Messages in this thread | | | Date | Wed, 20 Oct 2010 09:31:56 +0200 | From | Andi Kleen <> | Subject | Re: [PATCH 2/2]x86: spread tlb flush vector between nodes |
| |
> Maybe we should have a per_node memory infrastructure, so that we can > lower memory needs of currently per_cpu objects.
I have been looking at that :- for a lot of things per core data makes sense too.
Really a lot of the per CPU scaling we have today should be per core or per node to avoid explosion.
But for this particular case it doesn't help because you still need a mapping for each CPU.
-Andi -- ak@linux.intel.com -- Speaking for myself only.
| |