Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot

Committed: https://skia.googlesource.com/skia/+/9db43ac4ee1a83a4f7b332fe6c00f592b6237349

[mtklein_C, 2015-11-30 13:52:39]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.

Other small tweaks here:
   - Sk4f::toBytes() (one at a time) is rather faster with _mm_shuffle_epi8
   - Sk4f::FromBytes() is rather faster with _mm_cvtepu8_epi32

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

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.

Other small tweaks here:
   - Sk4f::toBytes() (one at a time) is rather faster with _mm_shuffle_epi8
   - Sk4f::FromBytes() is rather faster with _mm_cvtepu8_epi32

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[mtklein, 2015-11-30 13:54:55]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.

Other small tweaks here:
   - Sk4f::toBytes() (one at a time) is rather faster with _mm_shuffle_epi8
   - Sk4f::FromBytes() is rather faster with _mm_cvtepu8_epi32

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[mtklein, 2015-11-30 22:26:06]

The CQ bit was checked by mtklein@google.com to run a CQ dry run

[commit-bot: I haz the power, 2015-11-30 22:26:19]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1483953002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1483953002/100001

[mtklein, 2015-11-30 22:27:33]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g.
  gradient_linear_clamp_3color: 147µs -> 103µs

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[mtklein, 2015-11-30 22:38:10]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g.
  gradient_linear_clamp_3color: 147µs -> 103µs

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g.
  gradient_linear_clamp_3color: x86 147µs -> 103µs, arm64 2.03ms -> 648µs

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[commit-bot: I haz the power, 2015-11-30 22:38:32]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-11-30 22:38:32]

Dry run: This issue passed the CQ dry run.

[mtklein, 2015-11-30 22:42:10]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g.
  gradient_linear_clamp_3color: x86 147µs -> 103µs, arm64 2.03ms -> 648µs

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs
 arm64 2.03ms -> 648µs

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[mtklein, 2015-11-30 22:45:58]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs
 arm64 2.03ms -> 648µs

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs
 arm64 2.03ms -> 648µs
 armv7 1.12ms -> 489µs  (same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[mtklein, 2015-11-30 22:46:55]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs
 arm64 2.03ms -> 648µs
 armv7 1.12ms -> 489µs  (same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[mtklein_C, 2015-11-30 22:47:18]

mtklein@chromium.org changed reviewers:
+ reed@google.com
- mtklein@google.com

[mtklein, 2015-11-30 23:19:05]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot,Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot
==========

[mtklein, 2015-11-30 23:19:09]

The CQ bit was checked by mtklein@google.com to run a CQ dry run

[commit-bot: I haz the power, 2015-11-30 23:19:22]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1483953002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1483953002/100001

[commit-bot: I haz the power, 2015-11-30 23:19:41]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-11-30 23:19:41]

Dry run: Try jobs failed on following builders:
  Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot on client.skia
(JOB_FAILED, no build URL)

[mtklein, 2015-11-30 23:20:01]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot,Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot
==========

[mtklein, 2015-11-30 23:20:06]

The CQ bit was checked by mtklein@google.com to run a CQ dry run

[commit-bot: I haz the power, 2015-11-30 23:20:13]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1483953002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1483953002/100001

[commit-bot: I haz the power, 2015-12-01 12:35:27]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-12-01 12:35:29]

Dry run: This issue passed the CQ dry run.

[reed1, 2015-12-01 14:34:26]

wow, killer improvement on arm64!

lgtm

[mtklein, 2015-12-01 15:09:46]

On 2015/12/01 at 14:34:26, reed wrote:
> wow, killer improvement on arm64!
> 
> lgtm

Or rather, why is arm64 so bad?

[mtklein, 2015-12-01 15:09:52]

The CQ bit was checked by mtklein@google.com

[commit-bot: I haz the power, 2015-12-01 15:09:56]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1483953002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1483953002/100001

[commit-bot: I haz the power, 2015-12-01 15:10:24]

Description was changed from

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot
==========

to

==========
Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

This is a big speedup for float -> byte.  E.g. gradient_linear_clamp_3color:
 x86-64 147µs -> 103µs  (Broadwell MBP)
 arm64 2.03ms -> 648µs  (Galaxy S6)
 armv7 1.12ms -> 489µs  (Galaxy S6, same device!)

BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot

Committed:
https://skia.googlesource.com/skia/+/9db43ac4ee1a83a4f7b332fe6c00f592b6237349
==========

[commit-bot: I haz the power, 2015-12-01 15:10:25]

Committed patchset #6 (id:100001) as
https://skia.googlesource.com/skia/+/9db43ac4ee1a83a4f7b332fe6c00f592b6237349