Re: Still seeing "calcru: runtime went backwards" messages

From: John Baldwin <jhb_at_freebsd.org>
Date: Wed, 1 Mar 2006 17:24:06 -0500
On Wednesday 01 March 2006 17:15, Poul-Henning Kamp wrote:
> In message <44061AFC.4080109_at_rogers.com>, Mike Jakubik writes:
> 
> 
> >Mar  1 17:04:27 fbsd kernel: calcru: runtime went backwards from 6995 
> >usec to 6994 usec for pid 662 (perl)
> 
> Backwards steps of this small nature are to be expected from the
> cpu_tick (TSC) calibration in kern_tc.c
> 
> It is possible that they can be eliminated with a more precise (and
> therefore timeconsuming) calibration calculation.

Maybe we could make the dynamic flag for the tsc controllable via tunable?
If I have a server machine without any fancy pentium-m cpufreq type stuff,
then I would rather just use the tsc frequency snapshot taken at boot and
just stick with that as the static frequency then have a bunch of warnings
on the console if the max freq changes sometime later.  Also, note that
I am still getting a bunch of the backwards messages on my DS20 Alpha which
is _not_ using the TSC. :)  (I should probably fix Alpha to use its own
version of the TSC at some point, but for now I want to get it to stop
spitting out the warnings when using the timecounters as that shouldn't
be causing any warnings).

-- 
John Baldwin <jhb_at_FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org
Received on Wed Mar 01 2006 - 21:23:13 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:52 UTC