Supporting high precision 1/8-pel motion vectors

This is the initial patch for supporting 1/8th pel
motion. Currently if we configure with enable-high-precision-mv,
all motion vectors would default to 1/8 pel. Encode and
decode syncs fine with the current code. In the next phase
the code will be refactored so that we can choose the 1/8
pel mode adaptively at a frame/segment/mb level.

Derf results:
http://www.corp.google.com/~debargha/vp8_results/enhinterp_hpmv.html
(about 0.83% better than 8-tap interpoaltion)

Patch 3: Rebased. Also adding 1/16th pel interpolation for U and V

Patch 4: HD results.
http://www.corp.google.com/~debargha/vp8_results/enhinterp_hd_hpmv.html
Seems impressive (unless I am doing something wrong).

Patch 5: Added mmx/sse for bilateral filtering, as well as enforced
use of c-versions of subpel filters with 8-taps and 1/16th pel;
Also redesigned the 8-tap filters to reduce the cut-off in order to
introduce a denoising effect. There is a new configure option
sixteenth-subpel-uv which will use 1/16 th pel interpolation for
uv, if the motion vectors have 1/8 pel accuracy.

With the fixes the results are promising on the derf set. The enhanced
interpolation option with 8-taps alone gives 3% improvement over thei
derf set:
http://www.corp.google.com/~debargha/vp8_results/enhinterpn.html

Results on high precision mv and on the hd set are to follow.

Patch 6: Adding a missing condition for CONFIG_SIXTEENTH_SUBPEL_UV in
vp8/common/x86/x86_systemdependent.c

Patch 7: Cleaning up various debug messages.

Patch 8: Merge conflict

Change-Id: I5b1d844457aefd7414a9e4e0e06c6ed38fd8cc04

1 files changed, 37 insertions, 2 deletions

diff --git a/vp8/encoder/x86/variance_mmx.c b/vp8/encoder/x86/variance_mmx.c
index 92b695f17..b84d00034 100644
--- a/vp8/encoder/x86/variance_mmx.c
+++ b/vp8/encoder/x86/variance_mmx.c
@@ -204,6 +204,27 @@ unsigned int vp8_variance8x16_mmx(
 // the mmx function that does the bilinear filtering and var calculation //
 // int one pass                                                          //
 ///////////////////////////////////////////////////////////////////////////
+#if CONFIG_SIXTEENTH_SUBPEL_UV
+DECLARE_ALIGNED(16, const short, vp8_vp7_bilinear_filters_mmx[16][8]) =
+{
+    { 128, 128, 128, 128,  0,  0,  0,  0 },
+    { 120, 120, 120, 120,  8,  8,  8,  8 },
+    { 112, 112, 112, 112, 16, 16, 16, 16 },
+    { 104, 104, 104, 104, 24, 24, 24, 24 },
+    {  96, 96, 96, 96, 32, 32, 32, 32 },
+    {  88, 88, 88, 88, 40, 40, 40, 40 },
+    {  80, 80, 80, 80, 48, 48, 48, 48 },
+    {  72, 72, 72, 72, 56, 56, 56, 56 },
+    {  64, 64, 64, 64, 64, 64, 64, 64 },
+    {  56, 56, 56, 56, 72, 72, 72, 72 },
+    {  48, 48, 48, 48, 80, 80, 80, 80 },
+    {  40, 40, 40, 40, 88, 88, 88, 88 },
+    {  32, 32, 32, 32, 96, 96, 96, 96 },
+    {  24, 24, 24, 24, 104, 104, 104, 104 },
+    {  16, 16, 16, 16, 112, 112, 112, 112 },
+    {   8,  8,  8,  8, 120, 120, 120, 120 }
+};
+#else
 DECLARE_ALIGNED(16, const short, vp8_vp7_bilinear_filters_mmx[8][8]) =
 {
     { 128, 128, 128, 128,  0,  0,  0,  0 },
@@ -215,6 +236,7 @@ DECLARE_ALIGNED(16, const short, vp8_vp7_bilinear_filters_mmx[8][8]) =
     {  32, 32, 32, 32, 96, 96, 96, 96 },
     {  16, 16, 16, 16, 112, 112, 112, 112 }
 };
+#endif
 
 unsigned int vp8_sub_pixel_variance4x4_mmx
 (
@@ -279,7 +301,6 @@ unsigned int vp8_sub_pixel_variance16x16_mmx
     int xsum0, xsum1;
     unsigned int xxsum0, xxsum1;
 
-
     vp8_filter_block2d_bil_var_mmx(
         src_ptr, src_pixels_per_line,
         dst_ptr, dst_pixels_per_line, 16,
@@ -287,7 +308,6 @@ unsigned int vp8_sub_pixel_variance16x16_mmx
         &xsum0, &xxsum0
     );
 
-
     vp8_filter_block2d_bil_var_mmx(
         src_ptr + 8, src_pixels_per_line,
         dst_ptr + 8, dst_pixels_per_line, 16,
@@ -386,8 +406,13 @@ unsigned int vp8_variance_halfpixvar16x16_h_mmx(
     int  recon_stride,
     unsigned int *sse)
 {
+#if CONFIG_SIXTEENTH_SUBPEL_UV
+    return vp8_sub_pixel_variance16x16_mmx(src_ptr, source_stride, 8, 0,
+                                           ref_ptr, recon_stride, sse);
+#else
     return vp8_sub_pixel_variance16x16_mmx(src_ptr, source_stride, 4, 0,
                                            ref_ptr, recon_stride, sse);
+#endif
 }
 
 
@@ -398,8 +423,13 @@ unsigned int vp8_variance_halfpixvar16x16_v_mmx(
     int  recon_stride,
     unsigned int *sse)
 {
+#if CONFIG_SIXTEENTH_SUBPEL_UV
+    return vp8_sub_pixel_variance16x16_mmx(src_ptr, source_stride, 0, 8,
+                                           ref_ptr, recon_stride, sse);
+#else
     return vp8_sub_pixel_variance16x16_mmx(src_ptr, source_stride, 0, 4,
                                            ref_ptr, recon_stride, sse);
+#endif
 }
 
 
@@ -410,6 +440,11 @@ unsigned int vp8_variance_halfpixvar16x16_hv_mmx(
     int  recon_stride,
     unsigned int *sse)
 {
+#if CONFIG_SIXTEENTH_SUBPEL_UV
+    return vp8_sub_pixel_variance16x16_mmx(src_ptr, source_stride, 8, 8,
+                                           ref_ptr, recon_stride, sse);
+#else
     return vp8_sub_pixel_variance16x16_mmx(src_ptr, source_stride, 4, 4,
                                            ref_ptr, recon_stride, sse);
+#endif
 }