From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dominik Brodowski Subject: Re: [ACPI] _PDC method in DSDT Date: Mon, 23 Jun 2003 15:38:34 +0200 Sender: cpufreq-admin-1walMZg8u8rXmaaqVzeoHQ@public.gmane.org Message-ID: <20030623133834.GA2330@brodo.de> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Errors-To: cpufreq-admin-1walMZg8u8rXmaaqVzeoHQ@public.gmane.org List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: "Grover, Andrew" Cc: David Moore , "Adachi, Kenichi" , acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org, cpufreq-1walMZg8u8rXmaaqVzeoHQ@public.gmane.org List-Id: linux-acpi@vger.kernel.org On Mon, Jun 23, 2003 at 02:19:27AM -0700, Grover, Andrew wrote: > The spec intended FixedFunctionalHW registers to be an "escape hatch". > FFH means that while ACPI is used for perf state enumeration, the > transitions will be handled in a CPU-specific driver, not via methods > described in the ACPI spec. This division (use ACPI for enum, x for > control) isn't currently supported by cpufreq; ACPI either does both > enum and control, or neither. But it could easily be added to the cpufreq drivers. For example, the speedstep-centrino driver could call a int acpi_get_frequency_table(struct cpufreq_frequency_table **freq_table) which returns an array with the frequency in freq_table[i].frequency, and some sort of "index" in freq_table[i].index, e.g. what needs to be written into the IA32_PERF_CTL MSR. Dominik