From xemacs-m  Wed Sep 10 03:56:55 1997
Received: from fjo.nada.kth.se (fjo.nada.kth.se [130.237.225.42])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id DAA16616
	for <xemacs-beta@xemacs.org>; Wed, 10 Sep 1997 03:56:54 -0500 (CDT)
From: fjo@nada.kth.se
Received: from nada.kth.se (localhost [127.0.0.1])
	by fjo.nada.kth.se (8.8.7/8.6.9) with ESMTP
	id KAA00505;
	Wed, 10 Sep 1997 10:56:41 +0200 (MET DST)
Message-Id: <199709100856.KAA00505@fjo.nada.kth.se>
Date: Wed, 10 Sep 1997 10:56:38 +0200 (MET DST)
Subject: Re: xemacs not dying on logout
To: vroonhof@math.ethz.ch
cc: engelen@lucent.com, xemacs-beta@xemacs.org
In-Reply-To: <by202x1rt4.fsf@midget.math.ethz.ch>
MIME-Version: 1.0
Content-Type: TEXT/plain; CHARSET=US-ASCII

On 10 Sep, Jan Vroonhof wrote:
> cvengel@hzsbg01.lucent.com (Chris van Engelen) writes:
> 
>> I have had exactly the same problem with XEmacs under OpenWindows, CDE, fvwm,
>> fvwm2, and fvwm95, from the very first version which I started using (19.12),
>> up until the version of XEmacs I'm currently using (19.14).
> 
> Note that we also saw these problems in these versions but we do NOT
> see them in 19.15. Note also that Fredrik said he already tested some
> patches (probaly equivalent to yours) that didn't help.

Yes, this is the very patch I was referring to, It _is_ a part of
XEmacs-30.3-b18, though with an extra && variable to keep it from being 
default. I've tried this too without much success I'm afraid.

>> int
>> x_IO_error_handler (Display *disp)
>> {
> 
> Fredrik can you breakpoint on 'x_IO_error_handler' and
> io_error_delete_device with gdb to see whether they actually get
> called on exit.

Funny you should mention this. I did some testing the time I tried 
this patch before, and it suggested just as you say, that this piece
of code is never called.

I'll do some more systematic debugging.

> The infamous 19.13-19.14 frame freezing problem got fixed because I
> could reproduce it and was sufficiently annoyed to become a beta
> tester and fix it myself.

I get the point :-)

> P.S. Frederik I was trying to come up with a static binary for you but
> onfurtunately I do not seem to have all the support libraries in
> static form. I'll try again if we cannot resolve this.

Ok, thanks for all help. That goes for Chris van Engelen too.

	/fjo


