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

Issue 636368 link

Starred by 1 user
Status: Fixed
Owner:
█ hpayer@chromium.org
Last visit > 30 days ago
Closed: Aug 2016
Cc:
mlippautz@chromium.org
█ hpayer@chromium.org
u...@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

4.6%-7% regression in smoothness.top_25_smooth at 410714:410775

Project Member Reported by pmeenan@chromium.org, Aug 10 2016 
See the link to graphs below.

Comment 1 by pmeenan@chromium.org, Aug 10 2016 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=636368

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgtsOgvwkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgtsOg_wgM


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

chromium-rel-mac-retina
chromium-rel-mac11
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Aug 10 2016 


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


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@410721  18.7375  0.148033  5  good
chromium@410731  18.7135  0.165993  5  good
chromium@410734  18.7389  0.18555   5  good
chromium@410735  19.5639  0.228373  5  bad
chromium@410736  19.7735  0.145386  5  bad
chromium@410741  19.7608  0.309467  5  bad
chromium@410761  19.4865  0.488105  8  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 636368

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.top_25_smooth
Test Metric: frame_times/http___sports.yahoo.com_
Relative Change: 3.29%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/822
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004709762363913040


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5875658987667456

| 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, Aug 10 2016 


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


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@410721  18.7375  0.148033  5  good
chromium@410731  18.7135  0.165993  5  good
chromium@410734  18.7389  0.18555   5  good
chromium@410735  19.5639  0.228373  5  bad
chromium@410736  19.7735  0.145386  5  bad
chromium@410741  19.7608  0.309467  5  bad
chromium@410761  19.4865  0.488105  8  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 636368

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.top_25_smooth
Test Metric: frame_times/http___sports.yahoo.com_
Relative Change: 3.29%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/822
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004709762363913040


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5875658987667456

| 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 7 by pmeenan@chromium.org, Aug 10 2016 

Kicked new bisect
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Aug 11 2016

Cc: hpayer@chromium.org
Owner: hpayer@chromium.org

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

Hi hpayer@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 : [heap] Enforce finalization of marking at heap limit when optimizing for memory usage.
Author  : hpayer
Commit description:
  
BUG= chromium:634900 

Review-Url: https://codereview.chromium.org/2225363002
Cr-Commit-Position: refs/heads/master@{#38495}
Commit  : a7b7d691064f00feaa37946416e65c76d6b9926e
Date    : Tue Aug 09 13:09:14 2016


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev    N  Good?
chromium@410713                18.1932  0.11753    5  good
chromium@410725                18.2489  0.134039   5  good
chromium@410731                18.1461  0.101548   5  good
chromium@410734                18.1816  0.114248   8  good
chromium@410734,v8@9b6d753170  18.2229  0.0824238  5  good
chromium@410734,v8@2f6b26ed45  18.1883  0.0931068  8  good
chromium@410734,v8@c403ab7f57  18.2297  0.108671   8  good
chromium@410734,v8@a7b7d69106  18.858   0.367982   5  bad    <--
chromium@410735                19.1338  0.448812   5  bad
chromium@410736                18.9741  0.332958   8  bad
chromium@410737                18.8215  0.335662   5  bad
chromium@410761                18.8303  0.239556   5  bad

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 636368

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.top_25_smooth
Test Metric: frame_times/http___sports.yahoo.com_
Relative Change: 3.50%
Score: 98.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1562
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004689009760206048


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6386552863719424

| 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 10 by 42576172...@developer.gserviceaccount.com, Aug 11 2016 


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


===== SUSPECTED CL(s) =====
Subject : [heap] Enforce finalization of marking at heap limit when optimizing for memory usage.
Author  : hpayer
Commit description:
  
BUG= chromium:634900 

Review-Url: https://codereview.chromium.org/2225363002
Cr-Commit-Position: refs/heads/master@{#38495}
Commit  : a7b7d691064f00feaa37946416e65c76d6b9926e
Date    : Tue Aug 09 13:09:14 2016


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev    N  Good?
chromium@410721                18.7448  0.148734   5  good
chromium@410731                18.8172  0.176718   5  good
chromium@410734                18.7239  0.0906364  5  good
chromium@410734,v8@9b6d753170  18.7194  0.0750705  8  good
chromium@410734,v8@2f6b26ed45  18.8097  0.189655   8  good
chromium@410734,v8@c403ab7f57  18.7251  0.0905622  5  good
chromium@410734,v8@a7b7d69106  20.0078  0.196435   5  bad    <--
chromium@410735                19.432   0.359708   8  bad
chromium@410736                19.7575  0.219904   5  bad
chromium@410741                19.6469  0.345061   5  bad
chromium@410761                19.5678  0.342169   5  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 636368

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.top_25_smooth
Test Metric: frame_times/http___sports.yahoo.com_
Relative Change: 4.39%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/825
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004709762363913040


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5875658987667456

| 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 11 by hpayer@chromium.org, Aug 12 2016

Status: Started (was: Assigned)
A regression on the Android One (Svelte) device is expected. However, the other devices should not be impacted. I am investigating.

Comment 12 by hpayer@chromium.org, Aug 12 2016

Cc: u...@chromium.org mlippautz@chromium.org
The memory optimization triggers because isolate()->IsIsolateInBackground() is true. That should not be true in the given benchmarks.

Comment 13 by hpayer@chromium.org, Aug 12 2016 

The signal we are getting over the API seems wrong. As soon as you navigate from one web page to the other, IsolateInBackgroundNotification is invoked. However, the tab is still used in the foreground.
Project Member

Comment 14 by bugdroid1@chromium.org, Aug 12 2016 

The following revision refers to this bug:
  https://chromium.googlesource.com/v8/v8.git/+/b992a8e1899091e3c180a050c1a63fa1e29886d5

commit b992a8e1899091e3c180a050c1a63fa1e29886d5
Author: hpayer <hpayer@chromium.org>
Date: Fri Aug 12 08:13:35 2016

Fix wrong state of IsolateInBackgroundNotification and IsolateInForegroundNotification.

BUG= chromium:636368 , chromium:635965 , chromium:634900 

Review-Url: https://codereview.chromium.org/2245483004
Cr-Commit-Position: refs/heads/master@{#38598}

[modify] https://crrev.com/b992a8e1899091e3c180a050c1a63fa1e29886d5/src/isolate.cc

Comment 15 by hpayer@chromium.org, Aug 12 2016 

With the CL in #14, all graphs should recover except Android One. There the regression is intentional.

Comment 16 by hpayer@chromium.org, Aug 17 2016

Labels: TradeOff
Status: Fixed (was: Started)
All graph recovered except Android One. There we favor memory over latency and throughput.
►

Sign in to add a comment