Date: Fri, 15 Jun 2001 18:55:59 +0300 From: "Eli Zaretskii" Sender: halo1 AT zahav DOT net DOT il To: "Chris Wilkinson" Message-Id: <2945-Fri15Jun2001185559+0300-eliz@is.elta.co.il> X-Mailer: Emacs 20.6 (via feedmail 8.3.emacs20_6 I) and Blat ver 1.8.9 CC: djgpp AT delorie DOT com In-reply-to: <9gd57o$717$2@hecate.umd.edu> (chrisw@wam.umd.edu) Subject: Re: Link errors with Debug switch on References: <000801c0f518$4cddd300$5a3e0281 AT umd DOT edu> <9791-Fri15Jun2001104543+0300-eliz AT is DOT elta DOT co DOT il> <9gcu73$3s7$1 AT hecate DOT umd DOT edu> <9gd57o$717$2 AT hecate DOT umd DOT edu> Reply-To: djgpp AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk > From: "Chris Wilkinson" > Newsgroups: comp.os.msdos.djgpp > Date: Fri, 15 Jun 2001 10:14:06 -0400 > > The procedure "ipmpar" has a function prototype "int ipmpar(int*);", and is > called from dcdflib in several places with "b = ipmpar(&Kx);". dcdflib is > also a .c file, compiled and linked. Kx is a static int variable. I don't see anything wrong in ipmpar.c at the source level. What I suggest is to build the program with the -save-temps switch (and with all the other switches you normally compile and link). This will force GCC to leave the temporary files with the funny names like c:/djgpp/tmp/ccN89Gr5.o and c:/djgpp/tmp/ccDvDCh3.o instead of deleting them. (Note that these files are in the `tmp' subdirectory of the main DJGPP installation directory.) The trace from the -v switch will show you what .o file has the object code from ipmpar.o. Then type this from the command line: nm -A c:/djgpp/tmp/ccDvDCh3.o (assuming that this is the name of the object file into which ipmpar.c was compiled). Whatever `nm' prints, post that here. You might also compare the output of `nm' from files produced by compiling with and without -g, in case it gives some clues.