Filter send-only and recv-only as appropriate in WebRTC perf tests.

Filter send-only and recv-only as appropriate in WebRTC perf tests.

This should work around the problem where send-only stats appear and
disappear because of the race in 598623.

BUG=596398
Committed: https://crrev.com/569c778e5750343f2b0b8144a86dec762d61a703
Cr-Commit-Position: refs/heads/master@{#383744}

[phoglund_chromium, 2016-03-29 11:54:54]

phoglund@chromium.org changed reviewers:
+ kjellander@chromium.org

[kjellander_chromium, 2016-03-29 12:52:26]

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
File chrome/browser/media/webrtc_browsertest_perf.cc (right):

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
chrome/browser/media/webrtc_browsertest_perf.cc:260:
MaybePrintResultsForAudioSend(ssrc, pc_dict, modifier);
Shouldn't we continue to check the result of these methods as in the previous
implementation?

[phoglund_chromium, 2016-03-29 14:02:09]

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
File chrome/browser/media/webrtc_browsertest_perf.cc (right):

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
chrome/browser/media/webrtc_browsertest_perf.cc:260:
MaybePrintResultsForAudioSend(ssrc, pc_dict, modifier);
On 2016/03/29 12:52:26, kjellander (chromium) wrote:
> Shouldn't we continue to check the result of these methods as in the previous
> implementation?

That would be nice, but it seemed clunky to implement. In that case the two new
Print methods would need to coordinate among each other since one knows about
recv and one send. I didn't think it was worth it. I don't know if we care if a
new SSRC comes along, anyway. We do care if ssrc types disappear though, but
that's caught by data stoppage alerts.

[kjellander_chromium, 2016-03-29 14:10:31]

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
File chrome/browser/media/webrtc_browsertest_perf.cc (right):

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
chrome/browser/media/webrtc_browsertest_perf.cc:260:
MaybePrintResultsForAudioSend(ssrc, pc_dict, modifier);
On 2016/03/29 14:02:09, phoglund_chrome wrote:
> On 2016/03/29 12:52:26, kjellander (chromium) wrote:
> > Shouldn't we continue to check the result of these methods as in the
previous
> > implementation?
> 
> That would be nice, but it seemed clunky to implement. In that case the two
new
> Print methods would need to coordinate among each other since one knows about
> recv and one send. I didn't think it was worth it. I don't know if we care if
a
> new SSRC comes along, anyway. We do care if ssrc types disappear though, but
> that's caught by data stoppage alerts.

OK, but if we now never look at the result from the Maybe* methods, I suggest
removing the returning of the boolean value from them.

[phoglund_chromium, 2016-03-29 14:37:24]

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
File chrome/browser/media/webrtc_browsertest_perf.cc (right):

https://codereview.chromium.org/1839653003/diff/1/chrome/browser/media/webrtc...
chrome/browser/media/webrtc_browsertest_perf.cc:260:
MaybePrintResultsForAudioSend(ssrc, pc_dict, modifier);
On 2016/03/29 14:10:31, kjellander (chromium) wrote:
> On 2016/03/29 14:02:09, phoglund_chrome wrote:
> > On 2016/03/29 12:52:26, kjellander (chromium) wrote:
> > > Shouldn't we continue to check the result of these methods as in the
> previous
> > > implementation?
> > 
> > That would be nice, but it seemed clunky to implement. In that case the two
> new
> > Print methods would need to coordinate among each other since one knows
about
> > recv and one send. I didn't think it was worth it. I don't know if we care
if
> a
> > new SSRC comes along, anyway. We do care if ssrc types disappear though, but
> > that's caught by data stoppage alerts.
> 
> OK, but if we now never look at the result from the Maybe* methods, I suggest
> removing the returning of the boolean value from them.

Good point. Done

[kjellander_chromium, 2016-03-29 15:04:50]

lgtm

[phoglund_chromium, 2016-03-29 15:40:52]

The CQ bit was checked by phoglund@chromium.org

[commit-bot: I haz the power, 2016-03-29 15:41:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1839653003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1839653003/20001

[commit-bot: I haz the power, 2016-03-29 16:49:26]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-03-29 16:50:44]

Description was changed from

==========
Filter send-only and recv-only as appropriate in WebRTC perf tests.

This should work around the problem where send-only stats appear and
disappear because of the race in 598623.

BUG=596398
==========

to

==========
Filter send-only and recv-only as appropriate in WebRTC perf tests.

This should work around the problem where send-only stats appear and
disappear because of the race in 598623.

BUG=596398

Committed: https://crrev.com/569c778e5750343f2b0b8144a86dec762d61a703
Cr-Commit-Position: refs/heads/master@{#383744}
==========

[commit-bot: I haz the power, 2016-03-29 16:50:44]

Patchset 2 (id:??) landed as
https://crrev.com/569c778e5750343f2b0b8144a86dec762d61a703
Cr-Commit-Position: refs/heads/master@{#383744}