rv40: x86 SIMD for biweight

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>

1 files changed, 19 insertions, 0 deletions

diff --git a/libavcodec/x86/rv40dsp_init.c b/libavcodec/x86/rv40dsp_init.c
index b429de7293..3d6c6f0fa0 100644
--- a/libavcodec/x86/rv40dsp_init.c
+++ b/libavcodec/x86/rv40dsp_init.c
@@ -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
 }