2.6.0 kernel panic with megaraid driver
I get the following panic on a system with linux-2.6.0 when I create a significant amount of i/o on a lvm2 device using drives through a megaraid controller:
------------[ cut here ]------------ kernel BUG at include/linux/blkdev.h:550! invalid operand: 0000 [#1] CPU: 0 EIP: 0060:[<c0272077>] Not tainted EFLAGS: 00010046
EIP is at scsi_request_fn+0x368/0x3a8
eax: 00000000 ebx: d2ffbd90 ecx: 000040d5 edx: d2ffbd90
esi: f7701c00 edi: f755d000 ebp: f7734a00 esp: f7533e7c
ds: 007b es: 007b ss: 0068
Process scsi_eh_3 (pid: 31, threadinfo=f7532000 task=f76c26b0)
Stack: f7734a00 00000202 f7734b50 f7532000 f7532000 f7532000 f7532000 f7701d84
d2ffbd90 f7734a00 f755d000 00000202 c023022e f7734a00 d2ffbd90 00000292
e5a46e00 f7701c00 e5a46e00 f755d000 00001055 c0270bb1 f7734a00 d2ffbd90
Call Trace:
[<c023022e>] blk_insert_request+0x93/0xe1
[<c0270bb1>] scsi_queue_insert+0x75/0xa7
[<c026cd0e>] scsi_dispatch_cmd+0x1b0/0x21b
[<c026f530>] scsi_times_out+0x0/0x50
[<c026cfef>] scsi_softirq+0xb6/0xd1
[<c0127ca7>] do_softirq+0xc3/0xc5
[<c010d4c6>] do_IRQ+0x152/0x1a6
[<c010b70c>] common_interrupt+0x18/0x20
[<c027007b>] scsi_eh_bus_reset+0x28/0x10d
[<c0109c10>] __down_failed_interruptible+0x0/0xc
[<c0270af4>] .text.lock.scsi_error+0xb9/0xc1
[<c0270727>] scsi_error_handler+0x0/0x142
[<c0108b11>] kernel_thread_helper+0x5/0xb
Code: 0f 0b 26 02 1f 35 35 c0 e9 58 fd ff ff 89 5c 24 04 89 2c 24
<0>Kernel panic: Fatal exception in interrupt In interrupt handler - not syncing
The following megaraid driver output preceeds the panic on the console:
megaraid: ABORTING-a28b2 cmd=2a <c=0 t=11 l=0>
megaraid: ABORTING-a28b2[38], fw owner.
megaraid: ABORTING-a28b3 cmd=2a <c=0 t=11 l=0>
megaraid: ABORTING-a28b3[35], fw owner.
[a few more screens of this]
megaraid: reservation reset failed.
megaraid: RESET-a28b2 cmd=2a <c=0 t=11 l=0>
megaraid: RESET-a28b2[38], fw owner.
megaraid: reservation reset failed.
megaraid: RESET-a28b2 cmd=2a <c=0 t=11 l=0>
megaraid: RESET-a28b2[38], fw owner.
megaraid: reservation reset failed.
I've been able to reproduce the problem by untarring several large files on the lvm2 volume simultaneously.
Let me know if there's any additional information I can give.
Signature · 6 lines
Mark Wong - - [email protected] Open Source Development Lab Inc - A non-profit corporation 12725 SW Millikan Way - Suite 400 - Beaverton, OR 97005 (503) 626-2455 x 32 (office) (503) 626-2436 (fax) http://developer.osdl.org/markw/