From owner-freebsd-current@FreeBSD.ORG Sat Apr 24 02:39:00 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2297C16A4CF for ; Sat, 24 Apr 2004 02:39:00 -0700 (PDT) Received: from ultra36.almamedia.fi (ultra36.almamedia.fi [193.209.83.44]) by mx1.FreeBSD.org (Postfix) with ESMTP id D926243D5C for ; Sat, 24 Apr 2004 02:38:59 -0700 (PDT) (envelope-from koston@iki.fi) Received: by ultra36.almamedia.fi (Postfix, from userid 60001) id 5372C1DA46; Sat, 24 Apr 2004 12:38:58 +0300 (EEST) Received: from norsu.kameli.org (ip213-185-39-27.laajakaista.mtv3.fi [213.185.39.27]) by ultra36.almamedia.fi (Postfix) with SMTP id D4BC21DA38; Sat, 24 Apr 2004 12:38:57 +0300 (EEST) Received: by norsu.kameli.org (sSMTP sendmail emulation); Sat, 24 Apr 2004 11:40:43 +0000 From: "Atte Peltomaki" Date: Sat, 24 Apr 2004 11:40:43 +0000 To: Eirik Oeverby Message-ID: <20040424114043.GA25392@norsu.kameli.org> References: <200404201311.i3KDBY7C071099@anduin.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200404201311.i3KDBY7C071099@anduin.net> User-Agent: Mutt/1.4.2.1i cc: freebsd-current@freebsd.org Subject: Re: SMP weirdness in -CURRENT X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 24 Apr 2004 09:39:00 -0000 > So.. What's going on? Am I the only one who notices these things? I have a feeling this only happens when the system load is high, not when the user load is high. And the funny thing is - my openoffice build (in this case) is running with nice -n 10, so even its priority is low. > Is this related to SCHED_ULE, should I try with SCHED_BSD? I thought the real strengths of SCHED_ULE only shows on SMP, am I wrong? I have been experiencing same symptoms you described on two of my SMP boxes, long time now. And I'm really going to get to the bottom of this, tried switching NIC's, audio cards, UP kernel, with and without ACPI.. I'll let you know when I come up with something useful. At least the symptoms surface a lot worse when all the -current kernel debugging options are turned on (=more system cpu load) Atte Peltomaki http://kameli.org