Issue metadata
Sign in to add a comment
|
2.1%-6.6% regression in memory.top_10_mobile at 458726:458749 |
||||||||||||||||||||
Issue descriptionThe background cases look like a land/revert pattern, but that isn't paralleled by the foreground cases. Kicking off a bisect to dig into it.
,
Mar 23 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984291289798383344
,
Mar 24 2017
=== Auto-CCing suspected CL author msarda@chromium.org === Hi msarda@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 : msarda Commit : d625c6d1708edd48bada92dde69fc4905dcf1dd2 Date : Wed Mar 22 13:06:30 2017 Subject: Remove finch experiments for NewProfileManagement. Bisect Details Configuration: android_webview_nexus6_aosp_perf_bisect Benchmark : memory.top_10_mobile Metric : memory:webview:all_processes:reported_by_os:system_memory:java_heap:proportional_resident_size_avg/background/after_http_www_baidu_com_s_word_google Change : 7.27% | 1660586.66667 -> 1781248.0 Revision Result N chromium@458725 1660587 +- 150796 9 good chromium@458726 1789269 +- 3739.12 6 bad <-- chromium@458727 1787221 +- 4729.65 6 bad chromium@458728 1787221 +- 4729.65 6 bad chromium@458731 1787904 +- 5016.55 6 bad chromium@458737 1773568 +- 70231.8 9 bad chromium@458749 1781248 +- 47673.8 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 memory.top_10_mobile Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8984291289798383344 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=6076133612716032 | 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 24 2017
My CL is a removal of dead code, it should have no impact on performance (it is basically). I think there was probably something flaky in the bisect.
,
Jul 27 2017
This alert was found before M-60 branched. Closing as WontFix as this is believed to either be invalid or non-reproducible. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by jasontiller@chromium.org
, Mar 23 2017