Cisco Aironet 350 and FC2

[Matthew Saltzman]

On Tue, 27 Jul 2004, Steven Kugelmass wrote:

> Matthew Saltzman said:
> > On Sun, 25 Jul 2004, Mail Lists wrote:
> >
> >> On Sun, Jul 25, 2004 at 10:28:58PM -0400, Steven Kugelmass wrote:
> >> > Hello.  I'm having trouble with my airo card using the later kernel.
> >>  ...
> >>
> >> > kernel-2.6.6-1.435.2.3
> >> >     Hangs.  Plug it in, /var/log/messages says:
> >>   ...
> >> > That's it.  At that point, ifconfig hangs, iwconfig hangs.  The
> >> machine
> >>
> >>   I have similar trouble here - on both dell
> >>   lattitude c600 and thinktop t42. Very nasty.
> >>
> >>   Be nice to have aironet working again ...
> >
> > Is this the Aironet 350 Mini-PCI?  I have that working just fine on a
> > Thinkpad T41 with FC2 (though it took a bit of fiddling).
> >
> > Tell us more about what you've done to configure it and what you've tried
> > to fix it.  What firmware version are you using?
> >
> > --
> > 		Matthew Saltzman
> >
> > Clemson University Math Sciences
> > mjs AT clemson DOT edu
> > http://www.math.clemson.edu/~mjs
> >
>
>
> I'm using a pcmcia aironet card, model pcm-352.  What kind of fiddling
> did you have to do?

When I installed FC2, the detection didn't work correctly, and I got
incorrect modules for the onboard Ether as well as the Mini-PCI wireless.
I don't have a recipe for solving it, but I ended up deleting both
interfaces, putting aliases in /etc/modprobe.conf for both interfaces
(airo_cs for the wireless), and running kudzu and system-config-network a
few times to get the right combination of devices and drivers.  I may even
have edited /etc/sysconfig/network-scripts/ifcfg-ethX by hand to get the
HWADDR fields correct.

>
> I've tried the following firmware versions:
>
> $ ls
> total 792
> 132 45c42530.img*  132 45c52017.img*  132 45c54010.img*
> 132 45c50220.img*  132 45c53017.img*  132 45c54100.img*
>
> and none of them work with the later kernel.

Sorry, don't know much about the PC-card firmware.  The PC-card and
MiniPCI models must have a difference.  This is looking like a bug
introduced in the later kernels.



>
> --sdk
>
>
>
>

--
		Matthew Saltzman

Clemson University Math Sciences
mjs AT clemson DOT edu
http://www.math.clemson.edu/~mjs