loop-aes on 2.6.0-test6 and higher

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



	I've been having great success using loop-aes with my 2.4 and 2.6 test
series kernel but have recently run into problems with 2.6.0-test6 and
higher.  I've tried this with 4 different kernel versions above
2.6.0-test5-mm4 and the results are the nearly the same.  When I build
loop-aes it compiles fine but when I actually try to mount I get the
following:

loop: loaded (max 8 devices)
Unable to handle kernel NULL pointer dereference at virtual address
00000028
 printing eip:
f9b3626f
*pde = 00000000
Oops: 0000 [#1]
CPU:    0
EIP:    0060:[<f9b3626f>]    Not tainted VLI
EFLAGS: 00010282
EIP is at lo_ioctl+0x2b/0x132 [loop]
eax: 00000000   ebx: f54f3224   ecx: f54f3100   edx: f6e22b80
esi: f54f3230   edi: 00000003   ebp: 00004c00   esp: f6e4bf4c
ds: 007b   es: 007b   ss: 0068
Process mount (pid: 6764, threadinfo=f6e4a000 task=f55a5310)
Stack: f6e729c0 00000282 f6e729c0 c0148f17 00004c00 00000003 f9b36244
f54f3100 
       c02265e0 f54f3100 f6e22b80 00004c00 00000003 f7881d00 00000003
f6e22b80 
       00004c00 ffffffe7 c01637aa f54f3100 f6e22b80 00004c00 00000003
c01493a2 
Call Trace:
 [<c0148f17>] unmap_vma+0x40/0x7d
 [<f9b36244>] lo_ioctl+0x0/0x132 [loop]
 [<c02265e0>] blkdev_ioctl+0x92/0x3bd
 [<c01637aa>] sys_ioctl+0xd5/0x233
 [<c01493a2>] sys_munmap+0x44/0x64
 [<c02b774a>] sysenter_past_esp+0x43/0x65

Code: 83 ec 20 89 74 24 14 89 7c 24 18 89 6c 24 1c 89 5c 24 10 8b 44 24
24 8b 7c 24 30 8b 98 28 01 00 00 8b 6c 24 2c 8b 43 40 8d 73 0c <8b> 50
28 89 f1 ff 4b 0c 0f 88 7a 02 00 00 8b 43 40 8b 40 24 8b 

	The mount fails and I can't even remove the loop module because it's
still in use somewhere.  I'm using the mm series on an x86 gentoo box. 
I've recently "upgraded" to gcc 3.3.1-r5 and glibc 2.3.2-r6 and these
haven't helped any.  I'm using loop-aes 1.7e with the 20030922 update
patch but I'm still getting the same errors.  Just about all of my
software is bleeding edge so this could certainly be part of the problem
but I can still build a working cryptoloop module for 2.6.0-test5-mm4
(my last known working kernel for cryptoloop).  If there are any
unreleased patched out or anything I can do to help test this I would be
happy to contribute.
	Thank you,
	Dan Elder

-- 
----------------------------------------------
Dan Elder, Software Developer, Novacoast, Inc.
Voice: (805) 884-4152
Mobile: (805) 453-6563
E-mail: delder@xxxxxxxxxxxxx
----------------------------------------------

Attachment: signature.asc
Description: This is a digitally signed message part


[Index of Archives]     [Kernel]     [Linux Crypto]     [Gnu Crypto]     [Gnu Classpath]     [Netfilter]     [Bugtraq]
  Powered by Linux