From owner-freebsd-current  Sun Nov 29 20:32:24 1998
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id UAA15585
          for freebsd-current-outgoing; Sun, 29 Nov 1998 20:32:24 -0800 (PST)
          (envelope-from owner-freebsd-current@FreeBSD.ORG)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.15.68.22])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA15559;
          Sun, 29 Nov 1998 20:32:02 -0800 (PST)
          (envelope-from bde@godzilla.zeta.org.au)
Received: (from bde@localhost)
	by godzilla.zeta.org.au (8.8.7/8.8.7) id PAA20359;
	Mon, 30 Nov 1998 15:31:43 +1100
Date: Mon, 30 Nov 1998 15:31:43 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199811300431.PAA20359@godzilla.zeta.org.au>
To: bde@zeta.org.au, deischen@iworks.interworks.org, majordom@FreeBSD.ORG
Subject: Re: kern_clock.c (was: video mode switching has gone south)
Cc: current@FreeBSD.ORG, elias@cnetworks.net, green@unixhelp.org,
        phk@FreeBSD.ORG, seggers@semyam.dinoco.de
Sender: owner-freebsd-current@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

>> My system is running a Cyrix 686MX PR266, so this doesn't seem to be
>> limited to the AMD processors.
>
>Well, it affects my laptop also (Intel Pentium 200MHz MMX).  I do
>not run xntpd, but I am running the kernel at HZ=1000.

Increasing HZ certainly amplifies the old bug.  The timecounters will
cycle every NTIMECOUNTER/hz seconds, and problems occur if handling a
bunch of interrupts (or one heavweight interrupt) ever takes longer
than this.  You need to increase NTIMECOUNTER to 10 times what would
work for HZ=100.

Bruce

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