From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ivo van Doorn Subject: Re: [RESEND] [PATCH -next 2/2] acpi,rfkill,backlight: comapl-laptop update - use rfkill switch subsystem Date: Thu, 10 Jul 2008 15:17:37 +0200 Message-ID: <200807101517.38109.IvDoorn@gmail.com> References: <20080709231059.778d16f4@debian> <200807092357.28013.IvDoorn@gmail.com> <20080710011201.GC15729@khazad-dum.debian.net> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from wr-out-0506.google.com ([64.233.184.234]:12485 "EHLO wr-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754201AbYGJNIO (ORCPT ); Thu, 10 Jul 2008 09:08:14 -0400 Received: by wr-out-0506.google.com with SMTP id 69so2459915wri.5 for ; Thu, 10 Jul 2008 06:08:12 -0700 (PDT) In-Reply-To: <20080710011201.GC15729@khazad-dum.debian.net> Content-Disposition: inline Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: Henrique de Moraes Holschuh Cc: Cezary Jackiewicz , linux-acpi@vger.kernel.org, linux-kernel@vger.kernel.org, ak@linux.intel.com, Len Brown , Richard Purdie , Andrew Morton On Thursday 10 July 2008, Henrique de Moraes Holschuh wrote: > On Wed, 09 Jul 2008, Ivo van Doorn wrote: > > It is not mandatory if you are writing rfkill support for a driver that does not > > come with a rfkill switch. Such drivers can make use of the rfkill events produced > > by the hardware which does have such a switch. > > > > When the hardware does have the rfkill switch, then yes rfkill_force_state() is mandatory. > > The get_state() callback function is optional, and allows rfkill to differentiate > > between soft and hardblock. > > Do you want me to mark rfkill_force_state() as mandatory in the docs? It > *IS* the preferred way to deal with firmware/hardware-initiated state > changes, after all. Please do. Thanks. > The rfkill subsystem will limp along without it, even when there are > hardware rfkill lines... but no OSD function will work, as the system will > pick up the change only when someone reads or writes to the state > attribute... That reason alone is good enough for me to mark it mandatory for any drivers which features the rfkill key. :) Ivo