Message-Id: <328A6F7E.479@bud.cc.swin.edu.au>
Date: Thu, 14 Nov 1996 11:01:50 +1000
From: John Newbigin <newbs AT bud DOT cc DOT swin DOT edu DOT au>
Mime-Version: 1.0
To: djgpp AT delorie DOT com
Subject: Re: Adding Windows Support?
References: <Pine DOT SUN DOT 3 DOT 91 DOT 961103103410 DOT 27986A-100000 AT is> <55mi0g$e6m AT news DOT manawatu DOT gen DOT nz> <32829931 DOT 7EF7 AT ananke DOT amu DOT edu DOT pl>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Mark Habersack wrote:
> 
> Malcolm Taylor wrote:
> 
> > Well, there is a document somewhere around that describes the Win32
> > PE/COFF format. I do not have this anymore, but I did send it to a few
> > people a while ago (can't remember quite who, Bill Currie maybe?).
> I have the document which came with BC 4.52. It is from Micro$oft and
> has some bugs. It is inevitable to have the Win32 SDK header files to
> check the document with structures in them.
> 
> > Basicly it's a _very_ large project waiting for someone to lead it.
> > The project would consist of enabling the generation of PE/COFF format
> > (ie. write a new linker, change LD or make a COFF-GO32 to PE/COFF
> > converter), writing header files, writing startup code, modifying the
> > libc.a to work under Windows (maybe a new lib), finding a freeware
> > resource compiler ...
> I think that changing LD is not necessary. The COFF-PE converter for
> linked executables would be a fairly easy to program stuff. If anything
> might be modified here, it's the BFD library - someone would have to add
> the PE target to it..

I have already been able to create a PE exe from using LD, but I have problems calling DPMI 
functions from the windows code and importing and exporting functions seems to be a challenge but 
should still be possible without changing the compiler or linker.

If anyone wants a copy of what I have got, please e-mail me.

Newbs.