Date: Fri, 16 Feb 2001 08:12:28 +0200 From: "Eli Zaretskii" Sender: halo1 AT zahav DOT net DOT il To: "Matthew Conte" Message-Id: <3405-Fri16Feb2001081227+0200-eliz@is.elta.co.il> X-Mailer: Emacs 20.6 (via feedmail 8.3.emacs20_6 I) and Blat ver 1.8.6 CC: djgpp-workers AT delorie DOT com In-reply-to: <000901c097cf$691ff3a0$8c971918@nycap.rr.com> (zeus@nofrendo.org) Subject: Re: possible objcopy problem. References: <000901c097cf$691ff3a0$8c971918 AT nycap DOT rr DOT com> Reply-To: djgpp-workers AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk > From: "Matthew Conte" > Date: Thu, 15 Feb 2001 23:17:42 -0500 > > No, no I don't. The first two bytes are not $014c, they are $0000. That's > a Bad Thing. And what do you know, changing them to the correct COFF magic > number makes everything work correctly. > > This same problem also happens with sparC and elf32-i386. Didn't you say that it worked for you on some ELF system? I thought it _was_ elf32-i386, no? Anyway, this looks like a simple bug (or maybe misfeature) in Binutils. It should be a relatively simple matter to find in the source of objcopy where does it fail to output the signature. I will try to look into that when I have time, unless someone else does it sooner. > Eli, do you have a "hotter" connection into binutils development that could > raise interest about this bug? No, I don't have any special relationship with them. I'd imagine that, if you told there the COFF signature is missing, that would be specific enough to draw someone's attention.