Re: Purpose of the mm/slab.c changes

[Andrea Arcangeli]

On Sun, Sep 09, 2001 at 10:01:32AM -0700, Linus Torvalds wrote:
> (Doing per-CPU LIFO queues for the actual page allocator would potentially
> make page alloc/de-alloc much faster due to lower locking requirements
> too. So you might have a double performance win if anybody wants to try
> this out).

I recall I seen this one implemented during some auditing recently, I
think it's in the tux patch but I may remeber wrong.

Andrea