Imported Debian version 2.5.3~trusty1
[deb_ffmpeg.git] / ffmpeg / libavcodec / cavs.c
1 /*
2 * Chinese AVS video (AVS1-P2, JiZhun profile) decoder.
3 * Copyright (c) 2006 Stefan Gehrer <stefan.gehrer@gmx.de>
4 *
5 * This file is part of FFmpeg.
6 *
7 * FFmpeg is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
11 *
12 * FFmpeg is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 * Lesser General Public License for more details.
16 *
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with FFmpeg; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20 */
21
22 /**
23 * @file
24 * Chinese AVS video (AVS1-P2, JiZhun profile) decoder
25 * @author Stefan Gehrer <stefan.gehrer@gmx.de>
26 */
27
28 #include "avcodec.h"
29 #include "get_bits.h"
30 #include "golomb.h"
31 #include "h264chroma.h"
32 #include "idctdsp.h"
33 #include "internal.h"
34 #include "mathops.h"
35 #include "qpeldsp.h"
36 #include "cavs.h"
37
38 static const uint8_t alpha_tab[64] = {
39 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 3, 3,
40 4, 4, 5, 5, 6, 7, 8, 9, 10, 11, 12, 13, 15, 16, 18, 20,
41 22, 24, 26, 28, 30, 33, 33, 35, 35, 36, 37, 37, 39, 39, 42, 44,
42 46, 48, 50, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64
43 };
44
45 static const uint8_t beta_tab[64] = {
46 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2,
47 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6,
48 6, 7, 7, 7, 8, 8, 8, 9, 9, 10, 10, 11, 11, 12, 13, 14,
49 15, 16, 17, 18, 19, 20, 21, 22, 23, 23, 24, 24, 25, 25, 26, 27
50 };
51
52 static const uint8_t tc_tab[64] = {
53 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
54 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2,
55 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4,
56 5, 5, 5, 6, 6, 6, 7, 7, 7, 7, 8, 8, 8, 9, 9, 9
57 };
58
59 /** mark block as unavailable, i.e. out of picture
60 * or not yet decoded */
61 static const cavs_vector un_mv = { 0, 0, 1, NOT_AVAIL };
62
63 static const int8_t left_modifier_l[8] = { 0, -1, 6, -1, -1, 7, 6, 7 };
64 static const int8_t top_modifier_l[8] = { -1, 1, 5, -1, -1, 5, 7, 7 };
65 static const int8_t left_modifier_c[7] = { 5, -1, 2, -1, 6, 5, 6 };
66 static const int8_t top_modifier_c[7] = { 4, 1, -1, -1, 4, 6, 6 };
67
68 /*****************************************************************************
69 *
70 * in-loop deblocking filter
71 *
72 ****************************************************************************/
73
74 static inline int get_bs(cavs_vector *mvP, cavs_vector *mvQ, int b)
75 {
76 if ((mvP->ref == REF_INTRA) || (mvQ->ref == REF_INTRA))
77 return 2;
78 if((abs(mvP->x - mvQ->x) >= 4) ||
79 (abs(mvP->y - mvQ->y) >= 4) ||
80 (mvP->ref != mvQ->ref))
81 return 1;
82 if (b) {
83 mvP += MV_BWD_OFFS;
84 mvQ += MV_BWD_OFFS;
85 if((abs(mvP->x - mvQ->x) >= 4) ||
86 (abs(mvP->y - mvQ->y) >= 4) ||
87 (mvP->ref != mvQ->ref))
88 return 1;
89 }
90 return 0;
91 }
92
93 #define SET_PARAMS \
94 alpha = alpha_tab[av_clip(qp_avg + h->alpha_offset, 0, 63)]; \
95 beta = beta_tab[av_clip(qp_avg + h->beta_offset, 0, 63)]; \
96 tc = tc_tab[av_clip(qp_avg + h->alpha_offset, 0, 63)];
97
98 /**
99 * in-loop deblocking filter for a single macroblock
100 *
101 * boundary strength (bs) mapping:
102 *
103 * --4---5--
104 * 0 2 |
105 * | 6 | 7 |
106 * 1 3 |
107 * ---------
108 *
109 */
110 void ff_cavs_filter(AVSContext *h, enum cavs_mb mb_type)
111 {
112 uint8_t bs[8];
113 int qp_avg, alpha, beta, tc;
114 int i;
115
116 /* save un-deblocked lines */
117 h->topleft_border_y = h->top_border_y[h->mbx * 16 + 15];
118 h->topleft_border_u = h->top_border_u[h->mbx * 10 + 8];
119 h->topleft_border_v = h->top_border_v[h->mbx * 10 + 8];
120 memcpy(&h->top_border_y[h->mbx * 16], h->cy + 15 * h->l_stride, 16);
121 memcpy(&h->top_border_u[h->mbx * 10 + 1], h->cu + 7 * h->c_stride, 8);
122 memcpy(&h->top_border_v[h->mbx * 10 + 1], h->cv + 7 * h->c_stride, 8);
123 for (i = 0; i < 8; i++) {
124 h->left_border_y[i * 2 + 1] = *(h->cy + 15 + (i * 2 + 0) * h->l_stride);
125 h->left_border_y[i * 2 + 2] = *(h->cy + 15 + (i * 2 + 1) * h->l_stride);
126 h->left_border_u[i + 1] = *(h->cu + 7 + i * h->c_stride);
127 h->left_border_v[i + 1] = *(h->cv + 7 + i * h->c_stride);
128 }
129 if (!h->loop_filter_disable) {
130 /* determine bs */
131 if (mb_type == I_8X8)
132 memset(bs, 2, 8);
133 else {
134 memset(bs, 0, 8);
135 if (ff_cavs_partition_flags[mb_type] & SPLITV) {
136 bs[2] = get_bs(&h->mv[MV_FWD_X0], &h->mv[MV_FWD_X1], mb_type > P_8X8);
137 bs[3] = get_bs(&h->mv[MV_FWD_X2], &h->mv[MV_FWD_X3], mb_type > P_8X8);
138 }
139 if (ff_cavs_partition_flags[mb_type] & SPLITH) {
140 bs[6] = get_bs(&h->mv[MV_FWD_X0], &h->mv[MV_FWD_X2], mb_type > P_8X8);
141 bs[7] = get_bs(&h->mv[MV_FWD_X1], &h->mv[MV_FWD_X3], mb_type > P_8X8);
142 }
143 bs[0] = get_bs(&h->mv[MV_FWD_A1], &h->mv[MV_FWD_X0], mb_type > P_8X8);
144 bs[1] = get_bs(&h->mv[MV_FWD_A3], &h->mv[MV_FWD_X2], mb_type > P_8X8);
145 bs[4] = get_bs(&h->mv[MV_FWD_B2], &h->mv[MV_FWD_X0], mb_type > P_8X8);
146 bs[5] = get_bs(&h->mv[MV_FWD_B3], &h->mv[MV_FWD_X1], mb_type > P_8X8);
147 }
148 if (AV_RN64(bs)) {
149 if (h->flags & A_AVAIL) {
150 qp_avg = (h->qp + h->left_qp + 1) >> 1;
151 SET_PARAMS;
152 h->cdsp.cavs_filter_lv(h->cy, h->l_stride, alpha, beta, tc, bs[0], bs[1]);
153 qp_avg = (ff_cavs_chroma_qp[h->qp] + ff_cavs_chroma_qp[h->left_qp] + 1) >> 1;
154 SET_PARAMS;
155 h->cdsp.cavs_filter_cv(h->cu, h->c_stride, alpha, beta, tc, bs[0], bs[1]);
156 h->cdsp.cavs_filter_cv(h->cv, h->c_stride, alpha, beta, tc, bs[0], bs[1]);
157 }
158 qp_avg = h->qp;
159 SET_PARAMS;
160 h->cdsp.cavs_filter_lv(h->cy + 8, h->l_stride, alpha, beta, tc, bs[2], bs[3]);
161 h->cdsp.cavs_filter_lh(h->cy + 8 * h->l_stride, h->l_stride, alpha, beta, tc, bs[6], bs[7]);
162
163 if (h->flags & B_AVAIL) {
164 qp_avg = (h->qp + h->top_qp[h->mbx] + 1) >> 1;
165 SET_PARAMS;
166 h->cdsp.cavs_filter_lh(h->cy, h->l_stride, alpha, beta, tc, bs[4], bs[5]);
167 qp_avg = (ff_cavs_chroma_qp[h->qp] + ff_cavs_chroma_qp[h->top_qp[h->mbx]] + 1) >> 1;
168 SET_PARAMS;
169 h->cdsp.cavs_filter_ch(h->cu, h->c_stride, alpha, beta, tc, bs[4], bs[5]);
170 h->cdsp.cavs_filter_ch(h->cv, h->c_stride, alpha, beta, tc, bs[4], bs[5]);
171 }
172 }
173 }
174 h->left_qp = h->qp;
175 h->top_qp[h->mbx] = h->qp;
176 }
177
178 #undef SET_PARAMS
179
180 /*****************************************************************************
181 *
182 * spatial intra prediction
183 *
184 ****************************************************************************/
185
186 void ff_cavs_load_intra_pred_luma(AVSContext *h, uint8_t *top,
187 uint8_t **left, int block)
188 {
189 int i;
190
191 switch (block) {
192 case 0:
193 *left = h->left_border_y;
194 h->left_border_y[0] = h->left_border_y[1];
195 memset(&h->left_border_y[17], h->left_border_y[16], 9);
196 memcpy(&top[1], &h->top_border_y[h->mbx * 16], 16);
197 top[17] = top[16];
198 top[0] = top[1];
199 if ((h->flags & A_AVAIL) && (h->flags & B_AVAIL))
200 h->left_border_y[0] = top[0] = h->topleft_border_y;
201 break;
202 case 1:
203 *left = h->intern_border_y;
204 for (i = 0; i < 8; i++)
205 h->intern_border_y[i + 1] = *(h->cy + 7 + i * h->l_stride);
206 memset(&h->intern_border_y[9], h->intern_border_y[8], 9);
207 h->intern_border_y[0] = h->intern_border_y[1];
208 memcpy(&top[1], &h->top_border_y[h->mbx * 16 + 8], 8);
209 if (h->flags & C_AVAIL)
210 memcpy(&top[9], &h->top_border_y[(h->mbx + 1) * 16], 8);
211 else
212 memset(&top[9], top[8], 9);
213 top[17] = top[16];
214 top[0] = top[1];
215 if (h->flags & B_AVAIL)
216 h->intern_border_y[0] = top[0] = h->top_border_y[h->mbx * 16 + 7];
217 break;
218 case 2:
219 *left = &h->left_border_y[8];
220 memcpy(&top[1], h->cy + 7 * h->l_stride, 16);
221 top[17] = top[16];
222 top[0] = top[1];
223 if (h->flags & A_AVAIL)
224 top[0] = h->left_border_y[8];
225 break;
226 case 3:
227 *left = &h->intern_border_y[8];
228 for (i = 0; i < 8; i++)
229 h->intern_border_y[i + 9] = *(h->cy + 7 + (i + 8) * h->l_stride);
230 memset(&h->intern_border_y[17], h->intern_border_y[16], 9);
231 memcpy(&top[0], h->cy + 7 + 7 * h->l_stride, 9);
232 memset(&top[9], top[8], 9);
233 break;
234 }
235 }
236
237 void ff_cavs_load_intra_pred_chroma(AVSContext *h)
238 {
239 /* extend borders by one pixel */
240 h->left_border_u[9] = h->left_border_u[8];
241 h->left_border_v[9] = h->left_border_v[8];
242 if(h->flags & C_AVAIL) {
243 h->top_border_u[h->mbx*10 + 9] = h->top_border_u[h->mbx*10 + 11];
244 h->top_border_v[h->mbx*10 + 9] = h->top_border_v[h->mbx*10 + 11];
245 } else {
246 h->top_border_u[h->mbx * 10 + 9] = h->top_border_u[h->mbx * 10 + 8];
247 h->top_border_v[h->mbx * 10 + 9] = h->top_border_v[h->mbx * 10 + 8];
248 }
249 if((h->flags & A_AVAIL) && (h->flags & B_AVAIL)) {
250 h->top_border_u[h->mbx * 10] = h->left_border_u[0] = h->topleft_border_u;
251 h->top_border_v[h->mbx * 10] = h->left_border_v[0] = h->topleft_border_v;
252 } else {
253 h->left_border_u[0] = h->left_border_u[1];
254 h->left_border_v[0] = h->left_border_v[1];
255 h->top_border_u[h->mbx * 10] = h->top_border_u[h->mbx * 10 + 1];
256 h->top_border_v[h->mbx * 10] = h->top_border_v[h->mbx * 10 + 1];
257 }
258 }
259
260 static void intra_pred_vert(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
261 {
262 int y;
263 uint64_t a = AV_RN64(&top[1]);
264 for (y = 0; y < 8; y++)
265 *((uint64_t *)(d + y * stride)) = a;
266 }
267
268 static void intra_pred_horiz(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
269 {
270 int y;
271 uint64_t a;
272 for (y = 0; y < 8; y++) {
273 a = left[y + 1] * 0x0101010101010101ULL;
274 *((uint64_t *)(d + y * stride)) = a;
275 }
276 }
277
278 static void intra_pred_dc_128(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
279 {
280 int y;
281 uint64_t a = 0x8080808080808080ULL;
282 for (y = 0; y < 8; y++)
283 *((uint64_t *)(d + y * stride)) = a;
284 }
285
286 static void intra_pred_plane(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
287 {
288 int x, y, ia;
289 int ih = 0;
290 int iv = 0;
291 const uint8_t *cm = ff_crop_tab + MAX_NEG_CROP;
292
293 for (x = 0; x < 4; x++) {
294 ih += (x + 1) * (top[5 + x] - top[3 - x]);
295 iv += (x + 1) * (left[5 + x] - left[3 - x]);
296 }
297 ia = (top[8] + left[8]) << 4;
298 ih = (17 * ih + 16) >> 5;
299 iv = (17 * iv + 16) >> 5;
300 for (y = 0; y < 8; y++)
301 for (x = 0; x < 8; x++)
302 d[y * stride + x] = cm[(ia + (x - 3) * ih + (y - 3) * iv + 16) >> 5];
303 }
304
305 #define LOWPASS(ARRAY, INDEX) \
306 ((ARRAY[(INDEX) - 1] + 2 * ARRAY[(INDEX)] + ARRAY[(INDEX) + 1] + 2) >> 2)
307
308 static void intra_pred_lp(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
309 {
310 int x, y;
311 for (y = 0; y < 8; y++)
312 for (x = 0; x < 8; x++)
313 d[y * stride + x] = (LOWPASS(top, x + 1) + LOWPASS(left, y + 1)) >> 1;
314 }
315
316 static void intra_pred_down_left(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
317 {
318 int x, y;
319 for (y = 0; y < 8; y++)
320 for (x = 0; x < 8; x++)
321 d[y * stride + x] = (LOWPASS(top, x + y + 2) + LOWPASS(left, x + y + 2)) >> 1;
322 }
323
324 static void intra_pred_down_right(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
325 {
326 int x, y;
327 for (y = 0; y < 8; y++)
328 for (x = 0; x < 8; x++)
329 if (x == y)
330 d[y * stride + x] = (left[1] + 2 * top[0] + top[1] + 2) >> 2;
331 else if (x > y)
332 d[y * stride + x] = LOWPASS(top, x - y);
333 else
334 d[y * stride + x] = LOWPASS(left, y - x);
335 }
336
337 static void intra_pred_lp_left(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
338 {
339 int x, y;
340 for (y = 0; y < 8; y++)
341 for (x = 0; x < 8; x++)
342 d[y * stride + x] = LOWPASS(left, y + 1);
343 }
344
345 static void intra_pred_lp_top(uint8_t *d, uint8_t *top, uint8_t *left, int stride)
346 {
347 int x, y;
348 for (y = 0; y < 8; y++)
349 for (x = 0; x < 8; x++)
350 d[y * stride + x] = LOWPASS(top, x + 1);
351 }
352
353 #undef LOWPASS
354
355 static inline void modify_pred(const int8_t *mod_table, int *mode)
356 {
357 *mode = mod_table[*mode];
358 if (*mode < 0) {
359 av_log(NULL, AV_LOG_ERROR, "Illegal intra prediction mode\n");
360 *mode = 0;
361 }
362 }
363
364 void ff_cavs_modify_mb_i(AVSContext *h, int *pred_mode_uv)
365 {
366 /* save pred modes before they get modified */
367 h->pred_mode_Y[3] = h->pred_mode_Y[5];
368 h->pred_mode_Y[6] = h->pred_mode_Y[8];
369 h->top_pred_Y[h->mbx * 2 + 0] = h->pred_mode_Y[7];
370 h->top_pred_Y[h->mbx * 2 + 1] = h->pred_mode_Y[8];
371
372 /* modify pred modes according to availability of neighbour samples */
373 if (!(h->flags & A_AVAIL)) {
374 modify_pred(left_modifier_l, &h->pred_mode_Y[4]);
375 modify_pred(left_modifier_l, &h->pred_mode_Y[7]);
376 modify_pred(left_modifier_c, pred_mode_uv);
377 }
378 if (!(h->flags & B_AVAIL)) {
379 modify_pred(top_modifier_l, &h->pred_mode_Y[4]);
380 modify_pred(top_modifier_l, &h->pred_mode_Y[5]);
381 modify_pred(top_modifier_c, pred_mode_uv);
382 }
383 }
384
385 /*****************************************************************************
386 *
387 * motion compensation
388 *
389 ****************************************************************************/
390
391 static inline void mc_dir_part(AVSContext *h, AVFrame *pic, int chroma_height,
392 int delta, int list, uint8_t *dest_y,
393 uint8_t *dest_cb, uint8_t *dest_cr,
394 int src_x_offset, int src_y_offset,
395 qpel_mc_func *qpix_op,
396 h264_chroma_mc_func chroma_op, cavs_vector *mv)
397 {
398 const int mx = mv->x + src_x_offset * 8;
399 const int my = mv->y + src_y_offset * 8;
400 const int luma_xy = (mx & 3) + ((my & 3) << 2);
401 uint8_t *src_y = pic->data[0] + (mx >> 2) + (my >> 2) * h->l_stride;
402 uint8_t *src_cb = pic->data[1] + (mx >> 3) + (my >> 3) * h->c_stride;
403 uint8_t *src_cr = pic->data[2] + (mx >> 3) + (my >> 3) * h->c_stride;
404 int extra_width = 0;
405 int extra_height = extra_width;
406 const int full_mx = mx >> 2;
407 const int full_my = my >> 2;
408 const int pic_width = 16 * h->mb_width;
409 const int pic_height = 16 * h->mb_height;
410 int emu = 0;
411
412 if (!pic->data[0])
413 return;
414 if (mx & 7)
415 extra_width -= 3;
416 if (my & 7)
417 extra_height -= 3;
418
419 if (full_mx < 0 - extra_width ||
420 full_my < 0 - extra_height ||
421 full_mx + 16 /* FIXME */ > pic_width + extra_width ||
422 full_my + 16 /* FIXME */ > pic_height + extra_height) {
423 h->vdsp.emulated_edge_mc(h->edge_emu_buffer,
424 src_y - 2 - 2 * h->l_stride,
425 h->l_stride, h->l_stride,
426 16 + 5, 16 + 5 /* FIXME */,
427 full_mx - 2, full_my - 2,
428 pic_width, pic_height);
429 src_y = h->edge_emu_buffer + 2 + 2 * h->l_stride;
430 emu = 1;
431 }
432
433 // FIXME try variable height perhaps?
434 qpix_op[luma_xy](dest_y, src_y, h->l_stride);
435
436 if (emu) {
437 h->vdsp.emulated_edge_mc(h->edge_emu_buffer, src_cb,
438 h->c_stride, h->c_stride,
439 9, 9 /* FIXME */,
440 mx >> 3, my >> 3,
441 pic_width >> 1, pic_height >> 1);
442 src_cb = h->edge_emu_buffer;
443 }
444 chroma_op(dest_cb, src_cb, h->c_stride, chroma_height, mx & 7, my & 7);
445
446 if (emu) {
447 h->vdsp.emulated_edge_mc(h->edge_emu_buffer, src_cr,
448 h->c_stride, h->c_stride,
449 9, 9 /* FIXME */,
450 mx >> 3, my >> 3,
451 pic_width >> 1, pic_height >> 1);
452 src_cr = h->edge_emu_buffer;
453 }
454 chroma_op(dest_cr, src_cr, h->c_stride, chroma_height, mx & 7, my & 7);
455 }
456
457 static inline void mc_part_std(AVSContext *h, int chroma_height, int delta,
458 uint8_t *dest_y,
459 uint8_t *dest_cb,
460 uint8_t *dest_cr,
461 int x_offset, int y_offset,
462 qpel_mc_func *qpix_put,
463 h264_chroma_mc_func chroma_put,
464 qpel_mc_func *qpix_avg,
465 h264_chroma_mc_func chroma_avg,
466 cavs_vector *mv)
467 {
468 qpel_mc_func *qpix_op = qpix_put;
469 h264_chroma_mc_func chroma_op = chroma_put;
470
471 dest_y += x_offset * 2 + y_offset * h->l_stride * 2;
472 dest_cb += x_offset + y_offset * h->c_stride;
473 dest_cr += x_offset + y_offset * h->c_stride;
474 x_offset += 8 * h->mbx;
475 y_offset += 8 * h->mby;
476
477 if (mv->ref >= 0) {
478 AVFrame *ref = h->DPB[mv->ref].f;
479 mc_dir_part(h, ref, chroma_height, delta, 0,
480 dest_y, dest_cb, dest_cr, x_offset, y_offset,
481 qpix_op, chroma_op, mv);
482
483 qpix_op = qpix_avg;
484 chroma_op = chroma_avg;
485 }
486
487 if ((mv + MV_BWD_OFFS)->ref >= 0) {
488 AVFrame *ref = h->DPB[0].f;
489 mc_dir_part(h, ref, chroma_height, delta, 1,
490 dest_y, dest_cb, dest_cr, x_offset, y_offset,
491 qpix_op, chroma_op, mv + MV_BWD_OFFS);
492 }
493 }
494
495 void ff_cavs_inter(AVSContext *h, enum cavs_mb mb_type)
496 {
497 if (ff_cavs_partition_flags[mb_type] == 0) { // 16x16
498 mc_part_std(h, 8, 0, h->cy, h->cu, h->cv, 0, 0,
499 h->cdsp.put_cavs_qpel_pixels_tab[0],
500 h->h264chroma.put_h264_chroma_pixels_tab[0],
501 h->cdsp.avg_cavs_qpel_pixels_tab[0],
502 h->h264chroma.avg_h264_chroma_pixels_tab[0],
503 &h->mv[MV_FWD_X0]);
504 } else {
505 mc_part_std(h, 4, 0, h->cy, h->cu, h->cv, 0, 0,
506 h->cdsp.put_cavs_qpel_pixels_tab[1],
507 h->h264chroma.put_h264_chroma_pixels_tab[1],
508 h->cdsp.avg_cavs_qpel_pixels_tab[1],
509 h->h264chroma.avg_h264_chroma_pixels_tab[1],
510 &h->mv[MV_FWD_X0]);
511 mc_part_std(h, 4, 0, h->cy, h->cu, h->cv, 4, 0,
512 h->cdsp.put_cavs_qpel_pixels_tab[1],
513 h->h264chroma.put_h264_chroma_pixels_tab[1],
514 h->cdsp.avg_cavs_qpel_pixels_tab[1],
515 h->h264chroma.avg_h264_chroma_pixels_tab[1],
516 &h->mv[MV_FWD_X1]);
517 mc_part_std(h, 4, 0, h->cy, h->cu, h->cv, 0, 4,
518 h->cdsp.put_cavs_qpel_pixels_tab[1],
519 h->h264chroma.put_h264_chroma_pixels_tab[1],
520 h->cdsp.avg_cavs_qpel_pixels_tab[1],
521 h->h264chroma.avg_h264_chroma_pixels_tab[1],
522 &h->mv[MV_FWD_X2]);
523 mc_part_std(h, 4, 0, h->cy, h->cu, h->cv, 4, 4,
524 h->cdsp.put_cavs_qpel_pixels_tab[1],
525 h->h264chroma.put_h264_chroma_pixels_tab[1],
526 h->cdsp.avg_cavs_qpel_pixels_tab[1],
527 h->h264chroma.avg_h264_chroma_pixels_tab[1],
528 &h->mv[MV_FWD_X3]);
529 }
530 }
531
532 /*****************************************************************************
533 *
534 * motion vector prediction
535 *
536 ****************************************************************************/
537
538 static inline void scale_mv(AVSContext *h, int *d_x, int *d_y,
539 cavs_vector *src, int distp)
540 {
541 int den = h->scale_den[FFMAX(src->ref, 0)];
542
543 *d_x = (src->x * distp * den + 256 + FF_SIGNBIT(src->x)) >> 9;
544 *d_y = (src->y * distp * den + 256 + FF_SIGNBIT(src->y)) >> 9;
545 }
546
547 static inline void mv_pred_median(AVSContext *h,
548 cavs_vector *mvP,
549 cavs_vector *mvA,
550 cavs_vector *mvB,
551 cavs_vector *mvC)
552 {
553 int ax, ay, bx, by, cx, cy;
554 int len_ab, len_bc, len_ca, len_mid;
555
556 /* scale candidates according to their temporal span */
557 scale_mv(h, &ax, &ay, mvA, mvP->dist);
558 scale_mv(h, &bx, &by, mvB, mvP->dist);
559 scale_mv(h, &cx, &cy, mvC, mvP->dist);
560 /* find the geometrical median of the three candidates */
561 len_ab = abs(ax - bx) + abs(ay - by);
562 len_bc = abs(bx - cx) + abs(by - cy);
563 len_ca = abs(cx - ax) + abs(cy - ay);
564 len_mid = mid_pred(len_ab, len_bc, len_ca);
565 if (len_mid == len_ab) {
566 mvP->x = cx;
567 mvP->y = cy;
568 } else if (len_mid == len_bc) {
569 mvP->x = ax;
570 mvP->y = ay;
571 } else {
572 mvP->x = bx;
573 mvP->y = by;
574 }
575 }
576
577 void ff_cavs_mv(AVSContext *h, enum cavs_mv_loc nP, enum cavs_mv_loc nC,
578 enum cavs_mv_pred mode, enum cavs_block size, int ref)
579 {
580 cavs_vector *mvP = &h->mv[nP];
581 cavs_vector *mvA = &h->mv[nP-1];
582 cavs_vector *mvB = &h->mv[nP-4];
583 cavs_vector *mvC = &h->mv[nC];
584 const cavs_vector *mvP2 = NULL;
585
586 mvP->ref = ref;
587 mvP->dist = h->dist[mvP->ref];
588 if (mvC->ref == NOT_AVAIL || (nP == MV_FWD_X3) || (nP == MV_BWD_X3 ))
589 mvC = &h->mv[nP - 5]; // set to top-left (mvD)
590 if (mode == MV_PRED_PSKIP &&
591 (mvA->ref == NOT_AVAIL ||
592 mvB->ref == NOT_AVAIL ||
593 (mvA->x | mvA->y | mvA->ref) == 0 ||
594 (mvB->x | mvB->y | mvB->ref) == 0)) {
595 mvP2 = &un_mv;
596 /* if there is only one suitable candidate, take it */
597 } else if (mvA->ref >= 0 && mvB->ref < 0 && mvC->ref < 0) {
598 mvP2 = mvA;
599 } else if (mvA->ref < 0 && mvB->ref >= 0 && mvC->ref < 0) {
600 mvP2 = mvB;
601 } else if (mvA->ref < 0 && mvB->ref < 0 && mvC->ref >= 0) {
602 mvP2 = mvC;
603 } else if (mode == MV_PRED_LEFT && mvA->ref == ref) {
604 mvP2 = mvA;
605 } else if (mode == MV_PRED_TOP && mvB->ref == ref) {
606 mvP2 = mvB;
607 } else if (mode == MV_PRED_TOPRIGHT && mvC->ref == ref) {
608 mvP2 = mvC;
609 }
610 if (mvP2) {
611 mvP->x = mvP2->x;
612 mvP->y = mvP2->y;
613 } else
614 mv_pred_median(h, mvP, mvA, mvB, mvC);
615
616 if (mode < MV_PRED_PSKIP) {
617 mvP->x += get_se_golomb(&h->gb);
618 mvP->y += get_se_golomb(&h->gb);
619 }
620 set_mvs(mvP, size);
621 }
622
623 /*****************************************************************************
624 *
625 * macroblock level
626 *
627 ****************************************************************************/
628
629 /**
630 * initialise predictors for motion vectors and intra prediction
631 */
632 void ff_cavs_init_mb(AVSContext *h)
633 {
634 int i;
635
636 /* copy predictors from top line (MB B and C) into cache */
637 for (i = 0; i < 3; i++) {
638 h->mv[MV_FWD_B2 + i] = h->top_mv[0][h->mbx * 2 + i];
639 h->mv[MV_BWD_B2 + i] = h->top_mv[1][h->mbx * 2 + i];
640 }
641 h->pred_mode_Y[1] = h->top_pred_Y[h->mbx * 2 + 0];
642 h->pred_mode_Y[2] = h->top_pred_Y[h->mbx * 2 + 1];
643 /* clear top predictors if MB B is not available */
644 if (!(h->flags & B_AVAIL)) {
645 h->mv[MV_FWD_B2] = un_mv;
646 h->mv[MV_FWD_B3] = un_mv;
647 h->mv[MV_BWD_B2] = un_mv;
648 h->mv[MV_BWD_B3] = un_mv;
649 h->pred_mode_Y[1] = h->pred_mode_Y[2] = NOT_AVAIL;
650 h->flags &= ~(C_AVAIL | D_AVAIL);
651 } else if (h->mbx) {
652 h->flags |= D_AVAIL;
653 }
654 if (h->mbx == h->mb_width - 1) // MB C not available
655 h->flags &= ~C_AVAIL;
656 /* clear top-right predictors if MB C is not available */
657 if (!(h->flags & C_AVAIL)) {
658 h->mv[MV_FWD_C2] = un_mv;
659 h->mv[MV_BWD_C2] = un_mv;
660 }
661 /* clear top-left predictors if MB D is not available */
662 if (!(h->flags & D_AVAIL)) {
663 h->mv[MV_FWD_D3] = un_mv;
664 h->mv[MV_BWD_D3] = un_mv;
665 }
666 }
667
668 /**
669 * save predictors for later macroblocks and increase
670 * macroblock address
671 * @return 0 if end of frame is reached, 1 otherwise
672 */
673 int ff_cavs_next_mb(AVSContext *h)
674 {
675 int i;
676
677 h->flags |= A_AVAIL;
678 h->cy += 16;
679 h->cu += 8;
680 h->cv += 8;
681 /* copy mvs as predictors to the left */
682 for (i = 0; i <= 20; i += 4)
683 h->mv[i] = h->mv[i + 2];
684 /* copy bottom mvs from cache to top line */
685 h->top_mv[0][h->mbx * 2 + 0] = h->mv[MV_FWD_X2];
686 h->top_mv[0][h->mbx * 2 + 1] = h->mv[MV_FWD_X3];
687 h->top_mv[1][h->mbx * 2 + 0] = h->mv[MV_BWD_X2];
688 h->top_mv[1][h->mbx * 2 + 1] = h->mv[MV_BWD_X3];
689 /* next MB address */
690 h->mbidx++;
691 h->mbx++;
692 if (h->mbx == h->mb_width) { // New mb line
693 h->flags = B_AVAIL | C_AVAIL;
694 /* clear left pred_modes */
695 h->pred_mode_Y[3] = h->pred_mode_Y[6] = NOT_AVAIL;
696 /* clear left mv predictors */
697 for (i = 0; i <= 20; i += 4)
698 h->mv[i] = un_mv;
699 h->mbx = 0;
700 h->mby++;
701 /* re-calculate sample pointers */
702 h->cy = h->cur.f->data[0] + h->mby * 16 * h->l_stride;
703 h->cu = h->cur.f->data[1] + h->mby * 8 * h->c_stride;
704 h->cv = h->cur.f->data[2] + h->mby * 8 * h->c_stride;
705 if (h->mby == h->mb_height) { // Frame end
706 return 0;
707 }
708 }
709 return 1;
710 }
711
712 /*****************************************************************************
713 *
714 * frame level
715 *
716 ****************************************************************************/
717
718 int ff_cavs_init_pic(AVSContext *h)
719 {
720 int i;
721
722 /* clear some predictors */
723 for (i = 0; i <= 20; i += 4)
724 h->mv[i] = un_mv;
725 h->mv[MV_BWD_X0] = ff_cavs_dir_mv;
726 set_mvs(&h->mv[MV_BWD_X0], BLK_16X16);
727 h->mv[MV_FWD_X0] = ff_cavs_dir_mv;
728 set_mvs(&h->mv[MV_FWD_X0], BLK_16X16);
729 h->pred_mode_Y[3] = h->pred_mode_Y[6] = NOT_AVAIL;
730 h->cy = h->cur.f->data[0];
731 h->cu = h->cur.f->data[1];
732 h->cv = h->cur.f->data[2];
733 h->l_stride = h->cur.f->linesize[0];
734 h->c_stride = h->cur.f->linesize[1];
735 h->luma_scan[2] = 8 * h->l_stride;
736 h->luma_scan[3] = 8 * h->l_stride + 8;
737 h->mbx = h->mby = h->mbidx = 0;
738 h->flags = 0;
739
740 return 0;
741 }
742
743 /*****************************************************************************
744 *
745 * headers and interface
746 *
747 ****************************************************************************/
748
749 /**
750 * some predictions require data from the top-neighbouring macroblock.
751 * this data has to be stored for one complete row of macroblocks
752 * and this storage space is allocated here
753 */
754 void ff_cavs_init_top_lines(AVSContext *h)
755 {
756 /* alloc top line of predictors */
757 h->top_qp = av_mallocz(h->mb_width);
758 h->top_mv[0] = av_mallocz_array(h->mb_width * 2 + 1, sizeof(cavs_vector));
759 h->top_mv[1] = av_mallocz_array(h->mb_width * 2 + 1, sizeof(cavs_vector));
760 h->top_pred_Y = av_mallocz_array(h->mb_width * 2, sizeof(*h->top_pred_Y));
761 h->top_border_y = av_mallocz_array(h->mb_width + 1, 16);
762 h->top_border_u = av_mallocz_array(h->mb_width, 10);
763 h->top_border_v = av_mallocz_array(h->mb_width, 10);
764
765 /* alloc space for co-located MVs and types */
766 h->col_mv = av_mallocz_array(h->mb_width * h->mb_height,
767 4 * sizeof(cavs_vector));
768 h->col_type_base = av_mallocz(h->mb_width * h->mb_height);
769 h->block = av_mallocz(64 * sizeof(int16_t));
770 }
771
772 av_cold int ff_cavs_init(AVCodecContext *avctx)
773 {
774 AVSContext *h = avctx->priv_data;
775
776 ff_blockdsp_init(&h->bdsp, avctx);
777 ff_h264chroma_init(&h->h264chroma, 8);
778 ff_idctdsp_init(&h->idsp, avctx);
779 ff_videodsp_init(&h->vdsp, 8);
780 ff_cavsdsp_init(&h->cdsp, avctx);
781 ff_init_scantable_permutation(h->idsp.idct_permutation,
782 h->cdsp.idct_perm);
783 ff_init_scantable(h->idsp.idct_permutation, &h->scantable, ff_zigzag_direct);
784
785 h->avctx = avctx;
786 avctx->pix_fmt = AV_PIX_FMT_YUV420P;
787
788 h->cur.f = av_frame_alloc();
789 h->DPB[0].f = av_frame_alloc();
790 h->DPB[1].f = av_frame_alloc();
791 if (!h->cur.f || !h->DPB[0].f || !h->DPB[1].f) {
792 ff_cavs_end(avctx);
793 return AVERROR(ENOMEM);
794 }
795
796 h->luma_scan[0] = 0;
797 h->luma_scan[1] = 8;
798 h->intra_pred_l[INTRA_L_VERT] = intra_pred_vert;
799 h->intra_pred_l[INTRA_L_HORIZ] = intra_pred_horiz;
800 h->intra_pred_l[INTRA_L_LP] = intra_pred_lp;
801 h->intra_pred_l[INTRA_L_DOWN_LEFT] = intra_pred_down_left;
802 h->intra_pred_l[INTRA_L_DOWN_RIGHT] = intra_pred_down_right;
803 h->intra_pred_l[INTRA_L_LP_LEFT] = intra_pred_lp_left;
804 h->intra_pred_l[INTRA_L_LP_TOP] = intra_pred_lp_top;
805 h->intra_pred_l[INTRA_L_DC_128] = intra_pred_dc_128;
806 h->intra_pred_c[INTRA_C_LP] = intra_pred_lp;
807 h->intra_pred_c[INTRA_C_HORIZ] = intra_pred_horiz;
808 h->intra_pred_c[INTRA_C_VERT] = intra_pred_vert;
809 h->intra_pred_c[INTRA_C_PLANE] = intra_pred_plane;
810 h->intra_pred_c[INTRA_C_LP_LEFT] = intra_pred_lp_left;
811 h->intra_pred_c[INTRA_C_LP_TOP] = intra_pred_lp_top;
812 h->intra_pred_c[INTRA_C_DC_128] = intra_pred_dc_128;
813 h->mv[7] = un_mv;
814 h->mv[19] = un_mv;
815 return 0;
816 }
817
818 av_cold int ff_cavs_end(AVCodecContext *avctx)
819 {
820 AVSContext *h = avctx->priv_data;
821
822 av_frame_free(&h->cur.f);
823 av_frame_free(&h->DPB[0].f);
824 av_frame_free(&h->DPB[1].f);
825
826 av_freep(&h->top_qp);
827 av_freep(&h->top_mv[0]);
828 av_freep(&h->top_mv[1]);
829 av_freep(&h->top_pred_Y);
830 av_freep(&h->top_border_y);
831 av_freep(&h->top_border_u);
832 av_freep(&h->top_border_v);
833 av_freep(&h->col_mv);
834 av_freep(&h->col_type_base);
835 av_freep(&h->block);
836 av_freep(&h->edge_emu_buffer);
837 return 0;
838 }