SSE optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul

SSE optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul

Swizzle Runtime (Dell Venue 8)
Unpremul 0.17x
Premul   0.20x

PNG Decode Runtime on GrayAlpha Encoded PNGs (Dell Venue 8)
Unpremul Regular  0.91x
Unpremul ZeroInit 0.92x
Premul   Regular  0.84x
Premul   ZeroInit 0.85x

BUG=skia:4767
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666853002
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot

Committed: https://skia.googlesource.com/skia/+/095742419d0277a4fb0d499a05ff29b7506f1c5e

[msarett, 2016-02-03 22:21:19]

Description was changed from

==========
SSE optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul

Swizzle Runtime (Dell Venue 8)
Unpremul 0.17x
Premul   0.20x

PNG Decode Runtime on GrayAlpha Encoded PNGs (Dell Venue 8)
Unpremul Regular  0.91x
Unpremul ZeroInit 0.92x
Premul   Regular  0.84x
Premul   ZeroInit 0.85x

BUG=skia:4767
==========

to

==========
SSE optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul

Swizzle Runtime (Dell Venue 8)
Unpremul 0.17x
Premul   0.20x

PNG Decode Runtime on GrayAlpha Encoded PNGs (Dell Venue 8)
Unpremul Regular  0.91x
Unpremul ZeroInit 0.92x
Premul   Regular  0.84x
Premul   ZeroInit 0.85x

BUG=skia:4767
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[msarett, 2016-02-03 22:23:35]

msarett@google.com changed reviewers:
+ mtklein@google.com

[msarett, 2016-02-03 22:23:35]

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h
File src/opts/SkSwizzler_opts.h (right):

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h#...
src/opts/SkSwizzler_opts.h:583: static void grayA_to_RGBA(uint32_t dst[], const
void* vsrc, int count) {
pshufb was a little slower on mobile, but closer this time.

[mtklein, 2016-02-03 22:46:32]

lgtm

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h
File src/opts/SkSwizzler_opts.h (right):

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h#...
src/opts/SkSwizzler_opts.h:590: __m128i gg = _mm_or_si128(_mm_and_si128(ga,
mask), _mm_slli_epi16(ga, 8));
Might be easier to digest all together as

__m128i gg = _mm_or_si128(_mm_and_si128(ga, _mm_set1_epi16(0x00FF)),
                          _mm_slli_epi16(ga, 8));

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h#...
src/opts/SkSwizzler_opts.h:613: __m128i g0 = _mm_and_si128(grayA, mask);
ditto here, moving the mask inline,

__m128i g0 = _mm_and_si128(grayA, _mm_set1_epi16(0x00FF));

[msarett, 2016-02-03 22:59:50]

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h
File src/opts/SkSwizzler_opts.h (right):

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h#...
src/opts/SkSwizzler_opts.h:590: __m128i gg = _mm_or_si128(_mm_and_si128(ga,
mask), _mm_slli_epi16(ga, 8));
On 2016/02/03 22:46:32, mtklein wrote:
> Might be easier to digest all together as
> 
> __m128i gg = _mm_or_si128(_mm_and_si128(ga, _mm_set1_epi16(0x00FF)),
>                           _mm_slli_epi16(ga, 8));

Done.

https://codereview.chromium.org/1666853002/diff/1/src/opts/SkSwizzler_opts.h#...
src/opts/SkSwizzler_opts.h:613: __m128i g0 = _mm_and_si128(grayA, mask);
On 2016/02/03 22:46:32, mtklein wrote:
> ditto here, moving the mask inline,
> 
> __m128i g0 = _mm_and_si128(grayA, _mm_set1_epi16(0x00FF));

Done.

[msarett, 2016-02-03 22:59:53]

The CQ bit was checked by msarett@google.com

[msarett, 2016-02-03 22:59:54]

The patchset sent to the CQ was uploaded after l-g-t-m from mtklein@google.com
Link to the patchset: https://codereview.chromium.org/1666853002/#ps20001
(title: "Response to comments")

[commit-bot: I haz the power, 2016-02-03 23:00:02]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1666853002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1666853002/20001

[commit-bot: I haz the power, 2016-02-03 23:28:39]

Description was changed from

==========
SSE optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul

Swizzle Runtime (Dell Venue 8)
Unpremul 0.17x
Premul   0.20x

PNG Decode Runtime on GrayAlpha Encoded PNGs (Dell Venue 8)
Unpremul Regular  0.91x
Unpremul ZeroInit 0.92x
Premul   Regular  0.84x
Premul   ZeroInit 0.85x

BUG=skia:4767
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
SSE optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul

Swizzle Runtime (Dell Venue 8)
Unpremul 0.17x
Premul   0.20x

PNG Decode Runtime on GrayAlpha Encoded PNGs (Dell Venue 8)
Unpremul Regular  0.91x
Unpremul ZeroInit 0.92x
Premul   Regular  0.84x
Premul   ZeroInit 0.85x

BUG=skia:4767
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot

Committed:
https://skia.googlesource.com/skia/+/095742419d0277a4fb0d499a05ff29b7506f1c5e
==========

[commit-bot: I haz the power, 2016-02-03 23:28:40]

Committed patchset #2 (id:20001) as
https://skia.googlesource.com/skia/+/095742419d0277a4fb0d499a05ff29b7506f1c5e