nsproxy.c 6.22 KB
Newer Older
1 2 3 4 5 6 7 8 9
/*
 *  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.
10 11 12 13
 *
 *  Jun 2006 - namespaces support
 *             OpenVZ, SWsoft Inc.
 *             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>
Cedric Le Goater's avatar
Cedric Le Goater committed
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 27
#include <linux/file.h>
#include <linux/syscalls.h>
28

29 30
static struct kmem_cache *nsproxy_cachep;

31 32 33 34 35 36 37 38 39 40 41 42
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	= &init_pid_ns,
#ifdef CONFIG_NET
	.net_ns	= &init_net,
#endif
};
43

44
static inline struct nsproxy *create_nsproxy(void)
45
{
46
	struct nsproxy *nsproxy;
47

48 49 50 51
	nsproxy = kmem_cache_alloc(nsproxy_cachep, GFP_KERNEL);
	if (nsproxy)
		atomic_set(&nsproxy->count, 1);
	return nsproxy;
52 53 54
}

/*
55 56 57
 * 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.
58
 */
59
static struct nsproxy *create_new_namespaces(unsigned long flags,
60 61
	struct task_struct *tsk, struct user_namespace *user_ns,
	struct fs_struct *new_fs)
62
{
63
	struct nsproxy *new_nsp;
64
	int err;
65

66
	new_nsp = create_nsproxy();
67 68
	if (!new_nsp)
		return ERR_PTR(-ENOMEM);
69

70
	new_nsp->mnt_ns = copy_mnt_ns(flags, tsk->nsproxy->mnt_ns, user_ns, new_fs);
71 72
	if (IS_ERR(new_nsp->mnt_ns)) {
		err = PTR_ERR(new_nsp->mnt_ns);
73
		goto out_ns;
74
	}
75

76
	new_nsp->uts_ns = copy_utsname(flags, user_ns, tsk->nsproxy->uts_ns);
77 78
	if (IS_ERR(new_nsp->uts_ns)) {
		err = PTR_ERR(new_nsp->uts_ns);
79
		goto out_uts;
80
	}
81

82
	new_nsp->ipc_ns = copy_ipcs(flags, user_ns, tsk->nsproxy->ipc_ns);
83 84
	if (IS_ERR(new_nsp->ipc_ns)) {
		err = PTR_ERR(new_nsp->ipc_ns);
85
		goto out_ipc;
86
	}
87

88
	new_nsp->pid_ns = copy_pid_ns(flags, user_ns, tsk->nsproxy->pid_ns);
89 90
	if (IS_ERR(new_nsp->pid_ns)) {
		err = PTR_ERR(new_nsp->pid_ns);
91
		goto out_pid;
92
	}
93

94
	new_nsp->net_ns = copy_net_ns(flags, user_ns, tsk->nsproxy->net_ns);
95 96 97 98 99
	if (IS_ERR(new_nsp->net_ns)) {
		err = PTR_ERR(new_nsp->net_ns);
		goto out_net;
	}

100 101
	return new_nsp;

102
out_net:
103 104
	if (new_nsp->pid_ns)
		put_pid_ns(new_nsp->pid_ns);
105 106 107 108 109 110 111 112 113 114
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:
115
	kmem_cache_free(nsproxy_cachep, new_nsp);
116
	return ERR_PTR(err);
117 118 119 120 121 122
}

/*
 * called from clone.  This now handles copy for nsproxy and all
 * namespaces therein.
 */
123
int copy_namespaces(unsigned long flags, struct task_struct *tsk)
124 125
{
	struct nsproxy *old_ns = tsk->nsproxy;
126
	struct user_namespace *user_ns = task_cred_xxx(tsk, user_ns);
127 128
	struct nsproxy *new_ns;
	int err = 0;
129 130 131 132 133 134

	if (!old_ns)
		return 0;

	get_nsproxy(old_ns);

135
	if (!(flags & (CLONE_NEWNS | CLONE_NEWUTS | CLONE_NEWIPC |
136
				CLONE_NEWPID | CLONE_NEWNET)))
137 138
		return 0;

139
	if (!ns_capable(user_ns, CAP_SYS_ADMIN)) {
140
		err = -EPERM;
141 142 143
		goto out;
	}

144 145 146 147 148 149 150 151 152 153 154 155
	/*
	 * 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) && (flags & CLONE_SYSVSEM)) {
		err = -EINVAL;
		goto out;
	}

156
	new_ns = create_new_namespaces(flags, tsk, user_ns, tsk->fs);
157 158 159 160
	if (IS_ERR(new_ns)) {
		err = PTR_ERR(new_ns);
		goto out;
	}
Cedric Le Goater's avatar
Cedric Le Goater committed
161

162
	tsk->nsproxy = new_ns;
163

164
out:
165
	put_nsproxy(old_ns);
166
	return err;
167 168 169 170
}

void free_nsproxy(struct nsproxy *ns)
{
Cedric Le Goater's avatar
Cedric Le Goater committed
171 172 173 174 175 176 177 178
	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)
		put_pid_ns(ns->pid_ns);
179
	put_net(ns->net_ns);
180
	kmem_cache_free(nsproxy_cachep, ns);
181
}
182 183 184

/*
 * Called from unshare. Unshare all the namespaces part of nsproxy.
185
 * On success, returns the new nsproxy.
186 187
 */
int unshare_nsproxy_namespaces(unsigned long unshare_flags,
188
	struct nsproxy **new_nsp, struct cred *new_cred, struct fs_struct *new_fs)
189
{
190
	struct user_namespace *user_ns;
191 192
	int err = 0;

193
	if (!(unshare_flags & (CLONE_NEWNS | CLONE_NEWUTS | CLONE_NEWIPC |
194
			       CLONE_NEWNET | CLONE_NEWPID)))
195 196
		return 0;

197 198
	user_ns = new_cred ? new_cred->user_ns : current_user_ns();
	if (!ns_capable(user_ns, CAP_SYS_ADMIN))
199 200
		return -EPERM;

201
	*new_nsp = create_new_namespaces(unshare_flags, current, user_ns,
202
					 new_fs ? new_fs : current->fs);
203
	if (IS_ERR(*new_nsp)) {
204
		err = PTR_ERR(*new_nsp);
205 206 207 208
		goto out;
	}

out:
209 210
	return err;
}
211

212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238
void switch_task_namespaces(struct task_struct *p, struct nsproxy *new)
{
	struct nsproxy *ns;

	might_sleep();

	ns = p->nsproxy;

	rcu_assign_pointer(p->nsproxy, new);

	if (ns && atomic_dec_and_test(&ns->count)) {
		/*
		 * wait for others to get what they want from this nsproxy.
		 *
		 * cannot release this nsproxy via the call_rcu() since
		 * put_mnt_ns() will want to sleep
		 */
		synchronize_rcu();
		free_nsproxy(ns);
	}
}

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

239 240 241 242 243
SYSCALL_DEFINE2(setns, int, fd, int, nstype)
{
	const struct proc_ns_operations *ops;
	struct task_struct *tsk = current;
	struct nsproxy *new_nsproxy;
244
	struct proc_ns *ei;
245 246 247 248 249 250 251 252
	struct file *file;
	int err;

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

	err = -EINVAL;
253
	ei = get_proc_ns(file_inode(file));
254 255 256 257
	ops = ei->ns_ops;
	if (nstype && (ops->type != nstype))
		goto out;

258
	new_nsproxy = create_new_namespaces(0, tsk, current_user_ns(), tsk->fs);
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274
	if (IS_ERR(new_nsproxy)) {
		err = PTR_ERR(new_nsproxy);
		goto out;
	}

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

275
int __init nsproxy_cache_init(void)
276
{
277
	nsproxy_cachep = KMEM_CACHE(nsproxy, SLAB_PANIC);
278 279
	return 0;
}