"virtual/gpu/fast/canvas/OffscreenCanvas-empty-image-source.html" is flaky |
|||||
Issue description"virtual/gpu/fast/canvas/OffscreenCanvas-empty-image-source.html" is flaky. This issue was created automatically by the chromium-try-flakes app. Please find the right owner to fix the respective test/step and assign this issue to them. If the step/test is infrastructure-related, please add Infra-Troopers label and change issue status to Untriaged. When done, please remove the issue from Sheriff Bug Queue by removing the Sheriff-Chromium label. We have detected 4 recent flakes. List of all flakes can be found at https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNySgsSBUZsYWtlIj92aXJ0dWFsL2dwdS9mYXN0L2NhbnZhcy9PZmZzY3JlZW5DYW52YXMtZW1wdHktaW1hZ2Utc291cmNlLmh0bWwM. Flaky tests should be disabled within 30 minutes unless culprit CL is found and reverted. Please see more details here: https://sites.google.com/a/chromium.org/dev/developers/tree-sheriffs/sheriffing-bug-queues#triaging-auto-filed-flakiness-bugs
,
Nov 7 2016
,
Nov 7 2016
The test was created for a special case when the image pattern is created out of an empty OffscreenCanvas and it was running fine for about two months (https://codereview.chromium.org/2344573002). It starts to become flaky since Nov 4 and it only happens to GPU virtual test when [4196:5156:1104/142234:3452161:INFO:grsurface.cpp(42)] e:\b\c\b\win\src\third_party\skia\src\gpu\grsurface.cpp:42: fatal error: "assert(colorBytes > 0)" junov@, xidachen@: Is there any recent change to the GPU-accelerated OffscreenCanvas?
,
Nov 7 2016
I looked at the dashboard here: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=virtual%2Fgpu%2Ffast%2Fcanvas%2FOffscreenCanvas-empty-image-source.html Looks like it was crashing on Nov.4th and 5th, and now it is back in normal, let's keep an eye on it for the next day or two, just to make sure that it is fine.
,
Nov 7 2016
More info: This particular skia-deps-roll: https://codereview.chromium.org/2470913008/ was reverted, because it causes this test to crash. It was re-land here: https://codereview.chromium.org/2481603002/ It might have to do with the flaky. Will definitely keep an eye on that.
,
Nov 7 2016
Actually, could you please take a look at this Robert? The original skia roll here: https://codereview.chromium.org/2470913008/ seems to cause the crash with crash log: [4728:4732:1104/113437:481387:INFO:grsurface.cpp(42)] e:\b\c\b\win_layout\src\third_party\skia\src\gpu\grsurface.cpp:42: fatal error: "assert(colorBytes > 0)" The above roll was re-land here: https://codereview.chromium.org/2481603002/ Could your change here: https://chromium.googlesource.com/skia.git/+/ccd3c8937fce4bb28df19533ed043cad209e277d causes the crash?
,
Nov 7 2016
Chrome appears to passing Skia a 0 width or height. In these cases we would prefer to return a nullptr rather than a 0 sized GrSurface. For the time being I have disabled the assert and will attempt to make Chrome be able to deal with a returned nullptr.
,
Nov 14 2016
Marking as fixed for now (since the assert has disabled) so the branch isn't held up. |
|||||
►
Sign in to add a comment |
|||||
Comment 1 Deleted