af_inet.c 42.7 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.
 *
 *		PF_INET protocol family socket handler.
 *
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
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Florian La Roche, <flla@stud.uni-sb.de>
 *		Alan Cox, <A.Cox@swansea.ac.uk>
 *
 * Changes (see also sock.c)
 *
 *		piggy,
 *		Karl Knutson	:	Socket protocol table
 *		A.N.Kuznetsov	:	Socket death error in accept().
 *		John Richardson :	Fix non blocking error in connect()
 *					so sockets that fail to connect
 *					don't return -EINPROGRESS.
 *		Alan Cox	:	Asynchronous I/O support
 *		Alan Cox	:	Keep correct socket pointer on sock
 *					structures
 *					when accept() ed
 *		Alan Cox	:	Semantics of SO_LINGER aren't state
 *					moved to close when you look carefully.
 *					With this fixed and the accept bug fixed
 *					some RPC stuff seems happier.
 *		Niibe Yutaka	:	4.4BSD style write async I/O
 *		Alan Cox,
 *		Tony Gale 	:	Fixed reuse semantics.
 *		Alan Cox	:	bind() shouldn't abort existing but dead
 *					sockets. Stops FTP netin:.. I hope.
 *		Alan Cox	:	bind() works correctly for RAW sockets.
 *					Note that FreeBSD at least was broken
 *					in this respect so be careful with
 *					compatibility tests...
 *		Alan Cox	:	routing cache support
 *		Alan Cox	:	memzero the socket structure for
 *					compactness.
 *		Matt Day	:	nonblock connect error handler
 *		Alan Cox	:	Allow large numbers of pending sockets
 *					(eg for big web sites), but only if
 *					specifically application requested.
 *		Alan Cox	:	New buffering throughout IP. Used
 *					dumbly.
 *		Alan Cox	:	New buffering now used smartly.
 *		Alan Cox	:	BSD rather than common sense
 *					interpretation of listen.
 *		Germano Caronni	:	Assorted small races.
 *		Alan Cox	:	sendmsg/recvmsg basic support.
 *		Alan Cox	:	Only sendmsg/recvmsg now supported.
 *		Alan Cox	:	Locked down bind (see security list).
 *		Alan Cox	:	Loosened bind a little.
 *		Mike McLagan	:	ADD/DEL DLCI Ioctls
 *	Willy Konynenberg	:	Transparent proxying support.
 *		David S. Miller	:	New socket lookup architecture.
 *					Some other random speedups.
 *		Cyrus Durgin	:	Cleaned up file for kmod hacks.
 *		Andi Kleen	:	Fix inet_stream_connect TCP race.
 *
 *		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.
 */

H
Herbert Xu 已提交
68
#include <linux/err.h>
L
Linus Torvalds 已提交
69 70 71 72 73 74 75 76 77 78 79
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/in.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/timer.h>
#include <linux/string.h>
#include <linux/sockios.h>
#include <linux/net.h>
80
#include <linux/capability.h>
L
Linus Torvalds 已提交
81 82 83 84 85 86 87
#include <linux/fcntl.h>
#include <linux/mm.h>
#include <linux/interrupt.h>
#include <linux/stat.h>
#include <linux/init.h>
#include <linux/poll.h>
#include <linux/netfilter_ipv4.h>
88
#include <linux/random.h>
89
#include <linux/slab.h>
L
Linus Torvalds 已提交
90 91 92 93 94 95

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

#include <linux/inet.h>
#include <linux/igmp.h>
96
#include <linux/inetdevice.h>
L
Linus Torvalds 已提交
97
#include <linux/netdevice.h>
H
Herbert Xu 已提交
98
#include <net/checksum.h>
L
Linus Torvalds 已提交
99 100 101 102 103
#include <net/ip.h>
#include <net/protocol.h>
#include <net/arp.h>
#include <net/route.h>
#include <net/ip_fib.h>
104
#include <net/inet_connection_sock.h>
L
Linus Torvalds 已提交
105 106
#include <net/tcp.h>
#include <net/udp.h>
107
#include <net/udplite.h>
108
#include <net/ping.h>
L
Linus Torvalds 已提交
109 110 111 112 113 114 115
#include <linux/skbuff.h>
#include <net/sock.h>
#include <net/raw.h>
#include <net/icmp.h>
#include <net/ipip.h>
#include <net/inet_common.h>
#include <net/xfrm.h>
116
#include <net/net_namespace.h>
L
Linus Torvalds 已提交
117 118 119 120 121 122 123 124 125 126 127
#ifdef CONFIG_IP_MROUTE
#include <linux/mroute.h>
#endif


/* The inetsw table contains everything that inet_create needs to
 * build a new socket.
 */
static struct list_head inetsw[SOCK_MAX];
static DEFINE_SPINLOCK(inetsw_lock);

128 129 130
struct ipv4_config ipv4_config;
EXPORT_SYMBOL(ipv4_config);

L
Linus Torvalds 已提交
131 132 133 134 135 136 137 138 139
/* New destruction routine */

void inet_sock_destruct(struct sock *sk)
{
	struct inet_sock *inet = inet_sk(sk);

	__skb_queue_purge(&sk->sk_receive_queue);
	__skb_queue_purge(&sk->sk_error_queue);

H
Hideo Aoki 已提交
140 141
	sk_mem_reclaim(sk);

L
Linus Torvalds 已提交
142
	if (sk->sk_type == SOCK_STREAM && sk->sk_state != TCP_CLOSE) {
E
Eric Dumazet 已提交
143
		pr_err("Attempt to release TCP socket in state %d %p\n",
L
Linus Torvalds 已提交
144 145 146 147
		       sk->sk_state, sk);
		return;
	}
	if (!sock_flag(sk, SOCK_DEAD)) {
E
Eric Dumazet 已提交
148
		pr_err("Attempt to release alive inet socket %p\n", sk);
L
Linus Torvalds 已提交
149 150 151
		return;
	}

152 153 154 155
	WARN_ON(atomic_read(&sk->sk_rmem_alloc));
	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
	WARN_ON(sk->sk_wmem_queued);
	WARN_ON(sk->sk_forward_alloc);
L
Linus Torvalds 已提交
156

157
	kfree(rcu_dereference_protected(inet->inet_opt, 1));
E
Eric Dumazet 已提交
158
	dst_release(rcu_dereference_check(sk->sk_dst_cache, 1));
159
	sk_refcnt_debug_dec(sk);
L
Linus Torvalds 已提交
160
}
E
Eric Dumazet 已提交
161
EXPORT_SYMBOL(inet_sock_destruct);
L
Linus Torvalds 已提交
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178

/*
 *	The routines beyond this point handle the behaviour of an AF_INET
 *	socket object. Mostly it punts to the subprotocols of IP to do
 *	the work.
 */

/*
 *	Automatically bind an unbound socket.
 */

static int inet_autobind(struct sock *sk)
{
	struct inet_sock *inet;
	/* We may need to bind the socket. */
	lock_sock(sk);
	inet = inet_sk(sk);
E
Eric Dumazet 已提交
179
	if (!inet->inet_num) {
L
Linus Torvalds 已提交
180 181 182 183
		if (sk->sk_prot->get_port(sk, 0)) {
			release_sock(sk);
			return -EAGAIN;
		}
E
Eric Dumazet 已提交
184
		inet->inet_sport = htons(inet->inet_num);
L
Linus Torvalds 已提交
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
	}
	release_sock(sk);
	return 0;
}

/*
 *	Move a socket into listening state.
 */
int inet_listen(struct socket *sock, int backlog)
{
	struct sock *sk = sock->sk;
	unsigned char old_state;
	int err;

	lock_sock(sk);

	err = -EINVAL;
	if (sock->state != SS_UNCONNECTED || sock->type != SOCK_STREAM)
		goto out;

	old_state = sk->sk_state;
	if (!((1 << old_state) & (TCPF_CLOSE | TCPF_LISTEN)))
		goto out;

	/* Really, if the socket is already in listen state
	 * we can only allow the backlog to be adjusted.
	 */
	if (old_state != TCP_LISTEN) {
213
		err = inet_csk_listen_start(sk, backlog);
L
Linus Torvalds 已提交
214 215 216 217 218 219 220 221 222 223
		if (err)
			goto out;
	}
	sk->sk_max_ack_backlog = backlog;
	err = 0;

out:
	release_sock(sk);
	return err;
}
E
Eric Dumazet 已提交
224
EXPORT_SYMBOL(inet_listen);
L
Linus Torvalds 已提交
225

226
u32 inet_ehash_secret __read_mostly;
227 228
EXPORT_SYMBOL(inet_ehash_secret);

229 230 231
/*
 * inet_ehash_secret must be set exactly once
 */
232 233
void build_ehash_secret(void)
{
234
	u32 rnd;
235

236 237 238
	do {
		get_random_bytes(&rnd, sizeof(rnd));
	} while (rnd == 0);
239 240

	cmpxchg(&inet_ehash_secret, 0, rnd);
241 242 243
}
EXPORT_SYMBOL(build_ehash_secret);

244 245 246
static inline int inet_netns_ok(struct net *net, int protocol)
{
	int hash;
247
	const struct net_protocol *ipprot;
248

E
Eric Dumazet 已提交
249
	if (net_eq(net, &init_net))
250 251 252 253 254 255 256 257 258 259 260
		return 1;

	hash = protocol & (MAX_INET_PROTOS - 1);
	ipprot = rcu_dereference(inet_protos[hash]);

	if (ipprot == NULL)
		/* raw IP is OK */
		return 1;
	return ipprot->netns_ok;
}

L
Linus Torvalds 已提交
261 262 263 264
/*
 *	Create an inet socket.
 */

265 266
static int inet_create(struct net *net, struct socket *sock, int protocol,
		       int kern)
L
Linus Torvalds 已提交
267 268 269 270 271 272 273
{
	struct sock *sk;
	struct inet_protosw *answer;
	struct inet_sock *inet;
	struct proto *answer_prot;
	unsigned char answer_flags;
	char answer_no_check;
274
	int try_loading_module = 0;
275
	int err;
L
Linus Torvalds 已提交
276

E
Eric Dumazet 已提交
277 278 279
	if (unlikely(!inet_ehash_secret))
		if (sock->type != SOCK_RAW && sock->type != SOCK_DGRAM)
			build_ehash_secret();
280

L
Linus Torvalds 已提交
281 282 283
	sock->state = SS_UNCONNECTED;

	/* Look for the requested type/protocol pair. */
284
lookup_protocol:
285
	err = -ESOCKTNOSUPPORT;
L
Linus Torvalds 已提交
286
	rcu_read_lock();
287
	list_for_each_entry_rcu(answer, &inetsw[sock->type], list) {
L
Linus Torvalds 已提交
288

289
		err = 0;
L
Linus Torvalds 已提交
290 291 292 293 294 295 296 297 298 299 300 301 302
		/* Check the non-wild match. */
		if (protocol == answer->protocol) {
			if (protocol != IPPROTO_IP)
				break;
		} else {
			/* Check for the two wild cases. */
			if (IPPROTO_IP == protocol) {
				protocol = answer->protocol;
				break;
			}
			if (IPPROTO_IP == answer->protocol)
				break;
		}
303
		err = -EPROTONOSUPPORT;
L
Linus Torvalds 已提交
304 305
	}

306
	if (unlikely(err)) {
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
		if (try_loading_module < 2) {
			rcu_read_unlock();
			/*
			 * Be more specific, e.g. net-pf-2-proto-132-type-1
			 * (net-pf-PF_INET-proto-IPPROTO_SCTP-type-SOCK_STREAM)
			 */
			if (++try_loading_module == 1)
				request_module("net-pf-%d-proto-%d-type-%d",
					       PF_INET, protocol, sock->type);
			/*
			 * Fall back to generic, e.g. net-pf-2-proto-132
			 * (net-pf-PF_INET-proto-IPPROTO_SCTP)
			 */
			else
				request_module("net-pf-%d-proto-%d",
					       PF_INET, protocol);
			goto lookup_protocol;
		} else
			goto out_rcu_unlock;
	}

L
Linus Torvalds 已提交
328
	err = -EPERM;
329
	if (sock->type == SOCK_RAW && !kern && !capable(CAP_NET_RAW))
L
Linus Torvalds 已提交
330 331
		goto out_rcu_unlock;

332 333 334 335
	err = -EAFNOSUPPORT;
	if (!inet_netns_ok(net, protocol))
		goto out_rcu_unlock;

L
Linus Torvalds 已提交
336 337 338 339 340 341
	sock->ops = answer->ops;
	answer_prot = answer->prot;
	answer_no_check = answer->no_check;
	answer_flags = answer->flags;
	rcu_read_unlock();

342
	WARN_ON(answer_prot->slab == NULL);
L
Linus Torvalds 已提交
343 344

	err = -ENOBUFS;
345
	sk = sk_alloc(net, PF_INET, GFP_KERNEL, answer_prot);
L
Linus Torvalds 已提交
346 347 348 349 350 351 352 353 354
	if (sk == NULL)
		goto out;

	err = 0;
	sk->sk_no_check = answer_no_check;
	if (INET_PROTOSW_REUSE & answer_flags)
		sk->sk_reuse = 1;

	inet = inet_sk(sk);
355
	inet->is_icsk = (INET_PROTOSW_ICSK & answer_flags) != 0;
L
Linus Torvalds 已提交
356

357 358
	inet->nodefrag = 0;

L
Linus Torvalds 已提交
359
	if (SOCK_RAW == sock->type) {
E
Eric Dumazet 已提交
360
		inet->inet_num = protocol;
L
Linus Torvalds 已提交
361 362 363 364 365 366 367 368 369
		if (IPPROTO_RAW == protocol)
			inet->hdrincl = 1;
	}

	if (ipv4_config.no_pmtu_disc)
		inet->pmtudisc = IP_PMTUDISC_DONT;
	else
		inet->pmtudisc = IP_PMTUDISC_WANT;

E
Eric Dumazet 已提交
370
	inet->inet_id = 0;
L
Linus Torvalds 已提交
371 372 373 374 375 376 377 378 379 380

	sock_init_data(sock, sk);

	sk->sk_destruct	   = inet_sock_destruct;
	sk->sk_protocol	   = protocol;
	sk->sk_backlog_rcv = sk->sk_prot->backlog_rcv;

	inet->uc_ttl	= -1;
	inet->mc_loop	= 1;
	inet->mc_ttl	= 1;
381
	inet->mc_all	= 1;
L
Linus Torvalds 已提交
382 383 384
	inet->mc_index	= 0;
	inet->mc_list	= NULL;

385
	sk_refcnt_debug_inc(sk);
L
Linus Torvalds 已提交
386

E
Eric Dumazet 已提交
387
	if (inet->inet_num) {
L
Linus Torvalds 已提交
388 389 390 391 392
		/* It assumes that any protocol which allows
		 * the user to assign a number at socket
		 * creation time automatically
		 * shares.
		 */
E
Eric Dumazet 已提交
393
		inet->inet_sport = htons(inet->inet_num);
L
Linus Torvalds 已提交
394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
		/* Add to protocol hash chains. */
		sk->sk_prot->hash(sk);
	}

	if (sk->sk_prot->init) {
		err = sk->sk_prot->init(sk);
		if (err)
			sk_common_release(sk);
	}
out:
	return err;
out_rcu_unlock:
	rcu_read_unlock();
	goto out;
}


/*
 *	The peer socket should always be NULL (or else). When we call this
 *	function we are destroying the object and from then on nobody
 *	should refer to it.
 */
int inet_release(struct socket *sock)
{
	struct sock *sk = sock->sk;

	if (sk) {
		long timeout;

423
		sock_rps_reset_flow(sk);
T
Tom Herbert 已提交
424

L
Linus Torvalds 已提交
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
		/* Applications forget to leave groups before exiting */
		ip_mc_drop_socket(sk);

		/* If linger is set, we don't return until the close
		 * is complete.  Otherwise we return immediately. The
		 * actually closing is done the same either way.
		 *
		 * If the close is due to the process exiting, we never
		 * linger..
		 */
		timeout = 0;
		if (sock_flag(sk, SOCK_LINGER) &&
		    !(current->flags & PF_EXITING))
			timeout = sk->sk_lingertime;
		sock->sk = NULL;
		sk->sk_prot->close(sk, timeout);
	}
	return 0;
}
E
Eric Dumazet 已提交
444
EXPORT_SYMBOL(inet_release);
L
Linus Torvalds 已提交
445 446

/* It is off by default, see below. */
447
int sysctl_ip_nonlocal_bind __read_mostly;
E
Eric Dumazet 已提交
448
EXPORT_SYMBOL(sysctl_ip_nonlocal_bind);
L
Linus Torvalds 已提交
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467

int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
{
	struct sockaddr_in *addr = (struct sockaddr_in *)uaddr;
	struct sock *sk = sock->sk;
	struct inet_sock *inet = inet_sk(sk);
	unsigned short snum;
	int chk_addr_ret;
	int err;

	/* If the socket has its own bind function then use it. (RAW) */
	if (sk->sk_prot->bind) {
		err = sk->sk_prot->bind(sk, uaddr, addr_len);
		goto out;
	}
	err = -EINVAL;
	if (addr_len < sizeof(struct sockaddr_in))
		goto out;

468
	if (addr->sin_family != AF_INET) {
469 470 471
		/* Compatibility games : accept AF_UNSPEC (mapped to AF_INET)
		 * only if s_addr is INADDR_ANY.
		 */
472
		err = -EAFNOSUPPORT;
473 474 475
		if (addr->sin_family != AF_UNSPEC ||
		    addr->sin_addr.s_addr != htonl(INADDR_ANY))
			goto out;
476
	}
477

478
	chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr);
L
Linus Torvalds 已提交
479 480 481 482 483 484 485 486 487 488

	/* Not specified by any standard per-se, however it breaks too
	 * many applications when removed.  It is unfortunate since
	 * allowing applications to make a non-local bind solves
	 * several problems with systems using dynamic addressing.
	 * (ie. your servers still start up even if your ISDN link
	 *  is temporarily down)
	 */
	err = -EADDRNOTAVAIL;
	if (!sysctl_ip_nonlocal_bind &&
489
	    !(inet->freebind || inet->transparent) &&
A
Al Viro 已提交
490
	    addr->sin_addr.s_addr != htonl(INADDR_ANY) &&
L
Linus Torvalds 已提交
491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511
	    chk_addr_ret != RTN_LOCAL &&
	    chk_addr_ret != RTN_MULTICAST &&
	    chk_addr_ret != RTN_BROADCAST)
		goto out;

	snum = ntohs(addr->sin_port);
	err = -EACCES;
	if (snum && snum < PROT_SOCK && !capable(CAP_NET_BIND_SERVICE))
		goto out;

	/*      We keep a pair of addresses. rcv_saddr is the one
	 *      used by hash lookups, and saddr is used for transmit.
	 *
	 *      In the BSD API these are the same except where it
	 *      would be illegal to use them (multicast/broadcast) in
	 *      which case the sending device address is used.
	 */
	lock_sock(sk);

	/* Check these errors (active socket, double bind). */
	err = -EINVAL;
E
Eric Dumazet 已提交
512
	if (sk->sk_state != TCP_CLOSE || inet->inet_num)
L
Linus Torvalds 已提交
513 514
		goto out_release_sock;

E
Eric Dumazet 已提交
515
	inet->inet_rcv_saddr = inet->inet_saddr = addr->sin_addr.s_addr;
L
Linus Torvalds 已提交
516
	if (chk_addr_ret == RTN_MULTICAST || chk_addr_ret == RTN_BROADCAST)
E
Eric Dumazet 已提交
517
		inet->inet_saddr = 0;  /* Use device */
L
Linus Torvalds 已提交
518 519 520

	/* Make sure we are allowed to bind here. */
	if (sk->sk_prot->get_port(sk, snum)) {
E
Eric Dumazet 已提交
521
		inet->inet_saddr = inet->inet_rcv_saddr = 0;
L
Linus Torvalds 已提交
522 523 524 525
		err = -EADDRINUSE;
		goto out_release_sock;
	}

E
Eric Dumazet 已提交
526
	if (inet->inet_rcv_saddr)
L
Linus Torvalds 已提交
527 528 529
		sk->sk_userlocks |= SOCK_BINDADDR_LOCK;
	if (snum)
		sk->sk_userlocks |= SOCK_BINDPORT_LOCK;
E
Eric Dumazet 已提交
530 531 532
	inet->inet_sport = htons(inet->inet_num);
	inet->inet_daddr = 0;
	inet->inet_dport = 0;
L
Linus Torvalds 已提交
533 534 535 536 537 538 539
	sk_dst_reset(sk);
	err = 0;
out_release_sock:
	release_sock(sk);
out:
	return err;
}
E
Eric Dumazet 已提交
540
EXPORT_SYMBOL(inet_bind);
L
Linus Torvalds 已提交
541 542 543 544 545 546

int inet_dgram_connect(struct socket *sock, struct sockaddr * uaddr,
		       int addr_len, int flags)
{
	struct sock *sk = sock->sk;

547 548
	if (addr_len < sizeof(uaddr->sa_family))
		return -EINVAL;
L
Linus Torvalds 已提交
549 550 551
	if (uaddr->sa_family == AF_UNSPEC)
		return sk->sk_prot->disconnect(sk, flags);

E
Eric Dumazet 已提交
552
	if (!inet_sk(sk)->inet_num && inet_autobind(sk))
L
Linus Torvalds 已提交
553 554 555
		return -EAGAIN;
	return sk->sk_prot->connect(sk, (struct sockaddr *)uaddr, addr_len);
}
E
Eric Dumazet 已提交
556
EXPORT_SYMBOL(inet_dgram_connect);
L
Linus Torvalds 已提交
557 558 559 560 561

static long inet_wait_for_connect(struct sock *sk, long timeo)
{
	DEFINE_WAIT(wait);

E
Eric Dumazet 已提交
562
	prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
L
Linus Torvalds 已提交
563 564 565 566 567 568 569 570 571 572 573 574

	/* Basic assumption: if someone sets sk->sk_err, he _must_
	 * change state of the socket from TCP_SYN_*.
	 * Connect() does not allow to get error notifications
	 * without closing the socket.
	 */
	while ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) {
		release_sock(sk);
		timeo = schedule_timeout(timeo);
		lock_sock(sk);
		if (signal_pending(current) || !timeo)
			break;
E
Eric Dumazet 已提交
575
		prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
L
Linus Torvalds 已提交
576
	}
E
Eric Dumazet 已提交
577
	finish_wait(sk_sleep(sk), &wait);
L
Linus Torvalds 已提交
578 579 580 581 582 583 584 585 586 587 588 589 590 591
	return timeo;
}

/*
 *	Connect to a remote host. There is regrettably still a little
 *	TCP 'magic' in here.
 */
int inet_stream_connect(struct socket *sock, struct sockaddr *uaddr,
			int addr_len, int flags)
{
	struct sock *sk = sock->sk;
	int err;
	long timeo;

592 593 594
	if (addr_len < sizeof(uaddr->sa_family))
		return -EINVAL;

L
Linus Torvalds 已提交
595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622
	lock_sock(sk);

	if (uaddr->sa_family == AF_UNSPEC) {
		err = sk->sk_prot->disconnect(sk, flags);
		sock->state = err ? SS_DISCONNECTING : SS_UNCONNECTED;
		goto out;
	}

	switch (sock->state) {
	default:
		err = -EINVAL;
		goto out;
	case SS_CONNECTED:
		err = -EISCONN;
		goto out;
	case SS_CONNECTING:
		err = -EALREADY;
		/* Fall out of switch with err, set for this state */
		break;
	case SS_UNCONNECTED:
		err = -EISCONN;
		if (sk->sk_state != TCP_CLOSE)
			goto out;

		err = sk->sk_prot->connect(sk, uaddr, addr_len);
		if (err < 0)
			goto out;

623
		sock->state = SS_CONNECTING;
L
Linus Torvalds 已提交
624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668

		/* Just entered SS_CONNECTING state; the only
		 * difference is that return value in non-blocking
		 * case is EINPROGRESS, rather than EALREADY.
		 */
		err = -EINPROGRESS;
		break;
	}

	timeo = sock_sndtimeo(sk, flags & O_NONBLOCK);

	if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) {
		/* Error code is set above */
		if (!timeo || !inet_wait_for_connect(sk, timeo))
			goto out;

		err = sock_intr_errno(timeo);
		if (signal_pending(current))
			goto out;
	}

	/* Connection was closed by RST, timeout, ICMP error
	 * or another process disconnected us.
	 */
	if (sk->sk_state == TCP_CLOSE)
		goto sock_error;

	/* sk->sk_err may be not zero now, if RECVERR was ordered by user
	 * and error was received after socket entered established state.
	 * Hence, it is handled normally after connect() return successfully.
	 */

	sock->state = SS_CONNECTED;
	err = 0;
out:
	release_sock(sk);
	return err;

sock_error:
	err = sock_error(sk) ? : -ECONNABORTED;
	sock->state = SS_UNCONNECTED;
	if (sk->sk_prot->disconnect(sk, flags))
		sock->state = SS_DISCONNECTING;
	goto out;
}
E
Eric Dumazet 已提交
669
EXPORT_SYMBOL(inet_stream_connect);
L
Linus Torvalds 已提交
670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685

/*
 *	Accept a pending connection. The TCP layer now gives BSD semantics.
 */

int inet_accept(struct socket *sock, struct socket *newsock, int flags)
{
	struct sock *sk1 = sock->sk;
	int err = -EINVAL;
	struct sock *sk2 = sk1->sk_prot->accept(sk1, flags, &err);

	if (!sk2)
		goto do_err;

	lock_sock(sk2);

686
	sock_rps_record_flow(sk2);
687 688
	WARN_ON(!((1 << sk2->sk_state) &
		  (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT | TCPF_CLOSE)));
L
Linus Torvalds 已提交
689 690 691 692 693 694 695 696 697

	sock_graft(sk2, newsock);

	newsock->state = SS_CONNECTED;
	err = 0;
	release_sock(sk2);
do_err:
	return err;
}
E
Eric Dumazet 已提交
698
EXPORT_SYMBOL(inet_accept);
L
Linus Torvalds 已提交
699 700 701 702 703 704 705 706 707 708


/*
 *	This does both peername and sockname.
 */
int inet_getname(struct socket *sock, struct sockaddr *uaddr,
			int *uaddr_len, int peer)
{
	struct sock *sk		= sock->sk;
	struct inet_sock *inet	= inet_sk(sk);
709
	DECLARE_SOCKADDR(struct sockaddr_in *, sin, uaddr);
L
Linus Torvalds 已提交
710 711 712

	sin->sin_family = AF_INET;
	if (peer) {
E
Eric Dumazet 已提交
713
		if (!inet->inet_dport ||
L
Linus Torvalds 已提交
714 715 716
		    (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_SYN_SENT)) &&
		     peer == 1))
			return -ENOTCONN;
E
Eric Dumazet 已提交
717 718
		sin->sin_port = inet->inet_dport;
		sin->sin_addr.s_addr = inet->inet_daddr;
L
Linus Torvalds 已提交
719
	} else {
E
Eric Dumazet 已提交
720
		__be32 addr = inet->inet_rcv_saddr;
L
Linus Torvalds 已提交
721
		if (!addr)
E
Eric Dumazet 已提交
722 723
			addr = inet->inet_saddr;
		sin->sin_port = inet->inet_sport;
L
Linus Torvalds 已提交
724 725 726 727 728 729
		sin->sin_addr.s_addr = addr;
	}
	memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
	*uaddr_len = sizeof(*sin);
	return 0;
}
E
Eric Dumazet 已提交
730
EXPORT_SYMBOL(inet_getname);
L
Linus Torvalds 已提交
731 732 733 734 735 736

int inet_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
		 size_t size)
{
	struct sock *sk = sock->sk;

737
	sock_rps_record_flow(sk);
T
Tom Herbert 已提交
738

L
Linus Torvalds 已提交
739
	/* We may need to bind the socket. */
740 741
	if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
	    inet_autobind(sk))
L
Linus Torvalds 已提交
742 743 744 745
		return -EAGAIN;

	return sk->sk_prot->sendmsg(iocb, sk, msg, size);
}
E
Eric Dumazet 已提交
746
EXPORT_SYMBOL(inet_sendmsg);
L
Linus Torvalds 已提交
747

748 749
ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
		      size_t size, int flags)
L
Linus Torvalds 已提交
750 751 752
{
	struct sock *sk = sock->sk;

753
	sock_rps_record_flow(sk);
T
Tom Herbert 已提交
754

L
Linus Torvalds 已提交
755
	/* We may need to bind the socket. */
756 757
	if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
	    inet_autobind(sk))
L
Linus Torvalds 已提交
758 759 760 761 762 763
		return -EAGAIN;

	if (sk->sk_prot->sendpage)
		return sk->sk_prot->sendpage(sk, page, offset, size, flags);
	return sock_no_sendpage(sock, page, offset, size, flags);
}
764
EXPORT_SYMBOL(inet_sendpage);
L
Linus Torvalds 已提交
765

T
Tom Herbert 已提交
766 767 768 769 770 771 772
int inet_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
		 size_t size, int flags)
{
	struct sock *sk = sock->sk;
	int addr_len = 0;
	int err;

773
	sock_rps_record_flow(sk);
T
Tom Herbert 已提交
774 775 776 777 778 779 780 781

	err = sk->sk_prot->recvmsg(iocb, sk, msg, size, flags & MSG_DONTWAIT,
				   flags & ~MSG_DONTWAIT, &addr_len);
	if (err >= 0)
		msg->msg_namelen = addr_len;
	return err;
}
EXPORT_SYMBOL(inet_recvmsg);
L
Linus Torvalds 已提交
782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835

int inet_shutdown(struct socket *sock, int how)
{
	struct sock *sk = sock->sk;
	int err = 0;

	/* This should really check to make sure
	 * the socket is a TCP socket. (WHY AC...)
	 */
	how++; /* maps 0->1 has the advantage of making bit 1 rcvs and
		       1->2 bit 2 snds.
		       2->3 */
	if ((how & ~SHUTDOWN_MASK) || !how)	/* MAXINT->0 */
		return -EINVAL;

	lock_sock(sk);
	if (sock->state == SS_CONNECTING) {
		if ((1 << sk->sk_state) &
		    (TCPF_SYN_SENT | TCPF_SYN_RECV | TCPF_CLOSE))
			sock->state = SS_DISCONNECTING;
		else
			sock->state = SS_CONNECTED;
	}

	switch (sk->sk_state) {
	case TCP_CLOSE:
		err = -ENOTCONN;
		/* Hack to wake up other listeners, who can poll for
		   POLLHUP, even on eg. unconnected UDP sockets -- RR */
	default:
		sk->sk_shutdown |= how;
		if (sk->sk_prot->shutdown)
			sk->sk_prot->shutdown(sk, how);
		break;

	/* Remaining two branches are temporary solution for missing
	 * close() in multithreaded environment. It is _not_ a good idea,
	 * but we have no choice until close() is repaired at VFS level.
	 */
	case TCP_LISTEN:
		if (!(how & RCV_SHUTDOWN))
			break;
		/* Fall through */
	case TCP_SYN_SENT:
		err = sk->sk_prot->disconnect(sk, O_NONBLOCK);
		sock->state = err ? SS_DISCONNECTING : SS_UNCONNECTED;
		break;
	}

	/* Wake up anyone sleeping in poll. */
	sk->sk_state_change(sk);
	release_sock(sk);
	return err;
}
E
Eric Dumazet 已提交
836
EXPORT_SYMBOL(inet_shutdown);
L
Linus Torvalds 已提交
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851

/*
 *	ioctl() calls you can issue on an INET socket. Most of these are
 *	device configuration and stuff and very rarely used. Some ioctls
 *	pass on to the socket itself.
 *
 *	NOTE: I like the idea of a module for the config stuff. ie ifconfig
 *	loads the devconfigure module does its configuring and unloads it.
 *	There's a good 20K of config code hanging around the kernel.
 */

int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
{
	struct sock *sk = sock->sk;
	int err = 0;
852
	struct net *net = sock_net(sk);
L
Linus Torvalds 已提交
853 854

	switch (cmd) {
E
Eric Dumazet 已提交
855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889
	case SIOCGSTAMP:
		err = sock_get_timestamp(sk, (struct timeval __user *)arg);
		break;
	case SIOCGSTAMPNS:
		err = sock_get_timestampns(sk, (struct timespec __user *)arg);
		break;
	case SIOCADDRT:
	case SIOCDELRT:
	case SIOCRTMSG:
		err = ip_rt_ioctl(net, cmd, (void __user *)arg);
		break;
	case SIOCDARP:
	case SIOCGARP:
	case SIOCSARP:
		err = arp_ioctl(net, cmd, (void __user *)arg);
		break;
	case SIOCGIFADDR:
	case SIOCSIFADDR:
	case SIOCGIFBRDADDR:
	case SIOCSIFBRDADDR:
	case SIOCGIFNETMASK:
	case SIOCSIFNETMASK:
	case SIOCGIFDSTADDR:
	case SIOCSIFDSTADDR:
	case SIOCSIFPFLAGS:
	case SIOCGIFPFLAGS:
	case SIOCSIFFLAGS:
		err = devinet_ioctl(net, cmd, (void __user *)arg);
		break;
	default:
		if (sk->sk_prot->ioctl)
			err = sk->sk_prot->ioctl(sk, cmd, arg);
		else
			err = -ENOIOCTLCMD;
		break;
L
Linus Torvalds 已提交
890 891 892
	}
	return err;
}
E
Eric Dumazet 已提交
893
EXPORT_SYMBOL(inet_ioctl);
L
Linus Torvalds 已提交
894

895 896 897 898 899 900 901 902 903 904 905 906 907
#ifdef CONFIG_COMPAT
int inet_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
{
	struct sock *sk = sock->sk;
	int err = -ENOIOCTLCMD;

	if (sk->sk_prot->compat_ioctl)
		err = sk->sk_prot->compat_ioctl(sk, cmd, arg);

	return err;
}
#endif

908
const struct proto_ops inet_stream_ops = {
909 910 911 912 913 914 915 916 917 918 919 920 921 922
	.family		   = PF_INET,
	.owner		   = THIS_MODULE,
	.release	   = inet_release,
	.bind		   = inet_bind,
	.connect	   = inet_stream_connect,
	.socketpair	   = sock_no_socketpair,
	.accept		   = inet_accept,
	.getname	   = inet_getname,
	.poll		   = tcp_poll,
	.ioctl		   = inet_ioctl,
	.listen		   = inet_listen,
	.shutdown	   = inet_shutdown,
	.setsockopt	   = sock_common_setsockopt,
	.getsockopt	   = sock_common_getsockopt,
923
	.sendmsg	   = inet_sendmsg,
T
Tom Herbert 已提交
924
	.recvmsg	   = inet_recvmsg,
925
	.mmap		   = sock_no_mmap,
926
	.sendpage	   = inet_sendpage,
J
Jens Axboe 已提交
927
	.splice_read	   = tcp_splice_read,
928
#ifdef CONFIG_COMPAT
929 930
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
931
	.compat_ioctl	   = inet_compat_ioctl,
932
#endif
L
Linus Torvalds 已提交
933
};
E
Eric Dumazet 已提交
934
EXPORT_SYMBOL(inet_stream_ops);
L
Linus Torvalds 已提交
935

936
const struct proto_ops inet_dgram_ops = {
937 938 939 940 941 942 943 944 945 946 947 948 949 950 951
	.family		   = PF_INET,
	.owner		   = THIS_MODULE,
	.release	   = inet_release,
	.bind		   = inet_bind,
	.connect	   = inet_dgram_connect,
	.socketpair	   = sock_no_socketpair,
	.accept		   = sock_no_accept,
	.getname	   = inet_getname,
	.poll		   = udp_poll,
	.ioctl		   = inet_ioctl,
	.listen		   = sock_no_listen,
	.shutdown	   = inet_shutdown,
	.setsockopt	   = sock_common_setsockopt,
	.getsockopt	   = sock_common_getsockopt,
	.sendmsg	   = inet_sendmsg,
T
Tom Herbert 已提交
952
	.recvmsg	   = inet_recvmsg,
953 954
	.mmap		   = sock_no_mmap,
	.sendpage	   = inet_sendpage,
955
#ifdef CONFIG_COMPAT
956 957
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
958
	.compat_ioctl	   = inet_compat_ioctl,
959
#endif
L
Linus Torvalds 已提交
960
};
E
Eric Dumazet 已提交
961
EXPORT_SYMBOL(inet_dgram_ops);
L
Linus Torvalds 已提交
962 963 964 965 966

/*
 * For SOCK_RAW sockets; should be the same as inet_dgram_ops but without
 * udp_poll
 */
967
static const struct proto_ops inet_sockraw_ops = {
968 969 970 971 972 973 974 975 976 977 978 979 980 981 982
	.family		   = PF_INET,
	.owner		   = THIS_MODULE,
	.release	   = inet_release,
	.bind		   = inet_bind,
	.connect	   = inet_dgram_connect,
	.socketpair	   = sock_no_socketpair,
	.accept		   = sock_no_accept,
	.getname	   = inet_getname,
	.poll		   = datagram_poll,
	.ioctl		   = inet_ioctl,
	.listen		   = sock_no_listen,
	.shutdown	   = inet_shutdown,
	.setsockopt	   = sock_common_setsockopt,
	.getsockopt	   = sock_common_getsockopt,
	.sendmsg	   = inet_sendmsg,
T
Tom Herbert 已提交
983
	.recvmsg	   = inet_recvmsg,
984 985
	.mmap		   = sock_no_mmap,
	.sendpage	   = inet_sendpage,
986
#ifdef CONFIG_COMPAT
987 988
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
989
	.compat_ioctl	   = inet_compat_ioctl,
990
#endif
L
Linus Torvalds 已提交
991 992
};

993
static const struct net_proto_family inet_family_ops = {
L
Linus Torvalds 已提交
994 995 996 997 998 999 1000 1001 1002 1003
	.family = PF_INET,
	.create = inet_create,
	.owner	= THIS_MODULE,
};

/* Upon startup we insert all the elements in inetsw_array[] into
 * the linked list inetsw.
 */
static struct inet_protosw inetsw_array[] =
{
1004 1005 1006 1007 1008 1009 1010
	{
		.type =       SOCK_STREAM,
		.protocol =   IPPROTO_TCP,
		.prot =       &tcp_prot,
		.ops =        &inet_stream_ops,
		.no_check =   0,
		.flags =      INET_PROTOSW_PERMANENT |
1011
			      INET_PROTOSW_ICSK,
1012 1013 1014 1015 1016 1017 1018 1019 1020
	},

	{
		.type =       SOCK_DGRAM,
		.protocol =   IPPROTO_UDP,
		.prot =       &udp_prot,
		.ops =        &inet_dgram_ops,
		.no_check =   UDP_CSUM_DEFAULT,
		.flags =      INET_PROTOSW_PERMANENT,
L
Linus Torvalds 已提交
1021
       },
1022

1023 1024 1025 1026 1027 1028 1029 1030
       {
		.type =       SOCK_DGRAM,
		.protocol =   IPPROTO_ICMP,
		.prot =       &ping_prot,
		.ops =        &inet_dgram_ops,
		.no_check =   UDP_CSUM_DEFAULT,
		.flags =      INET_PROTOSW_REUSE,
       },
L
Linus Torvalds 已提交
1031 1032

       {
1033 1034 1035 1036 1037 1038
	       .type =       SOCK_RAW,
	       .protocol =   IPPROTO_IP,	/* wild card */
	       .prot =       &raw_prot,
	       .ops =        &inet_sockraw_ops,
	       .no_check =   UDP_CSUM_DEFAULT,
	       .flags =      INET_PROTOSW_REUSE,
L
Linus Torvalds 已提交
1039 1040 1041
       }
};

1042
#define INETSW_ARRAY_LEN ARRAY_SIZE(inetsw_array)
L
Linus Torvalds 已提交
1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076

void inet_register_protosw(struct inet_protosw *p)
{
	struct list_head *lh;
	struct inet_protosw *answer;
	int protocol = p->protocol;
	struct list_head *last_perm;

	spin_lock_bh(&inetsw_lock);

	if (p->type >= SOCK_MAX)
		goto out_illegal;

	/* If we are trying to override a permanent protocol, bail. */
	answer = NULL;
	last_perm = &inetsw[p->type];
	list_for_each(lh, &inetsw[p->type]) {
		answer = list_entry(lh, struct inet_protosw, list);

		/* Check only the non-wild match. */
		if (INET_PROTOSW_PERMANENT & answer->flags) {
			if (protocol == answer->protocol)
				break;
			last_perm = lh;
		}

		answer = NULL;
	}
	if (answer)
		goto out_permanent;

	/* Add the new entry after the last permanent entry if any, so that
	 * the new entry does not override a permanent entry when matched with
	 * a wild-card protocol. But it is allowed to override any existing
1077
	 * non-permanent entry.  This means that when we remove this entry, the
L
Linus Torvalds 已提交
1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096
	 * system automatically returns to the old behavior.
	 */
	list_add_rcu(&p->list, last_perm);
out:
	spin_unlock_bh(&inetsw_lock);

	return;

out_permanent:
	printk(KERN_ERR "Attempt to override permanent protocol %d.\n",
	       protocol);
	goto out;

out_illegal:
	printk(KERN_ERR
	       "Ignoring attempt to register invalid socket type %d.\n",
	       p->type);
	goto out;
}
E
Eric Dumazet 已提交
1097
EXPORT_SYMBOL(inet_register_protosw);
L
Linus Torvalds 已提交
1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112

void inet_unregister_protosw(struct inet_protosw *p)
{
	if (INET_PROTOSW_PERMANENT & p->flags) {
		printk(KERN_ERR
		       "Attempt to unregister permanent protocol %d.\n",
		       p->protocol);
	} else {
		spin_lock_bh(&inetsw_lock);
		list_del_rcu(&p->list);
		spin_unlock_bh(&inetsw_lock);

		synchronize_net();
	}
}
E
Eric Dumazet 已提交
1113
EXPORT_SYMBOL(inet_unregister_protosw);
L
Linus Torvalds 已提交
1114

1115 1116 1117 1118
/*
 *      Shall we try to damage output packets if routing dev changes?
 */

1119
int sysctl_ip_dynaddr __read_mostly;
1120 1121 1122 1123

static int inet_sk_reselect_saddr(struct sock *sk)
{
	struct inet_sock *inet = inet_sk(sk);
E
Eric Dumazet 已提交
1124 1125
	__be32 old_saddr = inet->inet_saddr;
	__be32 daddr = inet->inet_daddr;
1126
	struct flowi4 *fl4;
1127 1128
	struct rtable *rt;
	__be32 new_saddr;
1129
	struct ip_options_rcu *inet_opt;
1130

1131 1132 1133 1134
	inet_opt = rcu_dereference_protected(inet->inet_opt,
					     sock_owned_by_user(sk));
	if (inet_opt && inet_opt->opt.srr)
		daddr = inet_opt->opt.faddr;
1135 1136

	/* Query new route. */
1137 1138
	fl4 = &inet->cork.fl.u.ip4;
	rt = ip_route_connect(fl4, daddr, 0, RT_CONN_FLAGS(sk),
1139 1140 1141 1142
			      sk->sk_bound_dev_if, sk->sk_protocol,
			      inet->inet_sport, inet->inet_dport, sk, false);
	if (IS_ERR(rt))
		return PTR_ERR(rt);
1143

1144
	sk_setup_caps(sk, &rt->dst);
1145

1146
	new_saddr = fl4->saddr;
1147 1148 1149 1150 1151

	if (new_saddr == old_saddr)
		return 0;

	if (sysctl_ip_dynaddr > 1) {
1152 1153
		printk(KERN_INFO "%s(): shifting inet->saddr from %pI4 to %pI4\n",
		       __func__, &old_saddr, &new_saddr);
1154 1155
	}

E
Eric Dumazet 已提交
1156
	inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173

	/*
	 * XXX The only one ugly spot where we need to
	 * XXX really change the sockets identity after
	 * XXX it has entered the hashes. -DaveM
	 *
	 * Besides that, it does not check for connection
	 * uniqueness. Wait for troubles.
	 */
	__sk_prot_rehash(sk);
	return 0;
}

int inet_sk_rebuild_header(struct sock *sk)
{
	struct inet_sock *inet = inet_sk(sk);
	struct rtable *rt = (struct rtable *)__sk_dst_check(sk, 0);
A
Al Viro 已提交
1174
	__be32 daddr;
1175
	struct ip_options_rcu *inet_opt;
1176
	struct flowi4 *fl4;
1177 1178 1179 1180 1181 1182 1183
	int err;

	/* Route is OK, nothing to do. */
	if (rt)
		return 0;

	/* Reroute. */
1184 1185
	rcu_read_lock();
	inet_opt = rcu_dereference(inet->inet_opt);
E
Eric Dumazet 已提交
1186
	daddr = inet->inet_daddr;
1187 1188 1189
	if (inet_opt && inet_opt->opt.srr)
		daddr = inet_opt->opt.faddr;
	rcu_read_unlock();
1190 1191
	fl4 = &inet->cork.fl.u.ip4;
	rt = ip_route_output_ports(sock_net(sk), fl4, sk, daddr, inet->inet_saddr,
1192 1193 1194
				   inet->inet_dport, inet->inet_sport,
				   sk->sk_protocol, RT_CONN_FLAGS(sk),
				   sk->sk_bound_dev_if);
1195 1196
	if (!IS_ERR(rt)) {
		err = 0;
1197
		sk_setup_caps(sk, &rt->dst);
1198 1199 1200
	} else {
		err = PTR_ERR(rt);

1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217
		/* Routing failed... */
		sk->sk_route_caps = 0;
		/*
		 * Other protocols have to map its equivalent state to TCP_SYN_SENT.
		 * DCCP maps its DCCP_REQUESTING state to TCP_SYN_SENT. -acme
		 */
		if (!sysctl_ip_dynaddr ||
		    sk->sk_state != TCP_SYN_SENT ||
		    (sk->sk_userlocks & SOCK_BINDADDR_LOCK) ||
		    (err = inet_sk_reselect_saddr(sk)) != 0)
			sk->sk_err_soft = -err;
	}

	return err;
}
EXPORT_SYMBOL(inet_sk_rebuild_header);

1218 1219
static int inet_gso_send_check(struct sk_buff *skb)
{
1220
	const struct iphdr *iph;
1221
	const struct net_protocol *ops;
1222 1223 1224 1225 1226 1227 1228
	int proto;
	int ihl;
	int err = -EINVAL;

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

1229
	iph = ip_hdr(skb);
1230 1231 1232 1233 1234 1235 1236
	ihl = iph->ihl * 4;
	if (ihl < sizeof(*iph))
		goto out;

	if (unlikely(!pskb_may_pull(skb, ihl)))
		goto out;

1237 1238
	__skb_pull(skb, ihl);
	skb_reset_transport_header(skb);
1239
	iph = ip_hdr(skb);
1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252
	proto = iph->protocol & (MAX_INET_PROTOS - 1);
	err = -EPROTONOSUPPORT;

	rcu_read_lock();
	ops = rcu_dereference(inet_protos[proto]);
	if (likely(ops && ops->gso_send_check))
		err = ops->gso_send_check(skb);
	rcu_read_unlock();

out:
	return err;
}

1253
static struct sk_buff *inet_gso_segment(struct sk_buff *skb, u32 features)
H
Herbert Xu 已提交
1254 1255 1256
{
	struct sk_buff *segs = ERR_PTR(-EINVAL);
	struct iphdr *iph;
1257
	const struct net_protocol *ops;
H
Herbert Xu 已提交
1258 1259 1260
	int proto;
	int ihl;
	int id;
1261
	unsigned int offset = 0;
H
Herbert Xu 已提交
1262

1263 1264 1265
	if (!(features & NETIF_F_V4_CSUM))
		features &= ~NETIF_F_SG;

1266 1267 1268 1269 1270 1271 1272 1273 1274
	if (unlikely(skb_shinfo(skb)->gso_type &
		     ~(SKB_GSO_TCPV4 |
		       SKB_GSO_UDP |
		       SKB_GSO_DODGY |
		       SKB_GSO_TCP_ECN |
		       0)))
		goto out;

	if (unlikely(!pskb_may_pull(skb, sizeof(*iph))))
H
Herbert Xu 已提交
1275 1276
		goto out;

1277
	iph = ip_hdr(skb);
H
Herbert Xu 已提交
1278 1279 1280 1281
	ihl = iph->ihl * 4;
	if (ihl < sizeof(*iph))
		goto out;

1282
	if (unlikely(!pskb_may_pull(skb, ihl)))
H
Herbert Xu 已提交
1283 1284
		goto out;

1285 1286
	__skb_pull(skb, ihl);
	skb_reset_transport_header(skb);
1287
	iph = ip_hdr(skb);
H
Herbert Xu 已提交
1288 1289 1290 1291 1292 1293
	id = ntohs(iph->id);
	proto = iph->protocol & (MAX_INET_PROTOS - 1);
	segs = ERR_PTR(-EPROTONOSUPPORT);

	rcu_read_lock();
	ops = rcu_dereference(inet_protos[proto]);
1294
	if (likely(ops && ops->gso_segment))
1295
		segs = ops->gso_segment(skb, features);
H
Herbert Xu 已提交
1296 1297
	rcu_read_unlock();

1298
	if (!segs || IS_ERR(segs))
H
Herbert Xu 已提交
1299 1300 1301 1302
		goto out;

	skb = segs;
	do {
1303
		iph = ip_hdr(skb);
1304 1305 1306 1307 1308 1309 1310 1311
		if (proto == IPPROTO_UDP) {
			iph->id = htons(id);
			iph->frag_off = htons(offset >> 3);
			if (skb->next != NULL)
				iph->frag_off |= htons(IP_MF);
			offset += (skb->len - skb->mac_len - iph->ihl * 4);
		} else
			iph->id = htons(id++);
H
Herbert Xu 已提交
1312 1313
		iph->tot_len = htons(skb->len - skb->mac_len);
		iph->check = 0;
1314
		iph->check = ip_fast_csum(skb_network_header(skb), iph->ihl);
H
Herbert Xu 已提交
1315 1316 1317 1318 1319 1320
	} while ((skb = skb->next));

out:
	return segs;
}

H
Herbert Xu 已提交
1321 1322 1323
static struct sk_buff **inet_gro_receive(struct sk_buff **head,
					 struct sk_buff *skb)
{
1324
	const struct net_protocol *ops;
H
Herbert Xu 已提交
1325 1326
	struct sk_buff **pp = NULL;
	struct sk_buff *p;
1327
	const struct iphdr *iph;
1328 1329
	unsigned int hlen;
	unsigned int off;
1330
	unsigned int id;
H
Herbert Xu 已提交
1331 1332 1333
	int flush = 1;
	int proto;

1334 1335 1336 1337 1338 1339 1340 1341
	off = skb_gro_offset(skb);
	hlen = off + sizeof(*iph);
	iph = skb_gro_header_fast(skb, off);
	if (skb_gro_header_hard(skb, hlen)) {
		iph = skb_gro_header_slow(skb, hlen, off);
		if (unlikely(!iph))
			goto out;
	}
H
Herbert Xu 已提交
1342 1343 1344 1345 1346 1347 1348 1349

	proto = iph->protocol & (MAX_INET_PROTOS - 1);

	rcu_read_lock();
	ops = rcu_dereference(inet_protos[proto]);
	if (!ops || !ops->gro_receive)
		goto out_unlock;

H
Herbert Xu 已提交
1350
	if (*(u8 *)iph != 0x45)
H
Herbert Xu 已提交
1351 1352 1353 1354 1355
		goto out_unlock;

	if (unlikely(ip_fast_csum((u8 *)iph, iph->ihl)))
		goto out_unlock;

1356 1357
	id = ntohl(*(__be32 *)&iph->id);
	flush = (u16)((ntohl(*(__be32 *)iph) ^ skb_gro_len(skb)) | (id ^ IP_DF));
1358
	id >>= 16;
H
Herbert Xu 已提交
1359 1360 1361 1362 1363 1364 1365 1366 1367

	for (p = *head; p; p = p->next) {
		struct iphdr *iph2;

		if (!NAPI_GRO_CB(p)->same_flow)
			continue;

		iph2 = ip_hdr(p);

H
Herbert Xu 已提交
1368 1369
		if ((iph->protocol ^ iph2->protocol) |
		    (iph->tos ^ iph2->tos) |
1370 1371
		    ((__force u32)iph->saddr ^ (__force u32)iph2->saddr) |
		    ((__force u32)iph->daddr ^ (__force u32)iph2->daddr)) {
H
Herbert Xu 已提交
1372 1373 1374 1375 1376 1377
			NAPI_GRO_CB(p)->same_flow = 0;
			continue;
		}

		/* All fields must match except length and checksum. */
		NAPI_GRO_CB(p)->flush |=
H
Herbert Xu 已提交
1378 1379
			(iph->ttl ^ iph2->ttl) |
			((u16)(ntohs(iph2->id) + NAPI_GRO_CB(p)->count) ^ id);
H
Herbert Xu 已提交
1380 1381 1382 1383 1384

		NAPI_GRO_CB(p)->flush |= flush;
	}

	NAPI_GRO_CB(skb)->flush |= flush;
1385 1386
	skb_gro_pull(skb, sizeof(*iph));
	skb_set_transport_header(skb, skb_gro_offset(skb));
H
Herbert Xu 已提交
1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400

	pp = ops->gro_receive(head, skb);

out_unlock:
	rcu_read_unlock();

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

	return pp;
}

static int inet_gro_complete(struct sk_buff *skb)
{
1401
	const struct net_protocol *ops;
H
Herbert Xu 已提交
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422
	struct iphdr *iph = ip_hdr(skb);
	int proto = iph->protocol & (MAX_INET_PROTOS - 1);
	int err = -ENOSYS;
	__be16 newlen = htons(skb->len - skb_network_offset(skb));

	csum_replace2(&iph->check, iph->tot_len, newlen);
	iph->tot_len = newlen;

	rcu_read_lock();
	ops = rcu_dereference(inet_protos[proto]);
	if (WARN_ON(!ops || !ops->gro_complete))
		goto out_unlock;

	err = ops->gro_complete(skb);

out_unlock:
	rcu_read_unlock();

	return err;
}

1423
int inet_ctl_sock_create(struct sock **sk, unsigned short family,
1424 1425
			 unsigned short type, unsigned char protocol,
			 struct net *net)
1426
{
1427 1428
	struct socket *sock;
	int rc = sock_create_kern(family, type, protocol, &sock);
1429 1430

	if (rc == 0) {
1431 1432
		*sk = sock->sk;
		(*sk)->sk_allocation = GFP_ATOMIC;
1433 1434 1435 1436
		/*
		 * Unhash it so that IP input processing does not even see it,
		 * we do not wish this socket to see incoming packets.
		 */
1437
		(*sk)->sk_prot->unhash(*sk);
1438 1439

		sk_change_net(*sk, net);
1440 1441 1442 1443 1444
	}
	return rc;
}
EXPORT_SYMBOL_GPL(inet_ctl_sock_create);

1445
unsigned long snmp_fold_field(void __percpu *mib[], int offt)
1446 1447
{
	unsigned long res = 0;
E
Eric Dumazet 已提交
1448
	int i, j;
1449 1450

	for_each_possible_cpu(i) {
E
Eric Dumazet 已提交
1451 1452
		for (j = 0; j < SNMP_ARRAY_SZ; j++)
			res += *(((unsigned long *) per_cpu_ptr(mib[j], i)) + offt);
1453 1454 1455 1456 1457
	}
	return res;
}
EXPORT_SYMBOL_GPL(snmp_fold_field);

1458 1459 1460 1461 1462 1463 1464 1465
#if BITS_PER_LONG==32

u64 snmp_fold_field64(void __percpu *mib[], int offt, size_t syncp_offset)
{
	u64 res = 0;
	int cpu;

	for_each_possible_cpu(cpu) {
E
Eric Dumazet 已提交
1466
		void *bhptr;
1467
		struct u64_stats_sync *syncp;
E
Eric Dumazet 已提交
1468
		u64 v;
1469 1470
		unsigned int start;

E
Eric Dumazet 已提交
1471
		bhptr = per_cpu_ptr(mib[0], cpu);
1472 1473 1474
		syncp = (struct u64_stats_sync *)(bhptr + syncp_offset);
		do {
			start = u64_stats_fetch_begin_bh(syncp);
E
Eric Dumazet 已提交
1475
			v = *(((u64 *) bhptr) + offt);
1476 1477
		} while (u64_stats_fetch_retry_bh(syncp, start));

E
Eric Dumazet 已提交
1478
		res += v;
1479 1480 1481 1482 1483 1484
	}
	return res;
}
EXPORT_SYMBOL_GPL(snmp_fold_field64);
#endif

1485
int snmp_mib_init(void __percpu *ptr[2], size_t mibsize, size_t align)
1486 1487
{
	BUG_ON(ptr == NULL);
1488
	ptr[0] = __alloc_percpu(mibsize, align);
1489
	if (!ptr[0])
E
Eric Dumazet 已提交
1490 1491
		return -ENOMEM;
#if SNMP_ARRAY_SZ == 2
1492
	ptr[1] = __alloc_percpu(mibsize, align);
E
Eric Dumazet 已提交
1493 1494 1495 1496 1497 1498
	if (!ptr[1]) {
		free_percpu(ptr[0]);
		ptr[0] = NULL;
		return -ENOMEM;
	}
#endif
1499 1500 1501 1502
	return 0;
}
EXPORT_SYMBOL_GPL(snmp_mib_init);

E
Eric Dumazet 已提交
1503
void snmp_mib_free(void __percpu *ptr[SNMP_ARRAY_SZ])
1504
{
E
Eric Dumazet 已提交
1505 1506
	int i;

1507
	BUG_ON(ptr == NULL);
E
Eric Dumazet 已提交
1508 1509 1510 1511
	for (i = 0; i < SNMP_ARRAY_SZ; i++) {
		free_percpu(ptr[i]);
		ptr[i] = NULL;
	}
1512 1513 1514
}
EXPORT_SYMBOL_GPL(snmp_mib_free);

L
Linus Torvalds 已提交
1515
#ifdef CONFIG_IP_MULTICAST
1516
static const struct net_protocol igmp_protocol = {
L
Linus Torvalds 已提交
1517
	.handler =	igmp_rcv,
1518
	.netns_ok =	1,
L
Linus Torvalds 已提交
1519 1520 1521
};
#endif

1522
static const struct net_protocol tcp_protocol = {
L
Linus Torvalds 已提交
1523 1524
	.handler =	tcp_v4_rcv,
	.err_handler =	tcp_v4_err,
1525
	.gso_send_check = tcp_v4_gso_send_check,
H
Herbert Xu 已提交
1526
	.gso_segment =	tcp_tso_segment,
H
Herbert Xu 已提交
1527 1528
	.gro_receive =	tcp4_gro_receive,
	.gro_complete =	tcp4_gro_complete,
L
Linus Torvalds 已提交
1529
	.no_policy =	1,
1530
	.netns_ok =	1,
L
Linus Torvalds 已提交
1531 1532
};

1533
static const struct net_protocol udp_protocol = {
L
Linus Torvalds 已提交
1534 1535
	.handler =	udp_rcv,
	.err_handler =	udp_err,
1536 1537
	.gso_send_check = udp4_ufo_send_check,
	.gso_segment = udp4_ufo_fragment,
L
Linus Torvalds 已提交
1538
	.no_policy =	1,
1539
	.netns_ok =	1,
L
Linus Torvalds 已提交
1540 1541
};

1542
static const struct net_protocol icmp_protocol = {
L
Linus Torvalds 已提交
1543
	.handler =	icmp_rcv,
1544
	.err_handler =	ping_err,
1545
	.no_policy =	1,
1546
	.netns_ok =	1,
L
Linus Torvalds 已提交
1547 1548
};

1549 1550
static __net_init int ipv4_mib_init_net(struct net *net)
{
1551
	if (snmp_mib_init((void __percpu **)net->mib.tcp_statistics,
1552 1553
			  sizeof(struct tcp_mib),
			  __alignof__(struct tcp_mib)) < 0)
1554
		goto err_tcp_mib;
1555
	if (snmp_mib_init((void __percpu **)net->mib.ip_statistics,
1556 1557
			  sizeof(struct ipstats_mib),
			  __alignof__(struct ipstats_mib)) < 0)
1558
		goto err_ip_mib;
1559
	if (snmp_mib_init((void __percpu **)net->mib.net_statistics,
1560 1561
			  sizeof(struct linux_mib),
			  __alignof__(struct linux_mib)) < 0)
1562
		goto err_net_mib;
1563
	if (snmp_mib_init((void __percpu **)net->mib.udp_statistics,
1564 1565
			  sizeof(struct udp_mib),
			  __alignof__(struct udp_mib)) < 0)
1566
		goto err_udp_mib;
1567
	if (snmp_mib_init((void __percpu **)net->mib.udplite_statistics,
1568 1569
			  sizeof(struct udp_mib),
			  __alignof__(struct udp_mib)) < 0)
1570
		goto err_udplite_mib;
1571
	if (snmp_mib_init((void __percpu **)net->mib.icmp_statistics,
1572 1573
			  sizeof(struct icmp_mib),
			  __alignof__(struct icmp_mib)) < 0)
1574
		goto err_icmp_mib;
1575
	if (snmp_mib_init((void __percpu **)net->mib.icmpmsg_statistics,
1576 1577
			  sizeof(struct icmpmsg_mib),
			  __alignof__(struct icmpmsg_mib)) < 0)
1578
		goto err_icmpmsg_mib;
1579 1580

	tcp_mib_init(net);
1581
	return 0;
1582

1583
err_icmpmsg_mib:
1584
	snmp_mib_free((void __percpu **)net->mib.icmp_statistics);
1585
err_icmp_mib:
1586
	snmp_mib_free((void __percpu **)net->mib.udplite_statistics);
1587
err_udplite_mib:
1588
	snmp_mib_free((void __percpu **)net->mib.udp_statistics);
1589
err_udp_mib:
1590
	snmp_mib_free((void __percpu **)net->mib.net_statistics);
1591
err_net_mib:
1592
	snmp_mib_free((void __percpu **)net->mib.ip_statistics);
1593
err_ip_mib:
1594
	snmp_mib_free((void __percpu **)net->mib.tcp_statistics);
1595 1596
err_tcp_mib:
	return -ENOMEM;
1597 1598 1599 1600
}

static __net_exit void ipv4_mib_exit_net(struct net *net)
{
1601 1602 1603 1604 1605 1606 1607
	snmp_mib_free((void __percpu **)net->mib.icmpmsg_statistics);
	snmp_mib_free((void __percpu **)net->mib.icmp_statistics);
	snmp_mib_free((void __percpu **)net->mib.udplite_statistics);
	snmp_mib_free((void __percpu **)net->mib.udp_statistics);
	snmp_mib_free((void __percpu **)net->mib.net_statistics);
	snmp_mib_free((void __percpu **)net->mib.ip_statistics);
	snmp_mib_free((void __percpu **)net->mib.tcp_statistics);
1608 1609 1610 1611 1612 1613 1614
}

static __net_initdata struct pernet_operations ipv4_mib_ops = {
	.init = ipv4_mib_init_net,
	.exit = ipv4_mib_exit_net,
};

L
Linus Torvalds 已提交
1615 1616
static int __init init_ipv4_mibs(void)
{
1617
	return register_pernet_subsys(&ipv4_mib_ops);
L
Linus Torvalds 已提交
1618 1619 1620 1621
}

static int ipv4_proc_init(void);

1622 1623 1624 1625
/*
 *	IP protocol layer initialiser
 */

1626
static struct packet_type ip_packet_type __read_mostly = {
1627
	.type = cpu_to_be16(ETH_P_IP),
1628
	.func = ip_rcv,
1629
	.gso_send_check = inet_gso_send_check,
H
Herbert Xu 已提交
1630
	.gso_segment = inet_gso_segment,
H
Herbert Xu 已提交
1631 1632
	.gro_receive = inet_gro_receive,
	.gro_complete = inet_gro_complete,
1633 1634
};

L
Linus Torvalds 已提交
1635 1636 1637 1638 1639 1640 1641
static int __init inet_init(void)
{
	struct sk_buff *dummy_skb;
	struct inet_protosw *q;
	struct list_head *r;
	int rc = -EINVAL;

1642
	BUILD_BUG_ON(sizeof(struct inet_skb_parm) > sizeof(dummy_skb->cb));
L
Linus Torvalds 已提交
1643

1644 1645 1646 1647
	sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL);
	if (!sysctl_local_reserved_ports)
		goto out;

L
Linus Torvalds 已提交
1648 1649
	rc = proto_register(&tcp_prot, 1);
	if (rc)
1650
		goto out_free_reserved_ports;
L
Linus Torvalds 已提交
1651 1652 1653 1654 1655 1656 1657 1658 1659

	rc = proto_register(&udp_prot, 1);
	if (rc)
		goto out_unregister_tcp_proto;

	rc = proto_register(&raw_prot, 1);
	if (rc)
		goto out_unregister_udp_proto;

1660 1661 1662 1663
	rc = proto_register(&ping_prot, 1);
	if (rc)
		goto out_unregister_raw_proto;

L
Linus Torvalds 已提交
1664
	/*
1665
	 *	Tell SOCKET that we are alive...
L
Linus Torvalds 已提交
1666 1667
	 */

1668
	(void)sock_register(&inet_family_ops);
L
Linus Torvalds 已提交
1669

1670 1671 1672 1673
#ifdef CONFIG_SYSCTL
	ip_static_sysctl_init();
#endif

L
Linus Torvalds 已提交
1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701
	/*
	 *	Add all the base protocols.
	 */

	if (inet_add_protocol(&icmp_protocol, IPPROTO_ICMP) < 0)
		printk(KERN_CRIT "inet_init: Cannot add ICMP protocol\n");
	if (inet_add_protocol(&udp_protocol, IPPROTO_UDP) < 0)
		printk(KERN_CRIT "inet_init: Cannot add UDP protocol\n");
	if (inet_add_protocol(&tcp_protocol, IPPROTO_TCP) < 0)
		printk(KERN_CRIT "inet_init: Cannot add TCP protocol\n");
#ifdef CONFIG_IP_MULTICAST
	if (inet_add_protocol(&igmp_protocol, IPPROTO_IGMP) < 0)
		printk(KERN_CRIT "inet_init: Cannot add IGMP protocol\n");
#endif

	/* Register the socket-side information for inet_create. */
	for (r = &inetsw[0]; r < &inetsw[SOCK_MAX]; ++r)
		INIT_LIST_HEAD(r);

	for (q = inetsw_array; q < &inetsw_array[INETSW_ARRAY_LEN]; ++q)
		inet_register_protosw(q);

	/*
	 *	Set the ARP module up
	 */

	arp_init();

1702 1703 1704
	/*
	 *	Set the IP module up
	 */
L
Linus Torvalds 已提交
1705 1706 1707

	ip_init();

1708
	tcp_v4_init();
L
Linus Torvalds 已提交
1709 1710 1711 1712

	/* Setup TCP slab cache for open requests. */
	tcp_init();

H
Hideo Aoki 已提交
1713 1714 1715
	/* Setup UDP memory threshold */
	udp_init();

1716 1717
	/* Add UDP-Lite (RFC 3828) */
	udplite4_register();
L
Linus Torvalds 已提交
1718

1719 1720
	ping_init();

L
Linus Torvalds 已提交
1721 1722 1723 1724
	/*
	 *	Set the ICMP layer up
	 */

1725 1726
	if (icmp_init() < 0)
		panic("Failed to create the ICMP control socket.\n");
L
Linus Torvalds 已提交
1727 1728 1729 1730 1731

	/*
	 *	Initialise the multicast router
	 */
#if defined(CONFIG_IP_MROUTE)
W
Wang Chen 已提交
1732 1733
	if (ip_mr_init())
		printk(KERN_CRIT "inet_init: Cannot init ipv4 mroute\n");
L
Linus Torvalds 已提交
1734 1735 1736
#endif
	/*
	 *	Initialise per-cpu ipv4 mibs
1737
	 */
L
Linus Torvalds 已提交
1738

S
Stephen Hemminger 已提交
1739
	if (init_ipv4_mibs())
W
Wang Chen 已提交
1740
		printk(KERN_CRIT "inet_init: Cannot init ipv4 mibs\n");
1741

L
Linus Torvalds 已提交
1742 1743 1744 1745
	ipv4_proc_init();

	ipfrag_init();

1746 1747
	dev_add_pack(&ip_packet_type);

L
Linus Torvalds 已提交
1748 1749 1750
	rc = 0;
out:
	return rc;
1751 1752
out_unregister_raw_proto:
	proto_unregister(&raw_prot);
L
Linus Torvalds 已提交
1753 1754
out_unregister_udp_proto:
	proto_unregister(&udp_prot);
1755 1756
out_unregister_tcp_proto:
	proto_unregister(&tcp_prot);
1757 1758
out_free_reserved_ports:
	kfree(sysctl_local_reserved_ports);
L
Linus Torvalds 已提交
1759 1760 1761
	goto out;
}

1762
fs_initcall(inet_init);
L
Linus Torvalds 已提交
1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776

/* ------------------------------------------------------------------------ */

#ifdef CONFIG_PROC_FS
static int __init ipv4_proc_init(void)
{
	int rc = 0;

	if (raw_proc_init())
		goto out_raw;
	if (tcp4_proc_init())
		goto out_tcp;
	if (udp4_proc_init())
		goto out_udp;
1777 1778
	if (ping_proc_init())
		goto out_ping;
L
Linus Torvalds 已提交
1779 1780 1781 1782 1783
	if (ip_misc_proc_init())
		goto out_misc;
out:
	return rc;
out_misc:
1784 1785
	ping_proc_exit();
out_ping:
L
Linus Torvalds 已提交
1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804
	udp4_proc_exit();
out_udp:
	tcp4_proc_exit();
out_tcp:
	raw_proc_exit();
out_raw:
	rc = -ENOMEM;
	goto out;
}

#else /* CONFIG_PROC_FS */
static int __init ipv4_proc_init(void)
{
	return 0;
}
#endif /* CONFIG_PROC_FS */

MODULE_ALIAS_NETPROTO(PF_INET);