FFmpeg: libswscale/x86/swscale

00001 /*
00002  * Copyright (C) 2001-2011 Michael Niedermayer <michaelni@gmx.at>
00003  *
00004  * This file is part of FFmpeg.
00005  *
00006  * FFmpeg is free software; you can redistribute it and/or
00007  * modify it under the terms of the GNU Lesser General Public
00008  * License as published by the Free Software Foundation; either
00009  * version 2.1 of the License, or (at your option) any later version.
00010  *
00011  * FFmpeg is distributed in the hope that it will be useful,
00012  * but WITHOUT ANY WARRANTY; without even the implied warranty of
00013  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
00014  * Lesser General Public License for more details.
00015  *
00016  * You should have received a copy of the GNU Lesser General Public
00017  * License along with FFmpeg; if not, write to the Free Software
00018  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
00019  */
00020 
00021 #include <inttypes.h>
00022 #include "config.h"
00023 #include "libswscale/swscale.h"
00024 #include "libswscale/swscale_internal.h"
00025 #include "libavutil/intreadwrite.h"
00026 #include "libavutil/x86_cpu.h"
00027 #include "libavutil/cpu.h"
00028 #include "libavutil/pixdesc.h"
00029 
00030 DECLARE_ASM_CONST(8, uint64_t, bF8)=       0xF8F8F8F8F8F8F8F8LL;
00031 DECLARE_ASM_CONST(8, uint64_t, bFC)=       0xFCFCFCFCFCFCFCFCLL;
00032 DECLARE_ASM_CONST(8, uint64_t, w10)=       0x0010001000100010LL;
00033 DECLARE_ASM_CONST(8, uint64_t, w02)=       0x0002000200020002LL;
00034 DECLARE_ASM_CONST(8, uint64_t, bm00001111)=0x00000000FFFFFFFFLL;
00035 DECLARE_ASM_CONST(8, uint64_t, bm00000111)=0x0000000000FFFFFFLL;
00036 DECLARE_ASM_CONST(8, uint64_t, bm11111000)=0xFFFFFFFFFF000000LL;
00037 DECLARE_ASM_CONST(8, uint64_t, bm01010101)=0x00FF00FF00FF00FFLL;
00038 
00039 const DECLARE_ALIGNED(8, uint64_t, ff_dither4)[2] = {
00040     0x0103010301030103LL,
00041     0x0200020002000200LL,};
00042 
00043 const DECLARE_ALIGNED(8, uint64_t, ff_dither8)[2] = {
00044     0x0602060206020602LL,
00045     0x0004000400040004LL,};
00046 
00047 DECLARE_ASM_CONST(8, uint64_t, b16Mask)=   0x001F001F001F001FLL;
00048 DECLARE_ASM_CONST(8, uint64_t, g16Mask)=   0x07E007E007E007E0LL;
00049 DECLARE_ASM_CONST(8, uint64_t, r16Mask)=   0xF800F800F800F800LL;
00050 DECLARE_ASM_CONST(8, uint64_t, b15Mask)=   0x001F001F001F001FLL;
00051 DECLARE_ASM_CONST(8, uint64_t, g15Mask)=   0x03E003E003E003E0LL;
00052 DECLARE_ASM_CONST(8, uint64_t, r15Mask)=   0x7C007C007C007C00LL;
00053 
00054 DECLARE_ALIGNED(8, const uint64_t, ff_M24A)         = 0x00FF0000FF0000FFLL;
00055 DECLARE_ALIGNED(8, const uint64_t, ff_M24B)         = 0xFF0000FF0000FF00LL;
00056 DECLARE_ALIGNED(8, const uint64_t, ff_M24C)         = 0x0000FF0000FF0000LL;
00057 
00058 #ifdef FAST_BGR2YV12
00059 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2YCoeff)   = 0x000000210041000DULL;
00060 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2UCoeff)   = 0x0000FFEEFFDC0038ULL;
00061 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2VCoeff)   = 0x00000038FFD2FFF8ULL;
00062 #else
00063 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2YCoeff)   = 0x000020E540830C8BULL;
00064 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2UCoeff)   = 0x0000ED0FDAC23831ULL;
00065 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2VCoeff)   = 0x00003831D0E6F6EAULL;
00066 #endif /* FAST_BGR2YV12 */
00067 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2YOffset)  = 0x1010101010101010ULL;
00068 DECLARE_ALIGNED(8, const uint64_t, ff_bgr2UVOffset) = 0x8080808080808080ULL;
00069 DECLARE_ALIGNED(8, const uint64_t, ff_w1111)        = 0x0001000100010001ULL;
00070 
00071 DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toY1Coeff) = 0x0C88000040870C88ULL;
00072 DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toY2Coeff) = 0x20DE4087000020DEULL;
00073 DECLARE_ASM_CONST(8, uint64_t, ff_rgb24toY1Coeff) = 0x20DE0000408720DEULL;
00074 DECLARE_ASM_CONST(8, uint64_t, ff_rgb24toY2Coeff) = 0x0C88408700000C88ULL;
00075 DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toYOffset) = 0x0008010000080100ULL;
00076 
00077 DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toUV)[2][4] = {
00078     {0x38380000DAC83838ULL, 0xECFFDAC80000ECFFULL, 0xF6E40000D0E3F6E4ULL, 0x3838D0E300003838ULL},
00079     {0xECFF0000DAC8ECFFULL, 0x3838DAC800003838ULL, 0x38380000D0E33838ULL, 0xF6E4D0E30000F6E4ULL},
00080 };
00081 
00082 DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toUVOffset)= 0x0040010000400100ULL;
00083 
00084 //MMX versions
00085 #if HAVE_MMX
00086 #undef RENAME
00087 #define COMPILE_TEMPLATE_MMX2 0
00088 #define RENAME(a) a ## _MMX
00089 #include "swscale_template.c"
00090 #endif
00091 
00092 //MMX2 versions
00093 #if HAVE_MMX2
00094 #undef RENAME
00095 #undef COMPILE_TEMPLATE_MMX2
00096 #define COMPILE_TEMPLATE_MMX2 1
00097 #define RENAME(a) a ## _MMX2
00098 #include "swscale_template.c"
00099 #endif
00100 
00101 void updateMMXDitherTables(SwsContext *c, int dstY, int lumBufIndex, int chrBufIndex,
00102                            int lastInLumBuf, int lastInChrBuf)
00103 {
00104     const int dstH= c->dstH;
00105     const int flags= c->flags;
00106     int16_t **lumPixBuf= c->lumPixBuf;
00107     int16_t **chrUPixBuf= c->chrUPixBuf;
00108     int16_t **alpPixBuf= c->alpPixBuf;
00109     const int vLumBufSize= c->vLumBufSize;
00110     const int vChrBufSize= c->vChrBufSize;
00111     int16_t *vLumFilterPos= c->vLumFilterPos;
00112     int16_t *vChrFilterPos= c->vChrFilterPos;
00113     int16_t *vLumFilter= c->vLumFilter;
00114     int16_t *vChrFilter= c->vChrFilter;
00115     int32_t *lumMmxFilter= c->lumMmxFilter;
00116     int32_t *chrMmxFilter= c->chrMmxFilter;
00117     int32_t av_unused *alpMmxFilter= c->alpMmxFilter;
00118     const int vLumFilterSize= c->vLumFilterSize;
00119     const int vChrFilterSize= c->vChrFilterSize;
00120     const int chrDstY= dstY>>c->chrDstVSubSample;
00121     const int firstLumSrcY= vLumFilterPos[dstY]; //First line needed as input
00122     const int firstChrSrcY= vChrFilterPos[chrDstY]; //First line needed as input
00123 
00124     c->blueDither= ff_dither8[dstY&1];
00125     if (c->dstFormat == PIX_FMT_RGB555 || c->dstFormat == PIX_FMT_BGR555)
00126         c->greenDither= ff_dither8[dstY&1];
00127     else
00128         c->greenDither= ff_dither4[dstY&1];
00129     c->redDither= ff_dither8[(dstY+1)&1];
00130     if (dstY < dstH - 2) {
00131         const int16_t **lumSrcPtr= (const int16_t **)(void*) lumPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize;
00132         const int16_t **chrUSrcPtr= (const int16_t **)(void*) chrUPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
00133         const int16_t **alpSrcPtr= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? (const int16_t **)(void*) alpPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize : NULL;
00134         int i;
00135         if (flags & SWS_ACCURATE_RND) {
00136             int s= APCK_SIZE / 8;
00137             for (i=0; i<vLumFilterSize; i+=2) {
00138                 *(const void**)&lumMmxFilter[s*i              ]= lumSrcPtr[i  ];
00139                 *(const void**)&lumMmxFilter[s*i+APCK_PTR2/4  ]= lumSrcPtr[i+(vLumFilterSize>1)];
00140                 lumMmxFilter[s*i+APCK_COEF/4  ]=
00141                 lumMmxFilter[s*i+APCK_COEF/4+1]= vLumFilter[dstY*vLumFilterSize + i    ]
00142                 + (vLumFilterSize>1 ? vLumFilter[dstY*vLumFilterSize + i + 1]<<16 : 0);
00143                 if (CONFIG_SWSCALE_ALPHA && alpPixBuf) {
00144                     *(const void**)&alpMmxFilter[s*i              ]= alpSrcPtr[i  ];
00145                     *(const void**)&alpMmxFilter[s*i+APCK_PTR2/4  ]= alpSrcPtr[i+(vLumFilterSize>1)];
00146                     alpMmxFilter[s*i+APCK_COEF/4  ]=
00147                     alpMmxFilter[s*i+APCK_COEF/4+1]= lumMmxFilter[s*i+APCK_COEF/4  ];
00148                 }
00149             }
00150             for (i=0; i<vChrFilterSize; i+=2) {
00151                 *(const void**)&chrMmxFilter[s*i              ]= chrUSrcPtr[i  ];
00152                 *(const void**)&chrMmxFilter[s*i+APCK_PTR2/4  ]= chrUSrcPtr[i+(vChrFilterSize>1)];
00153                 chrMmxFilter[s*i+APCK_COEF/4  ]=
00154                 chrMmxFilter[s*i+APCK_COEF/4+1]= vChrFilter[chrDstY*vChrFilterSize + i    ]
00155                 + (vChrFilterSize>1 ? vChrFilter[chrDstY*vChrFilterSize + i + 1]<<16 : 0);
00156             }
00157         } else {
00158             for (i=0; i<vLumFilterSize; i++) {
00159                 *(const void**)&lumMmxFilter[4*i+0]= lumSrcPtr[i];
00160                 lumMmxFilter[4*i+2]=
00161                 lumMmxFilter[4*i+3]=
00162                 ((uint16_t)vLumFilter[dstY*vLumFilterSize + i])*0x10001;
00163                 if (CONFIG_SWSCALE_ALPHA && alpPixBuf) {
00164                     *(const void**)&alpMmxFilter[4*i+0]= alpSrcPtr[i];
00165                     alpMmxFilter[4*i+2]=
00166                     alpMmxFilter[4*i+3]= lumMmxFilter[4*i+2];
00167                 }
00168             }
00169             for (i=0; i<vChrFilterSize; i++) {
00170                 *(const void**)&chrMmxFilter[4*i+0]= chrUSrcPtr[i];
00171                 chrMmxFilter[4*i+2]=
00172                 chrMmxFilter[4*i+3]=
00173                 ((uint16_t)vChrFilter[chrDstY*vChrFilterSize + i])*0x10001;
00174             }
00175         }
00176     }
00177 }
00178 
00179 #if HAVE_MMX2
00180 static void yuv2yuvX_sse3(const int16_t *filter, int filterSize,
00181                            const int16_t **src, uint8_t *dest, int dstW,
00182                            const uint8_t *dither, int offset)
00183 {
00184     if(((int)dest) & 15){
00185         return yuv2yuvX_MMX2(filter, filterSize, src, dest, dstW, dither, offset);
00186     }
00187     if (offset) {
00188         __asm__ volatile("movq       (%0), %%xmm3\n\t"
00189                          "movdqa    %%xmm3, %%xmm4\n\t"
00190                          "psrlq       $24, %%xmm3\n\t"
00191                          "psllq       $40, %%xmm4\n\t"
00192                          "por       %%xmm4, %%xmm3\n\t"
00193                          :: "r"(dither)
00194                          );
00195     } else {
00196         __asm__ volatile("movq       (%0), %%xmm3\n\t"
00197                          :: "r"(dither)
00198                          );
00199     }
00200     __asm__ volatile(
00201         "pxor      %%xmm0, %%xmm0\n\t"
00202         "punpcklbw %%xmm0, %%xmm3\n\t"
00203         "psraw        $4, %%xmm3\n\t"
00204         "movdqa    %%xmm3, %%xmm4\n\t"
00205         "movdqa    %%xmm3, %%xmm7\n\t"
00206         "movl %3, %%ecx\n\t"
00207         "mov                                 %0, %%"REG_d"  \n\t"\
00208         "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
00209         ".p2align                             4             \n\t" /* FIXME Unroll? */\
00210         "1:                                                 \n\t"\
00211         "movddup                  8(%%"REG_d"), %%xmm0      \n\t" /* filterCoeff */\
00212         "movdqa              (%%"REG_S", %%"REG_c", 2), %%xmm2      \n\t" /* srcData */\
00213         "movdqa            16(%%"REG_S", %%"REG_c", 2), %%xmm5      \n\t" /* srcData */\
00214         "add                                $16, %%"REG_d"  \n\t"\
00215         "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
00216         "test                         %%"REG_S", %%"REG_S"  \n\t"\
00217         "pmulhw                           %%xmm0, %%xmm2      \n\t"\
00218         "pmulhw                           %%xmm0, %%xmm5      \n\t"\
00219         "paddw                            %%xmm2, %%xmm3      \n\t"\
00220         "paddw                            %%xmm5, %%xmm4      \n\t"\
00221         " jnz                                1b             \n\t"\
00222         "psraw                               $3, %%xmm3      \n\t"\
00223         "psraw                               $3, %%xmm4      \n\t"\
00224         "packuswb                         %%xmm4, %%xmm3      \n\t"
00225         "movntdq                          %%xmm3, (%1, %%"REG_c")\n\t"
00226         "add                         $16, %%"REG_c"         \n\t"\
00227         "cmp                          %2, %%"REG_c"         \n\t"\
00228         "movdqa    %%xmm7, %%xmm3\n\t"
00229         "movdqa    %%xmm7, %%xmm4\n\t"
00230         "mov                                 %0, %%"REG_d"  \n\t"\
00231         "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
00232         "jb                                  1b             \n\t"\
00233         :: "g" (filter),
00234            "r" (dest-offset), "g" ((x86_reg)(dstW+offset)), "m" (offset)
00235         : "%"REG_d, "%"REG_S, "%"REG_c
00236     );
00237 }
00238 #endif
00239 
00240 #define SCALE_FUNC(filter_n, from_bpc, to_bpc, opt) \
00241 extern void ff_hscale ## from_bpc ## to ## to_bpc ## _ ## filter_n ## _ ## opt( \
00242                                                 SwsContext *c, int16_t *data, \
00243                                                 int dstW, const uint8_t *src, \
00244                                                 const int16_t *filter, \
00245                                                 const int16_t *filterPos, int filterSize);
00246 
00247 #define SCALE_FUNCS(filter_n, opt) \
00248     SCALE_FUNC(filter_n,  8, 15, opt); \
00249     SCALE_FUNC(filter_n,  9, 15, opt); \
00250     SCALE_FUNC(filter_n, 10, 15, opt); \
00251     SCALE_FUNC(filter_n, 14, 15, opt); \
00252     SCALE_FUNC(filter_n, 16, 15, opt); \
00253     SCALE_FUNC(filter_n,  8, 19, opt); \
00254     SCALE_FUNC(filter_n,  9, 19, opt); \
00255     SCALE_FUNC(filter_n, 10, 19, opt); \
00256     SCALE_FUNC(filter_n, 14, 19, opt); \
00257     SCALE_FUNC(filter_n, 16, 19, opt)
00258 
00259 #define SCALE_FUNCS_MMX(opt) \
00260     SCALE_FUNCS(4, opt); \
00261     SCALE_FUNCS(8, opt); \
00262     SCALE_FUNCS(X, opt)
00263 
00264 #define SCALE_FUNCS_SSE(opt) \
00265     SCALE_FUNCS(4, opt); \
00266     SCALE_FUNCS(8, opt); \
00267     SCALE_FUNCS(X4, opt); \
00268     SCALE_FUNCS(X8, opt)
00269 
00270 #if ARCH_X86_32
00271 SCALE_FUNCS_MMX(mmx);
00272 #endif
00273 SCALE_FUNCS_SSE(sse2);
00274 SCALE_FUNCS_SSE(ssse3);
00275 SCALE_FUNCS_SSE(sse4);
00276 
00277 #define VSCALEX_FUNC(size, opt) \
00278 extern void ff_yuv2planeX_ ## size ## _ ## opt(const int16_t *filter, int filterSize, \
00279                                                const int16_t **src, uint8_t *dest, int dstW, \
00280                                                const uint8_t *dither, int offset)
00281 #define VSCALEX_FUNCS(opt1, opt2) \
00282     VSCALEX_FUNC(8,  opt1); \
00283     VSCALEX_FUNC(9,  opt2); \
00284     VSCALEX_FUNC(10, opt2)
00285 
00286 #if ARCH_X86_32
00287 VSCALEX_FUNCS(mmx,  mmx2);
00288 #endif
00289 VSCALEX_FUNCS(sse2, sse2);
00290 VSCALEX_FUNCS(sse4, sse4);
00291 VSCALEX_FUNC(16, sse4);
00292 VSCALEX_FUNCS(avx,  avx);
00293 
00294 #define VSCALE_FUNC(size, opt) \
00295 extern void ff_yuv2plane1_ ## size ## _ ## opt(const int16_t *src, uint8_t *dst, int dstW, \
00296                                                const uint8_t *dither, int offset)
00297 #define VSCALE_FUNCS(opt1, opt2) \
00298     VSCALE_FUNC(8,  opt1); \
00299     VSCALE_FUNC(9,  opt2); \
00300     VSCALE_FUNC(10, opt2); \
00301     VSCALE_FUNC(16, opt1)
00302 
00303 #if ARCH_X86_32
00304 VSCALE_FUNCS(mmx, mmx2);
00305 #endif
00306 VSCALE_FUNCS(sse2, sse2);
00307 VSCALE_FUNC(16, sse4);
00308 VSCALE_FUNCS(avx, avx);
00309 
00310 void ff_sws_init_swScale_mmx(SwsContext *c)
00311 {
00312     int cpu_flags = av_get_cpu_flags();
00313 
00314     if (cpu_flags & AV_CPU_FLAG_MMX)
00315         sws_init_swScale_MMX(c);
00316 #if HAVE_MMX2
00317     if (cpu_flags & AV_CPU_FLAG_MMX2)
00318         sws_init_swScale_MMX2(c);
00319     if (cpu_flags & AV_CPU_FLAG_SSE3){
00320         if(c->use_mmx_vfilter && !(c->flags & SWS_ACCURATE_RND))
00321             c->yuv2planeX = yuv2yuvX_sse3;
00322     }
00323 #endif
00324 
00325 #if HAVE_YASM
00326 #define ASSIGN_SCALE_FUNC2(hscalefn, filtersize, opt1, opt2) do { \
00327     if (c->srcBpc == 8) { \
00328         hscalefn = c->dstBpc <= 10 ? ff_hscale8to15_ ## filtersize ## _ ## opt2 : \
00329                                      ff_hscale8to19_ ## filtersize ## _ ## opt1; \
00330     } else if (c->srcBpc == 9) { \
00331         hscalefn = c->dstBpc <= 10 ? ff_hscale9to15_ ## filtersize ## _ ## opt2 : \
00332                                      ff_hscale9to19_ ## filtersize ## _ ## opt1; \
00333     } else if (c->srcBpc == 10) { \
00334         hscalefn = c->dstBpc <= 10 ? ff_hscale10to15_ ## filtersize ## _ ## opt2 : \
00335                                      ff_hscale10to19_ ## filtersize ## _ ## opt1; \
00336     } else if (c->srcBpc == 14 || ((c->srcFormat==PIX_FMT_PAL8||isAnyRGB(c->srcFormat)) && av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1<15)) { \
00337         hscalefn = c->dstBpc <= 10 ? ff_hscale14to15_ ## filtersize ## _ ## opt2 : \
00338                                      ff_hscale14to19_ ## filtersize ## _ ## opt1; \
00339     } else { /* c->srcBpc == 16 */ \
00340         hscalefn = c->dstBpc <= 10 ? ff_hscale16to15_ ## filtersize ## _ ## opt2 : \
00341                                      ff_hscale16to19_ ## filtersize ## _ ## opt1; \
00342     } \
00343 } while (0)
00344 #define ASSIGN_MMX_SCALE_FUNC(hscalefn, filtersize, opt1, opt2) \
00345     switch (filtersize) { \
00346     case 4:  ASSIGN_SCALE_FUNC2(hscalefn, 4, opt1, opt2); break; \
00347     case 8:  ASSIGN_SCALE_FUNC2(hscalefn, 8, opt1, opt2); break; \
00348     default: ASSIGN_SCALE_FUNC2(hscalefn, X, opt1, opt2); break; \
00349     }
00350 #define ASSIGN_VSCALEX_FUNC(vscalefn, opt1, opt2, opt2chk, do_16_case) \
00351 switch(c->dstBpc){ \
00352     case 16:                                     /*do_16_case;*/                           break; \
00353     case 10: if (!isBE(c->dstFormat) && opt2chk) /*vscalefn = ff_yuv2planeX_10_ ## opt2;*/ break; \
00354     case 9:  if (!isBE(c->dstFormat) && opt2chk) /*vscalefn = ff_yuv2planeX_9_  ## opt2;*/ break; \
00355     default:                                     /*vscalefn = ff_yuv2planeX_8_  ## opt1;*/ break; \
00356     }
00357 #define ASSIGN_VSCALE_FUNC(vscalefn, opt1, opt2, opt2chk) \
00358     switch(c->dstBpc){ \
00359     case 16: if (!isBE(c->dstFormat))            vscalefn = ff_yuv2plane1_16_ ## opt1; break; \
00360     case 10: if (!isBE(c->dstFormat) && opt2chk) vscalefn = ff_yuv2plane1_10_ ## opt2; break; \
00361     case 9:  if (!isBE(c->dstFormat) && opt2chk) vscalefn = ff_yuv2plane1_9_  ## opt2;  break; \
00362     default:                                     vscalefn = ff_yuv2plane1_8_  ## opt1;  break; \
00363     }
00364 #if ARCH_X86_32
00365     if (cpu_flags & AV_CPU_FLAG_MMX) {
00366         ASSIGN_MMX_SCALE_FUNC(c->hyScale, c->hLumFilterSize, mmx, mmx);
00367         ASSIGN_MMX_SCALE_FUNC(c->hcScale, c->hChrFilterSize, mmx, mmx);
00368         ASSIGN_VSCALEX_FUNC(c->yuv2planeX, mmx, mmx2, cpu_flags & AV_CPU_FLAG_MMX2,);
00369         ASSIGN_VSCALE_FUNC(c->yuv2plane1, mmx, mmx2, cpu_flags & AV_CPU_FLAG_MMX2);
00370     }
00371 #endif
00372 #define ASSIGN_SSE_SCALE_FUNC(hscalefn, filtersize, opt1, opt2) \
00373     switch (filtersize) { \
00374     case 4:  ASSIGN_SCALE_FUNC2(hscalefn, 4, opt1, opt2); break; \
00375     case 8:  ASSIGN_SCALE_FUNC2(hscalefn, 8, opt1, opt2); break; \
00376     default: if (filtersize & 4) ASSIGN_SCALE_FUNC2(hscalefn, X4, opt1, opt2); \
00377              else                ASSIGN_SCALE_FUNC2(hscalefn, X8, opt1, opt2); \
00378              break; \
00379     }
00380     if (cpu_flags & AV_CPU_FLAG_SSE2) {
00381         ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, sse2, sse2);
00382         ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, sse2, sse2);
00383         ASSIGN_VSCALEX_FUNC(c->yuv2planeX, sse2, sse2, 1,);
00384         ASSIGN_VSCALE_FUNC(c->yuv2plane1, sse2, sse2, 1);
00385     }
00386     if (cpu_flags & AV_CPU_FLAG_SSSE3) {
00387         ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, ssse3, ssse3);
00388         ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, ssse3, ssse3);
00389     }
00390     if (cpu_flags & AV_CPU_FLAG_SSE4) {
00391         /* Xto15 don't need special sse4 functions */
00392         ASSIGN_SSE_SCALE_FUNC(c->hyScale, c->hLumFilterSize, sse4, ssse3);
00393         ASSIGN_SSE_SCALE_FUNC(c->hcScale, c->hChrFilterSize, sse4, ssse3);
00394         ASSIGN_VSCALEX_FUNC(c->yuv2planeX, sse4, sse4, 1,
00395                             if (!isBE(c->dstFormat)) c->yuv2planeX = ff_yuv2planeX_16_sse4);
00396         if (c->dstBpc == 16 && !isBE(c->dstFormat))
00397             c->yuv2plane1 = ff_yuv2plane1_16_sse4;
00398     }
00399 
00400     if (HAVE_AVX && cpu_flags & AV_CPU_FLAG_AVX) {
00401         ASSIGN_VSCALEX_FUNC(c->yuv2planeX, avx, avx, 1,);
00402         ASSIGN_VSCALE_FUNC(c->yuv2plane1, avx, avx, 1);
00403     }
00404 #endif
00405 }
libswscale/x86/swscale_mmx.c
