Avoid waiting on the SurfaceTexture until we need to.

Avoid waiting on the SurfaceTexture until we need to.

We can let the SurfaceTexture operations complete in the background
for back buffer rendering instead of waiting during frame delivery;
this frees up the GPU thread for other tasks.

Also reduces the wait time to be more inline with the UMA stats;
some devices like the MotoX don't reliably deliver this event when
the MediaCodec is underflowing. I've also seen the N7v2 drop this
event every now and then.

BUG=601066
TEST=no playback regressions, less waiting.

Committed: https://crrev.com/501f6225e00045e88ba56d7fe9deed6850c2906c
Cr-Commit-Position: refs/heads/master@{#391426}

[DaleCurtis, 2016-05-02 19:03:49]

Patchset #1 (id:1) has been deleted

[DaleCurtis, 2016-05-02 19:03:52]

Patchset #1 (id:20001) has been deleted

[DaleCurtis, 2016-05-02 19:03:55]

Patchset #1 (id:40001) has been deleted

[DaleCurtis, 2016-05-02 19:10:36]

Description was changed from

==========
Avoid waiting on the SurfaceTexture until we need to.

We can let the SurfaceTexture operations complete in the background
for back buffer rendering instead of waiting during frame delivery;
this frees up the GPU thread for other tasks.

BUG=xxx
TEST=xxx
==========

to

==========
Avoid waiting on the SurfaceTexture until we need to.

We can let the SurfaceTexture operations complete in the background
for back buffer rendering instead of waiting during frame delivery;
this frees up the GPU thread for other tasks.

Also reduces the wait time to be more inline with the UMA stats;
some devices like the MotoX don't reliably deliver this event when
the MediaCodec is underflowing. I've also seen the N7v2 drop this
event every now and then.

BUG=601066
TEST=no playback regressions, less waiting.
==========

[DaleCurtis, 2016-05-02 19:10:48]

dalecurtis@chromium.org changed reviewers:
+ liberato@chromium.org, watk@chromium.org

[watk, 2016-05-02 23:32:34]

lgtm

https://codereview.chromium.org/1924973004/diff/80001/media/gpu/avda_shared_s...
File media/gpu/avda_shared_state.h (right):

https://codereview.chromium.org/1924973004/diff/80001/media/gpu/avda_shared_s...
media/gpu/avda_shared_state.h:82: // rendering to a SurfaceTexture; this method
should
Comment abruptly ends :)

https://codereview.chromium.org/1924973004/diff/80001/media/gpu/avda_shared_s...
media/gpu/avda_shared_state.h:85: // by using this helper we can attempt to let
this process complete in a non
delete one "by"

[DaleCurtis, 2016-05-03 00:17:54]

https://codereview.chromium.org/1924973004/diff/80001/media/gpu/avda_shared_s...
File media/gpu/avda_shared_state.h (right):

https://codereview.chromium.org/1924973004/diff/80001/media/gpu/avda_shared_s...
media/gpu/avda_shared_state.h:82: // rendering to a SurfaceTexture; this method
should
On 2016/05/02 at 23:32:34, watk wrote:
> Comment abruptly ends :)

Done.

https://codereview.chromium.org/1924973004/diff/80001/media/gpu/avda_shared_s...
media/gpu/avda_shared_state.h:85: // by using this helper we can attempt to let
this process complete in a non
On 2016/05/02 at 23:32:34, watk wrote:
> delete one "by"

You passed the if if test!

[liberato (no reviews please), 2016-05-03 05:49:30]

nice.  lgtm.

-fl

https://codereview.chromium.org/1924973004/diff/100001/media/gpu/avda_shared_...
File media/gpu/avda_shared_state.h (right):

https://codereview.chromium.org/1924973004/diff/100001/media/gpu/avda_shared_...
media/gpu/avda_shared_state.h:126: // there has been no last call.
not quite accurate -- null if and only if a frame available signal is expected
but not waited for.

[DaleCurtis, 2016-05-04 00:25:29]

https://codereview.chromium.org/1924973004/diff/100001/media/gpu/avda_shared_...
File media/gpu/avda_shared_state.h (right):

https://codereview.chromium.org/1924973004/diff/100001/media/gpu/avda_shared_...
media/gpu/avda_shared_state.h:126: // there has been no last call.
On 2016/05/03 at 05:49:30, liberato wrote:
> not quite accurate -- null if and only if a frame available signal is expected
but not waited for.

Actually, I think your suggestion isn't accurate :) I think it's missing a "not"
at the beginning. In any case I've added some verbosity here which should make
this clearer.

[DaleCurtis, 2016-05-04 01:45:14]

The CQ bit was checked by dalecurtis@chromium.org

[DaleCurtis, 2016-05-04 01:45:14]

The patchset sent to the CQ was uploaded after l-g-t-m from watk@chromium.org,
liberato@chromium.org
Link to the patchset: https://codereview.chromium.org/1924973004/#ps120001
(title: "Comment fix.")

[commit-bot: I haz the power, 2016-05-04 01:45:36]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1924973004/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1924973004/120001

[commit-bot: I haz the power, 2016-05-04 02:43:40]

Committed patchset #4 (id:120001)

[commit-bot: I haz the power, 2016-05-04 02:44:35]

Description was changed from

==========
Avoid waiting on the SurfaceTexture until we need to.

We can let the SurfaceTexture operations complete in the background
for back buffer rendering instead of waiting during frame delivery;
this frees up the GPU thread for other tasks.

Also reduces the wait time to be more inline with the UMA stats;
some devices like the MotoX don't reliably deliver this event when
the MediaCodec is underflowing. I've also seen the N7v2 drop this
event every now and then.

BUG=601066
TEST=no playback regressions, less waiting.
==========

to

==========
Avoid waiting on the SurfaceTexture until we need to.

We can let the SurfaceTexture operations complete in the background
for back buffer rendering instead of waiting during frame delivery;
this frees up the GPU thread for other tasks.

Also reduces the wait time to be more inline with the UMA stats;
some devices like the MotoX don't reliably deliver this event when
the MediaCodec is underflowing. I've also seen the N7v2 drop this
event every now and then.

BUG=601066
TEST=no playback regressions, less waiting.

Committed: https://crrev.com/501f6225e00045e88ba56d7fe9deed6850c2906c
Cr-Commit-Position: refs/heads/master@{#391426}
==========

[commit-bot: I haz the power, 2016-05-04 02:44:36]

Patchset 4 (id:??) landed as
https://crrev.com/501f6225e00045e88ba56d7fe9deed6850c2906c
Cr-Commit-Position: refs/heads/master@{#391426}