2.6.x kernels & PCI-to-PCMCIA bridges [again]
jar
jar at pcuf.fi
Wed May 5 12:44:28 BST 2004
Pavel Roskin wrote:
> Do you kill cardmgr? If I do "cardctl eject", I can remove the socket
> driver (yenta_socket in my case) without killing cardmgr, which doesn't
> seem right. It confuses the kernel, but doesn't cause kernel panic for
> me. If I restart cardmgr after that it fails to adjust resources.
>
> You may want to describe the exact sequence of commands and post the stack
> trace.
I run e.g. "service pcmcia stop" script witch will first kill cardmgr
and then unload modules. The stop case goes as below:
echo -n "Shutting down PCMCIA services: "
if [ -s /var/run/cardmgr.pid ] ; then
PID=`cat /var/run/cardmgr.pid`
kill $PID
# Give cardmgr a few seconds to handle the signal
for N in 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 ; do
kill -0 $PID 2>/dev/null || break
sleep 2
done
fi
killall -q "CardBus Watcher"
######if grep -q "ds " /proc/modules ; then
/sbin/cardctl eject
/sbin/rmmod ds
/sbin/rmmod $PCIC 2>/dev/null || \
/sbin/rmmod yenta_socket 2>/dev/null
/sbin/rmmod pcmcia_core
######fi
echo "done."
rm -f /var/lock/subsys/pcmcia
# we do this because when we stop the service, the ide stop # script
gets run before the card disappears
/usr/sbin/updfstab
EXITCODE=0
;;
The the kernel panic happen:
[root at gw1 etc]# May 2 15:03:40 gw1 kernel: wlan0: Deauthenticate all
stations
May 2 15:03:40 gw1 kernel: wlan0: Deauthenticate all stations
May 2 15:03:40 gw1 arpwatch: pcap_loop: recvfrom: Network is down
May 2 15:03:40 gw1 arpwatch: pcap_loop: recvfrom: Network is down
May 2 15:03:40 gw1 kernel: device wlan0 left promiscuous mode
May 2 15:03:40 gw1 kernel: device wlan0 left promiscuous mode
May 2 15:03:41 gw1 kernel: hostap_cs: detach postponed, 'wlan0' still
locked
May 2 15:03:41 gw1 kernel: hostap_cs: detach postponed, 'wlan0' still
locked
May 2 15:03:41 gw1 kernel: divert: freeing divert_blk for wlan0
May 2 15:03:41 gw1 kernel: divert: freeing divert_blk for wlan0
May 2 15:03:41 gw1 kernel: hostap_cs: Driver unloaded
May 2 15:03:41 gw1 kernel: hostap_cs: Driver unloaded
May 2 15:03:41 gw1 kernel: hostap_crypt: unregistered algorithm 'NULL'
(deinit)
May 2 15:03:41 gw1 kernel: hostap_crypt: unregistered algorithm 'NULL'
(deinit)
May 2 15:03:41 gw1 cardmgr[1729]: exiting
May 2 15:03:41 gw1 cardmgr[1729]: exiting
Unable to handle kernel paging request at virtual address 6b6b6b6b
printing eip:
6b6b6b6b
*pde = 00000000
Oops: 0000 [#1]
CPU: 0
EIP: 0060:[<6b6b6b6b>] Not tainted
EFLAGS: 00010202 (2.6.5-1.339)
EIP is at 0x6b6b6b6b
eax: 6b6b6b6b ebx: 6b6b6b6b ecx: c331bee0 edx: c03b6fd0
esi: 6b6b6b6b edi: c03374a0 ebp: c03b6fd0 esp: c03b6fc8
ds: 007b es: 007b ss: 0068
Process hotplug (pid: 2158, threadinfo=c03b6000 task=c36180b0)
Stack: c012720e c03b6fd0 c03b6fd0 c03b6fd0 c03b6fd8 00000011 c03db3a8
0000000a
00000000 c0122f5d c0a7ef10 00000046 c03b7000 c0108c41
Call Trace:
[<c012720e>] run_timer_softirq+0x217/0x32b
[<c0122f5d>] __do_softirq+0x35/0x73
[<c0108c41>] do_softirq+0x46/0x4d
=======================
[<c010803d>] do_IRQ+0x2f5/0x301
[<c02e8f28>] common_interrupt+0x18/0x20
[<c0143d02>] check_poison_obj+0xa3/0x15b
[<c016d8fd>] getname+0x19/0xb2
[<c0145786>] kmem_cache_alloc+0x77/0x142
[<c016d8fd>] getname+0x19/0xb2
[<c015e11d>] sys_open+0xc/0x7d
[<c02e8d37>] syscall_call+0x7/0xb
Code: Bad EIP value.
<0>Kernel panic: Fatal exception in interrupt
In interrupt handler - not syncing
Regards,
Jar
More information about the linux-pcmcia
mailing list