GNU bug report logs - #16093
24.3.50; Crash of emacs --daemon on killing X server

Previous Next

Package: emacs;

Reported by: Nix <nix <at> esperi.org.uk>

Date: Mon, 9 Dec 2013 15:59:02 UTC

Severity: normal

Found in version 24.3.50

Fixed in version 24.4

Done: Glenn Morris <rgm <at> gnu.org>

Bug is archived. No further changes may be made.

Full log


View this message in rfc822 format

From: Nix <nix <at> esperi.org.uk>
To: Dmitry Antipov <dmantipov <at> yandex.ru>
Cc: 16093 <at> debbugs.gnu.org, Jan Djärv <jan.h.d <at> swipnet.se>
Subject: bug#16093: 24.3.50; Crash of emacs --daemon on killing X server
Date: Tue, 10 Dec 2013 23:39:33 +0000
On 9 Dec 2013, Dmitry Antipov stated:

> On 12/09/2013 07:58 PM, Nix wrote:
>> It is fairly clear that we should not be calling xftfont_close() if the
>> X server is not there! Why has this started happening?
>
> Hm... IIUC there is no convenient way to check whether X display connection
> is alive. But we can do some tricks, for example, try to select() or poll()
> ConnectionNumber() - if there is no EBADF, X server is hopefully present...

Well, it's never happened before. Something has clearly changed... maybe
we're just cleaning up fonts much more lazily than before? (An ugly hack
to "fix" it in that case might just be to trigger a garbage-collect when
the last frame on a given terminal goes away.)

The Xlib error handler should really be being called, which suggests
this is a bug in Xft as well: I'll chase that up. (But that can't be the
cause of this: I last updated my Xft a year and a half ago.)

-- 
NULL && (void)




This bug report was last modified 11 years and 192 days ago.

Previous Next


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