New issue
Advanced search Search tips

Issue 691547 link

Starred by 1 user
Status: Duplicate
Merged: issue 691033
Owner: ----
Closed: Feb 2017
Cc:
jgruber@google.com
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

15.3%-124.6% regression in v8.browsing_desktop at 449351:449546

Project Member Reported by jgruber@google.com, Feb 13 2017 
See the link to graphs below.
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Feb 13 2017

Mergedinto: 691033
Status: Duplicate (was: Untriaged)

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

Suspected Commit
  Author : japhet
  Commit : ffefd4173f178d76449fc51f5dc22bfd7e120e03
  Date   : Thu Feb 09 22:25:56 2017
  Subject: Actually handle the Vary header in blink, rather than punting on sight.

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : v8.browsing_desktop
  Metric       : v8-gc-latency-mark-compactor_sum/browse_media/browse_media_pinterest
  Change       : 161.85% | 31.752 -> 83.1433333333

Revision             Result                  N
chromium@449434      31.752 +- 9.53065       6      good
chromium@449436      31.7413 +- 6.86893      6      good
chromium@449437      32.5515 +- 12.278       6      good
chromium@449438      77.4503 +- 21.2693      6      bad       <--
chromium@449441      61.6985 +- 39.6633      6      bad
chromium@449448      80.7828 +- 20.6367      6      bad
chromium@449462      68.5453 +- 15.4217      6      bad
chromium@449490      77.622 +- 26.3358       6      bad
chromium@449546      83.1433 +- 26.2916      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=browse.media.pinterest v8.browsing_desktop

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987770732405846240

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6115269782536192


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!
Project Member

Comment 6 by 42576172...@developer.gserviceaccount.com, Feb 13 2017 


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

Suspected Commit
  Author : japhet
  Commit : ffefd4173f178d76449fc51f5dc22bfd7e120e03
  Date   : Thu Feb 09 22:25:56 2017
  Subject: Actually handle the Vary header in blink, rather than punting on sight.

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : v8.browsing_desktop
  Metric       : v8-gc-latency-mark-compactor_avg/browse_media/browse_media_pinterest
  Change       : 105.71% | 9.62925 -> 19.8085277778

Revision             Result                  N
chromium@449402      9.62925 +- 1.46205      6      good
chromium@449433      8.4485 +- 2.99447       6      good
chromium@449437      8.60849 +- 1.49547      9      good
chromium@449438      20.1749 +- 4.5376       6      bad       <--
chromium@449439      17.6569 +- 15.2485      9      bad
chromium@449441      19.7479 +- 5.2997       6      bad
chromium@449448      24.396 +- 13.2821       6      bad
chromium@449463      20.2862 +- 14.407       6      bad
chromium@449524      19.8085 +- 6.16056      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=browse.media.pinterest v8.browsing_desktop

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987770790405872400

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5711690764648448


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Feb 13 2017 


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

Suspected Commit
  Author : japhet
  Commit : ffefd4173f178d76449fc51f5dc22bfd7e120e03
  Date   : Thu Feb 09 22:25:56 2017
  Subject: Actually handle the Vary header in blink, rather than punting on sight.

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : v8.browsing_desktop
  Metric       : v8-gc-scavenger_sum/browse_media/browse_media_pinterest
  Change       : 13.79% | 142.775888889 -> 160.688666667

Revision             Result                  N
chromium@449434      142.776 +- 20.3252      9      good
chromium@449436      136.789 +- 9.23754      6      good
chromium@449437      137.208 +- 8.3417       6      good
chromium@449438      161.256 +- 14.7754      6      bad       <--
chromium@449441      167.029 +- 13.4439      6      bad
chromium@449448      163.407 +- 27.9287      9      bad
chromium@449462      161.653 +- 11.8336      6      bad
chromium@449490      161.802 +- 30.7193      9      bad
chromium@449546      160.689 +- 9.08748      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=browse.media.pinterest v8.browsing_desktop

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987770740989722512

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5849111330291712


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!
►

Sign in to add a comment