From mboxrd@z Thu Jan 1 00:00:00 1970 From: Igor Stoppa Subject: Re: cpufreq-set problems [Was: Re: Ottawa Linux Power Management Summit, June 25-26, 2007 - Minutes] Date: Mon, 29 Oct 2007 15:26:00 +0200 Message-ID: <1193664360.23273.24.camel@localhost.localdomain> References: <200709050426.04793.lenb@kernel.org> <20071028135446.GG26943@isilmar.linta.de> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20071028135446.GG26943@isilmar.linta.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-pm-bounces@lists.linux-foundation.org Errors-To: linux-pm-bounces@lists.linux-foundation.org To: ext Dominik Brodowski Cc: linux-pm@lists.linux-foundation.org List-Id: linux-pm@vger.kernel.org Hi, On Sun, 2007-10-28 at 14:54 +0100, ext Dominik Brodowski wrote: > Hi, > > On Wed, Sep 05, 2007 at 04:26:04AM -0400, Len Brown wrote: > > cpufreq_set_policy() doesn't match Nokia's needs as it is a 1-way > > notification, and there is no way to register constraints. > > Could you elaborate a bit on this? cpufreq_update_policy() and > cpufreq_register_notifier() allows to register constraints, and then you can > use cpufreq_update_policy(). we are using update policy, but the problem with the notifications as cpufreq handles them, is that we wanted to have a specific sequence of operations been done, while (if i'm not wrong) the cpufreq notifications do not provide this feature. Probably this is not really needed for traditional x86 systems, while SoC allow to know at design time what sort of devices will be present in the system. -- Cheers, Igor Igor Stoppa (Nokia Multimedia - CP - OSSO / Helsinki, Finland)