Fix mac raciness with initial needs begin frame state

Fix mac raciness with initial needs begin frame state

Because Mac doesn't delay begin frame messages (because it needs to
receive them during resize), this creates some startup raciness.
Fix this by just having the renderer host side assume that the
renderer is going to want begin frames (even before hearing about it),
so that if the SetNeedsBeginFrames(true) message is dropped it
won't matter.

Renderers when they start up send a SetNeedsBeginFrames(true) and
then a SetNeedsBeginFrames(false) once they've finished loading and
at that point the renderer and the host will be back in agreement
about the state.

BUG=657878
Committed: https://crrev.com/2051d912c9c52c61045908314d288a3ecca6c6c2
Cr-Commit-Position: refs/heads/master@{#427844}

[enne (OOO), 2016-10-26 18:32:23]

Description was changed from

==========
Fix mac raciness with initial begin frames

Because Mac doesn't delay begin frame messages (because it needs to
receive them during resize), this creates some startup raciness.
Fix this by just having the renderer host side assume that the
renderer is going to want begin frames (even before hearing about it),
so that if the SetNeedsBeginFrames(true) message is dropped it
won't matter.

BUG=657878
==========

to

==========
Fix mac raciness with initial needs begin frame state

Because Mac doesn't delay begin frame messages (because it needs to
receive them during resize), this creates some startup raciness.
Fix this by just having the renderer host side assume that the
renderer is going to want begin frames (even before hearing about it),
so that if the SetNeedsBeginFrames(true) message is dropped it
won't matter.

Renderers when they start up send a SetNeedsBeginFrames(true) and
then a SetNeedsBeginFrames(false) once they've finished loading and
at that point the renderer and the host will be back in agreement
about the state.

BUG=657878
==========

[enne (OOO), 2016-10-26 18:32:56]

enne@chromium.org changed reviewers:
+ ccameron@chromium.org

[enne (OOO), 2016-10-26 18:32:56]

PTAL

I think I can't remove the code in RenderMessageFilter, as it's still needed for
receiving the message during resize.

[enne (OOO), 2016-10-26 18:33:30]

The CQ bit was checked by enne@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-26 18:34:13]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[ccameron, 2016-10-26 18:42:39]

On 2016/10/26 18:32:56, enne wrote:
> PTAL
> 
> I think I can't remove the code in RenderMessageFilter, as it's still needed
for
> receiving the message during resize.

lgtm

Sorry, I think this going to keep biting us as we mojo-ify things. I'll see if
we can at least make it so that we only pump during resize, not at show and
hide.

[enne (OOO), 2016-10-26 18:52:08]

The CQ bit was unchecked by enne@chromium.org

[enne (OOO), 2016-10-26 18:52:14]

The CQ bit was checked by enne@chromium.org

[commit-bot: I haz the power, 2016-10-26 18:52:54]

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, 2016-10-26 20:05:30]

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

[commit-bot: I haz the power, 2016-10-26 20:05:31]

Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[enne (OOO), 2016-10-26 20:55:32]

On 2016/10/26 at 20:05:31, commit-bot wrote:
> Try jobs failed on following builders:
>   mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

Ah, this test was counting total messages and so an additional message made it
fail.  I added a flush and clear step to make this work properly before and
after my change.

[enne (OOO), 2016-10-26 20:55:35]

The CQ bit was checked by enne@chromium.org

[enne (OOO), 2016-10-26 20:55:35]

The patchset sent to the CQ was uploaded after l-g-t-m from
ccameron@chromium.org
Link to the patchset: https://codereview.chromium.org/2451953004/#ps20001
(title: "Fixes for ScrollWheelEndEventDelivery test")

[commit-bot: I haz the power, 2016-10-26 20:56:19]

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, 2016-10-26 22:11:18]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-10-26 22:13:50]

Description was changed from

==========
Fix mac raciness with initial needs begin frame state

Because Mac doesn't delay begin frame messages (because it needs to
receive them during resize), this creates some startup raciness.
Fix this by just having the renderer host side assume that the
renderer is going to want begin frames (even before hearing about it),
so that if the SetNeedsBeginFrames(true) message is dropped it
won't matter.

Renderers when they start up send a SetNeedsBeginFrames(true) and
then a SetNeedsBeginFrames(false) once they've finished loading and
at that point the renderer and the host will be back in agreement
about the state.

BUG=657878
==========

to

==========
Fix mac raciness with initial needs begin frame state

Because Mac doesn't delay begin frame messages (because it needs to
receive them during resize), this creates some startup raciness.
Fix this by just having the renderer host side assume that the
renderer is going to want begin frames (even before hearing about it),
so that if the SetNeedsBeginFrames(true) message is dropped it
won't matter.

Renderers when they start up send a SetNeedsBeginFrames(true) and
then a SetNeedsBeginFrames(false) once they've finished loading and
at that point the renderer and the host will be back in agreement
about the state.

BUG=657878

Committed: https://crrev.com/2051d912c9c52c61045908314d288a3ecca6c6c2
Cr-Commit-Position: refs/heads/master@{#427844}
==========

[commit-bot: I haz the power, 2016-10-26 22:13:51]

Patchset 2 (id:??) landed as
https://crrev.com/2051d912c9c52c61045908314d288a3ecca6c6c2
Cr-Commit-Position: refs/heads/master@{#427844}