Issue metadata
Sign in to add a comment
|
3.6%-149.1% regression in system_health.common_desktop at 545378:545519 |
||||||||||||||||||||
Issue description+tdresser: we only see this on one device config, and it seems to be coming back to a level it was at before. Will wait for a bisect, but worried this might be bimodal.
,
Mar 25 2018
📍 Pinpoint job started. https://pinpoint-dot-chromeperf.appspot.com/job/1297413d440000
,
Mar 25 2018
📍 Found a significant difference after 1 commit. https://pinpoint-dot-chromeperf.appspot.com/job/1297413d440000 Enforce GPU earlier on Mac to avoid GPU switching during startup. by zmo@chromium.org https://chromium.googlesource.com/chromium/src/+/ee6d9fdd6aa00ecc73a23d6da91425c3e776e498 Understanding performance regressions: http://g.co/ChromePerformanceRegressions
,
Mar 26 2018
Is this the Retina Mac? Before my CL, we have a bug so Chrome ends up starting on discrete GPU (so of course it runs faster). With my CL, Chrome correctly starts on integrated GPU. This is desired behavior and if perf regresses a bit, we'll have to live with it.
,
Mar 26 2018
Yep, this is the retina macbook with dual GPU.
,
Mar 26 2018
Thanks for confirmation. Then there is nothing we should do here.
,
Mar 26 2018
|
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Mar 25 2018