[Webkit-unassigned] [Bug 184391] [GTK] Web pages become blank/white

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=184391

Michael Catanzaro <mcatanzaro at igalia.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|Web pages become            |[GTK] Web pages become
                   |blank/white                 |blank/white

--- Comment #3 from Michael Catanzaro <mcatanzaro at igalia.com> ---
(In reply to li from comment #2)
> I've been affected by this bug since I upgraded webkitgtk from 2.18.6 to
> 2.20.0, and I can confirm that it hasn't been fixed.

Oh wow, I thought this was a regression in 2.21.1, since I don't remember seeing this issue until very recently, after I upgraded from 2.20.0.

(In reply to li from comment #2)
> I tried to generate a
> core dump, but it seems there isn't any crash. Maybe I'm wrong, I attached
> one gdb to each webkit process with gdb -p PID, and tried to reproduce the
> bug, when a webview became blank, all webkit processes were still running.  

Good job, then you know there's indeed no crash. (Of course, it's much easier to just check coredumpctl if you use a distro where that's enabled.)

I don't have any network process crashes either, so my original guess was wrong.

> Some steps that I found that may reproduce this bug: open more than one
> webviews/tabs all sharing one origin (in my case, they were all opened with
> javascript window.open), when closing one view, one of the other view that
> gets focus becomes blank.

Those are "related" views, then (two views that share the same web process). That seems really similar to bug #183199. But that's marked as a duplicate of bug #184040, which is a crash, and you've already determined there's no crash here. Furthermore, the crash fixed in bug #184040 occurs only "when using single shared process model or when the process limit is reached in multiple process model." I'm experiencing this bug in multiple secondary process mode without any process count limit.

There's one more issue: I see web pages becoming white *after* they had previously loaded content. That seems different than bug #183199.

-- 
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/20180508/96fe4320/attachment.html>