Issue metadata
Sign in to add a comment
|
1.7%-200% regression in media.tough_video_cases_tbmv2 at 482245:482435 |
||||||||||||||||||||||
Issue descriptionThese 276 alerts look valid (at least from the subset of the group whose graphs loaded). There are 202 other alerts in two groups (106+96) which have regression ranges that intersect thse 276's range, so I expect likely duplication upon successful bisection of these ~478 alerts.
,
Jun 28 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8975529971328606464
,
Jun 28 2017
Potential duplicates (similar regression range and metrics): bug 737599 , bug 737596
,
Jun 28 2017
=== BISECT JOB RESULTS === Bisect was unable to run to completion The bisect was able to narrow the range, you can try running with: good_revision: bfbaeb51e23a640bf32e084e34ad6e7e8339e98b bad_revision : 6fb10fe065b1cfe4c4aa691c31c03e7387a72d40 If failures persist contact the team (see below) and report the error. Bisect Details Configuration: winx64ati_perf_bisect Benchmark : media.tough_video_cases_tbmv2 Metric : memory:chrome:browser_process:reported_by_chrome:gpu:effective_size_avg/video.html?src_smpte_3840x2160_60fps_vp9.webm_seek Change : 200.00% | 2097088.0 -> 6291328.0 Revision Result N chromium@482296 2097088 +- 0.0 6 good chromium@482326 2097088 +- 0.0 6 good chromium@482334 2097088 +- 0.0 6 good chromium@482336 6291328 +- 0.0 6 bad chromium@482338 6291328 +- 0.0 6 bad chromium@482341 6291328 +- 0.0 6 bad chromium@482356 6291328 +- 0.0 6 bad chromium@482415 6291328 +- 0.0 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=video.html.src.smpte.3840x2160.60fps.vp9.webm.seek media.tough_video_cases_tbmv2 Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8975529971328606464 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5633318395052032 | 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!
,
Jun 28 2017
+crouleau@ -- Please take a look at the bisection failure in #4. Unless r482335 is the real culprit (unlikely, given bug 737599 and bug 737596 both had very similar metric alerts as this one, with culprit confirmed as r482336), this looks like another duplicate of bug 737396 . (I've reported #4 as bad bisect.) I'll kick off another bisect just to see if it can complete. If that next bisect fails, I recommend just duping this into bug 737396 .
,
Jun 28 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8975524465442821712
,
Jun 28 2017
(#6 started automatically)
,
Jun 28 2017
If it's only the two commits, we can just add the developers for each commit. rnephew@ in particular should know whether his change could have been the culprit since he understands Telemetry system. +treib@, rnephew@ FYI.
,
Jun 28 2017
Yeah, seems likely that this is mine too. That change adds some work on "page load finished", which is apparently not the best time to do extra work ;)
,
Jun 28 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Marc Treib Commit : 6fb10fe065b1cfe4c4aa691c31c03e7387a72d40 Date : Mon Jun 26 18:39:55 2017 Subject: Fieldtrial testing config for CaptureThumbnailOnLoadFinished Bisect Details Configuration: winx64ati_perf_bisect Benchmark : media.tough_video_cases_tbmv2 Metric : memory:chrome:browser_process:reported_by_chrome:gpu:effective_size_avg/video.html?src_smpte_3840x2160_60fps_vp9.webm_seek Change : 200.00% | 2097088.0 -> 6291328.0 Revision Result N chromium@482296 2097088 +- 0.0 6 good chromium@482326 2097088 +- 0.0 6 good chromium@482334 2097088 +- 0.0 6 good chromium@482335 2097088 +- 0.0 6 good chromium@482336 6291328 +- 0.0 6 bad <-- chromium@482338 6291328 +- 0.0 6 bad chromium@482341 6291328 +- 0.0 6 bad chromium@482356 6291328 +- 0.0 6 bad chromium@482415 6291328 +- 0.0 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=video.html.src.smpte.3840x2160.60fps.vp9.webm.seek media.tough_video_cases_tbmv2 Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8975524465442821712 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5633318395052032 | 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 wolenetz@chromium.org
, Jun 28 2017