use h264 MC functions for 2xX Xx2 blocks in snow too

Originally committed as revision 4824 to svn://svn.ffmpeg.org/ffmpeg/trunk

3 files changed, 86 insertions, 8 deletions

diff --git a/libavcodec/dsputil.c b/libavcodec/dsputil.c
index 3587ebf8b2..c7ae725c6f 100644
--- a/libavcodec/dsputil.c
+++ b/libavcodec/dsputil.c
@@ -1487,6 +1487,17 @@ H264_CHROMA_MC(avg_       , op_avg)
 #undef op_avg
 #undef op_put
 
+static inline void copy_block2(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
+{
+    int i;
+    for(i=0; i<h; i++)
+    {
+        ST16(dst   , LD16(src   ));
+        dst+=dstStride;
+        src+=srcStride;
+    }
+}
+
 static inline void copy_block4(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
 {
     int i;
@@ -2052,6 +2063,68 @@ QPEL_MC(0, avg_       , _       , op_avg)
 
 #if 1
 #define H264_LOWPASS(OPNAME, OP, OP2) \
+static void OPNAME ## h264_qpel2_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
+    const int h=2;\
+    uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+    int i;\
+    for(i=0; i<h; i++)\
+    {\
+        OP(dst[0], (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3]));\
+        OP(dst[1], (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4]));\
+        dst+=dstStride;\
+        src+=srcStride;\
+    }\
+}\
+\
+static void OPNAME ## h264_qpel2_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
+    const int w=2;\
+    uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+    int i;\
+    for(i=0; i<w; i++)\
+    {\
+        const int srcB= src[-2*srcStride];\
+        const int srcA= src[-1*srcStride];\
+        const int src0= src[0 *srcStride];\
+        const int src1= src[1 *srcStride];\
+        const int src2= src[2 *srcStride];\
+        const int src3= src[3 *srcStride];\
+        const int src4= src[4 *srcStride];\
+        OP(dst[0*dstStride], (src0+src1)*20 - (srcA+src2)*5 + (srcB+src3));\
+        OP(dst[1*dstStride], (src1+src2)*20 - (src0+src3)*5 + (srcA+src4));\
+        dst++;\
+        src++;\
+    }\
+}\
+\
+static void OPNAME ## h264_qpel2_hv_lowpass(uint8_t *dst, int16_t *tmp, uint8_t *src, int dstStride, int tmpStride, int srcStride){\
+    const int h=2;\
+    const int w=2;\
+    uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+    int i;\
+    src -= 2*srcStride;\
+    for(i=0; i<h+5; i++)\
+    {\
+        tmp[0]= (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3]);\
+        tmp[1]= (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4]);\
+        tmp+=tmpStride;\
+        src+=srcStride;\
+    }\
+    tmp -= tmpStride*(h+5-2);\
+    for(i=0; i<w; i++)\
+    {\
+        const int tmpB= tmp[-2*tmpStride];\
+        const int tmpA= tmp[-1*tmpStride];\
+        const int tmp0= tmp[0 *tmpStride];\
+        const int tmp1= tmp[1 *tmpStride];\
+        const int tmp2= tmp[2 *tmpStride];\
+        const int tmp3= tmp[3 *tmpStride];\
+        const int tmp4= tmp[4 *tmpStride];\
+        OP2(dst[0*dstStride], (tmp0+tmp1)*20 - (tmpA+tmp2)*5 + (tmpB+tmp3));\
+        OP2(dst[1*dstStride], (tmp1+tmp2)*20 - (tmp0+tmp3)*5 + (tmpA+tmp4));\
+        dst++;\
+        tmp++;\
+    }\
+}\
 static void OPNAME ## h264_qpel4_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
     const int h=4;\
     uint8_t *cm = cropTbl + MAX_NEG_CROP;\
@@ -2398,6 +2471,7 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc32_c(uint8_t *dst, uint8_t *src, i
 
 H264_LOWPASS(put_       , op_put, op2_put)
 H264_LOWPASS(avg_       , op_avg, op2_avg)
+H264_MC(put_, 2)
 H264_MC(put_, 4)
 H264_MC(put_, 8)
 H264_MC(put_, 16)
@@ -3879,6 +3953,7 @@ void dsputil_init(DSPContext* c, AVCodecContext *avctx)
     dspfunc(put_h264_qpel, 0, 16);
     dspfunc(put_h264_qpel, 1, 8);
     dspfunc(put_h264_qpel, 2, 4);
+    dspfunc(put_h264_qpel, 3, 2);
     dspfunc(avg_h264_qpel, 0, 16);
     dspfunc(avg_h264_qpel, 1, 8);
     dspfunc(avg_h264_qpel, 2, 4);
diff --git a/libavcodec/dsputil.h b/libavcodec/dsputil.h
index 969fdbcb69..fdcea72129 100644
--- a/libavcodec/dsputil.h
+++ b/libavcodec/dsputil.h
@@ -253,8 +253,8 @@ typedef struct DSPContext {
     h264_chroma_mc_func put_h264_chroma_pixels_tab[3];
     h264_chroma_mc_func avg_h264_chroma_pixels_tab[3];
 
-    qpel_mc_func put_h264_qpel_pixels_tab[3][16];
-    qpel_mc_func avg_h264_qpel_pixels_tab[3][16];
+    qpel_mc_func put_h264_qpel_pixels_tab[4][16];
+    qpel_mc_func avg_h264_qpel_pixels_tab[4][16];
 
     h264_weight_func weight_h264_pixels_tab[10];
     h264_biweight_func biweight_h264_pixels_tab[10];
@@ -510,6 +510,7 @@ struct unaligned_16 { uint16_t l; } __attribute__((packed));
 #define LD32(a) (((const struct unaligned_32 *) (a))->l)
 #define LD64(a) (((const struct unaligned_64 *) (a))->l)
 
+#define ST16(a, b) (((struct unaligned_16 *) (a))->l) = (b)
 #define ST32(a, b) (((struct unaligned_32 *) (a))->l) = (b)
 
 #else /* __GNUC__ */
diff --git a/libavcodec/snow.c b/libavcodec/snow.c
index 0aa56dba61..20d24f4fcb 100644
--- a/libavcodec/snow.c
+++ b/libavcodec/snow.c
@@ -2500,6 +2500,7 @@ static void pred_block(SnowContext *s, uint8_t *dst, uint8_t *src, uint8_t *tmp,
         int my= block->my*scale;
         const int dx= mx&15;
         const int dy= my&15;
+        const int tab_index= 3 - (b_w>>2) + (b_w>>4);
         sx += (mx>>4) - 2;
         sy += (my>>4) - 2;
         src += sx + sy*stride;
@@ -2511,17 +2512,18 @@ static void pred_block(SnowContext *s, uint8_t *dst, uint8_t *src, uint8_t *tmp,
         assert(b_w == b_h || 2*b_w == b_h || b_w == 2*b_h);
         assert(!(b_w&(b_w-1)));
         assert(b_w>1 && b_h>1);
-        if((dx&3) || (dy&3) || b_w==2 || b_h==2)
+        assert(tab_index>=0 && tab_index<4);
+        if((dx&3) || (dy&3))
             mc_block(dst, src, tmp, stride, b_w, b_h, dx, dy);
         else if(b_w==b_h)
-            s->dsp.put_h264_qpel_pixels_tab[2-(b_w>>3)][dy+(dx>>2)](dst,src + 2 + 2*stride,stride);
+            s->dsp.put_h264_qpel_pixels_tab[tab_index  ][dy+(dx>>2)](dst,src + 2 + 2*stride,stride);
         else if(b_w==2*b_h){
-            s->dsp.put_h264_qpel_pixels_tab[2-(b_h>>3)][dy+(dx>>2)](dst    ,src + 2       + 2*stride,stride);
-            s->dsp.put_h264_qpel_pixels_tab[2-(b_h>>3)][dy+(dx>>2)](dst+b_h,src + 2 + b_h + 2*stride,stride);
+            s->dsp.put_h264_qpel_pixels_tab[tab_index+1][dy+(dx>>2)](dst    ,src + 2       + 2*stride,stride);
+            s->dsp.put_h264_qpel_pixels_tab[tab_index+1][dy+(dx>>2)](dst+b_h,src + 2 + b_h + 2*stride,stride);
         }else{
             assert(2*b_w==b_h);
-            s->dsp.put_h264_qpel_pixels_tab[2-(b_w>>3)][dy+(dx>>2)](dst           ,src + 2 + 2*stride           ,stride);
-            s->dsp.put_h264_qpel_pixels_tab[2-(b_w>>3)][dy+(dx>>2)](dst+b_w*stride,src + 2 + 2*stride+b_w*stride,stride);
+            s->dsp.put_h264_qpel_pixels_tab[tab_index  ][dy+(dx>>2)](dst           ,src + 2 + 2*stride           ,stride);
+            s->dsp.put_h264_qpel_pixels_tab[tab_index  ][dy+(dx>>2)](dst+b_w*stride,src + 2 + 2*stride+b_w*stride,stride);
         }
     }
 }