New issue
Advanced search Search tips

Issue 721279 link

Starred by 2 users

Issue metadata

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



Sign in to add a comment

6% regression in memory.long_running_idle_gmail_tbmv2 at 470487:470517

Project Member Reported by hjd@google.com, May 11 2017

Issue description

See the link to graphs below.
 

Comment 1 by hjd@google.com, May 11 2017

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=721279

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgqqHIoAoM


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

android-nexus6
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, May 11 2017

Mergedinto: 721238
Status: Duplicate (was: Untriaged)

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

Suspected Commit
  Author : yhirano
  Commit : 91088a195969f5af61f404c9f02edc622cbe373a
  Date   : Wed May 10 05:49:31 2017
  Subject: Add Field Trial Testing Configuration for MojoLoading

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : memory.long_running_idle_gmail_tbmv2
  Metric       : memory:chrome:renderer_processes:reported_by_os:system_memory:private_dirty_size_max/memory:chrome:renderer_processes:reported_by_os:system_memory:private_dirty_size_max
  Change       : 5.01% | 133633601.333 -> 140332120.667

Revision             Result                    N
chromium@470486      133633601 +- 2223228      6      good
chromium@470488      134120505 +- 1970044      6      good
chromium@470489      133538755 +- 4031819      6      good
chromium@470490      141050377 +- 4439259      6      bad       <--
chromium@470494      142223722 +- 2606123      6      bad
chromium@470502      141072286 +- 8383894      6      bad
chromium@470517      140332121 +- 4018632      6      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 memory.long_running_idle_gmail_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8979900934846253328

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5792836424826880


| 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!
Cc: tzik@chromium.org
Components: Blink>Loader
Owner: yhirano@chromium.org
Status: Assigned (was: Duplicate)
Status: Duplicate (was: Assigned)
Status: Assigned (was: Duplicate)
Project Member

Comment 7 by bugdroid1@chromium.org, May 12 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/823d5c738187627cbe218f01ddac6f383bcaaba8

commit 823d5c738187627cbe218f01ddac6f383bcaaba8
Author: Yutaka Hirano <yhirano@chromium.org>
Date: Fri May 12 09:29:18 2017

Set URLResponseBodyConsumer::kMaxNumConsumedBytesInTask to 32K

This is a speculative fix for a perf regression.

Bug:  721279 
Change-Id: Ic91648ab18bb95bca8bf2c40372a72ebed3c98d3
Reviewed-on: https://chromium-review.googlesource.com/503951
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Yutaka Hirano <yhirano@chromium.org>
Cr-Commit-Position: refs/heads/master@{#471257}
[modify] https://crrev.com/823d5c738187627cbe218f01ddac6f383bcaaba8/content/child/url_response_body_consumer.h

Project Member

Comment 8 by bugdroid1@chromium.org, May 15 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f20d10a747ef558a18cdd6925df977352e477997

commit f20d10a747ef558a18cdd6925df977352e477997
Author: Yutaka Hirano <yhirano@chromium.org>
Date: Mon May 15 04:44:45 2017

Revert "Set URLResponseBodyConsumer::kMaxNumConsumedBytesInTask to 32K"

This reverts commit 823d5c738187627cbe218f01ddac6f383bcaaba8.

Reason for revert: This change worsen the perf regression.

Original change's description:
> Set URLResponseBodyConsumer::kMaxNumConsumedBytesInTask to 32K
> 
> This is a speculative fix for a perf regression.
> 
> Bug:  721279 
> Change-Id: Ic91648ab18bb95bca8bf2c40372a72ebed3c98d3
> Reviewed-on: https://chromium-review.googlesource.com/503951
> Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
> Commit-Queue: Yutaka Hirano <yhirano@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#471257}

TBR=kinuko@chromium.org,yhirano@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug:  721279 

Change-Id: I2607316f6ce47d2dcdf9bc583d7e6fb9e32f93d0
Reviewed-on: https://chromium-review.googlesource.com/505649
Reviewed-by: Yutaka Hirano <yhirano@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Yutaka Hirano <yhirano@chromium.org>
Cr-Commit-Position: refs/heads/master@{#471670}
[modify] https://crrev.com/f20d10a747ef558a18cdd6925df977352e477997/content/child/url_response_body_consumer.h

Comment 9 by jochen@chromium.org, May 22 2017

Cc: jochen@chromium.org
 Issue 725010  has been merged into this issue.
Project Member

Comment 10 by sheriffbot@chromium.org, Jul 17 2017

Labels: Hotlist-Google
Status: Duplicate (was: Assigned)

Sign in to add a comment