From: sandmann AT clio DOT rice DOT edu (Charles Sandmann) Message-Id: <10001131426.AA17375@clio.rice.edu> Subject: Re: upgrade chaos To: eliz AT is DOT elta DOT co DOT il (Eli Zaretskii) Date: Thu, 13 Jan 2000 08:26:23 -0600 (CST) Cc: djgpp AT delorie DOT com In-Reply-To: from "Eli Zaretskii" at Jan 13, 2000 09:26:14 AM X-Mailer: ELM [version 2.5 PL2] Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: djgpp AT delorie DOT com Errors-To: dj-admin AT delorie DOT com X-Mailing-List: djgpp AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk > Since in this case, the offending program is GCC, I usually tend to > assign the blame for such intermittent failures to faulty hardware > rather than to GCC's code. Since most platforms dzero memory used for the first time for security reasons - and the DPMI provider is free not to - you will see interesting non-reproducible bugs in DJGPP which are not seen elsewhere. There is a crt0 flag which allows you to zero the memory (which should do the stack too) and allows "fixing" these bugs by making the DJGPP environment look even more unix-like. Given the size and complexity of GCC, I would be surprised if there were not bugs like this in the code.