New issue
Advanced search Search tips

Issue 657207 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner:
Closed: Apr 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

24.6% regression in service_worker.service_worker_micro_benchmark at 425639:425661

Project Member Reported by horo@chromium.org, Oct 19 2016

Issue description

See the link to graphs below.
 

Comment 1 by horo@chromium.org, Oct 19 2016

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=657207

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg7feFsQoM


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

chromium-rel-win8-dual

Comment 3 by horo@chromium.org, Oct 19 2016

Components: Blink>ServiceWorker
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Oct 19 2016

Cc: mstarzinger@chromium.org
Owner: mstarzinger@chromium.org

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

Hi mstarzinger@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : [turbofan] Funnel class constructors through TurboFan.
Author  : mstarzinger
Commit description:
  
This makes optimization of all class constructors (i.e. both base and
derived) go through TurboFan. Note some class constructors containing
Harmony language features (e.g. super constructor calls or accesses to
the new.target value) were already going through TurboFan before.

R=bmeurer@chromium.org
BUG= v8:5458 

Review-Url: https://codereview.chromium.org/2397723002
Cr-Commit-Position: refs/heads/master@{#40342}
Commit  : 3c6bbe07e8033b25e1f64d2e46aadd586f5ed32b
Date    : Mon Oct 17 08:58:25 2016


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev   N   Good?
chromium@425638                2.45625  0.186255  8   good
chromium@425650                2.47125  0.194365  12  good
chromium@425656                2.56458  0.155906  12  good
chromium@425658                2.50688  0.265665  8   good
chromium@425658,v8@87ae05c7c4  2.46667  0.215157  12  good
chromium@425658,v8@3c6bbe07e8  2.98625  0.592614  12  bad    <--
chromium@425658,v8@9768ff47e7  2.96563  0.530784  8   bad
chromium@425658,v8@e0b4b1b63f  4.97917  6.52844   12  bad
chromium@425659                3.11062  0.394094  8   bad
chromium@425661                2.98562  0.491619  8   bad

Bisect job ran on: win_8_perf_bisect
Bug ID: 657207

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests service_worker.service_worker_micro_benchmark
Test Metric: concurrent_1_response_99_percentile/concurrent_1_response_99_percentile
Relative Change: 31.93%
Score: 95.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_8_perf_bisect/builds/2253
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8998419307747396672


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5886320822452224

| 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 Tests>AutoBisect.  Thank you!
Project Member

Comment 6 by 42576172...@developer.gserviceaccount.com, Oct 19 2016


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : [turbofan] Funnel class constructors through TurboFan.
Author  : mstarzinger
Commit description:
  
This makes optimization of all class constructors (i.e. both base and
derived) go through TurboFan. Note some class constructors containing
Harmony language features (e.g. super constructor calls or accesses to
the new.target value) were already going through TurboFan before.

R=bmeurer@chromium.org
BUG= v8:5458 

Review-Url: https://codereview.chromium.org/2397723002
Cr-Commit-Position: refs/heads/master@{#40342}
Commit  : 3c6bbe07e8033b25e1f64d2e46aadd586f5ed32b
Date    : Mon Oct 17 08:58:25 2016


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev    N   Good?
chromium@425638                2.377    0.160842   5   good
chromium@425650                2.461    0.0953546  5   good
chromium@425656                2.46125  0.157077   12  good
chromium@425658                2.4325   0.13931    8   good
chromium@425658,v8@87ae05c7c4  2.35     0.165702   8   good
chromium@425658,v8@3c6bbe07e8  3.019    0.448879   5   bad    <--
chromium@425658,v8@9768ff47e7  3.07938  0.53383    8   bad
chromium@425658,v8@e0b4b1b63f  3.166    0.626901   5   bad
chromium@425659                3.01625  0.624125   12  bad
chromium@425661                2.9825   0.550448   8   bad

Bisect job ran on: win_8_perf_bisect
Bug ID: 657207

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests service_worker.service_worker_micro_benchmark
Test Metric: concurrent_1_response_99_percentile/concurrent_1_response_99_percentile
Relative Change: 22.09%
Score: 99.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_8_perf_bisect/builds/2254
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8998419258365012624


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6381553368694784

| 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 Tests>AutoBisect.  Thank you!
Owner: mvstan...@chromium.org
We'll target improving performance of TurboFan class constructors over the next quarter.

Comment 9 by horo@chromium.org, Oct 20 2016

Status: Assigned (was: Untriaged)

Comment 10 by horo@chromium.org, Oct 20 2016

 Issue 657690  has been merged into this issue.
Status: WontFix (was: Assigned)
This has recovered.

Sign in to add a comment