Date: Fri, 21 Apr 2000 03:37:51 -0400 (EDT)
Message-Id: <200004210737.DAA16966@indy.delorie.com>
From: Eli Zaretskii <eliz AT delorie DOT com>
To: Andrew Hakman <hakmana AT hotmail DOT com>
CC: djgpp AT delorie DOT com
In-reply-to: <38FFA0BC.FF6986A8@hotmail.com> (message from Andrew Hakman on
	Fri, 21 Apr 2000 00:29:28 GMT)
Subject: Re: 387 Emulation
References:  <38FFA0BC DOT FF6986A8 AT hotmail DOT com>
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: Andrew Hakman <hakmana AT hotmail DOT com>
> Newsgroups: comp.os.msdos.djgpp
> Date: Fri, 21 Apr 2000 00:29:28 GMT
> 
> [reposted, since I don't think anyone ever saw it!]

I did see it, but didn't reply, since I didn't think you were asking
something.

> Yes, I did disable the q87 emulator from the autoexec.bat before trying
> the version with wmemu.a lnked in.
> 
> By print them, do you mean cout<<?

Yes.

> If so, have tried that. Have couted
> the individual parts of the assignment. Doesn't matter which one I cout
> first, get page fault on either (how could BOTH variables be at an
> invalid address? this is weird!)

I agree that this is weird.  Perhaps there's a bug somewhere in the
DJGPP wrapper functions which makes them incompatible with emulation,
but I cannot see any such bug.

It might be a good idea to try to write a test program that uses
_go32_* wrappers for real-mode callback, but for some interrupt that's
different from what your present program used, and also uses FP
computations in the foreground, and see if that program, too, crashes
on an FPU-less machine.  That would tell us whether the DJGPP library
might be a culprit here.