diff options
author | 2015-03-03 07:46:15 -0800 | |
---|---|---|
committer | 2015-03-03 07:46:15 -0800 | |
commit | 60d2a32b2dbbaabf4a0c133c8d3ff5ad888b8e5e (patch) | |
tree | b5f30fb2ea6ae04b1f5c60d943274fea2878c466 /src/opts/SkPMFloat_neon.h | |
parent | cff10b21a9934afc540d121b493b204335829589 (diff) |
Make SkPMFloats store floats in [0,255] instead of [0,1].
This pushes the cost of the *255 and *1/255 conversions onto only those code
paths that need it. We're not doing it any more efficiently than can be done
with Sk4f.
In microbenchmark isolation, this is about a 15% speedup.
BUG=skia:
NOPRESUBMIT=true
Review URL: https://codereview.chromium.org/973603002
Diffstat (limited to 'src/opts/SkPMFloat_neon.h')
-rw-r--r-- | src/opts/SkPMFloat_neon.h | 16 |
1 files changed, 6 insertions, 10 deletions
diff --git a/src/opts/SkPMFloat_neon.h b/src/opts/SkPMFloat_neon.h index 004b6588dd..15ba3a58e4 100644 --- a/src/opts/SkPMFloat_neon.h +++ b/src/opts/SkPMFloat_neon.h @@ -2,11 +2,10 @@ #include <arm_neon.h> // For set(), we widen our 8 bit components (fix8) to 8-bit components in 16 bits (fix8_16), -// then widen those to 8-bit-in-32-bits (fix8_32), convert those to floats (scaled), -// then finally scale those down from [0.0f, 255.0f] to [0.0f, 1.0f] into fColor. +// then widen those to 8-bit-in-32-bits (fix8_32), and finally convert those to floats. -// get() and clamped() do the opposite, working from [0.0f, 1.0f] floats to [0.0f, 255.0f], -// to 8-bit-in-32-bit, to 8-bit-in-16-bit, back down to 8-bit components. +// get() and clamped() do the opposite, working from floats to 8-bit-in-32-bit, +// to 8-bit-in-16-bit, back down to 8-bit components. // clamped() uses vqmovn to clamp while narrowing instead of just narrowing with vmovn. inline void SkPMFloat::set(SkPMColor c) { @@ -14,15 +13,13 @@ inline void SkPMFloat::set(SkPMColor c) { uint8x8_t fix8 = (uint8x8_t)vdup_n_u32(c); uint16x8_t fix8_16 = vmovl_u8(fix8); uint32x4_t fix8_32 = vmovl_u16(vget_low_u16(fix8_16)); - float32x4_t scaled = vcvtq_f32_u32(fix8_32); - vst1q_f32(fColor, vmulq_f32(scaled, vdupq_n_f32(1.0f/255.0f))); + vst1q_f32(fColor, vcvtq_f32_u32(fix8_32)); SkASSERT(this->isValid()); } inline SkPMColor SkPMFloat::get() const { SkASSERT(this->isValid()); - float32x4_t scaled = vmulq_f32(vld1q_f32(fColor), vdupq_n_f32(255.0f)); - uint32x4_t fix8_32 = vcvtq_u32_f32(scaled); + uint32x4_t fix8_32 = vcvtq_u32_f32(vld1q_f32(fColor)); uint16x4_t fix8_16 = vmovn_u32(fix8_32); uint8x8_t fix8 = vmovn_u16(vcombine_u16(fix8_16, vdup_n_u16(0))); SkPMColor c = vget_lane_u32((uint32x2_t)fix8, 0); @@ -31,8 +28,7 @@ inline SkPMColor SkPMFloat::get() const { } inline SkPMColor SkPMFloat::clamped() const { - float32x4_t scaled = vmulq_f32(vld1q_f32(fColor), vdupq_n_f32(255.0f)); - uint32x4_t fix8_32 = vcvtq_u32_f32(scaled); + uint32x4_t fix8_32 = vcvtq_u32_f32(vld1q_f32(fColor)); uint16x4_t fix8_16 = vqmovn_u32(fix8_32); uint8x8_t fix8 = vqmovn_u16(vcombine_u16(fix8_16, vdup_n_u16(0))); SkPMColor c = vget_lane_u32((uint32x2_t)fix8, 0); |