ca6094acb55db951
94891Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94892Date: Thu Jan 3 23:08:49 2008 -0500
94893
94894 devPrivates rework: Free callback lists after use.
94895
94896commit de18703d2a25999e391d11b4c82ee018fb87372d
94897Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94898Date: Thu Jan 3 23:07:24 2008 -0500
94899
94900 dix: Fix bug+leak in callback manager DeleteCallbackList function.
94901
94902commit 20eb26f9d149993ae360a2cbd1b536b68c9f4069
94903Author: Alex Deucher <alex@samba.(none)>
94904Date: Thu Jan 3 22:53:36 2008 -0500
94905
94906 Fix potential crasher in xf86CrtcRotate()
94907
94908 xf86CrtcRotate() is called by randr 1.2 drivers via xf86CrtcSetMode() or xf86SetDesiredModes()
94909 during ScreenInit() at which point pScrn->pScreen is not set. If a user specifies a rotation
94910 in their config file pScrn->pScreen is dereferenced and boom.
94911
94912commit 249b9b30c11e03bb113b37cf4712bf88debf6f05
94913Author: Peter Hutterer <peter@cs.unisa.edu.au>
94914Date: Fri Jan 4 13:49:48 2008 +1030
94915
94916 Xi: remove file for the obsolete ChangePointerKeyboardPairing request handling.
94917
94918commit e4a214e40d35aa957c83b86b4e4abc86c22fbde0
94919Author: Peter Hutterer <peter@cs.unisa.edu.au>
94920Date: Fri Jan 4 13:32:53 2008 +1030
94921
94922 dix: don't free device's devPrivates manually, dixFreePrivates does it.
94923
94924 Merge detritus from last pull.
94925
94926commit cd0603c2dc5ee000ebce66056bc1a72f99bfb617
94927Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94928Date: Thu Jan 3 21:41:02 2008 -0500
94929
94930 Bug #13765: Heap corruption in XC-SECURITY extension code.
94931
94932commit e46d559739e020dc7f6fcbdc6d1fb39c57aab4b1
94933Author: Jeremy Huddleston <jeremy@yuffie.local>
94934Date: Sat Dec 22 15:09:12 2007 -0800
94935
94936 XQuartz: Cleaned up color map configuration.
94937 8 bit color still doesn't work, but the -depth command line argument now works properly.
94938 (cherry picked from commit 6765949c27c053d22882f54337cfd09203aa5383)
94939
94940commit 17a9714a6789a389d52dbb40fd1eed1e24c04d64
94941Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
94942Date: Thu Jan 3 14:46:54 2008 -0500
94943
94944 Bug #13794: Update MBE extension devPrivates to new interface.
94945
94946commit b2da44c76d68a76f20e90bccb268ebf65e132b49
94947Author: Peter Hutterer <peter@cs.unisa.edu.au>
94948Date: Thu Jan 3 18:51:44 2008 +1030
94949
94950 dix: DoChangeKeyboardControl shouldn't be using inputInfo.keyboard.
94951
94952commit 9cc41b88dfebb3f1e0a7cb5c224146e651a2cda4
94953Author: Peter Hutterer <peter@cs.unisa.edu.au>
94954Date: Thu Jan 3 18:47:28 2008 +1030
94955
94956 Xi: ProcXSetClientPointer should directly check for the client from the id.
94957
94958 The window ID includes the client ID anyway, so we might as well just look up
94959 the client directly instead of trying to get the window first and the client
94960 from the window.
94961
94962 This also fixes a possible issue with XACE. If the client had permission to
94963 write on the client but not on the window, the previous approach would have
94964 failed.
94965
94966commit 521fdc28d86d091495da3558d26ab4a938250ffe
94967Author: Peter Hutterer <peter@cs.unisa.edu.au>
94968Date: Thu Jan 3 18:43:18 2008 +1030
94969
94970 xfixes: switch a few inputInfo.pointer to PickPointer
94971
94972commit 37194b13554f0d36343cf73324b128b15bb6a338
94973Author: Peter Hutterer <peter@cs.unisa.edu.au>
94974Date: Thu Jan 3 18:36:33 2008 +1030
94975
94976 Xi: switch to using "rc" (instead of "err") as returncode.
94977
94978 All the rest of XI uses rc and returns rc in case of error, so make
94979 mpx-related stuff comply. This stops the rest of XI sending the error
94980 manually.
94981 This is just a cosmetic change to be in line with the rest.
94982
94983commit c7e9b67c547c97b2c943eab555b4fe2844e8a91f
94984Author: Peter Hutterer <peter@cs.unisa.edu.au>
94985Date: Thu Jan 3 18:09:56 2008 +1030
94986
94987 dix: free ClassesRec when a BadAlloc occurs during CoreKeyboardProc.
94988
94989commit 8da83836b60f7cdb75d08482f4311fa0e2ab4e1d
94990Merge: eace889 ae869fc
94991Author: Peter Hutterer <peter@cs.unisa.edu.au>
94992Date: Thu Jan 3 17:04:54 2008 +1030
94993
94994 Merge branch 'master' into mpx
94995
94996 Conflicts:
94997
94998 XTrap/xtrapddmi.c
94999 Xext/security.c
95000 Xext/xprint.c
95001 Xext/xtest.c
95002 Xext/xvdisp.c
95003 Xi/exevents.c
95004 Xi/grabdevb.c
95005 Xi/grabdevk.c
95006 Xi/opendev.c
95007 Xi/ungrdev.c
95008 Xi/ungrdevb.c
95009 Xi/ungrdevk.c
95010 dix/cursor.c
95011 dix/devices.c
95012 dix/dixutils.c
95013 dix/events.c
95014 dix/getevents.c
95015 dix/main.c
95016 dix/window.c
95017 hw/xfree86/ramdac/xf86Cursor.c
95018 include/dix.h
95019 include/input.h
95020 include/inputstr.h
95021 mi/midispcur.c
95022 mi/miinitext.c
95023 mi/misprite.c
95024 render/animcur.c
95025 xfixes/cursor.c
95026 xkb/xkbAccessX.c
95027
95028commit ccf6636d2ca8acdaaeb8da34db507a10a082b0de
95029Author: Alan Coopersmith <alan.coopersmith@sun.com>
95030Date: Wed Jan 2 19:28:33 2008 -0800
95031
95032 Kill xf86getpagesize even harder (dummylib & ioport)
95033
95034commit dfd682b582636a36345144bcf835e3ee46718d90
95035Author: Alan Coopersmith <alan.coopersmith@sun.com>
95036Date: Wed Jan 2 19:27:22 2008 -0800
95037
95038 X.Org bug 4947/Sun bug 6646626: Xv extension not byte-swapping properly
95039
95040 X.Org Bugzilla #4947 <https://bugs.freedesktop.org/show_bug.cgi?id=4947>
95041 Sun bug 6646626 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6646626>
95042
95043 Don't use swapped data after swapping it. When done swapping data,
95044 send the swapped data, not the address of the pointer to it, to the client.
95045
95046commit f6666dcc3b1ac60f850ea53c357a9ef61672a52a
95047Author: Alan Coopersmith <alan.coopersmith@sun.com>
95048Date: Wed Jan 2 19:19:55 2008 -0800
95049
95050 Add dixAllocatePrivate stub to dummylib for utils
95051
95052 Normally not necessary, except when building non-optimized/debug causes
95053 the inline functions from private.h to appear in os-support/libxorgos.la
95054
95055commit 73f422996016107d5f53492e4197bb05ed9c4bb9
95056Author: Alan Coopersmith <alan.coopersmith@sun.com>
95057Date: Wed Jan 2 19:17:54 2008 -0800
95058
95059 Fix names/types of new vuidMouse{Get,Set}ScreenPrivates macros
95060
95061commit 895073f6b41d9313cfe748232c492c5e9f76b443
95062Author: Alan Coopersmith <alan.coopersmith@sun.com>
95063Date: Wed Jan 2 18:09:26 2008 -0800
95064
95065 Restore include & typedef needed by dtrace
95066
95067 996b621bec1bbc4fb21970c75eaec62053bc6ccb deleted a couple lines too many
95068
95069commit 306fde4082044dfecbedd9af41e660bafb3ce438
95070Author: Brian <brian@i915.localnet.net>
95071Date: Tue Jan 1 09:27:44 2008 -0700
95072
95073 regenerated to add framebuffer object tokens (bug 13800)
95074
95075commit 5c362c2eb2cfdf1f6d667a3e64a0a7bc4942c950
95076Author: Brian <brian@i915.localnet.net>
95077Date: Tue Jan 1 09:07:48 2008 -0700
95078
95079 regenerated, adds GL_MAX_3D_TEXTURE_SIZE (see bug 13811)
95080
95081commit 85365ddf16e2b954d8249b380df53337420ed684
95082Author: Adam Jackson <ajax@redhat.com>
95083Date: Fri Dec 28 17:35:54 2007 -0500
95084
95085 EDID 1.4: If given a native pixel format, use it when inferring virtual.
95086
95087commit 312b30cb03e439644ea10e08fa93268116333f0d
95088Author: Adam Jackson <ajax@redhat.com>
95089Date: Fri Dec 28 17:18:30 2007 -0500
95090
95091 EDID 1.4: First detailed mode is always preferred.
95092
95093 ... so act like it in the modelist generator, not just the parser.
95094
95095commit 9dbb73033ae60e7ab85f1469a696e2a52f0cb0fe
95096Author: Colin Harrison <colin.harrison@virgin.net>
95097Date: Sat Dec 29 00:02:16 2007 +0200
95098
95099 Rootless: Fix lvalue error from devPrivates change
95100
95101 Instead of trying to use an invalid expression as an lvalue, use the function
95102 call instead.
95103
95104commit 36ff05771b0699aa80ad718e24097bc25cb4fb00
95105Author: Jurij Smakov <jurij@wooyd.org>
95106Date: Fri Dec 28 23:59:53 2007 +0200
95107
95108 GL: Add GLX compile flags lost in modular X server changes
95109
95110 RISC chips that trap on unaligned loads and stores need to
95111 define __GLX_ALIGN64. This used to get added to the cflags
95112 in the old *.cf files but it no longer does in the modular
95113 X server.
95114
95115 Also, Alpha needs to pass -mieee to the compiler as well.
95116
95117 This is a simple backport of a patch that debian, and probably other
95118 distributions, have been applying forever. To the best of my
95119 knowledge the patch was written by Jurij Smakov. See Debian bug
95120 number #388125.
95121
95122 I just checked and this has been rotting for more than a year in
95123 freedesktop bugzilla as #8392.
95124
95125 Signed-off-by: David S. Miller <davem@davemloft.net>
95126
95127commit bae459cfc4f17a5ec5f2810e9f913e3ad2d8b8d4
95128Author: Adam Jackson <ajax@redhat.com>
95129Date: Fri Dec 28 16:50:18 2007 -0500
95130
95131 Don't carp on EDID 1.4 blocks anymore.
95132
95133 Also whine more loudly when we get something other than 1.x.
95134
95135commit 70b2d6cfeb3bcb7b862a2ae29f6ef7cb84d69486
95136Author: Adam Jackson <ajax@redhat.com>
95137Date: Fri Dec 28 16:39:00 2007 -0500
95138
95139 Check the gamma value, not its address.
95140
95141commit 592d814ee09e86e283116a7a1052762c8398e8e5
95142Author: Adam Jackson <ajax@redhat.com>
95143Date: Fri Dec 28 16:37:23 2007 -0500
95144
95145 EDID 1.4: Additional semantics for display feature bits.
95146
95147 First mode is _always_ preferred in 1.4; the bit that used to mean this
95148 now means that the preferred mode is also the native pixel format. The
95149 old "is GTF" bit now means "is continuous-frequency" instead.
95150
95151 Section 3.6.4, Table 3.14: Feature Support, Notes 4 and 5.
95152
95153commit 322d0103aee317500057c80d542d7270d69a5731
95154Author: Adam Jackson <ajax@redhat.com>
95155Date: Fri Dec 28 16:28:44 2007 -0500
95156
95157 EDID 1.4: Alternate color encodings for digital inputs.
95158
95159 Section 3.6.4, Table 3.14: Feature support.
95160
95161commit f1f43caf7e26a84dbacd4e5d7d47c8b4e4982836
95162Author: Adam Jackson <ajax@redhat.com>
95163Date: Fri Dec 28 16:12:11 2007 -0500
95164
95165 EDID 1.4: Allow for gamma definition in extension blocks.
95166
95167 Section 3.6.3, Table 3.13: Display Transfer Characteristics (Gamma)
95168
95169commit 861ee38817523a647e6be10d7e8fe26f66054217
95170Author: Adam Jackson <ajax@redhat.com>
95171Date: Fri Dec 28 16:06:45 2007 -0500
95172
95173 EDID 1.4: Additional aspect ratio semantics for screen size fields.
95174
95175 Section 3.6.2, Table 3.12: Horizontal and Vertical Screen Size or
95176 Aspect Ratio.
95177
95178commit 14b5c8a447db0395fb14b2d404eafb1d8e4fb817
95179Author: Adam Jackson <ajax@redhat.com>
95180Date: Fri Dec 28 15:52:42 2007 -0500
95181
95182 EDID 1.4: Extended support for digital interfaces.
95183
95184 Section 3.6.1, Table 3.11: Video Input Definition.
95185
95186commit a948216dccb5ee577a50a42035dc9bc49d0a00c6
95187Author: Adam Jackson <ajax@redhat.com>
95188Date: Fri Dec 28 15:00:41 2007 -0500
95189
95190 EDID 1.4: Decode CVT 3-byte codes and add them to the mode pool.
95191
95192 Section 3.10.3.8: CVT 3 Byte Code Descriptor Definition.
95193
95194commit f6df66cc89bcd0a0be2e7bca05839fdd428c1d4c
95195Author: Adam Jackson <ajax@redhat.com>
95196Date: Fri Dec 28 13:55:39 2007 -0500
95197
95198 EDID 1.4: Trivial support for new detailed sections.
95199
95200 Nothing actually decoded yet, but at least we print what they are.
95201
95202 New in EDID 1.4:
95203 - Color Management Data (0xF9), Section 3.10.3.7
95204 - CVT 3 Byte Code Descriptor (0xF8), Section 3.10.3.8
95205 - Established Timings III Descriptor (0xF7), section 3.10.3.9
95206 - Manufacturer-specified data tag (0x00 - 0x0F), section 3.10.3.12
95207
95208commit bac3ecde39cc914ab515991234b7dc2138005b84
95209Author: Adam Jackson <ajax@redhat.com>
95210Date: Fri Dec 28 13:33:39 2007 -0500
95211
95212 EDID 1.4: Allow for sync range offsets.
95213
95214 Table 3.26: Display Range Limits & Timing Descriptor Block Definition
95215
95216commit 3b23dd9fd43a28033d0af7b02088b0c6ca433158
95217Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95218Date: Fri Dec 28 13:29:45 2007 -0500
95219
95220 xselinux: Fix whitespace warnings.
95221
95222commit 643c52be32c187a0fdb9a031b1e31d97cd551339
95223Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95224Date: Fri Dec 28 13:26:26 2007 -0500
95225
95226 xselinux: Remove "X" prefix on remaining functions and strings.
95227
95228 Should be evident from the context.
95229
95230commit f4bc333fc1e8d9fa9911771d2072df4df741c553
95231Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95232Date: Fri Dec 28 11:56:54 2007 -0500
95233
95234 xselinux: don't FatalError on an invalid class mapping, just disable support.
95235
95236commit f3780ece528ed3ead809ba6a388fa0f8aab2a775
95237Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95238Date: Thu Dec 20 20:32:07 2007 -0500
95239
95240 xselinux: Implement swapped protocol request logic.
95241
95242commit 1bbf64ab115e2a1121d6f9c0830b1b977f025178
95243Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95244Date: Thu Dec 20 20:31:34 2007 -0500
95245
95246 xselinux: Remove unnecessary structure definition.
95247
95248commit 938da5ee389975f910721f1c2cebc2dcec793117
95249Author: Adam Jackson <ajax@redhat.com>
95250Date: Fri Dec 28 12:23:44 2007 -0500
95251
95252 Add HDMI and DisplayPort connector types.
95253
95254commit 24105cf6582201a94bc39aeac5a795297018aeb5
95255Author: Daniel Stone <daniel@fooishbar.org>
95256Date: Fri Dec 28 15:49:50 2007 +0200
95257
95258 Input: Don't reinit devices
95259
95260 If a device is already initialised (i.e. the virtual core devices) during
95261 IASD, don't init them again. This fixes a leak.
95262
95263commit 941058f8da0d725f909dc97f68c32ce244a9dc0a
95264Author: Daniel Stone <daniel@fooishbar.org>
95265Date: Fri Dec 28 15:48:57 2007 +0200
95266
95267 KDrive: Xephyr: Don't leak screen damage structure
95268
95269commit b2f6cd290c43b88f0d08fb29f8657618a067d2a0
95270Author: Daniel Stone <daniel@fooishbar.org>
95271Date: Fri Dec 28 15:48:25 2007 +0200
95272
95273 OS: Don't leak connection translation table on regeneration
95274
95275commit 190a0506243b39cd8dfc0e12068e3a3f416330f1
95276Author: Daniel Stone <daniel@fooishbar.org>
95277Date: Fri Dec 28 15:47:57 2007 +0200
95278
95279 Config: HAL: Don't leak options on failure to add device
95280
95281 This showed up in Xephyr in particular, which denies new device requests.
95282
95283commit f44fd3f9e41bf467360ace93ef5b532d8f61fb2c
95284Author: Daniel Stone <daniel@fooishbar.org>
95285Date: Fri Dec 28 15:47:21 2007 +0200
95286
95287 Config: D-Bus: Don't leak timers
95288
95289 TimerCancel doesn't free the timer: you need TimerFree for that.
95290
95291commit ae869fc7669764729e13fdd70149ed636753f2a3
95292Author: David S. Miller <davem@davemloft.net>
95293Date: Tue Dec 25 22:42:50 2007 -0800
95294
95295 [SBUS]: Fix build, use getpagesize() instead of xf86getpagesize().
95296
95297 xf86getpagesize() was removed, but this one call site was
95298 not fixed up.
95299
95300 Signed-off-by: David S. Miller <davem@davemloft.net>
95301
95302commit 009f1e4e55200425de2fe0dbc1f0ac0f431fb4cf
95303Author: Fatih Aşıcı <fatih@pardus.org.tr>
95304Date: Tue Dec 25 23:09:49 2007 +0200
95305
95306 Config: Don't forget to add xkb_rules option
95307
95308commit 389e8917f66a489455f1d5c70f44c262717538ad
95309Author: Fatih Aşıcı <fatih@pardus.org.tr>
95310Date: Tue Dec 25 22:59:24 2007 +0200
95311
95312 Config: Fix a memory leak
95313
95314commit 743008a4812d6b046211ebcf4eab202687b458d5
95315Author: Adam Jackson <ajax@redhat.com>
95316Date: Sun Dec 23 14:27:14 2007 -0500
95317
95318 Report serverClient resources in the X-Resource extension.
95319
95320commit beb29c605b8c66e1a18b89668aa421c1519645f6
95321Author: Jeremy Huddleston <jeremy@yuffie.local>
95322Date: Fri Dec 21 02:09:01 2007 -0800
95323
95324 XQuartz: *REALLY* use CFStringCreateWithCString
95325 I need sleep! Why am I making these stupid mistakes... sorry for pointless commit spam. ugg.
95326 (cherry picked from commit b16351fc6457aabead328472d16dc25789032940)
95327
95328commit 5dd895efa305954e2695aa22a9e49acfb65b4d5e
95329Author: Jeremy Huddleston <jeremy@yuffie.local>
95330Date: Fri Dec 21 02:06:47 2007 -0800
95331
95332 XQuartz: Use CFStringCreateWithCString
95333 (cherry picked from commit 79782b0e14761dcf5d6635b8eec161b74f06763a)
95334
95335commit 2c24231fc2027cf5034bb1b6636332687f586726
95336Author: Jeremy Huddleston <jeremy@yuffie.local>
95337Date: Fri Dec 21 01:57:43 2007 -0800
95338
95339 XQuartz: Reduce code duplication in X11.app
95340 (cherry picked from commit b81809cd91a9f90b7f2de77b1dcf514cee87c32d)
95341
95342commit f3042a63be0748bb60567144276d2c61b75ba0b7
95343Author: Jeremy Huddleston <jeremy@yuffie.local>
95344Date: Fri Dec 21 01:24:06 2007 -0800
95345
95346 XQuartz: Handle Pseudorami init in miinitext
95347 (cherry picked from commit a585c94fedd4ecbc87524703c01bb128fc2aa951)
95348
95349commit fa9680a7305d7f906da1bdeb40a0863ef66316e6
95350Author: Jeremy Huddleston <jeremy@yuffie.local>
95351Date: Thu Dec 20 19:38:20 2007 -0800
95352
95353 XQuartz: Added localization.
95354 (cherry picked from commit 7a5cc7bfbb296a2c41a580b063324c448f7131db)
95355
95356commit 603a8b73d46d59e5f9f0be39be8317f3fadfe7e6
95357Author: Jeremy Huddleston <jeremy@yuffie.local>
95358Date: Thu Dec 20 18:29:57 2007 -0800
95359
95360 XQuartz: Cleaned up command line arguments.
95361
95362commit 4cf3002b6020024f2fc2ed0cc40a872a066e482d
95363Author: Jeremy Huddleston <jeremy@yuffie.local>
95364Date: Thu Dec 20 18:08:40 2007 -0800
95365
95366 XQuartz: Kill off server-main.c and launcher-main.c
95367 Now using xinit
95368 (cherry picked from commit 5d6ae3d299f72df714117948b3d31dcbddf6c0bc)
95369
95370commit 1f74bef1ad1399323fc0d2e309b808bf32c622e4
95371Author: Jeremy Huddleston <jeremy@yuffie.local>
95372Date: Thu Dec 20 17:33:38 2007 -0800
95373
95374 XQuartz: Don't fork to exec app_to_run.
95375 Plus other housecleaning...
95376 (cherry picked from commit ae302db472f127be082d05b418ede332fae8ccc5)
95377
95378commit 2d15d439f844d4016f169664a338595c11b91b77
95379Author: Jeremy Huddleston <jeremy@yuffie.local>
95380Date: Thu Dec 20 15:46:40 2007 -0800
95381
95382 Xquartz: Use X11ControllerMain()
95383 (cherry picked from commit a9ac932543374aa2540f5a12cc85ef82c85b0e0c)
95384
95385commit 1393a97ea97b5f7d7b90e3e8c58b5996b600e0c6
95386Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95387Date: Thu Dec 20 16:23:35 2007 -0500
95388
95389 xselinux: Send AVC messages to audit system instead of log file/stderr.
95390
95391commit 42802a8e6b3d3795acc4f8b7597ea5a48619b5cd
95392Author: Jeremy Huddleston <jeremy@yuffie.local>
95393Date: Thu Dec 20 13:17:30 2007 -0800
95394
95395 Xquartz: General Cleanup
95396 General code cleanup, whitespace, dead code removal, added missing prototypes.
95397 Made Xquartz come to foreground later in startup, so it doesn't appear for Xquartz -version
95398 (cherry picked from commit 36922e8ff4316c93843aa3fe959cf8df3c7d5892)
95399
95400commit 7ef7727b800fa4715b80a82850d65b88fde5fe6c
95401Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com>
95402Date: Thu Dec 20 10:11:26 2007 +1000
95403
95404 entity sharing: make !shareable entity non-fatal.
95405
95406 Just because the entity isn't shareable, we should bring down the server.
95407 Just ignore the extra screen and keep going.
95408
95409commit 50e80c39870adfdc84fdbc00dddf1362117ad443
95410Author: Peter Hutterer <peter@cs.unisa.edu.au>
95411Date: Wed Dec 19 16:20:36 2007 +1030
95412
95413 include: never overwrite realInputProc with enqueueInputProc. Bug #13511
95414
95415 In some cases (triggered by a key repeat during a sync grab) XKB unwrapping
95416 can overwrite the device's realInputProc with the enqueueInputProc. When the
95417 grab is released and the events are replayed, we end up in an infinite loop.
95418 Each event is replayed and in replaying pushed to the end of the queue again.
95419
95420 This fix is a hack only. It ensures that the realInputProc is never
95421 overwritten with the enqueueInputProc.
95422
95423 This fixes Bug #13511 (https://bugs.freedesktop.org/show_bug.cgi?id=13511)
95424 (cherry picked from commit eace88989c3b65d5c20e9f37ea9b23c7c8e19335)
95425
95426commit d0308b64655360517d83e07e866d103c3f2b389d
95427Author: Peter Hutterer <peter@cs.unisa.edu.au>
95428Date: Wed Dec 19 18:18:10 2007 +1030
95429
95430 Xi: specify correct struct when calculating size of GetDeviceControl reply.
95431
95432 This doesn't change much, as the struct previously given has the same size as
95433 the ones now anyway. Still, we should be pendantic.
95434
95435 Thanks to Simon Thum for reporting.
95436
95437commit 66b00029e587cec628d0041179a301e888277f8e
95438Author: Peter Hutterer <peter@cs.unisa.edu.au>
95439Date: Wed Dec 19 18:10:50 2007 +1030
95440
95441 Xext: remove redefinition of Bool.
95442
95443 Thanks to Simon Thum.
95444
95445commit eace88989c3b65d5c20e9f37ea9b23c7c8e19335
95446Author: Peter Hutterer <peter@cs.unisa.edu.au>
95447Date: Wed Dec 19 16:20:36 2007 +1030
95448
95449 include: never overwrite realInputProc with enqueueInputProc. Bug #13511
95450
95451 In some cases (triggered by a key repeat during a sync grab) XKB unwrapping
95452 can overwrite the device's realInputProc with the enqueueInputProc. When the
95453 grab is released and the events are replayed, we end up in an infinite loop.
95454 Each event is replayed and in replaying pushed to the end of the queue again.
95455
95456 This fix is a hack only. It ensures that the realInputProc is never
95457 overwritten with the enqueueInputProc.
95458
95459 This fixes Bug #13511 (https://bugs.freedesktop.org/show_bug.cgi?id=13511)
95460
95461commit bcad2a5a24f30cfdf9eca31915ed5a55ed094285
95462Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95463Date: Tue Dec 18 20:19:26 2007 -0500
95464
95465 XACE: Too many arguments to selection access hook.
95466
95467commit 7721d3e9217b41aab3a0ee5eaa52f5b53cbb07db
95468Author: Adam Jackson <ajax@redhat.com>
95469Date: Tue Dec 18 19:14:26 2007 -0500
95470
95471 Reference cvt and gtf in the xorg.conf man page.
95472
95473commit 51fab1eb30691c503f1b4dc98b465f2bc2e1394e
95474Author: Sam Lau <sam.lau@sun.com>
95475Date: Tue Dec 18 11:38:47 2007 -0800
95476
95477 Sun bug 6278039: Xevie checking wrong size in swapped XevieSelectInput requests
95478
95479 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6278039>
95480
95481commit 97c82ce0510808ea9d8a37a0a121e750f6dd8158
95482Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95483Date: Mon Dec 17 23:11:29 2007 -0500
95484
95485 XACE: Restore the old background None behavior in response to bug #13683.
95486
95487 From the X11 protocol spec:
95488
95489 "If background None is specified, the window has no defined background."
95490
95491 This means that toolkits and apps cannot rely on the "transparent" nature
95492 of the current implementation! At some point before the next release,
95493 XACE will switch back to a solid background as the default.
95494
95495commit 9f6ae61ad12cc2813d04405458e1ca5aed8a539e
95496Author: Peter Hutterer <peter@cs.unisa.edu.au>
95497Date: Tue Dec 18 13:57:07 2007 +1030
95498
95499 dix: set the correct number of valuators in valuator events.
95500
95501 (first_valuator + num_valuators) must never be larger than the number of axes,
95502 otherwise DIX freaks out. And from looking at libXI, anything larger than 6 is
95503 wrong too.
95504
95505commit 6a5c3e04fa43b98ccffd69ad86dd781602f88d0b
95506Author: Dave Airlie <airlied@redhat.com>
95507Date: Mon Dec 17 14:59:12 2007 +1000
95508
95509 mi: set the private key to a unique non-zero value
95510
95511commit a18d28a5efbe6021d6c800506cece28a73545aad
95512Author: Dave Airlie <airlied@redhat.com>
95513Date: Mon Dec 17 13:49:16 2007 +1000
95514
95515 damn then my cut-n-paste ate my end of lines...
95516
95517commit d096bbd01bf7c7e15b5a2c582718f3333e063ddc
95518Author: Dave Airlie <airlied@redhat.com>
95519Date: Mon Dec 17 13:45:15 2007 +1000
95520
95521 Xquartz ate my DMX - thanks
95522
95523commit bf20c4374aeb5160a0dc372df9b49f1bbc05f078
95524Author: Jeremy Huddleston <jeremy@yuffie.local>
95525Date: Sun Dec 16 01:14:32 2007 -0800
95526
95527 Xquartz: Removed launchd plist and x11-exec. The relevant code is now in xinit.
95528 (cherry picked from commit 767b4c9d9daa5d0ea59ac1f0d70146798da631cb)
95529
95530commit 58c2898b62fbf0d8e0f175de7cc208dc29d93788
95531Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
95532Date: Sun Dec 16 01:21:45 2007 +0100
95533
95534 xfree86: permit access to io port 0xffff on the hurd
95535
95536commit b4ef8885e1697b83a0dcc9f7fe79155f19241798
95537Author: Jeremy Huddleston <jeremy@yuffie.local>
95538Date: Sat Dec 15 14:00:19 2007 -0800
95539
95540 Xquartz: Fixed launchd detection
95541
95542commit ff5abc72fcc459d7eac663e5f8e4d40b28749841
95543Author: Otavio Salvador <otavio@ossystems.com.br>
95544Date: Fri Dec 14 17:59:29 2007 -0200
95545
95546 registry: XREGISTRY_UNKNOWN needs to be defined even if XREGISTRY isn't enabled
95547
95548 In case XREGISTRY isn't enabled, XREGISTRY_UNKNOWN is used but it's
95549 not being available. It's now always available.
95550
95551commit a3f7f7b60e391e6106f5db40b3fe5fbc67ccd836
95552Author: Ben Byer <bbyer@bbyer.apple.com>
95553Date: Thu Dec 13 20:45:14 2007 -0800
95554
95555 clarified debug message to indicate that we're sleeping
95556 (in case we get reports about slow launch times, this will
95557 help clarify what's happening)
95558 (cherry picked from commit 2eea3483cf893f8f81bacd434b31408dfb38cb06)
95559
95560commit e0e59b3bbc4d8e7ac3934a6f6a9e4a15b328c475
95561Author: Ben Byer <bbyer@bbyer.apple.com>
95562Date: Thu Dec 13 20:44:33 2007 -0800
95563
95564 we need to link against CoreServices, not ApplicationServices
95565 (cherry picked from commit ba4d2096e7953ef5b971682f0e28535da968acb1)
95566
95567commit 062d9234e233fc4c1c617f59093da973c9d3e2ce
95568Author: Ben Byer <bbyer@bbyer.apple.com>
95569Date: Thu Dec 13 20:40:27 2007 -0800
95570
95571 fixed bug in x11-exec that prevent icon from showing up
95572 (cherry picked from commit e1f4a0c20d3a52d98954c4b28d0ec4d44564bc32)
95573
95574commit 95c02adea80a14e18bb51876bc1418eccdade31d
95575Author: Jeremy Huddleston <jeremy@yuffie.local>
95576Date: Fri Dec 14 15:21:40 2007 -0800
95577
95578 Xquartz: Fixed cpprules include
95579
95580commit 86730337001ba4db6d77fe42406695e32784b157
95581Author: Otavio Salvador <otavio@ossystems.com.br>
95582Date: Fri Dec 14 08:46:35 2007 -0200
95583
95584 kdrive/ati: use operating system input devices
95585
95586commit e110255501e2f699709e6978f5e52d3be96333c8
95587Author: Otavio Salvador <otavio@ossystems.com.br>
95588Date: Fri Dec 14 08:45:09 2007 -0200
95589
95590 kdrive/vesa: use operating system input devices
95591
95592commit 863ba390e9fdf0d37cdf03bf5eebe7fdfe6288f5
95593Author: Otavio Salvador <otavio@ossystems.com.br>
95594Date: Fri Dec 14 00:03:13 2007 -0200
95595
95596 kdrive/fbdev: use operating system input devices
95597
95598commit ca59d3f7bdb5f3724ff45ea57912c0b1098a73d6
95599Author: Arkadiusz Miskiewicz <arekm@maven.pl>
95600Date: Thu Dec 13 00:09:08 2007 +0200
95601
95602 Xprint: Clean up generated files
95603
95604 Remember to clean generated wrapper files.
95605
95606commit 5b02a6ca5b31db69d08f2f452494c0f93a6260d9
95607Author: Bartosz Fabianowski <freebsd@chillt.de>
95608Date: Fri Dec 7 02:38:14 2007 +0000
95609
95610 Input: Fix proximity events with valuators
95611
95612 Initialise num_events to 1, so we always send a proximity event, and then
95613 optionally valuator events. Also make sure mieq can deal with valuator
95614 events sent after proximity events.
95615
95616commit a14a143832be844b4b890b0160ccb9fc8293c28c
95617Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95618Date: Fri Dec 14 00:57:16 2007 -0500
95619
95620 Bump server version for devPrivates rework / XACE.
95621
95622commit a2df51f8e95a814c54b806814020155ac8bd177d
95623Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95624Date: Fri Dec 14 00:53:54 2007 -0500
95625
95626 Bump video driver ABI and extension ABI for devPrivates rework.
95627
95628commit 86b2e59bfb79bd042a13c35fbb4ccecec576f629
95629Merge: 1c1a4bc cb0d7e2
95630Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95631Date: Fri Dec 14 00:32:24 2007 -0500
95632
95633 Merge branch 'master' into XACE-SELINUX
95634
95635 Conflicts:
95636
95637 configure.ac
95638
95639 This is the last pull before merging to master.
95640
95641commit 1c1a4bc970be061484bb8dcccf945eb08144c656
95642Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95643Date: Thu Dec 13 19:51:40 2007 -0500
95644
95645 devPrivates rework: more cleanup from previous merge operation.
95646
95647commit cb0d7e2c2692a332e2bd5495478ebf9a6cd601d0
95648Author: Jeremy Huddleston <jeremy@yuffie.local>
95649Date: Thu Dec 13 16:23:46 2007 -0800
95650
95651 Xquartz: Don't hardcode libexec dir
95652 (cherry picked from commit 67b479ef80cb740a24981335eb8d596744168a62)
95653
95654commit c39212fd7353fc1a07a30bade90f78356c748e2d
95655Author: Jeremy Huddleston <jeremy@yuffie.local>
95656Date: Thu Dec 13 15:56:31 2007 -0800
95657
95658 Xquartz: Don't hardcode LaunchAgents dir
95659 (cherry picked from commit 07a12d71fefd78c380078efa835700f2868ab204)
95660
95661commit 82e1aff9fbc1d15e3451707e3ccbf4b13eedda94
95662Author: Ben Byer <bbyer@bbyer.apple.com>
95663Date: Thu Dec 13 15:57:39 2007 -0800
95664
95665 Modified X11 plist to use x11-exec
95666 (cherry picked from commit 7d9a11329e476f45e4d9f9aebcb43469321347c7)
95667
95668commit 1a5910588a60af0c136595e2457d897d9e54ac88
95669Author: Ben Byer <bbyer@bbyer.apple.com>
95670Date: Thu Dec 13 15:55:28 2007 -0800
95671
95672 created x11-exec wrapper, which uses LaunchServices to find
95673 (and then exec) X11.app
95674 (cherry picked from commit fc04c9759b30d062111d4a7f3f411ed0f18cbde4)
95675
95676commit c8feb73f5841e7812d8dfdb91f1064e2033d028c
95677Merge: 9a7ce57 a125ce4
95678Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95679Date: Thu Dec 13 18:38:25 2007 -0500
95680
95681 Merge branch 'master' into XACE-SELINUX
95682
95683 Conflicts:
95684
95685 Xext/EVI.c
95686 Xext/bigreq.c
95687 Xext/cup.c
95688 Xext/dpms.c
95689 Xext/fontcache.c
95690 Xext/mitmisc.c
95691 Xext/xcmisc.c
95692 Xext/xf86bigfont.c
95693 Xext/xtest.c
95694 configure.ac
95695 dbe/dbe.c
95696 hw/darwin/darwin.h
95697 hw/darwin/darwinEvents.c
95698 hw/darwin/iokit/xfIOKit.h
95699 hw/darwin/iokit/xfIOKitCursor.c
95700 hw/darwin/quartz/fullscreen/fullscreen.c
95701 hw/darwin/quartz/fullscreen/quartzCursor.c
95702 hw/darwin/quartz/quartz.c
95703 hw/darwin/quartz/quartzCommon.h
95704 hw/darwin/quartz/quartzCursor.c
95705 hw/darwin/quartz/xpr/dri.c
95706 hw/darwin/quartz/xpr/dristruct.h
95707 hw/darwin/quartz/xpr/xprCursor.c
95708 hw/darwin/quartz/xpr/xprFrame.c
95709 hw/xfree86/modes/xf86RandR12.c
95710 include/cursor.h
95711 miext/rootless/rootlessCommon.h
95712 miext/rootless/rootlessScreen.c
95713 miext/rootless/rootlessWindow.c
95714 render/picturestr.h
95715
95716 Trying to pick up the pieces from the darwin churn here...
95717
95718commit efcdc0d7010f4e6ec833842cb010a07068edf7ab
95719Author: Adam Jackson <ajax@redhat.com>
95720Date: Thu Dec 13 15:38:41 2007 -0500
95721
95722 Correct the documentation comments in xf86Modes.c
95723
95724 Most of those functions do not, in fact, work with circular mode lists,
95725 and by this point the API isn't really "proposed" anymore.
95726
95727commit 1768af38c737f4c14d32f587b51a8ec3d3d6ed5f
95728Author: Adam Jackson <ajax@redhat.com>
95729Date: Thu Dec 13 15:06:18 2007 -0500
95730
95731 Add infrastructure for validating modes by memory bandwidth.
95732
95733commit 4359193aaa522599c502d012b9c163e993c01d79
95734Author: Adam Jackson <ajax@redhat.com>
95735Date: Thu Dec 13 10:59:48 2007 -0500
95736
95737 Explain a confusing #ifdef.
95738
95739commit 8cedbb0a53d47b12f03edb726db9d5879c8a63a4
95740Author: Adam Jackson <ajax@redhat.com>
95741Date: Thu Dec 13 10:57:35 2007 -0500
95742
95743 Clean up some #if 0.
95744
95745commit 9a7ce573636e349ee2967991c7cc1407e80ae524
95746Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
95747Date: Wed Dec 12 20:44:59 2007 -0500
95748
95749 xselinux: Add new protocol for setting device create context.
95750
95751commit a125ce4a84f5fb5934fefebd7cfb22a83180874d
95752Author: Aaron Plattner <aplattner@nvidia.com>
95753Date: Wed Dec 12 12:20:54 2007 -0800
95754
95755 Fix dist by including modeline2c.awk.
95756
95757 This was broken by commit cb44b6121c4b7b9dd7ff4ff52aaab914c82ff013, which
95758 removed modeline2c.pl from EXTRA_DIST without adding modeline2c.awk.
95759
95760commit 671592343701d8174a70f1ffb9c818784ea3af7a
95761Author: Aaron Plattner <aplattner@nvidia.com>
95762Date: Wed Dec 12 10:59:15 2007 -0800
95763
95764 Get rid of xf86DefModes.c.
95765
95766 It's out of date and not included in the build. Instead, xf86DefModeSet.c is
95767 built from vesamodes and extramodes using modeline2c.awk and *that's* what gets
95768 built.
95769
95770commit d86e7f2ecc0a0129a722ffaca93dfc7c0c60e842
95771Author: Peter Hutterer <peter@cs.unisa.edu.au>
95772Date: Fri Dec 7 20:40:21 2007 +1030
95773
95774 dix: when getting a bogus event, at least print out the type.
95775
95776commit eab0c4e49015fe96f6d985316f9c5fa28a7eb1fe
95777Author: Jeremy Huddleston <jeremy@yuffie.local>
95778Date: Mon Dec 10 20:57:24 2007 -0800
95779
95780 Xquartz: Pre-process Xquartz man page
95781 (cherry picked from commit dec2633c41dd0adf73264afdf251a6522d6ae76a)
95782
95783commit 1ff945a8e43e622b39b360ee49efd6ae3b77be67
95784Author: Jeremy Huddleston <jeremy@yuffie.local>
95785Date: Mon Dec 10 20:47:48 2007 -0800
95786
95787 Xquartz: Make Alt work with Xmodmap again
95788 (cherry picked from commit 0e017177dcca7185716ca760dcce9ddedc7bfef9)
95789
95790commit 7d61893b49569a72bccb63f1ae8c9ce4ef4e354f
95791Author: Jeremy Huddleston <jeremy@yuffie.local>
95792Date: Mon Dec 10 20:33:30 2007 -0800
95793
95794 Xquartz: Removed Xplugin.h from EXTRA_DIST
95795 (cherry picked from commit a746383eca77c9b9ea2cba0cf1c8fc39c0f7d536)
95796
95797commit 8f2eff643bf421bc4233fbaa2409b75d9f80d147
95798Author: Ben Byer <bbyer@bbyer.local>
95799Date: Sat Dec 8 23:34:40 2007 -0800
95800
95801 remove Xplugin.h, because we should use the one in /usr/include
95802 (cherry picked from commit 3e881032f35f774ff9638678d7e3f77c81f62976)
95803
95804commit 7b573ed43672b1fac7b4e6df85a657942ab4cba6
95805Author: Jeremy Huddleston <jeremy@yuffie.local>
95806Date: Sun Dec 9 12:02:04 2007 -0800
95807
95808 Xquartz: Added missing link to libconfig.a
95809 (cherry picked from commit 14ec1cf1cb7ebc183c05e13f9c2b4b4eed679ff3)
95810
95811commit cd13c4ba5b7a1bdfb419cb492a96a72dccf2681e
95812Author: Jeremy Huddleston <jeremy@yuffie.local>
95813Date: Sat Dec 8 13:18:17 2007 -0800
95814
95815 .gitignore: added hw/xquartz/bundle/org.x.X11.plist
95816
95817commit 020b0e92b039d6ddaea0bbdb890b6a01037bf9b6
95818Author: Jeremy Huddleston <jeremy@yuffie.local>
95819Date: Sat Dec 8 11:49:37 2007 -0800
95820
95821 Xquartz Added quartzKeyboard.h to EXTRA_DIST
95822 (cherry picked from commit 37c9781fdb672229ceab101b080762e15512943f)
95823
95824commit 5e016fa9b2bf28971ed1794f4706c6538b1d411c
95825Author: Ben Byer <bbyer@bbyer.local>
95826Date: Sat Dec 8 06:12:46 2007 -0800
95827
95828 Added darwinEvents.h to EXTRA_DIST
95829 (cherry picked from commit 45e5247564c423a2bf02cfec1993155858c91a14)
95830
95831commit 6bb5dacc1710cdbededb9b28ba89a184ecd0931c
95832Author: Jeremy Huddleston <jeremy@yuffie.local>
95833Date: Sat Dec 8 01:41:37 2007 -0800
95834
95835 Xquartz: Actually add quartzForeground.[hc]
95836 Sorry for the commit spam... I'm tired and was overly quick to commit... forgot to include a neccessary file.
95837 (cherry picked from commit e564b7aeaab63e4c943445275af680b3b5898a94)
95838
95839commit 02df03667052fa6a4e0405b91a005dc48e9b39c4
95840Author: Jeremy Huddleston <jeremy@yuffie.local>
95841Date: Sat Dec 8 01:28:26 2007 -0800
95842
95843 Xquartz: Actually, it should be org.x.X11 for case-sensitive FS
95844 (cherry picked from commit c5ccb98d5d461c8a22fc0f3942a607ac90e1e37e)
95845
95846commit 740cc54f081393d4ffe1a3e91c9e504dfaee3fe9
95847Author: Jeremy Huddleston <jeremy@yuffie.local>
95848Date: Sat Dec 8 01:24:58 2007 -0800
95849
95850 Xquartz: Use org.x.X11 instead of com.apple.X11 for preferences
95851 Fixed inconsistency so preferences get read from the correct source.
95852 (cherry picked from commit a74c38bd9f28735acd602d359d7ca6357aed1e93)
95853
95854commit 41a0aeaae9b7b2f8cc2468fd1f3ee11287d34828
95855Author: Jeremy Huddleston <jeremy@yuffie.local>
95856Date: Sat Dec 8 00:13:47 2007 -0800
95857
95858 XQuartz: Fixed "Multiple Dock Icons"
95859 BAM!
95860 (cherry picked from commit d0dca8a88506f50b51f41f99a2f1feb6954c8a31)
95861 (cherry picked from commit 0502955a2af487b51bf22916ac02e497c2d96aba)
95862
95863commit a1b0346853720e98963910b82603c5cda72bb7f9
95864Author: Jeremy Huddleston <jeremy@yuffie.local>
95865Date: Fri Dec 7 23:26:11 2007 -0800
95866
95867 XQuartz: Cleaned up configure, X11.app path in launchd script
95868 Don't hardcode X11.app's path in the launchd plist.
95869 Only install the launchd plist if we --enable-launchd.
95870 (cherry picked from commit 6b74c535dc331d1d621b2541492a3336f69d70a2)
95871
95872commit 1157cfcc5a4e2a7299a4c48df04a1cc8d5093906
95873Author: Ben Byer <bbyer@bbyer.local>
95874Date: Fri Dec 7 21:55:42 2007 -0800
95875
95876 Just a couple of small uninitialized pointer fixes
95877 (cherry picked from commit d12b650362da100ceaecb7e859cd4ef1908d4407)
95878
95879commit 0ad1c359c5b0be63748f5c630c97be88a8cc92ce
95880Author: Jeremy Huddleston <jeremy@dhcp-38-248.EECS.Berkeley.EDU>
95881Date: Fri Dec 7 18:54:58 2007 -0800
95882
95883 Darwin: Use __APPLE__ instead of __DARWIN__
95884 (cherry picked from commit 54654815fa5e59b25cfd1fa72610120b72c10175)
95885
95886commit 85ed0bb44011312dfaa9f2dc31642a0f89ec0bd3
95887Author: Brice Goglin <bgoglin@debian.org>
95888Date: Sat Dec 8 02:53:27 2007 +0100
95889
95890 Add a missing linebreak after LoadModule: "foo"
95891
95892 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=454742
95893
95894commit 4fc288a13f825db942c9dcd64f4abd0265652faf
95895Author: Alan Coopersmith <alan.coopersmith@sun.com>
95896Date: Fri Dec 7 17:28:37 2007 -0800
95897
95898 Check for <sys/sdt.h> as well when determining to enable dtrace probes
95899
95900 Avoids auto-detecting dtrace is present on systems with the ISDN trace tool
95901 named dtrace installed, but not the dynamic tracing facility named dtrace
95902
95903commit 67907904f094c803d5faf6fa2ce23c01f9a5a521
95904Author: Ben Byer <bbyer@bbyer.local>
95905Date: Fri Dec 7 01:51:53 2007 -0800
95906
95907 fixed pathname in GL/apple/Makefile.am
95908 (cherry picked from commit b6357cec6d837226009c0d2b69026027da36656e)
95909
95910commit 56f5066d477836a975122f4e5748c0f4fb790175
95911Author: Ben Byer <bbyer@bbyer.apple.com>
95912Date: Thu Dec 6 20:51:32 2007 -0800
95913
95914 ALLOCATE_LOCAL is dangerous on Darwin due to memory layout differences
95915 from Linux, so let's define NO_ALLOCA.
95916 (cherry picked from commit 7caf51d1a5a86ae884e0087795636222c082962c)
95917
95918commit c238ef06a270c0c1d48cdb9175b6d5815c7c2a49
95919Author: Jeremy Huddleston <jeremy@yuffie.local>
95920Date: Wed Dec 5 21:36:34 2007 -0800
95921
95922 Darwin: Dead coded removal
95923 Kill off assert macro
95924 (cherry picked from commit d6493abedb2caf03b2bc3a6440b637df67eff081)
95925
95926commit 3a058456e0ce531b21ae2b37be8868b8e0d56e56
95927Author: Peter Hutterer <peter@cs.unisa.edu.au>
95928Date: Fri Dec 7 20:23:36 2007 +1030
95929
95930 mi: unify ErrorFs, prepend [mi]
95931
95932commit c5acf401dad6fbca47f2bb870185c1d08ac947a5
95933Author: Peter Hutterer <peter@cs.unisa.edu.au>
95934Date: Fri Dec 7 20:19:43 2007 +1030
95935
95936 mi: unify last two ErrorF's in mieq.c
95937
95938commit 117458d2db49efd3f04432ff45871b44c7d4ad57
95939Author: Peter Hutterer <peter@cs.unisa.edu.au>
95940Date: Fri Dec 7 20:18:49 2007 +1030
95941
95942 mi: change infamous "Tossed event ..." error for something more explanatory.
95943
95944 Few whitespace errors fixed.
95945 Two ErrorF's prefixed with [mi].
95946
95947commit 8a8239f2e21795602fcff5281833b350e6b2a286
95948Author: Jeremy Huddleston <jeremy@yuffie.local>
95949Date: Wed Dec 5 21:23:36 2007 -0800
95950
95951 Darwin: Renamed DarwinModeBlah to QuartzBlah
95952 (cherry picked from commit 08ebf86d379b1ddfb08df65d29aea5df66de4327)
95953
95954commit 540439a966cce3fc68a7e4bffdb5bcab1b20725f
95955Author: Jeremy Huddleston <jeremy@yuffie.local>
95956Date: Wed Dec 5 20:55:06 2007 -0800
95957
95958 .gitignore: Added Xcode user files
95959
95960commit bc65a243930e4b02f06a861495420b0a120eae8c
95961Author: Jeremy Huddleston <jeremy@yuffie.local>
95962Date: Wed Dec 5 19:43:49 2007 -0800
95963
95964 Darwin: Flattened quartz into darwin, renamed darwin xquartz
95965 Leaving xpr unflattened since we want modularity to replace that with xpc (XPluginComposite) at some point
95966 (cherry picked from commit 48e6a75fbdd0fee86e364f02ace83f20b312a2b2)
95967
95968commit e00f7061b22001989edf5bd38c2d0cc1566fdd19
95969Author: Jeremy Huddleston <jeremy@yuffie.local>
95970Date: Tue Dec 4 23:18:37 2007 -0800
95971
95972 Darwin: Cleaned up keyboard interface headers
95973 (cherry picked from commit 141f69dc3d8d6e7d8ff65607f43700ac11243041)
95974
95975commit cb44b6121c4b7b9dd7ff4ff52aaab914c82ff013
95976Author: Andrew Oakley <andrew@ado.is-a-geek.net>
95977Date: Wed Dec 5 20:23:05 2007 -0500
95978
95979 Fix commit aa0dfb3f42f19bb351ca7f1a9507ff5ec4590e96
95980 From bugzilla bug 13467¹:
95981
95982 Currently the xserver fails to build without this (now deleted) file, as the
95983 Makefile tries to distribute it. The patch simply removes the reference to
95984 modeline2c.pl.
95985
95986 1] http://bugs.freedesktop.org/show_bug.cgi?id=13467
95987
95988 Signed-off-by: James Cloos <cloos@jhcloos.com>
95989
95990commit 320abd7d1d906807448fa01ad3377daf707f46cc
95991Author: Daniel Stone <daniel@fooishbar.org>
95992Date: Wed Dec 5 19:37:48 2007 +0000
95993
95994 XKB: Actions: Don't run certain actions on the core keyboard
95995
95996 Don't run VT switches, terminations, or anything, on the core keyboard: only
95997 run actions which affect the keyboard state. If we get an action such as VT
95998 switch, just swallow the event.
95999
96000commit 85dd8efac1bc0715f03c99d261b1c5d0980623e1
96001Author: Daniel Stone <daniel@fooishbar.org>
96002Date: Wed Dec 5 19:36:59 2007 +0000
96003
96004 WaitForSomething: Ignore EAGAIN
96005
96006 If select ever returns EAGAIN, don't bother complaining.
96007
96008commit b7f3618f3933a810778093fd47564a1e3bf3fde6
96009Author: Rich Coe <Richard.Coe@med.ge.com>
96010Date: Wed Dec 5 19:36:37 2007 +0000
96011
96012 OS: Connection: Keep trying select while it gets interrupted (bug #9240)
96013
96014 If we got interrupted (EINTR or EAGAIN) during select, just try again, rather
96015 than shutting clients down on either of these errors.
96016
96017commit d8b2cad3771a09860e7be1726f67e684cf7caeec
96018Author: Rich Coe <Richard.Coe@med.ge.com>
96019Date: Wed Dec 5 19:31:07 2007 +0000
96020
96021 OS: Connection: Don't shut down disappeared clients (bug #7876)
96022
96023 If a client disappears in the middle of CheckConnections (presumably
96024 because its appgroup leader disappears), then don't attempt to shut it down
96025 a second time, when it's already vanished.
96026
96027commit a8e27a108abeba73b2888da4e0604008f4b02045
96028Author: Kanru Chen <koster@debian.org.tw>
96029Date: Mon Dec 3 12:46:45 2007 +0000
96030
96031 Config: HAL: Fix XKB option parsing
96032
96033 Actually combine the XKB options into a string, rather than just repeatedly
96034 writing a comma.
96035
96036commit 2d723bbd0d36f6d7763b4df3298d40720f97fdd0
96037Author: Peter Harris <peter.harris@hummingbird.com>
96038Date: Mon Oct 29 18:05:19 2007 -0400
96039
96040 Add missing swaps in panoramiXSwap.c
96041
96042commit 0fccb24aa978b838cf0fb008e9695837e612c529
96043Author: Daniel Stone <daniel@fooishbar.org>
96044Date: Fri Nov 30 20:35:26 2007 +0200
96045
96046 ProcessOtherEvent: Don't do double translation of button events
96047
96048 We already deal with the button mapping in GetPointerEvents, so don't
96049 do the remapping again in ProcessOtherEvent.
96050
96051commit c6cfcd408df3e44d0094946c0a7d2fa944b4d2d1
96052Author: Hong Liu <hong.liu@intel.com>
96053Date: Wed Dec 5 17:48:28 2007 +0100
96054
96055 Bug 13308: Verify and reject obviously broken modes.
96056
96057commit 8d0efe4c2a48047680af40e5f6d639f426902e07
96058Author: Jeremy Huddleston <jeremy@yuffie.local>
96059Date: Tue Dec 4 17:59:13 2007 -0800
96060
96061 Darwin: Rework build system to more accurately reveal code infrastructure and facilitate future modularity.
96062 (cherry picked from commit e8399fd4d66a2b77b770c277e2fa424229a721b2)
96063
96064commit cc98a8e2415f12c7a90fd846d1ec858068e8c796
96065Author: Jeremy Huddleston <jeremy@yuffie.local>
96066Date: Mon Dec 3 23:59:19 2007 -0800
96067
96068 Darwin: RIP dumpkeymap, cr, and fullscreen
96069 Taking out the trash.
96070 We don't need dumpkeymap since we'll be getting keymaps straight from the OS. .Xmodmap should be sufficient for any user-needed changes. If this is not
96071 the case, please let us know, so we can address any problems you have.
96072 fullscreen never worked AFAIK
96073 cr isn't being used and xpr is much better.
96074 (cherry picked from commit e41af2967e885466c4d194fa4c3b358e6be37c30)
96075
96076commit 13af2d1efcc83d1412a4c727afddd97577b00f32
96077Author: Adam Jackson <ajax@redhat.com>
96078Date: Tue Dec 4 17:36:21 2007 -0500
96079
96080 Restore xf86getsecs() as not having an ANSI equivalent.
96081
96082commit 447cd5d411875b62eb1a501bf00e604225b74d26
96083Author: Peter Hutterer <peter@cs.unisa.edu.au>
96084Date: Tue Dec 4 19:09:37 2007 +1030
96085
96086 dix: update comments about inputInfo.pointer.
96087
96088commit f44d7dcb5fefca3ddfd45e75d0bd0b47ec785e48
96089Author: Peter Hutterer <peter@cs.unisa.edu.au>
96090Date: Tue Dec 4 19:07:46 2007 +1030
96091
96092 dix: change the filters to be per-device.
96093
96094 If we have one global filter, one pointer may change the filter value and
96095 affect another pointer.
96096
96097 Reproduceable effect:
96098 blackbox and xterm, start dragging xterm then click anywhere with the other
96099 pointer (attached to different masterd device!). The button release resets
96100 the filter[Motion_Filter(button)] value, thus stopping dragging and no event
96101 is sent to the client anymore.
96102 Having the filters set per device gets around this.
96103
96104commit 0931f40bf1bd6e00b8d95968d761a495b2c9a46c
96105Author: Peter Hutterer <peter@cs.unisa.edu.au>
96106Date: Tue Dec 4 17:08:56 2007 +1030
96107
96108 dix: comments, whitespaces, copyright fixes.
96109
96110 Removing my copyright message for now, should eventually be in line with the
96111 rest of the messages.
96112
96113commit 09c0c1a3cc4807813774a3c0e28a7ba9a87bb5c7
96114Author: Peter Hutterer <peter@cs.unisa.edu.au>
96115Date: Tue Dec 4 16:51:52 2007 +1030
96116
96117 dix: remove #ifdef XINPUT.
96118
96119 No escaping XINPUT anymore.
96120
96121commit f8d7729df388c142624def36ba6d8c3b15922018
96122Author: Jeremy Huddleston <jeremy@yuffie.local>
96123Date: Mon Dec 3 20:20:05 2007 -0800
96124
96125 Darwin: Combine launcher and server X11.app
96126 This should hopefully eliminate confusion some people have over which X11.app is which.
96127 Now BOTH are in /A/U/X11.app and we intelligently determine whether to execute our app_to_run
96128 or launch the server. If arguments are given, we launch the server. Otherwise if we can
96129 connect to an X DISPLAY, we execute app_to_run. Otherwise, we launch the server.
96130
96131 (cherry picked from commit e7026216ccaa8e4fb073800ba947c9909d4faada)
96132
96133commit 678f786715d76e972f8a77807c9caf3e90c24418
96134Author: Dave Airlie <airlied@redhat.com>
96135Date: Tue Dec 4 12:24:47 2007 +1100
96136
96137 xf86crtc: oh mon could be NULL, so check before quirks
96138
96139commit a9df4bb555fd91707a68794c2dce24fb06e6cf64
96140Author: Dave Airlie <airlied@redhat.com>
96141Date: Tue Dec 4 12:17:29 2007 +1100
96142
96143 xf86Crtc: pass correct parameter.
96144
96145 quite how this has worked I've no idea.
96146
96147commit fe25f897c62bb324660217e15dbd3091c808dbba
96148Author: Adam Jackson <ajax@redhat.com>
96149Date: Mon Dec 3 18:34:40 2007 -0500
96150
96151 xf86getpagesize -> getpagesize elsewhere in os-support/
96152
96153commit b84f2833a681585162b8dabfb02ff62e7e0ef4d6
96154Author: Alan Coopersmith <alan.coopersmith@sun.com>
96155Date: Mon Dec 3 14:52:17 2007 -0800
96156
96157 xf86getpagesize() -> getpagesize() in os-support/solaris/sun_bios.c
96158
96159commit aa0dfb3f42f19bb351ca7f1a9507ff5ec4590e96
96160Author: James Cloos <cloos@jhcloos.com>
96161Date: Mon Dec 3 16:57:58 2007 -0500
96162
96163 Remove Perl dependency from the build
96164 From bugzilla bug 13467¹:
96165
96166 The modeline2c script is the only part of the Xorg server that requires Perl.
96167 [This] is a simpler replacement that works with any normal AWK.
96168
96169 1] http://bugs.freedesktop.org/show_bug.cgi?id=13467
96170
96171 Bug was posted by Joerg Sonnenberger <joerg@NetBSD.org>.
96172
96173commit 60086d90168265795e07a60939e9e2fe95c6e15c
96174Author: Alan Coopersmith <alan.coopersmith@sun.com>
96175Date: Mon Dec 3 11:30:58 2007 -0800
96176
96177 Use pkg-config to get -I, -L & -R flags needed for OpenSSL
96178
96179 Still just uses -lcrypto instead of the full library list from --libs
96180
96181commit b77ca7cc9c23184c4ab367baf1b3ed0acf27c269
96182Author: Alan Coopersmith <alan.coopersmith@sun.com>
96183Date: Mon Dec 3 11:29:54 2007 -0800
96184
96185 Use _X_EXPORT instead of __attribute__((visibility("default")))
96186
96187commit 1faba797cbfe1a4804b7ea6b47e1ca9d4e4324e4
96188Author: Adam Jackson <ajax@redhat.com>
96189Date: Mon Dec 3 14:12:58 2007 -0500
96190
96191 Death to libcwrapper.
96192
96193 This has been deprecated since 1.1. Since we're breaking ABI again anyway,
96194 remove it entirely.
96195
96196commit 28b93d74a11a1064d68a214fcaa7b0aede864a38
96197Author: Peter Hutterer <peter@cs.unisa.edu.au>
96198Date: Mon Dec 3 14:32:53 2007 +1030
96199
96200 ephyr: miPointerWarpCursor needs the device parameter.
96201
96202 Thanks to Mitchell Quille for spotting it.
96203
96204commit 004876355b43fb4d3c2bc5653a7dc1bfd3f985ee
96205Author: Peter Hutterer <peter@cs.unisa.edu.au>
96206Date: Fri Nov 30 13:56:04 2007 +1030
96207
96208 dix: Remove RegisterPairingClient and UnregisterPairingClient.
96209
96210commit f4dc521b38560c8f93b614316a3a5511941a93a9
96211Author: Jeremy Huddleston <jeremy@tifa.local>
96212Date: Sun Dec 2 18:21:40 2007 -0800
96213
96214 Darwin: Added {/,/System/}Library/Fonts to DEFAULT_FONT_PATH
96215 (cherry picked from commit b0069b04dddaa2df6d4cdf86f96fd8a2a257e47e)
96216
96217commit 0fff01f5660fb3bb9284f97c45dc76154435d02b
96218Author: Adam Jackson <ajax@aspartame.nwnk.net>
96219Date: Sun Dec 2 14:15:36 2007 -0500
96220
96221 Fix swapped Xv dispatch under Xinerama.
96222
96223 Same-endian dispatch was properly calling through the Xinerama wrapping,
96224 but other-endian dispatch wasn't.
96225
96226commit fa47910045c3700d8d668b5e214e5ffc1e8dc3e7
96227Author: Adam Jackson <ajax@redhat.com>
96228Date: Sun Dec 2 12:39:05 2007 -0500
96229
96230 Clean up many #if 0.
96231
96232commit 83ba1e167c1473ac7d85239a6ee5ed629353cb16
96233Author: Ben Byer <bbyer@bbyer.local>
96234Date: Sat Dec 1 18:28:19 2007 -0800
96235
96236 added missing call to xcb_connect()
96237 (cherry picked from commit dc2fb323ee11f081d447605be151024f9e2487f9)
96238
96239commit 9ad4560b3cbd42e647d6227746d4d037616d57cf
96240Author: Jeremy Huddleston <jeremy@tifa.local>
96241Date: Sat Dec 1 16:23:23 2007 -0800
96242
96243 Darwin: Alt is Mode_switch Switching to Mode_switch to maintain compatibility with Tiger X11.
96244 (cherry picked from commit 8a76c99c0ebbaf7375f3a9c75c4f7921a79024da)
96245
96246commit f83d758dcc4878849a851c8466f6fa16b2b7cd8e
96247Author: Jeremy Huddleston <jeremy@yuffie.local>
96248Date: Fri Nov 30 16:11:15 2007 -0800
96249
96250 Darwin: properly implemented xcb check for stale sockets
96251 (cherry picked from commit f543cb8fbb3d9213cb03396f4252ab9821319993)
96252
96253commit f54b28eeba119c42d0fcccfbe295306dd670221a
96254Author: Jeremy Huddleston <jeremy@yuffie.local>
96255Date: Fri Nov 30 16:09:23 2007 -0800
96256
96257 Darwin: Undo focus-hack which didn't work right.
96258
96259commit f30abe30c5fea10e680aa12f3fe37ee8ce1a0201
96260Author: Dave Airlie <airlied@linux.ie>
96261Date: Fri Nov 30 13:52:06 2007 +1000
96262
96263 edid quirk for MAX 0x77e monitor
96264
96265 From RH bugzilla 306441
96266
96267commit 6216abe0c1a94efa4b2f0d14e88278f128304cd2
96268Author: Peter Hutterer <peter@cs.unisa.edu.au>
96269Date: Fri Nov 30 11:34:24 2007 +1030
96270
96271 dix: avoid activating the VCP/VCK twice.
96272
96273 This may set dev->key-xkbinfo to NULL, causing a segfault in xkb code lateron.
96274 Spotted by David Huggins-Daines.
96275
96276commit 9eb8ea9e615ebd8b568711eeca36aa84aa781ad3
96277Author: Peter Hutterer <peter@cs.unisa.edu.au>
96278Date: Fri Nov 30 11:32:59 2007 +1030
96279
96280 dix: only freeze the paired MD on a grab, not all other devices.
96281
96282commit 8a079be0dd0f2ce37868988cde4ac8895522b088
96283Author: Jeremy Huddleston <jeremy@yuffie.local>
96284Date: Thu Nov 29 02:19:22 2007 -0800
96285
96286 Darwin: #ifdefs around dix-config.h include and NDEBUG/assert.h workaround.
96287 (cherry picked from commit d2b768890f0878ae4e3fec8f7219e82b79256133)
96288
96289commit 38397560612424b5b348f34c1a0bea8c47a574be
96290Author: Jeremy Huddleston <jeremy@yuffie.local>
96291Date: Wed Nov 28 23:07:41 2007 -0800
96292
96293 Darwin: Removed support for darwinSwapAltMeta
96294 (cherry picked from commit 3d153c8fa40986d194b7701f5eafa0080e32399a)
96295
96296commit 89c3dfe41e3a17a4f27b20e23623dc5777670feb
96297Author: Dave Airlie <airlied@redhat.com>
96298Date: Thu Nov 29 19:57:24 2007 +1100
96299
96300 modes: use xf86RandR12Index to stop illegal access
96301
96302 xf86RandR12Index set to -1, and if initialised it gets 0 or higher.
96303 This allows the server to start with xinerama turned on with only one head
96304
96305commit 725710fd0bc990b2c35e4c76128ef1c668013299
96306Author: Dave Airlie <airlied@redhat.com>
96307Date: Thu Nov 29 19:40:53 2007 +1100
96308
96309 randr: make randr code not segfault when xinerama set
96310
96311commit e4fe0a3cb789b8757f1c80f606dfe32bccada582
96312Author: Peter Hutterer <peter@cs.unisa.edu.au>
96313Date: Thu Nov 29 09:55:24 2007 +1030
96314
96315 xkb: swap a LookupKeyboardDevice over in favour of GetPairedDevice
96316
96317 The former always returns the VCK, which is obviously wrong if we have
96318 multiple devices.
96319
96320commit 23b8ca8a373d919225de9739af7b064f650eceec
96321Author: Adam Jackson <ajax@redhat.com>
96322Date: Tue Nov 27 13:20:40 2007 -0500
96323
96324 RANDR 1.2: Only enable unknown outputs if there are no connected outputs.
96325
96326 Otherwise you end up with a confusing initial geometry, and xscreensaver
96327 and friends get very angry.
96328
96329commit edebe76cfdb31072d18a6fcd3ee8f1d95006855f
96330Author: Peter Hutterer <peter@cs.unisa.edu.au>
96331Date: Tue Nov 27 10:22:44 2007 +1030
96332
96333 Xi: set DeviceXXXState's length fields to the correct size of the struct.
96334
96335 Setting it to the size of a pointer is an interesting but equally wrong
96336 approach. Luckily Xlib never used this field anyway so nobody got hurt so far.
96337
96338 Spotted by Simon Thum.
96339 (cherry picked from commit 0f2398d06ce591724e388b3270800c5e22b3de2d)
96340
96341commit 0f2398d06ce591724e388b3270800c5e22b3de2d
96342Author: Peter Hutterer <peter@cs.unisa.edu.au>
96343Date: Tue Nov 27 10:22:44 2007 +1030
96344
96345 Xi: set DeviceXXXState's length fields to the correct size of the struct.
96346
96347 Setting it to the size of a pointer is an interesting but equally wrong
96348 approach. Luckily Xlib never used this field anyway so nobody got hurt so far.
96349
96350 Spotted by Simon Thum.
96351
96352commit 601307615e4955be23fd86a057285074242ad83e
96353Author: Jeremy Huddleston <jeremy@yuffie.local>
96354Date: Mon Nov 26 13:04:57 2007 -0800
96355
96356 Darwin,Rootless: Makefile cleanup
96357 (cherry picked from commit 9c6d8a035b712b219833653ac637b89703a9b0c3)
96358
96359commit 8503072e1c2b89dca786d4afb72aa60a170d2fbd
96360Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96361Date: Mon Nov 26 16:52:41 2007 -0500
96362
96363 registry: add missing include statement.
96364
96365commit 996b621bec1bbc4fb21970c75eaec62053bc6ccb
96366Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96367Date: Mon Nov 26 15:59:44 2007 -0500
96368
96369 registry: swap out the DTRACE XErrorDB stuff for the new registry call.
96370
96371commit 54cb729ecc2d366c1af836cb3d2ffc8e864e9b79
96372Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96373Date: Mon Nov 26 15:59:01 2007 -0500
96374
96375 registry: Add a call for DTRACE compatibility.
96376
96377commit decd5a7c605e42c99b6a4523c8e1833b859d9b24
96378Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96379Date: Mon Nov 26 15:26:49 2007 -0500
96380
96381 registry: Rebase registry to use the server config file of protocol names.
96382
96383commit 9b0e72c8d960d056276f5fa93f3cc2872825711e
96384Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96385Date: Mon Nov 26 15:26:04 2007 -0500
96386
96387 registry: Add a great big list of protocol names, like the XErrorDB that
96388 ships with Xlib. This is considered temporary, until server-side XCB can
96389 solve the problem programmatically.
96390
96391commit c0f9e204baf0218466973868c5ea6ed0f78e6b8b
96392Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96393Date: Mon Nov 26 15:24:15 2007 -0500
96394
96395 registry: rename the SERVERCONFIGdir and relocate it to /usr/lib/xorg
96396 by default.
96397
96398commit a344920ae86c1970e4cc34ee91e2f2008d490c49
96399Author: Adam Jackson <ajax@redhat.com>
96400Date: Mon Nov 26 11:53:08 2007 -0500
96401
96402 Allow Virtual to be specified globally in the Screen section.
96403
96404 The Display subsections are optional, and it's confusing to need to create
96405 them just to set a Virtual size.
96406
96407commit c6c284e64b1f537a3243856cf78cf3f2324e4c2b
96408Author: Matthias Hopf <mhopf@suse.de>
96409Date: Mon Nov 26 15:38:20 2007 +0100
96410
96411 Initialize Mode with 0 in xf86RandRModeConvert.
96412
96413 Asking for trouble if non-initialized values contain random data.
96414
96415commit a4edfbef022f9635c2c9b9eb229cb622834dc68c
96416Author: Peter Hutterer <peter@cs.unisa.edu.au>
96417Date: Sat Nov 24 16:16:48 2007 +1030
96418
96419 mi: only call UpdateSpriteForScreen if the screen actually changed.
96420
96421 If we call it unconditionally, we flip the dev->spriteInfo->sprite permanently
96422 when using XTestFakeInput (once in CheckMotion as called from the
96423 processInputProc, another time in UpdateSpriteForScreen when we actually warp
96424 the cursor). USFS also updates to the screen's root window unconditionally,
96425 which is not really what we want if we haven't changed screen at all.
96426
96427commit bf3198c8c56289244c58d36c6869442479fd3481
96428Author: Peter Hutterer <peter@cs.unisa.edu.au>
96429Date: Sat Nov 24 15:00:57 2007 +1030
96430
96431 dix: fix typo
96432
96433commit 5dabe448bda68a483bf444a4adfed2b25b30f600
96434Author: Peter Hutterer <peter@cs.unisa.edu.au>
96435Date: Sat Nov 24 15:00:43 2007 +1030
96436
96437 dix: Add special treatment of NotifyUngrab for enter/leave events.
96438
96439 In some cases (e.g. using x2x) the previous model broke, with a window ending
96440 not counting down to 0 anymore. Special treatment for NotifyUngrab seems to
96441 help here.
96442
96443 Current solution: If a NotifyGrab is to be sent ignore it. If a NotifyUngrab
96444 enter is sent, only adjust the semaphore if it is on 0. Likewise, do the same
96445 for a NotifyUngrab leave if the semaphore is on 1. This seems to work alright
96446 so far.
96447
96448commit 33b94da6327d3423b4ebc1a58d5894c9904e67c9
96449Author: Keith Packard <keithp@keithp.com>
96450Date: Fri Nov 23 16:01:11 2007 -0800
96451
96452 Re-add call to XFixesExtensionInit for static servers.
96453
96454 This reverts a portion of bcbaf2a0ce34b6c5e41d2831b8b87dbd0617a89b which
96455 removed the call to XFixesExtensionInit and some cpp lines.
96456
96457commit f6401f944d327cc5d9a7ee0bbdf4f7fc8eaa31e8
96458Author: Matthias Hopf <mhopf@suse.de>
96459Date: Fri Nov 23 16:12:49 2007 +0100
96460
96461 Don't segfault if referring to a relative output where no modes survived.
96462
96463commit fa19e84714aa84a2f2e817e363d6440349d0b619
96464Author: Matthias Hopf <mhopf@suse.de>
96465Date: Tue Nov 20 16:54:50 2007 +0100
96466
96467 Fix initial placement of LeftOf and Above.
96468
96469commit 184e571957f697f2a125dc9c9da0c7dfb92c2cd9
96470Author: Matthias Hopf <mhopf@suse.de>
96471Date: Tue Nov 20 13:05:26 2007 +0100
96472
96473 Adjust offsets of modes that do not fit virtual screen size.
96474
96475 Fixes memory corruption if a too small "Virtual" was specified in xorg.conf
96476 for the selected multi-monitor configuration.
96477
96478commit a80e64f1503a4d8b11c4a6608d296422c69e3e8b
96479Author: Peter Hutterer <peter@cs.unisa.edu.au>
96480Date: Sat Nov 17 22:50:07 2007 +0100
96481
96482 XKB: Generate correct key repeat events (bug #13114)
96483
96484 Make sure we send the correct event for the type of device when we're
96485 sending key repeat events, which stops repeats being sent to incorrect
96486 windows.
96487
96488commit 63351df0eec320aa3ce27d4d2ee6bcdb58aa2d92
96489Author: Jeremy Huddleston <jeremy@tifa.local>
96490Date: Thu Nov 22 18:02:07 2007 -0800
96491
96492 Darwin: Fix compilation/linking problems
96493
96494commit bf4ef4da759c01e6794ed28ba4988a2c8ee049bf
96495Author: Jeremy Huddleston <jeremy@tifa.local>
96496Date: Thu Nov 22 17:45:15 2007 -0800
96497
96498 Darwin: Remove the PaintWindow optimization which snuck back in.
96499
96500commit 2082e7aa878fe1221fd50895a9de1f408b3157a8
96501Author: Jeremy Huddleston <jeremy@tifa.local>
96502Date: Thu Nov 22 17:18:48 2007 -0800
96503
96504 Rootless: Remove the PaintWindow optimization which snuck back in.
96505
96506commit a751bc12bee1d4d2ed35e3a0c64d9c8c9bf30a82
96507Author: Jeremy Huddleston <jeremy@tifa.local>
96508Date: Thu Nov 22 13:53:00 2007 -0800
96509
96510 Rootless: Imported changes made in xorg-server-1.2-apple branch
96511
96512commit 59c7ca6586e7c20e28ad407ca9a0883c4d621d64
96513Author: Jeremy Huddleston <jeremy@tifa.local>
96514Date: Thu Nov 22 13:29:15 2007 -0800
96515
96516 Darwin: Added missing Makefile.am
96517
96518commit 23596291c30a85e38c00aef2c01b46d561e2916e
96519Author: Jeremy Huddleston <jeremy@tifa.local>
96520Date: Thu Nov 22 13:17:44 2007 -0800
96521
96522 Darwin: More syncing witn xorg-server-1.2-apple
96523
96524commit 4e18c626350c7c2e0fb540aa64a98957699f3abe
96525Author: Jeremy Huddleston <jeremy@tifa.local>
96526Date: Thu Nov 22 12:21:59 2007 -0800
96527
96528 Rootless: Pulled in changes from fb{Blt,Fill} into rl{Blt,Fill}
96529 (cherry picked from commit 3f857e129df7ce492191e0c51b8e53eaf6179366)
96530 (cherry picked from commit 70374a58937d7a6f01c210bd6ac66cafb63e895a)
96531
96532commit ed9524d36e42a310bb128284f2b507f76b8c40d9
96533Author: Jeremy Huddleston <jeremy@tifa.local>
96534Date: Thu Nov 22 01:07:02 2007 -0800
96535
96536 Darwin: Copied over missing file (Localizable.strings) from xorg-server-1.2-apple
96537
96538commit 4d9cef197b12548e0716dab3557e48311519e325
96539Author: Jeremy Huddleston <jeremy@tifa.local>
96540Date: Thu Nov 22 00:35:09 2007 -0800
96541
96542 Darwin: Misc cleanups to line up with xorg-server-1.2-apple
96543
96544commit 5e950123daa167c9ffe289b3bd89e3bd288da0e3
96545Author: Jeremy Huddleston <jeremy@tifa.local>
96546Date: Wed Nov 21 23:32:00 2007 -0800
96547
96548 Darwin: Removed cvs tags from Xquartz man page
96549
96550commit 3a2f714eea475a13cde65921e24c7ee3f70ffc3c
96551Author: Jeremy Huddleston <jeremy@tifa.local>
96552Date: Wed Nov 21 23:30:37 2007 -0800
96553
96554 Darwin: Removed .svn dir
96555
96556commit 691da031319dc59b9496101358c267f317abfd1e
96557Author: Peter Hutterer <peter@cs.unisa.edu.au>
96558Date: Thu Nov 22 17:44:39 2007 +1030
96559
96560 Xi: allocate motion history for master device.
96561
96562 We're still missing out on the actual content of the history but at least this
96563 way we don't segfault.
96564
96565commit 602de4f70b6f4aab93b514f3a01917bd5d4ad640
96566Author: Jeremy Huddleston <jeremy@tifa.local>
96567Date: Wed Nov 21 16:53:10 2007 -0800
96568
96569 Darwin: Use UTF8String since lossyCString is deprecated
96570 (cherry picked from commit 1786f9464af51ff606a612aec6fe420fa9688a28)
96571
96572commit 13666e287c347aab2a5e9d8ee5f6bb29b9b85171
96573Author: Ben Byer <bbyer@bbyer.local>
96574Date: Sun Nov 18 17:44:12 2007 -0800
96575
96576 Darwin: Added some DEBUG_LOG sauce to the XP_EVENT handling code
96577 (cherry picked from commit ec84a4cef66a2b46ed71f9758c434ea629d2f270)
96578
96579commit 829b6641bd64c352e1e8a7c619f84dedbdb07a09
96580Author: Ben Byer <bbyer@bbyer.local>
96581Date: Sun Nov 18 17:43:40 2007 -0800
96582
96583 Darwin: Disabled ALT_IS_MODE_SWITCH
96584 (cherry picked from commit fd181254f85543558190140787dc7b41f6cf90db)
96585
96586commit 8486f8af91b477c7bcb8438a0e9a72d0c11d1d63
96587Author: Ben Byer <bbyer@bbyer.local>
96588Date: Thu Nov 15 02:25:50 2007 -0800
96589
96590 Darwin: Added a lightweight debugging facility to support troubleshooting
96591 (for example) the stuck modifier key issue
96592 (cherry picked from commit 0e0b452d10c0af55497c3299b5f3db45d5b381cb)
96593
96594commit 74214a9f42b931f99d83ddb4efb3720881a2de16
96595Author: Ben Byer <bbyer@bbyer.local>
96596Date: Thu Nov 15 00:56:54 2007 -0800
96597
96598 Darwin: Patch to avert (some) damage / rootless crashes, courtesy of Ken Thomases
96599 (cherry picked from commit 148a87ff20aa5e7a6d839610aa14fa1a31505c4a)
96600
96601commit f5f833b80609f1f98c93113183bd2b1bab3bfec9
96602Author: Ben Byer <bbyer@bbyer.local>
96603Date: Sun Nov 11 04:30:34 2007 -0800
96604
96605 Darwin: These changes are necessary, yet not sufficient, to get 8-bit indexed
96606 color mode working in Xquartz.
96607 (cherry picked from commit a415f62f5289fae99ea9b0038d21fad7695b1336)
96608
96609commit 8358334180a4f8c1e73fc5647a62bcd3539dee45
96610Author: Ben Byer <bbyer@bbyer.local>
96611Date: Sun Nov 11 04:30:00 2007 -0800
96612
96613 Darwin: Fixed the call to xp_init so that we now receive Motion notifications even
96614 if X is not the active application.
96615 fixes <rdar://problem/5167664> xeyes dead until window activation
96616 (cherry picked from commit c7573379a85a1480cc51650075078e41dafe56af)
96617
96618commit 512dee90878e552ad1b2bb5b27366707f6464f28
96619Author: Ben Byer <bbyer@bbyer.apple.com>
96620Date: Thu Nov 8 22:17:38 2007 -0800
96621
96622 Darwin: fix for spurious "Are you sure you want to quit?" message
96623 (cherry picked from commit 30cbfc786e4fedda3fe070bacceabe1d9212d00b)
96624
96625commit 28e73e99a9a59223963312c5dd43ce5566d1db9d
96626Author: Ben Byer <bbyer@bbyer.apple.com>
96627Date: Thu Nov 8 22:12:41 2007 -0800
96628
96629 Darwin: Adding "fake RandR" support from old X11.app
96630 (cherry picked from commit 633490c4e8dab30af7ecbe1bef076c22ad5f5da9)
96631
96632commit 01b70afaac0990b41d1fb6fadbfd64df1486b669
96633Author: Ben Byer <bbyer@bbyer.local>
96634Date: Sat Nov 3 05:34:19 2007 -0700
96635
96636 Darwin: Initial support for Spaces -- if you use Expose to drag an X11
96637 window to another Space, it will work correctly (as opposed
96638 to just leaving a ghost window). We accomplish this by listening
96639 for the notification from Xplugin that our window has been moved,
96640 and then we ask X11 to move the window to the new location.
96641 (cherry picked from commit 2d50ea8013e7c1639d570e227b53b037fb567565)
96642
96643commit b39edc01a6588697b65f831e8ab1dbb24cbe7b24
96644Author: Ben Byer <bbyer@bbyer.local>
96645Date: Wed Oct 31 23:46:50 2007 -0700
96646
96647 Darwin: Swap modifier keys for buttons 2 and 3 -- now Option-click is the middle click
96648 (cherry picked from commit 0aa61293b62aeb69a93b2035d0aef8644343eed3)
96649
96650commit 606a8dc73d91a198d72d249934dc027a23f4c338
96651Author: Ben Byer <bbyer@bbyer.local>
96652Date: Wed Oct 31 03:39:47 2007 -0700
96653
96654 Darwin: Trap Deactivate messages and release modifiers to avoid "stuck shift lock" (etc) bugs
96655 (cherry picked from commit 2b189a99330eb465fa0d17020fb1db1e38829151)
96656
96657commit 4c18ef4331aaee268431a3ba50991f0312b82870
96658Author: Ben Byer <bbyer@bbyer.local>
96659Date: Wed Oct 31 03:22:18 2007 -0700
96660
96661 Darwin: Workaround for a bug where the holding down Command to make a "fake"
96662 button 2 click would actually result in a Command-2 chord.
96663 (I.e. it wasn't releasing Command before clicking the fake button.)
96664 (cherry picked from commit 0d5dd5dffa4c5ce3f54dfe53720a39d524dc8e37)
96665
96666commit f9269bebae27bbc9d0e03e02943166b83946623d
96667Author: Peter Hutterer <peter@cs.unisa.edu.au>
96668Date: Thu Nov 22 17:30:14 2007 +1030
96669
96670 DeviceIntRec: move lastx/lasty from valuator into DeviceIntRec.
96671
96672 We free the ValuatorClassRec quite regularly. If a SIGIO is handled while
96673 we're swapping device classes, we can bring the server down when we try to
96674 access lastx/lasty of the master device.
96675
96676commit 7f2972d47a5d74fe92268c6d609b1eb6ad845824
96677Author: Jeremy Huddleston <jeremy@tifa.local>
96678Date: Wed Nov 21 21:59:59 2007 -0800
96679
96680 Darwin: Really add launcher this time
96681
96682commit bcbaf2a0ce34b6c5e41d2831b8b87dbd0617a89b
96683Author: Jeremy Huddleston <jeremy@tifa.local>
96684Date: Wed Nov 21 19:51:14 2007 -0800
96685
96686 Darwin: Dead code removal, Code cleanup, Added launcher
96687 Imported changes from xorg-server-1.2-apple to make master more current wrt
96688 file layout, build system changes, and dead code removal.
96689
96690commit 9ed43eff48201520797f89a12bb3b2f5819bd99f
96691Author: Peter Hutterer <peter@cs.unisa.edu.au>
96692Date: Wed Nov 21 17:16:27 2007 +1030
96693
96694 Xi: allow clients to specify pure client id in SetClientPointer.
96695
96696 If no window was found with the given ID, try if there's a client with the ID.
96697 Allows us to set the CP for apps that don't open windows (e.g. x2x).
96698
96699commit 33f15689922ad9f1193f803bc1636c82c23ab99e
96700Author: Peter Hutterer <peter@cs.unisa.edu.au>
96701Date: Wed Nov 21 16:06:44 2007 +1030
96702
96703 xtest: switch an inputInfo.pointer over to PickPointer.
96704
96705 Couple of whitespace fixes too.
96706
96707commit e5dd7a95791748c57cab75c0d8ba9d37f72edccf
96708Author: Peter Hutterer <peter@cs.unisa.edu.au>
96709Date: Wed Nov 21 15:59:31 2007 +1030
96710
96711 Xext: Scruffy the janitor don't like no "register" keywords.
96712
96713commit a55ec1a9f4b62139dc5e5462d79d47b330c27c79
96714Author: Alan Coopersmith <alan.coopersmith@sun.com>
96715Date: Tue Nov 20 18:31:03 2007 -0800
96716
96717 Restore checks for __i386 where needed for Sun compilers on Solaris
96718
96719commit 2f387d913aa76f1b6d21d8e2698be165301c6bc1
96720Author: Alan Coopersmith <alan.coopersmith@sun.com>
96721Date: Tue Nov 20 18:27:12 2007 -0800
96722
96723 Enable use of /dev/urandom on Solaris as well
96724
96725commit 4363d70c6b420648b501126d1fbdebfafc7ae09f
96726Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96727Date: Tue Nov 20 18:58:55 2007 -0500
96728
96729 registry: Fix some mistakes in the reversion of prior commits.
96730
96731commit 140a4660aca1c283613d5b62f51668b44b45baf6
96732Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96733Date: Tue Nov 20 18:49:30 2007 -0500
96734
96735 Revert "registry: Register XTrap extension protocol names."
96736
96737 This reverts commit b38a91993364aa80cfd99721e319e1458d9fb760.
96738
96739 Moving all the names into dix/registry.c
96740
96741commit ed8a39c48ab9dac085fcf58b9641364b5608f3f4
96742Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96743Date: Tue Nov 20 18:47:52 2007 -0500
96744
96745 Revert "registry: Register XKB extension protocol names."
96746
96747 This reverts commit a5cf3f21f712e46dbf9bca289e67be75f2b531d3.
96748
96749 Moving all the names into dix/registry.c
96750
96751commit 17b0c729b553e2f0f8f82497698b282a47db3326
96752Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96753Date: Tue Nov 20 18:46:43 2007 -0500
96754
96755 registry: Remove registry code from XInput extension.
96756
96757 Moving all the names into dix/registry.c
96758
96759commit e86852aff62a861823b8e419434e0401b8cdc8e0
96760Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96761Date: Tue Nov 20 18:44:56 2007 -0500
96762
96763 Revert "registry: Register XFixes extension protocol names."
96764
96765 This reverts commit 106758893b68033f14f69c4ee6591fb6a149ba37.
96766
96767 Moving all the names into dix/registry.c
96768
96769commit 5269da2bde3cf4feb12fa2bd87bff6ee6d8730a1
96770Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96771Date: Tue Nov 20 18:43:38 2007 -0500
96772
96773 Revert "registry: Register XvMC extension protocol names."
96774
96775 This reverts commit 853ea337bdad17f8f6ec7d940de14ce2cbbbf93e.
96776
96777 Moving all the names into dix/registry.c
96778
96779commit 03a86c8d5e20a6e47f3c294f0087f205cf2a72dd
96780Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96781Date: Tue Nov 20 18:42:19 2007 -0500
96782
96783 Revert "registry: Register Xv extension protocol names."
96784
96785 This reverts commit 12766c5b5ffdab95255a63b2c8421ee773fd43b5.
96786
96787 Moving all the names into dix/registry.c
96788
96789commit edcf490cdb965e2a5bfc0169c01732d2924da3ae
96790Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96791Date: Tue Nov 20 18:41:10 2007 -0500
96792
96793 registry: Remove registry code from XTest extension.
96794
96795 Moving all the names into dix/registry.c
96796
96797commit 5fea1ed50f37691a5273bf2897479781de808ff5
96798Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96799Date: Tue Nov 20 18:39:48 2007 -0500
96800
96801 registry: Remove registry code from SELinux extension.
96802
96803 Moving all the names into dix/registry.c
96804
96805commit 9a8af33718d085656a672e4c27df200485c84154
96806Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96807Date: Tue Nov 20 18:38:24 2007 -0500
96808
96809 Revert "registry: Register Resource extension protocol names."
96810
96811 This reverts commit 5c8b1a91726817816d20faefad21c7a68ab634cc.
96812
96813 Moving all the names into dix/registry.c
96814
96815commit e6023e0208fae8f19c566f9df1a8aa20494f40ab
96816Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96817Date: Tue Nov 20 18:36:49 2007 -0500
96818
96819 Revert "registry: Register XPrint extension protocol names."
96820
96821 This reverts commit f077578e42eee424b0e534774574c84af9d6f85b.
96822
96823 Moving all the names into dix/registry.c
96824
96825commit 277345fb7065d74c3b0d076382affb78cbe67569
96826Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96827Date: Tue Nov 20 18:35:57 2007 -0500
96828
96829 registry: Remove registry code from XF86Bigfont extension.
96830
96831 Moving all the names into dix/registry.c
96832
96833commit bf27edd365ffd275e5453f44d130eeacbfe0ecd9
96834Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96835Date: Tue Nov 20 18:34:14 2007 -0500
96836
96837 Revert "registry: Register EVIE extension protocol names."
96838
96839 This reverts commit 48891d5696f56711f23743cb03be39cf6b26c522.
96840
96841 Moving all the names into dix/registry.c
96842
96843commit 687427179420b18a55a1a02b8a9f2a32ea8eac8d
96844Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96845Date: Tue Nov 20 18:32:54 2007 -0500
96846
96847 registry: Remove registry code from XC-MISC extension.
96848
96849 Moving all the names into dix/registry.c
96850
96851commit 4b0274e8f712e51b18618a2a0bdbe03b17b9736b
96852Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96853Date: Tue Nov 20 18:25:15 2007 -0500
96854
96855 Revert "registry: Register SYNC extension protocol names."
96856
96857 This reverts commit 9f597f6c87e0b14cc382d8e5929e42f822db4329.
96858
96859 Moving all the names into dix/registry.c
96860
96861commit 4c7cf5aa4c802dcde895c723879a80a87620c0f7
96862Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96863Date: Tue Nov 20 18:23:57 2007 -0500
96864
96865 Revert "registry: Register SHM extension protocol names."
96866
96867 This reverts commit 2c9646ad4e65bb061d910c9e2b1a8a978f21fa17.
96868
96869 Moving all the names to dix/registry.c
96870
96871commit 67e82e306f67a215c6c89868cc1d3649747bd93d
96872Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96873Date: Tue Nov 20 18:22:59 2007 -0500
96874
96875 Revert "registry: Register SHAPE extension protocol names."
96876
96877 This reverts commit 4e274e90e16b1d954391e1af3e2074fb10f70ee7.
96878
96879 Moving all the names to dix/registry.c
96880
96881commit 8583bf78ad056ffe2d83b54e5c9a0a217e425a7b
96882Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96883Date: Tue Nov 20 18:21:09 2007 -0500
96884
96885 registry: Remove registry code from XC-SECURITY extension.
96886
96887 Moving all the names to dix/registry.c
96888
96889commit 55744d8e5d7bf1ff27cd25de54e14e799dd1a70a
96890Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96891Date: Tue Nov 20 18:19:44 2007 -0500
96892
96893 Revert "registry: Register MIT-SCREEN-SAVER extension protocol names."
96894
96895 This reverts commit 58c3240fcbec23aad122e1c340f6bb6d3b18f779.
96896
96897 Moving all the names into dix/registry.c
96898
96899commit 36ef45928c783292cef18acfdd83ae057826c989
96900Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96901Date: Tue Nov 20 18:18:01 2007 -0500
96902
96903 registry: Remove registry code from MIT-MISC extension.
96904
96905 Moving all the names to dix/registry.c
96906
96907commit 816e6e612e4bc3cea1e67e7ea79d5b640458011f
96908Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96909Date: Tue Nov 20 18:15:37 2007 -0500
96910
96911 Revert "registry: Register Multibuffer extension protocol names."
96912
96913 This reverts commit 3877faf7d9fe00ed634077e38a198ae4b91a2bb4.
96914
96915 Moving all the names into dix/registry.c
96916
96917commit 40a0da044e911ea51de003f3621331ffbe2842bc
96918Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96919Date: Tue Nov 20 18:13:43 2007 -0500
96920
96921 registry: Remove registry code from Fontcache extension.
96922
96923 Moving all the names into dix/registry.c
96924
96925commit 46412baf60ed639ddc1d5fb601f73a75e39737f7
96926Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96927Date: Tue Nov 20 18:11:06 2007 -0500
96928
96929 registry: Remove registry code from EVI extension.
96930
96931 Moving all the names into dix/registry.c
96932
96933commit 460c43032f05aad3f0f552901a52d199f61c7f4f
96934Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96935Date: Tue Nov 20 18:08:18 2007 -0500
96936
96937 registry: Remove registry code from DPMS extension.
96938
96939 Moving all the names into dix/registry.c
96940
96941commit 76e89d45b497d4afa4e60e1d0ec50b62f54f6b88
96942Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96943Date: Tue Nov 20 18:06:40 2007 -0500
96944
96945 registry: Remove registry code from TOG-CUP extension.
96946
96947 Moving all the names into dix/registry.c
96948
96949commit ce93c5772da52ab88faef7e5b661b681d5b60b1e
96950Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96951Date: Tue Nov 20 18:03:57 2007 -0500
96952
96953 registry: Remove registry code from BigRequests extension.
96954
96955 Moving all the names into dix/registry.c
96956
96957commit 0756d1271209e6ae14cc641dddca095271b43150
96958Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96959Date: Tue Nov 20 17:59:40 2007 -0500
96960
96961 Revert "registry: Register APPGROUP extension protocol names."
96962
96963 This reverts commit b504678ba5407a6fd8d47d051305f7c3d5606dfe.
96964
96965 Moving all the names into dix/registry.c
96966
96967commit 5aff37d1d69be493727856a29628bd782d50b90f
96968Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96969Date: Tue Nov 20 17:57:06 2007 -0500
96970
96971 Revert "registry: Register RENDER extension protocol names."
96972
96973 This reverts commit 8964c6d8e14ae47798762191e359b2bf138ca32e.
96974
96975 Moving all the names into dix/registry.c
96976
96977commit e585a2ddb495b50a53e15cccc368ca0858fc9d23
96978Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96979Date: Tue Nov 20 17:55:47 2007 -0500
96980
96981 Revert "registry: Register Record extension protocol names."
96982
96983 This reverts commit ea09c9acc8f0d5577f54c864ff88b7f03d93b2f4.
96984
96985 Moving all the names into dix/registry.c
96986
96987commit d4577e485367468227e031eb434b739eff7b5e9a
96988Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96989Date: Tue Nov 20 17:51:27 2007 -0500
96990
96991 Revert "registry: Register RANDR extension protocol names."
96992
96993 This reverts commit c827db57e4d9ca14c82b099dcfc9b7a0c0b5ba0a.
96994
96995 Moving all the names into dix/registry.c
96996
96997commit a541e826c9310d3051e53834833c6c3a08654148
96998Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
96999Date: Tue Nov 20 17:50:26 2007 -0500
97000
97001 Revert "registry: Register WINDOWSWM extension protocol names."
97002
97003 This reverts commit 4c3285c883cc50a91bc5262bbc9d073d816f860a.
97004
97005 Moving all the names into dix/registry.c
97006
97007commit 993595430bd0580ab4d936be6b70fb91b8bb1d16
97008Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97009Date: Tue Nov 20 17:48:46 2007 -0500
97010
97011 Revert "registry: Register XF86DRI extension protocol names."
97012
97013 This reverts commit b7786724080fd3928ef7b8c294346661d7ffd90b.
97014
97015 Moving all the names into dix/registry.c
97016
97017commit 6b73c215c9d612534af290230b2e914d42d819cd
97018Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97019Date: Tue Nov 20 17:47:30 2007 -0500
97020
97021 Revert "registry: Register XF86VidMode extension protocol names."
97022
97023 This reverts commit 960677e876c068400fb45e1764bb5470cd8c389f.
97024
97025 Moving all the names into dix/registry.c
97026
97027commit 8e2cd7a804664bbd2d03789dcd5c93223122e929
97028Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97029Date: Tue Nov 20 17:45:30 2007 -0500
97030
97031 Revert "registry: Register XF86Misc extension protocol names."
97032
97033 This reverts commit 2cd1b32b77e0ceeaccb3f01c4ac13a97c557668c.
97034
97035 Moving all the names into dix/registry.c
97036
97037commit 0356153a58cef87d655bccacd8e2cf03d577bd19
97038Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97039Date: Tue Nov 20 17:43:18 2007 -0500
97040
97041 Revert "registry: Register XF86DGA extension protocol names."
97042
97043 This reverts commit 3815284e899b61731b6a63c4ba14c5d773e24eb6.
97044
97045 Moving all the names into dix/registry.c
97046
97047commit de93c1e9df14577e158b6dc3ccec7ee48f592386
97048Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97049Date: Tue Nov 20 17:40:57 2007 -0500
97050
97051 Revert "registry: Register DMX extension protocol names."
97052
97053 This reverts commit 32f3f5a1e7654f8bb43ea16b9227b3994e616739.
97054
97055 Moving all the names into dix/registry.c
97056
97057commit 2d3e0cdf4bd7ab069bad7244ede7c2d489e92b17
97058Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97059Date: Tue Nov 20 17:39:56 2007 -0500
97060
97061 Revert "registry: Register APPLEDRI extension protocol names."
97062
97063 This reverts commit 3464b419230c6d17e940d967b567c5d2cb22d232.
97064
97065 Moving all the names into dix/registry.c
97066
97067commit 546d46224e355d4f00232da5538548e3c8853e40
97068Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97069Date: Tue Nov 20 17:37:48 2007 -0500
97070
97071 Revert "registry: Register XINERAMA extension protocol names."
97072
97073 This reverts commit b9f5ab98c8dea36dcce1ad15fd2e059a77e77c39.
97074
97075 Moving all the names into dix/registry.c
97076
97077commit fd2d83d5bf5b35c8a2b05f725486be166783921e
97078Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97079Date: Tue Nov 20 17:34:48 2007 -0500
97080
97081 Revert "registry: Register APPLEWM extension protocol names."
97082
97083 This reverts commit eee46b4681ec55297604b0425705f2b18381f7ca.
97084
97085 Moving all the names into dix/registry.c
97086
97087commit c934e1af27189571c1e7dd838872e380c3580eeb
97088Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97089Date: Tue Nov 20 17:32:35 2007 -0500
97090
97091 Revert "registry: Register DBE extension protocol names."
97092
97093 This reverts commit 2e1e5be1d9067816525aa13a1d818e8ca6899599.
97094
97095 Moving all the names into dix/registry.c
97096
97097commit b9ab6f300a46aa8879b11eac51857357cc379c2f
97098Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97099Date: Tue Nov 20 17:31:00 2007 -0500
97100
97101 Revert "registry: Register DAMAGE extension protocol names."
97102
97103 This reverts commit 20db50b4c44a14f7eeac2b1de17ada68482521da.
97104
97105 Moving all the names into dix/registry.c
97106
97107commit 26586a7ad5e999b34996d147fb43998deea89178
97108Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97109Date: Tue Nov 20 17:27:37 2007 -0500
97110
97111 Revert "registry: Register composite extension protocol names."
97112
97113 This reverts commit 166ef972febc00c665e1d5aeb68e75d7bbcf9879.
97114
97115 Moving all the names into dix/registry.c
97116
97117commit fae39db7957c0ebdc7af36f8d8f484473beb6d38
97118Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97119Date: Tue Nov 20 15:21:31 2007 -0500
97120
97121 devPrivates rework: put back some changes that were mistakenly removed
97122 during merge conflict resolution.
97123
97124commit 709c1a70c8c6a9e132bf0d92f78a12be72beee51
97125Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97126Date: Tue Nov 20 15:18:02 2007 -0500
97127
97128 Remove some duplicate include statements.
97129
97130commit be0cbe5a330f62cef47fffbc49e83b5e1637b7d0
97131Author: Dodji Seketeli <dodji@openedhand.com>
97132Date: Tue Nov 20 15:39:49 2007 +0100
97133
97134 kaa: update kaaCreatePixmap to support the new usage_int
97135
97136commit 5b0dfb73ca4699cc4b33720f10416de7440081b7
97137Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97138Date: Mon Nov 19 21:01:22 2007 -0500
97139
97140 devPrivates rework: put back a comment that was mistakenly removed
97141 during merge conflict resolution.
97142
97143commit 2d17f47cc7d6b174857617d31ad1b437d8e97d94
97144Merge: 60be452 ea9c63e
97145Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97146Date: Mon Nov 19 18:10:46 2007 -0500
97147
97148 Merge branch 'master' into XACE-SELINUX
97149
97150 Conflicts:
97151
97152 hw/xnest/Pixmap.c
97153 include/dix.h
97154
97155commit 60be452c2e88342f92a76ba5ec7d90b5b0211aaf
97156Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97157Date: Mon Nov 19 16:55:09 2007 -0500
97158
97159 xace: restore the old SaveScreens function and introduce new API, since the
97160 old version is called from drivers...
97161
97162commit a95bb52b4366d85fc049130c60af5c9e727c565b
97163Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97164Date: Mon Nov 19 16:34:38 2007 -0500
97165
97166 devPrivates rework: add missing include of dix/privates.h
97167
97168commit ea9c63e93b9bb731796e8a8de2d127e6cc720076
97169Author: Michel Dänzer <michel@tungstengraphics.com>
97170Date: Mon Nov 19 15:53:49 2007 +0100
97171
97172 DEFAULT_DPI was undefined here.
97173
97174commit db9ae863536fff80b5463d99e71dc47ae587980d
97175Author: Adam Jackson <ajax@aspartame.nwnk.net>
97176Date: Sun Nov 18 11:57:01 2007 -0500
97177
97178 Bump DEFAULT_DPI to 96.
97179
97180 75 is just nonsense.
97181
97182commit ee2d4626dca6e0d4fc6f524e5de4bdefa2ed43df
97183Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net>
97184Date: Sun Nov 18 11:56:31 2007 -0500
97185
97186 Bug #12932: Use DEFAULT_DPI in randr1.2 instead of hardcoded 96.
97187
97188commit a46c30c3be33ffb304a885503c8aaa78396ed3d9
97189Author: Jernej Azarija <azi.stdout@gmail.com>
97190Date: Sun Nov 18 11:44:36 2007 -0500
97191
97192 Bug #12531: RRModesForScreen can fail to allocate.
97193
97194commit fac7e7e4e1809e865b9b3cf5b7eb69ba9d3a3759
97195Author: Adam Jackson <ajax@aspartame.nwnk.net>
97196Date: Sun Nov 18 11:39:26 2007 -0500
97197
97198 Document the requirement for interleaved code and declarations.
97199
97200commit bad96e5a864e40fbd47265d7fb6eaa67c55fac11
97201Author: Peter Hutterer <peter@cs.unisa.edu.au>
97202Date: Sun Nov 18 15:02:24 2007 +1030
97203
97204 dix: getevents.c remove trailing whitespaces.
97205
97206commit 75e5e9f88c5f974d996dd81e672cd3709d361a44
97207Author: Peter Hutterer <peter@cs.unisa.edu.au>
97208Date: Sat Nov 17 13:46:01 2007 +1030
97209
97210 Xi: add missing XI_DeviceClassesChangedMask to XI filters array.
97211
97212commit d15339a92c4d689d2ab8a86e4f10107f3e45eff8
97213Author: Adam Jackson <ajax@aspartame.nwnk.net>
97214Date: Sat Nov 17 22:12:10 2007 -0500
97215
97216 Bug #9725: Don't look in root's $HOME for config files, that's just confusing.
97217
97218commit 748cfbc820f8cdeb544c54a6db495fecf2e2457b
97219Author: Adam Jackson <ajax@aspartame.nwnk.net>
97220Date: Sat Nov 17 21:23:05 2007 -0500
97221
97222 Disinfect mi/ of mfb.
97223
97224commit a969db091cab16a448f82782e85b3dd19c81627a
97225Author: Daniel Stone <daniel@fooishbar.org>
97226Date: Sat Nov 17 22:34:47 2007 +0100
97227
97228 XKB: Don't ring the bell when we don't have a BellProc (bug #13246)
97229
97230commit c89b543198d5ec56ff025bdd6bb7229523478e58
97231Author: Ben Skeggs <skeggsb@gmail.com>
97232Date: Sat Nov 17 18:20:49 2007 +1000
97233
97234 exa: set driverPriv to NULL before it might get used later with garbage
97235
97236commit 514ba4ca727f0b1076bc67500617722203d34daa
97237Author: Adam Jackson <ajax@aspartame.nwnk.net>
97238Date: Fri Nov 16 19:53:11 2007 -0500
97239
97240 Bug #1612: Use a stronger PRNG.
97241
97242 Currently just reads from /dev/urandom, and only on Linux.
97243
97244commit 12dd6e9911de187e0ebee86434e8c24a67c990f9
97245Author: Peter Hutterer <peter@cs.unisa.edu.au>
97246Date: Fri Nov 16 17:23:09 2007 +1030
97247
97248 dix: reset MD's devPrivate classes to NULL before device initialisation.
97249
97250 XkbInitKeyboardDefviceStruct may call FatalError if it fails. FatalError then
97251 cleans up all the devices, resulting in a segfault if the pointer is
97252 uninitialised.
97253
97254commit be3321c2e9fad228a9ee5fef47680a47bc9e39a4
97255Author: Peter Hutterer <peter@cs.unisa.edu.au>
97256Date: Fri Nov 16 12:12:41 2007 +1030
97257
97258 dix: Free both current classes and original classes when closing an MD.
97259
97260commit 3c39dd19ec2a53b8854279e3b03131098031473a
97261Author: Peter Hutterer <peter@cs.unisa.edu.au>
97262Date: Fri Nov 16 12:12:14 2007 +1030
97263
97264 Xi: Deep-copy full list of FeedbackClasses.
97265
97266 All feedback classes are linked lists and the whole list has to be duplicated,
97267 not just the first entry.
97268
97269 Xkb stuff still missing.
97270
97271commit 497862df2fcd67531fbe0f876c20a09884ee74df
97272Author: Peter Hutterer <peter@cs.unisa.edu.au>
97273Date: Fri Nov 16 11:20:22 2007 +1030
97274
97275 dix: explicitly float all attached SDs before closing down devices.
97276
97277 Some drivers flush on shutdown, if our SD is still attached we'd be trying to
97278 route an event through a non-existing device.
97279
97280commit 9de1ebe2a80164507cbe2ef688f284225e0ec808
97281Author: Peter Hutterer <peter@cs.unisa.edu.au>
97282Date: Fri Nov 16 10:45:28 2007 +1030
97283
97284 dix: Fix up class restoring when last SD disconnects.
97285
97286 Old code was fundamentally broken, fixes now are:
97287 - free the MDs current device classes
97288 - copy the device classes instead of flipping the pointers
97289 - check for the old MD, not the new one.
97290
97291commit 20fd4783247b1b93d9675dc36768dd1ed59ba2d3
97292Author: Adam Jackson <ajax@redhat.com>
97293Date: Thu Nov 15 17:01:33 2007 -0500
97294
97295 Small static cleanups on dix/
97296
97297commit 70e50fa51f05663f289eeeea4521e737e8e24bca
97298Author: Adam Jackson <ajax@redhat.com>
97299Date: Thu Nov 15 17:00:37 2007 -0500
97300
97301 Allocate RRCrtcRecs with calloc.
97302
97303commit 8d0cd1cd2c57ee5a2fc4d577d8182d66369f0617
97304Author: Aaron Plattner <aplattner@nvidia.com>
97305Date: Thu Nov 15 12:16:36 2007 -0800
97306
97307 Fix a really dumb typo.
97308
97309commit f797c96845a3fab37cda6839ebecf9ac5401fd6e
97310Author: Aaron Plattner <aplattner@nvidia.com>
97311Date: Thu Nov 15 12:12:02 2007 -0800
97312
97313 Save pixmap allocation hints into the PixmapRec.
97314
97315commit 6bc50de02108f822977fc7545da81fce95ea7ff4
97316Author: Adam Jackson <ajax@redhat.com>
97317Date: Thu Nov 15 15:03:58 2007 -0500
97318
97319 Simplify system resource range setup.
97320
97321 osRes only existed to get copied into Acc. Waste of effort.
97322
97323commit e1ff14a9246e12d42ce8ca5afbe3b957333a5620
97324Author: Adam Jackson <ajax@redhat.com>
97325Date: Thu Nov 15 14:45:49 2007 -0500
97326
97327 Delete some dead code in X -configure.
97328
97329commit 01cfba75229f4b9bf1e4fe80814931acdacde14c
97330Author: Adam Jackson <ajax@redhat.com>
97331Date: Thu Nov 15 14:27:03 2007 -0500
97332
97333 Nuke the debugging allocator.
97334
97335 This has never been hooked up in the modular build, and can not possibly
97336 have built since before 6.7. Clearly no one's using it.
97337
97338commit c67b9c5fc33002b13a2360929a37f24169710f64
97339Author: Adam Jackson <ajax@redhat.com>
97340Date: Thu Nov 15 14:24:36 2007 -0500
97341
97342 Clean up some garbage in driver enumeration.
97343
97344commit 0706e5e790060fbf046cfaff295b78806b7841c6
97345Author: Adam Jackson <ajax@redhat.com>
97346Date: Thu Nov 15 14:15:09 2007 -0500
97347
97348 Eliminate some redundancy in autoconfiguration.
97349
97350 We already synthesize Monitor and Module sections for you, no need to
97351 specify them explicitly in the fake config buffer.
97352
97353commit 2c01a49bf0a407bd5510bb9ceb4ef86a2cc36be9
97354Author: Adam Jackson <ajax@redhat.com>
97355Date: Thu Nov 15 13:32:59 2007 -0500
97356
97357 Don't sleep(1) at exit on any platform.
97358
97359commit 83926cb8bef6288b89e801c5e60b3f40e923e16e
97360Author: Peter Hutterer <peter@cs.unisa.edu.au>
97361Date: Thu Nov 15 17:41:07 2007 +1030
97362
97363 Xi: Only alloc modifierKeyMap if <slave device>->maxKeysPerModifier > 0
97364
97365 Sometimes (e.g. on my debian ppc box) maxKeysPerModifier of the SD is 0. So we
97366 try to malloc(0), bringing the whole server down with a FatalError because it
97367 looks as if the malloc failed. This is bad, so only alloc if we actually have
97368 something to alloc.
97369
97370commit 070195dbf88eb121e65f802e023aa37ed1f2c2ac
97371Author: Peter Hutterer <peter@cs.unisa.edu.au>
97372Date: Thu Nov 15 16:27:18 2007 +1030
97373
97374 Xi: fix up sloppy class copying causing segfaults.
97375
97376commit 53539688cab990a7df1851d64f3ee4e11920a86b
97377Author: Peter Hutterer <peter@cs.unisa.edu.au>
97378Date: Thu Nov 15 16:23:48 2007 +1030
97379
97380 dix: SetKeySymMap should alloc the map if dst doesn't have one already.
97381
97382commit b40646dc104fb03ea7cc0b27fae573aecaab486e
97383Author: Peter Hutterer <peter@cs.unisa.edu.au>
97384Date: Thu Nov 15 15:43:44 2007 +1030
97385
97386 dix: Add FreeDeviceClass and FreeFeedbackClass for centralised xfree.
97387
97388 Ensures that we only have one way of freeing a device class to avoid leaks in
97389 ChangeMasterDeviceClasses and other places.
97390
97391commit 4c9cc82fc4461d180ae2c2fbe50e7f98b0777f91
97392Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
97393Date: Thu Nov 15 01:46:11 2007 -0200
97394
97395 For some reason "-nozap" appeared twice. Weird.
97396
97397commit 18833d648fd7e1a5e962b93636bbbb38aca9c454
97398Author: Peter Hutterer <peter@cs.unisa.edu.au>
97399Date: Thu Nov 15 12:13:59 2007 +1030
97400
97401 Xi: reset xkb-stuff to NULL after copying (DeepCopyDeviceClasses)
97402
97403 Having two devices point to the same xkb stuff causes SIGABRTs.
97404
97405 Also, don't init a MD's xkbInfo unless the SD has an xkbInfo.
97406
97407commit 1635832c1635374033686d3a943b77adbd60bb98
97408Author: Peter Hutterer <peter@cs.unisa.edu.au>
97409Date: Thu Nov 15 11:35:07 2007 +1030
97410
97411 Revert "xkb: disable xkb key repeats (temporarily)"
97412
97413 This reverts commit 2b1d946392ce28b96941341778b2b526aa0fb126.
97414
97415commit b05246696d14bd35aa53b49302707b51206c72a6
97416Author: Peter Hutterer <peter@cs.unisa.edu.au>
97417Date: Thu Nov 15 11:31:43 2007 +1030
97418
97419 Xi: free XkbSrvLedInfos as well when freeing an MD's device classes.
97420
97421commit c758e5a664a52045ad419340044beebb6774a336
97422Author: Peter Hutterer <peter@cs.unisa.edu.au>
97423Date: Thu Nov 15 10:47:20 2007 +1030
97424
97425 dix: Make sure core MDs copy their initial classes before they are used.
97426
97427 Anything in dev->key, dev->valuator etc. of a MD must always be a copy of the
97428 original class. The intial classes of an MD (the ones set up before an SD is
97429 attached) as well, as we may have to restore them if no SD is attached
97430 anymore.
97431
97432commit a08665d4d3b0a7d567a90bb9bbfe4abafd6f3887
97433Author: Peter Hutterer <peter@cs.unisa.edu.au>
97434Date: Thu Nov 15 10:42:29 2007 +1030
97435
97436 mi: don't call SwitchCoreKeyboard, we switch during event processing now.
97437
97438commit 64711a094893e83764bbeda538c6e877ebe2af79
97439Author: Peter Hutterer <peter@cs.unisa.edu.au>
97440Date: Thu Nov 15 10:41:34 2007 +1030
97441
97442 Xi: When switching MD classes, make a deep copy instead of pointer flip.
97443
97444 Turns out it's really really hard synchronising device state across multiple
97445 duplicated events if they all share the same struct. So instead of doing so,
97446 when the SD changes deep-copy all it's classes into the MD. The MD then has
97447 the same capabilities, but the state can be set separately. This should fix
97448 xkb, key state, repeat etc. problems.
97449
97450 Updating the device state allows us to remove the SwitchCoreKeyboard from the
97451 event gathering, it's all done during event processing now.
97452
97453commit 3dde66f96b9b8431381871cf85266da3ec57a0d4
97454Author: Adam Jackson <ajax@redhat.com>
97455Date: Wed Nov 14 15:10:59 2007 -0500
97456
97457 Start 1.4.99
97458
97459commit c3897ca7099fc007b4134a8fabd4c707f99f2ac7
97460Author: Adam Jackson <ajax@redhat.com>
97461Date: Fri Nov 9 13:55:32 2007 -0500
97462
97463 Add -pogo option for init/teardown performance testing.
97464
97465commit cecac794451b793871f297b91a11d3b52eeb6d1b
97466Author: Adam Jackson <ajax@redhat.com>
97467Date: Thu Nov 8 17:25:36 2007 -0500
97468
97469 Don't sleep(1) at server exit.
97470
97471commit 1603130236c55ddabc3854d4ba62d544debcf1f5
97472Merge: f207e69 f7dd0c7
97473Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97474Date: Wed Nov 14 13:35:50 2007 -0500
97475
97476 Merge branch 'master' into XACE-SELINUX
97477
97478 Conflicts:
97479
97480 Xext/xace.c
97481 Xext/xace.h
97482
97483commit f207e69d62bc04c7f254347b03e6d8fa8b569d66
97484Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97485Date: Wed Nov 14 12:23:29 2007 -0500
97486
97487 xselinux: adjust receive hook to use new synthetic_event class.
97488
97489commit cc4586df600af571815d5cdda2028c0d074c8469
97490Author: Peter Hutterer <peter@cs.unisa.edu.au>
97491Date: Wed Nov 14 14:27:52 2007 +1030
97492
97493 Xi: toggle the public.devPrivate as well when switching device classes.
97494
97495 The master needs to have the same devPrivate as the slave, in case a client
97496 issues a request that goes down to the driver.
97497 Example: if a driver wants to ring the keyboard bell, it'll pick a keyboard.
97498 The KeyClassPtr will direct it to the matching method in the driver, but
97499 because the MD doesn't have the devPrivate set the driver segfaults.
97500 Even if all drivers were updated to not dereference the nullpointer, nothing
97501 would actually ever happen.
97502
97503 To avoid this, we flip the master's public.devPrivate to the last SDs
97504 devPrivate.
97505
97506commit b44c1118f3bab6d5f28fa42e0c322fbaec005012
97507Author: Peter Hutterer <peter@cs.unisa.edu.au>
97508Date: Wed Nov 14 14:08:21 2007 +1030
97509
97510 dix: Return Success from ProcBell if no BellProc is set.
97511
97512 We must not return BadDevice, this causes applications to crash. If no
97513 BellProc is set, just quietly do nothing and report a Success.
97514
97515commit 240b10fa9c32510a380a73a8acdd8267c81e538b
97516Author: Peter Hutterer <peter@cs.unisa.edu.au>
97517Date: Wed Nov 14 12:18:08 2007 +1030
97518
97519 Xi: Move updating the device state from POE into separate function.
97520
97521 POE now only deals with processing the event and calling the appropriate
97522 delivery methods. Actually modifying the device state is done in
97523 UpdateDeviceState. This separation should make it easier to avoid setting the
97524 state twice when master events are processed.
97525
97526commit 0bdfeb55fc559ffe0334df69ba6e9e711b26dc18
97527Author: Peter Hutterer <peter@cs.unisa.edu.au>
97528Date: Tue Nov 13 17:31:20 2007 +1030
97529
97530 Xi: remove RegisterPairingClient handling. Deprecated with device hierarchy.
97531
97532commit e96d926d64f7fb63f6bf2aa0ea0a8440a188947b
97533Author: Peter Hutterer <peter@cs.unisa.edu.au>
97534Date: Tue Nov 13 17:14:35 2007 +1030
97535
97536 Xi: remove trailing whitespaces.
97537
97538commit 5031238aad2b6b7511aab0f9d15edfbdd2b4cce7
97539Author: Peter Hutterer <peter@cs.unisa.edu.au>
97540Date: Tue Nov 13 16:58:23 2007 +1030
97541
97542 dix: remove trailing/wrong whitespaces from devices.c and events.c
97543
97544commit c703849e79391bb1bb01a994bb8d5cf2eb64e48b
97545Author: Peter Hutterer <peter@cs.unisa.edu.au>
97546Date: Tue Nov 13 16:50:18 2007 +1030
97547
97548 dix: Attach spriteInfoRec to same memory block as the DeviceIntRec.
97549
97550 No need to alloc a separate one, we never do anything special with it anyway.
97551
97552commit 51239f87ce42ad564ceee1761980391947294511
97553Author: Peter Hutterer <peter@cs.unisa.edu.au>
97554Date: Tue Nov 13 11:26:16 2007 +1030
97555
97556 dix: Send MappingNotify when keyboard maps change.
97557
97558 If a slave device is attached to a master device, then we need to send a
97559 mapping notify to the client.
97560 Mapping notify needs to be sent if
97561 - different slave device but on same master
97562 - different master
97563
97564 This gives you funny behaviour with the ClientPointer. When a
97565 MappingNotify is sent to the client, the client usually responds with a
97566 GetKeyboardMapping. This will retrieve the ClientPointer's keyboard mapping,
97567 regardless of which keyboard sent the last mapping notify request. So
97568 depending on the CP setting, your keyboard may change layout in each app...
97569
97570commit 2b1d946392ce28b96941341778b2b526aa0fb126
97571Author: Peter Hutterer <peter@cs.unisa.edu.au>
97572Date: Tue Nov 13 09:51:33 2007 +1030
97573
97574 xkb: disable xkb key repeats (temporarily)
97575
97576 Haven't quite figured out yet how to make these repeats work. Because we share
97577 the class between devices, the key state is already set when we process the
97578 master device's event, causing a repeat on each event.
97579
97580commit f7dd0c72b8f861f4d5443a43d1013e3fe3db43ca
97581Author: Matthias Hopf <mhopf@suse.de>
97582Date: Mon Nov 12 15:11:03 2007 +0100
97583
97584 Only clear crtc of output if it is the one we're actually working on.
97585
97586 Upon recreation of the RandR internal data structures in RRCrtcNotify() the
97587 crtc of an output could be NULLed if the crtc was shared (cloned) between two
97588 outputs and one of them got another crtc assigned.
97589
97590commit 70b4087c4dd1904d9d655f4afb9dfcea4f137f7a
97591Author: Peter Hutterer <peter@cs.unisa.edu.au>
97592Date: Mon Nov 12 13:10:39 2007 +1030
97593
97594 dix: don't unconditionally update valuators during key events.
97595
97596 Master may not have valuators, device may not have valuators.
97597
97598commit 23365d28651f7942fdafb889bcbbd019470a4274
97599Author: Peter Hutterer <peter@cs.unisa.edu.au>
97600Date: Mon Nov 12 13:08:38 2007 +1030
97601
97602 dix: allow grab modifier device to be NULL.
97603
97604 This can happen if we check for a passive core grab and our device is a
97605 floating slave device. Doesn't really change anything as SDs can't send core
97606 events but it stops the server from segfaulting.
97607
97608commit 5a7a65a3c978a65e8ff39d0cc9878527ec42adc9
97609Author: Peter Hutterer <peter@cs.unisa.edu.au>
97610Date: Mon Nov 12 12:37:39 2007 +1030
97611
97612 mi: avoid SIGABRT by setting master_event to NULL.
97613
97614commit a05f43bf3e9629df98e93c366d4327f20ed81e6c
97615Author: Peter Hutterer <peter@cs.unisa.edu.au>
97616Date: Mon Nov 12 11:35:18 2007 +1030
97617
97618 dix: When the last slave is removed, set master to the original classes.
97619
97620 DeviceClassesChangedEvent is sent to the client, where device == new slave.
97621
97622commit 45f884d79c0eebaa1eb24d7db76c1177f6b710c9
97623Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97624Date: Fri Nov 9 14:45:27 2007 -0500
97625
97626 xselinux: add new synthetic_event security class, and fix registry code.
97627
97628commit b092856baba5bd43b23950f23236b5cc3ce78c1e
97629Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97630Date: Fri Nov 9 14:45:02 2007 -0500
97631
97632 registry: Register XC-SECURITY extension protocol names.
97633
97634commit 7a81bafc9bc7048560b17483e6addf58469a05d0
97635Author: Peter Hutterer <peter@cs.unisa.edu.au>
97636Date: Fri Nov 9 23:10:24 2007 +1030
97637
97638 Xi, dix: Add ability to change MD classes + send event when doing so.
97639
97640 Each time a different slave device sends through a master, an
97641 DeviceClassesChangedEvent is enqueued. When this event is processed, all
97642 classes of the matching master device are changed, and the event is sent to
97643 the clients.
97644
97645 Next time the master is queried, it thus shows the evclasses of the last slave
97646 device. The original classes are stored in the devPrivates.
97647
97648 TODO: if all slave devices are removed, the master's original classes need to
97649 be restored.
97650
97651commit c0a05805783ee3d38fbcc0fb45f4aa3c511785f0
97652Author: Peter Hutterer <peter@cs.unisa.edu.au>
97653Date: Fri Nov 9 23:07:10 2007 +1030
97654
97655 Xext: add simple macro to easily check a generic event's type.
97656
97657commit f48087b6c33c1f84bf2cfc0744b1c38697321c07
97658Author: Kristian Høgsberg <krh@temari.boston.redhat.com>
97659Date: Fri Nov 9 05:49:26 2007 -0500
97660
97661 Regenerate GLX dispatch code for recent gl_API.xml changes (#12935).
97662
97663commit f2a3728868376a3646832d4af3a29549ce0b8f5d
97664Author: Ben Byer <bbyer@bbyer.apple.com>
97665Date: Thu Nov 8 18:49:05 2007 -0800
97666
97667 Patch to rootless code that should fix many crashes. Credit to Ken Thomases
97668 at CodeWeavers for the patch. From his description:
97669
97670 Fix a display bug with the X server. The Generic Rootless extension
97671 installs overrides for certain GC (graphics context) operations. Within
97672 these overrides, they temporarily uninstall themselves, perform their work,
97673 and then reinstall themselves. Except sometimes they would return early
97674 and wouldn't reinstall themselves when they should. Now they do in all cases.
97675
97676 Fix a bug in RootlessCopyWindow where early returns could leave the screen's
97677 dispatch table entry for CopyWindow unwrapped. We think that this is
97678 another case (hopefully the last) of the rootless drawing bug.
97679
97680commit 338c1aedbdf3964e542947140f7c50d58542cf12
97681Author: Ben Byer <bbyer@bbyer.local>
97682Date: Wed Nov 7 03:56:44 2007 -0800
97683
97684 formatting fixes
97685
97686commit bd269d0d783d418ef99363478fdf849fd89eed76
97687Author: Ben Byer <bbyer@bbyer.local>
97688Date: Wed Nov 7 03:55:08 2007 -0800
97689
97690 Fix for off-by-one error in menu bar height calculation -- props to Nicholas Riley!
97691
97692commit b4d14484056e6f4a7374fc1acf3f223be4bd116f
97693Author: Ben Byer <bbyer@bbyer.local>
97694Date: Wed Nov 7 03:10:52 2007 -0800
97695
97696 Undo some last-minute breakage in xpr.h
97697
97698commit 05d5b9baa05a4ba14a4383d8a981bc327d99290c
97699Author: Ben Byer <bbyer@bbyer.local>
97700Date: Wed Nov 7 02:28:49 2007 -0800
97701
97702 removed debugging output
97703
97704commit 9a8abcfa6d6d0cdc17be02a3443a7e116eb07d07
97705Author: Ben Byer <bbyer@bbyer.local>
97706Date: Wed Nov 7 02:22:39 2007 -0800
97707
97708 Fixed focus problem (clicking on an X11 window that sits behind
97709 an Aqua window would not always bring it to the top of the stack.
97710
97711commit b34d2ffc38002f7c4980c138f57e9a828cd79c37
97712Author: Ben Byer <bbyer@bbyer.local>
97713Date: Wed Nov 7 01:56:37 2007 -0800
97714
97715 formatting changes.
97716
97717commit 50dac9b2cb3b40810fb79253adc0265a838a497b
97718Author: Ben Byer <bbyer@bbyer.local>
97719Date: Wed Nov 7 01:35:48 2007 -0800
97720
97721 Fixed Spaces issue, correctly -- dragging an X window from one
97722 Space to another in Expose mode now works.
97723
97724commit ce7cfbe261b7fd4fcd09d1a4a61344d1555a71f2
97725Author: Ben Byer <bbyer@bbyer.apple.com>
97726Date: Thu Nov 8 20:10:51 2007 -0800
97727
97728 Fixed check to refer to DarwinApp, not all Darwin targets
97729
97730commit d68bd5510437c1fd3850e020f7cd90901fae8e1b
97731Author: Ben Byer <bbyer@bbyer.apple.com>
97732Date: Thu Nov 8 20:08:49 2007 -0800
97733
97734 fixing GLX in Xquartz
97735
97736commit a6ac9002956767fefa37aac95513e21ac5246d15
97737Author: Ben Byer <bbyer@bbyer.apple.com>
97738Date: Mon Nov 5 20:25:10 2007 -0800
97739
97740 formatting cleanup
97741
97742commit 67e96be13cdb45be31db121ce216295cd9496d20
97743Author: Ben Byer <bbyer@bbyer.apple.com>
97744Date: Mon Nov 5 20:01:34 2007 -0800
97745
97746 Fixed logic error that prevent JIS (Japanese) keyboard layouts from
97747 being detected.
97748
97749commit 154fb6417e5d0bae5191984beac5ce045ce754bb
97750Author: Ben Byer <bbyer@bbyer.local>
97751Date: Sat Nov 3 05:34:19 2007 -0700
97752
97753 Initial support for Spaces -- if you use Expose to drag an X11
97754 window to another Space, it will work correctly (as opposed
97755 to just leaving a ghost window). We accomplish this by listening
97756 for the notification from Xplugin that our window has been moved,
97757 and then we ask X11 to move the window to the new location.
97758
97759commit 5bbc468b702f62d7c91d41aabcc27eeb553f6959
97760Author: Peter Hutterer <peter@cs.unisa.edu.au>
97761Date: Fri Nov 9 11:33:27 2007 +1030
97762
97763 dix: grabbing an attached SD sets it floating for the duration of the grab.
97764
97765commit 676b26ca3e9b142cf007af3f439aa1993f2247c4
97766Author: Peter Hutterer <peter@cs.unisa.edu.au>
97767Date: Thu Nov 8 17:54:17 2007 +1030
97768
97769 Xi: notify the clients when the device hierarchy has been changed.
97770
97771commit cb75f09146a3c17b1a67b636bbf7229c65c83b35
97772Author: Peter Hutterer <peter@cs.unisa.edu.au>
97773Date: Fri Nov 9 11:29:18 2007 +1030
97774
97775 dix: calloc root window's Generic Event mask, stops segfaults.
97776
97777commit 169f83e366f678ac5441ad21beb84c9b8c65d28e
97778Author: Ben Byer <bbyer@bbyer.local>
97779Date: Sun Nov 4 19:14:10 2007 -0800
97780
97781 Disable deferred updates in xp_init to fix performance problems
97782 -- thanks to Eric Gouriou for pointing out the issue
97783
97784commit 1c6cb353f77747c101ce47716ff1fa055fbf85a4
97785Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97786Date: Thu Nov 8 16:46:49 2007 -0500
97787
97788 Restore the XC-SECURITY option in configure.ac, but disabled by default.
97789
97790commit 9d03cad1446c27b397c198cf6247e71e46bc9e6d
97791Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97792Date: Thu Nov 8 16:41:47 2007 -0500
97793
97794 Remove SecurityPolicy file and associated references in the manpages.
97795
97796commit 8b5d21cc1d1f4e9d20e5d5eca44cb1e60a419763
97797Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97798Date: Thu Nov 8 16:32:42 2007 -0500
97799
97800 Rework of the XC-SECURITY extension. The gen-auth protocol has not changed,
97801 but the XC-QUERY-SECURITY-1 authorization method and the SecurityPolicy
97802 configuration file have been removed. The semantics of the trusted vs.
97803 untrusted split have been changed. This will be documented in a future
97804 commit.
97805
97806commit 476a9d85f819f454a6901ccb7eb028d1c563c341
97807Author: Dodji Seketeli <dodji@openedhand.com>
97808Date: Thu Nov 8 09:11:05 2007 +0100
97809
97810 Xephyr: do not AM_CONDITIONAL inside a shell if branch
97811
97812commit 59b304d8a24fecd094296feb217823f0c73d6f82
97813Author: Peter Hutterer <peter@cs.unisa.edu.au>
97814Date: Thu Nov 8 15:44:18 2007 +1030
97815
97816 dix, mi: stop segfaults when a floating device emits events.
97817
97818 For pointers: don't try to set master->valuator fields if there is no master.
97819 For keyboards: check if device is valid before trying to access the fields in
97820 miPointerGetScreen (btw. this disables DGA events for floating keyboards).
97821
97822 Also stop the hideous number of ErrorFs if we request the paired device for a
97823 floating dev.
97824
97825commit 3063f0c6679bdbea13f360cff8d09a88b9871da9
97826Author: Peter Hutterer <peter@cs.unisa.edu.au>
97827Date: Thu Nov 8 11:51:03 2007 +1030
97828
97829 mi: change the device id to the master devices' when duplicating the event.
97830
97831 Move the event duplication into a CopyGetMasterEvent(), makes the code a bit
97832 better to read.
97833
97834commit 0e9ef65fa583bf2393dd0fda82df6f092387b425
97835Author: Keith Packard <keithp@koto.keithp.com>
97836Date: Wed Nov 7 16:33:10 2007 -0800
97837
97838 Don't frob timers unless SmartSchedule is running
97839
97840commit 180220f284823c486e7001ef00f1279dc548c9c7
97841Author: Peter Hutterer <peter@cs.unisa.edu.au>
97842Date: Thu Nov 8 10:00:35 2007 +1030
97843
97844 Xi: return BadDevice for slave devices on ProcXSetClientPointer
97845
97846 Also clean up to return error codes instead of sending the error manually.
97847
97848commit 184a7b8917a15bb2c719153b9b016c03aab42101
97849Merge: a8808ac 0b72905
97850Author: Peter Hutterer <peter@cs.unisa.edu.au>
97851Date: Thu Nov 8 09:34:35 2007 +1030
97852
97853 Merge branch 'mpx' into mdsd
97854
97855 Conflicts:
97856
97857 Xi/opendev.c
97858
97859commit e94ab0b502160376d1956345196f7e1c7e2e886f
97860Merge: 26e1fc7 9bee1c6
97861Author: Dodji Seketeli <dodji@openedhand.com>
97862Date: Wed Nov 7 19:04:24 2007 +0100
97863
97864 sync with 'master'
97865
97866commit 9bee1c6912817f65bbb8cf4078f0ad016d9d51cb
97867Author: Michel Dänzer <michel@tungstengraphics.com>
97868Date: Wed Nov 7 18:56:45 2007 +0100
97869
97870 EXA: Disable problematic optimization of dest pixmap migration by default.
97871
97872 Also add some code comments about these optimizations.
97873
97874commit 26e1fc7b42de850d69fba89703ffddd36480b997
97875Author: Dodji Seketeli <dodji@openedhand.com>
97876Date: Wed Nov 7 18:48:23 2007 +0100
97877
97878 Xephyr: don't use Xv or GL when those are disabled.
97879
97880commit 950f9995d11aff2c51139b34fb27eba594f2bd20
97881Author: Dodji Seketeli <dodji@openedhand.com>
97882Date: Wed Nov 7 18:43:16 2007 +0100
97883
97884 Xnest: fix lib dependancy to make libtool happy
97885
97886 This fixes an undefined symbol error happening when compiling
97887 the server with the --disable-xv configure switch.
97888 Basically, xnest was linking against
97889 @XSERVER_LIBS@ and @XNEST_LIBS@ and the order of the libraries
97890 given to the linker at the end of the process was bogus.
97891
97892 * configure.ac: make XNEST_LIBS contain the $XSERVER_LIBS re-ordered
97893 in such a way that the linker finds the symbols of all the libs contained
97894 in $XNEST_LIBS.
97895 * hw/xnest/Makefile.am: don't link against @XSERVER_LIBS@ anymore because
97896 XNEST_LIBS contains the right thing.
97897
97898commit 0b729051c04da7068f1e6dd319190bd0a362b2c0
97899Merge: b7ee005 d7c5e8b
97900Author: Peter Hutterer <peter@cs.unisa.edu.au>
97901Date: Wed Nov 7 15:37:23 2007 +1030
97902
97903 Merge branch 'master' into mpx
97904
97905 Conflicts:
97906
97907 Xi/extinit.c
97908 Xi/grabdev.c
97909 Xi/setmode.c
97910 Xi/ungrdev.c
97911 dix/devices.c
97912 dix/events.c
97913 dix/getevents.c
97914 include/dix.h
97915 mi/midispcur.c
97916 mi/misprite.c
97917 xkb/xkbActions.c
97918 xkb/xkbEvents.c
97919 xkb/xkbPrKeyEv.c
97920
97921commit e717409bae355df9a617a226f12fbb8c54ae77e5
97922Author: Daniel Stone <daniel@fooishbar.org>
97923Date: Tue Nov 6 21:36:13 2007 +0000
97924
97925 DIX/getevents: Document GetMaximumEventsNum() a little better
97926
97927 Note that the number returned by GMEN can _never_ change, and be a little more
97928 explicit about the figure for repeats.
97929
97930commit 66fe554a59bb7de37354b618945cd5f30d78250d
97931Author: Daniel Stone <daniel@fooishbar.org>
97932Date: Tue Nov 6 18:57:09 2007 +0000
97933
97934 COPYING: Collapse 'canonical license' into one statement
97935
97936 For a few of us, the license statement is identical, and the only variant is
97937 the copyright. For these, aggregate the copyrights, and only list the license
97938 once. Put this at the top, and note that this is more or less our agreed
97939 canonical license.
97940
97941commit fda832772b3e630037bf1b822534996154a50861
97942Author: Daniel Stone <daniel@fooishbar.org>
97943Date: Tue Nov 6 15:05:06 2007 +0000
97944
97945 .gitignore: Ignore build directories
97946
97947 Ignore directories people might use for building.
97948
97949commit 512bac25ec0e980968b93a2ebe88bd89bf99b697
97950Author: Daniel Stone <daniel@fooishbar.org>
97951Date: Tue Nov 6 14:52:03 2007 +0000
97952
97953 DIX: XKB: Set xkbInfo to NULL as well as freeing it (bug #10639)
97954
97955 XkbRemoveResourceClient wants to access xkbInfo if it exists, so make
97956 sure we NULL it after freeing it. It doesn't make much sense to move
97957 the RemoveResourceClient call first, as there's not much point in
97958 notifying clients while we're shutting the server down anyway.
97959
97960commit a8808ac3d093f33b39de109107d396fe0a02c4fc
97961Author: Peter Hutterer <peter@cs.unisa.edu.au>
97962Date: Wed Nov 7 11:42:58 2007 +1030
97963
97964 Xi: don't call CheckMotion for slave devices.
97965
97966 This essentially disables enter/leave for slave devices, but re-enables them
97967 for master devices. Which is a good thing after all.
97968
97969commit 15944b8f02752eedd61be34a6a29dd6b82a0ac97
97970Author: Peter Hutterer <peter@cs.unisa.edu.au>
97971Date: Wed Nov 7 11:33:24 2007 +1030
97972
97973 mi: duplicate event before processing it, so master has original values
97974
97975 Event values may get changed in the event processing stage, so we need to
97976 duplicate it to pump different events through for slave and master device.
97977
97978commit d7c5e8bfc1aecbd23a4cbb2eab08656587aac2e8
97979Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
97980Date: Tue Nov 6 16:26:09 2007 -0500
97981
97982 Modified performance patches from Arjan van de Ven <arjan@infradead.org>
97983
97984 Subject: [PATCH] fix some performance gaps in Xace
97985
97986 The XaceHook function is used in several hotpaths.
97987 The problem with it (performance wise) is twofold:
97988 * The XaceHook function has a big switch() statement for the hook number in it
97989 * The XaceHook function uses varargs to reassemble the final dispatch arguments again
97990
97991 Both are expensive operations... for something that is known at compile time
97992
97993 This patch turns the hotpath XaceHook call into a direct call to avoid
97994 the switch and varargs; this gives me over 10% performance gain
97995 on the x11perf benchmark.
97996
97997commit aaa50b64113b122aaebd46e3b78e3fb7a8d70500
97998Merge: 868e303 ddce48e
97999Author: Dodji Seketeli <dodji@openedhand.com>
98000Date: Tue Nov 6 16:25:40 2007 +0100
98001
98002 resync with 'master'
98003
98004commit 868e303630d8b84070c2f1fd8d6da2cef045b029
98005Author: Dodji Seketeli <dodji@openedhand.com>
98006Date: Tue Nov 6 16:24:46 2007 +0100
98007
98008 Xephyr: fix a crash when using xrandr twice
98009
98010 * hw/kdrive/ephyr/ephyr.c:
98011 (ephyrScreenFini): don't forget to
98012 free shadowfb data (if necessary) upon server is reset.
98013
98014commit ddce48ede036f3996f8e584b0012c396c5df42fb
98015Author: Elvis Pranskevichus <el@prans.net>
98016Date: Tue Nov 6 09:40:14 2007 +0000
98017
98018 Config: D-Bus: Fix dbus_bus_request_name failure check
98019
98020 The code in connect_hook incorrectly checks for dbus_bus_request_name failure.
98021 The dbus_bus_request_name error indicator is -1, not 0. This leads
98022 to subsequent assertion failure in libdbus.
98023
98024commit a52c9b2a59f27266557ff9d5d2c08492e04135a6
98025Merge: c7e18be 5833289
98026Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98027Date: Mon Nov 5 19:08:36 2007 -0500
98028
98029 Merge branch 'master' into XACE-SELINUX
98030
98031 Conflicts:
98032
98033 dix/dispatch.c
98034 dix/property.c
98035 hw/xfree86/common/xf86VidMode.c
98036 include/xkbsrv.h
98037 render/glyph.c
98038 xkb/xkbActions.c
98039
98040commit 58332894c061ae96d6a457f65266660f5f65e88b
98041Author: Adam Jackson <ajax@redhat.com>
98042Date: Mon Nov 5 17:17:25 2007 -0500
98043
98044 Export the server ABI versions from xorg-server.pc
98045
98046commit c7e18beb3c87eb1ada9b21c4ffacd11c1939c087
98047Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98048Date: Mon Nov 5 15:01:13 2007 -0500
98049
98050 xselinux: Register SELinux extension protocol names.
98051
98052commit 632c33c5c18b3e091c4fce98280af4d583e99640
98053Merge: 4113f04 dda10c9
98054Author: Dodji Seketeli <dodji@openedhand.com>
98055Date: Mon Nov 5 17:52:33 2007 +0100
98056
98057 pull 'master'
98058
98059commit 4113f040c587fc536adc693d7ee5a4c0a60b75d4
98060Author: Dodji Seketeli <dodji@openedhand.com>
98061Date: Mon Nov 5 17:33:01 2007 +0100
98062
98063 GL: fix crash at mesa destruction time
98064
98065 * GL/glx/glxglcore.c:
98066 (_glXMesaScreenDestroy): delete the same amount of visuals
98067 that those which were created in createMesaVisuals().
98068
98069commit dda10c9066a660b647384179f82e1da8e063264f
98070Author: Daniel Stone <daniel@fooishbar.org>
98071Date: Mon Nov 5 16:28:35 2007 +0000
98072
98073 Remove all traces of external RGB database (and Speedo)
98074
98075 Remove all references to an external RGB database (which hasn't been enabled
98076 for a very long time). Also get rid of some references to Speedo fonts.
98077
98078commit 6ff79ea5f7d3ff0c3b14e39849514784ccd40190
98079Author: Dodji Seketeli <dodji@openedhand.com>
98080Date: Mon Nov 5 16:46:49 2007 +0100
98081
98082 Xephyr: fix some DRI build breakage
98083
98084commit 5e363500c86042c394595e1a6633581eb8fcd1bb
98085Author: Daniel Stone <daniel@fooishbar.org>
98086Date: Mon Nov 5 14:38:28 2007 +0000
98087
98088 OS: Remove ALLOCATE_LOCAL from os.h
98089
98090 Remove ALLOCATE_LOCAL_FALLBACK and DEALLOCATE_LOCAL_FALLBACK from os.h, and
98091 remove the include of Xalloca.h as well.
98092
98093commit 83e5d9e75d0fa1135e2c9d8b59caee98c3291681
98094Author: Daniel Stone <daniel@fooishbar.org>
98095Date: Mon Nov 5 14:36:54 2007 +0000
98096
98097 DIX: Remove last alloca call
98098
98099 Replace with heap allocations.
98100
98101commit 1179ddea94efae6606162e9a1b0f2bf752ae4dd0
98102Author: Daniel Stone <daniel@fooishbar.org>
98103Date: Mon Nov 5 14:36:36 2007 +0000
98104
98105 EXA: Remove usage of alloca
98106
98107 Replace with heap allocations.
98108
98109commit 1eb6a1d0d2817cd0ce96fb27db3ceb2d0ff024af
98110Author: Daniel Stone <daniel@fooishbar.org>
98111Date: Mon Nov 5 14:18:35 2007 +0000
98112
98113 XTrap: Remove usage of alloca
98114
98115 Replace with xalloc/xfree.
98116
98117commit 934281126f6c602fa4bd0c2e29d8c9f44fe532b8
98118Author: Daniel Stone <daniel@fooishbar.org>
98119Date: Mon Nov 5 14:18:22 2007 +0000
98120
98121 Xext: Remove usage of alloca
98122
98123 Replace with heap allocations.
98124
98125commit fb32bb9839b615f7297fbfac2050bc216682f01c
98126Author: Daniel Stone <daniel@fooishbar.org>
98127Date: Mon Nov 5 14:17:54 2007 +0000
98128
98129 Xi: Remove usage of alloca
98130
98131 Replace with xalloc/xfree.
98132
98133commit 683ee1776d172035c465aa8fc84ccd53bb8ba7fd
98134Author: Daniel Stone <daniel@fooishbar.org>
98135Date: Mon Nov 5 14:14:25 2007 +0000
98136
98137 afb: Remove usage of alloca
98138
98139 Replace with heap allocations.
98140
98141commit ca75261beedc3e00767b3812a81b7dac4437f4a1
98142Author: Daniel Stone <daniel@fooishbar.org>
98143Date: Mon Nov 5 14:14:04 2007 +0000
98144
98145 cfb: Remove usage of alloca
98146
98147 Replace with xalloc/xfree.
98148
98149commit 914922fd6100a409a3dfd1c64511ed6bdc344bef
98150Author: Daniel Stone <daniel@fooishbar.org>
98151Date: Mon Nov 5 14:12:59 2007 +0000
98152
98153 DIX: Remove usage of alloca
98154
98155 Replace with heap allocations.
98156
98157commit 3b77689266e729411229ec83d2a90578ebc1d82f
98158Author: Daniel Stone <daniel@fooishbar.org>
98159Date: Mon Nov 5 14:12:34 2007 +0000
98160
98161 EXA: Remove usage of alloca
98162
98163 Replace with heap allocations.
98164
98165commit 34cdf06e4ccb243664005cc33009d8759a7f6e4d
98166Author: Daniel Stone <daniel@fooishbar.org>
98167Date: Mon Nov 5 14:12:22 2007 +0000
98168
98169 fb: Remove usage of alloca
98170
98171 Replace with heap storage.
98172
98173commit 733d42065f2c24505b3874ce51c18f6063c2b67e
98174Author: Daniel Stone <daniel@fooishbar.org>
98175Date: Mon Nov 5 14:11:59 2007 +0000
98176
98177 XFree86: Remove usage of alloca
98178
98179 Replace with heap allocations.
98180
98181commit caf545063457591f88e1f7bcd25dbd0342f44343
98182Author: Daniel Stone <daniel@fooishbar.org>
98183Date: Mon Nov 5 14:10:55 2007 +0000
98184
98185 KDrive: Remove usage of alloca
98186
98187 Replace with heap allocations.
98188
98189commit d57060f16714f5667722001bd1a4500059dd59e1
98190Author: Daniel Stone <daniel@fooishbar.org>
98191Date: Mon Nov 5 14:10:21 2007 +0000
98192
98193 DMX: Remove usage of alloca
98194
98195 Replace with heap allocations.
98196
98197commit def6f74f2d7342f85f3df2053e0b9c1ac483b51d
98198Author: Daniel Stone <daniel@fooishbar.org>
98199Date: Mon Nov 5 14:10:03 2007 +0000
98200
98201 Xvfb: Remove usage of alloca
98202
98203 Replace with heap allocations
98204
98205commit 259f4c36d581896ce605741b9e557a589013a9b8
98206Author: Daniel Stone <daniel@fooishbar.org>
98207Date: Mon Nov 5 14:09:49 2007 +0000
98208
98209 Xprint: Remove usage of alloca
98210
98211 Replace with heap-based allocations.
98212
98213commit 1c84337af0ac40498a53aa36289e2f6f5ff8b1c6
98214Author: Daniel Stone <daniel@fooishbar.org>
98215Date: Mon Nov 5 14:09:32 2007 +0000
98216
98217 Xwin: Remove usage of alloca
98218
98219 Replace with heap allocations.
98220
98221commit f7d5c292e44113ea8eb32e67e91cd02e520df5e3
98222Author: Daniel Stone <daniel@fooishbar.org>
98223Date: Mon Nov 5 14:09:14 2007 +0000
98224
98225 mfb: Remove usage of alloca
98226
98227 Replace with heap-based allocations.
98228
98229commit be9ee17f960cc3d8a8f999cab1579e83d9aea520
98230Author: Daniel Stone <daniel@fooishbar.org>
98231Date: Mon Nov 5 14:08:51 2007 +0000
98232
98233 mi: Remove usage of alloca
98234
98235 Replace with heap allocations.
98236
98237commit 3633ae6efb57c5405c26e8ae132b9371e5f920de
98238Author: Daniel Stone <daniel@fooishbar.org>
98239Date: Mon Nov 5 14:04:27 2007 +0000
98240
98241 Damage: Remove usage of alloca
98242
98243 Replace with heap allocations.
98244
98245commit 2761c103311a1160bc483fd0367d654733df8598
98246Author: Daniel Stone <daniel@fooishbar.org>
98247Date: Mon Nov 5 14:03:26 2007 +0000
98248
98249 OS: Remove usage of alloca
98250
98251 Replace with heap allocations.
98252
98253commit 2d738efb959912d8a89864e41190e0064fa77906
98254Author: Daniel Stone <daniel@fooishbar.org>
98255Date: Mon Nov 5 14:01:11 2007 +0000
98256
98257 RandR: Remove usage of alloca
98258
98259 Replace with heap allocations.
98260
98261commit 3c1d2fdeff0ed61d86fa7d35cb0a61535d9b2816
98262Author: Daniel Stone <daniel@fooishbar.org>
98263Date: Mon Nov 5 14:00:40 2007 +0000
98264
98265 Record: Remove usage of alloca
98266
98267 Replace with xalloc/xfree.
98268
98269commit e0491f470e130147191388168e878e3a7348afaf
98270Author: Daniel Stone <daniel@fooishbar.org>
98271Date: Mon Nov 5 13:59:51 2007 +0000
98272
98273 Render: Remove usage of alloca
98274
98275 Replace it with heap-based allocations.
98276
98277commit 59774af86b851c7fb8989cef6c013522549000b8
98278Author: Daniel Stone <daniel@fooishbar.org>
98279Date: Mon Nov 5 13:59:15 2007 +0000
98280
98281 XKB: Remove usage of alloca
98282
98283 alloca has no way to return failure, and instead can possibly arbitrarily
98284 overflow the stack. Let's avoid that one.
98285
98286commit 6e4f5cf83f35ffebb51633ab30b1826e63e37223
98287Author: Ben Byer <bbyer@bbyer.local>
98288Date: Mon Nov 5 05:53:34 2007 -0800
98289
98290 changing ALLOCATE_LOCAL to xalloc to prevent stack overflow
98291
98292commit b1764ddf133cfdf979db62ee2491124a4798b55b
98293Author: Ben Byer <bbyer@bbyer.local>
98294Date: Mon Nov 5 05:44:54 2007 -0800
98295
98296 pulling more patches over from xorg-xserver-1.2-apple branch
98297
98298commit 10fde62fc88302f7d3b2546239b1679be249567c
98299Author: Ben Byer <bbyer@bbyer.local>
98300Date: Sun Nov 4 18:59:41 2007 -0800
98301
98302 more CVS tags
98303
98304commit 27ecb89250398f45564fe454e20bfdd66f62c8e4
98305Author: Ben Byer <bbyer@bbyer.local>
98306Date: Sun Nov 4 18:53:41 2007 -0800
98307
98308 Removed CVS tags.
98309
98310commit f2e310132fbe1520c1b5f3da4faa2d2d47835e72
98311Author: Aaron Plattner <aplattner@nvidia.com>
98312Date: Wed Oct 31 14:15:35 2007 -0700
98313
98314 Add CreatePixmap allocation hints.
98315
98316 These hints allow an acceleration architecture to optimize allocation of certain
98317 types of pixmaps, such as pixmaps that will serve as backing pixmaps for
98318 redirected windows.
98319
98320commit 3f1b6765aadf665ede8253464da19a5878f16e56
98321Author: Markku Vire <markku.vire@movial.fi>
98322Date: Thu Nov 1 22:43:04 2007 +0200
98323
98324 Config: HAL: Touchpads are pointers too
98325
98326 Treat touchpads -- not just mice -- as pointer devices.
98327
98328commit 41c3069f7cf28155f8e6cfe0c10a12a1f5f76c7d
98329Author: Mark Vytlacil <mrv@wi.rr.com>
98330Date: Thu Nov 1 21:05:43 2007 +0200
98331
98332 XFree86: Input: Save/restore errno around SIGIO (bug #10683)
98333
98334 Make sure errno is saved and restored from the SIGIO handler, so errors
98335 from system calls in input handlers don't break the interrupted code.
98336
98337commit ff9929ed48f2dec8b536d348e25e66a0bc4ac1a6
98338Author: Ben Byer <bbyer@bbyer.local>
98339Date: Sun Nov 4 05:13:19 2007 -0800
98340
98341 pulling in changes from xorg-server-1.2-apple branch
98342
98343commit 181468db92d44a58080fc9a76e46dfc7011bf9f1
98344Author: Ben Byer <bbyer@bbyer.local>
98345Date: Sun Nov 4 04:50:26 2007 -0800
98346
98347 add missing null-pointer checks
98348
98349commit 4d0f35c81df307609b29c0f12aa1b4c6ee6fd8d2
98350Author: Matthias Hopf <mhopf@suse.de>
98351Date: Fri Nov 2 19:27:32 2007 +0100
98352
98353 Always duplicate mode name when duplicating a mode.
98354
98355 If the originating mode didn't have a name, we would end up with the name of
98356 the original mode being setup correctly, but with the name of the copy still
98357 being NULL.
98358
98359commit dfb1cce28ac07d412598f7b9ab2ee908978b38e7
98360Author: Adam Jackson <ajax@redhat.com>
98361Date: Fri Nov 2 12:51:56 2007 -0400
98362
98363 Restore build of xf8_16bpp.
98364
98365 It doesn't link against cfb, so don't conditionalize its build against cfb.
98366
98367commit 7a0555e9bb59d02816803a1100f807d2d29f31d4
98368Author: Kristian Høgsberg <krh@redhat.com>
98369Date: Sun Oct 28 09:37:52 2007 +0100
98370
98371 Fix crash in xf86InitOrigins()
98372
98373 In a multihead setup, if only the first screen can be
98374 initialized, but the second screen is mentioned first in the
98375 ServerLayout section, the xf86InitOrigins() function will crash
98376 because the screen referred to in the e.g. "RightOf" part is
98377 non-existent.
98378
98379commit 5b41d4e60be35cfb96bedec0931fd5922823b4b9
98380Author: Adam Jackson <ajax@redhat.com>
98381Date: Sun Oct 28 09:37:52 2007 +0100
98382
98383 Don't filter modes away during VBE mode list construction.
98384
98385 Pass all VBE modes back up to the driver, on the assumption that it
98386 knows how to filter modes intelligently.
98387
98388commit c095da04fe7c73b6503ef5b93549b13796c51b22
98389Author: Adam Jackson <ajax@redhat.com>
98390Date: Sun Oct 28 09:37:52 2007 +0100
98391
98392 Fix magic number in fbdevhw
98393
98394 The transformation between fbdev and xfree86 mode timings needs to be
98395 invertible, otherwise Xen and other framebuffers that don't have real
98396 pixel clocks won't initialize.
98397
98398commit f4fe66f6767d1941317dc0280ac359421a152587
98399Author: Adam Jackson <ajax@redhat.com>
98400Date: Sun Oct 28 09:37:52 2007 +0100
98401
98402 Disable explicit commenting in Monitor section.
98403
98404commit 22f0e3a8b04e574047a51c8f928a007787303294
98405Author: Adam Jackson <ajax@redhat.com>
98406Date: Sun Oct 28 09:37:52 2007 +0100
98407
98408 Avoid PS/2 protocol probing for /dev/input/mice
98409
98410 The kernel will always upconvert to ExplorerPS/2 for all readers of
98411 /dev/input/mice, so don't waste time on startup trying to figure
98412 that out.
98413
98414commit b97518666dc32710fe69eee33ee56881dcff1bbc
98415Author: Adam Jackson <ajax@redhat.com>
98416Date: Sun Oct 28 09:37:52 2007 +0100
98417
98418 Fix accidental ABI usage in RANDR 1.2 drivers.
98419
98420 Due to RANDR 1.2, xf86findOptionValue and xf86nameCompare are now ABI.
98421 Make sure they're exported from the server.
98422
98423commit f01e149d1af14ef9ee0e8a6743ab6a08f3bb677c
98424Author: Adam Jackson <ajax@redhat.com>
98425Date: Thu Nov 1 15:41:11 2007 -0400
98426
98427 Move SIGUSR1 notification as late as possible.
98428
98429 If we inherited a signal mask from the parent process that ignores SIGUSR1,
98430 then we will send SIGUSR1 to the parent to indicate when we're ready to
98431 accept connections. Unfortunately, we send this notification way too
98432 early, right after creating the sockets rather than just before entering
98433 the main loop.
98434
98435 Move it to just before Dispatch() so we're not lying quite so much.
98436
98437commit a2ee5fe0c4e863c7ff4f644e5ac86f2793903103
98438Author: Adam Jackson <ajax@redhat.com>
98439Date: Thu Nov 1 13:46:20 2007 -0400
98440
98441 Enable MIT-SHM in Xdmx.
98442
98443 The extension is entirely DIX-level, no new DDX changes needed.
98444
98445commit 2338d5c9914e2a43c3a4f7ee0f4355ad0a1ad9e7
98446Author: Arjan van de Ven <arjan@linux.intel.com>
98447Date: Sun Oct 28 09:37:52 2007 +0100
98448
98449 reduce wakeups from smart scheduler
98450
98451 The smart scheduler itimer currently always fires after each request
98452 (which in turn causes the CPU to wake out of idle, burning precious
98453 power). Rather than doing this, just stop the timer before going into
98454 the select() portion of the WaitFor loop. It's a cheap system call, and
98455 it will only get called if there's no more commands batched up from the
98456 active fd.
98457
98458 This change also allows some of the functions to be simplified;
98459 setitimer() will only fail if it's passed invalid data, and we don't do
98460 that... so make it void and remove all the conditional code that deals
98461 with failure.
98462
98463 The change also allows us to remove a few variables that were used for
98464 housekeeping between the signal handler and the main loop.
98465
98466 Signed-off-by: Keith Packard <keithp@koto.keithp.com>
98467
98468commit 692654b4300e61a9481e6fa588bcb44a3c3ca150
98469Author: Kristian Høgsberg <krh@redhat.com>
98470Date: Mon Oct 29 18:13:58 2007 -0400
98471
98472 Set up visuals for the existing X visuals.
98473
98474 This makes the root visual a GLX capable visual again and adds a GLX visual
98475 for the COMPOSITE ARGB visual cleanly (as opposed to the hack we had before).
98476
98477commit a5546a99ac4da61aee0d49c55bcb38bbce9a96aa
98478Author: Kristian Høgsberg <krh@redhat.com>
98479Date: Mon Oct 29 18:06:41 2007 -0400
98480
98481 Change the GLX module to initialize after COMPOSITE.
98482
98483 This changes the module initalization order so that the GLX module initializes
98484 after COMPOSITE. The reason for this change is to be able to initialize a
98485 GLX visual config for the COMPOSITE ARGB visual.
98486
98487commit 8ead41388e36e21eea6fa0408c847f174911eab0
98488Author: Peter Hutterer <peter@cs.unisa.edu.au>
98489Date: Thu Sep 6 18:57:00 2007 +0930
98490
98491 xfree86: wrap keyboard devices for XKB.
98492
98493 Call ProcessOtherEvents first, then for all keyboard devices let them be
98494 wrapped by XKB. This way all XI events will go through XKB.
98495
98496 Note that the VCK is still not wrapped, so core events will bypass XKB.
98497
98498 (cherry picked from commit d627061b48ae06d27b37be209d67a3f4f2388dd3)
98499
98500commit e717cf08e99746761d74289c426bbd84176f4435
98501Author: Daniel Stone <daniel@fooishbar.org>
98502Date: Sat Oct 27 21:32:47 2007 +0300
98503
98504 XKB: Cope with all events in XkbProcessKeyboardEvent
98505
98506 Cope with Xi and pointer events in the (now increasingly misnamed)
98507 XkbProcessKeyboardEvent. If it's the wrong type, call through the wrapping
98508 chain to get out; else, process it.
98509
98510commit 9db8846fa53d91193bbfe541b244e2326440011d
98511Author: Daniel Stone <daniel@fooishbar.org>
98512Date: Sat Oct 27 21:31:39 2007 +0300
98513
98514 XKB: Don't update indicators on all devices, add missing include file
98515
98516 Don't get XkbUpdateIndicators to update the indicators on all our devices: we
98517 already deal with that ourselves.
98518 Add exevents.h include to get more (proto)types.
98519
98520commit ee3aa948eb8ed181d037294ed87df6ceec81684e
98521Author: Peter Hutterer <peter@cs.unisa.edu.au>
98522Date: Wed Sep 26 18:04:59 2007 +0930
98523
98524 xkb: Unwrap properly in ProcessPointerEvent.
98525
98526 Instead of hardcoding CoreProcessPointerEvent, actually try to unwrap properly
98527 and then call the unwrapped processInputProc. Seems to be a better idea,
98528 especially since it makes stuff actually work...
98529 (cherry picked from commit 8f9bf927e1beecf9b9ec8877131ec12c765e4d84)
98530
98531commit d3588a0aee33fbd233082f881c0d37152c6d4d8b
98532Author: Peter Hutterer <peter@cs.unisa.edu.au>
98533Date: Thu Sep 27 11:44:03 2007 +0930
98534
98535 xkb: xkbHandleActions: let wrapping take care of event delivery.
98536
98537 This is hopefully better than hardcodey calling CoreProcessPointerEvent.
98538 (cherry picked from commit 32d0440c7f6e604807cb14dd32349df6f22c903b)
98539
98540commit 352c5a311200bf491153fe9ef16126c5877a57bb
98541Author: Peter Hutterer <peter@cs.unisa.edu.au>
98542Date: Wed Sep 12 17:40:11 2007 +0930
98543
98544 dix: don't compress motion events from different devices (EventEnqueue)
98545
98546 (cherry picked from commit 8840829ab93c4eb62eb58753c015da5307133fe5)
98547
98548commit 8d3d027062c105b50863dce43b8070ec560bc12e
98549Author: Peter Hutterer <peter@cs.unisa.edu.au>
98550Date: Thu Sep 6 18:52:02 2007 +0930
98551
98552 dix: add XI event support to FixKeyState.
98553
98554 FixKeyState needs to be able to handle XI events, otherwise we get "impossible
98555 keyboard events" on server zaps and other special key combos.
98556 (cherry picked from commit 5ee409794ee604fcf84886f70429fc2d6b1ff4f1)
98557
98558commit 99e826e867c1c5520153c539ba07a884aec88d0c
98559Author: Peter Hutterer <peter@cs.unisa.edu.au>
98560Date: Thu Sep 6 18:49:57 2007 +0930
98561
98562 xkb: enable XI event processing for xkb.
98563
98564 XI events can now take the same processing paths as core events, and should do
98565 the correct state changes etc.
98566
98567 There's some cases where XKB will use KeyPress as type for an event to be
98568 delivered to the client. Stuck warnings in, not sure what the correct solution
98569 is yet.
98570
98571 (cherry picked from commit 6334d4e7be18de5f237c12a6dc20f75aa23477d0 with some
98572 additional compile fixes and non-MPX adaptations)
98573
98574commit 91077bfc50d54be37c217e377c55b6bf886a2fab
98575Author: Peter Hutterer <peter@cs.unisa.edu.au>
98576Date: Thu Jun 21 18:24:30 2007 +0930
98577
98578 Save processInputProc before wrapping it and restore it later, instead of
98579 using a hardcoded ProcessKeyboardEvent. Otherwise we lose the ability to
98580 process DeviceKeyEvents after the first key press.
98581
98582 This should be the correct fix now.
98583 (cherry picked from commit 4d5df14f2c4a3108a8c8adfcf4766c0d1a9daad2)
98584
98585commit 8b9481a113b56078191e2298bf590905978f6289
98586Author: Peter Hutterer <peter@cs.unisa.edu.au>
98587Date: Tue Sep 4 17:44:51 2007 +0930
98588
98589 xkb: Store the action filters per device in the XkbSrvInfoRec.
98590
98591 Using a global array for action filters is bad. If two keyboard hit a modifier
98592 at the same time, releaseing the first one will deactivate the filter and
98593 thus the second keyboard can never release the modifier again.
98594 (cherry picked from commit bfe6b4d2d9952a80f8dbc63eec974ef894e5c226)
98595
98596commit 493b83bd097372ae0023da9919da83af39e3fc1c
98597Author: Daniel Stone <daniel@fooishbar.org>
98598Date: Sat Oct 27 21:33:52 2007 +0300
98599
98600 XFree86: Remove ridiculous SIGIO debugging
98601
98602 YOU PRESSED A KEY
98603 AND AGAIN
98604 YOU RELEASED A KEY
98605 AND AGAIN
98606 YOU PRESSED A KEY
98607 AND AGAIN
98608
98609 ... not so much.
98610
98611commit a3d48de5f2b7eacf3193c60f0fb461912201210b
98612Author: Daniel Stone <daniel@fooishbar.org>
98613Date: Sat Oct 27 21:35:31 2007 +0300
98614
98615 XKB: Add more bits to xkbsrv.h
98616
98617 Add the device private index, given we use that in a macro here, and also the
98618 prototype for xkbUnwrapProc, since that's also useful.
98619
98620commit e29e69960d67aa4b7a4d1551af509dbac193f438
98621Author: Daniel Stone <daniel@fooishbar.org>
98622Date: Sat Oct 27 21:34:22 2007 +0300
98623
98624 Xi: Include XI protocol header in exevents.h
98625
98626 Make sure we have all the types we need to use this header.
98627
98628commit df57ae1639ba4f1719883c5bf868394e4748a022
98629Author: Daniel Stone <daniel@fooishbar.org>
98630Date: Sun Oct 28 15:46:26 2007 +0200
98631
98632 configure.ac/XFree86: Only build XF86Misc and XF86VidMode when appropriate
98633
98634 Don't build XF86Misc or XF86Vidmode in hw/xfree86/dixmod when it's been
98635 explicitly disabled in configure, or we don't have the proto modules
98636 installed.
98637
98638commit 3b7af72fe315c7c26c89838c0c5dacbe58765d0f
98639Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98640Date: Fri Oct 26 20:32:10 2007 -0400
98641
98642 xselinux: Add a SetDeviceContext request and stubs for more requests.
98643
98644commit 2251572062b2c25643671b8d5070de1c3f7ae976
98645Author: Aaron Plattner <aplattner@nvidia.com>
98646Date: Fri Oct 26 15:13:50 2007 -0700
98647
98648 Restore the CompositeGlyphs -> ps->Glyphs -> miGlyphs callchain to allow acceleration architectures to wrap above miGlyphs.
98649
98650commit 497aba894904b08b15bb19916e2a163f96212a7d
98651Author: Kristian Høgsberg <krh@redhat.com>
98652Date: Fri Oct 26 11:25:57 2007 -0400
98653
98654 Weed out some unused fluff from __GLXcontext.
98655
98656commit 0e749ceab421d72b48e18292c5ca0e337baf5ce8
98657Author: Bernardo Innocenti <bernie@codewiz.org>
98658Date: Sat Oct 20 20:31:57 2007 -0400
98659
98660 Include stddef.h for size_t.
98661
98662 This is required by a buggy version of the openssl/sha.h header
98663 which is distributed with Fedora 7.
98664
98665commit 80d8a3e1767609a92b0b85a7e92cea5f0748038a
98666Author: Kristian Høgsberg <krh@redhat.com>
98667Date: Fri Oct 26 10:19:57 2007 -0400
98668
98669 Quiet glx visual debug info.
98670
98671commit 27b13eeb56a5cbac00b381309604505002e77228
98672Author: Matthias Hopf <mhopf@suse.de>
98673Date: Fri Oct 26 13:15:07 2007 +0200
98674
98675 Nuke superfluous ",", added missing include for xf86i2c.h.
98676
98677commit 6d59bb5709a99ab60b482bbf3393ebffda7f9407
98678Author: Daniel Stone <daniel@fooishbar.org>
98679Date: Fri Oct 26 09:12:15 2007 +0300
98680
98681 XFree86 Misc/VidMode: Remove ridiculous debug ErrorFs
98682
98683 When we're building with --enable-debug, don't emit an ErrorF every time a
98684 function gets called.
98685
98686commit 5f9095f0d29bac0190d82c87a09cf32d6a34c17c
98687Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98688Date: Thu Oct 25 19:02:03 2007 -0400
98689
98690 registry: Remove synthetic bit from event types in lookup function.
98691
98692commit 8c6923018c7d71cd15d9cf4ef9e8528ef5ec7c2e
98693Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98694Date: Thu Oct 25 19:01:29 2007 -0400
98695
98696 xace: Add a "manage" access check when setting the Redirect event bits.
98697
98698commit 7d14ca59c5b942c09feaa2429c394cde9d8d3fd1
98699Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98700Date: Thu Oct 25 19:00:50 2007 -0400
98701
98702 xselinux: Don't include the client in the receive hook audit messages.
98703
98704commit f62277d421023b3150d3a1accb00a8206ab2bde3
98705Author: Kristian Høgsberg <krh@redhat.com>
98706Date: Thu Oct 25 18:48:39 2007 -0400
98707
98708 Clean up unused pVisual part of __GLXcontext.
98709
98710commit 30bcaa966d6b00f1630609a78db18dee683cc43d
98711Author: Kristian Høgsberg <krh@redhat.com>
98712Date: Thu Oct 25 18:48:06 2007 -0400
98713
98714 Make glx destroy path handle cases where the X window goes away first.
98715
98716commit 40de9fcf18930811dd5ae355c83275af887a9f83
98717Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98718Date: Thu Oct 25 12:35:01 2007 -0400
98719
98720 xselinux: Label the default device directly with the process context.
98721
98722commit b633d54b94d746d26e13014634d9f63bbb7e8f7d
98723Merge: 4b05f19 48ca596
98724Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98725Date: Thu Oct 25 12:19:30 2007 -0400
98726
98727 Merge branch 'master' into XACE-SELINUX
98728
98729 Conflicts:
98730
98731 GL/glx/glxscreens.c
98732 hw/xnest/Screen.c
98733 render/glyph.c
98734 render/glyphstr.h
98735 render/render.c
98736
98737commit 242f56f722243938e908d1957781ee53c2999783
98738Author: Naoki Hamada <nao@tom-yam.or.jp>
98739Date: Thu Oct 25 18:45:50 2007 +0300
98740
98741 Input: Fix key down test (bug #12858)
98742
98743 Fix the botched previous key_is_down test, which would give false positives.
98744 Also move key_autorepeats to a separate inline function.
98745
98746commit 4b05f19cb9e42d8c8eff5ca4e463f5bc2a05433d
98747Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98748Date: Wed Oct 24 19:59:58 2007 -0400
98749
98750 xselinux: Introduce a type transition when labeling events.
98751
98752commit 0d2ef187e77b12713d2a9661932fa01dba58a945
98753Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98754Date: Wed Oct 24 18:23:31 2007 -0400
98755
98756 xselinux: Add audit message fields for selection and event names.
98757
98758commit 48ca5961caee62f2980017a6bdc96a1b4c747727
98759Author: Matthias Hopf <mhopf@suse.de>
98760Date: Wed Oct 24 20:31:51 2007 +0200
98761
98762 Prefer configured DisplaySize to probed DDC data, if available.
98763
98764 Based on patch by Hong Liu <hong.liu@intel.com>.
98765
98766commit 0388a59a6ef212c497cc3f64d677b1ca5b410982
98767Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98768Date: Tue Oct 23 20:59:21 2007 -0400
98769
98770 Revert "registry: special case minor number when looking up core requests."
98771
98772 This reverts commit 31110d6837ee52fd654729d9e5c4b0c5395abab0.
98773
98774 This is handled properly by StandardMinorOpcode().
98775
98776commit 46521f529841e032e198e5df87974088548a68de
98777Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98778Date: Tue Oct 23 20:58:48 2007 -0400
98779
98780 xselinux: Add basic support for selection access control and redirection.
98781
98782 Probably not fully baked yet. It's difficult to test since so few apps
98783 actually follow the ICCCM with respect to cut & paste.
98784
98785commit b7ee005d327372c1e414ee6c526f9f7aee14bc86
98786Author: Peter Hutterer <peter@cs.unisa.edu.au>
98787Date: Wed Oct 24 10:26:12 2007 +0930
98788
98789 dix: Don't interfere grabs the interfering device is of different type.
98790
98791 A pointer device should be able to send events while the client has a core
98792 grab on a keyboard device, and likewise.
98793
98794commit 02508614b9f882ba1559fb4581dcf812118fdf89
98795Author: Peter Hutterer <peter@cs.unisa.edu.au>
98796Date: Wed Oct 24 10:09:18 2007 +0930
98797
98798 dix: check for core passive grabs on inferiors when replaying events.
98799
98800commit 825f09dffd94cfcd0562a01c5181998503851461
98801Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98802Date: Tue Oct 23 17:12:57 2007 -0400
98803
98804 xace: Still more changes to selection code. Removed the SelectionPtr from
98805 the hook - the hook only needs the Atom to control access to the selection
98806 object. Upgraded the SelectionCallback to take a client argument and
98807 additional type codes so that it can be used for redirection.
98808
98809commit 660557593ea961948722298ea8ffba83891c9914
98810Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98811Date: Tue Oct 23 14:46:37 2007 -0400
98812
98813 xselinux: Remove synthetic bit when looking up event type.
98814
98815commit d7db549db41a27aef28cff9bfb7973bc741f88b2
98816Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98817Date: Tue Oct 23 14:08:54 2007 -0400
98818
98819 xselinux: Unregister callbacks on server reset.
98820
98821commit 9e0a468af19d8e46330bcff37c9adc5e11d3aee7
98822Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98823Date: Tue Oct 23 13:35:30 2007 -0400
98824
98825 xace: try to pretend events were sent when a denial occurs.
98826
98827 Probably need to redo the error return paths in these functions
98828 at some point.
98829
98830commit d502521c3669f3f22b94c39a64ab63bfd92c6a97
98831Author: Pierre Willenbrock <pierre@pirsoft.dnsalias.org>
98832Date: Tue Oct 23 16:45:13 2007 +0200
98833
98834 EXA: Fix off-by-one in polyline drawing.
98835
98836commit a8a148919b84a293e3e7a49409ab833590357edb
98837Author: Matthias Hopf <mhopf@suse.de>
98838Date: Tue Oct 23 16:23:28 2007 +0200
98839
98840 Superfluous ','.
98841
98842commit 75f05086d04a90c3dcdcdd31bf79d7033708e3e0
98843Author: Matthias Hopf <mhopf@suse.de>
98844Date: Tue Oct 23 15:39:23 2007 +0200
98845
98846 Get rid of unnecessary GNU extended variadic macro.
98847
98848commit 7ef863720dc79107fc2cd17ce684366c87e001a4
98849Author: Peter Hutterer <peter@cs.unisa.edu.au>
98850Date: Tue Oct 23 17:28:03 2007 +0930
98851
98852 dix: Create new sprite for floating devices.
98853
98854 Floating devices get sprites, but still aren't spriteOwners. This prevents
98855 them from getting rendered, and also stops segfaulting.
98856 (not really solving the problems with keyboards though)
98857
98858commit d808d653d1fc16f1d5af76ab00fa862fb80fa3ba
98859Author: David Nusinow <dnusinow@debian.org>
98860Date: Mon Oct 22 21:30:04 2007 -0400
98861
98862 Define PCI_TXT_IDS_DIR more cleanly
98863
98864commit 29e0e180729a4f0cc020985a4de4c8bc4b9c7f5f
98865Author: Keith Packard <keithp@koto.keithp.com>
98866Date: Mon Oct 22 13:38:16 2007 -0700
98867
98868 Leave hardware-specified preferred modes alone when user preference exists.
98869
98870 Instead of removing the preference bit marking the hardware declared mode
98871 preference, leave it in place and just move the user preferred mode to the
98872 front of the list while marking it with the USERPREF bit which will cause it
98873 to be selected by the initial mode selection code.
98874
98875commit fbe19c66c36acfb484809111cf02579a3baf2f0f
98876Author: Michel Dänzer <michel@tungstengraphics.com>
98877Date: Mon Oct 22 18:28:03 2007 +0200
98878
98879 GLX: Fix leak of X pixmaps associated with GLX pixmaps.
98880
98881commit c6d36b1cee44a9cbb690dff62a4683d7f6fbf30c
98882Author: Michel Dänzer <michel@tungstengraphics.com>
98883Date: Mon Oct 22 18:28:03 2007 +0200
98884
98885 GLX: Don't crash on unused client array members when switching to/from console.
98886
98887commit be9b0e558d6172ef01ed880d47da897d2ceb25fa
98888Author: David Nusinow <dnusinow@debian.org>
98889Date: Sun Oct 21 23:49:14 2007 -0400
98890
98891 Add a note as to why the text file based PCI ID scheme is present at all
98892
98893commit b79b965bd9a96f79781e85c0428068caa1ba381b
98894Author: David Nusinow <dnusinow@debian.org>
98895Date: Sun Oct 21 23:46:54 2007 -0400
98896
98897 Fix compilation issue on FreeBSD (bug #12841)
98898
98899 Hide getline call by checking for glibc. If not, use fgetln instead. Even
98900 though this section is now #ifdef'ed for linux only, this should help make
98901 it more portable if non-linux folks end up wanting it.
98902
98903commit 7f5e71fa7347fb4ec2657ccc453831c04a0f288d
98904Author: David Nusinow <dnusinow@debian.org>
98905Date: Sun Oct 21 23:41:49 2007 -0400
98906
98907 Make PCI text file directory configurable
98908
98909commit ce7f6fe1268fef4f89aa21c7b44d73ecd98efe24
98910Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98911Date: Fri Oct 19 19:40:04 2007 -0400
98912
98913 xselinux: properly update sizes when dynamic arrays are resized...
98914
98915commit 4b14c9a9cd2033d3839c4ba364d41ab4c4b198ab
98916Author: Eric Anholt <eric@anholt.net>
98917Date: Fri Oct 19 16:34:54 2007 -0700
98918
98919 Replace calls to Glyphs screen hook with CompositeGlyphs and remove dead code.
98920
98921 Not all of the DDX/miext Glyphs hook implementations have been removed, but
98922 they should be.
98923
98924commit a3a95d3475dc91ed2e8a55bf484a6b3f2b5ac32a
98925Merge: a358b87 7e1cada
98926Author: Eric Anholt <eric@anholt.net>
98927Date: Fri Oct 19 15:44:17 2007 -0700
98928
98929 Merge branch 'master' into glyph-pixmaps
98930
98931 Conflicts:
98932
98933 configure.ac
98934
98935commit 12e889d202ac9849f534c51167cbfed91c32027a
98936Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
98937Date: Fri Oct 19 18:43:38 2007 -0400
98938
98939 xace: Bug fixes, name changes to selection access hooks and fields.
98940
98941commit 7e1cada6c6b9fa41ef3ead00bf7725d626dda193
98942Author: Eric Anholt <eric@anholt.net>
98943Date: Fri Oct 19 15:16:19 2007 -0700
98944
98945 Disable debian pci-id-list autoconfig code on non-linux.
98946
98947 It contains static paths, fails to build on non-glibc, and apparently just
98948 exists to support distributions managing binary drivers and open-source drivers
98949 together. Also restores previous code for fallback to vesa if nothing is
98950 detected.
98951
98952commit 89c6108531e603bdc81faf2ea860f318a2e94a39
98953Author: Kristian Høgsberg <krh@redhat.com>
98954Date: Fri Oct 19 16:21:54 2007 -0400
98955
98956 Fix software GL to provide a list of supported fbconfigs like the DRI case.
98957
98958commit 74e84b8d891632141a60e2d29463ab46f622df3f
98959Author: Matthias Hopf <mhopf@suse.de>
98960Date: Fri Oct 19 17:28:15 2007 +0200
98961
98962 Only issue XF86_APM_CAPABILITY_CHANGED for video change ACPI events (not e.g. brightness etc.).
98963
98964commit 6fdf3ddfef2707755cec0fc9cbce354fd472bb52
98965Author: Matthias Hopf <mhopf@suse.de>
98966Date: Fri Oct 19 17:20:21 2007 +0200
98967
98968 Fix ids of ACPI events according to ACPI spec.
98969
98970commit d9caa469b9bb4eb6125b890820853062fc2c4441
98971Author: Peter Hutterer <peter@cs.unisa.edu.au>
98972Date: Fri Oct 19 15:26:09 2007 +0930
98973
98974 Xi: allow VCP/VCK to be OpenDevice'd as well.
98975
98976 All devices deserve to be equal. Except master devices, they are a bit more
98977 equal than the others.
98978
98979commit 155e2c559ed0dbf31b6d39d48648a3ee22b37635
98980Author: Peter Hutterer <peter@cs.unisa.edu.au>
98981Date: Fri Oct 19 14:37:07 2007 +0930
98982
98983 Xi: advertise as XInput v2 capable
98984
98985commit 5ba738935f0d786e4670adf3b05ad42fc5e710fd
98986Author: Peter Hutterer <peter@cs.unisa.edu.au>
98987Date: Fri Oct 19 14:36:37 2007 +0930
98988
98989 Xi: remove ChangePointerKeyboardPairing in favour of ChangeDeviceHierarchy.
98990
98991commit 6dcde0e9c1d068d9fc4a772d29d1d4c6cc57aeb9
98992Author: Peter Hutterer <peter@cs.unisa.edu.au>
98993Date: Thu Oct 18 12:21:07 2007 +0930
98994
98995 Xi: set master device's id in ListDevices Reply.
98996
98997commit cfcc6e14b9a15f7979ba0df9c105cf563bef98fa
98998Author: Peter Hutterer <peter@cs.unisa.edu.au>
98999Date: Fri Oct 19 11:28:44 2007 +0930
99000
99001 Xi: return all master devices as type IsXPointer/Keyboard when listing devs.
99002
99003 Slave devices are reported as IsXExtensionPointer/Keyboard.
99004
99005commit 8e3c1dfc48930c455529313a42efa35e3b9071b2
99006Author: Kristian Høgsberg <krh@redhat.com>
99007Date: Thu Oct 18 21:01:51 2007 -0400
99008
99009 Introduce a new "GlxVisuals" option that controls which visuals are added.
99010
99011 Right now we default to "all" which gives us a situation much like before,
99012 but when the "typical" option is implemented, we can change the default and
99013 reduce the number of visuals the GLX module bloats the X server with.
99014
99015commit 9e3cfd950d279353f1795352ad2c4132aabc1bab
99016Author: Kristian Høgsberg <krh@redhat.com>
99017Date: Thu Oct 18 19:12:24 2007 -0400
99018
99019 Add code to set up GLX visuals and add one for the first double buffered FBConfig.
99020
99021commit 8fc635e6a8072c7cd2777d804dd6f8eda2eecc15
99022Author: Kristian Høgsberg <krh@redhat.com>
99023Date: Thu Oct 18 15:51:11 2007 -0400
99024
99025 Separate handling of GLX visuals and GLX FBConfigs.
99026
99027 XIDs for GLX visuals and FBConfigs used to be interchangable and the list of
99028 GLX visuals was identical to the list for FBConfigs. This patch splits handling
99029 of these two data types and allows the X server to pick and choose the FBConfigs
99030 that are exposed as visuals.
99031
99032commit 0af8180683247955ce4cfd48e6a5b4d00bbe618a
99033Author: Kristian Høgsberg <krh@redhat.com>
99034Date: Wed Oct 17 17:18:47 2007 -0400
99035
99036 Drop VisualConfigPriv support from the DRI module.
99037
99038 No DDX driver uses this, and this patch stops the DRI module from
99039 poking around GLX module data structures.
99040
99041commit 7d74690536b64f7b8e8036507ab7790807349c50
99042Author: Kristian Høgsberg <krh@redhat.com>
99043Date: Sun Oct 14 15:39:21 2007 -0400
99044
99045 Simplify and clean up GLX visual initialization.
99046
99047 Instead of the fragile setup where we filter the modes common between the
99048 DDX generated GLX visuals and the DRI driver generated fbconfigs, we now
99049 just take the fbconfigs returned by the DRI driver to be our supported set.
99050
99051commit ab4bce02a9457dd9c86b774fc74caf3dd6b287ca
99052Author: Eric Anholt <eric@anholt.net>
99053Date: Thu Oct 18 15:22:42 2007 -0700
99054
99055 Add a quirk for Philips 107P5 which lacks the preferred bit on detailed timing.
99056
99057 Also fix the prefer-large-75 quirk if the prefer-first-detailed bit was set,
99058 though it's not the case for the existing prefer-large-75 consumer.
99059
99060commit 55a96aa6b0995fda6660b7e78c85b955a62b9735
99061Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99062Date: Thu Oct 18 14:11:11 2007 -0400
99063
99064 xselinux: add basic event labeling.
99065
99066commit e974bc1233608ec09fbd40b12217925e4d2205aa
99067Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99068Date: Thu Oct 18 12:33:39 2007 -0400
99069
99070 xselinux: add hooks for send and receive access.
99071
99072commit 06eb830169afd0631a31e8846c7d2533c49ea378
99073Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99074Date: Thu Oct 18 12:31:14 2007 -0400
99075
99076 xace: Fix bug in AddPassiveGrabToList(), was using wrong GrabPtr.
99077
99078commit 6107a245035366fe762756b6aa05ac0e3a5482bb
99079Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99080Date: Thu Oct 18 12:24:55 2007 -0400
99081
99082 dix: Add client parameter to AddPassiveGrabsToList().
99083
99084commit ce50bfd3369686cfecee5a138bd84ef1107a249d
99085Author: Michel Dänzer <michel@tungstengraphics.com>
99086Date: Thu Oct 18 17:44:48 2007 +0200
99087
99088 EXA: Skip empty glyphs.
99089
99090commit 5d74416740de883b7ef0994afea4bbd4d3901be0
99091Author: Michel Dänzer <michel@tungstengraphics.com>
99092Date: Thu Oct 18 17:44:14 2007 +0200
99093
99094 EXA: Don't attempt to move in pixmaps that can't be accelerated.
99095
99096 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=12815 .
99097
99098commit 31110d6837ee52fd654729d9e5c4b0c5395abab0
99099Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99100Date: Thu Oct 18 10:30:44 2007 -0400
99101
99102 registry: special case minor number when looking up core requests.
99103
99104commit e3fd90ae9c3ddfc5d78e62614e311b73505d7ead
99105Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99106Date: Thu Oct 18 10:29:10 2007 -0400
99107
99108 registry: Add "X11:" prefix to core protocol names.
99109
99110commit aa340b2c7cbe9ddab53cff08c8ba165558209187
99111Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99112Date: Wed Oct 17 19:27:16 2007 -0400
99113
99114 xselinux: add hook for device acceses.
99115
99116commit 503f918f55d0cb29585d83b022bbb8dc29f446c5
99117Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99118Date: Wed Oct 17 19:14:15 2007 -0400
99119
99120 xselinux: Move functions around; add some more comments.
99121
99122commit 50b27e1ad2a98d36728dc8157492ef5c59c132cd
99123Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99124Date: Wed Oct 17 16:09:40 2007 -0400
99125
99126 devPrivates rework: update new GL/glxext code.
99127
99128 Need to merge so this type of thing stops happening.
99129
99130commit c3f7b862550fa8f46633162f83db8c27e46fc672
99131Merge: af4dde0 feac075
99132Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99133Date: Wed Oct 17 15:00:54 2007 -0400
99134
99135 Merge branch 'master' into XACE-SELINUX
99136
99137 Conflicts:
99138
99139 os/access.c
99140
99141commit af4dde0ac19ecec1d0ad988eb25b15401e7c6b36
99142Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99143Date: Wed Oct 17 14:13:02 2007 -0400
99144
99145 xselinux: Remove config file, this has been moved to the policy.
99146
99147commit baabae623b3658196b67a710dc72663c2105bf31
99148Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99149Date: Wed Oct 17 13:54:56 2007 -0400
99150
99151 xselinux: Started reworking extension using new XACE hooks.
99152
99153commit db66e66dbf26b91c655f1659859c022cc31f0db6
99154Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99155Date: Wed Oct 17 13:51:11 2007 -0400
99156
99157 xace: Add an access_mode field to the extension structure.
99158 This allows the same callback to be used for both extension hooks.
99159
99160commit e3a8cbe523bae8b771ad3c8ad497f4444f6d05d5
99161Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99162Date: Wed Oct 17 13:48:44 2007 -0400
99163
99164 xace: add creation/labeling hook to CreateRootWindow().
99165
99166commit feac0759522cbdc3e61ccfa373df735903c5cb27
99167Author: Keith Packard <keithp@koto.keithp.com>
99168Date: Wed Oct 17 11:42:28 2007 +0800
99169
99170 Make config file preferred mode override monitor preferred mode.
99171
99172 Add a new even-more-preferred bit to each mode which is used to make config
99173 file preferences selected instead of the monitor preferred mode.
99174
99175commit f2da10f7bc2ddb6ad2f18b793afc10d04b97c51c
99176Author: Keith Packard <keithp@koto.keithp.com>
99177Date: Wed Oct 17 10:50:22 2007 +0800
99178
99179 KDRIVE_LOCAL_LIBS includes some system libraries, not just internal x server libs
99180
99181commit 9aadde377991bfbd88524d02106bec3dedd6e7c9
99182Author: Peter Hutterer <peter@cs.unisa.edu.au>
99183Date: Wed Oct 17 12:05:50 2007 +0930
99184
99185 mi: don't call UndisplayCursor for non-sprite-owners.
99186
99187 Segfaults are bad.
99188
99189commit 8b7c4249d82d07e852d8d8c15c7ab9977dd1f31c
99190Author: Peter Hutterer <peter@cs.unisa.edu.au>
99191Date: Wed Oct 17 10:41:58 2007 +0930
99192
99193 mi: call processInputProc for master devices after slave event processing.
99194
99195 More work is needed to sort out grabs though.
99196
99197commit ae9fc10adec8f9bf0492d14d435f8f11e0163b27
99198Author: Peter Hutterer <peter@cs.unisa.edu.au>
99199Date: Wed Oct 17 10:41:04 2007 +0930
99200
99201 Xi: don't send core events for slave devices.
99202
99203commit de70cfdbe60eb6e7bf3e74dfd1ac34de554deff1
99204Author: Peter Hutterer <peter@cs.unisa.edu.au>
99205Date: Wed Oct 17 09:52:30 2007 +0930
99206
99207 mi: fix some macros to allow multiple cursors for master devices.
99208
99209 Macros defaulted to inputInfo.pointe rfor devices that weren't spriteOwners.
99210 Changed to take the device's master device now.
99211
99212 This includes sticking in a number of checks and warnings that cursor
99213 rendering won't be called for floating devices.
99214
99215commit 773f6491c1cc8819038e753d08c32ba213f80f8f
99216Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99217Date: Tue Oct 16 19:11:36 2007 -0400
99218
99219 xace: update the DeleteProperty prototype to include the client argument.
99220 This should have been part of 8f23d40068151ad85cde239d07031284f0b2c4dc.
99221
99222commit 9dde53ed179336c7b483c9a94a97182ad1777dfb
99223Author: Alan Hourihane <alanh@tungstengraphics.com>
99224Date: Tue Oct 16 19:55:56 2007 +0100
99225
99226 Fix walking of GLX providers.
99227
99228commit 70a5d33c9e41c077a8cd92abd43376e2956d3aed
99229Author: Michel Dänzer <michel@tungstengraphics.com>
99230Date: Tue Oct 16 12:46:07 2007 +0200
99231
99232 Always check the return value of __glXGetDrawable first.
99233
99234 Fixes spurious GLX protocol errors because __glXGetDrawable doesn't set the
99235 error code in case of success. Maybe it should, though.
99236
99237commit 1d9ebbac8c589cae7e4952083692b6d148def9bc
99238Author: Peter Hutterer <peter@cs.unisa.edu.au>
99239Date: Tue Oct 16 18:46:12 2007 +0930
99240
99241 dix: Make InitCoreDevices use AllocMasterDevice.
99242
99243 Also change naming a bit, append "pointer" and "keyboard" to master devices
99244 instead of -ptr and -keybd.
99245
99246commit 299573f4617c3b5599bb65069e96d050277b9471
99247Author: Peter Hutterer <peter@cs.unisa.edu.au>
99248Date: Tue Oct 16 18:40:15 2007 +0930
99249
99250 dix: add AllocMasterDevice for creation of new master devices.
99251
99252 Devices are initiated pretty much the same as the core devices.
99253
99254commit 5fe9bfd23f17b84c3afaa82f75a7c517c9f8e0d3
99255Author: Peter Hutterer <peter@cs.unisa.edu.au>
99256Date: Tue Oct 16 18:23:48 2007 +0930
99257
99258 xfree86: NIDR: don't call PairDevices explicitly.
99259 set isMaster to FALSE explicitly.
99260
99261 Pairing isn't necessary, attachment should be done in EnableDevices.
99262
99263commit 5eb033835e92ea951cc385fd709af9656b3772d8
99264Author: Peter Hutterer <peter@cs.unisa.edu.au>
99265Date: Tue Oct 16 18:03:52 2007 +0930
99266
99267 mi: don't exclude inputInfo.pointer from rendering.
99268
99269commit b697c4ed145968d3c3281bb85e628f1b068b09fc
99270Author: Peter Hutterer <peter@cs.unisa.edu.au>
99271Date: Tue Oct 16 17:47:06 2007 +0930
99272
99273 dix: CheckPassiveGrabOnWindow: only get paired device for non-keyboards.
99274
99275commit caa69d8f7b92b80669df02e1001409d7c653e7e4
99276Author: Peter Hutterer <peter@cs.unisa.edu.au>
99277Date: Tue Oct 16 17:38:13 2007 +0930
99278
99279 dix: fix detritus from adding lastSlave field.
99280
99281commit 90d077e537ac4cb7d79d67afcf3984a3e8d65fe8
99282Author: Peter Hutterer <peter@cs.unisa.edu.au>
99283Date: Tue Oct 16 17:37:33 2007 +0930
99284
99285 dix: GetPointerEvents: get state from master device before applying valuators.
99286
99287 We need to get lastx/y from the master device before we start applying
99288 acceleration etc. Otherwise we get jumping cursors, which is reasonably bad.
99289
99290commit b6ccf721b0a582150858d68f91967fb4e319c340
99291Author: Peter Hutterer <peter@cs.unisa.edu.au>
99292Date: Tue Oct 16 17:35:06 2007 +0930
99293
99294 include: add "lastSlave" field to DeviceIntRec.
99295
99296 Set to the last slave device that routed events through the master.
99297
99298commit e6bd8ae0608bd8379c5ac962f69cd0bcc54f9734
99299Author: Peter Hutterer <peter@cs.unisa.edu.au>
99300Date: Tue Oct 16 15:59:30 2007 +0930
99301
99302 dix: set the device's sprite when using AttachDevice
99303
99304 Sprite is set to the master device's sprite.
99305
99306commit 271d4c3d6255318aabb9ad1ea444f534ed456f0b
99307Author: Peter Hutterer <peter@cs.unisa.edu.au>
99308Date: Tue Oct 16 15:50:07 2007 +0930
99309
99310 xfree86: remove XI86_SHARED_POINTER flag.
99311
99312 Not needed anymore. By default, all devices are slaves of first master device.
99313
99314commit c7b878b90406781c97db751a9b454e2b6baee0e1
99315Author: Peter Hutterer <peter@cs.unisa.edu.au>
99316Date: Tue Oct 16 15:07:31 2007 +0930
99317
99318 dix: don't treat VCK/VCP separately in DevHasCursor, IsPointer/KeyboardDevice.
99319
99320commit e9f149fb56747f7eaa0f714dce97b067b007c47e
99321Author: Peter Hutterer <peter@cs.unisa.edu.au>
99322Date: Tue Oct 16 14:24:20 2007 +0930
99323
99324 Fix up detritus from removing GetPairedPointer/Keyboard.
99325
99326commit 840bde3d32f3627dfc3d7d8b6564a61a8014933b
99327Author: Peter Hutterer <peter@cs.unisa.edu.au>
99328Date: Tue Oct 16 13:09:35 2007 +0930
99329
99330 dix: GetPairedDevice: return dev paired with master for slave devices.
99331
99332commit 6c259a08d917f94fd7381453b625b07826d3ef9c
99333Author: Peter Hutterer <peter@cs.unisa.edu.au>
99334Date: Tue Oct 16 13:09:19 2007 +0930
99335
99336 include: remove unused GetPairedPointer/Keyboard declarations.
99337
99338commit fe97f7c54a1b42acd542696b6cdc9e83e89548f3
99339Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99340Date: Mon Oct 15 22:46:08 2007 -0400
99341
99342 registry: Add some missing #include's.
99343
99344commit 9ecbbf198f4cec398897736e173e7e8c56bf6f94
99345Author: Peter Hutterer <peter@cs.unisa.edu.au>
99346Date: Sun Oct 14 08:00:06 2007 +0930
99347
99348 dix: adjust PickPointer and PickKeyboard to get the first master device.
99349
99350 Simplifies it a lot, since we always have at least one master device
99351 available at all times, so less mucking around.
99352
99353commit 0c5f65ecd3ad11fbdb1cab3cb1d0eb4f33bb4e35
99354Author: Peter Hutterer <peter@cs.unisa.edu.au>
99355Date: Sun Oct 14 07:57:24 2007 +0930
99356
99357 dix: don't allow slave devices as ClientPointer
99358
99359commit 9ccc9ab6f29f68298b68cdb5c9b4bd57a095f05a
99360Author: Peter Hutterer <peter@cs.unisa.edu.au>
99361Date: Sun Oct 14 07:53:00 2007 +0930
99362
99363 dix: don't call CloseDevice on VCP/VCK separately.
99364
99365commit e4fd981b48723b77a6c1a528638a771b9cc35472
99366Author: Peter Hutterer <peter@cs.unisa.edu.au>
99367Date: Sun Oct 14 07:52:11 2007 +0930
99368
99369 dix: check for isMaster before calling DeviceCursorCleanup.
99370
99371commit 7503d1340726e0a529f04304d177ed2ceb8fbc91
99372Author: Peter Hutterer <peter@cs.unisa.edu.au>
99373Date: Sun Oct 14 07:50:13 2007 +0930
99374
99375 dix: don't check VCK and VCP separately when looking for device ids.
99376
99377 Both are part of the device list again, so we cover them there.
99378
99379commit 032e906711202d376af95b37bb0cdf14a3648256
99380Author: Peter Hutterer <peter@cs.unisa.edu.au>
99381Date: Tue Oct 16 11:57:42 2007 +0930
99382
99383 dix: Remove GetPairedPointer, GetPairedKeyboard in favour of GetPairedDevice.
99384
99385commit 853ea337bdad17f8f6ec7d940de14ce2cbbbf93e
99386Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99387Date: Mon Oct 15 22:13:02 2007 -0400
99388
99389 registry: Register XvMC extension protocol names.
99390
99391commit 58c3240fcbec23aad122e1c340f6bb6d3b18f779
99392Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99393Date: Mon Oct 15 22:09:47 2007 -0400
99394
99395 registry: Register MIT-SCREEN-SAVER extension protocol names.
99396
99397commit 3e07e73fefc100e491d1e465cb162373d1d82425
99398Author: Peter Hutterer <peter@cs.unisa.edu.au>
99399Date: Tue Oct 16 11:34:29 2007 +0930
99400
99401 dix: remove pairing/attachment from InitAndStartDevices.
99402
99403 If we enabled in the correct order, this has all been done already.
99404
99405commit 4e274e90e16b1d954391e1af3e2074fb10f70ee7
99406Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99407Date: Mon Oct 15 22:04:20 2007 -0400
99408
99409 registry: Register SHAPE extension protocol names.
99410
99411commit 9f597f6c87e0b14cc382d8e5929e42f822db4329
99412Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99413Date: Mon Oct 15 22:01:20 2007 -0400
99414
99415 registry: Register SYNC extension protocol names.
99416
99417commit 38baac71bdbb8c7e882e3e39133615cfed894a6b
99418Author: Peter Hutterer <peter@cs.unisa.edu.au>
99419Date: Sun Oct 14 04:21:42 2007 +0930
99420
99421 dix: Set bidirectional pairing, don't allow pairing with already paired devs.
99422
99423commit 70efd3d06a15093661933bda4ec21e306dece4a4
99424Author: Peter Hutterer <peter@cs.unisa.edu.au>
99425Date: Tue Oct 16 11:30:32 2007 +0930
99426
99427 dix: fix up Activate/Enable/Disable device.
99428
99429 Set isMaster for VCP/VCK.
99430 Init sprites for master pointer devices.
99431 Pair master kbds with master pointers (1:1 pairing!).
99432 Attach other devices to VCP/VCK.
99433
99434commit b504678ba5407a6fd8d47d051305f7c3d5606dfe
99435Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99436Date: Mon Oct 15 21:54:25 2007 -0400
99437
99438 registry: Register APPGROUP extension protocol names.
99439
99440commit 204f2dc89ef662b57400b128c30c15e8cf32f323
99441Author: Peter Hutterer <peter@cs.unisa.edu.au>
99442Date: Sun Oct 14 04:13:06 2007 +0930
99443
99444 include: add "isMaster" field to DeviceIntRec.
99445
99446 Set to TRUE for master devices.
99447
99448 Also fixing up comment for paired field in spriteInfo, will be set
99449 bidirectional from now on.
99450
99451commit bd7d5255ce4865b684f7d8bcf80ba9872a1af22d
99452Author: Peter Hutterer <peter@cs.unisa.edu.au>
99453Date: Tue Oct 16 11:18:31 2007 +0930
99454
99455 dix: add AttachDevice, needed to attach a slave device to a master device.
99456
99457 For now, we don't allow attaching slaves to other slaves, and we don't allow
99458 pairing slaves with other slaves.
99459 Pairing is for master keyboard->master pointer only.
99460 Attaching is for slave device->master device only.
99461
99462commit 6ec35a8cf539c900b334dd6df146b394f54e3706
99463Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99464Date: Mon Oct 15 21:46:56 2007 -0400
99465
99466 registry: Register BigRequests extension protocol names.
99467
99468commit 1254cc399c53eadcc32eeabf69990ed2526c7ae0
99469Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99470Date: Mon Oct 15 21:43:06 2007 -0400
99471
99472 registry: Register Fontcache extension protocol names.
99473
99474commit e987648cf2c21dcbd77dd9a71793090a48e4f521
99475Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99476Date: Mon Oct 15 21:38:56 2007 -0400
99477
99478 registry: Register EVI extension protocol names.
99479
99480commit be1565f6b8fb09eba9941d6f7c485bf5fb25fe7a
99481Author: Peter Hutterer <peter@cs.unisa.edu.au>
99482Date: Sun Oct 14 01:53:54 2007 +0930
99483
99484 dix: Set core devices to ProcessOtherEvents, set sendCore and XKB.
99485
99486 Even the virtual core devices should send through ProcessOtherEvents.
99487
99488commit ce9bf9a19185a36ac2f7ae75acd320ab8d03d247
99489Author: Peter Hutterer <peter@cs.unisa.edu.au>
99490Date: Sun Oct 14 01:50:40 2007 +0930
99491
99492 include: add "master" field to DeviceIntRec.
99493
99494 "master" points to the device this device is attached to. Event sent by the
99495 device will also be routed through the master.
99496 master and spriteOwner are mutually exclusive.
99497
99498commit f6226d3bfe1515058e2092e8662ae87825501209
99499Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99500Date: Mon Oct 15 21:35:12 2007 -0400
99501
99502 registry: Register TOG-CUP extension protocol names.
99503
99504commit 7e182a5d89d618e20dcc77850131690733322d39
99505Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99506Date: Mon Oct 15 21:13:39 2007 -0400
99507
99508 registry: Register MIT-MISC extension protocol names.
99509
99510commit 32f6171862461d17ebea58a2fb6ddd16ac71358c
99511Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99512Date: Mon Oct 15 21:10:14 2007 -0400
99513
99514 registry: Register XF86Bigfont extension protocol names.
99515
99516commit 12766c5b5ffdab95255a63b2c8421ee773fd43b5
99517Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99518Date: Mon Oct 15 21:00:46 2007 -0400
99519
99520 registry: Register Xv extension protocol names.
99521
99522commit 35ae03871af88b2f420dd83448011a077852d7a0
99523Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99524Date: Mon Oct 15 20:50:26 2007 -0400
99525
99526 registry: Register XC-MISC extension protocol names.
99527
99528commit 32fe282d5b8306514d641e15bc6d9fd4ab360977
99529Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99530Date: Mon Oct 15 20:45:18 2007 -0400
99531
99532 registry: Register XTest extension protocol names.
99533
99534commit 3877faf7d9fe00ed634077e38a198ae4b91a2bb4
99535Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99536Date: Mon Oct 15 20:35:13 2007 -0400
99537
99538 registry: Register Multibuffer extension protocol names.
99539
99540commit 16764a2d299c7c0c98002aadd52ab4a1a36758c3
99541Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99542Date: Mon Oct 15 20:29:23 2007 -0400
99543
99544 registry: Register DPMS extension protocol names.
99545
99546commit f077578e42eee424b0e534774574c84af9d6f85b
99547Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99548Date: Mon Oct 15 19:43:02 2007 -0400
99549
99550 registry: Register XPrint extension protocol names.
99551
99552commit 5c8b1a91726817816d20faefad21c7a68ab634cc
99553Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99554Date: Mon Oct 15 19:35:04 2007 -0400
99555
99556 registry: Register Resource extension protocol names.
99557
99558commit 48891d5696f56711f23743cb03be39cf6b26c522
99559Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99560Date: Mon Oct 15 19:32:01 2007 -0400
99561
99562 registry: Register EVIE extension protocol names.
99563
99564commit 2c9646ad4e65bb061d910c9e2b1a8a978f21fa17
99565Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99566Date: Mon Oct 15 19:18:03 2007 -0400
99567
99568 registry: Register SHM extension protocol names.
99569
99570commit 8964c6d8e14ae47798762191e359b2bf138ca32e
99571Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99572Date: Mon Oct 15 19:10:51 2007 -0400
99573
99574 registry: Register RENDER extension protocol names.
99575
99576commit c827db57e4d9ca14c82b099dcfc9b7a0c0b5ba0a
99577Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99578Date: Mon Oct 15 19:06:01 2007 -0400
99579
99580 registry: Register RANDR extension protocol names.
99581
99582commit 20db50b4c44a14f7eeac2b1de17ada68482521da
99583Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99584Date: Mon Oct 15 18:56:59 2007 -0400
99585
99586 registry: Register DAMAGE extension protocol names.
99587
99588commit b38a91993364aa80cfd99721e319e1458d9fb760
99589Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99590Date: Mon Oct 15 18:50:06 2007 -0400
99591
99592 registry: Register XTrap extension protocol names.
99593
99594commit 106758893b68033f14f69c4ee6591fb6a149ba37
99595Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99596Date: Mon Oct 15 18:31:40 2007 -0400
99597
99598 registry: Register XFixes extension protocol names.
99599
99600commit ea09c9acc8f0d5577f54c864ff88b7f03d93b2f4
99601Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99602Date: Mon Oct 15 18:25:12 2007 -0400
99603
99604 registry: Register Record extension protocol names.
99605
99606commit 2e1e5be1d9067816525aa13a1d818e8ca6899599
99607Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99608Date: Mon Oct 15 18:18:57 2007 -0400
99609
99610 registry: Register DBE extension protocol names.
99611
99612commit 4c3285c883cc50a91bc5262bbc9d073d816f860a
99613Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99614Date: Mon Oct 15 18:04:47 2007 -0400
99615
99616 registry: Register WINDOWSWM extension protocol names.
99617
99618commit 3815284e899b61731b6a63c4ba14c5d773e24eb6
99619Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99620Date: Mon Oct 15 17:57:59 2007 -0400
99621
99622 registry: Register XF86DGA extension protocol names.
99623
99624commit 2cd1b32b77e0ceeaccb3f01c4ac13a97c557668c
99625Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99626Date: Mon Oct 15 17:37:58 2007 -0400
99627
99628 registry: Register XF86Misc extension protocol names.
99629
99630commit 960677e876c068400fb45e1764bb5470cd8c389f
99631Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99632Date: Mon Oct 15 17:30:00 2007 -0400
99633
99634 registry: Register XF86VidMode extension protocol names.
99635
99636commit b7786724080fd3928ef7b8c294346661d7ffd90b
99637Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99638Date: Mon Oct 15 17:15:34 2007 -0400
99639
99640 registry: Register XF86DRI extension protocol names.
99641
99642commit eee46b4681ec55297604b0425705f2b18381f7ca
99643Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99644Date: Mon Oct 15 17:06:22 2007 -0400
99645
99646 registry: Register APPLEWM extension protocol names.
99647
99648commit b9f5ab98c8dea36dcce1ad15fd2e059a77e77c39
99649Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99650Date: Mon Oct 15 16:18:37 2007 -0400
99651
99652 registry: Register XINERAMA extension protocol names.
99653
99654commit 3464b419230c6d17e940d967b567c5d2cb22d232
99655Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99656Date: Mon Oct 15 15:45:12 2007 -0400
99657
99658 registry: Register APPLEDRI extension protocol names.
99659
99660commit 32f3f5a1e7654f8bb43ea16b9227b3994e616739
99661Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99662Date: Mon Oct 15 15:32:17 2007 -0400
99663
99664 registry: Register DMX extension protocol names.
99665
99666commit 166ef972febc00c665e1d5aeb68e75d7bbcf9879
99667Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99668Date: Mon Oct 15 15:22:02 2007 -0400
99669
99670 registry: Register composite extension protocol names.
99671
99672commit a5cf3f21f712e46dbf9bca289e67be75f2b531d3
99673Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99674Date: Mon Oct 15 14:27:32 2007 -0400
99675
99676 registry: Register XKB extension protocol names.
99677
99678commit 5277a6ff589b5ddb475b90e1aaf5dbd9172d9711
99679Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99680Date: Mon Oct 15 13:38:35 2007 -0400
99681
99682 registry: Register Input extension protocol names.
99683
99684commit 8b5078b7d9ec08a588a78eb49096b698c6f4916b
99685Author: Kristian Høgsberg <krh@redhat.com>
99686Date: Mon Oct 15 13:10:10 2007 -0400
99687
99688 Fix byte swapping for GetFBConfigs.
99689
99690commit c922d2eebe29f08f463ee76293dc9042712fb21c
99691Author: George Sapountzis <gsap7@yahoo.gr>
99692Date: Sun Oct 14 20:40:18 2007 +0300
99693
99694 glx: drop duplicate GetDrawableAttributesSGIX declarations.
99695
99696 They are officially autogenerated in indirect_dispatch.h now.
99697
99698commit 3d4eb17b38dcb1468493f3686dc5ea3623ef9a73
99699Author: Ben Byer <bbyer@bbyer.local>
99700Date: Sun Oct 14 18:07:03 2007 -0700
99701
99702 mass change from #ifdef i386 to #ifdef __i386__ to conform to ANSI
99703
99704commit ec0fc012e91e703bb399a380df2912f71957a220
99705Author: Kristian Høgsberg <krh@redhat.com>
99706Date: Sun Oct 14 14:59:12 2007 -0400
99707
99708 Fix GLX byteswapping.
99709
99710commit 927757e1028f45f7fd94b9a2ab35567e0f34b2a8
99711Merge: ccda4b6 17ffc34
99712Author: Kristian Høgsberg <krh@redhat.com>
99713Date: Fri Oct 12 19:28:39 2007 -0400
99714
99715 Merge branch 'master' of git://git.freedesktop.org/git/xorg/xserver
99716
99717commit ccda4b66bdfc179e661b24adc8ec72bb7db75b2b
99718Author: Kristian Høgsberg <krh@redhat.com>
99719Date: Mon Aug 27 14:43:48 2007 -0400
99720
99721 Implement ChangeDrawableAttributes so we can support GLX_EVENT_MASK.
99722
99723 We never need to actually send out the clobber event, so this should be
99724 sufficient for GLX 1.3.
99725
99726commit 695eb8b2e88abc9fa3a76d8da48c3214c7dd1f08
99727Author: Kristian Høgsberg <krh@redhat.com>
99728Date: Mon Aug 27 14:23:50 2007 -0400
99729
99730 Implement GLX pbuffers.
99731
99732commit 516c181f57367847c3f317f8f8f5cc3211026f4c
99733Author: Kristian Høgsberg <krh@redhat.com>
99734Date: Thu Aug 23 21:40:01 2007 -0400
99735
99736 Add dispatching for GLX_SGIX_pbuffer.
99737
99738commit d7ded953c4d263e3dd3374dc03eea19e80c05bc6
99739Author: Kristian Høgsberg <krh@redhat.com>
99740Date: Thu Aug 23 19:07:52 2007 -0400
99741
99742 Implment GLXPixmaps.
99743
99744commit 526f40434c86548830c4f72940462b6253fe9790
99745Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99746Date: Fri Oct 12 18:18:00 2007 -0400
99747
99748 NULL is not a valid argument to CreatePicture, please use serverClient
99749 as the client argument if no real client is creating the object.
99750
99751commit 2b0847c7aaf95b6b10f062dfff601032b2430a9b
99752Author: Kristian Høgsberg <krh@redhat.com>
99753Date: Wed Aug 8 10:07:03 2007 -0400
99754
99755 Implement damage reporting for AIGLX.
99756
99757commit 600028305cc047e824b18369a026f89c9eb7e437
99758Author: Kristian Høgsberg <krh@redhat.com>
99759Date: Thu May 31 22:55:47 2007 -0400
99760
99761 Move hyperpipe and swapbarrier extension funcs to screen private struct.
99762
99763 This gets rid of a couple more global arrays and gets the two extensions
99764 more in line with the general extension mechanism.
99765
99766commit 781515bb637de4fd79e3c83817cb6ffa8f2d8a4f
99767Author: Kristian Høgsberg <krh@redhat.com>
99768Date: Wed May 30 23:25:03 2007 -0400
99769
99770 Convert GLX module to use screen private indexes like everything else.
99771
99772commit 72a3d68a2f5abcd09fef1a55e976e1a5731d4b02
99773Author: Kristian Høgsberg <krh@redhat.com>
99774Date: Thu May 10 11:16:50 2007 -0400
99775
99776 Update the AIGLX DRI loader to use the new DRI interface.
99777
99778 This lets us drop a bunch of workarounds and use the new DRI extension
99779 mechanism.
99780
99781commit 17ffc34ad5e9a8e2269afef05f209701f1895c01
99782Author: Matthias Hopf <mhopf@suse.de>
99783Date: Fri Oct 12 15:33:31 2007 +0200
99784
99785 Drivers include miscstruct.h which requires pixman.h.
99786
99787commit eaf0e2a21c2cb14e19852e61a4521b3c240253af
99788Author: David Nusinow <dnusinow@debian.org>
99789Date: Thu Oct 11 22:31:24 2007 -0400
99790
99791 Fix another compiler warning
99792
99793commit a5089af726b6a4f833b95a31274743c994277e20
99794Author: David Nusinow <dnusinow@debian.org>
99795Date: Thu Oct 11 22:24:31 2007 -0400
99796
99797 Fix a warning about the control logic in xchomp()
99798
99799commit cdf29ff45a3cb45573c9d0cb8f82e6ee97953fb5
99800Author: David Nusinow <dnusinow@debian.org>
99801Date: Thu Oct 11 22:23:34 2007 -0400
99802
99803 Remove some unused variables
99804
99805commit 3367091f7fa14497aab40e668cad179e244eef81
99806Author: David Nusinow <dnusinow@debian.org>
99807Date: Thu Oct 11 22:21:38 2007 -0400
99808
99809 Separate choosing driver from the file-based implementation
99810
99811 This clears the implementation out of the way to prepare for development
99812 of a symbol-based resolution scheme
99813
99814commit 3aa41bcb8215c037512ddbd68a3f7bcad3b80a1f
99815Author: David Nusinow <dnusinow@debian.org>
99816Date: Thu Oct 11 21:29:13 2007 -0400
99817
99818 Remove obsolete error message define from parser
99819
99820commit e3e12221111886c4063d2da5d70d3830c56d39e2
99821Author: David Nusinow <dnusinow@debian.org>
99822Date: Thu Oct 11 21:27:07 2007 -0400
99823
99824 Don't bother validating the Device section of the conf file
99825
99826 All the previous tests can now be recovered from if not specified.
99827
99828commit 28ef7f59416677be380561709197b04df0479bef
99829Author: David Nusinow <dnusinow@debian.org>
99830Date: Thu Oct 11 20:56:46 2007 -0400
99831
99832 Re-enable validation of the screen section of xorg.conf
99833
99834 This also fixes a problem where the server can't find the device section
99835 when it is specified in the screen section.
99836
99837commit ea2d4dc468dcebe6d38e676469ec51ed1d43490b
99838Author: David Nusinow <dnusinow@debian.org>
99839Date: Thu Oct 11 20:26:02 2007 -0400
99840
99841 When there's no xorg.conf, use the video driver autoloading function
99842
99843commit fc092334ac0a323b80a9602cb8bf60ca9dee3bfa
99844Author: Eric Anholt <eric@anholt.net>
99845Date: Thu Oct 11 16:48:56 2007 -0700
99846
99847 Bug #10304,12784,11603: Add quirks for several physical size issues.
99848
99849 A lot of EDID writers apparently end up stuffing centimeters (like the
99850 maximum image size field) into the detailed timings, instead of millimeters.
99851 Some of them only get it wrong in one direction. Also, add a quirk to let
99852 us mark the largest 75hz mode as preferred, which will often be used for
99853 EDID 1.0 CRTs.
99854
99855commit 6adeba17301a309be2f34cd51eca84a13d5503fd
99856Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99857Date: Thu Oct 11 14:17:17 2007 -0400
99858
99859 dix: Add a new "registry" mechanism for registering string names of things.
99860 Supports protocol requests, events, and errors, and resource names.
99861 Modify XRES extension to use it.
99862
99863commit 45cc03726b49f6ad4afe6c3fb4ad65d1051928b4
99864Author: Matthias Hopf <mhopf@suse.de>
99865Date: Thu Oct 11 18:23:25 2007 +0200
99866
99867 Make mode checking more tolerant like in pre-RandR times.
99868
99869commit 8f23d40068151ad85cde239d07031284f0b2c4dc
99870Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99871Date: Wed Oct 10 19:56:03 2007 -0400
99872
99873 xace: move the property deletion hook inside the DeleteProperty function.
99874
99875commit 473bc6ec4c59e1a962b0b897c449a69aa5064ab0
99876Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99877Date: Wed Oct 10 19:43:12 2007 -0400
99878
99879 xace: remove the special-cased "ignore" functionality from the property code.
99880
99881 There will be no more faking of Success to hide things. XACE does not
99882 provide polyinstantiation.
99883
99884commit 7e9e01a4a34fa45521067d43c5bbff942dd5d51a
99885Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
99886Date: Wed Oct 10 17:40:22 2007 -0400
99887
99888 dix: pass a valid ClientPtr to SetFontPath in all cases.
99889
99890commit 8825d36a38c76ff8faf409c853061be2bd33430d
99891Author: David Nusinow <dnusinow@debian.org>
99892Date: Tue Oct 9 22:49:07 2007 -0400
99893
99894 Allow the user to not have a screen section
99895
99896 If none is present, a default one will be created. This will be attached
99897 to either the first device section in the xorg.conf (allowing you to
99898 specify something like using EXA without having a screen section) or a
99899 default screen section if none is present in the file.
99900
99901commit 8665cf3327b713f334b0483593a924c197f892f2
99902Author: David Nusinow <dnusinow@debian.org>
99903Date: Tue Oct 9 22:45:15 2007 -0400
99904
99905 Don't require that the screen explicitly attach the device section
99906
99907 This will allow the screen to not explicitly have a device section. If
99908 this is the case and there is a device section in the xorg.conf, the first
99909 one will be used. If there is no device section at all, a default one will
99910 be created that loads the automatically determined module.
99911
99912commit 37898b824fdc94735495e3494aa2b9a681d477b9
99913Author: David Nusinow <dnusinow@debian.org>
99914Date: Tue Oct 9 22:11:58 2007 -0400
99915
99916 Use the best guess heuristic as a fallback for autoloading
99917
99918 This prevents breaking the current behavior
99919
99920commit 847fbbf4d25132e6c1f99fcf82fe757dbca94a34
99921Author: David Nusinow <dnusinow@debian.org>
99922Date: Tue Oct 9 22:07:49 2007 -0400
99923
99924 Fix autoloading of drivers for pci-rework
99925
99926commit 6033d8150be3a115b90226eaa42f237bb0cf3369
99927Author: David Nusinow <dnusinow@debian.org>
99928Date: Tue Oct 9 21:17:27 2007 -0400
99929
99930 First pass at improved video driver autoloading
99931
99932 This is what we're currently shipping in Debian. Enables the ability for
99933 drivers to ship a text file listing PCI ID's they support, and have the
99934 server read them on startup when no driver is specified. This works, but
99935 isn't the final solution.
99936
99937commit 81d7b81146224f2b83278f5e21b3f9a36f30bd56
99938Author: David Nusinow <dnusinow@debian.org>
99939Date: Sat Aug 25 15:00:30 2007 -0400
99940
99941 Load intel instead of i810 when autoconfiguring
99942
99943commit fdef7be5c8d5989e0aa453d0a5b86d0a6952e960
99944Author: Alan Coopersmith <alan.coopersmith@sun.com>
99945Date: Tue Oct 9 18:44:04 2007 -0700
99946
99947 Sun bug 6589829: include zoneid of shm segment in access policy for MIT-SHM
99948
99949 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6589829>
99950 <http://www.opensolaris.org/os/community/arc/caselog/2007/500/>
99951
99952commit 49c8285e38e0e71c9ea43df63add08a4f2564ed1
99953Author: Christian Weisgerber <naddy@freebsd.org>
99954Date: Tue Oct 9 15:25:16 2007 -0700
99955
99956 FreeBSD/alpha server build requires libio (missed in modularization).
99957
99958commit 042e16458b751ad0fcfa1ebbdd74d7d48036b22a
99959Author: Peter Hutterer <peter@cs.unisa.edu.au>
99960Date: Tue Oct 9 23:07:51 2007 +0930
99961
99962 dix: change a few inputInfo.keyboard over to PickKeyboard()
99963
99964commit 6aab6a6e2ae0ef5fdedae09cf2cdc86f382e3a8a
99965Author: Dodji Seketeli <dodji@openedhand.com>
99966Date: Mon Oct 8 13:25:38 2007 +0200
99967
99968 EXA: dude, don't validation driver hooks twice !
99969
99970commit d42909b23a9be64a26aa349c1d6dcd8b228bffa7
99971Author: Peter Hutterer <peter@cs.unisa.edu.au>
99972Date: Fri Oct 5 13:04:10 2007 +0930
99973
99974 dix: when disabling a device, make sure all paired devices are re-paired.
99975
99976 We re-pair them with the VCP, not a real device! If we would do otherwise,
99977 somebody may change our keyboard focus and thus get us typing where we don't
99978 want to type.
99979
99980commit ff2351246da30b56a3a8f90654993c41b8eb3921
99981Author: Peter Hutterer <peter@cs.unisa.edu.au>
99982Date: Fri Oct 5 12:19:03 2007 +0930
99983
99984 dix: When pairing, only check for access right if client is set.
99985
99986 If the pairing client is not set, then the pairing is initiated internally
99987 (e.g. when a new keyboard device is configured). In this case we _must_ pair
99988 regardless of who is the pairing client.
99989
99990commit 4611f9568cdadf9c00f4b7ca4c77c8c6f07e94f8
99991Merge: ab11bad 3549a12
99992Author: Dodji Seketeli <dodji@openedhand.com>
99993Date: Thu Oct 4 13:07:50 2007 +0200
99994
99995 updated from 'master'
99996
99997commit ab11bad54707941eb41be62c025b983760ce3900
99998Author: Dodji Seketeli <dodji@openedhand.com>
99999Date: Thu Oct 4 13:05:29 2007 +0200
100000
100001 Xephyr: remove a potential crasher
100002
100003 * hw/kdrive/ephyr/hostx.c:
100004 (hostx_has_dri): be more defensive.
100005
100006commit 06c8977966e76255ce7798f4839ef6e3530264a8
100007Author: Dodji Seketeli <dodji@openedhand.com>
100008Date: Thu Oct 4 13:01:24 2007 +0200
100009
100010 Xephyr: better handling of the -nodri option
100011
100012 * hw/kdrive/ephyr/ephyr.c:
100013 (ephyrInitScreen): try and detect when the host has no
100014 DRI support. In that case, switch to the -nodri behaviour.
100015 When in the -nodri case, make sure not to skip glx visual
100016 initialisation.
100017
100018commit ea94f59e9b8824b1a638c63e06ca16f0efd43869
100019Author: Dodji Seketeli <dodji@openedhand.com>
100020Date: Thu Oct 4 12:54:26 2007 +0200
100021
100022 Xephyr: don't disable glx visual init by default
100023
100024 * hw/kdrive/ephyr/ephyrinit.c:
100025 (ddxProcessArgument): disabling visual init here
100026 is bad because it gets disabled even when we want
100027 to use software GL, leading to Xephyr :1 -nodri
100028 crashing in mesa.
100029
100030commit a7f210e6fcda14eae4de64186904c6c676c758ee
100031Author: Dodji Seketeli <dodji@openedhand.com>
100032Date: Thu Oct 4 12:51:08 2007 +0200
100033
100034 Xephyr: fix a compiler warning
100035
100036commit 5d6eac1251f1e17baa74d0893bb225b775e8ec2e
100037Author: Dodji Seketeli <dodji@openedhand.com>
100038Date: Thu Oct 4 12:47:05 2007 +0200
100039
100040 Xephyr: remove an unused variable
100041
100042commit 8018733c39c283a931df424dcfd11e7aefbcd8cf
100043Author: Dodji Seketeli <dodji@openedhand.com>
100044Date: Thu Oct 4 12:06:19 2007 +0200
100045
100046 Xephyr: better compilation without GLX
100047
100048 * hw/kdrive/ephyr/ephyrinit.c: protect GLX related code with GLXEXT
100049 macro
100050
100051commit 3549a1282365e69e70c7c2546cfa7d25923cce31
100052Author: Michel Dänzer <michel@tungstengraphics.com>
100053Date: Thu Oct 4 11:38:01 2007 +0200
100054
100055 EXA: Disable 1x1 pixmap fill optimization for drivers that handle pixmaps.
100056
100057 This reverts commit 1365aeff5499a051375e43a9fcbf54733ac93929.
100058
100059 It defeated the optimization for drivers that don't provide a CreatePixmap
100060 hook. The optimization makes no sense for drivers that do anyway, so disable
100061 it for them completely.
100062
100063commit 604ebb5a6de372e6a8a96e0ee997db7929126860
100064Merge: 1365aef 4ba76a7
100065Author: Dodji Seketeli <dodji@openedhand.com>
100066Date: Wed Oct 3 16:13:16 2007 +0200
100067
100068 Merge Xephyr-XV/GL stuff into master
100069
100070commit 4ba76a7e2b62d26f43c0e670de571afb75ec92f4
100071Author: Dodji Seketeli <dodji@openedhand.com>
100072Date: Wed Oct 3 13:03:34 2007 +0200
100073
100074 Xephyr: port XV/GL stuff of the new multiscreen architecture
100075
100076 We can now launch GL or XV apps in any of the
100077 Xephyr screens we want.
100078
100079 * hw/kdrive/ephyr/hostx.c,h:
100080 (hostx_get_window):
100081 (hostx_create_window): make these functions be screen
100082 number aware.
100083 * hw/kdrive/ephyr/XF86dri.c : fix some compiler warnings.
100084 * hw/kdrive/ephyr/ephyrdri.c:
100085 (ephyrDRIQueryDirectRenderingCapable),
100086 (ephyrDRIOpenConnection),
100087 (ephyrDRIAuthConnection),
100088 (ephyrDRICloseConnection),
100089 (ephyrDRIGetClientDriverName),
100090 (ephyrDRICreateContext),
100091 (ephyrDRIDestroyContext),
100092 (ephyrDRICreateDrawable),
100093 (ephyrDRIGetDrawableInfo),
100094 (ephyrDRIGetDeviceInfo): in all those functions, don't forward
100095 the screen number we receive - from the client - to the host X.
100096 We (Xephyr) are always targetting the same X display screen, which is
100097 the one Xephyr got launched against. So we enforce that in the code.
100098 * hw/kdrive/ephyr/ephyrdriext.c:
100099 (EphyrMirrorHostVisuals): make this duplicate the visuals of the host X
100100 default screen into a given Xephyr screen. This way we have a chance
100101 to update the visuals of all Xephyr screen to make them mirror those
100102 of the host X.
100103 (many other places): specify screen number where required by the api
100104 change in hostx.h.
100105 * hw/kdrive/ephyr/ephyrglxext.c: specify screen number where required
100106 by the api change in hostx.h
100107 * hw/kdrive/ephyr/ephyrhostglx.c: don't forward the screen number we
100108 receive - from the client - to the host X.
100109 We (Xephyr) are always targetting the same
100110 X display screen, which is
100111 the one Xephyr got launched against. So we enforce that in the code.
100112 * hw/kdrive/ephyr/ephyrhostvideo.c,h: take in account the screen number received
100113 from the client app. This is useful to know on which Xephyr screen we
100114 need to display video stuff.
100115 * hw/kdrive/ephyr/ephyrvideo.c: update this to reflect the API change
100116 in hw/kdrive/ephyr/ephyrhostvideo.h.
100117 (ephyrSetPortAttribute): when parameters are not valid
100118 - they exceed their validity range - send them to the host anyway
100119 and do not return an error to clients.
100120 Some host expose buggy validity range, so rejecting client for that
100121 is too harsh.
100122
100123commit ab88cb8d2e7c1410f9ed2be928b38f176b132e11
100124Author: Peter Hutterer <peter@cs.unisa.edu.au>
100125Date: Wed Oct 3 18:08:33 2007 +0930
100126
100127 Xext: add missing geint.h to Makefile.am (un-breaks make dist)
100128
100129commit 9f2b493e34e93881101f31e631901d3fe56da4f0
100130Author: Peter Hutterer <peter@cs.unisa.edu.au>
100131Date: Thu Sep 27 15:35:41 2007 +0930
100132
100133 xkb: remove some warning comments.
100134
100135 Obsolete with 340911d7243a7f1095d79b5b2dcfa81b145c2474.
100136
100137commit 1eebb03a3190947a8102f2ddc73766cf98d34c84
100138Author: Peter Hutterer <peter@cs.unisa.edu.au>
100139Date: Wed Oct 3 15:18:17 2007 +0930
100140
100141 dix: ignore passive grab if the client already has a grab on the device.
100142
100143 In some cases a button press may activate a passive core grab. If the client
100144 owning the passive grab already has a core grab on another device, don't
100145 actually activate it. Otherwise the client gets two simultaneous passive
100146 core grabs, and may never ungrab the device again (only if the other grab uses
100147 GrabModeSync).
100148
100149 Reproducable: fire up gnome-session, open up gnome-terminal. Click with the
100150 ClientPointer onto the window decoration, then click with another pointer onto
100151 an application icon in the panel. Drag the icon out, release the button and
100152 voila - you just lost your second mouse.
100153
100154commit 0b485067823620b5dbd9ef2b3e13bd35ad5a4410
100155Author: Peter Hutterer <peter@cs.unisa.edu.au>
100156Date: Wed Oct 3 14:22:55 2007 +0930
100157
100158 dix: send NotifyGrab/NotifyUngrab focus events regardless of semaphore state.
100159
100160 This is just papering over a problem. The whole focus system needs to be
100161 revised.
100162
100163commit 05106ac9839102c0e4a3ce5d9d83d19abf129f8a
100164Author: Peter Hutterer <peter@cs.unisa.edu.au>
100165Date: Wed Oct 3 11:33:10 2007 +0930
100166
100167 dix: change Enter/Leave semaphore handling to accommodate for NotifyGrab.
100168
100169 This is a half-assed attempt at getting rid of some enter-leave problems. When
100170 a grab is activated, the events didn't get sent before, leading to interesting
100171 results. This commit papers over it but doesn't actually fix it properly. The
100172 whole enter/leave (focusin/out) structure needs to be ripped out and changed
100173 for multiple devices.
100174
100175commit 1365aeff5499a051375e43a9fcbf54733ac93929
100176Author: Dave Airlie <airlied@redhat.com>
100177Date: Wed Oct 3 12:00:16 2007 +1000
100178
100179 exa: direct access to the pixmap sys ptr is bad if the pixmap isn't mapped
100180
100181commit 566dd3b7d789ba60d0adf33b3f729cfb02ff33cd
100182Author: Dave Airlie <airlied@redhat.com>
100183Date: Wed Oct 3 11:59:52 2007 +1000
100184
100185 exa: add new flags to denote driver handles all pixmap migration/hiding
100186
100187commit 1df1fee82d3565f6d8cfb91ce25a81c23a10a4b5
100188Author: Dave Airlie <airlied@redhat.com>
100189Date: Wed Oct 3 11:56:04 2007 +1000
100190
100191 exa: make sure we set fb_ptr to NULL
100192
100193commit a358b87f45ce75e5d013fc904a07dfe394f74387
100194Author: Eric Anholt <eric@anholt.net>
100195Date: Tue Oct 2 13:13:51 2007 -0700
100196
100197 Just link against libcrypto instead of relying on openssl pkg-config.
100198
100199 Also fix incorrect library inclusion by kdrive which broke the build.
100200
100201commit 439edc768eea17667846ce573c843b8377e0dfb4
100202Merge: c8ccf46 93ae6fe
100203Author: Eric Anholt <eric@anholt.net>
100204Date: Tue Oct 2 12:14:04 2007 -0700
100205
100206 Merge branch 'glyph-pixmaps'
100207
100208 Conflicts:
100209
100210 configure.ac
100211 exa/exa_render.c
100212
100213commit 59cebcd2e9302d15a52588ecafbbc2d2c5ae3a6c
100214Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
100215Date: Tue Oct 2 13:39:25 2007 -0400
100216
100217 xace: add creation hook for new input devices.
100218
100219 Unfortunately, more information is needed to properly label the device.
100220 This will come from the configuration file, the hotplug messages, etc.
100221 It will either have to be passed into this function,
100222 or this hook moved down into the callers.
100223
100224commit b77d272d7555c1e0f176ee74b8717030a6d6c7b0
100225Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
100226Date: Tue Oct 2 13:21:53 2007 -0400
100227
100228 xace: add hooks + new access codes: XTEST extension
100229
100230commit c8ccf469dcb2d7609b23faa8b4999c436263051e
100231Author: Keith Packard <keithp@koto.keithp.com>
100232Date: Tue Oct 2 08:59:40 2007 -0700
100233
100234 Do not build hw/xfree86/os-support/bus/ix86Pci.c on Linux.
100235
100236 Linux has custom PCI bus support functions, so this generic code is
100237 unnecessary.
100238
100239commit dcb4db1bf5b6fcfaed4e7162eb6c7a3a509e14bd
100240Author: Dodji Seketeli <dodji@openedhand.com>
100241Date: Wed Sep 19 15:46:09 2007 +0200
100242
100243 Xephyr: glx fixlet
100244
100245 * hw/kdrive/ephyr/ephyrhostglx.c:
100246 (ephyrHostGLXQueryVersion): properly query
100247 the host for its gxl version.
100248
100249commit 6a435b00003fb00930299a0e0810c93afc23a72e
100250Author: Dodji Seketeli <dodji@openedhand.com>
100251Date: Wed Sep 19 15:35:51 2007 +0200
100252
100253 Xephyr: check presence of extensions in host X
100254
100255 * hw/kdrive/ephyr/hostx.c,h:
100256 (hostx_has_xshape),
100257 (hostx_has_glx),
100258 (hostx_has_dri): added these new entry points
100259 * hw/kdrive/ephyr/ephyrdriext.c:
100260 (ephyrDRIExtensionInit):
100261 check presence of DRI and XShape extensions before
100262 trying to use them.
100263 * hw/kdrive/ephyr/ephyrglxext.c:
100264 (ephyrHijackGLXExtension):
100265 check presence of glx extension before we use it.
100266
100267commit 2b217fc055300d4c709dd27c4e8ec8166183993b
100268Author: Dodji Seketeli <dodji@openedhand.com>
100269Date: Wed Sep 19 15:00:43 2007 +0200
100270
100271 Xephyr: cleanup XF86dri.c a bit
100272
100273commit f731b9da7dbafad50dd5a1e8da28ce86ffab19b7
100274Author: Dodji Seketeli <dodji@openedhand.com>
100275Date: Tue Sep 18 18:41:18 2007 +0200
100276
100277 Xephyr: add -nodri and -noxv switches.
100278
100279 * hw/kdrive/ephyr/ephyr.c,ephyrinit.c::
100280 add -nodri and -noxv to disable either DRI or XV.
100281 When -nodri, mesa GL (non acceleretated) works.
100282
100283commit 50674391388b3cfe987782a3ad81b4f240f3db67
100284Author: Dodji Seketeli <dodji@openedhand.com>
100285Date: Tue Sep 18 18:37:49 2007 +0200
100286
100287 Kdrive: use fbcmap_mi.c instead of fbcmap.c
100288
100289 * hw/kdrive/src/Makefile.am: use fb/fbcmap_mi.c
100290 and not fb/fbcmap.c. This allows kdrive to take advantage of
100291 extensions redefining the entry points of micmap.c stuff.
100292 For instance it allows Xephyr to have a working GL, which is not
100293 possible otherwise, because GL redefines mInitVisualsProc
100294 to initialise its visuals.
100295
100296commit a5bd536869f646e4024befa84f6e677b990310bc
100297Author: Dodji Seketeli <dodji@openedhand.com>
100298Date: Tue Sep 18 18:28:25 2007 +0200
100299
100300 fbcmap_mi.c: add a missing api
100301
100302 * fb/fbcmap_mi.c:
100303 (fbSetVisualTypesAndMasks): added this entry point that was missing.
100304 This is useful so that servers using this entry point can use fbcmap_mi.c,
100305 and be obliged to stick to fbcmap.c. Note that fbcmap.c does implement this
100306 entry point. Up to now, kdrive based server could not use fbcmap_mi.c because this
100307 entry point was missing. The will allow Xephyr to properly use GL.
100308
100309commit 6307ee77091041af2d34e93bc164352d4438a808
100310Author: Dodji Seketeli <dodji@openedhand.com>
100311Date: Tue Sep 18 18:22:44 2007 +0200
100312
100313 GLX: allow skipping of visual init wrapping
100314
100315 * GL/glx/glxvisuals.c: added boolean to disable
100316 calling init_visuals(). This gives a chance to Xephyr
100317 to take over visuals manipulation and avoid a crash at
100318 server shutdown in __glXMesaScreenDestroy(), due to the fact
100319 that mesa might sees more visual than what it has actually created in
100320 init_visuals(). It might see more visuals because Xephyr can augment
100321 the number of visuals, dynamically.
100322 * os/utils.c: the boolean is actually defined here.
100323
100324commit ffe9ce6a843443606ad9367eca1645bf5e0635e5
100325Author: Dodji Seketeli <dodji@openedhand.com>
100326Date: Mon Sep 17 22:13:51 2007 +0200
100327
100328 Xephyr: fix a clipping issue xephyr-dri
100329
100330 * hw/kdrive/ephyr/ephyrdri.c:
100331 (ephyrDRIGetDrawableInfo): force the back clipping rects
100332 to equal the front clipping rects.
100333 * hw/kdrive/ephyr/ephyrdriext.c:
100334 (ProcXF86DRIGetDrawableInfo): properly overclip the clipping rects we
100335 got from the client. This bug fixes a clipping rect that was too
100336 small in height, basically. Also fix a possible mem corruption.
100337 * hw/kdrive/ephyr/hostx.c:
100338 (hostx_set_window_geometry): remove a useless XSync
100339
100340commit 79782726617d4b5a4f9b376f21936d035fc870e1
100341Author: Dodji Seketeli <dodji@openedhand.com>
100342Date: Mon Sep 17 18:15:03 2007 +0200
100343
100344 Xephyr: properly clip GL drawables in Xephyr
100345
100346commit 3816d64e50a59e0ef276ddd6cb250cab98a49392
100347Author: Dodji Seketeli <dodji@seketeli.org>
100348Date: Sun Sep 9 15:20:51 2007 +0200
100349
100350 Xephyr: oops, forgot to add ephyrdriext.h
100351
100352commit cf58781eee7d91a3bbca701a2c4523a36f8e0abb
100353Author: Dodji Seketeli <dodji@openedhand.com>
100354Date: Sat Sep 8 00:33:59 2007 +0200
100355
100356 Xephyr: make accelerated glxgears work in Xephyr
100357
100358 * hw/kdrive/ephyr/ephyr.c:
100359 (ephyrInitialize): cleanup ephyrDRI extension init.
100360 remove functions that belongs in ephyrdriext.c .
100361 * hw/kdrive/ephyr/ephyrdri.c:
100362 (ephyrDRICreateDrawable): create the drawable on the host X peer
100363 window, not on the host xephyr main window.
100364 (ephyrDRIGetDrawableInfo): get drawable info of the host X peer
100365 window.
100366 * hw/kdrive/ephyr/ephyrdriext.c: make ephyr DRI extention wrap
100367 a bunch of screen ops so that it can update the host X peer
100368 window whenever DRI bound drawable are moved in Xephyr.
100369 Also code the building blocks of the management of the
100370 host X window peer.
100371 * hw/kdrive/ephyr/hostx.c,h:
100372 (hostx_create_window): added this new entry point
100373 (hostx_destroy_window): ditto
100374 ()hostx_set_window_geometry): ditto
100375
100376commit 6d1e44d3d53b451d466d43197be589d0ecc4b9f6
100377Author: Dodji Seketeli <dodji@openedhand.com>
100378Date: Tue Sep 4 18:28:16 2007 +0200
100379
100380 Xephyr: fix a host X hang.
100381
100382 * hw/kdrive/ephyr/ephyrdri.c:
100383 (ephyrDRIGetDrawableInfo): quickly hook
100384 this into getting the drawable info from the host
100385 X server. For the time being, this only gets the drawable info
100386 of the Xephyr main window in the host. It should really get
100387 the info of a the peer drawable in the host X. So there should be a
100388 peer drawable to begin with.
100389 * hw/kdrive/ephyr/ephyrdriext.c:
100390 (ProcXF86DRIGetDrawableInfo): some cleanups. Properly get the
100391 the drawable info otherwise there is a host X hang.
100392 * hw/kdrive/ephyr/ephyrhostglx.c: do not
100393 (ephyrHostGLXQueryVersion): do not use C bindings of the glx protocol
100394 calls. Some of those actually access DRI context directly, resulting
100395 in the context having three clients. Instead all XF86DRI proto
100396 fowarding request should be coded by hand and only forward the
100397 protocol requests
100398
100399commit a39b57d2f784f14e3198c8506f5eb865bacc2440
100400Author: Dodji Seketeli <dodji@openedhand.com>
100401Date: Fri Aug 31 16:53:18 2007 +0200
100402
100403 Xephyr: add more logging to GLX forwarding
100404
100405 * hw/kdrive/ephyr/ephyrhostglx.c: added more logging.
100406
100407commit 9e192d2118246ad715e23cf5626a038274eb1eaa
100408Author: Dodji Seketeli <dodji@openedhand.com>
100409Date: Fri Aug 31 14:43:39 2007 +0200
100410
100411 Xephyr: better error handling in GLX forwarding
100412
100413 * hw/kdrive/ephyr/ephyrglxext.c:
100414 fixed various logging functions
100415 (ephyrGLXGetStringReal): make sure all the string is sent to clients
100416 including the ending zero.
100417 * hw/kdrive/ephyr/ephyrhostglx.c:
100418 (ephyrHostGLXGetStringFromServer): better error handling.
100419 (ephyrHostGLXSendClientInfo): ditto.
100420 (ephyrHostGLXMakeCurrent): ditto
100421
100422commit 4dd4be99df38b70d43b02d9cb19e09e0ed94462b
100423Author: Dodji Seketeli <dodji@seketeli.org>
100424Date: Thu Aug 30 23:54:49 2007 +0200
100425
100426 Xephyr: Make glxinfo work on the ATI R200 free driver.
100427
100428 * hw/kdrive/ephyr/ephyr.c:
100429 (EphyrDuplicateVisual): when duplicating the
100430 visual, copy the color component masks and the class
100431 from the hostX
100432 (EphyrMirrorHostVisuals): don't mix blue and green mask.
100433 * hw/kdrive/ephyr/ephyrdri.c: add more logs.
100434 (ephyrDRICreateDrawable): actually implement this.
100435 for the moment it creates a DRI drawable for the hostX window,
100436 no matter what drawable this call was issued for.
100437 (ephyrDRIGetDrawableInfo): actually implemented this.
100438 for the moment the drawable info queried for its attrs is the
100439 Xephyr main main window.
100440 * hw/kdrive/ephyr/ephyrdriext.c:
100441 (ProcXF86DRIGetDrawableInfo): properly hook this dispatch
100442 function to the ephyrDRIGetDrawableInfo() function.
100443 * hw/kdrive/ephyr/ephyrglxext.c: add a bunch of GLX implementation hooks
100444 here. Hijack some of the xserver GLX hooks with them. Still need to
100445 properly support byteswapped clients though.
100446 * hw/kdrive/ephyr/ephyrhostglx.c,h: actually implemented the protocol
100447 level forwarding functions used by the GLX entr points in
100448 ephyrglxext.c. Here as well, there are a bunch of them, but we are
100449 far from having implemented all the GLX calls.
100450 * hw/kdrive/ephyr/hostx.c,h:
100451 (hostx_get_window_attributes): added this new entry point
100452 (hostx_allocate_resource_id_peer): added this to keep track of
100453 resource IDs peers: one member of the peer is in Xephyr, the other
100454 is in host X.
100455 (hostx_get_resource_id_peer): ditto.
100456
100457commit 8c78df0ea36abc7d46c4f59670289f4280f75507
100458Author: Dodji Seketeli <dodji@seketeli.org>
100459Date: Thu Aug 30 22:31:47 2007 +0200
100460
100461 Xserver: fix potential crash
100462
100463 * dix/extension.c:
100464 (AddExtensionAlias): don't crash when failed to add
100465 extension alias.
100466
100467commit 024abe825cd702334266f6b8da289907bbd05970
100468Author: Dodji Seketeli <dodji@seketeli.org>
100469Date: Tue Aug 28 15:59:45 2007 +0200
100470
100471 Xephyr: add logging in the DRI extension
100472
100473commit 1dd589410c6c5e8b62b4d15ec681b01452ec68c6
100474Author: Dodji Seketeli <dodji@seketeli.org>
100475Date: Tue Aug 28 15:58:55 2007 +0200
100476
100477 mirror the visuals of the host X at startup.
100478
100479 * hw/kdrive/ephyr/ephyr.c: make Xephyr mirror
100480 the visuals of the host X upon startup. This
100481 is important for GLX client apps.
100482 * hw/kdrive/ephyr/hostx.c,h: add a hostx_get_visuals_info()
100483 to get the visuals of the host X.
100484
100485commit 26da625055a3876c93a7a005520657b971748c4e
100486Author: Dodji Seketeli <dodji@seketeli.org>
100487Date: Tue Aug 28 15:55:05 2007 +0200
100488
100489 proxy the glXGetFBConfigSGIX call
100490
100491 * hw/kdrive/ephyr/ephyrglxext.c:
100492 (ephyrGLXGetFBConfigsSGIX): proxy the GLXGetFBConfigsSGIX call.
100493 It is a vendor extension to get the visual configs as a list of
100494 name/value pairs.
100495 (ephyrHijackGLXExtension): hijack the VendorPriv_dispatch_info
100496 dispatch table to register our implementation of GLXGetFBConfigsSGIX
100497 (ephyrGLXGetFBConfigsSGIXReal): added this where the real
100498 implementation of GLXGetFBConfigsSGIX is. It support bytes swapping.
100499 (ephyrGLXGetFBConfigsSGIX,ephyrGLXGetFBConfigsSGIXSwap): these are
100500 the dispatch entry points. They just call
100501 ephyrGLXGetFBConfigsSGIXReal.
100502 * hw/kdrive/ephyr/ephyrhostglx.c,h: reorganize the proxies to get
100503 visual params from the host so that they clearly support the different
100504 methods of doing so.
100505
100506commit 5af73f98c46dfe5640185ca90feb20c39c6e06fa
100507Author: Dodji Seketeli <dodji@seketeli.org>
100508Date: Fri Aug 24 16:30:50 2007 +0200
100509
100510 EPHYR: Add an experiment proto proxy extension.
100511
100512 * hw/kdrive/ephyr/Makefile.am: add the proxy extension to
100513 ephyr. The proxy extension is an experimental extension that
100514 forwards protocol packets targeted at a given extension to the
100515 host X.
100516 * hw/kdrive/ephyr/ephyr.c: init proxy ext.
100517 * hw/kdrive/ephyr/ephyrhostproxy.c,h: added this new file as part of the
100518 proxy extension.
100519 * hw/kdrive/ephyr/ephyrproxyext.c,h: ditto
100520 * hw/kdrive/ephyr/hostx.c: add the hostx_get_get_extension_info() entry
100521 point.
100522
100523commit c06fa924b4781a35b86e4a78d95ff3e9d95b02d1
100524Author: Dodji Seketeli <dodji@seketeli.org>
100525Date: Thu Aug 23 11:53:02 2007 +0200
100526
100527 XEPHYR: more GLX/DRI proxying work.
100528
100529 * hw/kdrive/ephyr/XF86dri.c: re format this correctly.
100530 Make function decls honour the Ansi-C standard.
100531 * hw/kdrive/ephyr/ephyr.c: protect glx/dri related
100532 extension initialisation with the XEPHYR_DRI
100533 macro. Initialize the GLX ext hijacking
100534 at startup.
100535 * hw/kdrive/ephyr/ephyrdri.c: add more logging to ease debugging
100536 * hw/kdrive/ephyr/ephyrdriext.c: ditto. reformat.
100537 * hw/kdrive/ephyr/ephyrglxext.c,h: add this extension to
100538 proxy GLX requests to the host X. started to proxy those nedded to
100539 make glxinfo work with fglrx. Not yet finished.
100540 * hw/kdrive/ephyr/ephyrhostglx.c,h: put here the actual
100541 Xlib code used to hit the host X server because Xlib stuff cannot be
100542 mixed with xserver internal code, otherwise compilation erros due to
100543 type clashes happen. So no Xlib type should be exported by the
100544 entrypoints defined here.
100545
100546commit 0b8545144975bf7ed43c2564d01c697144eb1244
100547Author: Dodji Seketeli <dodji@seketeli.org>
100548Date: Thu Aug 16 17:30:13 2007 +0200
100549
100550 Implement ReputImage and StopVideo
100551
100552 * hw/kdrive/ephyr/ephyrhostvideo.c/h:
100553 (ephyrHostXVStopVideo): add this entry point.
100554 * hw/kdrive/ephyr/ephyrvideo.c:
100555 Basically add ReputImage and StopVideo implementations.
100556 Now, when other windows obscur the video window, the reclipping
100557 seems to be well handled using StopVideo and ReputImage.
100558 To do this, I was obliged to save the frame in PutImage, so
100559 that I could resend it un ReputImage.
100560
100561commit 810dc55866d1c2343512354646c7ab309ea1fad2
100562Author: Dodji Seketeli <dodji@seketeli.org>
100563Date: Thu Aug 16 17:11:22 2007 +0200
100564
100565 Ephyr-Xv: add a new line to a log
100566
100567 * hw/kdrive/ephyr/ephyrvideo.c:
100568 (ephyrQueryImageAttributes): add newline to log.
100569
100570commit 47e6dff89e48249828e828502e98951eee7f85bc
100571Author: Dodji Seketeli <dodji@seketeli.org>
100572Date: Thu Aug 16 17:07:36 2007 +0200
100573
100574 Xephyr-Xv: fix a crash when host X support several ports
100575
100576 * hw/kdrive/ephyr/ephyrvideo.c:
100577 (ephyrXVPrivQueryHostAdaptors): properly set
100578 port private luke. This fixes a crash when
100579 the host Xv supports multiple ports.
100580 Make sure number of ports cannot be zero.
100581
100582commit a38ad562a6f50e9d76a37917b936035215ea460e
100583Author: Dodji Seketeli <dodji@seketeli.org>
100584Date: Wed Aug 15 19:40:10 2007 +0200
100585
100586 make xephyr talk DRI protocol with hostX
100587
100588 * configure.ac,include/dix-config.h.in: define the XEPHYR_DRI macro.
100589 define it when --enable-xephyr and --enable-dri are both turned on.
100590 * hw/kdrive/ephyr/XF86dri.c: copy this from mesa source to enable
100591 Xephyr to talk DRI protocol the host X. In mesa, this is used by libGL.so to
100592 talk DRI protocol with the server.
100593 * hw/kdrive/ephyr/ephyr.c: finally initialise the DRI extension
100594 in the ephyrInitScreen() function.
100595 * hw/kdrive/ephyr/ephyrdri.c,ephyrdriext.c: safeguard the compilation
100596 using the XEPHYR_DRI macro.
100597
100598commit e4239a48075c77e6a8d2d5cb21f58dd67687482f
100599Author: Dodji Seketeli <dodji@openedhand.com>
100600Date: Tue Aug 14 22:48:41 2007 +0200
100601
100602 Initial dri forwarding big bricks.
100603
100604 * hw/kdrive/ephyr/ephyrdriext.c: added this to implement a DRI extension
100605 into Xephyr. Normally the DRI extension is only present in the
100606 xfree86 server, but I have ported it to Xephyr. The extension calls
100607 functions that declared/defined in ephyrdri.h ephyrdri.c that
100608 forwards the DRI calls to the host X. It does not work yet, as this
100609 entry is just to put the big bricks in place.
100610 * hw/kdrive/ephyr/ephyrdri.c,h: declaration & definition of the
100611 DRI client API that would hit the hostX server.
100612 * hw/kdrive/ephyr/GL/internal/dri_interface.h: added this, otherwise
100613 inclusion of /usr/include/X11/dri/xf86dri.h won't compile
100614
100615commit e01d3dd98d1b596e75d25f94dd89c7d41223011d
100616Author: Dodji Seketeli <dodji@openedhand.com>
100617Date: Thu Aug 9 10:55:10 2007 +0200
100618
100619 Support clipping region in PutImage.
100620
100621 * hw/kdrive/ephyr/ephyrhostvideo.c,h:
100622 (ephyrHostXVPutImage): make this support clipping region.
100623 The clipping region is propagated to host using XSetClipRectangles.
100624 This changes the API of ephyrHostXVPutImage.
100625 * hw/kdrive/ephyr/ephyrvideo.c:
100626 (ephyrPutImage): propagate the clipping region to the new
100627 ephyrHostXVPutImage() entry point.
100628
100629commit 39d3895469f07304d72800d8dcef6c7732f13d5f
100630Author: Dodji Seketeli <dodji@openedhand.com>
100631Date: Wed Aug 8 14:27:32 2007 +0200
100632
100633 make EphyrXVPriv be a singleton
100634
100635 * hw/kdrive/ephyr/ephyrvideo.c:
100636 (ephyrInitVideo) make the EphyrXVPriv object be a
100637 singleton instance, otherwise a new object is created at each
100638 generation.
100639
100640commit 4ed083095a13ec92bb4c8e705f26500f8312c138
100641Author: Dodji Seketeli <dodji@openedhand.com>
100642Date: Tue Aug 7 14:16:13 2007 +0200
100643
100644 add [Get/Put]Video and [Get/Put]Still support
100645
100646 * hw/kdrive/ephyr/ephyrhostvideo.c,h:
100647 (ephyrHostXVAdaptorHasXXX): fix these.
100648 (ephyrHostXVAdaptorHasGetVideo): added this
100649 (ephyrHostXVAdaptorHasGetStill): ditto
100650 (ephyrHostXVPutVideo): added this
100651 (ephyrHostXVGetVideo): ditto
100652 (ephyrHostXVPutStill): ditto
100653 (ephyrHostXVGetStill): ditto
100654 * hw/kdrive/ephyr/ephyrvideo.c:
100655 (ephyrPutVideo): implement this
100656 (ephyrGetVideo): ditto
100657 (ephyrPutStill): ditto
100658 (ephyrGetStill): ditto
100659 (ephyrXVPrivSetAdaptorsHooks): advertise GetVideo and GetStill
100660 when the host X supports it.
100661
100662commit 6f7961bc218169aaa0d10cfad0952adb7fac2f00
100663Author: Dodji Seketeli <dodji@openedhand.com>
100664Date: Mon Aug 6 18:38:54 2007 +0200
100665
100666 advertise only the hooks implemented by host X
100667
100668 * hw/kdrive/ephyr/ephyrhostvideo.c,h:
100669 (ephyrHostXVAdaptorHasPutVideo): detect if
100670 host X has the PutVideo call.
100671 (ephyrHostXVAdaptorHasPutStill): detect if
100672 host X has the PutStill call
100673 (ephyrHostXVAdaptorHasPutImage): detect if
100674 host X has the PutImage call
100675 * hw/kdrive/ephyr/ephyrvideo.c:
100676 (ephyrXVPrivQueryHostAdaptors): make sure to create
100677 atoms for attribute names otherwise subsequent
100678 calls to get/set attribute from clients won't work.
100679 (ephyrXVPrivSetAdaptorsHooks): don't hardwire advertising
100680 of the PutImage call. Instead, advertise the calls advertised
100681 by the host.
100682
100683commit 1de89239bdfa68ded2ef25fbb7d08369f527f7a9
100684Author: Dodji Seketeli <dodji@openedhand.com>
100685Date: Sat Aug 4 01:13:38 2007 +0200
100686
100687 don't crash when hostX reports an error.
100688
100689 * hw/kdrive/ephyr/ephyrhostvideo.c,h:
100690 (ephyrHostXVLogXErrorEvent): add this to
100691 log X error events. Heavily copied from libx11
100692 (ephyrHostXVErrorHandler): new x error handler that
100693 logs the error but does not exits.
100694 (ephyrHostXVInit): add this to be called at the beginning
100695 of xvideo lifetime. It sets an xerror handler that does not
100696 exit.
100697 * hw/kdrive/ephyr/ephyrvideo.c:
100698 (ephyrXVPrivIsAttrValueValid): this validates an attribute
100699 value.
100700 (ephyrSetPortAttribute): before setting an attribute,
100701 validate the new value so that we don't send a buggy
100702 request to host X.
100703 * hw/kdrive/ephyr/*.c: fix case in ephyrvideo code.
100704 * hw/kdrive/ephyr/ephyr.c: fix a typo
100705
100706commit 207714b60d18e94bbe2d755a431dbb29fe5677f7
100707Author: Dodji Seketeli <dodji@openedhand.com>
100708Date: Fri Aug 3 01:10:39 2007 +0200
100709
100710 first implementation of putimage
100711
100712 * hw/kdrive/ephyr/ephyrhostvideo.c,h:
100713 (EphyrHostXVPutImage): first implementation. does not
100714 support clipping regions yet.
100715 * hw/kdrive/ephyr/ephyrvideo.c:
100716 (DoSimpleClip): clip using a clipping box. Does not
100717 support regions yet.
100718 (EphyrPutImage): first implementation.
100719 Uses a simple clipping rectangle, no region yet.
100720 * hw/kdrive/ephyr/hostx.c:
100721 (hostx_get_window): added this to get the main
100722 window of the host x.
100723
100724commit aa478e09e25a1cb19b808e7e52b2ee155e34f03b
100725Author: Dodji Seketeli <dodji@openedhand.com>
100726Date: Thu Aug 2 12:49:36 2007 +0200
100727
100728 make properties setting/query code work and hopefully complete.
100729
100730 * hw/kdrive/ephyr/ephyrhostvideo.c,h:
100731 (EphyrHostXVQueryImageAttributes): add this call. It calls
100732 XvQueryBestSize xserver entry point. It uses the protocol
100733 level machinery because Xvlib does not expose that entry point
100734 as a C function.
100735 (EphyrHostXVQueryBestSize): added this wrapper around XvQueryBestSize().
100736 (EphyrHostGetAtom, EphyrHostGetAtomName): added this to get
100737 an atom or atom name from the host server
100738 * hw/kdrive/ephyr/ephyrvideo.c:
100739 (EphyrSetPortAttribute): convert the atom into an host server
100740 server atom before attacking the host server with it, because in
100741 in its current form, the input atom is only valid in xephyr.
100742 This fix makes this call work.
100743 (EphyrGetPortAttribute): ditto.
100744 (EphyrQueryBestSize): implement this.
100745 (EphyrQueryImageAttributes): implement this.
100746
100747commit 24c837c0be6f85a917582156ab61d6c9c69301f9
100748Author: Dodji Seketeli <dodji@openedhand.com>
100749Date: Tue Jul 31 23:18:52 2007 +0200
100750
100751 implement setportattribute and getportattribute calls
100752
100753commit b8cd313be467fd14b8e5c9ba68e1a2029c1e888b
100754Author: Dodji Seketeli <dodji@seketeli.org>
100755Date: Mon Jul 23 12:43:43 2007 +0200
100756
100757 make xvinfo work
100758
100759 * hw/kdrive/ephyr/ephyrhostvideo.c:
100760 (EphyrHostXVAdaptorGetVideoFormats): properly get visual class instead of
100761 returning the visual id.
100762 (EphyrHostXVQueryEncodings): properly copy the fields because simple casting does
100763 truncate some fields.
100764 (EphyrHostAttributesDelete): XFree the whole array instead of trying to free invidial members.
100765 * hw/kdrive/ephyr/ephyrvideo.c:
100766 (ephyrInitVideo): fix a typo
100767 (EphyrXVPrivQueryHostAdaptors): set XvWindowMask mask to adaptors type.
100768 use host adaptor name. Don't forget to set nImages field.
100769 (EphyrXVPrivRegisterAdaptors): report an error when KdXVScreenInit() fails.
100770
100771commit aa2fae7b75f741e57bc4a9b754b5ea7518d78f47
100772Author: Dodji Seketeli <dodji@seketeli.org>
100773Date: Sat Jul 21 23:56:13 2007 +0200
100774
100775 misc fixes.
100776
100777 *(EphyrHostXVQueryAdaptors): return the queried adaptors list
100778 * (EphyrHostXVQueryPortAttributes): return port attributes number
100779
100780commit 2bf7f3c223fd32cd4922f98eadda2e3f06d9e119
100781Author: Dodji Seketeli <dodji@seketeli.org>
100782Date: Sat Jul 21 17:55:12 2007 +0200
100783
100784 link and init xv
100785
100786commit 66b28532e533ff5688705c5286425e68d6403cee
100787Author: Dodji Seketeli <dodji@seketeli.org>
100788Date: Sat Jul 21 12:08:39 2007 +0200
100789
100790 initial commit of xv support work
100791
100792commit 1e8f8b36a56a884092ee01a0bb3f40436e30fa42
100793Author: Dodji Seketeli <dodji@seketeli.org>
100794Date: Sat Jul 21 18:22:20 2007 +0200
100795
100796 works with no adaptors
100797
100798 Debug the early code to make work when the host has no video adaptors
100799
100800commit 50ff2377cac50ae25218be5b30612fb8bf6046a6
100801Author: Dodji Seketeli <dodji@seketeli.org>
100802Date: Sat Jul 21 18:20:59 2007 +0200
100803
100804 add -verbosity <level> option to command line
100805
100806commit 8426eb2433f964ea19848f24dc99eb4588d10d1a
100807Author: Dodji Seketeli <dodji@seketeli.org>
100808Date: Sat Jul 21 17:55:12 2007 +0200
100809
100810 link and init xv
100811
100812commit 50a64c84e1556ef7f8efcc7f8655e0c5e023ef4d
100813Author: Dodji Seketeli <dodji@seketeli.org>
100814Date: Sat Jul 21 12:08:39 2007 +0200
100815
100816 initial commit of xv support work
100817
100818commit 95fadbd4022ec99f42ba78ec8a18a064903e8a7f
100819Author: Dodji Seketeli <dodji@seketeli.org>
100820Date: Mon Jul 16 17:37:21 2007 +0200
100821
100822 Add basic Exa driver entry point validation.
100823
100824 In exaDriverInit(), quickly check if the programmer forgot to set some
100825 mandadory driver hooks, in that case display a meaningful error message.
100826
100827commit e5e6514ffa0fd132e0cc1b15b94119e6e8755f43
100828Author: Andrew Christan <andrew.christian@nokia.com>
100829Date: Tue Oct 2 13:25:51 2007 +0200
100830
100831 Xephyr: add "multiscreen" suport
100832
100833 * This patch adds multiscreen support to Xephyr. For instance,
100834 the command line : "Xephyr :4 -ac -screen 320x240 -screen 640x480"
100835 will launch with two "screens" - namely two main windows.
100836 The first main window represents a screen that has the number :4.0, with
100837 a geometry of 320x240 pixels, and the second one represents a screen
100838 that has the number :4.1 with a geometry of 640x480.
100839 The command line: "DISPLAY=:4.1 xclock" will launch the xclock program
100840 on the second screen, for intance.
100841
100842 * this patch was edited by Dodji Seketeli <dodji@openedhand.com> for:
100843 - better style compliance with the rest of the Xephyr code
100844 - make sure Xephyr could be launched with no -screen option. By
100845 default that creates a default screen of 640x480 pixel like before
100846 - display full titles on the windows - with insctructions to grab
100847 keyboard and mouse - like before.
100848
100849commit 81692b628f41cb64329f3cccc0503fb216a2b8e3
100850Author: Dodji Seketeli <dodji@openedhand.com>
100851Date: Tue Oct 2 13:11:28 2007 +0200
100852
100853 Fix #12650: "windows not receiving mouse events in multiple screens context"
100854
100855 * dix/events.c, include/dix.h:
100856 (UpdateSpriteForScreen): added this to update the mouse sprite context
100857 when we switch from a pScreen to another.
100858 * mi/mipointer.c:
100859 (miPointerWarpCursor): as we are switching to a new pScreen,
100860 don't forget to update the mouse sprite context.
100861
100862commit f965a5f3454a95ddffb2faf9b291feff46305adf
100863Author: Peter Hutterer <peter@cs.unisa.edu.au>
100864Date: Tue Oct 2 16:20:27 2007 +0930
100865
100866 dix: Ignore devices when deleting passive core grabs.
100867
100868commit dc90d500085dd74e90073be008fc514bd97b160a
100869Author: Dave Airlie <airlied@redhat.com>
100870Date: Tue Oct 2 14:05:07 2007 +1000
100871
100872 exa: increase minor version number for pixmap allocation hooks
100873
100874commit f15af2ae60bb8503c336bc3cba0560bd314a34bc
100875Author: Kristian Høgsberg <krh@redhat.com>
100876Date: Fri Aug 3 16:33:33 2007 +1000
100877
100878 exa: only setup offscreen allocator if driver doesn't provide CreatePixmap
100879
100880commit ffb58f4fa8d86e87f831430b8627f27d85f971a9
100881Author: Dave Airlie <airlied@redhat.com>
100882Date: Tue Oct 2 14:03:39 2007 +1000
100883
100884 exa: add hooks for drivers to take over pixmap allocation
100885
100886 This adds hooks for the driver to access Create/DestroyPixmap and ModifyPixmapHe
100887 ader.
100888
100889 It allocates a 0 sized pixmap using fb and calls the driver routine to do
100890 work of allocating the actual memory.
100891
100892 ModifyPixmapHeader is mainly required for hooking the screen pixmap which
100893 isn't create by normal methods
100894
100895commit 2b075e97ec913ebe19290181e36bb169410b820b
100896Author: Peter Hutterer <peter@cs.unisa.edu.au>
100897Date: Tue Oct 2 13:28:40 2007 +0930
100898
100899 mi: switch core keyboard on XI events, not only on core events.
100900
100901 We only get core events through the EQ in exceptional cases, so make sure we
100902 actually swap the core keymap for XI events as well. Gives us back the ability
100903 to have multiple keyboard layouts simultaneously.
100904
100905commit 31555af0005a0bc4d7ef785214696ac84681e29f
100906Author: Dave Airlie <airlied@redhat.com>
100907Date: Tue Oct 2 13:32:57 2007 +1000
100908
100909 exa: add a pixmap private pointer for drivers to retrieve.
100910
100911commit 45ec6cd1fb242363c91ad8af1fd4a27a7f02621a
100912Author: Peter Hutterer <peter@cs.unisa.edu.au>
100913Date: Tue Oct 2 13:19:07 2007 +0930
100914
100915 xfree86: fix DGA to use the correct devices.
100916
100917 DGAStealXXXEvent modified to take in device argument.
100918
100919 The evdev driver only sends one valuator when only one axis changed. We need
100920 to check for DGA either way (xf86PostMotionEventP), otherwise we lose purely
100921 horizontal/vertical movements.
100922
100923 Note that DGA does not do XI events.
100924
100925commit 122ae65ed90195c584a770027b4d14ef65f72492
100926Author: Peter Hutterer <peter@cs.unisa.edu.au>
100927Date: Tue Oct 2 10:50:59 2007 +0930
100928
100929 xfree86: xf86DGA: some cleanup to use the correct devices instead of VCP/VCK.
100930
100931commit 54ce7436b2579bb20728eab7c8d460b531c378af
100932Author: Peter Hutterer <peter@cs.unisa.edu.au>
100933Date: Tue Oct 2 10:49:46 2007 +0930
100934
100935 xfree86: update all pointers when calling xf86SwitchMode
100936
100937 Center the frame around the first pointer found and then update all pointers
100938 on the same screen to move to the edges (if necessary).
100939
100940 Note: xf86WarpCursor needs to be modified, is using deprecated
100941 miPointerWarpCursor and will kill the server when called with
100942 inputInfo.pointer.
100943
100944commit 113011221a4fcbd70bc592930b35f20d55809f1e
100945Author: Peter Hutterer <peter@cs.unisa.edu.au>
100946Date: Mon Oct 1 21:42:01 2007 +0930
100947
100948 Xi, xfree86: Remove leftovers of the XI wrapper code.
100949
100950commit 50551ec693f40b91652fe4814e9fe2e1f9ab6517
100951Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
100952Date: Fri Sep 28 15:04:33 2007 -0400
100953
100954 xace: remove obsoleted DRAWABLE_ACCESS hook.
100955
100956commit 8b548657204000e18c7a38706a0071ae2f93159f
100957Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
100958Date: Fri Sep 28 13:34:18 2007 -0400
100959
100960 xace: add hooks + new access codes: XKB extension.
100961
100962 Removes "LookupKeyboardDevice" and "LookupPointerDevice" in favor of
100963 inputInfo.keyboard and inputInfo.pointer, respectively; all use cases
100964 are non-XI compliant anyway.
100965
100966commit 5c03d131815cfe2f78792277ab8352e69e830196
100967Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
100968Date: Fri Sep 28 08:02:00 2007 -0400
100969
100970 xace: add new hooks + access controls: XInput extension.
100971
100972 Introduces new dix API to lookup a device, dixLookupDevice(), which
100973 replaces LookupDeviceIntRec and LookupDevice.
100974
100975commit a511c445debbd13e8c48146ecd2d7c97e793f788
100976Author: Peter Hutterer <peter@cs.unisa.edu.au>
100977Date: Fri Sep 28 18:46:41 2007 +0930
100978
100979 Remove generation of core events, long live XI!
100980
100981 Let the drivers only generate XI events and put those into the event queue.
100982 When processing events, generate core events as needed. This fixes a number of
100983 problems with XKB and the DIX in general.
100984
100985 The previous approach was to put core events and XI events as separate events
100986 into the event queue. When being processed, the server had no knowledge of
100987 them coming from the same device state change. Anything that would then change
100988 the state of the device accordingly was in danger of changing it twice,
100989 leading to some funny (i.e. not funny at all) results.
100990
100991 Emulating core events while processing XI events fixes this, there is only one
100992 path that actually changes the device state now. Although we have to be
100993 careful when replaying events from synced devices, otherwise we may lose
100994 events.
100995
100996 Note: XI has precedence over core for passive grabs, but core events are
100997 delivered to the client first.
100998
100999 This removes the wrapping added in 340911d7243a7f1095d79b5b2dcfa81b145c2474
101000
101001commit be466d8df808c4e4067a7963617bc3f506768f2d
101002Author: Peter Hutterer <peter@cs.unisa.edu.au>
101003Date: Thu Sep 27 22:23:05 2007 +0930
101004
101005 dix: GetPairedKeyboard() always returns a valid keyboard (VCK if necessary).
101006
101007 We need it unconditionally in a few places, and the rest checked for NULL and
101008 then set it to VCK anyway. So, fixing up all callers to appreciate the defined
101009 return value.
101010
101011commit 53434edc3d306137d019d95189ecdf0dbff75205
101012Author: Peter Hutterer <peter@cs.unisa.edu.au>
101013Date: Thu Sep 27 22:14:09 2007 +0930
101014
101015 dix: take paired keyboard instead of VCK (CoreProcessPointerEvents)
101016
101017commit 2d93e69690d2c5d4a89a795ede6423796528e5df
101018Author: Alan Coopersmith <alan.coopersmith@sun.com>
101019Date: Thu Sep 27 16:47:06 2007 -0700
101020
101021 Rework local client id finding code to be more uniform
101022
101023commit c7ead3a68e5839cb92129e35b21f55007fba8445
101024Author: Alan Coopersmith <alan.coopersmith@sun.com>
101025Date: Thu Sep 27 16:13:28 2007 -0700
101026
101027 xorgcfg needs $(DIX_CFLAGS) for pixman-1 include path
101028
101029commit 62a9fb4cda988a896909a5620a68c51e46d0e403
101030Author: Alan Coopersmith <alan.coopersmith@sun.com>
101031Date: Thu Sep 27 16:12:29 2007 -0700
101032
101033 Fix PCI rework build on Solaris (copy what BSD does)
101034
101035commit cfe549d1ba28396c44af94133d8c1d323e3c7086
101036Author: Tilman Sauerbeck <tilman@code-monkey.de>
101037Date: Thu Sep 27 22:36:52 2007 +0200
101038
101039 EXA: Added back the maxPitchPixels initialization code.
101040
101041 This doesn't add real value yet, but it will be useful once I add code
101042 that splits large render operations into smaller parts if necessary.
101043
101044commit c11a27ef85674f1e77ffa7f083646b848e1dc0b8
101045Author: Alan Hourihane <alanh@tungstengraphics.com>
101046Date: Thu Sep 27 14:54:40 2007 +0100
101047
101048 Add the FB_ACCESS_WRAPPER checks
101049
101050commit e0bb33b3d2e4f54cf20853cde9a1664a7dbd0dc7
101051Author: Alan Hourihane <alanh@tungstengraphics.com>
101052Date: Thu Sep 27 14:50:28 2007 +0100
101053
101054 Fix bug #12286
101055
101056commit 6d5c1e0d896666bcb2b3c1de7bfa424f140be364
101057Author: Michel Dänzer <michel@tungstengraphics.com>
101058Date: Thu Sep 27 13:04:41 2007 +0200
101059
101060 EXA: Remove bogus pitch checks.
101061
101062 exaCreatePixmap should handle all cases correctly.
101063
101064commit 598698678b07cb3a9406a9ee98bd3186366949e7
101065Author: Michel Dänzer <michel@tungstengraphics.com>
101066Date: Thu Sep 27 13:07:09 2007 +0200
101067
101068 EXA: Punt for all fallbacks in exaFillRegion*.
101069
101070 Now that PaintWindow is gone, all callers already handle fallbacks.
101071
101072commit 006f6525057970a74382132237b2131286ad147c
101073Author: Michel Dänzer <michel@tungstengraphics.com>
101074Date: Thu Sep 27 13:08:41 2007 +0200
101075
101076 EXA: Make sure tile offsets passed to drivers are never negative.
101077
101078 Thanks to Björn Steinbrink for pointing out the problem on IRC.
101079
101080commit da7d9aa1fb60e13a59c9f842fed7aefc5b97c195
101081Author: Michel Dänzer <michel@tungstengraphics.com>
101082Date: Thu Sep 27 13:08:40 2007 +0200
101083
101084 EXA: Tile offscreen pixmap coordinate offsets are always 0.
101085
101086commit d6f4764bf5f3a601a0034ded039857e8ea5563b2
101087Author: Michel Dänzer <michel@tungstengraphics.com>
101088Date: Thu Sep 27 13:08:40 2007 +0200
101089
101090 EXA: Remove some clearly bogus code from exaCopyNtoN.
101091
101092 Not sure what I was thinking when I wrote this... it would cause the box
101093 coordinates to be off for exaCopyNtoNTwoDir or fallbacks.
101094
101095 Thanks to Tilman Sauerbeck for pointing out the problem on IRC and testing the
101096 fix.
101097
101098commit c7d6d1f589d729fa689d22d82fe30afbc6e1cacb
101099Author: Michel Dänzer <michel@tungstengraphics.com>
101100Date: Thu Sep 27 13:08:40 2007 +0200
101101
101102 EXA: Punt on fallback case not handled correctly in exaFillRegionTiled.
101103
101104 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=12520 .
101105
101106commit 1d938a80fd4fa58d1791c146b6b5c2dfe148dce7
101107Author: Alan Coopersmith <alan.coopersmith@sun.com>
101108Date: Wed Sep 26 19:42:46 2007 -0700
101109
101110 Remove unused pciAddrHostToBus functions from ix86Pci.c
101111
101112 Matches linuxPci.c changes made in 8279444a54f38c5e2e5d4c2d936a10d74990e0be
101113 Fixes compiler errors:
101114 "ix86Pci.c", line 194: too many struct/union initializers
101115 "ix86Pci.c", line 204: too many struct/union initializers
101116 "ix86Pci.c", line 214: too many struct/union initializers
101117
101118commit de06a47d4ad9c845b098438d9492a5f42483a2ad
101119Author: Peter Hutterer <peter@cs.unisa.edu.au>
101120Date: Thu Sep 27 11:45:03 2007 +0930
101121
101122 dix: GetMaximumEventsNum() needs to return 3 + MAX_VALUATOR_EVENTS.
101123
101124 Raw events can be generated for pointer events, so 2 + MVE isn't enough.
101125
101126commit 32d0440c7f6e604807cb14dd32349df6f22c903b
101127Author: Peter Hutterer <peter@cs.unisa.edu.au>
101128Date: Thu Sep 27 11:44:03 2007 +0930
101129
101130 xkb: xkbHandleActions: let wrapping take care of event delivery.
101131
101132 This is hopefully better than hardcodey calling CoreProcessPointerEvent.
101133
101134commit 27612748e0ec20f3a23839f0a12e39f598dd722c
101135Merge: b614614 3b114f2
101136Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101137Date: Wed Sep 26 07:47:29 2007 -0400
101138
101139 Merge branch 'master' into XACE-SELINUX
101140
101141 Conflicts:
101142
101143 dix/devices.c
101144
101145commit 3b114f2603fc2adeec7f5f8f20fe4870afb8dff1
101146Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101147Date: Tue Sep 25 07:55:17 2007 -0400
101148
101149 Input: Properly swap DevicePresenceNotify events.
101150
101151commit 31a7994ac7365562ef1d00e0a7b25f967a961a4e
101152Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101153Date: Mon Sep 24 13:33:01 2007 -0400
101154
101155 Input: Return errors to the dispatcher instead of sending them ourself.
101156
101157 Also fixed two "unused variable: stuff" warnings.
101158
101159commit 56ffc381d32687242dd094395fcf2216339bab2a
101160Author: Tilman Sauerbeck <tilman@code-monkey.de>
101161Date: Wed Sep 26 16:47:54 2007 +0200
101162
101163 EXA: Removed duplicated exaGetDrawablePixmap() calls.
101164
101165commit aa0325db7e8ab11d9418cae14f11c488f443ccaa
101166Author: Dodji Seketeli <dodji@openedhand.com>
101167Date: Wed Sep 26 15:26:42 2007 +0200
101168
101169 Xephyr: fix compiler warnings
101170
101171 * hw/kdrive/ephyr/ephyr_draw.c:
101172 (ephyrDownloadFromScreen),
101173 (ephyrUploadToScreen): fix compiler warnings
101174
101175commit 8f9bf927e1beecf9b9ec8877131ec12c765e4d84
101176Author: Peter Hutterer <peter@cs.unisa.edu.au>
101177Date: Wed Sep 26 18:04:59 2007 +0930
101178
101179 xkb: Unwrap properly in ProcessPointerEvent.
101180
101181 Instead of hardcoding CoreProcessPointerEvent, actually try to unwrap properly
101182 and then call the unwrapped processInputProc. Seems to be a better idea,
101183 especially since it makes stuff actually work...
101184
101185commit 3342b5ad47be25c6838321c0aafc28c329c308b5
101186Author: Peter Hutterer <peter@cs.unisa.edu.au>
101187Date: Wed Sep 26 18:03:21 2007 +0930
101188
101189 Xi: fix the wrapper code for processInputProc wrapping.
101190
101191 Followup to [1].
101192 If a core grab causes the device to freeze, it overwrites the processInputProc
101193 of the device. [1] would then overwrite this while unwrapping, the device
101194 does not thaw anymore.
101195 Changing this to only re-wrap if the processInputProc hasn't been changed
101196 during the event handling.
101197
101198 [1] 340911d7243a7f1095d79b5b2dcfa81b145c2474
101199
101200commit e2cb8515661b1f7826981931d82dee6e05529f04
101201Author: Peter Hutterer <peter@cs.unisa.edu.au>
101202Date: Wed Sep 26 18:01:17 2007 +0930
101203
101204 xfree86: pass in the XKB event processing proc instead of the XI one.
101205
101206 We need to pass ProcessPointerEvent or ProcessKeyboardEvent to
101207 XkbSetExtension, otherwise we lose the xkb layer.
101208
101209commit 27bc1a8fef2bfd3d62fb44f7c7eb0d463ed08632
101210Author: Peter Hutterer <peter@cs.unisa.edu.au>
101211Date: Wed Sep 26 15:24:41 2007 +0930
101212
101213 xkb: XkbFilterEvents: Remove unused variable compiler warning.
101214
101215commit bfc89c035542a10594f5f0cbde1c7e28b7d024a7
101216Author: Peter Hutterer <peter@cs.unisa.edu.au>
101217Date: Wed Sep 26 15:23:37 2007 +0930
101218
101219 xkb: unify ErrorFs. Prefix all with [xkb].
101220
101221 Output for XkbUseMsg intentionally skipped.
101222
101223commit 8b508f5d6bd0d3995294d5ff300a856754442999
101224Author: Peter Hutterer <peter@cs.unisa.edu.au>
101225Date: Wed Sep 26 15:19:53 2007 +0930
101226
101227 dix: unify ErrorFs. prepend all with [dix].
101228
101229commit 394f3c1dbee7270a1d930846d49278424c3072d4
101230Author: Peter Hutterer <peter@cs.unisa.edu.au>
101231Date: Wed Sep 26 15:19:37 2007 +0930
101232
101233 dix: GetPointerEvents: remove unused variable compiler warning.
101234
101235commit 55fd50273ea67eb99a0c8a830349851931298387
101236Author: Peter Hutterer <peter@cs.unisa.edu.au>
101237Date: Wed Sep 26 15:15:34 2007 +0930
101238
101239 Xi: unify ErrorFs. Prefix all with [Xi].
101240
101241commit 340911d7243a7f1095d79b5b2dcfa81b145c2474
101242Author: Peter Hutterer <peter@cs.unisa.edu.au>
101243Date: Wed Sep 26 15:12:49 2007 +0930
101244
101245 Wrap core event handling through ProcessOtherEvents.
101246
101247 When processing events from the EQ, _always_ call the processInputProc of the
101248 matching device. For XI devices, this proc is wrapped in three layers.
101249 Core event handling is wrapped by XI event handling, which is wrapped by XKB.
101250 A core event now passes through XKB -> XI -> DIX.
101251
101252 This gets rid of a sync'd grab problem: with the previous code, core events
101253 did disappear during a sync'd device grab on account of mieqProcessInputEvents
101254 calling the processInputProc of the VCP/VCK instead of the actual device. This
101255 lead to the event being processed as normal instead of being enqueued for
101256 later replaying.
101257
101258commit b2a4883bd89d406713d4f808e72721ecc1456d67
101259Author: Peter Hutterer <peter@cs.unisa.edu.au>
101260Date: Wed Sep 26 09:20:11 2007 +0930
101261
101262 dix: add comments to PlayReleasedEvents()
101263
101264 Also stick another warning in, more changes needed for XGE if panoramix is
101265 active.
101266
101267commit 275dc77c2972147407584323b866b8acc1654ea0
101268Author: Peter Hutterer <peter@cs.unisa.edu.au>
101269Date: Tue Sep 25 21:40:08 2007 +0930
101270
101271 dix: add comments to ComputeFreezes() and syncEvents struct.
101272
101273commit ae8b4f7dcf1291a2f5a0d0159f3e6089eea0578a
101274Author: Tilman Sauerbeck <tilman@code-monkey.de>
101275Date: Mon Sep 24 22:06:52 2007 +0200
101276
101277 EXA: Added pitch limitations.
101278
101279 Drivers can now specify the max pitches that the accelerator supports.
101280
101281commit b61461425eb15fcff2a58330d74fe5a5a1f226fc
101282Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101283Date: Tue Sep 25 09:56:00 2007 -0400
101284
101285 xace: add hooks + new access codes: XV extension.
101286
101287 May need to revisit this extension in the future, depending on observed use.
101288
101289commit 9bd04055a2175ec16756d3bf73ae03b5e163a28a
101290Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101291Date: Tue Sep 25 09:33:51 2007 -0400
101292
101293 xace: change prototype of VALIDATE_DRAWABLE_AND_GC macro to allow access
101294 mode to be passed to dixLookupDrawable.
101295
101296commit 988f446fe00ac2b92c484a4440613dd82191eadd
101297Author: Eric Anholt <eric@anholt.net>
101298Date: Mon Sep 24 20:23:35 2007 -0700
101299
101300 [EDID] Ignore reserved bits in deciding monitor vs detailed timing descriptor.
101301
101302 Even though they're defined to zero by the spec, we've seen an EDID block
101303 where the (empty) ASCII strings were stuffed in a byte early, leading to the
101304 descriptor being considered a detailed timing instead.
101305
101306commit 5d9f6cf51becdeb1d2be9bdeec4bb5d0c4dd0e8b
101307Author: Peter Hutterer <peter@cs.unisa.edu.au>
101308Date: Thu Sep 13 23:33:07 2007 +0930
101309
101310 dix: cosmetic change. Remove "pointer" variable, might as well use pDev.
101311
101312commit c4db4d1fcf7fa2ba802a208beee6b3adc83921f5
101313Author: Peter Hutterer <peter@cs.unisa.edu.au>
101314Date: Thu Sep 13 23:27:07 2007 +0930
101315
101316 include: remove register keywords.
101317
101318commit 1c25c46f3d5f98be8dbf36684aad4302b220e163
101319Author: Peter Hutterer <peter@cs.unisa.edu.au>
101320Date: Thu Sep 13 23:22:27 2007 +0930
101321
101322 dix: fix compiler warnings about unused variables.
101323
101324commit 68e0c4988e359e3c9da933946bc703cf8530bdbc
101325Author: Peter Hutterer <peter@cs.unisa.edu.au>
101326Date: Wed Sep 12 20:48:48 2007 +0930
101327
101328 dix: fix "possible use of uninitialized variable" warning.
101329
101330 The device passed through to UnrealizeCursor isn't used anyway, so setting it
101331 to NULL is enough.
101332
101333commit b484451eab261a919fe94c0c0f56877f7571fc9e
101334Author: Peter Hutterer <peter@cs.unisa.edu.au>
101335Date: Wed Sep 12 18:34:51 2007 +0930
101336
101337 dix: don't try to access "time" field for GenericEvents in PlayReleasedEvent.
101338
101339 GenericEvents can't be parsed to keyButtonPointer, and there's no guarantee
101340 that it has a time field anyway. PlayReleasedEvent needs to store the millis
101341 when we know it (core events, XI event) and just re-use them for GenericEvents.
101342
101343 Yes, this is a hack. But it looks like the time has zero significance anyway.
101344
101345commit 1c38eac442a3ab9e8469e944f382251957996c5a
101346Author: Peter Hutterer <peter@cs.unisa.edu.au>
101347Date: Wed Sep 12 18:09:57 2007 +0930
101348
101349 dix: fix comment for DeliverDeviceEvents.
101350
101351 DDE is not always called, grabbed or focused devices go through
101352 Deliver{Grabbed|Focus}Event first.
101353
101354 Thanks to Eamon Walsh for spotting it.
101355
101356commit 06188ce90d2a8146140773f9367f23404b464282
101357Author: Peter Hutterer <peter@cs.unisa.edu.au>
101358Date: Wed Sep 12 18:00:03 2007 +0930
101359
101360 dix: Make EnqueueEvent aware of GenericEvents.
101361
101362 GenericEvents can be > sizeof(xEvents), so we need to make sure we're
101363 allocating and copying enough memory for the event.
101364
101365commit 8840829ab93c4eb62eb58753c015da5307133fe5
101366Author: Peter Hutterer <peter@cs.unisa.edu.au>
101367Date: Wed Sep 12 17:40:11 2007 +0930
101368
101369 dix: don't compress motion events from different devices (EventEnqueue)
101370
101371commit 27ad5d74c20f01516a1bff73be283f8982fcf0fe
101372Author: Daniel Stone <daniel@fooishbar.org>
101373Date: Sun Sep 23 17:17:03 2007 +0300
101374
101375 Input: Generate XKB mapping changes for all core-sending devices (bug #12523)
101376
101377 When we change the mapping on a core device, make sure we propagate this
101378 through to XKB for all extended devices as well.
101379
101380commit 0e800ca4651a947ccef239e6fe7bf64aab92257c
101381Author: Daniel Stone <daniel@fooishbar.org>
101382Date: Sun Sep 23 12:43:31 2007 +0300
101383
101384 GetKeyboardEvents: Reject out-of-range keycodes (bug #12528)
101385
101386 We can only deal with keycodes between 8 and 255, so make sure that we never
101387 accept anything out of this range.
101388
101389commit e1860f241b7322c54c109cf5727df45b54b73916
101390Author: Alex Deucher <alex@botch2.(none)>
101391Date: Sat Sep 22 17:54:51 2007 -0400
101392
101393 Document xf86_crtc_clip_video_helper better.
101394
101395commit 19d30c1c771c0f5c9dfc0f4cfb084dd2e528d992
101396Author: Alex Deucher <alex@botch2.(none)>
101397Date: Sat Sep 22 17:51:27 2007 -0400
101398
101399 Revert "Fix possible crash if Xv window is outside of either crtc"
101400
101401 This reverts commit 92355314103d2193ca20786915eaf608922a36dc.
101402
101403 This changes ABI. False means there was a memory failure of some kind,
101404 not that the video shouldn't be displayed
101405
101406commit 92355314103d2193ca20786915eaf608922a36dc
101407Author: Michel Dänzer <daenzer@debian.org>
101408Date: Sat Sep 22 14:25:02 2007 -0400
101409
101410 Fix possible crash if Xv window is outside of either crtc
101411
101412commit edd88170836ec985ab1a59179de69d50644b8dff
101413Author: Tilman Sauerbeck <tilman@code-monkey.de>
101414Date: Sat Sep 22 13:02:08 2007 +0200
101415
101416 EXA: Check ABI compatibility earlier.
101417
101418 If the driver isn't compatible to the server, all bets are off anyway wrt
101419 the contents of the fields that we're validating, which can lead to bogus
101420 error messages.
101421
101422commit abe0a51f3f790f8c055289465e130177c4b647cc
101423Author: Ben Byer <bbyer@bbyer.apple.com>
101424Date: Fri Sep 21 17:07:36 2007 -0700
101425
101426 So, like, checking return codes of system calls (signal, etc) is good.
101427 Also, only restore an old signal handler if one was actually set
101428 (prevents the server from dying on OS X).
101429
101430commit eb82b19aa71333b46e927516cc228f25d3e05e4d
101431Author: Ben Byer <bbyer@bbyer.apple.com>
101432Date: Thu Sep 20 19:55:48 2007 -0700
101433
101434 XDarwin: build fix for X11.app
101435
101436commit 5b08932bfbb7e4612733fffd2acc9682c6856d90
101437Author: Ben Byer <bbyer@bbyer.apple.com>
101438Date: Thu Sep 20 19:27:31 2007 -0700
101439
101440 XDarwin: pulling in Indirect GLX fixes from downstream
101441
101442commit 742d5d6adc4d4a1b6fceeb3443d7931e107462f7
101443Author: Ben Byer <bbyer@bbyer.apple.com>
101444Date: Thu Sep 20 18:03:51 2007 -0700
101445
101446 XDarwin: Fixes to quartzKeyboard.c to prevent a crash using certain non-US keyboards
101447
101448commit 374bd88d1025835fe36ca0ff6620eb1d3b0a06b0
101449Author: Ben Byer <bbyer@bbyer.apple.com>
101450Date: Thu Sep 20 16:50:49 2007 -0700
101451
101452 XDarwin: Merging misc changes from downstream.
101453
101454 The only functional changes in this patch are a removal of use of
101455 Xtrans internals -- replaced by xcb, which doesn't seem to be used
101456 elsewhere in the server? Pity.
101457 Also, a fix to make all X11 windows pop to the front of the display
101458 when the X11.app icon is clicked -- currently takes two clicks,
101459 not sure why.
101460
101461commit 3a965fdadccea7beff09a28c9c0ef4b4975eae38
101462Author: Aaron Plattner <aplattner@nvidia.com>
101463Date: Thu Sep 20 16:22:24 2007 -0700
101464
101465 Don't segfault on shutdown if we never managed to connect to dbus.
101466
101467commit 50fa8722d35c12e5f0322cebe25faf99c39d4f50
101468Author: Aaron Plattner <aplattner@nvidia.com>
101469Date: Thu Sep 20 14:00:33 2007 -0700
101470
101471 Set noCompositeExtension to TRUE when failing to initialize the extension (e.g. when Xinerama is enabled).
101472
101473commit a247886b082cea93fa8f8980616a9c388ba70111
101474Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101475Date: Thu Sep 20 13:06:38 2007 -0400
101476
101477 xace: add hooks + new access codes: XF86-Bigfont extension
101478
101479commit f6532a81eec5f096e27285687964b77c17987f72
101480Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101481Date: Thu Sep 20 12:17:17 2007 -0400
101482
101483 xace: add hooks + new access codes: APPGROUP extension
101484
101485commit 82f7195a628cc7ec94abc0cfe5bae2be8af443bc
101486Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101487Date: Thu Sep 20 09:17:09 2007 -0400
101488
101489 xace: modifications to ChangeWindowAttributes special case: separate
101490 Receive and SetAttr. Refer to 963e69b8efc39369915e7f0c6f370ac0d5d2b60f
101491
101492commit 661b1328cf992d8855552677a94d60de1d8ce942
101493Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101494Date: Thu Sep 20 08:41:26 2007 -0400
101495
101496 xace: add hooks + new access codes: SYNC extension
101497
101498 May need to revisit this extension in the future, depending on observed use.
101499
101500commit 90bacdef723e1e49c72775144916750758d3568c
101501Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101502Date: Thu Sep 20 06:53:51 2007 -0400
101503
101504 xace: add hooks + new access codes: MIT-SHM extension
101505
101506commit 12a18cc8903fac53c3c77b23dd8093187594a4f3
101507Author: David Nolden <david.nolden.git@art-master.de>
101508Date: Fri Sep 14 22:42:19 2007 +0200
101509
101510 compilation-fix in debug mode
101511
101512commit e93cff52fed9074aa007c2e6ec6b578f69aef3cb
101513Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101514Date: Wed Sep 19 14:48:20 2007 -0400
101515
101516 xace: add hooks + new access codes: DOUBLE-BUFFER extension
101517
101518commit 082c0f7fb34458ebb303cf875d1d75686eca25e6
101519Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101520Date: Wed Sep 19 13:59:35 2007 -0400
101521
101522 devPrivates rework: move devPrivates field in drawable structure types
101523 to just below the DrawableRec. Wish there were a better way to do this
101524 but it has to be in the same place for all drawable types.
101525
101526commit 5b36b64192517e2470766ce7ff1d4dc04c936fad
101527Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101528Date: Wed Sep 19 11:11:54 2007 -0400
101529
101530 xace: add missing argument to hook call.
101531
101532commit 963e69b8efc39369915e7f0c6f370ac0d5d2b60f
101533Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101534Date: Wed Sep 19 11:11:41 2007 -0400
101535
101536 xace: add special-case for just setting the event mask on a window,
101537 this should only check "receive" permission, not "setattr" permission.
101538
101539commit c140d20f4ade973496dcbbd06453402ef0c0e85c
101540Author: Ben Byer <bbyer@bbyer.local>
101541Date: Wed Sep 19 05:58:22 2007 -0700
101542
101543 XDarwin: mass change from using xorg-config.h to dix-config.h cuz we're not using the X.org ddx anymore
101544
101545commit bcb5e3e6276ce9ae7a4dae7434cf2247764da078
101546Author: Ben Byer <bbyer@bbyer.local>
101547Date: Wed Sep 19 04:46:10 2007 -0700
101548
101549 XDARWIN: More fixes to Makefile.am
101550
101551 At least on my system (10.5 with the latest and greatest modules),
101552 Xquartz now builds out of the box. It doesn't quite work yet, but
101553 hey -- you have to start somewhere. ;)
101554
101555commit 97c150b61bbe436453b05d3c07cd2173870aac40
101556Merge: 57907e0 547ad21
101557Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101558Date: Wed Sep 19 07:25:55 2007 -0400
101559
101560 Merge branch 'master' into XACE-SELINUX
101561
101562 Conflicts:
101563
101564 afb/afbpntwin.c
101565 afb/afbscrinit.c
101566 afb/afbwindow.c
101567 cfb/cfb.h
101568 cfb/cfballpriv.c
101569 cfb/cfbscrinit.c
101570 cfb/cfbwindow.c
101571 configure.ac
101572 fb/wfbrename.h
101573 hw/xfree86/xf4bpp/ppcIO.c
101574 hw/xfree86/xf4bpp/ppcPntWin.c
101575 hw/xfree86/xf4bpp/ppcWindow.c
101576 hw/xfree86/xf8_32bpp/cfbscrinit.c
101577 mfb/mfb.h
101578 mfb/mfbpntwin.c
101579 mfb/mfbscrinit.c
101580 mfb/mfbwindow.c
101581 mi/miexpose.c
101582
101583 Note: conflicts caused by devPrivates rework vs. paintwindow changes.
101584
101585commit 5e209b21f3d6b3268fa7dab1e8df892d8352cc08
101586Author: Ben Byer <bbyer@bbyer.local>
101587Date: Wed Sep 19 03:44:46 2007 -0700
101588
101589 XNEST: removed obsolete hack to build under Darwin
101590
101591commit 378c7ebef444cdc9ae62ebf05c7111088a0c4bc1
101592Author: Ben Byer <bbyer@bbyer.local>
101593Date: Wed Sep 19 03:43:40 2007 -0700
101594
101595 XVFB: Removed obsolete hack to build on Darwin
101596
101597commit 7813392d1c9a38d01cfff17732278bb7798eee5d
101598Author: Ben Byer <bbyer@bbyer.local>
101599Date: Wed Sep 19 01:52:58 2007 -0700
101600
101601 XDarwin: changes to Makefile.am to use new conditionals
101602
101603commit 170fc77d9e73151f5c1c1f0f04598b3497db4284
101604Author: Dodji Seketeli <dodji@openedhand.com>
101605Date: Wed Sep 19 09:54:09 2007 +0200
101606
101607 xserver: don't force DTRACE detection by default
101608
101609 this fixes a breakage caused by 7a4ec34e256bf36b041c011a083916ad75a1d8bc.
101610 When running a non DTRACE aware system that is not darwin*, DTRACE was getting
101611 required. Now it is not anymore.
101612
101613commit 9b4f5157179a0d20756c16ea1f5130b171f72bb1
101614Author: Ben Byer <bbyer@bbyer.apple.com>
101615Date: Tue Sep 18 21:40:32 2007 -0700
101616
101617 XDarwin: We build many many different targets, let's break them out into autoconf conditionals
101618
101619commit 7a4ec34e256bf36b041c011a083916ad75a1d8bc
101620Author: Ben Byer <bbyer@bbyer.apple.com>
101621Date: Tue Sep 18 20:37:09 2007 -0700
101622
101623 XDarwin: Disable dtrace support on Darwin 9 (since it doesn't work ... yet ...)
101624
101625commit 3c19ec47b434d4ca84db58363cc053cc0b6aa413
101626Author: Ben Byer <bbyer@bbyer.apple.com>
101627Date: Tue Sep 18 20:15:44 2007 -0700
101628
101629 XDarwin: changed XF86BigFont support default from yes to auto so XDarwin can disable it
101630
101631commit 547ad2125ece93bbe01f6d09a3baf176ebd16bb3
101632Author: Michel Dänzer <michel@tungstengraphics.com>
101633Date: Mon Sep 17 20:33:56 2007 +0200
101634
101635 EXA: Make sure driver hooks get correct offscreen offsets from exaCopyDirty.
101636
101637 This should ensure the driver UploadTo/DownloadFromScreen hooks can always
101638 work as intended.
101639
101640commit 56cc24ffb21f7fd41f9ea9e8f969aa85021b9f53
101641Author: Alan Hourihane <alanh@tungstengraphics.com>
101642Date: Mon Sep 17 20:33:56 2007 +0200
101643
101644 EXA: Don't crash in ExaCheckPolyArc.
101645
101646 See https://bugs.freedesktop.org/show_bug.cgi?id=12286 .
101647
101648commit 7ac89060e02c7a803018afd580720f8326561fd8
101649Author: Michel Dänzer <michel@tungstengraphics.com>
101650Date: Mon Sep 17 20:33:55 2007 +0200
101651
101652 EXA: Don't crash in exaGetImage when swapped out.
101653
101654commit 3876c6c80534a6f7412ec806a2ba9ada22c5e505
101655Author: Tilman Sauerbeck <tilman@code-monkey.de>
101656Date: Mon Sep 17 18:47:45 2007 +0200
101657
101658 EXA: Fixed compiler warnings.
101659
101660commit d5738ff2e0f93df4729c075ce31a1041d580e50e
101661Author: Tilman Sauerbeck <tilman@code-monkey.de>
101662Date: Mon Sep 17 16:59:03 2007 +0200
101663
101664 EXA: Added missing exaPrepare/FinishAccess calls in ExaCheckPushPixels.
101665
101666commit a0dafd95e1e13a2f1b77ab9f82fd365a7be19de5
101667Author: Tilman Sauerbeck <tilman@code-monkey.de>
101668Date: Mon Sep 17 16:45:20 2007 +0200
101669
101670 Removed hw/xwin/winpntwin.c from Makefile.am.
101671
101672 winpntwin.c was removed in e4d11e58ce349dfe6af2f73ff341317f9b39684c.
101673
101674commit 7bd65577018a574970b767b67967b65fcd5c2bf0
101675Author: Tilman Sauerbeck <tilman@code-monkey.de>
101676Date: Sat Sep 15 14:01:57 2007 +0200
101677
101678 Initialize output->pendingProperties.
101679
101680commit b9f7aeb20015290a8acf938a11bf9272bf3527cf
101681Merge: e4d11e5 06d27f8
101682Author: Eric Anholt <eric@anholt.net>
101683Date: Thu Sep 13 00:15:45 2007 +0000
101684
101685 Merge remote branch 'origin/master' into paint-window
101686
101687 Conflicts:
101688
101689 mi/miexpose.c
101690
101691commit e4d11e58ce349dfe6af2f73ff341317f9b39684c
101692Author: Eric Anholt <eric@anholt.net>
101693Date: Wed Sep 12 13:58:46 2007 +0000
101694
101695 Remove the PaintWindow optimization.
101696
101697 This was an attempt to avoid scratch gc creation and validation for paintwin
101698 because that was expensive. This is not the case in current servers, and the
101699 danger of failure to implement it correctly (as seen in all previous
101700 implementations) is high enough to justify removing it. No performance
101701 difference detected with x11perf -create -move -resize -circulate on Xvfb.
101702 Leave the screen hooks for PaintWindow* in for now to avoid ABI change.
101703
101704commit 06d27f8045966c1fb154eafaff308a01b93f265b
101705Author: Keith Packard <keithp@koto.keithp.com>
101706Date: Wed Sep 12 23:57:30 2007 +0100
101707
101708 Try again to fix drawable and tile offsets in miPaintWindow
101709
101710 Many coordinate spaces are hard. Let's go drinking.
101711
101712commit dd3992eb86377684a5dbe86fa19c756a9e53cda2
101713Author: Keith Packard <keithp@koto.keithp.com>
101714Date: Wed Sep 12 22:39:31 2007 +0100
101715
101716 miPaintWindow draw to window for background.
101717
101718 Instead of drawing to window pixmap for everything, draw to window for
101719 background as that works for Xnest and Xdmx; draw to pixmap for borders
101720 which neither of those X servers use.
101721
101722commit 6da39c67905500ab2db00a45cda4a9f756cdde96
101723Author: Eric Anholt <eric@anholt.net>
101724Date: Wed Sep 12 13:23:13 2007 +0000
101725
101726 Fix build on FreeBSD after Popen changes.
101727
101728commit 257c8ed17f4f908e0d0d5e53aaf13aa3b1313f50
101729Author: Keith Packard <keithp@koto.keithp.com>
101730Date: Wed Sep 12 12:11:49 2007 +0100
101731
101732 Rewrite miPaintWindow to draw to window pixmap.
101733
101734 miPaintWindow was drawing to the root window, or (sometimes) drawing to the
101735 window after smashing the window clip list. This is losing, and easily fixed
101736 by just drawing to the window pixmap.
101737
101738commit 43dc41034735d84765233a78bd619f3f0e7ace75
101739Author: Brice Goglin <bgoglin@debian.org>
101740Date: Wed Sep 12 01:43:37 2007 +0200
101741
101742 Replace a non-ascii char with the corresponding groff escape in exa.man.pre
101743
101744commit a5b8053606d6e786cdcf6734f271acc05f9cc588
101745Author: Adam Jackson <ajax@benzedrine.nwnk.net>
101746Date: Tue Sep 11 11:37:06 2007 -0400
101747
101748 Ignore - not just block - SIGALRM around Popen()/Pclose().
101749
101750 Because our "popen" implementation uses stdio, and because nobody's stdio
101751 library is capable of surviving signals, we need to make absolutely sure
101752 that we hide the SIGALRM from the smart scheduler. Otherwise, when you
101753 open a menu in openoffice, and it recompiles XKB to deal with the
101754 accelerators, and you popen xkbcomp because we suck, then the scheduler
101755 will tell you you're taking forever doing something stupid, and the
101756 wait() code will get confused, and input will hang and your CPU usage
101757 slams to 100%. Down, not across.
101758
101759commit 2e3e08d31e908ceadeef16d6069cdaf8c61d7ed1
101760Author: Dodji Seketeli <dodji@openedhand.com>
101761Date: Tue Sep 11 08:50:41 2007 +0200
101762
101763 xserver: test presence of strlcpy in configure
101764
101765 * configure.ac: define HAVE_STRLCPY when function strlcat is present
101766 this serves at least to unbreak libxtrans code compilation.
101767
101768commit e8c73ac697aa472e10b4d8c093a715df0bf545af
101769Author: Dodji Seketeli <dodji@openedhand.com>
101770Date: Tue Sep 11 00:20:45 2007 +0200
101771
101772 Xephyr: fix compilation breakage.
101773
101774 * hw/kdrive/ephyr/ephyr_draw.c:
101775 (exaDDXDriverInit): pExaScr->hideOffscreenPixmapData does not exist
101776 anymore, so don't set it.
101777
101778commit 321e0a21600e418bbeb164043a9a21a0ff80cbe9
101779Author: Ben Byer <bbyer@bbyer.local>
101780Date: Sun Sep 9 16:28:41 2007 -0700
101781
101782 fixed cut-and-paste typo
101783
101784commit 0ff273fd1e2ea7242b3e6c5effb2a623ef32ec6f
101785Author: Ben Byer <bbyer@bbyer.apple.com>
101786Date: Thu Sep 6 18:52:26 2007 -0700
101787
101788 XDARWIN: fixes to make Xquartz build again
101789
101790commit f8637137ab43818d45d249b337820cbf427a05ec
101791Author: Ben Byer <bbyer@bbyer.local>
101792Date: Thu Sep 6 05:39:57 2007 -0700
101793
101794 XDARWIN: build fix for quartz/pseudoramiX.c
101795
101796commit e8093e15c7df7a3d5a9717bc9d7d7517b0743f29
101797Author: Michel Dänzer <michel@tungstengraphics.com>
101798Date: Mon Sep 3 13:52:29 2007 +0200
101799
101800 EXA: Exclude bits that will be overwritten from migration in exaCopyNtoN.
101801
101802 Also plug a region leak in exaPolyFillRect.
101803
101804commit e81af8ba643df3be53b0a46d9d4a0eaf21557c9e
101805Author: Michel Dänzer <michel@tungstengraphics.com>
101806Date: Mon Sep 3 13:14:29 2007 +0200
101807
101808 EXA: exaFillRegion{Solid,Tiled} improvements.
101809
101810 Use region to exclude bits that will be overwritten from migration.
101811
101812 Also make exaFillRegionSolid use the same logic as exaFillRegionTiled.
101813
101814commit 5f7da4da8de7449e1c2a4c679632a0b2a5858b7e
101815Author: Michel Dänzer <michel@tungstengraphics.com>
101816Date: Fri Aug 31 16:59:28 2007 +0200
101817
101818 EXA: Use exaShmPutImage for pushing glyphs to scratch pixmap in exaGlyphs.
101819
101820commit be922b30486abce3a8c13996d579b211a7b56f0e
101821Author: Michel Dänzer <michel@tungstengraphics.com>
101822Date: Thu Aug 30 13:59:07 2007 +0200
101823
101824 EXA: exa(Shm)PutImage improvements.
101825
101826 Improve exaShmPutImage performance and reuse its core in exaPutImage as it
101827 seems faster than the previous code when the driver doesn't provide an
101828 UploadToScreen hook.
101829
101830 Make sure all damage records are notified of the damage incurred by actual
101831 ShmPutImage calls.
101832
101833 Remove superfluous manual damage tracking for actual PutImage calls.
101834
101835commit ea92ea415665e294a1ba233e9a1d39b6daa0cee1
101836Author: Michel Dänzer <michel@tungstengraphics.com>
101837Date: Thu Aug 30 13:54:18 2007 +0200
101838
101839 EXA: exaGetImage improvements.
101840
101841 Use the new migration infrastructure to cache FB bits we need in the system
101842 copy, for the benefit of repeated calls.
101843
101844commit aa2ed73e0ec881947c969b67269e3206da4de359
101845Author: Michel Dänzer <michel@tungstengraphics.com>
101846Date: Thu Aug 30 13:50:42 2007 +0200
101847
101848 EXA: Remove superfluous manual damage tracking.
101849
101850 These should all be covered by damage wrappers.
101851
101852commit a634c9b03494ba80aeec28be19662ac96657cc23
101853Author: Michel Dänzer <michel@tungstengraphics.com>
101854Date: Thu Aug 30 13:48:03 2007 +0200
101855
101856 EXA: RENDER improvements.
101857
101858 Exclude bits that will be overwritten from migration.
101859
101860 Use exaGlyphs even when Composite can't be accelerated, to avoid PolyFillRect
101861 roundtrip via offscreen memory.
101862
101863 Initialize mask pixmap in exaGlyphs in FB in addition to system if the driver
101864 provides Composite hooks to avoid migration overhead.
101865
101866 Remove manual damage tracking where superfluous.
101867
101868commit 1f457ff3db24178eefecfbbf177aaf6554adb204
101869Author: Michel Dänzer <michel@tungstengraphics.com>
101870Date: Thu Aug 30 13:44:20 2007 +0200
101871
101872 EXA: Improvements for 1x1 pixmaps.
101873
101874 Initialize system and FB copy in exaFillRegionSolid and adapt
101875 exaGetPixmapFirstPixel to the new migration infrastructure.
101876
101877 This should mostly eliminate migration overhead for these, whether they are
101878 used for acceleration or fallbacks.
101879
101880commit 489bc7551ffc7360ba9648ca5c98b59c7e7a1fd1
101881Author: Michel Dänzer <michel@tungstengraphics.com>
101882Date: Thu Aug 30 13:37:53 2007 +0200
101883
101884 EXA: exaImageGlyphBlt improvements.
101885
101886 As we can't actually accelerate anything interesting here, just migrate out
101887 once and call fbSolidBoxClipped instead of taking a round trip via offscreen
101888 memory with exaSolidBoxClipped.
101889
101890 Reuse pending damage region for extents and to prevent any actual migration of
101891 pixmap contents when we're overwriting the whole pending damage region.
101892
101893 Remove superfluous manual damage tracking.
101894
101895commit 2e0895a4ba27c1308713022820444c8f57f7a69f
101896Author: Michel Dänzer <michel@tungstengraphics.com>
101897Date: Thu Aug 30 13:30:03 2007 +0200
101898
101899 EXA: Improvements for trapezoids and triangles.
101900
101901 Only migrate once in exaTrapezoids/Triangles instead of every time in
101902 exaRasterizeTrapezoid/AddTriangles. Adapt manual damage tracking to new
101903 infrastructure.
101904
101905 Also move definition of NeedsComponent() closer to where it's used.
101906
101907commit 6c9d7ed61bc4a19d21c53717b8af3d90b5d82ca9
101908Author: Michel Dänzer <michel@tungstengraphics.com>
101909Date: Thu Sep 6 13:10:16 2007 +0200
101910
101911 EXA: Hide pixmap pointer outside of exaPrepare/FinishAccess whenever possible.
101912
101913 We finally want to catch all cases where the pixmap pointer is dereferenced
101914 outside of exaPrepare/FinishAccess.
101915
101916 Also fix a couple of such cases exposed by this change.
101917
101918commit 962eddd7a2863a8475f5fd8107d3112df08d1172
101919Author: Michel Dänzer <michel@tungstengraphics.com>
101920Date: Wed Aug 29 19:55:22 2007 +0200
101921
101922 EXA: Support partial migration of pixmap contents between Sys and FB.
101923
101924 The initiator of migration can pass in a region that defines the relevant area
101925 of each source pixmap or the irrelevant area of the destination pixmap. By
101926 default, the pending damage region is assumed relevant for the destination
101927 pixmap, and everything for source pixmaps.
101928
101929 Thanks to Jarno Manninen for reassuring me that my own ideas for this were
101930 feasible and for providing additional ideas.
101931
101932commit f27931bdd26fc9a1e6bb5173b5537e32c51a98b3
101933Author: Michel Dänzer <michel@tungstengraphics.com>
101934Date: Wed Aug 29 19:41:52 2007 +0200
101935
101936 Add DamagePendingRegion.
101937
101938 DamagePendingRegion returns a pointer to the region of a drawable that will
101939 be damaged by the current operation for damage records that chose to get damage
101940 reported only at the end of the operation.
101941
101942commit 5c7ee3f47fa0c067102a17dee3f75a51cc0bdb3a
101943Author: Michel Dänzer <michel@tungstengraphics.com>
101944Date: Fri Aug 24 19:24:18 2007 +0200
101945
101946 EXA: Track valid bits in Sys and FB separately.
101947
101948 Also consolidate exaCopyDirtyToFb/Sys.
101949
101950commit 8cfcf9973c765f11d1b45b95b8091ef7e01d7f01
101951Author: Michel Dänzer <michel@tungstengraphics.com>
101952Date: Wed Sep 5 20:10:09 2007 +0200
101953
101954 EXA: Migrate out pixmap in exaPrepareAccess.
101955
101956 Also fix exaFinishAccessGC not to use the same index for tile and stipple.
101957
101958commit e510a77ba4d65d5d6ead514cd698f1b1e3f8a2b6
101959Author: Dave Airlie <airlied@linux.ie>
101960Date: Tue Jul 17 17:16:51 2007 +1000
101961
101962 EXA: Add a couple of missing exaPrepare/FinishAccess calls.
101963
101964commit 72b347e681f5667b68257822e7cec02ab4c9cb6d
101965Author: Peter Hutterer <peter@cs.unisa.edu.au>
101966Date: Fri Sep 7 14:31:19 2007 +0930
101967
101968 dix: Enabling devices must not overwrite existing sprites/pairing.
101969
101970 EnableDevices is (amongst others )called after a VT switch. We must not create
101971 a new sprite or re-pair the device, otherwise we lose the input device setup
101972 that we had before the VT switch.
101973
101974 This requires the devices to be in exactly the same order as before
101975 the VT switch. Removing a device while on a different VT is probably a bad
101976 idea.
101977
101978commit 57907e0943da0c3fd3bf6c128d210b544629ce72
101979Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
101980Date: Thu Sep 6 16:55:51 2007 -0400
101981
101982 devPrivates rework: register an offset for every resource type, use
101983 signed values so -1 actually works correctly, and provide a macro for
101984 adding an offset to a pointer.
101985
101986commit 84eb7e62248ddc2761af8cefe33d1b7147477528
101987Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
101988Date: Thu Sep 6 13:36:37 2007 +0100
101989
101990 Fix Xdmx build
101991
101992commit 35ec8b89f4cdf5ec48b292d47b2dad42d2fb9534
101993Author: Ben Byer <bbyer@bbyer.local>
101994Date: Thu Sep 6 05:34:14 2007 -0700
101995
101996 DDX changes for XDarwin
101997
101998 Creating a real DDX section for XDarwin instead of
101999 stuffing it into the XORG section in a half-assed manner.
102000
102001commit d627061b48ae06d27b37be209d67a3f4f2388dd3
102002Author: Peter Hutterer <peter@cs.unisa.edu.au>
102003Date: Thu Sep 6 18:57:00 2007 +0930
102004
102005 xfree86: wrap keyboard devices for XKB.
102006
102007 Call ProcessOtherEvents first, then for all keyboard devices let them be
102008 wrapped by XKB. This way all XI events will go through XKB.
102009
102010 Note that the VCK is still not wrapped, so core events will bypass XKB.
102011
102012commit 5ee409794ee604fcf84886f70429fc2d6b1ff4f1
102013Author: Peter Hutterer <peter@cs.unisa.edu.au>
102014Date: Thu Sep 6 18:52:02 2007 +0930
102015
102016 dix: add XI event support to FixKeyState.
102017
102018 FixKeyState needs to be able to handle XI events, otherwise we get "impossible
102019 keyboard events" on server zaps and other special key combos.
102020
102021commit 6334d4e7be18de5f237c12a6dc20f75aa23477d0
102022Author: Peter Hutterer <peter@cs.unisa.edu.au>
102023Date: Thu Sep 6 18:49:57 2007 +0930
102024
102025 xkb: enable XI event processing for xkb.
102026
102027 XI events can now take the same processing paths as core events, and should do
102028 the correct state changes etc.
102029
102030 There's some cases where XKB will use KeyPress as type for an event to be
102031 delivered to the client. Stuck warnings in, not sure what the correct solution
102032 is yet.
102033
102034commit 49dbe9a757a3d7a0b9ab318242c6cc0cbd4dd1f0
102035Author: Peter Hutterer <peter@cs.unisa.edu.au>
102036Date: Thu Sep 6 17:27:28 2007 +0930
102037
102038 dix: close virtual core devices after other devices.
102039
102040 If a device is paired with the VCP, deleting the VCP before the device will
102041 segfault the server when the sprite should get updated.
102042
102043commit 03680a384aa423ece75b658f00b96db2628c39fa
102044Author: Peter Hutterer <peter@cs.unisa.edu.au>
102045Date: Thu Sep 6 15:43:47 2007 +0930
102046
102047 dix: don't change the device struct while processing core events.
102048
102049 The device state needs to be changed while processing the XI event. Core
102050 events are always processed after XI, so by then the device is already set up
102051 properly. However, we now rely on DeviceButtonMotionMask to be equal to
102052 ButtonMotionMask. It already is, but stick a big fat warning in so nobody
102053 attempts to change it.
102054
102055 This commit disables XKB for the VCK, thus essentially for all devices.
102056 Temporarily anyway.
102057
102058commit f3f614cd01ae56d84d61b4f5c907c6edd7c8c6d3
102059Author: Ben Byer <bbyer@bbyer.local>
102060Date: Wed Sep 5 20:34:00 2007 -0700
102061
102062 Revert "configure.ac: exclude pci-access dependency from XDarwin"
102063
102064 This reverts commit 20c6677d1b5f8d77325dd878ffa3df1d0fb01864.
102065
102066commit bf1641b94cffa54b786e18eaeff3839d8790b9f2
102067Author: Ben Byer <bbyer@bbyer.local>
102068Date: Wed Sep 5 20:33:43 2007 -0700
102069
102070 XDARWIN: Add launchd support
102071
102072 This adds a bit of glue to configure.ac to support launchd detection;
102073 on OS X (or other platforms which choose to implement launchd), this allows
102074 the system to automagically start the Xserver as necessary to serve clients.
102075
102076commit 71c21dea748ea0dcad758679c40ee39192d170f9
102077Author: Ben Byer <bbyer@bbyer.apple.com>
102078Date: Wed Sep 5 18:45:50 2007 -0700
102079
102080 added HAVE_LAUNCHD check to configure.ac (mostly for OSX)
102081
102082commit 81c28ffd2b13a83770eadcfd7829d35d319d637f
102083Author: Daniel Stone <daniel@fooishbar.org>
102084Date: Wed Sep 5 17:46:23 2007 -0700
102085
102086 Fix key repeats during VT switch.
102087
102088 Add keyc->postdown, which represents the key state as of the last mieqEnqueue
102089 call, and use it when we need to know the posted state, instead of the
102090 processed state (keyc->down). Add small functions to getevents.c to query and
102091 modify key state in postdown and use them all through, eliminating previously
102092 broken uses.
102093
102094commit e332335241af28ef0ab66b102d0cbc4e5c73ac68
102095Merge: 7381e91 accd71b
102096Author: Ben Byer <bbyer@bbyer.apple.com>
102097Date: Wed Sep 5 15:51:23 2007 -0700
102098
102099 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/xserver
102100
102101commit 7381e9149e3cbb7e672070781a16e3f096202be9
102102Author: Ben Byer <bbyer@bbyer.apple.com>
102103Date: Wed Sep 5 15:51:11 2007 -0700
102104
102105 added an exclusion for setting XORG on darwin
102106
102107commit accd71bda6f958ea6892ad3a10879232d345774c
102108Author: Keith Packard <keithp@koto.keithp.com>
102109Date: Wed Sep 5 14:19:19 2007 -0700
102110
102111 Deliver correct event when releasing keys on VT switch.
102112
102113 In commit 41bb9fce47f6366cc3f7d45790f7883f74289b5a, the event delivery loop
102114 for Xinput enabled keyboards was changed and accidentally used the wrong
102115 index variable, causing random events to be delivered when returning from VT
102116 switch.
102117
102118 In addition, in commit aeba855b07832354f59678e20cc29a085e42bd99,
102119 SIGIO was blocked during delivery of these events, but not for the entire
102120 period the xf86Events array was being used. Block SIGIO for the whole loop
102121 to avoid other event delivery from trashing the key release events.
102122 (cherry picked from commit aa7ed1f5f35cd043bc38d985500aa0a32e857e84)
102123
102124commit 8b77dc7e808f61f1ed10fe05cf898bb47459a76d
102125Author: Ben Byer <bbyer@bbyer.apple.com>
102126Date: Wed Sep 5 14:48:38 2007 -0700
102127
102128 XDARWIN: build breakage fix
102129
102130commit 20c6677d1b5f8d77325dd878ffa3df1d0fb01864
102131Author: Ben Byer <bbyer@bbyer.apple.com>
102132Date: Wed Sep 5 14:31:01 2007 -0700
102133
102134 configure.ac: exclude pci-access dependency from XDarwin
102135
102136 XDarwin doesn't need any of this pci stuff since it doesn't talk directly to hardware,
102137 so it doesn't make sense to require it when building on OSX/Darwin.
102138
102139commit ff01e44e33fd072958fb0157dae072f1b1c88944
102140Merge: 8ba8c16 47300ed
102141Author: Ben Byer <bbyer@bbyer.apple.com>
102142Date: Wed Sep 5 13:56:08 2007 -0700
102143
102144 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/xserver
102145
102146commit 8ba8c16af773ec83a0b1c0661a23d746b401944e
102147Author: Ben Byer <bbyer@bbyer.apple.com>
102148Date: Wed Sep 5 13:46:30 2007 -0700
102149
102150 build fixes for XDarwin
102151
102152commit 47300ed2be59d0ba7ea9345b954bf3104877c095
102153Author: Eric Anholt <eric@anholt.net>
102154Date: Wed Sep 5 12:34:29 2007 -0700
102155
102156 Fix server version reporting to be the server package version.
102157
102158 Previously, the server version reported by xdpyinfo and Xorg -version would
102159 bear some vague resemblance to a X.Org katamari version, but in the presence
102160 of modularization (and client-server relationships with different katamari
102161 versions on each side) those numbers don't really make sense. Instead, just
102162 report the package version.
102163
102164 When branching a stable branch, master's version should be immediately updated
102165 to the endpoint of the stable branch plus a snapshot of 1 (for example,
102166 1.4.0.1 after server-1.4-branch). The stable branch should then be changed to
102167 RC0 at that time (1.3.99.0, for example).
102168
102169 This scheme was partially attempted for server 1.3, but lacked the appropriate
102170 master updates, thus why it had to be revisited now. While here, we can also
102171 remove a lot of versioning complexity since everything is based on the package
102172 version.
102173
102174commit 6c89d1237c4fdce961b30a8eaee964af5d56565e
102175Author: Dodji Seketeli <dodji@openedhand.com>
102176Date: Wed Sep 5 17:46:49 2007 +0200
102177
102178 Kdrive: unbreak kdrive linking
102179
102180 * configure.ac: re-sort Kdrive libs so that symbols get properly resolved.
102181 Basically, all some libs are present in both $KDRIVE_LIBS and $XSERVER_LIBS,
102182 and some libs orders are not correct. So I made sure Kdrive servers don't have
102183 to link against $KDRIVE_LIBS *and* $XSERVER_LIBS. They just have to link
102184 against $KDRIVE_LIBS now.
102185 * hw/kdrive/*/Makefile.am: update those makefile to reflect the change in configure.ac
102186
102187commit 0003ccfcdfae1b473aa024342304b84256d378b9
102188Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102189Date: Wed Sep 5 11:18:36 2007 -0400
102190
102191 xace: add new fields to resource access hook to allow parent resource objects
102192 to be passed in at create time. Also added a missing devPrivates initializer.
102193
102194commit 28e48cd8e6e4c412a49d7177daad6d3c93c28e08
102195Author: Ben Byer <bbyer@bbyer.local>
102196Date: Wed Sep 5 04:43:17 2007 -0700
102197
102198 Another pathname fix for event_status_driver.h
102199
102200commit b6c0697fd67323893a9ad3676c33f2f1ec48e15e
102201Author: Ben Byer <bbyer@bbyer.local>
102202Date: Wed Sep 5 04:40:03 2007 -0700
102203
102204 fixed path for event_status_driver.h
102205 (The path currently used is deprecated on Tiger and invalid on Leopard.)
102206
102207commit c4fff050836feeef8390b7197f1de39af2997811
102208Author: Peter Hutterer <peter@cs.unisa.edu.au>
102209Date: Wed Sep 5 16:19:45 2007 +0930
102210
102211 Revert "Input: Fix stuck modifiers (bug #11683)"
102212
102213 This reverts commit 6b055e5d9751e3679ff98065e43225ec8a960053.
102214
102215 MPX relies on the XI event being delivered before the core event. Device grabs
102216 break, amongst other things. I guess stuck modifiers need to be fixed some
102217 other way.
102218
102219 Conflicts:
102220
102221 dix/getevents.c
102222
102223commit bfe6b4d2d9952a80f8dbc63eec974ef894e5c226
102224Author: Peter Hutterer <peter@cs.unisa.edu.au>
102225Date: Tue Sep 4 17:44:51 2007 +0930
102226
102227 xkb: Store the action filters per device in the XkbSrvInfoRec.
102228
102229 Using a global array for action filters is bad. If two keyboard hit a modifier
102230 at the same time, releaseing the first one will deactivate the filter and
102231 thus the second keyboard can never release the modifier again.
102232
102233commit cc5c926267be099d793e6dfec17916f21c73c64d
102234Author: Peter Hutterer <peter@cs.unisa.edu.au>
102235Date: Thu Aug 30 15:51:22 2007 +0930
102236
102237 randr: RRPointerScreenConfigured needs to move all pointers.
102238
102239 Previous version only moved the VCP, causing "bogus pointer events" lateron.
102240 Now we run through the device list, updating each pointer separately if
102241 necessary.
102242
102243 Also stick a big warning into RRPointerMoved, not sure what device we need to
102244 work on here.
102245
102246commit 136fde2c7b5ed590bc6c63d11ede31e92dc679a2
102247Author: Ademar de Souza Reis Jr <ademar@mandriva.com.br>
102248Date: Tue Jul 3 13:44:28 2007 -0300
102249
102250 Add Xserver man section about catalogue:<dir> FPE
102251
102252 catalogue:<dir> FPEs were introduced in libXfont 1.2.9
102253
102254commit eb6a933dc60bec5601260794eeb973e946af37b6
102255Author: Ademar de Souza Reis Jr <ademar@mandriva.com.br>
102256Date: Tue Jul 3 13:44:03 2007 -0300
102257
102258 Add xorg.conf man section about catalogue:<dir> FPE
102259
102260 catalogue:<dir> FPEs were introduced in libXfont 1.2.9
102261
102262commit 205183a733237ea418a25c7423b689fcc8eae628
102263Author: Eric Anholt <eric@anholt.net>
102264Date: Tue Sep 4 16:19:22 2007 -0700
102265
102266 Fix driver build by including an appropriate Requires.private line on pixman.
102267
102268 We'd previously been substituting PIXMAN_CFLAGS, but we've got a better tool
102269 now, plus I deleted the PIXMAN_CFLAGS substitution without noticing.
102270
102271commit d67e210f3458b62d7d4a6032aabfda0004d661c1
102272Author: Gerte Hoogewerf <g.hoogewerf@gmail.com>
102273Date: Tue Sep 4 16:09:38 2007 -0700
102274
102275 Add stub symbols to make xprint build.
102276
102277commit f7f79724fdea0cc6fda0e90e56431df937d49335
102278Author: Eric Anholt <eric@anholt.net>
102279Date: Tue Sep 4 15:10:49 2007 -0700
102280
102281 Increase despair by fixing xprint build after my _DEPENDENCIES changes.
102282
102283commit e89d16be07e45e487913509788a9e8cb1ee09bc7
102284Author: Ian Romanick <idr@us.ibm.com>
102285Date: Tue Sep 4 14:49:49 2007 -0700
102286
102287 Revert part of 529acb175440969af9d7fa38aab8d7dea0dc2661 because libtool is smart.
102288
102289commit bf5948518763b5e21eff806a0a9abc5f7757fa10
102290Merge: 4062db4 735da3d
102291Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
102292Date: Tue Sep 4 22:40:31 2007 +0200
102293
102294 Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/xserver
102295
102296commit 4062db4020c671fc0921a3a4e7fe5d8dc1be2e1d
102297Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
102298Date: Tue Sep 4 22:39:14 2007 +0200
102299
102300 typo in comment
102301
102302commit ce9e83d913511fe619da42f805d7bcd1a2a60d90
102303Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102304Date: Tue Sep 4 14:01:55 2007 -0400
102305
102306 xace: add hooks + new access codes: Damage extension
102307
102308commit 735da3dcd09f59c40f68eca8c9a91c2e826cd1b9
102309Author: Egbert Eich <eich@freedesktop.org>
102310Date: Tue Sep 4 16:42:57 2007 +0200
102311
102312 Fixing xf86I2CGetScreenBuses(), now dereferencing the correct pointer.
102313
102314commit 7c5de093d499de63d39fe038d86da0a085262017
102315Author: Egbert Eich <eich@freedesktop.org>
102316Date: Tue Sep 4 15:07:17 2007 +0200
102317
102318 Fixing sig11 in xf86I2CGetScreenBuses().
102319
102320 Dereferencing a pointer once too often caused a
102321 sig11 in xf86I2CGetScreenBuses().
102322
102323commit 9adea807038b64292403ede982075fe1dcfd4c9a
102324Author: Hong Liu <hong.liu@intel.com>
102325Date: Tue Sep 4 08:46:46 2007 +0100
102326
102327 bgPixel (unsigned long) is 64-bit on x86_64, so -1 != 0xffffffff
102328
102329 This patch should fix bug 8080.
102330
102331commit 529acb175440969af9d7fa38aab8d7dea0dc2661
102332Author: Eric Anholt <eric@anholt.net>
102333Date: Sun Sep 2 15:16:01 2007 -0700
102334
102335 Fix Xorg build by listing circular dependency libraries twice.
102336
102337 One of these I introduced by listing dix and mi in the same library list to
102338 simplify other servers. The other had been hacked around using libosandcommon,
102339 which is now gone.
102340
102341commit c56930e6ca90a61d2f4cbd845f7d0a51a66f83cb
102342Author: Eric Anholt <eric@anholt.net>
102343Date: Fri Aug 31 18:27:41 2007 -0700
102344
102345 Remove backend.[ch] from neomagic to fix distcheck.
102346
102347commit f98dfec79dadb70fa7bba84e7335f92b3a73dc02
102348Author: Keith Packard <keithp@koto.keithp.com>
102349Date: Sat Sep 1 21:14:22 2007 -0700
102350
102351 [COMPOSITE] Composite used for pixmap population on redirect. (Bug #7447)
102352
102353 compNewPixmap copies bits from the parent window to the redirected child
102354 pixmap to populate the pixmap with reasonable data. It cannot always use
102355 CopyArea as that only works across matching depths. Use Composite when
102356 the depths do not match.
102357
102358commit 1afdf8b0a92437dffe84fa98b6083b3d8fd55e27
102359Author: Adam Jackson <ajax@redhat.com>
102360Date: Fri Aug 31 22:11:13 2007 -0700
102361
102362 [RANDR] Don't mark Xinerama as active if no crtcs are enabled. (bug #11504).
102363
102364 Clients expect any Xinerama-enabled screen to report at least one
102365 monitor, but with RandR, there may not be any enabled crtcs. In this case,
102366 tell the client that Xinerama is not active.
102367
102368commit 0dc2bb6101704d0fd25f36e2c3df79687f119f5b
102369Author: Marius Gedminas <mgedmin@b4net.lt>
102370Date: Fri Aug 31 21:36:37 2007 -0700
102371
102372 [RANDR] Compare only milliseconds of config time. (Bug #6502)
102373
102374 The timestamp transferred in the X protocol is a 32-bit number of
102375 milliseconds.
102376
102377 The timestamp stored in the server is a structure that contains two fields:
102378 months (!) and milliseconds.
102379
102380 When the server passes the config timestamp to the client, it discards the
102381 months part and sends only the milliseconds part.
102382
102383 When the server receives the config timestamp from the client, it tries to
102384 guess the "months" part by looking at the current time and then maybe adding
102385 or
102386 subtracting one. The guess is wrong after the server has been running long
102387 enough (several hours).
102388
102389 I have added two ErrorF calls around the 'if' statement that returns
102390 RRSetConfigInvalidConfigTimestamp in randr/randr.c and my Xorg.0.log has
102391 this:
102392
102393 randr request got good config time: 0:-2103495671
102394
102395 for the first few successful xrandr calls, and
102396
102397 randr request failed with RRSetConfigInvalidConfigTime: client passed
102398 1:-2103495671, server has 0:-2103495671
102399
102400 when it fails. The server has been running for 8 and a half hours.
102401
102402 The obvious fix would be to ignore the months field and only compare the
102403 milliseconds.
102404
102405commit 07630d897ef37cad8b79d073d9edc891d5a7bddd
102406Author: Eric Anholt <eric@anholt.net>
102407Date: Fri Aug 31 15:16:01 2007 -0700
102408
102409 Bug #7364: Require renderproto 0.9.3 on 64-bit, and fix build with it.
102410
102411commit ca82d4bddf235c9b68d51d68636bab40eafb9889
102412Author: Eric Anholt <eric@anholt.net>
102413Date: Fri Aug 31 13:00:23 2007 -0700
102414
102415 Bug #7186: Fix an excessive request size limitation that broke big-requests.
102416
102417 MAXBUFSIZE appears to be a leftover of some previous time. Instead, just
102418 use maxBigRequestSize when bigreqs are available (limiting buffers to ~16MB).
102419 When bigreqs are not available, needed won't be larger than the maximum
102420 size of a non-bigreqs request (256kB).
102421
102422commit a02db0d500cac20d0f0f107d27c064a175018421
102423Author: Eric Anholt <eric@anholt.net>
102424Date: Fri Aug 31 12:24:07 2007 -0700
102425
102426 Convert servers to using _DEPENDENCIES to ensure proper rebuilds.
102427
102428 This cleans up server Makefile.ams a little bit, but also means that people
102429 messing with configure.ac need to be careful with whether they put libraries
102430 in the _LIBS or _SYS_LIBS targets. Hopefully the comment in configure.ac will
102431 clarify the issues.
102432
102433commit c9ceb4878063ca22487c708d9d1f86e367f2cec8
102434Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102435Date: Fri Aug 31 11:03:54 2007 -0400
102436
102437 xace: add hooks + new access codes: Composite extension
102438
102439commit fd04b983db6a70bf747abe02ca07c1fbbaae6343
102440Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102441Date: Fri Aug 31 09:55:27 2007 -0400
102442
102443 xace: add hooks + new access codes: Render extension
102444
102445commit 3f42af8c0ef1e5379bc836f589e0cbee43c02ac5
102446Author: Peter Hutterer <peter@cs.unisa.edu.au>
102447Date: Thu Aug 30 18:22:12 2007 +0930
102448
102449 config: Use [config/dbus] consistently for error messages.
102450
102451commit 0fcde83d94507eadd9f99d4e6a63584b221c989c
102452Author: Peter Hutterer <peter@cs.unisa.edu.au>
102453Date: Thu Aug 30 18:20:20 2007 +0930
102454
102455 config: return BadValue to caller if add/remove doesn't have parameters.
102456
102457 If message iterator cannot be created, the caller didn't supply any
102458 parameters. Return BadValue, instead of dying a horrible death while being
102459 stuck in an endless loop.
102460
102461commit 87495fc7064d5e0a7575a0713b6895a4172df0fa
102462Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
102463Date: Thu Aug 30 21:57:41 2007 +0100
102464
102465 Allow yres_virtual to be greater for some kernel fbdev drivers.
102466 (temporary fix for older fbdev drivers)
102467
102468commit 1005b29cc6939851b40397cc9cd0de9476ad3046
102469Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102470Date: Thu Aug 30 14:48:24 2007 -0400
102471
102472 xace: Correct some access modes.
102473
102474commit d8ab2353cbd7694b556b7b9d550104dc8c460a1b
102475Author: Eric Anholt <eric@anholt.net>
102476Date: Thu Aug 30 10:39:53 2007 -0700
102477
102478 Remove dead xf86GetPciSizeFromOS and xf86GetPciOffsetFromOS.
102479
102480commit c2d80529fc7f514d80cf3cbed6f580cb999aca1b
102481Author: Eric Anholt <eric@anholt.net>
102482Date: Thu Aug 30 10:20:55 2007 -0700
102483
102484 Remove the now-dead PciAvoid symbol.
102485
102486commit 53f346b158fa8e10de5a8777fa6d8d86f918878b
102487Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102488Date: Thu Aug 30 13:20:04 2007 -0400
102489
102490 xace: add hooks + new access codes: SHAPE extension
102491
102492commit 766c693ef3637ee6fc402df594060ed2c1346761
102493Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102494Date: Thu Aug 30 13:06:28 2007 -0400
102495
102496 xace: add hooks + new access codes: MIT-SCREEN-SAVER extension
102497
102498commit cda92bbf12107865e93c03c71b901ef51466dc31
102499Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102500Date: Thu Aug 30 11:48:45 2007 -0400
102501
102502 xace: add hooks + new access codes: XFixes extension.
102503
102504 Required a new name argument to the selection access hook to handle
102505 XFixesSelectSelectionInput.
102506
102507commit 47ab4d648b31ea1d5800e0bc84cf5f25025bffe3
102508Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102509Date: Thu Aug 30 11:40:39 2007 -0400
102510
102511 devPrivates rework: convert CursorRec and CursorBits over to new interface.
102512
102513commit 1d11e4bc4ccb169fb23fc18583f0b648f0a6a4e0
102514Author: Egbert Eich <eich@freedesktop.org>
102515Date: Thu Aug 30 12:50:21 2007 +0200
102516
102517 Fixing a misleading comment which could suggest a GPL violation.
102518
102519 The author of the int10 code looked at the VBIOS POSTing code
102520 in DOSEMU to get some initial idea on how to POST a VBIOS.
102521 To give credit to the DOSEMU Team for this inspiration a comment
102522 was added to the code which could suggest that code from the
102523 GPLed DOSEMU was directly incorporated into this code.
102524 This patch should clearify the situation.
102525
102526commit 45efe85003195bd45501630cef08349abb180c3a
102527Author: Eric Anholt <eric@anholt.net>
102528Date: Wed Aug 29 16:05:51 2007 -0700
102529
102530 Remove stale changelogs from kdrive.
102531
102532commit 5aaf00190157114780ab51f7268b396459ed1cad
102533Author: Eric Anholt <eric@anholt.net>
102534Date: Wed Aug 29 16:05:23 2007 -0700
102535
102536 Add more generated files to .gitignore.
102537
102538commit 87295b66a972a2bd194a79af6aa4f715018fcded
102539Author: Eric Anholt <eric@anholt.net>
102540Date: Wed Aug 29 15:54:32 2007 -0700
102541
102542 Bug #9629: Remove badly-licensed neomagic kdrive files.
102543
102544 Licensing issues of these files include:
102545 - They claim to be licensed under the GPL, yet we haven't allowed that in the
102546 xserver repository in the past.
102547 - They refer the user to the top of the tree for GPL license text, yet it isn't
102548 there.
102549 - They claim to be derived from the (MIT-licensed) ati kdrive code, yet don't
102550 follow the licensing terms of those files.
102551
102552commit 4795df62456b73c6790f271e0a20a83c60496490
102553Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102554Date: Wed Aug 29 14:40:10 2007 -0400
102555
102556 xace: add hooks + new access codes: TOG-CUP extension.
102557
102558commit e39694789e31e221fc8dec44ace9c697daf7acad
102559Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102560Date: Wed Aug 29 14:16:46 2007 -0400
102561
102562 xace: drop map-window checking hook, add new hooks for controlling the
102563 delivery of events to windows and clients.
102564
102565 This is tentative. It's likely that an additional last-resort hook will
102566 be necessary for code that calls TryClientEvents or WriteEventsToClient
102567 directly. It's also possible that new xace machinery will be necessary
102568 to classify events and pull useful resource ID's out of them.
102569
102570 The failure case also needs some thinking through. Should event delivery
102571 "succeed" or should it report undeliverable?
102572
102573 Finally, XKB appears to call WriteToClient to pass events. Sigh.
102574
102575commit 41355a53c29bbf879da0c6ea562294fcc7ef89ff
102576Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102577Date: Tue Aug 28 15:10:20 2007 -0400
102578
102579 xace: add hooks + new access codes: core protocol input requests
102580
102581commit adf46b57ce6c69ab13a38b09a8104c802d54d052
102582Author: Eric Anholt <eric@anholt.net>
102583Date: Tue Aug 28 10:08:38 2007 -0700
102584
102585 Replace BSD custom PCI code with a stub implementation thanks to pciaccess.
102586
102587 Note that pciaccess doesn't yet have Net/OpenBSD support, but the relevant
102588 code should go there instead of disconnected code in the X Server.
102589
102590 While here, remove the now-disabled INCLUDE_XF86_NO_DOMAIN from the headers,
102591 and un-disable xf8StdAccResFromOS for those OSes without domain support which
102592 will need it.
102593
102594commit 4017d3190234e189a0bbd33193a148d4d3c7556b
102595Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102596Date: Tue Aug 28 09:28:25 2007 -0400
102597
102598 devPrivates rework: since API is already broken, switch everything
102599 over to new system.
102600
102601 Need to update documentation and address some remaining vestiges of
102602 old system such as CursorRec structure, fb "offman" structure, and
102603 FontRec privates.
102604
102605commit 85547073265ae9bc4ae3af920a6d3214fd1ca0c5
102606Merge: 860a09c 7d54399
102607Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
102608Date: Tue Aug 28 07:25:21 2007 -0400
102609
102610 Merge branch 'master' into XACE-SELINUX
102611
102612 Conflicts:
102613
102614 include/miscstruct.h
102615 mi/mibstore.c
102616 mi/midispcur.c
102617 os/Makefile.am
102618
102619commit 3fe67d23edaae3ddde20cd5f349aa5dfde1d26a3
102620Author: Eric Anholt <eric@anholt.net>
102621Date: Mon Aug 27 19:02:41 2007 -0700
102622
102623 Remove the BusAccWindows resource code which is now unused.
102624
102625 This was a bunch of poorly defined resource ranges per OS/platform combination
102626 which were supposed to represent what regions could potentially have resources
102627 allocated into them.
102628
102629commit 9e2112b2b56af099a7f380ece9b5c1d25b20cce4
102630Author: Eric Anholt <eric@anholt.net>
102631Date: Mon Aug 27 18:50:34 2007 -0700
102632
102633 Remove unused xf8GetBlock and xf86GetSparse entry points.
102634
102635commit 801c359574d08ff2d6ac75a3325ff522bc862e30
102636Author: Eric Anholt <eric@anholt.net>
102637Date: Mon Aug 27 15:46:05 2007 -0700
102638
102639 Fix fbdevhwstub for pci-rework.
102640
102641commit 7d54399cfdaa7f54e28828267a76b89c4e8e798f
102642Author: Keith Packard <keithp@koto.keithp.com>
102643Date: Sun Aug 26 22:11:45 2007 -0700
102644
102645 Add XSERVER_LIBPCIACCESS to xorg-server.h when using libpciaccess
102646
102647commit affda73a1d6e291516880dfbcb74b661374524c6
102648Author: Eugeniy Meshcheryakov <eugen@univ.kiev.ua>
102649Date: Mon Aug 27 00:41:03 2007 +1000
102650
102651 Xprint: fix handling of TrueType font name
102652
102653 Debian bug #272368
102654 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=272368
102655 In certain locales, some characters from some TrueType fonts were not
102656 appearing in the Xprint postscript output due to the font not being
102657 identified in the postscript output.
102658
102659commit 6f44a2c8a8d6e8f95681ebc7b2dd5ad9c3a02c6e
102660Author: Adam Jackson <ajax@benzedrine.nwnk.net>
102661Date: Sun Jul 29 14:16:10 2007 -0400
102662
102663 Refactor PictureInitIndexedFormats.
102664
102665 The plural version is now static, which is fine since it was only ever called
102666 from within picture post-init anyway. The body of the work is now done with
102667 a one-shot (public) function that operates on a single format at a time.
102668
102669commit ae7f71a8b3d6756161e55d998d6eec37d2695c98
102670Author: Adam Jackson <ajax@benzedrine.nwnk.net>
102671Date: Sat Aug 25 15:08:20 2007 -0400
102672
102673 Implement core protocol backing store exclusively in terms of Composite.
102674
102675 Composite's automatic redirection is a more general mechanism than the
102676 ad-hoc BS machinery, so it's much prettier to implement the one in terms
102677 of the other. Composite now wraps ChangeWindowAttributes and activates
102678 automatic redirection for windows with backing store requested. The old
102679 backing store infrastructure is completely gutted: ABI-visible structures
102680 retain the function pointers, but they never get called, and all the
102681 open-coded conditionals throughout the DIX layer to implement BS are gone.
102682
102683 Note that this is still not a strictly complete implementation of backing
102684 store, since Composite will throw the bits away on unmap and therefore
102685 WhenMapped and Always hints are equivalent.
102686
102687commit bf0883ae5081bd75569115a3eb27c6d3d336c9f2
102688Author: David Nusinow <dnusinow@debian.org>
102689Date: Sat Aug 25 14:53:17 2007 -0400
102690
102691 Fix bug in debugging info related to pci-rework merge
102692
102693commit 4eed88af8b7b6881b44ccf9f4a5c5875af6d2b78
102694Author: Ian Romanick <idr@us.ibm.com>
102695Date: Fri Aug 24 18:06:50 2007 -0700
102696
102697 Remove files made obsolete by pci-rework.
102698
102699commit 1ab4b3e183d04bb20fb5039f7d2671752ab24cf6
102700Author: Aaron Plattner <aplattner@nvidia.com>
102701Date: Fri Aug 24 16:00:31 2007 -0700
102702
102703 Bump video driver ABI for pci-rework.
102704
102705commit 91f358336f77c0e4f577be65cca977d17298e36c
102706Merge: b9a806f 8b6b40b
102707Author: Ian Romanick <idr@us.ibm.com>
102708Date: Fri Aug 24 15:04:21 2007 -0700
102709
102710 Merge branch 'pci-rework'
102711
102712commit b9a806f0b3d495c7616b469281e5892ae7f3f6b3
102713Author: Michel Dänzer <michel@tungstengraphics.com>
102714Date: Fri Aug 24 19:04:55 2007 +0200
102715
102716 exaPolyFillRect: pGC->alu doesn't matter with a single rectangle.
102717
102718commit d0f0d1092c7587a02404e1db07740e6334462ba6
102719Author: Michel Dänzer <michel@tungstengraphics.com>
102720Date: Fri Aug 24 14:10:13 2007 +0200
102721
102722 exaGetImage: Don't migrate pixmap out of FB with no DownloadFromScreen hook.
102723
102724 Based on the assumption that GetImage is relatively rare, so the overhead of
102725 the migration is probably bigger than any potential savings.
102726
102727commit 5d9e2c282145897008d7d941e2a0a3fdc71f2373
102728Author: Michel Dänzer <michel@tungstengraphics.com>
102729Date: Fri Aug 24 14:03:14 2007 +0200
102730
102731 EXA: Improve ShmPutImage.
102732
102733 Share as much code with exaPutImage as possible, and fall back to fbShmPutImage
102734 when that fails.
102735
102736commit 6085522d91e875c0e1ab8d4300e7378701c19b7c
102737Author: Michel Dänzer <michel@tungstengraphics.com>
102738Date: Fri Aug 24 14:02:35 2007 +0200
102739
102740 Export fbShmPutImage to modules.
102741
102742 To be used by EXA.
102743
102744commit c19f227b468d039c5ea136cc8a53c420da30263b
102745Author: Michel Dänzer <michel@tungstengraphics.com>
102746Date: Fri Aug 24 13:05:52 2007 +0200
102747
102748 EXA: Only mark offscreen memory as used when it really is.
102749
102750commit 095850596114178119a8cc854716ce0cc6e05121
102751Author: Michel Dänzer <michel@tungstengraphics.com>
102752Date: Fri Aug 24 13:04:48 2007 +0200
102753
102754 __glXDRIbindTexImage: Fail if no texture bound to pixmap's texture target.
102755
102756 We would most likely crash somewhere in Mesa if we tried to continue in this
102757 case.
102758
102759commit 8b6b40b7271acd81a9548f502c18f46f3b640640
102760Merge: ab7a6d8 3305d17
102761Author: Ian Romanick <idr@us.ibm.com>
102762Date: Thu Aug 23 18:19:17 2007 -0700
102763
102764 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
102765
102766 Conflicts:
102767
102768 hw/xfree86/common/xf86.h
102769 hw/xfree86/common/xf86Init.c
102770 hw/xfree86/common/xf86pciBus.c
102771 hw/xfree86/int10/generic.c
102772 hw/xfree86/int10/helper_exec.c
102773 hw/xfree86/loader/xf86sym.c
102774 hw/xfree86/os-support/bus/Pci.c
102775 hw/xfree86/os-support/bus/Pci.h
102776 hw/xfree86/os-support/bus/linuxPci.c
102777 hw/xfree86/os-support/linux/int10/linux.c
102778
102779commit 93ae6fe18c417a22f1fccb22add4890a20cae713
102780Author: Carl Worth <cworth@cworth.org>
102781Date: Thu Aug 23 16:33:05 2007 -0700
102782
102783 Avoid leaking a Pixmap for every glyph
102784
102785commit 3305d17195e3a0a5555300555bd7703312fa489f
102786Author: Matthieu Herrb <matthieu@deville.herrb.com>
102787Date: Thu Aug 23 22:48:19 2007 +0200
102788
102789 Fix indentation.
102790
102791commit a66c0f1dca2958835ff65a5b50579e3304ed316a
102792Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
102793Date: Thu Aug 23 22:11:56 2007 +0200
102794
102795 Remove an extra cast.
102796
102797 Thou should not apply patches manually without testing.
102798
102799commit 12d27cf33c6d963eae77795c0d247175907162a5
102800Author: Otto Moerbeek <otto@openbsd.org>
102801Date: Thu Aug 23 21:59:25 2007 +0200
102802
102803 A high resolution device that's moving fast can potentially generate
102804 an int overflow, making dx*dx+dy*dy negative. Now pow(negative,
102805 non-integer) yields NaN, so you loose. Use fp math to avoid that.
102806
102807commit ff089e6cae634ac3eb509abd448a250bcbb17275
102808Author: Brian Paul <brian.paul@tungstengraphics.com>
102809Date: Thu Aug 23 19:38:53 2007 +0200
102810
102811 glx: fix crash when freeing visuals
102812
102813 Don't set screen->num_vis to a value greater than the actual number of visuals.
102814
102815 X.Org Bug #10809 <http://bugs.freedesktop.org/show_bug.cgi?id=10809>
102816
102817commit 943dd6ad99670c283a6869ea6c5f751acbd73134
102818Author: Søren Sandmann Pedersen <sandmann@redhat.com>
102819Date: Thu Aug 23 12:15:03 2007 -0400
102820
102821 Revert "Revert "Require pixman 0.9.5; Use pixman_image_set_source_clipping() to fix"" since the pixman changes have been pushed now.
102822
102823 This reverts commit 57f7f2a5327a2d967a726bb4706e4f6b2f4b2cea.
102824
102825commit 13949f997289068354e83bc83e50d97b8232efb1
102826Author: Julien Cristau <jcristau@debian.org>
102827Date: Thu Aug 23 11:27:33 2007 +0200
102828
102829 Get rid of the type1 font module.
102830
102831 It is completely replaced by freetype these days.
102832
102833commit 88a9828ef906bba973debc191e35ea669b7ec271
102834Author: Peter Hutterer <peter@cs.unisa.edu.au>
102835Date: Thu Aug 23 18:02:10 2007 +0930
102836
102837 dix: Only check device events for possible ACLs.
102838
102839 We shouldn't be able to restrict events like Expose, etc. with device based
102840 ACLs. So we just ignore all non-input events when checking for permissions.
102841
102842commit 76bf3cd7b8c6189b6b08518cde00c8bd991bdfb7
102843Author: Dave Airlie <airlied@redhat.com>
102844Date: Thu Aug 23 16:22:03 2007 +1000
102845
102846 randr: fixup crtc and output destroy
102847
102848 if you are moving pointers, you want to move the pointers not just a byte
102849
102850commit 81f8b652d99ee0f7116c1e34aed0e585d23a91fb
102851Author: Alex Deucher <alex@botch2.(none)>
102852Date: Wed Aug 22 19:26:34 2007 -0400
102853
102854 Add _X_EXPORT to exported functions in hw/xfree86/modes/*
102855
102856 Also add missing exports to hw/xfree86/loader/xf86sym.c
102857
102858commit 57f7f2a5327a2d967a726bb4706e4f6b2f4b2cea
102859Author: Eric Anholt <eric@anholt.net>
102860Date: Wed Aug 22 09:02:03 2007 -0700
102861
102862 Revert "Require pixman 0.9.5; Use pixman_image_set_source_clipping() to fix"
102863
102864 The corresponding pixman code hasn't been pushed, so revert until the code is
102865 ready.
102866
102867 This reverts commit 53941c8e68014619d3ded7f8bc0f07d9a38bb9b1.
102868
102869commit d0dc9698ae4324d44ed4c0482d6858d0b73bff33
102870Author: Eric Anholt <eric@anholt.net>
102871Date: Wed Aug 22 09:00:45 2007 -0700
102872
102873 Revert "Fix <pixman.h> include to <pixman/pixman.h>"
102874
102875 The pixman headers have been located under pixman-1/ instead of pixman/ since
102876 around 2007-08-06, and pixman-1.pc has the updated include paths to account
102877 for this.
102878
102879 This reverts commit feb1b3e45513bd6eaa2e6a5ee536183f20d9cb68.
102880
102881commit bc2d516f16d94c805b4dfa8e5b9eef40ff0cbe98
102882Author: Eric Anholt <eric@anholt.net>
102883Date: Fri Aug 17 12:14:16 2007 -0700
102884
102885 Fix overly-restrictive integer overflow check in EXA pixmap creation.
102886
102887 The result was that at 32bpp, pixmaps of width 8192 or greater couldn't be
102888 created, due to treating a pitch value as a width.
102889
102890commit feb1b3e45513bd6eaa2e6a5ee536183f20d9cb68
102891Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
102892Date: Wed Aug 22 16:54:29 2007 +0100
102893
102894 Fix <pixman.h> include to <pixman/pixman.h>
102895
102896commit b6a7c0112c42a3287e53647c38b2c0c5bf8fefa0
102897Merge: 53941c8 6ef4ecd
102898Author: Søren Sandmann Pedersen <sandmann@redhat.com>
102899Date: Tue Aug 21 14:26:34 2007 -0400
102900
102901 Merge branch 'master' of ssh+git://sandmann@git.freedesktop.org/git/xorg/xserver
102902
102903commit 53941c8e68014619d3ded7f8bc0f07d9a38bb9b1
102904Author: Søren Sandmann Pedersen <sandmann@redhat.com>
102905Date: Tue Aug 21 14:26:14 2007 -0400
102906
102907 Require pixman 0.9.5; Use pixman_image_set_source_clipping() to fix
102908 bug 11620 (reported by Jens Stroebel.
102909
102910commit 6ef4ecd82670c37a354243166750d76a97959c8b
102911Author: Julien Cristau <jcristau@debian.org>
102912Date: Tue Aug 21 18:17:35 2007 +0200
102913
102914 config: fix default xkb model (pc105, not keyboard)
102915
102916commit 1834cfb4470341aace64a2fa47d04f85dbf98a47
102917Author: Adam Jackson <ajax@benzedrine.nwnk.net>
102918Date: Tue Aug 21 10:44:37 2007 -0400
102919
102920 Fix an error message to not point to @xfree86.org.
102921
102922commit 265a633cf1fcbf497d6916d9e22403dffdde2e07
102923Author: Keith Packard <keithp@koto.keithp.com>
102924Date: Sun Aug 19 20:29:37 2007 -0700
102925
102926 Screen size changing should leave FB alone when X is inactive.
102927
102928 xf86RandR12ScreenSetSize must protect calls to EnableDisableFBAccess with
102929 suitable vtSema checks to avoid invoking driver code while the X server is
102930 inactive.
102931
102932commit 7dc8531548cc9573e28bb04363dcbb3af5864c9a
102933Author: Keith Packard <keithp@koto.keithp.com>
102934Date: Sun Aug 19 20:28:05 2007 -0700
102935
102936 Ref count cursors used in hw/xfree86/modes code.
102937
102938 The multi-crtc cursor code in hw/xfree86/modes holds a reference to the
102939 current cursor. This reference must be correctly ref counted so the cursor
102940 is not freed out from underneath this code.
102941
102942commit 1f6ddae003ec65d6bc567831bf32bf75dfefdd6c
102943Author: Alex Deucher <alex@botch2.com>
102944Date: Tue Aug 21 00:37:33 2007 -0400
102945
102946 add xf86_crtc_clip_video_helper to xf86sym.c
102947
102948commit c839859d1bc35451923a2cbd5dfac4f3ca5eb3f9
102949Author: David Nusinow <dnusinow@debian.org>
102950Date: Mon Aug 20 21:09:27 2007 -0400
102951
102952 Move module defaults from the header to the source file.
102953
102954 This is where they should have been in the first place. All the rest of
102955 the code in the server defines such things in the source files, not the
102956 headers.
102957
102958commit 53c04351c462d2ae307684e50d5960debe1ee557
102959Author: Alex Deucher <alex@botch2.com>
102960Date: Mon Aug 20 19:46:38 2007 -0400
102961
102962 move intel crtc xv clipping helper to the xserver
102963
102964 The code is generic and can be used by any overlay-based card when
102965 adding randr 1.2 support. Tested on radeon.
102966
102967commit a1fe36b772f7edc162ea97368f86588c0fb77148
102968Author: Julien Cristau <jcristau@debian.org>
102969Date: Mon Aug 20 12:57:06 2007 +0200
102970
102971 xfree86: Fix build on Linux/alpha.
102972
102973 A bunch of CFLAGS had gone missing, so the build failed with errors like:
102974 ../../../../../hw/xfree86/os-support/linux/lnx_ev56.c:7:19: error: input.h: No such file or directory
102975 ../../../../../hw/xfree86/os-support/linux/lnx_ev56.c:8:24: error: scrnintstr.h: No such file or directory
102976
102977commit 2c1431a76e7219e3bd14fd7f7888a8bc4fea0f58
102978Author: Peter Hutterer <peter@cs.unisa.edu.au>
102979Date: Mon Aug 20 10:06:13 2007 +0930
102980
102981 dix: ProcChangeActivePointerGrab: make sure variable is initialised.
102982
102983 Thanks to Ben Close for spotting it.
102984
102985commit 65a49f0ca198e0366175367729a101211388b16b
102986Author: Blair Sadewitz <blair.sadewitz@gmail.com>
102987Date: Sun Aug 19 20:29:22 2007 +0200
102988
102989 Autoconfiguration of wsmouse for NetBSD.
102990
102991commit 3c448b0eb67337b56641e09a6d168aad6745e3ef
102992Author: Fredrik Höglund <fredrik@kde.org>
102993Date: Sat Aug 18 19:02:18 2007 +0200
102994
102995 EXA: Fix a couple of logic errors in exaGetPixmapFirstPixel.
102996
102997 The fb pointer would be left uninitialized when exaPixmapIsOffscreen
102998 returned false. When it returned true and the pixmap was damaged,
102999 fb would be initialized from the pixmap's devPrivate.ptr before the
103000 exaDoMigration and exaPrepareAccess calls, at which point
103001 devPrivate.ptr would still be pointing at offscreen memory.
103002
103003commit 23fbd5292d356067e85e1eec4eb4f743532b0503
103004Author: Alan Coopersmith <alan.coopersmith@sun.com>
103005Date: Fri Aug 17 15:29:16 2007 -0700
103006
103007 Actually build Secure RPC authentication support (missed in modularization)
103008
103009commit 6a32a96d8df184c3ace4847beb48fdcb846d2286
103010Author: Aaron Plattner <aplattner@nvidia.com>
103011Date: Thu Aug 16 17:43:29 2007 -0700
103012
103013 stride is in FbBits-sized chunks, but xoff is not.
103014
103015 Fixes corruption problems with composite rendering to redirected windows in
103016 depth 16.
103017
103018commit 32666d77227fcd2c066de16bf3c07366f92b0457
103019Author: Aaron Plattner <aplattner@nvidia.com>
103020Date: Thu Aug 16 14:57:18 2007 -0700
103021
103022 Bug #12015: Use the right offsets in the dst arguments of pixman_blt.
103023
103024commit daee59b1703ac07c2def9e9fecc479e59b93f761
103025Author: Fredrik Höglund <fredrik@kde.org>
103026Date: Wed Aug 15 19:19:11 2007 +0200
103027
103028 EXA: Wrap Trapezoids to prevent excessive migration of the alpha pixmap.
103029
103030 miTrapezoids creates an alpha pixmap and initializes the contents
103031 using PolyFillRect, which causes the pixmap to be moved in for
103032 acceleration. The subsequent call to RasterizeTrapezoid won't be
103033 accelerated by EXA, which causing the pixmap to be moved back out
103034 again.
103035
103036 By wrapping Trapezoids and using ExaCheckPolyFillRect instead of
103037 PolyFillRect to initialize the pixmap, we avoid this roundtrip.
103038
103039commit 860a09cfb8afc0a293c7eb5e01762724eb86847a
103040Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103041Date: Thu Aug 16 16:10:44 2007 -0400
103042
103043 devPrivates rework: Nevermind, can't const due to return value warnings.
103044
103045 This reverts commit 6fd0a0b08de912421718aca17fe34a55ae285ae7.
103046
103047commit 6fd0a0b08de912421718aca17fe34a55ae285ae7
103048Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103049Date: Thu Aug 16 16:01:42 2007 -0400
103050
103051 devPrivates rework: add const qualifier to key type.
103052
103053commit b2b7817497dd5da73d23ec9cc637c563041fc490
103054Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103055Date: Thu Aug 16 15:30:25 2007 -0400
103056
103057 devPrivates rework: use camelcase standard for name of key type.
103058
103059commit 0a994d4f859a4e48d41a90ed9d2a282bb528c555
103060Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103061Date: Thu Aug 16 12:54:35 2007 -0400
103062
103063 xace: add hooks + new access codes: core protocol selection requests
103064
103065commit 3ef2e9e623819c625a92f464fb14f1e5c181df42
103066Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103067Date: Thu Aug 16 12:27:48 2007 -0400
103068
103069 xace: add hooks + new access codes: core protocol pixmap requests
103070
103071commit be536b79f2a364399937314cfa6c88bf8188da9c
103072Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
103073Date: Thu Aug 16 11:23:28 2007 -0500
103074
103075 Update for support on NetBSD and DragonFly. From
103076 Joerg Sonnenberger and pkgsrc.
103077
103078commit 1d4bea6106d7a1c83e1dfe37fad8268589feaa0b
103079Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
103080Date: Thu Aug 16 11:20:12 2007 -0500
103081
103082 Add some more support for DragonFly. From Joerg Sonnenberger
103083 and pkgsrc.
103084
103085commit fe9bc481efb0821134e10760c23993c6a7386450
103086Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103087Date: Thu Aug 16 12:02:59 2007 -0400
103088
103089 xace: add hooks + new access codes: core protocol font requests
103090
103091commit e89301c8790df9fc49de13dd7c7f36e5340c0c31
103092Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103093Date: Thu Aug 16 10:57:49 2007 -0400
103094
103095 xace: add hooks + new access codes: core protocol client requests
103096
103097commit 5bee8db003a5d552ee1d85bb6c40a3cb93bd6b2b
103098Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103099Date: Thu Aug 16 10:44:51 2007 -0400
103100
103101 xace: drop background-none checking hook, add new hook for controlling
103102 access to other clients.
103103
103104commit b82557c9fb60f11fd2696c8fb2ae17b9dfd915ed
103105Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103106Date: Thu Aug 16 10:36:05 2007 -0400
103107
103108 xace: add hooks + new access codes: core protocol screensaver requests
103109
103110commit 568ae737d1d5d476a0bf85659d88910c4e0ef5e0
103111Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103112Date: Wed Aug 15 14:14:45 2007 -0400
103113
103114 xace: add hooks + new access codes: core protocol server requests
103115
103116commit 14c13b8d62eb37cba8a044daffcddec578ba1644
103117Author: Dave Jones <davej@redhat.com>
103118Date: Thu Aug 16 09:46:27 2007 +0200
103119
103120 Kdrive: fix nasty thinko in TslibEnable()
103121
103122commit 026534f945ae5652592a090a9d41375ca37ab618
103123Author: Alan Coopersmith <alan.coopersmith@sun.com>
103124Date: Wed Aug 15 16:47:53 2007 -0700
103125
103126 Update pci.ids to 2007-08-15 snapshot
103127
103128 Remove nvidia ids in extrapci.ids that are now in pci.ids
103129
103130commit 6cef7b9611297cb1d93cefe3890b26b69c87bce2
103131Author: Alan Coopersmith <alan.coopersmith@sun.com>
103132Date: Wed Aug 15 16:44:49 2007 -0700
103133
103134 Correct XErrorDB path and make it configurable (used by DTrace support)
103135
103136commit 3c9553ac2cac7f3a41966def44a50d722d7e645b
103137Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103138Date: Wed Aug 15 14:14:25 2007 -0400
103139
103140 xace: rename hostlist security hook to "server" as this hook will be used
103141 for other types of server access besides just the host list.
103142
103143commit dc84bb3418933297a8c005070902d9a91ed3d18f
103144Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103145Date: Wed Aug 15 14:13:53 2007 -0400
103146
103147 xace: add hooks + new access codes: core protocol cursor requests
103148
103149commit b424e01ec59d9600a02823f1522949325797268c
103150Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103151Date: Tue Aug 14 13:20:42 2007 -0400
103152
103153 xace: add hooks + new access codes: core protocol property requests
103154
103155commit 14d0397cded699378fa3c19f4e61dbab7d3a9b2c
103156Author: Peter Hutterer <peter@cs.unisa.edu.au>
103157Date: Wed Aug 15 22:12:29 2007 +0930
103158
103159 dix: ProcUngrabKeyboard: make sure grab is initialized.
103160
103161 This should fix the random segfaults with ProcUngrabKeyboard.
103162
103163 Thanks to David Nolden for spotting it.
103164
103165commit 0f9e89b4e309e570d7d366489d250ca2143f0ad7
103166Author: Fredrik Höglund <fredrik@kde.org>
103167Date: Tue Aug 14 22:47:49 2007 +0200
103168
103169 Fix the value comparisons in the IDLETIME wakeup handler.
103170
103171 LessThan/GreaterThan comparisons were used in the wakeup handler,
103172 and LessOrEqual/GreaterOrEqual in the block handler.
103173
103174 Change it to use LessOrEqual/GreaterOrEqual in both functions,
103175 since this is what XSyncNegativeComparison and
103176 XSyncPositiveComparison imply.
103177
103178commit 42d6112ec21949a336ee8b34469f2695273ee2d6
103179Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103180Date: Tue Aug 14 13:09:38 2007 -0400
103181
103182 xace: add hooks + new access codes: core protocol GC requests
103183
103184commit 9a183d7ba50e31afa133cc03aee7991517a283ea
103185Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103186Date: Tue Aug 14 11:39:26 2007 -0400
103187
103188 dix: remove caching of drawables and graphics contexts. The security checks
103189 simply bypass the cached values so they are unused.
103190
103191commit 2763056ab5ae31bed422a0948198d98c6ace6d55
103192Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103193Date: Mon Aug 13 13:40:47 2007 -0400
103194
103195 xace: add hooks + new access codes: core protocol window requests
103196
103197commit 6a195e816b9d60f728d77cc1c23538e7af00a879
103198Author: Kristian Høgsberg <krh@redhat.com>
103199Date: Mon Aug 13 10:43:48 2007 -0400
103200
103201 Revert "Implement damage tracking for AIGLX."
103202
103203 This reverts commit 2243b30e54df07892f75e3d65b687abe5b183cf3. The existing
103204 DRI interface doesn't let us get from a __DRIdrawable to the corresponding
103205 X drawable, and thus, we can't implement AIGLX damage tracking with the
103206 current interface.
103207
103208commit f367285fd5825e0adc271982a529c9904ad65c89
103209Merge: b1272ee ff4bd3a
103210Author: Peter Hutterer <peter@cs.unisa.edu.au>
103211Date: Sun Aug 12 15:31:10 2007 +0930
103212
103213 Merge branch 'master' into mpx
103214
103215 Conflicts:
103216
103217 Xi/exevents.c
103218 dix/devices.c
103219 dix/getevents.c
103220 include/dix.h
103221 mi/mieq.c
103222
103223commit 03f9da672466b9ab9a9814d784b8c44f1030587e
103224Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
103225Date: Sun Aug 12 03:07:04 2007 +0200
103226
103227 xfree86: Improve default mouse handling on the Hurd
103228
103229 Make /dev/mouse the default device. This makes Xorg works with empty
103230 or missing InputDevice sections.
103231
103232commit c5741438a3a171f493e9da32a6b39f73403f6993
103233Author: Alan Coopersmith <alan.coopersmith@sun.com>
103234Date: Fri Aug 10 16:13:55 2007 -0700
103235
103236 Only use evdev drivers in Xephyr #ifdef linux
103237
103238commit 59961e47df4ea621a6713a8c7d060555f8746c3a
103239Author: Alan Coopersmith <alan.coopersmith@sun.com>
103240Date: Fri Aug 10 16:08:41 2007 -0700
103241
103242 xorgcfg needs PIXMAN_CFLAGS in order to build libc_wrapper.c
103243
103244commit 2243b30e54df07892f75e3d65b687abe5b183cf3
103245Author: Kristian Høgsberg <krh@redhat.com>
103246Date: Fri Aug 10 15:53:05 2007 -0400
103247
103248 Implement damage tracking for AIGLX.
103249
103250commit ff4bd3addb48df3eacc4b121cc249a7f38eb981a
103251Author: Eric Anholt <eric@anholt.net>
103252Date: Wed Aug 8 14:24:42 2007 -0700
103253
103254 Fix the swapped decode of the EDID DTD h/v sync polarity fields.
103255
103256 As a result, we can remove the quirks that existed to flip the bits back around
103257 for us. This is not confirmed in all cases due to lack of bugs containing EDID
103258 blocks associated with the quirks, but is likely true.
103259
103260commit 2926cf1da7e4ed63573bfaecdd7e19beb3057d9b
103261Author: Gustavo Pichorim Boiko <boiko@mandriva.com>
103262Date: Thu Aug 2 18:09:52 2007 -0300
103263
103264 [PATCH] Allocate the right number of entries for saving crtcs
103265
103266commit b2dcfbca2441ca8c561f86a78a76ab59ecbb40e4
103267Author: Keith Packard <keithp@koto.keithp.com>
103268Date: Wed Aug 8 12:16:12 2007 -0700
103269
103270 RRScanOldConfig cannot use RRFirstOutput before output is configured.
103271
103272 RRFirstOutput returns the first active output, which won't be set until
103273 after RRScanOldConfig is finished running. Instead, just use the first
103274 output (which is the only output present with an old driver, after all).
103275
103276commit b1272eefd9a3e340d65c14903f337747ec82d021
103277Author: Peter Hutterer <peter@cs.unisa.edu.au>
103278Date: Wed Aug 8 15:00:02 2007 +0930
103279
103280 Fix typo from last commit. Oh well.
103281
103282commit c02128532e910e813fba94983733942d30c2d5cb
103283Author: Peter Hutterer <peter@cs.unisa.edu.au>
103284Date: Wed Aug 8 13:10:00 2007 +0930
103285
103286 dix: Allow flexible devices for passive core grabs.
103287
103288 A passive core grab doesn't specify the device, and is thus created with the
103289 ClientPointer as device. When this grab is activated later, don't actually
103290 activate the grab on the grab device, but rather change the device to the one
103291 that caused the grab to activate. Same procedure for keyboards.
103292
103293 Makes core apps _A LOT_ more useable and reduces the need to set the
103294 ClientPointer.
103295
103296 Only applies to core grabs!
103297
103298commit ab3f601149e15789edfb7c9a0c33387070279582
103299Author: Tiago Vignatti <tiagov@balalaika.(none)>
103300Date: Tue Aug 7 23:17:32 2007 -0300
103301
103302 Updates some piece of the dead mouse evdev code under the new hotplug scheme.
103303 I exported the evdev driver to Xephyr server. I'm running it using something
103304 like:
103305 $ ./hw/kdrive/ephyr/Xephyr :1 -mouse evdev,,device=/dev/input/event4 -keybd \
103306 evdev,,device=/dev/input/event1,xkbmodel=abnt2,xkblayout=br
103307
103308 It also closes /#5668.
103309
103310commit 7d1a749b210ba5b9f8d0e5a1feb9a9ef9fa4d992
103311Author: Tiago Vignatti <tiagov@balalaika.(none)>
103312Date: Tue Aug 7 22:59:12 2007 -0300
103313
103314 Export device path key options to be called by the command line of server.
103315
103316commit aee3588a4a6829326770c84b860061f47f2cbcae
103317Author: Tiago Vignatti <tiagov@balalaika.(none)>
103318Date: Tue Aug 7 22:49:07 2007 -0300
103319
103320 Update KdUseMsg() for completeness.
103321
103322commit 30259d5a4e95ff20b30807e5e207ab5995a3fdaf
103323Author: Daniel Stone <daniel@fooishbar.org>
103324Date: Tue Aug 7 20:58:49 2007 +0300
103325
103326 Hotplug: HAL: Fix error handling
103327
103328 Don't use our DBusError for property getting, because we simply don't care:
103329 this fixes D-Bus error spew to stderr. Thanks Michel Dänzer for debugging
103330 and testing.
103331
103332commit aef255425a3521d66c3405d34f7787628a22703e
103333Author: Daniel Stone <daniel@fooishbar.org>
103334Date: Tue Aug 7 16:37:42 2007 +0300
103335
103336 Config: HAL: Use input.xkb namespace
103337
103338 Use an explicit input.xkb.foo namespace, not input.xkb_foo.
103339
103340commit 838e59c02ec06446fc180fb9d86fa8793c7b9903
103341Author: Daniel Stone <daniel@fooishbar.org>
103342Date: Mon Aug 6 16:07:20 2007 +0300
103343
103344 configure.ac: Add $CONFIG_LIB to server libraries
103345
103346 Make sure all DDXes get $CONFIG_LIB. Build-tested with Xvfb and Xdmx.
103347
103348commit b4193a2eee80895c5641e77488df0e72a73a3d99
103349Author: Keith Packard <keithp@koto.keithp.com>
103350Date: Tue Aug 7 12:45:53 2007 -0700
103351
103352 RRScanOldConfig wasn't getting crtcs set correctly
103353
103354 The output crtc is set by RRCrtcNotify, which is called at the end of
103355 RRScanOldConfig. Several uses of output->crtc in this function were wrong.
103356
103357commit 2b93cbb5f8bac9b1b75f723baaa728430b5fefff
103358Author: Keith Packard <keithp@koto.keithp.com>
103359Date: Tue Aug 7 12:44:19 2007 -0700
103360
103361 Decrement mode count when removing RandR output mode.
103362
103363 Removing an output mode without decrementing the mode count scrambles the
103364 output mode array badly.
103365
103366commit fef4c7a6f1a1ef34233b36137bb66d9a657307fb
103367Author: Eric Anholt <eric@anholt.net>
103368Date: Tue Aug 7 09:01:14 2007 -0700
103369
103370 Fix driver build after pixman changes.
103371
103372commit 1339e57485db5a285cfbecbe0bba7154458680ad
103373Author: Tiago Vignatti <tiagov@balalaika.(none)>
103374Date: Tue Aug 7 04:24:34 2007 -0300
103375
103376 Fix typo.
103377
103378commit d9ee5f3e3a3a814ebcd257736c305b41139cc354
103379Author: Tiago Vignatti <tiagov@balalaika.(none)>
103380Date: Tue Aug 7 04:22:26 2007 -0300
103381
103382 Clean a little bit the code.
103383
103384commit 7a5eb3e96b74daaaeb6babf46b13d698280aa3f6
103385Author: Tiago Vignatti <tiagov@balalaika.(none)>
103386Date: Tue Aug 7 02:16:44 2007 -0300
103387
103388 Let xkb options be passed through command line in kdrive servers. I start my
103389 Xephyr using something like:
103390
103391 ./hw/kdrive/ephyr/Xephyr :1 -fp /usr/share/fonts/X11/misc/ -mouse ephyr -keybd ephyr,,xkblayout=br,xkbmodel=abnt2
103392
103393commit 955d5f6c0d14fae63bfe7c4ab39ee0a708919479
103394Author: Tiago Vignatti <tiagov@balalaika.(none)>
103395Date: Tue Aug 7 01:39:29 2007 -0300
103396
103397 When we call Xephyr with '-pointer' a new pointer is added inside the server
103398 and the Xephyr virtual mouse keeps alive. With this patch the semantic changes
103399 turning '-pointer' && 'Xephyr virtual mouse' always false.
103400
103401 Now we can open a device pointer and pass its options in Xephyr's command line
103402 without having other pointer unused.
103403
103404commit b8abeaf74ee8296d4bc3164a5a253624f984a6d4
103405Author: Peter Hutterer <peter@cs.unisa.edu.au>
103406Date: Tue Aug 7 12:32:46 2007 +0930
103407
103408 dix: get the paired keyboard for a passive grab (ProcGrabButton).
103409
103410 Taking the VCK is only correct if no physical device is connected, and even
103411 then it's not really a good idea.
103412
103413commit a0b87f87fb8753955505958bf3d438eef191302d
103414Author: Peter Hutterer <peter@cs.unisa.edu.au>
103415Date: Tue Aug 7 10:49:33 2007 +0930
103416
103417 dix: check for core event to determine if grab is a core grab (CreateGrab).
103418
103419 Checking for VCP/VCK is simply not a safe way to check if a grab is a core grab.
103420
103421commit aa3c6aaaab213200591d29ddb2921adfb87ee5b4
103422Author: Søren Sandmann Pedersen <sandmann@redhat.com>
103423Date: Mon Aug 6 19:00:59 2007 -0400
103424
103425 Require pixman-1 0.9.4, update pixman includes to new scheme
103426
103427commit 74feba4d77d74979a0ea478d666439ffc55001e5
103428Author: Aaron Plattner <aplattner@nvidia.com>
103429Date: Wed Aug 1 14:30:03 2007 -0700
103430
103431 Don't unwrap too early in libwfb for Composite.
103432
103433 Don't call fbFinishWrap until the pixman_image_t that stores the pointer is
103434 actually freed. This prevents corruption or crashes caused by accessing a
103435 wrapped pointer after the wrapping is torn down.
103436
103437commit f6aa2200f2fb4f4d4bb51e67d68e86aabcac0c4b
103438Author: Roland "Test-tools" Bär <roland@verifysoft.de>
103439Date: Mon Aug 6 12:37:52 2007 -0700
103440
103441 Probable off by one buffer overflow in .../xorgconfig/xorgconfig.c
103442
103443 X.Org Bug #11858 <http://bugs.freedesktop.org/show_bug.cgi?id=11858>
103444 Patch #11005 <http://bugs.freedesktop.org/attachment.cgi?id=11005>
103445
103446commit d744df32a15103aa14237175f506350d25b2fec0
103447Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103448Date: Mon Aug 6 12:23:21 2007 -0400
103449
103450 xace: add hooks + new access codes: core protocol colormap requests
103451
103452commit acc9a42c926a3f84159780de12ecc1dc6186068a
103453Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103454Date: Mon Aug 6 12:16:59 2007 -0400
103455
103456 Temporarily disable Security and SELinux extensions
103457 while changes to XACE are being made.
103458
103459commit 102df4f9bac59d95963572d1a7f31d1a064ca4ca
103460Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103461Date: Mon Aug 6 09:16:30 2007 -0400
103462
103463 xace: drop site-policy and declare-extension-security hooks, add 2 new hooks
103464 for controlling access to screens and screen savers.
103465
103466commit 9eddede039f6cbcc323b7e3e4e841c43d3ed4f43
103467Author: Peter Hutterer <peter@cs.unisa.edu.au>
103468Date: Mon Jul 16 18:01:00 2007 +0930
103469
103470 mieqEnqueue: Don't try to update the time for GenericEvents.
103471
103472 Doing so may overwrite the event's length field and cause havoc. Also check if
103473 realloc'd memory did actually return valid pointer.
103474
103475commit e717eb82dc2e55f852919312d04f5cfc8ee55bc8
103476Author: Dave Airlie <airlied@redhat.com>
103477Date: Thu Aug 2 10:50:01 2007 +1000
103478
103479 xserver: stop bcopy from going really slow
103480
103481 The outport is most likely unnecessary on any currently used hardware,
103482 the byte copy is necessary from what I know on IA64 and friends so leave it.
103483
103484 Add a new API entry point which lets a driver select the old behaviour if
103485 such a needs is ever found.
103486
103487 This gives me ~20% speed up on startup on 945 hardware.
103488
103489commit 600ef07113caa7a901c7d486bc8ebd1ae47f885c
103490Author: Tiago Vignatti <tiagov@balalaika.(none)>
103491Date: Fri Aug 3 15:33:41 2007 -0300
103492
103493 Fix kdrive command line parser.
103494
103495commit 375864cb74cced40ae688078b1f7750998972535
103496Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103497Date: Fri Aug 3 13:23:34 2007 -0400
103498
103499 security: drop support for XC-QUERY-SECURITY authorization method.
103500
103501commit d445d2f22b5c97fa010370f4ba9cb0555df4a853
103502Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103503Date: Fri Aug 3 10:56:18 2007 -0400
103504
103505 security: drop the "declare extension security" dix call. Use the
103506 SecurityPolicy configuration file instead.
103507
103508commit 0a71e1542a07abc5e32501973a7cf6de3f641317
103509Author: Carl Worth <cworth@cworth.org>
103510Date: Thu Aug 2 22:48:32 2007 -0700
103511
103512 Create a Picture as well as a Pixmap at the time of AllocateGlyph
103513
103514 This avoids some inefficiency in creating a temporary Picture
103515 for every glyph at rendering time. My measurements with an i965
103516 showed the previous patch causing a 10-15% slowdown for NoAccel
103517 and XAA cases, (while providing an 18% speedup for EXA).
103518
103519 With this change, the NoAccel and XAA performance regression is
103520 eliminated, and the overall EXA speedup, (before any of the
103521 glyphs-as-pixmaps work), is now 32%.
103522
103523commit a2af34d5a861982a03afad8e586bb0181b72bbd0
103524Author: Carl Worth <cworth@cworth.org>
103525Date: Wed Aug 1 15:48:30 2007 -0700
103526
103527 Use per-screen Pixmaps for glyphs
103528
103529 Instead of system-memory data which prevents accelerated
103530 compositing of glyphs, (at least without forcing an upload
103531 of the glyph data before compositing).
103532
103533commit 19b3b1fd8feb343a690331cafe88ef10b34b9d98
103534Author: Carl Worth <cworth@cworth.org>
103535Date: Tue Jul 31 17:04:13 2007 -0700
103536
103537 Use strong hash (SHA1) for glyphs
103538
103539 Using a cryptographically strong hash means that comparing the
103540 hash alone is sufficient for determining glyph equality (no need
103541 to compare the glyph bits directly). This will allow us to replace
103542 system-memory copies of the glyph bits, (which we've only been
103543 holding onto for comparisons), with Pixmaps.
103544
103545commit 516b96387b0e57b524a37a96da22dbeeeb041712
103546Author: Carl Worth <cworth@cworth.org>
103547Date: Mon Jul 30 17:31:47 2007 -0700
103548
103549 ProcRenderAddGlyphs: Avoid allocating a glyph just to find it cached
103550
103551 This is a cleanup without any real savings (yet). Previously, the
103552 implementation would allocate a new glyph, then (often) find it in
103553 the cache, and immediately discard the allocated object. This
103554 re-organization first uses a new FindGlyphByHash function and only
103555 allocates the glyph if nothing is found.
103556
103557 This isn't a real savings yet, since FindGlyphByHash currently still
103558 does a temporary glyph allocation, but this is expected to be replaced
103559 immediately as we switch to an alternate hashing mechanism (SHA1).
103560
103561commit 4c6abe1c7c8abcf203572bbf86b21d97ea4e756f
103562Author: Carl Worth <cworth@cworth.org>
103563Date: Mon Jul 30 21:43:20 2007 -0700
103564
103565 Split HashGlyph functionality out into HashGlyphInfoAndBits
103566
103567 This is in preparation for a future change that will take advantage
103568 of being able to compute a hash for a separate xGlyphInfo and chunk
103569 of bits without a combined Glyph object.
103570
103571commit 363d764ea32b938f3dff35df7cf3370363c04d5c
103572Author: Carl Worth <cworth@cworth.org>
103573Date: Mon Jul 30 15:10:11 2007 -0700
103574
103575 ProcRenderAddGlyphs: Take advantage of the for loops to simplify the code a bit
103576
103577commit dc8a528cd6b9a4da3e60fa31428c37f5b34a897f
103578Author: Carl Worth <cworth@cworth.org>
103579Date: Wed Jul 25 14:57:13 2007 -0700
103580
103581 ProcRenderAddGlyphs: Convert while loops to for loops where more natural
103582
103583commit e34fcd2bf42dbd72ab6ce2df80f2dcaa13416e74
103584Merge: 32c0dcc f3955c0
103585Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
103586Date: Thu Aug 2 14:27:03 2007 -0400
103587
103588 Merge branch 'master' into XACE-SELINUX
103589
103590 Conflicts:
103591
103592 dix/devices.c
103593 dix/property.c
103594 include/dix.h
103595
103596commit f3955c0a020b39021050cd33c20a17f14fc4b579
103597Author: Arkadiusz Miskiewicz <arekm@maven.pl>
103598Date: Wed Aug 1 21:04:22 2007 +0300
103599
103600 XFree86: xf1bpp: distclean generated files as well
103601
103602 Make sure we clean up after ourselves: not sure why distcheck didn't flag
103603 this one.
103604
103605commit a04c95f4446e5c169dea71019321d790ab4fa139
103606Author: Julien Cristau <jcristau@debian.org>
103607Date: Wed Aug 1 20:37:05 2007 +0300
103608
103609 configure.ac: Fix argument quoting for argv[]
103610
103611 m4 quoting. Yar boo sux.
103612
103613commit 99a88826e5e8cfa25c5f8a88c12799d33114729c
103614Author: Daniel Stone <daniel@fooishbar.org>
103615Date: Wed Aug 1 20:34:58 2007 +0300
103616
103617 configure.ac: Actually use -lrt in monotonic clock test
103618
103619 If we need -lrt to use clock_gettime, then make sure we link with it.
103620
103621commit 1c80e04f876e9254b93ef87eadfcff71234340c6
103622Author: Daniel Stone <daniel@fooishbar.org>
103623Date: Wed Aug 1 20:08:31 2007 +0300
103624
103625 configure.ac: Disable D-Bus config API support by default
103626
103627 This is problematic, so don't even bother with it unless someone wants it.
103628 respeclaration is dead, long live HAL.
103629
103630commit c46663367329615bd2c9b63e93c9534036e5a2ae
103631Author: Michel Dänzer <michel@tungstengraphics.com>
103632Date: Wed Aug 1 18:32:09 2007 +0200
103633
103634 GLX/DRI: Remove some unused variables.
103635
103636commit 17cb4f64e3c39725e83b1e311c09422d7e1c0e52
103637Author: Michel Dänzer <michel@tungstengraphics.com>
103638Date: Wed Aug 1 18:13:18 2007 +0200
103639
103640 GLX_EXT_texture_from_pixmap: Use client provided texture target when available.
103641
103642 This prevents situations where the server doesn't use the target the
103643 client thinks it does, usually resulting in the texture being sampled as all
103644 white.
103645
103646commit a4197db9504adae6af005b2218eee36b8af0d98b
103647Author: Daniel Stone <daniel@fooishbar.org>
103648Date: Wed Aug 1 14:04:51 2007 +0300
103649
103650 GL: GLX: Make sure glxbyteorder.h is distributed
103651
103652commit ad7421fc764e2b82e20d90f12225a03a1d636f18
103653Author: Daniel Stone <daniel@fooishbar.org>
103654Date: Wed Aug 1 08:30:00 2007 +0300
103655
103656 Bump version to 1.3.99.1 for development
103657
103658 This is not actually .1, just bumping for a different devel version.
103659
103660commit 43e71a54502d9ab28ece7f6296d1416d60948dad
103661Author: Daniel Stone <daniel@fooishbar.org>
103662Date: Wed Aug 1 08:16:35 2007 +0300
103663
103664 XFree86: xf1bpp: Fix previous build system commit
103665
103666 Amateur error.
103667
103668commit 6d6bc93b0a13c5356544561e326d4aedf33e61c2
103669Author: Daniel Stone <daniel@fooishbar.org>
103670Date: Wed Aug 1 08:11:22 2007 +0300
103671
103672 Build system: Add missing files
103673
103674 A couple of headers weren't added to the build.
103675
103676commit 505ec436af3a173e0ba32c6f14b4cf9837a553eb
103677Author: Daniel Stone <daniel@fooishbar.org>
103678Date: Wed Aug 1 08:11:08 2007 +0300
103679
103680 XFree86: Properly clean up after ourselves
103681
103682 CLEAN is not a useful variable. CLEANFILES/DISTCLEANFILES, on the other hand,
103683 are useful variables.
103684
103685commit 1ace9770fed4a2ba354ff06a96189428beb36088
103686Author: Daniel Stone <daniel@fooishbar.org>
103687Date: Wed Aug 1 08:10:38 2007 +0300
103688
103689 Build system: Non-dtrace distcheck hacks
103690
103691 automake 1.10 really wants foo.c for foo.O, so give it some dummy files to
103692 deal with if it really needs them.
103693
103694commit cacbdf18ee771d43228c2e96e8ef9a32251ceb55
103695Author: Daniel Drake <d.drake@mmm.com>
103696Date: Wed Aug 1 08:08:37 2007 +0300
103697
103698 Remove duplicated licenses
103699
103700 Some files had two copies of the same license.
103701
103702commit bd49332e4772bd57ffb76c829f0e4770ab876057
103703Author: Daniel Drake <d.drake@mmm.com>
103704Date: Wed Aug 1 08:07:08 2007 +0300
103705
103706 Add proper COPYING file
103707
103708 I went through the entire xorg-server distribution and aggregated all
103709 the licenses I could find (except the questionable GPL files, see my
103710 last mail).
103711
103712 There are many many permutations on essentially the same license terms,
103713 but I have been pedantic and treated slight differences as separate
103714 licenses.
103715
103716 Here is a description of the process I used:
103717
103718 tar xvjf /usr/portage/distfiles/xorg-server-1.1.1.tar.bz2
103719
103720 cd xorg-server-1.1.1
103721 find -name '*.c' -o -name '*.h' | xargs gvim
103722
103723 egrep -Rli "permission|copyright" * | grep -v "\.[ch]" \
103724 | grep -v "\.in$" | xargs gvim
103725
103726 cd ..
103727 tar xvjf /usr/portage/distfiles/xorg-server-1.3.0.0.tar.bz2
103728 diff -urNp xorg-server-1.1.1 xorg-server-1.3.0.0
103729
103730 git clone git://anongit.freedesktop.org/git/xorg/xserver
103731 cd xserver
103732 git diff xorg-server-1.3.0.0..
103733
103734 For each file, licenses have been aggregated as follows:
103735
103736 If 2 files have identical license text but different copyright notices,
103737 the copyright notices are aggregated and the license text
103738 is included only once.
103739
103740 Note that by identical I mean really identical, i.e.:
103741 'AUTHOR(S)' is not the same as 'AUTHORS'
103742 'KEITH PACKARD DISCLAIMS' is not the same as 'KEITH PACKARD AND COMPAQ
103743 DISCLAIM'
103744
103745 Otherwise, licenses and accompanying copyright notices have been
103746 stacked.
103747
103748 When going through the changes from 1.1.1 to 1.3.0.0 then HEAD, licenses
103749 have been added and removed (so I have reflected this since the original
103750 version of my COPYING file). It's slightly concerning to see that even
103751 between 1.3.0.0 and HEAD, new license permutations are being added. I'd
103752 suggest that a primary license be chosen and this would be indicated at
103753 the top of this COPYING file.
103754
103755commit 51b735394f0aa9f953f9c320617c7a56028ec458
103756Author: Daniel Drake <dsd@gentoo.org>
103757Date: Mon Apr 30 11:37:46 2007 -0400
103758
103759 [PATCH] xserver: Add COPYING terms
103760
103761 I went through the entire xorg-server distribution and aggregated all
103762 the licenses I could find (except the questionable GPL files, see my
103763 last mail).
103764
103765 There are many many permutations on essentially the same license terms,
103766 but I have been pedantic and treated slight differences as separate
103767 licenses.
103768
103769 Here is a description of the process I used:
103770
103771 tar xvjf /usr/portage/distfiles/xorg-server-1.1.1.tar.bz2
103772
103773 cd xorg-server-1.1.1
103774 find -name '*.c' -o -name '*.h' | xargs gvim
103775
103776 egrep -Rli "permission|copyright" * | grep -v "\.[ch]" \
103777 | grep -v "\.in$" | xargs gvim
103778
103779 cd ..
103780 tar xvjf /usr/portage/distfiles/xorg-server-1.3.0.0.tar.bz2
103781
103782commit 7fa58385724fa7f441107a1793b601ba3dcb1f4c
103783Author: Arkadiusz Miskiewicz <arekm@maven.pl>
103784Date: Wed Aug 1 08:01:28 2007 +0300
103785
103786 XFree86: xf1bpp: Fix parallel build
103787
103788 One of the constructs wasn't parallel-build safe: fix that.
103789
103790commit 18ab4d559409d4b682aab99fb75f8d861122eab6
103791Author: Daniel Stone <daniel@fooishbar.org>
103792Date: Wed Aug 1 07:27:53 2007 +0300
103793
103794 Darwin: Remove missing file
103795
103796 Xserver.m is missing and still hasn't been added, so just remove it for now.
103797
103798commit 0bd6fe7401b2524cf34793c0b0c642e3d32fae00
103799Author: Daniel Stone <daniel@fooishbar.org>
103800Date: Wed Aug 1 07:27:30 2007 +0300
103801
103802 Config: Add missing include
103803
103804commit 48b3034d13bbbb69072eb11f4579389cc32b0850
103805Author: Daniel Stone <daniel@fooishbar.org>
103806Date: Wed Aug 1 07:01:51 2007 +0300
103807
103808 Config: Add current FDI file
103809
103810 Add the FDI file we're using at the moment, until it gets into upstream HAL.
103811
103812commit 82b720cf3e09d8a6adcd40b25c4d48b34ba1ae80
103813Author: Daniel Stone <daniel@fooishbar.org>
103814Date: Wed Aug 1 06:57:11 2007 +0300
103815
103816 Config: Fix merge detritus
103817
103818commit 6b055e5d9751e3679ff98065e43225ec8a960053
103819Author: Daniel Stone <daniel@fooishbar.org>
103820Date: Wed Aug 1 06:55:36 2007 +0300
103821
103822 Input: Fix stuck modifiers (bug #11683)
103823
103824 Disclaimer: It's 6:51am. I'm trying to be as understandable as possible.
103825
103826 What was happening previously was this:
103827 * Press Alt
103828 * Extended event generated and processed: state is now Alt down once
103829 * Core event generated
103830 - keyboard switched: inherited state is Alt down once
103831 - event processed: Alt down twice
103832 * Release Alt
103833 * Extended event generated and processed: state is now null
103834 * Core event generated and processed: Alt down once
103835
103836 If we switch the order:
103837 * Press Alt
103838 * Core event generated:
103839 - keyboard switched: inherited state is null
103840 - event processed: Alt down once
103841 * Extended event generated and processed: state is now Alt down once
103842 * Release Alt
103843 * Core event generated and processed: state is now null
103844 * Extended event generated and processed: state is now null
103845
103846 When we carry over the previous state, it needs to be the _previous_ state
103847 (state and modifiersPerKey), assuming that we're going to catch now-core
103848 events for any of these. For example, if Ctrl is held down as we pivot, we
103849 need to carry Ctrl over with a count of one, for which an extended + core
103850 release will then clear. Carrying over the union of the previous state _and
103851 the state resulting from the immediate action_ was what broke things.
103852
103853commit 0e0174d45ecbeb7b6dddc4af53da9d6211038e0e
103854Author: Daniel Stone <daniel@fooishbar.org>
103855Date: Wed Aug 1 03:30:07 2007 +0300
103856
103857 XFree86: Allow disabling of HAL
103858
103859 If NoAutoAddDevices is given as a server flag, then no devices will be added
103860 from HAL events at all. If NoAutoEnableDevices is given, then the devices will
103861 be added (and the DevicePresenceNotify sent), but not enabled, thus leaving
103862 policy up to the client.
103863
103864commit cd8e99e56ec5d02026e401cc15e0f8d75f2a4727
103865Author: Daniel Stone <daniel@fooishbar.org>
103866Date: Wed Aug 1 03:29:12 2007 +0300
103867
103868 Input: Don't enable devices when we open them
103869
103870 Thanks to Xi's braindead design, it's otherwise impossible to query input
103871 devices without enabling them. Hurrah.
103872
103873commit 0a31db14b7c7c21ef550dbcc73a9f649f3613cbe
103874Author: Daniel Stone <daniel@fooishbar.org>
103875Date: Wed Aug 1 02:54:14 2007 +0300
103876
103877 Config: D-Bus core: Fix hook removal
103878
103879 Make sure we properly initialise the entire hook when adding it, and
103880 bust out when we're done removing.
103881
103882commit 89f628394f7d831f2ba1e45c5884c3983bef6031
103883Author: Daniel Stone <daniel@fooishbar.org>
103884Date: Wed Aug 1 02:08:02 2007 +0300
103885
103886 XFree86: Input: Fix whitespace
103887
103888commit aec0d06469a2fa7440fdd5ee03dc256a68704e77
103889Author: Aaron Plattner <aplattner@nvidia.com>
103890Date: Tue Jul 31 16:33:37 2007 -0700
103891
103892 Fix a crash when rotating the screen.
103893
103894 Remember output->crtc before setting a NULL mode because RRCrtcNotify now sets
103895 output->crtc to NULL. Use the saved crtc to set the new mode.
103896
103897commit a93033b0bc14ed0bb95c680ded26b63cfe5fd1d3
103898Author: Daniel Stone <daniel@fooishbar.org>
103899Date: Wed Aug 1 01:53:08 2007 +0300
103900
103901 XFree86: Module: Bump input version
103902
103903 config_info changes the size (and ordering) of DeviceIntRec, so bump the
103904 input major.
103905
103906commit 1150969b826e2bd6d8345fa245ed499f2e4cf101
103907Author: Daniel Stone <daniel@fooishbar.org>
103908Date: Wed Aug 1 01:52:20 2007 +0300
103909
103910 Convert all my license statements to the standard form
103911
103912 Convert all my license statements to the standard, accepted form:
103913 cf. <20070717142307.GD13478@fooishbar.org>
103914 http://lists.freedesktop.org/archives/xorg/2007-July/026451.html
103915
103916 keithp's license on configure.ac changed with his verbal permission.
103917
103918commit 8658f5d923a69fb55b4cd9e1e84c2d271679f6e2
103919Author: Daniel Stone <daniel@fooishbar.org>
103920Date: Wed Aug 1 01:10:50 2007 +0300
103921
103922 Hotplug: Add HAL support
103923
103924 Add support for HAL-based hotplugging, in which we just get the list of
103925 input devices and properties from HAL. Requires an FDI which is not yet
103926 in mainline HAL.
103927
103928commit aa75b3481724834da2f855d8dd2ff36074bd5706
103929Author: Daniel Stone <daniel@fooishbar.org>
103930Date: Wed Aug 1 01:09:07 2007 +0300
103931
103932 Hotplug: D-Bus: Dispatch harder
103933
103934 Dispatch until we've got nothing left to dispatch, since apparently
103935 dispatching will only ever fire a single message ...
103936
103937commit 4d238c5c67461ed747aa6c021d1532734f4c63fe
103938Author: Daniel Stone <daniel@fooishbar.org>
103939Date: Wed Aug 1 01:08:26 2007 +0300
103940
103941 Input: GetPointerEvents: Deny events from devices without valuators
103942
103943 For some reason, my keyboard has 25 mouse buttons, but zero valuators. This
103944 causes GPE to blow up spectacularly, trying to get (and set) co-ordinates from
103945 devices without valuators. For now, just prevent this from ever happening,
103946 and whack a dirty great FIXME in.
103947
103948commit 7c9e8fd56e1830f7a971187d14877ebbdf35c4b0
103949Author: Daniel Stone <daniel@fooishbar.org>
103950Date: Wed Aug 1 00:19:14 2007 +0300
103951
103952 Input: Allow enabling and disabling of devices
103953
103954 Add DEVICE_ENABLE to KDrive and XFree86 to allow us to enable and disable
103955 devices on the fly.
103956
103957commit 0afeb0241a83796575da827bd81375c99ff10af5
103958Author: Daniel Stone <daniel@fooishbar.org>
103959Date: Sun Jul 8 20:48:57 2007 +0300
103960
103961 DIX: Clean up null root cursor handling
103962
103963 Move the null root cursor handling out of main() and into CreateRootCursor.
103964
103965commit 62ec6d09b3adaea82ff52c8672e6f611c15ec56d
103966Author: Daniel Stone <daniel@fooishbar.org>
103967Date: Sun Jul 8 20:47:28 2007 +0300
103968
103969 dix.h: Remove duplicate ffs() prototype.
103970
103971commit 4d3379d418a781938358e511fd41deb4115a032c
103972Author: Daniel Stone <daniel@fooishbar.org>
103973Date: Sun Jul 8 14:31:35 2007 +0300
103974
103975 Fonts: Fix builtin fonts
103976
103977 Make sure the font path is always 'built-ins' when we use built-in fonts,
103978 rather than having it as a fixed path for a while, then clobbering it
103979 halfway through startup.
103980
103981commit 9ac7e8a559fe6008cafc95e8264680c50e72ba19
103982Author: Daniel Stone <daniel@fooishbar.org>
103983Date: Sun Jul 8 14:30:53 2007 +0300
103984
103985 Hotplug: D-Bus: API version 2
103986
103987 Use uint32s instead of int32s where practical, and add an API version
103988 request. Also, try to return all devices added, not just the first,
103989 and box device arguments.
103990
103991commit 1cdadc2f43d9069572814510d04b1a560c488fcb
103992Author: Daniel Stone <daniel@fooishbar.org>
103993Date: Sun Jul 8 14:28:58 2007 +0300
103994
103995 Hotplug: Separate D-Bus into core and hotplug API components
103996
103997 Break up D-Bus into two components: a D-Bus core that can be used by any
103998 part of the server (for the moment, just the D-Bus hotplug API, and the
103999 forthcoming HAL hotplug API), and the old D-Bus hotplug API.
104000
104001commit 8bfa41e1bf3f588780d7e9f6f900b1fde0570a7e
104002Author: Daniel Stone <daniel@fooishbar.org>
104003Date: Sun Jul 8 04:29:43 2007 +0300
104004
104005 gitignore: Add automake lex/yacc wrapper
104006
104007commit 06dd2748da8b7af343f6cab409b9f351567de5f3
104008Author: Daniel Stone <daniel@fooishbar.org>
104009Date: Sun Jul 8 00:27:40 2007 +0300
104010
104011 configure.ac: Properly check XFree86 proto modules
104012
104013 Not sure why these are conditionals, anyway. This one really needs
104014 revisiting, but at least causes configure, rather than the compilation,
104015 to bomb out.
104016
104017commit fd10312b4224197b937d9e696b53dc2a16c8912f
104018Author: Daniel Stone <daniel@fooishbar.org>
104019Date: Sun Jul 8 00:26:26 2007 +0300
104020
104021 configure.ac: Fix KDrive VESA/fbdev conditionals
104022
104023 Make sure we actually respect anything explicitly given on the configure
104024 line, instead of just stomping it with what we detect.
104025
104026commit f37612c6f2375ca904411e6caa0be19fa24f032c
104027Author: Nicolas Trangez <eikke@eikke.com>
104028Date: Sun Jul 8 00:23:57 2007 +0300
104029
104030 Hotplug: Remove unused function definition from hotplug.h
104031
104032 configDispatch hasn't been used in a long time.
104033
104034commit 951c058e7800308f7c472e77178c14400f45c1b3
104035Author: Aaron Plattner <aplattner@nvidia.com>
104036Date: Tue Jul 31 14:23:58 2007 -0700
104037
104038 Don't fail compScreenInit if the driver added its own alternate visuals.
104039
104040commit 722d73a0ef54c2ebd8ef38c4a6afa0e7c5aa3e30
104041Author: Dave Airlie <airlied@redhat.com>
104042Date: Tue Jul 31 10:34:56 2007 +1000
104043
104044 Revert "Fix RandR 1.2 conversion of two colour to ARGB cursor on MSB first platforms."
104045
104046 This reverts commit 0f057ebb272f0ee0b51b9ab37d4b07da0924fec4.
104047
104048 This screws my cursor up just starting a bare X server on Intel,
104049 I get the X more like <> than ><..
104050
104051commit 57b5b97a0710fc043b8a1c01d756cdb73dfe4567
104052Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104053Date: Sun Jul 29 11:02:47 2007 -0400
104054
104055 ReduceCompositeOp returns a Render op, not a boolean.
104056
104057commit f62beb6f3609e8b6e61325ac89017590811bbd07
104058Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104059Date: Fri Jul 27 13:23:15 2007 -0400
104060
104061 Remove all trace of Option "BiosLocation".
104062
104063 This code was deeply dangerous. If anyone actually had a use for this code,
104064 we should find a better way of doing it.
104065
104066commit 486fd4145aed93093d1f1655de40c0a8582bb8b1
104067Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104068Date: Fri Jul 27 13:10:39 2007 -0400
104069
104070 exaGetPixmapFirstPixel: avoid framebuffer readbacks if possible.
104071
104072 If the pixel in framebuffer memory isn't modified since we uploaded it, we
104073 can just read from the system memory copy, wihch avoids both a readback and
104074 an accelerator stall.
104075
104076 In principle this function is still wrong, and all the framebuffer pixel
104077 access should be going through (w)fb so we can get pixel layout corrections.
104078
104079commit 50cb6c7e4419e067c1f080d1de940811d21fc725
104080Author: Kristian Høgsberg <krh@redhat.com>
104081Date: Fri Jun 15 15:29:00 2007 -0400
104082
104083 Don't map the front buffer in libdri if the ddx driver doesn't set the size.
104084
104085 This lets drivers map the front buffer themselves
104086 by setting dontMapFramebuffer.
104087
104088commit cec793ef7a6dac9fa2a6538683e363a72672cde9
104089Author: Aaron Plattner <aplattner@nvidia.com>
104090Date: Thu Jul 26 11:49:46 2007 -0700
104091
104092 Include picturestr.h in xf86Crtc.h to pick up definition of PictTransform.
104093
104094commit 27845fe197b74bf453d99f352e83513e201fdaae
104095Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104096Date: Thu Jul 26 09:32:16 2007 -0400
104097
104098 libconfig shouldn't be an installed library.
104099
104100commit 276f8e2ca42eec982d16b86d67217d68ff98f81d
104101Author: Alan Coopersmith <alan.coopersmith@sun.com>
104102Date: Wed Jul 25 17:42:23 2007 -0700
104103
104104 Include comment/copyright/license for AC_DEFINE_DIR in acinclude.m4
104105
104106commit eba2be448bdd298ff2f7b8603bd9e976da1fdf72
104107Author: Brice Goglin <Brice.Goglin@ens-lyon.org>
104108Date: Wed Jul 25 20:53:45 2007 +0200
104109
104110 Minor fixes in cvt and gtf manpages
104111
104112 Reported by "A. Costa" <agcosta@gis.net> in
104113 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=432065
104114
104115commit 8d230319040f0a7f72231da2bf5ec97dc3612e21
104116Author: Gustavo Pichorim Boiko <boiko@mandriva.com>
104117Date: Tue Jul 24 16:19:19 2007 -0300
104118
104119 Fix the output->crtc initialization in the old randr setup
104120
104121commit 0f057ebb272f0ee0b51b9ab37d4b07da0924fec4
104122Author: Michel Dänzer <michel@tungstengraphics.com>
104123Date: Wed Jul 25 17:04:04 2007 +0200
104124
104125 Fix RandR 1.2 conversion of two colour to ARGB cursor on MSB first platforms.
104126
104127 Doesn't seem necessary to do anything here...
104128
104129commit 5b424b562eee863b11571de4cd0019cd9bc5b379
104130Author: Gustavo Pichorim Boiko <boiko@mandriva.com>
104131Date: Mon Jul 23 18:27:41 2007 -0300
104132
104133 Set the crtc before the output change is notified
104134
104135 Set the new randr crtc of the output before the output change notification is
104136 delivered to the clients.
104137 Remove RROutputSetCrtc as it is not really necessary. All we have to do is set
104138 the output's crtc on RRCrtcNotify
104139
104140commit 7da38bb6a15247948c90e00a59230453fcf13cbd
104141Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104142Date: Sat Jul 21 15:27:40 2007 -0400
104143
104144 Partial redundancy elimination in PropertyNotify generation.
104145
104146commit 0f91abd5c68eb044d09733d18ef0f6b8ed128200
104147Author: Julien Cristau <jcristau@debian.org>
104148Date: Thu Jul 19 20:37:26 2007 -0400
104149
104150 Fix alpha build failures
104151
104152 Don't include <asm/pci.h> in os-support/linux/lnx_axp.c, use "lnx.h" and
104153 <unistd.h> instead
104154
104155commit dc9c5196282ba61bd542e198dfe0d53d93181591
104156Author: Keith Packard <keithp@neko.keithp.com>
104157Date: Thu Jul 19 13:28:00 2007 -0700
104158
104159 Make PreferredMode option in config file override EDID mode preferences.
104160
104161 When the PreferredMode option is selected in the config file, remove the
104162 M_T_PREFERRED bit from all other preferred modes to force the config file
104163 mode to be selected.
104164
104165commit 73a93c5a6b68f7ba21f9e75f50b1032603a3b39e
104166Author: Keith Packard <keithp@neko.keithp.com>
104167Date: Thu Jul 19 13:26:36 2007 -0700
104168
104169 Query modes on disabled (but not ignored) outputs.
104170
104171 Code that disabled mode detection on disabled outputs would confuse
104172 applications by listing said outputs as connected but without any modes.
104173 This makes the disabled state in the config file affect only the initial
104174 configuration and not subsequent modifications by RandR.
104175
104176commit 9fc36a391c11170cde1a28f548a2cae5f6f20d5b
104177Author: Keith Packard <keithp@neko.keithp.com>
104178Date: Sat Jul 14 12:36:15 2007 -0700
104179
104180 Make pending property changes trigger mode setting.
104181
104182 The DDX code was ignoring pending properties for computing when mode setting
104183 was required. This meant that configurations differing only in property
104184 values would not cause the mode to be set.
104185
104186commit aed6569309223ecc7e26fa84e4d430e422455607
104187Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104188Date: Sat Jul 14 15:21:46 2007 -0400
104189
104190 Refactor how Composite adds visuals to the screen.
104191
104192 Besides being slightly simpler to read, it's now trivial to add a depth-16
104193 visual to a depth-24 screen just by adding a line for it in the alternate
104194 visual list. Visuals for indexed depths are slightly tricky still.
104195
104196commit 21bbd7d64b5f74915afd7a312e589654442f3461
104197Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104198Date: Tue Feb 6 21:42:50 2007 -0500
104199
104200 Delete some pre-dlloader debugging scaffolding.
104201
104202 If your loader is as bad as elfloader, then it makes sense for the
104203 server to have some stubs for you to assign to / break on. However it
104204 is no longer 1996.
104205
104206commit 1f71f0c0574bafb36da20fec669f9a1138c69a47
104207Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104208Date: Tue Feb 6 21:28:03 2007 -0500
104209
104210 Remove (long-)deprecated xf86EnablePciBusMaster.
104211
104212commit 0a63d874e9c2f4fe4b38839a744461f9d41040b2
104213Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104214Date: Tue Feb 6 21:22:49 2007 -0500
104215
104216 Always normalize the module name.
104217
104218commit 9a1c6afd12caf0143483f72bfbba0c4c3daaa6ff
104219Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104220Date: Tue Feb 6 21:19:50 2007 -0500
104221
104222 Remove dead code for screen crossing.
104223
104224commit 8ca2fe8914af1a67bf597f99025e5cbe9b08da57
104225Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104226Date: Tue Feb 6 21:11:13 2007 -0500
104227
104228 Delete dead module test code.
104229
104230commit e2413cc7cae4e578b8e9b408ea85bef596b03ea3
104231Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104232Date: Tue Feb 6 21:07:37 2007 -0500
104233
104234 Remove MEMDEBUG
104235
104236 This existed (but may not have worked) in the monolith, but is gone now.
104237
104238commit d1d65a84150dfbc3a4dbe108f237a85ab6e09bbb
104239Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104240Date: Tue Feb 6 21:01:08 2007 -0500
104241
104242 Dead ifdefs for BITMAP_SCANLINE_UNIT == 64
104243
104244 This appears to be a legacy of cfb24 not being smart enough to deal with this
104245 case. But since cfb24 unexists, die die die.
104246
104247commit cbe74394a5ed21ed80c0aab6eefd2716122cce11
104248Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104249Date: Tue Feb 6 20:44:34 2007 -0500
104250
104251 Nuke dead X -configure code.
104252
104253commit 5657fb065cc79ba3ca5a836f45637ba9894f9abf
104254Author: Dodji Seketeli <dodji@seketeli.org>
104255Date: Tue Jul 17 12:12:02 2007 +0200
104256
104257 exaDriverInit: Fail if pScreenInfo or a member of it is invalid.
104258
104259 EXA may attempt to use the invalid value and crash otherwise.
104260
104261commit bbe7ce10fa93017374d7a4611427b70a22d7507a
104262Author: Alan Coopersmith <alan.coopersmith@sun.com>
104263Date: Mon Jul 16 17:25:59 2007 -0700
104264
104265 Update pci.ids to 2007-07-16 snapshot
104266
104267 Remove nvidia ids in extrapci.ids that are now in pci.ids
104268 Add nvidia ids to extrapci.ids that are in xf86-video-nv but not pci.ids
104269
104270commit ac979c165128704116cd40086320b6edc79018e2
104271Author: Keith Packard <keithp@neko.keithp.com>
104272Date: Sat Jul 14 12:13:17 2007 -0700
104273
104274 MakeAtom needs length without trailing NUL. sizeof("string") includes NUL.
104275
104276 I made a mistake in some new code using MakeAtom, passing the size of the
104277 string instead of the length of the string. Figuring there might be other
104278 such mistakes, I reviewed the server code and found four bugs of the same
104279 form.
104280
104281commit 393171034c15d8a1b82232b8f9455a358035e932
104282Author: Keith Packard <keithp@neko.keithp.com>
104283Date: Sat Jul 14 09:03:40 2007 -0700
104284
104285 Add RandR reflection support.
104286
104287 Replace the ad-hoc transformation mechanisms with matrices.
104288 Prepares for more general transformation as well.
104289
104290commit 8773ad023eb28950eb0f802d2ca31a67f84adddc
104291Author: Keith Packard <keithp@neko.keithp.com>
104292Date: Sat Jul 14 08:47:50 2007 -0700
104293
104294 Screen size bounds check in ProcRRSetCrtcConfig not masking out reflections.
104295
104296 When checking how to validate the selected mode and position against the
104297 current screen size, the test against 90/270 rotation did not mask out
104298 reflection, so that when reflection was specified, the 90/270 test would
104299 never succeed. This caused incorrect bounds checking and would return
104300 an error to the user instead of rotating the screen.
104301
104302commit 881a620b4d6ea7a54af14c8f8fbe6924c9aa9291
104303Author: Keith Packard <keithp@neko.keithp.com>
104304Date: Sat Jul 14 08:45:10 2007 -0700
104305
104306 When sync'ing logfile, also flush it.
104307
104308 When the logfile is set to sync, the actual sync occurs whenever the log
104309 file is flushed. If the log file is not also set to flush, no syncing
104310 occurs.
104311
104312commit 031b009ea678809bf1ddca883c2082b304c408c9
104313Author: Alan Coopersmith <alan.coopersmith@sun.com>
104314Date: Fri Jul 13 14:54:45 2007 -0700
104315
104316 Use %S instead of %s for strftime seconds when printing build time
104317
104318commit c0e91777a9874fe2cd9a7e9180263f512c1e8f8d
104319Author: Alan Coopersmith <alan.coopersmith@sun.com>
104320Date: Thu Jul 12 16:37:11 2007 -0700
104321
104322 Add __SOL8__ to xorg-server.h.in since xf86-input-kbd needs it to build
104323
104324commit 6b4231e3b5b49b731c9a00930ae465fff8539831
104325Author: Alan Coopersmith <alan.coopersmith@sun.com>
104326Date: Thu Jul 12 16:36:27 2007 -0700
104327
104328 Use kbd driver when xorg.conf specifies "keyboard" or "Keyboard" (bug #11301)
104329
104330 X.Org Bug #11301 <https://bugs.freedesktop.org/show_bug.cgi?id=11301>
104331 Sun Bug #6560332 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6560332>
104332
104333commit 9fcb30ebf7b7b2137955f759e95c1d58c4f27a11
104334Author: Alan Coopersmith <alan.coopersmith@sun.com>
104335Date: Thu Jul 12 13:00:32 2007 -0700
104336
104337 Make SOLARIS_INOUT_ARCH substitutions work better with automake-1.10
104338
104339commit 7c0ca27f6dd0a800dc27429a33dbc8e133f9a9c1
104340Author: Alan Coopersmith <alan.coopersmith@sun.com>
104341Date: Wed Jul 11 17:15:29 2007 -0700
104342
104343 "fbpict.c", line 215: void function cannot return value
104344
104345commit 0a4e9311158ed3ecda0722640f860ace2f87a97e
104346Author: Hanno Boeck <hanno@gentoo.org>
104347Date: Thu Jul 12 10:17:07 2007 +1000
104348
104349 xnest: fix linking since dbus
104350
104351 Fixes bug 8955
104352
104353commit b2f9ca6ac400d426d7a1ef0162f7e7ce28288dd1
104354Author: Keith Packard <keithp@neko.keithp.com>
104355Date: Tue Jul 10 21:33:34 2007 -0700
104356
104357 Redirect fix: Manual + Automatic - Manual = Automatic
104358
104359 A window with redirect manual *and* redirect automatic which loses the
104360 manual redirecting client becomes redirect automatic.
104361
104362commit 561989f2f0fc31e3d3bf8df978a9cb3d4c85af59
104363Author: Keith Packard <keithp@neko.keithp.com>
104364Date: Tue Jul 10 21:06:51 2007 -0700
104365
104366 Generate ChangeLog file for make dist.
104367
104368 Copy Makefile.am snippet which generates a complete git change history to
104369 the ChangeLog file during the distribution generation process.
104370
104371commit e316fa59fea8b7b18cdf3a227890351a9567ec65
104372Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104373Date: Tue Jul 10 14:20:55 2007 -0400
104374
104375 Add per-monitor config file option for maximum pixel clock.
104376
104377commit ffdf8f3e452638e6b050fccabee465d6c0300f45
104378Author: Peter Hutterer <peter@cs.unisa.edu.au>
104379Date: Tue Jul 10 13:56:38 2007 +0930
104380
104381 DeliverEventsToWindow: ensure that genericMask has a defined value.
104382
104383commit 161624a5a45808fd56141dc2c64be729944f03ed
104384Author: Michel Dänzer <michel@tungstengraphics.com>
104385Date: Tue Jul 10 09:02:40 2007 +0200
104386
104387 GLX: Only build code dealing with GLXPixmap damage field when DRI is enabled.
104388
104389commit 4abd00dab7e648dab8172f6009371e4e63d0c521
104390Author: Michel Dänzer <michel@tungstengraphics.com>
104391Date: Tue Jul 10 09:02:08 2007 +0200
104392
104393 Make sure DRI drawables are cleaned up when client dies.
104394
104395 The previous scheme didn't work when the client didn't create the core drawable,
104396 e.g. the root or composite overlay window. Use refcounting via special client
104397 resources to fix that.
104398
104399commit 583e988b9f7cfb9293144c8309023c0dd1766715
104400Author: Peter Hutterer <peter@cs.unisa.edu.au>
104401Date: Tue Jul 10 09:19:02 2007 +0930
104402
104403 Install geext.h, otherwise drivers won't build.
104404
104405commit 9809715afaafee9baf2aef348c1ebda7e8b3f076
104406Author: Peter Hutterer <peter@cs.unisa.edu.au>
104407Date: Mon Jul 9 10:42:03 2007 +0930
104408
104409 Change CheckMotion to ignore non-pointer events but acknowledge XI events.
104410
104411 Call CheckMotion from ProcessOtherEvents() to make sure absolute XI events
104412 update the sprite before an event is sent.
104413
104414commit 62efc3951a96648cf975302aa6651cb67b87fa64
104415Author: Peter Hutterer <peter@cs.unisa.edu.au>
104416Date: Fri Jul 6 17:00:20 2007 +0930
104417
104418 DeliverGrabbedEvent: stop segfault when gemask == NULL.
104419
104420commit 3312e4dd5e055b2cb445b5d5c617aa7a611eedc1
104421Author: Peter Hutterer <peter@cs.unisa.edu.au>
104422Date: Fri Jul 6 16:41:52 2007 +0930
104423
104424 Call CheckMotion for all core events.
104425
104426 We need to do this to update the sprites x/y coordinate before we assemble a
104427 button event. Absolute devices may send a buttonEvent with valuators attached.
104428 If we don't update the sprite before assembling the event, the valuators are
104429 lost and the button press is delivered to the previous position of the device.
104430 Doesn't have any effect on relative devices.
104431
104432commit c1a6841a64576b7e688e9ca0d3e0db8acf52d4ae
104433Author: Paulo Ricardo Zanoni <prz05@c3sl.ufpr.br>
104434Date: Tue Jul 10 10:08:44 2007 +0930
104435
104436 ProcX{Change|Query}WindowAccess: change device list from char* to XID*.
104437
104438commit 5957aa6fdc580ccad4557eeefa0636ffad823f33
104439Author: Michel Dänzer <michel@tungstengraphics.com>
104440Date: Mon Jul 9 08:47:05 2007 +0200
104441
104442 Fix regression from recent composite changes.
104443
104444 One pWin->redirectDraw test was converted incorrectly, causing incorrect
104445 rendering in some cases.
104446
104447commit 5ccc09b18244f91a06b3bea20b02a97280d1a229
104448Author: Peter Hutterer <peter@cs.unisa.edu.au>
104449Date: Fri Jul 6 15:43:08 2007 +0930
104450
104451 Use the same struct for generic event masks throughout the code.
104452 Renaming those structs too.
104453
104454 Previously grabs were using a different struct than windows, which was
104455 reasonably stupid.
104456
104457commit bcb23527421578bd4c9397d4c2c19cbefa22fc59
104458Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104459Date: Thu Jul 5 15:56:25 2007 -0400
104460
104461 Clean up unused #ifdefs from fb.
104462
104463commit 9ff7ff2fda30f334515b16ef0867c1500c41bc0f
104464Author: Keith Packard <keithp@neko.keithp.com>
104465Date: Wed Jul 4 23:38:27 2007 -0700
104466
104467 Fix MEMORY SMASH in XkbCopyKeymap.
104468
104469 XkbCopyKeymap reallocates the destination keymap when it is not large enough
104470 to hold the source data. When reallocating the map->types data, it needs to
104471 zero out the new entries. The computation for where to start bzero'ing was
104472 accounting for the size of the data type twice, once implicitly in the
104473 pointer arithmetic, and once explicitly with '* sizeof (XkbKeyTypeRec)'.
104474 This would often lead to random memory corruption when the destination
104475 keymap had existing map->types data.
104476
104477commit 9131d560a0d42067cc4e726e445e060216c9acdc
104478Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104479Date: Thu Jul 5 02:47:34 2007 -0300
104480
104481 Postpone options variable assignment to fix segfault when we got a device but
104482 its driver is incorrect. Also if (!ki && !pi) can never be true.
104483
104484 This one also adds the device option field.
104485
104486commit 41b485d5507821e41c3281c3c565647ae7582101
104487Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104488Date: Thu Jul 5 02:40:07 2007 -0300
104489
104490 kdrive must to know that devices are unplugged.
104491
104492commit a92dc6b5295e4f352115fed2856169929819863f
104493Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104494Date: Thu Jul 5 02:28:14 2007 -0300
104495
104496 Remove redundant linking in kdrive. Fix configure.ac variable name and clean
104497 it up a little.
104498
104499commit 41b5155c8be75c4e171c0f64616cc09598b8ec54
104500Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104501Date: Thu Jul 5 01:57:41 2007 -0300
104502
104503 For each kdrive server put a dependencie on its own libraries.
104504
104505commit 16e429bcbf2f62cfc58162ab2857afb7376dda41
104506Author: Jonathan Lim <jlim@sgi.com>
104507Date: Wed Jul 4 20:08:49 2007 +0200
104508
104509 Bug 5000: Fix domain support for SGI Altix
104510
104511commit f106c04b627d9f57b38627971dc79c75129e66d6
104512Author: Keith Packard <keithp@neko.keithp.com>
104513Date: Tue Jul 3 14:47:19 2007 -0700
104514
104515 Have Composite always report server version.
104516
104517 It was reporting the lessor of the server and client versions, which doesn't
104518 make sense with the 0.4 semantic change in clipping.
104519
104520commit 4f88d68bdb90cc7d12170355105b4fd020acd306
104521Author: Keith Packard <keithp@neko.keithp.com>
104522Date: Tue Jul 3 14:43:17 2007 -0700
104523
104524 Force advertised Composite version to 0.4 instead of using header version.
104525
104526 Installed protocol header version may be newer than the server code base.
104527 Use internal version number for Composite extension to make sure the server
104528 doesn't advertise capabilities it doesn't support.
104529
104530commit 866f092ca0160a366add01b48ad03438926c4d16
104531Author: Keith Packard <keithp@neko.keithp.com>
104532Date: Tue Jul 3 14:29:11 2007 -0700
104533
104534 Make Composite manual redirect windows not clip their parent.
104535
104536 This patch changes the semantics of manual redirect windows so that they no
104537 longer affect the clip list of their parent. Doing this means the parent can
104538 draw to the area covered by the child without using IncludeInferiors. More
104539 importantly, this also means that the parent receives expose events when
104540 that region is damaged by other actions.
104541
104542commit 2a75c774975b50dd4e71b7dbea7bd65ca2984a43
104543Author: Dodji Seketeli <dodji@seketeli.org>
104544Date: Tue Jul 3 11:00:29 2007 +0200
104545
104546 ExaOffscreenMarkUsed: Don't crash when there's no offscreen memory.
104547
104548commit 0ede39a25cf5b0b6c2c89677f810c21ce42b95df
104549Author: Michel Dänzer <michel@tungstengraphics.com>
104550Date: Tue Jul 3 10:55:13 2007 +0200
104551
104552 Fix build when int10 doesn't use x86emu.
104553
104554commit 028a00bc518dc6908839e8ce7c50ab1837100945
104555Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104556Date: Mon Jul 2 18:41:55 2007 -0400
104557
104558 Make x86emu's I/O cycle tracing more useful.
104559
104560 Print debug messages only when the appropriate debug bit is set in the
104561 8086 state vector, so you can focus in on the call you're actually
104562 interested in.
104563
104564commit 00e8295b7e0c7c0ba97707903004272818e3d87d
104565Author: Gero Mudersbach <geroxp@web.de>
104566Date: Mon Jul 2 11:40:11 2007 -0700
104567
104568 Bug #10814: Add needed quirk for Samsung 225BW like the 226BW.
104569
104570commit 667e95f2e8389d9f23c50446d6d664eddd16d260
104571Author: Eric Anholt <eric@anholt.net>
104572Date: Mon Jul 2 11:36:11 2007 -0700
104573
104574 Correct the xf86EdidModes.c file description.
104575
104576commit 3de1f0d03b329b01856f664651db23ffefb58646
104577Author: Eric Anholt <eric@anholt.net>
104578Date: Tue May 29 10:08:58 2007 -0700
104579
104580 Fix documentation of association of outputs to monitor sections in xorg.conf(5)
104581
104582commit c1a16bdcfe7aa907fe78f27dc606a8e5a2699952
104583Author: Peter Hutterer <peter@cs.unisa.edu.au>
104584Date: Wed Jun 27 18:08:03 2007 +0930
104585
104586 Add 'evfill' field to GEExtensions.
104587 We need this to allow extensions to fill in extra data for an event before it
104588 is sent to the client. This is probably to be used like
104589 FillUpEventsFromWindow().
104590
104591commit 4d76075dbb618a47ff9fc15c4be2e2d34210fa8d
104592Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104593Date: Fri Jun 29 14:06:52 2007 -0400
104594
104595 Death to RCS tags.
104596
104597commit 2691c05fd647d9fa10f791ac397ecb9c423a076f
104598Author: Peter Hutterer <peter@cs.unisa.edu.au>
104599Date: Fri Jun 29 11:56:18 2007 +0930
104600
104601 Make sure window->optional is allocated before assigning it.
104602
104603 DeletePassiveGrabFromList() may remove the window optional, so we need to
104604 re-alloc it if it isn't there anymore.
104605
104606 Thanks to Colin Harrison for spotting the bug.
104607
104608commit f7f3fe7fe7233a2ffc43106c48f44cbbd82b7c19
104609Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104610Date: Thu Jun 28 18:59:05 2007 -0400
104611
104612 Remove the remnants of OS/2 support.
104613
104614 This has never worked in any modular server release, and as far as I know
104615 was never tested in 6.7 through 6.9.
104616
104617commit 8a06ff9ffa4816d192e58e43e7fe569b97b4dd7c
104618Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104619Date: Thu Jun 28 16:41:28 2007 -0400
104620
104621 Fix another usage of MAX_PCI_DEVICES.
104622
104623 Fixes cases where the VGA device is above the 128th device on the system.
104624
104625commit 928836a5abd85466e920eb487fab9ccb295e0c5b
104626Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104627Date: Thu Jun 28 16:29:28 2007 -0400
104628
104629 Bug #10770: "Inputdevs" isn't a valid config file keyword.
104630
104631commit 62f43d8b33f67d8f3d0bd65787ffae9e6b634d65
104632Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104633Date: Wed Jun 27 16:24:42 2007 -0300
104634
104635 Relink properly all kdrive servers when changes happen.
104636
104637commit 3860996d5666b76600b1537e2cbd58e36b086308
104638Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104639Date: Wed Jun 27 16:06:12 2007 -0300
104640
104641 Remove double-defined.
104642
104643commit 1e189ed1daab58c1de67d387306fde0a9e7984a4
104644Author: Aaron Plattner <aplattner@nvidia.com>
104645Date: Wed Jun 27 10:16:40 2007 -0700
104646
104647 Handle tileStride > 1 in fbEvenTile.
104648
104649 Patch courtesy of James Jones.
104650
104651commit bf1ad1aa4270dccf1540943d97e80b317c0adb56
104652Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104653Date: Wed Jun 27 09:19:28 2007 -0400
104654
104655 Add VBE PanelID support.
104656
104657 Originally found only in the i810 driver.
104658
104659commit d73835efda4995a310188537233a984f4b73628d
104660Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104661Date: Wed Jun 27 03:34:13 2007 -0300
104662
104663 More janitor work. Remove 'defined but not used' warnings from kdrive and some minor cosmetic.
104664
104665commit 38f8e536684193c3f70b23be22d818053c676072
104666Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104667Date: Wed Jun 27 03:19:37 2007 -0300
104668
104669 Janitor work. Remove 'defined but not used' warnings from xorg and other cosmetic.
104670
104671commit 1340f34ec98c41781164018d43bd7bb858d8132b
104672Author: Tiago Vignatti <vignatti@c3sl.ufpr.br>
104673Date: Wed Jun 27 03:04:55 2007 -0300
104674
104675 Fix kdrive XKB.
104676
104677commit 08779c62ba2f4cbfa47e3cf738daf8abd74b60cf
104678Author: Peter Hutterer <peter@cs.unisa.edu.au>
104679Date: Wed Jun 27 11:16:55 2007 +0930
104680
104681 Reset grab->next to NULL when we're creating a grab.
104682
104683 This is definitely necessary in CheckDeviceGrabAndHintWindow (thanks to Paulo
104684 Zanoni for spotting it). We're resetting it everywhere else too, just to be
104685 save.
104686
104687commit 49136d3e9fe2d065090ee90c18a49f65fb027c9f
104688Author: Peter Hutterer <peter@cs.unisa.edu.au>
104689Date: Tue Jun 26 19:43:54 2007 +0930
104690
104691 mieqEnqueue: use modulo for queue tail wrapping.
104692
104693 This was previously committed by Michael Daenzer, but was lost during a pull.
104694
104695commit 8396bf7c8b4eb6b8f2d1e1ef19887064c69591da
104696Author: Peter Hutterer <peter@cs.unisa.edu.au>
104697Date: Tue Jun 26 12:58:37 2007 +0930
104698
104699 Init genericMasks to 0, make sure list is NULL terminated.
104700
104701commit 9725516b4274ceaf9d6caf372c5439b4c9db2316
104702Author: Alan Coopersmith <alan.coopersmith@sun.com>
104703Date: Fri Jun 22 17:05:21 2007 -0700
104704
104705 Split checks for dtrace & getpeerucred()
104706
104707commit edb9ccf3ecb4e35a840aa13815979c5fbd73f32d
104708Author: Michel Dänzer <michel@tungstengraphics.com>
104709Date: Mon Jun 25 10:51:38 2007 +0200
104710
104711 Make sure DRIScreenPrivIndex is -1 when no DRI screen private is allocated.
104712
104713 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=11340 .
104714
104715commit c4e850a781e3b85631cb386d24efcca2a835d4c9
104716Author: Peter Hutterer <peter@cs.unisa.edu.au>
104717Date: Mon Jun 25 13:38:25 2007 +0930
104718
104719 Quickfix to stop core key events from doubling up.
104720
104721commit 4c601b904ee6fb01da3343ff9ef00d36f1341fcb
104722Author: Peter Hutterer <peter@cs.unisa.edu.au>
104723Date: Mon Jun 25 10:53:05 2007 +0930
104724
104725 configFiles(): don't return anything when declared as void.
104726
104727commit bec4e47d128ec40b58a2c9aae475f6a6fc4323c3
104728Author: Peter Hutterer <peter@cs.unisa.edu.au>
104729Date: Mon Jun 25 10:51:42 2007 +0930
104730
104731 NULL-terminate device list when synthesizing core devices.
104732
104733 This fix is required for 93ca526892c0d22afa05cce6496198c652043a19 to work.
104734
104735commit 8e5102b9f01821048e72e7f068193a0b3e1816f9
104736Author: Peter Hutterer <peter@cs.unisa.edu.au>
104737Date: Thu Jun 21 15:47:48 2007 +0930
104738
104739 Set the detail field for DeviceKeyEvents to the keycode.
104740
104741 (cherry picked from commit 0c33dc152e372cdc8ae59d9a5696b0774bcd03b7)
104742
104743commit 87564543d92c1ee1f8cb6fb9716a15d693e08cf5
104744Author: Peter Hutterer <peter@cs.unisa.edu.au>
104745Date: Tue Jun 19 18:20:05 2007 +0930
104746
104747 Only decrement buttonsDown when the button count is greater than 0.
104748
104749 Device drivers flush their buttons on device init and cause a button down
104750 event to be generated. If we unconditionally decrease the buttons, we won't be
104751 able to ever get a passive device grab.
104752
104753 Format documentation for CheckDeviceGrabs to make it readable.
104754 (cherry picked from commit 3e894974cdd6a75683d4601f71622d1da7ec4395)
104755
104756 Conflicts:
104757
104758 Xi/exevents.c
104759
104760commit 24ee89fd60f489f2d3af0399e0d667057df74d02
104761Author: Peter Hutterer <peter@cs.unisa.edu.au>
104762Date: Tue Jun 19 15:31:56 2007 +0930
104763
104764 Add a few comments to devices.c
104765
104766commit 93ca526892c0d22afa05cce6496198c652043a19
104767Author: Peter Hutterer <peter@cs.unisa.edu.au>
104768Date: Wed Jun 13 15:28:15 2007 +0930
104769
104770 Split up memory for devices configured in the config file.
104771
104772 If we're using a continuous block here, we segfault when a device removal
104773 triggers an xfree call.
104774
104775commit b141b85c254afff3ce2221d899787fab3dc295bd
104776Author: Peter Hutterer <peter@cs.unisa.edu.au>
104777Date: Wed Jun 13 15:26:03 2007 +0930
104778
104779 Check for identical grabs when adding a new passive grab. If an identical grab
104780
104781 exists, remove the old one and prepend the new one.
104782
104783 X.org Bug 2738 <https://bugs.freedesktop.org/show_bug.cgi?id=2738>
104784
104785commit 19cde59c41cf167cc609debfee75bfc015beac12
104786Author: Søren Sandmann Pedersen <sandmann@redhat.com>
104787Date: Fri Jun 22 00:38:50 2007 -0400
104788
104789 In fbFill() make sure the drawable is validated when pixman_fill() succeeds.
104790
104791 In fbSolidBoxClipped() don't return when pixman_fill() succeeds.
104792
104793commit d2177c80915f2fe2e8a5c948d4ba2fa51dbfbea2
104794Author: Keith Packard <keithp@neko.keithp.com>
104795Date: Fri Jun 22 02:08:21 2007 +0100
104796
104797 Skip driver mode detection/configuration when !vtSema.
104798
104799 When the server is not active, make sure the driver functions related to
104800 mode setting are not called.
104801
104802commit 32c0dcc8c0d1edba5d7e418fd2dc916847a4f069
104803Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
104804Date: Thu Jun 21 15:39:19 2007 -0400
104805
104806 xselinux: adjust the config file format to that expected by libselinux.
104807
104808 This file will eventually be moved out of the X source tree.
104809
104810commit 2030e9e5395be43bd8eab15b65c21ca4c2f1e619
104811Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
104812Date: Thu Jun 21 15:37:18 2007 -0400
104813
104814 xselinux: use new libselinux support for context labeling.
104815
104816 Remove all the config file parsing code and use the new lookup interface
104817 instead.
104818
104819commit 4d5df14f2c4a3108a8c8adfcf4766c0d1a9daad2
104820Author: Peter Hutterer <peter@cs.unisa.edu.au>
104821Date: Thu Jun 21 18:24:30 2007 +0930
104822
104823 Save processInputProc before wrapping it and restore it later, instead of
104824 using a hardcoded ProcessKeyboardEvent. Otherwise we lose the ability to
104825 process DeviceKeyEvents after the first key press.
104826
104827 This should be the correct fix now.
104828
104829commit f6f3322fc7562dbfccaf798b05d42ea222860ba8
104830Author: Peter Hutterer <peter@cs.unisa.edu.au>
104831Date: Thu Jun 21 18:16:21 2007 +0930
104832
104833 Revert "Save processInputProc before wrapping it and restore it later, instead of"
104834 Mixed up source trees, this was a very bad fix.
104835
104836 This reverts commit 8431f6083076cf1e701366767d8f8d32a628200f.
104837
104838commit 8431f6083076cf1e701366767d8f8d32a628200f
104839Author: Peter Hutterer <peter@cs.unisa.edu.au>
104840Date: Thu Jun 21 17:00:41 2007 +0930
104841
104842 Save processInputProc before wrapping it and restore it later, instead of
104843 using a hardcoded ProcessKeyboardEvent. Otherwise we lose the ability to
104844 process DeviceKeyEvents after the first key press.
104845
104846commit 0c33dc152e372cdc8ae59d9a5696b0774bcd03b7
104847Author: Peter Hutterer <peter@cs.unisa.edu.au>
104848Date: Thu Jun 21 15:47:48 2007 +0930
104849
104850 Set the detail field for DeviceKeyEvents to the keycode.
104851
104852commit e523859a952d49b20f3d10152cc0ef695d2c12a1
104853Author: Alan Coopersmith <alan.coopersmith@sun.com>
104854Date: Wed Jun 20 17:54:38 2007 -0700
104855
104856 Include module name in "already built-in" message
104857
104858commit 5138f710a1574fef6f553f3fe2fccac0620d2584
104859Author: Arcady Goldmints-Orlov <arcadyg@nvidia.com>
104860Date: Wed Jun 20 16:31:55 2007 -0700
104861
104862 Fixed fbSolidBoxClipped() to fill the right place.
104863
104864 Changed an X2 to a Y1.
104865
104866 Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
104867
104868commit c9b79a355845c895aca8303a39798264d80b6212
104869Author: Michel Dänzer <michel@tungstengraphics.com>
104870Date: Wed Jun 20 18:56:06 2007 +0200
104871
104872 exaPolyFillRect: Don't track damage explicitly.
104873
104874 All callers should already do it.
104875
104876 Also don't leak pReg.
104877
104878commit 87966c5d2889873cea6cbc16b7e4399490dfaec1
104879Author: Michel Dänzer <michel@tungstengraphics.com>
104880Date: Wed Jun 20 18:42:00 2007 +0200
104881
104882 exaGetImage: Don't migrate or try to accelerate for 1x1.
104883
104884 This is mainly to avoid wasting effort for XSync(), but just reading a single
104885 pixel directly is probably faster than DownloadFromScreen anyway. Though in
104886 light of the latter, even larger thresholds might be useful.
104887
104888 Also move the swappedOut check before the migration checks because migration
104889 can't actually occur when swapped out.
104890
104891commit 1f97a7647606226cde61d6ad7a94f2b0b571a06c
104892Merge: 5dee64f 2e7fef7
104893Author: Peter Hutterer <peter@cs.unisa.edu.au>
104894Date: Tue Jun 19 17:20:52 2007 +0930
104895
104896 Merge branch 'master' into mpx
104897
104898 Conflicts:
104899
104900 dix/devices.c
104901 hw/xfree86/common/xf86Xinput.c
104902 hw/xfree86/loader/xf86sym.c
104903 mi/mieq.c
104904
104905commit 40f27a2df4906d9ceb1c78f6163a62c497321535
104906Author: Michel Dänzer <michel@tungstengraphics.com>
104907Date: Tue Jun 19 09:11:16 2007 +0200
104908
104909 mieqEnqueue: Make local queue tail variables unsigned.
104910
104911 So the modulo arithmetic actually works as intended... thanks to Peter Hutterer
104912 for pointing out the problem.
104913
104914commit 5dee64fc99f34e091abce65d47c4b6f026ab4849
104915Author: Peter Hutterer <peter@cs.unisa.edu.au>
104916Date: Tue Jun 19 11:31:22 2007 +0930
104917
104918 Add a few comments to ProcXGrabDevice.
104919
104920commit 9e257029c760883c4ea0715d4fd06476f3fe8053
104921Author: Peter Hutterer <peter@cs.unisa.edu.au>
104922Date: Tue Jun 19 11:28:07 2007 +0930
104923
104924 Add implicitGrab field to GrabInfoRec.
104925 Is set when passive grab is implicit as result of a ButtonPress event. If this
104926 is the case, we need to store the XI mask as well as the core mask to ensure
104927 delivery of XI events during the grab's lifetime.
104928
104929 Remove all core grabs on other devices when client issues a GrabPointer or
104930 GrabKeyboard request. Let's assume that the client really only wants one
104931 device to interact, so this seems like a reasonable solution.
104932
104933commit ab7a6d860d4a275a810a64b1ba7b13726ed10575
104934Merge: e9130b8 2e7fef7
104935Author: Ian Romanick <idr@us.ibm.com>
104936Date: Mon Jun 18 16:51:13 2007 -0700
104937
104938 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
104939
104940 Conflicts:
104941
104942 hw/xfree86/os-support/bus/Pci.c
104943 hw/xfree86/os-support/bus/linuxPci.c
104944
104945commit 2e7fef7d0837939e822c40b6ac77e7f0e66d57bd
104946Author: Adam Jackson <ajax@benzedrine.nwnk.net>
104947Date: Mon Jun 18 12:08:39 2007 -0400
104948
104949 Make xf86{En,Dis}ableInterrupts no-ops on Linux.
104950
104951commit 831d3b7f8d053aba649c8d04af3bef96376bdc3a
104952Author: Lennart Buytenhek <buytenh@wantstofly.org>
104953Date: Mon Jun 18 12:05:55 2007 -0400
104954
104955 Compile fixes for Linux ARM platforms.
104956
104957commit 562ca3f2f9005e7c5ed0a24b0759051ded2173e9
104958Author: Zephaniah E. Hull <warp@agamemnon.b5>
104959Date: Mon Jun 18 12:00:49 2007 -0400
104960
104961 In NewInputDeviceRequest, only call EnableDevice if xf86Screens[0]->vtSema is
104962 true, preventing unwanted behavior in the case where a device is added while
104963 the user is in a different VT.
104964
104965commit 3e894974cdd6a75683d4601f71622d1da7ec4395
104966Author: Peter Hutterer <peter@cs.unisa.edu.au>
104967Date: Fri Jun 15 17:16:16 2007 +0930
104968
104969 Only decrement buttonsDown when the button count is greater than 0.
104970 Device drivers flush their buttons on device init and cause a button down
104971 event to be generated. If we unconditionally decrease the buttons, we won't be
104972 able to ever get a passive device grab.
104973
104974 Format documentation for CheckDeviceGrabs to make it readable.
104975
104976commit 42c2e14b254f6f882b3e79444360ab855db43e27
104977Author: Matthieu Herrb <matthieu@deville.herrb.com>
104978Date: Fri Jun 15 00:14:02 2007 +0200
104979
104980 swap xOrigin and yOrigin in SProcRenderSetPictureClipRectangles.
104981
104982 Fixes Xrender clipping rectangles when X server and client are of
104983 different endianness, shown by xterm 225 among others.
104984
104985commit 0aaf51546666f71e8d1711978bc7988ec2ecc7d9
104986Author: Peter Hutterer <peter@cs.unisa.edu.au>
104987Date: Tue Jun 12 16:55:26 2007 +0930
104988
104989 Zero deviceMask and genericMask when creating a grab via CreateGrab().
104990
104991commit 78179ae827bb5d19abb1340084362bc51ad5c1e5
104992Author: Søren Sandmann Pedersen <sandmann@redhat.com>
104993Date: Mon Jun 11 22:46:42 2007 -0400
104994
104995 Remove fbmmx.[ch] files
104996
104997commit eb2d7fe02f9cbca57b462bba05498e2d59316fbc
104998Author: Søren Sandmann Pedersen <sandmann@redhat.com>
104999Date: Mon Jun 11 22:43:01 2007 -0400
105000
105001 Replace fbFillmmx() with pixman_fill() and remove fbmmx.[ch]
105002
105003commit f52ae237d3eec79ccd64cdd77271aeacc37af70c
105004Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105005Date: Mon Jun 11 22:02:39 2007 -0400
105006
105007 Require pixman 0.9.2
105008
105009commit d1d85c04e248f46b1cf1b1d25fdd56aa69b8f0ee
105010Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105011Date: Mon Jun 11 21:25:42 2007 -0400
105012
105013 Delete fbBltmmx().
105014
105015commit e2a720c9a17dc860ee0a858c2b21fd71e86cdcd0
105016Merge: 878cac7 3f9adb1
105017Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
105018Date: Mon Jun 11 15:29:11 2007 -0400
105019
105020 Merge branch 'master' into XACE-SELINUX
105021
105022 Conflicts:
105023
105024 dix/dixutils.c
105025
105026commit 3f9adb18f127318d054f30a57e3a77176e14c692
105027Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105028Date: Mon Jun 11 15:19:27 2007 -0400
105029
105030 Port a few forgotten fbSolidFillmmx()'es to fbFillmmx(). Use pixman_blt() instead of fbBltmmx()
105031 in fbCopyNToN().
105032
105033commit 878cac71aa0018deee861b297638c0744dba631b
105034Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
105035Date: Mon Jun 11 14:19:37 2007 -0400
105036
105037 xselinux: use new libselinux support for private Flask definitions.
105038
105039 Removes indirect dependency on kernel headers.
105040
105041commit d06099b38e8445e6e31f5178ffefcc31a71080ef
105042Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105043Date: Mon Jun 11 13:28:24 2007 -0400
105044
105045 Remove fbCopyAreammx() and fbSolidFillmmx()
105046
105047commit d4a034370c8ae71b2cc4fe824ceee58b19624f35
105048Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105049Date: Mon Jun 11 13:26:24 2007 -0400
105050
105051 Split fbSolidFillmmx() into a new FbFillmmx() function. Call that from fbFill().
105052
105053commit 3210902a7334f3d8d6c18a34a3cb3f55803b0043
105054Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105055Date: Mon Jun 11 13:14:50 2007 -0400
105056
105057 Split fbCopyAreammx() into a new fbBltmmx() function; call this
105058
105059 function from fbCopyNToN().
105060
105061commit 8d5f4368eac1b259db3e61f877a4cc10f04efa2f
105062Author: Clark Rawlins <clark.rawlins@escient.com>
105063Date: Mon Jun 11 16:53:38 2007 +0200
105064
105065 Really make sure BUILD_TIME doesn't have a leading zero.
105066
105067 date +'%k%M%S' still gives a leading zero in the hour after midnight...
105068
105069 Add a leading 1 and remove it in xf86PrintBanner().
105070
105071commit 54e023cec07aa7e392da36e11d0a4667b8341370
105072Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105073Date: Mon Jun 11 09:16:46 2007 -0400
105074
105075 Don't pass regions to pixman_image_composite() anymore.
105076
105077commit 5cbec267b6426960c90f6bcff1d051af5084538c
105078Author: Michel Dänzer <michel@tungstengraphics.com>
105079Date: Mon Jun 11 12:38:41 2007 +0200
105080
105081 Make sure BUILD_TIME doesn't have a leading zero.
105082
105083 It causes the compiler to treat it as an octal constant instead of decimal as
105084 intended, which could even cause a build failure in the cases of 08 and 09.
105085
105086 Thanks to Clark Rawlins for pointing out the problem.
105087
105088commit 1aceec61ff203848576c47a1eab13f90a67d7176
105089Author: Michel Dänzer <michel@tungstengraphics.com>
105090Date: Mon Jun 11 09:23:19 2007 +0200
105091
105092 DRI: Clip cliprects obtained from DRIGetDrawableInfo to screen dimensions.
105093
105094 This is to avoid issues with redirected windows which are located partly or
105095 fully outside of a screen edge, resulting in unusual cliprects which the 3D
105096 drivers generally can't handle. The symptoms in such cases would be incorrect
105097 rendering or even crashes or hangs.
105098
105099commit 5d896e43fd056d935935b4eb66562791edc247a1
105100Author: Michel Dänzer <michel@tungstengraphics.com>
105101Date: Mon Jun 11 09:23:19 2007 +0200
105102
105103 DRITreeTraversal: Stop walking tree when we've seen all DRI windows.
105104
105105commit 0fb44c6f9a0415184818ba8357a21ff920e907dc
105106Author: Michel Dänzer <michel@tungstengraphics.com>
105107Date: Mon Jun 11 09:23:19 2007 +0200
105108
105109 DRI: Fix build warning.
105110
105111commit 644f7ddc0cb029e2ebca43742fd8a46a1a3f4c9f
105112Author: Michel Dänzer <michel@tungstengraphics.com>
105113Date: Mon Jun 11 09:23:18 2007 +0200
105114
105115 dixLookupClient: Use access parameter.
105116
105117commit 30a3297fed9af3a594aba0875a8f58a0a38b33fc
105118Author: Michel Dänzer <michel@tungstengraphics.com>
105119Date: Mon Jun 11 09:23:18 2007 +0200
105120
105121 mieq queue handling cleanups.
105122
105123 In particular, fix handling of wraparounds in mieqEnqueue.
105124
105125commit c1a49a9269f14b6975a1a2c751bb179757373f11
105126Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105127Date: Sun Jun 10 22:14:57 2007 -0400
105128
105129 GNU is wrong and ` is not left-quote.
105130
105131commit 67a0a4da1a225ee3bd6bbd1846f8141fe333c884
105132Author: Dave Airlie <airlied@linux.ie>
105133Date: Mon Jun 11 11:55:11 2007 +1000
105134
105135 update xproto dependency to at least 7.0.9
105136
105137commit c079cce9d884ab03f305b3fba4a4e1247c023480
105138Author: Daniel Ciocea <daniel.ciocea@eosystems.ro>
105139Date: Fri Jun 8 18:12:21 2007 -0700
105140
105141 Fix sync polarity on Samsung SyncMaster 205BW monitor.
105142
105143 need to use standard VESA sync polarity instead of the
105144 EDID provided -hsync -vsync values.
105145
105146commit 9c47b86bd9a4633fda5fd305a09ac8623187efa0
105147Author: Aaron Plattner <aplattner@nvidia.com>
105148Date: Thu Jun 7 13:57:12 2007 -0700
105149
105150 Add new fb symbols to wfbrename.h.
105151
105152 Avoids crashes when wfbComposite calls the wrong image_from_pict.
105153
105154commit 567b5bf765254a4ae9cc7711bb6acfa89a9fd61c
105155Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105156Date: Tue Jun 5 20:26:49 2007 -0400
105157
105158 Delete fbCompositeCopyAreammx()
105159
105160commit 49ed31c0b323dd8c5887a803c199875e6f2330d8
105161Author: Søren Sandmann Pedersen <sandmann@redhat.com>
105162Date: Tue Jun 5 17:44:21 2007 -0400
105163
105164 Remove most of the fast-path MMX operations from fbmmx. fbCopyAreammx
105165
105166 and fbSolidFillmmx are still needed by other code.
105167
105168commit dfbe32b5b828cc4e3da36a0e2e6ad641164eaa5e
105169Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105170Date: Mon Jun 4 18:07:00 2007 -0400
105171
105172 Remove the old Kerberos 5 authentication code.
105173
105174 Before you complain, this code hasn't seen material change since at least
105175 X11R6. It certainly does not build with any modern version of Kerberos.
105176 Anybody wanting krb5 auth to their X server should probably be using
105177 GSSAPI instead of internal krb5 API anyway.
105178
105179commit 793470a8356976ddd427280a738dfb6e1c0e4e70
105180Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
105181Date: Mon Jun 4 12:33:49 2007 -0400
105182
105183 dix: fix null pointer dereference in new resource lookup function.
105184
105185commit 75dece08fb72803d5116e6776e9f1534ff20e37b
105186Author: Zephaniah E. Hull <warp@agamemnon.b5>
105187Date: Mon Jun 4 09:09:20 2007 -0400
105188
105189 xf86PostMotionEvents[P] calls xf86SendDragEvents, xf86SendDragEvents
105190 unconditionally checks device->button->buttonsDown.
105191
105192 Let's make it possible to have a device with motion, but no buttons.
105193
105194 Without segfaulting.
105195
105196commit fbb9b203950e9d0e82574cde5b3e006b0e6b404f
105197Author: Zephaniah E. Hull <warp@agamemnon.b5>
105198Date: Mon Jun 4 06:59:42 2007 -0400
105199
105200 Let's not do a calloc and a free on every call to xf86PostMotionEvents.
105201
105202commit 3f4295e643ca56c40f33af7966e8efd367ef8749
105203Author: Zephaniah E. Hull <warp@agamemnon.b5>
105204Date: Mon Jun 4 06:48:06 2007 -0400
105205
105206 Add xf86PostMotionEventP, takes a pointer instead of a variable number of
105207 arguments.
105208
105209 Bump input ABI to 1.1 since we export this.
105210
105211commit a4f3473c88370b8411e016ebab619cffd33e58f9
105212Author: Zephaniah E. Hull <warp@agamemnon.b5>
105213Date: Mon Jun 4 06:39:02 2007 -0400
105214
105215 Fully init the AbsoluteClassRec in InitAbsoluteClassDeviceStruct.
105216 (Specificly, we were missing the screen field.)
105217
105218commit 0cbc3a4da2ddb6e4f30f60d2bc7f405d31aa554a
105219Author: Zephaniah E. Hull <warp@agamemnon.b5>
105220Date: Mon Jun 4 02:03:44 2007 -0400
105221
105222 Print the build time as well as the date if we can.
105223
105224commit 9a7aaeb3f6ff79af60fde91cd0575a54ba0b9587
105225Author: Daniel Stone <daniel@fooishbar.org>
105226Date: Mon May 28 13:54:16 2007 +0300
105227
105228 XFree86: Input: Assume core events per default
105229
105230 Assume that a device will be sending core events, unless explicitly
105231 specified otherwise.
105232
105233commit 94361cbba7f866144691f6f5e9251a550e0e0cb8
105234Author: Daniel Stone <daniel@fooishbar.org>
105235Date: Mon May 28 13:54:47 2007 +0300
105236
105237 XFree86: Input: Perform case-insensitive comparisons on option names
105238
105239commit e5ce982381c4092252d6b55fcefcc9a3cd21e656
105240Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
105241Date: Sun Jun 3 09:40:37 2007 +1000
105242
105243 Include pixman.h from fb.h or compile of some files will fail
105244
105245 Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
105246
105247commit 90eb22656c34d2d08a8dccaf05e6d081c56bd7f3
105248Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105249Date: Sat Jun 2 16:49:26 2007 -0400
105250
105251 Minor cleanup/robustification to config parsing.
105252
105253commit f6a983533bdc84752562ef0be25b320678bf08a1
105254Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105255Date: Sat Jun 2 16:37:39 2007 -0400
105256
105257 Don't warn about default behaviour when autoconfigging.
105258
105259commit 21e8f4eb02842f877336db08c332d8ee4a381ee0
105260Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105261Date: Sat Jun 2 16:13:01 2007 -0400
105262
105263 Don't print lack of DRI support as an error in AIGLX init.
105264
105265commit 0e1384d8318637f75d04d3d1b7600f7cad40117e
105266Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105267Date: Sat Jun 2 16:07:20 2007 -0400
105268
105269 Delete VDIF support; it was never used anyway.
105270
105271commit 66702f3c1c6c884e83744c72da173cc32f22b2f4
105272Author: Henry Zhao <henryz@localhost.localdomain>
105273Date: Fri Jun 1 23:55:40 2007 -0700
105274
105275 Need to use minPitch in miScanLineWidth() to get the shrinked
105276 linePitch.
105277
105278commit 26b21157cf934ae387b15faa9ebb060120e6a0d6
105279Author: Peter Hutterer <peter@cs.unisa.edu.au>
105280Date: Thu May 31 17:05:14 2007 +0930
105281
105282 Add a deviceMask to the GrabRec and don't interfere with passiveGrabs.
105283
105284 This quickfixes event delivery problems with XI events when a grab was on.
105285 deviceMask is only used when the grab was from a ButtonPress to preserve
105286 potential XI event masks.
105287
105288 This is not an ideal solution but it works until I have time to work on
105289 PassiveGrabs.
105290
105291commit fa877d7ff25c4ec45288e1fea70d4f5e1baf3ef3
105292Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
105293Date: Wed May 30 13:06:45 2007 +0100
105294
105295 Fix mode validation against the maximum X/Y values configured
105296 at server startup, and not against the virtual X/Y parameters
105297 as they can change.
105298
105299 This fixes an issue when canGrow is TRUE and modes get dropped
105300 when using the virtual X/Y parameters.
105301
105302commit 182ab3a3d5b28daa86e6d6155e76ce759687ae6d
105303Author: Peter Hutterer <peter@cs.unisa.edu.au>
105304Date: Wed May 30 17:44:31 2007 +0930
105305
105306 Fix up comments for event.c
105307
105308commit b5db863945fa8045995b3bf742c44e40f2650f04
105309Author: Paulo Ricardo Zanoni <prz05@c3sl.ufpr.br>
105310Date: Wed May 30 16:15:15 2007 +0930
105311
105312 Allow value None for windows when setting ClientPointer.
105313
105314 We need this for clients that need to set the ClientPointer but don't have a
105315 window on display yet. If used, it will set the device as the ClientPointer
105316 for the requesting client.
105317
105318commit 99eae8bea6724a24477375ad5b2d31cc4883cf6b
105319Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
105320Date: Tue May 29 22:04:36 2007 -0400
105321
105322 I/O enable/disable update for the Hurd
105323
105324commit 3c6f1428489c1f71acd41066ea73ef4ae7c60f17
105325Author: Julien Cristau <jcristau@debian.org>
105326Date: Tue May 29 22:01:30 2007 -0400
105327
105328 Make sure that the ramdac symbols are present in the server
105329
105330 The former ramdac module is now built into the server, so its symbols need to
105331 be explicitly exported to drivers (Debian #423129).
105332
105333commit ee20c481eede0954f4a8bef5113979b101863c32
105334Author: Matthieu Herrb <matthieu@deville.herrb.com>
105335Date: Tue May 29 14:54:27 2007 -0600
105336
105337 Remove wscons keyboard handling stuff that doesn't belong there anymore.
105338
105339commit 60de6c7ef9bdcee043f63e8e0d493e6feba6a9d0
105340Merge: 3a6549a 2f13b7c
105341Author: Matthieu Herrb <matthieu@deville.herrb.com>
105342Date: Tue May 29 12:14:49 2007 -0600
105343
105344 Merge branch 'master' of ssh://herrb@git.freedesktop.org/git/xorg/xserver
105345
105346commit 3a6549a163aba26bf4ac58b050c493fba0df14c6
105347Author: Matthieu Herrb <matthieu@deville.herrb.com>
105348Date: Tue May 29 12:14:23 2007 -0600
105349
105350 Make this build on OpenBSD
105351
105352commit 2f13b7c113c17239e382dd3640e9c29201d8ab1f
105353Author: Drew Parsons <dparsons@debian.org>
105354Date: Wed May 30 02:13:36 2007 +1000
105355
105356 Update Xprint build for pixman.
105357
105358 Xprt links libfb, which now uses pixman. Update configure.ac to
105359 require module $PIXMAN for XPRINT.
105360 Also, use $(top_builddir) to reference libfb.la and other local
105361 libraries, rather than using the relative reference ../..
105362
105363commit ba0b7d47ab0c24d5a29228f8af583044060464bd
105364Author: David Nusinow <dnusinow@debian.org>
105365Date: Mon May 28 21:57:04 2007 -0400
105366
105367 Fix for GNU/kFreeBSD
105368
105369commit 2267bf48b385c93243e26c3bb84ebb04c7fdb39f
105370Author: Bastian Blank <waldi@debian.org>
105371Date: Mon May 28 21:55:05 2007 -0400
105372
105373 Fixes for s390
105374
105375commit 857ddbb660a21cad1c16f4fb2dc8a904d6655304
105376Author: Eugene Konev <ejka@imfi.kspu.ru>
105377Date: Mon May 28 21:53:02 2007 -0400
105378
105379 Allow configurable serverconfigdir for security policy location
105380 Allow the location of the SERVERCONFIGdir variable to be defined at
105381 compile-time. This allows us to specify where the security policy will be
105382 located (Debian uses this to put it in /etc). The default is to the
105383 previous location.
105384
105385commit 78d01d1008973899d931ef44b47d5f0b5f220b0d
105386Author: Gerhard Tonn <gt@debian.org>
105387Date: Mon May 28 21:48:58 2007 -0400
105388
105389 Miscellaneous fixes for S/390.
105390
105391commit d98bd4bf908c2c51fcfd3a4c3230de17f2567244
105392Author: Branden Robinson <branden@debian.org>
105393Date: Mon May 28 21:44:59 2007 -0400
105394
105395 Overhaul xorg.conf manpage
105396
105397 Major stylistic cleanups, greatly expanded cross-reference ("SEE ALSO")
105398 section and some typo fixes.
105399
105400 This patch by Branden Robinson. Forward-ported by Fabio M. Di Nitto.
105401
105402commit 6bf8d5019313ee2251a44dfb7ad3435a3c6db7eb
105403Author: David Nusinow <dnusinow@debian.org>
105404Date: Mon May 28 21:42:10 2007 -0400
105405
105406 Read ROM in chunks
105407 This patch speeds up reads of the ROM by reading in large chunks rather
105408 than one byte at a time. This patch was by Dann Frazier.
105409
105410commit 6fdd134a0c3e6fdde9b089100e8783705c9cc6ac
105411Author: David Nusinow <dnusinow@debian.org>
105412Date: Mon May 28 21:39:12 2007 -0400
105413
105414 Fix up xnest manpage
105415 I believe this patch was originally by Branden Robinson
105416
105417commit 6a870992d81a6bacfa9d313c15784fdb281d474f
105418Author: Keith Packard <keithp@dulcimer.keithp.com>
105419Date: Fri May 25 20:33:08 2007 -0700
105420
105421 xf86XVFillKeyHelper assumed root clip never changed.
105422
105423 When the root window changed size, xf86XVFillKeyHelper would not revalidate
105424 the GC, leaving the clip at the old size causing lossage (and possibly
105425 memory corruption if the screen and frame buffer shrank).
105426
105427 Fixed by just using a scratch GC; saving memory, eliminating bugs and
105428 shrinking the code.
105429
105430commit 3c982bc1a49509dda7bc469b0eced44df02755b3
105431Author: Luo Jie <luojie@nlsde.buaa.edu.cn>
105432Date: Thu May 24 11:13:03 2007 -0700
105433
105434 Reinstate an apparently mis-deleted ';' from a for loop with no body.
105435
105436 Fixes an error returning "No core keyboard" with multiple keyboards.
105437
105438commit 4d7469f75fadfc4a59664e88e18eb304203670f4
105439Author: Luo Jie <luojie@nlsde.buaa.edu.cn>
105440Date: Thu May 24 11:04:06 2007 -0700
105441
105442 Fix a typo in using memcpy in xwin.
105443
105444commit 0b988450462ddb005311e68502357baf272e6371
105445Author: Luo Jie <luojie@nlsde.buaa.edu.cn>
105446Date: Thu May 24 11:02:28 2007 -0700
105447
105448 Fix os/utils.c compile with mingw.
105449
105450commit 1f48995d66c0072caa7e5ce2845be642221dd56d
105451Author: Luo Jie <luojie@nlsde.buaa.edu.cn>
105452Date: Thu May 24 11:01:15 2007 -0700
105453
105454 Fix build of composite, dix, and randr when Xinerama is disabled.
105455
105456commit 8f98be7db303bc3db650054efb86843c70114451
105457Author: Eric Anholt <eric@anholt.net>
105458Date: Thu May 24 11:00:04 2007 -0700
105459
105460 Fix bswap detection on BSD (mis-added '_' in function names).
105461
105462commit 9616a042855399f0ee9c6489ea824621ea5fee18
105463Author: Matthias Drochner <drochner@netbsd.org>
105464Date: Tue Apr 10 16:15:40 2007 -0700
105465
105466 Fix build on NetBSD/amd64.
105467
105468commit ddb26bccd275f4fc011f7a2be685d1ce58555a00
105469Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
105470Date: Thu May 24 12:20:24 2007 -0400
105471
105472 dix: Add a bunch of new access codes.
105473
105474 These were determined through an analysis of the core protocol and 35 of
105475 the most common protocol extensions. There remain four bits for future
105476 use.
105477
105478commit 649e7f82d8d4333443493056b81eb20d6cf022bc
105479Author: Michel Dänzer <michel@tungstengraphics.com>
105480Date: Thu May 24 12:10:05 2007 +0200
105481
105482 Consolidate portPriv->pDraw assignments into xf86XVEnlistPortInWindow.
105483
105484 This avoids a crash in xf86XVReputVideo and also cleans up the code slightly.
105485
105486commit 047bf3349bb697c73c95729a8bbf15f72605901f
105487Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105488Date: Wed May 23 16:56:05 2007 -0400
105489
105490 Delete trapezoid rendering code; replace with pixman calls
105491
105492commit 9d87ef4e0dff40ea39f1b209c67b90079fc79065
105493Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105494Date: Wed May 23 15:50:25 2007 -0400
105495
105496 - Make image_from_pict() non-static
105497 - Delete fbedge.c and fbedgeimp.h
105498 - Use pixman_rasterize_edges() in fbtrap.c
105499
105500commit 2a960c442bd7560630f52b55d82ec0517542ee5a
105501Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105502Date: Wed May 23 13:08:26 2007 -0400
105503
105504 Port renderedge.c to pixman
105505
105506commit 3ba3ede9bbdfc6376b6f6e0b6ce8280a05e6584d
105507Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105508Date: Wed May 23 12:56:04 2007 -0400
105509
105510 Add missing offsets for window coordinates - reported by Colin Harrison
105511
105512commit 2a4aa63a23ddd816b647b851a01865861827a7eb
105513Merge: 9cee4ec cc648e6
105514Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
105515Date: Tue May 22 14:50:26 2007 -0400
105516
105517 Merge branch 'master' into XACE-SELINUX
105518
105519 Conflicts:
105520
105521 include/miscstruct.h
105522
105523commit cc648e609d472472bac4a2e568eb3598b3690ba3
105524Author: Michel Dänzer <michel@tungstengraphics.com>
105525Date: Tue May 22 10:51:56 2007 +0200
105526
105527 EXA: Export ExaOffscreenMarkUsed.
105528
105529 Can be used to inform EXA that an offscreen area is used outside of EXA.
105530
105531commit e6a7198e7cd96f1fe0654cc6811a977821579258
105532Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105533Date: Tue May 22 10:51:55 2007 +0200
105534
105535 Bug #8991: Add glXGetDrawableAttributes dispatch; fix texture format therein.
105536
105537 Adapted to master branch by Michel Dänzer <michel@tungstengraphics.com>.
105538
105539commit 6324bfc468f7a645d2fee59f1c921a4328a4639f
105540Author: Michel Dänzer <michel@tungstengraphics.com>
105541Date: Tue May 22 10:51:53 2007 +0200
105542
105543 AIGLX: Zero-copy texture-from-pixmap.
105544
105545 When available, use the 2D driver texOffsetStart hook and the 3D driver
105546 setTexOffset hook to save the overhead of passing the pixmap data to
105547 glTex(Sub)Image.
105548
105549 The basic idea is to update the driver specific 'offset' for bound pixmaps
105550 before dispatching a GLX render request and to flush immediately afterwards
105551 if there are any pixmaps bound. This should ensure that the 3D driver can
105552 use pixmaps for texturing directly regardless of the X server moving them
105553 around.
105554
105555commit 5006d08d7fc56d3d380cc6b75297f94e8594eb54
105556Author: Michel Dänzer <michel@tungstengraphics.com>
105557Date: Tue May 22 10:51:52 2007 +0200
105558
105559 DRI: Add TexOffset driver hooks.
105560
105561 To be used by AIGLX for GLX_EXT_texture_from_pixmap without several data copies.
105562
105563 The texOffsetStart hook must make sure that the given pixmap is accessible by
105564 the GPU for texturing and return an 'offset' that can be used by the 3D
105565 driver for that purpose.
105566
105567 The texOffsetFinish hook is called when the pixmap is no longer being used for
105568 texturing.
105569
105570commit ff2eae86b6a8760befbbc5d605debebe7b024c05
105571Author: David Nusinow <dnusinow@debian.org>
105572Date: Mon May 21 19:50:04 2007 -0400
105573
105574 Fix boolean thinko that prevented working without a server layout
105575
105576commit 56fd92715567cd32e4b725b3791de9ac4e3879aa
105577Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105578Date: Mon May 21 20:00:25 2007 -0400
105579
105580 Remove fast path code from fbpict.c
105581
105582 Remove the various fast path functions from fbpict, and instead use
105583 pixman_image_composite().
105584
105585commit d0e8f474099dea40bbea555427772724ccb787d1
105586Author: Peter Hutterer <peter@cs.unisa.edu.au>
105587Date: Mon May 21 13:06:21 2007 +0930
105588
105589 Change GrabDevice's parameter name back to "coreGrab".
105590 Was changed during a global search/replace for
105591 5c680e94938c16e04a4349715cf11796b80400db.
105592
105593 Otherwise people like me introduce bugs (e.g. in
105594 333bab4b235801efd2b2de5b60df5b0c44048d3e)
105595
105596commit 167e1773de0fb566559d4362ff929eedcdb6d055
105597Author: Peter Hutterer <peter@cs.unisa.edu.au>
105598Date: Mon May 21 10:45:10 2007 +0930
105599
105600 Remove unused include "gestr.h"
105601
105602commit 7e2c935920cafadbd87c351f1a3239932864fb90
105603Author: Fredrik Höglund <fredrik@kde.org>
105604Date: Fri May 18 20:06:14 2007 +0200
105605
105606 Add a new IDLETIME system sync counter.
105607
105608 This counter exposes the time in milliseconds since the last
105609 input event. Clients such as screen savers and power managers
105610 can set an alarm on this counter to find out when the idle time
105611 reaches a certain value, without having to poll the server.
105612
105613commit 756acea23a0cc56c470bcd77c6f5638d923ab3d1
105614Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105615Date: Fri May 18 13:39:12 2007 -0400
105616
105617 Use pixman_image_set_indexed() to make 8 bit work
105618
105619commit 7916419a0092b8bf9713c0840f9e969950d7aa85
105620Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105621Date: Fri May 18 11:58:24 2007 -0400
105622
105623 Comment out setup of general MMX code
105624
105625commit 998164bac648756e5b5254aa36e075ae360d3972
105626Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105627Date: Fri May 18 11:36:20 2007 -0400
105628
105629 Move fbCompositeGeneral() to fbpict.c and remove fbcompose.c
105630
105631commit a2e3614eb8f0fa198615df492b03ff36bc9c1121
105632Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105633Date: Fri May 18 11:33:11 2007 -0400
105634
105635 Break image_from_pict() into a few subfunctions.
105636
105637commit c5ef84c325440af5fbdf9f44c3781d99a0392df9
105638Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105639Date: Thu May 17 21:31:08 2007 -0400
105640
105641 Make the general compositing code create a pixman image and call
105642
105643 pixman_image_composite(). Leave the general code commented out for now.
105644
105645commit 076d070e186afeb416976ae74fbfd50c86db10c5
105646Author: Keith Packard <keithp@neko.keithp.com>
105647Date: Thu May 17 20:24:18 2007 -0700
105648
105649 Use Screen block handler for rotation to draw under DRI lock.
105650
105651 DRI uses a non-screen block/wakeup handler which will be executed after the
105652 screen block handler finishes. To ensure that the rotation block handler is
105653 executed under the DRI lock, dynamically wrap the screen block handler for
105654 rotation.
105655
105656commit 915563eba530c5e2fdc2456cf1c7c3cc09b3add0
105657Author: Keith Packard <keithp@neko.keithp.com>
105658Date: Thu May 17 20:22:43 2007 -0700
105659
105660 Disable all outputs and crtcs at startup.
105661
105662 Leaving devices enabled during server startup can cause problems during the
105663 initial mode setting in the server, especially when they are used for
105664 different purposes by the X server than by the BIOS. Disabling all of them
105665 before any mode setting is attempted provides a stable base upon which the
105666 remaining mode setting operations can be built.
105667
105668commit 0375009a97c2ab7f0e0f0265463d45c0580388c6
105669Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105670Date: Thu May 17 12:59:24 2007 -0400
105671
105672 Remove excessive unrolling in fbCompositeSrc_x888x8x8888() and fix bug where
105673 the source alpha was used instead of 0xff.
105674
105675commit 0b4db74922299df785e6273fdb1bf65c38d36070
105676Author: Peter Hutterer <peter@cs.unisa.edu.au>
105677Date: Thu May 17 18:00:07 2007 +0930
105678
105679 Squish the flags into the upper bits of the mode field for EnterLeave events.
105680 This way we have enough space for the detail field.
105681
105682commit 333bab4b235801efd2b2de5b60df5b0c44048d3e
105683Author: Peter Hutterer <peter@cs.unisa.edu.au>
105684Date: Thu May 17 17:59:02 2007 +0930
105685
105686 Clean up, correct some comments.
105687 Send event type down with the RawEvents.
105688
105689commit bc334286b060bc8d0c829b18acebadf24fbdaf19
105690Author: Peter Hutterer <peter@cs.unisa.edu.au>
105691Date: Thu May 17 10:36:46 2007 +0930
105692
105693 Init device axis with -1,-1 minimum values.
105694
105695 This is needed to be able to cross to screens located east/north of the
105696 current active screen.
105697
105698commit 546465ee6aa6584780aec6357f32d205c807ae71
105699Author: Soren Sandmann Pedersen <sandmann@redhat.com>
105700Date: Wed May 16 17:42:04 2007 -0400
105701
105702 Make fbFetch_b8g8r8() actually write the read value to the buffer
105703
105704commit 0fcd17c9181901c419cc32bc24c07fe5a6934d81
105705Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105706Date: Tue May 15 17:59:13 2007 -0400
105707
105708 Use pixman short formats, revert the gradient_stop change
105709
105710commit f4c1d5fc28a5a7fe2592505350f9e2331f6049b7
105711Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105712Date: Tue May 15 17:12:22 2007 -0400
105713
105714 Use pixman types for transforms and vectors
105715
105716commit f2e30e7d0a1d075e7e83c5b5ceca9e4752951138
105717Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105718Date: Tue May 15 16:51:21 2007 -0400
105719
105720 Use the pixman fixed point types and macros
105721
105722commit 3da842bf930d7875599ca0c06cb4a09cfa987ac5
105723Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105724Date: Tue May 15 14:57:14 2007 -0400
105725
105726 Revert various fast path functions to their pre-pixman-merge state
105727 since they fail rendercheck. Remove their associated macros.
105728
105729 See bug 10903.
105730
105731commit 1568b6b6a0d7337f29c7b87cc46ae64b3b0f8fdf
105732Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105733Date: Sat May 12 20:33:23 2007 -0400
105734
105735 Port large amounts of the region code to pixman
105736
105737commit dde0ceac4ea7639d0096bfd26f37c5851778854c
105738Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105739Date: Sat May 12 17:41:47 2007 -0400
105740
105741 Add new InitRegions() function called from dix/main
105742
105743commit e037052ac522150786abf44d3a04c813cc490050
105744Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105745Date: Sat May 12 16:58:54 2007 -0400
105746
105747 Turn boxes and regions into typedefs for pixman types
105748
105749commit 8e56f5be4b70773c899f01b9ccd2e88d523327e4
105750Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105751Date: Fri May 11 11:45:37 2007 -0400
105752
105753 Add dependency on pixman 0.9.0
105754
105755commit a8c56372ba8aa36bac09877c478ff53ea5358df7
105756Author: Peter Hutterer <peter@cs.unisa.edu.au>
105757Date: Tue May 15 10:24:22 2007 +0930
105758
105759 Revert "Allow events to grabWindows event if the device is not grabbed."
105760
105761 This reverts commit 339b73e710a0920608a3fbcb20b406f0f6c4e0f6.
105762 Allowing those events can lead to race conditions.
105763
105764 Conflicts:
105765
105766 dix/events.c
105767
105768commit 81fc6a128b64ad412064405ed45db0175398e3f0
105769Author: Peter Hutterer <peter@cs.unisa.edu.au>
105770Date: Mon May 14 18:12:56 2007 +0930
105771
105772 Add ungrab support to ExtendedGrabDeviceRequest handling.
105773 Polish the code a bit.
105774
105775commit 5c680e94938c16e04a4349715cf11796b80400db
105776Author: Peter Hutterer <peter@cs.unisa.edu.au>
105777Date: Mon May 14 12:34:53 2007 +0930
105778
105779 Remove double-grab option. We only want one grab per device.
105780
105781commit 4dc973f08c54294e06bc831bd2839fd3aa24ecfb
105782Author: Peter Hutterer <peter@cs.unisa.edu.au>
105783Date: Mon May 14 10:42:03 2007 +0930
105784
105785 Add ExtendedGrabDevice handling.
105786 Add XGE handling in DeliverGrabbedEvent.
105787
105788 We can now grab something selecting XGE events, but the current code is a
105789 bit messy and doesn't work too well yet.
105790
105791commit a277f04ab08514462b7f10b4dd92eb326af85501
105792Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105793Date: Wed May 9 22:03:12 2007 -0400
105794
105795 Remove mfb and cfb from include paths where they're not needed.
105796
105797commit 20c5250e487e032d392e2e4624021fccb1bfb72c
105798Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105799Date: Wed May 9 21:49:44 2007 -0400
105800
105801 Use system copy of cbrt() if available.
105802
105803 Also move the replacement inline into miarc.c, since that's the only user.
105804
105805commit 8dcc37520d5e8c8b52cee81faa67fd5205548377
105806Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105807Date: Wed May 9 18:57:05 2007 -0400
105808
105809 Use _X_INLINE instead of ad-hoc #defines.
105810
105811commit 6ff239cb4e67c0a2ea497a1714e5585c1d941af3
105812Author: Adam Jackson <ajax@benzedrine.nwnk.net>
105813Date: Wed May 9 18:38:33 2007 -0400
105814
105815 Make the use of ICEIL slightly less ugly.
105816
105817commit 178d426311bb3c7160f72b5d95b0a137eda09ba9
105818Author: Colin Harrison <colin.harrison-at-virgin.net>
105819Date: Fri May 11 10:08:42 2007 +0100
105820
105821 Missing piece from bug 9808
105822
105823commit ebaa6c920c82401952a0ccc991b94574306449bd
105824Author: Matthias Hopf <mhopf@suse.de>
105825Date: Thu May 10 15:25:31 2007 +0200
105826
105827 Disable Simba PCI bridge routing code (Bug #8020).
105828
105829 The code in hw/xfree86/os-support/bus/sparcPci.c:simbaCheckBus()
105830 is trying to mimmick VGA routing by disabling I/O space responses
105831 behind the Simba PCI-PCI controller.
105832
105833 Unfortunately, doing this also happens to disable access to the
105834 IDE controller I/O space registers, thus crashing the system. The
105835 granularity of the I/O disabling in the Simba controller is not
105836 fine enough to disable VGA without also disabling the IDE controller
105837 registers.
105838
105839commit 86c4941727f7c673ae6bb88c67443fa25935c7f5
105840Author: Colin Harrison <colin.harrison-at-virgin.net>
105841Date: Wed May 9 16:55:27 2007 +0100
105842
105843 fix an occasional crash in GetWindowName() (bug: 9798)
105844
105845commit be44018a3c6172caf3e91c36ea321420d104e79f
105846Author: Colin Harrison <colin.harrison-at-virgin.net>
105847Date: Wed May 9 16:55:09 2007 +0100
105848
105849 Fix bad use of hwnd (bug: 9808)
105850
105851commit d3248b66a650c6c629cd66240e25004869217d2e
105852Author: Colin Harrison <colin.harrison-at-virgin.net>
105853Date: Wed May 9 16:54:46 2007 +0100
105854
105855 Migrate some code to the new mi apis
105856
105857commit 021e5df85d7c9373a2fed55512751d16e08128db
105858Author: David Nusinow <dnusinow@debian.org>
105859Date: Mon May 7 21:03:40 2007 -0400
105860
105861 Add more informative logging for module default loading
105862
105863 When the modules section is parsed, if a module is set to be loaded by
105864 default, this will be logged. If it is redundantly specified in xorg.conf,
105865 this will also be noted. None of this logging will happen if the xorg.conf
105866 lacks a modules section.
105867
105868commit 1b3a0508a7aee1c7b14cd62216b4727fcc9181d4
105869Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com>
105870Date: Sun May 6 01:30:59 2007 -0700
105871
105872 Fix documentation for Copy hook -- it can copy memory to the scanout
105873 buffer too.
105874
105875commit 030a578391c634bc68add6ada3f251cf3f8c3069
105876Author: David Nusinow <dnusinow@debian.org>
105877Date: Thu May 3 22:51:07 2007 -0400
105878
105879 Provide UseDefaultFontPath option
105880
105881 This provides a new option, UseDefaultFontPath. This option is enabled by
105882 default, and causes the X server to always append the default font path
105883 (defined at compile time) to the font path for the server. This will allow
105884 people to specify additional font paths if they want without breaking
105885 their font path, thus hopefully avoiding ye olde "fixed front" problem.
105886
105887 Because this option is a ServerFlag option, the ServerFlags need to be
105888 processed before the files section of the config file, so swap the order
105889 that they are processed.
105890
105891commit e91b9ddc7aa95abc2d4d314e8db204860771a099
105892Author: David Nusinow <dnusinow@debian.org>
105893Date: Thu May 3 22:00:23 2007 -0400
105894
105895 Improve modules loading defaults
105896
105897 Provide default modules that may be overrided easily. Previously the
105898 server would load a set of default modules, but only if none were
105899 specified in the xorg.conf, or if you didn't have a xorg.conf at all. This
105900 patch provides a default set and you can add only the "Load" instructions
105901 to xorg.conf that you want without losing the defaults. Similarly, if you
105902 don't want to load a module that's loaded by default, you can add "Disable
105903 modulename" to your xorg.conf (see man xorg.conf in this release for
105904 details). This allows for a minimal "Modules" section, where the user only
105905 need specify what they want to be different. See bug #10541 for more.
105906
105907 The list of default modules is taken from the set loaded by default when
105908 there was a xorg.conf containing no "Modules" section.
105909
105910 A potential problem for some users is that some users disable a module,
105911 most notably DRI, by commenting out the "Load" line in their xorg.conf.
105912 This needs to be changed to an uncommented "Disable" line, as DRI is
105913 loaded by default.
105914
105915commit d2f813f7db157fc83abc4b3726821c36ee7e40b1
105916Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105917Date: Wed May 2 19:10:22 2007 -0400
105918
105919 New fbWalkCompositeRegion() function
105920
105921 This new function walks the composite region and calls a rectangle
105922 compositing function on each compositing rectangle. Previously there
105923 were buggy duplicates of this code in fbcompose.c and
105924 miext/rootles/safealpha/safeAlphaPicture.c.
105925
105926commit e0959adcd8df2c61e98e76e708fceef9c7cd54eb
105927Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
105928Date: Tue May 1 13:41:48 2007 -0400
105929
105930 Add fbCompositeRect() as another special case in the switch of doom in fbpict.c
105931
105932 This is phase one of getting the two region walkers in fbcompose.c and
105933 fbpict.c merged together.
105934
105935commit cb22bdbe678a9948eda75d464d6701a729664ef0
105936Author: Peter Hutterer <peter@cs.unisa.edu.au>
105937Date: Thu May 3 10:05:58 2007 +0930
105938
105939 Add selected RamDac functions to xfree86LookupTab.
105940
105941 Adding those makes drivers for s3 and tga work again.
105942
105943commit c1e1d6b98a6708860e5b5f6e21d8d5b1d8ce9075
105944Author: Brian <brian@yutani.localnet.net>
105945Date: Wed May 2 15:55:40 2007 -0600
105946
105947 In __glXCreateARGBConfig(), insert the new GL mode at the _end_ of the linked list.
105948
105949 Previously, the new mode was added at the head of the list. This caused the
105950 positional correspondence between modes and the XMesaVisuals array to be off
105951 by one. The net result was GLX clients failing when they tried to use the
105952 last GLX mode/visual.
105953
105954 We still have the problem of DRI drivers not being able to use the extra
105955 mode/visual introduced by __glXCreateARGBConfig(). glXCreateContext fails
105956 with BadAlloc if it's attempted. This is also the source of the often-
105957 seen warning "libGL warning: 3D driver claims to not support visual xxx"
105958 Look into fixing that someday...
105959
105960commit bd0abb2844ef9faf28703e592cfebb886004234c
105961Author: Tilman Sauerbeck <tilman@code-monkey.de>
105962Date: Wed May 2 17:20:48 2007 +0200
105963
105964 Bug #10823: Fixed default OSNAME value.
105965
105966 We try to get OSNAME from uname by default now.
105967
105968commit 71fc5b3e9309182978ead676965d65ca93a4e3b9
105969Author: Keith Packard <keithp@keithp.com>
105970Date: Wed May 2 11:41:11 2007 +0200
105971
105972 Fix for a divide by zero that can be triggered by a malicious client.
105973
105974 Problem reported by Derek Abdine of rapid7.com. Thanks.
105975
105976commit c03d9a7217f9895feed8cfed3ede4752faf6187c
105977Author: Peter Hutterer <peter@cs.unisa.edu.au>
105978Date: Wed May 2 18:16:40 2007 +0930
105979
105980 Cleaning up a bit.
105981
105982 Register correct event in EventSwapVector.
105983 Fix up event swap for GE events, register XI's swap function at GE.
105984
105985commit e6fd4a24ebd205013b41e44aacbbfb847709d2fd
105986Author: Peter Hutterer <peter@cs.unisa.edu.au>
105987Date: Wed May 2 17:49:20 2007 +0930
105988
105989 Add handling for FakeDeviceEvent request.
105990 Fix a stupid bug from last commit, mask names were wrong.
105991
105992commit af25720598bb35d9d953ac8d9a07528289ebfb4a
105993Author: Peter Hutterer <peter@cs.unisa.edu.au>
105994Date: Wed May 2 13:51:16 2007 +0930
105995
105996 Add missing files for ProcXiSelectEvent.
105997
105998 Should have been a part of 5e439109292e54b5c4d1a7bc7b6ac0e42ee285f7. Oh well.
105999
106000commit 58c0fd29272fb6c2f193cff82fb0b0573c607eec
106001Author: Peter Hutterer <peter@cs.unisa.edu.au>
106002Date: Wed May 2 13:50:09 2007 +0930
106003
106004 Adjust EQ to support long events.
106005 Add RawDeviceEvent (pointers only for now).
106006
106007 This commit changes the event queue to use EventLists instead of xEvent
106008 arrays. Only EQ is affected, event delivery still uses xEvent* (look for
106009 comment in mieqProcessInputEvent).
106010
106011 RawDeviceEvents deliver driver information to the client, without clipping or
106012 acceleration.
106013
106014commit 0214d0b96a9b900a8cf5c7f2183add3a411441f3
106015Author: Peter Hutterer <peter@cs.unisa.edu.au>
106016Date: Tue May 1 15:49:41 2007 +0930
106017
106018 Start devices after windows are initialized.
106019
106020 This fixes the hotplug segfault introduced with the multihead changes and
106021 cleans up the code a bit as well.
106022
106023commit 11d0e2109b699714055e263e8963f2c39eded28b
106024Author: Peter Hutterer <peter@cs.unisa.edu.au>
106025Date: Tue May 1 14:16:36 2007 +0930
106026
106027 Correct enterleave semaphore value for root window.
106028
106029commit 8b42ba64e169a35df2a90528dcb0cd06be0f67a1
106030Author: Peter Hutterer <peter@cs.unisa.edu.au>
106031Date: Tue May 1 12:12:58 2007 +0930
106032
106033 Send PointerKeyboardPairingChangedNotifyEvent using XGE.
106034
106035 Needs updated inputproto and libXi.
106036
106037commit 325380adb20ec2b82e176d75599fd4bc97a3b918
106038Author: Peter Hutterer <peter@cs.unisa.edu.au>
106039Date: Tue May 1 11:02:05 2007 +0930
106040
106041 Check and re-set paired devices when initializing sprites.
106042
106043 If we don't do this, a device that is paired before a sprite has been
106044 initialized for the paired device will not actually get the right sprite and
106045 segfault the server on focus events. Happens for the VCK.
106046
106047commit b043a184509cfe5c4c4691ecde1a4a065e53d575
106048Author: Peter Hutterer <peter@cs.unisa.edu.au>
106049Date: Tue May 1 09:54:33 2007 +0930
106050
106051 Check VCP's and VCK's ids when searching for a new device id.
106052
106053 Both VCP and VCK are not in the inputInfo.devices list anymore, so we need to
106054 check them separately. If we don't do this, we end up re-using ids 0 and 1,
106055 causing all sorts of grief for clients.
106056
106057commit 873ef75b1e8c94d39670c981c4d830ab8bcc018b
106058Author: Colin Guthrie <gmane@colin.guthr.ie>
106059Date: Mon Apr 30 10:33:12 2007 -0600
106060
106061 fix __glXErrorCallBack() proto
106062
106063commit 6b33459bf5aac23c3ecc7002d091c02f327d907a
106064Merge: 18252a5 3c91a99
106065Author: Brian <brian@yutani.localnet.net>
106066Date: Mon Apr 30 10:26:19 2007 -0600
106067
106068 Merge branch 'master' of git+ssh://brianp@git.freedesktop.org/git/xorg/xserver
106069
106070commit 5e439109292e54b5c4d1a7bc7b6ac0e42ee285f7
106071Author: Peter Hutterer <peter@cs.unisa.edu.au>
106072Date: Mon Apr 30 12:57:42 2007 +0930
106073
106074 Add GenericEvent extension to Xext.
106075
106076 This adds (unconditional) support for the GE extension. Anything from now on
106077 that sends events in MPX will have to use the GE extension. No GE, no MPX
106078 events. GE is not actually used yet from anywhere with this commit.
106079
106080 You will need to update x11proto, xextproto, libX11, libXext and xcb to the
106081 matching xge branches. Things will _NOT_ work without the updated protocol
106082 headers and libraries.
106083
106084commit 3c91a993e8c752002adf85c317216e1487c20780
106085Author: Michel Dänzer <michel@tungstengraphics.com>
106086Date: Sun Apr 29 23:49:41 2007 +0200
106087
106088 EXA: Fix OffscreenValidate build with DEBUG_OFFSCREEN enabled.
106089
106090commit a261e1325057974d58440812b93c00c0caa4423a
106091Author: Michel Dänzer <michel@tungstengraphics.com>
106092Date: Sun Apr 29 23:49:35 2007 +0200
106093
106094 EXA: Remove DrawableDirty.
106095
106096 Convert the remaining callers to PixmapDirty.
106097
106098commit b1b6674a919943a8ac37e54d02e8d0d23a642b1d
106099Author: Michel Dänzer <michel@tungstengraphics.com>
106100Date: Sun Apr 29 23:49:28 2007 +0200
106101
106102 EXA: FillSpans improvements.
106103
106104 * Don't need to track damage.
106105 * Always migrate for fallbacks.
106106
106107commit 584697a2231ac782f362a925e1489c15483a8791
106108Author: Michel Dänzer <michel@tungstengraphics.com>
106109Date: Sun Apr 29 23:49:09 2007 +0200
106110
106111 EXA: SolidBoxClipped improvements.
106112
106113 * Centralize handling of fallbacks and damage tracking.
106114 * Always migrate for fallbacks.
106115
106116commit 982d7c2c0b948ba04c8eefa475d660981e6ed9f9
106117Author: Michel Dänzer <michel@tungstengraphics.com>
106118Date: Sun Apr 29 23:48:59 2007 +0200
106119
106120 EXA: CopyNtoN improvements.
106121
106122 * Centralize handling of fallbacks and damage tracking.
106123 * Always migrate for fallbacks.
106124
106125commit d2245386eed200e77a8c84bdda36ab29e39fd593
106126Author: Michel Dänzer <michel@tungstengraphics.com>
106127Date: Sun Apr 29 23:48:31 2007 +0200
106128
106129 EXA: GetImage improvements.
106130
106131 Only migrate when appropriate. In particular, don't migrate to offscreen in the
106132 no-fallback case as copying from system memory should usually be as fast if not
106133 faster than DownloadFromScreen, in particular if the bits need to be uploaded
106134 to offscreen first.
106135
106136commit 0880aaac9c83019fec2e3d32871f74c7a407f8b3
106137Author: Michel Dänzer <michel@tungstengraphics.com>
106138Date: Sun Apr 29 23:48:19 2007 +0200
106139
106140 EXA: PutImage improvements.
106141
106142 * Migrate for fallbacks when appropriate.
106143 * Add damage tracking in ExaCheckPutImage.
106144
106145commit 7fca16901187ade48e83e6a2684ef464b1912357
106146Author: Michel Dänzer <michel@tungstengraphics.com>
106147Date: Sun Apr 29 23:48:11 2007 +0200
106148
106149 EXA: ImageGlyphBlt improvements.
106150
106151 * Don't waste effort on invisible glyphs.
106152 * Only track damage for bounding box instead of each glyph separately.
106153 * Always migrate for fallbacks.
106154
106155commit a8d6ebdf9338dc2f6ff9a532e6fec460a70d3b1e
106156Author: Michel Dänzer <michel@tungstengraphics.com>
106157Date: Sun Apr 29 23:47:53 2007 +0200
106158
106159 EXA: Defer to FillRegionTiled in Composite when possible.
106160
106161 Committed separately as this case is hard to hit and has only been tested
106162 lightly.
106163
106164commit 81b055605c34b5823f6c5f63cc0f92f43c6b7252
106165Author: Michel Dänzer <michel@tungstengraphics.com>
106166Date: Sun Apr 29 23:47:43 2007 +0200
106167
106168 EXA: Composite improvements.
106169
106170 * Defer to simpler hooks in more cases (inspired by XAA behaviour).
106171 * Move damage tracking from lower to higher level functions.
106172 * Always migrate for fallbacks.
106173
106174commit ce317a5b76c053f449122c46e1372bf8e067cb4c
106175Author: Michel Dänzer <michel@tungstengraphics.com>
106176Date: Sun Apr 29 23:47:16 2007 +0200
106177
106178 EXA: Glyphs improvements.
106179
106180 * Don't waste effort on invisible glyphs.
106181 * Add damage tracking where necessary.
106182 * Always migrate for fallbacks.
106183
106184commit 0c8905ebc91cf654facef84ee52231a358deec5c
106185Author: Michel Dänzer <michel@tungstengraphics.com>
106186Date: Sun Apr 29 23:47:08 2007 +0200
106187
106188 EXA: PolyFillRect improvements.
106189
106190 * Convert rects to region and use it for damage tracking.
106191 * When possible, defer to exaFillRegion{Solid,Tiled} using converted region.
106192 * Always migrate for fallbacks.
106193 * Move damage tracking out of ExaCheckPolyFillRect.
106194
106195commit 567f18a09bfb05f448be40c7ebe0f210f955601c
106196Author: Michel Dänzer <michel@tungstengraphics.com>
106197Date: Sun Apr 29 23:46:49 2007 +0200
106198
106199 EXA: FillRegion{Solid,Tiled} improvements.
106200
106201 * Support planemasks, different ALUs and arbitrary tile origin.
106202 * Leave damage tracking and non-trivial fallbacks to callers.
106203 * Always migrate for fallbacks.
106204
106205 This is in preparation for using these from more other functions.
106206
106207commit e869573b52fac69fb88cea120daaeec59c7a3461
106208Author: Michel Dänzer <michel@tungstengraphics.com>
106209Date: Sun Apr 29 23:45:48 2007 +0200
106210
106211 EXA: exaAssertNotDirty improvements.
106212
106213 * Return early if the valid region is empty or the pixmap is pinned.
106214 * Fix loop for several cliprects.
106215
106216commit d3f8667341bfe6dc7d0258c4ad69377f37d88d95
106217Author: Michel Dänzer <michel@tungstengraphics.com>
106218Date: Sun Apr 29 23:44:27 2007 +0200
106219
106220 EXA: Fix exaEnableDisableFBAccess for nested disables and enables.
106221
106222commit 5e4b3232dafe3b0dec65bf639bebaba4774210b7
106223Author: Michel Dänzer <michel@tungstengraphics.com>
106224Date: Sun Apr 29 23:38:22 2007 +0200
106225
106226 Fix fbCompositeTrans_0888xnx0888 build for wfb on big endian.
106227
106228commit 2866e0bac9b8dd3892c5e68abcfc6c97cebaf88a
106229Author: Michel Dänzer <michel@tungstengraphics.com>
106230Date: Sun Apr 29 23:38:13 2007 +0200
106231
106232 Fix a couple of picture repeat fields incorrectly compared to RepeatNormal.
106233
106234commit 78a20455356ccc310f73cfc65ad65a7677eee7e5
106235Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106236Date: Fri Apr 27 15:20:24 2007 -0400
106237
106238 Pixman merging
106239
106240 More msvc++ porting
106241
106242commit 6c4f1826bf2c5f30f5fe6e489a02b6375478b380
106243Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106244Date: Fri Apr 27 08:13:08 2007 -0400
106245
106246 Bug fix in fbCompositeIn_nx8x8888
106247
106248 Make sure both halves of the dst word is set to zero when the masks
106249 are both 0.
106250
106251commit f28eea0647f007c2e2415ecc6fceef46201faad4
106252Merge: 339b73e ae04f2c
106253Author: Peter Hutterer <peter@cs.unisa.edu.au>
106254Date: Fri Apr 27 16:34:36 2007 +0930
106255
106256 Merge branch 'master' into mpx
106257
106258 Conflicts:
106259
106260 dix/devices.c
106261 dix/events.c
106262
106263commit 339b73e710a0920608a3fbcb20b406f0f6c4e0f6
106264Author: Peter Hutterer <peter@cs.unisa.edu.au>
106265Date: Fri Apr 27 13:24:27 2007 +0930
106266
106267 Allow events to grabWindows event if the device is not grabbed.
106268
106269 This kinda makes popup windows useable if the WM doesn't set the
106270 ClientPointer. Kinda.
106271
106272commit ae04f2cb0a068cdc1e519627bf745de0c9e4a85a
106273Author: Adam Jackson <ajax@benzedrine.nwnk.net>
106274Date: Thu Apr 26 15:28:04 2007 -0400
106275
106276 Fix the 'relink' target for kdrive servers.
106277
106278commit 6c8152d6ee9eeb21a68a8bbfed1540939e5bcd1f
106279Author: Adam Jackson <ajax@benzedrine.nwnk.net>
106280Date: Thu Apr 26 14:59:04 2007 -0400
106281
106282 Remove old edid_modes.c, it lives in hw/xfree86/modes/ now.
106283
106284commit 2208c6087d6bffcb24a30891a56430e28735874c
106285Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106286Date: Thu Apr 26 14:40:30 2007 -0400
106287
106288 Change expand_alpha_rev to expand_alpha in mmxSaturateU
106289
106290commit a300ef84cee26febfbe08c497d0d063588130bdd
106291Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106292Date: Thu Apr 26 14:37:53 2007 -0400
106293
106294 Fix typo in previous commit
106295
106296commit 0ebe48be59368b55c618f60d4656300bd7f52ed9
106297Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106298Date: Thu Apr 26 14:36:32 2007 -0400
106299
106300 Pixman merge
106301
106302 - Changes to support MS Visual C++
106303
106304 - use inline instead of __inline__
106305
106306 - Fix rounding errors (Billy Biggs, from xserver via pixman)
106307
106308commit 4fe918b38553133c27e5ae672e5c43984a9bbaea
106309Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106310Date: Thu Apr 26 12:41:01 2007 -0400
106311
106312 Fix bug 8871 - scrolling corruption with a compositing manager
106313
106314 Call miHandleExposures() in CopyArea/CopyPlane explicitly in cw to
106315 generate GraphicsExposes correctly.
106316
106317commit 0ff7c94fcf6497ee8575f81cf97eeeb3a857739e
106318Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106319Date: Thu Apr 26 10:56:02 2007 -0400
106320
106321 Pixman merge
106322
106323 Make use of fbCompositeSrcAdd_8888x8x8mmx
106324
106325commit 701ccb4a22cfd646ccb7f19b7b3a476aeb5ce2da
106326Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106327Date: Thu Apr 26 10:49:06 2007 -0400
106328
106329 Pixman merge
106330
106331 - Remove stray default label
106332
106333 - Integrate new MMX ops SolidMaskSrc_nx8x8888mmx, In_8x8mmx, and
106334 In_nx8x8mmx
106335
106336 - Formatting changes to reduce diff noise
106337
106338commit a54ef54db19dcd36ed86b33cff2bc369f9690a15
106339Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106340Date: Thu Apr 26 10:24:25 2007 -0400
106341
106342 Pixman merge
106343
106344 Make sure fbCompositeSrc_x888x8x8888mmx and fbCompositeSrc_8888x8x8888mmx
106345 are used when possible.
106346
106347commit 3571b8e65b0857322d12e291305cfe29ea497c3c
106348Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106349Date: Thu Apr 26 09:45:11 2007 -0400
106350
106351 More pixman merging
106352
106353 - Move some code around to minimize diff noise
106354
106355 - Add mmx ops from pixman that never made it into X
106356
106357 - Merge Jeff Muizelaar's bugfixes to fbCompositeSrc_8888x8x8888mmx and
106358 fbCompositeSrc_x888x8x8888mmx.
106359
106360commit cfc01115af4136b2dad8218ba6b389513a356a2e
106361Author: Peter Hutterer <peter@cs.unisa.edu.au>
106362Date: Thu Apr 26 22:18:35 2007 +0930
106363
106364 Fix cursor rendering for multi-head.
106365
106366 Before putting anything on the screen, check if the GC was made for the
106367 ScreenRec we want to render to. If not, toss the GC and create a new one. This
106368 is not the best solution but it does the job for now. Same thing for ARGB
106369 cursors except that it's even uglier.
106370
106371 Also remember the screen the cursor was rendered to and check for the right
106372 screen in the BlockHandler, SourceValidate and a few others. Only remove or
106373 restore the cursor if we are rendering to the same screen, otherwise we get
106374 artefacts that are both funky and really annoying.
106375
106376commit 82f97e1c0cc15b050edc82a8f3b9a423d6cf5fe7
106377Author: Paulo Ricardo Zanoni <prz05@c3sl.ufpr.br>
106378Date: Thu Apr 26 15:58:50 2007 +0930
106379
106380 Enable event delivery for multiple heads.
106381
106382 Requires moving the spriteTrace into the DeviceIntRec and adjusting a few
106383 functions to take in device argument, most notably XYToWindow().
106384
106385 Cursor rendering on the second screen is busted.
106386
106387commit c0346e57e6d3857994f7af76060c502c2fdea294
106388Author: Peter Hutterer <peter@cs.unisa.edu.au>
106389Date: Thu Apr 26 12:02:45 2007 +0930
106390
106391 Require inputproto 1.4.2.
106392
106393 Requirement was introduced with c7e2ba0c9b9b1fc1aed8f91f86471c4c8e650b78.
106394
106395commit 67347739b0571b2978468e8088480b105f505ad2
106396Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106397Date: Wed Apr 25 14:19:39 2007 -0400
106398
106399 Don't treat convolution filters as transformations.
106400
106401 Some rearrangement of code to get it closer to pixman.
106402
106403commit c056ce95d89ef1df57edf47149fc34cd3925496e
106404Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106405Date: Wed Apr 25 13:21:47 2007 -0400
106406
106407 Port MSVC++ CPU detection code from pixman. (Vladimir Vukicevic).
106408
106409commit c19ece1d8c32dc81740a4036a642661f54064e75
106410Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106411Date: Wed Apr 25 12:34:19 2007 -0400
106412
106413 Integrate optimization from xserver from David Reveman where repeats
106414
106415 get handled by fbFetchTransformed() rather than in the region walking
106416 code.
106417
106418commit 48c73dfc369fdf8f6023436ebe82bb604f76bb80
106419Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106420Date: Wed Apr 25 12:09:22 2007 -0400
106421
106422 Add function fbCompositeSrcAdd_8888x8x8(), and fix a bug where
106423
106424 srcRepeat = FALSE would be set in the wrong place.
106425
106426commit 66ba3d758a368bf83d75bab8b08bdb6b34925e40
106427Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106428Date: Wed Apr 25 10:31:38 2007 -0400
106429
106430 Various fixes from xserver via pixman (Billy Biggs)
106431
106432commit c09e68ce30dabd6b7068b163b9d2382d85d0d0bc
106433Author: Adam Jackson <ajax@benzedrine.nwnk.net>
106434Date: Wed Apr 25 16:46:26 2007 -0400
106435
106436 Paper over a crash at exit during GLX teardown.
106437
106438commit 9c80eda826448822328bb678a7d284cc43fffb17
106439Author: Adam Jackson <ajax@benzedrine.nwnk.net>
106440Date: Wed Apr 25 16:35:04 2007 -0400
106441
106442 Disable RANDR's fake Xinerama protocol when there's more than one screen.
106443
106444 ... in the protocol sense. Xinerama doesn't have any provision for more
106445 than one protocol screen each with its own geometry.
106446
106447 Red Hat bug #231257.
106448
106449commit d322608dc929d5f8cda07a53143a4f28423e0460
106450Author: Adam Jackson <ajax@benzedrine.nwnk.net>
106451Date: Wed Apr 25 16:29:48 2007 -0400
106452
106453 Fix a buffer overrun on machines with excessively large PCI busses.
106454
106455 Formerly we sized an array with a compile time constant, then initialized
106456 its size to the same constant, but the Linux PCI init code would increase
106457 that "constant". So if you happened to have more than 128 PCI devices,
106458 you'd happily scribble into whatever variables happened to be in .bss
106459 after that array.
106460
106461 Only really fixed for Linux atm. Other OSes will simply (still) fail to
106462 work on video devices above the 128th PCI device.
106463
106464commit 3ba1e8ab6d69566e1a3f8f0eb4605631aeffc8e5
106465Author: Aaron Plattner <aplattner@nvidia.com>
106466Date: Tue Apr 24 17:20:14 2007 -0700
106467
106468 Include xf86Rename.h in xf86RandR12.h.
106469
106470commit 0a2fe443d25b1ca25349aba3f748df986952e20f
106471Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106472Date: Tue Apr 24 19:02:44 2007 -0400
106473
106474 Use READ/WRITE macros for new functions introduced in previous commits.
106475
106476commit 7e16da7b78c422f96387502b9cc29eaa1741543f
106477Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106478Date: Tue Apr 24 18:15:34 2007 -0400
106479
106480 Remove #if 0'ed blocks
106481
106482commit 18252a515d4989b983a3b7389636045e06d0f246
106483Author: Brian <brian@yutani.localnet.net>
106484Date: Tue Apr 24 14:10:09 2007 -0600
106485
106486 bump release date to reflect input code updates
106487
106488commit 2d9a7a768747ca39a800475f12c424c298018dc6
106489Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106490Date: Tue Apr 24 14:46:59 2007 -0400
106491
106492 From pixman (Jeff Muizelaar)
106493
106494 Fix up the fast-path compositing operators; those are useful for
106495 sources without alpha, but can't be used for sources with
106496 alpha. Also, replaced fbCompositeSrcSrc_nxn with call to fbBlt as
106497 this function must handle 1, 4, 8, 16, 24, 32 bpp objects. Would
106498 be nice to optimize fbBlt for common cases involving 8, 16, 24 and
106499 32bpp.
106500
106501 From Keith Packard.
106502
106503commit fde4a5adf02d3067a064ebf6bdd666aa5784cfe9
106504Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106505Date: Tue Apr 24 13:30:43 2007 -0400
106506
106507 From xserver via pixman (Jeff Muizelaar)
106508
106509 Add some optimizations from jaymz. Also adds some compile
106510 warnings that will hopefully go awa y as we continue merging.
106511
106512commit 13e1d5ea55b0a3b7729316c8e37d3d8fca2075b5
106513Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106514Date: Tue Apr 24 12:59:18 2007 -0400
106515
106516 Fix format vs formatCode in previous commit
106517
106518commit 077a5d4555676d5775e990468a697b6890c6d609
106519Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106520Date: Tue Apr 24 12:57:55 2007 -0400
106521
106522 Add functions fbCompositeSrcSrc_nxn() and fbCompositeTrans_0565xnx0565
106523 from xserver via pixman. Add READ/WRITE and fbFinishAccess as
106524 appropriate.
106525
106526commit 09436fb7c38a9819bde770c4c21143591671c4d7
106527Author: Peter Hutterer <peter@cs.unisa.edu.au>
106528Date: Tue Apr 24 22:52:33 2007 +0930
106529
106530 Disable devices before removing, remove unrecoverable devices.
106531
106532commit c7e2ba0c9b9b1fc1aed8f91f86471c4c8e650b78
106533Author: Peter Hutterer <peter@cs.unisa.edu.au>
106534Date: Tue Apr 24 21:34:47 2007 +0930
106535
106536 Use DevicePresence events to tell the client about enabled/disabled devices.
106537
106538 Include the device id in the event sent to the client.
106539
106540commit ce099a9b78195540ec251a6a3dbe26019c1a686d
106541Author: Brian <brian@yutani.localnet.net>
106542Date: Mon Apr 23 12:34:01 2007 -0600
106543
106544 fix bug in which maxKeysPerModifier wasn't getting set
106545
106546commit b5e1f7869b2f12a1c2baa7f699ae609fc9ad50aa
106547Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106548Date: Mon Apr 23 14:16:30 2007 -0400
106549
106550 Remove #if 0'ed merge leftovers
106551
106552commit 84838268b34661d598f8e4856fab355f414930d9
106553Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106554Date: Mon Apr 23 13:19:54 2007 -0400
106555
106556 Gradient fixes
106557
106558 * Port fix for bug 7685 from pixman. Patch by Carl Worth
106559
106560 * Add projective version of radial gradient code.
106561
106562 * Make sure that all Pict*Gradient types have PictGradient as prefix,
106563 since code in various places relies on that.
106564
106565commit 38d14e858980a1b0c087344d24bf6aebf755663c
106566Author: Aaron Plattner <aplattner@nvidia.com>
106567Date: Sun Apr 22 18:04:27 2007 -0700
106568
106569 Adjust the screen pixmap's dimensions in xf86RandR12ScreenSetSize.
106570
106571commit ca784df84e07227a4cc0a1add079884f557b7a00
106572Author: Aaron Plattner <aplattner@nvidia.com>
106573Date: Sun Apr 22 16:26:01 2007 -0700
106574
106575 Fix unbalanced fbGetDrawable added in commit 0a9239ec.
106576
106577commit 55bd8668e7d4100579bcd6c16a804d9f25267070
106578Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106579Date: Fri Apr 20 14:53:37 2007 -0400
106580
106581 Remove #if 0'ed leftovers from merge
106582
106583commit 41dd7ab067adde8f66cd9f74c5a6570c325518a5
106584Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106585Date: Fri Apr 20 14:51:40 2007 -0400
106586
106587 Fix gradient walker to not reset needlessly
106588
106589 Previously the gradient walker was doing excessive resets, (such
106590 as on every pixel in constant-colored regions or outside the
106591 gradient with CAIRO_EXTEND_NONE). Don't do that.
106592
106593 Carl Worth, from pixman
106594
106595commit c1b73f0f2acd56b423b91a04f1e1b3cdcad0069f
106596Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106597Date: Fri Apr 20 14:34:13 2007 -0400
106598
106599 Fixing gradient repeat mode computations in previous patch. From David
106600
106601 Turner.
106602
106603commit 38f718799c68995c2d9a1680355bd55fd925009e
106604Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106605Date: Fri Apr 20 13:59:11 2007 -0400
106606
106607 Remove a few memory references in fbFetchTransformed
106608
106609commit 9c4b14d4f6a1fe018acd64789434216cd1560a4a
106610Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106611Date: Fri Apr 20 13:23:58 2007 -0400
106612
106613 Integrate David Turner's gradient optimizations from pixman
106614
106615commit cd2c1714eb4946bf7b1fc194fe074f8024a2ec23
106616Author: Brian <brian@yutani.localnet.net>
106617Date: Sat Apr 21 12:40:51 2007 -0600
106618
106619 add slang_mem.c
106620
106621commit 67545333ec0b08db783e94e9e3ec55873dea19a3
106622Author: Brian <brian@yutani.localnet.net>
106623Date: Sat Apr 21 12:40:33 2007 -0600
106624
106625 replace occlude.c w/ queryobj.c
106626
106627commit 39bc8bb0fdc854dcf9bbc0857fec84d50fa4f3b2
106628Author: Aaron Plattner <aplattner@nvidia.com>
106629Date: Fri Apr 20 14:22:42 2007 -0700
106630
106631 Don't call xf86RandR12TellChanged if it doesn't exist. Add some exports to xf86Rename.h.
106632
106633commit 3daa5c1a991d659b1386a09e33b044470d489cb3
106634Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106635Date: Fri Apr 20 09:43:16 2007 -0400
106636
106637 Fix pixman bug 5777, patch by David Reveman
106638
106639commit 9c2e955f6792e80fb84f848ed9e6ebbfd79f7130
106640Author: Brian <brian@yutani.localnet.net>
106641Date: Fri Apr 20 07:21:19 2007 -0600
106642
106643 regenerated to add GL_CLIENT_ATTRIB_STACK_DEPTH (bug 9823)
106644
106645commit 96ef0f78438b60436c3940817980a3ab4070c7e8
106646Author: Keith Packard <keithp@neko.keithp.com>
106647Date: Thu Apr 19 17:39:51 2007 -0700
106648
106649 Disable SourceValidate in rotation to capture cursor.
106650
106651 SourceValidate is used exclusively by the software cursor code to pull the
106652 cursor off of the screen before using the screen as a source operand. This
106653 eliminates the software cursor from the frame buffer while painting the
106654 rotated image though. Disabling this function by temporarily setting the
106655 screen function pointer to NULL causes the cursor image to be captured.
106656 (cherry picked from commit 05e1c45ade9c558820685bfd2541617a2e8de816)
106657
106658commit 7ca4baffb5569ea12b578a4a3f69e93d272d6c6d
106659Author: Keith Packard <keithp@neko.keithp.com>
106660Date: Thu Apr 19 17:37:18 2007 -0700
106661
106662 Was accidentally disabling rotation updates in mode set.
106663
106664 Setting a mode on an unrotated CRTC was causing all of the rotation updates
106665 to be disabled; the loop looking for active rotation wasn't actually looking
106666 at each crtc, it was looking at the modified crtc many times.
106667 (cherry picked from commit 8b217dee3a6c46b13fc9571a4a9a95bc55686cdb)
106668
106669commit 806a537e644d8cc9e53f3ac52efb49453e5aa1fb
106670Author: Keith Packard <keithp@neko.keithp.com>
106671Date: Thu Apr 19 17:49:34 2007 -0700
106672
106673 Revert "Suppress software cursor removal during rotated shadow buffer drawing."
106674
106675 This reverts commit 999b681cf3973af4191506e49cde06963b11a774.
106676 Replacing this with simpler code that just disables SourceValidate
106677 during rotation redisplay.
106678
106679commit 0a9239ec258828ec1da6c208634a55fc4053d7da
106680Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
106681Date: Thu Apr 19 18:19:34 2007 -0400
106682
106683 Merge David Reveman's gradient optimization patch from pixman
106684
106685commit d0e55774e0da641ba85c5173f27f68de27372747
106686Author: Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
106687Date: Thu Apr 19 11:39:53 2007 +0200
106688
106689 libdri: Make sure the new DRIInfo keepFDOpen member is honoured.
106690
106691commit e1f0b3e70b696d7ea4cf9e6ed30d751e7fdbc577
106692Author: Peter Hutterer <peter@cs.unisa.edu.au>
106693Date: Thu Apr 19 12:00:24 2007 +0930
106694
106695 config: Return errors as negative numbers, device ids as positive numbers.
106696 Update dbus-api documentation.
106697
106698commit c6972c893359f8fa7631ae674330f3f4f7010ba0
106699Author: Peter Hutterer <peter@cs.unisa.edu.au>
106700Date: Wed Apr 18 12:10:05 2007 +0930
106701
106702 Change dbus 'listDevices' call to not require an argument.
106703 Update dbus-api documentation, plug memory leak on dbus reply error.
106704
106705commit d61ed6c8a2823a3532439d5cb9f355129c93f523
106706Author: Peter Hutterer <peter@cs.unisa.edu.au>
106707Date: Thu Apr 19 18:08:55 2007 +0930
106708
106709 Check for NULL devices in CloseDevice().
106710
106711commit 9c30f7422121a0443c8d612d06181e17d8af9730
106712Author: Peter Hutterer <peter@cs.unisa.edu.au>
106713Date: Thu Apr 19 12:00:24 2007 +0930
106714
106715 config: Return errors as negative numbers, device ids as positive numbers.
106716 Update dbus-api documentation.
106717
106718commit 999b681cf3973af4191506e49cde06963b11a774
106719Author: Eric Anholt <eric@anholt.net>
106720Date: Wed Apr 18 14:33:27 2007 -0700
106721
106722 Suppress software cursor removal during rotated shadow buffer drawing.
106723
106724commit 28bb34eec63bf3c98f38ba7fc044f6419aaa3307
106725Author: Eric Anholt <eric@anholt.net>
106726Date: Wed Apr 18 13:48:28 2007 -0700
106727
106728 Belatedly bump XORG_VERSION for 7.2.
106729
106730commit 5d8e8a7f4b3226bffd9e4d6d9326688f475b0183
106731Author: Eric Anholt <eric@anholt.net>
106732Date: Wed Apr 18 13:22:26 2007 -0700
106733
106734 Remove libminimi build.
106735
106736 It appears to have been a leftover of a previous incarnation of the build
106737 system that didn't handle miinitext.c well.
106738
106739commit 53fb42e65c2b2ff58a4a324b7f05cff8a587720a
106740Author: Erik Andrén <erik.andren@gmail.com>
106741Date: Tue Apr 17 21:34:47 2007 -0700
106742
106743 Syncmaster 226 monitor needs 60Hz refresh (#10545).
106744
106745 I've managed to solve my own bug (#10545) by applying the following
106746 patch to the xserver.
106747
106748 Please apply.
106749
106750 <Conspiracy mode on>
106751 This monitor is "Vista Certified". I wonder if this is a pure coincidence...
106752 <Conspiracy mode off>
106753
106754 With kind regards
106755 Erik Andrén
106756 (cherry picked from commit a63704f14a1d97b9a00fef6fa290e74e51b9732b)
106757
106758commit ae75afcb1b5419102c5be10b8826ceed50d2ef5d
106759Author: Peter Hutterer <peter@cs.unisa.edu.au>
106760Date: Wed Apr 18 12:10:05 2007 +0930
106761
106762 Change dbus 'listDevices' call to not require an argument.
106763 Update dbus-api documentation, plug memory leak on dbus reply error.
106764
106765commit 9cee4ec5e6e06d23aafb302494b082c77ade4623
106766Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
106767Date: Tue Apr 17 16:01:56 2007 -0400
106768
106769 xace: change the semantics of the return value of XACE hooks to allow
106770 arbitrary X status codes instead of just TRUE/FALSE.
106771
106772 The dix layer in most cases still does not propagate the return value of
106773 XACE hooks back to the client, however. There is more error propagation
106774 work to do.
106775
106776commit 47bd311e3dcc501cbb202ce79a55ac32e9db50f2
106777Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
106778Date: Tue Apr 17 13:46:55 2007 -0400
106779
106780 security: remove debugging code.
106781
106782commit 1f06d32ef58749d0f0c062193d237ee98f60e90f
106783Merge: cf962a8 fc162c6
106784Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
106785Date: Tue Apr 17 12:10:22 2007 -0400
106786
106787 Merge branch 'master' into XACE-SELINUX
106788
106789 Conflicts:
106790
106791 dix/privates.c
106792
106793commit b6aec7f6f906a18d13586d63afabf1ee4fbb11c3
106794Author: Peter Hutterer <peter@cs.unisa.edu.au>
106795Date: Tue Apr 17 16:51:16 2007 +0930
106796
106797 Change FocusIn/Out semantics to match Enter/Leave semantics.
106798
106799commit 451d5464b4e8a2516b8a4598b3c4eb14656be90e
106800Author: Peter Hutterer <peter@cs.unisa.edu.au>
106801Date: Tue Apr 17 11:17:04 2007 +0930
106802
106803 Change enter/leave semantics for events with detail Notify{Nonlinear}Virtual.
106804
106805 Core enter/leave events with detail Notify{Ancestor|Inferior|Nonlinear} are
106806 only sent for the first/last pointer to enter/leave. Events with detail
106807 Notify{Nonlinear}Virtual are sent at all times, but not to those windows that
106808 currently have one or more pointers inside their boundaries.
106809
106810commit fc162c6cfa06f0b012743d6d79cef45cf0166229
106811Author: Keith Packard <keithp@neko.keithp.com>
106812Date: Mon Apr 16 09:55:58 2007 -0700
106813
106814 Allow outputs to be explicitly enabled in config, overriding detect.
106815
106816 Option "Enable" "True" will force the server to enable an output at startup
106817 time, even if the output is not connected. This also causes the default
106818 modes to be added for this output, allowing even sync ranges to be used to
106819 pick out standard modes.
106820 (cherry picked from commit a3d73ba2cb7e13a6d129cd88d6a7f7d756e2ced2)
106821
106822commit c41e3bd713206c0bbd8ab8cef4c83eb7ba7e1c3c
106823Author: Keith Packard <keithp@neko.keithp.com>
106824Date: Mon Apr 16 09:53:42 2007 -0700
106825
106826 Use default screen monitor for one of the outputs.
106827
106828 By default, use the screen monitor section for output 0, however, a driver
106829 can change which output gets the screen monitor by calling
106830 xf86OutputUseScreenMonitor.
106831 (cherry picked from commit f4a8e54caf6b9431711383a39f55a18e7fd654f4)
106832
106833commit 97a2c2579c56c304705c934f3b536473645747df
106834Author: Keith Packard <keithp@neko.keithp.com>
106835Date: Mon Apr 16 09:39:47 2007 -0700
106836
106837 Using wrong log level in extension to built-in message
106838
106839 was: typo in built-in module log message
106840 (cherry picked from commit 00cfd1f765895b4d1b2234f3203727a8871b64b0)
106841
106842commit deda7791dfa34d0563c8d7fa2a0660ac27e6858c
106843Author: Brian <brian@yutani.localnet.net>
106844Date: Mon Apr 16 11:35:22 2007 -0600
106845
106846 remove sources deleted in Mesa
106847
106848commit 02d42f344ce020c9b84723671cb9c68d5c064933
106849Author: Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
106850Date: Mon Apr 16 17:24:53 2007 +0200
106851
106852 Changes for single-entity multi-screen DRI.
106853
106854 The entity (device) has a locking SAREA and a master file descriptor
106855 that optionally isn't closed between server generation.
106856
106857 The locking SAREA contains the device hardware lock.
106858 Each DRI screen creates an new SAREA containing the drawable lock,
106859 drawable-and private info, the drawable SAREA.
106860
106861 The first screen optionally shares its drawable SAREA with the
106862 device SAREA.
106863
106864 Default is to close the master descriptor between server generations,
106865 and to share the drawable SAREA of the first screen with the device locking
106866 SAREA. Thus we should (hopefully) have full backwards compatibility.
106867
106868 Mesa changes to support single-device multiple screens are pending.
106869
106870commit b5823ea3e1ed5a0449d44da05165a46719dcf287
106871Author: Keith Packard <keithp@neko.keithp.com>
106872Date: Sun Apr 15 22:59:19 2007 -0300
106873
106874 RandR 1.2 spec says CRTC info contains screen-relative geometry.
106875
106876 Was reporting mode size instead of adjusting for rotation.
106877 (cherry picked from commit e2e7c47a528447e90cff6cf10d2ce457742ef48d)
106878
106879commit cc4eb1c7ea1bace7ed69cfd80c99d22933282ae1
106880Author: Keith Packard <keithp@neko.keithp.com>
106881Date: Fri Apr 13 15:04:29 2007 -0300
106882
106883 Add quirk for Acer AL1706 monitor to force 60hz refresh.
106884
106885 This Acer monitor reports support for 75hz refresh via EDID, and yet when
106886 that rate is delivered, the monitor does not sync and reports out of range.
106887 Use the existing 60hz quirk for this monitor.
106888 (cherry picked from commit 1328a288e9030a472a915077160f090d1afd4126)
106889
106890commit eba81a0a01f8a61151d8bf9f3d83bda85ca26e73
106891Author: George Sapountzis <gsap7@yahoo.gr>
106892Date: Sat Apr 14 18:30:09 2007 +0300
106893
106894 glx: move __glXMesaProvider from GLcore module to glx module.
106895
106896 This treats the GLcore provider similar to DRI provider, using a subset of
106897 XMesa as the GLcore interface.
106898
106899commit 6b040b79f0e247b6f2da8f7d239443743e96de67
106900Author: George Sapountzis <gsap7@yahoo.gr>
106901Date: Sat Apr 14 18:29:52 2007 +0300
106902
106903 glx: drop xmesaP.h include from xf86glx.c
106904
106905 The declarations for the xfree86-specific XMesa functions were moved up to
106906 xmesa.h, requires Mesa as of 2007-04-13.
106907
106908commit 2c833f60acb3dc358815a99cd295ef7fc695c45d
106909Author: George Sapountzis <gsap7@yahoo.gr>
106910Date: Sat Apr 14 18:29:25 2007 +0300
106911
106912 glx: drop stray CAPI define.
106913
106914 SI imports/exports were dropped from Mesa.
106915
106916commit ea27b09d3a973d9a4943f205b24940b8624cf405
106917Author: Peter Hutterer <peter@cs.unisa.edu.au>
106918Date: Fri Apr 13 17:15:22 2007 +0930
106919
106920 Deliver FocusIn/FocusOut events for all devices.
106921 Don't use VCK in SetInputFocus and DeliverFocusedEvent.
106922
106923commit 7ccebc50b98ac175fdbdfaab081bcead62e60ee3
106924Author: Peter Hutterer <peter@cs.unisa.edu.au>
106925Date: Fri Apr 13 13:08:44 2007 +0930
106926
106927 Documentation for events.c.
106928
106929commit 2bb3c81fc1ca98c909309f3af7e9c8ca6b695657
106930Author: Peter Hutterer <peter@cs.unisa.edu.au>
106931Date: Thu Apr 12 16:35:43 2007 +0930
106932
106933 Use paired kbd instead of VCK for EnterLeaveEvents, sanitize variable naming.
106934
106935commit e7b47b1758ed20e75ee267b3a09e91bf2861f6fd
106936Author: Peter Hutterer <peter@cs.unisa.edu.au>
106937Date: Thu Apr 12 16:24:42 2007 +0930
106938
106939 Change core enter/leave semantics for multiple pointers.
106940
106941 Send EnterNotify when first device enters the window, LeaveNotify when the
106942 last device leaves the window. Additional devices will not cause additional
106943 Enter/LeaveNotifies.
106944
106945commit 547d720938b3668666d60110d79b150b1e9325c6
106946Author: Peter Hutterer <peter@cs.unisa.edu.au>
106947Date: Thu Apr 12 11:16:19 2007 +0930
106948
106949 Remove workaround for a NULL core pointer in ProcessInputEvents.
106950
106951 This workaround is obsolete with 33a5d9605e3e282f6aa1921d7321a2a12ef02c42
106952
106953commit d4dad6f84f82a4ade5005c3aa93511c1295875b8
106954Merge: f1f8b56 33a5d96
106955Author: Peter Hutterer <peter@cs.unisa.edu.au>
106956Date: Thu Apr 12 11:11:03 2007 +0930
106957
106958 Merge branch 'master' into mpx
106959
106960 Conflicts:
106961
106962 configure.ac
106963 dix/events.c
106964 hw/xfree86/common/xf86Xinput.c
106965
106966commit 33a5d9605e3e282f6aa1921d7321a2a12ef02c42
106967Author: Daniel Stone <daniel@fooishbar.org>
106968Date: Wed Apr 11 18:28:57 2007 +0300
106969
106970 XFree86: DGA: Don't call ProcessInputEvents from CloseScreen
106971
106972 By the time CloseScreen gets called, we can't call ProcessInputEvents, as
106973 the event queue will get unhappy. So just unregister our hooks instantly,
106974 and hope that they don't get called.
106975
106976commit 0910540e4322bba72a2fa0a907072eab2547a7b6
106977Author: Remigiusz Marcinkiewicz <enleth@enleth.com>
106978Date: Wed Apr 11 01:09:26 2007 +0300
106979
106980 Config: Extend D-BUS API
106981
106982 Return device ID where available.
106983 Add listDevices call, which does what it says on the box.
106984
106985commit aecbc712144dd1aaf462bd758821438b1d22d957
106986Author: Remigiusz Marcinkiewicz <enleth@enleth.com>
106987Date: Wed Apr 11 00:38:16 2007 +0300
106988
106989 Input: Allow a pointer to a device to be returned in NIDR
106990
106991 Allow a pointer to the first device added to be returned, so we know which
106992 device(s) were added by the NIDR call.
106993
106994commit 4f05f9591e5492c72f3856bd7a2ff13378f59f2b
106995Author: Magnus Vigerlöf <Magnus.Vigerlof@home.se>
106996Date: Tue Apr 10 23:57:48 2007 +0300
106997
106998 Input: Always add devices with first available ID
106999
107000 Scan the device list when adding a new device, and make sure we can use
107001 the first available ID, instead of always incrementing.
107002
107003commit 20674dcbb2373a0af287883bc008fb6fb23d4466
107004Author: Magnus Vigerlöf <Magnus.Vigerlof@home.se>
107005Date: Tue Apr 10 23:55:36 2007 +0300
107006
107007 Config: Fix memory leaks
107008
107009 Fix memory leaks that could occur along the error path.
107010
107011commit 82962bbae2b4fda274625d1712ef839ce1ab9dc8
107012Author: Magnus Vigerlöf <Magnus.Vigerlof@home.se>
107013Date: Tue Apr 10 23:54:32 2007 +0300
107014
107015 Input: Add DeleteInputDeviceRequest
107016
107017 Add DIDR, which asks the DDX to remove a device, analogous to
107018 NewInputDeviceRequest. Only implemented for XFree86 at the moment.
107019
107020commit 7b82a836c66ba88566255052caff63577e1a0384
107021Author: Magnus Vigerlöf <Magnus.Vigerlof@home.se>
107022Date: Tue Apr 10 23:52:08 2007 +0300
107023
107024 XFree86: Fix memory leaks, option parsing, in NewInputDeviceRequest
107025
107026 Plugged some possible memory leaks, and added some more checks on the
107027 options, particular for driver/identifier. Added an unwind.
107028
107029commit 4771fa8747791498e504d73afccfb5833499a38b
107030Author: Magnus Vigerlöf <Magnus.Vigerlof@home.se>
107031Date: Tue Apr 10 23:48:00 2007 +0300
107032
107033 XFree86: Fix memory leak in option parsing
107034
107035 Fix option parsing functions and callers thereof to not leak memory.
107036
107037commit 07c56abf84080c020a3e7b7703a447c7f996975c
107038Author: Magnus Vigerlöf <Magnus.Vigerlof@home.se>
107039Date: Tue Apr 10 23:43:58 2007 +0300
107040
107041 Input: Plug memory leak in device free
107042
107043 Remember to also free the motion history, if we're using the DIX-managed
107044 history.
107045
107046commit e92743bc9839c36914a44f3e5bc8cd85773ac794
107047Author: Daniel Stone <daniel@fooishbar.org>
107048Date: Sun Apr 8 14:02:02 2007 +0300
107049
107050 getevents: Copy modifier state from extended to core devices
107051
107052 Make core events carry the same modifier state as the extended events, so
107053 that holding down Ctrl on keyboard A and pressing Q on keyboard B won't
107054 cause your app to quit.
107055
107056commit e49f836d6fa2768cd6d2a6d0227b5dbf516013dc
107057Author: Daniel Stone <daniel@fooishbar.org>
107058Date: Sun Apr 8 13:56:41 2007 +0300
107059
107060 mieq: Use larger default queue size
107061
107062 Use a default queue size of 512 rather than 256, else Xephyr is too slow
107063 without a host cursor, so events get stuck in the queue.
107064
107065commit 4aae2de74b9224bac2b2e2522637dac09abc3837
107066Author: Jared D. McNeill <jmcneill@netbsd.org>
107067Date: Tue Apr 10 12:57:15 2007 -0700
107068
107069 Add a real xf86EnableIO/xf86DisableIO for NetBSD/PPC.
107070
107071commit f77a8ea849d171a8ca00b2b7334866ace1ffbf73
107072Author: Keith Packard <keithp@neko.keithp.com>
107073Date: Mon Apr 9 14:29:46 2007 -0700
107074
107075 Rotate screen size as needed from RandR 1.1 change requests.
107076
107077 Screen size must reflect rotated mode size when setting rotated mode using
107078 RandR 1.1 SetScreenConfig request.
107079 (cherry picked from commit efcec7dbd3c2736c7b421d29c4d37e231aa681d2)
107080
107081commit bcf17df69a232e5e84609abacdca36274316e170
107082Author: Keith Packard <keithp@neko.keithp.com>
107083Date: Mon Apr 9 14:12:27 2007 -0700
107084
107085 Disable CRTC when SetSingleMode has no matching mode. Update RandR as well.
107086
107087 xf86SetSingleMode tries to resize all crtcs to match the selected mode. When
107088 a CRTC has no matching mode, it now disables the CRTC (instead of crashing).
107089
107090 Also, poke the RandR extension when xf86SetSingleMode is done so that
107091 appropriate events can be delivered, and so that future RandR queries return
107092 correct information.
107093 (cherry picked from commit dc6c4f6989f87149d8605604f4514f5cbf11de67)
107094
107095commit 67e1c98895a566f927e1ae2384d56cfca104f971
107096Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107097Date: Mon Apr 9 19:08:52 2007 -0400
107098
107099 Look for the PCI ROM file elsewhere in sysfs.
107100
107101 /sys/devices reflects the bus topology, and we don't care that much.
107102 Easier (and more reliable) to just look in /sys/bus/pci/devices, which
107103 is a flat view.
107104
107105commit a08d5157f70567a0aa9583d4a15e62437340cf34
107106Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107107Date: Mon Apr 9 19:04:56 2007 -0400
107108
107109 VT activate or waitactive are fatal if they fail.
107110
107111 Also, be sure to waitactive on the way down, to make sure we're off the VT
107112 before exiting.
107113
107114commit 72b477f964c748a1ab668781643cc11877f19738
107115Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107116Date: Mon Apr 9 18:59:01 2007 -0400
107117
107118 Don't write out empty sections from the parser.
107119
107120commit 1f6741db19d4c91b1eacb497dff1814acb1bf0c3
107121Author: Stefan Huehner <stefan@huehner.org>
107122Date: Mon Apr 9 14:33:15 2007 -0700
107123
107124 Bug #10560: Code-Cleanup: function declarations () -> (void)
107125
107126 X.Org Bugzilla #10560: <https://bugs.freedesktop.org/show_bug.cgi?id=10560>
107127 Patch #9511 <https://bugs.freedesktop.org/attachment.cgi?id=9511>
107128
107129commit f24391dbfd12a84253dfec794ee7884afd52e197
107130Author: Keith Packard <keithp@neko.keithp.com>
107131Date: Mon Apr 9 12:30:31 2007 -0700
107132
107133 In AIGLX EnterVT processing, invoke driver EnterVT before resuming glx.
107134
107135 As the driver EnterVT function generally re-enables the hardware and
107136 prepares it for rendering, it must be called before any gl functions are
107137 called which could touch the hardware.
107138
107139commit 4c2e28b0916b5f75cfefb6df9fa0a7a09675539a
107140Author: Keith Packard <keithp@neko.keithp.com>
107141Date: Mon Apr 9 12:28:53 2007 -0700
107142
107143 Add setrlimit call in -core option to make dumps occur.
107144
107145 Default core size limit for most environments is 0, which disables core
107146 dumps. Add code in the -core option processing path to set the core limit to
107147 the maximum value.
107148
107149commit f1f8b562aaaa6ec32ab0d0697f964d92d6d536a4
107150Author: Peter Hutterer <peter@cs.unisa.edu.au>
107151Date: Mon Apr 9 19:31:59 2007 +0930
107152
107153 Alloc sprite memory in devices' devPrivates, allow undisplaying cursors.
107154
107155 Improve memory usage by allocating the sprite's memory only to devices that
107156 actually have a sprite and provide means to remove a device's cursor from the
107157 screen (more hotplugging, yay!).
107158 This commit breaks ScreenRec's ABI.
107159
107160commit 7cef789fa13ae53bfba6dc7b5a7928b7362b2522
107161Author: Peter Hutterer <peter@cs.unisa.edu.au>
107162Date: Mon Apr 9 18:37:48 2007 +0930
107163
107164 Close down virtual core devices when closing all devices.
107165
107166commit 6081b8c76f7d934bd4e9584a2f6d55636c5289d2
107167Author: Peter Hutterer <peter@cs.unisa.edu.au>
107168Date: Mon Apr 9 18:35:47 2007 +0930
107169
107170 Don't explicitly init sprite for VCP, EnableDevice() will take care of that.
107171
107172commit 00b1d52f4103a07568dcebcdaa860b9b51f19b4d
107173Author: Peter Hutterer <peter@cs.unisa.edu.au>
107174Date: Mon Apr 9 18:32:00 2007 +0930
107175
107176 Adding some comments to devices.c.
107177
107178commit aa77ffb510abe004802ab9acc6996e4c6fe3ebb2
107179Author: Peter Hutterer <peter@cs.unisa.edu.au>
107180Date: Mon Apr 9 18:27:22 2007 +0930
107181
107182 Fix: pick new ClientPointer when device is closed.
107183
107184commit 4beeab8424774ea4c3142f29b90e33f1fc7cb154
107185Author: Daniel Stone <daniel@fooishbar.org>
107186Date: Sun Apr 8 13:39:06 2007 +0300
107187
107188 XFree86: Treat evdev and vmmouse as mouse drivers (bug #10512, #10559)
107189
107190 When we see an evdev or vmmouse section, assume that it's a mouse, and
107191 don't add a default mouse device. This will break users who have an
107192 evdev keyboard section but no mouse, and want the mouse to get added
107193 by default.
107194
107195commit 0a6ac992363343487dfe0a0fc985ea55bd448382
107196Author: Brian <brian@yutani.localnet.net>
107197Date: Sat Apr 7 12:41:57 2007 -0600
107198
107199 regenerated to fix bug 10371
107200
107201commit 7e385598613778de14c0feea0d32f17d7aa66a8e
107202Author: George Sapountzis <gsap7@yahoo.gr>
107203Date: Fri Apr 6 13:38:12 2007 +0300
107204
107205 GLcore: fix after moving xf86glx_util.[hc] to Mesa.
107206
107207commit cf962a849db2b259ca558c6265ea7e3328a4d312
107208Merge: 5ad5625 5a804f2
107209Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107210Date: Thu Apr 5 14:41:39 2007 -0400
107211
107212 Merge branch 'master' into XACE-SELINUX
107213
107214 Conflicts:
107215
107216 configure.ac
107217
107218commit 5ad562565ac8ef9257da3afb0de1ae4f90f80fe9
107219Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107220Date: Thu Apr 5 14:18:05 2007 -0400
107221
107222 devPrivates rework: properly free devPrivates on compatibility structures,
107223 type pixmap. Requires ddx'es to call the free function from DestroyPixmap.
107224
107225commit 5a804f2e97ab59745482660a635b801ac2b9e769
107226Author: George Sapountzis <gsap7@yahoo.gr>
107227Date: Thu Apr 5 19:14:31 2007 +0300
107228
107229 configure: minor cosmetic, move GLX extension options together.
107230
107231commit a4e2fc703484fffed8dd50c1b4b24c564be4d3cd
107232Author: George Sapountzis <gsap7@yahoo.gr>
107233Date: Thu Apr 5 19:13:47 2007 +0300
107234
107235 glx: Remove stray __GLinterface.
107236
107237 __GLinterface was droped from glcore.h
107238
107239commit 38ca7d388c47c4800c74442172d6595a9b3dfcc7
107240Author: George Sapountzis <gsap7@yahoo.gr>
107241Date: Thu Apr 5 19:13:14 2007 +0300
107242
107243 glx: fix symlink, glcontextmodes.c was moved to mesa/src/glx/
107244
107245commit 1d550bb2c5cb5b3e588f0e0b68a421dc1cb8bd7c
107246Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107247Date: Thu Apr 5 12:12:58 2007 -0400
107248
107249 devPrivates rework: minor fix; use calloc and avoid initialization.
107250
107251commit 5c4deb71a1cb981ea7e2e25d2b3a1179f27efa5a
107252Author: Peter Hutterer <peter@cs.unisa.edu.au>
107253Date: Thu Apr 5 14:21:46 2007 +0930
107254
107255 Synthesize WarpPointer events for the actual device rather than the VCP.
107256
107257commit ed75b056511ccb429c48c6c55d14dc7ae79e75a3
107258Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107259Date: Wed Apr 4 12:00:15 2007 -0400
107260
107261 dix: add new, combined resource lookup function. Move all dix lookup API
107262 deprecated so far to a new file dix/deprecated.c. Remove the deprecation
107263 warnings for the time being.
107264
107265commit 63e46e4fc3e98751f2edbed9c79ef3d5dc2dadc6
107266Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107267Date: Wed Apr 4 15:59:51 2007 -0400
107268
107269 devPrivates rework: properly free devPrivates on compatibility structures,
107270 excluding pixmap.
107271
107272commit f8482967ae8080f49dd1bbb0b79cc65020df679f
107273Author: Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
107274Date: Wed Apr 4 12:28:48 2007 +0200
107275
107276 Add an EXA driver callback to determine whether a pixmap is
107277
107278 "offscreen" in exa terms, which means accessible to the GPU.
107279 Bump exa minor. The change is backwards-compatible.
107280
107281commit 7f36ba57062096e9c67889d5b7dd64e2fb0004e7
107282Author: Peter Hutterer <peter@cs.unisa.edu.au>
107283Date: Wed Apr 4 17:38:10 2007 +0930
107284
107285 Allow for multiple animated cursors.
107286
107287commit 676188531b2bd7679ee1831b5c517df3e6d215c5
107288Author: Peter Hutterer <peter@cs.unisa.edu.au>
107289Date: Wed Apr 4 17:35:06 2007 +0930
107290
107291 Unrealize cursor only once, not per device.
107292
107293commit c10df5b967d4da4e11786520317e2917de5541fa
107294Author: Aaron Plattner <aplattner@nvidia.com>
107295Date: Tue Apr 3 15:47:01 2007 -0700
107296
107297 Swap RRScreenChangeNotifyEvent dimensions when the screen has one crtc and it's rotated.
107298
107299 RandR 1.1 clients expect the size fields in this event to be the unrotated
107300 dimensions of the screen. This behavior is "weird", but that's the way the old
107301 code worked so we need to be bug-compatible with it.
107302
107303commit 1cb84768f376b477a08a558854609b0743f2bd29
107304Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107305Date: Tue Apr 3 15:31:16 2007 -0400
107306
107307 security: rewrite to use new devPrivates support.
107308
107309commit 14aea12cadef647369e44639ff5024dd7034570a
107310Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107311Date: Tue Apr 3 15:23:56 2007 -0400
107312
107313 xace: forgot one of the hook call arguments. Add it.
107314
107315commit 353e19fd5e18ad55a0dd12a7b63f6af9df7bfe6b
107316Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107317Date: Tue Apr 3 14:06:02 2007 -0400
107318
107319 devPrivates rework: zero out newly allocated private space.
107320
107321commit e1dea151db6405e12d991feacba9446320739ee8
107322Author: Brian <brian@yutani.localnet.net>
107323Date: Tue Apr 3 11:21:50 2007 -0600
107324
107325 Implement a minor hack in dmxCheckFunctionKeys() to detect special keys.
107326
107327 Keep track of status of (left) alt/ctrl keys so that ctrl-alt-q to exit
107328 can be detected. Not ideal, but works for now.
107329
107330commit 0ee40c935750e25a9e178cdd70f6b2c667e79344
107331Author: Brian <brian@yutani.localnet.net>
107332Date: Tue Apr 3 09:31:00 2007 -0600
107333
107334 s/intead/instead/
107335
107336commit 23974f20bf0e0c2786cc75af026af5484f6dc331
107337Author: Brian <brian@yutani.localnet.net>
107338Date: Tue Apr 3 09:30:24 2007 -0600
107339
107340 add DDXRingBell() stub to solve link problem
107341
107342commit a240c039c47c0be22ea5e100692307b26d938747
107343Author: Brian <brian@yutani.localnet.net>
107344Date: Tue Apr 3 09:27:57 2007 -0600
107345
107346 Split the xserver/fb/fbcmap.c file into two files.
107347
107348 Now, fbcmap_mi.c contains the fb functions which just wrap mi functions.
107349 Previously, these were in fbcmap.c and compiled when XFree86Server was defined.
107350 Now, clients of fbcmap should either use fbcmap.c or fbcmap_mi.c and not worry
107351 about setting the XFree86Server symbol.
107352
107353commit 1cc8db72816cd079f30255046e10043c350bf683
107354Merge: 645d87c a39f297
107355Author: Matthieu Herrb <matthieu@roadrock.(none)>
107356Date: Tue Apr 3 16:04:45 2007 +0200
107357
107358 Merge branch 'master' of ssh://herrb@git.freedesktop.org/git/xorg/xserver
107359
107360commit 645d87cf8ef724d4591614f9994cdc4d7549a7a8
107361Author: Matthieu Herrb <matthieu@roadrock.(none)>
107362Date: Tue Apr 3 15:47:18 2007 +0200
107363
107364 CVE-2007-1003: XC-MISC Extension ProcXCMiscGetXIDList() Memory Corruption
107365
107366commit 207f710d16337839c2427aa16fad70a49834153e
107367Author: Peter Hutterer <peter@cs.unisa.edu.au>
107368Date: Tue Apr 3 19:02:14 2007 +0930
107369
107370 Delete all grabs from all devices when deleting a window.
107371
107372commit 5f382c9c7fbda6eccf8e76c28a90b55ff2f0aef3
107373Merge: 9b0b340 f09ee16
107374Author: Peter Hutterer <peter@cs.unisa.edu.au>
107375Date: Tue Apr 3 15:07:18 2007 +0930
107376
107377 Merge branch 'master' into mpx
107378
107379commit f09ee168e234d0a9416e15f2916726b975cad293
107380Author: Peter Hutterer <peter@cs.unisa.edu.au>
107381Date: Tue Apr 3 15:07:00 2007 +0930
107382
107383 Revert "o fix minor error in comment for GetPointerEvents()"
107384
107385 This reverts commit 96ce17aa94413c4b8bcb61cae71167050130a307.
107386
107387commit 9b0b3406682b5a3161e6c3895771523214c37207
107388Author: Peter Hutterer <peter@cs.unisa.edu.au>
107389Date: Tue Apr 3 15:06:26 2007 +0930
107390
107391 Make GrabDevice differ between core grabs and device grabs.
107392
107393commit a39f297ada4fa87c858395ae2aacefac5f8fba05
107394Author: Keith Packard <keithp@viola.jf.intel.com>
107395Date: Mon Apr 2 14:15:36 2007 -0700
107396
107397 Don't erase current crtc for outputs on CloseScreen
107398
107399 Erasing this variable causes some outputs (SDVO on intel) to fail
107400 to be correctly reset at server reset time.
107401 (cherry picked from commit 56262a4ee943f328d089a8eb4aa70b9a4bd5d135)
107402
107403commit 11797ffdcc22160317a5ebbc9291472570a51c6d
107404Author: Eric Anholt <eric@anholt.net>
107405Date: Mon Apr 2 18:21:58 2007 -0700
107406
107407 Move modes/ debugging output under Option "ModeDebug" in the Device section.
107408
107409commit e44f106ffc796c025abdfb66717c06db8b12b4e4
107410Author: Brian <brian@yutani.localnet.net>
107411Date: Mon Apr 2 16:26:15 2007 -0600
107412
107413 clean-up, debug code
107414
107415commit 8d8bc8927760fad631bef83fa2841b455ff6d511
107416Author: Brian <brian@yutani.localnet.net>
107417Date: Mon Apr 2 16:21:57 2007 -0600
107418
107419 fix formatting
107420
107421commit 3e482de7b145a5eed79b81c30c359fe43647824a
107422Author: Brian <brian@yutani.localnet.net>
107423Date: Mon Apr 2 15:38:15 2007 -0600
107424
107425 checkpoint: more clean-up
107426
107427commit 76a7a5ca1f068c27c9b5fbd49d5a1da80ed6f488
107428Author: Brian <brian@yutani.localnet.net>
107429Date: Mon Apr 2 15:24:05 2007 -0600
107430
107431 formatting fixes
107432
107433commit 69baad321d35dae0bfa535be0c6ed2131fed1e60
107434Author: Brian <brian@yutani.localnet.net>
107435Date: Mon Apr 2 15:21:22 2007 -0600
107436
107437 clean-up dmxCoreMotion() dmxCoreMotion2()
107438
107439commit 44eb15adeee3b299677070f39625daa53679bd13
107440Author: Brian <brian@yutani.localnet.net>
107441Date: Mon Apr 2 15:12:04 2007 -0600
107442
107443 checkpoint clean-up
107444
107445commit 0f873a9d4f02b399c37b4058c6a9a2e21aa205e8
107446Author: Brian <brian@yutani.localnet.net>
107447Date: Mon Apr 2 14:51:38 2007 -0600
107448
107449 remove some debug code
107450
107451commit 3a0ce1084a18e17a3c8a009d99c228652b8763a9
107452Author: Brian <brian@yutani.localnet.net>
107453Date: Mon Apr 2 14:51:21 2007 -0600
107454
107455 for completeness, init dummy's min/maxval[1] values (vertical axis)
107456
107457commit 12016f20f7f5365f30cfbeb05568b3fb89759e5a
107458Author: Brian <brian@yutani.localnet.net>
107459Date: Mon Apr 2 14:50:48 2007 -0600
107460
107461 As for normal mouse device, init valuator maxval[] to real values, not zero.
107462
107463commit 0aaf28e5633a59563b89a2e42d19fabc84adc3ed
107464Author: Brian <brian@yutani.localnet.net>
107465Date: Mon Apr 2 12:41:30 2007 -0600
107466
107467 In dmxBackendMouGetInfo() initialize the info->minval[], maxval[] arrays to the size of the backend display.
107468
107469 It seems that the changes to X input exposed a problem that wasn't detected
107470 before. The axis clipping code in GetPointerEvents() uses those limits to
107471 constrain the pointer's coordinate range. The max was zero so the pointer
107472 couldn't move.
107473
107474commit 0013bf6ddb3867c9a504603434d8c2ec83f3f3bc
107475Author: Brian <brian@yutani.localnet.net>
107476Date: Mon Apr 2 12:39:04 2007 -0600
107477
107478 undo 1280 valuator hack
107479
107480commit 08a88d1803f672555141011e082fbc0edeedcf05
107481Author: Brian <brian@yutani.localnet.net>
107482Date: Mon Apr 2 12:28:14 2007 -0600
107483
107484 Pass num_valuators=0 for ButtonPress/Release. This seems to fix the button coordinate problem
107485
107486commit 70683e338dacc48e3adf489d66ec33b29dfc3b77
107487Author: Brian <brian@yutani.localnet.net>
107488Date: Mon Apr 2 12:26:27 2007 -0600
107489
107490 formatting fixes
107491
107492commit e8777a91f37d828b9df839bf3d9cf2f954bdddb0
107493Merge: a120547 96ce17a
107494Author: Peter Hutterer <peter@cs.unisa.edu.au>
107495Date: Mon Apr 2 15:36:26 2007 +0930
107496
107497 Merge branch 'master' into mpx
107498
107499 Conflicts:
107500
107501 Xi/closedev.c
107502 Xi/exevents.c
107503 Xi/extinit.c
107504 Xi/listdev.c
107505 dix/window.c
107506 hw/xfree86/common/xf86Xinput.c
107507 include/extinit.h
107508 mi/mipointer.c
107509
107510commit 96ce17aa94413c4b8bcb61cae71167050130a307
107511Author: Peter Hutterer <peter@cs.unisa.edu.au>
107512Date: Wed Nov 15 17:50:02 2006 +1030
107513
107514 o fix minor error in comment for GetPointerEvents()
107515
107516commit a12054757d21edacc1c24c3077b9214726652829
107517Author: Peter Hutterer <peter@cs.unisa.edu.au>
107518Date: Mon Apr 2 11:20:44 2007 +0930
107519
107520 Xi: fix ProcXListInputDevices to include the virtual core devices.
107521
107522commit f2808005f4ee72c5fd7f5f3dcca181306485113e
107523Author: Alberto Mardegan <mardy@users.sourceforge.net>
107524Date: Sat Mar 31 16:51:24 2007 +0200
107525
107526 Bug #6620: Fixed a missing 'else' in ATIPseudoDMAInit().
107527
107528 Before this, we'd write some registers twice on R200 hardware and also
107529 possibly end up with a bad value in atis->cce_pri_size.
107530
107531commit 5257b32e492bd2082bef6a4cd0fea03ce093c0f8
107532Author: Aaron Plattner <aplattner@nvidia.com>
107533Date: Wed Mar 28 15:51:24 2007 -0700
107534
107535 Bump video driver ABI to 2.0 for cw change (commit 6ed08949af4f7ac09170d3d9581e4092b24a84ee).
107536
107537commit 73fdc16bc4f4e21ff604b3f9ded23b40398fb1b6
107538Author: Brian <brian@yutani.localnet.net>
107539Date: Fri Mar 30 16:07:26 2007 -0600
107540
107541 formatting fixes
107542
107543commit ebdc8ce5c108dc3b6b0004e7c7939d1a5bef8676
107544Author: Brian <brian@yutani.localnet.net>
107545Date: Fri Mar 30 16:05:46 2007 -0600
107546
107547 Checkpoint DMX updates: things are working much better now, but still not 100% right.
107548
107549 Use new dmxCoreMotion2() function which enqueues motion events with
107550 GetPointerEvents()/mieqEnqueue().
107551 The clipAxis() code in GetPointerEvents() is causing some grief. The
107552 limits seem to have always been (0,0) according to the original calls
107553 to InitValuatorAxisStruct() in dmxinputinit.c.
107554 Terrible hack for now: Call InitValuatorAxisStruct() with hard-coded max
107555 values of 1280 (my screen width).
107556
107557commit e9130b8bac73a0843d5ff6b2216eccfb3e094a48
107558Merge: 2c6d471 92e8cdb
107559Author: Ian Romanick <idr@us.ibm.com>
107560Date: Fri Mar 30 13:39:15 2007 -0700
107561
107562 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
107563
107564commit 3c7413e0c2f87e154aa8aa4a83bd585a6d1091e8
107565Author: Brian <brian@yutani.localnet.net>
107566Date: Fri Mar 30 14:07:04 2007 -0600
107567
107568 Tweak some parameters, etc. Things seem a little better now, but still a ways to go.
107569
107570commit 7989dacdcb1449b10d7733dda11cd96e260e9fae
107571Author: Brian <brian@yutani.localnet.net>
107572Date: Fri Mar 30 13:44:24 2007 -0600
107573
107574 num_valuators=1 for GetPointerEvents(), hack ButtonPress/Release position
107575
107576commit 1ea842960fddbc6363cc6e7f914d70ba45525a6b
107577Author: Brian <brian@yutani.localnet.net>
107578Date: Fri Mar 30 13:43:15 2007 -0600
107579
107580 more debug
107581
107582commit 92e8cdbd32b0d86cabd4ad88e3240bf90c018b9a
107583Author: Brian <brian@yutani.localnet.net>
107584Date: Fri Mar 30 13:19:33 2007 -0600
107585
107586 Checkpoint fixes to DMX for X input changes.
107587
107588 Xdmx builds and runs now.
107589 Keyboard seems OK, and mouse pointer moves, but everything else is flakey.
107590 Something is still seriously wrong.
107591
107592commit d92da3d5f309392ac398c0975ef17bb04312d5e2
107593Author: Brian <brian@yutani.localnet.net>
107594Date: Fri Mar 30 12:56:34 2007 -0600
107595
107596 more formatting fixes
107597
107598commit 44acb2517d9fb07790d9d799aa9cc727d1b7d35c
107599Author: Brian <brian@yutani.localnet.net>
107600Date: Fri Mar 30 12:54:22 2007 -0600
107601
107602 Fix some bad formatting.
107603
107604 Doing this:
107605 if (something) stmt;
107606 is evil if you're debugging and want to break on stmt!
107607
107608commit 9f24798af50896cc3262c1201f75c10a688f2a83
107609Author: Brian <brian@yutani.localnet.net>
107610Date: Fri Mar 30 12:49:34 2007 -0600
107611
107612 ompile fbcmap.c w/ -DXFree86Server instead of linking libfbcmap.a.
107613
107614 The former works, the later doesn't (DMX blows up on visuals/pixel formats).
107615 This undos Daniel's patch, which undid my prev patch. Revisit someday.
107616
107617commit 76756f27561c6386cba0d338441e8ec7b98500ce
107618Author: George Sapountzis <gsap7@yahoo.gr>
107619Date: Thu Nov 30 04:20:32 2006 +0200
107620
107621 Make xf86glx.c unaware of Mesa internals
107622
107623 Use newly added XMesaCopyContext() and drop the GlxSetRenderTables() call
107624 for Xgl, as this is now done inside XMesaForceCurrent(). This leaves xmesaP.h
107625 but only for the declarations of the three XMesa/XFree86 functions. Also,
107626 GlxSetRenderTables() stays but is only used in hw/xgl/glxext/ .
107627
107628 Also drop xf86glxint.h, no longer used.
107629
107630 Depends on mesa commit 7439a36785b6a2783e80a40a96c09db8f56dc2bc of 2007-03-30.
107631
107632commit 63d8f01819ef44ea3bf0d4fb20ba1d698ae91cd2
107633Author: Peter Hutterer <peter@cs.unisa.edu.au>
107634Date: Fri Mar 30 14:59:43 2007 +0930
107635
107636 dix: Fix wrong cursor refcount.
107637
107638 Calloc cursor struct to ensure devPrivates are zeroed out and don't
107639 increase the refcnt for devices automatically when allocating a new
107640 cursor. Use new DeviceIsPointerType() to detect if device is a pointer
107641 _before_ device has been activated and can thus be identified and set
107642 up grab functions accordingly. This way we can increase the refcnt
107643 when we get a pointer grab.
107644
107645commit 20e4314b178e1a093bce85e93329d4bcfb4f4210
107646Author: Peter Hutterer <peter@cs.unisa.edu.au>
107647Date: Thu Mar 29 19:12:32 2007 +0930
107648
107649 dix: Try to get closer to the correct refcnt for cursors.
107650
107651 This commit introduces a memory leak, as the refcnt never seems to get
107652 down to 0 and thus cursors will not be freed. Solves the problems with
107653 GNOME segfaulting every 30 seconds though.
107654
107655commit 307d2b57bbfcc281656011533627bea6ab98189e
107656Author: Peter Hutterer <peter@cs.unisa.edu.au>
107657Date: Thu Mar 29 15:23:41 2007 +0930
107658
107659 Xi: remove 'register' keywords.
107660
107661commit 0c607f4b1902f993597e09ea7a2c83bc1cca9b96
107662Author: Peter Hutterer <peter@cs.unisa.edu.au>
107663Date: Thu Mar 29 14:43:50 2007 +0930
107664
107665 dix: Pick keyboard instead of using virtual keyboard in Grab/UngrabKey.
107666
107667commit 82a8b99a6c46018885600011913267d8af9dfe13
107668Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107669Date: Wed Mar 28 15:17:02 2007 -0400
107670
107671 Move the XAA private indices to be static.
107672
107673 Technically this is an ABI break, if you aren't smart enough to be using the
107674 getter functions. Cope.
107675
107676commit 8c7f56d92d8471ee059c14d322af5f7f555dd5c6
107677Author: Tomas Janousek <tomi@nomi.cz>
107678Date: Wed Mar 28 14:46:30 2007 -0400
107679
107680 Bug #10296: Fix timer rescheduling.
107681
107682commit 327bc332a61294209d39286228199f54bdde73d1
107683Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107684Date: Wed Mar 28 13:00:03 2007 -0400
107685
107686 xace: minor comment fixes.
107687
107688commit 299ff4c82998d2a32204bfbecde4993dfbd3d4a5
107689Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107690Date: Wed Mar 28 12:57:11 2007 -0400
107691
107692 xace: provide creation-time resource hook call in CreateWindow().
107693
107694commit 99b220969f2f8ba829bc8294b27ca90fd9311be4
107695Merge: 728ad64 82a8b99
107696Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107697Date: Wed Mar 28 12:54:04 2007 -0400
107698
107699 Merge branch 'master' into XACE-SELINUX
107700
107701commit 5ba4d9eedf1b4ce4795bf910cd184872e2d9b3fc
107702Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107703Date: Wed Mar 28 12:03:19 2007 -0400
107704
107705 Refuse to create tiny modes from EDID detailed timing.
107706
107707commit 2c6d47108880584f1221ff86c6c8947627f9f607
107708Merge: c4fe1bc 8522044
107709Author: Ian Romanick <idr@us.ibm.com>
107710Date: Wed Mar 28 08:08:04 2007 -0700
107711
107712 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
107713
107714 Conflicts:
107715
107716 hw/xfree86/Makefile.am
107717 hw/xfree86/dri/dri.c
107718
107719commit 85220446359a75ea2c359b418b4051c04eea739c
107720Author: Daniel Stone <daniel@fooishbar.org>
107721Date: Wed Mar 28 13:03:32 2007 +0300
107722
107723 GL: Update for Mesa changes
107724 Added s_fragprog.c to fix the build.
107725
107726commit 1af2ef0b25fd8017a3271e624a5f1548f02b09f9
107727Author: Eric Anholt <eric@anholt.net>
107728Date: Tue Mar 27 13:13:45 2007 -0700
107729
107730 Enable Composite by default now that it disables itself in the known bad cases.
107731
107732commit 0bfc3cc22db94ec6867596606fe93228e315c847
107733Author: Eric Anholt <eric@anholt.net>
107734Date: Tue Mar 27 13:12:21 2007 -0700
107735
107736 Disable composite when Xinerama is active.
107737
107738 It will likely take a decent bit of work to make that work right.
107739
107740commit 5e7936371c9e1ac48e19bf1e9e3f71f037fd9b5d
107741Author: Eric Anholt <eric@anholt.net>
107742Date: Mon Mar 26 20:18:18 2007 -0700
107743
107744 Disable Composite when the screen's visual is pseudocolor.
107745
107746 Rendering fails badly in this case, and I don't care enough to fix it.
107747
107748commit 8afc7e2eb3ebec48d3879bf269143259c8bc18c8
107749Author: Eric Anholt <eric@anholt.net>
107750Date: Mon Mar 26 15:55:38 2007 -0700
107751
107752 Refuse to initialize Composite if Render is not present.
107753
107754 Composite relies on the presence of Render, in particular for the automatic
107755 compositing.
107756
107757commit 6ed08949af4f7ac09170d3d9581e4092b24a84ee
107758Author: Eric Anholt <eric@anholt.net>
107759Date: Tue Mar 27 17:31:28 2007 -0700
107760
107761 Move libcw setup to the only renderer requiring it (XAA).
107762
107763 Additionally, protect libcw setup behind checks for Render, to avoid
107764 segfaulting if Render isn't available (xnest).
107765
107766 The previous setup was an ABI-preserving dance, which is better nuked now.
107767 Now, anything that needs libcw must explicitly initialize it, and
107768 miDisableCompositeWrapper (previously only called by EXA and presumably binary
107769 drivers) is gone.
107770
107771commit e76b6349516d5d1c8f7167d6f5419e0d06a546c3
107772Author: Eric Anholt <eric@anholt.net>
107773Date: Mon Mar 26 16:04:50 2007 -0700
107774
107775 Fix indentation of fakexa help text.
107776
107777commit 728ad64d979fd9e5cca28e8c15118c18d707c431
107778Merge: 0331c6a 6a0bed1
107779Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107780Date: Tue Mar 27 13:16:40 2007 -0400
107781
107782 Merge branch 'master' into XACE-SELINUX
107783
107784commit 6a0bed16e80a91891cee6c7033c90875bc2af193
107785Author: Michel Dänzer <michel@tungstengraphics.com>
107786Date: Tue Mar 27 16:51:12 2007 +0200
107787
107788 Fix typo in GL/mesa/shader/slang/Makefile.am.
107789
107790commit b8f846a9dfc6697d59ad5482ba7c9d738875318e
107791Author: Dave Airlie <airlied@linux.ie>
107792Date: Tue Mar 27 14:17:40 2007 +1000
107793
107794 gl: oops dodgy s appeared pointed out by jcristau on irc..
107795
107796commit a63ee90bc2d490f6c5c1802c164391963cf6c1d9
107797Author: Dave Airlie <airlied@pegasus.(none)>
107798Date: Tue Mar 27 11:05:52 2007 +1000
107799
107800 gl: update for latest mesa glsl-compiler merge
107801
107802commit d387a3ddf76716791e5e8b8f0954ca0df3c579d6
107803Author: Dave Airlie <airlied@pegasus.(none)>
107804Date: Tue Mar 27 11:00:13 2007 +1000
107805
107806 fix loading of GLcore after recent loading changes
107807
107808commit 92ba435bd9aa7b6eca9aef8e5193576ef62fc9db
107809Author: Eric Anholt <eric@anholt.net>
107810Date: Mon Mar 26 12:44:58 2007 -0700
107811
107812 Update xorg.conf manpage for new RandR 1.2 monitor options.
107813
107814commit 0331c6a8fcba1dc27ef45aaf5f694799d0085413
107815Merge: 84a066c f7c5aa0
107816Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107817Date: Mon Mar 26 10:28:05 2007 -0400
107818
107819 Merge branch 'master' into XACE-SELINUX
107820
107821commit f7c5aa0dc0fa3569a2ee412c4f996960f936b6ed
107822Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107823Date: Mon Mar 26 10:21:44 2007 -0400
107824
107825 Remove dead NEED_DBE_BUF_BITS code.
107826
107827commit 2e3cc861f90415f200826bc71dab6298d759c42b
107828Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107829Date: Sun Mar 25 22:01:34 2007 -0400
107830
107831 Since ddc, i2c, and ramdac are in core now, remove their ModuleData stubs.
107832
107833commit e88fa75c9b468b88bb7b87b1da235c6eb2fe8164
107834Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107835Date: Sun Mar 18 17:39:08 2007 -0400
107836
107837 Static cleanup on Xi/
107838
107839commit 4b5802ddbd45271be3cadeae0a83a6742df2515b
107840Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107841Date: Sun Mar 25 17:57:54 2007 -0400
107842
107843 General DIX static and dead code cleanup.
107844
107845commit 04b87d6dfae02e4ecdb5216d12c6cdafd1e8c2b4
107846Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107847Date: Sun Mar 25 17:57:22 2007 -0400
107848
107849 Static and dead code cleaup for Xext/
107850
107851commit af769892a91c9af59de53ca3bcd77fc4967daffb
107852Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107853Date: Sun Mar 25 17:56:32 2007 -0400
107854
107855 Static and dead code cleanup from mi/
107856
107857commit 62224e39727fd6f1cf11a461983662f615a9fea1
107858Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107859Date: Sun Mar 25 17:55:15 2007 -0400
107860
107861 Static cleanup for xf86 ddx.
107862
107863commit e8bc1988d9ff10b65717574175f70df3c4d6334d
107864Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107865Date: Sun Mar 25 15:13:05 2007 -0400
107866
107867 Un-staticise VTSwitchEnabled, since kbd wants it apparently.
107868
107869commit 70e493d223b1e943e652191150bd0b7e1a6ebcfb
107870Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107871Date: Sun Mar 25 14:55:28 2007 -0400
107872
107873 Static and dead code cleanup over afb/
107874
107875commit f36bf1a3e4ce9465ea4a6159c209924a3cafbe58
107876Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107877Date: Sun Mar 25 12:28:13 2007 -0400
107878
107879 Delete a dead file.
107880
107881commit 9a0f25de7ca3c68af867b38936103d17daa92ac6
107882Author: Adam Jackson <ajax@benzedrine.nwnk.net>
107883Date: Sun Mar 25 12:27:01 2007 -0400
107884
107885 Static cleanups, dead code deletion.
107886
107887commit ac2356843e38b3400142bc54b65393c12976fc07
107888Author: Peter Hutterer <peter@cs.unisa.edu.au>
107889Date: Sun Mar 25 09:41:33 2007 +0930
107890
107891 dix: Increase allocation size for core keyboard keymap to avoid buffer overrun when copying keymap from extension devices.
107892
107893commit 5f9c10e8da0d9d96cc00a38cdf9a8c3030063e03
107894Author: Peter Hutterer <peter@cs.unisa.edu.au>
107895Date: Sun Mar 25 15:00:03 2007 +0930
107896
107897 Clean up DeviceIntPtr, prepare for Server 1.4 ABI. Move sprite-related
107898 information into a new SpriteInfoRec, remove isMPDev flag.
107899
107900commit e54311395e1e7ae458c8ce59da0914706c4a04f6
107901Author: Peter Hutterer <peter@cs.unisa.edu.au>
107902Date: Sun Mar 25 09:41:33 2007 +0930
107903
107904 dix: Increase allocation size for core keyboard keymap to avoid buffer
107905 overrun when copying keymap from extension devices.
107906
107907commit 1072b88a8f352484e70bc749e300c936e5600480
107908Author: Dave Airlie <airlied@linux.ie>
107909Date: Sun Mar 25 10:06:00 2007 +1000
107910
107911 loader: fix already built-in message
107912
107913commit 804080a7096347d48c686f2c8fbfd06326bce400
107914Author: Keith Packard <keithp@neko.keithp.com>
107915Date: Fri Mar 23 23:41:36 2007 -0700
107916
107917 Make pending properties force mode set. And, remove AttachScreen calls.
107918
107919 Yes, two changes in one commit. Sorry 'bout that.
107920
107921 The first change ensures that when pending property values have been
107922 changed, a mode set to the current mode will actually do something, rather
107923 than being identified as a no-op. In addition, the driver no longer needs to
107924 manage the migration of pending to current values, that is handled both
107925 within the xf86 mode setting code (to deal with non-RandR changes) as well
107926 as within the RandR extension itself.
107927
107928 The second change eliminates the two-call Create/AttachScreen stuff that was
107929 done in a failed attempt to create RandR resources before the screen
107930 structures were allocated. Merging these back into the Create function is
107931 cleaner.
107932 (cherry picked from commit 57e87e0d006cbf1f5b175fe02eeb981f741d92f0)
107933
107934 Conflicts:
107935
107936 randr/randrstr.h
107937 randr/rrcrtc.c
107938
107939 I think master and server-1.3-branch are more in sync now.
107940
107941commit 1f77120775dc05fc84a00dd55190af2fa50ae509
107942Author: Keith Packard <keithp@neko.keithp.com>
107943Date: Fri Mar 23 14:39:10 2007 -0700
107944
107945 Ensure that crtc desired values track most recent mode.
107946
107947 desiredX and desiredY were not recorded during xf86InitialConfiguration.
107948 desiredX, desiredY and desiredRotation were not recorded during
107949 xf86SetSingleMode.
107950 (cherry picked from commit 36e5227215e0912ddf8a010db042467f00efe0fc)
107951
107952commit 84a066cc88fe4326ddacd04ab5e1158a80571c33
107953Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
107954Date: Fri Mar 23 10:33:53 2007 -0400
107955
107956 xace: pass serverClient as default argument to dixChangeWindowProperty
107957 instead of NullClient.
107958
107959commit 476f2b5aefa518262b69e487555e6094818d857a
107960Author: Keith Packard <keithp@guitar.keithp.com>
107961Date: Fri Mar 23 01:17:14 2007 -0700
107962
107963 Incorrect extra memory copy in RRChangeOutputProperty.
107964
107965 Left over from previous version of the code, this memmove will break when
107966 the mode is not Replace.
107967 (cherry picked from commit 945aa0aa556429b50dea8e8ebc0008304b093eb7)
107968
107969commit 7093367c3976bef5b9d219d9f2a7dc7dd3eeb091
107970Author: Keith Packard <keithp@guitar.keithp.com>
107971Date: Fri Mar 23 01:05:55 2007 -0700
107972
107973 Fix Pending property API, adding RRPostPendingProperty.
107974
107975 Pending Properties take effect when the driver says they do, so provide an
107976 API to tell DIX when a property effect is made. Also, allow driver
107977 to reject property values in RRChangeOutputProperty.
107978 (cherry picked from commit 8eb288fbd69e2ffd02521d2c6a964c8180d08ec8)
107979
107980commit 86d76390eb182f271f5fa5dc19205e97a867f7e7
107981Author: Keith Packard <keithp@guitar.keithp.com>
107982Date: Fri Mar 23 01:03:40 2007 -0700
107983
107984 Make sure RandR events are delivered from RRCrtcSet.
107985
107986 Some paths were skipping the event delivery stage.
107987 (cherry picked from commit 9ca7ba5d6012295a77ed773c656e786440da973d)
107988
107989commit 510eaa346e68fd82c852c7b41fb0e2c5be12da78
107990Author: Keith Packard <keithp@guitar.keithp.com>
107991Date: Fri Mar 23 00:59:11 2007 -0700
107992
107993 Clean up xf86CrtcRec and xf86OutputRec objects at CloseScreen.
107994
107995 Erase pointers to structures which are freed at server reset time.
107996 (cherry picked from commit 492c768065f49306a2194a88edf96b85de0ff4ff)
107997
107998commit 479b2be4badab0a67b1f091feb83c1364e27d783
107999Author: Keith Packard <keithp@guitar.keithp.com>
108000Date: Fri Mar 23 00:57:18 2007 -0700
108001
108002 Clear allocated RandR screen private structure.
108003
108004 Use xcalloc instead of xalloc when allocating this structure to ensure
108005 consistent contents at startup.
108006 (cherry picked from commit 16f4c0c1750824f2e5a001cef82a4122a7a2beb0)
108007
108008commit b63e0d2545bb75e14d9de019a88f31e20a2f7377
108009Author: Keith Packard <keithp@guitar.keithp.com>
108010Date: Tue Mar 20 07:17:27 2007 -0700
108011
108012 Clean up Rotate state on server reset.
108013
108014 The rotation state is stored in the xf86_config structure which is not
108015 re-initialized at server reset time. Clean it up at CloseScreen time.
108016 (cherry picked from commit f8db7665dcd7af78ca4db2461e0bf787ec662cb1)
108017
108018commit e1cc68add0bcdd5e0e4e15cf6ee8a3da136d3534
108019Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108020Date: Thu Mar 22 17:33:16 2007 -0400
108021
108022 xace: drop the name argument from the property callback.
108023
108024commit c9fb8a35332d101897607d8f06ed5a6512eac7cf
108025Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108026Date: Thu Mar 22 17:23:26 2007 -0400
108027
108028 dix: move access codes to separate header file, add DixCreateAccess.
108029
108030commit 1b766ffc0647d5e9a9bf6938d33548d977b5535e
108031Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108032Date: Thu Mar 22 15:55:35 2007 -0400
108033
108034 dix: reorganize property code to better support xace hook; requires new API for
108035 changing a property, dixChangeWindowProperty, taking an additional client argument.
108036
108037commit 1b58304ac837735920747ed0f0d10ba331bdaeb7
108038Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108039Date: Thu Mar 22 13:06:50 2007 -0400
108040
108041 xace: add new argument to property hook for property structure itself.
108042
108043commit 5486be4898766205149fadce71529724eb78fbf3
108044Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108045Date: Thu Mar 22 10:59:21 2007 -0400
108046
108047 dix: devPrivates support for PropertyRec.
108048
108049commit f8cd19bd1692d12a6047c088a626a6ae715d469d
108050Author: Peter Hutterer <peter@cs.unisa.edu.au>
108051Date: Thu Mar 22 17:47:01 2007 +1030
108052
108053 Xi: add GetClientPointer and GetPointerPairing requests.
108054
108055 dix: use PickKeyboard for ProcSetInputFocus and ProcGetInputFocus instead
108056 of the core keyboard.
108057
108058commit a3296d111dc4d76aa3afa7e338cbab93eb390ec4
108059Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108060Date: Wed Mar 21 17:01:26 2007 -0400
108061
108062 xace: add access_mode argument to selection hook.
108063
108064commit 4c1fb8069d5dd30a73277698503e9dcc2e9d64c6
108065Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108066Date: Wed Mar 21 16:17:14 2007 -0400
108067
108068 dix: add new selection fields supporting redirection.
108069 This is a minor ABI break.
108070
108071commit 4fa482b4be1150bcffeabb64d018c00ac5951e41
108072Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108073Date: Wed Mar 21 14:49:56 2007 -0400
108074
108075 xace: bump major version since the hooks have changed.
108076
108077commit 9c144f8ac5cea25deaa543767dbaf371d029c608
108078Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108079Date: Wed Mar 21 14:39:00 2007 -0400
108080
108081 xace: add XACE_SELECTION_ACCESS hook for selection redirection/access.
108082
108083commit 3e9f7a5504ab41d845e88f293d8498c963d8a7d8
108084Author: Daniel Stone <daniel@fooishbar.org>
108085Date: Wed Mar 21 02:35:31 2007 +0200
108086
108087 XFree86 DGA: Guard against NULL pointer dereferences.
108088 Ass, u, me ...
108089
108090commit f292de2ef13dc994a38029cee9e2642576893332
108091Author: Daniel Stone <daniel@fooishbar.org>
108092Date: Wed Mar 21 02:04:12 2007 +0200
108093
108094 XKB: Fix size_syms calculation bug
108095
108096 Apparently it needed to be nSyms*15/10, not *12/10; make it match the
108097 other allocation code.
108098
108099commit f34b9a20b0181d3c2641c305e91180711afbd4b9
108100Author: Daniel Stone <daniel@fooishbar.org>
108101Date: Wed Mar 21 02:03:37 2007 +0200
108102
108103 XKB: Be a tiny bit more conservative with type allocation
108104
108105 Make sure size_types will _always_ be 0 if we don't have any types.
108106
108107commit 021fc5cb2cb4a7972b4a6fcb570c1da92787d68d
108108Author: Adam Jackson <ajax@benzedrine.nwnk.net>
108109Date: Sun Mar 18 16:31:19 2007 -0400
108110
108111 Static markup and dead code cull over xkb/.
108112
108113 The former <X11/extensions/XKBsrv.h> has been pulled into the server now as
108114 include/xkbsrv.h, and the world updated to look for it in the new place,
108115 since it made no sense to define server API in an extension header. Any
108116 further work along this line will need to do similar things with XKBgeom.h
108117 and friends.
108118
108119commit 9398d62f27ee1b287e4458fd8b011c10f7b59efd
108120Author: Daniel Stone <daniel@fooishbar.org>
108121Date: Wed Mar 21 00:18:24 2007 +0200
108122
108123 XFree86 input: Add backwards compatibility for motion history
108124 Add the old motion history API back, as a shim around the new mi API.
108125
108126commit 0f75c47e0c5f4b2778930a6fabf894fc1dffd9d3
108127Author: Daniel Stone <daniel@fooishbar.org>
108128Date: Wed Mar 21 00:12:02 2007 +0200
108129
108130 xfree86 input: Re-enable DGA support
108131 Re-enable DGA support for relative mouse motion.
108132
108133commit 80d29475b9a2ebbb303a8e324e09a15c528d5556
108134Author: Daniel Stone <daniel@fooishbar.org>
108135Date: Wed Mar 21 00:10:38 2007 +0200
108136
108137 mieq: Allow event handlers for arbitrary events to be set
108138 Allow arbitrary events to use mieq by letting custom handlers be set.
108139
108140commit b8df961843a95b29258ae9c5d46ccfc620d8de1c
108141Author: Alan Coopersmith <alan.coopersmith@sun.com>
108142Date: Mon Mar 19 18:03:26 2007 -0700
108143
108144 Define XF86PM on Solaris x86 builds now that we have sun_apm.c
108145
108146commit 2945deba1d4a7dce4f6dd0c568297a1c537fdfb4
108147Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108148Date: Mon Mar 19 17:09:10 2007 -0400
108149
108150 xace: drop XACE_WINDOW_INIT hook, it has been superseded by ResourceStateCallback.
108151
108152commit 78c962da76efe644b8d485265f1ecdda84b45d27
108153Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108154Date: Mon Mar 19 17:04:51 2007 -0400
108155
108156 xselinux: use the new ResourceStateCallback instead of the XACE_WINDOW_INIT hook.
108157
108158commit 6a89106e9c963a495fd40427d242ba0abd44f764
108159Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108160Date: Mon Mar 19 16:51:29 2007 -0400
108161
108162 xselinux + security: remove confusing CALLBACK macro.
108163
108164commit 0acf993e2878d0866fd4e2db8252c1ea6841e494
108165Author: Peter Hutterer <peter@cs.unisa.edu.au>
108166Date: Mon Mar 19 16:11:01 2007 +1030
108167
108168 dix: Move sprite initalization/pairing from ActivateDevice to
108169 EnableDevice.
108170 zero out spriteTrace, otherwise GetCurrentRootWindow() may return
108171 invalid data.
108172
108173 Evdev adds previously hotplugged devices on its own when rescanning,
108174 and ActivateDevice does not get called for those devices. Sprites need
108175 to be initialized or paired before the device gets started, so the
108176 safest place to do so is EnableDevice (which actually does get called
108177 by the evdev driver).
108178
108179commit 015d728bcde5b16a72a976579755421e3023626b
108180Merge: 65b0eb6 b167583
108181Author: Peter Hutterer <peter@cs.unisa.edu.au>
108182Date: Mon Mar 19 09:42:56 2007 +1030
108183
108184 Merge branch 'master' into mpx
108185
108186 Conflicts:
108187
108188 dix/devices.c
108189 dix/events.c
108190 mi/misprite.c
108191
108192commit 720f302d241e88e6e9f2962207da1aa9a79728b7
108193Author: Keith Packard <keithp@neko.keithp.com>
108194Date: Sat Mar 17 20:14:05 2007 -0700
108195
108196 Slow down DDC I2C bus using a RiseFallTime of 20us for old monitors.
108197
108198 This time value makes the bus run slowly enough for even the least reliable
108199 of monitors. Thanks to Pavel Troller for finding the necessary change.
108200
108201commit b5a8a71e64c76b8dd42962cbd7984215c6ce4aa8
108202Author: Keith Packard <keithp@neko.keithp.com>
108203Date: Sat Mar 17 17:26:11 2007 -0700
108204
108205 Remove extra (and wrong) I2C ByteTimeout setting in DDC code.
108206
108207 The DDC code sets the I2C timeouts to VESA standards, except that it had an
108208 extra setting of the ByteTimeout value which was wrong (off by a factor of
108209 50). Removing this should help DDC work on many more monitors. Note that the
108210 Intel driver duplicated these settings, along with the error. Yay for cult
108211 and paste coding.
108212
108213commit 2489dae9f7def788910eee5733931392df83a0d6
108214Author: Keith Packard <keithp@guitar.keithp.com>
108215Date: Thu Mar 15 20:26:07 2007 -0700
108216
108217 Correct ref counting of RRMode structures
108218
108219 RRModes are referenced by the resource db, RROutput and RRCrtc structures.
108220 Ensure that the mode reference count is decremented each time a reference is
108221 lost from one of these sources. The missing destroys were in
108222 RRCrtcDestroyResource and RROutputDestroyResource, which only happen at
108223 server reset time, so modes would be unavailable in subsequent server
108224 generations.
108225
108226commit 9d0c3b52f25df89738fb1a62ccffda8c8cbb4689
108227Author: Keith Packard <keithp@guitar.keithp.com>
108228Date: Tue Feb 20 23:04:26 2007 -0800
108229
108230 Eliminate RRModeRec devPrivate field.
108231
108232 The xf86 mode setting code was mis-using this field to try and store a
108233 pointer to a DisplayModeRec, however, each output has its own copy of every
108234 DisplayModeRec leaving the one in in the RRModeRec devPrivate field pointing
108235 at a random DisplayModeRec.
108236
108237 Instead of attempting to rectify this, eliminating the devPrivate entirely
108238 turned out to be very easy; the DDX code now accepts an arbitrary RRModeRec
108239 structure and set that to the hardware, converting it on the fly to a
108240 DisplayModeRec as needed.
108241 (cherry picked from commit 3506b9376c2b0db09bfff58d64e07af88a6e8195)
108242
108243commit 2c93083edd29a65e73bb2e8eff9d353e92845c9b
108244Author: Keith Packard <keithp@guitar.keithp.com>
108245Date: Sun Feb 18 23:49:38 2007 -0800
108246
108247 Add support for user-defined modelines in RandR.
108248
108249 The RandR protocol spec has several requests in support of user-defined
108250 modes, but the implementation was stubbed out inside the X server. Fill out
108251 the DIX portion and start on the xf86 DDX portion. It might be necessary to
108252 add more code to the DDX to insert the user-defined modes into the output
108253 mode list.
108254 (cherry picked from commit 63cc2a51ef87130c632a874672a8c9167f14314e)
108255
108256 Conflicts:
108257
108258 randr/randrstr.h
108259
108260 Updated code to work in master with recent security API changes.
108261
108262commit b167583fe234a536aa6187e17f21da85be4f0b7f
108263Author: Peter Hutterer <peter@cs.unisa.edu.au>
108264Date: Wed Nov 15 17:50:02 2006 +1030
108265
108266 o fix minor error in comment for GetPointerEvents()
108267
108268commit 65b0eb60b0b9e6405aa2e3d4c712ec4d2f0da070
108269Author: Peter Hutterer <peter@cs.unisa.edu.au>
108270Date: Fri Mar 16 15:52:24 2007 +1030
108271
108272 Xi: move IsPointerDevice into dix.
108273 dix: IsPointerDevice and IsKeyboardDevice, use same ways to identify type
108274 of device as XI does for the XListInputDevices reply.
108275 Autopair each non-pointer device with VCP when activating, pair with
108276 real device after activation.
108277 Don't return non-keyboard devices when calling GetPairedKeyboard or
108278 PickKeyboard, otherwise we segfault for 'evdev brain'.
108279
108280commit 3bffb281260476d2f74f0bf451d85d2f7cacd6c4
108281Author: Keith Packard <keithp@neko.keithp.com>
108282Date: Thu Mar 15 16:16:16 2007 -0700
108283
108284 Don't wedge when rotating more than one CRTC.
108285
108286 Rotation block handler was re-registering the rotation damage structure,
108287 creating an infinite loop in the damage code. Track registration of the
108288 damage structure to avoid this.
108289 (cherry picked from commit b14f003b0ed1252766c9e3b1c086ea2809521047)
108290
108291commit 9562b6abe1da566cf73a08c4f4c4339fb67fbc71
108292Author: Keith Packard <keithp@neko.keithp.com>
108293Date: Thu Mar 15 10:50:45 2007 -0700
108294
108295 Allow xf86_reload_cursors during server init.
108296
108297 xf86_reload_cursors is supposed to be called from the crtc mode setting
108298 commit hook; as that happens during server initialization, check for this
108299 case.
108300 (cherry picked from commit 5b77bf2d020b1ee56c1c5f2db089a8f7f64a76a6)
108301
108302commit 3b71b0f89f1db837da91650baa0ef4bb7ef2e98f
108303Author: Eric Anholt <eric@anholt.net>
108304Date: Thu Mar 15 13:21:00 2007 -0700
108305
108306 Set the RandR version returned, rather than just passing the proto's version.
108307
108308commit 2fe74ef339c3a4902ae8214f5a0454662895422c
108309Author: Matthias Hopf <mhopf@suse.de>
108310Date: Thu Mar 15 16:56:01 2007 +0100
108311
108312 Fix calculations in x86 emulator for the long long case (Andreas Schwab).
108313
108314commit 456f429ad655ab4d80e30c58291d801966a613b2
108315Author: Peter Hutterer <peter@cs.unisa.edu.au>
108316Date: Thu Mar 15 21:48:19 2007 +1030
108317
108318 dix: fix: don't return when device is same as argument.
108319 fix: set access default rule to 0 on init.
108320
108321commit ae75019ccf1edac9e8be31b6a96293624f672ccb
108322Author: Keith Packard <keithp@guitar.keithp.com>
108323Date: Wed Mar 14 23:59:29 2007 -0700
108324
108325 Create driver-independent CRTC-based cursor layer.
108326
108327 This moves most of the cursor management code out of the intel driver and
108328 into the general server code. Of course, the hope is that this code will be
108329 useful for other driver writers as well.
108330
108331 Check out xf86Crtc.h for the usage information, making sure you add the
108332 needed hooks to the crtc funcs structure for your driver.
108333 (cherry picked from commit 4d81c99a4660a0bf9014f789de55edabd185bd14)
108334
108335commit c5b07fb717289f61b54d7b093421bcb92124b839
108336Author: Peter Hutterer <peter@cs.unisa.edu.au>
108337Date: Wed Mar 14 17:42:30 2007 +1030
108338
108339 kdrive: update kdrive to new mi API with additional device argument. Includes
108340 an update of ati_cursor.c and i810_cursor.c
108341
108342commit c3eb248cf93a3afd9168acfb88254606beddd7a8
108343Author: Peter Hutterer <peter@cs.unisa.edu.au>
108344Date: Wed Mar 14 13:21:33 2007 +1030
108345
108346 dix: fix: don't skip first item in list when seeking for a device cursor.
108347
108348commit ce9409acebd994cf1809050f2cc8e1db66130f12
108349Author: Peter Hutterer <peter@cs.unisa.edu.au>
108350Date: Tue Mar 13 19:16:56 2007 +1030
108351
108352 dix: Get the state from the paired device and the real device to fill the
108353 state field in the event, rather than using the virtual devices.
108354 ProcessPointerEvent: name argument "device" instead of "other".
108355 Add GetPairedKeyboard().
108356
108357commit 4bf1b280f7cb676ec2b172f26dd2ad9bac2eb2ca
108358Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
108359Date: Fri Mar 9 14:18:14 2007 +0000
108360
108361 Set pScreen on context
108362
108363commit c366b82bd50066019cf82b3464445d5bc27d6f9f
108364Author: Jay Estabrook <Jay.Estabrook@hp.com>
108365Date: Fri Mar 9 12:26:55 2007 +0000
108366
108367 Ensure domain is stripped from the bus ID.
108368
108369commit 577464af4362e5a32cf7165b5128655dd86c6200
108370Author: Peter Hutterer <peter@cs.unisa.edu.au>
108371Date: Fri Mar 9 17:13:05 2007 +1030
108372
108373 dix: restore commit b3b2a6a0d43d1724e04d69588f8a55c3270e5523 that for some
108374 reason got wiped.
108375 fix ProcGrabKeyboard to use PickKeyboard
108376 fix PickKeyboard to actually work.
108377
108378commit 07806f4081f8dcd3b5733b590eb6e5b4cae734ad
108379Author: Peter Hutterer <peter@cs.unisa.edu.au>
108380Date: Fri Mar 9 15:45:25 2007 +1030
108381
108382 Xi: Add SetClientPointer request handling.
108383
108384commit 405483496538f1c82cbd7fe1e76c5d94e1a90525
108385Author: Peter Hutterer <peter@cs.unisa.edu.au>
108386Date: Fri Mar 9 14:16:23 2007 +1030
108387
108388 mi: remove 'register' keywords.
108389
108390commit fe05ba75a10ec080e7ec34bff6936103185586b3
108391Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108392Date: Thu Mar 8 12:14:30 2007 -0500
108393
108394 devPrivates rework: pass address of pointer to private callbacks instead of
108395 the pointer itself.
108396
108397commit 18339375cd332f0ab1cbdade3dcd9140212ce1ca
108398Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108399Date: Thu Mar 8 12:14:06 2007 -0500
108400
108401 xselinux: remove context validation function for now.
108402
108403commit b9cff1670f29949a5bc41afc19aca443f434febb
108404Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108405Date: Thu Mar 8 12:13:52 2007 -0500
108406
108407 Add ResourceStateCallback similar in function to ClientStateCallback.
108408
108409commit 2fcb45eb5dc1803b372df8b5765f6a43bea83611
108410Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108411Date: Thu Mar 8 12:13:36 2007 -0500
108412
108413 devPrivates rework: redo field offset registration, drop RC_PRIVATES class.
108414
108415commit 947f8d249bac61beb10669d935888c4c280b5062
108416Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108417Date: Thu Mar 8 12:13:18 2007 -0500
108418
108419 devPrivates rework: redo interface again, dropping parent and type parameters
108420 as well as preallocation routine.
108421
108422commit c45f6762080ef00b41d9f73441a9f0e605253008
108423Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108424Date: Wed Mar 7 11:22:42 2007 -0500
108425
108426 devPrivates rework: hook up new mechanism in backwards-compatibility mode
108427 on existing structures that support devPrivates.
108428
108429commit b3b2a6a0d43d1724e04d69588f8a55c3270e5523
108430Author: Peter Hutterer <peter@cs.unisa.edu.au>
108431Date: Thu Mar 8 22:16:17 2007 +1030
108432
108433 dix: Check if client has a grab on any other device before mouse/kbd
108434 event is delivered to the client. Client's don't cope well with
108435 receiving core events from other devices when having a grab on a
108436 pointer or keyboard.
108437
108438commit 63169ce52d354b4345dcfc46b89f0ea88379718f
108439Author: Peter Hutterer <peter@cs.unisa.edu.au>
108440Date: Thu Mar 8 17:50:19 2007 +1030
108441
108442 dix: remove 'register' keyword for all variables.
108443
108444commit ceca5670fee99b5feceaa2453f1ac32d1bfe7dcd
108445Author: Peter Hutterer <peter@cs.unisa.edu.au>
108446Date: Thu Mar 8 17:17:23 2007 +1030
108447
108448 dix: Each device needs to differ between a core grab and an XI grab,
108449 otherwise a Xi grab may overwrite or release a core grab.
108450 Replace grab and associates with coreGrab and deviceGrab structures,
108451 adjust rest of dix/Xi/etc to compile.
108452
108453 xfree86: Don't check for core devices, we'll have the virtual ones anyway.
108454 If we check, the first mouse device is duplicated and sends
108455 double events.
108456
108457commit 40ae4f246d8818410490236ab183204a84765629
108458Author: Keith Packard <keithp@guitar.keithp.com>
108459Date: Wed Mar 7 20:52:31 2007 -0800
108460
108461 Remove stale monitor data when output becomes disconnected.
108462
108463 Remove parsed EDID and EDID property from disconnected outputs.
108464 (cherry picked from commit ae9d5aa479dd50cc81b755079fcf96a0d02f135a)
108465
108466commit b5fde366e2e21234ac0b81222fd5c42ca3e49cba
108467Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108468Date: Wed Mar 7 12:29:55 2007 -0500
108469
108470 Properly free device devPrivates - memory leak fix.
108471
108472commit a3d2c5d622d9ca36d6fa2966aff09524e3ea39ac
108473Author: Adam Jackson <ajax@benzedrine.nwnk.net>
108474Date: Wed Mar 7 11:02:47 2007 -0500
108475
108476 XORG_VERSION_CURRENT, not XF86_VERSION_CURRENT.
108477
108478 If only this was the least wrong thing in this code.
108479
108480commit aaef4d6a4121d9341b670a0ce8fabc3b491049cf
108481Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108482Date: Wed Mar 7 09:57:02 2007 -0500
108483
108484 devPrivates rework: move reset functions into a single call.
108485
108486commit bb111291d854b4329e47367ce3c67e8a2785e6e9
108487Merge: 74175e0 e9bfb2b
108488Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108489Date: Wed Mar 7 09:03:46 2007 -0500
108490
108491 Merge branch 'master' into XACE-SELINUX
108492
108493commit 537bc2ead4d154552cbdc3a19e335f82af63792c
108494Author: Peter Hutterer <peter@cs.unisa.edu.au>
108495Date: Wed Mar 7 18:19:20 2007 +1030
108496
108497 Xi: Don't deactivate core grabs from non-core button/key releases.
108498
108499 dix: set coreGrab flag for grabs caused by GrabPointer and button presses.
108500 remove virtual core devices from device list, only real devices are in
108501 the list now.
108502 Auto-pair each keyboard with a real pointer if one is available,
108503 provides multiple keyboards automatically after startup if devices
108504 are configured.
108505 fix GuessFreePointerDevice() to do what it's supposed to do.
108506
108507 mi: fix: call miPointerMove from miPointerWarpCursor.
108508 fix: remove unused id field from miCursorInfoRec
108509 don't update sprite for virtual core pointer.
108510
108511commit e9bfb2b3d7dfaafd90d2ad0fa3d0e1acced4380b
108512Author: Keith Packard <keithp@neko.keithp.com>
108513Date: Tue Mar 6 23:19:30 2007 -0800
108514
108515 Add hw/xfree86/docs/README.modes, documenting new mode setting APIs.
108516
108517 This document covers both API and xorg.conf usage of the new mode setting
108518 APIs.
108519 (cherry picked from commit a59c31b0f7b94ed1f395c7586c37ef5fe7ba2a25)
108520
108521commit 72a23d88d73a8c72ed18847b004db05092d3e7be
108522Author: Keith Packard <keithp@guitar.keithp.com>
108523Date: Tue Mar 6 23:15:34 2007 -0800
108524
108525 Add xf86CrtcScreenInit to share initialization across drivers.
108526
108527 xf86CrtcScreenInit performs initialization that needs to happen at
108528 ScreenInit time.
108529 (cherry picked from commit 558a4f5588ad2ec11254e0b5d6ce9515b137369e)
108530
108531commit 81526232bc0119d2ec7b8590be4f78cf066ae359
108532Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108533Date: Tue Mar 6 17:19:11 2007 -0500
108534
108535 remove PIXPRIV check as this flag is always set.
108536
108537commit a7cd53deb99957dec27a55ffd75e548b322ae0ce
108538Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108539Date: Tue Mar 6 15:32:13 2007 -0500
108540
108541 remove PIXPRIV checks as this flag is always set.
108542
108543commit 024bbc7cbb924daaf3e305ddfc8e74509acd1e15
108544Author: Eric Anholt <eric@anholt.net>
108545Date: Tue Mar 6 16:18:59 2007 -0800
108546
108547 Bug #9931: Fix linear allocations with a non-1-byte granularity.
108548
108549 This was introduced in 83080809f9a1c1d24b0318e54632f25f5940da25. Instead of
108550 aligning the offset, it doubled it. Results were appropriately spectacular.
108551
108552commit 9d94c137596d3f9d9118ec70455b7a30b3582046
108553Author: Ben Byer <bbyer@apple.com>
108554Date: Tue Mar 6 11:09:30 2007 -0800
108555
108556 updated todo list
108557
108558commit 81d581e655fc989da3be4256b83849a63b8607b7
108559Merge: a05ffca d5aba03
108560Author: Ben Byer <bbyer@bbyer.(none)>
108561Date: Tue Mar 6 10:37:29 2007 -0800
108562
108563 Merge branch 'master' of git+ssh://bbyer@git.freedesktop.org/git/xorg/xserver
108564
108565commit a05ffca8dd0da9bdb5c1bf4c481028aeabf21e34
108566Author: Ben Byer <bbyer@bbyer.(none)>
108567Date: Tue Mar 6 10:36:51 2007 -0800
108568
108569 rewrote event handling, Xquartz now has working mouse and keyboard. use it\!
108570
108571commit d5aba03feff41722c72b4c6193f09d141cbf1678
108572Author: Drew Parsons <dparsons@debian.org>
108573Date: Tue Mar 6 23:53:23 2007 +1100
108574
108575 Xprint: shorten font filename to fit in tar length limit
108576
108577 The length of the Xprint font file NewCenturySchlbk-BoldItalic.pmf
108578 pushes the full path over the traditional 100 character limit for
108579 tarballs (when module version number is included). Shorten it to
108580 NewCentSchlbk-BoldItal.pmf to get back below the limit and rename
108581 other font files in that family to match.
108582
108583commit 3206e9225897989638ad553e1f392b918ac4d21f
108584Author: Ben Byer <bbyer@bbyer.(none)>
108585Date: Tue Mar 6 02:31:59 2007 -0800
108586
108587 moved new event-handling code from X11Application.m to darwinEvents.c in preparation for making all Darwin servers use it
108588
108589commit 0ccd1443fd6db397b42e5b99ce733ce1316c785e
108590Merge: ec1ef8a 9b6bb06
108591Author: Ben Byer <bbyer@bbyer.(none)>
108592Date: Tue Mar 6 01:04:50 2007 -0800
108593
108594 Merge branch 'master' of git+ssh://bbyer@git.freedesktop.org/git/xorg/xserver
108595
108596commit ec1ef8a56d6217ca2b04899043874ce0bcad9784
108597Author: Ben Byer <bbyer@bbyer.(none)>
108598Date: Tue Mar 6 00:57:23 2007 -0800
108599
108600 Fixed Darwin's Makefile.am to fix a problem building X11.app
108601
108602commit 9b6bb06f13a71f6078f762b4a78fa516faccb638
108603Author: Keith Packard <keithp@guitar.keithp.com>
108604Date: Mon Mar 5 23:49:35 2007 -0800
108605
108606 Allow relative positions to use output names or monitor identifiers.
108607
108608 Previous version used monitor identifiers if present, otherwise output
108609 names. That caused existing working configurations to break when additional
108610 information was added to the configuration file.
108611 (cherry picked from commit 3f5cedf00a82f08a433c95ffbb7f8ac69dcf6a50)
108612
108613commit bed76caa6caaea6a6598755b82a54425a9d9f73e
108614Author: Keith Packard <keithp@guitar.keithp.com>
108615Date: Mon Mar 5 23:36:00 2007 -0800
108616
108617 Use EDID data to set screen physical size at server startup.
108618
108619 Screen physical size is set to a random value before the RandR code gets
108620 control, override that and reset it to a value based on the compat_output
108621 physical size (if available). If that output has no physical size, just use
108622 96dpi as the default resolution and set the physical size as appropriate.
108623 (cherry picked from commit 843077f23a1b49bd712d931421753e3a09d4008c)
108624
108625commit 47f8361c3a64834587e54507653d8d5b258c2530
108626Author: Keith Packard <keithp@guitar.keithp.com>
108627Date: Mon Mar 5 22:07:01 2007 -0800
108628
108629 Add xf86SetDesiredModes to apply desired modes to crtcs.
108630
108631 xf86SetDesiredModes applies the desired modes to each crtc (as selected by
108632 xf86InitialConfiguration initially and modified by successful mode settings
108633 afterwards). For crtcs without a desired mode, pScrn->currentMode is used to
108634 select something workable.
108635 (cherry picked from commit bcade98ccaa18298d844a606cb44271f0254c185)
108636
108637commit 33d2cf93fb50464941e74efe246b10aee212223a
108638Author: Keith Packard <keithp@neko.keithp.com>
108639Date: Sat Mar 3 23:10:31 2007 -0800
108640
108641 Move xf86SetSingleMode into X server from intel driver.
108642
108643 This function applies a single mode to the screen (as from RandR 1.1,
108644 XFree86-VidModeExtension or XFree86-DGA) using a policy that selects one
108645 output to reconfigure to the requested mode and then makes all other outputs
108646 fit within that size.
108647 (cherry picked from commit 5a595c1f767a8d666348b845d18934aee0cfe38f)
108648
108649commit 689d52b6242434507a64a8fff27b01607628c393
108650Author: Jens Granseuer <jensgr@gmx.net>
108651Date: Mon Mar 5 15:31:44 2007 -0800
108652
108653 Bugzilla #7145: fix build with gcc 2.95
108654
108655 Bugzilla #7145: <http://bugs.freedesktop.org/show_bug.cgi?id=7145>
108656 Patch #8987: <http://bugs.freedesktop.org/attachment.cgi?id=8987>
108657
108658commit 74175e0af74c530cb712a6772d3c5d61d1be9748
108659Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108660Date: Mon Mar 5 12:34:37 2007 -0500
108661
108662 devPrivates rework: remove some debugging code from dixFreePrivates.
108663
108664commit 23fc429aad5b2721911862617772c314e1036bb0
108665Merge: e684824 fe7b8f4
108666Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108667Date: Mon Mar 5 12:18:54 2007 -0500
108668
108669 Merge branch 'master' into XACE-SELINUX
108670
108671commit e684824709fa8ffe03dde3c8dfbc58c267515a4f
108672Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108673Date: Thu Mar 1 15:00:02 2007 -0500
108674
108675 devPrivates rework: redo interface and implementation.
108676
108677commit fe7b8f4237874e3e45fe25a6bf06faddfa1ab8e1
108678Author: Ben Byer <bbyer@bbyer.(none)>
108679Date: Mon Mar 5 03:48:27 2007 -0800
108680
108681 began to factor out code to move to darwinEvents.c
108682
108683commit 537dc5ecde46d0525c503d1d2b39b6eb89a1298e
108684Author: Ben Byer <bbyer@bbyer.(none)>
108685Date: Mon Mar 5 02:30:56 2007 -0800
108686
108687 started moving new input code into darwinEvents.c so that it may be shared by the three servers
108688
108689commit 39aa79177196e21bcdbaf8e44adead9ef91e6ee5
108690Author: Peter Hutterer <peter@cs.unisa.edu.au>
108691Date: Mon Mar 5 15:31:16 2007 +1030
108692
108693 dix: Add GuessFreePointerDevice(). Runs through device list and tries to
108694 find a pointer that hasn't been paired yet.
108695
108696 xfree86: Use GuessFreePointerDevice() for newly connected non-sprite devices.
108697
108698commit 8ba5e8d82014b774a52f3e050ddbbb8bde4e0933
108699Author: Dave Airlie <airlied@linux.ie>
108700Date: Mon Mar 5 13:46:41 2007 +1100
108701
108702 add a standard connector type and name for us as an output property
108703
108704commit 2e31872e05c2408d53ba0182bcddc5dabb3615fe
108705Author: Dave Airlie <airlied@linux.ie>
108706Date: Mon Feb 26 09:40:00 2007 +1100
108707
108708 modes: add commit/prepare hooks
108709
108710commit 1f0075786fedde538a95e2f39681052e25021d88
108711Merge: 57aa5e9 12175b6
108712Author: Peter Hutterer <peter@cs.unisa.edu.au>
108713Date: Mon Mar 5 12:37:17 2007 +1030
108714
108715 Merge branch 'master' into mpx
108716
108717 Conflicts:
108718
108719 configure.ac
108720 dix/getevents.c
108721 hw/xfree86/ramdac/xf86Cursor.c
108722 mi/mipointer.c
108723 xkb/xkbUtils.c
108724
108725commit 06b01186f6ae17aafdd1f628c306466ddea9e065
108726Author: Keith Packard <keithp@neko.keithp.com>
108727Date: Sun Mar 4 17:15:24 2007 -0800
108728
108729 Remove debugging ErrorF from rotation code.
108730 (cherry picked from commit e6af7569f201842b4754aec6e72b30dc2daefdfb)
108731
108732commit c14507b6837387d867792a24778786311b2b38d5
108733Author: Keith Packard <keithp@neko.keithp.com>
108734Date: Sun Mar 4 17:06:37 2007 -0800
108735
108736 Handle non-zero origin rotated crtc. Damage crtc area on re-rotate.
108737
108738 Box transformation from source to dest area was broken, leaving the wrong
108739 areas painted when the crtc origin was non-zero.
108740
108741 When rotating from left to right, the pixmap doesn't get reallocated, and so
108742 no damage was left in the pixmap from xf86RotatePrepare. Separately damage
108743 the whole crtc area when this occurs to repaint the area.
108744 (cherry picked from commit 2a50ca2160bc05af1c24421ec079e902ff730277)
108745
108746commit 97978b515b7af5fbaaa32b1729e835f3bfb9f5c6
108747Author: Drew Parsons <drew@pug.localdomain>
108748Date: Sun Mar 4 16:28:54 2007 +1100
108749
108750 Xprint: fix font symlinks
108751
108752 Change symlinks to Xprint base fonts in model/PSdefault using local
108753 relative links. This facilitates moving the Xprint config files, for
108754 instance for FHS compliance placing data files in /usr/share rather
108755 than /usr/lib. Also ensures NewCenturySchlbk-BoldItalic.pmf is
108756 installed.
108757
108758commit 215e3691b76a63e6af19865790193b20b105ec5a
108759Author: Ben Byer <bbyer@xyzzy.local>
108760Date: Sat Mar 3 21:52:56 2007 -0800
108761
108762 stopped using XTrans internals in X11.app because they're apparently no longer public
108763
108764commit ea8dcc458ea8870126cf8d3e21cab9d63d094c5e
108765Author: Ben Byer <bbyer@xyzzy.local>
108766Date: Sat Mar 3 21:51:20 2007 -0800
108767
108768 Makefile fix for X11.app
108769
108770commit 18508212599bf0964c450c69b9790208e5d428be
108771Author: Ben Byer <bbyer@xyzzy.local>
108772Date: Sat Mar 3 21:41:33 2007 -0800
108773
108774 fixed X11.xcodeproj to get CFLAGS and LDFLAGS from autoconf script
108775
108776commit 7f2b9f3790456044d01bf8e6404f9a1239b41da6
108777Author: Ben Byer <bbyer@xyzzy.local>
108778Date: Sat Mar 3 19:27:53 2007 -0800
108779
108780 autoconf fixes for XDarwin (created DARWIN_LIBS)
108781
108782commit 12175b668a94e23994f724b366a691ec312cce69
108783Author: Peter Hutterer <peter@cs.unisa.edu.au>
108784Date: Wed Nov 15 17:50:02 2006 +1030
108785
108786 o fix minor error in comment for GetPointerEvents()
108787
108788commit ea1a72946d1aa4c256e6afb9d834c582ba4ac3a1
108789Author: Aaron Plattner <aplattner@nvidia.com>
108790Date: Wed Feb 28 14:26:47 2007 -0800
108791
108792 Add a canGrow argument to xf86InitialConfiguration.
108793
108794 canGrow indicates to the DDX that the driver can enlarge the desktop via the
108795 xf86_config->funcs->resize hook. If so, xf86InitialConfiguration will set
108796 virtual[XY] to match the configuration it chooses and will leave the crtc config
108797 size ranges alone. If FALSE, it will bloat the screen to fit the largest probed
108798 mode and also set the crtc config max size to limit the desktop to the initial
108799 virtual[XY] size.
108800
108801commit 04d15da95d608766c7832a7aa881be499c1395ba
108802Author: Aaron Plattner <aplattner@nvidia.com>
108803Date: Wed Feb 28 13:36:58 2007 -0800
108804
108805 Add a screen resize hook to xf86CrtcConfigRec.
108806
108807 This hook is called when the DDX needs to resize the screen. The driver is
108808 responsible for changing virtualX and virtualY, along with any other related
108809 screen properties (devPrivate.ptr, devKind, displayWidth, etc.).
108810
108811 Use the size range from the crtc config instead of randrp->virtual[XY] when
108812 reporting the min and max screen sizes to the DDX.
108813
108814commit b11dfac287d65de7b83f63749087cba4e8ddaf4a
108815Author: Matthias Hopf <mhopf@suse.de>
108816Date: Fri Mar 2 12:30:26 2007 +0100
108817
108818 Legacy framebuffer support wasn't compiled if Xorg wasn't explicitly enabled.
108819
108820commit 57aa5e908dc11d5d8c27ed705c526f1416c1e8ad
108821Author: Peter Hutterer <peter@cs.unisa.edu.au>
108822Date: Fri Mar 2 17:14:37 2007 +1030
108823
108824 dix, Xext, Xtrap, Xi: replace inputInfo.pointer with PickPointer where
108825 possible. More replacements to come.
108826
108827commit e43b8a4e40991ca6e545bda4cf9b9bd7a2bf22e2
108828Author: Peter Hutterer <peter@cs.unisa.edu.au>
108829Date: Fri Mar 2 15:21:46 2007 +1030
108830
108831 dix: Add ClientPointer to client, used for picking which pointer to use in
108832 ambiguious request. PickPointer and PickKeyboard are used for getting
108833 the appropriate pointer when situation is unclear.
108834 Fix some issues with InitializeSprite.
108835
108836 dix, xfree86: Remove last traces of InitSprite.
108837
108838commit 2dafc46e3d814e02b25e5a2fa2e931f0257402a8
108839Author: Ben Byer <bbyer@bbyer.apple.com>
108840Date: Thu Mar 1 17:44:39 2007 -0800
108841
108842 Fixed pointer events in Xquartz -- Keyboard events work, but
108843 the keycodes are incorrect.
108844
108845commit 74f1de1de9633119c2cf26086875717181c8a6f7
108846Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108847Date: Thu Mar 1 12:07:33 2007 -0500
108848
108849 devPrivates rework: unhook resource system; will try a different approach.
108850
108851commit 39ecd6fff4f946deebe310b4b26b171c842db223
108852Author: Ben Byer <bbyer@bbyer.(none)>
108853Date: Thu Mar 1 01:45:19 2007 -0800
108854
108855 Rewrote parts of the Xquartz event-handling code (thanks daniels and whot!)
108856 It should still be considered a work in progress, but mouse events almost work.
108857
108858commit ed7ccc481ad1caaa518cafe944c2327a5d0b6c65
108859Author: Ben Byer <bbyer@bbyer.(none)>
108860Date: Thu Mar 1 00:51:10 2007 -0800
108861
108862 AIGLX support for Darwin -- works well enough to run
108863 glxgears and glxinfo, but still needs more testing.
108864
108865commit 43bd35fcf85b51b7e5714844bade90184dbd746f
108866Author: Peter Hutterer <peter@cs.unisa.edu.au>
108867Date: Thu Mar 1 17:19:37 2007 +1030
108868
108869 dix: Remove InitSprite, work can be done by InitializeSprite.
108870 Free sprite struct if a spriteOwner is paired.
108871
108872 xfree86: Use PairDevices instead of passing booleans around when creating a
108873 sprite.
108874
108875 Xext: Switch back to using LookupPointer/KeyboardDevice instead of
108876 inputInfo.xyz.
108877
108878commit 90ca76ba28fcd8bed945e33cf9674784fa2eb533
108879Author: Jay Cotton <jay.cotton@sun.com>
108880Date: Wed Feb 28 17:40:58 2007 -0800
108881
108882 Add sun_apm.c for Suspend-and-Resume support on Solaris
108883
108884 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6205248>
108885
108886commit 06c3021aec720837bef432656e88ae9b4e35101d
108887Author: Aaron Plattner <aplattner@nvidia.com>
108888Date: Wed Feb 28 16:09:11 2007 -0800
108889
108890 Don't crash setting a NULL mode with a randr classic DDX. Also remember to update the screen size during modesets.
108891
108892commit 68c64ad7b1eea79c786b5a7f3459076780163a47
108893Author: Peter Hutterer <peter@cs.unisa.edu.au>
108894Date: Thu Mar 1 09:51:20 2007 +1030
108895
108896 Xext: Update device's lastx/lasty when sending a motion event with XTest.
108897
108898commit 8b245758845523d5f8f017bb9d0e9aa57b616c28
108899Author: Aaron Plattner <aplattner@nvidia.com>
108900Date: Mon Feb 26 17:45:40 2007 -0800
108901
108902 Return BadMatch if a client tries to clone non-cloneable outputs.
108903
108904commit 2a35d44b6d87f96a7ce90dc8f0142f48cc176e95
108905Author: Peter Hutterer <peter@cs.unisa.edu.au>
108906Date: Wed Feb 28 15:43:06 2007 +1030
108907
108908 Update device valuators on XTest motion event, otherwise the cursor jumps
108909 between coordinates stored in device and coordinates sent by the event.
108910
108911commit d9bcb22d199e8444b9762a35754e04d327dd5915
108912Merge: 272d934 c16343a
108913Author: Ben Byer <bbyer@bbyer.apple.com>
108914Date: Tue Feb 27 16:28:20 2007 -0800
108915
108916 Merge branch 'master' of git+ssh://bbyer@git.freedesktop.org/git/xorg/xserver
108917
108918commit 272d9341d0f7c3e9e0c9b9a8c0d4d8779cdcc5cf
108919Author: Ben Byer <bbyer@bbyer.apple.com>
108920Date: Tue Feb 27 16:27:26 2007 -0800
108921
108922 fix for hw/darwin/Makefile.am to properly use XSERVER_LIBS
108923
108924commit c16343ac2ca18391b21022b2edd02ad9f413d2b3
108925Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108926Date: Tue Feb 27 14:14:47 2007 -0500
108927
108928 Make mfb, cfb, and afb support configurable at build-time.
108929
108930commit 5680efc0d2baf0a9451e82e490e3690fc23dda0f
108931Author: Alan Coopersmith <alan.coopersmith@sun.com>
108932Date: Tue Feb 27 09:55:48 2007 -0800
108933
108934 Sun bug 6529003: Xorg should not be including <sys/immu.h> on Solaris
108935
108936 <sys/immu.h> was removed from the latest Solaris Nevada build, but it's
108937 been useless to Xorg for a long time (it only declared a couple of kernel
108938 variables)
108939 <http://bugs.opensolaris.org/view_bug.do?bug_id=6529003>
108940
108941commit ab0fc8c1ad7ea2dc3389a4a4bb1c45bbded5e7ad
108942Author: Ben Byer <bbyer@bbyer.(none)>
108943Date: Tue Feb 27 00:14:35 2007 -0800
108944
108945 verbiage corrected per daniels
108946
108947commit cdd4c84572cc3bdd004f8dca6d8b64e710344ac0
108948Author: Ben Byer <bbyer@bbyer.(none)>
108949Date: Mon Feb 26 23:57:02 2007 -0800
108950
108951 added hw/darwin/README.apple file with some todo items and props.
108952
108953commit 776d4d6587c57f94bca8732f915d07a0d4e137c8
108954Author: Ben Byer <bbyer@bbyer.(none)>
108955Date: Mon Feb 26 23:40:00 2007 -0800
108956
108957 X11.app now builds correctly
108958
108959commit 154d2c13f4ec22b7e6332808bbcd049d63784891
108960Author: Ben Byer <bbyer@bbyer.apple.com>
108961Date: Mon Feb 26 19:39:26 2007 -0800
108962
108963 more changes for X11.app
108964
108965commit fa06e11f972e2a75c84b2f1586997ffc1239cbd9
108966Author: Ben Byer <bbyer@bbyer.apple.com>
108967Date: Mon Feb 26 17:06:53 2007 -0800
108968
108969 added hw/darwin/apple directory, which contains source and data files to build
108970 a version of the X11.app shipped with Mac OS X, using the X.org server.
108971
108972commit a16360733ea393ec1fc267e88fc604d9d7534484
108973Author: Jay Estabrook <Jay.Estabrook@hp.com>
108974Date: Sun Feb 25 19:58:26 2007 +0000
108975
108976 Fix root bus/domain selection on alpha
108977
108978commit 566610680c4e1cab3e7fc7146adbeaba52fdd0ad
108979Author: Adam Jackson <ajax@benzedrine.nwnk.net>
108980Date: Fri Feb 23 15:20:35 2007 -0500
108981
108982 Don't install libi2c.a
108983
108984commit 81372f9096b952f4be545654b0b44ac37ef4f2c2
108985Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108986Date: Fri Feb 23 13:23:12 2007 -0500
108987
108988 devPrivates rework: hook up new interface in resource system; add new
108989 resource-adding function that takes an additional ScreenPtr argument.
108990
108991commit 16f2b8892d9ebcef6410a675d10549043223f617
108992Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108993Date: Fri Feb 23 13:20:43 2007 -0500
108994
108995 devPrivates rework: add new interface implementation.
108996
108997commit 779faccfb78648a9f7e70b77dcfa9f6e19559772
108998Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
108999Date: Fri Feb 23 13:19:53 2007 -0500
109000
109001 devPrivates rework: add dix header file containing new interface.
109002
109003commit bc4ae25433e38a25a2012f9f233d500665172a4b
109004Author: Peter Hutterer <peter@cs.unisa.edu.au>
109005Date: Fri Feb 23 12:06:51 2007 +1030
109006
109007 xfree86: Don't initialize sprites for devices that don't own a sprite (stops
109008 keyboard hotplug segfaults)
109009
109010commit af550ea91c451cf4f831c2413266a19d1f211d0e
109011Author: Alan Coopersmith <alan.coopersmith@sun.com>
109012Date: Thu Feb 22 14:38:40 2007 -0800
109013
109014 Move SecurityPolicy file format from Xserver(1) to it's own man page
109015
109016 Don't make users looking for Xserver information page through pages of
109017 details only interesting to the handful of people writing security policies.
109018
109019commit b1142cdbce76fed8cb22ba6d7ac027751dd56a76
109020Author: Brice Goglin <brice.goglin@ens-lyon.org>
109021Date: Thu Feb 22 12:26:04 2007 -0800
109022
109023 Bug #10034: 'man Xserver' typos: s/dqoute/dquote/
109024
109025 Bugzilla #10034: <http://bugs.freedesktop.org/show_bug.cgi?id=10034>
109026 Patch #8780: <http://bugs.freedesktop.org/attachment.cgi?id=8780>
109027
109028commit 4b8b0e377a27ec904b2028c89aed11c6416af26c
109029Author: Peter Hutterer <peter@cs.unisa.edu.au>
109030Date: Thu Feb 22 20:00:59 2007 +1030
109031
109032 Xi: Add access control request handling.
109033
109034 dix: New file access.c to handle all access control for devices.
109035
109036commit 9a3eb0357e779d5d5f76858f23667956c4c5d721
109037Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109038Date: Fri Feb 16 19:30:03 2007 -0500
109039
109040 devPrivates rework: add RC_PRIVATES class, make ResourceRec visible in
109041 the API, and add extra fields and structure supporting private storage.
109042
109043commit cd0af7a7856e8246e27acc5513d219a094211625
109044Author: Peter Hutterer <peter@cs.unisa.edu.au>
109045Date: Wed Feb 21 10:08:41 2007 +1030
109046
109047 Xi, dix: Only one client at a time can change the pointer-keyboard pairing,
109048 using xRegisterPairingClient request.
109049
109050commit 3344a4eda704edc7dc30037f095de277a60a70bb
109051Author: Michel Dänzer <michel@tungstengraphics.com>
109052Date: Thu Feb 15 16:27:50 2007 +0100
109053
109054 DRI: Make sure number of DRI windows is accurate in driver ClipNotify hook.
109055
109056 Always call DRI{De,In}creaseNumberVisible (which in turn calls
109057 DRIDriverClipNotify) after updating pDRIPriv->nrWindows.
109058
109059commit 3c7a27dc77595ad018bb7c4f7cef6bc178268cb6
109060Author: Michel Dänzer <michel@tungstengraphics.com>
109061Date: Wed Feb 14 16:17:18 2007 +0100
109062
109063 DRI: New ClipNotify driver hook.
109064
109065 The hook is called whenever the clipList of any DRI window changes, be it via
109066 DRIClipNotify, DRICreateDrawable or DRIDrawablePrivDelete. This allows the
109067 driver to keep track of which DRI windows are visible where.
109068
109069commit eedf148e5a1273ebbf4dc8dcac9c435712fc00ea
109070Author: Michel Dänzer <michel@tungstengraphics.com>
109071Date: Fri Feb 2 18:27:40 2007 +0100
109072
109073 Track number of visible DRI windows separately for transitions.
109074
109075 This allows e.g. doing page flipping with multiple DRI windows as long as
109076 only one of them is visible.
109077
109078commit 8a42af6a935b1cf0e15102e986bb527f4fab31a8
109079Author: Keith Packard <keithp@neko.keithp.com>
109080Date: Mon Feb 19 15:28:37 2007 -0800
109081
109082 Check for clientGone before sending events from XFixes (bug #1753).
109083
109084 Freeing resources during client closedown can cause cursor changes which
109085 attempt to send cursor events through the XFixes extension; a client in the
109086 process of closing down has no file to send events to, causing a crash when
109087 this path is hit.
109088
109089commit 4660eaaffb36f526f71d5847ae1309c10ee133c6
109090Author: Ben Byer <bbyer@bbyer.(none)>
109091Date: Sun Feb 18 14:09:51 2007 -0800
109092
109093 configure fixes for darwin
109094
109095commit 5631a67f648f5f151a849a918ee12871c71c32e9
109096Author: Keith Packard <keithp@neko.keithp.com>
109097Date: Fri Feb 16 10:06:22 2007 -0800
109098
109099 Don't set subpixel order during startup; the screen won't be ready.
109100
109101 in xf86CrtcSetMode, scrn->pScreen will be NULL during server startup time,
109102 so don't try to set the subpixel order. subpixel order will be set in the
109103 randr initialization anyways.
109104 (cherry picked from commit 5f6f8616d862ce4a37f6d3df4bdbc44fd21cc82a)
109105
109106commit 096965ec9c7514f0c9fc0d17e5166f2d26781f87
109107Author: Keith Packard <keithp@guitar.keithp.com>
109108Date: Fri Feb 16 02:17:11 2007 -0800
109109
109110 Ensure drivers can use new modes header files.
109111
109112 New modes header files required a few minor changes to be used by external
109113 drivers, the most notable of which is the publication of the config file
109114 parser header files.
109115
109116commit 55797dd252382d35ebab5d9e18a5e0e77872d775
109117Author: Keith Packard <keithp@guitar.keithp.com>
109118Date: Fri Feb 16 00:56:00 2007 -0800
109119
109120 Respect rotation in initial screen size computation.
109121
109122commit e4507825bf0328ea59673f2bbe652de3a9105c86
109123Author: Keith Packard <keithp@guitar.keithp.com>
109124Date: Fri Feb 16 00:41:29 2007 -0800
109125
109126 Enable startup-time rotation; change rotation pixmap creation API.
109127
109128 Add monitor "Rotate" option taking one of "normal", "left", "inverted" or
109129 "right". However, because initial mode selection is made before the screen
109130 is completely initialized, we cannot create the shadow pixmap object at this
109131 point. Pend the shadow pixmap creation until the block handler.
109132
109133 Note that this code is not completely functional yet.
109134
109135commit 8606aeb9b2ab2dafc11e64436db4d3a7e67dbcc0
109136Author: Keith Packard <keithp@guitar.keithp.com>
109137Date: Thu Feb 15 22:23:16 2007 -0800
109138
109139 RRConfigureOutputProperty is a variable length request.
109140
109141 Replace REQUEST_SIZE_MATCH with REQUEST_AT_LEAST_SIZE
109142
109143commit a88844eccb0e423e71d4fcb286866a026308babd
109144Author: Daniel Stone <daniel@fooishbar.org>
109145Date: Sat Feb 17 20:35:07 2007 +0200
109146
109147 configure.ac: disable dmx per default
109148
109149 Disable DMX until it gets ported to the new input API.
109150
109151commit e9a2cc7d9fcc73e16576be2522522cce675dc3f3
109152Author: Daniel Stone <daniel@fooishbar.org>
109153Date: Sat Feb 17 16:17:48 2007 +0200
109154
109155 config: error message cleanup
109156
109157 Demote failure to connect from ErrorF to DebugF.
109158
109159commit 81876bc5ddc2f3eda5078fe4bd101917fb32e586
109160Author: Ben Byer <bbyer@apple.com>
109161Date: Sat Feb 17 04:07:11 2007 -0800
109162
109163 oops, missed a spot
109164
109165commit d287b76471f66c9aea54f969d050b35643cb2501
109166Author: Ben Byer <bbyer@apple.com>
109167Date: Sat Feb 17 03:47:42 2007 -0800
109168
109169 cleaned up some linking ugliness in hw/darwin/quartz
109170
109171commit 81444486be4f182dde778bac6f7edcbfc4368482
109172Author: Ben Byer <bbyer@apple.com>
109173Date: Sat Feb 17 02:23:11 2007 -0800
109174
109175 autoconf goodness for XDarwin, courtesy of pogma
109176
109177commit cf4994b0db2fef4c10ce8804adef766bc5118daf
109178Author: Ben Byer <bbyer@apple.com>
109179Date: Sat Feb 17 01:21:43 2007 -0800
109180
109181 dix mods for Darwin
109182
109183commit cece0601571f6304e392a3a40505664544b249f3
109184Author: Ben Byer <bbyer@apple.com>
109185Date: Sat Feb 17 01:00:13 2007 -0800
109186
109187 build fix for configure.ac / BUILD_DARWIN, oops
109188
109189commit 00b0657b815b95964401c3e36eed54063afbd003
109190Author: Ben Byer <bbyer@bbyer.(none)>
109191Date: Sat Feb 17 00:55:32 2007 -0800
109192
109193 glx fixes for XDarwin
109194
109195commit 93777c7b96e560da087963040e372aecbfca7bbc
109196Author: Ben Byer <bbyer@bbyer.(none)>
109197Date: Sat Feb 17 00:22:39 2007 -0800
109198
109199 more patches to make the Quartz part of XDarwin work again
109200 (thanks Peter and Torrey!)
109201
109202commit 612144c811fdf06b7c03cf48a321388fe411acd4
109203Author: Ben Byer <bbyer@bbyer.(none)>
109204Date: Sat Feb 17 00:09:58 2007 -0800
109205
109206 More build fixes / updates for XDarwin:
109207 quartz/cr: "Cocoa Rootless" support (deprecated in favor of xpr?)
109208 quartz/fullscreen: Fullscreen support using Xplugin (not yet functional)
109209
109210commit 68d39d8571d8717d26cedc84015d537549520a14
109211Author: Daniel Stone <daniel@fooishbar.org>
109212Date: Fri Feb 16 23:02:13 2007 +0200
109213
109214 kdrive/ephyr: fix keysym type confusion once and for all
109215
109216 Take keysyms in as an XID in hostx_load_keymap() and explicitly
109217 convert them to CARD32 for loading into the server. Fixes Xephyr on
109218 AMD64, wa-hey.
109219
109220commit 5507cb885d861e974be240120ada2ace2a980a72
109221Author: Daniel Stone <daniel@fooishbar.org>
109222Date: Fri Feb 16 23:01:27 2007 +0200
109223
109224 kdrive: delete input debugging, yet again ...
109225
109226 I have no idea how this keeps on coming back.
109227
109228commit 2534f5a9027c196f677923aaa38fa9ed9917f73d
109229Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109230Date: Fri Feb 16 15:33:48 2007 -0500
109231
109232 Remove nasty function pointer type from DevUnion, return to documented type.
109233
109234commit 84efe23ae834dd3a4d3f3e08832b69469c7382aa
109235Author: Ben Byer <bbyer@bbyer.(none)>
109236Date: Fri Feb 16 04:37:38 2007 -0800
109237
109238 updated darwin/quartz/xpr (libXplugin interface for Mac OS X) support
109239
109240commit 5e7f7436a755a33e48ab91831cc6af710a8344ef
109241Author: Ben Byer <bbyer@bbyer.(none)>
109242Date: Fri Feb 16 04:12:26 2007 -0800
109243
109244 merged in miext/rootless changes for XDarwin support
109245
109246commit f350909d1696fcfda87e8f12c729254d762313c9
109247Author: Keith Packard <keithp@guitar.keithp.com>
109248Date: Thu Feb 15 21:50:48 2007 -0800
109249
109250 Kludge around duplicate code added in hw/xfree86/modes.
109251
109252 Code added in hw/xfree86/modes came from the server-1.3-branch.
109253 Portions of this code had previously been integrated into xf86Mode.c
109254 and edid_modes.c.
109255
109256 To preserve hw/xfree86/modes as much as possible, the duplicate code from
109257 the other files has been disabled; a more careful review would figure out
109258 where that code actually belonged.
109259
109260commit 258beebc77510f84fbea66d6ebf29c5097bd11db
109261Author: Keith Packard <keithp@guitar.keithp.com>
109262Date: Thu Feb 15 20:13:15 2007 -0800
109263
109264 Report correct RandR 1.0 sizeID. Report correct subpixel order.
109265
109266 RandR 1.0 sizeID must be computed the same way every time, so when reporting
109267 it in the ScreenChangeNotify event, just construct the usual 1.0 data block
109268 and use that.
109269
109270 subpixel geometry information can be computed by looking at the connected
109271 outputs and finding any with subpixel geometry and using one of those for
109272 the global screen subpixel geometry. This might be improved by reporting
109273 None if more than one screen has information and they conflict.
109274
109275commit ef6b1235fd7d6dc422e8a150c089496a8e648067
109276Author: Keith Packard <keithp@guitar.keithp.com>
109277Date: Thu Feb 15 11:27:35 2007 -0800
109278
109279 Allow new modes code to build inside drivers as well as server.
109280
109281 Use config.h for driver builds where xorg-config.h isn't available.
109282
109283commit 3dbe8f6b6ea32a9a137ad6e9235f74009b095bd8
109284Author: Tilman Sauerbeck <tilman@code-monkey.de>
109285Date: Thu Feb 15 17:51:01 2007 +0100
109286
109287 Distribute hw/xfree86/modes.
109288
109289commit d4eb4d065032112a38444e36f791cb468a5ca8f4
109290Author: Keith Packard <keithp@guitar.keithp.com>
109291Date: Thu Feb 15 20:36:20 2007 -0800
109292
109293 Merge crtc/output-based mode selection code.
109294
109295 This code comes from the intel driver, so there's no history in this tree.
109296
109297 As the crtc/output-based mode selection code uses ddc, the ddc and i2c
109298 modules have been merged into the server. Attempts to load them are safely
109299 ignored now.
109300
109301commit 37fe4c49dc3a5faf2d3d56112b6bd78453045f6a
109302Author: Peter Hutterer <peter@cs.unisa.edu.au>
109303Date: Fri Feb 16 09:57:57 2007 +1030
109304
109305 mi: Move WarpPointer event generation to miPointerMove to avoid duplicate
109306 events, cache event array allocation.
109307
109308commit c2f3f705f1db8ca78292912544a7e416116175f3
109309Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109310Date: Thu Feb 15 14:38:24 2007 -0500
109311
109312 Bug #6988: Change behavior of Security extension per user feature request.
109313
109314commit 811675733e97416c990e6dc9c19271b43d96248d
109315Author: Daniel Stone <daniel@fooishbar.org>
109316Date: Thu Feb 15 19:09:00 2007 +0200
109317
109318 os: fix client privates leak
109319 Minor leak here. Oops.
109320
109321commit 8f6961d385bda92703f18090cff551409d2710c9
109322Author: Daniel Stone <daniel@fooishbar.org>
109323Date: Thu Feb 15 19:08:46 2007 +0200
109324
109325 configure.ac: add xdarwin stubs
109326 Add stub AM_CONDITIONALs to at least fix the build.
109327
109328commit a3b62623b8aac56b219633bdb2c2f6de19b0580b
109329Author: Daniel Stone <daniel@fooishbar.org>
109330Date: Thu Feb 15 17:07:42 2007 +0200
109331
109332 change versioning for new server version scheme
109333 See:
109334 http://xorg.freedesktop.org/wiki/XDC2007Notes#head-2719037a1905516c45cf74f0e155c8703221e446
109335
109336commit f452507ca9209598401d15da0039aaa4e1fed1a3
109337Merge: 0463283 8116757
109338Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109339Date: Thu Feb 15 09:47:20 2007 -0500
109340
109341 Merge branch 'master' into my-XACE-SELINUX
109342
109343commit 04632835b7402456fdcf6c8fa01970cd2804e27c
109344Merge: 88f89b9 9ecf79c
109345Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109346Date: Thu Feb 15 09:27:05 2007 -0500
109347
109348 Merge branch 'master' into my-XACE-SELINUX
109349
109350commit 0f6dd4aea6176507dbe1c90c950d332fecbcaacb
109351Author: Daniel Stone <daniel@fooishbar.org>
109352Date: Thu Feb 15 16:14:57 2007 +0200
109353
109354 kdrive/ephyr: free screen struct
109355 Free screen->driver on screenFini, instead of just leaking it.
109356
109357commit 9ecf79ca0111dd899ca88dd54156f71013220fcc
109358Author: Ben Byer <bbyer@apple.com>
109359Date: Thu Feb 15 05:22:21 2007 -0800
109360
109361 Beginnings of an update Darwin driver
109362
109363commit 136bb4874aadf4a731d7eb8671e8bb641f9980a7
109364Author: Ben Byer <bbyer@apple.com>
109365Date: Thu Feb 15 05:14:38 2007 -0800
109366
109367 iokit support for XDarwin
109368
109369commit 3ead1afe78d2913f08c8144cb2d3813c6b159488
109370Author: Ben Byer <bbyer@apple.com>
109371Date: Thu Feb 15 05:09:29 2007 -0800
109372
109373 Beginning of patches to add XDarwin support to the modular tree;
109374 special thanks to Torrey Lyons and Peter O'Gorman for making this possible.
109375
109376 This is the automake framework for the XDarwin.app interface files.
109377
109378commit f6c3b9fa97ccf85e96f15435d564a1c261e40532
109379Author: Peter Hutterer <peter@cs.unisa.edu.au>
109380Date: Thu Feb 15 13:51:20 2007 +1030
109381
109382 xfree86: mouse and keyboard hotplugging support. New mouse devices will get
109383 new cursors, keyboard devices use VCP's focus until paired.
109384
109385commit d570ff7c81858a3174686b46a088f67563b4a2d5
109386Author: Peter Hutterer <peter@cs.unisa.edu.au>
109387Date: Wed Feb 14 17:09:33 2007 +1030
109388
109389 fix: WarpCursor needs to send MotionNotify.
109390
109391commit c4fe1bcce1c1e4822e688959b331b47a051d6e0a
109392Merge: e540d57 d570ff7
109393Author: Ian Romanick <idr@us.ibm.com>
109394Date: Wed Feb 14 15:39:52 2007 -0800
109395
109396 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109397
109398commit e540d572c5acba877b3ce01e7b31e399dac2d44a
109399Author: Ian Romanick <idr@us.ibm.com>
109400Date: Wed Feb 14 15:38:10 2007 -0800
109401
109402 Merge fix missed on previous commit.
109403
109404commit 81aa7f059d3cfd8d28420b7932b8ff7e06d67979
109405Author: Eric Anholt <eric@anholt.net>
109406Date: Wed Feb 14 12:48:15 2007 -0800
109407
109408 Add missing dirty marking in a couple of fallback cases in the exaGlyphs path.
109409
109410commit a5f19c5150a7b3dc2ff3ad759ee1a6ab0ad8925c
109411Author: Eric Anholt <eric@anholt.net>
109412Date: Wed Feb 14 10:39:46 2007 -0800
109413
109414 Mark sync when UploadToScreen succeeds in exaGlyphs().
109415
109416commit a492d494f51caf15a5cb979dc335387486c105d1
109417Author: Alan Coopersmith <alan.coopersmith@sun.com>
109418Date: Tue Feb 13 18:32:59 2007 -0800
109419
109420 Update Xvfb man page: remove monolith build instructions, use /var/tmp instead of /usr/tmp
109421
109422commit 25104ed2e6050f86691104b34eeaf3147eb83c4a
109423Author: Peter Hutterer <peter@cs.unisa.edu.au>
109424Date: Wed Feb 14 11:26:53 2007 +1030
109425
109426 dix: fix typo in condition.
109427
109428commit 33ef546b942977bb2573156eaba5269350d16f99
109429Author: Peter Hutterer <peter@cs.unisa.edu.au>
109430Date: Wed Feb 14 11:18:43 2007 +1030
109431
109432 dix: Clean up GetPointerEvents. If pDev == inputInfo.pointer, assume that
109433 we do not want Xi events.
109434
109435 mi: Remove POINTER_CORE_ONLY flag.
109436
109437commit a23b0b069cac8a48e2b306b2095515d75f647705
109438Author: Adam Jackson <ajax@benzedrine.nwnk.net>
109439Date: Mon Feb 12 17:50:00 2007 -0500
109440
109441 Typo fix.
109442
109443commit d21c95f80bdba2f29eedd57fb0b00e580391c08e
109444Author: Adam Jackson <ajax@benzedrine.nwnk.net>
109445Date: Mon Feb 12 17:22:39 2007 -0500
109446
109447 Hook up --with-builderstring for vendor build identification.
109448
109449commit 46784d24c11767455a4986449a8037295912dcee
109450Author: Adam Jackson <ajax@benzedrine.nwnk.net>
109451Date: Mon Feb 12 17:18:29 2007 -0500
109452
109453 Remove spurious LIBADD from xf4bpp
109454
109455commit 16eb7254f861465f988ae3861ac3449c2c966062
109456Merge: 3814862 c4b7e9d
109457Author: Ian Romanick <idr@us.ibm.com>
109458Date: Fri Feb 9 12:38:49 2007 -0800
109459
109460 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109461
109462commit c4b7e9d1c16797c3e4b1200b40aceab5696a7fb8
109463Author: Aaron Plattner <aplattner@nvidia.com>
109464Date: Tue Feb 6 14:57:22 2007 -0800
109465
109466 Add an RDTSC implementation to the x86 emulator.
109467
109468 This instruction is being used in some debug VBIOSes. This implementation
109469 doesn't even try to be accurate. Instead, it just increments the counter by a
109470 fixed amount every time an rdtsc instruction in encountered, to avoid divides by
109471 zero.
109472
109473commit 262b9b104a04e55969593ef96a16004e53ecd00a
109474Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
109475Date: Tue Feb 6 17:30:22 2007 -0500
109476
109477 Use the new 8888x0565mmx function in fbpict.c
109478
109479commit 876b806ec09d5ff0c6cd19df91006c4eefedfaa6
109480Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
109481Date: Tue Feb 6 17:16:23 2007 -0500
109482
109483 Reapply patch to fix AMD CPU detection
109484
109485commit 13568d2aa43da4216bbcb46e1125ff28c323ac54
109486Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
109487Date: Tue Feb 6 17:12:01 2007 -0500
109488
109489 Revert "Fix for AMD cpu detection. Bug 9614, Dan Williams."
109490
109491 This reverts commit b2cd3b133748cc5aa541905a703a6fdb1cbbb1e6 since
109492 unrelated changes in fbpict.c broke the build.
109493
109494commit 5a3334410367a2186b2c667fa1eb6cf0baf93e95
109495Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
109496Date: Tue Feb 6 17:11:01 2007 -0500
109497
109498 Add new fbCompositeSrc_8888x0565mmx() function, based on patch by Dan
109499 Williams. Bug 9682.
109500
109501commit b2cd3b133748cc5aa541905a703a6fdb1cbbb1e6
109502Author: Soren Sandmann Pedersen <ssp@dhcp83-218.boston.redhat.com>
109503Date: Tue Feb 6 16:43:37 2007 -0500
109504
109505 Fix for AMD cpu detection. Bug 9614, Dan Williams.
109506
109507 Credit for the fixes in this patch goes to:
109508
109509 Marco Gritti <mpg at redhat dot com>
109510 Jordan Crouse <jordan dot crouse at amd dot com>
109511
109512commit 7cd73b00a29469d37a7f526cc9b35cc927a13ae2
109513Author: Peter Hutterer <peter@cs.unisa.edu.au>
109514Date: Mon Feb 5 14:48:48 2007 +1030
109515
109516 dix: If POINTER_CORE_ONLY is set, GetPointerEvents() only creates a core event
109517
109518 mi: fix: Create motion event in miPointerMoved
109519
109520commit 760a38c4c7ab66ae653d3acb92f5cda4bd44edd6
109521Author: Daniel Stone <daniel@fooishbar.org>
109522Date: Mon Feb 5 03:39:36 2007 +0200
109523
109524 XkbCopyKeymap: fix copy-and-waste accident
109525
109526 When we reallocated modmap, we accidentally clobbered syms with the
109527 result, leaving syms definitely too small, and modmap also potentially too
109528 small (as well as not actually allocated anymore).
109529
109530commit 17d85387d1e6851d35474b65929e268ca64ef65b
109531Author: Daniel Stone <daniel@fooishbar.org>
109532Date: Thu Jan 18 15:23:57 2007 +1100
109533
109534 dmx, vfb, xnest: fix fbcmap compilation
109535
109536 Don't always define XFree86Server, but only for damn fbcmap.c.
109537 Split fbcmap.c into its own library to achieve this.
109538
109539commit a309c936bb79e2fea04a96ce33c7da99ed902484
109540Author: Peter Hutterer <peter@cs.unisa.edu.au>
109541Date: Mon Feb 5 11:38:44 2007 +1030
109542
109543 xkb: Daniel's patch to stop evdev keyboard segfaults.
109544
109545 mi: Remove quickfix from ba547d4f6a2707f51f0d826389d2d418fb62367a
109546
109547commit 236f04b638e7d4d1656c6bedd8a6e8d7cec285ec
109548Author: Dave Airlie <airlied@linux.ie>
109549Date: Mon Feb 5 09:09:12 2007 +1100
109550
109551 remove array_cache from everywhere
109552
109553commit eb228e8d1eaa78911541b2fec5d04a74c1299718
109554Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
109555Date: Sun Feb 4 22:06:59 2007 +0000
109556
109557 clean up more of the vbo fallout
109558
109559commit fb1bc1c65b88527b42a0e4abed23e5ddaae711b7
109560Author: Dave Airlie <airlied@linux.ie>
109561Date: Sun Feb 4 18:39:58 2007 +1100
109562
109563 add vbo to .gitignore
109564
109565commit d8e148ec841d340327e6813127b0e0ffc4db712d
109566Author: Dave Airlie <airlied@linux.ie>
109567Date: Sun Feb 4 18:39:04 2007 +1100
109568
109569 update xserver for vbo code in mesa
109570
109571commit 5dcad9e9d7d9993d65f989219bee94a060bbf476
109572Author: Alan Coopersmith <alan.coopersmith@sun.com>
109573Date: Fri Feb 2 14:44:55 2007 -0800
109574
109575 Fix bus error on startup in 64-bit Xephyr
109576
109577 hostx_get_visual_masks takes unsigned long * arguments, but was being
109578 passed pointers to CARD32's.
109579
109580commit 170a55022ebc7b148bff93886eda152a0d5ce79a
109581Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
109582Date: Fri Feb 2 20:56:12 2007 +0000
109583
109584 remove file
109585
109586commit e6a505be84f5f72349d6860dc5a5058367516019
109587Author: Dan Nicholson <dan@conor.dwcab.com>
109588Date: Fri Feb 2 20:53:01 2007 +0000
109589
109590 The array_cache sources don't exist anymore in the Mesa tree,
109591 so we shouldn't try to build them.
109592
109593commit af20485ec370801f2aabfaeae17bbd030a849bd1
109594Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
109595Date: Fri Feb 2 19:14:46 2007 +0000
109596
109597 Remove array_cache for recent Mesa changes
109598
109599commit cf5b29d75dad7c74543f49f010c817623a3df747
109600Author: George Sapountzis <gsap7@yahoo.gr>
109601Date: Fri Feb 2 12:57:38 2007 +0200
109602
109603 dmx: drop leftover __GLXdrawablePrivateRec struct.
109604
109605commit 4f2f3233c808fd86bf9f6c09937feda9e0b367fd
109606Author: Eric Anholt <eric@anholt.net>
109607Date: Thu Feb 1 15:10:29 2007 -0800
109608
109609 Fix the size expectations of xRRSetCrtcGamma.
109610
109611 It was using REQUEST_SIZE_MATCH (client request length must equal request size)
109612 rather than REQUEST_AT_LEAST_SIZE (client request length must be at least
109613 big enough for request size), and this request has data following the request
109614 structure.
109615
109616commit 8274ea6aa97b06a56b7468c3908894c0ff72b687
109617Author: Eric Anholt <eric@anholt.net>
109618Date: Thu Feb 1 12:15:54 2007 -0800
109619
109620 Set the Damage version supported in the server, instead of using damageproto.
109621
109622 This was caught by distributions upgrading damageproto to 1.1, before the
109623 server they had supported it. The server then advertised the new version
109624 without supporting the protocol.
109625
109626commit 8bce182568f14edfb03911d8c5d791fd83bb6222
109627Author: Eric Anholt <eric@anholt.net>
109628Date: Mon Jan 29 17:30:59 2007 -0800
109629
109630 Restore a few important lines killed in the previous commit.
109631
109632 Typical results were failure to sync, and a black screen.
109633
109634commit 3814862a869ee83d307eb01225d5949039f435d8
109635Merge: a216de9 31f2d4a
109636Author: Ian Romanick <idr@us.ibm.com>
109637Date: Mon Jan 29 15:14:31 2007 -0800
109638
109639 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109640
109641 Conflicts:
109642
109643 hw/xfree86/os-support/bus/linuxPci.c
109644
109645commit 31f2d4a57e04f5ea635fbb50c508405c4fc37b65
109646Author: Eric Anholt <eric@anholt.net>
109647Date: Mon Jan 29 09:39:33 2007 -0800
109648
109649 Bug #9680: Remove bogus blank length limiting in xf86SetModeCrtc().
109650
109651 Our modes typically come from EDID or default modes, and when the monitor
109652 asks for a specific mode, deciding to tweak it usually results in incorrect
109653 display. And if the user is specifying a mode by hand, tweaking it then is
109654 still pretty rude.
109655
109656 Reviewed by: ajax
109657
109658commit 4aaaa70d1b52346213fad46777c006a93c4ece5d
109659Author: Peter Hutterer <peter@cs.unisa.edu.au>
109660Date: Mon Jan 29 18:23:24 2007 +1030
109661
109662 Xi: Adding ChangePointerKeyboardPairing request
109663 Adding PointerKeyboardPairingChanged event
109664 Correct error values for XWarpDevicePointer
109665
109666 dix: Adding device argument to SendMappingNotify
109667 Adding spriteOwner flag to devices
109668
109669commit f3418b52dcf2ab4982504856ab9fae3e726ee6d2
109670Author: Peter Hutterer <peter@cs.unisa.edu.au>
109671Date: Mon Jan 29 16:10:03 2007 +1030
109672
109673 mi: Fix cursor rendering issues.
109674
109675commit 15a81b6325d359990017b8e9f17ce18a7eff1354
109676Merge: 3759254 cc3d66d
109677Author: Peter Hutterer <peter@cs.unisa.edu.au>
109678Date: Sun Jan 28 17:18:57 2007 +1030
109679
109680 Merge branch 'master'
109681
109682commit cc3d66d4a53fee8f1d940bb8c7f032c94b9c25f6
109683Author: Peter Hutterer <peter@cs.unisa.edu.au>
109684Date: Wed Nov 15 17:50:02 2006 +1030
109685
109686 o fix minor error in comment for GetPointerEvents()
109687
109688commit 3759254c0967ae83bdcbf097d9e8a58733c8e6f9
109689Merge: ba547d4 ca5ebe3
109690Author: Peter Hutterer <peter@cs.unisa.edu.au>
109691Date: Sun Jan 28 17:05:55 2007 +1030
109692
109693 Merge branch 'master'
109694
109695commit 1627af54497bee659ea30f2850b39cbbf576e22d
109696Author: Jonathan Lim <jlim@sgi.com>
109697Date: Fri Jan 26 13:00:45 2007 +0100
109698
109699 Call linuxPciOpenFile() for r/w access if applicable.
109700
109701 Currently, the call to linuxPciOpenFile() is always made for read
109702 only access which causes the subsequent mmap call to fail when the
109703 memory is mapped read/write.
109704
109705 Xorg #9692
109706
109707commit a216de9b7ff55e2b73c487d037f248f00bd2e63b
109708Merge: 24506ea cf7ca9d
109709Author: Ian Romanick <idr@us.ibm.com>
109710Date: Thu Jan 25 10:17:32 2007 -0800
109711
109712 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109713
109714commit cf7ca9d09cba14d107152a5179de38e5ef7bd784
109715Author: Alan Coopersmith <alan.coopersmith@sun.com>
109716Date: Wed Jan 24 20:20:48 2007 -0800
109717
109718 Plug memory leak in doLoadModule()
109719
109720commit 5abd50e37ceda134897891ed32e05215db67e0b4
109721Author: Alan Coopersmith <alan.coopersmith@sun.com>
109722Date: Wed Jan 24 18:54:38 2007 -0800
109723
109724 Correct help lines for configure's --with-vendor-name flags
109725
109726commit b32a40817fc0e2ac2edf2fa22a8813087fce2e7b
109727Author: Alan Coopersmith <alan.coopersmith@sun.com>
109728Date: Wed Jan 24 16:29:49 2007 -0800
109729
109730 Correct variable descriptions in comment for SecurityCheckResourceIDAccess
109731
109732commit 24506ea65be4cb29c5e1486aa0a529a40ce5c230
109733Author: Ian Romanick <idr@us.ibm.com>
109734Date: Wed Jan 24 15:33:49 2007 -0800
109735
109736 Move xf86ReadLegacyBIOS to the one place that uses it.
109737
109738 xf86ReadLegacyBIOS is only used by one function in int10/generic.c.
109739 Move a generic implementation of that function there, rename it to
109740 read_legcay_BIOS, and delete all remnants of it from all other places.
109741
109742commit fdb3a0798f0d17e72ec7293d59a7a81b5ffdf95b
109743Merge: 39b51e1 a53586e
109744Author: Ian Romanick <idr@us.ibm.com>
109745Date: Wed Jan 24 14:49:39 2007 -0800
109746
109747 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109748
109749commit a53586eebc166e35c1f48942205832810061daee
109750Author: Eric Anholt <eric@anholt.net>
109751Date: Wed Jan 24 13:36:25 2007 -0800
109752
109753 Warning fix for RRCrtcSetRotations().
109754
109755commit 7a12952fd437b105ea0d013d680f9c3a775a183c
109756Author: Eric Anholt <eric@anholt.net>
109757Date: Wed Jan 24 13:34:29 2007 -0800
109758
109759 Bug #7639: Only swap out pixmaps (rather than everything) on VT switch in EXA.
109760
109761 This is a new behavior for version 2.1 of EXA, and only takes effect if the
109762 driver has requested that. Otherwise, the previous behavior remains the same.
109763
109764commit b6b855932109b4bc3454f07bef8cb079d79ca369
109765Author: Keith Packard <keithp@keithp.com>
109766Date: Thu Jan 25 00:29:20 2007 +0800
109767
109768 Make Xinearama screen information reflect CRTC rotation.
109769
109770commit 39b51e1bfc4924fc3bda4342ef9c6c0125d9f4fa
109771Merge: 637b19b 788cfce
109772Author: Ian Romanick <idr@us.ibm.com>
109773Date: Tue Jan 23 13:25:56 2007 -0800
109774
109775 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109776
109777commit 788cfce911793a26aed16f38f30678ecee82c873
109778Author: Michel Dänzer <michel@tungstengraphics.com>
109779Date: Tue Jan 23 10:15:22 2007 +0100
109780
109781 Bump video driver ABI version to 1.2.
109782
109783 This is necessary because server-1.2-branch bumped to 1.1 for xf86CVTMode and
109784 we have xf86XVFillKeyHelperDrawable on top of that.
109785
109786commit 637b19b3ee7345c2d449ce94ceabe998d817c02e
109787Author: Ian Romanick <idr@us.ibm.com>
109788Date: Mon Jan 22 09:58:33 2007 -0800
109789
109790 Eliminate pciControlBridge, pciGetBridgeBuses, and pciGetBridgeResources.
109791
109792commit a232e4ae938fa3a68f7614995676eda7929260db
109793Author: Ian Romanick <idr@us.ibm.com>
109794Date: Mon Jan 22 09:16:45 2007 -0800
109795
109796 Convert xf86GetPciHostConfigFromTag to get_parent_bridge.
109797
109798 Convert xf86GetPciHostConfigFromTag to a new function called
109799 get_parent_bridge. This name better represents what
109800 xf86GetPciHostConfigFromTag is used for: walking up the lists of PCI
109801 bridges from a device.
109802
109803commit 844560a02fa6c4ce18ea2af3ec27beaa60b7af11
109804Author: Ian Romanick <idr@us.ibm.com>
109805Date: Mon Jan 22 09:13:59 2007 -0800
109806
109807 Elminiate ARCH_PCI_OS_INIT.
109808
109809 Never, ever use the ix86Pci stuff on Linux. This renders the whole
109810 ARCH_PCI_OS_INIT mechanism useless. Remove it.
109811
109812commit 7dfb3cea913b02a6b36c308d1330bd40abe6b41f
109813Merge: c279d5f 2dc8662
109814Author: Ian Romanick <idr@us.ibm.com>
109815Date: Mon Jan 22 08:44:36 2007 -0800
109816
109817 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109818
109819commit 2dc866252c84ed0e7b3afa25e8a5312f448d405b
109820Author: Eric Anholt <eric@anholt.net>
109821Date: Mon Jan 22 08:41:50 2007 +0800
109822
109823 Really fix optimized render cases being hit when they shouldn't.
109824
109825 I don't know how this define slipped in there. Fixes
109826 6fdfd9dad91d7b7aa292f8c4d268dd27c34de8d3.
109827
109828commit 88f89b9ac1b92a0916c46488350ff68c3ffdd490
109829Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109830Date: Fri Jan 19 19:15:49 2007 -0500
109831
109832 Policy updates: use x prefix in property and ext types.
109833
109834commit 2fb8b7f8199c35ae0870cb54b40ee28a4e01d479
109835Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109836Date: Fri Jan 19 19:14:51 2007 -0500
109837
109838 Split ObjectSIDByLabel into two functions since property labeling now
109839 involves an additional compute_create lookup.
109840
109841commit 0d6d373af95d0004d33b987d14ad7e04dd5d2003
109842Author: Alan Coopersmith <alan.coopersmith@sun.com>
109843Date: Fri Jan 19 14:52:23 2007 -0800
109844
109845 Update Xserver man page to match commit ed33c7c98ad0c542e9e2dd6caa3f84879c21dd61
109846
109847 Remove unused -xkbdb and -noloadxkb options. Rename -ar1 and -ar2 to
109848 -ardelay and -arinterval, respectively.
109849
109850commit c279d5fdc53612a90f33fafe9e9c59ced008dd51
109851Merge: 88f248e 14d6a9b
109852Author: Ian Romanick <idr@us.ibm.com>
109853Date: Fri Jan 19 12:59:54 2007 -0800
109854
109855 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
109856
109857commit 700fccf863593cbea1691789f1f1cafc08a32fee
109858Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109859Date: Fri Jan 19 14:56:38 2007 -0500
109860
109861 Remove the root window context line from the configuration file.
109862 This context will be derived through a type_transition rule instead.
109863
109864commit 3a9791b456f35adb252a9059b19265c6c447f1ba
109865Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
109866Date: Fri Jan 19 14:53:09 2007 -0500
109867
109868 Policy updates.
109869
109870commit 14d6a9b327381a6bb2dac59c62728e5fd0f0bcfb
109871Author: Michel Dänzer <michel@tungstengraphics.com>
109872Date: Fri Jan 19 18:30:21 2007 +0100
109873
109874 fbdevhw: Only deal with RGB weight if default visual is True- or DirectColor.
109875
109876commit 27a01e100bff21ac0b70c6d72071d7226fc91264
109877Author: Michel Dänzer <michel@tungstengraphics.com>
109878Date: Fri Jan 19 18:28:05 2007 +0100
109879
109880 fbdevhw: Consider mode set equal to mode requested if virtual width is larger.
109881
109882commit 65f4690ecb4576f60396fcccff8e5bd5d4b6645f
109883Author: Michel Dänzer <michel@tungstengraphics.com>
109884Date: Fri Jan 19 17:54:03 2007 +0100
109885
109886 __glXDRIscreenProbe: Use drmOpen/CloseOnce.
109887
109888 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=9275 . Based on patch from
109889 Alan Swanson.
109890
109891commit 8b3a591cd39f2d51209dc71b641cac79663e1b16
109892Author: Alan Coopersmith <alan.coopersmith@sun.com>
109893Date: Thu Jan 18 16:03:30 2007 -0800
109894
109895 Update pci.ids to 2007-01-18 snapshot
109896
109897 (includes a whole bunch of ATI device id updates)
109898
109899commit 0f0c321adf2850b3d7aafe281362bfe424cb0ca1
109900Author: Alan Coopersmith <alan.coopersmith@sun.com>
109901Date: Thu Jan 18 15:31:53 2007 -0800
109902
109903 Make xf1bpp build correctly with compilers that don't support -include
109904
109905commit a811e92104028ae60ba69f73e32ee1e0533b088c
109906Author: Eric Anholt <eric@anholt.net>
109907Date: Thu Jan 18 14:28:01 2007 -0800
109908
109909 Account for CRTC rotation in the cursor containment code.
109910
109911commit df147c10ce597c56c16cbca552e8a3e3ecb3cdaa
109912Author: Alan Coopersmith <alan.coopersmith@sun.com>
109913Date: Wed Jan 17 16:47:07 2007 -0800
109914
109915 Xserver man page: remove bc, add -wr
109916
109917commit 2dfd1aab244a2c8da3b62b522b9a8434e474af17
109918Author: Alan Coopersmith <alan.coopersmith@sun.com>
109919Date: Wed Jan 17 14:39:28 2007 -0800
109920
109921 Always include compiler.h in cfbmskbits.h instead of checking #ifdef XFREE86
109922
109923commit 42a48786acf54f83167de4f561526986d4e27033
109924Author: Eric Anholt <eric@anholt.net>
109925Date: Wed Jan 17 14:34:42 2007 -0800
109926
109927 Add a setter for randr_crtc->rotations.
109928
109929commit 88f248e67143175532cbafd6faf8fc6df97c7894
109930Author: Ian Romanick <idr@us.ibm.com>
109931Date: Wed Jan 17 14:04:43 2007 -0800
109932
109933 Replace PciBusId with 'struct pci_device *'
109934
109935 There's no need to store the slot information for a PCI device as its
109936 ID. Instead, skip the middle man and just store a pointer to the
109937 pci_device structure.
109938
109939commit 0361611080267727f570e17f2212aaa890223f6e
109940Author: Ian Romanick <idr@us.ibm.com>
109941Date: Wed Jan 17 13:06:04 2007 -0800
109942
109943 Convert xf86IsolateDevice to a 'struct pci_slot_match'.
109944
109945 This change was accidentally not committed on the previous commit.
109946
109947commit c1808f148405a28256e0480d6f8714691b0e964e
109948Author: Ian Romanick <idr@us.ibm.com>
109949Date: Wed Jan 17 13:02:27 2007 -0800
109950
109951 Convert xf86IsolateDevice to a 'struct pci_slot_match'.
109952
109953commit d2f8c42c48ecc398d224d9e3b280a66042e68664
109954Author: Ian Romanick <idr@us.ibm.com>
109955Date: Wed Jan 17 12:59:17 2007 -0800
109956
109957 Convert xf86FindPrimaryDevice to use a static buffer and snprintf.
109958
109959 Rather than allocate a 9 byte buffer on each invocation, use a static
109960 16 byte buffer. Use snprintf for safety. This commit should probably
109961 be cherry-picked to the trunk.
109962
109963commit 28976bebec8b6475d0703f8356edd116e25d40b2
109964Author: Ian Romanick <idr@us.ibm.com>
109965Date: Wed Jan 17 12:15:09 2007 -0800
109966
109967 Eliminate unused fields in pciAccRec.
109968
109969commit f36a447d20c0a89b78c5899eeb303cab568ebac6
109970Author: Ian Romanick <idr@us.ibm.com>
109971Date: Tue Jan 16 14:49:19 2007 -0800
109972
109973 Eliminate use of Tag field.
109974
109975 The Tag field was removed from the int10 structures in commit
109976 ca9c41e09ddb72d150d86f5d372eb07eeca7c1a2. This file was over looked.
109977
109978commit cde17015dff1ced2aabb8b76c08f9110237821a5
109979Author: Eric Anholt <eric@anholt.net>
109980Date: Tue Jan 16 13:01:03 2007 -0800
109981
109982 When changing a non-pending property, call the screen rrOutputSetProperty hook.
109983
109984commit e3add7c8ecbb2a0a662860f208f6ae7d1857c717
109985Author: Eric Anholt <eric@anholt.net>
109986Date: Tue Jan 16 12:59:34 2007 -0800
109987
109988 Don't forget to add the property we configure to the properties list.
109989
109990commit 780b55ec6f97864b50f00170b201e93eba188f1b
109991Author: Ian Romanick <idr@us.ibm.com>
109992Date: Mon Jan 15 15:30:16 2007 -0800
109993
109994 Fix accidental commit (bug) in linuxOpenLegacy.
109995
109996commit 31bd8d150749c548eb70404395b4a52d46412f43
109997Author: Ian Romanick <idr@us.ibm.com>
109998Date: Mon Jan 15 15:19:30 2007 -0800
109999
110000 Fix domain insanity.
110001
110002 Eliminate xf86GetPciDomain. The domain from libpciaccess is the
110003 domain. Period. This means that 0 is a valid domain. Make sure that
110004 INCLUDE_XF86_NO_DOMAIN is *not* set. Always run in "domain mode,"
110005 even if the only domain possible is 0.
110006
110007commit 079a6a55a516778871a8b2f8e9c8dd9162105a76
110008Author: Ian Romanick <idr@us.ibm.com>
110009Date: Mon Jan 15 11:39:24 2007 -0800
110010
110011 Convert to new pci_device_cfg_write_u* interface.
110012
110013commit d8bd0c41de11d12976a5ea9f185b7ac6c7060b71
110014Author: Ian Romanick <idr@us.ibm.com>
110015Date: Mon Jan 15 11:38:22 2007 -0800
110016
110017 Convert to new pci_device_cfg_write_u* interface.
110018
110019commit a87af8218f97da76433a5192d5e51aca0d5c3ed0
110020Author: Ian Romanick <idr@us.ibm.com>
110021Date: Mon Jan 15 11:21:37 2007 -0800
110022
110023 Remove unnecessary files from the sources list.
110024
110025commit 1c326bf75fc0d2c39507834d159d031291eddac4
110026Author: Ian Romanick <idr@us.ibm.com>
110027Date: Fri Jan 12 12:07:19 2007 -0800
110028
110029 Eliminate unused INCLUDE_XF86_MAP_PCI_MEM define.
110030
110031commit 3936caa00535d0b7543be5a4b315562111a6b3ab
110032Author: Ian Romanick <idr@us.ibm.com>
110033Date: Fri Jan 12 10:59:35 2007 -0800
110034
110035 Refactor pciTagConvertRange2Host and pciConvertRange2Host.
110036
110037commit 1a493509e30412a0f6c6f8f8e47a446678dbbc1a
110038Author: Ian Romanick <idr@us.ibm.com>
110039Date: Fri Jan 12 10:58:59 2007 -0800
110040
110041 Fix typos in conditionally compiled code.
110042
110043commit 8279444a54f38c5e2e5d4c2d936a10d74990e0be
110044Author: Ian Romanick <idr@us.ibm.com>
110045Date: Fri Jan 12 10:41:03 2007 -0800
110046
110047 Eliminate unused pciAddrHostToBus infrastructure.
110048
110049commit e32fa10a04a15ce1b15c6faa7a64e68ee3906d59
110050Author: Ian Romanick <idr@us.ibm.com>
110051Date: Fri Jan 12 10:33:17 2007 -0800
110052
110053 Remove prototype for pciGetBaseSize.
110054
110055commit 7411f5c389db73a6731d853d603029b8cf0a66ba
110056Author: Ian Romanick <idr@us.ibm.com>
110057Date: Fri Jan 12 10:32:04 2007 -0800
110058
110059 Eliminate pciHostAddrToBusAddr from interface.
110060
110061 pciHostAddrToBusAddr is no longer used anywhere in the tree. Remove
110062 it with extreme prejudice.
110063
110064commit 60c5e023e8ddaae8e011f80736e7cc3e8c12583d
110065Author: Ian Romanick <idr@us.ibm.com>
110066Date: Fri Jan 12 10:24:03 2007 -0800
110067
110068 Eliminate some dead code.
110069
110070commit dd12a94885bad9c9d064dcf7e0d9a7a7114ae3e6
110071Author: Ian Romanick <idr@us.ibm.com>
110072Date: Fri Jan 12 10:21:25 2007 -0800
110073
110074 Eliminate xf86GetPciEntity from public interface.
110075
110076 xf86GetPciEntity was folded into the only place (xf86CheckPciSlot) in
110077 the tree that still called it.
110078
110079commit c3016a29964761478305cf2f5241c563ff3eab0a
110080Author: Ian Romanick <idr@us.ibm.com>
110081Date: Fri Jan 12 09:56:00 2007 -0800
110082
110083 Missed file from previous commit.
110084
110085 Convert xf86ReadLegacyVideoBIOS to take a 'struct pci_device *'
110086 parameter instead of a PCITAG.
110087
110088commit ca9c41e09ddb72d150d86f5d372eb07eeca7c1a2
110089Author: Ian Romanick <idr@us.ibm.com>
110090Date: Thu Jan 11 21:09:20 2007 -0800
110091
110092 Convert int10 and vgaHW to use 'struct pci_device' instead of PCITAG.
110093
110094 Convert all uses of PCITAG in int10 and vgaHW to 'struct pci_device'.
110095 This allows the conversion of xf86ReadLegacyVideoBIOS and
110096 xf86MapDomainMemory to 'struct pci_device' from PCITAG.
110097
110098commit ba547d4f6a2707f51f0d826389d2d418fb62367a
110099Author: Peter Hutterer <peter@cs.unisa.edu.au>
110100Date: Fri Jan 12 14:42:02 2007 +1030
110101
110102 mi: Quick fix to stop segfault on non-core keyboards.
110103
110104commit 5e2f8dee6331645fcbd2734c43698eb4f9c5b116
110105Author: Peter Hutterer <peter@cs.unisa.edu.au>
110106Date: Thu Jan 11 14:05:39 2007 +1030
110107
110108 mi: remove core pointer special handling. No event has core pointer as
110109 device now.
110110
110111 dix: zero pSprite memory, stop segfaults on server restart.
110112
110113commit 7fccec91c46baac4f8d2965180dc535b4eb7d65c
110114Author: Eric Anholt <eric@anholt.net>
110115Date: Wed Jan 10 13:10:43 2007 -0800
110116
110117 Bug #9555: Always define _GNU_SOURCE in glibc environments.
110118
110119 This keeps us from having to define _POSIX_C_SOURCE, _BSD_SOURCE, and
110120 _XOPEN_SORUCE to get the C environment we want in different places. It also
110121 fixes the build on linux due to RTLD_DEFAULT having not been defined.
110122
110123commit 78f9592c112d4245f6119b98c244bbb4cae3e5aa
110124Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
110125Date: Wed Jan 10 16:04:20 2007 +0000
110126
110127 lnx_ev56.c has to be built with -mcpu=ev56. Fix it.
110128
110129commit ea82333dc1799a7e013205adbc89874bc8d273cb
110130Author: Peter Hutterer <peter@cs.unisa.edu.au>
110131Date: Wed Jan 10 14:52:13 2007 +1030
110132
110133 Xi: add missing call for SProcXChangeDeviceCursor
110134 add DeviceEnterNotify and DeviceLeaveNotify
110135 fix: QueryDevicePointer crashed if called with keyboard device
110136
110137commit 6a2fb2928714ce77ee342cdc23a1178e5e766cf2
110138Author: Eric Anholt <eric@anholt.net>
110139Date: Tue Jan 9 16:34:40 2007 -0800
110140
110141 Track rename of DamagePost -> DamageAdd.
110142
110143commit da09964a931cc64d05ab571bf545fdad35a6d395
110144Merge: e1f73d2 e3aa6ad
110145Author: Ian Romanick <idr@us.ibm.com>
110146Date: Tue Jan 9 15:27:34 2007 -0800
110147
110148 Merge branch 'origin' into pci-rework
110149
110150 Conflicts:
110151
110152 hw/xfree86/int10/generic.c
110153
110154commit e3aa6ad201eb20862c11c000e76206e317a96dc9
110155Author: Matthieu Herrb <matthieu@blues.laas.fr>
110156Date: Tue Jan 9 14:14:19 2007 +0100
110157
110158 Multiple integer overflows in dbe and render extensions
110159 CVE IDs: CVE-2006-6101 CVE-2006-6102 CVE-2006-6103
110160
110161commit 359d20532bdcef6a540a551578d000afbb609c2d
110162Author: Michel Dänzer <michel@tungstengraphics.com>
110163Date: Tue Jan 9 09:53:45 2007 +0100
110164
110165 Require glproto >= 1.4.8 for GLX.
110166
110167 It builds against 1.4.7 as well, but it hardcodes the GLX_EXT_tfp tokens that
110168 were finalized in 1.4.8, so GLX_EXT_tfp breaks if the client side was built
110169 against an older glproto. This will hopefully alert people to rebuild other
110170 components (in particular Mesa) against the new glproto as well.
110171
110172commit 88740c4855babedbea420b5e1b35ae105d1f1026
110173Author: Alan Coopersmith <alan.coopersmith@sun.com>
110174Date: Mon Jan 8 17:36:07 2007 -0800
110175
110176 Use PKG_CHECK_EXISTS(libdrm) to determine if DRI should be enabled on Solaris
110177
110178commit 282a4dcaabc5f0cd6f7d3819aa648333b93b265e
110179Author: Michel Dänzer <michel@tungstengraphics.com>
110180Date: Mon Jan 8 19:22:41 2007 +0100
110181
110182 Attempt to fix drawable type checks in dixLookupDrawable().
110183
110184 Not sure this is 100% correct either, but it fixes at least one reproducible
110185 crasher where it returned a pixmap to dixLookupWindow().
110186
110187commit a7ab7932938820a795bb6fb8e0444e0824433b99
110188Author: Peter Hutterer <peter@cs.unisa.edu.au>
110189Date: Mon Jan 8 17:31:16 2007 +1030
110190
110191 xfree86: fix XI86_SHARED_POINTER flag setting
110192
110193commit 95e1a88050dde61e9b2407428042a43e47b46e18
110194Author: Peter Hutterer <peter@cs.unisa.edu.au>
110195Date: Mon Jan 8 12:31:18 2007 +1030
110196
110197 Xi: Adding ChangeDeviceCursor request
110198
110199commit 0b73a7eb17fd848c6bdc6a65ba835aa2cbfc3cfd
110200Author: Eric Anholt <eric@anholt.net>
110201Date: Fri Jan 5 18:12:04 2007 -0800
110202
110203 Add support for the DamagePost (XDamage 1.1) request.
110204
110205 This makes damageproto >= 1.1 a requirement to build.
110206
110207commit dfb2c10413e22afd8d486a982870f874326d5ef4
110208Author: Ian Romanick <idr@us.ibm.com>
110209Date: Fri Jan 5 10:15:09 2007 -0800
110210
110211 Add missing #else from previous commits.
110212
110213commit f90c3e226b105bf77beb94723fc08bdff14834be
110214Author: Ian Romanick <idr@us.ibm.com>
110215Date: Thu Jan 4 15:38:16 2007 -0800
110216
110217 Re-regenerate from Mesa scripts.
110218
110219 DO NOT HAND EDIT THESE FILES! For cryin' out loud, there's even a
110220 comment to that effect in the file's header...
110221
110222commit b7ca5d14ce7ba410b0dab5c2289f6d7b75e763df
110223Author: Ian Romanick <idr@us.ibm.com>
110224Date: Thu Jan 4 15:37:33 2007 -0800
110225
110226 Incorporate new byte-order related configure changes.
110227
110228commit 8dd5771a1b91c331860b667fb18e484452000aad
110229Merge: 45aa26c 7d2ec92
110230Author: Ian Romanick <idr@us.ibm.com>
110231Date: Thu Jan 4 15:01:38 2007 -0800
110232
110233 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
110234
110235 Conflicts:
110236
110237 GL/glx/indirect_dispatch_swap.c
110238 GL/glx/swap_interval.c
110239
110240commit 45aa26ccb4f61c2919ce2475d0907c6e1b177da2
110241Author: Ian Romanick <idr@us.ibm.com>
110242Date: Thu Jan 4 14:55:51 2007 -0800
110243
110244 Regenerate from Mesa scripts.
110245
110246 Regenerate source files from Mesa scripts. This causes the generated
110247 files to use glxbyteorder.h.
110248
110249commit 6d603bb47ff9d238637adbf30c6e9697e6e7e6fa
110250Author: Ian Romanick <idr@us.ibm.com>
110251Date: Thu Jan 4 14:49:26 2007 -0800
110252
110253 Add new header file containing byte-order wrappers.
110254
110255 Move the byte-order related wrappers out of the individual source
110256 files into a dedicated header file. Modify the single hand-coded
110257 source file that uses the byte-order wrappers to use the new header
110258 file.
110259
110260commit 7d2ec92170ebbdfa10a05734cb7cfaac97d19d65
110261Author: Eric Anholt <eric@anholt.net>
110262Date: Thu Jan 4 12:24:48 2007 -0800
110263
110264 Keep track of how many visuals we set up for GLcore, to avoid an invalid free.
110265
110266 The proper fix would involve actually setting up the ARGB visual for GLcore,
110267 but I just want the server to not crash at exit.
110268
110269commit aab2ca204279b638c7e5bb6b8427c58be9704c57
110270Author: Eric Anholt <eric@anholt.net>
110271Date: Thu Dec 21 09:16:19 2006 -0800
110272
110273 Try dlsym(RTLD_DEFAULT) first when finding symbols.
110274
110275 The previous mechanism failed when finding drm symbols now that libdrm has
110276 moved to being linked by libdri instead of being linked into the server.
110277
110278commit 2fd4626fa6969b84d8e2f9db16d6e2d44c4bc499
110279Author: Alan Coopersmith <alan.coopersmith@sun.com>
110280Date: Wed Jan 3 15:44:55 2007 -0800
110281
110282 Make GLX byteswap macros more portable
110283
110284 - Use autoconf tests instead of platform-specific #ifdef's to decide
110285 which macros to use.
110286 - Provide fallbacks for platforms like Solaris that don't provide any
110287 of the existing known forms.
110288
110289commit 66fa87292ef26bd0f464481287f3af992cd5741c
110290Author: Aaron Plattner <aplattner@nvidia.com>
110291Date: Wed Jan 3 10:27:07 2007 -0800
110292
110293 Fix BSF and BSR instructions in the x86 emulator.
110294
110295 Patch courtesy of Michael Yaroslavtsev.
110296
110297commit 3b5b7ef5c2ab1d196806f6359e0972fd78d204dd
110298Author: Fredrik Höglund <fredrik@kde.org>
110299Date: Wed Jan 3 21:05:35 2007 +0100
110300
110301 Move the code for resetting the DPMS mode in response to input events,
110302 from WaitForSomething to mieqProcessInputEvents.
110303
110304 mieqProcessInputEvents already handles resetting the screen saver.
110305
110306commit 953a9ef949b4c57d28daeec57031fe1ce368c27c
110307Author: Keith Packard <keithp@guitar.keithp.com>
110308Date: Thu Dec 21 23:50:39 2006 -0800
110309
110310 Track physical screen size and send out updates when that changes.
110311
110312 Events and internal data structures need to be updated whenever the physical
110313 or pixel size of the screen changes. The code was ignoring the physical
110314 size, so changing only that would not be registered anywhere.
110315 (cherry picked from f42e3cea236fa0091ed398a818fc8e17b0e1b3df commit)
110316
110317commit e79602fca2f2cced66136729cdda4d356b0bdda0
110318Author: Keith Packard <keithp@neko.keithp.com>
110319Date: Sat Dec 30 21:52:22 2006 -0800
110320
110321 Use RRScreenSetSizeRange in 1.0 compat. Check RRGetInfo for error.
110322
110323 The RRScreenSizeSetRange function is used externally for 1.2 API drivers,
110324 but can also be used in the 1.0 compatibility code. This also ensures that
110325 the right changed bits are set so that clients are correctly notified when
110326 the range changes.
110327
110328 RRGetInfo can return an error, use that to return BadAlloc to clients
110329 instead of blindly going on with various requests.
110330 (cherry picked from f05dd384d38c76dd9662933a03625dfef5b1c81f commit)
110331
110332commit dc5eb4523298f966bd5fd9ae6672160034b5e82c
110333Author: Michel Dänzer <michel@tungstengraphics.com>
110334Date: Sun Dec 31 17:59:44 2006 +0100
110335
110336 fbdevhw: Override RGB offsets and masks after setting initial mode.
110337
110338 This is a hack, but it should be a NOP for all the setups that worked before
110339 and actually seems to fix some others...
110340
110341 Based on a patch by Peter Teichmann from
110342 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=338241 .
110343
110344commit d077c0da470ab7291e8d838eaace57b066477d6f
110345Author: Michel Dänzer <michel@tungstengraphics.com>
110346Date: Sun Dec 31 17:23:31 2006 +0100
110347
110348 fbdevhw: Use displayWidth for fbdev virtual width when appropriate.
110349
110350 The fbdev API doesn't allow setting the pitch explicitly, so we have to set
110351 the virtual width to the pitch we're using for drawing. This fixes corruption
110352 after changing the virtual width with RandR.
110353
110354commit c385bcf0bde38dd869f7065f859dd4b4126f5690
110355Author: Michel Dänzer <michel@tungstengraphics.com>
110356Date: Sat Dec 30 16:44:31 2006 +0100
110357
110358 fbdevhw: Fix some issues with the previous commit.
110359
110360 Fix a TRACE_ENTER typo and only update the internal fbdev mode state cache
110361 after actually setting a mode.
110362
110363commit f6815cb68b0f6698497348fc6e4214dacef33b95
110364Author: Michel Dänzer <michel@tungstengraphics.com>
110365Date: Sat Dec 30 10:18:28 2006 +0100
110366
110367 fbdevhw: Consolidate modeset ioctl calling, report failure if it modifies mode.
110368
110369 The fbdev API allows the driver to 'accept' modes it doesn't really support by
110370 modifying it to the nearest supported mode. Without this check, e.g. vesafb
110371 would appear to accept all modes, even though it actually can't set any modes
110372 other than the bootup mode at all.
110373
110374commit 083b790515faaf134a78abc4b0a7ef0d6ea5db75
110375Author: Eric Anholt <eric@anholt.net>
110376Date: Thu Dec 28 13:21:25 2006 -0800
110377
110378 Switch the default migration heuristic for EXA to "always".
110379
110380 This has been what has been used the most successfully post-damagetrack.
110381 The current thinking is that:
110382 1) We should be able to accelerate basically everything. So we don't need to
110383 try to migrate trees of pixmaps permanently out of framebuffer to speed
110384 CPU drawing up.
110385 2) Migration is cheaper in the thrashing case, so we don't want to go to a lot
110386 of effort to try (and fail badly) to find a working set.
110387
110388commit cfbc7379f0232bb336461f6d2a8496d3d0763e7e
110389Author: Eric Anholt <eric@anholt.net>
110390Date: Thu Dec 28 13:15:11 2006 -0800
110391
110392 Export exaMove{In,Out}Pixmap().
110393
110394commit 683ca3f7afaf15fd3ca7918f6175b5a9e4a6f05b
110395Merge: 05f9150 9563b2e
110396Author: Eric Anholt <eric@anholt.net>
110397Date: Wed Dec 27 16:11:31 2006 -0800
110398
110399 Merge branch 'exa-damagetrack'
110400
110401commit 05f915050cad72d4fb39cbb886be57beeac18749
110402Author: Peter Hutterer <peter@cs.unisa.edu.au>
110403Date: Wed Dec 27 16:38:06 2006 +0000
110404
110405 dix/events: take screen number, not pointer, in PostSyntheticMotion
110406 Since we were using PostSyntheticMotion incorrectly anyway, update the
110407 declared API to match.
110408
110409commit c1674660a7115ebf993dcde78f4e45f756e4c951
110410Author: Daniel Stone <daniel@fooishbar.org>
110411Date: Sun Dec 24 06:28:44 2006 +0200
110412
110413 os: test for userland, not kernel
110414
110415 It doesn't matter which kernel we're running on, the relevant part when
110416 dealing with includes is what our userland is.
110417
110418commit 83080809f9a1c1d24b0318e54632f25f5940da25
110419Author: Marc Aurele La France <tsi@ualberta.ca>
110420Date: Sun Dec 24 06:28:21 2006 +0200
110421
110422 xfree86: deal with pitch that isn't a multiple of the granularity
110423
110424 When the pitch isn't a multiple of the granularity, allocate more space to
110425 compensate.
110426
110427commit 329f6417275bb1201ba66c29b202028eeab3a355
110428Author: Daniel Stone <daniel@fooishbar.org>
110429Date: Sat Dec 9 22:51:59 2006 +0200
110430
110431 XkbCopyKeymap: make sure sym_interpret is always valid
110432
110433 Make sure we're not copying sym_interpret across from an empty source.
110434
110435commit 43a4376bd72ef1b6486cddb60ad57b2d6169292a
110436Merge: 4b1c9ac d9e079d
110437Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110438Date: Fri Dec 22 13:53:55 2006 -0500
110439
110440 Merge branch 'master' into my-XACE-SELINUX
110441
110442commit d9e079d2a385203fdd18d958cfc19d759cab4ba8
110443Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110444Date: Fri Dec 22 13:07:09 2006 -0500
110445
110446 Zero out client devPrivates on allocation.
110447
110448commit 4b1c9ac3d13767e395b47e76b37f9f3a569e7be1
110449Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110450Date: Fri Dec 22 13:04:50 2006 -0500
110451
110452 Policy updates.
110453
110454commit cd71e861830081807e5b93ae89c73c17986c6330
110455Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110456Date: Wed Dec 20 13:45:24 2006 -0500
110457
110458 Naming change: Security*Access -> Dix*Access.
110459 Clarify some error message strings.
110460
110461commit 9fd2f167ec02ee170b87f02ddca89eba0b2e2389
110462Author: Peter Hutterer <peter@cs.unisa.edu.au>
110463Date: Wed Dec 20 16:30:36 2006 +1030
110464
110465 xfree86: fix xf86ProcessCommonOptions() core and shared pointer assignment
110466
110467 TAG: mpx-0.3.0
110468
110469commit 056c919d849a78c8d507bfcecfa74c0b0165751a
110470Author: Peter Hutterer <peter@cs.unisa.edu.au>
110471Date: Wed Dec 20 14:49:17 2006 +1030
110472
110473 removing MPX extension files
110474
110475commit b55e1239ac3b2962a33d8af1f911cc0f01f40f23
110476Author: Peter Hutterer <peter@cs.unisa.edu.au>
110477Date: Wed Dec 20 14:43:38 2006 +1030
110478
110479 xfree86: fix xf86ActivateDevice(): only set to MP device if not a keyboard
110480
110481commit 202b46eb6bf0d5f94973c2bf1e4ebe9d154eadbf
110482Author: Peter Hutterer <peter@cs.unisa.edu.au>
110483Date: Wed Dec 20 13:05:53 2006 +1030
110484
110485 xfree86: Changing "IsMPDevice" to "SharedPointer" option. Devices will default
110486 to MP devices.
110487
110488 Xi: Adding QueryDevicePointer request/reply
110489 Adding WarpDevicePointer request/reply
110490
110491commit e437f357b6850a6c87ca6696870b3abd40e5b8ed
110492Author: Alan Coopersmith <alan.coopersmith@sun.com>
110493Date: Tue Dec 19 16:38:34 2006 -0800
110494
110495 xorg.conf man page should say "XFree86-DGA", not "Xorg-DGA"
110496
110497commit a81dbaae0597492f0245080c6f5af7158e05e2d0
110498Merge: fb6d676 d442998
110499Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110500Date: Tue Dec 19 16:11:19 2006 -0500
110501
110502 Merge branch 'master' into my-XACE-SELINUX
110503
110504commit d442998e39611be6805ea261f2286a2fd00f49b1
110505Author: Eric Anholt <eric@anholt.net>
110506Date: Fri Dec 8 13:35:36 2006 -0800
110507
110508 Only do the _POSIX_C_SOURCE hackery on linux where it's required.
110509
110510 On other OSes, the nasty hack was resulting in prototypes being hidden, so just
110511 don't do it.
110512
110513commit 228b9f77696190e47d4c96d6e0809bf645751557
110514Author: Eric Anholt <eric@anholt.net>
110515Date: Fri Dec 8 13:32:22 2006 -0800
110516
110517 Include sys/select.h to get FD_ISSET.
110518
110519commit 9563b2eea2f61246b6a9e14e00c701f693efa4e1
110520Author: Michel Dänzer <michel@tungstengraphics.com>
110521Date: Tue Dec 19 18:57:22 2006 +0100
110522
110523 EXA: Lots of damage tracking fixes.
110524
110525 Mostly due to exaDrawableDirty() now calculating the backing pixmap coordinates
110526 internally, for cases where they aren't trivially known. There's a new
110527 exaPixmapDirty() function for the other cases.
110528
110529commit 467c00cf450826e0bf06fe94470ec193af625d68
110530Author: George Sapountzis <gsap7@yahoo.gr>
110531Date: Tue Dec 19 18:45:25 2006 +0100
110532
110533 exaGlyphs: mark dirty for software path also.
110534
110535 This affects drivers with no UploadToScreen or UploadToScreen failures.
110536
110537commit 4334860e69e7d5b156082bd05c7a86708e5bad4c
110538Merge: 7e47176 fdcc22c
110539Author: Michel Dänzer <michel@tungstengraphics.com>
110540Date: Tue Dec 19 16:29:26 2006 +0100
110541
110542 Merge branch 'master' into exa-damagetrack
110543
110544 Conflicts:
110545
110546 exa/exa_accel.c
110547 exa/exa_migration.c
110548
110549commit fdcc22ca1704d3519156c66804528c21b04fea65
110550Author: Michel Dänzer <michel@tungstengraphics.com>
110551Date: Tue Dec 19 16:11:17 2006 +0100
110552
110553 exaCopyNtoN: Fix usage of 'dx' and 'dy' instead of 'reverse' and 'upsidedown'.
110554
110555commit 67c2a86e59e915d9a5681e9d233478cfea3e51ed
110556Author: Michel Dänzer <michel@tungstengraphics.com>
110557Date: Tue Dec 19 15:44:18 2006 +0100
110558
110559 EXA: Compare backing pixmaps instead of drawables against driver limits.
110560
110561 The driver operations are always contained within the backing pixmaps, it
110562 doesn't matter if the drawables are bigger.
110563
110564commit 6b1e354dbb6e8ed9f2c654bbe7f8bbf241843d1c
110565Author: Eric Anholt <eric@anholt.net>
110566Date: Tue Dec 19 15:24:19 2006 +0100
110567
110568 EXA: Disable SHM pixmaps.
110569
110570 See https://bugs.freedesktop.org/show_bug.cgi?id=6772 .
110571
110572commit ca5ebe3971d8ebcfed00c5ebcd026cdd0ce0c6ba
110573Merge: 4d07b50 1b029fd
110574Author: Peter Hutterer <peter@cs.unisa.edu.au>
110575Date: Tue Dec 19 10:51:01 2006 +1030
110576
110577 Merge branch 'master' of git://anongit.freedesktop.org/git/xorg/xserver
110578
110579commit 2d0a63126b3d3a17005b7e122617ee0c5f44a55b
110580Merge: d8b5394 1b029fd
110581Author: Peter Hutterer <peter@cs.unisa.edu.au>
110582Date: Tue Dec 19 10:31:40 2006 +1030
110583
110584 Merge branch 'master' of git://anongit.freedesktop.org/git/xorg/xserver
110585
110586commit 1b029fd896b76096905c516925ce0214fe14632c
110587Author: Alan Coopersmith <alan.coopersmith@sun.com>
110588Date: Mon Dec 18 14:51:04 2006 -0800
110589
110590 Xorg & Xserver man page updates for 1.2 release
110591
110592 - Added -extension & +extension to Xserver man page
110593 - Changed Xorg synopsis from X11R6 to X11R7
110594 - Clarified Xorg ancestry description
110595 - Moved Solaris to free/Open Source OS list
110596 - Removed references to MetroLink module loader & getconfig
110597 - Converted (1) to (__appmansuffix__) in a few more places
110598 - Replaced http://www.freedesktop.org/cvs/ with http://gitweb.freedesktop.org/
110599
110600commit d8b5394eda9d92b7193004931caa6c24a337b2e6
110601Author: Peter Hutterer <peter@cs.unisa.edu.au>
110602Date: Mon Dec 18 17:35:34 2006 +1030
110603
110604 Cleaning up ifdef MPX from remaining files, removing it from configure.ac.
110605 Removing building mpx extension from Makefile.am
110606
110607commit 03c554283e49e449fd1282cf32564d15d9fd2c77
110608Author: Peter Hutterer <peter@cs.unisa.edu.au>
110609Date: Mon Dec 18 15:33:54 2006 +1030
110610
110611 mi: removing MPX ifdefs
110612 global sprite renaming in mipointer and misprite
110613 fixed: multiple remove cursor call in miSpriteInstallColormap
110614 fixed: wrong core pointer usage in miSpriteStoreColors
110615
110616 dix: bugfix in CheckCursorConfinement
110617 removing MPX ifdefs
110618 removing MPX event generation (using Xi solely now)
110619 bugfix GrabDevice: uninitialized field in grab struct caused segfault
110620
110621 xfree86: removing MPX fdefs
110622
110623 Xi: removing MPX protocol events
110624
110625commit a5fcf1e5e7452c9be82f63b6c2be2a25c4109523
110626Author: James Steven Supancic III <arrummzen@arrummzen.net>
110627Date: Sat Dec 16 12:02:38 2006 -0500
110628
110629 Fix RENDER issues (bug #7555) and implement RENDER add/remove screen
110630 support (bug #8485).
110631
110632commit c92f7bef54fa737766d65fe32c200f405f39228c
110633Author: Kevin E Martin <kem@freedesktop.org>
110634Date: Sat Dec 16 12:01:49 2006 -0500
110635
110636 For Xvfb, Xnest and Xprt, compile fbcmap.c with -DXFree86Server
110637
110638commit 012807356883128fde58bb2d4f91dd356d6418fc
110639Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110640Date: Fri Dec 15 18:27:16 2006 -0500
110641
110642 Add loud warnings to deprecated lookup functions.
110643 Hopefully this will alert external driver maintainers.
110644
110645commit ab1d5b0c31a1cfce95ab6b1d06f209f2c44e19ac
110646Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110647Date: Fri Dec 15 17:26:58 2006 -0500
110648
110649 Convert callers of LookupClient() to dixLookupClient().
110650
110651commit f11dafaafc68f5cff1a1538d9566907786d8ab72
110652Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110653Date: Fri Dec 15 16:51:58 2006 -0500
110654
110655 Convert callers of SecurityLookupDrawable() to dixLookupDrawable().
110656
110657commit 10aabb729d1586db344f9c1abdf1cf45e7ddaa7a
110658Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110659Date: Fri Dec 15 16:36:29 2006 -0500
110660
110661 Convert callers of LookupDrawable() to dixLookupDrawable().
110662
110663commit 25d5e0a629f82d95bd71daf9a920a70e095b5188
110664Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110665Date: Fri Dec 15 15:50:46 2006 -0500
110666
110667 Convert callers of SecurityLookupWindow() to dixLookupWindow().
110668
110669commit 04c721854fbf1bd6379c165a53fab2bdc09961c0
110670Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110671Date: Fri Dec 15 14:11:40 2006 -0500
110672
110673 Convert callers of LookupWindow() to dixLookupWindow().
110674
110675commit 670bbb87310503fcc17203cecfa6f4f2f5db51d2
110676Author: Keith Packard <keithp@neko.keithp.com>
110677Date: Wed Dec 13 01:21:32 2006 -0800
110678
110679 RandR 1.2 rotation code must adjust width/height.
110680
110681 Mode lines reflect the monitor mode, not the projected size into the frame
110682 buffer. Flip width/height around so that the dimensions are oriented
110683 correctly.
110684 (cherry picked from 612a8e61803da8db0e305cbb093696b8e4284572 commit)
110685
110686commit 6c6901434ab469dd03b79fc98cd4a2b64d339305
110687Author: Keith Packard <keithp@mandolin.keithp.com>
110688Date: Wed Dec 13 00:58:54 2006 -0800
110689
110690 RandR 1.0 refresh rates unscrambled. SetScreenConfig uses RRCrtcSet right.
110691
110692 RandR 1.0 refresh rates were scrambled when working with a 1.2 driver that
110693 returned sizes in a mixed order. SetScreenConfig was treating RRCrtcSet as
110694 returning an RandR status instead of a Bool.
110695 (cherry picked from 6dc711833d7387372012fdff1ce1df3aefa2d234 commit)
110696
110697commit 628c7daeb12713d28e85e6b49fa037a7748dff83
110698Author: Keith Packard <keithp@guitar.keithp.com>
110699Date: Tue Dec 12 22:59:03 2006 -0800
110700
110701 RandR: config time updates when hardware config changes.
110702
110703 The config time in the RandR protocol reflects when the hardware state has
110704 changed. It was getting changed anytime the driver changed the usage
110705 of the hardware as well.
110706 (cherry picked from 98d18a6578130adb411ca4bcc776fcb7e07f189f commit)
110707
110708commit d742025f435f3eb7458cf8284d59300bc9a850aa
110709Author: Keith Packard <keithp@guitar.keithp.com>
110710Date: Tue Dec 12 20:16:49 2006 -0800
110711
110712 RandR mode list needs both output and crtc modes.
110713
110714 When an output no longer reports the current mode, it must still be included
110715 in the list advertised by the X server. Walk the crtcs to ensure it is
110716 included.
110717 (cherry picked from 78689d0d6630afcbcd3ce5394d12c2564a489f45 commit)
110718
110719commit 9e32bf98bc9ab17a137664d01f59a8f426f7ff3b
110720Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110721Date: Thu Dec 14 19:31:58 2006 -0500
110722
110723 Remove now-unused macro definitions from dix.h.
110724
110725commit 00f0705b3bb444ac934fc902cd23130f1777eab2
110726Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110727Date: Thu Dec 14 19:15:21 2006 -0500
110728
110729 Remove instances of macros SECURITY_VERIFY_GEOMETRABLE and SECURITY_VERIFY_GC.
110730
110731commit 5e334f06a1ef89891f9df2a371e4662340bec26b
110732Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110733Date: Thu Dec 14 18:27:09 2006 -0500
110734
110735 Remove instances of macros VERIFY_GEOMETRABLE and VERIFY_GC.
110736
110737commit 51b69ff499c05f59cb1e577c4e8abf6f7f283b3e
110738Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110739Date: Thu Dec 14 17:53:43 2006 -0500
110740
110741 Remove instances of macro SECURITY_VERIFY_DRAWABLE.
110742
110743commit 0cf75e74322e2b6a6efc7acf892e04365fde503b
110744Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110745Date: Thu Dec 14 17:27:12 2006 -0500
110746
110747 Remove instances of macros LOOKUP_DRAWABLE and VERIFY_DRAWABLE.
110748
110749commit ab1886df73b73360fa3bd7ce8e01affc074cbc8d
110750Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110751Date: Thu Dec 14 15:42:19 2006 -0500
110752
110753 Add new, combined dix lookup functions (tweak).
110754
110755commit 60cdc592fe042c03ceb5d4c3344acfbbf5d8ae28
110756Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110757Date: Thu Dec 14 14:46:03 2006 -0500
110758
110759 Add new, combined dix lookup functions.
110760
110761commit 6c46645cfc1afda8aeabfe0ed4d9342673b702f1
110762Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110763Date: Thu Dec 14 14:45:42 2006 -0500
110764
110765 Naming change: Security*Access -> Dix*Access
110766
110767commit a96788058e48f3e97c66da6dafd58828f0c2ce53
110768Author: Peter Hutterer <peter@cs.unisa.edu.au>
110769Date: Thu Dec 14 10:50:18 2006 +1030
110770
110771 mi: removing MPX ifdefs
110772 global sprite renaming in mipointer and misprite
110773 fixed: multiple remove cursor call in miSpriteInstallColormap
110774 fixed: wrong core pointer usage in miSpriteStoreColors
110775
110776 dix: bugfix in CheckCursorConfinement
110777 removing MPX ifdefs
110778 removing MPX event generation (using Xi solely now)
110779
110780commit b88ad820fac81d0dfd557a384bf0406e8893e7af
110781Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
110782Date: Wed Dec 13 12:13:11 2006 +0000
110783
110784 Set Int10Current->Tag for the linux native int10 module
110785 Fixes bug #9296
110786 (cherry picked from 731952c561a3972d09d1315f4fd31466e459ccb9 commit)
110787
110788commit fb6d676de5aa606d943715437a12a68d9a41f386
110789Author: Eamon Walsh <ewalsh@moss-huskies.epoch.ncsc.mil>
110790Date: Tue Dec 12 16:17:51 2006 -0500
110791
110792 Add xserver object class to list of object classes.
110793
110794commit 7b90944258eba66b61328480759833ad7589bcca
110795Author: Eamon Walsh <ewalsh@moss-huskies.epoch.ncsc.mil>
110796Date: Tue Dec 12 15:59:38 2006 -0500
110797
110798 Change MLS levels in config file contexts to more sane defaults.
110799
110800commit 568c09481e5d62091d032837171a36f409f39379
110801Author: Eamon Walsh <ewalsh@moss-huskies.epoch.ncsc.mil>
110802Date: Tue Dec 12 15:59:08 2006 -0500
110803
110804 Split AssignClientState() into two routines, new routine is server-specific.
110805
110806commit 62e7722ebd8d341a23eb56cb330eeb00e6a975a6
110807Merge: e124806 f3c6090
110808Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110809Date: Tue Dec 12 13:54:33 2006 -0500
110810
110811 Merge branch 'XACE-SELINUX' into my-XACE-SELINUX
110812
110813commit e124806994675e16ca8e3937388f2cadeb529fc3
110814Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110815Date: Tue Dec 12 13:35:22 2006 -0500
110816
110817 Remove trailing whitespace (whitespace police).
110818
110819commit ca77c121075a9de1f47d42f6aaf91c20185231de
110820Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110821Date: Tue Dec 12 13:26:52 2006 -0500
110822
110823 Naming change: Security*Operation -> Xace*Operation
110824
110825commit a60da1db7cced28c07960a713eb18deb45beb432
110826Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110827Date: Wed Nov 29 22:19:57 2006 -0500
110828
110829 Add required root window context to config file.
110830
110831commit 5719afe6d3a246985709e6f045617c1e16a7da51
110832Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110833Date: Thu Oct 26 20:20:57 2006 -0400
110834
110835 Change symbol in libaudit library test.
110836
110837commit 354c80da66af141e8ba6d75fed75a0f482987956
110838Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110839Date: Thu Oct 5 16:07:26 2006 -0400
110840
110841 Improve error handling, messages during initialization.
110842
110843commit 3714d9149928754afcd6b2466a1371ca32e17985
110844Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110845Date: Wed Oct 4 16:23:35 2006 -0400
110846
110847 Experimental window property holding security context.
110848
110849commit 7f16c38ae2b47b195609d8fedefb7b28f612b2d4
110850Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110851Date: Fri Sep 8 15:31:18 2006 -0400
110852
110853 Add SELinux extension configure-time support.
110854
110855commit a7f4bbea87ada1d699bfd9e3b6a98f06191650f6
110856Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110857Date: Fri Sep 8 15:17:01 2006 -0400
110858
110859 Add SELinux extension to the module/extension loader.
110860
110861commit 28e80cd65b1207b123c02f895851bb6d207aa3c1
110862Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110863Date: Fri Sep 8 15:15:19 2006 -0400
110864
110865 Add SELinux extension configure-time support.
110866
110867commit 83aad2be8a80890f349c2f9caf84786333f7cc8c
110868Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
110869Date: Fri Sep 8 15:11:04 2006 -0400
110870
110871 Add SELinux extension source files.
110872
110873commit 81281cb298a5825bc7a2e692375a86199293bbbe
110874Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
110875Date: Tue Dec 12 11:28:24 2006 +0000
110876
110877 Fix bad commit
110878
110879commit 792e0f71c6a435b2e28f8a4cdcc790f3b982e62c
110880Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
110881Date: Mon Dec 11 14:54:49 2006 +0000
110882
110883 Fix Xming fails to use xkb bug
110884 bug #5049 (Colin Harrison)
110885
110886commit 27d4b84f268ac21601f7f52a7e257f70753396b3
110887Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
110888Date: Mon Dec 11 14:50:08 2006 +0000
110889
110890 Fix Tooltip from minimized clients
110891
110892 Bug #3678 (Colin Harrison)
110893
110894commit ae3c24da34cd8eeb77a6389861856fd51e0841f5
110895Author: Peter Hutterer <peter@cs.unisa.edu.au>
110896Date: Mon Dec 11 18:09:59 2006 +1030
110897
110898 dix: Moving SpriteRec into DeviceIntRec
110899 removing global sprite structure
110900 beginning to remove MPX ifdefs
110901
110902 xnest: Fix to make xnest compile again
110903
110904commit fb8364bca30fe9268e807b0a9a3ebf875ee1fce2
110905Author: Adam Jackson <ajax@benzedrine.nwnk.net>
110906Date: Sun Dec 10 11:24:05 2006 -0500
110907
110908 Accept EDID > 1.3 but < 2.0 if we find it, assume it's compatible.
110909
110910commit e1f73d220873fa091695e46b7fcd008663a95a6f
110911Merge: 27d682f d56249a
110912Author: Ian Romanick <idr@us.ibm.com>
110913Date: Fri Dec 8 17:24:15 2006 -0800
110914
110915 Merge branch 'master' of ssh+git://idr@git.freedesktop.org/git/xorg/xserver into pci-rework
110916
110917 Conflicts:
110918
110919 hw/xfree86/common/xf86Configure.c
110920 hw/xfree86/common/xf86Helper.c
110921 hw/xfree86/common/xf86pciBus.c
110922 hw/xfree86/int10/helper_exec.c
110923 hw/xfree86/os-support/bus/Pci.c
110924 hw/xfree86/os-support/bus/linuxPci.c
110925 hw/xfree86/os-support/linux/lnx_pci.c
110926 hw/xfree86/scanpci/Makefile.am
110927 hw/xfree86/utils/pcitweak/Makefile.am
110928 hw/xfree86/utils/scanpci/Makefile.am
110929
110930commit d56249a15ead51ad4d2117d5538ada24af05b693
110931Merge: f1f8df1 ec84f72
110932Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
110933Date: Fri Dec 8 15:52:37 2006 -0600
110934
110935 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
110936
110937commit f1f8df1889ab656bb57596e2f85408f15f42cd5d
110938Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
110939Date: Fri Dec 8 15:51:44 2006 -0600
110940
110941 For MANDEFS, also replace __mandir__ for $(mandir) which includes
110942 the prefix. Noticed wrong path to man pages on both Ubuntu and on
110943 NetBSD with pkgsrc.
110944
110945commit ec84f72d077eaf2e7768a1f5398f65a5e1714d08
110946Author: Daniel Stone <daniel@fooishbar.org>
110947Date: Fri Dec 8 21:31:47 2006 +0200
110948
110949 XkbCopyKeymap: always initialise map and preserve
110950
110951 If we don't have map or preserve in the source map, make sure the
110952 destination is initialised to NULL, and freed if it's a valid type.
110953
110954commit e59aeac1ff13ea53f44dba1ac7800f37f4532ca1
110955Author: Daniel Stone <daniel@fooishbar.org>
110956Date: Wed Dec 6 23:44:06 2006 +0200
110957
110958 xfree86: remove stray debug line
110959
110960commit 729fca33a417ae3dfb180caf0ea8946ef9eee1df
110961Author: Michel Dänzer <michel@tungstengraphics.com>
110962Date: Thu Dec 7 12:09:18 2006 +0100
110963
110964 Revert "xfree86 DDX: Delete DDX screens in ddxGiveUp()."
110965
110966 This reverts commit a6381e69845f58d2b3282992b1f881015190f1bc.
110967
110968 See https://bugs.freedesktop.org/show_bug.cgi?id=9224 .
110969
110970commit 4ea6dfb984063117eef5c2f931205b1c3eb3108b
110971Author: Daniel Stone <daniel@fooishbar.org>
110972Date: Wed Dec 6 23:24:39 2006 +0200
110973
110974 whitespace police
110975
110976commit 4cba1a1ebfbdd7ab489b1b1ffb6656cbe88eb61e
110977Author: Daniel Stone <daniel@fooishbar.org>
110978Date: Wed Dec 6 23:21:38 2006 +0200
110979
110980 config: bus reconnect support
110981
110982 Add support for reconnecting to the bus when it restarts.
110983
110984commit 0c5dab5c8eaa174f28054b9d20244a709c015210
110985Author: Daniel Stone <daniel@fooishbar.org>
110986Date: Wed Dec 6 23:21:15 2006 +0200
110987
110988 config: move config.h to hotplug.h
110989
110990 Also, move configInitialise to after OsInit, since the next commit will
110991 make it use a timer.
110992
110993commit 72e7f2ac6cf0db474d0defa7918f2a3ba76c0e46
110994Author: Daniel Stone <daniel@fooishbar.org>
110995Date: Wed Dec 6 23:18:52 2006 +0200
110996
110997 GetPointerEvents: always send valuator events for MotionNotify
110998
110999 Always chase a DeviceMotionNotify event with a DeviceValuator, which is
111000 not required in the spec, but will silently break the lib if you don't
111001 include.
111002
111003commit c458a70d650bd62b8f4706f022d1f3f347636db1
111004Author: Daniel Stone <daniel@fooishbar.org>
111005Date: Wed Dec 6 20:30:44 2006 +0200
111006
111007 GetPointerEvents: fix typo
111008
111009 Fix typo that resulted in inverted axes when using an absolute positioning
111010 device that didn't report y, and thus relied on the previous value.
111011
111012commit edabf45425f9ed79547f918cc0dfff4c268de386
111013Author: Daniel Stone <daniel@fooishbar.org>
111014Date: Sat Dec 2 16:37:19 2006 +0200
111015
111016 configure.ac: add CONFIG_LIB to Xvfb
111017
111018commit 8724af248cd6c93182fecd060fed09a556361080
111019Author: Daniel Stone <daniel@fooishbar.org>
111020Date: Sat Dec 2 16:20:34 2006 +0200
111021
111022 kdrive/mga: fix compiler warning
111023
111024 Change a case that only made one test for an if, preventing the compiler
111025 warning about all other PICT_* types being unsupported.
111026
111027commit 2f0a800ffdc881cdb3adf84f1ed97bbb63cba34c
111028Author: Daniel Stone <daniel@fooishbar.org>
111029Date: Mon Nov 27 22:22:53 2006 +0200
111030
111031 config: move to block/wakeup handler
111032
111033commit 99378b58dbc63160382ad9c41f9cb0dd2a24e9d1
111034Author: Daniel Stone <daniel@fooishbar.org>
111035Date: Mon Nov 27 22:22:33 2006 +0200
111036
111037 kdrive/tslib: remove vendor-specific hacks
111038
111039 Parts of the KDrive merge accidentally contained a bunch of
111040 vendor-specific hacks; sorry.
111041
111042commit 8884a73a3f4efa8276c5e38b9573201574c4f1f6
111043Author: Daniel Stone <daniel@fooishbar.org>
111044Date: Mon Nov 27 22:39:56 2006 +0200
111045
111046 xfree86/input: re-add support for disabling drag events
111047
111048commit f2903c12bb4bb0b7c94b96c55af8fa55507f9d7d
111049Author: Daniel Stone <daniel@fooishbar.org>
111050Date: Tue Nov 7 11:13:32 2006 +0200
111051
111052 SyntheticMotion: don't dereference sprite.screen when not using Xinerama
111053
111054 (cherry picked from aa052e43c6c293e14f78837e00c6b7581f9713bb commit)
111055
111056commit d17ec01e8395a8f14b75a10c8bf082b3f5a4fb36
111057Author: Daniel Stone <daniel@fooishbar.org>
111058Date: Fri Dec 1 00:41:41 2006 +0200
111059
111060 remove CID support (bug #5553)
111061
111062 Remove CID from all our fontpaths.
111063 (cherry picked from 69820a10e33e4582c192360996e866007114639d commit)
111064
111065commit abe5e079af715713097ab0daad29a3e9f523c398
111066Author: Alan Coopersmith <alan.coopersmith@sun.com>
111067Date: Wed Dec 6 07:58:03 2006 -0800
111068
111069 Update pci.ids to 2006-12-06 from pciids.sf.net
111070
111071commit 724f9cb578086e8483a2d0636dd6eb05d664d31c
111072Author: Aaron Plattner <aplattner@nvidia.com>
111073Date: Tue Dec 5 13:44:05 2006 -0800
111074
111075 Bug #9219: Use pWin->viewable instead of pWin->realized to catch InputOnly windows too.
111076
111077commit 3690de9b1b0902d395bc7d071fc05ebc8f75be2b
111078Author: Aaron Plattner <aplattner@nvidia.com>
111079Date: Tue Dec 5 12:42:12 2006 -0800
111080
111081 Bug #9219: Return BadMatch when trying to name the backing pixmap of an unrealized window.
111082
111083 Before this change, ProcCompositeNameWindowPixmap would name the screen pixmap
111084 if !pWin->realized.
111085
111086commit eb1d9f51af19dab20a95c2830ca1c508d8ee66bb
111087Author: Peter Hutterer <peter@cs.unisa.edu.au>
111088Date: Tue Dec 5 18:50:19 2006 +1030
111089
111090 Ironing some glitches caused by the merge
111091
111092commit 3c8f166022a2d19af14eb670fa382503ba1451cd
111093Merge: 2a51149 f9f7d7f
111094Author: Peter Hutterer <peter@cs.unisa.edu.au>
111095Date: Tue Dec 5 18:28:33 2006 +1030
111096
111097 Merge branch 'master' of git://anongit.freedesktop.org/git/xorg/xserver into mpx
111098
111099 Conflicts:
111100
111101 dix/events.c
111102
111103commit 2a511492f9a235d3f5390ab87b6c643cbea92d88
111104Author: Peter Hutterer <peter@cs.unisa.edu.au>
111105Date: Tue Dec 5 15:45:18 2006 +1030
111106
111107 Xi: adding MPX checks to ProcessOtherEvents
111108
111109 mpx: adding QueryPointer request and reply
111110 adjusting names of requests to Xlib compatible naming (xMPX*** instead of
111111 mpx***)
111112
111113commit f9f7d7f3be53c808abb5eaceb7a1abc55744a210
111114Author: Alan Coopersmith <alan.coopersmith@sun.com>
111115Date: Mon Dec 4 13:36:30 2006 -0800
111116
111117 Check for __sparc as well as __sparc__ for compatibility with Sun cc
111118
111119 (gcc defines __sparc__, Sun cc defines __sparc)
111120
111121commit ac90ce58ba1da3ed605adf75f4d54c34b578c402
111122Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
111123Date: Fri Dec 1 21:12:21 2006 -0500
111124
111125 Naming change: Security*Operation -> Xace*Operation
111126
111127commit f44f14fe564d834568a0afefba944223a73ea0f5
111128Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
111129Date: Fri Dec 1 20:48:15 2006 -0500
111130
111131 Define calls away when not building XACE, allowing ifdef's to be removed.
111132
111133commit a5d6499d666fea4a9988118ddd3a5e4c9cfcc32c
111134Author: Alan Coopersmith <alan.coopersmith@sun.com>
111135Date: Fri Dec 1 16:42:37 2006 -0800
111136
111137 Fix syntax error in configure check for SYSV_IPC that broke with Sun cc
111138
111139commit 89b2aa9be81613cb1a06bd535bf50ecf2a00208d
111140Author: Keith Packard <keithp@guitar.keithp.com>
111141Date: Mon Nov 27 21:40:24 2006 -0800
111142
111143 Destroying RandR crtc or output overwrites memory.
111144
111145 RRCrtcDestroyResource and RROutputDestroyResource had matching
111146 bugs that would overwrite memory past the end of the storage
111147 of the crtc or output arrays. Oops.
111148 (cherry picked from 4202b23ed86405a4cebfdcf239df1b023c1d10ca commit)
111149
111150commit 23ba72323af785516db6cbcf6c1b2fa907a8232f
111151Author: Keith Packard <keithp@ukulele.keithp.com>
111152Date: Thu Nov 30 23:16:42 2006 -0800
111153
111154 RandR ListOutputProperties has nAtoms element, not nProperties
111155
111156 Earlier RandR 1.2 encoding revisions used 8-bit nProperties field.
111157 Final RandR 1.2 spec uses 16-bit nAtoms field instead.
111158 (cherry picked from 66b6358a393972946f16394918db2401c51dc5ed commit)
111159
111160commit 416f1bb99a6d4557f2863ae941868e47e11bbb3b
111161Author: Peter Hutterer <peter@cs.unisa.edu.au>
111162Date: Fri Dec 1 15:56:52 2006 +1030
111163
111164 mpx: SelectEvents and GetEventBase requests.
111165 Some renaming and cleaning up in extinit.c
111166 MPXLastEvent added
111167
111168 Xi: ShouldFreeInputMask() from XI is not static any more, used in mpx
111169
111170 dix: GetPointerEvents() allocates MPX event for MPX devices.
111171 DeliverDeviceEvents() caters for MPX devices.
111172
111173commit ccd804c6c01cdfffe938fa5336be9b5668a6f0c0
111174Merge: 9423ac1 82912ad
111175Author: Drew Parsons <drew@pug.localdomain>
111176Date: Fri Dec 1 15:21:57 2006 +1100
111177
111178 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
111179
111180commit 9423ac134a2a9800787c4978d384827cc4e61fc8
111181Author: Drew Parsons <drew@pug.localdomain>
111182Date: Fri Dec 1 15:21:05 2006 +1100
111183
111184 GNU/FreeBSD support in GLX: include byteswap.h for FreeBSD systems
111185 defining __GLIBC__, as done already for Linux and Hurd (cf. bug #5613).
111186 Also includes some more __GNU__ checks on top of those made in
111187 commit ade4bf09076d13dbf3549c0a2d987a0afe76d5c1.
111188 Thanks Petr Salinger (Debian bug #400869).
111189
111190commit 82912ad7709e8cf4a5f8a9fa6b47f789842a3fe9
111191Author: Alan Coopersmith <alan.coopersmith@sun.com>
111192Date: Thu Nov 30 20:13:52 2006 -0800
111193
111194 Make solaris version of xf86OSRingBell return void like other OS'es
111195
111196commit 8956f63a941bf0a5f157d47b33e6221601a75040
111197Author: Alan Coopersmith <alan.coopersmith@sun.com>
111198Date: Thu Nov 30 19:53:29 2006 -0800
111199
111200 Tell automake to STFU about the *.O files
111201
111202 automake will not stop whining about the *.O files not being in normal library
111203 name format, so just tell automake they are PROGRAMS so it builds them without
111204 bitching.
111205
111206commit f3c60900e575e65254cd2576cc6c90b97c8f63ae
111207Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
111208Date: Wed Nov 29 22:19:57 2006 -0500
111209
111210 Add required root window context to config file.
111211
111212commit 143558500a89a5c887c75c03798bae68c953cac9
111213Merge: e3d3d29 ee9bdd3
111214Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
111215Date: Wed Nov 29 20:41:41 2006 -0500
111216
111217 Merge branch 'master' into my-XACE-SELINUX
111218
111219commit ee9bdd3f4a14a42fb0747acc420966f0f669129b
111220Author: Aaron Plattner <aplattner@nvidia.com>
111221Date: Wed Nov 29 12:01:43 2006 -0800
111222
111223 Add DIX_CFLAGS to util builds.
111224
111225 Fixes a build breakage when $(top_srcdir) != $(top_builddir) because
111226 -I$(top_srcdir)/include is missing for the cvt, ioport, pcitweak, and scanpci
111227 builds.
111228
111229 Signed-off-by: Adam Jackson <ajax@nwnk.net>
111230
111231commit a6381e69845f58d2b3282992b1f881015190f1bc
111232Author: Michel Dänzer <michel@tungstengraphics.com>
111233Date: Wed Nov 29 19:25:09 2006 +0100
111234
111235 xfree86 DDX: Delete DDX screens in ddxGiveUp().
111236
111237 This allows video drivers to clean up in the FreeScreen hook things they set up
111238 in the PreInit hook.
111239
111240commit afd6af65510b0147f16f08314045b3506bfa9d89
111241Author: Peter Hutterer <peter@cs.unisa.edu.au>
111242Date: Wed Nov 29 17:55:15 2006 +1030
111243
111244 mpx: Infrastructure for MPX extension, GetExtensionVersion request works.
111245
111246commit b0c8558b9d9a9984c0067960392e28f5a7622b29
111247Author: Keith Packard <keithp@neko.keithp.com>
111248Date: Sun Nov 26 19:31:48 2006 -0800
111249
111250 Ensure RandR resource types are registered before resources are created.
111251
111252 Now that resources can be created during server initialization, make sure
111253 the crtc, output and mode resource types are created before attempting to
111254 create associated resources.
111255 (cherry picked from commit ec83d674167e7045d5317b179c9998e3172a26dc)
111256
111257commit 6245e9dd4719c5dc15ff45d49cf626123794038b
111258Author: Keith Packard <keithp@neko.keithp.com>
111259Date: Tue Nov 21 16:52:28 2006 -0800
111260
111261 Allocate correct size for RRPropertyRec (oops).
111262
111263 Neglected to change the allocation size from sizeof (PropertyRec) to
111264 sizeof (RRPropertyRec). Lots of fun crashes this way.
111265 (cherry picked from commit 0626eb8e5c9fa05de6bdc9aa0c654f5148bf7cff)
111266
111267commit 24abce8032940e96bb2ccf9e463a7fff6f36283a
111268Author: Keith Packard <keithp@neko.keithp.com>
111269Date: Tue Nov 21 01:15:26 2006 -0800
111270
111271 Change RandR property datatype to include pending/valid values.
111272
111273 This patch tracks the protocol changes which introduce more complex
111274 semantics for RandR output properties including pending and valid value
111275 information.
111276 (cherry picked from commit af55c65bea40669fdc038aa34c6a1ec9ecb33e87)
111277
111278commit f62ac3ec39c6593df476985c630e499864c19c72
111279Author: Eric Anholt <eric@anholt.net>
111280Date: Tue Nov 28 10:31:40 2006 -0800
111281
111282 Separate DDC mode list creation from MonPtr creation.
111283
111284 This will be used by the intel driver, and likely other RandR 1.2 drivers.
111285
111286commit fbd09443385c533416fa530399d54f130afaf985
111287Author: Eric Anholt <eric@anholt.net>
111288Date: Tue Nov 28 10:15:51 2006 -0800
111289
111290 Replace bad mode name-setting code with xf86SetModeDefaultName().
111291
111292commit 4ad0bde661be2af4a17771d66066d49736e85cbe
111293Author: Eric Anholt <eric@anholt.net>
111294Date: Tue Nov 28 10:12:02 2006 -0800
111295
111296 Clean up a bunch of long lines and trailing whitespace.
111297
111298commit 05778432dc6e688bc0beff0c20ffd7e27b74888e
111299Author: Eric Anholt <eric@anholt.net>
111300Date: Tue Nov 28 10:07:57 2006 -0800
111301
111302 Move code to get a mode list from EDID data from ddcProperty.c to edid_modes.c.
111303
111304commit 38ecc66cd9c61346a46697bbf1d8319f4f6f9800
111305Author: Eric Anholt <eric@anholt.net>
111306Date: Tue Nov 28 10:06:15 2006 -0800
111307
111308 Typo that was missed in testing.
111309
111310commit 834e4b079866594b50be64ae79f3cb2a5baa2070
111311Author: Matthias Hopf <mhopf@suse.de>
111312Date: Tue Nov 28 18:57:13 2006 +0100
111313
111314 Fix potential NULL pointer access in timer code.
111315
111316 https://bugzilla.novell.com/show_bug.cgi?id=223718
111317
111318commit bdf718907b3202815a04ec28a868689708f43da9
111319Author: Peter Hutterer <peter@cs.unisa.edu.au>
111320Date: Tue Nov 28 18:06:34 2006 +1030
111321
111322 dix: missing parameter to SetCursorPosition in CheckMotion ifndef MPX
111323
111324 mi: missing include in misprite.c ifndef MPX
111325
111326commit e88bc0e55a6937ff0b47fe38f60bfa8991196578
111327Author: Peter Hutterer <peter@cs.unisa.edu.au>
111328Date: Tue Nov 28 14:34:36 2006 +1030
111329
111330 mi: Bugfix: FreeCursor ignores negative refcounts.
111331
111332 mieq: Bugfix: Motion notifies did overwrite DeviceMotionNotifies
111333
111334commit 5dbcd34a0a6c0d10dbfea8fdc9d7dfe7a0261b19
111335Author: Eric Anholt <eric@anholt.net>
111336Date: Mon Nov 27 16:26:14 2006 -0800
111337
111338 Register dependency on new RandR protocol.
111339
111340commit 16f8f10dc2106bc6253b2d89a1f8efee8d80e2ba
111341Author: Eric Anholt <eric@anholt.net>
111342Date: Mon Nov 27 16:21:31 2006 -0800
111343
111344 Move mode handling helpers from ddcProperty.c to xf86Mode.c.
111345
111346commit b4b0d901d98371a8aa7b17d195e18e83e2a6a618
111347Merge: 64de3ba d6cd031
111348Author: Eric Anholt <eric@anholt.net>
111349Date: Mon Nov 27 15:43:15 2006 -0800
111350
111351 Merge branch 'randr-1.2'
111352
111353 Conflicts:
111354
111355 dix/events.c
111356 dix/getevents.c
111357 hw/xfree86/common/xf86Mode.c
111358 hw/xfree86/dri/Makefile.am
111359 hw/xfree86/os-support/drm/xf86drm.c
111360 hw/xfree86/os-support/xf86drm.h
111361
111362commit d6cd0313c7f23f32c9c7dda00ff739e772bf7db3
111363Author: Eric Anholt <eric@anholt.net>
111364Date: Mon Nov 27 14:46:50 2006 -0800
111365
111366 Add some mode helper functions from the intel driver.
111367
111368 This also removes static from some other functions that had been copied out
111369 to at least the intel driver, but perhaps others that were doing mode list
111370 handling.
111371
111372commit 85ea408e38f452bcf9109f14fd1016d159692a99
111373Author: Peter Hutterer <peter@cs.unisa.edu.au>
111374Date: Mon Nov 27 18:07:16 2006 +1030
111375
111376 mi: malloc checks for misprite and midispcur, code cleanup
111377 removed IsMPDev()
111378
111379commit 023da4e804ad260eaf0c50d2ec7213dd514e2113
111380Author: Peter Hutterer <peter@cs.unisa.edu.au>
111381Date: Mon Nov 27 15:54:33 2006 +1030
111382
111383 mi: mipointer.c cleanup, reducing code duplication, using MPHasCursor()
111384 instead of IsMPDev()
111385
111386commit a2340a73687dfd93affb6de28a33d73fd808c96b
111387Author: Peter Hutterer <peter@cs.unisa.edu.au>
111388Date: Mon Nov 27 14:47:23 2006 +1030
111389
111390 mi: free allocated memory in miDCCloseScreen and in miDCInitialize in case
111391 of errors.
111392 Some copyright notices.
111393 misprite.c passes DeviceIntPtr around and down to DC (previously
111394 miCursorInfoPtr and the plain device id).
111395 Large cleanup in misprite.c to avoid code duplication.
111396
111397commit 1c7568b8a1417257fa67c7fca69aa253099b9461
111398Author: Peter Hutterer <peter@cs.unisa.edu.au>
111399Date: Thu Nov 23 17:15:14 2006 +1030
111400
111401 mi: closing memory leak, miPointer is freed in miPointerCloseScreen
111402 bugfix: uninitialized pPointer in miPointerGetPosition ifndef MPX
111403 adding DeviceIntPtr parameter to ScreenRec's cursor functions.
111404 cleanup of miPointer code to use same scheme in each function
111405
111406 dix: MPHasCursor() function determines checking whether to invoke
111407 cursor rendering.
111408
111409 animcur: adding DeviceIntPtr parameter to cursor functions but animcur relies
111410 on the core pointer right now.
111411
111412 xfixes: adding DeviceIntPtr parameter to cursor functions but xfixes relies on
111413 the core pointer right now.
111414
111415 rac: adding DeviceIntPtr parameter to cursor functions but RAC relies on
111416 the core pointer right now.
111417
111418 ramdac: adding DeviceIntPtr parameter to cursor functions but ramdac relies on
111419 the core pointer right now.
111420
111421commit 64de3baf85f6df274f71f736016f0848567cd9f6
111422Author: Aaron Plattner <aplattner@nvidia.com>
111423Date: Wed Nov 22 14:46:51 2006 -0800
111424
111425 Add a -showDefaultLibPath option.
111426
111427 A corollary to the previous change, this option prints $libdir.
111428
111429commit 0a2a6e4070718b90af7ca0e047f028e0cabdfb9d
111430Author: Aaron Plattner <aplattner@nvidia.com>
111431Date: Fri Nov 17 18:27:23 2006 -0800
111432
111433 Add a -showDefaultModulePath option.
111434
111435 As discussed on the mailing list, people would rather have an X command-line
111436 option to print the module path so installers can know where to put modules,
111437 rather than the installers using `pkg-config --variable=moduledir xorg-server`,
111438 since some distros choose not to install xorg-server.pc.
111439
111440commit 61832cb94c8a4d62cddb92188caeed86519e9d62
111441Author: Brian <brian@i915.localnet.net>
111442Date: Wed Nov 22 09:16:43 2006 -0700
111443
111444 Regenerated from Mesa w/ latest gl_API.xml file
111445
111446commit 61863f09d22935406371e92bb75173d55ff9b29f
111447Author: Brian <brian@i915.localnet.net>
111448Date: Wed Nov 22 09:16:17 2006 -0700
111449
111450 Regenerated from Mesa, fixes glMap* protocol problem (bug 8899)
111451
111452commit f52d53e060d0468b658a565688d1f92c156405c7
111453Author: Peter Hutterer <peter@cs.unisa.edu.au>
111454Date: Wed Nov 22 17:58:57 2006 +1030
111455
111456 dix: DeliverEventsToWindow adjusted for MPX
111457
111458 mi: sprite debugging disabled
111459
111460commit 8c16f21d5df08b028e93440d632939a5fdbf5876
111461Author: Peter Hutterer <peter@cs.unisa.edu.au>
111462Date: Wed Nov 22 17:06:47 2006 +1030
111463
111464 dix GetSpriteWindow() adjusted for MPX
111465 added device dependency for MPX to DoEnterLeaveEvents,
111466 EnterLeaveEvents, EnterNotifies, LeaveNotifies and
111467 FixUpEventFromWindow.
111468
111469commit 70383105de279df553874efa56b37a7b3e426ea1
111470Author: Peter Hutterer <peter@cs.unisa.edu.au>
111471Date: Wed Nov 22 15:27:16 2006 +1030
111472
111473 mi: added DevToSprite() conversion function
111474 fixed miSpriteRealize to use passed pDev argument instead of looping.
111475
111476 dix: changed sprite dependency and added MPX functionality to parts of
111477 events.c (XineramaChangeToCursor, XineramaCheckPhysLimits,
111478 XineramaConstrainCursor)
111479
111480 Xi: fix: accessing other->buttons for keyboard segfaulted the server
111481
111482commit 33ff4cec5ff7533ec725f71d357c096dfb0acb79
111483Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
111484Date: Tue Nov 21 21:23:17 2006 +0000
111485
111486 Issue CloseDownDevices() in os/log.c and remove from dix/main.c.
111487
111488 This ensures that all calls to FatalError() will shutdown the input devices.
111489
111490commit efd4f3c6ffec804c68ba5df17cc117da264fb7c4
111491Author: Peter Hutterer <peter@cs.unisa.edu.au>
111492Date: Tue Nov 21 20:31:30 2006 +1030
111493
111494 dix: WindowsRestructured() calls CheckMotion() on all devices
111495
111496 mi: core pointer was checked twice in miSpriteReportDamage,
111497 miSpriteInstallColormap, miSpriteStoreColors, miSpriteSaveDoomedAreas
111498 and miSpriteRealiseCursor
111499 using damage bug (see comment in file) to avoid artefacts
111500
111501commit 9db851c22d4befb95bfb074b96620261d8e32ac9
111502Author: Peter Hutterer <peter@cs.unisa.edu.au>
111503Date: Tue Nov 21 18:15:04 2006 +1030
111504
111505 dix: moved sprite from static to be pointer of type SpritePtr
111506 added SpriteRecs for MPX devices
111507 changed sprite dependency and added MPX functionality to parts
111508 of events.c (ConfineToShape, PostNewCursor, XineramaCheckMotion,
111509 CheckMotion, XineramaChangeToCursor, ChangeToCursor, CheckPhysLimits,
111510 PointerConfinedToScreen)
111511 added DeviceIntRec param to GetSpritePosition(). This required some
111512 minor changes in ddx, xtest, xkb and xfixes.
111513
111514 mi: changed miPointer to pointer instead of static struct.
111515
111516commit fd8b9a6eaee28a453a00af14195b81ef1890a577
111517Author: Peter Hutterer <peter@cs.unisa.edu.au>
111518Date: Tue Nov 21 15:24:43 2006 +1030
111519
111520 TAG: MPX_BEFORE_DIX_API_BREAK
111521
111522commit 447efdc01d4f8922205ba26e52c9e5ff9f5cb90e
111523Author: Peter Hutterer <peter@cs.unisa.edu.au>
111524Date: Tue Nov 21 14:46:00 2006 +1030
111525
111526 mi: constrain MPX cursors to screen
111527
111528commit 1107d8bea22eea26f7b666f1c00a53360b7d237e
111529Author: Peter Hutterer <peter@cs.unisa.edu.au>
111530Date: Tue Nov 21 14:00:46 2006 +1030
111531
111532 mi: added device ID to miSpriteCursorFuncRec functions
111533 added MPX code to midispcur.c
111534
111535 BUG: Core pointer changes whenever MPX cursor changes shape.
111536
111537commit 3bad452d12d620df91b1f37d06c650ded96bfafe
111538Author: Peter Hutterer <peter@cs.unisa.edu.au>
111539Date: Tue Nov 21 11:26:21 2006 +1030
111540
111541 mi: added MPX to miSpriteReportDamage
111542 added id field to miCursorInfoPtr, required to pass through to miDC
111543 core pointer uses mpCursors array as well.
111544 added miDCBufferRec for future use with MPX
111545
111546 TAG: MPX_BEFORE_MIDC_API_BREAK
111547
111548commit b1bbdf464d2f8799fa3429c1c27ca998c5f3e94a
111549Author: Peter Hutterer <peter@cs.unisa.edu.au>
111550Date: Mon Nov 20 17:58:30 2006 +1030
111551
111552 mi: added MPX to miSprite functions.
111553
111554 NOTE: This version will render multiple cursors but with an artefact each time
111555 when devices alternate.
111556
111557commit 7b4b619d4c268804199f9f4775eb56604f0b80e8
111558Author: Peter Hutterer <peter@cs.unisa.edu.au>
111559Date: Mon Nov 20 13:10:07 2006 +1030
111560
111561 mi: moved core pointer information from miSpriteScreenRec into a
111562 miCursorInfoRec.
111563 bugfix in mipointer, semicolon after preprocessor statement
111564
111565commit ca2874b273232d9f51881b1cd754ed6847bfaf47
111566Merge: c9a5f9d e2f6dac
111567Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
111568Date: Sat Nov 18 19:56:32 2006 +0000
111569
111570 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
111571
111572commit c9a5f9d3911c2e6a8f81b0721c9eb914fb7719b5
111573Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
111574Date: Sat Nov 18 19:55:42 2006 +0000
111575
111576 Shutdown input devices if FatalError occurs during startup.
111577 Fixes Xdmx problems when the input device has been initialized
111578 and the keyboard map has been destroyed.
111579
111580commit e2f6dacc736527790ed8e304698678afc17f71c6
111581Author: Dave Airlie <airlied@linux.ie>
111582Date: Sat Nov 18 11:26:55 2006 +1100
111583
111584 dri: setup libdrm hooks as early as possible.
111585
111586 This is due to radeon doing drm stuff before DRI extension gets initialised
111587
111588commit 90bea69dfdd63f58ee326887359ad1b35a31cb5f
111589Author: Peter Hutterer <peter@cs.unisa.edu.au>
111590Date: Fri Nov 17 18:03:43 2006 +1030
111591
111592 mi:
111593 added miCursorInfoRec to contain info of the MPX cursors.
111594 calling miUpdatePointerSprite() from event queue for MPX devices.
111595 adding device-specific processing to miPointer*() functions.
111596
111597 dix: Call to SetCursorPosition in CheckMotion() temporarily disabled.
111598
111599 xfree86/common: call to miPointerUpdateSprite() disabled, is done from the EQ
111600
111601 NOTE: This build will not display cursor images.
111602 BUG: The second mouse does to take correct x coordinates.
111603
111604commit 94e106a898c44daeb228bcb031b85f996ddc96b0
111605Author: Peter Hutterer <peter@cs.unisa.edu.au>
111606Date: Fri Nov 17 15:10:37 2006 +1030
111607
111608 dix: moved isMPdev field to end of _DeviceIntRec structure
111609
111610 mi:
111611 added miMPPointers array to mipointer.c
111612 added DeviceIntPtr to all miPointerSpriteFuncs. Coming from miPointer
111613 we use inputInfo.pointer as standard value. ABI BREAK!
111614
111615 ramdac:
111616 forcing failed HW Cursor initialisation. MPX needs software rendering.
111617 changes to use new miPointerSpriteFunc (this required externing
111618 inputInfo, should probably be fixed at a later point).
111619
111620 RAC: changes to use new miPointerSpriteFuncs.
111621
111622commit ef47d9c3ba63e9a6243fe5c81ccc60c8246352b4
111623Author: Keith Packard <keithp@neko.keithp.com>
111624Date: Thu Nov 16 13:50:48 2006 -0800
111625
111626 Reduce calls to RRGetInfo.
111627
111628 RRGetInfo can be expensive. Don't invoke it when quering Xinerama
111629 information or setting a new CRTC configuration.
111630 (cherry picked from commit b5aa9eb8e6eda36856a075f4b008c33f6c706bad)
111631
111632commit 07b26e690cd9a4fc626132feed0702515cbe5a88
111633Author: Keith Packard <keithp@neko.keithp.com>
111634Date: Thu Nov 16 09:48:33 2006 -0800
111635
111636 Remove RandR output options.
111637
111638 RandR output options are now expected to be handled by properties instead.
111639 (cherry picked from commit 8b2a7e94a1dc2776ab2cfaaebb309be02502602a)
111640
111641commit f17e3c34dfd1f1418440bdebf45764e4dbf550f0
111642Author: George Sapountzis <gsap7@yahoo.gr>
111643Date: Thu Nov 16 02:18:03 2006 +0200
111644
111645 Fix GL context destruction with AIGLX.
111646
111647 The logic for freeing GL contexts introduced by "Fix AIGLX VT switching." is
111648 inverted. As it is now, GL context destruction is deferred for glxDRIEnterVT().
111649
111650commit ae3c9ad4abe66784d7ee474455003d2745699286
111651Author: Bjorn Helgaas <bjorn.helgaas@hp.com>
111652Date: Thu Nov 16 17:29:06 2006 +0100
111653
111654 Bug 9041: Check the return code in xf86MapDomainMemory().
111655
111656commit 5388423eb05daefcc71067b6ab96b6e57c44ef5c
111657Author: Peter Hutterer <peter@cs.unisa.edu.au>
111658Date: Thu Nov 16 20:43:27 2006 +1030
111659
111660 mieq: EQ processing handles MP devices
111661
111662 global: MPX define added to xorg-server.h.in
111663
111664 xfree86/common: small fix to avoid byte overflow
111665
111666commit c957a16180810fbd58526e1a670d82384fb4a61d
111667Author: Peter Hutterer <peter@cs.unisa.edu.au>
111668Date: Thu Nov 16 15:11:14 2006 +1030
111669
111670 dix: added isMPdev field to _DeviceIntRec, is used in GetPointerEvents()
111671
111672 mieq: avoid merging events from different devices in mieqEnqueue()
111673
111674 xfree86/common
111675 isMPdev field used from xf86ActivateDevice(), xf86PostMotionEvent()
111676 and xf86PostButtonEvent()
111677
111678commit 7ce2dc57844872cc340b60387d4d1cb3d982633e
111679Author: Peter Hutterer <peter@cs.unisa.edu.au>
111680Date: Thu Nov 16 12:51:08 2006 +1030
111681
111682 dix:
111683 merge with code cleanup from master
111684 GetPointerEvents treats events in the same way as XINPUT devices when flag
111685 has POINTER_MULTIPOINTER set.
111686
111687 xfree86/common:
111688 added XI86_MP_DEVICE flag and parsing in xf86ProcessCommonOptions
111689 added POINTER_MULTIPOINTER define. Is used in xf86PostMotionEvent and
111690 xf86PostButtonEvent for the flags that are passed into GetPointerEvents()
111691
111692 global:
111693 added flags to configure.ac to enable/disable MPX define
111694 added flags to dix-config.h.in to define MPX
111695
111696commit 4d07b503727ba86315b90d6f45b06ed6185724b2
111697Author: Peter Hutterer <peter@cs.unisa.edu.au>
111698Date: Thu Nov 16 10:55:15 2006 +1030
111699
111700 dix: cleaned up GetPointerEvents() to avoid code duplication.
111701
111702commit 6ff7f2ad6a5e2e769244590578e6809974b5235d
111703Author: Eric Anholt <eric@anholt.net>
111704Date: Tue Nov 7 13:13:53 2006 -0800
111705
111706 Fix build on FreeBSD after input-hotplug.
111707 (cherry picked from commit 4e6e4baead6c565363abbcd9e06cc685be121596)
111708
111709commit bffb1f9b67c8d3d264c0c3fd703cf7e57dcdf7b7
111710Author: Peter Hutterer <peter@cs.unisa.edu.au>
111711Date: Wed Nov 15 17:53:13 2006 +1030
111712
111713 removed unused variables drag and local from xf86PostMotionEvent
111714
111715commit f737cd925efe452787443b039ae65b8d85081bb3
111716Author: Peter Hutterer <peter@cs.unisa.edu.au>
111717Date: Wed Nov 15 17:50:02 2006 +1030
111718
111719 o fix minor error in comment for GetPointerEvents()
111720
111721commit 2eab230d9bd3f73ffe1b5a42111f89e85904ee11
111722Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
111723Date: Tue Nov 14 16:37:18 2006 -0600
111724
111725 For NetBSD, define PCVT_SUPPORT (System has PCVT console).
111726
111727 Noticed by Joerg Sonnenberger. This fixes problem with console
111728 switching.
111729
111730 This was in original imake NetBSD.cf.
111731
111732commit 26d2e45bdb0cf4d18ba7b0365425da49d60b3d5c
111733Author: Matthias Hopf <mhopf@suse.de>
111734Date: Tue Nov 14 15:33:07 2006 +0100
111735
111736 Bug #9023: Only check mice for "mouse" or "void" if identifier is != NULL.
111737
111738commit a724b7f1302ba7a59f140b521f13d2ddf0fcf9bf
111739Merge: 6facd95 f80a8ae
111740Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
111741Date: Mon Nov 13 20:32:26 2006 -0600
111742
111743 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
111744
111745commit 6facd958c2e7098cb68eac2810a88b8cd40f19c0
111746Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
111747Date: Mon Nov 13 20:23:06 2006 -0600
111748
111749 On DragonFLy, default to /dev/sysmouse (just like on FreeBSD).
111750
111751 TODO: use autoconf to handle this so the code is not platform
111752 based but feature based.
111753
111754commit f80a8ae6e6820378ea70ac0849cee8bf3055e0e8
111755Author: Adam Jackson <ajax@benzedrine.nwnk.net>
111756Date: Mon Nov 13 18:26:05 2006 -0500
111757
111758 Disable Xprt server build by default.
111759
111760commit 5bfe7f4dfc6ab51b6790d74faf11574385234423
111761Author: George Sapountzis <gsap7@yahoo.gr>
111762Date: Mon Nov 13 01:54:32 2006 +0200
111763
111764 DRI: call drmSetServerInfo() before drmOpen().
111765
111766 Also, remove some libdrm remnants.
111767
111768commit acb5ff4c73ac4d52201d7c421f488e2ead5c8b9c
111769Author: Jurij Smakov <jurij@wooyd.org>
111770Date: Sat Nov 11 14:09:15 2006 +0200
111771
111772 ffs: handle 0 argument (bug #8968)
111773
111774 Handle an argument of 0 in ffs(), instead of looping indefinitely.
111775 Add an ffs prototype to dix.h, and add includes to ffs.c.
111776 (cherry picked from 34164e551e4c3909322d50b09835ca4ac1d49d68 commit)
111777
111778commit ca094684196886a4a1c10273049fae0705a3edc2
111779Author: Alan Coopersmith <alan.coopersmith@sun.com>
111780Date: Fri Nov 10 18:02:05 2006 -0800
111781
111782 Correct symlink-mesa.sh usage message
111783
111784commit 28337cc1060bc08adef81c304dd9ed02d0a0e37b
111785Author: Eric Anholt <eric@anholt.net>
111786Date: Thu Nov 9 19:51:17 2006 -0800
111787
111788 Fix typo before the last commit.
111789
111790commit e1720b1089328af80ca0cb85e5289ced05263f95
111791Author: Eric Anholt <eric@anholt.net>
111792Date: Thu Nov 9 18:53:37 2006 -0800
111793
111794 Bug #8868: Remove drm from SUBDIRS now that the directory is gone.
111795
111796commit 0dee48b8af3e054228aef0d15c1cb1c9e23790cc
111797Author: Keith Packard <keithp@mandolin.keithp.com>
111798Date: Wed Nov 8 23:17:55 2006 -0800
111799
111800 Add RRInit function to create resource types for RR objects.
111801
111802 To allow RandR objects to be created before the screen object exists,
111803 the resource types must be registered with the resource database.
111804 A driver wishing to create RandR objects must call RRInit before doing so.
111805
111806 Also, fix a segfault when setting Output data before it is associated with a
111807 screen.
111808
111809commit ec77a95a02329a2ee3a94d7de9d2a234aecb9ca0
111810Author: Keith Packard <keithp@mandolin.keithp.com>
111811Date: Wed Nov 8 21:36:35 2006 -0800
111812
111813 Allow RandR objects to be created before the associated ScreenRec.
111814
111815 xf86 drivers need to create RandR object in the PreInit stage,
111816 before the ScreenRec is allocated. Changing the RandR DIX code
111817 to permit this required the addition of functions that later associate the
111818 objects with the related screen.
111819
111820 An additional change is that modes are now global, and no longer associated
111821 with a specific screen. This change actually makes mode management cleaner
111822 as there is no more per-screen list of modes to deal with.
111823
111824 This changes the RandR 1.2 ABI/API for drivers.
111825
111826commit fd91630b73100e9d77ccb492c52807448bc772a5
111827Author: Dave Airlie <airlied@linux.ie>
111828Date: Thu Nov 9 09:30:33 2006 +1100
111829
111830 make X server use system libdrm - this requires libdrm >= 2.3.0
111831
111832 This patch blacklists the load "drm" line, moves some functions in dri module
111833 links dri module against libdrm, and removes the X copy of libdrm
111834
111835commit 0409e1627a167db2efc1355f292d3c02a6989ffc
111836Author: Adam Jackson <ajax@benzedrine.nwnk.net>
111837Date: Wed Nov 8 16:17:20 2006 -0500
111838
111839 'make dist' fixes.
111840
111841commit 7e0aeebb8f8a5dff3cb4d88756e535dd70edeec4
111842Author: Adam Jackson <ajax@benzedrine.nwnk.net>
111843Date: Wed Nov 8 12:03:37 2006 -0500
111844
111845 Bug #6786: Use separate defines for server's Fixes support level.
111846
111847commit b5d09d4adb8088719ff494a4281a793717046576
111848Author: Rich Coe <Richard.Coe@med.ge.com>
111849Date: Wed Nov 8 18:10:14 2006 +0200
111850
111851 CheckConnections: don't close down the server client (bug #7876)
111852 When an appgroup is shutting down, the list of clients can change, so make
111853 sure we're not trying to shut the server down.
111854
111855commit 0567a6337b84fa045b5732e98203f488274aa2a2
111856Author: Bram Verweij <amverweij@gmail.com>
111857Date: Wed Nov 8 18:00:52 2006 +0200
111858
111859 xfree86/linux acpi: fix tokenising
111860 Split on a space, rather on the 'video' string, as strtok takes a char,
111861 not a string.
111862
111863commit e7900d68c3ee657158813f0650886d680c0a9a3c
111864Author: Daniel Stone <daniel@fooishbar.org>
111865Date: Wed Nov 8 15:36:47 2006 +0200
111866
111867 remove trailing whitespace
111868 Whitespace police in full effect.
111869
111870commit 2035f115b7db3c4b7deabeab0d814b3107d6ef30
111871Author: Daniel Stone <daniel@fooishbar.org>
111872Date: Wed Nov 8 15:28:10 2006 +0200
111873
111874 xfree86: remove xf86Keymap.h
111875
111876commit 809e2841aaa54662a42498aacec558bc813bba1c
111877Author: Daniel Stone <daniel@fooishbar.org>
111878Date: Wed Nov 8 15:27:58 2006 +0200
111879
111880 xfree86: add general handler, port ACPI to it (bug #5665)
111881
111882 Add 'general' handler registration, which will not trigger DPMS when an
111883 event comes in.
111884 Make ACPI use this.
111885
111886commit b5438f7fb2879e0226b578f60b22a490e73c3a83
111887Author: Daniel Stone <daniel@fooishbar.org>
111888Date: Wed Nov 8 15:24:58 2006 +0200
111889
111890 Get*Events: massive reorganisation
111891
111892 Reorganise the code logically, and put more comments in.
111893 Clip valuators in proximity calls.
111894
111895commit b0a7443ca8bd224994f98c1c1e25de88f2573224
111896Author: Daniel Stone <daniel@fooishbar.org>
111897Date: Wed Nov 8 15:23:52 2006 +0200
111898
111899 configure.ac: axe redundant SDL test
111900
111901commit 332f179db7c38310db1f752d45f51b6d8301fd30
111902Author: Daniel Stone <daniel@fooishbar.org>
111903Date: Wed Nov 8 15:23:35 2006 +0200
111904
111905 kdrive/sdl: fix compilation
111906
111907commit 988757a44197c91027109076e2e33ff6510ed71d
111908Author: Daniel Stone <daniel@fooishbar.org>
111909Date: Wed Nov 8 14:06:51 2006 +0200
111910
111911 config: error messages non-fatal, and before we free them
111912
111913 Move error messages before we free the error structure, and make them all
111914 non-fatal.
111915
111916commit e2b78df5800fb4e3f0ed01b38a1974ba3577949c
111917Author: Daniel Stone <daniel@fooishbar.org>
111918Date: Tue Nov 7 11:13:56 2006 +0200
111919
111920 xkb: warning fix
111921
111922commit b55007d8cc9d20baa23d5de67683e414c827d3e5
111923Author: Daniel Stone <daniel@fooishbar.org>
111924Date: Tue Nov 7 11:13:32 2006 +0200
111925
111926 SyntheticMotion: don't dereference sprite.screen when not using Xinerama
111927
111928commit f93d10ce9bb4a6de83b561f44fb7b046def16234
111929Author: Daniel Stone <daniel@fooishbar.org>
111930Date: Mon Nov 6 18:33:45 2006 +0200
111931
111932 dix: remove staggeringly broken vendor workarounds
111933
111934 Dear SGI,
111935 No.
111936
111937 Scant regards,
111938 Daniel
111939
111940commit 389275d240e4ba19d62fda0f138a45c7ecb245ff
111941Author: Daniel Stone <daniel@fooishbar.org>
111942Date: Sun Nov 5 02:47:59 2006 +0200
111943
111944 XkbCopyKeymap: don't iterate broken types, or dereference null pointers
111945
111946 Don't iterate invalid destination types (>= num_types) when coping key
111947 types.
111948 Don't free key_aliases if it's NULL (theoretical, but sure).
111949 Make sure dst's label_font gets allocated if it's NULL.
111950 (Thanks, Chris Lee.)
111951
111952commit d585b4189aff8d7952847f75b19b4f092ab3b88b
111953Author: Daniel Stone <daniel@fooishbar.org>
111954Date: Sat Nov 4 21:47:55 2006 +0200
111955
111956 xkb: fix uninitialised warning
111957
111958 Fix uninitialised warning with memset(); we never actually use it
111959 uninitialised, but gcc doesn't know that.
111960
111961commit b6d7b537ed8975363ad0f7c4180a62822358e418
111962Author: Daniel Stone <daniel@fooishbar.org>
111963Date: Sat Nov 4 21:43:22 2006 +0200
111964
111965 os: fix sun extensions test
111966
111967 'else if' is not very valid, plus the logic is kind of broken, so just
111968 move it outside the ifdef in the first place.
111969
111970commit 8ba0c7b62c78dead722b0c8aa414f37bac4414b7
111971Author: Daniel Stone <daniel@fooishbar.org>
111972Date: Sat Nov 4 21:41:03 2006 +0200
111973
111974 xace: avoid 'unused variable pScreen'
111975
111976 Initialise pScreen explicitly, as REGION_* macros ignore pScreen.
111977
111978commit c3ea1f7db494365032526dc06a7283384bd0ecd1
111979Author: Daniel Stone <daniel@fooishbar.org>
111980Date: Sat Nov 4 21:38:31 2006 +0200
111981
111982 dix/mi: still more warning fixes
111983
111984 Fix up prototypes for PrintChildren and PrintWindowTree in the dix.
111985 Make miPrintRegion be unconditionally defined, and move the prototype into
111986 regionstr.h.
111987 Change a bunch of ScreenPtr pScreen = foo; to
111988 ScreenPtr pScreen; pScreen = foo; in window.c, so we avoid unused variable
111989 references (as inline REGION_* doesn't reference pScreen).
111990
111991commit 3a9b96425851b495503bd2eb0fd0d01c08f6a097
111992Author: Daniel Stone <daniel@fooishbar.org>
111993Date: Sat Nov 4 21:33:09 2006 +0200
111994
111995 dix: add missing prototypes
111996
111997 Add missing prototype for ffs, and include headers from ffs.c.
111998 Move PostSyntheticMotion prototype to input.h.
111999
112000commit 6716488fa256798070017232405b107d5c985479
112001Author: Daniel Stone <daniel@fooishbar.org>
112002Date: Sat Nov 4 21:30:23 2006 +0200
112003
112004 dix: remove unused debug code
112005
112006commit 51813d77bfb84609a58a98e678efe9b6c0bf5503
112007Author: Daniel Stone <daniel@fooishbar.org>
112008Date: Sat Nov 4 21:29:05 2006 +0200
112009
112010 dix: remove unused variable
112011
112012commit 844090a5b557705dd0adce2b7ed98813b5104d85
112013Author: Daniel Stone <daniel@fooishbar.org>
112014Date: Sat Nov 4 21:21:29 2006 +0200
112015
112016 xfree86/xf86misc: warning fixes
112017
112018commit 61b570d0c1eb448f0aa08b4598118f0d43bc7345
112019Author: Daniel Stone <daniel@fooishbar.org>
112020Date: Sat Nov 4 21:21:09 2006 +0200
112021
112022 xfree86/os-support: update prototype for OSRingBell
112023
112024commit c51fadc07d938f6a3edfd5620170fcb7d6486a11
112025Author: Daniel Stone <daniel@fooishbar.org>
112026Date: Sat Nov 4 21:20:45 2006 +0200
112027
112028 mieq: annotate with some more comments
112029
112030commit 578899139f133746634a7bf8845e25362b5dfca2
112031Author: Daniel Stone <daniel@fooishbar.org>
112032Date: Sat Nov 4 20:35:55 2006 +0200
112033
112034 RemoveGeneralSocket: don't touch EnabledDevices
112035
112036 RemoveGeneralSocket. Harmless, but.
112037
112038commit ae58d349c1cf5d63ad3616c485baa858350978d5
112039Author: Laurence Withers <l@lwithers.me.uk>
112040Date: Sat Nov 4 19:34:37 2006 +0200
112041
112042 CreateColormap: fix return value (bug #7083)
112043
112044 Return BadMatch when an unsupported visual type is given, not BadValue --
112045 this is correct according to the spec.
112046
112047commit 96f78e3886791b723ccd9ba40bea701603537b0c
112048Author: Erik Andren <erik.andren@gmail.com>
112049Date: Sat Nov 4 19:29:49 2006 +0200
112050
112051 remove XFree86 changelogs (bug #7262)
112052
112053 Without being able to tie these to specific commits, the text changelog is
112054 useless, as well as being huge.
112055
112056commit 5a40448f2d0ac2c86c617bebe3fb649174bf0d7f
112057Author: Eric Anholt <eric@anholt.net>
112058Date: Tue Nov 7 15:48:05 2006 -0800
112059
112060 A couple more cases of error message before freeing strings.
112061
112062commit 05f1c302460a14c8fa9a943a12d69adcd3c30d58
112063Merge: 3e7e0e3 46af6d1
112064Author: Adam Jackson <ajax@benzedrine.nwnk.net>
112065Date: Tue Nov 7 18:42:54 2006 -0500
112066
112067 Merge branch 'autoconfig-for-7.2'
112068
112069commit 3e7e0e35094d09e0e764818ed125314be75be01a
112070Author: Eric Anholt <eric@anholt.net>
112071Date: Tue Nov 7 14:13:23 2006 -0800
112072
112073 Report the error before freeing the error strings.
112074
112075commit 4e6e4baead6c565363abbcd9e06cc685be121596
112076Author: Eric Anholt <eric@anholt.net>
112077Date: Tue Nov 7 13:13:53 2006 -0800
112078
112079 Fix build on FreeBSD after input-hotplug.
112080
112081commit 20e9144c0746943624ff77a61791b8596f3f8458
112082Author: Keith Packard <keithp@neko.keithp.com>
112083Date: Tue Nov 7 12:49:28 2006 -0800
112084
112085 Add $(DIX_CFLAGS) to remaining Makefile.am files
112086
112087commit 5e946dd853a4ebc2722ae023429ce5797de3d7a6
112088Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
112089Date: Tue Nov 7 13:50:19 2006 -0500
112090
112091 Bug #8937: Extension setup functions not called on server resets
112092
112093commit 1dcda4f3c56214464c0b6123fea6daa69aae69fc
112094Author: Keith Packard <keithp@neko.keithp.com>
112095Date: Tue Nov 7 01:29:51 2006 -0800
112096
112097 Avoid dereferencing sprite.screen when Xinerama is not running. (#8925)
112098
112099 With Xinerama support built into the X server but not in use,
112100 sprite.screen is NULL and yet the SyntheticMotion
112101 macro would dereference it. Avoid that by just passing sprite.screen
112102 to PostSyntheticMotion which can then dereference it when Xinerama is
112103 enabled.
112104
112105 Also, define PostSyntheticMotion in dixevents.h and include dixevents.h in
112106 getevents.c
112107
112108commit c20d3bf7533da0bf26beaf7d8c359d18edbd70e8
112109Merge: 028bbdc 3d39c02
112110Author: Keith Packard <keithp@neko.keithp.com>
112111Date: Tue Nov 7 01:21:28 2006 -0800
112112
112113 Merge branch 'origin' into randr-1.2
112114
112115commit 3d39c02fe6aaa602c52f1d4f0ea6cd3bd000cf9f
112116Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
112117Date: Mon Nov 6 21:25:52 2006 -0500
112118
112119 More work on Bug #8875: revert previous fix and try using client argument
112120 instead of serverClient. Also don't use totalClientSize as it is not
112121 initialized until after the first call to InitClient.
112122
112123commit 75fe0670eb1f71144246f1c20759d58788bbee00
112124Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
112125Date: Mon Nov 6 15:30:25 2006 -0500
112126
112127 whitespace adjust
112128
112129commit 0539d9cf2423fc0bed6f5c413beba3080f8abd85
112130Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
112131Date: Mon Nov 6 15:29:17 2006 -0500
112132
112133 Bug #8875: Security extension causes Xorg to core dump on server reset
112134
112135commit aa0261a98e9d5b1349b33e2639bd83c556dd4000
112136Author: Juliusz Chroboczek <jch@pps.jussieu.fr>
112137Date: Mon Nov 6 02:29:49 2006 +0100
112138
112139 Improve vm86 error handling in Xvesa.
112140
112141commit 6b2c65fdd169037c6ede250d4a8fec3d29a080ae
112142Author: Juliusz Chroboczek <jch@pps.jussieu.fr>
112143Date: Mon Nov 6 00:30:09 2006 +0100
112144
112145 Fix typo in Xvesa: incorrect reporting of DAC capabilities.
112146
112147commit 028bbdc0417173803695808ba9f48498519273a3
112148Merge: 50504c6 8deaaa3
112149Author: Keith Packard <keithp@neko.keithp.com>
112150Date: Sat Nov 4 17:46:26 2006 -0800
112151
112152 Merge master back to randr-1.2
112153
112154commit 50504c68e1d407232cf83465981b235e542ef31f
112155Merge: 8b87ce1 cde8806
112156Author: Keith Packard <keithp@neko.keithp.com>
112157Date: Sat Nov 4 17:43:19 2006 -0800
112158
112159 Merge branch 'randr-1.2-origin' into randr-1.2
112160
112161commit 8b87ce19741753eafbd99e7093bc3dea8f26e838
112162Author: Keith Packard <keithp@neko.keithp.com>
112163Date: Sat Nov 4 17:41:25 2006 -0800
112164
112165 Allow X server to build against libdrm 2.1
112166
112167commit 7ffbe9d232e3a4621a204448d67e434736465cbe
112168Author: Keith Packard <keithp@neko.keithp.com>
112169Date: Sat Nov 4 17:41:09 2006 -0800
112170
112171 Add DIX_CFLAGS to hw/vfb/Makefile.am
112172
112173commit 2db62bce0725ba2d88cbe40fc440b6bda45046f3
112174Author: Keith Packard <keithp@neko.keithp.com>
112175Date: Sat Nov 4 17:40:34 2006 -0800
112176
112177 Define fbHasVisualTypes in fb.h as it is exported
112178
112179commit 8deaaa312ad7f9b492a2ae8ad17d74650112c25c
112180Author: Bernhard Rosenkraenzer <bero@arklinux.org>
112181Date: Sat Nov 4 18:59:39 2006 +0200
112182
112183 automake: avoid use of reserved _SOURCES keyword (bug #8866)
112184 Avoid using _SOURCES unless we're directly referencing a program or
112185 library to be built; use _SRCS instead. Shuts automake 1.10 up.
112186
112187commit f72927d26cd112d321f7bf187df3c740b3129d22
112188Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
112189Date: Sat Nov 4 19:00:57 2006 +0200
112190
112191 xfree86/hurd: remove OsMouseProc (bug #5613)
112192 Remove OsMouseProc, let the mouse driver deal with it.
112193
112194commit 0273610578485564c3c0be11b336b6554cc31b43
112195Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
112196Date: Sat Nov 4 19:02:04 2006 +0200
112197
112198 xfree86/hurd: add bell support, remove SERVER/LOADER defines (bug #5613)
112199 Add no-op bell ringing support, and remove obsolete @SERVER_DEFINES@ and
112200 @LOADER_DEFINES@ from CFLAGS.
112201
112202commit ade4bf09076d13dbf3549c0a2d987a0afe76d5c1
112203Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
112204Date: Sat Nov 4 19:03:13 2006 +0200
112205
112206 mesa/indirect: include byteswap.h on GNU userland (bug #5613)
112207 Include byteswap.h on all GNU-userland systems (including with the Hurd
112208 and FreeBSD kernels), not just Linux.
112209
112210commit 412e93349e1656c9650115328af4be0e59a66f74
112211Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
112212Date: Sat Nov 4 19:05:02 2006 +0200
112213
112214 kdrive: make building of Linux support conditional (bug #5613)
112215 Only try to build Linux support on Linux. We should probably disable all
112216 OS-dependent DDXes if we don't have a workable OS (and only build
112217 Xephyr/Xfake), but that's future work.
112218
112219commit cde8806c2930788ba8076e94651d391e45f3ccdb
112220Author: Eric Anholt <eric@anholt.net>
112221Date: Fri Nov 3 16:36:34 2006 -0800
112222
112223 Don't bump the refcnt if the new mode is NULL.
112224
112225commit 97fd471a627be185bee8cda3f709cfccea3fa12d
112226Author: Aaron Plattner <aplattner@nvidia.com>
112227Date: Fri Oct 27 12:36:56 2006 -0700
112228
112229 Fix standard VESA modes.
112230
112231 The built-in mode timings were off slightly for the 640x480@60, 640x480@72,
112232 and 1024x768@75 modes.
112233
112234commit 22ee2e4e1f1d9fd9ca9f25c9bf25370034b771d4
112235Merge: 49a2668 bd0c829
112236Author: Alan Coopersmith <alan.coopersmith@sun.com>
112237Date: Fri Nov 3 12:54:56 2006 -0800
112238
112239 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
112240
112241commit 49a26681b2bdd95ed65c425f1fa1441d2f092a6e
112242Author: Alan Coopersmith <alan.coopersmith@sun.com>
112243Date: Fri Nov 3 12:54:43 2006 -0800
112244
112245 Add DTrace probe points for X server <-> client communications
112246
112247 See http://people.freedesktop.org/~alanc/dtrace/ for more details
112248
112249commit bd0c829654903ca45543dfa59cda967c4fafd8ac
112250Author: Bjorn Helgaas <bjorn.helgaas@hp.com>
112251Date: Fri Nov 3 18:54:06 2006 +0100
112252
112253 Do not map full 0-1MB legacy range
112254
112255 If we're mapping something in the "legacy range" (0-1Mb), we shouldn't
112256 expand the requested range to the entire 0-1Mb range. Typically this
112257 is for mapping the VGA frame buffer, and some platforms support mmap of
112258 the frame buffer but not the entire 0-1Mb range.
112259
112260 For example, HP sx1000 and sx2000 ia64 platforms can have memory from
112261 0-0x9ffff, VGA frame buffer from 0xa0000-0xbffff, and memory from
112262 0xc0000-0xfffff. On these platforms, we can't map the entire 0-1Mb
112263 range with the same attribute because the memory only supports WB,
112264 while the frame buffer supports only UC. But an mmap of just the
112265 frame buffer should work fine.
112266
112267commit c1828a8ff51c8db326c47e6710f4f42fab94fb6d
112268Author: Egbert Eich <eich@freedesktop.org>
112269Date: Fri Nov 3 18:32:48 2006 +0100
112270
112271 Fixing mach64 driver bailing out on ia64
112272
112273 Mach64 driver bails out on ia64 because it cannot map device
112274 memory. It turns out that some bogus and unneeded code attempts
112275 to find the root bridge of the device and fails to do so proberly
112276 as there this host-to-pci bridge is not existant. This code has
112277 been around for years although it completely unclear what it had
112278 been intended for. Fixing this by eliminating the bogus code.
112279
112280commit d50fc413b39f52663b46084c28e81fc4933a7b49
112281Author: Matthias Hopf <mhopf@suse.de>
112282Date: Thu Nov 2 18:53:41 2006 +0100
112283
112284 Fix device path in altixPCI.c to be domain aware.
112285
112286commit 1d731fc54a2cf5d3f353d8ee1c7c4989df27f011
112287Author: Matthias Hopf <mhopf@suse.de>
112288Date: Thu Nov 2 18:50:15 2006 +0100
112289
112290 Add domain support to linuxPciOpenFile().
112291
112292 Loosely based on patch from David S. Miller <davem@davemloft.net>
112293 See also bug #2368.
112294
112295commit caaa113acf4144fd47a1ac93ca440d78d1983e54
112296Author: Matthias Hopf <mhopf@suse.de>
112297Date: Thu Nov 2 18:38:45 2006 +0100
112298
112299 Fixing domain support for ia64
112300
112301 ia64 specific functions + defines.
112302 Still uses /proc interface for some scaning code.
112303 Based on code from Egbert Eich <eich@freedesktop.org>.
112304
112305commit 16c5043fc0c84b14323cd211c2645106455ac320
112306Author: Matthias Hopf <mhopf@suse.de>
112307Date: Thu Nov 2 18:22:09 2006 +0100
112308
112309 Fix 2 warnings.
112310
112311commit 072c022e731c3aadf34096f16364e29df47280d2
112312Author: Matthias Hopf <mhopf@suse.de>
112313Date: Thu Nov 2 17:58:19 2006 +0100
112314
112315 Fix obviously wrong boundary checks + cleanup unused vars.
112316
112317 Also disable compilation of code that is no longer used anywhere in the whole
112318 Xserver tree.
112319
112320commit 5afc6c1a14fea2966017493b045fa7209faeb8eb
112321Author: Matthias Hopf <mhopf@suse.de>
112322Date: Thu Nov 2 15:42:03 2006 +0100
112323
112324 Added linux 2.6 compatible domain aware device scanning code.
112325
112326 Additional scanning code uses the /sys interface of 2.6 kernels.
112327 Cleaned up the use of tags and already split domain/bus/dev/fn.
112328
112329commit 6319f7d713971f70f06166480f069eca3bcace36
112330Author: Egbert Eich <eich@freedesktop.org>
112331Date: Thu Nov 2 12:50:52 2006 +0100
112332
112333 Make int10 fully domain aware.
112334
112335commit f4dd2665b0f9aa9c00a5152c73bc72cb7514eeb5
112336Author: Matthias Hopf <mhopf@suse.de>
112337Date: Thu Nov 2 12:36:12 2006 +0100
112338
112339 Added missing domain stripping in already domain aware code.
112340
112341commit 46901063e8edc82b67989f4e5eec39d17c67dc98
112342Author: Matthias Hopf <mhopf@suse.de>
112343Date: Thu Nov 2 12:25:03 2006 +0100
112344
112345 Build with -D_PC on ix86 only.
112346
112347commit 543b397277d1f03b8091e44812010abcd5d80102
112348Merge: 4056e6e 645d057
112349Author: Keith Packard <keithp@neko.keithp.com>
112350Date: Thu Nov 2 19:00:35 2006 -0800
112351
112352 Merge branch 'origin' into randr-1.2
112353
112354commit 037f23e6f8fbe6e6fc8e71ed21958fc553df72d0
112355Author: Alan Coopersmith <alan.coopersmith@sun.com>
112356Date: Thu Nov 2 18:30:58 2006 -0800
112357
112358 Convert Xprt DDX to new motion history api
112359
112360commit 7dc54a40e900cbea1e509620623b091d54a3c2d1
112361Author: Alan Coopersmith <alan.coopersmith@sun.com>
112362Date: Thu Nov 2 16:38:47 2006 -0800
112363
112364 Remove references to xf86Info.kbd* from solaris code
112365
112366commit e46f7f78b362e76f5a553184e3f5ec7e109aa39d
112367Author: Alan Coopersmith <alan.coopersmith@sun.com>
112368Date: Thu Nov 2 16:38:12 2006 -0800
112369
112370 sun_bell.c needs to #include "xf86_OSlib.h"
112371
112372commit 58bf9a142d1957f4d77038ee4ce7b1116b1f7955
112373Author: Alan Coopersmith <alan.coopersmith@sun.com>
112374Date: Thu Nov 2 14:43:02 2006 -0800
112375
112376 PostSyntheticMotion needs to be extern, not static, since it's in getevents.c
112377
112378commit 1ecd45fb8e4250fb51daa2bdf1a960af0f8b53fb
112379Author: Daniel Stone <daniel@fooishbar.org>
112380Date: Thu Nov 2 04:51:03 2006 +0200
112381
112382 Makefile.am: add config to DIST_SUBDIRS as well
112383
112384commit 01afa533aa872d1a101a41153f95d800e68fea3e
112385Author: Daniel Stone <daniel@fooishbar.org>
112386Date: Thu Nov 2 04:32:37 2006 +0200
112387
112388 xfree86/os-support: axe more unused files
112389
112390commit 57c1409151cb1f6e0e528fb92ebda58f86f12c1a
112391Author: Daniel Stone <daniel@fooishbar.org>
112392Date: Thu Nov 2 04:21:06 2006 +0200
112393
112394 xfree86/os-support: remove unused file from dist
112395
112396commit de5a4c63747a417cdece919f4fb5a4004a3ee7bb
112397Author: Daniel Stone <daniel@fooishbar.org>
112398Date: Thu Nov 2 04:18:33 2006 +0200
112399
112400 xkb: note that we allow full xi interaction
112401 We now allow maps to be set (etc) on different keyboards, so stop putting
112402 XkbXI_KeyboardsMask in unsupported.
112403
112404commit 1d65429a9e03871969552d0c31b022546cc46b12
112405Author: Daniel Stone <daniel@fooishbar.org>
112406Date: Thu Nov 2 04:12:55 2006 +0200
112407
112408 xfree86: don't attempt to enable and disable non-DIX devices
112409 Don't try to enable and disable devices with no entry in the DIX, such as
112410 the evdev brain.
112411
112412commit 64139c1950ea825c0a0124abc5f88499e91f797f
112413Author: Daniel Stone <daniel@fooishbar.org>
112414Date: Thu Nov 2 03:22:09 2006 +0200
112415
112416 bump to 1.2.99.0
112417
112418commit 18c246a13b887b865de6a17e6cd1c259b9bc383d
112419Merge: 794f2e7 4843d82
112420Author: Daniel Stone <daniel@fooishbar.org>
112421Date: Thu Nov 2 03:21:37 2006 +0200
112422
112423 Merge branch 'input-hotplug'
112424
112425commit 4843d823f4d38d8bd468ce3a8feddbff229ed416
112426Merge: ba9f513 a7b944f
112427Author: Daniel Stone <daniel@fooishbar.org>
112428Date: Thu Nov 2 03:18:13 2006 +0200
112429
112430 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
112431
112432commit ba9f5138fc32a7a7b97bcf941bc92751b7c6c2c0
112433Author: Daniel Stone <daniel@fooishbar.org>
112434Date: Thu Nov 2 03:16:10 2006 +0200
112435
112436 xfree86: allow starting with no input devices
112437 Add a server flag (AllowEmptyInput), which will inhibit adding the
112438 standard keyboard and mouse drivers, if there are no input devices in the
112439 config file.
112440
112441commit be291a6d9764cf29a7d9a8114d47d9f41ce856e9
112442Merge: a2d6242 6fdfd9d
112443Author: Daniel Stone <daniel@fooishbar.org>
112444Date: Thu Nov 2 03:15:25 2006 +0200
112445
112446 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
112447
112448commit a7b944f0d96c3e0e15e75378a04def1ac96089fb
112449Author: Alan Coopersmith <alan.coopersmith@sun.com>
112450Date: Wed Nov 1 16:17:49 2006 -0800
112451
112452 If getpeerucred() is available, include pid & zoneid in audit messages too
112453
112454commit fbfb35189ef6666707097704b43e052cb2f919ae
112455Author: Alan Coopersmith <alan.coopersmith@sun.com>
112456Date: Wed Nov 1 15:11:48 2006 -0800
112457
112458 Bug #1997: AUDIT messages should contain uid for local accesses
112459
112460 <https://bugs.freedesktop.org/show_bug.cgi?id=1997>
112461
112462commit a2d6242106bb3a440faa9cad157e0120dbfa7b6e
112463Author: Daniel Stone <daniel@fooishbar.org>
112464Date: Thu Nov 2 00:46:33 2006 +0200
112465
112466 kdrive:remove Change{Keyboard,Pointer}Device
112467 This was removed in the DIX, so just axe it.
112468
112469commit d9a5e3e964b6c91fbca88b50674fce9660b972a4
112470Author: Daniel Stone <daniel@fooishbar.org>
112471Date: Thu Nov 2 00:45:23 2006 +0200
112472
112473 XkbCopyKeymap: be more careful with levels, allocate compat/geom
112474 Take various extra precautions with copying levels across (thanks Chris
112475 Lee for a gdb session), including allocating when we don't already have a
112476 coherent map.
112477 Only free type components if they're present.
112478 Allocate geometry and compat components if we don't already have them in
112479 the dest map.
112480
112481commit 6fdfd9dad91d7b7aa292f8c4d268dd27c34de8d3
112482Author: Eric Anholt <eric@anholt.net>
112483Date: Wed Nov 1 14:29:59 2006 -0800
112484
112485 Fix several cases where optimized paths were hit when they shouldn't be.
112486
112487 This fixes a number of rendercheck cases.
112488
112489commit 40f84793bca40dcc6883d51aefa1bda44bd1ac61
112490Author: Alan Coopersmith <alan.coopersmith@sun.com>
112491Date: Wed Nov 1 14:34:46 2006 -0800
112492
112493 Propogate $LIBS for xtrans, clock_gettime, libm, etc. to libs used for each server
112494
112495commit d7d931abe01a8cf555b027f2bcfcccd5e9053e52
112496Author: Daniel Stone <daniel@fooishbar.org>
112497Date: Wed Nov 1 23:48:58 2006 +0200
112498
112499 configure.ac: remove check for rate/period
112500 Keyboard stuff is now handled in the kbd driver.
112501
112502commit 81728558a044fdde0e1d63da7b6314755f77296e
112503Author: Daniel Stone <daniel@fooishbar.org>
112504Date: Wed Nov 1 23:10:26 2006 +0200
112505
112506 input: add non-keyboard bell ringing function
112507 Add a generic 'ring the bell' function (console bell on Linux and BSD,
112508 /dev/audio on Solaris), and add DDX functions for this. Make this the
112509 core keyboard's bell.
112510 Port Xvfb and Xnest to this.
112511 Port XFree86 to this, with OS-specific hooks for Linux, BSD, and Solaris
112512 taken from foo_io.c in the old layer.
112513
112514commit 3df454719f9cbf6a046cb7458019ec621b3b42ee
112515Author: Daniel Stone <daniel@fooishbar.org>
112516Date: Wed Nov 1 23:02:57 2006 +0200
112517
112518 kdrive: add KdRingBell prototype
112519
112520commit 4056e6e79a4e37101d298ae29139c83d3816368b
112521Author: Keith Packard <keithp@mandolin.keithp.com>
112522Date: Wed Nov 1 00:29:46 2006 -0800
112523
112524 Move physical size from mode to output.
112525
112526 Modes can be shared across different sized monitors this way.
112527
112528 Also caught some missing byteswapping and an incorrect return type.
112529
112530commit c03311a1e78daa291477a67b1bb7206772108c5d
112531Author: Alan Coopersmith <alan.coopersmith@sun.com>
112532Date: Tue Oct 31 16:05:48 2006 -0800
112533
112534 Fix automake error: BUILT_SOURCES was defined multiple times on Solaris
112535
112536commit a2434ec5f3c9dc79d1f05c2d704a82a766718ed4
112537Author: Alan Coopersmith <alan.coopersmith@sun.com>
112538Date: Tue Oct 31 15:57:59 2006 -0800
112539
112540 Make _POSIX_C_SOURCE hack work with Solaris headers
112541
112542 Solaris headers are very literal - if you ask for POSIX_C_SOURCE 199309L,
112543 they limit to only the functions in that standard and no more, unless you
112544 also specify __EXTENSIONS__ to allow functions beyond the standard base.
112545
112546commit 645d0576205532a3610ae351267d5b84d76236bd
112547Author: Matthieu Herrb <matthieu.herrb@laas.fr>
112548Date: Sun Oct 29 18:19:56 2006 +0100
112549
112550 Handle building in a separate objdir
112551
112552commit 59584c375f4e4b2670a92002ecb7a78a0bc50cce
112553Author: Matthieu Herrb <matthieu.herrb@laas.fr>
112554Date: Sun Oct 29 17:49:46 2006 +0100
112555
112556 kill GNU-make'ism.
112557
112558commit 0107320fac0913aae2cb169992e31c670b4bd2f7
112559Merge: 06b6b97 a34446f
112560Author: Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
112561Date: Sun Oct 29 15:23:35 2006 +0100
112562
112563 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
112564
112565commit 06b6b971d065226b983ba25da7ea8236ec37df04
112566Author: Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
112567Date: Sun Oct 29 15:22:37 2006 +0100
112568
112569 Make sure we have 64-bit file-offsets in libdrm.
112570
112571commit 644ee2434a4e09f5baea00a486911f8c24b9a261
112572Author: Daniel Stone <daniel@fooishbar.org>
112573Date: Sun Oct 29 03:52:11 2006 +0300
112574
112575 kdrive: avoid null dereferences
112576 Avoid jumping into NULL pointers for OS or card functions which do not
112577 exist.
112578
112579commit 738d2e88171bcce8782d670a1cda9a1d941d7977
112580Author: Daniel Stone <daniel@fooishbar.org>
112581Date: Sun Oct 29 03:48:02 2006 +0300
112582
112583 kdrive: move bell ringing into an OS function
112584 Move the bell into an OS function, and use that if it's declared; else,
112585 fall back to using the driver's function.
112586 Remove the Linux keyboard bell function; just move it into the OS layer.
112587 Use named initialisers when converting the old structures, and eliminate
112588 unused functions.
112589
112590commit 96e32805d12fc36f0fa0926dbfb0dd8a5cadb739
112591Author: Daniel Stone <daniel@fooishbar.org>
112592Date: Sun Oct 29 03:43:34 2006 +0300
112593
112594 Xi: disallow changing core keyboard and pointer
112595 Just short-circuit the change core keyboard/pointer requests.
112596
112597commit 68f595ca6c7883e030947b7f95c50e92aa733f2b
112598Author: Daniel Stone <daniel@fooishbar.org>
112599Date: Sun Oct 29 03:41:34 2006 +0300
112600
112601 GetTimeInMillis: use correct units for clock_gettime
112602 Make sure we're treating the nanoseconds as a long, not an int, so we
112603 don't overflow.
112604
112605commit 51a06b3c44509c72279b5cfcf2b52b9a35c461b0
112606Author: Daniel Stone <daniel@fooishbar.org>
112607Date: Sun Oct 29 03:40:57 2006 +0300
112608
112609 WaitForSomething: only rewind when delta is more than 250ms
112610 Only rewind time when we're more than (original delta + 250ms) away from
112611 executing the timer.
112612 When we're walking the timer list, use a goto to iterate all of them from
112613 the start again, since timers may drop out of the list.
112614 Don't bother trying to be smart in TimerSet, we'll pick it up in
112615 WaitForSomething anyway.
112616
112617commit a34446f5b3d90714969a90583c49cb1eae1c9651
112618Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
112619Date: Fri Oct 27 13:43:43 2006 -0400
112620
112621 Add missing file to list in Makefile.am
112622
112623commit 92d04e746bd9b8ad3ee217c165ace20468e079cf
112624Author: Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
112625Date: Fri Oct 27 18:26:30 2006 +0200
112626
112627 Import libdrm functions for the drm memory manager.
112628
112629commit e3d3d29db5eef057b4e8294377c9c3147436bc2f
112630Merge: 92387e9 004d00e
112631Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
112632Date: Thu Oct 26 20:21:16 2006 -0400
112633
112634 Merge branch 'master' into my-XACE-SELINUX
112635
112636commit 92387e99d085b0b081fcedb2f20304eb0ac536b1
112637Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
112638Date: Thu Oct 26 20:20:57 2006 -0400
112639
112640 Change symbol in libaudit library test.
112641
112642commit 196c5836f463c28f633bbba847f59acd5935359d
112643Author: Daniel Stone <daniel@fooishbar.org>
112644Date: Fri Oct 27 01:27:31 2006 +0300
112645
112646 CoreKeyboardProc: annotate with FIXME
112647 Setting an empty keymap by default isn't wildly useful.
112648
112649commit f9a1e456f8a4eaa1a9c71fd0fe5231140975c22d
112650Author: Daniel Stone <daniel@fooishbar.org>
112651Date: Fri Oct 27 01:25:39 2006 +0300
112652
112653 CoreKeyboardProc: don't leak keymap and modmap
112654 SetKeySymsMap does a copy here, so try not to leak them.
112655
112656commit a5be65401769fabcb5001dc63035c69f9e4a2712
112657Author: Daniel Stone <daniel@fooishbar.org>
112658Date: Fri Oct 27 01:25:21 2006 +0300
112659
112660 mieqEnqueue: only compare DEVICE_BITS of deviceid
112661 Only compare DEVICE_BITS of the two deviceids, so we don't decide that
112662 a valuator event isn't for us, because (id | MORE_EVENTS) != id.
112663
112664commit 85212eb504f860b054eb0f0a5029fed86cb8d1c0
112665Author: Daniel Stone <daniel@fooishbar.org>
112666Date: Fri Oct 27 01:23:58 2006 +0300
112667
112668 getValuatorEvents: make sure we put MORE_EVENTS in the right places
112669 Make sure we put MORE_EVENTS in with the device id if there are, in fact,
112670 more valuator events coming.
112671
112672commit 794f2e7291ccb4e48f9fbfc8f08302e3aac0f79f
112673Author: Myron Stowe <myron.stowe@hp.com>
112674Date: Thu Oct 26 20:38:58 2006 +0300
112675
112676 xfree86: re-enable chipset-specific drivers for Linux/ia64
112677 Re-enable chipset-specific support for Linux/ia64, by linking in
112678 lnx_ia64.c.
112679
112680commit 8c0556e7cb1de8c387ddd886a03a8f8afff1fd0e
112681Merge: cdc8a4b 004d00e
112682Author: Daniel Stone <daniel@fooishbar.org>
112683Date: Thu Oct 26 15:21:22 2006 +0300
112684
112685 Merge branch 'master' into input-hotplug
112686
112687commit 004d00e6689f452fc9fdf91f5ffc6d6aed697d54
112688Author: Daniel Stone <daniel@fooishbar.org>
112689Date: Thu Oct 26 01:10:08 2006 +0300
112690
112691 GetTimeInMillis: simplify monotonic test
112692 We don't actually need to get the CPU clock ID, which means we don't need
112693 the monotonic_usable test. Since there's now only one branch, the
112694 compiler will treat that as likely, so we don't need xproto 7.0.9 anymore.
112695
112696 The fallthrough to gettimeofday() is preserved.
112697
112698commit cdc8a4b7b2f099b8860a54c5c9f488e6f7c4913a
112699Merge: 3da918a d285833
112700Author: Daniel Stone <daniel@fooishbar.org>
112701Date: Thu Oct 26 00:28:30 2006 +0300
112702
112703 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
112704
112705commit d285833290316cb5dd1e7f1e52c96be3e9cf21cd
112706Author: Daniel Stone <daniel@fooishbar.org>
112707Date: Wed Oct 25 23:57:00 2006 +0300
112708
112709 GetTimeInMillis: spuport monotonic clock
112710 Add support for CLOCK_MONOTONIC from clock_gettime, and use that in
112711 GetTimeInMillis() if available, falling back to the old gettimeofday()
112712 implementation.
112713
112714 This is _slightly_ faster on some 64-bit architectures, and _slightly_
112715 slower on others (though barely measurable).
112716
112717commit d3e57faffee63df1424a209d0418d3a712f91ae6
112718Author: Daniel Stone <daniel@fooishbar.org>
112719Date: Wed Oct 25 23:55:43 2006 +0300
112720
112721 WaitForSomething: allow time to rewind
112722 If time rewinds dramatically, reset all the timers to fix their expiry.
112723
112724commit e21604914dccece6bc64c69b55512d1f1a969235
112725Author: Keith Packard <keithp@neko.keithp.com>
112726Date: Wed Oct 25 09:48:23 2006 -0700
112727
112728 Merge master back in and clean up some unfinished code (closes 8745)
112729
112730commit becbda6d519a11c2c211afb8d46f9ea1a2676bc3
112731Author: Michel Dänzer <michel@tungstengraphics.com>
112732Date: Wed Oct 25 11:39:27 2006 +0200
112733
112734 Fix test for Option "IgnoreABI".
112735
112736 This option has plenty of potential for wasting the time of bug triagers
112737 without pretending it's always on.
112738
112739commit 59511974db843fa7e11133894cfc10c20fdaf60f
112740Merge: 054f8cd 598ac7a
112741Author: Keith Packard <keithp@neko.keithp.com>
112742Date: Tue Oct 24 17:26:20 2006 -0700
112743
112744 Merge branch 'master' into randr-1.2
112745
112746commit 598ac7a83698327b607084abaebcbd22f8d25fbb
112747Merge: 828c34e 948a97b
112748Author: Keith Packard <keithp@neko.keithp.com>
112749Date: Tue Oct 24 17:23:12 2006 -0700
112750
112751 Merge branch 'origin'
112752
112753commit 828c34e83ccdf3bcd2844d5af8b0cac4164b04ab
112754Author: Keith Packard <keithp@neko.keithp.com>
112755Date: Tue Oct 24 17:23:02 2006 -0700
112756
112757 Byte swap RRSelectInput enable flags.
112758
112759commit 3da918a16c8908fdfaf89f2a1bcaec19e01528a9
112760Author: Daniel Stone <daniel@fooishbar.org>
112761Date: Wed Oct 25 02:22:07 2006 +0300
112762
112763 mipointer: remember to update pointer location
112764 Update pointer location so it doesn't get quickly reset by the next
112765 pointer update.
112766
112767commit aabc087998e680c2fcf0ebc1c5022c1fe8f58f0c
112768Author: Daniel Stone <daniel@fooishbar.org>
112769Date: Wed Oct 25 02:21:39 2006 +0300
112770
112771 GetPointerEvents: always ensure correct number of events
112772 Ensure correct number of valuator events are returned, and that we always
112773 increment events correctly.
112774
112775commit 65cd5aa4d754624566c2263015f1a018d137fce1
112776Author: Daniel Stone <daniel@fooishbar.org>
112777Date: Wed Oct 25 01:12:45 2006 +0300
112778
112779 kdrive/input: remove unnecessary #ifdef XINPUTs
112780 Xi is now mandatory, so don't bother with the ifdefs.
112781
112782commit 0514d53e10b3521bb708a9cbde4bab525248eadb
112783Author: Daniel Stone <daniel@fooishbar.org>
112784Date: Wed Oct 25 01:10:44 2006 +0300
112785
112786 xfree86: remove motion history handling
112787 Remove motion history handling, as we now deal with this in the DIX.
112788
112789commit b1debebf8fe20ded20ba27e871fd1a6a9de029e3
112790Author: Daniel Stone <daniel@fooishbar.org>
112791Date: Wed Oct 25 01:10:20 2006 +0300
112792
112793 mi: remove mi motion history
112794 This is now unneeded as we do motion history in the DIX.
112795
112796commit b9e180e632d04bf685ade9e32bd0b20882794486
112797Author: Daniel Stone <daniel@fooishbar.org>
112798Date: Wed Oct 25 01:09:19 2006 +0300
112799
112800 port all users to the new DIX motion history API
112801 Port KDrive, Xvfb, and Xnest, as well as the virtual core devices, to the
112802 new motion history API. Make GetPointerEvents also update the history.
112803
112804commit 5b38eb69cdaa154791c7f74e35dbe4d3256b19bd
112805Author: Daniel Stone <daniel@fooishbar.org>
112806Date: Wed Oct 25 01:08:29 2006 +0300
112807
112808 dix: add motion history support
112809 Add motion history support (sort of based on the XFree86 DDX's
112810 implementation) to the DIX.
112811
112812commit 11fb58be77ac163844e494b2b0a260cf28a7ecd1
112813Author: Daniel Stone <daniel@fooishbar.org>
112814Date: Wed Oct 25 01:07:36 2006 +0300
112815
112816 miscellaneous warning fixes
112817 Use the correct type for time, and fix the mi prototype of EnqueueEvent.
112818
112819commit 2a74b8a91da1a98669993078f7fe9081f2d743ce
112820Author: Daniel Stone <daniel@fooishbar.org>
112821Date: Wed Oct 25 01:04:53 2006 +0300
112822
112823 xfree86: re-bump input abi to 1.0, yet again
112824
112825commit 37d1fffe79c35ada056ce9a56292c000014fe48a
112826Author: Daniel Stone <daniel@fooishbar.org>
112827Date: Tue Oct 24 23:06:57 2006 +0300
112828
112829 inputstr: try to keep device structs the same size
112830 Try to make sure DeviceIntRec and friends stay the same size,
112831 regardless of whether or not XKB is enabled.
112832
112833commit 9f9ac01a819ee96fb5be5d7d346c91f461bf3979
112834Author: Daniel Stone <daniel@fooishbar.org>
112835Date: Tue Oct 24 23:01:05 2006 +0300
112836
112837 inputstr: fix indentation
112838
112839commit 27d682fa030501611a5b52f41f29738134b45c0b
112840Author: Ian Romanick <idr@us.ibm.com>
112841Date: Tue Oct 24 09:04:14 2006 -0700
112842
112843 Fix from HEAD accidentally reverted during merge.
112844
112845commit 8b909135664abb69ada522e8e39495d9eca717ac
112846Merge: 21291d6 948a97b
112847Author: Ian Romanick <idr@us.ibm.com>
112848Date: Tue Oct 24 08:57:59 2006 -0700
112849
112850 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
112851
112852 Conflicts:
112853
112854 hw/xfree86/Makefile.am
112855 hw/xfree86/common/xf86.h
112856 hw/xfree86/common/xf86DoScanPci.c
112857 hw/xfree86/os-support/bus/linuxPci.c
112858 hw/xfree86/scanpci/extrapci.ids
112859 hw/xfree86/scanpci/pci.ids
112860 hw/xfree86/scanpci/pciid2c.pl
112861 hw/xfree86/scanpci/xf86PciStdIds.h
112862
112863commit 0cd6a3d8efb5cc1ce4f85ab95bcdf4fb66c7245d
112864Author: Daniel Stone <daniel@fooishbar.org>
112865Date: Mon Oct 23 06:56:07 2006 +0300
112866
112867 xfree86/input: add proximity support, free valuators
112868 Re-add support for proximity events, and remember to both va_end our
112869 varargs, and free our valuators.
112870
112871commit cccf7ae0ff24d0c84b5144c457f3f86bbbc36e12
112872Author: Daniel Stone <daniel@fooishbar.org>
112873Date: Mon Oct 23 06:55:21 2006 +0300
112874
112875 GetProximityEvents: add (untested) function
112876 Add untested first guess at what GetProximityEvents should look like.
112877
112878commit 35fa4ac12b3da33f81e2a12bc9661ed075f323ed
112879Author: Daniel Stone <daniel@fooishbar.org>
112880Date: Mon Oct 23 06:23:45 2006 +0300
112881
112882 GetPointerEvents: break into separate functions
112883 Break out clipAxis, clipValuators, and getValuatorEvents, into
112884 separate functions, to be used by the proximity event code.
112885
112886commit a7c2d9a15dc2ff253bb69c3b0738ad993521b9c7
112887Author: Daniel Stone <daniel@fooishbar.org>
112888Date: Mon Oct 23 06:08:27 2006 +0300
112889
112890 kdrive: numerous warning fixes
112891
112892commit c8f76fb3a473a022d497bd0acd6c84f58fc6efbe
112893Author: Daniel Stone <daniel@fooishbar.org>
112894Date: Mon Oct 23 05:12:15 2006 +0300
112895
112896 xfree86/input: disallow pointer device changes, fix drag event calculation
112897 Don't allow users to change the core pointer.
112898 Fix xf86SendDragEvents to check the device button state, not the core
112899 pointer's.
112900 Remove unused xf86CheckButton.
112901
112902commit c5b5b046e86b159a9d32451308f38246cc4587f7
112903Merge: fab1d37 948a97b
112904Author: Daniel Stone <daniel@fooishbar.org>
112905Date: Mon Oct 23 02:58:30 2006 +0300
112906
112907 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
112908
112909commit fab1d37ecbeee9777ad1485b0cc62b55042d5014
112910Author: Daniel Stone <daniel@fooishbar.org>
112911Date: Mon Oct 23 02:57:00 2006 +0300
112912
112913 xfree86/options: don't blindly dereference NULL pointers
112914 Sure, conf_idev should be initialised, but if it's not, try to make the
112915 best of a bad situation.
112916
112917commit 523fe64ba16cf7a40f4095432d81300726d83e8d
112918Author: Daniel Stone <daniel@fooishbar.org>
112919Date: Mon Oct 23 02:56:28 2006 +0300
112920
112921 kdrive/lots of DDXes: make stubs use KdOsAddInputDrivers()
112922 In theory, removing LinuxPciScan should make these drivers OS-independent.
112923
112924commit 57f98e2b72d5c7fea93a9f6da86228ffb4856bf3
112925Author: Daniel Stone <daniel@fooishbar.org>
112926Date: Mon Oct 23 02:55:41 2006 +0300
112927
112928 kdrive/tslib: port to new absolute API, remove debugging
112929 Port to new absolute device (not touchscreen) API, and remove some
112930 leftover debugging.
112931
112932commit a6dc629052f0ad509cfa30e98217043a63c09552
112933Author: Daniel Stone <daniel@fooishbar.org>
112934Date: Mon Oct 23 02:54:59 2006 +0300
112935
112936 kdrive/os: add KdOsAddInputDrivers
112937 Add KdOsAddInputDrivers, which adds all relevant input drivers.
112938
112939 Could possibly be refactored to KdAddInputDrivers, which called through
112940 OsFuncs to a new function, if it existed.
112941
112942commit ac240b00ff6fc354c1aa641406ae8b49244c0b8b
112943Author: Daniel Stone <daniel@fooishbar.org>
112944Date: Mon Oct 23 02:54:07 2006 +0300
112945
112946 kdrive/input: use Absolute instead of Touchscreen
112947 Adjust to suit Zepheniah's new absolute device API.
112948
112949commit b1ba4b3e8ad427af1fa8618b0bd839f741ec2ce3
112950Author: Daniel Stone <daniel@fooishbar.org>
112951Date: Mon Oct 23 02:53:15 2006 +0300
112952
112953 kdrive/neomagic: include vesa.h, use DebugF
112954
112955commit 29f28dd1a8a7ed07e945a67946e3510f4b32d12a
112956Author: Daniel Stone <daniel@fooishbar.org>
112957Date: Mon Oct 23 02:52:52 2006 +0300
112958
112959 configure.ac: add XSERVER_LIBS to XNEST_LIBS
112960
112961commit 2f33f4065d89ae2b6fdda43c7105d72f89920cae
112962Author: Daniel Stone <daniel@fooishbar.org>
112963Date: Mon Oct 23 02:52:35 2006 +0300
112964
112965 Xnest: port to new input API
112966 Port Xnest to Get{Pointer,Keyboard}Events, plus the new mieq API.
112967
112968commit 08928afb0500d46b0caa0a1d1244dee2ed80e6a0
112969Author: Daniel Stone <daniel@fooishbar.org>
112970Date: Mon Oct 23 02:51:52 2006 +0300
112971
112972 Xnest: disable XKB, reshuffle code
112973 Disable XKB, as we can't yet use it; move Composite disabling to
112974 ddxInitGlobals, along with XKB.
112975
112976commit cd3b16a57efaf89108054f18a94c91e2dd74fafa
112977Author: Daniel Stone <daniel@fooishbar.org>
112978Date: Mon Oct 23 02:51:13 2006 +0300
112979
112980 Xvfb: port to new mieq API
112981
112982commit bf4df9b73f0c1a84093aaf9a2e2cbc56fb341c60
112983Author: Daniel Stone <daniel@fooishbar.org>
112984Date: Mon Oct 23 02:50:53 2006 +0300
112985
112986 include: move POINTER_* flags from inputstr.h to input.h
112987 Given they're just numeric constants, they should be included in
112988 input.h, not inputstr.h.
112989
112990commit 4dd91c45abea9fb561a5acb10290e29487df6722
112991Author: Daniel Stone <daniel@fooishbar.org>
112992Date: Mon Oct 23 02:50:03 2006 +0300
112993
112994 miinitext: Xi and XKB are not hardware-only extensions
112995 Xi is now a required extension, and XKB can be used without hardware,
112996 so include them both when NO_HW_ONLY_EXTS is defined.
112997
112998commit 562096a012f4bb8f44d5ec6320a32f4010c189e4
112999Author: Daniel Stone <daniel@fooishbar.org>
113000Date: Mon Oct 23 02:49:22 2006 +0300
113001
113002 XkbCopyKeymap: increment shapes and outlines when copying
113003 Remember to increment the source and destination shapes when copying,
113004 instead of just endlessly copying the first one.
113005
113006commit eec182259112fba240751f974f7e5ca09fce8b9d
113007Author: Daniel Stone <daniel@fooishbar.org>
113008Date: Mon Oct 23 02:48:30 2006 +0300
113009
113010 dix/getevents: move SyntheticMotion to getevents.c
113011 Mostly, this is just a cheesy hack to ensure that getevents.o gets
113012 included when linking. Sigh.
113013
113014commit bc701a14292da5abfb601e3a040651a74f46df8f
113015Author: Daniel Stone <daniel@fooishbar.org>
113016Date: Mon Oct 23 00:08:32 2006 +0300
113017
113018 dix/getevents: cosmetic cleanups, remove keymap copy from GKVE
113019 Remove keymap copy from GetKeyboardValuatorEvents, as
113020 SwitchCoreKeyboard now takes care of this for us.
113021 Remove unused variable and function prototype.
113022 Update comments to be as informative as possible.
113023
113024commit b03e2f7029506640a8fe5cb88818b329c23503ff
113025Author: Daniel Stone <daniel@fooishbar.org>
113026Date: Sun Oct 22 19:56:49 2006 +0300
113027
113028 xi: fix NIDR return yet again
113029 For a one-line function, it was pretty broken.
113030
113031commit 90de7ce25a84cfe6c6790f9af2bc2399d25b9b9c
113032Author: Daniel Stone <daniel@fooishbar.org>
113033Date: Sun Oct 22 19:54:36 2006 +0300
113034
113035 xi: fix return type for NIDR
113036
113037commit f46dc272913ffb6b5b234a7ec6f4ba5cae44a831
113038Author: Daniel Stone <daniel@fooishbar.org>
113039Date: Sun Oct 22 19:51:35 2006 +0300
113040
113041 xi: add NewInputDeviceRequest to stubs
113042
113043commit 31a6307b7ba5adaa96deb8101ddfcda0262f537d
113044Author: Daniel Stone <daniel@fooishbar.org>
113045Date: Sun Oct 22 19:49:31 2006 +0300
113046
113047 xi: change DEVICE_TOUCHSCREEN to ABS_{AREA_CALIB} for stubs
113048
113049commit eae6594d03a606ddf1f433b5897b5938aa940c1e
113050Author: Daniel Stone <daniel@fooishbar.org>
113051Date: Sun Oct 22 16:39:44 2006 +0300
113052
113053 Xi: swap control in DevicePresenceNotify
113054
113055commit be21630164e865eca72ff2a686a38ae4e30fd79c
113056Author: Daniel Stone <daniel@fooishbar.org>
113057Date: Sun Oct 22 16:33:02 2006 +0300
113058
113059 dix, Xi: make use of deviceid in DevicePresenceNotify
113060 Use the deviceid and control fields in DevicePresenceNotify since
113061 the last push to inputproto to send a DPN whenever a control changes
113062 on a device.
113063
113064commit f08b6b2367705cb5b60e996e6328197430bf1919
113065Author: Daniel Stone <daniel@fooishbar.org>
113066Date: Sun Oct 22 12:30:02 2006 +0300
113067
113068 kdrive: change DEVICE_TOUCHSCREEN to DEVICE_ABS_{CALIB,AREA}
113069
113070commit 77e724585f6c53feb55475b94d8cfcb6acf1159b
113071Author: Daniel Stone <daniel@fooishbar.org>
113072Date: Fri Oct 20 00:44:46 2006 +0300
113073
113074 minor formatting fixes
113075
113076commit 948a97b97e93cee05a23f1f1ae699c5e181bc8ce
113077Author: Drew Parsons <dparsons@debian.org>
113078Date: Sat Oct 21 23:09:22 2006 +1000
113079
113080 Minor typo fix to xorg.conf man page.
113081
113082commit 04554a3adcddc32de5fdb0b3122da0bcdd4c24a9
113083Author: Drew Parsons <dparsons@debian.org>
113084Date: Sat Oct 21 22:06:43 2006 +1000
113085
113086 Minor typos in Xserver man page.
113087
113088commit e26a494f417c3c700636ee68892c3015b2e0f27a
113089Merge: 736b0d5 aeba855
113090Author: Zephaniah E. Hull <warp@agamemnon.b5>
113091Date: Sat Oct 21 04:26:14 2006 -0400
113092
113093 Merge branch 'input-hotplug' of git+ssh://git.freedesktop.org/git/xorg/xserver
113094
113095commit 736b0d5078597abbda80444deef852879260af90
113096Author: Zephaniah E. Hull <warp@agamemnon.b5>
113097Date: Sat Oct 21 04:24:49 2006 -0400
113098
113099 DEVICE_TOUCHSCREEN becomes DEVICE_ABS_CALIB.
113100
113101 Update the DEVICE_ABS_CALIB stuff to include the new elements.
113102
113103 New DEVICE_ABS_AREA support.
113104
113105 dev->touchscreen becomes dev->absolute, with _CALIB and _AREA stuff in it.
113106
113107 Update xfree86 to compile with this, kdrive needs an update too.
113108
113109commit edd5f1745461f995670969cb736d1569ca94643f
113110Author: Carl Switzky <Carl.Switzky@Sun.COM>
113111Date: Thu Oct 19 17:30:54 2006 -0700
113112
113113 Add ast driver/device info to Xorg server & config utilities
113114
113115commit aeba855b07832354f59678e20cc29a085e42bd99
113116Author: Daniel Stone <daniel@fooishbar.org>
113117Date: Fri Oct 20 00:30:28 2006 +0300
113118
113119 move keymap copy to event processing, from enqueuing
113120
113121 Move the keymap copying to event processing time (in
113122 ProcessInputEvents), instead of being at event enqueuing time.
113123 Break SetCore{Pointer,Keyboard} out into separate functions.
113124 Change mieqEnqueue to take a device pointer, that asks for the
113125 _original_ device associated with this event.
113126
113127commit a8d3dad9d9f2b9053843e655abe463a68ba8dcb7
113128Author: Daniel Stone <daniel@fooishbar.org>
113129Date: Fri Oct 20 00:28:40 2006 +0300
113130
113131 xi: add DEVICE_ENABLE control
113132
113133 Add DEVICE_ENABLE control, which allows runtime enabling and disabling
113134 of specific devices.
113135
113136commit b0780312d80ea4af0136227f90fdd7ada3db71c5
113137Author: Alan Coopersmith <alan.coopersmith@sun.com>
113138Date: Thu Oct 19 13:51:53 2006 -0700
113139
113140 Pre-release message should tell users to check git, not CVS, for updates
113141
113142commit c5dc997baf57ffa08025efadbbaf761296ce4bc4
113143Author: Joshua Baergen <joshuabaergen@gentoo.org>
113144Date: Thu Oct 19 11:14:26 2006 -0700
113145
113146 Create xorg.conf.example (Gentoo bug #138623).
113147
113148commit d029c8f1b72019446a5c873f55ffa43504b03dfb
113149Author: Alan Coopersmith <alan.coopersmith@sun.com>
113150Date: Wed Oct 18 18:11:06 2006 -0700
113151
113152 Use getisax() instead of asm code to determine available x86 ISA extensions on Solaris
113153
113154commit a8a0abdbea0573c861a5af9d58f3ce66790455ca
113155Author: Daniel Stone <daniel@fooishbar.org>
113156Date: Wed Oct 18 10:59:07 2006 +0300
113157
113158 config/dbus: always unref the connection, not close
113159
113160commit 80642f37d40216035786eaf490952d16f6b5f597
113161Author: Adam Jackson <ajax@benzedrine.nwnk.net>
113162Date: Tue Oct 17 14:53:28 2006 -0400
113163
113164 Stop building xorgcfg by default.
113165
113166commit 5e17cde27b064174584d478130b0f95dcef78deb
113167Author: Matthias Hopf <mhopf@suse.de>
113168Date: Tue Oct 17 17:06:44 2006 +0200
113169
113170 StorePixels() macro could create invalid *x++=*x... code - fixed.
113171
113172commit 205c6788d7a34704e36b23f1a93d89e9b986266a
113173Author: Daniel Stone <daniel@fooishbar.org>
113174Date: Mon Oct 16 23:48:09 2006 +0300
113175
113176 config/dbus: properly initialise vtable
113177
113178 Properly initialise the vtable, so we don't end up with an unregister_function
113179 pointing to god knows where.
113180
113181commit 85ac2f16abe9f6e88b4e71609da334d336a9a600
113182Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
113183Date: Mon Oct 16 12:39:05 2006 +0100
113184
113185 Small modification to blocking signals when switching modes.
113186
113187commit 0901eec87ee9f3a2a067695bdbd569ff42149879
113188Author: Michel Dänzer <michel@tungstengraphics.com>
113189Date: Sun Oct 15 16:57:09 2006 +0200
113190
113191 Fix __glXDRIbindTexImage() for 32 bpp on big endian platforms.
113192
113193commit a232693c8c2a206aac47c07b133c071938204e0b
113194Author: Michel Dänzer <michel@tungstengraphics.com>
113195Date: Sun Oct 15 16:48:59 2006 +0200
113196
113197 Add per-drawable Xv colour key helper function.
113198
113199 This allows overlay Xv adaptors to work slightly better with compositing
113200 managers.
113201
113202 Bump the video driver ABI minor so drivers only need to check for this at build
113203 time.
113204
113205commit 5563861ab7e56ec891cfce6b34af43fec53ccee3
113206Author: Alan Coopersmith <alan.coopersmith@sun.com>
113207Date: Fri Oct 13 19:05:28 2006 -0700
113208
113209 Make sure xorgcfg files are included even when dist made with --disable-xorgcfg
113210
113211commit e1dd1904c6c7ce33f347d822272831d54a6497c8
113212Author: Alan Coopersmith <alan.coopersmith@sun.com>
113213Date: Thu Oct 19 13:51:53 2006 -0700
113214
113215 Pre-release message should tell users to check git, not CVS, for updates
113216
113217commit 62d24097932708fbbb62a23614fe63b4b7acf3bd
113218Author: Joshua Baergen <joshuabaergen@gentoo.org>
113219Date: Thu Oct 19 11:14:26 2006 -0700
113220
113221 Create xorg.conf.example (Gentoo bug #138623).
113222
113223commit 357b37b3826fa6e9878c0bd895164259c2ed3c0d
113224Author: Alan Coopersmith <alan.coopersmith@sun.com>
113225Date: Wed Oct 18 18:11:06 2006 -0700
113226
113227 Use getisax() instead of asm code to determine available x86 ISA extensions on Solaris
113228
113229commit 1b1698af41b9038d9f9dbf521737d0baab5a2237
113230Author: Zephaniah E. Hull <warp@agamemnon.b5>
113231Date: Wed Oct 18 04:57:22 2006 -0400
113232
113233 Pass SetDeviceValuators down to the driver.
113234 NOTE: This changes the LocalDeviceRec struct, which breaks input drivers.
113235
113236commit 5eca750fe2f3f243fb352271ad8da196af0cb16a
113237Author: Adam Jackson <ajax@benzedrine.nwnk.net>
113238Date: Tue Oct 17 14:53:28 2006 -0400
113239
113240 Stop building xorgcfg by default.
113241
113242commit df979b75c8cd8a7e0566aea58031bb9b8f5cd3d3
113243Author: Matthias Hopf <mhopf@suse.de>
113244Date: Tue Oct 17 17:06:44 2006 +0200
113245
113246 StorePixels() macro could create invalid *x++=*x... code - fixed.
113247
113248commit d430e76a161c963169067875c3654f5fd8f42b19
113249Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
113250Date: Mon Oct 16 12:39:05 2006 +0100
113251
113252 Small modification to blocking signals when switching modes.
113253
113254commit 6dd4fc4652f942724039dc2317c560ea7276ab59
113255Author: Daniel Stone <daniel@fooishbar.org>
113256Date: Mon Oct 16 00:22:00 2006 +0300
113257
113258 xkb: fix virtual modmap size computation
113259 Compute virtual modmap size bounded by nVModMapKeys-1, rather than
113260 nVModMapKeys.
113261
113262 This is sort of a best guess. The other way seems a little more
113263 logical, but also leads to segfaults pretty quickly if you hammer
113264 GetMap hard enough. So let's try this one.
113265
113266commit a484ba15277e66e7ef9b21b238dcbf760695bc63
113267Author: Daniel Stone <daniel@fooishbar.org>
113268Date: Sun Oct 15 23:47:34 2006 +0300
113269
113270 XkbCopyKeymap: copy server vmods, and name atoms
113271 Copy server->vmods, and all the atoms in names.
113272
113273commit ad355fecee3965be576596aeed5da54d776edf1d
113274Author: Daniel Stone <daniel@fooishbar.org>
113275Date: Sun Oct 15 21:59:06 2006 +0300
113276
113277 xkb: make sure we set the map on the right device, not necessarily core
113278 Forgot that all XKB requests took a device spec: the comparison of
113279 'if working on the core keyboard, does this device send core events; or,
113280 is this device the core keyboard?' was broken. Instead, what we want is
113281 'if working on the core keyboard, does this device send core events; or,
113282 is this device the one we're working on?'.
113283
113284commit 4ae7745a0dc86de6346409a69c1e396e0b954514
113285Author: Daniel Stone <daniel@fooishbar.org>
113286Date: Sun Oct 15 21:48:01 2006 +0300
113287
113288 xfree86 input: always open devices on NewInputDeviceRequest
113289
113290commit acd8419948003032056a56d46adbef7c35e7739c
113291Author: Daniel Stone <daniel@fooishbar.org>
113292Date: Sun Oct 15 20:42:31 2006 +0300
113293
113294 config: unref connection, don't close it
113295 Just unref the connection instead of explicitly closing it (thanks, Rob
113296 McQueen).
113297 Add a commented-out unregister_object_path call: unfortunately, when we
113298 call it, libdbus segfaults. But if we don't unregister the path, we
113299 can't register it again. So regenerations are broken either way, but a
113300 little less violently like this.
113301
113302commit fc9b5f84b244ea08480b73bd15ac919b875800fb
113303Author: Daniel Stone <daniel@fooishbar.org>
113304Date: Sun Oct 15 20:01:01 2006 +0300
113305
113306 dix/devices: add devices in proper forward order
113307 Add devices in forward order with the normal linked list convention.
113308 Previously, AddInputDevice would add all the devices in reverse order to
113309 off_devices, before they were added again in reverse order to devices with
113310 EnableDevice.
113311 This just makes both work in forward order, which provides the ordering as
113312 you'd expect when hotplugging devices (i.e. adds them to the head, not the
113313 tail).
113314
113315commit ec35e7198debf938f1115f584e675ce5995743e3
113316Author: Daniel Stone <daniel@fooishbar.org>
113317Date: Sun Oct 15 19:44:49 2006 +0300
113318
113319 config: add replies and dbus api documentation
113320 Add replies, which use standard X error values, to the two currently-supported
113321 input configuration requests.
113322 Document the D-BUS API we use.
113323 Make sure we free everything when we encounter an error.
113324 Add a _source option to all incoming requests, noting that it came from a
113325 client.
113326 Reject all requests to add a device where an option name contains an
113327 underscore.
113328
113329commit 7e4717683d6c08d1e490a60b7493a94bbc57bf8d
113330Author: Michel Dänzer <michel@tungstengraphics.com>
113331Date: Sun Oct 15 18:12:28 2006 +0200
113332
113333 exaDrawableDirty: Fix initialization of BoxRec.
113334
113335 This will hopefully fix the partial window corruption experienced by some
113336 people.
113337
113338commit 3ad1642f1bbaa5f96558cdf3384b40f7122f8781
113339Author: Michel Dänzer <michel@tungstengraphics.com>
113340Date: Sun Oct 15 16:57:09 2006 +0200
113341
113342 Fix __glXDRIbindTexImage() for 32 bpp on big endian platforms.
113343
113344commit f9bfee50981006a2c58d3f73e2b0d123bb2a41b7
113345Author: Michel Dänzer <michel@tungstengraphics.com>
113346Date: Sun Oct 15 16:48:59 2006 +0200
113347
113348 Add per-drawable Xv colour key helper function.
113349
113350 This allows overlay Xv adaptors to work slightly better with compositing
113351 managers.
113352
113353 Bump the video driver ABI minor so drivers only need to check for this at build
113354 time.
113355
113356commit a05044cfc14a8bc6cc31236dcecada60bec09924
113357Author: Daniel Stone <daniel@fooishbar.org>
113358Date: Sat Oct 14 22:14:56 2006 +0300
113359
113360 xkb: better support of XkbDfltXIId
113361 XKB.h specifies that XkbDfltXIId should be used where the client doesn't
113362 care about the device identifier. We take this to mean core devices,
113363 where practical.
113364
113365commit 4d8030076ed1a7680bdfcb7b89af1045bdc40304
113366Author: Daniel Stone <daniel@fooishbar.org>
113367Date: Sat Oct 14 22:14:07 2006 +0300
113368
113369 dix: move GetKeyboardEvents/GetPointerEvents to a new file, export symbols
113370 Move GKE and GPE to a separate file, to help stem the events.c explosion.
113371 Mark GKE/GKVE/GPE as _X_EXPORT.
113372
113373commit 6afc7c284690b1e2bb7544b5bc4f31a3f6a05519
113374Author: Daniel Stone <daniel@fooishbar.org>
113375Date: Sat Oct 14 15:54:35 2006 +0300
113376
113377 dix/devices: remove XACE merge debris
113378
113379commit 93302452e737bd91a893eb495592538d40d921e5
113380Author: Daniel Stone <daniel@fooishbar.org>
113381Date: Sat Oct 14 15:54:12 2006 +0300
113382
113383 XkbCopyKeymap: add geometry support
113384 Add a first cut at geometry support, which seems to generally work.
113385
113386commit b9108a13fc126d97c0393f911a1d9292563444ce
113387Author: Alan Coopersmith <alan.coopersmith@sun.com>
113388Date: Fri Oct 13 19:05:28 2006 -0700
113389
113390 Make sure xorgcfg files are included even when dist made with --disable-xorgcfg
113391
113392commit 054f8cd2675a80b14bc1ce266377fcfee2335cee
113393Author: Keith Packard <keithp@neko.keithp.com>
113394Date: Fri Oct 13 17:34:53 2006 -0700
113395
113396 Limit pointer to valid crtc areas. Add event swapping. Fix change tracking.
113397
113398 Add function to keep pointer within valid crtc areas.
113399 Finish event delivery and swapping code.
113400 Separate configuration from layout changes to send correct events.
113401
113402commit 335b503c5e7041bb0c44611e496d1c46f554e630
113403Merge: bd3d93b cf948b7
113404Author: Daniel Stone <daniel@fooishbar.org>
113405Date: Fri Oct 13 18:10:45 2006 +0300
113406
113407 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
113408
113409commit cf948b7b04dfeb61a294889027b9a54f6b9b478e
113410Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
113411Date: Fri Oct 13 14:48:10 2006 +0100
113412
113413 Block IO on switchmode just as we do for VT switching
113414 to avoid cursor movements signalling io.
113415
113416commit 4b53a5e4b147ab748b28dffc1d330b7148d8efa6
113417Merge: 23f6f08 042d4a4
113418Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
113419Date: Wed Oct 11 20:50:31 2006 -0400
113420
113421 Merge branch 'master' into my-XACE-SELINUX
113422
113423commit a6df780e4d3f90cc699b0b864bade03d6f15afd7
113424Author: Ian Romanick <idr@us.ibm.com>
113425Date: Wed Oct 11 15:41:22 2006 -0700
113426
113427 Fix bug #8608.
113428
113429 Regenerate files based on recent commits to Mesa (that fix Mesa bug
113430 that return pixel data.
113431
113432commit 042d4a407d0eca9131d0420b0b9189559aac326d
113433Author: Aaron Plattner <aplattner@nvidia.com>
113434Date: Mon Oct 9 16:32:11 2006 -0700
113435
113436 Bug #8459: Don't forget to include AM_CFLAGS in libfb_la_CFLAGS on non-MMX platforms.
113437
113438 Reported by Edgar Toernig.
113439
113440commit 1b94c117e0f294ef2f89bf24d45ba7a8e45efe35
113441Author: Matthias Hopf <mhopf@suse.de>
113442Date: Tue Oct 10 19:37:22 2006 +0200
113443
113444 Fixing probably long-standing bug in domain support.
113445
113446commit 6776c0f0e9677a65ec36ceef9618ff701b99596c
113447Author: Alan Coopersmith <alan.coopersmith@sun.com>
113448Date: Mon Oct 9 17:24:37 2006 -0700
113449
113450 Use bash on Solaris to run symlink-mesa.sh
113451
113452 symlink-mesa.sh won't run with Solaris Bourne Shell (/bin/sh) so explicitly
113453 run it with /usr/bin/bash instead
113454 (cherry picked from cde68728860179dc84e615ccb378ce992513fd62 commit)
113455
113456commit bd3d93be82d91e4cf35ae317dfd658d1706257ea
113457Author: Daniel Stone <daniel@fooishbar.org>
113458Date: Sun Oct 8 23:21:12 2006 +0300
113459
113460 xkb: remove random broken vendor workarounds
113461
113462commit 33406da096b4ae21134484113b280e07d0c8f0d9
113463Author: Daniel Stone <daniel@fooishbar.org>
113464Date: Sun Oct 8 23:20:56 2006 +0300
113465
113466 GetKeyboardEvents: add first_valuator argument to GKVE
113467
113468commit ef7e05e9de57b9c9c27ed3307eede6d8fc6c1af3
113469Author: Daniel Stone <daniel@fooishbar.org>
113470Date: Sun Oct 8 21:42:15 2006 +0300
113471
113472 xfree86/xinput: pass first_valuator params, don't clip to first screen bound
113473 Pass first_valuator to GetPointerEvents/GetKeyboardValuatorEvents.
113474 Don't clip axis bounds to that of screen 0 when initialising axes.
113475
113476commit 1546a398144619a14ff06aaf84ebc1bf293eac66
113477Author: Daniel Stone <daniel@fooishbar.org>
113478Date: Sun Oct 8 21:40:53 2006 +0300
113479
113480 GetKeyboardValuatorEvents: also take first_valuator param
113481 Take a first_valuator parameter, which specifies the first valuator.
113482
113483commit b05a11478edc7e6d1e38ef7f8d6788c7bd917493
113484Author: Daniel Stone <daniel@fooishbar.org>
113485Date: Sun Oct 8 21:23:12 2006 +0300
113486
113487 doc/extensions: rename to c-extensions
113488 The old name could be somewhat confusing.
113489
113490commit 41bb9fce47f6366cc3f7d45790f7883f74289b5a
113491Author: Daniel Stone <daniel@fooishbar.org>
113492Date: Sun Oct 8 20:34:32 2006 +0300
113493
113494 mipointer: take device arguments, split miPointerAbsoluteCursor
113495 Update mipointer API to take a device argument to (almost) all functions,
113496 and split miPointerAbsoluteCursor into a couple of separate functions.
113497 Remove miPointerAbsoluteCursor call from mieq, as we now deal with it in
113498 GetPointerEvents.
113499 Make miPointerSetPosition (successor of miPointerAbsoluteCursor) take
113500 pointers to x and y, so it can return the clipped values.
113501 Modify callers of miPointer*() functions to generally use the new
113502 functions.
113503 This should fix things with multi-head setups.
113504
113505commit be8dfafd1d58b27bbfd953fc1216311523353db1
113506Author: Daniel Stone <daniel@fooishbar.org>
113507Date: Sun Oct 8 20:32:16 2006 +0300
113508
113509 warning cleanups
113510 Fix still more warnings.
113511
113512commit ca474e0920dd29ebe7ccf346cddc526732ad01ba
113513Author: Daniel Stone <daniel@fooishbar.org>
113514Date: Sun Oct 8 20:30:49 2006 +0300
113515
113516 Xi: move SendEventToAllWindows and FindInterestedChildren to exevents
113517 Move SendEventToAllWindows and FindInterestedChildren from chgptr to exevents,
113518 so the DIX can more easily use it.
113519 Clean up two warnings (type mismatch, unused variable) in exevents.c.
113520
113521commit c2fab469b66f2796c541e911202faa411d116b04
113522Author: Daniel Stone <daniel@fooishbar.org>
113523Date: Sun Oct 8 18:26:26 2006 +0300
113524
113525 dix/devices: clean up debugging
113526
113527commit 4493acb88c59721f7807093a3ed3c39396c2076d
113528Author: Daniel Stone <daniel@fooishbar.org>
113529Date: Sun Oct 8 17:51:03 2006 +0300
113530
113531 xkb: add FIXMEs to procedures which need to act on all core devices
113532 Add FIXME comments above request handlers which need to act on all core-sending
113533 devices if called on the core keyboard.
113534
113535commit ef68273f5bdb27a492ec0b69548ec4fbede46c08
113536Author: Daniel Stone <daniel@fooishbar.org>
113537Date: Sun Oct 8 17:44:37 2006 +0300
113538
113539 mi/mipointer: deprecate functions which don't take a device
113540 Deprecate all mi pointer functions which don't take a device argument, and
113541 replace them with versions which do, in preparation for MPX.
113542
113543commit 6eab4c55890660089067da0e944256b1ed3a8c67
113544Author: Daniel Stone <daniel@fooishbar.org>
113545Date: Sun Oct 8 17:24:33 2006 +0300
113546
113547 doc/extensions: document C extension use in the X server
113548
113549commit 80cdd26581508dd17c5d0a5739cd540113996bbb
113550Author: Daniel Stone <daniel@fooishbar.org>
113551Date: Sun Oct 8 17:23:54 2006 +0300
113552
113553 mi/pointer: mark public pointer functions as deprecated
113554 Deprecate miPointer functions which don't take a device pointer. Pointer
113555 movement should be handled through GetPointerEvents, and functions which
113556 take a device as an argument (e.g. miPointerPosition) will be added.
113557
113558commit 97030b6c6b0fb6ff629ae31e483704d0a2207a53
113559Author: Daniel Stone <daniel@fooishbar.org>
113560Date: Sun Oct 8 17:07:05 2006 +0300
113561
113562 config: fix compilation
113563 Accidentally built with --disable-config, didn't notice that the previous
113564 commit to clean up the debugging broke things horribly.
113565
113566commit 14b157bdb1f2cd5feba03ba0815d7c5b2dd6633f
113567Author: Daniel Stone <daniel@fooishbar.org>
113568Date: Sun Oct 8 17:04:12 2006 +0300
113569
113570 include: actually declare DebugF
113571 DebugF is ErrorF when DEBUG is defined, else a no-op.
113572
113573commit 9e37de193f5d7412ffd8de76d5eed0158c0a3609
113574Author: Daniel Stone <daniel@fooishbar.org>
113575Date: Sun Oct 8 16:32:15 2006 +0300
113576
113577 configure.ac: reactivate warnings when building with gcc
113578 We were inadvertently stomping XSERVER_CFLAGS after adding the warnings, so
113579 move them after we do that.
113580
113581commit b559cbb1601f93cb03ea3dcfb2c5ca94ee6b73bb
113582Author: Daniel Stone <daniel@fooishbar.org>
113583Date: Sun Oct 8 16:23:14 2006 +0300
113584
113585 dix/CoreProcessKeyboardEvent: remove debugging for every key event
113586 Also change #ifdef DEBUG/ErrorF/#endif to DebugF in FixKeyState.
113587
113588commit 3ae4d250185e71a0a218c062426f92b9b1adbf05
113589Author: Daniel Stone <daniel@fooishbar.org>
113590Date: Sun Oct 8 16:20:42 2006 +0300
113591
113592 xfree86 Xinput: remove still more excessive debugging
113593 There isn't any more debugging left for input events in the XFree86 DDX.
113594
113595commit 58314756aeecbb8fb04706c3e04d98e9ac531a02
113596Author: Daniel Stone <daniel@fooishbar.org>
113597Date: Sun Oct 8 16:18:05 2006 +0300
113598
113599 GetPointerEvents: add first_valuator parameter
113600 Add a first_valuator parameter. Looks correct by inspection, but untested
113601 with first_valuator != 0 as yet.
113602
113603commit 84f5d2291c1fe92fd8358e999e909bf3aab86c98
113604Author: Daniel Stone <daniel@fooishbar.org>
113605Date: Sun Oct 8 15:30:24 2006 +0300
113606
113607 GetPointerEvents: fix relatively harmless typo
113608 Change !(cp->button || !cp->valuator) to (!cp->button || !cp->valuator).
113609
113610commit cfc3e9ede2dc83741bd38bf3df13f096ecb8adc0
113611Author: Daniel Stone <daniel@fooishbar.org>
113612Date: Sun Oct 8 15:27:52 2006 +0300
113613
113614 config: remove excessive debugging
113615
113616commit 8d8e7f8bae4099f9e90ef9aac687607dae1d32bf
113617Author: Daniel Stone <daniel@fooishbar.org>
113618Date: Sun Oct 8 15:26:54 2006 +0300
113619
113620 kdrive/input: remove excessive debugging in NIDR
113621
113622commit 22a836fafd39a8ef413826dc2c94bc5f96990e2d
113623Author: Daniel Stone <daniel@fooishbar.org>
113624Date: Sat Oct 7 14:16:51 2006 +0300
113625
113626 xfree86/loader: bump input major ABI version
113627 Bump input major ABI version to 1.0, since we removed the OS keyboard
113628 layer.
113629
113630commit 4c342246300e06bdf5c9c62cc1d2f6aa57a524db
113631Merge: 8382234 49a70c8
113632Author: Alan Coopersmith <alan.coopersmith@sun.com>
113633Date: Fri Oct 6 18:01:13 2006 -0700
113634
113635 Merge branch 'XACE-modular'
113636
113637commit 3686cd0fbf56d883f2f3b3fda11ffba1058b74e4
113638Author: Daniel Stone <daniel@fooishbar.org>
113639Date: Fri Oct 6 17:20:42 2006 +0300
113640
113641 xkb: make XkbSetControls work on all core-sending devices
113642
113643commit 7b4dc171b036107cfba87a1a16bf692b982005a5
113644Author: Daniel Stone <daniel@fooishbar.org>
113645Date: Fri Oct 6 16:26:54 2006 +0300
113646
113647 xkb: remove unused #ifndef
113648
113649commit ab56f0c5b516269bb99ae8b5f479e49e61a3af76
113650Author: Daniel Stone <daniel@fooishbar.org>
113651Date: Fri Oct 6 16:12:36 2006 +0300
113652
113653 xkb: simplify core device loop in GetKeyboardByName
113654
113655commit 4b6e2f12f7296e17b2850f36b3adcf8156125cbe
113656Author: Daniel Stone <daniel@fooishbar.org>
113657Date: Fri Oct 6 16:08:21 2006 +0300
113658
113659 xkb: make LatchLockGroup work on all core-sending devices
113660 Apply the settings to all devices sending core events, if we're working on the
113661 core keyboard.
113662
113663commit ebf9b3bbbb04acb78cdf8a84e47a96755fbfe854
113664Author: Daniel Stone <daniel@fooishbar.org>
113665Date: Fri Oct 6 14:17:59 2006 +0300
113666
113667 xkb: update all core-sending keyboards on GetKeyboardByName
113668 Update the keymaps of all keyboards which send core events on
113669 GetKeyboardByName; still a few other procedures which need this treatment.
113670
113671commit 1178796a4dff5ebf0bd9fb3cacb35be9709b41e5
113672Author: Keith Packard <keithp@neko.keithp.com>
113673Date: Thu Oct 5 22:31:35 2006 -0700
113674
113675 Add preferred modes for each output. Round vrefresh. Deliver crtc events.
113676
113677commit 23f6f08b7b5c9a4297fd223d232a7e9f45376550
113678Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
113679Date: Thu Oct 5 16:07:26 2006 -0400
113680
113681 Improve error handling, messages during initialization.
113682
113683commit de63a469dcd0a8ae98554bca540ac0106cccf2a5
113684Merge: 9c7440b 8382234
113685Author: Daniel Stone <daniel@fooishbar.org>
113686Date: Thu Oct 5 20:29:19 2006 +0300
113687
113688 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
113689
113690commit b4659faf9b455b44ac8e691cc7a8fc00a967c80b
113691Merge: c4f30c6 8382234
113692Author: Keith Packard <keithp@neko.keithp.com>
113693Date: Wed Oct 4 14:46:04 2006 -0700
113694
113695 Merge branch 'master' into randr-1.2
113696
113697commit df351f1efbcc95f94c719fcf993c480155c511e9
113698Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
113699Date: Wed Oct 4 16:23:35 2006 -0400
113700
113701 Experimental window property holding security context.
113702
113703commit c4f30c63538e1451f15ed1991439869127d9b148
113704Author: Keith Packard <keithp@guitar.keithp.com>
113705Date: Tue Oct 3 21:06:11 2006 -0700
113706
113707 Add mode origins and output options. Fix memmoves in resource free funcs.
113708
113709 Output options and mode origins both affected driver ABI. memmove mistakes
113710 were causing 'Freeing resource which isn't there' messages.
113711
113712 Prune unused non-user defined modes from available list now.
113713
113714commit 8382234a7f318057de66490299d63807cefb5201
113715Author: Alan Coopersmith <alan.coopersmith@sun.com>
113716Date: Tue Oct 3 17:49:48 2006 -0700
113717
113718 Update pci.ids to Tue 2006-10-03 daily snapshot
113719
113720commit 96edf7b853c1045d12d957a2957a11879100a2de
113721Author: Alan Coopersmith <alan.coopersmith@sun.com>
113722Date: Tue Oct 3 17:48:50 2006 -0700
113723
113724 Don't insert RCS tag into generated pci id header files
113725
113726commit 75e0cc41b201b9ceb2615e7cec0dd5b4c136b343
113727Merge: 6926776 20c0cee
113728Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
113729Date: Tue Oct 3 13:58:03 2006 -0400
113730
113731 Merge branch 'my-XACE-modular' into my-XACE-SELINUX
113732
113733commit df800d87e04ce984a8a9ab4252ac6478ab1e4426
113734Author: Alan Coopersmith <alan.coopersmith@sun.com>
113735Date: Mon Oct 2 11:28:47 2006 -0700
113736
113737 Add (void) casts to clear compiler errors about ?: results having type mismatch
113738
113739commit 9c7440bdf5a4ecd113e102004c804a2ba354c422
113740Author: Daniel Stone <daniel@fooishbar.org>
113741Date: Mon Oct 2 20:58:33 2006 +0300
113742
113743 xkb: remove the world's most staggeringly broken vendor workaround
113744 Certain versions of LynxOS needed to sleep up to five seconds for closing a
113745 pipe to actually, y'know, be useful.
113746
113747commit d7c89c7c1c8c1e110345d9d8d300adbf5fe5804a
113748Author: Daniel Stone <daniel@fooishbar.org>
113749Date: Mon Oct 2 02:15:36 2006 +0300
113750
113751 symlink-mesa.sh: expand *.{c,h}
113752
113753commit d6ea96b13e2ea01c51998c41ae2a3677bdedf61c
113754Author: Ivan Pascal <pascal@info.tsu.ru>
113755Date: Mon Oct 2 02:17:14 2006 +0300
113756
113757 xkb: fix wrapping when switching between groups
113758 Use XkbCharToInt as that's what we're doing.
113759
113760commit 3c98cebb6e954855528794fec46830f456cbdec1
113761Merge: fa1ac94 2cf1098
113762Author: Daniel Stone <daniel@fooishbar.org>
113763Date: Mon Oct 2 02:18:17 2006 +0300
113764
113765 Merge branch 'input-hotplug' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
113766
113767commit fa1ac94178cf976d4c8dae9a4dc8703303a62d4b
113768Author: Ivan Pascal <pascal@info.tsu.ru>
113769Date: Mon Oct 2 02:17:14 2006 +0300
113770
113771 xkb: fix wrapping when switching between groups
113772 Use XkbCharToInt as that's what we're doing.
113773
113774commit c31672e6aab168262bd7824a8082ecdf841fc3c0
113775Author: Daniel Stone <daniel@fooishbar.org>
113776Date: Mon Oct 2 02:15:36 2006 +0300
113777
113778 symlink-mesa.sh: expand *.{c,h}
113779
113780commit 2cf1098436d6b4382d9ed3f6b88214d37bdd8ddb
113781Author: Daniel Stone <daniel@fooishbar.org>
113782Date: Sat Sep 30 17:05:46 2006 +0300
113783
113784 dix/events, mi/eq: remove utterly ridiculous debugging
113785 Remove debugging which can cause long-lived Xorg logs to grow well above 1GB
113786 if built with --enable-debug.
113787
113788commit 20c0ceedfbce9ae5c70703900b52973917ac4cd0
113789Merge: 49a70c8 84eb2c0
113790Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
113791Date: Fri Sep 29 15:37:00 2006 -0400
113792
113793 Merge branch 'master' into my-XACE-modular
113794
113795commit 518db35ca3f569e7cb95dbddeddb93f3691de498
113796Merge: 5d99e05 84eb2c0
113797Author: Daniel Stone <daniel@fooishbar.org>
113798Date: Fri Sep 29 00:35:21 2006 +0300
113799
113800 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
113801
113802commit 5d99e05f05a42a82a9f02844df9bfebaa673759d
113803Merge: ad631af 5893375
113804Author: Daniel Stone <daniel@fooishbar.org>
113805Date: Fri Sep 29 00:35:07 2006 +0300
113806
113807 Merge branch 'input-hotplug' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
113808
113809commit ad631afcf3fbc74024cddaaacd05d38addc047a2
113810Author: Daniel Stone <daniel@fooishbar.org>
113811Date: Fri Sep 29 00:34:23 2006 +0300
113812
113813 make core keyboard/pointer map functions act on all core-sending devices
113814 Make Set{Keyboard,Modifier,Pointer}Mapping act on all devices which send core
113815 events.
113816 Change LegalModifier to accept a DeviceIntPtr.
113817
113818commit 84eb2c0a06de60e88e14bb03fabe661d7cd8f1d3
113819Author: Brian <brian@yutani.localnet.net>
113820Date: Thu Sep 28 15:09:40 2006 -0600
113821
113822 Replace hard-coded filesnames with loops (all .c and .h files).
113823 Should fix problems with Mesa adding/removing source files, for the most part.
113824 Patch by Dan Nicholson.
113825
113826commit 4bc5dc2854e33bf343cdea44a3c3b4c41f6f4145
113827Merge: cf6e968 f9542e7
113828Author: Aaron Plattner <aplattner@nvidia.com>
113829Date: Thu Sep 28 13:27:13 2006 -0700
113830
113831 Merge branch 'wfb'
113832
113833commit f9542e749544c7a3084fd72ecc6642ca3262f7c7
113834Author: Aaron Plattner <aplattner@nvidia.com>
113835Date: Thu Sep 28 13:26:54 2006 -0700
113836
113837 Restore the global xx* symbols and add them to wfbrename.h.
113838
113839 I don't think they're ever used, but it's best to just leave them alone for ABI
113840 compatibility.
113841
113842commit cf6e9687ffcc52af0d64e9098186570719a575a2
113843Author: Jeff Smith <whydoubt@yahoo.com>
113844Date: Thu Sep 28 15:34:17 2006 -0400
113845
113846 Bug #8449: Yet another Mesa symlink script resync.
113847
113848commit a5d639cd87f30f9b3d765352d27253940f33b2b7
113849Author: Daniel Stone <daniel@fooishbar.org>
113850Date: Wed Sep 27 16:15:27 2006 +0300
113851
113852 remove merge detritus
113853
113854commit 2206a92a97901977910a6e39b4174ca805f9f4a7
113855Author: Michel Dänzer <michel@tungstengraphics.com>
113856Date: Wed Aug 30 19:15:55 2006 +0200
113857
113858 Push information about cliprects of DRI windows to the DRM.
113859
113860commit 54d371e7a4ebab79a0f616669e2f601d8370cef3
113861Author: Michel Dänzer <michel@tungstengraphics.com>
113862Date: Wed Aug 30 19:12:17 2006 +0200
113863
113864 Add wrapper for new ioctl to update drawable information in the DRM.
113865
113866commit ad321fad4b9ab3a2c70cfff37ca6c8faaa5cce9c
113867Merge: 5e9d33f f7c1d94
113868Author: Daniel Stone <daniel@fooishbar.org>
113869Date: Sun Sep 24 17:56:43 2006 +0300
113870
113871 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
113872
113873commit f7c1d942416db8d0d4c5a21f5ece1ccacb926b69
113874Author: Brian <brian@yutani.localnet.net>
113875Date: Sat Sep 23 10:38:10 2006 -0600
113876
113877 Check for visual==NULL in dmxBECreateColormap() before calling XCreateColormap()
113878 to prevent potential segfault.
113879
113880commit 945b7c63946f5257d0f9b0dcf2f8f4882fb2c6f8
113881Author: Brian <brian@yutani.localnet.net>
113882Date: Sat Sep 23 10:35:25 2006 -0600
113883
113884 The fbcmap.c file used by Xdmx _must_ be compiled with XFree86Server defined.
113885 Otherwise, Xdmx generates a slew of protocol errors.
113886
113887commit 891e9c3e6cbd0869a57395b96c8e18ff522c2bb4
113888Author: Brian <brian@yutani.localnet.net>
113889Date: Sat Sep 23 10:28:24 2006 -0600
113890
113891 Replace broken DMXDBG3() with DMXDBG2()
113892
113893commit a10039a100dfe5f87e29e9cc4fa656176e0890f9
113894Author: David Nusinow <dnusinow@debian.org>
113895Date: Thu Sep 21 23:58:32 2006 -0400
113896
113897 Allow the xfree86 ddx utils to be optionally built. Patch by Eugene Konev.
113898
113899commit ce78b0cd2b1c35d60eb5683a1d00222aa4797c79
113900Author: Adam Jackson <ajax@benzedrine.nwnk.net>
113901Date: Thu Sep 21 20:42:47 2006 -0400
113902
113903 Close with Pclose() that which we open with Popen().
113904
113905commit c1655f0fd457f9bdf0857c5e0904639925bb01f1
113906Author: Aaron Plattner <aplattner@nvidia.com>
113907Date: Thu Sep 21 14:45:17 2006 -0700
113908
113909 Bug 8386: Grow parser buffers to fit an entire line if it's longer than CONFIG_BUF_LEN.
113910
113911commit b36fde9257263fa502147df37e8331184c323e14
113912Author: Keith Packard <keithp@guitar.keithp.com>
113913Date: Thu Sep 21 09:52:04 2006 -0700
113914
113915 When no mode is specified, don't validate mode-specific parameters.
113916
113917commit 8b4ed47c5d39f219866e3c72fa973c6fc4c70f18
113918Author: Drew Parsons <dparsons@debian.org>
113919Date: Thu Sep 21 22:19:44 2006 +1000
113920
113921 * Install Xprint's Xsession script to $(sysconfdir)/X11/Xsession.d
113922 * Removing outdated references to CDE and dt, rename script to
113923 92xprint-xpserverlist.
113924
113925commit 219546fd76750f358ffb6738f17b9237c58c15a6
113926Author: Keith Packard <keithp@guitar.keithp.com>
113927Date: Wed Sep 20 22:43:05 2006 -0700
113928
113929 Steal Xinerama code from SiS driver. Add missing files.
113930
113931 Provide a Xinerama implementation when DIX version isn't enabled. This
113932 version exposes each crtc as a separate 'screen' and reports the size of
113933 that patch. The extension also sends ConfigureNotify events to the root
113934 window whenever crtcs change so that applications will re-fetch xinerama
113935 information. This actually works for metacity.
113936
113937commit bde0a4c12cb393a6d7f1552b067624da1b0502ae
113938Author: Keith Packard <keithp@guitar.keithp.com>
113939Date: Wed Sep 20 19:42:34 2006 -0700
113940
113941 RRSetCrtcConfig status fix. RRGetScreenResources timestamp fix.
113942
113943 RRSetCrtcConfig was returning the wrong status values.
113944 RRGetScreenResources was always returning currentTime.
113945
113946commit d812f486a01a6276aed7b4ebd3cd8eb8ddfe10d3
113947Author: Donnie Berkholz <donnie@supernova.(none)>
113948Date: Wed Sep 20 15:39:39 2006 -0700
113949
113950 Really fix sparc on 64-bit kernel/32-bit userland.
113951
113952 Commit b3a3020fd018df8bc5a8193d36e1a1c7ae8af8ba used a sparc64 ifdef instead of
113953 sparc. But for 32-bit userland, __sparc64__ is not defined so the wrong code is
113954 used.
113955
113956commit 09f7499851bd2f2eba1e30460c61c7a82ed9e853
113957Author: Keith Packard <keithp@guitar.keithp.com>
113958Date: Wed Sep 20 13:15:20 2006 -0700
113959
113960 typo
113961
113962commit 9f870e0aa1ada238d6a0cd099996e8c47f6ba1d9
113963Author: Keith Packard <keithp@guitar.keithp.com>
113964Date: Wed Sep 20 13:14:53 2006 -0700
113965
113966 When setting output state, leave output unchanged when setting to current.
113967
113968commit d16e83413e7e06adebd629d04de57bbedd8c3765
113969Author: Aaron Plattner <aplattner@nvidia.com>
113970Date: Wed Sep 20 12:47:17 2006 -0700
113971
113972 Hide or rename more global symbols to avoid clashes with libfb.
113973
113974 Rename composeFunctions, xxSetup, and xxPrintVisuals. Hide the other xx*
113975 symbols by making them static.
113976
113977commit d08718d8fd31477e90f13b9e122504c515b46ee0
113978Author: Keith Packard <keithp@guitar.keithp.com>
113979Date: Wed Sep 20 12:05:52 2006 -0700
113980
113981 Avoid calling xalloc(0). Change rrScreenSizeSet to rrScreenSetSize.
113982
113983commit ef1f3248cb5fff0a02c0059f865c4d931eba23a6
113984Author: Keith Packard <keithp@guitar.keithp.com>
113985Date: Tue Sep 19 22:48:54 2006 -0700
113986
113987 Split out 1.0-style info and new property routines to their own files.
113988
113989commit 07112adb0802d28488de5a495aa61bb3cfc280b6
113990Author: Keith Packard <keithp@guitar.keithp.com>
113991Date: Tue Sep 19 00:46:27 2006 -0700
113992
113993 RRGetScreenResources and RRGetOutputInfo are working now.
113994
113995 Removed separate id field in RRModeRec.
113996 Pull screen subpixel order from Render extension.
113997 Implement RGetScreenResources and RRGetOutputInfo
113998
113999commit afe5e9483b352ed06075ed68a6ffa50799194e2d
114000Author: Keith Packard <keithp@guitar.keithp.com>
114001Date: Mon Sep 18 12:18:22 2006 -0700
114002
114003 RandR working with old clients and old API.
114004
114005commit 2be1ac15aee592782d7693b8de2c3815478a094e
114006Author: Keith Packard <keithp@guitar.keithp.com>
114007Date: Mon Sep 18 12:11:18 2006 -0700
114008
114009 Remove smashing of CFLAGS from server build.
114010
114011 CFLAGS is a user variable, extracted from the environment at configure time
114012 and settable by the user at build time. We must not override this variable.
114013
114014commit bf07893947cfca945598e194ed416fda6162b11c
114015Author: Keith Packard <keithp@neko.keithp.com>
114016Date: Sun Sep 17 23:03:23 2006 -0700
114017
114018 Split out RandR dispatch code from randr.c to rr*dispatch.c.
114019
114020 More disassembly to ease ongoing development.
114021
114022commit 3e745745fecef1cb59e53bde52ded311b51e1dac
114023Author: Keith Packard <keithp@neko.keithp.com>
114024Date: Sat Sep 16 23:21:37 2006 -0700
114025
114026 Split RandR implementation into separate files.
114027
114028 RandR is getting too big to live in one file; split into one file per object
114029 type (crtc, mode, screen), leaving the rest of the code in randr.c.
114030
114031 Code is slowly approaching the point where it will drop-in as a replacement
114032 for the old 1.0 implementation.
114033
114034commit d17fb9672e238a089e463ac74cc4cd3325b67e1f
114035Author: Keith Packard <keithp@neko.keithp.com>
114036Date: Sat Sep 16 21:44:42 2006 -0700
114037
114038 Start moving to new randr 1.2 definition
114039
114040commit 8dec74321d916f204f8182f1b93a65defbe50e78
114041Author: Keith Packard <keithp@neko.keithp.com>
114042Date: Mon Jul 17 14:43:07 2006 -0400
114043
114044 Successful legacy RandR API/Protocol emulation for query.
114045
114046 These changes clean up minor errors to make it possible to list the
114047 available modes for a monitor using legacy APIs in both the X server DDX and
114048 RandR protocol. Setting modes is untested, so it probably doesn't work.
114049
114050commit cab3a0145f2483fe43b5db5f5dd2076db9757fe5
114051Author: Keith Packard <keithp@neko.keithp.com>
114052Date: Mon Jul 17 01:21:11 2006 -0400
114053
114054 RandR: New data structure, old API. At least it compiles now
114055
114056commit d95c758630f4aacec339a7ec80d2c4a9d7de1e4a
114057Author: Keith Packard <keithp@neko.keithp.com>
114058Date: Sat Jul 1 19:46:38 2006 -0700
114059
114060 Preliminary RandR 1.2 work
114061
114062commit f057de4f73fa593fa3fc5f05f65b89e76273b158
114063Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114064Date: Sat Sep 16 03:49:11 2006 -0400
114065
114066 Don't install librac.a.
114067
114068 Thanks, automake.
114069
114070commit 6926776f0ecd1e8e81c5c40ccd3a97227bc44dcb
114071Merge: 6950267 49a70c8
114072Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114073Date: Fri Sep 15 15:28:13 2006 -0400
114074
114075 Merge branch 'my-XACE-modular' into my-XACE-SELINUX
114076
114077commit 49a70c8570b03aff8239324a2474918a6fbc52a0
114078Merge: d1110c5 05231e3
114079Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114080Date: Fri Sep 15 15:26:57 2006 -0400
114081
114082 Merge branch 'master' into my-XACE-modular
114083
114084commit 46af6d1e953f1eefb6edbba3d29fb9700e42c2bb
114085Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114086Date: Thu Sep 14 19:28:44 2006 -0400
114087
114088 Always believe the monitor when it reports a reduced-blanking mode.
114089
114090 CVT reduced blanking modes are typically only seen on digital connections to
114091 LCDs, but there are some monitors that report them as supported over the
114092 VGA connector too, which is perfectly legitimate, electrically speaking.
114093
114094commit 63acf18b7e4ce3a9f7deab3a9088a1c41cab0191
114095Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114096Date: Thu Sep 14 19:26:37 2006 -0400
114097
114098 In xf86MatchPciInstances, fail gracefully when there's no PCI device at all.
114099
114100 This allows the autoconfig logic to fall through sanely on non-PCI machines,
114101 which importantly includes Xen virtual machines.
114102
114103commit a8f9936f55c5364bb02e8c3187507eb1f70e2ef2
114104Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114105Date: Thu Sep 14 19:24:41 2006 -0400
114106
114107 Prefer driver-provided modes when matching name strings to modelines.
114108
114109 Well, kinda. Strictly we prefer M_T_BUILTIN strongest since those are modes
114110 where the driver has said it absolutely can't do anything else (VBE). Then
114111 we look for user-defined modes, ie, modelines from the config file. Then
114112 we consider modes reported by the monitor via EDID. Finally if nothing has
114113 matched yet we consider the default mode pool.
114114
114115 Within each of the above-mentioned classes, modes with the M_T_PREFERRED bit
114116 take priority over other modes in the same class.
114117
114118 This logic ensures that the timings sent to the monitor exactly match the
114119 timings it reported as supported, which occasionally don't match the numbers
114120 you might get for that mode from CVT or GTF.
114121
114122commit 81ef1b6d6063c20db4963abf7b7848e235aa4ebb
114123Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114124Date: Thu Sep 14 19:18:58 2006 -0400
114125
114126 Mark EDID modes as driver modes. Infer virtual size from driver modes.
114127
114128 This allows the server to guess an appropriate initial virtual size and
114129 resolution. The heuristic is to select the largest driver-reported mode
114130 that matches the monitor's physical aspect ratio. We revalidate this
114131 estimate after mode validation, since we may have filtered away all
114132 modes that would fill that size.
114133
114134 Also, the EDID preferred timing is now marked as M_T_PREFERRED as well.
114135
114136commit 43d9edd31e31b33b9da4a50d8ab05004881c8d5a
114137Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114138Date: Thu Sep 14 19:09:02 2006 -0400
114139
114140 Attempt to add the 'mouse' driver in more situations.
114141
114142 Always add a mouse driver instance configured to send core events, unless
114143 a core pointer already exists using either the mouse or void drivers. This
114144 handles the laptop case where the config file only specifies, say,
114145 synaptics, which causes the touchpad to work but not the pointing stick.
114146 We don't double-instantiate the mouse driver to avoid the mouse moving twice
114147 as fast, and we skip this logic when the user asked for a void core pointer
114148 since that probably means they want to run with no pointer at all.
114149
114150commit 739224d05eb4f356c9cab9dcb8a44a8d78287765
114151Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114152Date: Thu Sep 14 19:03:32 2006 -0400
114153
114154 Load the default module set when no Module section is given in the config.
114155
114156 Also, synchronize that list with the list for the pseudoconfig file used
114157 when starting with no config file. These really need to be better unified.
114158
114159commit beac2bf1e48e6b77dbf7d95f086abc5abcd90cf0
114160Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114161Date: Thu Sep 14 19:01:13 2006 -0400
114162
114163 Expand the default sync ranges to be large enough for 800x600@60.
114164
114165commit 71a15a7d5721073eccb3a275f353b3aa584c0d68
114166Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114167Date: Thu Sep 14 19:00:10 2006 -0400
114168
114169 Publish the raw EDID block as a property on the root window.
114170
114171 This was removed in the patch for bug #5386, but is still useful.
114172
114173commit 7939c8dfb7c7bed4febcdc12922fb2e17619ea36
114174Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114175Date: Thu Sep 14 18:57:57 2006 -0400
114176
114177 Bump the default pixel depth to 24, and default bpp to 32.
114178
114179commit 72af975f9c8de0ff6796f1ce4b76dcf841d21e99
114180Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114181Date: Thu Sep 14 18:56:34 2006 -0400
114182
114183 Fix up EDID blocks where the max pixclock exceeds the preferred mode clock.
114184
114185 Base EDID only lets you specify the maximum dotclock in tens of MHz, which
114186 is too fuzzy for some monitors. 1600x1200@60 is just over 160MHz, but if
114187 the monitor really can't handle any mode at 170MHz, then 160 is more
114188 correct. Fix up the EDID block before the driver can see it in this case,
114189 so we don't spuriously reject modes.
114190
114191commit d05e0a97bb704a4986cf638487205da759c4ce17
114192Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114193Date: Thu Sep 14 18:49:12 2006 -0400
114194
114195 Enable DPMS by default.
114196
114197commit 334f7db9f653113d5d46236911d7de2ec4173f28
114198Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114199Date: Thu Sep 14 18:46:10 2006 -0400
114200
114201 Allow hsync and vsync ranges to be overridden independently again.
114202
114203commit ced46e17777b635df9371c4cfaec3f8968b4dbcf
114204Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114205Date: Thu Sep 14 18:41:59 2006 -0400
114206
114207 Record all standard timings from EDID as modes, instead of just the first five.
114208
114209commit d89fee68d0e49211871cd9eb3893ed55c1d478a6
114210Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114211Date: Thu Sep 14 18:41:15 2006 -0400
114212
114213 Record the maximum dot clock of the monitor, and filter by it.
114214
114215commit fa8ef7166839a7435e0017683f3e3c7f7904b285
114216Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114217Date: Thu Sep 14 18:33:00 2006 -0400
114218
114219 Don't translate monitor gamma to X gamma.
114220
114221 The X gamma is used to set the output ramp of the card. Setting a 2.2 output
114222 gamma going into a 2.2 monitor gives an effective gamma of 4.84, which is
114223 very much not what you want.
114224
114225commit 511c60bc7399b07c267d686a969880e5ec92408a
114226Author: Luc Verhaegen <libv@skynet.be>
114227Date: Thu Sep 14 18:30:36 2006 -0400
114228
114229 Bug #5386: Synthesize modelines from EDID data.
114230
114231commit 05231e336db8f959c15dda518641976f061df1a6
114232Author: Ian Romanick <idr@us.ibm.com>
114233Date: Thu Sep 14 14:13:39 2006 -0700
114234
114235 Use correct opcodes for GLX_EXT_texture_from_pixmap.
114236
114237 Regenerate from glX_API.xml 1.3 from Mesa. The glproto package and libGL
114238 (from Mesa) must also be updated.
114239
114240commit 0a62840e2ce25e5c2554e7e5ab4c9c5b96899e2d
114241Author: Bill Nottingham <notting@redhat.com>
114242Date: Wed Sep 13 15:40:23 2006 -0700
114243
114244 Bug 7641: fix comment written to Xorg.conf (s/VertSync/VertRefresh/)
114245
114246 X.Org Bugzilla #7641 <https://bugs.freedesktop.org/show_bug.cgi?id=7641>
114247 Patch #6349 <https://bugs.freedesktop.org/attachment.cgi?id=6349>
114248
114249commit 182e5e0f4ba4c98a34bc52bdf4032ba315fe80ad
114250Author: Drew Parsons <dparsons@debian.org>
114251Date: Tue Sep 12 14:30:46 2006 +1000
114252
114253 Xprint: revert installation of /etc/X11/Xsession.d/cde_xsessiond_xprint.sh
114254 pending resolution of #8232.
114255
114256commit 594d4019c613b0f4bf8f48cc074ecc3c8366f1d7
114257Author: Tilman Sauerbeck <tilman@code-monkey.de>
114258Date: Tue Sep 12 01:15:40 2006 +0200
114259
114260 transformIsIdentity() now doesn't accept a zero matrix as the identity.
114261
114262 Added a non-zero test for one of the diagonal values.
114263
114264commit fc30370d14125f86ee1192890a184881fa139546
114265Author: Tilman Sauerbeck <tilman@code-monkey.de>
114266Date: Mon Sep 11 19:43:09 2006 +0200
114267
114268 Bug #8226: Fixed SetPictureTransform()'s handling of the argument matrix.
114269
114270 It now recognizes scaled variants of the identity matrix, too.
114271
114272commit 2b357e9a2f9038cf9cd07da908e3103a3d0965c9
114273Author: Donnie Berkholz <dberkholz@gentoo.org>
114274Date: Sun Sep 10 22:17:20 2006 -0700
114275
114276 If we're installing libxf86config, install headers needed to build against it.
114277
114278commit 58933757862c458e2da39bd464e50d9c0e41b313
114279Author: Zephaniah E. Hull <warp@agamemnon.b5>
114280Date: Sun Sep 10 15:50:51 2006 -0400
114281
114282 Warning fix, and a syntax fix in a #if 0 section of code.
114283
114284commit 0a3740a0000191e3039fe183ae51b938d0548340
114285Author: Zephaniah E. Hull <warp@agamemnon.b5>
114286Date: Sun Sep 10 15:49:25 2006 -0400
114287
114288 Typo correction, 'i' is not a '1', so no longer crash on some Xi requests.
114289
114290commit 8d709f0280b458515b32c2b87938749428e5c149
114291Author: Zephaniah E. Hull <warp@agamemnon.b5>
114292Date: Sun Sep 10 15:48:35 2006 -0400
114293
114294 Remove a merge artifact so that we can compile.
114295
114296commit b3a3020fd018df8bc5a8193d36e1a1c7ae8af8ba
114297Author: Jesse Barnes <jbarnes@nietzche.virtuousgeek.org>
114298Date: Sun Sep 10 11:13:18 2006 -0700
114299
114300 the new PCI mapping routines are broken on sparc64 (in fact they look
114301 broken for any 32 bit X server running on a 64 bit kernel) so #ifdef
114302 them out for now. the PCI rework tree will make all this crap go away,
114303 so I think we can tolerate the extra #ifdef for the next release.
114304
114305commit 60db190ecfce52cbfa888c0af3210634f9186bed
114306Merge: 5e9d33f 6525610
114307Author: Zephaniah E. Hull <warp@agamemnon.b5>
114308Date: Sun Sep 10 03:49:17 2006 -0400
114309
114310 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
114311
114312commit 65256109bb8f5a26704ed960e1dd113981df5787
114313Author: Drew Parsons <dparsons@debian.org>
114314Date: Sun Sep 10 17:40:37 2006 +1000
114315
114316 * Define XPSERVERLIST with `/etc/init.d/xprint get_xpserverlist`
114317 instead of `/bin/sh /etc/init.d/xprint get_xpserverlist`
114318 - allows the initscript to set its own different shell under #!
114319 - allows disabling of XPSERVERLIST by making the script non-executable
114320 * Allow files to be installed by using dist_*_DATA instead of EXTRA_DIST.
114321 Also, use dist_*_SCRIPTS to install scripts.
114322 * Fix minor typos in man pages.
114323
114324commit 6950267dd690ef8e29b1c32a157dd64c9b79c06d
114325Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114326Date: Fri Sep 8 15:31:18 2006 -0400
114327
114328 Add SELinux extension configure-time support.
114329
114330commit fb34c02861ab3629c1c85c156e73b158518db7c7
114331Merge: 9c503f0 d1110c5
114332Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114333Date: Fri Sep 8 15:30:12 2006 -0400
114334
114335 Merge branch 'my-XACE-modular' into my-XACE-SELINUX
114336
114337commit d1110c5c83a7f439158f369ab2f3ae614fa9d2a5
114338Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114339Date: Fri Sep 8 15:28:48 2006 -0400
114340
114341 Generalize the handling of configuration files that ship with extensions.
114342
114343commit 9deb579dc9366590203afe0576bf88643ab36c89
114344Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114345Date: Fri Sep 8 15:25:17 2006 -0400
114346
114347 Zero out newly created ExtensionEntry structures, but only after the
114348 devPrivates have been initialized.
114349
114350commit cec392656cda1c938d5462e1949e6eef489f9168
114351Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114352Date: Fri Sep 8 15:24:23 2006 -0400
114353
114354 Zero out newly allocated ClientRec structures.
114355 This is required to initialize the devPrivates to a known state.
114356
114357commit 0fba09cdfcc78161f5c92bef6cca53e5309656bd
114358Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114359Date: Fri Sep 8 15:23:06 2006 -0400
114360
114361 Include dix-config.h.
114362
114363commit c93877100eb98647c5b6b8556730d54677f730b6
114364Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114365Date: Fri Sep 8 15:21:57 2006 -0400
114366
114367 Don't need to allocate memory now that devPrivates are being used.
114368
114369commit 9c503f09ce78d952d0ece77c424e42b6df3fa9ad
114370Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114371Date: Fri Sep 8 15:17:01 2006 -0400
114372
114373 Add SELinux extension to the module/extension loader.
114374
114375commit 9aa44e3e4c321f42d8e64f83c7f0932470593c26
114376Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114377Date: Fri Sep 8 15:15:19 2006 -0400
114378
114379 Add SELinux extension configure-time support.
114380
114381commit 150eabc4c8a08c81c48493583f922a1240b7e91c
114382Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114383Date: Fri Sep 8 15:11:04 2006 -0400
114384
114385 Add SELinux extension source files.
114386
114387commit 86450998da616e3d00d4d6293acc35eccc2061e7
114388Author: Kristian Høgsberg <krh@redhat.com>
114389Date: Thu Sep 7 15:35:16 2006 -0400
114390
114391 Fix AIGLX VT switching.
114392
114393 See https://bugs.freedesktop.org/show_bug.cgi?id=7916
114394
114395 There may be a simpler, less intrusive fix that involves just rearranging
114396 DRI locking between 2D and 3D drivers around VT switch.
114397
114398commit 5e9d33fe87f9d24e55c468d4b2bb761c9efdb26a
114399Merge: 629798c 64479ff
114400Author: Daniel Stone <daniel@fooishbar.org>
114401Date: Thu Sep 7 15:43:31 2006 +0300
114402
114403 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
114404
114405commit 629798c73ad76a77fa6a55bc6403fd9b95ade2bb
114406Author: Daniel Stone <daniel@fooishbar.org>
114407Date: Thu Sep 7 15:17:04 2006 +0300
114408
114409 XkbCopyKeymap/SrvXkbCopyKeymap: free geom harder, add cheery comments
114410 Unconditionally free geometry when copying the keymap (so we have none on
114411 core, oh well), add a couple of heartening comments.
114412
114413commit 64479fffa22581cc7d753065c33eda5520b7db9a
114414Author: Ian Romanick <idr@us.ibm.com>
114415Date: Wed Sep 6 16:13:21 2006 -0700
114416
114417 Remove prototypes and externs for non-existant functions and variables.
114418
114419commit a0179281a6522ec59830e8f2549633741bc56e10
114420Author: Ian Romanick <idr@us.ibm.com>
114421Date: Wed Sep 6 15:45:48 2006 -0700
114422
114423 Remove prototypes for non-existant functions.
114424
114425commit 8356be492c6b46abdffa08b13836571ed872e16f
114426Author: Michel Dänzer <michel@tungstengraphics.com>
114427Date: Wed Sep 6 15:20:55 2006 +0200
114428
114429 Make sure _XSERVER64 is defined when it should be and gets tested.
114430
114431commit f6ce0839ba5b73247097826d28f7388fe248ec0c
114432Author: Michel Dänzer <michel@tungstengraphics.com>
114433Date: Wed Sep 6 13:18:02 2006 +0200
114434
114435 Fix #include paths for fontcacheproto headers.
114436
114437commit f39fd4242902eaa862321d39337f429dd14ebacf
114438Author: Aaron Plattner <aplattner@nvidia.com>
114439Date: Tue Sep 5 15:23:54 2006 -0700
114440
114441 (unsigned long)(1 << 31) = bad news on x86_64.
114442 (cherry picked from 410e5b1d738ba47b36778e6cbed44023a27ce259 commit)
114443
114444commit 410e5b1d738ba47b36778e6cbed44023a27ce259
114445Author: Aaron Plattner <aplattner@nvidia.com>
114446Date: Tue Sep 5 15:23:54 2006 -0700
114447
114448 (unsigned long)(1 << 31) = bad news on x86_64.
114449
114450commit 0b81fccd2ee4e054e5cffb739de07460ff2c13f7
114451Merge: 20c4ac6 c281351
114452Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114453Date: Tue Sep 5 18:03:25 2006 -0400
114454
114455 Merge branch 'master' into my-XACE-modular
114456
114457 Conflicts:
114458
114459 configure.ac
114460
114461commit c2813514cf7b1a36caa848cbc2ceef99cf2eb769
114462Author: Ian Romanick <idr@us.ibm.com>
114463Date: Thu Aug 31 15:36:13 2006 -0700
114464
114465 Add missing file from previous commit.
114466
114467commit 0f9cfb2f752a9010ff07f4b2bd891db0cc30b8e6
114468Author: Ian Romanick <idr@us.ibm.com>
114469Date: Thu Aug 31 13:54:10 2006 -0700
114470
114471 Implement GLX_SGI_swap_control.
114472
114473 Regenerate from glX_API.xml 1.2. Add infrastructure to support
114474 GLX_SGI_swap_control for AIGLX when the DRI driver enables it. Tested
114475 with R300.
114476
114477commit a9ef5862919313582f72fc0cfb5ab0af4df6507e
114478Author: Ian Romanick <idr@us.ibm.com>
114479Date: Thu Aug 31 13:47:50 2006 -0700
114480
114481 Fix problems with vertex program protocol
114482
114483 There were two sets of bugs in the vertex program (ARB and NV)
114484 protocol. First, several of the ARB functions were missing the
114485 'doubles_in_order="true"' annotation. Second, after the ARB decided
114486 that glVertexAttrib*ARB functions must not alias fixed-function state
114487 for GLSL, Nvidia re-assigned GLX protocol opcodes for
114488 glVertexAttrib*NV (circa Septeber 2004). For some reason gl_API.xml
114489 was never updated to reflect this, and the updated version of the
114490 GL_NV_vertex_program spec never made into the registry.
114491
114492 This is just a server-side regeneration from gl_API.xml version 1.68.
114493
114494commit 69d5becce4ca2cfc8f8de53672ed54a47de62164
114495Author: Matthew Allum <mallum@openedhand.com>
114496Date: Thu Aug 31 17:30:24 2006 +0100
114497
114498 Fix previous commit breaking other kdrives pulling in fbdev.a
114499
114500commit fd609956f27d76ee76ac8623787f0fc8633a5546
114501Author: Matthew Allum <mallum@openedhand.com>
114502Date: Thu Aug 31 17:18:57 2006 +0100
114503
114504 Add framebuffer device command line switch for Xfbdev
114505
114506commit 2fb7b8795a9a36cce61f6449f6ca26ffd1b071f0
114507Author: Ian Romanick <idr@us.ibm.com>
114508Date: Tue Aug 29 16:35:32 2006 -0700
114509
114510 Minor extension tweaks.
114511
114512 GLX_EXT_texture_from_pixmap should always be enabled.
114513 GLX_SGI_video_sync is only for direct rendering and should never
114514 appear in the server's string.
114515
114516commit 1c8851ad491dd02d1c79e620b46384956838ed42
114517Merge: d59b52f 5ddbf4b
114518Author: Ian Romanick <idr@us.ibm.com>
114519Date: Tue Aug 29 16:34:04 2006 -0700
114520
114521 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
114522
114523commit d59b52fc08f2d80b38993e383e61c3eeb0bb0763
114524Author: Ian Romanick <idr@us.ibm.com>
114525Date: Tue Aug 29 14:40:13 2006 -0700
114526
114527 Make sure unsupported extensions are disabled.
114528
114529 GLX protocol isn't supported for GLX_SGI_swap_control or
114530 GLX_SGI_video_sync. Remove them from the list of available extensions
114531 until they are supported.
114532
114533commit db6d04d4b87fd9b6409a3ddf0479a88440c2eda1
114534Author: Ian Romanick <idr@us.ibm.com>
114535Date: Tue Aug 29 14:35:08 2006 -0700
114536
114537 Add support for AIGLX drivers to enable GLX extensions that they support.
114538
114539commit 5ddbf4bcd46fe0d3d682668c2748c712fea410ae
114540Author: Matthew Allum <mallum@openedhand.com>
114541Date: Tue Aug 29 22:07:15 2006 +0100
114542
114543 Re-add support for tslib (1.0 release) and Xcalibrate extension.
114544
114545commit 4524a2bf6f22c871ed109b027a065f0262137dc5
114546Author: Daniel Stone <daniel@fooishbar.org>
114547Date: Tue Aug 29 23:49:26 2006 +0300
114548
114549 configure.ac: move tslib from KDRIVE_PURE_LIBS to KDRIVE_LIBS
114550 Yeah. That was dumb.
114551
114552commit 942b4369990a255257f66835caf8671432c405a3
114553Merge: 77d315b 393dc0a
114554Author: Ian Romanick <idr@us.ibm.com>
114555Date: Tue Aug 29 13:30:37 2006 -0700
114556
114557 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
114558
114559commit 77d315bd2fd2f9014c831d313efbe5821189177c
114560Author: Ian Romanick <idr@us.ibm.com>
114561Date: Tue Aug 29 13:30:20 2006 -0700
114562
114563 Remove __glXNoSuchRenderOpcode because it is no longer used.
114564
114565commit 260c3f32b69c98f8fc5360f860f69d32c19f04a7
114566Author: Daniel Stone <daniel@fooishbar.org>
114567Date: Tue Aug 29 23:18:12 2006 +0300
114568
114569 configure.ac: fix XSDL test
114570
114571commit 393dc0a3388d56186181b2bd9bcc1d267747e709
114572Author: Daniel Stone <daniel@fooishbar.org>
114573Date: Tue Aug 29 22:53:54 2006 +0300
114574
114575 kdrive: remove @KDRIVE_LIBS@ from Xfoo_DEPENDENCIES
114576
114577commit cff23616fe45e10c6786a303c8dcfc0a80463a53
114578Author: Daniel Stone <daniel@fooishbar.org>
114579Date: Tue Aug 29 22:44:09 2006 +0300
114580
114581 configure.ac: allow disabling of XSDL
114582
114583commit 89d272bb183e85715d8e6047929fb2d912033d82
114584Author: Daniel Stone <daniel.stone@nokia.com>
114585Date: Tue Aug 29 15:05:31 2006 +0300
114586
114587 [PATCH] kdrive/linux keyboard: remove more debugging spew
114588
114589commit bd6f539ff9409aa7d9056fabe120b457b0a15997
114590Author: Daniel Stone <daniel.stone@nokia.com>
114591Date: Tue Aug 29 13:21:58 2006 +0300
114592
114593 [PATCH] kdrive/linux keyboard: silence excessive debugging noise
114594
114595commit 5436fce09003e20744a388fa4ae49007c9cf8ede
114596Author: Daniel Stone <daniel.stone@nokia.com>
114597Date: Tue Aug 29 13:21:40 2006 +0300
114598
114599 [PATCH] GetKeyboardValutorEvents: be even more careful
114600
114601 Don't accept devices without a keyboard feedback class.
114602
114603commit 0eb7299f445455a7bcacf2410e83227b23259675
114604Author: Daniel Stone <daniel.stone@nokia.com>
114605Date: Tue Aug 29 13:19:12 2006 +0300
114606
114607 [PATCH] XkbCopyKeymap: still more range fixes
114608
114609 Make sure we don't stomp preserve if it doesn't already exist, and fix a
114610 couple of range-related thinkos in level name copying.
114611
114612commit 7fa3383e3c8eea7d1eb0e556393f2431cf8e6ed2
114613Merge: 8d77d44 ebbdc13
114614Author: Daniel Stone <daniel@fooishbar.org>
114615Date: Tue Aug 29 15:16:01 2006 +0300
114616
114617 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
114618
114619commit ebbdc1342a243b301723390696f742dc91f59764
114620Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114621Date: Mon Aug 28 18:17:32 2006 -0400
114622
114623 Remove calls to LoaderCheckUnresolved(), since it's now a stub.
114624
114625commit 8d77d44fda3aacbae62864a3620e09095b79e92d
114626Merge: d6f36bd 2fde560
114627Author: Daniel Stone <daniel@fooishbar.org>
114628Date: Sun Aug 27 23:08:49 2006 +0300
114629
114630 Merge branch 'origin' into input-hotplug
114631
114632commit 20c4ac6e038607ebbf6c04639670514c016d8597
114633Merge: 13c6713 8d4f21a
114634Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114635Date: Fri Aug 25 18:49:46 2006 -0400
114636
114637 Merge branch 'my-XACE-SELINUX' into my-XACE-modular
114638
114639commit 13c6713c82763a85c725c998b37ad02156d803ba
114640Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114641Date: Fri Aug 25 18:17:01 2006 -0400
114642
114643 Add four new XACE hooks: auditing, key event notification, window init
114644
114645commit 2fde560bbb9c1148f26fd969dc30c4e736672b7c
114646Author: Ian Romanick <idr@umwelt.(none)>
114647Date: Fri Aug 25 13:01:51 2006 -0700
114648
114649 Enable GL_EXT_texture_filter_anisotropic and GL_EXT_blend_equation_separate.
114650
114651 Re-generate from gl_API.xml 1.65. This provides the missing bits for
114652 GL_EXT_texture_filter_anisotropic and GL_EXT_blend_equation_separate.
114653 Enable those extensions.
114654
114655commit e2d529963ed40b5f113cf82c17809d241cd4aac1
114656Author: Ian Romanick <idr@umwelt.(none)>
114657Date: Fri Aug 25 12:05:16 2006 -0700
114658
114659 Enable vertex and fragment programs.
114660
114661 Implement glGetProgramStringARB and glGetProgramStringNV. With these
114662 functions implemented, GL_ARB_{vertex,fragment}_program,
114663 GL_NV_{vertex,fragment}_program, and related extensions can be enabled.
114664
114665commit 21291d6ca7188e5733ed0c93215ee0f1f0f90cc9
114666Merge: 1a5561c cd2da4e
114667Author: Ian Romanick <idr@umwelt.(none)>
114668Date: Fri Aug 25 09:34:21 2006 -0700
114669
114670 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
114671
114672commit d6f36bd28009881ef7f7a20cdadb3808d808ed97
114673Author: Daniel Stone <daniel@fooishbar.org>
114674Date: Fri Aug 25 12:43:17 2006 +0300
114675
114676 xfree86/parser: use 'kbd' driver when 'keyboard' specified
114677 Now that we've completely ditched the old driver, we should probably make a
114678 best-effort attempt to keep configs working.
114679
114680commit 7c4167f0d6b33c9c602b04fcfd246fd3aeddd709
114681Merge: 393f834 cd2da4e
114682Author: Daniel Stone <daniel@fooishbar.org>
114683Date: Fri Aug 25 11:15:33 2006 +0300
114684
114685 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
114686
114687commit 393f8347edcccfc24d8902a86ea9def7ada2537b
114688Author: Daniel Stone <daniel@fooishbar.org>
114689Date: Fri Aug 25 10:46:32 2006 +0300
114690
114691 xorgconf.cpp: kbd, not keyboard
114692
114693commit c33e39c86be2010b169ffbd8adbe53b93222dc5f
114694Author: Ian Romanick <idr@umwelt.(none)>
114695Date: Thu Aug 24 20:33:57 2006 -0700
114696
114697 Finish support for GL_ARB_texture_compression.
114698
114699 Fill in __glXDisp_GetCompressedTexImageARB and
114700 __glXDispSwap_GetCompressedTexImageARB to finish support for
114701 GL_ARB_texture_compression. With this extension (and the related
114702 compression extensions), the server-side GLX supports all of the
114703 protocol for GL 1.4. w00t!
114704
114705 The bad news is that this has received only minimal testing, and Mesa
114706 does not contain any good tests for GL_ARB_texture_compression.
114707
114708commit cd2da4e41eae233b50f8830d9a8f5d1d916a5a1b
114709Author: Ian Romanick <idr@umwelt.(none)>
114710Date: Thu Aug 24 18:00:16 2006 -0700
114711
114712 Remove GL/glx/g_disptab.c, GL/glx/g_disptab_EXT.c, and
114713 GL/glx/g_disptab_EXT.h. Unfortunately GL/glx/g_disptab.h has to be
114714 kept around a bit longer.
114715
114716commit a29e6dd2d2d45c18c52737bb3b7945aafcea5032
114717Author: Ian Romanick <idr@umwelt.(none)>
114718Date: Thu Aug 24 17:58:52 2006 -0700
114719
114720 Add some missing bits of GL_SGI_color_table.
114721
114722commit ae608b2071d882966e9c7ede71f846b1ecec0b23
114723Merge: 2c86527 b879356
114724Author: Ian Romanick <idr@umwelt.(none)>
114725Date: Thu Aug 24 14:56:33 2006 -0700
114726
114727 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
114728
114729commit 2c865277fe1d056981d1020e1af001d2319252c0
114730Author: Ian Romanick <idr@umwelt.(none)>
114731Date: Thu Aug 24 14:54:49 2006 -0700
114732
114733 Regenerate from gl_API.xml 1.63. Enable extensions.
114734
114735 gl_API.xml 1.63 corrects some problems with GLX protocol for
114736 GL_EXT_paletted_texture and GL_SGI_color_table. Regenerate from that
114737 file, and enable those extensions and GL_EXT_shared_texture_palette.
114738
114739commit 7d5de5c6657304246473d7ddd5c29bb0c7a3bc34
114740Author: Ian Romanick <idr@umwelt.(none)>
114741Date: Thu Aug 24 14:49:46 2006 -0700
114742
114743 Regenerate from gl_API.xml 1.62. Functions move, no real changes.
114744
114745commit 3a36b0a24aa9e9e238faa7f00100f59800f5142b
114746Merge: db1ab1b b879356
114747Author: Daniel Stone <daniel@fooishbar.org>
114748Date: Thu Aug 24 23:35:28 2006 +0300
114749
114750 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
114751
114752commit db1ab1bdb2f79eca593fe247056309a16ebd29c6
114753Author: Daniel Stone <daniel@fooishbar.org>
114754Date: Thu Aug 24 23:33:59 2006 +0300
114755
114756 XkbCopyKeymap: fix various range issues
114757 Fix a bunch of range issues caused by incorrect assumptions (e.g. that the
114758 design was at least halfway sensible), and copy types by hand, instead of
114759 just blindly memcpy()ing the lot, since it itself cleverly contains a ton
114760 of allocated pointers.
114761
114762commit 5fb8d947bb88d715b9b236342885c445cb5a9387
114763Author: Daniel Stone <daniel@fooishbar.org>
114764Date: Thu Aug 24 23:16:43 2006 +0300
114765
114766 configure.ac: more thinkos
114767 Fix auto tests for vidmode and xf86dga. I win at life.
114768
114769commit 4e37c07ba6e5d299d4f8922dc6cf054c814f7baf
114770Author: Daniel Stone <daniel@fooishbar.org>
114771Date: Thu Aug 24 23:16:17 2006 +0300
114772
114773 config: clean up debugging messages, make failure to acquire name fatal
114774 Bomb with FatalError when we can't acquire the bus and name.
114775 Clean up a bunch of debugging ErrorFs to be hidden behind #ifdef DEBUG.
114776
114777commit b879356ce96929d02bcb75b9aa24b17ac7e28125
114778Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114779Date: Thu Aug 24 15:50:15 2006 -0400
114780
114781 More #ifdef USE_DEPRECATED_KEYBOARD_DRIVER.
114782
114783commit 4ed311cf1c29090c53e474a3001c5702ff8409df
114784Merge: 73e58ad b29b236
114785Author: Matthias Hopf <mhopf@suse.de>
114786Date: Thu Aug 24 20:17:10 2006 +0200
114787
114788 Merge branch 'master' of git://anongit.freedesktop.org/git/xorg/xserver
114789
114790commit b29b236d88789fd45d823a55dbedb393bb134c5b
114791Author: Lukáš Turek <8an@centrum.cz>
114792Date: Thu Aug 24 15:57:09 2006 +0200
114793
114794 Adapt to Mesa header name change.
114795
114796commit ce4a0a4ddafd3833d7025f83ed3729915c8aba70
114797Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
114798Date: Thu Aug 24 13:56:22 2006 +0100
114799
114800 Apply patch in bug #7919, blit improvements in
114801 multiwindow mode for Xming/CygwinX
114802
114803commit 866ca1f929c95689bac9f0a0b3478f7b4d77214b
114804Author: Daniel Stone <daniel@fooishbar.org>
114805Date: Thu Aug 24 15:46:44 2006 +0300
114806
114807 configure.ac: fix XF86VidMode test
114808
114809commit 2b06c69c8feaf3bdc065635ee711efa45b3033b3
114810Author: Daniel Stone <daniel@fooishbar.org>
114811Date: Thu Aug 24 14:51:26 2006 +0300
114812
114813 GKVE: pass correct arguments to XkbCopyKeymap
114814 Fix horrendous thinko. Indicators now work perfectly.
114815
114816commit 4adf9af313c9f63b6ad734e174efe1d36ddb5813
114817Merge: 33af05d 67bd672
114818Author: Daniel Stone <daniel@fooishbar.org>
114819Date: Thu Aug 24 10:59:33 2006 +0300
114820
114821 Merge branch 'master' into input-hotplug
114822
114823commit 67bd672c880869ef625ae0c0163c3ec1eba46abf
114824Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
114825Date: Thu Aug 24 08:47:06 2006 +0100
114826
114827 Fix typo
114828
114829commit 733c4beb16c2c4ad9e9a4ea9a85b09fc5062a775
114830Author: David Nusinow <dnusinow@debian.org>
114831Date: Wed Aug 23 22:39:42 2006 +0000
114832
114833 Add xorg.conf IgnoreABI option which does the same thing as -ignoreABI
114834
114835commit b983773d446cef6a0948ca264ed48126e404ae9a
114836Merge: 0623d36 d9a8656
114837Author: Ian Romanick <idr@umwelt.(none)>
114838Date: Wed Aug 23 17:16:50 2006 -0700
114839
114840 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
114841
114842commit 0623d3643fc28ebc514b2ca872c985d0cf0c753a
114843Author: Ian Romanick <idr@umwelt.(none)>
114844Date: Wed Aug 23 17:16:02 2006 -0700
114845
114846 Fix the sorting of the extension string. Add a few extensions that
114847 are supported by the new code. A few of these were actually supported
114848 before but weren't advertised.
114849
114850commit 5d2caacff570dd68bb3fb05e776e02515b2a9da0
114851Author: Ian Romanick <idr@umwelt.(none)>
114852Date: Wed Aug 23 16:47:00 2006 -0700
114853
114854 Refector __glXDisp_Render and __glXDispSwap_Render to DoRender.
114855 Refector __glXDisp_RenderLarge and __glXDispSwap_RenderLarge to
114856 DoRenderLarge.
114857
114858commit 866bb3f34046045c9fa0744db1d76e035b3da9c7
114859Author: Ian Romanick <idr@umwelt.(none)>
114860Date: Wed Aug 23 16:41:53 2006 -0700
114861
114862 Memo to myself: Whenever a Makefile.am changes, autogen.sh must be
114863 re-run. This is especially true if the change is to remove a source
114864 file.
114865
114866 Fix RenderLarge to actually use the new protocol decode tables.
114867
114868commit d9a86566c21afd7985673f3ed851b055d9dac46f
114869Author: Alan Coopersmith <alan.coopersmith@sun.com>
114870Date: Wed Aug 23 16:15:19 2006 -0700
114871
114872 Add LOCALCONN to dix-config.h template for xtrans
114873
114874commit f6fd7d8f8393f93705e76b2b2777a0d9bcafa991
114875Author: Ian Romanick <idr@umwelt.(none)>
114876Date: Wed Aug 23 16:05:37 2006 -0700
114877
114878 Convert protocol decode tabels for Render and RenderLarge to use nice,
114879 compact N-way search trees generated by scripts in Mesa.
114880
114881commit 7ae82b5fc8721be78b43a322bbf2c46aac08b8cf
114882Author: Ian Romanick <idr@umwelt.(none)>
114883Date: Wed Aug 23 16:00:48 2006 -0700
114884
114885 Fix __glXDispatchInfo::dispatch_functions and
114886 __glXDispatchInfo::size_table. dispatch_functions had the const in
114887 the wrong place, and size_table was declared as an array of two
114888 pointers to int_fast16_t instead of a pointer to an array of 2
114889 int_fast16_t. cdecl to the rescue!
114890
114891commit 39a620d17809dc71fb5ad61a955fe3c442f90a05
114892Author: Ian Romanick <idr@umwelt.(none)>
114893Date: Wed Aug 23 14:24:34 2006 -0700
114894
114895 Rename __glXDrawArraysSize to __glXDrawArraysReqSize. This makes its
114896 name match the pattern of all the other functions in
114897 __glXRenderSizeTable.
114898
114899commit 86406455f0e5fc977431948611e9bb5fda1e1d46
114900Author: Ian Romanick <idr@umwelt.(none)>
114901Date: Wed Aug 23 13:30:59 2006 -0700
114902
114903 Re-generated files after a fix to glX_API.xml (in Mesa).
114904
114905commit d7a7f12361d31001bbd9394a57de029ef0b934b8
114906Author: Ian Romanick <idr@umwelt.(none)>
114907Date: Wed Aug 23 13:30:13 2006 -0700
114908
114909 Convert protocol decode tables for Single, VendorPrivate, and
114910 VendorPrivateWithReply message to use nice, compact N-way search trees
114911 generated by scripts in Mesa.
114912
114913 The Render protocol decode tables are next...
114914
114915commit bdec9680fa74dd23cf319d09af1940f8cf71a5b1
114916Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114917Date: Wed Aug 23 14:43:23 2006 -0400
114918
114919 Make sure Composite is never enabled for Xnest.
114920
114921commit 9f2a108051aad9b024ab737b45fc12290a113e37
114922Author: Adam Jackson <ajax@benzedrine.nwnk.net>
114923Date: Wed Aug 23 14:38:34 2006 -0400
114924
114925 Make 'Xvfb -render' also disable Composite, lest we segfault on startup.
114926
114927commit 33af05d58f1f4f021036e9ce4b60fd76dbaebe73
114928Author: Daniel Stone <daniel@fooishbar.org>
114929Date: Wed Aug 23 19:05:50 2006 +0300
114930
114931 XkbCopyKeymap: use correct range for MapNotify
114932 We haven't copied {min,max}_key_code by the time the notifies run, so use
114933 src instead of dst to determine number of keys, et al.
114934
114935commit 6323a11d0db4d3cf0317af83f0362730142f5325
114936Author: Daniel Stone <daniel@fooishbar.org>
114937Date: Wed Aug 23 18:53:04 2006 +0300
114938
114939 XkbCopyKeymap: optionally send NewKeyboardNotify/MapNotify events
114940 Optionally send a NewKeyboardNotify or MapNotify event when copying the
114941 keymap; modify GetKeyboardValuatorEvents to make use of this.
114942
114943commit 728fbadd16a748b45c80bc2c65c46f82cf803578
114944Author: Daniel Stone <daniel@fooishbar.org>
114945Date: Wed Aug 23 14:33:59 2006 +0300
114946
114947 gitignore: ignore vi swap files
114948
114949commit 8f8487ff997670a4af0293fed77ff920cfc39fb1
114950Author: Daniel Stone <daniel@fooishbar.org>
114951Date: Wed Aug 23 14:33:41 2006 +0300
114952
114953 xkb/gkve: copy XKB map, not pointer-assign
114954 Write a new function to copy an XKB map (does everything but geometry at
114955 the moment), and use that instead of nasty pointer assignments.
114956
114957commit 52ba722e4c89c052609b4fc62e965d92778aa2dd
114958Merge: 9138d5a 0554125
114959Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
114960Date: Mon Aug 21 18:49:31 2006 -0400
114961
114962 Merge branch 'XACE-modular' into my-XACE-modular
114963
114964commit 05541259bdb0dfaab015a01caa3722b7a1b782e2
114965Merge: c2535f6 a1ac044
114966Author: Alan Coopersmith <alan.coopersmith@sun.com>
114967Date: Mon Aug 21 13:07:41 2006 -0700
114968
114969 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into XACE-modular
114970
114971commit a56b98bb047003a05e26ca9365c212a2da7ac200
114972Author: Daniel Stone <daniel.stone@nokia.com>
114973Date: Fri Aug 18 18:03:41 2006 +0300
114974
114975 dix: enable null root cursor
114976
114977 Enable a blank root cursor, selectable with --enable-null-root-cursor at
114978 configure time.
114979
114980commit 0704bb298cc826cd117815898c6bc015a693c2c9
114981Merge: c140369 a1ac044
114982Author: Daniel Stone <daniel@fooishbar.org>
114983Date: Fri Aug 18 17:30:14 2006 +0300
114984
114985 Merge branch 'master' into input-hotplug
114986
114987commit a1ac0440bba690368aa4226468ce571be1a09d95
114988Author: Daniel Stone <daniel@fooishbar.org>
114989Date: Fri Aug 18 17:30:00 2006 +0300
114990
114991 dix: fix whiteroot thinko
114992 Note to self: run git update-index _after_ testing, not just before.
114993
114994commit c14036977fef7b8787c0b68f5262fa0b6a2834f5
114995Author: Daniel Stone <daniel@fooishbar.org>
114996Date: Fri Aug 18 17:24:34 2006 +0300
114997
114998 input.h: add InitCoreDevices prototype
114999
115000commit 1c2cb30cd88ba4453f9da339025f8ff39f7f5412
115001Merge: 633b6a6 70ddd0f
115002Author: Daniel Stone <daniel@fooishbar.org>
115003Date: Fri Aug 18 17:05:50 2006 +0300
115004
115005 Merge branch 'origin' into input-hotplug
115006
115007commit 19f673b7788d32c220e7e06734f1074b0e4a999c
115008Merge: cb0a565 70ddd0f
115009Author: Daniel Stone <daniel@fooishbar.org>
115010Date: Fri Aug 18 17:05:41 2006 +0300
115011
115012 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
115013
115014commit cb0a565d2b2cf8823abbd77b4426cc2237731dc1
115015Author: Daniel Stone <daniel@fooishbar.org>
115016Date: Fri Aug 18 17:04:48 2006 +0300
115017
115018 dix: add whiteroot flag
115019 Add a -wr option to use a white root window, and use a BackPixel rather
115020 than BackPixmap for both white and black root windows.
115021
115022commit 70ddd0f39d5118db72a1a4e473cbfb502f1ed9ec
115023Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
115024Date: Fri Aug 18 14:43:10 2006 +0100
115025
115026 Fix bug #5735, Serious flaw in CygwinX clipboard
115027 integration prevents paste from X to Windows apps
115028 (Brett Stahlman & Colin Harrison)
115029
115030commit 708b225689b5a4ba9ffe3372b584b715ef9eacdc
115031Merge: e1f4565 ee5e2cb
115032Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
115033Date: Fri Aug 18 09:13:52 2006 +0100
115034
115035 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
115036
115037commit e1f4565be5ce80be4655e81f77f4073fa3fbf8d0
115038Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
115039Date: Fri Aug 18 09:11:48 2006 +0100
115040
115041 Fix bug #7302, make Xn.hosts work from the
115042 Microsoft Windows install directory on Xming.
115043 (Colin Harrison)
115044
115045commit 1880defe4eaba02f9585b154d0883235eabc6d11
115046Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
115047Date: Fri Aug 18 09:09:53 2006 +0100
115048
115049 Fix bug #7281, clipboard viewer should not
115050 call SetClipboard viewer when bogus in Xming/CygwinX
115051 (Colin Harrison)
115052
115053commit a1a8e4f7f5917f537eb3dd51d3d6fa3e129236ce
115054Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
115055Date: Fri Aug 18 09:08:12 2006 +0100
115056
115057 Fix bug #7280, round title corner background
115058 should be transparent not black in Xming/CygwinX
115059 (Colin Harrison)
115060
115061commit ee5e2cbd2bee610a95facc6b486c4a5070973099
115062Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115063Date: Thu Aug 17 17:29:32 2006 -0400
115064
115065 Un-cut-and-paste the mode rejection message.
115066
115067commit 43e42eef1f5a22703eb64fc9cffecde036ea38e0
115068Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115069Date: Thu Aug 17 16:22:07 2006 -0400
115070
115071 Fix default mouse device on Linux, again.
115072
115073 It would be really nice if we only did this in one place instead of 40.
115074
115075commit 633b6a69f560c0a77dcff78fdef5fcf0041e2e04
115076Merge: 95dbfcf 7da5144
115077Author: Daniel Stone <daniel@fooishbar.org>
115078Date: Thu Aug 17 21:25:14 2006 +0300
115079
115080 Merge branch 'master' into input-hotplug
115081
115082commit 7da51447eaab34292e14077fb5a48e6b2e587781
115083Author: Daniel Stone <daniel@fooishbar.org>
115084Date: Thu Aug 17 21:24:07 2006 +0300
115085
115086 events.c: fix tiny XEvIE thinko
115087 Make sure xeviehot gets updated after the ConfineToShape() call.
115088
115089commit 5d082f05632906c29296a44ef5c3a4962c0cbe62
115090Author: Daniel Stone <daniel@fooishbar.org>
115091Date: Thu Aug 17 21:18:18 2006 +0300
115092
115093 events.c: make XEvIE a little less verbose
115094 Change a lot of:
115095 #ifdef XEVIE
115096 xeviehot.x =
115097 #endif
115098 sprite.hot.x = ...
115099 #ifdef XEVIE
115100 xeviehot.y =
115101 #endif
115102 sprite.hot.y = ...
115103 to one single
115104 #ifdef XEVIE
115105 xeviehot.x = sprite.hot.x;
115106 xeviehot.y = sprite.hot.y;
115107 #endif
115108 at the end of the functions.
115109
115110commit 95dbfcf8828c041c218145afc87d21a6c9c7bc02
115111Author: Daniel Stone <daniel@fooishbar.org>
115112Date: Thu Aug 17 21:18:18 2006 +0300
115113
115114 events.c: make XEvIE a little less verbose
115115 Change a lot of:
115116 #ifdef XEVIE
115117 xeviehot.x =
115118 #endif
115119 sprite.hot.x = ...
115120 #ifdef XEVIE
115121 xeviehot.y =
115122 #endif
115123 sprite.hot.y = ...
115124 to one single
115125 #ifdef XEVIE
115126 xeviehot.x = sprite.hot.x;
115127 xeviehot.y = sprite.hot.y;
115128 #endif
115129 at the end of the functions.
115130
115131commit c6c39afde3e5f43b623ca6b52162b83c98a28d45
115132Author: Daniel Stone <daniel@fooishbar.org>
115133Date: Thu Aug 17 21:13:09 2006 +0300
115134
115135 dix/events.c: add YAFIXME
115136 Add another FIXME to the cacaphony of XXX and FIXMEs in this file.
115137
115138commit f9624e0109cf12b6af43fb4235aaa0b54340a4bb
115139Author: Daniel Stone <daniel@fooishbar.org>
115140Date: Thu Aug 17 16:09:51 2006 +0300
115141
115142 kdrive/input: verify SIGIO with --enable-debug
115143
115144commit 73e58adda96c1d1b5176d819107faa7697c3eb94
115145Author: Matthias Hopf <mhopf@suse.de>
115146Date: Wed Aug 16 18:17:58 2006 +0200
115147
115148 Fixed segfault w/ broken Xinerama configs.
115149
115150commit cd3f744b1f983f71476db99c050045d981c5f5b2
115151Author: Daniel Stone <daniel@fooishbar.org>
115152Date: Tue Aug 15 15:54:13 2006 +0300
115153
115154 fix missing brace, trim unused variables
115155
115156commit 5d073697adb3864133fa3221b82ab8d2f4a59758
115157Author: Daniel Stone <daniel@fooishbar.org>
115158Date: Tue Aug 15 15:37:10 2006 +0300
115159
115160 kdrive/input: minor warning cleanups
115161 And also a compiler error fix when VERIFY_SIGIO is defined.
115162
115163commit 47c1c948e69cfba950ad37a3133fa2db0bd0ff2c
115164Author: Daniel Stone <daniel@fooishbar.org>
115165Date: Tue Aug 15 15:25:16 2006 +0300
115166
115167 kdrive/input: only run special key behaviours on non-XKB
115168 Only attempt to manually deal with special key behaviours (e.g. terminating
115169 the server) when not using XKB, and leave locking behaviour up to GKVE.
115170
115171commit 34228d8b280ef105a0c60b8de5dacf70a5ce24b5
115172Author: Daniel Stone <daniel@fooishbar.org>
115173Date: Tue Aug 15 15:23:53 2006 +0300
115174
115175 GPE: fix absolute button events / GKVE: (non-XKB) don't repeat lock keys
115176 Fix absolute button events in GPE, where we would previously send valuator
115177 events without bumping numEvents accordingly, causing the core event to
115178 go missing.
115179 In the non-XKB path in GKVE, implement proper lock behaviour (one press to
115180 enable, one press to disable, discard releases).
115181 Fix debug_events prototype.
115182
115183commit d003bada3352ec7d734498c4c732904876a9d1e2
115184Merge: d6433be a815b9b
115185Author: Daniel Stone <daniel@fooishbar.org>
115186Date: Sat Aug 12 22:48:55 2006 +0300
115187
115188 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115189
115190commit a815b9b990e068f02d9cbba2b17f2cc3a30a9310
115191Merge: 37943e2 984babe
115192Author: George Sapountzis <gsap7@yahoo.gr>
115193Date: Sat Aug 12 21:58:33 2006 +0300
115194
115195 Merge branch 'master' of git+ssh://gsap7@git.freedesktop.org/git/xorg/xserver
115196
115197commit d6433be3cca807dd78fbb1f45d9ba0212283083d
115198Merge: 2bf9e3d 984babe
115199Author: Daniel Stone <daniel@fooishbar.org>
115200Date: Sat Aug 12 21:50:52 2006 +0300
115201
115202 Merge branch 'master' into input-hotplug
115203
115204commit 2bf9e3dc1ec5fd7bf84a4a96899e5663a721d4a4
115205Author: Daniel Stone <daniel@fooishbar.org>
115206Date: Sat Aug 12 21:50:39 2006 +0300
115207
115208 make DIX more tolerant of devices without a CtrlProc (Debian #269860)
115209 Return BadDevice on client requests for devices without a CtrlProc, instead
115210 of tanking horribly.
115211
115212commit 984babe86bf82002b4d6589b2750c7b5a5489bd5
115213Author: Daniel Stone <daniel@fooishbar.org>
115214Date: Sat Aug 12 21:43:38 2006 +0300
115215
115216 remove obsolete vendor defines
115217 Remove random behaviour changes for SGI and MetroLink.
115218
115219commit 37943e2f1abc6709ff739000372b0394d5cd18c5
115220Author: George Sapountzis <gsap7@yahoo.gr>
115221Date: Sat Aug 12 20:54:33 2006 +0300
115222
115223 Call exaTryComponentAlphaHelper() for solid src also.
115224
115225 Also, rename to exaTryMagicTwoPassCompositeHelper() as it is now called for
115226 non-component-alpha masks also, and add function description from
115227 http://anholt.livejournal.com/32058.html.
115228
115229commit f7919c287936f55569c2301ebb1b5f52358e70fa
115230Author: Bastian Blank <waldi@debian.org>
115231Date: Sat Aug 12 20:43:25 2006 +0300
115232
115233 xfree86: don't do legacy IO on ARM or S/390 (Debian #362641)
115234 Don't attempt to poke legacy IO ranges on ARM or S/390.
115235
115236commit 59dcc62906d8ee597cd43aa307f414cb47995cea
115237Author: Daniel Stone <daniel@fooishbar.org>
115238Date: Sat Aug 12 20:39:08 2006 +0300
115239
115240 xfree86: remove Xqueue support completely
115241
115242commit e641000b98e7c2e92e3c801eaa42aa15d5c16ad0
115243Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
115244Date: Sat Aug 12 19:41:59 2006 +0300
115245
115246 xfree86: add Hurd support (#5613)
115247 Add support for GNU/Hurd.
115248
115249commit 5a3488ccac8e5dabd9fc98bc41ef178ead1b2faf
115250Author: Daniel Stone <daniel@fooishbar.org>
115251Date: Sat Aug 12 19:25:06 2006 +0300
115252
115253 configure.ac: fix execinfo.h test (Debian #363218)
115254 Define HAVE_EXECINFO_H as well as HAVE_BACKTRACE, when we find execinfo.h.
115255
115256commit 26c3cd1c9e3f52548389817a6d89a377e20c4269
115257Merge: 008aa7e c4951e0
115258Author: Daniel Stone <daniel@fooishbar.org>
115259Date: Sat Aug 12 18:58:18 2006 +0300
115260
115261 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115262
115263commit 008aa7eb6ed090384e6c44f861410e317c78a1cd
115264Author: Daniel Stone <daniel@fooishbar.org>
115265Date: Sat Aug 12 18:56:05 2006 +0300
115266
115267 completely remove OS keyboard layer
115268 Completely axe the keyboard layer from os-support.
115269
115270commit c4951e0a6b6cf3eeee710cc5cda1d9bc929ee3d7
115271Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115272Date: Thu Aug 10 20:49:06 2006 -0400
115273
115274 Fix a mode sanity check to not break reduced-blanking setups (LCDs).
115275
115276commit e1921f014b102e3eecf3b41972f8672cf23264d6
115277Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115278Date: Thu Aug 10 20:43:15 2006 -0400
115279
115280 Rename some mode tokens to better reflect their use.
115281
115282 Per #5386, M_T_EDID -> M_T_DRIVER, since it's really for any driver-detected
115283 mode. Also add M_T_PREFERRED bit, to select a 'best' mode out of a set.
115284
115285commit c2535f67923bde0bfb0e72363467110806e2f40f
115286Merge: c0cb8d1 db82e12
115287Author: Alan Coopersmith <alan.coopersmith@sun.com>
115288Date: Thu Aug 10 10:37:59 2006 -0700
115289
115290 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into XACE-modular
115291
115292commit 9525c2709ea3245c6518d4c3b5a0a4afff37181d
115293Author: Daniel Stone <daniel@fooishbar.org>
115294Date: Thu Aug 10 20:29:57 2006 +0300
115295
115296 configure.ac: fix xephyr conditionals
115297 Fix Xephyr build conditions, allowing it to actually be disabled.
115298
115299commit 6d8d4abaaacf08140b673472d985117d448a62e7
115300Author: Daniel Stone <daniel@fooishbar.org>
115301Date: Thu Aug 10 20:28:06 2006 +0300
115302
115303 configure.ac: allow conditional building of XF86{DGA,Misc,VidMode}
115304 Allow conditional building of the above three extensions, defaulting to
115305 auto.
115306
115307commit 1a5561c4ecccaf32b03c41373adf376100d457aa
115308Merge: 190f229 db82e12
115309Author: Ian Romanick <idr@umwelt.(none)>
115310Date: Thu Aug 10 10:20:37 2006 -0700
115311
115312 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
115313
115314commit cec284f2b3e948deb9e56a1a8519fddf693ab952
115315Author: Daniel Stone <daniel@fooishbar.org>
115316Date: Thu Aug 10 18:03:58 2006 +0300
115317
115318 kdrive: properly ifdef composite enabling
115319
115320commit 45bce556e8665412b9f6e89f88ed5bedb41de1ba
115321Author: Daniel Stone <daniel@fooishbar.org>
115322Date: Thu Aug 10 18:02:47 2006 +0300
115323
115324 GetMaximumEventsNum: be more conservative
115325 Be slightly more conservative in our maximum event count if we're using
115326 XKB (and thus don't need to count the extra repeat events).
115327
115328commit 172d45b9b75f95c997d1e9358040eead496e2a06
115329Merge: 3832a3d db82e12
115330Author: Daniel Stone <daniel@fooishbar.org>
115331Date: Thu Aug 10 14:14:54 2006 +0300
115332
115333 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115334
115335commit 3832a3d7db0f3b5d7167e3b3b5ea6d6b3d96351a
115336Author: Daniel Stone <daniel@fooishbar.org>
115337Date: Thu Aug 10 14:13:51 2006 +0300
115338
115339 GKVE: don't repeat modifiers when using XKB
115340 Make sure we don't ever repeat modifiers (previously was repeating when
115341 using XKB); only do explicit KP/KR repeats in the non-XKB case. XKB will
115342 take care of repeating when we're using it.
115343
115344commit 9f188416bb6b4837d4c3f8773053d5eee0ff0ee1
115345Author: Daniel Stone <daniel@fooishbar.org>
115346Date: Thu Aug 10 14:00:34 2006 +0300
115347
115348 core devices: clear devicePrivates on close
115349
115350commit 539d1f33475484d35fb5a377efc76dba2d868e3f
115351Author: Daniel Stone <daniel@fooishbar.org>
115352Date: Thu Aug 10 14:00:14 2006 +0300
115353
115354 GKVE/GPE: have DDX allocate events
115355 Don't allocate events on every GKE/GKVE/GPE call, just have the DDX manage
115356 it instead. Introduce GetMaximumEventsNum(), which is the maximum number
115357 of events these functions will ever produce.
115358
115359commit db82e12fac5eaa16a39fc1bd0bc31ad95089dc95
115360Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115361Date: Wed Aug 9 14:55:17 2006 -0400
115362
115363 Remove TargetRefresh option from the autoconfig logic.
115364
115365 The default target of 75Hz is almost always wrong for LCDs.
115366
115367commit fcd4167e8913f77bdf9e17a6955d0f2a9f4eeb10
115368Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115369Date: Wed Aug 9 14:48:51 2006 -0400
115370
115371 Remove the bc flag from the -help text, since it's gone.
115372
115373commit 767f372dd02232469f9fd804b811a17eaf762e1e
115374Merge: c4f5de6 462bb61
115375Author: Tilman Sauerbeck <tilman@code-monkey.de>
115376Date: Wed Aug 9 20:23:30 2006 +0200
115377
115378 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
115379
115380commit c4f5de6cc3b935025829af971b0b8010c1ecfedb
115381Author: Tilman Sauerbeck <tilman@code-monkey.de>
115382Date: Wed Aug 9 20:21:52 2006 +0200
115383
115384 Sanitized glxdri's Block/Wakeuphandler calling.
115385
115386 __glXDRIleaveServer() and _enterServer() used to call DRIDoBlockHandler
115387 (resp DRIDoWakeupHandler) directly. They are now calling DRIBlockHandler
115388 (resp DRIWakeupHandler) to account for driver specific block/wakeup
115389 hooks.
115390
115391commit 5506b4ad200745236f997c121e8200179c47b749
115392Merge: 4be9abb 462bb61
115393Author: Daniel Stone <daniel@fooishbar.org>
115394Date: Wed Aug 9 07:21:01 2006 +0300
115395
115396 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115397
115398commit 4be9abb8504b3761b5f3a01851e4eb3da86c76e2
115399Author: Daniel Stone <daniel@fooishbar.org>
115400Date: Wed Aug 9 07:20:16 2006 +0300
115401
115402 kdrive: remove ddx_DEPENDENCIES
115403 Remove foo_DEPENDENCIES as they weren't guaranteed to just be libs,
115404 and loader arguments (-lfoo, -Lfoo) might've crept in.
115405
115406commit 462bb61b0fe968fae1b99cf98ec6f7de09105dcd
115407Author: Aaron Plattner <aplattner@nvidia.com>
115408Date: Tue Aug 8 18:07:22 2006 -0700
115409
115410 Add CompositeRegisterAlternateVisuals.
115411
115412 This provides drivers the ability to add their own alternate visuals and then
115413 register them with Composite for implicit redirection.
115414
115415commit 190f229ed77d87797e0f2f6762c86b3ad3a3dcbe
115416Author: Ian Romanick <idr@umwelt.(none)>
115417Date: Tue Aug 8 16:47:32 2006 -0700
115418
115419 Make xf86PciVideoInfo static since it is only used within this file.
115420
115421 Dummy out all of the PCI bus and device access control functions. We
115422 need a better way to do this, and that will probably be in
115423 libpciaccess and / or the kernel.
115424
115425 Refactor xf86GetPciInfoForEntity to use pci_device_find_by_slot.
115426
115427 Refector xf86CheckPciSlot to use xf86GetPciEntity.
115428
115429 Eliminate disablePciBios and the one place that calls it.
115430
115431commit 5508f7646f9754b054e961a1025b7a52913b563b
115432Author: Ian Romanick <idr@umwelt.(none)>
115433Date: Tue Aug 8 16:43:31 2006 -0700
115434
115435 Remove prototypes from some functions that were previously removed.
115436
115437commit 9df53d903ed68073bf7d2c2a275b6f6556a85c0f
115438Author: Ian Romanick <idr@umwelt.(none)>
115439Date: Tue Aug 8 16:42:23 2006 -0700
115440
115441 Rename xf86ReadDomainMemory to xf86ReadLegacyVideoBIOS, since that's
115442 what it is actually used for. Modify a few routines in linuxPci.c to
115443 take pci_device structures as parameters in stead of PCITAGs.
115444
115445commit fe351a711ef55c3ae1e784d4551147c080eda109
115446Author: Daniel Stone <daniel@fooishbar.org>
115447Date: Tue Aug 8 14:54:10 2006 +0300
115448
115449 GKVE: send XkbMapNotify, not XkbNewKeyboardNotify
115450 Sending MapNotify is more correct in this case than NKN, so do that.
115451
115452commit 31089816317f27c668b12a15c74fdd226a8df9f7
115453Merge: ab3ebfe 12dbd8a
115454Author: Daniel Stone <daniel@fooishbar.org>
115455Date: Tue Aug 8 12:01:12 2006 +0300
115456
115457 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115458
115459commit ab3ebfefdb7d21aba72a5030f6f93bf4fafed709
115460Author: Tiago Vignatti <tv02@c3sl.ufpr.br>
115461Date: Tue Aug 8 12:00:34 2006 +0300
115462
115463 xorgcfg: fix compilation error
115464 Add missing parentheses to IS_KBDDRIV macros.
115465
115466commit 12dbd8a02f797ad57a1be683a02a1fcb1ca35438
115467Author: Daniel Stone <daniel@fooishbar.org>
115468Date: Mon Aug 7 23:43:40 2006 +0300
115469
115470 remove optional R3 backwards compatibility
115471 Remove the permitOldBugs flag, which enabled backwards compatbility with
115472 broken R2/R3 era clients.
115473
115474commit 7721ee308fbbb6fc9c969f15fe04b3346c04f843
115475Author: Daniel Stone <daniel@fooishbar.org>
115476Date: Mon Aug 7 23:03:02 2006 +0300
115477
115478 config client: fix minor race with event queue
115479 Fix a small race whereby you could remove a device while events from it
115480 were still in the queue, by calling ProcessInputEvents immediately before
115481 RemoveDevice, to (hopefully) flush the event queue.
115482
115483commit 458c63a84110f64c7fce397a531a3a779c2239a2
115484Author: Daniel Stone <daniel@fooishbar.org>
115485Date: Mon Aug 7 23:02:17 2006 +0300
115486
115487 enable adding devices after removal of a middle device
115488 Allow new devices to be added after a device that _wasn't_ the last on the
115489 list was removed, by ensuring inputInfo.numDevices always increases, and
115490 never decreases.
115491
115492commit a31d11a7a8485cdc799f76c4d407d3b7f7c9b350
115493Author: Daniel Stone <daniel@fooishbar.org>
115494Date: Mon Aug 7 23:01:23 2006 +0300
115495
115496 GKVE: get rid of bogus double-release check
115497 Get rid of a bogus double-release check which broke non-XKB servers.
115498
115499commit bedc4ecf23c7150e3156e0d24602ed3bc3977225
115500Author: Daniel Stone <daniel@fooishbar.org>
115501Date: Mon Aug 7 23:00:45 2006 +0300
115502
115503 xephyr: aid input debugging
115504 Add the 'ephyr' mouse and keyboard drivers to the driver list so we can
115505 re-add devices.
115506 Set the names properly in Ephyr{Keyboard,Mouse}Init, not in InitInput.
115507
115508commit baf93b3abe1e88d82ee6a3d6939f50f96ded271a
115509Author: Daniel Stone <daniel@fooishbar.org>
115510Date: Mon Aug 7 21:12:45 2006 +0300
115511
115512 kdrive: move map initialisation to KdNewPointer
115513 Do a linear n -> n initialisation on the map up until KD_MAX_BUTTON in
115514 KdNewPointer, moving it out of both KdParsePointer, and KdPointerProc.
115515 Also remove dead pointer acceleration code.
115516
115517commit ccb53340b66a778abf10182fd88a7d699207fb84
115518Author: Daniel Stone <daniel@fooishbar.org>
115519Date: Mon Aug 7 21:12:00 2006 +0300
115520
115521 ephyr: cleanup
115522 Remove extraneous KdAddPointerDriver call.
115523
115524commit d1c18af27e0aed73104743afb4bf4b8d3d1186cf
115525Author: Daniel Stone <daniel@fooishbar.org>
115526Date: Mon Aug 7 21:11:38 2006 +0300
115527
115528 GPE: use button map for DBP/DBR, not just BP/BR
115529 Make sure we use the button map for extended events, not just core.
115530
115531commit 9b7ecbd1dd8d092221897e29c85f3306c7367716
115532Author: Daniel Stone <daniel@fooishbar.org>
115533Date: Mon Aug 7 21:09:32 2006 +0300
115534
115535 kdrive: prevent overrun in map
115536 We actually need n + 1 elements for the mouse button map, not n.
115537
115538commit eb6e8d4042252b13328dbb122e0e6186796a80ac
115539Author: Daniel Stone <daniel@fooishbar.org>
115540Date: Mon Aug 7 21:05:37 2006 +0300
115541
115542 kdrive: increase maximum number of buttons
115543 Increase KD_MAX_BUTTONS to 32.
115544
115545commit 1c72290cdf4d9b214e1b9c0526cb7cb8641051f3
115546Author: Aaron Plattner <aplattner@nvidia.com>
115547Date: Mon Aug 7 09:57:58 2006 -0700
115548
115549 Use DrawablePtrs instead of PixmapPtrs for Prepare/Finish access.
115550
115551 Also, define some wfb functions even if FB_ACCESS_WRAPPER is not defined. This allows a client to use libfb and libwfb at the same time.
115552
115553commit afcad4ad99bbfc8bdcd0f4fdd70e072108410d30
115554Author: Daniel Stone <daniel@fooishbar.org>
115555Date: Mon Aug 7 18:11:05 2006 +0300
115556
115557 xfree86 ddx: always free GKE/GPE events
115558 free() events we get passed back from GKE and GPE so we don't just, er,
115559 leak them all. *cough*.
115560
115561commit 98fdf874eeadd5b37413922d8afba8415d0c56bb
115562Author: Daniel Stone <daniel@fooishbar.org>
115563Date: Mon Aug 7 16:51:39 2006 +0300
115564
115565 move all autorepeat logic to DIX
115566 Move core autorepeat logic for keyboards down to the DIX, remove it from
115567 KDrive.
115568
115569commit 5c7001fef8ffc6e3d8585a37d3f79a9495be8ed0
115570Author: Daniel Stone <daniel@fooishbar.org>
115571Date: Mon Aug 7 16:51:09 2006 +0300
115572
115573 memcpy() events in
115574 memcpy events into our event structure instead of doing pointer assignment.
115575
115576commit c85e64cba1d2d88f676ca7cf23b52a6f8219e90e
115577Merge: a406f6b f54b71b
115578Author: Daniel Stone <daniel@fooishbar.org>
115579Date: Mon Aug 7 15:54:55 2006 +0300
115580
115581 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115582
115583commit f54b71b772a1f587394ae3968782b611e52f0e2d
115584Author: David Nusinow <david@nee.(none)>
115585Date: Sun Aug 6 18:11:00 2006 +0000
115586
115587 Document enable/disable flag for AIGLX in xorg.conf manpage.
115588
115589commit a406f6bfeaa46e3236f7ab46813fe6c30b936a35
115590Author: Daniel Stone <daniel@fooishbar.org>
115591Date: Fri Aug 4 12:40:19 2006 +0300
115592
115593 mieq: don't leak events
115594 free all events posted through mieqEnqueue.
115595
115596commit 997ba45b192f21810099ed888792a45f1677a9ce
115597Author: Daniel Stone <daniel@fooishbar.org>
115598Date: Fri Aug 4 11:18:16 2006 +0300
115599
115600 fix incorrect button test
115601 Test for n (1..nButtons) being under nButtons, not button (1..(1<<nButtons)).
115602
115603commit 9138d5a51e411f598bc0e75a3e73d2c16187a518
115604Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
115605Date: Thu Aug 3 18:24:04 2006 -0400
115606
115607 Make SecurityLookupIDBy* part of the base functionality.
115608
115609commit 45c229f526bf1dafb5e81b50d700449ba4e1613d
115610Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
115611Date: Thu Aug 3 14:26:06 2006 -0400
115612
115613 Remove LBX code.
115614
115615commit 96e45626c43b7674b66e0258b0b1730d5ce71357
115616Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
115617Date: Wed Aug 2 20:29:59 2006 -0400
115618
115619 Rebase Security extension to use devPrivates for storing security state.
115620
115621commit 83ebf61ec03ff31005375900bee1e55e0e694c4b
115622Author: Ian Romanick <idr@umwelt.(none)>
115623Date: Wed Aug 2 13:33:33 2006 -0700
115624
115625 Eliminate the last few places outside xf86pciBus.c that use
115626 xf86PciVideoInfo. In all cases this means converting the code to use
115627 the pci_device_iterator interfaces.
115628
115629commit 5bfb2ee9652e5ef6d076ef8b6df54baaa43b8e0c
115630Author: Ian Romanick <idr@umwelt.(none)>
115631Date: Wed Aug 2 13:29:21 2006 -0700
115632
115633 Remove the last remants of the pci{Read,Write}{Long,Word,Byte}
115634 functionality. This also allows the removal of PCI_CPU, PCI_CPU16,
115635 and a few other dangling bits of cruft.
115636
115637commit a0f2e1cae46f189ed97e2d92b485d315b3d2627a
115638Author: Ian Romanick <idr@umwelt.(none)>
115639Date: Wed Aug 2 10:54:41 2006 -0700
115640
115641 Move xf86FindPciDeviceVendor and xf86FindPciClass from xf86pciBus.c to
115642 xf86int10.c. Refactor common code from those functions to do_find.
115643
115644commit de8234606f87ce79d016f7ddeabdf57c4ad212f3
115645Author: Ian Romanick <idr@umwelt.(none)>
115646Date: Wed Aug 2 10:44:11 2006 -0700
115647
115648 Modify xf86FindPciDeviceVendor and xf86FindPciClass to use the
115649 pci_interator interface instead of the xf86PciVideoInfo array.
115650
115651commit 3c23dec5962b8b81ae838fe0ee2c7b0a789f5386
115652Author: Eamon Walsh <ewalsh@tycho.nsa.gov>
115653Date: Wed Aug 2 13:39:49 2006 -0400
115654
115655 Call ClientStateCallback on serverClient devPrivates initialization.
115656
115657commit 23f44df9009023e77508f03ac4a7595c7e3d40cb
115658Author: Ian Romanick <idr@umwelt.(none)>
115659Date: Wed Aug 2 09:55:32 2006 -0700
115660
115661 Remove pciConfigPtr and all of the associated cruft.
115662
115663commit 966ebd3d2d84b440e89504d055a0e937303ed11d
115664Merge: a941766 39169fd
115665Author: Ian Romanick <idr@umwelt.(none)>
115666Date: Wed Aug 2 08:14:58 2006 -0700
115667
115668 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
115669
115670commit ee02e647882a4be29e1130bd79904ee79ed6b802
115671Author: Aaron Plattner <aplattner@nvidia.com>
115672Date: Tue Aug 1 13:45:43 2006 -0700
115673
115674 Wrap libwfb memory access.
115675
115676 Use the READ and WRITE macros to wrap memory accesses that could be in video
115677 memory. Add MEMCPY_WRAPPED and MEMSET_WRAPPED macros to wrap memcpy and
115678 memset, respectively.
115679
115680commit 39169fd373b97f34923f6494d697d9429d0b8aa3
115681Author: Matthew Allum <mallum@polystyrene.(none)>
115682Date: Tue Aug 1 13:39:22 2006 +0100
115683
115684 Back out 'mystery' spurious host window hints.
115685
115686commit f737cc38baea6af8bf284c9e207e60a7d90eebe1
115687Author: Eamon Walsh <ewalsh@moss-uranus.epoch.ncsc.mil>
115688Date: Mon Jul 31 19:58:42 2006 -0400
115689
115690 Adding devPrivates support to the ExtensionEntry structure.
115691
115692commit b04d64854712678701d5243aacf5cc93444cfadc
115693Author: Eamon Walsh <ewalsh@moss-uranus.epoch.ncsc.mil>
115694Date: Mon Jul 31 19:35:08 2006 -0400
115695
115696 Added devPrivates support to the ExtensionEntry structure.
115697
115698commit a94176627cdd6122ffadb618995f9bcec872609a
115699Author: Ian Romanick <idr@umwelt.(none)>
115700Date: Mon Jul 31 14:33:58 2006 -0700
115701
115702 Eliminate all uses of the businfo field of pciDevice / pciConfigPtr.
115703
115704commit a69335dc299be6de8b82ed34de1cb30f1255feb4
115705Author: Aaron Plattner <aplattner@nvidia.com>
115706Date: Mon Jul 31 14:15:55 2006 -0700
115707
115708 Make ReadMemoryProcPtr take a const pointer.
115709
115710 Fixes some warnings when using READ with a const pointer.
115711
115712commit 26c9587f90806cce1c0d2a98e656e9c311a34ee5
115713Author: Ian Romanick <idr@umwelt.(none)>
115714Date: Mon Jul 31 13:54:36 2006 -0700
115715
115716 Build fixes for last pull from HEAD.
115717
115718commit 65dc25d8f86b962f2adc765a9ff327419f98092b
115719Merge: 69533fc b74c845
115720Author: Ian Romanick <idr@umwelt.(none)>
115721Date: Mon Jul 31 12:30:30 2006 -0700
115722
115723 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
115724
115725commit b74c845a1233f78b841ff8840272c50873300c20
115726Merge: 3112a6c 02daa6b
115727Author: Ian Romanick <idr@umwelt.(none)>
115728Date: Mon Jul 31 10:26:06 2006 -0700
115729
115730 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
115731
115732commit 3112a6c4f26d5e9258b8def7ce4109b4bd408c67
115733Author: Ian Romanick <idr@umwelt.(none)>
115734Date: Mon Jul 31 10:25:31 2006 -0700
115735
115736 Noting uses libdummy.a, so don't build it. Only libdummy-nonserver.a
115737 is actually used.
115738
115739commit 02daa6bb103e53e5a33db2bb6acbe57d0bf2c30e
115740Author: Matthew Allum <mallum@polystyrene.(none)>
115741Date: Mon Jul 31 17:32:05 2006 +0100
115742
115743 Improve XRes to;
115744 - Better estimate general pixmap memory usage.
115745 - Account for pixmaps shared between clients.
115746 - Account for window background and border pixmaps,
115747 and GC stripple and tile pixmaps.
115748
115749commit 69533fc04f38c67fd424776c5191a4ba57ba8c8f
115750Merge: ef1aeca 24051ef
115751Author: Ian Romanick <idr@umwelt.(none)>
115752Date: Mon Jul 31 09:26:04 2006 -0700
115753
115754 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
115755
115756 Conflicts:
115757
115758 hw/xfree86/Makefile.am
115759 hw/xfree86/common/xf86AutoConfig.c
115760
115761commit 24051ef97406f28c102cf46a78223400b61fdae2
115762Author: Daniel Stone <daniel@fooishbar.org>
115763Date: Sun Jul 30 12:15:33 2006 +0300
115764
115765 remove filename that's too long for tar
115766
115767commit ecb7d43a76d507d04891ab7f189b23be5eccda51
115768Author: Daniel Stone <daniel@fooishbar.org>
115769Date: Sun Jul 30 11:52:41 2006 +0300
115770
115771 add sym.h to sources
115772
115773commit bf2d7499c84c94f228d03b21448f5688b3cda1a8
115774Author: Daniel Stone <daniel@fooishbar.org>
115775Date: Sun Jul 30 11:17:02 2006 +0300
115776
115777 add securitysrv.h
115778
115779commit e87e68634d8eb66ab783e2802e2d5d12ff1031be
115780Author: Daniel Stone <daniel@fooishbar.org>
115781Date: Sun Jul 30 11:11:59 2006 +0300
115782
115783 remove .cvsignores from EXTRA_DIST
115784
115785commit ed0c807de9f07468385fcbd2e8a9c0737759a461
115786Author: Daniel Stone <daniel@fooishbar.org>
115787Date: Sun Jul 30 11:08:54 2006 +0300
115788
115789 bump to 1.1.99.3
115790
115791commit a68dc013a33d867e65a7e76b3eec5947b862a5b4
115792Author: Daniel Stone <daniel@fooishbar.org>
115793Date: Sun Jul 30 11:08:47 2006 +0300
115794
115795 remove README (which doesn't exist) from EXTRA_DIST
115796
115797commit 87fe85f38b6f781bf0e2eb555526e3d77779f9fa
115798Merge: 3518e2d 654619d
115799Author: Daniel Stone <daniel@fooishbar.org>
115800Date: Sun Jul 30 10:51:34 2006 +0300
115801
115802 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115803
115804commit 654619d76c779606f2315782fc01d1410399fa3b
115805Author: Kevin E Martin <kem@freedesktop.org>
115806Date: Fri Jul 28 17:16:32 2006 -0400
115807
115808 Revert xkb changes that broke XkbGetKeyboard()
115809
115810commit 79016d4036786b091a9b9d1133a6cdfedd6c277d
115811Author: Luc Verhaegen <libv@skynet.be>
115812Date: Fri Jul 28 16:02:02 2006 -0400
115813
115814 Bug #5386 (partial): Move CVT mode generator from cvt(1) to server core, and
115815 export it from the X server to modules.
115816
115817commit e6ae1612be519ee6224d354244d076d85d44a750
115818Author: Aaron Plattner <aplattner@nvidia.com>
115819Date: Thu Jul 27 18:24:59 2006 -0700
115820
115821 Add fbHasVisualTypes and fbSetVisualTypesAndMasks to wfbrename.h and include -DXFree86Server.
115822
115823commit 2a4ceb09ed5a09dc5763754ab865ec23df91ac9f
115824Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115825Date: Wed Jul 26 19:39:17 2006 -0400
115826
115827 Remove dead function prototypes.
115828
115829commit 990a4009057e068f41d20b95aa0c59357185650d
115830Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115831Date: Wed Jul 26 19:03:39 2006 -0400
115832
115833 Remove getconfig horrorshow. Replace with a static built-in rule list for now.
115834
115835commit 377a581ddf5e428a368efb1b59fcb317666fecdd
115836Author: Aaron Plattner <aplattner@nvidia.com>
115837Date: Tue Jul 25 15:27:31 2006 -0700
115838
115839 Switch to using void* pointers.
115840
115841 Pass the size of the data pointed to by src or dst as an argument to
115842 wfb{Read,Write}Memory. This allows one set of macros to be used with any size
115843 pointer. Assumes that sizeof(FbBits) >= sizeof(FbStip).
115844
115845commit a4005c15fbb48231cb958c32b2c791a2d23a135a
115846Author: Aaron Plattner <aplattner@nvidia.com>
115847Date: Mon Jul 10 18:58:09 2006 -0700
115848
115849 Add framebuffer access wrapper infrastructure.
115850
115851 Create fbPrepareAccess macros to call into the driver to set up the
115852 wfbReadMemory and wfbWriteWemory pointers. Call these from fbGetDrawable and
115853 fbGetStipDrawable.
115854
115855 Add the READ and WRITE macros, which expand to simple memory accesses for fb,
115856 and calls through the function pointers for wfb.
115857
115858 Add fbFinishAccess macro to give the driver an opportunity to clean up. Add
115859 calls to this in the appropriate places.
115860
115861commit 319efac445cebda5a2ac1db67efebe54bc47ba9d
115862Author: Aaron Plattner <aplattner@nvidia.com>
115863Date: Fri Jul 7 18:45:30 2006 -0700
115864
115865 Prefix all of the exported symbols in libwfb.so with "wfb".
115866
115867 For now, just #define all of the exported symbols in wfbrename.h. Later,
115868 we should add FBPREFIX() around the exported symbols and use -fvisiblity=hidden
115869 to hide the rest.
115870
115871commit 7608a63ff7409f399c9a26962a304b84196a1868
115872Author: Aaron Plattner <aplattner@nvidia.com>
115873Date: Thu Jul 6 17:05:21 2006 -0700
115874
115875 Build infrastructure for libwfb.so.
115876
115877 Builds fb/* twice, defining FB_ACCESS_WRAPPER for libwfb.la. Define a macro,
115878 FBPREFIX(X) which expands to fbX for libfb.la and wfbX for libwfb.la. Use the
115879 macro on [w]fbModuleData so the new module loads.
115880
115881commit 39158e98acb29e97a2682d4a37385f9141b484c4
115882Author: Adam Jackson <ajax@benzedrine.nwnk.net>
115883Date: Wed Jul 26 18:39:28 2006 -0400
115884
115885 Remove another latent PowerMAX hunk.
115886
115887 All your favorite running jokes of 2005, today!
115888
115889commit ef1aecaaf6cd7e9e4f9d0c6373664e774500bc13
115890Author: Ian Romanick <idr@umwelt.(none)>
115891Date: Wed Jul 26 13:25:13 2006 -0700
115892
115893 Pass correct pointer to xf86MapLegacyIO.
115894
115895commit 3518e2d0debc97e2bacdefe604b280e7fdfdd216
115896Merge: eb7733a 3821f6a
115897Author: Daniel Stone <daniel@fooishbar.org>
115898Date: Wed Jul 26 11:29:21 2006 +0300
115899
115900 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
115901
115902commit eb7733a48a92405660d5d2ab60913b62c30daaed
115903Author: Daniel Stone <daniel@fooishbar.org>
115904Date: Wed Jul 26 11:28:45 2006 +0300
115905
115906 kdrive: drop excessive NewInputDeviceRequest debugging
115907
115908commit 3821f6aeaa714582ee0a631de96c6e7cfd96303e
115909Author: Kristian Høgsberg <krh@redhat.com>
115910Date: Wed Jul 26 01:56:02 2006 -0400
115911
115912 Unlibc-wrap DMX glxscreens.c and fix tag-removal typo.
115913
115914commit 28ba8d56912ae56d7b9835188f621b0a491add30
115915Author: Ian Romanick <idr@umwelt.(none)>
115916Date: Tue Jul 25 15:38:09 2006 -0700
115917
115918 Gut anything having to do with the PciBusPtr (or PciBusRec) type.
115919 Bump required version of libpciaccess to 0.5.0 so that
115920 pci_device_get_bridge_buses can be used.
115921
115922commit 380b51d605a82c98082a2cebd70a0d1d0735eaa7
115923Merge: 82f6b7e 8977b07
115924Author: Ian Romanick <idr@umwelt.(none)>
115925Date: Tue Jul 25 11:30:04 2006 -0700
115926
115927 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
115928
115929 Conflicts:
115930
115931 hw/xfree86/common/xf86Init.c
115932 hw/xfree86/int10/pci.c
115933 hw/xfree86/scanpci/xf86PciData.h
115934 hw/xfree86/scanpci/xf86PciStdIds.h
115935 hw/xfree86/scanpci/xf86PciStr.h
115936 hw/xfree86/scanpci/xf86ScanPci.h
115937 hw/xfree86/utils/pcitweak/pcitweak.c
115938 hw/xfree86/utils/scanpci/scanpci.c
115939
115940 Re-removed most of the conflicting files.
115941
115942commit 4ea475924c557ad0819b796f5369e5e669465709
115943Author: Daniel Stone <daniel@fooishbar.org>
115944Date: Tue Jul 25 20:00:48 2006 +0300
115945
115946 re-add OpenedHand copyright
115947 Ae-add OpenedHand copyright, accidentally dropped in KDrive new world
115948 order patch. Sorry guys.
115949
115950commit 82f6b7e75e04e3fb951ab71ad3c553c8b34ea995
115951Author: Ian Romanick <idr@umwelt.(none)>
115952Date: Mon Jul 24 13:52:58 2006 -0700
115953
115954 Don't re-look-up pointers that are already stored in structures.
115955
115956commit 2fd6b995ff927c8ca7267c201f535564b82e691f
115957Author: Ian Romanick <idr@umwelt.(none)>
115958Date: Mon Jul 24 13:33:34 2006 -0700
115959
115960 Global s/xf86MapDomainIO/xf86MapLegacyIO/. Let's call a duck a duck,
115961 okay? Since xf86MapLegacyIO is called from only one place, cut the
115962 parameter list down to the one parameter that actually conveys some
115963 information: the one that gives a PCI device. Change from using a
115964 PCITAG to a pci_device.
115965
115966commit aed6fe0bb12c68b94e564252bc03594728ed8c5b
115967Author: Ian Romanick <idr@umwelt.(none)>
115968Date: Mon Jul 24 13:13:05 2006 -0700
115969
115970 Refactor linuxGetSizesStruct to take a pci_device pointer instead of a
115971 PCITAG. Modify xf86BusAccWindowsFromOS and xf86AccResFromOS to call
115972 linuxGetSizesStruct directly with a pci_device pointer. Remove
115973 linuxGetSizes.
115974
115975commit e48762799248eb7e16ea2c0df1561ae1430f2112
115976Author: Ian Romanick <idr@umwelt.(none)>
115977Date: Mon Jul 24 12:23:23 2006 -0700
115978
115979 Modify xf86BusAccWindowsFromOS and xf86AccResFromOS to use the
115980 libpciaccess interfaces.
115981
115982commit 4b474cbc1a0fe17da2438a15291f04ee67ea28f8
115983Author: Ian Romanick <idr@umwelt.(none)>
115984Date: Mon Jul 24 12:17:27 2006 -0700
115985
115986 The functions xf86PciBusAccWindowsFromOS and xf86BusAccWindowsFromOS
115987 are identical. Just have one call the other.
115988
115989commit 7dff8d98a9ed11f92806cb8c4b3f10e01c5d063f
115990Author: Ian Romanick <idr@umwelt.(none)>
115991Date: Mon Jul 24 11:56:37 2006 -0700
115992
115993 Remove source files that have been dead since day-1 of this branch.
115994
115995commit eb95128f615e2f3fab17dcafd4fc260c6d4d0644
115996Author: Ian Romanick <idr@umwelt.(none)>
115997Date: Mon Jul 24 11:53:37 2006 -0700
115998
115999 Changes that were missed on the previous commit. Perhaps oddness
116000 caused by doing git-update-index at the wrong time.
116001
116002commit 05f8b772aec2b05d92078bb60a07bd8413386363
116003Author: Ian Romanick <idr@umwelt.(none)>
116004Date: Mon Jul 24 11:23:03 2006 -0700
116005
116006 ScanPciDisplayPCICardInfo now uses pci_device interface for PCI
116007 scanning. Log messages simplified to make the code shorter and less
116008 convoluted. ScanPciDisplayPCICardInfo is now void since it was only
116009 called from one place with a constant parameter.
116010
116011commit ca3f4fc1b0c21a0620ab1eb35c199cd55d795095
116012Author: Daniel Stone <daniel@fooishbar.org>
116013Date: Sun Jul 23 19:02:12 2006 -0400
116014
116015 add fallback ChangeDeviceControl, allow XOpenDevice on closed device
116016 Add a fallback ChangeDeviceControl, which allows the attributes we know about
116017 so far.
116018 Allow XOpenDevice on closed devices.
116019
116020commit e73e5e2a4d8f22889d840a7719479f9af686cb9c
116021Merge: a73cef1 8977b07
116022Author: Daniel Stone <daniel@fooishbar.org>
116023Date: Sat Jul 22 13:56:30 2006 -0400
116024
116025 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
116026
116027commit 8977b07434d75ca396d236dc1324f0c862b633c7
116028Author: Dave Airlie <airlied@linux.ie>
116029Date: Sun Jul 23 03:36:47 2006 +1000
116030
116031 glx: fix typo from tag removal
116032
116033commit a73cef1f005ca66db18e952e676ee5a21b829700
116034Merge: 672ca15 70869fc
116035Author: Daniel Stone <daniel@fooishbar.org>
116036Date: Sat Jul 22 12:07:22 2006 -0400
116037
116038 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
116039
116040commit 70869fc6777f87cd05794446caa739e3d9a91ffe
116041Author: Adam Jackson <ajax@theobromine.nwnk.net>
116042Date: Fri Jul 21 23:39:37 2006 -0400
116043
116044 Yet more dead code.
116045
116046commit 7c1b2ee7a8238c267bc97e78bbff204dc7723dd3
116047Author: Adam Jackson <ajax@theobromine.nwnk.net>
116048Date: Fri Jul 21 23:35:13 2006 -0400
116049
116050 static markup, more dead code.
116051
116052commit 1c4f90b1d05d4c49279f3224a6dd94850a6bd8d0
116053Author: Adam Jackson <ajax@theobromine.nwnk.net>
116054Date: Fri Jul 21 23:03:21 2006 -0400
116055
116056 Open-coded path checks make baby Jesus cry.
116057
116058commit 114264584ca43091a5e07282566a30a6378a1502
116059Author: Adam Jackson <ajax@theobromine.nwnk.net>
116060Date: Fri Jul 21 22:55:41 2006 -0400
116061
116062 Remove a useless open() of the module we're about to load.
116063
116064commit 985611d5cd079f97da700c7b8e898d33da004be0
116065Author: Adam Jackson <ajax@theobromine.nwnk.net>
116066Date: Fri Jul 21 22:37:59 2006 -0400
116067
116068 Delete some long-unused testing code.
116069
116070commit 525257eb98b46209c69c8576daa6bd7afc1f48a9
116071Author: Ian Romanick <idr@umwelt.(none)>
116072Date: Fri Jul 21 19:16:38 2006 -0700
116073
116074 Fix stupid typos.
116075
116076commit 672ca156bfb11440e6e234650bfba9d38e1edb52
116077Merge: d14d91f 6cf844a
116078Author: Daniel Stone <daniel@fooishbar.org>
116079Date: Fri Jul 21 19:58:42 2006 -0400
116080
116081 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
116082
116083commit 6cf844ab69926b6d23619a12c97734af3881ba67
116084Author: Daniel Stone <daniel@fooishbar.org>
116085Date: Fri Jul 21 19:57:28 2006 -0400
116086
116087 loader: walk directory paths with readdir(), don't stat() everything
116088 Walk the directories with readdir, and don't stat everything we can
116089 find. Thanks to davej for the public humiliation reminding me to go back
116090 and re-fix this one.
116091
116092commit d14d91f094c3897c889f6aafb66d738820dae0aa
116093Author: Daniel Stone <daniel@fooishbar.org>
116094Date: Fri Jul 21 19:57:28 2006 -0400
116095
116096 loader: walk directory paths with readdir(), don't stat() everything
116097 Walk the directories with readdir, and don't stat everything we can
116098 find. Thanks to davej for the public humiliation reminding me to go back
116099 and re-fix this one.
116100
116101commit e18d34f4238e13e226b0407fa2f5f77d2038de39
116102Author: Ian Romanick <idr@umwelt.(none)>
116103Date: Fri Jul 21 16:47:45 2006 -0700
116104
116105 Make the various implementations of xf86ExtendedInitInt10 use the
116106 libpciaccess interfaces. This eliminates all calls to mapPciRom,
116107 which in turn allows the elimination of hw/xfree86/int10/pci.c.
116108
116109commit 87a6346bf7f086b5f98b2b2ecd52f27efe864e56
116110Merge: b73fb2a 0486d39
116111Author: Daniel Stone <daniel@fooishbar.org>
116112Date: Fri Jul 21 19:36:25 2006 -0400
116113
116114 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
116115
116116commit 0486d3966d2888ef86d36c19f31bdbc2a3e8c652
116117Author: Daniel Stone <daniel@fooishbar.org>
116118Date: Fri Jul 21 19:35:04 2006 -0400
116119
116120 fix kbproto dependency
116121 Depend on kbproto >= 1.0.3, for unconditional definition of
116122 XkbSA_XFree86Private.
116123
116124commit b73fb2ae35a82e0bdd48f01132e971fb84946ff1
116125Merge: e7ac27a aff404f
116126Author: Daniel Stone <daniel@fooishbar.org>
116127Date: Fri Jul 21 19:30:26 2006 -0400
116128
116129 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
116130
116131commit e7ac27ad81efbea6128b3cec443ca98e228d14ad
116132Author: Daniel Stone <daniel@fooishbar.org>
116133Date: Fri Jul 21 19:29:28 2006 -0400
116134
116135 revert accidental deletion of lnx_io.c; re-delete lnx_kbd.c
116136 Thinko'd which file to remove after merging from master.
116137
116138commit 81913a12910e39d7ea6af8657c1c66cc6791cd65
116139Author: Daniel Stone <daniel@fooishbar.org>
116140Date: Fri Jul 21 19:10:26 2006 -0400
116141
116142 remove undead files from master
116143 Remove dead files which worked their way back in when merging from master.
116144 Ugh.
116145
116146commit 7465010d59ec435bd00b738f0cef766b352dc7eb
116147Merge: 1d31ed7 0aaac95
116148Author: Daniel Stone <daniel@fooishbar.org>
116149Date: Fri Jul 21 19:05:41 2006 -0400
116150
116151 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into input-hotplug
116152
116153commit 1d31ed778284082e1060bff63317c94581d9eb9b
116154Author: Daniel Stone <daniel@fooishbar.org>
116155Date: Fri Jul 21 19:02:52 2006 -0400
116156
116157 xephyr: load keysyms at init, not enable
116158 Load keysyms at init time, not enable, so we don't get the wrong map width.
116159
116160commit 63dfaa1d5ba556e09314ec914936e5471aab94b0
116161Author: Adam Jackson <ajax@theobromine.nwnk.net>
116162Date: Fri Jul 21 18:47:18 2006 -0400
116163
116164 Delete internal usage of the symbol ref/req lists.
116165
116166commit bca9364f3f4a2376edbcf57a34f704ce28be21ba
116167Author: Adam Jackson <ajax@theobromine.nwnk.net>
116168Date: Fri Jul 21 18:41:46 2006 -0400
116169
116170 Remove the loader's required and referenced symbol lists, dead code.
116171
116172commit 07ad92d2c4cb07db8487d76efc822fd7b88137cb
116173Merge: 18624a9 22db3fd
116174Author: Ian Romanick <idr@umwelt.(none)>
116175Date: Fri Jul 21 15:25:35 2006 -0700
116176
116177 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116178
116179 Conflicts:
116180
116181 hw/xfree86/common/xf86DoScanPci.c
116182 hw/xfree86/common/xf86Init.c
116183 hw/xfree86/common/xf86pciBus.c
116184
116185commit aff404f293ed86a44a093a51a9f11e79e6c3f4f6
116186Author: Adam Jackson <ajax@theobromine.nwnk.net>
116187Date: Fri Jul 21 18:24:37 2006 -0400
116188
116189 Detach xf4bpp from cfb.
116190
116191commit 0aaac95b0d12089b256c97f6ff955c8c229ae095
116192Author: Adam Jackson <ajax@theobromine.nwnk.net>
116193Date: Fri Jul 21 17:56:00 2006 -0400
116194
116195 Remove RCS tags. Fix Xprint makefile braindamage.
116196
116197commit eeaad0e956640aac653d194a992df7792e4abcbb
116198Author: Aaron Plattner <aplattner@nvidia.com>
116199Date: Thu Jul 20 18:19:07 2006 -0700
116200
116201 Fix the RandR failure path for rotated screens.
116202
116203commit 22db3fdb54d2f7f6b72638b46c186af6db04e214
116204Merge: 2f98841 93cd538
116205Author: Ian Romanick <idr@umwelt.(none)>
116206Date: Fri Jul 21 13:55:37 2006 -0700
116207
116208 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
116209
116210commit 3eeb62e8f587732e6b433c2b9c6879eb26a3f1b4
116211Author: Kristian Høgsberg <krh@redhat.com>
116212Date: Fri Jul 21 16:33:28 2006 -0400
116213
116214 bug #890: completely remove deprecated keyboard driver
116215 Remove all remnants of the old built-in keyboard driver.
116216
116217commit 60ea7b51fe2b8a19a08b63db48504971a7a50ee6
116218Author: Daniel Stone <daniel@fooishbar.org>
116219Date: Fri Jul 21 15:23:37 2006 -0400
116220
116221 xorg ddx: move to new input API, remove old keyboard driver
116222 Remove most of the rest of the old keyboard driver.
116223 Move to the new Get{Keyboard,Pointer}Events API, which is mostly
116224 complete at this stage: just missing the proximity events.
116225
116226commit d32dc8bf19e1071fc30af2f0bc6a6699b351f39e
116227Author: Daniel Stone <daniel@fooishbar.org>
116228Date: Thu Jul 20 18:40:47 2006 -0400
116229
116230 be more careful in IVAS
116231
116232 Don't walk off the end of a NULL pointer in InitValuatorAxisStruct.
116233
116234commit 7711c56d2e2aeae4dcd6d9297bc144c1cb3cfab1
116235Author: Daniel Stone <daniel.stone@nokia.com>
116236Date: Thu Jul 20 18:38:57 2006 -0400
116237
116238 xephyr: fix keymap, pointer; dix: fix multiple axes
116239
116240 Initialise our axes properly in the DIX, and make sure we don't
116241 unnecessarily clip maxval when it's not set.
116242 Fix keymap copying in Xephyr (to some degree: it's still broken),
116243 and set nAxes and nButtons properly.
116244
116245commit f18c3122a57df9770087e5fc70ac488552222233
116246Author: Daniel Stone <daniel.stone@nokia.com>
116247Date: Thu Jul 20 16:49:17 2006 -0400
116248
116249 sanitise debug output
116250
116251 Don't spit out huge chunks of pointless ephemera, especially without
116252 --enable-debug.
116253
116254commit 463e0fe35bec3c91b19be9aacf34babb146a88c9
116255Author: Daniel Stone <daniel.stone@nokia.com>
116256Date: Thu Jul 20 16:45:15 2006 -0400
116257
116258 update KDrive to new input API
116259
116260 Update KDrive to fit the new API (mieqInit and InitPointerDeviceStruct), and
116261 include InitTouchscreenDeviceStruct in the DIX.
116262
116263commit 1987af8c498a1bf394a8951ca6d5b0b7f7a35188
116264Author: Daniel Stone <daniel.stone@nokia.com>
116265Date: Thu Jul 20 16:39:54 2006 -0400
116266
116267 add virtual core devices to DIX
116268
116269 Add virtual core devices, with proper keymaps etc, to the DIX.
116270
116271commit 737e6e4836a6af26fedc22cda8e7d366b52c8fa7
116272Author: Kristian Høgsberg <krh@redhat.com>
116273Date: Wed Jul 19 20:10:48 2006 -0400
116274
116275 define SDevicePresenceNotifyEvent prototype
116276
116277commit 7f36cc533e7f6ae44e973c5f00f9bfec7c6b7b50
116278Author: Daniel Stone <daniel@fooishbar.org>
116279Date: Wed Jul 19 20:09:13 2006 -0400
116280
116281 distribute config.h
116282
116283commit e896195eab726a2b307200958308eda8c93dd3cd
116284Author: Daniel Stone <daniel.stone@nokia.com>
116285Date: Wed Jul 19 20:05:33 2006 -0400
116286
116287 remove extraneous font debugging code
116288
116289 Some of it didn't compile, and some of it was just unnecessary.
116290
116291commit 99c57674c002c5e88c6db34488a27b05004c9197
116292Author: Daniel Stone <daniel.stone@nokia.com>
116293Date: Wed Jul 19 19:59:11 2006 -0400
116294
116295 avoid using font servers with built-in fonts
116296
116297commit ecfad74c48f633916305bcc25baaaad74aa52b78
116298Author: Daniel Stone <daniel.stone@nokia.com>
116299Date: Wed Jul 19 19:55:13 2006 -0400
116300
116301 add support for built-in fonts
116302
116303 Use --enable-builtin-fonts to only use built-in fonts, and avoid loading
116304 fonts.
116305
116306commit 0a2068d123520d35818c38a555ae3ba06d8ca7fb
116307Author: Daniel Stone <daniel.stone@nokia.com>
116308Date: Wed Jul 19 17:29:23 2006 -0400
116309
116310 Xi: add XExtension{Keyboard,Pointer} types
116311
116312 Report XExtensionKeyboard for non-core keyboards, and XExtensionPointer for
116313 non-core pointers/mice.
116314
116315commit c7577f9b88aac84d59404f29d994ee7af583d33b
116316Author: Kristian Høgsberg <krh@redhat.com>
116317Date: Wed Jul 19 17:27:58 2006 -0400
116318
116319 Xi: add DevicePresenceNotify
116320
116321 Add support for DevicePresenceNotify events.
116322
116323commit 3a23e499017d5823157806029263edac53c663fd
116324Author: Daniel Stone <daniel.stone@nokia.com>
116325Date: Wed Jul 19 17:00:23 2006 -0400
116326
116327 make XInput mandatory
116328
116329 Always build Xi, since GetPointerEvents/GetKeyboardEvents relies on it.
116330
116331commit 02d09105113fb9b560a770fe15f7bb041165831c
116332Author: Daniel Stone <daniel.stone@nokia.com>
116333Date: Wed Jul 19 16:51:04 2006 -0400
116334
116335 new KDrive input world order
116336
116337 Convert KDrive to GPE/GKE interface.
116338 Add first-class drivers and enumerate every device separately through
116339 Xi, instead of lamely attempting to aggregate them.
116340 Add XKB support to the Linux keyboard driver.
116341 Add 'thumb button' support to the tslib driver.
116342 Rejig InitInput, so each DDX has to add a list of drivers it supports.
116343 Support NewInputDeviceRequest, et al.
116344
116345commit a274e7296b1bdd6f6c921f28b087610cec9548e0
116346Author: Daniel Stone <daniel.stone@nokia.com>
116347Date: Wed Jul 19 13:56:23 2006 -0400
116348
116349 add GetPointerEvents/GetKeyboardEvents framework
116350
116351 Add GetPointerEvents (with XFree86 pointer acceleration) and GetKeyboardEvents
116352 to the DIX. Extend the ValuatorClass structure to account for same.
116353
116354commit b308dbf273f8c26361b0fee7aca64aec3245f60b
116355Author: Daniel Stone <daniel.stone@nokia.com>
116356Date: Wed Jul 19 12:15:18 2006 -0400
116357
116358 add DEVICE_TOUCHSCREEN and DEVICE_CORE Xi controls (DeviceIntRec ABI break)
116359
116360 Add DEVICE_TOUCHSCREEN and DEVICE_CORE controls to the Xi code, and the
116361 TouchscreenClassRec and a coreEvents flag, to toggle propagation of core
116362 events.
116363
116364commit c9a3d9baa81ceb940032ffe529d9eadf2d202ab2
116365Author: Daniel Stone <daniel@fooishbar.org>
116366Date: Wed Jul 19 11:41:16 2006 -0400
116367
116368 xorg DDX: implement NewInputDeviceRequest
116369
116370 Implement NewInputDeviceRequest for Xorg, mainly written by Kristian Høgsberg.
116371 Move MatchInput to xf86Helper.c, as xf86LookupInputDriver.
116372
116373commit 02a95311568e24e1055ea52c7df8cb7aa3f38ad0
116374Author: Daniel Stone <daniel@fooishbar.org>
116375Date: Wed Jul 19 10:05:12 2006 -0400
116376
116377 add basic D-BUS configuration mechanism
116378
116379 Also move LookupDeviceIntRec into the DIX, and add InputOption type, and
116380 NewInputDeviceRequest prototype (DIX requests DDX to add a device). Does not
116381 link without an implemented NIDR.
116382
116383commit 93cd53860c3aca182a0a02543c41b5d71d65926b
116384Author: Daniel Stone <daniel.stone@nokia.com>
116385Date: Thu Jul 20 16:52:31 2006 -0400
116386
116387 kdrive: allow debugging
116388
116389commit cd0874dda1c30ef91a7d2b3cd455676422599ccf
116390Author: Daniel Stone <daniel.stone@nokia.com>
116391Date: Wed Jul 19 20:13:02 2006 -0400
116392
116393 never define MEMBUG
116394
116395 Definining MEMBUG causes allocations to randomly fail.
116396
116397commit 093943d4d02f1dbc8935b8cf835866a6e3885193
116398Author: Daniel Stone <daniel.stone@nokia.com>
116399Date: Wed Jul 19 20:09:55 2006 -0400
116400
116401 define DEBUG in DIX
116402
116403 Which makes #ifdef DEBUG actually useful. Incredible.
116404
116405commit 68b0678254240a984db9adefefb0cf68e9bfd4e4
116406Author: Daniel Stone <daniel.stone@nokia.com>
116407Date: Wed Jul 19 20:08:32 2006 -0400
116408
116409 exa: only disable cw when COMPOSITE is built
116410
116411commit 27df2eda795681c9f05e2907d74e2c102d3441e4
116412Author: Daniel Stone <daniel.stone@nokia.com>
116413Date: Wed Jul 19 16:18:28 2006 -0400
116414
116415 fix KdXvCopyPackedData to actually work
116416
116417 Remove extraneous bit shift in KdXvCopyPackedData, so it's actually
116418 useful.
116419
116420commit 00b24f119f03da86fa98ffea545c5b041810ce53
116421Author: Daniel Stone <daniel@fooishbar.org>
116422Date: Wed Jul 19 17:01:20 2006 -0400
116423
116424 fix minor typo
116425
116426commit f8a7a1e40c14a85ebde11c5854c07a8d529d38b9
116427Author: Daniel Stone <daniel.stone@nokia.com>
116428Date: Wed Jul 19 17:06:00 2006 -0400
116429
116430 fix XEvIE build without XKB
116431
116432 Don't unconditionally play with XKB stuff in XEvIE.
116433
116434commit 2f98841fde6bad807967ed15e954291240714198
116435Author: Ian Romanick <idr@umwelt.(none)>
116436Date: Thu Jul 20 16:08:27 2006 -0700
116437
116438 Remove unused variable.
116439
116440commit 985c34bf06af70a7296db8307899a17347a25558
116441Author: Adam Jackson <ajax@theobromine.nwnk.net>
116442Date: Thu Jul 20 17:33:13 2006 -0400
116443
116444 Remove the DDXTIME conditional, for being unused.
116445
116446commit c69c00d6523a35232a32e54a533811fc2b37815a
116447Merge: 4636935 84683f1
116448Author: Ian Romanick <idr@umwelt.(none)>
116449Date: Thu Jul 20 12:08:38 2006 -0700
116450
116451 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
116452
116453commit 84683f19b4d1c712281036bcabf8dc623e64b26a
116454Author: Daniel Stone <daniel@fooishbar.org>
116455Date: Tue Jul 18 18:16:12 2006 -0400
116456
116457 get rid of XFree86LOADER, XFree86Server, XFree86Module, and IN_MODULE
116458 Get rid of almost all uses of these definitions. They're still defined for
116459 delinquent out-of-tree drivers, and also for the Mesa build. As well as
116460 for miinitext.c. But largely gone.
116461
116462commit 881953813c7307f2aac4057b48d233e5f4a574cd
116463Author: Adam Jackson <ajax@theobromine.nwnk.net>
116464Date: Mon Jul 17 13:50:38 2006 -0400
116465
116466 Fix the Linux ACPI reopen code to use a repeating timer, rather than a
116467 one-shot sleep-and-reopen attempt.
116468
116469commit f029e9a32dcaa95b84e08ec173a0cc78fd92bdbf
116470Author: Kristian Høgsberg <krh@redhat.com>
116471Date: Sat Jul 15 22:05:38 2006 -0400
116472
116473 Un-glx-libcwrap DMX GLX proxy so it works without GL/include.
116474
116475commit 46369350d40819ecc2a9f37ed4aaa95866b80997
116476Author: Ian Romanick <idr@umwelt.(none)>
116477Date: Fri Jul 14 15:51:55 2006 -0700
116478
116479 Add some const qualifiers to serveral function parameters.
116480
116481commit 490fb304599b1f24b36439e5c1397781e7d2f612
116482Author: Ian Romanick <idr@umwelt.(none)>
116483Date: Fri Jul 14 15:26:56 2006 -0700
116484
116485 Rearrange code in xf86int10ParseBiosLocation to use strncasecmp. This
116486 eliminates the need for the first use of xstrdup in this function.
116487 The second use of xstrdup was *never* necessary and has also been
116488 eliminated.
116489
116490commit d3ee49bcbafe4b4e6b308686020847e978473779
116491Author: Ian Romanick <idr@umwelt.(none)>
116492Date: Fri Jul 14 15:13:35 2006 -0700
116493
116494 Refactor common code from the generic.c and linux.c version of
116495 xf86ExtendedInitInt10 to xf86int10GetBiosLocationType and
116496 xf86int10GetBiosSegment.
116497
116498 These changes were tested on MGA hardware on x86-64 with various
116499 combinations of InitPrimary and BiosLocation.
116500
116501commit 8793c7fd4ba7d1b3e2eff3f2c18d042ee9bb3f62
116502Author: Ian Romanick <idr@umwelt.(none)>
116503Date: Fri Jul 14 09:11:39 2006 -0700
116504
116505 Refactor identical xf86InitInt10 function from generic.c and linux.c
116506 to helper_exec.c.
116507
116508commit 1450fd596433f7adfe3d0798dc2ddceb9d0a9034
116509Author: Ian Romanick <idr@umwelt.(none)>
116510Date: Fri Jul 14 09:10:32 2006 -0700
116511
116512 Trivial refactor of libint10_la_SOURCES.
116513
116514commit 18624a9b7cf88e769c53cc3a876b3647c5d5057c
116515Author: Ian Romanick <idr@umwelt.(none)>
116516Date: Thu Jul 13 21:10:48 2006 -0700
116517
116518 Modify linuxGetSizesStruct to use pci_device instead of pciConfigPtr.
116519
116520commit 12595a77d6c2f037f48dfe751bdc10505b3317df
116521Merge: 37838de 4c225a3
116522Author: Ian Romanick <idr@umwelt.(none)>
116523Date: Thu Jul 13 12:48:03 2006 -0700
116524
116525 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116526
116527commit 4c225a3a8b2e7e5e5510347d8473f1318bbac769
116528Author: James Steven Supancic III <arrummzen@arrummzen.net>
116529Date: Thu Jul 13 10:03:57 2006 -0400
116530
116531 Bug #7482: Fix Xdmx's Render code to match reality; fixes BadLength client
116532 crashes.
116533
116534commit bb3aa94845a74d7718ba9539bb76203ec82957fc
116535Author: Drew Parsons <dparsons@debian.org>
116536Date: Tue Jul 11 18:26:55 2006 -0700
116537
116538 Bug #7346: Disable Composite extension in Xprt
116539
116540 Xorg Bug #7346 <https://bugs.freedesktop.org/show_bug.cgi?id=7346>
116541 Patch #6184 <https://bugs.freedesktop.org/attachment.cgi?id=6184>
116542
116543commit 10f3e32726d5b4981abd1a3a022e5b4f219fb41e
116544Author: Gustavo Pichorim Boiko <boiko@mandriva.com>
116545Date: Mon Jul 10 16:37:53 2006 -0700
116546
116547 Fix the configure check for the --{enable,disable}-dpms option
116548
116549commit 2194d99d9ca3c607e0f5ddd911ee3df536d77564
116550Author: Matthieu Herrb <matthieu.herrb@laas.fr>
116551Date: Sun Jul 9 16:16:08 2006 +0200
116552
116553 Replace GNU make-ism.
116554
116555commit 90a9b82272446fdaebe71c966325fd7e670f75c6
116556Author: Matthew Allum <mallum@polystyrene.(none)>
116557Date: Sat Jul 8 21:57:07 2006 +0100
116558
116559 Fix crash in Xephyr when running on host X with keymap width < 4 ( i.e xvnc )
116560
116561commit 21e3e3ca298dce22e5fad6ef38aa6fe9736a1d3b
116562Merge: e805621 39b2f7b
116563Author: Matthew Allum <mallum@polystyrene.(none)>
116564Date: Sat Jul 8 21:13:52 2006 +0100
116565
116566 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
116567
116568commit e8056218944e873135c93039d1e9646d51364467
116569Author: Matthew Allum <mallum@polystyrene.(none)>
116570Date: Sat Jul 8 21:10:58 2006 +0100
116571
116572 Add support to Xephyr for simulating 8bit grayscale.
116573
116574commit 39b2f7b2182aedb1ab45415efb4c263012ace512
116575Author: Tilman Sauerbeck <tilman@code-monkey.de>
116576Date: Sat Jul 8 19:55:53 2006 +0200
116577
116578 Bug #3042: Use autoconf to get the correct name of a struct member.
116579
116580 This allows us to remove the kernel version ifdefs from the code, which
116581 are ugly and broken.
116582
116583commit 63f13e01ee6e7df1753f2113f4cff9538596be0a
116584Author: Tilman Sauerbeck <tilman@code-monkey.de>
116585Date: Sat Jul 8 11:33:44 2006 +0200
116586
116587 Bug #7097: do case-insensitive comparison for some hotkeys.
116588
116589 xkb's strcasecmp implementation has been moved to the dix so it's now
116590 safe to just use strcasecmp().
116591
116592commit 5416f90e9c939027005fc01fa3ce3df56919ae0d
116593Author: Kristian Høgsberg <krh@redhat.com>
116594Date: Thu Jul 6 21:22:34 2006 -0400
116595
116596 Implement GLX_MESA_copy_sub_buffer.
116597
116598commit b84374b2917a91a7732e780ffab6a29c807a3ecc
116599Author: Kristian Høgsberg <krh@redhat.com>
116600Date: Thu Jul 6 02:28:55 2006 -0400
116601
116602 Add GLX_MESA_copy_sub_buffer marshalling support.
116603
116604commit 2152e2d364bdd179cf218cde446c763d8c8bb833
116605Author: Ed Catmur <ed@catmur.co.uk>
116606Date: Thu Jul 6 17:45:36 2006 -0700
116607
116608 Remove hardcoded 'lib' in XPRINTDIR.
116609
116610commit 233c004641483a75985e09fea5416ab2c7a97fc4
116611Author: Ed Catmur <ed@catmur.co.uk>
116612Date: Thu Jul 6 17:41:53 2006 -0700
116613
116614 Generate xprint.pre from xprint.cpp.
116615
116616commit b3e4d1d1f4bcf900146d7b8cd19e008209294663
116617Author: Adam Jackson <ajax@benzedrine.nwnk.net>
116618Date: Thu Jul 6 18:51:29 2006 -0400
116619
116620 PPC64 build fix.
116621
116622commit f847f3c0a956c0aec8ade8e32f770daae147f40b
116623Author: Keith Packard <keithp@neko.keithp.com>
116624Date: Thu Jul 6 15:33:31 2006 -0700
116625
116626 xorgcfg now needs libxkbui version 1.0.2 or better.
116627
116628 Old versions of libxkbui use XkbStrCaseCmp which doesn't exist any longer;
116629 the server fails to link with that version of the libkbui library, so
116630 require the current version (1.0.2).
116631
116632commit 27ffd7e03c05dd2083a10e5acebd2b385d729eeb
116633Author: Keith Packard <keithp@neko.keithp.com>
116634Date: Thu Jul 6 14:43:10 2006 -0700
116635
116636 GL proto version 1.4.7 needed for texture-from-pixmap defines.
116637
116638 GLX_TEXTURE_TARGET_EXT and GLX_TEXTURE_RECTANGLE_EXT are defined as a part
116639 of the new texture from pixmap GL extension in gl proto version 1.4.7 and
116640 are now used by the X server code.
116641
116642commit d7a96dd6f18e54c26fc5881772d033ac078db3a1
116643Author: Adam Jackson <ajax@benzedrine.nwnk.net>
116644Date: Thu Jul 6 17:39:14 2006 -0400
116645
116646 Switch the default mouse device on Linux to /dev/input/mice.
116647
116648commit 37838de62edc474f44c0a48bd56470a6c23f8956
116649Merge: fd49a0d 8d07ee0
116650Author: Ian Romanick <idr@umwelt.(none)>
116651Date: Thu Jul 6 13:01:57 2006 -0700
116652
116653 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116654
116655 Conflicts:
116656
116657 hw/xfree86/os-support/bus/linuxPci.c
116658
116659commit 8d07ee070ecf0d403d9d27c80764d343b80af6f0
116660Author: Ian Romanick <idr@umwelt.(none)>
116661Date: Thu Jul 6 12:48:51 2006 -0700
116662
116663 Refactor linuxGetIOSize and linuxGetSizes. Eliminate the unnecessary
116664 optimization in the search loop.
116665
116666commit 704e645207d88a2d0a372cf69f6abd778ed4c30b
116667Author: Adam Jackson <ajax@benzedrine.nwnk.net>
116668Date: Thu Jul 6 14:22:33 2006 -0400
116669
116670 Remind dlloader that it needs to search the global scope as well as the
116671 loaded modules. Fixes LoaderSymbol() on symbols provided by the server.
116672 Spotted by Aaron Plattner.
116673
116674commit fd49a0dcc13280cf195bd6c7ee0f23d0840cb665
116675Merge: e66e7b4 28b95fd
116676Author: Ian Romanick <idr@umwelt.(none)>
116677Date: Thu Jul 6 09:52:15 2006 -0700
116678
116679 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116680
116681commit 28b95fd9d1c2f078aaaac75c310a27b17c74a6fc
116682Author: Kristian Høgsberg <krh@redhat.com>
116683Date: Thu Jul 6 03:25:38 2006 -0400
116684
116685 Drop unused GL/include subtree.
116686
116687 Woo, less Makefile's to generate.
116688
116689commit fc1a55671dceae0e2a701e044ff8203fae5eb1ba
116690Author: Kristian Høgsberg <krh@redhat.com>
116691Date: Thu Jun 29 04:35:45 2006 -0400
116692
116693 Implement glXCreateWindow and glXDestroyWindow.
116694
116695commit ee012588d28b468bd41da8d216210f8cb2bf8cb5
116696Author: Kristian Høgsberg <krh@redhat.com>
116697Date: Thu Jun 29 04:25:54 2006 -0400
116698
116699 Move __GLXdrawable lookup and creation into GetDrawableOrPixmap.
116700
116701 Also refactors __glXSwapBuffers to use GetDrawableOrPixmap for
116702 getting the __GLXdrawable. This patch paves the way for GLXWindows
116703 with XIDs different from the X Windows they are created for, a
116704 prerequisite for glXCreateWindow().
116705
116706commit 8b5bc6a9ab487fdea754266b120c686d75d9e100
116707Author: Kristian Høgsberg <krh@redhat.com>
116708Date: Thu Jun 29 00:05:01 2006 -0400
116709
116710 Drop global GLX error integer variables and use __glXError() instead.
116711
116712 Also drop glxerror.h (__glXError is now declared in glxserver.h)
116713 and global.c (last remaining globals are in glxext.c now).
116714
116715 With this change we now support all GLX 1.3 error codes.
116716
116717commit 7cf3ec7b59223f15314a0629f122ecb796678421
116718Author: Kristian Høgsberg <krh@redhat.com>
116719Date: Wed Jun 28 17:00:23 2006 -0400
116720
116721 Move createDrawable from __GLXcontext to __GLXscreen.
116722
116723commit eea8efe4516750b2505b52ebc9f769f5e8a6f94c
116724Author: Kristian Høgsberg <krh@redhat.com>
116725Date: Wed Jun 28 15:59:01 2006 -0400
116726
116727 Add marshalling for GLX 1.3 requests.
116728
116729 Also, hook up glXGetDrawableAttributes and glXQueryContext to existing
116730 DoGetDrawableAttributes and __glXQueryContextInfoEXT.
116731
116732commit e66e7b48e15543f3753951a19376f0446759963a
116733Merge: a9d4842 eb35f81
116734Author: Ian Romanick <idr@umwelt.(none)>
116735Date: Wed Jul 5 15:21:36 2006 -0700
116736
116737 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116738
116739commit eb35f812a5b65adcc5f6cbb91b31b69cae5d7f3d
116740Author: Greg Kroah-Hartman <gregkh@suse.de>
116741Date: Wed Jul 5 13:27:26 2006 -0700
116742
116743 add another file to .gitignore
116744
116745commit 863f5cc31b747bc9f2fcd6a9e20c613a11733bf4
116746Author: Greg Kroah-Hartman <gregkh@suse.de>
116747Date: Wed Jul 5 13:26:34 2006 -0700
116748
116749 fix compiler warning in hw/xfree86/common/xf86Config.c
116750
116751commit f059b61ab3af25b03c704669eddb838d3ce4366c
116752Author: Greg Kroah-Hartman <gregkh@suse.de>
116753Date: Wed Jul 5 11:47:25 2006 -0700
116754
116755 fix compiler warning about xnestRecolorCursor() not being defined
116756
116757commit cc3e99f747586f9d32622e5a682de39891b1fcba
116758Author: Greg Kroah-Hartman <gregkh@suse.de>
116759Date: Wed Jul 5 10:13:19 2006 -0700
116760
116761 fix some more compiler warnings due to defines being declared differently
116762
116763commit 59836c0f2abee3339e1aa30dacadb82e477943d6
116764Author: Greg Kroah-Hartman <gregkh@suse.de>
116765Date: Wed Jul 5 09:30:48 2006 -0700
116766
116767 fix wrong function pointer type in hw/dmx/dmxcmap.c
116768
116769commit 12563db59dd613ecc926e3bed9534152ebc0a2fb
116770Author: Eric Anholt <anholt@FreeBSD.org>
116771Date: Mon Jul 3 12:52:27 2006 -0700
116772
116773 Revert "Optimize out computing a gradient pixel if the mask value is 0."
116774
116775 This reverts cf46242e337481cd3b9b39d77dd621d2a63b11f9 commit. It wasn't meant
116776 to be pushed to master yet, and doesn't work.
116777
116778commit 002e28c12c74aa63777f65cbfb382c2bfd0d6850
116779Author: Eric Anholt <anholt@FreeBSD.org>
116780Date: Mon Jul 3 12:48:12 2006 -0700
116781
116782 Correct AGP memory deallocation argument on *BSD.
116783
116784 This fixes leaks and eventual crashes with RandR resizing on Intel.
116785
116786commit cf46242e337481cd3b9b39d77dd621d2a63b11f9
116787Author: Eric Anholt <anholt@FreeBSD.org>
116788Date: Wed Jun 28 18:35:59 2006 +0200
116789
116790 Optimize out computing a gradient pixel if the mask value is 0.
116791
116792 Obtained from: kdrive CVS (DavidR XGL fb/ megapatch)
116793
116794commit a838fb70c52a829872680f6a2a2e7dd6d2dc9247
116795Author: Eric Anholt <anholt@FreeBSD.org>
116796Date: Mon Jul 3 19:22:36 2006 +0200
116797
116798 Bump server version to 7.1.99.2 for gradient and repeat fixes.
116799
116800commit 25d871d98462f0481ee419295ddc94b8c79dc881
116801Author: Eric Anholt <anholt@FreeBSD.org>
116802Date: Mon Jul 3 19:22:26 2006 +0200
116803
116804 Fix source picture filter check for multiple screens.
116805
116806 Now, we only check for filter commonality if we're operating on a source
116807 picture, and we compare the id (screen-independent index of the filter name)
116808 rather than the pointer to the filter (per-screen state).
116809
116810commit 7106a77df37c06d2b5568eceeb9297096bff3137
116811Author: Eric Anholt <anholt@FreeBSD.org>
116812Date: Sun Jul 2 12:41:35 2006 +0200
116813
116814 Fix bugs in support for new repeatTypes in XAA and EXA.
116815
116816 EXA now won't pass pictures with new repeatTypes to drivers. We can add a flag
116817 for them to support it at a later time.
116818
116819commit f5e92542a14f51029347b6476e4e4af69144930b
116820Author: Eric Anholt <anholt@FreeBSD.org>
116821Date: Fri Jun 30 12:03:47 2006 +0200
116822
116823 Bug #7366: Fix crashes when setting filters on source pictures.
116824
116825 Now, filters may only be set on source pictures when the filter is common to
116826 all screens. Also, like SetPictureTransform, ChangePictureFilter is now not
116827 called on source pictures.
116828
116829commit 6ef457913955d4289081c7d07d528963ccf5272c
116830Author: Eric Anholt <anholt@FreeBSD.org>
116831Date: Fri Jun 30 03:01:14 2006 +0200
116832
116833 Bug #7366: Fix two crashes in operations on source pictures.
116834
116835 A screen's ChangePictureTransform now isn't called when changing the transform,
116836 as source pictures aren't associated with screens. Also, attempting to set
116837 an AlphaMap to a source picture will fail with BadMatch just like a Window
116838 would, preventing another crash.
116839
116840commit 50a3e1ad18c815a5adafee22beccdf970bae62d6
116841Author: Rudo Thomas <rudo@matfyz.cz>
116842Date: Sat Jul 1 12:34:36 2006 -0700
116843
116844 Missing close parenthesis in one of the setuid() fixes.
116845
116846commit 124a81eb389dfa510ac07ca93ee17c4c9d6e56ea
116847Merge: d3d6c5f 179737d
116848Author: Keith Packard <keithp@neko.keithp.com>
116849Date: Sat Jul 1 11:12:50 2006 -0700
116850
116851 Merge branch 'origin'
116852
116853commit d3d6c5f4d05e0ca5b566e19657e0fe2b3898482a
116854Author: Paul Mackerras <paulus@samba.org>
116855Date: Sat Jul 1 11:10:18 2006 -0700
116856
116857 Bug #7381: Coordinates get wrapped in accelerated line drawing on pixmap
116858
116859 XAAPolylinesWideSolid was adding the drawable origin onto each element in the
116860 pPts array. Since the values got stored back into the pPts array, they got
116861 truncated to 16 bits, causing the overflow I saw. This patch avoids storing
116862 the coords back into the pPts array (and actually reduces the size of the
116863 code too :). Now the 32-bit sum of coords + origin doesn't get truncated to
116864 16 bits, and the problem is solved.
116865
116866commit 179737d4a07ed10a734fe017b5680f8e78ffda96
116867Author: Jens Granseuer <jensgr@gmx.net>
116868Date: Wed Jun 7 01:46:00 2006 -0700
116869
116870 Bug 7145: fix build with gcc 2.95 & other c89 compilers
116871
116872 Move variable declarations to start of blocks as required by c89
116873
116874commit a9d4842893b65232d61fe4c63f074d30e81021d2
116875Merge: 66cac9b 6bd4c25
116876Author: Ian Romanick <idr@umwelt.(none)>
116877Date: Fri Jun 30 11:49:29 2006 -0700
116878
116879 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116880
116881commit 6bd4c254396cb0f4e8ae21ff455ebb15cd9f4f10
116882Author: Martin Bochnig <mb1x@gmx.com>
116883Date: Mon Jun 26 01:52:24 2006 +0200
116884
116885 Updated Solaris aperture driver to build on sun4v & amd64 kernels
116886 Updated for Solaris 10 changes to DDI
116887
116888commit 66cac9b17498497d939bfb7a4075bcee5a485e5f
116889Merge: 7c4e540 54d9acd
116890Author: Ian Romanick <idr@umwelt.(none)>
116891Date: Wed Jun 28 14:11:49 2006 -0700
116892
116893 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116894
116895commit 54d9acd5113318274e291abab4554b8e678227df
116896Author: Kristian Høgsberg <krh@redhat.com>
116897Date: Tue Jun 27 19:44:52 2006 -0400
116898
116899 Add damage tracking to GLX_EXT_tfp implementation.
116900
116901 - Only update when pixmap content actually change;
116902 - Only update the regions that acutally changed.
116903
116904 This is a worthwhile optimization, but it doesn't completely remove
116905 the bottleneck, as mesa still uploads then entire texture whenever
116906 it changes.
116907
116908commit adfe8e7437ff739f54d1d074008e8cc0e3bcb4d3
116909Author: Eric Anholt <anholt@FreeBSD.org>
116910Date: Tue Jun 27 21:49:00 2006 +0200
116911
116912 Bump server release to 7.1.99.1.
116913
116914 This will be important for a couple of cairo workaround tests.
116915
116916commit 63c169e3b1f7d6a7375a414fcd50cce32358a525
116917Author: Eric Anholt <anholt@FreeBSD.org>
116918Date: Tue Jun 27 04:11:47 2006 +0200
116919
116920 Fix MMX Saturate implementation.
116921
116922 The code was expanding the source blend factor from the wrong channel. Fixes
116923 cairo's clip-operator test.
116924
116925commit 7c4e54025345455f420cd2abaa7a40679689d3cb
116926Merge: 9d0a15d ff6b59a
116927Author: Ian Romanick <idr@umwelt.(none)>
116928Date: Mon Jun 26 14:39:44 2006 -0700
116929
116930 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver into pci-rework
116931
116932commit ff6b59a0dbadbe61a53e48c23965d3073d95791b
116933Merge: b3c8693 48c8715
116934Author: Alan Coopersmith <alan.coopersmith@sun.com>
116935Date: Mon Jun 26 13:02:33 2006 -0700
116936
116937 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
116938
116939commit b3c869304cd85af034aa9debaa874e29d14fcbe6
116940Author: Peter Breitenlohner <peb@mppmu.mpg.de>
116941Date: Mon Jun 26 10:48:44 2006 -0700
116942
116943 Free small, one-time memory leak in xdmcp -from handling
116944
116945 Part of Patch #6046 <https://bugs.freedesktop.org/attachment.cgi?id=6046>
116946
116947commit 48c871564d493203d434d5da015903399287f619
116948Author: Eric Anholt <anholt@FreeBSD.org>
116949Date: Mon Jun 26 15:57:32 2006 +0200
116950
116951 Move EXA_PM_IS_SOLID() to the public API, since drivers will want it frequently.
116952
116953commit afb84c2fca56887b3bfe7aa93f337c49b087acdc
116954Author: George Fufutos <fufutos610@hotmail.com>
116955Date: Sat Jun 24 15:23:14 2006 +0200
116956
116957 Bug #6911: Check return value of exaGetPixelFromRGBA().
116958
116959commit 930b9a069a425818d4e9965f53be7df1f4d7d422
116960Author: Michel Dänzer <michel@tungstengraphics.com>
116961Date: Sat Jun 24 15:21:17 2006 +0200
116962
116963 Bug #6818: Avoid infinite loop in exaLog2() with negative arguments.
116964
116965commit 36756fdb2ddc154b406f664a6af0f38d26e6973d
116966Author: Michel Dänzer <michel@tungstengraphics.com>
116967Date: Sat Jun 24 15:09:24 2006 +0200
116968
116969 Make sure <execinfo.h> is actually included when needed.
116970
116971 configure only defines HAVE_BACKTRACE, not HAVE_EXECINFO_H.
116972
116973 This could cause problems on platforms where the size of a pointer is greater
116974 than that of an integer, see
116975 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=363218 .
116976
116977commit 4426215a6e99f84550aaac23ac9c2018668bfbc1
116978Author: Michel Dänzer <michel@tungstengraphics.com>
116979Date: Sat Jun 24 15:02:56 2006 +0200
116980
116981 Bug #7213: Fix the XFree86-DRI extension for byte-swapped clients.
116982
116983 These clients are by definition non-local and thus not direct rendering
116984 capable, but they still need the QueryVersion and QueryDirectRenderingCapable
116985 requests to find out cleanly.
116986
116987commit a195a3debca02572d9f7d7a9976b5bf67acc5d08
116988Author: Michel Dänzer <michel@tungstengraphics.com>
116989Date: Sat Jun 24 14:54:52 2006 +0200
116990
116991 Fix byte swapping in some GLX requests.
116992
116993commit bc6cfde19887eff7a07dc739ffa29609fb55b83d
116994Author: Eric Anholt <anholt@FreeBSD.org>
116995Date: Fri Jun 23 20:07:34 2006 -0700
116996
116997 Use correct OSNAME setting so we can find os-specific modules like libdrm.
116998
116999commit c7ac485a59709572307b9a4a9abacc52c7021b65
117000Author: Eric Anholt <anholt@FreeBSD.org>
117001Date: Wed Jun 21 09:34:55 2006 -0700
117002
117003 Remove the default case from fbcompose.c switches which should cover all cases.
117004
117005 Instead, stick the NULL return default case afterwards, so that the compiler can
117006 warn us when we've got unimplemented cases. Removes some unimplemented and
117007 unused 8bpp, depth 4 picture format names.
117008
117009commit 2cf1f39ca974c81a2f52d2f7509aa3d098a87176
117010Author: Eric Anholt <anholt@FreeBSD.org>
117011Date: Wed Jun 21 09:30:59 2006 -0700
117012
117013 Add a manpage for EXA.
117014
117015commit 9d0a15dca56b14821cad474f9db9c332acb3b1e6
117016Merge: 6c514ca d67fd10
117017Author: Ian Romanick <idr@umwelt.(none)>
117018Date: Thu Jun 22 16:21:10 2006 -0700
117019
117020 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117021
117022 Conflicts:
117023
117024 hw/xfree86/loader/xf86sym.c
117025
117026commit 6c514ca14c2326c2cf82b73196429789d11e4a01
117027Author: Ian Romanick <idr@umwelt.(none)>
117028Date: Thu Jun 22 15:51:16 2006 -0700
117029
117030 Use XSERVER_LIBS instead of XORG_LIBS for -lpciaccess. The former is
117031 only for libraries that are built as part of the xserver build.
117032
117033commit d67fd106968e371d8be3966ed5ecdd3c69f36e3a
117034Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117035Date: Thu Jun 22 12:47:51 2006 -0400
117036
117037 Add kdrive servers to .gitignore
117038
117039commit bf17c6dede1c0cf1edee10f2cc7e1e619b944d3a
117040Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117041Date: Thu Jun 22 12:24:09 2006 -0400
117042
117043 Drop libz from the server's link line, it was only needed for LBX.
117044
117045commit c0cb8d1fb80540e093da54da3ee2f55bdf139274
117046Author: Alan Coopersmith <alan.coopersmith@sun.com>
117047Date: Wed Jun 21 18:12:41 2006 -0700
117048
117049 Use XACE, not XCSECURITY to decide if SecurityLookup* are exported
117050
117051commit 3177dc498a955cd58cd6054a7c7e69724db4a59b
117052Merge: 3f19803 91dcac5
117053Author: Alan Coopersmith <alan.coopersmith@sun.com>
117054Date: Wed Jun 21 18:06:06 2006 -0700
117055
117056 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117057
117058commit 3f19803e0b1adc66e695f63f915b8dc85eb84215
117059Author: Alan Coopersmith <alan.coopersmith@sun.com>
117060Date: Wed Jun 21 18:05:51 2006 -0700
117061
117062 APPGROUP requires both X-ACE & XC-SECURITY now
117063
117064commit 91dcac5295486cc55a34ad91704bfa483bd31eeb
117065Merge: d8135eb 77c947b
117066Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117067Date: Wed Jun 21 20:49:30 2006 -0400
117068
117069 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117070
117071commit d8135eb9e414bf6957f64c5102ee0ef7c2404c6f
117072Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117073Date: Wed Jun 21 20:49:21 2006 -0400
117074
117075 Unbreak unbreaking the loader. Re-add the symbol reference lists so that the
117076 linker will include everything it's supposed to. This is a terrible solution,
117077 but ld semantics don't let you do anything better.
117078
117079commit f83cee0338eca095ad601374a87775be823a2565
117080Author: Eamon Walsh <ewalsh@epoch.ncsc.mil>
117081Date: Wed May 5 20:07:37 2004 +0000
117082
117083 Modify XC-SECURITY and XC-APPGROUP extensions to work with XACE
117084
117085commit 15c9002d68a7eeb02a6db1f231af7a18a3cf7512
117086Author: Alan Coopersmith <alan.coopersmith@sun.com>
117087Date: Wed Jun 21 16:24:20 2006 -0700
117088
117089 Correct ifdef - should be XACE, not XSECURITY
117090
117091commit 90af38fa0c46c2081d2becac262a614c26ba6ef1
117092Merge: 3e098ef 77c947b
117093Author: Alan Coopersmith <alan.coopersmith@sun.com>
117094Date: Wed Jun 21 16:23:31 2006 -0700
117095
117096 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117097
117098 Conflicts:
117099
117100 Xext/appgroup.c
117101
117102commit 3e098efa35ba70ad4d5699af3130a3b02e1cb06e
117103Author: Alan Coopersmith <alan.coopersmith@sun.com>
117104Date: Wed Jun 21 16:21:03 2006 -0700
117105
117106 Correct ifdef - should be XACE, not XCSECURITY
117107
117108commit 77c947b900faf34f425eef1549d8210c475e093b
117109Author: Alan Coopersmith <alan.coopersmith@sun.com>
117110Date: Wed Jun 21 16:18:41 2006 -0700
117111
117112 Move Xserver internal API for appgroup from Xagsrv.h to appgroup.h
117113
117114 (Since all use is inside the xserver module, might as well keep the header
117115 in the Xserver module instead of in proto/XExt to allow easier synchronization
117116 of updates.)
117117
117118commit aa131c51e06e735d7b54cadf51205f703a3de4ef
117119Merge: 1708b0c 77aa701
117120Author: Ian Romanick <idr@umwelt.(none)>
117121Date: Wed Jun 21 14:39:52 2006 -0700
117122
117123 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117124
117125commit 1708b0c7d962b80073500eb1c483447bbf965bc9
117126Author: Ian Romanick <idr@umwelt.(none)>
117127Date: Wed Jun 21 14:39:39 2006 -0700
117128
117129 Eliminate "ugly" spaces around parenthesis. Add const qualifier.
117130 General code clean-ups.
117131
117132commit a3c26d01a16e4ae20a5d1acf9d888e95760c3da1
117133Author: Ian Romanick <idr@umwelt.(none)>
117134Date: Wed Jun 21 14:38:07 2006 -0700
117135
117136 Move declaration of "j" inside #ifdef block where it is used to
117137 eliminate compiler warning.
117138
117139commit 40aafaf154ba8a2cba857689d1481be2d4a610c2
117140Author: Greg Kroah-Hartman <gregkh@suse.de>
117141Date: Wed Jun 21 14:36:09 2006 -0700
117142
117143 fix a number of compiler warnings in os/*
117144
117145commit 77aa701e898c6525322cc4029d95167dd9f6e618
117146Author: Greg Kroah-Hartman <gregkh@suse.de>
117147Date: Wed Jun 21 14:16:48 2006 -0700
117148
117149 add some function prototypes to hw/xprint/DiPrint.h as they are exported
117150
117151commit 865884d050e1778180f7677e15f9ee1625ea4bb2
117152Author: Greg Kroah-Hartman <gregkh@suse.de>
117153Date: Wed Jun 21 14:16:28 2006 -0700
117154
117155 fix compiler warning in dix/xpstubs.c when XPRINT is not defined
117156
117157commit ed18d776f02e2ab235954501ef64936af9f9d909
117158Author: Kristian Høgsberg <krh@redhat.com>
117159Date: Wed Jun 21 16:22:14 2006 -0400
117160
117161 Fix #2488 for fb too: sample pixel center when transforming.
117162
117163commit ea5e0eabd1303a55d8fc10f44d21a3d371ce8919
117164Author: Matthias Hopf <mhopf@suse.de>
117165Date: Wed Jun 21 17:08:51 2006 +0200
117166
117167 Bug 4320: Fastpath corner case improvement for Composite.
117168
117169commit 9af315a9be19b48faa1249e0575cbe3d1e31dec5
117170Author: Eamon Walsh <ewalsh@epoch.ncsc.mil>
117171Date: Wed May 5 20:07:37 2004 +0000
117172
117173 Modify XC-SECURITY and XC-APPGROUP extensions to work with XACE
117174
117175commit f68ecfa482b2c2037f929710310c7b9ce6fe9291
117176Author: Alan Coopersmith <alan.coopersmith@sun.com>
117177Date: Tue Jun 20 19:01:34 2006 -0700
117178
117179 Add X-ACE to build system
117180
117181 - Added --disable-xace to configure.ac and issue configure error if trying
117182 to build XC-Security without X-ACE
117183 - Added XACE #define to dix-config.h
117184 - Added X-ACE sources to Xext/Makefile.am
117185
117186commit a46c06dab8392cf8012c7cc0b916de9a9e569671
117187Merge: 49b368c d44b2a0
117188Author: Alan Coopersmith <alan.coopersmith@sun.com>
117189Date: Tue Jun 20 18:40:18 2006 -0700
117190
117191 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117192
117193 Conflicts:
117194
117195 Xext/appgroup.c
117196 Xext/security.c
117197 dix/devices.c
117198 dix/dispatch.c
117199 dix/dixutils.c
117200 dix/events.c
117201 dix/extension.c
117202 dix/property.c
117203 dix/window.c
117204 os/access.c
117205
117206commit d44b2a0a57fb89741173c31676af0ccc822387dc
117207Author: Alan Coopersmith <alan.coopersmith@sun.com>
117208Date: Tue Jun 20 18:22:51 2006 -0700
117209
117210 Move Xserver API for security extension to securitysrv.h
117211
117212commit a54435946544a039fc333bb5e3438501d0d1ffc6
117213Author: Alan Coopersmith <alan.coopersmith@sun.com>
117214Date: Tue Jun 20 18:14:27 2006 -0700
117215
117216 Move Xserver API for security extension to securitysrv.h
117217
117218commit 49b368c0bb04816c4a3579071c596b2398cae3ec
117219Merge: 63f6e6b 481d401
117220Author: Alan Coopersmith <alanc@alf.(none)>
117221Date: Tue Jun 20 16:22:39 2006 -0700
117222
117223 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117224
117225commit 481d4012e74d9b0e98911f0ae02700ecf4cfc5ac
117226Author: Alan Coopersmith <alanc@alf.(none)>
117227Date: Tue Jun 20 16:16:19 2006 -0700
117228
117229 Don't add -ldl to XORG_LIBS if it's not needed for dlopen
117230
117231commit 63f6e6bbfd0d3677e29621af982c9392ead98dd7
117232Merge: 88ede2c 6df52fb
117233Author: Alan Coopersmith <alanc@alf.(none)>
117234Date: Tue Jun 20 16:03:34 2006 -0700
117235
117236 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117237
117238commit 6df52fb7745c185c0168060f69cc6b4f5315914e
117239Author: Alan Coopersmith <alanc@alf.(none)>
117240Date: Tue Jun 20 16:02:55 2006 -0700
117241
117242 Delete code that's been inside #if 0 since X11R6.7.
117243
117244commit 75398cff82b56c0a8fc301e84e2087f59a61a360
117245Merge: 1f91b57 2b58685
117246Author: Ian Romanick <idr@umwelt.(none)>
117247Date: Tue Jun 20 15:49:47 2006 -0700
117248
117249 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117250
117251 Conflicts:
117252
117253 hw/xfree86/loader/Makefile.am
117254 hw/xfree86/loader/xf86sym.c
117255
117256commit 88ede2cec79281a43cecb43ee6dec65770f82530
117257Merge: 227a319 2b58685
117258Author: Alan Coopersmith <alanc@alf.(none)>
117259Date: Tue Jun 20 15:29:55 2006 -0700
117260
117261 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117262
117263commit 227a3193405147fbbee2971cc15bac92cc13285a
117264Author: Eamon Walsh <ewalsh@epoch.ncsc.mil>
117265Date: Tue Jun 1 21:09:25 2004 +0000
117266
117267 Add XACE extension source files.
117268
117269 (Copied from XACE-SELINUX branch in Xorg monolith CVS since these were never
117270 imported to modular cvs or git trees.)
117271
117272commit 1f91b5726c721778de0627c0139100744db06934
117273Author: Ian Romanick <idr@umwelt.(none)>
117274Date: Tue Jun 20 15:03:13 2006 -0700
117275
117276 Fix a compiler warning about varibles being used without being set.
117277
117278 Add some const qualifiers.
117279
117280 Remove some "ugly" space around parenthesis.
117281
117282commit 2b58685402e70f123e131fd8146e6083ff5214a4
117283Author: Matthieu Herrb <matthieu@deville.herrb.com>
117284Date: Tue Jun 20 21:07:53 2006 +0200
117285
117286 Check setuid() return value. Bugzilla #7116.
117287
117288commit 4365d16c8629e824973ee0c357efdfbfd28b672c
117289Author: Kristan Høgsberg <krh@redhat.com>
117290Date: Mon Jun 19 22:13:22 2006 -0400
117291
117292 Pull over convolution filter fixes from xgl-0-0-1 branch.
117293
117294 Cherry-picking patches:
117295
117296 8a5ea68800b9d7dca90ff4e573ad8533852f1ea3 and
117297 4d755fe14274a7293748ce9aa666ab85df6297c5
117298
117299commit f818e0ab60da3779ab2602c6e6d3ff261b50917e
117300Author: Eamon Walsh <ewalsh@epoch.ncsc.mil>
117301Date: Wed May 5 20:07:37 2004 +0000
117302
117303 Modify XC-SECURITY and XC-APPGROUP extensions to work with XACE
117304 (cherry picked from 0106715000196c7b349a0b4494b61545f0f5e138 commit)
117305
117306commit 0707eb33d6826e1300a905edea28c12134600b12
117307Merge: 37f0ae0 98d17bb
117308Author: Alan Coopersmith <alanc@alf.(none)>
117309Date: Mon Jun 19 17:09:51 2006 -0700
117310
117311 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117312
117313commit 37f0ae02457bd22b27f8f30a373e5cb19f2fbaea
117314Author: Eamon Walsh <ewalsh@epoch.ncsc.mil>
117315Date: Wed May 5 20:04:52 2004 +0000
117316
117317 Replace XC-SECURITY code with XACE security hooks
117318 (cherry picked from 8526cd6395490b03b279f1962df777fb0e4a9878 commit)
117319
117320commit 98d17bba716619e3402bd74c7c3e1c01d055ef6e
117321Author: Alan Coopersmith <alanc@alf.(none)>
117322Date: Mon Jun 19 17:07:59 2006 -0700
117323
117324 Tell git to ignore emacs *~ droppings and git .msg files
117325
117326commit 569c808a2375be71f835ee8693605487484bd22e
117327Author: Eric Anholt <anholt@FreeBSD.org>
117328Date: Mon Jun 19 16:42:09 2006 -0700
117329
117330 Fix crash when using PICT_x4a4 by supplying an appropriate fbFetchPixel_x4a4.
117331
117332commit 8d9ccc90a54c786ca4ba5620ab0a965e3f3bc8ea
117333Author: Eamon Walsh <ewalsh@epoch.ncsc.mil>
117334Date: Wed May 5 20:15:41 2004 +0000
117335
117336 Add XACE extension
117337 (partially cherry picked from 8d4f21ab53c44ca48501d6211ea6db0c0b8af916 commit)
117338
117339commit 55426650417df4ec22ea0e2a67f6074f0ac1d54e
117340Author: Eric Anholt <anholt@FreeBSD.org>
117341Date: Mon Jun 19 15:04:46 2006 -0700
117342
117343 Clean up gcc warnings from picture format CARD32 -> enum change.
117344
117345commit 520c80f4b807ae6419e70fe2b524532465b509ac
117346Author: Eric Anholt <anholt@FreeBSD.org>
117347Date: Mon Jun 19 14:40:27 2006 -0700
117348
117349 Don't forget to step the rows when verifying the equivalence of fb/sys areas.
117350
117351 This is only used by fakexa, but we would have missed some errors without this
117352 fix.
117353
117354commit e793f0eeee3e9c83b6a7b50d451fb6db12839087
117355Author: Eric Anholt <anholt@FreeBSD.org>
117356Date: Mon Jun 19 14:06:02 2006 -0700
117357
117358 Correct component ordering when fetching [ax]4b4g4r4 pixels.
117359
117360 Noticed by: rendercheck
117361
117362commit e1672a12eb70836a2ceec803d505294897ae8cd2
117363Author: Eric Anholt <anholt@FreeBSD.org>
117364Date: Mon Jun 19 13:20:56 2006 -0700
117365
117366 Convert PICT_* names from #defines to an enum to aid in debugging.
117367
117368commit 9742d55c820a260a42a4537502295931d4529deb
117369Author: Greg Kroah-Hartman <gregkh@suse.de>
117370Date: Mon Jun 19 14:40:14 2006 -0700
117371
117372 update .gitignore to handle Xprint move
117373
117374commit d97a21acb878bc4e5e6542912fbd820503bba312
117375Author: Greg Kroah-Hartman <gregkh@suse.de>
117376Date: Mon Jun 19 14:36:54 2006 -0700
117377
117378 fix compiler warnings in hw/xfree86/i2c/fi1236.c
117379
117380commit 29c78321e86956c4ce0c1c899d82557f927e04da
117381Author: Greg Kroah-Hartman <gregkh@suse.de>
117382Date: Mon Jun 19 14:36:41 2006 -0700
117383
117384 fix compiler warning in hw/xfree86/i2c/tda9850.c
117385
117386commit 9f2793551f335e5fb08990fc8bb9e05e0ffb68d5
117387Author: Greg Kroah-Hartman <gregkh@suse.de>
117388Date: Mon Jun 19 11:50:47 2006 -0700
117389
117390 fix compiler warning in hw/vfb/InitOutput.c
117391
117392commit b20ae5ddb7682bafcee6f8bf0c8208a3f70b882b
117393Author: Greg Kroah-Hartman <gregkh@suse.de>
117394Date: Mon Jun 19 11:38:52 2006 -0700
117395
117396 fix compiler warnings in hw/xfree86/xf4bpp/ppcGC.c
117397
117398commit 4d258f31967141e3c4a6e4abbef89ffa717e85aa
117399Author: Greg Kroah-Hartman <gregkh@suse.de>
117400Date: Mon Jun 19 11:27:47 2006 -0700
117401
117402 fix compiler warnings in XTrap/xtrapdi.c
117403
117404commit 870cecb72c2cba44dc64cb202917453603c8f287
117405Author: Greg Kroah-Hartman <gregkh@suse.de>
117406Date: Mon Jun 19 11:22:42 2006 -0700
117407
117408 fix compiler warnings in XTrap/xtrapdiswp.c
117409
117410commit a28652f9c35fbc009245382a5cc2a022f42366fc
117411Author: Adam Jackson <ajax@theobromine.nwnk.net>
117412Date: Mon Jun 19 00:57:18 2006 -0400
117413
117414 Another round of loader sense-beating. Remove the (unused) server export
117415 lists, a really bad hash table, the last vestiges of the other backends,
117416 and some miscellaneous cleanups. Good for dropping 300k from the size of
117417 the built server on x86.
117418
117419commit 98a602fab1f307a07a96868d7dae12b6f8d7f405
117420Author: Adam Jackson <ajax@theobromine.nwnk.net>
117421Date: Sun Jun 18 23:54:04 2006 -0400
117422
117423 Don't bother building RAC as a module, that's just absurd.
117424
117425commit 76aaf7eae7409162c5ed2963f2e27d019cb30263
117426Author: Eric Anholt <anholt@FreeBSD.org>
117427Date: Sun Jun 18 19:47:29 2006 -0700
117428
117429 Add a couple of (doxygen) comments I wrote while looking at modesetting.
117430
117431commit 6aaf0e5b581b06fc73e56f863a26cd9d684eb9c0
117432Author: Eric Anholt <anholt@FreeBSD.org>
117433Date: Sun Jun 18 19:12:15 2006 -0700
117434
117435 Add options to disable EXA acceleration for Composite/UTS/DFS, and always print
117436
117437 out how much memory EXA is managing for offscreen pixmaps.
117438
117439commit 21ef7e17ef6dca177461c9438b9df707a4d664a2
117440Author: Eric Anholt <anholt@FreeBSD.org>
117441Date: Sun Jun 18 18:57:55 2006 -0700
117442
117443 Add some missing .gitignore stuff for Mesa symlinks and other generated files.
117444
117445commit 71fbda8049f64c7fefae8ab817fb5f37ee2ee134
117446Author: Adam Jackson <ajax@theobromine.nwnk.net>
117447Date: Sun Jun 18 21:07:28 2006 -0400
117448
117449 Xprint/ -> hw/xprint
117450 XpConfig -> hw/xprint/config
117451
117452commit 868e2cab706e317618646e064b0559d4e68c7b32
117453Author: Eric Anholt <anholt@FreeBSD.org>
117454Date: Fri Jun 16 10:17:51 2006 -0700
117455
117456 Add explicit dependencies (Xorg_DEPENDENCIES = ) on the internal libraries
117457 (such as libcw.la) that we link into the server, causing it to be rebuild
117458 automatically when they're updated. Some system libraries are included, but
117459 don't appear to cause any harm. You would think this would be automatic...
117460
117461commit 53f74b6aa95fe57fda45fd8a051595e772f00402
117462Author: Eric Anholt <anholt@FreeBSD.org>
117463Date: Fri Jun 16 10:14:30 2006 -0700
117464
117465 Bugzilla #5120, #7246: In CW's GC ops, validate the backing GC against the
117466 backing drawable if the serial numbers differ. Fixes crash in XAA which
117467 occurred when the DDX bumped the serial number on the backing drawable and
117468 expected it to get re-validated, and we didn't because the wrapped drawable
117469 hadn't been bumped.
117470
117471commit 72ac20c0e878ad48ff61f7e846cbf1005a4eb435
117472Merge: fc3e814 b900883
117473Author: Ian Romanick <idr@localhost.localdomain>
117474Date: Tue Jun 13 21:13:02 2006 -0700
117475
117476 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117477
117478commit b90088321e6ef84970aa97d7c851af93f49bf4b7
117479Author: Ian Romanick <idr@localhost.localdomain>
117480Date: Mon Jun 12 15:22:31 2006 -0700
117481
117482 Add arrayobj.c to the Makefile as well.
117483
117484commit f9f33b72e34eaeccea2a20f4a3dd68c2dbefc90e
117485Author: Michel Dänzer <michel@tungstengraphics.com>
117486Date: Mon Jun 12 20:19:11 2006 +0200
117487
117488 Track per-drawable damage to minimize UTS and DFS transfers.
117489
117490 Based on work by Eric Anholt.
117491
117492commit 6060b612de6b41f872d034c6130770c1d189d0a3
117493Author: Eric Anholt <eric@anholt.net>
117494Date: Mon Jun 12 20:12:31 2006 +0200
117495
117496 Provide option to report damage after operation is complete.
117497
117498commit fc3e81473a98d3b932faa73c0c01c21366a48a9e
117499Merge: 4bedaad 041ef23
117500Author: Ian Romanick <idr@localhost.localdomain>
117501Date: Mon Jun 12 09:41:06 2006 -0700
117502
117503 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117504
117505commit 041ef23192b193b87f6cfc3e74e2e77f9f47cd4b
117506Author: Ian Romanick <idr@localhost.localdomain>
117507Date: Mon Jun 12 09:39:18 2006 -0700
117508
117509 Add new Mesa files arrayobj.c and arrayobj.h.
117510
117511commit 4bedaad17dc75dfbada91f47e1007a8c090300a9
117512Author: Ian Romanick <idr@localhost.localdomain>
117513Date: Mon Jun 12 09:29:05 2006 -0700
117514
117515 Add new Mesa files arrayobj.c and arrayobj.h.
117516
117517commit 4e3c9ab8c90d474a2224dbfa0807bc01fc012f7c
117518Author: Ian Romanick <idr@localhost.localdomain>
117519Date: Fri Jun 9 16:56:54 2006 -0700
117520
117521 Remove some static declarations for things that were removed in
117522 previous commits.
117523
117524commit 8b6596b1e1864dd79a3c53ee3ceb7a8e9bdb6ec7
117525Author: Ian Romanick <idr@localhost.localdomain>
117526Date: Fri Jun 9 16:08:06 2006 -0700
117527
117528 Move probe_devices_from_device_sections,
117529 add_matching_devices_to_configure_list, and check_for_matching_devices
117530 from xf86Helper.c to xf86Init.c. These functions are only called from
117531 a single place in xf86Init.c, so it makes sense for them to live
117532 there. They were originally in xf86Helper.c because they evolved out
117533 another function that was already in that file.
117534
117535commit 7810d87cd6ae9316d88f6a2e7696deb8837cec27
117536Merge: dcfaa9f caad8b7
117537Author: Ian Romanick <idr@localhost.localdomain>
117538Date: Fri Jun 9 15:16:50 2006 -0700
117539
117540 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117541
117542commit caad8b724b97074e41de447fe77dda189f287a26
117543Author: Greg Kroah-Hartman <gregkh@suse.de>
117544Date: Fri Jun 9 11:24:57 2006 -0700
117545
117546 fix compiler warnings in record/set.c
117547
117548 (note this only fixed up the function definitions for the static functions
117549 which can not cause any abi incompatibility)
117550
117551commit 51489bb5ed86cb6aa07e26a13618765c29f913e4
117552Author: Greg Kroah-Hartman <gregkh@suse.de>
117553Date: Fri Jun 9 10:59:47 2006 -0700
117554
117555 more .gitignore updates
117556
117557commit c4d251bd3e88cf8dfd6872537dbe30c07344b196
117558Author: Greg Kroah-Hartman <gregkh@suse.de>
117559Date: Fri Jun 9 10:46:18 2006 -0700
117560
117561 updated .gitignore with more pre-generated files
117562
117563commit 9f31ef83be61a900c701fcbc9a43ffae40ca7005
117564Author: Greg Kroah-Hartman <gregkh@suse.de>
117565Date: Fri Jun 9 10:44:46 2006 -0700
117566
117567 fix compiler warning in hw/xfree86/loader/loaderProcs.h
117568
117569commit 490ffc205a7714145cac0c63efeb6374ea28141f
117570Author: Greg Kroah-Hartman <gregkh@suse.de>
117571Date: Fri Jun 9 10:44:25 2006 -0700
117572
117573 fix compiler warning in hw/xfree86/loader/loadmod.c
117574
117575commit 6119845d1ff832ea2b7c9cbe7ed6c6637cdcf305
117576Author: Greg Kroah-Hartman <gregkh@suse.de>
117577Date: Fri Jun 9 10:14:08 2006 -0700
117578
117579 fix compiler warnings in hw/xnest/Keyboard.c
117580
117581commit 1a7335ff932baa59a3283c50dd6007d81989b7e3
117582Author: Greg Kroah-Hartman <gregkh@suse.de>
117583Date: Fri Jun 9 10:02:08 2006 -0700
117584
117585 remove unneeded externs from Xprint/ps/psout.c
117586
117587commit 9583859d538394e98ac1f38b8e6f0997e321621d
117588Author: Greg Kroah-Hartman <gregkh@suse.de>
117589Date: Fri Jun 9 09:59:22 2006 -0700
117590
117591 remove unneeded "extern" in Xprint/ddxInit.c
117592
117593commit cb9e29c184474edd75645e3b52e22a097a242e40
117594Author: Greg Kroah-Hartman <gregkh@suse.de>
117595Date: Fri Jun 9 09:54:50 2006 -0700
117596
117597 removed unneeded extern in Xprint/ps/PsWindow.c
117598
117599commit 96c19a3ec1b7e43782d373b8015fa0ca24cb4f5b
117600Author: Greg Kroah-Hartman <gregkh@suse.de>
117601Date: Fri Jun 9 09:51:28 2006 -0700
117602
117603 put function prototype for ShapeExtensionInit() in proper place
117604
117605 based on FIXME in mi/miinitext.c
117606
117607commit dcfaa9f4f8fb10b75be50c9a49ccd88caf6d0f6a
117608Merge: af9afe4 576e6fb
117609Author: Ian Romanick <idr@localhost.localdomain>
117610Date: Thu Jun 8 15:03:00 2006 -0700
117611
117612 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117613
117614commit 576e6fb1124a47493371210adf99d6f2076e72c5
117615Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117616Date: Thu Jun 8 17:49:02 2006 -0400
117617
117618 Bug #7120: Multimonitor (non-Xinerama) support for xwin servers.
117619 (Tom Whittock)
117620
117621commit d42cf4a2a4980fdd29fb15c4fd8fddda67b36256
117622Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117623Date: Thu Jun 8 17:46:53 2006 -0400
117624
117625 Bug #7121: fix clipboard handling in Xming (Colin Harrison)
117626
117627commit af9afe4cc50df973bea2293a803f7dfdefaa6411
117628Merge: caffac3 3930da3
117629Author: Ian Romanick <idr@localhost.localdomain>
117630Date: Thu Jun 8 11:37:09 2006 -0700
117631
117632 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117633
117634commit 3930da3f6209312dd0f10aba0b16ef45996a07fe
117635Author: Greg Kroah-Hartman <gregkh@suse.de>
117636Date: Thu Jun 8 11:27:29 2006 -0700
117637
117638 fix compiler warnings in Xprint/ps/psout.c
117639
117640commit c496a3b9c981dc079fcc6c0ac4db3aa912b3dcf1
117641Author: Greg Kroah-Hartman <gregkh@suse.de>
117642Date: Thu Jun 8 11:23:53 2006 -0700
117643
117644 fix compiler warning in Xprint/ps/PsImageUtil.c
117645
117646commit ea24b5a25c2544f3b3de6480da125edb23a6b3a9
117647Author: Greg Kroah-Hartman <gregkh@suse.de>
117648Date: Thu Jun 8 11:22:01 2006 -0700
117649
117650 add proper function prototypes for Xprint/ps/PsCache.c to Xprint/ps/Ps.h
117651
117652commit ee2bb4d1929e20436cf0e830ece02fe07db2d524
117653Author: Greg Kroah-Hartman <gregkh@suse.de>
117654Date: Thu Jun 8 11:15:34 2006 -0700
117655
117656 fix compiler warnings in Xprint/ps/PsArea.c
117657
117658commit 92303d534a91cb0ea30e4cd0f639efd70b9739b4
117659Author: Greg Kroah-Hartman <gregkh@suse.de>
117660Date: Thu Jun 8 11:12:16 2006 -0700
117661
117662 fix compiler warning in Xprint/ps/PsGC.c
117663
117664commit cf6169f9e99e6e8ab264f284cfa13cb379b36207
117665Author: Greg Kroah-Hartman <gregkh@suse.de>
117666Date: Thu Jun 8 11:09:56 2006 -0700
117667
117668 fix compiler warnings in Xprint/ps/PsFonts.c
117669
117670commit 1abc7f96edf37a1e2c766b9cdba7fc9b2cb06d19
117671Author: Greg Kroah-Hartman <gregkh@suse.de>
117672Date: Thu Jun 8 11:06:51 2006 -0700
117673
117674 fix compiler warnings in Xprint/ps/PsInit.c
117675
117676commit 2dc291384c550badf55542ae645240e166676848
117677Author: Greg Kroah-Hartman <gregkh@suse.de>
117678Date: Thu Jun 8 11:04:40 2006 -0700
117679
117680 fixed compiler warnings in Xprint/ps/PsPixmap.c
117681
117682commit aef092e0290143c2b8b1cb98fdf55c9630032aaf
117683Author: Greg Kroah-Hartman <gregkh@suse.de>
117684Date: Thu Jun 8 10:53:17 2006 -0700
117685
117686 fix compiler warnings in Xprint/ps/PsPolygon.c
117687
117688commit 511b231ded61159ebd70cab020ca1ca003fd0784
117689Author: Greg Kroah-Hartman <gregkh@suse.de>
117690Date: Thu Jun 8 10:41:34 2006 -0700
117691
117692 fix compiler warnings in Xprint/ps/PsPrint.c
117693
117694commit 35fccb0068e8d73d1e6a16aefdc771506e620f83
117695Author: Greg Kroah-Hartman <gregkh@suse.de>
117696Date: Thu Jun 8 10:40:24 2006 -0700
117697
117698 remove some compiler warnings in Xprint/ps/PsText.c
117699
117700 Note that one of the existing warnings is pointing out a
117701 real bug (uninitialized use for fontPage in PsPolyText16())
117702 if anyone really cares about this code.
117703
117704commit caffac38432009e8dc01e7d3e72a7fa6fba2f078
117705Merge: c5b3aa1 9fa7372
117706Author: Ian Romanick <idr@localhost.localdomain>
117707Date: Thu Jun 8 10:38:43 2006 -0700
117708
117709 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117710
117711commit 1ac30947f4a222ba78558eddf8e5f03cec31f613
117712Author: Greg Kroah-Hartman <gregkh@suse.de>
117713Date: Thu Jun 8 10:35:18 2006 -0700
117714
117715 fix compiler warnings in Xprint/ps/PsWindow.c
117716
117717commit 9fa73721f0c3df73e508da909a5665f47a54cb57
117718Author: Greg Kroah-Hartman <gregkh@suse.de>
117719Date: Thu Jun 8 10:27:28 2006 -0700
117720
117721 fix up EnableDisableExtension() and EnableDisableExtensionError() prototypes
117722
117723commit c405659626477f2009603d8c0e381b8b62277bb6
117724Author: Greg Kroah-Hartman <gregkh@suse.de>
117725Date: Thu Jun 8 10:19:24 2006 -0700
117726
117727 comment out QualifyName in Xprint/Init.c which is not used anymore
117728
117729 If someone else wants to delete this function, that's fine with me too.
117730
117731commit a940b851faba569e36983f7885aefa72f7bf2ade
117732Author: Greg Kroah-Hartman <gregkh@suse.de>
117733Date: Thu Jun 8 10:17:53 2006 -0700
117734
117735 fix function pointer warning in Xprint/Init.c
117736
117737commit 9e0c82386ae389bcc296a5ad44e996790b033ad3
117738Author: Greg Kroah-Hartman <gregkh@suse.de>
117739Date: Thu Jun 8 10:16:37 2006 -0700
117740
117741 fix already-defined warning in Xprint/Init.c
117742
117743commit 40fb7eecaf14a76f35ac2dc350ad2fffdaf6e0d0
117744Author: Greg Kroah-Hartman <gregkh@suse.de>
117745Date: Thu Jun 8 10:11:17 2006 -0700
117746
117747 fix noDamageExtension warning in Xprint/Init.c
117748
117749 Also took out duplicate definition of this variable in mi/miinitext.c
117750
117751commit 494895e0fbbf0a71bc535c0a2358c9db54c95c5a
117752Author: Greg Kroah-Hartman <gregkh@suse.de>
117753Date: Thu Jun 8 09:43:44 2006 -0700
117754
117755 Properly define dispatchExceptionAtReset to fix compiler warnings
117756
117757commit d90eecf40ea768b2bf6340f15bb0af9dab2f3cf3
117758Author: Greg Kroah-Hartman <gregkh@suse.de>
117759Date: Thu Jun 8 09:11:41 2006 -0700
117760
117761 add some missing function prototypes to Xprint/AttrValid.h to fix compiler warnings
117762
117763commit bccde1609153dee93f6fe5a138fc0c0f2fe08212
117764Author: Greg Kroah-Hartman <gregkh@suse.de>
117765Date: Thu Jun 8 09:01:59 2006 -0700
117766
117767 fix incompatible pointer warning in Xprint/ddxInit.c
117768
117769commit 60bd8893d50ed1da9b94f4b96a07ea432e23f467
117770Author: Greg Kroah-Hartman <gregkh@suse.de>
117771Date: Thu Jun 8 09:01:21 2006 -0700
117772
117773 Properly #ifdef out ddxBeforeReset() to fix compiler warning
117774
117775commit ac21e6a594eac69101aa8920d70a9d60412b57f6
117776Author: Greg Kroah-Hartman <gregkh@suse.de>
117777Date: Thu Jun 8 09:00:42 2006 -0700
117778
117779 remove unused variable warning in Xprint/ddxInit.c
117780
117781commit 71dd44b0ad617dd36ce4ed328f9e1e8c5ef713a5
117782Author: Greg Kroah-Hartman <gregkh@suse.de>
117783Date: Wed Jun 7 19:17:26 2006 -0700
117784
117785 Fixed up most "warning: function declaration isn't a prototype" warnings from Xprint/
117786
117787commit 839305bac98856a2bb1d96691e4dcf49db229f90
117788Author: Greg Kroah-Hartman <gregkh@suse.de>
117789Date: Wed Jun 7 19:12:23 2006 -0700
117790
117791 Remove unused variables from Xprint/attributes.c
117792
117793commit c5b3aa120bf96ee169d07c33a2698499944a1624
117794Author: Ian Romanick <idr@localhost.localdomain>
117795Date: Wed Jun 7 17:46:23 2006 -0700
117796
117797 Added a comment about some suspicious code.
117798
117799commit 50112dfcc82dd476727fa453b9e3d1852e06d678
117800Author: Ian Romanick <idr@localhost.localdomain>
117801Date: Wed Jun 7 17:46:05 2006 -0700
117802
117803 Fixed severly broken calls to pci_device_find_by_slot. I mis-read the
117804 tag bits when I originally coded it. I mistakenly thought that the
117805 original code masked off the PCI function bits, so I just always
117806 passed 0 as the function. That was a mistake.
117807
117808commit f9e7128d948a0b62e6f03295305e0080d9233fd8
117809Author: Ian Romanick <idr@localhost.localdomain>
117810Date: Wed Jun 7 17:44:06 2006 -0700
117811
117812 Added missing include of Pci.h. This caused the server to terminate
117813 when the symbol PCI_DOM_FROM_TAG couldn't be resolved at run-time.
117814
117815commit 23182315f34e3a0065739b43b15d4560e75b41a1
117816Merge: bc05158 757f40f
117817Author: Ian Romanick <idr@localhost.localdomain>
117818Date: Wed Jun 7 17:03:48 2006 -0700
117819
117820 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117821
117822commit 7a40ac2585028860730ffcd333eb3fe9de63680c
117823Author: Greg Kroah-Hartman <gregkh@suse.de>
117824Date: Wed Jun 7 16:23:45 2006 -0700
117825
117826 remove a bunch of unused variables in Xprint/Init.c (fixing the compiler warnings)
117827
117828commit 163980138cc0bfc9124456781b3dc45a49e2a129
117829Author: Greg Kroah-Hartman <gregkh@suse.de>
117830Date: Wed Jun 7 16:11:20 2006 -0700
117831
117832 remove some unused local variables in Xprint/Oid.c
117833
117834commit 8e41640db884a4633b598d0a52b269e6547c8bf0
117835Author: Greg Kroah-Hartman <gregkh@suse.de>
117836Date: Wed Jun 7 15:56:43 2006 -0700
117837
117838 add bison generated files to .gitignore
117839
117840commit 78f4ab6b89fca3086b9c9471b40c11c23fbb6142
117841Author: Greg Kroah-Hartman <gregkh@suse.de>
117842Date: Wed Jun 7 14:12:40 2006 -0700
117843
117844 Fix compiler warning about undefined ReinitializeRootWindow function
117845
117846commit 757f40fca50a99377e437949ee77b983c8cd6087
117847Author: Greg Kroah-Hartman <gregkh@suse.de>
117848Date: Wed Jun 7 14:09:13 2006 -0700
117849
117850 updated the .gitignore file with more auto-generated files
117851
117852commit bc05158a5ab00c548e4b04b5638afe07bc702260
117853Merge: 46f55f5 cc46580
117854Author: Ian Romanick <idr@localhost.localdomain>
117855Date: Wed Jun 7 14:09:12 2006 -0700
117856
117857 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/xserver
117858
117859commit 46f55f5dead5d70cdff30531d80a72f6be042315
117860Author: Ian Romanick <idr@localhost.localdomain>
117861Date: Wed Jun 7 14:09:02 2006 -0700
117862
117863 Initial batch of changes for PCI rework. All future changes will be
117864 tracked individually.
117865
117866commit cc465800ddca5fb6c9ec09fdfa8f1f05359cf396
117867Author: Greg Kroah-Hartman <gregkh@suse.de>
117868Date: Wed Jun 7 14:03:35 2006 -0700
117869
117870 Fix compiler warnings about SetVendorRelease and SetVendorString
117871
117872commit 785c9789704ed142fe98cd17b5995e4a95b7141f
117873Merge: 21ebcfd 36d786e
117874Author: Greg Kroah-Hartman <gregkh@suse.de>
117875Date: Wed Jun 7 13:20:21 2006 -0700
117876
117877 Merge ../xserver
117878
117879commit 36d786e9f051c5c95c1cc8c098c84e118ed3cc85
117880Author: Greg Kroah-Hartman <gregkh@suse.de>
117881Date: Wed Jun 7 12:47:50 2006 -0700
117882
117883 add more files to .gitignore
117884
117885commit 8f5aa38abf1158a789b5528df9d98826342e30cf
117886Author: Greg Kroah-Hartman <gregkh@suse.de>
117887Date: Wed Jun 7 12:33:44 2006 -0700
117888
117889 fix compiler warning about XKB_IN_SERVER redefinition
117890
117891commit 101ae616962c355388722e05ab8413eb5f5c3402
117892Author: Greg Kroah-Hartman <gregkh@suse.de>
117893Date: Wed Jun 7 12:06:22 2006 -0700
117894
117895 Add PanoramiXExtensionDisabledHack to globals.h as it was missing.
117896
117897commit e5b72bd9c6fb06640a5de4031be0dc9b04b4b215
117898Author: Greg Kroah-Hartman <gregkh@suse.de>
117899Date: Wed Jun 7 12:05:39 2006 -0700
117900
117901 Remove 3 compiler warnings in the Xext/xevie.c file
117902
117903commit e3c11f66516521959127b9ab8fd88cc4c954f5bb
117904Author: Greg Kroah-Hartman <gregkh@suse.de>
117905Date: Wed Jun 7 12:05:01 2006 -0700
117906
117907 Added first cut at a .gitignore file to make using git easier.
117908
117909commit 21ebcfd7027b2a6182d4065e56a2ef814f5181ae
117910Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117911Date: Wed Jun 7 14:17:31 2006 -0400
117912
117913 Demolish now-unused loader functions.
117914
117915commit f90761b06eaa5fa44fe85289e54eed5f47eff3b9
117916Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117917Date: Wed Jun 7 13:58:24 2006 -0400
117918
117919 Add a token for EDID-supplied modes.
117920
117921commit d00aa6b8559d3e5f70c6558ce0abd12f7d758491
117922Author: Adam Jackson <ajax@benzedrine.nwnk.net>
117923Date: Wed Jun 7 11:44:36 2006 -0400
117924
117925 Delete a (now misleading) message from the crash handler.
117926
117927commit f8535edec736cf19740bd41ed2adfe531f2c26ac
117928Author: Keith Packard <keithp@neko.keithp.com>
117929Date: Tue Jun 6 10:29:34 2006 -0700
117930
117931 Remove ChangeLog file.
117932
117933commit 8444bb77c91cf8a23d32b3cc9749e2a3d3f9f9eb
117934Author: Daniel Stone <daniel@fooishbar.org>
117935Date: Mon Jun 5 20:22:06 2006 +0000
117936
117937 When we can, bound the maximum number of PCI devices to attempt to scan, by
117938 the number found on the system. Only implemented for Linux right now.
117939
117940commit 11cf4d2fde9219e8d6ca427acae89a0c9f5d71b1
117941Author: Keith Packard <keithp@keithp.com>
117942Date: Mon Jun 5 07:15:23 2006 +0000
117943
117944 Update AC_DEFINE_DIR to version compatible with CVS autoconf which requires
117945 double eval to avoid leaving ${prefix} in output
117946
117947commit 52fc7c8dc70226cc7f03454e9be86a627672295f
117948Author: Adam Jackson <ajax@nwnk.net>
117949Date: Mon Jun 5 03:00:24 2006 +0000
117950
117951 Ensure all *ModuleData symbols are marked _X_EXPORT. Start removing
117952 XFree86LOADER ifdefs, non-loadable hasn't been supported for a while
117953 now. Remove completely gratuitious REMOVE_LOADER_CHECK_MODULE_INFO
117954 ifdefs surrounding a call to a function added in XFree86 4.1 (!).
117955 Miscellaneous static markings.
117956
117957commit d22582dc5a070f72d4653e24d1e4ebe4a112276e
117958Author: Adam Jackson <ajax@nwnk.net>
117959Date: Sun Jun 4 16:13:26 2006 +0000
117960
117961 Remove a stray LBX reference.
117962
117963commit ddc6b99505e227f99585b8c2392da524022a73e6
117964Author: Daniel Stone <daniel@fooishbar.org>
117965Date: Sat Jun 3 11:24:33 2006 +0000
117966
117967 Bump to 1.1.99.2.
117968
117969commit 4fd668940f2155c4b06d24b6da8bcccd01f66f4c
117970Author: Daniel Stone <daniel@fooishbar.org>
117971Date: Sat Jun 3 10:58:37 2006 +0000
117972
117973 Bug #6619: Fix disappearing hardware cursor. (Colin Harrison)
117974
117975commit ee71cb61f8da29bcf36ea4b199d629e34f89b119
117976Author: Daniel Stone <daniel@fooishbar.org>
117977Date: Sat Jun 3 10:54:38 2006 +0000
117978
117979 Bug #6956: Fix crash when removing session leader before its children.
117980 (Rich Coe)
117981
117982commit cd384af3058fe15077c57eccdffed3b61e261e7f
117983Author: Daniel Stone <daniel@fooishbar.org>
117984Date: Sat Jun 3 10:50:23 2006 +0000
117985
117986 Completely remove relocation pointer table.
117987
117988commit 8e8c6faecddbe014d8760822e1f705b43a00fa33
117989Author: Daniel Stone <daniel@fooishbar.org>
117990Date: Sat Jun 3 10:48:37 2006 +0000
117991
117992 Add support for kFreeBSD systems. (Robert Millan, Aurelien Jarno)
117993
117994commit 5b703f847d166176920077c1e6ba1d9559fc8481
117995Author: Daniel Stone <daniel@fooishbar.org>
117996Date: Fri Jun 2 12:05:32 2006 +0000
117997
117998 Fix type confusion.
117999
118000commit 56f21bda1ce95741c88c423b60bd709eef26eb12
118001Author: Daniel Stone <daniel@fooishbar.org>
118002Date: Thu Jun 1 22:30:52 2006 +0000
118003
118004 Bug #6583: Only open /proc/bus/pci/devices once. (Bill Nottingham)
118005
118006commit a9ed5a87902a839a5a135af03db78f113b18bd86
118007Author: Daniel Stone <daniel@fooishbar.org>
118008Date: Thu Jun 1 22:06:41 2006 +0000
118009
118010 Kill LBX, too.
118011
118012commit df6da66525836d515f408a82f1a13ca5251ff0f7
118013Author: Daniel Stone <daniel@fooishbar.org>
118014Date: Thu Jun 1 20:56:39 2006 +0000
118015
118016 Simplify the unsupported XI function list.
118017
118018commit 97203f1cf6e5b7c6389f69cbb1b75ac675d09531
118019Author: Daniel Stone <daniel@fooishbar.org>
118020Date: Thu Jun 1 20:41:21 2006 +0000
118021
118022 Clean up a warning, and remove excess multiple-suffix code.
118023
118024commit 6d594ebc667afd404556ec3e108c810946b20ac5
118025Author: Daniel Stone <daniel@fooishbar.org>
118026Date: Thu Jun 1 20:22:39 2006 +0000
118027
118028 Ditch more alternate-loader braindamage.
118029
118030commit c9468177486833d521ec62c7b0266b4be8200de7
118031Author: Daniel Stone <daniel@fooishbar.org>
118032Date: Thu Jun 1 20:18:30 2006 +0000
118033
118034 Kill a.out, COFF and ELF loaders with FIRE.
118035
118036commit a3a4221495dfe4cc0a3874a08dd5364ef45a7f2e
118037Author: Adam Jackson <ajax@nwnk.net>
118038Date: Thu Jun 1 19:53:06 2006 +0000
118039
118040 Add bitmap to the ignored module list.
118041
118042commit ee689c104287140db38bbd26959ab1b4847c168e
118043Author: Daniel Stone <daniel@fooishbar.org>
118044Date: Thu Jun 1 19:49:55 2006 +0000
118045
118046 Ignore requests to load GLcore and speedo.
118047
118048commit 07c731a2bc21e6b98f28a2c0ebc42f01b67b824b
118049Author: Adam Jackson <ajax@nwnk.net>
118050Date: Thu Jun 1 19:37:53 2006 +0000
118051
118052 Bug #5089: Die, libbitmap, die!
118053
118054commit 32be08ba7242da74de5defd6a4dcb536a273f57a
118055Author: Daniel Stone <daniel@fooishbar.org>
118056Date: Thu Jun 1 19:22:38 2006 +0000
118057
118058 Remove horrendously ugly DDX backward-compatibility.
118059
118060commit d81edb9e00680e3c0001f343fa1d0c310b86cb93
118061Author: Daniel Stone <daniel@fooishbar.org>
118062Date: Thu Jun 1 19:22:01 2006 +0000
118063
118064 Forgot to remove this one too.
118065
118066commit a73e0f8cdfec1c9199ffe696146ba7d677c4c10d
118067Author: Daniel Stone <daniel@fooishbar.org>
118068Date: Thu Jun 1 18:47:47 2006 +0000
118069
118070 Die XTESTEXT1, die!
118071
118072commit fc5ca97284ef237a91f6adb433148ff57a673c08
118073Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118074Date: Tue May 30 22:56:36 2006 +0000
118075
118076 Powerpc machines also need ioperm_noop.c
118077
118078commit 07b168c8d8b2d79a4e7cf8dc5124eafc0f2bbcd0
118079Author: Adam Jackson <ajax@nwnk.net>
118080Date: Tue May 30 16:10:59 2006 +0000
118081
118082 Properly document the DPMS, SyncOnGreen, and TargetRefresh options.
118083
118084commit 107defd920d9b1eb52b15e8ca8665bc48bb933d6
118085Author: Kristian Høgsberg <krh@redhat.com>
118086Date: Mon May 29 19:53:18 2006 +0000
118087
118088 Remove superfluous definition of tfp tokens.
118089
118090commit fd8bde8bb0f9d796b3464973b53285c0a6d22a31
118091Author: Daniel Stone <daniel@fooishbar.org>
118092Date: Mon May 29 11:14:03 2006 +0000
118093
118094 Remove -xkbmap argument.
118095
118096commit db0680cf70b8367e1f8a7fff9c0f6ec414db0542
118097Author: Daniel Stone <daniel@fooishbar.org>
118098Date: Mon May 29 09:26:32 2006 +0000
118099
118100 Minor #include cleanups.
118101
118102commit c38aab293a06f43c04f14223f94f822d91d73396
118103Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118104Date: Sat May 27 23:15:05 2006 +0000
118105
118106 check buffersize before reading next char, fixes the "mouse stuck at left"
118107 bug (bugzilla #3113) for BSD systems using wscons. (Dale Rahn).
118108
118109commit 2102fdd0a58e18aeaa842b2ec73b6071970fafb7
118110Author: Jeremy C. Reed <reed@reedmedia.net>
118111Date: Fri May 26 00:12:18 2006 +0000
118112
118113 RGB color database and XErrorDB install to "share" not "lib" by default (by
118114 app/rgb and libX11).
118115 TODO: They are customizable, so maybe cpprules.in should be extended.
118116
118117commit 354086d7e8e13fc7acbcc6603ca2a03c8cc806ee
118118Author: Jeremy C. Reed <reed@reedmedia.net>
118119Date: Thu May 25 23:32:33 2006 +0000
118120
118121 Don't do fixup_video_driver_list if no drivers. (I had signal 11 and core
118122 dump when drivers not installed yet.)
118123 TODO: fix fixup_video_driver_list to handle NULL argument.
118124
118125commit c6b1cff43238deded11dc58945778aec3d844598
118126Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118127Date: Thu May 25 13:27:42 2006 +0000
118128
118129 update to build against Mesa CVS HEAD (Carlos Eduardo Rodrigues Diogenes).
118130
118131commit 99724c16c9c58eb3e20ba91c79464747b8ee3fcf
118132Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118133Date: Thu May 25 10:14:39 2006 +0000
118134
118135 typo
118136
118137commit 5c7aef148de23f39027fda647bbb53bb5b992683
118138Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118139Date: Thu May 25 09:57:47 2006 +0000
118140
118141 Don't destroy a pixmap twice on server exit (bugzilla #4247).
118142
118143commit 693079442377daa2dc48bf318a0d7dd256cb2738
118144Author: Jeremy C. Reed <reed@reedmedia.net>
118145Date: Wed May 24 22:58:03 2006 +0000
118146
118147 NetBSD also has curses.h instead of ncurses.h. (TODO: Maybe instead of
118148 checking for SCO, UnixWare, Sun SVR4 and NetBSD, the curses.h versus
118149 ncurses.h should be tested in configure.)
118150
118151commit 9477e5e0be04f4f6fa311de8b3693fbc8a082fbf
118152Author: Jeremy C. Reed <reed@reedmedia.net>
118153Date: Wed May 24 22:43:13 2006 +0000
118154
118155 On NetBSD, add -li386 for i386 and -lx86_64 for x86_64 to SYS_LIBS. (x86_64
118156 on NetBSD not tested.)
118157
118158commit b3031532ca96b22e81863202efb8bbcb9d701fac
118159Author: Jeremy C. Reed <reed@reedmedia.net>
118160Date: Wed May 24 20:11:38 2006 +0000
118161
118162 Remove tolower() which was missed in the _XkbStrCaseCmp/strcmp changes.
118163
118164commit eb696f72a0819edef550bce9ff55730c02f70452
118165Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
118166Date: Tue May 23 16:17:09 2006 +0000
118167
118168 Add AC_SUBST([SOLARIS_ASM_CFLAGS]) that I forgot when splitting them out of
118169 the XORG_CFLAGS.
118170
118171commit bc0c56c407117d1545e20d21f7d30eb3472d618b
118172Author: Adam Jackson <ajax@nwnk.net>
118173Date: Mon May 22 15:47:56 2006 +0000
118174
118175 Bug #6924: Restore the ABI for DrawableRec and ColormapRec to the state
118176 they were in prior to the fix for #6438. Based on a patch from Andy
118177 Ritger.
118178
118179commit cc3b882bd141218052cdde0144fc2a707ceee83d
118180Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
118181Date: Sat May 20 00:55:44 2006 +0000
118182
118183 Update to 2006-05-19 snapshot
118184 Remove entries now present in pci.ids
118185 Regenerate from updated pci.ids & extrapci.ids
118186
118187commit 0b2c2b6bba5b578d5f0ac2d709d5d2ce7cb32bef
118188Author: Adam Jackson <ajax@nwnk.net>
118189Date: Fri May 19 20:29:47 2006 +0000
118190
118191 Remove dead DPMS timer functions from the installed headers. (Fredrik
118192 Höglund)
118193
118194commit deebf6bd51117c01a3217f134bd952481b9e41ab
118195Author: Adam Jackson <ajax@nwnk.net>
118196Date: Thu May 18 23:52:51 2006 +0000
118197
118198 Bug #4139: Fix a BAR remapping bug that could lead to IERR and system hang.
118199 (Egbert Eich)
118200
118201commit 91239d83f4e27835cf871348b5ff6c892bd4f4f4
118202Author: Adam Jackson <ajax@nwnk.net>
118203Date: Thu May 18 23:52:41 2006 +0000
118204
118205 file getemptypci.c was initially added on branch server-1_1-branch.
118206
118207commit c11cfcfaffc79be5686c666f881c4c08f69e1b86
118208Author: Adam Jackson <ajax@nwnk.net>
118209Date: Thu May 18 23:48:57 2006 +0000
118210
118211 Bug #6377: Ignore disabled BARs, and allow matching BARs aligned to less
118212 than 16 bytes. (Felix Kühling, ATI)
118213
118214commit fa9a49a92db52de968d7147c71c6b9a8fd480f1e
118215Author: Adam Jackson <ajax@nwnk.net>
118216Date: Thu May 18 18:18:41 2006 +0000
118217
118218 Bug #5877: Avoid burning CPU when acpid dies. Require acpid to be running
118219 for ACPI support on Linux. Minor errno handling fixes. (Valery
118220 Inozemtsev, Adam Jackson)
118221
118222commit 7893dadb2f6df218a4f4ea30a41c1aa9838da1f0
118223Author: Adam Jackson <ajax@nwnk.net>
118224Date: Thu May 18 14:39:59 2006 +0000
118225
118226 Bug #6827: Fix texel fetch in fbFetchTransformed to avoid crashes. Still
118227 not 100% correct, but better than 7.0. (Radek Doulik, Matthias Hopf)
118228
118229commit 2892dd6d2e34957650ef1630a94d471dfa71f888
118230Author: Daniel Stone <daniel@fooishbar.org>
118231Date: Wed May 17 16:20:03 2006 +0000
118232
118233 Make Xv symbols conditional. (Enrico Wiegelt)
118234
118235commit a317bf482257f0e1b612dec7961fdfa564f0b9f2
118236Author: Daniel Stone <daniel@fooishbar.org>
118237Date: Wed May 17 15:00:18 2006 +0000
118238
118239 Make DBE support conditional. (Enrico Wiegelt)
118240
118241commit 321dbed5f5a857a23525167ab85d4d7699429132
118242Author: Adam Jackson <ajax@nwnk.net>
118243Date: Mon May 15 18:27:18 2006 +0000
118244
118245 Bug #5209: Fix APM/ACPI support, again. (Michel Dänzer, Valery Inozemtsev)
118246
118247commit dd38d3bd673cf830c2cd591fe1245909aa729892
118248Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118249Date: Sun May 14 16:22:24 2006 +0000
118250
118251 Don't use AM_CONDITIONAL inside shell conditionals. Bugzilla #6916.
118252
118253commit 6558ba4e62dba99f2a4d830f1c16f8d1c255b316
118254Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
118255Date: Fri May 12 21:53:28 2006 +0000
118256
118257 Fri May 12 17:51:26 2006 Søren Sandmann <sandmann@redhat.com>
118258 Keep track of the previous item and update its next pointer when deleting
118259 from the linked list.
118260
118261commit 9e239a0df7ededb50de091e5271cbfddd2b683c3
118262Author: Matthias Hopf <mhopf@suse.de>
118263Date: Thu May 11 11:23:43 2006 +0000
118264
118265 Bug #5796: Wrong fastpath selection for repeating sources.
118266
118267commit 9db5d2dfc33e384ea4db1b7cbc377e0b05cfb3ff
118268Author: Matthias Hopf <mhopf@suse.de>
118269Date: Thu May 11 10:18:08 2006 +0000
118270
118271 Bug #4320: Improved XAA Composite fastpath.
118272
118273commit 9a2a63ca3ff30d15e82a29e75a3720ba5b446978
118274Author: Adam Jackson <ajax@nwnk.net>
118275Date: Wed May 10 15:44:27 2006 +0000
118276
118277 Bug #3561: Crash fix in the Record extension. (Paul Anderson)
118278
118279commit ab1a0249ba5e3174f18a1db212bc511fd7d74cb0
118280Author: Adam Jackson <ajax@nwnk.net>
118281Date: Tue May 9 22:36:01 2006 +0000
118282
118283 Bug #6867: Yet another Render crash fix. (Michel Dänzer)
118284
118285commit 24310f827b71009c7510a674d2f92ced89847e37
118286Author: Adam Jackson <ajax@nwnk.net>
118287Date: Tue May 9 18:12:50 2006 +0000
118288
118289 Revert accidental commit
118290
118291commit 86ffb46358965509aa3ee536f15cb5a4e5e04426
118292Author: Adam Jackson <ajax@nwnk.net>
118293Date: Tue May 9 18:04:29 2006 +0000
118294
118295 Bug #5209: Re-enable building APM and ACPI support. (Michel Dänzer)
118296
118297commit b46d6a44fa97a3e66de828385026b7f84d9e59b8
118298Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
118299Date: Wed May 3 23:45:16 2006 +0000
118300
118301 Make Xephyr build on Solaris:
118302 Add #include <sys/file.h> on Solaris for FNONBLOCK/FASYNC definitions
118303 hw/kdriver/linux/Makefile.am Move agp.c & agp.h to KDRIVE_HW_SOURCES since
118304 they're not needed for Xephyr-only builds
118305 Add -lrt to XEPHYR_LIBS if needed to get nanosleep().
118306
118307commit fc91ca069dd55490b99b096f029e0864b049120c
118308Author: Adam Jackson <ajax@nwnk.net>
118309Date: Wed May 3 17:50:10 2006 +0000
118310
118311 Redact a few mentions of speedo font support.
118312
118313commit b9a9cf618566bdd796556b8a1f31949f66184352
118314Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118315Date: Tue May 2 14:09:30 2006 +0000
118316
118317 Typo in ALLLOCATE_LOCAL() arguments, causing mis-computation of the buffer.
118318 Bugzilla #6642.
118319
118320commit 1e8a594957d84a37e66183e9c0cb9d42b62bdb24
118321Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
118322Date: Tue May 2 01:37:25 2006 +0000
118323
118324 Fix Solaris build with Sun compilers to work when exa is built before
118325 hw/xfree86/os-support/solaris (as it is by default now).
118326
118327commit 724dbc2f8bbe2f21bf16f20ca7b8bb555516626c
118328Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
118329Date: Tue May 2 01:30:37 2006 +0000
118330
118331 Use min() [defined in include/misc.h] instead of MIN() [not defined in any
118332 Xorg header].
118333
118334commit ad124742ae2a265a54a4a7ac91709ec6fd6ced34
118335Author: Adam Jackson <ajax@nwnk.net>
118336Date: Sun Apr 30 20:33:27 2006 +0000
118337
118338 Remove stray mfb/cfb references.
118339
118340commit 291e89d4f2a4bb3177b2dfe6421680e23f120b8e
118341Author: Adam Jackson <ajax@nwnk.net>
118342Date: Sun Apr 30 19:16:14 2006 +0000
118343
118344 Remove NEED_LINEHELPER BC cruft for pre-R6 DDXes.
118345
118346commit fdcacc5a4bc1d6cb7347fc66041a9c686c5c74d1
118347Author: Eric Anholt <anholt@freebsd.org>
118348Date: Fri Apr 28 03:27:12 2006 +0000
118349
118350 Add Polylines and PolyPoint acceleration as well. This is primarily to
118351 clean up fallback debugging output, so I can focus on more imporant
118352 cases. Performance is comparable but without hardware stalls, and
118353 passes Xlib9.
118354
118355commit 005529a1c9c9e78f06565dff081f03b74988081e
118356Author: Eric Anholt <anholt@freebsd.org>
118357Date: Fri Apr 28 03:26:30 2006 +0000
118358
118359 Correct some bugs causing performance issues in the "Smart" scheme.
118360
118361commit 8738bc295bba229e36d064713e0c28aa8720c494
118362Author: Eric Anholt <anholt@freebsd.org>
118363Date: Thu Apr 27 20:27:27 2006 +0000
118364
118365 Improve EXA fallback debugging output to include the locations of pixmaps.
118366 This is being used in tracking down recent compositing performance
118367 regressions.
118368
118369commit 83b061776a57025076fc26d6d01fe2e049c2243b
118370Author: Eric Anholt <anholt@freebsd.org>
118371Date: Thu Apr 27 19:17:34 2006 +0000
118372
118373 Add trivial PolySegment acceleration for 0-width horizontal/vertical lines,
118374 which still happen somewhat frequently and were cluttering up my
118375 fallback debugging output. x11perf says it's a major performance win in
118376 those cases (though probably irrelevant), and it passes Xlib9.
118377
118378commit 69164ec00c749787dd59d5913ec6b3d159ad74d7
118379Author: Eric Anholt <anholt@freebsd.org>
118380Date: Thu Apr 27 02:15:19 2006 +0000
118381
118382 In drawing glyphs, shortcut our way to exaComposite instead of going
118383 through the whole CompositePicture stack and doing things like
118384 computing damage over again. This is a sizeable win for text drawing
118385 with a compmgr. Also avoid calling down into the server for dealing
118386 with the scratch pixmap when we are able to do UploadToScreen
118387 successfully and never need it.
118388
118389commit 3d4ca57b69c40d27fe191170d0819013f8cc4947
118390Author: Eric Anholt <anholt@freebsd.org>
118391Date: Wed Apr 26 18:27:40 2006 +0000
118392
118393 Add a helper for the Component Alpha Over case, which breaks the operation
118394 down into an OutReverse and an Add. Turn off the fallback to software
118395 glyphs when component alpha, now that we expect all (new) drivers to be
118396 able to support it. Also, make Xephyr fall back in the CA Over case to
118397 exercise this code. This speeds up my rgb24text and ls -lR in
118398 gnome-terminal by a factor of 5.
118399
118400commit 26fa45b64258894201496f921eccb0cb7028c28c
118401Author: Dave Airlie <airlied@linux.ie>
118402Date: Wed Apr 26 11:40:58 2006 +0000
118403
118404 Bug #6751: Use the Linux PCI ROM interface on Linux properly. The old
118405 domain code, still modified BARs not a good idea, Just talk to sysfs,
118406 if 0 read, fallback to old methods.
118407
118408commit c339b221d3f59130a39e63d4cec3de7e3de95bf3
118409Author: Dave Airlie <airlied@linux.ie>
118410Date: Wed Apr 26 11:31:07 2006 +0000
118411
118412 Bug #6750: This patch detects Intel bridges that are transparent but aren't
118413 reported as such. From the Linux kernel fixups. This patch also removes
118414 the reserved BIOS area from the area to allocate resources in.
118415
118416commit 6d156c044085881c6ecbd8a13521c0a26df71c30
118417Author: Eric Anholt <anholt@freebsd.org>
118418Date: Wed Apr 26 01:33:15 2006 +0000
118419
118420 Improve the migration debugging output.
118421
118422commit eaed7545a25a08b3223bf620c2ab6f80fe7cfdf6
118423Author: Eric Anholt <anholt@freebsd.org>
118424Date: Wed Apr 26 01:32:55 2006 +0000
118425
118426 Fix a bug in the intersection computation that could concievably cause
118427 incorrect results to be returned (but would probably usually be
118428 over-conservative).
118429
118430commit 5d00859c6e7d4b7ebce56f438ec4993334de2328
118431Author: Eric Anholt <anholt@freebsd.org>
118432Date: Tue Apr 25 23:56:17 2006 +0000
118433
118434 Bug #4668: Check if the lists of glyphs don't have any intersecting glyphs,
118435 and if they all have a maskFormat matching the format of the actual
118436 glyphs If so, we can avoid the temporary pixmap for accumulating
118437 glyphs, which reduces the number of operations done, and makes it
118438 easier on the migration system. This fixes some significant performance
118439 issues, particularly with subpixel antialiasing. Note that it does
118440 increase the amount of damage computation which is done, so is not
118441 always a win with a compositing manager running.
118442
118443commit 074dc9a023b3967ce00aa42c26a7c988423afe8a
118444Author: Eric Anholt <anholt@freebsd.org>
118445Date: Tue Apr 25 16:47:23 2006 +0000
118446
118447 Add an option to verify at the point of migration that pixmaps which aren't
118448 marked dirty are in fact not dirty. This will hopefully help catch
118449 issues like the previous commit. Leave it on in fakexa.
118450
118451commit 702d9226d57ec1584de2e8a85c268795650b1094
118452Author: Eric Anholt <anholt@freebsd.org>
118453Date: Tue Apr 25 15:46:04 2006 +0000
118454
118455 Don't forget to mark the drawable as dirty in exaPutImage(). Fixes
118456 corruption with drivers that have UTS. (Michel Dänzer)
118457
118458commit 39ca0867c7dd6f3bdecc52aec8df435946682098
118459Author: Donnie Berkholz <spyderous@gentoo.org>
118460Date: Mon Apr 24 05:44:06 2006 +0000
118461
118462 Remove another reference to cfb16.
118463
118464commit 81f3c2937df6230542f3223c201da8c41ff59945
118465Author: Donnie Berkholz <spyderous@gentoo.org>
118466Date: Mon Apr 24 05:39:47 2006 +0000
118467
118468 strlcpy() doesn't exist on Linux, so use the implementation in os/.
118469
118470commit a715634d23fb3124261dbbd8d7d4e6522551bb9b
118471Author: Matthieu Herrb <matthieu.herrb@laas.fr>
118472Date: Sun Apr 23 13:14:50 2006 +0000
118473
118474 Don't access free memory after unloading a module. Bugzilla #4168.
118475
118476commit 79dc6892610c9f8385cde4f0d601cc7481225c16
118477Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
118478Date: Sat Apr 22 03:22:17 2006 +0000
118479
118480 When reading the kernel keyboard mapping (readKernelMapping in
118481 os-support/linux/lnx_KbdMap.c) we overrun the usefully-named global
118482 array 'map', scribbling on other random static variables elsewhere.
118483 This is fixed by changing the size of at2lnx. (David Woodhouse). Bug
118484 #5169
118485
118486commit b37c515320dc8df2b1d160cc3f37d6bfac109b91
118487Author: Keith Packard <keithp@keithp.com>
118488Date: Wed Apr 19 21:56:13 2006 +0000
118489
118490 Add support for x4a4 format (depth 4 at 8bpp). Bug #6325.
118491
118492commit c947d796aad0b81d661a10b787deed967376da79
118493Author: Eric Anholt <anholt@freebsd.org>
118494Date: Tue Apr 18 19:18:43 2006 +0000
118495
118496 Missed in previous commit: Add a new migration scheme, called "Smart" for
118497 lack of a better name. This one behaves somewhat between Greedy and
118498 Always. It moves in if we can accelerate, unless the destination is
118499 clean and shouldn't be kept in framebuffer according to the score, in
118500 which case we migrate out (and force-migrate anything where migration
118501 is free). This should help fix lack of acceleration for drivers without
118502 UTS since removing exaAsyncPixmapGCOps, and has removed one performance
118503 trap with Radeon I'd noticed. It is the new default.
118504
118505commit b17a4de83e7ab18bef29ae898195889638f1cc6a
118506Author: Eric Anholt <anholt@freebsd.org>
118507Date: Tue Apr 18 19:14:07 2006 +0000
118508
118509 Add a new migration scheme, called "Smart" for lack of a better name. This
118510 one behaves somewhat between Greedy and Always. It moves in if we can
118511 accelerate, unless the destination is clean and shouldn't be kept in
118512 framebuffer according to the score, in which case we migrate out (and
118513 force-migrate anything where migration is free). This should help fix
118514 lack of acceleration for drivers without UTS since removing
118515 exaAsyncPixmapGCOps, and has removed one performance trap with Radeon
118516 I'd noticed. It is the new default.
118517
118518commit 771b366abe0bc060592b548612ec413291e14bf0
118519Author: Eric Anholt <anholt@freebsd.org>
118520Date: Tue Apr 18 18:50:35 2006 +0000
118521
118522 Fix exaGetPixmapFirstPixel to migrate as unaccelerated. Also adds a bit of
118523 fallback debugging info to PolyGlyphBlt.
118524
118525commit 782d61a03176264e0a9eb222ae97be3d175cf0ab
118526Author: Donnie Berkholz <spyderous@gentoo.org>
118527Date: Tue Apr 18 17:28:44 2006 +0000
118528
118529 Update all prefixes in config tools, so they mostly work again. Also, make
118530 RgbPath commented out when using xorgconfig. Start using
118531 /dev/input/mice as the default mouse location on Linux.
118532
118533commit 0f065059dcaf9c452f1cdec115f619f697fd71cc
118534Author: Donnie Berkholz <spyderous@gentoo.org>
118535Date: Mon Apr 17 07:27:43 2006 +0000
118536
118537 Wrap a couple more SDK headers in if XORG, as per Dave Airlie's commit on
118538 2006-01-18.
118539
118540commit ba632f697a782cd47870705b6cecaac2c60d30ff
118541Author: Donnie Berkholz <spyderous@gentoo.org>
118542Date: Mon Apr 17 07:10:31 2006 +0000
118543
118544 Fix kdrive build by linking in libexa before KDRIVE_LIBS.
118545
118546commit 0e62d92d5b809bc3a6503e9bc386cf961fc22557
118547Author: Adam Jackson <ajax@nwnk.net>
118548Date: Fri Apr 14 23:43:32 2006 +0000
118549
118550 Coverity #804: Another leak on OOM path.
118551
118552commit d61219aaadf9e4aa83644a69627d3a1d3282c95f
118553Author: Adam Jackson <ajax@nwnk.net>
118554Date: Fri Apr 14 23:38:11 2006 +0000
118555
118556 Coverity #806: Another memory leak on OOM path.
118557
118558commit 1b04e313920447e4c1f42bdd5a61f188d463210c
118559Author: Adam Jackson <ajax@nwnk.net>
118560Date: Fri Apr 14 23:32:22 2006 +0000
118561
118562 Coverity #847, #848, #849: Three more memory leaks.
118563
118564commit 6545051902f2ce00c98bd1373f97ebc942667e9c
118565Author: Adam Jackson <ajax@nwnk.net>
118566Date: Fri Apr 14 23:10:59 2006 +0000
118567
118568 Coverity #1003, #1004: Two more useless null checks.
118569
118570commit ab1d420022fb09d36a0d6ad948c38147c65b9adf
118571Author: Adam Jackson <ajax@nwnk.net>
118572Date: Fri Apr 14 23:09:38 2006 +0000
118573
118574 Coverity #1005: Avoid a null deref.
118575
118576commit d01e0956a8903fb41e8a34c78973b9b2860b6446
118577Author: Adam Jackson <ajax@nwnk.net>
118578Date: Fri Apr 14 23:08:10 2006 +0000
118579
118580 Coverity #1007: Fix a silly null check.
118581
118582commit 6d29f659318364afe046dc242d6f506ce40a944a
118583Author: Luc Verhaegen <libv@skynet.be>
118584Date: Fri Apr 14 23:01:35 2006 +0000
118585
118586 CVT means Coordinated Video Timing instead of Common.
118587
118588commit 82b6ea1a4b414426072bf001daeb3e9de0e93589
118589Author: Adam Jackson <ajax@nwnk.net>
118590Date: Fri Apr 14 22:51:19 2006 +0000
118591
118592 Bug #6580: Don't install xf86drm.h, that's libdrm's job.
118593
118594commit aefa347bded9a3179ab139d0ccddce314040e9b9
118595Author: Ian Romanick <idr@us.ibm.com>
118596Date: Thu Apr 13 21:08:25 2006 +0000
118597
118598 Fix build for added file to Mesa CVS. This is always fun. :(
118599
118600commit 6aadd454e70d83921685b58bf57ec30d95920734
118601Author: Daniel Stone <daniel@fooishbar.org>
118602Date: Mon Apr 10 10:11:19 2006 +0000
118603
118604 Fix stupid thinko.
118605
118606commit c9f6e60d42dec82d06995c05a2a011c338cadd87
118607Author: Daniel Stone <daniel@fooishbar.org>
118608Date: Mon Apr 10 08:50:33 2006 +0000
118609
118610 Coverity #826: Fix potential memory leak.
118611
118612commit 1357af2474be9a3bce7ee2350fd4252eee89a3b1
118613Author: Daniel Stone <daniel@fooishbar.org>
118614Date: Sun Apr 9 17:39:10 2006 +0000
118615
118616 Coverity #340: Fix potential NULL dereference. Clean up proliferation of
118617 'register int n' in loops of ProcXkbGetNames.
118618
118619commit f324be00c547effc698ae6679d12ffe90bd90e43
118620Author: Daniel Stone <daniel@fooishbar.org>
118621Date: Sun Apr 9 17:28:42 2006 +0000
118622
118623 Coverity #324: Fix potential NULL dereference. (Alan Coopersmith)
118624
118625commit 7637aa17f21e26d979fbb210a638d6751c98b1eb
118626Author: Daniel Stone <daniel@fooishbar.org>
118627Date: Sun Apr 9 17:26:17 2006 +0000
118628
118629 Coverity #169: Fix potential fgets() into NULL (?!?).
118630
118631commit d5bc41b88272b4a3a1841cc1189720b0549db215
118632Author: Daniel Stone <daniel@fooishbar.org>
118633Date: Sun Apr 9 17:15:51 2006 +0000
118634
118635 Coverity #323, #445, #446, #447: Fix potential NULL dereferences.
118636
118637commit 2387bfa5ff5ed82f3f732fb9152c1ea95850a914
118638Author: Aaron Plattner <aplattner@nvidia.com>
118639Date: Fri Apr 7 18:56:04 2006 +0000
118640
118641 Bump the ABI versions. Due to Glyph privates and the XV update below, the
118642 video driver ABI needs to be bumped to 1.0. The rest of the ABI minor
118643 versions were bumped to include the LoaderGetABIVersion function.
118644 Add a DrawblePtr argument to the XV hooks. This allows drivers to determine
118645 that the target window is redirected and draw to the appropriate place.
118646
118647commit dc43909219fe2a4d03139638814b89032b2921b9
118648Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
118649Date: Fri Apr 7 17:49:32 2006 +0000
118650
118651 Fri Apr 7 13:46:45 2006 Søren Sandmann <sandmann@redhat.com>
118652 Use FreeResource instead of deleteCompOverlayClient()
118653
118654commit 94e7213d594dbbb53a6bb05d1dab7514c4ff5350
118655Author: Adam Jackson <ajax@nwnk.net>
118656Date: Fri Apr 7 16:08:50 2006 +0000
118657
118658 Remove libc wrapper types from Xisb interfaces.
118659
118660commit 47bdc9528c2dd4ea9d59a0944c023173ea7a7a66
118661Author: Daniel Stone <daniel@fooishbar.org>
118662Date: Fri Apr 7 16:07:50 2006 +0000
118663
118664 Coverity #844, #845, #846: Fix memory leaks.
118665
118666commit 2c90c3bfef8563f739a72bb645dd52b35b6ff6d5
118667Author: Daniel Stone <daniel@fooishbar.org>
118668Date: Fri Apr 7 15:57:17 2006 +0000
118669
118670 Coverity #987: Avoid potential NULL dereference.
118671
118672commit 843146cfbaef234e13df9a62b6f0232a5efdf7f0
118673Author: Daniel Stone <daniel@fooishbar.org>
118674Date: Fri Apr 7 15:53:21 2006 +0000
118675
118676 Coverity #1216: Fix double-close of file on error.
118677
118678commit 5dacc822327689c0f096093756473c96fba67d76
118679Author: Keith Packard <keithp@keithp.com>
118680Date: Fri Apr 7 02:20:11 2006 +0000
118681
118682 Coverity #333, #334 - eliminate unncessary test for always true condition
118683 in fbEvenStipple.
118684
118685commit 75a9afdbf42e4196471774102e1758f18866bec6
118686Author: Adam Jackson <ajax@nwnk.net>
118687Date: Fri Apr 7 01:53:43 2006 +0000
118688
118689 Coverity #488: Avoid smashing an array on malformed config files.
118690
118691commit 20c1ef2cc30abe45eeaf5b0833cbc0095ed05c02
118692Author: Adam Jackson <ajax@nwnk.net>
118693Date: Fri Apr 7 01:50:07 2006 +0000
118694
118695 Coverity #769: Fix a potential memory leak for systems that allocate on
118696 malloc(0)
118697
118698commit 5ef711032b821be82fd7281fe64872bcbaff0327
118699Author: Adam Jackson <ajax@nwnk.net>
118700Date: Fri Apr 7 01:41:00 2006 +0000
118701
118702 Coverity #838: Plug two more memory leaks.
118703
118704commit 69477ea4b6e666940c5dd4422bedfa6432dead04
118705Author: Adam Jackson <ajax@nwnk.net>
118706Date: Fri Apr 7 01:37:11 2006 +0000
118707
118708 Coverity #837: Fix another another memory leak.
118709
118710commit b472ce7307dd88a21c7713a2b127e34f5c2bc817
118711Author: Adam Jackson <ajax@nwnk.net>
118712Date: Fri Apr 7 01:35:43 2006 +0000
118713
118714 Coverity #836: Fix another memory leak.
118715
118716commit 9c84ed5f8d9eded1a8b509c9cad1ca0ebcf2166a
118717Author: Adam Jackson <ajax@nwnk.net>
118718Date: Fri Apr 7 01:34:29 2006 +0000
118719
118720 Coverity #835: Plug memory leak in extension section parsing.
118721
118722commit 12924d0da36ad2266bb040caac58534c07e85261
118723Author: Adam Jackson <ajax@nwnk.net>
118724Date: Fri Apr 7 01:29:39 2006 +0000
118725
118726 Coverity #812: Fix parser memory leak.
118727
118728commit 49abff79957799e9229d5c0226ee1b0d7505003d
118729Author: Adam Jackson <ajax@nwnk.net>
118730Date: Fri Apr 7 01:26:33 2006 +0000
118731
118732 Coverity #818: Avoid memory leak on error path.
118733
118734commit bda292120fc97f890c1f58a31177c0f7c0bfa048
118735Author: Adam Jackson <ajax@nwnk.net>
118736Date: Fri Apr 7 01:23:50 2006 +0000
118737
118738 Coverity #985: Avoid segfault on malloc failure.
118739
118740commit 536628bb4bcb0a0d749e0c01412a5eb5d6d24063
118741Author: Adam Jackson <ajax@nwnk.net>
118742Date: Fri Apr 7 01:18:01 2006 +0000
118743
118744 Coverity #1037: Sanity check idx before use.
118745
118746commit 53e97ce4ddd993248561c245143b61915ea254b5
118747Author: Adam Jackson <ajax@nwnk.net>
118748Date: Thu Apr 6 22:04:12 2006 +0000
118749
118750 missed a line while removing cfb16
118751
118752commit 4ae12636694af05cee4287b119bde08e9ceaa8aa
118753Author: Adam Jackson <ajax@nwnk.net>
118754Date: Thu Apr 6 18:59:11 2006 +0000
118755
118756 Remove cfb16, no longer used.
118757
118758commit e1fc15a85fb367ee9afd63c920c3327c3f45158d
118759Author: Fredrik Höglund <fredrik@kde.org>
118760Date: Wed Apr 5 21:08:45 2006 +0000
118761
118762 Put the screensaver extension back in the Xext module.
118763 Move the screenSaverSuspended variable to DIX globals.
118764 Restore the old link order for the Xorg and Xdmx binaries.
118765
118766commit 383c2e1e9ec54ab9de356993ad552c1aa6ec094f
118767Author: Ian Romanick <idr@us.ibm.com>
118768Date: Wed Apr 5 19:52:12 2006 +0000
118769
118770 Include fbmmx.h in fb/fbwindow.c when USE_MMX is defined. Fixes build
118771 problem on x86-64 resulting from fbHaveMMX being a macro instead of a
118772 function on that platform.
118773
118774commit 4697da177d545a2f8bb6fd0d6588a1c40532c339
118775Author: Adam Jackson <ajax@nwnk.net>
118776Date: Tue Apr 4 18:30:28 2006 +0000
118777
118778 Initial checkin
118779
118780commit 83ea57bcc82f478a7ecdcd6ed73ca4be01cd9c26
118781Author: Adam Jackson <ajax@nwnk.net>
118782Date: Tue Apr 4 14:39:06 2006 +0000
118783
118784 Bug #5729: Convert xf8_16bpp to fb. chips(4) users please test.
118785
118786commit 4c7da861185080d15b3ff4301af4af0e85a71f93
118787Author: Adam Jackson <ajax@nwnk.net>
118788Date: Tue Apr 4 14:17:04 2006 +0000
118789
118790 Bug #5300: Fix missing spaces in the Build OS line in the log. (Egmont
118791 Koblinger)
118792
118793commit fb6f61b50f1c701041680e49f6a406a6603f1577
118794Author: Adam Jackson <ajax@nwnk.net>
118795Date: Tue Apr 4 12:36:16 2006 +0000
118796
118797 Bug #4806: Dump the raw EDID contents in hex to the log file for better
118798 debugging. (Philip Prindeville)
118799
118800commit 14af50371c7f23855781924cdf6afa6ab7566a87
118801Author: Adam Jackson <ajax@nwnk.net>
118802Date: Mon Apr 3 22:00:06 2006 +0000
118803
118804 Bug #2142: Make font path logging more readable. (Eduard Fuchs)
118805
118806commit 373f9f92566290d979730c09c9c5c5d50e23390c
118807Author: Adam Jackson <ajax@nwnk.net>
118808Date: Mon Apr 3 21:45:54 2006 +0000
118809
118810 Bug #4766: Convert all Xprint drivers to fb.
118811
118812commit d9b8bfbfafe8758ceb629606607e37546d51ca52
118813Author: Adam Jackson <ajax@nwnk.net>
118814Date: Mon Apr 3 21:16:30 2006 +0000
118815
118816 Bug #5478: More use of fbSOlidFillmmx. (Jim Huang)
118817
118818commit b0e67782653033c6518944adfbf23e466bd8bc39
118819Author: Adam Jackson <ajax@nwnk.net>
118820Date: Mon Apr 3 19:50:15 2006 +0000
118821
118822 Bug #6346: Build fix when using gcc -mno-sse. (Jonathan Adamczewski)
118823
118824commit 66500819b1ca730a7b1df400a8368a08cbe49335
118825Author: Daniel Stone <daniel@fooishbar.org>
118826Date: Mon Apr 3 11:37:30 2006 +0000
118827
118828 Bug #1358: Make ISO_Prev_Group cycle/wrap as ISO_Next_Group does.
118829
118830commit 2a6c11aa3b06f13dad94f3441c7184e6720a2bf4
118831Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
118832Date: Mon Apr 3 09:12:28 2006 +0000
118833
118834 Fix a server crash due to memsetting beyond allocated memory when running
118835 GL applications.
118836
118837commit f6ca2b3ea92b7fe98408c51a17a590435e808b1d
118838Author: Adam Jackson <ajax@nwnk.net>
118839Date: Mon Apr 3 02:15:55 2006 +0000
118840
118841 Coverity #38: Dead branch elimination.
118842
118843commit 9b9dd747d8f4697c6d5c947c160d5991c7c8fde5
118844Author: Adam Jackson <ajax@nwnk.net>
118845Date: Mon Apr 3 02:13:47 2006 +0000
118846
118847 Coverity #75: Dead variable elimination.
118848
118849commit 3f87aeefb4be3ac23ae636d3756ffdc446eaa62d
118850Author: Adam Jackson <ajax@nwnk.net>
118851Date: Mon Apr 3 02:12:11 2006 +0000
118852
118853 Coverity #82: Dead variable elimination.
118854
118855commit 61926dbe592468076f8c9a666f0098d067d2213e
118856Author: Adam Jackson <ajax@nwnk.net>
118857Date: Mon Apr 3 02:09:05 2006 +0000
118858
118859 Coverity #271: Fix an unbelievably boneheaded NULL chase.
118860
118861commit 7ef95da8a3e22e710882590fc47d56893159cb5d
118862Author: Adam Jackson <ajax@nwnk.net>
118863Date: Mon Apr 3 01:51:54 2006 +0000
118864
118865 Coverity #616: Fix a rare memory leak.
118866
118867commit 01ebd633017249c496f378df511586c973d49708
118868Author: Adam Jackson <ajax@nwnk.net>
118869Date: Mon Apr 3 01:43:33 2006 +0000
118870
118871 Coverity #833: Fix a rather nasty memory leak.
118872
118873commit a01f17d6dec02f80144e108f748783cb4e429ebb
118874Author: Adam Jackson <ajax@nwnk.net>
118875Date: Mon Apr 3 01:35:05 2006 +0000
118876
118877 Coverity #983: Move some risky debugging code inside #ifdef DEBUG.
118878
118879commit c03cfca3806f45948627715b25b46839a07be979
118880Author: Adam Jackson <ajax@nwnk.net>
118881Date: Mon Apr 3 01:31:59 2006 +0000
118882
118883 Coverity #986: Prevent a NULL chase.
118884
118885commit 07ecf49521973bbb205b199c39e1171f1163df2b
118886Author: Adam Jackson <ajax@nwnk.net>
118887Date: Mon Apr 3 01:28:11 2006 +0000
118888
118889 Coverity #992: Prevent a NULL chase.
118890
118891commit c6b3b3354c2d9139b19b132051d434e97dd19715
118892Author: Adam Jackson <ajax@nwnk.net>
118893Date: Sun Apr 2 22:51:42 2006 +0000
118894
118895 Bump to 1.1.99.1.
118896
118897commit 7e085f52b6f07c076bd3bcfdce27c17d14d7822e
118898Author: Kristian Høgsberg <krh@redhat.com>
118899Date: Sun Apr 2 22:31:13 2006 +0000
118900
118901 Use xf86LoaderCheckSymbol to check for DRI symbols instead of dlsym,
118902 avoiding RTLD_DEFAULT. (__glXDRIscreenProbe): Change GLX-DRI to AIGLX
118903 in LogMessage for consitency.
118904
118905commit b2097b99a2e6cc045ee9b6d80946bc06c4d9302c
118906Author: Adam Jackson <ajax@nwnk.net>
118907Date: Sun Apr 2 21:45:03 2006 +0000
118908
118909 ../stub
118910
118911commit 4e3a4cfdd1d7153eb88aab05ed02ddb32601ae93
118912Author: Eric Anholt <anholt@freebsd.org>
118913Date: Sun Apr 2 06:22:05 2006 +0000
118914
118915 Use RTLD_DEFAULT, rather than relying on NULL happening to map to it as it
118916 does on Linux.
118917
118918commit 323fec20292fc5ad90bfee9015ecccdc13c968ad
118919Author: Adam Jackson <ajax@nwnk.net>
118920Date: Sun Apr 2 00:46:20 2006 +0000
118921
118922 Reorder link order for Xdmx to fix new screensaver variable reference
118923 properly; remove previous awful hack.
118924
118925commit a605b9ffd3c2e7d227e35b911761f720bf07b7e6
118926Author: Adam Jackson <ajax@nwnk.net>
118927Date: Sun Apr 2 00:09:43 2006 +0000
118928
118929 Fix some includes to point into X11/fonts/ properly.
118930
118931commit e5b1d38e142807b59ce4ec89764c949f707ec541
118932Author: Adam Jackson <ajax@nwnk.net>
118933Date: Sat Apr 1 23:53:33 2006 +0000
118934
118935 Disable Xprint freetype support momentarily. Needs ttf2pt1.c, which exists
118936 in the monolith but has an advertising clause in the license.
118937
118938commit ccca76b8083b83825fa16483b44e8926a35412bb
118939Author: Eric Anholt <anholt@freebsd.org>
118940Date: Sat Apr 1 23:41:23 2006 +0000
118941
118942 Clean up warnings and a debug printf.
118943
118944commit 6afa814ab16f351b2eb787e5bf481a1f9738b391
118945Author: Eric Anholt <anholt@freebsd.org>
118946Date: Sat Apr 1 23:28:17 2006 +0000
118947
118948 Pull out fb's tile handling during fbValidateGC so we can do the necessary
118949 exaPrepare/FinishAccess()es. Revealed by xtest with fakexa.
118950
118951commit 277f612d4eeb89adb8ccda4e8fd3d211d8d1705e
118952Author: Adam Jackson <ajax@nwnk.net>
118953Date: Sat Apr 1 23:19:08 2006 +0000
118954
118955 Hack around the new screensaver variable for DMX, which is otherwise
118956 blissfully ignorant of it.
118957
118958commit 5f95146fcfcae60cc29265799ba3b851647105d6
118959Author: Eric Anholt <anholt@freebsd.org>
118960Date: Sat Apr 1 22:35:16 2006 +0000
118961
118962 Export exaPrepare/FinishGC to the rest of EXA, and use it in the ImageGlyph
118963 implementation to avoid unprepared access to the tile. Also, relocate
118964 the fbGetDrawable to avoid using a stale dest pointer after
118965 exaSolidBoxClipped() may have migrated it. Revealed by xtest.
118966
118967commit c720ffe875e4b2038746ff9b4767f8b90db0a307
118968Author: Eric Anholt <anholt@freebsd.org>
118969Date: Sat Apr 1 22:17:44 2006 +0000
118970
118971 Use fb's depth-to-planemask computation, which doesn't suffer from getting
118972 a 1 planemask at depth 32. Fixes Get/PutImage xtest tests.
118973
118974commit 5c0a2088e229d05c38e5df7daea45af0d7db7daf
118975Author: Daniel Stone <daniel@fooishbar.org>
118976Date: Sat Apr 1 21:49:44 2006 +0000
118977
118978 Bug #6428: Fix off-by-one error when walking off the end of the vmodmap
118979 list.
118980
118981commit 1e764feab595b781dab22d6e41c26f118c9d41b5
118982Author: Daniel Stone <daniel@fooishbar.org>
118983Date: Sat Apr 1 21:20:31 2006 +0000
118984
118985 Bug #5801: Check for MTRR support under Linux. Minor refactoring of MTRR
118986 checks for other OSes.
118987
118988commit 978c7b14a18caffde5600480824d04492fc32aef
118989Author: Daniel Stone <daniel@fooishbar.org>
118990Date: Sat Apr 1 21:02:40 2006 +0000
118991
118992 Make Xprint AC_ARG_ENABLEs and AC_ARG_WITHs unconditional also.
118993
118994commit 71a6f2ef6c1138c5c6918a54dfb856183f4f242c
118995Author: Daniel Stone <daniel@fooishbar.org>
118996Date: Sat Apr 1 20:58:42 2006 +0000
118997
118998 Unconditionally run XP_USE_FREETYPE AM_CONDITIONAL, not only in the Xprint
118999 path.
119000
119001commit d1e90113fc32b6ddc4dbe1a074763c31bc133e75
119002Author: Eric Anholt <anholt@freebsd.org>
119003Date: Fri Mar 31 23:22:29 2006 +0000
119004
119005 Don't attempt to Prepare/FinishAccess NULL pDrawables. Exposed by new
119006 gradient testing in rendercheck.
119007
119008commit 2e38fedd29e7e55d01e3edce6a73b8ceaac17911
119009Author: Eric Anholt <anholt@freebsd.org>
119010Date: Fri Mar 31 19:41:28 2006 +0000
119011
119012 Add an option to EXA for the DDX to request that EXA hide the pixmap's
119013 devPrivate.ptr when pointing at offscreen memory, outside of
119014 exaPrepare/FinishAccess(). This was used with fakexa to find (by NULL
119015 dereference) many instances of un-Prepared CPU access to the
119016 framebuffer:
119017 - GC tiles used in several ops when fillStyle == FillTiled were never
119018 Prepared.
119019 - Migration could lead to un-Prepared access to mask data in render's
119020 Trapezoids and Triangles
119021 - PutImage's UploadToScreen failure fallback failed to Prepare.
119022
119023commit f480dc797b51f080f912efc7867d6d8e50be074c
119024Author: Eric Anholt <anholt@freebsd.org>
119025Date: Fri Mar 31 19:25:42 2006 +0000
119026
119027 Revert mistaken commit to exa_unaccel.c. Should have been to
119028 exa_offscreen.c: Correct a typo in debug-only offscreen validation
119029 code. (Wang Zhenyu)
119030
119031commit 1a8167c1baa767fc056d1e17d96d0ea98a5f3b17
119032Author: Eric Anholt <anholt@freebsd.org>
119033Date: Fri Mar 31 19:16:51 2006 +0000
119034
119035 Correct a typo in debug-only offscreen validation code. (Wang Zhenyu)
119036
119037commit 7ea30b507f4ce5ce20fbfaca80f7d5b53a99eb1d
119038Author: Fredrik Höglund <fredrik@kde.org>
119039Date: Fri Mar 31 18:49:38 2006 +0000
119040
119041 Move the screensaver extension from module to builtins.
119042 Add the server side implementation of the ScreenSaverSuspend request.
119043 Require scrnsaverproto >= 1.1, and change the linking order of the Xorg
119044 static libs.
119045
119046commit acca49b1a5a6c034f3b9d51d9016b8a7d43da809
119047Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
119048Date: Fri Mar 31 17:39:35 2006 +0000
119049
119050 Fri Mar 31 12:37:16 2006 Søren Sandmann <sandmann@redhat.com>
119051 Fix copyright statement
119052
119053commit b074ce22470ba0a51eda2af7100d09a260a1e8bb
119054Author: Egbert Eich <eich@suse.de>
119055Date: Fri Mar 31 15:11:51 2006 +0000
119056
119057 fixed typo.
119058
119059commit 710bb2e6c8b2874406e48fa8ad24539290c98d41
119060Author: Daniel Stone <daniel@fooishbar.org>
119061Date: Fri Mar 31 14:52:57 2006 +0000
119062
119063 Reindent with -cbi0.
119064
119065commit 7c44bb8c49656133eae675377edea55322d254ca
119066Author: Daniel Stone <daniel@fooishbar.org>
119067Date: Fri Mar 31 07:33:34 2006 +0000
119068
119069 Simplify XkbWriteXKBKeymapForNames a bit, and remove debug spew.
119070
119071commit 4c317bbc1259fa555dc5d5278226b21c42845c0c
119072Author: Daniel Stone <daniel@fooishbar.org>
119073Date: Fri Mar 31 07:21:41 2006 +0000
119074
119075 Add full FreeType support for Xprint. (Drew Parsons)
119076
119077commit 759033703ce17b20d57756206f48a7ae410a50d1
119078Author: Eric Anholt <anholt@freebsd.org>
119079Date: Thu Mar 30 21:44:36 2006 +0000
119080
119081 Remove the exaAsyncPixmapGCOps mostly-unaccelerated ops vector, and always
119082 plug in the accelerated one, even if the destination pixmap is
119083 currently offscreen. This was a leftover from when kaa originally got
119084 accelerated offscreen pixmap support, and its only concievable use was
119085 to avoid a little overhead on ops to in-system pixmaps that weren't
119086 going to get migrated. At this point, we probably care more about just
119087 getting everything accelerated that we easily can, which should happen
119088 with the new migration support.
119089
119090commit b9203dc068ccd4c0d22d49a94b910783432b96a8
119091Author: Eric Anholt <anholt@freebsd.org>
119092Date: Thu Mar 30 21:25:43 2006 +0000
119093
119094 Don't do an extra fallback path for CopyWindow while swappedOut, since
119095 exaCopyNtoN takes care of the fallback anyway, and we don't care about
119096 the performance of this path.
119097
119098commit 5c04610f8aeceed9ec7cd0ca8c5eb314cacc3c25
119099Author: Eric Anholt <anholt@freebsd.org>
119100Date: Thu Mar 30 21:21:59 2006 +0000
119101
119102 Add a dependency on EXA, so it rebuilds when the library does. The manual
119103 indicated I shouldn't do this, but experience indicates I should.
119104
119105commit 8ec42a10ff04e51e8d0b4cffb15064d901bc398d
119106Author: Kristian Høgsberg <krh@redhat.com>
119107Date: Thu Mar 30 20:08:44 2006 +0000
119108
119109 Mark the ARGB FBConfig as nonconforming to prevent drivers and apps from
119110 falling over.
119111 Add @GLX_DEFINES@ so GLcore gets compiled with TLS support if configured.
119112 Only destroy the mesa buffer if it got initialized.
119113
119114commit 08e319091fae7a60ae9fa757659cfde2966af9e9
119115Author: Egbert Eich <eich@suse.de>
119116Date: Thu Mar 30 18:53:41 2006 +0000
119117
119118 Added notice to last ChangeLog entry
119119 Fixes for some vsw4 failures on 64bit BE platforms such as PPC64 and s390x.
119120 Provided by Hong Bo Peng of IBM (slightly modified). Patches try to
119121 resolve some of the careless mixtures of ulong and uint (which are
119122 different size on
119123 64bit). > This patch will break the driver ABI! < Bugzilla #6438.
119124
119125commit 9da1d2257d02155cc8b4541cf5fcb4e64d756945
119126Author: Egbert Eich <eich@suse.de>
119127Date: Thu Mar 30 18:48:11 2006 +0000
119128
119129 Fixes for some vsw4 failures on 64bit BE platforms such as PPC64 and s390x.
119130 Provided by Hong Bo Peng of IBM (slightly modified). Patches try to
119131 resolve some of the careless mixtures of ulong and uint (which are
119132 different size on
119133 64bit). Bugzilla #6438.
119134
119135commit 6d7ad353bafe914f0b50887daaeaae89ada6ebd3
119136Author: Kristian Høgsberg <krh@redhat.com>
119137Date: Thu Mar 30 18:29:53 2006 +0000
119138
119139 Regenerate these files using updated scripts to avoid unused variable
119140 warnings.
119141
119142commit 2153fa97482bae5737def3ecd4fe1cdc03834991
119143Author: Eric Anholt <anholt@freebsd.org>
119144Date: Thu Mar 30 05:24:27 2006 +0000
119145
119146 Bug #2986: Add PutImage acceleration for the ZPixmap, planeMask ~=
119147 FB_ALLONES, bitsPerPixel >= 8, GXcopy cases. With the radeon driver on
119148 my machine, this gives about 10% speedup in PutImage
119149 10x10 and 500x500, and 40% speedup for 10x10 ShmPutImage, up to 65%
119150 improvement in 500x500 ShmPutImage. Also fixes a crasher in GetImage
119151 that slipped in at the last minute.
119152
119153commit 3cf46cc1e32efc0e4be1d88be111ba0438e0f021
119154Author: Eric Anholt <anholt@freebsd.org>
119155Date: Thu Mar 30 05:15:58 2006 +0000
119156
119157 Add an UploadToScreen implementation, for testing PutImage support, and
119158 make the DownloadFromScreen more robust.
119159
119160commit e799dd68e2bd0fa8ac3c344111fb12e1f32d4c10
119161Author: Eric Anholt <anholt@freebsd.org>
119162Date: Wed Mar 29 22:25:17 2006 +0000
119163
119164 Bug #2986: Add acceleration of GetImage using DownloadFromScreen for the
119165 ZPixmap, planeMask ~= FB_ALLONES, bitsPerPixel >= 8 case. I'm pretty
119166 convinced that this is the only case that we care about at all. Tested
119167 with xwd -root and xwd on a gnome-terminal, in a composited environment
119168 or not.
119169
119170commit 4bb5ab0b4453208573b91b334940f190a8f7210a
119171Author: Eric Anholt <anholt@freebsd.org>
119172Date: Wed Mar 29 22:03:18 2006 +0000
119173
119174 Add a DownloadFromScreen implementation, used for testing GetImage
119175 acceleration, and set the migration scheme to Always on init (since
119176 this is all for testing, and Always should make migration happen more
119177 frequently than Greedy).
119178
119179commit e31e8ace1043eab340d6b60a6e98b23ebf102786
119180Author: Deron Johnson <deron.johnson@sun.com>
119181Date: Wed Mar 29 17:51:54 2006 +0000
119182
119183 Fix composite overlay window bug 6411
119184
119185commit ff6f88348c7498e83b0b143ef3737fd6eb0995e4
119186Author: Adam Jackson <ajax@nwnk.net>
119187Date: Wed Mar 29 01:05:09 2006 +0000
119188
119189 More warning cleanup.
119190
119191commit 52d9ce7f4fc599d30dec2e61fc1720597043d91c
119192Author: Kristian Høgsberg <krh@redhat.com>
119193Date: Tue Mar 28 21:45:14 2006 +0000
119194
119195 Fix another typo.
119196
119197commit 7df64898eac46a487e8eab2af7213d133b9ca419
119198Author: Kristian Høgsberg <krh@redhat.com>
119199Date: Tue Mar 28 07:46:04 2006 +0000
119200
119201 Fix a couple of typos.
119202
119203commit bd283c2464e2c0e1fd0aca1dedff0f39c2564c34
119204Author: Aaron Plattner <aplattner@nvidia.com>
119205Date: Tue Mar 28 07:21:50 2006 +0000
119206
119207 Add a new export, LoaderGetABIVersion. This function allows modules to
119208 query the versions directly instead of having to guess. Bug #6416: Add
119209 LoaderGetABIVersion.
119210
119211commit a06342eccc76035ff859fee4d283b288c90ee923
119212Author: Kristian Høgsberg <krh@redhat.com>
119213Date: Tue Mar 28 02:57:07 2006 +0000
119214
119215 Add --enable-glx-tls ./configure option to enable use of TLS for storing
119216 current GL context. Use this option to let AIGLX load DRI drivers
119217 compiled for TLS.
119218
119219commit 77531dfb9f9f3ca0e38ad0555ee3735d6f28cf19
119220Author: Adam Jackson <ajax@nwnk.net>
119221Date: Tue Mar 28 01:22:01 2006 +0000
119222
119223 Silence some editorializing in the configure help text.
119224
119225commit 7deaaa797cf8e7ca71e9b34fa6f413d1ed2b3dab
119226Author: Adam Jackson <ajax@nwnk.net>
119227Date: Tue Mar 28 01:21:00 2006 +0000
119228
119229 Big old pile of warning fixes.
119230
119231commit 7342dbe4b2108827eaf30993ceeecbd828da2290
119232Author: Adam Jackson <ajax@nwnk.net>
119233Date: Tue Mar 28 00:18:31 2006 +0000
119234
119235 Remove long-dead screen region code.
119236
119237commit 0e88cefbfecbff0c7dd606ce0caca840f45cbc0d
119238Author: Daniel Stone <daniel@fooishbar.org>
119239Date: Mon Mar 27 23:03:47 2006 +0000
119240
119241 Prune XKB code to only what we need to run the server. Remove dead
119242 !XKB_IN_SERVER codepaths. Remove HAVE_CONFIG_H codepaths.
119243
119244commit 5be8a66d324f3d5840b134ad29069eace64e6f12
119245Author: Daniel Stone <daniel@fooishbar.org>
119246Date: Mon Mar 27 22:28:32 2006 +0000
119247
119248 Fix remnants of previous busted _XkbStrCaseCmp commit.
119249
119250commit 9e202dfe40e2bdd66f461a6ba531e927f82096ae
119251Author: Daniel Stone <daniel@fooishbar.org>
119252Date: Mon Mar 27 22:25:56 2006 +0000
119253
119254 Remove remnants of XkbCF code.
119255
119256commit 7257590651328f89d23e80da1ec6241542a660cd
119257Author: Daniel Stone <daniel@fooishbar.org>
119258Date: Mon Mar 27 21:15:06 2006 +0000
119259
119260 Move XFree86 DDX XKB actions into dixmods.
119261
119262commit d7b9e2b0e9d6889ea6b05e63892e612f4e5f19f5
119263Author: Daniel Stone <daniel@fooishbar.org>
119264Date: Sat Mar 25 23:09:50 2006 +0000
119265
119266 Bug #3819: Remove open-coding of strcasecmp.
119267
119268commit b3570dd94aa72f94e537a17680150e91e7712f5a
119269Author: Daniel Stone <daniel@fooishbar.org>
119270Date: Sat Mar 25 22:37:58 2006 +0000
119271
119272 Remove INITARGS braindamage, change to void; add XkbExtensionInit prototype
119273 to xkb.h. Explicitly initialise nTypes in xkb.c.
119274
119275commit 1ef60ce8ebb681b3cfb5e515be5c187c0442dcda
119276Author: Daniel Stone <daniel@fooishbar.org>
119277Date: Sat Mar 25 22:35:48 2006 +0000
119278
119279 Really remove all DDX pre-config code.
119280
119281commit ec10f70b2114e5369a5b2f34b084dcf55634dcb4
119282Author: Daniel Stone <daniel@fooishbar.org>
119283Date: Sat Mar 25 21:52:49 2006 +0000
119284
119285 Remove XkbCF DDX configuration code.
119286
119287commit aae4238360b842ac34dc8ee16e165a1821f9a801
119288Author: Daniel Stone <daniel@fooishbar.org>
119289Date: Sat Mar 25 20:17:58 2006 +0000
119290
119291 Fix two glaring unconditional-NULL-dereferences.
119292
119293commit a68c11bb1d7c5419004a1714e49dffac57304e78
119294Author: Adam Jackson <ajax@nwnk.net>
119295Date: Sat Mar 25 19:52:05 2006 +0000
119296
119297 Mark everything in xf86sym.c as _X_EXPORT.
119298
119299commit ae935832facfa81a9689882406ecca74b0346790
119300Author: Fredrik Höglund <fredrik@kde.org>
119301Date: Fri Mar 24 20:50:13 2006 +0000
119302
119303 Refactored the screensaver and DPMS timer code to use the screensaver timer
119304 for both screensaver and DPMS. Removed the SetDPMSTimers() and
119305 FreeDPMSTimers() functions.
119306
119307commit d1746ec0f0c8a0b750f390e7a7faf21b67683f4a
119308Author: Kristian Høgsberg <krh@redhat.com>
119309Date: Fri Mar 24 17:58:39 2006 +0000
119310
119311 Make sure DRI module is loaded before calling DRI functions.
119312
119313commit f1616508c95d12dfaad2cfd61b40228b3dba6f60
119314Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
119315Date: Thu Mar 23 23:54:08 2006 +0000
119316
119317 Add ast driver/pci id (Carl Switzky, Sun Microsystems)
119318
119319commit 6d2896b384e17512e8f12036daabcd575d21f804
119320Author: Kristian Høgsberg <krh@redhat.com>
119321Date: Wed Mar 22 22:49:52 2006 +0000
119322
119323 Improve error logging.
119324
119325commit 5449634e3c9428005aba5b3322ced7e86c62f185
119326Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
119327Date: Wed Mar 22 21:37:49 2006 +0000
119328
119329 Wed Mar 22 16:28:46 2006 Søren Sandmann <sandmann@redhat.com>
119330 Use inline assembly for copy area, since gcc doesn't generate movq
119331 instructions.
119332
119333commit 5b3084c64f7bd1232603ffb3e985600b8d045453
119334Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
119335Date: Wed Mar 22 21:13:08 2006 +0000
119336
119337 Wed Mar 22 16:05:09 2006 Søren Sandmann <sandmann@redhat.com>
119338 Use inline assembly for solid fills, since gcc doesn't use the movq
119339 instructions.
119340
119341commit a08e5e0c68baaf85b0fc3ecde74a6bcf80bcd4bf
119342Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
119343Date: Wed Mar 22 18:44:26 2006 +0000
119344
119345 Wed Mar 22 13:42:44 2006 Søren Sandmann <sandmann@redhat.com>
119346 Patch by Keith Packard to make sure redirected windows don't get considered
119347 "FullyObscured".
119348
119349commit 966d93ef6d1f2ed02f3b81b5bf5a1ebbdd48c93d
119350Author: Kristian Høgsberg <krh@redhat.com>
119351Date: Tue Mar 21 22:54:38 2006 +0000
119352
119353 Make the server distcheck and tag 1.0.99.1 snapshot.
119354 Bump CVS version to 1.0.99.1.
119355 Distcheck fixes.
119356
119357commit 8e3ad87d01c102591c7dc25614f6ac10e444a1b1
119358Author: Kristian Høgsberg <krh@redhat.com>
119359Date: Tue Mar 21 22:32:13 2006 +0000
119360
119361 #include indirect_dispatch to get prototypes for FBO functions.
119362 Fix a couple of warnings.
119363
119364commit dcc43d57cbe9d2b65384fe9ba2e4e4fbb43cb0a1
119365Author: Donnie Berkholz <spyderous@gentoo.org>
119366Date: Mon Mar 20 20:10:29 2006 +0000
119367
119368 Finish glx_ansic.h wrapper changes to make Xvfb and Xnest link again.
119369
119370commit 9509c6799e31e96677b6d07bdf24ea91ddd30020
119371Author: Adam Jackson <ajax@nwnk.net>
119372Date: Mon Mar 20 19:32:18 2006 +0000
119373
119374 dead file removal
119375
119376commit 61a020265c5915e3d671d5b2047b81a5d15594c3
119377Author: Adam Jackson <ajax@nwnk.net>
119378Date: Mon Mar 20 18:43:18 2006 +0000
119379
119380 Bug #5549: Fix build for sparc64. (Matthieu Herrb)
119381
119382commit 6eb4e2303aaab8d64e3f6cbc0bbee55689bdcb82
119383Author: Adam Jackson <ajax@nwnk.net>
119384Date: Mon Mar 20 14:01:05 2006 +0000
119385
119386 Bug #6213: Check geteuid's return value, not its address, otherwise
119387 unprivileged users can set the modulepath and run arbitrary code. Patch
119388 from Matthieu Herrb. (CVE-2006-0745, Coverity #4)
119389
119390commit 8c1bb37d0649b269b78c457b8b41ff59a41d89af
119391Author: Daniel Stone <daniel@fooishbar.org>
119392Date: Fri Mar 17 08:55:07 2006 +0000
119393
119394 Typo fix, reindent.
119395
119396commit 2d2d38d17cc2558f8a41166a4a1578bc4c663c37
119397Author: Kristian Høgsberg <krh@redhat.com>
119398Date: Fri Mar 17 01:47:25 2006 +0000
119399
119400 Check for glproto when building GLX and make sure we have at least 1.4.6.
119401 Drop glx_ansic.h wrapper and call xalloc, xrealloc, xfree and str-funcs
119402 directly.
119403
119404commit 2c11cde3367fcd22740b577a4364b1e41cf3e1d2
119405Author: Kristian Høgsberg <krh@redhat.com>
119406Date: Fri Mar 17 00:35:18 2006 +0000
119407
119408 More patches from David Reveman:
119409 Add GL_ARB_texture_non_power_of_two, GL_EXT_framebuffer_object and
119410 GL_NV_texture_env_combine4 extensions.
119411 Add __GLXcontext destructor and flush context cache there and on
119412 loseCurrent.
119413 Chain back to new __GLXcontext destructor. (__glXMesaContextForceCurrent):
119414 Set render table on forceCurrent. (init_screen_visuals): Index pVis
119415 array correctly. (GlxGetMesaProvider): Add this.
119416 Hook up FBO marshalling.
119417
119418commit 14aafc258cd774cf937f9798a888c2d3c97ccacf
119419Author: Eric Anholt <anholt@freebsd.org>
119420Date: Thu Mar 16 18:43:55 2006 +0000
119421
119422 Change EXA so that exaMoveOutPixmap() retains the framebuffer copy of the
119423 pixmap, and damage is tracked so that a later exaMoveInPixmap won't
119424 result in an upload if no upload is necessary. This will likely improve
119425 the performance of the "Always" migration scheme significantly, and is
119426 a step in the path to more exact damage tracking between framebuffer
119427 and system memory.
119428
119429commit d0d336efd58896718f31a400651bacd9b769fb5a
119430Author: Daniel Stone <daniel@fooishbar.org>
119431Date: Thu Mar 16 16:29:17 2006 +0000
119432
119433 Add support for ZX2 PCI-E local bus adaptors. (Alex Williamson, HP)
119434 Use soft timeout register to avoid MCAs when probing for non-existent local
119435 bus adaptors on ZX2. (Alex Williamson, HP)
119436
119437commit 175980580e572745a9a381b4432e3ba0457d3ba3
119438Author: Adam Jackson <ajax@nwnk.net>
119439Date: Wed Mar 15 23:05:53 2006 +0000
119440
119441 Bump to requiring fixesproto >= 4.0 and compositeproto >= 0.3.
119442
119443commit 6fe377af5a82deb6f8b0f3b75414335e7845caac
119444Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119445Date: Wed Mar 15 21:25:38 2006 +0000
119446
119447 - OpenBSD needs -Wl,-export-dynamic to export symbols from main executable
119448 to modules.
119449 - Probe for OpenBSD aperture driver and define HAS_APERTURE_DRV
119450 accordingly.
119451
119452commit 21f7f2fb113ee4f9cd011c3cc2d45d43bbdd35fa
119453Author: Felix Kuehling <fxkuehl@gmx.de>
119454Date: Wed Mar 15 18:43:32 2006 +0000
119455
119456 Enable correct handling of the BTS instruction (opcode 0f ab) The code was
119457 there but #ifdefed out. Insead of BTS, BT was executed. This patch
119458 enables the BTS function and hooks it up the the correct opcode. (ATI
119459 Technologies Inc.)
119460
119461commit b726aa502a871c700bc42b5325abf2c6820ff756
119462Author: Felix Kuehling <fxkuehl@gmx.de>
119463Date: Wed Mar 15 18:37:44 2006 +0000
119464
119465 Update to build against Mesa CVS HEAD.
119466
119467commit c74464d92cd673ff0669375757caab798cc57e95
119468Author: Eric Anholt <anholt@freebsd.org>
119469Date: Wed Mar 15 16:59:45 2006 +0000
119470
119471 Don't let pinned pixmaps get migrated in when using the "Always" migration
119472 scheme. This notably keeps the visible screen from getting migrated in
119473 to a new location in framebuffer.
119474 Reported by: Michel Dänzer.
119475
119476commit b9c43cde1e368903786977b06368d5e36db9ffe8
119477Author: Adam Jackson <ajax@nwnk.net>
119478Date: Wed Mar 15 16:56:10 2006 +0000
119479
119480 Coverity #1042, 1043: Nuke some dead variables.
119481
119482commit 5e106a71b9f8077216d41619402952b0005dd8a4
119483Author: Adam Jackson <ajax@nwnk.net>
119484Date: Wed Mar 15 16:49:04 2006 +0000
119485
119486 Coverity #807: Fix a memory leak in XFixesExpandRegion.
119487
119488commit a3ef63696cac950b2520e7c85564befc0a830fde
119489Author: Adam Jackson <ajax@nwnk.net>
119490Date: Wed Mar 15 16:36:31 2006 +0000
119491
119492 Coverity #490: Fix a range check in xf86vidmode extension.
119493
119494commit 152090ce442e94de1ae920208a92931af6493c8c
119495Author: Adam Jackson <ajax@nwnk.net>
119496Date: Wed Mar 15 16:33:12 2006 +0000
119497
119498 Coverity #487: Check version number correctly.
119499
119500commit 72cc6307257fcbb800267464487bf918ee674328
119501Author: Adam Jackson <ajax@nwnk.net>
119502Date: Wed Mar 15 16:32:05 2006 +0000
119503
119504 Coverity #491: Check version number correctly.
119505
119506commit 460f2ea4a594a53536f34c4ad27795fceec50bcc
119507Author: Adam Jackson <ajax@nwnk.net>
119508Date: Wed Mar 15 16:21:04 2006 +0000
119509
119510 Coverity #794: Fix a highly unlikely memory leak.
119511
119512commit 116d158e85ec43577ff69aeb3271ab1f888500c9
119513Author: Adam Jackson <ajax@nwnk.net>
119514Date: Wed Mar 15 16:16:24 2006 +0000
119515
119516 Coverity #269: Compare the requested ABI class against the ABI class of the
119517 module, not the module class.
119518
119519commit d8221a9b70a11606a0f7e1f69afee6049d7f182f
119520Author: Adam Jackson <ajax@nwnk.net>
119521Date: Wed Mar 15 16:11:34 2006 +0000
119522
119523 Coverity #484: Fix an off-by-one in module refcounting.
119524
119525commit 6bb2dc02a7cffd6ed7dd28e88d584920a4150749
119526Author: Adam Jackson <ajax@nwnk.net>
119527Date: Wed Mar 15 16:01:47 2006 +0000
119528
119529 Coverity #337: Remove useless NULL check.
119530
119531commit 1e5c0842af99027cc6c30a16f967d8b60c9a894d
119532Author: Adam Jackson <ajax@nwnk.net>
119533Date: Wed Mar 15 15:34:57 2006 +0000
119534
119535 Coverity #1053: Nuke a dead variable.
119536
119537commit 7314d16cde4c3f99d9d9f1d539f0c5ff4942e653
119538Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
119539Date: Wed Mar 15 03:18:42 2006 +0000
119540
119541 Fix DRIExtensionInit() to not register callbacks when it hasn't been
119542 initialized for the current server generation. Fixes a problem where it
119543 would use stale private index and blow up in colorful ways if no driver
119544 called DRIScreenInit() on the second generation (which happens due to a
119545 bug in radeon that i'll fix separately). Note: clearing the index in
119546 DRIReset() wouldn't work as DRIReset() is called before the
119547 CloseScreen() chain
119548
119549commit 02d80a0de93f7592e69065b0fbe5820dcdebdb44
119550Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
119551Date: Wed Mar 15 03:12:32 2006 +0000
119552
119553 Make xf86 linear allocator smarter when dealing with alignment constraints
119554 when falling back to X/Y allocations. Fixes various problems of Xv
119555 allocation failures, notably with "nv" driver.
119556
119557commit c1601717d536419693b3ef6e8a3d69b9f2fdc2b3
119558Author: Eric Anholt <anholt@freebsd.org>
119559Date: Wed Mar 15 01:20:08 2006 +0000
119560
119561 Add a new migration scheme, "always", which will move pixmaps to their
119562 desired location always (unless they don't fit in FB, in which case
119563 they all get moved out for software rendering). The default remains as
119564 before, but can be controlled by the MigrationHeuristic xorg.conf
119565 option (which is intentionally not documented, as it may be
119566 short-lived). This is part of the exa-damagetrack work, which appears
119567 stable in testing with fakexa, unlike the work as a whole.
119568
119569commit a90cff266cc81993ed804fb320c1dbfe5e0d4787
119570Author: Eric Anholt <anholt@freebsd.org>
119571Date: Wed Mar 15 00:13:52 2006 +0000
119572
119573 Add more doxygen documentation, including notes on WaitMarker() and
119574 MarkSync() that I noticed were needed while reading the VIA driver.
119575
119576commit 693e42114f1127528448126d78a5209dd1198d8d
119577Author: Eric Anholt <anholt@freebsd.org>
119578Date: Tue Mar 14 21:30:12 2006 +0000
119579
119580 Move migration logic to a new function, exaDoMigration(). This is largely a
119581 manual conversion to allow for different migration schemes to be
119582 implemented reasonably, but does include some minor improvements such
119583 as accounting for pinned pixmaps not being acceleratable, and for our
119584 current GetImage and GetSpans not being accelerated.
119585
119586commit d30905478078036383977ae9d4a3685c2e2c642f
119587Author: Eric Anholt <anholt@freebsd.org>
119588Date: Tue Mar 14 20:38:06 2006 +0000
119589
119590 Pull code for getting the (0,0) pixel from a pixmap out to a separate
119591 function, since it gets repeated (with bad error handling, in one
119592 case).
119593
119594commit 01aa209f2056ef04e3f2735756a0f8b4a67a3d87
119595Author: Kristian Høgsberg <krh@redhat.com>
119596Date: Tue Mar 14 19:32:27 2006 +0000
119597
119598 Bail out early if screen doesn't support DRI.
119599
119600commit 0cc34266d6e84bb491fcf9aa74e34615b2fca4fc
119601Author: Deron Johnson <deron.johnson@sun.com>
119602Date: Mon Mar 13 22:43:42 2006 +0000
119603
119604 Updated ChangeLog for my latest composite and xfixes changes.
119605
119606commit 450018f48b2796345a4eaccbb94c1971ebd30114
119607Author: Deron Johnson <deron.johnson@sun.com>
119608Date: Mon Mar 13 21:59:55 2006 +0000
119609
119610 Part 3 of 3 (Other parts are in proto and lib) Composite Version 0.3:
119611 CompositeGetOverlayWindow, CompositeReleaseOverlayWindow Xfixes Version
119612 4.0: XFixesHideCursor, XFixesShowCursor
119613
119614commit e5956f49b217b0ee9c9f35b6a58f339a8d22b1d7
119615Author: Kristian Høgsberg <krh@redhat.com>
119616Date: Mon Mar 13 01:54:59 2006 +0000
119617
119618 First batch of AIGLX fixes from David Reveman.
119619 Add getter for Mesa provider.
119620 Export this for Xgl.
119621 Move resource tracking out of drawable constructor to allow wrapping.
119622 Use corrent reply size #define.
119623 Add this function. (DoGetDrawableAttributes): Fix array length.
119624
119625commit eb63e50d95da4e1e08fc6fcec46ac63d5e3b7bf4
119626Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119627Date: Sun Mar 12 17:14:03 2006 +0000
119628
119629 Fix build when AIGLX is false.
119630
119631commit 9ed3463450469c3108e0be7e4baabc0a403a78b2
119632Author: Eric Anholt <anholt@freebsd.org>
119633Date: Sun Mar 12 03:04:52 2006 +0000
119634
119635 Improve doxygen formatting, and attempt to clarify the 1:1 ratio of
119636 successful PrepareCopy()s to DoneCopy()s.
119637
119638commit 9a7fba5fd07c8831d0acab8d901605de537ae273
119639Author: Eric Anholt <anholt@freebsd.org>
119640Date: Sun Mar 12 03:02:26 2006 +0000
119641
119642 Make exaCopyNtoNTwoDir() call DoneCopy() at the end of each string of
119643 consecutive Copy() calls (rather than exactly once at the end of the
119644 function).
119645 Reviewed by: jbarnes
119646
119647commit c3342c8000f6d2bfb61e2cf95e028d11b59698fa
119648Author: Kristian Høgsberg <krh@redhat.com>
119649Date: Sun Mar 12 00:11:34 2006 +0000
119650
119651 Merge accel_indirect branch to HEAD.
119652
119653commit b1b731c28630965d9e2defe62d1108270dc8264c
119654Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
119655Date: Sat Mar 11 02:43:51 2006 +0000
119656
119657 Fix buffer size checks to prevent 2-byte buffer overflows. (Coverity #480,
119658 #481, #482, #483)
119659
119660commit fc0772de36315f19f5b57220db69f48a3b1fdc9a
119661Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
119662Date: Sat Mar 11 02:10:14 2006 +0000
119663
119664 Add HAS_MMAP for Xvfb
119665 Fix Xvfb option parsing to exit on bad arguments, not just issue error
119666 messages and continue on. (Coverity #492)
119667
119668commit f2ecbb30187000547a98ca7cbaee433ea4ba8fe3
119669Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
119670Date: Sat Mar 11 01:58:32 2006 +0000
119671
119672 Pass sizeof the correct buffer to XmuSnprintf. (Coverity #489)
119673
119674commit d6955798489813ef77cca13cf5f5c67d49e6dece
119675Author: Eric Anholt <anholt@freebsd.org>
119676Date: Fri Mar 10 21:36:24 2006 +0000
119677
119678 If fakexa is enabled, create a larger buffer in the Ximage, but keep the
119679 same width/height for front-buffer drawing. The fakexa code then uses
119680 this extra space for offscreen pixmaps. Note that this tones down the
119681 absurdity of fakexa's offscreen pixmap alignment requirements (odd
119682 alignment is too weird, so stick with "24", which is still strange but
119683 exists out there). It also fixes a couple of bugs in the fakexa
119684 implementation revealed by using offscreen pixmaps.
119685
119686commit 5b1a7b478f072f56e836f2d4c0fbc1985842e2bb
119687Author: Eric Anholt <anholt@freebsd.org>
119688Date: Fri Mar 10 21:32:34 2006 +0000
119689
119690 Move the exaDrawableDirty in exaPrepareAccess to exaFinishAccess, which is
119691 after the drawing is done. Previously, a failed PrepareAccess could
119692 have migrated and cleared the dirty flag before the damage was ever
119693 done.
119694
119695commit ffdbb547becc71f1cfdd035d0d6c71539f185fb1
119696Author: Eric Anholt <anholt@freebsd.org>
119697Date: Fri Mar 10 08:06:42 2006 +0000
119698
119699 Coverity #1011: Remove a useless NULL check on a pVbe that had been
119700 dereferenced many times before.
119701
119702commit 1bc72dce5f8bc40e369e69b684816fdaaa07da43
119703Author: Eric Anholt <anholt@freebsd.org>
119704Date: Fri Mar 10 08:03:24 2006 +0000
119705
119706 Coverity #857: Fix resource leak in error path by freeing earlier.
119707
119708commit 55f677d600370b19d62ef821025481f2be6f5edb
119709Author: Eric Anholt <anholt@freebsd.org>
119710Date: Fri Mar 10 07:58:27 2006 +0000
119711
119712 Coverity #813, #814, #815, #816: Fix resource leaks in error paths of
119713 config parsing code.
119714
119715commit 2bd41105496b729395fbcf97f09581eb0efb3510
119716Author: Eric Anholt <anholt@freebsd.org>
119717Date: Fri Mar 10 07:45:25 2006 +0000
119718
119719 Document the restriction on PrepareAccess() failure, from discussion with
119720 benh.
119721
119722commit 21dcd0304879f38ea8ea01ba88e7cc7783771adf
119723Author: Jeremy C. Reed <reed@reedmedia.net>
119724Date: Fri Mar 10 01:34:45 2006 +0000
119725
119726 Just like FreeBSD, let DragonFly's default mouse Device be /dev/sysmouse
119727 (since /dev/mouse don't even exist by default).
119728
119729commit 9a99afdfb292f303f914039952fdd772eed9e03a
119730Author: Jeremy C. Reed <reed@reedmedia.net>
119731Date: Fri Mar 10 01:22:26 2006 +0000
119732
119733 Add DragonFly support. (It is like FreeBSD.)
119734 This patch is from DragonFly developer Joerg Sonnenberger and the pkgsrc
119735 collection.
119736 I tested using /dev/sysmouse with moused using my serial /dev/cuaa0.
119737
119738commit 7a0f7f739804bc7d9c5562701abee8d134878977
119739Author: Eric Anholt <anholt@freebsd.org>
119740Date: Thu Mar 9 23:29:44 2006 +0000
119741
119742 Coverity #349: Fall back to software early if pSrc->pDrawable is NULL, or
119743 pMask is non-NULL but pMask->pDrawable is NULL. This prevents NULL
119744 dereferences on gradients and other Pictures which have no pDrawable.
119745
119746commit 8a3ff42abb726d1604af39b4653ede5f760b7e69
119747Author: Eric Anholt <anholt@freebsd.org>
119748Date: Thu Mar 9 23:25:35 2006 +0000
119749
119750 Commit changes missed in last commit (mis-typed path and didn't notice): Do
119751 a first pass of doxygen documentation of EXA. This removes the
119752 corresponding pieces of exa-driver.txt, which were becoming stale.
119753 Hopefully the documentation will stay much more up-to-date this way.
119754 Many thanks to jbarnes for writing exa-driver.txt which was used a lot
119755 in writing this documentation.
119756
119757commit ab35c3fbc135bafdfc5057ef5d6227ca3534ed26
119758Author: Eric Anholt <anholt@freebsd.org>
119759Date: Thu Mar 9 23:18:15 2006 +0000
119760
119761 Do a first pass of doxygen documentation of EXA. This removes the
119762 corresponding pieces of exa-driver.txt, which were becoming stale.
119763 Hopefully the documentation will stay much more up-to-date this way.
119764 Many thanks to jbarnes for writing exa-driver.txt which was used a lot
119765 in writing this documentation.
119766
119767commit d8f8bfeccef0750d79f852b9ae7152e841227d5a
119768Author: Matthias Hopf <mhopf@suse.de>
119769Date: Thu Mar 9 14:23:57 2006 +0000
119770
119771 Do Xorg configure checks for Xgl only as well
119772
119773commit 2822cbc1fb2271844e7ae10c3629aaa940ae4042
119774Author: Eric Anholt <anholt@freebsd.org>
119775Date: Thu Mar 9 06:04:07 2006 +0000
119776
119777 Rearrange EXA driver structures so that there's a hope of maintaining ABI
119778 when extending the driver interface. The card and accel structures are
119779 merged into the ExaDriverRec, which is to be allocated using
119780 exaDriverAlloc(). The driver structure also grows exa_major and
119781 exa_minor, which drivers fill in and have checked by EXA
119782 (double-checking that the driver really did check that the EXA version
119783 was correct). Removes exaInitCard(), which is replaced by the driver
119784 filling in the rec by hand, and the exaGetVersion() and related
119785 EXA_*VERSION which are replaced by always using the XFree86 loadable
119786 module versioning.
119787
119788commit 65aa33f9173b1554924437685698f7c5f645a3c4
119789Author: Lars Knoll <lars@trolltech.com>
119790Date: Wed Mar 8 06:19:37 2006 +0000
119791
119792 render/picture.c Initialize the format of a source picture to
119793 PICT_a8r8g8b8. Fixes a failure in the gradients test of rendercheck. In
119794 the long term we could do better by setting the format to something
119795 without alpha whenever the gradient doesn't contain colors with alpha.
119796 This triggers a reduction of the over operation to a pure source
119797 operation.
119798
119799commit cb5090e8d60f4e9780c859faeea5c24587f6bee7
119800Author: Eric Anholt <anholt@freebsd.org>
119801Date: Wed Mar 8 03:32:07 2006 +0000
119802
119803 Bug #6150: Do the obvious fix of an insane sanity check in
119804 xf86InitFBManager. (Julio M. Merino Vidal)
119805
119806commit 2e6f801fe1a749f6a4db2cfd8a43abec5caceae0
119807Author: Ian Romanick <idr@us.ibm.com>
119808Date: Tue Mar 7 23:58:22 2006 +0000
119809
119810 Numerous amounts refactoring and comment adding (see ChangeLog for file by
119811 file details). The primary intention for these changes is to pave the
119812 way for the new device probing and PCI configuration code that I'm
119813 working on.
119814
119815commit b7d2dfc1e5e07051732303731ff3e4e76852dd94
119816Author: Eric Anholt <anholt@freebsd.org>
119817Date: Tue Mar 7 20:06:15 2006 +0000
119818
119819 Add appropriate MIT license. Oops.
119820
119821commit 9d8c0e4bcbb111e860b7c3c33c224c22589006b1
119822Author: Eric Anholt <anholt@freebsd.org>
119823Date: Tue Mar 7 19:57:46 2006 +0000
119824
119825 Add a new flag to ephyr, "-fakexa", which turns on an EXA acceleration
119826 implementation that calls fb to get its work done. The purpose is to
119827 have a trusted EXA driver for use with testing changes to the core of
119828 EXA. However, fakexa has not received much testing yet, lacks offscreen
119829 pixmaps support, and doesn't reliably provide garbage when EXA doesn't
119830 get its syncing right. All of these should be fixed soon.
119831
119832commit 0a3d6c739968bf5af81fc0e8ea7211c20d52080b
119833Author: Eric Anholt <anholt@freebsd.org>
119834Date: Tue Mar 7 19:49:31 2006 +0000
119835
119836 Remove stale EXA files, which failed to get removed during the move to
119837 top-level, somehow.
119838
119839commit 68a8963f726cb92624665669813b6d952d53556e
119840Author: Luc Verhaegen <libv@skynet.be>
119841Date: Tue Mar 7 16:00:57 2006 +0000
119842
119843 Fix cvt -r check again. CH7011 TV encoder had 800x600 PAL hit the check.
119844
119845commit 0693083335185ce05ee64546151f3fc43ce98575
119846Author: Lars Knoll <lars@trolltech.com>
119847Date: Mon Mar 6 21:00:09 2006 +0000
119848
119849 render/picture.c Correctly initialize devPrivates variable in source only
119850 pictures to 0
119851 miext/cw/cw.h Don't try to access devPrivates of source only pictures
119852
119853commit 448997ebcd2bab02be1059b07b91b63b0d05d268
119854Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119855Date: Sun Mar 5 16:43:10 2006 +0000
119856
119857 Only output SetClientVersion message if verbosity > 1, like other
119858 extensions do
119859
119860commit d921173833cc207380eb08b6675393f5e8139d5f
119861Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119862Date: Sun Mar 5 16:35:08 2006 +0000
119863
119864 define SYS_LIBS to hold system dependant libraries that may needed. and add
119865 it to libraries list where needed. Update ChangeLog for previous
119866 changes too
119867
119868commit 82cbd2ee0d20225b9edbb5246c8ed116b4614e1a
119869Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119870Date: Sun Mar 5 16:33:17 2006 +0000
119871
119872 Don't hard-code -DUSE_DEV_IO here. configure generates the proper OS
119873 specific values here.
119874
119875commit b56a1513d27f84dcd55f3dc6053f183aa6f7855b
119876Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119877Date: Sun Mar 5 16:32:40 2006 +0000
119878
119879 Definitions for bswapxx() macros on OpenBSD.
119880
119881commit 4335868476af7c821c64def52b102b93ae91f8b0
119882Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119883Date: Sun Mar 5 16:13:21 2006 +0000
119884
119885 Fix build with non GNU make.
119886
119887commit b2f8f410c0bb8bc24039b2a593f8a2a483659914
119888Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
119889Date: Fri Mar 3 09:54:54 2006 +0000
119890
119891 https://bugs.freedesktop.org/show_bug.cgi?id=4341 Make Xming error messages
119892 more meaningful.
119893
119894commit 29237c1977e454511e0d0244c68d34d572b68458
119895Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
119896Date: Fri Mar 3 09:50:55 2006 +0000
119897
119898 https://bugs.freedesktop.org/show_bug.cgi?id=4538 Fix mouse button release
119899 on multiwindows scrolling.
119900
119901commit 06f01623fde61f1a11c2c1ecfae6a4c346473b05
119902Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
119903Date: Fri Mar 3 09:43:42 2006 +0000
119904
119905 https://bugs.freedesktop.org/show_bug.cgi?id=5138 Check for NULL pointer
119906
119907commit 054c291b274b238893e408e070aef13a7933400b
119908Author: Felix Kuehling <fxkuehl@gmx.de>
119909Date: Thu Mar 2 18:35:08 2006 +0000
119910
119911 Fix build against Mesa CVS HEAD: added s_blit.c to symlink-mesa.sh.
119912
119913commit c1a82b9554028640dc4e08f042f1a8faf3372627
119914Author: Brian Paul <brian.paul@tungstengraphics.com>
119915Date: Thu Mar 2 03:43:26 2006 +0000
119916
119917 added s_blit.c file
119918
119919commit 5f4d11c8d926cf396e0a8e203e14a8e1e123e011
119920Author: Jesse Barnes <jbarnes@virtuousgeek.org>
119921Date: Wed Mar 1 16:31:53 2006 +0000
119922
119923 fix spelling error, document EXA_TWO_BITBLT_DIRECTIONS device flag
119924
119925commit 044a3abb382a4850722c391f04d09d3160790814
119926Author: Jesse Barnes <jbarnes@virtuousgeek.org>
119927Date: Wed Mar 1 16:28:34 2006 +0000
119928
119929 Add accelerated two directional blt support to EXA
119930
119931commit 96ca329382141fd50dccb1cc35a71a333d80bce4
119932Author: Ian Romanick <idr@us.ibm.com>
119933Date: Tue Feb 28 23:07:09 2006 +0000
119934
119935 Remove redundant definition of struct Inst. Safeguard xf86AddDriver against
119936 future additions to DriverRec.
119937
119938commit 1cfa9f647e0241f4b9e56556b128d7bfd987eaca
119939Author: Daniel Stone <daniel@fooishbar.org>
119940Date: Tue Feb 28 16:55:26 2006 +0000
119941
119942 Bug #5216: Allow options to appear with other components.
119943
119944commit e3b6b95f29cb2ea00b4290d694c5e202b8d180ad
119945Author: Adam Jackson <ajax@nwnk.net>
119946Date: Tue Feb 28 16:26:16 2006 +0000
119947
119948 Bug #5627: Fix Xprint font symlinking. (TIlman Sauerbeck)
119949
119950commit e7f0b84fa7bd0c40cb456ec4e447103442c8dae3
119951Author: Jesse Barnes <jbarnes@virtuousgeek.org>
119952Date: Tue Feb 28 05:20:20 2006 +0000
119953
119954 fix exaInitCard by making it a real function
119955
119956commit 088e5768faa90fe16de41b135b1111b5d25c64ad
119957Author: Felix Kuehling <fxkuehl@gmx.de>
119958Date: Mon Feb 27 18:12:24 2006 +0000
119959
119960 Fixing the Mesa build again, sigh. Add slang_execute_x86.c. Add
119961 -I../shader/slang to swrast INCLUDES.
119962
119963commit 345d99c972cac67f2cdc38750e4ba2dea1cdb360
119964Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
119965Date: Mon Feb 27 16:19:39 2006 +0000
119966
119967 Typo fixes (Nicholas Joly, XFree86 bugzilla #1658)
119968
119969commit 6b08a5013b4e9e350ba461c9a59d30bb41feef8f
119970Author: Jesse Barnes <jbarnes@virtuousgeek.org>
119971Date: Sat Feb 25 20:26:49 2006 +0000
119972
119973 EXA driver doc cleanups and additions.
119974
119975commit f41ec003f39c575299429897d4287233184583ad
119976Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch>
119977Date: Sat Feb 25 01:17:10 2006 +0000
119978
119979 Add two radeon pci ids (one is used for a radeon mobility X700 XL in a
119980 medion notebook, the other is for a AIW X800 VE)
119981
119982commit 01a0bf881ada03ca3c27bdef7423c760c3bc2f9c
119983Author: Matthieu Herrb <matthieu.herrb@laas.fr>
119984Date: Fri Feb 24 17:01:57 2006 +0000
119985
119986 OpenBSD supports PCVT and WSCONS. PCCONS is long gone.
119987
119988commit d3e1587c20c155b7873b6646ddf0b96f806f8a7f
119989Author: Kristian Høgsberg <krh@redhat.com>
119990Date: Fri Feb 24 16:50:42 2006 +0000
119991
119992 file glxdri.c was initially added on branch accel_indirect_branch.
119993
119994commit 5d9a620726d2b0ad89625574478d2fd4536485b0
119995Author: Adam Jackson <ajax@nwnk.net>
119996Date: Thu Feb 23 19:25:57 2006 +0000
119997
119998 Remove redundant composite op reduction, done in Render now.
119999
120000commit 028d6903f674fa77617f333b25356710d1682b05
120001Author: Alexander Gottwald <alexander.gottwald@s1999.tu-chemnitz.de>
120002Date: Wed Feb 22 16:29:07 2006 +0000
120003
120004 Bug #5978: Added missing swap of input variables. Added missing cases for
120005 GL_SECONDARY_COLOR_ARRAY and GL_FOG_COORD_ARRAY (Colin McDonald)
120006
120007commit 43324132afcbb6b231efcc24ec72ee44678d5771
120008Author: Alexander Gottwald <alexander.gottwald@s1999.tu-chemnitz.de>
120009Date: Wed Feb 22 16:08:56 2006 +0000
120010
120011 Added Mesa include directory
120012
120013commit 43fbcc28c960ce5abe1d3223441c3dc5a10cde27
120014Author: Jeremy C. Reed <reed@reedmedia.net>
120015Date: Wed Feb 22 02:47:00 2006 +0000
120016
120017 Update to build against Mesa HEAD. (Thank you Felix for feedback on xorg
120018 list.)
120019
120020commit cfdacab33a62b47f22bb54683e3ca20ec9824864
120021Author: Ian Romanick <idr@us.ibm.com>
120022Date: Tue Feb 21 00:02:08 2006 +0000
120023
120024 Eliminate unused PCI BIOS reading functionality. The old code used several
120025 function pointers to implement a level of flexability that was never
120026 used. The code also had unused support for extracting a single image
120027 type from a larger expansion ROM.
120028 Fix the spelling of PCI_BIOS_OPEN_FIRMWARE.
120029 Fix a couple errors in #ifdef debug code.
120030 These changes have been tested on x86 and x86-64 Linux.
120031
120032commit 5fd0f94006775e2271107c960dfa3314dddd9a5f
120033Author: Ian Romanick <idr@us.ibm.com>
120034Date: Mon Feb 20 23:45:50 2006 +0000
120035
120036 Eliminate all the code for querying the PCI class from the PCI ID database.
120037 Class information is not, and never has been, stored there. Therefore,
120038 this is just a bunch of elaborate code to read 0x00000000.
120039 This has received testing on x86 and x86-64 Linux.
120040
120041commit 6d7083bd69724586338d79784655328f1fcd2ae5
120042Author: Adam Jackson <ajax@nwnk.net>
120043Date: Mon Feb 20 22:16:49 2006 +0000
120044
120045 indent fixes (OMG SO UGLY), and nuke old RCS keywords.
120046
120047commit 5480c537cee79b324736eac3e438a4713dfa1036
120048Author: Adam Jackson <ajax@nwnk.net>
120049Date: Mon Feb 20 21:50:49 2006 +0000
120050
120051 ANSIfy Xi/. Mostly automated via protoize(1).
120052
120053commit cc42e153c2437fe9c94b0c20e1b56277474d94d0
120054Author: Felix Kuehling <fxkuehl@gmx.de>
120055Date: Mon Feb 20 03:00:09 2006 +0000
120056
120057 Update to build against Mesa HEAD.
120058
120059commit ed3ea887a6b1c9fdc83895c40da34076121f05e9
120060Author: Eric Anholt <anholt@freebsd.org>
120061Date: Thu Feb 16 21:49:51 2006 +0000
120062
120063 Fix the encoding of DES's name.
120064
120065commit ea0b3f65f77a78df1671b09739c00762a8875607
120066Author: Eric Anholt <anholt@freebsd.org>
120067Date: Thu Feb 16 21:45:12 2006 +0000
120068
120069 Fix build of Xorg by putting xf86bigfont back into builtin instead of
120070 module sources list.
120071
120072commit d97f29be22e22e6f5bc23229ffa5ef087f992c8c
120073Author: Eric Anholt <anholt@freebsd.org>
120074Date: Thu Feb 16 21:35:32 2006 +0000
120075
120076 Bug #5453: Don't forget to still do AM_CONDITIONAL for XVMC even if XV is
120077 disabled, and also force XVMC disabled if XV is disabled. (Dag-Erling
120078 Sm�rgrav)
120079
120080commit d33c2e0d1079f93f4ba8b28d19950d384a9e7c32
120081Author: Keith Packard <keithp@keithp.com>
120082Date: Thu Feb 16 19:36:39 2006 +0000
120083
120084 Add oscolor.h
120085
120086commit 60d4839a2d373cc6d8c0d5004284494d3a994a63
120087Author: Keith Packard <keithp@keithp.com>
120088Date: Thu Feb 16 19:03:07 2006 +0000
120089
120090 Forgot to include this file in previous patch
120091
120092commit 8987b2c1efc9a4667e278e6ba411772ba2a4a4e6
120093Author: Keith Packard <keithp@keithp.com>
120094Date: Thu Feb 16 07:17:31 2006 +0000
120095
120096 Make more extensions optional in build (for kdrive). Fix kdrive build for
120097 actual hardware. Fix kdrive pointer signed/unsigned types. Add
120098 kdrive-required YX rotation functions. Replace rgb text file loading
120099 with static rgb color table.
120100
120101commit c8acb342695936db062c966029019a458d45459e
120102Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
120103Date: Thu Feb 16 06:52:12 2006 +0000
120104
120105 Fix an occasional crash on VT switches: the server would save a pointer to
120106 the current cursor when disabling FB access and would try to restore
120107 that cursor when re-enabling. However, that cursor might have been
120108 destroyed in between. This fixes it by updating the saved cursor
120109 pointer when a cursor is set and vtSema is FALSE.
120110
120111commit c845e152f52b3e8cef579797c5c8834ad2fd2cd5
120112Author: Eric Anholt <anholt@freebsd.org>
120113Date: Thu Feb 16 05:46:08 2006 +0000
120114
120115 Bug #5871: Drop special build infrastructure left over from libcwrapper
120116 times. (George Fufutos)
120117
120118commit c2f685e64464ccf86ed47ae37f74bf46877e0739
120119Author: Dave Airlie <airlied@linux.ie>
120120Date: Thu Feb 16 05:18:20 2006 +0000
120121
120122 Fix XSERVER64 should be _XSERVER64
120123
120124commit 1bbd5e49b1dcc2e3f9785bd2bb58f946b9998099
120125Author: Eric Anholt <anholt@freebsd.org>
120126Date: Thu Feb 16 04:59:45 2006 +0000
120127
120128 Bug #5869: Remove traces of EXTMODULE define, which doesn't appear to be
120129 useful any more. (George Fufutos)
120130
120131commit c03b06bdf04fa8500d0f85314c7268848b4d50be
120132Author: Eric Anholt <anholt@freebsd.org>
120133Date: Thu Feb 16 04:39:00 2006 +0000
120134
120135 Bug #5888: Remove orphaned laymodule.c from miext/layer removal. (George
120136 Fufutos)
120137
120138commit 3b32e902c7a12aa2320da27d984029cde28fd8c3
120139Author: Eric Anholt <anholt@freebsd.org>
120140Date: Thu Feb 16 01:49:23 2006 +0000
120141
120142 Add entry missed in my last commit.
120143
120144commit c4767794ef5b014ae25fe8541e72348ecfb1ee49
120145Author: Zephaniah E. Hull <warp@aehallh.com>
120146Date: Thu Feb 16 01:03:09 2006 +0000
120147
120148 Export xf86ActivateDevice, used by the evdev driver.
120149
120150commit dc0354104cb4057dfcc7b2ccb8e2ae8474d70b15
120151Author: Eric Anholt <anholt@freebsd.org>
120152Date: Thu Feb 16 00:14:11 2006 +0000
120153
120154 Move EXA implementation up to the top level and remove its XFree86
120155 dependencies. It was nearly abstract enough already to be used by
120156 multiple DDXes. This will be useful for EXA development through
120157 providing a fake acceleration implementation within Xephyr, so that
120158 testing can be done on new EXA code without worrying about buggy
120159 drivers.
120160
120161commit c170aa830d0ce3dbff6b30081e04c3f91bf921be
120162Author: Eric Anholt <anholt@freebsd.org>
120163Date: Wed Feb 15 23:27:40 2006 +0000
120164
120165 Forced commit to note repocopy from hw/xfree86/exa/
120166
120167commit 6770f1bdb145e7a6c431d0523f10d12155f58273
120168Author: Eric Anholt <anholt@freebsd.org>
120169Date: Wed Feb 15 21:09:14 2006 +0000
120170
120171 Define NO_LIBCWRAPPER in dix-config.h, and rely on Mesa including
120172 dix-config.h if DIX_HAVE_CONFIG_H is defined to get it and _XSERVER64,
120173 instead of defining things like this per directory.
120174
120175commit 7d7fc927cd90146788780477b8e3379d91c3b910
120176Author: Adam Jackson <ajax@nwnk.net>
120177Date: Wed Feb 15 20:47:44 2006 +0000
120178
120179 Remove a few #ifdef vms; whatever problem that was solving should assuredly
120180 be solved some other way.
120181
120182commit f105b8da11fcf337512b3c39da3368f98da07a33
120183Author: Adam Jackson <ajax@nwnk.net>
120184Date: Wed Feb 15 20:44:13 2006 +0000
120185
120186 Mark everything in dixsym.c as _X_EXPORT.
120187
120188commit 010d6effa6fa210251b12459882e88aeee82c2c0
120189Author: Adam Jackson <ajax@nwnk.net>
120190Date: Wed Feb 15 19:15:32 2006 +0000
120191
120192 Mark everything in {ext,font}sym.c as _X_EXPORT.
120193
120194commit 3fe482c77e7b3e46739d011d8bbdee527d7a42fc
120195Author: Adam Jackson <ajax@nwnk.net>
120196Date: Wed Feb 15 19:05:55 2006 +0000
120197
120198 Mark everything in misym.c as _X_EXPORT.
120199
120200commit 6ad4325b87889e1aada9333d750b7bb586c38b52
120201Author: Kristian Høgsberg <krh@redhat.com>
120202Date: Wed Feb 15 18:26:45 2006 +0000
120203
120204 Update to build against mesa head.
120205
120206commit 50e2ff9a2500078ebbd833fddab0d93f3a50b6a6
120207Author: Eric Anholt <anholt@freebsd.org>
120208Date: Wed Feb 15 03:20:55 2006 +0000
120209
120210 Remove the waitSync from KdDisableScreen and push it off to drivers'
120211 disableAccel hook, which is more correct anyway. This makes kdrive.c
120212 not have any knowledge of kaa, opening the way for using exa from
120213 kdrive.
120214
120215commit 0446aafa9467f43515fb578d50f45e2c3153c8cf
120216Author: Eric Anholt <anholt@freebsd.org>
120217Date: Wed Feb 15 03:07:23 2006 +0000
120218
120219 Avoid some more libcwrapper damage that prevented kdrive linking.
120220
120221commit 5c9b6f0fb01252d704de1bbdf3015dee7f956593
120222Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
120223Date: Tue Feb 14 08:14:42 2006 +0000
120224
120225 DRIGetSecs() would call getsecs() when XFree86LOADER is defined, relying on
120226 the wrappers to provide it. Wrapper gone, and getsecs doesn't exist on
120227 linux so it now blows up. Fixes it by just calling gettimeofday() in
120228 all cases instead.
120229
120230commit 049dca0f43eb2179d2c61033a17ff1a89f8fb689
120231Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
120232Date: Tue Feb 14 08:11:41 2006 +0000
120233
120234 Remove useless line of code that contained a bug and triggered a gcc
120235 warning. This variable will be overriden before being used anyway.
120236 (Bugzilla #5595)
120237
120238commit 1132d0e6102d4564f70f0e8c98854e3acf25b109
120239Author: Dave Airlie <airlied@linux.ie>
120240Date: Tue Feb 14 06:27:59 2006 +0000
120241
120242 update to latest Mesa CVS HEAD
120243
120244commit bb8c36690ab411c11aa8dd3d4520d513eb8f9091
120245Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120246Date: Tue Feb 14 04:20:37 2006 +0000
120247
120248 Bump shadow module version number to 1.1.0 from 1.0.0
120249
120250commit cc9dfab0b31c7956f99d1f1b9c195065b5e18c29
120251Author: Adam Jackson <ajax@nwnk.net>
120252Date: Mon Feb 13 18:57:38 2006 +0000
120253
120254 (Reverted)
120255
120256commit 83dd6241c8cd81e8d897bd17588ada92a945e647
120257Author: Adam Jackson <ajax@nwnk.net>
120258Date: Mon Feb 13 18:55:44 2006 +0000
120259
120260 Reverted, did nothing anyway, I'm not smart today.
120261
120262commit 4a7f6f53cad541e8c5042a6472e3b3886fc9b7e6
120263Author: Adam Jackson <ajax@nwnk.net>
120264Date: Mon Feb 13 18:09:51 2006 +0000
120265
120266 Further op reduction when both src and dst alpha are absent.
120267
120268commit 28ced9f3e0dd4bd81067f590a1d64ba0844edb06
120269Author: Eric Anholt <anholt@freebsd.org>
120270Date: Mon Feb 13 05:29:00 2006 +0000
120271
120272 Add missing ChangeLog text for r1.2 of GL/glx/indirect_reqsize.h
120273
120274commit 4839e91fcab4c344e672154a447d8c7035fce1f4
120275Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
120276Date: Mon Feb 13 05:03:13 2006 +0000
120277
120278 HAS_MKSTEMP vs. HAVE_MKSTEMP (From Fredrik Höglund)
120279
120280commit 1a4f20541a9f4f41f444d826d743899ea2dee2db
120281Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
120282Date: Mon Feb 13 04:56:27 2006 +0000
120283
120284 Fix linux build without libc wrappers (From Fredrik Höglund)
120285
120286commit 2dc7b5e0d96a187bfbb355caa788f0fdcd88eaad
120287Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
120288Date: Mon Feb 13 04:43:40 2006 +0000
120289
120290 Move call to xf86WrapperInit() to OsVendorInit() in xf86Init.c and remove
120291 stubs in other DDX.
120292
120293commit 34d0b9228f46c2f87be74dddc9c7d97aab091d03
120294Author: Eric Anholt <anholt@freebsd.org>
120295Date: Sun Feb 12 20:53:35 2006 +0000
120296
120297 Simplify ops that would use the alpha channel when an alpha channel is
120298 always 1.0, and short circuit PictOpDst for good measure.
120299
120300commit 5f45776ef3b9256bea44842d1c50f269422531a1
120301Author: Eric Anholt <anholt@freebsd.org>
120302Date: Sun Feb 12 10:30:47 2006 +0000
120303
120304 Add missing HAVE_DIX_CONFIG_H which caused issues with mismatched screen
120305 structure interpretations, and remove a bunch of unused junk from
120306 kdrive-config.h. Xephyr almost works on my amd64.
120307
120308commit 5249416d091d59c248c8dda44529b8aa4910b1a0
120309Author: Eric Anholt <anholt@freebsd.org>
120310Date: Sat Feb 11 22:40:50 2006 +0000
120311
120312 Add stub xf86WrapperInits so that the servers will build even if os/ was
120313 built with XFree86LOADER set.
120314
120315commit a2a5254675a6b7ef0f7da9caa76c028b7c526502
120316Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
120317Date: Sat Feb 11 19:16:51 2006 +0000
120318
120319 Add <string.h>, <stdlib.h>, and <stdio.h> to clear undefined function
120320 warnings after the removal of libcwrapper headers.
120321
120322commit d6337c83241f0fa4bb03039a9767b58d8a1a7c91
120323Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
120324Date: Sat Feb 11 17:42:31 2006 +0000
120325
120326 <sys/limits.h> -> <limits.h> so we can compile on non-BSD OS'es
120327
120328commit 4fafba61d5402d4e4d2c21ba1be3ed8969b99334
120329Author: Eric Anholt <anholt@freebsd.org>
120330Date: Sat Feb 11 03:03:45 2006 +0000
120331
120332 Remove libcwrapper damage from GLX (requires fresh Mesa HEAD), and get it
120333 compiling in kdrive.
120334
120335commit c3d14036729fd186d4ec7ca1de603e1f2d174e2f
120336Author: Eric Anholt <anholt@freebsd.org>
120337Date: Fri Feb 10 22:00:30 2006 +0000
120338
120339 Remove libcwrapper usage from xorg server modules. The libcwrapper is only
120340 of (marginal) use in the drivers, and that usage remains.
120341
120342commit a8cec1b656f57746758613213de1d6e5acb79451
120343Author: Eric Anholt <anholt@freebsd.org>
120344Date: Fri Feb 10 09:00:02 2006 +0000
120345
120346 Merge from kdrive: use RECT_PRIM to avoid tearing in xvideo.
120347
120348commit efc3fab7f4b29f56fffd21304c64c03a48aa5b4b
120349Author: Eric Anholt <anholt@freebsd.org>
120350Date: Fri Feb 10 07:52:05 2006 +0000
120351
120352 Make kdrive (i.e. Xephyr only) buildable on FreeBSD and probably other OSes
120353 without linux VT switching, fbdev, and vm86 support.
120354
120355commit fa3a65e33d8c893c7867ea507afc7caa1361aa9c
120356Author: Eric Anholt <anholt@freebsd.org>
120357Date: Fri Feb 10 07:47:21 2006 +0000
120358
120359 Remove the include of X11/misc.h, which fails to compile and isn't
120360 necessary.
120361
120362commit d875bdb2756b082ce93bd86016c369ea85c04d17
120363Author: Jeremy C. Reed <reed@reedmedia.net>
120364Date: Fri Feb 3 02:44:19 2006 +0000
120365
120366 hw/xfree86/os-support/xf86_OSlib.h Fix sysmouse handling on DragonFly,
120367 mostly garbage arrived. From Joerg Sonnenberger.
120368
120369commit 0946bb9427695a4314e5c43de573b3a75a18e466
120370Author: Jeremy C. Reed <reed@reedmedia.net>
120371Date: Fri Feb 3 02:37:52 2006 +0000
120372
120373 Fix sysmouse handling on DragonFly, mostly garbage arrived. From Joerg
120374 Sonnenberger.
120375
120376commit 5e2a7af23bd0f46fbddca34098cb297be58b7a55
120377Author: Eric Anholt <anholt@freebsd.org>
120378Date: Thu Feb 2 21:07:06 2006 +0000
120379
120380 Move the frequently-repeated code to get the pixmap that backs a drawable
120381 to a new function, exaGetDrawablePixmap().
120382
120383commit ee3c7ccb175752dbeaed6b0113d0819b3fcd2398
120384Author: Eric Anholt <anholt@freebsd.org>
120385Date: Thu Feb 2 20:51:54 2006 +0000
120386
120387 Remove more debugging leftovers.
120388
120389commit 3366b6836572461209bb2f8aa28d9e662067dc54
120390Author: Eric Anholt <anholt@freebsd.org>
120391Date: Thu Feb 2 20:09:14 2006 +0000
120392
120393 Rearrange and rename EXA code to be a bit more logically organized. Also
120394 removes a little bit of debugging leftovers. Summary:
120395 exa.c -> exa.c (miscellaneous code) exa_accel.c (all acceleration code)
120396 exa_migration.c (migration logic) exaasync.c -> exa_unaccel.c (software
120397 fallbacks) exapict.c -> exa_render.c (render extension stuff)
120398 exaoffscreen.c -> exa_offscreen.c exaPriv.h -> exa_priv.h
120399
120400commit 25d4ff870d49533d82a77f144722ff7934d52e0b
120401Author: Eric Anholt <anholt@freebsd.org>
120402Date: Thu Feb 2 20:04:15 2006 +0000
120403
120404 Forced commit to note repocopies: exa.c -> exa_accel.c exa_migration.c
120405 exaasync.c -> exa_unaccel.c exapict.c -> exa_render.c exaoffscreen.c ->
120406 exa_offscreen.c exaPriv.h -> exa_priv.h
120407
120408commit 2ab487d4d20e4e34e73cc6d87f41bf0836c7a8af
120409Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120410Date: Wed Feb 1 22:20:05 2006 +0000
120411
120412 Add a new function RRGetRotation() which does exactly the same thing as
120413 xf86GetRotation(), but allows for drivers to provide their own RandR
120414 implementation. xf86GetRotation could be obsoleted by this change.
120415
120416commit f4898b409376803c9a9dd8475bdd5576ff1cc59d
120417Author: Luc Verhaegen <libv@skynet.be>
120418Date: Tue Jan 31 14:49:43 2006 +0000
120419
120420 Further bug #5386 fixes: Fix some problems with the EDID code: Some
120421 bitoffsets were wrong. Unknown Detailed Sections weren't handled
120422 properly and defaulted to Detailed Timing.
120423
120424commit 437b385ce4cc3ff00e14d3d39f4a2f6c8f0c67a0
120425Author: Luc Verhaegen <libv@skynet.be>
120426Date: Tue Jan 31 13:55:01 2006 +0000
120427
120428 Bring the cvt utility up to date with bug #5386 changes. Fix 2 issues with
120429 the generator routine: the allocated modeline wasn't nulled and
120430 mode->name's \0 wasn't copied over. PrintModeLine was rewritten and
120431 HDisplay gets rounded up to character width instead of refused.
120432
120433commit 8f3c69dcf17691f71bca7b0a2cd34f7788a97b8c
120434Author: Luc Verhaegen <libv@skynet.be>
120435Date: Tue Jan 31 13:04:02 2006 +0000
120436
120437 Accept modes with less than 25% horizontal blanking again (you can push old
120438 gtf timing to below 25%), only stop cvt reduced blanking. Users should
120439 be free to blow up their monitors if they so choose.
120440
120441commit 701b63cf1dcd3e49602114fb1dde45a74b4e1122
120442Author: Donnie Berkholz <spyderous@gentoo.org>
120443Date: Mon Jan 30 20:04:56 2006 +0000
120444
120445 Update to build against Mesa trunk.
120446
120447commit dd50015b05b901fe0c60717512c854389610aea2
120448Author: Eric Anholt <anholt@freebsd.org>
120449Date: Sat Jan 28 02:20:37 2006 +0000
120450
120451 Add libc_wrapper support for random(), which will be used in upcoming EXA
120452 memory manager work.
120453
120454commit 3d1667278ff309d7f8e61a6d330f712bae5bcd41
120455Author: Eric Anholt <anholt@freebsd.org>
120456Date: Sat Jan 28 00:37:52 2006 +0000
120457
120458 Remove leftover variables for cfb24 build, and finish commenting out Xglx
120459 standalone stuff, which some versions of automake get whiny about.
120460
120461commit ab01eb247f9e5d7c9995bf2d6432358cd64bf11d
120462Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120463Date: Fri Jan 27 12:27:34 2006 +0000
120464
120465 update pci ids
120466
120467commit 261aa4403c77203f8f02b399ddd382c731dda324
120468Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120469Date: Thu Jan 26 09:04:22 2006 +0000
120470
120471 remove that, and will put it in a i810 driver specific Changelog
120472
120473commit 94e678fd014c61d12591d7398b6591f24c3d71f1
120474Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120475Date: Thu Jan 26 08:49:19 2006 +0000
120476
120477 add changelog for i810 updates
120478
120479commit c5e93182905332383ca3ef5db3f334cec69c8dda
120480Author: Donnie Berkholz <spyderous@gentoo.org>
120481Date: Thu Jan 26 04:32:45 2006 +0000
120482
120483 Really allow linking against Mesa trunk to work.
120484
120485commit 0dc0f17f27f99da79c99031b41b0c0e95ef035f5
120486Author: Adam Jackson <ajax@nwnk.net>
120487Date: Thu Jan 26 04:10:43 2006 +0000
120488
120489 Speed up checkout and autogen by removing disused iplan2p4 and ilbm.
120490
120491commit 023d2b4e3c392eed1f149dc5b13a83429cd052a3
120492Author: Daniel Stone <daniel@fooishbar.org>
120493Date: Thu Jan 26 00:23:44 2006 +0000
120494
120495 Add forgotten HAVE_BACKTRACE define.
120496
120497commit 14fdd81614cdd6ef7e01976a43da8b6a3bf8386e
120498Author: Adam Jackson <ajax@nwnk.net>
120499Date: Wed Jan 25 23:05:26 2006 +0000
120500
120501 Remove xf8_32wid, it's dead code with no maintainer interest. Also remove
120502 cfb24, since xf8_32wid was the only user.
120503
120504commit 2e28f4104ddf94a8f9a70fe6b2a2a6859ffedc8f
120505Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120506Date: Tue Jan 24 22:05:33 2006 +0000
120507
120508 Allow current trunk to build against Mesa trunk
120509
120510commit 0d9ed2624fe8fb95c57930da523351556ba11351
120511Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120512Date: Mon Jan 23 22:01:34 2006 +0000
120513
120514 Commit #4633 - Initial mouse pointer incorrect with EXA which also fixes
120515 the repaint of the cursor image with randr events.
120516
120517commit 9148d8700b7c5afc2644e5820c57c509378f93ce
120518Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120519Date: Mon Jan 23 13:59:14 2006 +0000
120520
120521 Commit slight variation of bug #5460 which is the merge of the new shadow
120522 code from kdrive.
120523
120524commit cfd3988ed906ab48ca4362256f8dbb8852d7ca0a
120525Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120526Date: Mon Jan 23 13:58:19 2006 +0000
120527
120528 wrap with if XORG / endif
120529
120530commit f51ecc66e9ad6d2c3541b1dafa7659da5a0a3a86
120531Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120532Date: Mon Jan 23 13:54:59 2006 +0000
120533
120534 #include "gcstruct.h"
120535
120536commit 80f45fa4dfa011c2ae7bcb34f87aafb91763f1fe
120537Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120538Date: Mon Jan 23 13:54:34 2006 +0000
120539
120540 add damage.h & damagestr.h to SDK headers
120541
120542commit af5b3ea4b3df9e9c6dd6993c5e7238a366a3f508
120543Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
120544Date: Thu Jan 19 14:51:09 2006 +0000
120545
120546 add randrstr.h to sdk_HEADERS
120547
120548commit 03ebd37baba2f5af3ab502ff02ec14c15859dc3f
120549Author: Daniel Stone <daniel@fooishbar.org>
120550Date: Thu Jan 19 12:25:01 2006 +0000
120551
120552 Make error() bomb with exit code 1, not 0.
120553
120554commit adce1f16e0d815e5c762407da3544a7d2eff9303
120555Author: Eric Anholt <anholt@freebsd.org>
120556Date: Thu Jan 19 00:06:57 2006 +0000
120557
120558 Only try to use byteswap.h on linux. Assume that everyone else (thinking of
120559 BSDs here) has sys/endian.h, and use macros as appropriate for the
120560 names. This should probably be in a gloabl header.
120561
120562commit 1c3f8727b2349c9b988eaa744f11366322d42538
120563Author: Adam Jackson <ajax@nwnk.net>
120564Date: Wed Jan 18 19:42:56 2006 +0000
120565
120566 More kdrive merge, fast path fbBlt to use memcpy() when possible. Good for
120567 -5% to 60% speedup on XGetImage, and 0% to 10% speedup on copies within
120568 host memory. Based on work by Jaymz Julian.
120569
120570commit e70b64b93024d05519014fb1b76fe26bd9f3a496
120571Author: Dave Airlie <airlied@linux.ie>
120572Date: Wed Jan 18 07:15:55 2006 +0000
120573
120574 Updated xgl code drop from Novell + xserver tree changes
120575
120576commit b5356e0afaf2b660c8905f63d5fdcb03402b81c5
120577Author: Dave Airlie <airlied@linux.ie>
120578Date: Wed Jan 18 07:00:50 2006 +0000
120579
120580 typo in last change
120581
120582commit 506eca5f57b960a6650c3387047a6ae8a22181e9
120583Author: Dave Airlie <airlied@linux.ie>
120584Date: Wed Jan 18 06:56:52 2006 +0000
120585
120586 Wrap sdk_HEADERS in if XORG as otherwise installing non-xorg servers
120587 breaks.
120588
120589commit 8ccf4f2b8fdb5e57d2ec5f2d54731fbf83fb9d8c
120590Author: Dave Airlie <airlied@linux.ie>
120591Date: Wed Jan 18 06:49:17 2006 +0000
120592
120593 This is a fix from David Reveman from the xserver tree, Make fbPadPixmap
120594 work with negative stride
120595
120596commit a1f9262c6acd195c0fcf5f602d5ca0c252993521
120597Author: Kristian Høgsberg <krh@redhat.com>
120598Date: Tue Jan 17 21:27:49 2006 +0000
120599
120600 file glxvisuals.c was initially added on branch accel_indirect_glx.
120601
120602commit c2dedf4d17f8a5b1a8037fd2b4e29122ef78945a
120603Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
120604Date: Thu Jan 12 22:14:56 2006 +0000
120605
120606 Thu Jan 12 17:09:18 2006 Søren Sandmann <sandmann@redhat.com>
120607 Add new functions to enable and disable events on Map and Unmap.
120608 Use them here to make sure Composite redirect doesn't cause Map/UnmapNotify
120609 events that would confuse window managers.
120610
120611commit 847d83ec3c90c5b298eaf19ba55251b4a30f4155
120612Author: Ian Romanick <idr@us.ibm.com>
120613Date: Thu Jan 12 00:21:59 2006 +0000
120614
120615 Bug #2996: libglx / libGLcore should use a dispatch table
120616 Port all changes from the (monolithic) accelerated_indirect-0-0-1 branch to
120617 the modular trunk. This will break the Darwin and cygwin builds. Other
120618 than the changes to symlink-mesa.sh and the various Makefile.am files,
120619 to code is identical to what's in the branch.
120620 Reviewed by: airlied, krh
120621
120622commit c56e9a8849ce8dd5c09732ae1860e409e7886690
120623Author: Daniel Stone <daniel@fooishbar.org>
120624Date: Tue Jan 10 03:24:53 2006 +0000
120625
120626 Bomb when symlink-mesa.sh continues, instead of silently failing.
120627
120628commit 2949c705f11f8710301555c039bcecbe748cabd9
120629Author: Daniel Stone <daniel@fooishbar.org>
120630Date: Tue Jan 10 03:23:05 2006 +0000
120631
120632 Add all subdirs used to DIST_SUBDIRS, and files to EXTRA_DIST.
120633 Attempt to build xeglmodule.c, not xglxmodule.c.
120634 Add xf86Sbus.h to EXTRA_DIST, as _HEADERS doesn't appear to get the same
120635 treatment as _SOURCES in terms of automatically DISTing.
120636
120637commit 4fc9eb592a446ad5711bdaa82c60e9fe010fd76a
120638Author: Daniel Stone <daniel@fooishbar.org>
120639Date: Tue Jan 10 02:32:20 2006 +0000
120640
120641 Bomb out if symlink-mesa.sh failed.
120642
120643commit 890ec849479db2510a9b4bc5e5e2f7978ca37b83
120644Author: Daniel Stone <daniel@fooishbar.org>
120645Date: Tue Jan 10 02:30:56 2006 +0000
120646
120647 Add xgl to DIST_SUBDIRS.
120648
120649commit 53dbd00a75313ec5301ca95b2e91d5d02bdaf820
120650Author: Adam Jackson <ajax@nwnk.net>
120651Date: Sun Jan 8 23:43:54 2006 +0000
120652
120653 Remove remaining #ifdef DPSEXT stanzas.
120654
120655commit 7fc9bc44e099f8f046bf707cb87ef7d736933f80
120656Author: Adam Jackson <ajax@nwnk.net>
120657Date: Sat Jan 7 01:29:05 2006 +0000
120658
120659 Compile fix, again, stupid non-clean builds
120660
120661commit 9d62d1e6903ccc095f784279a699b3f40a8f0cf8
120662Author: Adam Jackson <ajax@nwnk.net>
120663Date: Sat Jan 7 00:45:17 2006 +0000
120664
120665 Bug #5218: Don't crash on unconfigured interfaces. (Andrei Barbu)
120666
120667commit 3c58072956c28ebc3ca2eb50c1ff09823e1219d7
120668Author: Adam Jackson <ajax@nwnk.net>
120669Date: Sat Jan 7 00:33:41 2006 +0000
120670
120671 One more build fix.
120672
120673commit 07303c1b42afd1ada98cbc11d1ba616d366017fb
120674Author: Adam Jackson <ajax@nwnk.net>
120675Date: Sat Jan 7 00:05:46 2006 +0000
120676
120677 This version will actually compile
120678
120679commit 7f46aba35ee482e9b28ecc81d1a99d423fc88a70
120680Author: Adam Jackson <ajax@nwnk.net>
120681Date: Fri Jan 6 23:52:23 2006 +0000
120682
120683 Compile fix
120684
120685commit 25babf2791ad42101a86ba2a0f14564328256ee2
120686Author: Adam Jackson <ajax@nwnk.net>
120687Date: Fri Jan 6 23:36:53 2006 +0000
120688
120689 Missed file.
120690
120691commit 13c9e0c094c4e34cd1e43a7cc08b2dca39a32412
120692Author: Adam Jackson <ajax@nwnk.net>
120693Date: Fri Jan 6 23:06:15 2006 +0000
120694
120695 Bug #5525: Build a working Xprt. (Drew Parsons)
120696
120697commit fe0c838b5d8bc8d9cf5a686bb7d3e90682d2d19a
120698Author: Adam Jackson <ajax@nwnk.net>
120699Date: Fri Jan 6 18:06:02 2006 +0000
120700
120701 Move drawable lock acquisition into DRIClipNotify from DRIValidateTree, so
120702 we only take it when clipping a DRI drawable instead of on every tree
120703 update. Note drawable lock acquisition per- screen instead of globally,
120704 and drop it in BlockHandler if necessary.
120705
120706commit 39ce5f1544029412f4060f3e89ce1d87222ef42b
120707Author: Adam Jackson <ajax@nwnk.net>
120708Date: Fri Jan 6 17:05:26 2006 +0000
120709
120710 Remove unused X11R4 DDX compatibility function miClipNotify.
120711
120712commit 07ecb969d7eb8d4ab0bb0b8a55a5f40f3c8ec5e3
120713Author: Eric Anholt <anholt@freebsd.org>
120714Date: Wed Jan 4 03:29:15 2006 +0000
120715
120716 Forced commit to note that glyph privates commit was:
120717 Obtained from: xserver tree (David Reveman)
120718
120719commit b6b88d2f62d8c596171f487dd25fbdbc85d0c5a8
120720Author: Eric Anholt <anholt@freebsd.org>
120721Date: Wed Jan 4 00:05:16 2006 +0000
120722
120723 Correct rounding in divide-by-255 code. Obtained from xserver.
120724
120725commit b9c0ae867e1b52186c26841a77745f7f5a0a76dd
120726Author: Eric Anholt <anholt@freebsd.org>
120727Date: Tue Jan 3 22:36:46 2006 +0000
120728
120729 Remove the manual AddFilter for convolution, which I'm pretty sure
120730 shouldn't be necessary due to it already happening from
120731 PictureSetDefaultFilters.
120732
120733commit 601ab861b46a62b0742ffd3e937c4fab129664f0
120734Author: Eric Anholt <anholt@freebsd.org>
120735Date: Tue Jan 3 22:06:23 2006 +0000
120736
120737 Add glyph privates for Xgl, which uses them to implement a glyph cache. EXA
120738 would probably also like to do this. This breaks module ABI for EXA and
120739 XAA, and likely breaks proprietary drivers as well.
120740
120741commit 1729fc882ceec392331566c95efd5968fe9e97fd
120742Author: Eric Anholt <anholt@freebsd.org>
120743Date: Sat Dec 31 08:06:00 2005 +0000
120744
120745 Change REGION_INIT(pScreen, &foo, NullBox, 0) to REGION_NULL(pScreen,
120746 &foo). While it is no longer (or rather, once again not) required as of
120747 regionstr.h r1.4, it matches the style of the rest of the xorg code.
120748
120749commit c25536a7937b11a5347bfb8796d5cb6eb0445b51
120750Author: Eric Anholt <anholt@freebsd.org>
120751Date: Sat Dec 31 08:01:31 2005 +0000
120752
120753 Initialize the fourcc value in stack-allocated glitz_pixel_format_t
120754 structures. Greatly reduces the number of uninitialized-value accesses
120755 during Xgl startup according to valgrind. Allocating and filling these
120756 in by hand on the stack seems very shady to me.
120757
120758commit e6dab3d7c429a2d30d31f188c4554e870011e051
120759Author: Eric Anholt <anholt@freebsd.org>
120760Date: Fri Dec 30 12:05:47 2005 +0000
120761
120762 Fix the AC_TRY_RUN for sys/linker.h which had no hope due to lacking a
120763 main() to instead use a nice AC_CHECK_HEADERS that works. Also, fix the
120764 nearby SYSV IPC check which was lacking an argument and giving bogus
120765 results (it's "AC_TRY_LINK(includes, main, yes, no)").
120766
120767commit eef16c36ad6e90fd8eaad4d8bdbc1205bc28a66f
120768Author: Eric Anholt <anholt@freebsd.org>
120769Date: Fri Dec 30 05:44:14 2005 +0000
120770
120771 Add #undef BSD44SOCKETS, without which the listener socket on FreeBSD would
120772 be created without the port number due to xtrans's define not being
120773 used when xtrans was compiled in the X Server.
120774
120775commit 7aa0ea23bc9b8df582fe06f2bc39dcfe34583c7e
120776Author: Eric Anholt <anholt@freebsd.org>
120777Date: Fri Dec 30 04:11:42 2005 +0000
120778
120779 Add an empty all-local target for FreeBSD make's sake, which doesn't deal
120780 with .PHONY.
120781
120782commit 49a9249239b0dd105b83a101db6e32549978f0d5
120783Author: Adam Jackson <ajax@nwnk.net>
120784Date: Thu Dec 29 21:11:41 2005 +0000
120785
120786 Make kdInputMachine static const, shrinks .data a bit.
120787
120788commit 6d7ee4167d9daeef9b793789a70aa724c4fe6bf4
120789Author: Adam Jackson <ajax@nwnk.net>
120790Date: Thu Dec 29 20:54:08 2005 +0000
120791
120792 Style fix, make SCREEN_EPILOGUE two arguments instead of three.
120793
120794commit b1efb3810cfea8116d76bae3ff3acfee521f4793
120795Author: Adam Jackson <ajax@nwnk.net>
120796Date: Thu Dec 29 20:42:54 2005 +0000
120797
120798 Missed one
120799
120800commit 858b4bc14a7adc7314ce5043c7ff8ca1891dd0c9
120801Author: Adam Jackson <ajax@nwnk.net>
120802Date: Thu Dec 29 20:38:29 2005 +0000
120803
120804 missed changelog entry:
120805 Enough build fixes to get {sdl,ephyr,fake} to link.
120806
120807commit 54922aeadb29df0a1819afd3f616131aa56e140a
120808Author: Adam Jackson <ajax@nwnk.net>
120809Date: Thu Dec 29 20:29:26 2005 +0000
120810
120811 Disable building LBX by default.
120812
120813commit fc69a2e729532ee062af6676fb187a89f7fffe8a
120814Author: Dave Airlie <airlied@linux.ie>
120815Date: Thu Dec 29 08:42:49 2005 +0000
120816
120817 Remove PowerMAX_OS define it never worked , it'll never work again,
120818 finishes running joke of OLS2005
120819
120820commit 25d3852b833bc0b61ce8313ce116251a2602b827
120821Author: Eric Anholt <anholt@freebsd.org>
120822Date: Thu Dec 29 00:44:20 2005 +0000
120823
120824 Undo spamming of libv's ChangeLog entry.
120825
120826commit d6646307f063b938c44d6193a2e8872e178aa90f
120827Author: Eric Anholt <anholt@freebsd.org>
120828Date: Thu Dec 29 00:19:33 2005 +0000
120829
120830 Add necessary *CONFIG_H declarations and a force-off of XF86* extensions in
120831 miinitext in the XGL case. Prevents mismatched structure sizes on my
120832 _XSERVER64 machine. At this point, with the uncommitted render/ diffs,
120833 Xglx starts up but displays badly.
120834
120835commit 3664c3ebf348d07ae3fe301fb8720adf32cf6d64
120836Author: Luc Verhaegen <libv@skynet.be>
120837Date: Wed Dec 28 15:22:21 2005 +0000
120838
120839 Bug #5153: standalone CVT modeline generator.
120840 - add hw/xfree86/utils/cvt/, cvt.c, cvt.man.pre and Makefile.am.
120841 - Adjust configure.ac and hw/xfree86/utils/Makefile.am for cvt.
120842 - Add MonPtr->reducedblanking and Option "ReducedBlanking" to the Monitor
120843 section.
120844 - Check for reduced blanking in xf86CheckModeForMonitor and disallow modes
120845 with less than 25% blanking otherwise.
120846 - Fix some warnings in hw/xfree86/common/xf86Config.c.
120847
120848commit 08708f7d616a7c0a596fb71dd7acd76d45257fec
120849Author: Eric Anholt <anholt@freebsd.org>
120850Date: Wed Dec 28 11:48:14 2005 +0000
120851
120852 Merge from xserver: Make the existing ChangePictureFilter and
120853 ChangePictureTransform screen hooks get called at appropriate times.
120854
120855commit 31d0fdde19598ce9375cc9638ad4e2c8b5af8d9a
120856Author: Adam Jackson <ajax@nwnk.net>
120857Date: Wed Dec 28 10:46:56 2005 +0000
120858
120859 Remove a debugging printf
120860
120861commit 64ac7401ad5022462279dff4dcfb12844c9857ae
120862Author: Eric Anholt <anholt@freebsd.org>
120863Date: Wed Dec 28 10:43:02 2005 +0000
120864
120865 Fix a copy'n'paste-o that would result in mis-rounding of the results of
120866 several composite operators in A and G channels.
120867
120868commit 9ceffb6b92e55f0d74d71489900d43940f14dfd9
120869Author: Eric Anholt <anholt@freebsd.org>
120870Date: Wed Dec 28 10:37:17 2005 +0000
120871
120872 Merge from xserver a header for shm's server internal functions, which are
120873 called by DDXes. Desired by XGL, and should be used in other locations,
120874 probably.
120875
120876commit b1b40ed6a87f72d07825624730cca193d5a89baf
120877Author: Eric Anholt <anholt@freebsd.org>
120878Date: Wed Dec 28 10:31:46 2005 +0000
120879
120880 Initial commit of XGL build infrastructure and XGL code changes for
120881 building within the xorg server tree. Requires additional, uncommitted
120882 dix changes to successfully build, and successful running is still yet
120883 to happen.
120884
120885commit 36061c75ae42aa733cde9b3fd05e0c8280b31655
120886Author: Eric Anholt <anholt@freebsd.org>
120887Date: Wed Dec 28 10:10:59 2005 +0000
120888
120889 Forced commit to note repocopy from xserver CVS as of a few minutes ago.
120890
120891commit 05c139d4cdfd11d39c0168d0c80ac1dbdd069b4c
120892Author: Adam Jackson <ajax@nwnk.net>
120893Date: Wed Dec 28 10:02:54 2005 +0000
120894
120895 Enough build fixes to get {sdl,ephyr,fake} to link.
120896
120897commit aeb770f645e2d591b255ec4ab06addcb1beafa5f
120898Author: Dave Airlie <airlied@linux.ie>
120899Date: Wed Dec 28 02:43:50 2005 +0000
120900
120901 recommit previous changes to evdev.c
120902
120903commit ce7c0c89375ec74f89ae5727998fd75fb768d280
120904Author: Dave Airlie <airlied@linux.ie>
120905Date: Wed Dec 28 01:57:11 2005 +0000
120906
120907 some updates for default colormap install
120908
120909commit 022aa1127c7dcd133ce73dbc12a10bfba8b1ed6e
120910Author: Adam Jackson <ajax@nwnk.net>
120911Date: Wed Dec 28 01:01:06 2005 +0000
120912
120913 Get Xsdl closer to linking.
120914
120915commit 27d79ab2bcebb634d0b69c851c72283a7514eb0c
120916Author: Adam Jackson <ajax@nwnk.net>
120917Date: Tue Dec 27 23:03:15 2005 +0000
120918
120919 s/XSERVER/KDRIVE/
120920
120921commit 72817714a0787536ce8e8ad0d5473dea0f1c1abe
120922Author: Adam Jackson <ajax@nwnk.net>
120923Date: Tue Dec 27 23:01:27 2005 +0000
120924
120925 Remove Imakefiles.
120926
120927commit 9dd0af6cb4e2c8976ada57a4f4ed16faae090a9d
120928Author: Adam Jackson <ajax@nwnk.net>
120929Date: Tue Dec 27 08:31:37 2005 +0000
120930
120931 Skeletal kdrive build system. Totall non-functional atm.
120932
120933commit 7fd73d2953cf9449c15462cf4bf67639db64f997
120934Author: Adam Jackson <ajax@nwnk.net>
120935Date: Tue Dec 27 08:29:50 2005 +0000
120936
120937 Build fixes: XSERVER_LIBS -> KDRIVE_LIBS, config.h -> kdrive-config.h
120938
120939commit 2f3ac6e5fcbd0e954a094fb6b975d7c8816c44b7
120940Author: Adam Jackson <ajax@nwnk.net>
120941Date: Tue Dec 27 08:26:03 2005 +0000
120942
120943 Start importing kdrive.
120944
120945commit 6798fd0170f4225ce4e69148978533fcee9bdc34
120946Author: Adam Jackson <ajax@nwnk.net>
120947Date: Mon Dec 26 19:13:52 2005 +0000
120948
120949 Bug #4190: Add a rule for 'make relink' since automake sucks.
120950
120951commit 8fc4ea8620913776a903ee2b4f22c306d5778623
120952Author: Adam Jackson <ajax@nwnk.net>
120953Date: Mon Dec 26 18:55:09 2005 +0000
120954
120955 Nuke unsupported NDBM routines. Shrink the hash table a bit, over
120956 25% of the buckets were going empty.
120957
120958commit ed33c7c98ad0c542e9e2dd6caa3f84879c21dd61
120959Author: Daniel Stone <daniel@fooishbar.org>
120960Date: Mon Dec 26 04:23:58 2005 +0000
120961
120962 Remove unused -xkbdb and -noloadxkb options. Rename -ar1 and -ar2 to
120963 -ardelay and -arinterval, respectively. Remove XKB banner from help text.
120964
120965commit 7e3cb9a09ac422179be89773f7fb14a462d25434
120966Author: Adam Jackson <ajax@nwnk.net>
120967Date: Sun Dec 25 22:25:15 2005 +0000
120968
120969 Remove unused layer module.
120970
120971commit 9b083369ded2258cbc8ac2058e06ec8a3b171178
120972Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
120973Date: Fri Dec 23 20:11:12 2005 +0000
120974
120975 Change list of X server man pages in "See Also" section to list the ones
120976 actually included and remove the ones that are no longer.
120977
120978commit 5fd978b1e7bce9169f87712a4a7c2c36a68ac00a
120979Author: Daniel Stone <daniel@fooishbar.org>
120980Date: Fri Dec 23 07:40:44 2005 +0000
120981
120982 Make LBX configuration default to auto.
120983
120984commit 5230e86b1cc841bfb35806618052aa835b7eb7e7
120985Author: Dave Airlie <airlied@linux.ie>
120986Date: Fri Dec 23 04:13:37 2005 +0000
120987
120988 fix up xglglx.c
120989
120990commit feb735c5bb0cd391136f1c73476703dff82dc9b0
120991Author: Dave Airlie <airlied@linux.ie>
120992Date: Fri Dec 23 02:07:58 2005 +0000
120993
120994 Well there were a couple of snapshots later than CVS available outside of
120995 Novell, so I've done a crazy merge to try and get them into a workable
120996 CVS, I suspect I may have failed.. there is a pre-xgldrop-merge tag if
120997 I did.
120998
120999commit ade104ce5a016623c1ce97b0d52b531185b35baf
121000Author: Dave Airlie <airlied@linux.ie>
121001Date: Fri Dec 23 01:51:40 2005 +0000
121002
121003 check drawable is available
121004
121005commit 7fb521e80d6e2c05e9475e74fbf80bfbe74cda95
121006Author: Dave Airlie <airlied@linux.ie>
121007Date: Fri Dec 23 01:50:04 2005 +0000
121008
121009 from davidr's tree update some fixes
121010
121011commit c88a3145d057ab72466a3ea8b789bf419e4efc33
121012Author: Dave Airlie <airlied@linux.ie>
121013Date: Fri Dec 23 01:49:21 2005 +0000
121014
121015 from davidr's tree if source picture defined return
121016
121017commit c59508566f11982aa3f4be383597d0e6178718c2
121018Author: Dave Airlie <airlied@linux.ie>
121019Date: Fri Dec 23 01:13:28 2005 +0000
121020
121021 fix glx up for newest glitz interface
121022
121023commit d822bc159672e7327054e572b659ae7dde040e83
121024Author: Dave Airlie <airlied@linux.ie>
121025Date: Fri Dec 23 00:08:35 2005 +0000
121026
121027 make xgl code at least build against latest glitz.
121028
121029commit 6e2086395d99081d8d682b90cec650a06e41fc2c
121030Author: Dave Airlie <airlied@linux.ie>
121031Date: Thu Dec 22 23:32:49 2005 +0000
121032
121033 fixups for newer glitz API
121034
121035commit f3ae42c0fd910b7f9feb9be91ccb056bce0cd999
121036Author: Dave Airlie <airlied@linux.ie>
121037Date: Thu Dec 22 23:31:15 2005 +0000
121038
121039 small fix towards new glitz interface
121040
121041commit 2af7e94eab6847159a3439301ecc93c62a12b1a0
121042Author: Eric Anholt <anholt@freebsd.org>
121043Date: Thu Dec 22 13:54:08 2005 +0000
121044
121045 Adjust the rules for auto-generating some source files, so that they work
121046 on both GNU make and FreeBSD's make.
121047
121048commit 0d7ec5c7d9b451066a079fe56bcc9722341a91ff
121049Author: Kevin E Martin <kem@kem.org>
121050Date: Wed Dec 21 02:30:08 2005 +0000
121051
121052 Update package version for X11R7 release.
121053
121054commit b37e738d5f4e1769bdee98acca788aeeb1556bcc
121055Author: Adam Jackson <ajax@nwnk.net>
121056Date: Tue Dec 20 21:40:19 2005 +0000
121057
121058 Fix an fb regression on A8 pictures. (Fredrik Höglund)
121059
121060commit 03d37eb03864cfc1a2f8d239d5a4c8341bf274f7
121061Author: Adam Jackson <ajax@nwnk.net>
121062Date: Tue Dec 20 21:34:21 2005 +0000
121063
121064 Bug #5359: Fix a segfault (Mark Kettenis)
121065
121066commit 7b89b643c12fa0f7a662b3ff76e05ece53101312
121067Author: Adam Jackson <ajax@nwnk.net>
121068Date: Mon Dec 19 16:44:21 2005 +0000
121069
121070 Bug #5116: Refer DRI section details to dri.fd.o.
121071
121072commit 3ef3add90351e3cb7b54dbcedc234bc5d3d65f1c
121073Author: Adam Jackson <ajax@nwnk.net>
121074Date: Mon Dec 19 16:34:07 2005 +0000
121075
121076 Stub COPYING files
121077
121078commit 3566307c8d44f89622ea51169f67c79092cb56d1
121079Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121080Date: Mon Dec 19 09:18:29 2005 +0000
121081
121082 Fix typos.
121083
121084commit 137447c5f3c6f1914ac869297f823ae93ce428ac
121085Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121086Date: Thu Dec 15 01:54:45 2005 +0000
121087
121088 Update to 2005-12-14 snapshot from pciids.sf.net
121089
121090commit 2cf86fce41e3fd2ac48c5088da11e19077e42e65
121091Author: Kevin E Martin <kem@kem.org>
121092Date: Thu Dec 15 00:20:27 2005 +0000
121093
121094 Update package version number for final X11R7 release candidate. Update
121095 release string to 6.99.99.904.
121096
121097commit f1ba3b4f33a928a3a59538799b3863de5c87e70e
121098Author: Adam Jackson <ajax@nwnk.net>
121099Date: Wed Dec 14 20:11:16 2005 +0000
121100
121101 Bug #4718: Command line flag to disable ACPI.
121102
121103commit cf605eb91619a8c0589a08674ffc3e018471b3fc
121104Author: Adam Jackson <ajax@nwnk.net>
121105Date: Tue Dec 13 17:35:26 2005 +0000
121106
121107 Build libglx correctly when not building the Xorg DDX.
121108
121109commit b076dd787ff71c4b385ab4e2e4eb367f3de378f6
121110Author: Adam Jackson <ajax@nwnk.net>
121111Date: Tue Dec 13 17:34:06 2005 +0000
121112
121113 Spell it XINERAMA_SRCS, not PANORAMIX_SRCS.
121114
121115commit 3666dbb5f3e06fa6a72def64556d64cf73141777
121116Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121117Date: Mon Dec 12 23:33:55 2005 +0000
121118
121119 Remove unnecessary include of dgaproc.h that broke Solaris builds.
121120
121121commit 438a5549f08ab03443d45dd46323579a2f2e4ba2
121122Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121123Date: Mon Dec 12 03:06:18 2005 +0000
121124
121125 Bugzilla #4715 <https://bugs.freedesktop.org/show_bug.cgi?id=4715> Files in
121126 xserver/xorg/Xext not included in tarball after make dist
121127
121128commit 62f3ef930adc7edd49b27dd1f7b0f51bc8bc0afa
121129Author: Adam Jackson <ajax@nwnk.net>
121130Date: Fri Dec 9 18:35:21 2005 +0000
121131
121132 Bug #5258: Restore binary compatibility with 6.8.2's PictureRec. (Aaron
121133 Plattner)
121134
121135commit b99dea9dcf99f907a3536c0db1c39cc67931a5b1
121136Author: Adam Jackson <ajax@nwnk.net>
121137Date: Fri Dec 9 18:32:46 2005 +0000
121138
121139 Bug #4935: Fix includes. (Eric Anholt)
121140
121141commit f4957ee94810b471110deebf03d7413399b45db3
121142Author: Adam Jackson <ajax@nwnk.net>
121143Date: Fri Dec 9 18:30:51 2005 +0000
121144
121145 Bug #4809: Re-fix that doesn't break distcheck. (Alan Coopersmith)
121146
121147commit de22d0c2264bd6dbacbbb4160d09c7e84ad37e70
121148Author: Adam Jackson <ajax@nwnk.net>
121149Date: Fri Dec 9 15:30:05 2005 +0000
121150
121151 Fix a thinko so the code matches the comment
121152
121153commit 6fcb049cd0d2291da5943176716d1f7bbb85fdc2
121154Author: Adam Jackson <ajax@nwnk.net>
121155Date: Fri Dec 9 06:49:39 2005 +0000
121156
121157 Bug #1288: Additional refactor of the driver probe logic to keep ati loaded
121158 before atimisc.
121159
121160commit 17ac5e9fec1e07bd18ae1407043c300cb4695ede
121161Author: Adam Jackson <ajax@nwnk.net>
121162Date: Fri Dec 9 05:36:41 2005 +0000
121163
121164 Push the fallback drivers to the end of the list so driver probe order
121165 stays useful.
121166
121167commit 80ea67e37980d07438749f1aa4dfdd7ee1086799
121168Author: Adam Jackson <ajax@nwnk.net>
121169Date: Fri Dec 9 03:59:41 2005 +0000
121170
121171 Bug #4361: Change driver probe logic to read the driver list from disk
121172 instead of using a compile-time array.
121173
121174commit 7fa2d11d85d43f42aa9c02f8d772c91d1b04df43
121175Author: Adam Jackson <ajax@nwnk.net>
121176Date: Fri Dec 9 03:57:41 2005 +0000
121177
121178 Bug #4361: Define XF86CONFIGFILE properly so config file generation works
121179
121180commit 26b41ff43959a07a778bc3d6e4db8da036f09de3
121181Author: Kevin E Martin <kem@kem.org>
121182Date: Fri Dec 9 03:02:21 2005 +0000
121183
121184 Fix sgml docs build.
121185
121186commit d6f98cbdb8fb74c504a92939d3741420eeed7110
121187Author: Adam Jackson <ajax@nwnk.net>
121188Date: Thu Dec 8 19:33:09 2005 +0000
121189
121190 Bug #3944: Fix 24bpp packed pixel. (Søren Sandmann Pedersen)
121191
121192commit f9ccebe8c5cd674c08fe8ed860d1c456e42c937e
121193Author: Adam Jackson <ajax@nwnk.net>
121194Date: Thu Dec 8 19:27:13 2005 +0000
121195
121196 Bug #4928: Fix compilation for Alpha. (Stefaan DeRoeck)
121197
121198commit 3a6bdf0715b994d6ecaa5b6e448695a8a8ec7d72
121199Author: Kevin E Martin <kem@kem.org>
121200Date: Thu Dec 8 19:21:12 2005 +0000
121201
121202 Add configure option to set the top level font dir.
121203
121204commit 008c2dd5e4614e6a21123ee3a2ac9c5d3bafa97a
121205Author: Kevin E Martin <kem@kem.org>
121206Date: Thu Dec 8 17:55:19 2005 +0000
121207
121208 Add configure options to allow hard-coded paths to be changed.
121209
121210commit 39189c2b86a4c2ab5f3f161d423eb072356668e5
121211Author: Kevin E Martin <kem@kem.org>
121212Date: Thu Dec 8 17:54:40 2005 +0000
121213
121214 Allow hard-coded paths to be configurable.
121215
121216commit 20c0ebe7b3feb85abf9bf140b7799aafc6f59513
121217Author: Kevin E Martin <kem@kem.org>
121218Date: Wed Dec 7 16:18:02 2005 +0000
121219
121220 Change to use the app-defaults default dir configured in libXt.
121221
121222commit 4a39354e14c3c360046b04ea0d4825832b05df05
121223Author: Kevin E Martin <kem@kem.org>
121224Date: Tue Dec 6 22:48:51 2005 +0000
121225
121226 Change *man_SOURCES ==> *man_PRE to fix autotools warnings.
121227
121228commit 84faf8dc9747bc4f1db5ebc2f23e17cf1460e2e9
121229Author: Adam Jackson <ajax@nwnk.net>
121230Date: Tue Dec 6 16:22:47 2005 +0000
121231
121232 Bug #5230: Fix whitespace bugs.
121233
121234commit 9439297b7bc07dcb90f0d01da09eea1bac3d42ff
121235Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121236Date: Tue Dec 6 15:50:35 2005 +0000
121237
121238 Bugzilla #5219 <https://bugs.freedesktop.org/show_bug.cgi?id=5219> Make
121239 sure all optional sources are included in EXTRA_DIST, even if they
121240 aren't used on the platform the distballs are made on.
121241
121242commit f259fd680caccb59546d7788704e46e51a9c6146
121243Author: Adam Jackson <ajax@nwnk.net>
121244Date: Sat Dec 3 22:47:47 2005 +0000
121245
121246 Disable the xf8_32wid logic for now, breaks distcheck
121247
121248commit 26f9c4305660c2b3dc7fe8d214bcdd3c24e1b198
121249Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121250Date: Sat Dec 3 17:04:45 2005 +0000
121251
121252 Bugzilla #4809 <https://bugs.freedesktop.org/show_bug.cgi?id=4809> Patch
121253 #3908 <https://bugs.freedesktop.org/attachment.cgi?id=3908> xf8_32wid
121254 and cfb24 only need to be built on sparc
121255
121256commit 98231c6b38c98976f4ac2b9417ecfbc37a8cbe9a
121257Author: Kevin E Martin <kem@kem.org>
121258Date: Sat Dec 3 05:47:25 2005 +0000
121259
121260 Update package version number for X11R7 RC3 release. Update release string
121261 to 6.99.99.903 (i.e., 7.0 RC3).
121262
121263commit 7c00afd0ec94e491f1a9ef32d6543ed51ea3319d
121264Author: Kevin E Martin <kem@kem.org>
121265Date: Fri Dec 2 06:02:45 2005 +0000
121266
121267 Define XFree86Server only where it is required.
121268
121269commit 924518605b613eb66aa569877fa9f131e6f2a2fd
121270Author: Kevin E Martin <kem@kem.org>
121271Date: Thu Dec 1 23:39:00 2005 +0000
121272
121273 Fix GL build when srcdir != builddir (Donnie Berkholz).
121274
121275commit 14b9315379fe8c783013906616d868f93fd51c83
121276Author: Kevin E Martin <kem@kem.org>
121277Date: Thu Dec 1 22:06:49 2005 +0000
121278
121279 Add missing XvExtension and XvMCExtension defines.
121280
121281commit df8fa21d3189e20260328b88cc8a86224a9b1ebf
121282Author: Kevin E Martin <kem@kem.org>
121283Date: Thu Dec 1 16:20:09 2005 +0000
121284
121285 Fix typo: xorg_bus_linuxbsdpci ==> xorg_bus_linuxpci
121286
121287commit ccfaf82367c9d057fd8314ce36b47f0a8eb696b6
121288Author: Eric Anholt <anholt@freebsd.org>
121289Date: Thu Dec 1 05:04:07 2005 +0000
121290
121291 Bug #5160: Fix the modular build to try to use the same logic for choosing
121292 the architecture/os-specific bus support as monolithic.
121293
121294commit 9c0bd9687fe7d20f2f0793332ae0db06f035eb23
121295Author: Adam Jackson <ajax@nwnk.net>
121296Date: Wed Nov 30 22:59:22 2005 +0000
121297
121298 Import libdrm 2.0
121299
121300commit 4ec0b623b6ab5f8a1e5af2cc3d839251acf81ce2
121301Author: Adam Jackson <ajax@nwnk.net>
121302Date: Wed Nov 30 02:36:25 2005 +0000
121303
121304 Bug #5093: Fix fb for non-SSE machines. (Xavier Bachelot)
121305
121306commit ed826d563cba82c516fd41f6a29ee50aa1fe6c6a
121307Author: Adam Jackson <ajax@nwnk.net>
121308Date: Tue Nov 29 23:34:30 2005 +0000
121309
121310 Only build dlloader modules by default.
121311
121312commit da5d66f2ff27b21fe5c39a4abb4f627edd707f1d
121313Author: Kevin E Martin <kem@kem.org>
121314Date: Tue Nov 29 16:39:33 2005 +0000
121315
121316 Fix usage of XFree86LOADER/XFree86Module/IN_MODULE and update loadable
121317 module builds to reflect this change.
121318
121319commit 51a721a6dbb42702347aad3115147e4922fc1a25
121320Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121321Date: Mon Nov 28 22:05:09 2005 +0000
121322
121323 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4
121324 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir)
121325
121326commit 381931b15b15d0a2ec384b0c22864412c44f9c6e
121327Author: Kevin E Martin <kem@kem.org>
121328Date: Wed Nov 23 07:14:46 2005 +0000
121329
121330 Add configure option to install libxf86config.a (disabled by default).
121331
121332commit 594ca0966e8fd5992ebf95170cc42e19c698fec6
121333Author: Eric Anholt <anholt@freebsd.org>
121334Date: Tue Nov 22 02:11:00 2005 +0000
121335
121336 Bug #5118: Use "rm -f" instead of "$(RM)", which isn't always defined.
121337
121338commit 1c8c1179c0789e3e134d31a62dbb88bfdb594b26
121339Author: Felix Kuehling <fxkuehl@gmx.de>
121340Date: Mon Nov 21 04:24:07 2005 +0000
121341
121342 Fix Xprt library dependencies in the case that Xprint is auto-detected by
121343 configure.
121344
121345commit a1f110bda80bb3b8e4f602385ca5ccd96cf3f786
121346Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121347Date: Sun Nov 20 23:01:02 2005 +0000
121348
121349 Make sure XKM_OUTPUT_DIR (used in code) ends in / (so paths don't get hosed
121350 when appending file names) but XKB_COMPILED_DIR (used in Makefiles)
121351 does not so install-sh -d doesn't get confused when the directory
121352 already exists.
121353
121354commit 385730d23944c24dd9af45b27f62c1161abc48b2
121355Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121356Date: Sun Nov 20 04:15:15 2005 +0000
121357
121358 Add xext to list of modules xorgcfg depends on.
121359
121360commit 63aa96c08a8390621b017ea498c88cf88152024b
121361Author: Kevin E Martin <kem@kem.org>
121362Date: Sat Nov 19 07:15:50 2005 +0000
121363
121364 Update pkgconfig files to separate library build-time dependencies from
121365 application build-time dependencies, and update package deps to work
121366 with separate build roots.
121367
121368commit d3b6653a2892e8c929c79fe3ace19ac9d8366fc4
121369Author: Adam Jackson <ajax@nwnk.net>
121370Date: Sat Nov 19 03:53:04 2005 +0000
121371
121372 Bug #4824: Build XTrap support by default, matching monolith.
121373
121374commit 627ac1fe1dbcbc070575da7bee9e686a7dce5262
121375Author: Eric Anholt <anholt@freebsd.org>
121376Date: Fri Nov 18 23:34:04 2005 +0000
121377
121378 Bug #5060: Fix non-Linux DRI on 64 bit post Linux 32/64 changes.
121379
121380commit e3ec048ff2fe0ee0862472e9b147b7ce488ea898
121381Author: Adam Jackson <ajax@nwnk.net>
121382Date: Fri Nov 18 22:43:50 2005 +0000
121383
121384 Bug #4928: Unbreak Makefile.am for Alpha chips. (Stefan DeRoeck)
121385
121386commit de95d8ee197a0bb738037195997d754a20e10254
121387Author: Adam Jackson <ajax@nwnk.net>
121388Date: Fri Nov 18 18:02:24 2005 +0000
121389
121390 Bug #4859: Don't segfault on bad DDC read. (Tony Houghton)
121391
121392commit 21f7d03dbc347f6bf97a40671275ac75df15bd10
121393Author: Adam Jackson <ajax@nwnk.net>
121394Date: Wed Nov 16 07:28:19 2005 +0000
121395
121396 Fix builds when not building the Xorg DDX.
121397
121398commit fb2d9df869af0c96f1488ef7cf364e01a9d28f3f
121399Author: Adam Jackson <ajax@nwnk.net>
121400Date: Tue Nov 15 00:29:23 2005 +0000
121401
121402 Make fb build on darwin/ppc without addition #define hacks
121403
121404commit 16b315affa30e34b9bab81778978484137a5d9bb
121405Author: Kevin E Martin <kem@kem.org>
121406Date: Mon Nov 14 21:04:12 2005 +0000
121407
121408 Use glcontextmodes.[ch] from Mesa.
121409
121410commit 0c110c80e7afbef50bb354cf1df30123ed048250
121411Author: Kevin E Martin <kem@kem.org>
121412Date: Mon Nov 14 20:18:03 2005 +0000
121413
121414 Add GL_CFLAGS so that GLX can find its proto headers when using separate
121415 build dirs.
121416
121417commit fc81c13e4dafb0eb818879454ee7ae3fa3dae6d0
121418Author: Kean Johnson <kean@armory.com>
121419Date: Mon Nov 14 18:49:30 2005 +0000
121420
121421 Dont prevent SCO platforms for using the default ZAxisMapping now that the
121422 OS layer correctly sends z-axis events when the wheel button is used.
121423
121424commit bd9fb533b31c2427d854199fa59dccd357cf874b
121425Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121426Date: Mon Nov 14 00:01:34 2005 +0000
121427
121428 Default xkb-output directory needs trailing slash.
121429
121430commit 267cbffa41fffff69c692911d128462f5bab2a69
121431Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121432Date: Sun Nov 13 20:53:24 2005 +0000
121433
121434 Bug #5019 <https://bugs.freedesktop.org/show_bug.cgi?id=5019> xserver
121435 installs manpages into 'man1' instead of 'man1x'
121436
121437commit 3179d29b8212c197634d81fbeb8dd2e8df995735
121438Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121439Date: Sat Nov 12 18:03:34 2005 +0000
121440
121441 use RGB_DB not RGB_PATH as that's what configure defines (Jürg Billeter
121442 <j@bitron.ch>)
121443
121444commit 0e7e4c7064df64c29b1a0ccd84fba1be7c748f18
121445Author: Kevin E Martin <kem@kem.org>
121446Date: Thu Nov 10 04:59:21 2005 +0000
121447
121448 Fix typo to enable DGA support.
121449 EXTMODULE is required to build DGA support into extmod.
121450
121451commit e4554db8f87c6a39a3087186395972000bd2085c
121452Author: Kean Johnson <kean@armory.com>
121453Date: Thu Nov 10 02:41:20 2005 +0000
121454
121455 Dont pass wheel mouse buttons as real buttons, map them as Z-Axis movement
121456 on SCO and USL. Re-instate the ZAxisMapping default for the mouse
121457 driver.
121458
121459commit 1b26fe6d2092c202141a0371f47ef1cd7c66ec00
121460Author: Kevin E Martin <kem@kem.org>
121461Date: Wed Nov 9 21:28:54 2005 +0000
121462
121463 Update package version number for X11R7 RC2 release. Update release string
121464 to 6.99.99.902 (i.e., 7.0 RC2).
121465
121466commit f886e632b8dab1bfa0de42b9759a8284ecd9b94f
121467Author: Matthias Hopf <mhopf@suse.de>
121468Date: Wed Nov 9 17:05:41 2005 +0000
121469
121470 Bug #4915: ButtonMapping option which allows to define arbitrary button
121471 mappings (including left-handed mouse etc.). Fixed incorrect usage of
121472 non-reversed, but ZAxisMapped buttons for state detection. Nuked unused
121473 part of reverseMap.
121474
121475commit a25871ae52dd5ce094ba8c1b2021dd027d3e71bd
121476Author: Kevin E Martin <kem@kem.org>
121477Date: Wed Nov 9 01:00:46 2005 +0000
121478
121479 DRM 1.0.5 import
121480
121481commit c9709c0a38af46368726857f7261cbeb84e53911
121482Author: Kevin E Martin <kem@kem.org>
121483Date: Tue Nov 8 22:47:57 2005 +0000
121484
121485 Add newly checked in files to Xorg server build.
121486 Fix release date. Enable DGA extension by default.
121487
121488commit f8430a1b8651f4b52d9d3b54694a60d929b48925
121489Author: Kristian Høgsberg <krh@redhat.com>
121490Date: Tue Nov 8 19:04:56 2005 +0000
121491
121492 Bug #2880, add functions for byte and word level access to pci config
121493 space.
121494 Fix broken utf8 again.
121495
121496commit 5390c7ab05d23f64e6d9afaa558be246a6d6e1b4
121497Author: Kean Johnson <kean@armory.com>
121498Date: Tue Nov 8 06:33:30 2005 +0000
121499
121500 See ChangeLog entry 2005-11-07 for details.
121501
121502commit f5814bf3fff5352ed6edef4c58aadf2d4593f094
121503Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121504Date: Tue Nov 8 03:12:43 2005 +0000
121505
121506 Don't reference noXkbExtension when building without XKB. (Bob Terek - Sun
121507 Microsystems)
121508
121509commit e73cdba865f36ebf78c2dc4ff674b4d9bfe85013
121510Author: Kevin E Martin <kem@kem.org>
121511Date: Mon Nov 7 21:03:49 2005 +0000
121512
121513 Fix Xvfb to work properly in depth 15 mode. Fixes XTS5 XCloseDisplay-3
121514 server crash.
121515
121516commit 890ed0e082e048fa8daf48229b40558381bd131d
121517Author: Thomas Hellstrom <unichrome@shipmail.org>
121518Date: Sun Nov 6 16:40:59 2005 +0000
121519
121520 Fix a bug where a system memory pixmap got a wrong address if memcpy()
121521 fallback was used for downloading from screen.
121522
121523commit 70aedcf32a0c924fd073f5b36d20813e8323026b
121524Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121525Date: Sat Nov 5 18:56:50 2005 +0000
121526
121527 Bug #4948: <https://bugs.freedesktop.org/show_bug.cgi?id=4948> Incorrect
121528 URL in log file for Xorg CVS. Also fixed wording of statement to not
121529 refer to monolithic CVS since modular uses the same code, so it was
121530 appearing in modular builds too.
121531
121532commit 89c661d61f1b9c70a08237476fa1f7f42c1783ab
121533Author: Kevin E Martin <kem@kem.org>
121534Date: Fri Nov 4 21:37:32 2005 +0000
121535
121536 Only use fbCopyAreammx if planemask is FB_ALLONES (fixes XTS5 XCopyArea
121537 tests 22 and 23).
121538
121539commit 0b150a05e6fadca7ee8240697d6cbeadea0c53b3
121540Author: Ian Romanick <idr@us.ibm.com>
121541Date: Thu Nov 3 17:12:53 2005 +0000
121542
121543 Whitespace change just to make sure I created the branch correctly.
121544
121545commit 90cf8e339b71c2f8f2d7a362e6e1ca8078d7f4fd
121546Author: Kevin E Martin <kem@kem.org>
121547Date: Thu Nov 3 17:08:06 2005 +0000
121548
121549 Fix vendor string and release version reported by the servers. Enable
121550 security, lbx and xevie extensions to give us parity with monolithic
121551 tree.
121552
121553commit f23defeef285b4a5bb58405589294bd557c9bb01
121554Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121555Date: Thu Nov 3 16:57:01 2005 +0000
121556
121557 Use APP_MAN_SUFFIX for Xserver man page instead of hardcoding section 1
121558
121559commit f5daec674aeb4fe6ccbc95ead8a319bbeb368d9f
121560Author: Kevin E Martin <kem@kem.org>
121561Date: Wed Nov 2 15:56:40 2005 +0000
121562
121563 Add support for enabling/disabling DBE (part of generic enable/disable
121564 extension support in the server).
121565
121566commit a311bfa73afa1af76f81958d23bc8e0c631d6828
121567Author: Kevin E Martin <kem@kem.org>
121568Date: Wed Nov 2 15:53:57 2005 +0000
121569
121570 Fix support for enabling/disabling extensions loaded from modules.
121571
121572commit 462a2407d540eac831c9be4dcee8a16aa1cea6ac
121573Author: Kevin E Martin <kem@kem.org>
121574Date: Tue Nov 1 15:01:51 2005 +0000
121575
121576 Add xorg-server.m4 for driver dependency checking.
121577 Update pkgcheck depedencies to work with separate build roots.
121578
121579commit 56101c9d6ec3585a0a8550da4b83dd399e3bcce6
121580Author: Kevin E Martin <kem@kem.org>
121581Date: Mon Oct 31 05:45:40 2005 +0000
121582
121583 Fix fd leak by closing them in the ACPI code instead of just using
121584 shutdown.
121585
121586commit 7993486e80711bd6f6f5b6c2b1f2ac32bfba735b
121587Author: Thomas Winischhofer <thomas@winischhofer.net>
121588Date: Sun Oct 30 17:38:49 2005 +0000
121589
121590 RandR: Add a driver func to let the driver determine the physical size of a
121591 screen size (display mode). Useful for faked widescreen modes, modes
121592 which are scaled by the driver, etc. This really helps fixing RandR's
121593 sometimes dumb DPI assumptions.
121594
121595commit c818d3a1a5439c54fc687927a99d69712602ed5e
121596Author: Thomas Winischhofer <thomas@winischhofer.net>
121597Date: Sun Oct 30 09:27:06 2005 +0000
121598
121599 Add xf86RandRSetNewVirtualAndDimensions to loader symlist
121600
121601commit fdbb3ea60949a12eb2f4805d16e8acc2348e39c7
121602Author: Thomas Winischhofer <thomas@winischhofer.net>
121603Date: Sat Oct 29 21:31:23 2005 +0000
121604
121605 Add function for drivers to change RandR's idea of the virtual screen size.
121606 (This allows drivers to reserve a larger virtual size at start and
121607 change it later)
121608
121609commit e921eec1c6d6ce32630977bd876c529a7c694459
121610Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121611Date: Sat Oct 29 00:12:33 2005 +0000
121612
121613 Make X -> Xorg symlink at install time.
121614
121615commit f842c229d4c4dbd5c01364f9e99709bedfd32be6
121616Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
121617Date: Fri Oct 28 16:01:17 2005 +0000
121618
121619 build fix on alpha
121620
121621commit 7416fd61a17a70a2c27c4b1d19796955c296dc7a
121622Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121623Date: Thu Oct 27 21:03:27 2005 +0000
121624
121625 Improved stack trace dump code for Solaris - try fork & exec of pstack
121626 first so we can see the names of non-exported symbols that aren't
121627 visible to walkcontext/dladdr1 code.
121628
121629commit b588bdfe2ac3758d7188706078d79fa276a303e3
121630Author: Dave Airlie <airlied@linux.ie>
121631Date: Sat Oct 22 04:38:50 2005 +0000
121632
121633 programs/Xserver/GL/mesa/X/xf86glx.c: Missing initializer in xf86glx.c
121634 spotted while debugging something else.
121635
121636commit 59279da49806b032027bb54410bc2513d21e3d9e
121637Author: Adam Jackson <ajax@nwnk.net>
121638Date: Fri Oct 21 19:06:13 2005 +0000
121639
121640 Bug #1429: Report input device type correctly. (Stéphane VOLTZ)
121641
121642commit 4a8072011895e6f472e429af7503fc07e0561144
121643Author: Adam Jackson <ajax@nwnk.net>
121644Date: Fri Oct 21 18:50:09 2005 +0000
121645
121646 Bug #4730: Byte-swap the pixmap ID correctly. (Neil Campbell)
121647
121648commit 81e913d3106066de73792f59f3e50e2b5458c567
121649Author: Adam Jackson <ajax@nwnk.net>
121650Date: Fri Oct 21 18:23:33 2005 +0000
121651
121652 Bug #4840: Typo, x$xRES -> x$RES. (George Fufutos)
121653
121654commit f5a58178347878e0409b592330a07867bea02bef
121655Author: Ian Romanick <idr@us.ibm.com>
121656Date: Thu Oct 20 23:24:47 2005 +0000
121657
121658 Make sure that the __gl*_size prototypes are seen in all the places that
121659 they need to be seen.
121660
121661commit 279cf9f79da5778b6e14ecc437379d73e3bec5b0
121662Author: Donnie Berkholz <spyderous@gentoo.org>
121663Date: Thu Oct 20 22:41:28 2005 +0000
121664
121665 Bug #4817 <https://bugs.freedesktop.org/show_bug.cgi?id=4817> Restore '='
121666 to '==' in test for mmx_capable.
121667
121668commit 1f43d218cc24358a0379535ed517c23011633c31
121669Author: Thomas Winischhofer <thomas@winischhofer.net>
121670Date: Thu Oct 20 21:45:40 2005 +0000
121671
121672 EXA: The "optimization" for using a fill operation instead of 1x1 copies
121673 checked the destination drawable's dimensions (!) instead of the
121674 tile's. Really....
121675
121676commit 15f56b203dbc14ea59885d40fd4bed3da9e8e190
121677Author: Adam Jackson <ajax@nwnk.net>
121678Date: Thu Oct 20 18:52:51 2005 +0000
121679
121680 Move xf86XTrapModule.c to dixmods, guess at a build system.
121681
121682commit da43c778f4a831061ad2c8b8a312b7a54c9cd79e
121683Author: Adam Jackson <ajax@nwnk.net>
121684Date: Wed Oct 19 22:45:54 2005 +0000
121685
121686 Bug #3224: Degrade XKB fallback message to X_WARNING.
121687
121688commit 4ebd26f04b32f1b09e0759f1a83437d0b1c4d646
121689Author: Adam Jackson <ajax@nwnk.net>
121690Date: Wed Oct 19 22:36:22 2005 +0000
121691
121692 Bug #3196: Fix Load foo.so syntax.
121693
121694commit af211a9bc1bcab0aa631558e5d6ce013095f9802
121695Author: Adam Jackson <ajax@nwnk.net>
121696Date: Wed Oct 19 22:30:09 2005 +0000
121697
121698 Fix PCI bus scan on ia64 E8870 chipsets.
121699
121700commit 5744308e2957781449bfe6fee9b465617a88384d
121701Author: Kevin E Martin <kem@kem.org>
121702Date: Tue Oct 18 22:06:54 2005 +0000
121703
121704 Update package version number for RC1 release. Update release string to
121705 6.99.99.901 (i.e., 7.0 RC1).
121706
121707commit 2769c3e72c470b472dae013e256a7ee73c3e53f2
121708Author: Adam Jackson <ajax@nwnk.net>
121709Date: Tue Oct 18 19:43:48 2005 +0000
121710
121711 Fix distcheck by forcing Xorg to be installed before chmod/chown.
121712
121713commit dd0d010e9c34278f968be486a6c5c91e021b6609
121714Author: Adam Jackson <ajax@nwnk.net>
121715Date: Tue Oct 18 19:14:08 2005 +0000
121716
121717 Fix parallel builds my ensuring libdmxconfig builds first.
121718
121719commit 79e6ac79f983b6cbd88a868dfd2235d9cbe75e8b
121720Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121721Date: Tue Oct 18 07:18:21 2005 +0000
121722
121723 Don't use $< in explicit rules since neither BSD nor Solaris make allow
121724 that.
121725
121726commit fb282ef43a1936dcdefa57f16a8363b2adaf983b
121727Author: Aaron Plattner <aplattner@nvidia.com>
121728Date: Tue Oct 18 04:03:01 2005 +0000
121729
121730 Add miext/damage so misym.c can export DamageDamageRegion.
121731
121732commit 959db6028d232dc76396cb658aa48d3b4e605aed
121733Author: Aaron Plattner <aplattner@nvidia.com>
121734Date: Tue Oct 18 04:02:31 2005 +0000
121735
121736 Export DamageDamageRegion. Not bumping the ABI version since we did that
121737 already for this release.
121738
121739commit b61c828b0455ec1d4a7ffb54b5ac9b65764a458b
121740Author: Kevin E Martin <kem@kem.org>
121741Date: Tue Oct 18 02:23:58 2005 +0000
121742
121743 DRM 20051017 import
121744
121745commit d6a40bcd4a745b5d6d1070deb696b21d128ca0fe
121746Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121747Date: Tue Oct 18 00:32:55 2005 +0000
121748
121749 Change default install dir for app-default files from
121750 $(sysconfdir)/X11/app-defaults to $(libdir)/X11/app-defaults to match
121751 the monolith & allow localization
121752
121753commit e7007f7d51c9e1d39118865fefb1716c579a70bd
121754Author: Adam Jackson <ajax@nwnk.net>
121755Date: Mon Oct 17 22:42:03 2005 +0000
121756
121757 More automake 1.7 braindamage: use mkdir -p, not .
121758
121759commit 151ba8b67fd88a721f9f72d3019212b22f5cd3e2
121760Author: Adam Jackson <ajax@nwnk.net>
121761Date: Mon Oct 17 22:25:58 2005 +0000
121762
121763 Work around automake-1.7 braindamage by providing an explicit rule for
121764 XOrgCfg.
121765
121766commit eec3df1503e561aff6656e15c73b25a0bba1b06b
121767Author: Kristian Høgsberg <krh@redhat.com>
121768Date: Mon Oct 17 17:11:12 2005 +0000
121769
121770 Fix whitespace in AS_HELP_STRING uses, convert all help strings to use
121771 AS_HELP_STRING.
121772
121773commit 1859c62607d567aa05334df1662f7249c983f793
121774Author: Kevin E Martin <kem@kem.org>
121775Date: Mon Oct 17 07:18:59 2005 +0000
121776
121777 include/dix-config.h.in Add support for more extensions
121778 Add missing files to EXTRA_DIST
121779
121780commit ccfe9e7e9b49cbbf7c50fbf1a5c33178f27f79eb
121781Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121782Date: Sun Oct 16 21:57:34 2005 +0000
121783
121784 Link Xprint config directories in $(C_LOCALE) list to C locale dir, not
121785 en_US
121786
121787commit a7d6a4fb321415b8aaad72760ff8a1ca3fd077f9
121788Author: Donnie Berkholz <spyderous@gentoo.org>
121789Date: Sun Oct 16 03:02:53 2005 +0000
121790
121791 Change '==' to portable '='.
121792
121793commit c2e461c7e970830ea430de3e5f352d144e9f0239
121794Author: Kevin E Martin <kem@kem.org>
121795Date: Sat Oct 15 20:44:44 2005 +0000
121796
121797 Fix typo and add new Makefiles to AC_OUTPUT
121798 Fix typos
121799 Add xorg.conf.man to CLEANFILES
121800 Add missing files to EXTRA_DIST
121801
121802commit 34b7b57b3c80507f63a542c6adb4b5c8ed80b642
121803Author: Kristian Høgsberg <krh@redhat.com>
121804Date: Sat Oct 15 19:34:28 2005 +0000
121805
121806 Doh, remember to add this file.
121807
121808commit 744aa34ca5228ea176cc56a7bdd48bbf5f29b0b5
121809Author: Eric Anholt <anholt@freebsd.org>
121810Date: Sat Oct 15 02:19:09 2005 +0000
121811
121812 Add an additional meaning to the "dirty" flag. Now, if !dirty && !area, the
121813 pixmaps's contents are undefined, so we won't need to upload the
121814 undefined contents in MoveIn. Use the ExaCheck* for async ops as well,
121815 so that dirty is always tracked. While the performance impact for my ls
121816 -lR test was not significant (though the avoiding-upload path was being
121817 hit), it's likely to be important for the upcoming Get/PutImage
121818 acceleration from ajax.
121819
121820commit 21e7339c1eead1148eea462bc99cf8faf02c8d39
121821Author: Kristian Høgsberg <krh@redhat.com>
121822Date: Fri Oct 14 22:44:56 2005 +0000
121823
121824 Hook up lbx.
121825
121826commit d62943c040fd3d45079c9918c57f74f993b585d4
121827Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121828Date: Fri Oct 14 22:19:51 2005 +0000
121829
121830 Set default font path to match the default in the monolith so fonts are
121831 actually found.
121832
121833commit 0ee70f53ef9b05052ee079560df107d05a9c5407
121834Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121835Date: Fri Oct 14 22:01:46 2005 +0000
121836
121837 Install Xorg & xorg.conf man pages even when not building docs
121838
121839commit 0676a2874a62a3661a718cdf21e75ffc77197ad9
121840Author: Kristian Høgsberg <krh@redhat.com>
121841Date: Fri Oct 14 20:01:36 2005 +0000
121842
121843 Add sysv and sco os-support subdirs and add simple EXTRA_DIST Makefile.am
121844 in those dirs. Remove unsupported os-support subdirs (bsdi, dgux, hurd,
121845 nto, os2, pmax, qnx4) that have no maintainer and we don't dist.
121846 Add Options.
121847
121848commit 57abb5b171b2fe88252aeb788463e533106d66b9
121849Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
121850Date: Fri Oct 14 08:29:16 2005 +0000
121851
121852 remove reference to non-existent agpgart.h
121853
121854commit 7e3e9ed97ba25bb84286f97fe6882a37c9aa7e25
121855Author: Donnie Berkholz <spyderous@gentoo.org>
121856Date: Fri Oct 14 06:10:06 2005 +0000
121857
121858 Add XTRAP_LIB to XPRINT_EXTENSIONS to fix xprint build when xtrap is
121859 enabled.
121860
121861commit 821584fcd3bf83f3aaacd35e54323f71d976db44
121862Author: Donnie Berkholz <spyderous@gentoo.org>
121863Date: Fri Oct 14 05:36:39 2005 +0000
121864
121865 Require glproto >= 1.4.1 if building DRI with GLX. This fixes a build
121866 failure on a number of hyperpipe functions.
121867
121868commit 8df7628a2ad93edf8271f13e0b43c0fa8f766668
121869Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121870Date: Fri Oct 14 00:41:51 2005 +0000
121871
121872 Remove reference to XF86config-4. Add xorgcfg(1) to See Also list.
121873
121874commit b54c8154ca19edce00b9c6379d5daf94268bade1
121875Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121876Date: Fri Oct 14 00:34:49 2005 +0000
121877
121878 Set substitutions needed in xorgconfig man page.
121879
121880commit 1df705e465a103c94ffbb9fe97bdbe6b0aefc746
121881Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121882Date: Thu Oct 13 20:30:38 2005 +0000
121883
121884 AC_SUBST VENDOR_STRING & VENDOR_RELEASE for xorgcfg's app-defaults file
121885
121886commit b349b20d783252d5126451142419aae554f9b776
121887Author: Kristian Høgsberg <krh@redhat.com>
121888Date: Thu Oct 13 18:08:24 2005 +0000
121889
121890 Dist NOTES.
121891 Dist helper shell scripts.
121892 Dist XAA.HOWTO and a few more unused C files.
121893 Dist xorgconf.cpp.
121894 Fix DIST_EXTRA typo.
121895 Clean yacc and lex generated files only during make maintainer-clean as we
121896 don't expect users to have those tools installed.
121897
121898commit 35a767590e481b15ae66dccc2dd91098992b2751
121899Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
121900Date: Thu Oct 13 01:13:58 2005 +0000
121901
121902 Fix stupid mistake in yesterday's allocator commit, would cause exa to
121903 consider a random available memory size
121904
121905commit a16dabd05ee7ec97877f07bd40ed83c01e72fc22
121906Author: Eric Anholt <anholt@freebsd.org>
121907Date: Wed Oct 12 11:15:44 2005 +0000
121908
121909 Remove an RM line that appears unnecessary and was breaking the build at
121910 xf86DefModeSet.c with FreeBSD make, where RM was undefined. While here,
121911 make the build of xf86DefModeSet.c depend on its sources, so it'll
121912 rebuild properly, and make it a normal CLEANFILE rather than a
121913 DISTCLEANFILE, since the intention seems to be to build it at the
121914 user's build time.
121915
121916commit b819c8378fbf29f185332e8435a80eb35991cd1f
121917Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
121918Date: Wed Oct 12 08:22:31 2005 +0000
121919
121920 remove unneeded line of code
121921
121922commit e573b272bf2b06fb62d0306ddc966f3230ead967
121923Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
121924Date: Wed Oct 12 07:46:36 2005 +0000
121925
121926 Use proper access size when reading pixel based on bpp of the source pixmap
121927
121928commit 55efb41f6cc064763cbfd3ee2a1239dc46cb109a
121929Author: Eric Anholt <anholt@freebsd.org>
121930Date: Wed Oct 12 07:35:20 2005 +0000
121931
121932 If a window background is a 1x1 pixmap, read the value out and go to
121933 exaFillRegionSolid rather than sending piles and piles of Copies to the
121934 hardware.
121935
121936commit fce11fdf03acc1f3f1dafb79fc8fff0251cf5473
121937Author: Kevin E Martin <kem@kem.org>
121938Date: Wed Oct 12 02:11:06 2005 +0000
121939
121940 Fix typo (DIST_EXTRA -> EXTRA_DIST)
121941
121942commit 12994b9afbc18bfb7209f677abf673415c9ddf15
121943Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
121944Date: Tue Oct 11 23:11:37 2005 +0000
121945
121946 Fix a couple of bugs in the offscreen allocator. One mostly harmless was
121947 causing our search loop for evictable blocks to possibly skip a good
121948 candiate, and another was the allocator would occasionally use
121949 area->offset as if it was the base of the pixmap, while for a pixmap
121950 that is not in available state, it is not. This caused some funny
121951 miscalculation leading to overlapping pixmaps and accesses beyond the
121952 end of the framebuffer. To make things cleared, I renamed save_offset
121953 to base_offset, made sure it's the one used everywhere in the
121954 allocator, and only align "offset" for the client at the end of
121955 exaOffscreenAlloc().
121956
121957commit 8444a1f3918b0433f89cae31673ab63628b4543d
121958Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
121959Date: Tue Oct 11 21:01:04 2005 +0000
121960
121961 missed commit
121962
121963commit 3b683b63eed603ae58a8cddab48eb81f7ba0dbdf
121964Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
121965Date: Tue Oct 11 20:12:24 2005 +0000
121966
121967 missed ChangeLog entry for previous commit
121968
121969commit b4450f3242ab408e80bc3d6d5d1cf6765f3e5339
121970Author: Thomas Winischhofer <thomas@winischhofer.net>
121971Date: Tue Oct 11 19:03:02 2005 +0000
121972
121973 Add DGAReInitModes, Part 2
121974
121975commit d91d18e1d6d663244288748ab86a35a6c151a535
121976Author: Thomas Winischhofer <thomas@winischhofer.net>
121977Date: Tue Oct 11 19:02:18 2005 +0000
121978
121979 Add DGAReInitModes in order to allow the driver to change the list of
121980 supported DGA modes. (Part 1)
121981
121982commit c1a2abadfbb862cbaac3e23d0c1317ce5473ebdd
121983Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
121984Date: Tue Oct 11 14:50:47 2005 +0000
121985
121986 fix a typo
121987
121988commit 697f64a22ac5a7742a0022605a1074351296d4f8
121989Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
121990Date: Tue Oct 11 14:50:03 2005 +0000
121991
121992 check randrp has been initialized
121993
121994commit 2828d92c6ca400b603b6a20a221d9c858732292f
121995Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
121996Date: Tue Oct 11 14:45:01 2005 +0000
121997
121998 programs/Xserver/hw/xfree86/common/xf86RandR.c
121999 programs/Xserver/hw/xfree86/loader/xf86sym.c Add a new function
122000 xf86GetRotation to allow third party modules to obtain the current
122001 rotation.
122002
122003commit cad18ec979e38ef80a606f0e4abf2142b9d0d2b1
122004Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
122005Date: Mon Oct 10 10:07:47 2005 +0000
122006
122007 don't move x or y depending on the screen size change
122008
122009commit 7f72f94aa4f0655b8aab6c67eef2a5f5ac4b418f
122010Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
122011Date: Mon Oct 10 09:31:49 2005 +0000
122012
122013 rework that again
122014
122015commit 7c1d9a31a36552467d194e7d009c17dc526256c2
122016Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
122017Date: Mon Oct 10 09:24:28 2005 +0000
122018
122019 a furthur tweak to the randr cursor position fix
122020
122021commit 13f958fbe8420e406f24c01d320f29002ee860b7
122022Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
122023Date: Mon Oct 10 05:58:41 2005 +0000
122024
122025 Add missing {Prepare,Finish}Access() wrappers for the tile pixmap in the
122026 fallback case
122027
122028commit d82aeb55ca3b6abe4cafa7b9c39777a5f67308e5
122029Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122030Date: Sun Oct 9 23:47:52 2005 +0000
122031
122032 Bug #3254 <https://bugs.freedesktop.org/show_bug.cgi?id=3254> Make sure
122033 screensaver & DPMS timeouts don't overflow when multiplied by
122034 MILLI_PER_MIN. (Reported by Zachary J. Slater)
122035
122036commit 29b5f846d261976f466d2c7181d6a75de670066b
122037Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122038Date: Sun Oct 9 17:47:34 2005 +0000
122039
122040 Bug #4715 <https://bugs.freedesktop.org/show_bug.cgi?id=4715> Add
122041 SecurityPolicy to EXTRA_DIST (Bill Crawford)
122042
122043commit 046234b3ebdfe221de9e87d70d287f69a6f59d6e
122044Author: Eric Anholt <anholt@freebsd.org>
122045Date: Sun Oct 9 02:03:22 2005 +0000
122046
122047 Don't try the accelerated glyphs path for component-alpha text (which I
122048 don't expect drivers to be able to accelerate without exa assistance).
122049 Instead, drop back to plain old miGlyphs for a 62.5% +/- 1.5% reduction
122050 in runtime of my ls -lR test (n=5) with component alpha. While a
122051 reasonable approach would seem to be making a better test to see
122052 whether the entire path would be accelerated and force migration
122053 appropriately, my attempt at this made the situation much worse.
122054
122055commit 526d1502df8db6799c9d1155b86ce79cef90872b
122056Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
122057Date: Fri Oct 7 21:29:39 2005 +0000
122058
122059 another update to the RandR fix (thanks Aaron)
122060
122061commit 5a71a5667eb5b01e0f65f9310f4af1f6c5711ab7
122062Author: Kristian Høgsberg <krh@redhat.com>
122063Date: Fri Oct 7 19:01:10 2005 +0000
122064
122065 Add Xprint init scrips to EXTRA_DIST.
122066
122067commit 470213753b158225b44a39a872599344acbc7101
122068Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
122069Date: Fri Oct 7 18:15:08 2005 +0000
122070
122071 update the last RandR fix
122072
122073commit 348242f35aeb2869ef390241035b5f3266fc0288
122074Author: Alan Hourihane <alanh@fairlite.demon.co.uk>
122075Date: Fri Oct 7 15:39:52 2005 +0000
122076
122077 programs/Xserver/hw/xfree86/common/xf86RandR.c Use PointerMoved instead of
122078 SetCursorPosition, as PointerMoved will call AdjustFrame to reposition
122079 the window if necessary and avoid the cursor ending up offscreen.
122080
122081commit 578e18d11b3d61449c1dd7eba04e1748f19c68f3
122082Author: Kevin E Martin <kem@kem.org>
122083Date: Fri Oct 7 14:27:47 2005 +0000
122084
122085 Add darwin to dist
122086 Include missing docs in EXTRA_DIST
122087 Include headers and other files in dist
122088
122089commit 148df64a05d69adaac4b0f3684b846eb1da60219
122090Author: Kevin E Martin <kem@kem.org>
122091Date: Fri Oct 7 04:11:02 2005 +0000
122092
122093 Add README.compiled to dist tarball
122094
122095commit ff258ac2783203ed2a7698894d951391d1aecebc
122096Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
122097Date: Thu Oct 6 23:45:29 2005 +0000
122098
122099 Clients tend to set picture->repeat when not necessary. Most HW cannot
122100 accelerate repeat NPOT thus triggering software fallback (this is the
122101 case with gnome desktop for example). This adds a simple optimisation
122102 to exa that removes "repeat" when it's obviously useless, that is, the
122103 single picture instance covers the entire rectangle beeing used
122104
122105commit e4ed43c3a6c248ba2b82b8bbf29da537a68407e6
122106Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
122107Date: Thu Oct 6 22:25:35 2005 +0000
122108
122109 symlink.sh:
122110 New files linked:
122111 xorgconf.cpp Options
122112 usb.3 usb_hid_usages
122113 lynx_ppc.S
122114 BUSmemcpy.S IODelay.S PortIO.S SlowBcopy.S
122115 sun_inout.s
122116 xaaTEGlyphBlt.S
122117 xkbcomp/compiled/README
122118 New files excluded:
122119 All of lib/GL/apple
122120 xlibi18n/*/*.mapfile
122121 xxserver/xorg/configure.ac, xkb/Makefile.am:
122122 Install README.compiled in the xkb output dir
122123
122124commit 1614a31a9dad9482ae4526c194c2bae1c4993f8f
122125Author: Eric Anholt <anholt@freebsd.org>
122126Date: Thu Oct 6 21:55:41 2005 +0000
122127
122128 Bug #4699: Correct some memory leaks in EXA and damage related to region
122129 handling.
122130
122131commit cd9ff6aec81e04bbfe14364407ccb28df05fc063
122132Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122133Date: Thu Oct 6 20:16:13 2005 +0000
122134
122135 cpp processing for Xvfb man page
122136
122137commit 370b8c8f1cb1a3531d52ea3b430852a0d76b2a4c
122138Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122139Date: Thu Oct 6 20:14:43 2005 +0000
122140
122141 App-defaults file not supposed to have .ad suffix when installed Fix cpp
122142 rules to set needed flags for app-defaults file
122143
122144commit 2770233069d3845c681bea8eccff22e92254487e
122145Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122146Date: Thu Oct 6 19:59:26 2005 +0000
122147
122148 Don't build "ev" example on systems without <linux/input.h>
122149
122150commit 460145a5d52b5325fa5e920cee3699fcf7dd9afe
122151Author: Kristian Høgsberg <krh@redhat.com>
122152Date: Thu Oct 6 19:37:39 2005 +0000
122153
122154 Add cpconfig.c to EXTRA_DIST.
122155 Add CURSOR.NOTES to EXTRA_DIST.
122156 Add extrapci.ids to EXTRA_DIST and fix xf86PciIds.h rule.
122157
122158commit e63f76caa1b1342422567fdcb9f8af24792c8ca1
122159Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122160Date: Thu Oct 6 17:55:54 2005 +0000
122161
122162 Sun bug #6326551: xkbSetDetectableAutoRepeat broken when using XEvIE
122163 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6326551>
122164 (Derek Wang, Sun Microsystems)
122165
122166commit 5f30a7b10286b4f55821acd4eb5580a8f5a3c56a
122167Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>
122168Date: Thu Oct 6 08:08:04 2005 +0000
122169
122170 Bug #4689: Treat DirectColor as TrueColor in Render. It fixes some crashes
122171 with xcompmgr when using apps that use a DirectColor visual for their
122172 windows
122173
122174commit 9000c0321baf1e25e1796e6a333aad0e5a22cbe2
122175Author: Kevin E Martin <kem@kem.org>
122176Date: Thu Oct 6 04:05:30 2005 +0000
122177
122178 Install correct man page and add to dist
122179
122180commit 9b894df44b575f768a2400d044d8c1eb6ef2ec97
122181Author: Kevin E Martin <kem@kem.org>
122182Date: Thu Oct 6 02:40:41 2005 +0000
122183
122184 Include dmx-config.h for modular build
122185 Use <X11/extensions/dmxext.h> intead of "dmxext.h"
122186
122187commit 30c1369bf5816ffd7bd52d9a9dbcb72500684e2f
122188Author: Kevin E Martin <kem@kem.org>
122189Date: Thu Oct 6 02:35:22 2005 +0000
122190
122191 Add support for building DMX config and examples programs Add missing files
122192 to EXTRA_DIST Install Xdmx man page
122193
122194commit 1f9b6dc1ccd999c90ba825cf5fbdfa29770224a6
122195Author: Kevin E Martin <kem@kem.org>
122196Date: Thu Oct 6 00:34:29 2005 +0000
122197
122198 Clean up generated files to pass distcheck
122199 Clean up generated files to pass distcheck Distribute getconfig.man.pre,
122200 not getconfig.man
122201
122202commit 61cd478b545de0313271cf6852e2df770e8f5914
122203Author: Adam Jackson <ajax@nwnk.net>
122204Date: Wed Oct 5 22:39:41 2005 +0000
122205
122206 Bug #3652: Server-side GLX support for GLX_SGIX_swap_barrier and
122207 GLX_SGIX_hyperpipe extensions. (Eric Kunze, SGI)
122208
122209commit e891d9c078bd31447ae3e1fc7f8c15953b0bb916
122210Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122211Date: Wed Oct 5 22:19:09 2005 +0000
122212
122213 Update to 2005-10-05 snapshot from pciids.sf.net (includes a couple new
122214 Radeon id's).
122215
122216commit dc6ac8e46f80157960a24a1be1fb83f22dff45a0
122217Author: Kristian Høgsberg <krh@redhat.com>
122218Date: Wed Oct 5 21:38:40 2005 +0000
122219
122220 Add DGA configure option and add various files that we should be dist'ing.
122221 Simplify xf86DefModeSet.c rule a bit.
122222
122223commit 8391eaa4aa1ae3744ad8c45f5d148ba362d2c9dd
122224Author: Adam Jackson <ajax@nwnk.net>
122225Date: Wed Oct 5 21:13:49 2005 +0000
122226
122227 Preprocess and install XOrgCfg.ad as in the monolith.
122228
122229commit a9df169f108b15d312421e498675cd2e48206660
122230Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122231Date: Wed Oct 5 17:27:58 2005 +0000
122232
122233 Add missing $(DESTDIR) to custom install target
122234
122235commit a6cbe0776fcc8fb19a2bf2ecef41559eed6e5cef
122236Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122237Date: Wed Oct 5 16:39:09 2005 +0000
122238
122239 Fix the rest of the XFree86 DDX options that require an argument to say so
122240 instead of reporting "unrecognized option" when the argument is
122241 missing. Also give correct error instead of "unrecognized option" for
122242 options only available to root.
122243
122244commit 8c524f9966d2a167ea71dd81e235140e0db31471
122245Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122246Date: Wed Oct 5 15:33:40 2005 +0000
122247
122248 Xdmx & Xprint also need xau & xdmcp module dependencies
122249
122250commit 9f3ad65251832631630f7e587b409b750a144bd3
122251Author: Luc Verhaegen <libv@skynet.be>
122252Date: Wed Oct 5 07:27:52 2005 +0000
122253
122254 Fix lnx_pci.c's xf86GetOSOffsetFromPCI return value. Clears up the resource
122255 ranges awkwardness and the "INVALID MEM ALLOCATION" warning.
122256
122257commit da989e988cc96c0ec4f07fceb4c36b30c2e37f4a
122258Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122259Date: Wed Oct 5 02:18:10 2005 +0000
122260
122261 Xnest depends on xdmcp & xau modules too
122262
122263commit 9e8b5f3d478ca18a9ff9c26745de77c91a5d36d9
122264Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122265Date: Wed Oct 5 01:38:50 2005 +0000
122266
122267 Make Xorg -config stop lying to people and claiming it doesn't exist when
122268 you fail to specify a file name. Also, include it in the list of
122269 available options for non-root users when listing all available flags.
122270
122271commit 34a8411ede185553f1387ee0bf534cf77b0fc004
122272Author: Adam Jackson <ajax@nwnk.net>
122273Date: Wed Oct 5 00:55:08 2005 +0000
122274
122275 Bug #4038: Unbreak the SYSVIPC check for cross builds. (Detlef Vollman)
122276
122277commit f47f00ab747563678c8625de5e5b2a588660064e
122278Author: Eric Anholt <anholt@freebsd.org>
122279Date: Tue Oct 4 11:24:09 2005 +0000
122280
122281 Mark the temporary pixmap dirty if UploadToScreen succeeds. Failure to do
122282 so resulted in a solid black glyph if the font rendering actually
122283 resulted in a fallback (subpixel AA, for example) and the temporary got
122284 migrated after 10 or so glyphs.
122285
122286commit 89a1a91b88b94b341075bc208941337ce11465b7
122287Author: Aaron Plattner <aplattner@nvidia.com>
122288Date: Tue Oct 4 07:42:21 2005 +0000
122289
122290 Add miext/cw to the module loader include path so that misym.c can export
122291 miDisableCompositeWrapper.
122292
122293commit 43625a47063c246e7bf9d687caded0b7e2ea0dc6
122294Author: Aaron Plattner <aplattner@nvidia.com>
122295Date: Tue Oct 4 07:31:53 2005 +0000
122296
122297 Bump the video driver module ABI minor version to 8 so modules statically
122298 linked against miDisableCompositeWrapper won't load on older servers.
122299 #include "cw.h" instead of #include "cw/cw.h"
122300
122301commit ca57db6fc1e6100c47ad935d626fdd490ed6116e
122302Author: Aaron Plattner <aplattner@nvidia.com>
122303Date: Tue Oct 4 04:30:33 2005 +0000
122304
122305 Export miDisableCompositeWrapper.
122306
122307commit b2e451b93c20efc49a6cc565239432b2c705fe37
122308Author: Eric Anholt <anholt@freebsd.org>
122309Date: Tue Oct 4 03:44:14 2005 +0000
122310
122311 Correct the test for whether projective transform is necessary. Also, use
122312 "affine" to describe the variable (universally) on suggestion from
122313 vektor. Corrects a rendercheck failure.
122314
122315commit cdded97a0ad717f4f9120b37d2687fa661696c9b
122316Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122317Date: Tue Oct 4 00:45:42 2005 +0000
122318
122319 Add #include <xorg-config.h> for modular build
122320
122321commit 4ae4fc7d51aeb0f27bed52f7e6a346745f3ea453
122322Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122323Date: Tue Oct 4 00:43:16 2005 +0000
122324
122325 Add gtf to xserver/xorg/hw/xfree86/utils
122326
122327commit d51962378ef6371456e034d6d7f6780e05bc1207
122328Author: Adam Jackson <ajax@nwnk.net>
122329Date: Mon Oct 3 19:31:50 2005 +0000
122330
122331 Bug #3781: Only use fbCopyAreammx when alu == GXcopy. Originally Gentoo bug
122332 #96053, patch by bartron@gmx.net.
122333
122334commit 6d4b350dee9495e54e6e5492815885f1d8455ac9
122335Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122336Date: Mon Oct 3 16:46:14 2005 +0000
122337
122338 Bug #3815 <https://bugs.freedesktop.org/show_bug.cgi?id=3815> Patch #3463
122339 <https://bugs.freedesktop.org/attachment.cgi?id=3463> GNU/kFreeBSD
122340 Xserver support (Robert Millan)
122341
122342commit 22b4200b01310e7b4743ef0b3541c3053a2d8279
122343Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122344Date: Mon Oct 3 15:41:10 2005 +0000
122345
122346 Whoops, need to be in DIST_SUBDIRS too.
122347
122348commit 14a2bd33307fd937804a9fbb03787ec30858a05c
122349Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122350Date: Mon Oct 3 15:37:57 2005 +0000
122351
122352 Add missing ] (Dawid Gajownik)
122353 Add getconfig (Dawid Gajownik)
122354
122355commit 84141fc299b03b5552be093f9b698a85bc670d65
122356Author: Eric Anholt <anholt@freebsd.org>
122357Date: Mon Oct 3 11:43:55 2005 +0000
122358
122359 Merge r1.36 of fbcompose.c from xserver CVS: Special case projective
122360 transforms so we can avoid doing the expensive
122361 64-bit math. Unroll the bilinear interpolation loops for an extra boost. I
122362 tested this with the up/downscaling cairo-benchmarks with Xvfb and saw
122363 a 12% +/- 4% decrease in time taken to run them.
122364
122365commit c024262eae4e00567ccb66a59b4d572621233cbc
122366Author: Eric Anholt <anholt@freebsd.org>
122367Date: Mon Oct 3 10:20:29 2005 +0000
122368
122369 Merge r1.2 of fbedge.c from xserver CVS: Optimize spans where the same
122370 value is being added to multiple pixels. This improves the speed of
122371 rendering wide trapezoids. I tested this with a small set of xlibs
122372 cairo-benchmarks with Xvfb and saw a 4% decrease in time taken to run
122373 them.
122374
122375commit a7e3c6fa8ceb6a3a423377aa32ab0da5a6ab9286
122376Author: Adam Jackson <ajax@nwnk.net>
122377Date: Mon Oct 3 06:31:48 2005 +0000
122378
122379 Real configure check for execinfo.h (Yuri Vasilevski)
122380
122381commit e3d2a7d57bc57453d66aa63ca7fe4d910b64737c
122382Author: Adam Jackson <ajax@nwnk.net>
122383Date: Mon Oct 3 06:29:14 2005 +0000
122384
122385 Bug #4393: uClibc lies and defines __GLIBC__ even though it's not source
122386 compatible with glibc, so the backtrace support check fails. Work
122387 around this by wrapping the code in a configure check for execinfo.h,
122388 and emulate detection for the monolith. (Yuri Vasilevski)
122389
122390commit 5037d3441d65f1fb6493c3b55137ef1b5eddd6b0
122391Author: Alan Coopersmith <Alan.Coopersmith@sun.com>
122392Date: Sun Oct 2 22:17:38 2005 +0000
122393
122394 Bug #1465 <https://bugs.freedesktop.org/show_bug.cgi?id=1465>
122395 /etc/init.d/Xprint should use PROJECTROOT from build (Grzegorz
122396