Date: Sun, 2 Apr 2000 08:57:45 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Martin Str|mberg <ams AT ludd DOT luth DOT se>
cc: djgpp-workers AT delorie DOT com
Subject: Re: restrict
In-Reply-To: <200003261110.NAA07744@father.ludd.luth.se>
Message-ID: <Pine.SUN.3.91.1000402085639.8988O-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Sun, 26 Mar 2000, Martin Str|mberg wrote:

> According to Eli Zaretskii:
> > 
> >    How should we handle the new keyword restrict, until GCC supports it?
> > 
> > I imagine you checked this, but I'm surprised: I thought GCC 2.95.x
> > already supported `restrict'.  Perhaps it's a good idea to ask on
> > gnu.gcc.bug.
> 
> Hmm. I see I have 2.8.1. So are we going to require people to use at
> least 2.95?

It doesn't make sense to support C99 features with any version of the
compiler before 2.9X, since they didn't support C99.

If this is only for back-compatibility with older compilers, we could
define `restrict' away based on the value of __GNUC__, if that becomes
an issue.

However, I'm not even sure it's a good idea to have those `restrict'
keywords in the headers, or to compile the library with them being
visible to the compiler: it might introduce bugs or unexpected
misfeatures into user programs.  All we gain in return is some code
efficiency.  Comments?