Issue metadata
Sign in to add a comment
|
A zero-to-nonzero to 2.4% regression in media.tough_video_cases at 492969:493190 |
||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Aug 11 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971534388864861824
,
Aug 11 2017
=== BISECT JOB RESULTS === Perf regression found but unable to continue Bisect was stopped because a commit couldn't be classified as either good or bad. Bisect Details Configuration: winx64nvidia_perf_bisect Benchmark : media.tough_video_cases Metric : dropped_frame_count/video.html?src_crowd1080.webm Revision Result N chromium@493056 0.333333 +- 2.94392 21 good chromium@493100 2.52381 +- 16.5299 21 unknown chromium@493144 5.09524 +- 22.9741 21 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.crowd1080.webm media.tough_video_cases More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8971534388864861824 For feedback, file a bug with component Speed>Bisection
,
Aug 11 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971529789038459952
,
Aug 11 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971529774741286240
,
Aug 11 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971529755889300528
,
Aug 11 2017
=== Auto-CCing suspected CL author hubbe@google.com === Hi hubbe@google.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 : Fredrik Hubinette Commit : 91983d3c528c0181129ee51a09077fda019e3e32 Date : Wed Aug 09 19:14:38 2017 Subject: Add test and fixes for loading many videos on the same page. Bisect Details Configuration: mac_retina_perf_bisect Benchmark : media.tough_video_cases Metric : time_to_play/video.html?src_garden2_10s.mp4 Change : 918.95% | 436.8975 -> 4451.74666667 Revision Result N chromium@492988 436.898 +- 53.8354 6 good chromium@493063 437.653 +- 31.1015 6 good chromium@493082 441.034 +- 44.3682 6 good chromium@493087 448.941 +- 36.3846 6 good chromium@493088 442.868 +- 17.9285 6 good chromium@493089 4540.23 +- 959.749 6 bad <-- chromium@493091 4615.71 +- 1002.74 6 bad chromium@493100 4459.24 +- 940.581 6 bad chromium@493137 4451.75 +- 946.789 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=video.html.src.garden2.10s.mp4 media.tough_video_cases More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8971529774741286240 For feedback, file a bug with component Speed>Bisection
,
Aug 11 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Fredrik Hubinette Commit : 91983d3c528c0181129ee51a09077fda019e3e32 Date : Wed Aug 09 19:14:38 2017 Subject: Add test and fixes for loading many videos on the same page. Bisect Details Configuration: winx64nvidia_perf_bisect Benchmark : media.tough_video_cases Metric : time_to_play/video.html?src_crowd1080_vp9.webm Change : 3099.77% | 59.6916666667 -> 1909.99666667 Revision Result N chromium@493056 59.6917 +- 9.5884 6 good chromium@493078 88.0983 +- 153.651 6 good chromium@493084 106.476 +- 251.821 6 good chromium@493087 174.762 +- 618.018 6 good chromium@493088 75.5608 +- 91.6843 6 good chromium@493089 1972.35 +- 1305.96 6 bad <-- chromium@493100 2088.24 +- 724.662 6 bad chromium@493144 1910.0 +- 1656.56 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.crowd1080.vp9.webm media.tough_video_cases More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8971529789038459952 For feedback, file a bug with component Speed>Bisection
,
Aug 11 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Fredrik Hubinette Commit : 91983d3c528c0181129ee51a09077fda019e3e32 Date : Wed Aug 09 19:14:38 2017 Subject: Add test and fixes for loading many videos on the same page. Bisect Details Configuration: win_8_perf_bisect Benchmark : media.tough_video_cases Metric : time_to_play/video.html?src_crowd1080.webm Change : 54.96% | 449.254285714 -> 696.148888889 Revision Result N chromium@493054 449.254 +- 4089.19 14 good chromium@493073 166.747 +- 139.378 6 good chromium@493083 143.205 +- 51.4699 6 good chromium@493088 152.44 +- 88.6325 6 good chromium@493089 699.14 +- 355.004 6 bad <-- chromium@493090 537.364 +- 504.917 6 bad chromium@493092 703.944 +- 336.705 6 bad chromium@493129 696.149 +- 920.546 9 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=video.html.src.crowd1080.webm media.tough_video_cases More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8971529755889300528 For feedback, file a bug with component Speed>Bisection
,
Aug 11 2017
,
Aug 12 2017
,
Aug 16 2017
,
Aug 16 2017
,
Aug 17 2017
Looks like the graph recovered. +dalecurtis@ because this is interesting. The large a time_to_play increase is pretty exciting!
,
Aug 17 2017
,
Aug 17 2017
Hubbe@ reverted his CL because of this bug but neglected to reference this bug in his change: https://chromium.googlesource.com/chromium/src/+/91983d3c528c0181129ee51a09077fda019e3e32 |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Aug 11 2017