Convert Color32 code to perfect blend.

Before we commit to blend_256_round_alt, let's make sure blend_perfect is
really slower in practice (i.e. regresses on perf.skia.org).

blend_perfect is really the most desirable algorithm if we can afford it.  Not
only is it correct, but it's easy to think about and break into correct pieces:
for instance, its div255() doesn't require any coordination with the multiply.

This looks like a 30% hit according to microbenches.  That said, microbenches
said my previous change would be a 20-25% perf improvement, but it didn't end
up showing a significant effect at a high level.

As for correctness, I see a bunch of off-by-1 compared to blend_256_round_alt
(exactly what we'd expect), and one off-by-3 in a GM that looks like it has a
bunch of overdraw.

BUG=skia:

Review URL: https://codereview.chromium.org/1098913002

1 files changed, 12 insertions, 15 deletions

diff --git a/src/core/SkBlitRow_D32.cpp b/src/core/SkBlitRow_D32.cpp
index ac01e427bf..36bfa54095 100644
--- a/src/core/SkBlitRow_D32.cpp
+++ b/src/core/SkBlitRow_D32.cpp
@@ -142,11 +142,8 @@ SkBlitRow::Proc32 SkBlitRow::ColorProcFactory() {
 
 #define SK_SUPPORT_LEGACY_COLOR32_MATHx
 
-// Color32 and its SIMD specializations use the blend_256_round_alt algorithm
-// from tests/BlendTest.cpp.  It's not quite perfect, but it's never wrong in the
-// interesting edge cases, and it's quite a bit faster than blend_perfect.
-//
-// blend_256_round_alt is our currently blessed algorithm.  Please use it or an analogous one.
+// Color32 and its SIMD specializations use the blend_perfect algorithm from tests/BlendTest.cpp.
+// An acceptable alternative is blend_256_round_alt, which is faster but not quite perfect.
 void SkBlitRow::Color32(SkPMColor* SK_RESTRICT dst,
                         const SkPMColor* SK_RESTRICT src,
                         int count, SkPMColor color) {
@@ -156,19 +153,19 @@ void SkBlitRow::Color32(SkPMColor* SK_RESTRICT dst,
     }
 
     unsigned invA = 255 - SkGetPackedA32(color);
-#ifdef SK_SUPPORT_LEGACY_COLOR32_MATH  // blend_256_plus1_trunc, busted
-    unsigned round = 0;
-#else                          // blend_256_round_alt, good
-    invA += invA >> 7;
-    unsigned round = (128 << 16) + (128 << 0);
-#endif
-
     while (count --> 0) {
         // Our math is 16-bit, so we can do a little bit of SIMD in 32-bit registers.
         const uint32_t mask = 0x00FF00FF;
-        uint32_t rb = (((*src >> 0) & mask) * invA + round) >> 8,  // _r_b
-                 ag = (((*src >> 8) & mask) * invA + round) >> 0;  // a_g_
-        *dst = color + ((rb & mask) | (ag & ~mask));
+        uint32_t rb = (((*src >> 0) & mask) * invA),  // r_b_
+                 ag = (((*src >> 8) & mask) * invA);  // a_g_
+    #ifndef SK_SUPPORT_LEGACY_COLOR32_MATH
+        uint32_t round = (128 << 16) + (128 << 0);
+        rb += round;
+        ag += round;
+        rb += (rb & ~mask) >> 8;
+        ag += (ag & ~mask) >> 8;
+    #endif
+        *dst = color + (((rb>>8) & mask) | ((ag>>0) & ~mask));
         src++;
         dst++;
     }