[Webkit-unassigned] [Bug 183348] REGRESSION(r222772): [GTK][WPE] WebProcess from WebKitGtk+ 2.19.9x SIGSEVs in WebKit::WebProcess::ensureNetworkProcessConnection() at Source/WebKit/WebProcess/WebProcess.cpp:1127

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=183348

--- Comment #36 from Carlos Garcia Campos <cgarcia at igalia.com> ---
(In reply to Alex Christensen from comment #34)
> Comment on attachment 337489 [details]
> Patch
> 
> I think it's a bad idea to add code that's implicitly port-specific like
> this.  Why are you so sure you want to keep r222772?  Why are these sync
> messages failing?  Could they be made more robust?

I can add platform ifdefs instead, but the change is actually required when connection doesn't exit on sync message failure, it's specific to GTK and WPE, because we disable that option. I want to keep r222772 because it has worked fine since we added it (except for this very specific case of creating and destroying a web view very quickly). This particular sync message fails because the UI process closes the connections while the web page is still being initialized, so it can't connect to the network process. I don't see why we need to have the same behavior than Apple here, and we already discussed this in bug #168126. We really need to exit cleanly to clean up resources, while Apple doesn't. Most of those resources are multimedia related, not our fault and not under our control (GStreamer), but we end up leaking then and we can avoid it easily.

-- 
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/20180417/be1791ba/attachment.html>