From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758553Ab0IHJMr (ORCPT ); Wed, 8 Sep 2010 05:12:47 -0400 Received: from opensource.wolfsonmicro.com ([80.75.67.52]:58883 "EHLO opensource2.wolfsonmicro.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752714Ab0IHJMp (ORCPT ); Wed, 8 Sep 2010 05:12:45 -0400 Date: Wed, 8 Sep 2010 10:12:43 +0100 From: Mark Brown To: Jassi Brar Cc: Jassi Brar , Grant Likely , David Brownell , spi-devel-general@lists.sourceforge.net, patches@opensource.wolfsonmicro.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH] spi/spi_s3c64xx: Move to subsys_initcall() Message-ID: <20100908091242.GB31253@rakim.wolfsonmicro.main> References: <1283855357-15179-1-git-send-email-broonie@opensource.wolfsonmicro.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-Cookie: Crazee Edeee, his prices are INSANE!!! User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Sep 08, 2010 at 01:55:39PM +0900, Jassi Brar wrote: > On Tue, Sep 7, 2010 at 7:29 PM, Mark Brown > wrote: > > Allow the use of the S3C64xx SPI controller with things like PMICs by > > moving the init up to subsys_initcall(). > Couldn't any user ever need to load it as a module? > If no, we might as well drop the s3c64xx_spi_exit and s3c64xx_spi_remove This doesn't prevent building as a module - when built as a module subsys_initcall() is identical to module_init(), the change will only affect the order in which things are done when the code is built into the kernel otherwise it's a noop. > as well and save space. Rather going a step further, shouldn't then all > spi drivers be that way? Two steps further, why not every 'bus-driver' ? Yes, we probably do need to do the same thing for all embedded SPI controllers (as has already happened with embedded I2C controllers), though it's less pressing since outside of a few CPUs it's much less common for things like PMICs to be on SPI than I2C.