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

Issue 719371 link

Starred by 1 user
Status: WontFix
Owner:
█ hpayer@chromium.org
Last visit > 30 days ago
Closed: Jul 2017
Cc:
█ hpayer@chromium.org
u...@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

46.9%-168.6% regression in v8.runtimestats.browsing_desktop_classic at 469124:469822

Project Member Reported by ulan@google.com, May 8 2017 
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=719371

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyqqx_ggM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgytzHpgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgiqjT0ggM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgypWLpQsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyuyZtgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgitWWiAkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgypW6sAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyui2uAkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyriFogoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgytypqgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyuqKpAsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyvGbvgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgiuPEuwkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgivW8tgsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyoqJqQsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyryQtwsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgiui7gAsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgysDQ_wgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyryzqQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgiqzMzwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyoTkqQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgiqOrswoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyrbOtwsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg8ovyggoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyqqxvgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgis6q8goM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgiv-htQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgis6q8ggM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgisnaqAkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgyrievQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgiuPSpgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg8tmXqggM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgivXRvgoM


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

chromium-rel-mac-retina
chromium-rel-mac11
chromium-rel-mac11-pro
linux-release
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, May 8 2017 


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

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


Bisect Details
  Configuration: linux_perf_bisect
  Benchmark    : v8.runtimestats.browsing_desktop_classic
  Metric       : v8-gc-memory-mark-compactor_sum/browse_media/browse_media_youtube


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.media.youtube v8.runtimestats.browsing_desktop_classic

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980176434926976720

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6351777496039424


| 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 5 by 42576172...@developer.gserviceaccount.com, May 8 2017 


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

Error: BUILD_FAILURE

The bisect was able to narrow the range, you can try running with:
  good_revision: 25959abf7948b7caac1ec844ed332b5fc4aa7ebb
  bad_revision : 8833af23e9788c725b0887446915e4820383a385

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


Bisect Details
  Configuration: linux_perf_bisect
  Benchmark    : v8.runtimestats.browsing_desktop_classic
  Metric       : v8-gc-memory-mark-compactor_sum/browse_media/browse_media_youtube
  Change       : 161.87% | 19.301 -> 50.5441666667

Revision                           Result                  N
chromium@469601                    19.301 +- 4.44851       6      good
chromium@469647                    21.5272 +- 5.74187      6      good
chromium@469653                    23.8143 +- 3.70657      6      good
chromium@469654                    22.5323 +- 3.94598      6      good
chromium@469654,v8@25959abf79      20.1482 +- 5.27358      6      good
chromium@469654,v8@8833af23e9      54.7903 +- 11.5852      6      bad
chromium@469655                    52.1385 +- 14.8653      6      bad
chromium@469656                    50.4327 +- 7.8512       6      bad
chromium@469659                    52.0125 +- 19.2217      6      bad
chromium@469670                    53.2207 +- 4.52249      6      bad
chromium@469692                    50.5442 +- 11.2879      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.media.youtube v8.runtimestats.browsing_desktop_classic

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980175224380171760

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6351777496039424


| 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 6 by u...@chromium.org, May 12 2017

Labels: Performance-Sheriff-V8
Project Member

Comment 7 by sheriffbot@chromium.org, Jul 17 2017

Labels: Hotlist-Google
Project Member

Comment 10 by 42576172...@developer.gserviceaccount.com, Jul 24 2017

Cc: hpayer@chromium.org
Owner: hpayer@chromium.org

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

Hi hpayer@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 : hpayer
  Commit : 8833af23e9788c725b0887446915e4820383a385
  Date   : Fri May 05 11:24:02 2017
  Subject: [heap] Memory reducer calls GC with kGCCallbackFlagCollectAllExternalMemory flag.

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : v8.browsing_desktop
  Metric       : v8-gc-total_max/browse_media/browse_media_youtube
  Change       : 94.05% | 6.06383333333 -> 11.7666666667

Revision                           Result                  N
chromium@469626                    6.06383 +- 1.01764      6      good
chromium@469644                    7.04 +- 4.04943         6      good
chromium@469653                    6.34067 +- 1.43235      6      good
chromium@469654                    6.36217 +- 3.12005      6      good
chromium@469654,v8@25959abf79      6.45567 +- 2.00675      6      good
chromium@469654,v8@b7b3781b87      6.63467 +- 2.08178      6      good
chromium@469654,v8@8833af23e9      11.3782 +- 3.35327      6      bad       <--
chromium@469655                    11.7198 +- 2.19266      6      bad
chromium@469657                    12.0867 +- 1.54767      6      bad
chromium@469661                    11.959 +- 2.62418       6      bad
chromium@469695                    11.7667 +- 3.15327      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.youtube 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/8973174419329121680


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

Comment 11 by 42576172...@developer.gserviceaccount.com, Jul 24 2017 


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

Suspected Commit
  Author : hpayer
  Commit : 8833af23e9788c725b0887446915e4820383a385
  Date   : Fri May 05 11:24:02 2017
  Subject: [heap] Memory reducer calls GC with kGCCallbackFlagCollectAllExternalMemory flag.

Bisect Details
  Configuration: mac_retina_perf_bisect
  Benchmark    : v8.runtimestats.browsing_desktop_classic
  Metric       : v8-gc-memory-mark-compactor_avg/browse_media/browse_media_youtube
  Change       : 61.24% | 6.829 -> 11.0112166667

Revision                           Result                   N
chromium@469635                    6.829 +- 1.95757         6      good
chromium@469648                    6.70156 +- 1.89128       6      good
chromium@469652                    6.70278 +- 0.828727      6      good
chromium@469654                    6.46931 +- 1.02664       6      good
chromium@469654,v8@25959abf79      6.19355 +- 1.29106       5      good
chromium@469654,v8@b7b3781b87      5.77087 +- 0.768329      6      good
chromium@469654,v8@8833af23e9      10.4876 +- 2.10612       6      bad       <--
chromium@469655                    10.1176 +- 1.48042       6      bad
chromium@469661                    10.3528 +- 1.6996        6      bad
chromium@469686                    10.3468 +- 0.991007      6      bad
chromium@469736                    11.0112 +- 2.1548        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.media.youtube v8.runtimestats.browsing_desktop_classic

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8973174213989546176


For feedback, file a bug with component Speed>Bisection

Comment 12 by u...@chromium.org, Jul 25 2017

Status: WontFix (was: Assigned)
The regression is expected tradeoff between memory improvement and GC time increase
►

Sign in to add a comment