Re: [PATCH] slub: correctly bootstrap boot caches

From: Glauber Costa
Date: Fri Feb 22 2013 - 10:44:57 EST


On 02/22/2013 07:39 PM, Christoph Lameter wrote:
> On Fri, 22 Feb 2013, Glauber Costa wrote:
>
>> As I've mentioned in the description, the real bug is from partial slabs
>> being temporarily in the cpu_slab during a recent allocation and
>> therefore unreachable through the partial list.
>
> The bootstrap code does not use cpu slabs but goes directly to the slab
> pages. See early_kmem_cache_node_alloc.
>

That differs from what I am seeing here.
I can trace an early __slab_alloc allocation from
the kmem_cache_node cache, very likely coming from the kmem_cache boot
cache creation. It takes the page out of the partial list and moves it
to the cpu_slab. After that, that particular page becomes unreachable
for bootstrap.

At this point, we are already slab_state == PARTIAL, while
init_kmem_cache_nodes will only differentiate against slab_state == DOWN.



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