From owner-freebsd-bugs Tue Nov 30 21:40: 7 1999 Delivered-To: freebsd-bugs@freebsd.org Received: from freefall.freebsd.org (freefall.FreeBSD.ORG [204.216.27.21]) by hub.freebsd.org (Postfix) with ESMTP id A6FC515A6C for ; Tue, 30 Nov 1999 21:40:04 -0800 (PST) (envelope-from gnats@FreeBSD.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.9.3/8.9.2) id VAA84219; Tue, 30 Nov 1999 21:40:04 -0800 (PST) (envelope-from gnats@FreeBSD.org) Date: Tue, 30 Nov 1999 21:40:04 -0800 (PST) Message-Id: <199912010540.VAA84219@freefall.freebsd.org> To: freebsd-bugs@FreeBSD.org Cc: From: "Kenneth D. Merry" Subject: Re: kern/15198: CD changer autodetection doesn't work, causing SCSI timeout [PATCH] Reply-To: "Kenneth D. Merry" Sender: owner-freebsd-bugs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org The following reply was made to PR kern/15198; it has been noted by GNATS. From: "Kenneth D. Merry" To: gdonl@tsc.tdk.com Cc: FreeBSD-gnats-submit@FreeBSD.ORG Subject: Re: kern/15198: CD changer autodetection doesn't work, causing SCSI timeout [PATCH] Date: Tue, 30 Nov 1999 22:37:50 -0700 (MST) Don Lewis wrote... > > >Number: 15198 > >Category: kern > >Synopsis: CD changer autodetection doesn't work, causing SCSI timeout [PATCH] [ ... ] > FreeBSD 3.3-STABLE of the 19991129 vintage with a PIONEER CD-ROM DRM-6324X > LUN-based CD-ROM changer connected via an aic SCSI controller. > > >Description: > > A SCSI bus timeout occurs when the driver attempts to access the > second slot in the CD changer (cd2). If the cdX devices are listed > in /etc/fstab, the machine totally wedges. If no attempt is made > to mount the cdX devices, the machine is otherwise usable. [ ... ] > cd1 at aic0 bus 0 target 6 lun 0 > cd1: Removable CD-ROM SCSI-2 device > cd1: 5.000MB/s transfers (5.000MHz, offset 8) > cd1: cd present [12205 x 2048 byte records] > (cd2:aic0:0:6:1): ccb 0xc0871800 - timed out, phase 84 > cd2 at aic0 bus 0 target 6 lun 1 > cd2: Removable CD-ROM SCSI-2 device > cd2: 5.000MB/s transfers (5.000MHz, offset 8) > cd2: cd present [36007 x 2048 byte records] [ ... ] > >Fix: > > Adding a quirk entry to the driver fixes the problem, even though the > comments say this shouldn't be necessary. Indeed, I'm certainly surprised that adding a quirk fixed the problem. I suppose this device must take a long time to respond to a read capacity. With most changers, it's no big deal if the changer gets switched into the changer scheduling code with lun 1 instead of lun 0. You usually don't see any sort of trouble. With the patch, do the lockups still happen when you boot the machine? I wouldn't expect the patch to have much effect on that, since all LUNs of the changer will be run through the changer scheduling mechanism once the second lun is detected... Unless the CDs are being mounted as they are found. In any case, the lockup shouldn't happen, although finding the lockup might be rather difficult. Can you break into the debugger when the lockup happens and get a stack trace? Ken -- Kenneth Merry ken@kdm.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message