I got a report from a user who periodically sees
an oops in this function in our 2.6.14.3 based update.
Does this look familiar to anyone?
Unable to handle kernel NULL pointer dereference
printing eip:
*pde = 00426001
Oops: 0000 [#1]
SMP
Modules linked in: sata_nv libata scsi_mod e1000 nfs lockd nfs_acl sunrpc
CPU: 0
EIP: 0060:[<e09528e6>] Not tainted VLI
EFLAGS: 00010286 (2.6.14-1.1644_FC4smp)
EIP is at scsi_run_queue+0x10/0xaf [scsi_mod]
eax: 00000000 ebx: dd98007c ecx: dffef880 edx: 00000001
esi: ddc99e00 edi: 00000246 ebp: dd9071fc esp: c042af14
ds: 007b es: 007b ss: 0068
Process ksoftirqd/0 (pid: 3, threadinfo=c042a000 task=dfc41ab0)
Stack: dd9071fc dd98007c ddc99e00 00000246 dd9071fc e0952a7d ddc99e00 00000000
00000000 dd98007c e0952e88 00000001 e088260b dcd3a570 c15e3380 dcd3a570
00000000 00000000 00040000 00000024 dd9071fc 00000000 00000000 00000292
Call Trace:
[<e0952a7d>] scsi_end_request+0x83/0xb0 [scsi_mod]
[<e0952e88>] scsi_io_completion+0x29e/0x4d2 [scsi_mod]
[<e088260b>] e1000_clean_rx_irq+0x95/0x4f1 [e1000]
[<e094dcb2>] scsi_finish_command+0x82/0xb5 [scsi_mod]
[<e094db97>] scsi_softirq+0xc0/0x133 [scsi_mod]
[<c02bdf4e>] net_rx_action+0xb7/0x1bb
[<c01258c2>] __do_softirq+0x72/0xdc
[<c0105c43>] do_softirq+0x4b/0x4f
=======================
[<c0125ec2>] ksoftirqd+0x9c/0xe8
[<c0125e26>] ksoftirqd+0x0/0xe8
[<c0133d89>] kthread+0x93/0x97
[<c0133cf6>] kthread+0x0/0x97
[<c0101d5d>] kernel_thread_helper+0x5/0xb
Code: c5 8f df 8b 14 24 8b 42 44 e8 37 b6 9c df 89 44 24 04 89 d8 e8 2e b6 ff ff
eb b1 55 57 56 53 83 ec \
04 89 04 24 8b 80 10 01 00 00 <8b> 38 80 b8 85 01 00 00 00 0f 88 86 00 00 00 8b
47 44 e8 03 b6
The user mentions that the workstation is actually diskless, so I'm
puzzled why we're doing SCSI IO at all, but nevertheless...
I've asked for more info.
Dave
-
: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
