GNU bug report logs - #56487
xgselect race condition leading to abort when USE_GTK not defined

Previous Next

Full log

Message #51 received at 56487 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefankangas <at> gmail.com>
To: Tom Gillespie <tgbugs <at> gmail.com>
Cc: luangruo <at> yahoo.com, Eli Zaretskii <eliz <at> gnu.org>, 56487 <at> debbugs.gnu.org
Subject: Re: bug#56487: xgselect race condition leading to abort when USE_GTK
 not defined
Date: Tue, 25 Feb 2025 23:47:41 +0000

Tom Gillespie <tgbugs <at> gmail.com> writes:

>> Ping.  Any news here?
>
> I can confirm that the issue is still present in 30.0.93
> and that the original patch still works against 30.0.93.
>
> Unfortunately I have not had a chance to dig deeper
> or to split out a smaller patch for release_select_lock.
>
> Looking at the original patch again I assume the
> portion that would be useful is the part dealing with
> __atomic_sub_fetch and threads_holding_glib_lock?

AFAIU, yes.  IOW, the changes in release_select_lock.

> I will have to refresh my memory about the expected
> behavior for the release_select_lock changes because
> I don't think I had a test case that was able to differentiate
> that change independent of the various GTK related changes.

OK, thanks.  Please let us know when you make any progress.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.