tcp.c 75.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7
/*
 * 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).
 *
8
 * Authors:	Ross Biro
L
Linus Torvalds 已提交
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247
 *		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>
 *
 * Fixes:
 *		Alan Cox	:	Numerous verify_area() calls
 *		Alan Cox	:	Set the ACK bit on a reset
 *		Alan Cox	:	Stopped it crashing if it closed while
 *					sk->inuse=1 and was trying to connect
 *					(tcp_err()).
 *		Alan Cox	:	All icmp error handling was broken
 *					pointers passed where wrong and the
 *					socket was looked up backwards. Nobody
 *					tested any icmp error code obviously.
 *		Alan Cox	:	tcp_err() now handled properly. It
 *					wakes people on errors. poll
 *					behaves and the icmp error race
 *					has gone by moving it into sock.c
 *		Alan Cox	:	tcp_send_reset() fixed to work for
 *					everything not just packets for
 *					unknown sockets.
 *		Alan Cox	:	tcp option processing.
 *		Alan Cox	:	Reset tweaked (still not 100%) [Had
 *					syn rule wrong]
 *		Herp Rosmanith  :	More reset fixes
 *		Alan Cox	:	No longer acks invalid rst frames.
 *					Acking any kind of RST is right out.
 *		Alan Cox	:	Sets an ignore me flag on an rst
 *					receive otherwise odd bits of prattle
 *					escape still
 *		Alan Cox	:	Fixed another acking RST frame bug.
 *					Should stop LAN workplace lockups.
 *		Alan Cox	: 	Some tidyups using the new skb list
 *					facilities
 *		Alan Cox	:	sk->keepopen now seems to work
 *		Alan Cox	:	Pulls options out correctly on accepts
 *		Alan Cox	:	Fixed assorted sk->rqueue->next errors
 *		Alan Cox	:	PSH doesn't end a TCP read. Switched a
 *					bit to skb ops.
 *		Alan Cox	:	Tidied tcp_data to avoid a potential
 *					nasty.
 *		Alan Cox	:	Added some better commenting, as the
 *					tcp is hard to follow
 *		Alan Cox	:	Removed incorrect check for 20 * psh
 *	Michael O'Reilly	:	ack < copied bug fix.
 *	Johannes Stille		:	Misc tcp fixes (not all in yet).
 *		Alan Cox	:	FIN with no memory -> CRASH
 *		Alan Cox	:	Added socket option proto entries.
 *					Also added awareness of them to accept.
 *		Alan Cox	:	Added TCP options (SOL_TCP)
 *		Alan Cox	:	Switched wakeup calls to callbacks,
 *					so the kernel can layer network
 *					sockets.
 *		Alan Cox	:	Use ip_tos/ip_ttl settings.
 *		Alan Cox	:	Handle FIN (more) properly (we hope).
 *		Alan Cox	:	RST frames sent on unsynchronised
 *					state ack error.
 *		Alan Cox	:	Put in missing check for SYN bit.
 *		Alan Cox	:	Added tcp_select_window() aka NET2E
 *					window non shrink trick.
 *		Alan Cox	:	Added a couple of small NET2E timer
 *					fixes
 *		Charles Hedrick :	TCP fixes
 *		Toomas Tamm	:	TCP window fixes
 *		Alan Cox	:	Small URG fix to rlogin ^C ack fight
 *		Charles Hedrick	:	Rewrote most of it to actually work
 *		Linus		:	Rewrote tcp_read() and URG handling
 *					completely
 *		Gerhard Koerting:	Fixed some missing timer handling
 *		Matthew Dillon  :	Reworked TCP machine states as per RFC
 *		Gerhard Koerting:	PC/TCP workarounds
 *		Adam Caldwell	:	Assorted timer/timing errors
 *		Matthew Dillon	:	Fixed another RST bug
 *		Alan Cox	:	Move to kernel side addressing changes.
 *		Alan Cox	:	Beginning work on TCP fastpathing
 *					(not yet usable)
 *		Arnt Gulbrandsen:	Turbocharged tcp_check() routine.
 *		Alan Cox	:	TCP fast path debugging
 *		Alan Cox	:	Window clamping
 *		Michael Riepe	:	Bug in tcp_check()
 *		Matt Dillon	:	More TCP improvements and RST bug fixes
 *		Matt Dillon	:	Yet more small nasties remove from the
 *					TCP code (Be very nice to this man if
 *					tcp finally works 100%) 8)
 *		Alan Cox	:	BSD accept semantics.
 *		Alan Cox	:	Reset on closedown bug.
 *	Peter De Schrijver	:	ENOTCONN check missing in tcp_sendto().
 *		Michael Pall	:	Handle poll() after URG properly in
 *					all cases.
 *		Michael Pall	:	Undo the last fix in tcp_read_urg()
 *					(multi URG PUSH broke rlogin).
 *		Michael Pall	:	Fix the multi URG PUSH problem in
 *					tcp_readable(), poll() after URG
 *					works now.
 *		Michael Pall	:	recv(...,MSG_OOB) never blocks in the
 *					BSD api.
 *		Alan Cox	:	Changed the semantics of sk->socket to
 *					fix a race and a signal problem with
 *					accept() and async I/O.
 *		Alan Cox	:	Relaxed the rules on tcp_sendto().
 *		Yury Shevchuk	:	Really fixed accept() blocking problem.
 *		Craig I. Hagan  :	Allow for BSD compatible TIME_WAIT for
 *					clients/servers which listen in on
 *					fixed ports.
 *		Alan Cox	:	Cleaned the above up and shrank it to
 *					a sensible code size.
 *		Alan Cox	:	Self connect lockup fix.
 *		Alan Cox	:	No connect to multicast.
 *		Ross Biro	:	Close unaccepted children on master
 *					socket close.
 *		Alan Cox	:	Reset tracing code.
 *		Alan Cox	:	Spurious resets on shutdown.
 *		Alan Cox	:	Giant 15 minute/60 second timer error
 *		Alan Cox	:	Small whoops in polling before an
 *					accept.
 *		Alan Cox	:	Kept the state trace facility since
 *					it's handy for debugging.
 *		Alan Cox	:	More reset handler fixes.
 *		Alan Cox	:	Started rewriting the code based on
 *					the RFC's for other useful protocol
 *					references see: Comer, KA9Q NOS, and
 *					for a reference on the difference
 *					between specifications and how BSD
 *					works see the 4.4lite source.
 *		A.N.Kuznetsov	:	Don't time wait on completion of tidy
 *					close.
 *		Linus Torvalds	:	Fin/Shutdown & copied_seq changes.
 *		Linus Torvalds	:	Fixed BSD port reuse to work first syn
 *		Alan Cox	:	Reimplemented timers as per the RFC
 *					and using multiple timers for sanity.
 *		Alan Cox	:	Small bug fixes, and a lot of new
 *					comments.
 *		Alan Cox	:	Fixed dual reader crash by locking
 *					the buffers (much like datagram.c)
 *		Alan Cox	:	Fixed stuck sockets in probe. A probe
 *					now gets fed up of retrying without
 *					(even a no space) answer.
 *		Alan Cox	:	Extracted closing code better
 *		Alan Cox	:	Fixed the closing state machine to
 *					resemble the RFC.
 *		Alan Cox	:	More 'per spec' fixes.
 *		Jorge Cwik	:	Even faster checksumming.
 *		Alan Cox	:	tcp_data() doesn't ack illegal PSH
 *					only frames. At least one pc tcp stack
 *					generates them.
 *		Alan Cox	:	Cache last socket.
 *		Alan Cox	:	Per route irtt.
 *		Matt Day	:	poll()->select() match BSD precisely on error
 *		Alan Cox	:	New buffers
 *		Marc Tamsky	:	Various sk->prot->retransmits and
 *					sk->retransmits misupdating fixed.
 *					Fixed tcp_write_timeout: stuck close,
 *					and TCP syn retries gets used now.
 *		Mark Yarvis	:	In tcp_read_wakeup(), don't send an
 *					ack if state is TCP_CLOSED.
 *		Alan Cox	:	Look up device on a retransmit - routes may
 *					change. Doesn't yet cope with MSS shrink right
 *					but it's a start!
 *		Marc Tamsky	:	Closing in closing fixes.
 *		Mike Shaver	:	RFC1122 verifications.
 *		Alan Cox	:	rcv_saddr errors.
 *		Alan Cox	:	Block double connect().
 *		Alan Cox	:	Small hooks for enSKIP.
 *		Alexey Kuznetsov:	Path MTU discovery.
 *		Alan Cox	:	Support soft errors.
 *		Alan Cox	:	Fix MTU discovery pathological case
 *					when the remote claims no mtu!
 *		Marc Tamsky	:	TCP_CLOSE fix.
 *		Colin (G3TNE)	:	Send a reset on syn ack replies in
 *					window but wrong (fixes NT lpd problems)
 *		Pedro Roque	:	Better TCP window handling, delayed ack.
 *		Joerg Reuter	:	No modification of locked buffers in
 *					tcp_do_retransmit()
 *		Eric Schenk	:	Changed receiver side silly window
 *					avoidance algorithm to BSD style
 *					algorithm. This doubles throughput
 *					against machines running Solaris,
 *					and seems to result in general
 *					improvement.
 *	Stefan Magdalinski	:	adjusted tcp_readable() to fix FIONREAD
 *	Willy Konynenberg	:	Transparent proxying support.
 *	Mike McLagan		:	Routing by source
 *		Keith Owens	:	Do proper merging with partial SKB's in
 *					tcp_do_sendmsg to avoid burstiness.
 *		Eric Schenk	:	Fix fast close down bug with
 *					shutdown() followed by close().
 *		Andi Kleen 	:	Make poll agree with SIGIO
 *	Salvatore Sanfilippo	:	Support SO_LINGER with linger == 1 and
 *					lingertime == 0 (RFC 793 ABORT Call)
 *	Hirokazu Takahashi	:	Use copy_from_user() instead of
 *					csum_and_copy_from_user() if possible.
 *
 *		This program is free software; you can redistribute it and/or
 *		modify it under the terms of the GNU General Public License
 *		as published by the Free Software Foundation; either version
 *		2 of the License, or(at your option) any later version.
 *
 * Description of States:
 *
 *	TCP_SYN_SENT		sent a connection request, waiting for ack
 *
 *	TCP_SYN_RECV		received a connection request, sent ack,
 *				waiting for final ack in three-way handshake.
 *
 *	TCP_ESTABLISHED		connection established
 *
 *	TCP_FIN_WAIT1		our side has shutdown, waiting to complete
 *				transmission of remaining buffered data
 *
 *	TCP_FIN_WAIT2		all buffered data sent, waiting for remote
 *				to shutdown
 *
 *	TCP_CLOSING		both sides have shutdown but we still have
 *				data we have to finish sending
 *
 *	TCP_TIME_WAIT		timeout to catch resent junk before entering
 *				closed, can only be entered from FIN_WAIT2
 *				or CLOSING.  Required because the other end
 *				may not have gotten our last ACK causing it
 *				to retransmit the data packet (which we ignore)
 *
 *	TCP_CLOSE_WAIT		remote side has shutdown and is waiting for
 *				us to finish writing our data and to shutdown
 *				(we have to close() to move on to LAST_ACK)
 *
 *	TCP_LAST_ACK		out side has shutdown after remote has
 *				shutdown.  There may still be data in our
 *				buffer that we have to finish sending
 *
 *	TCP_CLOSE		socket is finished
 */

248
#include <linux/kernel.h>
L
Linus Torvalds 已提交
249 250 251 252 253 254
#include <linux/module.h>
#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/poll.h>
#include <linux/init.h>
#include <linux/fs.h>
J
Jens Axboe 已提交
255
#include <linux/skbuff.h>
256
#include <linux/scatterlist.h>
J
Jens Axboe 已提交
257 258 259
#include <linux/splice.h>
#include <linux/net.h>
#include <linux/socket.h>
L
Linus Torvalds 已提交
260 261
#include <linux/random.h>
#include <linux/bootmem.h>
262 263
#include <linux/highmem.h>
#include <linux/swap.h>
264
#include <linux/cache.h>
H
Herbert Xu 已提交
265
#include <linux/err.h>
266
#include <linux/crypto.h>
L
Linus Torvalds 已提交
267 268 269 270 271

#include <net/icmp.h>
#include <net/tcp.h>
#include <net/xfrm.h>
#include <net/ip.h>
C
Chris Leech 已提交
272
#include <net/netdma.h>
J
Jens Axboe 已提交
273
#include <net/sock.h>
L
Linus Torvalds 已提交
274 275 276 277

#include <asm/uaccess.h>
#include <asm/ioctls.h>

278
int sysctl_tcp_fin_timeout __read_mostly = TCP_FIN_TIMEOUT;
L
Linus Torvalds 已提交
279

280
struct percpu_counter tcp_orphan_count;
281 282
EXPORT_SYMBOL_GPL(tcp_orphan_count);

283 284 285
int sysctl_tcp_mem[3] __read_mostly;
int sysctl_tcp_wmem[3] __read_mostly;
int sysctl_tcp_rmem[3] __read_mostly;
L
Linus Torvalds 已提交
286 287 288 289 290 291 292

EXPORT_SYMBOL(sysctl_tcp_mem);
EXPORT_SYMBOL(sysctl_tcp_rmem);
EXPORT_SYMBOL(sysctl_tcp_wmem);

atomic_t tcp_memory_allocated;	/* Current allocated memory. */
EXPORT_SYMBOL(tcp_memory_allocated);
293 294 295 296 297

/*
 * Current number of TCP sockets.
 */
struct percpu_counter tcp_sockets_allocated;
L
Linus Torvalds 已提交
298 299
EXPORT_SYMBOL(tcp_sockets_allocated);

J
Jens Axboe 已提交
300 301 302 303 304 305 306 307 308
/*
 * TCP splice context
 */
struct tcp_splice_state {
	struct pipe_inode_info *pipe;
	size_t len;
	unsigned int flags;
};

L
Linus Torvalds 已提交
309 310 311
/*
 * Pressure flag: try to collapse.
 * Technical note: it is used by multiple contexts non atomically.
312
 * All the __sk_mem_schedule() is of this nature: accounting
L
Linus Torvalds 已提交
313 314
 * is strict, actions are advisory and have some latency.
 */
315
int tcp_memory_pressure __read_mostly;
L
Linus Torvalds 已提交
316 317 318

EXPORT_SYMBOL(tcp_memory_pressure);

319
void tcp_enter_memory_pressure(struct sock *sk)
L
Linus Torvalds 已提交
320 321
{
	if (!tcp_memory_pressure) {
P
Pavel Emelyanov 已提交
322
		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMEMORYPRESSURES);
L
Linus Torvalds 已提交
323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343
		tcp_memory_pressure = 1;
	}
}

EXPORT_SYMBOL(tcp_enter_memory_pressure);

/*
 *	Wait for a TCP event.
 *
 *	Note that we don't need to lock the socket, as the upper poll layers
 *	take care of normal races (between the test and the event) and we don't
 *	go look at any of the socket buffers directly.
 */
unsigned int tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
{
	unsigned int mask;
	struct sock *sk = sock->sk;
	struct tcp_sock *tp = tcp_sk(sk);

	poll_wait(file, sk->sk_sleep, wait);
	if (sk->sk_state == TCP_LISTEN)
344
		return inet_csk_listen_poll(sk);
L
Linus Torvalds 已提交
345 346

	/* Socket is not locked. We are protected from async events
347 348
	 * by poll logic and correct handling of state changes
	 * made by other threads is impossible in any case.
L
Linus Torvalds 已提交
349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
	 */

	mask = 0;
	if (sk->sk_err)
		mask = POLLERR;

	/*
	 * POLLHUP is certainly not done right. But poll() doesn't
	 * have a notion of HUP in just one direction, and for a
	 * socket the read side is more interesting.
	 *
	 * Some poll() documentation says that POLLHUP is incompatible
	 * with the POLLOUT/POLLWR flags, so somebody should check this
	 * all. But careful, it tends to be safer to return too many
	 * bits than too few, and you can easily break real applications
	 * if you don't tell them that something has hung up!
	 *
	 * Check-me.
	 *
	 * Check number 1. POLLHUP is _UNMASKABLE_ event (see UNIX98 and
	 * our fs/select.c). It means that after we received EOF,
	 * poll always returns immediately, making impossible poll() on write()
	 * in state CLOSE_WAIT. One solution is evident --- to set POLLHUP
	 * if and only if shutdown has been made in both directions.
	 * Actually, it is interesting to look how Solaris and DUX
374
	 * solve this dilemma. I would prefer, if POLLHUP were maskable,
L
Linus Torvalds 已提交
375 376
	 * then we could set it on SND_SHUTDOWN. BTW examples given
	 * in Stevens' books assume exactly this behaviour, it explains
377
	 * why POLLHUP is incompatible with POLLOUT.	--ANK
L
Linus Torvalds 已提交
378 379 380 381 382 383 384
	 *
	 * NOTE. Check for TCP_CLOSE is added. The goal is to prevent
	 * blocking on fresh not-connected or disconnected socket. --ANK
	 */
	if (sk->sk_shutdown == SHUTDOWN_MASK || sk->sk_state == TCP_CLOSE)
		mask |= POLLHUP;
	if (sk->sk_shutdown & RCV_SHUTDOWN)
385
		mask |= POLLIN | POLLRDNORM | POLLRDHUP;
L
Linus Torvalds 已提交
386 387 388

	/* Connected? */
	if ((1 << sk->sk_state) & ~(TCPF_SYN_SENT | TCPF_SYN_RECV)) {
389 390 391 392 393 394 395
		int target = sock_rcvlowat(sk, 0, INT_MAX);

		if (tp->urg_seq == tp->copied_seq &&
		    !sock_flag(sk, SOCK_URGINLINE) &&
		    tp->urg_data)
			target--;

L
Linus Torvalds 已提交
396 397 398
		/* Potential race condition. If read of tp below will
		 * escape above sk->sk_state, we can be illegally awaken
		 * in SYN_* states. */
399
		if (tp->rcv_nxt - tp->copied_seq >= target)
L
Linus Torvalds 已提交
400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446
			mask |= POLLIN | POLLRDNORM;

		if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
			if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk)) {
				mask |= POLLOUT | POLLWRNORM;
			} else {  /* send SIGIO later */
				set_bit(SOCK_ASYNC_NOSPACE,
					&sk->sk_socket->flags);
				set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);

				/* Race breaker. If space is freed after
				 * wspace test but before the flags are set,
				 * IO signal will be lost.
				 */
				if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk))
					mask |= POLLOUT | POLLWRNORM;
			}
		}

		if (tp->urg_data & TCP_URG_VALID)
			mask |= POLLPRI;
	}
	return mask;
}

int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int answ;

	switch (cmd) {
	case SIOCINQ:
		if (sk->sk_state == TCP_LISTEN)
			return -EINVAL;

		lock_sock(sk);
		if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))
			answ = 0;
		else if (sock_flag(sk, SOCK_URGINLINE) ||
			 !tp->urg_data ||
			 before(tp->urg_seq, tp->copied_seq) ||
			 !before(tp->urg_seq, tp->rcv_nxt)) {
			answ = tp->rcv_nxt - tp->copied_seq;

			/* Subtract 1, if FIN is in queue. */
			if (answ && !skb_queue_empty(&sk->sk_receive_queue))
				answ -=
447
		       tcp_hdr((struct sk_buff *)sk->sk_receive_queue.prev)->fin;
L
Linus Torvalds 已提交
448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465
		} else
			answ = tp->urg_seq - tp->copied_seq;
		release_sock(sk);
		break;
	case SIOCATMARK:
		answ = tp->urg_data && tp->urg_seq == tp->copied_seq;
		break;
	case SIOCOUTQ:
		if (sk->sk_state == TCP_LISTEN)
			return -EINVAL;

		if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))
			answ = 0;
		else
			answ = tp->write_seq - tp->snd_una;
		break;
	default:
		return -ENOIOCTLCMD;
466
	}
L
Linus Torvalds 已提交
467 468 469 470 471 472 473 474 475 476 477 478 479 480 481

	return put_user(answ, (int __user *)arg);
}

static inline void tcp_mark_push(struct tcp_sock *tp, struct sk_buff *skb)
{
	TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
	tp->pushed_seq = tp->write_seq;
}

static inline int forced_push(struct tcp_sock *tp)
{
	return after(tp->write_seq, tp->pushed_seq + (tp->max_window >> 1));
}

482
static inline void skb_entail(struct sock *sk, struct sk_buff *skb)
L
Linus Torvalds 已提交
483
{
484
	struct tcp_sock *tp = tcp_sk(sk);
485 486 487 488 489 490
	struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);

	skb->csum    = 0;
	tcb->seq     = tcb->end_seq = tp->write_seq;
	tcb->flags   = TCPCB_FLAG_ACK;
	tcb->sacked  = 0;
L
Linus Torvalds 已提交
491
	skb_header_release(skb);
492
	tcp_add_write_queue_tail(sk, skb);
493 494
	sk->sk_wmem_queued += skb->truesize;
	sk_mem_charge(sk, skb->truesize);
495
	if (tp->nonagle & TCP_NAGLE_PUSH)
496
		tp->nonagle &= ~TCP_NAGLE_PUSH;
L
Linus Torvalds 已提交
497 498 499 500 501
}

static inline void tcp_mark_urg(struct tcp_sock *tp, int flags,
				struct sk_buff *skb)
{
I
Ilpo Järvinen 已提交
502
	if (flags & MSG_OOB)
L
Linus Torvalds 已提交
503 504 505
		tp->snd_up = tp->write_seq;
}

506 507
static inline void tcp_push(struct sock *sk, int flags, int mss_now,
			    int nonagle)
L
Linus Torvalds 已提交
508
{
509 510
	struct tcp_sock *tp = tcp_sk(sk);

511 512
	if (tcp_send_head(sk)) {
		struct sk_buff *skb = tcp_write_queue_tail(sk);
L
Linus Torvalds 已提交
513 514 515
		if (!(flags & MSG_MORE) || forced_push(tp))
			tcp_mark_push(tp, skb);
		tcp_mark_urg(tp, flags, skb);
516
		__tcp_push_pending_frames(sk, mss_now,
L
Linus Torvalds 已提交
517 518 519 520
					  (flags & MSG_MORE) ? TCP_NAGLE_CORK : nonagle);
	}
}

521 522
static int tcp_splice_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb,
				unsigned int offset, size_t len)
J
Jens Axboe 已提交
523 524
{
	struct tcp_splice_state *tss = rd_desc->arg.data;
525
	int ret;
J
Jens Axboe 已提交
526

527 528
	ret = skb_splice_bits(skb, offset, tss->pipe, min(rd_desc->count, len),
			      tss->flags);
529 530 531
	if (ret > 0)
		rd_desc->count -= ret;
	return ret;
J
Jens Axboe 已提交
532 533 534 535 536 537 538
}

static int __tcp_splice_read(struct sock *sk, struct tcp_splice_state *tss)
{
	/* Store TCP splice context information in read_descriptor_t. */
	read_descriptor_t rd_desc = {
		.arg.data = tss,
539
		.count	  = tss->len,
J
Jens Axboe 已提交
540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619
	};

	return tcp_read_sock(sk, &rd_desc, tcp_splice_data_recv);
}

/**
 *  tcp_splice_read - splice data from TCP socket to a pipe
 * @sock:	socket to splice from
 * @ppos:	position (not valid)
 * @pipe:	pipe to splice to
 * @len:	number of bytes to splice
 * @flags:	splice modifier flags
 *
 * Description:
 *    Will read pages from given socket and fill them into a pipe.
 *
 **/
ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos,
			struct pipe_inode_info *pipe, size_t len,
			unsigned int flags)
{
	struct sock *sk = sock->sk;
	struct tcp_splice_state tss = {
		.pipe = pipe,
		.len = len,
		.flags = flags,
	};
	long timeo;
	ssize_t spliced;
	int ret;

	/*
	 * We can't seek on a socket input
	 */
	if (unlikely(*ppos))
		return -ESPIPE;

	ret = spliced = 0;

	lock_sock(sk);

	timeo = sock_rcvtimeo(sk, flags & SPLICE_F_NONBLOCK);
	while (tss.len) {
		ret = __tcp_splice_read(sk, &tss);
		if (ret < 0)
			break;
		else if (!ret) {
			if (spliced)
				break;
			if (sock_flag(sk, SOCK_DONE))
				break;
			if (sk->sk_err) {
				ret = sock_error(sk);
				break;
			}
			if (sk->sk_shutdown & RCV_SHUTDOWN)
				break;
			if (sk->sk_state == TCP_CLOSE) {
				/*
				 * This occurs when user tries to read
				 * from never connected socket.
				 */
				if (!sock_flag(sk, SOCK_DONE))
					ret = -ENOTCONN;
				break;
			}
			if (!timeo) {
				ret = -EAGAIN;
				break;
			}
			sk_wait_data(sk, &timeo);
			if (signal_pending(current)) {
				ret = sock_intr_errno(timeo);
				break;
			}
			continue;
		}
		tss.len -= ret;
		spliced += ret;

620 621
		if (!timeo)
			break;
J
Jens Axboe 已提交
622 623 624 625
		release_sock(sk);
		lock_sock(sk);

		if (sk->sk_err || sk->sk_state == TCP_CLOSE ||
626
		    (sk->sk_shutdown & RCV_SHUTDOWN) ||
J
Jens Axboe 已提交
627 628 629 630 631 632 633 634 635 636 637 638
		    signal_pending(current))
			break;
	}

	release_sock(sk);

	if (spliced)
		return spliced;

	return ret;
}

639
struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp)
640 641 642 643 644 645 646 647
{
	struct sk_buff *skb;

	/* The TCP header must be at least 32-bit aligned.  */
	size = ALIGN(size, 4);

	skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp);
	if (skb) {
648
		if (sk_wmem_schedule(sk, skb->truesize)) {
649 650 651 652 653 654 655 656 657
			/*
			 * Make sure that we have exactly size bytes
			 * available to the caller, no more, no less.
			 */
			skb_reserve(skb, skb_tailroom(skb) - size);
			return skb;
		}
		__kfree_skb(skb);
	} else {
658
		sk->sk_prot->enter_memory_pressure(sk);
659 660 661 662 663
		sk_stream_moderate_sndbuf(sk);
	}
	return NULL;
}

I
Ilpo Järvinen 已提交
664 665 666 667
static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now,
				       int large_allowed)
{
	struct tcp_sock *tp = tcp_sk(sk);
668
	u32 xmit_size_goal, old_size_goal;
I
Ilpo Järvinen 已提交
669 670 671 672 673 674 675 676 677 678

	xmit_size_goal = mss_now;

	if (large_allowed && sk_can_gso(sk)) {
		xmit_size_goal = ((sk->sk_gso_max_size - 1) -
				  inet_csk(sk)->icsk_af_ops->net_header_len -
				  inet_csk(sk)->icsk_ext_hdr_len -
				  tp->tcp_header_len);

		xmit_size_goal = tcp_bound_to_half_wnd(tp, xmit_size_goal);
679 680 681 682 683 684 685 686 687 688 689

		/* We try hard to avoid divides here */
		old_size_goal = tp->xmit_size_goal_segs * mss_now;

		if (likely(old_size_goal <= xmit_size_goal &&
			   old_size_goal + mss_now > xmit_size_goal)) {
			xmit_size_goal = old_size_goal;
		} else {
			tp->xmit_size_goal_segs = xmit_size_goal / mss_now;
			xmit_size_goal = tp->xmit_size_goal_segs * mss_now;
		}
I
Ilpo Järvinen 已提交
690 691
	}

692
	return max(xmit_size_goal, mss_now);
I
Ilpo Järvinen 已提交
693 694 695 696 697 698 699 700 701 702 703 704
}

static int tcp_send_mss(struct sock *sk, int *size_goal, int flags)
{
	int mss_now;

	mss_now = tcp_current_mss(sk);
	*size_goal = tcp_xmit_size_goal(sk, mss_now, !(flags & MSG_OOB));

	return mss_now;
}

L
Linus Torvalds 已提交
705 706 707 708
static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffset,
			 size_t psize, int flags)
{
	struct tcp_sock *tp = tcp_sk(sk);
709
	int mss_now, size_goal;
L
Linus Torvalds 已提交
710 711 712 713 714 715 716 717 718 719 720
	int err;
	ssize_t copied;
	long timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);

	/* Wait for a connection to finish. */
	if ((1 << sk->sk_state) & ~(TCPF_ESTABLISHED | TCPF_CLOSE_WAIT))
		if ((err = sk_stream_wait_connect(sk, &timeo)) != 0)
			goto out_err;

	clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);

I
Ilpo Järvinen 已提交
721
	mss_now = tcp_send_mss(sk, &size_goal, flags);
L
Linus Torvalds 已提交
722 723 724 725
	copied = 0;

	err = -EPIPE;
	if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN))
726
		goto out_err;
L
Linus Torvalds 已提交
727 728

	while (psize > 0) {
729
		struct sk_buff *skb = tcp_write_queue_tail(sk);
L
Linus Torvalds 已提交
730 731 732 733 734
		struct page *page = pages[poffset / PAGE_SIZE];
		int copy, i, can_coalesce;
		int offset = poffset % PAGE_SIZE;
		int size = min_t(size_t, psize, PAGE_SIZE - offset);

735
		if (!tcp_send_head(sk) || (copy = size_goal - skb->len) <= 0) {
L
Linus Torvalds 已提交
736 737 738 739
new_segment:
			if (!sk_stream_memory_free(sk))
				goto wait_for_sndbuf;

740
			skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation);
L
Linus Torvalds 已提交
741 742 743
			if (!skb)
				goto wait_for_memory;

744
			skb_entail(sk, skb);
745
			copy = size_goal;
L
Linus Torvalds 已提交
746 747 748 749 750 751 752 753 754 755 756
		}

		if (copy > size)
			copy = size;

		i = skb_shinfo(skb)->nr_frags;
		can_coalesce = skb_can_coalesce(skb, i, page, offset);
		if (!can_coalesce && i >= MAX_SKB_FRAGS) {
			tcp_mark_push(tp, skb);
			goto new_segment;
		}
757
		if (!sk_wmem_schedule(sk, copy))
L
Linus Torvalds 已提交
758
			goto wait_for_memory;
759

L
Linus Torvalds 已提交
760 761 762 763 764 765 766 767 768 769 770
		if (can_coalesce) {
			skb_shinfo(skb)->frags[i - 1].size += copy;
		} else {
			get_page(page);
			skb_fill_page_desc(skb, i, page, offset, copy);
		}

		skb->len += copy;
		skb->data_len += copy;
		skb->truesize += copy;
		sk->sk_wmem_queued += copy;
771
		sk_mem_charge(sk, copy);
772
		skb->ip_summed = CHECKSUM_PARTIAL;
L
Linus Torvalds 已提交
773 774
		tp->write_seq += copy;
		TCP_SKB_CB(skb)->end_seq += copy;
775
		skb_shinfo(skb)->gso_segs = 0;
L
Linus Torvalds 已提交
776 777 778 779 780 781 782 783 784

		if (!copied)
			TCP_SKB_CB(skb)->flags &= ~TCPCB_FLAG_PSH;

		copied += copy;
		poffset += copy;
		if (!(psize -= copy))
			goto out;

785
		if (skb->len < size_goal || (flags & MSG_OOB))
L
Linus Torvalds 已提交
786 787 788 789
			continue;

		if (forced_push(tp)) {
			tcp_mark_push(tp, skb);
790
			__tcp_push_pending_frames(sk, mss_now, TCP_NAGLE_PUSH);
791
		} else if (skb == tcp_send_head(sk))
L
Linus Torvalds 已提交
792 793 794 795 796 797 798
			tcp_push_one(sk, mss_now);
		continue;

wait_for_sndbuf:
		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
wait_for_memory:
		if (copied)
799
			tcp_push(sk, flags & ~MSG_MORE, mss_now, TCP_NAGLE_PUSH);
L
Linus Torvalds 已提交
800 801 802 803

		if ((err = sk_stream_wait_memory(sk, &timeo)) != 0)
			goto do_error;

I
Ilpo Järvinen 已提交
804
		mss_now = tcp_send_mss(sk, &size_goal, flags);
L
Linus Torvalds 已提交
805 806 807 808
	}

out:
	if (copied)
809
		tcp_push(sk, flags, mss_now, tp->nonagle);
L
Linus Torvalds 已提交
810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825
	return copied;

do_error:
	if (copied)
		goto out;
out_err:
	return sk_stream_error(sk, flags, err);
}

ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset,
		     size_t size, int flags)
{
	ssize_t res;
	struct sock *sk = sock->sk;

	if (!(sk->sk_route_caps & NETIF_F_SG) ||
826
	    !(sk->sk_route_caps & NETIF_F_ALL_CSUM))
L
Linus Torvalds 已提交
827 828 829 830 831 832 833 834 835 836 837 838 839
		return sock_no_sendpage(sock, page, offset, size, flags);

	lock_sock(sk);
	TCP_CHECK_TIMER(sk);
	res = do_tcp_sendpages(sk, &page, offset, size, flags);
	TCP_CHECK_TIMER(sk);
	release_sock(sk);
	return res;
}

#define TCP_PAGE(sk)	(sk->sk_sndmsg_page)
#define TCP_OFF(sk)	(sk->sk_sndmsg_off)

840
static inline int select_size(struct sock *sk)
L
Linus Torvalds 已提交
841
{
842
	struct tcp_sock *tp = tcp_sk(sk);
843
	int tmp = tp->mss_cache;
L
Linus Torvalds 已提交
844

845
	if (sk->sk_route_caps & NETIF_F_SG) {
846
		if (sk_can_gso(sk))
847 848 849 850 851 852 853 854 855
			tmp = 0;
		else {
			int pgbreak = SKB_MAX_HEAD(MAX_TCP_HEADER);

			if (tmp >= pgbreak &&
			    tmp <= pgbreak + (MAX_SKB_FRAGS - 1) * PAGE_SIZE)
				tmp = pgbreak;
		}
	}
L
Linus Torvalds 已提交
856 857 858 859

	return tmp;
}

860
int tcp_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
L
Linus Torvalds 已提交
861 862
		size_t size)
{
863
	struct sock *sk = sock->sk;
L
Linus Torvalds 已提交
864 865 866 867
	struct iovec *iov;
	struct tcp_sock *tp = tcp_sk(sk);
	struct sk_buff *skb;
	int iovlen, flags;
868
	int mss_now, size_goal;
L
Linus Torvalds 已提交
869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885
	int err, copied;
	long timeo;

	lock_sock(sk);
	TCP_CHECK_TIMER(sk);

	flags = msg->msg_flags;
	timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);

	/* Wait for a connection to finish. */
	if ((1 << sk->sk_state) & ~(TCPF_ESTABLISHED | TCPF_CLOSE_WAIT))
		if ((err = sk_stream_wait_connect(sk, &timeo)) != 0)
			goto out_err;

	/* This should be in poll */
	clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);

I
Ilpo Järvinen 已提交
886
	mss_now = tcp_send_mss(sk, &size_goal, flags);
L
Linus Torvalds 已提交
887 888 889 890 891 892 893 894

	/* Ok commence sending. */
	iovlen = msg->msg_iovlen;
	iov = msg->msg_iov;
	copied = 0;

	err = -EPIPE;
	if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN))
895
		goto out_err;
L
Linus Torvalds 已提交
896 897 898 899 900 901 902 903 904 905

	while (--iovlen >= 0) {
		int seglen = iov->iov_len;
		unsigned char __user *from = iov->iov_base;

		iov++;

		while (seglen > 0) {
			int copy;

906
			skb = tcp_write_queue_tail(sk);
L
Linus Torvalds 已提交
907

908
			if (!tcp_send_head(sk) ||
909
			    (copy = size_goal - skb->len) <= 0) {
L
Linus Torvalds 已提交
910 911 912 913 914 915 916 917

new_segment:
				/* Allocate new segment. If the interface is SG,
				 * allocate skb fitting to single page.
				 */
				if (!sk_stream_memory_free(sk))
					goto wait_for_sndbuf;

918 919
				skb = sk_stream_alloc_skb(sk, select_size(sk),
						sk->sk_allocation);
L
Linus Torvalds 已提交
920 921 922 923 924 925
				if (!skb)
					goto wait_for_memory;

				/*
				 * Check whether we can use HW checksum.
				 */
926
				if (sk->sk_route_caps & NETIF_F_ALL_CSUM)
927
					skb->ip_summed = CHECKSUM_PARTIAL;
L
Linus Torvalds 已提交
928

929
				skb_entail(sk, skb);
930
				copy = size_goal;
L
Linus Torvalds 已提交
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967
			}

			/* Try to append data to the end of skb. */
			if (copy > seglen)
				copy = seglen;

			/* Where to copy to? */
			if (skb_tailroom(skb) > 0) {
				/* We have some space in skb head. Superb! */
				if (copy > skb_tailroom(skb))
					copy = skb_tailroom(skb);
				if ((err = skb_add_data(skb, from, copy)) != 0)
					goto do_fault;
			} else {
				int merge = 0;
				int i = skb_shinfo(skb)->nr_frags;
				struct page *page = TCP_PAGE(sk);
				int off = TCP_OFF(sk);

				if (skb_can_coalesce(skb, i, page, off) &&
				    off != PAGE_SIZE) {
					/* We can extend the last page
					 * fragment. */
					merge = 1;
				} else if (i == MAX_SKB_FRAGS ||
					   (!i &&
					   !(sk->sk_route_caps & NETIF_F_SG))) {
					/* Need to add new fragment and cannot
					 * do this because interface is non-SG,
					 * or because all the page slots are
					 * busy. */
					tcp_mark_push(tp, skb);
					goto new_segment;
				} else if (page) {
					if (off == PAGE_SIZE) {
						put_page(page);
						TCP_PAGE(sk) = page = NULL;
968
						off = 0;
L
Linus Torvalds 已提交
969
					}
970
				} else
971
					off = 0;
972 973 974 975

				if (copy > PAGE_SIZE - off)
					copy = PAGE_SIZE - off;

976
				if (!sk_wmem_schedule(sk, copy))
977
					goto wait_for_memory;
L
Linus Torvalds 已提交
978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021

				if (!page) {
					/* Allocate new cache page. */
					if (!(page = sk_stream_alloc_page(sk)))
						goto wait_for_memory;
				}

				/* Time to copy data. We are close to
				 * the end! */
				err = skb_copy_to_page(sk, from, skb, page,
						       off, copy);
				if (err) {
					/* If this page was new, give it to the
					 * socket so it does not get leaked.
					 */
					if (!TCP_PAGE(sk)) {
						TCP_PAGE(sk) = page;
						TCP_OFF(sk) = 0;
					}
					goto do_error;
				}

				/* Update the skb. */
				if (merge) {
					skb_shinfo(skb)->frags[i - 1].size +=
									copy;
				} else {
					skb_fill_page_desc(skb, i, page, off, copy);
					if (TCP_PAGE(sk)) {
						get_page(page);
					} else if (off + copy < PAGE_SIZE) {
						get_page(page);
						TCP_PAGE(sk) = page;
					}
				}

				TCP_OFF(sk) = off + copy;
			}

			if (!copied)
				TCP_SKB_CB(skb)->flags &= ~TCPCB_FLAG_PSH;

			tp->write_seq += copy;
			TCP_SKB_CB(skb)->end_seq += copy;
1022
			skb_shinfo(skb)->gso_segs = 0;
L
Linus Torvalds 已提交
1023 1024 1025 1026 1027 1028

			from += copy;
			copied += copy;
			if ((seglen -= copy) == 0 && iovlen == 0)
				goto out;

1029
			if (skb->len < size_goal || (flags & MSG_OOB))
L
Linus Torvalds 已提交
1030 1031 1032 1033
				continue;

			if (forced_push(tp)) {
				tcp_mark_push(tp, skb);
1034
				__tcp_push_pending_frames(sk, mss_now, TCP_NAGLE_PUSH);
1035
			} else if (skb == tcp_send_head(sk))
L
Linus Torvalds 已提交
1036 1037 1038 1039 1040 1041 1042
				tcp_push_one(sk, mss_now);
			continue;

wait_for_sndbuf:
			set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
wait_for_memory:
			if (copied)
1043
				tcp_push(sk, flags & ~MSG_MORE, mss_now, TCP_NAGLE_PUSH);
L
Linus Torvalds 已提交
1044 1045 1046 1047

			if ((err = sk_stream_wait_memory(sk, &timeo)) != 0)
				goto do_error;

I
Ilpo Järvinen 已提交
1048
			mss_now = tcp_send_mss(sk, &size_goal, flags);
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053
		}
	}

out:
	if (copied)
1054
		tcp_push(sk, flags, mss_now, tp->nonagle);
L
Linus Torvalds 已提交
1055 1056 1057 1058 1059 1060
	TCP_CHECK_TIMER(sk);
	release_sock(sk);
	return copied;

do_fault:
	if (!skb->len) {
1061 1062 1063 1064 1065
		tcp_unlink_write_queue(skb, sk);
		/* It is the one place in all of TCP, except connection
		 * reset, where we can be unlinking the send_head.
		 */
		tcp_check_send_head(sk, skb);
1066
		sk_wmem_free_skb(sk, skb);
L
Linus Torvalds 已提交
1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083
	}

do_error:
	if (copied)
		goto out;
out_err:
	err = sk_stream_error(sk, flags, err);
	TCP_CHECK_TIMER(sk);
	release_sock(sk);
	return err;
}

/*
 *	Handle reading urgent data. BSD has very simple semantics for
 *	this, no blocking and very strange errors 8)
 */

1084
static int tcp_recv_urg(struct sock *sk, struct msghdr *msg, int len, int flags)
L
Linus Torvalds 已提交
1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 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
{
	struct tcp_sock *tp = tcp_sk(sk);

	/* No URG data to read. */
	if (sock_flag(sk, SOCK_URGINLINE) || !tp->urg_data ||
	    tp->urg_data == TCP_URG_READ)
		return -EINVAL;	/* Yes this is right ! */

	if (sk->sk_state == TCP_CLOSE && !sock_flag(sk, SOCK_DONE))
		return -ENOTCONN;

	if (tp->urg_data & TCP_URG_VALID) {
		int err = 0;
		char c = tp->urg_data;

		if (!(flags & MSG_PEEK))
			tp->urg_data = TCP_URG_READ;

		/* Read urgent data. */
		msg->msg_flags |= MSG_OOB;

		if (len > 0) {
			if (!(flags & MSG_TRUNC))
				err = memcpy_toiovec(msg->msg_iov, &c, 1);
			len = 1;
		} else
			msg->msg_flags |= MSG_TRUNC;

		return err ? -EFAULT : len;
	}

	if (sk->sk_state == TCP_CLOSE || (sk->sk_shutdown & RCV_SHUTDOWN))
		return 0;

	/* Fixed the recv(..., MSG_OOB) behaviour.  BSD docs and
	 * the available implementations agree in this case:
	 * this call should never block, independent of the
	 * blocking state of the socket.
	 * Mike <pall@rz.uni-karlsruhe.de>
	 */
	return -EAGAIN;
}

/* Clean up the receive buffer for full frames taken by the user,
 * then send an ACK if necessary.  COPIED is the number of bytes
 * tcp_recvmsg has given to the user so far, it speeds up the
 * calculation of whether or not we must ACK for the sake of
 * a window update.
 */
1134
void tcp_cleanup_rbuf(struct sock *sk, int copied)
L
Linus Torvalds 已提交
1135 1136 1137 1138 1139 1140 1141
{
	struct tcp_sock *tp = tcp_sk(sk);
	int time_to_ack = 0;

#if TCP_DEBUG
	struct sk_buff *skb = skb_peek(&sk->sk_receive_queue);

1142
	WARN_ON(skb && !before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq));
L
Linus Torvalds 已提交
1143 1144
#endif

1145 1146
	if (inet_csk_ack_scheduled(sk)) {
		const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
1147 1148
		   /* Delayed ACKs frequently hit locked sockets during bulk
		    * receive. */
1149
		if (icsk->icsk_ack.blocked ||
L
Linus Torvalds 已提交
1150
		    /* Once-per-two-segments ACK was not sent by tcp_input.c */
1151
		    tp->rcv_nxt - tp->rcv_wup > icsk->icsk_ack.rcv_mss ||
L
Linus Torvalds 已提交
1152 1153 1154 1155 1156 1157
		    /*
		     * If this read emptied read buffer, we send ACK, if
		     * connection is not bidirectional, user drained
		     * receive buffer and there was a small segment
		     * in queue.
		     */
1158 1159 1160 1161 1162
		    (copied > 0 &&
		     ((icsk->icsk_ack.pending & ICSK_ACK_PUSHED2) ||
		      ((icsk->icsk_ack.pending & ICSK_ACK_PUSHED) &&
		       !icsk->icsk_ack.pingpong)) &&
		      !atomic_read(&sk->sk_rmem_alloc)))
L
Linus Torvalds 已提交
1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196
			time_to_ack = 1;
	}

	/* We send an ACK if we can now advertise a non-zero window
	 * which has been raised "significantly".
	 *
	 * Even if window raised up to infinity, do not send window open ACK
	 * in states, where we will not receive more. It is useless.
	 */
	if (copied > 0 && !time_to_ack && !(sk->sk_shutdown & RCV_SHUTDOWN)) {
		__u32 rcv_window_now = tcp_receive_window(tp);

		/* Optimize, __tcp_select_window() is not cheap. */
		if (2*rcv_window_now <= tp->window_clamp) {
			__u32 new_window = __tcp_select_window(sk);

			/* Send ACK now, if this read freed lots of space
			 * in our buffer. Certainly, new_window is new window.
			 * We can advertise it now, if it is not less than current one.
			 * "Lots" means "at least twice" here.
			 */
			if (new_window && new_window >= 2 * rcv_window_now)
				time_to_ack = 1;
		}
	}
	if (time_to_ack)
		tcp_send_ack(sk);
}

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

1197
	NET_INC_STATS_USER(sock_net(sk), LINUX_MIB_TCPPREQUEUED);
L
Linus Torvalds 已提交
1198 1199 1200 1201 1202

	/* RX process wants to run with disabled BHs, though it is not
	 * necessary */
	local_bh_disable();
	while ((skb = __skb_dequeue(&tp->ucopy.prequeue)) != NULL)
P
Peter Zijlstra 已提交
1203
		sk_backlog_rcv(sk, skb);
L
Linus Torvalds 已提交
1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216
	local_bh_enable();

	/* Clear memory counter. */
	tp->ucopy.memory = 0;
}

static inline struct sk_buff *tcp_recv_skb(struct sock *sk, u32 seq, u32 *off)
{
	struct sk_buff *skb;
	u32 offset;

	skb_queue_walk(&sk->sk_receive_queue, skb) {
		offset = seq - TCP_SKB_CB(skb)->seq;
1217
		if (tcp_hdr(skb)->syn)
L
Linus Torvalds 已提交
1218
			offset--;
1219
		if (offset < skb->len || tcp_hdr(skb)->fin) {
L
Linus Torvalds 已提交
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
			*off = offset;
			return skb;
		}
	}
	return NULL;
}

/*
 * This routine provides an alternative to tcp_recvmsg() for routines
 * that would like to handle copying from skbuffs directly in 'sendfile'
 * fashion.
 * Note:
 *	- It is assumed that the socket was locked by the caller.
 *	- The routine does not block.
 *	- At present, there is no support for reading OOB data
 *	  or for 'peeking' the socket using this routine
 *	  (although both would be easy to implement).
 */
int tcp_read_sock(struct sock *sk, read_descriptor_t *desc,
		  sk_read_actor_t recv_actor)
{
	struct sk_buff *skb;
	struct tcp_sock *tp = tcp_sk(sk);
	u32 seq = tp->copied_seq;
	u32 offset;
	int copied = 0;

	if (sk->sk_state == TCP_LISTEN)
		return -ENOTCONN;
	while ((skb = tcp_recv_skb(sk, seq, &offset)) != NULL) {
		if (offset < skb->len) {
1251 1252
			int used;
			size_t len;
L
Linus Torvalds 已提交
1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263

			len = skb->len - offset;
			/* Stop reading if we hit a patch of urgent data */
			if (tp->urg_data) {
				u32 urg_offset = tp->urg_seq - seq;
				if (urg_offset < len)
					len = urg_offset;
				if (!len)
					break;
			}
			used = recv_actor(desc, skb, offset, len);
1264 1265 1266 1267 1268
			if (used < 0) {
				if (!copied)
					copied = used;
				break;
			} else if (used <= len) {
L
Linus Torvalds 已提交
1269 1270 1271 1272
				seq += used;
				copied += used;
				offset += used;
			}
1273 1274 1275 1276 1277 1278 1279 1280
			/*
			 * If recv_actor drops the lock (e.g. TCP splice
			 * receive) the skb pointer might be invalid when
			 * getting here: tcp_collapse might have deleted it
			 * while aggregating skbs from the socket queue.
			 */
			skb = tcp_recv_skb(sk, seq-1, &offset);
			if (!skb || (offset+1 != skb->len))
L
Linus Torvalds 已提交
1281 1282
				break;
		}
1283
		if (tcp_hdr(skb)->fin) {
1284
			sk_eat_skb(sk, skb, 0);
L
Linus Torvalds 已提交
1285 1286 1287
			++seq;
			break;
		}
1288
		sk_eat_skb(sk, skb, 0);
L
Linus Torvalds 已提交
1289 1290 1291 1292 1293 1294 1295 1296
		if (!desc->count)
			break;
	}
	tp->copied_seq = seq;

	tcp_rcv_space_adjust(sk);

	/* Clean up data we have read: This will do ACK frames. */
1297
	if (copied > 0)
1298
		tcp_cleanup_rbuf(sk, copied);
L
Linus Torvalds 已提交
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321
	return copied;
}

/*
 *	This routine copies from a sock struct into the user buffer.
 *
 *	Technical note: in 2.3 we work on _locked_ socket, so that
 *	tricks with *seq access order and skb->users are not required.
 *	Probably, code can be easily improved even more.
 */

int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
		size_t len, int nonblock, int flags, int *addr_len)
{
	struct tcp_sock *tp = tcp_sk(sk);
	int copied = 0;
	u32 peek_seq;
	u32 *seq;
	unsigned long used;
	int err;
	int target;		/* Read at least this many bytes */
	long timeo;
	struct task_struct *user_recv = NULL;
C
Chris Leech 已提交
1322
	int copied_early = 0;
1323
	struct sk_buff *skb;
L
Linus Torvalds 已提交
1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346

	lock_sock(sk);

	TCP_CHECK_TIMER(sk);

	err = -ENOTCONN;
	if (sk->sk_state == TCP_LISTEN)
		goto out;

	timeo = sock_rcvtimeo(sk, nonblock);

	/* Urgent data needs to be handled specially. */
	if (flags & MSG_OOB)
		goto recv_urg;

	seq = &tp->copied_seq;
	if (flags & MSG_PEEK) {
		peek_seq = tp->copied_seq;
		seq = &peek_seq;
	}

	target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);

C
Chris Leech 已提交
1347 1348 1349
#ifdef CONFIG_NET_DMA
	tp->ucopy.dma_chan = NULL;
	preempt_disable();
1350
	skb = skb_peek_tail(&sk->sk_receive_queue);
A
Andrew Morton 已提交
1351 1352 1353 1354 1355 1356 1357 1358
	{
		int available = 0;

		if (skb)
			available = TCP_SKB_CB(skb)->seq + skb->len - (*seq);
		if ((available < target) &&
		    (len > sysctl_tcp_dma_copybreak) && !(flags & MSG_PEEK) &&
		    !sysctl_tcp_low_latency &&
1359
		    dma_find_channel(DMA_MEMCPY)) {
A
Andrew Morton 已提交
1360 1361 1362 1363 1364 1365 1366
			preempt_enable_no_resched();
			tp->ucopy.pinned_list =
					dma_pin_iovec_pages(msg->msg_iov, len);
		} else {
			preempt_enable_no_resched();
		}
	}
C
Chris Leech 已提交
1367 1368
#endif

L
Linus Torvalds 已提交
1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397
	do {
		u32 offset;

		/* Are we at urgent data? Stop if we have read anything or have SIGURG pending. */
		if (tp->urg_data && tp->urg_seq == *seq) {
			if (copied)
				break;
			if (signal_pending(current)) {
				copied = timeo ? sock_intr_errno(timeo) : -EAGAIN;
				break;
			}
		}

		/* Next get a buffer. */

		skb = skb_peek(&sk->sk_receive_queue);
		do {
			if (!skb)
				break;

			/* Now that we have two receive queues this
			 * shouldn't happen.
			 */
			if (before(*seq, TCP_SKB_CB(skb)->seq)) {
				printk(KERN_INFO "recvmsg bug: copied %X "
				       "seq %X\n", *seq, TCP_SKB_CB(skb)->seq);
				break;
			}
			offset = *seq - TCP_SKB_CB(skb)->seq;
1398
			if (tcp_hdr(skb)->syn)
L
Linus Torvalds 已提交
1399 1400 1401
				offset--;
			if (offset < skb->len)
				goto found_ok_skb;
1402
			if (tcp_hdr(skb)->fin)
L
Linus Torvalds 已提交
1403
				goto found_fin_ok;
1404
			WARN_ON(!(flags & MSG_PEEK));
L
Linus Torvalds 已提交
1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417
			skb = skb->next;
		} while (skb != (struct sk_buff *)&sk->sk_receive_queue);

		/* Well, if we have backlog, try to process it now yet. */

		if (copied >= target && !sk->sk_backlog.tail)
			break;

		if (copied) {
			if (sk->sk_err ||
			    sk->sk_state == TCP_CLOSE ||
			    (sk->sk_shutdown & RCV_SHUTDOWN) ||
			    !timeo ||
1418
			    signal_pending(current))
L
Linus Torvalds 已提交
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
				break;
		} else {
			if (sock_flag(sk, SOCK_DONE))
				break;

			if (sk->sk_err) {
				copied = sock_error(sk);
				break;
			}

			if (sk->sk_shutdown & RCV_SHUTDOWN)
				break;

			if (sk->sk_state == TCP_CLOSE) {
				if (!sock_flag(sk, SOCK_DONE)) {
					/* This occurs when user tries to read
					 * from never connected socket.
					 */
					copied = -ENOTCONN;
					break;
				}
				break;
			}

			if (!timeo) {
				copied = -EAGAIN;
				break;
			}

			if (signal_pending(current)) {
				copied = sock_intr_errno(timeo);
				break;
			}
		}

1454
		tcp_cleanup_rbuf(sk, copied);
L
Linus Torvalds 已提交
1455

1456
		if (!sysctl_tcp_low_latency && tp->ucopy.task == user_recv) {
L
Linus Torvalds 已提交
1457 1458 1459 1460 1461 1462 1463 1464 1465
			/* Install new reader */
			if (!user_recv && !(flags & (MSG_TRUNC | MSG_PEEK))) {
				user_recv = current;
				tp->ucopy.task = user_recv;
				tp->ucopy.iov = msg->msg_iov;
			}

			tp->ucopy.len = len;

1466 1467
			WARN_ON(tp->copied_seq != tp->rcv_nxt &&
				!(flags & (MSG_PEEK | MSG_TRUNC)));
L
Linus Torvalds 已提交
1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494

			/* Ugly... If prequeue is not empty, we have to
			 * process it before releasing socket, otherwise
			 * order will be broken at second iteration.
			 * More elegant solution is required!!!
			 *
			 * Look: we have the following (pseudo)queues:
			 *
			 * 1. packets in flight
			 * 2. backlog
			 * 3. prequeue
			 * 4. receive_queue
			 *
			 * Each queue can be processed only if the next ones
			 * are empty. At this point we have empty receive_queue.
			 * But prequeue _can_ be not empty after 2nd iteration,
			 * when we jumped to start of loop because backlog
			 * processing added something to receive_queue.
			 * We cannot release_sock(), because backlog contains
			 * packets arrived _after_ prequeued ones.
			 *
			 * Shortly, algorithm is clear --- to process all
			 * the queues in order. We could make it more directly,
			 * requeueing packets from backlog to prequeue, if
			 * is not empty. It is more elegant, but eats cycles,
			 * unfortunately.
			 */
1495
			if (!skb_queue_empty(&tp->ucopy.prequeue))
L
Linus Torvalds 已提交
1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507
				goto do_prequeue;

			/* __ Set realtime policy in scheduler __ */
		}

		if (copied >= target) {
			/* Do not sleep, just process backlog. */
			release_sock(sk);
			lock_sock(sk);
		} else
			sk_wait_data(sk, &timeo);

C
Chris Leech 已提交
1508 1509 1510 1511
#ifdef CONFIG_NET_DMA
		tp->ucopy.wakeup = 0;
#endif

L
Linus Torvalds 已提交
1512 1513 1514 1515 1516 1517
		if (user_recv) {
			int chunk;

			/* __ Restore normal policy in scheduler __ */

			if ((chunk = len - tp->ucopy.len) != 0) {
1518
				NET_ADD_STATS_USER(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMBACKLOG, chunk);
L
Linus Torvalds 已提交
1519 1520 1521 1522 1523
				len -= chunk;
				copied += chunk;
			}

			if (tp->rcv_nxt == tp->copied_seq &&
1524
			    !skb_queue_empty(&tp->ucopy.prequeue)) {
L
Linus Torvalds 已提交
1525 1526 1527 1528
do_prequeue:
				tcp_prequeue_process(sk);

				if ((chunk = len - tp->ucopy.len) != 0) {
1529
					NET_ADD_STATS_USER(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk);
L
Linus Torvalds 已提交
1530 1531 1532 1533 1534 1535 1536 1537
					len -= chunk;
					copied += chunk;
				}
			}
		}
		if ((flags & MSG_PEEK) && peek_seq != tp->copied_seq) {
			if (net_ratelimit())
				printk(KERN_DEBUG "TCP(%s:%d): Application bug, race in MSG_PEEK.\n",
1538
				       current->comm, task_pid_nr(current));
L
Linus Torvalds 已提交
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
			peek_seq = tp->copied_seq;
		}
		continue;

	found_ok_skb:
		/* Ok so how much can we use? */
		used = skb->len - offset;
		if (len < used)
			used = len;

		/* Do we have urgent data here? */
		if (tp->urg_data) {
			u32 urg_offset = tp->urg_seq - *seq;
			if (urg_offset < used) {
				if (!urg_offset) {
					if (!sock_flag(sk, SOCK_URGINLINE)) {
						++*seq;
						offset++;
						used--;
						if (!used)
							goto skip_copy;
					}
				} else
					used = urg_offset;
			}
		}

		if (!(flags & MSG_TRUNC)) {
C
Chris Leech 已提交
1567 1568
#ifdef CONFIG_NET_DMA
			if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
1569
				tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY);
C
Chris Leech 已提交
1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599

			if (tp->ucopy.dma_chan) {
				tp->ucopy.dma_cookie = dma_skb_copy_datagram_iovec(
					tp->ucopy.dma_chan, skb, offset,
					msg->msg_iov, used,
					tp->ucopy.pinned_list);

				if (tp->ucopy.dma_cookie < 0) {

					printk(KERN_ALERT "dma_cookie < 0\n");

					/* Exception. Bailout! */
					if (!copied)
						copied = -EFAULT;
					break;
				}
				if ((offset + used) == skb->len)
					copied_early = 1;

			} else
#endif
			{
				err = skb_copy_datagram_iovec(skb, offset,
						msg->msg_iov, used);
				if (err) {
					/* Exception. Bailout! */
					if (!copied)
						copied = -EFAULT;
					break;
				}
L
Linus Torvalds 已提交
1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611
			}
		}

		*seq += used;
		copied += used;
		len -= used;

		tcp_rcv_space_adjust(sk);

skip_copy:
		if (tp->urg_data && after(tp->copied_seq, tp->urg_seq)) {
			tp->urg_data = 0;
1612
			tcp_fast_path_check(sk);
L
Linus Torvalds 已提交
1613 1614 1615 1616
		}
		if (used + offset < skb->len)
			continue;

1617
		if (tcp_hdr(skb)->fin)
L
Linus Torvalds 已提交
1618
			goto found_fin_ok;
C
Chris Leech 已提交
1619 1620 1621 1622
		if (!(flags & MSG_PEEK)) {
			sk_eat_skb(sk, skb, copied_early);
			copied_early = 0;
		}
L
Linus Torvalds 已提交
1623 1624 1625 1626 1627
		continue;

	found_fin_ok:
		/* Process the FIN. */
		++*seq;
C
Chris Leech 已提交
1628 1629 1630 1631
		if (!(flags & MSG_PEEK)) {
			sk_eat_skb(sk, skb, copied_early);
			copied_early = 0;
		}
L
Linus Torvalds 已提交
1632 1633 1634 1635
		break;
	} while (len > 0);

	if (user_recv) {
1636
		if (!skb_queue_empty(&tp->ucopy.prequeue)) {
L
Linus Torvalds 已提交
1637 1638 1639 1640 1641 1642 1643
			int chunk;

			tp->ucopy.len = copied > 0 ? len : 0;

			tcp_prequeue_process(sk);

			if (copied > 0 && (chunk = len - tp->ucopy.len) != 0) {
1644
				NET_ADD_STATS_USER(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk);
L
Linus Torvalds 已提交
1645 1646 1647 1648 1649 1650 1651 1652 1653
				len -= chunk;
				copied += chunk;
			}
		}

		tp->ucopy.task = NULL;
		tp->ucopy.len = 0;
	}

C
Chris Leech 已提交
1654 1655 1656 1657 1658 1659 1660
#ifdef CONFIG_NET_DMA
	if (tp->ucopy.dma_chan) {
		dma_cookie_t done, used;

		dma_async_memcpy_issue_pending(tp->ucopy.dma_chan);

		while (dma_async_memcpy_complete(tp->ucopy.dma_chan,
1661 1662
						 tp->ucopy.dma_cookie, &done,
						 &used) == DMA_IN_PROGRESS) {
C
Chris Leech 已提交
1663 1664 1665
			/* do partial cleanup of sk_async_wait_queue */
			while ((skb = skb_peek(&sk->sk_async_wait_queue)) &&
			       (dma_async_is_complete(skb->dma_cookie, done,
1666
						      used) == DMA_SUCCESS)) {
C
Chris Leech 已提交
1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681
				__skb_dequeue(&sk->sk_async_wait_queue);
				kfree_skb(skb);
			}
		}

		/* Safe to free early-copied skbs now */
		__skb_queue_purge(&sk->sk_async_wait_queue);
		tp->ucopy.dma_chan = NULL;
	}
	if (tp->ucopy.pinned_list) {
		dma_unpin_iovec_pages(tp->ucopy.pinned_list);
		tp->ucopy.pinned_list = NULL;
	}
#endif

L
Linus Torvalds 已提交
1682 1683 1684 1685 1686
	/* According to UNIX98, msg_name/msg_namelen are ignored
	 * on connected socket. I was just happy when found this 8) --ANK
	 */

	/* Clean up data we have read: This will do ACK frames. */
1687
	tcp_cleanup_rbuf(sk, copied);
L
Linus Torvalds 已提交
1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698

	TCP_CHECK_TIMER(sk);
	release_sock(sk);
	return copied;

out:
	TCP_CHECK_TIMER(sk);
	release_sock(sk);
	return err;

recv_urg:
1699
	err = tcp_recv_urg(sk, msg, len, flags);
L
Linus Torvalds 已提交
1700 1701 1702
	goto out;
}

I
Ilpo Järvinen 已提交
1703 1704 1705 1706 1707 1708 1709
void tcp_set_state(struct sock *sk, int state)
{
	int oldstate = sk->sk_state;

	switch (state) {
	case TCP_ESTABLISHED:
		if (oldstate != TCP_ESTABLISHED)
P
Pavel Emelyanov 已提交
1710
			TCP_INC_STATS(sock_net(sk), TCP_MIB_CURRESTAB);
I
Ilpo Järvinen 已提交
1711 1712 1713 1714
		break;

	case TCP_CLOSE:
		if (oldstate == TCP_CLOSE_WAIT || oldstate == TCP_ESTABLISHED)
P
Pavel Emelyanov 已提交
1715
			TCP_INC_STATS(sock_net(sk), TCP_MIB_ESTABRESETS);
I
Ilpo Järvinen 已提交
1716 1717 1718 1719

		sk->sk_prot->unhash(sk);
		if (inet_csk(sk)->icsk_bind_hash &&
		    !(sk->sk_userlocks & SOCK_BINDPORT_LOCK))
1720
			inet_put_port(sk);
I
Ilpo Järvinen 已提交
1721 1722
		/* fall through */
	default:
1723
		if (oldstate == TCP_ESTABLISHED)
P
Pavel Emelyanov 已提交
1724
			TCP_DEC_STATS(sock_net(sk), TCP_MIB_CURRESTAB);
I
Ilpo Järvinen 已提交
1725 1726 1727 1728 1729 1730 1731 1732
	}

	/* Change state AFTER socket is unhashed to avoid closed
	 * socket sitting in hash tables.
	 */
	sk->sk_state = state;

#ifdef STATE_TRACE
1733
	SOCK_DEBUG(sk, "TCP sk=%p, State %s -> %s\n", sk, statename[oldstate], statename[state]);
I
Ilpo Järvinen 已提交
1734 1735 1736 1737
#endif
}
EXPORT_SYMBOL_GPL(tcp_set_state);

L
Linus Torvalds 已提交
1738 1739 1740 1741 1742 1743 1744
/*
 *	State processing on a close. This implements the state shift for
 *	sending our FIN frame. Note that we only send a FIN for some
 *	states. A shutdown() may have already sent the FIN, or we may be
 *	closed.
 */

1745
static const unsigned char new_state[16] = {
L
Linus Torvalds 已提交
1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772
  /* current state:        new state:      action:	*/
  /* (Invalid)		*/ TCP_CLOSE,
  /* TCP_ESTABLISHED	*/ TCP_FIN_WAIT1 | TCP_ACTION_FIN,
  /* TCP_SYN_SENT	*/ TCP_CLOSE,
  /* TCP_SYN_RECV	*/ TCP_FIN_WAIT1 | TCP_ACTION_FIN,
  /* TCP_FIN_WAIT1	*/ TCP_FIN_WAIT1,
  /* TCP_FIN_WAIT2	*/ TCP_FIN_WAIT2,
  /* TCP_TIME_WAIT	*/ TCP_CLOSE,
  /* TCP_CLOSE		*/ TCP_CLOSE,
  /* TCP_CLOSE_WAIT	*/ TCP_LAST_ACK  | TCP_ACTION_FIN,
  /* TCP_LAST_ACK	*/ TCP_LAST_ACK,
  /* TCP_LISTEN		*/ TCP_CLOSE,
  /* TCP_CLOSING	*/ TCP_CLOSING,
};

static int tcp_close_state(struct sock *sk)
{
	int next = (int)new_state[sk->sk_state];
	int ns = next & TCP_STATE_MASK;

	tcp_set_state(sk, ns);

	return next & TCP_ACTION_FIN;
}

/*
 *	Shutdown the sending side of a connection. Much like close except
1773
 *	that we don't receive shut down or sock_set_flag(sk, SOCK_DEAD).
L
Linus Torvalds 已提交
1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798
 */

void tcp_shutdown(struct sock *sk, int how)
{
	/*	We need to grab some memory, and put together a FIN,
	 *	and then put it into the queue to be sent.
	 *		Tim MacKenzie(tym@dibbler.cs.monash.edu.au) 4 Dec '92.
	 */
	if (!(how & SEND_SHUTDOWN))
		return;

	/* If we've already sent a FIN, or it's a closed state, skip this. */
	if ((1 << sk->sk_state) &
	    (TCPF_ESTABLISHED | TCPF_SYN_SENT |
	     TCPF_SYN_RECV | TCPF_CLOSE_WAIT)) {
		/* Clear out any half completed packets.  FIN if needed. */
		if (tcp_close_state(sk))
			tcp_send_fin(sk);
	}
}

void tcp_close(struct sock *sk, long timeout)
{
	struct sk_buff *skb;
	int data_was_unread = 0;
H
Herbert Xu 已提交
1799
	int state;
L
Linus Torvalds 已提交
1800 1801 1802 1803 1804 1805 1806 1807

	lock_sock(sk);
	sk->sk_shutdown = SHUTDOWN_MASK;

	if (sk->sk_state == TCP_LISTEN) {
		tcp_set_state(sk, TCP_CLOSE);

		/* Special case. */
1808
		inet_csk_listen_stop(sk);
L
Linus Torvalds 已提交
1809 1810 1811 1812 1813 1814 1815 1816 1817 1818

		goto adjudge_to_death;
	}

	/*  We need to flush the recv. buffs.  We do this only on the
	 *  descriptor close, not protocol-sourced closes, because the
	 *  reader process may not have drained the data yet!
	 */
	while ((skb = __skb_dequeue(&sk->sk_receive_queue)) != NULL) {
		u32 len = TCP_SKB_CB(skb)->end_seq - TCP_SKB_CB(skb)->seq -
1819
			  tcp_hdr(skb)->fin;
L
Linus Torvalds 已提交
1820 1821 1822 1823
		data_was_unread += len;
		__kfree_skb(skb);
	}

1824
	sk_mem_reclaim(sk);
L
Linus Torvalds 已提交
1825

1826 1827 1828 1829 1830 1831
	/* As outlined in RFC 2525, section 2.17, we send a RST here because
	 * data was lost. To witness the awful effects of the old behavior of
	 * always doing a FIN, run an older 2.1.x kernel or 2.0.x, start a bulk
	 * GET in an FTP client, suspend the process, wait for the client to
	 * advertise a zero window, then kill -9 the FTP client, wheee...
	 * Note: timeout is always zero in such a case.
L
Linus Torvalds 已提交
1832 1833 1834
	 */
	if (data_was_unread) {
		/* Unread data was tossed, zap the connection. */
1835
		NET_INC_STATS_USER(sock_net(sk), LINUX_MIB_TCPABORTONCLOSE);
L
Linus Torvalds 已提交
1836 1837 1838 1839 1840
		tcp_set_state(sk, TCP_CLOSE);
		tcp_send_active_reset(sk, GFP_KERNEL);
	} else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) {
		/* Check zero linger _after_ checking for unread data. */
		sk->sk_prot->disconnect(sk, 0);
1841
		NET_INC_STATS_USER(sock_net(sk), LINUX_MIB_TCPABORTONDATA);
L
Linus Torvalds 已提交
1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873
	} else if (tcp_close_state(sk)) {
		/* We FIN if the application ate all the data before
		 * zapping the connection.
		 */

		/* RED-PEN. Formally speaking, we have broken TCP state
		 * machine. State transitions:
		 *
		 * TCP_ESTABLISHED -> TCP_FIN_WAIT1
		 * TCP_SYN_RECV	-> TCP_FIN_WAIT1 (forget it, it's impossible)
		 * TCP_CLOSE_WAIT -> TCP_LAST_ACK
		 *
		 * are legal only when FIN has been sent (i.e. in window),
		 * rather than queued out of window. Purists blame.
		 *
		 * F.e. "RFC state" is ESTABLISHED,
		 * if Linux state is FIN-WAIT-1, but FIN is still not sent.
		 *
		 * The visible declinations are that sometimes
		 * we enter time-wait state, when it is not required really
		 * (harmless), do not send active resets, when they are
		 * required by specs (TCP_ESTABLISHED, TCP_CLOSE_WAIT, when
		 * they look as CLOSING or LAST_ACK for Linux)
		 * Probably, I missed some more holelets.
		 * 						--ANK
		 */
		tcp_send_fin(sk);
	}

	sk_stream_wait_close(sk, timeout);

adjudge_to_death:
H
Herbert Xu 已提交
1874 1875 1876 1877
	state = sk->sk_state;
	sock_hold(sk);
	sock_orphan(sk);

L
Linus Torvalds 已提交
1878 1879 1880 1881 1882 1883 1884 1885 1886
	/* It is the last release_sock in its life. It will remove backlog. */
	release_sock(sk);


	/* Now socket is owned by kernel and we acquire BH lock
	   to finish close. No need to check for user refs.
	 */
	local_bh_disable();
	bh_lock_sock(sk);
1887
	WARN_ON(sock_owned_by_user(sk));
L
Linus Torvalds 已提交
1888

H
Herbert Xu 已提交
1889 1890
	percpu_counter_inc(sk->sk_prot->orphan_count);

H
Herbert Xu 已提交
1891 1892 1893
	/* Have we already been destroyed by a softirq or backlog? */
	if (state != TCP_CLOSE && sk->sk_state == TCP_CLOSE)
		goto out;
L
Linus Torvalds 已提交
1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913

	/*	This is a (useful) BSD violating of the RFC. There is a
	 *	problem with TCP as specified in that the other end could
	 *	keep a socket open forever with no application left this end.
	 *	We use a 3 minute timeout (about the same as BSD) then kill
	 *	our end. If they send after that then tough - BUT: long enough
	 *	that we won't make the old 4*rto = almost no time - whoops
	 *	reset mistake.
	 *
	 *	Nope, it was not mistake. It is really desired behaviour
	 *	f.e. on http servers, when such sockets are useless, but
	 *	consume significant resources. Let's do it with special
	 *	linger2	option.					--ANK
	 */

	if (sk->sk_state == TCP_FIN_WAIT2) {
		struct tcp_sock *tp = tcp_sk(sk);
		if (tp->linger2 < 0) {
			tcp_set_state(sk, TCP_CLOSE);
			tcp_send_active_reset(sk, GFP_ATOMIC);
1914 1915
			NET_INC_STATS_BH(sock_net(sk),
					LINUX_MIB_TCPABORTONLINGER);
L
Linus Torvalds 已提交
1916
		} else {
1917
			const int tmo = tcp_fin_time(sk);
L
Linus Torvalds 已提交
1918 1919

			if (tmo > TCP_TIMEWAIT_LEN) {
1920 1921
				inet_csk_reset_keepalive_timer(sk,
						tmo - TCP_TIMEWAIT_LEN);
L
Linus Torvalds 已提交
1922 1923 1924 1925 1926 1927 1928
			} else {
				tcp_time_wait(sk, TCP_FIN_WAIT2, tmo);
				goto out;
			}
		}
	}
	if (sk->sk_state != TCP_CLOSE) {
1929 1930 1931
		int orphan_count = percpu_counter_read_positive(
						sk->sk_prot->orphan_count);

1932
		sk_mem_reclaim(sk);
1933
		if (tcp_too_many_orphans(sk, orphan_count)) {
L
Linus Torvalds 已提交
1934 1935 1936 1937 1938
			if (net_ratelimit())
				printk(KERN_INFO "TCP: too many of orphaned "
				       "sockets\n");
			tcp_set_state(sk, TCP_CLOSE);
			tcp_send_active_reset(sk, GFP_ATOMIC);
1939 1940
			NET_INC_STATS_BH(sock_net(sk),
					LINUX_MIB_TCPABORTONMEMORY);
L
Linus Torvalds 已提交
1941 1942 1943 1944
		}
	}

	if (sk->sk_state == TCP_CLOSE)
1945
		inet_csk_destroy_sock(sk);
L
Linus Torvalds 已提交
1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965
	/* Otherwise, socket is reprieved until protocol close. */

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

/* These states need RST on ABORT according to RFC793 */

static inline int tcp_need_reset(int state)
{
	return (1 << state) &
	       (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT | TCPF_FIN_WAIT1 |
		TCPF_FIN_WAIT2 | TCPF_SYN_RECV);
}

int tcp_disconnect(struct sock *sk, int flags)
{
	struct inet_sock *inet = inet_sk(sk);
1966
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
1967 1968 1969 1970 1971 1972 1973 1974 1975
	struct tcp_sock *tp = tcp_sk(sk);
	int err = 0;
	int old_state = sk->sk_state;

	if (old_state != TCP_CLOSE)
		tcp_set_state(sk, TCP_CLOSE);

	/* ABORT function of RFC793 */
	if (old_state == TCP_LISTEN) {
1976
		inet_csk_listen_stop(sk);
L
Linus Torvalds 已提交
1977 1978 1979
	} else if (tcp_need_reset(old_state) ||
		   (tp->snd_nxt != tp->write_seq &&
		    (1 << old_state) & (TCPF_CLOSING | TCPF_LAST_ACK))) {
S
Stephen Hemminger 已提交
1980
		/* The last check adjusts for discrepancy of Linux wrt. RFC
L
Linus Torvalds 已提交
1981 1982 1983 1984 1985 1986 1987 1988 1989
		 * states
		 */
		tcp_send_active_reset(sk, gfp_any());
		sk->sk_err = ECONNRESET;
	} else if (old_state == TCP_SYN_SENT)
		sk->sk_err = ECONNRESET;

	tcp_clear_xmit_timers(sk);
	__skb_queue_purge(&sk->sk_receive_queue);
1990
	tcp_write_queue_purge(sk);
L
Linus Torvalds 已提交
1991
	__skb_queue_purge(&tp->out_of_order_queue);
C
Chris Leech 已提交
1992 1993 1994
#ifdef CONFIG_NET_DMA
	__skb_queue_purge(&sk->sk_async_wait_queue);
#endif
L
Linus Torvalds 已提交
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

	inet->dport = 0;

	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
		inet_reset_saddr(sk);

	sk->sk_shutdown = 0;
	sock_reset_flag(sk, SOCK_DONE);
	tp->srtt = 0;
	if ((tp->write_seq += tp->max_window + 2) == 0)
		tp->write_seq = 1;
2006
	icsk->icsk_backoff = 0;
L
Linus Torvalds 已提交
2007
	tp->snd_cwnd = 2;
2008
	icsk->icsk_probes_out = 0;
L
Linus Torvalds 已提交
2009 2010 2011
	tp->packets_out = 0;
	tp->snd_ssthresh = 0x7fffffff;
	tp->snd_cwnd_cnt = 0;
2012
	tp->bytes_acked = 0;
2013
	tcp_set_ca_state(sk, TCP_CA_Open);
L
Linus Torvalds 已提交
2014
	tcp_clear_retrans(tp);
2015
	inet_csk_delack_init(sk);
2016
	tcp_init_send_head(sk);
2017
	memset(&tp->rx_opt, 0, sizeof(tp->rx_opt));
L
Linus Torvalds 已提交
2018 2019
	__sk_dst_reset(sk);

2020
	WARN_ON(inet->num && !icsk->icsk_bind_hash);
L
Linus Torvalds 已提交
2021 2022 2023 2024 2025 2026 2027 2028

	sk->sk_error_report(sk);
	return err;
}

/*
 *	Socket option code for TCP.
 */
2029 2030
static int do_tcp_setsockopt(struct sock *sk, int level,
		int optname, char __user *optval, int optlen)
L
Linus Torvalds 已提交
2031 2032
{
	struct tcp_sock *tp = tcp_sk(sk);
2033
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2034 2035 2036
	int val;
	int err = 0;

2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050
	/* This is a string value all the others are int's */
	if (optname == TCP_CONGESTION) {
		char name[TCP_CA_NAME_MAX];

		if (optlen < 1)
			return -EINVAL;

		val = strncpy_from_user(name, optval,
					min(TCP_CA_NAME_MAX-1, optlen));
		if (val < 0)
			return -EFAULT;
		name[val] = 0;

		lock_sock(sk);
2051
		err = tcp_set_congestion_control(sk, name);
2052 2053 2054 2055
		release_sock(sk);
		return err;
	}

L
Linus Torvalds 已提交
2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086
	if (optlen < sizeof(int))
		return -EINVAL;

	if (get_user(val, (int __user *)optval))
		return -EFAULT;

	lock_sock(sk);

	switch (optname) {
	case TCP_MAXSEG:
		/* Values greater than interface MTU won't take effect. However
		 * at the point when this call is done we typically don't yet
		 * know which interface is going to be used */
		if (val < 8 || val > MAX_TCP_WINDOW) {
			err = -EINVAL;
			break;
		}
		tp->rx_opt.user_mss = val;
		break;

	case TCP_NODELAY:
		if (val) {
			/* TCP_NODELAY is weaker than TCP_CORK, so that
			 * this option on corked socket is remembered, but
			 * it is not activated until cork is cleared.
			 *
			 * However, when TCP_NODELAY is set we make
			 * an explicit push, which overrides even TCP_CORK
			 * for currently queued segments.
			 */
			tp->nonagle |= TCP_NAGLE_OFF|TCP_NAGLE_PUSH;
2087
			tcp_push_pending_frames(sk);
L
Linus Torvalds 已提交
2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110
		} else {
			tp->nonagle &= ~TCP_NAGLE_OFF;
		}
		break;

	case TCP_CORK:
		/* When set indicates to always queue non-full frames.
		 * Later the user clears this option and we transmit
		 * any pending partial frames in the queue.  This is
		 * meant to be used alongside sendfile() to get properly
		 * filled frames when the user (for example) must write
		 * out headers with a write() call first and then use
		 * sendfile to send out the data parts.
		 *
		 * TCP_CORK can be set together with TCP_NODELAY and it is
		 * stronger than TCP_NODELAY.
		 */
		if (val) {
			tp->nonagle |= TCP_NAGLE_CORK;
		} else {
			tp->nonagle &= ~TCP_NAGLE_CORK;
			if (tp->nonagle&TCP_NAGLE_OFF)
				tp->nonagle |= TCP_NAGLE_PUSH;
2111
			tcp_push_pending_frames(sk);
L
Linus Torvalds 已提交
2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127
		}
		break;

	case TCP_KEEPIDLE:
		if (val < 1 || val > MAX_TCP_KEEPIDLE)
			err = -EINVAL;
		else {
			tp->keepalive_time = val * HZ;
			if (sock_flag(sk, SOCK_KEEPOPEN) &&
			    !((1 << sk->sk_state) &
			      (TCPF_CLOSE | TCPF_LISTEN))) {
				__u32 elapsed = tcp_time_stamp - tp->rcv_tstamp;
				if (tp->keepalive_time > elapsed)
					elapsed = tp->keepalive_time - elapsed;
				else
					elapsed = 0;
2128
				inet_csk_reset_keepalive_timer(sk, elapsed);
L
Linus Torvalds 已提交
2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147
			}
		}
		break;
	case TCP_KEEPINTVL:
		if (val < 1 || val > MAX_TCP_KEEPINTVL)
			err = -EINVAL;
		else
			tp->keepalive_intvl = val * HZ;
		break;
	case TCP_KEEPCNT:
		if (val < 1 || val > MAX_TCP_KEEPCNT)
			err = -EINVAL;
		else
			tp->keepalive_probes = val;
		break;
	case TCP_SYNCNT:
		if (val < 1 || val > MAX_TCP_SYNCNT)
			err = -EINVAL;
		else
2148
			icsk->icsk_syn_retries = val;
L
Linus Torvalds 已提交
2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160
		break;

	case TCP_LINGER2:
		if (val < 0)
			tp->linger2 = -1;
		else if (val > sysctl_tcp_fin_timeout / HZ)
			tp->linger2 = 0;
		else
			tp->linger2 = val * HZ;
		break;

	case TCP_DEFER_ACCEPT:
2161 2162 2163 2164 2165 2166 2167 2168 2169
		icsk->icsk_accept_queue.rskq_defer_accept = 0;
		if (val > 0) {
			/* Translate value in seconds to number of
			 * retransmits */
			while (icsk->icsk_accept_queue.rskq_defer_accept < 32 &&
			       val > ((TCP_TIMEOUT_INIT / HZ) <<
				       icsk->icsk_accept_queue.rskq_defer_accept))
				icsk->icsk_accept_queue.rskq_defer_accept++;
			icsk->icsk_accept_queue.rskq_defer_accept++;
L
Linus Torvalds 已提交
2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186
		}
		break;

	case TCP_WINDOW_CLAMP:
		if (!val) {
			if (sk->sk_state != TCP_CLOSE) {
				err = -EINVAL;
				break;
			}
			tp->window_clamp = 0;
		} else
			tp->window_clamp = val < SOCK_MIN_RCVBUF / 2 ?
						SOCK_MIN_RCVBUF / 2 : val;
		break;

	case TCP_QUICKACK:
		if (!val) {
2187
			icsk->icsk_ack.pingpong = 1;
L
Linus Torvalds 已提交
2188
		} else {
2189
			icsk->icsk_ack.pingpong = 0;
L
Linus Torvalds 已提交
2190 2191
			if ((1 << sk->sk_state) &
			    (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT) &&
2192 2193
			    inet_csk_ack_scheduled(sk)) {
				icsk->icsk_ack.pending |= ICSK_ACK_PUSHED;
2194
				tcp_cleanup_rbuf(sk, 1);
L
Linus Torvalds 已提交
2195
				if (!(val & 1))
2196
					icsk->icsk_ack.pingpong = 1;
L
Linus Torvalds 已提交
2197 2198 2199 2200
			}
		}
		break;

2201 2202 2203 2204 2205 2206 2207
#ifdef CONFIG_TCP_MD5SIG
	case TCP_MD5SIG:
		/* Read the IP->Key mappings from userspace */
		err = tp->af_specific->md5_parse(sk, optval, optlen);
		break;
#endif

L
Linus Torvalds 已提交
2208 2209 2210
	default:
		err = -ENOPROTOOPT;
		break;
2211 2212
	}

L
Linus Torvalds 已提交
2213 2214 2215 2216
	release_sock(sk);
	return err;
}

2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228
int tcp_setsockopt(struct sock *sk, int level, int optname, char __user *optval,
		   int optlen)
{
	struct inet_connection_sock *icsk = inet_csk(sk);

	if (level != SOL_TCP)
		return icsk->icsk_af_ops->setsockopt(sk, level, optname,
						     optval, optlen);
	return do_tcp_setsockopt(sk, level, optname, optval, optlen);
}

#ifdef CONFIG_COMPAT
2229 2230
int compat_tcp_setsockopt(struct sock *sk, int level, int optname,
			  char __user *optval, int optlen)
2231
{
2232 2233 2234
	if (level != SOL_TCP)
		return inet_csk_compat_setsockopt(sk, level, optname,
						  optval, optlen);
2235 2236
	return do_tcp_setsockopt(sk, level, optname, optval, optlen);
}
2237 2238

EXPORT_SYMBOL(compat_tcp_setsockopt);
2239 2240
#endif

L
Linus Torvalds 已提交
2241 2242 2243 2244
/* Return information about state of tcp endpoint in API format. */
void tcp_get_info(struct sock *sk, struct tcp_info *info)
{
	struct tcp_sock *tp = tcp_sk(sk);
2245
	const struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2246 2247 2248 2249 2250
	u32 now = tcp_time_stamp;

	memset(info, 0, sizeof(*info));

	info->tcpi_state = sk->sk_state;
2251
	info->tcpi_ca_state = icsk->icsk_ca_state;
2252
	info->tcpi_retransmits = icsk->icsk_retransmits;
2253
	info->tcpi_probes = icsk->icsk_probes_out;
2254
	info->tcpi_backoff = icsk->icsk_backoff;
L
Linus Torvalds 已提交
2255 2256 2257

	if (tp->rx_opt.tstamp_ok)
		info->tcpi_options |= TCPI_OPT_TIMESTAMPS;
2258
	if (tcp_is_sack(tp))
L
Linus Torvalds 已提交
2259 2260 2261 2262 2263
		info->tcpi_options |= TCPI_OPT_SACK;
	if (tp->rx_opt.wscale_ok) {
		info->tcpi_options |= TCPI_OPT_WSCALE;
		info->tcpi_snd_wscale = tp->rx_opt.snd_wscale;
		info->tcpi_rcv_wscale = tp->rx_opt.rcv_wscale;
2264
	}
L
Linus Torvalds 已提交
2265 2266 2267 2268

	if (tp->ecn_flags&TCP_ECN_OK)
		info->tcpi_options |= TCPI_OPT_ECN;

2269 2270
	info->tcpi_rto = jiffies_to_usecs(icsk->icsk_rto);
	info->tcpi_ato = jiffies_to_usecs(icsk->icsk_ack.ato);
2271
	info->tcpi_snd_mss = tp->mss_cache;
2272
	info->tcpi_rcv_mss = icsk->icsk_ack.rcv_mss;
L
Linus Torvalds 已提交
2273

2274 2275 2276 2277 2278 2279 2280
	if (sk->sk_state == TCP_LISTEN) {
		info->tcpi_unacked = sk->sk_ack_backlog;
		info->tcpi_sacked = sk->sk_max_ack_backlog;
	} else {
		info->tcpi_unacked = tp->packets_out;
		info->tcpi_sacked = tp->sacked_out;
	}
L
Linus Torvalds 已提交
2281 2282 2283 2284 2285
	info->tcpi_lost = tp->lost_out;
	info->tcpi_retrans = tp->retrans_out;
	info->tcpi_fackets = tp->fackets_out;

	info->tcpi_last_data_sent = jiffies_to_msecs(now - tp->lsndtime);
2286
	info->tcpi_last_data_recv = jiffies_to_msecs(now - icsk->icsk_ack.lrcvtime);
L
Linus Torvalds 已提交
2287 2288
	info->tcpi_last_ack_recv = jiffies_to_msecs(now - tp->rcv_tstamp);

2289
	info->tcpi_pmtu = icsk->icsk_pmtu_cookie;
L
Linus Torvalds 已提交
2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305
	info->tcpi_rcv_ssthresh = tp->rcv_ssthresh;
	info->tcpi_rtt = jiffies_to_usecs(tp->srtt)>>3;
	info->tcpi_rttvar = jiffies_to_usecs(tp->mdev)>>2;
	info->tcpi_snd_ssthresh = tp->snd_ssthresh;
	info->tcpi_snd_cwnd = tp->snd_cwnd;
	info->tcpi_advmss = tp->advmss;
	info->tcpi_reordering = tp->reordering;

	info->tcpi_rcv_rtt = jiffies_to_usecs(tp->rcv_rtt_est.rtt)>>3;
	info->tcpi_rcv_space = tp->rcvq_space.space;

	info->tcpi_total_retrans = tp->total_retrans;
}

EXPORT_SYMBOL_GPL(tcp_get_info);

2306 2307
static int do_tcp_getsockopt(struct sock *sk, int level,
		int optname, char __user *optval, int __user *optlen)
L
Linus Torvalds 已提交
2308
{
2309
	struct inet_connection_sock *icsk = inet_csk(sk);
L
Linus Torvalds 已提交
2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322
	struct tcp_sock *tp = tcp_sk(sk);
	int val, len;

	if (get_user(len, optlen))
		return -EFAULT;

	len = min_t(unsigned int, len, sizeof(int));

	if (len < 0)
		return -EINVAL;

	switch (optname) {
	case TCP_MAXSEG:
2323
		val = tp->mss_cache;
L
Linus Torvalds 已提交
2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342
		if (!val && ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)))
			val = tp->rx_opt.user_mss;
		break;
	case TCP_NODELAY:
		val = !!(tp->nonagle&TCP_NAGLE_OFF);
		break;
	case TCP_CORK:
		val = !!(tp->nonagle&TCP_NAGLE_CORK);
		break;
	case TCP_KEEPIDLE:
		val = (tp->keepalive_time ? : sysctl_tcp_keepalive_time) / HZ;
		break;
	case TCP_KEEPINTVL:
		val = (tp->keepalive_intvl ? : sysctl_tcp_keepalive_intvl) / HZ;
		break;
	case TCP_KEEPCNT:
		val = tp->keepalive_probes ? : sysctl_tcp_keepalive_probes;
		break;
	case TCP_SYNCNT:
2343
		val = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries;
L
Linus Torvalds 已提交
2344 2345 2346 2347 2348 2349 2350
		break;
	case TCP_LINGER2:
		val = tp->linger2;
		if (val >= 0)
			val = (val ? : sysctl_tcp_fin_timeout) / HZ;
		break;
	case TCP_DEFER_ACCEPT:
2351 2352
		val = !icsk->icsk_accept_queue.rskq_defer_accept ? 0 :
			((TCP_TIMEOUT_INIT / HZ) << (icsk->icsk_accept_queue.rskq_defer_accept - 1));
L
Linus Torvalds 已提交
2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372
		break;
	case TCP_WINDOW_CLAMP:
		val = tp->window_clamp;
		break;
	case TCP_INFO: {
		struct tcp_info info;

		if (get_user(len, optlen))
			return -EFAULT;

		tcp_get_info(sk, &info);

		len = min_t(unsigned int, len, sizeof(info));
		if (put_user(len, optlen))
			return -EFAULT;
		if (copy_to_user(optval, &info, len))
			return -EFAULT;
		return 0;
	}
	case TCP_QUICKACK:
2373
		val = !icsk->icsk_ack.pingpong;
L
Linus Torvalds 已提交
2374
		break;
2375 2376 2377 2378 2379 2380 2381

	case TCP_CONGESTION:
		if (get_user(len, optlen))
			return -EFAULT;
		len = min_t(unsigned int, len, TCP_CA_NAME_MAX);
		if (put_user(len, optlen))
			return -EFAULT;
2382
		if (copy_to_user(optval, icsk->icsk_ca_ops->name, len))
2383 2384
			return -EFAULT;
		return 0;
L
Linus Torvalds 已提交
2385 2386
	default:
		return -ENOPROTOOPT;
2387
	}
L
Linus Torvalds 已提交
2388 2389 2390 2391 2392 2393 2394 2395

	if (put_user(len, optlen))
		return -EFAULT;
	if (copy_to_user(optval, &val, len))
		return -EFAULT;
	return 0;
}

2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407
int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval,
		   int __user *optlen)
{
	struct inet_connection_sock *icsk = inet_csk(sk);

	if (level != SOL_TCP)
		return icsk->icsk_af_ops->getsockopt(sk, level, optname,
						     optval, optlen);
	return do_tcp_getsockopt(sk, level, optname, optval, optlen);
}

#ifdef CONFIG_COMPAT
2408 2409
int compat_tcp_getsockopt(struct sock *sk, int level, int optname,
			  char __user *optval, int __user *optlen)
2410
{
2411 2412 2413
	if (level != SOL_TCP)
		return inet_csk_compat_getsockopt(sk, level, optname,
						  optval, optlen);
2414 2415
	return do_tcp_getsockopt(sk, level, optname, optval, optlen);
}
2416 2417

EXPORT_SYMBOL(compat_tcp_getsockopt);
2418
#endif
L
Linus Torvalds 已提交
2419

2420
struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features)
H
Herbert Xu 已提交
2421 2422 2423 2424 2425
{
	struct sk_buff *segs = ERR_PTR(-EINVAL);
	struct tcphdr *th;
	unsigned thlen;
	unsigned int seq;
2426
	__be32 delta;
H
Herbert Xu 已提交
2427
	unsigned int oldlen;
2428
	unsigned int mss;
H
Herbert Xu 已提交
2429 2430 2431 2432

	if (!pskb_may_pull(skb, sizeof(*th)))
		goto out;

2433
	th = tcp_hdr(skb);
H
Herbert Xu 已提交
2434 2435 2436 2437 2438 2439 2440
	thlen = th->doff * 4;
	if (thlen < sizeof(*th))
		goto out;

	if (!pskb_may_pull(skb, thlen))
		goto out;

2441
	oldlen = (u16)~skb->len;
H
Herbert Xu 已提交
2442 2443
	__skb_pull(skb, thlen);

2444 2445 2446 2447
	mss = skb_shinfo(skb)->gso_size;
	if (unlikely(skb->len <= mss))
		goto out;

2448 2449
	if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) {
		/* Packet is from an untrusted source, reset gso_segs. */
2450 2451 2452 2453 2454 2455 2456 2457 2458 2459
		int type = skb_shinfo(skb)->gso_type;

		if (unlikely(type &
			     ~(SKB_GSO_TCPV4 |
			       SKB_GSO_DODGY |
			       SKB_GSO_TCP_ECN |
			       SKB_GSO_TCPV6 |
			       0) ||
			     !(type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))))
			goto out;
2460

2461
		skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(skb->len, mss);
2462 2463 2464 2465 2466

		segs = NULL;
		goto out;
	}

2467
	segs = skb_segment(skb, features);
H
Herbert Xu 已提交
2468 2469 2470
	if (IS_ERR(segs))
		goto out;

2471
	delta = htonl(oldlen + (thlen + mss));
H
Herbert Xu 已提交
2472 2473

	skb = segs;
2474
	th = tcp_hdr(skb);
H
Herbert Xu 已提交
2475 2476 2477 2478 2479
	seq = ntohl(th->seq);

	do {
		th->fin = th->psh = 0;

2480 2481
		th->check = ~csum_fold((__force __wsum)((__force u32)th->check +
				       (__force u32)delta));
2482
		if (skb->ip_summed != CHECKSUM_PARTIAL)
2483 2484 2485
			th->check =
			     csum_fold(csum_partial(skb_transport_header(skb),
						    thlen, skb->csum));
H
Herbert Xu 已提交
2486

2487
		seq += mss;
H
Herbert Xu 已提交
2488
		skb = skb->next;
2489
		th = tcp_hdr(skb);
H
Herbert Xu 已提交
2490 2491 2492 2493 2494

		th->seq = htonl(seq);
		th->cwr = 0;
	} while (skb->next);

2495
	delta = htonl(oldlen + (skb->tail - skb->transport_header) +
2496
		      skb->data_len);
2497 2498
	th->check = ~csum_fold((__force __wsum)((__force u32)th->check +
				(__force u32)delta));
2499
	if (skb->ip_summed != CHECKSUM_PARTIAL)
2500 2501
		th->check = csum_fold(csum_partial(skb_transport_header(skb),
						   thlen, skb->csum));
H
Herbert Xu 已提交
2502 2503 2504 2505

out:
	return segs;
}
H
Herbert Xu 已提交
2506
EXPORT_SYMBOL(tcp_tso_segment);
H
Herbert Xu 已提交
2507

H
Herbert Xu 已提交
2508 2509 2510 2511 2512 2513
struct sk_buff **tcp_gro_receive(struct sk_buff **head, struct sk_buff *skb)
{
	struct sk_buff **pp = NULL;
	struct sk_buff *p;
	struct tcphdr *th;
	struct tcphdr *th2;
2514
	unsigned int len;
H
Herbert Xu 已提交
2515 2516 2517 2518
	unsigned int thlen;
	unsigned int flags;
	unsigned int mss = 1;
	int flush = 1;
H
Herbert Xu 已提交
2519
	int i;
H
Herbert Xu 已提交
2520

2521 2522
	th = skb_gro_header(skb, sizeof(*th));
	if (unlikely(!th))
H
Herbert Xu 已提交
2523 2524 2525 2526 2527 2528
		goto out;

	thlen = th->doff * 4;
	if (thlen < sizeof(*th))
		goto out;

2529 2530
	th = skb_gro_header(skb, thlen);
	if (unlikely(!th))
H
Herbert Xu 已提交
2531 2532
		goto out;

2533
	skb_gro_pull(skb, thlen);
H
Herbert Xu 已提交
2534

2535
	len = skb_gro_len(skb);
H
Herbert Xu 已提交
2536 2537 2538 2539 2540 2541 2542 2543
	flags = tcp_flag_word(th);

	for (; (p = *head); head = &p->next) {
		if (!NAPI_GRO_CB(p)->same_flow)
			continue;

		th2 = tcp_hdr(p);

H
Herbert Xu 已提交
2544
		if ((th->source ^ th2->source) | (th->dest ^ th2->dest)) {
H
Herbert Xu 已提交
2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558
			NAPI_GRO_CB(p)->same_flow = 0;
			continue;
		}

		goto found;
	}

	goto out_check_final;

found:
	flush = NAPI_GRO_CB(p)->flush;
	flush |= flags & TCP_FLAG_CWR;
	flush |= (flags ^ tcp_flag_word(th2)) &
		  ~(TCP_FLAG_CWR | TCP_FLAG_FIN | TCP_FLAG_PSH);
H
Herbert Xu 已提交
2559 2560 2561 2562
	flush |= (th->ack_seq ^ th2->ack_seq) | (th->window ^ th2->window);
	for (i = sizeof(*th); !flush && i < thlen; i += 4)
		flush |= *(u32 *)((u8 *)th + i) ^
			 *(u32 *)((u8 *)th2 + i);
H
Herbert Xu 已提交
2563

H
Herbert Xu 已提交
2564
	mss = skb_shinfo(p)->gso_size;
H
Herbert Xu 已提交
2565

2566
	flush |= (len > mss) | !len;
H
Herbert Xu 已提交
2567
	flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq);
H
Herbert Xu 已提交
2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578

	if (flush || skb_gro_receive(head, skb)) {
		mss = 1;
		goto out_check_final;
	}

	p = *head;
	th2 = tcp_hdr(p);
	tcp_flag_word(th2) |= flags & (TCP_FLAG_FIN | TCP_FLAG_PSH);

out_check_final:
2579
	flush = len < mss;
H
Herbert Xu 已提交
2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590
	flush |= flags & (TCP_FLAG_URG | TCP_FLAG_PSH | TCP_FLAG_RST |
			  TCP_FLAG_SYN | TCP_FLAG_FIN);

	if (p && (!NAPI_GRO_CB(skb)->same_flow || flush))
		pp = head;

out:
	NAPI_GRO_CB(skb)->flush |= flush;

	return pp;
}
H
Herbert Xu 已提交
2591
EXPORT_SYMBOL(tcp_gro_receive);
H
Herbert Xu 已提交
2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607

int tcp_gro_complete(struct sk_buff *skb)
{
	struct tcphdr *th = tcp_hdr(skb);

	skb->csum_start = skb_transport_header(skb) - skb->head;
	skb->csum_offset = offsetof(struct tcphdr, check);
	skb->ip_summed = CHECKSUM_PARTIAL;

	skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;

	if (th->cwr)
		skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;

	return 0;
}
H
Herbert Xu 已提交
2608
EXPORT_SYMBOL(tcp_gro_complete);
H
Herbert Xu 已提交
2609

2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633
#ifdef CONFIG_TCP_MD5SIG
static unsigned long tcp_md5sig_users;
static struct tcp_md5sig_pool **tcp_md5sig_pool;
static DEFINE_SPINLOCK(tcp_md5sig_pool_lock);

static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool)
{
	int cpu;
	for_each_possible_cpu(cpu) {
		struct tcp_md5sig_pool *p = *per_cpu_ptr(pool, cpu);
		if (p) {
			if (p->md5_desc.tfm)
				crypto_free_hash(p->md5_desc.tfm);
			kfree(p);
			p = NULL;
		}
	}
	free_percpu(pool);
}

void tcp_free_md5sig_pool(void)
{
	struct tcp_md5sig_pool **pool = NULL;

2634
	spin_lock_bh(&tcp_md5sig_pool_lock);
2635 2636 2637 2638
	if (--tcp_md5sig_users == 0) {
		pool = tcp_md5sig_pool;
		tcp_md5sig_pool = NULL;
	}
2639
	spin_unlock_bh(&tcp_md5sig_pool_lock);
2640 2641 2642 2643 2644 2645
	if (pool)
		__tcp_free_md5sig_pool(pool);
}

EXPORT_SYMBOL(tcp_free_md5sig_pool);

A
Adrian Bunk 已提交
2646
static struct tcp_md5sig_pool **__tcp_alloc_md5sig_pool(void)
2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681
{
	int cpu;
	struct tcp_md5sig_pool **pool;

	pool = alloc_percpu(struct tcp_md5sig_pool *);
	if (!pool)
		return NULL;

	for_each_possible_cpu(cpu) {
		struct tcp_md5sig_pool *p;
		struct crypto_hash *hash;

		p = kzalloc(sizeof(*p), GFP_KERNEL);
		if (!p)
			goto out_free;
		*per_cpu_ptr(pool, cpu) = p;

		hash = crypto_alloc_hash("md5", 0, CRYPTO_ALG_ASYNC);
		if (!hash || IS_ERR(hash))
			goto out_free;

		p->md5_desc.tfm = hash;
	}
	return pool;
out_free:
	__tcp_free_md5sig_pool(pool);
	return NULL;
}

struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(void)
{
	struct tcp_md5sig_pool **pool;
	int alloc = 0;

retry:
2682
	spin_lock_bh(&tcp_md5sig_pool_lock);
2683 2684 2685
	pool = tcp_md5sig_pool;
	if (tcp_md5sig_users++ == 0) {
		alloc = 1;
2686
		spin_unlock_bh(&tcp_md5sig_pool_lock);
2687 2688
	} else if (!pool) {
		tcp_md5sig_users--;
2689
		spin_unlock_bh(&tcp_md5sig_pool_lock);
2690 2691 2692
		cpu_relax();
		goto retry;
	} else
2693
		spin_unlock_bh(&tcp_md5sig_pool_lock);
2694 2695 2696 2697

	if (alloc) {
		/* we cannot hold spinlock here because this may sleep. */
		struct tcp_md5sig_pool **p = __tcp_alloc_md5sig_pool();
2698
		spin_lock_bh(&tcp_md5sig_pool_lock);
2699 2700
		if (!p) {
			tcp_md5sig_users--;
2701
			spin_unlock_bh(&tcp_md5sig_pool_lock);
2702 2703 2704 2705 2706
			return NULL;
		}
		pool = tcp_md5sig_pool;
		if (pool) {
			/* oops, it has already been assigned. */
2707
			spin_unlock_bh(&tcp_md5sig_pool_lock);
2708 2709 2710
			__tcp_free_md5sig_pool(p);
		} else {
			tcp_md5sig_pool = pool = p;
2711
			spin_unlock_bh(&tcp_md5sig_pool_lock);
2712 2713 2714 2715 2716 2717 2718 2719 2720 2721
		}
	}
	return pool;
}

EXPORT_SYMBOL(tcp_alloc_md5sig_pool);

struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu)
{
	struct tcp_md5sig_pool **p;
2722
	spin_lock_bh(&tcp_md5sig_pool_lock);
2723 2724 2725
	p = tcp_md5sig_pool;
	if (p)
		tcp_md5sig_users++;
2726
	spin_unlock_bh(&tcp_md5sig_pool_lock);
2727 2728 2729 2730 2731
	return (p ? *per_cpu_ptr(p, cpu) : NULL);
}

EXPORT_SYMBOL(__tcp_get_md5sig_pool);

2732 2733 2734
void __tcp_put_md5sig_pool(void)
{
	tcp_free_md5sig_pool();
2735 2736 2737
}

EXPORT_SYMBOL(__tcp_put_md5sig_pool);
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 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794

int tcp_md5_hash_header(struct tcp_md5sig_pool *hp,
			struct tcphdr *th)
{
	struct scatterlist sg;
	int err;

	__sum16 old_checksum = th->check;
	th->check = 0;
	/* options aren't included in the hash */
	sg_init_one(&sg, th, sizeof(struct tcphdr));
	err = crypto_hash_update(&hp->md5_desc, &sg, sizeof(struct tcphdr));
	th->check = old_checksum;
	return err;
}

EXPORT_SYMBOL(tcp_md5_hash_header);

int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *hp,
			  struct sk_buff *skb, unsigned header_len)
{
	struct scatterlist sg;
	const struct tcphdr *tp = tcp_hdr(skb);
	struct hash_desc *desc = &hp->md5_desc;
	unsigned i;
	const unsigned head_data_len = skb_headlen(skb) > header_len ?
				       skb_headlen(skb) - header_len : 0;
	const struct skb_shared_info *shi = skb_shinfo(skb);

	sg_init_table(&sg, 1);

	sg_set_buf(&sg, ((u8 *) tp) + header_len, head_data_len);
	if (crypto_hash_update(desc, &sg, head_data_len))
		return 1;

	for (i = 0; i < shi->nr_frags; ++i) {
		const struct skb_frag_struct *f = &shi->frags[i];
		sg_set_page(&sg, f->page, f->size, f->page_offset);
		if (crypto_hash_update(desc, &sg, f->size))
			return 1;
	}

	return 0;
}

EXPORT_SYMBOL(tcp_md5_hash_skb_data);

int tcp_md5_hash_key(struct tcp_md5sig_pool *hp, struct tcp_md5sig_key *key)
{
	struct scatterlist sg;

	sg_init_one(&sg, key->key, key->keylen);
	return crypto_hash_update(&hp->md5_desc, &sg, key->keylen);
}

EXPORT_SYMBOL(tcp_md5_hash_key);

2795 2796
#endif

A
Andi Kleen 已提交
2797 2798
void tcp_done(struct sock *sk)
{
2799
	if (sk->sk_state == TCP_SYN_SENT || sk->sk_state == TCP_SYN_RECV)
2800
		TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_ATTEMPTFAILS);
A
Andi Kleen 已提交
2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813

	tcp_set_state(sk, TCP_CLOSE);
	tcp_clear_xmit_timers(sk);

	sk->sk_shutdown = SHUTDOWN_MASK;

	if (!sock_flag(sk, SOCK_DEAD))
		sk->sk_state_change(sk);
	else
		inet_csk_destroy_sock(sk);
}
EXPORT_SYMBOL_GPL(tcp_done);

2814
extern struct tcp_congestion_ops tcp_reno;
L
Linus Torvalds 已提交
2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828

static __initdata unsigned long thash_entries;
static int __init set_thash_entries(char *str)
{
	if (!str)
		return 0;
	thash_entries = simple_strtoul(str, &str, 0);
	return 1;
}
__setup("thash_entries=", set_thash_entries);

void __init tcp_init(void)
{
	struct sk_buff *skb = NULL;
2829
	unsigned long nr_pages, limit;
2830
	int order, i, max_share;
L
Linus Torvalds 已提交
2831

2832
	BUILD_BUG_ON(sizeof(struct tcp_skb_cb) > sizeof(skb->cb));
L
Linus Torvalds 已提交
2833

2834
	percpu_counter_init(&tcp_sockets_allocated, 0);
2835
	percpu_counter_init(&tcp_orphan_count, 0);
2836 2837 2838
	tcp_hashinfo.bind_bucket_cachep =
		kmem_cache_create("tcp_bind_bucket",
				  sizeof(struct inet_bind_bucket), 0,
2839
				  SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
L
Linus Torvalds 已提交
2840 2841 2842 2843 2844 2845

	/* Size and allocate the main established and bind bucket
	 * hash tables.
	 *
	 * The methodology is similar to that of the buffer cache.
	 */
2846
	tcp_hashinfo.ehash =
L
Linus Torvalds 已提交
2847
		alloc_large_system_hash("TCP established",
2848
					sizeof(struct inet_ehash_bucket),
L
Linus Torvalds 已提交
2849 2850
					thash_entries,
					(num_physpages >= 128 * 1024) ?
2851
					13 : 15,
2852
					0,
2853
					&tcp_hashinfo.ehash_size,
L
Linus Torvalds 已提交
2854
					NULL,
2855
					thash_entries ? 0 : 512 * 1024);
2856 2857
	tcp_hashinfo.ehash_size = 1 << tcp_hashinfo.ehash_size;
	for (i = 0; i < tcp_hashinfo.ehash_size; i++) {
2858 2859
		INIT_HLIST_NULLS_HEAD(&tcp_hashinfo.ehash[i].chain, i);
		INIT_HLIST_NULLS_HEAD(&tcp_hashinfo.ehash[i].twchain, i);
L
Linus Torvalds 已提交
2860
	}
2861 2862
	if (inet_ehash_locks_alloc(&tcp_hashinfo))
		panic("TCP: failed to alloc ehash_locks");
2863
	tcp_hashinfo.bhash =
L
Linus Torvalds 已提交
2864
		alloc_large_system_hash("TCP bind",
2865
					sizeof(struct inet_bind_hashbucket),
2866
					tcp_hashinfo.ehash_size,
L
Linus Torvalds 已提交
2867
					(num_physpages >= 128 * 1024) ?
2868
					13 : 15,
2869
					0,
2870
					&tcp_hashinfo.bhash_size,
L
Linus Torvalds 已提交
2871 2872
					NULL,
					64 * 1024);
2873 2874 2875 2876
	tcp_hashinfo.bhash_size = 1 << tcp_hashinfo.bhash_size;
	for (i = 0; i < tcp_hashinfo.bhash_size; i++) {
		spin_lock_init(&tcp_hashinfo.bhash[i].lock);
		INIT_HLIST_HEAD(&tcp_hashinfo.bhash[i].chain);
L
Linus Torvalds 已提交
2877 2878 2879 2880 2881 2882
	}

	/* Try to be a bit smarter and adjust defaults depending
	 * on available memory.
	 */
	for (order = 0; ((1 << order) << PAGE_SHIFT) <
2883
			(tcp_hashinfo.bhash_size * sizeof(struct inet_bind_hashbucket));
L
Linus Torvalds 已提交
2884 2885
			order++)
		;
2886
	if (order >= 4) {
2887
		tcp_death_row.sysctl_max_tw_buckets = 180000;
L
Linus Torvalds 已提交
2888 2889 2890
		sysctl_tcp_max_orphans = 4096 << (order - 4);
		sysctl_max_syn_backlog = 1024;
	} else if (order < 3) {
2891
		tcp_death_row.sysctl_max_tw_buckets >>= (3 - order);
L
Linus Torvalds 已提交
2892 2893 2894 2895
		sysctl_tcp_max_orphans >>= (3 - order);
		sysctl_max_syn_backlog = 128;
	}

2896 2897 2898 2899
	/* Set the pressure threshold to be a fraction of global memory that
	 * is up to 1/2 at 256 MB, decreasing toward zero with the amount of
	 * memory, with a floor of 128 pages.
	 */
2900 2901 2902
	nr_pages = totalram_pages - totalhigh_pages;
	limit = min(nr_pages, 1UL<<(28-PAGE_SHIFT)) >> (20-PAGE_SHIFT);
	limit = (limit * (nr_pages >> (20-PAGE_SHIFT))) >> (PAGE_SHIFT-11);
2903 2904 2905
	limit = max(limit, 128UL);
	sysctl_tcp_mem[0] = limit / 4 * 3;
	sysctl_tcp_mem[1] = limit;
2906
	sysctl_tcp_mem[2] = sysctl_tcp_mem[0] * 2;
L
Linus Torvalds 已提交
2907

2908
	/* Set per-socket limits to no more than 1/128 the pressure threshold */
2909 2910 2911
	limit = ((unsigned long)sysctl_tcp_mem[1]) << (PAGE_SHIFT - 7);
	max_share = min(4UL*1024*1024, limit);

2912
	sysctl_tcp_wmem[0] = SK_MEM_QUANTUM;
2913 2914 2915
	sysctl_tcp_wmem[1] = 16*1024;
	sysctl_tcp_wmem[2] = max(64*1024, max_share);

2916
	sysctl_tcp_rmem[0] = SK_MEM_QUANTUM;
2917 2918
	sysctl_tcp_rmem[1] = 87380;
	sysctl_tcp_rmem[2] = max(87380, max_share);
L
Linus Torvalds 已提交
2919 2920 2921

	printk(KERN_INFO "TCP: Hash tables configured "
	       "(established %d bind %d)\n",
2922
	       tcp_hashinfo.ehash_size, tcp_hashinfo.bhash_size);
2923 2924

	tcp_register_congestion_control(&tcp_reno);
L
Linus Torvalds 已提交
2925 2926 2927 2928 2929 2930 2931 2932 2933 2934
}

EXPORT_SYMBOL(tcp_close);
EXPORT_SYMBOL(tcp_disconnect);
EXPORT_SYMBOL(tcp_getsockopt);
EXPORT_SYMBOL(tcp_ioctl);
EXPORT_SYMBOL(tcp_poll);
EXPORT_SYMBOL(tcp_read_sock);
EXPORT_SYMBOL(tcp_recvmsg);
EXPORT_SYMBOL(tcp_sendmsg);
J
Jens Axboe 已提交
2935
EXPORT_SYMBOL(tcp_splice_read);
L
Linus Torvalds 已提交
2936 2937 2938
EXPORT_SYMBOL(tcp_sendpage);
EXPORT_SYMBOL(tcp_setsockopt);
EXPORT_SYMBOL(tcp_shutdown);