At 15:11 Uhr +0100 5.11.2005, Martin Husemann wrote:
>On Sat, Nov 05, 2005 at 01:11:14PM +0100, Hauke Fath wrote:
>> db{1}> ps
>>  PID           PPID     PGRP        UID S   FLAGS LWPS          COMMAND
>>WAIT
>>  29348            1    29348          0 2  0x4002    1           reboot
>>stp4020
>
>This has to be the stp4020_delay() call at the end of the
>stp4020_chip_socket_disable() function. Do you have HZ option in your
>kernel config?

No.

>This could make the "ticks" calculation overflow in line
>1119 of stp4020.c. Otherwise I do not see a good reason for tsleep() to
>sleep too long here.

Hmmm...
	hauke

--
"It's never straight up and down"     (DEVO)