thorpej@zembu.com said:
> You can optimize your traversal of the tables in pmap_remove(), and if
> you're clever (like the i386 pmap is, esp. in the MP branch kernel),
> you can defer all the TLB operations until the end (the i386 just does
> a full non-global TLB flush if there are > 16 TLB invalidations
> pending).


If we could defer the unmappings until after the pmap has been changed to 
a different one (or switch to a different pmap before starting the 
unmapping), we could eliminate the cache flush on process death entirely.

Not sure how we might achieve this though.

R.