gpu: Fix --disable-shader-name-hashing.

gpu: Fix --disable-shader-name-hashing.

This switch was broken at some point in the past. Restore it by
adding it to the list of supported switches for the gpu process.

BUG=715117
R=kbr@chromium.org

Review-Url: https://codereview.chromium.org/2843753002
Cr-Commit-Position: refs/heads/master@{#467014}
Committed: https://chromium.googlesource.com/chromium/src/+/cce4a756f84d5b8928269f7342aaccdf68be910c

[Jamie Madill, 2017-04-25 15:26:32]

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

[Jamie Madill, 2017-04-25 15:26:39]

PTAL Ken.

[commit-bot: I haz the power, 2017-04-25 15:26:59]

Dry run: CQ is trying da patch.

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

[Ken Russell (switch to Gerrit), 2017-04-25 16:27:34]

lgtm

We should add a test for this. One could be added to
webgl_conformance_integration_test.py -- by cloning either
_RunExtensionCoverageTest or _RunExtensionTest as a new test and running one
specific page with a specific browser command line argument. The page would use
WEBGL_debug_shaders to fetch the processed shader and verify that one of the
names isn't hashed.

Note though that the RestartBrowserIfNecessaryWithArgs mechanism should really
be refactored up from e.g. pixel_integration_test.py and other places to
gpu_integration_test.py first, so that webgl_conformance_integration_test.py
could pick it up.

Anyway, that's a larger task than this fix so I'm not suggesting to do it, but
this is an outline for how to do so if you want to prevent this from regressing
again.

[commit-bot: I haz the power, 2017-04-25 16:28:23]

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

[commit-bot: I haz the power, 2017-04-25 16:28:24]

Dry run: This issue passed the CQ dry run.

[Jamie Madill, 2017-04-25 16:34:57]

On 2017/04/25 16:27:34, Ken Russell wrote:
> lgtm
> 
> We should add a test for this. One could be added to
> webgl_conformance_integration_test.py -- by cloning either
> _RunExtensionCoverageTest or _RunExtensionTest as a new test and running one
> specific page with a specific browser command line argument. The page would
use
> WEBGL_debug_shaders to fetch the processed shader and verify that one of the
> names isn't hashed.
> 
> Note though that the RestartBrowserIfNecessaryWithArgs mechanism should really
> be refactored up from e.g. pixel_integration_test.py and other places to
> gpu_integration_test.py first, so that webgl_conformance_integration_test.py
> could pick it up.
> 
> Anyway, that's a larger task than this fix so I'm not suggesting to do it, but
> this is an outline for how to do so if you want to prevent this from
regressing
> again.

I'll defer that work for now if it's alright. Since it's a developer-only
feature
that doesn't see too much use I won't be too upset if it breaks again.

Thanks!

[Jamie Madill, 2017-04-25 16:35:02]

The CQ bit was checked by jmadill@chromium.org

[commit-bot: I haz the power, 2017-04-25 16:35:28]

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, 2017-04-25 16:40:11]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1493138102421330, "parent_rev":
"3d9408b382d593aa59f01dc5cb9ae8f0bc564e1d", "commit_rev":
"cce4a756f84d5b8928269f7342aaccdf68be910c"}

[commit-bot: I haz the power, 2017-04-25 16:40:20]

Description was changed from

==========
gpu: Fix --disable-shader-name-hashing.

This switch was broken at some point in the past. Restore it by
adding it to the list of supported switches for the gpu process.

BUG=715117
R=kbr@chromium.org
==========

to

==========
gpu: Fix --disable-shader-name-hashing.

This switch was broken at some point in the past. Restore it by
adding it to the list of supported switches for the gpu process.

BUG=715117
R=kbr@chromium.org

Review-Url: https://codereview.chromium.org/2843753002
Cr-Commit-Position: refs/heads/master@{#467014}
Committed:
https://chromium.googlesource.com/chromium/src/+/cce4a756f84d5b8928269f7342aa...
==========

[commit-bot: I haz the power, 2017-04-25 16:40:21]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/cce4a756f84d5b8928269f7342aa...

[Ken Russell (switch to Gerrit), 2017-04-25 16:50:08]

On 2017/04/25 16:34:57, Jamie Madill wrote:
> On 2017/04/25 16:27:34, Ken Russell wrote:
> > lgtm
> > 
> > We should add a test for this. One could be added to
> > webgl_conformance_integration_test.py -- by cloning either
> > _RunExtensionCoverageTest or _RunExtensionTest as a new test and running one
> > specific page with a specific browser command line argument. The page would
> use
> > WEBGL_debug_shaders to fetch the processed shader and verify that one of the
> > names isn't hashed.
> > 
> > Note though that the RestartBrowserIfNecessaryWithArgs mechanism should
really
> > be refactored up from e.g. pixel_integration_test.py and other places to
> > gpu_integration_test.py first, so that webgl_conformance_integration_test.py
> > could pick it up.
> > 
> > Anyway, that's a larger task than this fix so I'm not suggesting to do it,
but
> > this is an outline for how to do so if you want to prevent this from
> regressing
> > again.
> 
> I'll defer that work for now if it's alright. Since it's a developer-only
> feature
> that doesn't see too much use I won't be too upset if it breaks again.

Yes, that's fine. :) I only realized what a mountain of work it would be when
writing it up.