From owner-freebsd-acpi@FreeBSD.ORG Fri Jan 9 05:51:08 2009 Return-Path: Delivered-To: freebsd-acpi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CC42D106564A for ; Fri, 9 Jan 2009 05:51:08 +0000 (UTC) (envelope-from yanefbsd@gmail.com) Received: from rv-out-0506.google.com (rv-out-0506.google.com [209.85.198.227]) by mx1.freebsd.org (Postfix) with ESMTP id 9F29B8FC08 for ; Fri, 9 Jan 2009 05:51:08 +0000 (UTC) (envelope-from yanefbsd@gmail.com) Received: by rv-out-0506.google.com with SMTP id b25so10653814rvf.43 for ; Thu, 08 Jan 2009 21:51:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:cc:in-reply-to:mime-version:content-type :content-transfer-encoding:content-disposition:references; bh=8jsfNm2THnI2X+90tTukp7Bg7kLCVXzvq9n8LK298aU=; b=GQbCzx/5vfdxp2teAS7ct4LznvaAxy+begovT0XKTGR/QF+m29VPT93wdMN24BuodT 0yQcGY89/EVBzHFRtBWcSF2VAk+aXiFmXc2CfUsg17twEwsYQy+8G4CT7X+CX19QY+lq g5Swl8SqEhCsLEpj8yBfsfzLyNQHAEWrVM7FI= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version :content-type:content-transfer-encoding:content-disposition :references; b=lSk14RmiuZit3ppAo0dJu1KY0bZt8yMe6g3xZc99TZQ6QYGIpAPLtTXnmJyHR+LFDt LCG6V/yK23Nd8L59JU9+ipq01mjeMH6XxaFp/Ic5KVSlwvWkVhXRA6EthRR9AsbKvK6v detfwYY6wbar75FBDdk1wJRQ2kgZKN+MTAVvc= Received: by 10.140.204.7 with SMTP id b7mr12540470rvg.117.1231480268223; Thu, 08 Jan 2009 21:51:08 -0800 (PST) Received: by 10.140.135.2 with HTTP; Thu, 8 Jan 2009 21:51:08 -0800 (PST) Message-ID: <7d6fde3d0901082151t4901824cv84eb6c2be2658b15@mail.gmail.com> Date: Thu, 8 Jan 2009 21:51:08 -0800 From: "Garrett Cooper" To: "Gabriel Lavoie" In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: Cc: freebsd-acpi@freebsd.org Subject: Re: EST (Enhanced Intel SpeedStep(R) Technology) on amd64 X-BeenThere: freebsd-acpi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: ACPI and power management development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Jan 2009 05:51:09 -0000 On Thu, Jan 8, 2009 at 3:51 PM, Gabriel Lavoie wrote: > Hello, > I recently built a small home server using an Inten 45nm E5200 @ > 2.5 GHz. With FreeBSD 7.0, the "est" driver had problems pooling for > the CPU frequency/voltage pairs and automatically disabled itself. It > left the CPU at the frequency the BIOS put it after startup, 1.25 GHz > because I have EIST enabled. I had to manually change the frequency or > use powerd to put it at maximum. Now I upgraded to FreeBSD 7.1 and > everything in my systems seems supported correctly. The est driver now > recognize the CPU and load the correct information. Also, after > bootup, the frequency of the CPU is at max (2.5 GHz). What I would > like is to get the behaviour of EIST found under Windows and Linux > where the CPU is automatically downclocked to 1.25 GHz when the system > isn't under any load, but comes back to 2.5 GHz as soon as there is > some load. I also get this behaviour under Linux on my Core 2 Duo and > I really like it. What I found interesting is that it seems the Intel > 45nm CPUs use under 5W of power when they are idle and EIST > downclocked them. I would really like to take profit of this low power > consumption. I found about "estctrl" on this page > http://www.daemonology.net/freebsd-est/ but it seems outdated and when > I try to build the port, it tells me that it is only supported on the > i386 architecture. Why not amd64? > > Thanks > > Gabriel > > -- > Gabriel Lavoie > glavoie@gmail.com Maybe this comment says why (from .../sys/i386/cpufreq/est.c)? /* * Frequency (MHz) and voltage (mV) settings. Data from the * Intel Pentium M Processor Datasheet (Order Number 252612), Table 5. * * Dothan processors have multiple VID#s with different settings for * each VID#. Since we can't uniquely identify this info * without undisclosed methods from Intel, we can't support newer * processors with this table method. If ACPI Px states are supported, * we get info from them. */ Colin could give you the definite answer, but I'm pretty sure that's it... Cheers, -Garrett