From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932748Ab2HVVnm (ORCPT ); Wed, 22 Aug 2012 17:43:42 -0400 Received: from caramon.arm.linux.org.uk ([78.32.30.218]:43404 "EHLO caramon.arm.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932369Ab2HVVnk (ORCPT ); Wed, 22 Aug 2012 17:43:40 -0400 Date: Wed, 22 Aug 2012 22:43:23 +0100 From: Russell King - ARM Linux To: Arnd Bergmann Cc: linux-arm-kernel@lists.infradead.org, linaro-kernel@lists.linaro.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 4/4] [RFC] ARM: treewide: manually change more mach-*/*.h includes Message-ID: <20120822214323.GA12255@n2100.arm.linux.org.uk> References: <201208221253.07278.arnd@arndb.de> <201208221301.19439.arnd@arndb.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201208221301.19439.arnd@arndb.de> User-Agent: Mutt/1.5.19 (2009-01-05) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Aug 22, 2012 at 01:01:19PM +0000, Arnd Bergmann wrote: > These are lots of device drivers that include machine > specific header files from ARM platforms and that are > not easily scriptable. The changes have been found through > manual inspection and should cause no visible changes > because of the build script that maps the old names to the > new ones. And you end up missing a bunch of drivers in the process, such as the sa1100 RTC driver which is shared between sa11x0 and PXA. I wonder how many of the Samsung platforms get broken by this as well... In any case, what we _should_ be doing here as well is moving the headers included by drivers for platform data out of the arch/arm/mach/ subtrees and into include/linux/platform_data. That will substantially reduce the amount of mach/ headers to deal with. Where I want this to get to is not a persistence of the existing crappy situation where platforms stuff platform data definitions into arch/arm, but instead put them in the right place. That means providing them with a good reason why the existing solution won't work, and not allowing mach/ includes in drivers is a very good way to achieve that.