From owner-freebsd-acpi@FreeBSD.ORG Thu Jul 3 15:54:27 2008 Return-Path: Delivered-To: acpi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id C90D5106567F for ; Thu, 3 Jul 2008 15:54:27 +0000 (UTC) (envelope-from nate@root.org) Received: from root.org (root.org [67.118.192.226]) by mx1.freebsd.org (Postfix) with ESMTP id 92C4C8FC1C for ; Thu, 3 Jul 2008 15:54:27 +0000 (UTC) (envelope-from nate@root.org) Received: (qmail 1775 invoked from network); 3 Jul 2008 15:27:47 -0000 Received: from ppp-71-139-14-56.dsl.snfc21.pacbell.net (HELO ?10.0.5.18?) (nate-mail@71.139.14.56) by root.org with ESMTPA; 3 Jul 2008 15:27:47 -0000 Message-ID: <486CEFEC.2070107@root.org> Date: Thu, 03 Jul 2008 08:27:40 -0700 From: Nate Lawson User-Agent: Thunderbird 2.0.0.14 (Windows/20080421) MIME-Version: 1.0 To: David Wolfskill References: <20080627235319.GP70792@bunrab.catwhisker.org> In-Reply-To: <20080627235319.GP70792@bunrab.catwhisker.org> X-Enigmail-Version: 0.95.6 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: acpi@freebsd.org Subject: Re: How/why would dev.cpu.0.freq_levels change??!? 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: Thu, 03 Jul 2008 15:54:27 -0000 David Wolfskill wrote: > My laptop is a Dell Inspiron 8200; I (ab)use it moderately heavily: > this includes tracking RELENG_6, RELENG_7, & HEAD on it, daily. > > Lately there have been some times when "make buildworld" for RELENG_6 > has taken a lot longer than it used to ... and I noticed that the > fans were on, even though it was running fairly cool (around 50C; > during a "make buildworld, around 85C is more common) -- and that > the machine was typically "topping out" at half speed (1200 MHz). > > During these times, querying dev.cpu.0.freq_levels would yield a list > that did, ini fact, max out at 1200 MHz, when I know that it has gone up > to 2400 MHz in the past. The ACPI spec describes how the _PSS levels can change dynamically at runtime. For instance, if you go off AC, the system might offer a lower power state. I'm guessing you're using the acpi_perf driver. Usually the system generates a notify when that happens though so I was surprised there was no dmesg. I'm not sure how well we support changing levels yet. -- Nate