Date: Mon, 2 Nov 1998 13:12:14 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Martin Stromberg <Martin DOT Stromberg AT lu DOT erisoft DOT se>
cc: DJGPP-WORKERS <djgpp-workers AT delorie DOT com>
Subject: Re: specs file
In-Reply-To: <199811021033.LAA03038@mars.lu.erisoft.se>
Message-ID: <Pine.SUN.3.91.981102130737.11877B-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com


On Mon, 2 Nov 1998, Martin Stromberg wrote:

> I'm not sure I understand what the problem is, but is it that lib/specs
> defines __DJGPP_MINOR et al. while gcc 2.8.* doesn't and instead uses 
> lib/gcc-*/specs?

I think GCC 2.8.1 *does* define it.  The problem is that, since GCC and 
libc.a can now be released independently, you will have to edit specs to 
change __DJGPP__ to reflect the new version.  I think it is a bad idea to 
tell people to edit specs.

> One solution would be if lib/gcc-*/specs included lib/specs if there was 
> such a feature.

AFAIK, there is no such feature in the current specs syntax.

> On the other hand if a program needs to know __DJGPP_MINOR and such, it
> really should include some header (version.h?) for that.

ANSI C allows for a program to be written that doesn't include any 
headers, and still uses library functions.  The problem is how to make
__DJGPP__ and __DJGPP_MINOR__ be defined in such a program.  For 
exxample, i386 and __MSDOS__ *will* be defined in such a program.