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

Issue 770860 link

Starred by 1 user
Status: Duplicate
Merged: issue 776506
Owner:
dmazz...@chromium.org
Closed: Nov 2017
Cc:
█ nednguyen@chromium.org
charliea@chromium.org
majidvp@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

4.8%-144.7% regression in system_health.common_desktop at 504379:504545

Project Member Reported by majidvp@chromium.org, Oct 2 2017 
I suspect "Add accessibility coverage to system_health benchmark." [1]. In particular, the regression is improves later but regresses again. This fits nicely with landing of [1], and its revert and re-landing.

Hopefully the bisect points out the culprit.

[1] https://chromium.googlesource.com/chromium/src/+/a5edee6703dcbcdad0a19ef19a4636641c9af7b9
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Oct 2 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=770860

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=37a5771522449e44b58e62271924c803a109b29b5d676ffb4db608fc7f782f51


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

chromium-rel-mac-retina
chromium-rel-mac11
chromium-rel-mac11-pro
chromium-rel-mac12
chromium-rel-mac12-mini-8gb
chromium-rel-win10
chromium-rel-win7-dual
chromium-rel-win7-gpu-ati
chromium-rel-win7-gpu-nvidia
chromium-rel-win7-x64-dual
chromium-rel-win8-dual
win-high-dpi
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Oct 2 2017 


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : cpu_time_percentage_avg/browse_social/browse_social_twitter_infinite_scroll


To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.social.twitter.infinite.scroll system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966813178655557840


For feedback, file a bug with component Speed>Bisection

Comment 6 by sullivan@chromium.org, Oct 4 2017

Cc: majidvp@chromium.org
 Issue 770859  has been merged into this issue.
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Oct 4 2017 


=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : cpu_time_percentage_avg/load_news/load_news_wikipedia

Revision             Result                     N
chromium@504402      0.102844 +- 0.0347449      21      good
chromium@504493      0.103014 +- 0.0289169      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.news.wikipedia system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966712009604860064


For feedback, file a bug with component Speed>Bisection
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Oct 4 2017 


=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : cpu_time_percentage_avg/browse_social/browse_social_twitter

Revision             Result                     N
chromium@504382      0.162726 +- 0.0165585      21      good
chromium@504476      0.163765 +- 0.0101405      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.social.twitter system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966711968208713920


For feedback, file a bug with component Speed>Bisection

Comment 9 by primiano@chromium.org, Oct 13 2017

Owner: dmazz...@chromium.org
Status: Assigned (was: Untriaged)
Yeah the revert reland pattern matches perfectly with the shape of the regression.
This seems really https://chromium-review.googlesource.com/663208

Comment 10 by charliea@chromium.org, Nov 1 2017

Status: WontFix (was: Assigned)
Going to close this as WontFix, as it's vanishingly unlikely that a regression from 10/2 is now going to get reverted. I'll reach out to dmazzoni@ offline about what happened.

Comment 11 by dmazz...@chromium.org, Nov 1 2017

Mergedinto: 776506
Status: Duplicate (was: WontFix)
Hi - I'm pretty sure this is  bug 776506  - my change that added new accessibility benchmarks did trigger the overall performance regression, but the underlying issue was that new code added to allow individual stories to use custom flags had a bug and "leaked" those flags to other stories. This is now fixed.
►

Sign in to add a comment