[linux-block.git] / kernel / bpf / net_namespace.c

// SPDX-License-Identifier: GPL-2.0

#include <linux/bpf.h>
#include <linux/filter.h>
#include <net/net_namespace.h>

/*
 * Functions to manage BPF programs attached to netns
 */

struct bpf_netns_link {
	struct bpf_link	link;
	enum bpf_attach_type type;
	enum netns_bpf_attach_type netns_type;

	/* We don't hold a ref to net in order to auto-detach the link
	 * when netns is going away. Instead we rely on pernet
	 * pre_exit callback to clear this pointer. Must be accessed
	 * with netns_bpf_mutex held.
	 */
	struct net *net;
};

/* Protects updates to netns_bpf */
DEFINE_MUTEX(netns_bpf_mutex);

/* Must be called with netns_bpf_mutex held. */
static void __net_exit bpf_netns_link_auto_detach(struct bpf_link *link)
{
	struct bpf_netns_link *net_link =
		container_of(link, struct bpf_netns_link, link);

	net_link->net = NULL;
}

static void bpf_netns_link_release(struct bpf_link *link)
{
	struct bpf_netns_link *net_link =
		container_of(link, struct bpf_netns_link, link);
	enum netns_bpf_attach_type type = net_link->netns_type;
	struct net *net;

	/* Link auto-detached by dying netns. */
	if (!net_link->net)
		return;

	mutex_lock(&netns_bpf_mutex);

	/* Recheck after potential sleep. We can race with cleanup_net
	 * here, but if we see a non-NULL struct net pointer pre_exit
	 * has not happened yet and will block on netns_bpf_mutex.
	 */
	net = net_link->net;
	if (!net)
		goto out_unlock;

	net->bpf.links[type] = NULL;
	RCU_INIT_POINTER(net->bpf.progs[type], NULL);

out_unlock:
	mutex_unlock(&netns_bpf_mutex);
}

static void bpf_netns_link_dealloc(struct bpf_link *link)
{
	struct bpf_netns_link *net_link =
		container_of(link, struct bpf_netns_link, link);

	kfree(net_link);
}

static int bpf_netns_link_update_prog(struct bpf_link *link,
				      struct bpf_prog *new_prog,
				      struct bpf_prog *old_prog)
{
	struct bpf_netns_link *net_link =
		container_of(link, struct bpf_netns_link, link);
	enum netns_bpf_attach_type type = net_link->netns_type;
	struct net *net;
	int ret = 0;

	if (old_prog && old_prog != link->prog)
		return -EPERM;
	if (new_prog->type != link->prog->type)
		return -EINVAL;

	mutex_lock(&netns_bpf_mutex);

	net = net_link->net;
	if (!net || !check_net(net)) {
		/* Link auto-detached or netns dying */
		ret = -ENOLINK;
		goto out_unlock;
	}

	old_prog = xchg(&link->prog, new_prog);
	rcu_assign_pointer(net->bpf.progs[type], new_prog);
	bpf_prog_put(old_prog);

out_unlock:
	mutex_unlock(&netns_bpf_mutex);
	return ret;
}

static int bpf_netns_link_fill_info(const struct bpf_link *link,
				    struct bpf_link_info *info)
{
	const struct bpf_netns_link *net_link =
		container_of(link, struct bpf_netns_link, link);
	unsigned int inum = 0;
	struct net *net;

	mutex_lock(&netns_bpf_mutex);
	net = net_link->net;
	if (net && check_net(net))
		inum = net->ns.inum;
	mutex_unlock(&netns_bpf_mutex);

	info->netns.netns_ino = inum;
	info->netns.attach_type = net_link->type;
	return 0;
}

static void bpf_netns_link_show_fdinfo(const struct bpf_link *link,
				       struct seq_file *seq)
{
	struct bpf_link_info info = {};

	bpf_netns_link_fill_info(link, &info);
	seq_printf(seq,
		   "netns_ino:\t%u\n"
		   "attach_type:\t%u\n",
		   info.netns.netns_ino,
		   info.netns.attach_type);
}

static const struct bpf_link_ops bpf_netns_link_ops = {
	.release = bpf_netns_link_release,
	.dealloc = bpf_netns_link_dealloc,
	.update_prog = bpf_netns_link_update_prog,
	.fill_link_info = bpf_netns_link_fill_info,
	.show_fdinfo = bpf_netns_link_show_fdinfo,
};

int netns_bpf_prog_query(const union bpf_attr *attr,
			 union bpf_attr __user *uattr)
{
	__u32 __user *prog_ids = u64_to_user_ptr(attr->query.prog_ids);
	u32 prog_id, prog_cnt = 0, flags = 0;
	enum netns_bpf_attach_type type;
	struct bpf_prog *attached;
	struct net *net;

	if (attr->query.query_flags)
		return -EINVAL;

	type = to_netns_bpf_attach_type(attr->query.attach_type);
	if (type < 0)
		return -EINVAL;

	net = get_net_ns_by_fd(attr->query.target_fd);
	if (IS_ERR(net))
		return PTR_ERR(net);

	rcu_read_lock();
	attached = rcu_dereference(net->bpf.progs[type]);
	if (attached) {
		prog_cnt = 1;
		prog_id = attached->aux->id;
	}
	rcu_read_unlock();

	put_net(net);

	if (copy_to_user(&uattr->query.attach_flags, &flags, sizeof(flags)))
		return -EFAULT;
	if (copy_to_user(&uattr->query.prog_cnt, &prog_cnt, sizeof(prog_cnt)))
		return -EFAULT;

	if (!attr->query.prog_cnt || !prog_ids || !prog_cnt)
		return 0;

	if (copy_to_user(prog_ids, &prog_id, sizeof(u32)))
		return -EFAULT;

	return 0;
}

int netns_bpf_prog_attach(const union bpf_attr *attr, struct bpf_prog *prog)
{
	enum netns_bpf_attach_type type;
	struct net *net;
	int ret;

	type = to_netns_bpf_attach_type(attr->attach_type);
	if (type < 0)
		return -EINVAL;

	net = current->nsproxy->net_ns;
	mutex_lock(&netns_bpf_mutex);

	/* Attaching prog directly is not compatible with links */
	if (net->bpf.links[type]) {
		ret = -EEXIST;
		goto out_unlock;
	}

	switch (type) {
	case NETNS_BPF_FLOW_DISSECTOR:
		ret = flow_dissector_bpf_prog_attach(net, prog);
		break;
	default:
		ret = -EINVAL;
		break;
	}
out_unlock:
	mutex_unlock(&netns_bpf_mutex);

	return ret;
}

/* Must be called with netns_bpf_mutex held. */
static int __netns_bpf_prog_detach(struct net *net,
				   enum netns_bpf_attach_type type)
{
	struct bpf_prog *attached;

	/* Progs attached via links cannot be detached */
	if (net->bpf.links[type])
		return -EINVAL;

	attached = rcu_dereference_protected(net->bpf.progs[type],
					     lockdep_is_held(&netns_bpf_mutex));
	if (!attached)
		return -ENOENT;
	RCU_INIT_POINTER(net->bpf.progs[type], NULL);
	bpf_prog_put(attached);
	return 0;
}

int netns_bpf_prog_detach(const union bpf_attr *attr)
{
	enum netns_bpf_attach_type type;
	int ret;

	type = to_netns_bpf_attach_type(attr->attach_type);
	if (type < 0)
		return -EINVAL;

	mutex_lock(&netns_bpf_mutex);
	ret = __netns_bpf_prog_detach(current->nsproxy->net_ns, type);
	mutex_unlock(&netns_bpf_mutex);

	return ret;
}

static int netns_bpf_link_attach(struct net *net, struct bpf_link *link,
				 enum netns_bpf_attach_type type)
{
	struct bpf_prog *prog;
	int err;

	mutex_lock(&netns_bpf_mutex);

	/* Allow attaching only one prog or link for now */
	if (net->bpf.links[type]) {
		err = -E2BIG;
		goto out_unlock;
	}
	/* Links are not compatible with attaching prog directly */
	prog = rcu_dereference_protected(net->bpf.progs[type],
					 lockdep_is_held(&netns_bpf_mutex));
	if (prog) {
		err = -EEXIST;
		goto out_unlock;
	}

	switch (type) {
	case NETNS_BPF_FLOW_DISSECTOR:
		err = flow_dissector_bpf_prog_attach(net, link->prog);
		break;
	default:
		err = -EINVAL;
		break;
	}
	if (err)
		goto out_unlock;

	net->bpf.links[type] = link;

out_unlock:
	mutex_unlock(&netns_bpf_mutex);
	return err;
}

int netns_bpf_link_create(const union bpf_attr *attr, struct bpf_prog *prog)
{
	enum netns_bpf_attach_type netns_type;
	struct bpf_link_primer link_primer;
	struct bpf_netns_link *net_link;
	enum bpf_attach_type type;
	struct net *net;
	int err;

	if (attr->link_create.flags)
		return -EINVAL;

	type = attr->link_create.attach_type;
	netns_type = to_netns_bpf_attach_type(type);
	if (netns_type < 0)
		return -EINVAL;

	net = get_net_ns_by_fd(attr->link_create.target_fd);
	if (IS_ERR(net))
		return PTR_ERR(net);

	net_link = kzalloc(sizeof(*net_link), GFP_USER);
	if (!net_link) {
		err = -ENOMEM;
		goto out_put_net;
	}
	bpf_link_init(&net_link->link, BPF_LINK_TYPE_NETNS,
		      &bpf_netns_link_ops, prog);
	net_link->net = net;
	net_link->type = type;
	net_link->netns_type = netns_type;

	err = bpf_link_prime(&net_link->link, &link_primer);
	if (err) {
		kfree(net_link);
		goto out_put_net;
	}

	err = netns_bpf_link_attach(net, &net_link->link, netns_type);
	if (err) {
		bpf_link_cleanup(&link_primer);
		goto out_put_net;
	}

	put_net(net);
	return bpf_link_settle(&link_primer);

out_put_net:
	put_net(net);
	return err;
}

static void __net_exit netns_bpf_pernet_pre_exit(struct net *net)
{
	enum netns_bpf_attach_type type;
	struct bpf_link *link;

	mutex_lock(&netns_bpf_mutex);
	for (type = 0; type < MAX_NETNS_BPF_ATTACH_TYPE; type++) {
		link = net->bpf.links[type];
		if (link)
			bpf_netns_link_auto_detach(link);
		else
			__netns_bpf_prog_detach(net, type);
	}
	mutex_unlock(&netns_bpf_mutex);
}

static struct pernet_operations netns_bpf_pernet_ops __net_initdata = {
	.pre_exit = netns_bpf_pernet_pre_exit,
};

static int __init netns_bpf_init(void)
{
	return register_pernet_subsys(&netns_bpf_pernet_ops);
}

subsys_initcall(netns_bpf_init);
Commit	Line	Data
b27f7bb5 JS	1	// SPDX-License-Identifier: GPL-2.0
	2
	3	#include <linux/bpf.h>
	4	#include <linux/filter.h>
	5	#include <net/net_namespace.h>
	6
	7	/*
	8	* Functions to manage BPF programs attached to netns
	9	*/
	10
7f045a49 JS	11	struct bpf_netns_link {
	12	struct bpf_link link;
	13	enum bpf_attach_type type;
	14	enum netns_bpf_attach_type netns_type;
	15
	16	/* We don't hold a ref to net in order to auto-detach the link
	17	* when netns is going away. Instead we rely on pernet
	18	* pre_exit callback to clear this pointer. Must be accessed
	19	* with netns_bpf_mutex held.
	20	*/
	21	struct net *net;
	22	};
	23
b27f7bb5 JS	24	/* Protects updates to netns_bpf */
	25	DEFINE_MUTEX(netns_bpf_mutex);
	26
7f045a49 JS	27	/* Must be called with netns_bpf_mutex held. */
	28	static void __net_exit bpf_netns_link_auto_detach(struct bpf_link *link)
	29	{
	30	struct bpf_netns_link *net_link =
	31	container_of(link, struct bpf_netns_link, link);
	32
	33	net_link->net = NULL;
	34	}
	35
	36	static void bpf_netns_link_release(struct bpf_link *link)
	37	{
	38	struct bpf_netns_link *net_link =
	39	container_of(link, struct bpf_netns_link, link);
	40	enum netns_bpf_attach_type type = net_link->netns_type;
	41	struct net *net;
	42
	43	/* Link auto-detached by dying netns. */
	44	if (!net_link->net)
	45	return;
	46
	47	mutex_lock(&netns_bpf_mutex);
	48
	49	/* Recheck after potential sleep. We can race with cleanup_net
	50	* here, but if we see a non-NULL struct net pointer pre_exit
	51	* has not happened yet and will block on netns_bpf_mutex.
	52	*/
	53	net = net_link->net;
	54	if (!net)
	55	goto out_unlock;
	56
	57	net->bpf.links[type] = NULL;
	58	RCU_INIT_POINTER(net->bpf.progs[type], NULL);
	59
	60	out_unlock:
	61	mutex_unlock(&netns_bpf_mutex);
	62	}
	63
	64	static void bpf_netns_link_dealloc(struct bpf_link *link)
	65	{
	66	struct bpf_netns_link *net_link =
	67	container_of(link, struct bpf_netns_link, link);
	68
	69	kfree(net_link);
	70	}
	71
	72	static int bpf_netns_link_update_prog(struct bpf_link *link,
	73	struct bpf_prog *new_prog,
	74	struct bpf_prog *old_prog)
	75	{
	76	struct bpf_netns_link *net_link =
	77	container_of(link, struct bpf_netns_link, link);
	78	enum netns_bpf_attach_type type = net_link->netns_type;
	79	struct net *net;
	80	int ret = 0;
	81
	82	if (old_prog && old_prog != link->prog)
	83	return -EPERM;
	84	if (new_prog->type != link->prog->type)
	85	return -EINVAL;
	86
	87	mutex_lock(&netns_bpf_mutex);
	88
	89	net = net_link->net;
	90	if (!net \|\| !check_net(net)) {
91	/* Link auto-detached or netns dying */
92	ret = -ENOLINK;
93	goto out_unlock;
94	}
95
96	old_prog = xchg(&link->prog, new_prog);
97	rcu_assign_pointer(net->bpf.progs[type], new_prog);
98	bpf_prog_put(old_prog);
99
100	out_unlock:
101	mutex_unlock(&netns_bpf_mutex);
102	return ret;
103	}
104
105	static int bpf_netns_link_fill_info(const struct bpf_link *link,
106	struct bpf_link_info *info)
107	{
108	const struct bpf_netns_link *net_link =
109	container_of(link, struct bpf_netns_link, link);
110	unsigned int inum = 0;
111	struct net *net;
112
113	mutex_lock(&netns_bpf_mutex);
114	net = net_link->net;
115	if (net && check_net(net))
116	inum = net->ns.inum;
117	mutex_unlock(&netns_bpf_mutex);
118
119	info->netns.netns_ino = inum;
120	info->netns.attach_type = net_link->type;
121	return 0;
122	}
123
124	static void bpf_netns_link_show_fdinfo(const struct bpf_link *link,
125	struct seq_file *seq)
126	{
127	struct bpf_link_info info = {};
128
129	bpf_netns_link_fill_info(link, &info);
130	seq_printf(seq,
131	"netns_ino:\t%u\n"
132	"attach_type:\t%u\n",
133	info.netns.netns_ino,
134	info.netns.attach_type);
135	}
136
137	static const struct bpf_link_ops bpf_netns_link_ops = {
138	.release = bpf_netns_link_release,
139	.dealloc = bpf_netns_link_dealloc,
140	.update_prog = bpf_netns_link_update_prog,
141	.fill_link_info = bpf_netns_link_fill_info,
142	.show_fdinfo = bpf_netns_link_show_fdinfo,
143	};
144
b27f7bb5 JS	145	int netns_bpf_prog_query(const union bpf_attr *attr,
	146	union bpf_attr __user *uattr)
	147	{
	148	__u32 __user *prog_ids = u64_to_user_ptr(attr->query.prog_ids);
	149	u32 prog_id, prog_cnt = 0, flags = 0;
	150	enum netns_bpf_attach_type type;
	151	struct bpf_prog *attached;
	152	struct net *net;
	153
	154	if (attr->query.query_flags)
	155	return -EINVAL;
	156
	157	type = to_netns_bpf_attach_type(attr->query.attach_type);
	158	if (type < 0)
	159	return -EINVAL;
	160
	161	net = get_net_ns_by_fd(attr->query.target_fd);
	162	if (IS_ERR(net))
	163	return PTR_ERR(net);
	164
	165	rcu_read_lock();
	166	attached = rcu_dereference(net->bpf.progs[type]);
	167	if (attached) {
	168	prog_cnt = 1;
	169	prog_id = attached->aux->id;
	170	}
	171	rcu_read_unlock();
	172
	173	put_net(net);
	174
	175	if (copy_to_user(&uattr->query.attach_flags, &flags, sizeof(flags)))
	176	return -EFAULT;
	177	if (copy_to_user(&uattr->query.prog_cnt, &prog_cnt, sizeof(prog_cnt)))
	178	return -EFAULT;
	179
	180	if (!attr->query.prog_cnt \|\| !prog_ids \|\| !prog_cnt)
	181	return 0;
	182
	183	if (copy_to_user(prog_ids, &prog_id, sizeof(u32)))
	184	return -EFAULT;
	185
	186	return 0;
	187	}
	188
	189	int netns_bpf_prog_attach(const union bpf_attr attr, struct bpf_prog prog)
	190	{
	191	enum netns_bpf_attach_type type;
	192	struct net *net;
	193	int ret;
	194
	195	type = to_netns_bpf_attach_type(attr->attach_type);
	196	if (type < 0)
	197	return -EINVAL;
	198
	199	net = current->nsproxy->net_ns;
	200	mutex_lock(&netns_bpf_mutex);
7f045a49 JS	201
	202	/* Attaching prog directly is not compatible with links */
	203	if (net->bpf.links[type]) {
	204	ret = -EEXIST;
	205	goto out_unlock;
	206	}
	207
b27f7bb5 JS	208	switch (type) {
	209	case NETNS_BPF_FLOW_DISSECTOR:
	210	ret = flow_dissector_bpf_prog_attach(net, prog);
	211	break;
	212	default:
	213	ret = -EINVAL;
	214	break;
	215	}
7f045a49	216	out_unlock:
b27f7bb5 JS	217	mutex_unlock(&netns_bpf_mutex);
	218
	219	return ret;
	220	}
	221
	222	/* Must be called with netns_bpf_mutex held. */
	223	static int __netns_bpf_prog_detach(struct net *net,
	224	enum netns_bpf_attach_type type)
	225	{
	226	struct bpf_prog *attached;
	227
7f045a49 JS	228	/* Progs attached via links cannot be detached */
	229	if (net->bpf.links[type])
	230	return -EINVAL;
	231
b27f7bb5 JS	232	attached = rcu_dereference_protected(net->bpf.progs[type],
	233	lockdep_is_held(&netns_bpf_mutex));
	234	if (!attached)
	235	return -ENOENT;
	236	RCU_INIT_POINTER(net->bpf.progs[type], NULL);
	237	bpf_prog_put(attached);
	238	return 0;
	239	}
	240
	241	int netns_bpf_prog_detach(const union bpf_attr *attr)
	242	{
	243	enum netns_bpf_attach_type type;
	244	int ret;
	245
	246	type = to_netns_bpf_attach_type(attr->attach_type);
	247	if (type < 0)
	248	return -EINVAL;
	249
	250	mutex_lock(&netns_bpf_mutex);
	251	ret = __netns_bpf_prog_detach(current->nsproxy->net_ns, type);
	252	mutex_unlock(&netns_bpf_mutex);
	253
	254	return ret;
	255	}
	256
7f045a49 JS	257	static int netns_bpf_link_attach(struct net net, struct bpf_link link,
	258	enum netns_bpf_attach_type type)
	259	{
	260	struct bpf_prog *prog;
	261	int err;
	262
	263	mutex_lock(&netns_bpf_mutex);
	264
	265	/* Allow attaching only one prog or link for now */
	266	if (net->bpf.links[type]) {
	267	err = -E2BIG;
	268	goto out_unlock;
	269	}
	270	/* Links are not compatible with attaching prog directly */
	271	prog = rcu_dereference_protected(net->bpf.progs[type],
	272	lockdep_is_held(&netns_bpf_mutex));
	273	if (prog) {
	274	err = -EEXIST;
	275	goto out_unlock;
	276	}
	277
	278	switch (type) {
	279	case NETNS_BPF_FLOW_DISSECTOR:
	280	err = flow_dissector_bpf_prog_attach(net, link->prog);
	281	break;
	282	default:
	283	err = -EINVAL;
	284	break;
	285	}
	286	if (err)
	287	goto out_unlock;
	288
	289	net->bpf.links[type] = link;
	290
	291	out_unlock:
	292	mutex_unlock(&netns_bpf_mutex);
	293	return err;
	294	}
	295
	296	int netns_bpf_link_create(const union bpf_attr attr, struct bpf_prog prog)
	297	{
	298	enum netns_bpf_attach_type netns_type;
	299	struct bpf_link_primer link_primer;
	300	struct bpf_netns_link *net_link;
	301	enum bpf_attach_type type;
	302	struct net *net;
	303	int err;
	304
	305	if (attr->link_create.flags)
	306	return -EINVAL;
	307
	308	type = attr->link_create.attach_type;
	309	netns_type = to_netns_bpf_attach_type(type);
	310	if (netns_type < 0)
	311	return -EINVAL;
	312
	313	net = get_net_ns_by_fd(attr->link_create.target_fd);
	314	if (IS_ERR(net))
	315	return PTR_ERR(net);
	316
	317	net_link = kzalloc(sizeof(*net_link), GFP_USER);
	318	if (!net_link) {
	319	err = -ENOMEM;
	320	goto out_put_net;
321	}
322	bpf_link_init(&net_link->link, BPF_LINK_TYPE_NETNS,
323	&bpf_netns_link_ops, prog);
324	net_link->net = net;
325	net_link->type = type;
326	net_link->netns_type = netns_type;
327
328	err = bpf_link_prime(&net_link->link, &link_primer);
329	if (err) {
330	kfree(net_link);
331	goto out_put_net;
332	}
333
334	err = netns_bpf_link_attach(net, &net_link->link, netns_type);
335	if (err) {
336	bpf_link_cleanup(&link_primer);
337	goto out_put_net;
338	}
339
340	put_net(net);
341	return bpf_link_settle(&link_primer);
342
343	out_put_net:
344	put_net(net);
345	return err;
346	}
347
b27f7bb5 JS	348	static void __net_exit netns_bpf_pernet_pre_exit(struct net *net)
	349	{
	350	enum netns_bpf_attach_type type;
7f045a49	351	struct bpf_link *link;
b27f7bb5 JS	352
b27f7bb5 JS	353	mutex_lock(&netns_bpf_mutex);
7f045a49 JS	354	for (type = 0; type < MAX_NETNS_BPF_ATTACH_TYPE; type++) {
	355	link = net->bpf.links[type];
	356	if (link)
	357	bpf_netns_link_auto_detach(link);
	358	else
	359	__netns_bpf_prog_detach(net, type);
	360	}
b27f7bb5 JS	361	mutex_unlock(&netns_bpf_mutex);
	362	}
	363
	364	static struct pernet_operations netns_bpf_pernet_ops __net_initdata = {
	365	.pre_exit = netns_bpf_pernet_pre_exit,
	366	};
	367
	368	static int __init netns_bpf_init(void)
	369	{
	370	return register_pernet_subsys(&netns_bpf_pernet_ops);
	371	}
	372
	373	subsys_initcall(netns_bpf_init);