[Webkit-unassigned] [Bug 138468] [GTK] WebProcess crashes when quickly attempting many DnD operations

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=138468

--- Comment #36 from Mario Sanchez Prada <mario at webkit.org> ---
(In reply to comment #35)
> [...]
> It's kind of weird, though, that the ondragend callback is not the only
> thing failing on the Mac now: when I had the timeout I could also see the
> prints from every call to runNextStep(), which are gone now as well.

Looks like the issue in the mac-wk2 is expected after all, as I can see a whole lot of tests being skipped already in the TestExpectations file for that platform, with a reference to bug 42194 ("WebKitTestRunner needs a more-complete implementation of eventSender"), which explains the timeout.

As for the failures still reported for the other mac bots, they seem to happen due to the last print happening out of order, which likely means we will still a timeout in there, even if an small one. So, I'm going to try a new patch introducing zero-seconds timeouts to simulate an on_idle() callback, as it's done in other drag-and-drop tests. Let's see if that helps...

I'll also add the test to the mac-wk2 TestExpectations file now, with a reference to bug 42194.

> Anyway, I agree it's probably not the most efficient use of time to keep
> fighting the EWS, but still I'd like to give it one more shot before adding
> it to the TestExpectations Mac file and calling it a day: tonight I'm hoping
> to do some investigation on my mac mini, see if I can figure out what's
> wrong. If after that I still have no clue, then I'll do what you just said.

JFTR, I've reduced the new layout test to remove the WKT/DRT specific stuff and run it manually on my Mac mini, and everything worked as expected, which kind of confirms this is an issue with the helper tools for testing, and not with WK or the test itself.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20160125/2d678d0b/attachment.html>