Bug 207060 – [ Mojave wk2 Release ] imported/w3c/web-platform-tests/html/webappapis/timers/type-long-setinterval.html flaky failure

• Status: RESOLVED
• Resolution: DUPLICATE
• Component: WebCore Misc.

[Jason Lawrence]

imported/w3c/web-platform-tests/html/webappapis/timers/type-long-setinterval.html

Description:
This test is lightly flaky failing on Mojave wk2 release.

History:
https://results.webkit.org/?suite=layout-tests&test=imported%2Fw3c%2Fweb-platform-tests%2Fhtml%2Fwebappapis%2Ftimers%2Ftype-long-setinterval.html&limit=50000&platform=mac

Diff:
--- /Volumes/Data/slave/mojave-release-tests-wk2/build/layout-test-results/imported/w3c/web-platform-tests/html/webappapis/timers/type-long-setinterval-expected.txt
+++ /Volumes/Data/slave/mojave-release-tests-wk2/build/layout-test-results/imported/w3c/web-platform-tests/html/webappapis/timers/type-long-setinterval-actual.txt
@@ -1,3 +1,4 @@
+CONSOLE MESSAGE: line 3463: Error: assert_unreached: Reached unreachable code
 
 PASS Type long timeout for setInterval

[Radar WebKit Bug Importer]

<rdar://problem/59068308>

[Jason Lawrence]

Created attachment 389386 [details]
Update Test Expectations

[Truitt Savell]

Comment on attachment 389386 [details]
Update Test Expectations

Clearing flags on attachment: 389386

Committed r255514: <https://trac.webkit.org/changeset/255514>

[Alexey Proskuryakov]

> interval = setInterval(next, Math.pow(2, 32));
> setTimeout(assert_unreached, 100);

I think that this tests a zero delay interval vs. a 100 ms delay timeout. Are these supposed to be on the same timeline?

[Alexey Proskuryakov]

I guess clamping makes it test something different from what the author intended, but flakiness still seems surprising to me.

[Darin Adler]

(In reply to Alexey Proskuryakov from comment #4)
> > interval = setInterval(next, Math.pow(2, 32));
> > setTimeout(assert_unreached, 100);
> 
> I think that this tests a zero delay interval vs. a 100 ms delay timeout.
> Are these supposed to be on the same timeline?

Yes, setInterval and setTimeout should be on the same timeline.

[Ryan Haddad]

*** This bug has been marked as a duplicate of bug 172052 ***