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

Issue 674595 link

Starred by 1 user
Status: WontFix
Owner:
gsat...@chromium.org
Closed: Jul 2017
Cc:
leszeks@chromium.org
gsat...@chromium.org
mythria@chromium.org
adamk@chromium.org
rmcilroy@chromium.org
█ rnep...@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

2.8%-12.4% regression in v8.browsing_desktop_ignition at 424082:437889

Project Member Reported by rmcilroy@chromium.org, Dec 15 2016 
See the link to graphs below.
Project Member

Comment 4 by 42576172...@developer.gserviceaccount.com, Dec 15 2016 


===== BISECT JOB RESULTS =====
Status: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev   N   Good?
chromium@436761  95185185  34615438  21  good
chromium@436860  93118888  34708647  21  bad

Bisect job ran on: winx64intel_perf_bisect
Bug ID: 674595

Test Command: 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.pinterest v8.browsing_desktop_ignition
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:effective_size_avg/browse_media/browse_media_pinterest
Relative Change: 2.17%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1256
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993192263240769184


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5904460956565504

| 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 5 by 42576172...@developer.gserviceaccount.com, Dec 15 2016 


===== BISECT JOB RESULTS =====
Status: failed


=== Bisection aborted ===
The bisect was aborted because Infra Failure: Step('Working on revision chromium@436848,v8@1f53653e6c.install ChromePublic.apk') returned 2
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@436761  2211791  26484.1  6  good
chromium@436813  2216638  25659.7  6  good
chromium@436839  2215852  25714.2  6  good
chromium@436846  2209305  21347.5  6  good
chromium@436848  2211345  17263.5  6  good
chromium@436849  2268285  10716.9  6  bad
chromium@436852  2263796  17580.5  6  bad
chromium@436865  2260892  23807.2  6  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 674595

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.news.washingtonpost v8.browsing_mobile_ignition
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:code_and_metadata_size_max/browse_news/browse_news_washingtonpost
Relative Change: 2.22%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1877
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993192249241245408


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5821477222350848

| 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!

Comment 6 by hablich@chromium.org, Dec 20 2016

Cc: mythria@chromium.org leszeks@chromium.org
Status: Available (was: Untriaged)
Mythria@ and leszeks@, could you please take a look?
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Dec 20 2016

Cc: rnep...@chromium.org
Owner: rnep...@chromium.org

=== PERF REGRESSION ===


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

Hi rnephew@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


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


===== SUSPECTED CL(s) =====
Subject : [System Health] Make pinterest/tumblr stories scroll to elements off screen.
Author  : rnephew
Commit description:
  
The system health stories for these pages require scrolling down to interact
with different elements that start off screen.

BUG=646392,  649392 

Review-Url: https://codereview.chromium.org/2414153004
Cr-Commit-Position: refs/heads/master@{#429475}
Commit  : d26f4378989c7cb41cba9b215464c180bd46b7e7
Date    : Thu Nov 03 00:33:08 2016


===== TESTED REVISIONS =====
Revision         Mean       Std Dev   N   Good?
chromium@424080  93819936   14552745  13  good
chromium@427538  91195046   16402396  6   good
chromium@429267  93738465   10481619  6   good
chromium@429375  91972430   4818265   6   good
chromium@429429  93038676   5655275   6   good
chromium@429456  89283895   12574159  6   good
chromium@429470  90506587   8904294   6   good
chromium@429474  90421318   5450105   6   good
chromium@429475  105164686  4682625   6   bad    <--
chromium@429476  106399915  11705570  6   bad
chromium@429477  103253297  5810340   6   bad
chromium@429483  103941603  1949681   6   bad
chromium@429699  106525819  3946181   6   bad
chromium@430131  104758020  8352645   9   bad
chromium@430995  107800036  4870120   6   bad
chromium@437910  101885465  16208881  9   bad

Bisect job ran on: winx64intel_perf_bisect
Bug ID: 674595

Test Command: 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.pinterest v8.browsing_desktop_ignition
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:effective_size_avg/browse_media/browse_media_pinterest
Relative Change: 8.60%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1261
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8992743412681283536


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5310906743914496

| 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!

Comment 9 by rnep...@chromium.org, Dec 20 2016 

The browse:media:pinterest increase is to be expected since we are changing how the story works drastically.

The washington post increase shouldn't happen. I'm going to bisect on only that page and see if I can find the culprit of that.
Project Member

Comment 11 by 42576172...@developer.gserviceaccount.com, Dec 21 2016

Cc: gsat...@chromium.org
Owner: gsat...@chromium.org

=== PERF REGRESSION ===


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

Hi gsathya@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


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


===== SUSPECTED CL(s) =====
Subject : Reland Create JSPromise (patchset #16 id:300001 of https://codereview.chromium.org/2536463002/ )"
Author  : gsathya
Commit description:
  
This reverts commit 4c7cccf9f40c063e2d6410fc35664a0efd1acb46.

BUG= v8:5343 

Review-Url: https://codereview.chromium.org/2554943002
Cr-Commit-Position: refs/heads/master@{#41534}
Commit  : 52e2c154ff18ae0e7ca807a5ad539750297c31b3
Date    : Tue Dec 06 18:43:18 2016


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev  N  Good?
chromium@436759                2214030  23702.3  6  good
chromium@436813                2207343  14057.8  6  good
chromium@436840                2214267  21342.3  6  good
chromium@436847                2217775  15335.0  6  good
chromium@436848                2217941  11033.1  6  good
chromium@436848,v8@900d1982cd  2210698  21963.8  6  good
chromium@436848,v8@52e2c154ff  2287629  88429.9  6  bad    <--
chromium@436848,v8@1f53653e6c  2264959  10784.2  6  bad
chromium@436849                2265023  16155.2  6  bad
chromium@436850                2275961  48021.4  6  bad
chromium@436853                2267315  6485.53  6  bad
chromium@436866                2262497  14849.4  6  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 674595

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.news.washingtonpost v8.browsing_mobile_ignition
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:code_and_metadata_size_max/browse_news/browse_news_washingtonpost
Relative Change: 2.19%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1893
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8992722788014504736


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5874817025179648

| 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!

Comment 12 by gsat...@chromium.org, Dec 21 2016

Cc: adamk@chromium.org
Certain code paths in Promises regressed as part of the piecemeal port to TF. The porting should be done by end of next week, so we'll know if these go away. 

Meanwhile, I'll try to reproduce this locally.

Comment 13 by benhenry@google.com, Jul 14 2017

Status: WontFix (was: Available)
This regression is too old to be actionable.
►

Sign in to add a comment