From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeremy Fitzhardinge Subject: Re: my dothan didn't work with cpufreq... Date: Wed, 21 Jul 2004 18:55:20 -0700 Sender: cpufreq-bounces@www.linux.org.uk Message-ID: <1090461320.13505.3.camel@localhost> References: <40F2FA8B.10307@lifl.fr> <20040713094937.GB8124@dominikbrodowski.de> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20040713094937.GB8124@dominikbrodowski.de> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: cpufreq-bounces+glkc-cpufreq=gmane.org@www.linux.org.uk Content-Type: text/plain; charset="us-ascii" To: Dominik Brodowski Cc: cpufreq@www.linux.org.uk On Tue, 2004-07-13 at 11:49 +0200, Dominik Brodowski wrote: > Maybe. The frequency/voltage pairs for Dothans are different to those of > Banias. A built-in table isn't possible for Dothans, IIRC, as even if you > know the CPUID you can't know for 100% which pairs are valid -- it depends > on some other info not in the CPUID. I've been meaning to look into this to see if there's some way to add a table method. It's moderately tricky because there's two steppings of Dothan, and the specs are not obvious to me (they seem to list 4 voltages per frequency point, which is confusing). Anyway, I have a good collection of /proc/cpuinfos now, so I can pick through them to see what common elements there are. I guess it would also be good to get dumps from the ACPI tables. What's the best way to do that? [ And for those who've mailed me and got nothing back - I'm sorry about that. I have been getting a lot of reports - people seem to be buying a lot of laptops... ] J