2 * Copyright (C) 2001-2011 Michael Niedermayer <michaelni@gmx.at>
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 #include "libswscale/swscale.h"
24 #include "libswscale/swscale_internal.h"
25 #include "libavutil/attributes.h"
26 #include "libavutil/avassert.h"
27 #include "libavutil/intreadwrite.h"
28 #include "libavutil/x86/asm.h"
29 #include "libavutil/x86/cpu.h"
30 #include "libavutil/cpu.h"
31 #include "libavutil/pixdesc.h"
37 DECLARE_ASM_CONST(8, uint64_t, bF8)= 0xF8F8F8F8F8F8F8F8LL;
38 DECLARE_ASM_CONST(8, uint64_t, bFC)= 0xFCFCFCFCFCFCFCFCLL;
39 DECLARE_ASM_CONST(8, uint64_t, w10)= 0x0010001000100010LL;
40 DECLARE_ASM_CONST(8, uint64_t, w02)= 0x0002000200020002LL;
42 const DECLARE_ALIGNED(8, uint64_t, ff_dither4)[2] = {
44 0x0200020002000200LL,};
46 const DECLARE_ALIGNED(8, uint64_t, ff_dither8)[2] = {
48 0x0004000400040004LL,};
50 DECLARE_ASM_CONST(8, uint64_t, b16Mask)= 0x001F001F001F001FLL;
51 DECLARE_ASM_CONST(8, uint64_t, g16Mask)= 0x07E007E007E007E0LL;
52 DECLARE_ASM_CONST(8, uint64_t, r16Mask)= 0xF800F800F800F800LL;
53 DECLARE_ASM_CONST(8, uint64_t, b15Mask)= 0x001F001F001F001FLL;
54 DECLARE_ASM_CONST(8, uint64_t, g15Mask)= 0x03E003E003E003E0LL;
55 DECLARE_ASM_CONST(8, uint64_t, r15Mask)= 0x7C007C007C007C00LL;
57 DECLARE_ALIGNED(8, const uint64_t, ff_M24A) = 0x00FF0000FF0000FFLL;
58 DECLARE_ALIGNED(8, const uint64_t, ff_M24B) = 0xFF0000FF0000FF00LL;
59 DECLARE_ALIGNED(8, const uint64_t, ff_M24C) = 0x0000FF0000FF0000LL;
61 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2YOffset) = 0x1010101010101010ULL;
62 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2UVOffset) = 0x8080808080808080ULL;
63 DECLARE_ALIGNED(8, const uint64_t, ff_w1111) = 0x0001000100010001ULL;
69 #define COMPILE_TEMPLATE_MMXEXT 0
70 #define RENAME(a) a ## _mmx
71 #include "swscale_template.c"
75 #if HAVE_MMXEXT_INLINE
77 #undef COMPILE_TEMPLATE_MMXEXT
78 #define COMPILE_TEMPLATE_MMXEXT 1
79 #define RENAME(a) a ## _mmxext
80 #include "swscale_template.c"
83 void ff_updateMMXDitherTables(SwsContext *c, int dstY, int lumBufIndex, int chrBufIndex,
84 int lastInLumBuf, int lastInChrBuf)
86 const int dstH= c->dstH;
87 const int flags= c->flags;
89 SwsPlane *lumPlane = &c->slice[c->numSlice-2].plane[0];
90 SwsPlane *chrUPlane = &c->slice[c->numSlice-2].plane[1];
91 SwsPlane *alpPlane = &c->slice[c->numSlice-2].plane[3];
93 int16_t **lumPixBuf= c->lumPixBuf;
94 int16_t **chrUPixBuf= c->chrUPixBuf;
95 int16_t **alpPixBuf= c->alpPixBuf;
96 const int vLumBufSize= c->vLumBufSize;
97 const int vChrBufSize= c->vChrBufSize;
99 int hasAlpha = c->alpPixBuf != NULL;
100 int32_t *vLumFilterPos= c->vLumFilterPos;
101 int32_t *vChrFilterPos= c->vChrFilterPos;
102 int16_t *vLumFilter= c->vLumFilter;
103 int16_t *vChrFilter= c->vChrFilter;
104 int32_t *lumMmxFilter= c->lumMmxFilter;
105 int32_t *chrMmxFilter= c->chrMmxFilter;
106 int32_t av_unused *alpMmxFilter= c->alpMmxFilter;
107 const int vLumFilterSize= c->vLumFilterSize;
108 const int vChrFilterSize= c->vChrFilterSize;
109 const int chrDstY= dstY>>c->chrDstVSubSample;
110 const int firstLumSrcY= vLumFilterPos[dstY]; //First line needed as input
111 const int firstChrSrcY= vChrFilterPos[chrDstY]; //First line needed as input
113 c->blueDither= ff_dither8[dstY&1];
114 if (c->dstFormat == AV_PIX_FMT_RGB555 || c->dstFormat == AV_PIX_FMT_BGR555)
115 c->greenDither= ff_dither8[dstY&1];
117 c->greenDither= ff_dither4[dstY&1];
118 c->redDither= ff_dither8[(dstY+1)&1];
119 if (dstY < dstH - 2) {
121 const int16_t **lumSrcPtr = (const int16_t **)(void*) lumPlane->line + firstLumSrcY - lumPlane->sliceY;
122 const int16_t **chrUSrcPtr = (const int16_t **)(void*) chrUPlane->line + firstChrSrcY - chrUPlane->sliceY;
123 const int16_t **alpSrcPtr = (CONFIG_SWSCALE_ALPHA && c->alpPixBuf) ? (const int16_t **)(void*) alpPlane->line + firstLumSrcY - alpPlane->sliceY : NULL;
125 const int16_t **lumSrcPtr= (const int16_t **)(void*) lumPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize;
126 const int16_t **chrUSrcPtr= (const int16_t **)(void*) chrUPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
127 const int16_t **alpSrcPtr= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? (const int16_t **)(void*) alpPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize : NULL;
130 if (firstLumSrcY < 0 || firstLumSrcY + vLumFilterSize > c->srcH) {
132 const int16_t **tmpY = (const int16_t **) lumPlane->tmp;
134 const int16_t **tmpY = (const int16_t **) lumPixBuf + 2 * vLumBufSize;
136 int neg = -firstLumSrcY, i, end = FFMIN(c->srcH - firstLumSrcY, vLumFilterSize);
137 for (i = 0; i < neg; i++)
138 tmpY[i] = lumSrcPtr[neg];
139 for ( ; i < end; i++)
140 tmpY[i] = lumSrcPtr[i];
141 for ( ; i < vLumFilterSize; i++)
147 const int16_t **tmpA = (const int16_t **) alpPlane->tmp;
149 const int16_t **tmpA = (const int16_t **) alpPixBuf + 2 * vLumBufSize;
151 for (i = 0; i < neg; i++)
152 tmpA[i] = alpSrcPtr[neg];
153 for ( ; i < end; i++)
154 tmpA[i] = alpSrcPtr[i];
155 for ( ; i < vLumFilterSize; i++)
156 tmpA[i] = tmpA[i - 1];
160 if (firstChrSrcY < 0 || firstChrSrcY + vChrFilterSize > c->chrSrcH) {
162 const int16_t **tmpU = (const int16_t **) chrUPlane->tmp;
164 const int16_t **tmpU = (const int16_t **) chrUPixBuf + 2 * vChrBufSize;
166 int neg = -firstChrSrcY, i, end = FFMIN(c->chrSrcH - firstChrSrcY, vChrFilterSize);
167 for (i = 0; i < neg; i++) {
168 tmpU[i] = chrUSrcPtr[neg];
170 for ( ; i < end; i++) {
171 tmpU[i] = chrUSrcPtr[i];
173 for ( ; i < vChrFilterSize; i++) {
174 tmpU[i] = tmpU[i - 1];
179 if (flags & SWS_ACCURATE_RND) {
180 int s= APCK_SIZE / 8;
181 for (i=0; i<vLumFilterSize; i+=2) {
182 *(const void**)&lumMmxFilter[s*i ]= lumSrcPtr[i ];
183 *(const void**)&lumMmxFilter[s*i+APCK_PTR2/4 ]= lumSrcPtr[i+(vLumFilterSize>1)];
184 lumMmxFilter[s*i+APCK_COEF/4 ]=
185 lumMmxFilter[s*i+APCK_COEF/4+1]= vLumFilter[dstY*vLumFilterSize + i ]
186 + (vLumFilterSize>1 ? vLumFilter[dstY*vLumFilterSize + i + 1]<<16 : 0);
187 if (CONFIG_SWSCALE_ALPHA && hasAlpha) {
188 *(const void**)&alpMmxFilter[s*i ]= alpSrcPtr[i ];
189 *(const void**)&alpMmxFilter[s*i+APCK_PTR2/4 ]= alpSrcPtr[i+(vLumFilterSize>1)];
190 alpMmxFilter[s*i+APCK_COEF/4 ]=
191 alpMmxFilter[s*i+APCK_COEF/4+1]= lumMmxFilter[s*i+APCK_COEF/4 ];
194 for (i=0; i<vChrFilterSize; i+=2) {
195 *(const void**)&chrMmxFilter[s*i ]= chrUSrcPtr[i ];
196 *(const void**)&chrMmxFilter[s*i+APCK_PTR2/4 ]= chrUSrcPtr[i+(vChrFilterSize>1)];
197 chrMmxFilter[s*i+APCK_COEF/4 ]=
198 chrMmxFilter[s*i+APCK_COEF/4+1]= vChrFilter[chrDstY*vChrFilterSize + i ]
199 + (vChrFilterSize>1 ? vChrFilter[chrDstY*vChrFilterSize + i + 1]<<16 : 0);
202 for (i=0; i<vLumFilterSize; i++) {
203 *(const void**)&lumMmxFilter[4*i+0]= lumSrcPtr[i];
206 ((uint16_t)vLumFilter[dstY*vLumFilterSize + i])*0x10001U;
207 if (CONFIG_SWSCALE_ALPHA && hasAlpha) {
208 *(const void**)&alpMmxFilter[4*i+0]= alpSrcPtr[i];
210 alpMmxFilter[4*i+3]= lumMmxFilter[4*i+2];
213 for (i=0; i<vChrFilterSize; i++) {
214 *(const void**)&chrMmxFilter[4*i+0]= chrUSrcPtr[i];
217 ((uint16_t)vChrFilter[chrDstY*vChrFilterSize + i])*0x10001U;
224 static void yuv2yuvX_sse3(const int16_t *filter, int filterSize,
225 const int16_t **src, uint8_t *dest, int dstW,
226 const uint8_t *dither, int offset)
228 if(((uintptr_t)dest) & 15){
229 yuv2yuvX_mmxext(filter, filterSize, src, dest, dstW, dither, offset);
233 #define MAIN_FUNCTION \
234 "pxor %%xmm0, %%xmm0 \n\t" \
235 "punpcklbw %%xmm0, %%xmm3 \n\t" \
236 "movd %4, %%xmm1 \n\t" \
237 "punpcklwd %%xmm1, %%xmm1 \n\t" \
238 "punpckldq %%xmm1, %%xmm1 \n\t" \
239 "punpcklqdq %%xmm1, %%xmm1 \n\t" \
240 "psllw $3, %%xmm1 \n\t" \
241 "paddw %%xmm1, %%xmm3 \n\t" \
242 "psraw $4, %%xmm3 \n\t" \
243 "movdqa %%xmm3, %%xmm4 \n\t" \
244 "movdqa %%xmm3, %%xmm7 \n\t" \
245 "movl %3, %%ecx \n\t" \
246 "mov %0, %%"REG_d" \n\t"\
247 "mov (%%"REG_d"), %%"REG_S" \n\t"\
248 ".p2align 4 \n\t" /* FIXME Unroll? */\
250 "movddup 8(%%"REG_d"), %%xmm0 \n\t" /* filterCoeff */\
251 "movdqa (%%"REG_S", %%"REG_c", 2), %%xmm2 \n\t" /* srcData */\
252 "movdqa 16(%%"REG_S", %%"REG_c", 2), %%xmm5 \n\t" /* srcData */\
253 "add $16, %%"REG_d" \n\t"\
254 "mov (%%"REG_d"), %%"REG_S" \n\t"\
255 "test %%"REG_S", %%"REG_S" \n\t"\
256 "pmulhw %%xmm0, %%xmm2 \n\t"\
257 "pmulhw %%xmm0, %%xmm5 \n\t"\
258 "paddw %%xmm2, %%xmm3 \n\t"\
259 "paddw %%xmm5, %%xmm4 \n\t"\
261 "psraw $3, %%xmm3 \n\t"\
262 "psraw $3, %%xmm4 \n\t"\
263 "packuswb %%xmm4, %%xmm3 \n\t"\
264 "movntdq %%xmm3, (%1, %%"REG_c")\n\t"\
265 "add $16, %%"REG_c" \n\t"\
266 "cmp %2, %%"REG_c" \n\t"\
267 "movdqa %%xmm7, %%xmm3 \n\t" \
268 "movdqa %%xmm7, %%xmm4 \n\t" \
269 "mov %0, %%"REG_d" \n\t"\
270 "mov (%%"REG_d"), %%"REG_S" \n\t"\
275 "movq %5, %%xmm3 \n\t"
276 "movdqa %%xmm3, %%xmm4 \n\t"
277 "psrlq $24, %%xmm3 \n\t"
278 "psllq $40, %%xmm4 \n\t"
279 "por %%xmm4, %%xmm3 \n\t"
282 "r" (dest-offset), "g" ((x86_reg)(dstW+offset)), "m" (offset),
283 "m"(filterSize), "m"(((uint64_t *) dither)[0])
284 : XMM_CLOBBERS("%xmm0" , "%xmm1" , "%xmm2" , "%xmm3" , "%xmm4" , "%xmm5" , "%xmm7" ,)
285 "%"REG_d, "%"REG_S, "%"REG_c
289 "movq %5, %%xmm3 \n\t"
292 "r" (dest-offset), "g" ((x86_reg)(dstW+offset)), "m" (offset),
293 "m"(filterSize), "m"(((uint64_t *) dither)[0])
294 : XMM_CLOBBERS("%xmm0" , "%xmm1" , "%xmm2" , "%xmm3" , "%xmm4" , "%xmm5" , "%xmm7" ,)
295 "%"REG_d, "%"REG_S, "%"REG_c
301 #endif /* HAVE_INLINE_ASM */
303 #define SCALE_FUNC(filter_n, from_bpc, to_bpc, opt) \
304 void ff_hscale ## from_bpc ## to ## to_bpc ## _ ## filter_n ## _ ## opt( \
305 SwsContext *c, int16_t *data, \
306 int dstW, const uint8_t *src, \
307 const int16_t *filter, \
308 const int32_t *filterPos, int filterSize)
310 #define SCALE_FUNCS(filter_n, opt) \
311 SCALE_FUNC(filter_n, 8, 15, opt); \
312 SCALE_FUNC(filter_n, 9, 15, opt); \
313 SCALE_FUNC(filter_n, 10, 15, opt); \
314 SCALE_FUNC(filter_n, 12, 15, opt); \
315 SCALE_FUNC(filter_n, 14, 15, opt); \
316 SCALE_FUNC(filter_n, 16, 15, opt); \
317 SCALE_FUNC(filter_n, 8, 19, opt); \
318 SCALE_FUNC(filter_n, 9, 19, opt); \
319 SCALE_FUNC(filter_n, 10, 19, opt); \
320 SCALE_FUNC(filter_n, 12, 19, opt); \
321 SCALE_FUNC(filter_n, 14, 19, opt); \
322 SCALE_FUNC(filter_n, 16, 19, opt)
324 #define SCALE_FUNCS_MMX(opt) \
325 SCALE_FUNCS(4, opt); \
326 SCALE_FUNCS(8, opt); \
329 #define SCALE_FUNCS_SSE(opt) \
330 SCALE_FUNCS(4, opt); \
331 SCALE_FUNCS(8, opt); \
332 SCALE_FUNCS(X4, opt); \
336 SCALE_FUNCS_MMX(mmx);
338 SCALE_FUNCS_SSE(sse2);
339 SCALE_FUNCS_SSE(ssse3);
340 SCALE_FUNCS_SSE(sse4);
342 #define VSCALEX_FUNC(size, opt) \
343 void ff_yuv2planeX_ ## size ## _ ## opt(const int16_t *filter, int filterSize, \
344 const int16_t **src, uint8_t *dest, int dstW, \
345 const uint8_t *dither, int offset)
346 #define VSCALEX_FUNCS(opt) \
347 VSCALEX_FUNC(8, opt); \
348 VSCALEX_FUNC(9, opt); \
349 VSCALEX_FUNC(10, opt)
352 VSCALEX_FUNCS(mmxext);
356 VSCALEX_FUNC(16, sse4);
359 #define VSCALE_FUNC(size, opt) \
360 void ff_yuv2plane1_ ## size ## _ ## opt(const int16_t *src, uint8_t *dst, int dstW, \
361 const uint8_t *dither, int offset)
362 #define VSCALE_FUNCS(opt1, opt2) \
363 VSCALE_FUNC(8, opt1); \
364 VSCALE_FUNC(9, opt2); \
365 VSCALE_FUNC(10, opt2); \
366 VSCALE_FUNC(16, opt1)
369 VSCALE_FUNCS(mmx, mmxext);
371 VSCALE_FUNCS(sse2, sse2);
372 VSCALE_FUNC(16, sse4);
373 VSCALE_FUNCS(avx, avx);
375 #define INPUT_Y_FUNC(fmt, opt) \
376 void ff_ ## fmt ## ToY_ ## opt(uint8_t *dst, const uint8_t *src, \
377 const uint8_t *unused1, const uint8_t *unused2, \
378 int w, uint32_t *unused)
379 #define INPUT_UV_FUNC(fmt, opt) \
380 void ff_ ## fmt ## ToUV_ ## opt(uint8_t *dstU, uint8_t *dstV, \
381 const uint8_t *unused0, \
382 const uint8_t *src1, \
383 const uint8_t *src2, \
384 int w, uint32_t *unused)
385 #define INPUT_FUNC(fmt, opt) \
386 INPUT_Y_FUNC(fmt, opt); \
387 INPUT_UV_FUNC(fmt, opt)
388 #define INPUT_FUNCS(opt) \
389 INPUT_FUNC(uyvy, opt); \
390 INPUT_FUNC(yuyv, opt); \
391 INPUT_UV_FUNC(nv12, opt); \
392 INPUT_UV_FUNC(nv21, opt); \
393 INPUT_FUNC(rgba, opt); \
394 INPUT_FUNC(bgra, opt); \
395 INPUT_FUNC(argb, opt); \
396 INPUT_FUNC(abgr, opt); \
397 INPUT_FUNC(rgb24, opt); \
398 INPUT_FUNC(bgr24, opt)
407 av_cold void ff_sws_init_swscale_x86(SwsContext *c)
409 int cpu_flags = av_get_cpu_flags();
412 if (INLINE_MMX(cpu_flags))
413 sws_init_swscale_mmx(c);
415 #if HAVE_MMXEXT_INLINE
416 if (INLINE_MMXEXT(cpu_flags))
417 sws_init_swscale_mmxext(c);
418 if (cpu_flags & AV_CPU_FLAG_SSE3){
419 if(c->use_mmx_vfilter && !(c->flags & SWS_ACCURATE_RND))
420 c->yuv2planeX = yuv2yuvX_sse3;
424 #define ASSIGN_SCALE_FUNC2(hscalefn, filtersize, opt1, opt2) do { \
425 if (c->srcBpc == 8) { \
426 hscalefn = c->dstBpc <= 14 ? ff_hscale8to15_ ## filtersize ## _ ## opt2 : \
427 ff_hscale8to19_ ## filtersize ## _ ## opt1; \
428 } else if (c->srcBpc == 9) { \
429 hscalefn = c->dstBpc <= 14 ? ff_hscale9to15_ ## filtersize ## _ ## opt2 : \
430 ff_hscale9to19_ ## filtersize ## _ ## opt1; \
431 } else if (c->srcBpc == 10) { \
432 hscalefn = c->dstBpc <= 14 ? ff_hscale10to15_ ## filtersize ## _ ## opt2 : \
433 ff_hscale10to19_ ## filtersize ## _ ## opt1; \
434 } else if (c->srcBpc == 12) { \
435 hscalefn = c->dstBpc <= 14 ? ff_hscale12to15_ ## filtersize ## _ ## opt2 : \
436 ff_hscale12to19_ ## filtersize ## _ ## opt1; \
437 } else if (c->srcBpc == 14 || ((c->srcFormat==AV_PIX_FMT_PAL8||isAnyRGB(c->srcFormat)) && av_pix_fmt_desc_get(c->srcFormat)->comp[0].depth<16)) { \
438 hscalefn = c->dstBpc <= 14 ? ff_hscale14to15_ ## filtersize ## _ ## opt2 : \
439 ff_hscale14to19_ ## filtersize ## _ ## opt1; \
440 } else { /* c->srcBpc == 16 */ \
441 av_assert0(c->srcBpc == 16);\
442 hscalefn = c->dstBpc <= 14 ? ff_hscale16to15_ ## filtersize ## _ ## opt2 : \
443 ff_hscale16to19_ ## filtersize ## _ ## opt1; \
446 #define ASSIGN_MMX_SCALE_FUNC(hscalefn, filtersize, opt1, opt2) \
447 switch (filtersize) { \
448 case 4: ASSIGN_SCALE_FUNC2(hscalefn, 4, opt1, opt2); break; \
449 case 8: ASSIGN_SCALE_FUNC2(hscalefn, 8, opt1, opt2); break; \
450 default: ASSIGN_SCALE_FUNC2(hscalefn, X, opt1, opt2); break; \
452 #define ASSIGN_VSCALEX_FUNC(vscalefn, opt, do_16_case, condition_8bit) \
454 case 16: do_16_case; break; \
455 case 10: if (!isBE(c->dstFormat)) vscalefn = ff_yuv2planeX_10_ ## opt; break; \
456 case 9: if (!isBE(c->dstFormat)) vscalefn = ff_yuv2planeX_9_ ## opt; break; \
457 case 8: if ((condition_8bit) && !c->use_mmx_vfilter) vscalefn = ff_yuv2planeX_8_ ## opt; break; \
459 #define ASSIGN_VSCALE_FUNC(vscalefn, opt1, opt2, opt2chk) \
461 case 16: if (!isBE(c->dstFormat)) vscalefn = ff_yuv2plane1_16_ ## opt1; break; \
462 case 10: if (!isBE(c->dstFormat) && opt2chk) vscalefn = ff_yuv2plane1_10_ ## opt2; break; \
463 case 9: if (!isBE(c->dstFormat) && opt2chk) vscalefn = ff_yuv2plane1_9_ ## opt2; break; \
464 case 8: vscalefn = ff_yuv2plane1_8_ ## opt1; break; \
465 default: av_assert0(c->dstBpc>8); \
467 #define case_rgb(x, X, opt) \
468 case AV_PIX_FMT_ ## X: \
469 c->lumToYV12 = ff_ ## x ## ToY_ ## opt; \
470 if (!c->chrSrcHSubSample) \
471 c->chrToYV12 = ff_ ## x ## ToUV_ ## opt; \
474 if (EXTERNAL_MMX(cpu_flags)) {
475 ASSIGN_MMX_SCALE_FUNC(c->hyScale, c->hLumFilterSize, mmx, mmx);
476 ASSIGN_MMX_SCALE_FUNC(c->hcScale, c->hChrFilterSize, mmx, mmx);
477 ASSIGN_VSCALE_FUNC(c->yuv2plane1, mmx, mmxext, cpu_flags & AV_CPU_FLAG_MMXEXT);
479 switch (c->srcFormat) {
481 c->lumToYV12 = ff_yuyvToY_mmx;
483 c->alpToYV12 = ff_uyvyToY_mmx;
485 case AV_PIX_FMT_YUYV422:
486 c->lumToYV12 = ff_yuyvToY_mmx;
487 c->chrToYV12 = ff_yuyvToUV_mmx;
489 case AV_PIX_FMT_UYVY422:
490 c->lumToYV12 = ff_uyvyToY_mmx;
491 c->chrToYV12 = ff_uyvyToUV_mmx;
493 case AV_PIX_FMT_NV12:
494 c->chrToYV12 = ff_nv12ToUV_mmx;
496 case AV_PIX_FMT_NV21:
497 c->chrToYV12 = ff_nv21ToUV_mmx;
499 case_rgb(rgb24, RGB24, mmx);
500 case_rgb(bgr24, BGR24, mmx);
501 case_rgb(bgra, BGRA, mmx);
502 case_rgb(rgba, RGBA, mmx);
503 case_rgb(abgr, ABGR, mmx);
504 case_rgb(argb, ARGB, mmx);
509 if (EXTERNAL_MMXEXT(cpu_flags)) {
510 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, mmxext, , 1);
512 #endif /* ARCH_X86_32 */
513 #define ASSIGN_SSE_SCALE_FUNC(hscalefn, filtersize, opt1, opt2) \
514 switch (filtersize) { \
515 case 4: ASSIGN_SCALE_FUNC2(hscalefn, 4, opt1, opt2); break; \
516 case 8: ASSIGN_SCALE_FUNC2(hscalefn, 8, opt1, opt2); break; \
517 default: if (filtersize & 4) ASSIGN_SCALE_FUNC2(hscalefn, X4, opt1, opt2); \
518 else ASSIGN_SCALE_FUNC2(hscalefn, X8, opt1, opt2); \
521 if (EXTERNAL_SSE2(cpu_flags)) {
522 ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, sse2, sse2);
523 ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, sse2, sse2);
524 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, sse2, ,
525 HAVE_ALIGNED_STACK || ARCH_X86_64);
526 ASSIGN_VSCALE_FUNC(c->yuv2plane1, sse2, sse2, 1);
528 switch (c->srcFormat) {
530 c->lumToYV12 = ff_yuyvToY_sse2;
532 c->alpToYV12 = ff_uyvyToY_sse2;
534 case AV_PIX_FMT_YUYV422:
535 c->lumToYV12 = ff_yuyvToY_sse2;
536 c->chrToYV12 = ff_yuyvToUV_sse2;
538 case AV_PIX_FMT_UYVY422:
539 c->lumToYV12 = ff_uyvyToY_sse2;
540 c->chrToYV12 = ff_uyvyToUV_sse2;
542 case AV_PIX_FMT_NV12:
543 c->chrToYV12 = ff_nv12ToUV_sse2;
545 case AV_PIX_FMT_NV21:
546 c->chrToYV12 = ff_nv21ToUV_sse2;
548 case_rgb(rgb24, RGB24, sse2);
549 case_rgb(bgr24, BGR24, sse2);
550 case_rgb(bgra, BGRA, sse2);
551 case_rgb(rgba, RGBA, sse2);
552 case_rgb(abgr, ABGR, sse2);
553 case_rgb(argb, ARGB, sse2);
558 if (EXTERNAL_SSSE3(cpu_flags)) {
559 ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, ssse3, ssse3);
560 ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, ssse3, ssse3);
561 switch (c->srcFormat) {
562 case_rgb(rgb24, RGB24, ssse3);
563 case_rgb(bgr24, BGR24, ssse3);
568 if (EXTERNAL_SSE4(cpu_flags)) {
569 /* Xto15 don't need special sse4 functions */
570 ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, sse4, ssse3);
571 ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, sse4, ssse3);
572 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, sse4,
573 if (!isBE(c->dstFormat)) c->yuv2planeX = ff_yuv2planeX_16_sse4,
574 HAVE_ALIGNED_STACK || ARCH_X86_64);
575 if (c->dstBpc == 16 && !isBE(c->dstFormat))
576 c->yuv2plane1 = ff_yuv2plane1_16_sse4;
579 if (EXTERNAL_AVX(cpu_flags)) {
580 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, avx, ,
581 HAVE_ALIGNED_STACK || ARCH_X86_64);
582 ASSIGN_VSCALE_FUNC(c->yuv2plane1, avx, avx, 1);
584 switch (c->srcFormat) {
585 case AV_PIX_FMT_YUYV422:
586 c->chrToYV12 = ff_yuyvToUV_avx;
588 case AV_PIX_FMT_UYVY422:
589 c->chrToYV12 = ff_uyvyToUV_avx;
591 case AV_PIX_FMT_NV12:
592 c->chrToYV12 = ff_nv12ToUV_avx;
594 case AV_PIX_FMT_NV21:
595 c->chrToYV12 = ff_nv21ToUV_avx;
597 case_rgb(rgb24, RGB24, avx);
598 case_rgb(bgr24, BGR24, avx);
599 case_rgb(bgra, BGRA, avx);
600 case_rgb(rgba, RGBA, avx);
601 case_rgb(abgr, ABGR, avx);
602 case_rgb(argb, ARGB, avx);