[PATCH v3 6/8] workingset: memcg: sleep when flushing stats in workingset_refault()

[Yosry Ahmed]

In workingset_refault(), we call
mem_cgroup_flush_stats_atomic_ratelimited() to read accurate stats
within an RCU read section and with sleeping disallowed. Move the
call above the RCU read section to make it non-atomic.

Flushing is an expensive operation that scales with the number of cpus
and the number of cgroups in the system, so avoid doing it atomically
where possible.

Since workingset_refault() is the only caller of
mem_cgroup_flush_stats_atomic_ratelimited(), just make it non-atomic,
and rename it to mem_cgroup_flush_stats_ratelimited().

Signed-off-by: Yosry Ahmed <yosryahmed@xxxxxxxxxx>
Acked-by: Shakeel Butt <shakeelb@xxxxxxxxxx>
Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxxx>
---
 include/linux/memcontrol.h | 4 ++--
 mm/memcontrol.c            | 4 ++--
 mm/workingset.c            | 5 +++--
 3 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index b424ba3ebd09..a4bc3910a2eb 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -1038,7 +1038,7 @@ static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
 
 void mem_cgroup_flush_stats(void);
 void mem_cgroup_flush_stats_atomic(void);
-void mem_cgroup_flush_stats_atomic_ratelimited(void);
+void mem_cgroup_flush_stats_ratelimited(void);
 
 void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
 			      int val);
@@ -1540,7 +1540,7 @@ static inline void mem_cgroup_flush_stats_atomic(void)
 {
 }
 
-static inline void mem_cgroup_flush_stats_atomic_ratelimited(void)
+static inline void mem_cgroup_flush_stats_ratelimited(void)
 {
 }
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index a2ce3aa10d94..361c0bbf7283 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -673,10 +673,10 @@ void mem_cgroup_flush_stats_atomic(void)
 		do_flush_stats(true);
 }
 
-void mem_cgroup_flush_stats_atomic_ratelimited(void)
+void mem_cgroup_flush_stats_ratelimited(void)
 {
 	if (time_after64(jiffies_64, READ_ONCE(flush_next_time)))
-		mem_cgroup_flush_stats_atomic();
+		mem_cgroup_flush_stats();
 }
 
 static void flush_memcg_stats_dwork(struct work_struct *w)
diff --git a/mm/workingset.c b/mm/workingset.c
index dab0c362b9e3..3025beee9b34 100644
--- a/mm/workingset.c
+++ b/mm/workingset.c
@@ -406,6 +406,9 @@ void workingset_refault(struct folio *folio, void *shadow)
 	unpack_shadow(shadow, &memcgid, &pgdat, &eviction, &workingset);
 	eviction <<= bucket_order;
 
+	/* Flush stats (and potentially sleep) before holding RCU read lock */
+	mem_cgroup_flush_stats_ratelimited();
+
 	rcu_read_lock();
 	/*
 	 * Look up the memcg associated with the stored ID. It might
@@ -461,8 +464,6 @@ void workingset_refault(struct folio *folio, void *shadow)
 	lruvec = mem_cgroup_lruvec(memcg, pgdat);
 
 	mod_lruvec_state(lruvec, WORKINGSET_REFAULT_BASE + file, nr);
-
-	mem_cgroup_flush_stats_atomic_ratelimited();
 	/*
 	 * Compare the distance to the existing workingset size. We
 	 * don't activate pages that couldn't stay resident even if
-- 
2.40.0.348.gf938b09366-goog