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

Issue 850399 link

Starred by 1 user

Issue metadata

Status: Assigned
Owner:
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

1%-11.7% regression in system_health.memory_mobile at 1528178519:1528246901

Project Member Reported by mlippautz@chromium.org, Jun 7 2018

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=850399

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=e8d55dde0810c2725ab5c1e87e7d163094933c6a92437145394316b210c7ae04


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

health-plan-clankium-low-end-phone
health-plan-clankium-phone
health-plan-webview-low-end-phone
perf-go-phone-1024
perf-go-phone-512
perf-go-webview-phone

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Please try rerunning the bisect.


If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: clankium-phone-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/load_social/load_social_twitter
  Change       : 12.21% | 2385250.66667 -> 2676381.33333

Revision                                                     Result                  N
android-chrome@435fb32486                                    2385251 +- 25.3824      6       good
android-chrome@2d0aa27954                                    2404817 +- 26.3717      6       good
android-chrome@d46272ab3e                                    2404817 +- 30.217       6       good
android-chrome@d46272ab3e,chromium@564289                    2401537 +- 6784.8       6       good
android-chrome@d46272ab3e,chromium@564298                    2404307 +- 41.1566      6       good
android-chrome@d46272ab3e,chromium@564301                    2404318 +- 29.5838      6       good
android-chrome@d46272ab3e,chromium@564302                    2403135 +- 4409.4       14      good
android-chrome@d46272ab3e,chromium@564302,v8@515cc07d28      2392149 +- 44.4282      6       bad
android-chrome@d46272ab3e,chromium@564302,v8@f34937caef      2628084 +- 91360.8      9       bad
android-chrome@d46272ab3e,chromium@564302,v8@66f3e8f64d      2593463 +- 119716       21      bad
android-chrome@d46272ab3e,chromium@564303                    2658277 +- 731.758      14      bad
android-chrome@d46272ab3e,chromium@564307                    2658482 +- 196.118      6       bad
android-chrome@d46272ab3e,chromium@564325                    2658665 +- 364.037      6       bad
android-chrome@d46272ab3e,chromium@564361                    2660231 +- 4175.07      6       bad
android-chrome@2385ec28b7                                    2676381 +- 556.634      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-chrome --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.social.twitter 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/8944400651862794304


For feedback, file a bug with component Speed>Bisection
Components: Blink>JavaScript>GC
Owner: u...@chromium.org
Status: Assigned (was: Untriaged)
Mhm, the changelog for V8:
  https://chromium.googlesource.com/v8/v8/+log/d86c2be0..3c891581

66f3e8f64d was still bad, which leaves us with 4 commits:
  51c9123 Remove always-true --harmony-promise-finally runtime flag by Mathias Bynens · 6 days ago
  675c1de [torque] Remove try/catch (only support try/label) by Daniel Clifford · 6 days ago
  94aac00 PPC/s390: [wasm] Make stack check independent of the Isolate. by Junliang Yan · 6 days ago
  aaa700b Revert "[heap] Remove independent handles" by Ulan Degenbaev · 6 days ago

Assigning based on aaa700b.

Ulan, this should not happen, right? I thought this is just dead code that has been revived.

Sign in to add a comment