From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pavel Machek Subject: Re: [linux-pm] [PATCH v2 1/2] Input: gpio-keys - allow platform to specify exact irq flags Date: Thu, 10 Dec 2009 11:13:10 +0100 Message-ID: <20091210101310.GE2446@ucw.cz> References: <87einfltp3.fsf@tac.ki.iif.hu> <20091206084704.GC2766@ucw.cz> <20091208042251.GB11147@core.coreip.homeip.net> <1260277410.19669.84.camel@localhost> <20091208174218.GB14271@core.coreip.homeip.net> <1260343860.19669.1189.camel@localhost> <20091209180345.GE4456@core.coreip.homeip.net> <20091209210816.GD3480@elf.ucw.cz> <20091209214841.GE10138@core.coreip.homeip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from atrey.karlin.mff.cuni.cz ([195.113.26.193]:50565 "EHLO atrey.karlin.mff.cuni.cz" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933594AbZLJKip (ORCPT ); Thu, 10 Dec 2009 05:38:45 -0500 Content-Disposition: inline In-Reply-To: <20091209214841.GE10138@core.coreip.homeip.net> Sender: linux-input-owner@vger.kernel.org List-Id: linux-input@vger.kernel.org To: Dmitry Torokhov Cc: Artem Bityutskiy , Ferenc Wagner , Alan Stern , linux-pm@lists.linux-foundation.org, Mika Westerberg , "linux-input@vger.kernel.org" Hi! > > > > > Now, there is a issue of waking up userspace task, additional scheduling > > > > > and keeping CPU running longer than necessary for "uninteresting" keys. > > > > > This can be solved by implementing a subscription model which allows > > > > > filtering uninteresing events on a per-client basis at evdev level. > > > > > > > > Right. And for gpio_keys, this would be dine on the driver level. > > > > > > But the semantics are different - if done on driver level you'd be > > > affecting _all_ consumers of the device; what I want to be done only > > > affects owner of the file descriptor. > > > > Well, if _all_ consumers decide to ignore some key, we should be able > > to ignore it at driver level. > > The intesection of drivers allowing shut off individual buttons with > all consumers agreeing on not using a key would be pretty miniscule. Well; people want to do that on specialized hw/embedded sw. It could work there. > > And actually it may make some sense -- I do not think disabling irq > > during runtime is worth the effort, but disabling wakeup source and > > getting rid of unneccessary wakeup when system is suspended is > > probably worth it. > > I don't believe Mika's patches touched wakeup in any way. So it has > been strictly about wakig up processor to service that interrupt so far. I think it makes sense for wakeups; I doubt it makes sense for runtime - the power savings are just not big enough. -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html