hi,

this looks fine too.

you might add an assertion in pve_getpmap() to verify that pm_obj
is the first member of struct pmap instead of just commenting it.

the tmppage/tmppve thing is goofy, but it seems worth it to reclaim
the memory.

-Chuck


On Sun, Nov 02, 2003 at 05:20:20PM +0900, YAMAMOTO Takashi wrote:
> hi,
> 
> attached patch shrinks pv_entry by eliminating pv_pmap member.
> while this increases cost to get pmap pointer a little,
> i think size of pv_entry is more important.
> 
> any comments?
> 
> YAMAMOTO Takashi