From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tony Lindgren Subject: Re: [RFC 1/3] pinctrl: add a driver for NVIDIA Tegra Date: Sun, 11 Dec 2011 11:36:25 -0800 Message-ID: <20111211193625.GE31337@atomide.com> References: <1323382390-14892-1-git-send-email-swarren@nvidia.com> <1323382390-14892-2-git-send-email-swarren@nvidia.com> <20111209170722.GO31337@atomide.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-tegra-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Linus Walleij Cc: Stephen Warren , linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-tegra@vger.kernel.org * Linus Walleij [111209 15:17]: > On Fri, Dec 9, 2011 at 6:07 PM, Tony Lindgren wrot= e: >=20 > > - We currently have hard time supporting pin groups with pins > > =C2=A0coming from multiple pinmux driver instances >=20 > Pin groups are per definiton per-controller inctance. That is a > given side effect of not having a global pin space. >=20 > But you can create two groups of the same name on two > pin controllers so that e.g. pinctrl.0 and pinctrl.1 have > a group named "foo", then map both to a certain device > in the board pinmux map. >=20 > You will have to request them individually however, like > with two hogs or two pinmux_get() calls specifying the > name of each map, else you will just get the first one. >=20 > Not that I've tried it, but that's how it's supposed to work... Yes a little bit of trickery is needed there, I'll look into it more as we already have two controller instances on omap4. Regards, Tony