libvpx: Pull from upstream

source/libvpx/vpx_dsp/vpx_dsp_rtcd_defs.pl

 sub vpx_dsp_forward_decls() {
 print <<EOF
 /*
  * DSP
  */
 
 #include "vpx/vpx_integer.h"
 #include "vpx_dsp/vpx_dsp_common.h"
 
 EOF
@@ skipping 36 matching lines @@
   $sse2_x86_64 = 'sse2';
   $ssse3_x86_64 = 'ssse3';
   $avx_x86_64 = 'avx';
   $avx2_x86_64 = 'avx2';
 }
 
 #
 # Intra prediction
 #
 
-if (vpx_config("CONFIG_VP9") eq "yes") {
+if ((vpx_config("CONFIG_VP9") eq "yes") || (vpx_config("CONFIG_VP10") eq "yes")) {
   add_proto qw/void vpx_d207_predictor_4x4/, "uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left";
   specialize qw/vpx_d207_predictor_4x4/, "$ssse3_x86inc";
 
   add_proto qw/void vpx_d45_predictor_4x4/, "uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left";
   specialize qw/vpx_d45_predictor_4x4 neon/, "$ssse3_x86inc";
 
   add_proto qw/void vpx_d63_predictor_4x4/, "uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left";
   specialize qw/vpx_d63_predictor_4x4/, "$ssse3_x86inc";
 
   add_proto qw/void vpx_h_predictor_4x4/, "uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left";
@@ skipping 294 matching lines @@
 
     add_proto qw/void vpx_highbd_dc_top_predictor_32x32/, "uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd";
     specialize qw/vpx_highbd_dc_top_predictor_32x32/;
 
     add_proto qw/void vpx_highbd_dc_left_predictor_32x32/, "uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd";
     specialize qw/vpx_highbd_dc_left_predictor_32x32/;
 
     add_proto qw/void vpx_highbd_dc_128_predictor_32x32/, "uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd";
     specialize qw/vpx_highbd_dc_128_predictor_32x32/;
   }  # CONFIG_VP9_HIGHBITDEPTH
-}  # CONFIG_VP9
+}  # CONFIG_VP9 || CONFIG_VP10
 
 #
 # Sub Pixel Filters
 #
 add_proto qw/void vpx_convolve_copy/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve_copy neon dspr2 msa/, "$sse2_x86inc";
 
 add_proto qw/void vpx_convolve_avg/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve_avg neon dspr2 msa/, "$sse2_x86inc";
 
 add_proto qw/void vpx_convolve8/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve8 sse2 ssse3 neon dspr2 msa/, "$avx2_ssse3";
 
 add_proto qw/void vpx_convolve8_horiz/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve8_horiz sse2 ssse3 neon dspr2 msa/, "$avx2_ssse3";
 
 add_proto qw/void vpx_convolve8_vert/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve8_vert sse2 ssse3 neon dspr2 msa/, "$avx2_ssse3";
 
 add_proto qw/void vpx_convolve8_avg/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve8_avg sse2 ssse3 neon dspr2 msa/;
 
 add_proto qw/void vpx_convolve8_avg_horiz/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve8_avg_horiz sse2 ssse3 neon dspr2 msa/;
 
 add_proto qw/void vpx_convolve8_avg_vert/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
 specialize qw/vpx_convolve8_avg_vert sse2 ssse3 neon dspr2 msa/;
 
+add_proto qw/void vpx_scaled_2d/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
+specialize qw/vpx_scaled_2d ssse3/;
+
+add_proto qw/void vpx_scaled_horiz/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
+specialize qw/vpx_scaled_horiz/;
+
+add_proto qw/void vpx_scaled_vert/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
+specialize qw/vpx_scaled_vert/;
+
+add_proto qw/void vpx_scaled_avg_2d/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
+specialize qw/vpx_scaled_avg_2d/;
+
+add_proto qw/void vpx_scaled_avg_horiz/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
+specialize qw/vpx_scaled_avg_horiz/;
+
+add_proto qw/void vpx_scaled_avg_vert/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h";
+specialize qw/vpx_scaled_avg_vert/;
+
 if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
   #
   # Sub Pixel Filters
   #
   add_proto qw/void vpx_highbd_convolve_copy/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps";
   specialize qw/vpx_highbd_convolve_copy/;
 
   add_proto qw/void vpx_highbd_convolve_avg/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps";
   specialize qw/vpx_highbd_convolve_avg/;
 
@@ skipping 13 matching lines @@
   specialize qw/vpx_highbd_convolve8_avg_horiz/, "$sse2_x86_64";
 
   add_proto qw/void vpx_highbd_convolve8_avg_vert/, "const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps";
   specialize qw/vpx_highbd_convolve8_avg_vert/, "$sse2_x86_64";
 }  # CONFIG_VP9_HIGHBITDEPTH
 
 #
 # Loopfilter
 #
 add_proto qw/void vpx_lpf_vertical_16/, "uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh";
-specialize qw/vpx_lpf_vertical_16 sse2 neon_asm msa/;
+specialize qw/vpx_lpf_vertical_16 sse2 neon_asm dspr2 msa/;
 $vpx_lpf_vertical_16_neon_asm=vpx_lpf_vertical_16_neon;
 
 add_proto qw/void vpx_lpf_vertical_16_dual/, "uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh";
-specialize qw/vpx_lpf_vertical_16_dual sse2 neon_asm msa/;
+specialize qw/vpx_lpf_vertical_16_dual sse2 neon_asm dspr2 msa/;
 $vpx_lpf_vertical_16_dual_neon_asm=vpx_lpf_vertical_16_dual_neon;
 
 add_proto qw/void vpx_lpf_vertical_8/, "uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count";
-specialize qw/vpx_lpf_vertical_8 sse2 neon msa/;
+specialize qw/vpx_lpf_vertical_8 sse2 neon dspr2 msa/;
 
 add_proto qw/void vpx_lpf_vertical_8_dual/, "uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1";
-specialize qw/vpx_lpf_vertical_8_dual sse2 neon_asm msa/;
+specialize qw/vpx_lpf_vertical_8_dual sse2 neon_asm dspr2 msa/;
 $vpx_lpf_vertical_8_dual_neon_asm=vpx_lpf_vertical_8_dual_neon;
 
 add_proto qw/void vpx_lpf_vertical_4/, "uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count";
-specialize qw/vpx_lpf_vertical_4 mmx neon msa/;
+specialize qw/vpx_lpf_vertical_4 mmx neon dspr2 msa/;
 
 add_proto qw/void vpx_lpf_vertical_4_dual/, "uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1";
-specialize qw/vpx_lpf_vertical_4_dual sse2 neon msa/;
+specialize qw/vpx_lpf_vertical_4_dual sse2 neon dspr2 msa/;
 
 add_proto qw/void vpx_lpf_horizontal_16/, "uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count";
-specialize qw/vpx_lpf_horizontal_16 sse2 avx2 neon_asm msa/;
+specialize qw/vpx_lpf_horizontal_16 sse2 avx2 neon_asm dspr2 msa/;
 $vpx_lpf_horizontal_16_neon_asm=vpx_lpf_horizontal_16_neon;
 
 add_proto qw/void vpx_lpf_horizontal_8/, "uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count";
-specialize qw/vpx_lpf_horizontal_8 sse2 neon msa/;
+specialize qw/vpx_lpf_horizontal_8 sse2 neon dspr2 msa/;
 
 add_proto qw/void vpx_lpf_horizontal_8_dual/, "uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1";
-specialize qw/vpx_lpf_horizontal_8_dual sse2 neon_asm msa/;
+specialize qw/vpx_lpf_horizontal_8_dual sse2 neon_asm dspr2 msa/;
 $vpx_lpf_horizontal_8_dual_neon_asm=vpx_lpf_horizontal_8_dual_neon;
 
 add_proto qw/void vpx_lpf_horizontal_4/, "uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count";
-specialize qw/vpx_lpf_horizontal_4 mmx neon msa/;
+specialize qw/vpx_lpf_horizontal_4 mmx neon dspr2 msa/;
 
 add_proto qw/void vpx_lpf_horizontal_4_dual/, "uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1";
-specialize qw/vpx_lpf_horizontal_4_dual sse2 neon msa/;
+specialize qw/vpx_lpf_horizontal_4_dual sse2 neon dspr2 msa/;
 
 if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
   add_proto qw/void vpx_highbd_lpf_vertical_16/, "uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd";
   specialize qw/vpx_highbd_lpf_vertical_16 sse2/;
 
   add_proto qw/void vpx_highbd_lpf_vertical_16_dual/, "uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd";
   specialize qw/vpx_highbd_lpf_vertical_16_dual sse2/;
 
   add_proto qw/void vpx_highbd_lpf_vertical_8/, "uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count, int bd";
   specialize qw/vpx_highbd_lpf_vertical_8 sse2/;
@@ skipping 23 matching lines @@
   specialize qw/vpx_highbd_lpf_horizontal_4_dual sse2/;
 }  # CONFIG_VP9_HIGHBITDEPTH
 
 #
 # Encoder functions.
 #
 
 #
 # Forward transform
 #
-if (vpx_config("CONFIG_VP9_ENCODER") eq "yes") {
+if ((vpx_config("CONFIG_VP9_ENCODER") eq "yes") || (vpx_config("CONFIG_VP10_ENCODER") eq "yes")) {
 if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
   add_proto qw/void vpx_fdct4x4/, "const int16_t *input, tran_low_t *output, int stride";
   specialize qw/vpx_fdct4x4 sse2/;
 
   add_proto qw/void vpx_fdct4x4_1/, "const int16_t *input, tran_low_t *output, int stride";
   specialize qw/vpx_fdct4x4_1 sse2/;
 
   add_proto qw/void vpx_fdct8x8/, "const int16_t *input, tran_low_t *output, int stride";
   specialize qw/vpx_fdct8x8 sse2/;
 
@@ skipping 59 matching lines @@
 
   add_proto qw/void vpx_fdct32x32/, "const int16_t *input, tran_low_t *output, int stride";
   specialize qw/vpx_fdct32x32 sse2 avx2 msa/;
 
   add_proto qw/void vpx_fdct32x32_rd/, "const int16_t *input, tran_low_t *output, int stride";
   specialize qw/vpx_fdct32x32_rd sse2 avx2 msa/;
 
   add_proto qw/void vpx_fdct32x32_1/, "const int16_t *input, tran_low_t *output, int stride";
   specialize qw/vpx_fdct32x32_1 sse2 msa/;
 }  # CONFIG_VP9_HIGHBITDEPTH
-}  # CONFIG_VP9_ENCODER
+}  # CONFIG_VP9_ENCODER || CONFIG_VP10_ENCODER
 
 #
 # Inverse transform
-if (vpx_config("CONFIG_VP9") eq "yes") {
+if ((vpx_config("CONFIG_VP9") eq "yes") || (vpx_config("CONFIG_VP10") eq "yes")) {
 if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
   # Note as optimized versions of these functions are added we need to add a check to ensure
   # that when CONFIG_EMULATE_HARDWARE is on, it defaults to the C versions only.
-  add_proto qw/void vp9_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct4x4_1_add/;
-
-  add_proto qw/void vp9_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct4x4_16_add/;
-
-  add_proto qw/void vp9_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct8x8_1_add/;
-
-  add_proto qw/void vp9_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct8x8_64_add/;
-
-  add_proto qw/void vp9_idct8x8_12_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct8x8_12_add/;
-
-  add_proto qw/void vp9_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct16x16_1_add/;
-
-  add_proto qw/void vp9_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct16x16_256_add/;
-
-  add_proto qw/void vp9_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct16x16_10_add/;
-
-  add_proto qw/void vp9_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct32x32_1024_add/;
-
-  add_proto qw/void vp9_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct32x32_34_add/;
-
-  add_proto qw/void vp9_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_idct32x32_1_add/;
-
-  add_proto qw/void vp9_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_iwht4x4_1_add/;
-
-  add_proto qw/void vp9_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-  specialize qw/vp9_iwht4x4_16_add/;
-
-  add_proto qw/void vp9_highbd_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_idct4x4_1_add/;
-
-  add_proto qw/void vp9_highbd_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_idct8x8_1_add/;
-
-  add_proto qw/void vp9_highbd_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_idct16x16_1_add/;
-
-  add_proto qw/void vp9_highbd_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_idct32x32_1024_add/;
-
-  add_proto qw/void vp9_highbd_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_idct32x32_34_add/;
-
-  add_proto qw/void vp9_highbd_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_idct32x32_1_add/;
-
-  add_proto qw/void vp9_highbd_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_iwht4x4_1_add/;
-
-  add_proto qw/void vp9_highbd_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-  specialize qw/vp9_highbd_iwht4x4_16_add/;
+  add_proto qw/void vpx_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct4x4_1_add/;
+
+  add_proto qw/void vpx_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct4x4_16_add/;
+
+  add_proto qw/void vpx_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct8x8_1_add/;
+
+  add_proto qw/void vpx_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct8x8_64_add/;
+
+  add_proto qw/void vpx_idct8x8_12_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct8x8_12_add/;
+
+  add_proto qw/void vpx_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct16x16_1_add/;
+
+  add_proto qw/void vpx_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct16x16_256_add/;
+
+  add_proto qw/void vpx_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct16x16_10_add/;
+
+  add_proto qw/void vpx_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct32x32_1024_add/;
+
+  add_proto qw/void vpx_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct32x32_34_add/;
+
+  add_proto qw/void vpx_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_idct32x32_1_add/;
+
+  add_proto qw/void vpx_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_iwht4x4_1_add/;
+
+  add_proto qw/void vpx_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+  specialize qw/vpx_iwht4x4_16_add/;
+
+  add_proto qw/void vpx_highbd_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_idct4x4_1_add/;
+
+  add_proto qw/void vpx_highbd_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_idct8x8_1_add/;
+
+  add_proto qw/void vpx_highbd_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_idct16x16_1_add/;
+
+  add_proto qw/void vpx_highbd_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_idct32x32_1024_add/;
+
+  add_proto qw/void vpx_highbd_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_idct32x32_34_add/;
+
+  add_proto qw/void vpx_highbd_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_idct32x32_1_add/;
+
+  add_proto qw/void vpx_highbd_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_iwht4x4_1_add/;
+
+  add_proto qw/void vpx_highbd_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+  specialize qw/vpx_highbd_iwht4x4_16_add/;
 
   # Force C versions if CONFIG_EMULATE_HARDWARE is 1
   if (vpx_config("CONFIG_EMULATE_HARDWARE") eq "yes") {
-    add_proto qw/void vp9_highbd_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct4x4_16_add/;
-
-    add_proto qw/void vp9_highbd_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct8x8_64_add/;
-
-    add_proto qw/void vp9_highbd_idct8x8_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct8x8_10_add/;
-
-    add_proto qw/void vp9_highbd_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct16x16_256_add/;
-
-    add_proto qw/void vp9_highbd_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct16x16_10_add/;
+    add_proto qw/void vpx_highbd_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct4x4_16_add/;
+
+    add_proto qw/void vpx_highbd_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct8x8_64_add/;
+
+    add_proto qw/void vpx_highbd_idct8x8_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct8x8_10_add/;
+
+    add_proto qw/void vpx_highbd_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct16x16_256_add/;
+
+    add_proto qw/void vpx_highbd_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct16x16_10_add/;
   } else {
-    add_proto qw/void vp9_highbd_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct4x4_16_add sse2/;
-
-    add_proto qw/void vp9_highbd_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct8x8_64_add sse2/;
-
-    add_proto qw/void vp9_highbd_idct8x8_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct8x8_10_add sse2/;
-
-    add_proto qw/void vp9_highbd_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct16x16_256_add sse2/;
-
-    add_proto qw/void vp9_highbd_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
-    specialize qw/vp9_highbd_idct16x16_10_add sse2/;
+    add_proto qw/void vpx_highbd_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct4x4_16_add sse2/;
+
+    add_proto qw/void vpx_highbd_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct8x8_64_add sse2/;
+
+    add_proto qw/void vpx_highbd_idct8x8_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct8x8_10_add sse2/;
+
+    add_proto qw/void vpx_highbd_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct16x16_256_add sse2/;
+
+    add_proto qw/void vpx_highbd_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride, int bd";
+    specialize qw/vpx_highbd_idct16x16_10_add sse2/;
   }  # CONFIG_EMULATE_HARDWARE
 } else {
   # Force C versions if CONFIG_EMULATE_HARDWARE is 1
   if (vpx_config("CONFIG_EMULATE_HARDWARE") eq "yes") {
-    add_proto qw/void vp9_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct4x4_1_add/;
-
-    add_proto qw/void vp9_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct4x4_16_add/;
-
-    add_proto qw/void vp9_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct8x8_1_add/;
-
-    add_proto qw/void vp9_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct8x8_64_add/;
-
-    add_proto qw/void vp9_idct8x8_12_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct8x8_12_add/;
-
-    add_proto qw/void vp9_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct16x16_1_add/;
-
-    add_proto qw/void vp9_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct16x16_256_add/;
-
-    add_proto qw/void vp9_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct16x16_10_add/;
-
-    add_proto qw/void vp9_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct32x32_1024_add/;
-
-    add_proto qw/void vp9_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct32x32_34_add/;
-
-    add_proto qw/void vp9_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct32x32_1_add/;
-
-    add_proto qw/void vp9_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_iwht4x4_1_add/;
-
-    add_proto qw/void vp9_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_iwht4x4_16_add/;
+    add_proto qw/void vpx_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct4x4_1_add/;
+
+    add_proto qw/void vpx_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct4x4_16_add/;
+
+    add_proto qw/void vpx_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct8x8_1_add/;
+
+    add_proto qw/void vpx_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct8x8_64_add/;
+
+    add_proto qw/void vpx_idct8x8_12_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct8x8_12_add/;
+
+    add_proto qw/void vpx_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct16x16_1_add/;
+
+    add_proto qw/void vpx_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct16x16_256_add/;
+
+    add_proto qw/void vpx_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct16x16_10_add/;
+
+    add_proto qw/void vpx_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct32x32_1024_add/;
+
+    add_proto qw/void vpx_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct32x32_34_add/;
+
+    add_proto qw/void vpx_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct32x32_1_add/;
+
+    add_proto qw/void vpx_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_iwht4x4_1_add/;
+
+    add_proto qw/void vpx_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_iwht4x4_16_add/;
   } else {
-    add_proto qw/void vp9_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct4x4_1_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct4x4_16_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct8x8_1_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct8x8_64_add sse2 neon dspr2 msa/, "$ssse3_x86_64_x86inc";
-
-    add_proto qw/void vp9_idct8x8_12_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct8x8_12_add sse2 neon dspr2 msa/, "$ssse3_x86_64_x86inc";
-
-    add_proto qw/void vp9_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct16x16_1_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct16x16_256_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct16x16_10_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct32x32_1024_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct32x32_34_add sse2 neon_asm dspr2 msa/;
+    add_proto qw/void vpx_idct4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct4x4_1_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_idct4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct4x4_16_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_idct8x8_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct8x8_1_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_idct8x8_64_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct8x8_64_add sse2 neon dspr2 msa/, "$ssse3_x86_64_x86inc";
+
+    add_proto qw/void vpx_idct8x8_12_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct8x8_12_add sse2 neon dspr2 msa/, "$ssse3_x86_64_x86inc";
+
+    add_proto qw/void vpx_idct16x16_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct16x16_1_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_idct16x16_256_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct16x16_256_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_idct16x16_10_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct16x16_10_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_idct32x32_1024_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct32x32_1024_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_idct32x32_34_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct32x32_34_add sse2 neon_asm dspr2 msa/;
     # Need to add 34 eob idct32x32 neon implementation.
-    $vp9_idct32x32_34_add_neon_asm=vp9_idct32x32_1024_add_neon;
-
-    add_proto qw/void vp9_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_idct32x32_1_add sse2 neon dspr2 msa/;
-
-    add_proto qw/void vp9_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_iwht4x4_1_add msa/;
-
-    add_proto qw/void vp9_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
-    specialize qw/vp9_iwht4x4_16_add msa/, "$sse2_x86inc";
+    $vpx_idct32x32_34_add_neon_asm=vpx_idct32x32_1024_add_neon;
+
+    add_proto qw/void vpx_idct32x32_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_idct32x32_1_add sse2 neon dspr2 msa/;
+
+    add_proto qw/void vpx_iwht4x4_1_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_iwht4x4_1_add msa/;
+
+    add_proto qw/void vpx_iwht4x4_16_add/, "const tran_low_t *input, uint8_t *dest, int dest_stride";
+    specialize qw/vpx_iwht4x4_16_add msa/, "$sse2_x86inc";
   }  # CONFIG_EMULATE_HARDWARE
 }  # CONFIG_VP9_HIGHBITDEPTH
-}  # CONFIG_VP9
+}  # CONFIG_VP9 || CONFIG_VP10
 
 #
 # Quantization
 #
-if (vpx_config("CONFIG_VP9_ENCODER") eq "yes") {
+if ((vpx_config("CONFIG_VP9_ENCODER") eq "yes") || (vpx_config("CONFIG_VP10_ENCODER") eq "yes")) {
 if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
-  add_proto qw/void vp9_quantize_b/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
-  specialize qw/vp9_quantize_b/;
-
-  add_proto qw/void vp9_quantize_b_32x32/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
-  specialize qw/vp9_quantize_b_32x32/;
-
-  add_proto qw/void vp9_highbd_quantize_b/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
-  specialize qw/vp9_highbd_quantize_b sse2/;
-
-  add_proto qw/void vp9_highbd_quantize_b_32x32/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
-  specialize qw/vp9_highbd_quantize_b_32x32 sse2/;
+  add_proto qw/void vpx_quantize_b/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
+  specialize qw/vpx_quantize_b/;
+
+  add_proto qw/void vpx_quantize_b_32x32/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
+  specialize qw/vpx_quantize_b_32x32/;
+
+  add_proto qw/void vpx_highbd_quantize_b/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
+  specialize qw/vpx_highbd_quantize_b sse2/;
+
+  add_proto qw/void vpx_highbd_quantize_b_32x32/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
+  specialize qw/vpx_highbd_quantize_b_32x32 sse2/;
 } else {
-  add_proto qw/void vp9_quantize_b/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
-  specialize qw/vp9_quantize_b sse2/, "$ssse3_x86_64_x86inc";
-
-  add_proto qw/void vp9_quantize_b_32x32/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
-  specialize qw/vp9_quantize_b_32x32/, "$ssse3_x86_64_x86inc";
-}  # CONFIG_VP9_ENCODER
+  add_proto qw/void vpx_quantize_b/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
+  specialize qw/vpx_quantize_b sse2/, "$ssse3_x86_64_x86inc";
+
+  add_proto qw/void vpx_quantize_b_32x32/, "const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, const int16_t *zbin_ptr, const int16_t *round_ptr, const int16_t *quant_ptr, const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const int16_t *scan, const int16_t *iscan";
+  specialize qw/vpx_quantize_b_32x32/, "$ssse3_x86_64_x86inc";
 }  # CONFIG_VP9_HIGHBITDEPTH
+}  # CONFIG_VP9_ENCODER || CONFIG_VP10_ENCODER
 
 if (vpx_config("CONFIG_ENCODERS") eq "yes") {
 #
 # Block subtraction
 #
 add_proto qw/void vpx_subtract_block/, "int rows, int cols, int16_t *diff_ptr, ptrdiff_t diff_stride, const uint8_t *src_ptr, ptrdiff_t src_stride, const uint8_t *pred_ptr, ptrdiff_t pred_stride";
 specialize qw/vpx_subtract_block neon msa/, "$sse2_x86inc";
 
 #
 # Single block SAD
@@ skipping 167 matching lines @@
 
 add_proto qw/void vpx_sad8x4x4d/, "const uint8_t *src_ptr, int src_stride, const uint8_t * const ref_ptr[], int ref_stride, uint32_t *sad_array";
 specialize qw/vpx_sad8x4x4d msa/, "$sse2_x86inc";
 
 add_proto qw/void vpx_sad4x8x4d/, "const uint8_t *src_ptr, int src_stride, const uint8_t * const ref_ptr[], int ref_stride, uint32_t *sad_array";
 specialize qw/vpx_sad4x8x4d msa/, "$sse_x86inc";
 
 add_proto qw/void vpx_sad4x4x4d/, "const uint8_t *src_ptr, int src_stride, const uint8_t * const ref_ptr[], int ref_stride, uint32_t *sad_array";
 specialize qw/vpx_sad4x4x4d msa/, "$sse_x86inc";
 
+#
+# Structured Similarity (SSIM)
+#
+if (vpx_config("CONFIG_INTERNAL_STATS") eq "yes") {
+    add_proto qw/void vpx_ssim_parms_8x8/, "const uint8_t *s, int sp, const uint8_t *r, int rp, uint32_t *sum_s, uint32_t *sum_r, uint32_t *sum_sq_s, uint32_t *sum_sq_r, uint32_t *sum_sxr";
+    specialize qw/vpx_ssim_parms_8x8/, "$sse2_x86_64";
+
+    add_proto qw/void vpx_ssim_parms_16x16/, "const uint8_t *s, int sp, const uint8_t *r, int rp, uint32_t *sum_s, uint32_t *sum_r, uint32_t *sum_sq_s, uint32_t *sum_sq_r, uint32_t *sum_sxr";
+    specialize qw/vpx_ssim_parms_16x16/, "$sse2_x86_64";
+}
+
 if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
   #
   # Block subtraction
   #
   add_proto qw/void vpx_highbd_subtract_block/, "int rows, int cols, int16_t *diff_ptr, ptrdiff_t diff_stride, const uint8_t *src_ptr, ptrdiff_t src_stride, const uint8_t *pred_ptr, ptrdiff_t pred_stride, int bd";
   specialize qw/vpx_highbd_subtract_block/;
 
   #
   # Single block SAD
   #
@@ skipping 166 matching lines @@
 
   add_proto qw/void vpx_highbd_sad8x4x4d/, "const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, uint32_t *sad_array";
   specialize qw/vpx_highbd_sad8x4x4d/, "$sse2_x86inc";
 
   add_proto qw/void vpx_highbd_sad4x8x4d/, "const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, uint32_t *sad_array";
   specialize qw/vpx_highbd_sad4x8x4d/, "$sse2_x86inc";
 
   add_proto qw/void vpx_highbd_sad4x4x4d/, "const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, uint32_t *sad_array";
   specialize qw/vpx_highbd_sad4x4x4d/, "$sse2_x86inc";
 
+  #
+  # Structured Similarity (SSIM)
+  #
+  if (vpx_config("CONFIG_INTERNAL_STATS") eq "yes") {
+    add_proto qw/void vpx_highbd_ssim_parms_8x8/, "const uint16_t *s, int sp, const uint16_t *r, int rp, uint32_t *sum_s, uint32_t *sum_r, uint32_t *sum_sq_s, uint32_t *sum_sq_r, uint32_t *sum_sxr";
+    specialize qw/vpx_highbd_ssim_parms_8x8/;
+  }
 }  # CONFIG_VP9_HIGHBITDEPTH
 }  # CONFIG_ENCODERS
 
 if (vpx_config("CONFIG_ENCODERS") eq "yes" || vpx_config("CONFIG_POSTPROC") eq "yes" || vpx_config("CONFIG_VP9_POSTPROC") eq "yes") {
 
 #
 # Variance
 #
 add_proto qw/unsigned int vpx_variance64x64/, "const uint8_t *src_ptr, int source_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse";
   specialize qw/vpx_variance64x64 sse2 avx2 neon msa/;
@@ skipping 510 matching lines @@
   add_proto qw/uint32_t vpx_highbd_8_sub_pixel_avg_variance8x4/, "const uint8_t *src_ptr, int source_stride, int xoffset, int  yoffset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred";
   specialize qw/vpx_highbd_8_sub_pixel_avg_variance8x4/, "$sse2_x86inc";
 
   add_proto qw/uint32_t vpx_highbd_8_sub_pixel_avg_variance4x8/, "const uint8_t *src_ptr, int source_stride, int xoffset, int  yoffset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred";
   add_proto qw/uint32_t vpx_highbd_8_sub_pixel_avg_variance4x4/, "const uint8_t *src_ptr, int source_stride, int xoffset, int  yoffset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred";
 
 }  # CONFIG_VP9_HIGHBITDEPTH
 }  # CONFIG_ENCODERS || CONFIG_POSTPROC || CONFIG_VP9_POSTPROC
 
 1;