memcg, cgroup: kill css_id

index 297462b..2bd052d 100644
@@ -161,7 +161,7 @@ struct cgroup_name {
 struct cgroup {
 	unsigned long flags;		/* "unsigned long" so bitops work */
 
-	int id;				/* ida allocated in-hierarchy ID */
+	int id;				/* idr allocated in-hierarchy ID */
 
 	/*
 	 * We link our 'sibling' struct into our parent's 'children'.

@@ -322,7 +322,7 @@ struct cgroupfs_root {
 	unsigned long flags;
 
 	/* IDs for cgroups in this hierarchy */
-	struct ida cgroup_ida;
+	struct idr cgroup_idr;
 
 	/* The path to use for release notifications. */
 	char release_agent_path[PATH_MAX];

index 345fac8..a26e083 100644
@@ -866,8 +866,6 @@ static void cgroup_free_fn(struct work_struct *work)
 	 */
 	dput(cgrp->parent->dentry);
 
-	ida_simple_remove(&cgrp->root->cgroup_ida, cgrp->id);
-
 	/*
 	 * Drop the active superblock reference that we took when we
 	 * created the cgroup. This will free cgrp->root, if we are

@@ -1379,6 +1377,7 @@ static void init_cgroup_root(struct cgroupfs_root *root)
 	cgrp->root = root;
 	RCU_INIT_POINTER(cgrp->name, &root_cgroup_name);
 	init_cgroup_housekeeping(cgrp);
+	idr_init(&root->cgroup_idr);
 }
 
 static int cgroup_init_root_id(struct cgroupfs_root *root, int start, int end)

@@ -1451,7 +1450,6 @@ static struct cgroupfs_root *cgroup_root_from_opts(struct cgroup_sb_opts *opts)
 	 */
 	root->subsys_mask = opts->subsys_mask;
 	root->flags = opts->flags;
-	ida_init(&root->cgroup_ida);
 	if (opts->release_agent)
 		strcpy(root->release_agent_path, opts->release_agent);
 	if (opts->name)

@@ -1467,7 +1465,7 @@ static void cgroup_free_root(struct cgroupfs_root *root)
 		/* hierarhcy ID shoulid already have been released */
 		WARN_ON_ONCE(root->hierarchy_id);
 
-		ida_destroy(&root->cgroup_ida);
+		idr_destroy(&root->cgroup_idr);
 		kfree(root);
 	}
 }

@@ -1582,6 +1580,11 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type,
 		mutex_lock(&cgroup_mutex);
 		mutex_lock(&cgroup_root_mutex);
 
+		root_cgrp->id = idr_alloc(&root->cgroup_idr, root_cgrp,
+					   0, 1, GFP_KERNEL);
+		if (root_cgrp->id < 0)
+			goto unlock_drop;
+
 		/* Check for name clashes with existing mounts */
 		ret = -EBUSY;
 		if (strlen(root->name))

@@ -4273,7 +4276,11 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,
 		goto err_free_cgrp;
 	rcu_assign_pointer(cgrp->name, name);
 
-	cgrp->id = ida_simple_get(&root->cgroup_ida, 1, 0, GFP_KERNEL);
+	/*
+	 * Temporarily set the pointer to NULL, so idr_find() won't return
+	 * a half-baked cgroup.
+	 */
+	cgrp->id = idr_alloc(&root->cgroup_idr, NULL, 1, 0, GFP_KERNEL);
 	if (cgrp->id < 0)
 		goto err_free_name;
 

@@ -4371,6 +4378,8 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,
 		}
 	}
 
+	idr_replace(&root->cgroup_idr, cgrp, cgrp->id);
+
 	err = cgroup_addrm_files(cgrp, NULL, cgroup_base_files, true);
 	if (err)
 		goto err_destroy;

@@ -4397,7 +4406,7 @@ err_free_all:
 	/* Release the reference count that we took on the superblock */
 	deactivate_super(sb);
 err_free_id:
-	ida_simple_remove(&root->cgroup_ida, cgrp->id);
+	idr_remove(&root->cgroup_idr, cgrp->id);
 err_free_name:
 	kfree(rcu_dereference_raw(cgrp->name));
 err_free_cgrp:

@@ -4590,6 +4599,14 @@ static void cgroup_offline_fn(struct work_struct *work)
 	/* delete this cgroup from parent->children */
 	list_del_rcu(&cgrp->sibling);
 
+	/*
+	 * We should remove the cgroup object from idr before its grace
+	 * period starts, so we won't be looking up a cgroup while the
+	 * cgroup is being freed.
+	 */
+	idr_remove(&cgrp->root->cgroup_idr, cgrp->id);
+	cgrp->id = -1;
+
 	dput(d);
 
 	set_bit(CGRP_RELEASABLE, &parent->flags);

@@ -4915,6 +4932,10 @@ int __init cgroup_init(void)
 
 	BUG_ON(cgroup_init_root_id(&cgroup_dummy_root, 0, 1));
 
+	err = idr_alloc(&cgroup_dummy_root.cgroup_idr, cgroup_dummy_top,
+			0, 1, GFP_KERNEL);
+	BUG_ON(err < 0);
+
 	mutex_unlock(&cgroup_root_mutex);
 	mutex_unlock(&cgroup_mutex);
 

index 2bd052d..8c107e9 100644
@@ -161,7 +161,13 @@ struct cgroup_name {
 struct cgroup {
 	unsigned long flags;		/* "unsigned long" so bitops work */
 
-	int id;				/* idr allocated in-hierarchy ID */
+	/*
+	 * idr allocated in-hierarchy ID.
+	 *
+	 * The ID of the root cgroup is always 0, and a new cgroup
+	 * will be assigned with a smallest available ID.
+	 */
+	int id;
 
 	/*
 	 * We link our 'sibling' struct into our parent's 'children'.

index 8c107e9..4ef8452 100644
@@ -720,6 +720,24 @@ static inline struct cgroup* task_cgroup(struct task_struct *task,
 	return task_subsys_state(task, subsys_id)->cgroup;
 }
 
+/**
+ * cgroup_from_id - lookup cgroup by id
+ * @ss: cgroup subsys to be looked into
+ * @id: the cgroup id
+ *
+ * Returns the cgroup if there's valid one with @id, otherwise returns NULL.
+ * Should be called under rcu_read_lock().
+ */
+static inline struct cgroup *cgroup_from_id(struct cgroup_subsys *ss, int id)
+{
+#ifdef CONFIG_PROVE_RCU
+	rcu_lockdep_assert(rcu_read_lock_held() ||
+			   lockdep_is_held(&cgroup_mutex),
+			   "cgroup_from_id() needs proper protection");
+#endif
+	return idr_find(&ss->root->cgroup_idr, id);
+}
+
 struct cgroup *cgroup_next_sibling(struct cgroup *pos);
 
 /**

index d12ca6f..626c426 100644
@@ -1434,7 +1434,7 @@ bool __mem_cgroup_same_or_subtree(const struct mem_cgroup *root_memcg,
 		return true;
 	if (!root_memcg->use_hierarchy || !memcg)
 		return false;
-	return css_is_ancestor(&memcg->css, &root_memcg->css);
+	return cgroup_is_descendant(memcg->css.cgroup, root_memcg->css.cgroup);
 }
 
 static bool mem_cgroup_same_or_subtree(const struct mem_cgroup *root_memcg,

index 626c426..62541f5 100644
@@ -512,6 +512,25 @@ static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
 	return (memcg == root_mem_cgroup);
 }
 
+static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
+{
+	/*
+	 * The ID of the root cgroup is 0, but memcg treat 0 as an
+	 * invalid ID, so we return (cgroup_id + 1).
+	 */
+	return memcg->css.cgroup->id + 1;
+}
+
+static inline struct mem_cgroup *mem_cgroup_from_id(unsigned short id)
+{
+	struct cgroup *cgrp;
+
+	cgrp = cgroup_from_id(&mem_cgroup_subsys, id - 1);
+	if (!cgrp)
+		return NULL;
+	return mem_cgroup_from_cont(cgrp);
+}
+
 /* Writing them here to avoid exposing memcg's inner layout */
 #if defined(CONFIG_INET) && defined(CONFIG_MEMCG_KMEM)
 

@@ -2821,15 +2840,10 @@ static void __mem_cgroup_cancel_local_charge(struct mem_cgroup *memcg,
  */
 static struct mem_cgroup *mem_cgroup_lookup(unsigned short id)
 {
-	struct cgroup_subsys_state *css;
-
 	/* ID 0 is unused ID */
 	if (!id)
 		return NULL;
-	css = css_lookup(&mem_cgroup_subsys, id);
-	if (!css)
-		return NULL;
-	return mem_cgroup_from_css(css);
+	return mem_cgroup_from_id(id);
 }
 
 struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page)

@@ -4328,7 +4342,7 @@ mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent, bool swapout)
 	 * css_get() was called in uncharge().
 	 */
 	if (do_swap_account && swapout && memcg)
-		swap_cgroup_record(ent, css_id(&memcg->css));
+		swap_cgroup_record(ent, mem_cgroup_id(memcg));
 }
 #endif
 

@@ -4380,8 +4394,8 @@ static int mem_cgroup_move_swap_account(swp_entry_t entry,
 {
 	unsigned short old_id, new_id;
 
-	old_id = css_id(&from->css);
-	new_id = css_id(&to->css);
+	old_id = mem_cgroup_id(from);
+	new_id = mem_cgroup_id(to);
 
 	if (swap_cgroup_cmpxchg(entry, old_id, new_id) == old_id) {
 		mem_cgroup_swap_statistics(from, false);

@@ -6542,7 +6556,7 @@ static enum mc_target_type get_mctgt_type(struct vm_area_struct *vma,
 	}
 	/* There is a swap entry and a page doesn't exist or isn't charged */
 	if (ent.val && !ret &&
-			css_id(&mc.from->css) == lookup_swap_cgroup_id(ent)) {
+	    mem_cgroup_id(mc.from) == lookup_swap_cgroup_id(ent)) {
 		ret = MC_TARGET_SWAP;
 		if (target)
 			target->ent = ent;

index 62541f5..ad7d7c9 100644
@@ -512,6 +512,12 @@ static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
 	return (memcg == root_mem_cgroup);
 }
 
+/*
+ * We restrict the id in the range of [1, 65535], so it can fit into
+ * an unsigned short.
+ */
+#define MEM_CGROUP_ID_MAX	USHRT_MAX
+
 static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
 {
 	/*

@@ -6248,6 +6254,9 @@ mem_cgroup_css_alloc(struct cgroup *cont)
 	long error = -ENOMEM;
 	int node;
 
+	if (cont->id > MEM_CGROUP_ID_MAX)
+		return ERR_PTR(-ENOSPC);
+
 	memcg = mem_cgroup_alloc();
 	if (!memcg)
 		return ERR_PTR(error);

index ad7d7c9..6b77ab6 100644
@@ -608,16 +608,11 @@ static void disarm_sock_keys(struct mem_cgroup *memcg)
 #ifdef CONFIG_MEMCG_KMEM
 /*
  * This will be the memcg's index in each cache's ->memcg_params->memcg_caches.
- * There are two main reasons for not using the css_id for this:
- *  1) this works better in sparse environments, where we have a lot of memcgs,
- *     but only a few kmem-limited. Or also, if we have, for instance, 200
- *     memcgs, and none but the 200th is kmem-limited, we'd have to have a
- *     200 entry array for that.
- *
- *  2) In order not to violate the cgroup API, we would like to do all memory
- *     allocation in ->create(). At that point, we haven't yet allocated the
- *     css_id. Having a separate index prevents us from messing with the cgroup
- *     core for this
+ * The main reason for not using cgroup id for this:
+ *  this works better in sparse environments, where we have a lot of memcgs,
+ *  but only a few kmem-limited. Or also, if we have, for instance, 200
+ *  memcgs, and none but the 200th is kmem-limited, we'd have to have a
+ *  200 entry array for that.
  *
  * The current size of the caches array is stored in
  * memcg_limited_groups_array_size.  It will double each time we have to

@@ -632,14 +627,14 @@ int memcg_limited_groups_array_size;
  * cgroups is a reasonable guess. In the future, it could be a parameter or
  * tunable, but that is strictly not necessary.
  *
- * MAX_SIZE should be as large as the number of css_ids. Ideally, we could get
+ * MAX_SIZE should be as large as the number of cgrp_ids. Ideally, we could get
  * this constant directly from cgroup, but it is understandable that this is
  * better kept as an internal representation in cgroup.c. In any case, the
- * css_id space is not getting any smaller, and we don't have to necessarily
+ * cgrp_id space is not getting any smaller, and we don't have to necessarily
  * increase ours as well if it increases.
  */
 #define MEMCG_CACHES_MIN_SIZE 4
-#define MEMCG_CACHES_MAX_SIZE 65535
+#define MEMCG_CACHES_MAX_SIZE MEM_CGROUP_ID_MAX
 
 /*
  * A lot of the calls to the cache allocation functions are expected to be

@@ -6188,7 +6183,6 @@ static void __mem_cgroup_free(struct mem_cgroup *memcg)
 	size_t size = memcg_size();
 
 	mem_cgroup_remove_from_trees(memcg);
-	free_css_id(&mem_cgroup_subsys, &memcg->css);
 
 	for_each_node(node)
 		free_mem_cgroup_per_zone_info(memcg, node);

@@ -6985,7 +6979,6 @@ struct cgroup_subsys mem_cgroup_subsys = {
 	.bind = mem_cgroup_bind,
 	.base_cftypes = mem_cgroup_files,
 	.early_init = 0,
-	.use_id = 1,
 };
 
 #ifdef CONFIG_MEMCG_SWAP

index 4ef8452..c07e175 100644
@@ -597,11 +597,6 @@ struct cgroup_subsys {
 	int subsys_id;
 	int disabled;
 	int early_init;
-	/*
-	 * True if this subsys uses ID. ID is not available before cgroup_init()
-	 * (not available in early_init time.)
-	 */
-	bool use_id;
 
 	/*
 	 * If %false, this subsystem is properly hierarchical -

@@ -627,9 +622,6 @@ struct cgroup_subsys {
 	 */
 	struct cgroupfs_root *root;
 	struct list_head sibling;
-	/* used when use_id == true */
-	struct idr idr;
-	spinlock_t id_lock;
 
 	/* list of cftype_sets */
 	struct list_head cftsets;

@@ -874,35 +866,6 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan);
 int cgroup_attach_task_all(struct task_struct *from, struct task_struct *);
 int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from);
 
-/*
- * CSS ID is ID for cgroup_subsys_state structs under subsys. This only works
- * if cgroup_subsys.use_id == true. It can be used for looking up and scanning.
- * CSS ID is assigned at cgroup allocation (create) automatically
- * and removed when subsys calls free_css_id() function. This is because
- * the lifetime of cgroup_subsys_state is subsys's matter.
- *
- * Looking up and scanning function should be called under rcu_read_lock().
- * Taking cgroup_mutex is not necessary for following calls.
- * But the css returned by this routine can be "not populated yet" or "being
- * destroyed". The caller should check css and cgroup's status.
- */
-
-/*
- * Typically Called at ->destroy(), or somewhere the subsys frees
- * cgroup_subsys_state.
- */
-void free_css_id(struct cgroup_subsys *ss, struct cgroup_subsys_state *css);
-
-/* Find a cgroup_subsys_state which has given ID */
-
-struct cgroup_subsys_state *css_lookup(struct cgroup_subsys *ss, int id);
-
-/* Returns true if root is ancestor of cg */
-bool css_is_ancestor(struct cgroup_subsys_state *cg,
-		     const struct cgroup_subsys_state *root);
-
-/* Get id and depth of css */
-unsigned short css_id(struct cgroup_subsys_state *css);
 struct cgroup_subsys_state *cgroup_css_from_dir(struct file *f, int id);
 
 #else /* !CONFIG_CGROUPS */

318 lines not on this page · raw message