Toru Nishimura wrote:

> >> Will the following hack make any performance difference with yours?
> >
> > Only tested in a R4400, but if anything just slightly worse.
> 
> Um, I expected it might be.  The real issue is the design and usage of
> pmap_protect() in pmap.c It's used for dual purposes and called in
> inefficient way at runtime.

Early tests show no apprecable difference on a 5000/240 either.  It
hasn't finished yet, but the fork/exec/sh, mmap and page fault tests
have finished.

Simon.