Re: [BUG] at mm/slab.c:3320

["Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>]

On Mon, Jan 07, 2008 at 09:38:51PM -0800, Christoph Lameter wrote:
> On Tue, 8 Jan 2008, KAMEZAWA Hiroyuki wrote:
> 
> > In usual alloc_pages() allocator, this is done by zonelist fallback.
> 
> Hmmm... __cache_alloc_node does:
> 
>     if (unlikely(!cachep->nodelists[nodeid])) {
>                 /* Node not bootstrapped yet */
>                 ptr = fallback_alloc(cachep, flags);
>                 goto out;
>         }
> 
> So kmalloc_node does the correct fallback.
> 
> Kmalloc does not fall back but relies on numa_node_id() referring to a 
> node that has ZONE_NORMAL memory. Sigh.
> 
> cache_alloc_refill:
> 
>         node = numa_node_id();
> 
>         check_irq_off();
>         ac = cpu_cache_get(cachep);
> retry:
>         batchcount = ac->batchcount;
>         if (!ac->touched && batchcount > BATCHREFILL_LIMIT) {
>                 /*
>                  * If there was little recent activity on this cache, then
>                  * perform only a partial refill.  Otherwise we could generate
>                  * refill bouncing.
>                  */
>                 batchcount = BATCHREFILL_LIMIT;
>         }
>         l3 = cachep->nodelists[node];
> 
> 	BUG_ON(ac->avail > 0 || !l3);
> 	^^^^ triggers
> 
> 
> > complicated ?
> 
> Hmm.. We could check for l3 == NULL and fail in that case? The 
> ___cache_alloc would fail and __do_cache_alloc would call 
> ___cache_alloc_node whicvh would provide the correct fallback.
> 
> Doesd this fix it?

Will test and get back to you. Waiting for the machine to be free.

-aneesh

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>