>Jun 19 15:26:53 garp /netbsd: nogo1, Data abort: 'Translation fault
>(section)' status = 005 address = 00400028 PC = f00cf49c

Ok this is because the binary you are running is an RiscBSD alpha2 binary
NetBSD/arm32 and RiscBSD 1.1 or later to not use alpha2 binaries. The kernels
that I have been putting online have an kern/exec_aout.c patch that allows
alpha2 binaries to work. Any alpha2 binaries should be recompiled.
>From NetBSD/RiscBSD 1.2 I will not longer be building this patch into any
release kernels.

The difference is the address of the text area.
In the alpha1,2 releases of RiscBSD this was 0x00400000 which was purely for
the convienience of being able to share the L2 page table that mapped the
system page (0x00000000) between every process.

This was changed prior to merging into the NetBSD source tree and from the beta
release onwards the text area starts at 0x1000. Now a separate page table is
faulted in per process and the system page is mapped in during cpu_fork().

>   Also xsysstats can't read context switches and ehter packets (It just
>registers them as 0)
>
>  I have applied no patches to the netbsd-current source.  Are they still
>necessary?

Is that xsysstats from the xutils or did you build it. The one in the xutils
set needs upgrade as it has not been recompiled since the changes to the
networking code so this could effect the network stuff. Not sure on the context
switches ...

The only patch that you really need to apply is the one to sys/proc.h to
declare p_nice as a signed char rather than a char.
This change has yet to be sorted out in the MI code.

Cheers,
				Mark

-- 
Mark Brinicombe				amb@physig.ph.kcl.ac.uk
Research Associate			http://www.ph.kcl.ac.uk/~amb/
Department of Physics			tel: 0171 873 2894
King's College London			fax: 0171 873 2716