From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bo Shen Date: Mon, 29 Oct 2012 06:38:17 +0000 Subject: Re: [PATCH 4/5] ASoC: sam9g20_wm8731: convert to device tree support Message-Id: <508E2459.5090707@atmel.com> List-Id: References: <1350901079-18307-1-git-send-email-voice.shen@atmel.com> <1350901079-18307-4-git-send-email-voice.shen@atmel.com> <20121027221204.GL4564@opensource.wolfsonmicro.com> In-Reply-To: <20121027221204.GL4564@opensource.wolfsonmicro.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Mark Brown Cc: alsa-devel@alsa-project.org, devicetree-discuss@lists.ozlabs.org, nicolas.ferre@atmel.com, linux-sound@vger.kernel.org, plagnioj@jcrosoft.com, jm.lin@atmel.com, linux-arm-kernel@lists.infradead.org On 10/28/2012 6:12, Mark Brown wrote: > On Mon, Oct 22, 2012 at 06:17:58PM +0800, Bo Shen wrote: >> Covert sam9g20 wm8731 to device tree support >> And enable it through dts file >> >> Tested on sam9g20 EK board > > This needs to add binding documentation for the board. I will add the binding documentation in next version. >> + dai: dai { >> + compatible = "atmel,atmel-ssc-dai"; >> + atmel,dai-master = <&ssc0>; >> + }; > > This looks wrong - this is a Linux-specific virtual device sitting on > top of the SSC which is the actual physical device. The usual patterns > would be something like have the machine driver register the DAI based > on the SSC specified in the bindings (much like how you're handling the > platform already). I have a question for this, do all nodes in dts file should be physical device? The SSC part is a little different with other SoC family. So, I think, if I keep the ssc driver in driver/misc folder as the library code. And create new code into sound/soc/atmel for ssc which used for audio, would this be acceptable? If so, the framework will be the same with other SoC family.