From mboxrd@z Thu Jan 1 00:00:00 1970 From: John Crispin Subject: Re: [PATCH V5 1/2] regulator: Add document for MT6323 regulator Date: Tue, 9 Feb 2016 13:45:17 +0100 Message-ID: <56B9DF5D.2010501@openwrt.org> References: <1454874037-38905-1-git-send-email-blogic@openwrt.org> <20160209124133.GE13270@sirena.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20160209124133.GE13270@sirena.org.uk> Sender: linux-kernel-owner@vger.kernel.org To: Mark Brown Cc: Liam Girdwood , Matthias Brugger , chen.zhong@mediatek.com, HenryC.Chen@mediatek.com, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-mediatek@lists.infradead.org, devicetree@vger.kernel.org List-Id: devicetree@vger.kernel.org On 09/02/2016 13:41, Mark Brown wrote: > On Sun, Feb 07, 2016 at 08:40:36PM +0100, John Crispin wrote: > >> +Required properties: >> +- compatible: "mediatek,mt6323-regulator" > > I'm really not happy with MFD subfunctions like this which add no > information over the parent device name appearing directly in the > binding, it results in us putting Linux specifics about how we split the > device up into the driver. It's not like this could realistically be > used separately to the parent device. Just have the MFD create the > device directly or if the individual regulators are reusable IPs which > can share a driver then describe them as such in the DT. > Hi Mark, How would you like it to be. i am failing to understand that from your comment. John