Support arguments and driver bug workarounds in command_buffer_gles2

Support arguments and driver bug workarounds in command_buffer_gles2

Add the possibility to pass Chromium command-line arguments to
the command_buffer_gles2 libary. The arguments can be passed
through CHROME_COMMAND_BUFFER_ARGS environment variable.

Initialize the driver bug workarounds. This makes the libarary
to behave more similarly to actual Chrome.

Makes it possible to test GL path rendering with Skia through
command buffer, since the --enable-gl-path-rendering command-line
flag can be set.

BUG=skia:2992
Committed: https://crrev.com/ec6196f4e367774ced1e6034562dc19663fece45
Cr-Commit-Position: refs/heads/master@{#366349}

[Kimmo Kinnunen, 2015-12-08 07:44:39]

kkinnunen@nvidia.com changed reviewers:
+ hendrikw@chromium.org, sievers@chromium.org

[Kimmo Kinnunen, 2015-12-17 13:48:52]

kkinnunen@nvidia.com changed reviewers:
+ piman@chromium.org

[Kimmo Kinnunen, 2015-12-17 13:48:52]

piman@: would it be possible for you to look at the patch, or suggest somebody
willing to look at egl / command_buffer_gles2_support -related code? My current
added reviewers do not respond.

[hendrikw, 2015-12-17 18:39:07]

On 2015/12/17 13:48:52, Kimmo Kinnunen wrote:
> piman@: would it be possible for you to look at the patch, or suggest somebody
> willing to look at egl / command_buffer_gles2_support -related code? My
current
> added reviewers do not respond.

piman's on vacation, sievers might be too, and I'm not on chrome anymore (and
I'm not an owner of this anyway).

[hendrikw, 2015-12-17 18:39:18]

https://codereview.chromium.org/1509833002/diff/1/gpu/gles2_conform_support/e...
File gpu/gles2_conform_support/egl/egl.cc (right):

https://codereview.chromium.org/1509833002/diff/1/gpu/gles2_conform_support/e...
gpu/gles2_conform_support/egl/egl.cc:121: if (args.size() == 0) {
Nit:
 if (args.empty())

https://codereview.chromium.org/1509833002/diff/1/gpu/gles2_conform_support/e...
gpu/gles2_conform_support/egl/egl.cc:125: for (size_t i = 0; i < args.size();
++i) {
Nit:
range-for instead >>  for (auto& arg : args) ...

[piman, 2015-12-18 06:42:16]

LGTM after Hendrik's nits are addressed.
(not in vacation just yet, just in a different timezone ^^)

[Kimmo Kinnunen, 2015-12-18 08:07:05]

https://codereview.chromium.org/1509833002/diff/1/gpu/gles2_conform_support/e...
File gpu/gles2_conform_support/egl/egl.cc (right):

https://codereview.chromium.org/1509833002/diff/1/gpu/gles2_conform_support/e...
gpu/gles2_conform_support/egl/egl.cc:121: if (args.size() == 0) {
On 2015/12/17 18:39:18, hendrikw wrote:
> Nit:
>  if (args.empty())

Done.

https://codereview.chromium.org/1509833002/diff/1/gpu/gles2_conform_support/e...
gpu/gles2_conform_support/egl/egl.cc:125: for (size_t i = 0; i < args.size();
++i) {
On 2015/12/17 18:39:18, hendrikw wrote:
> Nit:
> range-for instead >>  for (auto& arg : args) ...

The index is used in the lhs of the assignment.

[Kimmo Kinnunen, 2015-12-18 09:21:19]

>
https://codereview.chromium.org/1509833002/diff/1/gpu/gles2_conform_support/e...
> gpu/gles2_conform_support/egl/egl.cc:125: for (size_t i = 0; i < args.size();
> ++i) {
> On 2015/12/17 18:39:18, hendrikw wrote:
> > Nit:
> > range-for instead >>  for (auto& arg : args) ...
> 
> The index is used in the lhs of the assignment.

Taking the liberty and pushing the commit button even with one nit unaddressed.

[Kimmo Kinnunen, 2015-12-18 09:21:22]

The CQ bit was checked by kkinnunen@nvidia.com

[Kimmo Kinnunen, 2015-12-18 09:21:23]

The patchset sent to the CQ was uploaded after l-g-t-m from piman@chromium.org
Link to the patchset: https://codereview.chromium.org/1509833002/#ps20001
(title: "address review comment")

[commit-bot: I haz the power, 2015-12-18 09:22:09]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1509833002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1509833002/20001

[commit-bot: I haz the power, 2015-12-18 09:25:59]

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

[commit-bot: I haz the power, 2015-12-18 09:25:59]

Try jobs failed on following builders:
  android_chromium_gn_compile_rel on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_chro...)
  linux_chromium_gn_chromeos_rel on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[Kimmo Kinnunen, 2015-12-18 10:06:40]

piman@: could you still look at the gpu/BUILD.gn change?

[piman, 2015-12-18 14:05:10]

LGTM. Last chance before I'm gone :)

[Kimmo Kinnunen, 2015-12-21 07:39:26]

The CQ bit was checked by kkinnunen@nvidia.com

[commit-bot: I haz the power, 2015-12-21 07:39:35]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1509833002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1509833002/40001

[commit-bot: I haz the power, 2015-12-21 08:56:47]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-12-21 08:57:22]

Description was changed from

==========
Support arguments and driver bug workarounds in command_buffer_gles2

Add the possibility to pass Chromium command-line arguments to
the command_buffer_gles2 libary. The arguments can be passed
through CHROME_COMMAND_BUFFER_ARGS environment variable.

Initialize the driver bug workarounds. This makes the libarary
to behave more similarly to actual Chrome.

Makes it possible to test GL path rendering with Skia through
command buffer, since the --enable-gl-path-rendering command-line
flag can be set.

BUG=skia:2992
==========

to

==========
Support arguments and driver bug workarounds in command_buffer_gles2

Add the possibility to pass Chromium command-line arguments to
the command_buffer_gles2 libary. The arguments can be passed
through CHROME_COMMAND_BUFFER_ARGS environment variable.

Initialize the driver bug workarounds. This makes the libarary
to behave more similarly to actual Chrome.

Makes it possible to test GL path rendering with Skia through
command buffer, since the --enable-gl-path-rendering command-line
flag can be set.

BUG=skia:2992

Committed: https://crrev.com/ec6196f4e367774ced1e6034562dc19663fece45
Cr-Commit-Position: refs/heads/master@{#366349}
==========

[commit-bot: I haz the power, 2015-12-21 08:57:22]

Patchset 3 (id:??) landed as
https://crrev.com/ec6196f4e367774ced1e6034562dc19663fece45
Cr-Commit-Position: refs/heads/master@{#366349}