SkNx: add Sk4u

src/core/SkRasterPipelineBlitter.cpp

 /*
  * Copyright 2016 Google Inc.
  *
  * Use of this source code is governed by a BSD-style license that can be
  * found in the LICENSE file.
  */
 
 #include "SkBlitter.h"
 #include "SkColor.h"
 #include "SkColorFilter.h"
@@ skipping 279 matching lines @@
     dg = { sk_linear_from_srgb[(ptr[0] >> SK_G32_SHIFT) & 0xff],
            sk_linear_from_srgb[(ptr[1] >> SK_G32_SHIFT) & 0xff],
            sk_linear_from_srgb[(ptr[2] >> SK_G32_SHIFT) & 0xff],
            sk_linear_from_srgb[(ptr[3] >> SK_G32_SHIFT) & 0xff] };
 
     db = { sk_linear_from_srgb[(ptr[0] >> SK_B32_SHIFT) & 0xff],
            sk_linear_from_srgb[(ptr[1] >> SK_B32_SHIFT) & 0xff],
            sk_linear_from_srgb[(ptr[2] >> SK_B32_SHIFT) & 0xff],
            sk_linear_from_srgb[(ptr[3] >> SK_B32_SHIFT) & 0xff] };
 
-    // TODO: this >> doesn't really need mask if we make it logical instead of arithmetic.
-    da = SkNx_cast<float>((Sk4i::Load(ptr) >> SK_A32_SHIFT) & 0xff) * (1/255.0f);
+    da = SkNx_cast<float>(Sk4u::Load(ptr) >> SK_A32_SHIFT) * (1/255.0f);
 
     st->next(x, r,g,b,a, dr,dg,db,da);
 }
 
 // Tail variant of load_d_srgb() handling 1 pixel at a time.
 static void SK_VECTORCALL load_d_srgb_1(SkRasterPipeline::Stage* st, size_t x,
                                         Sk4f  r, Sk4f  g, Sk4f  b, Sk4f  a,
                                         Sk4f dr, Sk4f dg, Sk4f db, Sk4f da) {
     auto ptr = st->ctx<const uint32_t*>() + x;
 
@@ skipping 180 matching lines @@
             case SkMask::kLCD16_Format:
                 p.append(lerp_lcd16, lerp_lcd16_1, mask.getAddrLCD16(x,y)-x);
                 break;
             default: break;
         }
         this->append_store(&p, dst);
 
         p.run(x, clip.width());
     }
 }