Re: Re: remote shutdown

On Thu, 14 Aug 2003 16:17:42 -0400 (EDT)
Donald Becker <[email protected]> wrote:

> On Thu, 14 Aug 2003, Erik Thiele wrote:
> 
> > right now i tried the ftp://ftp.scyld.com/pub/network/3c59x.c
> > with it's pci-scan.[ch] and kern-compat.h
> 
> OK, you seem to be actually interested in solving what is now evident as
> a hard problem.
> 
> The tool I used to develop and test WoL is 'pci-config'
> You can get it from
>    ftp://ftp.scyld.com/pub/diag/pci-config.c
> 
> [[ Hmmm, there is actually a newer version, but the only changes are for
> the Operton. ]]
> 
> This code allows reading the PCI configuration space registers.
> I use it to monitoring if the card is trying to raise a wake-up sign.
> It has an extra piece of code expecially for the 3Com chips enabled by
>     --set-WOL 
> With this you can
>   Find the board index by running 'pci-config'
>   Run 'pci-config -#<INDEX>' to see the current board state
>   Test that ether-wake triggers the PME (Power Management Event) bit
>   load, start, stop and unload the driver
>   pci-config -#<N> --set-WOL
>   test again to see if  ether-wake triggers the PME bit.


attached is the output of pci-config
then pci-config -a -#1
...
pci-config -a -#9
then
cat /proc/pci

i tried to put all cards to sleep, and sent ether-wake packets.
i did not find a "PME" bit.??
only one card could be put to sleep, but it had no effects other than
it's state switched to D3. i.e. the network was still online.

there is no card with a "boot rom", even though i am sitting
in front of that machine and it has no mass storage devices
attached and definitly booted by bootrom :)


i blindly did --set-wol on all cards and then poweroff -f.
the card staid on-line but etherwake didn't work


???


cu
erik


-- 
Erik Thiele
Horst Thiele Maschinenbau-Hydraulische Geräte GmbH
Im Kampfrad 2 - 74196 Neuenstadt
Tel.: 07139/4801-19
Fax.: 07139/4801-29
email: [email protected]
Internet: http://www.thiele-hydraulik.de/