[deliverable/linux.git] / include / net / net_namespace.h

/*
 * Operations on the network namespace
 */
#ifndef __NET_NET_NAMESPACE_H
#define __NET_NET_NAMESPACE_H

#include <linux/atomic.h>
#include <linux/workqueue.h>
#include <linux/list.h>
#include <linux/sysctl.h>

#include <net/flow.h>
#include <net/netns/core.h>
#include <net/netns/mib.h>
#include <net/netns/unix.h>
#include <net/netns/packet.h>
#include <net/netns/ipv4.h>
#include <net/netns/ipv6.h>
#include <net/netns/ieee802154_6lowpan.h>
#include <net/netns/sctp.h>
#include <net/netns/dccp.h>
#include <net/netns/netfilter.h>
#include <net/netns/x_tables.h>
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
#include <net/netns/conntrack.h>
#endif
#include <net/netns/nftables.h>
#include <net/netns/xfrm.h>
#include <net/netns/mpls.h>
#include <linux/ns_common.h>
#include <linux/idr.h>
#include <linux/skbuff.h>

struct user_namespace;
struct proc_dir_entry;
struct net_device;
struct sock;
struct ctl_table_header;
struct net_generic;
struct sock;
struct netns_ipvs;


#define NETDEV_HASHBITS    8
#define NETDEV_HASHENTRIES (1 << NETDEV_HASHBITS)

struct net {
	atomic_t		passive;	/* To decided when the network
						 * namespace should be freed.
						 */
	atomic_t		count;		/* To decided when the network
						 *  namespace should be shut down.
						 */
	spinlock_t		rules_mod_lock;

	atomic64_t		cookie_gen;

	struct list_head	list;		/* list of network namespaces */
	struct list_head	cleanup_list;	/* namespaces on death row */
	struct list_head	exit_list;	/* Use only net_mutex */

	struct user_namespace   *user_ns;	/* Owning user namespace */
	spinlock_t		nsid_lock;
	struct idr		netns_ids;

	struct ns_common	ns;

	struct proc_dir_entry 	*proc_net;
	struct proc_dir_entry 	*proc_net_stat;

#ifdef CONFIG_SYSCTL
	struct ctl_table_set	sysctls;
#endif

	struct sock 		*rtnl;			/* rtnetlink socket */
	struct sock		*genl_sock;

	struct list_head 	dev_base_head;
	struct hlist_head 	*dev_name_head;
	struct hlist_head	*dev_index_head;
	unsigned int		dev_base_seq;	/* protected by rtnl_mutex */
	int			ifindex;
	unsigned int		dev_unreg_count;

	/* core fib_rules */
	struct list_head	rules_ops;


	struct net_device       *loopback_dev;          /* The loopback */
	struct netns_core	core;
	struct netns_mib	mib;
	struct netns_packet	packet;
	struct netns_unix	unx;
	struct netns_ipv4	ipv4;
#if IS_ENABLED(CONFIG_IPV6)
	struct netns_ipv6	ipv6;
#endif
#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN)
	struct netns_ieee802154_lowpan	ieee802154_lowpan;
#endif
#if defined(CONFIG_IP_SCTP) || defined(CONFIG_IP_SCTP_MODULE)
	struct netns_sctp	sctp;
#endif
#if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE)
	struct netns_dccp	dccp;
#endif
#ifdef CONFIG_NETFILTER
	struct netns_nf		nf;
	struct netns_xt		xt;
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
	struct netns_ct		ct;
#endif
#if defined(CONFIG_NF_TABLES) || defined(CONFIG_NF_TABLES_MODULE)
	struct netns_nftables	nft;
#endif
#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
	struct netns_nf_frag	nf_frag;
#endif
	struct sock		*nfnl;
	struct sock		*nfnl_stash;
#if IS_ENABLED(CONFIG_NETFILTER_NETLINK_ACCT)
	struct list_head        nfnl_acct_list;
#endif
#if IS_ENABLED(CONFIG_NF_CT_NETLINK_TIMEOUT)
	struct list_head	nfct_timeout_list;
#endif
#endif
#ifdef CONFIG_WEXT_CORE
	struct sk_buff_head	wext_nlevents;
#endif
	struct net_generic __rcu	*gen;

	/* Note : following structs are cache line aligned */
#ifdef CONFIG_XFRM
	struct netns_xfrm	xfrm;
#endif
#if IS_ENABLED(CONFIG_IP_VS)
	struct netns_ipvs	*ipvs;
#endif
#if IS_ENABLED(CONFIG_MPLS)
	struct netns_mpls	mpls;
#endif
	struct sock		*diag_nlsk;
	atomic_t		fnhe_genid;
};

#include <linux/seq_file_net.h>

/* Init's network namespace */
extern struct net init_net;

#ifdef CONFIG_NET_NS
struct net *copy_net_ns(unsigned long flags, struct user_namespace *user_ns,
			struct net *old_net);

#else /* CONFIG_NET_NS */
#include <linux/sched.h>
#include <linux/nsproxy.h>
static inline struct net *copy_net_ns(unsigned long flags,
	struct user_namespace *user_ns, struct net *old_net)
{
	if (flags & CLONE_NEWNET)
		return ERR_PTR(-EINVAL);
	return old_net;
}
#endif /* CONFIG_NET_NS */


extern struct list_head net_namespace_list;

struct net *get_net_ns_by_pid(pid_t pid);
struct net *get_net_ns_by_fd(int pid);

#ifdef CONFIG_SYSCTL
void ipx_register_sysctl(void);
void ipx_unregister_sysctl(void);
#else
#define ipx_register_sysctl()
#define ipx_unregister_sysctl()
#endif

#ifdef CONFIG_NET_NS
void __put_net(struct net *net);

static inline struct net *get_net(struct net *net)
{
	atomic_inc(&net->count);
	return net;
}

static inline struct net *maybe_get_net(struct net *net)
{
	/* Used when we know struct net exists but we
	 * aren't guaranteed a previous reference count
	 * exists.  If the reference count is zero this
	 * function fails and returns NULL.
	 */
	if (!atomic_inc_not_zero(&net->count))
		net = NULL;
	return net;
}

static inline void put_net(struct net *net)
{
	if (atomic_dec_and_test(&net->count))
		__put_net(net);
}

static inline
int net_eq(const struct net *net1, const struct net *net2)
{
	return net1 == net2;
}

void net_drop_ns(void *);

#else

static inline struct net *get_net(struct net *net)
{
	return net;
}

static inline void put_net(struct net *net)
{
}

static inline struct net *maybe_get_net(struct net *net)
{
	return net;
}

static inline
int net_eq(const struct net *net1, const struct net *net2)
{
	return 1;
}

#define net_drop_ns NULL
#endif


typedef struct {
#ifdef CONFIG_NET_NS
	struct net *net;
#endif
} possible_net_t;

static inline void write_pnet(possible_net_t *pnet, struct net *net)
{
#ifdef CONFIG_NET_NS
	pnet->net = net;
#endif
}

static inline struct net *read_pnet(const possible_net_t *pnet)
{
#ifdef CONFIG_NET_NS
	return pnet->net;
#else
	return &init_net;
#endif
}

#define for_each_net(VAR)				\
	list_for_each_entry(VAR, &net_namespace_list, list)

#define for_each_net_rcu(VAR)				\
	list_for_each_entry_rcu(VAR, &net_namespace_list, list)

#ifdef CONFIG_NET_NS
#define __net_init
#define __net_exit
#define __net_initdata
#define __net_initconst
#else
#define __net_init	__init
#define __net_exit	__ref
#define __net_initdata	__initdata
#define __net_initconst	__initconst
#endif

int peernet2id_alloc(struct net *net, struct net *peer);
int peernet2id(struct net *net, struct net *peer);
bool peernet_has_id(struct net *net, struct net *peer);
struct net *get_net_ns_by_id(struct net *net, int id);

struct pernet_operations {
	struct list_head list;
	int (*init)(struct net *net);
	void (*exit)(struct net *net);
	void (*exit_batch)(struct list_head *net_exit_list);
	int *id;
	size_t size;
};

/*
 * Use these carefully.  If you implement a network device and it
 * needs per network namespace operations use device pernet operations,
 * otherwise use pernet subsys operations.
 *
 * Network interfaces need to be removed from a dying netns _before_
 * subsys notifiers can be called, as most of the network code cleanup
 * (which is done from subsys notifiers) runs with the assumption that
 * dev_remove_pack has been called so no new packets will arrive during
 * and after the cleanup functions have been called.  dev_remove_pack
 * is not per namespace so instead the guarantee of no more packets
 * arriving in a network namespace is provided by ensuring that all
 * network devices and all sockets have left the network namespace
 * before the cleanup methods are called.
 *
 * For the longest time the ipv4 icmp code was registered as a pernet
 * device which caused kernel oops, and panics during network
 * namespace cleanup.   So please don't get this wrong.
 */
int register_pernet_subsys(struct pernet_operations *);
void unregister_pernet_subsys(struct pernet_operations *);
int register_pernet_device(struct pernet_operations *);
void unregister_pernet_device(struct pernet_operations *);

struct ctl_table;
struct ctl_table_header;

#ifdef CONFIG_SYSCTL
int net_sysctl_init(void);
struct ctl_table_header *register_net_sysctl(struct net *net, const char *path,
					     struct ctl_table *table);
void unregister_net_sysctl_table(struct ctl_table_header *header);
#else
static inline int net_sysctl_init(void) { return 0; }
static inline struct ctl_table_header *register_net_sysctl(struct net *net,
	const char *path, struct ctl_table *table)
{
	return NULL;
}
static inline void unregister_net_sysctl_table(struct ctl_table_header *header)
{
}
#endif

static inline int rt_genid_ipv4(struct net *net)
{
	return atomic_read(&net->ipv4.rt_genid);
}

static inline void rt_genid_bump_ipv4(struct net *net)
{
	atomic_inc(&net->ipv4.rt_genid);
}

extern void (*__fib6_flush_trees)(struct net *net);
static inline void rt_genid_bump_ipv6(struct net *net)
{
	if (__fib6_flush_trees)
		__fib6_flush_trees(net);
}

#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN)
static inline struct netns_ieee802154_lowpan *
net_ieee802154_lowpan(struct net *net)
{
	return &net->ieee802154_lowpan;
}
#endif

/* For callers who don't really care about whether it's IPv4 or IPv6 */
static inline void rt_genid_bump_all(struct net *net)
{
	rt_genid_bump_ipv4(net);
	rt_genid_bump_ipv6(net);
}

static inline int fnhe_genid(struct net *net)
{
	return atomic_read(&net->fnhe_genid);
}

static inline void fnhe_genid_bump(struct net *net)
{
	atomic_inc(&net->fnhe_genid);
}

#endif /* __NET_NET_NAMESPACE_H */
Commit	Line	Data
	1	/*
	2	* Operations on the network namespace
	3	*/
	4	#ifndef __NET_NET_NAMESPACE_H
	5	#define __NET_NET_NAMESPACE_H
	6
	7	#include <linux/atomic.h>
	8	#include <linux/workqueue.h>
	9	#include <linux/list.h>
	10	#include <linux/sysctl.h>
	11
	12	#include <net/flow.h>
	13	#include <net/netns/core.h>
	14	#include <net/netns/mib.h>
	15	#include <net/netns/unix.h>
	16	#include <net/netns/packet.h>
	17	#include <net/netns/ipv4.h>
	18	#include <net/netns/ipv6.h>
	19	#include <net/netns/ieee802154_6lowpan.h>
	20	#include <net/netns/sctp.h>
	21	#include <net/netns/dccp.h>
	22	#include <net/netns/netfilter.h>
	23	#include <net/netns/x_tables.h>
	24	#if defined(CONFIG_NF_CONNTRACK) \|\| defined(CONFIG_NF_CONNTRACK_MODULE)
	25	#include <net/netns/conntrack.h>
	26	#endif
	27	#include <net/netns/nftables.h>
	28	#include <net/netns/xfrm.h>
	29	#include <net/netns/mpls.h>
	30	#include <linux/ns_common.h>
	31	#include <linux/idr.h>
	32	#include <linux/skbuff.h>
	33
	34	struct user_namespace;
	35	struct proc_dir_entry;
	36	struct net_device;
	37	struct sock;
	38	struct ctl_table_header;
	39	struct net_generic;
	40	struct sock;
	41	struct netns_ipvs;
	42
	43
	44	#define NETDEV_HASHBITS 8
	45	#define NETDEV_HASHENTRIES (1 << NETDEV_HASHBITS)
	46
	47	struct net {
	48	atomic_t passive; /* To decided when the network
	49	* namespace should be freed.
	50	*/
	51	atomic_t count; /* To decided when the network
	52	* namespace should be shut down.
	53	*/
	54	spinlock_t rules_mod_lock;
	55
	56	atomic64_t cookie_gen;
	57
	58	struct list_head list; /* list of network namespaces */
	59	struct list_head cleanup_list; /* namespaces on death row */
	60	struct list_head exit_list; /* Use only net_mutex */
	61
	62	struct user_namespace user_ns; / Owning user namespace */
	63	spinlock_t nsid_lock;
	64	struct idr netns_ids;
	65
	66	struct ns_common ns;
	67
	68	struct proc_dir_entry *proc_net;
	69	struct proc_dir_entry *proc_net_stat;
	70
	71	#ifdef CONFIG_SYSCTL
	72	struct ctl_table_set sysctls;
	73	#endif
	74
	75	struct sock rtnl; / rtnetlink socket */
	76	struct sock *genl_sock;
	77
	78	struct list_head dev_base_head;
	79	struct hlist_head *dev_name_head;
	80	struct hlist_head *dev_index_head;
	81	unsigned int dev_base_seq; /* protected by rtnl_mutex */
	82	int ifindex;
	83	unsigned int dev_unreg_count;
	84
	85	/* core fib_rules */
	86	struct list_head rules_ops;
	87
	88
	89	struct net_device loopback_dev; / The loopback */
	90	struct netns_core core;
	91	struct netns_mib mib;
	92	struct netns_packet packet;
	93	struct netns_unix unx;
	94	struct netns_ipv4 ipv4;
	95	#if IS_ENABLED(CONFIG_IPV6)
	96	struct netns_ipv6 ipv6;
	97	#endif
	98	#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN)
	99	struct netns_ieee802154_lowpan ieee802154_lowpan;
	100	#endif
	101	#if defined(CONFIG_IP_SCTP) \|\| defined(CONFIG_IP_SCTP_MODULE)
	102	struct netns_sctp sctp;
	103	#endif
	104	#if defined(CONFIG_IP_DCCP) \|\| defined(CONFIG_IP_DCCP_MODULE)
	105	struct netns_dccp dccp;
	106	#endif
	107	#ifdef CONFIG_NETFILTER
	108	struct netns_nf nf;
	109	struct netns_xt xt;
	110	#if defined(CONFIG_NF_CONNTRACK) \|\| defined(CONFIG_NF_CONNTRACK_MODULE)
	111	struct netns_ct ct;
	112	#endif
	113	#if defined(CONFIG_NF_TABLES) \|\| defined(CONFIG_NF_TABLES_MODULE)
	114	struct netns_nftables nft;
	115	#endif
	116	#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
	117	struct netns_nf_frag nf_frag;
	118	#endif
	119	struct sock *nfnl;
	120	struct sock *nfnl_stash;
	121	#if IS_ENABLED(CONFIG_NETFILTER_NETLINK_ACCT)
	122	struct list_head nfnl_acct_list;
	123	#endif
	124	#if IS_ENABLED(CONFIG_NF_CT_NETLINK_TIMEOUT)
	125	struct list_head nfct_timeout_list;
	126	#endif
	127	#endif
	128	#ifdef CONFIG_WEXT_CORE
	129	struct sk_buff_head wext_nlevents;
	130	#endif
	131	struct net_generic __rcu *gen;
	132
	133	/* Note : following structs are cache line aligned */
	134	#ifdef CONFIG_XFRM
	135	struct netns_xfrm xfrm;
	136	#endif
	137	#if IS_ENABLED(CONFIG_IP_VS)
	138	struct netns_ipvs *ipvs;
	139	#endif
	140	#if IS_ENABLED(CONFIG_MPLS)
	141	struct netns_mpls mpls;
	142	#endif
	143	struct sock *diag_nlsk;
	144	atomic_t fnhe_genid;
	145	};
	146
	147	#include <linux/seq_file_net.h>
	148
	149	/* Init's network namespace */
	150	extern struct net init_net;
	151
	152	#ifdef CONFIG_NET_NS
	153	struct net copy_net_ns(unsigned long flags, struct user_namespace user_ns,
	154	struct net *old_net);
	155
	156	#else /* CONFIG_NET_NS */
	157	#include <linux/sched.h>
	158	#include <linux/nsproxy.h>
	159	static inline struct net *copy_net_ns(unsigned long flags,
	160	struct user_namespace user_ns, struct net old_net)
	161	{
	162	if (flags & CLONE_NEWNET)
	163	return ERR_PTR(-EINVAL);
	164	return old_net;
	165	}
	166	#endif /* CONFIG_NET_NS */
	167
	168
	169	extern struct list_head net_namespace_list;
	170
	171	struct net *get_net_ns_by_pid(pid_t pid);
	172	struct net *get_net_ns_by_fd(int pid);
	173
	174	#ifdef CONFIG_SYSCTL
	175	void ipx_register_sysctl(void);
	176	void ipx_unregister_sysctl(void);
	177	#else
	178	#define ipx_register_sysctl()
	179	#define ipx_unregister_sysctl()
	180	#endif
	181
	182	#ifdef CONFIG_NET_NS
	183	void __put_net(struct net *net);
	184
	185	static inline struct net get_net(struct net net)
	186	{
	187	atomic_inc(&net->count);
	188	return net;
	189	}
	190
	191	static inline struct net maybe_get_net(struct net net)
	192	{
	193	/* Used when we know struct net exists but we
	194	* aren't guaranteed a previous reference count
	195	* exists. If the reference count is zero this
	196	* function fails and returns NULL.
	197	*/
	198	if (!atomic_inc_not_zero(&net->count))
	199	net = NULL;
	200	return net;
	201	}
	202
	203	static inline void put_net(struct net *net)
	204	{
	205	if (atomic_dec_and_test(&net->count))
	206	__put_net(net);
	207	}
	208
	209	static inline
	210	int net_eq(const struct net net1, const struct net net2)
	211	{
	212	return net1 == net2;
	213	}
	214
	215	void net_drop_ns(void *);
	216
	217	#else
	218
	219	static inline struct net get_net(struct net net)
	220	{
	221	return net;
	222	}
	223
	224	static inline void put_net(struct net *net)
	225	{
	226	}
	227
	228	static inline struct net maybe_get_net(struct net net)
	229	{
	230	return net;
	231	}
	232
	233	static inline
	234	int net_eq(const struct net net1, const struct net net2)
	235	{
	236	return 1;
	237	}
	238
	239	#define net_drop_ns NULL
	240	#endif
	241
	242
	243	typedef struct {
	244	#ifdef CONFIG_NET_NS
	245	struct net *net;
	246	#endif
	247	} possible_net_t;
	248
	249	static inline void write_pnet(possible_net_t pnet, struct net net)
	250	{
	251	#ifdef CONFIG_NET_NS
	252	pnet->net = net;
	253	#endif
	254	}
	255
	256	static inline struct net read_pnet(const possible_net_t pnet)
	257	{
	258	#ifdef CONFIG_NET_NS
	259	return pnet->net;
	260	#else
	261	return &init_net;
	262	#endif
	263	}
	264
	265	#define for_each_net(VAR) \
	266	list_for_each_entry(VAR, &net_namespace_list, list)
	267
	268	#define for_each_net_rcu(VAR) \
	269	list_for_each_entry_rcu(VAR, &net_namespace_list, list)
	270
	271	#ifdef CONFIG_NET_NS
	272	#define __net_init
	273	#define __net_exit
	274	#define __net_initdata
	275	#define __net_initconst
	276	#else
	277	#define __net_init __init
	278	#define __net_exit __ref
	279	#define __net_initdata __initdata
	280	#define __net_initconst __initconst
	281	#endif
	282
	283	int peernet2id_alloc(struct net net, struct net peer);
	284	int peernet2id(struct net net, struct net peer);
	285	bool peernet_has_id(struct net net, struct net peer);
	286	struct net get_net_ns_by_id(struct net net, int id);
	287
	288	struct pernet_operations {
	289	struct list_head list;
	290	int (init)(struct net net);
	291	void (exit)(struct net net);
	292	void (exit_batch)(struct list_head net_exit_list);
	293	int *id;
	294	size_t size;
	295	};
	296
	297	/*
	298	* Use these carefully. If you implement a network device and it
	299	* needs per network namespace operations use device pernet operations,
	300	* otherwise use pernet subsys operations.
	301	*
	302	* Network interfaces need to be removed from a dying netns _before_
	303	* subsys notifiers can be called, as most of the network code cleanup
	304	* (which is done from subsys notifiers) runs with the assumption that
	305	* dev_remove_pack has been called so no new packets will arrive during
	306	* and after the cleanup functions have been called. dev_remove_pack
	307	* is not per namespace so instead the guarantee of no more packets
	308	* arriving in a network namespace is provided by ensuring that all
	309	* network devices and all sockets have left the network namespace
	310	* before the cleanup methods are called.
	311	*
	312	* For the longest time the ipv4 icmp code was registered as a pernet
	313	* device which caused kernel oops, and panics during network
	314	* namespace cleanup. So please don't get this wrong.
	315	*/
	316	int register_pernet_subsys(struct pernet_operations *);
	317	void unregister_pernet_subsys(struct pernet_operations *);
	318	int register_pernet_device(struct pernet_operations *);
	319	void unregister_pernet_device(struct pernet_operations *);
	320
	321	struct ctl_table;
	322	struct ctl_table_header;
	323
	324	#ifdef CONFIG_SYSCTL
	325	int net_sysctl_init(void);
	326	struct ctl_table_header register_net_sysctl(struct net net, const char *path,
	327	struct ctl_table *table);
	328	void unregister_net_sysctl_table(struct ctl_table_header *header);
	329	#else
	330	static inline int net_sysctl_init(void) { return 0; }
	331	static inline struct ctl_table_header register_net_sysctl(struct net net,
	332	const char path, struct ctl_table table)
	333	{
	334	return NULL;
	335	}
	336	static inline void unregister_net_sysctl_table(struct ctl_table_header *header)
	337	{
	338	}
	339	#endif
	340
	341	static inline int rt_genid_ipv4(struct net *net)
	342	{
	343	return atomic_read(&net->ipv4.rt_genid);
	344	}
	345
	346	static inline void rt_genid_bump_ipv4(struct net *net)
	347	{
	348	atomic_inc(&net->ipv4.rt_genid);
	349	}
	350
	351	extern void (__fib6_flush_trees)(struct net net);
	352	static inline void rt_genid_bump_ipv6(struct net *net)
	353	{
	354	if (__fib6_flush_trees)
	355	__fib6_flush_trees(net);
	356	}
	357
	358	#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN)
	359	static inline struct netns_ieee802154_lowpan *
	360	net_ieee802154_lowpan(struct net *net)
	361	{
	362	return &net->ieee802154_lowpan;
	363	}
	364	#endif
	365
	366	/* For callers who don't really care about whether it's IPv4 or IPv6 */
	367	static inline void rt_genid_bump_all(struct net *net)
	368	{
	369	rt_genid_bump_ipv4(net);
	370	rt_genid_bump_ipv6(net);
	371	}
	372
	373	static inline int fnhe_genid(struct net *net)
	374	{
	375	return atomic_read(&net->fnhe_genid);
	376	}
	377
	378	static inline void fnhe_genid_bump(struct net *net)
	379	{
	380	atomic_inc(&net->fnhe_genid);
	381	}
	382
	383	#endif /* __NET_NET_NAMESPACE_H */