| Message ID | 20181009184732.762-2-hannes@cmpxchg.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | mm: workingset & shrinker fixes | expand |
On Tue, 2018-10-09 at 14:47 -0400, Johannes Weiner wrote: > The shadow shrinker is invoked per NUMA node, but the shadow limit > enforced for cgroups is based on the page counter, which isn't NUMA > aware. Instead of shrinking shadow pages to desired_size, we end up > with desired_size * nr_online_nodes. > > Switch to NUMA-aware lru and slab counters to approximate cgroup > size. > > Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@surriel.com>
diff --git a/mm/workingset.c b/mm/workingset.c index 1d111913929d..e5c70bc94077 100644 --- a/mm/workingset.c +++ b/mm/workingset.c @@ -418,9 +418,15 @@ static unsigned long count_shadow_nodes(struct shrinker *shrinker, * PAGE_SIZE / xa_nodes / node_entries * 8 / PAGE_SIZE */ #ifdef CONFIG_MEMCG - if (sc->memcg) - pages = page_counter_read(&sc->memcg->memory); - else + if (sc->memcg) { + struct lruvec *lruvec; + + pages = mem_cgroup_node_nr_lru_pages(sc->memcg, sc->nid, + LRU_ALL); + lruvec = mem_cgroup_lruvec(NODE_DATA(sc->nid), sc->memcg); + pages += lruvec_page_state(lruvec, NR_SLAB_RECLAIMABLE); + pages += lruvec_page_state(lruvec, NR_SLAB_UNRECLAIMABLE); + } else #endif pages = node_present_pages(sc->nid);
The shadow shrinker is invoked per NUMA node, but the shadow limit enforced for cgroups is based on the page counter, which isn't NUMA aware. Instead of shrinking shadow pages to desired_size, we end up with desired_size * nr_online_nodes. Switch to NUMA-aware lru and slab counters to approximate cgroup size. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> --- mm/workingset.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)