Remove uses of external begin frame sources

Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp doesn't create an output surface so no longer needs special
case code for an external begin frame source that it also doesn't use.

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Committed: https://crrev.com/077ba4863c28cbcb1049c24148aeae09f1dfa646
Cr-Commit-Position: refs/heads/master@{#418040}

[enne (OOO), 2016-08-31 21:59:26]

Description was changed from

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp now gets a StubBeginFrameSource which never does anything.
(Previously, output surfaces in Blimp would have just never provided a
begin frame source, which is functionally equivalent.)

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.
==========

to

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp now gets a StubBeginFrameSource which never does anything.
(Previously, output surfaces in Blimp would have just never provided a
begin frame source, which is functionally equivalent.)

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

[enne (OOO), 2016-08-31 22:00:25]

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

[commit-bot: I haz the power, 2016-08-31 22:02:13]

Dry run: 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, 2016-08-31 22:35:25]

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

[commit-bot: I haz the power, 2016-08-31 22:35:26]

Dry run: Try jobs failed on following builders:
  cast_shell_linux on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/cast_shell_linu...)

[enne (OOO), 2016-09-09 19:47:29]

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

[commit-bot: I haz the power, 2016-09-09 19:48:10]

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

[enne (OOO), 2016-09-09 20:30:49]

enne@chromium.org changed reviewers:
+ danakj@chromium.org, sievers@chromium.org

[commit-bot: I haz the power, 2016-09-09 21:09:01]

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

[commit-bot: I haz the power, 2016-09-09 21:09:02]

Dry run: This issue passed the CQ dry run.

[danakj, 2016-09-09 21:20:47]

LGTM

https://codereview.chromium.org/2299003002/diff/20001/content/renderer/render...
File content/renderer/render_thread_impl.cc (right):

https://codereview.chromium.org/2299003002/diff/20001/content/renderer/render...
content/renderer/render_thread_impl.cc:1597: // Blimp drives itself and doesn't
need a real begin frame source.
Does blimp server actually make an output surface?

[enne (OOO), 2016-09-09 22:06:09]

https://codereview.chromium.org/2299003002/diff/20001/content/renderer/render...
File content/renderer/render_thread_impl.cc (right):

https://codereview.chromium.org/2299003002/diff/20001/content/renderer/render...
content/renderer/render_thread_impl.cc:1597: // Blimp drives itself and doesn't
need a real begin frame source.
On 2016/09/09 at 21:20:47, danakj wrote:
> Does blimp server actually make an output surface?

Oh! I thought it did, but it looks like it doesn't.

[enne (OOO), 2016-09-09 22:22:57]

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

[commit-bot: I haz the power, 2016-09-09 22:23:14]

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

[enne (OOO), 2016-09-09 22:23:28]

Description was changed from

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp now gets a StubBeginFrameSource which never does anything.
(Previously, output surfaces in Blimp would have just never provided a
begin frame source, which is functionally equivalent.)

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

to

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp doesn't create an output surface so no longer needs special
case code for an external begin frame source that it also doesn't use.

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

[commit-bot: I haz the power, 2016-09-09 23:20:20]

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

[commit-bot: I haz the power, 2016-09-09 23:20:21]

Dry run: This issue passed the CQ dry run.

[enne (OOO), 2016-09-12 17:36:35]

Description was changed from

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp doesn't create an output surface so no longer needs special
case code for an external begin frame source that it also doesn't use.

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

to

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp doesn't create an output surface so no longer needs special
case code for an external begin frame source that it also doesn't use.

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

[no sievers, 2016-09-12 19:30:18]

content lgtm

[enne (OOO), 2016-09-12 19:52:41]

The CQ bit was checked by enne@chromium.org

[enne (OOO), 2016-09-12 19:52:42]

The patchset sent to the CQ was uploaded after l-g-t-m from danakj@chromium.org
Link to the patchset: https://codereview.chromium.org/2299003002/#ps60001
(title: "Remove Blimp conditional")

[commit-bot: I haz the power, 2016-09-12 19:53:18]

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, 2016-09-12 21:00:12]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-09-12 21:01:41]

Description was changed from

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp doesn't create an output surface so no longer needs special
case code for an external begin frame source that it also doesn't use.

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

to

==========
Remove uses of external begin frame sources

With this change, all output surfaces in the renderer provide begin
frame sources directly rather than via an additional parameter to
the LayerTreeHost.

The functional difference between these two paths is that the external
begin frame source exists for the lifetime of the compositor, whereas
the output surface begin frame source is only provided between
BindToClient / DetachFromClient.  The cc::Scheduler is already robust
to these changes and the browser compositor already uses this logic.

The output surface path is more useful in the browser where the output
surface may change which begin frame source it is providing.  However,
this patch just modifies the renderer to use the same path as the
browser so there's only one way of doing things.

In this patch, any time a CompositorOutputSurface or
SynchronousCompositorOutputSurface would be used, RenderThreadImpl
passes it the CompositorExternalBeginFrameSource that would have been
used as the external begin frame source parameter.

The other cases of output surface creation do not need an external begin
frame source.  Layout tests composite synchronously without a
cc::Scheduler and so do not need one.  ui::OutputSurface for mus
renderers internally creates its own default begin frame source (for now).
Blimp doesn't create an output surface so no longer needs special
case code for an external begin frame source that it also doesn't use.

The external begin frame source setting and plumbing is not cleaned up
in this patch and will be done in a followup.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Committed: https://crrev.com/077ba4863c28cbcb1049c24148aeae09f1dfa646
Cr-Commit-Position: refs/heads/master@{#418040}
==========

[commit-bot: I haz the power, 2016-09-12 21:01:43]

Patchset 4 (id:??) landed as
https://crrev.com/077ba4863c28cbcb1049c24148aeae09f1dfa646
Cr-Commit-Position: refs/heads/master@{#418040}