mm/slab: fix the theoretical race by holding proper lock

While processing concurrent allocation, SLAB could be contended a lot
because it did a lots of work with holding a lock.  This patchset try to
reduce the number of critical section to reduce lock contention.  Major
changes are lockless decision to allocate more slab and lockless cpu
cache refill from the newly allocated slab.

Below is the result of concurrent allocation/free in slab allocation
benchmark made by Christoph a long time ago.  I make the output simpler.
The number shows cycle count during alloc/free respectively so less is
better.

  * Before
  Kmalloc N*alloc N*free(32): Average=365/806
  Kmalloc N*alloc N*free(64): Average=452/690
  Kmalloc N*alloc N*free(128): Average=736/886
  Kmalloc N*alloc N*free(256): Average=1167/985
  Kmalloc N*alloc N*free(512): Average=2088/1125
  Kmalloc N*alloc N*free(1024): Average=4115/1184
  Kmalloc N*alloc N*free(2048): Average=8451/1748
  Kmalloc N*alloc N*free(4096): Average=16024/2048

  * After
  Kmalloc N*alloc N*free(32): Average=344/792
  Kmalloc N*alloc N*free(64): Average=347/882
  Kmalloc N*alloc N*free(128): Average=390/959
  Kmalloc N*alloc N*free(256): Average=393/1067
  Kmalloc N*alloc N*free(512): Average=683/1229
  Kmalloc N*alloc N*free(1024): Average=1295/1325
  Kmalloc N*alloc N*free(2048): Average=2513/1664
  Kmalloc N*alloc N*free(4096): Average=4742/2172

It shows that performance improves greatly (roughly more than 50%) for
the object class whose size is more than 128 bytes.

This patch (of 11):

If we don't hold neither the slab_mutex nor the node lock, node's shared
array cache could be freed and re-populated.  If __kmem_cache_shrink()
is called at the same time, it will call drain_array() with n->shared
without holding node lock so problem can happen.  This patch fix the
situation by holding the node lock before trying to drain the shared
array.

In addition, add a debug check to confirm that n->shared access race
doesn't exist.

Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/slab.c

@@ -2180,6 +2180,11 @@ static void check_irq_on(void)
 	BUG_ON(irqs_disabled());
 }
 
+static void check_mutex_acquired(void)
+{
+	BUG_ON(!mutex_is_locked(&slab_mutex));
+}
+
 static void check_spinlock_acquired(struct kmem_cache *cachep)
 {
 #ifdef CONFIG_SMP
@@ -2199,13 +2204,27 @@ static void check_spinlock_acquired_node(struct kmem_cache *cachep, int node)
 #else
 #define check_irq_off()	do { } while(0)
 #define check_irq_on()	do { } while(0)
+#define check_mutex_acquired()	do { } while(0)
 #define check_spinlock_acquired(x) do { } while(0)
 #define check_spinlock_acquired_node(x, y) do { } while(0)
 #endif
 
-static void drain_array(struct kmem_cache *cachep, struct kmem_cache_node *n,
+static void drain_array_locked(struct kmem_cache *cachep, struct array_cache *ac,
-			struct array_cache *ac,
+				int node, bool free_all, struct list_head *list)
-			int force, int node);
+{
+	int tofree;
+
+	if (!ac || !ac->avail)
+		return;
+
+	tofree = free_all ? ac->avail : (ac->limit + 4) / 5;
+	if (tofree > ac->avail)
+		tofree = (ac->avail + 1) / 2;
+
+	free_block(cachep, ac->entry, tofree, node, list);
+	ac->avail -= tofree;
+	memmove(ac->entry, &(ac->entry[tofree]), sizeof(void *) * ac->avail);
+}
 
 static void do_drain(void *arg)
 {
@@ -2229,6 +2248,7 @@ static void drain_cpu_caches(struct kmem_cache *cachep)
 {
 	struct kmem_cache_node *n;
 	int node;
+	LIST_HEAD(list);
 
 	on_each_cpu(do_drain, cachep, 1);
 	check_irq_on();
@@ -2236,8 +2256,13 @@ static void drain_cpu_caches(struct kmem_cache *cachep)
 		if (n->alien)
 			drain_alien_cache(cachep, n->alien);
 
-	for_each_kmem_cache_node(cachep, node, n)
+	for_each_kmem_cache_node(cachep, node, n) {
-		drain_array(cachep, n, n->shared, 1, node);
+		spin_lock_irq(&n->list_lock);
+		drain_array_locked(cachep, n->shared, node, true, &list);
+		spin_unlock_irq(&n->list_lock);
+
+		slabs_destroy(cachep, &list);
+	}
 }
 
 /*
@@ -3869,29 +3894,26 @@ skip_setup:
  * if drain_array() is used on the shared array.
  */
 static void drain_array(struct kmem_cache *cachep, struct kmem_cache_node *n,
-			 struct array_cache *ac, int force, int node)
+			 struct array_cache *ac, int node)
 {
 	LIST_HEAD(list);
-	int tofree;
+
+	/* ac from n->shared can be freed if we don't hold the slab_mutex. */
+	check_mutex_acquired();
 
 	if (!ac || !ac->avail)
 		return;
-	if (ac->touched && !force) {
+
+	if (ac->touched) {
 		ac->touched = 0;
-	} else {
+		return;
-		spin_lock_irq(&n->list_lock);
-		if (ac->avail) {
-			tofree = force ? ac->avail : (ac->limit + 4) / 5;
-			if (tofree > ac->avail)
-				tofree = (ac->avail + 1) / 2;
-			free_block(cachep, ac->entry, tofree, node, &list);
-			ac->avail -= tofree;
-			memmove(ac->entry, &(ac->entry[tofree]),
-				sizeof(void *) * ac->avail);
-		}
-		spin_unlock_irq(&n->list_lock);
-		slabs_destroy(cachep, &list);
 	}
+
+	spin_lock_irq(&n->list_lock);
+	drain_array_locked(cachep, ac, node, false, &list);
+	spin_unlock_irq(&n->list_lock);
+
+	slabs_destroy(cachep, &list);
 }
 
 /**
@@ -3929,7 +3951,7 @@ static void cache_reap(struct work_struct *w)
 
 		reap_alien(searchp, n);
 
-		drain_array(searchp, n, cpu_cache_get(searchp), 0, node);
+		drain_array(searchp, n, cpu_cache_get(searchp), node);
 
 		/*
 		 * These are racy checks but it does not matter
@@ -3940,7 +3962,7 @@ static void cache_reap(struct work_struct *w)
 
 		n->next_reap = jiffies + REAPTIMEOUT_NODE;
 
-		drain_array(searchp, n, n->shared, 0, node);
+		drain_array(searchp, n, n->shared, node);
 
 		if (n->free_touched)
 			n->free_touched = 0;