VC1: add and use avg_no_rnd chroma MC functions

Originally committed as revision 18518 to svn://svn.ffmpeg.org/ffmpeg/trunk

libavcodec/dsputil.c

@@ -1703,6 +1703,30 @@ static void put_no_rnd_vc1_chroma_mc8_c(uint8_t *dst/*align 8*/, uint8_t *src/*a
     }
 }
 
+static void avg_no_rnd_vc1_chroma_mc8_c(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y){
+    const int A=(8-x)*(8-y);
+    const int B=(  x)*(8-y);
+    const int C=(8-x)*(  y);
+    const int D=(  x)*(  y);
+    int i;
+
+    assert(x<8 && y<8 && x>=0 && y>=0);
+
+    for(i=0; i<h; i++)
+    {
+        dst[0] = avg2(dst[0], ((A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1] + 32 - 4) >> 6));
+        dst[1] = avg2(dst[1], ((A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2] + 32 - 4) >> 6));
+        dst[2] = avg2(dst[2], ((A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3] + 32 - 4) >> 6));
+        dst[3] = avg2(dst[3], ((A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4] + 32 - 4) >> 6));
+        dst[4] = avg2(dst[4], ((A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5] + 32 - 4) >> 6));
+        dst[5] = avg2(dst[5], ((A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6] + 32 - 4) >> 6));
+        dst[6] = avg2(dst[6], ((A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7] + 32 - 4) >> 6));
+        dst[7] = avg2(dst[7], ((A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8] + 32 - 4) >> 6));
+        dst+= stride;
+        src+= stride;
+    }
+}
+
 #define QPEL_MC(r, OPNAME, RND, OP) \
 static void OPNAME ## mpeg4_qpel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h){\
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
@@ -4490,6 +4514,7 @@ void dsputil_init(DSPContext* c, AVCodecContext *avctx)
     c->avg_h264_chroma_pixels_tab[1]= avg_h264_chroma_mc4_c;
     c->avg_h264_chroma_pixels_tab[2]= avg_h264_chroma_mc2_c;
     c->put_no_rnd_vc1_chroma_pixels_tab[0]= put_no_rnd_vc1_chroma_mc8_c;
+    c->avg_no_rnd_vc1_chroma_pixels_tab[0]= avg_no_rnd_vc1_chroma_mc8_c;
 
     c->weight_h264_pixels_tab[0]= weight_h264_pixels16x16_c;
     c->weight_h264_pixels_tab[1]= weight_h264_pixels16x8_c;

libavcodec/dsputil.h

@@ -319,6 +319,7 @@ typedef struct DSPContext {
     h264_chroma_mc_func avg_h264_chroma_pixels_tab[3];
     /* This is really one func used in VC-1 decoding */
     h264_chroma_mc_func put_no_rnd_vc1_chroma_pixels_tab[3];
+    h264_chroma_mc_func avg_no_rnd_vc1_chroma_pixels_tab[3];
 
     qpel_mc_func put_h264_qpel_pixels_tab[4][16];
     qpel_mc_func avg_h264_qpel_pixels_tab[4][16];

libavcodec/vc1.c

@@ -1925,8 +1925,13 @@ static void vc1_interp_mc(VC1Context *v)
     uvdxy = ((uvmy & 3) << 2) | (uvmx & 3);
     uvmx = (uvmx&3)<<1;
     uvmy = (uvmy&3)<<1;
-    dsp->avg_h264_chroma_pixels_tab[0](s->dest[1], srcU, s->uvlinesize, 8, uvmx, uvmy);
+    if(!v->rnd){
-    dsp->avg_h264_chroma_pixels_tab[0](s->dest[2], srcV, s->uvlinesize, 8, uvmx, uvmy);
+        dsp->avg_h264_chroma_pixels_tab[0](s->dest[1], srcU, s->uvlinesize, 8, uvmx, uvmy);
+        dsp->avg_h264_chroma_pixels_tab[0](s->dest[2], srcV, s->uvlinesize, 8, uvmx, uvmy);
+    }else{
+        dsp->avg_no_rnd_vc1_chroma_pixels_tab[0](s->dest[1], srcU, s->uvlinesize, 8, uvmx, uvmy);
+        dsp->avg_no_rnd_vc1_chroma_pixels_tab[0](s->dest[2], srcV, s->uvlinesize, 8, uvmx, uvmy);
+    }
 }
 
 static av_always_inline int scale_mv(int value, int bfrac, int inv, int qs)

libavcodec/x86/dsputil_mmx.c

@@ -2790,6 +2790,8 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             c->avg_rv40_chroma_pixels_tab[0]= avg_rv40_chroma_mc8_mmx2;
             c->avg_rv40_chroma_pixels_tab[1]= avg_rv40_chroma_mc4_mmx2;
 
+            c->avg_no_rnd_vc1_chroma_pixels_tab[0]= avg_vc1_chroma_mc8_mmx2_nornd;
+
             c->avg_h264_chroma_pixels_tab[0]= avg_h264_chroma_mc8_mmx2_rnd;
             c->avg_h264_chroma_pixels_tab[1]= avg_h264_chroma_mc4_mmx2;
             c->avg_h264_chroma_pixels_tab[2]= avg_h264_chroma_mc2_mmx2;
@@ -2940,6 +2942,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             H264_QPEL_FUNCS(3, 2, ssse3);
             H264_QPEL_FUNCS(3, 3, ssse3);
             c->put_no_rnd_vc1_chroma_pixels_tab[0]= put_vc1_chroma_mc8_ssse3_nornd;
+            c->avg_no_rnd_vc1_chroma_pixels_tab[0]= avg_vc1_chroma_mc8_ssse3_nornd;
             c->put_h264_chroma_pixels_tab[0]= put_h264_chroma_mc8_ssse3_rnd;
             c->avg_h264_chroma_pixels_tab[0]= avg_h264_chroma_mc8_ssse3_rnd;
             c->put_h264_chroma_pixels_tab[1]= put_h264_chroma_mc4_ssse3;

libavcodec/x86/h264dsp_mmx.c

@@ -2129,6 +2129,10 @@ static void avg_h264_chroma_mc8_mmx2_rnd(uint8_t *dst/*align 8*/, uint8_t *src/*
 {
     avg_h264_chroma_generic_mc8_mmx2(dst, src, stride, h, x, y, h264_rnd_reg);
 }
+static void avg_vc1_chroma_mc8_mmx2_nornd(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y)
+{
+    avg_h264_chroma_generic_mc8_mmx2(dst, src, stride, h, x, y, h264_rnd_reg+2);
+}
 static void avg_h264_chroma_mc4_mmx2(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y)
 {
     avg_h264_chroma_generic_mc4_mmx2(dst, src, stride, h, x, y, h264_rnd_reg);
@@ -2191,6 +2195,10 @@ static void avg_h264_chroma_mc8_ssse3_rnd(uint8_t *dst/*align 8*/, uint8_t *src/
 {
     avg_h264_chroma_mc8_ssse3(dst, src, stride, h, x, y, 1);
 }
+static void avg_vc1_chroma_mc8_ssse3_nornd(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y)
+{
+    avg_h264_chroma_mc8_ssse3(dst, src, stride, h, x, y, 0);
+}
 #undef AVG_OP
 #undef H264_CHROMA_MC8_TMPL
 #undef H264_CHROMA_MC4_TMPL