Merge remote-tracking branch 'qatar/master'

* qatar/master:
  pixfmt: support more yuva formats
  swscale: support gray to 9bit and 10bit formats
  configure: rewrite print_config() function using awk
  FATE: fix (AD)PCM test dependencies broken in e519990
  Use ptrdiff_t instead of int for intra pred "stride" function parameter.
  x86: use PRED4x4/8x8/8x8L/16x16 macros to declare intrapred prototypes.

Conflicts:
	libavcodec/h264pred.c
	libavcodec/h264pred_template.c
	libavutil/pixfmt.h
	libswscale/swscale_unscaled.c
	tests/ref/lavfi/pixdesc
	tests/ref/lavfi/pixfmts_copy
	tests/ref/lavfi/pixfmts_null
	tests/ref/lavfi/pixfmts_scale
	tests/ref/lavfi/pixfmts_vflip

Merged-by: Michael Niedermayer <michaelni@gmx.at>

7 files changed, 501 insertions, 331 deletions

diff --git a/libavcodec/arm/h264pred_init_arm.c b/libavcodec/arm/h264pred_init_arm.c
index bd2ed0a5df..31714d73d5 100644
--- a/libavcodec/arm/h264pred_init_arm.c
+++ b/libavcodec/arm/h264pred_init_arm.c
@@ -23,25 +23,25 @@
 #include "libavutil/arm/cpu.h"
 #include "libavcodec/h264pred.h"
 
-void ff_pred16x16_vert_neon(uint8_t *src, int stride);
-void ff_pred16x16_hor_neon(uint8_t *src, int stride);
-void ff_pred16x16_plane_neon(uint8_t *src, int stride);
-void ff_pred16x16_dc_neon(uint8_t *src, int stride);
-void ff_pred16x16_128_dc_neon(uint8_t *src, int stride);
-void ff_pred16x16_left_dc_neon(uint8_t *src, int stride);
-void ff_pred16x16_top_dc_neon(uint8_t *src, int stride);
+void ff_pred16x16_vert_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred16x16_hor_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred16x16_plane_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred16x16_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred16x16_128_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred16x16_left_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred16x16_top_dc_neon(uint8_t *src, ptrdiff_t stride);
 
-void ff_pred8x8_vert_neon(uint8_t *src, int stride);
-void ff_pred8x8_hor_neon(uint8_t *src, int stride);
-void ff_pred8x8_plane_neon(uint8_t *src, int stride);
-void ff_pred8x8_dc_neon(uint8_t *src, int stride);
-void ff_pred8x8_128_dc_neon(uint8_t *src, int stride);
-void ff_pred8x8_left_dc_neon(uint8_t *src, int stride);
-void ff_pred8x8_top_dc_neon(uint8_t *src, int stride);
-void ff_pred8x8_l0t_dc_neon(uint8_t *src, int stride);
-void ff_pred8x8_0lt_dc_neon(uint8_t *src, int stride);
-void ff_pred8x8_l00_dc_neon(uint8_t *src, int stride);
-void ff_pred8x8_0l0_dc_neon(uint8_t *src, int stride);
+void ff_pred8x8_vert_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_hor_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_plane_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_128_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_left_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_top_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_l0t_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_0lt_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_l00_dc_neon(uint8_t *src, ptrdiff_t stride);
+void ff_pred8x8_0l0_dc_neon(uint8_t *src, ptrdiff_t stride);
 
 static void ff_h264_pred_init_neon(H264PredContext *h, int codec_id, const int bit_depth, const int chroma_format_idc)
 {
diff --git a/libavcodec/h264pred.c b/libavcodec/h264pred.c
index 5619efd109..f83924da34 100644
--- a/libavcodec/h264pred.c
+++ b/libavcodec/h264pred.c
@@ -48,7 +48,9 @@
 #include "h264pred_template.c"
 #undef BIT_DEPTH
 
-static void pred4x4_vertical_vp8_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_vertical_vp8_c(uint8_t *src, const uint8_t *topright,
+                                   ptrdiff_t stride)
+{
     const unsigned lt = src[-1-1*stride];
     LOAD_TOP_EDGE
     LOAD_TOP_RIGHT_EDGE
@@ -63,7 +65,9 @@ static void pred4x4_vertical_vp8_c(uint8_t *src, const uint8_t *topright, int st
     AV_WN32A(src+3*stride, v);
 }
 
-static void pred4x4_horizontal_vp8_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_horizontal_vp8_c(uint8_t *src, const uint8_t *topright,
+                                     ptrdiff_t stride)
+{
     const unsigned lt = src[-1-1*stride];
     LOAD_LEFT_EDGE
 
@@ -73,7 +77,9 @@ static void pred4x4_horizontal_vp8_c(uint8_t *src, const uint8_t *topright, int
     AV_WN32A(src+3*stride, ((l2 + 2*l3 + l3 + 2) >> 2)*0x01010101);
 }
 
-static void pred4x4_down_left_svq3_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_down_left_svq3_c(uint8_t *src, const uint8_t *topright,
+                                     ptrdiff_t stride)
+{
     LOAD_TOP_EDGE
     LOAD_LEFT_EDGE
 
@@ -95,7 +101,9 @@ static void pred4x4_down_left_svq3_c(uint8_t *src, const uint8_t *topright, int
     src[3+3*stride]=(l3 + t3)>>1;
 }
 
-static void pred4x4_down_left_rv40_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_down_left_rv40_c(uint8_t *src, const uint8_t *topright,
+                                     ptrdiff_t stride)
+{
     LOAD_TOP_EDGE
     LOAD_TOP_RIGHT_EDGE
     LOAD_LEFT_EDGE
@@ -119,7 +127,10 @@ static void pred4x4_down_left_rv40_c(uint8_t *src, const uint8_t *topright, int
     src[3+3*stride]=(t6 + t7 + 1 + l6 + l7 + 1)>>2;
 }
 
-static void pred4x4_down_left_rv40_nodown_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_down_left_rv40_nodown_c(uint8_t *src,
+                                            const uint8_t *topright,
+                                            ptrdiff_t stride)
+{
     LOAD_TOP_EDGE
     LOAD_TOP_RIGHT_EDGE
     LOAD_LEFT_EDGE
@@ -142,8 +153,11 @@ static void pred4x4_down_left_rv40_nodown_c(uint8_t *src, const uint8_t *toprigh
     src[3+3*stride]=(t6 + t7 + 1 + 2*l3 + 1)>>2;
 }
 
-static void pred4x4_vertical_left_rv40(uint8_t *src, const uint8_t *topright, int stride,
-                                       const int l0, const int l1, const int l2, const int l3, const int l4){
+static void pred4x4_vertical_left_rv40(uint8_t *src, const uint8_t *topright,
+                                       ptrdiff_t stride,
+                                       const int l0, const int l1, const int l2,
+                                       const int l3, const int l4)
+{
     LOAD_TOP_EDGE
     LOAD_TOP_RIGHT_EDGE
 
@@ -165,20 +179,27 @@ static void pred4x4_vertical_left_rv40(uint8_t *src, const uint8_t *topright, in
     src[3+3*stride]=(t4 + 2*t5 + t6 + 2)>>2;
 }
 
-static void pred4x4_vertical_left_rv40_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_vertical_left_rv40_c(uint8_t *src, const uint8_t *topright,
+                                         ptrdiff_t stride)
+{
     LOAD_LEFT_EDGE
     LOAD_DOWN_LEFT_EDGE
 
     pred4x4_vertical_left_rv40(src, topright, stride, l0, l1, l2, l3, l4);
 }
 
-static void pred4x4_vertical_left_rv40_nodown_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_vertical_left_rv40_nodown_c(uint8_t *src,
+                                                const uint8_t *topright,
+                                                ptrdiff_t stride)
+{
     LOAD_LEFT_EDGE
 
     pred4x4_vertical_left_rv40(src, topright, stride, l0, l1, l2, l3, l3);
 }
 
-static void pred4x4_vertical_left_vp8_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_vertical_left_vp8_c(uint8_t *src, const uint8_t *topright,
+                                        ptrdiff_t stride)
+{
     LOAD_TOP_EDGE
     LOAD_TOP_RIGHT_EDGE
 
@@ -200,7 +221,9 @@ static void pred4x4_vertical_left_vp8_c(uint8_t *src, const uint8_t *topright, i
     src[3+3*stride]=(t5 + 2*t6 + t7 + 2)>>2;
 }
 
-static void pred4x4_horizontal_up_rv40_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_horizontal_up_rv40_c(uint8_t *src, const uint8_t *topright,
+                                         ptrdiff_t stride)
+{
     LOAD_LEFT_EDGE
     LOAD_DOWN_LEFT_EDGE
     LOAD_TOP_EDGE
@@ -224,7 +247,10 @@ static void pred4x4_horizontal_up_rv40_c(uint8_t *src, const uint8_t *topright,
     src[3+3*stride]=(l4 + 2*l5 + l6 + 2)>>2;
 }
 
-static void pred4x4_horizontal_up_rv40_nodown_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_horizontal_up_rv40_nodown_c(uint8_t *src,
+                                                const uint8_t *topright,
+                                                ptrdiff_t stride)
+{
     LOAD_LEFT_EDGE
     LOAD_TOP_EDGE
     LOAD_TOP_RIGHT_EDGE
@@ -247,7 +273,9 @@ static void pred4x4_horizontal_up_rv40_nodown_c(uint8_t *src, const uint8_t *top
     src[3+3*stride]=l3;
 }
 
-static void pred4x4_tm_vp8_c(uint8_t *src, const uint8_t *topright, int stride){
+static void pred4x4_tm_vp8_c(uint8_t *src, const uint8_t *topright,
+                             ptrdiff_t stride)
+{
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP - src[-1-stride];
     uint8_t *top = src-stride;
     int y;
@@ -262,15 +290,18 @@ static void pred4x4_tm_vp8_c(uint8_t *src, const uint8_t *topright, int stride){
     }
 }
 
-static void pred16x16_plane_svq3_c(uint8_t *src, int stride){
+static void pred16x16_plane_svq3_c(uint8_t *src, ptrdiff_t stride)
+{
     pred16x16_plane_compat_8_c(src, stride, 1, 0);
 }
 
-static void pred16x16_plane_rv40_c(uint8_t *src, int stride){
+static void pred16x16_plane_rv40_c(uint8_t *src, ptrdiff_t stride)
+{
     pred16x16_plane_compat_8_c(src, stride, 0, 1);
 }
 
-static void pred16x16_tm_vp8_c(uint8_t *src, int stride){
+static void pred16x16_tm_vp8_c(uint8_t *src, ptrdiff_t stride)
+{
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP - src[-1-stride];
     uint8_t *top = src-stride;
     int y;
@@ -297,7 +328,8 @@ static void pred16x16_tm_vp8_c(uint8_t *src, int stride){
     }
 }
 
-static void pred8x8_left_dc_rv40_c(uint8_t *src, int stride){
+static void pred8x8_left_dc_rv40_c(uint8_t *src, ptrdiff_t stride)
+{
     int i;
     unsigned dc0;
 
@@ -312,7 +344,8 @@ static void pred8x8_left_dc_rv40_c(uint8_t *src, int stride){
     }
 }
 
-static void pred8x8_top_dc_rv40_c(uint8_t *src, int stride){
+static void pred8x8_top_dc_rv40_c(uint8_t *src, ptrdiff_t stride)
+{
     int i;
     unsigned dc0;
 
@@ -327,7 +360,8 @@ static void pred8x8_top_dc_rv40_c(uint8_t *src, int stride){
     }
 }
 
-static void pred8x8_dc_rv40_c(uint8_t *src, int stride){
+static void pred8x8_dc_rv40_c(uint8_t *src, ptrdiff_t stride)
+{
     int i;
     unsigned dc0 = 0;
 
@@ -348,7 +382,8 @@ static void pred8x8_dc_rv40_c(uint8_t *src, int stride){
     }
 }
 
-static void pred8x8_tm_vp8_c(uint8_t *src, int stride){
+static void pred8x8_tm_vp8_c(uint8_t *src, ptrdiff_t stride)
+{
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP - src[-1-stride];
     uint8_t *top = src-stride;
     int y;
@@ -370,7 +405,9 @@ static void pred8x8_tm_vp8_c(uint8_t *src, int stride){
 /**
  * Set the intra prediction function pointers.
  */
-void ff_h264_pred_init(H264PredContext *h, int codec_id, const int bit_depth, const int chroma_format_idc){
+void ff_h264_pred_init(H264PredContext *h, int codec_id, const int bit_depth,
+                       const int chroma_format_idc)
+{
 //    MpegEncContext * const s = &h->s;
 
 #undef FUNC
diff --git a/libavcodec/h264pred.h b/libavcodec/h264pred.h
index d68f39bf8c..33f39447df 100644
--- a/libavcodec/h264pred.h
+++ b/libavcodec/h264pred.h
@@ -90,21 +90,23 @@
  * Context for storing H.264 prediction functions
  */
 typedef struct H264PredContext {
-    void(*pred4x4[9 + 3 + 3])(uint8_t *src, const uint8_t *topright, int stride); //FIXME move to dsp?
-    void(*pred8x8l[9 + 3])(uint8_t *src, int topleft, int topright, int stride);
-    void(*pred8x8[4 + 3 + 4])(uint8_t *src, int stride);
-    void(*pred16x16[4 + 3 + 2])(uint8_t *src, int stride);
+    void(*pred4x4[9 + 3 + 3])(uint8_t *src, const uint8_t *topright,
+                              ptrdiff_t stride);
+    void(*pred8x8l[9 + 3])(uint8_t *src, int topleft, int topright,
+                           ptrdiff_t stride);
+    void(*pred8x8[4 + 3 + 4])(uint8_t *src, ptrdiff_t stride);
+    void(*pred16x16[4 + 3 + 2])(uint8_t *src, ptrdiff_t stride);
 
     void(*pred4x4_add[2])(uint8_t *pix /*align  4*/,
-                          const DCTELEM *block /*align 16*/, int stride);
+                          const DCTELEM *block /*align 16*/, ptrdiff_t stride);
     void(*pred8x8l_add[2])(uint8_t *pix /*align  8*/,
-                           const DCTELEM *block /*align 16*/, int stride);
+                           const DCTELEM *block /*align 16*/, ptrdiff_t stride);
     void(*pred8x8_add[3])(uint8_t *pix /*align  8*/,
                           const int *block_offset,
-                          const DCTELEM *block /*align 16*/, int stride);
+                          const DCTELEM *block /*align 16*/, ptrdiff_t stride);
     void(*pred16x16_add[3])(uint8_t *pix /*align 16*/,
                             const int *block_offset,
-                            const DCTELEM *block /*align 16*/, int stride);
+                            const DCTELEM *block /*align 16*/, ptrdiff_t stride);
 } H264PredContext;
 
 void ff_h264_pred_init(H264PredContext *h, int codec_id,
diff --git a/libavcodec/h264pred_template.c b/libavcodec/h264pred_template.c
index 3a1b1cf94e..f08fdf4ba4 100644
--- a/libavcodec/h264pred_template.c
+++ b/libavcodec/h264pred_template.c
@@ -29,7 +29,9 @@
 
 #include "bit_depth_template.c"
 
-static void FUNCC(pred4x4_vertical)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_vertical)(uint8_t *_src, const uint8_t *topright,
+                                    ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const pixel4 a= AV_RN4PA(src-stride);
@@ -40,7 +42,9 @@ static void FUNCC(pred4x4_vertical)(uint8_t *_src, const uint8_t *topright, int
     AV_WN4PA(src+3*stride, a);
 }
 
-static void FUNCC(pred4x4_horizontal)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_horizontal)(uint8_t *_src, const uint8_t *topright,
+                                      ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     AV_WN4PA(src+0*stride, PIXEL_SPLAT_X4(src[-1+0*stride]));
@@ -49,7 +53,9 @@ static void FUNCC(pred4x4_horizontal)(uint8_t *_src, const uint8_t *topright, in
     AV_WN4PA(src+3*stride, PIXEL_SPLAT_X4(src[-1+3*stride]));
 }
 
-static void FUNCC(pred4x4_dc)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_dc)(uint8_t *_src, const uint8_t *topright,
+                              ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const int dc= (  src[-stride] + src[1-stride] + src[2-stride] + src[3-stride]
@@ -62,7 +68,9 @@ static void FUNCC(pred4x4_dc)(uint8_t *_src, const uint8_t *topright, int _strid
     AV_WN4PA(src+3*stride, a);
 }
 
-static void FUNCC(pred4x4_left_dc)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_left_dc)(uint8_t *_src, const uint8_t *topright,
+                                   ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const int dc= (  src[-1+0*stride] + src[-1+1*stride] + src[-1+2*stride] + src[-1+3*stride] + 2) >>2;
@@ -74,7 +82,9 @@ static void FUNCC(pred4x4_left_dc)(uint8_t *_src, const uint8_t *topright, int _
     AV_WN4PA(src+3*stride, a);
 }
 
-static void FUNCC(pred4x4_top_dc)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_top_dc)(uint8_t *_src, const uint8_t *topright,
+                                  ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const int dc= (  src[-stride] + src[1-stride] + src[2-stride] + src[3-stride] + 2) >>2;
@@ -86,7 +96,9 @@ static void FUNCC(pred4x4_top_dc)(uint8_t *_src, const uint8_t *topright, int _s
     AV_WN4PA(src+3*stride, a);
 }
 
-static void FUNCC(pred4x4_128_dc)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_128_dc)(uint8_t *_src, const uint8_t *topright,
+                                  ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const pixel4 a = PIXEL_SPLAT_X4(1<<(BIT_DEPTH-1));
@@ -97,7 +109,9 @@ static void FUNCC(pred4x4_128_dc)(uint8_t *_src, const uint8_t *topright, int _s
     AV_WN4PA(src+3*stride, a);
 }
 
-static void FUNCC(pred4x4_127_dc)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_127_dc)(uint8_t *_src, const uint8_t *topright,
+                                  ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const pixel4 a = PIXEL_SPLAT_X4((1<<(BIT_DEPTH-1))-1);
@@ -108,7 +122,9 @@ static void FUNCC(pred4x4_127_dc)(uint8_t *_src, const uint8_t *topright, int _s
     AV_WN4PA(src+3*stride, a);
 }
 
-static void FUNCC(pred4x4_129_dc)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_129_dc)(uint8_t *_src, const uint8_t *topright,
+                                  ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const pixel4 a = PIXEL_SPLAT_X4((1<<(BIT_DEPTH-1))+1);
@@ -144,7 +160,9 @@ static void FUNCC(pred4x4_129_dc)(uint8_t *_src, const uint8_t *topright, int _s
     const unsigned av_unused t2 = src[ 2-1*stride];\
     const unsigned av_unused t3 = src[ 3-1*stride];\
 
-static void FUNCC(pred4x4_down_right)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_down_right)(uint8_t *_src, const uint8_t *topright,
+                                      ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const int lt= src[-1-1*stride];
@@ -169,7 +187,9 @@ static void FUNCC(pred4x4_down_right)(uint8_t *_src, const uint8_t *topright, in
     src[3+0*stride]=(t1 + 2*t2 + t3 + 2)>>2;
 }
 
-static void FUNCC(pred4x4_down_left)(uint8_t *_src, const uint8_t *_topright, int _stride){
+static void FUNCC(pred4x4_down_left)(uint8_t *_src, const uint8_t *_topright,
+                                     ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     const pixel *topright = (const pixel*)_topright;
     int stride = _stride>>(sizeof(pixel)-1);
@@ -195,7 +215,10 @@ static void FUNCC(pred4x4_down_left)(uint8_t *_src, const uint8_t *_topright, in
     src[3+3*stride]=(t6 + 3*t7 + 2)>>2;
 }
 
-static void FUNCC(pred4x4_vertical_right)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_vertical_right)(uint8_t *_src,
+                                          const uint8_t *topright,
+                                          ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const int lt= src[-1-1*stride];
@@ -220,7 +243,10 @@ static void FUNCC(pred4x4_vertical_right)(uint8_t *_src, const uint8_t *topright
     src[0+3*stride]=(l0 + 2*l1 + l2 + 2)>>2;
 }
 
-static void FUNCC(pred4x4_vertical_left)(uint8_t *_src, const uint8_t *_topright, int _stride){
+static void FUNCC(pred4x4_vertical_left)(uint8_t *_src,
+                                         const uint8_t *_topright,
+                                         ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     const pixel *topright = (const pixel*)_topright;
     int stride = _stride>>(sizeof(pixel)-1);
@@ -245,7 +271,9 @@ static void FUNCC(pred4x4_vertical_left)(uint8_t *_src, const uint8_t *_topright
     src[3+3*stride]=(t4 + 2*t5 + t6 + 2)>>2;
 }
 
-static void FUNCC(pred4x4_horizontal_up)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_horizontal_up)(uint8_t *_src, const uint8_t *topright,
+                                         ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     LOAD_LEFT_EDGE
@@ -268,7 +296,10 @@ static void FUNCC(pred4x4_horizontal_up)(uint8_t *_src, const uint8_t *topright,
     src[3+3*stride]=l3;
 }
 
-static void FUNCC(pred4x4_horizontal_down)(uint8_t *_src, const uint8_t *topright, int _stride){
+static void FUNCC(pred4x4_horizontal_down)(uint8_t *_src,
+                                           const uint8_t *topright,
+                                           ptrdiff_t _stride)
+{
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
     const int lt= src[-1-1*stride];
@@ -293,7 +324,8 @@ static void FUNCC(pred4x4_horizontal_down)(uint8_t *_src, const uint8_t *toprigh
     src[1+3*stride]=(l1 + 2*l2 + l3 + 2)>>2;
 }
 
-static void FUNCC(pred16x16_vertical)(uint8_t *_src, int _stride){
+static void FUNCC(pred16x16_vertical)(uint8_t *_src, ptrdiff_t _stride)
+{
     int i;
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
@@ -310,7 +342,8 @@ static void FUNCC(pred16x16_vertical)(uint8_t *_src, int _stride){
     }
 }
 
-static void FUNCC(pred16x16_horizontal)(uint8_t *_src, int stride){
+static void FUNCC(pred16x16_horizontal)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     pixel *src = (pixel*)_src;
     stride >>= sizeof(pixel)-1;
@@ -334,7 +367,8 @@ static void FUNCC(pred16x16_horizontal)(uint8_t *_src, int stride){
         src += stride;\
     }
 
-static void FUNCC(pred16x16_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred16x16_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i, dc=0;
     pixel *src = (pixel*)_src;
     pixel4 dcsplat;
@@ -352,7 +386,8 @@ static void FUNCC(pred16x16_dc)(uint8_t *_src, int stride){
     PREDICT_16x16_DC(dcsplat);
 }
 
-static void FUNCC(pred16x16_left_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred16x16_left_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i, dc=0;
     pixel *src = (pixel*)_src;
     pixel4 dcsplat;
@@ -366,7 +401,8 @@ static void FUNCC(pred16x16_left_dc)(uint8_t *_src, int stride){
     PREDICT_16x16_DC(dcsplat);
 }
 
-static void FUNCC(pred16x16_top_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred16x16_top_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i, dc=0;
     pixel *src = (pixel*)_src;
     pixel4 dcsplat;
@@ -381,7 +417,8 @@ static void FUNCC(pred16x16_top_dc)(uint8_t *_src, int stride){
 }
 
 #define PRED16x16_X(n, v) \
-static void FUNCC(pred16x16_##n##_dc)(uint8_t *_src, int stride){\
+static void FUNCC(pred16x16_##n##_dc)(uint8_t *_src, ptrdiff_t stride)\
+{\
     int i;\
     pixel *src = (pixel*)_src;\
     stride >>= sizeof(pixel)-1;\
@@ -392,12 +429,16 @@ PRED16x16_X(127, (1<<(BIT_DEPTH-1))-1)
 PRED16x16_X(128, (1<<(BIT_DEPTH-1))+0)
 PRED16x16_X(129, (1<<(BIT_DEPTH-1))+1)
 
-static inline void FUNCC(pred16x16_plane_compat)(uint8_t *p_src, int p_stride, const int svq3, const int rv40){
+static inline void FUNCC(pred16x16_plane_compat)(uint8_t *_src,
+                                                 ptrdiff_t _stride,
+                                                 const int svq3,
+                                                 const int rv40)
+{
   int i, j, k;
   int a;
   INIT_CLIP
-  pixel *src = (pixel*)p_src;
-  int stride = p_stride>>(sizeof(pixel)-1);
+  pixel *src = (pixel*)_src;
+  int stride = _stride>>(sizeof(pixel)-1);
   const pixel * const src0 = src +7-stride;
   const pixel *       src1 = src +8*stride-1;
   const pixel *       src2 = src1-2*stride;    // == src+6*stride-1;
@@ -437,11 +478,13 @@ static inline void FUNCC(pred16x16_plane_compat)(uint8_t *p_src, int p_stride, c
   }
 }
 
-static void FUNCC(pred16x16_plane)(uint8_t *src, int stride){
+static void FUNCC(pred16x16_plane)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred16x16_plane_compat)(src, stride, 0, 0);
 }
 
-static void FUNCC(pred8x8_vertical)(uint8_t *_src, int _stride){
+static void FUNCC(pred8x8_vertical)(uint8_t *_src, ptrdiff_t _stride)
+{
     int i;
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
@@ -454,7 +497,8 @@ static void FUNCC(pred8x8_vertical)(uint8_t *_src, int _stride){
     }
 }
 
-static void FUNCC(pred8x16_vertical)(uint8_t *_src, int _stride){
+static void FUNCC(pred8x16_vertical)(uint8_t *_src, ptrdiff_t _stride)
+{
     int i;
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
@@ -467,7 +511,8 @@ static void FUNCC(pred8x16_vertical)(uint8_t *_src, int _stride){
     }
 }
 
-static void FUNCC(pred8x8_horizontal)(uint8_t *_src, int stride){
+static void FUNCC(pred8x8_horizontal)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     pixel *src = (pixel*)_src;
     stride >>= sizeof(pixel)-1;
@@ -479,7 +524,8 @@ static void FUNCC(pred8x8_horizontal)(uint8_t *_src, int stride){
     }
 }
 
-static void FUNCC(pred8x16_horizontal)(uint8_t *_src, int stride){
+static void FUNCC(pred8x16_horizontal)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     pixel *src = (pixel*)_src;
     stride >>= sizeof(pixel)-1;
@@ -491,7 +537,8 @@ static void FUNCC(pred8x16_horizontal)(uint8_t *_src, int stride){
 }
 
 #define PRED8x8_X(n, v)\
-static void FUNCC(pred8x8_##n##_dc)(uint8_t *_src, int stride){\
+static void FUNCC(pred8x8_##n##_dc)(uint8_t *_src, ptrdiff_t stride)\
+{\
     int i;\
     const pixel4 a = PIXEL_SPLAT_X4(v);\
     pixel *src = (pixel*)_src;\
@@ -506,12 +553,14 @@ PRED8x8_X(127, (1<<(BIT_DEPTH-1))-1)
 PRED8x8_X(128, (1<<(BIT_DEPTH-1))+0)
 PRED8x8_X(129, (1<<(BIT_DEPTH-1))+1)
 
-static void FUNCC(pred8x16_128_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred8x16_128_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     FUNCC(pred8x8_128_dc)(_src, stride);
     FUNCC(pred8x8_128_dc)(_src+8*stride, stride);
 }
 
-static void FUNCC(pred8x8_left_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred8x8_left_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     int dc0, dc2;
     pixel4 dc0splat, dc2splat;
@@ -536,12 +585,14 @@ static void FUNCC(pred8x8_left_dc)(uint8_t *_src, int stride){
     }
 }
 
-static void FUNCC(pred8x16_left_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred8x16_left_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     FUNCC(pred8x8_left_dc)(_src, stride);
     FUNCC(pred8x8_left_dc)(_src+8*stride, stride);
 }
 
-static void FUNCC(pred8x8_top_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred8x8_top_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     int dc0, dc1;
     pixel4 dc0splat, dc1splat;
@@ -566,7 +617,8 @@ static void FUNCC(pred8x8_top_dc)(uint8_t *_src, int stride){
     }
 }
 
-static void FUNCC(pred8x16_top_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred8x16_top_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     int dc0, dc1;
     pixel4 dc0splat, dc1splat;
@@ -587,7 +639,8 @@ static void FUNCC(pred8x16_top_dc)(uint8_t *_src, int stride){
     }
 }
 
-static void FUNCC(pred8x8_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred8x8_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     int dc0, dc1, dc2;
     pixel4 dc0splat, dc1splat, dc2splat, dc3splat;
@@ -615,7 +668,8 @@ static void FUNCC(pred8x8_dc)(uint8_t *_src, int stride){
     }
 }
 
-static void FUNCC(pred8x16_dc)(uint8_t *_src, int stride){
+static void FUNCC(pred8x16_dc)(uint8_t *_src, ptrdiff_t stride)
+{
     int i;
     int dc0, dc1, dc2, dc3, dc4;
     pixel4 dc0splat, dc1splat, dc2splat, dc3splat, dc4splat, dc5splat, dc6splat, dc7splat;
@@ -658,51 +712,60 @@ static void FUNCC(pred8x16_dc)(uint8_t *_src, int stride){
 }
 
 //the following 4 function should not be optimized!
-static void FUNC(pred8x8_mad_cow_dc_l0t)(uint8_t *src, int stride){
+static void FUNC(pred8x8_mad_cow_dc_l0t)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x8_top_dc)(src, stride);
     FUNCC(pred4x4_dc)(src, NULL, stride);
 }
 
-static void FUNC(pred8x16_mad_cow_dc_l0t)(uint8_t *src, int stride){
+static void FUNC(pred8x16_mad_cow_dc_l0t)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x16_top_dc)(src, stride);
     FUNCC(pred4x4_dc)(src, NULL, stride);
 }
 
-static void FUNC(pred8x8_mad_cow_dc_0lt)(uint8_t *src, int stride){
+static void FUNC(pred8x8_mad_cow_dc_0lt)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x8_dc)(src, stride);
     FUNCC(pred4x4_top_dc)(src, NULL, stride);
 }
 
-static void FUNC(pred8x16_mad_cow_dc_0lt)(uint8_t *src, int stride){
+static void FUNC(pred8x16_mad_cow_dc_0lt)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x16_dc)(src, stride);
     FUNCC(pred4x4_top_dc)(src, NULL, stride);
 }
 
-static void FUNC(pred8x8_mad_cow_dc_l00)(uint8_t *src, int stride){
+static void FUNC(pred8x8_mad_cow_dc_l00)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x8_left_dc)(src, stride);
     FUNCC(pred4x4_128_dc)(src + 4*stride                  , NULL, stride);
     FUNCC(pred4x4_128_dc)(src + 4*stride + 4*sizeof(pixel), NULL, stride);
 }
 
-static void FUNC(pred8x16_mad_cow_dc_l00)(uint8_t *src, int stride){
+static void FUNC(pred8x16_mad_cow_dc_l00)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x16_left_dc)(src, stride);
     FUNCC(pred4x4_128_dc)(src + 4*stride                  , NULL, stride);
     FUNCC(pred4x4_128_dc)(src + 4*stride + 4*sizeof(pixel), NULL, stride);
 }
 
-static void FUNC(pred8x8_mad_cow_dc_0l0)(uint8_t *src, int stride){
+static void FUNC(pred8x8_mad_cow_dc_0l0)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x8_left_dc)(src, stride);
     FUNCC(pred4x4_128_dc)(src                  , NULL, stride);
     FUNCC(pred4x4_128_dc)(src + 4*sizeof(pixel), NULL, stride);
 }
 
-static void FUNC(pred8x16_mad_cow_dc_0l0)(uint8_t *src, int stride){
+static void FUNC(pred8x16_mad_cow_dc_0l0)(uint8_t *src, ptrdiff_t stride)
+{
     FUNCC(pred8x16_left_dc)(src, stride);
     FUNCC(pred4x4_128_dc)(src                  , NULL, stride);
     FUNCC(pred4x4_128_dc)(src + 4*sizeof(pixel), NULL, stride);
 }
 
-static void FUNCC(pred8x8_plane)(uint8_t *_src, int _stride){
+static void FUNCC(pred8x8_plane)(uint8_t *_src, ptrdiff_t _stride)
+{
   int j, k;
   int a;
   INIT_CLIP
@@ -737,7 +800,8 @@ static void FUNCC(pred8x8_plane)(uint8_t *_src, int _stride){
   }
 }
 
-static void FUNCC(pred8x16_plane)(uint8_t *_src, int _stride){
+static void FUNCC(pred8x16_plane)(uint8_t *_src, ptrdiff_t _stride)
+{
   int j, k;
   int a;
   INIT_CLIP
@@ -816,14 +880,16 @@ static void FUNCC(pred8x16_plane)(uint8_t *_src, int _stride){
         src += stride; \
     }
 
-static void FUNCC(pred8x8l_128_dc)(uint8_t *_src, int has_topleft, int has_topright, int _stride)
+static void FUNCC(pred8x8l_128_dc)(uint8_t *_src, int has_topleft,
+                                   int has_topright, ptrdiff_t _stride)
 {
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
 
     PREDICT_8x8_DC(PIXEL_SPLAT_X4(1<<(BIT_DEPTH-1)));
 }
-static void FUNCC(pred8x8l_left_dc)(uint8_t *_src, int has_topleft, int has_topright, int _stride)
+static void FUNCC(pred8x8l_left_dc)(uint8_t *_src, int has_topleft,
+                                    int has_topright, ptrdiff_t _stride)
 {
     pixel *src = (pixel*)_src;
     int stride = _stride>>(sizeof(pixel)-1);
@@ -832,19 +898,21 @@ static void FUNCC(pred8x8l_left_dc)(uint8_t *_src, int has_topleft, int has_topr
     const pixel4 dc = PIXEL_SPLAT_X4((l0+l1+l2+l3+l4+l5+l6+l7+4) >> 3);
     PREDICT_8x8_DC(dc);
 }
-static void FUNCC(pred8x8l_top_dc)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_top_dc)(uint8_t *_src, int has_topleft,
+                                   int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
 
     PREDICT_8x8_LOAD_TOP;
     const pixel4 dc = PIXEL_SPLAT_X4((t0+t1+t2+t3+t4+t5+t6+t7+4) >> 3);
     PREDICT_8x8_DC(dc);
 }
-static void FUNCC(pred8x8l_dc)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_dc)(uint8_t *_src, int has_topleft,
+                               int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
 
     PREDICT_8x8_LOAD_LEFT;
     PREDICT_8x8_LOAD_TOP;
@@ -852,10 +920,11 @@ static void FUNCC(pred8x8l_dc)(uint8_t *p_src, int has_topleft, int has_topright
                                      +t0+t1+t2+t3+t4+t5+t6+t7+8) >> 4);
     PREDICT_8x8_DC(dc);
 }
-static void FUNCC(pred8x8l_horizontal)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_horizontal)(uint8_t *_src, int has_topleft,
+                                       int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
     pixel4 a;
 
     PREDICT_8x8_LOAD_LEFT;
@@ -865,7 +934,8 @@ static void FUNCC(pred8x8l_horizontal)(uint8_t *p_src, int has_topleft, int has_
     ROW(0); ROW(1); ROW(2); ROW(3); ROW(4); ROW(5); ROW(6); ROW(7);
 #undef ROW
 }
-static void FUNCC(pred8x8l_vertical)(uint8_t *_src, int has_topleft, int has_topright, int _stride)
+static void FUNCC(pred8x8l_vertical)(uint8_t *_src, int has_topleft,
+                                     int has_topright, ptrdiff_t _stride)
 {
     int y;
     pixel *src = (pixel*)_src;
@@ -888,10 +958,11 @@ static void FUNCC(pred8x8l_vertical)(uint8_t *_src, int has_topleft, int has_top
         AV_WN4PA(((pixel4*)(src+y*stride))+1, b);
     }
 }
-static void FUNCC(pred8x8l_down_left)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_down_left)(uint8_t *_src, int has_topleft,
+                                      int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
     PREDICT_8x8_LOAD_TOP;
     PREDICT_8x8_LOAD_TOPRIGHT;
     SRC(0,0)= (t0 + 2*t1 + t2 + 2) >> 2;
@@ -910,10 +981,11 @@ static void FUNCC(pred8x8l_down_left)(uint8_t *p_src, int has_topleft, int has_t
     SRC(6,7)=SRC(7,6)= (t13 + 2*t14 + t15 + 2) >> 2;
     SRC(7,7)= (t14 + 3*t15 + 2) >> 2;
 }
-static void FUNCC(pred8x8l_down_right)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_down_right)(uint8_t *_src, int has_topleft,
+                                       int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
     PREDICT_8x8_LOAD_TOP;
     PREDICT_8x8_LOAD_LEFT;
     PREDICT_8x8_LOAD_TOPLEFT;
@@ -933,10 +1005,11 @@ static void FUNCC(pred8x8l_down_right)(uint8_t *p_src, int has_topleft, int has_
     SRC(6,0)=SRC(7,1)= (t4 + 2*t5 + t6 + 2) >> 2;
     SRC(7,0)= (t5 + 2*t6 + t7 + 2) >> 2;
 }
-static void FUNCC(pred8x8l_vertical_right)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_vertical_right)(uint8_t *_src, int has_topleft,
+                                           int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
     PREDICT_8x8_LOAD_TOP;
     PREDICT_8x8_LOAD_LEFT;
     PREDICT_8x8_LOAD_TOPLEFT;
@@ -963,10 +1036,11 @@ static void FUNCC(pred8x8l_vertical_right)(uint8_t *p_src, int has_topleft, int
     SRC(7,1)= (t5 + 2*t6 + t7 + 2) >> 2;
     SRC(7,0)= (t6 + t7 + 1) >> 1;
 }
-static void FUNCC(pred8x8l_horizontal_down)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_horizontal_down)(uint8_t *_src, int has_topleft,
+                                            int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
     PREDICT_8x8_LOAD_TOP;
     PREDICT_8x8_LOAD_LEFT;
     PREDICT_8x8_LOAD_TOPLEFT;
@@ -993,10 +1067,11 @@ static void FUNCC(pred8x8l_horizontal_down)(uint8_t *p_src, int has_topleft, int
     SRC(6,0)= (t5 + 2*t4 + t3 + 2) >> 2;
     SRC(7,0)= (t6 + 2*t5 + t4 + 2) >> 2;
 }
-static void FUNCC(pred8x8l_vertical_left)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_vertical_left)(uint8_t *_src, int has_topleft,
+                                          int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
     PREDICT_8x8_LOAD_TOP;
     PREDICT_8x8_LOAD_TOPRIGHT;
     SRC(0,0)= (t0 + t1 + 1) >> 1;
@@ -1022,10 +1097,11 @@ static void FUNCC(pred8x8l_vertical_left)(uint8_t *p_src, int has_topleft, int h
     SRC(7,6)= (t10 + t11 + 1) >> 1;
     SRC(7,7)= (t10 + 2*t11 + t12 + 2) >> 2;
 }
-static void FUNCC(pred8x8l_horizontal_up)(uint8_t *p_src, int has_topleft, int has_topright, int p_stride)
+static void FUNCC(pred8x8l_horizontal_up)(uint8_t *_src, int has_topleft,
+                                          int has_topright, ptrdiff_t _stride)
 {
-    pixel *src = (pixel*)p_src;
-    int stride = p_stride>>(sizeof(pixel)-1);
+    pixel *src = (pixel*)_src;
+    int stride = _stride>>(sizeof(pixel)-1);
     PREDICT_8x8_LOAD_LEFT;
     SRC(0,0)= (l0 + l1 + 1) >> 1;
     SRC(1,0)= (l0 + 2*l1 + l2 + 2) >> 2;
@@ -1056,10 +1132,12 @@ static void FUNCC(pred8x8l_horizontal_up)(uint8_t *p_src, int has_topleft, int h
 #undef PL
 #undef SRC
 
-static void FUNCC(pred4x4_vertical_add)(uint8_t *p_pix, const DCTELEM *p_block, int stride){
+static void FUNCC(pred4x4_vertical_add)(uint8_t *_pix, const DCTELEM *_block,
+                                        ptrdiff_t stride)
+{
     int i;
-    pixel *pix = (pixel*)p_pix;
-    const dctcoef *block = (const dctcoef*)p_block;
+    pixel *pix = (pixel*)_pix;
+    const dctcoef *block = (const dctcoef*)_block;
     stride >>= sizeof(pixel)-1;
     pix -= stride;
     for(i=0; i<4; i++){
@@ -1073,10 +1151,12 @@ static void FUNCC(pred4x4_vertical_add)(uint8_t *p_pix, const DCTELEM *p_block,
     }
 }
 
-static void FUNCC(pred4x4_horizontal_add)(uint8_t *p_pix, const DCTELEM *p_block, int stride){
+static void FUNCC(pred4x4_horizontal_add)(uint8_t *_pix, const DCTELEM *_block,
+                                          ptrdiff_t stride)
+{
     int i;
-    pixel *pix = (pixel*)p_pix;
-    const dctcoef *block = (const dctcoef*)p_block;
+    pixel *pix = (pixel*)_pix;
+    const dctcoef *block = (const dctcoef*)_block;
     stride >>= sizeof(pixel)-1;
     for(i=0; i<4; i++){
         pixel v = pix[-1];
@@ -1089,10 +1169,12 @@ static void FUNCC(pred4x4_horizontal_add)(uint8_t *p_pix, const DCTELEM *p_block
     }
 }
 
-static void FUNCC(pred8x8l_vertical_add)(uint8_t *p_pix, const DCTELEM *p_block, int stride){
+static void FUNCC(pred8x8l_vertical_add)(uint8_t *_pix, const DCTELEM *_block,
+                                         ptrdiff_t stride)
+{
     int i;
-    pixel *pix = (pixel*)p_pix;
-    const dctcoef *block = (const dctcoef*)p_block;
+    pixel *pix = (pixel*)_pix;
+    const dctcoef *block = (const dctcoef*)_block;
     stride >>= sizeof(pixel)-1;
     pix -= stride;
     for(i=0; i<8; i++){
@@ -1110,10 +1192,12 @@ static void FUNCC(pred8x8l_vertical_add)(uint8_t *p_pix, const DCTELEM *p_block,
     }
 }
 
-static void FUNCC(pred8x8l_horizontal_add)(uint8_t *p_pix, const DCTELEM *p_block, int stride){
+static void FUNCC(pred8x8l_horizontal_add)(uint8_t *_pix, const DCTELEM *_block,
+                                           ptrdiff_t stride)
+{
     int i;
-    pixel *pix = (pixel*)p_pix;
-    const dctcoef *block = (const dctcoef*)p_block;
+    pixel *pix = (pixel*)_pix;
+    const dctcoef *block = (const dctcoef*)_block;
     stride >>= sizeof(pixel)-1;
     for(i=0; i<8; i++){
         pixel v = pix[-1];
@@ -1130,25 +1214,36 @@ static void FUNCC(pred8x8l_horizontal_add)(uint8_t *p_pix, const DCTELEM *p_bloc
     }
 }
 
-static void FUNCC(pred16x16_vertical_add)(uint8_t *pix, const int *block_offset, const DCTELEM *block, int stride){
+static void FUNCC(pred16x16_vertical_add)(uint8_t *pix, const int *block_offset,
+                                          const DCTELEM *block,
+                                          ptrdiff_t stride)
+{
     int i;
     for(i=0; i<16; i++)
         FUNCC(pred4x4_vertical_add)(pix + block_offset[i], block + i*16*sizeof(pixel), stride);
 }
 
-static void FUNCC(pred16x16_horizontal_add)(uint8_t *pix, const int *block_offset, const DCTELEM *block, int stride){
+static void FUNCC(pred16x16_horizontal_add)(uint8_t *pix,
+                                            const int *block_offset,
+                                            const DCTELEM *block,
+                                            ptrdiff_t stride)
+{
     int i;
     for(i=0; i<16; i++)
         FUNCC(pred4x4_horizontal_add)(pix + block_offset[i], block + i*16*sizeof(pixel), stride);
 }
 
-static void FUNCC(pred8x8_vertical_add)(uint8_t *pix, const int *block_offset, const DCTELEM *block, int stride){
+static void FUNCC(pred8x8_vertical_add)(uint8_t *pix, const int *block_offset,
+                                        const DCTELEM *block, ptrdiff_t stride)
+{
     int i;
     for(i=0; i<4; i++)
         FUNCC(pred4x4_vertical_add)(pix + block_offset[i], block + i*16*sizeof(pixel), stride);
 }
 
-static void FUNCC(pred8x16_vertical_add)(uint8_t *pix, const int *block_offset, const DCTELEM *block, int stride){
+static void FUNCC(pred8x16_vertical_add)(uint8_t *pix, const int *block_offset,
+                                         const DCTELEM *block, ptrdiff_t stride)
+{
     int i;
     for(i=0; i<4; i++)
         FUNCC(pred4x4_vertical_add)(pix + block_offset[i], block + i*16*sizeof(pixel), stride);
@@ -1156,13 +1251,19 @@ static void FUNCC(pred8x16_vertical_add)(uint8_t *pix, const int *block_offset,
         FUNCC(pred4x4_vertical_add)(pix + block_offset[i+4], block + i*16*sizeof(pixel), stride);
 }
 
-static void FUNCC(pred8x8_horizontal_add)(uint8_t *pix, const int *block_offset, const DCTELEM *block, int stride){
+static void FUNCC(pred8x8_horizontal_add)(uint8_t *pix, const int *block_offset,
+                                          const DCTELEM *block,
+                                          ptrdiff_t stride)
+{
     int i;
     for(i=0; i<4; i++)
         FUNCC(pred4x4_horizontal_add)(pix + block_offset[i], block + i*16*sizeof(pixel), stride);
 }
 
-static void FUNCC(pred8x16_horizontal_add)(uint8_t *pix, const int *block_offset, const DCTELEM *block, int stride){
+static void FUNCC(pred8x16_horizontal_add)(uint8_t *pix,
+                                           const int *block_offset,
+                                           const DCTELEM *block, ptrdiff_t stride)
+{
     int i;
     for(i=0; i<4; i++)
         FUNCC(pred4x4_horizontal_add)(pix + block_offset[i], block + i*16*sizeof(pixel), stride);
diff --git a/libavcodec/raw.c b/libavcodec/raw.c
index 5f8bf33b79..e23dbeadf9 100644
--- a/libavcodec/raw.c
+++ b/libavcodec/raw.c
@@ -147,6 +147,25 @@ const PixelFormatTag ff_raw_pix_fmt_tags[] = {
     { AV_PIX_FMT_YUVA444P,    MKTAG('Y', '4',  0 ,  8 ) },
     { AV_PIX_FMT_GRAY8A,      MKTAG('Y', '2',  0 ,  8 ) },
 
+    { AV_PIX_FMT_YUVA420P9LE,  MKTAG('Y', '4', 11 ,  9 ) },
+    { AV_PIX_FMT_YUVA420P9BE,  MKTAG( 9 , 11 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA422P9LE,  MKTAG('Y', '4', 10 ,  9 ) },
+    { AV_PIX_FMT_YUVA422P9BE,  MKTAG( 9 , 10 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA444P9LE,  MKTAG('Y', '4',  0 ,  9 ) },
+    { AV_PIX_FMT_YUVA444P9BE,  MKTAG( 9 ,  0 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA420P10LE, MKTAG('Y', '4', 11 , 10 ) },
+    { AV_PIX_FMT_YUVA420P10BE, MKTAG(10 , 11 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA422P10LE, MKTAG('Y', '4', 10 , 10 ) },
+    { AV_PIX_FMT_YUVA422P10BE, MKTAG(10 , 10 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA444P10LE, MKTAG('Y', '4',  0 , 10 ) },
+    { AV_PIX_FMT_YUVA444P10BE, MKTAG(10 ,  0 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA420P16LE, MKTAG('Y', '4', 11 , 16 ) },
+    { AV_PIX_FMT_YUVA420P16BE, MKTAG(16 , 11 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA422P16LE, MKTAG('Y', '4', 10 , 16 ) },
+    { AV_PIX_FMT_YUVA422P16BE, MKTAG(16 , 10 , '4', 'Y') },
+    { AV_PIX_FMT_YUVA444P16LE, MKTAG('Y', '4',  0 , 16 ) },
+    { AV_PIX_FMT_YUVA444P16BE, MKTAG(16 ,  0 , '4', 'Y') },
+
     /* quicktime */
     { AV_PIX_FMT_YUV420P, MKTAG('R', '4', '2', '0') }, /* Radius DV YUV PAL */
     { AV_PIX_FMT_YUV411P, MKTAG('R', '4', '1', '1') }, /* Radius DV YUV NTSC */
diff --git a/libavcodec/x86/h264_intrapred.asm b/libavcodec/x86/h264_intrapred.asm
index 609cb2303d..21dbf5d9cd 100644
--- a/libavcodec/x86/h264_intrapred.asm
+++ b/libavcodec/x86/h264_intrapred.asm
@@ -53,7 +53,7 @@ cextern pw_32
 ; void pred16x16_vertical(uint8_t *src, int stride)
 ;-----------------------------------------------------------------------------
 
-cglobal pred16x16_vertical_mmx, 2,3
+cglobal pred16x16_vertical_8_mmx, 2,3
     sub   r0, r1
     mov   r2, 8
     movq mm0, [r0+0]
@@ -68,7 +68,7 @@ cglobal pred16x16_vertical_mmx, 2,3
     jg .loop
     REP_RET
 
-cglobal pred16x16_vertical_sse, 2,3
+cglobal pred16x16_vertical_8_sse, 2,3
     sub   r0, r1
     mov   r2, 4
     movaps xmm0, [r0]
@@ -88,7 +88,7 @@ cglobal pred16x16_vertical_sse, 2,3
 ;-----------------------------------------------------------------------------
 
 %macro PRED16x16_H 0
-cglobal pred16x16_horizontal, 2,3
+cglobal pred16x16_horizontal_8, 2,3
     mov       r2, 8
 %if cpuflag(ssse3)
     mova      m2, [pb_3]
@@ -130,7 +130,7 @@ INIT_XMM
 ;-----------------------------------------------------------------------------
 
 %macro PRED16x16_DC 0
-cglobal pred16x16_dc, 2,7
+cglobal pred16x16_dc_8, 2,7
     mov       r4, r0
     sub       r0, r1
     pxor      mm0, mm0
@@ -193,7 +193,7 @@ INIT_XMM
 ;-----------------------------------------------------------------------------
 
 %macro PRED16x16_TM_MMX 0
-cglobal pred16x16_tm_vp8, 2,5
+cglobal pred16x16_tm_vp8_8, 2,5
     sub        r0, r1
     pxor      mm7, mm7
     movq      mm0, [r0+0]
@@ -234,7 +234,7 @@ INIT_MMX mmx2
 PRED16x16_TM_MMX
 INIT_MMX
 
-cglobal pred16x16_tm_vp8_sse2, 2,6,6
+cglobal pred16x16_tm_vp8_8_sse2, 2,6,6
     sub          r0, r1
     pxor       xmm2, xmm2
     movdqa     xmm0, [r0]
@@ -274,7 +274,7 @@ cglobal pred16x16_tm_vp8_sse2, 2,6,6
 ;-----------------------------------------------------------------------------
 
 %macro H264_PRED16x16_PLANE 1
-cglobal pred16x16_plane_%1, 2,9,7
+cglobal pred16x16_plane_%1_8, 2,9,7
     mov          r2, r1           ; +stride
     neg          r1               ; -stride
 
@@ -556,7 +556,7 @@ INIT_XMM
 ;-----------------------------------------------------------------------------
 
 %macro H264_PRED8x8_PLANE 0
-cglobal pred8x8_plane, 2,9,7
+cglobal pred8x8_plane_8, 2,9,7
     mov          r2, r1           ; +stride
     neg          r1               ; -stride
 
@@ -730,7 +730,7 @@ INIT_XMM
 ; void pred8x8_vertical(uint8_t *src, int stride)
 ;-----------------------------------------------------------------------------
 
-cglobal pred8x8_vertical_mmx, 2,2
+cglobal pred8x8_vertical_8_mmx, 2,2
     sub    r0, r1
     movq  mm0, [r0]
 %rep 3
@@ -747,7 +747,7 @@ cglobal pred8x8_vertical_mmx, 2,2
 ;-----------------------------------------------------------------------------
 
 %macro PRED8x8_H 0
-cglobal pred8x8_horizontal, 2,3
+cglobal pred8x8_horizontal_8, 2,3
     mov       r2, 4
 %if cpuflag(ssse3)
     mova      m2, [pb_3]
@@ -774,7 +774,7 @@ INIT_MMX
 ;-----------------------------------------------------------------------------
 ; void pred8x8_top_dc_mmxext(uint8_t *src, int stride)
 ;-----------------------------------------------------------------------------
-cglobal pred8x8_top_dc_mmxext, 2,5
+cglobal pred8x8_top_dc_8_mmxext, 2,5
     sub         r0, r1
     movq       mm0, [r0]
     pxor       mm1, mm1
@@ -809,7 +809,7 @@ cglobal pred8x8_top_dc_mmxext, 2,5
 ;-----------------------------------------------------------------------------
 
 INIT_MMX
-cglobal pred8x8_dc_mmxext, 2,5
+cglobal pred8x8_dc_8_mmxext, 2,5
     sub       r0, r1
     pxor      m7, m7
     movd      m0, [r0+0]
@@ -869,7 +869,7 @@ cglobal pred8x8_dc_mmxext, 2,5
 ; void pred8x8_dc_rv40(uint8_t *src, int stride)
 ;-----------------------------------------------------------------------------
 
-cglobal pred8x8_dc_rv40_mmxext, 2,7
+cglobal pred8x8_dc_rv40_8_mmxext, 2,7
     mov       r4, r0
     sub       r0, r1
     pxor      mm0, mm0
@@ -906,7 +906,7 @@ cglobal pred8x8_dc_rv40_mmxext, 2,7
 ;-----------------------------------------------------------------------------
 
 %macro PRED8x8_TM_MMX 0
-cglobal pred8x8_tm_vp8, 2,6
+cglobal pred8x8_tm_vp8_8, 2,6
     sub        r0, r1
     pxor      mm7, mm7
     movq      mm0, [r0]
@@ -946,7 +946,7 @@ INIT_MMX mmx2
 PRED8x8_TM_MMX
 INIT_MMX
 
-cglobal pred8x8_tm_vp8_sse2, 2,6,4
+cglobal pred8x8_tm_vp8_8_sse2, 2,6,4
     sub          r0, r1
     pxor       xmm1, xmm1
     movq       xmm0, [r0]
@@ -974,7 +974,7 @@ cglobal pred8x8_tm_vp8_sse2, 2,6,4
     jg .loop
     REP_RET
 
-cglobal pred8x8_tm_vp8_ssse3, 2,3,6
+cglobal pred8x8_tm_vp8_8_ssse3, 2,3,6
     sub          r0, r1
     movdqa     xmm4, [tm_shuf]
     pxor       xmm1, xmm1
@@ -1016,7 +1016,7 @@ cglobal pred8x8_tm_vp8_ssse3, 2,3,6
 ; void pred8x8l_top_dc(uint8_t *src, int has_topleft, int has_topright, int stride)
 ;-----------------------------------------------------------------------------
 %macro PRED8x8L_TOP_DC 1
-cglobal pred8x8l_top_dc_%1, 4,4
+cglobal pred8x8l_top_dc_8_%1, 4,4
     sub          r0, r3
     pxor        mm7, mm7
     movq        mm0, [r0-8]
@@ -1073,7 +1073,7 @@ PRED8x8L_TOP_DC ssse3
 ;-----------------------------------------------------------------------------
 
 %macro PRED8x8L_DC 1
-cglobal pred8x8l_dc_%1, 4,5
+cglobal pred8x8l_dc_8_%1, 4,5
     sub          r0, r3
     lea          r4, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -1176,7 +1176,7 @@ PRED8x8L_DC ssse3
 ;-----------------------------------------------------------------------------
 
 %macro PRED8x8L_HORIZONTAL 1
-cglobal pred8x8l_horizontal_%1, 4,4
+cglobal pred8x8l_horizontal_8_%1, 4,4
     sub          r0, r3
     lea          r2, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -1248,7 +1248,7 @@ PRED8x8L_HORIZONTAL ssse3
 ;-----------------------------------------------------------------------------
 
 %macro PRED8x8L_VERTICAL 1
-cglobal pred8x8l_vertical_%1, 4,4
+cglobal pred8x8l_vertical_8_%1, 4,4
     sub          r0, r3
     movq        mm0, [r0-8]
     movq        mm3, [r0]
@@ -1300,7 +1300,7 @@ PRED8x8L_VERTICAL ssse3
 
 INIT_MMX
 %define PALIGNR PALIGNR_MMX
-cglobal pred8x8l_down_left_mmxext, 4,5
+cglobal pred8x8l_down_left_8_mmxext, 4,5
     sub          r0, r3
     movq        mm0, [r0-8]
     movq        mm3, [r0]
@@ -1408,7 +1408,7 @@ cglobal pred8x8l_down_left_mmxext, 4,5
     RET
 
 %macro PRED8x8L_DOWN_LEFT 1
-cglobal pred8x8l_down_left_%1, 4,4
+cglobal pred8x8l_down_left_8_%1, 4,4
     sub          r0, r3
     movq        mm0, [r0-8]
     movq        mm3, [r0]
@@ -1503,7 +1503,7 @@ PRED8x8L_DOWN_LEFT ssse3
 
 INIT_MMX
 %define PALIGNR PALIGNR_MMX
-cglobal pred8x8l_down_right_mmxext, 4,5
+cglobal pred8x8l_down_right_8_mmxext, 4,5
     sub          r0, r3
     lea          r4, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -1635,7 +1635,7 @@ cglobal pred8x8l_down_right_mmxext, 4,5
     RET
 
 %macro PRED8x8L_DOWN_RIGHT 1
-cglobal pred8x8l_down_right_%1, 4,5
+cglobal pred8x8l_down_right_8_%1, 4,5
     sub          r0, r3
     lea          r4, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -1757,7 +1757,7 @@ PRED8x8L_DOWN_RIGHT ssse3
 
 INIT_MMX
 %define PALIGNR PALIGNR_MMX
-cglobal pred8x8l_vertical_right_mmxext, 4,5
+cglobal pred8x8l_vertical_right_8_mmxext, 4,5
     sub          r0, r3
     lea          r4, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -1864,7 +1864,7 @@ cglobal pred8x8l_vertical_right_mmxext, 4,5
     RET
 
 %macro PRED8x8L_VERTICAL_RIGHT 1
-cglobal pred8x8l_vertical_right_%1, 4,5,7
+cglobal pred8x8l_vertical_right_8_%1, 4,5,7
     ; manually spill XMM registers for Win64 because
     ; the code here is initialized with INIT_MMX
     WIN64_SPILL_XMM 7
@@ -1986,7 +1986,7 @@ PRED8x8L_VERTICAL_RIGHT ssse3
 ;-----------------------------------------------------------------------------
 
 %macro PRED8x8L_VERTICAL_LEFT 1
-cglobal pred8x8l_vertical_left_%1, 4,4
+cglobal pred8x8l_vertical_left_8_%1, 4,4
     sub          r0, r3
     movq        mm0, [r0-8]
     movq        mm3, [r0]
@@ -2077,7 +2077,7 @@ PRED8x8L_VERTICAL_LEFT ssse3
 ;-----------------------------------------------------------------------------
 
 %macro PRED8x8L_HORIZONTAL_UP 1
-cglobal pred8x8l_horizontal_up_%1, 4,4
+cglobal pred8x8l_horizontal_up_8_%1, 4,4
     sub          r0, r3
     lea          r2, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -2166,7 +2166,7 @@ PRED8x8L_HORIZONTAL_UP ssse3
 
 INIT_MMX
 %define PALIGNR PALIGNR_MMX
-cglobal pred8x8l_horizontal_down_mmxext, 4,5
+cglobal pred8x8l_horizontal_down_8_mmxext, 4,5
     sub          r0, r3
     lea          r4, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -2281,7 +2281,7 @@ cglobal pred8x8l_horizontal_down_mmxext, 4,5
     RET
 
 %macro PRED8x8L_HORIZONTAL_DOWN 1
-cglobal pred8x8l_horizontal_down_%1, 4,5
+cglobal pred8x8l_horizontal_down_8_%1, 4,5
     sub          r0, r3
     lea          r4, [r0+r3*2]
     movq        mm0, [r0+r3*1-8]
@@ -2415,7 +2415,7 @@ PRED8x8L_HORIZONTAL_DOWN ssse3
 ; void pred4x4_dc_mmxext(uint8_t *src, const uint8_t *topright, int stride)
 ;-----------------------------------------------------------------------------
 
-cglobal pred4x4_dc_mmxext, 3,5
+cglobal pred4x4_dc_8_mmxext, 3,5
     pxor   mm7, mm7
     mov     r4, r0
     sub     r0, r2
@@ -2445,7 +2445,7 @@ cglobal pred4x4_dc_mmxext, 3,5
 ;-----------------------------------------------------------------------------
 
 %macro PRED4x4_TM_MMX 0
-cglobal pred4x4_tm_vp8, 3,6
+cglobal pred4x4_tm_vp8_8, 3,6
     sub        r0, r2
     pxor      mm7, mm7
     movd      mm0, [r0]
@@ -2486,7 +2486,7 @@ INIT_MMX mmx2
 PRED4x4_TM_MMX
 INIT_MMX
 
-cglobal pred4x4_tm_vp8_ssse3, 3,3
+cglobal pred4x4_tm_vp8_8_ssse3, 3,3
     sub         r0, r2
     movq       mm6, [tm_shuf]
     pxor       mm1, mm1
@@ -2526,7 +2526,7 @@ cglobal pred4x4_tm_vp8_ssse3, 3,3
 ;-----------------------------------------------------------------------------
 
 INIT_MMX
-cglobal pred4x4_vertical_vp8_mmxext, 3,3
+cglobal pred4x4_vertical_vp8_8_mmxext, 3,3
     sub       r0, r2
     movd      m1, [r0-1]
     movd      m0, [r0]
@@ -2545,7 +2545,7 @@ cglobal pred4x4_vertical_vp8_mmxext, 3,3
 ; void pred4x4_down_left_mmxext(uint8_t *src, const uint8_t *topright, int stride)
 ;-----------------------------------------------------------------------------
 INIT_MMX
-cglobal pred4x4_down_left_mmxext, 3,3
+cglobal pred4x4_down_left_8_mmxext, 3,3
     sub       r0, r2
     movq      m1, [r0]
     punpckldq m1, [r1]
@@ -2572,7 +2572,7 @@ cglobal pred4x4_down_left_mmxext, 3,3
 ;-----------------------------------------------------------------------------
 
 INIT_MMX
-cglobal pred4x4_vertical_left_mmxext, 3,3
+cglobal pred4x4_vertical_left_8_mmxext, 3,3
     sub       r0, r2
     movq      m1, [r0]
     punpckldq m1, [r1]
@@ -2597,7 +2597,7 @@ cglobal pred4x4_vertical_left_mmxext, 3,3
 ;-----------------------------------------------------------------------------
 
 INIT_MMX
-cglobal pred4x4_horizontal_up_mmxext, 3,3
+cglobal pred4x4_horizontal_up_8_mmxext, 3,3
     sub       r0, r2
     lea       r1, [r0+r2*2]
     movd      m0, [r0+r2*1-4]
@@ -2631,7 +2631,7 @@ cglobal pred4x4_horizontal_up_mmxext, 3,3
 
 INIT_MMX
 %define PALIGNR PALIGNR_MMX
-cglobal pred4x4_horizontal_down_mmxext, 3,3
+cglobal pred4x4_horizontal_down_8_mmxext, 3,3
     sub       r0, r2
     lea       r1, [r0+r2*2]
     movh      m0, [r0-4]      ; lt ..
@@ -2667,7 +2667,7 @@ cglobal pred4x4_horizontal_down_mmxext, 3,3
 
 INIT_MMX
 %define PALIGNR PALIGNR_MMX
-cglobal pred4x4_vertical_right_mmxext, 3,3
+cglobal pred4x4_vertical_right_8_mmxext, 3,3
     sub     r0, r2
     lea     r1, [r0+r2*2]
     movh    m0, [r0]                    ; ........t3t2t1t0
@@ -2698,7 +2698,7 @@ cglobal pred4x4_vertical_right_mmxext, 3,3
 
 INIT_MMX
 %define PALIGNR PALIGNR_MMX
-cglobal pred4x4_down_right_mmxext, 3,3
+cglobal pred4x4_down_right_8_mmxext, 3,3
     sub       r0, r2
     lea       r1, [r0+r2*2]
     movq      m1, [r1-8]
diff --git a/libavcodec/x86/h264_intrapred_init.c b/libavcodec/x86/h264_intrapred_init.c
index 2ce02f0f6c..83a4ca98c1 100644
--- a/libavcodec/x86/h264_intrapred_init.c
+++ b/libavcodec/x86/h264_intrapred_init.c
@@ -23,7 +23,9 @@
 #include "libavcodec/h264pred.h"
 
 #define PRED4x4(TYPE, DEPTH, OPT) \
-void ff_pred4x4_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, const uint8_t *topright, int stride);
+void ff_pred4x4_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, \
+                                                    const uint8_t *topright, \
+                                                    ptrdiff_t stride);
 
 PRED4x4(dc, 10, mmx2)
 PRED4x4(down_left, 10, sse2)
@@ -42,7 +44,8 @@ PRED4x4(horizontal_down, 10, ssse3)
 PRED4x4(horizontal_down, 10, avx)
 
 #define PRED8x8(TYPE, DEPTH, OPT) \
-void ff_pred8x8_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, int stride);
+void ff_pred8x8_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, \
+                                                    ptrdiff_t stride);
 
 PRED8x8(dc, 10, mmx2)
 PRED8x8(dc, 10, sse2)
@@ -52,7 +55,10 @@ PRED8x8(vertical, 10, sse2)
 PRED8x8(horizontal, 10, sse2)
 
 #define PRED8x8L(TYPE, DEPTH, OPT)\
-void ff_pred8x8l_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, int has_topleft, int has_topright, int stride);
+void ff_pred8x8l_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, \
+                                                     int has_topleft, \
+                                                     int has_topright, \
+                                                     ptrdiff_t stride);
 
 PRED8x8L(dc, 10, sse2)
 PRED8x8L(dc, 10, avx)
@@ -79,7 +85,8 @@ PRED8x8L(horizontal_up, 10, ssse3)
 PRED8x8L(horizontal_up, 10, avx)
 
 #define PRED16x16(TYPE, DEPTH, OPT)\
-void ff_pred16x16_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, int stride);
+void ff_pred16x16_ ## TYPE ## _ ## DEPTH ## _ ## OPT (uint8_t *src, \
+                                                      ptrdiff_t stride);
 
 PRED16x16(dc, 10, mmx2)
 PRED16x16(dc, 10, sse2)
@@ -94,79 +101,83 @@ PRED16x16(vertical, 10, sse2)
 PRED16x16(horizontal, 10, mmx2)
 PRED16x16(horizontal, 10, sse2)
 
-void ff_pred16x16_vertical_mmx     (uint8_t *src, int stride);
-void ff_pred16x16_vertical_sse     (uint8_t *src, int stride);
-void ff_pred16x16_horizontal_mmx   (uint8_t *src, int stride);
-void ff_pred16x16_horizontal_mmx2  (uint8_t *src, int stride);
-void ff_pred16x16_horizontal_ssse3 (uint8_t *src, int stride);
-void ff_pred16x16_dc_mmx2          (uint8_t *src, int stride);
-void ff_pred16x16_dc_sse2          (uint8_t *src, int stride);
-void ff_pred16x16_dc_ssse3         (uint8_t *src, int stride);
-void ff_pred16x16_plane_h264_mmx   (uint8_t *src, int stride);
-void ff_pred16x16_plane_h264_mmx2  (uint8_t *src, int stride);
-void ff_pred16x16_plane_h264_sse2  (uint8_t *src, int stride);
-void ff_pred16x16_plane_h264_ssse3 (uint8_t *src, int stride);
-void ff_pred16x16_plane_rv40_mmx   (uint8_t *src, int stride);
-void ff_pred16x16_plane_rv40_mmx2  (uint8_t *src, int stride);
-void ff_pred16x16_plane_rv40_sse2  (uint8_t *src, int stride);
-void ff_pred16x16_plane_rv40_ssse3 (uint8_t *src, int stride);
-void ff_pred16x16_plane_svq3_mmx   (uint8_t *src, int stride);
-void ff_pred16x16_plane_svq3_mmx2  (uint8_t *src, int stride);
-void ff_pred16x16_plane_svq3_sse2  (uint8_t *src, int stride);
-void ff_pred16x16_plane_svq3_ssse3 (uint8_t *src, int stride);
-void ff_pred16x16_tm_vp8_mmx       (uint8_t *src, int stride);
-void ff_pred16x16_tm_vp8_mmx2      (uint8_t *src, int stride);
-void ff_pred16x16_tm_vp8_sse2      (uint8_t *src, int stride);
-void ff_pred8x8_top_dc_mmxext      (uint8_t *src, int stride);
-void ff_pred8x8_dc_rv40_mmxext     (uint8_t *src, int stride);
-void ff_pred8x8_dc_mmxext          (uint8_t *src, int stride);
-void ff_pred8x8_vertical_mmx       (uint8_t *src, int stride);
-void ff_pred8x8_horizontal_mmx     (uint8_t *src, int stride);
-void ff_pred8x8_horizontal_mmx2    (uint8_t *src, int stride);
-void ff_pred8x8_horizontal_ssse3   (uint8_t *src, int stride);
-void ff_pred8x8_plane_mmx          (uint8_t *src, int stride);
-void ff_pred8x8_plane_mmx2         (uint8_t *src, int stride);
-void ff_pred8x8_plane_sse2         (uint8_t *src, int stride);
-void ff_pred8x8_plane_ssse3        (uint8_t *src, int stride);
-void ff_pred8x8_tm_vp8_mmx         (uint8_t *src, int stride);
-void ff_pred8x8_tm_vp8_mmx2        (uint8_t *src, int stride);
-void ff_pred8x8_tm_vp8_sse2        (uint8_t *src, int stride);
-void ff_pred8x8_tm_vp8_ssse3       (uint8_t *src, int stride);
-void ff_pred8x8l_top_dc_mmxext     (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_top_dc_ssse3      (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_dc_mmxext         (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_dc_ssse3          (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_horizontal_mmxext (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_horizontal_ssse3  (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_vertical_mmxext   (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_vertical_ssse3    (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_down_left_mmxext  (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_down_left_sse2    (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_down_left_ssse3   (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_down_right_mmxext (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_down_right_sse2   (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_down_right_ssse3  (uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_vertical_right_mmxext(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_vertical_right_sse2(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_vertical_right_ssse3(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_vertical_left_sse2(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_vertical_left_ssse3(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_horizontal_up_mmxext(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_horizontal_up_ssse3(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_horizontal_down_mmxext(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_horizontal_down_sse2(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred8x8l_horizontal_down_ssse3(uint8_t *src, int has_topleft, int has_topright, int stride);
-void ff_pred4x4_dc_mmxext          (uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_down_left_mmxext   (uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_down_right_mmxext  (uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_vertical_left_mmxext(uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_vertical_right_mmxext(uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_horizontal_up_mmxext(uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_horizontal_down_mmxext(uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_tm_vp8_mmx         (uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_tm_vp8_mmx2        (uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_tm_vp8_ssse3       (uint8_t *src, const uint8_t *topright, int stride);
-void ff_pred4x4_vertical_vp8_mmxext(uint8_t *src, const uint8_t *topright, int stride);
+/* 8-bit versions */
+PRED16x16(vertical, 8, mmx)
+PRED16x16(vertical, 8, sse)
+PRED16x16(horizontal, 8, mmx)
+PRED16x16(horizontal, 8, mmx2)
+PRED16x16(horizontal, 8, ssse3)
+PRED16x16(dc, 8, mmx2)
+PRED16x16(dc, 8, sse2)
+PRED16x16(dc, 8, ssse3)
+PRED16x16(plane_h264, 8, mmx)
+PRED16x16(plane_h264, 8, mmx2)
+PRED16x16(plane_h264, 8, sse2)
+PRED16x16(plane_h264, 8, ssse3)
+PRED16x16(plane_rv40, 8, mmx)
+PRED16x16(plane_rv40, 8, mmx2)
+PRED16x16(plane_rv40, 8, sse2)
+PRED16x16(plane_rv40, 8, ssse3)
+PRED16x16(plane_svq3, 8, mmx)
+PRED16x16(plane_svq3, 8, mmx2)
+PRED16x16(plane_svq3, 8, sse2)
+PRED16x16(plane_svq3, 8, ssse3)
+PRED16x16(tm_vp8, 8, mmx)
+PRED16x16(tm_vp8, 8, mmx2)
+PRED16x16(tm_vp8, 8, sse2)
+
+PRED8x8(top_dc, 8, mmxext)
+PRED8x8(dc_rv40, 8, mmxext)
+PRED8x8(dc, 8, mmxext)
+PRED8x8(vertical, 8, mmx)
+PRED8x8(horizontal, 8, mmx)
+PRED8x8(horizontal, 8, mmx2)
+PRED8x8(horizontal, 8, ssse3)
+PRED8x8(plane, 8, mmx)
+PRED8x8(plane, 8, mmx2)
+PRED8x8(plane, 8, sse2)
+PRED8x8(plane, 8, ssse3)
+PRED8x8(tm_vp8, 8, mmx)
+PRED8x8(tm_vp8, 8, mmx2)
+PRED8x8(tm_vp8, 8, sse2)
+PRED8x8(tm_vp8, 8, ssse3)
+
+PRED8x8L(top_dc, 8, mmxext)
+PRED8x8L(top_dc, 8, ssse3)
+PRED8x8L(dc, 8, mmxext)
+PRED8x8L(dc, 8, ssse3)
+PRED8x8L(horizontal, 8, mmxext)
+PRED8x8L(horizontal, 8, ssse3)
+PRED8x8L(vertical, 8, mmxext)
+PRED8x8L(vertical, 8, ssse3)
+PRED8x8L(down_left, 8, mmxext)
+PRED8x8L(down_left, 8, sse2)
+PRED8x8L(down_left, 8, ssse3)
+PRED8x8L(down_right, 8, mmxext)
+PRED8x8L(down_right, 8, sse2)
+PRED8x8L(down_right, 8, ssse3)
+PRED8x8L(vertical_right, 8, mmxext)
+PRED8x8L(vertical_right, 8, sse2)
+PRED8x8L(vertical_right, 8, ssse3)
+PRED8x8L(vertical_left, 8, sse2)
+PRED8x8L(vertical_left, 8, ssse3)
+PRED8x8L(horizontal_up, 8, mmxext)
+PRED8x8L(horizontal_up, 8, ssse3)
+PRED8x8L(horizontal_down, 8, mmxext)
+PRED8x8L(horizontal_down, 8, sse2)
+PRED8x8L(horizontal_down, 8, ssse3)
+
+PRED4x4(dc, 8, mmxext)
+PRED4x4(down_left, 8, mmxext)
+PRED4x4(down_right, 8, mmxext)
+PRED4x4(vertical_left, 8, mmxext)
+PRED4x4(vertical_right, 8, mmxext)
+PRED4x4(horizontal_up, 8, mmxext)
+PRED4x4(horizontal_down, 8, mmxext)
+PRED4x4(tm_vp8, 8, mmx)
+PRED4x4(tm_vp8, 8, mmx2)
+PRED4x4(tm_vp8, 8, ssse3)
+PRED4x4(vertical_vp8, 8, mmxext)
 
 void ff_h264_pred_init_x86(H264PredContext *h, int codec_id, const int bit_depth, const int chroma_format_idc)
 {
@@ -174,136 +185,136 @@ void ff_h264_pred_init_x86(H264PredContext *h, int codec_id, const int bit_depth
 
     if (bit_depth == 8) {
         if (EXTERNAL_MMX(mm_flags)) {
-            h->pred16x16[VERT_PRED8x8         ] = ff_pred16x16_vertical_mmx;
-            h->pred16x16[HOR_PRED8x8          ] = ff_pred16x16_horizontal_mmx;
+            h->pred16x16[VERT_PRED8x8         ] = ff_pred16x16_vertical_8_mmx;
+            h->pred16x16[HOR_PRED8x8          ] = ff_pred16x16_horizontal_8_mmx;
             if (chroma_format_idc == 1) {
-                h->pred8x8  [VERT_PRED8x8     ] = ff_pred8x8_vertical_mmx;
-                h->pred8x8  [HOR_PRED8x8      ] = ff_pred8x8_horizontal_mmx;
+                h->pred8x8  [VERT_PRED8x8     ] = ff_pred8x8_vertical_8_mmx;
+                h->pred8x8  [HOR_PRED8x8      ] = ff_pred8x8_horizontal_8_mmx;
             }
             if (codec_id == AV_CODEC_ID_VP8) {
-                h->pred16x16[PLANE_PRED8x8    ] = ff_pred16x16_tm_vp8_mmx;
-                h->pred8x8  [PLANE_PRED8x8    ] = ff_pred8x8_tm_vp8_mmx;
-                h->pred4x4  [TM_VP8_PRED      ] = ff_pred4x4_tm_vp8_mmx;
+                h->pred16x16[PLANE_PRED8x8    ] = ff_pred16x16_tm_vp8_8_mmx;
+                h->pred8x8  [PLANE_PRED8x8    ] = ff_pred8x8_tm_vp8_8_mmx;
+                h->pred4x4  [TM_VP8_PRED      ] = ff_pred4x4_tm_vp8_8_mmx;
             } else {
                 if (chroma_format_idc == 1)
-                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_mmx;
+                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_8_mmx;
                 if (codec_id == AV_CODEC_ID_SVQ3) {
                     if (mm_flags & AV_CPU_FLAG_CMOV)
-                        h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_svq3_mmx;
+                        h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_svq3_8_mmx;
                 } else if (codec_id == AV_CODEC_ID_RV40) {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_rv40_mmx;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_rv40_8_mmx;
                 } else {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_h264_mmx;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_h264_8_mmx;
                 }
             }
         }
 
         if (EXTERNAL_MMXEXT(mm_flags)) {
-            h->pred16x16[HOR_PRED8x8            ] = ff_pred16x16_horizontal_mmx2;
-            h->pred16x16[DC_PRED8x8             ] = ff_pred16x16_dc_mmx2;
+            h->pred16x16[HOR_PRED8x8            ] = ff_pred16x16_horizontal_8_mmx2;
+            h->pred16x16[DC_PRED8x8             ] = ff_pred16x16_dc_8_mmx2;
             if (chroma_format_idc == 1)
-                h->pred8x8[HOR_PRED8x8          ] = ff_pred8x8_horizontal_mmx2;
-            h->pred8x8l [TOP_DC_PRED            ] = ff_pred8x8l_top_dc_mmxext;
-            h->pred8x8l [DC_PRED                ] = ff_pred8x8l_dc_mmxext;
-            h->pred8x8l [HOR_PRED               ] = ff_pred8x8l_horizontal_mmxext;
-            h->pred8x8l [VERT_PRED              ] = ff_pred8x8l_vertical_mmxext;
-            h->pred8x8l [DIAG_DOWN_RIGHT_PRED   ] = ff_pred8x8l_down_right_mmxext;
-            h->pred8x8l [VERT_RIGHT_PRED        ] = ff_pred8x8l_vertical_right_mmxext;
-            h->pred8x8l [HOR_UP_PRED            ] = ff_pred8x8l_horizontal_up_mmxext;
-            h->pred8x8l [DIAG_DOWN_LEFT_PRED    ] = ff_pred8x8l_down_left_mmxext;
-            h->pred8x8l [HOR_DOWN_PRED          ] = ff_pred8x8l_horizontal_down_mmxext;
-            h->pred4x4  [DIAG_DOWN_RIGHT_PRED   ] = ff_pred4x4_down_right_mmxext;
-            h->pred4x4  [VERT_RIGHT_PRED        ] = ff_pred4x4_vertical_right_mmxext;
-            h->pred4x4  [HOR_DOWN_PRED          ] = ff_pred4x4_horizontal_down_mmxext;
-            h->pred4x4  [DC_PRED                ] = ff_pred4x4_dc_mmxext;
+                h->pred8x8[HOR_PRED8x8          ] = ff_pred8x8_horizontal_8_mmx2;
+            h->pred8x8l [TOP_DC_PRED            ] = ff_pred8x8l_top_dc_8_mmxext;
+            h->pred8x8l [DC_PRED                ] = ff_pred8x8l_dc_8_mmxext;
+            h->pred8x8l [HOR_PRED               ] = ff_pred8x8l_horizontal_8_mmxext;
+            h->pred8x8l [VERT_PRED              ] = ff_pred8x8l_vertical_8_mmxext;
+            h->pred8x8l [DIAG_DOWN_RIGHT_PRED   ] = ff_pred8x8l_down_right_8_mmxext;
+            h->pred8x8l [VERT_RIGHT_PRED        ] = ff_pred8x8l_vertical_right_8_mmxext;
+            h->pred8x8l [HOR_UP_PRED            ] = ff_pred8x8l_horizontal_up_8_mmxext;
+            h->pred8x8l [DIAG_DOWN_LEFT_PRED    ] = ff_pred8x8l_down_left_8_mmxext;
+            h->pred8x8l [HOR_DOWN_PRED          ] = ff_pred8x8l_horizontal_down_8_mmxext;
+            h->pred4x4  [DIAG_DOWN_RIGHT_PRED   ] = ff_pred4x4_down_right_8_mmxext;
+            h->pred4x4  [VERT_RIGHT_PRED        ] = ff_pred4x4_vertical_right_8_mmxext;
+            h->pred4x4  [HOR_DOWN_PRED          ] = ff_pred4x4_horizontal_down_8_mmxext;
+            h->pred4x4  [DC_PRED                ] = ff_pred4x4_dc_8_mmxext;
             if (codec_id == AV_CODEC_ID_VP8 || codec_id == AV_CODEC_ID_H264) {
-                h->pred4x4  [DIAG_DOWN_LEFT_PRED] = ff_pred4x4_down_left_mmxext;
+                h->pred4x4  [DIAG_DOWN_LEFT_PRED] = ff_pred4x4_down_left_8_mmxext;
             }
             if (codec_id == AV_CODEC_ID_SVQ3 || codec_id == AV_CODEC_ID_H264) {
-                h->pred4x4  [VERT_LEFT_PRED     ] = ff_pred4x4_vertical_left_mmxext;
+                h->pred4x4  [VERT_LEFT_PRED     ] = ff_pred4x4_vertical_left_8_mmxext;
             }
             if (codec_id != AV_CODEC_ID_RV40) {
-                h->pred4x4  [HOR_UP_PRED        ] = ff_pred4x4_horizontal_up_mmxext;
+                h->pred4x4  [HOR_UP_PRED        ] = ff_pred4x4_horizontal_up_8_mmxext;
             }
             if (codec_id == AV_CODEC_ID_SVQ3 || codec_id == AV_CODEC_ID_H264) {
                 if (chroma_format_idc == 1) {
-                    h->pred8x8[TOP_DC_PRED8x8   ] = ff_pred8x8_top_dc_mmxext;
-                    h->pred8x8[DC_PRED8x8       ] = ff_pred8x8_dc_mmxext;
+                    h->pred8x8[TOP_DC_PRED8x8   ] = ff_pred8x8_top_dc_8_mmxext;
+                    h->pred8x8[DC_PRED8x8       ] = ff_pred8x8_dc_8_mmxext;
                 }
             }
             if (codec_id == AV_CODEC_ID_VP8) {
-                h->pred16x16[PLANE_PRED8x8      ] = ff_pred16x16_tm_vp8_mmx2;
-                h->pred8x8  [DC_PRED8x8         ] = ff_pred8x8_dc_rv40_mmxext;
-                h->pred8x8  [PLANE_PRED8x8      ] = ff_pred8x8_tm_vp8_mmx2;
-                h->pred4x4  [TM_VP8_PRED        ] = ff_pred4x4_tm_vp8_mmx2;
-                h->pred4x4  [VERT_PRED          ] = ff_pred4x4_vertical_vp8_mmxext;
+                h->pred16x16[PLANE_PRED8x8      ] = ff_pred16x16_tm_vp8_8_mmx2;
+                h->pred8x8  [DC_PRED8x8         ] = ff_pred8x8_dc_rv40_8_mmxext;
+                h->pred8x8  [PLANE_PRED8x8      ] = ff_pred8x8_tm_vp8_8_mmx2;
+                h->pred4x4  [TM_VP8_PRED        ] = ff_pred4x4_tm_vp8_8_mmx2;
+                h->pred4x4  [VERT_PRED          ] = ff_pred4x4_vertical_vp8_8_mmxext;
             } else {
                 if (chroma_format_idc == 1)
-                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_mmx2;
+                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_8_mmx2;
                 if (codec_id == AV_CODEC_ID_SVQ3) {
-                    h->pred16x16[PLANE_PRED8x8  ] = ff_pred16x16_plane_svq3_mmx2;
+                    h->pred16x16[PLANE_PRED8x8  ] = ff_pred16x16_plane_svq3_8_mmx2;
                 } else if (codec_id == AV_CODEC_ID_RV40) {
-                    h->pred16x16[PLANE_PRED8x8  ] = ff_pred16x16_plane_rv40_mmx2;
+                    h->pred16x16[PLANE_PRED8x8  ] = ff_pred16x16_plane_rv40_8_mmx2;
                 } else {
-                    h->pred16x16[PLANE_PRED8x8  ] = ff_pred16x16_plane_h264_mmx2;
+                    h->pred16x16[PLANE_PRED8x8  ] = ff_pred16x16_plane_h264_8_mmx2;
                 }
             }
         }
 
         if (EXTERNAL_SSE(mm_flags)) {
-            h->pred16x16[VERT_PRED8x8] = ff_pred16x16_vertical_sse;
+            h->pred16x16[VERT_PRED8x8] = ff_pred16x16_vertical_8_sse;
         }
 
         if (EXTERNAL_SSE2(mm_flags)) {
-            h->pred16x16[DC_PRED8x8           ] = ff_pred16x16_dc_sse2;
-            h->pred8x8l [DIAG_DOWN_LEFT_PRED  ] = ff_pred8x8l_down_left_sse2;
-            h->pred8x8l [DIAG_DOWN_RIGHT_PRED ] = ff_pred8x8l_down_right_sse2;
-            h->pred8x8l [VERT_RIGHT_PRED      ] = ff_pred8x8l_vertical_right_sse2;
-            h->pred8x8l [VERT_LEFT_PRED       ] = ff_pred8x8l_vertical_left_sse2;
-            h->pred8x8l [HOR_DOWN_PRED        ] = ff_pred8x8l_horizontal_down_sse2;
+            h->pred16x16[DC_PRED8x8           ] = ff_pred16x16_dc_8_sse2;
+            h->pred8x8l [DIAG_DOWN_LEFT_PRED  ] = ff_pred8x8l_down_left_8_sse2;
+            h->pred8x8l [DIAG_DOWN_RIGHT_PRED ] = ff_pred8x8l_down_right_8_sse2;
+            h->pred8x8l [VERT_RIGHT_PRED      ] = ff_pred8x8l_vertical_right_8_sse2;
+            h->pred8x8l [VERT_LEFT_PRED       ] = ff_pred8x8l_vertical_left_8_sse2;
+            h->pred8x8l [HOR_DOWN_PRED        ] = ff_pred8x8l_horizontal_down_8_sse2;
             if (codec_id == AV_CODEC_ID_VP8) {
-                h->pred16x16[PLANE_PRED8x8    ] = ff_pred16x16_tm_vp8_sse2;
-                h->pred8x8  [PLANE_PRED8x8    ] = ff_pred8x8_tm_vp8_sse2;
+                h->pred16x16[PLANE_PRED8x8    ] = ff_pred16x16_tm_vp8_8_sse2;
+                h->pred8x8  [PLANE_PRED8x8    ] = ff_pred8x8_tm_vp8_8_sse2;
             } else {
                 if (chroma_format_idc == 1)
-                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_sse2;
+                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_8_sse2;
                 if (codec_id == AV_CODEC_ID_SVQ3) {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_svq3_sse2;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_svq3_8_sse2;
                 } else if (codec_id == AV_CODEC_ID_RV40) {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_rv40_sse2;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_rv40_8_sse2;
                 } else {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_h264_sse2;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_h264_8_sse2;
                 }
             }
         }
 
         if (EXTERNAL_SSSE3(mm_flags)) {
-            h->pred16x16[HOR_PRED8x8          ] = ff_pred16x16_horizontal_ssse3;
-            h->pred16x16[DC_PRED8x8           ] = ff_pred16x16_dc_ssse3;
+            h->pred16x16[HOR_PRED8x8          ] = ff_pred16x16_horizontal_8_ssse3;
+            h->pred16x16[DC_PRED8x8           ] = ff_pred16x16_dc_8_ssse3;
             if (chroma_format_idc == 1)
-                h->pred8x8  [HOR_PRED8x8      ] = ff_pred8x8_horizontal_ssse3;
-            h->pred8x8l [TOP_DC_PRED          ] = ff_pred8x8l_top_dc_ssse3;
-            h->pred8x8l [DC_PRED              ] = ff_pred8x8l_dc_ssse3;
-            h->pred8x8l [HOR_PRED             ] = ff_pred8x8l_horizontal_ssse3;
-            h->pred8x8l [VERT_PRED            ] = ff_pred8x8l_vertical_ssse3;
-            h->pred8x8l [DIAG_DOWN_LEFT_PRED  ] = ff_pred8x8l_down_left_ssse3;
-            h->pred8x8l [DIAG_DOWN_RIGHT_PRED ] = ff_pred8x8l_down_right_ssse3;
-            h->pred8x8l [VERT_RIGHT_PRED      ] = ff_pred8x8l_vertical_right_ssse3;
-            h->pred8x8l [VERT_LEFT_PRED       ] = ff_pred8x8l_vertical_left_ssse3;
-            h->pred8x8l [HOR_UP_PRED          ] = ff_pred8x8l_horizontal_up_ssse3;
-            h->pred8x8l [HOR_DOWN_PRED        ] = ff_pred8x8l_horizontal_down_ssse3;
+                h->pred8x8  [HOR_PRED8x8      ] = ff_pred8x8_horizontal_8_ssse3;
+            h->pred8x8l [TOP_DC_PRED          ] = ff_pred8x8l_top_dc_8_ssse3;
+            h->pred8x8l [DC_PRED              ] = ff_pred8x8l_dc_8_ssse3;
+            h->pred8x8l [HOR_PRED             ] = ff_pred8x8l_horizontal_8_ssse3;
+            h->pred8x8l [VERT_PRED            ] = ff_pred8x8l_vertical_8_ssse3;
+            h->pred8x8l [DIAG_DOWN_LEFT_PRED  ] = ff_pred8x8l_down_left_8_ssse3;
+            h->pred8x8l [DIAG_DOWN_RIGHT_PRED ] = ff_pred8x8l_down_right_8_ssse3;
+            h->pred8x8l [VERT_RIGHT_PRED      ] = ff_pred8x8l_vertical_right_8_ssse3;
+            h->pred8x8l [VERT_LEFT_PRED       ] = ff_pred8x8l_vertical_left_8_ssse3;
+            h->pred8x8l [HOR_UP_PRED          ] = ff_pred8x8l_horizontal_up_8_ssse3;
+            h->pred8x8l [HOR_DOWN_PRED        ] = ff_pred8x8l_horizontal_down_8_ssse3;
             if (codec_id == AV_CODEC_ID_VP8) {
-                h->pred8x8  [PLANE_PRED8x8    ] = ff_pred8x8_tm_vp8_ssse3;
-                h->pred4x4  [TM_VP8_PRED      ] = ff_pred4x4_tm_vp8_ssse3;
+                h->pred8x8  [PLANE_PRED8x8    ] = ff_pred8x8_tm_vp8_8_ssse3;
+                h->pred4x4  [TM_VP8_PRED      ] = ff_pred4x4_tm_vp8_8_ssse3;
             } else {
                 if (chroma_format_idc == 1)
-                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_ssse3;
+                    h->pred8x8  [PLANE_PRED8x8] = ff_pred8x8_plane_8_ssse3;
                 if (codec_id == AV_CODEC_ID_SVQ3) {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_svq3_ssse3;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_svq3_8_ssse3;
                 } else if (codec_id == AV_CODEC_ID_RV40) {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_rv40_ssse3;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_rv40_8_ssse3;
                 } else {
-                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_h264_ssse3;
+                    h->pred16x16[PLANE_PRED8x8] = ff_pred16x16_plane_h264_8_ssse3;
                 }
             }
         }