diff options
author | 2016-07-14 11:02:09 -0700 | |
---|---|---|
committer | 2016-07-14 11:02:09 -0700 | |
commit | 3296bee70d074bb8094b3229dbe12fa016657e90 (patch) | |
tree | a6dc47b3805301e3416e3b9a2e361ac9d0217f15 /src/core/SkSpanProcs.cpp | |
parent | 7cf36ccb441956e9e35bc4c9b08bdf8bafd9d4bb (diff) |
Expand _01 half<->float limitation to _finite. Simplify.
It's become clear we need to sometimes deal with values <0 or >1.
I'm not yet convinced we care about NaN or +-inf.
We had some fairly clever tricks and optimizations here for NEON
and SSE. I've thrown them out in favor of a single implementation.
If we find the specializations mattered, we can certainly figure out
how to extend them to this new range/domain.
This happens to add a vectorized float -> half for ARMv7, which was
missing from the _01 version. (The SSE strategy was not portable to
platforms that flush denorm floats to zero.)
I've tested the full float range for FloatToHalf on my desktop and a 5x.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2145663003
CQ_INCLUDE_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
Review-Url: https://codereview.chromium.org/2145663003
Diffstat (limited to 'src/core/SkSpanProcs.cpp')
-rw-r--r-- | src/core/SkSpanProcs.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/core/SkSpanProcs.cpp b/src/core/SkSpanProcs.cpp index dddf7419e2..87dcbc0ee7 100644 --- a/src/core/SkSpanProcs.cpp +++ b/src/core/SkSpanProcs.cpp @@ -42,7 +42,7 @@ static void load_f16(const SkPixmap& src, int x, int y, SkPM4f span[], int count SkASSERT(src.addr64(x + count - 1, y)); for (int i = 0; i < count; ++i) { - SkHalfToFloat_01(addr[i]).store(span[i].fVec); + SkHalfToFloat_finite(addr[i]).store(span[i].fVec); } } |