Switch Speech Recognition to asynchronous callback-based AudioManager interactions.

Switch Speech Recognition to asynchronous callback-based AudioManager interactions.

1) This introduces AudioSystem, which provides an asynchronous interface to AudioManager and eventually will replace AudioManager.
The plan is to gradually populate it with all the required methods and switch all AudioManager clients to it,
in preparation for moving AudioManager functionality to Mojo audio service.

2) SpeechRecognizerImpl now uses this interface, which required adding an extra state to its FSM.

3) SpeechRecognizerImpl unit tests are updated to take into account the new threading model and the new state.

BUG=687981, 672468
CQ_INCLUDE_TRYBOTS=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

Review-Url: https://codereview.chromium.org/2675713002
Cr-Commit-Position: refs/heads/master@{#448273}
Committed: https://chromium.googlesource.com/chromium/src/+/ef762c904b723cb488bad237300a5976c29327d3

[o1ka, 2017-02-02 14:53:37]

Description was changed from

==========
Switch Speech Recognition to asyncrhonous callback-based AudioManager
interactions.

1) AudioSystem interface is introduced, which provides an asynchronous interface
to AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager to Mojo audio service.

2) SpeechRecognizerImpl is switched to this interface, which required adding an
extra state ot its FSM.

3) SpeechRecognizerImpl unit tests aer updated to take into account the new
threading model and the new state.

BUG=687981,672468
==========

to

==========
Switch Speech Recognition to asyncrhonous callback-based AudioManager
interactions.

1) AudioSystem interface is introduced, which provides an asynchronous interface
to AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager to Mojo audio service.

2) SpeechRecognizerImpl is switched to this interface, which required adding an
extra state ot its FSM.

3) SpeechRecognizerImpl unit tests aer updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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
==========

[o1ka, 2017-02-02 14:54:30]

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

[commit-bot: I haz the power, 2017-02-02 14:54:37]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[o1ka, 2017-02-02 15:00:31]

Description was changed from

==========
Switch Speech Recognition to asyncrhonous callback-based AudioManager
interactions.

1) AudioSystem interface is introduced, which provides an asynchronous interface
to AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager to Mojo audio service.

2) SpeechRecognizerImpl is switched to this interface, which required adding an
extra state ot its FSM.

3) SpeechRecognizerImpl unit tests aer updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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

==========
Switch Speech Recognition to asyncrhonous callback-based AudioManager
interactions.

1) AudioSystem interface is introduced, which provides an asynchronous interface
to AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager to Mojo audio service.

2) SpeechRecognizerImpl is switched to this interface, which required adding an
extra state ot its FSM.

3) SpeechRecognizerImpl unit tests aer updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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
==========

[o1ka, 2017-02-02 15:34:11]

olka@chromium.org changed reviewers:
+ tommi@chromium.org

[o1ka, 2017-02-02 15:34:12]

Tommi - PTAL.

[commit-bot: I haz the power, 2017-02-02 16:01:29]

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

[commit-bot: I haz the power, 2017-02-02 16:01:30]

Dry run: Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[tommi (sloooow) - chröme, 2017-02-02 16:14:18]

Description was changed from

==========
Switch Speech Recognition to asyncrhonous callback-based AudioManager
interactions.

1) AudioSystem interface is introduced, which provides an asynchronous interface
to AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager to Mojo audio service.

2) SpeechRecognizerImpl is switched to this interface, which required adding an
extra state ot its FSM.

3) SpeechRecognizerImpl unit tests aer updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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

==========
Switch Speech Recognition to asynchronous callback-based AudioManager
interactions.

1) AudioSystem interface is introduced, which provides an asynchronous interface
to AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager to Mojo audio service.

2) SpeechRecognizerImpl is switched to this interface, which required adding an
extra state to its FSM.

3) SpeechRecognizerImpl unit tests are updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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
==========

[tommi (sloooow) - chröme, 2017-02-02 16:27:18]

https://codereview.chromium.org/2675713002/diff/1/content/browser/speech/spee...
File content/browser/speech/speech_recognizer_impl.cc (right):

https://codereview.chromium.org/2675713002/diff/1/content/browser/speech/spee...
content/browser/speech/speech_recognizer_impl.cc:562:
GetAudioSystem()->GetInputStreamParameters(
Instead of adding AudioSystem etc, could you post to the audio thread here, call
GetInputParameters() and post pack to the IO thread when you're done to complete
the state transition?  Maybe I'm missing something but it feels like we're
adding a lot of complexity that we can avoid adding.

https://codereview.chromium.org/2675713002/diff/1/media/audio/audio_system.h
File media/audio/audio_system.h (right):

https://codereview.chromium.org/2675713002/diff/1/media/audio/audio_system.h#...
media/audio/audio_system.h:19: class MEDIA_EXPORT AudioSystem {
I feel that adding a new class at the same level as AudioManager, will cause
confusion.  Can we instead add GetInputStreamParametersAsync to AudioManager? 
Alternatively, we could rename the old one to 'Deprecated' and use
GetInputStreamParameters where we use the async variant.

[o1ka, 2017-02-02 16:27:51]

On 2017/02/02 15:34:12, o1ka (CET) wrote:
> Tommi - PTAL.

Ok, there is an issue with AudioManager destruction in unit tests on Mac; I'll
take care of it.

[o1ka, 2017-02-02 16:32:22]

https://codereview.chromium.org/2675713002/diff/1/media/audio/audio_system.h
File media/audio/audio_system.h (right):

https://codereview.chromium.org/2675713002/diff/1/media/audio/audio_system.h#...
media/audio/audio_system.h:19: class MEDIA_EXPORT AudioSystem {
On 2017/02/02 16:27:18, tommi (krómíum) wrote:
> I feel that adding a new class at the same level as AudioManager, will cause
> confusion.  Can we instead add GetInputStreamParametersAsync to AudioManager? 
> Alternatively, we could rename the old one to 'Deprecated' and use
> GetInputStreamParameters where we use the async variant.

This is a staged switch to Mojo audio service (see CL description). AudioManager
interface won't be public soon.

[o1ka, 2017-02-02 17:07:02]

On 2017/02/02 16:27:51, o1ka (CET) wrote:
> On 2017/02/02 15:34:12, o1ka (CET) wrote:
> > Tommi - PTAL.
> 
> Ok, there is an issue with AudioManager destruction in unit tests on Mac; I'll
> take care of it.

Oh well, I should not have used AudioManagerDeleter anywhere in  tests, since it
has mac specific stuff and also casts AudioManager to MacAudioManager.
Will fix it tomorrow.

[o1ka, 2017-02-03 10:44:05]

Patchset #3 (id:40001) has been deleted

[o1ka, 2017-02-03 12:09:22]

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

[commit-bot: I haz the power, 2017-02-03 12:09:26]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[o1ka, 2017-02-03 12:10:11]

On 2017/02/02 17:07:02, o1ka (CET) wrote:
> On 2017/02/02 16:27:51, o1ka (CET) wrote:
> > On 2017/02/02 15:34:12, o1ka (CET) wrote:
> > > Tommi - PTAL.
> > 
> > Ok, there is an issue with AudioManager destruction in unit tests on Mac;
I'll
> > take care of it.
> 
> Oh well, I should not have used AudioManagerDeleter anywhere in  tests, since
it
> has mac specific stuff and also casts AudioManager to MacAudioManager.
> Will fix it tomorrow.

Now it should work - Tommi PTAL, thanks!

[commit-bot: I haz the power, 2017-02-03 14:25:46]

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

[commit-bot: I haz the power, 2017-02-03 14:25:47]

Dry run: This issue passed the CQ dry run.

[tommi (sloooow) - chröme, 2017-02-05 19:38:29]

Description was changed from

==========
Switch Speech Recognition to asynchronous callback-based AudioManager
interactions.

1) AudioSystem interface is introduced, which provides an asynchronous interface
to AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager to Mojo audio service.

2) SpeechRecognizerImpl is switched to this interface, which required adding an
extra state to its FSM.

3) SpeechRecognizerImpl unit tests are updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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

==========
Switch Speech Recognition to asynchronous callback-based AudioManager
interactions.

1) This introduces AudioSystem, which provides an asynchronous interface to
AudioManager and eventually will replace AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager functionality to Mojo audio service.

2) SpeechRecognizerImpl now uses this interface, which required adding an extra
state to its FSM.

3) SpeechRecognizerImpl unit tests are updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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
==========

[tommi (sloooow) - chröme, 2017-02-05 20:14:55]

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
File content/browser/speech/speech_recognizer_impl.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
content/browser/speech/speech_recognizer_impl.cc:117: media::AudioSystem*
SpeechRecognizerImpl::audio_system_for_tests_ = nullptr;
Just checking - Is it still necessary for this to be static?

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
File content/browser/speech/speech_recognizer_impl_unittest.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
content/browser/speech/speech_recognizer_impl_unittest.cc:46: :
audio_thread_("AudioThread"),
nit: SpeechAudioThread?  Nice improvement to really have a separate thread btw.

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
content/browser/speech/speech_recognizer_impl_unittest.cc:246: //
recognizer_->StopAudioCapture();
remove?

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_system.h
File media/audio/audio_system.h (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system.h:31: virtual void GetInputStreamParameters(const
std::string& device_id,
Since we're making a new interface, would it make sense for this to be const?

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
File media/audio/audio_system_impl.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.cc:23: // TODO(olka) : remove this when
AudioManager::GetInputStreamParameters()
nit: TODO(foo):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.cc:50: void
AudioSystemImpl::GetInputStreamParameters(
is it the intent that this method can be called on any thread?

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
File media/audio/audio_system_impl.h (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.h:17: ~AudioSystemImpl();
override?

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.h:30: AudioManager* audio_manager_;
Can this be:
AudioManager* const audio_manager_;

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
File media/audio/audio_system_impl_unittest.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl_unittest.cc:24: : use_audio_thread_(GetParam()),
audio_thread_("AudioThread") {
nit: AudioSystemThread? (just for disambiguation)

[tommi (sloooow) - chröme, 2017-02-06 08:36:42]

given that all of my comments are minor (and assuming the commented out line,
shouldn't be there), lgtm.  I'm going to be unavailable for at least half of the
day so if there are more changes to be reviewed, can you ask Max to take a look?

[o1ka, 2017-02-06 11:54:44]

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

[commit-bot: I haz the power, 2017-02-06 11:54:59]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[o1ka, 2017-02-06 12:06:08]

Thanks! The comments addressed.

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
File content/browser/speech/speech_recognizer_impl.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
content/browser/speech/speech_recognizer_impl.cc:117: media::AudioSystem*
SpeechRecognizerImpl::audio_system_for_tests_ = nullptr;
On 2017/02/05 20:14:54, tommi (krómíum) wrote:
> Just checking - Is it still necessary for this to be static?

Yes, it's used in speech_recognition_browsertest.cc to substitute the real audio
system. I added a comment for that.

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
File content/browser/speech/speech_recognizer_impl_unittest.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
content/browser/speech/speech_recognizer_impl_unittest.cc:46: :
audio_thread_("AudioThread"),
On 2017/02/05 20:14:54, tommi (krómíum) wrote:
> nit: SpeechAudioThread?  Nice improvement to really have a separate thread
btw.

Done.

https://codereview.chromium.org/2675713002/diff/80001/content/browser/speech/...
content/browser/speech/speech_recognizer_impl_unittest.cc:246: //
recognizer_->StopAudioCapture();
On 2017/02/05 20:14:54, tommi (krómíum) wrote:
> remove?

Done.

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_system.h
File media/audio/audio_system.h (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system.h:31: virtual void GetInputStreamParameters(const
std::string& device_id,
On 2017/02/05 20:14:54, tommi (krómíum) wrote:
> Since we're making a new interface, would it make sense for this to be const?

Right! Done.

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
File media/audio/audio_system_impl.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.cc:23: // TODO(olka) : remove this when
AudioManager::GetInputStreamParameters()
On 2017/02/05 20:14:54, tommi (krómíum) wrote:
> nit: TODO(foo):

Done.

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.cc:50: void
AudioSystemImpl::GetInputStreamParameters(
On 2017/02/05 20:14:54, tommi (krómíum) wrote:
> is it the intent that this method can be called on any thread?

Yes, it can be called from UI and IO threads, and UI==audio thread on Mac.
I removed GetTaskRunner() from the interface for now. When/if somebody needs to
post AudioSystem calls to the audio thread, we'll need to re-conciser
AudioSystem lifetime, since now AudioSystem is deleted on IO thread and the
audio thread (and its tasks) outlives AudioSystem (this concern is not
applicable in Mac case).

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
File media/audio/audio_system_impl.h (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.h:17: ~AudioSystemImpl();
On 2017/02/05 20:14:55, tommi (krómíum) wrote:
> override?

Oops. Done.

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl.h:30: AudioManager* audio_manager_;
On 2017/02/05 20:14:55, tommi (krómíum) wrote:
> Can this be:
> AudioManager* const audio_manager_;

Done.

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
File media/audio/audio_system_impl_unittest.cc (right):

https://codereview.chromium.org/2675713002/diff/80001/media/audio/audio_syste...
media/audio/audio_system_impl_unittest.cc:24: : use_audio_thread_(GetParam()),
audio_thread_("AudioThread") {
On 2017/02/05 20:14:55, tommi (krómíum) wrote:
> nit: AudioSystemThread? (just for disambiguation)

Done.

[commit-bot: I haz the power, 2017-02-06 13:26:08]

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

[commit-bot: I haz the power, 2017-02-06 13:26:09]

Dry run: This issue passed the CQ dry run.

[o1ka, 2017-02-06 14:01:35]

avi@chromium.org: Could you RS changes in content/browser/browser_main_loop.* ?
Thanks!

[o1ka, 2017-02-06 14:02:34]

olka@chromium.org changed reviewers:
+ avi@chromium.org

[o1ka, 2017-02-06 14:03:39]

mailto:avi@chromium.org: Could you RS changes in
content/browser/browser_main_loop.* ?
Thanks!

(and now actually adding avi@ as a reviewer)

[Avi (use Gerrit), 2017-02-06 16:14:21]

lgtm

https://codereview.chromium.org/2675713002/diff/100001/content/browser/browse...
File content/browser/browser_main_loop.cc (right):

https://codereview.chromium.org/2675713002/diff/100001/content/browser/browse...
content/browser/browser_main_loop.cc:1541: speech_recognition_manager_.reset(new
SpeechRecognitionManagerImpl(
base::MakeUnique?

[o1ka, 2017-02-06 16:40:06]

Thanks!

https://codereview.chromium.org/2675713002/diff/100001/content/browser/browse...
File content/browser/browser_main_loop.cc (right):

https://codereview.chromium.org/2675713002/diff/100001/content/browser/browse...
content/browser/browser_main_loop.cc:1541: speech_recognition_manager_.reset(new
SpeechRecognitionManagerImpl(
On 2017/02/06 16:14:21, Avi wrote:
> base::MakeUnique?

Unfortunately this won't compile:
for some reason SpeechRecognitionImpl has a protected constructor;
BrowserMainLoop is its friend.
- Changing this behavior is out of scope of the CL.

[o1ka, 2017-02-06 16:40:44]

The CQ bit was checked by olka@chromium.org

[o1ka, 2017-02-06 16:40:45]

The patchset sent to the CQ was uploaded after l-g-t-m from tommi@chromium.org
Link to the patchset: https://codereview.chromium.org/2675713002/#ps100001
(title: "review comments addressed")

[commit-bot: I haz the power, 2017-02-06 16:41:04]

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, 2017-02-06 16:45:20]

CQ is committing da patch.

Bot data: {"patchset_id": 100001, "attempt_start_ts": 1486399244587980,
"parent_rev": "63f6b6af9eb41af69c24db2c03a1be7e176a7a5f", "commit_rev":
"ef762c904b723cb488bad237300a5976c29327d3"}

[commit-bot: I haz the power, 2017-02-06 16:45:56]

Description was changed from

==========
Switch Speech Recognition to asynchronous callback-based AudioManager
interactions.

1) This introduces AudioSystem, which provides an asynchronous interface to
AudioManager and eventually will replace AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager functionality to Mojo audio service.

2) SpeechRecognizerImpl now uses this interface, which required adding an extra
state to its FSM.

3) SpeechRecognizerImpl unit tests are updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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

==========
Switch Speech Recognition to asynchronous callback-based AudioManager
interactions.

1) This introduces AudioSystem, which provides an asynchronous interface to
AudioManager and eventually will replace AudioManager.
The plan is to gradually populate it with all the required methods and switch
all AudioManager clients to it,
in preparation for moving AudioManager functionality to Mojo audio service.

2) SpeechRecognizerImpl now uses this interface, which required adding an extra
state to its FSM.

3) SpeechRecognizerImpl unit tests are updated to take into account the new
threading model and the new state.

BUG=687981,672468
CQ_INCLUDE_TRYBOTS=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

Review-Url: https://codereview.chromium.org/2675713002
Cr-Commit-Position: refs/heads/master@{#448273}
Committed:
https://chromium.googlesource.com/chromium/src/+/ef762c904b723cb488bad237300a...
==========

[commit-bot: I haz the power, 2017-02-06 16:45:57]

Committed patchset #5 (id:100001) as
https://chromium.googlesource.com/chromium/src/+/ef762c904b723cb488bad237300a...