Get audio_unittests passing with h/w on windows.

Get audio_unittests passing with h/w on windows.

Ensure that audio_unittests pass with --require-audio-hardware-for-testing.
 * Convert tests to STA COM threads, not MTA.
 * Do COM thread initialization in AudioManager.
 * Disable exclusive mode tests.
 * Convert a couple of tests to use ABORT_AUDIO_TEST_IF_NOT.

BUG=416594
TEST=audio_unittests --require-audio-hardware-for-testing

Committed: https://crrev.com/c99b195f6abb12e617e8e14ef95b6403fc504f9f
Cr-Commit-Position: refs/heads/master@{#329917}

[watk, 2015-05-06 18:22:29]

watk@chromium.org changed reviewers:
+ dalecurtis@chromium.org

[DaleCurtis, 2015-05-07 02:11:12]

lgtm

https://codereview.chromium.org/1130753005/diff/1/media/audio/audio_manager.cc
File media/audio/audio_manager.cc (right):

https://codereview.chromium.org/1130753005/diff/1/media/audio/audio_manager.c...
media/audio/audio_manager.cc:131: base::win::ScopedCOMInitializer::kMTA));
Did using a regular initializer (i.e. w/ MTA) fail?

[watk, 2015-05-07 18:13:36]

On 2015/05/07 02:11:12, DaleCurtis wrote:
> lgtm
> 
> https://codereview.chromium.org/1130753005/diff/1/media/audio/audio_manager.cc
> File media/audio/audio_manager.cc (right):
> 
>
https://codereview.chromium.org/1130753005/diff/1/media/audio/audio_manager.c...
> media/audio/audio_manager.cc:131: base::win::ScopedCOMInitializer::kMTA));
> Did using a regular initializer (i.e. w/ MTA) fail?

Thanks! Do you mean without MTA? I saw that we're using MTA for the AudioThread
so I figured we should be consistent. Perhaps I'm misunderstanding though.
(https://chromium.googlesource.com/chromium/src/+/2fa37a05b45e8d50d39295464271...)

[DaleCurtis, 2015-05-07 18:31:08]

On 2015/05/07 18:13:36, watk wrote:
> On 2015/05/07 02:11:12, DaleCurtis wrote:
> > lgtm
> > 
> >
https://codereview.chromium.org/1130753005/diff/1/media/audio/audio_manager.cc
> > File media/audio/audio_manager.cc (right):
> > 
> >
>
https://codereview.chromium.org/1130753005/diff/1/media/audio/audio_manager.c...
> > media/audio/audio_manager.cc:131: base::win::ScopedCOMInitializer::kMTA));
> > Did using a regular initializer (i.e. w/ MTA) fail?
> 
> Thanks! Do you mean without MTA? I saw that we're using MTA for the
AudioThread
> so I figured we should be consistent. Perhaps I'm misunderstanding though.
>
(https://chromium.googlesource.com/chromium/src/+/2fa37a05b45e8d50d39295464271...)

No I did mean STA, since this call to the audiomanager in production happens on
the browser thread which is STA. It'd be nice to have them similar to catch
issues early. However it may not be possible since many tests use that same
thread for executing MTA based COM calls.

[watk, 2015-05-13 23:16:46]

Turns out nothing breaks if we use STA, so converted to it! Thanks

[watk, 2015-05-14 20:12:29]

The CQ bit was checked by watk@chromium.org

[watk, 2015-05-14 20:12:29]

The patchset sent to the CQ was uploaded after l-g-t-m from
dalecurtis@chromium.org
Link to the patchset: https://codereview.chromium.org/1130753005/#ps20001
(title: "MTA -> STA")

[commit-bot: I haz the power, 2015-05-14 20:13:47]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1130753005/20001

[commit-bot: I haz the power, 2015-05-14 20:24:33]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-05-14 20:26:20]

Patchset 2 (id:??) landed as
https://crrev.com/c99b195f6abb12e617e8e14ef95b6403fc504f9f
Cr-Commit-Position: refs/heads/master@{#329917}