Re: memcg: fix fatal livelock in kswapd

From: Paul Menage
Date: Mon May 02 2011 - 18:54:01 EST


[ Adding the memcg maintainers ]

On Mon, May 2, 2011 at 1:07 PM, James Bottomley
<James.Bottomley@xxxxxxxxxxxxxxxxxxxxx> wrote:
> The fatal livelock in kswapd, reported in this thread:
>
> http://marc.info/?t=130392066000001
>
> Is mitigateable if we prevent the cgroups code being so aggressive in
> its zone shrinking (by reducing it's default shrink from 0 [everything]
> to DEF_PRIORITY [some things]).  This will have an obvious knock on
> effect to cgroup accounting, but it's better than hanging systems.
>
> Signed-off-by: James Bottomley <James.Bottomley@xxxxxxx>
>
> ---
>
> From 74b62fc417f07e1411d98181631e4e097c8e3e68 Mon Sep 17 00:00:00 2001
> From: James Bottomley <James.Bottomley@xxxxxxxxxxxxxxxxxxxxx>
> Date: Mon, 2 May 2011 14:56:29 -0500
> Subject: [PATCH] vmscan: move containers scan back to default priority
>
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index f6b435c..46cde92 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -2173,8 +2173,12 @@ unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem,
>         * if we don't reclaim here, the shrink_zone from balance_pgdat
>         * will pick up pages from other mem cgroup's as well. We hack
>         * the priority and make it zero.
> +        *
> +        * FIXME: jejb: zero here was causing a livelock in the
> +        * shrinker so changed to DEF_PRIORITY to fix this. Now need to
> +        * sort out cgroup accounting.
>         */
> -       shrink_zone(0, zone, &sc);
> +       shrink_zone(DEF_PRIORITY, zone, &sc);
>
>        trace_mm_vmscan_memcg_softlimit_reclaim_end(sc.nr_reclaimed);
>
>
>
>
--
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/