Hi,

Probably the thing that could win you most is accelerating cache
flushing. AFAIR the RiscPC does not have a "fast area," so one possible
solution was that it reads a hunk of kernel text to do the job (amb? I
really don't remember any more...). With the kernel being loaded low in
memory, if this is the case then you've pretty much lost.

AFAIR the cache sync routines, in the cpufunc_asm.S file, referenced
some simple address parameters for the addresses of the clean areas. You
possibly do a the quick hack and pmap_enter a mapping from the fast
memory for a few pages and then point the routine to this instead. This
might be a miimal-effort way of winning something.

	Neil