sock.h 59.6 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>
58
#include <linux/jump_label.h>
Linus Torvalds's avatar
Linus Torvalds committed
59
60

#include <linux/filter.h>
61
#include <linux/rculist_nulls.h>
62
#include <linux/poll.h>
Linus Torvalds's avatar
Linus Torvalds committed
63

64
#include <linux/atomic.h>
Linus Torvalds's avatar
Linus Torvalds committed
65
66
67
#include <net/dst.h>
#include <net/checksum.h>

68
69
struct cgroup;
struct cgroup_subsys;
70
#ifdef CONFIG_NET
Glauber Costa's avatar
Glauber Costa committed
71
72
int mem_cgroup_sockets_init(struct cgroup *cgrp, struct cgroup_subsys *ss);
void mem_cgroup_sockets_destroy(struct cgroup *cgrp, struct cgroup_subsys *ss);
73
74
75
76
77
78
79
80
81
82
83
#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
84
85
86
87
88
89
90
91
92
93
94
95
/*
 * 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
96
/* Validate arguments and do nothing */
97
98
static inline __printf(2, 3)
void SOCK_DEBUG(struct sock *sk, const char *msg, ...)
99
100
{
}
Linus Torvalds's avatar
Linus Torvalds committed
101
102
103
104
105
106
107
108
#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;
109
	int			owned;
Linus Torvalds's avatar
Linus Torvalds committed
110
	wait_queue_head_t	wq;
111
112
113
114
115
116
117
118
119
	/*
	 * 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
120
121
122
} socket_lock_t;

struct sock;
123
struct proto;
124
struct net;
Linus Torvalds's avatar
Linus Torvalds committed
125
126

/**
127
 *	struct sock_common - minimal network layer representation of sockets
128
129
 *	@skc_daddr: Foreign IPv4 addr
 *	@skc_rcv_saddr: Bound local IPv4 addr
Eric Dumazet's avatar
Eric Dumazet committed
130
 *	@skc_hash: hash value used with various protocol lookup tables
131
 *	@skc_u16hashes: two u16 hash values used by UDP lookup tables
132
133
134
135
136
 *	@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
137
 *	@skc_portaddr_node: second hash linkage for UDP/UDP-Lite protocol
138
 *	@skc_prot: protocol handlers inside a network family
139
 *	@skc_net: reference to the network namespace of this socket
140
141
142
143
 *	@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
144
145
 *
 *	This is the minimal network layer representation of sockets, the header
146
147
 *	for struct sock and struct inet_timewait_sock.
 */
Linus Torvalds's avatar
Linus Torvalds committed
148
struct sock_common {
149
150
	/* skc_daddr and skc_rcv_saddr must be grouped :
	 * cf INET_MATCH() and INET_TW_MATCH()
Eric Dumazet's avatar
Eric Dumazet committed
151
	 */
152
153
	__be32			skc_daddr;
	__be32			skc_rcv_saddr;
Eric Dumazet's avatar
Eric Dumazet committed
154

155
156
157
158
	union  {
		unsigned int	skc_hash;
		__u16		skc_u16hashes[2];
	};
Eric Dumazet's avatar
Eric Dumazet committed
159
160
161
162
	unsigned short		skc_family;
	volatile unsigned char	skc_state;
	unsigned char		skc_reuse;
	int			skc_bound_dev_if;
163
164
165
166
	union {
		struct hlist_node	skc_bind_node;
		struct hlist_nulls_node skc_portaddr_node;
	};
167
	struct proto		*skc_prot;
168
#ifdef CONFIG_NET_NS
169
	struct net	 	*skc_net;
170
#endif
171
172
173
174
	/*
	 * fields between dontcopy_begin/dontcopy_end
	 * are not copied in sock_copy()
	 */
175
	/* private: */
176
	int			skc_dontcopy_begin[0];
177
	/* public: */
178
179
180
181
182
183
	union {
		struct hlist_node	skc_node;
		struct hlist_nulls_node skc_nulls_node;
	};
	int			skc_tx_queue_mapping;
	atomic_t		skc_refcnt;
184
	/* private: */
185
	int                     skc_dontcopy_end[0];
186
	/* public: */
Linus Torvalds's avatar
Linus Torvalds committed
187
188
};

Glauber Costa's avatar
Glauber Costa committed
189
struct cg_proto;
Linus Torvalds's avatar
Linus Torvalds committed
190
191
/**
  *	struct sock - network layer representation of sockets
192
  *	@__sk_common: shared layout with inet_timewait_sock
193
194
195
196
  *	@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
197
  *	@sk_wq: sock wait queue and async head
198
199
200
201
202
203
  *	@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
204
  *	@sk_async_wait_queue: DMA copied packets
205
206
207
208
209
  *	@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
210
  *	@sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE,
211
  *		   %SO_OOBINLINE settings, %SO_TIMESTAMPING settings
212
213
  *	@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
214
  *	@sk_route_nocaps: forbidden route capabilities (e.g NETIF_F_GSO_MASK)
215
  *	@sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4)
216
  *	@sk_gso_max_size: Maximum GSO segment size to build
217
218
219
220
  *	@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
221
222
  *	@sk_prot_creator: sk_prot of original sock creator (see ipv6_setsockopt,
  *			  IPV6_ADDRFORM for instance)
223
  *	@sk_err: last error
Wang Chen's avatar
Wang Chen committed
224
225
  *	@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
226
  *	@sk_drops: raw/udp drops counter
227
228
229
  *	@sk_ack_backlog: current listen backlog
  *	@sk_max_ack_backlog: listen backlog set in listen()
  *	@sk_priority: %SO_PRIORITY setting
230
  *	@sk_cgrp_prioidx: socket group's priority map index
231
232
  *	@sk_type: socket type (%SOCK_STREAM, etc)
  *	@sk_protocol: which protocol this socket belongs in this network family
233
234
  *	@sk_peer_pid: &struct pid for this socket's peer
  *	@sk_peer_cred: %SO_PEERCRED setting
235
236
237
  *	@sk_rcvlowat: %SO_RCVLOWAT setting
  *	@sk_rcvtimeo: %SO_RCVTIMEO setting
  *	@sk_sndtimeo: %SO_SNDTIMEO setting
238
  *	@sk_rxhash: flow hash received from netif layer
239
240
241
242
243
244
245
246
247
  *	@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
248
  *	@sk_security: used by security modules
249
  *	@sk_mark: generic packet mark
250
  *	@sk_classid: this socket's cgroup classid
Glauber Costa's avatar
Glauber Costa committed
251
  *	@sk_cgrp: this socket's cgroup-specific proto data
252
253
254
255
256
257
258
  *	@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
259
260
261
 */
struct sock {
	/*
262
	 * Now struct inet_timewait_sock also uses sock_common, so please just
Linus Torvalds's avatar
Linus Torvalds committed
263
264
265
	 * don't add nothing before this first member (__sk_common) --acme
	 */
	struct sock_common	__sk_common;
Eric Dumazet's avatar
Eric Dumazet committed
266
267
268
#define sk_node			__sk_common.skc_node
#define sk_nulls_node		__sk_common.skc_nulls_node
#define sk_refcnt		__sk_common.skc_refcnt
269
#define sk_tx_queue_mapping	__sk_common.skc_tx_queue_mapping
Eric Dumazet's avatar
Eric Dumazet committed
270

271
272
#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
273
#define sk_hash			__sk_common.skc_hash
Linus Torvalds's avatar
Linus Torvalds committed
274
275
276
277
278
#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
279
#define sk_prot			__sk_common.skc_prot
280
#define sk_net			__sk_common.skc_net
Linus Torvalds's avatar
Linus Torvalds committed
281
	socket_lock_t		sk_lock;
Eric Dumazet's avatar
Eric Dumazet committed
282
	struct sk_buff_head	sk_receive_queue;
283
284
285
286
	/*
	 * 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
287
288
289
	 * Note : rmem_alloc is in this structure to fill a hole
	 * on 64bit arches, not because its logically part of
	 * backlog.
290
291
	 */
	struct {
Eric Dumazet's avatar
Eric Dumazet committed
292
293
294
295
		atomic_t	rmem_alloc;
		int		len;
		struct sk_buff	*head;
		struct sk_buff	*tail;
296
	} sk_backlog;
Eric Dumazet's avatar
Eric Dumazet committed
297
298
299
300
301
302
303
304
305
#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;
306
	struct socket_wq __rcu	*sk_wq;
Eric Dumazet's avatar
Eric Dumazet committed
307
308
309
310
311

#ifdef CONFIG_NET_DMA
	struct sk_buff_head	sk_async_wait_queue;
#endif

312
#ifdef CONFIG_XFRM
Linus Torvalds's avatar
Linus Torvalds committed
313
	struct xfrm_policy	*sk_policy[2];
314
#endif
Eric Dumazet's avatar
Eric Dumazet committed
315
316
	unsigned long 		sk_flags;
	struct dst_entry	*sk_dst_cache;
Eric Dumazet's avatar
Eric Dumazet committed
317
	spinlock_t		sk_dst_lock;
Linus Torvalds's avatar
Linus Torvalds committed
318
319
	atomic_t		sk_wmem_alloc;
	atomic_t		sk_omem_alloc;
320
	int			sk_sndbuf;
Linus Torvalds's avatar
Linus Torvalds committed
321
	struct sk_buff_head	sk_write_queue;
Eric Dumazet's avatar
Eric Dumazet committed
322
323
324
325
326
327
328
	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
329
	int			sk_wmem_queued;
Al Viro's avatar
Al Viro committed
330
	gfp_t			sk_allocation;
331
332
	netdev_features_t	sk_route_caps;
	netdev_features_t	sk_route_nocaps;
333
	int			sk_gso_type;
334
	unsigned int		sk_gso_max_size;
335
	int			sk_rcvlowat;
Linus Torvalds's avatar
Linus Torvalds committed
336
337
	unsigned long	        sk_lingertime;
	struct sk_buff_head	sk_error_queue;
338
	struct proto		*sk_prot_creator;
Linus Torvalds's avatar
Linus Torvalds committed
339
340
341
342
343
344
	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;
345
346
347
#ifdef CONFIG_CGROUPS
	__u32			sk_cgrp_prioidx;
#endif
348
349
	struct pid		*sk_peer_pid;
	const struct cred	*sk_peer_cred;
Linus Torvalds's avatar
Linus Torvalds committed
350
351
352
353
	long			sk_rcvtimeo;
	long			sk_sndtimeo;
	void			*sk_protinfo;
	struct timer_list	sk_timer;
354
	ktime_t			sk_stamp;
Linus Torvalds's avatar
Linus Torvalds committed
355
356
357
358
359
	struct socket		*sk_socket;
	void			*sk_user_data;
	struct page		*sk_sndmsg_page;
	struct sk_buff		*sk_send_head;
	__u32			sk_sndmsg_off;
360
	__s32			sk_peek_off;
Linus Torvalds's avatar
Linus Torvalds committed
361
	int			sk_write_pending;
Alexey Dobriyan's avatar
Alexey Dobriyan committed
362
#ifdef CONFIG_SECURITY
Linus Torvalds's avatar
Linus Torvalds committed
363
	void			*sk_security;
Alexey Dobriyan's avatar
Alexey Dobriyan committed
364
#endif
365
	__u32			sk_mark;
366
	u32			sk_classid;
Glauber Costa's avatar
Glauber Costa committed
367
	struct cg_proto		*sk_cgrp;
Linus Torvalds's avatar
Linus Torvalds committed
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);
};

377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
static inline int sk_peek_offset(struct sock *sk, int flags)
{
	if ((flags & MSG_PEEK) && (sk->sk_peek_off >= 0))
		return sk->sk_peek_off;
	else
		return 0;
}

static inline void sk_peek_offset_bwd(struct sock *sk, int val)
{
	if (sk->sk_peek_off >= 0) {
		if (sk->sk_peek_off >= val)
			sk->sk_peek_off -= val;
		else
			sk->sk_peek_off = 0;
	}
}

static inline void sk_peek_offset_fwd(struct sock *sk, int val)
{
	if (sk->sk_peek_off >= 0)
		sk->sk_peek_off += val;
}

Linus Torvalds's avatar
Linus Torvalds committed
401
402
403
/*
 * Hashed lists helper routines
 */
Li Zefan's avatar
Li Zefan committed
404
405
406
407
408
static inline struct sock *sk_entry(const struct hlist_node *node)
{
	return hlist_entry(node, struct sock, sk_node);
}

409
static inline struct sock *__sk_head(const struct hlist_head *head)
Linus Torvalds's avatar
Linus Torvalds committed
410
411
412
413
{
	return hlist_entry(head->first, struct sock, sk_node);
}

414
static inline struct sock *sk_head(const struct hlist_head *head)
Linus Torvalds's avatar
Linus Torvalds committed
415
416
417
418
{
	return hlist_empty(head) ? NULL : __sk_head(head);
}

419
420
421
422
423
424
425
426
427
428
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);
}

429
static inline struct sock *sk_next(const struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
430
431
432
433
434
{
	return sk->sk_node.next ?
		hlist_entry(sk->sk_node.next, struct sock, sk_node) : NULL;
}

435
436
437
438
439
440
441
442
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;
}

443
static inline int sk_unhashed(const struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
444
445
446
447
{
	return hlist_unhashed(&sk->sk_node);
}

448
static inline int sk_hashed(const struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
449
{
Akinobu Mita's avatar
Akinobu Mita committed
450
	return !sk_unhashed(sk);
Linus Torvalds's avatar
Linus Torvalds committed
451
452
453
454
455
456
457
}

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

458
459
460
461
462
static __inline__ void sk_nulls_node_init(struct hlist_nulls_node *node)
{
	node->pprev = NULL;
}

Linus Torvalds's avatar
Linus Torvalds committed
463
464
465
466
467
static __inline__ void __sk_del_node(struct sock *sk)
{
	__hlist_del(&sk->sk_node);
}

468
/* NB: equivalent to hlist_del_init_rcu */
Linus Torvalds's avatar
Linus Torvalds committed
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
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;
}
509
#define sk_del_node_init_rcu(sk)	sk_del_node_init(sk)
Linus Torvalds's avatar
Linus Torvalds committed
510

511
static __inline__ int __sk_nulls_del_node_init_rcu(struct sock *sk)
512
513
{
	if (sk_hashed(sk)) {
514
		hlist_nulls_del_init_rcu(&sk->sk_nulls_node);
515
516
517
518
519
		return 1;
	}
	return 0;
}

520
static __inline__ int sk_nulls_del_node_init_rcu(struct sock *sk)
521
{
522
	int rc = __sk_nulls_del_node_init_rcu(sk);
523
524
525
526
527
528
529
530
531

	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
532
533
534
535
536
537
538
539
540
541
542
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);
}

543
544
545
546
547
548
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);
}

549
static __inline__ void __sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list)
550
{
551
	hlist_nulls_add_head_rcu(&sk->sk_nulls_node, list);
552
553
}

554
static __inline__ void sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list)
555
556
{
	sock_hold(sk);
557
	__sk_nulls_add_node_rcu(sk, list);
558
559
}

Linus Torvalds's avatar
Linus Torvalds committed
560
561
562
563
564
565
566
567
568
569
570
571
572
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)
573
574
#define sk_for_each_rcu(__sk, node, list) \
	hlist_for_each_entry_rcu(__sk, node, list, sk_node)
575
576
577
578
#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
579
580
581
#define sk_for_each_from(__sk, node) \
	if (__sk && ({ node = &(__sk)->sk_node; 1; })) \
		hlist_for_each_entry_from(__sk, node, sk_node)
582
583
584
#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
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
#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 */
604
	SOCK_RCVTSTAMPNS, /* %SO_TIMESTAMPNS setting */
Linus Torvalds's avatar
Linus Torvalds committed
605
606
	SOCK_LOCALROUTE, /* route locally only, %SO_DONTROUTE setting */
	SOCK_QUEUE_SHRUNK, /* write queue has been shrunk recently */
607
608
609
610
611
612
613
	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
614
	SOCK_FASYNC, /* fasync() active */
615
	SOCK_RXQ_OVFL,
616
	SOCK_ZEROCOPY, /* buffers from userspace */
617
	SOCK_WIFI_STATUS, /* push wifi status to userspace */
Linus Torvalds's avatar
Linus Torvalds committed
618
619
};

Ralf Baechle's avatar
Ralf Baechle committed
620
621
622
623
624
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
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
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)
{
652
	return sk->sk_ack_backlog > sk->sk_max_ack_backlog;
Linus Torvalds's avatar
Linus Torvalds committed
653
654
655
656
657
658
659
}

/*
 * Compute minimal free write space needed to queue new packets.
 */
static inline int sk_stream_min_wspace(struct sock *sk)
{
660
	return sk->sk_wmem_queued >> 1;
Linus Torvalds's avatar
Linus Torvalds committed
661
662
663
664
665
666
667
668
669
670
671
672
673
674
}

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
675
/* OOB backlog add */
Zhu Yi's avatar
Zhu Yi committed
676
static inline void __sk_add_backlog(struct sock *sk, struct sk_buff *skb)
677
{
Eric Dumazet's avatar
Eric Dumazet committed
678
679
680
681
682
683
	/* 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
684
		sk->sk_backlog.tail->next = skb;
Eric Dumazet's avatar
Eric Dumazet committed
685
686

	sk->sk_backlog.tail = skb;
687
688
	skb->next = NULL;
}
Linus Torvalds's avatar
Linus Torvalds committed
689

690
691
/*
 * Take into account size of receive queue and backlog queue
Eric Dumazet's avatar
Eric Dumazet committed
692
693
 * Do not take into account this skb truesize,
 * to allow even a single big packet to come.
694
695
696
697
698
 */
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
699
	return qsize > sk->sk_rcvbuf;
700
701
}

Zhu Yi's avatar
Zhu Yi committed
702
/* The per-socket spinlock must be held here. */
703
static inline __must_check int sk_add_backlog(struct sock *sk, struct sk_buff *skb)
Zhu Yi's avatar
Zhu Yi committed
704
{
705
	if (sk_rcvqueues_full(sk, skb))
Zhu Yi's avatar
Zhu Yi committed
706
707
		return -ENOBUFS;

Zhu Yi's avatar
Zhu Yi committed
708
	__sk_add_backlog(sk, skb);
Zhu Yi's avatar
Zhu Yi committed
709
710
711
712
	sk->sk_backlog.len += skb->truesize;
	return 0;
}

713
714
715
716
717
static inline int sk_backlog_rcv(struct sock *sk, struct sk_buff *skb)
{
	return sk->sk_backlog_rcv(sk, skb);
}

718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
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
}

742
743
static inline void sock_rps_save_rxhash(struct sock *sk,
					const struct sk_buff *skb)
744
745
{
#ifdef CONFIG_RPS
746
	if (unlikely(sk->sk_rxhash != skb->rxhash)) {
747
		sock_rps_reset_flow(sk);
748
		sk->sk_rxhash = skb->rxhash;
749
750
751
752
	}
#endif
}

753
754
755
756
757
758
759
760
static inline void sock_rps_reset_rxhash(struct sock *sk)
{
#ifdef CONFIG_RPS
	sock_rps_reset_flow(sk);
	sk->sk_rxhash = 0;
#endif
}

761
762
763
764
765
766
767
768
769
770
771
#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
772
773
774
775
776
777
778
779
780

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);

781
struct request_sock_ops;
782
struct timewait_sock_ops;
783
struct inet_hashinfo;
784
struct raw_hashinfo;
785
struct module;
786

Linus Torvalds's avatar
Linus Torvalds committed
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
/* 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);
804
	void			(*destroy)(struct sock *sk);
Linus Torvalds's avatar
Linus Torvalds committed
805
806
807
	void			(*shutdown)(struct sock *sk, int how);
	int			(*setsockopt)(struct sock *sk, int level, 
					int optname, char __user *optval,
808
					unsigned int optlen);
Linus Torvalds's avatar
Linus Torvalds committed
809
810
811
	int			(*getsockopt)(struct sock *sk, int level, 
					int optname, char __user *optval, 
					int __user *option);  	 
812
#ifdef CONFIG_COMPAT
813
814
815
	int			(*compat_setsockopt)(struct sock *sk,
					int level,
					int optname, char __user *optval,
816
					unsigned int optlen);
817
818
819
820
	int			(*compat_getsockopt)(struct sock *sk,
					int level,
					int optname, char __user *optval,
					int __user *option);
821
822
	int			(*compat_ioctl)(struct sock *sk,
					unsigned int cmd, unsigned long arg);
823
#endif
Linus Torvalds's avatar
Linus Torvalds committed
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
	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
841
	void			(*rehash)(struct sock *sk);
Linus Torvalds's avatar
Linus Torvalds committed
842
	int			(*get_port)(struct sock *sk, unsigned short snum);
843
	void			(*clear_sk)(struct sock *sk, int size);
Linus Torvalds's avatar
Linus Torvalds committed
844

845
	/* Keeping track of sockets in use */
846
#ifdef CONFIG_PROC_FS
847
	unsigned int		inuse_idx;
848
#endif
849

Linus Torvalds's avatar
Linus Torvalds committed
850
	/* Memory pressure */
851
	void			(*enter_memory_pressure)(struct sock *sk);
Eric Dumazet's avatar
Eric Dumazet committed
852
	atomic_long_t		*memory_allocated;	/* Current allocated memory. */
853
	struct percpu_counter	*sockets_allocated;	/* Current number of sockets. */
Linus Torvalds's avatar
Linus Torvalds committed
854
855
856
	/*
	 * Pressure flag: try to collapse.
	 * Technical note: it is used by multiple contexts non atomically.
857
	 * All the __sk_mem_schedule() is of this nature: accounting
Linus Torvalds's avatar
Linus Torvalds committed
858
859
860
	 * is strict, actions are advisory and have some latency.
	 */
	int			*memory_pressure;
Eric Dumazet's avatar
Eric Dumazet committed
861
	long			*sysctl_mem;
Linus Torvalds's avatar
Linus Torvalds committed
862
863
864
	int			*sysctl_wmem;
	int			*sysctl_rmem;
	int			max_header;
865
	bool			no_autobind;
Linus Torvalds's avatar
Linus Torvalds committed
866

867
	struct kmem_cache	*slab;
Linus Torvalds's avatar
Linus Torvalds committed
868
	unsigned int		obj_size;
869
	int			slab_flags;
Linus Torvalds's avatar
Linus Torvalds committed
870

871
	struct percpu_counter	*orphan_count;
872

873
	struct request_sock_ops	*rsk_prot;
874
	struct timewait_sock_ops *twsk_prot;
875

876
877
	union {
		struct inet_hashinfo	*hashinfo;
878
		struct udp_table	*udp_table;
879
		struct raw_hashinfo	*raw_hash;
880
	} h;
881

Linus Torvalds's avatar
Linus Torvalds committed
882
883
884
885
886
	struct module		*owner;

	char			name[32];

	struct list_head	node;
887
888
889
#ifdef SOCK_REFCNT_DEBUG
	atomic_t		socks;
#endif
Glauber Costa's avatar
Glauber Costa committed
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
#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
921
922
923
924
925
};

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

926
927
928
929
930
931
932
933
934
935
936
937
938
#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
939
inline void sk_refcnt_debug_release(const struct sock *sk)
940
941
942
943
944
945
946
947
948
949
950
{
	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 */

951
#if defined(CONFIG_CGROUP_MEM_RES_CTLR_KMEM) && defined(CONFIG_NET)
Glauber Costa's avatar
Glauber Costa committed
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
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


969
970
971
972
973
974
975
976
977
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
978
979
980
981

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

982
983
984
985
986
987
988
	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
989
990
991
992
	if (!memory_pressure)
		return;

	if (*memory_pressure)
993
		*memory_pressure = 0;
Glauber Costa's avatar
Glauber Costa committed
994
995
996
997
998
999
1000
1001
1002
1003

	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;
	}

1004
1005
1006
1007
}

static inline void sk_enter_memory_pressure(struct sock *sk)
{
Glauber Costa's avatar
Glauber Costa committed
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
	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);
1020
1021
1022
1023
1024
}

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
1025
1026
	if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
		prot = sk->sk_cgrp->sysctl_mem;
1027
1028
1029
	return prot[index];
}

Glauber Costa's avatar
Glauber Costa committed
1030
1031
1032
1033
1034
1035
1036
static inline void memcg_memory_allocated_add(struct cg_proto *prot,
					      unsigned long amt,
					      int *parent_status)
{
	struct res_counter *fail;
	int ret;

1037
1038
	ret = res_counter_charge_nofail(prot->memory_allocated,
					amt << PAGE_SHIFT, &fail);
Glauber Costa's avatar
Glauber Costa committed
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
	if (ret < 0)
		*parent_status = OVER_LIMIT;
}

static inline void memcg_memory_allocated_sub(struct cg_proto *prot,
					      unsigned long amt)
{
	res_counter_uncharge(prot->memory_allocated, amt << PAGE_SHIFT);
}

static inline u64 memcg_memory_allocated_read(struct cg_proto *prot)
{
	u64 ret;
	ret = res_counter_read_u64(prot->memory_allocated, RES_USAGE);
	return ret >> PAGE_SHIFT;
}

1056
1057
1058
1059
static inline long
sk_memory_allocated(const struct sock *sk)
{
	struct proto *prot = sk->sk_prot;
Glauber Costa's avatar
Glauber Costa committed
1060
1061
1062
	if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
		return memcg_memory_allocated_read(sk->sk_cgrp);

1063
1064
1065
1066
	return atomic_long_read(prot->memory_allocated);
}

static inline long
Glauber Costa's avatar
Glauber Costa committed
1067
sk_memory_allocated_add(struct sock *sk, int amt, int *parent_status)
1068
1069
{
	struct proto *prot = sk->sk_prot;
Glauber Costa's avatar
Glauber Costa committed
1070
1071
1072
1073
1074
1075
1076
1077

	if (mem_cgroup_sockets_enabled && sk->sk_cgrp) {
		memcg_memory_allocated_add(sk->sk_cgrp, amt, parent_status);
		/* update the root cgroup regardless */
		atomic_long_add_return(amt, prot->memory_allocated);
		return memcg_memory_allocated_read(sk->sk_cgrp);
	}

1078
1079
1080
1081
	return atomic_long_add_return(amt, prot->memory_allocated);
}

static inline void
1082
sk_memory_allocated_sub(struct sock *sk, int amt)
1083
1084
{
	struct proto *prot = sk->sk_prot;
Glauber Costa's avatar
Glauber Costa committed
1085

1086
	if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
Glauber Costa's avatar
Glauber Costa committed
1087
1088
		memcg_memory_allocated_sub(sk->sk_cgrp, amt);

1089
1090
1091
1092
1093
1094
	atomic_long_sub(amt, prot->memory_allocated);
}

static inline void sk_sockets_allocated_dec(struct sock *sk)
{
	struct proto *prot = sk->sk_prot;
Glauber Costa's avatar
Glauber Costa committed
1095
1096
1097
1098
1099
1100
1101
1102

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

		for (; cg_proto; cg_proto = parent_cg_proto(prot, cg_proto))
			percpu_counter_dec(cg_proto->sockets_allocated);
	}

1103
1104
1105
1106
1107
1108
	percpu_counter_dec(prot->sockets_allocated);
}

static inline void sk_sockets_allocated_inc(struct sock *sk)
{
	struct proto *prot = sk->sk_prot;
Glauber Costa's avatar
Glauber Costa committed
1109
1110
1111
1112
1113
1114
1115
1116

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

		for (; cg_proto; cg_proto = parent_cg_proto(prot, cg_proto))
			percpu_counter_inc(cg_proto->sockets_allocated);
	}

1117
1118
1119
1120
1121
1122
1123
1124
	percpu_counter_inc(prot->sockets_allocated);
}

static inline int
sk_sockets_allocated_read_positive(struct sock *sk)
{
	struct proto *prot = sk->sk_prot;

Glauber Costa's avatar
Glauber Costa committed
1125
1126
1127
	if (mem_cgroup_sockets_enabled && sk->sk_cgrp)
		return percpu_counter_sum_positive(sk->sk_cgrp->sockets_allocated);

1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
	return percpu_counter_sum_positive(prot->sockets_allocated);
}

static inline int
proto_sockets_allocated_sum_positive(struct proto *prot)
{
	return percpu_counter_sum_positive(prot->sockets_allocated);
}

static inline long
proto_memory_allocated(struct proto *prot)
{
	return atomic_long_read(prot->memory_allocated);
}

static inline bool
proto_memory_pressure(struct proto *prot)
{
	if (!prot->memory_pressure)
		return false;
	return !!*prot->memory_pressure;
}

1151
1152

#ifdef CONFIG_PROC_FS
Linus Torvalds's avatar
Linus Torvalds committed
1153
/* Called with local bh disabled */
1154
1155
extern void sock_prot_inuse_add(struct net *net, struct proto *prot, int inc);
extern int sock_prot_inuse_get(struct net *net, struct proto *proto);
1156
#else
1157
1158
static void inline sock_prot_inuse_add(struct net *net, struct proto *prot,
		int inc)
1159
1160
1161
1162
{
}
#endif

Linus Torvalds's avatar
Linus Torvalds committed
1163

1164
1165
1166
1167
1168
1169
1170
1171
1172
/* With per-bucket locks this operation is not-atomic, so that
 * this version is not worse.
 */
static inline void __sk_prot_rehash(struct sock *sk)
{
	sk->sk_prot->unhash(sk);
	sk->sk_prot->hash(sk);
}

1173
1174
void sk_prot_clear_portaddr_nulls(struct sock *sk, int size);

Linus Torvalds's avatar
Linus Torvalds committed
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
/* About 10 seconds */
#define SOCK_DESTROY_TIME (10*HZ)

/* Sockets 0-1023 can't be bound to unless you are superuser */
#define PROT_SOCK	1024

#define SHUTDOWN_MASK	3
#define RCV_SHUTDOWN	1
#define SEND_SHUTDOWN	2

#define SOCK_SNDBUF_LOCK	1
#define SOCK_RCVBUF_LOCK	2
#define SOCK_BINDADDR_LOCK	4
#define SOCK_BINDPORT_LOCK	8

/* sock_iocb: used to kick off async processing of socket ios */
struct sock_iocb {
	struct list_head	list;

	int			flags;
	int			size;
	struct socket		*sock;
	struct sock		*sk;
	struct scm_cookie	*scm;
	struct msghdr		*msg, async_msg;
	struct kiocb		*kiocb;
};

static inline struct sock_iocb *kiocb_to_siocb(struct kiocb *iocb)
{
	return (struct sock_iocb *)iocb->private;
}

static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)
{
	return si->kiocb;
}

struct socket_alloc {
	struct socket socket;
	struct inode vfs_inode;
};

static inline struct socket *SOCKET_I(struct inode *inode)
{
	return &container_of(inode, struct socket_alloc, vfs_inode)->socket;
}

static inline struct inode *SOCK_INODE(struct socket *socket)
{
	return &container_of(socket, struct socket_alloc, socket)->vfs_inode;
}

1228
1229
1230
1231
1232
/*
 * Functions for memory accounting
 */
extern int __sk_mem_schedule(struct sock *sk, int size, int kind);
extern void __sk_mem_reclaim(struct sock *sk);
Linus Torvalds's avatar
Linus Torvalds committed
1233

1234
1235
1236
1237
#define SK_MEM_QUANTUM ((int)PAGE_SIZE)
#define SK_MEM_QUANTUM_SHIFT ilog2(SK_MEM_QUANTUM)
#define SK_MEM_SEND	0
#define SK_MEM_RECV	1
Linus Torvalds's avatar
Linus Torvalds committed
1238

1239
static inline int sk_mem_pages(int amt)
Linus Torvalds's avatar
Linus Torvalds committed
1240
{
1241
	return (amt + SK_MEM_QUANTUM - 1) >> SK_MEM_QUANTUM_SHIFT;
Linus Torvalds's avatar
Linus Torvalds committed
1242
1243
}

1244
static inline int sk_has_account(struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
1245
{
1246
1247
	/* return true if protocol supports memory accounting */
	return !!sk->sk_prot->memory_allocated;
Linus Torvalds's avatar
Linus Torvalds committed
1248
1249
}

1250
static inline int sk_wmem_schedule(struct sock *sk, int size)
Linus Torvalds's avatar
Linus Torvalds committed
1251
{
1252
1253
1254
1255
	if (!sk_has_account(sk))
		return 1;
	return size <= sk->sk_forward_alloc ||
		__sk_mem_schedule(sk, size, SK_MEM_SEND);
Linus Torvalds's avatar
Linus Torvalds committed
1256
1257
}

1258
static inline int sk_rmem_schedule(struct sock *sk, int size)
1259
{
1260
1261
	if (!sk_has_account(sk))
		return 1;
1262
	return size <= sk->sk_forward_alloc ||
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
		__sk_mem_schedule(sk, size, SK_MEM_RECV);
}

static inline void sk_mem_reclaim(struct sock *sk)
{
	if (!sk_has_account(sk))
		return;
	if (sk->sk_forward_alloc >= SK_MEM_QUANTUM)
		__sk_mem_reclaim(sk);
}

1274
1275
1276
1277
1278
1279
1280
1281
static inline void sk_mem_reclaim_partial(struct sock *sk)
{
	if (!sk_has_account(sk))
		return;
	if (sk->sk_forward_alloc > SK_MEM_QUANTUM)
		__sk_mem_reclaim(sk);
}

1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
static inline void sk_mem_charge(struct sock *sk, int size)
{
	if (!sk_has_account(sk))
		return;
	sk->sk_forward_alloc -= size;
}

static inline void sk_mem_uncharge(struct sock *sk, int size)
{
	if (!sk_has_account(sk))
		return;
	sk->sk_forward_alloc += size;
}

static inline void sk_wmem_free_skb(struct sock *sk, struct sk_buff *skb)
{
	sock_set_flag(sk, SOCK_QUEUE_SHRUNK);
	sk->sk_wmem_queued -= skb->truesize;
	sk_mem_uncharge(sk, skb->truesize);
	__kfree_skb(skb);
1302
1303
}

Linus Torvalds's avatar
Linus Torvalds committed
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
/* Used by processes to "lock" a socket state, so that
 * interrupts and bottom half handlers won't change it
 * from under us. It essentially blocks any incoming
 * packets, so that we won't get any new data or any
 * packets that change the state of the socket.
 *
 * While locked, BH processing will add new packets to
 * the backlog queue.  This queue is processed by the
 * owner of the socket lock right before it is released.
 *
 * Since ~2.3.5 it is also exclusive sleep lock serializing
 * accesses from user process context.
 */
1317
#define sock_owned_by_user(sk)	((sk)->sk_lock.owned)
Linus Torvalds's avatar
Linus Torvalds committed
1318

1319
1320
1321
1322
1323
1324
1325
1326
1327
/*
 * Macro so as to not evaluate some arguments when
 * lockdep is not enabled.
 *
 * Mark both the sk_lock and the sk_lock.slock as a
 * per-address-family lock class.
 */
#define sock_lock_init_class_and_name(sk, sname, skey, name, key) 	\
do {									\
1328
	sk->sk_lock.owned = 0;						\
1329
1330
1331
1332
1333
1334
1335
1336
1337
	init_waitqueue_head(&sk->sk_lock.wq);				\
	spin_lock_init(&(sk)->sk_lock.slock);				\
	debug_check_no_locks_freed((void *)&(sk)->sk_lock,		\
			sizeof((sk)->sk_lock));				\
	lockdep_set_class_and_name(&(sk)->sk_lock.slock,		\
		       	(skey), (sname));				\
	lockdep_init_map(&(sk)->sk_lock.dep_map, (name), (key), 0);	\
} while (0)

Harvey Harrison's avatar
Harvey Harrison committed
1338
extern void lock_sock_nested(struct sock *sk, int subclass);
1339
1340
1341
1342
1343
1344

static inline void lock_sock(struct sock *sk)
{
	lock_sock_nested(sk, 0);
}

Harvey Harrison's avatar
Harvey Harrison committed
1345
extern void release_sock(struct sock *sk);
Linus Torvalds's avatar
Linus Torvalds committed
1346
1347
1348

/* BH context may only use the following locking interface. */
#define bh_lock_sock(__sk)	spin_lock(&((__sk)->sk_lock.slock))