New issue
Advanced search Search tips

Issue 672865 link

Starred by 1 user
Status: Fixed
Owner:
u...@chromium.org
Closed: Feb 2017
Cc:
simonhatch@chromium.org
majidvp@chromium.org
u...@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression

Blocked on:
issue 672529


Sign in to add a comment

82.3%-445.3% regression in v8.infinite_scroll_tbmv2 at 437222:437244

Project Member Reported by majidvp@google.com, Dec 9 2016 
See the link to graphs below.
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Dec 9 2016 


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


=== Bisection aborted ===
The bisect was aborted because Infra Failure: Step('Working on revision chromium@437229,v8@93b87c89f2.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@437229  22.5615  33.4707  20  good
chromium@437230  69.6307  201.231  20  bad
chromium@437231  78.0322  216.956  20  bad
chromium@437233  72.5648  206.904  20  bad

Bisect job ran on: android_nexus6_perf_bisect
Bug ID: 672865

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests v8.infinite_scroll-ignition_tbmv2
Test Metric: v8-gc-latency-mark-compactor_max/v8-gc-latency-mark-compactor_max
Relative Change: 221.63%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2816
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993738634272698640


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4511458660974592

| 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 4 by sullivan@chromium.org, Dec 9 2016

Cc: simonhatch@chromium.org
+simonhatch, looks like this didn't descend into the v8 roll at r437230

Comment 5 by simonhatch@chromium.org, Dec 9 2016

Blockedon: 672529
Think this is https://bugs.chromium.org/p/chromium/issues/detail?id=672529, the builders are generating invalid builds.
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Dec 9 2016 


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


=== Bisection aborted ===
The bisect was aborted because Infra Failure: Step('Working on revision chromium@437229,v8@93b87c89f2.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@437229  23.0596  36.1958  24  good
chromium@437230  75.1566  244.722  24  bad
chromium@437231  72.4998  225.171  24  bad
chromium@437233  78.1396  254.926  24  bad

Bisect job ran on: android_nexus6_perf_bisect
Bug ID: 672865

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests v8.infinite_scroll-ignition_tbmv2
Test Metric: v8-gc-latency-mark-compactor_max/v8-gc-latency-mark-compactor_max
Relative Change: 238.86%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2818
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993738634272698640


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4511458660974592

| 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 9 by 42576172...@developer.gserviceaccount.com, Dec 16 2016 


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


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

===== TESTED REVISIONS =====
Revision                       Mean     Std Dev  N   Good?
chromium@437229                25.0913  41.3145  24  good
chromium@437229,v8@ebe9419228  32.8689  93.4887  24  unknown
chromium@437229,v8@ebb1aff1e6  65.7846  217.382  24  bad
chromium@437229,v8@93b87c89f2  73.2608  252.241  24  bad
chromium@437230                68.6917  200.741  24  bad

Bisect job ran on: android_nexus6_perf_bisect
Bug ID: 672865

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests v8.infinite_scroll-ignition_tbmv2
Test Metric: v8-gc-latency-mark-compactor_max/v8-gc-latency-mark-compactor_max
Relative Change: 173.77%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2833
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993101199135534080


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5778970333675520

| 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 11 by 42576172...@developer.gserviceaccount.com, Jan 10 2017 


=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : v8.infinite_scroll_tbmv2
  Metric       : v8-gc-latency-mark-compactor_max/v8-gc-latency-mark-compactor_max

Revision             Result                  N
chromium@437222      6.88303 +- 48.6812      30      good
chromium@437235      14.5441 +- 98.9811      30      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests v8.infinite_scroll_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8993738634272698640

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4511458660974592


| 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 13 by 42576172...@developer.gserviceaccount.com, Jan 18 2017

Cc: u...@chromium.org
Owner: u...@chromium.org

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

Hi ulan@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 : ulan
  Commit : ebb1aff1e62728848377a0fefc35f1f247195751
  Date   : Wed Dec 07 16:33:18 2016
  Subject: [heap] Do not delay mark-compact in Heap::CollectGarbage if incremental

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : v8.infinite_scroll-ignition_tbmv2
  Metric       : v8-gc-latency-mark-compactor_avg/v8-gc-latency-mark-compactor_avg
  Change       : 96.26% | 15.8369472222 -> 31.0816361111

Revision                           Result                  N
chromium@437228                    15.8369 +- 3.09362      6      good
chromium@437229                    16.1222 +- 3.19456      6      good
chromium@437229,v8@ebe9419228      18.2943 +- 5.1647       6      good
chromium@437229,v8@582cdddeb4      18.0491 +- 7.46399      9      good
chromium@437229,v8@ebb1aff1e6      29.4826 +- 15.4823      6      bad       <--
chromium@437229,v8@93b87c89f2      34.2825 +- 22.445       6      bad
chromium@437230                    32.7181 +- 10.4665      6      bad
chromium@437232                    29.8539 +- 13.9609      6      bad
chromium@437235                    31.4204 +- 7.10639      6      bad
chromium@437241                    31.0816 +- 9.1701       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 v8.infinite_scroll-ignition_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8993738634272698640

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4511458660974592


| 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 14 by chrishtr@chromium.org, Jan 23 2017

Components: Blink>JavaScript

Comment 15 by hablich@chromium.org, Jan 30 2017

Status: Assigned (was: Untriaged)

Comment 16 by hablich@chromium.org, Jan 30 2017 

Not sure if this impact was expected.

Comment 17 by alexclarke@chromium.org, Feb 17 2017

Status: Fixed (was: Assigned)
This metric looks like it's recovered.
►

Sign in to add a comment