Subject: Re: Re: iBook2 build problems
From: Michael Blakeley (mike@blakeley.com)
Date: Mon Jun 18 2001 - 10:02:22 MDT
> From: Derrik Pates <dpates@dsdk12.net>
> To: <mol-general@lists.maconlinux.org>
> Message-ID: <Pine.LNX.4.33.0106170954150.17929-100000@andromeda.dsdk12.net>
> MIME-Version: 1.0
> Content-Type: TEXT/PLAIN; charset=US-ASCII
> Subject: Re: iBook2 build problems
>
> On Sun, 17 Jun 2001, Michael Blakeley wrote:
>
> > ...now if I can just get the airport working.
>
> You may have to jump to a 2.4 kernel, unless the airport driver that's
> built against the orinoco_cs driver works with 2.2.x (which it doesn't
> appear to). If you're daring enough to try it, you can get BenH's current
> tree by doing the following:
>
> mkdir linux-2.4-benh
> rsync -avzc penguinppc.org::linux-2.4-benh linux-2.4-benh/
>
> Then you'll have to setup either a dummy or tap/tun interface, and use
> iptables to masq the MoL session out to the real world. Due to some design
> issues with 802.11, you won't be able to just piggyback the MoL session
> over the wireless interface directly.
Odd, 'cause it seems to be working without that tap-dance :-).
I went to a benh rsync kernel (NB: it's very important to compile
without SMP - for me, this took some doing:
rsync as above
edit Ben's .config to remove SMP from the .config
# I also removed Altivec and added Joliet
cp .config /tmp/$$ && make distclean && cp /tmp/$$ .config
make oldconfig # answering N to SMP
make dep && make && make modules && make modules_install
cp vmlinux /boot/ && cp System.map /boot/ # or wherever you like
).
Then I rebooted, tested the aiport, unplugged the ethernet cable,
added eth1 to the molrc, and ran startmol. MOL _thinks_ it's attached
to the ethernet - that confused me a little. But look: no wires.
Come to think of it, there's a console message about my MOL IP when I
run startmol. Maybe MOL does the mapping itself now?
-- Mike
This archive was generated by hypermail 2a24 : Mon Jun 18 2001 - 09:08:15 MDT