From owner-freebsd-current@FreeBSD.ORG Wed Oct 6 15:30:02 2010 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E79FA106564A for ; Wed, 6 Oct 2010 15:30:02 +0000 (UTC) (envelope-from oberman@es.net) Received: from mailgw.es.net (mail1.es.net [IPv6:2001:400:201:1::2]) by mx1.freebsd.org (Postfix) with ESMTP id CAF8A8FC14 for ; Wed, 6 Oct 2010 15:30:02 +0000 (UTC) Received: from ptavv.es.net (ptavv.es.net [IPv6:2001:400:910::29]) by mailgw.es.net (8.14.3/8.14.3) with ESMTP id o96FTwGU029437 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Wed, 6 Oct 2010 08:29:58 -0700 Received: from ptavv.es.net (localhost [127.0.0.1]) by ptavv.es.net (Tachyon Server) with ESMTP id 40A0D1CC3E; Wed, 6 Oct 2010 08:29:58 -0700 (PDT) To: Bruce Cran In-reply-to: Your message of "Wed, 06 Oct 2010 07:18:38 BST." <201010060718.39137.bruce@cran.org.uk> Date: Wed, 06 Oct 2010 08:29:58 -0700 From: "Kevin Oberman" Message-Id: <20101006152958.40A0D1CC3E@ptavv.es.net> Cc: freebsd-current@freebsd.org, Alexander Kojevnikov Subject: Re: sysctl debug.cpufreq.highest X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Oct 2010 15:30:03 -0000 > From: Bruce Cran > Date: Wed, 6 Oct 2010 07:18:38 +0100 > Sender: owner-freebsd-current@freebsd.org > > On Wednesday 06 October 2010 03:02:36 Alexander Kojevnikov wrote: > > > This variable allows to underclock the CPU, which is useful when > > building quiet or low-power systems. > > You can use the new -m and -M switches to powerd to control the minimum and > maximum frequencies instead. Ack! Did this really make it into the code? This is NOT the answer. the answer is to get rid of the useless CPU throttling and TCC which is the real cause of this. Neither of these methods was designed as a power management mechanism. They were designed to keep the processor from over-heating. They both can do what they were designed for quite well, but they can provide only very limited power savings and can actually result in higher power consumption in some cases. (I reported on my research into this several years ago, but the definitive work was done by mav@ and can be read on the FreeBSD wiki at http://wiki.freebsd.org/TuningPowerConsumption If we would just get rid of this (or at least turn it off by default), this whole problems requiring -m would go away. (I can see real use for -M, though.) -- R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: oberman@es.net Phone: +1 510 486-8634 Key fingerprint:059B 2DDF 031C 9BA3 14A4 EADA 927D EBB3 987B 3751