Date: Tue, 22 Jan 2008 09:47:56 +0100 From: Gerrit =?ISO-8859-1?Q?K=FChn?= <gerrit@pmp.uni-hannover.de> To: Jeremy Chadwick <koitsu@freebsd.org> Cc: freebsd-stable@freebsd.org, freebsd-mobile@freebsd.org Subject: Re: T7200 CPU not detected by est Message-ID: <20080122094756.2d8f0bdf.gerrit@pmp.uni-hannover.de> In-Reply-To: <20080121170102.GA46419@eos.sc1.parodius.com> References: <20080121171606.8eb5f189.gerrit@pmp.uni-hannover.de> <20080121170102.GA46419@eos.sc1.parodius.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 21 Jan 2008 09:01:02 -0800 Jeremy Chadwick <koitsu@FreeBSD.org> wrote about Re: T7200 CPU not detected by est: JC> > Jan 18 23:18:14 comet kernel: est1: <Enhanced SpeedStep Frequency JC> > Control> on cpu1 Jan 18 23:18:14 comet kernel: est: CPU supports JC> > Control> Enhanced Speedstep, but is not recognized. JC> > Jan 18 23:18:14 comet kernel: est: cpu_vendor GenuineIntel, msr JC> > 6130c2906000c29 Jan 18 23:18:14 comet kernel: device_attach: est1 JC> > attach returned 6 JC> I see identical behaviour on our Supermicro PDSMI+ systems, using E6420 JC> CPUs, so I don't believe the problem is specific to your motherboard or JC> certain Intel CPU models: It is definitely not bound to certain mainboards, because I see it on several different ones. However, I have only T-series CPUs to test. JC> CPU: Intel(R) Core(TM)2 CPU 6420 @ 2.13GHz (2128.01-MHz JC> 686-class CPU) acpi0: <PTLTD RSDT> on motherboard JC> acpi0: [ITHREAD] JC> acpi0: Power Button (fixed) JC> cpu0: <ACPI CPU> on acpi0 JC> est0: <Enhanced SpeedStep Frequency Control> on cpu0 JC> est: CPU supports Enhanced Speedstep, but is not recognized. JC> est: cpu_vendor GenuineIntel, msr 82a082a0600082a JC> device_attach: est0 attach returned 6 JC> cpu1: <ACPI CPU> on acpi0 JC> est1: <Enhanced SpeedStep Frequency Control> on cpu1 JC> est: CPU supports Enhanced Speedstep, but is not recognized. JC> est: cpu_vendor GenuineIntel, msr 82a082a0600082a JC> device_attach: est1 attach returned 6 Ok, so it's probably neither specific for CPUs nor for the mainbaords; however, up to now all CPUs with this problem are Core2 CPUs. JC> In the case of our servers, we usually turn EIST off (this one JC> particular box has it enabled) because of the above problem -- but I'd JC> much rather have it turned on to help save power. For a laptop or JC> workstation, however, I can see this being an incredibly important JC> feature. I run low power workstations here which are intended to be used in a lab environment, and I would very much like to have working power-saving features (otherwise the whole setup is quite useless). Can I somehow help debugging this, should I file a PR or are there any further recommended things to do? cu Gerrit
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080122094756.2d8f0bdf.gerrit>