tcp_output.c 59.4 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6 7 8 9
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Implementation of the Transmission Control Protocol(TCP).
 *
 * Version:	$Id: tcp_output.c,v 1.146 2002/02/01 22:01:04 davem Exp $
 *
10
 * Authors:	Ross Biro
Linus Torvalds's avatar
Linus Torvalds committed
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
 *		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	:	Retransmit queue handled by TCP.
 *				:	Fragmentation on mtu decrease
 *				:	Segment collapse on retransmit
 *				:	AF independence
 *
 *		Linus Torvalds	:	send_delayed_ack
 *		David S. Miller	:	Charge memory using the right skb
 *					during syn/ack processing.
 *		David S. Miller :	Output engine completely rewritten.
 *		Andrea Arcangeli:	SYNACK carry ts_recent in tsecr.
 *		Cacophonix Gaul :	draft-minshall-nagle-01
 *		J Hadi Salim	:	ECN support
 *
 */

#include <net/tcp.h>

#include <linux/compiler.h>
#include <linux/module.h>
#include <linux/smp_lock.h>

/* People can turn this off for buggy TCP's found in printers etc. */
int sysctl_tcp_retrans_collapse = 1;

/* This limits the percentage of the congestion window which we
 * will allow a single TSO frame to consume.  Building TSO frames
 * which are too large can cause TCP streams to be bursty.
 */
52
int sysctl_tcp_tso_win_divisor = 3;
Linus Torvalds's avatar
Linus Torvalds committed
53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

static inline void update_send_head(struct sock *sk, struct tcp_sock *tp,
				    struct sk_buff *skb)
{
	sk->sk_send_head = skb->next;
	if (sk->sk_send_head == (struct sk_buff *)&sk->sk_write_queue)
		sk->sk_send_head = NULL;
	tp->snd_nxt = TCP_SKB_CB(skb)->end_seq;
	tcp_packets_out_inc(sk, tp, skb);
}

/* SND.NXT, if window was not shrunk.
 * If window has been shrunk, what should we make? It is not clear at all.
 * Using SND.UNA we will fail to open window, SND.NXT is out of window. :-(
 * Anything in between SND.UNA...SND.UNA+SND.WND also can be already
 * invalid. OK, let's make this for now:
 */
static inline __u32 tcp_acceptable_seq(struct sock *sk, struct tcp_sock *tp)
{
	if (!before(tp->snd_una+tp->snd_wnd, tp->snd_nxt))
		return tp->snd_nxt;
	else
		return tp->snd_una+tp->snd_wnd;
}

/* Calculate mss to advertise in SYN segment.
 * RFC1122, RFC1063, draft-ietf-tcpimpl-pmtud-01 state that:
 *
 * 1. It is independent of path mtu.
 * 2. Ideally, it is maximal possible segment size i.e. 65535-40.
 * 3. For IPv4 it is reasonable to calculate it from maximal MTU of
 *    attached devices, because some buggy hosts are confused by
 *    large MSS.
 * 4. We do not make 3, we advertise MSS, calculated from first
 *    hop device mtu, but allow to raise it to ip_rt_min_advmss.
 *    This may be overridden via information stored in routing table.
 * 5. Value 65535 for MSS is valid in IPv6 and means "as large as possible,
 *    probably even Jumbo".
 */
static __u16 tcp_advertise_mss(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct dst_entry *dst = __sk_dst_get(sk);
	int mss = tp->advmss;

	if (dst && dst_metric(dst, RTAX_ADVMSS) < mss) {
		mss = dst_metric(dst, RTAX_ADVMSS);
		tp->advmss = mss;
	}

	return (__u16)mss;
}

/* RFC2861. Reset CWND after idle period longer RTO to "restart window".
 * This is the first part of cwnd validation mechanism. */
108
static void tcp_cwnd_restart(struct sock *sk, struct dst_entry *dst)
Linus Torvalds's avatar
Linus Torvalds committed
109
{
110
	struct tcp_sock *tp = tcp_sk(sk);
Linus Torvalds's avatar
Linus Torvalds committed
111 112 113 114
	s32 delta = tcp_time_stamp - tp->lsndtime;
	u32 restart_cwnd = tcp_init_cwnd(tp, dst);
	u32 cwnd = tp->snd_cwnd;

115
	tcp_ca_event(sk, CA_EVENT_CWND_RESTART);
Linus Torvalds's avatar
Linus Torvalds committed
116

117
	tp->snd_ssthresh = tcp_current_ssthresh(sk);
Linus Torvalds's avatar
Linus Torvalds committed
118 119
	restart_cwnd = min(restart_cwnd, cwnd);

120
	while ((delta -= inet_csk(sk)->icsk_rto) > 0 && cwnd > restart_cwnd)
Linus Torvalds's avatar
Linus Torvalds committed
121 122 123 124 125 126 127 128 129
		cwnd >>= 1;
	tp->snd_cwnd = max(cwnd, restart_cwnd);
	tp->snd_cwnd_stamp = tcp_time_stamp;
	tp->snd_cwnd_used = 0;
}

static inline void tcp_event_data_sent(struct tcp_sock *tp,
				       struct sk_buff *skb, struct sock *sk)
{
130 131
	struct inet_connection_sock *icsk = inet_csk(sk);
	const u32 now = tcp_time_stamp;
Linus Torvalds's avatar
Linus Torvalds committed
132

133 134
	if (!tp->packets_out && (s32)(now - tp->lsndtime) > icsk->icsk_rto)
		tcp_cwnd_restart(sk, __sk_dst_get(sk));
Linus Torvalds's avatar
Linus Torvalds committed
135 136 137 138 139 140

	tp->lsndtime = now;

	/* If it is a reply for ato after last received
	 * packet, enter pingpong mode.
	 */
141 142
	if ((u32)(now - icsk->icsk_ack.lrcvtime) < icsk->icsk_ack.ato)
		icsk->icsk_ack.pingpong = 1;
Linus Torvalds's avatar
Linus Torvalds committed
143 144
}

145
static __inline__ void tcp_event_ack_sent(struct sock *sk, unsigned int pkts)
Linus Torvalds's avatar
Linus Torvalds committed
146
{
147 148
	tcp_dec_quickack_mode(sk, pkts);
	inet_csk_clear_xmit_timer(sk, ICSK_TIME_DACK);
Linus Torvalds's avatar
Linus Torvalds committed
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192
}

/* Determine a window scaling and initial window to offer.
 * Based on the assumption that the given amount of space
 * will be offered. Store the results in the tp structure.
 * NOTE: for smooth operation initial space offering should
 * be a multiple of mss if possible. We assume here that mss >= 1.
 * This MUST be enforced by all callers.
 */
void tcp_select_initial_window(int __space, __u32 mss,
			       __u32 *rcv_wnd, __u32 *window_clamp,
			       int wscale_ok, __u8 *rcv_wscale)
{
	unsigned int space = (__space < 0 ? 0 : __space);

	/* If no clamp set the clamp to the max possible scaled window */
	if (*window_clamp == 0)
		(*window_clamp) = (65535 << 14);
	space = min(*window_clamp, space);

	/* Quantize space offering to a multiple of mss if possible. */
	if (space > mss)
		space = (space / mss) * mss;

	/* NOTE: offering an initial window larger than 32767
	 * will break some buggy TCP stacks. We try to be nice.
	 * If we are not window scaling, then this truncates
	 * our initial window offering to 32k. There should also
	 * be a sysctl option to stop being nice.
	 */
	(*rcv_wnd) = min(space, MAX_TCP_WINDOW);
	(*rcv_wscale) = 0;
	if (wscale_ok) {
		/* Set window scaling on max possible window
		 * See RFC1323 for an explanation of the limit to 14 
		 */
		space = max_t(u32, sysctl_tcp_rmem[2], sysctl_rmem_max);
		while (space > 65535 && (*rcv_wscale) < 14) {
			space >>= 1;
			(*rcv_wscale)++;
		}
	}

	/* Set initial window to value enough for senders,
193
	 * following RFC2414. Senders, not following this RFC,
Linus Torvalds's avatar
Linus Torvalds committed
194 195 196
	 * will be satisfied with 2.
	 */
	if (mss > (1<<*rcv_wscale)) {
197 198
		int init_cwnd = 4;
		if (mss > 1460*3)
Linus Torvalds's avatar
Linus Torvalds committed
199
			init_cwnd = 2;
200 201
		else if (mss > 1460)
			init_cwnd = 3;
Linus Torvalds's avatar
Linus Torvalds committed
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267
		if (*rcv_wnd > init_cwnd*mss)
			*rcv_wnd = init_cwnd*mss;
	}

	/* Set the clamp no higher than max representable value */
	(*window_clamp) = min(65535U << (*rcv_wscale), *window_clamp);
}

/* Chose a new window to advertise, update state in tcp_sock for the
 * socket, and return result with RFC1323 scaling applied.  The return
 * value can be stuffed directly into th->window for an outgoing
 * frame.
 */
static __inline__ u16 tcp_select_window(struct sock *sk)
{
	struct tcp_sock *tp = tcp_sk(sk);
	u32 cur_win = tcp_receive_window(tp);
	u32 new_win = __tcp_select_window(sk);

	/* Never shrink the offered window */
	if(new_win < cur_win) {
		/* Danger Will Robinson!
		 * Don't update rcv_wup/rcv_wnd here or else
		 * we will not be able to advertise a zero
		 * window in time.  --DaveM
		 *
		 * Relax Will Robinson.
		 */
		new_win = cur_win;
	}
	tp->rcv_wnd = new_win;
	tp->rcv_wup = tp->rcv_nxt;

	/* Make sure we do not exceed the maximum possible
	 * scaled window.
	 */
	if (!tp->rx_opt.rcv_wscale)
		new_win = min(new_win, MAX_TCP_WINDOW);
	else
		new_win = min(new_win, (65535U << tp->rx_opt.rcv_wscale));

	/* RFC1323 scaling applied */
	new_win >>= tp->rx_opt.rcv_wscale;

	/* If we advertise zero window, disable fast path. */
	if (new_win == 0)
		tp->pred_flags = 0;

	return new_win;
}


/* This routine actually transmits TCP packets queued in by
 * tcp_do_sendmsg().  This is used by both the initial
 * transmission and possible later retransmissions.
 * All SKB's seen here are completely headerless.  It is our
 * job to build the TCP header, and pass the packet down to
 * IP so it can do the same plus pass the packet off to the
 * device.
 *
 * We are working here with either a clone of the original
 * SKB, or a fresh unique copy made by the retransmit engine.
 */
static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb)
{
	if (skb != NULL) {
268
		const struct inet_connection_sock *icsk = inet_csk(sk);
Linus Torvalds's avatar
Linus Torvalds committed
269 270 271 272 273 274 275 276 277 278 279 280 281 282
		struct inet_sock *inet = inet_sk(sk);
		struct tcp_sock *tp = tcp_sk(sk);
		struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
		int tcp_header_size = tp->tcp_header_len;
		struct tcphdr *th;
		int sysctl_flags;
		int err;

		BUG_ON(!tcp_skb_pcount(skb));

#define SYSCTL_FLAG_TSTAMPS	0x1
#define SYSCTL_FLAG_WSCALE	0x2
#define SYSCTL_FLAG_SACK	0x4

283
		/* If congestion control is doing timestamping */
284
		if (icsk->icsk_ca_ops->rtt_sample)
285
			__net_timestamp(skb);
286

Linus Torvalds's avatar
Linus Torvalds committed
287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
		sysctl_flags = 0;
		if (tcb->flags & TCPCB_FLAG_SYN) {
			tcp_header_size = sizeof(struct tcphdr) + TCPOLEN_MSS;
			if(sysctl_tcp_timestamps) {
				tcp_header_size += TCPOLEN_TSTAMP_ALIGNED;
				sysctl_flags |= SYSCTL_FLAG_TSTAMPS;
			}
			if(sysctl_tcp_window_scaling) {
				tcp_header_size += TCPOLEN_WSCALE_ALIGNED;
				sysctl_flags |= SYSCTL_FLAG_WSCALE;
			}
			if(sysctl_tcp_sack) {
				sysctl_flags |= SYSCTL_FLAG_SACK;
				if(!(sysctl_flags & SYSCTL_FLAG_TSTAMPS))
					tcp_header_size += TCPOLEN_SACKPERM_ALIGNED;
			}
		} else if (tp->rx_opt.eff_sacks) {
			/* A SACK is 2 pad bytes, a 2 byte header, plus
			 * 2 32-bit sequence numbers for each SACK block.
			 */
			tcp_header_size += (TCPOLEN_SACK_BASE_ALIGNED +
					    (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK));
		}
		
311
		if (tcp_packets_in_flight(tp) == 0)
312
			tcp_ca_event(sk, CA_EVENT_TX_START);
Linus Torvalds's avatar
Linus Torvalds committed
313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358

		th = (struct tcphdr *) skb_push(skb, tcp_header_size);
		skb->h.th = th;
		skb_set_owner_w(skb, sk);

		/* Build TCP header and checksum it. */
		th->source		= inet->sport;
		th->dest		= inet->dport;
		th->seq			= htonl(tcb->seq);
		th->ack_seq		= htonl(tp->rcv_nxt);
		*(((__u16 *)th) + 6)	= htons(((tcp_header_size >> 2) << 12) | tcb->flags);
		if (tcb->flags & TCPCB_FLAG_SYN) {
			/* RFC1323: The window in SYN & SYN/ACK segments
			 * is never scaled.
			 */
			th->window	= htons(tp->rcv_wnd);
		} else {
			th->window	= htons(tcp_select_window(sk));
		}
		th->check		= 0;
		th->urg_ptr		= 0;

		if (tp->urg_mode &&
		    between(tp->snd_up, tcb->seq+1, tcb->seq+0xFFFF)) {
			th->urg_ptr		= htons(tp->snd_up-tcb->seq);
			th->urg			= 1;
		}

		if (tcb->flags & TCPCB_FLAG_SYN) {
			tcp_syn_build_options((__u32 *)(th + 1),
					      tcp_advertise_mss(sk),
					      (sysctl_flags & SYSCTL_FLAG_TSTAMPS),
					      (sysctl_flags & SYSCTL_FLAG_SACK),
					      (sysctl_flags & SYSCTL_FLAG_WSCALE),
					      tp->rx_opt.rcv_wscale,
					      tcb->when,
		      			      tp->rx_opt.ts_recent);
		} else {
			tcp_build_and_update_options((__u32 *)(th + 1),
						     tp, tcb->when);

			TCP_ECN_send(sk, tp, skb, tcp_header_size);
		}
		tp->af_specific->send_check(sk, th, skb->len, skb);

		if (tcb->flags & TCPCB_FLAG_ACK)
359
			tcp_event_ack_sent(sk, tcp_skb_pcount(skb));
Linus Torvalds's avatar
Linus Torvalds committed
360 361 362 363 364 365 366 367 368 369

		if (skb->len != tcp_header_size)
			tcp_event_data_sent(tp, skb, sk);

		TCP_INC_STATS(TCP_MIB_OUTSEGS);

		err = tp->af_specific->queue_xmit(skb, 0);
		if (err <= 0)
			return err;

370
		tcp_enter_cwr(sk);
Linus Torvalds's avatar
Linus Torvalds committed
371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406

		/* NET_XMIT_CN is special. It does not guarantee,
		 * that this packet is lost. It tells that device
		 * is about to start to drop packets or already
		 * drops some packets of the same priority and
		 * invokes us to send less aggressively.
		 */
		return err == NET_XMIT_CN ? 0 : err;
	}
	return -ENOBUFS;
#undef SYSCTL_FLAG_TSTAMPS
#undef SYSCTL_FLAG_WSCALE
#undef SYSCTL_FLAG_SACK
}


/* This routine just queue's the buffer 
 *
 * NOTE: probe0 timer is not checked, do not forget tcp_push_pending_frames,
 * otherwise socket can stall.
 */
static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb)
{
	struct tcp_sock *tp = tcp_sk(sk);

	/* Advance write_seq and place onto the write_queue. */
	tp->write_seq = TCP_SKB_CB(skb)->end_seq;
	skb_header_release(skb);
	__skb_queue_tail(&sk->sk_write_queue, skb);
	sk_charge_skb(sk, skb);

	/* Queue it, remembering where we must start sending. */
	if (sk->sk_send_head == NULL)
		sk->sk_send_head = skb;
}

407
static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int mss_now)
408
{
409
	if (skb->len <= mss_now ||
410 411 412 413 414 415 416 417 418
	    !(sk->sk_route_caps & NETIF_F_TSO)) {
		/* Avoid the costly divide in the normal
		 * non-TSO case.
		 */
		skb_shinfo(skb)->tso_segs = 1;
		skb_shinfo(skb)->tso_size = 0;
	} else {
		unsigned int factor;

419 420
		factor = skb->len + (mss_now - 1);
		factor /= mss_now;
421
		skb_shinfo(skb)->tso_segs = factor;
422
		skb_shinfo(skb)->tso_size = mss_now;
Linus Torvalds's avatar
Linus Torvalds committed
423 424 425 426 427 428 429 430
	}
}

/* Function to create two new TCP segments.  Shrinks the given segment
 * to the specified size and appends a new segment with the rest of the
 * packet to the list.  This won't be called frequently, I hope. 
 * Remember, these are still headerless SKBs at this point.
 */
431
int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss_now)
Linus Torvalds's avatar
Linus Torvalds committed
432 433 434
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *buff;
435
	int nsize, old_factor;
Linus Torvalds's avatar
Linus Torvalds committed
436 437
	u16 flags;

438
	BUG_ON(len > skb->len);
Linus Torvalds's avatar
Linus Torvalds committed
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462
	nsize = skb_headlen(skb) - len;
	if (nsize < 0)
		nsize = 0;

	if (skb_cloned(skb) &&
	    skb_is_nonlinear(skb) &&
	    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
		return -ENOMEM;

	/* Get a new skb... force flag on. */
	buff = sk_stream_alloc_skb(sk, nsize, GFP_ATOMIC);
	if (buff == NULL)
		return -ENOMEM; /* We'll just try again later. */
	sk_charge_skb(sk, buff);

	/* Correct the sequence numbers. */
	TCP_SKB_CB(buff)->seq = TCP_SKB_CB(skb)->seq + len;
	TCP_SKB_CB(buff)->end_seq = TCP_SKB_CB(skb)->end_seq;
	TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(buff)->seq;

	/* PSH and FIN should only be set in the second packet. */
	flags = TCP_SKB_CB(skb)->flags;
	TCP_SKB_CB(skb)->flags = flags & ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
	TCP_SKB_CB(buff)->flags = flags;
463
	TCP_SKB_CB(buff)->sacked = TCP_SKB_CB(skb)->sacked;
Linus Torvalds's avatar
Linus Torvalds committed
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484
	TCP_SKB_CB(skb)->sacked &= ~TCPCB_AT_TAIL;

	if (!skb_shinfo(skb)->nr_frags && skb->ip_summed != CHECKSUM_HW) {
		/* Copy and checksum data tail into the new buffer. */
		buff->csum = csum_partial_copy_nocheck(skb->data + len, skb_put(buff, nsize),
						       nsize, 0);

		skb_trim(skb, len);

		skb->csum = csum_block_sub(skb->csum, buff->csum, len);
	} else {
		skb->ip_summed = CHECKSUM_HW;
		skb_split(skb, buff, len);
	}

	buff->ip_summed = skb->ip_summed;

	/* Looks stupid, but our code really uses when of
	 * skbs, which it never sent before. --ANK
	 */
	TCP_SKB_CB(buff)->when = TCP_SKB_CB(skb)->when;
485
	buff->tstamp = skb->tstamp;
Linus Torvalds's avatar
Linus Torvalds committed
486

487 488
	old_factor = tcp_skb_pcount(skb);

Linus Torvalds's avatar
Linus Torvalds committed
489
	/* Fix up tso_factor for both original and new SKB.  */
490 491
	tcp_set_skb_tso_segs(sk, skb, mss_now);
	tcp_set_skb_tso_segs(sk, buff, mss_now);
Linus Torvalds's avatar
Linus Torvalds committed
492

493 494 495
	/* If this packet has been sent out already, we must
	 * adjust the various packet counters.
	 */
496
	if (!before(tp->snd_nxt, TCP_SKB_CB(buff)->end_seq)) {
497 498
		int diff = old_factor - tcp_skb_pcount(skb) -
			tcp_skb_pcount(buff);
Linus Torvalds's avatar
Linus Torvalds committed
499

500
		tp->packets_out -= diff;
501 502 503 504 505 506

		if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)
			tp->sacked_out -= diff;
		if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS)
			tp->retrans_out -= diff;

507 508 509 510
		if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) {
			tp->lost_out -= diff;
			tp->left_out -= diff;
		}
511

512
		if (diff > 0) {
513 514 515 516 517 518 519 520
			/* Adjust Reno SACK estimate. */
			if (!tp->rx_opt.sack_ok) {
				tp->sacked_out -= diff;
				if ((int)tp->sacked_out < 0)
					tp->sacked_out = 0;
				tcp_sync_left_out(tp);
			}

521 522 523 524
			tp->fackets_out -= diff;
			if ((int)tp->fackets_out < 0)
				tp->fackets_out = 0;
		}
Linus Torvalds's avatar
Linus Torvalds committed
525 526 527
	}

	/* Link BUFF into the send queue. */
528
	skb_header_release(buff);
David S. Miller's avatar
David S. Miller committed
529
	__skb_append(skb, buff, &sk->sk_write_queue);
Linus Torvalds's avatar
Linus Torvalds committed
530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590

	return 0;
}

/* This is similar to __pskb_pull_head() (it will go to core/skbuff.c
 * eventually). The difference is that pulled data not copied, but
 * immediately discarded.
 */
static unsigned char *__pskb_trim_head(struct sk_buff *skb, int len)
{
	int i, k, eat;

	eat = len;
	k = 0;
	for (i=0; i<skb_shinfo(skb)->nr_frags; i++) {
		if (skb_shinfo(skb)->frags[i].size <= eat) {
			put_page(skb_shinfo(skb)->frags[i].page);
			eat -= skb_shinfo(skb)->frags[i].size;
		} else {
			skb_shinfo(skb)->frags[k] = skb_shinfo(skb)->frags[i];
			if (eat) {
				skb_shinfo(skb)->frags[k].page_offset += eat;
				skb_shinfo(skb)->frags[k].size -= eat;
				eat = 0;
			}
			k++;
		}
	}
	skb_shinfo(skb)->nr_frags = k;

	skb->tail = skb->data;
	skb->data_len -= len;
	skb->len = skb->data_len;
	return skb->tail;
}

int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len)
{
	if (skb_cloned(skb) &&
	    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
		return -ENOMEM;

	if (len <= skb_headlen(skb)) {
		__skb_pull(skb, len);
	} else {
		if (__pskb_trim_head(skb, len-skb_headlen(skb)) == NULL)
			return -ENOMEM;
	}

	TCP_SKB_CB(skb)->seq += len;
	skb->ip_summed = CHECKSUM_HW;

	skb->truesize	     -= len;
	sk->sk_wmem_queued   -= len;
	sk->sk_forward_alloc += len;
	sock_set_flag(sk, SOCK_QUEUE_SHRUNK);

	/* Any change of skb->len requires recalculation of tso
	 * factor and mss.
	 */
	if (tcp_skb_pcount(skb) > 1)
591
		tcp_set_skb_tso_segs(sk, skb, tcp_current_mss(sk, 1));
Linus Torvalds's avatar
Linus Torvalds committed
592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648

	return 0;
}

/* This function synchronize snd mss to current pmtu/exthdr set.

   tp->rx_opt.user_mss is mss set by user by TCP_MAXSEG. It does NOT counts
   for TCP options, but includes only bare TCP header.

   tp->rx_opt.mss_clamp is mss negotiated at connection setup.
   It is minumum of user_mss and mss received with SYN.
   It also does not include TCP options.

   tp->pmtu_cookie is last pmtu, seen by this function.

   tp->mss_cache is current effective sending mss, including
   all tcp options except for SACKs. It is evaluated,
   taking into account current pmtu, but never exceeds
   tp->rx_opt.mss_clamp.

   NOTE1. rfc1122 clearly states that advertised MSS
   DOES NOT include either tcp or ip options.

   NOTE2. tp->pmtu_cookie and tp->mss_cache are READ ONLY outside
   this function.			--ANK (980731)
 */

unsigned int tcp_sync_mss(struct sock *sk, u32 pmtu)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int mss_now;

	/* Calculate base mss without TCP options:
	   It is MMS_S - sizeof(tcphdr) of rfc1122
	 */
	mss_now = pmtu - tp->af_specific->net_header_len - sizeof(struct tcphdr);

	/* Clamp it (mss_clamp does not include tcp options) */
	if (mss_now > tp->rx_opt.mss_clamp)
		mss_now = tp->rx_opt.mss_clamp;

	/* Now subtract optional transport overhead */
	mss_now -= tp->ext_header_len;

	/* Then reserve room for full set of TCP options and 8 bytes of data */
	if (mss_now < 48)
		mss_now = 48;

	/* Now subtract TCP options size, not including SACKs */
	mss_now -= tp->tcp_header_len - sizeof(struct tcphdr);

	/* Bound mss with half of window */
	if (tp->max_window && mss_now > (tp->max_window>>1))
		mss_now = max((tp->max_window>>1), 68U - tp->tcp_header_len);

	/* And store cached results */
	tp->pmtu_cookie = pmtu;
649
	tp->mss_cache = mss_now;
Linus Torvalds's avatar
Linus Torvalds committed
650 651 652 653 654 655 656 657 658 659 660

	return mss_now;
}

/* Compute the current effective MSS, taking SACKs and IP options,
 * and even PMTU discovery events into account.
 *
 * LARGESEND note: !urg_mode is overkill, only frames up to snd_up
 * cannot be large. However, taking into account rare use of URG, this
 * is not a big flaw.
 */
661
unsigned int tcp_current_mss(struct sock *sk, int large_allowed)
Linus Torvalds's avatar
Linus Torvalds committed
662 663 664
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct dst_entry *dst = __sk_dst_get(sk);
665 666 667 668 669 670 671 672 673 674
	u32 mss_now;
	u16 xmit_size_goal;
	int doing_tso = 0;

	mss_now = tp->mss_cache;

	if (large_allowed &&
	    (sk->sk_route_caps & NETIF_F_TSO) &&
	    !tp->urg_mode)
		doing_tso = 1;
Linus Torvalds's avatar
Linus Torvalds committed
675 676 677 678 679 680 681

	if (dst) {
		u32 mtu = dst_mtu(dst);
		if (mtu != tp->pmtu_cookie)
			mss_now = tcp_sync_mss(sk, mtu);
	}

682 683 684
	if (tp->rx_opt.eff_sacks)
		mss_now -= (TCPOLEN_SACK_BASE_ALIGNED +
			    (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK));
Linus Torvalds's avatar
Linus Torvalds committed
685

686
	xmit_size_goal = mss_now;
Linus Torvalds's avatar
Linus Torvalds committed
687

688 689 690
	if (doing_tso) {
		xmit_size_goal = 65535 -
			tp->af_specific->net_header_len -
Linus Torvalds's avatar
Linus Torvalds committed
691 692
			tp->ext_header_len - tp->tcp_header_len;

693 694 695 696
		if (tp->max_window &&
		    (xmit_size_goal > (tp->max_window >> 1)))
			xmit_size_goal = max((tp->max_window >> 1),
					     68U - tp->tcp_header_len);
Linus Torvalds's avatar
Linus Torvalds committed
697

698
		xmit_size_goal -= (xmit_size_goal % mss_now);
Linus Torvalds's avatar
Linus Torvalds committed
699
	}
700
	tp->xmit_size_goal = xmit_size_goal;
Linus Torvalds's avatar
Linus Torvalds committed
701 702 703 704

	return mss_now;
}

705 706 707 708 709 710 711 712 713 714 715 716 717 718 719
/* Congestion window validation. (RFC2861) */

static inline void tcp_cwnd_validate(struct sock *sk, struct tcp_sock *tp)
{
	__u32 packets_out = tp->packets_out;

	if (packets_out >= tp->snd_cwnd) {
		/* Network is feed fully. */
		tp->snd_cwnd_used = 0;
		tp->snd_cwnd_stamp = tcp_time_stamp;
	} else {
		/* Network starves. */
		if (tp->packets_out > tp->snd_cwnd_used)
			tp->snd_cwnd_used = tp->packets_out;

720
		if ((s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= inet_csk(sk)->icsk_rto)
721 722 723 724
			tcp_cwnd_application_limited(sk);
	}
}

725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
static unsigned int tcp_window_allows(struct tcp_sock *tp, struct sk_buff *skb, unsigned int mss_now, unsigned int cwnd)
{
	u32 window, cwnd_len;

	window = (tp->snd_una + tp->snd_wnd - TCP_SKB_CB(skb)->seq);
	cwnd_len = mss_now * cwnd;
	return min(window, cwnd_len);
}

/* Can at least one segment of SKB be sent right now, according to the
 * congestion window rules?  If so, return how many segments are allowed.
 */
static inline unsigned int tcp_cwnd_test(struct tcp_sock *tp, struct sk_buff *skb)
{
	u32 in_flight, cwnd;

	/* Don't be strict about the congestion window for the final FIN.  */
	if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)
		return 1;

	in_flight = tcp_packets_in_flight(tp);
	cwnd = tp->snd_cwnd;
	if (in_flight < cwnd)
		return (cwnd - in_flight);

	return 0;
}

/* This must be invoked the first time we consider transmitting
 * SKB onto the wire.
 */
756
static inline int tcp_init_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int mss_now)
757 758 759
{
	int tso_segs = tcp_skb_pcount(skb);

760 761 762 763
	if (!tso_segs ||
	    (tso_segs > 1 &&
	     skb_shinfo(skb)->tso_size != mss_now)) {
		tcp_set_skb_tso_segs(sk, skb, mss_now);
764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840
		tso_segs = tcp_skb_pcount(skb);
	}
	return tso_segs;
}

static inline int tcp_minshall_check(const struct tcp_sock *tp)
{
	return after(tp->snd_sml,tp->snd_una) &&
		!after(tp->snd_sml, tp->snd_nxt);
}

/* Return 0, if packet can be sent now without violation Nagle's rules:
 * 1. It is full sized.
 * 2. Or it contains FIN. (already checked by caller)
 * 3. Or TCP_NODELAY was set.
 * 4. Or TCP_CORK is not set, and all sent packets are ACKed.
 *    With Minshall's modification: all sent small packets are ACKed.
 */

static inline int tcp_nagle_check(const struct tcp_sock *tp,
				  const struct sk_buff *skb, 
				  unsigned mss_now, int nonagle)
{
	return (skb->len < mss_now &&
		((nonagle&TCP_NAGLE_CORK) ||
		 (!nonagle &&
		  tp->packets_out &&
		  tcp_minshall_check(tp))));
}

/* Return non-zero if the Nagle test allows this packet to be
 * sent now.
 */
static inline int tcp_nagle_test(struct tcp_sock *tp, struct sk_buff *skb,
				 unsigned int cur_mss, int nonagle)
{
	/* Nagle rule does not apply to frames, which sit in the middle of the
	 * write_queue (they have no chances to get new data).
	 *
	 * This is implemented in the callers, where they modify the 'nonagle'
	 * argument based upon the location of SKB in the send queue.
	 */
	if (nonagle & TCP_NAGLE_PUSH)
		return 1;

	/* Don't use the nagle rule for urgent data (or for the final FIN).  */
	if (tp->urg_mode ||
	    (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN))
		return 1;

	if (!tcp_nagle_check(tp, skb, cur_mss, nonagle))
		return 1;

	return 0;
}

/* Does at least the first segment of SKB fit into the send window? */
static inline int tcp_snd_wnd_test(struct tcp_sock *tp, struct sk_buff *skb, unsigned int cur_mss)
{
	u32 end_seq = TCP_SKB_CB(skb)->end_seq;

	if (skb->len > cur_mss)
		end_seq = TCP_SKB_CB(skb)->seq + cur_mss;

	return !after(end_seq, tp->snd_una + tp->snd_wnd);
}

/* This checks if the data bearing packet SKB (usually sk->sk_send_head)
 * should be put on the wire right now.  If so, it returns the number of
 * packets allowed by the congestion window.
 */
static unsigned int tcp_snd_test(struct sock *sk, struct sk_buff *skb,
				 unsigned int cur_mss, int nonagle)
{
	struct tcp_sock *tp = tcp_sk(sk);
	unsigned int cwnd_quota;

841
	tcp_init_tso_segs(sk, skb, cur_mss);
842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877

	if (!tcp_nagle_test(tp, skb, cur_mss, nonagle))
		return 0;

	cwnd_quota = tcp_cwnd_test(tp, skb);
	if (cwnd_quota &&
	    !tcp_snd_wnd_test(tp, skb, cur_mss))
		cwnd_quota = 0;

	return cwnd_quota;
}

static inline int tcp_skb_is_last(const struct sock *sk, 
				  const struct sk_buff *skb)
{
	return skb->next == (struct sk_buff *)&sk->sk_write_queue;
}

int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp)
{
	struct sk_buff *skb = sk->sk_send_head;

	return (skb &&
		tcp_snd_test(sk, skb, tcp_current_mss(sk, 1),
			     (tcp_skb_is_last(sk, skb) ?
			      TCP_NAGLE_PUSH :
			      tp->nonagle)));
}

/* Trim TSO SKB to LEN bytes, put the remaining data into a new packet
 * which is put after SKB on the list.  It is very much like
 * tcp_fragment() except that it may make several kinds of assumptions
 * in order to speed up the splitting operation.  In particular, we
 * know that all the data is in scatter-gather pages, and that the
 * packet has never been sent out before (and thus is not cloned).
 */
878
static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, unsigned int mss_now)
879 880 881 882 883 884
{
	struct sk_buff *buff;
	int nlen = skb->len - len;
	u16 flags;

	/* All of a TSO frame must be composed of paged data.  */
885 886
	if (skb->len != skb->data_len)
		return tcp_fragment(sk, skb, len, mss_now);
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

	buff = sk_stream_alloc_pskb(sk, 0, 0, GFP_ATOMIC);
	if (unlikely(buff == NULL))
		return -ENOMEM;

	buff->truesize = nlen;
	skb->truesize -= nlen;

	/* Correct the sequence numbers. */
	TCP_SKB_CB(buff)->seq = TCP_SKB_CB(skb)->seq + len;
	TCP_SKB_CB(buff)->end_seq = TCP_SKB_CB(skb)->end_seq;
	TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(buff)->seq;

	/* PSH and FIN should only be set in the second packet. */
	flags = TCP_SKB_CB(skb)->flags;
	TCP_SKB_CB(skb)->flags = flags & ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
	TCP_SKB_CB(buff)->flags = flags;

	/* This packet was never sent out yet, so no SACK bits. */
	TCP_SKB_CB(buff)->sacked = 0;

	buff->ip_summed = skb->ip_summed = CHECKSUM_HW;
	skb_split(skb, buff, len);

	/* Fix up tso_factor for both original and new SKB.  */
912 913
	tcp_set_skb_tso_segs(sk, skb, mss_now);
	tcp_set_skb_tso_segs(sk, buff, mss_now);
914 915 916

	/* Link BUFF into the send queue. */
	skb_header_release(buff);
David S. Miller's avatar
David S. Miller committed
917
	__skb_append(skb, buff, &sk->sk_write_queue);
918 919 920 921 922 923 924 925 926 927 928

	return 0;
}

/* Try to defer sending, if possible, in order to minimize the amount
 * of TSO splitting we do.  View it as a kind of TSO Nagle test.
 *
 * This algorithm is from John Heffner.
 */
static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_buff *skb)
{
929
	const struct inet_connection_sock *icsk = inet_csk(sk);
930 931 932 933 934
	u32 send_win, cong_win, limit, in_flight;

	if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)
		return 0;

935
	if (icsk->icsk_ca_state != TCP_CA_Open)
936 937
		return 0;

938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972
	in_flight = tcp_packets_in_flight(tp);

	BUG_ON(tcp_skb_pcount(skb) <= 1 ||
	       (tp->snd_cwnd <= in_flight));

	send_win = (tp->snd_una + tp->snd_wnd) - TCP_SKB_CB(skb)->seq;

	/* From in_flight test above, we know that cwnd > in_flight.  */
	cong_win = (tp->snd_cwnd - in_flight) * tp->mss_cache;

	limit = min(send_win, cong_win);

	if (sysctl_tcp_tso_win_divisor) {
		u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache);

		/* If at least some fraction of a window is available,
		 * just use it.
		 */
		chunk /= sysctl_tcp_tso_win_divisor;
		if (limit >= chunk)
			return 0;
	} else {
		/* Different approach, try not to defer past a single
		 * ACK.  Receiver should ACK every other full sized
		 * frame, so if we have space for more than 3 frames
		 * then send now.
		 */
		if (limit > tcp_max_burst(tp) * tp->mss_cache)
			return 0;
	}

	/* Ok, it looks like it is advisable to defer.  */
	return 1;
}

Linus Torvalds's avatar
Linus Torvalds committed
973 974 975 976 977 978 979
/* This routine writes packets to the network.  It advances the
 * send_head.  This happens as incoming acks open up the remote
 * window for us.
 *
 * Returns 1, if no segments are in flight and we have queued segments, but
 * cannot send anything now because of SWS or another problem.
 */
980
static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
Linus Torvalds's avatar
Linus Torvalds committed
981 982
{
	struct tcp_sock *tp = tcp_sk(sk);
983
	struct sk_buff *skb;
984 985
	unsigned int tso_segs, sent_pkts;
	int cwnd_quota;
Linus Torvalds's avatar
Linus Torvalds committed
986 987 988 989 990

	/* If we are closed, the bytes will have to remain here.
	 * In time closedown will finish, we empty the write queue and all
	 * will be happy.
	 */
991 992
	if (unlikely(sk->sk_state == TCP_CLOSE))
		return 0;
Linus Torvalds's avatar
Linus Torvalds committed
993

994
	sent_pkts = 0;
995
	while ((skb = sk->sk_send_head)) {
996 997
		unsigned int limit;

998
		tso_segs = tcp_init_tso_segs(sk, skb, mss_now);
999
		BUG_ON(!tso_segs);
1000

1001 1002 1003 1004 1005 1006 1007
		cwnd_quota = tcp_cwnd_test(tp, skb);
		if (!cwnd_quota)
			break;

		if (unlikely(!tcp_snd_wnd_test(tp, skb, mss_now)))
			break;

1008 1009 1010 1011 1012 1013 1014 1015 1016
		if (tso_segs == 1) {
			if (unlikely(!tcp_nagle_test(tp, skb, mss_now,
						     (tcp_skb_is_last(sk, skb) ?
						      nonagle : TCP_NAGLE_PUSH))))
				break;
		} else {
			if (tcp_tso_should_defer(sk, tp, skb))
				break;
		}
1017

1018
		limit = mss_now;
1019
		if (tso_segs > 1) {
1020 1021
			limit = tcp_window_allows(tp, skb,
						  mss_now, cwnd_quota);
1022 1023 1024

			if (skb->len < limit) {
				unsigned int trim = skb->len % mss_now;
1025

1026 1027 1028
				if (trim)
					limit = skb->len - trim;
			}
1029
		}
Linus Torvalds's avatar
Linus Torvalds committed
1030

1031 1032 1033 1034
		if (skb->len > limit &&
		    unlikely(tso_fragment(sk, skb, limit, mss_now)))
			break;

1035
		TCP_SKB_CB(skb)->when = tcp_time_stamp;
1036

1037
		if (unlikely(tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC))))
1038
			break;
Linus Torvalds's avatar
Linus Torvalds committed
1039

1040 1041 1042 1043
		/* Advance the send_head.  This one is sent out.
		 * This call will increment packets_out.
		 */
		update_send_head(sk, tp, skb);
Linus Torvalds's avatar
Linus Torvalds committed
1044

1045
		tcp_minshall_update(tp, mss_now, skb);
1046
		sent_pkts++;
1047
	}
Linus Torvalds's avatar
Linus Torvalds committed
1048

1049
	if (likely(sent_pkts)) {
1050 1051
		tcp_cwnd_validate(sk, tp);
		return 0;
Linus Torvalds's avatar
Linus Torvalds committed
1052
	}
1053
	return !tp->packets_out && sk->sk_send_head;
Linus Torvalds's avatar
Linus Torvalds committed
1054 1055
}

1056 1057 1058 1059 1060
/* Push out any pending frames which were held back due to
 * TCP_CORK or attempt at coalescing tiny packets.
 * The socket must be locked by the caller.
 */
void __tcp_push_pending_frames(struct sock *sk, struct tcp_sock *tp,
1061
			       unsigned int cur_mss, int nonagle)
1062 1063 1064 1065
{
	struct sk_buff *skb = sk->sk_send_head;

	if (skb) {
1066
		if (tcp_write_xmit(sk, cur_mss, nonagle))
1067 1068 1069 1070
			tcp_check_probe_timer(sk, tp);
	}
}

1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081
/* Send _single_ skb sitting at the send head. This function requires
 * true push pending frames to setup probe timer etc.
 */
void tcp_push_one(struct sock *sk, unsigned int mss_now)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb = sk->sk_send_head;
	unsigned int tso_segs, cwnd_quota;

	BUG_ON(!skb || skb->len < mss_now);

1082
	tso_segs = tcp_init_tso_segs(sk, skb, mss_now);
1083 1084 1085
	cwnd_quota = tcp_snd_test(sk, skb, mss_now, TCP_NAGLE_PUSH);

	if (likely(cwnd_quota)) {
1086 1087
		unsigned int limit;

1088 1089
		BUG_ON(!tso_segs);

1090
		limit = mss_now;
1091
		if (tso_segs > 1) {
1092 1093
			limit = tcp_window_allows(tp, skb,
						  mss_now, cwnd_quota);
1094 1095 1096 1097 1098 1099 1100 1101 1102

			if (skb->len < limit) {
				unsigned int trim = skb->len % mss_now;

				if (trim)
					limit = skb->len - trim;
			}
		}

1103 1104 1105 1106
		if (skb->len > limit &&
		    unlikely(tso_fragment(sk, skb, limit, mss_now)))
			return;

1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117
		/* Send it out now. */
		TCP_SKB_CB(skb)->when = tcp_time_stamp;

		if (likely(!tcp_transmit_skb(sk, skb_clone(skb, sk->sk_allocation)))) {
			update_send_head(sk, tp, skb);
			tcp_cwnd_validate(sk, tp);
			return;
		}
	}
}

Linus Torvalds's avatar
Linus Torvalds committed
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 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171
/* This function returns the amount that we can raise the
 * usable window based on the following constraints
 *  
 * 1. The window can never be shrunk once it is offered (RFC 793)
 * 2. We limit memory per socket
 *
 * RFC 1122:
 * "the suggested [SWS] avoidance algorithm for the receiver is to keep
 *  RECV.NEXT + RCV.WIN fixed until:
 *  RCV.BUFF - RCV.USER - RCV.WINDOW >= min(1/2 RCV.BUFF, MSS)"
 *
 * i.e. don't raise the right edge of the window until you can raise
 * it at least MSS bytes.
 *
 * Unfortunately, the recommended algorithm breaks header prediction,
 * since header prediction assumes th->window stays fixed.
 *
 * Strictly speaking, keeping th->window fixed violates the receiver
 * side SWS prevention criteria. The problem is that under this rule
 * a stream of single byte packets will cause the right side of the
 * window to always advance by a single byte.
 * 
 * Of course, if the sender implements sender side SWS prevention
 * then this will not be a problem.
 * 
 * BSD seems to make the following compromise:
 * 
 *	If the free space is less than the 1/4 of the maximum
 *	space available and the free space is less than 1/2 mss,
 *	then set the window to 0.
 *	[ Actually, bsd uses MSS and 1/4 of maximal _window_ ]
 *	Otherwise, just prevent the window from shrinking
 *	and from being larger than the largest representable value.
 *
 * This prevents incremental opening of the window in the regime
 * where TCP is limited by the speed of the reader side taking
 * data out of the TCP receive queue. It does nothing about
 * those cases where the window is constrained on the sender side
 * because the pipeline is full.
 *
 * BSD also seems to "accidentally" limit itself to windows that are a
 * multiple of MSS, at least until the free space gets quite small.
 * This would appear to be a side effect of the mbuf implementation.
 * Combining these two algorithms results in the observed behavior
 * of having a fixed window size at almost all times.
 *
 * Below we obtain similar behavior by forcing the offered window to
 * a multiple of the mss when it is feasible to do so.
 *
 * Note, we don't "adjust" for TIMESTAMP or SACK option bytes.
 * Regular options like TIMESTAMP are taken into account.
 */
u32 __tcp_select_window(struct sock *sk)
{
1172
	struct inet_connection_sock *icsk = inet_csk(sk);
Linus Torvalds's avatar
Linus Torvalds committed
1173 1174 1175 1176 1177 1178 1179
	struct tcp_sock *tp = tcp_sk(sk);
	/* MSS for the peer's data.  Previous verions used mss_clamp
	 * here.  I don't know if the value based on our guesses
	 * of peer's MSS is better for the performance.  It's more correct
	 * but may be worse for the performance because of rcv_mss
	 * fluctuations.  --SAW  1998/11/1
	 */
1180
	int mss = icsk->icsk_ack.rcv_mss;
Linus Torvalds's avatar
Linus Torvalds committed
1181 1182 1183 1184 1185 1186 1187 1188
	int free_space = tcp_space(sk);
	int full_space = min_t(int, tp->window_clamp, tcp_full_space(sk));
	int window;

	if (mss > full_space)
		mss = full_space; 

	if (free_space < full_space/2) {
1189
		icsk->icsk_ack.quick = 0;
Linus Torvalds's avatar
Linus Torvalds committed
1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263

		if (tcp_memory_pressure)
			tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U*tp->advmss);

		if (free_space < mss)
			return 0;
	}

	if (free_space > tp->rcv_ssthresh)
		free_space = tp->rcv_ssthresh;

	/* Don't do rounding if we are using window scaling, since the
	 * scaled window will not line up with the MSS boundary anyway.
	 */
	window = tp->rcv_wnd;
	if (tp->rx_opt.rcv_wscale) {
		window = free_space;

		/* Advertise enough space so that it won't get scaled away.
		 * Import case: prevent zero window announcement if
		 * 1<<rcv_wscale > mss.
		 */
		if (((window >> tp->rx_opt.rcv_wscale) << tp->rx_opt.rcv_wscale) != window)
			window = (((window >> tp->rx_opt.rcv_wscale) + 1)
				  << tp->rx_opt.rcv_wscale);
	} else {
		/* Get the largest window that is a nice multiple of mss.
		 * Window clamp already applied above.
		 * If our current window offering is within 1 mss of the
		 * free space we just keep it. This prevents the divide
		 * and multiply from happening most of the time.
		 * We also don't do any window rounding when the free space
		 * is too small.
		 */
		if (window <= free_space - mss || window > free_space)
			window = (free_space/mss)*mss;
	}

	return window;
}

/* Attempt to collapse two adjacent SKB's during retransmission. */
static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int mss_now)
{
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *next_skb = skb->next;

	/* The first test we must make is that neither of these two
	 * SKB's are still referenced by someone else.
	 */
	if (!skb_cloned(skb) && !skb_cloned(next_skb)) {
		int skb_size = skb->len, next_skb_size = next_skb->len;
		u16 flags = TCP_SKB_CB(skb)->flags;

		/* Also punt if next skb has been SACK'd. */
		if(TCP_SKB_CB(next_skb)->sacked & TCPCB_SACKED_ACKED)
			return;

		/* Next skb is out of window. */
		if (after(TCP_SKB_CB(next_skb)->end_seq, tp->snd_una+tp->snd_wnd))
			return;

		/* Punt if not enough space exists in the first SKB for
		 * the data in the second, or the total combined payload
		 * would exceed the MSS.
		 */
		if ((next_skb_size > skb_tailroom(skb)) ||
		    ((skb_size + next_skb_size) > mss_now))
			return;

		BUG_ON(tcp_skb_pcount(skb) != 1 ||
		       tcp_skb_pcount(next_skb) != 1);

		/* Ok.  We will be able to collapse the packet. */
David S. Miller's avatar
David S. Miller committed
1264
		__skb_unlink(next_skb, &sk->sk_write_queue);
Linus Torvalds's avatar
Linus Torvalds committed
1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311

		memcpy(skb_put(skb, next_skb_size), next_skb->data, next_skb_size);

		if (next_skb->ip_summed == CHECKSUM_HW)
			skb->ip_summed = CHECKSUM_HW;

		if (skb->ip_summed != CHECKSUM_HW)
			skb->csum = csum_block_add(skb->csum, next_skb->csum, skb_size);

		/* Update sequence range on original skb. */
		TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(next_skb)->end_seq;

		/* Merge over control information. */
		flags |= TCP_SKB_CB(next_skb)->flags; /* This moves PSH/FIN etc. over */
		TCP_SKB_CB(skb)->flags = flags;

		/* All done, get rid of second SKB and account for it so
		 * packet counting does not break.
		 */
		TCP_SKB_CB(skb)->sacked |= TCP_SKB_CB(next_skb)->sacked&(TCPCB_EVER_RETRANS|TCPCB_AT_TAIL);
		if (TCP_SKB_CB(next_skb)->sacked&TCPCB_SACKED_RETRANS)
			tp->retrans_out -= tcp_skb_pcount(next_skb);
		if (TCP_SKB_CB(next_skb)->sacked&TCPCB_LOST) {
			tp->lost_out -= tcp_skb_pcount(next_skb);
			tp->left_out -= tcp_skb_pcount(next_skb);
		}
		/* Reno case is special. Sigh... */
		if (!tp->rx_opt.sack_ok && tp->sacked_out) {
			tcp_dec_pcount_approx(&tp->sacked_out, next_skb);
			tp->left_out -= tcp_skb_pcount(next_skb);
		}

		/* Not quite right: it can be > snd.fack, but
		 * it is better to underestimate fackets.
		 */
		tcp_dec_pcount_approx(&tp->fackets_out, next_skb);
		tcp_packets_out_dec(tp, next_skb);
		sk_stream_free_skb(sk, next_skb);
	}
}

/* Do a simple retransmit without using the backoff mechanisms in
 * tcp_timer. This is used for path mtu discovery. 
 * The socket is already locked here.
 */ 
void tcp_simple_retransmit(struct sock *sk)
{
1312
	const struct inet_connection_sock *icsk = inet_csk(sk);
Linus Torvalds's avatar
Linus Torvalds committed
1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb;
	unsigned int mss = tcp_current_mss(sk, 0);
	int lost = 0;

	sk_stream_for_retrans_queue(skb, sk) {
		if (skb->len > mss && 
		    !(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED)) {
			if (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS) {
				TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
				tp->retrans_out -= tcp_skb_pcount(skb);
			}
			if (!(TCP_SKB_CB(skb)->sacked&TCPCB_LOST)) {
				TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
				tp->lost_out += tcp_skb_pcount(skb);
				lost = 1;
			}
		}
	}

	if (!lost)
		return;

	tcp_sync_left_out(tp);

 	/* Don't muck with the congestion window here.
	 * Reason is that we do not increase amount of _data_
	 * in network, but units changed and effective
	 * cwnd/ssthresh really reduced now.
	 */
1343
	if (icsk->icsk_ca_state != TCP_CA_Loss) {
Linus Torvalds's avatar
Linus Torvalds committed
1344
		tp->high_seq = tp->snd_nxt;
1345
		tp->snd_ssthresh = tcp_current_ssthresh(sk);
Linus Torvalds's avatar
Linus Torvalds committed
1346 1347
		tp->prior_ssthresh = 0;
		tp->undo_marker = 0;
1348
		tcp_set_ca_state(sk, TCP_CA_Loss);
Linus Torvalds's avatar
Linus Torvalds committed
1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386
	}
	tcp_xmit_retransmit_queue(sk);
}

/* This retransmits one SKB.  Policy decisions and retransmit queue
 * state updates are done by the caller.  Returns non-zero if an
 * error occurred which prevented the send.
 */
int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
{
	struct tcp_sock *tp = tcp_sk(sk);
 	unsigned int cur_mss = tcp_current_mss(sk, 0);
	int err;

	/* Do not sent more than we queued. 1/4 is reserved for possible
	 * copying overhead: frgagmentation, tunneling, mangling etc.
	 */
	if (atomic_read(&sk->sk_wmem_alloc) >
	    min(sk->sk_wmem_queued + (sk->sk_wmem_queued >> 2), sk->sk_sndbuf))
		return -EAGAIN;

	if (before(TCP_SKB_CB(skb)->seq, tp->snd_una)) {
		if (before(TCP_SKB_CB(skb)->end_seq, tp->snd_una))
			BUG();
		if (tcp_trim_head(sk, skb, tp->snd_una - TCP_SKB_CB(skb)->seq))
			return -ENOMEM;
	}

	/* If receiver has shrunk his window, and skb is out of
	 * new window, do not retransmit it. The exception is the
	 * case, when window is shrunk to zero. In this case
	 * our retransmit serves as a zero window probe.
	 */
	if (!before(TCP_SKB_CB(skb)->seq, tp->snd_una+tp->snd_wnd)
	    && TCP_SKB_CB(skb)->seq != tp->snd_una)
		return -EAGAIN;

	if (skb->len > cur_mss) {
1387
		if (tcp_fragment(sk, skb, cur_mss, cur_mss))
Linus Torvalds's avatar
Linus Torvalds committed
1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467
			return -ENOMEM; /* We'll try again later. */
	}

	/* Collapse two adjacent packets if worthwhile and we can. */
	if(!(TCP_SKB_CB(skb)->flags & TCPCB_FLAG_SYN) &&
	   (skb->len < (cur_mss >> 1)) &&
	   (skb->next != sk->sk_send_head) &&
	   (skb->next != (struct sk_buff *)&sk->sk_write_queue) &&
	   (skb_shinfo(skb)->nr_frags == 0 && skb_shinfo(skb->next)->nr_frags == 0) &&
	   (tcp_skb_pcount(skb) == 1 && tcp_skb_pcount(skb->next) == 1) &&
	   (sysctl_tcp_retrans_collapse != 0))
		tcp_retrans_try_collapse(sk, skb, cur_mss);

	if(tp->af_specific->rebuild_header(sk))
		return -EHOSTUNREACH; /* Routing failure or similar. */

	/* Some Solaris stacks overoptimize and ignore the FIN on a
	 * retransmit when old data is attached.  So strip it off
	 * since it is cheap to do so and saves bytes on the network.
	 */
	if(skb->len > 0 &&
	   (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) &&
	   tp->snd_una == (TCP_SKB_CB(skb)->end_seq - 1)) {
		if (!pskb_trim(skb, 0)) {
			TCP_SKB_CB(skb)->seq = TCP_SKB_CB(skb)->end_seq - 1;
			skb_shinfo(skb)->tso_segs = 1;
			skb_shinfo(skb)->tso_size = 0;
			skb->ip_summed = CHECKSUM_NONE;
			skb->csum = 0;
		}
	}

	/* Make a copy, if the first transmission SKB clone we made
	 * is still in somebody's hands, else make a clone.
	 */
	TCP_SKB_CB(skb)->when = tcp_time_stamp;

	err = tcp_transmit_skb(sk, (skb_cloned(skb) ?
				    pskb_copy(skb, GFP_ATOMIC):
				    skb_clone(skb, GFP_ATOMIC)));

	if (err == 0) {
		/* Update global TCP statistics. */
		TCP_INC_STATS(TCP_MIB_RETRANSSEGS);

		tp->total_retrans++;

#if FASTRETRANS_DEBUG > 0
		if (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS) {
			if (net_ratelimit())
				printk(KERN_DEBUG "retrans_out leaked.\n");
		}
#endif
		TCP_SKB_CB(skb)->sacked |= TCPCB_RETRANS;
		tp->retrans_out += tcp_skb_pcount(skb);

		/* Save stamp of the first retransmit. */
		if (!tp->retrans_stamp)
			tp->retrans_stamp = TCP_SKB_CB(skb)->when;

		tp->undo_retrans++;

		/* snd_nxt is stored to detect loss of retransmitted segment,
		 * see tcp_input.c tcp_sacktag_write_queue().
		 */
		TCP_SKB_CB(skb)->ack_seq = tp->snd_nxt;
	}
	return err;
}

/* This gets called after a retransmit timeout, and the initially
 * retransmitted data is acknowledged.  It tries to continue
 * resending the rest of the retransmit queue, until either
 * we've sent it all or the congestion window limit is reached.
 * If doing SACK, the first ACK which comes back for a timeout
 * based retransmit packet might feed us FACK information again.
 * If so, we use it to avoid unnecessarily retransmissions.
 */
void tcp_xmit_retransmit_queue(struct sock *sk)
{
1468
	const struct inet_connection_sock *icsk = inet_csk(sk);
Linus Torvalds's avatar
Linus Torvalds committed
1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb;
	int packet_cnt = tp->lost_out;

	/* First pass: retransmit lost packets. */
	if (packet_cnt) {
		sk_stream_for_retrans_queue(skb, sk) {
			__u8 sacked = TCP_SKB_CB(skb)->sacked;

			/* Assume this retransmit will generate
			 * only one packet for congestion window
			 * calculation purposes.  This works because
			 * tcp_retransmit_skb() will chop up the
			 * packet to be MSS sized and all the
			 * packet counting works out.
			 */
			if (tcp_packets_in_flight(tp) >= tp->snd_cwnd)
				return;

			if (sacked&TCPCB_LOST) {
				if (!(sacked&(TCPCB_SACKED_ACKED|TCPCB_SACKED_RETRANS))) {
					if (tcp_retransmit_skb(sk, skb))
						return;
1492
					if (icsk->icsk_ca_state != TCP_CA_Loss)
Linus Torvalds's avatar
Linus Torvalds committed
1493 1494 1495 1496 1497 1498
						NET_INC_STATS_BH(LINUX_MIB_TCPFASTRETRANS);
					else
						NET_INC_STATS_BH(LINUX_MIB_TCPSLOWSTARTRETRANS);

					if (skb ==
					    skb_peek(&sk->sk_write_queue))
1499
						inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
1500 1501
									  inet_csk(sk)->icsk_rto,
									  TCP_RTO_MAX);
Linus Torvalds's avatar
Linus Torvalds committed
1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513
				}

				packet_cnt -= tcp_skb_pcount(skb);
				if (packet_cnt <= 0)
					break;
			}
		}
	}

	/* OK, demanded retransmission is finished. */

	/* Forward retransmissions are possible only during Recovery. */
1514
	if (icsk->icsk_ca_state != TCP_CA_Recovery)
Linus Torvalds's avatar
Linus Torvalds committed
1515 1516 1517 1518