Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 15 Aug 2004 11:03:35 +0200
From:      des@des.no (=?iso-8859-1?q?Dag-Erling_Sm=F8rgrav?=)
To:        Doug White <dwhite@gumbysoft.com>
Cc:        Andrew Gallatin <gallatin@cs.duke.edu>
Subject:   Re: Is the TSC timecounter safe on SMP system?
Message-ID:  <xzpbrhchixk.fsf@dwp.des.no>
In-Reply-To: <20040813103710.F93695@carver.gumbysoft.com> (Doug White's message of "Fri, 13 Aug 2004 10:38:29 -0700 (PDT)")
References:  <16668.61707.474283.639200@grasshopper.cs.duke.edu> <20040813103710.F93695@carver.gumbysoft.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Doug White <dwhite@gumbysoft.com> writes:
> On Fri, 13 Aug 2004, Andrew Gallatin wrote:
> > I have a system where the TSC timecounter is quite a bit more accurate
> > (or perhaps its just much cheaper) than the ACPI timecounter.  This is a
> > single CPU, HTT system running an SMP kernel.
> > [...]
> > 1) Is it safe to switch to TSC?
> If you like your ticks bouncing around; you'll get different values
> depending on which CPU you read.

Not necessarily; some SMP motherboards keep the TSCs synchronized.  On
those systems, you can set the kern.timecounter.smp_tsc tunable to a
non-zero value to make the TSC eligible.  It defaults to 1 on single-
CPU systems, but the TSC code can't tell the difference between an
HTT-enabled single-CPU box and a dual-CPU box, so you have to force it
on HTT-enabled systems.

DES
--=20
Dag-Erling Sm=F8rgrav - des@des.no



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?xzpbrhchixk.fsf>