From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262503AbUEOOUv (ORCPT ); Sat, 15 May 2004 10:20:51 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262538AbUEOOUv (ORCPT ); Sat, 15 May 2004 10:20:51 -0400 Received: from h00207813f68b.ne.client2.attbi.com ([24.91.60.206]:20624 "EHLO buddha.badbelly.com") by vger.kernel.org with ESMTP id S262503AbUEOOUs (ORCPT ); Sat, 15 May 2004 10:20:48 -0400 Subject: Re: Burning CDs with a CD-ROM is a bad idea From: Gregory Boyce To: Jens Axboe Cc: linux-kernel@vger.kernel.org In-Reply-To: <20040515104752.GC24600@suse.de> References: <20040515104752.GC24600@suse.de> Content-Type: text/plain Message-Id: <1084630641.1540.8.camel@qube.badbelly.com> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.6 Date: Sat, 15 May 2004 10:17:21 -0400 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, 2004-05-15 at 06:47, Jens Axboe wrote: > On Fri, May 14 2004, gboyce@badbelly.com wrote: > > Hello folks, > > > > Yesterday I made a slight thinko while attempting to burn a CD. Rather > > than specifying dev=/dev/hdd, I add dev=/dev/hdc, which is my CD-ROM > > drive rather than my cd burner. Whoops! > > > > Now, I believe I've done this before, and recieved an error message. > > However, in this particular case with 2.6.6, the system behaved a bit > > different. > > > > bio 00000000, biotail 00000000, buffer 00000000, data 00000000, len 0 > > cdb: 1e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > > hdc: cdrom_pc_intr: The drive appears confused (ireason = 0x02) > > ide-cd: cmd 0x1e timed out > > hdc: lost interrupt > > Looks like one of the burning commands confused the drive so much, that > it now refuses to do anything (the above command is a simple medium > removal prevention command, doesn't even need a data transfer). > > So I don't think this is a kernel problem. Well maybe we could reset the > device and see if it recovers (do you see any resets in the log?) There was a reset early on, but it was before the majority of errors. Here is the begining of problem: May 13 16:38:14 cthulhu hdc: DMA interrupt recovery May 13 16:38:14 cthulhu hdc: lost interrupt May 13 16:38:14 cthulhu hdc: status timeout: status=0xd0 { Busy } May 13 16:38:14 cthulhu hdc: status timeout: error=0x00 May 13 16:38:14 cthulhu hdc: DMA disabled May 13 16:38:14 cthulhu hdc: drive not ready for command May 13 16:38:15 cthulhu hdc: ATAPI reset complete May 13 16:38:15 cthulhu cdrom_pc_intr, write: dev hdc: flags = REQ_STARTED REQ_PC May 13 16:38:15 cthulhu sector 0, nr/cnr 0/0 May 13 16:38:15 cthulhu bio 00000000, biotail 00000000, buffer 00000000, data 00000000, len 0 May 13 16:38:15 cthulhu cdb: 1e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 May 13 16:38:15 cthulhu hdc: cdrom_pc_intr: The drive appears confused (ireason = 0x02) May 13 16:39:15 cthulhu ide-cd: cmd 0x1e timed out May 13 16:39:15 cthulhu hdc: lost interrupt May 13 16:39:15 cthulhu cdrom_pc_intr, write: dev hdc: flags = REQ_STARTED REQ_PC REQ_FAILED May 13 16:39:15 cthulhu sector 0, nr/cnr 0/0 May 13 16:39:15 cthulhu bio 00000000, biotail 00000000, buffer 00000000, data 00000000, len 0 May 13 16:39:15 cthulhu cdb: 1e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 May 13 16:39:15 cthulhu hdc: cdrom_pc_intr: The drive appears confused (ireason = 0x02) There are an additional 700K of errors like this before I finally rebooted the machine the next day. The machine also refused to let me burn CDs on the actual CD burner during this time. -- Greg