From: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx> Global reclaim aims to reduce the amount of memory used on a given node or set of nodes. Migrating pages to another node serves this purpose. memcg reclaim is different. Its goal is to reduce the total memory consumption of the entire memcg, across all nodes. Migration does not assist memcg reclaim because it just moves page contents between nodes rather than actually reducing memory consumption. Signed-off-by: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx> Signed-off-by: "Huang, Ying" <ying.huang@xxxxxxxxx> Suggested-by: Yang Shi <yang.shi@xxxxxxxxxxxxxxxxx> Reviewed-by: Yang Shi <shy828301@xxxxxxxxx> Cc: Michal Hocko <mhocko@xxxxxxxx> Cc: Wei Xu <weixugc@xxxxxxxxxx> Cc: osalvador <osalvador@xxxxxxx> Cc: Zi Yan <ziy@xxxxxxxxxx> Cc: David Rientjes <rientjes@xxxxxxxxxx> Cc: Dan Williams <dan.j.williams@xxxxxxxxx> Cc: David Hildenbrand <david@xxxxxxxxxx> -- changes from 20210618: * Avoid to scan anon lists for demotion for cgroup reclaim. --- mm/vmscan.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index fce43c7970d7..e3cc356369fd 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -540,8 +540,9 @@ static inline bool can_reclaim_anon_pages(struct mem_cgroup *memcg, * The page can not be swapped. * * Can it be reclaimed from this node via demotion? + * But it's pointless to do demotion in memcg reclaim. */ - if ((!sc || !sc->no_demotion) && + if ((!sc || (!sc->no_demotion && !cgroup_reclaim(sc))) && next_demotion_node(node_id) != NUMA_NO_NODE) return true; @@ -1276,6 +1277,9 @@ static bool migrate_demote_page_ok(struct page *page, if (sc->no_demotion) return false; + /* It is pointless to do demotion in memcg reclaim */ + if (cgroup_reclaim(sc)) + return false; next_nid = next_demotion_node(page_to_nid(page)); if (next_nid == NUMA_NO_NODE) return false; @@ -2759,7 +2763,7 @@ static bool anon_can_be_aged(struct pglist_data *pgdat, return true; /* Also valuable if anon pages can be demoted: */ - if (!sc->no_demotion && + if (!sc->no_demotion && !cgroup_reclaim(sc) && next_demotion_node(pgdat->node_id) != NUMA_NO_NODE) return true; -- 2.30.2
