From owner-freebsd-current  Mon Sep  4 22:34:18 1995
Return-Path: current-owner
Received: (from majordom@localhost)
          by freefall.freebsd.org (8.6.11/8.6.6) id WAA02936
          for current-outgoing; Mon, 4 Sep 1995 22:34:18 -0700
Received: from who.cdrom.com (who.cdrom.com [192.216.222.3])
          by freefall.freebsd.org (8.6.11/8.6.6) with ESMTP id WAA02817
          ; Mon, 4 Sep 1995 22:33:16 -0700
Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211])
          by who.cdrom.com (8.6.11/8.6.11) with ESMTP id LAA13989
          ; Mon, 4 Sep 1995 11:11:19 -0700
Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA21773; Mon, 4 Sep 1995 11:04:41 -0700
From: Terry Lambert <terry@lambert.org>
Message-Id: <199509041804.LAA21773@phaeton.artisoft.com>
Subject: Re: eBones Changes coming soon!
To: bde@zeta.org.au (Bruce Evans)
Date: Mon, 4 Sep 1995 11:04:41 -0700 (MST)
Cc: mal@algonet.se, terry@lambert.org, committers@FreeBSD.org,
        current@FreeBSD.org, mark@grondar.za
In-Reply-To: <199509040805.SAA14279@godzilla.zeta.org.au> from "Bruce Evans" at Sep 4, 95 06:05:11 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1277      
Sender: current-owner@FreeBSD.org
Precedence: bulk

> >The extern declarations have (and should have) __P(( int a, char *b))
> >wrappers on them so that an ANSI compiler can do type checking that
> >would be in the linker if anyone could build a linker with any brains
> >(besides Fred Fish's M88k one, I mean).
> 
> An ANSI compiler must also do type conversions of the args if a
> prototype is in scope, and it can use a parameter passing convention
> better suited to the arg types.

It can do this if there is a prototype in scope; the declaration itself
doesn't need it.  Yeah, this means you need to be careful to include the
prototype containing header in the file declaring the function, but
this is hardly a hardship.

> >The only valid argument for prototypes in the function declarations
> >themselves is so you can compile the kernel with C++, which is not
> 
> The following arguments are sometimes given:
> 
> 1) For static functions, it the prototype need not be declared separately
>    from the function if the function is defined before it is referenced.
> 2) It is possible to pass chars and shorts without promotion.
> 
> These are actually counter-arguments :-).

8-).


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.