> On Tue, 2 Oct 2007, Yasunori Goto wrote:
quoted text
> 
> > Do you mean that just nr_slabs should be checked like followings?
> > I'm not sure this is enough.
> > 
> >     :
> > if (s->node[nid]) {
> > 	n = get_node(s, nid);
> > 	if (!atomic_read(&n->nr_slabs)) {
> > 		s->node[nid] = NULL;
> > 		kmem_cache_free(kmalloc_caches, n);
> > 	}
> > }
> >     :
> >     :
> 
> That would work. But it would be better to shrink the cache first. The 
> first 2 slabs on a node may be empty and the shrinking will remove those. 
> If you do not shrink then the code may falsely assume that there are 
> objects on the node.

I'm sorry, but I don't think I understand what you mean... :-(
Could you explain more? 

Which slabs should be shrinked? kmem_cache_node and kmem_cache_cpu?

I think kmem_cache_cpu should be disabled by cpu hotplug,
not memory/node hotplug. Basically, cpu should be offlined before
memory offline on the node.

Sorry, I'm confusing now...

-- 
Yasunori Goto 


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