Reland Mojo-based waiting for IPC::SyncChannel

Reland Mojo-based waiting for IPC::SyncChannel

BUG=612500
TBR=jam@chromium.org

Committed: https://crrev.com/3c23629163a036c17017c4c2d16a0ac77f62d469
Cr-Commit-Position: refs/heads/master@{#404233}

[Ken Rockot(use gerrit already), 2016-06-28 01:31:48]

rockot@chromium.org changed reviewers:
+ amistry@chromium.org, sammc@chromium.org

[Ken Rockot(use gerrit already), 2016-06-28 01:31:48]

FYI, here's the patch that's been landed a few times. It leads to renderer hangs
as well as browser hangs (particularly when waiting for sync replies from the
gpu process). All the hangs I've seen so far are calls to
mojo::SyncHandleRegistry::WatchAllHandles coming from IPC::SyncChannel. Haven't
seen any from SyncMessageFilter yet.

[Ken Rockot(use gerrit already), 2016-06-28 01:42:12]

Also some relevant bugs with dumps, though not terribly useful imho

https://bugs.chromium.org/p/chromium/issues/detail?id=622982

https://bugs.chromium.org/p/chromium/issues/detail?id=622934

[Ken Rockot(use gerrit already), 2016-07-07 18:20:13]

Description was changed from

==========
Reland Mojo-based waiting for IPC::SyncChannel

BUG=612500
==========

to

==========
Reland Mojo-based waiting for IPC::SyncChannel

BUG=612500
TBR=jam@chromium.org
==========

[Ken Rockot(use gerrit already), 2016-07-07 18:22:11]

Anand suggested this may have only been highlighting existing hangs. After
looking into it more I do see browser UI thread hangs on sync GPU IPC in about
equal proportion with and without this change (see go/crash/34358b6600000000 for
example).

Given that, I've rebased and am going to reland once again.

[Ken Rockot(use gerrit already), 2016-07-07 18:22:34]

The CQ bit was checked by rockot@chromium.org

[commit-bot: I haz the power, 2016-07-07 18:23:49]

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, 2016-07-07 18:44:19]

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

[commit-bot: I haz the power, 2016-07-07 18:44:20]

Try jobs failed on following builders:
  chromeos_daisy_chromium_compile_only_ng on master.tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_daisy_...)

[Ken Rockot(use gerrit already), 2016-07-07 19:59:00]

The CQ bit was checked by rockot@chromium.org

[commit-bot: I haz the power, 2016-07-07 19:59:36]

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, 2016-07-07 20:26:16]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-07-07 20:26:18]

CQ bit was unchecked.

[commit-bot: I haz the power, 2016-07-07 20:27:38]

Description was changed from

==========
Reland Mojo-based waiting for IPC::SyncChannel

BUG=612500
TBR=jam@chromium.org
==========

to

==========
Reland Mojo-based waiting for IPC::SyncChannel

BUG=612500
TBR=jam@chromium.org

Committed: https://crrev.com/3c23629163a036c17017c4c2d16a0ac77f62d469
Cr-Commit-Position: refs/heads/master@{#404233}
==========

[commit-bot: I haz the power, 2016-07-07 20:27:39]

Patchset 2 (id:??) landed as
https://crrev.com/3c23629163a036c17017c4c2d16a0ac77f62d469
Cr-Commit-Position: refs/heads/master@{#404233}

[Ken Rockot(use gerrit already), 2016-07-08 05:32:46]

For some additional context about the GPU hangs: http://crbug.com/620259