Date: Thu, 5 Jun 1997 08:35:03 -0400 (EDT)
Message-Id: <199706051235.IAA14788@delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: eliz AT is DOT elta DOT co DOT il
CC: peter AT mcs DOT nl, djgpp AT delorie DOT com
In-reply-to: <Pine.SUN.3.91.970604094652.2271E-100000@is> (message from Eli
	Zaretskii on Wed, 4 Jun 1997 09:47:31 +0300 (IDT))
Subject: Re: Why COFF ??
Precedence: bulk


> From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
> On Tue, 3 Jun 1997, Peter Zijlstra wrote:
> > Why does the DJGPP port of GNU CC use the COFF object format ?
> 
> The reason for this is mostly historical: the way that DJGPP started
> (from a hacked Unix libc that used COFF) and evolved, it was easier to
> keep compatibility between old and new tools by using the same object
> file format.

Actually, the first DJGPP was a.out format.  The switch to COFF
happened with the support for DPMI, because a.out format had data
starting at the 4Mb mark (not configurable) and many DPMI hosts barfed
when we asked for 4+ Mb.

COFF format was the next best choice then, and it allows the data to
start right after the code, so we switched.