From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from an-out-0708.google.com (an-out-0708.google.com [209.85.132.241]) by ozlabs.org (Postfix) with ESMTP id 00E85DDF38 for ; Sun, 25 May 2008 14:47:09 +1000 (EST) Received: by an-out-0708.google.com with SMTP id c34so316783anc.78 for ; Sat, 24 May 2008 21:47:08 -0700 (PDT) Message-ID: Date: Sat, 24 May 2008 22:47:08 -0600 From: "Grant Likely" Sender: glikely@secretlab.ca To: "David Brownell" Subject: Re: [RFC] OpenFirmware bindings for the MMC-over-SPI driver In-Reply-To: <200805241256.32639.david-b@pacbell.net> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 References: <20080523182754.GA26891@polina.dev.rtsoft.ru> <200805241256.32639.david-b@pacbell.net> Cc: David Brownell , Gary Jennejohn , linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org, Guennadi Liakhovetski , Pierre Ossman List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Sat, May 24, 2008 at 1:56 PM, David Brownell wrote: > On Friday 23 May 2008, Anton Vorontsov wrote: >> >> This is second attempt to write the OpenFirmware bindings for the >> MMC-over-SPI (and SPI bindings in general). > > Summary: an OF-specific wrapper around the mmc_spi platform code. > > I think a wrapper to encapsulate all the OF-specific knowledge makes > much sense here. > > The only thing that looks odd to me about this is that the wrapper > is a spi_device rather than an of_device. To me it makes more sense > to just have an of_device setting up the right spi_device. (Though > maybe I missed some discussion about why that can't work.) It's not so much that I can't work; more like it's not necessary. of_platform bus is by no means the prescribed way to work with the device tree. In fact, there is talk about moving away from of_platform bus entirely and using platform_device/spi_device/i2c_device/etc directly instead since the of_platform bus is mostly a clone of the platform bus with different device binding semantics. As Anton's patch shows, it is straight forward to add a binding that can extract the platform data out of the device tree without the overhead of adding an additional of_platform bus device & driver pair. Cheers, g. -- Grant Likely, B.Sc., P.Eng. Secret Lab Technologies Ltd.