Adds Media.Audio.InputStartupSuccessMac UMA stat for Mac OS X

Adds Media.Audio.InputStartupSuccessMac UMA stat for Mac OS X

BUG=549021
Committed: https://crrev.com/ca6592b92bb146fdd737034d7addb281cab800b4
Cr-Commit-Position: refs/heads/master@{#357129}

[henrika (OOO until Aug 14), 2015-10-29 15:56:51]

Description was changed from

==========
Adds Media.Audio.InputStartupSuccessMac UMA stat for Mac OS X

BUG=
==========

to

==========
Adds Media.Audio.InputStartupSuccessMac UMA stat for Mac OS X

BUG=549021
==========

[henrika (OOO until Aug 14), 2015-10-30 09:43:25]

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

[henrika (OOO until Aug 14), 2015-10-30 09:43:53]

Tommi, PTAL.

I will add the XML part when you are OK with the rest.

[henrika (OOO until Aug 14), 2015-10-30 11:59:08]

henrika@chromium.org changed reviewers:
+ asvitkine@chromium.org

[henrika (OOO until Aug 14), 2015-10-30 11:59:52]

Adding Alexei Svitkine for XML parts.

[Alexei Svitkine (slow), 2015-10-30 15:06:40]

histograms.xml lgtm

[tommi (sloooow) - chröme, 2015-10-30 15:08:44]

I think the 'main thread' vs 'audio thread' concepts may be throwing me off.

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
File media/audio/mac/audio_low_latency_input_mac.cc (right):

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.cc:272:
base::TimeDelta::FromSeconds(kInputCallbackStartTimeoutInSeconds), this,
I think this uses base::Unretained() and due to that there may be a race since
the timer is created on the audio thread, destroyed there as well, but executes
on the browser thread (right?) and there it uses an unretained pointer to the
AUAudioInputStream object.  So if the timer is executing at the time the stream
is being deleted, we would have a use after free problem.

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.cc:499: if
(!audio_input->GetInputCallbackIsActive())
is this check necessary?

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.cc:762:
DCHECK(thread_checker_.CalledOnValidThread());
hmm... this suggests that we're actually not on the main thread but on the audio
thread, right?

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
File media/audio/mac/audio_low_latency_input_mac.h (right):

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.h:124: // Checks if a strem was
started successfully and the audio unit also starts
strem -> stream

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.h:131: base::ThreadChecker
thread_checker_;
nice

[tommi (sloooow) - chröme, 2015-10-30 15:50:03]

lgtm with nits addressed. for reference here's the audiomanagerbase check for
the main thread:
https://code.google.com/p/chromium/codesearch#chromium/src/media/audio/audio_...

[henrika (OOO until Aug 14), 2015-10-30 15:56:36]

Thanks. Landing.

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
File media/audio/mac/audio_low_latency_input_mac.cc (right):

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.cc:272:
base::TimeDelta::FromSeconds(kInputCallbackStartTimeoutInSeconds), this,
Discussed off line. Special case for Mac which does not create a unique
AudioThread but uses the main browser thread instead.

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.cc:499: if
(!audio_input->GetInputCallbackIsActive())
Not really; removed. Now only calls Set.

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.cc:762:
DCHECK(thread_checker_.CalledOnValidThread());
For the record, we are on the main thread. Special case for Mac.

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
File media/audio/mac/audio_low_latency_input_mac.h (right):

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.h:124: // Checks if a strem was
started successfully and the audio unit also starts
On 2015/10/30 15:08:44, tommi wrote:
> strem -> stream

Done.

https://codereview.chromium.org/1407333005/diff/40001/media/audio/mac/audio_l...
media/audio/mac/audio_low_latency_input_mac.h:131: base::ThreadChecker
thread_checker_;
Thanks

[henrika (OOO until Aug 14), 2015-10-30 15:57:59]

The CQ bit was checked by henrika@chromium.org

[henrika (OOO until Aug 14), 2015-10-30 15:57:59]

The patchset sent to the CQ was uploaded after l-g-t-m from tommi@chromium.org,
asvitkine@chromium.org
Link to the patchset: https://codereview.chromium.org/1407333005/#ps60001
(title: "Review comments from tommi@")

[commit-bot: I haz the power, 2015-10-30 15:58:36]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1407333005/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1407333005/60001

[commit-bot: I haz the power, 2015-10-30 17:08:00]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-10-30 17:08:43]

Patchset 4 (id:??) landed as
https://crrev.com/ca6592b92bb146fdd737034d7addb281cab800b4
Cr-Commit-Position: refs/heads/master@{#357129}