[Webkit-unassigned] [Bug 241583] Make sure WebPageProxy doesn't leak though strong references in async IPC callbacks

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=241583

--- Comment #15 from Yury Semikhatsky <yurys at chromium.org> ---
(In reply to Michael Catanzaro from comment #14)
> OK... in that case, I think we can close this now. Do you agree, Yury?
> 
The bug with WebPageProxy (and web process) leak was exposed by the reverted patch and now that it has been reverted the original repro might not trigger the issue, but I believe the underlying issue of preventing a process from suspending while it has pending tasks is still there. So I'd keep this bug open to track progress on that.

The other issue mentioned above (if the web process is stopped and the UI processes gets killed the former will leak) can probably be moved into another bug. It might make sense to have an api to disable process suspension logic for the embedders that care about possible leaks (we do).


> Non-Cocoa ports still probably need a more robust ownership model for the
> network process. The Cocoa setup of one singleton process, rather than one
> process per data store, seems superior to me. But improving this would be a
> different problem for a different bug. Agreed?

I may be missing something but it doesn't seem directly related to this bug report and would be a topic of a different bug.

-- 
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/20220817/cc8cc6cb/attachment.htm>