From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner+w=401wt.eu-S1756667AbYEXOkW@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1756667AbYEXOkW (ORCPT <rfc822;w@1wt.eu>);
	Sat, 24 May 2008 10:40:22 -0400
Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753650AbYEXOkK
	(ORCPT <rfc822;linux-kernel-outgoing>);
	Sat, 24 May 2008 10:40:10 -0400
Received: from 1wt.eu ([62.212.114.60]:1037 "EHLO 1wt.eu"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1752976AbYEXOkJ (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Sat, 24 May 2008 10:40:09 -0400
Date: Sat, 24 May 2008 16:39:50 +0200
From: Willy Tarreau <w@1wt.eu>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Sam Ravnborg <sam@ravnborg.org>, "H. Peter Anvin" <hpa@zytor.com>,
       Steve French <smfrench@gmail.com>, lkml <linux-kernel@vger.kernel.org>
Subject: Re: kernel coding style for if ... else which cross #ifdef
Message-ID: <20080524143950.GG6749@1wt.eu>
References: <524f69650805231211r315be4e4u5890aa0f914bcb4f@mail.gmail.com> <48374D3F.1080502@zytor.com> <20080524054301.GA3773@uranus.ravnborg.org> <4837AAE2.9090102@zytor.com> <20080524064201.GA4133@uranus.ravnborg.org> <4837E89D.9040008@goop.org> <20080524112704.GA7292@uranus.ravnborg.org> <483827CE.9080200@goop.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <483827CE.9080200@goop.org>
User-Agent: Mutt/1.5.11
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Sat, May 24, 2008 at 03:35:58PM +0100, Jeremy Fitzhardinge wrote:
> Sam Ravnborg wrote:
> >I assume you wanted to say:
> >  
> >>   #undef  CONFIG_FOO
> >>   #define CFG_FOO   1
> >>   #define CONFIG_FOO_MODULE
> >>   #define CFG_FOO_MODULE 1
> >>    
> >Because then the CONFIG_* is not changed
> >and we do not want to change that.
> >  
> 
> Yeah, I didn't intend to change the meaning of CONFIG_FOO.
> 
> >I'm not fully convinced about:
> >  
> >>   #define CFG_FOO   1
> >>    
> >But on the other hand it is only in odd
> >cases we distingush between built-in and module.
> >So it makes most sense.
> >  
> 
> I think CONFIG_ and CFG_ should be exact parallels, so if CONFIG_FOO is 
> undefined, CFG_FOO should be 0.
> 
> >>Not sure what CFG_* should be for string/numeric options.  Probably "1" 
> >>if the value is defined, "0" if not, with CONFIG_* being the actual 
> >>value (so a CONFIG_ value of 0 is distinguishable from not defined).
> >>    
> >For non-boolean/tristate values we simply skip CFG_ values - thats
> >the most simple approach.
> 
> I suppose, but it might be useful to know whether a constant is present:
> 
> 	if (CFG_THINGY_LIMIT && x > CONFIG_THINGY_LIMIT) {...}
> 
> (which fails if CONFIG_THINGY_LIMIT is undefined, so I guess it still 
> doesn't work very well).

You still have the possibility to use the "-0" trick :

 	if (CFG_THINGY_LIMIT && x > (CONFIG_THINGY_LIMIT-0)) {...}

Willy