Subject: Re: RFC: est.c driver synced with OpenBSD.
To: Steven M. Bellovin <>
From: Quentin Garnier <>
List: tech-kern
Date: 09/01/2006 14:42:14
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Fri, Sep 01, 2006 at 08:35:58AM -0400, Steven M. Bellovin wrote:
> On Fri, 01 Sep 2006 14:01:47 +1000, Simon Burge <> wrote:
> >=20
> > And blind luck wins!
> >=20
> > I've got a little script that cycles through the frequencies every 2
> > seconds, and prints some info gathered from the kernel.  I then started
> > a "make -j4" to keep the system busy.  Note here that the MSR_PERF_CTL
> > value of 0x613 is the value for the lowest CPU speed (1000MHz).
> Let me see if I understand the situation correctly.  If we had good ACPI
> support -- and for laptops, there was a lot of progress a few months ago,
> though it seems to have stalled; I'm one feature away from being able to
> use it on my laptop -- we could read the speed tables directly.  However,
> we don't have that, so we need to resort to heuristics to guess what the
> speeds should be?  Is that a fair summary?  Or is the data just not
> present on the machine in any form?

The ACPI table doesn't always have information about that, anyway.  For
example, a lot of desktop CPUs support EST, but you never find
information in the APCI table because the motherboard maker simply
doesn't control what CPU you put in it.  Also, I was surprised but the
table in Simon's laptop doesn't contain that information either.  So
even if we had the few bits to get the information from the ACPI table,
we'd still need static tables for other situations.

Quentin Garnier - -
"When I find the controls, I'll go where I like, I'll know where I want
to be, but maybe for now I'll stay right here on a silent sea."
KT Tunstall, Silent Sea, Eye to the Telescope, 2004.

Content-Type: application/pgp-signature
Content-Disposition: inline

Version: GnuPG v1.4.3 (NetBSD)