Merge "Adding sse2 variant for vp9_mse{8x8, 8x16, 16x8}."

vp9/common/vp9_rtcd_defs.pl

@@ -693,16 +693,16 @@ add_proto qw/void vp9_sad4x4x4d/, "const uint8_t *src_ptr, int  src_stride, cons
 specialize qw/vp9_sad4x4x4d sse/;
 
 add_proto qw/unsigned int vp9_mse16x16/, "const uint8_t *src_ptr, int  source_stride, const uint8_t *ref_ptr, int  recon_stride, unsigned int *sse";
-specialize qw/vp9_mse16x16 avx2/, "$sse2_x86inc";
+specialize qw/vp9_mse16x16 sse2 avx2/;
 
 add_proto qw/unsigned int vp9_mse8x16/, "const uint8_t *src_ptr, int  source_stride, const uint8_t *ref_ptr, int  recon_stride, unsigned int *sse";
-specialize qw/vp9_mse8x16/;
+specialize qw/vp9_mse8x16 sse2/;
 
 add_proto qw/unsigned int vp9_mse16x8/, "const uint8_t *src_ptr, int  source_stride, const uint8_t *ref_ptr, int  recon_stride, unsigned int *sse";
-specialize qw/vp9_mse16x8/;
+specialize qw/vp9_mse16x8 sse2/;
 
 add_proto qw/unsigned int vp9_mse8x8/, "const uint8_t *src_ptr, int  source_stride, const uint8_t *ref_ptr, int  recon_stride, unsigned int *sse";
-specialize qw/vp9_mse8x8/;
+specialize qw/vp9_mse8x8 sse2/;
 
 add_proto qw/unsigned int vp9_get_mb_ss/, "const int16_t *";
 specialize qw/vp9_get_mb_ss sse2/;

vp9/encoder/x86/vp9_variance_sse2.c

@@ -217,14 +217,6 @@ unsigned int vp9_variance16x16_sse2(const unsigned char *src, int src_stride,
   return *sse - (((unsigned int)sum * sum) >> 8);
 }
 
-unsigned int vp9_mse16x16_sse2(const unsigned char *src, int src_stride,
-                               const unsigned char *ref, int ref_stride,
-                               unsigned int *sse) {
-  int sum;
-  vp9_get16x16var_sse2(src, src_stride, ref, ref_stride, sse, &sum);
-  return *sse;
-}
-
 unsigned int vp9_variance32x32_sse2(const uint8_t *src, int src_stride,
                                     const uint8_t *ref, int ref_stride,
                                     unsigned int *sse) {
@@ -279,6 +271,34 @@ unsigned int vp9_variance32x64_sse2(const uint8_t *src, int src_stride,
   return *sse - (((int64_t)sum * sum) >> 11);
 }
 
+unsigned int vp9_mse8x8_sse2(const uint8_t *src, int src_stride,
+                             const uint8_t *ref, int ref_stride,
+                             unsigned int *sse) {
+  vp9_variance8x8_sse2(src, src_stride, ref, ref_stride, sse);
+  return *sse;
+}
+
+unsigned int vp9_mse8x16_sse2(const uint8_t *src, int src_stride,
+                              const uint8_t *ref, int ref_stride,
+                              unsigned int *sse) {
+  vp9_variance8x16_sse2(src, src_stride, ref, ref_stride, sse);
+  return *sse;
+}
+
+unsigned int vp9_mse16x8_sse2(const uint8_t *src, int src_stride,
+                              const uint8_t *ref, int ref_stride,
+                              unsigned int *sse) {
+  vp9_variance16x8_sse2(src, src_stride, ref, ref_stride, sse);
+  return *sse;
+}
+
+unsigned int vp9_mse16x16_sse2(const uint8_t *src, int src_stride,
+                               const uint8_t *ref, int ref_stride,
+                               unsigned int *sse) {
+  vp9_variance16x16_sse2(src, src_stride, ref, ref_stride, sse);
+  return *sse;
+}
+
 #define DECL(w, opt) \
 int vp9_sub_pixel_variance##w##xh_##opt(const uint8_t *src, \
                                         ptrdiff_t src_stride, \