mm: vmscan: fix IO/refault regression in cache workingset transition

Since commit 59dc76b0d4df ("mm: vmscan: reduce size of inactive file
list") we noticed bigger IO spikes during changes in cache access
patterns.

The patch in question shrunk the inactive list size to leave more room
for the current workingset in the presence of streaming IO.  However,
workingset transitions that previously happened on the inactive list are
now pushed out of memory and incur more refaults to complete.

This patch disables active list protection when refaults are being
observed.  This accelerates workingset transitions, and allows more of
the new set to establish itself from memory, without eating into the
ability to protect the established workingset during stable periods.

The workloads that were measurably affected for us were hit pretty bad
by it, with refault/majfault rates doubling and tripling during cache
transitions, and the machines sustaining half-hour periods of 100% IO
utilization, where they'd previously have sub-minute peaks at 60-90%.

Stateful services that handle user data tend to be more conservative
with kernel upgrades.  As a result we hit most page cache issues with
some delay, as was the case here.

The severity seemed to warrant a stable tag.

Fixes: 59dc76b0d4df ("mm: vmscan: reduce size of inactive file list")
Link: http://lkml.kernel.org/r/20170404220052.27593-1-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>	[4.7+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

2 files changed, 63 insertions, 3 deletions

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index c5ebb32fef49..cfa91a3ca0ca 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -57,6 +57,9 @@ enum mem_cgroup_stat_index {
 	MEMCG_SLAB_RECLAIMABLE,
 	MEMCG_SLAB_UNRECLAIMABLE,
 	MEMCG_SOCK,
+	MEMCG_WORKINGSET_REFAULT,
+	MEMCG_WORKINGSET_ACTIVATE,
+	MEMCG_WORKINGSET_NODERECLAIM,
 	MEMCG_NR_STAT,
 };
 
@@ -495,6 +498,40 @@ extern int do_swap_account;
 void lock_page_memcg(struct page *page);
 void unlock_page_memcg(struct page *page);
 
+static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg,
+						 enum mem_cgroup_stat_index idx)
+{
+	long val = 0;
+	int cpu;
+
+	for_each_possible_cpu(cpu)
+		val += per_cpu(memcg->stat->count[idx], cpu);
+
+	if (val < 0)
+		val = 0;
+
+	return val;
+}
+
+static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg,
+				   enum mem_cgroup_stat_index idx, int val)
+{
+	if (!mem_cgroup_disabled())
+		this_cpu_add(memcg->stat->count[idx], val);
+}
+
+static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg,
+				   enum mem_cgroup_stat_index idx)
+{
+	mem_cgroup_update_stat(memcg, idx, 1);
+}
+
+static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg,
+				   enum mem_cgroup_stat_index idx)
+{
+	mem_cgroup_update_stat(memcg, idx, -1);
+}
+
 /**
  * mem_cgroup_update_page_stat - update page state statistics
  * @page: the page
@@ -509,14 +546,14 @@ void unlock_page_memcg(struct page *page);
  *   if (TestClearPageState(page))
  *     mem_cgroup_update_page_stat(page, state, -1);
  *   unlock_page(page) or unlock_page_memcg(page)
+ *
+ * Kernel pages are an exception to this, since they'll never move.
  */
 static inline void mem_cgroup_update_page_stat(struct page *page,
 				 enum mem_cgroup_stat_index idx, int val)
 {
-	VM_BUG_ON(!(rcu_read_lock_held() || PageLocked(page)));
-
 	if (page->mem_cgroup)
-		this_cpu_add(page->mem_cgroup->stat->count[idx], val);
+		mem_cgroup_update_stat(page->mem_cgroup, idx, val);
 }
 
 static inline void mem_cgroup_inc_page_stat(struct page *page,
@@ -741,6 +778,27 @@ static inline bool mem_cgroup_oom_synchronize(bool wait)
 	return false;
 }
 
+static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg,
+						 enum mem_cgroup_stat_index idx)
+{
+	return 0;
+}
+
+static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg,
+				   enum mem_cgroup_stat_index idx, int val)
+{
+}
+
+static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg,
+				   enum mem_cgroup_stat_index idx)
+{
+}
+
+static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg,
+				   enum mem_cgroup_stat_index idx)
+{
+}
+
 static inline void mem_cgroup_update_page_stat(struct page *page,
 					       enum mem_cgroup_stat_index idx,
 					       int nr)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 446cf68c1c09..e0c3c5e3d8a0 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -225,6 +225,8 @@ struct lruvec {
 	struct zone_reclaim_stat	reclaim_stat;
 	/* Evictions & activations on the inactive file list */
 	atomic_long_t			inactive_age;
+	/* Refaults at the time of last reclaim cycle */
+	unsigned long			refaults;
 #ifdef CONFIG_MEMCG
 	struct pglist_data *pgdat;
 #endif