Issue metadata
Sign in to add a comment
|
10.3%-745.6% regression in blink_perf.{bindings,events,parser,pywebsocket} at 378467:378564 |
||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Mar 2 2016
=== Auto-CCing suspected CL author dgozman@chromium.org === Hi dgozman@chromium.org, the bisect results pointed to your CL below as possibly causing a regression. Please have a look at this info and see whether your CL be related. ===== BISECT JOB RESULTS ===== Status: completed ===== SUSPECTED CL(s) ===== Subject : [DevTools] Postpone more agents instantiation until attached. Author : dgozman Commit description: To support multiple attached devtools, we have to instantiate individual agents on demand, rather than on creation. This patch moves most of agents, with only worker and inspector to go. BUG= 590878 Review URL: https://codereview.chromium.org/1746393003 Cr-Commit-Position: refs/heads/master@{#378509} Commit : b547858514737d59af3c55b2b06b4e511624c304 Date : Tue Mar 01 19:17:51 2016 ===== TESTED REVISIONS ===== Revision Mean Value Std. Dev. Num Values Good? chromium@378507 18.76 1.470714 5 good chromium@378508 19.5 0.863134 5 good chromium@378509 98.4 3.040559 5 bad chromium@378510 96.64 1.74442 5 bad chromium@378513 98.32 1.834939 5 bad chromium@378519 99.78 4.563113 5 bad Bisect job ran on: linux_perf_bisect Bug ID: 591406 Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --also-run-disabled-tests blink_perf.pywebsocket Test Metric: XHR-send-blob-worker-sync-verify/XHR-send-blob-worker-sync-verify Relative Change: 431.88% Score: 99.9 Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6266 Job details: https://chromeperf.appspot.com/buildbucket_job_status/9019284174577641936 | 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 label Cr-Tests-AutoBisect. Thank you!
,
Mar 2 2016
dgozman@, would your CL have affected blink_perf.pywebsocket perf test results? The bisect appears to have reproduced a ~400% increase in *-verify time (~20 ms to 90 ms) across all platforms. https://chromeperf.appspot.com/group_report?bug_id=591406
,
Mar 3 2016
pfeldman@, dgozman@, do you know why CL might have affected some Telemetry performance test results?
,
Mar 3 2016
I will look into this today.
,
Mar 3 2016
I don't see how this may affect performance, but let's revert and see whether it helps.
,
Mar 3 2016
Thanks! After the revert, if that CL did affect performance test results on the chromium.perf waterfall, we should be able to see a change within a few hours on most/all of the graphs (https://chromeperf.appspot.com/group_report?bug_id=591406).
,
Mar 4 2016
This CL likely leaks memory, see: https://bugs.chromium.org/p/chromium/issues/detail?id=591314
,
Mar 4 2016
,
Mar 4 2016
I clicked through random graphs, they all recovered. Marking as fixed. Please reopen if some graphs are still in a bad shape. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by qyears...@chromium.org
, Mar 2 2016