tcp_minisocks.c 21.0 KB
Newer Older
L
Linus Torvalds 已提交
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.
 *
 *		Implementation of the Transmission Control Protocol(TCP).
 *
 * Version:	$Id: tcp_minisocks.c,v 1.15 2002/02/01 22:01:04 davem Exp $
 *
10
 * Authors:	Ross Biro
L
Linus Torvalds 已提交
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
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Mark Evans, <evansmp@uhura.aston.ac.uk>
 *		Corey Minyard <wf-rch!minyard@relay.EU.net>
 *		Florian La Roche, <flla@stud.uni-sb.de>
 *		Charles Hedrick, <hedrick@klinzhai.rutgers.edu>
 *		Linus Torvalds, <torvalds@cs.helsinki.fi>
 *		Alan Cox, <gw4pts@gw4pts.ampr.org>
 *		Matthew Dillon, <dillon@apollo.west.oic.com>
 *		Arnt Gulbrandsen, <agulbra@nvg.unit.no>
 *		Jorge Cwik, <jorge@laser.satlink.net>
 */

#include <linux/config.h>
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/sysctl.h>
#include <linux/workqueue.h>
#include <net/tcp.h>
#include <net/inet_common.h>
#include <net/xfrm.h>

#ifdef CONFIG_SYSCTL
#define SYNC_INIT 0 /* let the user enable it */
#else
#define SYNC_INIT 1
#endif

int sysctl_tcp_syncookies = SYNC_INIT; 
int sysctl_tcp_abort_on_overflow;

41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
struct inet_timewait_death_row tcp_death_row = {
	.sysctl_max_tw_buckets = NR_FILE * 2,
	.period		= TCP_TIMEWAIT_LEN / INET_TWDR_TWKILL_SLOTS,
	.death_lock	= SPIN_LOCK_UNLOCKED,
	.hashinfo	= &tcp_hashinfo,
	.tw_timer	= TIMER_INITIALIZER(inet_twdr_hangman, 0,
					    (unsigned long)&tcp_death_row),
	.twkill_work	= __WORK_INITIALIZER(tcp_death_row.twkill_work,
					     inet_twdr_twkill_work,
					     &tcp_death_row),
/* Short-time timewait calendar */

	.twcal_hand	= -1,
	.twcal_timer	= TIMER_INITIALIZER(inet_twdr_twcal_tick, 0,
					    (unsigned long)&tcp_death_row),
};

EXPORT_SYMBOL_GPL(tcp_death_row);

L
Linus Torvalds 已提交
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97
static __inline__ int tcp_in_window(u32 seq, u32 end_seq, u32 s_win, u32 e_win)
{
	if (seq == s_win)
		return 1;
	if (after(end_seq, s_win) && before(seq, e_win))
		return 1;
	return (seq == e_win && seq == end_seq);
}

/* 
 * * Main purpose of TIME-WAIT state is to close connection gracefully,
 *   when one of ends sits in LAST-ACK or CLOSING retransmitting FIN
 *   (and, probably, tail of data) and one or more our ACKs are lost.
 * * What is TIME-WAIT timeout? It is associated with maximal packet
 *   lifetime in the internet, which results in wrong conclusion, that
 *   it is set to catch "old duplicate segments" wandering out of their path.
 *   It is not quite correct. This timeout is calculated so that it exceeds
 *   maximal retransmission timeout enough to allow to lose one (or more)
 *   segments sent by peer and our ACKs. This time may be calculated from RTO.
 * * When TIME-WAIT socket receives RST, it means that another end
 *   finally closed and we are allowed to kill TIME-WAIT too.
 * * Second purpose of TIME-WAIT is catching old duplicate segments.
 *   Well, certainly it is pure paranoia, but if we load TIME-WAIT
 *   with this semantics, we MUST NOT kill TIME-WAIT state with RSTs.
 * * If we invented some more clever way to catch duplicates
 *   (f.e. based on PAWS), we could truncate TIME-WAIT to several RTOs.
 *
 * The algorithm below is based on FORMAL INTERPRETATION of RFCs.
 * When you compare it to RFCs, please, read section SEGMENT ARRIVES
 * from the very beginning.
 *
 * NOTE. With recycling (and later with fin-wait-2) TW bucket
 * is _not_ stateless. It means, that strictly speaking we must
 * spinlock it. I do not want! Well, probability of misbehaviour
 * is ridiculously low and, seems, we could use some mb() tricks
 * to avoid misread sequence numbers, states etc.  --ANK
 */
enum tcp_tw_status
98 99
tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
			   const struct tcphdr *th)
L
Linus Torvalds 已提交
100
{
101
	struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw);
L
Linus Torvalds 已提交
102 103 104 105
	struct tcp_options_received tmp_opt;
	int paws_reject = 0;

	tmp_opt.saw_tstamp = 0;
106
	if (th->doff > (sizeof(*th) >> 2) && tcptw->tw_ts_recent_stamp) {
L
Linus Torvalds 已提交
107 108 109
		tcp_parse_options(skb, &tmp_opt, 0);

		if (tmp_opt.saw_tstamp) {
110 111
			tmp_opt.ts_recent	= tcptw->tw_ts_recent;
			tmp_opt.ts_recent_stamp	= tcptw->tw_ts_recent_stamp;
L
Linus Torvalds 已提交
112 113 114 115 116 117 118 119 120 121
			paws_reject = tcp_paws_check(&tmp_opt, th->rst);
		}
	}

	if (tw->tw_substate == TCP_FIN_WAIT2) {
		/* Just repeat all the checks of tcp_rcv_state_process() */

		/* Out of window, send ACK */
		if (paws_reject ||
		    !tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq,
122 123
				   tcptw->tw_rcv_nxt,
				   tcptw->tw_rcv_nxt + tcptw->tw_rcv_wnd))
L
Linus Torvalds 已提交
124 125 126 127 128
			return TCP_TW_ACK;

		if (th->rst)
			goto kill;

129
		if (th->syn && !before(TCP_SKB_CB(skb)->seq, tcptw->tw_rcv_nxt))
L
Linus Torvalds 已提交
130 131 132
			goto kill_with_rst;

		/* Dup ACK? */
133
		if (!after(TCP_SKB_CB(skb)->end_seq, tcptw->tw_rcv_nxt) ||
L
Linus Torvalds 已提交
134
		    TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq) {
135
			inet_twsk_put(tw);
L
Linus Torvalds 已提交
136 137 138 139 140 141 142
			return TCP_TW_SUCCESS;
		}

		/* New data or FIN. If new data arrive after half-duplex close,
		 * reset.
		 */
		if (!th->fin ||
143
		    TCP_SKB_CB(skb)->end_seq != tcptw->tw_rcv_nxt + 1) {
L
Linus Torvalds 已提交
144
kill_with_rst:
145
			inet_twsk_deschedule(tw, &tcp_death_row);
146
			inet_twsk_put(tw);
L
Linus Torvalds 已提交
147 148 149 150
			return TCP_TW_RST;
		}

		/* FIN arrived, enter true time-wait state. */
151 152
		tw->tw_substate	  = TCP_TIME_WAIT;
		tcptw->tw_rcv_nxt = TCP_SKB_CB(skb)->end_seq;
L
Linus Torvalds 已提交
153
		if (tmp_opt.saw_tstamp) {
154 155
			tcptw->tw_ts_recent_stamp = xtime.tv_sec;
			tcptw->tw_ts_recent	  = tmp_opt.rcv_tsval;
L
Linus Torvalds 已提交
156 157 158 159 160
		}

		/* I am shamed, but failed to make it more elegant.
		 * Yes, it is direct reference to IP, which is impossible
		 * to generalize to IPv6. Taking into account that IPv6
S
Stephen Hemminger 已提交
161
		 * do not understand recycling in any case, it not
L
Linus Torvalds 已提交
162 163
		 * a big problem in practice. --ANK */
		if (tw->tw_family == AF_INET &&
164
		    tcp_death_row.sysctl_tw_recycle && tcptw->tw_ts_recent_stamp &&
L
Linus Torvalds 已提交
165
		    tcp_v4_tw_remember_stamp(tw))
166 167
			inet_twsk_schedule(tw, &tcp_death_row, tw->tw_timeout,
					   TCP_TIMEWAIT_LEN);
L
Linus Torvalds 已提交
168
		else
169 170
			inet_twsk_schedule(tw, &tcp_death_row, TCP_TIMEWAIT_LEN,
					   TCP_TIMEWAIT_LEN);
L
Linus Torvalds 已提交
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191
		return TCP_TW_ACK;
	}

	/*
	 *	Now real TIME-WAIT state.
	 *
	 *	RFC 1122:
	 *	"When a connection is [...] on TIME-WAIT state [...]
	 *	[a TCP] MAY accept a new SYN from the remote TCP to
	 *	reopen the connection directly, if it:
	 *	
	 *	(1)  assigns its initial sequence number for the new
	 *	connection to be larger than the largest sequence
	 *	number it used on the previous connection incarnation,
	 *	and
	 *
	 *	(2)  returns to TIME-WAIT state if the SYN turns out 
	 *	to be an old duplicate".
	 */

	if (!paws_reject &&
192
	    (TCP_SKB_CB(skb)->seq == tcptw->tw_rcv_nxt &&
L
Linus Torvalds 已提交
193 194 195 196
	     (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq || th->rst))) {
		/* In window segment, it may be only reset or bare ack. */

		if (th->rst) {
S
Stephen Hemminger 已提交
197
			/* This is TIME_WAIT assassination, in two flavors.
L
Linus Torvalds 已提交
198 199 200 201 202
			 * Oh well... nobody has a sufficient solution to this
			 * protocol bug yet.
			 */
			if (sysctl_tcp_rfc1337 == 0) {
kill:
203
				inet_twsk_deschedule(tw, &tcp_death_row);
204
				inet_twsk_put(tw);
L
Linus Torvalds 已提交
205 206 207
				return TCP_TW_SUCCESS;
			}
		}
208 209
		inet_twsk_schedule(tw, &tcp_death_row, TCP_TIMEWAIT_LEN,
				   TCP_TIMEWAIT_LEN);
L
Linus Torvalds 已提交
210 211

		if (tmp_opt.saw_tstamp) {
212 213
			tcptw->tw_ts_recent	  = tmp_opt.rcv_tsval;
			tcptw->tw_ts_recent_stamp = xtime.tv_sec;
L
Linus Torvalds 已提交
214 215
		}

216
		inet_twsk_put(tw);
L
Linus Torvalds 已提交
217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237
		return TCP_TW_SUCCESS;
	}

	/* Out of window segment.

	   All the segments are ACKed immediately.

	   The only exception is new SYN. We accept it, if it is
	   not old duplicate and we are not in danger to be killed
	   by delayed old duplicates. RFC check is that it has
	   newer sequence number works at rates <40Mbit/sec.
	   However, if paws works, it is reliable AND even more,
	   we even may relax silly seq space cutoff.

	   RED-PEN: we violate main RFC requirement, if this SYN will appear
	   old duplicate (i.e. we receive RST in reply to SYN-ACK),
	   we must return socket to time-wait state. It is not good,
	   but not fatal yet.
	 */

	if (th->syn && !th->rst && !th->ack && !paws_reject &&
238 239 240 241
	    (after(TCP_SKB_CB(skb)->seq, tcptw->tw_rcv_nxt) ||
	     (tmp_opt.saw_tstamp &&
	      (s32)(tcptw->tw_ts_recent - tmp_opt.rcv_tsval) < 0))) {
		u32 isn = tcptw->tw_snd_nxt + 65535 + 2;
L
Linus Torvalds 已提交
242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
		if (isn == 0)
			isn++;
		TCP_SKB_CB(skb)->when = isn;
		return TCP_TW_SYN;
	}

	if (paws_reject)
		NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);

	if(!th->rst) {
		/* In this case we must reset the TIMEWAIT timer.
		 *
		 * If it is ACKless SYN it may be both old duplicate
		 * and new good SYN with random sequence number <rcv_nxt.
		 * Do not reschedule in the last case.
		 */
		if (paws_reject || th->ack)
259 260
			inet_twsk_schedule(tw, &tcp_death_row, TCP_TIMEWAIT_LEN,
					   TCP_TIMEWAIT_LEN);
L
Linus Torvalds 已提交
261 262 263 264 265 266

		/* Send ACK. Note, we do not put the bucket,
		 * it will be released by caller.
		 */
		return TCP_TW_ACK;
	}
267
	inet_twsk_put(tw);
L
Linus Torvalds 已提交
268 269 270 271 272 273 274 275
	return TCP_TW_SUCCESS;
}

/* 
 * Move a socket to time-wait or dead fin-wait-2 state.
 */ 
void tcp_time_wait(struct sock *sk, int state, int timeo)
{
276
	struct inet_timewait_sock *tw = NULL;
277
	const struct inet_connection_sock *icsk = inet_csk(sk);
278
	const struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
279 280
	int recycle_ok = 0;

281
	if (tcp_death_row.sysctl_tw_recycle && tp->rx_opt.ts_recent_stamp)
282
		recycle_ok = icsk->icsk_af_ops->remember_stamp(sk);
L
Linus Torvalds 已提交
283

284
	if (tcp_death_row.tw_count < tcp_death_row.sysctl_max_tw_buckets)
285
		tw = inet_twsk_alloc(sk, state);
L
Linus Torvalds 已提交
286

287 288
	if (tw != NULL) {
		struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw);
289
		const int rto = (icsk->icsk_rto << 2) - (icsk->icsk_rto >> 1);
290

L
Linus Torvalds 已提交
291
		tw->tw_rcv_wscale	= tp->rx_opt.rcv_wscale;
292 293 294 295 296
		tcptw->tw_rcv_nxt	= tp->rcv_nxt;
		tcptw->tw_snd_nxt	= tp->snd_nxt;
		tcptw->tw_rcv_wnd	= tcp_receive_window(tp);
		tcptw->tw_ts_recent	= tp->rx_opt.ts_recent;
		tcptw->tw_ts_recent_stamp = tp->rx_opt.ts_recent_stamp;
L
Linus Torvalds 已提交
297 298 299 300

#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
		if (tw->tw_family == PF_INET6) {
			struct ipv6_pinfo *np = inet6_sk(sk);
301
			struct inet6_timewait_sock *tw6;
L
Linus Torvalds 已提交
302

303 304 305 306
			tw->tw_ipv6_offset = inet6_tw_offset(sk->sk_prot);
			tw6 = inet6_twsk((struct sock *)tw);
			ipv6_addr_copy(&tw6->tw_v6_daddr, &np->daddr);
			ipv6_addr_copy(&tw6->tw_v6_rcv_saddr, &np->rcv_saddr);
307
			tw->tw_ipv6only = np->ipv6only;
308
		}
L
Linus Torvalds 已提交
309 310
#endif
		/* Linkage updates. */
311
		__inet_twsk_hashdance(tw, sk, &tcp_hashinfo);
L
Linus Torvalds 已提交
312 313 314 315 316 317 318 319 320 321 322 323 324

		/* Get the TIME_WAIT timeout firing. */
		if (timeo < rto)
			timeo = rto;

		if (recycle_ok) {
			tw->tw_timeout = rto;
		} else {
			tw->tw_timeout = TCP_TIMEWAIT_LEN;
			if (state == TCP_TIME_WAIT)
				timeo = TCP_TIMEWAIT_LEN;
		}

325 326
		inet_twsk_schedule(tw, &tcp_death_row, timeo,
				   TCP_TIMEWAIT_LEN);
327
		inet_twsk_put(tw);
L
Linus Torvalds 已提交
328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346
	} else {
		/* Sorry, if we're out of memory, just CLOSE this
		 * socket up.  We've got bigger problems than
		 * non-graceful socket closings.
		 */
		if (net_ratelimit())
			printk(KERN_INFO "TCP: time wait bucket table overflow\n");
	}

	tcp_update_metrics(sk);
	tcp_done(sk);
}

/* This is not only more efficient than what we used to do, it eliminates
 * a lot of code duplication between IPv4/IPv6 SYN recv processing. -DaveM
 *
 * Actually, we could lots of memory writes here. tp of listening
 * socket contains all necessary default parameters.
 */
347
struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, struct sk_buff *skb)
L
Linus Torvalds 已提交
348
{
349
	struct sock *newsk = inet_csk_clone(sk, req, GFP_ATOMIC);
L
Linus Torvalds 已提交
350

351
	if (newsk != NULL) {
352
		const struct inet_request_sock *ireq = inet_rsk(req);
353
		struct tcp_request_sock *treq = tcp_rsk(req);
354
		struct inet_connection_sock *newicsk = inet_csk(sk);
L
Linus Torvalds 已提交
355 356 357 358 359
		struct tcp_sock *newtp;

		/* Now setup tcp_sock */
		newtp = tcp_sk(newsk);
		newtp->pred_flags = 0;
360
		newtp->rcv_nxt = treq->rcv_isn + 1;
361
		newtp->snd_nxt = newtp->snd_una = newtp->snd_sml = treq->snt_isn + 1;
L
Linus Torvalds 已提交
362 363 364

		tcp_prequeue_init(newtp);

365
		tcp_init_wl(newtp, treq->snt_isn, treq->rcv_isn);
L
Linus Torvalds 已提交
366 367 368

		newtp->srtt = 0;
		newtp->mdev = TCP_TIMEOUT_INIT;
369
		newicsk->icsk_rto = TCP_TIMEOUT_INIT;
L
Linus Torvalds 已提交
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384

		newtp->packets_out = 0;
		newtp->left_out = 0;
		newtp->retrans_out = 0;
		newtp->sacked_out = 0;
		newtp->fackets_out = 0;
		newtp->snd_ssthresh = 0x7fffffff;

		/* 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
		 */
		newtp->snd_cwnd = 2;
		newtp->snd_cwnd_cnt = 0;
385
		newtp->bytes_acked = 0;
L
Linus Torvalds 已提交
386 387 388 389

		newtp->frto_counter = 0;
		newtp->frto_highmark = 0;

390
		newicsk->icsk_ca_ops = &tcp_init_congestion_ops;
391

392
		tcp_set_ca_state(newsk, TCP_CA_Open);
L
Linus Torvalds 已提交
393 394
		tcp_init_xmit_timers(newsk);
		skb_queue_head_init(&newtp->out_of_order_queue);
395 396
		newtp->rcv_wup = treq->rcv_isn + 1;
		newtp->write_seq = treq->snt_isn + 1;
L
Linus Torvalds 已提交
397
		newtp->pushed_seq = newtp->write_seq;
398
		newtp->copied_seq = treq->rcv_isn + 1;
L
Linus Torvalds 已提交
399 400 401 402 403 404 405 406 407 408

		newtp->rx_opt.saw_tstamp = 0;

		newtp->rx_opt.dsack = 0;
		newtp->rx_opt.eff_sacks = 0;

		newtp->rx_opt.num_sacks = 0;
		newtp->urg_data = 0;

		if (sock_flag(newsk, SOCK_KEEPOPEN))
409 410
			inet_csk_reset_keepalive_timer(newsk,
						       keepalive_time_when(newtp));
L
Linus Torvalds 已提交
411

412 413
		newtp->rx_opt.tstamp_ok = ireq->tstamp_ok;
		if((newtp->rx_opt.sack_ok = ireq->sack_ok) != 0) {
L
Linus Torvalds 已提交
414 415 416 417 418 419
			if (sysctl_tcp_fack)
				newtp->rx_opt.sack_ok |= 2;
		}
		newtp->window_clamp = req->window_clamp;
		newtp->rcv_ssthresh = req->rcv_wnd;
		newtp->rcv_wnd = req->rcv_wnd;
420
		newtp->rx_opt.wscale_ok = ireq->wscale_ok;
L
Linus Torvalds 已提交
421
		if (newtp->rx_opt.wscale_ok) {
422 423
			newtp->rx_opt.snd_wscale = ireq->snd_wscale;
			newtp->rx_opt.rcv_wscale = ireq->rcv_wscale;
L
Linus Torvalds 已提交
424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439
		} else {
			newtp->rx_opt.snd_wscale = newtp->rx_opt.rcv_wscale = 0;
			newtp->window_clamp = min(newtp->window_clamp, 65535U);
		}
		newtp->snd_wnd = ntohs(skb->h.th->window) << newtp->rx_opt.snd_wscale;
		newtp->max_window = newtp->snd_wnd;

		if (newtp->rx_opt.tstamp_ok) {
			newtp->rx_opt.ts_recent = req->ts_recent;
			newtp->rx_opt.ts_recent_stamp = xtime.tv_sec;
			newtp->tcp_header_len = sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED;
		} else {
			newtp->rx_opt.ts_recent_stamp = 0;
			newtp->tcp_header_len = sizeof(struct tcphdr);
		}
		if (skb->len >= TCP_MIN_RCVMSS+newtp->tcp_header_len)
440
			newicsk->icsk_ack.last_seg_size = skb->len - newtp->tcp_header_len;
L
Linus Torvalds 已提交
441 442 443 444 445 446 447 448 449 450
		newtp->rx_opt.mss_clamp = req->mss;
		TCP_ECN_openreq_child(newtp, req);

		TCP_INC_STATS_BH(TCP_MIB_PASSIVEOPENS);
	}
	return newsk;
}

/* 
 *	Process an incoming packet for SYN_RECV sockets represented
451
 *	as a request_sock.
L
Linus Torvalds 已提交
452 453 454
 */

struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
455 456
			   struct request_sock *req,
			   struct request_sock **prev)
L
Linus Torvalds 已提交
457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479
{
	struct tcphdr *th = skb->h.th;
	u32 flg = tcp_flag_word(th) & (TCP_FLAG_RST|TCP_FLAG_SYN|TCP_FLAG_ACK);
	int paws_reject = 0;
	struct tcp_options_received tmp_opt;
	struct sock *child;

	tmp_opt.saw_tstamp = 0;
	if (th->doff > (sizeof(struct tcphdr)>>2)) {
		tcp_parse_options(skb, &tmp_opt, 0);

		if (tmp_opt.saw_tstamp) {
			tmp_opt.ts_recent = req->ts_recent;
			/* We do not store true stamp, but it is not required,
			 * it can be estimated (approximately)
			 * from another data.
			 */
			tmp_opt.ts_recent_stamp = xtime.tv_sec - ((TCP_TIMEOUT_INIT/HZ)<<req->retrans);
			paws_reject = tcp_paws_check(&tmp_opt, th->rst);
		}
	}

	/* Check for pure retransmitted SYN. */
480
	if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn &&
L
Linus Torvalds 已提交
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499
	    flg == TCP_FLAG_SYN &&
	    !paws_reject) {
		/*
		 * RFC793 draws (Incorrectly! It was fixed in RFC1122)
		 * this case on figure 6 and figure 8, but formal
		 * protocol description says NOTHING.
		 * To be more exact, it says that we should send ACK,
		 * because this segment (at least, if it has no data)
		 * is out of window.
		 *
		 *  CONCLUSION: RFC793 (even with RFC1122) DOES NOT
		 *  describe SYN-RECV state. All the description
		 *  is wrong, we cannot believe to it and should
		 *  rely only on common sense and implementation
		 *  experience.
		 *
		 * Enforce "SYN-ACK" according to figure 8, figure 6
		 * of RFC793, fixed by RFC1122.
		 */
500
		req->rsk_ops->rtx_syn_ack(sk, req, NULL);
L
Linus Torvalds 已提交
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
		return NULL;
	}

	/* Further reproduces section "SEGMENT ARRIVES"
	   for state SYN-RECEIVED of RFC793.
	   It is broken, however, it does not work only
	   when SYNs are crossed.

	   You would think that SYN crossing is impossible here, since
	   we should have a SYN_SENT socket (from connect()) on our end,
	   but this is not true if the crossed SYNs were sent to both
	   ends by a malicious third party.  We must defend against this,
	   and to do that we first verify the ACK (as per RFC793, page
	   36) and reset if it is invalid.  Is this a true full defense?
	   To convince ourselves, let us consider a way in which the ACK
	   test can still pass in this 'malicious crossed SYNs' case.
	   Malicious sender sends identical SYNs (and thus identical sequence
	   numbers) to both A and B:

		A: gets SYN, seq=7
		B: gets SYN, seq=7

	   By our good fortune, both A and B select the same initial
	   send sequence number of seven :-)

		A: sends SYN|ACK, seq=7, ack_seq=8
		B: sends SYN|ACK, seq=7, ack_seq=8

	   So we are now A eating this SYN|ACK, ACK test passes.  So
	   does sequence test, SYN is truncated, and thus we consider
	   it a bare ACK.

533 534 535 536
	   If icsk->icsk_accept_queue.rskq_defer_accept, we silently drop this
	   bare ACK.  Otherwise, we create an established connection.  Both
	   ends (listening sockets) accept the new incoming connection and try
	   to talk to each other. 8-)
L
Linus Torvalds 已提交
537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552

	   Note: This case is both harmless, and rare.  Possibility is about the
	   same as us discovering intelligent life on another plant tomorrow.

	   But generally, we should (RFC lies!) to accept ACK
	   from SYNACK both here and in tcp_rcv_state_process().
	   tcp_rcv_state_process() does not, hence, we do not too.

	   Note that the case is absolutely generic:
	   we cannot optimize anything here without
	   violating protocol. All the checks must be made
	   before attempt to create socket.
	 */

	/* RFC793 page 36: "If the connection is in any non-synchronized state ...
	 *                  and the incoming segment acknowledges something not yet
S
Stephen Hemminger 已提交
553
	 *                  sent (the segment carries an unacceptable ACK) ...
L
Linus Torvalds 已提交
554 555 556 557 558
	 *                  a reset is sent."
	 *
	 * Invalid ACK: reset will be sent by listening socket
	 */
	if ((flg & TCP_FLAG_ACK) &&
559
	    (TCP_SKB_CB(skb)->ack_seq != tcp_rsk(req)->snt_isn + 1))
L
Linus Torvalds 已提交
560 561 562 563 564 565 566 567 568 569
		return sk;

	/* Also, it would be not so bad idea to check rcv_tsecr, which
	 * is essentially ACK extension and too early or too late values
	 * should cause reset in unsynchronized states.
	 */

	/* RFC793: "first check sequence number". */

	if (paws_reject || !tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq,
570
					  tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) {
L
Linus Torvalds 已提交
571 572
		/* Out of window: send ACK and drop. */
		if (!(flg & TCP_FLAG_RST))
573
			req->rsk_ops->send_ack(skb, req);
L
Linus Torvalds 已提交
574 575 576 577 578 579 580
		if (paws_reject)
			NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);
		return NULL;
	}

	/* In sequence, PAWS is OK. */

581
	if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, tcp_rsk(req)->rcv_isn + 1))
L
Linus Torvalds 已提交
582 583
			req->ts_recent = tmp_opt.rcv_tsval;

584
		if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn) {
L
Linus Torvalds 已提交
585
			/* Truncate SYN, it is out of window starting
586
			   at tcp_rsk(req)->rcv_isn + 1. */
L
Linus Torvalds 已提交
587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602
			flg &= ~TCP_FLAG_SYN;
		}

		/* RFC793: "second check the RST bit" and
		 *	   "fourth, check the SYN bit"
		 */
		if (flg & (TCP_FLAG_RST|TCP_FLAG_SYN))
			goto embryonic_reset;

		/* ACK sequence verified above, just make sure ACK is
		 * set.  If ACK not set, just silently drop the packet.
		 */
		if (!(flg & TCP_FLAG_ACK))
			return NULL;

		/* If TCP_DEFER_ACCEPT is set, drop bare ACK. */
603 604
		if (inet_csk(sk)->icsk_accept_queue.rskq_defer_accept &&
		    TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) {
605
			inet_rsk(req)->acked = 1;
L
Linus Torvalds 已提交
606 607 608 609 610 611 612 613 614
			return NULL;
		}

		/* OK, ACK is valid, create big socket and
		 * feed this segment to it. It will repeat all
		 * the tests. THIS SEGMENT MUST MOVE SOCKET TO
		 * ESTABLISHED STATE. If it will be dropped after
		 * socket is created, wait for troubles.
		 */
615 616
		child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb,
								 req, NULL);
L
Linus Torvalds 已提交
617 618 619
		if (child == NULL)
			goto listen_overflow;

620 621
		inet_csk_reqsk_queue_unlink(sk, req, prev);
		inet_csk_reqsk_queue_removed(sk, req);
L
Linus Torvalds 已提交
622

623
		inet_csk_reqsk_queue_add(sk, req, child);
L
Linus Torvalds 已提交
624 625 626 627
		return child;

	listen_overflow:
		if (!sysctl_tcp_abort_on_overflow) {
628
			inet_rsk(req)->acked = 1;
L
Linus Torvalds 已提交
629 630 631 632 633 634
			return NULL;
		}

	embryonic_reset:
		NET_INC_STATS_BH(LINUX_MIB_EMBRYONICRSTS);
		if (!(flg & TCP_FLAG_RST))
635
			req->rsk_ops->send_reset(skb);
L
Linus Torvalds 已提交
636

637
		inet_csk_reqsk_queue_drop(sk, req, prev);
L
Linus Torvalds 已提交
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
		return NULL;
}

/*
 * Queue segment on the new socket if the new socket is active,
 * otherwise we just shortcircuit this and continue with
 * the new socket.
 */

int tcp_child_process(struct sock *parent, struct sock *child,
		      struct sk_buff *skb)
{
	int ret = 0;
	int state = child->sk_state;

	if (!sock_owned_by_user(child)) {
		ret = tcp_rcv_state_process(child, skb, skb->h.th, skb->len);

		/* Wakeup parent, send SIGIO */
		if (state == TCP_SYN_RECV && child->sk_state != state)
			parent->sk_data_ready(parent, 0);
	} else {
		/* Alas, it is possible again, because we do lookup
		 * in main socket hash table and lock on listening
		 * socket does not protect us more.
		 */
		sk_add_backlog(child, skb);
	}

	bh_unlock_sock(child);
	sock_put(child);
	return ret;
}

EXPORT_SYMBOL(tcp_check_req);
EXPORT_SYMBOL(tcp_child_process);
EXPORT_SYMBOL(tcp_create_openreq_child);
EXPORT_SYMBOL(tcp_timewait_state_process);