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

Issue 783772 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Nov 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

75.2% improvement in loading.mobile at 514955:515069

Project Member Reported by alexclarke@chromium.org, Nov 10 2017

Issue description

Suspiciously large improvement, did something break?
 
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Nov 10 2017

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=783772

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=9f50ef967324476694115a29b8407cc1135ae34b7ccaa9e2bbbf9297fd1f8074


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

android-nexus7v2
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Nov 10 2017

Cc: crouleau@chromium.org
Owner: crouleau@chromium.org
Status: Assigned (was: Untriaged)

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

Hi crouleau@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 : Caleb Rouleau
  Commit : 5daa21ce99f8f70abe1202689a510d9f03e5a3e1
  Date   : Wed Nov 08 23:25:07 2017
  Subject: Have pages edit grouping keys themselves.

Bisect Details
  Configuration: android_nexus7_perf_bisect
  Benchmark    : loading.mobile
  Metric       : timeToFirstMeaningfulPaint_avg/Regular-3G/Dawn
  Change       : 58.92% | 3714.0755 -> 1525.69583333

Revision             Result                  N
chromium@514954      3714.08 +- 2671.63      6      good
chromium@514983      3189.26 +- 969.065      6      good
chromium@514998      3153.96 +- 578.422      6      good
chromium@514999      1399.24 +- 210.666      6      bad       <--
chromium@515000      1507.14 +- 245.611      6      bad
chromium@515002      1401.78 +- 232.047      6      bad
chromium@515005      1437.1 +- 249.836       6      bad
chromium@515012      1459.01 +- 243.21       6      bad
chromium@515069      1525.7 +- 225.411       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 --story-filter=Dawn loading.mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8963304428513199360


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

Comment 4 by 42576172...@developer.gserviceaccount.com, Nov 10 2017

 Issue 783773  has been merged into this issue.
Cc: ksakamoto@chromium.org nednguyen@chromium.org kouhei@chromium.org
Not sure why this is happening. My change should not have had any impact on the values. I am investigating.

+cc my reviewers.
I forgot to pass in traffic_setting into Page since I took it out of kwargs: https://cs.chromium.org/chromium/src/tools/perf/page_sets/page_cycler_story.py?sq=package:chromium&dr=C&l=22
Oops. Can you revert & reland?
Cc: -nednguyen@chromium.org nedngu...@google.com
Problem with revert and reland is that I would have to revert the catapult CL. I'll just fix this
To test my fix, I am running locally. Before fix I ran all the dawn stories (using $ tools/perf/run_benchmark loading.mobile --browser=android-chrome-dev --story-filter=Dawn ) and got this:
avg: 802.606 ms
max: 878.540 ms
min: 740.572 ms
std: 57.683 ms

The results were not bimodal even though some of them are with traffic_setting set and some are with traffic_setting being NONE.

After my fix:
avg: 2,492.178 ms
max: 4,335.327 ms
min: 804.854 ms
std: 1,944.113 ms

And the graph shows bimodal results. It looks like the correct results again.

Change is at https://chromium-review.googlesource.com/#/c/chromium/src/+/764489

Sorry for breaking this!

Project Member

Comment 11 by bugdroid1@chromium.org, Nov 10 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/46cf2b51bb3b56831ded3ffb60400f51091400ea

commit 46cf2b51bb3b56831ded3ffb60400f51091400ea
Author: Caleb Rouleau <crouleau@chromium.org>
Date: Fri Nov 10 22:23:42 2017

[Telemetry] give traffic_setting to parent. Fix Bug.

To test my fix, I am running locally. Before fix I ran all the dawn stories
(using $ tools/perf/run_benchmark loading.mobile --browser=android-chrome-dev
 --story-filter=Dawn ) and got this:
avg: 802.606 ms
max: 878.540 ms
min: 740.572 ms
std: 57.683 ms

The results were not bimodal even though some of them are with traffic_setting
 set and some are with traffic_setting being NONE.

After my fix:
avg: 2,492.178 ms
max: 4,335.327 ms
min: 804.854 ms
std: 1,944.113 ms

And the graph shows bimodal results. It looks like the correct results again.


Bug:  783772 
Change-Id: I5249125a29923f2a72dd22f203e5fbfe5f6ba340
Reviewed-on: https://chromium-review.googlesource.com/764489
Commit-Queue: Caleb Rouleau <crouleau@chromium.org>
Reviewed-by: Ned Nguyen <nednguyen@google.com>
Cr-Commit-Position: refs/heads/master@{#515714}
[modify] https://crrev.com/46cf2b51bb3b56831ded3ffb60400f51091400ea/tools/perf/page_sets/page_cycler_story.py

Comment 12 Deleted

Project Member

Comment 13 by 42576172...@developer.gserviceaccount.com, Nov 11 2017

 Issue 783620  has been merged into this issue.
Project Member

Comment 14 by 42576172...@developer.gserviceaccount.com, Nov 11 2017

 Issue 783739  has been merged into this issue.
Project Member

Comment 15 by 42576172...@developer.gserviceaccount.com, Nov 11 2017

 Issue 783906  has been merged into this issue.
Status: Fixed (was: Assigned)

Sign in to add a comment