Re: Kernel 2.4.18 panic on orinoco null event
Hi,
On Sat, Jul 13, 2002, Hanasaki JiJi wrote:
> Thanks for the quick response.. its been quite awhile since I have done
> kernel work. Any idea how I can track down what did the tainting and
Identify set of currently loaded modules ('lsmod')
Reboot the machine (only way to clear tainting). Confirm with 'lsmod' what
kernels are loaded and that it's still not tainted. Then load each of above
modules (from before reboot), and observer tainted status after each, noting
which one taints kernel.
> take it out? This is a kernel built from the src at kernel.org. How
> the heck could I not have the src?
I presumed that the orinoco was a binary driver that you got from vendor,
not from master kernel tree. If that's not the case and you do have source
for it, it's probably just the case (unless one of your other drivers
tainted the kernel) that one of the modules does not have the properly
listed license export (MODULE_LICENSE). Like the following example from
hci_usb.c:
MODULE_AUTHOR("Maxim Krasnyansky <maxk@qualcomm.com>");
MODULE_DESCRIPTION("BlueZ HCI USB driver ver " VERSION);
MODULE_LICENSE("GPL");
<pause>
Yeah, now that I look further, './net/wireless/orinoco.c' looks just fine on
licensing:
static char version[] __initdata = "orinoco.c 0.09b (David Gibson <hermes@gibson.dropbear.id.au> and others)";
MODULE_AUTHOR("David Gibson <hermes@gibson.dropbear.id.au>");
MODULE_DESCRIPTION("Driver for Lucent Orinoco, Prism II based and similar wireless cards");
#ifdef MODULE_LICENSE
MODULE_LICENSE("Dual MPL/GPL");
#endif
So, it was a false alarm, but you'll definitely have to look elsewhere for
your tainting, and then reproduce the kernel panic in an untainted kernel
for the lk dev people to really follow up your report.
Take care.
HTH,
Daniel
> Daniel Freedman wrote:
> >Hi,
> >
> >I don't have much to specifically advise you on this Orinoco issue, but...
> >
> >On Sat, Jul 13, 2002, Hanasaki JiJi wrote:
> >
> >>Help please?!? The wireless card in my laptop + its driver is causing a
> >>kernel panic.
> >>
> >>HW:
> >> laptop with built-in 802.11b card
> >> P4 1.7
> >>OS:
> >> Woody & kernel 2.4.18
> >>
> >>running dhclient-2-2-x causes the following error:
> >>
> >>Unable to handle kernel NULL ponter dereference at virtual
> >>address 0000 0078
> >>Tainted: P
> >
> >
> >In case you weren't aware, the tainted flag above signifies that your
> >kernel
> >is using binary-only (or at least, non-compatible licensed-) drivers, which
> >means that the kernel-developers are probably going to have a hard time
> >helping track this down (since they don't have source for driver).
> >
> >They'll probably ask if you can reproduce it in scenarios when your kernel
> >is not tainted (it stays so till reboot: check with
> >'cat /proc/sys/kernel/tainted')
> >
> >Sorry I can't be of more help, and take care,
> >
> >Daniel
> >
> >
> >>Kernel panic: killing interrupt handler!
> >>eth1: Null event in orinoco_interrupt!
> >>
> >>Thank you.
--
Daniel A. Freedman
Laboratory for Atomic and Solid State Physics
Department of Physics
Cornell University
--
To UNSUBSCRIBE, email to debian-user-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: