Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin AT sources DOT redhat DOT com From: Chris Faylor Date: Tue, 17 Oct 2000 11:48:03 -0400 To: cygwin AT sources DOT redhat DOT com Subject: Re: rxvt SEGV (Win98, 1.1.5s/2000-10-08) Message-ID: <20001017114803.A9581@cygnus.com> Reply-To: cygwin AT sources DOT redhat DOT com Mail-Followup-To: cygwin AT sources DOT redhat DOT com References: <1824-Wed11Oct2000224542+0100-starksb AT ebi DOT ac DOT uk> <20001012003918 DOT A16885 AT cygnus DOT com> <9884-Sat14Oct2000001744+0100-starksb AT ebi DOT ac DOT uk> <20001017032858 DOT 30497 DOT qmail AT lizard DOT curl DOT com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.6i In-Reply-To: <20001017032858.30497.qmail@lizard.curl.com>; from jik@curl.com on Mon, Oct 16, 2000 at 11:28:58PM -0400 On Mon, Oct 16, 2000 at 11:28:58PM -0400, Jonathan Kamens wrote: >(Note: I do not subscribe to the "cygwin" list. If you send a >response to this message, please make sure to CC me at jik AT curl DOT com. >Thank you.) > >> Date: Sat, 14 Oct 2000 00:17:43 +0100 >> From: David Starks-Browning >> >> On Thursday 12 Oct 00, Chris Faylor writes: >> > On Wed, Oct 11, 2000 at 10:45:42PM +0100, David Starks-Browning wrote: >> > >I've run Bub's rxvt-2.7.2-w32c from >> > >on Win98 without problem, under Cygwin-1.1.4. With the 2000-10-08 >> > >snapshot, rxvt now crashes with a stackdump upon exit. Other than >> > >that, it works OK. >> > >> > I've checked in a fix for this. >> >> Thanks Chris, I just checked it against the 2000-10-12 snapshot and >> the problem is gone. > >Alas, it just got worse for us on Windows NT 4.0. The 20001012, >20001013 and 20001014 snapshots all die horrible deaths with multiple >"handle_exceptions" errors when we drop them into place on top of >Cygwin 1.1.4 and try to run builds of our product, which use a GNU >autoconf configure script and GNU make. > >I suspect that the fact that we run our builds on multiprocessor >machines has a lot to do with this. > >What can we do to help get this problem debugged and fixed? You can help by dropping the "horrible deaths" hyperbole and provide the actual output that you're seeing. I appreciate that you are willing to help, but I don't understand why you didn't take at least an initial stab at providing some details to your problem. If you're really motivated, you can build cygwin yourself and set CYGWIN=error_start=x:/path/to/gdb.exe . This will cause cygwin to pop up a gdb window when it encounters an error. If you then issue the following commands: thread 1 bt It should be a little clearer what is going on. Barring that, strace output would also probably help pinpoint the problem: strace -om:/tmp/strace.out make (this will probably be a big file) Please do this with the most recent snapshot. That would be 2000-10-16. My main development machine is 300MHZ P2 SMP, so it's not just as simple as this being an SMP problem. You can also provide 'cygcheck -r -s -v' output. cgf -- Want to unsubscribe from this list? Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com