Issue metadata
Sign in to add a comment
|
61% regression in rendering.desktop at 585366:585398 |
||||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Aug 23
📍 Pinpoint job started. https://pinpoint-dot-chromeperf.appspot.com/job/17898461640000
,
Aug 23
📍 Found significant differences after each of 2 commits. https://pinpoint-dot-chromeperf.appspot.com/job/17898461640000 Enable Perfetto by default for all telemetry tests by oysteine@google.com https://chromium.googlesource.com/chromium/src/+/32c8ebbeeaee6dfdf46ae51d4c9c5b67dd465fed 76.94 → 100.6 (+23.63) Skip login flow when replaying top_real_world_desktop pages by vovoy@chromium.org https://chromium.googlesource.com/chromium/src/+/dbf0f2f517312568dfbb417c6be7420ba1c68f81 100.6 → 133.4 (+32.79) Understanding performance regressions: http://g.co/ChromePerformanceRegressions
,
Aug 24
,
Aug 27
"Skip login flow when replaying top_real_world_desktop pages" changed the test sequence, regression of tasks_per_frame_total_all is expected. As other major metrics of rendering.desktop are not affected, I think it's OK to set this issue won't fix. e.g. no regression on mac-10_13_laptop_high_end-perf/rendering.desktop / mean_frame_time / linkedin_2018 https://chromeperf.appspot.com/report?sid=43c5a7af750b4505ae51ecd2ea94a275005a47ba3132fefc32689bd90cd6282e
,
Aug 28
Agree, this should be fine. If anything it's now measuring something closer to what a typical (logged in) user would actually experience.
,
Aug 29
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Aug 23