2 * thirdpel DSP functions
4 * This file is part of FFmpeg.
6 * FFmpeg is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
11 * FFmpeg is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with FFmpeg; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
23 * thirdpel DSP functions
28 #include "libavutil/attributes.h"
32 #include "pel_template.c"
34 static inline void put_tpel_pixels_mc00_c(uint8_t *dst
, const uint8_t *src
,
35 int stride
, int width
, int height
)
39 put_pixels2_8_c(dst
, src
, stride
, height
);
42 put_pixels4_8_c(dst
, src
, stride
, height
);
45 put_pixels8_8_c(dst
, src
, stride
, height
);
48 put_pixels16_8_c(dst
, src
, stride
, height
);
53 static inline void put_tpel_pixels_mc10_c(uint8_t *dst
, const uint8_t *src
,
54 int stride
, int width
, int height
)
58 for (i
= 0; i
< height
; i
++) {
59 for (j
= 0; j
< width
; j
++)
60 dst
[j
] = ((2 * src
[j
] + src
[j
+ 1] + 1) *
67 static inline void put_tpel_pixels_mc20_c(uint8_t *dst
, const uint8_t *src
,
68 int stride
, int width
, int height
)
72 for (i
= 0; i
< height
; i
++) {
73 for (j
= 0; j
< width
; j
++)
74 dst
[j
] = ((src
[j
] + 2 * src
[j
+ 1] + 1) *
81 static inline void put_tpel_pixels_mc01_c(uint8_t *dst
, const uint8_t *src
,
82 int stride
, int width
, int height
)
86 for (i
= 0; i
< height
; i
++) {
87 for (j
= 0; j
< width
; j
++)
88 dst
[j
] = ((2 * src
[j
] + src
[j
+ stride
] + 1) *
95 static inline void put_tpel_pixels_mc11_c(uint8_t *dst
, const uint8_t *src
,
96 int stride
, int width
, int height
)
100 for (i
= 0; i
< height
; i
++) {
101 for (j
= 0; j
< width
; j
++)
102 dst
[j
] = ((4 * src
[j
] + 3 * src
[j
+ 1] +
103 3 * src
[j
+ stride
] + 2 * src
[j
+ stride
+ 1] + 6) *
110 static inline void put_tpel_pixels_mc12_c(uint8_t *dst
, const uint8_t *src
,
111 int stride
, int width
, int height
)
115 for (i
= 0; i
< height
; i
++) {
116 for (j
= 0; j
< width
; j
++)
117 dst
[j
] = ((3 * src
[j
] + 2 * src
[j
+ 1] +
118 4 * src
[j
+ stride
] + 3 * src
[j
+ stride
+ 1] + 6) *
125 static inline void put_tpel_pixels_mc02_c(uint8_t *dst
, const uint8_t *src
,
126 int stride
, int width
, int height
)
130 for (i
= 0; i
< height
; i
++) {
131 for (j
= 0; j
< width
; j
++)
132 dst
[j
] = ((src
[j
] + 2 * src
[j
+ stride
] + 1) *
139 static inline void put_tpel_pixels_mc21_c(uint8_t *dst
, const uint8_t *src
,
140 int stride
, int width
, int height
)
144 for (i
= 0; i
< height
; i
++) {
145 for (j
= 0; j
< width
; j
++)
146 dst
[j
] = ((3 * src
[j
] + 4 * src
[j
+ 1] +
147 2 * src
[j
+ stride
] + 3 * src
[j
+ stride
+ 1] + 6) *
154 static inline void put_tpel_pixels_mc22_c(uint8_t *dst
, const uint8_t *src
,
155 int stride
, int width
, int height
)
159 for (i
= 0; i
< height
; i
++) {
160 for (j
= 0; j
< width
; j
++)
161 dst
[j
] = ((2 * src
[j
] + 3 * src
[j
+ 1] +
162 3 * src
[j
+ stride
] + 4 * src
[j
+ stride
+ 1] + 6) *
169 static inline void avg_tpel_pixels_mc00_c(uint8_t *dst
, const uint8_t *src
,
170 int stride
, int width
, int height
)
174 avg_pixels2_8_c(dst
, src
, stride
, height
);
177 avg_pixels4_8_c(dst
, src
, stride
, height
);
180 avg_pixels8_8_c(dst
, src
, stride
, height
);
183 avg_pixels16_8_c(dst
, src
, stride
, height
);
188 static inline void avg_tpel_pixels_mc10_c(uint8_t *dst
, const uint8_t *src
,
189 int stride
, int width
, int height
)
193 for (i
= 0; i
< height
; i
++) {
194 for (j
= 0; j
< width
; j
++)
196 (((2 * src
[j
] + src
[j
+ 1] + 1) *
197 683) >> 11) + 1) >> 1;
203 static inline void avg_tpel_pixels_mc20_c(uint8_t *dst
, const uint8_t *src
,
204 int stride
, int width
, int height
)
208 for (i
= 0; i
< height
; i
++) {
209 for (j
= 0; j
< width
; j
++)
211 (((src
[j
] + 2 * src
[j
+ 1] + 1) *
212 683) >> 11) + 1) >> 1;
218 static inline void avg_tpel_pixels_mc01_c(uint8_t *dst
, const uint8_t *src
,
219 int stride
, int width
, int height
)
223 for (i
= 0; i
< height
; i
++) {
224 for (j
= 0; j
< width
; j
++)
226 (((2 * src
[j
] + src
[j
+ stride
] + 1) *
227 683) >> 11) + 1) >> 1;
233 static inline void avg_tpel_pixels_mc11_c(uint8_t *dst
, const uint8_t *src
,
234 int stride
, int width
, int height
)
238 for (i
= 0; i
< height
; i
++) {
239 for (j
= 0; j
< width
; j
++)
241 (((4 * src
[j
] + 3 * src
[j
+ 1] +
242 3 * src
[j
+ stride
] + 2 * src
[j
+ stride
+ 1] + 6) *
243 2731) >> 15) + 1) >> 1;
249 static inline void avg_tpel_pixels_mc12_c(uint8_t *dst
, const uint8_t *src
,
250 int stride
, int width
, int height
)
254 for (i
= 0; i
< height
; i
++) {
255 for (j
= 0; j
< width
; j
++)
257 (((3 * src
[j
] + 2 * src
[j
+ 1] +
258 4 * src
[j
+ stride
] + 3 * src
[j
+ stride
+ 1] + 6) *
259 2731) >> 15) + 1) >> 1;
265 static inline void avg_tpel_pixels_mc02_c(uint8_t *dst
, const uint8_t *src
,
266 int stride
, int width
, int height
)
270 for (i
= 0; i
< height
; i
++) {
271 for (j
= 0; j
< width
; j
++)
273 (((src
[j
] + 2 * src
[j
+ stride
] + 1) *
274 683) >> 11) + 1) >> 1;
280 static inline void avg_tpel_pixels_mc21_c(uint8_t *dst
, const uint8_t *src
,
281 int stride
, int width
, int height
)
285 for (i
= 0; i
< height
; i
++) {
286 for (j
= 0; j
< width
; j
++)
288 (((3 * src
[j
] + 4 * src
[j
+ 1] +
289 2 * src
[j
+ stride
] + 3 * src
[j
+ stride
+ 1] + 6) *
290 2731) >> 15) + 1) >> 1;
296 static inline void avg_tpel_pixels_mc22_c(uint8_t *dst
, const uint8_t *src
,
297 int stride
, int width
, int height
)
301 for (i
= 0; i
< height
; i
++) {
302 for (j
= 0; j
< width
; j
++)
304 (((2 * src
[j
] + 3 * src
[j
+ 1] +
305 3 * src
[j
+ stride
] + 4 * src
[j
+ stride
+ 1] + 6) *
306 2731) >> 15) + 1) >> 1;
312 av_cold
void ff_tpeldsp_init(TpelDSPContext
*c
)
314 c
->put_tpel_pixels_tab
[ 0] = put_tpel_pixels_mc00_c
;
315 c
->put_tpel_pixels_tab
[ 1] = put_tpel_pixels_mc10_c
;
316 c
->put_tpel_pixels_tab
[ 2] = put_tpel_pixels_mc20_c
;
317 c
->put_tpel_pixels_tab
[ 4] = put_tpel_pixels_mc01_c
;
318 c
->put_tpel_pixels_tab
[ 5] = put_tpel_pixels_mc11_c
;
319 c
->put_tpel_pixels_tab
[ 6] = put_tpel_pixels_mc21_c
;
320 c
->put_tpel_pixels_tab
[ 8] = put_tpel_pixels_mc02_c
;
321 c
->put_tpel_pixels_tab
[ 9] = put_tpel_pixels_mc12_c
;
322 c
->put_tpel_pixels_tab
[10] = put_tpel_pixels_mc22_c
;
324 c
->avg_tpel_pixels_tab
[ 0] = avg_tpel_pixels_mc00_c
;
325 c
->avg_tpel_pixels_tab
[ 1] = avg_tpel_pixels_mc10_c
;
326 c
->avg_tpel_pixels_tab
[ 2] = avg_tpel_pixels_mc20_c
;
327 c
->avg_tpel_pixels_tab
[ 4] = avg_tpel_pixels_mc01_c
;
328 c
->avg_tpel_pixels_tab
[ 5] = avg_tpel_pixels_mc11_c
;
329 c
->avg_tpel_pixels_tab
[ 6] = avg_tpel_pixels_mc21_c
;
330 c
->avg_tpel_pixels_tab
[ 8] = avg_tpel_pixels_mc02_c
;
331 c
->avg_tpel_pixels_tab
[ 9] = avg_tpel_pixels_mc12_c
;
332 c
->avg_tpel_pixels_tab
[10] = avg_tpel_pixels_mc22_c
;