Re: 2.6.14-rc1-git-now still dying in mm/slab - this time line 1849

[Christoph Lameter]

On Mon, 19 Sep 2005, Petr Vandrovec wrote:

> I've thought that this is problem, but as far as I can tell while this is
> problem it does not happen here.  Just free_block() finds that pointer it
> got from caller belongs to the slab that belongs to the CPU#1/node#1
> while caller obtained lock on CPU#0/node#0 structures.  Which suggests
> that drain_array_locked() was issued with node #0 while array_cache->entry
> it got contains blocks which belong to node #1.  Which I cannot explain.

That can happen if node 0 runs out of memory and the page_allocator falls 
back to take memory from node 1 for node 0 requests.

Maybe we have a problem here.

-
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/