Issue metadata
Sign in to add a comment
|
8.1%-41.1% regression (Execution speed) in v8.runtimestats.browsing_desktop at 455495:455817 |
||||||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Mar 13 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8985245588612589184
,
Mar 13 2017
,
Mar 13 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8985245521796685056
,
Mar 13 2017
That looks like the anticipated change in the performance profile: More time spent executing JavaScript, less time in Compilation and ICs. Overall a small net win. Example: https://chromeperf.appspot.com/report?sid=87c972e75d611e2adc18911993f0506a94a884d74744c565427a702379685ca7
,
Mar 13 2017
,
Mar 13 2017
Re #5, should we just make this wontfix then?
,
Mar 13 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Michael Hablich Commit : 7c3936c67a6361371d9a00d83e2c03ecc603ce76 Date : Thu Mar 09 08:05:42 2017 Subject: Version 5.9.34.1 (Turn on I+TF) Bisect Details Configuration: winx64intel_perf_bisect Benchmark : v8.runtimestats.browsing_desktop Metric : JavaScript:duration_avg/browse_media/browse_media_youtube Change : 24.65% | 501.798333333 -> 625.484333333 Revision Result N chromium@455707 501.798 +- 53.5957 6 good chromium@455717 508.44 +- 20.0501 6 good chromium@455719 517.525 +- 35.4576 6 good chromium@455719,v8@7c3936c67a 623.464 +- 33.5665 6 bad <-- chromium@455719,v8@fbffc377e3 694.453 +- 379.676 6 bad chromium@455720 623.956 +- 38.8809 6 bad chromium@455722 628.4 +- 46.5368 6 bad chromium@455726 638.262 +- 66.446 6 bad chromium@455744 623.075 +- 26.7482 6 bad chromium@455781 625.484 +- 21.6812 6 bad To Run This Test src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.media.youtube v8.runtimestats.browsing_desktop Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8985245588612589184 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=6461026788179968 | 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 13 2017
Re #7: If the alerts are as expected, yes.
,
Mar 13 2017
Filed a bug about autobisect (#8) btw: https://github.com/catapult-project/catapult/issues/3372
,
Mar 13 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Michael Hablich Commit : 7c3936c67a6361371d9a00d83e2c03ecc603ce76 Date : Thu Mar 09 08:05:42 2017 Subject: Version 5.9.34.1 (Turn on I+TF) Bisect Details Configuration: mac_retina_perf_bisect Benchmark : v8.runtimestats.browsing_desktop Metric : JavaScript:duration_avg/browse_media/browse_media_youtube Change : 29.04% | 898.343 -> 1159.183 Revision Result N chromium@455719 898.343 +- 77.9937 6 good chromium@455719,v8@7c3936c67a 1170.55 +- 96.8662 6 bad <-- chromium@455719,v8@fbffc377e3 1163.8 +- 40.9735 6 bad chromium@455720 1156.29 +- 58.8088 6 bad chromium@455721 1174.53 +- 62.7686 6 bad chromium@455723 1170.78 +- 100.307 6 bad chromium@455726 1154.87 +- 75.1395 6 bad chromium@455732 1166.21 +- 35.0946 6 bad chromium@455744 1169.81 +- 78.3937 6 bad chromium@455768 1190.78 +- 115.214 6 bad chromium@455816 1159.18 +- 45.7426 6 bad To Run This Test src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.media.youtube v8.runtimestats.browsing_desktop Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8985245521796685056 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=6432107162763264 | 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! |
|||||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||||
Comment 1 by hablich@chromium.org
, Mar 13 2017