From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benoit Cousson Subject: Re: [PATCHv2] Input: omap4-keypad: Add pinctrl support Date: Tue, 23 Oct 2012 11:35:06 +0200 Message-ID: <508664CA.7000601@ti.com> References: <1350911580-20307-1-git-send-email-sourav.poddar@ti.com> <20121022155028.GA13791@core.coreip.homeip.net> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Linus Walleij Cc: Sourav Poddar , Dmitry Torokhov , tony@atomide.com, linux-omap@vger.kernel.org, linux-kernel@vger.kernel.org, devicetree-discuss@lists.ozlabs.org, linux-arm-kernel@lists.infradead.org, linux-input@vger.kernel.org, Felipe Balbi List-Id: linux-input@vger.kernel.org Hi Linus, On 10/23/2012 11:13 AM, Linus Walleij wrote: > On Mon, Oct 22, 2012 at 5:50 PM, Dmitry Torokhov > wrote: >> Hi Sourav, >> >> On Mon, Oct 22, 2012 at 06:43:00PM +0530, Sourav Poddar wrote: >>> Adapt keypad to use pinctrl framework. >>> >>> Tested on omap4430 sdp with 3.7-rc1 kernel. >> >> I do not see anything in the driver that would directly use pinctrl. Is >> there a better place to select default pin configuration; maybe when >> instantiating platform device? > > The option is to use pinctrl hogs. Then the pins will be taken, > muxed and configured by the pin controller itself. > > Another option (not implemented) is to use bus notifiers. > > (I wrote about this in some other thread but can't find it now.) > > Each approach above come with its own set of problems. > > If the driver need to handle multiple states like default/idle/sleep > it is IMO better to put the handling into the driver, so if that > is what is going to happen also to this driver it could be a good > idea to actually implement that code upfront and include in > this submission so as to show that this driver is really going > to exercise its pins. > > But it's also a question of conformity: if other drivers in the > system is using different states and this is the only one > using a single "default" state, then it doesn't make sense > to have just one driver get its pins using hogs, it's just > inconsistent. > > So Sourav, please tell us a bit about your plans for this > and other drivers! Yeah, this idea is to handle pinctrl from all the drivers, and potentially change the mode during suspend when it is relevant. Regards, Benoit