From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ducrot Bruno Subject: Re: can't switch CPU frequency while running on batterypower Date: Wed, 7 Jan 2004 16:55:20 +0100 Sender: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Message-ID: <20040107155520.GP14031@poupinou.org> References: <20040106103944.Q26693@root.org> <32785.192.168.36.2.1073420997.squirrel@defiant.homedns.org> <20040106133725.E27491@root.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20040106133725.E27491-Y6VGUYTwhu0@public.gmane.org> Errors-To: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: Nate Lawson Cc: Defiant , acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: linux-acpi@vger.kernel.org On Tue, Jan 06, 2004 at 01:39:44PM -0800, Nate Lawson wrote: > On Tue, 6 Jan 2004, Defiant wrote: > > Nate Lawson wrote: > > > This is most likely a problem with re-evaluation of _PSS after it has been > > > changed. I don't remember if Linux uses the notify on the processor > > > object to re-evaluate _PSS. If Linux does this correctly, it may be that > > > the underlying firmware has a bug and you should use a more specific > > > cpufreq driver (i.e. SpeedStep or whatever). Some systems have a problem > > > where they don't properly export firmware information to the ACPI > > > performance states interface. > > > > uhm, I don't know why, but it is working now, using > > CONFIG_X86_P4_CLOCKMOD instead of CONFIG_X86_SPEEDSTEP_ICH, even it is a > > Pentium 4m, not Pentium 4. > > Sure. In general, it's better to use a hw-specific cpufreq driver instead > of the generic ACPI one, even if the latter seems to work. Many chipsets > don't export the hw-specific information to ACPI correctly and thus ACPI > performance states cpufreq should be a last resort. If they still don't > work, you should try multiple hw-specific drivers, as you did. > What is strange though is that speedstep-ich is a non-acpi driver. -- Ducrot Bruno -- Which is worse: ignorance or apathy? -- Don't know. Don't care. ------------------------------------------------------- This SF.net email is sponsored by: Perforce Software. Perforce is the Fast Software Configuration Management System offering advanced branching capabilities and atomic changes on 50+ platforms. Free Eval! http://www.perforce.com/perforce/loadprog.html