New issue
Advanced search Search tips

Issue 717518 link

Starred by 1 user
Status: Duplicate
Owner: ----
Closed: May 2017
Cc:
alessiob@chromium.org
ossu@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

68.2% regression in browser_tests at 467955:467961

Reported by alessiob@webrtc.org, May 2 2017

Comment 1 by alessiob@webrtc.org, May 2 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=717518

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg0p6esgoM


Bot(s) for this bug's original alert(s):

chromium-webrtc-rel-7

Comment 2 by kwiberg@chromium.org, May 2 2017

Status: Untriaged (was: Assigned)
Unlikely. That CL makes changes to some unit tests that don't use any audio codecs. The regression is about a use of G722 in an entirely separate group of tests.

Comment 3 by kwiberg@chromium.org, May 2 2017

Owner: ----

Comment 4 by kwiberg@chromium.org, May 2 2017

Cc: ossu@chromium.org
CC ossu@, who landed https://chromium.googlesource.com/external/webrtc/trunk/webrtc/+/fe5f71cd91461634b920c92123fbd2a457e4901a and https://chromium.googlesource.com/external/webrtc/trunk/webrtc/+/30af12bd054bd5440d79f43006d730d14fe820c5 in the suspected range. Oskar, could either of those have caused G722 encoding and decoding to take more time?

Comment 5 by ossu@chromium.org, May 2 2017

Mergedinto: 716457
Status: Duplicate (was: Untriaged)
If I read the graphs correctly, this is about send_ and receive_rate, not encoding time. It's the same as chromium:716457 - i.e. CNG now works with G722 as well, so we send less data.

Closing as dupe.
►

Sign in to add a comment