Add more logging and fix PTS overflow for HW decoder.

Add more logging and fix PTS overflow for HW decoder.

- Reduce maximum pending frames for H.264 decoder to 8.
- Log data for next 2 output frames every time frame drop
happens or decoder drain is triggered.
- When timeout happens for dequeueInputBuffer call try to
drain the decoder and get input buffer one more time.
- Fix PTS values overflow.

Committed: https://crrev.com/ae95ff32ff6920660aacf0978a3cb6c4be595a40
Cr-Commit-Position: refs/heads/master@{#11492}

[AlexG, 2016-02-03 23:57:55]

glaznev@webrtc.org changed reviewers:
+ perkj@webrtc.org

[AlexG, 2016-02-03 23:57:56]

PTAL

[perkj_webrtc, 2016-02-04 07:22:09]

can you please add link to bug? 

lgtm

https://codereview.webrtc.org/1661203002/diff/1/talk/app/webrtc/java/jni/andr...
File talk/app/webrtc/java/jni/androidmediadecoder_jni.cc (right):

https://codereview.webrtc.org/1661203002/diff/1/talk/app/webrtc/java/jni/andr...
talk/app/webrtc/java/jni/androidmediadecoder_jni.cc:575: if
(!DeliverPendingOutputs(jni, kMediaCodecPollMs)) {
why would this help?  
Is  the problem that while ((frames_received_ > frames_decoded_ +
max_pending_frames  include both pending input and output buffers?

https://codereview.webrtc.org/1661203002/diff/1/talk/app/webrtc/java/jni/andr...
talk/app/webrtc/java/jni/androidmediadecoder_jni.cc:601:
static_cast<int64_t>(frames_received_) * 1000000 / codec_.maxFramerate);
good catch.

[AlexG, 2016-02-04 18:47:24]

https://codereview.webrtc.org/1661203002/diff/1/talk/app/webrtc/java/jni/andr...
File talk/app/webrtc/java/jni/androidmediadecoder_jni.cc (right):

https://codereview.webrtc.org/1661203002/diff/1/talk/app/webrtc/java/jni/andr...
talk/app/webrtc/java/jni/androidmediadecoder_jni.cc:575: if
(!DeliverPendingOutputs(jni, kMediaCodecPollMs)) {
On 2016/02/04 07:22:08, perkj_webrtc wrote:
> why would this help?  
> Is  the problem that while ((frames_received_ > frames_decoded_ +
> max_pending_frames  include both pending input and output buffers?

Yes, I think it is hard to predict how many actual input buffers decoder has -
length of inputBuffers in MEdiaCodecVideoDecoder.java can be 4, but decoder will
be ok to accept 5 or more - I guess it just copies input frame to its internal
memory.

[AlexG, 2016-02-04 18:48:22]

The CQ bit was checked by glaznev@webrtc.org

[AlexG, 2016-02-04 18:48:22]

The patchset sent to the CQ was uploaded after l-g-t-m from perkj@webrtc.org
Link to the patchset: https://codereview.webrtc.org/1661203002/#ps20001 (title:
"Minor fix")

[commit-bot: I haz the power, 2016-02-04 18:48:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1661203002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1661203002/20001

[commit-bot: I haz the power, 2016-02-04 19:47:17]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-04 19:47:25]

Description was changed from

==========
Add more logging and fix PTS overflow for HW decoder.

- Reduce maximum pending frames for H.264 decoder to 8.
- Log data for next 2 output frames every time frame drop
happens or decoder drain is triggered.
- When timeout happens for dequeueInputBuffer call try to
drain the decoder and get input buffer one more time.
- Fix PTS values overflow.
==========

to

==========
Add more logging and fix PTS overflow for HW decoder.

- Reduce maximum pending frames for H.264 decoder to 8.
- Log data for next 2 output frames every time frame drop
happens or decoder drain is triggered.
- When timeout happens for dequeueInputBuffer call try to
drain the decoder and get input buffer one more time.
- Fix PTS values overflow.

Committed: https://crrev.com/ae95ff32ff6920660aacf0978a3cb6c4be595a40
Cr-Commit-Position: refs/heads/master@{#11492}
==========

[commit-bot: I haz the power, 2016-02-04 19:47:26]

Patchset 2 (id:??) landed as
https://crrev.com/ae95ff32ff6920660aacf0978a3cb6c4be595a40
Cr-Commit-Position: refs/heads/master@{#11492}