cryptoloop hard lockup 2.6.0-test5-mm4

[Peter Lieverdink]

I just had a play with an article from an old LinuxJournal, to see if I 
could set up a cryptoloop homedir on my Debian unstable box. Found making 
an fs on the loop device locks the kernel hard. (need reset button) When 
using loopback without crypto it all works fine.

Ayone else have the same experience?

Using util-linux 2.12, aes (same prob with twofish, blowfish).

Crash dump is attached, as is kernel config. When not running uml-switch, 
the crash still occurs but doesn't print the VM: messages. (I don't have a 
swap partition anyway, so what's that all about?)

- Peter.kahlua:~# modprobe cryptoloop
loop: loaded (max 8 devices)
kahlua:~# modprobe aes
kahlua:~# losetup -e aes -k 128 /dev/loop0 /home/cafuego.img
kahlua:~# mkfs -t ext2 /dev/loop0
mke2fs 1.35-WIP (21-Aug-2003)
Filesystem label=
OS type: Linux
Block size=1024 (log=0)
Fragment size=1024 (log=0)
131072 inodes, 524288 blocks
26214 blocks (5.00%) reserved for the super user
First data block=1
64 block groups
8192 blocks per group, 8192 fragments per group
2048 inodes per group
Superblock backups stored on blocks:
        8193, 24577, 40961, 57345, 73729, 204801, 221185, 401409

Writing inode tables: done
Writing superblocks and filesystem accounting information: do_wp_page: bogus page at address 0804d056
VM: killing process uml_switch
swap_dup: Bad swap file entry 90885fdb
VM: killing process uml_switch
swap_dup: Bad swap file entry 286be58f
VM: killing process uml_switch
Debug: sleeping fucntion called from invalid context at include/linux/rwsem.h:43
in_atomic():0, irqs_disabled():1
Call Trace:

Attachment: config-2.6.0-test5-mm4
Description: Binary data