Re: vm lock contention reduction

[Rik van Riel <riel@conectiva.com.br>]

On Thu, 4 Jul 2002, Andrew Morton wrote:

> If the machine is instead full of anon pages then everything is still
> crap because the page reclaim code is scanning zillions of pages and not
> doing much useful with them.

This is something that can be fixed with rmap, because the
kernel _will_ be able to do something useful with the anon
pages.

Now we just need to get Arjan to tune the O(1) page launder
thing he was looking at ;)

Rik
-- 
Bravely reimplemented by the knights who say "NIH".

http://www.surriel.com/		http://distro.conectiva.com/

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/