Re: Minutes from LSE Call March 7

[William Lee Irwin III <wli@holomorphy.com>]

On Fri, Mar 07, 2003 at 10:34:29AM -0800, Hanna Linder wrote:
> Bill is using some simplified heuristics to search
> for pages to fault in. Turns out those heuristics 
> suck so he needs to go in and do a different set.
> The ones originally done in Hugh's patch did something
> in the order of scanning acros an entire vma looking
> for pte's pointing to a particular page. It didnt
> have any alignment restrictions. Bill does have
> alignment restrictions and Hugh's solution would
> break down pretty quickly (kernel compiles swapping).

The bit about Hugh's heuristics is backward; the heuristics he used
for 2.4.x were very effective. It's my homegrown heuristics that are
breaking down very quickly wrt. performance and fragmentation.


-- wli