Re: smart media changing problem

Gerry Tool wrote:

> hub.c: USB hub found
> hub.c: 4 ports detected
> hub.c: USB hub found
> hub.c: 2 ports detected
> hub.c: USB hub found
> hub.c: 2 ports detected
> hub.c: USB hub found
> hub.c: 2 ports detected
> hub.c: USB hub found
> hub.c: 2 ports detected

That's a lot of USB ports :)

> hub.c: USB new device connect on bus4/1, assigned device number 2
> usb.c: USB device 2 (vend/prod 0x4b8/0x10b) is not claimed by any active driver.
> hub.c: USB new device connect on bus4/2, assigned device number 3
> usb.c: USB device 3 (vend/prod 0x66b/0x105) is not claimed by any active driver.
 > hub.c: USB new device connect on bus2/1, assigned device number 2
> usb.c: USB device 2 (vend/prod 0xaec/0x5010) is not claimed by any active driver.

A lot of unrecognized devices. It might help to disconnect them
for a while. Probably not, but maybe.

> Initializing USB Mass Storage driver...
> usb.c: registered new driver usb-storage
> scsi1 : SCSI emulation for USB Mass Storage devices
>   Vendor: Olympus   Model: Camedia MAUSB-2   Rev: 0100
>   Type:   Direct-Access                      ANSI SCSI revision: 02

Seams detected OK.

 > Attached scsi removable disk sda at scsi1, channel 0, id 0, lun 0

Knows it's removable too.
I wonder if issuing and eject might help?

> usb-uhci.c: interrupt, status 3, frame# 494

Could be an issue. USB debugging code might be handy here.

> SCSI device sda: 131072 512-byte hdwr sectors (67 MB)
> sda: Write Protect is on
>  sda: sda1
> usb-uhci.c: interrupt, status 2, frame# 167
>  I/O error: dev 08:01, sector 0
> FAT: unable to read boot sector

It tried to read the partition table, and couldn't.
I/O error reading sector 0.

This looks like the problem Mark C. was having with
his USB attached camera. It was noted on the kernel-list
that Smart media doesn't alway put the partition table
in sector 0, and some don't even implement the first few
sectors.

> On Thu, 2002-09-19 at 03:49, Dennis Gilmore wrote:
> [root@gstpc gerry]# /sbin/fdisk -l /dev/sda
> 
> Disk /dev/sda: 4 heads, 16 sectors, 256 cylinders
> Units = cylinders of 64 * 512 bytes
> 
>    Device Boot    Start       End    Blocks   Id  System
> /dev/sda1   *         1       250      7987+   1  FAT12
> [root@gstpc gerry]# mount /mnt/smedia
> mount: block device /dev/sda1 is write-protected, mounting read-only
> [root@gstpc gerry]# umount /mnt/smedia
> 
> 	remove and reinsert the card
> 
> [root@gstpc gerry]# mount /mnt/smedia
> mount: block device /dev/sda1 is write-protected, mounting read-only
> mount: wrong fs type, bad option, bad superblock on /dev/sda1,
>        or too many mounted file systems

If you checked dmesg here, I'd expect to se the I/O error
for sector 0 again.

> [root@gstpc gerry]# /sbin/fdisk -l /dev/sda
> 
> Disk /dev/sda: 4 heads, 16 sectors, 256 cylinders
> Units = cylinders of 64 * 512 bytes
> 
>    Device Boot    Start       End    Blocks   Id  System
> /dev/sda1   *         1       250      7987+   1  FAT12

I wonder if the kernel is giving the old partition data to
fdisk. Some sort of caching in the SCSI layer maybe.

I did notice that in the above dmesg, the SCSI layer never rescanned
the device. I had a removable SCSI disk sometime ago (SyQuest)
and always used the eject command. I think the kernel rescanned
the device when I inserted a disk. It probably should do the same here.

	-Thomas