Bug 229532 – REGRESSION (r281516): [AppleSilicon WK2] fast/loader/reload-zero-byte-plugin.html is timing out

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Ryan Haddad]

fast/loader/reload-zero-byte-plugin.html has started consistently timing out with no output on Big Sur WK2 bots

https://results.webkit.org/?suite=layout-tests&test=fast%2Floader%2Freload-zero-byte-plugin.html

[Ryan Haddad]

Created attachment 436458 [details]
sample from bot

[Radar WebKit Bug Importer]

<rdar://problem/82370692>

[Ryan Haddad]

I cannot reproduce this with a build of r281515, but I can with r281516.

[Ryan Haddad]

This is only occurring on Apple Silicon bots, and my machine is also an AS one.

[Chris Dumez]

Committed r281611 (240967@main): <https://commits.webkit.org/240967@main>

[Chris Dumez]

Reopening since I merely skipped the test for now.

[Chris Dumez]

Created attachment 436462 [details]
WIP patch

Patch for EWS since I don't have an Apple Silicon device handy.

[Ryan Haddad]

(In reply to Chris Dumez from comment #7)
> Created attachment 436462 [details]
> WIP patch
> 
> Patch for EWS since I don't have an Apple Silicon device handy.
This did not resolve the issue for me locally.

[Chris Dumez]

(In reply to Ryan Haddad from comment #8)
> (In reply to Chris Dumez from comment #7)
> > Created attachment 436462 [details]
> > WIP patch
> > 
> > Patch for EWS since I don't have an Apple Silicon device handy.
> This did not resolve the issue for me locally.

Thanks for checking. Mac-AS-debug-wk2 is green so it seems I cannot rely on EWS after all?

[Chris Dumez]

I built with UBSan and ran this test but there was nothing interesting in the output.

[Chris Dumez]

Created attachment 436509 [details]
WIP patch

[Ryan Haddad]

(In reply to Chris Dumez from comment #9)
> (In reply to Ryan Haddad from comment #8)
> > (In reply to Chris Dumez from comment #7)
> > > Created attachment 436462 [details]
> > > WIP patch
> > > 
> > > Patch for EWS since I don't have an Apple Silicon device handy.
> > This did not resolve the issue for me locally.
> 
> Thanks for checking. Mac-AS-debug-wk2 is green so it seems I cannot rely on
> EWS after all?
Hm, I’ll try another build..

[Chris Dumez]

(In reply to Ryan Haddad from comment #12)
> (In reply to Chris Dumez from comment #9)
> > (In reply to Ryan Haddad from comment #8)
> > > (In reply to Chris Dumez from comment #7)
> > > > Created attachment 436462 [details]
> > > > WIP patch
> > > > 
> > > > Patch for EWS since I don't have an Apple Silicon device handy.
> > > This did not resolve the issue for me locally.
> > 
> > Thanks for checking. Mac-AS-debug-wk2 is green so it seems I cannot rely on
> > EWS after all?
> Hm, I’ll try another build..

Would you be able to provide me with the output of:
log stream --system | grep ProcessSwapping

When you run the test?

I am curious if process swapping is actually happening for some reason (it shouldn't in this test).

[Chris Dumez]

Committed r281632 (240988@main): <https://commits.webkit.org/240988@main>

[Chris Dumez]

"Big Sur E wk2 Release on Macmini9,1 with arm64" is green again. Still waiting on "Big Sur E wk2 Debug on Macmini9,1 with arm64" to process the patch but odds are the partial revert worked.