Hi,
a few days ago I got an oops in the crypt module. This is with an
nvidia-tainted kernel, so I didn't file a bugzilla report, but
Geert and Milan suggested that I post it here if the oops isn't
related to the nvidia driver. Of course, I can't be sure, but
I've been using both nvidia-drivers and dm/crypt since years
and never got this problem, and there's no indication in the bt
that nvidia is any way involved.
I couldn't reproduce the problem with or without the nvidia
driver (173.14.17 for an FX5500), nor can I say which of my
input exactly caused it (done by a script), but it was when
"destroying" a loop-mounted ~4GB file, ext3 formatted and aes
encrypted, so one of these:
umount ...
cryptsetup remove ...
losetup -d ...
This has worked for years, so it's probably not some recent
user error. The next time I mounted/unmounted the file it
worked flawlessly, as it used to. System:
Intel P4, 2.4G
Linux 2.6.28.5 (vanilla)
gcc 4.3.1 (openSuSE 11.0)
libc 2.8
All modules compiled in, except for the nvidia module and
scsi_wait_scan (for historic reasons).
One thing might be of interest: I had the (desktop-)system
hibernated a few times in a row (2 or 3 times?). Or maybe
it was just bitflipping due to cosmic rays? :-)
BUG: unable to handle kernel NULL pointer dereference at 00000004
IP: [<c0158a75>] mempool_free+0xd/0x9a
*pde = 00000000
Oops: 0000 [#1] PREEMPT
last sysfs file: /sys/devices/virtual/block/dm-1/range
Modules linked in: nvidia(P) scsi_wait_scan
Pid: 19181, comm: loop1 Tainted: P (2.6.28.5 #4) MS-6788
EIP: 0060:[<c0158a75>] EFLAGS: 00010286 CPU: 0
EIP is at mempool_free+0xd/0x9a
EAX: d36af0d4 EBX: 00000000 ECX: 00000001 EDX: 00000000
ESI: d36af0d4 EDI: 00000000 EBP: 00000000 ESP: f621bf58
DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
Process loop1 (pid: 19181, ti=f621a000 task=d98537b0 task.ti=f621a000)
Stack:
c03b3a94 c03b3bb6 00000000 c0198dfb f71a13f0 c032ca0a 00336000 00000000
00000000 f723dd08 f723dc00 c72c1780 00001000 00000001 00000001 00000000
c032c570 00000000 00000000 00000000 f621bfb4 08005000 00000000 f723dc00
Call Trace:
[<c03b3a94>] crypt_dec_pending+0x28/0x4e
[<c03b3bb6>] crypt_endio+0x0/0xde
[<c0198dfb>] bio_endio+0x19/0x30
[<c032ca0a>] loop_thread+0x2f6/0x42d
[<c032c570>] do_lo_send_aops+0x0/0x1a4
[<c032c714>] loop_thread+0x0/0x42d
[<c0131c38>] kthread+0x37/0x5f
[<c0131c01>] kthread+0x0/0x5f
[<c0103d93>] kernel_thread_helper+0x7/0x14
Code: 08 08 74 05 e8 79 45 37 00 8b 04 24 8b 50 0c 89 f0 8b 0c 24 ff 51 14
31 c0 e9 90 fe ff ff 56 53 83 ec 04 89 c6 89 d3 85 c0 74 54 <8b> 42
04 3b 02 7d 6b 9c 59 fa 89 e0 25 00 e0 ff ff 83 40 14 01×
EIP: [<c0158a75>] mempool_free+0xd/0x9a SS:ESP 0068:f621bf58
---[ end trace 916d6e0bbba638c2 ]---
m.
--
To unsubscribe from this list: send the line "unsubscribe linux-crypto" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
