[Webkit-unassigned] [Bug 230556] REGRESSION(241918 at main): [WPE][GTK] New test is timing out on bots

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=230556

--- Comment #1 from Michael Catanzaro <mcatanzaro at gnome.org> ---
(In reply to Michael Catanzaro from comment #0)
> Unfortunately the new test added in 241918 at main is timing out on the bots. :/
> 
> It works perfectly fine for me locally, so I cannot do much about this.

Ah wait, this is wrong. I got confused:

 * In the last version of *my* patch from https://bugs.webkit.org/show_bug.cgi?id=229116#c26, the test was timing out on the EWS, but worked for me locally. 
 * But in the patch that Alex landed, it cannot possibly work because WebKitWebContext is still using only the WebURLSchemeTask identifier in its map. That's not enough, because the task identifier is not unique in the web process.

IMO we should change the WebURLSchemeTask::identifier to return std::pair<WebCore::ResourceLoaderIdentifier, WebPageProxyID>, since that is now the *real* identifier type that the UI process must use everywhere.

After making this change, it's possible the test will still fail on the bots. We'll just have to find out.

-- 
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/20210921/7e6d9063/attachment.htm>