New issue
Advanced search Search tips

Issue 719360 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Jul 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

51% regression in v8.browsing_desktop at 468947:469011

Project Member Reported by ulan@google.com, May 8 2017

Issue description

See the link to graphs below.
 

Comment 1 by ulan@google.com, May 8 2017

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=719360

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyrSWugoM


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

chromium-rel-win10

=== Auto-CCing suspected CL author ulan@chromium.org ===

Hi ulan@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


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

Suspected Commit
  Author : ulan
  Commit : c6816cd87d1d31dd78218a5f004d6db6202fb1fc
  Date   : Tue May 02 17:03:31 2017
  Subject: [heap] Implement simple concurrent marking deque.

Bisect Details
  Configuration: winx64_10_perf_bisect
  Benchmark    : v8.browsing_desktop
  Metric       : v8-gc-incremental-step_sum/browse_search/browse_search_google_india
  Change       : 77.09% | 22.6325 -> 40.0803333333

Revision                           Result                  N
chromium@468946                    22.6325 +- 10.4262      6      good
chromium@468963                    20.4718 +- 2.18655      6      good
chromium@468971                    20.8407 +- 1.03305      6      good
chromium@468973                    20.2317 +- 1.65494      6      good
chromium@468973,v8@63a40cae7c      20.8325 +- 1.76438      6      good
chromium@468973,v8@66f6954064      20.239 +- 2.24304       6      good
chromium@468973,v8@c6816cd87d      40.3792 +- 1.72622      6      bad       <--
chromium@468973,v8@e393093a74      40.9935 +- 1.73723      6      bad
chromium@468973,v8@94ca3b6809      40.2068 +- 1.91343      6      bad
chromium@468974                    40.7923 +- 3.45632      6      bad
chromium@468975                    40.0013 +- 1.25467      6      bad
chromium@468979                    40.9877 +- 3.23786      6      bad
chromium@469011                    40.0803 +- 1.47279      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.search.google.india v8.browsing_desktop

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980177365780048768

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6301410280341504


| 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!

Comment 4 by u...@chromium.org, May 12 2017

Labels: Performance-Sheriff-V8
Project Member

Comment 5 by sheriffbot@chromium.org, Jul 17 2017

Labels: Hotlist-Google

Comment 6 by u...@chromium.org, Jul 24 2017

Status: Fixed (was: Assigned)
The bug in CL #3 (accidentally enabling concurrent marking queue) was fixed.

Sign in to add a comment