1%-99.5% regression in media.tough_video_cases_tbmv2 at 478949:479285 |
|||||
Issue descriptionSee the link to graphs below.
,
Jun 15 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8976701326873853664
,
Jun 16 2017
=== Auto-CCing suspected CL author sunnyps@chromium.org === Hi sunnyps@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 : Sunny Sachanandani Commit : f9a0da96d26ec1fe2121a8eef62f7bd7e9330555 Date : Tue Jun 13 22:16:40 2017 Subject: Enable gpu scheduler on waterfall and developer builds. Bisect Details Configuration: winx64nvidia_perf_bisect Benchmark : media.tough_video_cases_tbmv2 Metric : memory:chrome:gpu_process:reported_by_chrome:gpu:effective_size_avg/video.html?src_tulip2.ogg_type_audio_seek Change : 104.04% | 236230.222222 -> 481997.333333 Revision Result N chromium@479117 236230 +- 231765 9 good chromium@479159 263544 +- 0.0 6 good chromium@479170 219853 +- 239304 6 good chromium@479173 219853 +- 239304 6 good chromium@479174 219853 +- 239304 6 good chromium@479175 525688 +- 0.0 6 bad <-- chromium@479180 525688 +- 0.0 6 bad chromium@479201 481997 +- 239304 6 bad chromium@479285 481997 +- 239304 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.tulip2.ogg.type.audio.seek media.tough_video_cases_tbmv2 Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8976701326873853664 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5746867683459072 | 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 16 2017
,
Jun 16 2017
Issue 733744 has been merged into this issue.
,
Jun 17 2017
Issue 733706 has been merged into this issue.
,
Jun 22 2017
,
Jun 24 2017
There are three classes of regressions: 1. gpu:effective_size_avg which is because of a race between the memory dump and cc's resource cleanup and is being tracked in issue 733725 and 734853 2. thread/frame time regressions which is being tracked in issue 736553 3. v8 heap code space regression of ~10-20 KB which is this bug. (probably the least important for now)
,
Jun 25 2017
,
Jun 28 2017
Issue 735649 has been merged into this issue.
,
Jun 30 2017
,
Sep 6 2017
Marking P3 based on comment in #8 that this issue is "(probably the least important for now)" |
|||||
►
Sign in to add a comment |
|||||
Comment 1 by dalecur...@chromium.org
, Jun 15 2017