1 /*
2 * Copyright (C) 2001-2011 Michael Niedermayer <michaelni@gmx.at>
3 *
4 * This file is part of FFmpeg.
5 *
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.
10 *
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.
15 *
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
19 */
20
21 #include <inttypes.h>
22 #include "config.h"
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/cpu.h"
29 #include "libavutil/cpu.h"
30 #include "libavutil/mem_internal.h"
31 #include "libavutil/pixdesc.h"
32
33 const DECLARE_ALIGNED(8, uint64_t, ff_dither4)[2] = {
34 0x0103010301030103LL,
35 0x0200020002000200LL,};
36
37 const DECLARE_ALIGNED(8, uint64_t, ff_dither8)[2] = {
38 0x0602060206020602LL,
39 0x0004000400040004LL,};
40
41 #if HAVE_INLINE_ASM
42
43 #define DITHER1XBPP
44
45 DECLARE_ASM_CONST(8, uint64_t, bF8)= 0xF8F8F8F8F8F8F8F8LL;
46 DECLARE_ASM_CONST(8, uint64_t, bFC)= 0xFCFCFCFCFCFCFCFCLL;
47
48 DECLARE_ASM_ALIGNED(8, const uint64_t, ff_M24A) = 0x00FF0000FF0000FFLL;
49 DECLARE_ASM_ALIGNED(8, const uint64_t, ff_M24B) = 0xFF0000FF0000FF00LL;
50 DECLARE_ASM_ALIGNED(8, const uint64_t, ff_M24C) = 0x0000FF0000FF0000LL;
51
52 DECLARE_ASM_ALIGNED(8, const uint64_t, ff_bgr2YOffset) = 0x1010101010101010ULL;
53 DECLARE_ASM_ALIGNED(8, const uint64_t, ff_bgr2UVOffset) = 0x8080808080808080ULL;
54 DECLARE_ASM_ALIGNED(8, const uint64_t, ff_w1111) = 0x0001000100010001ULL;
55
56
57 //MMX versions
58 #if HAVE_MMX_INLINE
59 #undef RENAME
60 #define COMPILE_TEMPLATE_MMXEXT 0
61 #define RENAME(a) a ## _mmx
62 #include "swscale_template.c"
63 #endif
64
65 // MMXEXT versions
66 #if HAVE_MMXEXT_INLINE
67 #undef RENAME
68 #undef COMPILE_TEMPLATE_MMXEXT
69 #define COMPILE_TEMPLATE_MMXEXT 1
70 #define RENAME(a) a ## _mmxext
71 #include "swscale_template.c"
72 #endif
73
ff_updateMMXDitherTables(SwsContext * c,int dstY)74 void ff_updateMMXDitherTables(SwsContext *c, int dstY)
75 {
76 const int dstH= c->dstH;
77 const int flags= c->flags;
78
79 SwsPlane *lumPlane = &c->slice[c->numSlice-2].plane[0];
80 SwsPlane *chrUPlane = &c->slice[c->numSlice-2].plane[1];
81 SwsPlane *alpPlane = &c->slice[c->numSlice-2].plane[3];
82
83 int hasAlpha = c->needAlpha;
84 int32_t *vLumFilterPos= c->vLumFilterPos;
85 int32_t *vChrFilterPos= c->vChrFilterPos;
86 int16_t *vLumFilter= c->vLumFilter;
87 int16_t *vChrFilter= c->vChrFilter;
88 int32_t *lumMmxFilter= c->lumMmxFilter;
89 int32_t *chrMmxFilter= c->chrMmxFilter;
90 int32_t av_unused *alpMmxFilter= c->alpMmxFilter;
91 const int vLumFilterSize= c->vLumFilterSize;
92 const int vChrFilterSize= c->vChrFilterSize;
93 const int chrDstY= dstY>>c->chrDstVSubSample;
94 const int firstLumSrcY= vLumFilterPos[dstY]; //First line needed as input
95 const int firstChrSrcY= vChrFilterPos[chrDstY]; //First line needed as input
96
97 c->blueDither= ff_dither8[dstY&1];
98 if (c->dstFormat == AV_PIX_FMT_RGB555 || c->dstFormat == AV_PIX_FMT_BGR555)
99 c->greenDither= ff_dither8[dstY&1];
100 else
101 c->greenDither= ff_dither4[dstY&1];
102 c->redDither= ff_dither8[(dstY+1)&1];
103 if (dstY < dstH - 2) {
104 const int16_t **lumSrcPtr = (const int16_t **)(void*) lumPlane->line + firstLumSrcY - lumPlane->sliceY;
105 const int16_t **chrUSrcPtr = (const int16_t **)(void*) chrUPlane->line + firstChrSrcY - chrUPlane->sliceY;
106 const int16_t **alpSrcPtr = (CONFIG_SWSCALE_ALPHA && hasAlpha) ? (const int16_t **)(void*) alpPlane->line + firstLumSrcY - alpPlane->sliceY : NULL;
107
108 int i;
109 if (firstLumSrcY < 0 || firstLumSrcY + vLumFilterSize > c->srcH) {
110 const int16_t **tmpY = (const int16_t **) lumPlane->tmp;
111
112 int neg = -firstLumSrcY, i, end = FFMIN(c->srcH - firstLumSrcY, vLumFilterSize);
113 for (i = 0; i < neg; i++)
114 tmpY[i] = lumSrcPtr[neg];
115 for ( ; i < end; i++)
116 tmpY[i] = lumSrcPtr[i];
117 for ( ; i < vLumFilterSize; i++)
118 tmpY[i] = tmpY[i-1];
119 lumSrcPtr = tmpY;
120
121 if (alpSrcPtr) {
122 const int16_t **tmpA = (const int16_t **) alpPlane->tmp;
123 for (i = 0; i < neg; i++)
124 tmpA[i] = alpSrcPtr[neg];
125 for ( ; i < end; i++)
126 tmpA[i] = alpSrcPtr[i];
127 for ( ; i < vLumFilterSize; i++)
128 tmpA[i] = tmpA[i - 1];
129 alpSrcPtr = tmpA;
130 }
131 }
132 if (firstChrSrcY < 0 || firstChrSrcY + vChrFilterSize > c->chrSrcH) {
133 const int16_t **tmpU = (const int16_t **) chrUPlane->tmp;
134 int neg = -firstChrSrcY, i, end = FFMIN(c->chrSrcH - firstChrSrcY, vChrFilterSize);
135 for (i = 0; i < neg; i++) {
136 tmpU[i] = chrUSrcPtr[neg];
137 }
138 for ( ; i < end; i++) {
139 tmpU[i] = chrUSrcPtr[i];
140 }
141 for ( ; i < vChrFilterSize; i++) {
142 tmpU[i] = tmpU[i - 1];
143 }
144 chrUSrcPtr = tmpU;
145 }
146
147 if (flags & SWS_ACCURATE_RND) {
148 int s= APCK_SIZE / 8;
149 for (i=0; i<vLumFilterSize; i+=2) {
150 *(const void**)&lumMmxFilter[s*i ]= lumSrcPtr[i ];
151 *(const void**)&lumMmxFilter[s*i+APCK_PTR2/4 ]= lumSrcPtr[i+(vLumFilterSize>1)];
152 lumMmxFilter[s*i+APCK_COEF/4 ]=
153 lumMmxFilter[s*i+APCK_COEF/4+1]= vLumFilter[dstY*vLumFilterSize + i ]
154 + (vLumFilterSize>1 ? vLumFilter[dstY*vLumFilterSize + i + 1] * (1 << 16) : 0);
155 if (CONFIG_SWSCALE_ALPHA && hasAlpha) {
156 *(const void**)&alpMmxFilter[s*i ]= alpSrcPtr[i ];
157 *(const void**)&alpMmxFilter[s*i+APCK_PTR2/4 ]= alpSrcPtr[i+(vLumFilterSize>1)];
158 alpMmxFilter[s*i+APCK_COEF/4 ]=
159 alpMmxFilter[s*i+APCK_COEF/4+1]= lumMmxFilter[s*i+APCK_COEF/4 ];
160 }
161 }
162 for (i=0; i<vChrFilterSize; i+=2) {
163 *(const void**)&chrMmxFilter[s*i ]= chrUSrcPtr[i ];
164 *(const void**)&chrMmxFilter[s*i+APCK_PTR2/4 ]= chrUSrcPtr[i+(vChrFilterSize>1)];
165 chrMmxFilter[s*i+APCK_COEF/4 ]=
166 chrMmxFilter[s*i+APCK_COEF/4+1]= vChrFilter[chrDstY*vChrFilterSize + i ]
167 + (vChrFilterSize>1 ? vChrFilter[chrDstY*vChrFilterSize + i + 1] * (1 << 16) : 0);
168 }
169 } else {
170 for (i=0; i<vLumFilterSize; i++) {
171 *(const void**)&lumMmxFilter[4*i+0]= lumSrcPtr[i];
172 lumMmxFilter[4*i+2]=
173 lumMmxFilter[4*i+3]=
174 ((uint16_t)vLumFilter[dstY*vLumFilterSize + i])*0x10001U;
175 if (CONFIG_SWSCALE_ALPHA && hasAlpha) {
176 *(const void**)&alpMmxFilter[4*i+0]= alpSrcPtr[i];
177 alpMmxFilter[4*i+2]=
178 alpMmxFilter[4*i+3]= lumMmxFilter[4*i+2];
179 }
180 }
181 for (i=0; i<vChrFilterSize; i++) {
182 *(const void**)&chrMmxFilter[4*i+0]= chrUSrcPtr[i];
183 chrMmxFilter[4*i+2]=
184 chrMmxFilter[4*i+3]=
185 ((uint16_t)vChrFilter[chrDstY*vChrFilterSize + i])*0x10001U;
186 }
187 }
188 }
189 }
190 #endif /* HAVE_INLINE_ASM */
191
192 #define YUV2YUVX_FUNC_MMX(opt, step) \
193 void ff_yuv2yuvX_ ##opt(const int16_t *filter, int filterSize, int srcOffset, \
194 uint8_t *dest, int dstW, \
195 const uint8_t *dither, int offset); \
196 static void yuv2yuvX_ ##opt(const int16_t *filter, int filterSize, \
197 const int16_t **src, uint8_t *dest, int dstW, \
198 const uint8_t *dither, int offset) \
199 { \
200 if(dstW > 0) \
201 ff_yuv2yuvX_ ##opt(filter, filterSize - 1, 0, dest - offset, dstW + offset, dither, offset); \
202 return; \
203 }
204
205 #define YUV2YUVX_FUNC(opt, step) \
206 void ff_yuv2yuvX_ ##opt(const int16_t *filter, int filterSize, int srcOffset, \
207 uint8_t *dest, int dstW, \
208 const uint8_t *dither, int offset); \
209 static void yuv2yuvX_ ##opt(const int16_t *filter, int filterSize, \
210 const int16_t **src, uint8_t *dest, int dstW, \
211 const uint8_t *dither, int offset) \
212 { \
213 int remainder = (dstW % step); \
214 int pixelsProcessed = dstW - remainder; \
215 if(((uintptr_t)dest) & 15){ \
216 yuv2yuvX_mmx(filter, filterSize, src, dest, dstW, dither, offset); \
217 return; \
218 } \
219 if(pixelsProcessed > 0) \
220 ff_yuv2yuvX_ ##opt(filter, filterSize - 1, 0, dest - offset, pixelsProcessed + offset, dither, offset); \
221 if(remainder > 0){ \
222 ff_yuv2yuvX_mmx(filter, filterSize - 1, pixelsProcessed, dest - offset, pixelsProcessed + remainder + offset, dither, offset); \
223 } \
224 return; \
225 }
226
227 #if HAVE_MMX_EXTERNAL
228 YUV2YUVX_FUNC_MMX(mmx, 16)
229 #endif
230 #if HAVE_MMXEXT_EXTERNAL
231 YUV2YUVX_FUNC_MMX(mmxext, 16)
232 #endif
233 #if HAVE_SSE3_EXTERNAL
234 YUV2YUVX_FUNC(sse3, 32)
235 #endif
236 #if HAVE_AVX2_EXTERNAL
237 YUV2YUVX_FUNC(avx2, 64)
238 #endif
239
240 #define SCALE_FUNC(filter_n, from_bpc, to_bpc, opt) \
241 void ff_hscale ## from_bpc ## to ## to_bpc ## _ ## filter_n ## _ ## opt( \
242 SwsContext *c, int16_t *data, \
243 int dstW, const uint8_t *src, \
244 const int16_t *filter, \
245 const int32_t *filterPos, int filterSize)
246
247 #define SCALE_FUNCS(filter_n, opt) \
248 SCALE_FUNC(filter_n, 8, 15, opt); \
249 SCALE_FUNC(filter_n, 9, 15, opt); \
250 SCALE_FUNC(filter_n, 10, 15, opt); \
251 SCALE_FUNC(filter_n, 12, 15, opt); \
252 SCALE_FUNC(filter_n, 14, 15, opt); \
253 SCALE_FUNC(filter_n, 16, 15, opt); \
254 SCALE_FUNC(filter_n, 8, 19, opt); \
255 SCALE_FUNC(filter_n, 9, 19, opt); \
256 SCALE_FUNC(filter_n, 10, 19, opt); \
257 SCALE_FUNC(filter_n, 12, 19, opt); \
258 SCALE_FUNC(filter_n, 14, 19, opt); \
259 SCALE_FUNC(filter_n, 16, 19, opt)
260
261 #define SCALE_FUNCS_MMX(opt) \
262 SCALE_FUNCS(4, opt); \
263 SCALE_FUNCS(8, opt); \
264 SCALE_FUNCS(X, opt)
265
266 #define SCALE_FUNCS_SSE(opt) \
267 SCALE_FUNCS(4, opt); \
268 SCALE_FUNCS(8, opt); \
269 SCALE_FUNCS(X4, opt); \
270 SCALE_FUNCS(X8, opt)
271
272 #if ARCH_X86_32
273 SCALE_FUNCS_MMX(mmx);
274 #endif
275 SCALE_FUNCS_SSE(sse2);
276 SCALE_FUNCS_SSE(ssse3);
277 SCALE_FUNCS_SSE(sse4);
278
279 #define VSCALEX_FUNC(size, opt) \
280 void ff_yuv2planeX_ ## size ## _ ## opt(const int16_t *filter, int filterSize, \
281 const int16_t **src, uint8_t *dest, int dstW, \
282 const uint8_t *dither, int offset)
283 #define VSCALEX_FUNCS(opt) \
284 VSCALEX_FUNC(8, opt); \
285 VSCALEX_FUNC(9, opt); \
286 VSCALEX_FUNC(10, opt)
287
288 #if ARCH_X86_32
289 VSCALEX_FUNCS(mmxext);
290 #endif
291 VSCALEX_FUNCS(sse2);
292 VSCALEX_FUNCS(sse4);
293 VSCALEX_FUNC(16, sse4);
294 VSCALEX_FUNCS(avx);
295
296 #define VSCALE_FUNC(size, opt) \
297 void ff_yuv2plane1_ ## size ## _ ## opt(const int16_t *src, uint8_t *dst, int dstW, \
298 const uint8_t *dither, int offset)
299 #define VSCALE_FUNCS(opt1, opt2) \
300 VSCALE_FUNC(8, opt1); \
301 VSCALE_FUNC(9, opt2); \
302 VSCALE_FUNC(10, opt2); \
303 VSCALE_FUNC(16, opt1)
304
305 #if ARCH_X86_32
306 VSCALE_FUNCS(mmx, mmxext);
307 #endif
308 VSCALE_FUNCS(sse2, sse2);
309 VSCALE_FUNC(16, sse4);
310 VSCALE_FUNCS(avx, avx);
311
312 #define INPUT_Y_FUNC(fmt, opt) \
313 void ff_ ## fmt ## ToY_ ## opt(uint8_t *dst, const uint8_t *src, \
314 const uint8_t *unused1, const uint8_t *unused2, \
315 int w, uint32_t *unused)
316 #define INPUT_UV_FUNC(fmt, opt) \
317 void ff_ ## fmt ## ToUV_ ## opt(uint8_t *dstU, uint8_t *dstV, \
318 const uint8_t *unused0, \
319 const uint8_t *src1, \
320 const uint8_t *src2, \
321 int w, uint32_t *unused)
322 #define INPUT_FUNC(fmt, opt) \
323 INPUT_Y_FUNC(fmt, opt); \
324 INPUT_UV_FUNC(fmt, opt)
325 #define INPUT_FUNCS(opt) \
326 INPUT_FUNC(uyvy, opt); \
327 INPUT_FUNC(yuyv, opt); \
328 INPUT_UV_FUNC(nv12, opt); \
329 INPUT_UV_FUNC(nv21, opt); \
330 INPUT_FUNC(rgba, opt); \
331 INPUT_FUNC(bgra, opt); \
332 INPUT_FUNC(argb, opt); \
333 INPUT_FUNC(abgr, opt); \
334 INPUT_FUNC(rgb24, opt); \
335 INPUT_FUNC(bgr24, opt)
336
337 #if ARCH_X86_32
338 INPUT_FUNCS(mmx);
339 #endif
340 INPUT_FUNCS(sse2);
341 INPUT_FUNCS(ssse3);
342 INPUT_FUNCS(avx);
343
344 #if ARCH_X86_64
345 #define YUV2NV_DECL(fmt, opt) \
346 void ff_yuv2 ## fmt ## cX_ ## opt(enum AVPixelFormat format, const uint8_t *dither, \
347 const int16_t *filter, int filterSize, \
348 const int16_t **u, const int16_t **v, \
349 uint8_t *dst, int dstWidth)
350
351 YUV2NV_DECL(nv12, avx2);
352 YUV2NV_DECL(nv21, avx2);
353 #endif
354
ff_sws_init_swscale_x86(SwsContext * c)355 av_cold void ff_sws_init_swscale_x86(SwsContext *c)
356 {
357 int cpu_flags = av_get_cpu_flags();
358
359 #if HAVE_MMX_INLINE
360 if (INLINE_MMX(cpu_flags))
361 sws_init_swscale_mmx(c);
362 #endif
363 #if HAVE_MMXEXT_INLINE
364 if (INLINE_MMXEXT(cpu_flags))
365 sws_init_swscale_mmxext(c);
366 #endif
367 if(c->use_mmx_vfilter && !(c->flags & SWS_ACCURATE_RND)) {
368 #if HAVE_MMX_EXTERNAL
369 if (EXTERNAL_MMX(cpu_flags))
370 c->yuv2planeX = yuv2yuvX_mmx;
371 #endif
372 #if HAVE_MMXEXT_EXTERNAL
373 if (EXTERNAL_MMXEXT(cpu_flags))
374 c->yuv2planeX = yuv2yuvX_mmxext;
375 #endif
376 #if HAVE_SSE3_EXTERNAL
377 if (EXTERNAL_SSE3(cpu_flags))
378 c->yuv2planeX = yuv2yuvX_sse3;
379 #endif
380 #if HAVE_AVX2_EXTERNAL
381 if (EXTERNAL_AVX2_FAST(cpu_flags))
382 c->yuv2planeX = yuv2yuvX_avx2;
383 #endif
384 }
385
386 #define ASSIGN_SCALE_FUNC2(hscalefn, filtersize, opt1, opt2) do { \
387 if (c->srcBpc == 8) { \
388 hscalefn = c->dstBpc <= 14 ? ff_hscale8to15_ ## filtersize ## _ ## opt2 : \
389 ff_hscale8to19_ ## filtersize ## _ ## opt1; \
390 } else if (c->srcBpc == 9) { \
391 hscalefn = c->dstBpc <= 14 ? ff_hscale9to15_ ## filtersize ## _ ## opt2 : \
392 ff_hscale9to19_ ## filtersize ## _ ## opt1; \
393 } else if (c->srcBpc == 10) { \
394 hscalefn = c->dstBpc <= 14 ? ff_hscale10to15_ ## filtersize ## _ ## opt2 : \
395 ff_hscale10to19_ ## filtersize ## _ ## opt1; \
396 } else if (c->srcBpc == 12) { \
397 hscalefn = c->dstBpc <= 14 ? ff_hscale12to15_ ## filtersize ## _ ## opt2 : \
398 ff_hscale12to19_ ## filtersize ## _ ## opt1; \
399 } 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)) { \
400 hscalefn = c->dstBpc <= 14 ? ff_hscale14to15_ ## filtersize ## _ ## opt2 : \
401 ff_hscale14to19_ ## filtersize ## _ ## opt1; \
402 } else { /* c->srcBpc == 16 */ \
403 av_assert0(c->srcBpc == 16);\
404 hscalefn = c->dstBpc <= 14 ? ff_hscale16to15_ ## filtersize ## _ ## opt2 : \
405 ff_hscale16to19_ ## filtersize ## _ ## opt1; \
406 } \
407 } while (0)
408 #define ASSIGN_MMX_SCALE_FUNC(hscalefn, filtersize, opt1, opt2) \
409 switch (filtersize) { \
410 case 4: ASSIGN_SCALE_FUNC2(hscalefn, 4, opt1, opt2); break; \
411 case 8: ASSIGN_SCALE_FUNC2(hscalefn, 8, opt1, opt2); break; \
412 default: ASSIGN_SCALE_FUNC2(hscalefn, X, opt1, opt2); break; \
413 }
414 #define ASSIGN_VSCALEX_FUNC(vscalefn, opt, do_16_case, condition_8bit) \
415 switch(c->dstBpc){ \
416 case 16: do_16_case; break; \
417 case 10: if (!isBE(c->dstFormat) && c->dstFormat != AV_PIX_FMT_P010LE) vscalefn = ff_yuv2planeX_10_ ## opt; break; \
418 case 9: if (!isBE(c->dstFormat)) vscalefn = ff_yuv2planeX_9_ ## opt; break; \
419 case 8: if ((condition_8bit) && !c->use_mmx_vfilter) vscalefn = ff_yuv2planeX_8_ ## opt; break; \
420 }
421 #define ASSIGN_VSCALE_FUNC(vscalefn, opt1, opt2, opt2chk) \
422 switch(c->dstBpc){ \
423 case 16: if (!isBE(c->dstFormat)) vscalefn = ff_yuv2plane1_16_ ## opt1; break; \
424 case 10: if (!isBE(c->dstFormat) && c->dstFormat != AV_PIX_FMT_P010LE && opt2chk) vscalefn = ff_yuv2plane1_10_ ## opt2; break; \
425 case 9: if (!isBE(c->dstFormat) && opt2chk) vscalefn = ff_yuv2plane1_9_ ## opt2; break; \
426 case 8: vscalefn = ff_yuv2plane1_8_ ## opt1; break; \
427 default: av_assert0(c->dstBpc>8); \
428 }
429 #define case_rgb(x, X, opt) \
430 case AV_PIX_FMT_ ## X: \
431 c->lumToYV12 = ff_ ## x ## ToY_ ## opt; \
432 if (!c->chrSrcHSubSample) \
433 c->chrToYV12 = ff_ ## x ## ToUV_ ## opt; \
434 break
435 #if ARCH_X86_32
436 if (EXTERNAL_MMX(cpu_flags)) {
437 ASSIGN_MMX_SCALE_FUNC(c->hyScale, c->hLumFilterSize, mmx, mmx);
438 ASSIGN_MMX_SCALE_FUNC(c->hcScale, c->hChrFilterSize, mmx, mmx);
439 ASSIGN_VSCALE_FUNC(c->yuv2plane1, mmx, mmxext, cpu_flags & AV_CPU_FLAG_MMXEXT);
440
441 switch (c->srcFormat) {
442 case AV_PIX_FMT_YA8:
443 c->lumToYV12 = ff_yuyvToY_mmx;
444 if (c->needAlpha)
445 c->alpToYV12 = ff_uyvyToY_mmx;
446 break;
447 case AV_PIX_FMT_YUYV422:
448 c->lumToYV12 = ff_yuyvToY_mmx;
449 c->chrToYV12 = ff_yuyvToUV_mmx;
450 break;
451 case AV_PIX_FMT_UYVY422:
452 c->lumToYV12 = ff_uyvyToY_mmx;
453 c->chrToYV12 = ff_uyvyToUV_mmx;
454 break;
455 case AV_PIX_FMT_NV12:
456 c->chrToYV12 = ff_nv12ToUV_mmx;
457 break;
458 case AV_PIX_FMT_NV21:
459 c->chrToYV12 = ff_nv21ToUV_mmx;
460 break;
461 case_rgb(rgb24, RGB24, mmx);
462 case_rgb(bgr24, BGR24, mmx);
463 case_rgb(bgra, BGRA, mmx);
464 case_rgb(rgba, RGBA, mmx);
465 case_rgb(abgr, ABGR, mmx);
466 case_rgb(argb, ARGB, mmx);
467 default:
468 break;
469 }
470 }
471 if (EXTERNAL_MMXEXT(cpu_flags)) {
472 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, mmxext, , 1);
473 }
474 #endif /* ARCH_X86_32 */
475 #define ASSIGN_SSE_SCALE_FUNC(hscalefn, filtersize, opt1, opt2) \
476 switch (filtersize) { \
477 case 4: ASSIGN_SCALE_FUNC2(hscalefn, 4, opt1, opt2); break; \
478 case 8: ASSIGN_SCALE_FUNC2(hscalefn, 8, opt1, opt2); break; \
479 default: if (filtersize & 4) ASSIGN_SCALE_FUNC2(hscalefn, X4, opt1, opt2); \
480 else ASSIGN_SCALE_FUNC2(hscalefn, X8, opt1, opt2); \
481 break; \
482 }
483 if (EXTERNAL_SSE2(cpu_flags)) {
484 ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, sse2, sse2);
485 ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, sse2, sse2);
486 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, sse2, ,
487 HAVE_ALIGNED_STACK || ARCH_X86_64);
488 ASSIGN_VSCALE_FUNC(c->yuv2plane1, sse2, sse2, 1);
489
490 switch (c->srcFormat) {
491 case AV_PIX_FMT_YA8:
492 c->lumToYV12 = ff_yuyvToY_sse2;
493 if (c->needAlpha)
494 c->alpToYV12 = ff_uyvyToY_sse2;
495 break;
496 case AV_PIX_FMT_YUYV422:
497 c->lumToYV12 = ff_yuyvToY_sse2;
498 c->chrToYV12 = ff_yuyvToUV_sse2;
499 break;
500 case AV_PIX_FMT_UYVY422:
501 c->lumToYV12 = ff_uyvyToY_sse2;
502 c->chrToYV12 = ff_uyvyToUV_sse2;
503 break;
504 case AV_PIX_FMT_NV12:
505 c->chrToYV12 = ff_nv12ToUV_sse2;
506 break;
507 case AV_PIX_FMT_NV21:
508 c->chrToYV12 = ff_nv21ToUV_sse2;
509 break;
510 case_rgb(rgb24, RGB24, sse2);
511 case_rgb(bgr24, BGR24, sse2);
512 case_rgb(bgra, BGRA, sse2);
513 case_rgb(rgba, RGBA, sse2);
514 case_rgb(abgr, ABGR, sse2);
515 case_rgb(argb, ARGB, sse2);
516 default:
517 break;
518 }
519 }
520 if (EXTERNAL_SSSE3(cpu_flags)) {
521 ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, ssse3, ssse3);
522 ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, ssse3, ssse3);
523 switch (c->srcFormat) {
524 case_rgb(rgb24, RGB24, ssse3);
525 case_rgb(bgr24, BGR24, ssse3);
526 default:
527 break;
528 }
529 }
530 if (EXTERNAL_SSE4(cpu_flags)) {
531 /* Xto15 don't need special sse4 functions */
532 ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, sse4, ssse3);
533 ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, sse4, ssse3);
534 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, sse4,
535 if (!isBE(c->dstFormat)) c->yuv2planeX = ff_yuv2planeX_16_sse4,
536 HAVE_ALIGNED_STACK || ARCH_X86_64);
537 if (c->dstBpc == 16 && !isBE(c->dstFormat))
538 c->yuv2plane1 = ff_yuv2plane1_16_sse4;
539 }
540
541 if (EXTERNAL_AVX(cpu_flags)) {
542 ASSIGN_VSCALEX_FUNC(c->yuv2planeX, avx, ,
543 HAVE_ALIGNED_STACK || ARCH_X86_64);
544 ASSIGN_VSCALE_FUNC(c->yuv2plane1, avx, avx, 1);
545
546 switch (c->srcFormat) {
547 case AV_PIX_FMT_YUYV422:
548 c->chrToYV12 = ff_yuyvToUV_avx;
549 break;
550 case AV_PIX_FMT_UYVY422:
551 c->chrToYV12 = ff_uyvyToUV_avx;
552 break;
553 case AV_PIX_FMT_NV12:
554 c->chrToYV12 = ff_nv12ToUV_avx;
555 break;
556 case AV_PIX_FMT_NV21:
557 c->chrToYV12 = ff_nv21ToUV_avx;
558 break;
559 case_rgb(rgb24, RGB24, avx);
560 case_rgb(bgr24, BGR24, avx);
561 case_rgb(bgra, BGRA, avx);
562 case_rgb(rgba, RGBA, avx);
563 case_rgb(abgr, ABGR, avx);
564 case_rgb(argb, ARGB, avx);
565 default:
566 break;
567 }
568 }
569
570 #if ARCH_X86_64
571 if (EXTERNAL_AVX2_FAST(cpu_flags)) {
572 switch (c->dstFormat) {
573 case AV_PIX_FMT_NV12:
574 case AV_PIX_FMT_NV24:
575 c->yuv2nv12cX = ff_yuv2nv12cX_avx2;
576 break;
577 case AV_PIX_FMT_NV21:
578 case AV_PIX_FMT_NV42:
579 c->yuv2nv12cX = ff_yuv2nv21cX_avx2;
580 break;
581 default:
582 break;
583 }
584 }
585 #endif
586 }
587