net_namespace.h 8.99 KB
Newer Older
1 2 3 4 5 6
/*
 * Operations on the network namespace
 */
#ifndef __NET_NET_NAMESPACE_H
#define __NET_NET_NAMESPACE_H

Arun Sharma's avatar
Arun Sharma committed
7
#include <linux/atomic.h>
8 9
#include <linux/workqueue.h>
#include <linux/list.h>
10
#include <linux/sysctl.h>
11

12
#include <net/netns/core.h>
Pavel Emelyanov's avatar
Pavel Emelyanov committed
13
#include <net/netns/mib.h>
14
#include <net/netns/unix.h>
15
#include <net/netns/packet.h>
16
#include <net/netns/ipv4.h>
17
#include <net/netns/ipv6.h>
18
#include <net/netns/sctp.h>
19
#include <net/netns/dccp.h>
20
#include <net/netns/netfilter.h>
21
#include <net/netns/x_tables.h>
22 23 24
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
#include <net/netns/conntrack.h>
#endif
25
#include <net/netns/nftables.h>
26
#include <net/netns/xfrm.h>
27

28
struct user_namespace;
29
struct proc_dir_entry;
30
struct net_device;
31
struct sock;
32
struct ctl_table_header;
33
struct net_generic;
Johannes Berg's avatar
Johannes Berg committed
34
struct sock;
35
struct netns_ipvs;
36

37 38 39 40

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

41
struct net {
42 43 44
	atomic_t		passive;	/* To decided when the network
						 * namespace should be freed.
						 */
45
	atomic_t		count;		/* To decided when the network
46
						 *  namespace should be shut down.
47
						 */
48
#ifdef NETNS_REFCNT_DEBUG
49 50 51
	atomic_t		use_count;	/* To track references we
						 * destroy on demand
						 */
52
#endif
53 54
	spinlock_t		rules_mod_lock;

55
	struct list_head	list;		/* list of network namespaces */
56
	struct list_head	cleanup_list;	/* namespaces on death row */
57
	struct list_head	exit_list;	/* Use only net_mutex */
58

59 60
	struct user_namespace   *user_ns;	/* Owning user namespace */

61 62
	unsigned int		proc_inum;

63 64
	struct proc_dir_entry 	*proc_net;
	struct proc_dir_entry 	*proc_net_stat;
65

66 67 68
#ifdef CONFIG_SYSCTL
	struct ctl_table_set	sysctls;
#endif
69

70 71
	struct sock 		*rtnl;			/* rtnetlink socket */
	struct sock		*genl_sock;
72

73 74 75
	struct list_head 	dev_base_head;
	struct hlist_head 	*dev_name_head;
	struct hlist_head	*dev_index_head;
76
	unsigned int		dev_base_seq;	/* protected by rtnl_mutex */
77
	int			ifindex;
78
	unsigned int		dev_unreg_count;
79

80 81 82
	/* core fib_rules */
	struct list_head	rules_ops;

83

84
	struct net_device       *loopback_dev;          /* The loopback */
85
	struct netns_core	core;
Pavel Emelyanov's avatar
Pavel Emelyanov committed
86
	struct netns_mib	mib;
87
	struct netns_packet	packet;
88
	struct netns_unix	unx;
89
	struct netns_ipv4	ipv4;
90
#if IS_ENABLED(CONFIG_IPV6)
91 92
	struct netns_ipv6	ipv6;
#endif
93 94 95
#if defined(CONFIG_IP_SCTP) || defined(CONFIG_IP_SCTP_MODULE)
	struct netns_sctp	sctp;
#endif
96 97 98
#if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE)
	struct netns_dccp	dccp;
#endif
99
#ifdef CONFIG_NETFILTER
100
	struct netns_nf		nf;
101
	struct netns_xt		xt;
102 103
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
	struct netns_ct		ct;
104
#endif
105 106 107
#if defined(CONFIG_NF_TABLES) || defined(CONFIG_NF_TABLES_MODULE)
	struct netns_nftables	nft;
#endif
108 109
#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
	struct netns_nf_frag	nf_frag;
110
#endif
111 112
	struct sock		*nfnl;
	struct sock		*nfnl_stash;
113
#endif
Johannes Berg's avatar
Johannes Berg committed
114
#ifdef CONFIG_WEXT_CORE
115
	struct sk_buff_head	wext_nlevents;
116
#endif
Eric Dumazet's avatar
Eric Dumazet committed
117
	struct net_generic __rcu	*gen;
118 119 120 121 122

	/* Note : following structs are cache line aligned */
#ifdef CONFIG_XFRM
	struct netns_xfrm	xfrm;
#endif
123
#if IS_ENABLED(CONFIG_IP_VS)
124
	struct netns_ipvs	*ipvs;
125
#endif
126
	struct sock		*diag_nlsk;
127
	atomic_t		fnhe_genid;
128 129
};

130 131 132 133 134 135 136
/*
 * ifindex generation is per-net namespace, and loopback is
 * always the 1st device in ns (see net_dev_init), thus any
 * loopback device should get ifindex 1
 */

#define LOOPBACK_IFINDEX	1
137

138 139
#include <linux/seq_file_net.h>

140
/* Init's network namespace */
141
extern struct net init_net;
142

143
#ifdef CONFIG_NET_NS
144 145
struct net *copy_net_ns(unsigned long flags, struct user_namespace *user_ns,
			struct net *old_net);
146

147 148 149
#else /* CONFIG_NET_NS */
#include <linux/sched.h>
#include <linux/nsproxy.h>
150 151
static inline struct net *copy_net_ns(unsigned long flags,
	struct user_namespace *user_ns, struct net *old_net)
152
{
153 154 155
	if (flags & CLONE_NEWNET)
		return ERR_PTR(-EINVAL);
	return old_net;
156
}
157
#endif /* CONFIG_NET_NS */
158 159 160


extern struct list_head net_namespace_list;
161

162 163
struct net *get_net_ns_by_pid(pid_t pid);
struct net *get_net_ns_by_fd(int pid);
164

165 166 167 168 169 170 171 172
#ifdef CONFIG_SYSCTL
void ipx_register_sysctl(void);
void ipx_unregister_sysctl(void);
#else
#define ipx_register_sysctl()
#define ipx_unregister_sysctl()
#endif

173
#ifdef CONFIG_NET_NS
174
void __put_net(struct net *net);
175 176 177 178 179 180 181

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

182 183 184 185 186 187 188 189 190 191 192 193
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;
}

194 195 196 197 198 199
static inline void put_net(struct net *net)
{
	if (atomic_dec_and_test(&net->count))
		__put_net(net);
}

200 201 202 203 204
static inline
int net_eq(const struct net *net1, const struct net *net2)
{
	return net1 == net2;
}
205

206
void net_drop_ns(void *);
207

208
#else
209

210 211 212 213 214 215 216 217 218
static inline struct net *get_net(struct net *net)
{
	return net;
}

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

219 220 221 222 223 224 225 226 227 228
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;
}
229 230

#define net_drop_ns NULL
231 232 233 234
#endif


#ifdef NETNS_REFCNT_DEBUG
235 236
static inline struct net *hold_net(struct net *net)
{
237 238
	if (net)
		atomic_inc(&net->use_count);
239 240 241 242 243
	return net;
}

static inline void release_net(struct net *net)
{
244 245
	if (net)
		atomic_dec(&net->use_count);
246
}
247 248
#else
static inline struct net *hold_net(struct net *net)
249 250 251
{
	return net;
}
252

253
static inline void release_net(struct net *net)
254 255
{
}
256
#endif
257

258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
#ifdef CONFIG_NET_NS

static inline void write_pnet(struct net **pnet, struct net *net)
{
	*pnet = net;
}

static inline struct net *read_pnet(struct net * const *pnet)
{
	return *pnet;
}

#else

#define write_pnet(pnet, net)	do { (void)(net);} while (0)
#define read_pnet(pnet)		(&init_net)

#endif
276

277 278 279
#define for_each_net(VAR)				\
	list_for_each_entry(VAR, &net_namespace_list, list)

280 281 282
#define for_each_net_rcu(VAR)				\
	list_for_each_entry_rcu(VAR, &net_namespace_list, list)

283 284 285
#ifdef CONFIG_NET_NS
#define __net_init
#define __net_exit
286
#define __net_initdata
287
#define __net_initconst
288 289 290
#else
#define __net_init	__init
#define __net_exit	__exit_refok
291
#define __net_initdata	__initdata
292
#define __net_initconst	__initconst
293
#endif
294 295 296 297 298

struct pernet_operations {
	struct list_head list;
	int (*init)(struct net *net);
	void (*exit)(struct net *net);
299
	void (*exit_batch)(struct list_head *net_exit_list);
300 301
	int *id;
	size_t size;
302 303
};

Eric W. Biederman's avatar
Eric W. Biederman committed
304 305 306 307 308
/*
 * 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.
 *
309 310 311 312 313 314 315 316 317
 * 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.
Eric W. Biederman's avatar
Eric W. Biederman committed
318 319 320 321 322
 *
 * 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.
 */
323 324 325 326
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 *);
327

328 329
struct ctl_table;
struct ctl_table_header;
330

331
#ifdef CONFIG_SYSCTL
332 333 334 335
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);
336 337 338 339 340 341 342 343 344 345 346 347
#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

fan.du's avatar
fan.du committed
348
static inline int rt_genid_ipv4(struct net *net)
349
{
fan.du's avatar
fan.du committed
350
	return atomic_read(&net->ipv4.rt_genid);
351 352
}

fan.du's avatar
fan.du committed
353
static inline void rt_genid_bump_ipv4(struct net *net)
354
{
fan.du's avatar
fan.du committed
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383
	atomic_inc(&net->ipv4.rt_genid);
}

#if IS_ENABLED(CONFIG_IPV6)
static inline int rt_genid_ipv6(struct net *net)
{
	return atomic_read(&net->ipv6.rt_genid);
}

static inline void rt_genid_bump_ipv6(struct net *net)
{
	atomic_inc(&net->ipv6.rt_genid);
}
#else
static inline int rt_genid_ipv6(struct net *net)
{
	return 0;
}

static inline void rt_genid_bump_ipv6(struct net *net)
{
}
#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);
384
}
385

386 387 388 389 390 391 392 393 394 395
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);
}

396
#endif /* __NET_NET_NAMESPACE_H */