3 files changed, 45 insertions, 2 deletions

diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index d85889710f9b..fd5be240c0b7 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -40,6 +40,7 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
                 PGSCAN_ZONE_RECLAIM_FAILED,
 #endif
                 PGINODESTEAL, SLABS_SCANNED, KSWAPD_STEAL, KSWAPD_INODESTEAL,
+                KSWAPD_PREMATURE_FAST, KSWAPD_PREMATURE_SLOW,
                 PAGEOUTRUN, ALLOCSTALL, PGROTATED,
 #ifdef CONFIG_HUGETLB_PAGE
                 HTLB_BUDDY_PGALLOC, HTLB_BUDDY_PGALLOC_FAIL,
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 61d3a9a0d96f..e176bd3936da 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1904,6 +1904,24 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont,
 }
 #endif
 
+/* is kswapd sleeping prematurely? */
+static int sleeping_prematurely(int order, long remaining)
+{
+        struct zone *zone;
+
+        /* If a direct reclaimer woke kswapd within HZ/10, it's premature */
+        if (remaining)
+                return 1;
+
+        /* If after HZ/10, a zone is below the high mark, it's premature */
+        for_each_populated_zone(zone)
+                if (!zone_watermark_ok(zone, order, high_wmark_pages(zone),
+                                                                0, 0))
+                        return 1;
+
+        return 0;
+}
+
 /*
  * For kswapd, balance_pgdat() will work across all this node's zones until
  * they are all at high_wmark_pages(zone).
@@ -2185,8 +2203,30 @@ static int kswapd(void *p)
                          */
                         order = new_order;
                 } else {
-                        if (!freezing(current) && !kthread_should_stop())
-                                schedule();
+                        if (!freezing(current) && !kthread_should_stop()) {
+                                long remaining = 0;
+
+                                /* Try to sleep for a short interval */
+                                if (!sleeping_prematurely(order, remaining)) {
+                                        remaining = schedule_timeout(HZ/10);
+                                        finish_wait(&pgdat->kswapd_wait, &wait);
+                                        prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
+                                }
+
+                                /*
+                                 * After a short sleep, check if it was a
+                                 * premature sleep. If not, then go fully
+                                 * to sleep until explicitly woken up
+                                 */
+                                if (!sleeping_prematurely(order, remaining))
+                                        schedule();
+                                else {
+                                        if (remaining)
+                                                count_vm_event(KSWAPD_PREMATURE_FAST);
+                                        else
+                                                count_vm_event(KSWAPD_PREMATURE_SLOW);
+                                }
+                        }
 
                         order = pgdat->kswapd_max_order;
                 }
diff --git a/mm/vmstat.c b/mm/vmstat.c
index dad2327e4580..63ab71455c5b 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -683,6 +683,8 @@ static const char * const vmstat_text[] = {
         "slabs_scanned",
         "kswapd_steal",
         "kswapd_inodesteal",
+        "kswapd_slept_prematurely_fast",
+        "kswapd_slept_prematurely_slow",
         "pageoutrun",
         "allocstall",