From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932759AbWJGH1L (ORCPT ); Sat, 7 Oct 2006 03:27:11 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932760AbWJGH1L (ORCPT ); Sat, 7 Oct 2006 03:27:11 -0400 Received: from www.tammen.de ([62.206.99.154]:59311 "EHLO mail.tammen.de") by vger.kernel.org with ESMTP id S932759AbWJGH1K convert rfc822-to-8bit (ORCPT ); Sat, 7 Oct 2006 03:27:10 -0400 From: Heinz Ulrich Stille To: linux-kernel@vger.kernel.org Subject: it821x eats CPU? Date: Sat, 7 Oct 2006 09:27:13 +0200 User-Agent: KMail/1.9.4 X-Face: 4mP]9aFSZN&|I]TOVXU]ip[QEBP^RNKxLkj?@0A{KRjs9>QuA{bW MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8BIT Content-Disposition: inline Message-Id: <200610070927.13713.hus@design-d.de> X-DSPAM-Result: Whitelisted X-DSPAM-Processed: Sat Oct 7 09:27:08 2006 X-DSPAM-Confidence: 0.8517 X-DSPAM-Improbability: 1 in 575 chance of being spam X-DSPAM-Probability: 0.0000 X-DSPAM-Signature: 452756cc46252227431997 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi! All I wanted was a cheap IDE controller to attach a spool disk for my backups. What I got was a RAID controller with the it8212 chip. Ok, I thought, just ignore that. The module was loaded and the disk mounted without a hitch. But as soon as the backup started writing to the disk, CPU usage went up to about 100%, a lot (rough estimage 50%) of that "system". Strange enough in itself, but it did not only affect the one process actually accessing that disk, but apparently all processes doing I/O (or perhaps doing I/O to any IDE device?). Most notably one ogg123 (feeding on hold music to our pbx), which usually uses about 5% CPU, went up to 25%... Even if the driver is strange, what takes up the "user" CPU cycles? Looking through the logs I notices that it821x was in "smart" mode, so I restarted the system with "noraid=1" to get into "pass through". Now everything is back to normal. A large dd did about 40MB/s without disturbing other processes. Wasn't smart mode the one supposed to be the one reducing CPU load? BTW, my hardware setup (lspci): 00:00.0 Host bridge: nVidia Corporation nForce2 AGP (different version?) (rev c1) 00:00.1 RAM memory: nVidia Corporation nForce2 Memory Controller 1 (rev c1) 00:00.2 RAM memory: nVidia Corporation nForce2 Memory Controller 4 (rev c1) 00:00.3 RAM memory: nVidia Corporation nForce2 Memory Controller 3 (rev c1) 00:00.4 RAM memory: nVidia Corporation nForce2 Memory Controller 2 (rev c1) 00:00.5 RAM memory: nVidia Corporation nForce2 Memory Controller 5 (rev c1) 00:01.0 ISA bridge: nVidia Corporation nForce2 ISA Bridge (rev a4) 00:01.1 SMBus: nVidia Corporation nForce2 SMBus (MCP) (rev a2) 00:02.0 USB Controller: nVidia Corporation nForce2 USB Controller (rev a4) 00:02.1 USB Controller: nVidia Corporation nForce2 USB Controller (rev a4) 00:02.2 USB Controller: nVidia Corporation nForce2 USB Controller (rev a4) 00:04.0 Ethernet controller: nVidia Corporation nForce2 Ethernet Controller (rev a1) 00:05.0 Multimedia audio controller: nVidia Corporation nForce Audio Processing Unit (rev a2) 00:06.0 Multimedia audio controller: nVidia Corporation nForce2 AC97 Audio Controler (MCP) (rev a1) 00:08.0 PCI bridge: nVidia Corporation nForce2 External PCI Bridge (rev a3) 00:09.0 IDE interface: nVidia Corporation nForce2 IDE (rev a2) 00:1e.0 PCI bridge: nVidia Corporation nForce2 AGP (rev c1) 01:06.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8129 (rev 10) 01:07.0 SCSI storage controller: Adaptec AHA-2940U/UW/D / AIC-7881U (rev 01) 01:08.0 Multimedia controller: Philips Semiconductors TriMedia TM-1300 (rev 82) 01:09.0 RAID bus controller: Silicon Image, Inc. SiI 3112 [SATALink/SATARaid] Serial ATA Controller (rev 02) 01:0a.0 RAID bus controller: (rev 13) 01:0b.0 Mass storage controller: Silicon Image, Inc. SiI 3114 [SATALink/SATARaid] Serial ATA Controller (rev 02) 02:00.0 VGA compatible controller: nVidia Corporation NV11 [GeForce2 MX/MX 400] (rev b2) The one with a name to long for the buffer is the it8212. Root partition is on a raid 1 spanning hda and hdc, the spool disk is hde. MfG, Ulrich -- Heinz Ulrich Stille / Tel.: +49-541-9400473 / Fax: +49-541-9400450 design_d gmbh / Wilhelmstr. 16 / 49076 Osnabrück / www.design-d.de