Android Java Surfaces can now be looked up via GpuSurfaceTracker

Android Java Surfaces can now be looked up via GpuSurfaceTracker

AndroidVideoDecodeAccelerator needs to be able to retrieve
registered Surfaces in order to decode to them. Previously this was
only possible to do from the GPU process through
SurfaceTextureManagerImpl, but now it can be done from the browser
through GpuSurfaceTracker. This enables fullscreen video decoding with
AVDA when in-process-gpu is enabled.

BUG=601952
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Committed: https://crrev.com/2b597036f28ef872717bdfb0c337e43ae11170a7
Cr-Commit-Position: refs/heads/master@{#386531}

[watk, 2016-04-08 22:28:56]

Description was changed from

==========
Android Java Surfaces can now be looked up via GpuSurfaceTracker

AndroidVideoDecodeAccelerator needs to be able to retrieve
registered Surfaces in order to decode to them. Previously this was
only possible to do from the GPU process through
SurfaceTextureManagerImpl, but now it can be done from the browser
through GpuSurfaceTracker. This enables fullscreen video decoding with
AVDA when in-process-gpu is enabled.

BUG=601952
==========

to

==========
Android Java Surfaces can now be looked up via GpuSurfaceTracker

AndroidVideoDecodeAccelerator needs to be able to retrieve
registered Surfaces in order to decode to them. Previously this was
only possible to do from the GPU process through
SurfaceTextureManagerImpl, but now it can be done from the browser
through GpuSurfaceTracker. This enables fullscreen video decoding with
AVDA when in-process-gpu is enabled.

BUG=601952
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[watk, 2016-04-08 22:30:56]

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

[watk, 2016-04-08 22:30:56]

Hey sievers, PTAL. I wasn't thinking about in-process-gpu when I first added
AcquireJavaSurface..

[no sievers, 2016-04-11 18:03:41]

lgtm

[watk, 2016-04-11 18:11:01]

watk@chromium.org changed reviewers:
+ wfh@chromium.org

[watk, 2016-04-11 18:11:02]

wfh@chromium.org: Please review changes in gpu/ipc

[Will Harris, 2016-04-11 21:03:15]

lgtm, seems this is just hooking the APIs up in common, was there another CL
that actually transferred something new over IPC?

[watk, 2016-04-11 21:55:19]

On 2016/04/11 21:03:15, Will Harris wrote:
> lgtm, seems this is just hooking the APIs up in common, was there another CL
> that actually transferred something new over IPC?

Yeah, there's no new IPC in this one. The CL that added new IPC is here:
https://codereview.chromium.org/1655083002

[Will Harris, 2016-04-11 21:57:25]

On 2016/04/11 21:55:19, watk wrote:
> On 2016/04/11 21:03:15, Will Harris wrote:
> > lgtm, seems this is just hooking the APIs up in common, was there another CL
> > that actually transferred something new over IPC?
> 
> Yeah, there's no new IPC in this one. The CL that added new IPC is here:
> https://codereview.chromium.org/1655083002

okay thanks, if possible it helps to use same security reviewer when reviewing
two similar CLs in the same project, so they have all the context. thanks!

[watk, 2016-04-11 21:59:28]

On 2016/04/11 21:57:25, Will Harris wrote:
> On 2016/04/11 21:55:19, watk wrote:
> > On 2016/04/11 21:03:15, Will Harris wrote:
> > > lgtm, seems this is just hooking the APIs up in common, was there another
CL
> > > that actually transferred something new over IPC?
> > 
> > Yeah, there's no new IPC in this one. The CL that added new IPC is here:
> > https://codereview.chromium.org/1655083002
> 
> okay thanks, if possible it helps to use same security reviewer when reviewing
> two similar CLs in the same project, so they have all the context. thanks!

That makes sense. Will do from now on. :)

[watk, 2016-04-11 22:12:56]

The CQ bit was checked by watk@chromium.org

[commit-bot: I haz the power, 2016-04-11 22:13:57]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1874023002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1874023002/1

[commit-bot: I haz the power, 2016-04-12 00:33:57]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-04-12 00:35:12]

Description was changed from

==========
Android Java Surfaces can now be looked up via GpuSurfaceTracker

AndroidVideoDecodeAccelerator needs to be able to retrieve
registered Surfaces in order to decode to them. Previously this was
only possible to do from the GPU process through
SurfaceTextureManagerImpl, but now it can be done from the browser
through GpuSurfaceTracker. This enables fullscreen video decoding with
AVDA when in-process-gpu is enabled.

BUG=601952
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Android Java Surfaces can now be looked up via GpuSurfaceTracker

AndroidVideoDecodeAccelerator needs to be able to retrieve
registered Surfaces in order to decode to them. Previously this was
only possible to do from the GPU process through
SurfaceTextureManagerImpl, but now it can be done from the browser
through GpuSurfaceTracker. This enables fullscreen video decoding with
AVDA when in-process-gpu is enabled.

BUG=601952
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Committed: https://crrev.com/2b597036f28ef872717bdfb0c337e43ae11170a7
Cr-Commit-Position: refs/heads/master@{#386531}
==========

[commit-bot: I haz the power, 2016-04-12 00:35:13]

Patchset 1 (id:??) landed as
https://crrev.com/2b597036f28ef872717bdfb0c337e43ae11170a7
Cr-Commit-Position: refs/heads/master@{#386531}