New issue
Advanced search Search tips

Issue 737390 link

Starred by 1 user
Status: WontFix
Owner:
u...@chromium.org
Closed: Jul 2017
Cc:
sullivan@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

11.1%-13.7% regression in v8.mobile_infinite_scroll_tbmv2 at 481758:481869

Project Member Reported by sullivan@chromium.org, Jun 28 2017 
See the link to graphs below.

Comment 1 by sullivan@chromium.org, Jun 28 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=737390

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgrq3wuwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgjt2r3QoM


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

android-nexus6
chromium-rel-win7-dual
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Jun 28 2017 


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

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : v8.infinite_scroll_tbmv2
  Metric       : v8-gc-scavenger_avg/discourse

Revision             Result                   N
chromium@481757      2.44994 +- 0.702973      21      good
chromium@481820      2.48208 +- 0.649734      21      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=discourse v8.infinite_scroll_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8975580022361651392

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6433588108591104


| 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, Jun 28 2017 


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

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : v8.infinite_scroll_tbmv2
  Metric       : v8-gc-scavenger_avg/discourse

Revision             Result                   N
chromium@481757      1.44272 +- 0.920285      21      good
chromium@481820      1.52607 +- 1.09148       21      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=discourse v8.infinite_scroll_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8975566113944451648

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6433588108591104


| 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, Jun 28 2017 


=== BISECT JOB RESULTS ===
Perf regression found but unable to narrow commit range

Build failures prevented the bisect from narrowing the range further.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : v8.mobile_infinite_scroll_tbmv2
  Metric       : v8-gc-scavenger_avg/flickr
  Change       : 5.07% | 6.65588629202 -> 6.99341342787

Suspected Commit Range
  3 commits in range
  https://chromium.googlesource.com/v8/v8.git/+log/e65aeada27fbf147f5758efe7f7c90f48bd0cae5..1541f422a7f57e5c5eb3c39477f62a75a184f63e


Revision                           Result                   N
chromium@481835                    6.65589 +- 1.04653       14       good
chromium@481844                    6.67856 +- 1.17608       14       good
chromium@481846                    6.73597 +- 1.16037       14       good
chromium@481847                    6.73522 +- 0.471354      9        good
chromium@481847,v8@e65aeada27      6.68538 +- 0.963062      14       good
chromium@481847,v8@a37f95ff54      ---                      ---      build failure
chromium@481847,v8@116363258d      ---                      ---      build failure
chromium@481847,v8@1541f422a7      6.75354 +- 0.773847      14       bad
chromium@481848                    7.10705 +- 1.25941       14       bad
chromium@481852                    7.22816 +- 0.698703      9        bad
chromium@481869                    6.99341 +- 0.796454      14       bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=flickr v8.mobile_infinite_scroll_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8975515679445150000

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6113389169868800


| 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 8 by sullivan@chromium.org, Jun 29 2017 

Weird, all the "bad" and "build failure CLs in the v8 roll look like test/build changes to me:
https://chromium.googlesource.com/v8/v8/+log/4162e53c..a7e622ba

Maybe the bisect didn't bisect the v8 roll right? Trying again.
Project Member

Comment 10 by 42576172...@developer.gserviceaccount.com, Jun 29 2017 


=== BISECT JOB RESULTS ===
Perf regression found but unable to narrow commit range

Build failures prevented the bisect from narrowing the range further.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : v8.mobile_infinite_scroll_tbmv2
  Metric       : v8-gc-scavenger_avg/flickr
  Change       : 6.34% | 6.71090802988 -> 7.13654647785

Suspected Commit Range
  3 commits in range
  Mismatching LKGR/FKBR depots, unable to provide handy url.
  good_revision: v8@1541f422a7f57e5c5eb3c39477f62a75a184f63e
  bad_revision : chromium@f588d973c5b3e43ef304cb6a51a4b6e64c6d0aa9


Revision                           Result                   N
chromium@481835                    6.71091 +- 0.67658       14       good
chromium@481844                    6.5826 +- 0.375731       6        good
chromium@481846                    6.64938 +- 0.480285      9        good
chromium@481847                    6.72345 +- 0.646167      14       good
chromium@481847,v8@e65aeada27      6.70649 +- 0.887525      9        good
chromium@481847,v8@1541f422a7      6.74584 +- 1.02182       14       good
chromium@481847,v8@07b1113252      ---                      ---      build failure
chromium@481847,v8@a7e622ba33      ---                      ---      build failure
chromium@481848                    7.12666 +- 0.919702      14       bad
chromium@481852                    7.01213 +- 0.272219      6        bad
chromium@481869                    7.13655 +- 0.828811      9        bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=flickr v8.mobile_infinite_scroll_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8975494998673234624

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5850946669641728


| 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 11 by mustaq@chromium.org, Jul 10 2017 

Most of these bugs re v8.infinite_scroll_tbmv2 should be handled through  Issue 724262  .

Comment 12 by u...@chromium.org, Jul 11 2017

Owner: u...@chromium.org

Comment 13 by u...@chromium.org, Jul 24 2017

Status: WontFix (was: Untriaged)
The regression is dominated by the effect of 512K semi-space.
https://chromeperf.appspot.com/group_report?bug_id=737390
►

Sign in to add a comment