From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-50783-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 8520 invoked by alias); 26 Apr 2002 17:34:48 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
Received: (qmail 8497 invoked from network); 26 Apr 2002 17:34:46 -0000
Received: from unknown (HELO nondot.org) (64.5.103.85)
  by sources.redhat.com with SMTP; 26 Apr 2002 17:34:46 -0000
Received: by nondot.org (Postfix, from userid 501)
	id A97FB11883; Fri, 26 Apr 2002 12:31:39 -0500 (CDT)
Received: from localhost (localhost [127.0.0.1])
	by nondot.org (Postfix) with ESMTP
	id A530D1183E; Fri, 26 Apr 2002 12:31:39 -0500 (CDT)
Date: Fri, 26 Apr 2002 10:35:00 -0000
From: Chris Lattner <sabre@nondot.org>
To: Magnus Fromreide <magfr@lysator.liu.se>
Cc: Kris Warkentin <kewarken@qnx.com>, <gcc@gcc.gnu.org>
Subject: Re: pure and const functions
In-Reply-To: <Pine.GSO.4.33.0204261927480.10853-100000@proton.lysator.liu.se>
Message-ID: <Pine.LNX.4.30.0204261230020.3264-100000@nondot.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-SW-Source: 2002-04/txt/msg01435.txt.bz2

On Fri, 26 Apr 2002, Magnus Fromreide wrote:

> On Fri, 26 Apr 2002, Chris Lattner wrote:
> > But those presumably would not be pure or const, would they?
>
> Possibly, but given the definitions in the original post fprintf is pure
> and it is also partial since you can't foresee what odd devices there
> might be to write to.

fprintf isn't pure though, it modifies global buffers, calls syscalls, and
otherwise has a lot of side effects...

I wouldn't want:

fprintf(stdout, "hello world\n");

to be eliminated because I don't use the return value of fprintf.  :)

-Chris

http://www.nondot.org/~sabre/os/
http://www.nondot.org/~sabre/Projects/