Wake up other streams/task runners after removing ended stream

Wake up other streams/task runners after removing ended stream

BUG=internal b/19560149

Committed: https://crrev.com/bb537f10a592f899d157529a26545240933be51c
Cr-Commit-Position: refs/heads/master@{#330400}

[servolk, 2015-05-13 20:19:31]

servolk@chromium.org changed reviewers:
+ damienv@chromium.org, gunsch@chromium.org, lcwu@chromium.org

[damienv1, 2015-05-13 22:23:12]

https://codereview.chromium.org/1127183008/diff/1/chromecast/media/cma/base/b...
File chromecast/media/cma/base/balanced_media_task_runner_factory.cc (right):

https://codereview.chromium.org/1127183008/diff/1/chromecast/media/cma/base/b...
chromecast/media/cma/base/balanced_media_task_runner_factory.cc:248:
task_runners_.erase(media_task_runner);
Note: the lock should not be applied on OnNewTask:

{
  base::AutoLock auto_lock(lock_);
  task_runners_.erase(media_task_runner);
}
OnNewTask();

[servolk, 2015-05-13 22:25:45]

https://codereview.chromium.org/1127183008/diff/1/chromecast/media/cma/base/b...
File chromecast/media/cma/base/balanced_media_task_runner_factory.cc (right):

https://codereview.chromium.org/1127183008/diff/1/chromecast/media/cma/base/b...
chromecast/media/cma/base/balanced_media_task_runner_factory.cc:248:
task_runners_.erase(media_task_runner);
On 2015/05/13 22:23:11, damienv1 wrote:
> Note: the lock should not be applied on OnNewTask:
> 
> {
>   base::AutoLock auto_lock(lock_);
>   task_runners_.erase(media_task_runner);
> }
> OnNewTask();

Done.

[gunsch, 2015-05-13 22:28:22]

lgtm

[gunsch, 2015-05-13 22:28:43]

Any chance you can add a unit test for this? (We already have a
BalancedMediaTaskRunnerTest file)

[servolk, 2015-05-18 18:48:36]

On 2015/05/13 22:28:43, gunsch wrote:
> Any chance you can add a unit test for this? (We already have a
> BalancedMediaTaskRunnerTest file)

Adding a unit test for this is not entirely trivial, I need to become more
familiar with the current tests first. So for now I've created an internal bug
b/21269746 to track adding unit tests for this case, and for the case of two
DemuxerStreamAdapters where two streams are of different lengths.

[servolk, 2015-05-18 18:49:11]

The CQ bit was checked by servolk@chromium.org

[commit-bot: I haz the power, 2015-05-18 18:50:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1127183008/20001

[commit-bot: I haz the power, 2015-05-18 20:30:37]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-05-18 22:42:08]

Patchset 2 (id:??) landed as
https://crrev.com/bb537f10a592f899d157529a26545240933be51c
Cr-Commit-Position: refs/heads/master@{#330400}