Sk4x4f: Simplify x86 down to SSE2.

Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performs most uniformly
    across different chips.

This makes Sk4x4f fast on Windows and Linux, and may help mobile x86.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1817353005

Committed: https://skia.googlesource.com/skia/+/1443c6920c4b7aa80811c30ed9cdc81395d5df4f

[mtklein_C, 2016-03-23 02:11:11]

Description was changed from

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performst most uniformly
    across different chips.

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

to

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performst most uniformly
    across different chips.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[mtklein_C, 2016-03-23 02:11:42]

Description was changed from

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performst most uniformly
    across different chips.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performs most uniformly
    across different chips.

This makes Sk4x4f fast(er) on Windows, Linux, and mobile x86.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[mtklein_C, 2016-03-23 02:12:00]

Description was changed from

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performs most uniformly
    across different chips.

This makes Sk4x4f fast(er) on Windows, Linux, and mobile x86.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performs most uniformly
    across different chips.

This makes Sk4x4f fast on Windows and Linux, and may help mobile x86.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[mtklein_C, 2016-03-23 02:13:04]

The CQ bit was checked by mtklein@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-03-23 02:13:12]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1817353005/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1817353005/1

[mtklein_C, 2016-03-23 02:22:06]

mtklein@chromium.org changed reviewers:
+ fmalita@chromium.org

[mtklein_C, 2016-03-23 02:22:06]

The CQ bit was checked by mtklein@chromium.org

[commit-bot: I haz the power, 2016-03-23 02:22:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1817353005/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1817353005/20001

[commit-bot: I haz the power, 2016-03-23 02:22:17]

Note for Reviewers:
The CQ is waiting for an approval.
If you believe that the CL is not ready yet, or if you would like to L-G-T-M
with comments then please uncheck the CQ checkbox.

Waiting for LGTM from valid reviewer(s) till 2016-03-23 08:22 UTC

[commit-bot: I haz the power, 2016-03-23 02:23:19]

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

[commit-bot: I haz the power, 2016-03-23 02:23:20]

Try jobs failed on following builders:
  Test-Ubuntu-GCC-ShuttleA-GPU-GTX660-x86_64-Release-Trybot on client.skia
(JOB_FAILED,
http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-ShuttleA-GPU...)
  Build-Mac-Clang-x86_64-Release-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Mac-Clang-x86_...)
  Build-Ubuntu-Clang-x86_64-Debug-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-Clang-x...)
  Build-Ubuntu-GCC-x86_64-Release-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-x86...)

[mtklein, 2016-03-23 02:24:10]

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

[commit-bot: I haz the power, 2016-03-23 02:24:20]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1817353005/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1817353005/40001

[commit-bot: I haz the power, 2016-03-23 02:34:56]

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

[commit-bot: I haz the power, 2016-03-23 02:34:57]

Dry run: This issue passed the CQ dry run.

[mtklein_C, 2016-03-23 16:33:09]

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

[mtklein_C, 2016-03-23 16:33:09]

+Herb

[f(malita), 2016-03-23 16:51:08]

The CQ bit was checked by fmalita@chromium.org

[f(malita), 2016-03-23 16:51:09]

lgtm

[commit-bot: I haz the power, 2016-03-23 16:51:17]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1817353005/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1817353005/40001

[commit-bot: I haz the power, 2016-03-23 16:52:17]

Description was changed from

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performs most uniformly
    across different chips.

This makes Sk4x4f fast on Windows and Linux, and may help mobile x86.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Sk4x4f: Simplify x86 down to SSE2.

  - This drops the minimum requirement for Sk4x4f on x86 to SSE2 by
    removing calls to _mm_shuffle_epi8().  Instead we use good old
    shifting and masking.

  - Performance is very similar to SSSE3, close enough I'm having trouble
    telling which is faster.  I think we should let ourselves circle back
    on whether we need an SSSE3 version later.  When possible it's nice
    to stick to SSE2: it's most available, and performs most uniformly
    across different chips.

This makes Sk4x4f fast on Windows and Linux, and may help mobile x86.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/1443c6920c4b7aa80811c30ed9cdc81395d5df4f
==========

[commit-bot: I haz the power, 2016-03-23 16:52:18]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/1443c6920c4b7aa80811c30ed9cdc81395d5df4f

[herb_g, 2016-03-23 16:56:35]

You may be able to move the shifts into the mults and divs for normalization.

[herb_g, 2016-03-23 16:56:51]

lgtm