Issue metadata
Sign in to add a comment
|
1%-32.8% regression in memory.top_10_mobile_stress at 454720:454995 |
||||||||||||||||||||
Issue descriptionsomething affected a ton of tests.
,
Mar 8 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8985654489430880224
,
Mar 9 2017
=== Auto-CCing suspected CL author ulan@chromium.org === Hi ulan@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 : ulan Commit : af55a29706703121f83837d0c9020649bb9ec6ce Date : Thu Mar 02 18:27:49 2017 Subject: [heap] Do not delay mark-compact by doing scavenge. Bisect Details Configuration: android_webview_nexus6_aosp_perf_bisect Benchmark : system_health.memory_mobile Metric : memory:webview:all_processes:reported_by_chrome:v8:effective_size_avg/background_media/background_media_imgur Change : 30.99% | 12602436.8889 -> 15546644.0 Revision Result N chromium@454750 12602437 +- 686866 9 good chromium@454788 12582110 +- 837883 14 good chromium@454793 12702834 +- 682022 14 good chromium@454793,v8@59c9e6ff69 12718161 +- 321361 14 good chromium@454793,v8@802d367df2 12630485 +- 811485 14 good chromium@454793,v8@ff8b1abb1a 12608270 +- 1314689 14 good chromium@454793,v8@12ce15c35b 12650699 +- 610536 9 good chromium@454793,v8@d003c2ad9c 12484969 +- 450026 6 good chromium@454793,v8@af55a29706 16154719 +- 3809560 9 bad <-- chromium@454793,v8@386e5a1149 15478738 +- 9117979 21 bad chromium@454794 15273249 +- 9634477 21 bad chromium@454796 15927735 +- 5851684 14 bad chromium@454797 15933191 +- 6092353 14 bad chromium@454805 15504891 +- 9247497 21 bad chromium@454822 15635203 +- 6865819 14 bad chromium@454900 15546644 +- 5574582 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=background.media.imgur system_health.memory_mobile Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8985654489430880224 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5516293098176512 | 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!
,
Mar 9 2017
,
Mar 13 2017
The CL changes GC timing. Some regressions are expected. The graphs have either recovered or even better than before thanks to TF+I launch. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by benhenry@google.com
, Mar 8 2017