tcp_ipv4.c 56.1 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 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
/*
 * 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.
 *
 *		Implementation of the Transmission Control Protocol(TCP).
 *
 * Version:	$Id: tcp_ipv4.c,v 1.240 2002/02/01 22:01:04 davem Exp $
 *
 *		IPv4 specific functions
 *
 *
 *		code split from:
 *		linux/ipv4/tcp.c
 *		linux/ipv4/tcp_input.c
 *		linux/ipv4/tcp_output.c
 *
 *		See tcp.c for author information
 *
 *	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.
 */

/*
 * Changes:
 *		David S. Miller	:	New socket lookup architecture.
 *					This code is dedicated to John Dyson.
 *		David S. Miller :	Change semantics of established hash,
 *					half is devoted to TIME_WAIT sockets
 *					and the rest go in the other half.
 *		Andi Kleen :		Add support for syncookies and fixed
 *					some bugs: ip options weren't passed to
 *					the TCP layer, missed a check for an
 *					ACK bit.
 *		Andi Kleen :		Implemented fast path mtu discovery.
 *	     				Fixed many serious bugs in the
39
 *					request_sock handling and moved
L
Linus Torvalds 已提交
40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
 *					most of it into the af independent code.
 *					Added tail drop and some other bugfixes.
 *					Added new listen sematics.
 *		Mike McLagan	:	Routing by source
 *	Juan Jose Ciarlante:		ip_dynaddr bits
 *		Andi Kleen:		various fixes.
 *	Vitaly E. Lavrov	:	Transparent proxy revived after year
 *					coma.
 *	Andi Kleen		:	Fix new listen.
 *	Andi Kleen		:	Fix accept error reporting.
 *	YOSHIFUJI Hideaki @USAGI and:	Support IPV6_V6ONLY socket option, which
 *	Alexey Kuznetsov		allow both IPv4 and IPv6 sockets to bind
 *					a single port at the same time.
 */

#include <linux/config.h>

#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/module.h>
#include <linux/random.h>
#include <linux/cache.h>
#include <linux/jhash.h>
#include <linux/init.h>
#include <linux/times.h>

#include <net/icmp.h>
67
#include <net/inet_hashtables.h>
L
Linus Torvalds 已提交
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91
#include <net/tcp.h>
#include <net/ipv6.h>
#include <net/inet_common.h>
#include <net/xfrm.h>

#include <linux/inet.h>
#include <linux/ipv6.h>
#include <linux/stddef.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>

extern int sysctl_ip_dynaddr;
int sysctl_tcp_tw_reuse;
int sysctl_tcp_low_latency;

/* Check TCP sequence numbers in ICMP packets. */
#define ICMP_MIN_LENGTH 8

/* Socket used for sending RSTs */
static struct socket *tcp_socket;

void tcp_v4_send_check(struct sock *sk, struct tcphdr *th, int len,
		       struct sk_buff *skb);

92 93 94 95 96
struct inet_hashinfo __cacheline_aligned tcp_hashinfo = {
	.lhash_lock	= RW_LOCK_UNLOCKED,
	.lhash_users	= ATOMIC_INIT(0),
	.lhash_wait	= __WAIT_QUEUE_HEAD_INITIALIZER(tcp_hashinfo.lhash_wait),
	.portalloc_lock	= SPIN_LOCK_UNLOCKED,
97
	.port_rover	= 1024 - 1,
L
Linus Torvalds 已提交
98 99 100 101 102 103 104 105 106
};

/*
 * This array holds the first and last local port number.
 * For high-usage systems, use sysctl to change this to
 * 32768-61000
 */
int sysctl_local_port_range[2] = { 1024, 4999 };

107
static inline int inet_csk_bind_conflict(struct sock *sk, struct inet_bind_bucket *tb)
L
Linus Torvalds 已提交
108
{
109
	const u32 sk_rcv_saddr = inet_rcv_saddr(sk);
L
Linus Torvalds 已提交
110 111 112 113 114 115
	struct sock *sk2;
	struct hlist_node *node;
	int reuse = sk->sk_reuse;

	sk_for_each_bound(sk2, node, &tb->owners) {
		if (sk != sk2 &&
116
		    !inet_v6_ipv6only(sk2) &&
L
Linus Torvalds 已提交
117 118 119 120 121
		    (!sk->sk_bound_dev_if ||
		     !sk2->sk_bound_dev_if ||
		     sk->sk_bound_dev_if == sk2->sk_bound_dev_if)) {
			if (!reuse || !sk2->sk_reuse ||
			    sk2->sk_state == TCP_LISTEN) {
122
				const u32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
L
Linus Torvalds 已提交
123 124 125 126 127 128 129 130 131 132 133 134
				if (!sk2_rcv_saddr || !sk_rcv_saddr ||
				    sk2_rcv_saddr == sk_rcv_saddr)
					break;
			}
		}
	}
	return node != NULL;
}

/* Obtain a reference to a local port for the given sock,
 * if snum is zero it means select any available local port.
 */
135 136
int inet_csk_get_port(struct inet_hashinfo *hashinfo,
		      struct sock *sk, unsigned short snum)
L
Linus Torvalds 已提交
137
{
138
	struct inet_bind_hashbucket *head;
L
Linus Torvalds 已提交
139
	struct hlist_node *node;
140
	struct inet_bind_bucket *tb;
L
Linus Torvalds 已提交
141 142 143 144 145 146 147 148 149
	int ret;

	local_bh_disable();
	if (!snum) {
		int low = sysctl_local_port_range[0];
		int high = sysctl_local_port_range[1];
		int remaining = (high - low) + 1;
		int rover;

150 151
		spin_lock(&hashinfo->portalloc_lock);
		if (hashinfo->port_rover < low)
152 153
			rover = low;
		else
154
			rover = hashinfo->port_rover;
L
Linus Torvalds 已提交
155 156
		do {
			rover++;
157
			if (rover > high)
L
Linus Torvalds 已提交
158
				rover = low;
159
			head = &hashinfo->bhash[inet_bhashfn(rover, hashinfo->bhash_size)];
L
Linus Torvalds 已提交
160
			spin_lock(&head->lock);
161
			inet_bind_bucket_for_each(tb, node, &head->chain)
L
Linus Torvalds 已提交
162 163 164 165 166 167
				if (tb->port == rover)
					goto next;
			break;
		next:
			spin_unlock(&head->lock);
		} while (--remaining > 0);
168 169
		hashinfo->port_rover = rover;
		spin_unlock(&hashinfo->portalloc_lock);
L
Linus Torvalds 已提交
170

171 172 173 174 175 176
		/* Exhausted local port range during search?  It is not
		 * possible for us to be holding one of the bind hash
		 * locks if this test triggers, because if 'remaining'
		 * drops to zero, we broke out of the do/while loop at
		 * the top level, not from the 'break;' statement.
		 */
L
Linus Torvalds 已提交
177
		ret = 1;
178
		if (unlikely(remaining <= 0))
L
Linus Torvalds 已提交
179 180 181 182 183 184 185
			goto fail;

		/* OK, here is the one we will use.  HEAD is
		 * non-NULL and we hold it's mutex.
		 */
		snum = rover;
	} else {
186
		head = &hashinfo->bhash[inet_bhashfn(snum, hashinfo->bhash_size)];
L
Linus Torvalds 已提交
187
		spin_lock(&head->lock);
188
		inet_bind_bucket_for_each(tb, node, &head->chain)
L
Linus Torvalds 已提交
189 190 191 192 193 194 195 196 197 198 199 200 201 202
			if (tb->port == snum)
				goto tb_found;
	}
	tb = NULL;
	goto tb_not_found;
tb_found:
	if (!hlist_empty(&tb->owners)) {
		if (sk->sk_reuse > 1)
			goto success;
		if (tb->fastreuse > 0 &&
		    sk->sk_reuse && sk->sk_state != TCP_LISTEN) {
			goto success;
		} else {
			ret = 1;
203
			if (inet_csk_bind_conflict(sk, tb))
L
Linus Torvalds 已提交
204 205 206 207 208
				goto fail_unlock;
		}
	}
tb_not_found:
	ret = 1;
209
	if (!tb && (tb = inet_bind_bucket_create(hashinfo->bind_bucket_cachep, head, snum)) == NULL)
L
Linus Torvalds 已提交
210 211 212 213 214 215 216 217 218 219
		goto fail_unlock;
	if (hlist_empty(&tb->owners)) {
		if (sk->sk_reuse && sk->sk_state != TCP_LISTEN)
			tb->fastreuse = 1;
		else
			tb->fastreuse = 0;
	} else if (tb->fastreuse &&
		   (!sk->sk_reuse || sk->sk_state == TCP_LISTEN))
		tb->fastreuse = 0;
success:
220
	if (!inet_csk(sk)->icsk_bind_hash)
221
		inet_bind_hash(sk, tb, snum);
222
	BUG_TRAP(inet_csk(sk)->icsk_bind_hash == tb);
L
Linus Torvalds 已提交
223 224 225 226 227 228 229 230 231
 	ret = 0;

fail_unlock:
	spin_unlock(&head->lock);
fail:
	local_bh_enable();
	return ret;
}

232 233 234 235 236
static int tcp_v4_get_port(struct sock *sk, unsigned short snum)
{
	return inet_csk_get_port(&tcp_hashinfo, sk, snum);
}

L
Linus Torvalds 已提交
237 238
static void tcp_v4_hash(struct sock *sk)
{
239
	inet_hash(&tcp_hashinfo, sk);
L
Linus Torvalds 已提交
240 241 242 243
}

void tcp_unhash(struct sock *sk)
{
244
	inet_unhash(&tcp_hashinfo, sk);
L
Linus Torvalds 已提交
245 246 247 248 249 250 251 252 253 254 255 256
}

static inline __u32 tcp_v4_init_sequence(struct sock *sk, struct sk_buff *skb)
{
	return secure_tcp_sequence_number(skb->nh.iph->daddr,
					  skb->nh.iph->saddr,
					  skb->h.th->dest,
					  skb->h.th->source);
}

/* called with local bh disabled */
static int __tcp_v4_check_established(struct sock *sk, __u16 lport,
257
				      struct inet_timewait_sock **twp)
L
Linus Torvalds 已提交
258 259 260 261 262
{
	struct inet_sock *inet = inet_sk(sk);
	u32 daddr = inet->rcv_saddr;
	u32 saddr = inet->daddr;
	int dif = sk->sk_bound_dev_if;
263 264
	INET_ADDR_COOKIE(acookie, saddr, daddr)
	const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
265 266
	const int hash = inet_ehashfn(daddr, lport, saddr, inet->dport, tcp_hashinfo.ehash_size);
	struct inet_ehash_bucket *head = &tcp_hashinfo.ehash[hash];
L
Linus Torvalds 已提交
267
	struct sock *sk2;
268 269
	const struct hlist_node *node;
	struct inet_timewait_sock *tw;
L
Linus Torvalds 已提交
270 271 272 273

	write_lock(&head->lock);

	/* Check TIME-WAIT sockets first. */
274
	sk_for_each(sk2, node, &(head + tcp_hashinfo.ehash_size)->chain) {
275
		tw = inet_twsk(sk2);
L
Linus Torvalds 已提交
276

277 278
		if (INET_TW_MATCH(sk2, acookie, saddr, daddr, ports, dif)) {
			const struct tcp_timewait_sock *tcptw = tcp_twsk(sk2);
L
Linus Torvalds 已提交
279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294
			struct tcp_sock *tp = tcp_sk(sk);

			/* With PAWS, it is safe from the viewpoint
			   of data integrity. Even without PAWS it
			   is safe provided sequence spaces do not
			   overlap i.e. at data rates <= 80Mbit/sec.

			   Actually, the idea is close to VJ's one,
			   only timestamp cache is held not per host,
			   but per port pair and TW bucket is used
			   as state holder.

			   If TW bucket has been already destroyed we
			   fall back to VJ's scheme and use initial
			   timestamp retrieved from peer table.
			 */
295
			if (tcptw->tw_ts_recent_stamp &&
L
Linus Torvalds 已提交
296 297
			    (!twp || (sysctl_tcp_tw_reuse &&
				      xtime.tv_sec -
298 299 300
				      tcptw->tw_ts_recent_stamp > 1))) {
				tp->write_seq = tcptw->tw_snd_nxt + 65535 + 2;
				if (tp->write_seq == 0)
L
Linus Torvalds 已提交
301
					tp->write_seq = 1;
302 303
				tp->rx_opt.ts_recent	   = tcptw->tw_ts_recent;
				tp->rx_opt.ts_recent_stamp = tcptw->tw_ts_recent_stamp;
L
Linus Torvalds 已提交
304 305 306 307 308 309 310 311 312 313
				sock_hold(sk2);
				goto unique;
			} else
				goto not_unique;
		}
	}
	tw = NULL;

	/* And established part... */
	sk_for_each(sk2, node, &head->chain) {
314
		if (INET_MATCH(sk2, acookie, saddr, daddr, ports, dif))
L
Linus Torvalds 已提交
315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
			goto not_unique;
	}

unique:
	/* Must record num and sport now. Otherwise we will see
	 * in hash table socket with a funny identity. */
	inet->num = lport;
	inet->sport = htons(lport);
	sk->sk_hashent = hash;
	BUG_TRAP(sk_unhashed(sk));
	__sk_add_node(sk, &head->chain);
	sock_prot_inc_use(sk->sk_prot);
	write_unlock(&head->lock);

	if (twp) {
		*twp = tw;
		NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
	} else if (tw) {
		/* Silly. Should hash-dance instead... */
		tcp_tw_deschedule(tw);
		NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);

337
		inet_twsk_put(tw);
L
Linus Torvalds 已提交
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359
	}

	return 0;

not_unique:
	write_unlock(&head->lock);
	return -EADDRNOTAVAIL;
}

static inline u32 connect_port_offset(const struct sock *sk)
{
	const struct inet_sock *inet = inet_sk(sk);

	return secure_tcp_port_ephemeral(inet->rcv_saddr, inet->daddr, 
					 inet->dport);
}

/*
 * Bind a port for a connect operation and hash it.
 */
static inline int tcp_v4_hash_connect(struct sock *sk)
{
360 361 362
	const unsigned short snum = inet_sk(sk)->num;
 	struct inet_bind_hashbucket *head;
 	struct inet_bind_bucket *tb;
L
Linus Torvalds 已提交
363 364 365 366 367 368 369 370 371 372 373
	int ret;

 	if (!snum) {
 		int low = sysctl_local_port_range[0];
 		int high = sysctl_local_port_range[1];
		int range = high - low;
 		int i;
		int port;
		static u32 hint;
		u32 offset = hint + connect_port_offset(sk);
		struct hlist_node *node;
374
 		struct inet_timewait_sock *tw = NULL;
L
Linus Torvalds 已提交
375 376 377 378

 		local_bh_disable();
		for (i = 1; i <= range; i++) {
			port = low + (i + offset) % range;
379
 			head = &tcp_hashinfo.bhash[inet_bhashfn(port, tcp_hashinfo.bhash_size)];
L
Linus Torvalds 已提交
380 381 382 383 384 385
 			spin_lock(&head->lock);

 			/* Does not bother with rcv_saddr checks,
 			 * because the established check is already
 			 * unique enough.
 			 */
386
			inet_bind_bucket_for_each(tb, node, &head->chain) {
L
Linus Torvalds 已提交
387 388 389 390 391 392 393 394 395 396 397 398
 				if (tb->port == port) {
 					BUG_TRAP(!hlist_empty(&tb->owners));
 					if (tb->fastreuse >= 0)
 						goto next_port;
 					if (!__tcp_v4_check_established(sk,
									port,
									&tw))
 						goto ok;
 					goto next_port;
 				}
 			}

399
 			tb = inet_bind_bucket_create(tcp_hashinfo.bind_bucket_cachep, head, port);
L
Linus Torvalds 已提交
400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417
 			if (!tb) {
 				spin_unlock(&head->lock);
 				break;
 			}
 			tb->fastreuse = -1;
 			goto ok;

 		next_port:
 			spin_unlock(&head->lock);
 		}
 		local_bh_enable();

 		return -EADDRNOTAVAIL;

ok:
		hint += i;

 		/* Head lock still held and bh's disabled */
418
 		inet_bind_hash(sk, tb, port);
L
Linus Torvalds 已提交
419 420
		if (sk_unhashed(sk)) {
 			inet_sk(sk)->sport = htons(port);
421
 			__inet_hash(&tcp_hashinfo, sk, 0);
L
Linus Torvalds 已提交
422 423 424 425 426
 		}
 		spin_unlock(&head->lock);

 		if (tw) {
 			tcp_tw_deschedule(tw);
427
 			inet_twsk_put(tw);
L
Linus Torvalds 已提交
428 429 430 431 432 433
 		}

		ret = 0;
		goto out;
 	}

434
 	head = &tcp_hashinfo.bhash[inet_bhashfn(snum, tcp_hashinfo.bhash_size)];
435
 	tb  = inet_csk(sk)->icsk_bind_hash;
L
Linus Torvalds 已提交
436 437
	spin_lock_bh(&head->lock);
	if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
438
		__inet_hash(&tcp_hashinfo, sk, 0);
L
Linus Torvalds 已提交
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539
		spin_unlock_bh(&head->lock);
		return 0;
	} else {
		spin_unlock(&head->lock);
		/* No definite answer... Walk to established hash table */
		ret = __tcp_v4_check_established(sk, snum, NULL);
out:
		local_bh_enable();
		return ret;
	}
}

/* This will initiate an outgoing connection. */
int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
	struct inet_sock *inet = inet_sk(sk);
	struct tcp_sock *tp = tcp_sk(sk);
	struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
	struct rtable *rt;
	u32 daddr, nexthop;
	int tmp;
	int err;

	if (addr_len < sizeof(struct sockaddr_in))
		return -EINVAL;

	if (usin->sin_family != AF_INET)
		return -EAFNOSUPPORT;

	nexthop = daddr = usin->sin_addr.s_addr;
	if (inet->opt && inet->opt->srr) {
		if (!daddr)
			return -EINVAL;
		nexthop = inet->opt->faddr;
	}

	tmp = ip_route_connect(&rt, nexthop, inet->saddr,
			       RT_CONN_FLAGS(sk), sk->sk_bound_dev_if,
			       IPPROTO_TCP,
			       inet->sport, usin->sin_port, sk);
	if (tmp < 0)
		return tmp;

	if (rt->rt_flags & (RTCF_MULTICAST | RTCF_BROADCAST)) {
		ip_rt_put(rt);
		return -ENETUNREACH;
	}

	if (!inet->opt || !inet->opt->srr)
		daddr = rt->rt_dst;

	if (!inet->saddr)
		inet->saddr = rt->rt_src;
	inet->rcv_saddr = inet->saddr;

	if (tp->rx_opt.ts_recent_stamp && inet->daddr != daddr) {
		/* Reset inherited state */
		tp->rx_opt.ts_recent	   = 0;
		tp->rx_opt.ts_recent_stamp = 0;
		tp->write_seq		   = 0;
	}

	if (sysctl_tcp_tw_recycle &&
	    !tp->rx_opt.ts_recent_stamp && rt->rt_dst == daddr) {
		struct inet_peer *peer = rt_get_peer(rt);

		/* VJ's idea. We save last timestamp seen from
		 * the destination in peer table, when entering state TIME-WAIT
		 * and initialize rx_opt.ts_recent from it, when trying new connection.
		 */

		if (peer && peer->tcp_ts_stamp + TCP_PAWS_MSL >= xtime.tv_sec) {
			tp->rx_opt.ts_recent_stamp = peer->tcp_ts_stamp;
			tp->rx_opt.ts_recent = peer->tcp_ts;
		}
	}

	inet->dport = usin->sin_port;
	inet->daddr = daddr;

	tp->ext_header_len = 0;
	if (inet->opt)
		tp->ext_header_len = inet->opt->optlen;

	tp->rx_opt.mss_clamp = 536;

	/* Socket identity is still unknown (sport may be zero).
	 * However we set state to SYN-SENT and not releasing socket
	 * lock select source port, enter ourselves into the hash tables and
	 * complete initialization after this.
	 */
	tcp_set_state(sk, TCP_SYN_SENT);
	err = tcp_v4_hash_connect(sk);
	if (err)
		goto failure;

	err = ip_route_newports(&rt, inet->sport, inet->dport, sk);
	if (err)
		goto failure;

	/* OK, now commit destination to socket.  */
540
	sk_setup_caps(sk, &rt->u.dst);
L
Linus Torvalds 已提交
541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565

	if (!tp->write_seq)
		tp->write_seq = secure_tcp_sequence_number(inet->saddr,
							   inet->daddr,
							   inet->sport,
							   usin->sin_port);

	inet->id = tp->write_seq ^ jiffies;

	err = tcp_connect(sk);
	rt = NULL;
	if (err)
		goto failure;

	return 0;

failure:
	/* This unhashes the socket and releases the local port, if necessary. */
	tcp_set_state(sk, TCP_CLOSE);
	ip_rt_put(rt);
	sk->sk_route_caps = 0;
	inet->dport = 0;
	return err;
}

566
static inline int inet_iif(const struct sk_buff *skb)
L
Linus Torvalds 已提交
567 568 569 570
{
	return ((struct rtable *)skb->dst)->rt_iif;
}

571 572
static inline u32 inet_synq_hash(const u32 raddr, const u16 rport,
				 const u32 rnd, const u16 synq_hsize)
L
Linus Torvalds 已提交
573
{
574
	return jhash_2words(raddr, (u32)rport, rnd) & (synq_hsize - 1);
L
Linus Torvalds 已提交
575 576
}

577 578 579 580
struct request_sock *inet_csk_search_req(const struct sock *sk,
					 struct request_sock ***prevp,
					 const __u16 rport, const __u32 raddr,
					 const __u32 laddr)
L
Linus Torvalds 已提交
581
{
582 583
	const struct inet_connection_sock *icsk = inet_csk(sk);
	struct listen_sock *lopt = icsk->icsk_accept_queue.listen_opt;
584
	struct request_sock *req, **prev;
L
Linus Torvalds 已提交
585

586 587
	for (prev = &lopt->syn_table[inet_synq_hash(raddr, rport, lopt->hash_rnd,
						    lopt->nr_table_entries)];
L
Linus Torvalds 已提交
588 589
	     (req = *prev) != NULL;
	     prev = &req->dl_next) {
590 591 592 593 594
		const struct inet_request_sock *ireq = inet_rsk(req);

		if (ireq->rmt_port == rport &&
		    ireq->rmt_addr == raddr &&
		    ireq->loc_addr == laddr &&
595
		    AF_INET_FAMILY(req->rsk_ops->family)) {
L
Linus Torvalds 已提交
596 597 598 599 600 601 602 603 604
			BUG_TRAP(!req->sk);
			*prevp = prev;
			break;
		}
	}

	return req;
}

605
static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req)
L
Linus Torvalds 已提交
606
{
607 608 609 610
	struct inet_connection_sock *icsk = inet_csk(sk);
	struct listen_sock *lopt = icsk->icsk_accept_queue.listen_opt;
	const u32 h = inet_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port,
				     lopt->hash_rnd, lopt->nr_table_entries);
L
Linus Torvalds 已提交
611

612 613
	reqsk_queue_hash_req(&icsk->icsk_accept_queue, h, req, TCP_TIMEOUT_INIT);
	inet_csk_reqsk_queue_added(sk, TCP_TIMEOUT_INIT);
L
Linus Torvalds 已提交
614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698
}


/*
 * This routine does path mtu discovery as defined in RFC1191.
 */
static inline void do_pmtu_discovery(struct sock *sk, struct iphdr *iph,
				     u32 mtu)
{
	struct dst_entry *dst;
	struct inet_sock *inet = inet_sk(sk);
	struct tcp_sock *tp = tcp_sk(sk);

	/* We are not interested in TCP_LISTEN and open_requests (SYN-ACKs
	 * send out by Linux are always <576bytes so they should go through
	 * unfragmented).
	 */
	if (sk->sk_state == TCP_LISTEN)
		return;

	/* We don't check in the destentry if pmtu discovery is forbidden
	 * on this route. We just assume that no packet_to_big packets
	 * are send back when pmtu discovery is not active.
     	 * There is a small race when the user changes this flag in the
	 * route, but I think that's acceptable.
	 */
	if ((dst = __sk_dst_check(sk, 0)) == NULL)
		return;

	dst->ops->update_pmtu(dst, mtu);

	/* Something is about to be wrong... Remember soft error
	 * for the case, if this connection will not able to recover.
	 */
	if (mtu < dst_mtu(dst) && ip_dont_fragment(sk, dst))
		sk->sk_err_soft = EMSGSIZE;

	mtu = dst_mtu(dst);

	if (inet->pmtudisc != IP_PMTUDISC_DONT &&
	    tp->pmtu_cookie > mtu) {
		tcp_sync_mss(sk, mtu);

		/* Resend the TCP packet because it's
		 * clear that the old packet has been
		 * dropped. This is the new "fast" path mtu
		 * discovery.
		 */
		tcp_simple_retransmit(sk);
	} /* else let the usual retransmit timer handle it */
}

/*
 * This routine is called by the ICMP module when it gets some
 * sort of error condition.  If err < 0 then the socket should
 * be closed and the error returned to the user.  If err > 0
 * it's just the icmp type << 8 | icmp code.  After adjustment
 * header points to the first 8 bytes of the tcp header.  We need
 * to find the appropriate port.
 *
 * The locking strategy used here is very "optimistic". When
 * someone else accesses the socket the ICMP is just dropped
 * and for some paths there is no check at all.
 * A more general error queue to queue errors for later handling
 * is probably better.
 *
 */

void tcp_v4_err(struct sk_buff *skb, u32 info)
{
	struct iphdr *iph = (struct iphdr *)skb->data;
	struct tcphdr *th = (struct tcphdr *)(skb->data + (iph->ihl << 2));
	struct tcp_sock *tp;
	struct inet_sock *inet;
	int type = skb->h.icmph->type;
	int code = skb->h.icmph->code;
	struct sock *sk;
	__u32 seq;
	int err;

	if (skb->len < (iph->ihl << 2) + 8) {
		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
		return;
	}

699
	sk = inet_lookup(&tcp_hashinfo, iph->daddr, th->dest, iph->saddr,
700
			 th->source, inet_iif(skb));
L
Linus Torvalds 已提交
701 702 703 704 705
	if (!sk) {
		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
		return;
	}
	if (sk->sk_state == TCP_TIME_WAIT) {
706
		inet_twsk_put((struct inet_timewait_sock *)sk);
L
Linus Torvalds 已提交
707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754
		return;
	}

	bh_lock_sock(sk);
	/* If too many ICMPs get dropped on busy
	 * servers this needs to be solved differently.
	 */
	if (sock_owned_by_user(sk))
		NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);

	if (sk->sk_state == TCP_CLOSE)
		goto out;

	tp = tcp_sk(sk);
	seq = ntohl(th->seq);
	if (sk->sk_state != TCP_LISTEN &&
	    !between(seq, tp->snd_una, tp->snd_nxt)) {
		NET_INC_STATS(LINUX_MIB_OUTOFWINDOWICMPS);
		goto out;
	}

	switch (type) {
	case ICMP_SOURCE_QUENCH:
		/* Just silently ignore these. */
		goto out;
	case ICMP_PARAMETERPROB:
		err = EPROTO;
		break;
	case ICMP_DEST_UNREACH:
		if (code > NR_ICMP_UNREACH)
			goto out;

		if (code == ICMP_FRAG_NEEDED) { /* PMTU discovery (RFC1191) */
			if (!sock_owned_by_user(sk))
				do_pmtu_discovery(sk, iph, info);
			goto out;
		}

		err = icmp_err_convert[code].errno;
		break;
	case ICMP_TIME_EXCEEDED:
		err = EHOSTUNREACH;
		break;
	default:
		goto out;
	}

	switch (sk->sk_state) {
755
		struct request_sock *req, **prev;
L
Linus Torvalds 已提交
756 757 758 759
	case TCP_LISTEN:
		if (sock_owned_by_user(sk))
			goto out;

760 761
		req = inet_csk_search_req(sk, &prev, th->dest,
					  iph->daddr, iph->saddr);
L
Linus Torvalds 已提交
762 763 764 765 766 767 768 769
		if (!req)
			goto out;

		/* ICMPs are not backlogged, hence we cannot get
		   an established socket here.
		 */
		BUG_TRAP(!req->sk);

770
		if (seq != tcp_rsk(req)->snt_isn) {
L
Linus Torvalds 已提交
771 772 773 774 775 776 777 778 779 780
			NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
			goto out;
		}

		/*
		 * Still in SYN_RECV, just remove it silently.
		 * There is no good way to pass the error to the newly
		 * created socket, and POSIX does not want network
		 * errors returned from accept().
		 */
781
		inet_csk_reqsk_queue_drop(sk, req, prev);
L
Linus Torvalds 已提交
782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950
		goto out;

	case TCP_SYN_SENT:
	case TCP_SYN_RECV:  /* Cannot happen.
			       It can f.e. if SYNs crossed.
			     */
		if (!sock_owned_by_user(sk)) {
			TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS);
			sk->sk_err = err;

			sk->sk_error_report(sk);

			tcp_done(sk);
		} else {
			sk->sk_err_soft = err;
		}
		goto out;
	}

	/* If we've already connected we will keep trying
	 * until we time out, or the user gives up.
	 *
	 * rfc1122 4.2.3.9 allows to consider as hard errors
	 * only PROTO_UNREACH and PORT_UNREACH (well, FRAG_FAILED too,
	 * but it is obsoleted by pmtu discovery).
	 *
	 * Note, that in modern internet, where routing is unreliable
	 * and in each dark corner broken firewalls sit, sending random
	 * errors ordered by their masters even this two messages finally lose
	 * their original sense (even Linux sends invalid PORT_UNREACHs)
	 *
	 * Now we are in compliance with RFCs.
	 *							--ANK (980905)
	 */

	inet = inet_sk(sk);
	if (!sock_owned_by_user(sk) && inet->recverr) {
		sk->sk_err = err;
		sk->sk_error_report(sk);
	} else	{ /* Only an error on timeout */
		sk->sk_err_soft = err;
	}

out:
	bh_unlock_sock(sk);
	sock_put(sk);
}

/* This routine computes an IPv4 TCP checksum. */
void tcp_v4_send_check(struct sock *sk, struct tcphdr *th, int len,
		       struct sk_buff *skb)
{
	struct inet_sock *inet = inet_sk(sk);

	if (skb->ip_summed == CHECKSUM_HW) {
		th->check = ~tcp_v4_check(th, len, inet->saddr, inet->daddr, 0);
		skb->csum = offsetof(struct tcphdr, check);
	} else {
		th->check = tcp_v4_check(th, len, inet->saddr, inet->daddr,
					 csum_partial((char *)th,
						      th->doff << 2,
						      skb->csum));
	}
}

/*
 *	This routine will send an RST to the other tcp.
 *
 *	Someone asks: why I NEVER use socket parameters (TOS, TTL etc.)
 *		      for reset.
 *	Answer: if a packet caused RST, it is not for a socket
 *		existing in our system, if it is matched to a socket,
 *		it is just duplicate segment or bug in other side's TCP.
 *		So that we build reply only basing on parameters
 *		arrived with segment.
 *	Exception: precedence violation. We do not implement it in any case.
 */

static void tcp_v4_send_reset(struct sk_buff *skb)
{
	struct tcphdr *th = skb->h.th;
	struct tcphdr rth;
	struct ip_reply_arg arg;

	/* Never send a reset in response to a reset. */
	if (th->rst)
		return;

	if (((struct rtable *)skb->dst)->rt_type != RTN_LOCAL)
		return;

	/* Swap the send and the receive. */
	memset(&rth, 0, sizeof(struct tcphdr));
	rth.dest   = th->source;
	rth.source = th->dest;
	rth.doff   = sizeof(struct tcphdr) / 4;
	rth.rst    = 1;

	if (th->ack) {
		rth.seq = th->ack_seq;
	} else {
		rth.ack = 1;
		rth.ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin +
				    skb->len - (th->doff << 2));
	}

	memset(&arg, 0, sizeof arg);
	arg.iov[0].iov_base = (unsigned char *)&rth;
	arg.iov[0].iov_len  = sizeof rth;
	arg.csum = csum_tcpudp_nofold(skb->nh.iph->daddr,
				      skb->nh.iph->saddr, /*XXX*/
				      sizeof(struct tcphdr), IPPROTO_TCP, 0);
	arg.csumoffset = offsetof(struct tcphdr, check) / 2;

	ip_send_reply(tcp_socket->sk, skb, &arg, sizeof rth);

	TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
	TCP_INC_STATS_BH(TCP_MIB_OUTRSTS);
}

/* The code following below sending ACKs in SYN-RECV and TIME-WAIT states
   outside socket context is ugly, certainly. What can I do?
 */

static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack,
			    u32 win, u32 ts)
{
	struct tcphdr *th = skb->h.th;
	struct {
		struct tcphdr th;
		u32 tsopt[3];
	} rep;
	struct ip_reply_arg arg;

	memset(&rep.th, 0, sizeof(struct tcphdr));
	memset(&arg, 0, sizeof arg);

	arg.iov[0].iov_base = (unsigned char *)&rep;
	arg.iov[0].iov_len  = sizeof(rep.th);
	if (ts) {
		rep.tsopt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
				     (TCPOPT_TIMESTAMP << 8) |
				     TCPOLEN_TIMESTAMP);
		rep.tsopt[1] = htonl(tcp_time_stamp);
		rep.tsopt[2] = htonl(ts);
		arg.iov[0].iov_len = sizeof(rep);
	}

	/* Swap the send and the receive. */
	rep.th.dest    = th->source;
	rep.th.source  = th->dest;
	rep.th.doff    = arg.iov[0].iov_len / 4;
	rep.th.seq     = htonl(seq);
	rep.th.ack_seq = htonl(ack);
	rep.th.ack     = 1;
	rep.th.window  = htons(win);

	arg.csum = csum_tcpudp_nofold(skb->nh.iph->daddr,
				      skb->nh.iph->saddr, /*XXX*/
				      arg.iov[0].iov_len, IPPROTO_TCP, 0);
	arg.csumoffset = offsetof(struct tcphdr, check) / 2;

	ip_send_reply(tcp_socket->sk, skb, &arg, arg.iov[0].iov_len);

	TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
}

static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
{
951 952
	struct inet_timewait_sock *tw = inet_twsk(sk);
	const struct tcp_timewait_sock *tcptw = tcp_twsk(sk);
L
Linus Torvalds 已提交
953

954 955
	tcp_v4_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale, tcptw->tw_ts_recent);
L
Linus Torvalds 已提交
956

957
	inet_twsk_put(tw);
L
Linus Torvalds 已提交
958 959
}

960
static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req)
L
Linus Torvalds 已提交
961
{
962
	tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd,
L
Linus Torvalds 已提交
963 964 965
			req->ts_recent);
}

966 967
struct dst_entry* inet_csk_route_req(struct sock *sk,
				     const struct request_sock *req)
L
Linus Torvalds 已提交
968 969
{
	struct rtable *rt;
970 971
	const struct inet_request_sock *ireq = inet_rsk(req);
	struct ip_options *opt = inet_rsk(req)->opt;
L
Linus Torvalds 已提交
972 973 974 975
	struct flowi fl = { .oif = sk->sk_bound_dev_if,
			    .nl_u = { .ip4_u =
				      { .daddr = ((opt && opt->srr) ?
						  opt->faddr :
976 977
						  ireq->rmt_addr),
					.saddr = ireq->loc_addr,
L
Linus Torvalds 已提交
978
					.tos = RT_CONN_FLAGS(sk) } },
979
			    .proto = sk->sk_protocol,
L
Linus Torvalds 已提交
980 981
			    .uli_u = { .ports =
				       { .sport = inet_sk(sk)->sport,
982
					 .dport = ireq->rmt_port } } };
L
Linus Torvalds 已提交
983 984 985 986 987 988 989 990 991 992 993 994 995 996 997

	if (ip_route_output_flow(&rt, &fl, sk, 0)) {
		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
		return NULL;
	}
	if (opt && opt->is_strictroute && rt->rt_dst != rt->rt_gateway) {
		ip_rt_put(rt);
		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
		return NULL;
	}
	return &rt->u.dst;
}

/*
 *	Send a SYN-ACK after having received an ACK.
998
 *	This still operates on a request_sock only, not on a big
L
Linus Torvalds 已提交
999 1000
 *	socket.
 */
1001
static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
L
Linus Torvalds 已提交
1002 1003
			      struct dst_entry *dst)
{
1004
	const struct inet_request_sock *ireq = inet_rsk(req);
L
Linus Torvalds 已提交
1005 1006 1007 1008
	int err = -1;
	struct sk_buff * skb;

	/* First, grab a route. */
1009
	if (!dst && (dst = inet_csk_route_req(sk, req)) == NULL)
L
Linus Torvalds 已提交
1010 1011 1012 1013 1014 1015 1016 1017
		goto out;

	skb = tcp_make_synack(sk, dst, req);

	if (skb) {
		struct tcphdr *th = skb->h.th;

		th->check = tcp_v4_check(th, skb->len,
1018 1019
					 ireq->loc_addr,
					 ireq->rmt_addr,
L
Linus Torvalds 已提交
1020 1021 1022
					 csum_partial((char *)th, skb->len,
						      skb->csum));

1023 1024 1025
		err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr,
					    ireq->rmt_addr,
					    ireq->opt);
L
Linus Torvalds 已提交
1026 1027 1028 1029 1030 1031 1032 1033 1034 1035
		if (err == NET_XMIT_CN)
			err = 0;
	}

out:
	dst_release(dst);
	return err;
}

/*
1036
 *	IPv4 request_sock destructor.
L
Linus Torvalds 已提交
1037
 */
1038
static void tcp_v4_reqsk_destructor(struct request_sock *req)
L
Linus Torvalds 已提交
1039
{
1040 1041
	if (inet_rsk(req)->opt)
		kfree(inet_rsk(req)->opt);
L
Linus Torvalds 已提交
1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056
}

static inline void syn_flood_warning(struct sk_buff *skb)
{
	static unsigned long warntime;

	if (time_after(jiffies, (warntime + HZ * 60))) {
		warntime = jiffies;
		printk(KERN_INFO
		       "possible SYN flooding on port %d. Sending cookies.\n",
		       ntohs(skb->h.th->dest));
	}
}

/*
1057
 * Save and compile IPv4 options into the request_sock if needed.
L
Linus Torvalds 已提交
1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077
 */
static inline struct ip_options *tcp_v4_save_options(struct sock *sk,
						     struct sk_buff *skb)
{
	struct ip_options *opt = &(IPCB(skb)->opt);
	struct ip_options *dopt = NULL;

	if (opt && opt->optlen) {
		int opt_size = optlength(opt);
		dopt = kmalloc(opt_size, GFP_ATOMIC);
		if (dopt) {
			if (ip_options_echo(dopt, skb)) {
				kfree(dopt);
				dopt = NULL;
			}
		}
	}
	return dopt;
}

1078
struct request_sock_ops tcp_request_sock_ops = {
L
Linus Torvalds 已提交
1079
	.family		=	PF_INET,
1080
	.obj_size	=	sizeof(struct tcp_request_sock),
L
Linus Torvalds 已提交
1081
	.rtx_syn_ack	=	tcp_v4_send_synack,
1082 1083
	.send_ack	=	tcp_v4_reqsk_send_ack,
	.destructor	=	tcp_v4_reqsk_destructor,
L
Linus Torvalds 已提交
1084 1085 1086 1087 1088
	.send_reset	=	tcp_v4_send_reset,
};

int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
{
1089
	struct inet_request_sock *ireq;
L
Linus Torvalds 已提交
1090
	struct tcp_options_received tmp_opt;
1091
	struct request_sock *req;
L
Linus Torvalds 已提交
1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110
	__u32 saddr = skb->nh.iph->saddr;
	__u32 daddr = skb->nh.iph->daddr;
	__u32 isn = TCP_SKB_CB(skb)->when;
	struct dst_entry *dst = NULL;
#ifdef CONFIG_SYN_COOKIES
	int want_cookie = 0;
#else
#define want_cookie 0 /* Argh, why doesn't gcc optimize this :( */
#endif

	/* Never answer to SYNs send to broadcast or multicast */
	if (((struct rtable *)skb->dst)->rt_flags &
	    (RTCF_BROADCAST | RTCF_MULTICAST))
		goto drop;

	/* TW buckets are converted to open requests without
	 * limitations, they conserve resources and peer is
	 * evidently real one.
	 */
1111
	if (inet_csk_reqsk_queue_is_full(sk) && !isn) {
L
Linus Torvalds 已提交
1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124
#ifdef CONFIG_SYN_COOKIES
		if (sysctl_tcp_syncookies) {
			want_cookie = 1;
		} else
#endif
		goto drop;
	}

	/* Accept backlog is full. If we have already queued enough
	 * of warm entries in syn queue, drop request. It is better than
	 * clogging syn queue with openreqs with exponentially increasing
	 * timeout.
	 */
1125
	if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1)
L
Linus Torvalds 已提交
1126 1127
		goto drop;

1128
	req = reqsk_alloc(&tcp_request_sock_ops);
L
Linus Torvalds 已提交
1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
	if (!req)
		goto drop;

	tcp_clear_options(&tmp_opt);
	tmp_opt.mss_clamp = 536;
	tmp_opt.user_mss  = tcp_sk(sk)->rx_opt.user_mss;

	tcp_parse_options(skb, &tmp_opt, 0);

	if (want_cookie) {
		tcp_clear_options(&tmp_opt);
		tmp_opt.saw_tstamp = 0;
	}

	if (tmp_opt.saw_tstamp && !tmp_opt.rcv_tsval) {
		/* Some OSes (unknown ones, but I see them on web server, which
		 * contains information interesting only for windows'
		 * users) do not send their stamp in SYN. It is easy case.
		 * We simply do not advertise TS support.
		 */
		tmp_opt.saw_tstamp = 0;
		tmp_opt.tstamp_ok  = 0;
	}
	tmp_opt.tstamp_ok = tmp_opt.saw_tstamp;

	tcp_openreq_init(req, &tmp_opt, skb);

1156 1157 1158 1159
	ireq = inet_rsk(req);
	ireq->loc_addr = daddr;
	ireq->rmt_addr = saddr;
	ireq->opt = tcp_v4_save_options(sk, skb);
L
Linus Torvalds 已提交
1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181
	if (!want_cookie)
		TCP_ECN_create_request(req, skb->h.th);

	if (want_cookie) {
#ifdef CONFIG_SYN_COOKIES
		syn_flood_warning(skb);
#endif
		isn = cookie_v4_init_sequence(sk, skb, &req->mss);
	} else if (!isn) {
		struct inet_peer *peer = NULL;

		/* VJ's idea. We save last timestamp seen
		 * from the destination in peer table, when entering
		 * state TIME-WAIT, and check against it before
		 * accepting new connection request.
		 *
		 * If "isn" is not zero, this request hit alive
		 * timewait bucket, so that all the necessary checks
		 * are made in the function processing timewait state.
		 */
		if (tmp_opt.saw_tstamp &&
		    sysctl_tcp_tw_recycle &&
1182
		    (dst = inet_csk_route_req(sk, req)) != NULL &&
L
Linus Torvalds 已提交
1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194
		    (peer = rt_get_peer((struct rtable *)dst)) != NULL &&
		    peer->v4daddr == saddr) {
			if (xtime.tv_sec < peer->tcp_ts_stamp + TCP_PAWS_MSL &&
			    (s32)(peer->tcp_ts - req->ts_recent) >
							TCP_PAWS_WINDOW) {
				NET_INC_STATS_BH(LINUX_MIB_PAWSPASSIVEREJECTED);
				dst_release(dst);
				goto drop_and_free;
			}
		}
		/* Kill the following clause, if you dislike this way. */
		else if (!sysctl_tcp_syncookies &&
1195
			 (sysctl_max_syn_backlog - inet_csk_reqsk_queue_len(sk) <
L
Linus Torvalds 已提交
1196 1197 1198 1199 1200 1201 1202 1203 1204 1205
			  (sysctl_max_syn_backlog >> 2)) &&
			 (!peer || !peer->tcp_ts_stamp) &&
			 (!dst || !dst_metric(dst, RTAX_RTT))) {
			/* Without syncookies last quarter of
			 * backlog is filled with destinations,
			 * proven to be alive.
			 * It means that we continue to communicate
			 * to destinations, already remembered
			 * to the moment of synflood.
			 */
H
Heikki Orsila 已提交
1206 1207 1208 1209 1210
			LIMIT_NETDEBUG(printk(KERN_DEBUG "TCP: drop open "
					      "request from %u.%u."
					      "%u.%u/%u\n",
					      NIPQUAD(saddr),
					      ntohs(skb->h.th->source)));
L
Linus Torvalds 已提交
1211 1212 1213 1214 1215 1216
			dst_release(dst);
			goto drop_and_free;
		}

		isn = tcp_v4_init_sequence(sk, skb);
	}
1217
	tcp_rsk(req)->snt_isn = isn;
L
Linus Torvalds 已提交
1218 1219 1220 1221 1222

	if (tcp_v4_send_synack(sk, req, dst))
		goto drop_and_free;

	if (want_cookie) {
1223
	   	reqsk_free(req);
L
Linus Torvalds 已提交
1224 1225 1226 1227 1228 1229
	} else {
		tcp_v4_synq_add(sk, req);
	}
	return 0;

drop_and_free:
1230
	reqsk_free(req);
L
Linus Torvalds 已提交
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241
drop:
	TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS);
	return 0;
}


/*
 * The three way handshake has completed - we got a valid synack -
 * now create the new socket.
 */
struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
1242
				  struct request_sock *req,
L
Linus Torvalds 已提交
1243 1244
				  struct dst_entry *dst)
{
1245
	struct inet_request_sock *ireq;
L
Linus Torvalds 已提交
1246 1247 1248 1249 1250 1251 1252
	struct inet_sock *newinet;
	struct tcp_sock *newtp;
	struct sock *newsk;

	if (sk_acceptq_is_full(sk))
		goto exit_overflow;

1253
	if (!dst && (dst = inet_csk_route_req(sk, req)) == NULL)
L
Linus Torvalds 已提交
1254 1255 1256 1257 1258 1259
		goto exit;

	newsk = tcp_create_openreq_child(sk, req, skb);
	if (!newsk)
		goto exit;

1260
	sk_setup_caps(newsk, dst);
L
Linus Torvalds 已提交
1261 1262 1263

	newtp		      = tcp_sk(newsk);
	newinet		      = inet_sk(newsk);
1264 1265 1266 1267 1268 1269
	ireq		      = inet_rsk(req);
	newinet->daddr	      = ireq->rmt_addr;
	newinet->rcv_saddr    = ireq->loc_addr;
	newinet->saddr	      = ireq->loc_addr;
	newinet->opt	      = ireq->opt;
	ireq->opt	      = NULL;
1270
	newinet->mc_index     = inet_iif(skb);
L
Linus Torvalds 已提交
1271 1272 1273 1274 1275 1276 1277 1278 1279 1280
	newinet->mc_ttl	      = skb->nh.iph->ttl;
	newtp->ext_header_len = 0;
	if (newinet->opt)
		newtp->ext_header_len = newinet->opt->optlen;
	newinet->id = newtp->write_seq ^ jiffies;

	tcp_sync_mss(newsk, dst_mtu(dst));
	newtp->advmss = dst_metric(dst, RTAX_ADVMSS);
	tcp_initialize_rcv_mss(newsk);

1281
	__inet_hash(&tcp_hashinfo, newsk, 0);
1282
	__inet_inherit_port(&tcp_hashinfo, sk, newsk);
L
Linus Torvalds 已提交
1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298

	return newsk;

exit_overflow:
	NET_INC_STATS_BH(LINUX_MIB_LISTENOVERFLOWS);
exit:
	NET_INC_STATS_BH(LINUX_MIB_LISTENDROPS);
	dst_release(dst);
	return NULL;
}

static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
{
	struct tcphdr *th = skb->h.th;
	struct iphdr *iph = skb->nh.iph;
	struct sock *nsk;
1299
	struct request_sock **prev;
L
Linus Torvalds 已提交
1300
	/* Find possible connection requests. */
1301 1302
	struct request_sock *req = inet_csk_search_req(sk, &prev, th->source,
						       iph->saddr, iph->daddr);
L
Linus Torvalds 已提交
1303 1304 1305
	if (req)
		return tcp_check_req(sk, skb, req, prev);

1306 1307
	nsk = __inet_lookup_established(&tcp_hashinfo, skb->nh.iph->saddr,
					th->source, skb->nh.iph->daddr,
1308
					ntohs(th->dest), inet_iif(skb));
L
Linus Torvalds 已提交
1309 1310 1311 1312 1313 1314

	if (nsk) {
		if (nsk->sk_state != TCP_TIME_WAIT) {
			bh_lock_sock(nsk);
			return nsk;
		}
1315
		inet_twsk_put((struct inet_timewait_sock *)nsk);
L
Linus Torvalds 已提交
1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333
		return NULL;
	}

#ifdef CONFIG_SYN_COOKIES
	if (!th->rst && !th->syn && th->ack)
		sk = cookie_v4_check(sk, skb, &(IPCB(skb)->opt));
#endif
	return sk;
}

static int tcp_v4_checksum_init(struct sk_buff *skb)
{
	if (skb->ip_summed == CHECKSUM_HW) {
		skb->ip_summed = CHECKSUM_UNNECESSARY;
		if (!tcp_v4_check(skb->h.th, skb->len, skb->nh.iph->saddr,
				  skb->nh.iph->daddr, skb->csum))
			return 0;

H
Heikki Orsila 已提交
1334
		LIMIT_NETDEBUG(printk(KERN_DEBUG "hw tcp v4 csum failed\n"));
L
Linus Torvalds 已提交
1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449
		skb->ip_summed = CHECKSUM_NONE;
	}
	if (skb->len <= 76) {
		if (tcp_v4_check(skb->h.th, skb->len, skb->nh.iph->saddr,
				 skb->nh.iph->daddr,
				 skb_checksum(skb, 0, skb->len, 0)))
			return -1;
		skb->ip_summed = CHECKSUM_UNNECESSARY;
	} else {
		skb->csum = ~tcp_v4_check(skb->h.th, skb->len,
					  skb->nh.iph->saddr,
					  skb->nh.iph->daddr, 0);
	}
	return 0;
}


/* The socket must have it's spinlock held when we get
 * here.
 *
 * We have a potential double-lock case here, so even when
 * doing backlog processing we use the BH locking scheme.
 * This is because we cannot sleep with the original spinlock
 * held.
 */
int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
{
	if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */
		TCP_CHECK_TIMER(sk);
		if (tcp_rcv_established(sk, skb, skb->h.th, skb->len))
			goto reset;
		TCP_CHECK_TIMER(sk);
		return 0;
	}

	if (skb->len < (skb->h.th->doff << 2) || tcp_checksum_complete(skb))
		goto csum_err;

	if (sk->sk_state == TCP_LISTEN) {
		struct sock *nsk = tcp_v4_hnd_req(sk, skb);
		if (!nsk)
			goto discard;

		if (nsk != sk) {
			if (tcp_child_process(sk, nsk, skb))
				goto reset;
			return 0;
		}
	}

	TCP_CHECK_TIMER(sk);
	if (tcp_rcv_state_process(sk, skb, skb->h.th, skb->len))
		goto reset;
	TCP_CHECK_TIMER(sk);
	return 0;

reset:
	tcp_v4_send_reset(skb);
discard:
	kfree_skb(skb);
	/* Be careful here. If this function gets more complicated and
	 * gcc suffers from register pressure on the x86, sk (in %ebx)
	 * might be destroyed here. This current version compiles correctly,
	 * but you have been warned.
	 */
	return 0;

csum_err:
	TCP_INC_STATS_BH(TCP_MIB_INERRS);
	goto discard;
}

/*
 *	From tcp_input.c
 */

int tcp_v4_rcv(struct sk_buff *skb)
{
	struct tcphdr *th;
	struct sock *sk;
	int ret;

	if (skb->pkt_type != PACKET_HOST)
		goto discard_it;

	/* Count it even if it's bad */
	TCP_INC_STATS_BH(TCP_MIB_INSEGS);

	if (!pskb_may_pull(skb, sizeof(struct tcphdr)))
		goto discard_it;

	th = skb->h.th;

	if (th->doff < sizeof(struct tcphdr) / 4)
		goto bad_packet;
	if (!pskb_may_pull(skb, th->doff * 4))
		goto discard_it;

	/* An explanation is required here, I think.
	 * Packet length and doff are validated by header prediction,
	 * provided case of th->doff==0 is elimineted.
	 * So, we defer the checks. */
	if ((skb->ip_summed != CHECKSUM_UNNECESSARY &&
	     tcp_v4_checksum_init(skb) < 0))
		goto bad_packet;

	th = skb->h.th;
	TCP_SKB_CB(skb)->seq = ntohl(th->seq);
	TCP_SKB_CB(skb)->end_seq = (TCP_SKB_CB(skb)->seq + th->syn + th->fin +
				    skb->len - th->doff * 4);
	TCP_SKB_CB(skb)->ack_seq = ntohl(th->ack_seq);
	TCP_SKB_CB(skb)->when	 = 0;
	TCP_SKB_CB(skb)->flags	 = skb->nh.iph->tos;
	TCP_SKB_CB(skb)->sacked	 = 0;

1450 1451
	sk = __inet_lookup(&tcp_hashinfo, skb->nh.iph->saddr, th->source,
			   skb->nh.iph->daddr, ntohs(th->dest),
1452
			   inet_iif(skb));
L
Linus Torvalds 已提交
1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503

	if (!sk)
		goto no_tcp_socket;

process:
	if (sk->sk_state == TCP_TIME_WAIT)
		goto do_time_wait;

	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
		goto discard_and_relse;

	if (sk_filter(sk, skb, 0))
		goto discard_and_relse;

	skb->dev = NULL;

	bh_lock_sock(sk);
	ret = 0;
	if (!sock_owned_by_user(sk)) {
		if (!tcp_prequeue(sk, skb))
			ret = tcp_v4_do_rcv(sk, skb);
	} else
		sk_add_backlog(sk, skb);
	bh_unlock_sock(sk);

	sock_put(sk);

	return ret;

no_tcp_socket:
	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
		goto discard_it;

	if (skb->len < (th->doff << 2) || tcp_checksum_complete(skb)) {
bad_packet:
		TCP_INC_STATS_BH(TCP_MIB_INERRS);
	} else {
		tcp_v4_send_reset(skb);
	}

discard_it:
	/* Discard frame. */
	kfree_skb(skb);
  	return 0;

discard_and_relse:
	sock_put(sk);
	goto discard_it;

do_time_wait:
	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) {
1504
		inet_twsk_put((struct inet_timewait_sock *) sk);
L
Linus Torvalds 已提交
1505 1506 1507 1508 1509
		goto discard_it;
	}

	if (skb->len < (th->doff << 2) || tcp_checksum_complete(skb)) {
		TCP_INC_STATS_BH(TCP_MIB_INERRS);
1510
		inet_twsk_put((struct inet_timewait_sock *) sk);
L
Linus Torvalds 已提交
1511 1512
		goto discard_it;
	}
1513 1514
	switch (tcp_timewait_state_process((struct inet_timewait_sock *)sk,
					   skb, th)) {
L
Linus Torvalds 已提交
1515
	case TCP_TW_SYN: {
1516 1517 1518
		struct sock *sk2 = inet_lookup_listener(&tcp_hashinfo,
							skb->nh.iph->daddr,
							ntohs(th->dest),
1519
							inet_iif(skb));
L
Linus Torvalds 已提交
1520
		if (sk2) {
1521 1522
			tcp_tw_deschedule((struct inet_timewait_sock *)sk);
			inet_twsk_put((struct inet_timewait_sock *)sk);
L
Linus Torvalds 已提交
1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585
			sk = sk2;
			goto process;
		}
		/* Fall through to ACK */
	}
	case TCP_TW_ACK:
		tcp_v4_timewait_ack(sk, skb);
		break;
	case TCP_TW_RST:
		goto no_tcp_socket;
	case TCP_TW_SUCCESS:;
	}
	goto discard_it;
}

static void v4_addr2sockaddr(struct sock *sk, struct sockaddr * uaddr)
{
	struct sockaddr_in *sin = (struct sockaddr_in *) uaddr;
	struct inet_sock *inet = inet_sk(sk);

	sin->sin_family		= AF_INET;
	sin->sin_addr.s_addr	= inet->daddr;
	sin->sin_port		= inet->dport;
}

/* VJ's idea. Save last timestamp seen from this destination
 * and hold it at least for normal timewait interval to use for duplicate
 * segment detection in subsequent connections, before they enter synchronized
 * state.
 */

int tcp_v4_remember_stamp(struct sock *sk)
{
	struct inet_sock *inet = inet_sk(sk);
	struct tcp_sock *tp = tcp_sk(sk);
	struct rtable *rt = (struct rtable *)__sk_dst_get(sk);
	struct inet_peer *peer = NULL;
	int release_it = 0;

	if (!rt || rt->rt_dst != inet->daddr) {
		peer = inet_getpeer(inet->daddr, 1);
		release_it = 1;
	} else {
		if (!rt->peer)
			rt_bind_peer(rt, 1);
		peer = rt->peer;
	}

	if (peer) {
		if ((s32)(peer->tcp_ts - tp->rx_opt.ts_recent) <= 0 ||
		    (peer->tcp_ts_stamp + TCP_PAWS_MSL < xtime.tv_sec &&
		     peer->tcp_ts_stamp <= tp->rx_opt.ts_recent_stamp)) {
			peer->tcp_ts_stamp = tp->rx_opt.ts_recent_stamp;
			peer->tcp_ts = tp->rx_opt.ts_recent;
		}
		if (release_it)
			inet_putpeer(peer);
		return 1;
	}

	return 0;
}

1586
int tcp_v4_tw_remember_stamp(struct inet_timewait_sock *tw)
L
Linus Torvalds 已提交
1587
{
1588
	struct inet_peer *peer = inet_getpeer(tw->tw_daddr, 1);
L
Linus Torvalds 已提交
1589 1590

	if (peer) {
1591 1592 1593
		const struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw);

		if ((s32)(peer->tcp_ts - tcptw->tw_ts_recent) <= 0 ||
L
Linus Torvalds 已提交
1594
		    (peer->tcp_ts_stamp + TCP_PAWS_MSL < xtime.tv_sec &&
1595 1596 1597
		     peer->tcp_ts_stamp <= tcptw->tw_ts_recent_stamp)) {
			peer->tcp_ts_stamp = tcptw->tw_ts_recent_stamp;
			peer->tcp_ts	   = tcptw->tw_ts_recent;
L
Linus Torvalds 已提交
1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608
		}
		inet_putpeer(peer);
		return 1;
	}

	return 0;
}

struct tcp_func ipv4_specific = {
	.queue_xmit	=	ip_queue_xmit,
	.send_check	=	tcp_v4_send_check,
1609
	.rebuild_header	=	inet_sk_rebuild_header,
L
Linus Torvalds 已提交
1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630
	.conn_request	=	tcp_v4_conn_request,
	.syn_recv_sock	=	tcp_v4_syn_recv_sock,
	.remember_stamp	=	tcp_v4_remember_stamp,
	.net_header_len	=	sizeof(struct iphdr),
	.setsockopt	=	ip_setsockopt,
	.getsockopt	=	ip_getsockopt,
	.addr2sockaddr	=	v4_addr2sockaddr,
	.sockaddr_len	=	sizeof(struct sockaddr_in),
};

/* NOTE: A lot of things set to zero explicitly by call to
 *       sk_alloc() so need not be done here.
 */
static int tcp_v4_init_sock(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	skb_queue_head_init(&tp->out_of_order_queue);
	tcp_init_xmit_timers(sk);
	tcp_prequeue_init(tp);

1631
	inet_csk(sk)->icsk_rto = TCP_TIMEOUT_INIT;
L
Linus Torvalds 已提交
1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
	tp->mdev = TCP_TIMEOUT_INIT;

	/* So many TCP implementations out there (incorrectly) count the
	 * initial SYN frame in their delayed-ACK and congestion control
	 * algorithms that we must have the following bandaid to talk
	 * efficiently to them.  -DaveM
	 */
	tp->snd_cwnd = 2;

	/* See draft-stevens-tcpca-spec-01 for discussion of the
	 * initialization of these values.
	 */
	tp->snd_ssthresh = 0x7fffffff;	/* Infinity */
	tp->snd_cwnd_clamp = ~0;
1646
	tp->mss_cache = 536;
L
Linus Torvalds 已提交
1647 1648

	tp->reordering = sysctl_tcp_reordering;
1649
	tp->ca_ops = &tcp_init_congestion_ops;
L
Linus Torvalds 已提交
1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671

	sk->sk_state = TCP_CLOSE;

	sk->sk_write_space = sk_stream_write_space;
	sock_set_flag(sk, SOCK_USE_WRITE_QUEUE);

	tp->af_specific = &ipv4_specific;

	sk->sk_sndbuf = sysctl_tcp_wmem[1];
	sk->sk_rcvbuf = sysctl_tcp_rmem[1];

	atomic_inc(&tcp_sockets_allocated);

	return 0;
}

int tcp_v4_destroy_sock(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	tcp_clear_xmit_timers(sk);

1672 1673
	tcp_cleanup_congestion_control(tp);

L
Linus Torvalds 已提交
1674 1675 1676 1677 1678 1679 1680 1681 1682 1683
	/* Cleanup up the write buffer. */
  	sk_stream_writequeue_purge(sk);

	/* Cleans up our, hopefully empty, out_of_order_queue. */
  	__skb_queue_purge(&tp->out_of_order_queue);

	/* Clean prequeue, it must be empty really */
	__skb_queue_purge(&tp->ucopy.prequeue);

	/* Clean up a referenced TCP bind bucket. */
1684
	if (inet_csk(sk)->icsk_bind_hash)
1685
		inet_put_port(&tcp_hashinfo, sk);
L
Linus Torvalds 已提交
1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704

	/*
	 * If sendmsg cached page exists, toss it.
	 */
	if (sk->sk_sndmsg_page) {
		__free_page(sk->sk_sndmsg_page);
		sk->sk_sndmsg_page = NULL;
	}

	atomic_dec(&tcp_sockets_allocated);

	return 0;
}

EXPORT_SYMBOL(tcp_v4_destroy_sock);

#ifdef CONFIG_PROC_FS
/* Proc filesystem TCP sock list dumping. */

1705
static inline struct inet_timewait_sock *tw_head(struct hlist_head *head)
L
Linus Torvalds 已提交
1706 1707
{
	return hlist_empty(head) ? NULL :
1708
		list_entry(head->first, struct inet_timewait_sock, tw_node);
L
Linus Torvalds 已提交
1709 1710
}

1711
static inline struct inet_timewait_sock *tw_next(struct inet_timewait_sock *tw)
L
Linus Torvalds 已提交
1712 1713 1714 1715 1716 1717 1718
{
	return tw->tw_node.next ?
		hlist_entry(tw->tw_node.next, typeof(*tw), tw_node) : NULL;
}

static void *listening_get_next(struct seq_file *seq, void *cur)
{
1719
	struct inet_connection_sock *icsk;
L
Linus Torvalds 已提交
1720 1721 1722 1723 1724 1725
	struct hlist_node *node;
	struct sock *sk = cur;
	struct tcp_iter_state* st = seq->private;

	if (!sk) {
		st->bucket = 0;
1726
		sk = sk_head(&tcp_hashinfo.listening_hash[0]);
L
Linus Torvalds 已提交
1727 1728 1729 1730 1731 1732
		goto get_sk;
	}

	++st->num;

	if (st->state == TCP_SEQ_STATE_OPENREQ) {
1733
		struct request_sock *req = cur;
L
Linus Torvalds 已提交
1734

1735
	       	icsk = inet_csk(st->syn_wait_sk);
L
Linus Torvalds 已提交
1736 1737 1738
		req = req->dl_next;
		while (1) {
			while (req) {
1739
				if (req->rsk_ops->family == st->family) {
L
Linus Torvalds 已提交
1740 1741 1742 1743 1744 1745 1746 1747
					cur = req;
					goto out;
				}
				req = req->dl_next;
			}
			if (++st->sbucket >= TCP_SYNQ_HSIZE)
				break;
get_req:
1748
			req = icsk->icsk_accept_queue.listen_opt->syn_table[st->sbucket];
L
Linus Torvalds 已提交
1749 1750 1751
		}
		sk	  = sk_next(st->syn_wait_sk);
		st->state = TCP_SEQ_STATE_LISTENING;
1752
		read_unlock_bh(&icsk->icsk_accept_queue.syn_wait_lock);
L
Linus Torvalds 已提交
1753
	} else {
1754 1755 1756
	       	icsk = inet_csk(sk);
		read_lock_bh(&icsk->icsk_accept_queue.syn_wait_lock);
		if (reqsk_queue_len(&icsk->icsk_accept_queue))
L
Linus Torvalds 已提交
1757
			goto start_req;
1758
		read_unlock_bh(&icsk->icsk_accept_queue.syn_wait_lock);
L
Linus Torvalds 已提交
1759 1760 1761 1762 1763 1764 1765 1766
		sk = sk_next(sk);
	}
get_sk:
	sk_for_each_from(sk, node) {
		if (sk->sk_family == st->family) {
			cur = sk;
			goto out;
		}
1767 1768 1769
	       	icsk = inet_csk(sk);
		read_lock_bh(&icsk->icsk_accept_queue.syn_wait_lock);
		if (reqsk_queue_len(&icsk->icsk_accept_queue)) {
L
Linus Torvalds 已提交
1770 1771 1772 1773 1774 1775 1776
start_req:
			st->uid		= sock_i_uid(sk);
			st->syn_wait_sk = sk;
			st->state	= TCP_SEQ_STATE_OPENREQ;
			st->sbucket	= 0;
			goto get_req;
		}
1777
		read_unlock_bh(&icsk->icsk_accept_queue.syn_wait_lock);
L
Linus Torvalds 已提交
1778
	}
1779
	if (++st->bucket < INET_LHTABLE_SIZE) {
1780
		sk = sk_head(&tcp_hashinfo.listening_hash[st->bucket]);
L
Linus Torvalds 已提交
1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803
		goto get_sk;
	}
	cur = NULL;
out:
	return cur;
}

static void *listening_get_idx(struct seq_file *seq, loff_t *pos)
{
	void *rc = listening_get_next(seq, NULL);

	while (rc && *pos) {
		rc = listening_get_next(seq, rc);
		--*pos;
	}
	return rc;
}

static void *established_get_first(struct seq_file *seq)
{
	struct tcp_iter_state* st = seq->private;
	void *rc = NULL;

1804
	for (st->bucket = 0; st->bucket < tcp_hashinfo.ehash_size; ++st->bucket) {
L
Linus Torvalds 已提交
1805 1806
		struct sock *sk;
		struct hlist_node *node;
1807
		struct inet_timewait_sock *tw;
L
Linus Torvalds 已提交
1808 1809 1810 1811

		/* We can reschedule _before_ having picked the target: */
		cond_resched_softirq();

1812 1813
		read_lock(&tcp_hashinfo.ehash[st->bucket].lock);
		sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) {
L
Linus Torvalds 已提交
1814 1815 1816 1817 1818 1819 1820
			if (sk->sk_family != st->family) {
				continue;
			}
			rc = sk;
			goto out;
		}
		st->state = TCP_SEQ_STATE_TIME_WAIT;
1821 1822
		inet_twsk_for_each(tw, node,
				   &tcp_hashinfo.ehash[st->bucket + tcp_hashinfo.ehash_size].chain) {
L
Linus Torvalds 已提交
1823 1824 1825 1826 1827 1828
			if (tw->tw_family != st->family) {
				continue;
			}
			rc = tw;
			goto out;
		}
1829
		read_unlock(&tcp_hashinfo.ehash[st->bucket].lock);
L
Linus Torvalds 已提交
1830 1831 1832 1833 1834 1835 1836 1837 1838
		st->state = TCP_SEQ_STATE_ESTABLISHED;
	}
out:
	return rc;
}

static void *established_get_next(struct seq_file *seq, void *cur)
{
	struct sock *sk = cur;
1839
	struct inet_timewait_sock *tw;
L
Linus Torvalds 已提交
1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855
	struct hlist_node *node;
	struct tcp_iter_state* st = seq->private;

	++st->num;

	if (st->state == TCP_SEQ_STATE_TIME_WAIT) {
		tw = cur;
		tw = tw_next(tw);
get_tw:
		while (tw && tw->tw_family != st->family) {
			tw = tw_next(tw);
		}
		if (tw) {
			cur = tw;
			goto out;
		}
1856
		read_unlock(&tcp_hashinfo.ehash[st->bucket].lock);
L
Linus Torvalds 已提交
1857 1858 1859 1860 1861
		st->state = TCP_SEQ_STATE_ESTABLISHED;

		/* We can reschedule between buckets: */
		cond_resched_softirq();

1862 1863 1864
		if (++st->bucket < tcp_hashinfo.ehash_size) {
			read_lock(&tcp_hashinfo.ehash[st->bucket].lock);
			sk = sk_head(&tcp_hashinfo.ehash[st->bucket].chain);
L
Linus Torvalds 已提交
1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877
		} else {
			cur = NULL;
			goto out;
		}
	} else
		sk = sk_next(sk);

	sk_for_each_from(sk, node) {
		if (sk->sk_family == st->family)
			goto found;
	}

	st->state = TCP_SEQ_STATE_TIME_WAIT;
1878
	tw = tw_head(&tcp_hashinfo.ehash[st->bucket + tcp_hashinfo.ehash_size].chain);
L
Linus Torvalds 已提交
1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901
	goto get_tw;
found:
	cur = sk;
out:
	return cur;
}

static void *established_get_idx(struct seq_file *seq, loff_t pos)
{
	void *rc = established_get_first(seq);

	while (rc && pos) {
		rc = established_get_next(seq, rc);
		--pos;
	}		
	return rc;
}

static void *tcp_get_idx(struct seq_file *seq, loff_t pos)
{
	void *rc;
	struct tcp_iter_state* st = seq->private;

1902
	inet_listen_lock(&tcp_hashinfo);
L
Linus Torvalds 已提交
1903 1904 1905 1906
	st->state = TCP_SEQ_STATE_LISTENING;
	rc	  = listening_get_idx(seq, &pos);

	if (!rc) {
1907
		inet_listen_unlock(&tcp_hashinfo);
L
Linus Torvalds 已提交
1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939
		local_bh_disable();
		st->state = TCP_SEQ_STATE_ESTABLISHED;
		rc	  = established_get_idx(seq, pos);
	}

	return rc;
}

static void *tcp_seq_start(struct seq_file *seq, loff_t *pos)
{
	struct tcp_iter_state* st = seq->private;
	st->state = TCP_SEQ_STATE_LISTENING;
	st->num = 0;
	return *pos ? tcp_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
}

static void *tcp_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	void *rc = NULL;
	struct tcp_iter_state* st;

	if (v == SEQ_START_TOKEN) {
		rc = tcp_get_idx(seq, 0);
		goto out;
	}
	st = seq->private;

	switch (st->state) {
	case TCP_SEQ_STATE_OPENREQ:
	case TCP_SEQ_STATE_LISTENING:
		rc = listening_get_next(seq, v);
		if (!rc) {
1940
			inet_listen_unlock(&tcp_hashinfo);
L
Linus Torvalds 已提交
1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962
			local_bh_disable();
			st->state = TCP_SEQ_STATE_ESTABLISHED;
			rc	  = established_get_first(seq);
		}
		break;
	case TCP_SEQ_STATE_ESTABLISHED:
	case TCP_SEQ_STATE_TIME_WAIT:
		rc = established_get_next(seq, v);
		break;
	}
out:
	++*pos;
	return rc;
}

static void tcp_seq_stop(struct seq_file *seq, void *v)
{
	struct tcp_iter_state* st = seq->private;

	switch (st->state) {
	case TCP_SEQ_STATE_OPENREQ:
		if (v) {
1963 1964
			struct inet_connection_sock *icsk = inet_csk(st->syn_wait_sk);
			read_unlock_bh(&icsk->icsk_accept_queue.syn_wait_lock);
L
Linus Torvalds 已提交
1965 1966 1967
		}
	case TCP_SEQ_STATE_LISTENING:
		if (v != SEQ_START_TOKEN)
1968
			inet_listen_unlock(&tcp_hashinfo);
L
Linus Torvalds 已提交
1969 1970 1971 1972
		break;
	case TCP_SEQ_STATE_TIME_WAIT:
	case TCP_SEQ_STATE_ESTABLISHED:
		if (v)
1973
			read_unlock(&tcp_hashinfo.ehash[st->bucket].lock);
L
Linus Torvalds 已提交
1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039
		local_bh_enable();
		break;
	}
}

static int tcp_seq_open(struct inode *inode, struct file *file)
{
	struct tcp_seq_afinfo *afinfo = PDE(inode)->data;
	struct seq_file *seq;
	struct tcp_iter_state *s;
	int rc;

	if (unlikely(afinfo == NULL))
		return -EINVAL;

	s = kmalloc(sizeof(*s), GFP_KERNEL);
	if (!s)
		return -ENOMEM;
	memset(s, 0, sizeof(*s));
	s->family		= afinfo->family;
	s->seq_ops.start	= tcp_seq_start;
	s->seq_ops.next		= tcp_seq_next;
	s->seq_ops.show		= afinfo->seq_show;
	s->seq_ops.stop		= tcp_seq_stop;

	rc = seq_open(file, &s->seq_ops);
	if (rc)
		goto out_kfree;
	seq	     = file->private_data;
	seq->private = s;
out:
	return rc;
out_kfree:
	kfree(s);
	goto out;
}

int tcp_proc_register(struct tcp_seq_afinfo *afinfo)
{
	int rc = 0;
	struct proc_dir_entry *p;

	if (!afinfo)
		return -EINVAL;
	afinfo->seq_fops->owner		= afinfo->owner;
	afinfo->seq_fops->open		= tcp_seq_open;
	afinfo->seq_fops->read		= seq_read;
	afinfo->seq_fops->llseek	= seq_lseek;
	afinfo->seq_fops->release	= seq_release_private;
	
	p = proc_net_fops_create(afinfo->name, S_IRUGO, afinfo->seq_fops);
	if (p)
		p->data = afinfo;
	else
		rc = -ENOMEM;
	return rc;
}

void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo)
{
	if (!afinfo)
		return;
	proc_net_remove(afinfo->name);
	memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); 
}

2040
static void get_openreq4(struct sock *sk, struct request_sock *req,
L
Linus Torvalds 已提交
2041 2042
			 char *tmpbuf, int i, int uid)
{
2043
	const struct inet_request_sock *ireq = inet_rsk(req);
L
Linus Torvalds 已提交
2044 2045 2046 2047 2048
	int ttd = req->expires - jiffies;

	sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %u %d %p",
		i,
2049
		ireq->loc_addr,
L
Linus Torvalds 已提交
2050
		ntohs(inet_sk(sk)->sport),
2051 2052
		ireq->rmt_addr,
		ntohs(ireq->rmt_port),
L
Linus Torvalds 已提交
2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069
		TCP_SYN_RECV,
		0, 0, /* could print option size, but that is af dependent. */
		1,    /* timers active (only the expire timer) */
		jiffies_to_clock_t(ttd),
		req->retrans,
		uid,
		0,  /* non standard timer */
		0, /* open_requests have no inode */
		atomic_read(&sk->sk_refcnt),
		req);
}

static void get_tcp4_sock(struct sock *sp, char *tmpbuf, int i)
{
	int timer_active;
	unsigned long timer_expires;
	struct tcp_sock *tp = tcp_sk(sp);
2070
	const struct inet_connection_sock *icsk = inet_csk(sp);
L
Linus Torvalds 已提交
2071 2072 2073 2074 2075 2076
	struct inet_sock *inet = inet_sk(sp);
	unsigned int dest = inet->daddr;
	unsigned int src = inet->rcv_saddr;
	__u16 destp = ntohs(inet->dport);
	__u16 srcp = ntohs(inet->sport);

2077
	if (icsk->icsk_pending == ICSK_TIME_RETRANS) {
L
Linus Torvalds 已提交
2078
		timer_active	= 1;
2079 2080
		timer_expires	= icsk->icsk_timeout;
	} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
L
Linus Torvalds 已提交
2081
		timer_active	= 4;
2082
		timer_expires	= icsk->icsk_timeout;
L
Linus Torvalds 已提交
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096
	} else if (timer_pending(&sp->sk_timer)) {
		timer_active	= 2;
		timer_expires	= sp->sk_timer.expires;
	} else {
		timer_active	= 0;
		timer_expires = jiffies;
	}

	sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX "
			"%08X %5d %8d %lu %d %p %u %u %u %u %d",
		i, src, srcp, dest, destp, sp->sk_state,
		tp->write_seq - tp->snd_una, tp->rcv_nxt - tp->copied_seq,
		timer_active,
		jiffies_to_clock_t(timer_expires - jiffies),
2097
		icsk->icsk_retransmits,
L
Linus Torvalds 已提交
2098 2099 2100 2101
		sock_i_uid(sp),
		tp->probes_out,
		sock_i_ino(sp),
		atomic_read(&sp->sk_refcnt), sp,
2102 2103 2104
		icsk->icsk_rto,
		icsk->icsk_ack.ato,
		(icsk->icsk_ack.quick << 1) | icsk->icsk_ack.pingpong,
L
Linus Torvalds 已提交
2105 2106 2107 2108
		tp->snd_cwnd,
		tp->snd_ssthresh >= 0xFFFF ? -1 : tp->snd_ssthresh);
}

2109
static void get_timewait4_sock(struct inet_timewait_sock *tw, char *tmpbuf, int i)
L
Linus Torvalds 已提交
2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188
{
	unsigned int dest, src;
	__u16 destp, srcp;
	int ttd = tw->tw_ttd - jiffies;

	if (ttd < 0)
		ttd = 0;

	dest  = tw->tw_daddr;
	src   = tw->tw_rcv_saddr;
	destp = ntohs(tw->tw_dport);
	srcp  = ntohs(tw->tw_sport);

	sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %p",
		i, src, srcp, dest, destp, tw->tw_substate, 0, 0,
		3, jiffies_to_clock_t(ttd), 0, 0, 0, 0,
		atomic_read(&tw->tw_refcnt), tw);
}

#define TMPSZ 150

static int tcp4_seq_show(struct seq_file *seq, void *v)
{
	struct tcp_iter_state* st;
	char tmpbuf[TMPSZ + 1];

	if (v == SEQ_START_TOKEN) {
		seq_printf(seq, "%-*s\n", TMPSZ - 1,
			   "  sl  local_address rem_address   st tx_queue "
			   "rx_queue tr tm->when retrnsmt   uid  timeout "
			   "inode");
		goto out;
	}
	st = seq->private;

	switch (st->state) {
	case TCP_SEQ_STATE_LISTENING:
	case TCP_SEQ_STATE_ESTABLISHED:
		get_tcp4_sock(v, tmpbuf, st->num);
		break;
	case TCP_SEQ_STATE_OPENREQ:
		get_openreq4(st->syn_wait_sk, v, tmpbuf, st->num, st->uid);
		break;
	case TCP_SEQ_STATE_TIME_WAIT:
		get_timewait4_sock(v, tmpbuf, st->num);
		break;
	}
	seq_printf(seq, "%-*s\n", TMPSZ - 1, tmpbuf);
out:
	return 0;
}

static struct file_operations tcp4_seq_fops;
static struct tcp_seq_afinfo tcp4_seq_afinfo = {
	.owner		= THIS_MODULE,
	.name		= "tcp",
	.family		= AF_INET,
	.seq_show	= tcp4_seq_show,
	.seq_fops	= &tcp4_seq_fops,
};

int __init tcp4_proc_init(void)
{
	return tcp_proc_register(&tcp4_seq_afinfo);
}

void tcp4_proc_exit(void)
{
	tcp_proc_unregister(&tcp4_seq_afinfo);
}
#endif /* CONFIG_PROC_FS */

struct proto tcp_prot = {
	.name			= "TCP",
	.owner			= THIS_MODULE,
	.close			= tcp_close,
	.connect		= tcp_v4_connect,
	.disconnect		= tcp_disconnect,
2189
	.accept			= inet_csk_accept,
L
Linus Torvalds 已提交
2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210
	.ioctl			= tcp_ioctl,
	.init			= tcp_v4_init_sock,
	.destroy		= tcp_v4_destroy_sock,
	.shutdown		= tcp_shutdown,
	.setsockopt		= tcp_setsockopt,
	.getsockopt		= tcp_getsockopt,
	.sendmsg		= tcp_sendmsg,
	.recvmsg		= tcp_recvmsg,
	.backlog_rcv		= tcp_v4_do_rcv,
	.hash			= tcp_v4_hash,
	.unhash			= tcp_unhash,
	.get_port		= tcp_v4_get_port,
	.enter_memory_pressure	= tcp_enter_memory_pressure,
	.sockets_allocated	= &tcp_sockets_allocated,
	.memory_allocated	= &tcp_memory_allocated,
	.memory_pressure	= &tcp_memory_pressure,
	.sysctl_mem		= sysctl_tcp_mem,
	.sysctl_wmem		= sysctl_tcp_wmem,
	.sysctl_rmem		= sysctl_tcp_rmem,
	.max_header		= MAX_TCP_HEADER,
	.obj_size		= sizeof(struct tcp_sock),
2211
	.twsk_obj_size		= sizeof(struct tcp_timewait_sock),
2212
	.rsk_prot		= &tcp_request_sock_ops,
L
Linus Torvalds 已提交
2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232
};



void __init tcp_v4_init(struct net_proto_family *ops)
{
	int err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_TCP, &tcp_socket);
	if (err < 0)
		panic("Failed to create the TCP control socket.\n");
	tcp_socket->sk->sk_allocation   = GFP_ATOMIC;
	inet_sk(tcp_socket->sk)->uc_ttl = -1;

	/* Unhash it so that IP input processing does not even
	 * see it, we do not wish this socket to see incoming
	 * packets.
	 */
	tcp_socket->sk->sk_prot->unhash(tcp_socket->sk);
}

EXPORT_SYMBOL(ipv4_specific);
2233
EXPORT_SYMBOL(inet_bind_bucket_create);
L
Linus Torvalds 已提交
2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251
EXPORT_SYMBOL(tcp_hashinfo);
EXPORT_SYMBOL(tcp_prot);
EXPORT_SYMBOL(tcp_unhash);
EXPORT_SYMBOL(tcp_v4_conn_request);
EXPORT_SYMBOL(tcp_v4_connect);
EXPORT_SYMBOL(tcp_v4_do_rcv);
EXPORT_SYMBOL(tcp_v4_remember_stamp);
EXPORT_SYMBOL(tcp_v4_send_check);
EXPORT_SYMBOL(tcp_v4_syn_recv_sock);

#ifdef CONFIG_PROC_FS
EXPORT_SYMBOL(tcp_proc_register);
EXPORT_SYMBOL(tcp_proc_unregister);
#endif
EXPORT_SYMBOL(sysctl_local_port_range);
EXPORT_SYMBOL(sysctl_tcp_low_latency);
EXPORT_SYMBOL(sysctl_tcp_tw_reuse);