WebKit Bugzilla
New
Browse
Search+
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
NEW
286080
[ macOS Release WK2 ] fast/webgpu/nocrash/fuzz-284937.html never passes
https://bugs.webkit.org/show_bug.cgi?id=286080
Summary
[ macOS Release WK2 ] fast/webgpu/nocrash/fuzz-284937.html never passes
Robert Jenner
Reported
2025-01-16 11:08:37 PST
fast/webgpu/nocrash/fuzz-284937.html never passes. It only runs on macOS Release WK2 builds. An expectation for this test has been set as Pass Fail. However, it is a constant Timeout on Ventura and Sequoia Release on Apple Silicon. On Sonoma it flaky crashes on Intel, then fluctuates between flaky timeouts and flaky failures. HISTORY:
https://results.webkit.org/?suite=layout-tests&test=fast%2Fwebgpu%2Fnocrash%2Ffuzz-284937.html
According to the history, this test was introduced with this behavior. The test was introduced at:
https://commits.webkit.org/288256@main
TIMEOUT DIFF:
https://build.webkit.org/results/Apple-Sequoia-Release-AppleSilicon-WK2-Tests/289013@main%20(1311)/fast/webgpu/nocrash/fuzz-284937-pretty-diff.html
--- /Volumes/Data/worker/Apple-Sequoia-Release-AppleSilicon-WK2-Tests/build/layout-test-results/fast/webgpu/nocrash/fuzz-284937-expected.txt +++ /Volumes/Data/worker/Apple-Sequoia-Release-AppleSilicon-WK2-Tests/build/layout-test-results/fast/webgpu/nocrash/fuzz-284937-actual.txt @@ -1,62 +1,24 @@ CONSOLE MESSAGE: promises created -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: Unhandled Promise Rejection: GPUPipelineError -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: device0 lost! -CONSOLE MESSAGE: -CONSOLE MESSAGE: the end -CONSOLE MESSAGE: the end -CONSOLE MESSAGE: repro_284937.html -CONSOLE MESSAGE: notify done +FAIL: Timed out waiting for notifyDone to be called + layer at (0,0) size 785x4861 RenderView at (0,0) size 785x600 layer at (0,0) size 785x4861 RenderBlock {HTML} at (0,0) size 785x4861 [color=#99DDBBCC] [bgcolor=#102030E0] RenderBody {BODY} at (8,8) size 769x4845 - RenderHTMLCanvas {CANVAS} at (0,4395) size 300x150 - RenderImage {IMG} at (300,4543) size 26x2 CRASHLOG:
https://build.webkit.org/results/Apple-Sequoia-Release-WK2-Tests/289002@main%20(1309)/fast/webgpu/nocrash/fuzz-284937-crash-log.txt
No crash log found for com.apple.WebKit.GPU.Development:81192. stdout: stderr: com.apple.WebKit.GPU.Development terminated (pid 81192) for reason: crash
Attachments
Add attachment
proposed patch, testcase, etc.
Radar WebKit Bug Importer
Comment 1
2025-01-16 11:09:37 PST
<
rdar://problem/143065194
>
Robert Jenner
Comment 2
2025-01-16 11:11:02 PST
I can reproduce the timeout easily on macOS Sequoia release WK2 Apple Silicon ToT running the test as follows: run-webkit-tests fast/webgpu/nocrash/fuzz-284937.html I currently don't have an Intel Mac setup to repo the crash.
Robert Jenner
Comment 3
2025-01-16 11:21:39 PST
Pull request:
https://github.com/WebKit/WebKit/pull/39138
EWS
Comment 4
2025-01-16 11:24:00 PST
Test gardening commit
289016@main
(8ce15dbd046c): <
https://commits.webkit.org/289016@main
> Reviewed commits have been landed. Closing PR #39138 and removing active labels.
Ben Schwartz
Comment 5
2025-02-12 16:52:24 PST
This bug is made obsolete by arm64 test gardening changes in
290311@main
. Please file a new bug if issues persist on x86_64 machines.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug