Refactor and clean up on blend_mask6

Change-Id: Ie9188471e7dc07ab9c95b22f258b1662e895c533

diff --git a/test/blend_mask6_test.cc b/test/blend_mask6b_test.cc
similarity index 97%
rename from test/blend_mask6_test.cc
rename to test/blend_mask6b_test.cc
index 6afaad7344f41c8b67c58236b060014da0157b71..5cd7bf7acb8d1ee2761fff99772674dec742ebf6 100644 (file)
--- a/test/blend_mask6_test.cc
+++ b/test/blend_mask6b_test.cc
@@ -189,7 +189,7 @@ TEST_P(BlendMask6Test8B, ExtremeValues) {
 #if HAVE_SSE4_1
 INSTANTIATE_TEST_CASE_P(
   SSE4_1_C_COMPARE, BlendMask6Test8B,
-  ::testing::Values(make_tuple(&vpx_blend_mask6_c, &vpx_blend_mask6_sse4_1)));
+  ::testing::Values(make_tuple(&vpx_blend_mask6b_c, &vpx_blend_mask6b_sse4_1)));
 #endif  // HAVE_SSE4_1
 
 #if CONFIG_VP9_HIGHBITDEPTH
@@ -287,8 +287,8 @@ TEST_P(BlendMask6TestHBD, ExtremeValues) {
 #if HAVE_SSE4_1
 INSTANTIATE_TEST_CASE_P(
   SSE4_1_C_COMPARE, BlendMask6TestHBD,
-  ::testing::Values(make_tuple(&vpx_highbd_blend_mask6_c,
-                               &vpx_highbd_blend_mask6_sse4_1)));
+  ::testing::Values(make_tuple(&vpx_highbd_blend_mask6b_c,
+                               &vpx_highbd_blend_mask6b_sse4_1)));
 #endif  // HAVE_SSE4_1
 #endif  // CONFIG_VP9_HIGHBITDEPTH
 }  // namespace

diff --git a/test/test.mk b/test/test.mk
index fcd565c6b5d3cd8aa72d81b5862a95c49a017743..dd6abe123ce86d58760850c64a9ddb4778743670 100644 (file)
--- a/test/test.mk
+++ b/test/test.mk
@@ -182,7 +182,7 @@ LIBVPX_TEST_SRCS-$(CONFIG_VP10_ENCODER) += subtract_test.cc
 ifeq ($(CONFIG_EXT_INTER),yes)
 LIBVPX_TEST_SRCS-$(HAVE_SSSE3) += masked_variance_test.cc
 LIBVPX_TEST_SRCS-$(HAVE_SSSE3) += masked_sad_test.cc
-LIBVPX_TEST_SRCS-$(CONFIG_VP10_ENCODER) += blend_mask6_test.cc
+LIBVPX_TEST_SRCS-$(CONFIG_VP10_ENCODER) += blend_mask6b_test.cc
 LIBVPX_TEST_SRCS-$(CONFIG_VP10_ENCODER) += vp10_wedge_utils_test.cc
 endif
 

diff --git a/test/vp10_wedge_utils_test.cc b/test/vp10_wedge_utils_test.cc
index 930a5981d4b660c496184283bccf37816e71135d..4659c9aa81cfe785d841cc9f75101a0d47ab8017 100644 (file)
--- a/test/vp10_wedge_utils_test.cc
+++ b/test/vp10_wedge_utils_test.cc
@@ -104,7 +104,7 @@ TEST_F(WedgeUtilsSSEFuncTest, ResidualBlendingEquiv) {
       p1[j] = clamp(s[j] + rng_(33) - 16, 0, UINT8_MAX);
     }
 
-    vpx_blend_mask6(p, w, p0, w, p1, w, m, w, h, w, 0, 0);
+    vpx_blend_mask6b(p, w, p0, w, p1, w, m, w, h, w, 0, 0);
 
     vpx_subtract_block(h, w, r0, w, s, w, p0, w);
     vpx_subtract_block(h, w, r1, w, s, w, p1, w);

diff --git a/vp10/common/reconinter.c b/vp10/common/reconinter.c
index 70cf5e72cd7ccec8acbdabd149b115f3deacfa12..7d0065319d89c10eaa21bb1a4ed2b15b8304af4c 100644 (file)
--- a/vp10/common/reconinter.c
+++ b/vp10/common/reconinter.c
@@ -459,11 +459,11 @@ static void build_masked_compound_wedge_extend(
   const int subw = (2 << b_width_log2_lookup[sb_type]) == w;
   const uint8_t *mask = vp10_get_soft_mask(
      wedge_index, wedge_sign, sb_type, wedge_offset_x, wedge_offset_y);
-  vpx_blend_mask6(dst, dst_stride,
-                  src0, src0_stride,
-                  src1, src1_stride,
-                  mask, MASK_MASTER_STRIDE,
-                  h, w, subh, subw);
+  vpx_blend_mask6b(dst, dst_stride,
+                   src0, src0_stride,
+                   src1, src1_stride,
+                   mask, MASK_MASTER_STRIDE,
+                   h, w, subh, subw);
 }
 
 #if CONFIG_VP9_HIGHBITDEPTH
@@ -479,11 +479,11 @@ static void build_masked_compound_wedge_extend_highbd(
   const int subw = (2 << b_width_log2_lookup[sb_type]) == w;
   const uint8_t *mask = vp10_get_soft_mask(
       wedge_index, wedge_sign, sb_type, wedge_offset_x, wedge_offset_y);
-  vpx_highbd_blend_mask6(dst_8, dst_stride,
-                         src0_8, src0_stride,
-                         src1_8, src1_stride,
-                         mask, MASK_MASTER_STRIDE,
-                         h, w, subh, subw, bd);
+  vpx_highbd_blend_mask6b(dst_8, dst_stride,
+                          src0_8, src0_stride,
+                          src1_8, src1_stride,
+                          mask, MASK_MASTER_STRIDE,
+                          h, w, subh, subw, bd);
 }
 #endif  // CONFIG_VP9_HIGHBITDEPTH
 #endif  // CONFIG_SUPERTX
@@ -500,11 +500,11 @@ static void build_masked_compound_wedge(uint8_t *dst, int dst_stride,
   const int subw = (2 << b_width_log2_lookup[sb_type]) == w;
   const uint8_t *mask = vp10_get_contiguous_soft_mask(wedge_index, wedge_sign,
                                                       sb_type);
-  vpx_blend_mask6(dst, dst_stride,
-                  src0, src0_stride,
-                  src1, src1_stride,
-                  mask, 4 * num_4x4_blocks_wide_lookup[sb_type],
-                  h, w, subh, subw);
+  vpx_blend_mask6b(dst, dst_stride,
+                   src0, src0_stride,
+                   src1, src1_stride,
+                   mask, 4 * num_4x4_blocks_wide_lookup[sb_type],
+                   h, w, subh, subw);
 }
 
 #if CONFIG_VP9_HIGHBITDEPTH
@@ -520,11 +520,11 @@ static void build_masked_compound_wedge_highbd(uint8_t *dst_8, int dst_stride,
   const int subw = (2 << b_width_log2_lookup[sb_type]) == w;
   const uint8_t *mask = vp10_get_contiguous_soft_mask(wedge_index, wedge_sign,
                                                       sb_type);
-  vpx_highbd_blend_mask6(dst_8, dst_stride,
-                         src0_8, src0_stride,
-                         src1_8, src1_stride,
-                         mask, 4 * num_4x4_blocks_wide_lookup[sb_type],
-                         h, w, subh, subw, bd);
+  vpx_highbd_blend_mask6b(dst_8, dst_stride,
+                          src0_8, src0_stride,
+                          src1_8, src1_stride,
+                          mask, 4 * num_4x4_blocks_wide_lookup[sb_type],
+                          h, w, subh, subw, bd);
 }
 #endif  // CONFIG_VP9_HIGHBITDEPTH
 
@@ -1896,11 +1896,11 @@ static void combine_interintra(INTERINTRA_MODE mode,
                                                           bsize);
       const int subw = 2 * num_4x4_blocks_wide_lookup[bsize] == bw;
       const int subh = 2 * num_4x4_blocks_high_lookup[bsize] == bh;
-      vpx_blend_mask6(comppred, compstride,
-                      intrapred, intrastride,
-                      interpred, interstride,
-                      mask, 4 * num_4x4_blocks_wide_lookup[bsize],
-                      bh, bw, subh, subw);
+      vpx_blend_mask6b(comppred, compstride,
+                       intrapred, intrastride,
+                       interpred, interstride,
+                       mask, 4 * num_4x4_blocks_wide_lookup[bsize],
+                       bh, bw, subh, subw);
     }
     return;
   }
@@ -2036,11 +2036,11 @@ static void combine_interintra_highbd(INTERINTRA_MODE mode,
                                                           bsize);
       const int subh = 2 * num_4x4_blocks_high_lookup[bsize] == bh;
       const int subw = 2 * num_4x4_blocks_wide_lookup[bsize] == bw;
-      vpx_highbd_blend_mask6(comppred8, compstride,
-                             intrapred8, intrastride,
-                             interpred8, interstride,
-                             mask, bw,
-                             bh, bw, subh, subw, bd);
+      vpx_highbd_blend_mask6b(comppred8, compstride,
+                              intrapred8, intrastride,
+                              interpred8, interstride,
+                              mask, bw,
+                              bh, bw, subh, subw, bd);
     }
     return;
   }

diff --git a/vpx_dsp/blend_mask.h b/vpx_dsp/blend_mask.h
new file mode 100644 (file)
index 0000000..91c5f4d
--- /dev/null
+++ b/vpx_dsp/blend_mask.h
@@ -0,0 +1,17 @@
+/*
+*  Copyright (c) 2016 The WebM project authors. All Rights Reserved.
+*
+*  Use of this source code is governed by a BSD-style license
+*  that can be found in the LICENSE file in the root of the source
+*  tree. An additional intellectual property rights grant can be found
+*  in the file PATENTS.  All contributing project authors may
+*  be found in the AUTHORS file in the root of the source tree.
+*/
+
+#ifndef VPX_DSP_BLEND_MASK_H_
+#define VPX_DSP_BLEND_MASK_H_
+
+// Use blend_mask6b() for 6 bit masks
+#define MASK_BITS6   6
+
+#endif  // VPX_DSP_BLEND_MASK_H_

diff --git a/vpx_dsp/blend_mask6.c b/vpx_dsp/blend_mask6b.c
similarity index 80%
rename from vpx_dsp/blend_mask6.c
rename to vpx_dsp/blend_mask6b.c
index 584ee6a78c3ac44274e1af129942816c4f4333c8..abdefa300796422ee129902cfc68a07798522d8a 100644 (file)
--- a/vpx_dsp/blend_mask6.c
+++ b/vpx_dsp/blend_mask6b.c
@@ -12,17 +12,16 @@
 
 #include "vpx/vpx_integer.h"
 #include "vpx_ports/mem.h"
+#include "vpx_dsp/blend_mask.h"
 #include "vpx_dsp/vpx_dsp_common.h"
 
 #include "./vpx_dsp_rtcd.h"
 
-#define MASK_BITS 6
-
-void vpx_blend_mask6_c(uint8_t *dst, uint32_t dst_stride,
-                       uint8_t *src0, uint32_t src0_stride,
-                       uint8_t *src1, uint32_t src1_stride,
-                       const uint8_t *mask, uint32_t mask_stride,
-                       int h, int w, int subh, int subw) {
+void vpx_blend_mask6b_c(uint8_t *dst, uint32_t dst_stride,
+                        uint8_t *src0, uint32_t src0_stride,
+                        uint8_t *src1, uint32_t src1_stride,
+                        const uint8_t *mask, uint32_t mask_stride,
+                        int h, int w, int subh, int subw) {
   int i, j;
 
   assert(IMPLIES(src0 == dst, src0_stride == dst_stride));
@@ -37,10 +36,10 @@ void vpx_blend_mask6_c(uint8_t *dst, uint32_t dst_stride,
     for (i = 0; i < h; ++i)
       for (j = 0; j < w; ++j) {
         const int m0 = mask[i * mask_stride + j];
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   } else if (subw == 1 && subh == 1) {
     for (i = 0; i < h; ++i)
@@ -51,10 +50,10 @@ void vpx_blend_mask6_c(uint8_t *dst, uint32_t dst_stride,
                                mask[(2 * i) * mask_stride + (2 * j + 1)] +
                                mask[(2 * i + 1) * mask_stride + (2 * j + 1)],
                                2);
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   } else if (subw == 1 && subh == 0) {
     for (i = 0; i < h; ++i)
@@ -62,10 +61,10 @@ void vpx_blend_mask6_c(uint8_t *dst, uint32_t dst_stride,
         const int m0 =
             ROUND_POWER_OF_TWO(mask[i * mask_stride + (2 * j)] +
                                mask[i * mask_stride + (2 * j + 1)], 1);
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   } else {
     for (i = 0; i < h; ++i)
@@ -73,20 +72,20 @@ void vpx_blend_mask6_c(uint8_t *dst, uint32_t dst_stride,
         const int m0 =
             ROUND_POWER_OF_TWO(mask[(2 * i) * mask_stride + j] +
                                mask[(2 * i + 1) * mask_stride + j], 1);
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   }
 }
 
 #if CONFIG_VP9_HIGHBITDEPTH
-void vpx_highbd_blend_mask6_c(uint8_t *dst_8, uint32_t dst_stride,
-                              uint8_t *src0_8, uint32_t src0_stride,
-                              uint8_t *src1_8, uint32_t src1_stride,
-                              const uint8_t *mask, uint32_t mask_stride,
-                              int h, int w, int subh, int subw, int bd) {
+void vpx_highbd_blend_mask6b_c(uint8_t *dst_8, uint32_t dst_stride,
+                               uint8_t *src0_8, uint32_t src0_stride,
+                               uint8_t *src1_8, uint32_t src1_stride,
+                               const uint8_t *mask, uint32_t mask_stride,
+                               int h, int w, int subh, int subw, int bd) {
   int i, j;
   uint16_t *dst = CONVERT_TO_SHORTPTR(dst_8);
   uint16_t *src0 = CONVERT_TO_SHORTPTR(src0_8);
@@ -106,10 +105,10 @@ void vpx_highbd_blend_mask6_c(uint8_t *dst_8, uint32_t dst_stride,
     for (i = 0; i < h; ++i)
       for (j = 0; j < w; ++j) {
         const int m0 = mask[i * mask_stride + j];
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   } else if (subw == 1 && subh == 1) {
     for (i = 0; i < h; ++i)
@@ -120,10 +119,10 @@ void vpx_highbd_blend_mask6_c(uint8_t *dst_8, uint32_t dst_stride,
                                mask[(2 * i) * mask_stride + (2 * j + 1)] +
                                mask[(2 * i + 1) * mask_stride + (2 * j + 1)],
                                2);
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   } else if (subw == 1 && subh == 0) {
     for (i = 0; i < h; ++i)
@@ -131,10 +130,10 @@ void vpx_highbd_blend_mask6_c(uint8_t *dst_8, uint32_t dst_stride,
         const int m0 =
             ROUND_POWER_OF_TWO(mask[i * mask_stride + (2 * j)] +
                                mask[i * mask_stride + (2 * j + 1)], 1);
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   } else {
     for (i = 0; i < h; ++i)
@@ -142,10 +141,10 @@ void vpx_highbd_blend_mask6_c(uint8_t *dst_8, uint32_t dst_stride,
         const int m0 =
             ROUND_POWER_OF_TWO(mask[(2 * i) * mask_stride + j] +
                                mask[(2 * i + 1) * mask_stride + j], 1);
-        const int m1 = ((1 << MASK_BITS) - m0);
+        const int m1 = ((1 << MASK_BITS6) - m0);
         dst[i * dst_stride + j] =
             ROUND_POWER_OF_TWO(src0[i * src0_stride + j] * m0 +
-                               src1[i * src1_stride + j] * m1, MASK_BITS);
+                               src1[i * src1_stride + j] * m1, MASK_BITS6);
       }
   }
 }

diff --git a/vpx_dsp/vpx_dsp.mk b/vpx_dsp/vpx_dsp.mk
index 3c519b68b636787c59900fffb7ca13e22c728cf2..f48dd060a54b0faf3df33a5194df9ae5508bc2e8 100644 (file)
--- a/vpx_dsp/vpx_dsp.mk
+++ b/vpx_dsp/vpx_dsp.mk
@@ -71,8 +71,9 @@ DSP_SRCS-$(HAVE_DSPR2)  += mips/common_dspr2.c
 
 ifeq ($(CONFIG_VP10),yes)
 ifeq ($(CONFIG_EXT_INTER),yes)
-DSP_SRCS-yes            += blend_mask6.c
-DSP_SRCS-$(HAVE_SSE4_1) += x86/blend_mask6_sse4.c
+DSP_SRCS-yes            += blend_mask6b.c
+DSP_SRCS-yes            += blend_mask.h
+DSP_SRCS-$(HAVE_SSE4_1) += x86/blend_mask6b_sse4.c
 endif  #CONFIG_EXT_INTER
 endif  #CONFIG_VP10
 

diff --git a/vpx_dsp/vpx_dsp_rtcd_defs.pl b/vpx_dsp/vpx_dsp_rtcd_defs.pl
index e6309947f9beea95d1a065b55c905ff1dfedfe88..59f137e25e5a3b4b1f2574ee01641f3e26c6ad28 100644 (file)
--- a/vpx_dsp/vpx_dsp_rtcd_defs.pl
+++ b/vpx_dsp/vpx_dsp_rtcd_defs.pl
@@ -1385,12 +1385,12 @@ if (vpx_config("CONFIG_EXT_INTER") eq "yes") {
     }
   }
 
-  add_proto qw/void vpx_blend_mask6/, "uint8_t *dst, uint32_t dst_stride,  uint8_t *src0, uint32_t src0_stride, uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int h, int w, int suby, int subx";
-  specialize "vpx_blend_mask6", qw/sse4_1/;
+  add_proto qw/void vpx_blend_mask6b/, "uint8_t *dst, uint32_t dst_stride,  uint8_t *src0, uint32_t src0_stride, uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int h, int w, int suby, int subx";
+  specialize "vpx_blend_mask6b", qw/sse4_1/;
 
   if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
-    add_proto qw/void vpx_highbd_blend_mask6/, "uint8_t *dst, uint32_t dst_stride,  uint8_t *src0, uint32_t src0_stride, uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int h, int w, int suby, int subx, int bd";
-    specialize "vpx_highbd_blend_mask6", qw/sse4_1/;
+    add_proto qw/void vpx_highbd_blend_mask6b/, "uint8_t *dst, uint32_t dst_stride,  uint8_t *src0, uint32_t src0_stride, uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int h, int w, int suby, int subx, int bd";
+    specialize "vpx_highbd_blend_mask6b", qw/sse4_1/;
   }
 }
 

diff --git a/vpx_dsp/x86/blend_mask6_sse4.c b/vpx_dsp/x86/blend_mask6b_sse4.c
similarity index 81%
rename from vpx_dsp/x86/blend_mask6_sse4.c
rename to vpx_dsp/x86/blend_mask6b_sse4.c
index 28693a4a3d7e1febdf8e6535adc1e9df1eb8addf..0b128563275088091f18d630782e25b3310bdff0 100644 (file)
--- a/vpx_dsp/x86/blend_mask6_sse4.c
+++ b/vpx_dsp/x86/blend_mask6b_sse4.c
@@ -15,13 +15,12 @@
 #include "vpx/vpx_integer.h"
 #include "vpx_ports/mem.h"
 #include "vpx_dsp/vpx_dsp_common.h"
+#include "vpx_dsp/blend_mask.h"
 
 #include "vpx_dsp/x86/synonyms.h"
 
 #include "./vpx_dsp_rtcd.h"
 
-#define MASK_BITS 6
-
 //////////////////////////////////////////////////////////////////////////////
 // Common kernels
 //////////////////////////////////////////////////////////////////////////////
@@ -38,7 +37,7 @@ static INLINE __m128i blend_4(uint8_t*src0, uint8_t *src1,
 
   const __m128i v_sum_w = _mm_add_epi16(v_p0_w, v_p1_w);
 
-  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS);
+  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS6);
 
   return v_res_w;
 }
@@ -55,7 +54,7 @@ static INLINE __m128i blend_8(uint8_t*src0, uint8_t *src1,
 
   const __m128i v_sum_w = _mm_add_epi16(v_p0_w, v_p1_w);
 
-  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS);
+  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS6);
 
   return v_res_w;
 }
@@ -64,13 +63,13 @@ static INLINE __m128i blend_8(uint8_t*src0, uint8_t *src1,
 // No sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static void blend_mask6_w4_sse4_1(
+static void blend_mask6b_w4_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -92,13 +91,13 @@ static void blend_mask6_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_w8_sse4_1(
+static void blend_mask6b_w8_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -120,13 +119,13 @@ static void blend_mask6_w8_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_w16n_sse4_1(
+static void blend_mask6b_w16n_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -158,7 +157,7 @@ static void blend_mask6_w16n_sse4_1(
 // Horizontal sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static void blend_mask6_sx_w4_sse4_1(
+static void blend_mask6b_sx_w4_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
@@ -166,7 +165,7 @@ static void blend_mask6_sx_w4_sse4_1(
     int h, int w) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -190,7 +189,7 @@ static void blend_mask6_sx_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_sx_w8_sse4_1(
+static void blend_mask6b_sx_w8_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
@@ -198,7 +197,7 @@ static void blend_mask6_sx_w8_sse4_1(
     int h, int w) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -222,7 +221,7 @@ static void blend_mask6_sx_w8_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_sx_w16n_sse4_1(
+static void blend_mask6b_sx_w16n_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
@@ -230,7 +229,7 @@ static void blend_mask6_sx_w16n_sse4_1(
     int h, int w) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -265,13 +264,13 @@ static void blend_mask6_sx_w16n_sse4_1(
 // Vertical sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static void blend_mask6_sy_w4_sse4_1(
+static void blend_mask6b_sy_w4_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -296,13 +295,13 @@ static void blend_mask6_sy_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_sy_w8_sse4_1(
+static void blend_mask6b_sy_w8_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -327,14 +326,14 @@ static void blend_mask6_sy_w8_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_sy_w16n_sse4_1(
+static void blend_mask6b_sy_w16n_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   const __m128i v_zero = _mm_setzero_si128();
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -368,7 +367,7 @@ static void blend_mask6_sy_w16n_sse4_1(
 // Horizontal and Vertical sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static void blend_mask6_sx_sy_w4_sse4_1(
+static void blend_mask6b_sx_sy_w4_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
@@ -376,7 +375,7 @@ static void blend_mask6_sx_sy_w4_sse4_1(
     int h, int w) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -405,7 +404,7 @@ static void blend_mask6_sx_sy_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_sx_sy_w8_sse4_1(
+static void blend_mask6b_sx_sy_w8_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
@@ -413,7 +412,7 @@ static void blend_mask6_sx_sy_w8_sse4_1(
     int h, int w) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   (void)w;
 
@@ -442,7 +441,7 @@ static void blend_mask6_sx_sy_w8_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_sx_sy_w16n_sse4_1(
+static void blend_mask6b_sx_sy_w16n_sse4_1(
     uint8_t *dst, uint32_t dst_stride,
     uint8_t *src0, uint32_t src0_stride,
     uint8_t *src1, uint32_t src1_stride,
@@ -450,7 +449,7 @@ static void blend_mask6_sx_sy_w16n_sse4_1(
     int h, int w) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -495,7 +494,7 @@ static void blend_mask6_sx_sy_w16n_sse4_1(
 // Dispatch
 //////////////////////////////////////////////////////////////////////////////
 
-void vpx_blend_mask6_sse4_1(uint8_t *dst, uint32_t dst_stride,
+void vpx_blend_mask6b_sse4_1(uint8_t *dst, uint32_t dst_stride,
                             uint8_t *src0, uint32_t src0_stride,
                             uint8_t *src1, uint32_t src1_stride,
                             const uint8_t *mask, uint32_t mask_stride,
@@ -508,14 +507,14 @@ void vpx_blend_mask6_sse4_1(uint8_t *dst, uint32_t dst_stride,
 
   static blend_fn blend[3][2][2] = {  // width_index X subx X suby
     {     // w % 16 == 0
-      {blend_mask6_w16n_sse4_1, blend_mask6_sy_w16n_sse4_1},
-      {blend_mask6_sx_w16n_sse4_1, blend_mask6_sx_sy_w16n_sse4_1}
+      {blend_mask6b_w16n_sse4_1, blend_mask6b_sy_w16n_sse4_1},
+      {blend_mask6b_sx_w16n_sse4_1, blend_mask6b_sx_sy_w16n_sse4_1}
     }, {  // w == 4
-      {blend_mask6_w4_sse4_1, blend_mask6_sy_w4_sse4_1},
-      {blend_mask6_sx_w4_sse4_1, blend_mask6_sx_sy_w4_sse4_1}
+      {blend_mask6b_w4_sse4_1, blend_mask6b_sy_w4_sse4_1},
+      {blend_mask6b_sx_w4_sse4_1, blend_mask6b_sx_sy_w4_sse4_1}
     }, {  // w == 8
-      {blend_mask6_w8_sse4_1, blend_mask6_sy_w8_sse4_1},
-      {blend_mask6_sx_w8_sse4_1, blend_mask6_sx_sy_w8_sse4_1}
+      {blend_mask6b_w8_sse4_1, blend_mask6b_sy_w8_sse4_1},
+      {blend_mask6b_sx_w8_sse4_1, blend_mask6b_sx_sy_w8_sse4_1}
     }
   };
 
@@ -552,7 +551,7 @@ static INLINE __m128i blend_4_b10(uint16_t*src0, uint16_t *src1,
 
   const __m128i v_sum_w = _mm_add_epi16(v_p0_w, v_p1_w);
 
-  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS);
+  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS6);
 
   return v_res_w;
 }
@@ -567,7 +566,7 @@ static INLINE __m128i blend_8_b10(uint16_t*src0, uint16_t *src1,
 
   const __m128i v_sum_w = _mm_add_epi16(v_p0_w, v_p1_w);
 
-  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS);
+  const __m128i v_res_w = xx_roundn_epu16(v_sum_w, MASK_BITS6);
 
   return v_res_w;
 }
@@ -585,7 +584,7 @@ static INLINE __m128i blend_4_b12(uint16_t*src0, uint16_t *src1,
   const __m128i v_sum_d = _mm_madd_epi16(v_s01_w, v_m01_w);
 
   // Scale
-  const __m128i v_ssum_d = _mm_srli_epi32(v_sum_d, MASK_BITS - 1);
+  const __m128i v_ssum_d = _mm_srli_epi32(v_sum_d, MASK_BITS6 - 1);
 
   // Pack
   const __m128i v_pssum_d = _mm_packs_epi32(v_ssum_d, v_ssum_d);
@@ -612,8 +611,8 @@ static INLINE __m128i blend_8_b12(uint16_t*src0, uint16_t *src1,
   const __m128i v_sumh_d = _mm_madd_epi16(v_s01h_w, v_m01h_w);
 
   // Scale
-  const __m128i v_ssuml_d = _mm_srli_epi32(v_suml_d, MASK_BITS - 1);
-  const __m128i v_ssumh_d = _mm_srli_epi32(v_sumh_d, MASK_BITS - 1);
+  const __m128i v_ssuml_d = _mm_srli_epi32(v_suml_d, MASK_BITS6 - 1);
+  const __m128i v_ssumh_d = _mm_srli_epi32(v_sumh_d, MASK_BITS6 - 1);
 
   // Pack
   const __m128i v_pssum_d = _mm_packs_epi32(v_ssuml_d, v_ssumh_d);
@@ -628,13 +627,13 @@ static INLINE __m128i blend_8_b12(uint16_t*src0, uint16_t *src1,
 // No sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static INLINE void blend_mask6_bn_w4_sse4_1(
+static INLINE void blend_mask6b_bn_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, blend_unit_fn blend) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     const __m128i v_m0_b = xx_loadl_32(mask);
@@ -652,37 +651,37 @@ static INLINE void blend_mask6_bn_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_w4_sse4_1(
+static void blend_mask6b_b10_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                           src1_stride, mask, mask_stride, h,
-                           blend_4_b10);
+  blend_mask6b_bn_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                            src1_stride, mask, mask_stride, h,
+                            blend_4_b10);
 }
 
-static void blend_mask6_b12_w4_sse4_1(
+static void blend_mask6b_b12_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                           src1_stride, mask, mask_stride, h,
-                           blend_4_b12);
+  blend_mask6b_bn_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                            src1_stride, mask, mask_stride, h,
+                            blend_4_b12);
 }
 
-static inline void blend_mask6_bn_w8n_sse4_1(
+static inline void blend_mask6b_bn_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w, blend_unit_fn blend) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -702,33 +701,33 @@ static inline void blend_mask6_bn_w8n_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_w8n_sse4_1(
+static void blend_mask6b_b10_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                            src1_stride, mask, mask_stride, h, w,
-                            blend_8_b10);
+  blend_mask6b_bn_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                             src1_stride, mask, mask_stride, h, w,
+                             blend_8_b10);
 }
 
-static void blend_mask6_b12_w8n_sse4_1(
+static void blend_mask6b_b12_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                            src1_stride, mask, mask_stride, h, w,
-                            blend_8_b12);
+  blend_mask6b_bn_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                             src1_stride, mask, mask_stride, h, w,
+                             blend_8_b12);
 }
 
 //////////////////////////////////////////////////////////////////////////////
 // Horizontal sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static INLINE void blend_mask6_bn_sx_w4_sse4_1(
+static INLINE void blend_mask6b_bn_sx_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
@@ -736,7 +735,7 @@ static INLINE void blend_mask6_bn_sx_w4_sse4_1(
     int h, blend_unit_fn blend) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     const __m128i v_r_b = xx_loadl_64(mask);
@@ -756,31 +755,31 @@ static INLINE void blend_mask6_bn_sx_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_sx_w4_sse4_1(
+static void blend_mask6b_b10_sx_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_sx_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                              src1_stride, mask, mask_stride, h,
-                              blend_4_b10);
+  blend_mask6b_bn_sx_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                               src1_stride, mask, mask_stride, h,
+                               blend_4_b10);
 }
 
-static void blend_mask6_b12_sx_w4_sse4_1(
+static void blend_mask6b_b12_sx_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_sx_w4_sse4_1(dst, dst_stride, src0,  src0_stride, src1,
-                              src1_stride, mask, mask_stride, h,
-                              blend_4_b12);
+  blend_mask6b_bn_sx_w4_sse4_1(dst, dst_stride, src0,  src0_stride, src1,
+                               src1_stride, mask, mask_stride, h,
+                               blend_4_b12);
 }
 
-static INLINE void blend_mask6_bn_sx_w8n_sse4_1(
+static INLINE void blend_mask6b_bn_sx_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
@@ -788,7 +787,7 @@ static INLINE void blend_mask6_bn_sx_w8n_sse4_1(
     int h, int w, blend_unit_fn blend) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -810,39 +809,39 @@ static INLINE void blend_mask6_bn_sx_w8n_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_sx_w8n_sse4_1(
+static void blend_mask6b_b10_sx_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_sx_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                               src1_stride, mask, mask_stride, h, w,
-                               blend_8_b10);
+  blend_mask6b_bn_sx_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                src1_stride, mask, mask_stride, h, w,
+                                blend_8_b10);
 }
 
-static void blend_mask6_b12_sx_w8n_sse4_1(
+static void blend_mask6b_b12_sx_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_sx_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                               src1_stride, mask, mask_stride, h, w,
-                               blend_8_b12);
+  blend_mask6b_bn_sx_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                src1_stride, mask, mask_stride, h, w,
+                                blend_8_b12);
 }
 
 //////////////////////////////////////////////////////////////////////////////
 // Vertical sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static INLINE void blend_mask6_bn_sy_w4_sse4_1(
+static INLINE void blend_mask6b_bn_sy_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, blend_unit_fn blend) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     const __m128i v_ra_b = xx_loadl_32(mask);
@@ -863,37 +862,37 @@ static INLINE void blend_mask6_bn_sy_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_sy_w4_sse4_1(
+static void blend_mask6b_b10_sy_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                              src1_stride, mask, mask_stride, h,
-                              blend_4_b10);
+  blend_mask6b_bn_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                               src1_stride, mask, mask_stride, h,
+                               blend_4_b10);
 }
 
-static void blend_mask6_b12_sy_w4_sse4_1(
+static void blend_mask6b_b12_sy_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                              src1_stride, mask, mask_stride, h,
-                              blend_4_b12);
+  blend_mask6b_bn_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                               src1_stride, mask, mask_stride, h,
+                               blend_4_b12);
 }
 
-static INLINE void blend_mask6_bn_sy_w8n_sse4_1(
+static INLINE void blend_mask6b_bn_sy_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w, blend_unit_fn blend) {
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -916,33 +915,33 @@ static INLINE void blend_mask6_bn_sy_w8n_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_sy_w8n_sse4_1(
+static void blend_mask6b_b10_sy_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                               src1_stride, mask, mask_stride, h, w,
-                               blend_8_b10);
+  blend_mask6b_bn_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                src1_stride, mask, mask_stride, h, w,
+                                blend_8_b10);
 }
 
-static void blend_mask6_b12_sy_w8n_sse4_1(
+static void blend_mask6b_b12_sy_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                               src1_stride, mask, mask_stride, h, w,
-                               blend_8_b12);
+  blend_mask6b_bn_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                src1_stride, mask, mask_stride, h, w,
+                                blend_8_b12);
 }
 
 //////////////////////////////////////////////////////////////////////////////
 // Horizontal and Vertical sub-sampling
 //////////////////////////////////////////////////////////////////////////////
 
-static INLINE void blend_mask6_bn_sx_sy_w4_sse4_1(
+static INLINE void blend_mask6b_bn_sx_sy_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
@@ -950,7 +949,7 @@ static INLINE void blend_mask6_bn_sx_sy_w4_sse4_1(
     int h, blend_unit_fn blend) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     const __m128i v_ra_b = xx_loadl_64(mask);
@@ -975,31 +974,31 @@ static INLINE void blend_mask6_bn_sx_sy_w4_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_sx_sy_w4_sse4_1(
+static void blend_mask6b_b10_sx_sy_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_sx_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                                 src1_stride, mask, mask_stride, h,
-                                 blend_4_b10);
+  blend_mask6b_bn_sx_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                  src1_stride, mask, mask_stride, h,
+                                  blend_4_b10);
 }
 
-static void blend_mask6_b12_sx_sy_w4_sse4_1(
+static void blend_mask6b_b12_sx_sy_w4_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
   (void)w;
-  blend_mask6_bn_sx_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                                 src1_stride, mask, mask_stride, h,
-                                 blend_4_b12);
+  blend_mask6b_bn_sx_sy_w4_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                  src1_stride, mask, mask_stride, h,
+                                  blend_4_b12);
 }
 
-static INLINE void blend_mask6_bn_sx_sy_w8n_sse4_1(
+static INLINE void blend_mask6b_bn_sx_sy_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
@@ -1007,7 +1006,7 @@ static INLINE void blend_mask6_bn_sx_sy_w8n_sse4_1(
     int h, int w, blend_unit_fn blend) {
   const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff,
                                          0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff);
-  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS);
+  const __m128i v_maxval_w = _mm_set1_epi16(1 << MASK_BITS6);
 
   do {
     int c;
@@ -1034,37 +1033,37 @@ static INLINE void blend_mask6_bn_sx_sy_w8n_sse4_1(
   } while (--h);
 }
 
-static void blend_mask6_b10_sx_sy_w8n_sse4_1(
+static void blend_mask6b_b10_sx_sy_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_sx_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                                  src1_stride, mask, mask_stride, h, w,
-                                  blend_8_b10);
+  blend_mask6b_bn_sx_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                   src1_stride, mask, mask_stride, h, w,
+                                   blend_8_b10);
 }
 
-static void blend_mask6_b12_sx_sy_w8n_sse4_1(
+static void blend_mask6b_b12_sx_sy_w8n_sse4_1(
     uint16_t *dst, uint32_t dst_stride,
     uint16_t *src0, uint32_t src0_stride,
     uint16_t *src1, uint32_t src1_stride,
     const uint8_t *mask, uint32_t mask_stride,
     int h, int w) {
-  blend_mask6_bn_sx_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
-                                  src1_stride, mask, mask_stride, h, w,
-                                  blend_8_b12);
+  blend_mask6b_bn_sx_sy_w8n_sse4_1(dst, dst_stride, src0, src0_stride, src1,
+                                   src1_stride, mask, mask_stride, h, w,
+                                   blend_8_b12);
 }
 
 //////////////////////////////////////////////////////////////////////////////
 // Dispatch
 //////////////////////////////////////////////////////////////////////////////
 
-void vpx_highbd_blend_mask6_sse4_1(uint8_t *dst_8, uint32_t dst_stride,
-                                   uint8_t *src0_8, uint32_t src0_stride,
-                                   uint8_t *src1_8, uint32_t src1_stride,
-                                   const uint8_t *mask, uint32_t mask_stride,
-                                   int h, int w, int suby, int subx, int bd) {
+void vpx_highbd_blend_mask6b_sse4_1(uint8_t *dst_8, uint32_t dst_stride,
+                                    uint8_t *src0_8, uint32_t src0_stride,
+                                    uint8_t *src1_8, uint32_t src1_stride,
+                                    const uint8_t *mask, uint32_t mask_stride,
+                                    int h, int w, int suby, int subx, int bd) {
   uint16_t *const dst = CONVERT_TO_SHORTPTR(dst_8);
   uint16_t *const src0 = CONVERT_TO_SHORTPTR(src0_8);
   uint16_t *const src1 = CONVERT_TO_SHORTPTR(src1_8);
@@ -1078,20 +1077,20 @@ void vpx_highbd_blend_mask6_sse4_1(uint8_t *dst_8, uint32_t dst_stride,
   static blend_fn blend[2][2][2][2] = {  // bd_index X width_index X subx X suby
     {   // bd == 8 or 10
       {     // w % 8 == 0
-        {blend_mask6_b10_w8n_sse4_1, blend_mask6_b10_sy_w8n_sse4_1},
-        {blend_mask6_b10_sx_w8n_sse4_1, blend_mask6_b10_sx_sy_w8n_sse4_1}
+        {blend_mask6b_b10_w8n_sse4_1, blend_mask6b_b10_sy_w8n_sse4_1},
+        {blend_mask6b_b10_sx_w8n_sse4_1, blend_mask6b_b10_sx_sy_w8n_sse4_1}
       }, {  // w == 4
-        {blend_mask6_b10_w4_sse4_1, blend_mask6_b10_sy_w4_sse4_1},
-        {blend_mask6_b10_sx_w4_sse4_1, blend_mask6_b10_sx_sy_w4_sse4_1}
+        {blend_mask6b_b10_w4_sse4_1, blend_mask6b_b10_sy_w4_sse4_1},
+        {blend_mask6b_b10_sx_w4_sse4_1, blend_mask6b_b10_sx_sy_w4_sse4_1}
       }
     },
     {   // bd == 12
       {     // w % 8 == 0
-        {blend_mask6_b12_w8n_sse4_1, blend_mask6_b12_sy_w8n_sse4_1},
-        {blend_mask6_b12_sx_w8n_sse4_1, blend_mask6_b12_sx_sy_w8n_sse4_1}
+        {blend_mask6b_b12_w8n_sse4_1, blend_mask6b_b12_sy_w8n_sse4_1},
+        {blend_mask6b_b12_sx_w8n_sse4_1, blend_mask6b_b12_sx_sy_w8n_sse4_1}
       }, {  // w == 4
-        {blend_mask6_b12_w4_sse4_1, blend_mask6_b12_sy_w4_sse4_1},
-        {blend_mask6_b12_sx_w4_sse4_1, blend_mask6_b12_sx_sy_w4_sse4_1}
+        {blend_mask6b_b12_w4_sse4_1, blend_mask6b_b12_sy_w4_sse4_1},
+        {blend_mask6b_b12_sx_w4_sse4_1, blend_mask6b_b12_sx_sy_w4_sse4_1}
       }
     }
   };