From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate.crashing.org (gate.crashing.org [63.228.1.57]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTP id B00D22BDF2 for ; Wed, 3 Nov 2004 12:57:16 +1100 (EST) From: Benjamin Herrenschmidt To: Arrigo Benedetti In-Reply-To: <4187D1DE.90203@vision.caltech.edu> References: <4187D1DE.90203@vision.caltech.edu> Content-Type: text/plain Date: Wed, 03 Nov 2004 12:49:27 +1100 Message-Id: <1099446567.31630.21.camel@gaston> Mime-Version: 1.0 Cc: linuxppc-dev list Subject: Re: G5 clock frequency scaling List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Tue, 2004-11-02 at 10:28 -0800, Arrigo Benedetti wrote: > Does anyone know if the current IBM 970 chips in dual G5 Powermacs and > Xserves is under any clock frequency > scaling control? In other words, is the clock freuquency always > guardanteed to be the nominal, or it may change > based on temperature? If this is the case, does the linux kernel have > any control on it? The current desktop models have a "clock slewing" capability which isn't used by the kernel at this point. So the machine stays at the speed it had when booting. However, it's been observed that the firmware tend to boot at max speed when auto-booting, and at lowest speed (1.3Ghz generally observed) when entering the OF manual command line interface. Ben.