From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from az33egw02.freescale.net (az33egw02.freescale.net [192.88.158.103]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client CN "az33egw02.freescale.net", Issuer "Thawte Premium Server CA" (verified OK)) by ozlabs.org (Postfix) with ESMTP id 13A78DDED7 for ; Wed, 9 Jan 2008 05:01:40 +1100 (EST) Message-ID: <4783BA6E.50204@freescale.com> Date: Tue, 08 Jan 2008 11:01:18 -0700 From: John Rigby MIME-Version: 1.0 To: Olof Johansson Subject: Re: [PATCH 3/7] Basic Freescale MPC512x support References: <1199808093-15929-1-git-send-email-jrigby@freescale.com> <1199808093-15929-2-git-send-email-jrigby@freescale.com> <1199808093-15929-3-git-send-email-jrigby@freescale.com> <1199808093-15929-4-git-send-email-jrigby@freescale.com> <20080108174038.GA20819@lixom.net> In-Reply-To: <20080108174038.GA20819@lixom.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Cc: linuxppc-dev@ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Olof Johansson wrote: > Hi, > > > >> diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig >> index 66a3d8c..81c3f05 100644 >> --- a/arch/powerpc/Kconfig >> +++ b/arch/powerpc/Kconfig >> @@ -470,7 +470,7 @@ config PCI >> bool "PCI support" if 40x || CPM2 || PPC_83xx || PPC_85xx || PPC_86xx \ >> || PPC_MPC52xx || (EMBEDDED && (PPC_PSERIES || PPC_ISERIES)) \ >> || PPC_PS3 >> - default y if !40x && !CPM2 && !8xx && !PPC_83xx \ >> + default y if !40x && !CPM2 && !8xx && !PPC_512x && !PPC_83xx \ >> && !PPC_85xx && !PPC_86xx >> > > This is getting out of control. Not a comment to this specific patch, > but it's getting silly. > > Btw, why no PCI by default on this platform when it seemingly is default > on 5200? I thought they were fairly similar. > PCI has issues in first silicon so I left that out. Also contrary to popular rumor, 5121 is more like a 83xx that 52xx. The PCI and USB are very similar to 83xx. > >> diff --git a/arch/powerpc/platforms/512x/Kconfig b/arch/powerpc/platforms/512x/Kconfig >> new file mode 100644 >> index 0000000..399d279 >> --- /dev/null >> +++ b/arch/powerpc/platforms/512x/Kconfig >> @@ -0,0 +1,20 @@ >> +menu "Platform support" >> + depends on PPC_512x >> + >> +choice >> + prompt "Machine Type" >> + default MPC5121_ADS >> + >> +config MPC5121_ADS >> + bool "Freescale MPC5121E ADS" >> + select DEFAULT_UIMAGE >> + help >> + This option enables support for the MPC5121E ADS board. >> + >> +endchoice >> > > For new platforms it makes more sense to make it possible to build > multiplatform kernels instead of using 'choice'. > > I know some embedded board guys prefer to build a custom kernel for just > their board (since it means extra revenue for them to spin a kernel > for every customer), but it's still convenient to be able to build > multiplatform kernels. Especially since there's some companies out there > with intent to make end-user products with this cpu. > Grant made the same comment about multiplatform support. I'll change this. > >> @@ -0,0 +1,90 @@ >> +#include >> +#include >> +#include >> +#include >> +#include >> +#include >> +#include >> > > Stephen will tell you to include linux/of_platform.h instead. :-) > Oops, this patch was originally done before the open firmware unification. I guess checkpatch.pl needs an update too. > >> @@ -229,7 +229,7 @@ config NR_CPUS >> >> config NOT_COHERENT_CACHE >> bool >> - depends on 4xx || 8xx || E200 >> + depends on 4xx || 8xx || E200 || PPC_512x >> default y >> > > Why do you need this, when 5200 doesn't? > > > -Olof > >