Issue metadata
Sign in to add a comment
|
36.3% regression in blink_perf.canvas at 459455:459585 |
||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Mar 27 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983985679146588320
,
Mar 27 2017
=== Auto-CCing suspected CL author brianosman@google.com === Hi brianosman@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 : Brian Osman Commit : f4d443fd1bd8cffbcd7a07a48f3488ae871414e9 Date : Fri Mar 24 17:22:15 2017 Subject: Fix SkImage_Gpu::getROPixels Bisect Details Configuration: mac_10_12_perf_bisect Benchmark : blink_perf.canvas Metric : upload-video-to-sub-texture/upload-video-to-sub-texture Change : 35.09% | 209.221050047 -> 135.80250626 Revision Result N chromium@459454 209.221 +- 4.18694 6 good chromium@459520 207.543 +- 5.24825 6 good chromium@459537 206.665 +- 2.03972 6 good chromium@459541 206.208 +- 2.34584 6 good chromium@459542 204.482 +- 5.68949 6 good chromium@459542,skia@d3caa77f3c 207.378 +- 3.15895 6 good chromium@459542,skia@f388093309 204.329 +- 5.66557 6 good chromium@459542,skia@f4d443fd1b 136.905 +- 4.94866 6 bad <-- chromium@459543 137.76 +- 1.17287 6 bad chromium@459545 134.723 +- 5.36226 6 bad chromium@459553 138.348 +- 1.79645 6 bad chromium@459585 135.803 +- 4.90204 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 blink_perf.canvas Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8983985679146588320 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=6084909774405632 | 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 31 2017
The CL in question was reverted due to a WebGL correctness problem, so the perf numbers on this this have returned to their previous value. See also: https://bugs.chromium.org/p/skia/issues/detail?id=6433 |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by toyoshim@chromium.org
, Mar 27 2017