Issue metadata
Sign in to add a comment
|
15.3%-716% regression in rendering.mobile at 592220:592304 |
||||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Sep 20
📍 Pinpoint job started. https://pinpoint-dot-chromeperf.appspot.com/job/14981597640000
,
Sep 20
📍 Found a significant difference after 1 commit. https://pinpoint-dot-chromeperf.appspot.com/job/14981597640000 gpu: Expand oopr workaround to fallback to ES2 to all android versions. by khushalsagar@chromium.org https://chromium.googlesource.com/chromium/src/+/2b62fe723102af55ae3c3508d013655647af440b 0.4217 → 1.395 (+0.9734) Understanding performance regressions: http://g.co/ChromePerformanceRegressions Benchmark documentation link: https://bit.ly/rendering-benchmarks
,
Sep 20
The test which regressed is web_animation_value_type_path, I'm going to guess the fallback to ES2 changed the path rendering option in skia. This is still a WontFix. Its a temporary workaround for a GPU crash and it brings us back to the performance we had with GPU raster.
,
Sep 20
Issue 887289 has been merged into this issue. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Sep 20