From: Knut N. <knu...@un...> - 2003-03-30 23:27:58
|
Am Fre, 2003-03-28 um 21.59 schrieb Pavel Machek: > > Now, I would really appreciate any hint, how I can gather more > > information. Is there any chance to find out wether it actually jumps > > into wakeup.S after resume? > > Few leds on parallel port may do the trick... along with outb() in wakeup.S. Ole Rohne told me to use lcall $0xffff, $0 in wakeup.S, which I did in the first place and nothing happened. Since according to him this is supposed to reboot the computer I concluded that there is either something wrong with the wakeup-vector or with _WAK (there is no BFS in my DSDT). Thus I tried to jmp to the wakeup-vector from acpi_enter_sleep_state wich worked (besides it led to a general protection fault after a few instructions - I guess thats because I am not in real mode?!). So it seems it never even gets there when resuming. So I did - as you suggested - disassebmle the DSDT and had a look at the WAK method (attached) - but everything seems fine as far as I can tell...so I am totally stuck now. Next thing I wanted to try is to put system in real mode before jumping to the wakeup-vector just to have a look if that really is the problem... Any suggestions? Thank you for your help anyway. I really appreciate it. -Knut -- "Marriage is the alliance of two people,one of whom never remembers birthdays and the other never forgets them." - Ogden Nash |