sock.h 59.1 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
2
3
4
5
6
7
8
9
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Definitions for the AF_INET socket handler.
 *
 * Version:	@(#)sock.h	1.0.4	05/13/93
 *
10
 * Authors:	Ross Biro
Linus Torvalds's avatar
Linus Torvalds committed
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Corey Minyard <wf-rch!minyard@relay.EU.net>
 *		Florian La Roche <flla@stud.uni-sb.de>
 *
 * Fixes:
 *		Alan Cox	:	Volatiles in skbuff pointers. See
 *					skbuff comments. May be overdone,
 *					better to prove they can be removed
 *					than the reverse.
 *		Alan Cox	:	Added a zapped field for tcp to note
 *					a socket is reset and must stay shut up
 *		Alan Cox	:	New fields for options
 *	Pauline Middelink	:	identd support
 *		Alan Cox	:	Eliminate low level recv/recvfrom
 *		David S. Miller	:	New socket lookup architecture.
 *              Steve Whitehouse:       Default routines for sock_ops
 *              Arnaldo C. Melo :	removed net_pinfo, tp_pinfo and made
 *              			protinfo be just a void pointer, as the
 *              			protocol specific parts were moved to
 *              			respective headers and ipv4/v6, etc now
 *              			use private slabcaches for its socks
 *              Pedro Hortas	:	New flags field for socket options
 *
 *
 *		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; either version
 *		2 of the License, or (at your option) any later version.
 */
#ifndef _SOCK_H
#define _SOCK_H

43
#include <linux/hardirq.h>
44
#include <linux/kernel.h>
Linus Torvalds's avatar
Linus Torvalds committed
45
#include <linux/list.h>
46
#include <linux/list_nulls.h>
Linus Torvalds's avatar
Linus Torvalds committed
47
48
#include <linux/timer.h>
#include <linux/cache.h>
49
#include <linux/lockdep.h>
Linus Torvalds's avatar
Linus Torvalds committed
50
51
#include <linux/netdevice.h>
#include <linux/skbuff.h>	/* struct sk_buff */
Al Viro's avatar
Al Viro committed
52
#include <linux/mm.h>
Linus Torvalds's avatar
Linus Torvalds committed
53
#include <linux/security.h>
54
#include <linux/slab.h>
55
#include <linux/uaccess.h>
56
#include <linux/memcontrol.h>
Glauber Costa's avatar
Glauber Costa committed
57
#include <linux/res_counter.h>
Linus Torvalds's avatar
Linus Torvalds committed
58
59

#include <linux/filter.h>
60
#include <linux/rculist_nulls.h>
61
#include <linux/poll.h>
Linus Torvalds's avatar
Linus Torvalds committed
62

63
#include <linux/atomic.h>
Linus Torvalds's avatar
Linus Torvalds committed
64
65
66
#include <net/dst.h>
#include <net/checksum.h>

67
68
struct cgroup;
struct cgroup_subsys;
69
#ifdef CONFIG_NET
Glauber Costa's avatar
Glauber Costa committed
70
71
int mem_cgroup_sockets_init(struct cgroup *cgrp, struct cgroup_subsys *ss);
void mem_cgroup_sockets_destroy(struct cgroup *cgrp, struct cgroup_subsys *ss);
72
73
74
75
76
77
78
79
80
81
82
#else
static inline
int mem_cgroup_sockets_init(struct cgroup *cgrp, struct cgroup_subsys *ss)
{
	return 0;
}
static inline
void mem_cgroup_sockets_destroy(struct cgroup *cgrp, struct cgroup_subsys *ss)
{
}
#endif
Linus Torvalds's avatar
Linus Torvalds committed
83
84
85
86
87
88
89
90
91
92
93
94
/*
 * This structure really needs to be cleaned up.
 * Most of it is for TCP, and not used by any of
 * the other protocols.
 */

/* Define this to get the SOCK_DBG debugging facility. */
#define SOCK_DEBUGGING
#ifdef SOCK_DEBUGGING
#define SOCK_DEBUG(sk, msg...) do { if ((sk) && sock_flag((sk), SOCK_DBG)) \
					printk(KERN_DEBUG msg); } while (0)
#else
95
/* Validate arguments and do nothing */
96
97
static inline __printf(2, 3)
void SOCK_DEBUG(struct sock *sk, const char *msg, ...)
98
99
{
}
Linus Torvalds's avatar
Linus Torvalds committed
100
101
102
103
104
105
106
107
#endif

/* This is the per-socket lock.  The spinlock provides a synchronization
 * between user contexts and software interrupt processing, whereas the
 * mini-semaphore synchronizes multiple users amongst themselves.
 */
typedef struct {
	spinlock_t		slock;
108
	int			owned;
Linus Torvalds's avatar
Linus Torvalds committed
109
	wait_queue_head_t	wq;
110
111
112
113
114
115
116
117
118
	/*
	 * We express the mutex-alike socket_lock semantics
	 * to the lock validator by explicitly managing
	 * the slock as a lock variant (in addition to
	 * the slock itself):
	 */
#ifdef CONFIG_DEBUG_LOCK_ALLOC
	struct lockdep_map dep_map;
#endif
Linus Torvalds's avatar
Linus Torvalds committed
119
120
121
} socket_lock_t;

struct sock;
122
struct proto;
123
struct net;
Linus Torvalds's avatar
Linus Torvalds committed
124
125

/**
126
 *	struct sock_common - minimal network layer representation of sockets
127
128
 *	@skc_daddr: Foreign IPv4 addr
 *	@skc_rcv_saddr: Bound local IPv4 addr
Eric Dumazet's avatar
Eric Dumazet committed
129
 *	@skc_hash: hash value used with various protocol lookup tables
130
 *	@skc_u16hashes: two u16 hash values used by UDP lookup tables
131
132
133
134
135
 *	@skc_family: network address family
 *	@skc_state: Connection state
 *	@skc_reuse: %SO_REUSEADDR setting
 *	@skc_bound_dev_if: bound device index if != 0
 *	@skc_bind_node: bind hash linkage for various protocol lookup tables
136
 *	@skc_portaddr_node: second hash linkage for UDP/UDP-Lite protocol
137
 *	@skc_prot: protocol handlers inside a network family
138
 *	@skc_net: reference to the network namespace of this socket
139
140
141
142
 *	@skc_node: main hash linkage for various protocol lookup tables
 *	@skc_nulls_node: main hash linkage for TCP/UDP/UDP-Lite protocol
 *	@skc_tx_queue_mapping: tx queue number for this connection
 *	@skc_refcnt: reference count
143
144
 *
 *	This is the minimal network layer representation of sockets, the header
145
146
 *	for struct sock and struct inet_timewait_sock.
 */
Linus Torvalds's avatar
Linus Torvalds committed
147
struct sock_common {
148
149
	/* skc_daddr and skc_rcv_saddr must be grouped :
	 * cf INET_MATCH() and INET_TW_MATCH()
Eric Dumazet's avatar
Eric Dumazet committed
150
	 */
151
152
	__be32			skc_daddr;
	__be32			skc_rcv_saddr;
Eric Dumazet's avatar
Eric Dumazet committed
153

154
155
156
157
	union  {
		unsigned int	skc_hash;
		__u16		skc_u16hashes[2];
	};
Eric Dumazet's avatar
Eric Dumazet committed
158
159
160
161
	unsigned short		skc_family;
	volatile unsigned char	skc_state;
	unsigned char		skc_reuse;
	int			skc_bound_dev_if;
162
163
164
165
	union {
		struct hlist_node	skc_bind_node;
		struct hlist_nulls_node skc_portaddr_node;
	};
166
	struct proto		*skc_prot;
167
#ifdef CONFIG_NET_NS
168
	struct net	 	*skc_net;
169
#endif
170
171
172
173
	/*
	 * fields between dontcopy_begin/dontcopy_end
	 * are not copied in sock_copy()
	 */
174
	/* private: */
175
	int			skc_dontcopy_begin[0];
176
	/* public: */
177
178
179
180
181
182
	union {
		struct hlist_node	skc_node;
		struct hlist_nulls_node skc_nulls_node;
	};
	int			skc_tx_queue_mapping;
	atomic_t		skc_refcnt;
183
	/* private: */
184
	int                     skc_dontcopy_end[0];
185
	/* public: */
Linus Torvalds's avatar
Linus Torvalds committed
186
187
};

Glauber Costa's avatar
Glauber Costa committed
188
struct cg_proto;
Linus Torvalds's avatar
Linus Torvalds committed
189
190
/**
  *	struct sock - network layer representation of sockets
191
  *	@__sk_common: shared layout with inet_timewait_sock
192
193
194
195
  *	@sk_shutdown: mask of %SEND_SHUTDOWN and/or %RCV_SHUTDOWN
  *	@sk_userlocks: %SO_SNDBUF and %SO_RCVBUF settings
  *	@sk_lock:	synchronizer
  *	@sk_rcvbuf: size of receive buffer in bytes
196
  *	@sk_wq: sock wait queue and async head
197
198
199
200
201
202
  *	@sk_dst_cache: destination cache
  *	@sk_dst_lock: destination cache lock
  *	@sk_policy: flow policy
  *	@sk_receive_queue: incoming packets
  *	@sk_wmem_alloc: transmit queue bytes committed
  *	@sk_write_queue: Packet sending queue
203
  *	@sk_async_wait_queue: DMA copied packets
204
205
206
207
208
  *	@sk_omem_alloc: "o" is "option" or "other"
  *	@sk_wmem_queued: persistent queue size
  *	@sk_forward_alloc: space allocated forward
  *	@sk_allocation: allocation mode
  *	@sk_sndbuf: size of send buffer in bytes
Wang Chen's avatar
Wang Chen committed
209
  *	@sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE,
210
  *		   %SO_OOBINLINE settings, %SO_TIMESTAMPING settings
211
212
  *	@sk_no_check: %SO_NO_CHECK setting, wether or not checkup packets
  *	@sk_route_caps: route capabilities (e.g. %NETIF_F_TSO)
Eric Dumazet's avatar
Eric Dumazet committed
213
  *	@sk_route_nocaps: forbidden route capabilities (e.g NETIF_F_GSO_MASK)
214
  *	@sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4)
215
  *	@sk_gso_max_size: Maximum GSO segment size to build
216
217
218
219
  *	@sk_lingertime: %SO_LINGER l_linger setting
  *	@sk_backlog: always used with the per-socket spinlock held
  *	@sk_callback_lock: used with the callbacks in the end of this struct
  *	@sk_error_queue: rarely used
Wang Chen's avatar
Wang Chen committed
220
221
  *	@sk_prot_creator: sk_prot of original sock creator (see ipv6_setsockopt,
  *			  IPV6_ADDRFORM for instance)
222
  *	@sk_err: last error
Wang Chen's avatar
Wang Chen committed
223
224
  *	@sk_err_soft: errors that don't cause failure but are the cause of a
  *		      persistent failure not just 'timed out'
Eric Dumazet's avatar
Eric Dumazet committed
225
  *	@sk_drops: raw/udp drops counter
226
227
228
229
230
  *	@sk_ack_backlog: current listen backlog
  *	@sk_max_ack_backlog: listen backlog set in listen()
  *	@sk_priority: %SO_PRIORITY setting
  *	@sk_type: socket type (%SOCK_STREAM, etc)
  *	@sk_protocol: which protocol this socket belongs in this network family
231
232
  *	@sk_peer_pid: &struct pid for this socket's peer
  *	@sk_peer_cred: %SO_PEERCRED setting
233
234
235
  *	@sk_rcvlowat: %SO_RCVLOWAT setting
  *	@sk_rcvtimeo: %SO_RCVTIMEO setting
  *	@sk_sndtimeo: %SO_SNDTIMEO setting
236
  *	@sk_rxhash: flow hash received from netif layer
237
238
239
240
241
242
243
244
245
  *	@sk_filter: socket filtering instructions
  *	@sk_protinfo: private area, net family specific, when not using slab
  *	@sk_timer: sock cleanup timer
  *	@sk_stamp: time stamp of last packet received
  *	@sk_socket: Identd and reporting IO signals
  *	@sk_user_data: RPC layer private data
  *	@sk_sndmsg_page: cached page for sendmsg
  *	@sk_sndmsg_off: cached offset for sendmsg
  *	@sk_send_head: front of stuff to transmit
246
  *	@sk_security: used by security modules
247
  *	@sk_mark: generic packet mark
248
  *	@sk_classid: this socket's cgroup classid
Glauber Costa's avatar
Glauber Costa committed
249
  *	@sk_cgrp: this socket's cgroup-specific proto data
250
251
252
253
254
255
256
  *	@sk_write_pending: a write to stream socket waits to start
  *	@sk_state_change: callback to indicate change in the state of the sock
  *	@sk_data_ready: callback to indicate there is data to be processed
  *	@sk_write_space: callback to indicate there is bf sending space available
  *	@sk_error_report: callback to indicate errors (e.g. %MSG_ERRQUEUE)
  *	@sk_backlog_rcv: callback to process the backlog
  *	@sk_destruct: called at sock freeing time, i.e. when all refcnt == 0
Linus Torvalds's avatar
Linus Torvalds committed
257
258
259
 */
struct sock {
	/*
260
	 * Now struct inet_timewait_sock also uses sock_common, so please just
Linus Torvalds's avatar
Linus Torvalds committed
261
262
263
	 * don't add nothing before this first member (__sk_common) --acme
	 */
	struct sock_common	__sk_common;
Eric Dumazet's avatar
Eric Dumazet committed
264
265
266
#define sk_node			__sk_common.skc_node
#define sk_nulls_node		__sk_common.skc_nulls_node
#define sk_refcnt		__sk_common.skc_refcnt
267
#define sk_tx_queue_mapping	__sk_common.skc_tx_queue_mapping
Eric Dumazet's avatar
Eric Dumazet committed
268

269
270
#define sk_dontcopy_begin	__sk_common.skc_dontcopy_begin
#define sk_dontcopy_end		__sk_common.skc_dontcopy_end
Eric Dumazet's avatar
Eric Dumazet committed
271
#define sk_hash			__sk_common.skc_hash
Linus Torvalds's avatar
Linus Torvalds committed
272
273
274
275
276
#define sk_family		__sk_common.skc_family
#define sk_state		__sk_common.skc_state
#define sk_reuse		__sk_common.skc_reuse
#define sk_bound_dev_if		__sk_common.skc_bound_dev_if
#define sk_bind_node		__sk_common.skc_bind_node
277
#define sk_prot			__sk_common.skc_prot
278
#define sk_net			__sk_common.skc_net
Linus Torvalds's avatar
Linus Torvalds committed
279
	socket_lock_t		sk_lock;
Eric Dumazet's avatar
Eric Dumazet committed
280
	struct sk_buff_head	sk_receive_queue;
281
282
283
284
	/*
	 * The backlog queue is special, it is always used with
	 * the per-socket spinlock held and requires low latency
	 * access. Therefore we special case it's implementation.
Eric Dumazet's avatar
Eric Dumazet committed
285
286
287
	 * Note : rmem_alloc is in this structure to fill a hole
	 * on 64bit arches, not because its logically part of
	 * backlog.
288
289
	 */
	struct {
Eric Dumazet's avatar
Eric Dumazet committed
290
291
292
293
		atomic_t	rmem_alloc;
		int		len;
		struct sk_buff	*head;
		struct sk_buff	*tail;
294
	} sk_backlog;
Eric Dumazet's avatar
Eric Dumazet committed
295
296
297
298
299
300
301
302
303
#define sk_rmem_alloc sk_backlog.rmem_alloc
	int			sk_forward_alloc;
#ifdef CONFIG_RPS
	__u32			sk_rxhash;
#endif
	atomic_t		sk_drops;
	int			sk_rcvbuf;

	struct sk_filter __rcu	*sk_filter;
304
	struct socket_wq __rcu	*sk_wq;
Eric Dumazet's avatar
Eric Dumazet committed
305
306
307
308
309

#ifdef CONFIG_NET_DMA
	struct sk_buff_head	sk_async_wait_queue;
#endif

310
#ifdef CONFIG_XFRM
Linus Torvalds's avatar
Linus Torvalds committed
311
	struct xfrm_policy	*sk_policy[2];
312
#endif
Eric Dumazet's avatar
Eric Dumazet committed
313
314
	unsigned long 		sk_flags;
	struct dst_entry	*sk_dst_cache;
Eric Dumazet's avatar
Eric Dumazet committed
315
	spinlock_t		sk_dst_lock;
Linus Torvalds's avatar
Linus Torvalds committed
316
317
	atomic_t		sk_wmem_alloc;
	atomic_t		sk_omem_alloc;
318
	int			sk_sndbuf;
Linus Torvalds's avatar
Linus Torvalds committed
319
	struct sk_buff_head	sk_write_queue;
Eric Dumazet's avatar
Eric Dumazet committed
320
321
322
323
324
325
326
	kmemcheck_bitfield_begin(flags);
	unsigned int		sk_shutdown  : 2,
				sk_no_check  : 2,
				sk_userlocks : 4,
				sk_protocol  : 8,
				sk_type      : 16;
	kmemcheck_bitfield_end(flags);
Linus Torvalds's avatar
Linus Torvalds committed
327
	int			sk_wmem_queued;
Al Viro's avatar
Al Viro committed
328
	gfp_t			sk_allocation;
329
330
	netdev_features_t	sk_route_caps;
	netdev_features_t	sk_route_nocaps;
331
	int			sk_gso_type;
332
	unsigned int		sk_gso_max_size;
333
	int			sk_rcvlowat;
Linus Torvalds's avatar
Linus Torvalds committed
334
335
	unsigned long	        sk_lingertime;
	struct sk_buff_head	sk_error_queue;
336
	struct proto		*sk_prot_creator;
Linus Torvalds's avatar
Linus Torvalds committed
337
338
339
340
341
342
	rwlock_t		sk_callback_lock;
	int			sk_err,
				sk_err_soft;
	unsigned short		sk_ack_backlog;
	unsigned short		sk_max_ack_backlog;
	__u32			sk_priority;
343
344
345
#ifdef CONFIG_CGROUPS
	__u32			sk_cgrp_prioidx;
#endif
346
347
	struct pid		*sk_peer_pid;
	const struct cred	*sk_peer_cred;
Linus Torvalds's avatar
Linus Torvalds committed
348
349
350
351
	long			sk_rcvtimeo;
	long			sk_sndtimeo;
	void			*sk_protinfo;
	struct timer_list	sk_timer;
352
	ktime_t			sk_stamp;
Linus Torvalds's avatar
Linus Torvalds committed
353
354
355
356
357
358
	struct socket		*sk_socket;
	void			*sk_user_data;
	struct page		*sk_sndmsg_page;
	struct sk_buff		*sk_send_head;
	__u32			sk_sndmsg_off;
	int			sk_write_pending;
Alexey Dobriyan's avatar
Alexey Dobriyan committed
359
#ifdef CONFIG_SECURITY
Linus Torvalds's avatar
Linus Torvalds committed
360
	void			*sk_security;
Alexey Dobriyan's avatar
Alexey Dobriyan committed
361
#endif
362
	__u32			sk_mark;
363
	u32			sk_classid;
Glauber Costa's avatar
Glauber Costa committed
364
	struct cg_proto		*sk_cgrp;
Linus Torvalds's avatar
Linus Torvalds committed
365
366
367
368
369
370
371
372
373
374
375
376
	void			(*sk_state_change)(struct sock *sk);
	void			(*sk_data_ready)(struct sock *sk, int bytes);
	void			(*sk_write_space)(struct sock *sk);
	void			(*sk_error_report)(struct sock *sk);
  	int			(*sk_backlog_rcv)(struct sock *sk,
						  struct sk_buff *skb);  
	void                    (*sk_destruct)(struct sock *sk);
};

/*
 * Hashed lists helper routines
 */
Li Zefan's avatar
Li Zefan committed
377
378
379
380
381
static inline struct sock *sk_entry(const struct hlist_node *node)
{
	return hlist_entry(node, struct sock, sk_node);
}

382
static inline struct sock *__sk_head(const struct hlist_head *head)
Linus Torvalds's avatar
Linus Torvalds committed
383
384
385
386
{
	return hlist_entry(head->first, struct sock, sk_node);
}

387
static inline struct sock *sk_head(const struct hlist_head *head)
Linus Torvalds's avatar
Linus Torvalds committed
388
389
390
391
{
	return hlist_empty(head) ? NULL : __sk_head(head);
}

392
393
394
395
396
397
398
399
400
401
static inline struct sock *__sk_nulls_head(const struct hlist_nulls_head *head)
{
	return hlist_nulls_entry(head->first, struct sock, sk_nulls_node);
}

static inline struct sock *sk_nulls_head(const struct hlist_nulls_head *head)
{
	return hlist_nulls_empty(head) ? NULL : __sk_nulls_head(head);
}

402
static inline struct sock *sk_next(const struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
403
404
405
406
407
{
	return sk->sk_node.next ?
		hlist_entry(sk->sk_node.next, struct sock, sk_node) : NULL;
}

408
409
410
411
412
413
414
415
static inline struct sock *sk_nulls_next(const struct sock *sk)
{
	return (!is_a_nulls(sk->sk_nulls_node.next)) ?
		hlist_nulls_entry(sk->sk_nulls_node.next,
				  struct sock, sk_nulls_node) :
		NULL;
}

416
static inline int sk_unhashed(const struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
417
418
419
420
{
	return hlist_unhashed(&sk->sk_node);
}

421
static inline int sk_hashed(const struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
422
{
Akinobu Mita's avatar
Akinobu Mita committed
423
	return !sk_unhashed(sk);
Linus Torvalds's avatar
Linus Torvalds committed
424
425
426
427
428
429
430
}

static __inline__ void sk_node_init(struct hlist_node *node)
{
	node->pprev = NULL;
}

431
432
433
434
435
static __inline__ void sk_nulls_node_init(struct hlist_nulls_node *node)
{
	node->pprev = NULL;
}

Linus Torvalds's avatar
Linus Torvalds committed
436
437
438
439
440
static __inline__ void __sk_del_node(struct sock *sk)
{
	__hlist_del(&sk->sk_node);
}

441
/* NB: equivalent to hlist_del_init_rcu */
Linus Torvalds's avatar
Linus Torvalds committed
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
static __inline__ int __sk_del_node_init(struct sock *sk)
{
	if (sk_hashed(sk)) {
		__sk_del_node(sk);
		sk_node_init(&sk->sk_node);
		return 1;
	}
	return 0;
}

/* Grab socket reference count. This operation is valid only
   when sk is ALREADY grabbed f.e. it is found in hash table
   or a list and the lookup is made under lock preventing hash table
   modifications.
 */

static inline void sock_hold(struct sock *sk)
{
	atomic_inc(&sk->sk_refcnt);
}

/* Ungrab socket in the context, which assumes that socket refcnt
   cannot hit zero, f.e. it is true in context of any socketcall.
 */
static inline void __sock_put(struct sock *sk)
{
	atomic_dec(&sk->sk_refcnt);
}

static __inline__ int sk_del_node_init(struct sock *sk)
{
	int rc = __sk_del_node_init(sk);

	if (rc) {
		/* paranoid for a while -acme */
		WARN_ON(atomic_read(&sk->sk_refcnt) == 1);
		__sock_put(sk);
	}
	return rc;
}
482
#define sk_del_node_init_rcu(sk)	sk_del_node_init(sk)
Linus Torvalds's avatar
Linus Torvalds committed
483

484
static __inline__ int __sk_nulls_del_node_init_rcu(struct sock *sk)
485
486
{
	if (sk_hashed(sk)) {
487
		hlist_nulls_del_init_rcu(&sk->sk_nulls_node);
488
489
490
491
492
		return 1;
	}
	return 0;
}

493
static __inline__ int sk_nulls_del_node_init_rcu(struct sock *sk)
494
{
495
	int rc = __sk_nulls_del_node_init_rcu(sk);
496
497
498
499
500
501
502
503
504

	if (rc) {
		/* paranoid for a while -acme */
		WARN_ON(atomic_read(&sk->sk_refcnt) == 1);
		__sock_put(sk);
	}
	return rc;
}

Linus Torvalds's avatar
Linus Torvalds committed
505
506
507
508
509
510
511
512
513
514
515
static __inline__ void __sk_add_node(struct sock *sk, struct hlist_head *list)
{
	hlist_add_head(&sk->sk_node, list);
}

static __inline__ void sk_add_node(struct sock *sk, struct hlist_head *list)
{
	sock_hold(sk);
	__sk_add_node(sk, list);
}

516
517
518
519
520
521
static __inline__ void sk_add_node_rcu(struct sock *sk, struct hlist_head *list)
{
	sock_hold(sk);
	hlist_add_head_rcu(&sk->sk_node, list);
}

522
static __inline__ void __sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list)
523
{
524
	hlist_nulls_add_head_rcu(&sk->sk_nulls_node, list);
525
526
}

527
static __inline__ void sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list)
528
529
{
	sock_hold(sk);
530
	__sk_nulls_add_node_rcu(sk, list);
531
532
}

Linus Torvalds's avatar
Linus Torvalds committed
533
534
535
536
537
538
539
540
541
542
543
544
545
static __inline__ void __sk_del_bind_node(struct sock *sk)
{
	__hlist_del(&sk->sk_bind_node);
}

static __inline__ void sk_add_bind_node(struct sock *sk,
					struct hlist_head *list)
{
	hlist_add_head(&sk->sk_bind_node, list);
}

#define sk_for_each(__sk, node, list) \
	hlist_for_each_entry(__sk, node, list, sk_node)
546
547
#define sk_for_each_rcu(__sk, node, list) \
	hlist_for_each_entry_rcu(__sk, node, list, sk_node)
548
549
550
551
#define sk_nulls_for_each(__sk, node, list) \
	hlist_nulls_for_each_entry(__sk, node, list, sk_nulls_node)
#define sk_nulls_for_each_rcu(__sk, node, list) \
	hlist_nulls_for_each_entry_rcu(__sk, node, list, sk_nulls_node)
Linus Torvalds's avatar
Linus Torvalds committed
552
553
554
#define sk_for_each_from(__sk, node) \
	if (__sk && ({ node = &(__sk)->sk_node; 1; })) \
		hlist_for_each_entry_from(__sk, node, sk_node)
555
556
557
#define sk_nulls_for_each_from(__sk, node) \
	if (__sk && ({ node = &(__sk)->sk_nulls_node; 1; })) \
		hlist_nulls_for_each_entry_from(__sk, node, sk_nulls_node)
Linus Torvalds's avatar
Linus Torvalds committed
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
#define sk_for_each_safe(__sk, node, tmp, list) \
	hlist_for_each_entry_safe(__sk, node, tmp, list, sk_node)
#define sk_for_each_bound(__sk, node, list) \
	hlist_for_each_entry(__sk, node, list, sk_bind_node)

/* Sock flags */
enum sock_flags {
	SOCK_DEAD,
	SOCK_DONE,
	SOCK_URGINLINE,
	SOCK_KEEPOPEN,
	SOCK_LINGER,
	SOCK_DESTROY,
	SOCK_BROADCAST,
	SOCK_TIMESTAMP,
	SOCK_ZAPPED,
	SOCK_USE_WRITE_QUEUE, /* whether to call sk->sk_write_space in sock_wfree */
	SOCK_DBG, /* %SO_DEBUG setting */
	SOCK_RCVTSTAMP, /* %SO_TIMESTAMP setting */
577
	SOCK_RCVTSTAMPNS, /* %SO_TIMESTAMPNS setting */
Linus Torvalds's avatar
Linus Torvalds committed
578
579
	SOCK_LOCALROUTE, /* route locally only, %SO_DONTROUTE setting */
	SOCK_QUEUE_SHRUNK, /* write queue has been shrunk recently */
580
581
582
583
584
585
586
	SOCK_TIMESTAMPING_TX_HARDWARE,  /* %SOF_TIMESTAMPING_TX_HARDWARE */
	SOCK_TIMESTAMPING_TX_SOFTWARE,  /* %SOF_TIMESTAMPING_TX_SOFTWARE */
	SOCK_TIMESTAMPING_RX_HARDWARE,  /* %SOF_TIMESTAMPING_RX_HARDWARE */
	SOCK_TIMESTAMPING_RX_SOFTWARE,  /* %SOF_TIMESTAMPING_RX_SOFTWARE */
	SOCK_TIMESTAMPING_SOFTWARE,     /* %SOF_TIMESTAMPING_SOFTWARE */
	SOCK_TIMESTAMPING_RAW_HARDWARE, /* %SOF_TIMESTAMPING_RAW_HARDWARE */
	SOCK_TIMESTAMPING_SYS_HARDWARE, /* %SOF_TIMESTAMPING_SYS_HARDWARE */
Eric Dumazet's avatar
Eric Dumazet committed
587
	SOCK_FASYNC, /* fasync() active */
588
	SOCK_RXQ_OVFL,
589
	SOCK_ZEROCOPY, /* buffers from userspace */
590
	SOCK_WIFI_STATUS, /* push wifi status to userspace */
Linus Torvalds's avatar
Linus Torvalds committed
591
592
};

Ralf Baechle's avatar
Ralf Baechle committed
593
594
595
596
597
static inline void sock_copy_flags(struct sock *nsk, struct sock *osk)
{
	nsk->sk_flags = osk->sk_flags;
}

Linus Torvalds's avatar
Linus Torvalds committed
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
static inline void sock_set_flag(struct sock *sk, enum sock_flags flag)
{
	__set_bit(flag, &sk->sk_flags);
}

static inline void sock_reset_flag(struct sock *sk, enum sock_flags flag)
{
	__clear_bit(flag, &sk->sk_flags);
}

static inline int sock_flag(struct sock *sk, enum sock_flags flag)
{
	return test_bit(flag, &sk->sk_flags);
}

static inline void sk_acceptq_removed(struct sock *sk)
{
	sk->sk_ack_backlog--;
}

static inline void sk_acceptq_added(struct sock *sk)
{
	sk->sk_ack_backlog++;
}

static inline int sk_acceptq_is_full(struct sock *sk)
{
625
	return sk->sk_ack_backlog > sk->sk_max_ack_backlog;
Linus Torvalds's avatar
Linus Torvalds committed
626
627
628
629
630
631
632
}

/*
 * Compute minimal free write space needed to queue new packets.
 */
static inline int sk_stream_min_wspace(struct sock *sk)
{
633
	return sk->sk_wmem_queued >> 1;
Linus Torvalds's avatar
Linus Torvalds committed
634
635
636
637
638
639
640
641
642
643
644
645
646
647
}

static inline int sk_stream_wspace(struct sock *sk)
{
	return sk->sk_sndbuf - sk->sk_wmem_queued;
}

extern void sk_stream_write_space(struct sock *sk);

static inline int sk_stream_memory_free(struct sock *sk)
{
	return sk->sk_wmem_queued < sk->sk_sndbuf;
}

Zhu Yi's avatar
Zhu Yi committed
648
/* OOB backlog add */
Zhu Yi's avatar
Zhu Yi committed
649
static inline void __sk_add_backlog(struct sock *sk, struct sk_buff *skb)
650
{
Eric Dumazet's avatar
Eric Dumazet committed
651
652
653
654
655
656
	/* dont let skb dst not refcounted, we are going to leave rcu lock */
	skb_dst_force(skb);

	if (!sk->sk_backlog.tail)
		sk->sk_backlog.head = skb;
	else
657
		sk->sk_backlog.tail->next = skb;
Eric Dumazet's avatar
Eric Dumazet committed
658
659

	sk->sk_backlog.tail = skb;
660
661
	skb->next = NULL;
}
Linus Torvalds's avatar
Linus Torvalds committed
662

663
664
/*
 * Take into account size of receive queue and backlog queue
Eric Dumazet's avatar
Eric Dumazet committed
665
666
 * Do not take into account this skb truesize,
 * to allow even a single big packet to come.
667
668
669
670
671
 */
static inline bool sk_rcvqueues_full(const struct sock *sk, const struct sk_buff *skb)
{
	unsigned int qsize = sk->sk_backlog.len + atomic_read(&sk->sk_rmem_alloc);

Eric Dumazet's avatar
Eric Dumazet committed
672
	return qsize > sk->sk_rcvbuf;
673
674
}

Zhu Yi's avatar
Zhu Yi committed
675
/* The per-socket spinlock must be held here. */
676
static inline __must_check int sk_add_backlog(struct sock *sk, struct sk_buff *skb)
Zhu Yi's avatar
Zhu Yi committed
677
{
678
	if (sk_rcvqueues_full(sk, skb))
Zhu Yi's avatar
Zhu Yi committed
679
680
		return -ENOBUFS;

Zhu Yi's avatar
Zhu Yi committed
681
	__sk_add_backlog(sk, skb);
Zhu Yi's avatar
Zhu Yi committed
682
683
684
685
	sk->sk_backlog.len += skb->truesize;
	return 0;
}

686
687
688
689
690
static inline int sk_backlog_rcv(struct sock *sk, struct sk_buff *skb)
{
	return sk->sk_backlog_rcv(sk, skb);
}

691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
static inline void sock_rps_record_flow(const struct sock *sk)
{
#ifdef CONFIG_RPS
	struct rps_sock_flow_table *sock_flow_table;

	rcu_read_lock();
	sock_flow_table = rcu_dereference(rps_sock_flow_table);
	rps_record_sock_flow(sock_flow_table, sk->sk_rxhash);
	rcu_read_unlock();
#endif
}

static inline void sock_rps_reset_flow(const struct sock *sk)
{
#ifdef CONFIG_RPS
	struct rps_sock_flow_table *sock_flow_table;

	rcu_read_lock();
	sock_flow_table = rcu_dereference(rps_sock_flow_table);
	rps_reset_sock_flow(sock_flow_table, sk->sk_rxhash);
	rcu_read_unlock();
#endif
}

715
716
static inline void sock_rps_save_rxhash(struct sock *sk,
					const struct sk_buff *skb)
717
718
{
#ifdef CONFIG_RPS
719
	if (unlikely(sk->sk_rxhash != skb->rxhash)) {
720
		sock_rps_reset_flow(sk);
721
		sk->sk_rxhash = skb->rxhash;
722
723
724
725
	}
#endif
}

726
727
728
729
730
731
732
733
static inline void sock_rps_reset_rxhash(struct sock *sk)
{
#ifdef CONFIG_RPS
	sock_rps_reset_flow(sk);
	sk->sk_rxhash = 0;
#endif
}

734
735
736
737
738
739
740
741
742
743
744
#define sk_wait_event(__sk, __timeo, __condition)			\
	({	int __rc;						\
		release_sock(__sk);					\
		__rc = __condition;					\
		if (!__rc) {						\
			*(__timeo) = schedule_timeout(*(__timeo));	\
		}							\
		lock_sock(__sk);					\
		__rc = __condition;					\
		__rc;							\
	})
Linus Torvalds's avatar
Linus Torvalds committed
745
746
747
748
749
750
751
752
753

extern int sk_stream_wait_connect(struct sock *sk, long *timeo_p);
extern int sk_stream_wait_memory(struct sock *sk, long *timeo_p);
extern void sk_stream_wait_close(struct sock *sk, long timeo_p);
extern int sk_stream_error(struct sock *sk, int flags, int err);
extern void sk_stream_kill_queues(struct sock *sk);

extern int sk_wait_data(struct sock *sk, long *timeo);

754
struct request_sock_ops;
755
struct timewait_sock_ops;
756
struct inet_hashinfo;
757
struct raw_hashinfo;
758
struct module;
759

Linus Torvalds's avatar
Linus Torvalds committed
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
/* Networking protocol blocks we attach to sockets.
 * socket layer -> transport layer interface
 * transport -> network interface is defined by struct inet_proto
 */
struct proto {
	void			(*close)(struct sock *sk, 
					long timeout);
	int			(*connect)(struct sock *sk,
				        struct sockaddr *uaddr, 
					int addr_len);
	int			(*disconnect)(struct sock *sk, int flags);

	struct sock *		(*accept) (struct sock *sk, int flags, int *err);

	int			(*ioctl)(struct sock *sk, int cmd,
					 unsigned long arg);
	int			(*init)(struct sock *sk);
777
	void			(*destroy)(struct sock *sk);
Linus Torvalds's avatar
Linus Torvalds committed
778
779
780
	void			(*shutdown)(struct sock *sk, int how);
	int			(*setsockopt)(struct sock *sk, int level, 
					int optname, char __user *optval,
781
					unsigned int optlen);
Linus Torvalds's avatar
Linus Torvalds committed
782
783
784
	int			(*getsockopt)(struct sock *sk, int level, 
					int optname, char __user *optval, 
					int __user *option);  	 
785
#ifdef CONFIG_COMPAT
786
787
788
	int			(*compat_setsockopt)(struct sock *sk,
					int level,
					int optname, char __user *optval,
789
					unsigned int optlen);
790
791
792
793
	int			(*compat_getsockopt)(struct sock *sk,
					int level,
					int optname, char __user *optval,
					int __user *option);
794
795
	int			(*compat_ioctl)(struct sock *sk,
					unsigned int cmd, unsigned long arg);
796
#endif
Linus Torvalds's avatar
Linus Torvalds committed
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
	int			(*sendmsg)(struct kiocb *iocb, struct sock *sk,
					   struct msghdr *msg, size_t len);
	int			(*recvmsg)(struct kiocb *iocb, struct sock *sk,
					   struct msghdr *msg,
					size_t len, int noblock, int flags, 
					int *addr_len);
	int			(*sendpage)(struct sock *sk, struct page *page,
					int offset, size_t size, int flags);
	int			(*bind)(struct sock *sk, 
					struct sockaddr *uaddr, int addr_len);

	int			(*backlog_rcv) (struct sock *sk, 
						struct sk_buff *skb);

	/* Keeping track of sk's, looking them up, and port selection methods. */
	void			(*hash)(struct sock *sk);
	void			(*unhash)(struct sock *sk);
Eric Dumazet's avatar
Eric Dumazet committed
814
	void			(*rehash)(struct sock *sk);
Linus Torvalds's avatar
Linus Torvalds committed
815
	int			(*get_port)(struct sock *sk, unsigned short snum);
816
	void			(*clear_sk)(struct sock *sk, int size);
Linus Torvalds's avatar
Linus Torvalds committed
817

818
	/* Keeping track of sockets in use */
819
#ifdef CONFIG_PROC_FS
820
	unsigned int		inuse_idx;
821
#endif
822

Linus Torvalds's avatar
Linus Torvalds committed
823
	/* Memory pressure */
824
	void			(*enter_memory_pressure)(struct sock *sk);
Eric Dumazet's avatar
Eric Dumazet committed
825
	atomic_long_t		*memory_allocated;	/* Current allocated memory. */
826
	struct percpu_counter	*sockets_allocated;	/* Current number of sockets. */
Linus Torvalds's avatar
Linus Torvalds committed
827
828
829
	/*
	 * Pressure flag: try to collapse.
	 * Technical note: it is used by multiple contexts non atomically.
830
	 * All the __sk_mem_schedule() is of this nature: accounting
Linus Torvalds's avatar
Linus Torvalds committed
831
832
833
	 * is strict, actions are advisory and have some latency.
	 */
	int			*memory_pressure;
Eric Dumazet's avatar
Eric Dumazet committed
834
	long			*sysctl_mem;
Linus Torvalds's avatar
Linus Torvalds committed
835
836
837
	int			*sysctl_wmem;
	int			*sysctl_rmem;
	int			max_header;
838
	bool			no_autobind;
Linus Torvalds's avatar
Linus Torvalds committed
839

840
	struct kmem_cache	*slab;
Linus Torvalds's avatar
Linus Torvalds committed
841
	unsigned int		obj_size;
842
	int			slab_flags;
Linus Torvalds's avatar
Linus Torvalds committed
843

844
	struct percpu_counter	*orphan_count;
845

846
	struct request_sock_ops	*rsk_prot;
847
	struct timewait_sock_ops *twsk_prot;
848

849
850
	union {
		struct inet_hashinfo	*hashinfo;
851
		struct udp_table	*udp_table;
852
		struct raw_hashinfo	*raw_hash;
853
	} h;
854

Linus Torvalds's avatar
Linus Torvalds committed
855
856
857
858
859
	struct module		*owner;

	char			name[32];

	struct list_head	node;
860
861
862
#ifdef SOCK_REFCNT_DEBUG
	atomic_t		socks;
#endif
Glauber Costa's avatar
Glauber Costa committed
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM
	/*
	 * cgroup specific init/deinit functions. Called once for all
	 * protocols that implement it, from cgroups populate function.
	 * This function has to setup any files the protocol want to
	 * appear in the kmem cgroup filesystem.
	 */
	int			(*init_cgroup)(struct cgroup *cgrp,
					       struct cgroup_subsys *ss);
	void			(*destroy_cgroup)(struct cgroup *cgrp,
						  struct cgroup_subsys *ss);
	struct cg_proto		*(*proto_cgroup)(struct mem_cgroup *memcg);
#endif
};

struct cg_proto {
	void			(*enter_memory_pressure)(struct sock *sk);
	struct res_counter	*memory_allocated;	/* Current allocated memory. */
	struct percpu_counter	*sockets_allocated;	/* Current number of sockets. */
	int			*memory_pressure;
	long			*sysctl_mem;
	/*
	 * memcg field is used to find which memcg we belong directly
	 * Each memcg struct can hold more than one cg_proto, so container_of
	 * won't really cut.
	 *
	 * The elegant solution would be having an inverse function to
	 * proto_cgroup in struct proto, but that means polluting the structure
	 * for everybody, instead of just for memcg users.
	 */
	struct mem_cgroup	*memcg;
Linus Torvalds's avatar
Linus Torvalds committed
894
895
896
897
898
};

extern int proto_register(struct proto *prot, int alloc_slab);
extern void proto_unregister(struct proto *prot);

899
900
901
902
903
904
905
906
907
908
909
910
911
#ifdef SOCK_REFCNT_DEBUG
static inline void sk_refcnt_debug_inc(struct sock *sk)
{
	atomic_inc(&sk->sk_prot->socks);
}

static inline void sk_refcnt_debug_dec(struct sock *sk)
{
	atomic_dec(&sk->sk_prot->socks);
	printk(KERN_DEBUG "%s socket %p released, %d are still alive\n",
	       sk->sk_prot->name, sk, atomic_read(&sk->sk_prot->socks));
}

Glauber Costa's avatar
Glauber Costa committed
912
inline void sk_refcnt_debug_release(const struct sock *sk)
913
914
915
916
917
918
919
920
921
922
923
{
	if (atomic_read(&sk->sk_refcnt) != 1)
		printk(KERN_DEBUG "Destruction of the %s socket %p delayed, refcnt=%d\n",
		       sk->sk_prot->name, sk, atomic_read(&sk->sk_refcnt));
}
#else /* SOCK_REFCNT_DEBUG */
#define sk_refcnt_debug_inc(sk) do { } while (0)
#define sk_refcnt_debug_dec(sk) do { } while (0)
#define sk_refcnt_debug_release(sk) do { } while (0)
#endif /* SOCK_REFCNT_DEBUG */

Glauber Costa's avatar
Glauber Costa committed
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM
extern struct jump_label_key memcg_socket_limit_enabled;
static inline struct cg_proto *parent_cg_proto(struct proto *proto,
					       struct cg_proto *cg_proto)
{
	return proto->proto_cgroup(parent_mem_cgroup(cg_proto->memcg));
}
#define mem_cgroup_sockets_enabled static_branch(&memcg_socket_limit_enabled)
#else
#define mem_cgroup_sockets_enabled 0
static inline struct cg_proto *parent_cg_proto(struct proto *proto,
					       struct cg_proto *cg_proto)
{
	return NULL;
}
#endif


942
943
944
945
946
947
948
949
950
static inline bool sk_has_memory_pressure(const struct sock *sk)
{
	return sk->sk_prot->memory_pressure != NULL;
}

static inline bool sk_under_memory_pressure(const struct sock *sk)
{
	if (!sk->sk_prot->memory_pressure)
		return false;
Glauber Costa's avatar
Glauber Costa committed
951
952
953
954

	if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
		return !!*sk->sk_cgrp->memory_pressure;

955
956
957
958
959
960
961
	return !!*sk->sk_prot->memory_pressure;
}

static inline void sk_leave_memory_pressure(struct sock *sk)
{
	int *memory_pressure = sk->sk_prot->memory_pressure;

Glauber Costa's avatar
Glauber Costa committed
962
963
964
965
	if (!memory_pressure)
		return;

	if (*memory_pressure)
966
		*memory_pressure = 0;
Glauber Costa's avatar
Glauber Costa committed
967
968
969
970
971
972
973
974
975
976

	if (mem_cgroup_sockets_enabled && sk->sk_cgrp) {
		struct cg_proto *cg_proto = sk->sk_cgrp;
		struct proto *prot = sk->sk_prot;

		for (; cg_proto; cg_proto = parent_cg_proto(prot, cg_proto))
			if (*cg_proto->memory_pressure)
				*cg_proto->memory_pressure = 0;
	}

977
978
979
980
}

static inline void sk_enter_memory_pressure(struct sock *sk)
{
Glauber Costa's avatar
Glauber Costa committed
981
982
983
984
985
986
987
988
989
990
991
992
	if (!sk->sk_prot->enter_memory_pressure)
		return;

	if (mem_cgroup_sockets_enabled && sk->sk_cgrp) {
		struct cg_proto *cg_proto = sk->sk_cgrp;
		struct proto *prot = sk->sk_prot;

		for (; cg_proto; cg_proto = parent_cg_proto(prot, cg_proto))
			cg_proto->enter_memory_pressure(sk);
	}

	sk->sk_prot->enter_memory_pressure(sk);
993
994
995
996
997
}

static inline long sk_prot_mem_limits(const struct sock *sk, int index)
{
	long *prot = sk->sk_prot->sysctl_mem;
Glauber Costa's avatar
Glauber Costa committed
998
999
	if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
		prot = sk->sk_cgrp->sysctl_mem;
1000
	return prot[index];