Merge branch 'akpm' (more incoming from Andrew)

Merge second patch-bomb from Andrew Morton:

 - A little DM fix

 - the MM queue

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (154 commits)
  ksm: allocate roots when needed
  mm: cleanup "swapcache" in do_swap_page
  mm,ksm: swapoff might need to copy
  mm,ksm: FOLL_MIGRATION do migration_entry_wait
  ksm: shrink 32-bit rmap_item back to 32 bytes
  ksm: treat unstable nid like in stable tree
  ksm: add some comments
  tmpfs: fix mempolicy object leaks
  tmpfs: fix use-after-free of mempolicy object
  mm/fadvise.c: drain all pagevecs if POSIX_FADV_DONTNEED fails to discard all pages
  mm: export mmu notifier invalidates
  mm: accelerate mm_populate() treatment of THP pages
  mm: use long type for page counts in mm_populate() and get_user_pages()
  mm: accurately document nr_free_*_pages functions with code comments
  HWPOISON: change order of error_states[]'s elements
  HWPOISON: fix misjudgement of page_action() for errors on mlocked pages
  memcg: stop warning on memcg_propagate_kmem
  net: change type of virtio_chan->p9_max_pages
  vmscan: change type of vm_total_pages to unsigned long
  fs/nfsd: change type of max_delegations, nfsd_drc_max_mem and nfsd_drc_mem_used
  ...

1 files changed, 268 insertions, 205 deletions

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index fbb60b103e64..53b8201b31eb 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -120,6 +120,14 @@ static const char * const mem_cgroup_events_names[] = {
 	"pgmajfault",
 };
 
+static const char * const mem_cgroup_lru_names[] = {
+	"inactive_anon",
+	"active_anon",
+	"inactive_file",
+	"active_file",
+	"unevictable",
+};
+
 /*
  * Per memcg event counter is incremented at every pagein/pageout. With THP,
  * it will be incremated by the number of pages. This counter is used for
@@ -172,7 +180,7 @@ struct mem_cgroup_per_node {
 };
 
 struct mem_cgroup_lru_info {
-	struct mem_cgroup_per_node *nodeinfo[MAX_NUMNODES];
+	struct mem_cgroup_per_node *nodeinfo[0];
 };
 
 /*
@@ -276,17 +284,6 @@ struct mem_cgroup {
 	 */
 	struct res_counter kmem;
 	/*
-	 * Per cgroup active and inactive list, similar to the
-	 * per zone LRU lists.
-	 */
-	struct mem_cgroup_lru_info info;
-	int last_scanned_node;
-#if MAX_NUMNODES > 1
-	nodemask_t	scan_nodes;
-	atomic_t	numainfo_events;
-	atomic_t	numainfo_updating;
-#endif
-	/*
 	 * Should the accounting and control be hierarchical, per subtree?
 	 */
 	bool use_hierarchy;
@@ -349,8 +346,29 @@ struct mem_cgroup {
         /* Index in the kmem_cache->memcg_params->memcg_caches array */
 	int kmemcg_id;
 #endif
+
+	int last_scanned_node;
+#if MAX_NUMNODES > 1
+	nodemask_t	scan_nodes;
+	atomic_t	numainfo_events;
+	atomic_t	numainfo_updating;
+#endif
+	/*
+	 * Per cgroup active and inactive list, similar to the
+	 * per zone LRU lists.
+	 *
+	 * WARNING: This has to be the last element of the struct. Don't
+	 * add new fields after this point.
+	 */
+	struct mem_cgroup_lru_info info;
 };
 
+static size_t memcg_size(void)
+{
+	return sizeof(struct mem_cgroup) +
+		nr_node_ids * sizeof(struct mem_cgroup_per_node);
+}
+
 /* internal only representation about the status of kmem accounting. */
 enum {
 	KMEM_ACCOUNTED_ACTIVE = 0, /* accounted by this cgroup itself */
@@ -398,8 +416,8 @@ static bool memcg_kmem_test_and_clear_dead(struct mem_cgroup *memcg)
 
 /* Stuffs for move charges at task migration. */
 /*
- * Types of charges to be moved. "move_charge_at_immitgrate" is treated as a
- * left-shifted bitmap of these types.
+ * Types of charges to be moved. "move_charge_at_immitgrate" and
+ * "immigrate_flags" are treated as a left-shifted bitmap of these types.
  */
 enum move_type {
 	MOVE_CHARGE_TYPE_ANON,	/* private anonymous page and swap of it */
@@ -412,6 +430,7 @@ static struct move_charge_struct {
 	spinlock_t	  lock; /* for from, to */
 	struct mem_cgroup *from;
 	struct mem_cgroup *to;
+	unsigned long immigrate_flags;
 	unsigned long precharge;
 	unsigned long moved_charge;
 	unsigned long moved_swap;
@@ -424,14 +443,12 @@ static struct move_charge_struct {
 
 static bool move_anon(void)
 {
-	return test_bit(MOVE_CHARGE_TYPE_ANON,
-					&mc.to->move_charge_at_immigrate);
+	return test_bit(MOVE_CHARGE_TYPE_ANON, &mc.immigrate_flags);
 }
 
 static bool move_file(void)
 {
-	return test_bit(MOVE_CHARGE_TYPE_FILE,
-					&mc.to->move_charge_at_immigrate);
+	return test_bit(MOVE_CHARGE_TYPE_FILE, &mc.immigrate_flags);
 }
 
 /*
@@ -471,6 +488,13 @@ enum res_type {
 #define MEM_CGROUP_RECLAIM_SHRINK_BIT	0x1
 #define MEM_CGROUP_RECLAIM_SHRINK	(1 << MEM_CGROUP_RECLAIM_SHRINK_BIT)
 
+/*
+ * The memcg_create_mutex will be held whenever a new cgroup is created.
+ * As a consequence, any change that needs to protect against new child cgroups
+ * appearing has to hold it as well.
+ */
+static DEFINE_MUTEX(memcg_create_mutex);
+
 static void mem_cgroup_get(struct mem_cgroup *memcg);
 static void mem_cgroup_put(struct mem_cgroup *memcg);
 
@@ -627,6 +651,7 @@ static void drain_all_stock_async(struct mem_cgroup *memcg);
 static struct mem_cgroup_per_zone *
 mem_cgroup_zoneinfo(struct mem_cgroup *memcg, int nid, int zid)
 {
+	VM_BUG_ON((unsigned)nid >= nr_node_ids);
 	return &memcg->info.nodeinfo[nid]->zoneinfo[zid];
 }
 
@@ -1371,17 +1396,6 @@ int mem_cgroup_inactive_anon_is_low(struct lruvec *lruvec)
 	return inactive * inactive_ratio < active;
 }
 
-int mem_cgroup_inactive_file_is_low(struct lruvec *lruvec)
-{
-	unsigned long active;
-	unsigned long inactive;
-
-	inactive = mem_cgroup_get_lru_size(lruvec, LRU_INACTIVE_FILE);
-	active = mem_cgroup_get_lru_size(lruvec, LRU_ACTIVE_FILE);
-
-	return (active > inactive);
-}
-
 #define mem_cgroup_from_res_counter(counter, member)	\
 	container_of(counter, struct mem_cgroup, member)
 
@@ -1524,8 +1538,9 @@ static void move_unlock_mem_cgroup(struct mem_cgroup *memcg,
 	spin_unlock_irqrestore(&memcg->move_lock, *flags);
 }
 
+#define K(x) ((x) << (PAGE_SHIFT-10))
 /**
- * mem_cgroup_print_oom_info: Called from OOM with tasklist_lock held in read mode.
+ * mem_cgroup_print_oom_info: Print OOM information relevant to memory controller.
  * @memcg: The memory cgroup that went over limit
  * @p: Task that is going to be killed
  *
@@ -1543,8 +1558,10 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p)
 	 */
 	static char memcg_name[PATH_MAX];
 	int ret;
+	struct mem_cgroup *iter;
+	unsigned int i;
 
-	if (!memcg || !p)
+	if (!p)
 		return;
 
 	rcu_read_lock();
@@ -1563,7 +1580,7 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p)
 	}
 	rcu_read_unlock();
 
-	printk(KERN_INFO "Task in %s killed", memcg_name);
+	pr_info("Task in %s killed", memcg_name);
 
 	rcu_read_lock();
 	ret = cgroup_path(mem_cgrp, memcg_name, PATH_MAX);
@@ -1576,22 +1593,45 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p)
 	/*
 	 * Continues from above, so we don't need an KERN_ level
 	 */
-	printk(KERN_CONT " as a result of limit of %s\n", memcg_name);
+	pr_cont(" as a result of limit of %s\n", memcg_name);
 done:
 
-	printk(KERN_INFO "memory: usage %llukB, limit %llukB, failcnt %llu\n",
+	pr_info("memory: usage %llukB, limit %llukB, failcnt %llu\n",
 		res_counter_read_u64(&memcg->res, RES_USAGE) >> 10,
 		res_counter_read_u64(&memcg->res, RES_LIMIT) >> 10,
 		res_counter_read_u64(&memcg->res, RES_FAILCNT));
-	printk(KERN_INFO "memory+swap: usage %llukB, limit %llukB, "
-		"failcnt %llu\n",
+	pr_info("memory+swap: usage %llukB, limit %llukB, failcnt %llu\n",
 		res_counter_read_u64(&memcg->memsw, RES_USAGE) >> 10,
 		res_counter_read_u64(&memcg->memsw, RES_LIMIT) >> 10,
 		res_counter_read_u64(&memcg->memsw, RES_FAILCNT));
-	printk(KERN_INFO "kmem: usage %llukB, limit %llukB, failcnt %llu\n",
+	pr_info("kmem: usage %llukB, limit %llukB, failcnt %llu\n",
 		res_counter_read_u64(&memcg->kmem, RES_USAGE) >> 10,
 		res_counter_read_u64(&memcg->kmem, RES_LIMIT) >> 10,
 		res_counter_read_u64(&memcg->kmem, RES_FAILCNT));
+
+	for_each_mem_cgroup_tree(iter, memcg) {
+		pr_info("Memory cgroup stats");
+
+		rcu_read_lock();
+		ret = cgroup_path(iter->css.cgroup, memcg_name, PATH_MAX);
+		if (!ret)
+			pr_cont(" for %s", memcg_name);
+		rcu_read_unlock();
+		pr_cont(":");
+
+		for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) {
+			if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account)
+				continue;
+			pr_cont(" %s:%ldKB", mem_cgroup_stat_names[i],
+				K(mem_cgroup_read_stat(iter, i)));
+		}
+
+		for (i = 0; i < NR_LRU_LISTS; i++)
+			pr_cont(" %s:%luKB", mem_cgroup_lru_names[i],
+				K(mem_cgroup_nr_lru_pages(iter, BIT(i))));
+
+		pr_cont("\n");
+	}
 }
 
 /*
@@ -2256,6 +2296,17 @@ static void drain_local_stock(struct work_struct *dummy)
 	clear_bit(FLUSHING_CACHED_CHARGE, &stock->flags);
 }
 
+static void __init memcg_stock_init(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		struct memcg_stock_pcp *stock =
+					&per_cpu(memcg_stock, cpu);
+		INIT_WORK(&stock->work, drain_local_stock);
+	}
+}
+
 /*
  * Cache charges(val) which is from res_counter, to local per_cpu area.
  * This will be consumed by consume_stock() function, later.
@@ -4391,8 +4442,8 @@ void mem_cgroup_print_bad_page(struct page *page)
 
 	pc = lookup_page_cgroup_used(page);
 	if (pc) {
-		printk(KERN_ALERT "pc:%p pc->flags:%lx pc->mem_cgroup:%p\n",
-		       pc, pc->flags, pc->mem_cgroup);
+		pr_alert("pc:%p pc->flags:%lx pc->mem_cgroup:%p\n",
+			 pc, pc->flags, pc->mem_cgroup);
 	}
 }
 #endif
@@ -4719,6 +4770,33 @@ static void mem_cgroup_reparent_charges(struct mem_cgroup *memcg)
 }
 
 /*
+ * This mainly exists for tests during the setting of set of use_hierarchy.
+ * Since this is the very setting we are changing, the current hierarchy value
+ * is meaningless
+ */
+static inline bool __memcg_has_children(struct mem_cgroup *memcg)
+{
+	struct cgroup *pos;
+
+	/* bounce at first found */
+	cgroup_for_each_child(pos, memcg->css.cgroup)
+		return true;
+	return false;
+}
+
+/*
+ * Must be called with memcg_create_mutex held, unless the cgroup is guaranteed
+ * to be already dead (as in mem_cgroup_force_empty, for instance).  This is
+ * from mem_cgroup_count_children(), in the sense that we don't really care how
+ * many children we have; we only need to know if we have any.  It also counts
+ * any memcg without hierarchy as infertile.
+ */
+static inline bool memcg_has_children(struct mem_cgroup *memcg)
+{
+	return memcg->use_hierarchy && __memcg_has_children(memcg);
+}
+
+/*
  * Reclaims as many pages from the given memcg as possible and moves
  * the rest to the parent.
  *
@@ -4788,7 +4866,7 @@ static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
 	if (parent)
 		parent_memcg = mem_cgroup_from_cont(parent);
 
-	cgroup_lock();
+	mutex_lock(&memcg_create_mutex);
 
 	if (memcg->use_hierarchy == val)
 		goto out;
@@ -4803,7 +4881,7 @@ static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
 	 */
 	if ((!parent_memcg || !parent_memcg->use_hierarchy) &&
 				(val == 1 || val == 0)) {
-		if (list_empty(&cont->children))
+		if (!__memcg_has_children(memcg))
 			memcg->use_hierarchy = val;
 		else
 			retval = -EBUSY;
@@ -4811,7 +4889,7 @@ static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
 		retval = -EINVAL;
 
 out:
-	cgroup_unlock();
+	mutex_unlock(&memcg_create_mutex);
 
 	return retval;
 }
@@ -4896,8 +4974,6 @@ static int memcg_update_kmem_limit(struct cgroup *cont, u64 val)
 {
 	int ret = -EINVAL;
 #ifdef CONFIG_MEMCG_KMEM
-	bool must_inc_static_branch = false;
-
 	struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
 	/*
 	 * For simplicity, we won't allow this to be disabled.  It also can't
@@ -4910,18 +4986,11 @@ static int memcg_update_kmem_limit(struct cgroup *cont, u64 val)
 	 *
 	 * After it first became limited, changes in the value of the limit are
 	 * of course permitted.
-	 *
-	 * Taking the cgroup_lock is really offensive, but it is so far the only
-	 * way to guarantee that no children will appear. There are plenty of
-	 * other offenders, and they should all go away. Fine grained locking
-	 * is probably the way to go here. When we are fully hierarchical, we
-	 * can also get rid of the use_hierarchy check.
 	 */
-	cgroup_lock();
+	mutex_lock(&memcg_create_mutex);
 	mutex_lock(&set_limit_mutex);
 	if (!memcg->kmem_account_flags && val != RESOURCE_MAX) {
-		if (cgroup_task_count(cont) || (memcg->use_hierarchy &&
-						!list_empty(&cont->children))) {
+		if (cgroup_task_count(cont) || memcg_has_children(memcg)) {
 			ret = -EBUSY;
 			goto out;
 		}
@@ -4933,7 +5002,13 @@ static int memcg_update_kmem_limit(struct cgroup *cont, u64 val)
 			res_counter_set_limit(&memcg->kmem, RESOURCE_MAX);
 			goto out;
 		}
-		must_inc_static_branch = true;
+		static_key_slow_inc(&memcg_kmem_enabled_key);
+		/*
+		 * setting the active bit after the inc will guarantee no one
+		 * starts accounting before all call sites are patched
+		 */
+		memcg_kmem_set_active(memcg);
+
 		/*
 		 * kmem charges can outlive the cgroup. In the case of slab
 		 * pages, for instance, a page contain objects from various
@@ -4945,32 +5020,12 @@ static int memcg_update_kmem_limit(struct cgroup *cont, u64 val)
 		ret = res_counter_set_limit(&memcg->kmem, val);
 out:
 	mutex_unlock(&set_limit_mutex);
-	cgroup_unlock();
-
-	/*
-	 * We are by now familiar with the fact that we can't inc the static
-	 * branch inside cgroup_lock. See disarm functions for details. A
-	 * worker here is overkill, but also wrong: After the limit is set, we
-	 * must start accounting right away. Since this operation can't fail,
-	 * we can safely defer it to here - no rollback will be needed.
-	 *
-	 * The boolean used to control this is also safe, because
-	 * KMEM_ACCOUNTED_ACTIVATED guarantees that only one process will be
-	 * able to set it to true;
-	 */
-	if (must_inc_static_branch) {
-		static_key_slow_inc(&memcg_kmem_enabled_key);
-		/*
-		 * setting the active bit after the inc will guarantee no one
-		 * starts accounting before all call sites are patched
-		 */
-		memcg_kmem_set_active(memcg);
-	}
-
+	mutex_unlock(&memcg_create_mutex);
 #endif
 	return ret;
 }
 
+#ifdef CONFIG_MEMCG_KMEM
 static int memcg_propagate_kmem(struct mem_cgroup *memcg)
 {
 	int ret = 0;
@@ -4979,7 +5034,6 @@ static int memcg_propagate_kmem(struct mem_cgroup *memcg)
 		goto out;
 
 	memcg->kmem_account_flags = parent->kmem_account_flags;
-#ifdef CONFIG_MEMCG_KMEM
 	/*
 	 * When that happen, we need to disable the static branch only on those
 	 * memcgs that enabled it. To achieve this, we would be forced to
@@ -5005,10 +5059,10 @@ static int memcg_propagate_kmem(struct mem_cgroup *memcg)
 	mutex_lock(&set_limit_mutex);
 	ret = memcg_update_cache_sizes(memcg);
 	mutex_unlock(&set_limit_mutex);
-#endif
 out:
 	return ret;
 }
+#endif /* CONFIG_MEMCG_KMEM */
 
 /*
  * The user of this function is...
@@ -5148,15 +5202,14 @@ static int mem_cgroup_move_charge_write(struct cgroup *cgrp,
 
 	if (val >= (1 << NR_MOVE_TYPE))
 		return -EINVAL;
+
 	/*
-	 * We check this value several times in both in can_attach() and
-	 * attach(), so we need cgroup lock to prevent this value from being
-	 * inconsistent.
+	 * No kind of locking is needed in here, because ->can_attach() will
+	 * check this value once in the beginning of the process, and then carry
+	 * on with stale data. This means that changes to this value will only
+	 * affect task migrations starting after the change.
 	 */
-	cgroup_lock();
 	memcg->move_charge_at_immigrate = val;
-	cgroup_unlock();
-
 	return 0;
 }
 #else
@@ -5214,14 +5267,6 @@ static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 }
 #endif /* CONFIG_NUMA */
 
-static const char * const mem_cgroup_lru_names[] = {
-	"inactive_anon",
-	"active_anon",
-	"inactive_file",
-	"active_file",
-	"unevictable",
-};
-
 static inline void mem_cgroup_lru_names_not_uptodate(void)
 {
 	BUILD_BUG_ON(ARRAY_SIZE(mem_cgroup_lru_names) != NR_LRU_LISTS);
@@ -5335,18 +5380,17 @@ static int mem_cgroup_swappiness_write(struct cgroup *cgrp, struct cftype *cft,
 
 	parent = mem_cgroup_from_cont(cgrp->parent);
 
-	cgroup_lock();
+	mutex_lock(&memcg_create_mutex);
 
 	/* If under hierarchy, only empty-root can set this value */
-	if ((parent->use_hierarchy) ||
-	    (memcg->use_hierarchy && !list_empty(&cgrp->children))) {
-		cgroup_unlock();
+	if ((parent->use_hierarchy) || memcg_has_children(memcg)) {
+		mutex_unlock(&memcg_create_mutex);
 		return -EINVAL;
 	}
 
 	memcg->swappiness = val;
 
-	cgroup_unlock();
+	mutex_unlock(&memcg_create_mutex);
 
 	return 0;
 }
@@ -5672,17 +5716,16 @@ static int mem_cgroup_oom_control_write(struct cgroup *cgrp,
 
 	parent = mem_cgroup_from_cont(cgrp->parent);
 
-	cgroup_lock();
+	mutex_lock(&memcg_create_mutex);
 	/* oom-kill-disable is a flag for subhierarchy. */
-	if ((parent->use_hierarchy) ||
-	    (memcg->use_hierarchy && !list_empty(&cgrp->children))) {
-		cgroup_unlock();
+	if ((parent->use_hierarchy) || memcg_has_children(memcg)) {
+		mutex_unlock(&memcg_create_mutex);
 		return -EINVAL;
 	}
 	memcg->oom_kill_disable = val;
 	if (!val)
 		memcg_oom_recover(memcg);
-	cgroup_unlock();
+	mutex_unlock(&memcg_create_mutex);
 	return 0;
 }
 
@@ -5797,33 +5840,6 @@ static struct cftype mem_cgroup_files[] = {
 		.read_seq_string = memcg_numa_stat_show,
 	},
 #endif
-#ifdef CONFIG_MEMCG_SWAP
-	{
-		.name = "memsw.usage_in_bytes",
-		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_USAGE),
-		.read = mem_cgroup_read,
-		.register_event = mem_cgroup_usage_register_event,
-		.unregister_event = mem_cgroup_usage_unregister_event,
-	},
-	{
-		.name = "memsw.max_usage_in_bytes",
-		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_MAX_USAGE),
-		.trigger = mem_cgroup_reset,
-		.read = mem_cgroup_read,
-	},
-	{
-		.name = "memsw.limit_in_bytes",
-		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_LIMIT),
-		.write_string = mem_cgroup_write,
-		.read = mem_cgroup_read,
-	},
-	{
-		.name = "memsw.failcnt",
-		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_FAILCNT),
-		.trigger = mem_cgroup_reset,
-		.read = mem_cgroup_read,
-	},
-#endif
 #ifdef CONFIG_MEMCG_KMEM
 	{
 		.name = "kmem.limit_in_bytes",
@@ -5858,6 +5874,36 @@ static struct cftype mem_cgroup_files[] = {
 	{ },	/* terminate */
 };
 
+#ifdef CONFIG_MEMCG_SWAP
+static struct cftype memsw_cgroup_files[] = {
+	{
+		.name = "memsw.usage_in_bytes",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_USAGE),
+		.read = mem_cgroup_read,
+		.register_event = mem_cgroup_usage_register_event,
+		.unregister_event = mem_cgroup_usage_unregister_event,
+	},
+	{
+		.name = "memsw.max_usage_in_bytes",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_MAX_USAGE),
+		.trigger = mem_cgroup_reset,
+		.read = mem_cgroup_read,
+	},
+	{
+		.name = "memsw.limit_in_bytes",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_LIMIT),
+		.write_string = mem_cgroup_write,
+		.read = mem_cgroup_read,
+	},
+	{
+		.name = "memsw.failcnt",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_FAILCNT),
+		.trigger = mem_cgroup_reset,
+		.read = mem_cgroup_read,
+	},
+	{ },	/* terminate */
+};
+#endif
 static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node)
 {
 	struct mem_cgroup_per_node *pn;
@@ -5896,9 +5942,9 @@ static void free_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node)
 static struct mem_cgroup *mem_cgroup_alloc(void)
 {
 	struct mem_cgroup *memcg;
-	int size = sizeof(struct mem_cgroup);
+	size_t size = memcg_size();
 
-	/* Can be very big if MAX_NUMNODES is very big */
+	/* Can be very big if nr_node_ids is very big */
 	if (size < PAGE_SIZE)
 		memcg = kzalloc(size, GFP_KERNEL);
 	else
@@ -5935,7 +5981,7 @@ out_free:
 static void __mem_cgroup_free(struct mem_cgroup *memcg)
 {
 	int node;
-	int size = sizeof(struct mem_cgroup);
+	size_t size = memcg_size();
 
 	mem_cgroup_remove_from_trees(memcg);
 	free_css_id(&mem_cgroup_subsys, &memcg->css);
@@ -6017,19 +6063,7 @@ struct mem_cgroup *parent_mem_cgroup(struct mem_cgroup *memcg)
 }
 EXPORT_SYMBOL(parent_mem_cgroup);
 
-#ifdef CONFIG_MEMCG_SWAP
-static void __init enable_swap_cgroup(void)
-{
-	if (!mem_cgroup_disabled() && really_do_swap_account)
-		do_swap_account = 1;
-}
-#else
-static void __init enable_swap_cgroup(void)
-{
-}
-#endif
-
-static int mem_cgroup_soft_limit_tree_init(void)
+static void __init mem_cgroup_soft_limit_tree_init(void)
 {
 	struct mem_cgroup_tree_per_node *rtpn;
 	struct mem_cgroup_tree_per_zone *rtpz;
@@ -6040,8 +6074,7 @@ static int mem_cgroup_soft_limit_tree_init(void)
 		if (!node_state(node, N_NORMAL_MEMORY))
 			tmp = -1;
 		rtpn = kzalloc_node(sizeof(*rtpn), GFP_KERNEL, tmp);
-		if (!rtpn)
-			goto err_cleanup;
+		BUG_ON(!rtpn);
 
 		soft_limit_tree.rb_tree_per_node[node] = rtpn;
 
@@ -6051,23 +6084,12 @@ static int mem_cgroup_soft_limit_tree_init(void)
 			spin_lock_init(&rtpz->lock);
 		}
 	}
-	return 0;
-
-err_cleanup:
-	for_each_node(node) {
-		if (!soft_limit_tree.rb_tree_per_node[node])
-			break;
-		kfree(soft_limit_tree.rb_tree_per_node[node]);
-		soft_limit_tree.rb_tree_per_node[node] = NULL;
-	}
-	return 1;
-
 }
 
 static struct cgroup_subsys_state * __ref
 mem_cgroup_css_alloc(struct cgroup *cont)
 {
-	struct mem_cgroup *memcg, *parent;
+	struct mem_cgroup *memcg;
 	long error = -ENOMEM;
 	int node;
 
@@ -6081,24 +6103,44 @@ mem_cgroup_css_alloc(struct cgroup *cont)
 
 	/* root ? */
 	if (cont->parent == NULL) {
-		int cpu;
-		enable_swap_cgroup();
-		parent = NULL;
-		if (mem_cgroup_soft_limit_tree_init())
-			goto free_out;
 		root_mem_cgroup = memcg;
-		for_each_possible_cpu(cpu) {
-			struct memcg_stock_pcp *stock =
-						&per_cpu(memcg_stock, cpu);
-			INIT_WORK(&stock->work, drain_local_stock);
-		}
-	} else {
-		parent = mem_cgroup_from_cont(cont->parent);
-		memcg->use_hierarchy = parent->use_hierarchy;
-		memcg->oom_kill_disable = parent->oom_kill_disable;
+		res_counter_init(&memcg->res, NULL);
+		res_counter_init(&memcg->memsw, NULL);
+		res_counter_init(&memcg->kmem, NULL);
 	}
 
-	if (parent && parent->use_hierarchy) {
+	memcg->last_scanned_node = MAX_NUMNODES;
+	INIT_LIST_HEAD(&memcg->oom_notify);
+	atomic_set(&memcg->refcnt, 1);
+	memcg->move_charge_at_immigrate = 0;
+	mutex_init(&memcg->thresholds_lock);
+	spin_lock_init(&memcg->move_lock);
+
+	return &memcg->css;
+
+free_out:
+	__mem_cgroup_free(memcg);
+	return ERR_PTR(error);
+}
+
+static int
+mem_cgroup_css_online(struct cgroup *cont)
+{
+	struct mem_cgroup *memcg, *parent;
+	int error = 0;
+
+	if (!cont->parent)
+		return 0;
+
+	mutex_lock(&memcg_create_mutex);
+	memcg = mem_cgroup_from_cont(cont);
+	parent = mem_cgroup_from_cont(cont->parent);
+
+	memcg->use_hierarchy = parent->use_hierarchy;
+	memcg->oom_kill_disable = parent->oom_kill_disable;
+	memcg->swappiness = mem_cgroup_swappiness(parent);
+
+	if (parent->use_hierarchy) {
 		res_counter_init(&memcg->res, &parent->res);
 		res_counter_init(&memcg->memsw, &parent->memsw);
 		res_counter_init(&memcg->kmem, &parent->kmem);
@@ -6119,20 +6161,12 @@ mem_cgroup_css_alloc(struct cgroup *cont)
 		 * much sense so let cgroup subsystem know about this
 		 * unfortunate state in our controller.
 		 */
-		if (parent && parent != root_mem_cgroup)
+		if (parent != root_mem_cgroup)
 			mem_cgroup_subsys.broken_hierarchy = true;
 	}
-	memcg->last_scanned_node = MAX_NUMNODES;
-	INIT_LIST_HEAD(&memcg->oom_notify);
-
-	if (parent)
-		memcg->swappiness = mem_cgroup_swappiness(parent);
-	atomic_set(&memcg->refcnt, 1);
-	memcg->move_charge_at_immigrate = 0;
-	mutex_init(&memcg->thresholds_lock);
-	spin_lock_init(&memcg->move_lock);
 
 	error = memcg_init_kmem(memcg, &mem_cgroup_subsys);
+	mutex_unlock(&memcg_create_mutex);
 	if (error) {
 		/*
 		 * We call put now because our (and parent's) refcnts
@@ -6140,12 +6174,10 @@ mem_cgroup_css_alloc(struct cgroup *cont)
 		 * call __mem_cgroup_free, so return directly
 		 */
 		mem_cgroup_put(memcg);
-		return ERR_PTR(error);
+		if (parent->use_hierarchy)
+			mem_cgroup_put(parent);
 	}
-	return &memcg->css;
-free_out:
-	__mem_cgroup_free(memcg);
-	return ERR_PTR(error);
+	return error;
 }
 
 static void mem_cgroup_css_offline(struct cgroup *cont)
@@ -6281,7 +6313,7 @@ static struct page *mc_handle_swap_pte(struct vm_area_struct *vma,
 	 * Because lookup_swap_cache() updates some statistics counter,
 	 * we call find_get_page() with swapper_space directly.
 	 */
-	page = find_get_page(&swapper_space, ent.val);
+	page = find_get_page(swap_address_space(ent), ent.val);
 	if (do_swap_account)
 		entry->val = ent.val;
 
@@ -6322,7 +6354,7 @@ static struct page *mc_handle_file_pte(struct vm_area_struct *vma,
 		swp_entry_t swap = radix_to_swp_entry(page);
 		if (do_swap_account)
 			*entry = swap;
-		page = find_get_page(&swapper_space, swap.val);
+		page = find_get_page(swap_address_space(swap), swap.val);
 	}
 #endif
 	return page;
@@ -6532,8 +6564,15 @@ static int mem_cgroup_can_attach(struct cgroup *cgroup,
 	struct task_struct *p = cgroup_taskset_first(tset);
 	int ret = 0;
 	struct mem_cgroup *memcg = mem_cgroup_from_cont(cgroup);
+	unsigned long move_charge_at_immigrate;
 
-	if (memcg->move_charge_at_immigrate) {
+	/*
+	 * We are now commited to this value whatever it is. Changes in this
+	 * tunable will only affect upcoming migrations, not the current one.
+	 * So we need to save it, and keep it going.
+	 */
+	move_charge_at_immigrate  = memcg->move_charge_at_immigrate;
+	if (move_charge_at_immigrate) {
 		struct mm_struct *mm;
 		struct mem_cgroup *from = mem_cgroup_from_task(p);
 
@@ -6553,6 +6592,7 @@ static int mem_cgroup_can_attach(struct cgroup *cgroup,
 			spin_lock(&mc.lock);
 			mc.from = from;
 			mc.to = memcg;
+			mc.immigrate_flags = move_charge_at_immigrate;
 			spin_unlock(&mc.lock);
 			/* We set mc.moving_task later */
 
@@ -6747,6 +6787,7 @@ struct cgroup_subsys mem_cgroup_subsys = {
 	.name = "memory",
 	.subsys_id = mem_cgroup_subsys_id,
 	.css_alloc = mem_cgroup_css_alloc,
+	.css_online = mem_cgroup_css_online,
 	.css_offline = mem_cgroup_css_offline,
 	.css_free = mem_cgroup_css_free,
 	.can_attach = mem_cgroup_can_attach,
@@ -6757,19 +6798,6 @@ struct cgroup_subsys mem_cgroup_subsys = {
 	.use_id = 1,
 };
 
-/*
- * The rest of init is performed during ->css_alloc() for root css which
- * happens before initcalls.  hotcpu_notifier() can't be done together as
- * it would introduce circular locking by adding cgroup_lock -> cpu hotplug
- * dependency.  Do it from a subsys_initcall().
- */
-static int __init mem_cgroup_init(void)
-{
-	hotcpu_notifier(memcg_cpu_hotplug_callback, 0);
-	return 0;
-}
-subsys_initcall(mem_cgroup_init);
-
 #ifdef CONFIG_MEMCG_SWAP
 static int __init enable_swap_account(char *s)
 {
@@ -6782,4 +6810,39 @@ static int __init enable_swap_account(char *s)
 }
 __setup("swapaccount=", enable_swap_account);
 
+static void __init memsw_file_init(void)
+{
+	WARN_ON(cgroup_add_cftypes(&mem_cgroup_subsys, memsw_cgroup_files));
+}
+
+static void __init enable_swap_cgroup(void)
+{
+	if (!mem_cgroup_disabled() && really_do_swap_account) {
+		do_swap_account = 1;
+		memsw_file_init();
+	}
+}
+
+#else
+static void __init enable_swap_cgroup(void)
+{
+}
 #endif
+
+/*
+ * subsys_initcall() for memory controller.
+ *
+ * Some parts like hotcpu_notifier() have to be initialized from this context
+ * because of lock dependencies (cgroup_lock -> cpu hotplug) but basically
+ * everything that doesn't depend on a specific mem_cgroup structure should
+ * be initialized from here.
+ */
+static int __init mem_cgroup_init(void)
+{
+	hotcpu_notifier(memcg_cpu_hotplug_callback, 0);
+	enable_swap_cgroup();
+	mem_cgroup_soft_limit_tree_init();
+	memcg_stock_init();
+	return 0;
+}
+subsys_initcall(mem_cgroup_init);