From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754800Ab0CZW1x (ORCPT ); Fri, 26 Mar 2010 18:27:53 -0400 Received: from mail-gw0-f46.google.com ([74.125.83.46]:51052 "EHLO mail-gw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754648Ab0CZW1v convert rfc822-to-8bit (ORCPT ); Fri, 26 Mar 2010 18:27:51 -0400 MIME-Version: 1.0 In-Reply-To: <1269642001.3001.325.camel@iscandar.digidescorp.com> References: <1268843188-15292-1-git-send-email-steve@digidescorp.com> <1269642001.3001.325.camel@iscandar.digidescorp.com> From: Grant Likely Date: Fri, 26 Mar 2010 16:27:29 -0600 X-Google-Sender-Auth: d148c8b88ecfc880 Message-ID: Subject: Re: [microblaze-uclinux] Re: [PATCH 3/4] dmaengine: xlldma platform bus driver To: steve@digidescorp.com Cc: Dan Williams , linux-kernel@vger.kernel.org, monstr@monstr.eu, microblaze-uclinux@itee.uq.edu.au Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Mar 26, 2010 at 4:20 PM, Steven J. Magnani wrote: > On Fri, 2010-03-26 at 15:33 -0600, Grant Likely wrote: >> On Wed, Mar 17, 2010 at 10:26 AM, Steven J. Magnani >> wrote: >> > Platform bus attachment for the Xilinx MPMC Soft Direct Memory >> > Access Controller DMA engine. >> > >> > Signed-off-by: Steven J. Magnani >> > --- >> > diff -uprN a/drivers/dma/xlldma_plat.c b/drivers/dma/xlldma_plat.c >> > --- a/drivers/dma/xlldma_plat.c 1969-12-31 18:00:00.000000000 -0600 >> > +++ b/drivers/dma/xlldma_plat.c 2010-03-17 11:11:16.000000000 -0500 >> >> Not really worth putting into a separate file.  Just put this hunk in >> the xlldma.c file to keep a few more symbols out of the global >> namespace. > > I can certainly do that, but as someone who's had to add platform bus > attachments to drivers that only have OF attachments, I can say from > experience that the result ends up being rather messy. I thought > (explicitly) decoupling the bus logic from the driver might be an > improvement. Maybe it's not enough of a benefit. of_platform is geting merged with platform bus so this issue is going away. Just put it all in one file. g.