Issue metadata
Sign in to add a comment
|
1%-11.1% regression in memory.top_10_mobile_stress at 453587:453744 |
||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Mar 2 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8986223096034551184
,
Mar 3 2017
=== Auto-CCing suspected CL author verwaest@chromium.org === Hi verwaest@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 : Toon Verwaest Commit : 6477f0777f57736411a02837301451ffb0e349ff Date : Tue Feb 28 14:08:05 2017 Subject: [api] Mark api prototypes as prototype before configuring them Bisect Details Configuration: android_nexus5X_perf_bisect Benchmark : memory.top_10_mobile_stress Metric : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_www_baidu_com_s_word_google Change : 10.80% | 6534934.66667 -> 7240578.66667 Revision Result N chromium@453645 6534935 +- 44579.4 6 good chromium@453645,v8@44debc4f71 6535551 +- 12868.0 6 good chromium@453645,v8@a5dfa06213 6629605 +- 484622 6 good chromium@453645,v8@6477f0777f 7241897 +- 9413.23 6 bad <-- chromium@453645,v8@a35616e324 7239364 +- 5394.38 6 bad chromium@453646 7239621 +- 5192.11 6 bad chromium@453647 7127461 +- 622062 6 bad chromium@453649 7238591 +- 2791.01 6 bad chromium@453652 7237903 +- 7464.45 6 bad chromium@453658 7214377 +- 136678 6 bad chromium@453670 7241063 +- 10404.9 6 bad chromium@453695 7245631 +- 19935.1 6 bad chromium@453744 7240579 +- 6713.28 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 memory.top_10_mobile_stress Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8986223096034551184 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=4946449218928640 | 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 3 2017
I looked into a few of the regressions, and they were offset by improvements in other runs on the same session. This probably just changes GC timing, since I'm doing fewer object changes. The final object shape is the same.
,
Mar 11 2017
,
Mar 14 2017
Can you also explain the regression in ChromiumPerf/android-webview-nexus6/v8.runtimestats.browsing_mobile_turbo / v8-gc-scavenger_avg / browse_social / browse_social_twitter https://chromeperf.appspot.com/report?sid=f785a21ceddf65d74684594c158f11f88432e5f5e4371d6e055142780e525a49&rev=454900 |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by sullivan@chromium.org
, Mar 2 2017