From owner-freebsd-current@FreeBSD.ORG Thu Jan 24 15:50:01 2008 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 5359616A41A for ; Thu, 24 Jan 2008 15:50:01 +0000 (UTC) (envelope-from bruce@zuhause.org) Received: from mail.zuhause.org (216.243.156.193.real-time.com [216.243.156.193]) by mx1.freebsd.org (Postfix) with ESMTP id 2ADAD13C455 for ; Thu, 24 Jan 2008 15:50:01 +0000 (UTC) (envelope-from bruce@zuhause.org) Received: by mail.zuhause.org (Postfix, from userid 1001) id D65BD7C04; Thu, 24 Jan 2008 09:38:10 -0600 (CST) From: Bruce Albrecht MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <18328.45282.562906.708945@celery.zuhause.org> Date: Thu, 24 Jan 2008 09:38:10 -0600 To: freebsd-current@freebsd.org X-Mailer: VM 7.15 under 21.4 (patch 12) "Portable Code" XEmacs Lucid X-Mailman-Approved-At: Thu, 24 Jan 2008 17:08:42 +0000 Subject: Weird performance behaviour in 7.0 X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 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: Thu, 24 Jan 2008 15:50:01 -0000 This is my first Q6600 based system, and my first 7.0 system, and my first AMD-64 system and my first ZFS base system, and I'm seeing something really strange. Every now and then, processes run really slowly, like at 1/600th (or worse) than it ought to, but it racks up the CPU time as though it's running full tilt. For example, I have this little test program that just does a tight loop, and most times, it takes about 3 seconds to complete, but right now, it's taking about 2000 seconds to complete. Right now, it's consistently running slow, but sometimes it will run slow, but I can terminate it and start another one which will run at normal speed. Currently sysctl reports: dev.cpu.0.freq: 2400 dev.cpu.0.temperature: 55 dev.cpu.1.temperature: 57 dev.cpu.2.temperature: 51 dev.cpu.3.temperature: 54 What might be causing this? What information is pertinent for inclusion in a problem report?