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

Issue 719351 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner:
Last visit > 30 days ago
Closed: Jul 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

37.4% regression in v8.browsing_desktop at 468481:468573

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

Issue description

See the link to graphs below.
 

Comment 1 by u...@chromium.org, May 8 2017

 Issue 719352  has been merged into this issue.

Comment 2 by u...@chromium.org, May 8 2017

 Issue 719353  has been merged into this issue.

Comment 3 by u...@chromium.org, May 8 2017

 Issue 719350  has been merged into this issue.

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
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Jul 24 2017

Cc: avayvod@chromium.org
Owner: avayvod@chromium.org

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

Hi avayvod@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 : avayvod
  Commit : 27e76cce52a397ea93c9bcf38454c2b4013b7db9
  Date   : Tue May 02 01:23:30 2017
  Subject: [Media, Controls] Use a weak ref to the listener from callback.

Bisect Details
  Configuration: mac_pro_perf_bisect
  Benchmark    : v8.browsing_desktop
  Metric       : v8-gc-full-mark-compactor_avg/browse_news/browse_news_reddit
  Change       : 21.81% | 20.11195 -> 24.4980349206

Revision             Result                  N
chromium@468431      20.112 +- 1.27062       6      good
chromium@468493      20.1336 +- 1.43666      6      good
chromium@468524      20.308 +- 1.71204       6      good
chromium@468525      24.0137 +- 4.67681      6      bad       <--
chromium@468526      23.0425 +- 2.28832      6      bad
chromium@468528      23.455 +- 3.42572       6      bad
chromium@468532      24.1013 +- 3.71169      6      bad
chromium@468539      24.1175 +- 4.51858      6      bad
chromium@468554      24.498 +- 1.60822       6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.news.reddit v8.browsing_desktop

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8973173212111044144


For feedback, file a bug with component Speed>Bisection
Cc: x...@chromium.org m...@chromium.org mlamouri@chromium.org
Hm, I can understand Android regressions, but not desktop regressions. The cl shouldn't change anything on desktop.
Cc: -m...@chromium.org haraken@chromium.org
Hold on though, that's a different old cl that actually prevents a memory leak.

Comment 10 by u...@chromium.org, Jul 25 2017

Status: WontFix (was: Assigned)
Since this fixes memory leak, we have to take the GC performance regression.

Sign in to add a comment