mm: workingset: per-cgroup cache thrash detection

Cache thrash detection (see a528910e12ec "mm: thrash detection-based
file cache sizing" for details) currently only works on the system
level, not inside cgroups.  Worse, as the refaults are compared to the
global number of active cache, cgroups might wrongfully get all their
refaults activated when their pages are hotter than those of others.

Move the refault machinery from the zone to the lruvec, and then tag
eviction entries with the memcg ID.  This makes the thrash detection
work correctly inside cgroups.

[sergey.senozhatsky@gmail.com: do not return from workingset_activation() with locked rcu and page]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

2 files changed, 56 insertions, 13 deletions

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 8502fd4144eb..09b449849369 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -89,6 +89,10 @@ enum mem_cgroup_events_target {
 };
 
 #ifdef CONFIG_MEMCG
+
+#define MEM_CGROUP_ID_SHIFT	16
+#define MEM_CGROUP_ID_MAX	USHRT_MAX
+
 struct mem_cgroup_stat_cpu {
 	long count[MEMCG_NR_STAT];
 	unsigned long events[MEMCG_NR_EVENTS];
@@ -265,6 +269,11 @@ struct mem_cgroup {
 
 extern struct mem_cgroup *root_mem_cgroup;
 
+static inline bool mem_cgroup_disabled(void)
+{
+	return !cgroup_subsys_enabled(memory_cgrp_subsys);
+}
+
 /**
  * mem_cgroup_events - count memory events against a cgroup
  * @memcg: the memory cgroup
@@ -312,6 +321,28 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *,
 				   struct mem_cgroup_reclaim_cookie *);
 void mem_cgroup_iter_break(struct mem_cgroup *, struct mem_cgroup *);
 
+static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
+{
+	if (mem_cgroup_disabled())
+		return 0;
+
+	return memcg->css.id;
+}
+
+/**
+ * mem_cgroup_from_id - look up a memcg from an id
+ * @id: the id to look up
+ *
+ * Caller must hold rcu_read_lock() and use css_tryget() as necessary.
+ */
+static inline struct mem_cgroup *mem_cgroup_from_id(unsigned short id)
+{
+	struct cgroup_subsys_state *css;
+
+	css = css_from_id(id, &memory_cgrp_subsys);
+	return mem_cgroup_from_css(css);
+}
+
 /**
  * parent_mem_cgroup - find the accounting parent of a memcg
  * @memcg: memcg whose parent to find
@@ -353,11 +384,6 @@ static inline bool mm_match_cgroup(struct mm_struct *mm,
 struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page);
 ino_t page_cgroup_ino(struct page *page);
 
-static inline bool mem_cgroup_disabled(void)
-{
-	return !cgroup_subsys_enabled(memory_cgrp_subsys);
-}
-
 static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
 {
 	if (mem_cgroup_disabled())
@@ -502,8 +528,17 @@ void mem_cgroup_split_huge_fixup(struct page *head);
 #endif
 
 #else /* CONFIG_MEMCG */
+
+#define MEM_CGROUP_ID_SHIFT	0
+#define MEM_CGROUP_ID_MAX	0
+
 struct mem_cgroup;
 
+static inline bool mem_cgroup_disabled(void)
+{
+	return true;
+}
+
 static inline void mem_cgroup_events(struct mem_cgroup *memcg,
 				     enum mem_cgroup_events_index idx,
 				     unsigned int nr)
@@ -586,9 +621,16 @@ static inline void mem_cgroup_iter_break(struct mem_cgroup *root,
 {
 }
 
-static inline bool mem_cgroup_disabled(void)
+static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
 {
-	return true;
+	return 0;
+}
+
+static inline struct mem_cgroup *mem_cgroup_from_id(unsigned short id)
+{
+	WARN_ON_ONCE(id);
+	/* XXX: This should always return root_mem_cgroup */
+	return NULL;
 }
 
 static inline bool mem_cgroup_online(struct mem_cgroup *memcg)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 9fc23ab550a7..03cbdd906f55 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -212,10 +212,12 @@ struct zone_reclaim_stat {
 };
 
 struct lruvec {
-	struct list_head lists[NR_LRU_LISTS];
-	struct zone_reclaim_stat reclaim_stat;
+	struct list_head		lists[NR_LRU_LISTS];
+	struct zone_reclaim_stat	reclaim_stat;
+	/* Evictions & activations on the inactive file list */
+	atomic_long_t			inactive_age;
 #ifdef CONFIG_MEMCG
-	struct zone *zone;
+	struct zone			*zone;
 #endif
 };
 
@@ -490,9 +492,6 @@ struct zone {
 	spinlock_t		lru_lock;
 	struct lruvec		lruvec;
 
-	/* Evictions & activations on the inactive file list */
-	atomic_long_t		inactive_age;
-
 	/*
 	 * When free pages are below this point, additional steps are taken
 	 * when reading the number of free pages to avoid per-cpu counter
@@ -761,6 +760,8 @@ static inline struct zone *lruvec_zone(struct lruvec *lruvec)
 #endif
 }
 
+extern unsigned long lruvec_lru_size(struct lruvec *lruvec, enum lru_list lru);
+
 #ifdef CONFIG_HAVE_MEMORY_PRESENT
 void memory_present(int nid, unsigned long start, unsigned long end);
 #else