FFmpeg
Loading...
Searching...
No Matches
vp8dsp_init.c
Go to the documentation of this file.
1/*
2 * VP8 DSP functions x86-optimized
3 * Copyright (c) 2010 Ronald S. Bultje <rsbultje@gmail.com>
4 * Copyright (c) 2010 Fiona Glaser <fiona@x264.com>
5 *
6 * This file is part of FFmpeg.
7 *
8 * FFmpeg is free software; you can redistribute it and/or
9 * modify it under the terms of the GNU Lesser General Public
10 * License as published by the Free Software Foundation; either
11 * version 2.1 of the License, or (at your option) any later version.
12 *
13 * FFmpeg is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 * Lesser General Public License for more details.
17 *
18 * You should have received a copy of the GNU Lesser General Public
19 * License along with FFmpeg; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21 */
22
24#include "libavutil/cpu.h"
26#include "libavutil/x86/cpu.h"
27#include "libavcodec/vp8dsp.h"
28
29/*
30 * MC functions
31 */
32void ff_put_vp8_epel8_h4_sse2 (uint8_t *dst, ptrdiff_t dststride,
33 const uint8_t *src, ptrdiff_t srcstride,
34 int height, int mx, int my);
35void ff_put_vp8_epel8_h6_sse2 (uint8_t *dst, ptrdiff_t dststride,
36 const uint8_t *src, ptrdiff_t srcstride,
37 int height, int mx, int my);
38void ff_put_vp8_epel8_v4_sse2 (uint8_t *dst, ptrdiff_t dststride,
39 const uint8_t *src, ptrdiff_t srcstride,
40 int height, int mx, int my);
41void ff_put_vp8_epel8_v6_sse2 (uint8_t *dst, ptrdiff_t dststride,
42 const uint8_t *src, ptrdiff_t srcstride,
43 int height, int mx, int my);
44
45void ff_put_vp8_epel4_h4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
46 const uint8_t *src, ptrdiff_t srcstride,
47 int height, int mx, int my);
48void ff_put_vp8_epel4_h6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
49 const uint8_t *src, ptrdiff_t srcstride,
50 int height, int mx, int my);
51void ff_put_vp8_epel4_v4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
52 const uint8_t *src, ptrdiff_t srcstride,
53 int height, int mx, int my);
54void ff_put_vp8_epel4_v6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
55 const uint8_t *src, ptrdiff_t srcstride,
56 int height, int mx, int my);
57void ff_put_vp8_epel8_h4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
58 const uint8_t *src, ptrdiff_t srcstride,
59 int height, int mx, int my);
60void ff_put_vp8_epel8_h6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
61 const uint8_t *src, ptrdiff_t srcstride,
62 int height, int mx, int my);
63void ff_put_vp8_epel8_v4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
64 const uint8_t *src, ptrdiff_t srcstride,
65 int height, int mx, int my);
66void ff_put_vp8_epel8_v6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
67 const uint8_t *src, ptrdiff_t srcstride,
68 int height, int mx, int my);
69
70void ff_put_vp8_bilinear8_h_sse2 (uint8_t *dst, ptrdiff_t dststride,
71 const uint8_t *src, ptrdiff_t srcstride,
72 int height, int mx, int my);
73void ff_put_vp8_bilinear4_h_ssse3 (uint8_t *dst, ptrdiff_t dststride,
74 const uint8_t *src, ptrdiff_t srcstride,
75 int height, int mx, int my);
76void ff_put_vp8_bilinear8_h_ssse3 (uint8_t *dst, ptrdiff_t dststride,
77 const uint8_t *src, ptrdiff_t srcstride,
78 int height, int mx, int my);
79
80void ff_put_vp8_bilinear8_v_sse2 (uint8_t *dst, ptrdiff_t dststride,
81 const uint8_t *src, ptrdiff_t srcstride,
82 int height, int mx, int my);
83void ff_put_vp8_bilinear4_v_ssse3 (uint8_t *dst, ptrdiff_t dststride,
84 const uint8_t *src, ptrdiff_t srcstride,
85 int height, int mx, int my);
86void ff_put_vp8_bilinear8_v_ssse3 (uint8_t *dst, ptrdiff_t dststride,
87 const uint8_t *src, ptrdiff_t srcstride,
88 int height, int mx, int my);
89
90
91void ff_put_vp8_pixels8_sse2(uint8_t *dst, ptrdiff_t dststride,
92 const uint8_t *src, ptrdiff_t srcstride,
93 int height, int mx, int my);
94void ff_put_vp8_pixels16_sse(uint8_t *dst, ptrdiff_t dststride,
95 const uint8_t *src, ptrdiff_t srcstride,
96 int height, int mx, int my);
97
98#define TAP_W16(OPT, FILTERTYPE, TAPTYPE) \
99static void ff_put_vp8_ ## FILTERTYPE ## 16_ ## TAPTYPE ## _ ## OPT( \
100 uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, \
101 ptrdiff_t srcstride, int height, int mx, int my) \
102{ \
103 ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
104 dst, dststride, src, srcstride, height, mx, my); \
105 ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
106 dst + 8, dststride, src + 8, srcstride, height, mx, my); \
107}
108
109TAP_W16(sse2, epel, h6)
110TAP_W16(sse2, epel, v6)
111TAP_W16(sse2, bilinear, h)
112TAP_W16(sse2, bilinear, v)
113
114TAP_W16(ssse3, epel, h6)
115TAP_W16(ssse3, epel, v6)
116TAP_W16(ssse3, bilinear, h)
117TAP_W16(ssse3, bilinear, v)
118
119#define HVTAP(OPT, ALIGN, TAPNUMX, TAPNUMY, SIZE, MAXHEIGHT) \
120static void ff_put_vp8_epel ## SIZE ## _h ## TAPNUMX ## v ## TAPNUMY ## _ ## OPT( \
121 uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, \
122 ptrdiff_t srcstride, int height, int mx, int my) \
123{ \
124 LOCAL_ALIGNED(ALIGN, uint8_t, tmp, [SIZE * (MAXHEIGHT + TAPNUMY - 1)]); \
125 uint8_t *tmpptr = tmp + SIZE * (TAPNUMY / 2 - 1); \
126 src -= srcstride * (TAPNUMY / 2 - 1); \
127 ff_put_vp8_epel ## SIZE ## _h ## TAPNUMX ## _ ## OPT( \
128 tmp, SIZE, src, srcstride, height + TAPNUMY - 1, mx, my); \
129 ff_put_vp8_epel ## SIZE ## _v ## TAPNUMY ## _ ## OPT( \
130 dst, dststride, tmpptr, SIZE, height, mx, my); \
131}
132
133#define HVTAPSSE2(x, y, w) \
134HVTAP(sse2, 16, x, y, w, 16) \
135HVTAP(ssse3, 16, x, y, w, 16)
136
137HVTAPSSE2(4, 4, 8)
138HVTAPSSE2(4, 6, 8)
139HVTAPSSE2(6, 4, 8)
140HVTAPSSE2(6, 6, 8)
141HVTAPSSE2(6, 6, 16)
142
143HVTAP(ssse3, 16, 4, 4, 4, 8)
144HVTAP(ssse3, 16, 4, 6, 4, 8)
145HVTAP(ssse3, 16, 6, 4, 4, 8)
146HVTAP(ssse3, 16, 6, 6, 4, 8)
147
148#define HVBILIN(OPT, ALIGN, SIZE, MAXHEIGHT) \
149static void ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT( \
150 uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, \
151 ptrdiff_t srcstride, int height, int mx, int my) \
152{ \
153 LOCAL_ALIGNED(ALIGN, uint8_t, tmp, [SIZE * (MAXHEIGHT + 2)]); \
154 ff_put_vp8_bilinear ## SIZE ## _h_ ## OPT( \
155 tmp, SIZE, src, srcstride, height + 1, mx, my); \
156 ff_put_vp8_bilinear ## SIZE ## _v_ ## OPT( \
157 dst, dststride, tmp, SIZE, height, mx, my); \
158}
159
160HVBILIN(sse2, 8, 8, 16)
161HVBILIN(sse2, 8, 16, 16)
162HVBILIN(ssse3, 8, 4, 8)
163HVBILIN(ssse3, 8, 8, 16)
164HVBILIN(ssse3, 8, 16, 16)
165
166void ff_vp8_idct_dc_add_sse2(uint8_t *dst, int16_t block[16],
167 ptrdiff_t stride);
168void ff_vp8_idct_dc_add_sse4(uint8_t *dst, int16_t block[16],
169 ptrdiff_t stride);
170void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, int16_t block[4][16],
171 ptrdiff_t stride);
172void ff_vp8_idct_dc_add4uv_sse2(uint8_t *dst, int16_t block[4][16],
173 ptrdiff_t stride);
174void ff_vp8_luma_dc_wht_sse(int16_t block[4][4][16], int16_t dc[16]);
175void ff_vp8_idct_add_sse(uint8_t *dst, int16_t block[16], ptrdiff_t stride);
176
177#define DECLARE_LOOP_FILTER(NAME) \
178void ff_vp8_v_loop_filter_simple_ ## NAME(uint8_t *dst, \
179 ptrdiff_t stride, \
180 int flim); \
181void ff_vp8_h_loop_filter_simple_ ## NAME(uint8_t *dst, \
182 ptrdiff_t stride, \
183 int flim); \
184void ff_vp8_v_loop_filter16y_inner_ ## NAME (uint8_t *dst, \
185 ptrdiff_t stride, \
186 int e, int i, int hvt); \
187void ff_vp8_h_loop_filter16y_inner_ ## NAME (uint8_t *dst, \
188 ptrdiff_t stride, \
189 int e, int i, int hvt); \
190void ff_vp8_v_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, \
191 uint8_t *dstV, \
192 ptrdiff_t s, \
193 int e, int i, int hvt); \
194void ff_vp8_h_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, \
195 uint8_t *dstV, \
196 ptrdiff_t s, \
197 int e, int i, int hvt); \
198void ff_vp8_v_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, \
199 ptrdiff_t stride, \
200 int e, int i, int hvt); \
201void ff_vp8_h_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, \
202 ptrdiff_t stride, \
203 int e, int i, int hvt); \
204void ff_vp8_v_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, \
205 uint8_t *dstV, \
206 ptrdiff_t s, \
207 int e, int i, int hvt); \
208void ff_vp8_h_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, \
209 uint8_t *dstV, \
210 ptrdiff_t s, \
211 int e, int i, int hvt);
212
216
217#define VP8_LUMA_MC_FUNC(IDX, SIZE, OPT) \
218 c->put_vp8_epel_pixels_tab[IDX][0][2] = ff_put_vp8_epel ## SIZE ## _h6_ ## OPT; \
219 c->put_vp8_epel_pixels_tab[IDX][2][0] = ff_put_vp8_epel ## SIZE ## _v6_ ## OPT; \
220 c->put_vp8_epel_pixels_tab[IDX][2][2] = ff_put_vp8_epel ## SIZE ## _h6v6_ ## OPT
221
222#define VP8_MC_FUNC(IDX, SIZE, OPT) \
223 c->put_vp8_epel_pixels_tab[IDX][0][1] = ff_put_vp8_epel ## SIZE ## _h4_ ## OPT; \
224 c->put_vp8_epel_pixels_tab[IDX][1][0] = ff_put_vp8_epel ## SIZE ## _v4_ ## OPT; \
225 c->put_vp8_epel_pixels_tab[IDX][1][1] = ff_put_vp8_epel ## SIZE ## _h4v4_ ## OPT; \
226 c->put_vp8_epel_pixels_tab[IDX][1][2] = ff_put_vp8_epel ## SIZE ## _h6v4_ ## OPT; \
227 c->put_vp8_epel_pixels_tab[IDX][2][1] = ff_put_vp8_epel ## SIZE ## _h4v6_ ## OPT; \
228 VP8_LUMA_MC_FUNC(IDX, SIZE, OPT)
229
230#define VP8_BILINEAR_MC_FUNC(IDX, SIZE, OPT) \
231 c->put_vp8_bilinear_pixels_tab[IDX][0][1] = ff_put_vp8_bilinear ## SIZE ## _h_ ## OPT; \
232 c->put_vp8_bilinear_pixels_tab[IDX][0][2] = ff_put_vp8_bilinear ## SIZE ## _h_ ## OPT; \
233 c->put_vp8_bilinear_pixels_tab[IDX][1][0] = ff_put_vp8_bilinear ## SIZE ## _v_ ## OPT; \
234 c->put_vp8_bilinear_pixels_tab[IDX][1][1] = ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT; \
235 c->put_vp8_bilinear_pixels_tab[IDX][1][2] = ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT; \
236 c->put_vp8_bilinear_pixels_tab[IDX][2][0] = ff_put_vp8_bilinear ## SIZE ## _v_ ## OPT; \
237 c->put_vp8_bilinear_pixels_tab[IDX][2][1] = ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT; \
238 c->put_vp8_bilinear_pixels_tab[IDX][2][2] = ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT
239
240
242{
244
245 if (EXTERNAL_SSE(cpu_flags)) {
246 c->put_vp8_epel_pixels_tab[0][0][0] =
247 c->put_vp8_bilinear_pixels_tab[0][0][0] = ff_put_vp8_pixels16_sse;
248 }
249
251 c->put_vp8_epel_pixels_tab[1][0][0] =
252 c->put_vp8_bilinear_pixels_tab[1][0][0] = ff_put_vp8_pixels8_sse2;
253 VP8_LUMA_MC_FUNC(0, 16, sse2);
254 VP8_MC_FUNC(1, 8, sse2);
255 VP8_BILINEAR_MC_FUNC(0, 16, sse2);
256 VP8_BILINEAR_MC_FUNC(1, 8, sse2);
257 }
258
259 /* note that 4-tap width=16 functions are missing because w=16
260 * is only used for luma, and luma is always a copy or sixtap. */
262 VP8_LUMA_MC_FUNC(0, 16, ssse3);
263 VP8_MC_FUNC(1, 8, ssse3);
264 VP8_MC_FUNC(2, 4, ssse3);
265 VP8_BILINEAR_MC_FUNC(0, 16, ssse3);
266 VP8_BILINEAR_MC_FUNC(1, 8, ssse3);
267 VP8_BILINEAR_MC_FUNC(2, 4, ssse3);
268 }
269}
270
272{
274
275 if (EXTERNAL_SSE(cpu_flags)) {
276 c->vp8_idct_add = ff_vp8_idct_add_sse;
277 c->vp8_luma_dc_wht = ff_vp8_luma_dc_wht_sse;
278 }
279
281 c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_sse2;
282
283 c->vp8_v_loop_filter16y_inner = ff_vp8_v_loop_filter16y_inner_sse2;
284 c->vp8_v_loop_filter8uv_inner = ff_vp8_v_loop_filter8uv_inner_sse2;
285
286 c->vp8_v_loop_filter16y = ff_vp8_v_loop_filter16y_mbedge_sse2;
287 c->vp8_v_loop_filter8uv = ff_vp8_v_loop_filter8uv_mbedge_sse2;
288 }
289
291 c->vp8_idct_dc_add = ff_vp8_idct_dc_add_sse2;
292 c->vp8_idct_dc_add4y = ff_vp8_idct_dc_add4y_sse2;
293 c->vp8_idct_dc_add4uv = ff_vp8_idct_dc_add4uv_sse2;
294
295 c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_sse2;
296
297 c->vp8_h_loop_filter16y_inner = ff_vp8_h_loop_filter16y_inner_sse2;
298 c->vp8_h_loop_filter8uv_inner = ff_vp8_h_loop_filter8uv_inner_sse2;
299
300 c->vp8_h_loop_filter16y = ff_vp8_h_loop_filter16y_mbedge_sse2;
301 c->vp8_h_loop_filter8uv = ff_vp8_h_loop_filter8uv_mbedge_sse2;
302 }
303
305 c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_ssse3;
306 c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_ssse3;
307
308 c->vp8_v_loop_filter16y_inner = ff_vp8_v_loop_filter16y_inner_ssse3;
309 c->vp8_h_loop_filter16y_inner = ff_vp8_h_loop_filter16y_inner_ssse3;
310 c->vp8_v_loop_filter8uv_inner = ff_vp8_v_loop_filter8uv_inner_ssse3;
311 c->vp8_h_loop_filter8uv_inner = ff_vp8_h_loop_filter8uv_inner_ssse3;
312
313 c->vp8_v_loop_filter16y = ff_vp8_v_loop_filter16y_mbedge_ssse3;
314 c->vp8_h_loop_filter16y = ff_vp8_h_loop_filter16y_mbedge_ssse3;
315 c->vp8_v_loop_filter8uv = ff_vp8_v_loop_filter8uv_mbedge_ssse3;
316 c->vp8_h_loop_filter8uv = ff_vp8_h_loop_filter8uv_mbedge_ssse3;
317 }
318
320 c->vp8_idct_dc_add = ff_vp8_idct_dc_add_sse4;
321
322 c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_sse4;
323 c->vp8_h_loop_filter16y = ff_vp8_h_loop_filter16y_mbedge_sse4;
324 c->vp8_h_loop_filter8uv = ff_vp8_h_loop_filter8uv_mbedge_sse4;
325 }
326}
uint8_t ptrdiff_t const uint8_t ptrdiff_t int intptr_t intptr_t int int16_t const uint8_t ptrdiff_t srcstride
Definition dsp.h:88
uint8_t ptrdiff_t const uint8_t ptrdiff_t int intptr_t intptr_t my
Definition dsp.h:57
uint8_t ptrdiff_t const uint8_t ptrdiff_t int intptr_t mx
Definition dsp.h:57
uint8_t ptrdiff_t const uint8_t ptrdiff_t int intptr_t intptr_t int int16_t * dst
Definition dsp.h:87
static int16_t block[64]
Definition dct.c:125
Macro definitions for various function/variable attributes.
#define av_cold
Definition attributes.h:117
static atomic_int cpu_flags
Definition cpu.c:56
int av_get_cpu_flags(void)
Return the flags which specify extensions supported by the CPU.
Definition cpu.c:109
#define EXTERNAL_SSE(flags)
Definition cpu.h:52
#define EXTERNAL_SSSE3(flags)
Definition cpu.h:59
#define EXTERNAL_SSE2_SLOW(flags)
Definition cpu.h:55
#define EXTERNAL_SSE4(flags)
Definition cpu.h:62
#define EXTERNAL_SSE2(flags)
Definition cpu.h:53
#define stride
#define src
Definition vp8dsp.c:248
#define height
Definition dsp.h:89
VP8 compatible video decoder.
static double c[64]
void ff_put_vp8_epel8_h4_sse2(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
#define HVTAPSSE2(x, y, w)
#define HVTAP(OPT, ALIGN, TAPNUMX, TAPNUMY, SIZE, MAXHEIGHT)
#define DECLARE_LOOP_FILTER(NAME)
#define HVBILIN(OPT, ALIGN, SIZE, MAXHEIGHT)
void ff_put_vp8_bilinear8_h_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
#define VP8_LUMA_MC_FUNC(IDX, SIZE, OPT)
void ff_put_vp8_epel8_v4_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_vp8_luma_dc_wht_sse(int16_t block[4][4][16], int16_t dc[16])
void ff_put_vp8_epel8_v4_sse2(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_epel4_v4_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_epel8_h6_sse2(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_epel8_h6_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_pixels16_sse(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_pixels8_sse2(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_vp8_idct_dc_add_sse2(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
void ff_put_vp8_epel8_v6_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_epel8_v6_sse2(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride)
#define TAP_W16(OPT, FILTERTYPE, TAPTYPE)
Definition vp8dsp_init.c:98
void ff_put_vp8_bilinear4_h_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_bilinear8_v_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_epel4_h6_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_vp8_idct_add_sse(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
void ff_put_vp8_epel4_h4_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
av_cold void ff_vp8dsp_init_x86(VP8DSPContext *c)
void ff_put_vp8_bilinear8_v_sse2(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
#define VP8_MC_FUNC(IDX, SIZE, OPT)
void ff_put_vp8_bilinear4_v_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_put_vp8_epel4_v6_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
void ff_vp8_idct_dc_add_sse4(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
void ff_put_vp8_epel8_h4_ssse3(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
av_cold void ff_vp78dsp_init_x86(VP8DSPContext *c)
void ff_put_vp8_bilinear8_h_sse2(uint8_t *dst, ptrdiff_t dststride, const uint8_t *src, ptrdiff_t srcstride, int height, int mx, int my)
#define VP8_BILINEAR_MC_FUNC(IDX, SIZE, OPT)
void ff_vp8_idct_dc_add4uv_sse2(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride)