3 files changed, 31 insertions, 15 deletions

diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index bd4b34c03738..4415aa915681 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -702,7 +702,8 @@ The batch value of each per cpu pagelist is also updated as a result.  It is
 set to pcp->high/4.  The upper limit of batch is (PAGE_SHIFT * 8)
 
 The initial value is zero.  Kernel does not use this value at boot time to set
-the high water marks for each per cpu page list.
+the high water marks for each per cpu page list.  If the user writes '0' to this
+sysctl, it will revert to this default behavior.
 
 ==============================================================
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 7de6555cfea0..075d1903138f 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -136,7 +136,6 @@ static unsigned long dirty_bytes_min = 2 * PAGE_SIZE;
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
 static int maxolduid = 65535;
 static int minolduid;
-static int min_percpu_pagelist_fract = 8;
 
 static int ngroups_max = NGROUPS_MAX;
 static const int cap_last_cap = CAP_LAST_CAP;
@@ -1317,7 +1316,7 @@ static struct ctl_table vm_table[] = {
                 .maxlen         = sizeof(percpu_pagelist_fraction),
                 .mode           = 0644,
                 .proc_handler   = percpu_pagelist_fraction_sysctl_handler,
-                .extra1         = &min_percpu_pagelist_fract,
+                .extra1         = &zero,
         },
 #ifdef CONFIG_MMU
         {
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4f59fa29eda8..20d17f8266fe 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -69,6 +69,7 @@
 
 /* prevent >1 _updater_ of zone percpu pageset ->high and ->batch fields */
 static DEFINE_MUTEX(pcp_batch_high_lock);
+#define MIN_PERCPU_PAGELIST_FRACTION    (8)
 
 #ifdef CONFIG_USE_PERCPU_NUMA_NODE_ID
 DEFINE_PER_CPU(int, numa_node);
@@ -4145,7 +4146,7 @@ static void __meminit zone_init_free_lists(struct zone *zone)
         memmap_init_zone((size), (nid), (zone), (start_pfn), MEMMAP_EARLY)
 #endif
 
-static int __meminit zone_batchsize(struct zone *zone)
+static int zone_batchsize(struct zone *zone)
 {
 #ifdef CONFIG_MMU
         int batch;
@@ -4261,8 +4262,8 @@ static void pageset_set_high(struct per_cpu_pageset *p,
         pageset_update(&p->pcp, high, batch);
 }
 
-static void __meminit pageset_set_high_and_batch(struct zone *zone,
-                struct per_cpu_pageset *pcp)
+static void pageset_set_high_and_batch(struct zone *zone,
+                                       struct per_cpu_pageset *pcp)
 {
         if (percpu_pagelist_fraction)
                 pageset_set_high(pcp,
@@ -5881,23 +5882,38 @@ int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *table, int write,
         void __user *buffer, size_t *length, loff_t *ppos)
 {
         struct zone *zone;
-        unsigned int cpu;
+        int old_percpu_pagelist_fraction;
         int ret;
 
+        mutex_lock(&pcp_batch_high_lock);
+        old_percpu_pagelist_fraction = percpu_pagelist_fraction;
+
         ret = proc_dointvec_minmax(table, write, buffer, length, ppos);
-        if (!write || (ret < 0))
-                return ret;
+        if (!write || ret < 0)
+                goto out;
+
+        /* Sanity checking to avoid pcp imbalance */
+        if (percpu_pagelist_fraction &&
+            percpu_pagelist_fraction < MIN_PERCPU_PAGELIST_FRACTION) {
+                percpu_pagelist_fraction = old_percpu_pagelist_fraction;
+                ret = -EINVAL;
+                goto out;
+        }
+
+        /* No change? */
+        if (percpu_pagelist_fraction == old_percpu_pagelist_fraction)
+                goto out;
 
-        mutex_lock(&pcp_batch_high_lock);
         for_each_populated_zone(zone) {
-                unsigned long  high;
-                high = zone->managed_pages / percpu_pagelist_fraction;
+                unsigned int cpu;
+
                 for_each_possible_cpu(cpu)
-                        pageset_set_high(per_cpu_ptr(zone->pageset, cpu),
-                                         high);
+                        pageset_set_high_and_batch(zone,
+                                        per_cpu_ptr(zone->pageset, cpu));
         }
+out:
         mutex_unlock(&pcp_batch_high_lock);
-        return 0;
+        return ret;
 }
 
 int hashdist = HASHDIST_DEFAULT;