git.sesse.net Git - ffmpeg/blob - libswscale/swscale_template.c

   1 /*
   2  * Copyright (C) 2001-2003 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 modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (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
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License
  17  * 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  * The C code (not assembly, MMX, ...) of this file can be used
  21  * under the LGPL license.
  22  */
  23
  24 #undef REAL_MOVNTQ
  25 #undef MOVNTQ
  26 #undef PAVGB
  27 #undef PREFETCH
  28 #undef PREFETCHW
  29
  30 #if HAVE_AMD3DNOW
  31 #define PREFETCH  "prefetch"
  32 #define PREFETCHW "prefetchw"
  33 #elif HAVE_MMX2
  34 #define PREFETCH "prefetchnta"
  35 #define PREFETCHW "prefetcht0"
  36 #else
  37 #define PREFETCH  " # nop"
  38 #define PREFETCHW " # nop"
  39 #endif
  40
  41 #if HAVE_MMX2
  42 #define PAVGB(a,b) "pavgb " #a ", " #b " \n\t"
  43 #elif HAVE_AMD3DNOW
  44 #define PAVGB(a,b) "pavgusb " #a ", " #b " \n\t"
  45 #endif
  46
  47 #if HAVE_MMX2
  48 #define REAL_MOVNTQ(a,b) "movntq " #a ", " #b " \n\t"
  49 #else
  50 #define REAL_MOVNTQ(a,b) "movq " #a ", " #b " \n\t"
  51 #endif
  52 #define MOVNTQ(a,b)  REAL_MOVNTQ(a,b)
  53
  54 #if HAVE_ALTIVEC
  55 #include "ppc/swscale_altivec_template.c"
  56 #endif
  57
  58 #define YSCALEYUV2YV12X(x, offset, dest, width) \
  59     __asm__ volatile(\
  60     "xor                          %%"REG_a", %%"REG_a"  \n\t"\
  61     "movq             "VROUNDER_OFFSET"(%0), %%mm3      \n\t"\
  62     "movq                             %%mm3, %%mm4      \n\t"\
  63     "lea                     " offset "(%0), %%"REG_d"  \n\t"\
  64     "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
  65     ASMALIGN(4) /* FIXME Unroll? */\
  66     "1:                                                 \n\t"\
  67     "movq                      8(%%"REG_d"), %%mm0      \n\t" /* filterCoeff */\
  68     "movq   "  x "(%%"REG_S", %%"REG_a", 2), %%mm2      \n\t" /* srcData */\
  69     "movq 8+"  x "(%%"REG_S", %%"REG_a", 2), %%mm5      \n\t" /* srcData */\
  70     "add                                $16, %%"REG_d"  \n\t"\
  71     "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
  72     "test                         %%"REG_S", %%"REG_S"  \n\t"\
  73     "pmulhw                           %%mm0, %%mm2      \n\t"\
  74     "pmulhw                           %%mm0, %%mm5      \n\t"\
  75     "paddw                            %%mm2, %%mm3      \n\t"\
  76     "paddw                            %%mm5, %%mm4      \n\t"\
  77     " jnz                                1b             \n\t"\
  78     "psraw                               $3, %%mm3      \n\t"\
  79     "psraw                               $3, %%mm4      \n\t"\
  80     "packuswb                         %%mm4, %%mm3      \n\t"\
  81     MOVNTQ(%%mm3, (%1, %%REGa))\
  82     "add                                 $8, %%"REG_a"  \n\t"\
  83     "cmp                                 %2, %%"REG_a"  \n\t"\
  84     "movq             "VROUNDER_OFFSET"(%0), %%mm3      \n\t"\
  85     "movq                             %%mm3, %%mm4      \n\t"\
  86     "lea                     " offset "(%0), %%"REG_d"  \n\t"\
  87     "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
  88     "jb                                  1b             \n\t"\
  89     :: "r" (&c->redDither),\
  90     "r" (dest), "g" (width)\
  91     : "%"REG_a, "%"REG_d, "%"REG_S\
  92     );
  93
  94 #define YSCALEYUV2YV12X_ACCURATE(x, offset, dest, width) \
  95     __asm__ volatile(\
  96     "lea                     " offset "(%0), %%"REG_d"  \n\t"\
  97     "xor                          %%"REG_a", %%"REG_a"  \n\t"\
  98     "pxor                             %%mm4, %%mm4      \n\t"\
  99     "pxor                             %%mm5, %%mm5      \n\t"\
 100     "pxor                             %%mm6, %%mm6      \n\t"\
 101     "pxor                             %%mm7, %%mm7      \n\t"\
 102     "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
 103     ASMALIGN(4) \
 104     "1:                                                 \n\t"\
 105     "movq   "  x "(%%"REG_S", %%"REG_a", 2), %%mm0      \n\t" /* srcData */\
 106     "movq 8+"  x "(%%"REG_S", %%"REG_a", 2), %%mm2      \n\t" /* srcData */\
 107     "mov        "STR(APCK_PTR2)"(%%"REG_d"), %%"REG_S"  \n\t"\
 108     "movq   "  x "(%%"REG_S", %%"REG_a", 2), %%mm1      \n\t" /* srcData */\
 109     "movq                             %%mm0, %%mm3      \n\t"\
 110     "punpcklwd                        %%mm1, %%mm0      \n\t"\
 111     "punpckhwd                        %%mm1, %%mm3      \n\t"\
 112     "movq       "STR(APCK_COEF)"(%%"REG_d"), %%mm1      \n\t" /* filterCoeff */\
 113     "pmaddwd                          %%mm1, %%mm0      \n\t"\
 114     "pmaddwd                          %%mm1, %%mm3      \n\t"\
 115     "paddd                            %%mm0, %%mm4      \n\t"\
 116     "paddd                            %%mm3, %%mm5      \n\t"\
 117     "movq 8+"  x "(%%"REG_S", %%"REG_a", 2), %%mm3      \n\t" /* srcData */\
 118     "mov        "STR(APCK_SIZE)"(%%"REG_d"), %%"REG_S"  \n\t"\
 119     "add                  $"STR(APCK_SIZE)", %%"REG_d"  \n\t"\
 120     "test                         %%"REG_S", %%"REG_S"  \n\t"\
 121     "movq                             %%mm2, %%mm0      \n\t"\
 122     "punpcklwd                        %%mm3, %%mm2      \n\t"\
 123     "punpckhwd                        %%mm3, %%mm0      \n\t"\
 124     "pmaddwd                          %%mm1, %%mm2      \n\t"\
 125     "pmaddwd                          %%mm1, %%mm0      \n\t"\
 126     "paddd                            %%mm2, %%mm6      \n\t"\
 127     "paddd                            %%mm0, %%mm7      \n\t"\
 128     " jnz                                1b             \n\t"\
 129     "psrad                              $16, %%mm4      \n\t"\
 130     "psrad                              $16, %%mm5      \n\t"\
 131     "psrad                              $16, %%mm6      \n\t"\
 132     "psrad                              $16, %%mm7      \n\t"\
 133     "movq             "VROUNDER_OFFSET"(%0), %%mm0      \n\t"\
 134     "packssdw                         %%mm5, %%mm4      \n\t"\
 135     "packssdw                         %%mm7, %%mm6      \n\t"\
 136     "paddw                            %%mm0, %%mm4      \n\t"\
 137     "paddw                            %%mm0, %%mm6      \n\t"\
 138     "psraw                               $3, %%mm4      \n\t"\
 139     "psraw                               $3, %%mm6      \n\t"\
 140     "packuswb                         %%mm6, %%mm4      \n\t"\
 141     MOVNTQ(%%mm4, (%1, %%REGa))\
 142     "add                                 $8, %%"REG_a"  \n\t"\
 143     "cmp                                 %2, %%"REG_a"  \n\t"\
 144     "lea                     " offset "(%0), %%"REG_d"  \n\t"\
 145     "pxor                             %%mm4, %%mm4      \n\t"\
 146     "pxor                             %%mm5, %%mm5      \n\t"\
 147     "pxor                             %%mm6, %%mm6      \n\t"\
 148     "pxor                             %%mm7, %%mm7      \n\t"\
 149     "mov                        (%%"REG_d"), %%"REG_S"  \n\t"\
 150     "jb                                  1b             \n\t"\
 151     :: "r" (&c->redDither),\
 152     "r" (dest), "g" (width)\
 153     : "%"REG_a, "%"REG_d, "%"REG_S\
 154     );
 155
 156 #define YSCALEYUV2YV121 \
 157     "mov %2, %%"REG_a"                    \n\t"\
 158     ASMALIGN(4) /* FIXME Unroll? */\
 159     "1:                                   \n\t"\
 160     "movq  (%0, %%"REG_a", 2), %%mm0      \n\t"\
 161     "movq 8(%0, %%"REG_a", 2), %%mm1      \n\t"\
 162     "psraw                 $7, %%mm0      \n\t"\
 163     "psraw                 $7, %%mm1      \n\t"\
 164     "packuswb           %%mm1, %%mm0      \n\t"\
 165     MOVNTQ(%%mm0, (%1, %%REGa))\
 166     "add                   $8, %%"REG_a"  \n\t"\
 167     "jnc                   1b             \n\t"
 168
 169 #define YSCALEYUV2YV121_ACCURATE \
 170     "mov %2, %%"REG_a"                    \n\t"\
 171     "pcmpeqw %%mm7, %%mm7                 \n\t"\
 172     "psrlw                 $15, %%mm7     \n\t"\
 173     "psllw                  $6, %%mm7     \n\t"\
 174     ASMALIGN(4) /* FIXME Unroll? */\
 175     "1:                                   \n\t"\
 176     "movq  (%0, %%"REG_a", 2), %%mm0      \n\t"\
 177     "movq 8(%0, %%"REG_a", 2), %%mm1      \n\t"\
 178     "paddsw             %%mm7, %%mm0      \n\t"\
 179     "paddsw             %%mm7, %%mm1      \n\t"\
 180     "psraw                 $7, %%mm0      \n\t"\
 181     "psraw                 $7, %%mm1      \n\t"\
 182     "packuswb           %%mm1, %%mm0      \n\t"\
 183     MOVNTQ(%%mm0, (%1, %%REGa))\
 184     "add                   $8, %%"REG_a"  \n\t"\
 185     "jnc                   1b             \n\t"
 186
 187 /*
 188     :: "m" (-lumFilterSize), "m" (-chrFilterSize),
 189        "m" (lumMmxFilter+lumFilterSize*4), "m" (chrMmxFilter+chrFilterSize*4),
 190        "r" (dest), "m" (dstW),
 191        "m" (lumSrc+lumFilterSize), "m" (chrSrc+chrFilterSize)
 192     : "%eax", "%ebx", "%ecx", "%edx", "%esi"
 193 */
 194 #define YSCALEYUV2PACKEDX_UV \
 195     __asm__ volatile(\
 196     "xor                   %%"REG_a", %%"REG_a"     \n\t"\
 197     ASMALIGN(4)\
 198     "nop                                            \n\t"\
 199     "1:                                             \n\t"\
 200     "lea "CHR_MMX_FILTER_OFFSET"(%0), %%"REG_d"     \n\t"\
 201     "mov                 (%%"REG_d"), %%"REG_S"     \n\t"\
 202     "movq      "VROUNDER_OFFSET"(%0), %%mm3         \n\t"\
 203     "movq                      %%mm3, %%mm4         \n\t"\
 204     ASMALIGN(4)\
 205     "2:                                             \n\t"\
 206     "movq               8(%%"REG_d"), %%mm0         \n\t" /* filterCoeff */\
 207     "movq     (%%"REG_S", %%"REG_a"), %%mm2         \n\t" /* UsrcData */\
 208     "movq "AV_STRINGIFY(VOF)"(%%"REG_S", %%"REG_a"), %%mm5         \n\t" /* VsrcData */\
 209     "add                         $16, %%"REG_d"     \n\t"\
 210     "mov                 (%%"REG_d"), %%"REG_S"     \n\t"\
 211     "pmulhw                    %%mm0, %%mm2         \n\t"\
 212     "pmulhw                    %%mm0, %%mm5         \n\t"\
 213     "paddw                     %%mm2, %%mm3         \n\t"\
 214     "paddw                     %%mm5, %%mm4         \n\t"\
 215     "test                  %%"REG_S", %%"REG_S"     \n\t"\
 216     " jnz                         2b                \n\t"\
 217
 218 #define YSCALEYUV2PACKEDX_YA(offset,coeff,src1,src2,dst1,dst2) \
 219     "lea                "offset"(%0), %%"REG_d"     \n\t"\
 220     "mov                 (%%"REG_d"), %%"REG_S"     \n\t"\
 221     "movq      "VROUNDER_OFFSET"(%0), "#dst1"       \n\t"\
 222     "movq                    "#dst1", "#dst2"       \n\t"\
 223     ASMALIGN(4)\
 224     "2:                                             \n\t"\
 225     "movq               8(%%"REG_d"), "#coeff"      \n\t" /* filterCoeff */\
 226     "movq  (%%"REG_S", %%"REG_a", 2), "#src1"       \n\t" /* Y1srcData */\
 227     "movq 8(%%"REG_S", %%"REG_a", 2), "#src2"       \n\t" /* Y2srcData */\
 228     "add                         $16, %%"REG_d"            \n\t"\
 229     "mov                 (%%"REG_d"), %%"REG_S"     \n\t"\
 230     "pmulhw                 "#coeff", "#src1"       \n\t"\
 231     "pmulhw                 "#coeff", "#src2"       \n\t"\
 232     "paddw                   "#src1", "#dst1"       \n\t"\
 233     "paddw                   "#src2", "#dst2"       \n\t"\
 234     "test                  %%"REG_S", %%"REG_S"     \n\t"\
 235     " jnz                         2b                \n\t"\
 236
 237 #define YSCALEYUV2PACKEDX \
 238     YSCALEYUV2PACKEDX_UV \
 239     YSCALEYUV2PACKEDX_YA(LUM_MMX_FILTER_OFFSET,%%mm0,%%mm2,%%mm5,%%mm1,%%mm7) \
 240
 241 #define YSCALEYUV2PACKEDX_END                 \
 242     :: "r" (&c->redDither),                   \
 243         "m" (dummy), "m" (dummy), "m" (dummy),\
 244         "r" (dest), "m" (dstW)                \
 245     : "%"REG_a, "%"REG_d, "%"REG_S            \
 246     );
 247
 248 #define YSCALEYUV2PACKEDX_ACCURATE_UV \
 249     __asm__ volatile(\
 250     "xor %%"REG_a", %%"REG_a"                       \n\t"\
 251     ASMALIGN(4)\
 252     "nop                                            \n\t"\
 253     "1:                                             \n\t"\
 254     "lea "CHR_MMX_FILTER_OFFSET"(%0), %%"REG_d"     \n\t"\
 255     "mov                 (%%"REG_d"), %%"REG_S"     \n\t"\
 256     "pxor                      %%mm4, %%mm4         \n\t"\
 257     "pxor                      %%mm5, %%mm5         \n\t"\
 258     "pxor                      %%mm6, %%mm6         \n\t"\
 259     "pxor                      %%mm7, %%mm7         \n\t"\
 260     ASMALIGN(4)\
 261     "2:                                             \n\t"\
 262     "movq     (%%"REG_S", %%"REG_a"), %%mm0         \n\t" /* UsrcData */\
 263     "movq "AV_STRINGIFY(VOF)"(%%"REG_S", %%"REG_a"), %%mm2         \n\t" /* VsrcData */\
 264     "mov "STR(APCK_PTR2)"(%%"REG_d"), %%"REG_S"     \n\t"\
 265     "movq     (%%"REG_S", %%"REG_a"), %%mm1         \n\t" /* UsrcData */\
 266     "movq                      %%mm0, %%mm3         \n\t"\
 267     "punpcklwd                 %%mm1, %%mm0         \n\t"\
 268     "punpckhwd                 %%mm1, %%mm3         \n\t"\
 269     "movq "STR(APCK_COEF)"(%%"REG_d"),%%mm1         \n\t" /* filterCoeff */\
 270     "pmaddwd                   %%mm1, %%mm0         \n\t"\
 271     "pmaddwd                   %%mm1, %%mm3         \n\t"\
 272     "paddd                     %%mm0, %%mm4         \n\t"\
 273     "paddd                     %%mm3, %%mm5         \n\t"\
 274     "movq "AV_STRINGIFY(VOF)"(%%"REG_S", %%"REG_a"), %%mm3         \n\t" /* VsrcData */\
 275     "mov "STR(APCK_SIZE)"(%%"REG_d"), %%"REG_S"     \n\t"\
 276     "add           $"STR(APCK_SIZE)", %%"REG_d"     \n\t"\
 277     "test                  %%"REG_S", %%"REG_S"     \n\t"\
 278     "movq                      %%mm2, %%mm0         \n\t"\
 279     "punpcklwd                 %%mm3, %%mm2         \n\t"\
 280     "punpckhwd                 %%mm3, %%mm0         \n\t"\
 281     "pmaddwd                   %%mm1, %%mm2         \n\t"\
 282     "pmaddwd                   %%mm1, %%mm0         \n\t"\
 283     "paddd                     %%mm2, %%mm6         \n\t"\
 284     "paddd                     %%mm0, %%mm7         \n\t"\
 285     " jnz                         2b                \n\t"\
 286     "psrad                       $16, %%mm4         \n\t"\
 287     "psrad                       $16, %%mm5         \n\t"\
 288     "psrad                       $16, %%mm6         \n\t"\
 289     "psrad                       $16, %%mm7         \n\t"\
 290     "movq      "VROUNDER_OFFSET"(%0), %%mm0         \n\t"\
 291     "packssdw                  %%mm5, %%mm4         \n\t"\
 292     "packssdw                  %%mm7, %%mm6         \n\t"\
 293     "paddw                     %%mm0, %%mm4         \n\t"\
 294     "paddw                     %%mm0, %%mm6         \n\t"\
 295     "movq                      %%mm4, "U_TEMP"(%0)  \n\t"\
 296     "movq                      %%mm6, "V_TEMP"(%0)  \n\t"\
 297
 298 #define YSCALEYUV2PACKEDX_ACCURATE_YA(offset) \
 299     "lea                "offset"(%0), %%"REG_d"     \n\t"\
 300     "mov                 (%%"REG_d"), %%"REG_S"     \n\t"\
 301     "pxor                      %%mm1, %%mm1         \n\t"\
 302     "pxor                      %%mm5, %%mm5         \n\t"\
 303     "pxor                      %%mm7, %%mm7         \n\t"\
 304     "pxor                      %%mm6, %%mm6         \n\t"\
 305     ASMALIGN(4)\
 306     "2:                                             \n\t"\
 307     "movq  (%%"REG_S", %%"REG_a", 2), %%mm0         \n\t" /* Y1srcData */\
 308     "movq 8(%%"REG_S", %%"REG_a", 2), %%mm2         \n\t" /* Y2srcData */\
 309     "mov "STR(APCK_PTR2)"(%%"REG_d"), %%"REG_S"     \n\t"\
 310     "movq  (%%"REG_S", %%"REG_a", 2), %%mm4         \n\t" /* Y1srcData */\
 311     "movq                      %%mm0, %%mm3         \n\t"\
 312     "punpcklwd                 %%mm4, %%mm0         \n\t"\
 313     "punpckhwd                 %%mm4, %%mm3         \n\t"\
 314     "movq "STR(APCK_COEF)"(%%"REG_d"), %%mm4         \n\t" /* filterCoeff */\
 315     "pmaddwd                   %%mm4, %%mm0         \n\t"\
 316     "pmaddwd                   %%mm4, %%mm3         \n\t"\
 317     "paddd                     %%mm0, %%mm1         \n\t"\
 318     "paddd                     %%mm3, %%mm5         \n\t"\
 319     "movq 8(%%"REG_S", %%"REG_a", 2), %%mm3         \n\t" /* Y2srcData */\
 320     "mov "STR(APCK_SIZE)"(%%"REG_d"), %%"REG_S"     \n\t"\
 321     "add           $"STR(APCK_SIZE)", %%"REG_d"     \n\t"\
 322     "test                  %%"REG_S", %%"REG_S"     \n\t"\
 323     "movq                      %%mm2, %%mm0         \n\t"\
 324     "punpcklwd                 %%mm3, %%mm2         \n\t"\
 325     "punpckhwd                 %%mm3, %%mm0         \n\t"\
 326     "pmaddwd                   %%mm4, %%mm2         \n\t"\
 327     "pmaddwd                   %%mm4, %%mm0         \n\t"\
 328     "paddd                     %%mm2, %%mm7         \n\t"\
 329     "paddd                     %%mm0, %%mm6         \n\t"\
 330     " jnz                         2b                \n\t"\
 331     "psrad                       $16, %%mm1         \n\t"\
 332     "psrad                       $16, %%mm5         \n\t"\
 333     "psrad                       $16, %%mm7         \n\t"\
 334     "psrad                       $16, %%mm6         \n\t"\
 335     "movq      "VROUNDER_OFFSET"(%0), %%mm0         \n\t"\
 336     "packssdw                  %%mm5, %%mm1         \n\t"\
 337     "packssdw                  %%mm6, %%mm7         \n\t"\
 338     "paddw                     %%mm0, %%mm1         \n\t"\
 339     "paddw                     %%mm0, %%mm7         \n\t"\
 340     "movq               "U_TEMP"(%0), %%mm3         \n\t"\
 341     "movq               "V_TEMP"(%0), %%mm4         \n\t"\
 342
 343 #define YSCALEYUV2PACKEDX_ACCURATE \
 344     YSCALEYUV2PACKEDX_ACCURATE_UV \
 345     YSCALEYUV2PACKEDX_ACCURATE_YA(LUM_MMX_FILTER_OFFSET)
 346
 347 #define YSCALEYUV2RGBX \
 348     "psubw  "U_OFFSET"(%0), %%mm3       \n\t" /* (U-128)8*/\
 349     "psubw  "V_OFFSET"(%0), %%mm4       \n\t" /* (V-128)8*/\
 350     "movq            %%mm3, %%mm2       \n\t" /* (U-128)8*/\
 351     "movq            %%mm4, %%mm5       \n\t" /* (V-128)8*/\
 352     "pmulhw "UG_COEFF"(%0), %%mm3       \n\t"\
 353     "pmulhw "VG_COEFF"(%0), %%mm4       \n\t"\
 354 /* mm2=(U-128)8, mm3=ug, mm4=vg mm5=(V-128)8 */\
 355     "pmulhw "UB_COEFF"(%0), %%mm2       \n\t"\
 356     "pmulhw "VR_COEFF"(%0), %%mm5       \n\t"\
 357     "psubw  "Y_OFFSET"(%0), %%mm1       \n\t" /* 8(Y-16)*/\
 358     "psubw  "Y_OFFSET"(%0), %%mm7       \n\t" /* 8(Y-16)*/\
 359     "pmulhw  "Y_COEFF"(%0), %%mm1       \n\t"\
 360     "pmulhw  "Y_COEFF"(%0), %%mm7       \n\t"\
 361 /* mm1= Y1, mm2=ub, mm3=ug, mm4=vg mm5=vr, mm7=Y2 */\
 362     "paddw           %%mm3, %%mm4       \n\t"\
 363     "movq            %%mm2, %%mm0       \n\t"\
 364     "movq            %%mm5, %%mm6       \n\t"\
 365     "movq            %%mm4, %%mm3       \n\t"\
 366     "punpcklwd       %%mm2, %%mm2       \n\t"\
 367     "punpcklwd       %%mm5, %%mm5       \n\t"\
 368     "punpcklwd       %%mm4, %%mm4       \n\t"\
 369     "paddw           %%mm1, %%mm2       \n\t"\
 370     "paddw           %%mm1, %%mm5       \n\t"\
 371     "paddw           %%mm1, %%mm4       \n\t"\
 372     "punpckhwd       %%mm0, %%mm0       \n\t"\
 373     "punpckhwd       %%mm6, %%mm6       \n\t"\
 374     "punpckhwd       %%mm3, %%mm3       \n\t"\
 375     "paddw           %%mm7, %%mm0       \n\t"\
 376     "paddw           %%mm7, %%mm6       \n\t"\
 377     "paddw           %%mm7, %%mm3       \n\t"\
 378     /* mm0=B1, mm2=B2, mm3=G2, mm4=G1, mm5=R1, mm6=R2 */\
 379     "packuswb        %%mm0, %%mm2       \n\t"\
 380     "packuswb        %%mm6, %%mm5       \n\t"\
 381     "packuswb        %%mm3, %%mm4       \n\t"\
 382
 383 #define REAL_YSCALEYUV2PACKED(index, c) \
 384     "movq "CHR_MMX_FILTER_OFFSET"+8("#c"), %%mm0              \n\t"\
 385     "movq "LUM_MMX_FILTER_OFFSET"+8("#c"), %%mm1              \n\t"\
 386     "psraw                $3, %%mm0                           \n\t"\
 387     "psraw                $3, %%mm1                           \n\t"\
 388     "movq              %%mm0, "CHR_MMX_FILTER_OFFSET"+8("#c") \n\t"\
 389     "movq              %%mm1, "LUM_MMX_FILTER_OFFSET"+8("#c") \n\t"\
 390     "xor            "#index", "#index"                        \n\t"\
 391     ASMALIGN(4)\
 392     "1:                                 \n\t"\
 393     "movq     (%2, "#index"), %%mm2     \n\t" /* uvbuf0[eax]*/\
 394     "movq     (%3, "#index"), %%mm3     \n\t" /* uvbuf1[eax]*/\
 395     "movq "AV_STRINGIFY(VOF)"(%2, "#index"), %%mm5     \n\t" /* uvbuf0[eax+2048]*/\
 396     "movq "AV_STRINGIFY(VOF)"(%3, "#index"), %%mm4     \n\t" /* uvbuf1[eax+2048]*/\
 397     "psubw             %%mm3, %%mm2     \n\t" /* uvbuf0[eax] - uvbuf1[eax]*/\
 398     "psubw             %%mm4, %%mm5     \n\t" /* uvbuf0[eax+2048] - uvbuf1[eax+2048]*/\
 399     "movq "CHR_MMX_FILTER_OFFSET"+8("#c"), %%mm0    \n\t"\
 400     "pmulhw            %%mm0, %%mm2     \n\t" /* (uvbuf0[eax] - uvbuf1[eax])uvalpha1>>16*/\
 401     "pmulhw            %%mm0, %%mm5     \n\t" /* (uvbuf0[eax+2048] - uvbuf1[eax+2048])uvalpha1>>16*/\
 402     "psraw                $7, %%mm3     \n\t" /* uvbuf0[eax] - uvbuf1[eax] >>4*/\
 403     "psraw                $7, %%mm4     \n\t" /* uvbuf0[eax+2048] - uvbuf1[eax+2048] >>4*/\
 404     "paddw             %%mm2, %%mm3     \n\t" /* uvbuf0[eax]uvalpha1 - uvbuf1[eax](1-uvalpha1)*/\
 405     "paddw             %%mm5, %%mm4     \n\t" /* uvbuf0[eax+2048]uvalpha1 - uvbuf1[eax+2048](1-uvalpha1)*/\
 406     "movq  (%0, "#index", 2), %%mm0     \n\t" /*buf0[eax]*/\
 407     "movq  (%1, "#index", 2), %%mm1     \n\t" /*buf1[eax]*/\
 408     "movq 8(%0, "#index", 2), %%mm6     \n\t" /*buf0[eax]*/\
 409     "movq 8(%1, "#index", 2), %%mm7     \n\t" /*buf1[eax]*/\
 410     "psubw             %%mm1, %%mm0     \n\t" /* buf0[eax] - buf1[eax]*/\
 411     "psubw             %%mm7, %%mm6     \n\t" /* buf0[eax] - buf1[eax]*/\
 412     "pmulhw "LUM_MMX_FILTER_OFFSET"+8("#c"), %%mm0  \n\t" /* (buf0[eax] - buf1[eax])yalpha1>>16*/\
 413     "pmulhw "LUM_MMX_FILTER_OFFSET"+8("#c"), %%mm6  \n\t" /* (buf0[eax] - buf1[eax])yalpha1>>16*/\
 414     "psraw                $7, %%mm1     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 415     "psraw                $7, %%mm7     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 416     "paddw             %%mm0, %%mm1     \n\t" /* buf0[eax]yalpha1 + buf1[eax](1-yalpha1) >>16*/\
 417     "paddw             %%mm6, %%mm7     \n\t" /* buf0[eax]yalpha1 + buf1[eax](1-yalpha1) >>16*/\
 418
 419 #define YSCALEYUV2PACKED(index, c)  REAL_YSCALEYUV2PACKED(index, c)
 420
 421 #define REAL_YSCALEYUV2RGB_UV(index, c) \
 422     "xor            "#index", "#index"  \n\t"\
 423     ASMALIGN(4)\
 424     "1:                                 \n\t"\
 425     "movq     (%2, "#index"), %%mm2     \n\t" /* uvbuf0[eax]*/\
 426     "movq     (%3, "#index"), %%mm3     \n\t" /* uvbuf1[eax]*/\
 427     "movq "AV_STRINGIFY(VOF)"(%2, "#index"), %%mm5     \n\t" /* uvbuf0[eax+2048]*/\
 428     "movq "AV_STRINGIFY(VOF)"(%3, "#index"), %%mm4     \n\t" /* uvbuf1[eax+2048]*/\
 429     "psubw             %%mm3, %%mm2     \n\t" /* uvbuf0[eax] - uvbuf1[eax]*/\
 430     "psubw             %%mm4, %%mm5     \n\t" /* uvbuf0[eax+2048] - uvbuf1[eax+2048]*/\
 431     "movq "CHR_MMX_FILTER_OFFSET"+8("#c"), %%mm0    \n\t"\
 432     "pmulhw            %%mm0, %%mm2     \n\t" /* (uvbuf0[eax] - uvbuf1[eax])uvalpha1>>16*/\
 433     "pmulhw            %%mm0, %%mm5     \n\t" /* (uvbuf0[eax+2048] - uvbuf1[eax+2048])uvalpha1>>16*/\
 434     "psraw                $4, %%mm3     \n\t" /* uvbuf0[eax] - uvbuf1[eax] >>4*/\
 435     "psraw                $4, %%mm4     \n\t" /* uvbuf0[eax+2048] - uvbuf1[eax+2048] >>4*/\
 436     "paddw             %%mm2, %%mm3     \n\t" /* uvbuf0[eax]uvalpha1 - uvbuf1[eax](1-uvalpha1)*/\
 437     "paddw             %%mm5, %%mm4     \n\t" /* uvbuf0[eax+2048]uvalpha1 - uvbuf1[eax+2048](1-uvalpha1)*/\
 438     "psubw  "U_OFFSET"("#c"), %%mm3     \n\t" /* (U-128)8*/\
 439     "psubw  "V_OFFSET"("#c"), %%mm4     \n\t" /* (V-128)8*/\
 440     "movq              %%mm3, %%mm2     \n\t" /* (U-128)8*/\
 441     "movq              %%mm4, %%mm5     \n\t" /* (V-128)8*/\
 442     "pmulhw "UG_COEFF"("#c"), %%mm3     \n\t"\
 443     "pmulhw "VG_COEFF"("#c"), %%mm4     \n\t"\
 444     /* mm2=(U-128)8, mm3=ug, mm4=vg mm5=(V-128)8 */\
 445
 446 #define REAL_YSCALEYUV2RGB_YA(index, c, b1, b2) \
 447     "movq  ("#b1", "#index", 2), %%mm0     \n\t" /*buf0[eax]*/\
 448     "movq  ("#b2", "#index", 2), %%mm1     \n\t" /*buf1[eax]*/\
 449     "movq 8("#b1", "#index", 2), %%mm6     \n\t" /*buf0[eax]*/\
 450     "movq 8("#b2", "#index", 2), %%mm7     \n\t" /*buf1[eax]*/\
 451     "psubw             %%mm1, %%mm0     \n\t" /* buf0[eax] - buf1[eax]*/\
 452     "psubw             %%mm7, %%mm6     \n\t" /* buf0[eax] - buf1[eax]*/\
 453     "pmulhw "LUM_MMX_FILTER_OFFSET"+8("#c"), %%mm0  \n\t" /* (buf0[eax] - buf1[eax])yalpha1>>16*/\
 454     "pmulhw "LUM_MMX_FILTER_OFFSET"+8("#c"), %%mm6  \n\t" /* (buf0[eax] - buf1[eax])yalpha1>>16*/\
 455     "psraw                $4, %%mm1     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 456     "psraw                $4, %%mm7     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 457     "paddw             %%mm0, %%mm1     \n\t" /* buf0[eax]yalpha1 + buf1[eax](1-yalpha1) >>16*/\
 458     "paddw             %%mm6, %%mm7     \n\t" /* buf0[eax]yalpha1 + buf1[eax](1-yalpha1) >>16*/\
 459
 460 #define REAL_YSCALEYUV2RGB_COEFF(c) \
 461     "pmulhw "UB_COEFF"("#c"), %%mm2     \n\t"\
 462     "pmulhw "VR_COEFF"("#c"), %%mm5     \n\t"\
 463     "psubw  "Y_OFFSET"("#c"), %%mm1     \n\t" /* 8(Y-16)*/\
 464     "psubw  "Y_OFFSET"("#c"), %%mm7     \n\t" /* 8(Y-16)*/\
 465     "pmulhw  "Y_COEFF"("#c"), %%mm1     \n\t"\
 466     "pmulhw  "Y_COEFF"("#c"), %%mm7     \n\t"\
 467     /* mm1= Y1, mm2=ub, mm3=ug, mm4=vg mm5=vr, mm7=Y2 */\
 468     "paddw             %%mm3, %%mm4     \n\t"\
 469     "movq              %%mm2, %%mm0     \n\t"\
 470     "movq              %%mm5, %%mm6     \n\t"\
 471     "movq              %%mm4, %%mm3     \n\t"\
 472     "punpcklwd         %%mm2, %%mm2     \n\t"\
 473     "punpcklwd         %%mm5, %%mm5     \n\t"\
 474     "punpcklwd         %%mm4, %%mm4     \n\t"\
 475     "paddw             %%mm1, %%mm2     \n\t"\
 476     "paddw             %%mm1, %%mm5     \n\t"\
 477     "paddw             %%mm1, %%mm4     \n\t"\
 478     "punpckhwd         %%mm0, %%mm0     \n\t"\
 479     "punpckhwd         %%mm6, %%mm6     \n\t"\
 480     "punpckhwd         %%mm3, %%mm3     \n\t"\
 481     "paddw             %%mm7, %%mm0     \n\t"\
 482     "paddw             %%mm7, %%mm6     \n\t"\
 483     "paddw             %%mm7, %%mm3     \n\t"\
 484     /* mm0=B1, mm2=B2, mm3=G2, mm4=G1, mm5=R1, mm6=R2 */\
 485     "packuswb          %%mm0, %%mm2     \n\t"\
 486     "packuswb          %%mm6, %%mm5     \n\t"\
 487     "packuswb          %%mm3, %%mm4     \n\t"\
 488
 489 #define YSCALEYUV2RGB_YA(index, c, b1, b2) REAL_YSCALEYUV2RGB_YA(index, c, b1, b2)
 490
 491 #define YSCALEYUV2RGB(index, c) \
 492     REAL_YSCALEYUV2RGB_UV(index, c) \
 493     REAL_YSCALEYUV2RGB_YA(index, c, %0, %1) \
 494     REAL_YSCALEYUV2RGB_COEFF(c)
 495
 496 #define REAL_YSCALEYUV2PACKED1(index, c) \
 497     "xor            "#index", "#index"  \n\t"\
 498     ASMALIGN(4)\
 499     "1:                                 \n\t"\
 500     "movq     (%2, "#index"), %%mm3     \n\t" /* uvbuf0[eax]*/\
 501     "movq "AV_STRINGIFY(VOF)"(%2, "#index"), %%mm4     \n\t" /* uvbuf0[eax+2048]*/\
 502     "psraw                $7, %%mm3     \n\t" \
 503     "psraw                $7, %%mm4     \n\t" \
 504     "movq  (%0, "#index", 2), %%mm1     \n\t" /*buf0[eax]*/\
 505     "movq 8(%0, "#index", 2), %%mm7     \n\t" /*buf0[eax]*/\
 506     "psraw                $7, %%mm1     \n\t" \
 507     "psraw                $7, %%mm7     \n\t" \
 508
 509 #define YSCALEYUV2PACKED1(index, c)  REAL_YSCALEYUV2PACKED1(index, c)
 510
 511 #define REAL_YSCALEYUV2RGB1(index, c) \
 512     "xor            "#index", "#index"  \n\t"\
 513     ASMALIGN(4)\
 514     "1:                                 \n\t"\
 515     "movq     (%2, "#index"), %%mm3     \n\t" /* uvbuf0[eax]*/\
 516     "movq "AV_STRINGIFY(VOF)"(%2, "#index"), %%mm4     \n\t" /* uvbuf0[eax+2048]*/\
 517     "psraw                $4, %%mm3     \n\t" /* uvbuf0[eax] - uvbuf1[eax] >>4*/\
 518     "psraw                $4, %%mm4     \n\t" /* uvbuf0[eax+2048] - uvbuf1[eax+2048] >>4*/\
 519     "psubw  "U_OFFSET"("#c"), %%mm3     \n\t" /* (U-128)8*/\
 520     "psubw  "V_OFFSET"("#c"), %%mm4     \n\t" /* (V-128)8*/\
 521     "movq              %%mm3, %%mm2     \n\t" /* (U-128)8*/\
 522     "movq              %%mm4, %%mm5     \n\t" /* (V-128)8*/\
 523     "pmulhw "UG_COEFF"("#c"), %%mm3     \n\t"\
 524     "pmulhw "VG_COEFF"("#c"), %%mm4     \n\t"\
 525     /* mm2=(U-128)8, mm3=ug, mm4=vg mm5=(V-128)8 */\
 526     "movq  (%0, "#index", 2), %%mm1     \n\t" /*buf0[eax]*/\
 527     "movq 8(%0, "#index", 2), %%mm7     \n\t" /*buf0[eax]*/\
 528     "psraw                $4, %%mm1     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 529     "psraw                $4, %%mm7     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 530     "pmulhw "UB_COEFF"("#c"), %%mm2     \n\t"\
 531     "pmulhw "VR_COEFF"("#c"), %%mm5     \n\t"\
 532     "psubw  "Y_OFFSET"("#c"), %%mm1     \n\t" /* 8(Y-16)*/\
 533     "psubw  "Y_OFFSET"("#c"), %%mm7     \n\t" /* 8(Y-16)*/\
 534     "pmulhw  "Y_COEFF"("#c"), %%mm1     \n\t"\
 535     "pmulhw  "Y_COEFF"("#c"), %%mm7     \n\t"\
 536     /* mm1= Y1, mm2=ub, mm3=ug, mm4=vg mm5=vr, mm7=Y2 */\
 537     "paddw             %%mm3, %%mm4     \n\t"\
 538     "movq              %%mm2, %%mm0     \n\t"\
 539     "movq              %%mm5, %%mm6     \n\t"\
 540     "movq              %%mm4, %%mm3     \n\t"\
 541     "punpcklwd         %%mm2, %%mm2     \n\t"\
 542     "punpcklwd         %%mm5, %%mm5     \n\t"\
 543     "punpcklwd         %%mm4, %%mm4     \n\t"\
 544     "paddw             %%mm1, %%mm2     \n\t"\
 545     "paddw             %%mm1, %%mm5     \n\t"\
 546     "paddw             %%mm1, %%mm4     \n\t"\
 547     "punpckhwd         %%mm0, %%mm0     \n\t"\
 548     "punpckhwd         %%mm6, %%mm6     \n\t"\
 549     "punpckhwd         %%mm3, %%mm3     \n\t"\
 550     "paddw             %%mm7, %%mm0     \n\t"\
 551     "paddw             %%mm7, %%mm6     \n\t"\
 552     "paddw             %%mm7, %%mm3     \n\t"\
 553     /* mm0=B1, mm2=B2, mm3=G2, mm4=G1, mm5=R1, mm6=R2 */\
 554     "packuswb          %%mm0, %%mm2     \n\t"\
 555     "packuswb          %%mm6, %%mm5     \n\t"\
 556     "packuswb          %%mm3, %%mm4     \n\t"\
 557
 558 #define YSCALEYUV2RGB1(index, c)  REAL_YSCALEYUV2RGB1(index, c)
 559
 560 #define REAL_YSCALEYUV2PACKED1b(index, c) \
 561     "xor "#index", "#index"             \n\t"\
 562     ASMALIGN(4)\
 563     "1:                                 \n\t"\
 564     "movq     (%2, "#index"), %%mm2     \n\t" /* uvbuf0[eax]*/\
 565     "movq     (%3, "#index"), %%mm3     \n\t" /* uvbuf1[eax]*/\
 566     "movq "AV_STRINGIFY(VOF)"(%2, "#index"), %%mm5     \n\t" /* uvbuf0[eax+2048]*/\
 567     "movq "AV_STRINGIFY(VOF)"(%3, "#index"), %%mm4     \n\t" /* uvbuf1[eax+2048]*/\
 568     "paddw             %%mm2, %%mm3     \n\t" /* uvbuf0[eax] + uvbuf1[eax]*/\
 569     "paddw             %%mm5, %%mm4     \n\t" /* uvbuf0[eax+2048] + uvbuf1[eax+2048]*/\
 570     "psrlw                $8, %%mm3     \n\t" \
 571     "psrlw                $8, %%mm4     \n\t" \
 572     "movq  (%0, "#index", 2), %%mm1     \n\t" /*buf0[eax]*/\
 573     "movq 8(%0, "#index", 2), %%mm7     \n\t" /*buf0[eax]*/\
 574     "psraw                $7, %%mm1     \n\t" \
 575     "psraw                $7, %%mm7     \n\t"
 576 #define YSCALEYUV2PACKED1b(index, c)  REAL_YSCALEYUV2PACKED1b(index, c)
 577
 578 // do vertical chrominance interpolation
 579 #define REAL_YSCALEYUV2RGB1b(index, c) \
 580     "xor            "#index", "#index"  \n\t"\
 581     ASMALIGN(4)\
 582     "1:                                 \n\t"\
 583     "movq     (%2, "#index"), %%mm2     \n\t" /* uvbuf0[eax]*/\
 584     "movq     (%3, "#index"), %%mm3     \n\t" /* uvbuf1[eax]*/\
 585     "movq "AV_STRINGIFY(VOF)"(%2, "#index"), %%mm5     \n\t" /* uvbuf0[eax+2048]*/\
 586     "movq "AV_STRINGIFY(VOF)"(%3, "#index"), %%mm4     \n\t" /* uvbuf1[eax+2048]*/\
 587     "paddw             %%mm2, %%mm3     \n\t" /* uvbuf0[eax] + uvbuf1[eax]*/\
 588     "paddw             %%mm5, %%mm4     \n\t" /* uvbuf0[eax+2048] + uvbuf1[eax+2048]*/\
 589     "psrlw                $5, %%mm3     \n\t" /*FIXME might overflow*/\
 590     "psrlw                $5, %%mm4     \n\t" /*FIXME might overflow*/\
 591     "psubw  "U_OFFSET"("#c"), %%mm3     \n\t" /* (U-128)8*/\
 592     "psubw  "V_OFFSET"("#c"), %%mm4     \n\t" /* (V-128)8*/\
 593     "movq              %%mm3, %%mm2     \n\t" /* (U-128)8*/\
 594     "movq              %%mm4, %%mm5     \n\t" /* (V-128)8*/\
 595     "pmulhw "UG_COEFF"("#c"), %%mm3     \n\t"\
 596     "pmulhw "VG_COEFF"("#c"), %%mm4     \n\t"\
 597     /* mm2=(U-128)8, mm3=ug, mm4=vg mm5=(V-128)8 */\
 598     "movq  (%0, "#index", 2), %%mm1     \n\t" /*buf0[eax]*/\
 599     "movq 8(%0, "#index", 2), %%mm7     \n\t" /*buf0[eax]*/\
 600     "psraw                $4, %%mm1     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 601     "psraw                $4, %%mm7     \n\t" /* buf0[eax] - buf1[eax] >>4*/\
 602     "pmulhw "UB_COEFF"("#c"), %%mm2     \n\t"\
 603     "pmulhw "VR_COEFF"("#c"), %%mm5     \n\t"\
 604     "psubw  "Y_OFFSET"("#c"), %%mm1     \n\t" /* 8(Y-16)*/\
 605     "psubw  "Y_OFFSET"("#c"), %%mm7     \n\t" /* 8(Y-16)*/\
 606     "pmulhw  "Y_COEFF"("#c"), %%mm1     \n\t"\
 607     "pmulhw  "Y_COEFF"("#c"), %%mm7     \n\t"\
 608     /* mm1= Y1, mm2=ub, mm3=ug, mm4=vg mm5=vr, mm7=Y2 */\
 609     "paddw             %%mm3, %%mm4     \n\t"\
 610     "movq              %%mm2, %%mm0     \n\t"\
 611     "movq              %%mm5, %%mm6     \n\t"\
 612     "movq              %%mm4, %%mm3     \n\t"\
 613     "punpcklwd         %%mm2, %%mm2     \n\t"\
 614     "punpcklwd         %%mm5, %%mm5     \n\t"\
 615     "punpcklwd         %%mm4, %%mm4     \n\t"\
 616     "paddw             %%mm1, %%mm2     \n\t"\
 617     "paddw             %%mm1, %%mm5     \n\t"\
 618     "paddw             %%mm1, %%mm4     \n\t"\
 619     "punpckhwd         %%mm0, %%mm0     \n\t"\
 620     "punpckhwd         %%mm6, %%mm6     \n\t"\
 621     "punpckhwd         %%mm3, %%mm3     \n\t"\
 622     "paddw             %%mm7, %%mm0     \n\t"\
 623     "paddw             %%mm7, %%mm6     \n\t"\
 624     "paddw             %%mm7, %%mm3     \n\t"\
 625     /* mm0=B1, mm2=B2, mm3=G2, mm4=G1, mm5=R1, mm6=R2 */\
 626     "packuswb          %%mm0, %%mm2     \n\t"\
 627     "packuswb          %%mm6, %%mm5     \n\t"\
 628     "packuswb          %%mm3, %%mm4     \n\t"\
 629
 630 #define YSCALEYUV2RGB1b(index, c)  REAL_YSCALEYUV2RGB1b(index, c)
 631
 632 #define REAL_YSCALEYUV2RGB1_ALPHA(index) \
 633     "movq  (%1, "#index", 2), %%mm7     \n\t" /* abuf0[index  ]     */\
 634     "movq 8(%1, "#index", 2), %%mm1     \n\t" /* abuf0[index+4]     */\
 635     "psraw                $7, %%mm7     \n\t" /* abuf0[index  ] >>7 */\
 636     "psraw                $7, %%mm1     \n\t" /* abuf0[index+4] >>7 */\
 637     "packuswb          %%mm1, %%mm7     \n\t"
 638 #define YSCALEYUV2RGB1_ALPHA(index) REAL_YSCALEYUV2RGB1_ALPHA(index)
 639
 640 #define REAL_WRITEBGR32(dst, dstw, index, b, g, r, a, q0, q2, q3, t) \
 641     "movq       "#b", "#q2"     \n\t" /* B */\
 642     "movq       "#r", "#t"      \n\t" /* R */\
 643     "punpcklbw  "#g", "#b"      \n\t" /* GBGBGBGB 0 */\
 644     "punpcklbw  "#a", "#r"      \n\t" /* ARARARAR 0 */\
 645     "punpckhbw  "#g", "#q2"     \n\t" /* GBGBGBGB 2 */\
 646     "punpckhbw  "#a", "#t"      \n\t" /* ARARARAR 2 */\
 647     "movq       "#b", "#q0"     \n\t" /* GBGBGBGB 0 */\
 648     "movq      "#q2", "#q3"     \n\t" /* GBGBGBGB 2 */\
 649     "punpcklwd  "#r", "#q0"     \n\t" /* ARGBARGB 0 */\
 650     "punpckhwd  "#r", "#b"      \n\t" /* ARGBARGB 1 */\
 651     "punpcklwd  "#t", "#q2"     \n\t" /* ARGBARGB 2 */\
 652     "punpckhwd  "#t", "#q3"     \n\t" /* ARGBARGB 3 */\
 653 \
 654     MOVNTQ(   q0,   (dst, index, 4))\
 655     MOVNTQ(    b,  8(dst, index, 4))\
 656     MOVNTQ(   q2, 16(dst, index, 4))\
 657     MOVNTQ(   q3, 24(dst, index, 4))\
 658 \
 659     "add      $8, "#index"      \n\t"\
 660     "cmp "#dstw", "#index"      \n\t"\
 661     " jb      1b                \n\t"
 662 #define WRITEBGR32(dst, dstw, index, b, g, r, a, q0, q2, q3, t)  REAL_WRITEBGR32(dst, dstw, index, b, g, r, a, q0, q2, q3, t)
 663
 664 #define REAL_WRITERGB16(dst, dstw, index) \
 665     "pand "MANGLE(bF8)", %%mm2  \n\t" /* B */\
 666     "pand "MANGLE(bFC)", %%mm4  \n\t" /* G */\
 667     "pand "MANGLE(bF8)", %%mm5  \n\t" /* R */\
 668     "psrlq           $3, %%mm2  \n\t"\
 669 \
 670     "movq         %%mm2, %%mm1  \n\t"\
 671     "movq         %%mm4, %%mm3  \n\t"\
 672 \
 673     "punpcklbw    %%mm7, %%mm3  \n\t"\
 674     "punpcklbw    %%mm5, %%mm2  \n\t"\
 675     "punpckhbw    %%mm7, %%mm4  \n\t"\
 676     "punpckhbw    %%mm5, %%mm1  \n\t"\
 677 \
 678     "psllq           $3, %%mm3  \n\t"\
 679     "psllq           $3, %%mm4  \n\t"\
 680 \
 681     "por          %%mm3, %%mm2  \n\t"\
 682     "por          %%mm4, %%mm1  \n\t"\
 683 \
 684     MOVNTQ(%%mm2,  (dst, index, 2))\
 685     MOVNTQ(%%mm1, 8(dst, index, 2))\
 686 \
 687     "add             $8, "#index"   \n\t"\
 688     "cmp        "#dstw", "#index"   \n\t"\
 689     " jb             1b             \n\t"
 690 #define WRITERGB16(dst, dstw, index)  REAL_WRITERGB16(dst, dstw, index)
 691
 692 #define REAL_WRITERGB15(dst, dstw, index) \
 693     "pand "MANGLE(bF8)", %%mm2  \n\t" /* B */\
 694     "pand "MANGLE(bF8)", %%mm4  \n\t" /* G */\
 695     "pand "MANGLE(bF8)", %%mm5  \n\t" /* R */\
 696     "psrlq           $3, %%mm2  \n\t"\
 697     "psrlq           $1, %%mm5  \n\t"\
 698 \
 699     "movq         %%mm2, %%mm1  \n\t"\
 700     "movq         %%mm4, %%mm3  \n\t"\
 701 \
 702     "punpcklbw    %%mm7, %%mm3  \n\t"\
 703     "punpcklbw    %%mm5, %%mm2  \n\t"\
 704     "punpckhbw    %%mm7, %%mm4  \n\t"\
 705     "punpckhbw    %%mm5, %%mm1  \n\t"\
 706 \
 707     "psllq           $2, %%mm3  \n\t"\
 708     "psllq           $2, %%mm4  \n\t"\
 709 \
 710     "por          %%mm3, %%mm2  \n\t"\
 711     "por          %%mm4, %%mm1  \n\t"\
 712 \
 713     MOVNTQ(%%mm2,  (dst, index, 2))\
 714     MOVNTQ(%%mm1, 8(dst, index, 2))\
 715 \
 716     "add             $8, "#index"   \n\t"\
 717     "cmp        "#dstw", "#index"   \n\t"\
 718     " jb             1b             \n\t"
 719 #define WRITERGB15(dst, dstw, index)  REAL_WRITERGB15(dst, dstw, index)
 720
 721 #define WRITEBGR24OLD(dst, dstw, index) \
 722     /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */\
 723     "movq      %%mm2, %%mm1             \n\t" /* B */\
 724     "movq      %%mm5, %%mm6             \n\t" /* R */\
 725     "punpcklbw %%mm4, %%mm2             \n\t" /* GBGBGBGB 0 */\
 726     "punpcklbw %%mm7, %%mm5             \n\t" /* 0R0R0R0R 0 */\
 727     "punpckhbw %%mm4, %%mm1             \n\t" /* GBGBGBGB 2 */\
 728     "punpckhbw %%mm7, %%mm6             \n\t" /* 0R0R0R0R 2 */\
 729     "movq      %%mm2, %%mm0             \n\t" /* GBGBGBGB 0 */\
 730     "movq      %%mm1, %%mm3             \n\t" /* GBGBGBGB 2 */\
 731     "punpcklwd %%mm5, %%mm0             \n\t" /* 0RGB0RGB 0 */\
 732     "punpckhwd %%mm5, %%mm2             \n\t" /* 0RGB0RGB 1 */\
 733     "punpcklwd %%mm6, %%mm1             \n\t" /* 0RGB0RGB 2 */\
 734     "punpckhwd %%mm6, %%mm3             \n\t" /* 0RGB0RGB 3 */\
 735 \
 736     "movq      %%mm0, %%mm4             \n\t" /* 0RGB0RGB 0 */\
 737     "psrlq        $8, %%mm0             \n\t" /* 00RGB0RG 0 */\
 738     "pand "MANGLE(bm00000111)", %%mm4   \n\t" /* 00000RGB 0 */\
 739     "pand "MANGLE(bm11111000)", %%mm0   \n\t" /* 00RGB000 0.5 */\
 740     "por       %%mm4, %%mm0             \n\t" /* 00RGBRGB 0 */\
 741     "movq      %%mm2, %%mm4             \n\t" /* 0RGB0RGB 1 */\
 742     "psllq       $48, %%mm2             \n\t" /* GB000000 1 */\
 743     "por       %%mm2, %%mm0             \n\t" /* GBRGBRGB 0 */\
 744 \
 745     "movq      %%mm4, %%mm2             \n\t" /* 0RGB0RGB 1 */\
 746     "psrld       $16, %%mm4             \n\t" /* 000R000R 1 */\
 747     "psrlq       $24, %%mm2             \n\t" /* 0000RGB0 1.5 */\
 748     "por       %%mm4, %%mm2             \n\t" /* 000RRGBR 1 */\
 749     "pand "MANGLE(bm00001111)", %%mm2   \n\t" /* 0000RGBR 1 */\
 750     "movq      %%mm1, %%mm4             \n\t" /* 0RGB0RGB 2 */\
 751     "psrlq        $8, %%mm1             \n\t" /* 00RGB0RG 2 */\
 752     "pand "MANGLE(bm00000111)", %%mm4   \n\t" /* 00000RGB 2 */\
 753     "pand "MANGLE(bm11111000)", %%mm1   \n\t" /* 00RGB000 2.5 */\
 754     "por       %%mm4, %%mm1             \n\t" /* 00RGBRGB 2 */\
 755     "movq      %%mm1, %%mm4             \n\t" /* 00RGBRGB 2 */\
 756     "psllq       $32, %%mm1             \n\t" /* BRGB0000 2 */\
 757     "por       %%mm1, %%mm2             \n\t" /* BRGBRGBR 1 */\
 758 \
 759     "psrlq       $32, %%mm4             \n\t" /* 000000RG 2.5 */\
 760     "movq      %%mm3, %%mm5             \n\t" /* 0RGB0RGB 3 */\
 761     "psrlq        $8, %%mm3             \n\t" /* 00RGB0RG 3 */\
 762     "pand "MANGLE(bm00000111)", %%mm5   \n\t" /* 00000RGB 3 */\
 763     "pand "MANGLE(bm11111000)", %%mm3   \n\t" /* 00RGB000 3.5 */\
 764     "por       %%mm5, %%mm3             \n\t" /* 00RGBRGB 3 */\
 765     "psllq       $16, %%mm3             \n\t" /* RGBRGB00 3 */\
 766     "por       %%mm4, %%mm3             \n\t" /* RGBRGBRG 2.5 */\
 767 \
 768     MOVNTQ(%%mm0,   (dst))\
 769     MOVNTQ(%%mm2,  8(dst))\
 770     MOVNTQ(%%mm3, 16(dst))\
 771     "add         $24, "#dst"            \n\t"\
 772 \
 773     "add          $8, "#index"          \n\t"\
 774     "cmp     "#dstw", "#index"          \n\t"\
 775     " jb          1b                    \n\t"
 776
 777 #define WRITEBGR24MMX(dst, dstw, index) \
 778     /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */\
 779     "movq      %%mm2, %%mm1     \n\t" /* B */\
 780     "movq      %%mm5, %%mm6     \n\t" /* R */\
 781     "punpcklbw %%mm4, %%mm2     \n\t" /* GBGBGBGB 0 */\
 782     "punpcklbw %%mm7, %%mm5     \n\t" /* 0R0R0R0R 0 */\
 783     "punpckhbw %%mm4, %%mm1     \n\t" /* GBGBGBGB 2 */\
 784     "punpckhbw %%mm7, %%mm6     \n\t" /* 0R0R0R0R 2 */\
 785     "movq      %%mm2, %%mm0     \n\t" /* GBGBGBGB 0 */\
 786     "movq      %%mm1, %%mm3     \n\t" /* GBGBGBGB 2 */\
 787     "punpcklwd %%mm5, %%mm0     \n\t" /* 0RGB0RGB 0 */\
 788     "punpckhwd %%mm5, %%mm2     \n\t" /* 0RGB0RGB 1 */\
 789     "punpcklwd %%mm6, %%mm1     \n\t" /* 0RGB0RGB 2 */\
 790     "punpckhwd %%mm6, %%mm3     \n\t" /* 0RGB0RGB 3 */\
 791 \
 792     "movq      %%mm0, %%mm4     \n\t" /* 0RGB0RGB 0 */\
 793     "movq      %%mm2, %%mm6     \n\t" /* 0RGB0RGB 1 */\
 794     "movq      %%mm1, %%mm5     \n\t" /* 0RGB0RGB 2 */\
 795     "movq      %%mm3, %%mm7     \n\t" /* 0RGB0RGB 3 */\
 796 \
 797     "psllq       $40, %%mm0     \n\t" /* RGB00000 0 */\
 798     "psllq       $40, %%mm2     \n\t" /* RGB00000 1 */\
 799     "psllq       $40, %%mm1     \n\t" /* RGB00000 2 */\
 800     "psllq       $40, %%mm3     \n\t" /* RGB00000 3 */\
 801 \
 802     "punpckhdq %%mm4, %%mm0     \n\t" /* 0RGBRGB0 0 */\
 803     "punpckhdq %%mm6, %%mm2     \n\t" /* 0RGBRGB0 1 */\
 804     "punpckhdq %%mm5, %%mm1     \n\t" /* 0RGBRGB0 2 */\
 805     "punpckhdq %%mm7, %%mm3     \n\t" /* 0RGBRGB0 3 */\
 806 \
 807     "psrlq        $8, %%mm0     \n\t" /* 00RGBRGB 0 */\
 808     "movq      %%mm2, %%mm6     \n\t" /* 0RGBRGB0 1 */\
 809     "psllq       $40, %%mm2     \n\t" /* GB000000 1 */\
 810     "por       %%mm2, %%mm0     \n\t" /* GBRGBRGB 0 */\
 811     MOVNTQ(%%mm0, (dst))\
 812 \
 813     "psrlq       $24, %%mm6     \n\t" /* 0000RGBR 1 */\
 814     "movq      %%mm1, %%mm5     \n\t" /* 0RGBRGB0 2 */\
 815     "psllq       $24, %%mm1     \n\t" /* BRGB0000 2 */\
 816     "por       %%mm1, %%mm6     \n\t" /* BRGBRGBR 1 */\
 817     MOVNTQ(%%mm6, 8(dst))\
 818 \
 819     "psrlq       $40, %%mm5     \n\t" /* 000000RG 2 */\
 820     "psllq        $8, %%mm3     \n\t" /* RGBRGB00 3 */\
 821     "por       %%mm3, %%mm5     \n\t" /* RGBRGBRG 2 */\
 822     MOVNTQ(%%mm5, 16(dst))\
 823 \
 824     "add         $24, "#dst"    \n\t"\
 825 \
 826     "add          $8, "#index"  \n\t"\
 827     "cmp     "#dstw", "#index"  \n\t"\
 828     " jb          1b            \n\t"
 829
 830 #define WRITEBGR24MMX2(dst, dstw, index) \
 831     /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */\
 832     "movq "MANGLE(ff_M24A)", %%mm0 \n\t"\
 833     "movq "MANGLE(ff_M24C)", %%mm7 \n\t"\
 834     "pshufw $0x50, %%mm2, %%mm1 \n\t" /* B3 B2 B3 B2  B1 B0 B1 B0 */\
 835     "pshufw $0x50, %%mm4, %%mm3 \n\t" /* G3 G2 G3 G2  G1 G0 G1 G0 */\
 836     "pshufw $0x00, %%mm5, %%mm6 \n\t" /* R1 R0 R1 R0  R1 R0 R1 R0 */\
 837 \
 838     "pand   %%mm0, %%mm1        \n\t" /*    B2        B1       B0 */\
 839     "pand   %%mm0, %%mm3        \n\t" /*    G2        G1       G0 */\
 840     "pand   %%mm7, %%mm6        \n\t" /*       R1        R0       */\
 841 \
 842     "psllq     $8, %%mm3        \n\t" /* G2        G1       G0    */\
 843     "por    %%mm1, %%mm6        \n\t"\
 844     "por    %%mm3, %%mm6        \n\t"\
 845     MOVNTQ(%%mm6, (dst))\
 846 \
 847     "psrlq     $8, %%mm4        \n\t" /* 00 G7 G6 G5  G4 G3 G2 G1 */\
 848     "pshufw $0xA5, %%mm2, %%mm1 \n\t" /* B5 B4 B5 B4  B3 B2 B3 B2 */\
 849     "pshufw $0x55, %%mm4, %%mm3 \n\t" /* G4 G3 G4 G3  G4 G3 G4 G3 */\
 850     "pshufw $0xA5, %%mm5, %%mm6 \n\t" /* R5 R4 R5 R4  R3 R2 R3 R2 */\
 851 \
 852     "pand "MANGLE(ff_M24B)", %%mm1 \n\t" /* B5       B4        B3    */\
 853     "pand   %%mm7, %%mm3        \n\t" /*       G4        G3       */\
 854     "pand   %%mm0, %%mm6        \n\t" /*    R4        R3       R2 */\
 855 \
 856     "por    %%mm1, %%mm3        \n\t" /* B5    G4 B4     G3 B3    */\
 857     "por    %%mm3, %%mm6        \n\t"\
 858     MOVNTQ(%%mm6, 8(dst))\
 859 \
 860     "pshufw $0xFF, %%mm2, %%mm1 \n\t" /* B7 B6 B7 B6  B7 B6 B6 B7 */\
 861     "pshufw $0xFA, %%mm4, %%mm3 \n\t" /* 00 G7 00 G7  G6 G5 G6 G5 */\
 862     "pshufw $0xFA, %%mm5, %%mm6 \n\t" /* R7 R6 R7 R6  R5 R4 R5 R4 */\
 863 \
 864     "pand   %%mm7, %%mm1        \n\t" /*       B7        B6       */\
 865     "pand   %%mm0, %%mm3        \n\t" /*    G7        G6       G5 */\
 866     "pand "MANGLE(ff_M24B)", %%mm6 \n\t" /* R7       R6        R5    */\
 867 \
 868     "por    %%mm1, %%mm3        \n\t"\
 869     "por    %%mm3, %%mm6        \n\t"\
 870     MOVNTQ(%%mm6, 16(dst))\
 871 \
 872     "add      $24, "#dst"       \n\t"\
 873 \
 874     "add       $8, "#index"     \n\t"\
 875     "cmp  "#dstw", "#index"     \n\t"\
 876     " jb       1b               \n\t"
 877
 878 #if HAVE_MMX2
 879 #undef WRITEBGR24
 880 #define WRITEBGR24(dst, dstw, index)  WRITEBGR24MMX2(dst, dstw, index)
 881 #else
 882 #undef WRITEBGR24
 883 #define WRITEBGR24(dst, dstw, index)  WRITEBGR24MMX(dst, dstw, index)
 884 #endif
 885
 886 #define REAL_WRITEYUY2(dst, dstw, index) \
 887     "packuswb  %%mm3, %%mm3     \n\t"\
 888     "packuswb  %%mm4, %%mm4     \n\t"\
 889     "packuswb  %%mm7, %%mm1     \n\t"\
 890     "punpcklbw %%mm4, %%mm3     \n\t"\
 891     "movq      %%mm1, %%mm7     \n\t"\
 892     "punpcklbw %%mm3, %%mm1     \n\t"\
 893     "punpckhbw %%mm3, %%mm7     \n\t"\
 894 \
 895     MOVNTQ(%%mm1, (dst, index, 2))\
 896     MOVNTQ(%%mm7, 8(dst, index, 2))\
 897 \
 898     "add          $8, "#index"  \n\t"\
 899     "cmp     "#dstw", "#index"  \n\t"\
 900     " jb          1b            \n\t"
 901 #define WRITEYUY2(dst, dstw, index)  REAL_WRITEYUY2(dst, dstw, index)
 902
 903
 904 static inline void RENAME(yuv2yuvX)(SwsContext *c, const int16_t *lumFilter, const int16_t **lumSrc, int lumFilterSize,
 905                                     const int16_t *chrFilter, const int16_t **chrSrc, int chrFilterSize, const int16_t **alpSrc,
 906                                     uint8_t *dest, uint8_t *uDest, uint8_t *vDest, uint8_t *aDest, long dstW, long chrDstW)
 907 {
 908 #if HAVE_MMX
 909     if(!(c->flags & SWS_BITEXACT)){
 910         if (c->flags & SWS_ACCURATE_RND){
 911             if (uDest){
 912                 YSCALEYUV2YV12X_ACCURATE(   "0", CHR_MMX_FILTER_OFFSET, uDest, chrDstW)
 913                 YSCALEYUV2YV12X_ACCURATE(AV_STRINGIFY(VOF), CHR_MMX_FILTER_OFFSET, vDest, chrDstW)
 914             }
 915             if (CONFIG_SWSCALE_ALPHA && aDest){
 916                 YSCALEYUV2YV12X_ACCURATE(   "0", ALP_MMX_FILTER_OFFSET, aDest, dstW)
 917             }
 918
 919             YSCALEYUV2YV12X_ACCURATE("0", LUM_MMX_FILTER_OFFSET, dest, dstW)
 920         }else{
 921             if (uDest){
 922                 YSCALEYUV2YV12X(   "0", CHR_MMX_FILTER_OFFSET, uDest, chrDstW)
 923                 YSCALEYUV2YV12X(AV_STRINGIFY(VOF), CHR_MMX_FILTER_OFFSET, vDest, chrDstW)
 924             }
 925             if (CONFIG_SWSCALE_ALPHA && aDest){
 926                 YSCALEYUV2YV12X(   "0", ALP_MMX_FILTER_OFFSET, aDest, dstW)
 927             }
 928
 929             YSCALEYUV2YV12X("0", LUM_MMX_FILTER_OFFSET, dest, dstW)
 930         }
 931         return;
 932     }
 933 #endif
 934 #if HAVE_ALTIVEC
 935 yuv2yuvX_altivec_real(lumFilter, lumSrc, lumFilterSize,
 936                       chrFilter, chrSrc, chrFilterSize,
 937                       dest, uDest, vDest, dstW, chrDstW);
 938 #else //HAVE_ALTIVEC
 939 yuv2yuvXinC(lumFilter, lumSrc, lumFilterSize,
 940             chrFilter, chrSrc, chrFilterSize,
 941             alpSrc, dest, uDest, vDest, aDest, dstW, chrDstW);
 942 #endif //!HAVE_ALTIVEC
 943 }
 944
 945 static inline void RENAME(yuv2nv12X)(SwsContext *c, const int16_t *lumFilter, const int16_t **lumSrc, int lumFilterSize,
 946                                      const int16_t *chrFilter, const int16_t **chrSrc, int chrFilterSize,
 947                                      uint8_t *dest, uint8_t *uDest, int dstW, int chrDstW, int dstFormat)
 948 {
 949 yuv2nv12XinC(lumFilter, lumSrc, lumFilterSize,
 950              chrFilter, chrSrc, chrFilterSize,
 951              dest, uDest, dstW, chrDstW, dstFormat);
 952 }
 953
 954 static inline void RENAME(yuv2yuv1)(SwsContext *c, const int16_t *lumSrc, const int16_t *chrSrc, const int16_t *alpSrc,
 955                                     uint8_t *dest, uint8_t *uDest, uint8_t *vDest, uint8_t *aDest, long dstW, long chrDstW)
 956 {
 957     int i;
 958 #if HAVE_MMX
 959     if(!(c->flags & SWS_BITEXACT)){
 960         long p= 4;
 961         uint8_t *src[4]= {alpSrc + dstW, lumSrc + dstW, chrSrc + chrDstW, chrSrc + VOFW + chrDstW};
 962         uint8_t *dst[4]= {aDest, dest, uDest, vDest};
 963         x86_reg counter[4]= {dstW, dstW, chrDstW, chrDstW};
 964
 965         if (c->flags & SWS_ACCURATE_RND){
 966             while(p--){
 967                 if (dst[p]){
 968                     __asm__ volatile(
 969                         YSCALEYUV2YV121_ACCURATE
 970                         :: "r" (src[p]), "r" (dst[p] + counter[p]),
 971                         "g" (-counter[p])
 972                         : "%"REG_a
 973                     );
 974                 }
 975             }
 976         }else{
 977             while(p--){
 978                 if (dst[p]){
 979                     __asm__ volatile(
 980                         YSCALEYUV2YV121
 981                         :: "r" (src[p]), "r" (dst[p] + counter[p]),
 982                         "g" (-counter[p])
 983                         : "%"REG_a
 984                     );
 985                 }
 986             }
 987         }
 988         return;
 989     }
 990 #endif
 991     for (i=0; i<dstW; i++)
 992     {
 993         int val= (lumSrc[i]+64)>>7;
 994
 995         if (val&256){
 996             if (val<0) val=0;
 997             else       val=255;
 998         }
 999
1000         dest[i]= val;
1001     }
1002
1003     if (uDest)
1004         for (i=0; i<chrDstW; i++)
1005         {
1006             int u=(chrSrc[i       ]+64)>>7;
1007             int v=(chrSrc[i + VOFW]+64)>>7;
1008
1009             if ((u|v)&256){
1010                 if (u<0)        u=0;
1011                 else if (u>255) u=255;
1012                 if (v<0)        v=0;
1013                 else if (v>255) v=255;
1014             }
1015
1016             uDest[i]= u;
1017             vDest[i]= v;
1018         }
1019
1020     if (CONFIG_SWSCALE_ALPHA && aDest)
1021         for (i=0; i<dstW; i++){
1022             int val= (alpSrc[i]+64)>>7;
1023             aDest[i]= av_clip_uint8(val);
1024         }
1025 }
1026
1027
1028 /**
1029  * vertical scale YV12 to RGB
1030  */
1031 static inline void RENAME(yuv2packedX)(SwsContext *c, const int16_t *lumFilter, const int16_t **lumSrc, int lumFilterSize,
1032                                        const int16_t *chrFilter, const int16_t **chrSrc, int chrFilterSize,
1033                                        const int16_t **alpSrc, uint8_t *dest, long dstW, long dstY)
1034 {
1035 #if HAVE_MMX
1036     x86_reg dummy=0;
1037     if(!(c->flags & SWS_BITEXACT)){
1038         if (c->flags & SWS_ACCURATE_RND){
1039             switch(c->dstFormat){
1040             case PIX_FMT_RGB32:
1041                 if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf){
1042                     YSCALEYUV2PACKEDX_ACCURATE
1043                     YSCALEYUV2RGBX
1044                     "movq                      %%mm2, "U_TEMP"(%0)  \n\t"
1045                     "movq                      %%mm4, "V_TEMP"(%0)  \n\t"
1046                     "movq                      %%mm5, "Y_TEMP"(%0)  \n\t"
1047                     YSCALEYUV2PACKEDX_ACCURATE_YA(ALP_MMX_FILTER_OFFSET)
1048                     "movq               "Y_TEMP"(%0), %%mm5         \n\t"
1049                     "psraw                        $3, %%mm1         \n\t"
1050                     "psraw                        $3, %%mm7         \n\t"
1051                     "packuswb                  %%mm7, %%mm1         \n\t"
1052                     WRITEBGR32(%4, %5, %%REGa, %%mm3, %%mm4, %%mm5, %%mm1, %%mm0, %%mm7, %%mm2, %%mm6)
1053
1054                     YSCALEYUV2PACKEDX_END
1055                 }else{
1056                     YSCALEYUV2PACKEDX_ACCURATE
1057                     YSCALEYUV2RGBX
1058                     "pcmpeqd %%mm7, %%mm7 \n\t"
1059                     WRITEBGR32(%4, %5, %%REGa, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
1060
1061                     YSCALEYUV2PACKEDX_END
1062                 }
1063                 return;
1064             case PIX_FMT_BGR24:
1065                 YSCALEYUV2PACKEDX_ACCURATE
1066                 YSCALEYUV2RGBX
1067                 "pxor %%mm7, %%mm7 \n\t"
1068                 "lea (%%"REG_a", %%"REG_a", 2), %%"REG_c"\n\t" //FIXME optimize
1069                 "add %4, %%"REG_c"                        \n\t"
1070                 WRITEBGR24(%%REGc, %5, %%REGa)
1071
1072
1073                 :: "r" (&c->redDither),
1074                 "m" (dummy), "m" (dummy), "m" (dummy),
1075                 "r" (dest), "m" (dstW)
1076                 : "%"REG_a, "%"REG_c, "%"REG_d, "%"REG_S
1077                 );
1078                 return;
1079             case PIX_FMT_RGB555:
1080                 YSCALEYUV2PACKEDX_ACCURATE
1081                 YSCALEYUV2RGBX
1082                 "pxor %%mm7, %%mm7 \n\t"
1083                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1084 #ifdef DITHER1XBPP
1085                 "paddusb "BLUE_DITHER"(%0), %%mm2\n\t"
1086                 "paddusb "GREEN_DITHER"(%0), %%mm4\n\t"
1087                 "paddusb "RED_DITHER"(%0), %%mm5\n\t"
1088 #endif
1089
1090                 WRITERGB15(%4, %5, %%REGa)
1091                 YSCALEYUV2PACKEDX_END
1092                 return;
1093             case PIX_FMT_RGB565:
1094                 YSCALEYUV2PACKEDX_ACCURATE
1095                 YSCALEYUV2RGBX
1096                 "pxor %%mm7, %%mm7 \n\t"
1097                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1098 #ifdef DITHER1XBPP
1099                 "paddusb "BLUE_DITHER"(%0), %%mm2\n\t"
1100                 "paddusb "GREEN_DITHER"(%0), %%mm4\n\t"
1101                 "paddusb "RED_DITHER"(%0), %%mm5\n\t"
1102 #endif
1103
1104                 WRITERGB16(%4, %5, %%REGa)
1105                 YSCALEYUV2PACKEDX_END
1106                 return;
1107             case PIX_FMT_YUYV422:
1108                 YSCALEYUV2PACKEDX_ACCURATE
1109                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1110
1111                 "psraw $3, %%mm3    \n\t"
1112                 "psraw $3, %%mm4    \n\t"
1113                 "psraw $3, %%mm1    \n\t"
1114                 "psraw $3, %%mm7    \n\t"
1115                 WRITEYUY2(%4, %5, %%REGa)
1116                 YSCALEYUV2PACKEDX_END
1117                 return;
1118             }
1119         }else{
1120             switch(c->dstFormat)
1121             {
1122             case PIX_FMT_RGB32:
1123                 if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf){
1124                     YSCALEYUV2PACKEDX
1125                     YSCALEYUV2RGBX
1126                     YSCALEYUV2PACKEDX_YA(ALP_MMX_FILTER_OFFSET, %%mm0, %%mm3, %%mm6, %%mm1, %%mm7)
1127                     "psraw                        $3, %%mm1         \n\t"
1128                     "psraw                        $3, %%mm7         \n\t"
1129                     "packuswb                  %%mm7, %%mm1         \n\t"
1130                     WRITEBGR32(%4, %5, %%REGa, %%mm2, %%mm4, %%mm5, %%mm1, %%mm0, %%mm7, %%mm3, %%mm6)
1131                     YSCALEYUV2PACKEDX_END
1132                 }else{
1133                     YSCALEYUV2PACKEDX
1134                     YSCALEYUV2RGBX
1135                     "pcmpeqd %%mm7, %%mm7 \n\t"
1136                     WRITEBGR32(%4, %5, %%REGa, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
1137                     YSCALEYUV2PACKEDX_END
1138                 }
1139                 return;
1140             case PIX_FMT_BGR24:
1141                 YSCALEYUV2PACKEDX
1142                 YSCALEYUV2RGBX
1143                 "pxor                    %%mm7, %%mm7       \n\t"
1144                 "lea (%%"REG_a", %%"REG_a", 2), %%"REG_c"   \n\t" //FIXME optimize
1145                 "add                        %4, %%"REG_c"   \n\t"
1146                 WRITEBGR24(%%REGc, %5, %%REGa)
1147
1148                 :: "r" (&c->redDither),
1149                 "m" (dummy), "m" (dummy), "m" (dummy),
1150                 "r" (dest),  "m" (dstW)
1151                 : "%"REG_a, "%"REG_c, "%"REG_d, "%"REG_S
1152                 );
1153                 return;
1154             case PIX_FMT_RGB555:
1155                 YSCALEYUV2PACKEDX
1156                 YSCALEYUV2RGBX
1157                 "pxor %%mm7, %%mm7 \n\t"
1158                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1159 #ifdef DITHER1XBPP
1160                 "paddusb "BLUE_DITHER"(%0), %%mm2  \n\t"
1161                 "paddusb "GREEN_DITHER"(%0), %%mm4  \n\t"
1162                 "paddusb "RED_DITHER"(%0), %%mm5  \n\t"
1163 #endif
1164
1165                 WRITERGB15(%4, %5, %%REGa)
1166                 YSCALEYUV2PACKEDX_END
1167                 return;
1168             case PIX_FMT_RGB565:
1169                 YSCALEYUV2PACKEDX
1170                 YSCALEYUV2RGBX
1171                 "pxor %%mm7, %%mm7 \n\t"
1172                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1173 #ifdef DITHER1XBPP
1174                 "paddusb "BLUE_DITHER"(%0), %%mm2  \n\t"
1175                 "paddusb "GREEN_DITHER"(%0), %%mm4  \n\t"
1176                 "paddusb "RED_DITHER"(%0), %%mm5  \n\t"
1177 #endif
1178
1179                 WRITERGB16(%4, %5, %%REGa)
1180                 YSCALEYUV2PACKEDX_END
1181                 return;
1182             case PIX_FMT_YUYV422:
1183                 YSCALEYUV2PACKEDX
1184                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1185
1186                 "psraw $3, %%mm3    \n\t"
1187                 "psraw $3, %%mm4    \n\t"
1188                 "psraw $3, %%mm1    \n\t"
1189                 "psraw $3, %%mm7    \n\t"
1190                 WRITEYUY2(%4, %5, %%REGa)
1191                 YSCALEYUV2PACKEDX_END
1192                 return;
1193             }
1194         }
1195     }
1196 #endif /* HAVE_MMX */
1197 #if HAVE_ALTIVEC
1198     /* The following list of supported dstFormat values should
1199        match what's found in the body of ff_yuv2packedX_altivec() */
1200     if (!(c->flags & SWS_BITEXACT) && !c->alpPixBuf &&
1201        (c->dstFormat==PIX_FMT_ABGR  || c->dstFormat==PIX_FMT_BGRA  ||
1202         c->dstFormat==PIX_FMT_BGR24 || c->dstFormat==PIX_FMT_RGB24 ||
1203         c->dstFormat==PIX_FMT_RGBA  || c->dstFormat==PIX_FMT_ARGB))
1204             ff_yuv2packedX_altivec(c, lumFilter, lumSrc, lumFilterSize,
1205                                    chrFilter, chrSrc, chrFilterSize,
1206                                    dest, dstW, dstY);
1207     else
1208 #endif
1209         yuv2packedXinC(c, lumFilter, lumSrc, lumFilterSize,
1210                        chrFilter, chrSrc, chrFilterSize,
1211                        alpSrc, dest, dstW, dstY);
1212 }
1213
1214 /**
1215  * vertical bilinear scale YV12 to RGB
1216  */
1217 static inline void RENAME(yuv2packed2)(SwsContext *c, const uint16_t *buf0, const uint16_t *buf1, const uint16_t *uvbuf0, const uint16_t *uvbuf1,
1218                           const uint16_t *abuf0, const uint16_t *abuf1, uint8_t *dest, int dstW, int yalpha, int uvalpha, int y)
1219 {
1220     int  yalpha1=4095- yalpha;
1221     int uvalpha1=4095-uvalpha;
1222     int i;
1223
1224 #if HAVE_MMX
1225     if(!(c->flags & SWS_BITEXACT)){
1226         switch(c->dstFormat)
1227         {
1228             //Note 8280 == DSTW_OFFSET but the preprocessor can't handle that there :(
1229             case PIX_FMT_RGB32:
1230                 if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf){
1231 #if ARCH_X86_64
1232                     __asm__ volatile(
1233                     YSCALEYUV2RGB(%%REGBP, %5)
1234                     YSCALEYUV2RGB_YA(%%REGBP, %5, %6, %7)
1235                     "psraw                  $3, %%mm1       \n\t" /* abuf0[eax] - abuf1[eax] >>7*/
1236                     "psraw                  $3, %%mm7       \n\t" /* abuf0[eax] - abuf1[eax] >>7*/
1237                     "packuswb            %%mm7, %%mm1       \n\t"
1238                     WRITEBGR32(%4, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm1, %%mm0, %%mm7, %%mm3, %%mm6)
1239
1240                     :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "r" (dest),
1241                     "a" (&c->redDither)
1242                     ,"r" (abuf0), "r" (abuf1)
1243                     : "%"REG_BP
1244                     );
1245 #else
1246                     *(uint16_t **)(&c->u_temp)=abuf0;
1247                     *(uint16_t **)(&c->v_temp)=abuf1;
1248                     __asm__ volatile(
1249                     "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1250                     "mov        %4, %%"REG_b"               \n\t"
1251                     "push %%"REG_BP"                        \n\t"
1252                     YSCALEYUV2RGB(%%REGBP, %5)
1253                     "push                   %0              \n\t"
1254                     "push                   %1              \n\t"
1255                     "mov          "U_TEMP"(%5), %0          \n\t"
1256                     "mov          "V_TEMP"(%5), %1          \n\t"
1257                     YSCALEYUV2RGB_YA(%%REGBP, %5, %0, %1)
1258                     "psraw                  $3, %%mm1       \n\t" /* abuf0[eax] - abuf1[eax] >>7*/
1259                     "psraw                  $3, %%mm7       \n\t" /* abuf0[eax] - abuf1[eax] >>7*/
1260                     "packuswb            %%mm7, %%mm1       \n\t"
1261                     "pop                    %1              \n\t"
1262                     "pop                    %0              \n\t"
1263                     WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm1, %%mm0, %%mm7, %%mm3, %%mm6)
1264                     "pop %%"REG_BP"                         \n\t"
1265                     "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1266
1267                     :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1268                     "a" (&c->redDither)
1269                     );
1270 #endif
1271                 }else{
1272                     __asm__ volatile(
1273                     "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1274                     "mov        %4, %%"REG_b"               \n\t"
1275                     "push %%"REG_BP"                        \n\t"
1276                     YSCALEYUV2RGB(%%REGBP, %5)
1277                     "pcmpeqd %%mm7, %%mm7                   \n\t"
1278                     WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
1279                     "pop %%"REG_BP"                         \n\t"
1280                     "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1281
1282                     :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1283                     "a" (&c->redDither)
1284                     );
1285                 }
1286                 return;
1287             case PIX_FMT_BGR24:
1288                 __asm__ volatile(
1289                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1290                 "mov        %4, %%"REG_b"               \n\t"
1291                 "push %%"REG_BP"                        \n\t"
1292                 YSCALEYUV2RGB(%%REGBP, %5)
1293                 "pxor    %%mm7, %%mm7                   \n\t"
1294                 WRITEBGR24(%%REGb, 8280(%5), %%REGBP)
1295                 "pop %%"REG_BP"                         \n\t"
1296                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1297                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1298                 "a" (&c->redDither)
1299                 );
1300                 return;
1301             case PIX_FMT_RGB555:
1302                 __asm__ volatile(
1303                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1304                 "mov        %4, %%"REG_b"               \n\t"
1305                 "push %%"REG_BP"                        \n\t"
1306                 YSCALEYUV2RGB(%%REGBP, %5)
1307                 "pxor    %%mm7, %%mm7                   \n\t"
1308                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1309 #ifdef DITHER1XBPP
1310                 "paddusb "BLUE_DITHER"(%5), %%mm2      \n\t"
1311                 "paddusb "GREEN_DITHER"(%5), %%mm4      \n\t"
1312                 "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
1313 #endif
1314
1315                 WRITERGB15(%%REGb, 8280(%5), %%REGBP)
1316                 "pop %%"REG_BP"                         \n\t"
1317                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1318
1319                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1320                 "a" (&c->redDither)
1321                 );
1322                 return;
1323             case PIX_FMT_RGB565:
1324                 __asm__ volatile(
1325                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1326                 "mov        %4, %%"REG_b"               \n\t"
1327                 "push %%"REG_BP"                        \n\t"
1328                 YSCALEYUV2RGB(%%REGBP, %5)
1329                 "pxor    %%mm7, %%mm7                   \n\t"
1330                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1331 #ifdef DITHER1XBPP
1332                 "paddusb "BLUE_DITHER"(%5), %%mm2      \n\t"
1333                 "paddusb "GREEN_DITHER"(%5), %%mm4      \n\t"
1334                 "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
1335 #endif
1336
1337                 WRITERGB16(%%REGb, 8280(%5), %%REGBP)
1338                 "pop %%"REG_BP"                         \n\t"
1339                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1340                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1341                 "a" (&c->redDither)
1342                 );
1343                 return;
1344             case PIX_FMT_YUYV422:
1345                 __asm__ volatile(
1346                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1347                 "mov %4, %%"REG_b"                        \n\t"
1348                 "push %%"REG_BP"                        \n\t"
1349                 YSCALEYUV2PACKED(%%REGBP, %5)
1350                 WRITEYUY2(%%REGb, 8280(%5), %%REGBP)
1351                 "pop %%"REG_BP"                         \n\t"
1352                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1353                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1354                 "a" (&c->redDither)
1355                 );
1356                 return;
1357             default: break;
1358         }
1359     }
1360 #endif //HAVE_MMX
1361 YSCALE_YUV_2_ANYRGB_C(YSCALE_YUV_2_RGB2_C, YSCALE_YUV_2_PACKED2_C(void,0), YSCALE_YUV_2_GRAY16_2_C, YSCALE_YUV_2_MONO2_C)
1362 }
1363
1364 /**
1365  * YV12 to RGB without scaling or interpolating
1366  */
1367 static inline void RENAME(yuv2packed1)(SwsContext *c, const uint16_t *buf0, const uint16_t *uvbuf0, const uint16_t *uvbuf1,
1368                           const uint16_t *abuf0, uint8_t *dest, int dstW, int uvalpha, int dstFormat, int flags, int y)
1369 {
1370     const int yalpha1=0;
1371     int i;
1372
1373     const uint16_t *buf1= buf0; //FIXME needed for RGB1/BGR1
1374     const int yalpha= 4096; //FIXME ...
1375
1376     if (flags&SWS_FULL_CHR_H_INT)
1377     {
1378         c->yuv2packed2(c, buf0, buf0, uvbuf0, uvbuf1, abuf0, abuf0, dest, dstW, 0, uvalpha, y);
1379         return;
1380     }
1381
1382 #if HAVE_MMX
1383     if(!(flags & SWS_BITEXACT)){
1384         if (uvalpha < 2048) // note this is not correct (shifts chrominance by 0.5 pixels) but it is a bit faster
1385         {
1386             switch(dstFormat)
1387             {
1388             case PIX_FMT_RGB32:
1389                 if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf){
1390                     __asm__ volatile(
1391                     "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1392                     "mov        %4, %%"REG_b"               \n\t"
1393                     "push %%"REG_BP"                        \n\t"
1394                     YSCALEYUV2RGB1(%%REGBP, %5)
1395                     YSCALEYUV2RGB1_ALPHA(%%REGBP)
1396                     WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
1397                     "pop %%"REG_BP"                         \n\t"
1398                     "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1399
1400                     :: "c" (buf0), "d" (abuf0), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1401                     "a" (&c->redDither)
1402                     );
1403                 }else{
1404                     __asm__ volatile(
1405                     "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1406                     "mov        %4, %%"REG_b"               \n\t"
1407                     "push %%"REG_BP"                        \n\t"
1408                     YSCALEYUV2RGB1(%%REGBP, %5)
1409                     "pcmpeqd %%mm7, %%mm7                   \n\t"
1410                     WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
1411                     "pop %%"REG_BP"                         \n\t"
1412                     "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1413
1414                     :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1415                     "a" (&c->redDither)
1416                     );
1417                 }
1418                 return;
1419             case PIX_FMT_BGR24:
1420                 __asm__ volatile(
1421                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1422                 "mov        %4, %%"REG_b"               \n\t"
1423                 "push %%"REG_BP"                        \n\t"
1424                 YSCALEYUV2RGB1(%%REGBP, %5)
1425                 "pxor    %%mm7, %%mm7                   \n\t"
1426                 WRITEBGR24(%%REGb, 8280(%5), %%REGBP)
1427                 "pop %%"REG_BP"                         \n\t"
1428                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1429
1430                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1431                 "a" (&c->redDither)
1432                 );
1433                 return;
1434             case PIX_FMT_RGB555:
1435                 __asm__ volatile(
1436                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1437                 "mov        %4, %%"REG_b"               \n\t"
1438                 "push %%"REG_BP"                        \n\t"
1439                 YSCALEYUV2RGB1(%%REGBP, %5)
1440                 "pxor    %%mm7, %%mm7                   \n\t"
1441                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1442 #ifdef DITHER1XBPP
1443                 "paddusb "BLUE_DITHER"(%5), %%mm2      \n\t"
1444                 "paddusb "GREEN_DITHER"(%5), %%mm4      \n\t"
1445                 "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
1446 #endif
1447                 WRITERGB15(%%REGb, 8280(%5), %%REGBP)
1448                 "pop %%"REG_BP"                         \n\t"
1449                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1450
1451                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1452                 "a" (&c->redDither)
1453                 );
1454                 return;
1455             case PIX_FMT_RGB565:
1456                 __asm__ volatile(
1457                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1458                 "mov        %4, %%"REG_b"               \n\t"
1459                 "push %%"REG_BP"                        \n\t"
1460                 YSCALEYUV2RGB1(%%REGBP, %5)
1461                 "pxor    %%mm7, %%mm7                   \n\t"
1462                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1463 #ifdef DITHER1XBPP
1464                 "paddusb "BLUE_DITHER"(%5), %%mm2      \n\t"
1465                 "paddusb "GREEN_DITHER"(%5), %%mm4      \n\t"
1466                 "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
1467 #endif
1468
1469                 WRITERGB16(%%REGb, 8280(%5), %%REGBP)
1470                 "pop %%"REG_BP"                         \n\t"
1471                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1472
1473                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1474                 "a" (&c->redDither)
1475                 );
1476                 return;
1477             case PIX_FMT_YUYV422:
1478                 __asm__ volatile(
1479                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1480                 "mov        %4, %%"REG_b"               \n\t"
1481                 "push %%"REG_BP"                        \n\t"
1482                 YSCALEYUV2PACKED1(%%REGBP, %5)
1483                 WRITEYUY2(%%REGb, 8280(%5), %%REGBP)
1484                 "pop %%"REG_BP"                         \n\t"
1485                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1486
1487                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1488                 "a" (&c->redDither)
1489                 );
1490                 return;
1491             }
1492         }
1493         else
1494         {
1495             switch(dstFormat)
1496             {
1497             case PIX_FMT_RGB32:
1498                 if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf){
1499                     __asm__ volatile(
1500                     "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1501                     "mov        %4, %%"REG_b"               \n\t"
1502                     "push %%"REG_BP"                        \n\t"
1503                     YSCALEYUV2RGB1b(%%REGBP, %5)
1504                     YSCALEYUV2RGB1_ALPHA(%%REGBP)
1505                     WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
1506                     "pop %%"REG_BP"                         \n\t"
1507                     "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1508
1509                     :: "c" (buf0), "d" (abuf0), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1510                     "a" (&c->redDither)
1511                     );
1512                 }else{
1513                     __asm__ volatile(
1514                     "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1515                     "mov        %4, %%"REG_b"               \n\t"
1516                     "push %%"REG_BP"                        \n\t"
1517                     YSCALEYUV2RGB1b(%%REGBP, %5)
1518                     "pcmpeqd %%mm7, %%mm7                   \n\t"
1519                     WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
1520                     "pop %%"REG_BP"                         \n\t"
1521                     "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1522
1523                     :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1524                     "a" (&c->redDither)
1525                     );
1526                 }
1527                 return;
1528             case PIX_FMT_BGR24:
1529                 __asm__ volatile(
1530                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1531                 "mov        %4, %%"REG_b"               \n\t"
1532                 "push %%"REG_BP"                        \n\t"
1533                 YSCALEYUV2RGB1b(%%REGBP, %5)
1534                 "pxor    %%mm7, %%mm7                   \n\t"
1535                 WRITEBGR24(%%REGb, 8280(%5), %%REGBP)
1536                 "pop %%"REG_BP"                         \n\t"
1537                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1538
1539                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1540                 "a" (&c->redDither)
1541                 );
1542                 return;
1543             case PIX_FMT_RGB555:
1544                 __asm__ volatile(
1545                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1546                 "mov        %4, %%"REG_b"               \n\t"
1547                 "push %%"REG_BP"                        \n\t"
1548                 YSCALEYUV2RGB1b(%%REGBP, %5)
1549                 "pxor    %%mm7, %%mm7                   \n\t"
1550                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1551 #ifdef DITHER1XBPP
1552                 "paddusb "BLUE_DITHER"(%5), %%mm2      \n\t"
1553                 "paddusb "GREEN_DITHER"(%5), %%mm4      \n\t"
1554                 "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
1555 #endif
1556                 WRITERGB15(%%REGb, 8280(%5), %%REGBP)
1557                 "pop %%"REG_BP"                         \n\t"
1558                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1559
1560                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1561                 "a" (&c->redDither)
1562                 );
1563                 return;
1564             case PIX_FMT_RGB565:
1565                 __asm__ volatile(
1566                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1567                 "mov        %4, %%"REG_b"               \n\t"
1568                 "push %%"REG_BP"                        \n\t"
1569                 YSCALEYUV2RGB1b(%%REGBP, %5)
1570                 "pxor    %%mm7, %%mm7                   \n\t"
1571                 /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
1572 #ifdef DITHER1XBPP
1573                 "paddusb "BLUE_DITHER"(%5), %%mm2      \n\t"
1574                 "paddusb "GREEN_DITHER"(%5), %%mm4      \n\t"
1575                 "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
1576 #endif
1577
1578                 WRITERGB16(%%REGb, 8280(%5), %%REGBP)
1579                 "pop %%"REG_BP"                         \n\t"
1580                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1581
1582                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1583                 "a" (&c->redDither)
1584                 );
1585                 return;
1586             case PIX_FMT_YUYV422:
1587                 __asm__ volatile(
1588                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
1589                 "mov        %4, %%"REG_b"               \n\t"
1590                 "push %%"REG_BP"                        \n\t"
1591                 YSCALEYUV2PACKED1b(%%REGBP, %5)
1592                 WRITEYUY2(%%REGb, 8280(%5), %%REGBP)
1593                 "pop %%"REG_BP"                         \n\t"
1594                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
1595
1596                 :: "c" (buf0), "d" (buf1), "S" (uvbuf0), "D" (uvbuf1), "m" (dest),
1597                 "a" (&c->redDither)
1598                 );
1599                 return;
1600             }
1601         }
1602     }
1603 #endif /* HAVE_MMX */
1604     if (uvalpha < 2048)
1605     {
1606         YSCALE_YUV_2_ANYRGB_C(YSCALE_YUV_2_RGB1_C, YSCALE_YUV_2_PACKED1_C(void,0), YSCALE_YUV_2_GRAY16_1_C, YSCALE_YUV_2_MONO2_C)
1607     }else{
1608         YSCALE_YUV_2_ANYRGB_C(YSCALE_YUV_2_RGB1B_C, YSCALE_YUV_2_PACKED1B_C(void,0), YSCALE_YUV_2_GRAY16_1_C, YSCALE_YUV_2_MONO2_C)
1609     }
1610 }
1611
1612 //FIXME yuy2* can read up to 7 samples too much
1613
1614 static inline void RENAME(yuy2ToY)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)
1615 {
1616 #if HAVE_MMX
1617     __asm__ volatile(
1618     "movq "MANGLE(bm01010101)", %%mm2           \n\t"
1619     "mov                    %0, %%"REG_a"       \n\t"
1620     "1:                                         \n\t"
1621     "movq    (%1, %%"REG_a",2), %%mm0           \n\t"
1622     "movq   8(%1, %%"REG_a",2), %%mm1           \n\t"
1623     "pand                %%mm2, %%mm0           \n\t"
1624     "pand                %%mm2, %%mm1           \n\t"
1625     "packuswb            %%mm1, %%mm0           \n\t"
1626     "movq                %%mm0, (%2, %%"REG_a") \n\t"
1627     "add                    $8, %%"REG_a"       \n\t"
1628     " js                    1b                  \n\t"
1629     : : "g" ((x86_reg)-width), "r" (src+width*2), "r" (dst+width)
1630     : "%"REG_a
1631     );
1632 #else
1633     int i;
1634     for (i=0; i<width; i++)
1635         dst[i]= src[2*i];
1636 #endif
1637 }
1638
1639 static inline void RENAME(yuy2ToUV)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1, const uint8_t *src2, long width, uint32_t *unused)
1640 {
1641 #if HAVE_MMX
1642     __asm__ volatile(
1643     "movq "MANGLE(bm01010101)", %%mm4           \n\t"
1644     "mov                    %0, %%"REG_a"       \n\t"
1645     "1:                                         \n\t"
1646     "movq    (%1, %%"REG_a",4), %%mm0           \n\t"
1647     "movq   8(%1, %%"REG_a",4), %%mm1           \n\t"
1648     "psrlw                  $8, %%mm0           \n\t"
1649     "psrlw                  $8, %%mm1           \n\t"
1650     "packuswb            %%mm1, %%mm0           \n\t"
1651     "movq                %%mm0, %%mm1           \n\t"
1652     "psrlw                  $8, %%mm0           \n\t"
1653     "pand                %%mm4, %%mm1           \n\t"
1654     "packuswb            %%mm0, %%mm0           \n\t"
1655     "packuswb            %%mm1, %%mm1           \n\t"
1656     "movd                %%mm0, (%3, %%"REG_a") \n\t"
1657     "movd                %%mm1, (%2, %%"REG_a") \n\t"
1658     "add                    $4, %%"REG_a"       \n\t"
1659     " js                    1b                  \n\t"
1660     : : "g" ((x86_reg)-width), "r" (src1+width*4), "r" (dstU+width), "r" (dstV+width)
1661     : "%"REG_a
1662     );
1663 #else
1664     int i;
1665     for (i=0; i<width; i++)
1666     {
1667         dstU[i]= src1[4*i + 1];
1668         dstV[i]= src1[4*i + 3];
1669     }
1670 #endif
1671     assert(src1 == src2);
1672 }
1673
1674 /* This is almost identical to the previous, end exists only because
1675  * yuy2ToY/UV)(dst, src+1, ...) would have 100% unaligned accesses. */
1676 static inline void RENAME(uyvyToY)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)
1677 {
1678 #if HAVE_MMX
1679     __asm__ volatile(
1680     "mov                  %0, %%"REG_a"         \n\t"
1681     "1:                                         \n\t"
1682     "movq  (%1, %%"REG_a",2), %%mm0             \n\t"
1683     "movq 8(%1, %%"REG_a",2), %%mm1             \n\t"
1684     "psrlw                $8, %%mm0             \n\t"
1685     "psrlw                $8, %%mm1             \n\t"
1686     "packuswb          %%mm1, %%mm0             \n\t"
1687     "movq              %%mm0, (%2, %%"REG_a")   \n\t"
1688     "add                  $8, %%"REG_a"         \n\t"
1689     " js                  1b                    \n\t"
1690     : : "g" ((x86_reg)-width), "r" (src+width*2), "r" (dst+width)
1691     : "%"REG_a
1692     );
1693 #else
1694     int i;
1695     for (i=0; i<width; i++)
1696         dst[i]= src[2*i+1];
1697 #endif
1698 }
1699
1700 static inline void RENAME(uyvyToUV)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1, const uint8_t *src2, long width, uint32_t *unused)
1701 {
1702 #if HAVE_MMX
1703     __asm__ volatile(
1704     "movq "MANGLE(bm01010101)", %%mm4           \n\t"
1705     "mov                    %0, %%"REG_a"       \n\t"
1706     "1:                                         \n\t"
1707     "movq    (%1, %%"REG_a",4), %%mm0           \n\t"
1708     "movq   8(%1, %%"REG_a",4), %%mm1           \n\t"
1709     "pand                %%mm4, %%mm0           \n\t"
1710     "pand                %%mm4, %%mm1           \n\t"
1711     "packuswb            %%mm1, %%mm0           \n\t"
1712     "movq                %%mm0, %%mm1           \n\t"
1713     "psrlw                  $8, %%mm0           \n\t"
1714     "pand                %%mm4, %%mm1           \n\t"
1715     "packuswb            %%mm0, %%mm0           \n\t"
1716     "packuswb            %%mm1, %%mm1           \n\t"
1717     "movd                %%mm0, (%3, %%"REG_a") \n\t"
1718     "movd                %%mm1, (%2, %%"REG_a") \n\t"
1719     "add                    $4, %%"REG_a"       \n\t"
1720     " js                    1b                  \n\t"
1721     : : "g" ((x86_reg)-width), "r" (src1+width*4), "r" (dstU+width), "r" (dstV+width)
1722     : "%"REG_a
1723     );
1724 #else
1725     int i;
1726     for (i=0; i<width; i++)
1727     {
1728         dstU[i]= src1[4*i + 0];
1729         dstV[i]= src1[4*i + 2];
1730     }
1731 #endif
1732     assert(src1 == src2);
1733 }
1734
1735 #define BGR2Y(type, name, shr, shg, shb, maskr, maskg, maskb, RY, GY, BY, S)\
1736 static inline void RENAME(name)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)\
1737 {\
1738     int i;\
1739     for (i=0; i<width; i++)\
1740     {\
1741         int b= (((const type*)src)[i]>>shb)&maskb;\
1742         int g= (((const type*)src)[i]>>shg)&maskg;\
1743         int r= (((const type*)src)[i]>>shr)&maskr;\
1744 \
1745         dst[i]= (((RY)*r + (GY)*g + (BY)*b + (33<<((S)-1)))>>(S));\
1746     }\
1747 }
1748
1749 BGR2Y(uint32_t, bgr32ToY,16, 0, 0, 0x00FF, 0xFF00, 0x00FF, RY<< 8, GY   , BY<< 8, RGB2YUV_SHIFT+8)
1750 BGR2Y(uint32_t, rgb32ToY, 0, 0,16, 0x00FF, 0xFF00, 0x00FF, RY<< 8, GY   , BY<< 8, RGB2YUV_SHIFT+8)
1751 BGR2Y(uint16_t, bgr16ToY, 0, 0, 0, 0x001F, 0x07E0, 0xF800, RY<<11, GY<<5, BY    , RGB2YUV_SHIFT+8)
1752 BGR2Y(uint16_t, bgr15ToY, 0, 0, 0, 0x001F, 0x03E0, 0x7C00, RY<<10, GY<<5, BY    , RGB2YUV_SHIFT+7)
1753 BGR2Y(uint16_t, rgb16ToY, 0, 0, 0, 0xF800, 0x07E0, 0x001F, RY    , GY<<5, BY<<11, RGB2YUV_SHIFT+8)
1754 BGR2Y(uint16_t, rgb15ToY, 0, 0, 0, 0x7C00, 0x03E0, 0x001F, RY    , GY<<5, BY<<10, RGB2YUV_SHIFT+7)
1755
1756 static inline void RENAME(abgrToA)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused){
1757     int i;
1758     for (i=0; i<width; i++){
1759         dst[i]= src[4*i];
1760     }
1761 }
1762
1763 #define BGR2UV(type, name, shr, shg, shb, maska, maskr, maskg, maskb, RU, GU, BU, RV, GV, BV, S)\
1764 static inline void RENAME(name)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src, const uint8_t *dummy, long width, uint32_t *unused)\
1765 {\
1766     int i;\
1767     for (i=0; i<width; i++)\
1768     {\
1769         int b= (((const type*)src)[i]&maskb)>>shb;\
1770         int g= (((const type*)src)[i]&maskg)>>shg;\
1771         int r= (((const type*)src)[i]&maskr)>>shr;\
1772 \
1773         dstU[i]= ((RU)*r + (GU)*g + (BU)*b + (257<<((S)-1)))>>(S);\
1774         dstV[i]= ((RV)*r + (GV)*g + (BV)*b + (257<<((S)-1)))>>(S);\
1775     }\
1776 }\
1777 static inline void RENAME(name ## _half)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src, const uint8_t *dummy, long width, uint32_t *unused)\
1778 {\
1779     int i;\
1780     for (i=0; i<width; i++)\
1781     {\
1782         int pix0= ((const type*)src)[2*i+0];\
1783         int pix1= ((const type*)src)[2*i+1];\
1784         int g= (pix0&~(maskr|maskb))+(pix1&~(maskr|maskb));\
1785         int b= ((pix0+pix1-g)&(maskb|(2*maskb)))>>shb;\
1786         int r= ((pix0+pix1-g)&(maskr|(2*maskr)))>>shr;\
1787         g&= maskg|(2*maskg);\
1788 \
1789         g>>=shg;\
1790 \
1791         dstU[i]= ((RU)*r + (GU)*g + (BU)*b + (257<<(S)))>>((S)+1);\
1792         dstV[i]= ((RV)*r + (GV)*g + (BV)*b + (257<<(S)))>>((S)+1);\
1793     }\
1794 }
1795
1796 BGR2UV(uint32_t, bgr32ToUV,16, 0, 0, 0xFF000000, 0xFF0000, 0xFF00,   0x00FF, RU<< 8, GU   , BU<< 8, RV<< 8, GV   , BV<< 8, RGB2YUV_SHIFT+8)
1797 BGR2UV(uint32_t, rgb32ToUV, 0, 0,16, 0xFF000000,   0x00FF, 0xFF00, 0xFF0000, RU<< 8, GU   , BU<< 8, RV<< 8, GV   , BV<< 8, RGB2YUV_SHIFT+8)
1798 BGR2UV(uint16_t, bgr16ToUV, 0, 0, 0,          0,   0x001F, 0x07E0,   0xF800, RU<<11, GU<<5, BU    , RV<<11, GV<<5, BV    , RGB2YUV_SHIFT+8)
1799 BGR2UV(uint16_t, bgr15ToUV, 0, 0, 0,          0,   0x001F, 0x03E0,   0x7C00, RU<<10, GU<<5, BU    , RV<<10, GV<<5, BV    , RGB2YUV_SHIFT+7)
1800 BGR2UV(uint16_t, rgb16ToUV, 0, 0, 0,          0,   0xF800, 0x07E0,   0x001F, RU    , GU<<5, BU<<11, RV    , GV<<5, BV<<11, RGB2YUV_SHIFT+8)
1801 BGR2UV(uint16_t, rgb15ToUV, 0, 0, 0,          0,   0x7C00, 0x03E0,   0x001F, RU    , GU<<5, BU<<10, RV    , GV<<5, BV<<10, RGB2YUV_SHIFT+7)
1802
1803 #if HAVE_MMX
1804 static inline void RENAME(bgr24ToY_mmx)(uint8_t *dst, const uint8_t *src, long width, int srcFormat)
1805 {
1806
1807     if(srcFormat == PIX_FMT_BGR24){
1808         __asm__ volatile(
1809             "movq  "MANGLE(ff_bgr24toY1Coeff)", %%mm5       \n\t"
1810             "movq  "MANGLE(ff_bgr24toY2Coeff)", %%mm6       \n\t"
1811             :
1812         );
1813     }else{
1814         __asm__ volatile(
1815             "movq  "MANGLE(ff_rgb24toY1Coeff)", %%mm5       \n\t"
1816             "movq  "MANGLE(ff_rgb24toY2Coeff)", %%mm6       \n\t"
1817             :
1818         );
1819     }
1820
1821     __asm__ volatile(
1822         "movq  "MANGLE(ff_bgr24toYOffset)", %%mm4   \n\t"
1823         "mov                        %2, %%"REG_a"   \n\t"
1824         "pxor                    %%mm7, %%mm7       \n\t"
1825         "1:                                         \n\t"
1826         PREFETCH"               64(%0)              \n\t"
1827         "movd                     (%0), %%mm0       \n\t"
1828         "movd                    2(%0), %%mm1       \n\t"
1829         "movd                    6(%0), %%mm2       \n\t"
1830         "movd                    8(%0), %%mm3       \n\t"
1831         "add                       $12, %0          \n\t"
1832         "punpcklbw               %%mm7, %%mm0       \n\t"
1833         "punpcklbw               %%mm7, %%mm1       \n\t"
1834         "punpcklbw               %%mm7, %%mm2       \n\t"
1835         "punpcklbw               %%mm7, %%mm3       \n\t"
1836         "pmaddwd                 %%mm5, %%mm0       \n\t"
1837         "pmaddwd                 %%mm6, %%mm1       \n\t"
1838         "pmaddwd                 %%mm5, %%mm2       \n\t"
1839         "pmaddwd                 %%mm6, %%mm3       \n\t"
1840         "paddd                   %%mm1, %%mm0       \n\t"
1841         "paddd                   %%mm3, %%mm2       \n\t"
1842         "paddd                   %%mm4, %%mm0       \n\t"
1843         "paddd                   %%mm4, %%mm2       \n\t"
1844         "psrad                     $15, %%mm0       \n\t"
1845         "psrad                     $15, %%mm2       \n\t"
1846         "packssdw                %%mm2, %%mm0       \n\t"
1847         "packuswb                %%mm0, %%mm0       \n\t"
1848         "movd                %%mm0, (%1, %%"REG_a") \n\t"
1849         "add                        $4, %%"REG_a"   \n\t"
1850         " js                        1b              \n\t"
1851     : "+r" (src)
1852     : "r" (dst+width), "g" ((x86_reg)-width)
1853     : "%"REG_a
1854     );
1855 }
1856
1857 static inline void RENAME(bgr24ToUV_mmx)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src, long width, int srcFormat)
1858 {
1859     __asm__ volatile(
1860         "movq                    24+%4, %%mm6       \n\t"
1861         "mov                        %3, %%"REG_a"   \n\t"
1862         "pxor                    %%mm7, %%mm7       \n\t"
1863         "1:                                         \n\t"
1864         PREFETCH"               64(%0)              \n\t"
1865         "movd                     (%0), %%mm0       \n\t"
1866         "movd                    2(%0), %%mm1       \n\t"
1867         "punpcklbw               %%mm7, %%mm0       \n\t"
1868         "punpcklbw               %%mm7, %%mm1       \n\t"
1869         "movq                    %%mm0, %%mm2       \n\t"
1870         "movq                    %%mm1, %%mm3       \n\t"
1871         "pmaddwd                    %4, %%mm0       \n\t"
1872         "pmaddwd                  8+%4, %%mm1       \n\t"
1873         "pmaddwd                 16+%4, %%mm2       \n\t"
1874         "pmaddwd                 %%mm6, %%mm3       \n\t"
1875         "paddd                   %%mm1, %%mm0       \n\t"
1876         "paddd                   %%mm3, %%mm2       \n\t"
1877
1878         "movd                    6(%0), %%mm1       \n\t"
1879         "movd                    8(%0), %%mm3       \n\t"
1880         "add                       $12, %0          \n\t"
1881         "punpcklbw               %%mm7, %%mm1       \n\t"
1882         "punpcklbw               %%mm7, %%mm3       \n\t"
1883         "movq                    %%mm1, %%mm4       \n\t"
1884         "movq                    %%mm3, %%mm5       \n\t"
1885         "pmaddwd                    %4, %%mm1       \n\t"
1886         "pmaddwd                  8+%4, %%mm3       \n\t"
1887         "pmaddwd                 16+%4, %%mm4       \n\t"
1888         "pmaddwd                 %%mm6, %%mm5       \n\t"
1889         "paddd                   %%mm3, %%mm1       \n\t"
1890         "paddd                   %%mm5, %%mm4       \n\t"
1891
1892         "movq "MANGLE(ff_bgr24toUVOffset)", %%mm3       \n\t"
1893         "paddd                   %%mm3, %%mm0       \n\t"
1894         "paddd                   %%mm3, %%mm2       \n\t"
1895         "paddd                   %%mm3, %%mm1       \n\t"
1896         "paddd                   %%mm3, %%mm4       \n\t"
1897         "psrad                     $15, %%mm0       \n\t"
1898         "psrad                     $15, %%mm2       \n\t"
1899         "psrad                     $15, %%mm1       \n\t"
1900         "psrad                     $15, %%mm4       \n\t"
1901         "packssdw                %%mm1, %%mm0       \n\t"
1902         "packssdw                %%mm4, %%mm2       \n\t"
1903         "packuswb                %%mm0, %%mm0       \n\t"
1904         "packuswb                %%mm2, %%mm2       \n\t"
1905         "movd                %%mm0, (%1, %%"REG_a") \n\t"
1906         "movd                %%mm2, (%2, %%"REG_a") \n\t"
1907         "add                        $4, %%"REG_a"   \n\t"
1908         " js                        1b              \n\t"
1909     : "+r" (src)
1910     : "r" (dstU+width), "r" (dstV+width), "g" ((x86_reg)-width), "m"(ff_bgr24toUV[srcFormat == PIX_FMT_RGB24][0])
1911     : "%"REG_a
1912     );
1913 }
1914 #endif
1915
1916 static inline void RENAME(bgr24ToY)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)
1917 {
1918 #if HAVE_MMX
1919     RENAME(bgr24ToY_mmx)(dst, src, width, PIX_FMT_BGR24);
1920 #else
1921     int i;
1922     for (i=0; i<width; i++)
1923     {
1924         int b= src[i*3+0];
1925         int g= src[i*3+1];
1926         int r= src[i*3+2];
1927
1928         dst[i]= ((RY*r + GY*g + BY*b + (33<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT);
1929     }
1930 #endif /* HAVE_MMX */
1931 }
1932
1933 static inline void RENAME(bgr24ToUV)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1, const uint8_t *src2, long width, uint32_t *unused)
1934 {
1935 #if HAVE_MMX
1936     RENAME(bgr24ToUV_mmx)(dstU, dstV, src1, width, PIX_FMT_BGR24);
1937 #else
1938     int i;
1939     for (i=0; i<width; i++)
1940     {
1941         int b= src1[3*i + 0];
1942         int g= src1[3*i + 1];
1943         int r= src1[3*i + 2];
1944
1945         dstU[i]= (RU*r + GU*g + BU*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
1946         dstV[i]= (RV*r + GV*g + BV*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
1947     }
1948 #endif /* HAVE_MMX */
1949     assert(src1 == src2);
1950 }
1951
1952 static inline void RENAME(bgr24ToUV_half)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1, const uint8_t *src2, long width, uint32_t *unused)
1953 {
1954     int i;
1955     for (i=0; i<width; i++)
1956     {
1957         int b= src1[6*i + 0] + src1[6*i + 3];
1958         int g= src1[6*i + 1] + src1[6*i + 4];
1959         int r= src1[6*i + 2] + src1[6*i + 5];
1960
1961         dstU[i]= (RU*r + GU*g + BU*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
1962         dstV[i]= (RV*r + GV*g + BV*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
1963     }
1964     assert(src1 == src2);
1965 }
1966
1967 static inline void RENAME(rgb24ToY)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)
1968 {
1969 #if HAVE_MMX
1970     RENAME(bgr24ToY_mmx)(dst, src, width, PIX_FMT_RGB24);
1971 #else
1972     int i;
1973     for (i=0; i<width; i++)
1974     {
1975         int r= src[i*3+0];
1976         int g= src[i*3+1];
1977         int b= src[i*3+2];
1978
1979         dst[i]= ((RY*r + GY*g + BY*b + (33<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT);
1980     }
1981 #endif
1982 }
1983
1984 static inline void RENAME(rgb24ToUV)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1, const uint8_t *src2, long width, uint32_t *unused)
1985 {
1986 #if HAVE_MMX
1987     assert(src1==src2);
1988     RENAME(bgr24ToUV_mmx)(dstU, dstV, src1, width, PIX_FMT_RGB24);
1989 #else
1990     int i;
1991     assert(src1==src2);
1992     for (i=0; i<width; i++)
1993     {
1994         int r= src1[3*i + 0];
1995         int g= src1[3*i + 1];
1996         int b= src1[3*i + 2];
1997
1998         dstU[i]= (RU*r + GU*g + BU*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
1999         dstV[i]= (RV*r + GV*g + BV*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
2000     }
2001 #endif
2002 }
2003
2004 static inline void RENAME(rgb24ToUV_half)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1, const uint8_t *src2, long width, uint32_t *unused)
2005 {
2006     int i;
2007     assert(src1==src2);
2008     for (i=0; i<width; i++)
2009     {
2010         int r= src1[6*i + 0] + src1[6*i + 3];
2011         int g= src1[6*i + 1] + src1[6*i + 4];
2012         int b= src1[6*i + 2] + src1[6*i + 5];
2013
2014         dstU[i]= (RU*r + GU*g + BU*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
2015         dstV[i]= (RV*r + GV*g + BV*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
2016     }
2017 }
2018
2019
2020 static inline void RENAME(palToY)(uint8_t *dst, const uint8_t *src, long width, uint32_t *pal)
2021 {
2022     int i;
2023     for (i=0; i<width; i++)
2024     {
2025         int d= src[i];
2026
2027         dst[i]= pal[d] & 0xFF;
2028     }
2029 }
2030
2031 static inline void RENAME(palToUV)(uint8_t *dstU, uint8_t *dstV,
2032                                    const uint8_t *src1, const uint8_t *src2,
2033                                    long width, uint32_t *pal)
2034 {
2035     int i;
2036     assert(src1 == src2);
2037     for (i=0; i<width; i++)
2038     {
2039         int p= pal[src1[i]];
2040
2041         dstU[i]= p>>8;
2042         dstV[i]= p>>16;
2043     }
2044 }
2045
2046 static inline void RENAME(monowhite2Y)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)
2047 {
2048     int i, j;
2049     for (i=0; i<width/8; i++){
2050         int d= ~src[i];
2051         for(j=0; j<8; j++)
2052             dst[8*i+j]= ((d>>(7-j))&1)*255;
2053     }
2054 }
2055
2056 static inline void RENAME(monoblack2Y)(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)
2057 {
2058     int i, j;
2059     for (i=0; i<width/8; i++){
2060         int d= src[i];
2061         for(j=0; j<8; j++)
2062             dst[8*i+j]= ((d>>(7-j))&1)*255;
2063     }
2064 }
2065
2066 // bilinear / bicubic scaling
2067 static inline void RENAME(hScale)(int16_t *dst, int dstW, const uint8_t *src, int srcW, int xInc,
2068                                   const int16_t *filter, const int16_t *filterPos, long filterSize)
2069 {
2070 #if HAVE_MMX
2071     assert(filterSize % 4 == 0 && filterSize>0);
2072     if (filterSize==4) // Always true for upscaling, sometimes for down, too.
2073     {
2074         x86_reg counter= -2*dstW;
2075         filter-= counter*2;
2076         filterPos-= counter/2;
2077         dst-= counter/2;
2078         __asm__ volatile(
2079 #if defined(PIC)
2080         "push            %%"REG_b"              \n\t"
2081 #endif
2082         "pxor                %%mm7, %%mm7       \n\t"
2083         "push           %%"REG_BP"              \n\t" // we use 7 regs here ...
2084         "mov             %%"REG_a", %%"REG_BP"  \n\t"
2085         ASMALIGN(4)
2086         "1:                                     \n\t"
2087         "movzwl   (%2, %%"REG_BP"), %%eax       \n\t"
2088         "movzwl  2(%2, %%"REG_BP"), %%ebx       \n\t"
2089         "movq  (%1, %%"REG_BP", 4), %%mm1       \n\t"
2090         "movq 8(%1, %%"REG_BP", 4), %%mm3       \n\t"
2091         "movd      (%3, %%"REG_a"), %%mm0       \n\t"
2092         "movd      (%3, %%"REG_b"), %%mm2       \n\t"
2093         "punpcklbw           %%mm7, %%mm0       \n\t"
2094         "punpcklbw           %%mm7, %%mm2       \n\t"
2095         "pmaddwd             %%mm1, %%mm0       \n\t"
2096         "pmaddwd             %%mm2, %%mm3       \n\t"
2097         "movq                %%mm0, %%mm4       \n\t"
2098         "punpckldq           %%mm3, %%mm0       \n\t"
2099         "punpckhdq           %%mm3, %%mm4       \n\t"
2100         "paddd               %%mm4, %%mm0       \n\t"
2101         "psrad                  $7, %%mm0       \n\t"
2102         "packssdw            %%mm0, %%mm0       \n\t"
2103         "movd                %%mm0, (%4, %%"REG_BP")    \n\t"
2104         "add                    $4, %%"REG_BP"  \n\t"
2105         " jnc                   1b              \n\t"
2106
2107         "pop            %%"REG_BP"              \n\t"
2108 #if defined(PIC)
2109         "pop             %%"REG_b"              \n\t"
2110 #endif
2111         : "+a" (counter)
2112         : "c" (filter), "d" (filterPos), "S" (src), "D" (dst)
2113 #if !defined(PIC)
2114         : "%"REG_b
2115 #endif
2116         );
2117     }
2118     else if (filterSize==8)
2119     {
2120         x86_reg counter= -2*dstW;
2121         filter-= counter*4;
2122         filterPos-= counter/2;
2123         dst-= counter/2;
2124         __asm__ volatile(
2125 #if defined(PIC)
2126         "push             %%"REG_b"             \n\t"
2127 #endif
2128         "pxor                 %%mm7, %%mm7      \n\t"
2129         "push            %%"REG_BP"             \n\t" // we use 7 regs here ...
2130         "mov              %%"REG_a", %%"REG_BP" \n\t"
2131         ASMALIGN(4)
2132         "1:                                     \n\t"
2133         "movzwl    (%2, %%"REG_BP"), %%eax      \n\t"
2134         "movzwl   2(%2, %%"REG_BP"), %%ebx      \n\t"
2135         "movq   (%1, %%"REG_BP", 8), %%mm1      \n\t"
2136         "movq 16(%1, %%"REG_BP", 8), %%mm3      \n\t"
2137         "movd       (%3, %%"REG_a"), %%mm0      \n\t"
2138         "movd       (%3, %%"REG_b"), %%mm2      \n\t"
2139         "punpcklbw            %%mm7, %%mm0      \n\t"
2140         "punpcklbw            %%mm7, %%mm2      \n\t"
2141         "pmaddwd              %%mm1, %%mm0      \n\t"
2142         "pmaddwd              %%mm2, %%mm3      \n\t"
2143
2144         "movq  8(%1, %%"REG_BP", 8), %%mm1      \n\t"
2145         "movq 24(%1, %%"REG_BP", 8), %%mm5      \n\t"
2146         "movd      4(%3, %%"REG_a"), %%mm4      \n\t"
2147         "movd      4(%3, %%"REG_b"), %%mm2      \n\t"
2148         "punpcklbw            %%mm7, %%mm4      \n\t"
2149         "punpcklbw            %%mm7, %%mm2      \n\t"
2150         "pmaddwd              %%mm1, %%mm4      \n\t"
2151         "pmaddwd              %%mm2, %%mm5      \n\t"
2152         "paddd                %%mm4, %%mm0      \n\t"
2153         "paddd                %%mm5, %%mm3      \n\t"
2154         "movq                 %%mm0, %%mm4      \n\t"
2155         "punpckldq            %%mm3, %%mm0      \n\t"
2156         "punpckhdq            %%mm3, %%mm4      \n\t"
2157         "paddd                %%mm4, %%mm0      \n\t"
2158         "psrad                   $7, %%mm0      \n\t"
2159         "packssdw             %%mm0, %%mm0      \n\t"
2160         "movd                 %%mm0, (%4, %%"REG_BP")   \n\t"
2161         "add                     $4, %%"REG_BP" \n\t"
2162         " jnc                    1b             \n\t"
2163
2164         "pop             %%"REG_BP"             \n\t"
2165 #if defined(PIC)
2166         "pop              %%"REG_b"             \n\t"
2167 #endif
2168         : "+a" (counter)
2169         : "c" (filter), "d" (filterPos), "S" (src), "D" (dst)
2170 #if !defined(PIC)
2171         : "%"REG_b
2172 #endif
2173         );
2174     }
2175     else
2176     {
2177         uint8_t *offset = src+filterSize;
2178         x86_reg counter= -2*dstW;
2179         //filter-= counter*filterSize/2;
2180         filterPos-= counter/2;
2181         dst-= counter/2;
2182         __asm__ volatile(
2183         "pxor                  %%mm7, %%mm7     \n\t"
2184         ASMALIGN(4)
2185         "1:                                     \n\t"
2186         "mov                      %2, %%"REG_c" \n\t"
2187         "movzwl      (%%"REG_c", %0), %%eax     \n\t"
2188         "movzwl     2(%%"REG_c", %0), %%edx     \n\t"
2189         "mov                      %5, %%"REG_c" \n\t"
2190         "pxor                  %%mm4, %%mm4     \n\t"
2191         "pxor                  %%mm5, %%mm5     \n\t"
2192         "2:                                     \n\t"
2193         "movq                   (%1), %%mm1     \n\t"
2194         "movq               (%1, %6), %%mm3     \n\t"
2195         "movd (%%"REG_c", %%"REG_a"), %%mm0     \n\t"
2196         "movd (%%"REG_c", %%"REG_d"), %%mm2     \n\t"
2197         "punpcklbw             %%mm7, %%mm0     \n\t"
2198         "punpcklbw             %%mm7, %%mm2     \n\t"
2199         "pmaddwd               %%mm1, %%mm0     \n\t"
2200         "pmaddwd               %%mm2, %%mm3     \n\t"
2201         "paddd                 %%mm3, %%mm5     \n\t"
2202         "paddd                 %%mm0, %%mm4     \n\t"
2203         "add                      $8, %1        \n\t"
2204         "add                      $4, %%"REG_c" \n\t"
2205         "cmp                      %4, %%"REG_c" \n\t"
2206         " jb                      2b            \n\t"
2207         "add                      %6, %1        \n\t"
2208         "movq                  %%mm4, %%mm0     \n\t"
2209         "punpckldq             %%mm5, %%mm4     \n\t"
2210         "punpckhdq             %%mm5, %%mm0     \n\t"
2211         "paddd                 %%mm0, %%mm4     \n\t"
2212         "psrad                    $7, %%mm4     \n\t"
2213         "packssdw              %%mm4, %%mm4     \n\t"
2214         "mov                      %3, %%"REG_a" \n\t"
2215         "movd                  %%mm4, (%%"REG_a", %0)   \n\t"
2216         "add                      $4, %0        \n\t"
2217         " jnc                     1b            \n\t"
2218
2219         : "+r" (counter), "+r" (filter)
2220         : "m" (filterPos), "m" (dst), "m"(offset),
2221           "m" (src), "r" ((x86_reg)filterSize*2)
2222         : "%"REG_a, "%"REG_c, "%"REG_d
2223         );
2224     }
2225 #else
2226 #if HAVE_ALTIVEC
2227     hScale_altivec_real(dst, dstW, src, srcW, xInc, filter, filterPos, filterSize);
2228 #else
2229     int i;
2230     for (i=0; i<dstW; i++)
2231     {
2232         int j;
2233         int srcPos= filterPos[i];
2234         int val=0;
2235         //printf("filterPos: %d\n", filterPos[i]);
2236         for (j=0; j<filterSize; j++)
2237         {
2238             //printf("filter: %d, src: %d\n", filter[i], src[srcPos + j]);
2239             val += ((int)src[srcPos + j])*filter[filterSize*i + j];
2240         }
2241         //filter += hFilterSize;
2242         dst[i] = FFMIN(val>>7, (1<<15)-1); // the cubic equation does overflow ...
2243         //dst[i] = val>>7;
2244     }
2245 #endif /* HAVE_ALTIVEC */
2246 #endif /* HAVE_MMX */
2247 }
2248
2249 static inline void RENAME(hyscale_fast)(SwsContext *c, int16_t *dst,
2250                                         int dstWidth, const uint8_t *src, int srcW,
2251                                         int xInc)
2252 {
2253     int i;
2254     unsigned int xpos=0;
2255     for (i=0;i<dstWidth;i++)
2256     {
2257         register unsigned int xx=xpos>>16;
2258         register unsigned int xalpha=(xpos&0xFFFF)>>9;
2259         dst[i]= (src[xx]<<7) + (src[xx+1] - src[xx])*xalpha;
2260         xpos+=xInc;
2261     }
2262 }
2263
2264       // *** horizontal scale Y line to temp buffer
2265 static inline void RENAME(hyscale)(SwsContext *c, uint16_t *dst, long dstWidth, const uint8_t *src, int srcW, int xInc,
2266                                    int flags, const int16_t *hLumFilter,
2267                                    const int16_t *hLumFilterPos, int hLumFilterSize,
2268                                    int srcFormat, uint8_t *formatConvBuffer,
2269                                    uint32_t *pal, int isAlpha)
2270 {
2271     int32_t *mmx2FilterPos = c->lumMmx2FilterPos;
2272     int16_t *mmx2Filter = c->lumMmx2Filter;
2273     int canMMX2BeUsed = c->canMMX2BeUsed;
2274     void *funnyYCode = c->funnyYCode;
2275
2276     if (isAlpha) {
2277         if (srcFormat == PIX_FMT_RGB32   || srcFormat == PIX_FMT_BGR32  )
2278             src += 3;
2279     } else {
2280         if (srcFormat == PIX_FMT_RGB32_1 || srcFormat == PIX_FMT_BGR32_1)
2281             src += ALT32_CORR;
2282     }
2283
2284     if (c->hyscale_internal) {
2285         c->hyscale_internal(formatConvBuffer, src, srcW, pal);
2286         src= formatConvBuffer;
2287     }
2288
2289 #if HAVE_MMX
2290     // Use the new MMX scaler if the MMX2 one can't be used (it is faster than the x86 ASM one).
2291     if (!(flags&SWS_FAST_BILINEAR) || (!canMMX2BeUsed))
2292 #else
2293     if (!(flags&SWS_FAST_BILINEAR))
2294 #endif
2295     {
2296         c->hScale(dst, dstWidth, src, srcW, xInc, hLumFilter, hLumFilterPos, hLumFilterSize);
2297     }
2298     else // fast bilinear upscale / crap downscale
2299     {
2300 #if ARCH_X86 && CONFIG_GPL
2301 #if HAVE_MMX2
2302         int i;
2303 #if defined(PIC)
2304         uint64_t ebxsave __attribute__((aligned(8)));
2305 #endif
2306         if (canMMX2BeUsed)
2307         {
2308             __asm__ volatile(
2309 #if defined(PIC)
2310             "mov               %%"REG_b", %5        \n\t"
2311 #endif
2312             "pxor                  %%mm7, %%mm7     \n\t"
2313             "mov                      %0, %%"REG_c" \n\t"
2314             "mov                      %1, %%"REG_D" \n\t"
2315             "mov                      %2, %%"REG_d" \n\t"
2316             "mov                      %3, %%"REG_b" \n\t"
2317             "xor               %%"REG_a", %%"REG_a" \n\t" // i
2318             PREFETCH"        (%%"REG_c")            \n\t"
2319             PREFETCH"      32(%%"REG_c")            \n\t"
2320             PREFETCH"      64(%%"REG_c")            \n\t"
2321
2322 #if ARCH_X86_64
2323
2324 #define FUNNY_Y_CODE \
2325             "movl            (%%"REG_b"), %%esi     \n\t"\
2326             "call                    *%4            \n\t"\
2327             "movl (%%"REG_b", %%"REG_a"), %%esi     \n\t"\
2328             "add               %%"REG_S", %%"REG_c" \n\t"\
2329             "add               %%"REG_a", %%"REG_D" \n\t"\
2330             "xor               %%"REG_a", %%"REG_a" \n\t"\
2331
2332 #else
2333
2334 #define FUNNY_Y_CODE \
2335             "movl (%%"REG_b"), %%esi        \n\t"\
2336             "call         *%4                       \n\t"\
2337             "addl (%%"REG_b", %%"REG_a"), %%"REG_c" \n\t"\
2338             "add               %%"REG_a", %%"REG_D" \n\t"\
2339             "xor               %%"REG_a", %%"REG_a" \n\t"\
2340
2341 #endif /* ARCH_X86_64 */
2342
2343 FUNNY_Y_CODE
2344 FUNNY_Y_CODE
2345 FUNNY_Y_CODE
2346 FUNNY_Y_CODE
2347 FUNNY_Y_CODE
2348 FUNNY_Y_CODE
2349 FUNNY_Y_CODE
2350 FUNNY_Y_CODE
2351
2352 #if defined(PIC)
2353             "mov                      %5, %%"REG_b" \n\t"
2354 #endif
2355             :: "m" (src), "m" (dst), "m" (mmx2Filter), "m" (mmx2FilterPos),
2356             "m" (funnyYCode)
2357 #if defined(PIC)
2358             ,"m" (ebxsave)
2359 #endif
2360             : "%"REG_a, "%"REG_c, "%"REG_d, "%"REG_S, "%"REG_D
2361 #if !defined(PIC)
2362             ,"%"REG_b
2363 #endif
2364             );
2365             for (i=dstWidth-1; (i*xInc)>>16 >=srcW-1; i--) dst[i] = src[srcW-1]*128;
2366         }
2367         else
2368         {
2369 #endif /* HAVE_MMX2 */
2370         x86_reg xInc_shr16 = xInc >> 16;
2371         uint16_t xInc_mask = xInc & 0xffff;
2372         //NO MMX just normal asm ...
2373         __asm__ volatile(
2374         "xor %%"REG_a", %%"REG_a"            \n\t" // i
2375         "xor %%"REG_d", %%"REG_d"            \n\t" // xx
2376         "xorl    %%ecx, %%ecx                \n\t" // 2*xalpha
2377         ASMALIGN(4)
2378         "1:                                  \n\t"
2379         "movzbl    (%0, %%"REG_d"), %%edi    \n\t" //src[xx]
2380         "movzbl   1(%0, %%"REG_d"), %%esi    \n\t" //src[xx+1]
2381         "subl    %%edi, %%esi                \n\t" //src[xx+1] - src[xx]
2382         "imull   %%ecx, %%esi                \n\t" //(src[xx+1] - src[xx])*2*xalpha
2383         "shll      $16, %%edi                \n\t"
2384         "addl    %%edi, %%esi                \n\t" //src[xx+1]*2*xalpha + src[xx]*(1-2*xalpha)
2385         "mov        %1, %%"REG_D"            \n\t"
2386         "shrl       $9, %%esi                \n\t"
2387         "movw     %%si, (%%"REG_D", %%"REG_a", 2)   \n\t"
2388         "addw       %4, %%cx                 \n\t" //2*xalpha += xInc&0xFF
2389         "adc        %3, %%"REG_d"            \n\t" //xx+= xInc>>8 + carry
2390
2391         "movzbl    (%0, %%"REG_d"), %%edi    \n\t" //src[xx]
2392         "movzbl   1(%0, %%"REG_d"), %%esi    \n\t" //src[xx+1]
2393         "subl    %%edi, %%esi                \n\t" //src[xx+1] - src[xx]
2394         "imull   %%ecx, %%esi                \n\t" //(src[xx+1] - src[xx])*2*xalpha
2395         "shll      $16, %%edi                \n\t"
2396         "addl    %%edi, %%esi                \n\t" //src[xx+1]*2*xalpha + src[xx]*(1-2*xalpha)
2397         "mov        %1, %%"REG_D"            \n\t"
2398         "shrl       $9, %%esi                \n\t"
2399         "movw     %%si, 2(%%"REG_D", %%"REG_a", 2)  \n\t"
2400         "addw       %4, %%cx                 \n\t" //2*xalpha += xInc&0xFF
2401         "adc        %3, %%"REG_d"            \n\t" //xx+= xInc>>8 + carry
2402
2403
2404         "add        $2, %%"REG_a"            \n\t"
2405         "cmp        %2, %%"REG_a"            \n\t"
2406         " jb        1b                       \n\t"
2407
2408
2409         :: "r" (src), "m" (dst), "m" (dstWidth), "m" (xInc_shr16), "m" (xInc_mask)
2410         : "%"REG_a, "%"REG_d, "%ecx", "%"REG_D, "%esi"
2411         );
2412 #if HAVE_MMX2
2413         } //if MMX2 can't be used
2414 #endif
2415 #else
2416         c->hyscale_fast(c, dst, dstWidth, src, srcW, xInc);
2417 #endif /* ARCH_X86 */
2418     }
2419
2420     if(!isAlpha && c->srcRange != c->dstRange && !(isRGB(c->dstFormat) || isBGR(c->dstFormat))){
2421         int i;
2422         //FIXME all pal and rgb srcFormats could do this convertion as well
2423         //FIXME all scalers more complex than bilinear could do half of this transform
2424         if(c->srcRange){
2425             for (i=0; i<dstWidth; i++)
2426                 dst[i]= (dst[i]*14071 + 33561947)>>14;
2427         }else{
2428             for (i=0; i<dstWidth; i++)
2429                 dst[i]= (FFMIN(dst[i],30189)*19077 - 39057361)>>14;
2430         }
2431     }
2432 }
2433
2434 static inline void RENAME(hcscale_fast)(SwsContext *c, int16_t *dst,
2435                                         int dstWidth, const uint8_t *src1,
2436                                         const uint8_t *src2, int srcW, int xInc)
2437 {
2438     int i;
2439     unsigned int xpos=0;
2440     for (i=0;i<dstWidth;i++)
2441     {
2442         register unsigned int xx=xpos>>16;
2443         register unsigned int xalpha=(xpos&0xFFFF)>>9;
2444         dst[i]=(src1[xx]*(xalpha^127)+src1[xx+1]*xalpha);
2445         dst[i+VOFW]=(src2[xx]*(xalpha^127)+src2[xx+1]*xalpha);
2446         /* slower
2447         dst[i]= (src1[xx]<<7) + (src1[xx+1] - src1[xx])*xalpha;
2448         dst[i+VOFW]=(src2[xx]<<7) + (src2[xx+1] - src2[xx])*xalpha;
2449         */
2450         xpos+=xInc;
2451     }
2452 }
2453
2454 inline static void RENAME(hcscale)(SwsContext *c, uint16_t *dst, long dstWidth, const uint8_t *src1, const uint8_t *src2,
2455                                    int srcW, int xInc, int flags, const int16_t *hChrFilter,
2456                                    const int16_t *hChrFilterPos, int hChrFilterSize,
2457                                    int srcFormat, uint8_t *formatConvBuffer,
2458                                    uint32_t *pal)
2459 {
2460     int32_t *mmx2FilterPos = c->chrMmx2FilterPos;
2461     int16_t *mmx2Filter = c->chrMmx2Filter;
2462     int canMMX2BeUsed = c->canMMX2BeUsed;
2463     void *funnyUVCode = c->funnyUVCode;
2464
2465     if (isGray(srcFormat) || srcFormat==PIX_FMT_MONOBLACK || srcFormat==PIX_FMT_MONOWHITE)
2466         return;
2467
2468     if (srcFormat==PIX_FMT_RGB32_1) {
2469         src1 += ALT32_CORR;
2470         src2 += ALT32_CORR;
2471     }
2472
2473     if (c->hcscale_internal) {
2474         c->hcscale_internal(formatConvBuffer, formatConvBuffer+VOFW, src1, src2, srcW, pal);
2475         src1= formatConvBuffer;
2476         src2= formatConvBuffer+VOFW;
2477     }
2478
2479 #if HAVE_MMX
2480     // Use the new MMX scaler if the MMX2 one can't be used (it is faster than the x86 ASM one).
2481     if (!(flags&SWS_FAST_BILINEAR) || (!canMMX2BeUsed))
2482 #else
2483     if (!(flags&SWS_FAST_BILINEAR))
2484 #endif
2485     {
2486         c->hScale(dst     , dstWidth, src1, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize);
2487         c->hScale(dst+VOFW, dstWidth, src2, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize);
2488     }
2489     else // fast bilinear upscale / crap downscale
2490     {
2491 #if ARCH_X86 && CONFIG_GPL
2492 #if HAVE_MMX2
2493         int i;
2494 #if defined(PIC)
2495         uint64_t ebxsave __attribute__((aligned(8)));
2496 #endif
2497         if (canMMX2BeUsed)
2498         {
2499             __asm__ volatile(
2500 #if defined(PIC)
2501             "mov          %%"REG_b", %6         \n\t"
2502 #endif
2503             "pxor             %%mm7, %%mm7      \n\t"
2504             "mov                 %0, %%"REG_c"  \n\t"
2505             "mov                 %1, %%"REG_D"  \n\t"
2506             "mov                 %2, %%"REG_d"  \n\t"
2507             "mov                 %3, %%"REG_b"  \n\t"
2508             "xor          %%"REG_a", %%"REG_a"  \n\t" // i
2509             PREFETCH"   (%%"REG_c")             \n\t"
2510             PREFETCH" 32(%%"REG_c")             \n\t"
2511             PREFETCH" 64(%%"REG_c")             \n\t"
2512
2513 #if ARCH_X86_64
2514
2515 #define FUNNY_UV_CODE \
2516             "movl       (%%"REG_b"), %%esi      \n\t"\
2517             "call               *%4             \n\t"\
2518             "movl (%%"REG_b", %%"REG_a"), %%esi \n\t"\
2519             "add          %%"REG_S", %%"REG_c"  \n\t"\
2520             "add          %%"REG_a", %%"REG_D"  \n\t"\
2521             "xor          %%"REG_a", %%"REG_a"  \n\t"\
2522
2523 #else
2524
2525 #define FUNNY_UV_CODE \
2526             "movl       (%%"REG_b"), %%esi      \n\t"\
2527             "call               *%4             \n\t"\
2528             "addl (%%"REG_b", %%"REG_a"), %%"REG_c" \n\t"\
2529             "add          %%"REG_a", %%"REG_D"  \n\t"\
2530             "xor          %%"REG_a", %%"REG_a"  \n\t"\
2531
2532 #endif /* ARCH_X86_64 */
2533
2534 FUNNY_UV_CODE
2535 FUNNY_UV_CODE
2536 FUNNY_UV_CODE
2537 FUNNY_UV_CODE
2538             "xor          %%"REG_a", %%"REG_a"  \n\t" // i
2539             "mov                 %5, %%"REG_c"  \n\t" // src
2540             "mov                 %1, %%"REG_D"  \n\t" // buf1
2541             "add              $"AV_STRINGIFY(VOF)", %%"REG_D"  \n\t"
2542             PREFETCH"   (%%"REG_c")             \n\t"
2543             PREFETCH" 32(%%"REG_c")             \n\t"
2544             PREFETCH" 64(%%"REG_c")             \n\t"
2545
2546 FUNNY_UV_CODE
2547 FUNNY_UV_CODE
2548 FUNNY_UV_CODE
2549 FUNNY_UV_CODE
2550
2551 #if defined(PIC)
2552             "mov %6, %%"REG_b"    \n\t"
2553 #endif
2554             :: "m" (src1), "m" (dst), "m" (mmx2Filter), "m" (mmx2FilterPos),
2555             "m" (funnyUVCode), "m" (src2)
2556 #if defined(PIC)
2557             ,"m" (ebxsave)
2558 #endif
2559             : "%"REG_a, "%"REG_c, "%"REG_d, "%"REG_S, "%"REG_D
2560 #if !defined(PIC)
2561              ,"%"REG_b
2562 #endif
2563             );
2564             for (i=dstWidth-1; (i*xInc)>>16 >=srcW-1; i--)
2565             {
2566                 //printf("%d %d %d\n", dstWidth, i, srcW);
2567                 dst[i] = src1[srcW-1]*128;
2568                 dst[i+VOFW] = src2[srcW-1]*128;
2569             }
2570         }
2571         else
2572         {
2573 #endif /* HAVE_MMX2 */
2574             x86_reg xInc_shr16 = (x86_reg) (xInc >> 16);
2575             uint16_t xInc_mask = xInc & 0xffff;
2576             __asm__ volatile(
2577             "xor %%"REG_a", %%"REG_a"               \n\t" // i
2578             "xor %%"REG_d", %%"REG_d"               \n\t" // xx
2579             "xorl    %%ecx, %%ecx                   \n\t" // 2*xalpha
2580             ASMALIGN(4)
2581             "1:                                     \n\t"
2582             "mov        %0, %%"REG_S"               \n\t"
2583             "movzbl  (%%"REG_S", %%"REG_d"), %%edi  \n\t" //src[xx]
2584             "movzbl 1(%%"REG_S", %%"REG_d"), %%esi  \n\t" //src[xx+1]
2585             "subl    %%edi, %%esi                   \n\t" //src[xx+1] - src[xx]
2586             "imull   %%ecx, %%esi                   \n\t" //(src[xx+1] - src[xx])*2*xalpha
2587             "shll      $16, %%edi                   \n\t"
2588             "addl    %%edi, %%esi                   \n\t" //src[xx+1]*2*xalpha + src[xx]*(1-2*xalpha)
2589             "mov        %1, %%"REG_D"               \n\t"
2590             "shrl       $9, %%esi                   \n\t"
2591             "movw     %%si, (%%"REG_D", %%"REG_a", 2)   \n\t"
2592
2593             "movzbl    (%5, %%"REG_d"), %%edi       \n\t" //src[xx]
2594             "movzbl   1(%5, %%"REG_d"), %%esi       \n\t" //src[xx+1]
2595             "subl    %%edi, %%esi                   \n\t" //src[xx+1] - src[xx]
2596             "imull   %%ecx, %%esi                   \n\t" //(src[xx+1] - src[xx])*2*xalpha
2597             "shll      $16, %%edi                   \n\t"
2598             "addl    %%edi, %%esi                   \n\t" //src[xx+1]*2*xalpha + src[xx]*(1-2*xalpha)
2599             "mov        %1, %%"REG_D"               \n\t"
2600             "shrl       $9, %%esi                   \n\t"
2601             "movw     %%si, "AV_STRINGIFY(VOF)"(%%"REG_D", %%"REG_a", 2)   \n\t"
2602
2603             "addw       %4, %%cx                    \n\t" //2*xalpha += xInc&0xFF
2604             "adc        %3, %%"REG_d"               \n\t" //xx+= xInc>>8 + carry
2605             "add        $1, %%"REG_a"               \n\t"
2606             "cmp        %2, %%"REG_a"               \n\t"
2607             " jb        1b                          \n\t"
2608
2609 /* GCC 3.3 makes MPlayer crash on IA-32 machines when using "g" operand here,
2610    which is needed to support GCC 4.0. */
2611 #if ARCH_X86_64 && ((__GNUC__ > 3) || (__GNUC__ == 3 && __GNUC_MINOR__ >= 4))
2612             :: "m" (src1), "m" (dst), "g" (dstWidth), "m" (xInc_shr16), "m" (xInc_mask),
2613 #else
2614             :: "m" (src1), "m" (dst), "m" (dstWidth), "m" (xInc_shr16), "m" (xInc_mask),
2615 #endif
2616             "r" (src2)
2617             : "%"REG_a, "%"REG_d, "%ecx", "%"REG_D, "%esi"
2618             );
2619 #if HAVE_MMX2
2620         } //if MMX2 can't be used
2621 #endif
2622 #else
2623         c->hcscale_fast(c, dst, dstWidth, src1, src2, srcW, xInc);
2624 #endif /* ARCH_X86 */
2625     }
2626     if(c->srcRange != c->dstRange && !(isRGB(c->dstFormat) || isBGR(c->dstFormat))){
2627         int i;
2628         //FIXME all pal and rgb srcFormats could do this convertion as well
2629         //FIXME all scalers more complex than bilinear could do half of this transform
2630         if(c->srcRange){
2631             for (i=0; i<dstWidth; i++){
2632                 dst[i     ]= (dst[i     ]*1799 + 4081085)>>11; //1469
2633                 dst[i+VOFW]= (dst[i+VOFW]*1799 + 4081085)>>11; //1469
2634             }
2635         }else{
2636             for (i=0; i<dstWidth; i++){
2637                 dst[i     ]= (FFMIN(dst[i     ],30775)*4663 - 9289992)>>12; //-264
2638                 dst[i+VOFW]= (FFMIN(dst[i+VOFW],30775)*4663 - 9289992)>>12; //-264
2639             }
2640         }
2641     }
2642 }
2643
2644 static int RENAME(swScale)(SwsContext *c, uint8_t* src[], int srcStride[], int srcSliceY,
2645                            int srcSliceH, uint8_t* dst[], int dstStride[]){
2646
2647     /* load a few things into local vars to make the code more readable? and faster */
2648     const int srcW= c->srcW;
2649     const int dstW= c->dstW;
2650     const int dstH= c->dstH;
2651     const int chrDstW= c->chrDstW;
2652     const int chrSrcW= c->chrSrcW;
2653     const int lumXInc= c->lumXInc;
2654     const int chrXInc= c->chrXInc;
2655     const int dstFormat= c->dstFormat;
2656     const int srcFormat= c->srcFormat;
2657     const int flags= c->flags;
2658     int16_t *vLumFilterPos= c->vLumFilterPos;
2659     int16_t *vChrFilterPos= c->vChrFilterPos;
2660     int16_t *hLumFilterPos= c->hLumFilterPos;
2661     int16_t *hChrFilterPos= c->hChrFilterPos;
2662     int16_t *vLumFilter= c->vLumFilter;
2663     int16_t *vChrFilter= c->vChrFilter;
2664     int16_t *hLumFilter= c->hLumFilter;
2665     int16_t *hChrFilter= c->hChrFilter;
2666     int32_t *lumMmxFilter= c->lumMmxFilter;
2667     int32_t *chrMmxFilter= c->chrMmxFilter;
2668     int32_t *alpMmxFilter= c->alpMmxFilter;
2669     const int vLumFilterSize= c->vLumFilterSize;
2670     const int vChrFilterSize= c->vChrFilterSize;
2671     const int hLumFilterSize= c->hLumFilterSize;
2672     const int hChrFilterSize= c->hChrFilterSize;
2673     int16_t **lumPixBuf= c->lumPixBuf;
2674     int16_t **chrPixBuf= c->chrPixBuf;
2675     int16_t **alpPixBuf= c->alpPixBuf;
2676     const int vLumBufSize= c->vLumBufSize;
2677     const int vChrBufSize= c->vChrBufSize;
2678     uint8_t *formatConvBuffer= c->formatConvBuffer;
2679     const int chrSrcSliceY= srcSliceY >> c->chrSrcVSubSample;
2680     const int chrSrcSliceH= -((-srcSliceH) >> c->chrSrcVSubSample);
2681     int lastDstY;
2682     uint32_t *pal=c->pal_yuv;
2683
2684     /* vars which will change and which we need to store back in the context */
2685     int dstY= c->dstY;
2686     int lumBufIndex= c->lumBufIndex;
2687     int chrBufIndex= c->chrBufIndex;
2688     int lastInLumBuf= c->lastInLumBuf;
2689     int lastInChrBuf= c->lastInChrBuf;
2690
2691     if (isPacked(c->srcFormat)){
2692         src[0]=
2693         src[1]=
2694         src[2]=
2695         src[3]= src[0];
2696         srcStride[0]=
2697         srcStride[1]=
2698         srcStride[2]=
2699         srcStride[3]= srcStride[0];
2700     }
2701     srcStride[1]<<= c->vChrDrop;
2702     srcStride[2]<<= c->vChrDrop;
2703
2704     //printf("swscale %X %X %X -> %X %X %X\n", (int)src[0], (int)src[1], (int)src[2],
2705     //       (int)dst[0], (int)dst[1], (int)dst[2]);
2706
2707 #if 0 //self test FIXME move to a vfilter or something
2708     {
2709     static volatile int i=0;
2710     i++;
2711     if (srcFormat==PIX_FMT_YUV420P && i==1 && srcSliceH>= c->srcH)
2712         selfTest(src, srcStride, c->srcW, c->srcH);
2713     i--;
2714     }
2715 #endif
2716
2717     //printf("sws Strides:%d %d %d -> %d %d %d\n", srcStride[0],srcStride[1],srcStride[2],
2718     //dstStride[0],dstStride[1],dstStride[2]);
2719
2720     if (dstStride[0]%8 !=0 || dstStride[1]%8 !=0 || dstStride[2]%8 !=0 || dstStride[3]%8 != 0)
2721     {
2722         static int warnedAlready=0; //FIXME move this into the context perhaps
2723         if (flags & SWS_PRINT_INFO && !warnedAlready)
2724         {
2725             av_log(c, AV_LOG_WARNING, "Warning: dstStride is not aligned!\n"
2726                    "         ->cannot do aligned memory accesses anymore\n");
2727             warnedAlready=1;
2728         }
2729     }
2730
2731     /* Note the user might start scaling the picture in the middle so this
2732        will not get executed. This is not really intended but works
2733        currently, so people might do it. */
2734     if (srcSliceY ==0){
2735         lumBufIndex=0;
2736         chrBufIndex=0;
2737         dstY=0;
2738         lastInLumBuf= -1;
2739         lastInChrBuf= -1;
2740     }
2741
2742     lastDstY= dstY;
2743
2744     for (;dstY < dstH; dstY++){
2745         unsigned char *dest =dst[0]+dstStride[0]*dstY;
2746         const int chrDstY= dstY>>c->chrDstVSubSample;
2747         unsigned char *uDest=dst[1]+dstStride[1]*chrDstY;
2748         unsigned char *vDest=dst[2]+dstStride[2]*chrDstY;
2749         unsigned char *aDest=(CONFIG_SWSCALE_ALPHA && alpPixBuf) ? dst[3]+dstStride[3]*dstY : NULL;
2750
2751         const int firstLumSrcY= vLumFilterPos[dstY]; //First line needed as input
2752         const int firstChrSrcY= vChrFilterPos[chrDstY]; //First line needed as input
2753         const int lastLumSrcY= firstLumSrcY + vLumFilterSize -1; // Last line needed as input
2754         const int lastChrSrcY= firstChrSrcY + vChrFilterSize -1; // Last line needed as input
2755
2756         //printf("dstY:%d dstH:%d firstLumSrcY:%d lastInLumBuf:%d vLumBufSize: %d vChrBufSize: %d slice: %d %d vLumFilterSize: %d firstChrSrcY: %d vChrFilterSize: %d c->chrSrcVSubSample: %d\n",
2757         // dstY, dstH, firstLumSrcY, lastInLumBuf, vLumBufSize, vChrBufSize, srcSliceY, srcSliceH, vLumFilterSize, firstChrSrcY, vChrFilterSize,  c->chrSrcVSubSample);
2758         //handle holes (FAST_BILINEAR & weird filters)
2759         if (firstLumSrcY > lastInLumBuf) lastInLumBuf= firstLumSrcY-1;
2760         if (firstChrSrcY > lastInChrBuf) lastInChrBuf= firstChrSrcY-1;
2761         //printf("%d %d %d\n", firstChrSrcY, lastInChrBuf, vChrBufSize);
2762         assert(firstLumSrcY >= lastInLumBuf - vLumBufSize + 1);
2763         assert(firstChrSrcY >= lastInChrBuf - vChrBufSize + 1);
2764
2765         // Do we have enough lines in this slice to output the dstY line
2766         if (lastLumSrcY < srcSliceY + srcSliceH && lastChrSrcY < -((-srcSliceY - srcSliceH)>>c->chrSrcVSubSample))
2767         {
2768             //Do horizontal scaling
2769             while(lastInLumBuf < lastLumSrcY)
2770             {
2771                 uint8_t *src1= src[0]+(lastInLumBuf + 1 - srcSliceY)*srcStride[0];
2772                 uint8_t *src2= src[3]+(lastInLumBuf + 1 - srcSliceY)*srcStride[3];
2773                 lumBufIndex++;
2774                 //printf("%d %d %d %d\n", lumBufIndex, vLumBufSize, lastInLumBuf,  lastLumSrcY);
2775                 assert(lumBufIndex < 2*vLumBufSize);
2776                 assert(lastInLumBuf + 1 - srcSliceY < srcSliceH);
2777                 assert(lastInLumBuf + 1 - srcSliceY >= 0);
2778                 //printf("%d %d\n", lumBufIndex, vLumBufSize);
2779                 RENAME(hyscale)(c, lumPixBuf[ lumBufIndex ], dstW, src1, srcW, lumXInc,
2780                                 flags, hLumFilter, hLumFilterPos, hLumFilterSize,
2781                                 c->srcFormat, formatConvBuffer,
2782                                 pal, 0);
2783                 if (CONFIG_SWSCALE_ALPHA && alpPixBuf)
2784                     RENAME(hyscale)(c, alpPixBuf[ lumBufIndex ], dstW, src2, srcW, lumXInc,
2785                                     flags, hLumFilter, hLumFilterPos, hLumFilterSize,
2786                                     c->srcFormat, formatConvBuffer,
2787                                     pal, 1);
2788                 lastInLumBuf++;
2789             }
2790             while(lastInChrBuf < lastChrSrcY)
2791             {
2792                 uint8_t *src1= src[1]+(lastInChrBuf + 1 - chrSrcSliceY)*srcStride[1];
2793                 uint8_t *src2= src[2]+(lastInChrBuf + 1 - chrSrcSliceY)*srcStride[2];
2794                 chrBufIndex++;
2795                 assert(chrBufIndex < 2*vChrBufSize);
2796                 assert(lastInChrBuf + 1 - chrSrcSliceY < (chrSrcSliceH));
2797                 assert(lastInChrBuf + 1 - chrSrcSliceY >= 0);
2798                 //FIXME replace parameters through context struct (some at least)
2799
2800                 if (!(isGray(srcFormat) || isGray(dstFormat)))
2801                     RENAME(hcscale)(c, chrPixBuf[ chrBufIndex ], chrDstW, src1, src2, chrSrcW, chrXInc,
2802                                     flags, hChrFilter, hChrFilterPos, hChrFilterSize,
2803                                     c->srcFormat, formatConvBuffer,
2804                                     pal);
2805                 lastInChrBuf++;
2806             }
2807             //wrap buf index around to stay inside the ring buffer
2808             if (lumBufIndex >= vLumBufSize) lumBufIndex-= vLumBufSize;
2809             if (chrBufIndex >= vChrBufSize) chrBufIndex-= vChrBufSize;
2810         }
2811         else // not enough lines left in this slice -> load the rest in the buffer
2812         {
2813             /* printf("%d %d Last:%d %d LastInBuf:%d %d Index:%d %d Y:%d FSize: %d %d BSize: %d %d\n",
2814             firstChrSrcY,firstLumSrcY,lastChrSrcY,lastLumSrcY,
2815             lastInChrBuf,lastInLumBuf,chrBufIndex,lumBufIndex,dstY,vChrFilterSize,vLumFilterSize,
2816             vChrBufSize, vLumBufSize);*/
2817
2818             //Do horizontal scaling
2819             while(lastInLumBuf+1 < srcSliceY + srcSliceH)
2820             {
2821                 uint8_t *src1= src[0]+(lastInLumBuf + 1 - srcSliceY)*srcStride[0];
2822                 uint8_t *src2= src[3]+(lastInLumBuf + 1 - srcSliceY)*srcStride[3];
2823                 lumBufIndex++;
2824                 assert(lumBufIndex < 2*vLumBufSize);
2825                 assert(lastInLumBuf + 1 - srcSliceY < srcSliceH);
2826                 assert(lastInLumBuf + 1 - srcSliceY >= 0);
2827                 RENAME(hyscale)(c, lumPixBuf[ lumBufIndex ], dstW, src1, srcW, lumXInc,
2828                                 flags, hLumFilter, hLumFilterPos, hLumFilterSize,
2829                                 c->srcFormat, formatConvBuffer,
2830                                 pal, 0);
2831                 if (CONFIG_SWSCALE_ALPHA && alpPixBuf)
2832                     RENAME(hyscale)(c, alpPixBuf[ lumBufIndex ], dstW, src2, srcW, lumXInc,
2833                                     flags, hLumFilter, hLumFilterPos, hLumFilterSize,
2834                                     c->srcFormat, formatConvBuffer,
2835                                     pal, 1);
2836                 lastInLumBuf++;
2837             }
2838             while(lastInChrBuf+1 < (chrSrcSliceY + chrSrcSliceH))
2839             {
2840                 uint8_t *src1= src[1]+(lastInChrBuf + 1 - chrSrcSliceY)*srcStride[1];
2841                 uint8_t *src2= src[2]+(lastInChrBuf + 1 - chrSrcSliceY)*srcStride[2];
2842                 chrBufIndex++;
2843                 assert(chrBufIndex < 2*vChrBufSize);
2844                 assert(lastInChrBuf + 1 - chrSrcSliceY < chrSrcSliceH);
2845                 assert(lastInChrBuf + 1 - chrSrcSliceY >= 0);
2846
2847                 if (!(isGray(srcFormat) || isGray(dstFormat)))
2848                     RENAME(hcscale)(c, chrPixBuf[ chrBufIndex ], chrDstW, src1, src2, chrSrcW, chrXInc,
2849                             flags, hChrFilter, hChrFilterPos, hChrFilterSize,
2850                             c->srcFormat, formatConvBuffer,
2851                             pal);
2852                 lastInChrBuf++;
2853             }
2854             //wrap buf index around to stay inside the ring buffer
2855             if (lumBufIndex >= vLumBufSize) lumBufIndex-= vLumBufSize;
2856             if (chrBufIndex >= vChrBufSize) chrBufIndex-= vChrBufSize;
2857             break; //we can't output a dstY line so let's try with the next slice
2858         }
2859
2860 #if HAVE_MMX
2861         c->blueDither= ff_dither8[dstY&1];
2862         if (c->dstFormat == PIX_FMT_RGB555 || c->dstFormat == PIX_FMT_BGR555)
2863             c->greenDither= ff_dither8[dstY&1];
2864         else
2865             c->greenDither= ff_dither4[dstY&1];
2866         c->redDither= ff_dither8[(dstY+1)&1];
2867 #endif
2868         if (dstY < dstH-2)
2869         {
2870             const int16_t **lumSrcPtr= (const int16_t **) lumPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize;
2871             const int16_t **chrSrcPtr= (const int16_t **) chrPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
2872             const int16_t **alpSrcPtr= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? (const int16_t **) alpPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize : NULL;
2873 #if HAVE_MMX
2874             int i;
2875         if (flags & SWS_ACCURATE_RND){
2876             int s= APCK_SIZE / 8;
2877             for (i=0; i<vLumFilterSize; i+=2){
2878                 *(void**)&lumMmxFilter[s*i              ]= lumSrcPtr[i  ];
2879                 *(void**)&lumMmxFilter[s*i+APCK_PTR2/4  ]= lumSrcPtr[i+(vLumFilterSize>1)];
2880                           lumMmxFilter[s*i+APCK_COEF/4  ]=
2881                           lumMmxFilter[s*i+APCK_COEF/4+1]= vLumFilter[dstY*vLumFilterSize + i    ]
2882                     + (vLumFilterSize>1 ? vLumFilter[dstY*vLumFilterSize + i + 1]<<16 : 0);
2883                 if (CONFIG_SWSCALE_ALPHA && alpPixBuf){
2884                     *(void**)&alpMmxFilter[s*i              ]= alpSrcPtr[i  ];
2885                     *(void**)&alpMmxFilter[s*i+APCK_PTR2/4  ]= alpSrcPtr[i+(vLumFilterSize>1)];
2886                               alpMmxFilter[s*i+APCK_COEF/4  ]=
2887                               alpMmxFilter[s*i+APCK_COEF/4+1]= lumMmxFilter[s*i+APCK_COEF/4  ];
2888                 }
2889             }
2890             for (i=0; i<vChrFilterSize; i+=2){
2891                 *(void**)&chrMmxFilter[s*i              ]= chrSrcPtr[i  ];
2892                 *(void**)&chrMmxFilter[s*i+APCK_PTR2/4  ]= chrSrcPtr[i+(vChrFilterSize>1)];
2893                           chrMmxFilter[s*i+APCK_COEF/4  ]=
2894                           chrMmxFilter[s*i+APCK_COEF/4+1]= vChrFilter[chrDstY*vChrFilterSize + i    ]
2895                     + (vChrFilterSize>1 ? vChrFilter[chrDstY*vChrFilterSize + i + 1]<<16 : 0);
2896             }
2897         }else{
2898             for (i=0; i<vLumFilterSize; i++)
2899             {
2900                 lumMmxFilter[4*i+0]= (int32_t)lumSrcPtr[i];
2901                 lumMmxFilter[4*i+1]= (uint64_t)lumSrcPtr[i] >> 32;
2902                 lumMmxFilter[4*i+2]=
2903                 lumMmxFilter[4*i+3]=
2904                     ((uint16_t)vLumFilter[dstY*vLumFilterSize + i])*0x10001;
2905                 if (CONFIG_SWSCALE_ALPHA && alpPixBuf){
2906                     alpMmxFilter[4*i+0]= (int32_t)alpSrcPtr[i];
2907                     alpMmxFilter[4*i+1]= (uint64_t)alpSrcPtr[i] >> 32;
2908                     alpMmxFilter[4*i+2]=
2909                     alpMmxFilter[4*i+3]= lumMmxFilter[4*i+2];
2910                 }
2911             }
2912             for (i=0; i<vChrFilterSize; i++)
2913             {
2914                 chrMmxFilter[4*i+0]= (int32_t)chrSrcPtr[i];
2915                 chrMmxFilter[4*i+1]= (uint64_t)chrSrcPtr[i] >> 32;
2916                 chrMmxFilter[4*i+2]=
2917                 chrMmxFilter[4*i+3]=
2918                     ((uint16_t)vChrFilter[chrDstY*vChrFilterSize + i])*0x10001;
2919             }
2920         }
2921 #endif
2922             if (dstFormat == PIX_FMT_NV12 || dstFormat == PIX_FMT_NV21){
2923                 const int chrSkipMask= (1<<c->chrDstVSubSample)-1;
2924                 if (dstY&chrSkipMask) uDest= NULL; //FIXME split functions in lumi / chromi
2925                 c->yuv2nv12X(c,
2926                     vLumFilter+dstY*vLumFilterSize   , lumSrcPtr, vLumFilterSize,
2927                     vChrFilter+chrDstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
2928                     dest, uDest, dstW, chrDstW, dstFormat);
2929             }
2930             else if (isPlanarYUV(dstFormat) || dstFormat==PIX_FMT_GRAY8) //YV12 like
2931             {
2932                 const int chrSkipMask= (1<<c->chrDstVSubSample)-1;
2933                 if ((dstY&chrSkipMask) || isGray(dstFormat)) uDest=vDest= NULL; //FIXME split functions in lumi / chromi
2934                 if (vLumFilterSize == 1 && vChrFilterSize == 1) // unscaled YV12
2935                 {
2936                     int16_t *lumBuf = lumPixBuf[0];
2937                     int16_t *chrBuf= chrPixBuf[0];
2938                     int16_t *alpBuf= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? alpPixBuf[0] : NULL;
2939                     c->yuv2yuv1(c, lumBuf, chrBuf, alpBuf, dest, uDest, vDest, aDest, dstW, chrDstW);
2940                 }
2941                 else //General YV12
2942                 {
2943                     c->yuv2yuvX(c,
2944                         vLumFilter+dstY*vLumFilterSize   , lumSrcPtr, vLumFilterSize,
2945                         vChrFilter+chrDstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
2946                         alpSrcPtr, dest, uDest, vDest, aDest, dstW, chrDstW);
2947                 }
2948             }
2949             else
2950             {
2951                 assert(lumSrcPtr + vLumFilterSize - 1 < lumPixBuf + vLumBufSize*2);
2952                 assert(chrSrcPtr + vChrFilterSize - 1 < chrPixBuf + vChrBufSize*2);
2953                 if (vLumFilterSize == 1 && vChrFilterSize == 2) //unscaled RGB
2954                 {
2955                     int chrAlpha= vChrFilter[2*dstY+1];
2956                     if(flags & SWS_FULL_CHR_H_INT){
2957                         yuv2rgbXinC_full(c, //FIXME write a packed1_full function
2958                             vLumFilter+dstY*vLumFilterSize, lumSrcPtr, vLumFilterSize,
2959                             vChrFilter+dstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
2960                             alpSrcPtr, dest, dstW, dstY);
2961                     }else{
2962                         c->yuv2packed1(c, *lumSrcPtr, *chrSrcPtr, *(chrSrcPtr+1),
2963                             alpPixBuf ? *alpSrcPtr : NULL,
2964                             dest, dstW, chrAlpha, dstFormat, flags, dstY);
2965                     }
2966                 }
2967                 else if (vLumFilterSize == 2 && vChrFilterSize == 2) //bilinear upscale RGB
2968                 {
2969                     int lumAlpha= vLumFilter[2*dstY+1];
2970                     int chrAlpha= vChrFilter[2*dstY+1];
2971                     lumMmxFilter[2]=
2972                     lumMmxFilter[3]= vLumFilter[2*dstY   ]*0x10001;
2973                     chrMmxFilter[2]=
2974                     chrMmxFilter[3]= vChrFilter[2*chrDstY]*0x10001;
2975                     if(flags & SWS_FULL_CHR_H_INT){
2976                         yuv2rgbXinC_full(c, //FIXME write a packed2_full function
2977                             vLumFilter+dstY*vLumFilterSize, lumSrcPtr, vLumFilterSize,
2978                             vChrFilter+dstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
2979                             alpSrcPtr, dest, dstW, dstY);
2980                     }else{
2981                         c->yuv2packed2(c, *lumSrcPtr, *(lumSrcPtr+1), *chrSrcPtr, *(chrSrcPtr+1),
2982                             alpPixBuf ? *alpSrcPtr : NULL, alpPixBuf ? *(alpSrcPtr+1) : NULL,
2983                             dest, dstW, lumAlpha, chrAlpha, dstY);
2984                     }
2985                 }
2986                 else //general RGB
2987                 {
2988                     if(flags & SWS_FULL_CHR_H_INT){
2989                         yuv2rgbXinC_full(c,
2990                             vLumFilter+dstY*vLumFilterSize, lumSrcPtr, vLumFilterSize,
2991                             vChrFilter+dstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
2992                             alpSrcPtr, dest, dstW, dstY);
2993                     }else{
2994                         c->yuv2packedX(c,
2995                             vLumFilter+dstY*vLumFilterSize, lumSrcPtr, vLumFilterSize,
2996                             vChrFilter+dstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
2997                             alpSrcPtr, dest, dstW, dstY);
2998                     }
2999                 }
3000             }
3001         }
3002         else // hmm looks like we can't use MMX here without overwriting this array's tail
3003         {
3004             const int16_t **lumSrcPtr= (const int16_t **)lumPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize;
3005             const int16_t **chrSrcPtr= (const int16_t **)chrPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
3006             const int16_t **alpSrcPtr= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? (const int16_t **)alpPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize : NULL;
3007             if (dstFormat == PIX_FMT_NV12 || dstFormat == PIX_FMT_NV21){
3008                 const int chrSkipMask= (1<<c->chrDstVSubSample)-1;
3009                 if (dstY&chrSkipMask) uDest= NULL; //FIXME split functions in lumi / chromi
3010                 yuv2nv12XinC(
3011                     vLumFilter+dstY*vLumFilterSize   , lumSrcPtr, vLumFilterSize,
3012                     vChrFilter+chrDstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
3013                     dest, uDest, dstW, chrDstW, dstFormat);
3014             }
3015             else if (isPlanarYUV(dstFormat) || dstFormat==PIX_FMT_GRAY8) //YV12
3016             {
3017                 const int chrSkipMask= (1<<c->chrDstVSubSample)-1;
3018                 if ((dstY&chrSkipMask) || isGray(dstFormat)) uDest=vDest= NULL; //FIXME split functions in lumi / chromi
3019                 yuv2yuvXinC(
3020                     vLumFilter+dstY*vLumFilterSize   , lumSrcPtr, vLumFilterSize,
3021                     vChrFilter+chrDstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
3022                     alpSrcPtr, dest, uDest, vDest, aDest, dstW, chrDstW);
3023             }
3024             else
3025             {
3026                 assert(lumSrcPtr + vLumFilterSize - 1 < lumPixBuf + vLumBufSize*2);
3027                 assert(chrSrcPtr + vChrFilterSize - 1 < chrPixBuf + vChrBufSize*2);
3028                 if(flags & SWS_FULL_CHR_H_INT){
3029                     yuv2rgbXinC_full(c,
3030                         vLumFilter+dstY*vLumFilterSize, lumSrcPtr, vLumFilterSize,
3031                         vChrFilter+dstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
3032                         alpSrcPtr, dest, dstW, dstY);
3033                 }else{
3034                     yuv2packedXinC(c,
3035                         vLumFilter+dstY*vLumFilterSize, lumSrcPtr, vLumFilterSize,
3036                         vChrFilter+dstY*vChrFilterSize, chrSrcPtr, vChrFilterSize,
3037                         alpSrcPtr, dest, dstW, dstY);
3038                 }
3039             }
3040         }
3041     }
3042
3043     if ((dstFormat == PIX_FMT_YUVA420P) && !alpPixBuf)
3044         fillPlane(dst[3], dstStride[3], dstW, dstY-lastDstY, lastDstY, 255);
3045
3046 #if HAVE_MMX
3047     if (flags & SWS_CPU_CAPS_MMX2 )  __asm__ volatile("sfence":::"memory");
3048     /* On K6 femms is faster than emms. On K7 femms is directly mapped to emms. */
3049     if (flags & SWS_CPU_CAPS_3DNOW)  __asm__ volatile("femms" :::"memory");
3050     else                             __asm__ volatile("emms"  :::"memory");
3051 #endif
3052     /* store changed local vars back in the context */
3053     c->dstY= dstY;
3054     c->lumBufIndex= lumBufIndex;
3055     c->chrBufIndex= chrBufIndex;
3056     c->lastInLumBuf= lastInLumBuf;
3057     c->lastInChrBuf= lastInChrBuf;
3058
3059     return dstY - lastDstY;
3060 }
3061
3062 static void RENAME(sws_init_swScale)(SwsContext *c)
3063 {
3064     enum PixelFormat srcFormat = c->srcFormat;
3065
3066     c->yuv2nv12X    = RENAME(yuv2nv12X   );
3067     c->yuv2yuv1     = RENAME(yuv2yuv1    );
3068     c->yuv2yuvX     = RENAME(yuv2yuvX    );
3069     c->yuv2packed1  = RENAME(yuv2packed1 );
3070     c->yuv2packed2  = RENAME(yuv2packed2 );
3071     c->yuv2packedX  = RENAME(yuv2packedX );
3072
3073     c->hScale       = RENAME(hScale      );
3074
3075     c->hyscale_fast = RENAME(hyscale_fast);
3076     c->hcscale_fast = RENAME(hcscale_fast);
3077
3078     c->hcscale_internal = NULL;
3079     switch(srcFormat) {
3080         case PIX_FMT_YUYV422  : c->hcscale_internal = RENAME(yuy2ToUV); break;
3081         case PIX_FMT_UYVY422  : c->hcscale_internal = RENAME(uyvyToUV); break;
3082         case PIX_FMT_RGB8     :
3083         case PIX_FMT_BGR8     :
3084         case PIX_FMT_PAL8     :
3085         case PIX_FMT_BGR4_BYTE:
3086         case PIX_FMT_RGB4_BYTE: c->hcscale_internal = RENAME(palToUV); break;
3087     }
3088     if (c->chrSrcHSubSample) {
3089         switch(srcFormat) {
3090         case PIX_FMT_RGB32  :
3091         case PIX_FMT_RGB32_1: c->hcscale_internal = RENAME(bgr32ToUV_half); break;
3092         case PIX_FMT_BGR24  : c->hcscale_internal = RENAME(bgr24ToUV_half); break;
3093         case PIX_FMT_BGR565 : c->hcscale_internal = RENAME(bgr16ToUV_half); break;
3094         case PIX_FMT_BGR555 : c->hcscale_internal = RENAME(bgr15ToUV_half); break;
3095         case PIX_FMT_BGR32  :
3096         case PIX_FMT_BGR32_1: c->hcscale_internal = RENAME(rgb32ToUV_half); break;
3097         case PIX_FMT_RGB24  : c->hcscale_internal = RENAME(rgb24ToUV_half); break;
3098         case PIX_FMT_RGB565 : c->hcscale_internal = RENAME(rgb16ToUV_half); break;
3099         case PIX_FMT_RGB555 : c->hcscale_internal = RENAME(rgb15ToUV_half); break;
3100         }
3101     } else {
3102         switch(srcFormat) {
3103         case PIX_FMT_RGB32  :
3104         case PIX_FMT_RGB32_1: c->hcscale_internal = RENAME(bgr32ToUV); break;
3105         case PIX_FMT_BGR24  : c->hcscale_internal = RENAME(bgr24ToUV); break;
3106         case PIX_FMT_BGR565 : c->hcscale_internal = RENAME(bgr16ToUV); break;
3107         case PIX_FMT_BGR555 : c->hcscale_internal = RENAME(bgr15ToUV); break;
3108         case PIX_FMT_BGR32  :
3109         case PIX_FMT_BGR32_1: c->hcscale_internal = RENAME(rgb32ToUV); break;
3110         case PIX_FMT_RGB24  : c->hcscale_internal = RENAME(rgb24ToUV); break;
3111         case PIX_FMT_RGB565 : c->hcscale_internal = RENAME(rgb16ToUV); break;
3112         case PIX_FMT_RGB555 : c->hcscale_internal = RENAME(rgb15ToUV); break;
3113         }
3114     }
3115
3116     c->hyscale_internal = NULL;
3117     switch (srcFormat) {
3118     case PIX_FMT_YUYV422  :
3119     case PIX_FMT_GRAY16BE : c->hyscale_internal = RENAME(yuy2ToY); break;
3120     case PIX_FMT_UYVY422  :
3121     case PIX_FMT_GRAY16LE : c->hyscale_internal = RENAME(uyvyToY); break;
3122     case PIX_FMT_BGR24    : c->hyscale_internal = RENAME(bgr24ToY); break;
3123     case PIX_FMT_BGR565   : c->hyscale_internal = RENAME(bgr16ToY); break;
3124     case PIX_FMT_BGR555   : c->hyscale_internal = RENAME(bgr15ToY); break;
3125     case PIX_FMT_RGB24    : c->hyscale_internal = RENAME(rgb24ToY); break;
3126     case PIX_FMT_RGB565   : c->hyscale_internal = RENAME(rgb16ToY); break;
3127     case PIX_FMT_RGB555   : c->hyscale_internal = RENAME(rgb15ToY); break;
3128     case PIX_FMT_RGB8     :
3129     case PIX_FMT_BGR8     :
3130     case PIX_FMT_PAL8     :
3131     case PIX_FMT_BGR4_BYTE:
3132     case PIX_FMT_RGB4_BYTE: c->hyscale_internal = RENAME(palToY); break;
3133     case PIX_FMT_MONOBLACK: c->hyscale_internal = RENAME(monoblack2Y); break;
3134     case PIX_FMT_MONOWHITE: c->hyscale_internal = RENAME(monowhite2Y); break;
3135     }
3136     if (c->alpPixBuf) {
3137         switch (srcFormat) {
3138         case PIX_FMT_RGB32  :
3139         case PIX_FMT_RGB32_1:
3140         case PIX_FMT_BGR32  :
3141         case PIX_FMT_BGR32_1: c->hyscale_internal = RENAME(abgrToA); break;
3142         }
3143     } else {
3144         switch (srcFormat) {
3145         case PIX_FMT_RGB32  :
3146         case PIX_FMT_RGB32_1: c->hyscale_internal = RENAME(bgr32ToY); break;
3147         case PIX_FMT_BGR32  :
3148         case PIX_FMT_BGR32_1: c->hyscale_internal = RENAME(rgb32ToY); break;
3149         }
3150     }
3151 }