From owner-freebsd-current@FreeBSD.ORG  Thu Jan 26 18:50:56 2012
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Delivered-To: current@freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 5F21E106566B
	for <current@freebsd.org>; Thu, 26 Jan 2012 18:50:56 +0000 (UTC)
	(envelope-from luigi@onelab2.iet.unipi.it)
Received: from onelab2.iet.unipi.it (onelab2.iet.unipi.it [131.114.59.238])
	by mx1.freebsd.org (Postfix) with ESMTP id 271178FC12
	for <current@freebsd.org>; Thu, 26 Jan 2012 18:50:55 +0000 (UTC)
Received: by onelab2.iet.unipi.it (Postfix, from userid 275)
	id E1F267300A; Thu, 26 Jan 2012 20:08:17 +0100 (CET)
Date: Thu, 26 Jan 2012 20:08:17 +0100
From: Luigi Rizzo <rizzo@iet.unipi.it>
To: current@freebsd.org
Message-ID: <20120126190817.GA73310@onelab2.iet.unipi.it>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.4.2.3i
Cc: 
Subject: timeouts too long bug again...
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussions about the use of FreeBSD-current
	<freebsd-current.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>, 
	<mailto:freebsd-current-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-current>
List-Post: <mailto:freebsd-current@freebsd.org>
List-Help: <mailto:freebsd-current-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Jan 2012 18:50:56 -0000

In sep.2009 i noticed that usleep() select() and friends (including
the in-kernel callouts) would consistently take one tick longer
than they should, and committed a fix to sys/kern/kern_timeout.c
to HEAD and RELENG_8

http://svnweb.freebsd.org/base?view=revision&revision=197137
http://svnweb.freebsd.org/base?view=revision&revision=200510

Now it seems that the bug is there again, in some form:
on 9.0 (and someone reported this also no HEAD) the minimum sleep
interval is now 2 ticks. Callouts seem to be correct (at least,
dummynet delays are exact, and the sysctl
	net.inet.ip.dummynet.tick_adjustment: 2770
now does not grow at 500 HZ as it did in the past.

Any idea on what might have reintroduced the problem ?

(please spare us the story that usleep gives no guarantees
etc etc - we all know that, but FreeBSD is consistently 1 tick
above the minimum guaranteed interval)

cheers
luigi