Instrument ContextCacheController::OnIdle for postmortem reports

Instrument ContextCacheController::OnIdle for postmortem reports

Postmortem reports (reports sent for unclean chrome shutdowns) shows
a large fraction of reports in which the main thread is running
ContextCacheController::OnIdle. This CL adds additional adhoc
instrumentation to zoom in on the problem.

BUG=703342
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

[manzagop (departed), 2017-03-20 21:40:56]

Description was changed from

==========
Instrument ContextCacheController::OnIdle for postmortem reports

Postmortem reports (reports sent for unclean chrome shutdowns) shows
a large fraction of reports in which the main thread is running
ContextCacheController::OnIdle. This CL adds additional adhoc
instrumentation to zoom in on the problem.

BUG=703342
==========

to

==========
Instrument ContextCacheController::OnIdle for postmortem reports

Postmortem reports (reports sent for unclean chrome shutdowns) shows
a large fraction of reports in which the main thread is running
ContextCacheController::OnIdle. This CL adds additional adhoc
instrumentation to zoom in on the problem.

BUG=703342
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[manzagop (departed), 2017-03-20 21:40:57]

Description was changed from

==========
Instrument ContextCacheController::OnIdle for postmortem reports

Postmortem reports (reports sent for unclean chrome shutdowns) shows
a large fraction of reports in which the main thread is running
ContextCacheController::OnIdle. This CL adds additional adhoc
instrumentation to zoom in on the problem.

BUG=703342
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Instrument ContextCacheController::OnIdle for postmortem reports

Postmortem reports (reports sent for unclean chrome shutdowns) shows
a large fraction of reports in which the main thread is running
ContextCacheController::OnIdle. This CL adds additional adhoc
instrumentation to zoom in on the problem.

BUG=703342
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[manzagop (departed), 2017-03-20 21:49:04]

manzagop@chromium.org changed reviewers:
+ bcwhite@chromium.org

[manzagop (departed), 2017-03-20 21:49:04]

Please have a look!

[bcwhite, 2017-03-21 12:34:38]

https://codereview.chromium.org/2761093002/diff/1/cc/output/context_cache_con...
File cc/output/context_cache_controller.cc (right):

https://codereview.chromium.org/2761093002/diff/1/cc/output/context_cache_con...
cc/output/context_cache_controller.cc:164: base::debug::ScopedActivity
activity_free(0, 0, 0);
There is a default ctor that is (0,0,0) if you don't want to specify them...
though it's recommended that you do if you want to be able to identify them
during analysis.

https://codereview.chromium.org/2761093002/diff/1/gpu/command_buffer/client/g...
File gpu/command_buffer/client/gles2_implementation.cc (right):

https://codereview.chromium.org/2761093002/diff/1/gpu/command_buffer/client/g...
gpu/command_buffer/client/gles2_implementation.cc:424:
base::debug::ScopedActivity activity_flush(0, 0,
Do you want an ID for the second parameter to distinguish all these during
analysis?  See example in activity_tracker.h, around line 1200.

[manzagop (departed), 2017-03-21 14:00:23]

Thanks! Another look?

https://codereview.chromium.org/2761093002/diff/1/cc/output/context_cache_con...
File cc/output/context_cache_controller.cc (right):

https://codereview.chromium.org/2761093002/diff/1/cc/output/context_cache_con...
cc/output/context_cache_controller.cc:164: base::debug::ScopedActivity
activity_free(0, 0, 0);
On 2017/03/21 12:34:38, bcwhite wrote:
> There is a default ctor that is (0,0,0) if you don't want to specify them...
> though it's recommended that you do if you want to be able to identify them
> during analysis.

Done. (Another option is to rely on the activity's address.)

https://codereview.chromium.org/2761093002/diff/1/gpu/command_buffer/client/g...
File gpu/command_buffer/client/gles2_implementation.cc (right):

https://codereview.chromium.org/2761093002/diff/1/gpu/command_buffer/client/g...
gpu/command_buffer/client/gles2_implementation.cc:424:
base::debug::ScopedActivity activity_flush(0, 0,
On 2017/03/21 12:34:38, bcwhite wrote:
> Do you want an ID for the second parameter to distinguish all these during
> analysis?  See example in activity_tracker.h, around line 1200.

Done. Here and below.

[manzagop (departed), 2017-03-21 14:00:29]

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

[commit-bot: I haz the power, 2017-03-21 14:00:51]

Dry run: 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, 2017-03-21 15:08:35]

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

[commit-bot: I haz the power, 2017-03-21 15:08:36]

Dry run: This issue passed the CQ dry run.

[manzagop (departed), 2017-03-22 16:44:38]

Friendly ping.

[manzagop (departed), 2017-03-22 17:42:23]

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

[commit-bot: I haz the power, 2017-03-22 17:42:46]

Dry run: 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, 2017-03-22 18:53:43]

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

[commit-bot: I haz the power, 2017-03-22 18:53:44]

Dry run: This issue passed the CQ dry run.

[bcwhite, 2017-03-23 21:06:52]

lgtm