Best effort approach on repeated synchronization lag

Best effort approach on repeated synchronization lag

A very unlikely scenario in which the peer is not synchronizing for a long
time (e.g. 300 ms) which results in dropping cached buffers and restarting.

Since we have no control of the endpoint synchronization, we use the best
effort approach by clearing the FIFO and trying to catch up with the peer.

BUG=408940
Committed: https://crrev.com/ce1012f07a698becd14d11b5949b47bf425c624f
Cr-Commit-Position: refs/heads/master@{#299292}

[burnik, 2014-10-13 10:15:35]

burnik@chromium.org changed reviewers:
+ xians@chromium.org

[burnik, 2014-10-13 10:17:05]

I believe this does make more sense. :) PTAL

[burnik, 2014-10-13 10:25:17]

Rebase on master

[burnik, 2014-10-13 10:29:02]

Nit - remove DCHECK in unit test

[burnik, 2014-10-13 10:29:19]

Patchset #2 (id:100001) has been deleted

[no longer working on chromium, 2014-10-13 12:28:56]

lgtm with two nits.

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
File content/renderer/media/speech_recognition_audio_sink.cc (right):

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
content/renderer/media/speech_recognition_audio_sink.cc:132: DVLOG(1) << "Audio
FIFO overflow";
DLOG(ERROR) << "Audio FIFO overflow";

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
File content/renderer/media/speech_recognition_audio_sink_unittest.cc (right):

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
content/renderer/media/speech_recognition_audio_sink_unittest.cc:189:
MockSyncSocket* foreign_socket_;
change this foreign_socket_ to send_socket_ correspondingly.

[no longer working on chromium, 2014-10-13 12:28:58]

lgtm with two nits.

[burnik, 2014-10-13 12:36:41]

Change DVLOG(1) to DLOG(ERROR)

[burnik, 2014-10-13 12:37:06]

Nits addressed.

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
File content/renderer/media/speech_recognition_audio_sink.cc (right):

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
content/renderer/media/speech_recognition_audio_sink.cc:132: DVLOG(1) << "Audio
FIFO overflow";
On 2014/10/13 12:28:56, xians1 wrote:
> DLOG(ERROR) << "Audio FIFO overflow";

Done.

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
File content/renderer/media/speech_recognition_audio_sink_unittest.cc (right):

https://codereview.chromium.org/651873002/diff/200001/content/renderer/media/...
content/renderer/media/speech_recognition_audio_sink_unittest.cc:189:
MockSyncSocket* foreign_socket_;
On 2014/10/13 12:28:56, xians1 wrote:
> change this foreign_socket_ to send_socket_ correspondingly.

Done.

[burnik, 2014-10-13 12:38:08]

The CQ bit was checked by burnik@chromium.org

[commit-bot: I haz the power, 2014-10-13 12:39:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/651873002/350001

[commit-bot: I haz the power, 2014-10-13 13:33:51]

Committed patchset #3 (id:350001)

[commit-bot: I haz the power, 2014-10-13 13:34:38]

Patchset 3 (id:??) landed as
https://crrev.com/ce1012f07a698becd14d11b5949b47bf425c624f
Cr-Commit-Position: refs/heads/master@{#299292}