From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756512AbZE0UnR (ORCPT ); Wed, 27 May 2009 16:43:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754471AbZE0UnA (ORCPT ); Wed, 27 May 2009 16:43:00 -0400 Received: from 42.mail-out.ovh.net ([213.251.189.42]:56840 "HELO 42.mail-out.ovh.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1754600AbZE0Um7 (ORCPT ); Wed, 27 May 2009 16:42:59 -0400 Date: Wed, 27 May 2009 22:31:15 +0200 From: Jean-Christophe PLAGNIOL-VILLARD To: David Miller Cc: linux@arm.linux.org.uk, jonsmirl@gmail.com, scottwood@freescale.com, jacmet@sunsite.dk, r.schwebel@pengutronix.de, devicetree-discuss@ozlabs.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.arm.linux.org.uk, yuan-bo.ye@motorola.com, timur@freescale.com Subject: Re: [RFC] [PATCH] Device Tree on ARM platform Message-ID: <20090527203115.GB16219@game.jcrosoft.org> References: <9e4733910905271213k7f4b93e7i7e6f2af24d85f@mail.gmail.com> <20090527192116.GA31110@n2100.arm.linux.org.uk> <20090527193927.GD30039@game.jcrosoft.org> <20090527.132827.65331718.davem@davemloft.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090527.132827.65331718.davem@davemloft.net> User-Agent: Mutt/1.5.18 (2008-05-17) X-Ovh-Tracer-Id: 11942701787086564133 X-Ovh-Remote: 213.251.161.87 (ns32433.ovh.net) X-Ovh-Local: 213.186.33.20 (ns0.ovh.net) X-Spam-Check: DONE|H 0.5/N Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 13:28 Wed 27 May , David Miller wrote: > From: Jean-Christophe PLAGNIOL-VILLARD > Date: Wed, 27 May 2009 21:39:27 +0200 > > > when you have to support the same driver for non OF and OF platform it's > > really a pain in the ass > > We could embed the platform_device object into the of_platform_device > structure, it's just an implementation issue. > > But more to the point, I can tell you that anything platform_device > does can be implemented in terms of of_platform_device but the > opposite is not true. This is one of the problem I face not so long ago when I want to the boot count limit functionnality which is only availlble for of platform That's why I think having a simplest way to have driver availlable on the twice platform is important. It will be a win-win at the end Best Regards, J.