565 support for SIMD xfermodes

This uses the most basic approach possible:
  - to load an Sk4px from 565, convert to SkPMColors on the stack serially then load those SkPMColors.
  - to store an Sk4px to 565, store to SkPMColors on the stack then convert to 565 serially.

Clearly, we can optimize these loads and stores.  That's a TODO.

The code using SkPMFloat is the same idea but a little more long-term viable, as we're only operating on one pixel at a time anyway.  We could probably write 565 <-> SkPMFloat methods, but I'd rather not until it's really compelling.

The speedups are varied but similar across SSE and NEON: a few uninteresting, many 50% faster, some 2x faster, and SoftLight ~4x faster.

This will cause minor GM diffs, but I don't think any layout test changes.

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/942930dcaa51f66d82cdaf46ae62efebd16c8cd0

Committed: https://skia.googlesource.com/skia/+/860dcaa2ddfdadc050af4f943a84a9d499315066

Review URL: https://codereview.chromium.org/1245673002

1 files changed, 31 insertions, 0 deletions

diff --git a/src/opts/Sk4px_none.h b/src/opts/Sk4px_none.h
index ba13e58fb5..540edb821d 100644
--- a/src/opts/Sk4px_none.h
+++ b/src/opts/Sk4px_none.h
@@ -100,4 +100,35 @@ inline Sk4px Sk4px::zeroColors() const {
                  0,0,0, this->kth<15>());
 }
 
+inline Sk4px Sk4px::Load4(const SkPMColor16 src[4]) {
+    SkPMColor src32[4];
+    for (int i = 0; i < 4; i++) { src32[i] = SkPixel16ToPixel32(src[i]); }
+    return Load4(src32);
+}
+inline Sk4px Sk4px::Load2(const SkPMColor16 src[2]) {
+    SkPMColor src32[2];
+    for (int i = 0; i < 2; i++) { src32[i] = SkPixel16ToPixel32(src[i]); }
+    return Load2(src32);
+}
+inline Sk4px Sk4px::Load1(const SkPMColor16 src[1]) {
+    SkPMColor src32 = SkPixel16ToPixel32(src[0]);
+    return Load1(&src32);
+}
+
+inline void Sk4px::store4(SkPMColor16 dst[4]) const {
+    SkPMColor dst32[4];
+    this->store4(dst32);
+    for (int i = 0; i < 4; i++) { dst[i] = SkPixel32ToPixel16(dst32[i]); }
+}
+inline void Sk4px::store2(SkPMColor16 dst[2]) const {
+    SkPMColor dst32[2];
+    this->store2(dst32);
+    for (int i = 0; i < 2; i++) { dst[i] = SkPixel32ToPixel16(dst32[i]); }
+}
+inline void Sk4px::store1(SkPMColor16 dst[1]) const {
+    SkPMColor dst32;
+    this->store1(&dst32);
+    dst[0] = SkPixel32ToPixel16(dst32);
+}
+
 }  // namespace