*BSD News Article 17918


Return to BSD News archive

Newsgroups: comp.os.386bsd.questions
Path: sserve!newshost.anu.edu.au!munnari.oz.au!uunet!emba-news.uvm.edu!trantor.emba.uvm.edu!wollman
From: wollman@trantor.emba.uvm.edu (Garrett Wollman)
Subject: Re: clock gains time (or does it loose : -)
Message-ID: <1993Jul3.214618.18256@uvm.edu>
Sender: news@uvm.edu
Organization: University of Vermont, EMBA Computer Facility
References: <211rti$27e@urmel.informatik.rwth-aachen.de>
Date: Sat, 3 Jul 1993 21:46:18 GMT
Lines: 35

In article <211rti$27e@urmel.informatik.rwth-aachen.de> kuku@acds.physik.rwth-aachen.de writes:

>Is this - if it is as it is - a design flaw of 386bsd, that it cannot
>cope with any interrupt rate at any time? (also a possible cause for
>the silo overflows?)

No, it's a design flaw of quantum physics!  (Well, from one point of
view anyway...)

The problem here can be succinctly stated: crystals vary.  Unless you
want to shell out a large sum of money for a precisely calibrated
crystal, and keep your PC in a temperature-controlled room, your PC's
time will vary with environmental conditions and other problems.  This
also means that no two PCs are likely to have identical clocks.

The best way to determine just how far off your clock is, is to run
NTP synchronized to several good outside time sources.  My PC
(currently at stratum 3) has an NTP-reported drift of -59.36809 ppm,
which corresponds to a loss of five seconds per day---well within the
ability of NTP to correct.  If you find that your drift is more than
100 ppm in absolute value, then you should change the value of `tick'
in your kernel, using a debugger, the tickadj program, or by changing
the value in /sys/compile/YOURSYSTEM/param.c.

And don't feel bad if your clock is bad.  Because of a programming
error, most Sun-4 machines running SunOS 4.1.2 were far off the
mark...

-GAWollman

-- 
Garrett A. Wollman   | Shashish is simple, it's discreet, it's brief. ... 
wollman@emba.uvm.edu | Shashish is the bonding of hearts in spite of distance.
uvm-gen!wollman      | It is a bond more powerful than absence.  We like people
UVM disagrees.       | who like Shashish.  - Claude McKenzie + Florent Vollant