From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262790AbTLOAEm (ORCPT ); Sun, 14 Dec 2003 19:04:42 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262794AbTLOAEm (ORCPT ); Sun, 14 Dec 2003 19:04:42 -0500 Received: from deagol.email.Arizona.EDU ([128.196.133.142]:22660 "EHLO smtpgate.email.arizona.edu") by vger.kernel.org with ESMTP id S262790AbTLOAEl (ORCPT ); Sun, 14 Dec 2003 19:04:41 -0500 Subject: Strange issue with 2.6.0-test11 and magic packets From: Harry McGregor To: linux-kernel@vger.kernel.org Content-Type: text/plain Organization: Tucson Center Support Group - USGS Message-Id: <1071446598.30538.6.camel@Sony> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.5 Date: Sun, 14 Dec 2003 17:03:19 -0700 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, We have been having a strange issue while testing 2.6.0-test in one of our computer labs (actually, it's a Debian GNU/Linux based lab in an Elementary School!). The lab uses ether-wake to boot the computers, which is of course a bios level thing, and works quite nicely. The problem comes into play that when a computer happens to already be booted when a magic packet is sent to it's MAC address, CPU load goes to 100%. top shows the usages as being [events]. We did not experience this result of sending a magic packet while running 2.4.x kernels. Hardware is Intel i815e motherboards in Compaq Deskpro systems. Intel e100 driver built into the kernel. If anyone can shed some light on where we should start looking, or any tests you would like to see done, either follow up this post, or respond to me directly. Harry