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

Issue 777799 link

Starred by 2 users
Status: Duplicate
Merged: issue 778680
Owner: ----
Closed: Oct 2017
Cc:
simonhatch@chromium.org
█ kraynov@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

15.9% regression in system_health.common_desktop at 510613:510668

Project Member Reported by kraynov@chromium.org, Oct 24 2017 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Oct 24 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=777799

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=6596823d443ef5e3bb84eb8263fb3a76502f743791ee7a3c0978e16dc2604f35


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

chromium-rel-win7-gpu-intel
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Oct 24 2017 


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

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : cpu_time_percentage_avg/load_accessibility_media/load_accessibility_media_wikipedia

Revision             Result                     N
chromium@510612      0.124615 +- 0.0271691      21      good
chromium@510668      0.123238 +- 0.0407962      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.accessibility.media.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/8964860974884265824


For feedback, file a bug with component Speed>Bisection

Comment 5 by benhenry@chromium.org, Oct 26 2017 

Widened bisect range.
Project Member

Comment 6 by 42576172...@developer.gserviceaccount.com, Oct 26 2017 


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

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : cpu_time_percentage_avg/load_accessibility_media/load_accessibility_media_wikipedia

Revision             Result                     N
chromium@510225      0.143183 +- 0.0576796      21      good
chromium@510668      0.147792 +- 0.0304078      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.accessibility.media.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/8964633984767421616


For feedback, file a bug with component Speed>Bisection

Comment 7 by benhenry@chromium.org, Oct 30 2017

Cc: simonhatch@chromium.org
Hey Simon - what options do I have?

Comment 10 by simonhatch@chromium.org, Oct 30 2017 

For bisect? You widened the range, but didn't move the end commit. The alert point is a bit suspect, because there have been points earlier that were just as high. I kicked off 2, both with wider ranges, one with no story filter.

There's no reference build unfortunately, so hard to say if this is a real regression or not. Bot seems to be the same. If those bieects can't repro anything, might need to add the owners and start looking at the traces from the runs to see if there's any clue as to what regressed.
Project Member

Comment 11 by 42576172...@developer.gserviceaccount.com, Oct 31 2017

Mergedinto: 778680
Status: Duplicate (was: Untriaged)

=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Zhenyao Mo
  Commit : 0728a85617eae06098373349af80c28ce26b60fd
  Date   : Tue Oct 24 11:16:56 2017
  Subject: One step further into moving GPU feature decisions to GPU process

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : cpu_time_percentage_avg/load_accessibility_media/load_accessibility_media_wikipedia
  Change       : 33.17% | 0.122259420053 -> 0.0817020080253

Revision             Result                       N
chromium@510255      0.122259 +- 0.0126782        6      good
chromium@510699      0.123803 +- 0.0186317        6      good
chromium@510921      0.124588 +- 0.009974         6      good
chromium@511032      0.129519 +- 0.00610993       6      good
chromium@511088      0.1271 +- 0.0133112          6      good
chromium@511089      0.0817356 +- 0.0047637       6      bad       <--
chromium@511090      0.0825577 +- 0.00469825      6      bad
chromium@511092      0.0815656 +- 0.00432509      6      bad
chromium@511095      0.0815157 +- 0.00208991      6      bad
chromium@511102      0.0829184 +- 0.00521251      6      bad
chromium@511116      0.0823242 +- 0.00489476      6      bad
chromium@511143      0.081702 +- 0.00295939       6      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.accessibility.media.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/8964272364457617936


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

Comment 12 by 42576172...@developer.gserviceaccount.com, Oct 31 2017 


=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Error: INFRA_FAILURE

The bisect was able to narrow the range, you can try running with:
  good_revision: 01dfd4b9c14dabc48b22262afccdeebbf65aeeba
  bad_revision : 2701b19a9077c26a81f69af5768687951225daf4

If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : cpu_time_percentage_avg/load_accessibility_media/load_accessibility_media_wikipedia

Revision             Result                       N
chromium@510158      0.12563 +- 0.0105605         6      good
chromium@510651      0.127939 +- 0.00354606       6      good
chromium@511143      0.0822637 +- 0.00390896      6      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 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/8964272252384365984


For feedback, file a bug with component Speed>Bisection
►

Sign in to add a comment