From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jyri Sarha Subject: Re: [PATCH 3/3] ASoC: simple-card: Add system-clock-id DT parameter to dai nodes Date: Tue, 24 Jun 2014 12:59:20 +0300 Message-ID: <53A94BF8.4090808@ti.com> References: <20140621100959.GC5099@sirena.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20140621100959.GC5099@sirena.org.uk> Sender: linux-omap-owner@vger.kernel.org To: Mark Brown Cc: alsa-devel@alsa-project.org, linux-omap@vger.kernel.org, devicetree@vger.kernel.org, bcousson@baylibre.com, peter.ujfalusi@TI.COM, liam.r.girdwood@linux.intel.com, detheridge@TI.COM List-Id: alsa-devel@alsa-project.org On 06/21/2014 01:09 PM, Mark Brown wrote: > On Fri, Jun 13, 2014 at 01:04:38PM +0300, Jyri Sarha wrote: >> Select clk_id parameters for set_sysclk calls in the card init phase. > > This is making the ID defined by the drivers into ABI. It's not clear > that this is something we want to do, and we certainly don't want to do > it without updating the driver bindings. In principle what we want to > do here is have our clocks handled by the clock API, though that's not > trivial work. > I see, actually this is not something I would need anytime soon, but how about the previous "Add system-clock-direction DT parameter to dai nodes" patch? There is a clear common definitions for clock direction, so this should be ok, right? If the clock direction patch would go in I would not need to add more stuff to already quite messy davinci-evm driver. Best regards, Jyri