> In voimax.lists.linux-kernel, Rik van Riel wrote:
> >That's not the point, as long as page-tables can't be swapped
> >out, they will take up memory. Users sometimes have a memory
>
> If page-table swapping is implemented, what will happen to mlock():ed pages?
> Will the appropriate page tables be locked, too? I certainly hope so.. :)
Hmmm, quite a lot of loopholes with page-table swapping :(
Oh, and I've just calculated that a 4G mmap on /dev/zero will
take about 4megs of unswappable memory... Is pt swapping
worth it or not???
Other area's of vm-improvement are more important anyways...
(more graceful out-of-VM handling, better swapin-code (M)Ingo
had something?)
Rik.
----------
Send Linux memory-management wishes to me: I'm currently looking
for something to hack...