alternative LGPL-licensed, MMX-optimized YUV to RGB conversion routines

written by Kostya Shishkov

Originally committed as revision 31135 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
This commit is contained in:
Diego Biurrun 2010-05-04 09:11:01 +00:00
parent 4fe4bb6150
commit f4ea7c8911
4 changed files with 469 additions and 3 deletions

View File

@ -12,10 +12,9 @@ OBJS-$(ARCH_BFIN) += bfin/internal_bfin.o \
bfin/yuv2rgb_bfin.o
OBJS-$(CONFIG_MLIB) += mlib/yuv2rgb_mlib.o
OBJS-$(HAVE_ALTIVEC) += ppc/yuv2rgb_altivec.o
OBJS-$(HAVE_MMX) += x86/yuv2rgb_mmx.o
OBJS-$(HAVE_VIS) += sparc/yuv2rgb_vis.o
MMX-OBJS-$(CONFIG_GPL) += x86/yuv2rgb_mmx.o \
TESTPROGS = colorspace swscale
DIRS = bfin mlib ppc sparc x86

View File

@ -49,14 +49,22 @@ DECLARE_ASM_CONST(8, uint64_t, mmx_grnmask) = 0xfcfcfcfcfcfcfcfcULL;
#define HAVE_MMX2 0
#define HAVE_AMD3DNOW 0
#define RENAME(a) a ## _MMX
#if CONFIG_GPL
#include "yuv2rgb_template.c"
#else
#include "yuv2rgb_template2.c"
#endif
//MMX2 versions
#undef RENAME
#undef HAVE_MMX2
#define HAVE_MMX2 1
#define RENAME(a) a ## _MMX2
#if CONFIG_GPL
#include "yuv2rgb_template.c"
#else
#include "yuv2rgb_template2.c"
#endif
SwsFunc ff_yuv2rgb_init_mmx(SwsContext *c)
{

View File

@ -0,0 +1,459 @@
/*
* software YUV to RGB converter
*
* Copyright (C) 2001-2007 Michael Niedermayer
* (c) 2010 Konstantin Shishkov
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#undef MOVNTQ
#undef EMMS
#undef SFENCE
#if HAVE_AMD3DNOW
/* On K6 femms is faster than emms. On K7 femms is directly mapped to emms. */
#define EMMS "femms"
#else
#define EMMS "emms"
#endif
#if HAVE_MMX2
#define MOVNTQ "movntq"
#define SFENCE "sfence"
#else
#define MOVNTQ "movq"
#define SFENCE " # nop"
#endif
#define REG_BLUE "0"
#define REG_RED "1"
#define REG_GREEN "2"
#define REG_ALPHA "3"
#define YUV2RGB_LOOP(depth) \
h_size = (c->dstW + 7) & ~7; \
if (h_size * depth > FFABS(dstStride[0])) \
h_size -= 8; \
\
if (c->srcFormat == PIX_FMT_YUV422P) { \
srcStride[1] *= 2; \
srcStride[2] *= 2; \
} \
\
__asm__ volatile ("pxor %mm4, %mm4\n\t"); \
for (y = 0; y < srcSliceH; y++) { \
uint8_t *image = dst[0] + (y + srcSliceY) * dstStride[0]; \
const uint8_t *py = src[0] + y * srcStride[0]; \
const uint8_t *pu = src[1] + (y >> 1) * srcStride[1]; \
const uint8_t *pv = src[2] + (y >> 1) * srcStride[2]; \
x86_reg index = -h_size / 2; \
#define YUV2RGB_INITIAL_LOAD \
__asm__ volatile ( \
"movq (%5, %0, 2), %%mm6\n\t" \
"movd (%2, %0), %%mm0\n\t" \
"movd (%3, %0), %%mm1\n\t" \
"1: \n\t" \
/* YUV2RGB core
* Conversion is performed in usual way:
* R = Y' * Ycoef + Vred * V'
* G = Y' * Ycoef + Vgreen * V' + Ugreen * U'
* B = Y' * Ycoef + Ublue * U'
*
* where X' = X * 8 - Xoffset (multiplication is performed to increase
* precision a bit).
* Since it operates in YUV420 colorspace, Y component is additionally
* split into Y1 and Y2 for even and odd pixels.
*
* Input:
* mm0 - U (4 elems), mm1 - V (4 elems), mm6 - Y (8 elems), mm4 - zero register
* Output:
* mm1 - R, mm2 - G, mm0 - B
*/
#define YUV2RGB \
/* convert Y, U, V into Y1', Y2', U', V' */ \
"movq %%mm6, %%mm7\n\t" \
"punpcklbw %%mm4, %%mm0\n\t" \
"punpcklbw %%mm4, %%mm1\n\t" \
"pand "MANGLE(mmx_00ffw)", %%mm6\n\t" \
"psrlw $8, %%mm7\n\t" \
"psllw $3, %%mm0\n\t" \
"psllw $3, %%mm1\n\t" \
"psllw $3, %%mm6\n\t" \
"psllw $3, %%mm7\n\t" \
"psubsw "U_OFFSET"(%4), %%mm0\n\t" \
"psubsw "V_OFFSET"(%4), %%mm1\n\t" \
"psubw "Y_OFFSET"(%4), %%mm6\n\t" \
"psubw "Y_OFFSET"(%4), %%mm7\n\t" \
\
/* multiply by coefficients */ \
"movq %%mm0, %%mm2\n\t" \
"movq %%mm1, %%mm3\n\t" \
"pmulhw "UG_COEFF"(%4), %%mm2\n\t" \
"pmulhw "VG_COEFF"(%4), %%mm3\n\t" \
"pmulhw "Y_COEFF" (%4), %%mm6\n\t" \
"pmulhw "Y_COEFF" (%4), %%mm7\n\t" \
"pmulhw "UB_COEFF"(%4), %%mm0\n\t" \
"pmulhw "VR_COEFF"(%4), %%mm1\n\t" \
"paddsw %%mm3, %%mm2\n\t" \
/* now: mm0 = UB, mm1 = VR, mm2 = CG */ \
/* mm6 = Y1, mm7 = Y2 */ \
\
/* produce RGB */ \
"movq %%mm7, %%mm3\n\t" \
"movq %%mm7, %%mm5\n\t" \
"paddsw %%mm0, %%mm3\n\t" \
"paddsw %%mm1, %%mm5\n\t" \
"paddsw %%mm2, %%mm7\n\t" \
"paddsw %%mm6, %%mm0\n\t" \
"paddsw %%mm6, %%mm1\n\t" \
"paddsw %%mm6, %%mm2\n\t" \
\
/* pack and interleave even/odd pixels */ \
"packuswb %%mm0, %%mm0\n\t" \
"packuswb %%mm1, %%mm1\n\t" \
"packuswb %%mm2, %%mm2\n\t" \
"packuswb %%mm3, %%mm3\n\t" \
"packuswb %%mm5, %%mm5\n\t" \
"packuswb %%mm7, %%mm7\n\t" \
"punpcklbw %%mm3, %%mm0\n\t" \
"punpcklbw %%mm5, %%mm1\n\t" \
"punpcklbw %%mm7, %%mm2\n\t" \
#define YUV2RGB_ENDLOOP(depth) \
"movq 8 (%5, %0, 2), %%mm6\n\t" \
"movd 4 (%3, %0), %%mm1\n\t" \
"movd 4 (%2, %0), %%mm0\n\t" \
"add $"AV_STRINGIFY(depth * 8)", %1\n\t" \
"add $4, %0\n\t" \
"js 1b\n\t" \
#define YUV2RGB_OPERANDS \
: "+r" (index), "+r" (image) \
: "r" (pu - index), "r" (pv - index), "r"(&c->redDither), \
"r" (py - 2*index) \
); \
} \
#define YUV2RGB_OPERANDS_ALPHA \
: "+r" (index), "+r" (image) \
: "r" (pu - index), "r" (pv - index), "r"(&c->redDither), \
"r" (py - 2*index), "r" (pa - 2*index) \
); \
} \
#define YUV2RGB_ENDFUNC \
__asm__ volatile (SFENCE"\n\t"EMMS); \
return srcSliceH; \
#define RGB_PACK16(gmask, gshift, rshift) \
"pand "MANGLE(mmx_redmask)", %%mm0\n\t" \
"pand "MANGLE(mmx_redmask)", %%mm1\n\t" \
"psrlw $3, %%mm0\n\t" \
"pand "MANGLE(gmask)", %%mm2\n\t" \
"movq %%mm0, %%mm5\n\t" \
"movq %%mm1, %%mm6\n\t" \
"movq %%mm2, %%mm7\n\t" \
"punpcklbw %%mm4, %%mm0\n\t" \
"punpcklbw %%mm4, %%mm1\n\t" \
"punpcklbw %%mm4, %%mm2\n\t" \
"punpckhbw %%mm4, %%mm5\n\t" \
"punpckhbw %%mm4, %%mm6\n\t" \
"punpckhbw %%mm4, %%mm7\n\t" \
"psllw $"rshift", %%mm1\n\t" \
"psllw $"rshift", %%mm6\n\t" \
"psllw $"gshift", %%mm2\n\t" \
"psllw $"gshift", %%mm7\n\t" \
"por %%mm1, %%mm0\n\t" \
"por %%mm6, %%mm5\n\t" \
"por %%mm2, %%mm0\n\t" \
"por %%mm7, %%mm5\n\t" \
MOVNTQ " %%mm0, (%1)\n\t" \
MOVNTQ " %%mm5, 8(%1)\n\t" \
#define DITHER_RGB \
"paddusb "BLUE_DITHER"(%4), %%mm0\n\t" \
"paddusb "GREEN_DITHER"(%4), %%mm2\n\t" \
"paddusb "RED_DITHER"(%4), %%mm1\n\t" \
static inline int RENAME(yuv420_rgb15)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, h_size;
YUV2RGB_LOOP(2)
#ifdef DITHER1XBPP
c->blueDither = ff_dither8[y & 1];
c->greenDither = ff_dither8[y & 1];
c->redDither = ff_dither8[(y + 1) & 1];
#endif
YUV2RGB_INITIAL_LOAD
YUV2RGB
#ifdef DITHER1XBPP
DITHER_RGB
#endif
RGB_PACK16(mmx_redmask, "2", "7")
YUV2RGB_ENDLOOP(2)
YUV2RGB_OPERANDS
YUV2RGB_ENDFUNC
}
static inline int RENAME(yuv420_rgb16)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, h_size;
YUV2RGB_LOOP(2)
#ifdef DITHER1XBPP
c->blueDither = ff_dither8[y & 1];
c->greenDither = ff_dither4[y & 1];
c->redDither = ff_dither8[(y + 1) & 1];
#endif
YUV2RGB_INITIAL_LOAD
YUV2RGB
#ifdef DITHER1XBPP
DITHER_RGB
#endif
RGB_PACK16(mmx_grnmask, "3", "8")
YUV2RGB_ENDLOOP(2)
YUV2RGB_OPERANDS
YUV2RGB_ENDFUNC
}
#define RGB_PACK24(red, blue) \
/* generate first packed RGB octet */ \
"movq %%mm2, %%mm5\n\t" \
"movq %%mm"blue", %%mm6\n\t" \
"movq %%mm"red", %%mm7\n\t" \
"punpcklbw %%mm5, %%mm6\n\t" \
"punpcklbw %%mm4, %%mm7\n\t" \
"movq %%mm6, %%mm3\n\t" \
"punpcklwd %%mm7, %%mm6\n\t" \
"psrlq $32, %%mm3\n\t" \
"movq %%mm6, %%mm5\n\t" \
"psllq $40, %%mm6\n\t" \
"psllq $48, %%mm3\n\t" \
"psrlq $32, %%mm5\n\t" \
"psrlq $40, %%mm6\n\t" \
"psllq $24, %%mm5\n\t" \
"por %%mm3, %%mm6\n\t" \
"por %%mm5, %%mm6\n\t" \
MOVNTQ " %%mm6, (%1)\n\t" \
\
/* generate second packed RGB octet */ \
"movq %%mm"red", %%mm7\n\t" \
"movq %%mm2, %%mm5\n\t" \
"movq %%mm"blue", %%mm6\n\t" \
"punpcklbw %%mm4, %%mm7\n\t" \
"punpcklbw %%mm5, %%mm6\n\t" \
"movq %%mm7, %%mm3\n\t" \
"punpckhwd %%mm7, %%mm6\n\t" \
"psllq $16, %%mm3\n\t" \
"psrlq $32, %%mm6\n\t" \
"psrlq $48, %%mm3\n\t" \
"psllq $8, %%mm6\n\t" \
"movq %%mm"red", %%mm7\n\t" \
"por %%mm6, %%mm3\n\t" \
"movq %%mm"blue", %%mm6\n\t" \
"movq %%mm2, %%mm5\n\t" \
"punpckhbw %%mm4, %%mm7\n\t" \
"punpckhbw %%mm5, %%mm6\n\t" \
"movq %%mm6, %%mm5\n\t" \
"punpcklwd %%mm7, %%mm6\n\t" \
"psrlq $16, %%mm5\n\t" \
"psllq $56, %%mm5\n\t" \
"por %%mm5, %%mm3\n\t" \
"psllq $32, %%mm6\n\t" \
"por %%mm6, %%mm3\n\t" \
MOVNTQ " %%mm3, 8(%1)\n\t" \
\
/* generate third packed RGB octet */ \
"movq %%mm"red", %%mm7\n\t" \
"movq %%mm2, %%mm5\n\t" \
"movq %%mm2, %%mm3\n\t" \
"movq %%mm"blue", %%mm6\n\t" \
"punpckhbw %%mm"red", %%mm3\n\t" \
"punpckhbw %%mm4, %%mm7\n\t" \
"psllq $32, %%mm3\n\t" \
"punpckhbw %%mm5, %%mm6\n\t" \
"psrlq $48, %%mm3\n\t" \
"punpckhwd %%mm7, %%mm6\n\t" \
"movq %%mm6, %%mm7\n\t" \
"psrlq $32, %%mm6\n\t" \
"psllq $32, %%mm7\n\t" \
"psllq $40, %%mm6\n\t" \
"psrlq $16, %%mm7\n\t" \
"por %%mm6, %%mm3\n\t" \
"por %%mm7, %%mm3\n\t" \
MOVNTQ " %%mm3, 16(%1)\n\t" \
static inline int RENAME(yuv420_rgb24)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, h_size;
YUV2RGB_LOOP(3)
YUV2RGB_INITIAL_LOAD
YUV2RGB
RGB_PACK24(REG_BLUE, REG_RED)
YUV2RGB_ENDLOOP(3)
YUV2RGB_OPERANDS
YUV2RGB_ENDFUNC
}
static inline int RENAME(yuv420_bgr24)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, h_size;
YUV2RGB_LOOP(3)
YUV2RGB_INITIAL_LOAD
YUV2RGB
RGB_PACK24(REG_RED, REG_BLUE)
YUV2RGB_ENDLOOP(3)
YUV2RGB_OPERANDS
YUV2RGB_ENDFUNC
}
#define SET_EMPTY_ALPHA \
"pcmpeqd %%mm"REG_ALPHA", %%mm"REG_ALPHA"\n\t" /* set alpha to 0xFF */ \
#define LOAD_ALPHA \
"movq (%6, %0, 2), %%mm"REG_ALPHA"\n\t" \
#define RGB_PACK32(red, green, blue, alpha) \
"movq %%mm"blue", %%mm5\n\t" \
"movq %%mm"red", %%mm6\n\t" \
"punpckhbw %%mm"green", %%mm5\n\t" \
"punpcklbw %%mm"green", %%mm"blue"\n\t" \
"punpckhbw %%mm"alpha", %%mm6\n\t" \
"punpcklbw %%mm"alpha", %%mm"red"\n\t" \
"movq %%mm"blue", %%mm"green"\n\t" \
"movq %%mm5, %%mm"alpha"\n\t" \
"punpcklwd %%mm"red", %%mm"blue"\n\t" \
"punpckhwd %%mm"red", %%mm"green"\n\t" \
"punpcklwd %%mm6, %%mm5\n\t" \
"punpckhwd %%mm6, %%mm"alpha"\n\t" \
MOVNTQ " %%mm"blue", 0(%1)\n\t" \
MOVNTQ " %%mm"green", 8(%1)\n\t" \
MOVNTQ " %%mm5, 16(%1)\n\t" \
MOVNTQ " %%mm"alpha", 24(%1)\n\t" \
static inline int RENAME(yuv420_rgb32)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, h_size;
YUV2RGB_LOOP(4)
YUV2RGB_INITIAL_LOAD
YUV2RGB
SET_EMPTY_ALPHA
RGB_PACK32(REG_RED, REG_GREEN, REG_BLUE, REG_ALPHA)
YUV2RGB_ENDLOOP(4)
YUV2RGB_OPERANDS
YUV2RGB_ENDFUNC
}
static inline int RENAME(yuva420_rgb32)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
#if HAVE_7REGS
int y, h_size;
YUV2RGB_LOOP(4)
const uint8_t *pa = src[3] + y * srcStride[3];
YUV2RGB_INITIAL_LOAD
YUV2RGB
LOAD_ALPHA
RGB_PACK32(REG_RED, REG_GREEN, REG_BLUE, REG_ALPHA)
YUV2RGB_ENDLOOP(4)
YUV2RGB_OPERANDS_ALPHA
YUV2RGB_ENDFUNC
#endif
}
static inline int RENAME(yuv420_bgr32)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, h_size;
YUV2RGB_LOOP(4)
YUV2RGB_INITIAL_LOAD
YUV2RGB
SET_EMPTY_ALPHA
RGB_PACK32(REG_BLUE, REG_GREEN, REG_RED, REG_ALPHA)
YUV2RGB_ENDLOOP(4)
YUV2RGB_OPERANDS
YUV2RGB_ENDFUNC
}
static inline int RENAME(yuva420_bgr32)(SwsContext *c, const uint8_t *src[],
int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
#if HAVE_7REGS
int y, h_size;
YUV2RGB_LOOP(4)
const uint8_t *pa = src[3] + y * srcStride[3];
YUV2RGB_INITIAL_LOAD
YUV2RGB
LOAD_ALPHA
RGB_PACK32(REG_BLUE, REG_GREEN, REG_RED, REG_ALPHA)
YUV2RGB_ENDLOOP(4)
YUV2RGB_OPERANDS_ALPHA
YUV2RGB_ENDFUNC
#endif
}

View File

@ -543,7 +543,7 @@ CLOSEYUV2RGBFUNC(1)
SwsFunc ff_yuv2rgb_get_func_ptr(SwsContext *c)
{
SwsFunc t = NULL;
#if HAVE_MMX && CONFIG_GPL
#if HAVE_MMX
t = ff_yuv2rgb_init_mmx(c);
#endif
#if HAVE_VIS