Fix for unaligned automatic variable address in SSE routines

Fix for unaligned automatic variable address in SSE routines. This used to work on linux but crashed on windows when rendering antialiased glyphs. I tracked the problem down to an alignment issue.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1817813002

Committed: https://skia.googlesource.com/skia/+/f8ee67edb8e5fed34bf4d5208bce89ea8be1ac1e

[Marco A., 2016-03-19 21:29:34]

Description was changed from

==========
Fix for unaligned automatic variable address in SSE routines

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

to

==========
Fix for unaligned automatic variable address in SSE routines

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

[Marco A., 2016-03-19 21:32:28]

Description was changed from

==========
Fix for unaligned automatic variable address in SSE routines

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

to

==========
Fix for unaligned automatic variable address in SSE routines. This used to work
on linux but crashed on windows when rendering antialiased glyphs. I tracked the
problem down to an alignment issue.

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

[Marco A., 2016-03-19 21:32:28]

marco.diiga@gmail.com changed reviewers:
+ mtklein@google.com

[mtklein, 2016-03-20 20:22:19]

https://codereview.chromium.org/1817813002/diff/1/src/opts/SkOpts_sse41.cpp
File src/opts/SkOpts_sse41.cpp (right):

https://codereview.chromium.org/1817813002/diff/1/src/opts/SkOpts_sse41.cpp#n...
src/opts/SkOpts_sse41.cpp:59: return _mm_shuffle_epi8(xyzw,
_mm_load_si128((const __m128i*)mask));
Hmm.  It's odd that I wrote this function this way at all.  Let's just write

static __m128i replicate_coverage(__m128i xyzw) {
    return _mm_shuffle_epi8(xyzw, _mm_setr_epi8(0,0,0,0, 1,1,1,1, 2,2,2,2,
3,3,3,3));
}

[Marco A., 2016-03-20 22:42:15]

On 2016/03/20 20:22:19, mtklein wrote:
> https://codereview.chromium.org/1817813002/diff/1/src/opts/SkOpts_sse41.cpp
> File src/opts/SkOpts_sse41.cpp (right):
> 
>
https://codereview.chromium.org/1817813002/diff/1/src/opts/SkOpts_sse41.cpp#n...
> src/opts/SkOpts_sse41.cpp:59: return _mm_shuffle_epi8(xyzw,
> _mm_load_si128((const __m128i*)mask));
> Hmm.  It's odd that I wrote this function this way at all.  Let's just write
> 
> static __m128i replicate_coverage(__m128i xyzw) {
>     return _mm_shuffle_epi8(xyzw, _mm_setr_epi8(0,0,0,0, 1,1,1,1, 2,2,2,2,
> 3,3,3,3));
> }

Approved. Updated.

[mtklein, 2016-03-21 12:31:15]

The CQ bit was checked by mtklein@google.com

[mtklein, 2016-03-21 12:31:15]

lgtm

[commit-bot: I haz the power, 2016-03-21 12:31:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1817813002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1817813002/20001

[commit-bot: I haz the power, 2016-03-21 12:31:19]

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

[commit-bot: I haz the power, 2016-03-21 12:31:20]

The author marco.diiga@gmail.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[Marco A., 2016-03-21 12:35:08]

On 2016/03/21 12:31:20, commit-bot: I haz the power wrote:
> The author mailto:marco.diiga@gmail.com has not signed Google Contributor
License
> Agreement. Please visit https://cla.developers.google.com to sign and manage
> CLA.

Signed.

[mtklein, 2016-03-21 12:41:58]

The CQ bit was checked by mtklein@google.com

[commit-bot: I haz the power, 2016-03-21 12:42:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1817813002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1817813002/20001

[commit-bot: I haz the power, 2016-03-21 12:43:36]

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

[commit-bot: I haz the power, 2016-03-21 12:43:37]

Try jobs failed on following builders:
  skia_presubmit-Trybot on client.skia.fyi (JOB_FAILED,
http://build.chromium.org/p/client.skia.fyi/builders/skia_presubmit-Trybot/bu...)

[mtklein, 2016-03-21 12:55:02]

The CQ bit was checked by mtklein@google.com

[mtklein, 2016-03-21 12:55:03]

lgtm

[commit-bot: I haz the power, 2016-03-21 12:55:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1817813002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1817813002/40001

[commit-bot: I haz the power, 2016-03-21 13:04:47]

Description was changed from

==========
Fix for unaligned automatic variable address in SSE routines. This used to work
on linux but crashed on windows when rendering antialiased glyphs. I tracked the
problem down to an alignment issue.

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

to

==========
Fix for unaligned automatic variable address in SSE routines. This used to work
on linux but crashed on windows when rendering antialiased glyphs. I tracked the
problem down to an alignment issue.

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/+/f8ee67edb8e5fed34bf4d5208bce89ea8be1ac1e
==========

[commit-bot: I haz the power, 2016-03-21 13:04:48]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/f8ee67edb8e5fed34bf4d5208bce89ea8be1ac1e