request_sock.h 6.7 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
/*
 * NET		Generic infrastructure for Network protocols.
 *
 *		Definitions for request_sock 
 *
 * Authors:	Arnaldo Carvalho de Melo <acme@conectiva.com.br>
 *
 * 		From code originally in include/net/tcp.h
 *
 *		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 _REQUEST_SOCK_H
#define _REQUEST_SOCK_H

#include <linux/slab.h>
19
#include <linux/spinlock.h>
20
#include <linux/types.h>
21

22 23
#include <net/sock.h>

24
struct request_sock;
25 26 27 28
struct sk_buff;
struct dst_entry;
struct proto;

29
struct request_sock_ops {
30 31
	int		family;
	int		obj_size;
32
	kmem_cache_t	*slab;
33
	int		(*rtx_syn_ack)(struct sock *sk,
34
				       struct request_sock *req,
35 36
				       struct dst_entry *dst);
	void		(*send_ack)(struct sk_buff *skb,
37
				    struct request_sock *req);
38 39
	void		(*send_reset)(struct sock *sk,
				      struct sk_buff *skb);
40
	void		(*destructor)(struct request_sock *req);
41 42
};

43
/* struct request_sock - mini sock to represent a connection request
44
 */
45 46
struct request_sock {
	struct request_sock		*dl_next; /* Must be first member! */
47 48 49 50 51 52 53 54
	u16				mss;
	u8				retrans;
	u8				__pad;
	/* The following two fields can be easily recomputed I think -AK */
	u32				window_clamp; /* window clamp at creation time */
	u32				rcv_wnd;	  /* rcv_wnd offered first time */
	u32				ts_recent;
	unsigned long			expires;
55
	const struct request_sock_ops	*rsk_ops;
56
	struct sock			*sk;
57
	u32				secid;
58
	u32				peer_secid;
59 60
};

61
static inline struct request_sock *reqsk_alloc(const struct request_sock_ops *ops)
62
{
63
	struct request_sock *req = kmem_cache_alloc(ops->slab, SLAB_ATOMIC);
64 65

	if (req != NULL)
66
		req->rsk_ops = ops;
67 68 69 70

	return req;
}

71
static inline void __reqsk_free(struct request_sock *req)
72
{
73
	kmem_cache_free(req->rsk_ops->slab, req);
74 75
}

76
static inline void reqsk_free(struct request_sock *req)
77
{
78 79
	req->rsk_ops->destructor(req);
	__reqsk_free(req);
80 81
}

82 83
extern int sysctl_max_syn_backlog;

84
/** struct listen_sock - listen state
85 86 87
 *
 * @max_qlen_log - log_2 of maximal queued SYNs/REQUESTs
 */
88
struct listen_sock {
89 90 91 92 93 94
	u8			max_qlen_log;
	/* 3 bytes hole, try to use */
	int			qlen;
	int			qlen_young;
	int			clock_hand;
	u32			hash_rnd;
95
	u32			nr_table_entries;
96 97 98 99 100 101 102
	struct request_sock	*syn_table[0];
};

/** struct request_sock_queue - queue of request_socks
 *
 * @rskq_accept_head - FIFO head of established children
 * @rskq_accept_tail - FIFO tail of established children
103
 * @rskq_defer_accept - User waits for some data after accept()
104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
 * @syn_wait_lock - serializer
 *
 * %syn_wait_lock is necessary only to avoid proc interface having to grab the main
 * lock sock while browsing the listening hash (otherwise it's deadlock prone).
 *
 * This lock is acquired in read mode only from listening_get_next() seq_file
 * op and it's acquired in write mode _only_ from code that is actively
 * changing rskq_accept_head. All readers that are holding the master sock lock
 * don't need to grab this lock in read mode too as rskq_accept_head. writes
 * are always protected from the main sock lock.
 */
struct request_sock_queue {
	struct request_sock	*rskq_accept_head;
	struct request_sock	*rskq_accept_tail;
	rwlock_t		syn_wait_lock;
119 120
	u8			rskq_defer_accept;
	/* 3 bytes hole, try to pack */
121
	struct listen_sock	*listen_opt;
122 123 124
};

extern int reqsk_queue_alloc(struct request_sock_queue *queue,
125
			     unsigned int nr_table_entries);
126

127
static inline struct listen_sock *reqsk_queue_yank_listen_sk(struct request_sock_queue *queue)
128
{
129
	struct listen_sock *lopt;
130 131 132 133 134 135 136 137 138

	write_lock_bh(&queue->syn_wait_lock);
	lopt = queue->listen_opt;
	queue->listen_opt = NULL;
	write_unlock_bh(&queue->syn_wait_lock);

	return lopt;
}

139
static inline void __reqsk_queue_destroy(struct request_sock_queue *queue)
140 141 142 143
{
	kfree(reqsk_queue_yank_listen_sk(queue));
}

144 145
extern void reqsk_queue_destroy(struct request_sock_queue *queue);

146 147 148 149 150
static inline struct request_sock *
	reqsk_queue_yank_acceptq(struct request_sock_queue *queue)
{
	struct request_sock *req = queue->rskq_accept_head;

151
	queue->rskq_accept_head = NULL;
152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214
	return req;
}

static inline int reqsk_queue_empty(struct request_sock_queue *queue)
{
	return queue->rskq_accept_head == NULL;
}

static inline void reqsk_queue_unlink(struct request_sock_queue *queue,
				      struct request_sock *req,
				      struct request_sock **prev_req)
{
	write_lock(&queue->syn_wait_lock);
	*prev_req = req->dl_next;
	write_unlock(&queue->syn_wait_lock);
}

static inline void reqsk_queue_add(struct request_sock_queue *queue,
				   struct request_sock *req,
				   struct sock *parent,
				   struct sock *child)
{
	req->sk = child;
	sk_acceptq_added(parent);

	if (queue->rskq_accept_head == NULL)
		queue->rskq_accept_head = req;
	else
		queue->rskq_accept_tail->dl_next = req;

	queue->rskq_accept_tail = req;
	req->dl_next = NULL;
}

static inline struct request_sock *reqsk_queue_remove(struct request_sock_queue *queue)
{
	struct request_sock *req = queue->rskq_accept_head;

	BUG_TRAP(req != NULL);

	queue->rskq_accept_head = req->dl_next;
	if (queue->rskq_accept_head == NULL)
		queue->rskq_accept_tail = NULL;

	return req;
}

static inline struct sock *reqsk_queue_get_child(struct request_sock_queue *queue,
						 struct sock *parent)
{
	struct request_sock *req = reqsk_queue_remove(queue);
	struct sock *child = req->sk;

	BUG_TRAP(child != NULL);

	sk_acceptq_removed(parent);
	__reqsk_free(req);
	return child;
}

static inline int reqsk_queue_removed(struct request_sock_queue *queue,
				      struct request_sock *req)
{
215
	struct listen_sock *lopt = queue->listen_opt;
216 217 218 219 220 221 222 223 224

	if (req->retrans == 0)
		--lopt->qlen_young;

	return --lopt->qlen;
}

static inline int reqsk_queue_added(struct request_sock_queue *queue)
{
225
	struct listen_sock *lopt = queue->listen_opt;
226 227 228 229 230 231 232
	const int prev_qlen = lopt->qlen;

	lopt->qlen_young++;
	lopt->qlen++;
	return prev_qlen;
}

233
static inline int reqsk_queue_len(const struct request_sock_queue *queue)
234 235 236 237
{
	return queue->listen_opt != NULL ? queue->listen_opt->qlen : 0;
}

238
static inline int reqsk_queue_len_young(const struct request_sock_queue *queue)
239 240 241 242
{
	return queue->listen_opt->qlen_young;
}

243
static inline int reqsk_queue_is_full(const struct request_sock_queue *queue)
244 245 246 247 248 249
{
	return queue->listen_opt->qlen >> queue->listen_opt->max_qlen_log;
}

static inline void reqsk_queue_hash_req(struct request_sock_queue *queue,
					u32 hash, struct request_sock *req,
250
					unsigned long timeout)
251
{
252
	struct listen_sock *lopt = queue->listen_opt;
253 254 255 256 257 258 259 260 261 262 263

	req->expires = jiffies + timeout;
	req->retrans = 0;
	req->sk = NULL;
	req->dl_next = lopt->syn_table[hash];

	write_lock(&queue->syn_wait_lock);
	lopt->syn_table[hash] = req;
	write_unlock(&queue->syn_wait_lock);
}

264
#endif /* _REQUEST_SOCK_H */