Issue 660471 link

Starred by 2 users

Issue metadata

Status:	WontFix
Owner:	peah@chromium.org
Closed:	Nov 2016
Cc:	█ kjellander@chromium.org
EstimatedDays:	----
NextAction:	----
OS:	----
Pri:	2
Type:	Bug-Regression
M-54

Win perf regression in APM/level controller numbers at 14781

Project Member Reported by kjellander@chromium.org, Oct 28 2016

Issue description

These look a lot like  bug 659768  and  bug 659767  to me except those bugs are for Mac only and this is Windows. 

The revision this seems to trigger on doesn't make any sense though, since https://chromium.googlesource.com/external/webrtc/+/a8bec8d8e74e95998d3a18d7b7fed41b4c570e37 is a change that doesn't affect webrtc_perf_tests or any production code at all.

Per: should we just close this or should it be merged with any of the above mentioned bugs, even if it happens 12 revisions later?

Comment 1 by kjellander@chromium.org, Oct 28 2016

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=660471

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgvdyLswkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgvbTqowoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgvfmWrgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgvYGupgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg3fPl7AsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgvd3ntwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg_Z6svAsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgg5rarQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg_cydtAsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg_YuusgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg_dXrvgsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgg67btAoM


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

webrtc-win-large-tests

Comment 2 by phoglund@chromium.org, Nov 8 2016

Status: WontFix (was: Assigned)

This bugs contains both level controller changes and other random neteq_performance regressions. I don't think those are related. Also the pointed-out CL 14781 changes java code, so it can't be guilty. No changes in the vicinity look guilty either. I think the win-large-test regressions around this time were cosmic rays / bad luck / machine problems.

►

Issue 660471 link

Issue metadata

Win perf regression in APM/level controller numbers at 14781

Issue description

Comment 1 by kjellander@chromium.org, Oct 28 2016

Comment 1 Deleted

Comment 2 by phoglund@chromium.org, Nov 8 2016

Comment 2 Deleted

Sign in to add a comment