Bug 139357 – [GTK] many layout tests fail since the network process is used for the layout tests.

• Status: RESOLVED
• Resolution: FIXED
• Component: WebKitGTK

[Carlos Alberto Lopez Perez]

Since 175776 <http://trac.webkit.org/175776> the GTK port runs the layout tests using the network process.

Like [40-60] tests are failing since that.

A possible cause can be  (Quoting https://bugs.webkit.org/show_bug.cgi?id=138428#c8 )

> One big reason for these failures is that shared workers are disabled in multiple web process mode.
>

[Carlos Garcia Campos]

(In reply to comment #0)
> Since 175776 <http://trac.webkit.org/175776> the GTK port runs the layout
> tests using the network process.
> 
> Like [40-60] tests are failing since that.

I already fixed most (if not all of them)

> A possible cause can be  (Quoting
> https://bugs.webkit.org/show_bug.cgi?id=138428#c8 )
> 
> > One big reason for these failures is that shared workers are disabled in multiple web process mode.
> >

I skipped those.

[Carlos Alberto Lopez Perez]

Updated the GTK TestExpectations on https://trac.webkit.org/r176922 to refer to this bug number, also added 2 new tests to the list that were related.

[Marcos Chavarría Teijeiro (irc: chavaone)]

These tests are passing after r178310 where both expected files were updated.

js/dom/constructor-length.html 
fast/dom/call-a-constructor-as-a-function.html

Updating the GTK TestExpectations on https://bugs.webkit.org/show_bug.cgi?id=142665

[Marcos Chavarría Teijeiro (irc: chavaone)]

The layout test http/tests/appcache/different-https-origin-resource-main.html is crashing since r181074. 

Updating TestExpectations (bug 143701).

[Philippe Normand]

Only http/tests/media/video-accept-encoding.html fails now. I will aim to unskip it as part of bug 211228