New issue
Advanced search Search tips

Issue 866493 link

Starred by 1 user
Status: Duplicate
Owner:
sigurds@chromium.org
Closed: Jul 25
Cc:
perezju@chromium.org
kojii@chromium.org
sullivan@chromium.org
fs...@chromium.org
e...@chromium.org
sigurds@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

1.5%-4.3% regression in system_health.memory_mobile at 1531899977:1531917361

Project Member Reported by sullivan@chromium.org, Jul 23 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Jul 23 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=866493

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=2287f286dfd31e324655941b04726d250f2f87e27a03e19a6eb14695fb1c8337


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

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

Comment 3 by 42576172...@developer.gserviceaccount.com, Jul 24

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

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

Hi kojii@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 : Koji Ishii
  Commit : dbbee69d29d7fbf6858327a44d352c7f6090ebec
  Date   : Wed Jul 18 05:41:31 2018
  Subject: [auto] flag-specific rebaselines for space and tabs only differences

Bisect Details
  Configuration: webview-low-end-phone-perf-bisect
  Benchmark    : memory.top_10_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/background/after_http_www_baidu_com_s_word_google
  Change       : 4.33% | 13439320.0 -> 14021634.6667

Revision                                       Result                   N
android-chrome@74662f91c8                      13439320 +- 193665       9       good
android-chrome@1032e70048                      13816038 +- 206320       9       good
android-chrome@1032e70048,chromium@575953      13771608 +- 49719.4      6       good
android-chrome@1032e70048,chromium@575954      14035581 +- 252689       14      bad       <--
android-chrome@4e67b217c5                      14219722 +- 346657       9       bad
android-chrome@22c1f49355                      14021635 +- 416549       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=http.www.baidu.com.s.word.google memory.top_10_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8940199373497890496


For feedback, file a bug with component Speed>Bisection

Comment 5 by kojii@chromium.org, Jul 24

Cc: fs...@chromium.org e...@chromium.org
Owner: ----
The CL is only test changes, no code changes are included.

r575822 is suspicious, running bisect again according to:
https://chromium.googlesource.com/chromium/src/+/master/docs/speed/addressing_performance_regressions.md#If-you-don_t-believe-your-CL-could-be-the-cause
Project Member

Comment 6 by 42576172...@developer.gserviceaccount.com, Jul 24 


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

Bisect Details
  Configuration: go-webview-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_tools/load_tools_stackoverflow
  Change       : 0.41% | 22413168.381 -> 22505133.3333

Revision                       Result                  N
android-chrome@cf28be054f      22413168 +- 260286      21      good
android-chrome@74662f91c8      22505133 +- 315249      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=load.tools.stackoverflow 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/8940137080006025056


For feedback, file a bug with component Speed>Bisection

Comment 8 by kojii@chromium.org, Jul 24 

Most charts at https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICQjYHdsQkM

show a jump in similar range, but bisects in all bugs ( issue 865628 ,  issue 866475 ,  issue 866495 , and issue 866478) seem to be failing.

Comment 10 by sullivan@chromium.org, Jul 24 

Trying another bisect on stackoverflow, it seemed to have reproduced the issue but got confused by noise.
Project Member

Comment 11 by 42576172...@developer.gserviceaccount.com, Jul 24 


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

Bisect Details
  Configuration: go-webview-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_tools/load_tools_stackoverflow
  Change       : 0.58% | 22364113.9048 -> 22493674.2857

Revision                       Result                  N
android-chrome@cf28be054f      22364114 +- 349857      21      good
android-chrome@74662f91c8      22493674 +- 246359      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=load.tools.stackoverflow 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/8940121348016616768


For feedback, file a bug with component Speed>Bisection

Comment 12 by sullivan@chromium.org, Jul 24

Cc: perezju@chromium.org
+test owner perezju, any idea what to do here?

Comment 13 by kojii@chromium.org, Jul 24

Owner: sullivan@chromium.org
Is it possible to manually bisect for r575822? It's one of possible suspects. Emil told me that we have experts to do this.

Comment 14 by sullivan@chromium.org, Jul 24

Owner: perezju@chromium.org
Assigning to benchmark owner perezju for further investigation.
Project Member

Comment 15 by 42576172...@developer.gserviceaccount.com, Jul 24 


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

Bisect Details
  Configuration: go-webview-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_tools/load_tools_stackoverflow
  Change       : 0.76% | 22363431.2381 -> 22534097.9048

Revision                       Result                  N
android-chrome@cf28be054f      22363431 +- 300395      21      good
android-chrome@74662f91c8      22534098 +- 300888      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=load.tools.stackoverflow 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/8940109441960759152


For feedback, file a bug with component Speed>Bisection

Comment 16 by perezju@chromium.org, Jul 25

Owner: ----
Status: Available (was: Assigned)
Let's try bisecting again on one of the larger regressions.
Project Member

Comment 18 by 42576172...@developer.gserviceaccount.com, Jul 25

Cc: sigurds@chromium.org
Owner: sigurds@chromium.org
Status: Assigned (was: Available)

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

Hi sigurds@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 : Sigurd Schneider
  Commit : 46a78fbedfcd458f5d35097bdf1a9947644f4b0e
  Date   : Tue Jul 17 15:33:20 2018
  Subject: Revert "[embedded-builtins] Enable on all arches except x86 for benchmarks"

Bisect Details
  Configuration: webview-phone-perf-bisect
  Benchmark    : memory.top_10_mobile
  Metric       : memory:webview:all_processes:reported_by_os:system_memory:proportional_resident_size_avg/foreground/https_mobile_twitter_com_justinbieber_skip_interstitial_true
  Change       : 0.82% | 69977773.3333 -> 70548994.6667

Revision                                                     Result                   N
android-chrome@74662f91c8                                    69977773 +- 120752       9       good
android-chrome@74662f91c8,chromium@575926                    70075224 +- 244205       14      good
android-chrome@74662f91c8,chromium@575929                    69617496 +- 162170       6       good
android-chrome@74662f91c8,chromium@575929,v8@46a78fbedf      70354776 +- 201908       6       bad       <--
android-chrome@74662f91c8,chromium@575929,v8@e3a5b1e402      70566232 +- 246114       6       bad
android-chrome@74662f91c8,chromium@575929,v8@16af1baac4      70691843 +- 172262       6       bad
android-chrome@74662f91c8,chromium@575930                    70378840 +- 233826       6       bad
android-chrome@74662f91c8,chromium@575931                    70468952 +- 130284       6       bad
android-chrome@74662f91c8,chromium@575935                    70406829 +- 172427       9       bad
android-chrome@1032e70048                                    70311484 +- 238848       9       bad
android-chrome@4e67b217c5                                    70251864 +- 77840.2      6       bad
android-chrome@bdf6ddc146                                    70548995 +- 230370       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=https.mobile.twitter.com.justinbieber.skip.interstitial.true memory.top_10_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8940043309736985936


For feedback, file a bug with component Speed>Bisection

Comment 19 by perezju@chromium.org, Jul 25

Mergedinto: 865592
Status: Duplicate (was: Assigned)
►

Sign in to add a comment