RE: [microblaze-uclinux] EXT2 filesystem on RAMDISK

["Leonid" <Leonid@xxxxxxxxx>]

On Thursday, February 08, 2007 3:56 PM John Williams wrote:
> If no filesystem
> "claims" the ROOT_DEV, then you get the VFS panic you are reporting.

Yes, I understand that - nobody "claims" /dev/ram which has 01:00
ROOT_DEV. How filesystems claim certain devices? Where should I look?

> > I suspect it's something in the do_mount logic which is wrong - the
> > question is whether it's intrinsic uClinux bug or something that has
> > been introduced by Petalogix or just my other definitions are wrong.


> If you have compiled ext2 support into your kernel (double check -
there's > no error output here from ext2 subsystem so I'm dubious), then
the problem > is that the rootfs image itself is either corrupt, or not
where the kernel > expects it.

Ext2 support is in there - see image.elf content:
[leonid@mylinux petalinux-dist]$ /opt/mb/bin/mb-nm images/image.elf |
grep ext2_
2211a0fc t __exitcall_exit_ext2_fs
2214e1f8 ? __initcall_init_ext2_fs
22119a70 t exit_ext2_fs
2205608c T ext2_add_link
22057bd0 t ext2_alloc_block
...........................

I added some debug printouts to do_mount. root_mount_data is NULL for
some reason:

create_dev:416 name /dev/root devfs_name <NULL> dev 0x0100 err 0
mount_root Line 777 ROOT_DEV 0x04100
create_dev:416 name /dev/root devfs_name ram dev 0x0100 err 0
mount_block_root fs_names /dev/root name /dev/root flags 0x8000
mount_block_root type ext2 root_mount_data <NULL> err -22
Kernel panic: VFS: Unable to mount root fs on 01:00

Note that one ramdisk does already exist - /var is mounted there. Could
it be a problem?

Regards,

Leonid.

___________________________
microblaze-uclinux mailing list
microblaze-uclinux@xxxxxxxxxxxxxx
Project Home Page : http://www.itee.uq.edu.au/~jwilliams/mblaze-uclinux
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/