[GitHub/mt8127/android_kernel_alcatel_ttab.git] / kernel / pid_namespace.c

/*
 * Pid namespaces
 *
 * Authors:
 *    (C) 2007 Pavel Emelyanov <xemul@openvz.org>, OpenVZ, SWsoft Inc.
 *    (C) 2007 Sukadev Bhattiprolu <sukadev@us.ibm.com>, IBM
 *     Many thanks to Oleg Nesterov for comments and help
 *
 */

#include <linux/pid.h>
#include <linux/pid_namespace.h>
#include <linux/user_namespace.h>
#include <linux/syscalls.h>
#include <linux/err.h>
#include <linux/acct.h>
#include <linux/slab.h>
#include <linux/proc_ns.h>
#include <linux/reboot.h>
#include <linux/export.h>

struct pid_cache {
	int nr_ids;
	char name[16];
	struct kmem_cache *cachep;
	struct list_head list;
};

static LIST_HEAD(pid_caches_lh);
static DEFINE_MUTEX(pid_caches_mutex);
static struct kmem_cache *pid_ns_cachep;

/*
 * creates the kmem cache to allocate pids from.
 * @nr_ids: the number of numerical ids this pid will have to carry
 */

static struct kmem_cache *create_pid_cachep(int nr_ids)
{
	struct pid_cache *pcache;
	struct kmem_cache *cachep;

	mutex_lock(&pid_caches_mutex);
	list_for_each_entry(pcache, &pid_caches_lh, list)
		if (pcache->nr_ids == nr_ids)
			goto out;

	pcache = kmalloc(sizeof(struct pid_cache), GFP_KERNEL);
	if (pcache == NULL)
		goto err_alloc;

	snprintf(pcache->name, sizeof(pcache->name), "pid_%d", nr_ids);
	cachep = kmem_cache_create(pcache->name,
			sizeof(struct pid) + (nr_ids - 1) * sizeof(struct upid),
			0, SLAB_HWCACHE_ALIGN, NULL);
	if (cachep == NULL)
		goto err_cachep;

	pcache->nr_ids = nr_ids;
	pcache->cachep = cachep;
	list_add(&pcache->list, &pid_caches_lh);
out:
	mutex_unlock(&pid_caches_mutex);
	return pcache->cachep;

err_cachep:
	kfree(pcache);
err_alloc:
	mutex_unlock(&pid_caches_mutex);
	return NULL;
}

static void proc_cleanup_work(struct work_struct *work)
{
	struct pid_namespace *ns = container_of(work, struct pid_namespace, proc_work);
	pid_ns_release_proc(ns);
}

/* MAX_PID_NS_LEVEL is needed for limiting size of 'struct pid' */
#define MAX_PID_NS_LEVEL 32

static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns,
	struct pid_namespace *parent_pid_ns)
{
	struct pid_namespace *ns;
	unsigned int level = parent_pid_ns->level + 1;
	int i;
	int err;

	if (level > MAX_PID_NS_LEVEL) {
		err = -EINVAL;
		goto out;
	}

	err = -ENOMEM;
	ns = kmem_cache_zalloc(pid_ns_cachep, GFP_KERNEL);
	if (ns == NULL)
		goto out;

	ns->pidmap[0].page = kzalloc(PAGE_SIZE, GFP_KERNEL);
	if (!ns->pidmap[0].page)
		goto out_free;

	ns->pid_cachep = create_pid_cachep(level + 1);
	if (ns->pid_cachep == NULL)
		goto out_free_map;

	err = proc_alloc_inum(&ns->proc_inum);
	if (err)
		goto out_free_map;

	kref_init(&ns->kref);
	ns->level = level;
	ns->parent = get_pid_ns(parent_pid_ns);
	ns->user_ns = get_user_ns(user_ns);
	ns->nr_hashed = PIDNS_HASH_ADDING;
	INIT_WORK(&ns->proc_work, proc_cleanup_work);

	set_bit(0, ns->pidmap[0].page);
	atomic_set(&ns->pidmap[0].nr_free, BITS_PER_PAGE - 1);

	for (i = 1; i < PIDMAP_ENTRIES; i++)
		atomic_set(&ns->pidmap[i].nr_free, BITS_PER_PAGE);

	return ns;

out_free_map:
	kfree(ns->pidmap[0].page);
out_free:
	kmem_cache_free(pid_ns_cachep, ns);
out:
	return ERR_PTR(err);
}

static void destroy_pid_namespace(struct pid_namespace *ns)
{
	int i;

	proc_free_inum(ns->proc_inum);
	for (i = 0; i < PIDMAP_ENTRIES; i++)
		kfree(ns->pidmap[i].page);
	put_user_ns(ns->user_ns);
	kmem_cache_free(pid_ns_cachep, ns);
}

struct pid_namespace *copy_pid_ns(unsigned long flags,
	struct user_namespace *user_ns, struct pid_namespace *old_ns)
{
	if (!(flags & CLONE_NEWPID))
		return get_pid_ns(old_ns);
	if (task_active_pid_ns(current) != old_ns)
		return ERR_PTR(-EINVAL);
	return create_pid_namespace(user_ns, old_ns);
}

static void free_pid_ns(struct kref *kref)
{
	struct pid_namespace *ns;

	ns = container_of(kref, struct pid_namespace, kref);
	destroy_pid_namespace(ns);
}

void put_pid_ns(struct pid_namespace *ns)
{
	struct pid_namespace *parent;

	while (ns != &init_pid_ns) {
		parent = ns->parent;
		if (!kref_put(&ns->kref, free_pid_ns))
			break;
		ns = parent;
	}
}
EXPORT_SYMBOL_GPL(put_pid_ns);

void zap_pid_ns_processes(struct pid_namespace *pid_ns)
{
	int nr;
	int rc;
	struct task_struct *task, *me = current;
	int init_pids = thread_group_leader(me) ? 1 : 2;

	/* Don't allow any more processes into the pid namespace */
	disable_pid_allocation(pid_ns);

	/* Ignore SIGCHLD causing any terminated children to autoreap */
	spin_lock_irq(&me->sighand->siglock);
	me->sighand->action[SIGCHLD - 1].sa.sa_handler = SIG_IGN;
	spin_unlock_irq(&me->sighand->siglock);

	/*
	 * The last thread in the cgroup-init thread group is terminating.
	 * Find remaining pid_ts in the namespace, signal and wait for them
	 * to exit.
	 *
	 * Note:  This signals each threads in the namespace - even those that
	 * 	  belong to the same thread group, To avoid this, we would have
	 * 	  to walk the entire tasklist looking a processes in this
	 * 	  namespace, but that could be unnecessarily expensive if the
	 * 	  pid namespace has just a few processes. Or we need to
	 * 	  maintain a tasklist for each pid namespace.
	 *
	 */
	read_lock(&tasklist_lock);
	nr = next_pidmap(pid_ns, 1);
	while (nr > 0) {
		rcu_read_lock();

		task = pid_task(find_vpid(nr), PIDTYPE_PID);
		if (task && !__fatal_signal_pending(task))
			send_sig_info(SIGKILL, SEND_SIG_FORCED, task);

		rcu_read_unlock();

		nr = next_pidmap(pid_ns, nr);
	}
	read_unlock(&tasklist_lock);

	/* Firstly reap the EXIT_ZOMBIE children we may have. */
	do {
		clear_thread_flag(TIF_SIGPENDING);
		rc = sys_wait4(-1, NULL, __WALL, NULL);
	} while (rc != -ECHILD);

	/*
	 * sys_wait4() above can't reap the TASK_DEAD children.
	 * Make sure they all go away, see free_pid().
	 */
	for (;;) {
		set_current_state(TASK_UNINTERRUPTIBLE);
		if (pid_ns->nr_hashed == init_pids)
			break;
		schedule();
	}
	__set_current_state(TASK_RUNNING);

	if (pid_ns->reboot)
		current->signal->group_exit_code = pid_ns->reboot;

	acct_exit_ns(pid_ns);
	return;
}

#ifdef CONFIG_CHECKPOINT_RESTORE
static int pid_ns_ctl_handler(struct ctl_table *table, int write,
		void __user *buffer, size_t *lenp, loff_t *ppos)
{
	struct pid_namespace *pid_ns = task_active_pid_ns(current);
	struct ctl_table tmp = *table;

	if (write && !ns_capable(pid_ns->user_ns, CAP_SYS_ADMIN))
		return -EPERM;

	/*
	 * Writing directly to ns' last_pid field is OK, since this field
	 * is volatile in a living namespace anyway and a code writing to
	 * it should synchronize its usage with external means.
	 */

	tmp.data = &pid_ns->last_pid;
	return proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);
}

extern int pid_max;
static int zero = 0;
static struct ctl_table pid_ns_ctl_table[] = {
	{
		.procname = "ns_last_pid",
		.maxlen = sizeof(int),
		.mode = 0666, /* permissions are checked in the handler */
		.proc_handler = pid_ns_ctl_handler,
		.extra1 = &zero,
		.extra2 = &pid_max,
	},
	{ }
};
static struct ctl_path kern_path[] = { { .procname = "kernel", }, { } };
#endif	/* CONFIG_CHECKPOINT_RESTORE */

int reboot_pid_ns(struct pid_namespace *pid_ns, int cmd)
{
	if (pid_ns == &init_pid_ns)
		return 0;

	switch (cmd) {
	case LINUX_REBOOT_CMD_RESTART2:
	case LINUX_REBOOT_CMD_RESTART:
		pid_ns->reboot = SIGHUP;
		break;

	case LINUX_REBOOT_CMD_POWER_OFF:
	case LINUX_REBOOT_CMD_HALT:
		pid_ns->reboot = SIGINT;
		break;
	default:
		return -EINVAL;
	}

	read_lock(&tasklist_lock);
	force_sig(SIGKILL, pid_ns->child_reaper);
	read_unlock(&tasklist_lock);

	do_exit(0);

	/* Not reached */
	return 0;
}

static void *pidns_get(struct task_struct *task)
{
	struct pid_namespace *ns;

	rcu_read_lock();
	ns = task_active_pid_ns(task);
	if (ns)
		get_pid_ns(ns);
	rcu_read_unlock();

	return ns;
}

static void pidns_put(void *ns)
{
	put_pid_ns(ns);
}

static int pidns_install(struct nsproxy *nsproxy, void *ns)
{
	struct pid_namespace *active = task_active_pid_ns(current);
	struct pid_namespace *ancestor, *new = ns;

	if (!ns_capable(new->user_ns, CAP_SYS_ADMIN) ||
	    !nsown_capable(CAP_SYS_ADMIN))
		return -EPERM;

	/*
	 * Only allow entering the current active pid namespace
	 * or a child of the current active pid namespace.
	 *
	 * This is required for fork to return a usable pid value and
	 * this maintains the property that processes and their
	 * children can not escape their current pid namespace.
	 */
	if (new->level < active->level)
		return -EINVAL;

	ancestor = new;
	while (ancestor->level > active->level)
		ancestor = ancestor->parent;
	if (ancestor != active)
		return -EINVAL;

	put_pid_ns(nsproxy->pid_ns);
	nsproxy->pid_ns = get_pid_ns(new);
	return 0;
}

static unsigned int pidns_inum(void *ns)
{
	struct pid_namespace *pid_ns = ns;
	return pid_ns->proc_inum;
}

const struct proc_ns_operations pidns_operations = {
	.name		= "pid",
	.type		= CLONE_NEWPID,
	.get		= pidns_get,
	.put		= pidns_put,
	.install	= pidns_install,
	.inum		= pidns_inum,
};

static __init int pid_namespaces_init(void)
{
	pid_ns_cachep = KMEM_CACHE(pid_namespace, SLAB_PANIC);

#ifdef CONFIG_CHECKPOINT_RESTORE
	register_sysctl_paths(kern_path, pid_ns_ctl_table);
#endif
	return 0;
}

__initcall(pid_namespaces_init);
Commit	Line	Data
74bd59bb PE	1	/*
	2	* Pid namespaces
	3	*
	4	* Authors:
	5	* (C) 2007 Pavel Emelyanov <xemul@openvz.org>, OpenVZ, SWsoft Inc.
	6	* (C) 2007 Sukadev Bhattiprolu <sukadev@us.ibm.com>, IBM
	7	* Many thanks to Oleg Nesterov for comments and help
	8	*
	9	*/
	10
	11	#include <linux/pid.h>
	12	#include <linux/pid_namespace.h>
49f4d8b9	13	#include <linux/user_namespace.h>
74bd59bb PE	14	#include <linux/syscalls.h>
74bd59bb PE	15	#include <linux/err.h>
0b6b030f	16	#include <linux/acct.h>
5a0e3ad6	17	#include <linux/slab.h>
0bb80f24	18	#include <linux/proc_ns.h>
cf3f8921	19	#include <linux/reboot.h>
523a6a94	20	#include <linux/export.h>
74bd59bb	21
74bd59bb PE	22	struct pid_cache {
	23	int nr_ids;
	24	char name[16];
	25	struct kmem_cache *cachep;
	26	struct list_head list;
	27	};
	28
	29	static LIST_HEAD(pid_caches_lh);
	30	static DEFINE_MUTEX(pid_caches_mutex);
	31	static struct kmem_cache *pid_ns_cachep;
	32
	33	/*
	34	* creates the kmem cache to allocate pids from.
	35	* @nr_ids: the number of numerical ids this pid will have to carry
	36	*/
	37
	38	static struct kmem_cache *create_pid_cachep(int nr_ids)
	39	{
	40	struct pid_cache *pcache;
	41	struct kmem_cache *cachep;
	42
	43	mutex_lock(&pid_caches_mutex);
	44	list_for_each_entry(pcache, &pid_caches_lh, list)
	45	if (pcache->nr_ids == nr_ids)
	46	goto out;
	47
	48	pcache = kmalloc(sizeof(struct pid_cache), GFP_KERNEL);
	49	if (pcache == NULL)
	50	goto err_alloc;
	51
	52	snprintf(pcache->name, sizeof(pcache->name), "pid_%d", nr_ids);
	53	cachep = kmem_cache_create(pcache->name,
	54	sizeof(struct pid) + (nr_ids - 1) * sizeof(struct upid),
	55	0, SLAB_HWCACHE_ALIGN, NULL);
	56	if (cachep == NULL)
	57	goto err_cachep;
	58
	59	pcache->nr_ids = nr_ids;
	60	pcache->cachep = cachep;
	61	list_add(&pcache->list, &pid_caches_lh);
	62	out:
	63	mutex_unlock(&pid_caches_mutex);
	64	return pcache->cachep;
	65
	66	err_cachep:
	67	kfree(pcache);
	68	err_alloc:
	69	mutex_unlock(&pid_caches_mutex);
	70	return NULL;
	71	}
	72
0a01f2cc EB	73	static void proc_cleanup_work(struct work_struct *work)
	74	{
	75	struct pid_namespace *ns = container_of(work, struct pid_namespace, proc_work);
	76	pid_ns_release_proc(ns);
	77	}
	78
f2302505 AV	79	/* MAX_PID_NS_LEVEL is needed for limiting size of 'struct pid' */
	80	#define MAX_PID_NS_LEVEL 32
	81
49f4d8b9 EB	82	static struct pid_namespace create_pid_namespace(struct user_namespace user_ns,
49f4d8b9 EB	83	struct pid_namespace *parent_pid_ns)
74bd59bb PE	84	{
74bd59bb PE	85	struct pid_namespace *ns;
ed469a63	86	unsigned int level = parent_pid_ns->level + 1;
f2302505 AV	87	int i;
	88	int err;
	89
	90	if (level > MAX_PID_NS_LEVEL) {
	91	err = -EINVAL;
	92	goto out;
	93	}
74bd59bb	94
f2302505	95	err = -ENOMEM;
84406c15	96	ns = kmem_cache_zalloc(pid_ns_cachep, GFP_KERNEL);
74bd59bb PE	97	if (ns == NULL)
	98	goto out;
	99
	100	ns->pidmap[0].page = kzalloc(PAGE_SIZE, GFP_KERNEL);
	101	if (!ns->pidmap[0].page)
	102	goto out_free;
	103
	104	ns->pid_cachep = create_pid_cachep(level + 1);
	105	if (ns->pid_cachep == NULL)
	106	goto out_free_map;
	107
98f842e6 EB	108	err = proc_alloc_inum(&ns->proc_inum);
	109	if (err)
	110	goto out_free_map;
	111
74bd59bb	112	kref_init(&ns->kref);
74bd59bb	113	ns->level = level;
ed469a63	114	ns->parent = get_pid_ns(parent_pid_ns);
49f4d8b9	115	ns->user_ns = get_user_ns(user_ns);
c876ad76	116	ns->nr_hashed = PIDNS_HASH_ADDING;
0a01f2cc	117	INIT_WORK(&ns->proc_work, proc_cleanup_work);
74bd59bb PE	118
	119	set_bit(0, ns->pidmap[0].page);
	120	atomic_set(&ns->pidmap[0].nr_free, BITS_PER_PAGE - 1);
	121
84406c15	122	for (i = 1; i < PIDMAP_ENTRIES; i++)
74bd59bb	123	atomic_set(&ns->pidmap[i].nr_free, BITS_PER_PAGE);
74bd59bb PE	124
	125	return ns;
	126
	127	out_free_map:
	128	kfree(ns->pidmap[0].page);
	129	out_free:
	130	kmem_cache_free(pid_ns_cachep, ns);
	131	out:
4308eebb	132	return ERR_PTR(err);
74bd59bb PE	133	}
	134
	135	static void destroy_pid_namespace(struct pid_namespace *ns)
	136	{
	137	int i;
	138
98f842e6	139	proc_free_inum(ns->proc_inum);
74bd59bb PE	140	for (i = 0; i < PIDMAP_ENTRIES; i++)
74bd59bb PE	141	kfree(ns->pidmap[i].page);
49f4d8b9	142	put_user_ns(ns->user_ns);
74bd59bb PE	143	kmem_cache_free(pid_ns_cachep, ns);
	144	}
	145
49f4d8b9 EB	146	struct pid_namespace *copy_pid_ns(unsigned long flags,
49f4d8b9 EB	147	struct user_namespace user_ns, struct pid_namespace old_ns)
74bd59bb	148	{
74bd59bb	149	if (!(flags & CLONE_NEWPID))
dca4a979	150	return get_pid_ns(old_ns);
225778d6 EB	151	if (task_active_pid_ns(current) != old_ns)
225778d6 EB	152	return ERR_PTR(-EINVAL);
49f4d8b9	153	return create_pid_namespace(user_ns, old_ns);
74bd59bb PE	154	}
74bd59bb PE	155
bbc2e3ef	156	static void free_pid_ns(struct kref *kref)
74bd59bb	157	{
bbc2e3ef	158	struct pid_namespace *ns;
74bd59bb PE	159
74bd59bb PE	160	ns = container_of(kref, struct pid_namespace, kref);
74bd59bb	161	destroy_pid_namespace(ns);
bbc2e3ef	162	}
74bd59bb	163
bbc2e3ef CG	164	void put_pid_ns(struct pid_namespace *ns)
	165	{
	166	struct pid_namespace *parent;
	167
	168	while (ns != &init_pid_ns) {
	169	parent = ns->parent;
	170	if (!kref_put(&ns->kref, free_pid_ns))
	171	break;
	172	ns = parent;
	173	}
74bd59bb	174	}
bbc2e3ef	175	EXPORT_SYMBOL_GPL(put_pid_ns);
74bd59bb PE	176
	177	void zap_pid_ns_processes(struct pid_namespace *pid_ns)
	178	{
	179	int nr;
	180	int rc;
00c10bc1	181	struct task_struct task, me = current;
751c644b	182	int init_pids = thread_group_leader(me) ? 1 : 2;
00c10bc1	183
c876ad76 EB	184	/* Don't allow any more processes into the pid namespace */
	185	disable_pid_allocation(pid_ns);
	186
00c10bc1 EB	187	/* Ignore SIGCHLD causing any terminated children to autoreap */
	188	spin_lock_irq(&me->sighand->siglock);
	189	me->sighand->action[SIGCHLD - 1].sa.sa_handler = SIG_IGN;
	190	spin_unlock_irq(&me->sighand->siglock);
74bd59bb PE	191
	192	/*
	193	* The last thread in the cgroup-init thread group is terminating.
	194	* Find remaining pid_ts in the namespace, signal and wait for them
	195	* to exit.
	196	*
	197	* Note: This signals each threads in the namespace - even those that
	198	* belong to the same thread group, To avoid this, we would have
	199	* to walk the entire tasklist looking a processes in this
	200	* namespace, but that could be unnecessarily expensive if the
	201	* pid namespace has just a few processes. Or we need to
	202	* maintain a tasklist for each pid namespace.
	203	*
	204	*/
	205	read_lock(&tasklist_lock);
	206	nr = next_pidmap(pid_ns, 1);
	207	while (nr > 0) {
e4da026f SB	208	rcu_read_lock();
e4da026f SB	209
e4da026f	210	task = pid_task(find_vpid(nr), PIDTYPE_PID);
a02d6fd6 ON	211	if (task && !__fatal_signal_pending(task))
a02d6fd6 ON	212	send_sig_info(SIGKILL, SEND_SIG_FORCED, task);
e4da026f SB	213
	214	rcu_read_unlock();
	215
74bd59bb PE	216	nr = next_pidmap(pid_ns, nr);
	217	}
	218	read_unlock(&tasklist_lock);
	219
6347e900	220	/* Firstly reap the EXIT_ZOMBIE children we may have. */
74bd59bb PE	221	do {
	222	clear_thread_flag(TIF_SIGPENDING);
	223	rc = sys_wait4(-1, NULL, __WALL, NULL);
	224	} while (rc != -ECHILD);
	225
6347e900 EB	226	/*
6347e900 EB	227	* sys_wait4() above can't reap the TASK_DEAD children.
af4b8a83	228	* Make sure they all go away, see free_pid().
6347e900 EB	229	*/
6347e900 EB	230	for (;;) {
af4b8a83	231	set_current_state(TASK_UNINTERRUPTIBLE);
751c644b	232	if (pid_ns->nr_hashed == init_pids)
6347e900 EB	233	break;
	234	schedule();
	235	}
af4b8a83	236	__set_current_state(TASK_RUNNING);
6347e900	237
cf3f8921 DL	238	if (pid_ns->reboot)
	239	current->signal->group_exit_code = pid_ns->reboot;
	240
0b6b030f	241	acct_exit_ns(pid_ns);
74bd59bb PE	242	return;
	243	}
	244
98ed57ee	245	#ifdef CONFIG_CHECKPOINT_RESTORE
b8f566b0 PE	246	static int pid_ns_ctl_handler(struct ctl_table *table, int write,
	247	void __user buffer, size_t lenp, loff_t *ppos)
	248	{
49f4d8b9	249	struct pid_namespace *pid_ns = task_active_pid_ns(current);
b8f566b0 PE	250	struct ctl_table tmp = *table;
b8f566b0 PE	251
49f4d8b9	252	if (write && !ns_capable(pid_ns->user_ns, CAP_SYS_ADMIN))
b8f566b0 PE	253	return -EPERM;
	254
	255	/*
	256	* Writing directly to ns' last_pid field is OK, since this field
	257	* is volatile in a living namespace anyway and a code writing to
	258	* it should synchronize its usage with external means.
	259	*/
	260
49f4d8b9	261	tmp.data = &pid_ns->last_pid;
579035dc	262	return proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);
b8f566b0 PE	263	}
b8f566b0 PE	264
579035dc AV	265	extern int pid_max;
579035dc AV	266	static int zero = 0;
b8f566b0 PE	267	static struct ctl_table pid_ns_ctl_table[] = {
	268	{
	269	.procname = "ns_last_pid",
	270	.maxlen = sizeof(int),
	271	.mode = 0666, /* permissions are checked in the handler */
	272	.proc_handler = pid_ns_ctl_handler,
579035dc AV	273	.extra1 = &zero,
579035dc AV	274	.extra2 = &pid_max,
b8f566b0 PE	275	},
	276	{ }
	277	};
b8f566b0	278	static struct ctl_path kern_path[] = { { .procname = "kernel", }, { } };
98ed57ee	279	#endif /* CONFIG_CHECKPOINT_RESTORE */
b8f566b0	280
cf3f8921 DL	281	int reboot_pid_ns(struct pid_namespace *pid_ns, int cmd)
	282	{
	283	if (pid_ns == &init_pid_ns)
	284	return 0;
	285
	286	switch (cmd) {
	287	case LINUX_REBOOT_CMD_RESTART2:
	288	case LINUX_REBOOT_CMD_RESTART:
	289	pid_ns->reboot = SIGHUP;
	290	break;
	291
	292	case LINUX_REBOOT_CMD_POWER_OFF:
	293	case LINUX_REBOOT_CMD_HALT:
	294	pid_ns->reboot = SIGINT;
	295	break;
	296	default:
	297	return -EINVAL;
	298	}
	299
	300	read_lock(&tasklist_lock);
	301	force_sig(SIGKILL, pid_ns->child_reaper);
	302	read_unlock(&tasklist_lock);
	303
	304	do_exit(0);
	305
	306	/* Not reached */
	307	return 0;
	308	}
	309
57e8391d EB	310	static void pidns_get(struct task_struct task)
	311	{
	312	struct pid_namespace *ns;
	313
	314	rcu_read_lock();
a15c36b2 ON	315	ns = task_active_pid_ns(task);
	316	if (ns)
	317	get_pid_ns(ns);
57e8391d EB	318	rcu_read_unlock();
	319
	320	return ns;
	321	}
	322
	323	static void pidns_put(void *ns)
	324	{
	325	put_pid_ns(ns);
	326	}
	327
	328	static int pidns_install(struct nsproxy nsproxy, void ns)
	329	{
	330	struct pid_namespace *active = task_active_pid_ns(current);
	331	struct pid_namespace ancestor, new = ns;
	332
5e4a0847 EB	333	if (!ns_capable(new->user_ns, CAP_SYS_ADMIN) \|\|
5e4a0847 EB	334	!nsown_capable(CAP_SYS_ADMIN))
57e8391d EB	335	return -EPERM;
	336
	337	/*
	338	* Only allow entering the current active pid namespace
	339	* or a child of the current active pid namespace.
	340	*
	341	* This is required for fork to return a usable pid value and
	342	* this maintains the property that processes and their
	343	* children can not escape their current pid namespace.
	344	*/
	345	if (new->level < active->level)
	346	return -EINVAL;
	347
	348	ancestor = new;
	349	while (ancestor->level > active->level)
	350	ancestor = ancestor->parent;
	351	if (ancestor != active)
	352	return -EINVAL;
	353
	354	put_pid_ns(nsproxy->pid_ns);
	355	nsproxy->pid_ns = get_pid_ns(new);
	356	return 0;
	357	}
	358
98f842e6 EB	359	static unsigned int pidns_inum(void *ns)
	360	{
	361	struct pid_namespace *pid_ns = ns;
	362	return pid_ns->proc_inum;
	363	}
	364
57e8391d EB	365	const struct proc_ns_operations pidns_operations = {
	366	.name = "pid",
	367	.type = CLONE_NEWPID,
	368	.get = pidns_get,
	369	.put = pidns_put,
	370	.install = pidns_install,
98f842e6	371	.inum = pidns_inum,
57e8391d EB	372	};
57e8391d EB	373
74bd59bb PE	374	static __init int pid_namespaces_init(void)
	375	{
	376	pid_ns_cachep = KMEM_CACHE(pid_namespace, SLAB_PANIC);
98ed57ee CG	377
98ed57ee CG	378	#ifdef CONFIG_CHECKPOINT_RESTORE
b8f566b0	379	register_sysctl_paths(kern_path, pid_ns_ctl_table);
98ed57ee	380	#endif
74bd59bb PE	381	return 0;
	382	}
	383
	384	__initcall(pid_namespaces_init);