Bug 1917837

Summary:	JSC stress test stress/shared-array-buffer-sort-while-different-thread-is-modifying.js.default is flaky
Product:	[Fedora] Fedora	Reporter:	Michael Catanzaro <mcatanza>
Component:	webkit2gtk3	Assignee:	Michael Catanzaro <mcatanza>
Status:	CLOSED UPSTREAM	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	medium	Docs Contact:
Priority:	unspecified
Version:	32	CC:	erack, gnome-sig, mcatanza, tpopela
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2021-03-03 13:50:15 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description Michael Catanzaro 2021-01-19 13:54:03 UTC

JSC stress test stress/shared-array-buffer-sort-while-different-thread-is-modifying.js.default is flaky only on ppc64le. It failed on our ppc64le CI on: Jan 19, Jan 8, Jan 5, and Dec 13 (first known failure). The error looks like this:

Running stress/shared-array-buffer-sort-while-different-thread-is-modifying.js.default
stress/shared-array-buffer-sort-while-different-thread-is-modifying.js.default: Exception: JavaScript execution terminated.
stress/shared-array-buffer-sort-while-different-thread-is-modifying.js.default: ERROR: Unexpected exit code: 1
FAIL: stress/shared-array-buffer-sort-while-different-thread-is-modifying.js.default

which doesn't tell us very much.

Comment 1 Michael Catanzaro 2021-01-27 14:57:29 UTC

(In reply to Michael Catanzaro from comment #0)
> JSC stress test
> stress/shared-array-buffer-sort-while-different-thread-is-modifying.js.
> default is flaky only on ppc64le. It failed on our ppc64le CI on: Jan 19,
> Jan 8, Jan 5, and Dec 13 (first known failure).

Well surprise, yesterday (Jan 26) it failed on s390x as well. So the bug is probably not *truly* architecture-specific.

Comment 2 Michael Catanzaro 2021-01-29 14:20:01 UTC

OK, today it failed on both s390x and x86_64. Since it's failing on x86_64, I can report this to upstream. And it's clearly not a ppc64le issue after all, so removing from PPCTracker.

Comment 3 Dan Horák 2021-01-29 14:32:10 UTC

At least a bit good news from my point of view :-) I would suspect some bad timing or depending on how much loaded the worker is (they are all VMs) ...

Comment 4 Michael Catanzaro 2021-03-03 13:50:15 UTC

Possibly fixed by https://trac.webkit.org/changeset/273779/webkit. I'll reopen if I continue seeing test failures.