From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from nommos.sslcatacombnetworking.com (nommos.sslcatacombnetworking.com [67.18.224.114]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTP id 6F5EA67C7D for ; Wed, 15 Nov 2006 08:05:01 +1100 (EST) In-Reply-To: References: <4EC85F22-99A9-4B0F-987C-25B715235320@kernel.crashing.org> Mime-Version: 1.0 (Apple Message framework v752.2) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <3CC50148-9122-4133-A304-E2723077EE1D@kernel.crashing.org> From: Kumar Gala Subject: Re: [PATCH/RFC (take 3)] linkstation / kurobox support under arch/powerpc Date: Tue, 14 Nov 2006 15:05:11 -0600 To: Guennadi Liakhovetski Cc: linuxppc-dev@ozlabs.org, paulus@samba.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Nov 14, 2006, at 1:09 PM, Guennadi Liakhovetski wrote: > On Tue, 14 Nov 2006, Kumar Gala wrote: > >> Out of interest how are you booting this kernel build & dts on the >> linkstation/kurobox? u-boot or wrapper? > > u-boot. > >>> diff --git a/arch/powerpc/boot/dts/kuroboxHG.dts b/arch/powerpc/ >>> boot/dts/kuroboxHG.dts >>> new file mode 100644 >>> index 0000000..6c76ef6 >>> --- /dev/null >>> +++ b/arch/powerpc/boot/dts/kuroboxHG.dts >> >> would make more sense if this file was called linkstation.dts to >> match the >> Kconfig > > Well, as I wrote in the patch description, there are several NAS > devices > which can be hopefully made to run one kernel with just different > fdts. > The systems are LinkStation-I (I think, there are 2 variants thereof: > HDLAN and HGLAN), Kurobox (normal and HG version), Terastation > (several > variations). You can see them and other non-powerpc devices, e.g., > here: > http://linkstationwiki.net/index.php?title=Main_Page > The kernel patch traditionally was called "linkstation", that's > where the > CONFIG name comes from too. However, the device I have is a KuroboxHG, > therefore the dts file is called kuroboxHG.dts, as it is indeed system > specific. Makes sense? > Ok that makes sense, then KuroboxHG for the dts makes sense (lets just add a comment in the header saying it uses the linkstation kernel config) >>> + dma@80001100 { >>> + linux,phandle = <4100>; >>> + #interrupt-cells = <1>; >>> + #address-cells = <1>; >>> + #size-cells = <1>; >>> + device_type = "dma"; >>> + compatible = "fsl-dma"; >>> + clock-frequency = <0>; >>> + reg = <80001100 24>; >>> + interrupts = <6 0>; >>> + interrupt-parent = <4400>; >>> + }; >>> + >>> + dma@80001200 { >>> + linux,phandle = <4200>; >>> + #interrupt-cells = <1>; >>> + #address-cells = <1>; >>> + #size-cells = <1>; >>> + device_type = "dma"; >>> + compatible = "fsl-dma"; >>> + clock-frequency = <0>; >>> + reg = <80001200 24>; >>> + interrupts = <7 0>; >>> + interrupt-parent = <4400>; >>> + }; >> >> Can we drop the dma nodes? are you using them for something. I'd >> prefer not >> to specify what they look like until we figure out how to handle >> the various >> "features" between the families of FSL PPC's that share the DMA >> controller >> (10x, 824x, 83xx, 85xx) > > Sure, will do. > >>> + i2c@80003000 { >>> + linux,phandle = <4300>; >>> + device_type = "i2c"; >>> + compatible = "fsl-i2c"; >>> + clock-frequency = <0>; >> >> Any reason for adding this clock-frequency property? > > Copy-paste:-) from where? >>> + pic@80040000 { >>> + linux,phandle = <4400>; >>> + #interrupt-cells = <2>; >>> + #address-cells = <0>; >>> + device_type = "open-pic"; >>> + compatible = "chrp,open-pic"; >>> + interrupt-controller; >>> + reg = <80040000 40000>; >>> + clock-frequency = <0>; /* ??? */ >> >> Any reason for adding this clock-frequency property? > > Ditto. > >>> diff --git a/arch/powerpc/configs/kuroboxhg_defconfig b/arch/ >>> powerpc/configs/kuroboxhg_defconfig >>> new file mode 100644 >>> index 0000000..136632f >>> --- /dev/null >>> +++ b/arch/powerpc/configs/kuroboxhg_defconfig >> >> similar file rename comment linkstation_defconfig > > Well, it should eventually be linkstation_defconfig, but so far it > only > has drivers for kuroboxhg, but, perhaps, you're right. As the goal > is to > have 1 kernel for all with one defconfig, I can rename it now and just > hope it will be extended in the future as the patch is used / > tested on > more systems. Will rename, thanks. > > So, I would leave most parts with "linkstation" notation, apart > from the > .dts, will throw unneeded entries from the .dts, rename the > defconfig, but > keep the .dts name. Agree? Agreed, also update the comment/help in the Kconfig to make it clear that Kurobox and KuroboxHG are two different things. The Kurobox(HG) was a bit confusing to me at first glance, and since its help, no reason not be more specific :) - k