Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!cs.utexas.edu!uunet!mcsun!ukc!cam-cl!news
From: acn@cl.cam.ac.uk (Arthur Norman)
Newsgroups: comp.sys.ibm.pc.misc
Subject: Re: Wanted: GCC and RCS binaries
Summary: see lancs.pdsoft
Message-ID: <1991Mar27.110146.15500@cl.cam.ac.uk>
Date: 27 Mar 91 11:01:46 GMT
Sender: news@cl.cam.ac.uk (The news facility)
Organization: U of Cambridge Computer Lab, UK
Lines: 22

Expires: 
References: <1991Mar20.135146.1311@bnr.ca> <7056@spdcc.SPDCC.COM>
Sender: 
Reply-To: acn@cl.cam.ac.uk (Arthur Norman)
Followup-To: 
Distribution: 
Organization: U of Cambridge Comp Lab, UK
Keywords: 


There is a 386 GCC set of binaries with a DOS extender included
in the lancs.pdsoft (national public domain software archive).  It seems
to REQUIRE a 387 if you have any floating point values in the code you compile
as well as if your code uses FP to run.  The DOS extender (at least the version
installed there) is not compatible with XMS/VCPI etc., but by removing himem.sys
from my machine it worked OK.  The DOS extender is subject to FSF rules if you
use it with GCC, but otherwise attracts a royalty.  The position about
distributing code built with this system looks OK if your code accepts the FSF
manifesto, but may otherwise need thought. My code compiled and ran almost first
time I tried it using this compiler, but I only found GCC/386 last week.  Oh yes,
the DOS extender provides virtual memory for you, so it can run big programs
on a small 386 too.