>But your active page figures for the test under NetBSD 1.3 are rather high,
>don't you think?  My figures under NetBSD 1.2 -- with either behavior -- start
>several pages lower than yours with UVM.
>I wouldn't want to think you're leveraging this to over-promote UVM.... ;)

the number of active pages is totally dependent on the number of
programs you've run and how much data is cached in the VM system.
it isn't useful to compare the value (e.g. 4000 vs 600) ... what you 
want to compare is the difference between runs [i.e. "we gained 50
active pages while this program ran"].

just for your reference, in my previous message the number of active
pages for the NetBSD 1.3+oldvm system was greater than the UVM system
because the oldvm system was running X, netscape, and other stuff at
the time i took the measurements (while the UVM system i measured
was freshly booted and has a serial console).  note that doesn't 
really say anything about UVM vs. oldvm, so please don't read it that
way.

chuck