Reland "Respect full precision for RGB16 PNGs" (part 2)

This lands all the new xform hooks but no change to src/codec.
So the new decode features are turned off.

I'm relanding this in pieces to try to bisect a
strange MSAN error.

Original CL:
https://skia-review.googlesource.com/c/7085/

BUG=skia:

CQ_INCLUDE_TRYBOTS=skia.primary:Test-Ubuntu-Clang-GCE-CPU-AVX2-x86_64-Debug-MSAN,Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD,Build-Ubuntu-Clang-x86_64-Release-Fast


Change-Id: I451a2a29c73ca475e9e7a5ded58d4948d6b8be19
Reviewed-on: https://skia-review.googlesource.com/7277
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Matt Sarett <msarett@google.com>

1 files changed, 16 insertions, 0 deletions

diff --git a/src/core/SkNx.h b/src/core/SkNx.h
index a75495cd2a..5df575342c 100644
--- a/src/core/SkNx.h
+++ b/src/core/SkNx.h
@@ -68,6 +68,16 @@ struct SkNx {
         *c = SkNx{cl, ch};
         *d = SkNx{dl, dh};
     }
+    AI static void Load3(const void* vptr, SkNx* a, SkNx* b, SkNx* c) {
+        auto ptr = (const char*)vptr;
+        Half al, bl, cl,
+             ah, bh, ch;
+        Half::Load3(ptr                  , &al, &bl, &cl);
+        Half::Load3(ptr + 3*N/2*sizeof(T), &ah, &bh, &ch);
+        *a = SkNx{al, ah};
+        *b = SkNx{bl, bh};
+        *c = SkNx{cl, ch};
+    }
     AI static void Store4(void* vptr, const SkNx& a, const SkNx& b, const SkNx& c, const SkNx& d) {
         auto ptr = (char*)vptr;
         Half::Store4(ptr,                   a.fLo, b.fLo, c.fLo, d.fLo);
@@ -149,6 +159,12 @@ struct SkNx<1,T> {
         *c = Load(ptr + 2*sizeof(T));
         *d = Load(ptr + 3*sizeof(T));
     }
+    AI static void Load3(const void* vptr, SkNx* a, SkNx* b, SkNx* c) {
+        auto ptr = (const char*)vptr;
+        *a = Load(ptr + 0*sizeof(T));
+        *b = Load(ptr + 1*sizeof(T));
+        *c = Load(ptr + 2*sizeof(T));
+    }
     AI static void Store4(void* vptr, const SkNx& a, const SkNx& b, const SkNx& c, const SkNx& d) {
         auto ptr = (char*)vptr;
         a.store(ptr + 0*sizeof(T));