From owner-freebsd-hardware@FreeBSD.ORG Wed Sep 21 20:44:51 2005 Return-Path: X-Original-To: freebsd-hardware@freebsd.org Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 801AE16A440; Wed, 21 Sep 2005 20:44:50 +0000 (GMT) (envelope-from kono@kth.se) Received: from omega.nanophys.kth.se (omega.nanophys.kth.se [130.237.35.96]) by mx1.FreeBSD.org (Postfix) with ESMTP id 777C243D49; Wed, 21 Sep 2005 20:44:48 +0000 (GMT) (envelope-from kono@kth.se) Received: from omega.nanophys.kth.se (localhost [127.0.0.1]) by omega.nanophys.kth.se (8.13.4/8.13.1) with ESMTP id j8LKiqOt009479; Wed, 21 Sep 2005 22:44:52 +0200 (CEST) (envelope-from kono@kth.se) Received: from localhost (localhost [[UNIX: localhost]]) by omega.nanophys.kth.se (8.13.4/8.13.1/Submit) id j8LKipx7009478; Wed, 21 Sep 2005 22:44:51 +0200 (CEST) (envelope-from kono@kth.se) X-Authentication-Warning: omega.nanophys.kth.se: kono set sender to kono@kth.se using -f From: Alexander Konovalenko Organization: KTH To: freebsd-hardware@freebsd.org, freebsd-amd64@freebsd.org Date: Wed, 21 Sep 2005 22:44:51 +0200 User-Agent: KMail/1.8.2 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200509212244.51873.kono@kth.se> Cc: Subject: irq conflict of pcm0 & vge0 devices on Abit AV8 cause high CPU load X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: kono@kth.se List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 21 Sep 2005 20:44:53 -0000 Hi! The gigabit NIC (vge0) on Abit AV8 did not work well when I got this matherboard at work, so I disabled this interface in the BIOS and use PCI NIC (fxp0) instead. The problem was that after 3-4 days after boot vge0 suddenly reports "no carrier" and the only way to bring it back was to reboot again. My recent AMD64 X2 upgrade followed with BIOS upgrade (to v2.4) and apparently vge0 was enabled again. Then after couple of days I noticed high CPU (both cores) load: --------------------------------------------- systat: /0 /1 /2 /3 /4 /5 /6 /7 /8 /9 /10 Load Average ||||||||||||| /0 /10 /20 /30 /40 /50 /60 /70 /80 /90 /100 root irq22: pcm XXXXXXXXXXXXXXXXXXXXXXXXX --------------------------------------------- vmstat -i: interrupt total rate irq1: atkbd0 27286 0 irq0: clk 204987922 1000 irq6: fdc0 9 0 irq8: rtc 26234430 127 irq12: psm0 1967122 9 irq14: ata0 46 0 irq18: fxp0 6270446 30 irq20: atapci0 1760077 8 irq22: pcm0 vge0 918487906 4481 irq23: fwohci0 1 0 Total 1159735245 5658 --------------------------------------------- top -S: PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND 34 root -80 -199 0K 16K CPU1 1 214:29 98.05% 98.05% irq22: pcm0 I am not sure if there is a conflict between sound device pcm0 and vge0 but audio output generates low frequency noise, I tried to unload sound module but it did not work reporting "device busy". Probably I can solve this problem by rebooting and disabling vge0 in BIOS but I have two simulations running for couple of days and have no wish to interrupt them. My question is, how can I stop this irq22 process safely? Is it OK to just kill it? PS: I have 5.4-STABLE /Alexander Konovalenko +46-8-5537-8142 (office) +46-7-3752-2116 http://daemon.nanophys.kth.se/~kono Royal Institute of Technology (KTH) Nanostructure Physics Department, Albanova Roslagstullsbacken 21 10691 Stockholm Sweden