From mboxrd@z Thu Jan 1 00:00:00 1970 From: tony@atomide.com (Tony Lindgren) Date: Mon, 29 Jul 2013 02:31:25 -0700 Subject: [PATCH 3/4] pinctrl: Add support for additional dynamic states In-Reply-To: References: <20130716090310.5541.36777.stgit@localhost> <20130716090536.5541.36289.stgit@localhost> Message-ID: <20130729093125.GB7656@atomide.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org * Linus Walleij [130722 16:14]: > On Tue, Jul 16, 2013 at 11:05 AM, Tony Lindgren wrote: > > > To toggle dynamic states, let's add the optional active state in > > addition to the static default state. Then if the optional active > > state is defined, we can require that idle and sleep states cover > > the same pingroups as the active state. > > OK... > > > Then let's add pinctrl_check_dynamic() and pinctrl_select_dynamic() > > to use instead of pinctrl_select() to avoid breaking existing users. > > > > With pinctrl_check_dynamic() we can check that idle and sleep states > > match the active state for pingroups during init, and don't need to > > do it during runtime. > > I do not understand why this complexity need to be exposed outside > of the subsystem. Unfortunately it's mostly to deal with supporting the current behaviour of pinctrl_select_state() which is not quite suitable for runtime PM. > pinctrl_select_state() and the PM accessors are enough IMO. Why > should say a driver care about whether it is dynamic or not? I think we can make this all transparent to the consumer drivers for runtime PM. Basically drivers/base/pinctrl.c needs these for the checks because of the current pinctrl_select_state(). > Surely the checking and different paths for static/dynamic configurations > can be an intrinsic detail of the pinctrl subsystem, by adding flags and > members to private structs like struct pinctrl itself in worst case. I'll take a look if we can bury more things inside the pinctrl subsystem. > So I'm not buying into this, it looks like it is making things complicated > for consumers outside the subsystem for no reason. I don't think the consumer drivers eventually need to do much anything ideally. We're missing runtime PM related set_irq_wake() but that's a minor detail as we can initially keep the runtime PM related wake-up events always enabled. Regards, Tony