From: Endlisnis <s257m AT unb DOT ca>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Bug in GCC?
Date: Mon, 07 Sep 1998 18:36:00 -0300
Organization: NBTel Internet
Lines: 27
Message-ID: <35F451C0.717F8D94@unb.ca>
References: <35F296C8 DOT EE7B49C7 AT unb DOT ca> <35F2BF67 DOT 5EE4EF08 AT cartsys DOT com>
NNTP-Posting-Host: fctnts05c61.nbnet.nb.ca
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

Nate Eldredge wrote:
> > The program doesn't compile as I would expect.  All it was spose to do is
> > show how you could use inline asm to add 2 integers.  But, the return value
> > was always 4.  So I looked at the asm code generated for the function, it
> > reads from port 0x83 and does some loops and stuff.  All I wanted was 1 add.
> >         And the debugger (when the asm window is open) can't debug it right.  It
> > skips lines and weird things happen, registers change values without any
> > obvious reason.
> That is extraordinarily bizarre.  Could you:
> 
> * Find out *exactly* what switches RHIDE invokes GCC with.
gcc   -g   -Wall -m486     -c t2.cc -o t2.o
	I tried removing "-m486" and it worked perfectly.  It appears as though the
'-m486' switch sometimes makes programs 'wrong'.

> * Post the complete asm produced.
	I don't know how to do that.  I use RHIDE, I tried making the targetname
"test.s" and it did not work, stated: "Don't know how to build test.s from
(null)".  I also tried "test.S" with the same result.

-- 
     (\/) Endlisnis (\/)
          s257m AT unb DOT ca
          Endlisnis AT GeoCities DOT com
          Endlis AT nbnet DOT nb DOT ca