Messages in this thread | | | Date | Tue, 11 Mar 2008 23:21:58 -0700 (PDT) | From | Christoph Lameter <> | Subject | Re: [rfc][patch 2/3] slab: introduce SMP alignment |
| |
On Tue, 11 Mar 2008, Nick Piggin wrote:
> On Thu, Mar 06, 2008 at 09:26:24PM -0800, Christoph Lameter wrote: > > Internode alignment is 4k. If you need an object aligned like that then it > > may be better to go to the page allocator. > > I never got a reply about this. Again: how can the modern slab allocator > avoid the need for internode alignment in order to prevent cacheline > pingpong on vsmp? Because I'm going to resubmit the SMP_ALIGN patch.
You quoted my answer. Use the page allocator to allocate data that has an alignment requirement on a 4k boundary.
> Also, the last I heard about the HWCACHE_ALIGN from you is that it didn't > make sense... I see it's merged now, which I appreciate, but I think it is > really important that we're on the same page here, so let me know if it > still doesn't make sense.
I still think that the semantics are weird because it only works sometimes and then performs an alignment within a cacheline that improves the situation somewhat in some cases but does not give the user what he expected.
| |