From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-gx0-f228.google.com (mail-gx0-f228.google.com [209.85.217.228]) by ozlabs.org (Postfix) with ESMTP id 6A49DB7BC5 for ; Thu, 5 Nov 2009 18:55:36 +1100 (EST) Received: by gxk28 with SMTP id 28so1719937gxk.9 for ; Wed, 04 Nov 2009 23:55:35 -0800 (PST) MIME-Version: 1.0 Sender: glikely@secretlab.ca In-Reply-To: <4AF1FD46.9050101@austin.ibm.com> References: <4AF1FB29.50905@austin.ibm.com> <4AF1FD46.9050101@austin.ibm.com> From: Grant Likely Date: Thu, 5 Nov 2009 00:55:15 -0700 Message-ID: Subject: Re: [RFC PATCH 2/5] Merge dynamic OF code to of_dynamic.c To: Nathan Fontenot Content-Type: text/plain; charset=ISO-8859-1 Cc: linuxppc-dev@ozlabs.org, devicetree-discuss@lists.ozlabs.org, microblaze-uclinux@itee.uq.edu.au List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Hi Nathan, Thanks for the patches. Comments below. On Wed, Nov 4, 2009 at 3:16 PM, Nathan Fontenot wrot= e: > Creation of the OF dynamic device tree update code in drivers/of. =A0This > merges the common device tree updating routines to add/remove nodes and > properties from powerpc and microblaze. =A0All of the new code is conditi= onal > based on a new OF_DYNAMIC config option. Rather than one patch to create all the moved functions, and then subsequent patches to remove the duplicates from each arch, I've been using the pattern of one patch for each function or couple of functions to both remove from the old files and add to the new home. Would you be able to do the same for your patches here? The merging is complicated enough without having to track changes to a function between separate commits. Moving one function at a time will also make bisecting more friendly. Also, have you checked what impact these changes have on SPARC? > --- linux-next.orig/drivers/of/Makefile 2009-11-03 11:18:08.000000000 -06= 00 > +++ linux-next/drivers/of/Makefile =A0 =A0 =A02009-11-03 13:42:35.0000000= 00 -0600 > @@ -1,6 +1,7 @@ > obj-y =3D base.o > -obj-$(CONFIG_OF_DEVICE) +=3D device.o platform.o > -obj-$(CONFIG_OF_GPIO) =A0 +=3D gpio.o > -obj-$(CONFIG_OF_I2C) =A0 +=3D of_i2c.o > -obj-$(CONFIG_OF_SPI) =A0 +=3D of_spi.o > -obj-$(CONFIG_OF_MDIO) =A0+=3D of_mdio.o > +obj-$(CONFIG_OF_DEVICE) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0+=3D device.o pla= tform.o > +obj-$(CONFIG_OF_GPIO) =A0 =A0 =A0 =A0 =A0+=3D gpio.o > +obj-$(CONFIG_OF_I2C) =A0 =A0 =A0 =A0 =A0 +=3D of_i2c.o > +obj-$(CONFIG_OF_SPI) =A0 =A0 =A0 =A0 =A0 +=3D of_spi.o > +obj-$(CONFIG_OF_MDIO) =A0 =A0 =A0 =A0 =A0+=3D of_mdio.o > +obj-$(CONFIG_OF_DYNAMIC) =A0 =A0 =A0 +=3D of_dynamic.o Unrelated whitespace churn makes it hard to see what actually changed. Thanks, g. --=20 Grant Likely, B.Sc., P.Eng. Secret Lab Technologies Ltd.