New issue
Advanced search Search tips

Issue 691618 link

Starred by 1 user

Issue metadata

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

Blocked on:
issue 713747
issue 718928



Sign in to add a comment

2%-7.1% regression in system_health.memory_mobile at 448718:449661

Project Member Reported by majidvp@google.com, Feb 13 2017

Issue description

See the link to graphs below.
 
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Feb 14 2017

Cc: fdoray@chromium.org
Owner: fdoray@chromium.org

=== 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 : fdoray
  Commit : 99e56574537a8cf27af4c86f34e42fcad65e21c8
  Date   : Fri Feb 10 16:41:56 2017
  Subject: Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles
  Change       : 6.73% | 25020888.6667 -> 26484742.0

Revision             Result                   N
chromium@449621      25020889 +- 2040099      9      good
chromium@449631      25118832 +- 2884578      9      good
chromium@449632      25171321 +- 2514624      9      good
chromium@449633      26463986 +- 912915       9      bad       <--
chromium@449634      26475466 +- 849351       9      bad
chromium@449636      26419260 +- 1085255      9      bad
chromium@449641      26284567 +- 757608       9      bad
chromium@449661      26484742 +- 481487       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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987753985030334624

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5854658817425408


| 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 Speed>Bisection.  Thank you!

=== 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 : fdoray
  Commit : b83be4ca0e00b50b17619adf7f7de275455e9852
  Date   : Mon Feb 13 16:39:47 2017
  Subject: Use TaskScheduler instead of WorkerPool in v8_platform.cc.

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles
  Change       : 2.46% | 26749142.0 -> 26090996.6667

Revision             Result                   N
chromium@449948      26749142 +- 1223282      9       good
chromium@449970      26930359 +- 665071       9       good
chromium@449973      26857712 +- 1429043      14      good
chromium@449975      26878863 +- 1239035      9       good
chromium@449976      26218731 +- 726620       9       bad       <--
chromium@449981      26098203 +- 398479       6       bad
chromium@449992      26090997 +- 306896       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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8986192687224732144

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5834331567685632


| 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 Speed>Bisection.  Thank you!
Labels: Performance-Memory
Project Member

Comment 7 by bugdroid1@chromium.org, Apr 12 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/24e5d1ddee9fa61587b6f0279a25792d0208b9f1

commit 24e5d1ddee9fa61587b6f0279a25792d0208b9f1
Author: fdoray <fdoray@chromium.org>
Date: Wed Apr 12 19:58:13 2017

Revert of Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp (patchset #1 id:1 of https://codereview.chromium.org/2685673005/ )

Reason for revert:
Reverting to see the effect on memory benchmarks.
BUG= 691618 

Original issue's description:
> Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp
>
> This CL replaces base::WorkerPool::PostTask() with
> base::PostTaskWithTraits(). The following traits are used:
>
> Priority: Inherited (default)
>   The priority is inherited from the calling context (i.e. TaskTraits
>   are initialized with the priority of the current task).
>
> Shutdown behavior: CONTINUE_ON_SHUTDOWN
>   Tasks posted with this mode which have not started executing before
>   shutdown is initiated will never run. Tasks with this mode running at
>   shutdown will be ignored (the worker will not be joined).
>
>   Note: Tasks that were previously posted to base::WorkerPool should
>   use this shutdown behavior because this is how base::WorkerPool
>   handles all its tasks.
>
> *No* May Block (default):
>   Tasks may no block on synchronous file I/O operations or on sync
>   primitives.
>
> BUG= 659191 
>
> Review-Url: https://codereview.chromium.org/2685673005
> Cr-Commit-Position: refs/heads/master@{#449633}
> Committed: https://chromium.googlesource.com/chromium/src/+/99e56574537a8cf27af4c86f34e42fcad65e21c8

TBR=kinuko@chromium.org,haraken@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG= 659191 

Review-Url: https://codereview.chromium.org/2812263003
Cr-Commit-Position: refs/heads/master@{#464121}

[modify] https://crrev.com/24e5d1ddee9fa61587b6f0279a25792d0208b9f1/third_party/WebKit/Source/platform/threading/BackgroundTaskRunner.cpp
[modify] https://crrev.com/24e5d1ddee9fa61587b6f0279a25792d0208b9f1/third_party/WebKit/Source/platform/threading/DEPS

Project Member

Comment 9 by 42576172...@developer.gserviceaccount.com, Apr 25 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981426266274152112

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5842560452395008


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 16 by 42576172...@developer.gserviceaccount.com, Apr 26 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981310714284775520

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5842560452395008


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 17 by 42576172...@developer.gserviceaccount.com, Apr 27 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981249689144371360

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6333478804652032


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 19 by 42576172...@developer.gserviceaccount.com, Apr 27 2017

Cc: japhet@chromium.org
Owner: japhet@chromium.org

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

Hi japhet@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 : japhet
  Commit : ffefd4173f178d76449fc51f5dc22bfd7e120e03
  Date   : Thu Feb 09 22:25:56 2017
  Subject: Actually handle the Vary header in blink, rather than punting on sight.

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_media/load_media_9gag
  Change       : 4.77% | 35798815.1111 -> 37588539.5556

Revision             Result                   N
chromium@448717      35798815 +- 1584140      9       good
chromium@449178      35972659 +- 2218029      14      good
chromium@449408      35697439 +- 1754469      9       good
chromium@449423      36013286 +- 1606753      9       good
chromium@449431      35958787 +- 1525570      9       good
chromium@449435      36120692 +- 1703018      9       good
chromium@449437      35956170 +- 1410038      9       good
chromium@449438      37846132 +- 3558575      9       bad       <--
chromium@449523      37701391 +- 4526421      14      bad
chromium@449638      37588540 +- 3153954      9       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-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.media.9gag system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981249783504170224

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6340144057024512


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 21 by 42576172...@developer.gserviceaccount.com, Apr 28 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981250529118474368

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5816218310672384


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 23 by 42576172...@developer.gserviceaccount.com, Apr 28 2017

Owner: fdoray@chromium.org

=== 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 : fdoray
  Commit : 99e56574537a8cf27af4c86f34e42fcad65e21c8
  Date   : Fri Feb 10 16:41:56 2017
  Subject: Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp

Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_games/load_games_bubbles
  Change       : 1.57% | 41282997.7143 -> 41930084.4444

Revision             Result                   N
chromium@449627      41282998 +- 1837424      14      good
chromium@449631      41383935 +- 1809246      14      good
chromium@449632      41576252 +- 2143090      21      good
chromium@449633      42020936 +- 1480118      14      bad       <--
chromium@449635      42002943 +- 1121639      14      bad
chromium@449643      41993345 +- 740473       9       bad
chromium@449658      41930084 +- 687210       9       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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981249701391258368

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5007656223244288


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 24 by 42576172...@developer.gserviceaccount.com, Apr 28 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981249694915520640

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5907807951388672


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 25 by 42576172...@developer.gserviceaccount.com, Apr 28 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981197660266525232

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6333478804652032


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 28 by 42576172...@developer.gserviceaccount.com, Apr 29 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981116962616004032

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5842560452395008


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 29 by 42576172...@developer.gserviceaccount.com, Apr 29 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981090824078945248

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5816218310672384


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 30 by 42576172...@developer.gserviceaccount.com, Apr 29 2017


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : fdoray
  Commit : 99e56574537a8cf27af4c86f34e42fcad65e21c8
  Date   : Fri Feb 10 16:41:56 2017
  Subject: Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp

Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles
  Change       : 3.38% | 32310562.5714 -> 33399584.6667

Revision             Result                   N
chromium@449627      32310563 +- 2289776      14      good
chromium@449631      32179930 +- 1377226      6       good
chromium@449632      31637261 +- 1182335      6       good
chromium@449633      33364211 +- 1324812      6       bad       <--
chromium@449635      33483311 +- 1921698      9       bad
chromium@449643      33136248 +- 1321849      14      bad
chromium@449658      33399585 +- 1910278      9       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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981086468336632240

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5907807951388672


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 31 by 42576172...@developer.gserviceaccount.com, Apr 29 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980995211199124160

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6333478804652032


| 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 Speed>Bisection.  Thank you!
Project Member

Comment 33 by 42576172...@developer.gserviceaccount.com, Apr 30 2017


=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980899884625862400

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5816218310672384


| 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 Speed>Bisection.  Thank you!
 Issue 691659  has been merged into this issue.
The only graph that matters is:

ChromiumPerf/android-webview-nexus6/system_health.memory_mobile / memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg / load_games / load_games_bubbles

Other graphs have stopped reporting data.

This graph has recovered when r449633 "Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp" was reverted (r464121). Starting bisect to prove that the revert is responsible for the recovery.

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980705423501365360

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5271119247966208


| 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 Speed>Bisection.  Thank you!

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_games/load_games_bubbles


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=load.games.bubbles system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980685414903684896

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5271119247966208


| 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 Speed>Bisection.  Thank you!
Blockedon: 713747
Blocking on crbug.com/713747 for busted android_webview_nexus6_aosp_perf_bisect.
Blockedon: 718928
 crbug.com/718928  for the android_webview_arm64_aosp_perf_bisect issues.
Project Member

Comment 42 by bugdroid1@chromium.org, May 30 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f23174882c11f77a5b58cf37ab4051a2ad68a4bf

commit f23174882c11f77a5b58cf37ab4051a2ad68a4bf
Author: fdoray <fdoray@chromium.org>
Date: Tue May 30 17:37:56 2017

Reland of Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp (patchset #4 id:200001 of https://codereview.chromium.org/2812263003/ )

Reason for revert:
Relanding the patch set after the effect of reverting it has been analyzed.

Original issue's description:
> Revert of Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp (patchset #1 id:1 of https://codereview.chromium.org/2685673005/ )
>
> Reason for revert:
> Reverting to see the effect on memory benchmarks.
> BUG= 691618 
>
> Original issue's description:
> > Use TaskScheduler instead of WorkerPool in BackgroundTaskRunner.cpp
> >
> > This CL replaces base::WorkerPool::PostTask() with
> > base::PostTaskWithTraits(). The following traits are used:
> >
> > Priority: Inherited (default)
> >   The priority is inherited from the calling context (i.e. TaskTraits
> >   are initialized with the priority of the current task).
> >
> > Shutdown behavior: CONTINUE_ON_SHUTDOWN
> >   Tasks posted with this mode which have not started executing before
> >   shutdown is initiated will never run. Tasks with this mode running at
> >   shutdown will be ignored (the worker will not be joined).
> >
> >   Note: Tasks that were previously posted to base::WorkerPool should
> >   use this shutdown behavior because this is how base::WorkerPool
> >   handles all its tasks.
> >
> > *No* May Block (default):
> >   Tasks may no block on synchronous file I/O operations or on sync
> >   primitives.
> >
> > BUG= 659191 
> >
> > Review-Url: https://codereview.chromium.org/2685673005
> > Cr-Commit-Position: refs/heads/master@{#449633}
> > Committed: https://chromium.googlesource.com/chromium/src/+/99e56574537a8cf27af4c86f34e42fcad65e21c8
>
> TBR=kinuko@chromium.org,haraken@chromium.org
> # Not skipping CQ checks because original CL landed more than 1 days ago.
> BUG= 659191 
>
> Review-Url: https://codereview.chromium.org/2812263003
> Cr-Commit-Position: refs/heads/master@{#464121}
> Committed: https://chromium.googlesource.com/chromium/src/+/24e5d1ddee9fa61587b6f0279a25792d0208b9f1

TBR=kinuko@chromium.org,haraken@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG= 691618 

Review-Url: https://codereview.chromium.org/2914633002
Cr-Commit-Position: refs/heads/master@{#475582}

[modify] https://crrev.com/f23174882c11f77a5b58cf37ab4051a2ad68a4bf/third_party/WebKit/Source/platform/threading/BackgroundTaskRunner.cpp
[modify] https://crrev.com/f23174882c11f77a5b58cf37ab4051a2ad68a4bf/third_party/WebKit/Source/platform/threading/DEPS

Status: Fixed (was: Untriaged)
The suspected CLs move tasks from WorkerPool to TaskScheduler. This improved metrics and regressed others.

Sign in to add a comment