[deliverable/linux.git] / kernel / nsproxy.c

/*
 *  Copyright (C) 2006 IBM Corporation
 *
 *  Author: Serge Hallyn <serue@us.ibm.com>
 *
 *  This program is free software; you can redistribute it and/or
 *  modify it under the terms of the GNU General Public License as
 *  published by the Free Software Foundation, version 2 of the
 *  License.
 *
 *  Jun 2006 - namespaces support
 *             OpenVZ, SWsoft Inc.
 *             Pavel Emelianov <xemul@openvz.org>
 */

#include <linux/slab.h>
#include <linux/export.h>
#include <linux/nsproxy.h>
#include <linux/init_task.h>
#include <linux/mnt_namespace.h>
#include <linux/utsname.h>
#include <linux/pid_namespace.h>
#include <net/net_namespace.h>
#include <linux/ipc_namespace.h>
#include <linux/proc_ns.h>
#include <linux/file.h>
#include <linux/syscalls.h>
#include <linux/cgroup.h>

static struct kmem_cache *nsproxy_cachep;

struct nsproxy init_nsproxy = {
	.count			= ATOMIC_INIT(1),
	.uts_ns			= &init_uts_ns,
#if defined(CONFIG_POSIX_MQUEUE) || defined(CONFIG_SYSVIPC)
	.ipc_ns			= &init_ipc_ns,
#endif
	.mnt_ns			= NULL,
	.pid_ns_for_children	= &init_pid_ns,
#ifdef CONFIG_NET
	.net_ns			= &init_net,
#endif
#ifdef CONFIG_CGROUPS
	.cgroup_ns		= &init_cgroup_ns,
#endif
};

static inline struct nsproxy *create_nsproxy(void)
{
	struct nsproxy *nsproxy;

	nsproxy = kmem_cache_alloc(nsproxy_cachep, GFP_KERNEL);
	if (nsproxy)
		atomic_set(&nsproxy->count, 1);
	return nsproxy;
}

/*
 * Create new nsproxy and all of its the associated namespaces.
 * Return the newly created nsproxy.  Do not attach this to the task,
 * leave it to the caller to do proper locking and attach it to task.
 */
static struct nsproxy *create_new_namespaces(unsigned long flags,
	struct task_struct *tsk, struct user_namespace *user_ns,
	struct fs_struct *new_fs)
{
	struct nsproxy *new_nsp;
	int err;

	new_nsp = create_nsproxy();
	if (!new_nsp)
		return ERR_PTR(-ENOMEM);

	new_nsp->mnt_ns = copy_mnt_ns(flags, tsk->nsproxy->mnt_ns, user_ns, new_fs);
	if (IS_ERR(new_nsp->mnt_ns)) {
		err = PTR_ERR(new_nsp->mnt_ns);
		goto out_ns;
	}

	new_nsp->uts_ns = copy_utsname(flags, user_ns, tsk->nsproxy->uts_ns);
	if (IS_ERR(new_nsp->uts_ns)) {
		err = PTR_ERR(new_nsp->uts_ns);
		goto out_uts;
	}

	new_nsp->ipc_ns = copy_ipcs(flags, user_ns, tsk->nsproxy->ipc_ns);
	if (IS_ERR(new_nsp->ipc_ns)) {
		err = PTR_ERR(new_nsp->ipc_ns);
		goto out_ipc;
	}

	new_nsp->pid_ns_for_children =
		copy_pid_ns(flags, user_ns, tsk->nsproxy->pid_ns_for_children);
	if (IS_ERR(new_nsp->pid_ns_for_children)) {
		err = PTR_ERR(new_nsp->pid_ns_for_children);
		goto out_pid;
	}

	new_nsp->cgroup_ns = copy_cgroup_ns(flags, user_ns,
					    tsk->nsproxy->cgroup_ns);
	if (IS_ERR(new_nsp->cgroup_ns)) {
		err = PTR_ERR(new_nsp->cgroup_ns);
		goto out_cgroup;
	}

	new_nsp->net_ns = copy_net_ns(flags, user_ns, tsk->nsproxy->net_ns);
	if (IS_ERR(new_nsp->net_ns)) {
		err = PTR_ERR(new_nsp->net_ns);
		goto out_net;
	}

	return new_nsp;

out_net:
	put_cgroup_ns(new_nsp->cgroup_ns);
out_cgroup:
	if (new_nsp->pid_ns_for_children)
		put_pid_ns(new_nsp->pid_ns_for_children);
out_pid:
	if (new_nsp->ipc_ns)
		put_ipc_ns(new_nsp->ipc_ns);
out_ipc:
	if (new_nsp->uts_ns)
		put_uts_ns(new_nsp->uts_ns);
out_uts:
	if (new_nsp->mnt_ns)
		put_mnt_ns(new_nsp->mnt_ns);
out_ns:
	kmem_cache_free(nsproxy_cachep, new_nsp);
	return ERR_PTR(err);
}

/*
 * called from clone.  This now handles copy for nsproxy and all
 * namespaces therein.
 */
int copy_namespaces(unsigned long flags, struct task_struct *tsk)
{
	struct nsproxy *old_ns = tsk->nsproxy;
	struct user_namespace *user_ns = task_cred_xxx(tsk, user_ns);
	struct nsproxy *new_ns;

	if (likely(!(flags & (CLONE_NEWNS | CLONE_NEWUTS | CLONE_NEWIPC |
			      CLONE_NEWPID | CLONE_NEWNET |
			      CLONE_NEWCGROUP)))) {
		get_nsproxy(old_ns);
		return 0;
	}

	if (!ns_capable(user_ns, CAP_SYS_ADMIN))
		return -EPERM;

	/*
	 * CLONE_NEWIPC must detach from the undolist: after switching
	 * to a new ipc namespace, the semaphore arrays from the old
	 * namespace are unreachable.  In clone parlance, CLONE_SYSVSEM
	 * means share undolist with parent, so we must forbid using
	 * it along with CLONE_NEWIPC.
	 */
	if ((flags & (CLONE_NEWIPC | CLONE_SYSVSEM)) ==
		(CLONE_NEWIPC | CLONE_SYSVSEM)) 
		return -EINVAL;

	new_ns = create_new_namespaces(flags, tsk, user_ns, tsk->fs);
	if (IS_ERR(new_ns))
		return  PTR_ERR(new_ns);

	tsk->nsproxy = new_ns;
	return 0;
}

void free_nsproxy(struct nsproxy *ns)
{
	if (ns->mnt_ns)
		put_mnt_ns(ns->mnt_ns);
	if (ns->uts_ns)
		put_uts_ns(ns->uts_ns);
	if (ns->ipc_ns)
		put_ipc_ns(ns->ipc_ns);
	if (ns->pid_ns_for_children)
		put_pid_ns(ns->pid_ns_for_children);
	put_cgroup_ns(ns->cgroup_ns);
	put_net(ns->net_ns);
	kmem_cache_free(nsproxy_cachep, ns);
}

/*
 * Called from unshare. Unshare all the namespaces part of nsproxy.
 * On success, returns the new nsproxy.
 */
int unshare_nsproxy_namespaces(unsigned long unshare_flags,
	struct nsproxy **new_nsp, struct cred *new_cred, struct fs_struct *new_fs)
{
	struct user_namespace *user_ns;
	int err = 0;

	if (!(unshare_flags & (CLONE_NEWNS | CLONE_NEWUTS | CLONE_NEWIPC |
			       CLONE_NEWNET | CLONE_NEWPID | CLONE_NEWCGROUP)))
		return 0;

	user_ns = new_cred ? new_cred->user_ns : current_user_ns();
	if (!ns_capable(user_ns, CAP_SYS_ADMIN))
		return -EPERM;

	*new_nsp = create_new_namespaces(unshare_flags, current, user_ns,
					 new_fs ? new_fs : current->fs);
	if (IS_ERR(*new_nsp)) {
		err = PTR_ERR(*new_nsp);
		goto out;
	}

out:
	return err;
}

void switch_task_namespaces(struct task_struct *p, struct nsproxy *new)
{
	struct nsproxy *ns;

	might_sleep();

	task_lock(p);
	ns = p->nsproxy;
	p->nsproxy = new;
	task_unlock(p);

	if (ns && atomic_dec_and_test(&ns->count))
		free_nsproxy(ns);
}

void exit_task_namespaces(struct task_struct *p)
{
	switch_task_namespaces(p, NULL);
}

SYSCALL_DEFINE2(setns, int, fd, int, nstype)
{
	struct task_struct *tsk = current;
	struct nsproxy *new_nsproxy;
	struct file *file;
	struct ns_common *ns;
	int err;

	file = proc_ns_fget(fd);
	if (IS_ERR(file))
		return PTR_ERR(file);

	err = -EINVAL;
	ns = get_proc_ns(file_inode(file));
	if (nstype && (ns->ops->type != nstype))
		goto out;

	new_nsproxy = create_new_namespaces(0, tsk, current_user_ns(), tsk->fs);
	if (IS_ERR(new_nsproxy)) {
		err = PTR_ERR(new_nsproxy);
		goto out;
	}

	err = ns->ops->install(new_nsproxy, ns);
	if (err) {
		free_nsproxy(new_nsproxy);
		goto out;
	}
	switch_task_namespaces(tsk, new_nsproxy);
out:
	fput(file);
	return err;
}

int __init nsproxy_cache_init(void)
{
	nsproxy_cachep = KMEM_CACHE(nsproxy, SLAB_PANIC);
	return 0;
}
Commit	Line	Data
	1	/*
	2	* Copyright (C) 2006 IBM Corporation
	3	*
	4	* Author: Serge Hallyn <serue@us.ibm.com>
	5	*
	6	* This program is free software; you can redistribute it and/or
	7	* modify it under the terms of the GNU General Public License as
	8	* published by the Free Software Foundation, version 2 of the
	9	* License.
	10	*
	11	* Jun 2006 - namespaces support
	12	* OpenVZ, SWsoft Inc.
	13	* Pavel Emelianov <xemul@openvz.org>
	14	*/
	15
	16	#include <linux/slab.h>
	17	#include <linux/export.h>
	18	#include <linux/nsproxy.h>
	19	#include <linux/init_task.h>
	20	#include <linux/mnt_namespace.h>
	21	#include <linux/utsname.h>
	22	#include <linux/pid_namespace.h>
	23	#include <net/net_namespace.h>
	24	#include <linux/ipc_namespace.h>
	25	#include <linux/proc_ns.h>
	26	#include <linux/file.h>
	27	#include <linux/syscalls.h>
	28	#include <linux/cgroup.h>
	29
	30	static struct kmem_cache *nsproxy_cachep;
	31
	32	struct nsproxy init_nsproxy = {
	33	.count = ATOMIC_INIT(1),
	34	.uts_ns = &init_uts_ns,
	35	#if defined(CONFIG_POSIX_MQUEUE) \|\| defined(CONFIG_SYSVIPC)
	36	.ipc_ns = &init_ipc_ns,
	37	#endif
	38	.mnt_ns = NULL,
	39	.pid_ns_for_children = &init_pid_ns,
	40	#ifdef CONFIG_NET
	41	.net_ns = &init_net,
	42	#endif
	43	#ifdef CONFIG_CGROUPS
	44	.cgroup_ns = &init_cgroup_ns,
	45	#endif
	46	};
	47
	48	static inline struct nsproxy *create_nsproxy(void)
	49	{
	50	struct nsproxy *nsproxy;
	51
	52	nsproxy = kmem_cache_alloc(nsproxy_cachep, GFP_KERNEL);
	53	if (nsproxy)
	54	atomic_set(&nsproxy->count, 1);
	55	return nsproxy;
	56	}
	57
	58	/*
	59	* Create new nsproxy and all of its the associated namespaces.
	60	* Return the newly created nsproxy. Do not attach this to the task,
	61	* leave it to the caller to do proper locking and attach it to task.
	62	*/
	63	static struct nsproxy *create_new_namespaces(unsigned long flags,
	64	struct task_struct tsk, struct user_namespace user_ns,
	65	struct fs_struct *new_fs)
	66	{
	67	struct nsproxy *new_nsp;
	68	int err;
	69
	70	new_nsp = create_nsproxy();
	71	if (!new_nsp)
	72	return ERR_PTR(-ENOMEM);
	73
	74	new_nsp->mnt_ns = copy_mnt_ns(flags, tsk->nsproxy->mnt_ns, user_ns, new_fs);
	75	if (IS_ERR(new_nsp->mnt_ns)) {
	76	err = PTR_ERR(new_nsp->mnt_ns);
	77	goto out_ns;
	78	}
	79
	80	new_nsp->uts_ns = copy_utsname(flags, user_ns, tsk->nsproxy->uts_ns);
	81	if (IS_ERR(new_nsp->uts_ns)) {
	82	err = PTR_ERR(new_nsp->uts_ns);
	83	goto out_uts;
	84	}
	85
	86	new_nsp->ipc_ns = copy_ipcs(flags, user_ns, tsk->nsproxy->ipc_ns);
	87	if (IS_ERR(new_nsp->ipc_ns)) {
	88	err = PTR_ERR(new_nsp->ipc_ns);
	89	goto out_ipc;
	90	}
	91
	92	new_nsp->pid_ns_for_children =
	93	copy_pid_ns(flags, user_ns, tsk->nsproxy->pid_ns_for_children);
	94	if (IS_ERR(new_nsp->pid_ns_for_children)) {
	95	err = PTR_ERR(new_nsp->pid_ns_for_children);
	96	goto out_pid;
	97	}
	98
	99	new_nsp->cgroup_ns = copy_cgroup_ns(flags, user_ns,
	100	tsk->nsproxy->cgroup_ns);
	101	if (IS_ERR(new_nsp->cgroup_ns)) {
	102	err = PTR_ERR(new_nsp->cgroup_ns);
	103	goto out_cgroup;
	104	}
	105
	106	new_nsp->net_ns = copy_net_ns(flags, user_ns, tsk->nsproxy->net_ns);
	107	if (IS_ERR(new_nsp->net_ns)) {
	108	err = PTR_ERR(new_nsp->net_ns);
	109	goto out_net;
	110	}
	111
	112	return new_nsp;
	113
	114	out_net:
	115	put_cgroup_ns(new_nsp->cgroup_ns);
	116	out_cgroup:
	117	if (new_nsp->pid_ns_for_children)
	118	put_pid_ns(new_nsp->pid_ns_for_children);
	119	out_pid:
	120	if (new_nsp->ipc_ns)
	121	put_ipc_ns(new_nsp->ipc_ns);
	122	out_ipc:
	123	if (new_nsp->uts_ns)
	124	put_uts_ns(new_nsp->uts_ns);
	125	out_uts:
	126	if (new_nsp->mnt_ns)
	127	put_mnt_ns(new_nsp->mnt_ns);
	128	out_ns:
	129	kmem_cache_free(nsproxy_cachep, new_nsp);
	130	return ERR_PTR(err);
	131	}
	132
	133	/*
	134	* called from clone. This now handles copy for nsproxy and all
	135	* namespaces therein.
	136	*/
	137	int copy_namespaces(unsigned long flags, struct task_struct *tsk)
	138	{
	139	struct nsproxy *old_ns = tsk->nsproxy;
	140	struct user_namespace *user_ns = task_cred_xxx(tsk, user_ns);
	141	struct nsproxy *new_ns;
	142
	143	if (likely(!(flags & (CLONE_NEWNS \| CLONE_NEWUTS \| CLONE_NEWIPC \|
	144	CLONE_NEWPID \| CLONE_NEWNET \|
	145	CLONE_NEWCGROUP)))) {
	146	get_nsproxy(old_ns);
	147	return 0;
	148	}
	149
	150	if (!ns_capable(user_ns, CAP_SYS_ADMIN))
	151	return -EPERM;
	152
	153	/*
	154	* CLONE_NEWIPC must detach from the undolist: after switching
	155	* to a new ipc namespace, the semaphore arrays from the old
	156	* namespace are unreachable. In clone parlance, CLONE_SYSVSEM
	157	* means share undolist with parent, so we must forbid using
	158	* it along with CLONE_NEWIPC.
	159	*/
	160	if ((flags & (CLONE_NEWIPC \| CLONE_SYSVSEM)) ==
	161	(CLONE_NEWIPC \| CLONE_SYSVSEM))
	162	return -EINVAL;
	163
	164	new_ns = create_new_namespaces(flags, tsk, user_ns, tsk->fs);
	165	if (IS_ERR(new_ns))
	166	return PTR_ERR(new_ns);
	167
	168	tsk->nsproxy = new_ns;
	169	return 0;
	170	}
	171
	172	void free_nsproxy(struct nsproxy *ns)
	173	{
	174	if (ns->mnt_ns)
	175	put_mnt_ns(ns->mnt_ns);
	176	if (ns->uts_ns)
	177	put_uts_ns(ns->uts_ns);
	178	if (ns->ipc_ns)
	179	put_ipc_ns(ns->ipc_ns);
	180	if (ns->pid_ns_for_children)
	181	put_pid_ns(ns->pid_ns_for_children);
	182	put_cgroup_ns(ns->cgroup_ns);
	183	put_net(ns->net_ns);
	184	kmem_cache_free(nsproxy_cachep, ns);
	185	}
	186
	187	/*
	188	* Called from unshare. Unshare all the namespaces part of nsproxy.
	189	* On success, returns the new nsproxy.
	190	*/
	191	int unshare_nsproxy_namespaces(unsigned long unshare_flags,
	192	struct nsproxy *new_nsp, struct cred new_cred, struct fs_struct *new_fs)
	193	{
	194	struct user_namespace *user_ns;
	195	int err = 0;
	196
	197	if (!(unshare_flags & (CLONE_NEWNS \| CLONE_NEWUTS \| CLONE_NEWIPC \|
	198	CLONE_NEWNET \| CLONE_NEWPID \| CLONE_NEWCGROUP)))
	199	return 0;
	200
	201	user_ns = new_cred ? new_cred->user_ns : current_user_ns();
	202	if (!ns_capable(user_ns, CAP_SYS_ADMIN))
	203	return -EPERM;
	204
	205	*new_nsp = create_new_namespaces(unshare_flags, current, user_ns,
	206	new_fs ? new_fs : current->fs);
	207	if (IS_ERR(*new_nsp)) {
	208	err = PTR_ERR(*new_nsp);
	209	goto out;
	210	}
	211
	212	out:
	213	return err;
	214	}
	215
	216	void switch_task_namespaces(struct task_struct p, struct nsproxy new)
	217	{
	218	struct nsproxy *ns;
	219
	220	might_sleep();
	221
	222	task_lock(p);
	223	ns = p->nsproxy;
	224	p->nsproxy = new;
	225	task_unlock(p);
	226
	227	if (ns && atomic_dec_and_test(&ns->count))
	228	free_nsproxy(ns);
	229	}
	230
	231	void exit_task_namespaces(struct task_struct *p)
	232	{
	233	switch_task_namespaces(p, NULL);
	234	}
	235
	236	SYSCALL_DEFINE2(setns, int, fd, int, nstype)
	237	{
	238	struct task_struct *tsk = current;
	239	struct nsproxy *new_nsproxy;
	240	struct file *file;
	241	struct ns_common *ns;
	242	int err;
	243
	244	file = proc_ns_fget(fd);
	245	if (IS_ERR(file))
	246	return PTR_ERR(file);
	247
	248	err = -EINVAL;
	249	ns = get_proc_ns(file_inode(file));
	250	if (nstype && (ns->ops->type != nstype))
	251	goto out;
	252
	253	new_nsproxy = create_new_namespaces(0, tsk, current_user_ns(), tsk->fs);
	254	if (IS_ERR(new_nsproxy)) {
	255	err = PTR_ERR(new_nsproxy);
	256	goto out;
	257	}
	258
	259	err = ns->ops->install(new_nsproxy, ns);
	260	if (err) {
	261	free_nsproxy(new_nsproxy);
	262	goto out;
	263	}
	264	switch_task_namespaces(tsk, new_nsproxy);
	265	out:
	266	fput(file);
	267	return err;
	268	}
	269
	270	int __init nsproxy_cache_init(void)
	271	{
	272	nsproxy_cachep = KMEM_CACHE(nsproxy, SLAB_PANIC);
	273	return 0;
	274	}