New issue
Advanced search Search tips
Note: Color blocks (like █ or █) mean that a user may not be available. Tooltip shows the reason.

Issue 880985 link

Starred by 1 user
Status: Duplicate
Owner:
tguilbert@chromium.org
Closed: Sep 6
Cc:
█ webrtc-c...@skia-buildbots.google.com.iam.gserviceaccount.com
█ olka@chromium.org
tguilbert@chromium.org
johnchen@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

41.4%-106.2% regression in media.desktop at 588401:588424

Project Member Reported by tguilbert@chromium.org, Sep 5 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Sep 5 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=880985

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=2b6ed91188368f2db677dd87bb28f27fa078e7d8a87ad9adde85a9acd7e4cf3a


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

linux-perf
mac-10_12_laptop_low_end-perf
mac-10_13_laptop_high_end-perf

media.desktop - Benchmark documentation link:
  None
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Sep 5

Cc: johnchen@chromium.org
Owner: johnchen@chromium.org
Status: Assigned (was: Untriaged)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16e9541b640000

[ChromeDriver] Disable ChromeOptionsFunctionalTest by johnchen@chromium.org
https://chromium.googlesource.com/chromium/src/+/35893b77275d1bd4208bcb8596fb526790cefeb5
82.5 → 200.5 (+118)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Sep 5 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/11b80859640000

Comment 6 by johnchen@chromium.org, Sep 5

Owner: tguilbert@chromium.org
The CL identified in comment #3 couldn't have affected the perf of media benchmark. Running the pinpoint job with a wider range wasn't able to reproduce a difference. The perf curve also looks a little jumpy. Not sure what else I can do.
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Sep 6

Cc: webrtc-c...@skia-buildbots.google.com.iam.gserviceaccount.com olka@chromium.org
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/177bdf5d640000

Roll src/third_party/webrtc f18b35284288..689b5874d4fb (50 commits) by webrtc-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/ed774e4540ee3afb2dd4ac22ef3fb7bcc89f7e57
43.42 → 46.21 (+2.792)

Restoring field trial config for AudioService experiment on linux/mac by olka@chromium.org
https://chromium.googlesource.com/chromium/src/+/7ec94d0b3562d9414f716a1d6cb711d2fcf12c2f
50.97 → 84.97 (+34.01)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Benchmark documentation link:
  None

Comment 9 by tguilbert@chromium.org, Sep 6

Mergedinto: 870171
Status: Duplicate (was: Assigned)
►

Sign in to add a comment