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

Issue 669583 link

Starred by 1 user
Status: WontFix
Owner: ----
Closed: Apr 2017
Cc:
mlippautz@chromium.org
█ picksi@chromium.org
█ manzagop@chromium.org
simonhatch@chromium.org
hablich@chromium.org
sullivan@chromium.org
█ benhenry@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression

Blocked on:
issue 672529


Sign in to add a comment

1%-2.9% regression in memory.top_10_mobile at 434670:434725

Project Member Reported by benhenry@chromium.org, Nov 29 2016 
See graphs below.
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Nov 29 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@434689  7171780  827043   27  good
chromium@434692  7160263  832367   27  unknown
chromium@434695  7170697  833080   27  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 669583

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.yandex.ru.touchsearch.text.science memory.top_10_mobile_stress
Test Metric: memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/foreground/http_yandex_ru_touchsearch_text_science
Relative Change: 0.02%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1832
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994635818922524064


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4949165187006464

| 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, Nov 30 2016 


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


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@434673  7142026  839427   27  good
chromium@434683  7052702  570356   12  good
chromium@434684  7158264  857886   27  unknown
chromium@434685  7191447  760693   27  unknown
chromium@434688  7102341  690367   18  good
chromium@434689  7167197  812693   27  unknown
chromium@434690  7156928  684428   18  bad
chromium@434695  7172401  683825   18  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 669583

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.yandex.ru.touchsearch.text.science memory.top_10_mobile_stress
Test Metric: memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/foreground/http_yandex_ru_touchsearch_text_science
Relative Change: 0.43%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1833
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994611288092223520


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5701430965436416

| 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 6 by benhenry@google.com, Nov 30 2016

Cc: picksi@chromium.org
Owner: manzagop@chromium.org
Status: Assigned (was: Untriaged)
crrev.com/434689  -> https://chromium.googlesource.com/chromium/src/+/b5f74dbb3ccdc02be372cec383f3accd6b240f41

Assigning to the author of that CL. Looking at the CL, I couldn't tell why it would regress memory unless we've added a new library that didn't exist before.

Comment 7 by picksi@chromium.org, Nov 30 2016

Cc: sullivan@chromium.org
+Annie

Annie, tangential to this bug, it seems strange to me that alerts have been generated on this regression but not on the subsequent much larger regressions (as shown on the graphs here https://chromeperf.appspot.com/group_report?bug_id=669583)

Comment 8 by sullivan@chromium.org, Nov 30 2016 

Re #7: The dashboard waits for a minimum number of data points to come in before alerting on a regression, because many performance charts have unavoidable spikes. There hasn't been enough data yet to alert on the more recent large regressions.

Comment 9 by pmeenan@chromium.org, Dec 13 2016 

manzagop@ - ping.  Have you had a chance to look into this?

Comment 10 by manzagop@chromium.org, Dec 13 2016 

Yikes, sorry, I missed this.

I don't think my CL could have caused a regression. Aside from the includes, the code change is in a "#if !defined(OS_ANDROID)" block.

Comment 11 by manzagop@chromium.org, Dec 13 2016

Cc: manzagop@chromium.org
Owner: ----
Status: Available (was: Assigned)
Project Member

Comment 14 by 42576172...@developer.gserviceaccount.com, Dec 13 2016 


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


=== Bisection aborted ===
The bisect was aborted because Infra Failure: Step('Working on revision chromium@434689,v8@0a1dcadd05.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@434669  5207041  139.217  6  good
chromium@434687  5207240  767.833  6  good
chromium@434689  5207095  132.625  6  good
chromium@434690  5224928  10290.0  6  bad
chromium@434692  5224973  10340.3  6  bad
chromium@434696  5226796  190.578  6  bad
chromium@434704  5226777  182.574  6  bad

Bisect job ran on: android_nexus9_perf_bisect
Bug ID: 669583

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --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
Test Metric: memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/foreground/https_mobile_twitter_com_justinbieber_skip_interstitial_true
Relative Change: 0.38%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus9_perf_bisect/builds/2326
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993355484899643312


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5886751413895168

| 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 15 by 42576172...@developer.gserviceaccount.com, Dec 13 2016 


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


=== Bisection aborted ===
The bisect was aborted because Infra Failure: Step('Working on revision chromium@434689,v8@0a1dcadd05.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@434689  5278172  56908.5  6  good
chromium@434690  5341536  720.533  6  bad
chromium@434691  5341765  806.507  6  bad
chromium@434693  5344048  15951.8  6  bad
chromium@434697  5347371  19060.3  6  bad
chromium@434704  5344315  15510.7  6  bad

Bisect job ran on: android_nexus9_perf_bisect
Bug ID: 669583

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --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_stress
Test Metric: memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_https_mobile_twitter_com_justinbieber_skip_interstitial_true
Relative Change: 1.25%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus9_perf_bisect/builds/2327
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993355470335226240


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5786618814791680

| 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 16 by picksi@chromium.org, Dec 14 2016

Cc: simonhatch@chromium.org
+Simon FYI

Comment 17 by sullivan@chromium.org, Dec 14 2016

Blockedon: 672529

Comment 18 by sullivan@chromium.org, Dec 14 2016

Cc: hablich@chromium.org
The v8 roll in r434690 appears to have caused the memory regression, but we can't bisect into it due to  bug 672529 . +hablich in case he has any ideas on what in the roll we should look at while we are waiting.

Comment 19 by hablich@chromium.org, Dec 15 2016

Cc: mlippautz@chromium.org
Looks quite innocent. Adding memory sheriff.

Comment 20 by mlippautz@chromium.org, Dec 15 2016 

Yeah, looks minor.

Since the fix is apparently ready in  bug 672529  I'd suggest waiting until the bisect-into works again.
Project Member

Comment 22 by 42576172...@developer.gserviceaccount.com, Apr 10 2017 


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

Bisect Details
  Configuration: android_one_perf_bisect
  Benchmark    : memory.top_10_mobile_stress
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/foreground/http_yandex_ru_touchsearch_text_science

Revision             Result                 N
chromium@434689      7161842 +- 714135      21      good
chromium@434695      7164990 +- 736456      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-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.yandex.ru.touchsearch.text.science memory.top_10_mobile_stress

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982667528163452000

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4949165187006464


| 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 11 2017 


=== BISECT JOB RESULTS ===
Bisect failed unexpectedly

Bisect was aborted with the following:
  Infra Failure: Step('Working on revision chromium@434689,v8@0a1dcadd05.install ChromePublic.apk') returned 2


Bisect Details
  Configuration: android_nexus9_perf_bisect
  Benchmark    : memory.top_10_mobile_stress
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_https_mobile_twitter_com_justinbieber_skip_interstitial_true

Revision             Result                  N
chromium@434689      5278172 +- 56908.5      6      good
chromium@434690      5341536 +- 720.533      6      bad
chromium@434691      5341765 +- 806.507      6      bad
chromium@434693      5344048 +- 15951.8      6      bad
chromium@434697      5347371 +- 19060.3      6      bad
chromium@434704      5344315 +- 15510.7      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-chromium --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_stress

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982592352599630464

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5786618814791680


| 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!

Comment 25 by mlippautz@chromium.org, Apr 12 2017

Status: WontFix (was: Available)
Let's ignore this one. The following changes (in both directions) have been way more important than this regression.
►

Sign in to add a comment