tcp_input.c 135.1 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_input.c,v 1.243 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 41 42 43 44
 *		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>
 */

/*
 * Changes:
 *		Pedro Roque	:	Fast Retransmit/Recovery.
 *					Two receive queues.
 *					Retransmit queue handled by TCP.
 *					Better retransmit timer handling.
 *					New congestion avoidance.
 *					Header prediction.
 *					Variable renaming.
 *
 *		Eric		:	Fast Retransmit.
 *		Randy Scott	:	MSS option defines.
 *		Eric Schenk	:	Fixes to slow start algorithm.
 *		Eric Schenk	:	Yet another double ACK bug.
 *		Eric Schenk	:	Delayed ACK bug fixes.
 *		Eric Schenk	:	Floyd style fast retrans war avoidance.
 *		David S. Miller	:	Don't allow zero congestion window.
 *		Eric Schenk	:	Fix retransmitter so that it sends
 *					next packet on ack of previous packet.
 *		Andi Kleen	:	Moved open_request checking here
 *					and process RSTs for open_requests.
 *		Andi Kleen	:	Better prune_queue, and other fixes.
S
Stephen Hemminger 已提交
45
 *		Andrey Savochkin:	Fix RTT measurements in the presence of
L
Linus Torvalds 已提交
46 47 48 49 50 51 52
 *					timestamps.
 *		Andrey Savochkin:	Check sequence numbers correctly when
 *					removing SACKs due to in sequence incoming
 *					data segments.
 *		Andi Kleen:		Make sure we never ack data there is not
 *					enough room for. Also make this condition
 *					a fatal error if it might still happen.
53
 *		Andi Kleen:		Add tcp_measure_rcv_mss to make
L
Linus Torvalds 已提交
54
 *					connections with MSS<min(MTU,ann. MSS)
55
 *					work without delayed acks.
L
Linus Torvalds 已提交
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72
 *		Andi Kleen:		Process packets with PSH set in the
 *					fast path.
 *		J Hadi Salim:		ECN support
 *	 	Andrei Gurtov,
 *		Pasi Sarolahti,
 *		Panu Kuhlberg:		Experimental audit of TCP (re)transmission
 *					engine. Lots of bugs are found.
 *		Pasi Sarolahti:		F-RTO for dealing with spurious RTOs
 */

#include <linux/mm.h>
#include <linux/module.h>
#include <linux/sysctl.h>
#include <net/tcp.h>
#include <net/inet_common.h>
#include <linux/ipsec.h>
#include <asm/unaligned.h>
C
Chris Leech 已提交
73
#include <net/netdma.h>
L
Linus Torvalds 已提交
74

75 76 77 78 79 80 81 82 83
int sysctl_tcp_timestamps __read_mostly = 1;
int sysctl_tcp_window_scaling __read_mostly = 1;
int sysctl_tcp_sack __read_mostly = 1;
int sysctl_tcp_fack __read_mostly = 1;
int sysctl_tcp_reordering __read_mostly = TCP_FASTRETRANS_THRESH;
int sysctl_tcp_ecn __read_mostly;
int sysctl_tcp_dsack __read_mostly = 1;
int sysctl_tcp_app_win __read_mostly = 31;
int sysctl_tcp_adv_win_scale __read_mostly = 2;
L
Linus Torvalds 已提交
84

85 86 87 88 89
int sysctl_tcp_stdurg __read_mostly;
int sysctl_tcp_rfc1337 __read_mostly;
int sysctl_tcp_max_orphans __read_mostly = NR_FILE;
int sysctl_tcp_frto __read_mostly;
int sysctl_tcp_nometrics_save __read_mostly;
L
Linus Torvalds 已提交
90

91 92
int sysctl_tcp_moderate_rcvbuf __read_mostly = 1;
int sysctl_tcp_abc __read_mostly;
L
Linus Torvalds 已提交
93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114

#define FLAG_DATA		0x01 /* Incoming frame contained data.		*/
#define FLAG_WIN_UPDATE		0x02 /* Incoming ACK was a window update.	*/
#define FLAG_DATA_ACKED		0x04 /* This ACK acknowledged new data.		*/
#define FLAG_RETRANS_DATA_ACKED	0x08 /* "" "" some of which was retransmitted.	*/
#define FLAG_SYN_ACKED		0x10 /* This ACK acknowledged SYN.		*/
#define FLAG_DATA_SACKED	0x20 /* New SACK.				*/
#define FLAG_ECE		0x40 /* ECE in this ACK				*/
#define FLAG_DATA_LOST		0x80 /* SACK detected data lossage.		*/
#define FLAG_SLOWPATH		0x100 /* Do not skip RFC checks for window update.*/

#define FLAG_ACKED		(FLAG_DATA_ACKED|FLAG_SYN_ACKED)
#define FLAG_NOT_DUP		(FLAG_DATA|FLAG_WIN_UPDATE|FLAG_ACKED)
#define FLAG_CA_ALERT		(FLAG_DATA_SACKED|FLAG_ECE)
#define FLAG_FORWARD_PROGRESS	(FLAG_ACKED|FLAG_DATA_SACKED)

#define IsReno(tp) ((tp)->rx_opt.sack_ok == 0)
#define IsFack(tp) ((tp)->rx_opt.sack_ok & 2)
#define IsDSack(tp) ((tp)->rx_opt.sack_ok & 4)

#define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH)

115
/* Adapt the MSS value used to make delayed ack decision to the
L
Linus Torvalds 已提交
116
 * real world.
117
 */
S
Stephen Hemminger 已提交
118 119
static void tcp_measure_rcv_mss(struct sock *sk,
				const struct sk_buff *skb)
L
Linus Torvalds 已提交
120
{
121
	struct inet_connection_sock *icsk = inet_csk(sk);
122
	const unsigned int lss = icsk->icsk_ack.last_seg_size;
123
	unsigned int len;
L
Linus Torvalds 已提交
124

125
	icsk->icsk_ack.last_seg_size = 0;
L
Linus Torvalds 已提交
126 127 128 129

	/* skb->len may jitter because of SACKs, even if peer
	 * sends good full-sized frames.
	 */
H
Herbert Xu 已提交
130
	len = skb_shinfo(skb)->gso_size ?: skb->len;
131 132
	if (len >= icsk->icsk_ack.rcv_mss) {
		icsk->icsk_ack.rcv_mss = len;
L
Linus Torvalds 已提交
133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
	} else {
		/* Otherwise, we make more careful check taking into account,
		 * that SACKs block is variable.
		 *
		 * "len" is invariant segment length, including TCP header.
		 */
		len += skb->data - skb->h.raw;
		if (len >= TCP_MIN_RCVMSS + sizeof(struct tcphdr) ||
		    /* If PSH is not set, packet should be
		     * full sized, provided peer TCP is not badly broken.
		     * This observation (if it is correct 8)) allows
		     * to handle super-low mtu links fairly.
		     */
		    (len >= TCP_MIN_MSS + sizeof(struct tcphdr) &&
		     !(tcp_flag_word(skb->h.th)&TCP_REMNANT))) {
			/* Subtract also invariant (if peer is RFC compliant),
			 * tcp header plus fixed timestamp option length.
			 * Resulting "len" is MSS free of SACK jitter.
			 */
152 153
			len -= tcp_sk(sk)->tcp_header_len;
			icsk->icsk_ack.last_seg_size = len;
L
Linus Torvalds 已提交
154
			if (len == lss) {
155
				icsk->icsk_ack.rcv_mss = len;
L
Linus Torvalds 已提交
156 157 158
				return;
			}
		}
159 160
		if (icsk->icsk_ack.pending & ICSK_ACK_PUSHED)
			icsk->icsk_ack.pending |= ICSK_ACK_PUSHED2;
161
		icsk->icsk_ack.pending |= ICSK_ACK_PUSHED;
L
Linus Torvalds 已提交
162 163 164
	}
}

165
static void tcp_incr_quickack(struct sock *sk)
L
Linus Torvalds 已提交
166
{
167 168
	struct inet_connection_sock *icsk = inet_csk(sk);
	unsigned quickacks = tcp_sk(sk)->rcv_wnd / (2 * icsk->icsk_ack.rcv_mss);
L
Linus Torvalds 已提交
169 170 171

	if (quickacks==0)
		quickacks=2;
172 173
	if (quickacks > icsk->icsk_ack.quick)
		icsk->icsk_ack.quick = min(quickacks, TCP_MAX_QUICKACKS);
L
Linus Torvalds 已提交
174 175
}

176
void tcp_enter_quickack_mode(struct sock *sk)
L
Linus Torvalds 已提交
177
{
178 179 180 181
	struct inet_connection_sock *icsk = inet_csk(sk);
	tcp_incr_quickack(sk);
	icsk->icsk_ack.pingpong = 0;
	icsk->icsk_ack.ato = TCP_ATO_MIN;
L
Linus Torvalds 已提交
182 183 184 185 186 187
}

/* Send ACKs quickly, if "quick" count is not exhausted
 * and the session is not interactive.
 */

188
static inline int tcp_in_quickack_mode(const struct sock *sk)
L
Linus Torvalds 已提交
189
{
190 191
	const struct inet_connection_sock *icsk = inet_csk(sk);
	return icsk->icsk_ack.quick && !icsk->icsk_ack.pingpong;
L
Linus Torvalds 已提交
192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228
}

/* Buffer size and advertised window tuning.
 *
 * 1. Tuning sk->sk_sndbuf, when connection enters established state.
 */

static void tcp_fixup_sndbuf(struct sock *sk)
{
	int sndmem = tcp_sk(sk)->rx_opt.mss_clamp + MAX_TCP_HEADER + 16 +
		     sizeof(struct sk_buff);

	if (sk->sk_sndbuf < 3 * sndmem)
		sk->sk_sndbuf = min(3 * sndmem, sysctl_tcp_wmem[2]);
}

/* 2. Tuning advertised window (window_clamp, rcv_ssthresh)
 *
 * All tcp_full_space() is split to two parts: "network" buffer, allocated
 * forward and advertised in receiver window (tp->rcv_wnd) and
 * "application buffer", required to isolate scheduling/application
 * latencies from network.
 * window_clamp is maximal advertised window. It can be less than
 * tcp_full_space(), in this case tcp_full_space() - window_clamp
 * is reserved for "application" buffer. The less window_clamp is
 * the smoother our behaviour from viewpoint of network, but the lower
 * throughput and the higher sensitivity of the connection to losses. 8)
 *
 * rcv_ssthresh is more strict window_clamp used at "slow start"
 * phase to predict further behaviour of this connection.
 * It is used for two goals:
 * - to enforce header prediction at sender, even when application
 *   requires some significant "application buffer". It is check #1.
 * - to prevent pruning of receive queue because of misprediction
 *   of receiver window. Check #2.
 *
 * The scheme does not work when sender sends good segments opening
S
Stephen Hemminger 已提交
229
 * window and then starts to feed us spaghetti. But it should work
L
Linus Torvalds 已提交
230 231 232 233
 * in common situations. Otherwise, we have to rely on queue collapsing.
 */

/* Slow part of check#2. */
234 235
static int __tcp_grow_window(const struct sock *sk, struct tcp_sock *tp,
			     const struct sk_buff *skb)
L
Linus Torvalds 已提交
236 237 238
{
	/* Optimize this! */
	int truesize = tcp_win_from_space(skb->truesize)/2;
239
	int window = tcp_win_from_space(sysctl_tcp_rmem[2])/2;
L
Linus Torvalds 已提交
240 241 242

	while (tp->rcv_ssthresh <= window) {
		if (truesize <= skb->len)
243
			return 2 * inet_csk(sk)->icsk_ack.rcv_mss;
L
Linus Torvalds 已提交
244 245 246 247 248 249 250

		truesize >>= 1;
		window >>= 1;
	}
	return 0;
}

S
Stephen Hemminger 已提交
251 252
static void tcp_grow_window(struct sock *sk, struct tcp_sock *tp,
			    struct sk_buff *skb)
L
Linus Torvalds 已提交
253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269
{
	/* Check #1 */
	if (tp->rcv_ssthresh < tp->window_clamp &&
	    (int)tp->rcv_ssthresh < tcp_space(sk) &&
	    !tcp_memory_pressure) {
		int incr;

		/* Check #2. Increase window, if skb with such overhead
		 * will fit to rcvbuf in future.
		 */
		if (tcp_win_from_space(skb->truesize) <= skb->len)
			incr = 2*tp->advmss;
		else
			incr = __tcp_grow_window(sk, tp, skb);

		if (incr) {
			tp->rcv_ssthresh = min(tp->rcv_ssthresh + incr, tp->window_clamp);
270
			inet_csk(sk)->icsk_ack.quick |= 1;
L
Linus Torvalds 已提交
271 272 273 274 275 276 277 278 279 280 281 282
		}
	}
}

/* 3. Tuning rcvbuf, when connection enters established state. */

static void tcp_fixup_rcvbuf(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int rcvmem = tp->advmss + MAX_TCP_HEADER + 16 + sizeof(struct sk_buff);

	/* Try to select rcvbuf so that 4 mss-sized segments
S
Stephen Hemminger 已提交
283
	 * will fit to window and corresponding skbs will fit to our rcvbuf.
L
Linus Torvalds 已提交
284 285 286 287 288 289 290 291
	 * (was 3; 4 is minimum to allow fast retransmit to work.)
	 */
	while (tcp_win_from_space(rcvmem) < tp->advmss)
		rcvmem += 128;
	if (sk->sk_rcvbuf < 4 * rcvmem)
		sk->sk_rcvbuf = min(4 * rcvmem, sysctl_tcp_rmem[2]);
}

S
Stephen Hemminger 已提交
292
/* 4. Try to fixup all. It is made immediately after connection enters
L
Linus Torvalds 已提交
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330
 *    established state.
 */
static void tcp_init_buffer_space(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int maxwin;

	if (!(sk->sk_userlocks & SOCK_RCVBUF_LOCK))
		tcp_fixup_rcvbuf(sk);
	if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK))
		tcp_fixup_sndbuf(sk);

	tp->rcvq_space.space = tp->rcv_wnd;

	maxwin = tcp_full_space(sk);

	if (tp->window_clamp >= maxwin) {
		tp->window_clamp = maxwin;

		if (sysctl_tcp_app_win && maxwin > 4 * tp->advmss)
			tp->window_clamp = max(maxwin -
					       (maxwin >> sysctl_tcp_app_win),
					       4 * tp->advmss);
	}

	/* Force reservation of one segment. */
	if (sysctl_tcp_app_win &&
	    tp->window_clamp > 2 * tp->advmss &&
	    tp->window_clamp + tp->advmss > maxwin)
		tp->window_clamp = max(2 * tp->advmss, maxwin - tp->advmss);

	tp->rcv_ssthresh = min(tp->rcv_ssthresh, tp->window_clamp);
	tp->snd_cwnd_stamp = tcp_time_stamp;
}

/* 5. Recalculate window clamp after socket hit its memory bounds. */
static void tcp_clamp_window(struct sock *sk, struct tcp_sock *tp)
{
331
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
332

333
	icsk->icsk_ack.quick = 0;
L
Linus Torvalds 已提交
334

335 336 337 338 339 340
	if (sk->sk_rcvbuf < sysctl_tcp_rmem[2] &&
	    !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) &&
	    !tcp_memory_pressure &&
	    atomic_read(&tcp_memory_allocated) < sysctl_tcp_mem[0]) {
		sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc),
				    sysctl_tcp_rmem[2]);
L
Linus Torvalds 已提交
341
	}
342
	if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
L
Linus Torvalds 已提交
343 344 345
		tp->rcv_ssthresh = min(tp->window_clamp, 2U*tp->advmss);
}

S
Stephen Hemminger 已提交
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365

/* Initialize RCV_MSS value.
 * RCV_MSS is an our guess about MSS used by the peer.
 * We haven't any direct information about the MSS.
 * It's better to underestimate the RCV_MSS rather than overestimate.
 * Overestimations make us ACKing less frequently than needed.
 * Underestimations are more easy to detect and fix by tcp_measure_rcv_mss().
 */
void tcp_initialize_rcv_mss(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	unsigned int hint = min_t(unsigned int, tp->advmss, tp->mss_cache);

	hint = min(hint, tp->rcv_wnd/2);
	hint = min(hint, TCP_MIN_RCVMSS);
	hint = max(hint, TCP_MIN_MSS);

	inet_csk(sk)->icsk_ack.rcv_mss = hint;
}

L
Linus Torvalds 已提交
366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391
/* Receiver "autotuning" code.
 *
 * The algorithm for RTT estimation w/o timestamps is based on
 * Dynamic Right-Sizing (DRS) by Wu Feng and Mike Fisk of LANL.
 * <http://www.lanl.gov/radiant/website/pubs/drs/lacsi2001.ps>
 *
 * More detail on this code can be found at
 * <http://www.psc.edu/~jheffner/senior_thesis.ps>,
 * though this reference is out of date.  A new paper
 * is pending.
 */
static void tcp_rcv_rtt_update(struct tcp_sock *tp, u32 sample, int win_dep)
{
	u32 new_sample = tp->rcv_rtt_est.rtt;
	long m = sample;

	if (m == 0)
		m = 1;

	if (new_sample != 0) {
		/* If we sample in larger samples in the non-timestamp
		 * case, we could grossly overestimate the RTT especially
		 * with chatty applications or bulk transfer apps which
		 * are stalled on filesystem I/O.
		 *
		 * Also, since we are only going for a minimum in the
S
Stephen Hemminger 已提交
392
		 * non-timestamp case, we do not smooth things out
S
Stephen Hemminger 已提交
393
		 * else with timestamps disabled convergence takes too
L
Linus Torvalds 已提交
394 395 396 397 398 399 400 401
		 * long.
		 */
		if (!win_dep) {
			m -= (new_sample >> 3);
			new_sample += m;
		} else if (m < new_sample)
			new_sample = m << 3;
	} else {
S
Stephen Hemminger 已提交
402
		/* No previous measure. */
L
Linus Torvalds 已提交
403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424
		new_sample = m << 3;
	}

	if (tp->rcv_rtt_est.rtt != new_sample)
		tp->rcv_rtt_est.rtt = new_sample;
}

static inline void tcp_rcv_rtt_measure(struct tcp_sock *tp)
{
	if (tp->rcv_rtt_est.time == 0)
		goto new_measure;
	if (before(tp->rcv_nxt, tp->rcv_rtt_est.seq))
		return;
	tcp_rcv_rtt_update(tp,
			   jiffies - tp->rcv_rtt_est.time,
			   1);

new_measure:
	tp->rcv_rtt_est.seq = tp->rcv_nxt + tp->rcv_wnd;
	tp->rcv_rtt_est.time = tcp_time_stamp;
}

425
static inline void tcp_rcv_rtt_measure_ts(struct sock *sk, const struct sk_buff *skb)
L
Linus Torvalds 已提交
426
{
427
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
428 429
	if (tp->rx_opt.rcv_tsecr &&
	    (TCP_SKB_CB(skb)->end_seq -
430
	     TCP_SKB_CB(skb)->seq >= inet_csk(sk)->icsk_ack.rcv_mss))
L
Linus Torvalds 已提交
431 432 433 434 435 436 437 438 439 440 441 442
		tcp_rcv_rtt_update(tp, tcp_time_stamp - tp->rx_opt.rcv_tsecr, 0);
}

/*
 * This function should be called every time data is copied to user space.
 * It calculates the appropriate TCP receive buffer space.
 */
void tcp_rcv_space_adjust(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int time;
	int space;
443

L
Linus Torvalds 已提交
444 445
	if (tp->rcvq_space.time == 0)
		goto new_measure;
446

L
Linus Torvalds 已提交
447 448 449 450
	time = tcp_time_stamp - tp->rcvq_space.time;
	if (time < (tp->rcv_rtt_est.rtt >> 3) ||
	    tp->rcv_rtt_est.rtt == 0)
		return;
451

L
Linus Torvalds 已提交
452 453 454 455 456 457 458 459 460
	space = 2 * (tp->copied_seq - tp->rcvq_space.seq);

	space = max(tp->rcvq_space.space, space);

	if (tp->rcvq_space.space != space) {
		int rcvmem;

		tp->rcvq_space.space = space;

461 462
		if (sysctl_tcp_moderate_rcvbuf &&
		    !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
L
Linus Torvalds 已提交
463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485
			int new_clamp = space;

			/* Receive space grows, normalize in order to
			 * take into account packet headers and sk_buff
			 * structure overhead.
			 */
			space /= tp->advmss;
			if (!space)
				space = 1;
			rcvmem = (tp->advmss + MAX_TCP_HEADER +
				  16 + sizeof(struct sk_buff));
			while (tcp_win_from_space(rcvmem) < tp->advmss)
				rcvmem += 128;
			space *= rcvmem;
			space = min(space, sysctl_tcp_rmem[2]);
			if (space > sk->sk_rcvbuf) {
				sk->sk_rcvbuf = space;

				/* Make the window clamp follow along.  */
				tp->window_clamp = new_clamp;
			}
		}
	}
486

L
Linus Torvalds 已提交
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503
new_measure:
	tp->rcvq_space.seq = tp->copied_seq;
	tp->rcvq_space.time = tcp_time_stamp;
}

/* There is something which you must keep in mind when you analyze the
 * behavior of the tp->ato delayed ack timeout interval.  When a
 * connection starts up, we want to ack as quickly as possible.  The
 * problem is that "good" TCP's do slow start at the beginning of data
 * transmission.  The means that until we send the first few ACK's the
 * sender will sit on his end and only queue most of his data, because
 * he can only send snd_cwnd unacked packets at any given time.  For
 * each ACK we send, he increments snd_cwnd and transmits more of his
 * queue.  -DaveM
 */
static void tcp_event_data_recv(struct sock *sk, struct tcp_sock *tp, struct sk_buff *skb)
{
504
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
505 506
	u32 now;

507
	inet_csk_schedule_ack(sk);
L
Linus Torvalds 已提交
508

509
	tcp_measure_rcv_mss(sk, skb);
L
Linus Torvalds 已提交
510 511

	tcp_rcv_rtt_measure(tp);
512

L
Linus Torvalds 已提交
513 514
	now = tcp_time_stamp;

515
	if (!icsk->icsk_ack.ato) {
L
Linus Torvalds 已提交
516 517 518
		/* The _first_ data packet received, initialize
		 * delayed ACK engine.
		 */
519 520
		tcp_incr_quickack(sk);
		icsk->icsk_ack.ato = TCP_ATO_MIN;
L
Linus Torvalds 已提交
521
	} else {
522
		int m = now - icsk->icsk_ack.lrcvtime;
L
Linus Torvalds 已提交
523 524 525

		if (m <= TCP_ATO_MIN/2) {
			/* The fastest case is the first. */
526 527 528 529 530 531
			icsk->icsk_ack.ato = (icsk->icsk_ack.ato >> 1) + TCP_ATO_MIN / 2;
		} else if (m < icsk->icsk_ack.ato) {
			icsk->icsk_ack.ato = (icsk->icsk_ack.ato >> 1) + m;
			if (icsk->icsk_ack.ato > icsk->icsk_rto)
				icsk->icsk_ack.ato = icsk->icsk_rto;
		} else if (m > icsk->icsk_rto) {
S
Stephen Hemminger 已提交
532
			/* Too long gap. Apparently sender failed to
L
Linus Torvalds 已提交
533 534
			 * restart window, so that we send ACKs quickly.
			 */
535
			tcp_incr_quickack(sk);
L
Linus Torvalds 已提交
536 537 538
			sk_stream_mem_reclaim(sk);
		}
	}
539
	icsk->icsk_ack.lrcvtime = now;
L
Linus Torvalds 已提交
540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555

	TCP_ECN_check_ce(tp, skb);

	if (skb->len >= 128)
		tcp_grow_window(sk, tp, skb);
}

/* Called to compute a smoothed rtt estimate. The data fed to this
 * routine either comes from timestamps, or from segments that were
 * known _not_ to have been retransmitted [see Karn/Partridge
 * Proceedings SIGCOMM 87]. The algorithm is from the SIGCOMM 88
 * piece by Van Jacobson.
 * NOTE: the next three routines used to be one big routine.
 * To save cycles in the RFC 1323 implementation it was better to break
 * it up into three procedures. -- erics
 */
556
static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt)
L
Linus Torvalds 已提交
557
{
558
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
559 560 561 562 563
	long m = mrtt; /* RTT */

	/*	The following amusing code comes from Jacobson's
	 *	article in SIGCOMM '88.  Note that rtt and mdev
	 *	are scaled versions of rtt and mean deviation.
564
	 *	This is designed to be as fast as possible
L
Linus Torvalds 已提交
565 566 567 568 569 570 571
	 *	m stands for "measurement".
	 *
	 *	On a 1990 paper the rto value is changed to:
	 *	RTO = rtt + 4 * mdev
	 *
	 * Funny. This algorithm seems to be very broken.
	 * These formulae increase RTO, when it should be decreased, increase
S
Stephen Hemminger 已提交
572
	 * too slowly, when it should be increased quickly, decrease too quickly
L
Linus Torvalds 已提交
573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621
	 * etc. I guess in BSD RTO takes ONE value, so that it is absolutely
	 * does not matter how to _calculate_ it. Seems, it was trap
	 * that VJ failed to avoid. 8)
	 */
	if(m == 0)
		m = 1;
	if (tp->srtt != 0) {
		m -= (tp->srtt >> 3);	/* m is now error in rtt est */
		tp->srtt += m;		/* rtt = 7/8 rtt + 1/8 new */
		if (m < 0) {
			m = -m;		/* m is now abs(error) */
			m -= (tp->mdev >> 2);   /* similar update on mdev */
			/* This is similar to one of Eifel findings.
			 * Eifel blocks mdev updates when rtt decreases.
			 * This solution is a bit different: we use finer gain
			 * for mdev in this case (alpha*beta).
			 * Like Eifel it also prevents growth of rto,
			 * but also it limits too fast rto decreases,
			 * happening in pure Eifel.
			 */
			if (m > 0)
				m >>= 3;
		} else {
			m -= (tp->mdev >> 2);   /* similar update on mdev */
		}
		tp->mdev += m;	    	/* mdev = 3/4 mdev + 1/4 new */
		if (tp->mdev > tp->mdev_max) {
			tp->mdev_max = tp->mdev;
			if (tp->mdev_max > tp->rttvar)
				tp->rttvar = tp->mdev_max;
		}
		if (after(tp->snd_una, tp->rtt_seq)) {
			if (tp->mdev_max < tp->rttvar)
				tp->rttvar -= (tp->rttvar-tp->mdev_max)>>2;
			tp->rtt_seq = tp->snd_nxt;
			tp->mdev_max = TCP_RTO_MIN;
		}
	} else {
		/* no previous measure. */
		tp->srtt = m<<3;	/* take the measured time to be rtt */
		tp->mdev = m<<1;	/* make sure rto = 3*rtt */
		tp->mdev_max = tp->rttvar = max(tp->mdev, TCP_RTO_MIN);
		tp->rtt_seq = tp->snd_nxt;
	}
}

/* Calculate rto without backoff.  This is the second half of Van Jacobson's
 * routine referred to above.
 */
622
static inline void tcp_set_rto(struct sock *sk)
L
Linus Torvalds 已提交
623
{
624
	const struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
625 626 627 628 629 630 631 632
	/* Old crap is replaced with new one. 8)
	 *
	 * More seriously:
	 * 1. If rtt variance happened to be less 50msec, it is hallucination.
	 *    It cannot be less due to utterly erratic ACK generation made
	 *    at least by solaris and freebsd. "Erratic ACKs" has _nothing_
	 *    to do with delayed acks, because at cwnd>2 true delack timeout
	 *    is invisible. Actually, Linux-2.4 also generates erratic
S
Stephen Hemminger 已提交
633
	 *    ACKs in some circumstances.
L
Linus Torvalds 已提交
634
	 */
635
	inet_csk(sk)->icsk_rto = (tp->srtt >> 3) + tp->rttvar;
L
Linus Torvalds 已提交
636 637 638 639

	/* 2. Fixups made earlier cannot be right.
	 *    If we do not estimate RTO correctly without them,
	 *    all the algo is pure shit and should be replaced
S
Stephen Hemminger 已提交
640
	 *    with correct one. It is exactly, which we pretend to do.
L
Linus Torvalds 已提交
641 642 643 644 645 646
	 */
}

/* NOTE: clamping at TCP_RTO_MIN is not required, current algo
 * guarantees that rto is higher.
 */
647
static inline void tcp_bound_rto(struct sock *sk)
L
Linus Torvalds 已提交
648
{
649 650
	if (inet_csk(sk)->icsk_rto > TCP_RTO_MAX)
		inet_csk(sk)->icsk_rto = TCP_RTO_MAX;
L
Linus Torvalds 已提交
651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667
}

/* Save metrics learned by this TCP session.
   This function is called only, when TCP finishes successfully
   i.e. when it enters TIME-WAIT or goes from LAST-ACK to CLOSE.
 */
void tcp_update_metrics(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct dst_entry *dst = __sk_dst_get(sk);

	if (sysctl_tcp_nometrics_save)
		return;

	dst_confirm(dst);

	if (dst && (dst->flags&DST_HOST)) {
668
		const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
669 670
		int m;

671
		if (icsk->icsk_backoff || !tp->srtt) {
L
Linus Torvalds 已提交
672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719
			/* This session failed to estimate rtt. Why?
			 * Probably, no packets returned in time.
			 * Reset our results.
			 */
			if (!(dst_metric_locked(dst, RTAX_RTT)))
				dst->metrics[RTAX_RTT-1] = 0;
			return;
		}

		m = dst_metric(dst, RTAX_RTT) - tp->srtt;

		/* If newly calculated rtt larger than stored one,
		 * store new one. Otherwise, use EWMA. Remember,
		 * rtt overestimation is always better than underestimation.
		 */
		if (!(dst_metric_locked(dst, RTAX_RTT))) {
			if (m <= 0)
				dst->metrics[RTAX_RTT-1] = tp->srtt;
			else
				dst->metrics[RTAX_RTT-1] -= (m>>3);
		}

		if (!(dst_metric_locked(dst, RTAX_RTTVAR))) {
			if (m < 0)
				m = -m;

			/* Scale deviation to rttvar fixed point */
			m >>= 1;
			if (m < tp->mdev)
				m = tp->mdev;

			if (m >= dst_metric(dst, RTAX_RTTVAR))
				dst->metrics[RTAX_RTTVAR-1] = m;
			else
				dst->metrics[RTAX_RTTVAR-1] -=
					(dst->metrics[RTAX_RTTVAR-1] - m)>>2;
		}

		if (tp->snd_ssthresh >= 0xFFFF) {
			/* Slow start still did not finish. */
			if (dst_metric(dst, RTAX_SSTHRESH) &&
			    !dst_metric_locked(dst, RTAX_SSTHRESH) &&
			    (tp->snd_cwnd >> 1) > dst_metric(dst, RTAX_SSTHRESH))
				dst->metrics[RTAX_SSTHRESH-1] = tp->snd_cwnd >> 1;
			if (!dst_metric_locked(dst, RTAX_CWND) &&
			    tp->snd_cwnd > dst_metric(dst, RTAX_CWND))
				dst->metrics[RTAX_CWND-1] = tp->snd_cwnd;
		} else if (tp->snd_cwnd > tp->snd_ssthresh &&
720
			   icsk->icsk_ca_state == TCP_CA_Open) {
L
Linus Torvalds 已提交
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
			/* Cong. avoidance phase, cwnd is reliable. */
			if (!dst_metric_locked(dst, RTAX_SSTHRESH))
				dst->metrics[RTAX_SSTHRESH-1] =
					max(tp->snd_cwnd >> 1, tp->snd_ssthresh);
			if (!dst_metric_locked(dst, RTAX_CWND))
				dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_cwnd) >> 1;
		} else {
			/* Else slow start did not finish, cwnd is non-sense,
			   ssthresh may be also invalid.
			 */
			if (!dst_metric_locked(dst, RTAX_CWND))
				dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_ssthresh) >> 1;
			if (dst->metrics[RTAX_SSTHRESH-1] &&
			    !dst_metric_locked(dst, RTAX_SSTHRESH) &&
			    tp->snd_ssthresh > dst->metrics[RTAX_SSTHRESH-1])
				dst->metrics[RTAX_SSTHRESH-1] = tp->snd_ssthresh;
		}

		if (!dst_metric_locked(dst, RTAX_REORDERING)) {
			if (dst->metrics[RTAX_REORDERING-1] < tp->reordering &&
			    tp->reordering != sysctl_tcp_reordering)
				dst->metrics[RTAX_REORDERING-1] = tp->reordering;
		}
	}
}

/* Numbers are taken from RFC2414.  */
__u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst)
{
	__u32 cwnd = (dst ? dst_metric(dst, RTAX_INITCWND) : 0);

	if (!cwnd) {
753
		if (tp->mss_cache > 1460)
L
Linus Torvalds 已提交
754 755
			cwnd = 2;
		else
756
			cwnd = (tp->mss_cache > 1095) ? 3 : 4;
L
Linus Torvalds 已提交
757 758 759 760
	}
	return min_t(__u32, cwnd, tp->snd_cwnd_clamp);
}

S
Stephen Hemminger 已提交
761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781
/* Set slow start threshold and cwnd not falling to slow start */
void tcp_enter_cwr(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

	tp->prior_ssthresh = 0;
	tp->bytes_acked = 0;
	if (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) {
		tp->undo_marker = 0;
		tp->snd_ssthresh = inet_csk(sk)->icsk_ca_ops->ssthresh(sk);
		tp->snd_cwnd = min(tp->snd_cwnd,
				   tcp_packets_in_flight(tp) + 1U);
		tp->snd_cwnd_cnt = 0;
		tp->high_seq = tp->snd_nxt;
		tp->snd_cwnd_stamp = tcp_time_stamp;
		TCP_ECN_queue_cwr(tp);

		tcp_set_ca_state(sk, TCP_CA_CWR);
	}
}

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
/* Initialize metrics on socket. */

static void tcp_init_metrics(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct dst_entry *dst = __sk_dst_get(sk);

	if (dst == NULL)
		goto reset;

	dst_confirm(dst);

	if (dst_metric_locked(dst, RTAX_CWND))
		tp->snd_cwnd_clamp = dst_metric(dst, RTAX_CWND);
	if (dst_metric(dst, RTAX_SSTHRESH)) {
		tp->snd_ssthresh = dst_metric(dst, RTAX_SSTHRESH);
		if (tp->snd_ssthresh > tp->snd_cwnd_clamp)
			tp->snd_ssthresh = tp->snd_cwnd_clamp;
	}
	if (dst_metric(dst, RTAX_REORDERING) &&
	    tp->reordering != dst_metric(dst, RTAX_REORDERING)) {
		tp->rx_opt.sack_ok &= ~2;
		tp->reordering = dst_metric(dst, RTAX_REORDERING);
	}

	if (dst_metric(dst, RTAX_RTT) == 0)
		goto reset;

	if (!tp->srtt && dst_metric(dst, RTAX_RTT) < (TCP_TIMEOUT_INIT << 3))
		goto reset;

	/* Initial rtt is determined from SYN,SYN-ACK.
	 * The segment is small and rtt may appear much
	 * less than real one. Use per-dst memory
	 * to make it more realistic.
	 *
	 * A bit of theory. RTT is time passed after "normal" sized packet
S
Stephen Hemminger 已提交
819
	 * is sent until it is ACKed. In normal circumstances sending small
L
Linus Torvalds 已提交
820 821 822 823 824 825 826 827 828 829 830 831 832 833 834
	 * packets force peer to delay ACKs and calculation is correct too.
	 * The algorithm is adaptive and, provided we follow specs, it
	 * NEVER underestimate RTT. BUT! If peer tries to make some clever
	 * tricks sort of "quick acks" for time long enough to decrease RTT
	 * to low value, and then abruptly stops to do it and starts to delay
	 * ACKs, wait for troubles.
	 */
	if (dst_metric(dst, RTAX_RTT) > tp->srtt) {
		tp->srtt = dst_metric(dst, RTAX_RTT);
		tp->rtt_seq = tp->snd_nxt;
	}
	if (dst_metric(dst, RTAX_RTTVAR) > tp->mdev) {
		tp->mdev = dst_metric(dst, RTAX_RTTVAR);
		tp->mdev_max = tp->rttvar = max(tp->mdev, TCP_RTO_MIN);
	}
835 836 837
	tcp_set_rto(sk);
	tcp_bound_rto(sk);
	if (inet_csk(sk)->icsk_rto < TCP_TIMEOUT_INIT && !tp->rx_opt.saw_tstamp)
L
Linus Torvalds 已提交
838 839 840 841 842 843 844 845 846 847 848 849 850
		goto reset;
	tp->snd_cwnd = tcp_init_cwnd(tp, dst);
	tp->snd_cwnd_stamp = tcp_time_stamp;
	return;

reset:
	/* Play conservative. If timestamps are not
	 * supported, TCP will fail to recalculate correct
	 * rtt, if initial rto is too small. FORGET ALL AND RESET!
	 */
	if (!tp->rx_opt.saw_tstamp && tp->srtt) {
		tp->srtt = 0;
		tp->mdev = tp->mdev_max = tp->rttvar = TCP_TIMEOUT_INIT;
851
		inet_csk(sk)->icsk_rto = TCP_TIMEOUT_INIT;
L
Linus Torvalds 已提交
852 853 854
	}
}

855 856
static void tcp_update_reordering(struct sock *sk, const int metric,
				  const int ts)
L
Linus Torvalds 已提交
857
{
858
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
859 860 861 862 863 864 865 866 867 868 869 870 871 872
	if (metric > tp->reordering) {
		tp->reordering = min(TCP_MAX_REORDERING, metric);

		/* This exciting event is worth to be remembered. 8) */
		if (ts)
			NET_INC_STATS_BH(LINUX_MIB_TCPTSREORDER);
		else if (IsReno(tp))
			NET_INC_STATS_BH(LINUX_MIB_TCPRENOREORDER);
		else if (IsFack(tp))
			NET_INC_STATS_BH(LINUX_MIB_TCPFACKREORDER);
		else
			NET_INC_STATS_BH(LINUX_MIB_TCPSACKREORDER);
#if FASTRETRANS_DEBUG > 1
		printk(KERN_DEBUG "Disorder%d %d %u f%u s%u rr%d\n",
873
		       tp->rx_opt.sack_ok, inet_csk(sk)->icsk_ca_state,
L
Linus Torvalds 已提交
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
		       tp->reordering,
		       tp->fackets_out,
		       tp->sacked_out,
		       tp->undo_marker ? tp->undo_retrans : 0);
#endif
		/* Disable FACK yet. */
		tp->rx_opt.sack_ok &= ~2;
	}
}

/* This procedure tags the retransmission queue when SACKs arrive.
 *
 * We have three tag bits: SACKED(S), RETRANS(R) and LOST(L).
 * Packets in queue with these bits set are counted in variables
 * sacked_out, retrans_out and lost_out, correspondingly.
 *
 * Valid combinations are:
 * Tag  InFlight	Description
 * 0	1		- orig segment is in flight.
 * S	0		- nothing flies, orig reached receiver.
 * L	0		- nothing flies, orig lost by net.
 * R	2		- both orig and retransmit are in flight.
 * L|R	1		- orig is lost, retransmit is in flight.
 * S|R  1		- orig reached receiver, retrans is still in flight.
 * (L|S|R is logically valid, it could occur when L|R is sacked,
 *  but it is equivalent to plain S and code short-curcuits it to S.
 *  L|S is logically invalid, it would mean -1 packet in flight 8))
 *
 * These 6 states form finite state machine, controlled by the following events:
 * 1. New ACK (+SACK) arrives. (tcp_sacktag_write_queue())
 * 2. Retransmission. (tcp_retransmit_skb(), tcp_xmit_retransmit_queue())
 * 3. Loss detection event of one of three flavors:
 *	A. Scoreboard estimator decided the packet is lost.
 *	   A'. Reno "three dupacks" marks head of queue lost.
 *	   A''. Its FACK modfication, head until snd.fack is lost.
 *	B. SACK arrives sacking data transmitted after never retransmitted
 *	   hole was sent out.
 *	C. SACK arrives sacking SND.NXT at the moment, when the
 *	   segment was retransmitted.
 * 4. D-SACK added new rule: D-SACK changes any tag to S.
 *
 * It is pleasant to note, that state diagram turns out to be commutative,
 * so that we are allowed not to be bothered by order of our actions,
 * when multiple events arrive simultaneously. (see the function below).
 *
 * Reordering detection.
 * --------------------
 * Reordering metric is maximal distance, which a packet can be displaced
 * in packet stream. With SACKs we can estimate it:
 *
 * 1. SACK fills old hole and the corresponding segment was not
 *    ever retransmitted -> reordering. Alas, we cannot use it
 *    when segment was retransmitted.
 * 2. The last flaw is solved with D-SACK. D-SACK arrives
 *    for retransmitted and already SACKed segment -> reordering..
 * Both of these heuristics are not used in Loss state, when we cannot
 * account for retransmits accurately.
 */
static int
tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_una)
{
935
	const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
936 937
	struct tcp_sock *tp = tcp_sk(sk);
	unsigned char *ptr = ack_skb->h.raw + TCP_SKB_CB(ack_skb)->sacked;
938
	struct tcp_sack_block_wire *sp = (struct tcp_sack_block_wire *)(ptr+2);
939
	struct sk_buff *cached_skb;
L
Linus Torvalds 已提交
940 941 942 943 944
	int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3;
	int reord = tp->packets_out;
	int prior_fackets;
	u32 lost_retrans = 0;
	int flag = 0;
945
	int dup_sack = 0;
946
	int cached_fack_count;
L
Linus Torvalds 已提交
947
	int i;
948
	int first_sack_index;
L
Linus Torvalds 已提交
949 950 951 952 953

	if (!tp->sacked_out)
		tp->fackets_out = 0;
	prior_fackets = tp->fackets_out;

954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980
	/* Check for D-SACK. */
	if (before(ntohl(sp[0].start_seq), TCP_SKB_CB(ack_skb)->ack_seq)) {
		dup_sack = 1;
		tp->rx_opt.sack_ok |= 4;
		NET_INC_STATS_BH(LINUX_MIB_TCPDSACKRECV);
	} else if (num_sacks > 1 &&
			!after(ntohl(sp[0].end_seq), ntohl(sp[1].end_seq)) &&
			!before(ntohl(sp[0].start_seq), ntohl(sp[1].start_seq))) {
		dup_sack = 1;
		tp->rx_opt.sack_ok |= 4;
		NET_INC_STATS_BH(LINUX_MIB_TCPDSACKOFORECV);
	}

	/* D-SACK for already forgotten data...
	 * Do dumb counting. */
	if (dup_sack &&
			!after(ntohl(sp[0].end_seq), prior_snd_una) &&
			after(ntohl(sp[0].end_seq), tp->undo_marker))
		tp->undo_retrans--;

	/* Eliminate too old ACKs, but take into
	 * account more or less fresh ones, they can
	 * contain valid SACK info.
	 */
	if (before(TCP_SKB_CB(ack_skb)->ack_seq, prior_snd_una - tp->max_window))
		return 0;

981 982 983 984 985
	/* SACK fastpath:
	 * if the only SACK change is the increase of the end_seq of
	 * the first block then only apply that SACK block
	 * and use retrans queue hinting otherwise slowpath */
	flag = 1;
986 987 988
	for (i = 0; i < num_sacks; i++) {
		__be32 start_seq = sp[i].start_seq;
		__be32 end_seq = sp[i].end_seq;
989

990
		if (i == 0) {
991 992 993 994 995 996 997 998 999 1000
			if (tp->recv_sack_cache[i].start_seq != start_seq)
				flag = 0;
		} else {
			if ((tp->recv_sack_cache[i].start_seq != start_seq) ||
			    (tp->recv_sack_cache[i].end_seq != end_seq))
				flag = 0;
		}
		tp->recv_sack_cache[i].start_seq = start_seq;
		tp->recv_sack_cache[i].end_seq = end_seq;
	}
1001 1002 1003 1004 1005
	/* Clear the rest of the cache sack blocks so they won't match mistakenly. */
	for (; i < ARRAY_SIZE(tp->recv_sack_cache); i++) {
		tp->recv_sack_cache[i].start_seq = 0;
		tp->recv_sack_cache[i].end_seq = 0;
	}
1006

1007
	first_sack_index = 0;
1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018
	if (flag)
		num_sacks = 1;
	else {
		int j;
		tp->fastpath_skb_hint = NULL;

		/* order SACK blocks to allow in order walk of the retrans queue */
		for (i = num_sacks-1; i > 0; i--) {
			for (j = 0; j < i; j++){
				if (after(ntohl(sp[j].start_seq),
					  ntohl(sp[j+1].start_seq))){
B
Baruch Even 已提交
1019 1020 1021 1022 1023
					struct tcp_sack_block_wire tmp;

					tmp = sp[j];
					sp[j] = sp[j+1];
					sp[j+1] = tmp;
1024 1025 1026 1027

					/* Track where the first SACK block goes to */
					if (j == first_sack_index)
						first_sack_index = j+1;
1028 1029 1030 1031 1032 1033 1034 1035 1036
				}

			}
		}
	}

	/* clear flag as used for different purpose in following code */
	flag = 0;

1037 1038 1039 1040 1041 1042 1043 1044
	/* Use SACK fastpath hint if valid */
	cached_skb = tp->fastpath_skb_hint;
	cached_fack_count = tp->fastpath_cnt_hint;
	if (!cached_skb) {
		cached_skb = sk->sk_write_queue.next;
		cached_fack_count = 0;
	}

1045 1046 1047 1048 1049 1050
	for (i=0; i<num_sacks; i++, sp++) {
		struct sk_buff *skb;
		__u32 start_seq = ntohl(sp->start_seq);
		__u32 end_seq = ntohl(sp->end_seq);
		int fack_count;

1051 1052
		skb = cached_skb;
		fack_count = cached_fack_count;
L
Linus Torvalds 已提交
1053 1054 1055 1056 1057

		/* Event "B" in the comment above. */
		if (after(end_seq, tp->high_seq))
			flag |= FLAG_DATA_LOST;

1058
		sk_stream_for_retrans_queue_from(skb, sk) {
1059 1060
			int in_sack, pcount;
			u8 sacked;
L
Linus Torvalds 已提交
1061

1062 1063 1064 1065 1066 1067
			cached_skb = skb;
			cached_fack_count = fack_count;
			if (i == first_sack_index) {
				tp->fastpath_skb_hint = skb;
				tp->fastpath_cnt_hint = fack_count;
			}
1068

L
Linus Torvalds 已提交
1069 1070 1071
			/* The retransmission queue is always in order, so
			 * we can short-circuit the walk early.
			 */
1072
			if (!before(TCP_SKB_CB(skb)->seq, end_seq))
L
Linus Torvalds 已提交
1073 1074
				break;

1075 1076 1077
			in_sack = !after(start_seq, TCP_SKB_CB(skb)->seq) &&
				!before(end_seq, TCP_SKB_CB(skb)->end_seq);

1078 1079
			pcount = tcp_skb_pcount(skb);

1080 1081
			if (pcount > 1 && !in_sack &&
			    after(TCP_SKB_CB(skb)->end_seq, start_seq)) {
1082 1083
				unsigned int pkt_len;

1084 1085 1086 1087
				in_sack = !after(start_seq,
						 TCP_SKB_CB(skb)->seq);

				if (!in_sack)
1088 1089 1090 1091 1092
					pkt_len = (start_seq -
						   TCP_SKB_CB(skb)->seq);
				else
					pkt_len = (end_seq -
						   TCP_SKB_CB(skb)->seq);
1093
				if (tcp_fragment(sk, skb, pkt_len, skb_shinfo(skb)->gso_size))
1094 1095 1096 1097 1098
					break;
				pcount = tcp_skb_pcount(skb);
			}

			fack_count += pcount;
L
Linus Torvalds 已提交
1099

1100 1101
			sacked = TCP_SKB_CB(skb)->sacked;

L
Linus Torvalds 已提交
1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142
			/* Account D-SACK for retransmitted packet. */
			if ((dup_sack && in_sack) &&
			    (sacked & TCPCB_RETRANS) &&
			    after(TCP_SKB_CB(skb)->end_seq, tp->undo_marker))
				tp->undo_retrans--;

			/* The frame is ACKed. */
			if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una)) {
				if (sacked&TCPCB_RETRANS) {
					if ((dup_sack && in_sack) &&
					    (sacked&TCPCB_SACKED_ACKED))
						reord = min(fack_count, reord);
				} else {
					/* If it was in a hole, we detected reordering. */
					if (fack_count < prior_fackets &&
					    !(sacked&TCPCB_SACKED_ACKED))
						reord = min(fack_count, reord);
				}

				/* Nothing to do; acked frame is about to be dropped. */
				continue;
			}

			if ((sacked&TCPCB_SACKED_RETRANS) &&
			    after(end_seq, TCP_SKB_CB(skb)->ack_seq) &&
			    (!lost_retrans || after(end_seq, lost_retrans)))
				lost_retrans = end_seq;

			if (!in_sack)
				continue;

			if (!(sacked&TCPCB_SACKED_ACKED)) {
				if (sacked & TCPCB_SACKED_RETRANS) {
					/* If the segment is not tagged as lost,
					 * we do not clear RETRANS, believing
					 * that retransmission is still in flight.
					 */
					if (sacked & TCPCB_LOST) {
						TCP_SKB_CB(skb)->sacked &= ~(TCPCB_LOST|TCPCB_SACKED_RETRANS);
						tp->lost_out -= tcp_skb_pcount(skb);
						tp->retrans_out -= tcp_skb_pcount(skb);
1143 1144 1145

						/* clear lost hint */
						tp->retransmit_skb_hint = NULL;
L
Linus Torvalds 已提交
1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157
					}
				} else {
					/* New sack for not retransmitted frame,
					 * which was in hole. It is reordering.
					 */
					if (!(sacked & TCPCB_RETRANS) &&
					    fack_count < prior_fackets)
						reord = min(fack_count, reord);

					if (sacked & TCPCB_LOST) {
						TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
						tp->lost_out -= tcp_skb_pcount(skb);
1158 1159 1160

						/* clear lost hint */
						tp->retransmit_skb_hint = NULL;
L
Linus Torvalds 已提交
1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183
					}
				}

				TCP_SKB_CB(skb)->sacked |= TCPCB_SACKED_ACKED;
				flag |= FLAG_DATA_SACKED;
				tp->sacked_out += tcp_skb_pcount(skb);

				if (fack_count > tp->fackets_out)
					tp->fackets_out = fack_count;
			} else {
				if (dup_sack && (sacked&TCPCB_RETRANS))
					reord = min(fack_count, reord);
			}

			/* D-SACK. We can detect redundant retransmission
			 * in S|R and plain R frames and clear it.
			 * undo_retrans is decreased above, L|R frames
			 * are accounted above as well.
			 */
			if (dup_sack &&
			    (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS)) {
				TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
				tp->retrans_out -= tcp_skb_pcount(skb);
1184
				tp->retransmit_skb_hint = NULL;
L
Linus Torvalds 已提交
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194
			}
		}
	}

	/* Check for lost retransmit. This superb idea is
	 * borrowed from "ratehalving". Event "C".
	 * Later note: FACK people cheated me again 8),
	 * we have to account for reordering! Ugly,
	 * but should help.
	 */
1195
	if (lost_retrans && icsk->icsk_ca_state == TCP_CA_Recovery) {
L
Linus Torvalds 已提交
1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207
		struct sk_buff *skb;

		sk_stream_for_retrans_queue(skb, sk) {
			if (after(TCP_SKB_CB(skb)->seq, lost_retrans))
				break;
			if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una))
				continue;
			if ((TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS) &&
			    after(lost_retrans, TCP_SKB_CB(skb)->ack_seq) &&
			    (IsFack(tp) ||
			     !before(lost_retrans,
				     TCP_SKB_CB(skb)->ack_seq + tp->reordering *
1208
				     tp->mss_cache))) {
L
Linus Torvalds 已提交
1209 1210 1211
				TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
				tp->retrans_out -= tcp_skb_pcount(skb);

1212 1213 1214
				/* clear lost hint */
				tp->retransmit_skb_hint = NULL;

L
Linus Torvalds 已提交
1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
				if (!(TCP_SKB_CB(skb)->sacked&(TCPCB_LOST|TCPCB_SACKED_ACKED))) {
					tp->lost_out += tcp_skb_pcount(skb);
					TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
					flag |= FLAG_DATA_SACKED;
					NET_INC_STATS_BH(LINUX_MIB_TCPLOSTRETRANSMIT);
				}
			}
		}
	}

	tp->left_out = tp->sacked_out + tp->lost_out;

1227 1228
	if ((reord < tp->fackets_out) && icsk->icsk_ca_state != TCP_CA_Loss)
		tcp_update_reordering(sk, ((tp->fackets_out + 1) - reord), 0);
L
Linus Torvalds 已提交
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238

#if FASTRETRANS_DEBUG > 0
	BUG_TRAP((int)tp->sacked_out >= 0);
	BUG_TRAP((int)tp->lost_out >= 0);
	BUG_TRAP((int)tp->retrans_out >= 0);
	BUG_TRAP((int)tcp_packets_in_flight(tp) >= 0);
#endif
	return flag;
}

1239 1240 1241
/* F-RTO can only be used if these conditions are satisfied:
 *  - there must be some unsent new data
 *  - the advertised window should allow sending it
1242
 *  - TCP has never retransmitted anything other than head
1243
 */
1244
int tcp_use_frto(struct sock *sk)
1245 1246
{
	const struct tcp_sock *tp = tcp_sk(sk);
1247 1248 1249 1250 1251 1252
	struct sk_buff *skb;

	if (!sysctl_tcp_frto || !sk->sk_send_head ||
		after(TCP_SKB_CB(sk->sk_send_head)->end_seq,
		      tp->snd_una + tp->snd_wnd))
		return 0;
1253

1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
	/* Avoid expensive walking of rexmit queue if possible */
	if (tp->retrans_out > 1)
		return 0;

	skb = skb_peek(&sk->sk_write_queue)->next;	/* Skips head */
	sk_stream_for_retrans_queue_from(skb, sk) {
		if (TCP_SKB_CB(skb)->sacked&TCPCB_RETRANS)
			return 0;
		/* Short-circuit when first non-SACKed skb has been checked */
		if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED))
			break;
	}
	return 1;
1267 1268
}

1269 1270
/* RTO occurred, but do not yet enter Loss state. Instead, defer RTO
 * recovery a bit and use heuristics in tcp_process_frto() to detect if
1271 1272 1273 1274 1275
 * the RTO was spurious. Only clear SACKED_RETRANS of the head here to
 * keep retrans_out counting accurate (with SACK F-RTO, other than head
 * may still have that bit set); TCPCB_LOST and remaining SACKED_RETRANS
 * bits are handled if the Loss state is really to be entered (in
 * tcp_enter_frto_loss).
1276 1277 1278 1279
 *
 * Do like tcp_enter_loss() would; when RTO expires the second time it
 * does:
 *  "Reduce ssthresh if it has not yet been made inside this window."
L
Linus Torvalds 已提交
1280 1281 1282
 */
void tcp_enter_frto(struct sock *sk)
{
1283
	const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
1284 1285 1286
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb;

1287
	if ((!tp->frto_counter && icsk->icsk_ca_state <= TCP_CA_Disorder) ||
1288
	    tp->snd_una == tp->high_seq ||
1289 1290
	    ((icsk->icsk_ca_state == TCP_CA_Loss || tp->frto_counter) &&
	     !icsk->icsk_retransmits)) {
1291 1292 1293
		tp->prior_ssthresh = tcp_current_ssthresh(sk);
		tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
		tcp_ca_event(sk, CA_EVENT_FRTO);
L
Linus Torvalds 已提交
1294 1295 1296 1297 1298
	}

	tp->undo_marker = tp->snd_una;
	tp->undo_retrans = 0;

1299 1300
	skb = skb_peek(&sk->sk_write_queue);
	if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) {
1301
		TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
1302
		tp->retrans_out -= tcp_skb_pcount(skb);
L
Linus Torvalds 已提交
1303 1304 1305
	}
	tcp_sync_left_out(tp);

1306 1307
	tcp_set_ca_state(sk, TCP_CA_Disorder);
	tp->high_seq = tp->snd_nxt;
L
Linus Torvalds 已提交
1308
	tp->frto_highmark = tp->snd_nxt;
1309
	tp->frto_counter = 1;
L
Linus Torvalds 已提交
1310 1311 1312 1313 1314 1315
}

/* Enter Loss state after F-RTO was applied. Dupack arrived after RTO,
 * which indicates that we should follow the traditional RTO recovery,
 * i.e. mark everything lost and do go-back-N retransmission.
 */
1316
static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag)
L
Linus Torvalds 已提交
1317 1318 1319 1320 1321 1322 1323 1324
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb;
	int cnt = 0;

	tp->sacked_out = 0;
	tp->lost_out = 0;
	tp->fackets_out = 0;
1325
	tp->retrans_out = 0;
L
Linus Torvalds 已提交
1326 1327 1328

	sk_stream_for_retrans_queue(skb, sk) {
		cnt += tcp_skb_pcount(skb);
1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339
		/*
		 * Count the retransmission made on RTO correctly (only when
		 * waiting for the first ACK and did not get it)...
		 */
		if ((tp->frto_counter == 1) && !(flag&FLAG_DATA_ACKED)) {
			tp->retrans_out += tcp_skb_pcount(skb);
			/* ...enter this if branch just for the first segment */
			flag |= FLAG_DATA_ACKED;
		} else {
			TCP_SKB_CB(skb)->sacked &= ~(TCPCB_LOST|TCPCB_SACKED_RETRANS);
		}
L
Linus Torvalds 已提交
1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356
		if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED)) {

			/* Do not mark those segments lost that were
			 * forward transmitted after RTO
			 */
			if (!after(TCP_SKB_CB(skb)->end_seq,
				   tp->frto_highmark)) {
				TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
				tp->lost_out += tcp_skb_pcount(skb);
			}
		} else {
			tp->sacked_out += tcp_skb_pcount(skb);
			tp->fackets_out = cnt;
		}
	}
	tcp_sync_left_out(tp);

1357
	tp->snd_cwnd = tcp_packets_in_flight(tp) + allowed_segments;
L
Linus Torvalds 已提交
1358 1359 1360 1361 1362 1363 1364
	tp->snd_cwnd_cnt = 0;
	tp->snd_cwnd_stamp = tcp_time_stamp;
	tp->undo_marker = 0;
	tp->frto_counter = 0;

	tp->reordering = min_t(unsigned int, tp->reordering,
					     sysctl_tcp_reordering);
1365
	tcp_set_ca_state(sk, TCP_CA_Loss);
L
Linus Torvalds 已提交
1366 1367
	tp->high_seq = tp->frto_highmark;
	TCP_ECN_queue_cwr(tp);
1368 1369

	clear_all_retrans_hints(tp);
L
Linus Torvalds 已提交
1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390
}

void tcp_clear_retrans(struct tcp_sock *tp)
{
	tp->left_out = 0;
	tp->retrans_out = 0;

	tp->fackets_out = 0;
	tp->sacked_out = 0;
	tp->lost_out = 0;

	tp->undo_marker = 0;
	tp->undo_retrans = 0;
}

/* Enter Loss state. If "how" is not zero, forget all SACK information
 * and reset tags completely, otherwise preserve SACKs. If receiver
 * dropped its ofo queue, we will know this due to reneging detection.
 */
void tcp_enter_loss(struct sock *sk, int how)
{
1391
	const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
1392 1393 1394 1395 1396
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb;
	int cnt = 0;

	/* Reduce ssthresh if it has not yet been made inside this window. */
1397 1398 1399 1400 1401
	if (icsk->icsk_ca_state <= TCP_CA_Disorder || tp->snd_una == tp->high_seq ||
	    (icsk->icsk_ca_state == TCP_CA_Loss && !icsk->icsk_retransmits)) {
		tp->prior_ssthresh = tcp_current_ssthresh(sk);
		tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
		tcp_ca_event(sk, CA_EVENT_LOSS);
L
Linus Torvalds 已提交
1402 1403 1404 1405 1406
	}
	tp->snd_cwnd	   = 1;
	tp->snd_cwnd_cnt   = 0;
	tp->snd_cwnd_stamp = tcp_time_stamp;

1407
	tp->bytes_acked = 0;
L
Linus Torvalds 已提交
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
	tcp_clear_retrans(tp);

	/* Push undo marker, if it was plain RTO and nothing
	 * was retransmitted. */
	if (!how)
		tp->undo_marker = tp->snd_una;

	sk_stream_for_retrans_queue(skb, sk) {
		cnt += tcp_skb_pcount(skb);
		if (TCP_SKB_CB(skb)->sacked&TCPCB_RETRANS)
			tp->undo_marker = 0;
		TCP_SKB_CB(skb)->sacked &= (~TCPCB_TAGBITS)|TCPCB_SACKED_ACKED;
		if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED) || how) {
			TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_ACKED;
			TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
			tp->lost_out += tcp_skb_pcount(skb);
		} else {
			tp->sacked_out += tcp_skb_pcount(skb);
			tp->fackets_out = cnt;
		}
	}
	tcp_sync_left_out(tp);

	tp->reordering = min_t(unsigned int, tp->reordering,
					     sysctl_tcp_reordering);
1433
	tcp_set_ca_state(sk, TCP_CA_Loss);
L
Linus Torvalds 已提交
1434 1435
	tp->high_seq = tp->snd_nxt;
	TCP_ECN_queue_cwr(tp);
1436 1437

	clear_all_retrans_hints(tp);
L
Linus Torvalds 已提交
1438 1439
}

1440
static int tcp_check_sack_reneging(struct sock *sk)
L
Linus Torvalds 已提交
1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451
{
	struct sk_buff *skb;

	/* If ACK arrived pointing to a remembered SACK,
	 * it means that our remembered SACKs do not reflect
	 * real state of receiver i.e.
	 * receiver _host_ is heavily congested (or buggy).
	 * Do processing similar to RTO timeout.
	 */
	if ((skb = skb_peek(&sk->sk_write_queue)) != NULL &&
	    (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) {
1452
		struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
1453 1454 1455
		NET_INC_STATS_BH(LINUX_MIB_TCPSACKRENEGING);

		tcp_enter_loss(sk, 1);
1456
		icsk->icsk_retransmits++;
L
Linus Torvalds 已提交
1457
		tcp_retransmit_skb(sk, skb_peek(&sk->sk_write_queue));
1458
		inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
1459
					  icsk->icsk_rto, TCP_RTO_MAX);
L
Linus Torvalds 已提交
1460 1461 1462 1463 1464 1465 1466 1467 1468 1469
		return 1;
	}
	return 0;
}

static inline int tcp_fackets_out(struct tcp_sock *tp)
{
	return IsReno(tp) ? tp->sacked_out+1 : tp->fackets_out;
}

1470
static inline int tcp_skb_timedout(struct sock *sk, struct sk_buff *skb)
L
Linus Torvalds 已提交
1471
{
1472
	return (tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto);
L
Linus Torvalds 已提交
1473 1474 1475 1476 1477
}

static inline int tcp_head_timedout(struct sock *sk, struct tcp_sock *tp)
{
	return tp->packets_out &&
1478
	       tcp_skb_timedout(sk, skb_peek(&sk->sk_write_queue));
L
Linus Torvalds 已提交
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 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 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
}

/* Linux NewReno/SACK/FACK/ECN state machine.
 * --------------------------------------
 *
 * "Open"	Normal state, no dubious events, fast path.
 * "Disorder"   In all the respects it is "Open",
 *		but requires a bit more attention. It is entered when
 *		we see some SACKs or dupacks. It is split of "Open"
 *		mainly to move some processing from fast path to slow one.
 * "CWR"	CWND was reduced due to some Congestion Notification event.
 *		It can be ECN, ICMP source quench, local device congestion.
 * "Recovery"	CWND was reduced, we are fast-retransmitting.
 * "Loss"	CWND was reduced due to RTO timeout or SACK reneging.
 *
 * tcp_fastretrans_alert() is entered:
 * - each incoming ACK, if state is not "Open"
 * - when arrived ACK is unusual, namely:
 *	* SACK
 *	* Duplicate ACK.
 *	* ECN ECE.
 *
 * Counting packets in flight is pretty simple.
 *
 *	in_flight = packets_out - left_out + retrans_out
 *
 *	packets_out is SND.NXT-SND.UNA counted in packets.
 *
 *	retrans_out is number of retransmitted segments.
 *
 *	left_out is number of segments left network, but not ACKed yet.
 *
 *		left_out = sacked_out + lost_out
 *
 *     sacked_out: Packets, which arrived to receiver out of order
 *		   and hence not ACKed. With SACKs this number is simply
 *		   amount of SACKed data. Even without SACKs
 *		   it is easy to give pretty reliable estimate of this number,
 *		   counting duplicate ACKs.
 *
 *       lost_out: Packets lost by network. TCP has no explicit
 *		   "loss notification" feedback from network (for now).
 *		   It means that this number can be only _guessed_.
 *		   Actually, it is the heuristics to predict lossage that
 *		   distinguishes different algorithms.
 *
 *	F.e. after RTO, when all the queue is considered as lost,
 *	lost_out = packets_out and in_flight = retrans_out.
 *
 *		Essentially, we have now two algorithms counting
 *		lost packets.
 *
 *		FACK: It is the simplest heuristics. As soon as we decided
 *		that something is lost, we decide that _all_ not SACKed
 *		packets until the most forward SACK are lost. I.e.
 *		lost_out = fackets_out - sacked_out and left_out = fackets_out.
 *		It is absolutely correct estimate, if network does not reorder
 *		packets. And it loses any connection to reality when reordering
 *		takes place. We use FACK by default until reordering
 *		is suspected on the path to this destination.
 *
 *		NewReno: when Recovery is entered, we assume that one segment
 *		is lost (classic Reno). While we are in Recovery and
 *		a partial ACK arrives, we assume that one more packet
 *		is lost (NewReno). This heuristics are the same in NewReno
 *		and SACK.
 *
 *  Imagine, that's all! Forget about all this shamanism about CWND inflation
 *  deflation etc. CWND is real congestion window, never inflated, changes
 *  only according to classic VJ rules.
 *
 * Really tricky (and requiring careful tuning) part of algorithm
 * is hidden in functions tcp_time_to_recover() and tcp_xmit_retransmit_queue().
 * The first determines the moment _when_ we should reduce CWND and,
 * hence, slow down forward transmission. In fact, it determines the moment
 * when we decide that hole is caused by loss, rather than by a reorder.
 *
 * tcp_xmit_retransmit_queue() decides, _what_ we should retransmit to fill
 * holes, caused by lost packets.
 *
 * And the most logically complicated part of algorithm is undo
 * heuristics. We detect false retransmits due to both too early
 * fast retransmit (reordering) and underestimated RTO, analyzing
 * timestamps and D-SACKs. When we detect that some segments were
 * retransmitted by mistake and CWND reduction was wrong, we undo
 * window reduction and abort recovery phase. This logic is hidden
 * inside several functions named tcp_try_undo_<something>.
 */

/* This function decides, when we should leave Disordered state
 * and enter Recovery phase, reducing congestion window.
 *
 * Main question: may we further continue forward transmission
 * with the same cwnd?
 */
static int tcp_time_to_recover(struct sock *sk, struct tcp_sock *tp)
{
	__u32 packets_out;

1578 1579 1580 1581
	/* Do not perform any recovery during FRTO algorithm */
	if (tp->frto_counter)
		return 0;

L
Linus Torvalds 已提交
1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615
	/* Trick#1: The loss is proven. */
	if (tp->lost_out)
		return 1;

	/* Not-A-Trick#2 : Classic rule... */
	if (tcp_fackets_out(tp) > tp->reordering)
		return 1;

	/* Trick#3 : when we use RFC2988 timer restart, fast
	 * retransmit can be triggered by timeout of queue head.
	 */
	if (tcp_head_timedout(sk, tp))
		return 1;

	/* Trick#4: It is still not OK... But will it be useful to delay
	 * recovery more?
	 */
	packets_out = tp->packets_out;
	if (packets_out <= tp->reordering &&
	    tp->sacked_out >= max_t(__u32, packets_out/2, sysctl_tcp_reordering) &&
	    !tcp_may_send_now(sk, tp)) {
		/* We have nothing to send. This connection is limited
		 * either by receiver window or by application.
		 */
		return 1;
	}

	return 0;
}

/* If we receive more dupacks than we expected counting segments
 * in assumption of absent reordering, interpret this as reordering.
 * The only another reason could be bug in receiver TCP.
 */
1616
static void tcp_check_reno_reordering(struct sock *sk, const int addend)
L
Linus Torvalds 已提交
1617
{
1618
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
1619 1620 1621 1622 1623 1624 1625
	u32 holes;

	holes = max(tp->lost_out, 1U);
	holes = min(holes, tp->packets_out);

	if ((tp->sacked_out + holes) > tp->packets_out) {
		tp->sacked_out = tp->packets_out - holes;
1626
		tcp_update_reordering(sk, tp->packets_out + addend, 0);
L
Linus Torvalds 已提交
1627 1628 1629 1630 1631
	}
}

/* Emulate SACKs for SACKless connection: account for a new dupack. */

1632
static void tcp_add_reno_sack(struct sock *sk)
L
Linus Torvalds 已提交
1633
{
1634
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
1635
	tp->sacked_out++;
1636
	tcp_check_reno_reordering(sk, 0);
L
Linus Torvalds 已提交
1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650
	tcp_sync_left_out(tp);
}

/* Account for ACK, ACKing some data in Reno Recovery phase. */

static void tcp_remove_reno_sacks(struct sock *sk, struct tcp_sock *tp, int acked)
{
	if (acked > 0) {
		/* One ACK acked hole. The rest eat duplicate ACKs. */
		if (acked-1 >= tp->sacked_out)
			tp->sacked_out = 0;
		else
			tp->sacked_out -= acked-1;
	}
1651
	tcp_check_reno_reordering(sk, acked);
L
Linus Torvalds 已提交
1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665
	tcp_sync_left_out(tp);
}

static inline void tcp_reset_reno_sack(struct tcp_sock *tp)
{
	tp->sacked_out = 0;
	tp->left_out = tp->lost_out;
}

/* Mark head of queue up as lost. */
static void tcp_mark_head_lost(struct sock *sk, struct tcp_sock *tp,
			       int packets, u32 high_seq)
{
	struct sk_buff *skb;
1666
	int cnt;
L
Linus Torvalds 已提交
1667

1668 1669 1670 1671 1672 1673 1674 1675
	BUG_TRAP(packets <= tp->packets_out);
	if (tp->lost_skb_hint) {
		skb = tp->lost_skb_hint;
		cnt = tp->lost_cnt_hint;
	} else {
		skb = sk->sk_write_queue.next;
		cnt = 0;
	}
L
Linus Torvalds 已提交
1676

1677 1678 1679 1680 1681 1682 1683
	sk_stream_for_retrans_queue_from(skb, sk) {
		/* TODO: do this better */
		/* this is not the most efficient way to do this... */
		tp->lost_skb_hint = skb;
		tp->lost_cnt_hint = cnt;
		cnt += tcp_skb_pcount(skb);
		if (cnt > packets || after(TCP_SKB_CB(skb)->end_seq, high_seq))
L
Linus Torvalds 已提交
1684 1685 1686 1687
			break;
		if (!(TCP_SKB_CB(skb)->sacked&TCPCB_TAGBITS)) {
			TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
			tp->lost_out += tcp_skb_pcount(skb);
1688 1689 1690 1691 1692 1693 1694 1695 1696

			/* clear xmit_retransmit_queue hints
			 *  if this is beyond hint */
			if(tp->retransmit_skb_hint != NULL &&
			   before(TCP_SKB_CB(skb)->seq,
				  TCP_SKB_CB(tp->retransmit_skb_hint)->seq)) {

				tp->retransmit_skb_hint = NULL;
			}
L
Linus Torvalds 已提交
1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719
		}
	}
	tcp_sync_left_out(tp);
}

/* Account newly detected lost packet(s) */

static void tcp_update_scoreboard(struct sock *sk, struct tcp_sock *tp)
{
	if (IsFack(tp)) {
		int lost = tp->fackets_out - tp->reordering;
		if (lost <= 0)
			lost = 1;
		tcp_mark_head_lost(sk, tp, lost, tp->high_seq);
	} else {
		tcp_mark_head_lost(sk, tp, 1, tp->high_seq);
	}

	/* New heuristics: it is possible only after we switched
	 * to restart timer each time when something is ACKed.
	 * Hence, we can detect timed out packets during fast
	 * retransmit without falling to slow start.
	 */
1720
	if (!IsReno(tp) && tcp_head_timedout(sk, tp)) {
L
Linus Torvalds 已提交
1721 1722
		struct sk_buff *skb;

1723 1724 1725 1726 1727 1728 1729 1730
		skb = tp->scoreboard_skb_hint ? tp->scoreboard_skb_hint
			: sk->sk_write_queue.next;

		sk_stream_for_retrans_queue_from(skb, sk) {
			if (!tcp_skb_timedout(sk, skb))
				break;

			if (!(TCP_SKB_CB(skb)->sacked&TCPCB_TAGBITS)) {
L
Linus Torvalds 已提交
1731 1732
				TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
				tp->lost_out += tcp_skb_pcount(skb);
1733 1734 1735 1736 1737 1738 1739

				/* clear xmit_retrans hint */
				if (tp->retransmit_skb_hint &&
				    before(TCP_SKB_CB(skb)->seq,
					   TCP_SKB_CB(tp->retransmit_skb_hint)->seq))

					tp->retransmit_skb_hint = NULL;
L
Linus Torvalds 已提交
1740 1741
			}
		}
1742 1743 1744

		tp->scoreboard_skb_hint = skb;

L
Linus Torvalds 已提交
1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
		tcp_sync_left_out(tp);
	}
}

/* CWND moderation, preventing bursts due to too big ACKs
 * in dubious situations.
 */
static inline void tcp_moderate_cwnd(struct tcp_sock *tp)
{
	tp->snd_cwnd = min(tp->snd_cwnd,
			   tcp_packets_in_flight(tp)+tcp_max_burst(tp));
	tp->snd_cwnd_stamp = tcp_time_stamp;
}

1759 1760 1761 1762 1763 1764 1765 1766 1767 1768
/* Lower bound on congestion window is slow start threshold
 * unless congestion avoidance choice decides to overide it.
 */
static inline u32 tcp_cwnd_min(const struct sock *sk)
{
	const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops;

	return ca_ops->min_cwnd ? ca_ops->min_cwnd(sk) : tcp_sk(sk)->snd_ssthresh;
}

L
Linus Torvalds 已提交
1769
/* Decrease cwnd each second ack. */
1770
static void tcp_cwnd_down(struct sock *sk)
L
Linus Torvalds 已提交
1771
{
1772
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
1773 1774 1775 1776 1777
	int decr = tp->snd_cwnd_cnt + 1;

	tp->snd_cwnd_cnt = decr&1;
	decr >>= 1;

1778
	if (decr && tp->snd_cwnd > tcp_cwnd_min(sk))
L
Linus Torvalds 已提交
1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811
		tp->snd_cwnd -= decr;

	tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp)+1);
	tp->snd_cwnd_stamp = tcp_time_stamp;
}

/* Nothing was retransmitted or returned timestamp is less
 * than timestamp of the first retransmission.
 */
static inline int tcp_packet_delayed(struct tcp_sock *tp)
{
	return !tp->retrans_stamp ||
		(tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr &&
		 (__s32)(tp->rx_opt.rcv_tsecr - tp->retrans_stamp) < 0);
}

/* Undo procedures. */

#if FASTRETRANS_DEBUG > 1
static void DBGUNDO(struct sock *sk, struct tcp_sock *tp, const char *msg)
{
	struct inet_sock *inet = inet_sk(sk);
	printk(KERN_DEBUG "Undo %s %u.%u.%u.%u/%u c%u l%u ss%u/%u p%u\n",
	       msg,
	       NIPQUAD(inet->daddr), ntohs(inet->dport),
	       tp->snd_cwnd, tp->left_out,
	       tp->snd_ssthresh, tp->prior_ssthresh,
	       tp->packets_out);
}
#else
#define DBGUNDO(x...) do { } while (0)
#endif

1812
static void tcp_undo_cwr(struct sock *sk, const int undo)
L
Linus Torvalds 已提交
1813
{
1814 1815
	struct tcp_sock *tp = tcp_sk(sk);

L
Linus Torvalds 已提交
1816
	if (tp->prior_ssthresh) {
1817 1818 1819 1820
		const struct inet_connection_sock *icsk = inet_csk(sk);

		if (icsk->icsk_ca_ops->undo_cwnd)
			tp->snd_cwnd = icsk->icsk_ca_ops->undo_cwnd(sk);
L
Linus Torvalds 已提交
1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832
		else
			tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh<<1);

		if (undo && tp->prior_ssthresh > tp->snd_ssthresh) {
			tp->snd_ssthresh = tp->prior_ssthresh;
			TCP_ECN_withdraw_cwr(tp);
		}
	} else {
		tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh);
	}
	tcp_moderate_cwnd(tp);
	tp->snd_cwnd_stamp = tcp_time_stamp;
1833 1834 1835 1836

	/* There is something screwy going on with the retrans hints after
	   an undo */
	clear_all_retrans_hints(tp);
L
Linus Torvalds 已提交
1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851
}

static inline int tcp_may_undo(struct tcp_sock *tp)
{
	return tp->undo_marker &&
		(!tp->undo_retrans || tcp_packet_delayed(tp));
}

/* People celebrate: "We love our President!" */
static int tcp_try_undo_recovery(struct sock *sk, struct tcp_sock *tp)
{
	if (tcp_may_undo(tp)) {
		/* Happy end! We did not retransmit anything
		 * or our original transmission succeeded.
		 */
1852 1853 1854
		DBGUNDO(sk, tp, inet_csk(sk)->icsk_ca_state == TCP_CA_Loss ? "loss" : "retrans");
		tcp_undo_cwr(sk, 1);
		if (inet_csk(sk)->icsk_ca_state == TCP_CA_Loss)
L
Linus Torvalds 已提交
1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866
			NET_INC_STATS_BH(LINUX_MIB_TCPLOSSUNDO);
		else
			NET_INC_STATS_BH(LINUX_MIB_TCPFULLUNDO);
		tp->undo_marker = 0;
	}
	if (tp->snd_una == tp->high_seq && IsReno(tp)) {
		/* Hold old state until something *above* high_seq
		 * is ACKed. For Reno it is MUST to prevent false
		 * fast retransmits (RFC2582). SACK TCP is safe. */
		tcp_moderate_cwnd(tp);
		return 1;
	}
1867
	tcp_set_ca_state(sk, TCP_CA_Open);
L
Linus Torvalds 已提交
1868 1869 1870 1871 1872 1873 1874 1875
	return 0;
}

/* Try to undo cwnd reduction, because D-SACKs acked all retransmitted data */
static void tcp_try_undo_dsack(struct sock *sk, struct tcp_sock *tp)
{
	if (tp->undo_marker && !tp->undo_retrans) {
		DBGUNDO(sk, tp, "D-SACK");
1876
		tcp_undo_cwr(sk, 1);
L
Linus Torvalds 已提交
1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896
		tp->undo_marker = 0;
		NET_INC_STATS_BH(LINUX_MIB_TCPDSACKUNDO);
	}
}

/* Undo during fast recovery after partial ACK. */

static int tcp_try_undo_partial(struct sock *sk, struct tcp_sock *tp,
				int acked)
{
	/* Partial ACK arrived. Force Hoe's retransmit. */
	int failed = IsReno(tp) || tp->fackets_out>tp->reordering;

	if (tcp_may_undo(tp)) {
		/* Plain luck! Hole if filled with delayed
		 * packet, rather than with a retransmit.
		 */
		if (tp->retrans_out == 0)
			tp->retrans_stamp = 0;

1897
		tcp_update_reordering(sk, tcp_fackets_out(tp) + acked, 1);
L
Linus Torvalds 已提交
1898 1899

		DBGUNDO(sk, tp, "Hoe");
1900
		tcp_undo_cwr(sk, 0);
L
Linus Torvalds 已提交
1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919
		NET_INC_STATS_BH(LINUX_MIB_TCPPARTIALUNDO);

		/* So... Do not make Hoe's retransmit yet.
		 * If the first packet was delayed, the rest
		 * ones are most probably delayed as well.
		 */
		failed = 0;
	}
	return failed;
}

/* Undo during loss recovery after partial ACK. */
static int tcp_try_undo_loss(struct sock *sk, struct tcp_sock *tp)
{
	if (tcp_may_undo(tp)) {
		struct sk_buff *skb;
		sk_stream_for_retrans_queue(skb, sk) {
			TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
		}
1920 1921 1922

		clear_all_retrans_hints(tp);

L
Linus Torvalds 已提交
1923 1924 1925
		DBGUNDO(sk, tp, "partial loss");
		tp->lost_out = 0;
		tp->left_out = tp->sacked_out;
1926
		tcp_undo_cwr(sk, 1);
L
Linus Torvalds 已提交
1927
		NET_INC_STATS_BH(LINUX_MIB_TCPLOSSUNDO);
1928
		inet_csk(sk)->icsk_retransmits = 0;
L
Linus Torvalds 已提交
1929 1930
		tp->undo_marker = 0;
		if (!IsReno(tp))
1931
			tcp_set_ca_state(sk, TCP_CA_Open);
L
Linus Torvalds 已提交
1932 1933 1934 1935 1936
		return 1;
	}
	return 0;
}

1937
static inline void tcp_complete_cwr(struct sock *sk)
L
Linus Torvalds 已提交
1938
{
1939
	struct tcp_sock *tp = tcp_sk(sk);
1940
	tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh);
L
Linus Torvalds 已提交
1941
	tp->snd_cwnd_stamp = tcp_time_stamp;
1942
	tcp_ca_event(sk, CA_EVENT_COMPLETE_CWR);
L
Linus Torvalds 已提交
1943 1944 1945 1946 1947 1948 1949 1950 1951 1952
}

static void tcp_try_to_open(struct sock *sk, struct tcp_sock *tp, int flag)
{
	tp->left_out = tp->sacked_out;

	if (tp->retrans_out == 0)
		tp->retrans_stamp = 0;

	if (flag&FLAG_ECE)
1953
		tcp_enter_cwr(sk);
L
Linus Torvalds 已提交
1954

1955
	if (inet_csk(sk)->icsk_ca_state != TCP_CA_CWR) {
L
Linus Torvalds 已提交
1956 1957 1958 1959 1960
		int state = TCP_CA_Open;

		if (tp->left_out || tp->retrans_out || tp->undo_marker)
			state = TCP_CA_Disorder;

1961 1962
		if (inet_csk(sk)->icsk_ca_state != state) {
			tcp_set_ca_state(sk, state);
L
Linus Torvalds 已提交
1963 1964 1965 1966
			tp->high_seq = tp->snd_nxt;
		}
		tcp_moderate_cwnd(tp);
	} else {
1967
		tcp_cwnd_down(sk);
L
Linus Torvalds 已提交
1968 1969 1970
	}
}

J
John Heffner 已提交
1971 1972 1973 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
static void tcp_mtup_probe_failed(struct sock *sk)
{
	struct inet_connection_sock *icsk = inet_csk(sk);

	icsk->icsk_mtup.search_high = icsk->icsk_mtup.probe_size - 1;
	icsk->icsk_mtup.probe_size = 0;
}

static void tcp_mtup_probe_success(struct sock *sk, struct sk_buff *skb)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct inet_connection_sock *icsk = inet_csk(sk);

	/* FIXME: breaks with very large cwnd */
	tp->prior_ssthresh = tcp_current_ssthresh(sk);
	tp->snd_cwnd = tp->snd_cwnd *
		       tcp_mss_to_mtu(sk, tp->mss_cache) /
		       icsk->icsk_mtup.probe_size;
	tp->snd_cwnd_cnt = 0;
	tp->snd_cwnd_stamp = tcp_time_stamp;
	tp->rcv_ssthresh = tcp_current_ssthresh(sk);

	icsk->icsk_mtup.search_low = icsk->icsk_mtup.probe_size;
	icsk->icsk_mtup.probe_size = 0;
	tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
}


L
Linus Torvalds 已提交
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
/* Process an event, which can update packets-in-flight not trivially.
 * Main goal of this function is to calculate new estimate for left_out,
 * taking into account both packets sitting in receiver's buffer and
 * packets lost by network.
 *
 * Besides that it does CWND reduction, when packet loss is detected
 * and changes state of machine.
 *
 * It does _not_ decide what to send, it is made in function
 * tcp_xmit_retransmit_queue().
 */
static void
tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una,
		      int prior_packets, int flag)
{
2014
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2015 2016 2017 2018 2019 2020 2021
	struct tcp_sock *tp = tcp_sk(sk);
	int is_dupack = (tp->snd_una == prior_snd_una && !(flag&FLAG_NOT_DUP));

	/* Some technical things:
	 * 1. Reno does not count dupacks (sacked_out) automatically. */
	if (!tp->packets_out)
		tp->sacked_out = 0;
2022
	/* 2. SACK counts snd_fack in packets inaccurately. */
L
Linus Torvalds 已提交
2023 2024 2025
	if (tp->sacked_out == 0)
		tp->fackets_out = 0;

2026
	/* Now state machine starts.
L
Linus Torvalds 已提交
2027 2028 2029 2030 2031
	 * A. ECE, hence prohibit cwnd undoing, the reduction is required. */
	if (flag&FLAG_ECE)
		tp->prior_ssthresh = 0;

	/* B. In all the states check for reneging SACKs. */
2032
	if (tp->sacked_out && tcp_check_sack_reneging(sk))
L
Linus Torvalds 已提交
2033 2034 2035 2036 2037
		return;

	/* C. Process data loss notification, provided it is valid. */
	if ((flag&FLAG_DATA_LOST) &&
	    before(tp->snd_una, tp->high_seq) &&
2038
	    icsk->icsk_ca_state != TCP_CA_Open &&
L
Linus Torvalds 已提交
2039 2040 2041 2042 2043 2044 2045 2046 2047 2048
	    tp->fackets_out > tp->reordering) {
		tcp_mark_head_lost(sk, tp, tp->fackets_out-tp->reordering, tp->high_seq);
		NET_INC_STATS_BH(LINUX_MIB_TCPLOSS);
	}

	/* D. Synchronize left_out to current state. */
	tcp_sync_left_out(tp);

	/* E. Check state exit conditions. State can be terminated
	 *    when high_seq is ACKed. */
2049
	if (icsk->icsk_ca_state == TCP_CA_Open) {
2050
		BUG_TRAP(tp->retrans_out == 0);
L
Linus Torvalds 已提交
2051 2052
		tp->retrans_stamp = 0;
	} else if (!before(tp->snd_una, tp->high_seq)) {
2053
		switch (icsk->icsk_ca_state) {
L
Linus Torvalds 已提交
2054
		case TCP_CA_Loss:
2055
			icsk->icsk_retransmits = 0;
L
Linus Torvalds 已提交
2056 2057 2058 2059 2060 2061 2062 2063
			if (tcp_try_undo_recovery(sk, tp))
				return;
			break;

		case TCP_CA_CWR:
			/* CWR is to be held something *above* high_seq
			 * is ACKed for CWR bit to reach receiver. */
			if (tp->snd_una != tp->high_seq) {
2064 2065
				tcp_complete_cwr(sk);
				tcp_set_ca_state(sk, TCP_CA_Open);
L
Linus Torvalds 已提交
2066 2067 2068 2069 2070 2071 2072 2073 2074 2075
			}
			break;

		case TCP_CA_Disorder:
			tcp_try_undo_dsack(sk, tp);
			if (!tp->undo_marker ||
			    /* For SACK case do not Open to allow to undo
			     * catching for all duplicate ACKs. */
			    IsReno(tp) || tp->snd_una != tp->high_seq) {
				tp->undo_marker = 0;
2076
				tcp_set_ca_state(sk, TCP_CA_Open);
L
Linus Torvalds 已提交
2077 2078 2079 2080 2081 2082 2083 2084
			}
			break;

		case TCP_CA_Recovery:
			if (IsReno(tp))
				tcp_reset_reno_sack(tp);
			if (tcp_try_undo_recovery(sk, tp))
				return;
2085
			tcp_complete_cwr(sk);
L
Linus Torvalds 已提交
2086 2087 2088 2089 2090
			break;
		}
	}

	/* F. Process state. */
2091
	switch (icsk->icsk_ca_state) {
L
Linus Torvalds 已提交
2092 2093 2094
	case TCP_CA_Recovery:
		if (prior_snd_una == tp->snd_una) {
			if (IsReno(tp) && is_dupack)
2095
				tcp_add_reno_sack(sk);
L
Linus Torvalds 已提交
2096 2097 2098 2099 2100 2101 2102 2103 2104
		} else {
			int acked = prior_packets - tp->packets_out;
			if (IsReno(tp))
				tcp_remove_reno_sacks(sk, tp, acked);
			is_dupack = tcp_try_undo_partial(sk, tp, acked);
		}
		break;
	case TCP_CA_Loss:
		if (flag&FLAG_DATA_ACKED)
2105
			icsk->icsk_retransmits = 0;
L
Linus Torvalds 已提交
2106 2107 2108 2109 2110
		if (!tcp_try_undo_loss(sk, tp)) {
			tcp_moderate_cwnd(tp);
			tcp_xmit_retransmit_queue(sk);
			return;
		}
2111
		if (icsk->icsk_ca_state != TCP_CA_Open)
L
Linus Torvalds 已提交
2112 2113 2114 2115 2116 2117 2118
			return;
		/* Loss is undone; fall through to processing in Open state. */
	default:
		if (IsReno(tp)) {
			if (tp->snd_una != prior_snd_una)
				tcp_reset_reno_sack(tp);
			if (is_dupack)
2119
				tcp_add_reno_sack(sk);
L
Linus Torvalds 已提交
2120 2121
		}

2122
		if (icsk->icsk_ca_state == TCP_CA_Disorder)
L
Linus Torvalds 已提交
2123 2124 2125 2126 2127 2128 2129
			tcp_try_undo_dsack(sk, tp);

		if (!tcp_time_to_recover(sk, tp)) {
			tcp_try_to_open(sk, tp, flag);
			return;
		}

J
John Heffner 已提交
2130 2131 2132
		/* MTU probe failure: don't reduce cwnd */
		if (icsk->icsk_ca_state < TCP_CA_CWR &&
		    icsk->icsk_mtup.probe_size &&
2133
		    tp->snd_una == tp->mtu_probe.probe_seq_start) {
J
John Heffner 已提交
2134 2135 2136 2137 2138 2139 2140
			tcp_mtup_probe_failed(sk);
			/* Restores the reduction we did in tcp_mtup_probe() */
			tp->snd_cwnd++;
			tcp_simple_retransmit(sk);
			return;
		}

L
Linus Torvalds 已提交
2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152
		/* Otherwise enter Recovery state */

		if (IsReno(tp))
			NET_INC_STATS_BH(LINUX_MIB_TCPRENORECOVERY);
		else
			NET_INC_STATS_BH(LINUX_MIB_TCPSACKRECOVERY);

		tp->high_seq = tp->snd_nxt;
		tp->prior_ssthresh = 0;
		tp->undo_marker = tp->snd_una;
		tp->undo_retrans = tp->retrans_out;

2153
		if (icsk->icsk_ca_state < TCP_CA_CWR) {
L
Linus Torvalds 已提交
2154
			if (!(flag&FLAG_ECE))
2155 2156
				tp->prior_ssthresh = tcp_current_ssthresh(sk);
			tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
L
Linus Torvalds 已提交
2157 2158 2159
			TCP_ECN_queue_cwr(tp);
		}

2160
		tp->bytes_acked = 0;
L
Linus Torvalds 已提交
2161
		tp->snd_cwnd_cnt = 0;
2162
		tcp_set_ca_state(sk, TCP_CA_Recovery);
L
Linus Torvalds 已提交
2163 2164 2165 2166
	}

	if (is_dupack || tcp_head_timedout(sk, tp))
		tcp_update_scoreboard(sk, tp);
2167
	tcp_cwnd_down(sk);
L
Linus Torvalds 已提交
2168 2169 2170 2171
	tcp_xmit_retransmit_queue(sk);
}

/* Read draft-ietf-tcplw-high-performance before mucking
S
Stephen Hemminger 已提交
2172
 * with this code. (Supersedes RFC1323)
L
Linus Torvalds 已提交
2173
 */
2174
static void tcp_ack_saw_tstamp(struct sock *sk, int flag)
L
Linus Torvalds 已提交
2175 2176 2177 2178 2179 2180 2181 2182 2183 2184
{
	/* RTTM Rule: A TSecr value received in a segment is used to
	 * update the averaged RTT measurement only if the segment
	 * acknowledges some new data, i.e., only if it advances the
	 * left edge of the send window.
	 *
	 * See draft-ietf-tcplw-high-performance-00, section 3.3.
	 * 1998/04/10 Andrey V. Savochkin <saw@msu.ru>
	 *
	 * Changed: reset backoff as soon as we see the first valid sample.
S
Stephen Hemminger 已提交
2185
	 * If we do not, we get strongly overestimated rto. With timestamps
L
Linus Torvalds 已提交
2186 2187 2188 2189 2190
	 * samples are accepted even from very old segments: f.e., when rtt=1
	 * increases to 8, we retransmit 5 times and after 8 seconds delayed
	 * answer arrives rto becomes 120 seconds! If at least one of segments
	 * in window is lost... Voila.	 			--ANK (010210)
	 */
2191 2192
	struct tcp_sock *tp = tcp_sk(sk);
	const __u32 seq_rtt = tcp_time_stamp - tp->rx_opt.rcv_tsecr;
2193
	tcp_rtt_estimator(sk, seq_rtt);
2194 2195 2196
	tcp_set_rto(sk);
	inet_csk(sk)->icsk_backoff = 0;
	tcp_bound_rto(sk);
L
Linus Torvalds 已提交
2197 2198
}

2199
static void tcp_ack_no_tstamp(struct sock *sk, u32 seq_rtt, int flag)
L
Linus Torvalds 已提交
2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212
{
	/* We don't have a timestamp. Can only use
	 * packets that are not retransmitted to determine
	 * rtt estimates. Also, we must not reset the
	 * backoff for rto until we get a non-retransmitted
	 * packet. This allows us to deal with a situation
	 * where the network delay has increased suddenly.
	 * I.e. Karn's algorithm. (SIGCOMM '87, p5.)
	 */

	if (flag & FLAG_RETRANS_DATA_ACKED)
		return;

2213
	tcp_rtt_estimator(sk, seq_rtt);
2214 2215 2216
	tcp_set_rto(sk);
	inet_csk(sk)->icsk_backoff = 0;
	tcp_bound_rto(sk);
L
Linus Torvalds 已提交
2217 2218
}

2219
static inline void tcp_ack_update_rtt(struct sock *sk, const int flag,
2220
				      const s32 seq_rtt)
L
Linus Torvalds 已提交
2221
{
2222
	const struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
2223 2224
	/* Note that peer MAY send zero echo. In this case it is ignored. (rfc1323) */
	if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr)
2225
		tcp_ack_saw_tstamp(sk, flag);
L
Linus Torvalds 已提交
2226
	else if (seq_rtt >= 0)
2227
		tcp_ack_no_tstamp(sk, seq_rtt, flag);
L
Linus Torvalds 已提交
2228 2229
}

S
Stephen Hemminger 已提交
2230 2231
static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 rtt,
			   u32 in_flight, int good)
L
Linus Torvalds 已提交
2232
{
2233 2234 2235
	const struct inet_connection_sock *icsk = inet_csk(sk);
	icsk->icsk_ca_ops->cong_avoid(sk, ack, rtt, in_flight, good);
	tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp;
L
Linus Torvalds 已提交
2236 2237 2238 2239 2240 2241
}

/* Restart timer after forward progress on connection.
 * RFC2988 recommends to restart timer to now+rto.
 */

S
Stephen Hemminger 已提交
2242
static void tcp_ack_packets_out(struct sock *sk, struct tcp_sock *tp)
L
Linus Torvalds 已提交
2243 2244
{
	if (!tp->packets_out) {
2245
		inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS);
L
Linus Torvalds 已提交
2246
	} else {
2247
		inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, inet_csk(sk)->icsk_rto, TCP_RTO_MAX);
L
Linus Torvalds 已提交
2248 2249 2250 2251 2252 2253 2254
	}
}

static int tcp_tso_acked(struct sock *sk, struct sk_buff *skb,
			 __u32 now, __s32 *seq_rtt)
{
	struct tcp_sock *tp = tcp_sk(sk);
2255
	struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
L
Linus Torvalds 已提交
2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306
	__u32 seq = tp->snd_una;
	__u32 packets_acked;
	int acked = 0;

	/* If we get here, the whole TSO packet has not been
	 * acked.
	 */
	BUG_ON(!after(scb->end_seq, seq));

	packets_acked = tcp_skb_pcount(skb);
	if (tcp_trim_head(sk, skb, seq - scb->seq))
		return 0;
	packets_acked -= tcp_skb_pcount(skb);

	if (packets_acked) {
		__u8 sacked = scb->sacked;

		acked |= FLAG_DATA_ACKED;
		if (sacked) {
			if (sacked & TCPCB_RETRANS) {
				if (sacked & TCPCB_SACKED_RETRANS)
					tp->retrans_out -= packets_acked;
				acked |= FLAG_RETRANS_DATA_ACKED;
				*seq_rtt = -1;
			} else if (*seq_rtt < 0)
				*seq_rtt = now - scb->when;
			if (sacked & TCPCB_SACKED_ACKED)
				tp->sacked_out -= packets_acked;
			if (sacked & TCPCB_LOST)
				tp->lost_out -= packets_acked;
			if (sacked & TCPCB_URG) {
				if (tp->urg_mode &&
				    !before(seq, tp->snd_up))
					tp->urg_mode = 0;
			}
		} else if (*seq_rtt < 0)
			*seq_rtt = now - scb->when;

		if (tp->fackets_out) {
			__u32 dval = min(tp->fackets_out, packets_acked);
			tp->fackets_out -= dval;
		}
		tp->packets_out -= packets_acked;

		BUG_ON(tcp_skb_pcount(skb) == 0);
		BUG_ON(!before(scb->seq, scb->end_seq));
	}

	return acked;
}

2307
static u32 tcp_usrtt(struct timeval *tv)
2308
{
2309
	struct timeval now;
2310 2311

	do_gettimeofday(&now);
2312
	return (now.tv_sec - tv->tv_sec) * 1000000 + (now.tv_usec - tv->tv_usec);
2313
}
L
Linus Torvalds 已提交
2314 2315

/* Remove acknowledged frames from the retransmission queue. */
2316
static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p)
L
Linus Torvalds 已提交
2317 2318
{
	struct tcp_sock *tp = tcp_sk(sk);
2319
	const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2320 2321 2322 2323
	struct sk_buff *skb;
	__u32 now = tcp_time_stamp;
	int acked = 0;
	__s32 seq_rtt = -1;
2324
	u32 pkts_acked = 0;
2325 2326
	void (*rtt_sample)(struct sock *sk, u32 usrtt)
		= icsk->icsk_ca_ops->rtt_sample;
2327
	struct timeval tv = { .tv_sec = 0, .tv_usec = 0 };
L
Linus Torvalds 已提交
2328 2329 2330

	while ((skb = skb_peek(&sk->sk_write_queue)) &&
	       skb != sk->sk_send_head) {
2331
		struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
L
Linus Torvalds 已提交
2332 2333 2334 2335 2336 2337 2338
		__u8 sacked = scb->sacked;

		/* If our packet is before the ack sequence we can
		 * discard it as it's confirmed to have arrived at
		 * the other end.
		 */
		if (after(scb->end_seq, tp->snd_una)) {
2339 2340
			if (tcp_skb_pcount(skb) > 1 &&
			    after(tp->snd_una, scb->seq))
L
Linus Torvalds 已提交
2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354
				acked |= tcp_tso_acked(sk, skb,
						       now, &seq_rtt);
			break;
		}

		/* Initial outgoing SYN's get put onto the write_queue
		 * just like anything else we transmit.  It is not
		 * true data, and if we misinform our callers that
		 * this ACK acks real data, we will erroneously exit
		 * connection startup slow start one packet too
		 * quickly.  This is severely frowned upon behavior.
		 */
		if (!(scb->flags & TCPCB_FLAG_SYN)) {
			acked |= FLAG_DATA_ACKED;
2355
			++pkts_acked;
L
Linus Torvalds 已提交
2356 2357 2358 2359 2360
		} else {
			acked |= FLAG_SYN_ACKED;
			tp->retrans_stamp = 0;
		}

J
John Heffner 已提交
2361 2362
		/* MTU probing checks */
		if (icsk->icsk_mtup.probe_size) {
2363
			if (!after(tp->mtu_probe.probe_seq_end, TCP_SKB_CB(skb)->end_seq)) {
J
John Heffner 已提交
2364 2365 2366 2367
				tcp_mtup_probe_success(sk, skb);
			}
		}

L
Linus Torvalds 已提交
2368 2369 2370 2371 2372 2373
		if (sacked) {
			if (sacked & TCPCB_RETRANS) {
				if(sacked & TCPCB_SACKED_RETRANS)
					tp->retrans_out -= tcp_skb_pcount(skb);
				acked |= FLAG_RETRANS_DATA_ACKED;
				seq_rtt = -1;
2374
			} else if (seq_rtt < 0) {
L
Linus Torvalds 已提交
2375
				seq_rtt = now - scb->when;
2376
				skb_get_timestamp(skb, &tv);
2377
			}
L
Linus Torvalds 已提交
2378 2379 2380 2381 2382 2383 2384 2385 2386
			if (sacked & TCPCB_SACKED_ACKED)
				tp->sacked_out -= tcp_skb_pcount(skb);
			if (sacked & TCPCB_LOST)
				tp->lost_out -= tcp_skb_pcount(skb);
			if (sacked & TCPCB_URG) {
				if (tp->urg_mode &&
				    !before(scb->end_seq, tp->snd_up))
					tp->urg_mode = 0;
			}
2387
		} else if (seq_rtt < 0) {
L
Linus Torvalds 已提交
2388
			seq_rtt = now - scb->when;
2389
			skb_get_timestamp(skb, &tv);
2390
		}
L
Linus Torvalds 已提交
2391 2392
		tcp_dec_pcount_approx(&tp->fackets_out, skb);
		tcp_packets_out_dec(tp, skb);
D
David S. Miller 已提交
2393
		__skb_unlink(skb, &sk->sk_write_queue);
L
Linus Torvalds 已提交
2394
		sk_stream_free_skb(sk, skb);
2395
		clear_all_retrans_hints(tp);
L
Linus Torvalds 已提交
2396 2397 2398
	}

	if (acked&FLAG_ACKED) {
2399
		tcp_ack_update_rtt(sk, acked, seq_rtt);
L
Linus Torvalds 已提交
2400
		tcp_ack_packets_out(sk, tp);
2401 2402
		if (rtt_sample && !(acked & FLAG_RETRANS_DATA_ACKED))
			(*rtt_sample)(sk, tcp_usrtt(&tv));
2403

2404 2405
		if (icsk->icsk_ca_ops->pkts_acked)
			icsk->icsk_ca_ops->pkts_acked(sk, pkts_acked);
L
Linus Torvalds 已提交
2406 2407 2408 2409 2410 2411 2412
	}

#if FASTRETRANS_DEBUG > 0
	BUG_TRAP((int)tp->sacked_out >= 0);
	BUG_TRAP((int)tp->lost_out >= 0);
	BUG_TRAP((int)tp->retrans_out >= 0);
	if (!tp->packets_out && tp->rx_opt.sack_ok) {
2413
		const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2414 2415
		if (tp->lost_out) {
			printk(KERN_DEBUG "Leak l=%u %d\n",
2416
			       tp->lost_out, icsk->icsk_ca_state);
L
Linus Torvalds 已提交
2417 2418 2419 2420
			tp->lost_out = 0;
		}
		if (tp->sacked_out) {
			printk(KERN_DEBUG "Leak s=%u %d\n",
2421
			       tp->sacked_out, icsk->icsk_ca_state);
L
Linus Torvalds 已提交
2422 2423 2424 2425
			tp->sacked_out = 0;
		}
		if (tp->retrans_out) {
			printk(KERN_DEBUG "Leak r=%u %d\n",
2426
			       tp->retrans_out, icsk->icsk_ca_state);
L
Linus Torvalds 已提交
2427 2428 2429 2430 2431 2432 2433 2434 2435 2436
			tp->retrans_out = 0;
		}
	}
#endif
	*seq_rtt_p = seq_rtt;
	return acked;
}

static void tcp_ack_probe(struct sock *sk)
{
2437 2438
	const struct tcp_sock *tp = tcp_sk(sk);
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2439 2440 2441 2442 2443

	/* Was it a usable window open? */

	if (!after(TCP_SKB_CB(sk->sk_send_head)->end_seq,
		   tp->snd_una + tp->snd_wnd)) {
2444 2445
		icsk->icsk_backoff = 0;
		inet_csk_clear_xmit_timer(sk, ICSK_TIME_PROBE0);
L
Linus Torvalds 已提交
2446 2447 2448 2449
		/* Socket must be waked up by subsequent tcp_data_snd_check().
		 * This function is not for random using!
		 */
	} else {
2450
		inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
2451 2452
					  min(icsk->icsk_rto << icsk->icsk_backoff, TCP_RTO_MAX),
					  TCP_RTO_MAX);
L
Linus Torvalds 已提交
2453 2454 2455
	}
}

2456
static inline int tcp_ack_is_dubious(const struct sock *sk, const int flag)
L
Linus Torvalds 已提交
2457 2458
{
	return (!(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) ||
2459
		inet_csk(sk)->icsk_ca_state != TCP_CA_Open);
L
Linus Torvalds 已提交
2460 2461
}

2462
static inline int tcp_may_raise_cwnd(const struct sock *sk, const int flag)
L
Linus Torvalds 已提交
2463
{
2464
	const struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
2465
	return (!(flag & FLAG_ECE) || tp->snd_cwnd < tp->snd_ssthresh) &&
2466
		!((1 << inet_csk(sk)->icsk_ca_state) & (TCPF_CA_Recovery | TCPF_CA_CWR));
L
Linus Torvalds 已提交
2467 2468 2469 2470 2471
}

/* Check that window update is acceptable.
 * The function assumes that snd_una<=ack<=snd_next.
 */
2472 2473
static inline int tcp_may_update_window(const struct tcp_sock *tp, const u32 ack,
					const u32 ack_seq, const u32 nwin)
L
Linus Torvalds 已提交
2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503
{
	return (after(ack, tp->snd_una) ||
		after(ack_seq, tp->snd_wl1) ||
		(ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd));
}

/* Update our send window.
 *
 * Window update algorithm, described in RFC793/RFC1122 (used in linux-2.2
 * and in FreeBSD. NetBSD's one is even worse.) is wrong.
 */
static int tcp_ack_update_window(struct sock *sk, struct tcp_sock *tp,
				 struct sk_buff *skb, u32 ack, u32 ack_seq)
{
	int flag = 0;
	u32 nwin = ntohs(skb->h.th->window);

	if (likely(!skb->h.th->syn))
		nwin <<= tp->rx_opt.snd_wscale;

	if (tcp_may_update_window(tp, ack, ack_seq, nwin)) {
		flag |= FLAG_WIN_UPDATE;
		tcp_update_wl(tp, ack, ack_seq);

		if (tp->snd_wnd != nwin) {
			tp->snd_wnd = nwin;

			/* Note, it is the only place, where
			 * fast path is recovered for sending TCP.
			 */
2504
			tp->pred_flags = 0;
L
Linus Torvalds 已提交
2505 2506 2507 2508
			tcp_fast_path_check(sk, tp);

			if (nwin > tp->max_window) {
				tp->max_window = nwin;
2509
				tcp_sync_mss(sk, inet_csk(sk)->icsk_pmtu_cookie);
L
Linus Torvalds 已提交
2510 2511 2512 2513 2514 2515 2516 2517 2518
			}
		}
	}

	tp->snd_una = ack;

	return flag;
}

2519 2520 2521 2522 2523 2524
/* A very conservative spurious RTO response algorithm: reduce cwnd and
 * continue in congestion avoidance.
 */
static void tcp_conservative_spur_to_response(struct tcp_sock *tp)
{
	tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh);
2525
	tp->snd_cwnd_cnt = 0;
2526 2527 2528
	tcp_moderate_cwnd(tp);
}

2529 2530
/* F-RTO spurious RTO detection algorithm (RFC4138)
 *
2531 2532 2533
 * F-RTO affects during two new ACKs following RTO (well, almost, see inline
 * comments). State (ACK number) is kept in frto_counter. When ACK advances
 * window (but not to or beyond highest sequence sent before RTO):
2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552
 *   On First ACK,  send two new segments out.
 *   On Second ACK, RTO was likely spurious. Do spurious response (response
 *                  algorithm is not part of the F-RTO detection algorithm
 *                  given in RFC4138 but can be selected separately).
 * Otherwise (basically on duplicate ACK), RTO was (likely) caused by a loss
 * and TCP falls back to conventional RTO recovery.
 *
 * Rationale: if the RTO was spurious, new ACKs should arrive from the
 * original window even after we transmit two new data segments.
 *
 * F-RTO is implemented (mainly) in four functions:
 *   - tcp_use_frto() is used to determine if TCP is can use F-RTO
 *   - tcp_enter_frto() prepares TCP state on RTO if F-RTO is used, it is
 *     called when tcp_use_frto() showed green light
 *   - tcp_process_frto() handles incoming ACKs during F-RTO algorithm
 *   - tcp_enter_frto_loss() is called if there is not enough evidence
 *     to prove that the RTO is indeed spurious. It transfers the control
 *     from F-RTO to the conventional RTO recovery
 */
2553
static int tcp_process_frto(struct sock *sk, u32 prior_snd_una, int flag)
L
Linus Torvalds 已提交
2554 2555
{
	struct tcp_sock *tp = tcp_sk(sk);
2556

L
Linus Torvalds 已提交
2557
	tcp_sync_left_out(tp);
2558

2559 2560 2561 2562
	/* Duplicate the behavior from Loss state (fastretrans_alert) */
	if (flag&FLAG_DATA_ACKED)
		inet_csk(sk)->icsk_retransmits = 0;

2563
	if (!before(tp->snd_una, tp->frto_highmark)) {
2564
		tcp_enter_frto_loss(sk, tp->frto_counter + 1, flag);
2565
		return 1;
2566 2567
	}

2568 2569 2570 2571 2572
	/* RFC4138 shortcoming in step 2; should also have case c): ACK isn't
	 * duplicate nor advances window, e.g., opposite dir data, winupdate
	 */
	if ((tp->snd_una == prior_snd_una) && (flag&FLAG_NOT_DUP) &&
	    !(flag&FLAG_FORWARD_PROGRESS))
2573
		return 1;
2574

2575
	if (!(flag&FLAG_DATA_ACKED)) {
2576
		tcp_enter_frto_loss(sk, (tp->frto_counter == 1 ? 0 : 3), flag);
2577
		return 1;
L
Linus Torvalds 已提交
2578 2579 2580 2581
	}

	if (tp->frto_counter == 1) {
		tp->snd_cwnd = tcp_packets_in_flight(tp) + 2;
2582
		tp->frto_counter = 2;
2583
		return 1;
2584
	} else /* frto_counter == 2 */ {
2585
		tcp_conservative_spur_to_response(tp);
2586
		tp->frto_counter = 0;
L
Linus Torvalds 已提交
2587
	}
2588
	return 0;
L
Linus Torvalds 已提交
2589 2590 2591 2592 2593
}

/* This routine deals with incoming acks, but not outgoing ones. */
static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
{
2594
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2595 2596 2597 2598 2599 2600 2601
	struct tcp_sock *tp = tcp_sk(sk);
	u32 prior_snd_una = tp->snd_una;
	u32 ack_seq = TCP_SKB_CB(skb)->seq;
	u32 ack = TCP_SKB_CB(skb)->ack_seq;
	u32 prior_in_flight;
	s32 seq_rtt;
	int prior_packets;
2602
	int frto_cwnd = 0;
L
Linus Torvalds 已提交
2603 2604 2605 2606 2607 2608 2609 2610 2611 2612

	/* If the ack is newer than sent or older than previous acks
	 * then we can probably ignore it.
	 */
	if (after(ack, tp->snd_nxt))
		goto uninteresting_ack;

	if (before(ack, prior_snd_una))
		goto old_ack;

2613 2614 2615 2616 2617 2618 2619
	if (sysctl_tcp_abc) {
		if (icsk->icsk_ca_state < TCP_CA_CWR)
			tp->bytes_acked += ack - prior_snd_una;
		else if (icsk->icsk_ca_state == TCP_CA_Loss)
			/* we assume just one segment left network */
			tp->bytes_acked += min(ack - prior_snd_una, tp->mss_cache);
	}
2620

L
Linus Torvalds 已提交
2621 2622 2623 2624 2625 2626 2627 2628 2629
	if (!(flag&FLAG_SLOWPATH) && after(ack, prior_snd_una)) {
		/* Window is constant, pure forward advance.
		 * No more checks are required.
		 * Note, we use the fact that SND.UNA>=SND.WL2.
		 */
		tcp_update_wl(tp, ack, ack_seq);
		tp->snd_una = ack;
		flag |= FLAG_WIN_UPDATE;

2630
		tcp_ca_event(sk, CA_EVENT_FAST_ACK);
2631

L
Linus Torvalds 已提交
2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646
		NET_INC_STATS_BH(LINUX_MIB_TCPHPACKS);
	} else {
		if (ack_seq != TCP_SKB_CB(skb)->end_seq)
			flag |= FLAG_DATA;
		else
			NET_INC_STATS_BH(LINUX_MIB_TCPPUREACKS);

		flag |= tcp_ack_update_window(sk, tp, skb, ack, ack_seq);

		if (TCP_SKB_CB(skb)->sacked)
			flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una);

		if (TCP_ECN_rcv_ecn_echo(tp, skb->h.th))
			flag |= FLAG_ECE;

2647
		tcp_ca_event(sk, CA_EVENT_SLOW_ACK);
L
Linus Torvalds 已提交
2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661
	}

	/* We passed data and got it acked, remove any soft error
	 * log. Something worked...
	 */
	sk->sk_err_soft = 0;
	tp->rcv_tstamp = tcp_time_stamp;
	prior_packets = tp->packets_out;
	if (!prior_packets)
		goto no_queue;

	prior_in_flight = tcp_packets_in_flight(tp);

	/* See if we can take anything off of the retransmit queue. */
2662
	flag |= tcp_clean_rtx_queue(sk, &seq_rtt);
L
Linus Torvalds 已提交
2663 2664

	if (tp->frto_counter)
2665
		frto_cwnd = tcp_process_frto(sk, prior_snd_una, flag);
L
Linus Torvalds 已提交
2666

2667
	if (tcp_ack_is_dubious(sk, flag)) {
S
Stephen Hemminger 已提交
2668
		/* Advance CWND, if state allows this. */
2669 2670
		if ((flag & FLAG_DATA_ACKED) && !frto_cwnd &&
		    tcp_may_raise_cwnd(sk, flag))
2671
			tcp_cong_avoid(sk, ack,  seq_rtt, prior_in_flight, 0);
L
Linus Torvalds 已提交
2672 2673
		tcp_fastretrans_alert(sk, prior_snd_una, prior_packets, flag);
	} else {
2674
		if ((flag & FLAG_DATA_ACKED) && !frto_cwnd)
2675
			tcp_cong_avoid(sk, ack, seq_rtt, prior_in_flight, 1);
L
Linus Torvalds 已提交
2676 2677 2678 2679 2680 2681 2682 2683
	}

	if ((flag & FLAG_FORWARD_PROGRESS) || !(flag&FLAG_NOT_DUP))
		dst_confirm(sk->sk_dst_cache);

	return 1;

no_queue:
2684
	icsk->icsk_probes_out = 0;
L
Linus Torvalds 已提交
2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717

	/* If this ack opens up a zero window, clear backoff.  It was
	 * being used to time the probes, and is probably far higher than
	 * it needs to be for normal retransmission.
	 */
	if (sk->sk_send_head)
		tcp_ack_probe(sk);
	return 1;

old_ack:
	if (TCP_SKB_CB(skb)->sacked)
		tcp_sacktag_write_queue(sk, skb, prior_snd_una);

uninteresting_ack:
	SOCK_DEBUG(sk, "Ack %u out of %u:%u\n", ack, tp->snd_una, tp->snd_nxt);
	return 0;
}


/* Look for tcp options. Normally only called on SYN and SYNACK packets.
 * But, this can also be called on packets in the established flow when
 * the fast version below fails.
 */
void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, int estab)
{
	unsigned char *ptr;
	struct tcphdr *th = skb->h.th;
	int length=(th->doff*4)-sizeof(struct tcphdr);

	ptr = (unsigned char *)(th + 1);
	opt_rx->saw_tstamp = 0;

	while(length>0) {
2718
		int opcode=*ptr++;
L
Linus Torvalds 已提交
2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732
		int opsize;

		switch (opcode) {
			case TCPOPT_EOL:
				return;
			case TCPOPT_NOP:	/* Ref: RFC 793 section 3.1 */
				length--;
				continue;
			default:
				opsize=*ptr++;
				if (opsize < 2) /* "silly options" */
					return;
				if (opsize > length)
					return;	/* don't parse partial options */
2733
				switch(opcode) {
L
Linus Torvalds 已提交
2734 2735
				case TCPOPT_MSS:
					if(opsize==TCPOLEN_MSS && th->syn && !estab) {
2736
						u16 in_mss = ntohs(get_unaligned((__be16 *)ptr));
L
Linus Torvalds 已提交
2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763
						if (in_mss) {
							if (opt_rx->user_mss && opt_rx->user_mss < in_mss)
								in_mss = opt_rx->user_mss;
							opt_rx->mss_clamp = in_mss;
						}
					}
					break;
				case TCPOPT_WINDOW:
					if(opsize==TCPOLEN_WINDOW && th->syn && !estab)
						if (sysctl_tcp_window_scaling) {
							__u8 snd_wscale = *(__u8 *) ptr;
							opt_rx->wscale_ok = 1;
							if (snd_wscale > 14) {
								if(net_ratelimit())
									printk(KERN_INFO "tcp_parse_options: Illegal window "
									       "scaling value %d >14 received.\n",
									       snd_wscale);
								snd_wscale = 14;
							}
							opt_rx->snd_wscale = snd_wscale;
						}
					break;
				case TCPOPT_TIMESTAMP:
					if(opsize==TCPOLEN_TIMESTAMP) {
						if ((estab && opt_rx->tstamp_ok) ||
						    (!estab && sysctl_tcp_timestamps)) {
							opt_rx->saw_tstamp = 1;
2764 2765
							opt_rx->rcv_tsval = ntohl(get_unaligned((__be32 *)ptr));
							opt_rx->rcv_tsecr = ntohl(get_unaligned((__be32 *)(ptr+4)));
L
Linus Torvalds 已提交
2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783
						}
					}
					break;
				case TCPOPT_SACK_PERM:
					if(opsize==TCPOLEN_SACK_PERM && th->syn && !estab) {
						if (sysctl_tcp_sack) {
							opt_rx->sack_ok = 1;
							tcp_sack_reset(opt_rx);
						}
					}
					break;

				case TCPOPT_SACK:
					if((opsize >= (TCPOLEN_SACK_BASE + TCPOLEN_SACK_PERBLOCK)) &&
					   !((opsize - TCPOLEN_SACK_BASE) % TCPOLEN_SACK_PERBLOCK) &&
					   opt_rx->sack_ok) {
						TCP_SKB_CB(skb)->sacked = (ptr - 2) - (unsigned char *)th;
					}
2784 2785 2786 2787 2788 2789 2790 2791
#ifdef CONFIG_TCP_MD5SIG
				case TCPOPT_MD5SIG:
					/*
					 * The MD5 Hash has already been
					 * checked (see tcp_v{4,6}_do_rcv()).
					 */
					break;
#endif
2792 2793 2794 2795
				};
				ptr+=opsize-2;
				length-=opsize;
		};
L
Linus Torvalds 已提交
2796 2797 2798 2799 2800 2801
	}
}

/* Fast parse options. This hopes to only see timestamps.
 * If it is wrong it falls back on tcp_parse_options().
 */
S
Stephen Hemminger 已提交
2802 2803
static int tcp_fast_parse_options(struct sk_buff *skb, struct tcphdr *th,
				  struct tcp_sock *tp)
L
Linus Torvalds 已提交
2804 2805 2806 2807 2808 2809
{
	if (th->doff == sizeof(struct tcphdr)>>2) {
		tp->rx_opt.saw_tstamp = 0;
		return 0;
	} else if (tp->rx_opt.tstamp_ok &&
		   th->doff == (sizeof(struct tcphdr)>>2)+(TCPOLEN_TSTAMP_ALIGNED>>2)) {
2810 2811
		__be32 *ptr = (__be32 *)(th + 1);
		if (*ptr == htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
L
Linus Torvalds 已提交
2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869
				  | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP)) {
			tp->rx_opt.saw_tstamp = 1;
			++ptr;
			tp->rx_opt.rcv_tsval = ntohl(*ptr);
			++ptr;
			tp->rx_opt.rcv_tsecr = ntohl(*ptr);
			return 1;
		}
	}
	tcp_parse_options(skb, &tp->rx_opt, 1);
	return 1;
}

static inline void tcp_store_ts_recent(struct tcp_sock *tp)
{
	tp->rx_opt.ts_recent = tp->rx_opt.rcv_tsval;
	tp->rx_opt.ts_recent_stamp = xtime.tv_sec;
}

static inline void tcp_replace_ts_recent(struct tcp_sock *tp, u32 seq)
{
	if (tp->rx_opt.saw_tstamp && !after(seq, tp->rcv_wup)) {
		/* PAWS bug workaround wrt. ACK frames, the PAWS discard
		 * extra check below makes sure this can only happen
		 * for pure ACK frames.  -DaveM
		 *
		 * Not only, also it occurs for expired timestamps.
		 */

		if((s32)(tp->rx_opt.rcv_tsval - tp->rx_opt.ts_recent) >= 0 ||
		   xtime.tv_sec >= tp->rx_opt.ts_recent_stamp + TCP_PAWS_24DAYS)
			tcp_store_ts_recent(tp);
	}
}

/* Sorry, PAWS as specified is broken wrt. pure-ACKs -DaveM
 *
 * It is not fatal. If this ACK does _not_ change critical state (seqs, window)
 * it can pass through stack. So, the following predicate verifies that
 * this segment is not used for anything but congestion avoidance or
 * fast retransmit. Moreover, we even are able to eliminate most of such
 * second order effects, if we apply some small "replay" window (~RTO)
 * to timestamp space.
 *
 * All these measures still do not guarantee that we reject wrapped ACKs
 * on networks with high bandwidth, when sequence space is recycled fastly,
 * but it guarantees that such events will be very rare and do not affect
 * connection seriously. This doesn't look nice, but alas, PAWS is really
 * buggy extension.
 *
 * [ Later note. Even worse! It is buggy for segments _with_ data. RFC
 * states that events when retransmit arrives after original data are rare.
 * It is a blatant lie. VJ forgot about fast retransmit! 8)8) It is
 * the biggest problem on large power networks even with minor reordering.
 * OK, let's give it small replay window. If peer clock is even 1hz, it is safe
 * up to bandwidth of 18Gigabit/sec. 8) ]
 */

2870
static int tcp_disordered_ack(const struct sock *sk, const struct sk_buff *skb)
L
Linus Torvalds 已提交
2871
{
2872
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886
	struct tcphdr *th = skb->h.th;
	u32 seq = TCP_SKB_CB(skb)->seq;
	u32 ack = TCP_SKB_CB(skb)->ack_seq;

	return (/* 1. Pure ACK with correct sequence number. */
		(th->ack && seq == TCP_SKB_CB(skb)->end_seq && seq == tp->rcv_nxt) &&

		/* 2. ... and duplicate ACK. */
		ack == tp->snd_una &&

		/* 3. ... and does not update window. */
		!tcp_may_update_window(tp, ack, seq, ntohs(th->window) << tp->rx_opt.snd_wscale) &&

		/* 4. ... and sits in replay window. */
2887
		(s32)(tp->rx_opt.ts_recent - tp->rx_opt.rcv_tsval) <= (inet_csk(sk)->icsk_rto * 1024) / HZ);
L
Linus Torvalds 已提交
2888 2889
}

2890
static inline int tcp_paws_discard(const struct sock *sk, const struct sk_buff *skb)
L
Linus Torvalds 已提交
2891
{
2892
	const struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
2893 2894
	return ((s32)(tp->rx_opt.ts_recent - tp->rx_opt.rcv_tsval) > TCP_PAWS_WINDOW &&
		xtime.tv_sec < tp->rx_opt.ts_recent_stamp + TCP_PAWS_24DAYS &&
2895
		!tcp_disordered_ack(sk, skb));
L
Linus Torvalds 已提交
2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957
}

/* Check segment sequence number for validity.
 *
 * Segment controls are considered valid, if the segment
 * fits to the window after truncation to the window. Acceptability
 * of data (and SYN, FIN, of course) is checked separately.
 * See tcp_data_queue(), for example.
 *
 * Also, controls (RST is main one) are accepted using RCV.WUP instead
 * of RCV.NXT. Peer still did not advance his SND.UNA when we
 * delayed ACK, so that hisSND.UNA<=ourRCV.WUP.
 * (borrowed from freebsd)
 */

static inline int tcp_sequence(struct tcp_sock *tp, u32 seq, u32 end_seq)
{
	return	!before(end_seq, tp->rcv_wup) &&
		!after(seq, tp->rcv_nxt + tcp_receive_window(tp));
}

/* When we get a reset we do this. */
static void tcp_reset(struct sock *sk)
{
	/* We want the right error as BSD sees it (and indeed as we do). */
	switch (sk->sk_state) {
		case TCP_SYN_SENT:
			sk->sk_err = ECONNREFUSED;
			break;
		case TCP_CLOSE_WAIT:
			sk->sk_err = EPIPE;
			break;
		case TCP_CLOSE:
			return;
		default:
			sk->sk_err = ECONNRESET;
	}

	if (!sock_flag(sk, SOCK_DEAD))
		sk->sk_error_report(sk);

	tcp_done(sk);
}

/*
 * 	Process the FIN bit. This now behaves as it is supposed to work
 *	and the FIN takes effect when it is validly part of sequence
 *	space. Not before when we get holes.
 *
 *	If we are ESTABLISHED, a received fin moves us to CLOSE-WAIT
 *	(and thence onto LAST-ACK and finally, CLOSE, we never enter
 *	TIME-WAIT)
 *
 *	If we are in FINWAIT-1, a received FIN indicates simultaneous
 *	close and we go into CLOSING (and later onto TIME-WAIT)
 *
 *	If we are in FINWAIT-2, a received FIN moves us to TIME-WAIT.
 */
static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th)
{
	struct tcp_sock *tp = tcp_sk(sk);

2958
	inet_csk_schedule_ack(sk);
L
Linus Torvalds 已提交
2959 2960 2961 2962 2963 2964 2965 2966 2967

	sk->sk_shutdown |= RCV_SHUTDOWN;
	sock_set_flag(sk, SOCK_DONE);

	switch (sk->sk_state) {
		case TCP_SYN_RECV:
		case TCP_ESTABLISHED:
			/* Move to CLOSE_WAIT */
			tcp_set_state(sk, TCP_CLOSE_WAIT);
2968
			inet_csk(sk)->icsk_ack.pingpong = 1;
L
Linus Torvalds 已提交
2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022
			break;

		case TCP_CLOSE_WAIT:
		case TCP_CLOSING:
			/* Received a retransmission of the FIN, do
			 * nothing.
			 */
			break;
		case TCP_LAST_ACK:
			/* RFC793: Remain in the LAST-ACK state. */
			break;

		case TCP_FIN_WAIT1:
			/* This case occurs when a simultaneous close
			 * happens, we must ack the received FIN and
			 * enter the CLOSING state.
			 */
			tcp_send_ack(sk);
			tcp_set_state(sk, TCP_CLOSING);
			break;
		case TCP_FIN_WAIT2:
			/* Received a FIN -- send ACK and enter TIME_WAIT. */
			tcp_send_ack(sk);
			tcp_time_wait(sk, TCP_TIME_WAIT, 0);
			break;
		default:
			/* Only TCP_LISTEN and TCP_CLOSE are left, in these
			 * cases we should never reach this piece of code.
			 */
			printk(KERN_ERR "%s: Impossible, sk->sk_state=%d\n",
			       __FUNCTION__, sk->sk_state);
			break;
	};

	/* It _is_ possible, that we have something out-of-order _after_ FIN.
	 * Probably, we should reset in this case. For now drop them.
	 */
	__skb_queue_purge(&tp->out_of_order_queue);
	if (tp->rx_opt.sack_ok)
		tcp_sack_reset(&tp->rx_opt);
	sk_stream_mem_reclaim(sk);

	if (!sock_flag(sk, SOCK_DEAD)) {
		sk->sk_state_change(sk);

		/* Do not send POLL_HUP for half duplex close. */
		if (sk->sk_shutdown == SHUTDOWN_MASK ||
		    sk->sk_state == TCP_CLOSE)
			sk_wake_async(sk, 1, POLL_HUP);
		else
			sk_wake_async(sk, 1, POLL_IN);
	}
}

S
Stephen Hemminger 已提交
3023
static inline int tcp_sack_extend(struct tcp_sack_block *sp, u32 seq, u32 end_seq)
L
Linus Torvalds 已提交
3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034
{
	if (!after(seq, sp->end_seq) && !after(sp->start_seq, end_seq)) {
		if (before(seq, sp->start_seq))
			sp->start_seq = seq;
		if (after(end_seq, sp->end_seq))
			sp->end_seq = end_seq;
		return 1;
	}
	return 0;
}

S
Stephen Hemminger 已提交
3035
static void tcp_dsack_set(struct tcp_sock *tp, u32 seq, u32 end_seq)
L
Linus Torvalds 已提交
3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049
{
	if (tp->rx_opt.sack_ok && sysctl_tcp_dsack) {
		if (before(seq, tp->rcv_nxt))
			NET_INC_STATS_BH(LINUX_MIB_TCPDSACKOLDSENT);
		else
			NET_INC_STATS_BH(LINUX_MIB_TCPDSACKOFOSENT);

		tp->rx_opt.dsack = 1;
		tp->duplicate_sack[0].start_seq = seq;
		tp->duplicate_sack[0].end_seq = end_seq;
		tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + 1, 4 - tp->rx_opt.tstamp_ok);
	}
}

S
Stephen Hemminger 已提交
3050
static void tcp_dsack_extend(struct tcp_sock *tp, u32 seq, u32 end_seq)
L
Linus Torvalds 已提交
3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064
{
	if (!tp->rx_opt.dsack)
		tcp_dsack_set(tp, seq, end_seq);
	else
		tcp_sack_extend(tp->duplicate_sack, seq, end_seq);
}

static void tcp_send_dupack(struct sock *sk, struct sk_buff *skb)
{
	struct tcp_sock *tp = tcp_sk(sk);

	if (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq &&
	    before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKLOST);
3065
		tcp_enter_quickack_mode(sk);
L
Linus Torvalds 已提交
3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107

		if (tp->rx_opt.sack_ok && sysctl_tcp_dsack) {
			u32 end_seq = TCP_SKB_CB(skb)->end_seq;

			if (after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt))
				end_seq = tp->rcv_nxt;
			tcp_dsack_set(tp, TCP_SKB_CB(skb)->seq, end_seq);
		}
	}

	tcp_send_ack(sk);
}

/* These routines update the SACK block as out-of-order packets arrive or
 * in-order packets close up the sequence space.
 */
static void tcp_sack_maybe_coalesce(struct tcp_sock *tp)
{
	int this_sack;
	struct tcp_sack_block *sp = &tp->selective_acks[0];
	struct tcp_sack_block *swalk = sp+1;

	/* See if the recent change to the first SACK eats into
	 * or hits the sequence space of other SACK blocks, if so coalesce.
	 */
	for (this_sack = 1; this_sack < tp->rx_opt.num_sacks; ) {
		if (tcp_sack_extend(sp, swalk->start_seq, swalk->end_seq)) {
			int i;

			/* Zap SWALK, by moving every further SACK up by one slot.
			 * Decrease num_sacks.
			 */
			tp->rx_opt.num_sacks--;
			tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, 4 - tp->rx_opt.tstamp_ok);
			for(i=this_sack; i < tp->rx_opt.num_sacks; i++)
				sp[i] = sp[i+1];
			continue;
		}
		this_sack++, swalk++;
	}
}

S
Stephen Hemminger 已提交
3108
static inline void tcp_sack_swap(struct tcp_sack_block *sack1, struct tcp_sack_block *sack2)
L
Linus Torvalds 已提交
3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172
{
	__u32 tmp;

	tmp = sack1->start_seq;
	sack1->start_seq = sack2->start_seq;
	sack2->start_seq = tmp;

	tmp = sack1->end_seq;
	sack1->end_seq = sack2->end_seq;
	sack2->end_seq = tmp;
}

static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct tcp_sack_block *sp = &tp->selective_acks[0];
	int cur_sacks = tp->rx_opt.num_sacks;
	int this_sack;

	if (!cur_sacks)
		goto new_sack;

	for (this_sack=0; this_sack<cur_sacks; this_sack++, sp++) {
		if (tcp_sack_extend(sp, seq, end_seq)) {
			/* Rotate this_sack to the first one. */
			for (; this_sack>0; this_sack--, sp--)
				tcp_sack_swap(sp, sp-1);
			if (cur_sacks > 1)
				tcp_sack_maybe_coalesce(tp);
			return;
		}
	}

	/* Could not find an adjacent existing SACK, build a new one,
	 * put it at the front, and shift everyone else down.  We
	 * always know there is at least one SACK present already here.
	 *
	 * If the sack array is full, forget about the last one.
	 */
	if (this_sack >= 4) {
		this_sack--;
		tp->rx_opt.num_sacks--;
		sp--;
	}
	for(; this_sack > 0; this_sack--, sp--)
		*sp = *(sp-1);

new_sack:
	/* Build the new head SACK, and we're done. */
	sp->start_seq = seq;
	sp->end_seq = end_seq;
	tp->rx_opt.num_sacks++;
	tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, 4 - tp->rx_opt.tstamp_ok);
}

/* RCV.NXT advances, some SACKs should be eaten. */

static void tcp_sack_remove(struct tcp_sock *tp)
{
	struct tcp_sack_block *sp = &tp->selective_acks[0];
	int num_sacks = tp->rx_opt.num_sacks;
	int this_sack;

	/* Empty ofo queue, hence, all the SACKs are eaten. Clear. */
3173
	if (skb_queue_empty(&tp->out_of_order_queue)) {
L
Linus Torvalds 已提交
3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223
		tp->rx_opt.num_sacks = 0;
		tp->rx_opt.eff_sacks = tp->rx_opt.dsack;
		return;
	}

	for(this_sack = 0; this_sack < num_sacks; ) {
		/* Check if the start of the sack is covered by RCV.NXT. */
		if (!before(tp->rcv_nxt, sp->start_seq)) {
			int i;

			/* RCV.NXT must cover all the block! */
			BUG_TRAP(!before(tp->rcv_nxt, sp->end_seq));

			/* Zap this SACK, by moving forward any other SACKS. */
			for (i=this_sack+1; i < num_sacks; i++)
				tp->selective_acks[i-1] = tp->selective_acks[i];
			num_sacks--;
			continue;
		}
		this_sack++;
		sp++;
	}
	if (num_sacks != tp->rx_opt.num_sacks) {
		tp->rx_opt.num_sacks = num_sacks;
		tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, 4 - tp->rx_opt.tstamp_ok);
	}
}

/* This one checks to see if we can put data from the
 * out_of_order queue into the receive_queue.
 */
static void tcp_ofo_queue(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	__u32 dsack_high = tp->rcv_nxt;
	struct sk_buff *skb;

	while ((skb = skb_peek(&tp->out_of_order_queue)) != NULL) {
		if (after(TCP_SKB_CB(skb)->seq, tp->rcv_nxt))
			break;

		if (before(TCP_SKB_CB(skb)->seq, dsack_high)) {
			__u32 dsack = dsack_high;
			if (before(TCP_SKB_CB(skb)->end_seq, dsack_high))
				dsack_high = TCP_SKB_CB(skb)->end_seq;
			tcp_dsack_extend(tp, TCP_SKB_CB(skb)->seq, dsack);
		}

		if (!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt)) {
			SOCK_DEBUG(sk, "ofo packet was already received \n");
D
David S. Miller 已提交
3224
			__skb_unlink(skb, &tp->out_of_order_queue);
L
Linus Torvalds 已提交
3225 3226 3227 3228 3229 3230 3231
			__kfree_skb(skb);
			continue;
		}
		SOCK_DEBUG(sk, "ofo requeuing : rcv_next %X seq %X - %X\n",
			   tp->rcv_nxt, TCP_SKB_CB(skb)->seq,
			   TCP_SKB_CB(skb)->end_seq);

D
David S. Miller 已提交
3232
		__skb_unlink(skb, &tp->out_of_order_queue);
L
Linus Torvalds 已提交
3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305
		__skb_queue_tail(&sk->sk_receive_queue, skb);
		tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
		if(skb->h.th->fin)
			tcp_fin(skb, sk, skb->h.th);
	}
}

static int tcp_prune_queue(struct sock *sk);

static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
{
	struct tcphdr *th = skb->h.th;
	struct tcp_sock *tp = tcp_sk(sk);
	int eaten = -1;

	if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq)
		goto drop;

	__skb_pull(skb, th->doff*4);

	TCP_ECN_accept_cwr(tp, skb);

	if (tp->rx_opt.dsack) {
		tp->rx_opt.dsack = 0;
		tp->rx_opt.eff_sacks = min_t(unsigned int, tp->rx_opt.num_sacks,
						    4 - tp->rx_opt.tstamp_ok);
	}

	/*  Queue data for delivery to the user.
	 *  Packets in sequence go to the receive queue.
	 *  Out of sequence packets to the out_of_order_queue.
	 */
	if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt) {
		if (tcp_receive_window(tp) == 0)
			goto out_of_window;

		/* Ok. In sequence. In window. */
		if (tp->ucopy.task == current &&
		    tp->copied_seq == tp->rcv_nxt && tp->ucopy.len &&
		    sock_owned_by_user(sk) && !tp->urg_data) {
			int chunk = min_t(unsigned int, skb->len,
							tp->ucopy.len);

			__set_current_state(TASK_RUNNING);

			local_bh_enable();
			if (!skb_copy_datagram_iovec(skb, 0, tp->ucopy.iov, chunk)) {
				tp->ucopy.len -= chunk;
				tp->copied_seq += chunk;
				eaten = (chunk == skb->len && !th->fin);
				tcp_rcv_space_adjust(sk);
			}
			local_bh_disable();
		}

		if (eaten <= 0) {
queue_and_out:
			if (eaten < 0 &&
			    (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
			     !sk_stream_rmem_schedule(sk, skb))) {
				if (tcp_prune_queue(sk) < 0 ||
				    !sk_stream_rmem_schedule(sk, skb))
					goto drop;
			}
			sk_stream_set_owner_r(skb, sk);
			__skb_queue_tail(&sk->sk_receive_queue, skb);
		}
		tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
		if(skb->len)
			tcp_event_data_recv(sk, tp, skb);
		if(th->fin)
			tcp_fin(skb, sk, th);

3306
		if (!skb_queue_empty(&tp->out_of_order_queue)) {
L
Linus Torvalds 已提交
3307 3308 3309 3310 3311
			tcp_ofo_queue(sk);

			/* RFC2581. 4.2. SHOULD send immediate ACK, when
			 * gap in queue is filled.
			 */
3312
			if (skb_queue_empty(&tp->out_of_order_queue))
3313
				inet_csk(sk)->icsk_ack.pingpong = 0;
L
Linus Torvalds 已提交
3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333
		}

		if (tp->rx_opt.num_sacks)
			tcp_sack_remove(tp);

		tcp_fast_path_check(sk, tp);

		if (eaten > 0)
			__kfree_skb(skb);
		else if (!sock_flag(sk, SOCK_DEAD))
			sk->sk_data_ready(sk, 0);
		return;
	}

	if (!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt)) {
		/* A retransmit, 2nd most common case.  Force an immediate ack. */
		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKLOST);
		tcp_dsack_set(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq);

out_of_window:
3334 3335
		tcp_enter_quickack_mode(sk);
		inet_csk_schedule_ack(sk);
L
Linus Torvalds 已提交
3336 3337 3338 3339 3340 3341 3342 3343 3344
drop:
		__kfree_skb(skb);
		return;
	}

	/* Out of window. F.e. zero window probe. */
	if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt + tcp_receive_window(tp)))
		goto out_of_window;

3345
	tcp_enter_quickack_mode(sk);
L
Linus Torvalds 已提交
3346 3347 3348 3349 3350 3351 3352 3353

	if (before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
		/* Partial packet, seq < rcv_next < end_seq */
		SOCK_DEBUG(sk, "partial packet: rcv_next %X seq %X - %X\n",
			   tp->rcv_nxt, TCP_SKB_CB(skb)->seq,
			   TCP_SKB_CB(skb)->end_seq);

		tcp_dsack_set(tp, TCP_SKB_CB(skb)->seq, tp->rcv_nxt);
3354

L
Linus Torvalds 已提交
3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373
		/* If window is closed, drop tail of packet. But after
		 * remembering D-SACK for its head made in previous line.
		 */
		if (!tcp_receive_window(tp))
			goto out_of_window;
		goto queue_and_out;
	}

	TCP_ECN_check_ce(tp, skb);

	if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
	    !sk_stream_rmem_schedule(sk, skb)) {
		if (tcp_prune_queue(sk) < 0 ||
		    !sk_stream_rmem_schedule(sk, skb))
			goto drop;
	}

	/* Disable header prediction. */
	tp->pred_flags = 0;
3374
	inet_csk_schedule_ack(sk);
L
Linus Torvalds 已提交
3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397

	SOCK_DEBUG(sk, "out of order segment: rcv_next %X seq %X - %X\n",
		   tp->rcv_nxt, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq);

	sk_stream_set_owner_r(skb, sk);

	if (!skb_peek(&tp->out_of_order_queue)) {
		/* Initial out of order segment, build 1 SACK. */
		if (tp->rx_opt.sack_ok) {
			tp->rx_opt.num_sacks = 1;
			tp->rx_opt.dsack     = 0;
			tp->rx_opt.eff_sacks = 1;
			tp->selective_acks[0].start_seq = TCP_SKB_CB(skb)->seq;
			tp->selective_acks[0].end_seq =
						TCP_SKB_CB(skb)->end_seq;
		}
		__skb_queue_head(&tp->out_of_order_queue,skb);
	} else {
		struct sk_buff *skb1 = tp->out_of_order_queue.prev;
		u32 seq = TCP_SKB_CB(skb)->seq;
		u32 end_seq = TCP_SKB_CB(skb)->end_seq;

		if (seq == TCP_SKB_CB(skb1)->end_seq) {
D
David S. Miller 已提交
3398
			__skb_append(skb1, skb, &tp->out_of_order_queue);
L
Linus Torvalds 已提交
3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432

			if (!tp->rx_opt.num_sacks ||
			    tp->selective_acks[0].end_seq != seq)
				goto add_sack;

			/* Common case: data arrive in order after hole. */
			tp->selective_acks[0].end_seq = end_seq;
			return;
		}

		/* Find place to insert this segment. */
		do {
			if (!after(TCP_SKB_CB(skb1)->seq, seq))
				break;
		} while ((skb1 = skb1->prev) !=
			 (struct sk_buff*)&tp->out_of_order_queue);

		/* Do skb overlap to previous one? */
		if (skb1 != (struct sk_buff*)&tp->out_of_order_queue &&
		    before(seq, TCP_SKB_CB(skb1)->end_seq)) {
			if (!after(end_seq, TCP_SKB_CB(skb1)->end_seq)) {
				/* All the bits are present. Drop. */
				__kfree_skb(skb);
				tcp_dsack_set(tp, seq, end_seq);
				goto add_sack;
			}
			if (after(seq, TCP_SKB_CB(skb1)->seq)) {
				/* Partial overlap. */
				tcp_dsack_set(tp, seq, TCP_SKB_CB(skb1)->end_seq);
			} else {
				skb1 = skb1->prev;
			}
		}
		__skb_insert(skb, skb1, skb1->next, &tp->out_of_order_queue);
3433

L
Linus Torvalds 已提交
3434 3435 3436 3437 3438 3439 3440 3441
		/* And clean segments covered by new one as whole. */
		while ((skb1 = skb->next) !=
		       (struct sk_buff*)&tp->out_of_order_queue &&
		       after(end_seq, TCP_SKB_CB(skb1)->seq)) {
		       if (before(end_seq, TCP_SKB_CB(skb1)->end_seq)) {
			       tcp_dsack_extend(tp, TCP_SKB_CB(skb1)->seq, end_seq);
			       break;
		       }
D
David S. Miller 已提交
3442
		       __skb_unlink(skb1, &tp->out_of_order_queue);
L
Linus Torvalds 已提交
3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458
		       tcp_dsack_extend(tp, TCP_SKB_CB(skb1)->seq, TCP_SKB_CB(skb1)->end_seq);
		       __kfree_skb(skb1);
		}

add_sack:
		if (tp->rx_opt.sack_ok)
			tcp_sack_new_ofo_skb(sk, seq, end_seq);
	}
}

/* Collapse contiguous sequence of skbs head..tail with
 * sequence numbers start..end.
 * Segments with FIN/SYN are not collapsed (only because this
 * simplifies code)
 */
static void
D
David S. Miller 已提交
3459 3460 3461
tcp_collapse(struct sock *sk, struct sk_buff_head *list,
	     struct sk_buff *head, struct sk_buff *tail,
	     u32 start, u32 end)
L
Linus Torvalds 已提交
3462 3463 3464
{
	struct sk_buff *skb;

S
Stephen Hemminger 已提交
3465
	/* First, check that queue is collapsible and find
L
Linus Torvalds 已提交
3466 3467 3468 3469 3470
	 * the point where collapsing can be useful. */
	for (skb = head; skb != tail; ) {
		/* No new bits? It is possible on ofo queue. */
		if (!before(start, TCP_SKB_CB(skb)->end_seq)) {
			struct sk_buff *next = skb->next;
D
David S. Miller 已提交
3471
			__skb_unlink(skb, list);
L
Linus Torvalds 已提交
3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516
			__kfree_skb(skb);
			NET_INC_STATS_BH(LINUX_MIB_TCPRCVCOLLAPSED);
			skb = next;
			continue;
		}

		/* The first skb to collapse is:
		 * - not SYN/FIN and
		 * - bloated or contains data before "start" or
		 *   overlaps to the next one.
		 */
		if (!skb->h.th->syn && !skb->h.th->fin &&
		    (tcp_win_from_space(skb->truesize) > skb->len ||
		     before(TCP_SKB_CB(skb)->seq, start) ||
		     (skb->next != tail &&
		      TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb->next)->seq)))
			break;

		/* Decided to skip this, advance start seq. */
		start = TCP_SKB_CB(skb)->end_seq;
		skb = skb->next;
	}
	if (skb == tail || skb->h.th->syn || skb->h.th->fin)
		return;

	while (before(start, end)) {
		struct sk_buff *nskb;
		int header = skb_headroom(skb);
		int copy = SKB_MAX_ORDER(header, 0);

		/* Too big header? This can happen with IPv6. */
		if (copy < 0)
			return;
		if (end-start < copy)
			copy = end-start;
		nskb = alloc_skb(copy+header, GFP_ATOMIC);
		if (!nskb)
			return;
		skb_reserve(nskb, header);
		memcpy(nskb->head, skb->head, header);
		nskb->nh.raw = nskb->head + (skb->nh.raw-skb->head);
		nskb->h.raw = nskb->head + (skb->h.raw-skb->head);
		nskb->mac.raw = nskb->head + (skb->mac.raw-skb->head);
		memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
		TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start;
D
David S. Miller 已提交
3517
		__skb_insert(nskb, skb->prev, skb, list);
L
Linus Torvalds 已提交
3518 3519 3520 3521 3522 3523 3524
		sk_stream_set_owner_r(nskb, sk);

		/* Copy data, releasing collapsed skbs. */
		while (copy > 0) {
			int offset = start - TCP_SKB_CB(skb)->seq;
			int size = TCP_SKB_CB(skb)->end_seq - start;

3525
			BUG_ON(offset < 0);
L
Linus Torvalds 已提交
3526 3527 3528 3529 3530 3531 3532 3533 3534 3535
			if (size > 0) {
				size = min(copy, size);
				if (skb_copy_bits(skb, offset, skb_put(nskb, size), size))
					BUG();
				TCP_SKB_CB(nskb)->end_seq += size;
				copy -= size;
				start += size;
			}
			if (!before(start, TCP_SKB_CB(skb)->end_seq)) {
				struct sk_buff *next = skb->next;
D
David S. Miller 已提交
3536
				__skb_unlink(skb, list);
L
Linus Torvalds 已提交
3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571
				__kfree_skb(skb);
				NET_INC_STATS_BH(LINUX_MIB_TCPRCVCOLLAPSED);
				skb = next;
				if (skb == tail || skb->h.th->syn || skb->h.th->fin)
					return;
			}
		}
	}
}

/* Collapse ofo queue. Algorithm: select contiguous sequence of skbs
 * and tcp_collapse() them until all the queue is collapsed.
 */
static void tcp_collapse_ofo_queue(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb = skb_peek(&tp->out_of_order_queue);
	struct sk_buff *head;
	u32 start, end;

	if (skb == NULL)
		return;

	start = TCP_SKB_CB(skb)->seq;
	end = TCP_SKB_CB(skb)->end_seq;
	head = skb;

	for (;;) {
		skb = skb->next;

		/* Segment is terminated when we see gap or when
		 * we are at the end of all the queue. */
		if (skb == (struct sk_buff *)&tp->out_of_order_queue ||
		    after(TCP_SKB_CB(skb)->seq, end) ||
		    before(TCP_SKB_CB(skb)->end_seq, start)) {
D
David S. Miller 已提交
3572 3573
			tcp_collapse(sk, &tp->out_of_order_queue,
				     head, skb, start, end);
L
Linus Torvalds 已提交
3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597
			head = skb;
			if (skb == (struct sk_buff *)&tp->out_of_order_queue)
				break;
			/* Start new segment */
			start = TCP_SKB_CB(skb)->seq;
			end = TCP_SKB_CB(skb)->end_seq;
		} else {
			if (before(TCP_SKB_CB(skb)->seq, start))
				start = TCP_SKB_CB(skb)->seq;
			if (after(TCP_SKB_CB(skb)->end_seq, end))
				end = TCP_SKB_CB(skb)->end_seq;
		}
	}
}

/* Reduce allocated memory if we can, trying to get
 * the socket within its memory limits again.
 *
 * Return less than zero if we should start dropping frames
 * until the socket owning process reads some of the data
 * to stabilize the situation.
 */
static int tcp_prune_queue(struct sock *sk)
{
3598
	struct tcp_sock *tp = tcp_sk(sk);
L
Linus Torvalds 已提交
3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609

	SOCK_DEBUG(sk, "prune_queue: c=%x\n", tp->copied_seq);

	NET_INC_STATS_BH(LINUX_MIB_PRUNECALLED);

	if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)
		tcp_clamp_window(sk, tp);
	else if (tcp_memory_pressure)
		tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U * tp->advmss);

	tcp_collapse_ofo_queue(sk);
D
David S. Miller 已提交
3610 3611
	tcp_collapse(sk, &sk->sk_receive_queue,
		     sk->sk_receive_queue.next,
L
Linus Torvalds 已提交
3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622
		     (struct sk_buff*)&sk->sk_receive_queue,
		     tp->copied_seq, tp->rcv_nxt);
	sk_stream_mem_reclaim(sk);

	if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf)
		return 0;

	/* Collapsing did not help, destructive actions follow.
	 * This must not ever occur. */

	/* First, purge the out_of_order queue. */
3623 3624
	if (!skb_queue_empty(&tp->out_of_order_queue)) {
		NET_INC_STATS_BH(LINUX_MIB_OFOPRUNED);
L
Linus Torvalds 已提交
3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659
		__skb_queue_purge(&tp->out_of_order_queue);

		/* Reset SACK state.  A conforming SACK implementation will
		 * do the same at a timeout based retransmit.  When a connection
		 * is in a sad state like this, we care only about integrity
		 * of the connection not performance.
		 */
		if (tp->rx_opt.sack_ok)
			tcp_sack_reset(&tp->rx_opt);
		sk_stream_mem_reclaim(sk);
	}

	if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf)
		return 0;

	/* If we are really being abused, tell the caller to silently
	 * drop receive data on the floor.  It will get retransmitted
	 * and hopefully then we'll have sufficient space.
	 */
	NET_INC_STATS_BH(LINUX_MIB_RCVPRUNED);

	/* Massive buffer overcommit. */
	tp->pred_flags = 0;
	return -1;
}


/* RFC2861, slow part. Adjust cwnd, after it was not full during one rto.
 * As additional protections, we do not touch cwnd in retransmission phases,
 * and if application hit its sndbuf limit recently.
 */
void tcp_cwnd_application_limited(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

3660
	if (inet_csk(sk)->icsk_ca_state == TCP_CA_Open &&
L
Linus Torvalds 已提交
3661 3662
	    sk->sk_socket && !test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
		/* Limited by application or receiver window. */
3663 3664
		u32 init_win = tcp_init_cwnd(tp, __sk_dst_get(sk));
		u32 win_used = max(tp->snd_cwnd_used, init_win);
L
Linus Torvalds 已提交
3665
		if (win_used < tp->snd_cwnd) {
3666
			tp->snd_ssthresh = tcp_current_ssthresh(sk);
L
Linus Torvalds 已提交
3667 3668 3669 3670 3671 3672 3673
			tp->snd_cwnd = (tp->snd_cwnd + win_used) >> 1;
		}
		tp->snd_cwnd_used = 0;
	}
	tp->snd_cwnd_stamp = tcp_time_stamp;
}

S
Stephen Hemminger 已提交
3674
static int tcp_should_expand_sndbuf(struct sock *sk, struct tcp_sock *tp)
3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695
{
	/* If the user specified a specific send buffer setting, do
	 * not modify it.
	 */
	if (sk->sk_userlocks & SOCK_SNDBUF_LOCK)
		return 0;

	/* If we are under global TCP memory pressure, do not expand.  */
	if (tcp_memory_pressure)
		return 0;

	/* If we are under soft global TCP memory pressure, do not expand.  */
	if (atomic_read(&tcp_memory_allocated) >= sysctl_tcp_mem[0])
		return 0;

	/* If we filled the congestion window, do not expand.  */
	if (tp->packets_out >= tp->snd_cwnd)
		return 0;

	return 1;
}
L
Linus Torvalds 已提交
3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706

/* When incoming ACK allowed to free some skb from write_queue,
 * we remember this event in flag SOCK_QUEUE_SHRUNK and wake up socket
 * on the exit from tcp input handler.
 *
 * PROBLEM: sndbuf expansion does not work well with largesend.
 */
static void tcp_new_space(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);

3707
	if (tcp_should_expand_sndbuf(sk, tp)) {
3708
		int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) +
L
Linus Torvalds 已提交
3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720
			MAX_TCP_HEADER + 16 + sizeof(struct sk_buff),
		    demanded = max_t(unsigned int, tp->snd_cwnd,
						   tp->reordering + 1);
		sndmem *= 2*demanded;
		if (sndmem > sk->sk_sndbuf)
			sk->sk_sndbuf = min(sndmem, sysctl_tcp_wmem[2]);
		tp->snd_cwnd_stamp = tcp_time_stamp;
	}

	sk->sk_write_space(sk);
}

S
Stephen Hemminger 已提交
3721
static void tcp_check_space(struct sock *sk)
L
Linus Torvalds 已提交
3722 3723 3724 3725 3726 3727 3728 3729 3730
{
	if (sock_flag(sk, SOCK_QUEUE_SHRUNK)) {
		sock_reset_flag(sk, SOCK_QUEUE_SHRUNK);
		if (sk->sk_socket &&
		    test_bit(SOCK_NOSPACE, &sk->sk_socket->flags))
			tcp_new_space(sk);
	}
}

S
Stephen Hemminger 已提交
3731
static inline void tcp_data_snd_check(struct sock *sk, struct tcp_sock *tp)
L
Linus Torvalds 已提交
3732
{
3733
	tcp_push_pending_frames(sk, tp);
L
Linus Torvalds 已提交
3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744
	tcp_check_space(sk);
}

/*
 * Check if sending an ack is needed.
 */
static void __tcp_ack_snd_check(struct sock *sk, int ofo_possible)
{
	struct tcp_sock *tp = tcp_sk(sk);

	    /* More than one full frame received... */
3745
	if (((tp->rcv_nxt - tp->rcv_wup) > inet_csk(sk)->icsk_ack.rcv_mss
L
Linus Torvalds 已提交
3746 3747 3748 3749 3750
	     /* ... and right edge of window advances far enough.
	      * (tcp_recvmsg() will send ACK otherwise). Or...
	      */
	     && __tcp_select_window(sk) >= tp->rcv_wnd) ||
	    /* We ACK each frame or... */
3751
	    tcp_in_quickack_mode(sk) ||
L
Linus Torvalds 已提交
3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762
	    /* We have out of order data. */
	    (ofo_possible &&
	     skb_peek(&tp->out_of_order_queue))) {
		/* Then ack it now */
		tcp_send_ack(sk);
	} else {
		/* Else, send delayed ack. */
		tcp_send_delayed_ack(sk);
	}
}

S
Stephen Hemminger 已提交
3763
static inline void tcp_ack_snd_check(struct sock *sk)
L
Linus Torvalds 已提交
3764
{
3765
	if (!inet_csk_ack_scheduled(sk)) {
L
Linus Torvalds 已提交
3766 3767 3768 3769 3770 3771 3772 3773
		/* We sent a data segment already. */
		return;
	}
	__tcp_ack_snd_check(sk, 1);
}

/*
 *	This routine is only called when we have urgent data
S
Stephen Hemminger 已提交
3774
 *	signaled. Its the 'slow' part of tcp_urg. It could be
L
Linus Torvalds 已提交
3775 3776 3777 3778 3779 3780
 *	moved inline now as tcp_urg is only called from one
 *	place. We handle URGent data wrong. We have to - as
 *	BSD still doesn't use the correction from RFC961.
 *	For 1003.1g we should support a new option TCP_STDURG to permit
 *	either form (or just set the sysctl tcp_stdurg).
 */
3781

L
Linus Torvalds 已提交
3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818
static void tcp_check_urg(struct sock * sk, struct tcphdr * th)
{
	struct tcp_sock *tp = tcp_sk(sk);
	u32 ptr = ntohs(th->urg_ptr);

	if (ptr && !sysctl_tcp_stdurg)
		ptr--;
	ptr += ntohl(th->seq);

	/* Ignore urgent data that we've already seen and read. */
	if (after(tp->copied_seq, ptr))
		return;

	/* Do not replay urg ptr.
	 *
	 * NOTE: interesting situation not covered by specs.
	 * Misbehaving sender may send urg ptr, pointing to segment,
	 * which we already have in ofo queue. We are not able to fetch
	 * such data and will stay in TCP_URG_NOTYET until will be eaten
	 * by recvmsg(). Seems, we are not obliged to handle such wicked
	 * situations. But it is worth to think about possibility of some
	 * DoSes using some hypothetical application level deadlock.
	 */
	if (before(ptr, tp->rcv_nxt))
		return;

	/* Do we already have a newer (or duplicate) urgent pointer? */
	if (tp->urg_data && !after(ptr, tp->urg_seq))
		return;

	/* Tell the world about our new urgent pointer. */
	sk_send_sigurg(sk);

	/* We may be adding urgent data when the last byte read was
	 * urgent. To do this requires some care. We cannot just ignore
	 * tp->copied_seq since we would read the last urgent byte again
	 * as data, nor can we alter copied_seq until this data arrives
S
Stephen Hemminger 已提交
3819
	 * or we break the semantics of SIOCATMARK (and thus sockatmark())
L
Linus Torvalds 已提交
3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835
	 *
	 * NOTE. Double Dutch. Rendering to plain English: author of comment
	 * above did something sort of 	send("A", MSG_OOB); send("B", MSG_OOB);
	 * and expect that both A and B disappear from stream. This is _wrong_.
	 * Though this happens in BSD with high probability, this is occasional.
	 * Any application relying on this is buggy. Note also, that fix "works"
	 * only in this artificial test. Insert some normal data between A and B and we will
	 * decline of BSD again. Verdict: it is better to remove to trap
	 * buggy users.
	 */
	if (tp->urg_seq == tp->copied_seq && tp->urg_data &&
	    !sock_flag(sk, SOCK_URGINLINE) &&
	    tp->copied_seq != tp->rcv_nxt) {
		struct sk_buff *skb = skb_peek(&sk->sk_receive_queue);
		tp->copied_seq++;
		if (skb && !before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq)) {
D
David S. Miller 已提交
3836
			__skb_unlink(skb, &sk->sk_receive_queue);
L
Linus Torvalds 已提交
3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861
			__kfree_skb(skb);
		}
	}

	tp->urg_data   = TCP_URG_NOTYET;
	tp->urg_seq    = ptr;

	/* Disable header prediction. */
	tp->pred_flags = 0;
}

/* This is the 'fast' part of urgent handling. */
static void tcp_urg(struct sock *sk, struct sk_buff *skb, struct tcphdr *th)
{
	struct tcp_sock *tp = tcp_sk(sk);

	/* Check if we get a new urgent pointer - normally not. */
	if (th->urg)
		tcp_check_urg(sk,th);

	/* Do we wait for any urgent data? - normally not... */
	if (tp->urg_data == TCP_URG_NOTYET) {
		u32 ptr = tp->urg_seq - ntohl(th->seq) + (th->doff * 4) -
			  th->syn;

3862
		/* Is the urgent pointer pointing into this packet? */
L
Linus Torvalds 已提交
3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896
		if (ptr < skb->len) {
			u8 tmp;
			if (skb_copy_bits(skb, ptr, &tmp, 1))
				BUG();
			tp->urg_data = TCP_URG_VALID | tmp;
			if (!sock_flag(sk, SOCK_DEAD))
				sk->sk_data_ready(sk, 0);
		}
	}
}

static int tcp_copy_to_iovec(struct sock *sk, struct sk_buff *skb, int hlen)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int chunk = skb->len - hlen;
	int err;

	local_bh_enable();
	if (skb->ip_summed==CHECKSUM_UNNECESSARY)
		err = skb_copy_datagram_iovec(skb, hlen, tp->ucopy.iov, chunk);
	else
		err = skb_copy_and_csum_datagram_iovec(skb, hlen,
						       tp->ucopy.iov);

	if (!err) {
		tp->ucopy.len -= chunk;
		tp->copied_seq += chunk;
		tcp_rcv_space_adjust(sk);
	}

	local_bh_disable();
	return err;
}

3897
static __sum16 __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb)
L
Linus Torvalds 已提交
3898
{
3899
	__sum16 result;
L
Linus Torvalds 已提交
3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910

	if (sock_owned_by_user(sk)) {
		local_bh_enable();
		result = __tcp_checksum_complete(skb);
		local_bh_disable();
	} else {
		result = __tcp_checksum_complete(skb);
	}
	return result;
}

S
Stephen Hemminger 已提交
3911
static inline int tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb)
L
Linus Torvalds 已提交
3912 3913 3914 3915 3916
{
	return skb->ip_summed != CHECKSUM_UNNECESSARY &&
		__tcp_checksum_complete_user(sk, skb);
}

C
Chris Leech 已提交
3917 3918 3919 3920 3921 3922 3923 3924 3925
#ifdef CONFIG_NET_DMA
static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, int hlen)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int chunk = skb->len - hlen;
	int dma_cookie;
	int copied_early = 0;

	if (tp->ucopy.wakeup)
3926
		return 0;
C
Chris Leech 已提交
3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960

	if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
		tp->ucopy.dma_chan = get_softnet_dma();

	if (tp->ucopy.dma_chan && skb->ip_summed == CHECKSUM_UNNECESSARY) {

		dma_cookie = dma_skb_copy_datagram_iovec(tp->ucopy.dma_chan,
			skb, hlen, tp->ucopy.iov, chunk, tp->ucopy.pinned_list);

		if (dma_cookie < 0)
			goto out;

		tp->ucopy.dma_cookie = dma_cookie;
		copied_early = 1;

		tp->ucopy.len -= chunk;
		tp->copied_seq += chunk;
		tcp_rcv_space_adjust(sk);

		if ((tp->ucopy.len == 0) ||
		    (tcp_flag_word(skb->h.th) & TCP_FLAG_PSH) ||
		    (atomic_read(&sk->sk_rmem_alloc) > (sk->sk_rcvbuf >> 1))) {
			tp->ucopy.wakeup = 1;
			sk->sk_data_ready(sk, 0);
		}
	} else if (chunk > 0) {
		tp->ucopy.wakeup = 1;
		sk->sk_data_ready(sk, 0);
	}
out:
	return copied_early;
}
#endif /* CONFIG_NET_DMA */

L
Linus Torvalds 已提交
3961
/*
3962
 *	TCP receive function for the ESTABLISHED state.
L
Linus Torvalds 已提交
3963
 *
3964
 *	It is split into a fast path and a slow path. The fast path is
L
Linus Torvalds 已提交
3965 3966
 * 	disabled when:
 *	- A zero window was announced from us - zero window probing
3967
 *        is only handled properly in the slow path.
L
Linus Torvalds 已提交
3968 3969 3970 3971
 *	- Out of order segments arrived.
 *	- Urgent data is expected.
 *	- There is no buffer space left
 *	- Unexpected TCP flags/window values/header lengths are received
3972
 *	  (detected by checking the TCP header against pred_flags)
L
Linus Torvalds 已提交
3973 3974 3975 3976 3977
 *	- Data is sent in both directions. Fast path only supports pure senders
 *	  or pure receivers (this means either the sequence number or the ack
 *	  value must stay constant)
 *	- Unexpected TCP option.
 *
3978
 *	When these conditions are not satisfied it drops into a standard
L
Linus Torvalds 已提交
3979 3980
 *	receive procedure patterned after RFC793 to handle all cases.
 *	The first three cases are guaranteed by proper pred_flags setting,
3981
 *	the rest is checked inline. Fast processing is turned on in
L
Linus Torvalds 已提交
3982 3983 3984 3985 3986 3987 3988 3989 3990
 *	tcp_data_queue when everything is OK.
 */
int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
			struct tcphdr *th, unsigned len)
{
	struct tcp_sock *tp = tcp_sk(sk);

	/*
	 *	Header prediction.
3991
	 *	The code loosely follows the one in the famous
L
Linus Torvalds 已提交
3992
	 *	"30 instruction TCP receive" Van Jacobson mail.
3993 3994
	 *
	 *	Van's trick is to deposit buffers into socket queue
L
Linus Torvalds 已提交
3995 3996 3997 3998
	 *	on a device interrupt, to call tcp_recv function
	 *	on the receive process context and checksum and copy
	 *	the buffer to user space. smart...
	 *
3999
	 *	Our current scheme is not silly either but we take the
L
Linus Torvalds 已提交
4000 4001 4002 4003 4004 4005 4006
	 *	extra cost of the net_bh soft interrupt processing...
	 *	We do checksum and copy also but from device to kernel.
	 */

	tp->rx_opt.saw_tstamp = 0;

	/*	pred_flags is 0xS?10 << 16 + snd_wnd
S
Stephen Hemminger 已提交
4007
	 *	if header_prediction is to be made
L
Linus Torvalds 已提交
4008 4009
	 *	'S' will always be tp->tcp_header_len >> 2
	 *	'?' will be 0 for the fast path, otherwise pred_flags is 0 to
4010
	 *  turn it off	(when there are holes in the receive
L
Linus Torvalds 已提交
4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025
	 *	 space for instance)
	 *	PSH flag is ignored.
	 */

	if ((tcp_flag_word(th) & TCP_HP_BITS) == tp->pred_flags &&
		TCP_SKB_CB(skb)->seq == tp->rcv_nxt) {
		int tcp_header_len = tp->tcp_header_len;

		/* Timestamp header prediction: tcp_header_len
		 * is automatically equal to th->doff*4 due to pred_flags
		 * match.
		 */

		/* Check timestamp */
		if (tcp_header_len == sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) {
4026
			__be32 *ptr = (__be32 *)(th + 1);
L
Linus Torvalds 已提交
4027 4028

			/* No? Slow path! */
4029
			if (*ptr != htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
L
Linus Torvalds 已提交
4030 4031 4032 4033
					  | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP))
				goto slow_path;

			tp->rx_opt.saw_tstamp = 1;
4034
			++ptr;
L
Linus Torvalds 已提交
4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065
			tp->rx_opt.rcv_tsval = ntohl(*ptr);
			++ptr;
			tp->rx_opt.rcv_tsecr = ntohl(*ptr);

			/* If PAWS failed, check it more carefully in slow path */
			if ((s32)(tp->rx_opt.rcv_tsval - tp->rx_opt.ts_recent) < 0)
				goto slow_path;

			/* DO NOT update ts_recent here, if checksum fails
			 * and timestamp was corrupted part, it will result
			 * in a hung connection since we will drop all
			 * future packets due to the PAWS test.
			 */
		}

		if (len <= tcp_header_len) {
			/* Bulk data transfer: sender */
			if (len == tcp_header_len) {
				/* Predicted packet is in window by definition.
				 * seq == rcv_nxt and rcv_wup <= rcv_nxt.
				 * Hence, check seq<=rcv_wup reduces to:
				 */
				if (tcp_header_len ==
				    (sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) &&
				    tp->rcv_nxt == tp->rcv_wup)
					tcp_store_ts_recent(tp);

				/* We know that such packets are checksummed
				 * on entry.
				 */
				tcp_ack(sk, skb, 0);
4066
				__kfree_skb(skb);
4067
				tcp_data_snd_check(sk, tp);
L
Linus Torvalds 已提交
4068 4069 4070 4071 4072 4073 4074
				return 0;
			} else { /* Header too small */
				TCP_INC_STATS_BH(TCP_MIB_INERRS);
				goto discard;
			}
		} else {
			int eaten = 0;
C
Chris Leech 已提交
4075
			int copied_early = 0;
L
Linus Torvalds 已提交
4076

C
Chris Leech 已提交
4077 4078 4079 4080 4081 4082 4083 4084 4085 4086
			if (tp->copied_seq == tp->rcv_nxt &&
			    len - tcp_header_len <= tp->ucopy.len) {
#ifdef CONFIG_NET_DMA
				if (tcp_dma_try_early_copy(sk, skb, tcp_header_len)) {
					copied_early = 1;
					eaten = 1;
				}
#endif
				if (tp->ucopy.task == current && sock_owned_by_user(sk) && !copied_early) {
					__set_current_state(TASK_RUNNING);
L
Linus Torvalds 已提交
4087

C
Chris Leech 已提交
4088 4089 4090 4091
					if (!tcp_copy_to_iovec(sk, skb, tcp_header_len))
						eaten = 1;
				}
				if (eaten) {
L
Linus Torvalds 已提交
4092 4093 4094 4095 4096 4097 4098 4099 4100 4101
					/* Predicted packet is in window by definition.
					 * seq == rcv_nxt and rcv_wup <= rcv_nxt.
					 * Hence, check seq<=rcv_wup reduces to:
					 */
					if (tcp_header_len ==
					    (sizeof(struct tcphdr) +
					     TCPOLEN_TSTAMP_ALIGNED) &&
					    tp->rcv_nxt == tp->rcv_wup)
						tcp_store_ts_recent(tp);

4102
					tcp_rcv_rtt_measure_ts(sk, skb);
L
Linus Torvalds 已提交
4103 4104 4105 4106 4107

					__skb_pull(skb, tcp_header_len);
					tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
					NET_INC_STATS_BH(LINUX_MIB_TCPHPHITSTOUSER);
				}
C
Chris Leech 已提交
4108 4109
				if (copied_early)
					tcp_cleanup_rbuf(sk, skb->len);
L
Linus Torvalds 已提交
4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123
			}
			if (!eaten) {
				if (tcp_checksum_complete_user(sk, skb))
					goto csum_error;

				/* Predicted packet is in window by definition.
				 * seq == rcv_nxt and rcv_wup <= rcv_nxt.
				 * Hence, check seq<=rcv_wup reduces to:
				 */
				if (tcp_header_len ==
				    (sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) &&
				    tp->rcv_nxt == tp->rcv_wup)
					tcp_store_ts_recent(tp);

4124
				tcp_rcv_rtt_measure_ts(sk, skb);
L
Linus Torvalds 已提交
4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142

				if ((int)skb->truesize > sk->sk_forward_alloc)
					goto step5;

				NET_INC_STATS_BH(LINUX_MIB_TCPHPHITS);

				/* Bulk data transfer: receiver */
				__skb_pull(skb,tcp_header_len);
				__skb_queue_tail(&sk->sk_receive_queue, skb);
				sk_stream_set_owner_r(skb, sk);
				tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
			}

			tcp_event_data_recv(sk, tp, skb);

			if (TCP_SKB_CB(skb)->ack_seq != tp->snd_una) {
				/* Well, only one small jumplet in fast path... */
				tcp_ack(sk, skb, FLAG_DATA);
4143
				tcp_data_snd_check(sk, tp);
4144
				if (!inet_csk_ack_scheduled(sk))
L
Linus Torvalds 已提交
4145 4146 4147
					goto no_ack;
			}

4148
			__tcp_ack_snd_check(sk, 0);
L
Linus Torvalds 已提交
4149
no_ack:
C
Chris Leech 已提交
4150 4151 4152 4153 4154
#ifdef CONFIG_NET_DMA
			if (copied_early)
				__skb_queue_tail(&sk->sk_async_wait_queue, skb);
			else
#endif
L
Linus Torvalds 已提交
4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170
			if (eaten)
				__kfree_skb(skb);
			else
				sk->sk_data_ready(sk, 0);
			return 0;
		}
	}

slow_path:
	if (len < (th->doff<<2) || tcp_checksum_complete_user(sk, skb))
		goto csum_error;

	/*
	 * RFC1323: H1. Apply PAWS check first.
	 */
	if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
4171
	    tcp_paws_discard(sk, skb)) {
L
Linus Torvalds 已提交
4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217
		if (!th->rst) {
			NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);
			tcp_send_dupack(sk, skb);
			goto discard;
		}
		/* Resets are accepted even if PAWS failed.

		   ts_recent update must be made after we are sure
		   that the packet is in window.
		 */
	}

	/*
	 *	Standard slow path.
	 */

	if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
		/* RFC793, page 37: "In all states except SYN-SENT, all reset
		 * (RST) segments are validated by checking their SEQ-fields."
		 * And page 69: "If an incoming segment is not acceptable,
		 * an acknowledgment should be sent in reply (unless the RST bit
		 * is set, if so drop the segment and return)".
		 */
		if (!th->rst)
			tcp_send_dupack(sk, skb);
		goto discard;
	}

	if(th->rst) {
		tcp_reset(sk);
		goto discard;
	}

	tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);

	if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
		TCP_INC_STATS_BH(TCP_MIB_INERRS);
		NET_INC_STATS_BH(LINUX_MIB_TCPABORTONSYN);
		tcp_reset(sk);
		return 1;
	}

step5:
	if(th->ack)
		tcp_ack(sk, skb, FLAG_SLOWPATH);

4218
	tcp_rcv_rtt_measure_ts(sk, skb);
L
Linus Torvalds 已提交
4219 4220 4221 4222 4223 4224 4225

	/* Process urgent data. */
	tcp_urg(sk, skb, th);

	/* step 7: process the segment text */
	tcp_data_queue(sk, skb);

4226
	tcp_data_snd_check(sk, tp);
L
Linus Torvalds 已提交
4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241
	tcp_ack_snd_check(sk);
	return 0;

csum_error:
	TCP_INC_STATS_BH(TCP_MIB_INERRS);

discard:
	__kfree_skb(skb);
	return 0;
}

static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
					 struct tcphdr *th, unsigned len)
{
	struct tcp_sock *tp = tcp_sk(sk);
4242
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333
	int saved_clamp = tp->rx_opt.mss_clamp;

	tcp_parse_options(skb, &tp->rx_opt, 0);

	if (th->ack) {
		/* rfc793:
		 * "If the state is SYN-SENT then
		 *    first check the ACK bit
		 *      If the ACK bit is set
		 *	  If SEG.ACK =< ISS, or SEG.ACK > SND.NXT, send
		 *        a reset (unless the RST bit is set, if so drop
		 *        the segment and return)"
		 *
		 *  We do not send data with SYN, so that RFC-correct
		 *  test reduces to:
		 */
		if (TCP_SKB_CB(skb)->ack_seq != tp->snd_nxt)
			goto reset_and_undo;

		if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr &&
		    !between(tp->rx_opt.rcv_tsecr, tp->retrans_stamp,
			     tcp_time_stamp)) {
			NET_INC_STATS_BH(LINUX_MIB_PAWSACTIVEREJECTED);
			goto reset_and_undo;
		}

		/* Now ACK is acceptable.
		 *
		 * "If the RST bit is set
		 *    If the ACK was acceptable then signal the user "error:
		 *    connection reset", drop the segment, enter CLOSED state,
		 *    delete TCB, and return."
		 */

		if (th->rst) {
			tcp_reset(sk);
			goto discard;
		}

		/* rfc793:
		 *   "fifth, if neither of the SYN or RST bits is set then
		 *    drop the segment and return."
		 *
		 *    See note below!
		 *                                        --ANK(990513)
		 */
		if (!th->syn)
			goto discard_and_undo;

		/* rfc793:
		 *   "If the SYN bit is on ...
		 *    are acceptable then ...
		 *    (our SYN has been ACKed), change the connection
		 *    state to ESTABLISHED..."
		 */

		TCP_ECN_rcv_synack(tp, th);

		tp->snd_wl1 = TCP_SKB_CB(skb)->seq;
		tcp_ack(sk, skb, FLAG_SLOWPATH);

		/* Ok.. it's good. Set up sequence numbers and
		 * move to established.
		 */
		tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1;
		tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1;

		/* RFC1323: The window in SYN & SYN/ACK segments is
		 * never scaled.
		 */
		tp->snd_wnd = ntohs(th->window);
		tcp_init_wl(tp, TCP_SKB_CB(skb)->ack_seq, TCP_SKB_CB(skb)->seq);

		if (!tp->rx_opt.wscale_ok) {
			tp->rx_opt.snd_wscale = tp->rx_opt.rcv_wscale = 0;
			tp->window_clamp = min(tp->window_clamp, 65535U);
		}

		if (tp->rx_opt.saw_tstamp) {
			tp->rx_opt.tstamp_ok	   = 1;
			tp->tcp_header_len =
				sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED;
			tp->advmss	    -= TCPOLEN_TSTAMP_ALIGNED;
			tcp_store_ts_recent(tp);
		} else {
			tp->tcp_header_len = sizeof(struct tcphdr);
		}

		if (tp->rx_opt.sack_ok && sysctl_tcp_fack)
			tp->rx_opt.sack_ok |= 2;

J
John Heffner 已提交
4334
		tcp_mtup_init(sk);
4335
		tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
L
Linus Torvalds 已提交
4336 4337 4338 4339 4340 4341
		tcp_initialize_rcv_mss(sk);

		/* Remember, tcp_poll() does not lock socket!
		 * Change state from SYN-SENT only after copied_seq
		 * is initialized. */
		tp->copied_seq = tp->rcv_nxt;
R
Ralf Baechle 已提交
4342
		smp_mb();
L
Linus Torvalds 已提交
4343 4344
		tcp_set_state(sk, TCP_ESTABLISHED);

4345 4346
		security_inet_conn_established(sk, skb);

L
Linus Torvalds 已提交
4347
		/* Make sure socket is routed, for correct metrics.  */
4348
		icsk->icsk_af_ops->rebuild_header(sk);
L
Linus Torvalds 已提交
4349 4350 4351

		tcp_init_metrics(sk);

4352
		tcp_init_congestion_control(sk);
4353

L
Linus Torvalds 已提交
4354 4355 4356 4357 4358 4359 4360 4361
		/* Prevent spurious tcp_cwnd_restart() on first data
		 * packet.
		 */
		tp->lsndtime = tcp_time_stamp;

		tcp_init_buffer_space(sk);

		if (sock_flag(sk, SOCK_KEEPOPEN))
4362
			inet_csk_reset_keepalive_timer(sk, keepalive_time_when(tp));
L
Linus Torvalds 已提交
4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373

		if (!tp->rx_opt.snd_wscale)
			__tcp_fast_path_on(tp, tp->snd_wnd);
		else
			tp->pred_flags = 0;

		if (!sock_flag(sk, SOCK_DEAD)) {
			sk->sk_state_change(sk);
			sk_wake_async(sk, 0, POLL_OUT);
		}

4374 4375 4376
		if (sk->sk_write_pending ||
		    icsk->icsk_accept_queue.rskq_defer_accept ||
		    icsk->icsk_ack.pingpong) {
L
Linus Torvalds 已提交
4377 4378 4379 4380 4381 4382 4383
			/* Save one ACK. Data will be ready after
			 * several ticks, if write_pending is set.
			 *
			 * It may be deleted, but with this feature tcpdumps
			 * look so _wonderfully_ clever, that I was not able
			 * to stand against the temptation 8)     --ANK
			 */
4384
			inet_csk_schedule_ack(sk);
4385 4386
			icsk->icsk_ack.lrcvtime = tcp_time_stamp;
			icsk->icsk_ack.ato	 = TCP_ATO_MIN;
4387 4388
			tcp_incr_quickack(sk);
			tcp_enter_quickack_mode(sk);
4389 4390
			inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
						  TCP_DELACK_MAX, TCP_RTO_MAX);
L
Linus Torvalds 已提交
4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444

discard:
			__kfree_skb(skb);
			return 0;
		} else {
			tcp_send_ack(sk);
		}
		return -1;
	}

	/* No ACK in the segment */

	if (th->rst) {
		/* rfc793:
		 * "If the RST bit is set
		 *
		 *      Otherwise (no ACK) drop the segment and return."
		 */

		goto discard_and_undo;
	}

	/* PAWS check. */
	if (tp->rx_opt.ts_recent_stamp && tp->rx_opt.saw_tstamp && tcp_paws_check(&tp->rx_opt, 0))
		goto discard_and_undo;

	if (th->syn) {
		/* We see SYN without ACK. It is attempt of
		 * simultaneous connect with crossed SYNs.
		 * Particularly, it can be connect to self.
		 */
		tcp_set_state(sk, TCP_SYN_RECV);

		if (tp->rx_opt.saw_tstamp) {
			tp->rx_opt.tstamp_ok = 1;
			tcp_store_ts_recent(tp);
			tp->tcp_header_len =
				sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED;
		} else {
			tp->tcp_header_len = sizeof(struct tcphdr);
		}

		tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1;
		tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1;

		/* RFC1323: The window in SYN & SYN/ACK segments is
		 * never scaled.
		 */
		tp->snd_wnd    = ntohs(th->window);
		tp->snd_wl1    = TCP_SKB_CB(skb)->seq;
		tp->max_window = tp->snd_wnd;

		TCP_ECN_rcv_syn(tp, th);

J
John Heffner 已提交
4445
		tcp_mtup_init(sk);
4446
		tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
L
Linus Torvalds 已提交
4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483
		tcp_initialize_rcv_mss(sk);


		tcp_send_synack(sk);
#if 0
		/* Note, we could accept data and URG from this segment.
		 * There are no obstacles to make this.
		 *
		 * However, if we ignore data in ACKless segments sometimes,
		 * we have no reasons to accept it sometimes.
		 * Also, seems the code doing it in step6 of tcp_rcv_state_process
		 * is not flawless. So, discard packet for sanity.
		 * Uncomment this return to process the data.
		 */
		return -1;
#else
		goto discard;
#endif
	}
	/* "fifth, if neither of the SYN or RST bits is set then
	 * drop the segment and return."
	 */

discard_and_undo:
	tcp_clear_options(&tp->rx_opt);
	tp->rx_opt.mss_clamp = saved_clamp;
	goto discard;

reset_and_undo:
	tcp_clear_options(&tp->rx_opt);
	tp->rx_opt.mss_clamp = saved_clamp;
	return 1;
}


/*
 *	This function implements the receiving procedure of RFC 793 for
4484
 *	all states except ESTABLISHED and TIME_WAIT.
L
Linus Torvalds 已提交
4485 4486 4487
 *	It's called from both tcp_v4_rcv and tcp_v6_rcv and should be
 *	address independent.
 */
4488

L
Linus Torvalds 已提交
4489 4490 4491 4492
int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
			  struct tcphdr *th, unsigned len)
{
	struct tcp_sock *tp = tcp_sk(sk);
4493
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509
	int queued = 0;

	tp->rx_opt.saw_tstamp = 0;

	switch (sk->sk_state) {
	case TCP_CLOSE:
		goto discard;

	case TCP_LISTEN:
		if(th->ack)
			return 1;

		if(th->rst)
			goto discard;

		if(th->syn) {
4510
			if (icsk->icsk_af_ops->conn_request(sk, skb) < 0)
L
Linus Torvalds 已提交
4511 4512
				return 1;

4513 4514
			/* Now we have several options: In theory there is
			 * nothing else in the frame. KA9Q has an option to
L
Linus Torvalds 已提交
4515
			 * send data with the syn, BSD accepts data with the
4516 4517 4518
			 * syn up to the [to be] advertised window and
			 * Solaris 2.1 gives you a protocol error. For now
			 * we just ignore it, that fits the spec precisely
L
Linus Torvalds 已提交
4519 4520 4521
			 * and avoids incompatibilities. It would be nice in
			 * future to drop through and process the data.
			 *
4522
			 * Now that TTCP is starting to be used we ought to
L
Linus Torvalds 已提交
4523 4524
			 * queue this data.
			 * But, this leaves one open to an easy denial of
4525
			 * service attack, and SYN cookies can't defend
L
Linus Torvalds 已提交
4526
			 * against this problem. So, we drop the data
4527 4528
			 * in the interest of security over speed unless
			 * it's still in use.
L
Linus Torvalds 已提交
4529
			 */
4530 4531
			kfree_skb(skb);
			return 0;
L
Linus Torvalds 已提交
4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542
		}
		goto discard;

	case TCP_SYN_SENT:
		queued = tcp_rcv_synsent_state_process(sk, skb, th, len);
		if (queued >= 0)
			return queued;

		/* Do step6 onward by hand. */
		tcp_urg(sk, skb, th);
		__kfree_skb(skb);
4543
		tcp_data_snd_check(sk, tp);
L
Linus Torvalds 已提交
4544 4545 4546 4547
		return 0;
	}

	if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
4548
	    tcp_paws_discard(sk, skb)) {
L
Linus Torvalds 已提交
4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591
		if (!th->rst) {
			NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);
			tcp_send_dupack(sk, skb);
			goto discard;
		}
		/* Reset is accepted even if it did not pass PAWS. */
	}

	/* step 1: check sequence number */
	if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
		if (!th->rst)
			tcp_send_dupack(sk, skb);
		goto discard;
	}

	/* step 2: check RST bit */
	if(th->rst) {
		tcp_reset(sk);
		goto discard;
	}

	tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);

	/* step 3: check security and precedence [ignored] */

	/*	step 4:
	 *
	 *	Check for a SYN in window.
	 */
	if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
		NET_INC_STATS_BH(LINUX_MIB_TCPABORTONSYN);
		tcp_reset(sk);
		return 1;
	}

	/* step 5: check the ACK field */
	if (th->ack) {
		int acceptable = tcp_ack(sk, skb, FLAG_SLOWPATH);

		switch(sk->sk_state) {
		case TCP_SYN_RECV:
			if (acceptable) {
				tp->copied_seq = tp->rcv_nxt;
R
Ralf Baechle 已提交
4592
				smp_mb();
L
Linus Torvalds 已提交
4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616
				tcp_set_state(sk, TCP_ESTABLISHED);
				sk->sk_state_change(sk);

				/* Note, that this wakeup is only for marginal
				 * crossed SYN case. Passively open sockets
				 * are not waked up, because sk->sk_sleep ==
				 * NULL and sk->sk_socket == NULL.
				 */
				if (sk->sk_socket) {
					sk_wake_async(sk,0,POLL_OUT);
				}

				tp->snd_una = TCP_SKB_CB(skb)->ack_seq;
				tp->snd_wnd = ntohs(th->window) <<
					      tp->rx_opt.snd_wscale;
				tcp_init_wl(tp, TCP_SKB_CB(skb)->ack_seq,
					    TCP_SKB_CB(skb)->seq);

				/* tcp_ack considers this ACK as duplicate
				 * and does not calculate rtt.
				 * Fix it at least with timestamps.
				 */
				if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr &&
				    !tp->srtt)
4617
					tcp_ack_saw_tstamp(sk, 0);
L
Linus Torvalds 已提交
4618 4619 4620 4621 4622 4623 4624

				if (tp->rx_opt.tstamp_ok)
					tp->advmss -= TCPOLEN_TSTAMP_ALIGNED;

				/* Make sure socket is routed, for
				 * correct metrics.
				 */
4625
				icsk->icsk_af_ops->rebuild_header(sk);
L
Linus Torvalds 已提交
4626 4627 4628

				tcp_init_metrics(sk);

4629
				tcp_init_congestion_control(sk);
4630

L
Linus Torvalds 已提交
4631 4632 4633 4634 4635
				/* Prevent spurious tcp_cwnd_restart() on
				 * first data packet.
				 */
				tp->lsndtime = tcp_time_stamp;

J
John Heffner 已提交
4636
				tcp_mtup_init(sk);
L
Linus Torvalds 已提交
4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664
				tcp_initialize_rcv_mss(sk);
				tcp_init_buffer_space(sk);
				tcp_fast_path_on(tp);
			} else {
				return 1;
			}
			break;

		case TCP_FIN_WAIT1:
			if (tp->snd_una == tp->write_seq) {
				tcp_set_state(sk, TCP_FIN_WAIT2);
				sk->sk_shutdown |= SEND_SHUTDOWN;
				dst_confirm(sk->sk_dst_cache);

				if (!sock_flag(sk, SOCK_DEAD))
					/* Wake up lingering close() */
					sk->sk_state_change(sk);
				else {
					int tmo;

					if (tp->linger2 < 0 ||
					    (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq &&
					     after(TCP_SKB_CB(skb)->end_seq - th->fin, tp->rcv_nxt))) {
						tcp_done(sk);
						NET_INC_STATS_BH(LINUX_MIB_TCPABORTONDATA);
						return 1;
					}

4665
					tmo = tcp_fin_time(sk);
L
Linus Torvalds 已提交
4666
					if (tmo > TCP_TIMEWAIT_LEN) {
4667
						inet_csk_reset_keepalive_timer(sk, tmo - TCP_TIMEWAIT_LEN);
L
Linus Torvalds 已提交
4668 4669 4670 4671 4672 4673 4674
					} else if (th->fin || sock_owned_by_user(sk)) {
						/* Bad case. We could lose such FIN otherwise.
						 * It is not a big problem, but it looks confusing
						 * and not so rare event. We still can lose it now,
						 * if it spins in bh_lock_sock(), but it is really
						 * marginal case.
						 */
4675
						inet_csk_reset_keepalive_timer(sk, tmo);
L
Linus Torvalds 已提交
4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714
					} else {
						tcp_time_wait(sk, TCP_FIN_WAIT2, tmo);
						goto discard;
					}
				}
			}
			break;

		case TCP_CLOSING:
			if (tp->snd_una == tp->write_seq) {
				tcp_time_wait(sk, TCP_TIME_WAIT, 0);
				goto discard;
			}
			break;

		case TCP_LAST_ACK:
			if (tp->snd_una == tp->write_seq) {
				tcp_update_metrics(sk);
				tcp_done(sk);
				goto discard;
			}
			break;
		}
	} else
		goto discard;

	/* step 6: check the URG bit */
	tcp_urg(sk, skb, th);

	/* step 7: process the segment text */
	switch (sk->sk_state) {
	case TCP_CLOSE_WAIT:
	case TCP_CLOSING:
	case TCP_LAST_ACK:
		if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt))
			break;
	case TCP_FIN_WAIT1:
	case TCP_FIN_WAIT2:
		/* RFC 793 says to queue data in these states,
4715
		 * RFC 1122 says we MUST send a reset.
L
Linus Torvalds 已提交
4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726
		 * BSD 4.4 also does reset.
		 */
		if (sk->sk_shutdown & RCV_SHUTDOWN) {
			if (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq &&
			    after(TCP_SKB_CB(skb)->end_seq - th->fin, tp->rcv_nxt)) {
				NET_INC_STATS_BH(LINUX_MIB_TCPABORTONDATA);
				tcp_reset(sk);
				return 1;
			}
		}
		/* Fall through */
4727
	case TCP_ESTABLISHED:
L
Linus Torvalds 已提交
4728 4729 4730 4731 4732 4733 4734
		tcp_data_queue(sk, skb);
		queued = 1;
		break;
	}

	/* tcp_data could move socket to TIME-WAIT */
	if (sk->sk_state != TCP_CLOSE) {
4735
		tcp_data_snd_check(sk, tp);
L
Linus Torvalds 已提交
4736 4737 4738
		tcp_ack_snd_check(sk);
	}

4739
	if (!queued) {
L
Linus Torvalds 已提交
4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750
discard:
		__kfree_skb(skb);
	}
	return 0;
}

EXPORT_SYMBOL(sysctl_tcp_ecn);
EXPORT_SYMBOL(sysctl_tcp_reordering);
EXPORT_SYMBOL(tcp_parse_options);
EXPORT_SYMBOL(tcp_rcv_established);
EXPORT_SYMBOL(tcp_rcv_state_process);
S
Stephen Hemminger 已提交
4751
EXPORT_SYMBOL(tcp_initialize_rcv_mss);