vmstat: create separate function to fold per cpu diffs into local counters

The main idea behind this patchset is to reduce the vmstat update overhead
by avoiding interrupt enable/disable and the use of per cpu atomics.

This patch (of 3):

It is better to have a separate folding function because
refresh_cpu_vm_stats() also does other things like expire pages in the
page allocator caches.

If we have a separate function then refresh_cpu_vm_stats() is only called
from the local cpu which allows additional optimizations.

The folding function is only called when a cpu is being downed and
therefore no other processor will be accessing the counters.  Also
simplifies synchronization.

[akpm@linux-foundation.org: fix UP build]
Signed-off-by: Christoph Lameter <cl@linux.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
CC: Tejun Heo <tj@kernel.org>
Cc: Joonsoo Kim <js1304@gmail.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

include/linux/vmstat.h

@@ -198,7 +198,7 @@ extern void __inc_zone_state(struct zone *, enum zone_stat_item);
 extern void dec_zone_state(struct zone *, enum zone_stat_item);
 extern void __dec_zone_state(struct zone *, enum zone_stat_item);
 
-void refresh_cpu_vm_stats(int);
+void cpu_vm_stats_fold(int cpu);
 void refresh_zone_stat_thresholds(void);
 
 void drain_zonestat(struct zone *zone, struct per_cpu_pageset *);
@@ -255,6 +255,7 @@ static inline void __dec_zone_page_state(struct page *page,
 
 static inline void refresh_cpu_vm_stats(int cpu) { }
 static inline void refresh_zone_stat_thresholds(void) { }
+static inline void cpu_vm_stats_fold(int cpu) { }
 
 static inline void drain_zonestat(struct zone *zone,
 			struct per_cpu_pageset *pset) { }

mm/page_alloc.c

@@ -5435,7 +5435,7 @@ static int page_alloc_cpu_notify(struct notifier_block *self,
 		 * This is only okay since the processor is dead and cannot
 		 * race with what we are doing.
 		 */
-		refresh_cpu_vm_stats(cpu);
+		cpu_vm_stats_fold(cpu);
 	}
 	return NOTIFY_OK;
 }

mm/vmstat.c

@@ -415,11 +415,7 @@ EXPORT_SYMBOL(dec_zone_page_state);
 #endif
 
 /*
- * Update the zone counters for one cpu.
- *
- * The cpu specified must be either the current cpu or a processor that
- * is not online. If it is the current cpu then the execution thread must
- * be pinned to the current cpu.
+ * Update the zone counters for the current cpu.
  *
  * Note that refresh_cpu_vm_stats strives to only access
  * node local memory. The per cpu pagesets on remote zones are placed
@@ -432,7 +428,7 @@ EXPORT_SYMBOL(dec_zone_page_state);
  * with the global counters. These could cause remote node cache line
  * bouncing and will have to be only done when necessary.
  */
-void refresh_cpu_vm_stats(int cpu)
+static void refresh_cpu_vm_stats(int cpu)
 {
 	struct zone *zone;
 	int i;
@@ -493,6 +489,38 @@ void refresh_cpu_vm_stats(int cpu)
 			atomic_long_add(global_diff[i], &vm_stat[i]);
 }
 
+/*
+ * Fold the data for an offline cpu into the global array.
+ * There cannot be any access by the offline cpu and therefore
+ * synchronization is simplified.
+ */
+void cpu_vm_stats_fold(int cpu)
+{
+	struct zone *zone;
+	int i;
+	int global_diff[NR_VM_ZONE_STAT_ITEMS] = { 0, };
+
+	for_each_populated_zone(zone) {
+		struct per_cpu_pageset *p;
+
+		p = per_cpu_ptr(zone->pageset, cpu);
+
+		for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
+			if (p->vm_stat_diff[i]) {
+				int v;
+
+				v = p->vm_stat_diff[i];
+				p->vm_stat_diff[i] = 0;
+				atomic_long_add(v, &zone->vm_stat[i]);
+				global_diff[i] += v;
+			}
+	}
+
+	for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
+		if (global_diff[i])
+			atomic_long_add(global_diff[i], &vm_stat[i]);
+}
+
 /*
  * this is only called if !populated_zone(zone), which implies no other users of
  * pset->vm_stat_diff[] exsist.