Bug 2092

Summary: cannot mount usb mass storage device with fat
Product: IO/Storage Reporter: Michael PS (ps)
Component: SCSIAssignee: Mike Anderson (andmike)
Status: CLOSED PATCH_ALREADY_AVAILABLE    
Severity: normal    
Priority: P2    
Hardware: i386   
OS: Linux   
Kernel Version: 2.6.2 Subsystem:
Regression: --- Bisected commit-id:
Attachments: patch to 2.6.x kernel that solves the mount problem

Description Michael PS 2004-02-13 17:13:39 UTC 
Distribution:
Hardware Environment: compaq evo n800c laptop
Software Environment: slackware 9.1
Problem Description: cannot mount usb mass storage device with fat

Hi!  i have this iriver mp3 player with ums firmware that act as a usb mass
storage device driver and it got fat filesystem on a eprom of some sort without
any partition.

I can mount it ok under 2.6.1 and all other kernels before, but not under 2.6.2
or 2.6.3-rc2

bash-2.05b# mount -t vfat /dev/sda /mnt/sda
mount: /dev/sda: can't read superblock
bash-2.05b#

output of dmesg after failed mount command:

SCSI device sda: 1024000 512-byte hdwr sectors (524 MB)
sda: assuming Write Enabled
sda: assuming drive cache: write through
 sda:<3>Buffer I/O error on device sda, logical block 0
Buffer I/O error on device sda, logical block 0
 unable to read partition table
FAT: unable to read boot sector

And output of dmesg after pluging in the mp3 player:

Attached scsi removable disk sda at scsi1, channel 0, id 0, lun 0
Attached scsi generic sg0 at scsi1, channel 0, id 0, lun 0,  type 0
WARNING: USB Mass Storage data integrity not assured
USB Mass Storage device found at 3

regards Michael

Steps to reproduce:
Comment 1 Michael PS 2004-02-29 12:27:40 UTC 
Created attachment 2253 [details]
patch to 2.6.x kernel that solves the mount problem
Comment 2 Michael PS 2004-02-29 12:31:18 UTC 
correction:

The player can be mounted under 2.4.x kernel ok, but not under any 2.6.x kernel.

When applying the patch i attached it can be mounted ok under 2.6.x kernel.
I've found the patch in a mailing list i found on google.com

I'm not so good at this kernel hacking but i guess you would wan't to know this 
problem for future kernels.
Comment 3 Alexander Nyberg 2004-12-08 03:41:14 UTC 
In mail exchange Michael tells me it was most likely solved somewhere around
2.6.3 and works as expected in 2.6.9, closing.