generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

rv40: x86 SIMD for biweight

Browse Source 
Provide MMX, SSE2 and SSSE3 versions, with a fast-path when the weights are
multiples of 512 (which is often the case when the values round up nicely).

*_TIMER report for the 16x16 and 8x8 cases:
C:
9015 decicycles in 16, 524257 runs, 31 skips
2656 decicycles in 8, 524271 runs, 17 skips
MMX:
4156 decicycles in 16, 262090 runs, 54 skips
1206 decicycles in 8, 262131 runs, 13 skips
MMX on fast-path:
2760 decicycles in 16, 524222 runs, 66 skips
995 decicycles in 8, 524252 runs, 36 skips
SSE2:
2163 decicycles in 16, 262131 runs, 13 skips
832 decicycles in 8, 262137 runs, 7 skips
SSE2 with fast path:
1783 decicycles in 16, 524276 runs, 12 skips
711 decicycles in 8, 524283 runs, 5 skips
SSSE3:
2117 decicycles in 16, 262136 runs, 8 skips
814 decicycles in 8, 262143 runs, 1 skips
SSSE3 with fast path:
1315 decicycles in 16, 524285 runs, 3 skips
578 decicycles in 8, 524286 runs, 2 skips

This means around a 4% speedup for some sequences.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
This commit is contained in:
Christophe Gisquet
2012-01-12 00:11:15 +01:00
committed by Diego Biurrun
parent 91bafb52ae
commit e5c9de2ab7
3 changed files with 228 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
libavcodec/x86/rv40dsp_init.c
View File

@@ -40,6 +40,15 @@ void ff_avg_rv40_chroma_mc4_mmx2 (uint8_t *dst, uint8_t *src,
void ff_avg_rv40_chroma_mc4_3dnow(uint8_t *dst, uint8_t *src,
int stride, int h, int x, int y);
#define DECLARE_WEIGHT(opt) \
void ff_rv40_weight_func_16_##opt(uint8_t *dst, uint8_t *src1, uint8_t *src2, \
int w1, int w2, int stride); \
void ff_rv40_weight_func_8_##opt (uint8_t *dst, uint8_t *src1, uint8_t *src2, \
int w1, int w2, int stride);
DECLARE_WEIGHT(mmx)
DECLARE_WEIGHT(sse2)
DECLARE_WEIGHT(ssse3)
void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext *dsp)
{
#if HAVE_YASM
@@ -48,6 +57,8 @@ void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext *dsp)
if (mm_flags & AV_CPU_FLAG_MMX) {
c->put_chroma_pixels_tab[0] = ff_put_rv40_chroma_mc8_mmx;
c->put_chroma_pixels_tab[1] = ff_put_rv40_chroma_mc4_mmx;
c->rv40_weight_pixels_tab[0] = ff_rv40_weight_func_16_mmx;
c->rv40_weight_pixels_tab[1] = ff_rv40_weight_func_8_mmx;
}
if (mm_flags & AV_CPU_FLAG_MMX2) {
c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_mmx2;
@@ -56,5 +67,13 @@ void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext *dsp)
c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_3dnow;
c->avg_chroma_pixels_tab[1] = ff_avg_rv40_chroma_mc4_3dnow;
}
if (mm_flags & AV_CPU_FLAG_SSE2) {
c->rv40_weight_pixels_tab[0] = ff_rv40_weight_func_16_sse2;
c->rv40_weight_pixels_tab[1] = ff_rv40_weight_func_8_sse2;
}
if (mm_flags & AV_CPU_FLAG_SSSE3) {
c->rv40_weight_pixels_tab[0] = ff_rv40_weight_func_16_ssse3;
c->rv40_weight_pixels_tab[1] = ff_rv40_weight_func_8_ssse3;
}
#endif
}