Port S32A_opaque blit row to SkOpts.

src/opts/SkOpts_sse41.cpp

 /*
  * Copyright 2015 Google Inc.
  *
  * Use of this source code is governed by a BSD-style license that can be
  * found in the LICENSE file.
  */
 
 #include "SkOpts.h"
 
 #define SK_OPTS_NS sk_sse41
 #include "SkBlurImageFilter_opts.h"
+#include "SkBlitRow_opts.h"
 
 #ifndef SK_SUPPORT_LEGACY_X86_BLITS
 
-namespace sk_sse41 {
+namespace sk_sse41_new {

[djsollen, 2016/03/22 17:12:42]

why do we need to add _new?

[mtklein, 2016/03/22 17:39:28]

This is sort of dumb.

You know how the new SkBlitRow_opts.h code has a lot of
   // TODO: this math is bad and you should feel bad.
?  I started trying to fix some of those math problems a few months ago, and
that's the code here in this file.  The blit_row_color32 and blit_mask_d32_a8 in
this file should have correct math.

SkBlitRow_opts.h also has an older blit_row_color32.  Now that I'm mixing the
two files, I end up with two sk_sse41::blit_row_color32 functions, one older,
slower and incorrect, one newer faster and correct.  I've been dragging my heels
a little to rebase Blink and remove the old code.

 
 // An SSE register holding at most 64 bits of useful data in the low lanes.
 struct m64i {
     __m128i v;
     /*implicit*/ m64i(__m128i v) : v(v) {}
     operator __m128i() const { return v; }
 };
 
 // Load 4, 2, or 1 constant pixels or coverages (4x replicated).
 static __m128i next4(uint32_t val) { return _mm_set1_epi32(val); }
@@ skipping 178 matching lines @@
                 __m128i sc = scale(s,c),
                         AC = inv(alphas(sc));
                 return _mm_add_epi16(sc, scale(d,AC));
             }));
             dst += dstRB / sizeof(*dst);
             cov += covRB / sizeof(*cov);
         }
     }
 }
 
-}  // namespace sk_sse41
+}  // namespace sk_sse41_new
 
 #endif
 
 namespace SkOpts {
     void Init_sse41() {
         box_blur_xx = sk_sse41::box_blur_xx;
         box_blur_xy = sk_sse41::box_blur_xy;
         box_blur_yx = sk_sse41::box_blur_yx;
 
     #ifndef SK_SUPPORT_LEGACY_X86_BLITS
-        blit_row_color32 = sk_sse41::blit_row_color32;
-        blit_mask_d32_a8 = sk_sse41::blit_mask_d32_a8;
+        blit_row_color32 = sk_sse41_new::blit_row_color32;
+        blit_mask_d32_a8 = sk_sse41_new::blit_mask_d32_a8;
     #endif
+        blit_row_s32a_opaque = sk_sse41::blit_row_s32a_opaque;
     }
 }