From owner-FreeBSD-users-jp@jp.freebsd.org  Mon May 24 12:38:45 1999
Received: (from daemon@localhost)
	by castle.jp.freebsd.org (8.9.3+3.2W/8.7.3) id MAA35939;
	Mon, 24 May 1999 12:38:45 +0900 (JST)
	(envelope-from owner-FreeBSD-users-jp@jp.FreeBSD.org)
Received: from may.nue.org (may.nue.org [163.138.156.31])
	by castle.jp.freebsd.org (8.9.3+3.2W/8.7.3) with ESMTP id MAA35934
	for <freebsd-users-jp@jp.freebsd.org>; Mon, 24 May 1999 12:38:44 +0900 (JST)
	(envelope-from amagai-b@nue.org)
From: amagai-b@nue.org
Received: by may.nue.org (8.9.3/nue.mxn1); Mon, 24 May 1999 12:38:35 +0900 (JST) <199905240338.MAA19622@may.nue.org> from 
Date: Mon, 24 May 1999 12:38:35 +0900 (JST)
Message-Id: <199905240338.MAA19622@may.nue.org>
To: freebsd-users-jp@jp.freebsd.org
In-Reply-To: <19990523125643.23876.qmail@mbb.nifty.ne.jp>
Reply-To: FreeBSD-users-jp@jp.freebsd.org
Precedence: list
X-Distribute: distribute version 2.1 (Alpha) patchlevel 24e+990430
X-Sequence: FreeBSD-users-jp 42595
Subject: [FreeBSD-users-jp 42595] Re: [42573] negative time of XXXXXX
Errors-To: owner-FreeBSD-users-jp@jp.freebsd.org
Sender: owner-FreeBSD-users-jp@jp.freebsd.org
X-Originator: amagai-b@nue.org

> calcru: negative time of -693557420 usec for pid 3081 (perl5.00502)

$B$$$+$K$b$3$l$G$9$M(B. 

sysctl -w kern.timecounter.method=1

$B$r;n$7$F$_$?$i$$$+$,$G$7$g$&$+(B. 

$BE73$(B

================================================================

1.86 Sun Nov 29 20:31:02 1998 UTC by phk
CVS Tags: RELENG_3_1_0_RELEASE, RELENG_3_BP; Branch point for: RELENG_3
Diffs to 1.85

Make the previous behaviour the default, add a sysctl which you
can set if your hw/sw produces the "calcru negative..." message.

Setting the alternate method (sysctl -w kern.timecounter.method=1)
makes the the get{nano|micro}*() functions call the real thing at
resulting in a measurable but minor overhead.

I decided to NOT have the "calcru" change the method automatically
because you should be aware of this problem if you have it.

The problems currently seen, related to usleep and a few other corners
are fixed for both methods.

