From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752527Ab2DUMVH (ORCPT ); Sat, 21 Apr 2012 08:21:07 -0400 Received: from caramon.arm.linux.org.uk ([78.32.30.218]:51624 "EHLO caramon.arm.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751969Ab2DUMVF (ORCPT ); Sat, 21 Apr 2012 08:21:05 -0400 Date: Sat, 21 Apr 2012 13:20:45 +0100 From: Russell King - ARM Linux To: Arnd Bergmann Cc: linux-arm-kernel@lists.infradead.org, Alexander Shiyan , linux-kernel@vger.kernel.org Subject: Re: [PATCH 02/17] ARM clps711x: Added Cirrus Logic EP7312 CPU variant Message-ID: <20120421122044.GF24211@n2100.arm.linux.org.uk> References: <1334999490-27177-1-git-send-email-shc_work@mail.ru> <201204211121.49610.arnd@arndb.de> <20120421114803.GE24211@n2100.arm.linux.org.uk> <201204211211.36273.arnd@arndb.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201204211211.36273.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 Sat, Apr 21, 2012 at 12:11:35PM +0000, Arnd Bergmann wrote: > On Saturday 21 April 2012, Russell King - ARM Linux wrote: > > That's where we used to put truely hardware dependent stuff, rather than > > machine type stuff when the structure was first implemented. At that time > > we had a number of different mach-* directories wanting to share things > > like the acorn IOC/IOMD stuff, and there's also sharing of definitions > > between CLPS711x and EP93xx. > > > > That was long before we invented the plat-* stuff - but even that doesn't > > actually cover what asm/hardware does - eg, the SA1111 stuff shared > > between SA11x0 and PXA. > > Ah, thanks for the explanation. > > Since the clps7111 based asm/hardware/*.h files are not shared with ep93xx, > I guess it makes sense to move them to mach-clps711x/include/mach/*.h as part of > this cleanup series, right? Yes. > Out of these, only the first one is used in two places outside of > arch/arm/mach-clps711x, the serial and the fb driver, which could > be changed to use mach/hardware.h instead. The headers still have to end up being in an accessible place, and having drivers include the minimum required is a good idea. Merely moving them into include/mach and updating the include statements as far as I'd go. (and I've never liked the idea of mach/hardware.h being an "include the world for this platform" header file.) From mboxrd@z Thu Jan 1 00:00:00 1970 From: linux@arm.linux.org.uk (Russell King - ARM Linux) Date: Sat, 21 Apr 2012 13:20:45 +0100 Subject: [PATCH 02/17] ARM clps711x: Added Cirrus Logic EP7312 CPU variant In-Reply-To: <201204211211.36273.arnd@arndb.de> References: <1334999490-27177-1-git-send-email-shc_work@mail.ru> <201204211121.49610.arnd@arndb.de> <20120421114803.GE24211@n2100.arm.linux.org.uk> <201204211211.36273.arnd@arndb.de> Message-ID: <20120421122044.GF24211@n2100.arm.linux.org.uk> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Sat, Apr 21, 2012 at 12:11:35PM +0000, Arnd Bergmann wrote: > On Saturday 21 April 2012, Russell King - ARM Linux wrote: > > That's where we used to put truely hardware dependent stuff, rather than > > machine type stuff when the structure was first implemented. At that time > > we had a number of different mach-* directories wanting to share things > > like the acorn IOC/IOMD stuff, and there's also sharing of definitions > > between CLPS711x and EP93xx. > > > > That was long before we invented the plat-* stuff - but even that doesn't > > actually cover what asm/hardware does - eg, the SA1111 stuff shared > > between SA11x0 and PXA. > > Ah, thanks for the explanation. > > Since the clps7111 based asm/hardware/*.h files are not shared with ep93xx, > I guess it makes sense to move them to mach-clps711x/include/mach/*.h as part of > this cleanup series, right? Yes. > Out of these, only the first one is used in two places outside of > arch/arm/mach-clps711x, the serial and the fb driver, which could > be changed to use mach/hardware.h instead. The headers still have to end up being in an accessible place, and having drivers include the minimum required is a good idea. Merely moving them into include/mach and updating the include statements as far as I'd go. (and I've never liked the idea of mach/hardware.h being an "include the world for this platform" header file.)