Re: [MODSLAB 0/7] A modular slab allocator V1

[Christoph Lameter]

On Wed, 16 Aug 2006, Andi Kleen wrote:

> What other ones do we have?

vmalloc and the page allocators you derive from others using the 
framework.

> > 1. shrink_slab takes a function to move object. Using that
> >    function slabs can be defragmented to ease slab reclaim.
> 
> Does that help with the inefficient dcache/icache pruning? 

It will fix that if the dcache/icache would provide a move object
function.

> 
> > - No support for pagese
> 
> What does that mean?

That was just clutter. Sorry.

> 
> > Performance tests with AIM7 on an 8p Itanium machine (4 NUMA nodes)
> > (Memory spreading active which means that we do not take advantage of NUMA locality
> > in favor of load balancing)
> 
> Hmm, i'm not sure how allocator intensive AIM7 is. I guess networking
> would be a good test because it is very sensitive to allocator performance.
> Perhaps also check with the routing people on netdev -- they seem to be able
> to stress the allocator very much.

Yeah, well I think the implementations could be much more sophisticated 
and one should do better tests but I am not sure how much time I can spend 
on them.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/