From:	"A. Sinan Unur" <sinan DOT unur AT cornell DOT edu>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: malloc() bug?
Date:	Sun, 30 Nov 1997 05:22:12 -0500
Organization:	Cornell University (http://www.cornell.edu/)
Lines:	26
Sender:	asu1 AT cornell DOT edu (Verified)
Message-ID:	<34813E53.EE1F5F31@cornell.edu>
References:	<A299EE61924 AT brain1 DOT lstm DOT ruhr-uni-bochum DOT de> <65r720$f3t AT freenet-news DOT carleton DOT ca>
NNTP-Posting-Host:	cu-dialup-0024.cit.cornell.edu
Mime-Version:	1.0
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Paul Derbyshire wrote:
> 
> "Tom Demmer" (DEMMER AT brain1 DOT lstm DOT ruhr-uni-bochum DOT de) writes:
> > malloc() is pretty bug free and I never heard of SIGSEGVs inside
> > malloc when the user code was bug free.
> 
> I have.

what is your point. the example below is a case of user code (in this
case quake, _it_ is the user of malloc) having a bug. 
 
> On one 32-meg pentium I run Quake (the original, DJGPP-compiled) in a 
> W95 DOS box with 65535 DPMI memory, -winmem 24, everything is fine. On 
> the other computer with identical memory, I get SIGSEGV on startup at 
> the "mallocing 24,987,654" stage. Difference is, Quake 1.08 on the one 
> and 1.06 on the other (1.06 crashes). I suppose 1.06 might have a bug 
> though that they fixed in 1.08?
 
-- 
----------------------------------------------------------------------
A. Sinan Unur
Department of Policy Analysis and Management, College of Human Ecology,
Cornell University, Ithaca, NY 14853, USA

mailto:sinan DOT unur AT cornell DOT edu
http://www.people.cornell.edu/pages/asu1/

- Raw text -