namespace.c 3.88 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11
/*
 * linux/ipc/namespace.c
 * Copyright (C) 2006 Pavel Emelyanov <xemul@openvz.org> OpenVZ, SWsoft Inc.
 */

#include <linux/ipc.h>
#include <linux/msg.h>
#include <linux/ipc_namespace.h>
#include <linux/rcupdate.h>
#include <linux/nsproxy.h>
#include <linux/slab.h>
12 13
#include <linux/fs.h>
#include <linux/mount.h>
14
#include <linux/user_namespace.h>
15
#include <linux/proc_ns.h>
16 17 18

#include "util.h"

19
static struct ipc_namespace *create_ipc_ns(struct user_namespace *user_ns,
20
					   struct ipc_namespace *old_ns)
21 22
{
	struct ipc_namespace *ns;
23
	int err;
24 25 26

	ns = kmalloc(sizeof(struct ipc_namespace), GFP_KERNEL);
	if (ns == NULL)
27
		return ERR_PTR(-ENOMEM);
28

29 30 31 32 33 34
	err = proc_alloc_inum(&ns->proc_inum);
	if (err) {
		kfree(ns);
		return ERR_PTR(err);
	}

35 36 37
	atomic_set(&ns->count, 1);
	err = mq_init_ns(ns);
	if (err) {
38
		proc_free_inum(ns->proc_inum);
39 40 41
		kfree(ns);
		return ERR_PTR(err);
	}
42 43
	atomic_inc(&nr_ipc_ns);

44 45 46
	sem_init_ns(ns);
	msg_init_ns(ns);
	shm_init_ns(ns);
47

48
	ns->user_ns = get_user_ns(user_ns);
49

50 51 52
	return ns;
}

53
struct ipc_namespace *copy_ipcs(unsigned long flags,
54
	struct user_namespace *user_ns, struct ipc_namespace *ns)
55 56
{
	if (!(flags & CLONE_NEWIPC))
57
		return get_ipc_ns(ns);
58
	return create_ipc_ns(user_ns, ns);
59 60
}

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
/*
 * free_ipcs - free all ipcs of one type
 * @ns:   the namespace to remove the ipcs from
 * @ids:  the table of ipcs to free
 * @free: the function called to free each individual ipc
 *
 * Called for each kind of ipc when an ipc_namespace exits.
 */
void free_ipcs(struct ipc_namespace *ns, struct ipc_ids *ids,
	       void (*free)(struct ipc_namespace *, struct kern_ipc_perm *))
{
	struct kern_ipc_perm *perm;
	int next_id;
	int total, in_use;

Davidlohr Bueso's avatar
Davidlohr Bueso committed
76
	down_write(&ids->rwsem);
77 78 79 80 81 82 83

	in_use = ids->in_use;

	for (total = 0, next_id = 0; total < in_use; next_id++) {
		perm = idr_find(&ids->ipcs_idr, next_id);
		if (perm == NULL)
			continue;
Davidlohr Bueso's avatar
Davidlohr Bueso committed
84 85
		rcu_read_lock();
		ipc_lock_object(perm);
86 87 88
		free(ns, perm);
		total++;
	}
Davidlohr Bueso's avatar
Davidlohr Bueso committed
89
	up_write(&ids->rwsem);
90 91
}

92 93 94 95 96 97 98
static void free_ipc_ns(struct ipc_namespace *ns)
{
	sem_exit_ns(ns);
	msg_exit_ns(ns);
	shm_exit_ns(ns);
	atomic_dec(&nr_ipc_ns);

99
	put_user_ns(ns->user_ns);
100
	proc_free_inum(ns->proc_inum);
101
	kfree(ns);
102 103
}

104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
/*
 * put_ipc_ns - drop a reference to an ipc namespace.
 * @ns: the namespace to put
 *
 * If this is the last task in the namespace exiting, and
 * it is dropping the refcount to 0, then it can race with
 * a task in another ipc namespace but in a mounts namespace
 * which has this ipcns's mqueuefs mounted, doing some action
 * with one of the mqueuefs files.  That can raise the refcount.
 * So dropping the refcount, and raising the refcount when
 * accessing it through the VFS, are protected with mq_lock.
 *
 * (Clearly, a task raising the refcount on its own ipc_ns
 * needn't take mq_lock since it can't race with the last task
 * in the ipcns exiting).
 */
void put_ipc_ns(struct ipc_namespace *ns)
121
{
122 123 124 125 126 127 128
	if (atomic_dec_and_lock(&ns->count, &mq_lock)) {
		mq_clear_sbinfo(ns);
		spin_unlock(&mq_lock);
		mq_put_mnt(ns);
		free_ipc_ns(ns);
	}
}
129 130 131 132 133 134

static void *ipcns_get(struct task_struct *task)
{
	struct ipc_namespace *ns = NULL;
	struct nsproxy *nsproxy;

135 136
	task_lock(task);
	nsproxy = task->nsproxy;
137 138
	if (nsproxy)
		ns = get_ipc_ns(nsproxy->ipc_ns);
139
	task_unlock(task);
140 141 142 143 144 145 146 147 148

	return ns;
}

static void ipcns_put(void *ns)
{
	return put_ipc_ns(ns);
}

149
static int ipcns_install(struct nsproxy *nsproxy, void *new)
150
{
151
	struct ipc_namespace *ns = new;
152
	if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN) ||
153
	    !ns_capable(current_user_ns(), CAP_SYS_ADMIN))
154 155
		return -EPERM;

156 157 158 159 160 161 162
	/* Ditch state from the old ipc namespace */
	exit_sem(current);
	put_ipc_ns(nsproxy->ipc_ns);
	nsproxy->ipc_ns = get_ipc_ns(ns);
	return 0;
}

163 164 165 166 167 168 169
static unsigned int ipcns_inum(void *vp)
{
	struct ipc_namespace *ns = vp;

	return ns->proc_inum;
}

170 171 172 173 174 175
const struct proc_ns_operations ipcns_operations = {
	.name		= "ipc",
	.type		= CLONE_NEWIPC,
	.get		= ipcns_get,
	.put		= ipcns_put,
	.install	= ipcns_install,
176
	.inum		= ipcns_inum,
177
};