Here is what I could transcribe (manually):
Call trace: [<c012daee>] [<c012de34>] [c012e1a5>] [<c0109dd0>]
Unable to handle NULL pointer dereference at virtual address 00000004
current->tss.cr3= 00101000, %cr3=00101000
*pde=00000000
oops: 0002
CPU 0
EIP 0010 [<c0121d75>]
EFLAGS: 00010046
eax: c02c89c0 ebx: 00001eec ecx: ffffffff edx: 00000000
esi: 00003dd8 edi: c021394c ebp: 00000202 esp: c009ddfc
ds: 0018 es: 0018 ss: 0018
Process init (pid: 1, process nr: 1, stackpage = c009d000
...
Here's my configuration:
PPro 180 Mhz / 64 MB
hda: Western Digital 3.2 GB (not mounted at time of crash)
hdb: Mitsumi CD-R 8/4 under scsi emulation on /dev/scd0 (thhis is the
cdrom I was using cdparanoia on when the crash happened)
hdc: Quantum Fireball 13 GB (where / is)
hdd: Panasonic 4x cdrom under scsi emulation on /dev/scd1 (don't know if
it was mounted)
two PCI NE2000 cards (doing ip forwarding)
1 SB 16 ISA PnP
1 SB PCI 64
Before the crash, the driver has been acting a bit strangely (work
sometimes, sometimes doesn't). The crash occured exactly at the time I
tried to stop (^C) cdparanoia. Also, I have had other problems on this
machine which *might* be related:
1) What seems to be huge memory leaks in the kernel (I kill everything
(X, nfs, smb, ...), but the system is still using ~30 Meg more memory
(excluding cached data) than at boot and my system can hardly do
anything). It happened twice under 2.2.9, it's too soon to know if 2.2.10
fixed it.
2) I've found the following oops in a log (2.2.9):
Unable to handle kernel paging request at virtual address 001bd016
current->tss.cr3 = 03790000, %cr3 = 03790000
*pde = 00000000
Oops: 0002
CPU: 0
EIP: 0010:[opl3:opl3_detect_R67481767+-18308987/5620]
EFLAGS: 00010246
eax: c36b7300 ebx: 001bd016 ecx: 001bd016 edx: 00001606
esi: c0126884 edi: 0000020c ebp: c36b7300 esp: c3d25e5c
ds: 0018 es: 0018 ss: 0018
Process ld (pid: 21339, process nr: 96, stackpage=c3d25000)
If you need more info, just tell me. Or if you think of anything I can
try to find (reproduce) the bug.
Jean-Marc
-------------------------------------------------------------------------
Jean-Marc Valin O'TOOL'S COMMENTARY ON MURPHY'S LAWS
valj01@gel.usherb.ca -Murphy was an optimist
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/