Regression on webrtc.webrtc_smoothness |
|||||||||
Issue descriptionSome graphs for this regression. https://chromeperf.appspot.com/report?sid=a228dc894a4356f4988d20e720621b4f3e14fd38f7dd9073a6a6a01968786472
,
Mar 21 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984511700547534288
,
Mar 21 2017
=== Auto-CCing suspected CL author peary2@gmail.com === Hi peary2@gmail.com, 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 : peary2 Commit : 27c39dbef0b07b7cd62fb2476d0f0836115fe672 Date : Tue Mar 14 06:43:39 2017 Subject: Remove |remote| and |readonly| members of MediaStreamTrack. Bisect Details Configuration: linux_perf_bisect Benchmark : webrtc.webrtc_smoothness Metric : WebRTCRendering_fps/Action_Create_PeerConnection/720p_call_45s Change : 195.91% | 19.9711576955 -> 59.0975394236 Revision Result N chromium@456615 19.9712 +- 0.016381 9 good chromium@456630 19.9761 +- 0.0204001 9 good chromium@456637 19.9778 +- 0.0148313 9 good chromium@456638 19.972 +- 0.0139972 9 good chromium@456639 59.0945 +- 0.00943145 4 bad <-- chromium@456641 59.0906 +- 0.0120508 5 bad chromium@456644 59.0973 +- 0.0154127 7 bad chromium@456673 59.0942 +- 0.0123607 7 bad chromium@456731 59.0886 +- 0.00999405 5 bad chromium@456846 59.0919 +- 0.0108779 4 bad chromium@457076 59.0975 +- 0.0154988 4 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 webrtc.webrtc_smoothness Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8984511739974498512 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5681401287933952 | 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 22 2017
,
Mar 22 2017
I'm trying to revert the culprit CL, but will reland if it turns out that this was not really a problem.
,
Mar 22 2017
,
Mar 22 2017
,
Mar 22 2017
The revert failed.
,
Mar 23 2017
I would like to fix this bug as new patch. Could you give me an any idea?
,
Mar 23 2017
I fixed the rebase conflicts of the revert and landed it. Closing the bug as this will fix the performance regression. Let's continue the discussion about the removal of the attributes on Issue 598704 .
,
Mar 23 2017
,
Mar 23 2017
The revert didn't actually land. After passing all bots, it could not be applied due to another conflict.
,
Mar 23 2017
The revert finally landed. |
|||||||||
►
Sign in to add a comment |
|||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Mar 21 2017