Set content opacity of DFHAndroid surfacelayer based on renderer frame contents.

Set content opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Committed: https://crrev.com/862bef99a8b729f501689b0658f029464ff84135
Cr-Commit-Position: refs/heads/master@{#417131}

[jbauman, 2016-09-01 23:18:50]

Description was changed from

==========
Set opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
==========

to

==========
Set opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

[jbauman, 2016-09-01 23:18:52]

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

[commit-bot: I haz the power, 2016-09-01 23:19:21]

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

[jbauman, 2016-09-01 23:19:36]

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

[commit-bot: I haz the power, 2016-09-01 23:19:43]

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, 2016-09-02 00:18:48]

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

[commit-bot: I haz the power, 2016-09-02 00:18:49]

Dry run: This issue passed the CQ dry run.

[jbauman, 2016-09-02 00:33:57]

Description was changed from

==========
Set opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

to

==========
Set content opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

[jbauman, 2016-09-02 00:35:15]

jbauman@chromium.org changed reviewers:
+ sadrul@chromium.org, sievers@chromium.org

[jbauman, 2016-09-02 03:06:56]

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

[commit-bot: I haz the power, 2016-09-02 03:07:15]

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, 2016-09-02 03:42:36]

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

[commit-bot: I haz the power, 2016-09-02 03:42:37]

Dry run: This issue passed the CQ dry run.

[jbauman, 2016-09-02 04:40:34]

Apparently we have to change the background color in fullscreen when overlay
video is disabled, so added that.

[sadrul, 2016-09-02 18:29:13]

I assume you added me for ui/ changes? Would you mind adding an owner from
ui/android/ instead? (I am not a good reviewer for android specific changes)

[jbauman, 2016-09-02 19:46:21]

jbauman@chromium.org changed reviewers:
+ tedchoc@chromium.org
- sadrul@chromium.org

[no sievers, 2016-09-02 22:06:44]

sievers@chromium.org changed reviewers:
+ boliu@chromium.org

[no sievers, 2016-09-02 22:06:52]

lgtm

https://codereview.chromium.org/2302023003/diff/40001/content/browser/rendere...
File content/browser/renderer_host/render_widget_host_view_android.cc (right):

https://codereview.chromium.org/2302023003/diff/40001/content/browser/rendere...
content/browser/renderer_host/render_widget_host_view_android.cc:1031:
OnFrameMetadataUpdated(frame_metadata.Clone(), false);
+Bo for whether WebView needs to worry about this too.

https://codereview.chromium.org/2302023003/diff/40001/ui/android/delegated_fr...
File ui/android/delegated_frame_host_android.cc (right):

https://codereview.chromium.org/2302023003/diff/40001/ui/android/delegated_fr...
ui/android/delegated_frame_host_android.cc:175:
!current_frame_->has_transparent_background);
nit: maybe just make the readback layer always opaque? you're not getting any
video anyways.

[boliu, 2016-09-02 22:52:11]

https://codereview.chromium.org/2302023003/diff/40001/content/browser/rendere...
File content/browser/renderer_host/render_widget_host_view_android.cc (right):

https://codereview.chromium.org/2302023003/diff/40001/content/browser/rendere...
content/browser/renderer_host/render_widget_host_view_android.cc:1031:
OnFrameMetadataUpdated(frame_metadata.Clone(), false);
On 2016/09/02 22:06:52, sievers_OOOtil_Sep12 wrote:
> +Bo for whether WebView needs to worry about this too.

Don't think so. Webview disables video overlays because it has no benefit there.

[watk, 2016-09-02 23:07:35]

Yay! \o/ 

Btw, a field trial just landed for experimenting with not ever using Overlay
Fullscreen Video mode, and it was accidentally turned off by default (i.e., we
wouldn't exercise any of this code by default). If your checkout has that change
you can force it with --force-fieldtrials=OverlayFullscreenVideo/Enabled

(It's being reverted now)

[jbauman, 2016-09-03 01:12:18]

On 2016/09/02 22:06:52, sievers_OOOtil_Sep12 wrote:
> lgtm
> 
>
https://codereview.chromium.org/2302023003/diff/40001/content/browser/rendere...
> File content/browser/renderer_host/render_widget_host_view_android.cc (right):
> 
>
https://codereview.chromium.org/2302023003/diff/40001/content/browser/rendere...
> content/browser/renderer_host/render_widget_host_view_android.cc:1031:
> OnFrameMetadataUpdated(frame_metadata.Clone(), false);
> +Bo for whether WebView needs to worry about this too.
> 
>
https://codereview.chromium.org/2302023003/diff/40001/ui/android/delegated_fr...
> File ui/android/delegated_frame_host_android.cc (right):
> 
>
https://codereview.chromium.org/2302023003/diff/40001/ui/android/delegated_fr...
> ui/android/delegated_frame_host_android.cc:175:
> !current_frame_->has_transparent_background);
> nit: maybe just make the readback layer always opaque? you're not getting any
> video anyways.

I'd prefer to keep them consistent for now so we have fewer weird variations in
rendering.

[jbauman, 2016-09-07 22:05:39]

tedchoc@: ui/android OWNERS review

[Ted C, 2016-09-07 23:02:48]

On 2016/09/07 22:05:39, jbauman wrote:
> tedchoc@: ui/android OWNERS review

ui/android - lgtm

[jbauman, 2016-09-07 23:19:29]

The CQ bit was checked by jbauman@chromium.org

[commit-bot: I haz the power, 2016-09-07 23:20:18]

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, 2016-09-08 00:33:03]

Description was changed from

==========
Set content opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

to

==========
Set content opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

[commit-bot: I haz the power, 2016-09-08 00:33:04]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-09-08 00:34:50]

Description was changed from

==========
Set content opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
==========

to

==========
Set content opacity of DFHAndroid surfacelayer based on renderer frame contents.

The browser updates the overlay video mode flag before the renderer can
send a new frame. This means that when switching from fullscreen to
non-fullscreen the browser was expecting the renderer the fill the frame
completely and wasn't drawing the background, but it was accidentally
leaving part of the frame corrupted and uncleared.

BUG=631866
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Committed: https://crrev.com/862bef99a8b729f501689b0658f029464ff84135
Cr-Commit-Position: refs/heads/master@{#417131}
==========

[commit-bot: I haz the power, 2016-09-08 00:34:51]

Patchset 3 (id:??) landed as
https://crrev.com/862bef99a8b729f501689b0658f029464ff84135
Cr-Commit-Position: refs/heads/master@{#417131}