1 files changed, 62 insertions, 32 deletions

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index fb1ecfd2decd..03a1b3f754d6 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -211,6 +211,7 @@ static unsigned long have_free_callback __read_mostly;
 /* Ditto for the can_fork callback. */
 static unsigned long have_canfork_callback __read_mostly;
 
+static struct file_system_type cgroup2_fs_type;
 static struct cftype cgroup_dfl_base_files[];
 static struct cftype cgroup_legacy_base_files[];
 
@@ -236,6 +237,9 @@ static int cgroup_addrm_files(struct cgroup_subsys_state *css,
  */
 static bool cgroup_ssid_enabled(int ssid)
 {
+	if (CGROUP_SUBSYS_COUNT == 0)
+		return false;
+
 	return static_key_enabled(cgroup_subsys_enabled_key[ssid]);
 }
 
@@ -1647,10 +1651,6 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts)
 			all_ss = true;
 			continue;
 		}
-		if (!strcmp(token, "__DEVEL__sane_behavior")) {
-			opts->flags |= CGRP_ROOT_SANE_BEHAVIOR;
-			continue;
-		}
 		if (!strcmp(token, "noprefix")) {
 			opts->flags |= CGRP_ROOT_NOPREFIX;
 			continue;
@@ -1717,15 +1717,6 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts)
 			return -ENOENT;
 	}
 
-	if (opts->flags & CGRP_ROOT_SANE_BEHAVIOR) {
-		pr_warn("sane_behavior: this is still under development and its behaviors will change, proceed at your own risk\n");
-		if (nr_opts != 1) {
-			pr_err("sane_behavior: no other mount options allowed\n");
-			return -EINVAL;
-		}
-		return 0;
-	}
-
 	/*
 	 * If the 'all' option was specified select all the subsystems,
 	 * otherwise if 'none', 'name=' and a subsystem name options were
@@ -2004,6 +1995,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type,
 			 int flags, const char *unused_dev_name,
 			 void *data)
 {
+	bool is_v2 = fs_type == &cgroup2_fs_type;
 	struct super_block *pinned_sb = NULL;
 	struct cgroup_subsys *ss;
 	struct cgroup_root *root;
@@ -2020,6 +2012,17 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type,
 	if (!use_task_css_set_links)
 		cgroup_enable_task_cg_lists();
 
+	if (is_v2) {
+		if (data) {
+			pr_err("cgroup2: unknown option \"%s\"\n", (char *)data);
+			return ERR_PTR(-EINVAL);
+		}
+		cgrp_dfl_root_visible = true;
+		root = &cgrp_dfl_root;
+		cgroup_get(&root->cgrp);
+		goto out_mount;
+	}
+
 	mutex_lock(&cgroup_mutex);
 
 	/* First find the desired set of subsystems */
@@ -2027,15 +2030,6 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type,
 	if (ret)
 		goto out_unlock;
 
-	/* look for a matching existing root */
-	if (opts.flags & CGRP_ROOT_SANE_BEHAVIOR) {
-		cgrp_dfl_root_visible = true;
-		root = &cgrp_dfl_root;
-		cgroup_get(&root->cgrp);
-		ret = 0;
-		goto out_unlock;
-	}
-
 	/*
 	 * Destruction of cgroup root is asynchronous, so subsystems may
 	 * still be dying after the previous unmount.  Let's drain the
@@ -2146,9 +2140,10 @@ out_free:
 
 	if (ret)
 		return ERR_PTR(ret);
-
+out_mount:
 	dentry = kernfs_mount(fs_type, flags, root->kf_root,
-				CGROUP_SUPER_MAGIC, &new_sb);
+			      is_v2 ? CGROUP2_SUPER_MAGIC : CGROUP_SUPER_MAGIC,
+			      &new_sb);
 	if (IS_ERR(dentry) || !new_sb)
 		cgroup_put(&root->cgrp);
 
@@ -2191,6 +2186,12 @@ static struct file_system_type cgroup_fs_type = {
 	.kill_sb = cgroup_kill_sb,
 };
 
+static struct file_system_type cgroup2_fs_type = {
+	.name = "cgroup2",
+	.mount = cgroup_mount,
+	.kill_sb = cgroup_kill_sb,
+};
+
 /**
  * task_cgroup_path - cgroup path of a task in the first cgroup hierarchy
  * @task: target task
@@ -2498,6 +2499,14 @@ static void cgroup_migrate_add_src(struct css_set *src_cset,
 	lockdep_assert_held(&cgroup_mutex);
 	lockdep_assert_held(&css_set_lock);
 
+	/*
+	 * If ->dead, @src_set is associated with one or more dead cgroups
+	 * and doesn't contain any migratable tasks.  Ignore it early so
+	 * that the rest of migration path doesn't get confused by it.
+	 */
+	if (src_cset->dead)
+		return;
+
 	src_cgrp = cset_cgroup_from_root(src_cset, dst_cgrp->root);
 
 	if (!list_empty(&src_cset->mg_preload_node))
@@ -2677,7 +2686,8 @@ static int cgroup_procs_write_permission(struct task_struct *task,
 	 */
 	if (!uid_eq(cred->euid, GLOBAL_ROOT_UID) &&
 	    !uid_eq(cred->euid, tcred->uid) &&
-	    !uid_eq(cred->euid, tcred->suid))
+	    !uid_eq(cred->euid, tcred->suid) &&
+	    !ns_capable(tcred->user_ns, CAP_SYS_RESOURCE))
 		ret = -EACCES;
 
 	if (!ret && cgroup_on_dfl(dst_cgrp)) {
@@ -2713,9 +2723,10 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf,
 				    size_t nbytes, loff_t off, bool threadgroup)
 {
 	struct task_struct *tsk;
+	struct cgroup_subsys *ss;
 	struct cgroup *cgrp;
 	pid_t pid;
-	int ret;
+	int ssid, ret;
 
 	if (kstrtoint(strstrip(buf), 0, &pid) || pid < 0)
 		return -EINVAL;
@@ -2763,8 +2774,10 @@ out_unlock_rcu:
 	rcu_read_unlock();
 out_unlock_threadgroup:
 	percpu_up_write(&cgroup_threadgroup_rwsem);
+	for_each_subsys(ss, ssid)
+		if (ss->post_attach)
+			ss->post_attach();
 	cgroup_kn_unlock(of->kn);
-	cpuset_post_attach_flush();
 	return ret ?: nbytes;
 }
 
@@ -4681,14 +4694,15 @@ static void css_free_work_fn(struct work_struct *work)
 
 	if (ss) {
 		/* css free path */
+		struct cgroup_subsys_state *parent = css->parent;
 		int id = css->id;
 
-		if (css->parent)
-			css_put(css->parent);
-
 		ss->css_free(css);
 		cgroup_idr_remove(&ss->css_idr, id);
 		cgroup_put(cgrp);
+
+		if (parent)
+			css_put(parent);
 	} else {
 		/* cgroup free path */
 		atomic_dec(&cgrp->root->nr_cgrps);
@@ -4781,6 +4795,7 @@ static void init_and_link_css(struct cgroup_subsys_state *css,
 	memset(css, 0, sizeof(*css));
 	css->cgroup = cgrp;
 	css->ss = ss;
+	css->id = -1;
 	INIT_LIST_HEAD(&css->sibling);
 	INIT_LIST_HEAD(&css->children);
 	css->serial_nr = css_serial_nr_next++;
@@ -5131,6 +5146,7 @@ static int cgroup_destroy_locked(struct cgroup *cgrp)
 	__releases(&cgroup_mutex) __acquires(&cgroup_mutex)
 {
 	struct cgroup_subsys_state *css;
+	struct cgrp_cset_link *link;
 	int ssid;
 
 	lockdep_assert_held(&cgroup_mutex);
@@ -5151,11 +5167,18 @@ static int cgroup_destroy_locked(struct cgroup *cgrp)
 		return -EBUSY;
 
 	/*
-	 * Mark @cgrp dead.  This prevents further task migration and child
-	 * creation by disabling cgroup_lock_live_group().
+	 * Mark @cgrp and the associated csets dead.  The former prevents
+	 * further task migration and child creation by disabling
+	 * cgroup_lock_live_group().  The latter makes the csets ignored by
+	 * the migration path.
 	 */
 	cgrp->self.flags &= ~CSS_ONLINE;
 
+	spin_lock_bh(&css_set_lock);
+	list_for_each_entry(link, &cgrp->cset_links, cset_link)
+		link->cset->dead = true;
+	spin_unlock_bh(&css_set_lock);
+
 	/* initiate massacre of all css's */
 	for_each_css(css, ssid, cgrp)
 		kill_css(css);
@@ -5304,6 +5327,12 @@ int __init cgroup_init(void)
 	BUG_ON(cgroup_init_cftypes(NULL, cgroup_dfl_base_files));
 	BUG_ON(cgroup_init_cftypes(NULL, cgroup_legacy_base_files));
 
+	/*
+	 * The latency of the synchronize_sched() is too high for cgroups,
+	 * avoid it at the cost of forcing all readers into the slow path.
+	 */
+	rcu_sync_enter_start(&cgroup_threadgroup_rwsem.rss);
+
 	mutex_lock(&cgroup_mutex);
 
 	/* Add init_css_set to the hash table */
@@ -5359,6 +5388,7 @@ int __init cgroup_init(void)
 
 	WARN_ON(sysfs_create_mount_point(fs_kobj, "cgroup"));
 	WARN_ON(register_filesystem(&cgroup_fs_type));
+	WARN_ON(register_filesystem(&cgroup2_fs_type));
 	WARN_ON(!proc_create("cgroups", 0, NULL, &proc_cgroupstats_operations));
 
 	return 0;