freebsd stable - Aug 2005 - calcru: runtime went backwards and hard lockup in 5.4-STABLE

hey folks..
Im running on a compaq proliant dual 733Mhz computer with 1 gig o memory..
I was running 5.4-STABLE a few weeks ago with no problem.. I did a make world
about a
week ago, and now I see lots of messages in syslog..
Aug 22 17:20:40 monsterjam kernel: calcru: runtime went backwards from 77810370
usec to
77203
321 usec for pid 48 (swi5: clock sio)
Aug 22 17:20:40 monsterjam kernel: calcru: runtime went backwards from 17036511
usec to
16595
708 usec for pid 71 (schedcpu)
Aug 22 17:20:40 monsterjam kernel: calcru: runtime went backwards from 77810370
usec to
77203
365 usec for pid 48 (swi5: clock sio)
Aug 22 17:20:40 monsterjam kernel: calcru: runtime went backwards from 17036511
usec to
16595
708 usec for pid 71 (schedcpu)
Aug 22 17:20:40 monsterjam kernel: calcru: runtime went backwards from 77810370
usec to
77203
463 usec for pid 48 (swi5: clock sio)

and I see 
http://www.freebsd.org/doc/en_US.ISO8859-1/books/faq/troubleshoot.html#CALCRU-NEGATIVE
but this info seems to only apply to 4.x.

Is this really a problem? how can I squelch it?

the other reason I ask is because ive had 2 HARD lockups since upgrading. I have
since
compiled a debugging kernel..

(in my kernel config)
makeoptions    DEBUG=-g
options         GDB
options         DDB
options         KDB
(this look right?)

I have a serial connection hooked up to this server and I cannot even break into
the
debugger when it hangs.

regards,
Jason

Jason <jason@monsterjam.org> writes:
> and I see 
>
http://www.freebsd.org/doc/en_US.ISO8859-1/books/faq/troubleshoot.html#CALCRU-NEGATIVE
> but this info seems to only apply to 4.x.
Not really, but the values are making *huge* jumps back and forth, so
it's not just a matter of missing some interrupts in your case.
> Is this really a problem? how can I squelch it?
It does seem to be a problem.  It may be a hardware problem with your
timekeeping source.  Try changing the timecounter.