Re: [PATCH] memcg: remove mem_cgroup_reclaimable check from soft reclaim

From: Michal Hocko
Date: Tue Oct 21 2014 - 11:37:48 EST


On Tue 21-10-14 17:15:50, Vladimir Davydov wrote:
> mem_cgroup_reclaimable() checks whether a cgroup has reclaimable pages
> on *any* NUMA node. However, the only place where it's called is
> mem_cgroup_soft_reclaim(), which tries to reclaim memory from a
> *specific* zone. So the way how it's used is incorrect - it will return
> true even if the cgroup doesn't have pages on the zone we're scanning.
>
> I think we can get rid of this check completely, because
> mem_cgroup_shrink_node_zone(), which is called by
> mem_cgroup_soft_reclaim() if mem_cgroup_reclaimable() returns true, is
> equivalent to shrink_lruvec(), which exits almost immediately if the
> lruvec passed to it is empty. So there's no need to optimize anything
> here. Besides, we don't have such a check in the general scan path
> (shrink_zone) either.
>
> Signed-off-by: Vladimir Davydov <vdavydov@xxxxxxxxxxxxx>

yeah, let's ditch it. It doesn't make any sense without checking the
target zone and even then it is dubious bevause get_scan_count will give
us 0 target on an empty lru list.

Acked-by: Michal Hocko <mhocko@xxxxxxx>

Thanks

> ---
> mm/memcontrol.c | 43 -------------------------------------------
> 1 file changed, 43 deletions(-)
>
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 53393e27ff03..833b6a696aab 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -1799,52 +1799,11 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *memcg)
> memcg->last_scanned_node = node;
> return node;
> }
> -
> -/*
> - * Check all nodes whether it contains reclaimable pages or not.
> - * For quick scan, we make use of scan_nodes. This will allow us to skip
> - * unused nodes. But scan_nodes is lazily updated and may not cotain
> - * enough new information. We need to do double check.
> - */
> -static bool mem_cgroup_reclaimable(struct mem_cgroup *memcg, bool noswap)
> -{
> - int nid;
> -
> - /*
> - * quick check...making use of scan_node.
> - * We can skip unused nodes.
> - */
> - if (!nodes_empty(memcg->scan_nodes)) {
> - for (nid = first_node(memcg->scan_nodes);
> - nid < MAX_NUMNODES;
> - nid = next_node(nid, memcg->scan_nodes)) {
> -
> - if (test_mem_cgroup_node_reclaimable(memcg, nid, noswap))
> - return true;
> - }
> - }
> - /*
> - * Check rest of nodes.
> - */
> - for_each_node_state(nid, N_MEMORY) {
> - if (node_isset(nid, memcg->scan_nodes))
> - continue;
> - if (test_mem_cgroup_node_reclaimable(memcg, nid, noswap))
> - return true;
> - }
> - return false;
> -}
> -
> #else
> int mem_cgroup_select_victim_node(struct mem_cgroup *memcg)
> {
> return 0;
> }
> -
> -static bool mem_cgroup_reclaimable(struct mem_cgroup *memcg, bool noswap)
> -{
> - return test_mem_cgroup_node_reclaimable(memcg, 0, noswap);
> -}
> #endif
>
> static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg,
> @@ -1888,8 +1847,6 @@ static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg,
> }
> continue;
> }
> - if (!mem_cgroup_reclaimable(victim, false))
> - continue;
> total += mem_cgroup_shrink_node_zone(victim, gfp_mask, false,
> zone, &nr_scanned);
> *total_scanned += nr_scanned;
> --
> 1.7.10.4
>

--
Michal Hocko
SUSE Labs
--
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/