Issue metadata
Sign in to add a comment
|
33.5%-49.6% regression in media.mobile at 588274:588276 |
||||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Sep 4
📍 Pinpoint job started. https://pinpoint-dot-chromeperf.appspot.com/job/174cbb6d640000
,
Sep 4
📍 Found a significant difference after 1 commit. https://pinpoint-dot-chromeperf.appspot.com/job/174cbb6d640000 Revert "Use GL_COMMANDS_ISSUED_CHROMIUM query for shared memory GpuMemoryBuffers" by ericrk@chromium.org https://chromium.googlesource.com/chromium/src/+/6eb9eee331affe92c52a2926f57422bcbf40fa43 4.229e+06 → 6.326e+06 (+2.097e+06) Understanding performance regressions: http://g.co/ChromePerformanceRegressions
,
Sep 5
This should resolve if we can find a safe way to re-land the CL I reverted. Over to sunnyps@ who is working on this CL.
,
Sep 5
,
Sep 8
Issue 881911 has been merged into this issue.
,
Sep 18
,
Sep 20
The original CL had a bug which caused no queries to be created, and staging buffers to be reused aggressively and incorrectly. Re-landing the CL did not recover the graphs, but the improvement was due to a bug so WontFix. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Sep 4