raw.c 30.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2
/*
 *	RAW sockets for IPv6
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
 *
 *	Adapted from linux/net/ipv4/raw.c
 *
 *	$Id: raw.c,v 1.51 2002/02/01 22:01:04 davem Exp $
 *
 *	Fixes:
 *	Hideaki YOSHIFUJI	:	sin6_scope_id support
14
 *	YOSHIFUJI,H.@USAGI	:	raw checksum (RFC2292(bis) compliance)
L
Linus Torvalds 已提交
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
 *	Kazunori MIYAZAWA @USAGI:	change process style to use ip6_append_data
 *
 *	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/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/in6.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/icmpv6.h>
#include <linux/netfilter.h>
#include <linux/netfilter_ipv6.h>
34
#include <linux/skbuff.h>
L
Linus Torvalds 已提交
35 36 37
#include <asm/uaccess.h>
#include <asm/ioctls.h>

38
#include <net/net_namespace.h>
L
Linus Torvalds 已提交
39 40 41 42 43 44 45 46 47 48 49 50 51
#include <net/ip.h>
#include <net/sock.h>
#include <net/snmp.h>

#include <net/ipv6.h>
#include <net/ndisc.h>
#include <net/protocol.h>
#include <net/ip6_route.h>
#include <net/ip6_checksum.h>
#include <net/addrconf.h>
#include <net/transp_v6.h>
#include <net/udp.h>
#include <net/inet_common.h>
52
#include <net/tcp_states.h>
53
#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
54 55
#include <net/mip6.h>
#endif
56
#include <linux/mroute6.h>
L
Linus Torvalds 已提交
57

58
#include <net/raw.h>
L
Linus Torvalds 已提交
59 60 61 62 63 64
#include <net/rawv6.h>
#include <net/xfrm.h>

#include <linux/proc_fs.h>
#include <linux/seq_file.h>

65
static struct raw_hashinfo raw_v6_hashinfo = {
66
	.lock = __RW_LOCK_UNLOCKED(raw_v6_hashinfo.lock),
67
};
L
Linus Torvalds 已提交
68

69 70 71
static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk,
		unsigned short num, struct in6_addr *loc_addr,
		struct in6_addr *rmt_addr, int dif)
L
Linus Torvalds 已提交
72 73 74 75 76 77 78 79
{
	struct hlist_node *node;
	int is_multicast = ipv6_addr_is_multicast(loc_addr);

	sk_for_each_from(sk, node)
		if (inet_sk(sk)->num == num) {
			struct ipv6_pinfo *np = inet6_sk(sk);

80
			if (!net_eq(sock_net(sk), net))
81 82
				continue;

L
Linus Torvalds 已提交
83 84 85 86
			if (!ipv6_addr_any(&np->daddr) &&
			    !ipv6_addr_equal(&np->daddr, rmt_addr))
				continue;

87 88 89
			if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif)
				continue;

L
Linus Torvalds 已提交
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125
			if (!ipv6_addr_any(&np->rcv_saddr)) {
				if (ipv6_addr_equal(&np->rcv_saddr, loc_addr))
					goto found;
				if (is_multicast &&
				    inet6_mc_check(sk, loc_addr, rmt_addr))
					goto found;
				continue;
			}
			goto found;
		}
	sk = NULL;
found:
	return sk;
}

/*
 *	0 - deliver
 *	1 - block
 */
static __inline__ int icmpv6_filter(struct sock *sk, struct sk_buff *skb)
{
	struct icmp6hdr *icmph;
	struct raw6_sock *rp = raw6_sk(sk);

	if (pskb_may_pull(skb, sizeof(struct icmp6hdr))) {
		__u32 *data = &rp->filter.data[0];
		int bit_nr;

		icmph = (struct icmp6hdr *) skb->data;
		bit_nr = icmph->icmp6_type;

		return (data[bit_nr >> 5] & (1 << (bit_nr & 31))) != 0;
	}
	return 0;
}

126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
static int (*mh_filter)(struct sock *sock, struct sk_buff *skb);

int rawv6_mh_filter_register(int (*filter)(struct sock *sock,
					   struct sk_buff *skb))
{
	rcu_assign_pointer(mh_filter, filter);
	return 0;
}
EXPORT_SYMBOL(rawv6_mh_filter_register);

int rawv6_mh_filter_unregister(int (*filter)(struct sock *sock,
					     struct sk_buff *skb))
{
	rcu_assign_pointer(mh_filter, NULL);
	synchronize_rcu();
	return 0;
}
EXPORT_SYMBOL(rawv6_mh_filter_unregister);

#endif

L
Linus Torvalds 已提交
148 149 150 151 152 153 154
/*
 *	demultiplex raw sockets.
 *	(should consider queueing the skb in the sock receive_queue
 *	without calling rawv6.c)
 *
 *	Caller owns SKB so we must make clones.
 */
155
static int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
L
Linus Torvalds 已提交
156 157 158 159
{
	struct in6_addr *saddr;
	struct in6_addr *daddr;
	struct sock *sk;
160
	int delivered = 0;
L
Linus Torvalds 已提交
161
	__u8 hash;
162
	struct net *net;
L
Linus Torvalds 已提交
163

164
	saddr = &ipv6_hdr(skb)->saddr;
L
Linus Torvalds 已提交
165 166 167 168
	daddr = saddr + 1;

	hash = nexthdr & (MAX_INET_PROTOS - 1);

169 170
	read_lock(&raw_v6_hashinfo.lock);
	sk = sk_head(&raw_v6_hashinfo.ht[hash]);
L
Linus Torvalds 已提交
171 172 173 174

	if (sk == NULL)
		goto out;

175
	net = dev_net(skb->dev);
176
	sk = __raw_v6_lookup(net, sk, nexthdr, daddr, saddr, IP6CB(skb)->iif);
L
Linus Torvalds 已提交
177 178

	while (sk) {
179 180
		int filtered;

181
		delivered = 1;
182 183 184 185
		switch (nexthdr) {
		case IPPROTO_ICMPV6:
			filtered = icmpv6_filter(sk, skb);
			break;
186 187

#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
188
		case IPPROTO_MH:
189
		{
190 191 192 193 194 195
			/* XXX: To validate MH only once for each packet,
			 * this is placed here. It should be after checking
			 * xfrm policy, however it doesn't. The checking xfrm
			 * policy is placed in rawv6_rcv() because it is
			 * required for each socket.
			 */
196 197 198 199
			int (*filter)(struct sock *sock, struct sk_buff *skb);

			filter = rcu_dereference(mh_filter);
			filtered = filter ? filter(sk, skb) : 0;
200
			break;
201
		}
202 203 204 205 206 207 208 209 210
#endif
		default:
			filtered = 0;
			break;
		}

		if (filtered < 0)
			break;
		if (filtered == 0) {
L
Linus Torvalds 已提交
211 212 213
			struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC);

			/* Not releasing hash table! */
214 215
			if (clone) {
				nf_reset(clone);
L
Linus Torvalds 已提交
216
				rawv6_rcv(sk, clone);
217
			}
L
Linus Torvalds 已提交
218
		}
219
		sk = __raw_v6_lookup(net, sk_next(sk), nexthdr, daddr, saddr,
220
				     IP6CB(skb)->iif);
L
Linus Torvalds 已提交
221 222
	}
out:
223
	read_unlock(&raw_v6_hashinfo.lock);
224
	return delivered;
L
Linus Torvalds 已提交
225 226
}

227 228 229 230
int raw6_local_deliver(struct sk_buff *skb, int nexthdr)
{
	struct sock *raw_sk;

231
	raw_sk = sk_head(&raw_v6_hashinfo.ht[nexthdr & (MAX_INET_PROTOS - 1)]);
232 233 234 235 236 237
	if (raw_sk && !ipv6_raw_deliver(skb, nexthdr))
		raw_sk = NULL;

	return raw_sk != NULL;
}

L
Linus Torvalds 已提交
238 239 240 241 242 243
/* This cleans up af_inet6 a bit. -DaveM */
static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct sockaddr_in6 *addr = (struct sockaddr_in6 *) uaddr;
A
Al Viro 已提交
244
	__be32 v4addr = 0;
L
Linus Torvalds 已提交
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273
	int addr_type;
	int err;

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

	/* Raw sockets are IPv6 only */
	if (addr_type == IPV6_ADDR_MAPPED)
		return(-EADDRNOTAVAIL);

	lock_sock(sk);

	err = -EINVAL;
	if (sk->sk_state != TCP_CLOSE)
		goto out;

	/* Check if the address belongs to the host. */
	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;
			}
274

L
Linus Torvalds 已提交
275 276 277 278
			/* Binding to link-local address requires an interface */
			if (!sk->sk_bound_dev_if)
				goto out;

279
			dev = dev_get_by_index(sock_net(sk), sk->sk_bound_dev_if);
L
Linus Torvalds 已提交
280 281 282 283 284
			if (!dev) {
				err = -ENODEV;
				goto out;
			}
		}
285

L
Linus Torvalds 已提交
286 287 288 289 290 291
		/* 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))	{
			err = -EADDRNOTAVAIL;
292
			if (!ipv6_chk_addr(sock_net(sk), &addr->sin6_addr,
293
					   dev, 0)) {
L
Linus Torvalds 已提交
294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312
				if (dev)
					dev_put(dev);
				goto out;
			}
		}
		if (dev)
			dev_put(dev);
	}

	inet->rcv_saddr = inet->saddr = v4addr;
	ipv6_addr_copy(&np->rcv_saddr, &addr->sin6_addr);
	if (!(addr_type & IPV6_ADDR_MULTICAST))
		ipv6_addr_copy(&np->saddr, &addr->sin6_addr);
	err = 0;
out:
	release_sock(sk);
	return err;
}

313
static void rawv6_err(struct sock *sk, struct sk_buff *skb,
L
Linus Torvalds 已提交
314
	       struct inet6_skb_parm *opt,
315
	       int type, int code, int offset, __be32 info)
L
Linus Torvalds 已提交
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346
{
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
	int err;
	int harderr;

	/* Report error on raw socket, if:
	   1. User requested recverr.
	   2. Socket is connected (otherwise the error indication
	      is useless without recverr and error is hard.
	 */
	if (!np->recverr && sk->sk_state != TCP_ESTABLISHED)
		return;

	harderr = icmpv6_err_convert(type, code, &err);
	if (type == ICMPV6_PKT_TOOBIG)
		harderr = (np->pmtudisc == IPV6_PMTUDISC_DO);

	if (np->recverr) {
		u8 *payload = skb->data;
		if (!inet->hdrincl)
			payload += offset;
		ipv6_icmp_error(sk, skb, err, 0, ntohl(info), payload);
	}

	if (np->recverr || harderr) {
		sk->sk_err = err;
		sk->sk_error_report(sk);
	}
}

347 348 349 350 351 352
void raw6_icmp_error(struct sk_buff *skb, int nexthdr,
		int type, int code, int inner_offset, __be32 info)
{
	struct sock *sk;
	int hash;
	struct in6_addr *saddr, *daddr;
353
	struct net *net;
354

355
	hash = nexthdr & (RAW_HTABLE_SIZE - 1);
356

357 358
	read_lock(&raw_v6_hashinfo.lock);
	sk = sk_head(&raw_v6_hashinfo.ht[hash]);
359
	if (sk != NULL) {
360 361 362 363
		/* Note: ipv6_hdr(skb) != skb->data */
		struct ipv6hdr *ip6h = (struct ipv6hdr *)skb->data;
		saddr = &ip6h->saddr;
		daddr = &ip6h->daddr;
364
		net = dev_net(skb->dev);
365

366
		while ((sk = __raw_v6_lookup(net, sk, nexthdr, saddr, daddr,
367 368 369 370 371 372
						IP6CB(skb)->iif))) {
			rawv6_err(sk, skb, NULL, type, code,
					inner_offset, info);
			sk = sk_next(sk);
		}
	}
373
	read_unlock(&raw_v6_hashinfo.lock);
374 375
}

L
Linus Torvalds 已提交
376 377
static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb)
{
378
	if ((raw6_sk(sk)->checksum || sk->sk_filter) &&
379
	    skb_checksum_complete(skb)) {
W
Wang Chen 已提交
380
		atomic_inc(&sk->sk_drops);
381 382
		kfree_skb(skb);
		return 0;
L
Linus Torvalds 已提交
383 384 385 386
	}

	/* Charge it to the socket. */
	if (sock_queue_rcv_skb(sk,skb)<0) {
W
Wang Chen 已提交
387
		atomic_inc(&sk->sk_drops);
L
Linus Torvalds 已提交
388 389 390 391 392 393 394 395
		kfree_skb(skb);
		return 0;
	}

	return 0;
}

/*
396
 *	This is next to useless...
L
Linus Torvalds 已提交
397
 *	if we demultiplex in network layer we don't need the extra call
398 399
 *	just to queue the skb...
 *	maybe we could have the network decide upon a hint if it
L
Linus Torvalds 已提交
400 401 402 403 404 405 406
 *	should call raw_rcv for demultiplexing
 */
int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
{
	struct inet_sock *inet = inet_sk(sk);
	struct raw6_sock *rp = raw6_sk(sk);

407
	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) {
W
Wang Chen 已提交
408
		atomic_inc(&sk->sk_drops);
409 410 411
		kfree_skb(skb);
		return NET_RX_DROP;
	}
L
Linus Torvalds 已提交
412 413 414 415

	if (!rp->checksum)
		skb->ip_summed = CHECKSUM_UNNECESSARY;

416
	if (skb->ip_summed == CHECKSUM_COMPLETE) {
417
		skb_postpull_rcsum(skb, skb_network_header(skb),
418
				   skb_network_header_len(skb));
419 420
		if (!csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
				     &ipv6_hdr(skb)->daddr,
421
				     skb->len, inet->num, skb->csum))
L
Linus Torvalds 已提交
422 423
			skb->ip_summed = CHECKSUM_UNNECESSARY;
	}
424
	if (!skb_csum_unnecessary(skb))
425 426 427 428
		skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
							 &ipv6_hdr(skb)->daddr,
							 skb->len,
							 inet->num, 0));
L
Linus Torvalds 已提交
429 430

	if (inet->hdrincl) {
431
		if (skb_checksum_complete(skb)) {
W
Wang Chen 已提交
432
			atomic_inc(&sk->sk_drops);
L
Linus Torvalds 已提交
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459
			kfree_skb(skb);
			return 0;
		}
	}

	rawv6_rcv_skb(sk, skb);
	return 0;
}


/*
 *	This should be easy, if there is something there
 *	we return it, otherwise we block.
 */

static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk,
		  struct msghdr *msg, size_t len,
		  int noblock, int flags, int *addr_len)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)msg->msg_name;
	struct sk_buff *skb;
	size_t copied;
	int err;

	if (flags & MSG_OOB)
		return -EOPNOTSUPP;
460 461

	if (addr_len)
L
Linus Torvalds 已提交
462 463 464 465 466 467 468 469 470 471
		*addr_len=sizeof(*sin6);

	if (flags & MSG_ERRQUEUE)
		return ipv6_recv_error(sk, msg, len);

	skb = skb_recv_datagram(sk, flags, noblock, &err);
	if (!skb)
		goto out;

	copied = skb->len;
472 473 474 475
	if (copied > len) {
		copied = len;
		msg->msg_flags |= MSG_TRUNC;
	}
L
Linus Torvalds 已提交
476

477
	if (skb_csum_unnecessary(skb)) {
L
Linus Torvalds 已提交
478 479
		err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
	} else if (msg->msg_flags&MSG_TRUNC) {
480
		if (__skb_checksum_complete(skb))
L
Linus Torvalds 已提交
481 482 483 484 485 486 487 488 489 490 491 492 493
			goto csum_copy_err;
		err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
	} else {
		err = skb_copy_and_csum_datagram_iovec(skb, 0, msg->msg_iov);
		if (err == -EINVAL)
			goto csum_copy_err;
	}
	if (err)
		goto out_free;

	/* Copy the address. */
	if (sin6) {
		sin6->sin6_family = AF_INET6;
494
		sin6->sin6_port = 0;
495
		ipv6_addr_copy(&sin6->sin6_addr, &ipv6_hdr(skb)->saddr);
L
Linus Torvalds 已提交
496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516
		sin6->sin6_flowinfo = 0;
		sin6->sin6_scope_id = 0;
		if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL)
			sin6->sin6_scope_id = IP6CB(skb)->iif;
	}

	sock_recv_timestamp(msg, sk, skb);

	if (np->rxopt.all)
		datagram_recv_ctl(sk, msg, skb);

	err = copied;
	if (flags & MSG_TRUNC)
		err = skb->len;

out_free:
	skb_free_datagram(sk, skb);
out:
	return err;

csum_copy_err:
517
	skb_kill_datagram(sk, skb, flags);
L
Linus Torvalds 已提交
518 519 520 521 522

	/* Error for blocking case is chosen to masquerade
	   as some normal condition.
	 */
	err = (flags&MSG_DONTWAIT) ? -EAGAIN : -EHOSTUNREACH;
W
Wang Chen 已提交
523
	atomic_inc(&sk->sk_drops);
524
	goto out;
L
Linus Torvalds 已提交
525 526 527
}

static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl,
528
				     struct raw6_sock *rp)
L
Linus Torvalds 已提交
529 530 531
{
	struct sk_buff *skb;
	int err = 0;
532 533
	int offset;
	int len;
534
	int total_len;
535 536
	__wsum tmp_csum;
	__sum16 csum;
L
Linus Torvalds 已提交
537 538 539 540 541 542 543

	if (!rp->checksum)
		goto send;

	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
		goto out;

544
	offset = rp->offset;
545 546
	total_len = inet_sk(sk)->cork.length - (skb_network_header(skb) -
						skb->data);
547
	if (offset >= total_len - 1) {
L
Linus Torvalds 已提交
548
		err = -EINVAL;
549
		ip6_flush_pending_frames(sk);
L
Linus Torvalds 已提交
550 551 552 553 554 555 556 557 558 559
		goto out;
	}

	/* should be check HW csum miyazawa */
	if (skb_queue_len(&sk->sk_write_queue) == 1) {
		/*
		 * Only one fragment on the socket.
		 */
		tmp_csum = skb->csum;
	} else {
560
		struct sk_buff *csum_skb = NULL;
L
Linus Torvalds 已提交
561 562 563 564
		tmp_csum = 0;

		skb_queue_walk(&sk->sk_write_queue, skb) {
			tmp_csum = csum_add(tmp_csum, skb->csum);
565 566 567 568

			if (csum_skb)
				continue;

569
			len = skb->len - skb_transport_offset(skb);
570 571 572 573 574 575
			if (offset >= len) {
				offset -= len;
				continue;
			}

			csum_skb = skb;
L
Linus Torvalds 已提交
576
		}
577 578

		skb = csum_skb;
L
Linus Torvalds 已提交
579 580
	}

581
	offset += skb_transport_offset(skb);
582 583 584
	if (skb_copy_bits(skb, offset, &csum, 2))
		BUG();

L
Linus Torvalds 已提交
585
	/* in case cksum was not initialized */
586
	if (unlikely(csum))
587
		tmp_csum = csum_sub(tmp_csum, csum_unfold(csum));
588

589
	csum = csum_ipv6_magic(&fl->fl6_src,
590
				   &fl->fl6_dst,
591
				   total_len, fl->proto, tmp_csum);
592

593 594
	if (csum == 0 && fl->proto == IPPROTO_UDP)
		csum = CSUM_MANGLED_0;
L
Linus Torvalds 已提交
595

596 597
	if (skb_store_bits(skb, offset, &csum, 2))
		BUG();
L
Linus Torvalds 已提交
598 599 600 601 602 603 604 605

send:
	err = ip6_push_pending_frames(sk);
out:
	return err;
}

static int rawv6_send_hdrinc(struct sock *sk, void *from, int length,
606
			struct flowi *fl, struct rt6_info *rt,
L
Linus Torvalds 已提交
607 608
			unsigned int flags)
{
609
	struct ipv6_pinfo *np = inet6_sk(sk);
L
Linus Torvalds 已提交
610 611 612 613 614 615 616 617 618 619 620
	struct ipv6hdr *iph;
	struct sk_buff *skb;
	int err;

	if (length > rt->u.dst.dev->mtu) {
		ipv6_local_error(sk, EMSGSIZE, fl, rt->u.dst.dev->mtu);
		return -EMSGSIZE;
	}
	if (flags&MSG_PROBE)
		goto out;

621 622 623
	skb = sock_alloc_send_skb(sk,
				  length + LL_ALLOCATED_SPACE(rt->u.dst.dev) + 15,
				  flags & MSG_DONTWAIT, &err);
L
Linus Torvalds 已提交
624
	if (skb == NULL)
625
		goto error;
626
	skb_reserve(skb, LL_RESERVED_SPACE(rt->u.dst.dev));
L
Linus Torvalds 已提交
627 628

	skb->priority = sk->sk_priority;
629
	skb->mark = sk->sk_mark;
L
Linus Torvalds 已提交
630 631
	skb->dst = dst_clone(&rt->u.dst);

632 633
	skb_put(skb, length);
	skb_reset_network_header(skb);
634
	iph = ipv6_hdr(skb);
L
Linus Torvalds 已提交
635 636 637

	skb->ip_summed = CHECKSUM_NONE;

638
	skb->transport_header = skb->network_header;
L
Linus Torvalds 已提交
639 640 641 642
	err = memcpy_fromiovecend((void *)iph, from, 0, length);
	if (err)
		goto error_fault;

643
	IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS);
644
	err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
L
Linus Torvalds 已提交
645 646
		      dst_output);
	if (err > 0)
647
		err = np->recverr ? net_xmit_errno(err) : 0;
L
Linus Torvalds 已提交
648 649 650 651 652 653 654 655 656
	if (err)
		goto error;
out:
	return 0;

error_fault:
	err = -EFAULT;
	kfree_skb(skb);
error:
657
	IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
658
	return err;
L
Linus Torvalds 已提交
659 660
}

H
Heiko Carstens 已提交
661
static int rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg)
L
Linus Torvalds 已提交
662 663 664 665
{
	struct iovec *iov;
	u8 __user *type = NULL;
	u8 __user *code = NULL;
666
	u8 len = 0;
L
Linus Torvalds 已提交
667 668 669 670
	int probed = 0;
	int i;

	if (!msg->msg_iov)
H
Heiko Carstens 已提交
671
		return 0;
L
Linus Torvalds 已提交
672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692

	for (i = 0; i < msg->msg_iovlen; i++) {
		iov = &msg->msg_iov[i];
		if (!iov)
			continue;

		switch (fl->proto) {
		case IPPROTO_ICMPV6:
			/* check if one-byte field is readable or not. */
			if (iov->iov_base && iov->iov_len < 1)
				break;

			if (!type) {
				type = iov->iov_base;
				/* check if code field is readable or not. */
				if (iov->iov_len > 1)
					code = type + 1;
			} else if (!code)
				code = iov->iov_base;

			if (type && code) {
H
Heiko Carstens 已提交
693 694 695
				if (get_user(fl->fl_icmp_type, type) ||
				    get_user(fl->fl_icmp_code, code))
					return -EFAULT;
L
Linus Torvalds 已提交
696 697 698
				probed = 1;
			}
			break;
699 700 701 702 703 704
		case IPPROTO_MH:
			if (iov->iov_base && iov->iov_len < 1)
				break;
			/* check if type field is readable or not. */
			if (iov->iov_len > 2 - len) {
				u8 __user *p = iov->iov_base;
H
Heiko Carstens 已提交
705 706
				if (get_user(fl->fl_mh_type, &p[2 - len]))
					return -EFAULT;
707 708 709 710 711
				probed = 1;
			} else
				len += iov->iov_len;

			break;
L
Linus Torvalds 已提交
712 713 714 715 716 717 718
		default:
			probed = 1;
			break;
		}
		if (probed)
			break;
	}
H
Heiko Carstens 已提交
719
	return 0;
L
Linus Torvalds 已提交
720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736
}

static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
		   struct msghdr *msg, size_t len)
{
	struct ipv6_txoptions opt_space;
	struct sockaddr_in6 * sin6 = (struct sockaddr_in6 *) msg->msg_name;
	struct in6_addr *daddr, *final_p = NULL, final;
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct raw6_sock *rp = raw6_sk(sk);
	struct ipv6_txoptions *opt = NULL;
	struct ip6_flowlabel *flowlabel = NULL;
	struct dst_entry *dst = NULL;
	struct flowi fl;
	int addr_len = msg->msg_namelen;
	int hlimit = -1;
737
	int tclass = -1;
L
Linus Torvalds 已提交
738 739 740 741
	u16 proto;
	int err;

	/* Rough check on arithmetic overflow,
742
	   better check is made in ip6_append_data().
L
Linus Torvalds 已提交
743
	 */
744
	if (len > INT_MAX)
L
Linus Torvalds 已提交
745 746 747
		return -EMSGSIZE;

	/* Mirror BSD error message compatibility */
748
	if (msg->msg_flags & MSG_OOB)
L
Linus Torvalds 已提交
749 750 751
		return -EOPNOTSUPP;

	/*
752
	 *	Get and verify the address.
L
Linus Torvalds 已提交
753 754 755
	 */
	memset(&fl, 0, sizeof(fl));

756 757
	fl.mark = sk->sk_mark;

L
Linus Torvalds 已提交
758
	if (sin6) {
759
		if (addr_len < SIN6_LEN_RFC2133)
L
Linus Torvalds 已提交
760 761
			return -EINVAL;

762
		if (sin6->sin6_family && sin6->sin6_family != AF_INET6)
L
Linus Torvalds 已提交
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
			return(-EAFNOSUPPORT);

		/* port is the proto value [0..255] carried in nexthdr */
		proto = ntohs(sin6->sin6_port);

		if (!proto)
			proto = inet->num;
		else if (proto != inet->num)
			return(-EINVAL);

		if (proto > 255)
			return(-EINVAL);

		daddr = &sin6->sin6_addr;
		if (np->sndflow) {
			fl.fl6_flowlabel = sin6->sin6_flowinfo&IPV6_FLOWINFO_MASK;
			if (fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) {
				flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel);
				if (flowlabel == NULL)
					return -EINVAL;
				daddr = &flowlabel->dst;
			}
		}

		/*
		 * Otherwise it will be difficult to maintain
		 * sk->sk_dst_cache.
		 */
		if (sk->sk_state == TCP_ESTABLISHED &&
		    ipv6_addr_equal(daddr, &np->daddr))
			daddr = &np->daddr;

		if (addr_len >= sizeof(struct sockaddr_in6) &&
		    sin6->sin6_scope_id &&
		    ipv6_addr_type(daddr)&IPV6_ADDR_LINKLOCAL)
			fl.oif = sin6->sin6_scope_id;
	} else {
800
		if (sk->sk_state != TCP_ESTABLISHED)
L
Linus Torvalds 已提交
801
			return -EDESTADDRREQ;
802

L
Linus Torvalds 已提交
803 804 805 806 807 808 809 810 811 812 813 814 815
		proto = inet->num;
		daddr = &np->daddr;
		fl.fl6_flowlabel = np->flow_label;
	}

	if (fl.oif == 0)
		fl.oif = sk->sk_bound_dev_if;

	if (msg->msg_controllen) {
		opt = &opt_space;
		memset(opt, 0, sizeof(struct ipv6_txoptions));
		opt->tot_len = sizeof(struct ipv6_txoptions);

816
		err = datagram_send_ctl(sock_net(sk), msg, &fl, opt, &hlimit, &tclass);
L
Linus Torvalds 已提交
817 818 819 820 821 822 823 824 825 826 827 828 829 830
		if (err < 0) {
			fl6_sock_release(flowlabel);
			return err;
		}
		if ((fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) && !flowlabel) {
			flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel);
			if (flowlabel == NULL)
				return -EINVAL;
		}
		if (!(opt->opt_nflen|opt->opt_flen))
			opt = NULL;
	}
	if (opt == NULL)
		opt = np->opt;
831 832 833
	if (flowlabel)
		opt = fl6_merge_options(&opt_space, flowlabel, opt);
	opt = ipv6_fixup_options(&opt_space, opt);
L
Linus Torvalds 已提交
834 835

	fl.proto = proto;
H
Heiko Carstens 已提交
836 837 838
	err = rawv6_probe_proto_opt(&fl, msg);
	if (err)
		goto out;
839

840 841 842 843
	if (!ipv6_addr_any(daddr))
		ipv6_addr_copy(&fl.fl6_dst, daddr);
	else
		fl.fl6_dst.s6_addr[15] = 0x1; /* :: means loopback (BSD'ism) */
L
Linus Torvalds 已提交
844 845 846 847 848 849 850 851 852 853 854 855 856
	if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr))
		ipv6_addr_copy(&fl.fl6_src, &np->saddr);

	/* merge ip6_build_xmit from ip6_output */
	if (opt && opt->srcrt) {
		struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
		ipv6_addr_copy(&final, &fl.fl6_dst);
		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
		final_p = &final;
	}

	if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst))
		fl.oif = np->mcast_oif;
V
Venkat Yekkirala 已提交
857
	security_sk_classify_flow(sk, &fl);
L
Linus Torvalds 已提交
858 859 860 861 862 863 864

	err = ip6_dst_lookup(sk, &dst, &fl);
	if (err)
		goto out;
	if (final_p)
		ipv6_addr_copy(&fl.fl6_dst, final_p);

865
	if ((err = __xfrm_lookup(&dst, &fl, sk, XFRM_LOOKUP_WAIT)) < 0) {
866 867 868 869 870
		if (err == -EREMOTE)
			err = ip6_dst_blackhole(sk, &dst, &fl);
		if (err < 0)
			goto out;
	}
L
Linus Torvalds 已提交
871 872 873 874 875 876 877

	if (hlimit < 0) {
		if (ipv6_addr_is_multicast(&fl.fl6_dst))
			hlimit = np->mcast_hops;
		else
			hlimit = np->hop_limit;
		if (hlimit < 0)
878
			hlimit = ip6_dst_hoplimit(dst);
L
Linus Torvalds 已提交
879 880
	}

881
	if (tclass < 0) {
882
		tclass = np->tclass;
883 884 885 886
		if (tclass < 0)
			tclass = 0;
	}

L
Linus Torvalds 已提交
887 888 889 890 891 892 893 894
	if (msg->msg_flags&MSG_CONFIRM)
		goto do_confirm;

back_from_confirm:
	if (inet->hdrincl) {
		err = rawv6_send_hdrinc(sk, msg->msg_iov, len, &fl, (struct rt6_info*)dst, msg->msg_flags);
	} else {
		lock_sock(sk);
895 896 897
		err = ip6_append_data(sk, ip_generic_getfrag, msg->msg_iov,
			len, 0, hlimit, tclass, opt, &fl, (struct rt6_info*)dst,
			msg->msg_flags);
L
Linus Torvalds 已提交
898 899 900 901

		if (err)
			ip6_flush_pending_frames(sk);
		else if (!(msg->msg_flags & MSG_MORE))
902
			err = rawv6_push_pending_frames(sk, &fl, rp);
903
		release_sock(sk);
L
Linus Torvalds 已提交
904 905
	}
done:
906
	dst_release(dst);
907
out:
L
Linus Torvalds 已提交
908 909 910 911 912 913 914 915 916 917
	fl6_sock_release(flowlabel);
	return err<0?err:len;
do_confirm:
	dst_confirm(dst);
	if (!(msg->msg_flags & MSG_PROBE) || len)
		goto back_from_confirm;
	err = 0;
	goto done;
}

918
static int rawv6_seticmpfilter(struct sock *sk, int level, int optname,
L
Linus Torvalds 已提交
919 920 921 922 923 924 925 926 927 928 929
			       char __user *optval, int optlen)
{
	switch (optname) {
	case ICMPV6_FILTER:
		if (optlen > sizeof(struct icmp6_filter))
			optlen = sizeof(struct icmp6_filter);
		if (copy_from_user(&raw6_sk(sk)->filter, optval, optlen))
			return -EFAULT;
		return 0;
	default:
		return -ENOPROTOOPT;
930
	}
L
Linus Torvalds 已提交
931 932 933 934

	return 0;
}

935
static int rawv6_geticmpfilter(struct sock *sk, int level, int optname,
L
Linus Torvalds 已提交
936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954
			       char __user *optval, int __user *optlen)
{
	int len;

	switch (optname) {
	case ICMPV6_FILTER:
		if (get_user(len, optlen))
			return -EFAULT;
		if (len < 0)
			return -EINVAL;
		if (len > sizeof(struct icmp6_filter))
			len = sizeof(struct icmp6_filter);
		if (put_user(len, optlen))
			return -EFAULT;
		if (copy_to_user(optval, &raw6_sk(sk)->filter, len))
			return -EFAULT;
		return 0;
	default:
		return -ENOPROTOOPT;
955
	}
L
Linus Torvalds 已提交
956 957 958 959 960

	return 0;
}


961
static int do_rawv6_setsockopt(struct sock *sk, int level, int optname,
L
Linus Torvalds 已提交
962 963 964 965 966
			    char __user *optval, int optlen)
{
	struct raw6_sock *rp = raw6_sk(sk);
	int val;

967
	if (get_user(val, (int __user *)optval))
L
Linus Torvalds 已提交
968 969 970 971
		return -EFAULT;

	switch (optname) {
		case IPV6_CHECKSUM:
972 973 974 975 976 977 978 979 980 981 982 983 984
			if (inet_sk(sk)->num == IPPROTO_ICMPV6 &&
			    level == IPPROTO_IPV6) {
				/*
				 * RFC3542 tells that IPV6_CHECKSUM socket
				 * option in the IPPROTO_IPV6 level is not
				 * allowed on ICMPv6 sockets.
				 * If you want to set it, use IPPROTO_RAW
				 * level IPV6_CHECKSUM socket option
				 * (Linux extension).
				 */
				return -EINVAL;
			}

L
Linus Torvalds 已提交
985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003
			/* You may get strange result with a positive odd offset;
			   RFC2292bis agrees with me. */
			if (val > 0 && (val&1))
				return(-EINVAL);
			if (val < 0) {
				rp->checksum = 0;
			} else {
				rp->checksum = 1;
				rp->offset = val;
			}

			return 0;
			break;

		default:
			return(-ENOPROTOOPT);
	}
}

1004 1005
static int rawv6_setsockopt(struct sock *sk, int level, int optname,
			  char __user *optval, int optlen)
L
Linus Torvalds 已提交
1006 1007 1008 1009 1010 1011 1012 1013
{
	switch(level) {
		case SOL_RAW:
			break;

		case SOL_ICMPV6:
			if (inet_sk(sk)->num != IPPROTO_ICMPV6)
				return -EOPNOTSUPP;
1014
			return rawv6_seticmpfilter(sk, level, optname, optval,
L
Linus Torvalds 已提交
1015 1016 1017 1018 1019
						   optlen);
		case SOL_IPV6:
			if (optname == IPV6_CHECKSUM)
				break;
		default:
1020
			return ipv6_setsockopt(sk, level, optname, optval,
L
Linus Torvalds 已提交
1021
					       optlen);
1022 1023
	}

1024 1025 1026 1027 1028
	return do_rawv6_setsockopt(sk, level, optname, optval, optlen);
}

#ifdef CONFIG_COMPAT
static int compat_rawv6_setsockopt(struct sock *sk, int level, int optname,
1029
				   char __user *optval, int optlen)
1030
{
1031 1032 1033 1034 1035 1036 1037 1038 1039
	switch (level) {
	case SOL_RAW:
		break;
	case SOL_ICMPV6:
		if (inet_sk(sk)->num != IPPROTO_ICMPV6)
			return -EOPNOTSUPP;
		return rawv6_seticmpfilter(sk, level, optname, optval, optlen);
	case SOL_IPV6:
		if (optname == IPV6_CHECKSUM)
1040
			break;
1041 1042 1043
	default:
		return compat_ipv6_setsockopt(sk, level, optname,
					      optval, optlen);
1044
	}
1045 1046 1047 1048 1049 1050 1051 1052 1053
	return do_rawv6_setsockopt(sk, level, optname, optval, optlen);
}
#endif

static int do_rawv6_getsockopt(struct sock *sk, int level, int optname,
			    char __user *optval, int __user *optlen)
{
	struct raw6_sock *rp = raw6_sk(sk);
	int val, len;
L
Linus Torvalds 已提交
1054 1055 1056 1057 1058 1059

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

	switch (optname) {
	case IPV6_CHECKSUM:
1060 1061 1062 1063 1064
		/*
		 * We allow getsockopt() for IPPROTO_IPV6-level
		 * IPV6_CHECKSUM socket option on ICMPv6 sockets
		 * since RFC3542 is silent about it.
		 */
L
Linus Torvalds 已提交
1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083
		if (rp->checksum == 0)
			val = -1;
		else
			val = rp->offset;
		break;

	default:
		return -ENOPROTOOPT;
	}

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

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

1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101
static int rawv6_getsockopt(struct sock *sk, int level, int optname,
			  char __user *optval, int __user *optlen)
{
	switch(level) {
		case SOL_RAW:
			break;

		case SOL_ICMPV6:
			if (inet_sk(sk)->num != IPPROTO_ICMPV6)
				return -EOPNOTSUPP;
			return rawv6_geticmpfilter(sk, level, optname, optval,
						   optlen);
		case SOL_IPV6:
			if (optname == IPV6_CHECKSUM)
				break;
		default:
			return ipv6_getsockopt(sk, level, optname, optval,
					       optlen);
1102 1103
	}

1104 1105 1106 1107 1108
	return do_rawv6_getsockopt(sk, level, optname, optval, optlen);
}

#ifdef CONFIG_COMPAT
static int compat_rawv6_getsockopt(struct sock *sk, int level, int optname,
1109
				   char __user *optval, int __user *optlen)
1110
{
1111 1112 1113 1114 1115 1116 1117 1118 1119
	switch (level) {
	case SOL_RAW:
		break;
	case SOL_ICMPV6:
		if (inet_sk(sk)->num != IPPROTO_ICMPV6)
			return -EOPNOTSUPP;
		return rawv6_geticmpfilter(sk, level, optname, optval, optlen);
	case SOL_IPV6:
		if (optname == IPV6_CHECKSUM)
1120
			break;
1121 1122 1123
	default:
		return compat_ipv6_getsockopt(sk, level, optname,
					      optval, optlen);
1124
	}
1125 1126 1127 1128
	return do_rawv6_getsockopt(sk, level, optname, optval, optlen);
}
#endif

L
Linus Torvalds 已提交
1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141
static int rawv6_ioctl(struct sock *sk, int cmd, unsigned long arg)
{
	switch(cmd) {
		case SIOCOUTQ:
		{
			int amount = atomic_read(&sk->sk_wmem_alloc);
			return put_user(amount, (int __user *)arg);
		}
		case SIOCINQ:
		{
			struct sk_buff *skb;
			int amount = 0;

1142
			spin_lock_bh(&sk->sk_receive_queue.lock);
L
Linus Torvalds 已提交
1143 1144
			skb = skb_peek(&sk->sk_receive_queue);
			if (skb != NULL)
1145
				amount = skb->tail - skb->transport_header;
1146
			spin_unlock_bh(&sk->sk_receive_queue.lock);
L
Linus Torvalds 已提交
1147 1148 1149 1150
			return put_user(amount, (int __user *)arg);
		}

		default:
1151 1152 1153
#ifdef CONFIG_IPV6_MROUTE
			return ip6mr_ioctl(sk, cmd, (void __user *)arg);
#else
L
Linus Torvalds 已提交
1154
			return -ENOIOCTLCMD;
1155
#endif
L
Linus Torvalds 已提交
1156 1157 1158 1159 1160 1161 1162
	}
}

static void rawv6_close(struct sock *sk, long timeout)
{
	if (inet_sk(sk)->num == IPPROTO_RAW)
		ip6_ra_control(sk, -1, NULL);
1163
	ip6mr_sk_done(sk);
L
Linus Torvalds 已提交
1164 1165 1166 1167 1168
	sk_common_release(sk);
}

static int rawv6_init_sk(struct sock *sk)
{
1169 1170 1171 1172
	struct raw6_sock *rp = raw6_sk(sk);

	switch (inet_sk(sk)->num) {
	case IPPROTO_ICMPV6:
L
Linus Torvalds 已提交
1173 1174
		rp->checksum = 1;
		rp->offset   = 2;
1175 1176 1177 1178 1179 1180 1181
		break;
	case IPPROTO_MH:
		rp->checksum = 1;
		rp->offset   = 4;
		break;
	default:
		break;
L
Linus Torvalds 已提交
1182 1183 1184 1185 1186
	}
	return(0);
}

struct proto rawv6_prot = {
1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
	.name		   = "RAWv6",
	.owner		   = THIS_MODULE,
	.close		   = rawv6_close,
	.connect	   = ip6_datagram_connect,
	.disconnect	   = udp_disconnect,
	.ioctl		   = rawv6_ioctl,
	.init		   = rawv6_init_sk,
	.destroy	   = inet6_destroy_sock,
	.setsockopt	   = rawv6_setsockopt,
	.getsockopt	   = rawv6_getsockopt,
	.sendmsg	   = rawv6_sendmsg,
	.recvmsg	   = rawv6_recvmsg,
	.bind		   = rawv6_bind,
	.backlog_rcv	   = rawv6_rcv_skb,
1201 1202
	.hash		   = raw_hash_sk,
	.unhash		   = raw_unhash_sk,
1203
	.obj_size	   = sizeof(struct raw6_sock),
1204
	.h.raw_hash	   = &raw_v6_hashinfo,
1205
#ifdef CONFIG_COMPAT
1206 1207
	.compat_setsockopt = compat_rawv6_setsockopt,
	.compat_getsockopt = compat_rawv6_getsockopt,
1208
#endif
L
Linus Torvalds 已提交
1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
};

#ifdef CONFIG_PROC_FS
static void raw6_sock_seq_show(struct seq_file *seq, struct sock *sp, int i)
{
	struct ipv6_pinfo *np = inet6_sk(sp);
	struct in6_addr *dest, *src;
	__u16 destp, srcp;

	dest  = &np->daddr;
	src   = &np->rcv_saddr;
	destp = 0;
	srcp  = inet_sk(sp)->num;
	seq_printf(seq,
		   "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
W
Wang Chen 已提交
1224
		   "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %d\n",
L
Linus Torvalds 已提交
1225 1226 1227 1228 1229
		   i,
		   src->s6_addr32[0], src->s6_addr32[1],
		   src->s6_addr32[2], src->s6_addr32[3], srcp,
		   dest->s6_addr32[0], dest->s6_addr32[1],
		   dest->s6_addr32[2], dest->s6_addr32[3], destp,
1230
		   sp->sk_state,
L
Linus Torvalds 已提交
1231 1232 1233 1234 1235
		   atomic_read(&sp->sk_wmem_alloc),
		   atomic_read(&sp->sk_rmem_alloc),
		   0, 0L, 0,
		   sock_i_uid(sp), 0,
		   sock_i_ino(sp),
W
Wang Chen 已提交
1236
		   atomic_read(&sp->sk_refcnt), sp, atomic_read(&sp->sk_drops));
L
Linus Torvalds 已提交
1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
}

static int raw6_seq_show(struct seq_file *seq, void *v)
{
	if (v == SEQ_START_TOKEN)
		seq_printf(seq,
			   "  sl  "
			   "local_address                         "
			   "remote_address                        "
			   "st tx_queue rx_queue tr tm->when retrnsmt"
W
Wang Chen 已提交
1247
			   "   uid  timeout inode  drops\n");
L
Linus Torvalds 已提交
1248
	else
1249
		raw6_sock_seq_show(seq, v, raw_seq_private(seq)->bucket);
L
Linus Torvalds 已提交
1250 1251 1252
	return 0;
}

1253
static const struct seq_operations raw6_seq_ops = {
1254 1255 1256
	.start =	raw_seq_start,
	.next =		raw_seq_next,
	.stop =		raw_seq_stop,
L
Linus Torvalds 已提交
1257 1258 1259 1260 1261
	.show =		raw6_seq_show,
};

static int raw6_seq_open(struct inode *inode, struct file *file)
{
1262
	return raw_seq_open(inode, file, &raw_v6_hashinfo, &raw6_seq_ops);
L
Linus Torvalds 已提交
1263 1264
}

1265
static const struct file_operations raw6_seq_fops = {
L
Linus Torvalds 已提交
1266 1267 1268 1269
	.owner =	THIS_MODULE,
	.open =		raw6_seq_open,
	.read =		seq_read,
	.llseek =	seq_lseek,
1270
	.release =	seq_release_net,
L
Linus Torvalds 已提交
1271 1272
};

1273
static int raw6_init_net(struct net *net)
L
Linus Torvalds 已提交
1274
{
1275
	if (!proc_net_fops_create(net, "raw6", S_IRUGO, &raw6_seq_fops))
L
Linus Torvalds 已提交
1276
		return -ENOMEM;
1277

L
Linus Torvalds 已提交
1278 1279 1280
	return 0;
}

1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295
static void raw6_exit_net(struct net *net)
{
	proc_net_remove(net, "raw6");
}

static struct pernet_operations raw6_net_ops = {
	.init = raw6_init_net,
	.exit = raw6_exit_net,
};

int __init raw6_proc_init(void)
{
	return register_pernet_subsys(&raw6_net_ops);
}

L
Linus Torvalds 已提交
1296 1297
void raw6_proc_exit(void)
{
1298
	unregister_pernet_subsys(&raw6_net_ops);
L
Linus Torvalds 已提交
1299 1300
}
#endif	/* CONFIG_PROC_FS */
1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348

/* Same as inet6_dgram_ops, sans udp_poll.  */
static const struct proto_ops inet6_sockraw_ops = {
	.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		   = datagram_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,
#ifdef CONFIG_COMPAT
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
#endif
};

static struct inet_protosw rawv6_protosw = {
	.type		= SOCK_RAW,
	.protocol	= IPPROTO_IP,	/* wild card */
	.prot		= &rawv6_prot,
	.ops		= &inet6_sockraw_ops,
	.capability	= CAP_NET_RAW,
	.no_check	= UDP_CSUM_DEFAULT,
	.flags		= INET_PROTOSW_REUSE,
};

int __init rawv6_init(void)
{
	int ret;

	ret = inet6_register_protosw(&rawv6_protosw);
	if (ret)
		goto out;
out:
	return ret;
}

1349
void rawv6_exit(void)
1350 1351 1352
{
	inet6_unregister_protosw(&rawv6_protosw);
}