[Webkit-unassigned] [Bug 186638] [GTK] Tests that are failing on the EWS test queue

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Jun 18 07:51:26 PDT 2018


Adrian Perez <aperez at igalia.com> changed:

           What    |Removed                     |Added
                 CC|                            |aperez at igalia.com

--- Comment #16 from Adrian Perez <aperez at igalia.com> ---
(In reply to Michael Catanzaro from comment #14)
> (In reply to Carlos Alberto Lopez Perez from comment #11)
> > (In reply to Carlos Alberto Lopez Perez from comment #10)
> > 
> > > Regressions: Unexpected crashes (3)
> > >   fast/hidpi/broken-image-icon-very-hidpi.html [ Crash ]
> > >   http/tests/misc/cached-scripts.html [ Crash ]
> > >   legacy-animation-engine/fast/css/getFloatValueForUnit.html [ Crash ]
> Yes, as expected. I was planning to watch the EWS this weekend to try to
> find backtraces for the crashing tests, but it was broken all weekend.
> Anyway, I intend to help with this. I don't think it's too much effort to
> take the backtrace and report a bug when EWS finds a crashing test.
> > Crash log for fast/hidpi/broken-image-icon-very-hidpi.html obtained on the
> > EWS bot: http://termbin.com/a4of
> Weird... although you should probably install gnome-icon-theme, I think this
> is a WebKit bug worth reporting. Our current code assumes that we have
> missingImage icons in the GResource, which is apparently not true. See here:
> static Ref<Image> loadMissingImageIconFromTheme(const char* name)
> {
>     // ...
>     if (iconInfo) {
>         // Is the icon installed?
>         return WTFMove(icon);
>     }
>     return loadImageFromGResource(name);
> }
> That's not right, we shouldn't continue on to loadImageFromGResource if the
> image is not guaranteed to be in the resource.

I think that WebKitGTK+ should include within itself a copy of the
resources it uses which currently end up being loaded from the GNOME
theme, to avoid the runtime dependency on the “gnome-icon-theme”
package. Using icons which are part of GTK+ itself would still be
fine, of course. Should we open a new bug for this?

You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20180618/95c4a1dc/attachment.html>

More information about the webkit-unassigned mailing list