af_inet6.c 25.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2
/*
 *	PF_INET6 socket protocol family
3
 *	Linux INET6 implementation
L
Linus Torvalds 已提交
4 5
 *
 *	Authors:
6
 *	Pedro Roque		<roque@di.fc.ul.pt>
L
Linus Torvalds 已提交
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
 *
 *	Adapted from linux/net/ipv4/af_inet.c
 *
 *	$Id: af_inet6.c,v 1.66 2002/02/01 22:01:04 davem Exp $
 *
 * 	Fixes:
 *	piggy, Karl Knutson	:	Socket protocol table
 * 	Hideaki YOSHIFUJI	:	sin6_scope_id support
 * 	Arnaldo Melo		: 	check proc_net_create return, cleanups
 *
 *	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.
 */


#include <linux/module.h>
25
#include <linux/capability.h>
L
Linus Torvalds 已提交
26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/in.h>
#include <linux/kernel.h>
#include <linux/timer.h>
#include <linux/string.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/fcntl.h>
#include <linux/mm.h>
#include <linux/interrupt.h>
#include <linux/proc_fs.h>
#include <linux/stat.h>
#include <linux/init.h>

#include <linux/inet.h>
#include <linux/netdevice.h>
#include <linux/icmpv6.h>
45
#include <linux/netfilter_ipv6.h>
L
Linus Torvalds 已提交
46 47 48 49

#include <net/ip.h>
#include <net/ipv6.h>
#include <net/udp.h>
50
#include <net/udplite.h>
L
Linus Torvalds 已提交
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
#include <net/tcp.h>
#include <net/ipip.h>
#include <net/protocol.h>
#include <net/inet_common.h>
#include <net/transp_v6.h>
#include <net/ip6_route.h>
#include <net/addrconf.h>
#ifdef CONFIG_IPV6_TUNNEL
#include <net/ip6_tunnel.h>
#endif

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

MODULE_AUTHOR("Cast of dozens");
MODULE_DESCRIPTION("IPv6 protocol stack for Linux");
MODULE_LICENSE("GPL");

69
/* The inetsw6 table contains everything that inet6_create needs to
L
Linus Torvalds 已提交
70 71 72 73 74 75 76 77 78 79 80 81
 * build a new socket.
 */
static struct list_head inetsw6[SOCK_MAX];
static DEFINE_SPINLOCK(inetsw6_lock);

static __inline__ struct ipv6_pinfo *inet6_sk_generic(struct sock *sk)
{
	const int offset = sk->sk_prot->obj_size - sizeof(struct ipv6_pinfo);

	return (struct ipv6_pinfo *)(((u8 *)sk) + offset);
}

82
static int inet6_create(struct net *net, struct socket *sock, int protocol)
L
Linus Torvalds 已提交
83 84 85 86 87 88 89 90 91
{
	struct inet_sock *inet;
	struct ipv6_pinfo *np;
	struct sock *sk;
	struct list_head *p;
	struct inet_protosw *answer;
	struct proto *answer_prot;
	unsigned char answer_flags;
	char answer_no_check;
92 93
	int try_loading_module = 0;
	int err;
L
Linus Torvalds 已提交
94

95 96 97 98 99
	if (sock->type != SOCK_RAW &&
	    sock->type != SOCK_DGRAM &&
	    !inet_ehash_secret)
		build_ehash_secret();

L
Linus Torvalds 已提交
100 101
	/* Look for the requested type/protocol pair. */
	answer = NULL;
102 103
lookup_protocol:
	err = -ESOCKTNOSUPPORT;
L
Linus Torvalds 已提交
104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
	rcu_read_lock();
	list_for_each_rcu(p, &inetsw6[sock->type]) {
		answer = list_entry(p, struct inet_protosw, list);

		/* 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;
		}
121
		err = -EPROTONOSUPPORT;
L
Linus Torvalds 已提交
122 123 124
		answer = NULL;
	}

125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
	if (!answer) {
		if (try_loading_module < 2) {
			rcu_read_unlock();
			/*
			 * Be more specific, e.g. net-pf-10-proto-132-type-1
			 * (net-pf-PF_INET6-proto-IPPROTO_SCTP-type-SOCK_STREAM)
			 */
			if (++try_loading_module == 1)
				request_module("net-pf-%d-proto-%d-type-%d",
						PF_INET6, protocol, sock->type);
			/*
			 * Fall back to generic, e.g. net-pf-10-proto-132
			 * (net-pf-PF_INET6-proto-IPPROTO_SCTP)
			 */
			else
				request_module("net-pf-%d-proto-%d",
						PF_INET6, protocol);
			goto lookup_protocol;
		} else
			goto out_rcu_unlock;
	}

	err = -EPERM;
L
Linus Torvalds 已提交
148 149 150 151 152 153 154 155 156 157 158
	if (answer->capability > 0 && !capable(answer->capability))
		goto out_rcu_unlock;

	sock->ops = answer->ops;
	answer_prot = answer->prot;
	answer_no_check = answer->no_check;
	answer_flags = answer->flags;
	rcu_read_unlock();

	BUG_TRAP(answer_prot->slab != NULL);

159
	err = -ENOBUFS;
160
	sk = sk_alloc(net, PF_INET6, GFP_KERNEL, answer_prot);
L
Linus Torvalds 已提交
161 162 163 164 165
	if (sk == NULL)
		goto out;

	sock_init_data(sock, sk);

166
	err = 0;
L
Linus Torvalds 已提交
167 168 169 170 171
	sk->sk_no_check = answer_no_check;
	if (INET_PROTOSW_REUSE & answer_flags)
		sk->sk_reuse = 1;

	inet = inet_sk(sk);
172
	inet->is_icsk = (INET_PROTOSW_ICSK & answer_flags) != 0;
L
Linus Torvalds 已提交
173 174 175 176 177 178 179

	if (SOCK_RAW == sock->type) {
		inet->num = protocol;
		if (IPPROTO_RAW == protocol)
			inet->hdrincl = 1;
	}

180
	sk->sk_destruct		= inet_sock_destruct;
L
Linus Torvalds 已提交
181 182 183 184 185 186 187 188 189 190
	sk->sk_family		= PF_INET6;
	sk->sk_protocol		= protocol;

	sk->sk_backlog_rcv	= answer->prot->backlog_rcv;

	inet_sk(sk)->pinet6 = np = inet6_sk_generic(sk);
	np->hop_limit	= -1;
	np->mcast_hops	= -1;
	np->mc_loop	= 1;
	np->pmtudisc	= IPV6_PMTUDISC_WANT;
191
	np->ipv6only	= init_net.ipv6.sysctl.bindv6only;
192

L
Linus Torvalds 已提交
193 194 195 196 197 198 199 200 201 202 203 204 205 206
	/* Init the ipv4 part of the socket since we can have sockets
	 * using v6 API for ipv4.
	 */
	inet->uc_ttl	= -1;

	inet->mc_loop	= 1;
	inet->mc_ttl	= 1;
	inet->mc_index	= 0;
	inet->mc_list	= NULL;

	if (ipv4_config.no_pmtu_disc)
		inet->pmtudisc = IP_PMTUDISC_DONT;
	else
		inet->pmtudisc = IP_PMTUDISC_WANT;
207
	/*
208 209 210 211 212 213 214 215 216
	 * Increment only the relevant sk_prot->socks debug field, this changes
	 * the previous behaviour of incrementing both the equivalent to
	 * answer->prot->socks (inet6_sock_nr) and inet_sock_nr.
	 *
	 * This allows better debug granularity as we'll know exactly how many
	 * UDPv6, TCPv6, etc socks were allocated, not the sum of all IPv6
	 * transport protocol socks. -acme
	 */
	sk_refcnt_debug_inc(sk);
L
Linus Torvalds 已提交
217 218 219 220 221 222

	if (inet->num) {
		/* It assumes that any protocol which allows
		 * the user to assign a number at socket
		 * creation time automatically shares.
		 */
A
Al Viro 已提交
223
		inet->sport = htons(inet->num);
L
Linus Torvalds 已提交
224 225 226
		sk->sk_prot->hash(sk);
	}
	if (sk->sk_prot->init) {
227 228
		err = sk->sk_prot->init(sk);
		if (err) {
L
Linus Torvalds 已提交
229 230 231 232 233
			sk_common_release(sk);
			goto out;
		}
	}
out:
234
	return err;
L
Linus Torvalds 已提交
235 236 237 238 239 240 241 242 243 244 245 246 247
out_rcu_unlock:
	rcu_read_unlock();
	goto out;
}


/* bind for INET6 API */
int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
{
	struct sockaddr_in6 *addr=(struct sockaddr_in6 *)uaddr;
	struct sock *sk = sock->sk;
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
248
	struct net *net = sk->sk_net;
A
Al Viro 已提交
249
	__be32 v4addr = 0;
L
Linus Torvalds 已提交
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278
	unsigned short snum;
	int addr_type = 0;
	int err = 0;

	/* If the socket has its own bind function then use it. */
	if (sk->sk_prot->bind)
		return sk->sk_prot->bind(sk, uaddr, addr_len);

	if (addr_len < SIN6_LEN_RFC2133)
		return -EINVAL;
	addr_type = ipv6_addr_type(&addr->sin6_addr);
	if ((addr_type & IPV6_ADDR_MULTICAST) && sock->type == SOCK_STREAM)
		return -EINVAL;

	snum = ntohs(addr->sin6_port);
	if (snum && snum < PROT_SOCK && !capable(CAP_NET_BIND_SERVICE))
		return -EACCES;

	lock_sock(sk);

	/* Check these errors (active socket, double bind). */
	if (sk->sk_state != TCP_CLOSE || inet->num) {
		err = -EINVAL;
		goto out;
	}

	/* Check if the address belongs to the host. */
	if (addr_type == IPV6_ADDR_MAPPED) {
		v4addr = addr->sin6_addr.s6_addr32[3];
279
		if (inet_addr_type(net, v4addr) != RTN_LOCAL) {
L
Linus Torvalds 已提交
280 281 282 283 284 285 286 287 288 289 290 291 292 293 294
			err = -EADDRNOTAVAIL;
			goto out;
		}
	} else {
		if (addr_type != IPV6_ADDR_ANY) {
			struct net_device *dev = NULL;

			if (addr_type & IPV6_ADDR_LINKLOCAL) {
				if (addr_len >= sizeof(struct sockaddr_in6) &&
				    addr->sin6_scope_id) {
					/* Override any existing binding, if another one
					 * is supplied by user.
					 */
					sk->sk_bound_dev_if = addr->sin6_scope_id;
				}
295

L
Linus Torvalds 已提交
296 297 298 299 300
				/* Binding to link-local address requires an interface */
				if (!sk->sk_bound_dev_if) {
					err = -EINVAL;
					goto out;
				}
301
				dev = dev_get_by_index(net, sk->sk_bound_dev_if);
L
Linus Torvalds 已提交
302 303 304 305 306 307 308 309 310 311 312
				if (!dev) {
					err = -ENODEV;
					goto out;
				}
			}

			/* ipv4 addr of the socket is invalid.  Only the
			 * unspecified and mapped address have a v4 equivalent.
			 */
			v4addr = LOOPBACK4_IPV6;
			if (!(addr_type & IPV6_ADDR_MULTICAST))	{
313
				if (!ipv6_chk_addr(net, &addr->sin6_addr,
314
						   dev, 0)) {
L
Linus Torvalds 已提交
315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
					if (dev)
						dev_put(dev);
					err = -EADDRNOTAVAIL;
					goto out;
				}
			}
			if (dev)
				dev_put(dev);
		}
	}

	inet->rcv_saddr = v4addr;
	inet->saddr = v4addr;

	ipv6_addr_copy(&np->rcv_saddr, &addr->sin6_addr);
330

L
Linus Torvalds 已提交
331 332 333 334 335 336 337 338 339 340 341 342 343 344
	if (!(addr_type & IPV6_ADDR_MULTICAST))
		ipv6_addr_copy(&np->saddr, &addr->sin6_addr);

	/* Make sure we are allowed to bind here. */
	if (sk->sk_prot->get_port(sk, snum)) {
		inet_reset_saddr(sk);
		err = -EADDRINUSE;
		goto out;
	}

	if (addr_type != IPV6_ADDR_ANY)
		sk->sk_userlocks |= SOCK_BINDADDR_LOCK;
	if (snum)
		sk->sk_userlocks |= SOCK_BINDPORT_LOCK;
A
Al Viro 已提交
345
	inet->sport = htons(inet->num);
L
Linus Torvalds 已提交
346 347 348 349 350 351 352
	inet->dport = 0;
	inet->daddr = 0;
out:
	release_sock(sk);
	return err;
}

353 354
EXPORT_SYMBOL(inet6_bind);

L
Linus Torvalds 已提交
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370
int inet6_release(struct socket *sock)
{
	struct sock *sk = sock->sk;

	if (sk == NULL)
		return -EINVAL;

	/* Free mc lists */
	ipv6_sock_mc_close(sk);

	/* Free ac lists */
	ipv6_sock_ac_close(sk);

	return inet_release(sock);
}

371 372
EXPORT_SYMBOL(inet6_release);

L
Linus Torvalds 已提交
373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394
int inet6_destroy_sock(struct sock *sk)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct sk_buff *skb;
	struct ipv6_txoptions *opt;

	/* Release rx options */

	if ((skb = xchg(&np->pktoptions, NULL)) != NULL)
		kfree_skb(skb);

	/* Free flowlabels */
	fl6_free_socklist(sk);

	/* Free tx options */

	if ((opt = xchg(&np->opt, NULL)) != NULL)
		sock_kfree_s(sk, opt, opt->tot_len);

	return 0;
}

395 396
EXPORT_SYMBOL_GPL(inet6_destroy_sock);

L
Linus Torvalds 已提交
397 398 399
/*
 *	This does both peername and sockname.
 */
400

L
Linus Torvalds 已提交
401 402 403 404 405 406 407
int inet6_getname(struct socket *sock, struct sockaddr *uaddr,
		 int *uaddr_len, int peer)
{
	struct sockaddr_in6 *sin=(struct sockaddr_in6 *)uaddr;
	struct sock *sk = sock->sk;
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
408

L
Linus Torvalds 已提交
409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435
	sin->sin6_family = AF_INET6;
	sin->sin6_flowinfo = 0;
	sin->sin6_scope_id = 0;
	if (peer) {
		if (!inet->dport)
			return -ENOTCONN;
		if (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_SYN_SENT)) &&
		    peer == 1)
			return -ENOTCONN;
		sin->sin6_port = inet->dport;
		ipv6_addr_copy(&sin->sin6_addr, &np->daddr);
		if (np->sndflow)
			sin->sin6_flowinfo = np->flow_label;
	} else {
		if (ipv6_addr_any(&np->rcv_saddr))
			ipv6_addr_copy(&sin->sin6_addr, &np->saddr);
		else
			ipv6_addr_copy(&sin->sin6_addr, &np->rcv_saddr);

		sin->sin6_port = inet->sport;
	}
	if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL)
		sin->sin6_scope_id = sk->sk_bound_dev_if;
	*uaddr_len = sizeof(*sin);
	return(0);
}

436 437
EXPORT_SYMBOL(inet6_getname);

L
Linus Torvalds 已提交
438 439 440
int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
{
	struct sock *sk = sock->sk;
441
	struct net *net = sk->sk_net;
L
Linus Torvalds 已提交
442

443
	switch(cmd)
L
Linus Torvalds 已提交
444 445 446 447
	{
	case SIOCGSTAMP:
		return sock_get_timestamp(sk, (struct timeval __user *)arg);

448 449 450
	case SIOCGSTAMPNS:
		return sock_get_timestampns(sk, (struct timespec __user *)arg);

L
Linus Torvalds 已提交
451 452
	case SIOCADDRT:
	case SIOCDELRT:
453

454
		return(ipv6_route_ioctl(net, cmd, (void __user *)arg));
L
Linus Torvalds 已提交
455 456 457 458 459 460 461 462

	case SIOCSIFADDR:
		return addrconf_add_ifaddr((void __user *) arg);
	case SIOCDIFADDR:
		return addrconf_del_ifaddr((void __user *) arg);
	case SIOCSIFDSTADDR:
		return addrconf_set_dstaddr((void __user *) arg);
	default:
463 464 465
		if (!sk->sk_prot->ioctl)
			return -ENOIOCTLCMD;
		return sk->sk_prot->ioctl(sk, cmd, arg);
L
Linus Torvalds 已提交
466 467 468 469 470
	}
	/*NOTREACHED*/
	return(0);
}

471 472
EXPORT_SYMBOL(inet6_ioctl);

473
const struct proto_ops inet6_stream_ops = {
474 475 476 477 478 479 480 481 482 483 484 485 486 487
	.family		   = PF_INET6,
	.owner		   = THIS_MODULE,
	.release	   = inet6_release,
	.bind		   = inet6_bind,
	.connect	   = inet_stream_connect,	/* ok		*/
	.socketpair	   = sock_no_socketpair,	/* a do nothing	*/
	.accept		   = inet_accept,		/* ok		*/
	.getname	   = inet6_getname,
	.poll		   = tcp_poll,			/* ok		*/
	.ioctl		   = inet6_ioctl,		/* must change  */
	.listen		   = inet_listen,		/* ok		*/
	.shutdown	   = inet_shutdown,		/* ok		*/
	.setsockopt	   = sock_common_setsockopt,	/* ok		*/
	.getsockopt	   = sock_common_getsockopt,	/* ok		*/
488
	.sendmsg	   = tcp_sendmsg,		/* ok		*/
489 490 491
	.recvmsg	   = sock_common_recvmsg,	/* ok		*/
	.mmap		   = sock_no_mmap,
	.sendpage	   = tcp_sendpage,
492
	.splice_read	   = tcp_splice_read,
493
#ifdef CONFIG_COMPAT
494 495
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
496
#endif
L
Linus Torvalds 已提交
497 498
};

499
const struct proto_ops inet6_dgram_ops = {
500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517
	.family		   = PF_INET6,
	.owner		   = THIS_MODULE,
	.release	   = inet6_release,
	.bind		   = inet6_bind,
	.connect	   = inet_dgram_connect,	/* ok		*/
	.socketpair	   = sock_no_socketpair,	/* a do nothing	*/
	.accept		   = sock_no_accept,		/* a do nothing	*/
	.getname	   = inet6_getname,
	.poll		   = udp_poll,			/* ok		*/
	.ioctl		   = inet6_ioctl,		/* must change  */
	.listen		   = sock_no_listen,		/* ok		*/
	.shutdown	   = inet_shutdown,		/* ok		*/
	.setsockopt	   = sock_common_setsockopt,	/* ok		*/
	.getsockopt	   = sock_common_getsockopt,	/* ok		*/
	.sendmsg	   = inet_sendmsg,		/* ok		*/
	.recvmsg	   = sock_common_recvmsg,	/* ok		*/
	.mmap		   = sock_no_mmap,
	.sendpage	   = sock_no_sendpage,
518
#ifdef CONFIG_COMPAT
519 520
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
521
#endif
L
Linus Torvalds 已提交
522 523 524 525 526 527 528 529
};

static struct net_proto_family inet6_family_ops = {
	.family = PF_INET6,
	.create = inet6_create,
	.owner	= THIS_MODULE,
};

530
int inet6_register_protosw(struct inet_protosw *p)
L
Linus Torvalds 已提交
531 532 533 534
{
	struct list_head *lh;
	struct inet_protosw *answer;
	struct list_head *last_perm;
535 536
	int protocol = p->protocol;
	int ret;
L
Linus Torvalds 已提交
537 538 539

	spin_lock_bh(&inetsw6_lock);

540
	ret = -EINVAL;
L
Linus Torvalds 已提交
541 542 543 544 545
	if (p->type >= SOCK_MAX)
		goto out_illegal;

	/* If we are trying to override a permanent protocol, bail. */
	answer = NULL;
546
	ret = -EPERM;
L
Linus Torvalds 已提交
547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565
	last_perm = &inetsw6[p->type];
	list_for_each(lh, &inetsw6[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
566
	 * non-permanent entry.  This means that when we remove this entry, the
L
Linus Torvalds 已提交
567 568 569
	 * system automatically returns to the old behavior.
	 */
	list_add_rcu(&p->list, last_perm);
570
	ret = 0;
L
Linus Torvalds 已提交
571 572
out:
	spin_unlock_bh(&inetsw6_lock);
573
	return ret;
L
Linus Torvalds 已提交
574 575 576 577 578 579 580 581 582 583 584 585 586

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;
}

587 588
EXPORT_SYMBOL(inet6_register_protosw);

L
Linus Torvalds 已提交
589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604
void
inet6_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(&inetsw6_lock);
		list_del_rcu(&p->list);
		spin_unlock_bh(&inetsw6_lock);

		synchronize_net();
	}
}

605 606
EXPORT_SYMBOL(inet6_unregister_protosw);

607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627
int inet6_sk_rebuild_header(struct sock *sk)
{
	int err;
	struct dst_entry *dst;
	struct ipv6_pinfo *np = inet6_sk(sk);

	dst = __sk_dst_check(sk, np->dst_cookie);

	if (dst == NULL) {
		struct inet_sock *inet = inet_sk(sk);
		struct in6_addr *final_p = NULL, final;
		struct flowi fl;

		memset(&fl, 0, sizeof(fl));
		fl.proto = sk->sk_protocol;
		ipv6_addr_copy(&fl.fl6_dst, &np->daddr);
		ipv6_addr_copy(&fl.fl6_src, &np->saddr);
		fl.fl6_flowlabel = np->flow_label;
		fl.oif = sk->sk_bound_dev_if;
		fl.fl_ip_dport = inet->dport;
		fl.fl_ip_sport = inet->sport;
V
Venkat Yekkirala 已提交
628
		security_sk_classify_flow(sk, &fl);
629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649

		if (np->opt && np->opt->srcrt) {
			struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt;
			ipv6_addr_copy(&final, &fl.fl6_dst);
			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
			final_p = &final;
		}

		err = ip6_dst_lookup(sk, &dst, &fl);
		if (err) {
			sk->sk_route_caps = 0;
			return err;
		}
		if (final_p)
			ipv6_addr_copy(&fl.fl6_dst, final_p);

		if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) {
			sk->sk_err_soft = -err;
			return err;
		}

650
		__ip6_dst_store(sk, dst, NULL, NULL);
651 652 653 654 655 656 657
	}

	return 0;
}

EXPORT_SYMBOL_GPL(inet6_sk_rebuild_header);

658 659 660 661 662 663 664 665
int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct inet6_skb_parm *opt = IP6CB(skb);

	if (np->rxopt.all) {
		if ((opt->hop && (np->rxopt.bits.hopopts ||
				  np->rxopt.bits.ohopopts)) ||
666 667
		    ((IPV6_FLOWINFO_MASK &
		      *(__be32 *)skb_network_header(skb)) &&
668 669 670 671 672 673 674 675 676 677 678 679
		     np->rxopt.bits.rxflow) ||
		    (opt->srcrt && (np->rxopt.bits.srcrt ||
		     np->rxopt.bits.osrcrt)) ||
		    ((opt->dst1 || opt->dst0) &&
		     (np->rxopt.bits.dstopts || np->rxopt.bits.odstopts)))
			return 1;
	}
	return 0;
}

EXPORT_SYMBOL_GPL(ipv6_opt_accepted);

680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802
static struct inet6_protocol *ipv6_gso_pull_exthdrs(struct sk_buff *skb,
						    int proto)
{
	struct inet6_protocol *ops = NULL;

	for (;;) {
		struct ipv6_opt_hdr *opth;
		int len;

		if (proto != NEXTHDR_HOP) {
			ops = rcu_dereference(inet6_protos[proto]);

			if (unlikely(!ops))
				break;

			if (!(ops->flags & INET6_PROTO_GSO_EXTHDR))
				break;
		}

		if (unlikely(!pskb_may_pull(skb, 8)))
			break;

		opth = (void *)skb->data;
		len = ipv6_optlen(opth);

		if (unlikely(!pskb_may_pull(skb, len)))
			break;

		proto = opth->nexthdr;
		__skb_pull(skb, len);
	}

	return ops;
}

static int ipv6_gso_send_check(struct sk_buff *skb)
{
	struct ipv6hdr *ipv6h;
	struct inet6_protocol *ops;
	int err = -EINVAL;

	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
		goto out;

	ipv6h = ipv6_hdr(skb);
	__skb_pull(skb, sizeof(*ipv6h));
	err = -EPROTONOSUPPORT;

	rcu_read_lock();
	ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);
	if (likely(ops && ops->gso_send_check)) {
		skb_reset_transport_header(skb);
		err = ops->gso_send_check(skb);
	}
	rcu_read_unlock();

out:
	return err;
}

static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, int features)
{
	struct sk_buff *segs = ERR_PTR(-EINVAL);
	struct ipv6hdr *ipv6h;
	struct inet6_protocol *ops;

	if (!(features & NETIF_F_V6_CSUM))
		features &= ~NETIF_F_SG;

	if (unlikely(skb_shinfo(skb)->gso_type &
		     ~(SKB_GSO_UDP |
		       SKB_GSO_DODGY |
		       SKB_GSO_TCP_ECN |
		       SKB_GSO_TCPV6 |
		       0)))
		goto out;

	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
		goto out;

	ipv6h = ipv6_hdr(skb);
	__skb_pull(skb, sizeof(*ipv6h));
	segs = ERR_PTR(-EPROTONOSUPPORT);

	rcu_read_lock();
	ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);
	if (likely(ops && ops->gso_segment)) {
		skb_reset_transport_header(skb);
		segs = ops->gso_segment(skb, features);
	}
	rcu_read_unlock();

	if (unlikely(IS_ERR(segs)))
		goto out;

	for (skb = segs; skb; skb = skb->next) {
		ipv6h = ipv6_hdr(skb);
		ipv6h->payload_len = htons(skb->len - skb->mac_len -
					   sizeof(*ipv6h));
	}

out:
	return segs;
}

static struct packet_type ipv6_packet_type = {
	.type = __constant_htons(ETH_P_IPV6),
	.func = ipv6_rcv,
	.gso_send_check = ipv6_gso_send_check,
	.gso_segment = ipv6_gso_segment,
};

static int __init ipv6_packet_init(void)
{
	dev_add_pack(&ipv6_packet_type);
	return 0;
}

static void ipv6_packet_cleanup(void)
{
	dev_remove_pack(&ipv6_packet_type);
}

L
Linus Torvalds 已提交
803 804
static int __init init_ipv6_mibs(void)
{
805 806
	if (snmp_mib_init((void **)ipv6_statistics,
			  sizeof(struct ipstats_mib)) < 0)
L
Linus Torvalds 已提交
807
		goto err_ip_mib;
808 809
	if (snmp_mib_init((void **)icmpv6_statistics,
			  sizeof(struct icmpv6_mib)) < 0)
L
Linus Torvalds 已提交
810
		goto err_icmp_mib;
811
	if (snmp_mib_init((void **)icmpv6msg_statistics,
812
			  sizeof(struct icmpv6msg_mib)) < 0)
813
		goto err_icmpmsg_mib;
814
	if (snmp_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib)) < 0)
L
Linus Torvalds 已提交
815
		goto err_udp_mib;
816
#ifdef CONFIG_IP_UDPLITE
817 818
	if (snmp_mib_init((void **)udplite_stats_in6,
			  sizeof (struct udp_mib)) < 0)
819
		goto err_udplite_mib;
820
#endif
L
Linus Torvalds 已提交
821 822
	return 0;

823
#ifdef CONFIG_IP_UDPLITE
824
err_udplite_mib:
825
#endif
826
	snmp_mib_free((void **)udp_stats_in6);
L
Linus Torvalds 已提交
827
err_udp_mib:
828 829
	snmp_mib_free((void **)icmpv6msg_statistics);
err_icmpmsg_mib:
830
	snmp_mib_free((void **)icmpv6_statistics);
L
Linus Torvalds 已提交
831
err_icmp_mib:
832
	snmp_mib_free((void **)ipv6_statistics);
L
Linus Torvalds 已提交
833 834
err_ip_mib:
	return -ENOMEM;
835

L
Linus Torvalds 已提交
836 837 838 839
}

static void cleanup_ipv6_mibs(void)
{
840 841
	snmp_mib_free((void **)ipv6_statistics);
	snmp_mib_free((void **)icmpv6_statistics);
842
	snmp_mib_free((void **)icmpv6msg_statistics);
843
	snmp_mib_free((void **)udp_stats_in6);
844
#ifdef CONFIG_IP_UDPLITE
845
	snmp_mib_free((void **)udplite_stats_in6);
846
#endif
L
Linus Torvalds 已提交
847 848
}

849 850
static int inet6_net_init(struct net *net)
{
851
	net->ipv6.sysctl.bindv6only = 0;
852 853 854 855 856 857 858 859
	net->ipv6.sysctl.flush_delay = 0;
	net->ipv6.sysctl.ip6_rt_max_size = 4096;
	net->ipv6.sysctl.ip6_rt_gc_min_interval = HZ / 2;
	net->ipv6.sysctl.ip6_rt_gc_timeout = 60*HZ;
	net->ipv6.sysctl.ip6_rt_gc_interval = 30*HZ;
	net->ipv6.sysctl.ip6_rt_gc_elasticity = 9;
	net->ipv6.sysctl.ip6_rt_mtu_expires = 10*60*HZ;
	net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
860
	net->ipv6.sysctl.icmpv6_time = 1*HZ;
861

862 863 864 865 866 867 868 869 870 871 872 873 874
	return 0;
}

static void inet6_net_exit(struct net *net)
{
	return;
}

static struct pernet_operations inet6_net_ops = {
	.init = inet6_net_init,
	.exit = inet6_net_exit,
};

L
Linus Torvalds 已提交
875 876 877
static int __init inet6_init(void)
{
	struct sk_buff *dummy_skb;
878
	struct list_head *r;
L
Linus Torvalds 已提交
879 880
	int err;

881 882
	BUILD_BUG_ON(sizeof(struct inet6_skb_parm) > sizeof(dummy_skb->cb));

L
Linus Torvalds 已提交
883 884 885 886 887 888 889 890
	err = proto_register(&tcpv6_prot, 1);
	if (err)
		goto out;

	err = proto_register(&udpv6_prot, 1);
	if (err)
		goto out_unregister_tcp_proto;

891
#ifdef CONFIG_IP_UDPLITE
892
	err = proto_register(&udplitev6_prot, 1);
L
Linus Torvalds 已提交
893 894
	if (err)
		goto out_unregister_udp_proto;
895
#endif
L
Linus Torvalds 已提交
896

897 898 899 900
	err = proto_register(&rawv6_prot, 1);
	if (err)
		goto out_unregister_udplite_proto;

L
Linus Torvalds 已提交
901 902 903 904 905 906 907 908

	/* Register the socket-side information for inet6_create.  */
	for(r = &inetsw6[0]; r < &inetsw6[SOCK_MAX]; ++r)
		INIT_LIST_HEAD(r);

	/* We MUST register RAW sockets before we create the ICMP6,
	 * IGMP6, or NDISC control sockets.
	 */
909 910 911
	err = rawv6_init();
	if (err)
		goto out_unregister_raw_proto;
L
Linus Torvalds 已提交
912 913 914 915

	/* Register the family here so that the init calls below will
	 * be able to create sockets. (?? is this dangerous ??)
	 */
916 917
	err = sock_register(&inet6_family_ops);
	if (err)
918
		goto out_sock_register_fail;
L
Linus Torvalds 已提交
919 920 921 922

	/* Initialise ipv6 mibs */
	err = init_ipv6_mibs();
	if (err)
923
		goto out_unregister_sock;
924

L
Linus Torvalds 已提交
925 926 927 928 929 930 931
	/*
	 *	ipngwg API draft makes clear that the correct semantics
	 *	for TCP and UDP is to consider one TCP and UDP instance
	 *	in a host availiable by both INET and INET6 APIs and
	 *	able to communicate via both network protocols.
	 */

932 933 934
	err = register_pernet_subsys(&inet6_net_ops);
	if (err)
		goto register_pernet_fail;
935
	err = icmpv6_init();
L
Linus Torvalds 已提交
936 937
	if (err)
		goto icmp_fail;
938
	err = ndisc_init();
L
Linus Torvalds 已提交
939 940
	if (err)
		goto ndisc_fail;
941
	err = igmp6_init();
L
Linus Torvalds 已提交
942 943
	if (err)
		goto igmp_fail;
944 945 946
	err = ipv6_netfilter_init();
	if (err)
		goto netfilter_fail;
L
Linus Torvalds 已提交
947 948 949 950 951 952 953 954 955
	/* Create /proc/foo6 entries. */
#ifdef CONFIG_PROC_FS
	err = -ENOMEM;
	if (raw6_proc_init())
		goto proc_raw6_fail;
	if (tcp6_proc_init())
		goto proc_tcp6_fail;
	if (udp6_proc_init())
		goto proc_udp6_fail;
956 957
	if (udplite6_proc_init())
		goto proc_udplite6_fail;
L
Linus Torvalds 已提交
958 959 960 961 962 963 964 965
	if (ipv6_misc_proc_init())
		goto proc_misc6_fail;

	if (ac6_proc_init())
		goto proc_anycast6_fail;
	if (if6_proc_init())
		goto proc_if6_fail;
#endif
966 967 968
	err = ip6_route_init();
	if (err)
		goto ip6_route_fail;
969 970 971
	err = ip6_flowlabel_init();
	if (err)
		goto ip6_flowlabel_fail;
L
Linus Torvalds 已提交
972 973 974 975 976
	err = addrconf_init();
	if (err)
		goto addrconf_fail;

	/* Init v6 extension headers. */
977 978 979 980
	err = ipv6_exthdrs_init();
	if (err)
		goto ipv6_exthdrs_fail;

981 982 983
	err = ipv6_frag_init();
	if (err)
		goto ipv6_frag_fail;
L
Linus Torvalds 已提交
984 985

	/* Init v6 transport protocols. */
986 987 988
	err = udpv6_init();
	if (err)
		goto udpv6_fail;
989

990 991 992 993 994 995 996 997 998 999 1000
	err = udplitev6_init();
	if (err)
		goto udplitev6_fail;

	err = tcpv6_init();
	if (err)
		goto tcpv6_fail;

	err = ipv6_packet_init();
	if (err)
		goto ipv6_packet_fail;
1001 1002 1003 1004 1005 1006

#ifdef CONFIG_SYSCTL
	err = ipv6_sysctl_register();
	if (err)
		goto sysctl_fail;
#endif
L
Linus Torvalds 已提交
1007 1008 1009
out:
	return err;

1010 1011 1012 1013
#ifdef CONFIG_SYSCTL
sysctl_fail:
	ipv6_packet_cleanup();
#endif
1014 1015 1016 1017 1018 1019 1020 1021
ipv6_packet_fail:
	tcpv6_exit();
tcpv6_fail:
	udplitev6_exit();
udplitev6_fail:
	udpv6_exit();
udpv6_fail:
	ipv6_frag_exit();
1022 1023
ipv6_frag_fail:
	ipv6_exthdrs_exit();
1024 1025
ipv6_exthdrs_fail:
	addrconf_cleanup();
L
Linus Torvalds 已提交
1026 1027
addrconf_fail:
	ip6_flowlabel_cleanup();
1028
ip6_flowlabel_fail:
L
Linus Torvalds 已提交
1029
	ip6_route_cleanup();
1030
ip6_route_fail:
L
Linus Torvalds 已提交
1031 1032 1033 1034 1035 1036 1037
#ifdef CONFIG_PROC_FS
	if6_proc_exit();
proc_if6_fail:
	ac6_proc_exit();
proc_anycast6_fail:
	ipv6_misc_proc_exit();
proc_misc6_fail:
1038 1039
	udplite6_proc_exit();
proc_udplite6_fail:
L
Linus Torvalds 已提交
1040 1041 1042 1043 1044 1045 1046
	udp6_proc_exit();
proc_udp6_fail:
	tcp6_proc_exit();
proc_tcp6_fail:
	raw6_proc_exit();
proc_raw6_fail:
#endif
1047 1048
	ipv6_netfilter_fini();
netfilter_fail:
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053 1054
	igmp6_cleanup();
igmp_fail:
	ndisc_cleanup();
ndisc_fail:
	icmpv6_cleanup();
icmp_fail:
1055 1056
	unregister_pernet_subsys(&inet6_net_ops);
register_pernet_fail:
L
Linus Torvalds 已提交
1057
	cleanup_ipv6_mibs();
1058 1059
out_unregister_sock:
	sock_unregister(PF_INET6);
1060
	rtnl_unregister_all(PF_INET6);
1061 1062
out_sock_register_fail:
	rawv6_exit();
L
Linus Torvalds 已提交
1063 1064
out_unregister_raw_proto:
	proto_unregister(&rawv6_prot);
1065
out_unregister_udplite_proto:
1066
#ifdef CONFIG_IP_UDPLITE
1067
	proto_unregister(&udplitev6_prot);
L
Linus Torvalds 已提交
1068
out_unregister_udp_proto:
1069
#endif
L
Linus Torvalds 已提交
1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080
	proto_unregister(&udpv6_prot);
out_unregister_tcp_proto:
	proto_unregister(&tcpv6_prot);
	goto out;
}
module_init(inet6_init);

static void __exit inet6_exit(void)
{
	/* First of all disallow new sockets creation. */
	sock_unregister(PF_INET6);
1081 1082
	/* Disallow any further netlink messages */
	rtnl_unregister_all(PF_INET6);
1083

1084 1085 1086
#ifdef CONFIG_SYSCTL
	ipv6_sysctl_unregister();
#endif
1087
	udpv6_exit();
1088
#ifdef CONFIG_IP_UDPLITE
1089
	udplitev6_exit();
1090
#endif
1091 1092
	tcpv6_exit();

1093 1094
	/* Cleanup code parts. */
	ipv6_packet_cleanup();
1095
	ipv6_frag_exit();
1096
	ipv6_exthdrs_exit();
1097 1098 1099
	addrconf_cleanup();
	ip6_flowlabel_cleanup();
	ip6_route_cleanup();
L
Linus Torvalds 已提交
1100
#ifdef CONFIG_PROC_FS
1101 1102

	/* Cleanup code parts. */
L
Linus Torvalds 已提交
1103 1104
	if6_proc_exit();
	ac6_proc_exit();
1105 1106
	ipv6_misc_proc_exit();
	udplite6_proc_exit();
1107
	udp6_proc_exit();
1108 1109
	tcp6_proc_exit();
	raw6_proc_exit();
1110
#endif
1111
	ipv6_netfilter_fini();
1112
	igmp6_cleanup();
L
Linus Torvalds 已提交
1113 1114
	ndisc_cleanup();
	icmpv6_cleanup();
1115
	rawv6_exit();
1116

1117
	unregister_pernet_subsys(&inet6_net_ops);
L
Linus Torvalds 已提交
1118 1119
	cleanup_ipv6_mibs();
	proto_unregister(&rawv6_prot);
1120
#ifdef CONFIG_IP_UDPLITE
1121
	proto_unregister(&udplitev6_prot);
1122
#endif
L
Linus Torvalds 已提交
1123 1124 1125 1126 1127 1128
	proto_unregister(&udpv6_prot);
	proto_unregister(&tcpv6_prot);
}
module_exit(inet6_exit);

MODULE_ALIAS_NETPROTO(PF_INET6);