>Jonathan Stone writes:
>[...]
> > The point was to look at sharing that same vmspace/pmap (and avoid the
> > pmap_activate() costs) for processes like nfs_iod(), yet give them
> > back `a vmspace of their own' if/when they return to userland.
> > 

[...]

>Yes, that provides a noticable performance improvement when
>one has processes stuck in the kernel for a while (nfsiod).
>The sun3 and sun3x pmap modules use the "empty context" trick,
>which allows delaying the actual allocation of an MMU context
>until a process touches a user-space address.  Just FYI.

yup, thats what i was tryng to say you told me before;).  On the sun
mmu you do a lazy-evaluation of the MMU context-switch: you wait til
you take a userspace fault before allocating pmegs.  But thats exactly
what you _cannot_ do on the arm32, because it has only 1 mmu context
and a virtually-indexed cache (I-cache, at least).

there, it'd win to have a hook saying
   `this process will be in kernel-only context until it says otherwise.'
exactly so that you _can_ do a lazy MMU context-switch.

hope thats clearer.