__vectorcall needs SSE

__vectorcall needs SSE

We don't have builders in this mode, but Mozilla does.

BUG=skia:4563
TBR=reed@google.com
No public API changes.

Committed: https://skia.googlesource.com/skia/+/14102ca4cf0c1053b3cb26733723595d0e6f1f95

[mtklein, 2015-11-16 22:47:41]

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

[commit-bot: I haz the power, 2015-11-16 22:47:48]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1445283002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1445283002/1

[mtklein_C, 2015-11-16 22:48:18]

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

[lsalzman1, 2015-11-16 22:53:18]

I tried a similar change and it didn't work, as it requires specifically SSE2.
SSE1 is not sufficient to make VS eat it (i.e. _M_X86_FP must be >= 2).

[mtklein_C, 2015-11-16 22:57:23]

On 2015/11/16 at 22:53:18, lsalzman wrote:
> I tried a similar change and it didn't work, as it requires specifically SSE2.
SSE1 is not sufficient to make VS eat it (i.e. _M_X86_FP must be >= 2).

Weird, but probably no big deal.  I don't see us ever using SSE without things
from SSE2 like byte unpacking.

Have another look?

[lsalzman1, 2015-11-16 23:01:44]

I will try that change out locally - might take me a bit, but I'll see if I can
get to it today. There was a reason I went with the particular _M_X86_FP check
directly in the proposed patch, but it was a while ago so I can't remember
offhand. I'll see if checking the SSE level macros works this time around with
latest Skia.

[lsalzman1, 2015-11-16 23:38:39]

On 2015/11/16 23:01:44, lsalzman1 wrote:
> I will try that change out locally - might take me a bit, but I'll see if I
can
> get to it today. There was a reason I went with the particular _M_X86_FP check
> directly in the proposed patch, but it was a while ago so I can't remember
> offhand. I'll see if checking the SSE level macros works this time around with
> latest Skia.

Okay, I tested it with the SSE2 level check. Everything built. lgtm

[mtklein, 2015-11-17 00:02:47]

The CQ bit was checked by mtklein@google.com

[commit-bot: I haz the power, 2015-11-17 00:02:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1445283002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1445283002/20001

[mtklein, 2015-11-17 00:03:35]

Description was changed from

==========
__vectorcall needs SSE

We don't have builders in this mode, but Mozilla does.

BUG=skia:4563
==========

to

==========
__vectorcall needs SSE

We don't have builders in this mode, but Mozilla does.

BUG=skia:4563

TBR=reed@google.com
No public API changes.
==========

[commit-bot: I haz the power, 2015-11-17 00:03:43]

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

[commit-bot: I haz the power, 2015-11-17 00:03:44]

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, 2015-11-17 00:07:59]

The CQ bit was checked by mtklein@google.com

[commit-bot: I haz the power, 2015-11-17 00:08:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1445283002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1445283002/20001

[commit-bot: I haz the power, 2015-11-17 00:20:33]

Committed patchset #2 (id:20001) as
https://skia.googlesource.com/skia/+/14102ca4cf0c1053b3cb26733723595d0e6f1f95