Fix the EOS hang.

Fix the EOS hang and jerkiness in playback.

- output_picture_ should be cleared only after 
freeing the output buffers.

- pop out free_input_buffers_ in Flush. This was missing
and creates failure during FreeInputBuffers()

- new picture_buffers to be pushed back into
assigned_picture_buffers_. 
Otherwise display gets wrong texture

These changes will fix the hang at EOS.

BUG=none
TEST=gpu_video_tests

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=86254

[vhiremath, 2011-05-17 06:30:39]

Fix EOS hang for video playback.

[Ami GONE FROM CHROMIUM, 2011-05-17 15:22:46]

LGTM

[commit-bot: I haz the power, 2011-05-17 15:23:11]

Presubmit check for 7017006-2003 failed and returned exit status 1.

Running presubmit commit checks ...

** Presubmit Messages **
If this change has an associated bug, add BUG=[bug number].

If this change requires manual test instructions to QA team, add
TEST=[instructions].

** Presubmit ERRORS **
Missing LGTM from an OWNER for: content/gpu/omx_video_decode_accelerator.cc

[Ami GONE FROM CHROMIUM, 2011-05-17 15:25:31]

+brettw for OWNERS approval (and commit-box checking if LGTY).

[brettw, 2011-05-17 15:38:13]

It's not clear what the problem is. Can the change list description include
this?

[vhiremath, 2011-05-19 09:31:53]

On 2011/05/17 15:38:13, brettw wrote:
> It's not clear what the problem is. Can the change list description include
> this?

Uploaded the changes with more description.
Thank you.

[vhiremath, 2011-05-19 13:38:00]

Uploaded new patch and updated the description accordingly.
Thank you.

http://codereview.chromium.org/7017006/diff/6002/content/gpu/omx_video_decode...
File content/gpu/omx_video_decode_accelerator.cc (right):

http://codereview.chromium.org/7017006/diff/6002/content/gpu/omx_video_decode...
content/gpu/omx_video_decode_accelerator.cc:297:
assigned_picture_buffers_.end(),
Newer picture buffers should be inserted in the end. Not in the beginning. This
fixes the jerkiness in playback.

http://codereview.chromium.org/7017006/diff/6002/content/gpu/omx_video_decode...
content/gpu/omx_video_decode_accelerator.cc:859: return false;
This check is useful in case we get ReusePictureBuffer after Flush from app is
called.
The openmax component is closed and no need to even GetState of component when
it is already closed.

[brettw, 2011-05-19 16:38:19]

This LGTM from a high level (I don't understand the details so I'm trusting
Ami).

[commit-bot: I haz the power, 2011-05-19 18:19:16]

Try job failure for 7017006-10001 on mac:
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac&number...

[Ami GONE FROM CHROMIUM, 2011-05-19 19:31:23]

On 2011/05/19 18:19:16, commit-bot wrote:
> Try job failure for 7017006-10001 on mac:
>
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac&number...

Failure is unrelated flakiness.  Re-checked commit box.

[vhiremath, 2011-05-21 09:27:49]

I don't see try-server progress on this CL.

Do we need to re-check the commit box ?

Thank you,
|| Vishweshwar || 

-----Original Message-----
From: fischman@chromium.org [mailto:fischman@chromium.org] 
Sent: Friday, May 20, 2011 1:01 AM
To: Vishweshwar Hiremath; vmr@google.com; matti.heikkinen@gmail.com;
brettw@chromium.org; commit-bot@chromium.org
Cc: chromium-reviews@chromium.org; jam@chromium.org; apatrick@chromium.org;
Niranjan Wartikar
Subject: Re: Fix the EOS hang. (issue7017006)

On 2011/05/19 18:19:16, commit-bot wrote:
> Try job failure for 7017006-10001 on mac:

http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac&number...

Failure is unrelated flakiness.  Re-checked commit box.


http://codereview.chromium.org/7017006/
-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may
contain
confidential information.  Any unauthorized review, use, disclosure or
distribution
is prohibited.  If you are not the intended recipient, please contact the sender
by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------

[Ami GONE FROM CHROMIUM, 2011-05-21 15:55:56]

I did re-check the box after the previous failure.  Possibly the bot refuses
to attempt the same patchset more than once.  Can you re-upload the last
patchset and I'll re-check?

[vhiremath, 2011-05-23 04:25:34]

On 2011/05/21 15:55:56, Ami Fischman wrote:
> I did re-check the box after the previous failure.  Possibly the bot refuses
> to attempt the same patchset more than once.  Can you re-upload the last
> patchset and I'll re-check?

Submitted new patch set to enable re-checking.
Ami, please re-check the commit box.
Thank you.

[commit-bot: I haz the power, 2011-05-23 05:49:52]

Change committed as 86254