Re: 2.1.117, flaky cdrom, frozen mount
Erik Andersen (andersen@inconnect.com)
Wed, 26 Aug 1998 00:25:22 -0600
On Wed, Aug 26, 1998 at 03:50:48AM +0200, Peter Gervai wrote:
> On Tue, Aug 25, 1998 at 02:36:11PM -0600, Erik Andersen wrote:
> >
> > Ok, could you try again under 2.1.117 where you add
> > options cdrom debug=1
>
> Here you are.
>
> Aug 26 03:43:51 BRaiN kernel: hdd: ATAPI 4X CDROM drive, 120kB Cache
> Aug 26 03:43:51 BRaiN kernel: Uniform CDROM driver Revision: 2.13
> Aug 26 03:43:51 BRaiN kernel: cdrom: drive "/dev/hdd" registered
> Aug 26 03:43:51 BRaiN kernel: cdrom: entering cdrom_open
> Aug 26 03:43:51 BRaiN kernel: cdrom: entering open_for_data
> Aug 26 03:43:51 BRaiN kernel: cdrom: drive_status=4
> Aug 26 03:43:51 BRaiN kernel: cdrom: entering cdrom_count_tracks
> Aug 26 03:43:51 BRaiN kernel: cdrom: track 1: format=1, ctrl=4
> Aug 26 03:43:51 BRaiN kernel: cdrom: disc has 1 tracks: 0=audio 1=data 0=Cd-I 0=XA
> Aug 26 03:43:51 BRaiN kernel: cdrom: all seems well, opening the device.
> Aug 26 03:43:51 BRaiN kernel: cdrom: opening the device gave me 0.
> Aug 26 03:43:51 BRaiN kernel: cdrom: door locked.
> Aug 26 03:43:51 BRaiN kernel: cdrom: device opened sucessfully.
> Aug 26 03:43:51 BRaiN kernel: cdrom: Use count for "/dev/hdd" now 1
> Aug 26 03:43:51 BRaiN kernel: VFS: Disk change detected on device 16:40
> Aug 26 03:43:51 BRaiN kernel: cdrom: entering CDROMMULTISESSION
> Aug 26 03:43:51 BRaiN kernel: cdrom: CDROMMULTISESSION sucessful
> Aug 26 03:44:04 BRaiN kernel: hdd: timeout waiting for DMA
> Aug 26 03:44:04 BRaiN kernel: hdd: irq timeout: status=0x58 { DriveReady SeekComplete DataRequest }
> Aug 26 03:44:04 BRaiN kernel: hdd: DMA disabled
> Aug 26 03:44:04 BRaiN kernel: hdd: ATAPI reset complete
> Aug 26 03:44:04 BRaiN kernel: ATAPI device hdd:
> Aug 26 03:44:04 BRaiN kernel: Error: Unit attention -- (Sense key=0x06)
> Aug 26 03:44:04 BRaiN kernel: Power on, reset or bus device reset occurred --
> (asc=0x29, ascq=0x00)
> Aug 26 03:44:04 BRaiN kernel: ISO 9660 Extensions: RRIP_1991A
> Aug 26 03:44:04 BRaiN kernel: VFS: Disk change detected on device 16:40
>
> It seems not to be a problem of cdrom, but something around the IDE,
> maybe ide-cd. But my opinion doesn't count.
> I try to give a debug for it and see. (rebooting... *sigh*)
>
> bye,
> g
This is the _exact_ same problem (with death in the exact same place)
that the other guy who is reporting a problem has seen.
When the CDROMMULTISESSION call was sent down to the ide-cd driver by
the uniform cdrom driver, ide-cd then builds an ATAPI READ_TOC packet,
and sends it on to the ide layer, which stuffs the packet down to the
drive. When ide talks to the drive, the drive chokes and then the ide
layer does its best to get the drive back into a safe, sane state via a
reset.
The only thing I can think of is that recent changes to the ide layer
have somehow hosed things for the ide driver. Could you try a few older
kernels to see if you can narrow down when this started happening? What
compiler are you using to compile the kernel? If 2.7.2.x, have you tried
egcs? If egcs, have you tried gcc 2.7.2.x?
I have copied Andre M. Hedrick, the current ide driver guy, to see if he
has any ideas here.
Andre: we've got another one! Ugh. Any ideas where this could be going
wrong? I'm stumped on this one. ide-cd hasn't changed this stuff for a
long while, so I'm pretty sure this one is in ide. Did DMA or interrupt
handling change recently?
-Erik
--
Erik B. Andersen Web: http://www.inconnect.com/~andersen/
email: andersee@debian.org
--This message was written using 73% post-consumer electrons--
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html