You need to sign in or sign up before continuing.
af_inet.c 44.5 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.
 */

68 69
#define pr_fmt(fmt) "IPv4: " fmt

H
Herbert Xu 已提交
70
#include <linux/err.h>
L
Linus Torvalds 已提交
71 72 73 74 75 76 77 78 79 80 81
#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>
82
#include <linux/capability.h>
L
Linus Torvalds 已提交
83 84 85 86 87 88 89
#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>
90
#include <linux/random.h>
91
#include <linux/slab.h>
L
Linus Torvalds 已提交
92 93 94 95 96

#include <asm/uaccess.h>

#include <linux/inet.h>
#include <linux/igmp.h>
97
#include <linux/inetdevice.h>
L
Linus Torvalds 已提交
98
#include <linux/netdevice.h>
H
Herbert Xu 已提交
99
#include <net/checksum.h>
L
Linus Torvalds 已提交
100 101 102 103 104
#include <net/ip.h>
#include <net/protocol.h>
#include <net/arp.h>
#include <net/route.h>
#include <net/ip_fib.h>
105
#include <net/inet_connection_sock.h>
L
Linus Torvalds 已提交
106 107
#include <net/tcp.h>
#include <net/udp.h>
108
#include <net/udplite.h>
109
#include <net/ping.h>
L
Linus Torvalds 已提交
110 111 112 113 114 115 116
#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>
117
#include <net/net_namespace.h>
L
Linus Torvalds 已提交
118 119 120 121 122 123 124 125 126 127 128
#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);

129 130 131
struct ipv4_config ipv4_config;
EXPORT_SYMBOL(ipv4_config);

L
Linus Torvalds 已提交
132 133 134 135 136 137 138 139 140
/* 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
141 142
	sk_mem_reclaim(sk);

L
Linus Torvalds 已提交
143
	if (sk->sk_type == SOCK_STREAM && sk->sk_state != TCP_CLOSE) {
E
Eric Dumazet 已提交
144
		pr_err("Attempt to release TCP socket in state %d %p\n",
L
Linus Torvalds 已提交
145 146 147 148
		       sk->sk_state, sk);
		return;
	}
	if (!sock_flag(sk, SOCK_DEAD)) {
E
Eric Dumazet 已提交
149
		pr_err("Attempt to release alive inet socket %p\n", sk);
L
Linus Torvalds 已提交
150 151 152
		return;
	}

153 154 155 156
	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
157

158
	kfree(rcu_dereference_protected(inet->inet_opt, 1));
E
Eric Dumazet 已提交
159
	dst_release(rcu_dereference_check(sk->sk_dst_cache, 1));
D
David S. Miller 已提交
160
	dst_release(sk->sk_rx_dst);
161
	sk_refcnt_debug_dec(sk);
L
Linus Torvalds 已提交
162
}
E
Eric Dumazet 已提交
163
EXPORT_SYMBOL(inet_sock_destruct);
L
Linus Torvalds 已提交
164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180

/*
 *	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
181
	if (!inet->inet_num) {
L
Linus Torvalds 已提交
182 183 184 185
		if (sk->sk_prot->get_port(sk, 0)) {
			release_sock(sk);
			return -EAGAIN;
		}
E
Eric Dumazet 已提交
186
		inet->inet_sport = htons(inet->inet_num);
L
Linus Torvalds 已提交
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
	}
	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) {
215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234
		/* Check special setups for testing purpose to enable TFO w/o
		 * requiring TCP_FASTOPEN sockopt.
		 * Note that only TCP sockets (SOCK_STREAM) will reach here.
		 * Also fastopenq may already been allocated because this
		 * socket was in TCP_LISTEN state previously but was
		 * shutdown() (rather than close()).
		 */
		if ((sysctl_tcp_fastopen & TFO_SERVER_ENABLE) != 0 &&
		    inet_csk(sk)->icsk_accept_queue.fastopenq == NULL) {
			if ((sysctl_tcp_fastopen & TFO_SERVER_WO_SOCKOPT1) != 0)
				err = fastopen_init_queue(sk, backlog);
			else if ((sysctl_tcp_fastopen &
				  TFO_SERVER_WO_SOCKOPT2) != 0)
				err = fastopen_init_queue(sk,
				    ((uint)sysctl_tcp_fastopen) >> 16);
			else
				err = 0;
			if (err)
				goto out;
		}
235
		err = inet_csk_listen_start(sk, backlog);
L
Linus Torvalds 已提交
236 237 238 239 240 241 242 243 244 245
		if (err)
			goto out;
	}
	sk->sk_max_ack_backlog = backlog;
	err = 0;

out:
	release_sock(sk);
	return err;
}
E
Eric Dumazet 已提交
246
EXPORT_SYMBOL(inet_listen);
L
Linus Torvalds 已提交
247

248
u32 inet_ehash_secret __read_mostly;
249 250
EXPORT_SYMBOL(inet_ehash_secret);

251 252 253
/*
 * inet_ehash_secret must be set exactly once
 */
254 255
void build_ehash_secret(void)
{
256
	u32 rnd;
257

258 259 260
	do {
		get_random_bytes(&rnd, sizeof(rnd));
	} while (rnd == 0);
261 262

	cmpxchg(&inet_ehash_secret, 0, rnd);
263 264 265
}
EXPORT_SYMBOL(build_ehash_secret);

266
static inline int inet_netns_ok(struct net *net, __u8 protocol)
267
{
268
	const struct net_protocol *ipprot;
269

E
Eric Dumazet 已提交
270
	if (net_eq(net, &init_net))
271 272
		return 1;

273 274
	ipprot = rcu_dereference(inet_protos[protocol]);
	if (ipprot == NULL) {
275 276
		/* raw IP is OK */
		return 1;
277
	}
278 279 280
	return ipprot->netns_ok;
}

L
Linus Torvalds 已提交
281 282 283 284
/*
 *	Create an inet socket.
 */

285 286
static int inet_create(struct net *net, struct socket *sock, int protocol,
		       int kern)
L
Linus Torvalds 已提交
287 288 289 290 291 292 293
{
	struct sock *sk;
	struct inet_protosw *answer;
	struct inet_sock *inet;
	struct proto *answer_prot;
	unsigned char answer_flags;
	char answer_no_check;
294
	int try_loading_module = 0;
295
	int err;
L
Linus Torvalds 已提交
296

E
Eric Dumazet 已提交
297 298 299
	if (unlikely(!inet_ehash_secret))
		if (sock->type != SOCK_RAW && sock->type != SOCK_DGRAM)
			build_ehash_secret();
300

L
Linus Torvalds 已提交
301 302 303
	sock->state = SS_UNCONNECTED;

	/* Look for the requested type/protocol pair. */
304
lookup_protocol:
305
	err = -ESOCKTNOSUPPORT;
L
Linus Torvalds 已提交
306
	rcu_read_lock();
307
	list_for_each_entry_rcu(answer, &inetsw[sock->type], list) {
L
Linus Torvalds 已提交
308

309
		err = 0;
L
Linus Torvalds 已提交
310 311 312 313 314 315 316 317 318 319 320 321 322
		/* 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;
		}
323
		err = -EPROTONOSUPPORT;
L
Linus Torvalds 已提交
324 325
	}

326
	if (unlikely(err)) {
327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347
		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
348
	err = -EPERM;
349 350
	if (sock->type == SOCK_RAW && !kern &&
	    !ns_capable(net->user_ns, CAP_NET_RAW))
L
Linus Torvalds 已提交
351 352
		goto out_rcu_unlock;

353 354 355 356
	err = -EAFNOSUPPORT;
	if (!inet_netns_ok(net, protocol))
		goto out_rcu_unlock;

L
Linus Torvalds 已提交
357 358 359 360 361 362
	sock->ops = answer->ops;
	answer_prot = answer->prot;
	answer_no_check = answer->no_check;
	answer_flags = answer->flags;
	rcu_read_unlock();

363
	WARN_ON(answer_prot->slab == NULL);
L
Linus Torvalds 已提交
364 365

	err = -ENOBUFS;
366
	sk = sk_alloc(net, PF_INET, GFP_KERNEL, answer_prot);
L
Linus Torvalds 已提交
367 368 369 370 371 372
	if (sk == NULL)
		goto out;

	err = 0;
	sk->sk_no_check = answer_no_check;
	if (INET_PROTOSW_REUSE & answer_flags)
373
		sk->sk_reuse = SK_CAN_REUSE;
L
Linus Torvalds 已提交
374 375

	inet = inet_sk(sk);
376
	inet->is_icsk = (INET_PROTOSW_ICSK & answer_flags) != 0;
L
Linus Torvalds 已提交
377

378 379
	inet->nodefrag = 0;

L
Linus Torvalds 已提交
380
	if (SOCK_RAW == sock->type) {
E
Eric Dumazet 已提交
381
		inet->inet_num = protocol;
L
Linus Torvalds 已提交
382 383 384 385 386 387 388 389 390
		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
391
	inet->inet_id = 0;
L
Linus Torvalds 已提交
392 393 394 395 396 397 398 399 400 401

	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;
402
	inet->mc_all	= 1;
L
Linus Torvalds 已提交
403 404
	inet->mc_index	= 0;
	inet->mc_list	= NULL;
405
	inet->rcv_tos	= 0;
L
Linus Torvalds 已提交
406

407
	sk_refcnt_debug_inc(sk);
L
Linus Torvalds 已提交
408

E
Eric Dumazet 已提交
409
	if (inet->inet_num) {
L
Linus Torvalds 已提交
410 411 412 413 414
		/* 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
415
		inet->inet_sport = htons(inet->inet_num);
L
Linus Torvalds 已提交
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
		/* 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;

445
		sock_rps_reset_flow(sk);
T
Tom Herbert 已提交
446

L
Linus Torvalds 已提交
447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465
		/* 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
466
EXPORT_SYMBOL(inet_release);
L
Linus Torvalds 已提交
467 468

/* It is off by default, see below. */
469
int sysctl_ip_nonlocal_bind __read_mostly;
E
Eric Dumazet 已提交
470
EXPORT_SYMBOL(sysctl_ip_nonlocal_bind);
L
Linus Torvalds 已提交
471 472 473 474 475 476

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);
477
	struct net *net = sock_net(sk);
L
Linus Torvalds 已提交
478 479 480 481 482 483 484 485 486 487 488 489 490
	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;

491
	if (addr->sin_family != AF_INET) {
492 493 494
		/* Compatibility games : accept AF_UNSPEC (mapped to AF_INET)
		 * only if s_addr is INADDR_ANY.
		 */
495
		err = -EAFNOSUPPORT;
496 497 498
		if (addr->sin_family != AF_UNSPEC ||
		    addr->sin_addr.s_addr != htonl(INADDR_ANY))
			goto out;
499
	}
500

501
	chk_addr_ret = inet_addr_type(net, addr->sin_addr.s_addr);
L
Linus Torvalds 已提交
502 503 504 505 506 507 508 509 510 511

	/* 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 &&
512
	    !(inet->freebind || inet->transparent) &&
A
Al Viro 已提交
513
	    addr->sin_addr.s_addr != htonl(INADDR_ANY) &&
L
Linus Torvalds 已提交
514 515 516 517 518 519 520
	    chk_addr_ret != RTN_LOCAL &&
	    chk_addr_ret != RTN_MULTICAST &&
	    chk_addr_ret != RTN_BROADCAST)
		goto out;

	snum = ntohs(addr->sin_port);
	err = -EACCES;
521 522
	if (snum && snum < PROT_SOCK &&
	    !ns_capable(net->user_ns, CAP_NET_BIND_SERVICE))
L
Linus Torvalds 已提交
523 524 525 526 527 528 529 530 531 532 533 534 535
		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
536
	if (sk->sk_state != TCP_CLOSE || inet->inet_num)
L
Linus Torvalds 已提交
537 538
		goto out_release_sock;

E
Eric Dumazet 已提交
539
	inet->inet_rcv_saddr = inet->inet_saddr = addr->sin_addr.s_addr;
L
Linus Torvalds 已提交
540
	if (chk_addr_ret == RTN_MULTICAST || chk_addr_ret == RTN_BROADCAST)
E
Eric Dumazet 已提交
541
		inet->inet_saddr = 0;  /* Use device */
L
Linus Torvalds 已提交
542 543 544

	/* Make sure we are allowed to bind here. */
	if (sk->sk_prot->get_port(sk, snum)) {
E
Eric Dumazet 已提交
545
		inet->inet_saddr = inet->inet_rcv_saddr = 0;
L
Linus Torvalds 已提交
546 547 548 549
		err = -EADDRINUSE;
		goto out_release_sock;
	}

E
Eric Dumazet 已提交
550
	if (inet->inet_rcv_saddr)
L
Linus Torvalds 已提交
551 552 553
		sk->sk_userlocks |= SOCK_BINDADDR_LOCK;
	if (snum)
		sk->sk_userlocks |= SOCK_BINDPORT_LOCK;
E
Eric Dumazet 已提交
554 555 556
	inet->inet_sport = htons(inet->inet_num);
	inet->inet_daddr = 0;
	inet->inet_dport = 0;
L
Linus Torvalds 已提交
557 558 559 560 561 562 563
	sk_dst_reset(sk);
	err = 0;
out_release_sock:
	release_sock(sk);
out:
	return err;
}
E
Eric Dumazet 已提交
564
EXPORT_SYMBOL(inet_bind);
L
Linus Torvalds 已提交
565

D
Daniel Baluta 已提交
566
int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr,
L
Linus Torvalds 已提交
567 568 569 570
		       int addr_len, int flags)
{
	struct sock *sk = sock->sk;

571 572
	if (addr_len < sizeof(uaddr->sa_family))
		return -EINVAL;
L
Linus Torvalds 已提交
573 574 575
	if (uaddr->sa_family == AF_UNSPEC)
		return sk->sk_prot->disconnect(sk, flags);

E
Eric Dumazet 已提交
576
	if (!inet_sk(sk)->inet_num && inet_autobind(sk))
L
Linus Torvalds 已提交
577
		return -EAGAIN;
J
Joe Perches 已提交
578
	return sk->sk_prot->connect(sk, uaddr, addr_len);
L
Linus Torvalds 已提交
579
}
E
Eric Dumazet 已提交
580
EXPORT_SYMBOL(inet_dgram_connect);
L
Linus Torvalds 已提交
581

582
static long inet_wait_for_connect(struct sock *sk, long timeo, int writebias)
L
Linus Torvalds 已提交
583 584 585
{
	DEFINE_WAIT(wait);

E
Eric Dumazet 已提交
586
	prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
587
	sk->sk_write_pending += writebias;
L
Linus Torvalds 已提交
588 589 590 591 592 593 594 595 596 597 598 599

	/* 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
600
		prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
L
Linus Torvalds 已提交
601
	}
E
Eric Dumazet 已提交
602
	finish_wait(sk_sleep(sk), &wait);
603
	sk->sk_write_pending -= writebias;
L
Linus Torvalds 已提交
604 605 606 607 608 609 610
	return timeo;
}

/*
 *	Connect to a remote host. There is regrettably still a little
 *	TCP 'magic' in here.
 */
611 612
int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr,
			  int addr_len, int flags)
L
Linus Torvalds 已提交
613 614 615 616 617
{
	struct sock *sk = sock->sk;
	int err;
	long timeo;

618 619 620
	if (addr_len < sizeof(uaddr->sa_family))
		return -EINVAL;

L
Linus Torvalds 已提交
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
	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;

647
		sock->state = SS_CONNECTING;
L
Linus Torvalds 已提交
648 649 650 651 652 653 654 655 656 657 658 659

		/* 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)) {
660 661 662 663
		int writebias = (sk->sk_protocol == IPPROTO_TCP) &&
				tcp_sk(sk)->fastopen_req &&
				tcp_sk(sk)->fastopen_req->data ? 1 : 0;

L
Linus Torvalds 已提交
664
		/* Error code is set above */
665
		if (!timeo || !inet_wait_for_connect(sk, timeo, writebias))
L
Linus Torvalds 已提交
666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695
			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:
	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;
}
696 697 698 699 700 701 702 703 704 705 706 707
EXPORT_SYMBOL(__inet_stream_connect);

int inet_stream_connect(struct socket *sock, struct sockaddr *uaddr,
			int addr_len, int flags)
{
	int err;

	lock_sock(sock->sk);
	err = __inet_stream_connect(sock, uaddr, addr_len, flags);
	release_sock(sock->sk);
	return err;
}
E
Eric Dumazet 已提交
708
EXPORT_SYMBOL(inet_stream_connect);
L
Linus Torvalds 已提交
709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724

/*
 *	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);

725
	sock_rps_record_flow(sk2);
726
	WARN_ON(!((1 << sk2->sk_state) &
727 728
		  (TCPF_ESTABLISHED | TCPF_SYN_RECV |
		  TCPF_CLOSE_WAIT | TCPF_CLOSE)));
L
Linus Torvalds 已提交
729 730 731 732 733 734 735 736 737

	sock_graft(sk2, newsock);

	newsock->state = SS_CONNECTED;
	err = 0;
	release_sock(sk2);
do_err:
	return err;
}
E
Eric Dumazet 已提交
738
EXPORT_SYMBOL(inet_accept);
L
Linus Torvalds 已提交
739 740 741 742 743 744 745 746 747 748


/*
 *	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);
749
	DECLARE_SOCKADDR(struct sockaddr_in *, sin, uaddr);
L
Linus Torvalds 已提交
750 751 752

	sin->sin_family = AF_INET;
	if (peer) {
E
Eric Dumazet 已提交
753
		if (!inet->inet_dport ||
L
Linus Torvalds 已提交
754 755 756
		    (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_SYN_SENT)) &&
		     peer == 1))
			return -ENOTCONN;
E
Eric Dumazet 已提交
757 758
		sin->sin_port = inet->inet_dport;
		sin->sin_addr.s_addr = inet->inet_daddr;
L
Linus Torvalds 已提交
759
	} else {
E
Eric Dumazet 已提交
760
		__be32 addr = inet->inet_rcv_saddr;
L
Linus Torvalds 已提交
761
		if (!addr)
E
Eric Dumazet 已提交
762 763
			addr = inet->inet_saddr;
		sin->sin_port = inet->inet_sport;
L
Linus Torvalds 已提交
764 765 766 767 768 769
		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
770
EXPORT_SYMBOL(inet_getname);
L
Linus Torvalds 已提交
771 772 773 774 775 776

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

777
	sock_rps_record_flow(sk);
T
Tom Herbert 已提交
778

L
Linus Torvalds 已提交
779
	/* We may need to bind the socket. */
780 781
	if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
	    inet_autobind(sk))
L
Linus Torvalds 已提交
782 783 784 785
		return -EAGAIN;

	return sk->sk_prot->sendmsg(iocb, sk, msg, size);
}
E
Eric Dumazet 已提交
786
EXPORT_SYMBOL(inet_sendmsg);
L
Linus Torvalds 已提交
787

788 789
ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
		      size_t size, int flags)
L
Linus Torvalds 已提交
790 791 792
{
	struct sock *sk = sock->sk;

793
	sock_rps_record_flow(sk);
T
Tom Herbert 已提交
794

L
Linus Torvalds 已提交
795
	/* We may need to bind the socket. */
796 797
	if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
	    inet_autobind(sk))
L
Linus Torvalds 已提交
798 799 800 801 802 803
		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);
}
804
EXPORT_SYMBOL(inet_sendpage);
L
Linus Torvalds 已提交
805

T
Tom Herbert 已提交
806 807 808 809 810 811 812
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;

813
	sock_rps_record_flow(sk);
T
Tom Herbert 已提交
814 815 816 817 818 819 820 821

	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
822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875

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
876
EXPORT_SYMBOL(inet_shutdown);
L
Linus Torvalds 已提交
877 878 879 880 881 882 883 884 885 886 887 888 889 890 891

/*
 *	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;
892
	struct net *net = sock_net(sk);
L
Linus Torvalds 已提交
893 894

	switch (cmd) {
E
Eric Dumazet 已提交
895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929
	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
930 931 932
	}
	return err;
}
E
Eric Dumazet 已提交
933
EXPORT_SYMBOL(inet_ioctl);
L
Linus Torvalds 已提交
934

935
#ifdef CONFIG_COMPAT
936
static int inet_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
937 938 939 940 941 942 943 944 945 946 947
{
	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

948
const struct proto_ops inet_stream_ops = {
949 950 951 952 953 954 955 956 957 958 959 960 961 962
	.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,
963
	.sendmsg	   = inet_sendmsg,
T
Tom Herbert 已提交
964
	.recvmsg	   = inet_recvmsg,
965
	.mmap		   = sock_no_mmap,
966
	.sendpage	   = inet_sendpage,
J
Jens Axboe 已提交
967
	.splice_read	   = tcp_splice_read,
968
#ifdef CONFIG_COMPAT
969 970
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
971
	.compat_ioctl	   = inet_compat_ioctl,
972
#endif
L
Linus Torvalds 已提交
973
};
E
Eric Dumazet 已提交
974
EXPORT_SYMBOL(inet_stream_ops);
L
Linus Torvalds 已提交
975

976
const struct proto_ops inet_dgram_ops = {
977 978 979 980 981 982 983 984 985 986 987 988 989 990 991
	.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
992
	.recvmsg	   = inet_recvmsg,
993 994
	.mmap		   = sock_no_mmap,
	.sendpage	   = inet_sendpage,
995
#ifdef CONFIG_COMPAT
996 997
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
998
	.compat_ioctl	   = inet_compat_ioctl,
999
#endif
L
Linus Torvalds 已提交
1000
};
E
Eric Dumazet 已提交
1001
EXPORT_SYMBOL(inet_dgram_ops);
L
Linus Torvalds 已提交
1002 1003 1004 1005 1006

/*
 * For SOCK_RAW sockets; should be the same as inet_dgram_ops but without
 * udp_poll
 */
1007
static const struct proto_ops inet_sockraw_ops = {
1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
	.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
1023
	.recvmsg	   = inet_recvmsg,
1024 1025
	.mmap		   = sock_no_mmap,
	.sendpage	   = inet_sendpage,
1026
#ifdef CONFIG_COMPAT
1027 1028
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
1029
	.compat_ioctl	   = inet_compat_ioctl,
1030
#endif
L
Linus Torvalds 已提交
1031 1032
};

1033
static const struct net_proto_family inet_family_ops = {
L
Linus Torvalds 已提交
1034 1035 1036 1037 1038 1039 1040 1041 1042 1043
	.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[] =
{
1044 1045 1046 1047 1048 1049 1050
	{
		.type =       SOCK_STREAM,
		.protocol =   IPPROTO_TCP,
		.prot =       &tcp_prot,
		.ops =        &inet_stream_ops,
		.no_check =   0,
		.flags =      INET_PROTOSW_PERMANENT |
1051
			      INET_PROTOSW_ICSK,
1052 1053 1054 1055 1056 1057 1058 1059 1060
	},

	{
		.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
1061
       },
1062

1063 1064 1065 1066 1067 1068 1069 1070
       {
		.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
1071 1072

       {
1073 1074 1075 1076 1077 1078
	       .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
1079 1080 1081
       }
};

1082
#define INETSW_ARRAY_LEN ARRAY_SIZE(inetsw_array)
L
Linus Torvalds 已提交
1083 1084 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

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
1117
	 * non-permanent entry.  This means that when we remove this entry, the
L
Linus Torvalds 已提交
1118 1119 1120 1121 1122 1123 1124 1125 1126
	 * system automatically returns to the old behavior.
	 */
	list_add_rcu(&p->list, last_perm);
out:
	spin_unlock_bh(&inetsw_lock);

	return;

out_permanent:
J
Joe Perches 已提交
1127
	pr_err("Attempt to override permanent protocol %d\n", protocol);
L
Linus Torvalds 已提交
1128 1129 1130
	goto out;

out_illegal:
J
Joe Perches 已提交
1131
	pr_err("Ignoring attempt to register invalid socket type %d\n",
L
Linus Torvalds 已提交
1132 1133 1134
	       p->type);
	goto out;
}
E
Eric Dumazet 已提交
1135
EXPORT_SYMBOL(inet_register_protosw);
L
Linus Torvalds 已提交
1136 1137 1138 1139

void inet_unregister_protosw(struct inet_protosw *p)
{
	if (INET_PROTOSW_PERMANENT & p->flags) {
J
Joe Perches 已提交
1140
		pr_err("Attempt to unregister permanent protocol %d\n",
L
Linus Torvalds 已提交
1141 1142 1143 1144 1145 1146 1147 1148 1149
		       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
1150
EXPORT_SYMBOL(inet_unregister_protosw);
L
Linus Torvalds 已提交
1151

1152 1153 1154 1155
/*
 *      Shall we try to damage output packets if routing dev changes?
 */

1156
int sysctl_ip_dynaddr __read_mostly;
1157 1158 1159 1160

static int inet_sk_reselect_saddr(struct sock *sk)
{
	struct inet_sock *inet = inet_sk(sk);
E
Eric Dumazet 已提交
1161 1162
	__be32 old_saddr = inet->inet_saddr;
	__be32 daddr = inet->inet_daddr;
1163
	struct flowi4 *fl4;
1164 1165
	struct rtable *rt;
	__be32 new_saddr;
1166
	struct ip_options_rcu *inet_opt;
1167

1168 1169 1170 1171
	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;
1172 1173

	/* Query new route. */
1174 1175
	fl4 = &inet->cork.fl.u.ip4;
	rt = ip_route_connect(fl4, daddr, 0, RT_CONN_FLAGS(sk),
1176 1177 1178 1179
			      sk->sk_bound_dev_if, sk->sk_protocol,
			      inet->inet_sport, inet->inet_dport, sk, false);
	if (IS_ERR(rt))
		return PTR_ERR(rt);
1180

1181
	sk_setup_caps(sk, &rt->dst);
1182

1183
	new_saddr = fl4->saddr;
1184 1185 1186 1187 1188

	if (new_saddr == old_saddr)
		return 0;

	if (sysctl_ip_dynaddr > 1) {
J
Joe Perches 已提交
1189 1190
		pr_info("%s(): shifting inet->saddr from %pI4 to %pI4\n",
			__func__, &old_saddr, &new_saddr);
1191 1192
	}

E
Eric Dumazet 已提交
1193
	inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210

	/*
	 * 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
1211
	__be32 daddr;
1212
	struct ip_options_rcu *inet_opt;
1213
	struct flowi4 *fl4;
1214 1215 1216 1217 1218 1219 1220
	int err;

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

	/* Reroute. */
1221 1222
	rcu_read_lock();
	inet_opt = rcu_dereference(inet->inet_opt);
E
Eric Dumazet 已提交
1223
	daddr = inet->inet_daddr;
1224 1225 1226
	if (inet_opt && inet_opt->opt.srr)
		daddr = inet_opt->opt.faddr;
	rcu_read_unlock();
1227 1228
	fl4 = &inet->cork.fl.u.ip4;
	rt = ip_route_output_ports(sock_net(sk), fl4, sk, daddr, inet->inet_saddr,
1229 1230 1231
				   inet->inet_dport, inet->inet_sport,
				   sk->sk_protocol, RT_CONN_FLAGS(sk),
				   sk->sk_bound_dev_if);
1232 1233
	if (!IS_ERR(rt)) {
		err = 0;
1234
		sk_setup_caps(sk, &rt->dst);
1235 1236 1237
	} else {
		err = PTR_ERR(rt);

1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254
		/* 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);

1255 1256
static int inet_gso_send_check(struct sk_buff *skb)
{
1257
	const struct net_offload *ops;
1258
	const struct iphdr *iph;
1259 1260 1261 1262 1263 1264 1265
	int proto;
	int ihl;
	int err = -EINVAL;

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

1266
	iph = ip_hdr(skb);
1267 1268 1269 1270 1271 1272 1273
	ihl = iph->ihl * 4;
	if (ihl < sizeof(*iph))
		goto out;

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

1274 1275
	__skb_pull(skb, ihl);
	skb_reset_transport_header(skb);
1276
	iph = ip_hdr(skb);
1277
	proto = iph->protocol;
1278 1279 1280
	err = -EPROTONOSUPPORT;

	rcu_read_lock();
1281
	ops = rcu_dereference(inet_offloads[proto]);
1282 1283
	if (likely(ops && ops->callbacks.gso_send_check))
		err = ops->callbacks.gso_send_check(skb);
1284 1285 1286 1287 1288 1289
	rcu_read_unlock();

out:
	return err;
}

1290 1291
static struct sk_buff *inet_gso_segment(struct sk_buff *skb,
	netdev_features_t features)
H
Herbert Xu 已提交
1292 1293
{
	struct sk_buff *segs = ERR_PTR(-EINVAL);
1294
	const struct net_offload *ops;
1295
	struct iphdr *iph;
H
Herbert Xu 已提交
1296 1297 1298
	int proto;
	int ihl;
	int id;
1299
	unsigned int offset = 0;
H
Herbert Xu 已提交
1300

1301 1302 1303
	if (!(features & NETIF_F_V4_CSUM))
		features &= ~NETIF_F_SG;

1304 1305 1306 1307 1308 1309 1310 1311 1312
	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
1313 1314
		goto out;

1315
	iph = ip_hdr(skb);
H
Herbert Xu 已提交
1316 1317 1318 1319
	ihl = iph->ihl * 4;
	if (ihl < sizeof(*iph))
		goto out;

1320
	if (unlikely(!pskb_may_pull(skb, ihl)))
H
Herbert Xu 已提交
1321 1322
		goto out;

1323 1324
	__skb_pull(skb, ihl);
	skb_reset_transport_header(skb);
1325
	iph = ip_hdr(skb);
H
Herbert Xu 已提交
1326
	id = ntohs(iph->id);
1327
	proto = iph->protocol;
H
Herbert Xu 已提交
1328 1329 1330
	segs = ERR_PTR(-EPROTONOSUPPORT);

	rcu_read_lock();
1331
	ops = rcu_dereference(inet_offloads[proto]);
1332 1333
	if (likely(ops && ops->callbacks.gso_segment))
		segs = ops->callbacks.gso_segment(skb, features);
H
Herbert Xu 已提交
1334 1335
	rcu_read_unlock();

1336
	if (!segs || IS_ERR(segs))
H
Herbert Xu 已提交
1337 1338 1339 1340
		goto out;

	skb = segs;
	do {
1341
		iph = ip_hdr(skb);
1342 1343 1344 1345 1346 1347 1348 1349
		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
1350 1351
		iph->tot_len = htons(skb->len - skb->mac_len);
		iph->check = 0;
1352
		iph->check = ip_fast_csum(skb_network_header(skb), iph->ihl);
H
Herbert Xu 已提交
1353 1354 1355 1356 1357 1358
	} while ((skb = skb->next));

out:
	return segs;
}

H
Herbert Xu 已提交
1359 1360 1361
static struct sk_buff **inet_gro_receive(struct sk_buff **head,
					 struct sk_buff *skb)
{
1362
	const struct net_offload *ops;
H
Herbert Xu 已提交
1363 1364
	struct sk_buff **pp = NULL;
	struct sk_buff *p;
1365
	const struct iphdr *iph;
1366 1367
	unsigned int hlen;
	unsigned int off;
1368
	unsigned int id;
H
Herbert Xu 已提交
1369 1370 1371
	int flush = 1;
	int proto;

1372 1373 1374 1375 1376 1377 1378 1379
	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
1380

1381
	proto = iph->protocol;
H
Herbert Xu 已提交
1382 1383

	rcu_read_lock();
1384
	ops = rcu_dereference(inet_offloads[proto]);
1385
	if (!ops || !ops->callbacks.gro_receive)
H
Herbert Xu 已提交
1386 1387
		goto out_unlock;

H
Herbert Xu 已提交
1388
	if (*(u8 *)iph != 0x45)
H
Herbert Xu 已提交
1389 1390
		goto out_unlock;

1391
	if (unlikely(ip_fast_csum((u8 *)iph, 5)))
H
Herbert Xu 已提交
1392 1393
		goto out_unlock;

1394 1395
	id = ntohl(*(__be32 *)&iph->id);
	flush = (u16)((ntohl(*(__be32 *)iph) ^ skb_gro_len(skb)) | (id ^ IP_DF));
1396
	id >>= 16;
H
Herbert Xu 已提交
1397 1398 1399 1400 1401 1402 1403 1404 1405

	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
1406
		if ((iph->protocol ^ iph2->protocol) |
1407 1408
		    ((__force u32)iph->saddr ^ (__force u32)iph2->saddr) |
		    ((__force u32)iph->daddr ^ (__force u32)iph2->daddr)) {
H
Herbert Xu 已提交
1409 1410 1411 1412 1413 1414
			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
1415
			(iph->ttl ^ iph2->ttl) |
1416
			(iph->tos ^ iph2->tos) |
H
Herbert Xu 已提交
1417
			((u16)(ntohs(iph2->id) + NAPI_GRO_CB(p)->count) ^ id);
H
Herbert Xu 已提交
1418 1419 1420 1421 1422

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

	NAPI_GRO_CB(skb)->flush |= flush;
1423 1424
	skb_gro_pull(skb, sizeof(*iph));
	skb_set_transport_header(skb, skb_gro_offset(skb));
H
Herbert Xu 已提交
1425

1426
	pp = ops->callbacks.gro_receive(head, skb);
H
Herbert Xu 已提交
1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438

out_unlock:
	rcu_read_unlock();

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

	return pp;
}

static int inet_gro_complete(struct sk_buff *skb)
{
1439
	__be16 newlen = htons(skb->len - skb_network_offset(skb));
H
Herbert Xu 已提交
1440
	struct iphdr *iph = ip_hdr(skb);
1441
	const struct net_offload *ops;
1442
	int proto = iph->protocol;
H
Herbert Xu 已提交
1443 1444 1445 1446 1447 1448
	int err = -ENOSYS;

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

	rcu_read_lock();
1449
	ops = rcu_dereference(inet_offloads[proto]);
1450
	if (WARN_ON(!ops || !ops->callbacks.gro_complete))
H
Herbert Xu 已提交
1451 1452
		goto out_unlock;

1453
	err = ops->callbacks.gro_complete(skb);
H
Herbert Xu 已提交
1454 1455 1456 1457 1458 1459 1460

out_unlock:
	rcu_read_unlock();

	return err;
}

1461
int inet_ctl_sock_create(struct sock **sk, unsigned short family,
1462 1463
			 unsigned short type, unsigned char protocol,
			 struct net *net)
1464
{
1465 1466
	struct socket *sock;
	int rc = sock_create_kern(family, type, protocol, &sock);
1467 1468

	if (rc == 0) {
1469 1470
		*sk = sock->sk;
		(*sk)->sk_allocation = GFP_ATOMIC;
1471 1472 1473 1474
		/*
		 * Unhash it so that IP input processing does not even see it,
		 * we do not wish this socket to see incoming packets.
		 */
1475
		(*sk)->sk_prot->unhash(*sk);
1476 1477

		sk_change_net(*sk, net);
1478 1479 1480 1481 1482
	}
	return rc;
}
EXPORT_SYMBOL_GPL(inet_ctl_sock_create);

1483
unsigned long snmp_fold_field(void __percpu *mib[], int offt)
1484 1485
{
	unsigned long res = 0;
E
Eric Dumazet 已提交
1486
	int i, j;
1487 1488

	for_each_possible_cpu(i) {
E
Eric Dumazet 已提交
1489 1490
		for (j = 0; j < SNMP_ARRAY_SZ; j++)
			res += *(((unsigned long *) per_cpu_ptr(mib[j], i)) + offt);
1491 1492 1493 1494 1495
	}
	return res;
}
EXPORT_SYMBOL_GPL(snmp_fold_field);

1496 1497 1498 1499 1500 1501 1502 1503
#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
1504
		void *bhptr;
1505
		struct u64_stats_sync *syncp;
E
Eric Dumazet 已提交
1506
		u64 v;
1507 1508
		unsigned int start;

E
Eric Dumazet 已提交
1509
		bhptr = per_cpu_ptr(mib[0], cpu);
1510 1511 1512
		syncp = (struct u64_stats_sync *)(bhptr + syncp_offset);
		do {
			start = u64_stats_fetch_begin_bh(syncp);
E
Eric Dumazet 已提交
1513
			v = *(((u64 *) bhptr) + offt);
1514 1515
		} while (u64_stats_fetch_retry_bh(syncp, start));

E
Eric Dumazet 已提交
1516
		res += v;
1517 1518 1519 1520 1521 1522
	}
	return res;
}
EXPORT_SYMBOL_GPL(snmp_fold_field64);
#endif

1523
int snmp_mib_init(void __percpu *ptr[2], size_t mibsize, size_t align)
1524 1525
{
	BUG_ON(ptr == NULL);
1526
	ptr[0] = __alloc_percpu(mibsize, align);
1527
	if (!ptr[0])
E
Eric Dumazet 已提交
1528 1529
		return -ENOMEM;
#if SNMP_ARRAY_SZ == 2
1530
	ptr[1] = __alloc_percpu(mibsize, align);
E
Eric Dumazet 已提交
1531 1532 1533 1534 1535 1536
	if (!ptr[1]) {
		free_percpu(ptr[0]);
		ptr[0] = NULL;
		return -ENOMEM;
	}
#endif
1537 1538 1539 1540
	return 0;
}
EXPORT_SYMBOL_GPL(snmp_mib_init);

E
Eric Dumazet 已提交
1541
void snmp_mib_free(void __percpu *ptr[SNMP_ARRAY_SZ])
1542
{
E
Eric Dumazet 已提交
1543 1544
	int i;

1545
	BUG_ON(ptr == NULL);
E
Eric Dumazet 已提交
1546 1547 1548 1549
	for (i = 0; i < SNMP_ARRAY_SZ; i++) {
		free_percpu(ptr[i]);
		ptr[i] = NULL;
	}
1550 1551 1552
}
EXPORT_SYMBOL_GPL(snmp_mib_free);

L
Linus Torvalds 已提交
1553
#ifdef CONFIG_IP_MULTICAST
1554
static const struct net_protocol igmp_protocol = {
L
Linus Torvalds 已提交
1555
	.handler =	igmp_rcv,
1556
	.netns_ok =	1,
L
Linus Torvalds 已提交
1557 1558 1559
};
#endif

1560
static const struct net_protocol tcp_protocol = {
D
David S. Miller 已提交
1561 1562 1563 1564 1565
	.early_demux	=	tcp_v4_early_demux,
	.handler	=	tcp_v4_rcv,
	.err_handler	=	tcp_v4_err,
	.no_policy	=	1,
	.netns_ok	=	1,
L
Linus Torvalds 已提交
1566 1567
};

1568
static const struct net_offload tcp_offload = {
1569 1570 1571 1572 1573 1574
	.callbacks = {
		.gso_send_check	=	tcp_v4_gso_send_check,
		.gso_segment	=	tcp_tso_segment,
		.gro_receive	=	tcp4_gro_receive,
		.gro_complete	=	tcp4_gro_complete,
	},
1575 1576
};

1577
static const struct net_protocol udp_protocol = {
L
Linus Torvalds 已提交
1578 1579 1580
	.handler =	udp_rcv,
	.err_handler =	udp_err,
	.no_policy =	1,
1581
	.netns_ok =	1,
L
Linus Torvalds 已提交
1582 1583
};

1584
static const struct net_offload udp_offload = {
1585 1586 1587 1588
	.callbacks = {
		.gso_send_check = udp4_ufo_send_check,
		.gso_segment = udp4_ufo_fragment,
	},
1589 1590
};

1591
static const struct net_protocol icmp_protocol = {
L
Linus Torvalds 已提交
1592
	.handler =	icmp_rcv,
1593
	.err_handler =	ping_err,
1594
	.no_policy =	1,
1595
	.netns_ok =	1,
L
Linus Torvalds 已提交
1596 1597
};

1598 1599
static __net_init int ipv4_mib_init_net(struct net *net)
{
1600
	if (snmp_mib_init((void __percpu **)net->mib.tcp_statistics,
1601 1602
			  sizeof(struct tcp_mib),
			  __alignof__(struct tcp_mib)) < 0)
1603
		goto err_tcp_mib;
1604
	if (snmp_mib_init((void __percpu **)net->mib.ip_statistics,
1605 1606
			  sizeof(struct ipstats_mib),
			  __alignof__(struct ipstats_mib)) < 0)
1607
		goto err_ip_mib;
1608
	if (snmp_mib_init((void __percpu **)net->mib.net_statistics,
1609 1610
			  sizeof(struct linux_mib),
			  __alignof__(struct linux_mib)) < 0)
1611
		goto err_net_mib;
1612
	if (snmp_mib_init((void __percpu **)net->mib.udp_statistics,
1613 1614
			  sizeof(struct udp_mib),
			  __alignof__(struct udp_mib)) < 0)
1615
		goto err_udp_mib;
1616
	if (snmp_mib_init((void __percpu **)net->mib.udplite_statistics,
1617 1618
			  sizeof(struct udp_mib),
			  __alignof__(struct udp_mib)) < 0)
1619
		goto err_udplite_mib;
1620
	if (snmp_mib_init((void __percpu **)net->mib.icmp_statistics,
1621 1622
			  sizeof(struct icmp_mib),
			  __alignof__(struct icmp_mib)) < 0)
1623
		goto err_icmp_mib;
1624 1625 1626
	net->mib.icmpmsg_statistics = kzalloc(sizeof(struct icmpmsg_mib),
					      GFP_KERNEL);
	if (!net->mib.icmpmsg_statistics)
1627
		goto err_icmpmsg_mib;
1628 1629

	tcp_mib_init(net);
1630
	return 0;
1631

1632
err_icmpmsg_mib:
1633
	snmp_mib_free((void __percpu **)net->mib.icmp_statistics);
1634
err_icmp_mib:
1635
	snmp_mib_free((void __percpu **)net->mib.udplite_statistics);
1636
err_udplite_mib:
1637
	snmp_mib_free((void __percpu **)net->mib.udp_statistics);
1638
err_udp_mib:
1639
	snmp_mib_free((void __percpu **)net->mib.net_statistics);
1640
err_net_mib:
1641
	snmp_mib_free((void __percpu **)net->mib.ip_statistics);
1642
err_ip_mib:
1643
	snmp_mib_free((void __percpu **)net->mib.tcp_statistics);
1644 1645
err_tcp_mib:
	return -ENOMEM;
1646 1647 1648 1649
}

static __net_exit void ipv4_mib_exit_net(struct net *net)
{
1650
	kfree(net->mib.icmpmsg_statistics);
1651 1652 1653 1654 1655 1656
	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);
1657 1658 1659 1660 1661 1662 1663
}

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
1664 1665
static int __init init_ipv4_mibs(void)
{
1666
	return register_pernet_subsys(&ipv4_mib_ops);
L
Linus Torvalds 已提交
1667 1668 1669 1670
}

static int ipv4_proc_init(void);

1671 1672 1673 1674
/*
 *	IP protocol layer initialiser
 */

1675 1676
static struct packet_offload ip_packet_offload __read_mostly = {
	.type = cpu_to_be16(ETH_P_IP),
1677 1678 1679 1680 1681 1682
	.callbacks = {
		.gso_send_check = inet_gso_send_check,
		.gso_segment = inet_gso_segment,
		.gro_receive = inet_gro_receive,
		.gro_complete = inet_gro_complete,
	},
1683 1684
};

1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705
static int __init ipv4_offload_init(void)
{
	/*
	 * Add offloads
	 */
	if (inet_add_offload(&udp_offload, IPPROTO_UDP) < 0)
		pr_crit("%s: Cannot add UDP protocol offload\n", __func__);
	if (inet_add_offload(&tcp_offload, IPPROTO_TCP) < 0)
		pr_crit("%s: Cannot add TCP protocol offlaod\n", __func__);

	dev_add_offload(&ip_packet_offload);
	return 0;
}

fs_initcall(ipv4_offload_init);

static struct packet_type ip_packet_type __read_mostly = {
	.type = cpu_to_be16(ETH_P_IP),
	.func = ip_rcv,
};

L
Linus Torvalds 已提交
1706 1707 1708 1709 1710 1711
static int __init inet_init(void)
{
	struct inet_protosw *q;
	struct list_head *r;
	int rc = -EINVAL;

1712
	BUILD_BUG_ON(sizeof(struct inet_skb_parm) > FIELD_SIZEOF(struct sk_buff, cb));
L
Linus Torvalds 已提交
1713

1714 1715 1716 1717
	sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL);
	if (!sysctl_local_reserved_ports)
		goto out;

L
Linus Torvalds 已提交
1718 1719
	rc = proto_register(&tcp_prot, 1);
	if (rc)
1720
		goto out_free_reserved_ports;
L
Linus Torvalds 已提交
1721 1722 1723 1724 1725 1726 1727 1728 1729

	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;

1730 1731 1732 1733
	rc = proto_register(&ping_prot, 1);
	if (rc)
		goto out_unregister_raw_proto;

L
Linus Torvalds 已提交
1734
	/*
1735
	 *	Tell SOCKET that we are alive...
L
Linus Torvalds 已提交
1736 1737
	 */

1738
	(void)sock_register(&inet_family_ops);
L
Linus Torvalds 已提交
1739

1740 1741 1742 1743
#ifdef CONFIG_SYSCTL
	ip_static_sysctl_init();
#endif

G
Glauber Costa 已提交
1744 1745
	tcp_prot.sysctl_mem = init_net.ipv4.sysctl_tcp_mem;

L
Linus Torvalds 已提交
1746 1747 1748 1749 1750
	/*
	 *	Add all the base protocols.
	 */

	if (inet_add_protocol(&icmp_protocol, IPPROTO_ICMP) < 0)
J
Joe Perches 已提交
1751
		pr_crit("%s: Cannot add ICMP protocol\n", __func__);
L
Linus Torvalds 已提交
1752
	if (inet_add_protocol(&udp_protocol, IPPROTO_UDP) < 0)
J
Joe Perches 已提交
1753
		pr_crit("%s: Cannot add UDP protocol\n", __func__);
L
Linus Torvalds 已提交
1754
	if (inet_add_protocol(&tcp_protocol, IPPROTO_TCP) < 0)
J
Joe Perches 已提交
1755
		pr_crit("%s: Cannot add TCP protocol\n", __func__);
L
Linus Torvalds 已提交
1756 1757
#ifdef CONFIG_IP_MULTICAST
	if (inet_add_protocol(&igmp_protocol, IPPROTO_IGMP) < 0)
J
Joe Perches 已提交
1758
		pr_crit("%s: Cannot add IGMP protocol\n", __func__);
L
Linus Torvalds 已提交
1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773
#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();

1774 1775 1776
	/*
	 *	Set the IP module up
	 */
L
Linus Torvalds 已提交
1777 1778 1779

	ip_init();

1780
	tcp_v4_init();
L
Linus Torvalds 已提交
1781 1782 1783 1784

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

H
Hideo Aoki 已提交
1785 1786 1787
	/* Setup UDP memory threshold */
	udp_init();

1788 1789
	/* Add UDP-Lite (RFC 3828) */
	udplite4_register();
L
Linus Torvalds 已提交
1790

1791 1792
	ping_init();

L
Linus Torvalds 已提交
1793 1794 1795 1796
	/*
	 *	Set the ICMP layer up
	 */

1797 1798
	if (icmp_init() < 0)
		panic("Failed to create the ICMP control socket.\n");
L
Linus Torvalds 已提交
1799 1800 1801 1802 1803

	/*
	 *	Initialise the multicast router
	 */
#if defined(CONFIG_IP_MROUTE)
W
Wang Chen 已提交
1804
	if (ip_mr_init())
J
Joe Perches 已提交
1805
		pr_crit("%s: Cannot init ipv4 mroute\n", __func__);
L
Linus Torvalds 已提交
1806 1807 1808
#endif
	/*
	 *	Initialise per-cpu ipv4 mibs
1809
	 */
L
Linus Torvalds 已提交
1810

S
Stephen Hemminger 已提交
1811
	if (init_ipv4_mibs())
J
Joe Perches 已提交
1812
		pr_crit("%s: Cannot init ipv4 mibs\n", __func__);
1813

L
Linus Torvalds 已提交
1814 1815 1816 1817
	ipv4_proc_init();

	ipfrag_init();

1818 1819
	dev_add_pack(&ip_packet_type);

L
Linus Torvalds 已提交
1820 1821 1822
	rc = 0;
out:
	return rc;
1823 1824
out_unregister_raw_proto:
	proto_unregister(&raw_prot);
L
Linus Torvalds 已提交
1825 1826
out_unregister_udp_proto:
	proto_unregister(&udp_prot);
1827 1828
out_unregister_tcp_proto:
	proto_unregister(&tcp_prot);
1829 1830
out_free_reserved_ports:
	kfree(sysctl_local_reserved_ports);
L
Linus Torvalds 已提交
1831 1832 1833
	goto out;
}

1834
fs_initcall(inet_init);
L
Linus Torvalds 已提交
1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848

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

#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;
1849 1850
	if (ping_proc_init())
		goto out_ping;
L
Linus Torvalds 已提交
1851 1852 1853 1854 1855
	if (ip_misc_proc_init())
		goto out_misc;
out:
	return rc;
out_misc:
1856 1857
	ping_proc_exit();
out_ping:
L
Linus Torvalds 已提交
1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876
	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);