skcpu: sse4.1 floor, f16c f16<->f32
- floor with roundps is about 4.5x faster when available
- f16 srcover_n is similar to but a little faster than the version in https://codereview.chromium.org/1884683002. This new one fuses the dst load/stores into the f16<->f32 conversions:
+0x180 movups (%r15), %xmm1
+0x184 vcvtph2ps (%rbx), %xmm2
+0x189 movaps %xmm1, %xmm3
+0x18c shufps $255, %xmm3, %xmm3
+0x190 movaps %xmm0, %xmm4
+0x193 subps %xmm3, %xmm4
+0x196 mulps %xmm2, %xmm4
+0x199 addps %xmm1, %xmm4
+0x19c vcvtps2ph $0, %xmm4, (%rbx)
+0x1a2 addq $16, %r15
+0x1a6 addq $8, %rbx
+0x1aa decl %r14d
+0x1ad jne +0x180
If we decide to land this it'd be a good idea to convert most or all users of SkFloatToHalf_01 and SkHalfToFloat_01 over to the pointer-based versions.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1891513002
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
Committed: https://skia.googlesource.com/skia/+/cbe3c1af987d622ea67ef560d855b41bb14a0ce9
Committed: https://skia.googlesource.com/skia/+/3faf74b8364491ca806f523fbb1d8a97be592663
Review URL: https://codereview.chromium.org/1891513002
diff --git a/src/core/SkXfermodeF16.cpp b/src/core/SkXfermodeF16.cpp
index dfcefa2..2c6873f 100644
--- a/src/core/SkXfermodeF16.cpp
+++ b/src/core/SkXfermodeF16.cpp
@@ -134,15 +134,13 @@
static void srcover_n(const SkXfermode*, uint64_t dst[], const SkPM4f src[], int count,
const SkAlpha aa[]) {
for (int i = 0; i < count; ++i) {
- const Sk4f s4 = Sk4f::Load(src[i].fVec);
- const Sk4f dst_scale = Sk4f(1 - get_alpha(s4));
- const Sk4f d4 = SkHalfToFloat_01(dst[i]);
- const Sk4f r4 = s4 + d4 * dst_scale;
+ Sk4f s = Sk4f::Load(src+i),
+ d = SkHalfToFloat_01(dst+i),
+ r = s + d*(1.0f - SkNx_shuffle<3,3,3,3>(s));
if (aa) {
- dst[i] = SkFloatToHalf_01(lerp_by_coverage(r4, d4, aa[i]));
- } else {
- dst[i] = SkFloatToHalf_01(r4);
+ r = lerp_by_coverage(r, d, aa[i]);
}
+ SkFloatToHalf_01(r, dst+i);
}
}