Re: [PATCH 1/2] slub: fix leakage

[Willy Tarreau <w@1wt.eu>]

On Sat, Nov 03, 2007 at 05:10:54PM +0000, Hugh Dickins wrote:
> Slub has been quite leaky under load.  Taking mm_struct as an example, in
> a loop of swapping kernel builds, after the first iteration slabinfo shows:
> Name        Objects Objsize    Space Slabs/Part/Cpu  O/S O %Fr %Ef Flg
> mm_struct        55     840    73.7K         18/7/4    4 0  38  62 A
> but Objects and Partials steadily creep up - after the 340th iteration:
> mm_struct       110     840   188.4K        46/36/4    4 0  78  49 A
> 
> The culprit turns out to be __slab_alloc(), where it copes with the race
> that another task has assigned the cpu slab while we were allocating one.
> Don't rush off to load_freelist there: that assumes c->freelist is empty,
> and will lose all of its free slots when c->page->freelist is not empty.
> Instead just do a local allocation from c->freelist when it has one.
> 
> Which fixes the leakage: Objects and Partials then remain stable.
> 
> Signed-off-by: Hugh Dickins <hugh@veritas.com>
> ---
> I recommend this for 2.6.24-rc2 and 2.6.23-stable.

Hugh, you should also have CC'd linux-stable for this, otherwise there's
a risk that the patch remains unnoticed by the stable team (CC'd).

>  mm/slub.c |    5 +++++
>  1 file changed, 5 insertions(+)
> 
> --- 2.6.24-rc1/mm/slub.c	2007-10-24 07:16:04.000000000 +0100
> +++ linux/mm/slub.c	2007-11-03 13:22:31.000000000 +0000
> @@ -1525,6 +1525,11 @@ new_slab:
>  				 * want the current one since its cache hot
>  				 */
>  				discard_slab(s, new);
> +				if (c->freelist) {
> +					object = c->freelist;
> +					c->freelist = object[c->offset];
> +					return object;
> +				}
>  				slab_lock(c->page);
>  				goto load_freelist;
>  			}

Regards,
Willy