Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 17 May 2001 22:22:13 +1000 (EST)
From:      Bruce Evans <bde@zeta.org.au>
To:        Mike Smith <msmith@FreeBSD.ORG>
Cc:        freebsd-questions@FreeBSD.ORG, arch@FreeBSD.ORG
Subject:   Re: Gettimeofday Again... 
Message-ID:  <Pine.BSF.4.21.0105172158350.17368-100000@besplex.bde.org>
In-Reply-To: <200105170813.f4H8DhE01424@mass.dis.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 17 May 2001, Mike Smith wrote:

> > >     I don't change the timercounter method defaults, and I sure hope you
> > >     aren't advocating that people change their timecounter defaults.  If
> > >     the TSC is a reasonable default, the system should figure it out and
> > >     use it without requiring intervention.
> > 
> > It's only a reasonable default if apm (or possibly acpica) is configured
> > (and used).
> 
> The TSC is never a reasonable default; there is no good way to be certain 
> that the TSC is and/or will remain stable.  Even with ACPI, you can't be 
> entirely sure.

This must be why Linux uses it by default ;-).  See
linux/arch/i386/config.in, option CONFIG_X86_TSC.

Linux-2.4.1 still only uses it to give an offset from the last i8254
clock interrupt, like FreeBSD used to do 3+ years ago before timecounters.
This may limit the errors from the TSC frequency changing to between
-10 and 0 msec (hopefully the frequency is calibrated when it is as
large as possible; then if it slows down down you underestimate the
offset).

Bruce


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-arch" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0105172158350.17368-100000>