On Fri, Aug 22, 2003 at 06:17:40AM -0400, Russell O'Connor wrote: > Corinna Vinschen writes: > > It's a border case. I don't know if it's worth to wait for the fix. > > It would be nice if I could reproduce it under a debugger :-( > > If you are referring to "malloc returns overlapping buffers", I was > able to get it to fail under the debugger by increasing the malloc > size in the test case from 8000 to 16000. It then segfaults instead > of returning overlapping buffers, but that's presumably a bug too, and > likely a different manifestation of the same bug. > > I know that the stack trace for that case in my original post was not > too useful because there was no debugging information in the cygwin > dll, but I've since built cygwin from CVS and gotten a proper stack > trace. Unfortunately, it's at work and I won't be back there till > Monday. But if it would be useful to you I'll make an extra trip in.
Thanks for the offer but that's probably not necessary. I could already find a workaround (not a solution, though). It looks like you're right with your suspicion that mmap using malloc somehow messes up things but this needs further debugging. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developer mailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/