New issue
Advanced search Search tips

Issue 847997 link

Starred by 2 users
Status: WontFix
Owner:
perezju@chromium.org
Closed: Jul 26
Cc:
sullivan@chromium.org
fdoray@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

15.4%-15.9% regression in system_health.memory_mobile at 1526600925:1526611472

Project Member Reported by sullivan@chromium.org, May 30 2018 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, May 30 2018 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=847997

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=d29dd7e4a2d6e3e7f926d116257891bff66d0433a722e10bd0088b1573b6566a


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

health-plan-webview-low-end-phone
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, May 31 2018

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

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

Hi fdoray@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 : Francois Doray
  Commit : 0181d8b347f33ec1767ce0e244d73e97aff239ac
  Date   : Thu May 17 14:54:27 2018
  Subject: Fix TaskSchedulerWorkerPoolHistogramTest.NumTasksBeforeCleanup failure on Fuschia.

Bisect Details
  Configuration: webview-low-end-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:proportional_resident_size_avg/background_social/background_social_facebook
  Change       : 2.73% | 42246345.3333 -> 43401246.6667

Revision                                       Result                   N
android-chrome@8cea548794                      42246345 +- 1836161      6      good
android-chrome@8cea548794,chromium@559534      43406879 +- 49591.3      6      bad       <--
android-chrome@8cea548794,chromium@559535      43468489 +- 75661.5      6      bad
android-chrome@8cea548794,chromium@559536      43341001 +- 71479.1      6      bad
android-chrome@8cea548794,chromium@559539      43609119 +- 68219.2      6      bad
android-chrome@8cea548794,chromium@559545      43565087 +- 45806.9      6      bad
android-chrome@8cea548794,chromium@559557      43525663 +- 63061.8      6      bad
android-chrome@8cea548794,chromium@559580      43656052 +- 91508.6      6      bad
android-chrome@2c43d47261                      43401247 +- 32836.2      6      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=background.social.facebook system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8945072779599910240


For feedback, file a bug with component Speed>Bisection

Comment 4 by fdoray@chromium.org, Jun 1 2018

Owner: ----
Status: Untriaged (was: Assigned)
The suspected commit only touches a base_unittests file. Re-running bisect.

Comment 5 by fdoray@chromium.org, Jun 1 2018

Cc: -fdoray@chromium.org
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Jun 1 2018 


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

Bisect Details
  Configuration: webview-low-end-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:proportional_resident_size_avg/background_tools/background_tools_gmail
  Change       : 0.21% | 62157757.1429 -> 62287610.0952

Revision                       Result                   N
android-chrome@8cea548794      62157757 +- 2186782      21      good
android-chrome@2c43d47261      62287610 +- 343475       21      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=background.tools.gmail system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8944918556962798080


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

Comment 9 by 42576172...@developer.gserviceaccount.com, Jun 2 2018

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

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

Hi fdoray@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 : Francois Doray
  Commit : 6ac0dc16d307483b88be911c791d3fccfb8516f4
  Date   : Thu May 17 17:13:19 2018
  Subject: Allow SequencedTaskRunner argument in InterfacePtr::Bind().

Bisect Details
  Configuration: webview-low-end-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:proportional_resident_size_avg/background_social/background_social_facebook
  Change       : 1.12% | 43855902.6667 -> 43365455.4286

Revision                                       Result                   N
android-chrome@8cea548794                      43855903 +- 1547898      21      good
android-chrome@8cea548794,chromium@559580      45868745 +- 34856.1      6       good
android-chrome@8cea548794,chromium@559581      43875529 +- 45155.6      6       bad       <--
android-chrome@8cea548794,chromium@559582      44386676 +- 33801.3      6       bad
android-chrome@8cea548794,chromium@559583      44483444 +- 126335       6       bad
android-chrome@8cea548794,chromium@559586      44298271 +- 88127.5      6       bad
android-chrome@8cea548794,chromium@559592      44244852 +- 61681.0      6       bad
android-chrome@8cea548794,chromium@559604      44127092 +- 35594.9      6       bad
android-chrome@2c43d47261                      43365455 +- 647026       14      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=background.social.facebook system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8944918547949519504


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

Comment 13 by 42576172...@developer.gserviceaccount.com, Jun 27 2018 


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

Hi fdoray@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 : Francois Doray
  Commit : 0181d8b347f33ec1767ce0e244d73e97aff239ac
  Date   : Thu May 17 14:54:27 2018
  Subject: Fix TaskSchedulerWorkerPoolHistogramTest.NumTasksBeforeCleanup failure on Fuschia.

Bisect Details
  Configuration: webview-low-end-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:proportional_resident_size_avg/background_social/background_social_facebook
  Change       : 2.88% | 42008094.6667 -> 43219828.0

Revision                                       Result                   N
android-chrome@8cea548794                      42008095 +- 1956756      6      good
android-chrome@8cea548794,chromium@559534      45174473 +- 67694.6      6      bad       <--
android-chrome@8cea548794,chromium@559535      44124703 +- 42839.0      6      bad
android-chrome@8cea548794,chromium@559536      45546015 +- 89919.6      6      bad
android-chrome@8cea548794,chromium@559539      45504031 +- 51479.5      6      bad
android-chrome@8cea548794,chromium@559545      43787295 +- 59443.8      6      bad
android-chrome@8cea548794,chromium@559557      45358111 +- 105299       6      bad
android-chrome@8cea548794,chromium@559580      44939465 +- 62641.4      6      bad
android-chrome@2c43d47261                      43219828 +- 55131.7      6      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=background.social.facebook system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8942627671751307440


For feedback, file a bug with component Speed>Bisection
Possible duplicate of  crbug.com/847997 , but not merging issues due to multiple
culprits in destination issue.
Project Member

Comment 14 by 42576172...@developer.gserviceaccount.com, Jun 28 2018 


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

Bisect Details
  Configuration: webview-low-end-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:proportional_resident_size_avg/background_tools/background_tools_gmail
  Change       : 0.46% | 62091050.8571 -> 62377895.2

Revision                       Result                   N
android-chrome@8cea548794      62091051 +- 1750851      21      good
android-chrome@2c43d47261      62377895 +- 745104       20      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=background.tools.gmail system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8942627666459046784


For feedback, file a bug with component Speed>Bisection

Comment 15 by fdoray@chromium.org, Jul 9 

The suspected CL only touches a unit test file 0181d8b347f33ec1767ce0e244d73e97aff239ac.

I'm tempted to close with WontFix.

What would be the next steps to understand this?

Comment 16 by fdoray@chromium.org, Jul 25

Owner: sullivan@chromium.org
My CL changes a unit test file, so I highly doubt it can have an impact on Chrome memory usage. What should I do next?

Comment 17 by sullivan@chromium.org, Jul 25

Owner: perezju@chromium.org
Assigning to test owner for next steps.

Comment 18 by perezju@chromium.org, Jul 26

Status: WontFix (was: Assigned)
Yeah, looking at the graphs this looks like a regression to the mean.
►

Sign in to add a comment