Set ecx to 0 when calling cpuid.

Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=630077
Committed: https://crrev.com/c0fe8911a5076819ae903e184078245355d0f1b8
Cr-Commit-Position: refs/heads/master@{#441167}

[mtklein_C, 2017-01-03 15:05:06]

mtklein@chromium.org changed reviewers:
+ fbarchard@chromium.org, mark@chromium.org

[mtklein_C, 2017-01-03 15:08:02]

Here's the UMA chart that made me suspicious:

https://uma.googleplex.com/p/chrome/histograms?endDate=20170101&dayCount=1&hi...

[mtklein_C, 2017-01-03 15:17:08]

Description was changed from

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.
==========

to

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=667457,630077
==========

[mtklein_C, 2017-01-03 15:18:46]

Description was changed from

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=667457,630077
==========

to

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=630077
==========

[Mark Mentovai, 2017-01-03 17:29:49]

LGTM. The cpuidex approach discussed on bug 630077 would also be appropriate.

[mtklein_C, 2017-01-03 17:32:51]

On 2017/01/03 at 17:29:49, mark wrote:
> LGTM. The cpuidex approach discussed on bug 630077 would also be appropriate.

Yeah, as written this basically is MSVC's __cpuid(), which is documented to
clear ecx before the cpuid instruction.  I don't see any need yet for other
values of ecx than 0, but if we did we'd be calling __cpuidex() on Windows.

[mtklein_C, 2017-01-03 17:33:05]

The CQ bit was checked by mtklein@chromium.org

[commit-bot: I haz the power, 2017-01-03 17:33:15]

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, 2017-01-03 18:33:47]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1483464785757540, "parent_rev":
"0ca2cac0c886112079dc019f444b197dbb985cb1", "commit_rev":
"f8588e00f08a152adaa0335281b9aac6bb039a82"}

[commit-bot: I haz the power, 2017-01-03 18:34:32]

Description was changed from

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=630077
==========

to

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=630077

Review-Url: https://codereview.chromium.org/2611683002
==========

[commit-bot: I haz the power, 2017-01-03 18:34:34]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2017-01-03 18:37:34]

Description was changed from

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=630077

Review-Url: https://codereview.chromium.org/2611683002
==========

to

==========
Set ecx to 0 when calling cpuid.

This should fix AVX2 detection on Linux and Mac.

In order to read the extended feature flags that we want in cpu_info7, we must
set eax to 7 and ecx to 0.  At head we're setting eax to 7 but leaving ecx
unspecified.  This means we only detect AVX2 support if we get very lucky!

I noticed this when looking at UMA: Window shows a reasonable rate of AVX2,
Linux a miniscule one, and Mac zero.  That made no sense to me, typing this
CL on a Mac that supports AVX2.  :)

This makes base/cpu agree with SkCpu, which we've been using to detect AVX2 for
a good while now, in Chromium and elsewhere.  We use __cpuid_count(7,0,...),
which boils down to the same thing: eax=7, ecx=0.

And... clang format at the presubmit's insistence.

BUG=630077

Committed: https://crrev.com/c0fe8911a5076819ae903e184078245355d0f1b8
Cr-Commit-Position: refs/heads/master@{#441167}
==========

[commit-bot: I haz the power, 2017-01-03 18:37:35]

Patchset 1 (id:??) landed as
https://crrev.com/c0fe8911a5076819ae903e184078245355d0f1b8
Cr-Commit-Position: refs/heads/master@{#441167}

[fbarchard1, 2017-01-12 22:22:10]

fbarchard@google.com changed reviewers:
+ fbarchard@google.com

[fbarchard1, 2017-01-12 22:22:10]

lgtm

FYI I'm using __cpuidex in libyuv, but always with 0.