>I've recently undertaken an effort to clean up the `ae' Ethernet driver,
>splitting the NuBus attachment code out from the driver proper, and
>hopefully making it a little more reliable and easier to maintain.  In the
>process, however, I'm changing the code to make an assumption:  the
>Ethernet MAC address for the card is stored at the location pointed to by
>resource ID 0x80 in the card's functional resource, also ID 0x80.  The
>idea for this came from Ken Nakata's Kinetics EtherPort SE/30 support, and
>it looks like the two cards that I have both have the same resources
>defined.  Having said that, though, -don't blame Ken-.  If it breaks, it's
>entirely my fault.
>
>If it's any consolation, you can compile a kernel with
>`options AE_OLD_GET_ENADDR' to get back the old behavior.  You need only
>to recompile if_ae_nubus.o in your kernel compile directory for this
>change to take effect.
>
>I'd be interested to hear of any success or failure reports.  (Make sure
>you Cc: me on any replies to the list; it's difficult to wade through all
>that traffic on port-mac68k!)
>
>Thanks,
>
>--scott

In case anyone else is interested, the same asante ether board that I have
in my Mac IIci which works fine there, does not work when placed in a nubus
adapter inside my quadra 660av.  Granted I am not tracking up to the minute
-current, nor have I been able to keep track of all the zillions of kernels
and patches floating around out there, but with the kernel I built, it
seems to hang in the probe code some where in config_attach.  I am guessing
that this is probably an interupt problem, but I haven't and probably won't
have time to look at it till after the semester.  Just thought I would
bring it up in some one is poking around in there.

I can traces if anyone is interested.

---------
Christopher R. Bowman
crb@eng.umd.edu
My home page