Force spontaneous draws into the SW path

Force spontaneous draws into the SW path

Any unrequested draws will cause UI corruption if they hit the GL path.
The strategy is to disallow any such draws, but to enable diagnosis
force them into the SW path where they cause less UI damange but we can
log or assert as appropriate.

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=208014

[boliu, 2013-06-21 22:40:14]

lgtm

https://codereview.chromium.org/17553012/diff/1/content/browser/android/in_pr...
File content/browser/android/in_process/synchronous_compositor_output_surface.cc
(right):

https://codereview.chromium.org/17553012/diff/1/content/browser/android/in_pr...
content/browser/android/in_process/synchronous_compositor_output_surface.cc:226:
invoking_composite_ = true;
Use base::AutoReset?

[boliu, 2013-06-21 22:44:23]

https://codereview.chromium.org/17553012/diff/1/content/browser/android/in_pr...
File content/browser/android/in_process/synchronous_compositor_output_surface.cc
(right):

https://codereview.chromium.org/17553012/diff/1/content/browser/android/in_pr...
content/browser/android/in_process/synchronous_compositor_output_surface.cc:141:
void SynchronousCompositorOutputSurface::SwapBuffers(
Hmm...slightly tangent, but we could invalidate the view from here if
(!invoking_compositing), so a forced draw is always followed by a real draw...oh
the hacks

[joth, 2013-06-21 23:02:13]

thanks

https://codereview.chromium.org/17553012/diff/1/content/browser/android/in_pr...
File content/browser/android/in_process/synchronous_compositor_output_surface.cc
(right):

https://codereview.chromium.org/17553012/diff/1/content/browser/android/in_pr...
content/browser/android/in_process/synchronous_compositor_output_surface.cc:141:
void SynchronousCompositorOutputSurface::SwapBuffers(
On 2013/06/21 22:44:23, boliu wrote:
> Hmm...slightly tangent, but we could invalidate the view from here if
> (!invoking_compositing), so a forced draw is always followed by a real
draw...oh
> the hacks

yeah...... I'd rather not make it any more complicated. The code as it stands is
mostly funnelling us to the NOTREACHED() on line 64 anyway, so no point having
even more workaround beyond it.

https://codereview.chromium.org/17553012/diff/1/content/browser/android/in_pr...
content/browser/android/in_process/synchronous_compositor_output_surface.cc:226:
invoking_composite_ = true;
On 2013/06/21 22:40:14, boliu wrote:
> Use base::AutoReset?

Done.

Nice! Thought that existed, but I was searching the code for a ScopedFlibble
class...

[commit-bot: I haz the power, 2013-06-21 23:04:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/joth@chromium.org/17553012/3002

[commit-bot: I haz the power, 2013-06-22 02:46:48]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/joth@chromium.org/17553012/3002

[commit-bot: I haz the power, 2013-06-22 03:09:46]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/joth@chromium.org/17553012/3002

[commit-bot: I haz the power, 2013-06-22 03:22:57]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/joth@chromium.org/17553012/3002

[commit-bot: I haz the power, 2013-06-22 03:37:46]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/joth@chromium.org/17553012/3002

[commit-bot: I haz the power, 2013-06-22 07:06:20]

Change committed as 208014