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

["Pekka Enberg" <penberg@cs.helsinki.fi>]

Hi Christoph,

On Jan 10, 2008 2:02 AM, Christoph Lameter <clameter@sgi.com> wrote:
> New patch that also checks in alternate_node_alloc if the node has normal
> memory because we cannot call ____cache_alloc_node with an invalid node.

[snip]

> @@ -3439,8 +3442,14 @@ __do_cache_alloc(struct kmem_cache *cach
>          * We may just have run out of memory on the local node.
>          * ____cache_alloc_node() knows how to locate memory on other nodes
>          */
> -       if (!objp)
> -               objp = ____cache_alloc_node(cache, flags, numa_node_id());
> +       if (!objp) {
> +               int node_id = numa_node_id();
> +               if (likely(cache->nodelists[node_id])) /* fast path */
> +                       objp = ____cache_alloc_node(cache, flags, node_id);
> +               else /* this function can do good fallback */
> +                       objp = __cache_alloc_node(cache, flags, node_id,
> +                                       __builtin_return_address(0));
> +       }

But __cache_alloc_node() will call fallback_alloc() that does
cache_grow() for the node that doesn't have N_NORMAL_MEMORY, no?

Shouldn't we just revert 04231b3002ac53f8a64a7bd142fde3fa4b6808c6 for
2.6.24 as this is a clear regression from 2.6.23?

                        Pekka

--
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>