From owner-freebsd-stable@FreeBSD.ORG Thu Aug 17 05:31:50 2006 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3AFB516A4DA for ; Thu, 17 Aug 2006 05:31:50 +0000 (UTC) (envelope-from bill@lefebvre.org) Received: from atl04.ws-e.com (vh00.ws-e.com [69.61.31.90]) by mx1.FreeBSD.org (Postfix) with ESMTP id BE26343D46 for ; Thu, 17 Aug 2006 05:31:49 +0000 (GMT) (envelope-from bill@lefebvre.org) Received: from lilburn.lefebvre.org (ocee.groupsys.com [66.149.10.161]) by atl04.ws-e.com (8.13.6/8.13.6) with ESMTP id k7H5VnNb060532 for ; Thu, 17 Aug 2006 01:31:49 -0400 (EDT) (envelope-from bill@lefebvre.org) Received: from [192.168.0.110] (decatur.lefebvre.org [192.168.0.110]) by lilburn.lefebvre.org (8.13.3/8.12.8) with ESMTP id k7H5VmsB046386 for ; Thu, 17 Aug 2006 01:31:48 -0400 (EDT) (envelope-from bill@lefebvre.org) Message-ID: <44E3FF44.9070802@lefebvre.org> Date: Thu, 17 Aug 2006 01:31:48 -0400 From: Bill LeFebvre User-Agent: Thunderbird 1.5.0.2 (X11/20060425) MIME-Version: 1.0 To: freebsd-stable@freebsd.org References: <44E1F796.5070105@rogers.com> <20060815172728.GB88051@dan.emsphone.com> <44E39809.10104@mail.uni-mainz.de> In-Reply-To: <44E39809.10104@mail.uni-mainz.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-DCC-NIET-Metrics: lilburn.lefebvre.org 1080; env_From=2 Body=1 Fuz1=1 Fuz2=1 X-Scanned-By: MIMEDefang 2.43 Subject: Re: TOP shows above 100% WCPU usage X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Aug 2006 05:31:50 -0000 O. Hartmann wrote: > I use FreeBSD 6.1-STABLE/AMD on an pure 64Bit box/environment, single > CPU Athlon 3500+, and sometimes I can see a 100%+ usage of WCPU in > 'xine' or 'transmission'. So this is definitely not related to multiple > CPUs. WCPU is supposed to be weighted in some way to take swap time in to account. It's possible that the weighting calculation is adding a bit to a nearly 100% value. Try displaying just %CPU (the 'C' command I think?) and see if it is reporting over 100% for a normal (non-weighted) cpu percent. It shouldn't. If it is, then there's something weird about the way the value is being tracked by the kernel. Bill LeFebvre