Add clamp to sk_linear_to_srgb, reorder instructions

Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 346us
to2Dot2: 404us -> 387us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot

Committed: https://skia.googlesource.com/skia/+/a5ba329c81dd6e21f8d610b56416d9f45259fa42

[msarett, 2016-07-22 12:25:27]

Description was changed from

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems more
optimal to save clamping until the end (and not break up the mul
pipeline).
toSRGB:  371us -> 342us
to2Dot2: 404us -> 385us

Also, just for fun, optimize away shifts by 0.

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

to

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems more
optimal to save clamping until the end (and not break up the mul
pipeline).
toSRGB:  371us -> 342us
to2Dot2: 404us -> 385us

Also, just for fun, optimize away shifts by 0.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[msarett, 2016-07-22 12:29:41]

Description was changed from

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems more
optimal to save clamping until the end (and not break up the mul
pipeline).
toSRGB:  371us -> 342us
to2Dot2: 404us -> 385us

Also, just for fun, optimize away shifts by 0.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 342us
to2Dot2: 404us -> 385us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

Also, just for fun, optimize away shifts by 0.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[msarett, 2016-07-22 12:29:57]

Patchset #2 (id:20001) has been deleted

[msarett, 2016-07-22 12:35:07]

Description was changed from

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 342us
to2Dot2: 404us -> 385us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

Also, just for fun, optimize away shifts by 0.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 342us
to2Dot2: 404us -> 385us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[msarett, 2016-07-22 12:36:14]

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

[msarett, 2016-07-22 12:55:20]

Description was changed from

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 342us
to2Dot2: 404us -> 385us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 346us
to2Dot2: 404us -> 387us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

[msarett, 2016-07-22 12:55:30]

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

[commit-bot: I haz the power, 2016-07-22 12:55:33]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[mtklein, 2016-07-22 13:05:25]

lgtm

I do think the other use cases are going to be <0 or >1, so a clamp is probably
inevitable.

[commit-bot: I haz the power, 2016-07-22 13:11:48]

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

[commit-bot: I haz the power, 2016-07-22 13:11:49]

Dry run: Try jobs failed on following builders:
  Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Debug-Trybot on master.client.skia
(JOB_FAILED,
http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2...)

[msarett, 2016-07-22 13:13:44]

The CQ bit was checked by msarett@google.com

[commit-bot: I haz the power, 2016-07-22 13:13:50]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-07-22 13:23:02]

Description was changed from

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 346us
to2Dot2: 404us -> 387us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
==========

to

==========
Add clamp to sk_linear_to_srgb, reorder instructions

Improves performance for xforms toSRGB and to2Dot2.  Seems
more optimal to save clamping until the end.  That way we
don't stall the mul pipeline with a min/max.
toSRGB:  371us -> 346us
to2Dot2: 404us -> 387us

FWIW, it probably makes sense to clamp inside
sk_linear_to_srgb anyway.  If not, we should potentially
provide two versions (one that clamps and one that
doesn't).

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2173803002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot

Committed:
https://skia.googlesource.com/skia/+/a5ba329c81dd6e21f8d610b56416d9f45259fa42
==========

[commit-bot: I haz the power, 2016-07-22 13:23:03]

Committed patchset #3 (id:60001) as
https://skia.googlesource.com/skia/+/a5ba329c81dd6e21f8d610b56416d9f45259fa42