ipv6.h 28.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 *	Linux INET6 implementation
 *
 *	Authors:
 *	Pedro Roque		<roque@di.fc.ul.pt>
 *
 *	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.
 */

#ifndef _NET_IPV6_H
#define _NET_IPV6_H

#include <linux/ipv6.h>
#include <linux/hardirq.h>
E
Eric Dumazet 已提交
18
#include <linux/jhash.h>
19
#include <net/if_inet6.h>
L
Linus Torvalds 已提交
20 21
#include <net/ndisc.h>
#include <net/flow.h>
22
#include <net/flow_dissector.h>
L
Linus Torvalds 已提交
23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
#include <net/snmp.h>

#define SIN6_LEN_RFC2133	24

#define IPV6_MAXPLEN		65535

/*
 *	NextHeader field of IPv6 header
 */

#define NEXTHDR_HOP		0	/* Hop-by-hop option header. */
#define NEXTHDR_TCP		6	/* TCP segment. */
#define NEXTHDR_UDP		17	/* UDP message. */
#define NEXTHDR_IPV6		41	/* IPv6 in IPv6 */
#define NEXTHDR_ROUTING		43	/* Routing header. */
#define NEXTHDR_FRAGMENT	44	/* Fragmentation/reassembly header. */
X
xeb@mail.ru 已提交
39
#define NEXTHDR_GRE		47	/* GRE header. */
L
Linus Torvalds 已提交
40 41 42 43 44
#define NEXTHDR_ESP		50	/* Encapsulating security payload. */
#define NEXTHDR_AUTH		51	/* Authentication header. */
#define NEXTHDR_ICMP		58	/* ICMP for IPv6. */
#define NEXTHDR_NONE		59	/* No next header */
#define NEXTHDR_DEST		60	/* Destination options header. */
J
Joe Stringer 已提交
45
#define NEXTHDR_SCTP		132	/* SCTP message. */
46
#define NEXTHDR_MOBILITY	135	/* Mobility header. */
L
Linus Torvalds 已提交
47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90

#define NEXTHDR_MAX		255

#define IPV6_DEFAULT_HOPLIMIT   64
#define IPV6_DEFAULT_MCASTHOPS	1

/*
 *	Addr type
 *	
 *	type	-	unicast | multicast
 *	scope	-	local	| site	    | global
 *	v4	-	compat
 *	v4mapped
 *	any
 *	loopback
 */

#define IPV6_ADDR_ANY		0x0000U

#define IPV6_ADDR_UNICAST      	0x0001U	
#define IPV6_ADDR_MULTICAST    	0x0002U	

#define IPV6_ADDR_LOOPBACK	0x0010U
#define IPV6_ADDR_LINKLOCAL	0x0020U
#define IPV6_ADDR_SITELOCAL	0x0040U

#define IPV6_ADDR_COMPATv4	0x0080U

#define IPV6_ADDR_SCOPE_MASK	0x00f0U

#define IPV6_ADDR_MAPPED	0x1000U

/*
 *	Addr scopes
 */
#define IPV6_ADDR_MC_SCOPE(a)	\
	((a)->s6_addr[1] & 0x0f)	/* nonstandard */
#define __IPV6_ADDR_SCOPE_INVALID	-1
#define IPV6_ADDR_SCOPE_NODELOCAL	0x01
#define IPV6_ADDR_SCOPE_LINKLOCAL	0x02
#define IPV6_ADDR_SCOPE_SITELOCAL	0x05
#define IPV6_ADDR_SCOPE_ORGLOCAL	0x08
#define IPV6_ADDR_SCOPE_GLOBAL		0x0e

91 92 93 94 95 96 97 98 99 100
/*
 *	Addr flags
 */
#define IPV6_ADDR_MC_FLAG_TRANSIENT(a)	\
	((a)->s6_addr[1] & 0x10)
#define IPV6_ADDR_MC_FLAG_PREFIX(a)	\
	((a)->s6_addr[1] & 0x20)
#define IPV6_ADDR_MC_FLAG_RENDEZVOUS(a)	\
	((a)->s6_addr[1] & 0x40)

L
Linus Torvalds 已提交
101 102 103 104 105
/*
 *	fragmentation header
 */

struct frag_hdr {
A
Al Viro 已提交
106 107 108 109
	__u8	nexthdr;
	__u8	reserved;
	__be16	frag_off;
	__be32	identification;
L
Linus Torvalds 已提交
110 111
};

112 113
#define	IP6_MF		0x0001
#define	IP6_OFFSET	0xFFF8
L
Linus Torvalds 已提交
114

115 116 117
#define IP6_REPLY_MARK(net, mark) \
	((net)->ipv6.sysctl.fwmark_reflect ? (mark) : 0)

L
Linus Torvalds 已提交
118 119 120 121
#include <net/sock.h>

/* sysctls */
extern int sysctl_mld_max_msf;
122
extern int sysctl_mld_qrv;
123

124
#define _DEVINC(net, statname, modifier, idev, field)			\
125
({									\
126 127
	struct inet6_dev *_idev = (idev);				\
	if (likely(_idev != NULL))					\
128
		SNMP_INC_STATS##modifier((_idev)->stats.statname, (field)); \
129
	SNMP_INC_STATS##modifier((net)->mib.statname##_statistics, (field));\
130
})
131

132 133 134 135 136 137 138 139 140
/* per device counters are atomic_long_t */
#define _DEVINCATOMIC(net, statname, modifier, idev, field)		\
({									\
	struct inet6_dev *_idev = (idev);				\
	if (likely(_idev != NULL))					\
		SNMP_INC_STATS_ATOMIC_LONG((_idev)->stats.statname##dev, (field)); \
	SNMP_INC_STATS##modifier((net)->mib.statname##_statistics, (field));\
})

141 142 143 144 145 146 147 148 149
/* per device and per net counters are atomic_long_t */
#define _DEVINC_ATOMIC_ATOMIC(net, statname, idev, field)		\
({									\
	struct inet6_dev *_idev = (idev);				\
	if (likely(_idev != NULL))					\
		SNMP_INC_STATS_ATOMIC_LONG((_idev)->stats.statname##dev, (field)); \
	SNMP_INC_STATS_ATOMIC_LONG((net)->mib.statname##_statistics, (field));\
})

150
#define _DEVADD(net, statname, modifier, idev, field, val)		\
151 152 153 154
({									\
	struct inet6_dev *_idev = (idev);				\
	if (likely(_idev != NULL))					\
		SNMP_ADD_STATS##modifier((_idev)->stats.statname, (field), (val)); \
155
	SNMP_ADD_STATS##modifier((net)->mib.statname##_statistics, (field), (val));\
156 157
})

158 159 160 161 162 163 164 165
#define _DEVUPD(net, statname, modifier, idev, field, val)		\
({									\
	struct inet6_dev *_idev = (idev);				\
	if (likely(_idev != NULL))					\
		SNMP_UPD_PO_STATS##modifier((_idev)->stats.statname, field, (val)); \
	SNMP_UPD_PO_STATS##modifier((net)->mib.statname##_statistics, field, (val));\
})

166 167
/* MIBs */

168
#define IP6_INC_STATS(net, idev,field)		\
169
		_DEVINC(net, ipv6, 64, idev, field)
170
#define IP6_INC_STATS_BH(net, idev,field)	\
171
		_DEVINC(net, ipv6, 64_BH, idev, field)
172
#define IP6_ADD_STATS(net, idev,field,val)	\
173
		_DEVADD(net, ipv6, 64, idev, field, val)
174
#define IP6_ADD_STATS_BH(net, idev,field,val)	\
175
		_DEVADD(net, ipv6, 64_BH, idev, field, val)
176
#define IP6_UPD_PO_STATS(net, idev,field,val)   \
177
		_DEVUPD(net, ipv6, 64, idev, field, val)
178
#define IP6_UPD_PO_STATS_BH(net, idev,field,val)   \
179
		_DEVUPD(net, ipv6, 64_BH, idev, field, val)
180
#define ICMP6_INC_STATS(net, idev, field)	\
181
		_DEVINCATOMIC(net, icmpv6, , idev, field)
182
#define ICMP6_INC_STATS_BH(net, idev, field)	\
183
		_DEVINCATOMIC(net, icmpv6, _BH, idev, field)
184 185

#define ICMP6MSGOUT_INC_STATS(net, idev, field)		\
186
	_DEVINC_ATOMIC_ATOMIC(net, icmpv6msg, idev, field +256)
187
#define ICMP6MSGOUT_INC_STATS_BH(net, idev, field)	\
188
	_DEVINC_ATOMIC_ATOMIC(net, icmpv6msg, idev, field +256)
189
#define ICMP6MSGIN_INC_STATS_BH(net, idev, field)	\
190
	_DEVINC_ATOMIC_ATOMIC(net, icmpv6msg, idev, field)
191

E
Eric Dumazet 已提交
192
struct ip6_ra_chain {
L
Linus Torvalds 已提交
193 194 195 196 197 198 199 200 201 202 203 204 205 206
	struct ip6_ra_chain	*next;
	struct sock		*sk;
	int			sel;
	void			(*destructor)(struct sock *);
};

extern struct ip6_ra_chain	*ip6_ra_chain;
extern rwlock_t ip6_ra_lock;

/*
   This structure is prepared by protocol, when parsing
   ancillary data and passed to IPv6.
 */

E
Eric Dumazet 已提交
207
struct ipv6_txoptions {
208
	atomic_t		refcnt;
L
Linus Torvalds 已提交
209 210 211 212 213 214 215 216 217 218 219 220
	/* Length of this structure */
	int			tot_len;

	/* length of extension headers   */

	__u16			opt_flen;	/* after fragment hdr */
	__u16			opt_nflen;	/* before fragment hdr */

	struct ipv6_opt_hdr	*hopopt;
	struct ipv6_opt_hdr	*dst0opt;
	struct ipv6_rt_hdr	*srcrt;	/* Routing Header */
	struct ipv6_opt_hdr	*dst1opt;
221
	struct rcu_head		rcu;
L
Linus Torvalds 已提交
222 223 224
	/* Option buffer, as read by IPV6_PKTOPTIONS, starts here. */
};

E
Eric Dumazet 已提交
225
struct ip6_flowlabel {
226
	struct ip6_flowlabel __rcu *next;
A
Al Viro 已提交
227
	__be32			label;
228
	atomic_t		users;
L
Linus Torvalds 已提交
229 230 231
	struct in6_addr		dst;
	struct ipv6_txoptions	*opt;
	unsigned long		linger;
232
	struct rcu_head		rcu;
L
Linus Torvalds 已提交
233
	u8			share;
234 235 236 237
	union {
		struct pid *pid;
		kuid_t uid;
	} owner;
L
Linus Torvalds 已提交
238 239
	unsigned long		lastuse;
	unsigned long		expires;
240
	struct net		*fl_net;
L
Linus Torvalds 已提交
241 242
};

T
Tom Herbert 已提交
243 244 245 246
#define IPV6_FLOWINFO_MASK		cpu_to_be32(0x0FFFFFFF)
#define IPV6_FLOWLABEL_MASK		cpu_to_be32(0x000FFFFF)
#define IPV6_FLOWLABEL_STATELESS_FLAG	cpu_to_be32(0x00080000)

247
#define IPV6_TCLASS_MASK (IPV6_FLOWINFO_MASK & ~IPV6_FLOWLABEL_MASK)
248
#define IPV6_TCLASS_SHIFT	20
L
Linus Torvalds 已提交
249

E
Eric Dumazet 已提交
250
struct ipv6_fl_socklist {
251 252 253
	struct ipv6_fl_socklist	__rcu	*next;
	struct ip6_flowlabel		*fl;
	struct rcu_head			rcu;
L
Linus Torvalds 已提交
254 255
};

256 257 258 259 260 261
static inline struct ipv6_txoptions *txopt_get(const struct ipv6_pinfo *np)
{
	struct ipv6_txoptions *opt;

	rcu_read_lock();
	opt = rcu_dereference(np->opt);
262 263 264 265 266 267
	if (opt) {
		if (!atomic_inc_not_zero(&opt->refcnt))
			opt = NULL;
		else
			opt = rcu_pointer_handoff(opt);
	}
268 269 270 271 272 273 274 275 276 277
	rcu_read_unlock();
	return opt;
}

static inline void txopt_put(struct ipv6_txoptions *opt)
{
	if (opt && atomic_dec_and_test(&opt->refcnt))
		kfree_rcu(opt, rcu);
}

278 279 280 281 282 283
struct ip6_flowlabel *fl6_sock_lookup(struct sock *sk, __be32 label);
struct ipv6_txoptions *fl6_merge_options(struct ipv6_txoptions *opt_space,
					 struct ip6_flowlabel *fl,
					 struct ipv6_txoptions *fopt);
void fl6_free_socklist(struct sock *sk);
int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen);
284 285
int ipv6_flowlabel_opt_get(struct sock *sk, struct in6_flowlabel_req *freq,
			   int flags);
286 287
int ip6_flowlabel_init(void);
void ip6_flowlabel_cleanup(void);
L
Linus Torvalds 已提交
288 289 290 291 292 293 294

static inline void fl6_sock_release(struct ip6_flowlabel *fl)
{
	if (fl)
		atomic_dec(&fl->users);
}

295
void icmpv6_notify(struct sk_buff *skb, u8 type, u8 code, __be32 info);
296

297 298 299
int icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6,
			       struct icmp6hdr *thdr, int len);

300
int ip6_ra_control(struct sock *sk, int sel);
L
Linus Torvalds 已提交
301

302
int ipv6_parse_hopopts(struct sk_buff *skb);
L
Linus Torvalds 已提交
303

304 305 306 307 308 309 310
struct ipv6_txoptions *ipv6_dup_options(struct sock *sk,
					struct ipv6_txoptions *opt);
struct ipv6_txoptions *ipv6_renew_options(struct sock *sk,
					  struct ipv6_txoptions *opt,
					  int newtype,
					  struct ipv6_opt_hdr __user *newopt,
					  int newoptlen);
311 312
struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
					  struct ipv6_txoptions *opt);
L
Linus Torvalds 已提交
313

314 315
bool ipv6_opt_accepted(const struct sock *sk, const struct sk_buff *skb,
		       const struct inet6_skb_parm *opt);
316

317 318 319 320 321 322 323 324 325
static inline bool ipv6_accept_ra(struct inet6_dev *idev)
{
	/* If forwarding is enabled, RA are not accepted unless the special
	 * hybrid mode (accept_ra=2) is enabled.
	 */
	return idev->cnf.forwarding ? idev->cnf.accept_ra == 2 :
	    idev->cnf.accept_ra;
}

326 327 328
#if IS_ENABLED(CONFIG_IPV6)
static inline int ip6_frag_mem(struct net *net)
{
329
	return sum_frag_mem_limit(&net->ipv6.frags);
330 331
}
#endif
L
Linus Torvalds 已提交
332

333 334
#define IPV6_FRAG_HIGH_THRESH	(4 * 1024*1024)	/* 4194304 */
#define IPV6_FRAG_LOW_THRESH	(3 * 1024*1024)	/* 3145728 */
335
#define IPV6_FRAG_TIMEOUT	(60 * HZ)	/* 60 seconds */
L
Linus Torvalds 已提交
336

337
int __ipv6_addr_type(const struct in6_addr *addr);
338 339 340 341
static inline int ipv6_addr_type(const struct in6_addr *addr)
{
	return __ipv6_addr_type(addr) & 0xffff;
}
L
Linus Torvalds 已提交
342 343 344

static inline int ipv6_addr_scope(const struct in6_addr *addr)
{
345 346 347 348 349
	return __ipv6_addr_type(addr) & IPV6_ADDR_SCOPE_MASK;
}

static inline int __ipv6_addr_src_scope(int type)
{
E
Eric Dumazet 已提交
350
	return (type == IPV6_ADDR_ANY) ? __IPV6_ADDR_SCOPE_INVALID : (type >> 16);
351 352 353 354 355
}

static inline int ipv6_addr_src_scope(const struct in6_addr *addr)
{
	return __ipv6_addr_src_scope(__ipv6_addr_type(addr));
L
Linus Torvalds 已提交
356 357
}

358 359 360 361 362 363 364 365 366 367 368 369
static inline bool __ipv6_addr_needs_scope_id(int type)
{
	return type & IPV6_ADDR_LINKLOCAL ||
	       (type & IPV6_ADDR_MULTICAST &&
		(type & (IPV6_ADDR_LOOPBACK|IPV6_ADDR_LINKLOCAL)));
}

static inline __u32 ipv6_iface_scope_id(const struct in6_addr *addr, int iface)
{
	return __ipv6_addr_needs_scope_id(__ipv6_addr_type(addr)) ? iface : 0;
}

L
Linus Torvalds 已提交
370 371
static inline int ipv6_addr_cmp(const struct in6_addr *a1, const struct in6_addr *a2)
{
372
	return memcmp(a1, a2, sizeof(struct in6_addr));
L
Linus Torvalds 已提交
373 374
}

375
static inline bool
376 377 378
ipv6_masked_addr_cmp(const struct in6_addr *a1, const struct in6_addr *m,
		     const struct in6_addr *a2)
{
379 380 381 382 383 384 385 386
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
	const unsigned long *ul1 = (const unsigned long *)a1;
	const unsigned long *ulm = (const unsigned long *)m;
	const unsigned long *ul2 = (const unsigned long *)a2;

	return !!(((ul1[0] ^ ul2[0]) & ulm[0]) |
		  ((ul1[1] ^ ul2[1]) & ulm[1]));
#else
E
Eric Dumazet 已提交
387 388 389 390
	return !!(((a1->s6_addr32[0] ^ a2->s6_addr32[0]) & m->s6_addr32[0]) |
		  ((a1->s6_addr32[1] ^ a2->s6_addr32[1]) & m->s6_addr32[1]) |
		  ((a1->s6_addr32[2] ^ a2->s6_addr32[2]) & m->s6_addr32[2]) |
		  ((a1->s6_addr32[3] ^ a2->s6_addr32[3]) & m->s6_addr32[3]));
391
#endif
392 393
}

L
Linus Torvalds 已提交
394 395 396 397 398 399 400 401
static inline void ipv6_addr_prefix(struct in6_addr *pfx, 
				    const struct in6_addr *addr,
				    int plen)
{
	/* caller must guarantee 0 <= plen <= 128 */
	int o = plen >> 3,
	    b = plen & 0x7;

402
	memset(pfx->s6_addr, 0, sizeof(pfx->s6_addr));
L
Linus Torvalds 已提交
403
	memcpy(pfx->s6_addr, addr, o);
404
	if (b != 0)
L
Linus Torvalds 已提交
405 406 407
		pfx->s6_addr[o] = addr->s6_addr[o] & (0xff00 >> b);
}

A
Alexander Aring 已提交
408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
static inline void ipv6_addr_prefix_copy(struct in6_addr *addr,
					 const struct in6_addr *pfx,
					 int plen)
{
	/* caller must guarantee 0 <= plen <= 128 */
	int o = plen >> 3,
	    b = plen & 0x7;

	memcpy(addr->s6_addr, pfx, o);
	if (b != 0) {
		addr->s6_addr[o] &= ~(0xff00 >> b);
		addr->s6_addr[o] |= (pfx->s6_addr[o] & (0xff00 >> b));
	}
}

423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
static inline void __ipv6_addr_set_half(__be32 *addr,
					__be32 wh, __be32 wl)
{
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
#if defined(__BIG_ENDIAN)
	if (__builtin_constant_p(wh) && __builtin_constant_p(wl)) {
		*(__force u64 *)addr = ((__force u64)(wh) << 32 | (__force u64)(wl));
		return;
	}
#elif defined(__LITTLE_ENDIAN)
	if (__builtin_constant_p(wl) && __builtin_constant_p(wh)) {
		*(__force u64 *)addr = ((__force u64)(wl) << 32 | (__force u64)(wh));
		return;
	}
#endif
#endif
	addr[0] = wh;
	addr[1] = wl;
}

L
Linus Torvalds 已提交
443
static inline void ipv6_addr_set(struct in6_addr *addr, 
A
Al Viro 已提交
444 445
				     __be32 w1, __be32 w2,
				     __be32 w3, __be32 w4)
L
Linus Torvalds 已提交
446
{
447 448
	__ipv6_addr_set_half(&addr->s6_addr32[0], w1, w2);
	__ipv6_addr_set_half(&addr->s6_addr32[2], w3, w4);
L
Linus Torvalds 已提交
449 450
}

E
Eric Dumazet 已提交
451 452
static inline bool ipv6_addr_equal(const struct in6_addr *a1,
				   const struct in6_addr *a2)
L
Linus Torvalds 已提交
453
{
454 455 456 457 458 459
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
	const unsigned long *ul1 = (const unsigned long *)a1;
	const unsigned long *ul2 = (const unsigned long *)a2;

	return ((ul1[0] ^ ul2[0]) | (ul1[1] ^ ul2[1])) == 0UL;
#else
E
Eric Dumazet 已提交
460 461 462 463
	return ((a1->s6_addr32[0] ^ a2->s6_addr32[0]) |
		(a1->s6_addr32[1] ^ a2->s6_addr32[1]) |
		(a1->s6_addr32[2] ^ a2->s6_addr32[2]) |
		(a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0;
464
#endif
L
Linus Torvalds 已提交
465 466
}

467 468 469 470 471
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
static inline bool __ipv6_prefix_equal64_half(const __be64 *a1,
					      const __be64 *a2,
					      unsigned int len)
{
472
	if (len && ((*a1 ^ *a2) & cpu_to_be64((~0UL) << (64 - len))))
473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491
		return false;
	return true;
}

static inline bool ipv6_prefix_equal(const struct in6_addr *addr1,
				     const struct in6_addr *addr2,
				     unsigned int prefixlen)
{
	const __be64 *a1 = (const __be64 *)addr1;
	const __be64 *a2 = (const __be64 *)addr2;

	if (prefixlen >= 64) {
		if (a1[0] ^ a2[0])
			return false;
		return __ipv6_prefix_equal64_half(a1 + 1, a2 + 1, prefixlen - 64);
	}
	return __ipv6_prefix_equal64_half(a1, a2, prefixlen);
}
#else
492 493 494
static inline bool ipv6_prefix_equal(const struct in6_addr *addr1,
				     const struct in6_addr *addr2,
				     unsigned int prefixlen)
L
Linus Torvalds 已提交
495
{
496 497
	const __be32 *a1 = addr1->s6_addr32;
	const __be32 *a2 = addr2->s6_addr32;
498
	unsigned int pdw, pbi;
L
Linus Torvalds 已提交
499 500 501 502

	/* check complete u32 in prefix */
	pdw = prefixlen >> 5;
	if (pdw && memcmp(a1, a2, pdw << 2))
E
Eric Dumazet 已提交
503
		return false;
L
Linus Torvalds 已提交
504 505 506 507

	/* check incomplete u32 in prefix */
	pbi = prefixlen & 0x1f;
	if (pbi && ((a1[pdw] ^ a2[pdw]) & htonl((0xffffffff) << (32 - pbi))))
E
Eric Dumazet 已提交
508
		return false;
L
Linus Torvalds 已提交
509

E
Eric Dumazet 已提交
510
	return true;
L
Linus Torvalds 已提交
511
}
512
#endif
L
Linus Torvalds 已提交
513

514 515
struct inet_frag_queue;

516 517 518
enum ip6_defrag_users {
	IP6_DEFRAG_LOCAL_DELIVER,
	IP6_DEFRAG_CONNTRACK_IN,
519
	__IP6_DEFRAG_CONNTRACK_IN	= IP6_DEFRAG_CONNTRACK_IN + USHRT_MAX,
520
	IP6_DEFRAG_CONNTRACK_OUT,
521
	__IP6_DEFRAG_CONNTRACK_OUT	= IP6_DEFRAG_CONNTRACK_OUT + USHRT_MAX,
522
	IP6_DEFRAG_CONNTRACK_BRIDGE_IN,
523
	__IP6_DEFRAG_CONNTRACK_BRIDGE_IN = IP6_DEFRAG_CONNTRACK_BRIDGE_IN + USHRT_MAX,
524 525
};

526 527
struct ip6_create_arg {
	__be32 id;
528
	u32 user;
529 530
	const struct in6_addr *src;
	const struct in6_addr *dst;
531
	int iif;
532
	u8 ecn;
533 534
};

535 536
void ip6_frag_init(struct inet_frag_queue *q, const void *a);
bool ip6_frag_match(const struct inet_frag_queue *q, const void *a);
537

538 539 540 541 542 543 544 545 546 547 548 549 550 551
/*
 *	Equivalent of ipv4 struct ip
 */
struct frag_queue {
	struct inet_frag_queue	q;

	__be32			id;		/* fragment id		*/
	u32			user;
	struct in6_addr		saddr;
	struct in6_addr		daddr;

	int			iif;
	unsigned int		csum;
	__u16			nhoffset;
552
	u8			ecn;
553 554 555 556 557
};

void ip6_expire_frag_queue(struct net *net, struct frag_queue *fq,
			   struct inet_frags *frags);

E
Eric Dumazet 已提交
558
static inline bool ipv6_addr_any(const struct in6_addr *a)
L
Linus Torvalds 已提交
559
{
560 561 562 563 564
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
	const unsigned long *ul = (const unsigned long *)a;

	return (ul[0] | ul[1]) == 0UL;
#else
E
Eric Dumazet 已提交
565 566
	return (a->s6_addr32[0] | a->s6_addr32[1] |
		a->s6_addr32[2] | a->s6_addr32[3]) == 0;
567
#endif
L
Linus Torvalds 已提交
568 569
}

E
Eric Dumazet 已提交
570 571 572 573 574 575 576 577 578 579 580 581 582
static inline u32 ipv6_addr_hash(const struct in6_addr *a)
{
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
	const unsigned long *ul = (const unsigned long *)a;
	unsigned long x = ul[0] ^ ul[1];

	return (u32)(x ^ (x >> 32));
#else
	return (__force u32)(a->s6_addr32[0] ^ a->s6_addr32[1] ^
			     a->s6_addr32[2] ^ a->s6_addr32[3]);
#endif
}

E
Eric Dumazet 已提交
583
/* more secured version of ipv6_addr_hash() */
584
static inline u32 __ipv6_addr_jhash(const struct in6_addr *a, const u32 initval)
E
Eric Dumazet 已提交
585 586 587 588 589 590
{
	u32 v = (__force u32)a->s6_addr32[0] ^ (__force u32)a->s6_addr32[1];

	return jhash_3words(v,
			    (__force u32)a->s6_addr32[2],
			    (__force u32)a->s6_addr32[3],
591
			    initval);
E
Eric Dumazet 已提交
592 593
}

E
Eric Dumazet 已提交
594
static inline bool ipv6_addr_loopback(const struct in6_addr *a)
595
{
596
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
597
	const __be64 *be = (const __be64 *)a;
598

599
	return (be[0] | (be[1] ^ cpu_to_be64(1))) == 0UL;
600
#else
E
Eric Dumazet 已提交
601
	return (a->s6_addr32[0] | a->s6_addr32[1] |
602
		a->s6_addr32[2] | (a->s6_addr32[3] ^ cpu_to_be32(1))) == 0;
603
#endif
604 605
}

606 607 608 609
/*
 * Note that we must __force cast these to unsigned long to make sparse happy,
 * since all of the endian-annotated types are fixed size regardless of arch.
 */
E
Eric Dumazet 已提交
610
static inline bool ipv6_addr_v4mapped(const struct in6_addr *a)
B
Brian Haley 已提交
611
{
612 613
	return (
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
614
		*(unsigned long *)a |
615
#else
616
		(__force unsigned long)(a->s6_addr32[0] | a->s6_addr32[1]) |
617
#endif
618 619
		(__force unsigned long)(a->s6_addr32[2] ^
					cpu_to_be32(0x0000ffff))) == 0UL;
B
Brian Haley 已提交
620 621
}

622 623 624 625
/*
 * Check for a RFC 4843 ORCHID address
 * (Overlay Routable Cryptographic Hash Identifiers)
 */
E
Eric Dumazet 已提交
626
static inline bool ipv6_addr_orchid(const struct in6_addr *a)
627
{
E
Eric Dumazet 已提交
628
	return (a->s6_addr32[0] & htonl(0xfffffff0)) == htonl(0x20010010);
629 630
}

631 632 633 634 635
static inline bool ipv6_addr_is_multicast(const struct in6_addr *addr)
{
	return (addr->s6_addr32[0] & htonl(0xFF000000)) == htonl(0xFF000000);
}

636 637 638 639 640 641 642 643 644
static inline void ipv6_addr_set_v4mapped(const __be32 addr,
					  struct in6_addr *v4mapped)
{
	ipv6_addr_set(v4mapped,
			0, 0,
			htonl(0x0000FFFF),
			addr);
}

645 646 647 648
/*
 * find the first different bit between two addresses
 * length of address must be a multiple of 32bits
 */
649
static inline int __ipv6_addr_diff32(const void *token1, const void *token2, int addrlen)
650
{
651
	const __be32 *a1 = token1, *a2 = token2;
652 653 654 655 656
	int i;

	addrlen >>= 2;

	for (i = 0; i < addrlen; i++) {
657 658
		__be32 xb = a1[i] ^ a2[i];
		if (xb)
659
			return i * 32 + 31 - __fls(ntohl(xb));
660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677
	}

	/*
	 *	we should *never* get to this point since that 
	 *	would mean the addrs are equal
	 *
	 *	However, we do get to it 8) And exacly, when
	 *	addresses are equal 8)
	 *
	 *	ip route add 1111::/128 via ...
	 *	ip route add 1111::/64 via ...
	 *	and we are here.
	 *
	 *	Ideally, this function should stop comparison
	 *	at prefix length. It does not, but it is still OK,
	 *	if returned value is greater than prefix length.
	 *					--ANK (980803)
	 */
E
Eric Dumazet 已提交
678
	return addrlen << 5;
679 680
}

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
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
static inline int __ipv6_addr_diff64(const void *token1, const void *token2, int addrlen)
{
	const __be64 *a1 = token1, *a2 = token2;
	int i;

	addrlen >>= 3;

	for (i = 0; i < addrlen; i++) {
		__be64 xb = a1[i] ^ a2[i];
		if (xb)
			return i * 64 + 63 - __fls(be64_to_cpu(xb));
	}

	return addrlen << 6;
}
#endif

static inline int __ipv6_addr_diff(const void *token1, const void *token2, int addrlen)
{
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
	if (__builtin_constant_p(addrlen) && !(addrlen & 7))
		return __ipv6_addr_diff64(token1, token2, addrlen);
#endif
	return __ipv6_addr_diff32(token1, token2, addrlen);
}

708 709 710 711 712
static inline int ipv6_addr_diff(const struct in6_addr *a1, const struct in6_addr *a2)
{
	return __ipv6_addr_diff(a1, a2, sizeof(struct in6_addr));
}

713 714 715
__be32 ipv6_select_ident(struct net *net,
			 const struct in6_addr *daddr,
			 const struct in6_addr *saddr);
716
void ipv6_proxy_select_ident(struct net *net, struct sk_buff *skb);
717

718
int ip6_dst_hoplimit(struct dst_entry *dst);
719

720 721 722 723 724 725 726 727 728 729 730 731 732 733
static inline int ip6_sk_dst_hoplimit(struct ipv6_pinfo *np, struct flowi6 *fl6,
				      struct dst_entry *dst)
{
	int hlimit;

	if (ipv6_addr_is_multicast(&fl6->daddr))
		hlimit = np->mcast_hops;
	else
		hlimit = np->hop_limit;
	if (hlimit < 0)
		hlimit = ip6_dst_hoplimit(dst);
	return hlimit;
}

734 735 736 737 738 739 740 741 742 743 744 745 746 747
/* copy IPv6 saddr & daddr to flow_keys, possibly using 64bit load/store
 * Equivalent to :	flow->v6addrs.src = iph->saddr;
 *			flow->v6addrs.dst = iph->daddr;
 */
static inline void iph_to_flow_copy_v6addrs(struct flow_keys *flow,
					    const struct ipv6hdr *iph)
{
	BUILD_BUG_ON(offsetof(typeof(flow->addrs), v6addrs.dst) !=
		     offsetof(typeof(flow->addrs), v6addrs.src) +
		     sizeof(flow->addrs.v6addrs.src));
	memcpy(&flow->addrs.v6addrs, &iph->saddr, sizeof(flow->addrs.v6addrs));
	flow->control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;
}

748
#if IS_ENABLED(CONFIG_IPV6)
749 750 751 752 753 754 755 756 757

/* Sysctl settings for net ipv6.auto_flowlabels */
#define IP6_AUTO_FLOW_LABEL_OFF		0
#define IP6_AUTO_FLOW_LABEL_OPTOUT	1
#define IP6_AUTO_FLOW_LABEL_OPTIN	2
#define IP6_AUTO_FLOW_LABEL_FORCED	3

#define IP6_AUTO_FLOW_LABEL_MAX		IP6_AUTO_FLOW_LABEL_FORCED

758
#define IP6_DEFAULT_AUTO_FLOW_LABELS	IP6_AUTO_FLOW_LABEL_OPTOUT
759

760
static inline __be32 ip6_make_flowlabel(struct net *net, struct sk_buff *skb,
761 762
					__be32 flowlabel, bool autolabel,
					struct flowi6 *fl6)
763
{
764
	u32 hash;
765

766 767 768 769 770
	if (flowlabel ||
	    net->ipv6.sysctl.auto_flowlabels == IP6_AUTO_FLOW_LABEL_OFF ||
	    (!autolabel &&
	     net->ipv6.sysctl.auto_flowlabels != IP6_AUTO_FLOW_LABEL_FORCED))
		return flowlabel;
771

772
	hash = skb_get_hash_flowi6(skb, fl6);
773

774 775 776 777 778
	/* Since this is being sent on the wire obfuscate hash a bit
	 * to minimize possbility that any useful information to an
	 * attacker is leaked. Only lower 20 bits are relevant.
	 */
	rol32(hash, 16);
T
Tom Herbert 已提交
779

780 781 782 783
	flowlabel = (__force __be32)hash & IPV6_FLOWLABEL_MASK;

	if (net->ipv6.sysctl.flowlabel_state_ranges)
		flowlabel |= IPV6_FLOWLABEL_STATELESS_FLAG;
784 785 786

	return flowlabel;
}
787 788 789 790 791 792 793 794 795 796 797 798 799

static inline int ip6_default_np_autolabel(struct net *net)
{
	switch (net->ipv6.sysctl.auto_flowlabels) {
	case IP6_AUTO_FLOW_LABEL_OFF:
	case IP6_AUTO_FLOW_LABEL_OPTIN:
	default:
		return 0;
	case IP6_AUTO_FLOW_LABEL_OPTOUT:
	case IP6_AUTO_FLOW_LABEL_FORCED:
		return 1;
	}
}
800 801 802
#else
static inline void ip6_set_txhash(struct sock *sk) { }
static inline __be32 ip6_make_flowlabel(struct net *net, struct sk_buff *skb,
803 804
					__be32 flowlabel, bool autolabel,
					struct flowi6 *fl6)
805 806 807
{
	return flowlabel;
}
808 809 810 811
static inline int ip6_default_np_autolabel(struct net *net)
{
	return 0;
}
812 813
#endif

814

815 816 817 818 819 820
/*
 *	Header manipulation
 */
static inline void ip6_flow_hdr(struct ipv6hdr *hdr, unsigned int tclass,
				__be32 flowlabel)
{
821
	*(__be32 *)hdr = htonl(0x60000000 | (tclass << 20)) | flowlabel;
822 823
}

824 825 826 827 828
static inline __be32 ip6_flowinfo(const struct ipv6hdr *hdr)
{
	return *(__be32 *)hdr & IPV6_FLOWINFO_MASK;
}

F
Florent Fourcot 已提交
829 830 831 832 833
static inline __be32 ip6_flowlabel(const struct ipv6hdr *hdr)
{
	return *(__be32 *)hdr & IPV6_FLOWLABEL_MASK;
}

834 835 836 837
static inline u8 ip6_tclass(__be32 flowinfo)
{
	return ntohl(flowinfo & IPV6_TCLASS_MASK) >> IPV6_TCLASS_SHIFT;
}
838 839 840 841 842 843

static inline __be32 ip6_make_flowinfo(unsigned int tclass, __be32 flowlabel)
{
	return htonl(tclass << IPV6_TCLASS_SHIFT) | flowlabel;
}

L
Linus Torvalds 已提交
844 845 846 847 848 849 850 851
/*
 *	Prototypes exported by ipv6
 */

/*
 *	rcv function (called from netdevice level)
 */

852 853
int ipv6_rcv(struct sk_buff *skb, struct net_device *dev,
	     struct packet_type *pt, struct net_device *orig_dev);
L
Linus Torvalds 已提交
854

855
int ip6_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb);
856

L
Linus Torvalds 已提交
857 858 859
/*
 *	upper-layer output functions
 */
860
int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6,
861 862 863 864 865 866 867 868 869 870 871 872 873 874 875
	     struct ipv6_txoptions *opt, int tclass);

int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr);

int ip6_append_data(struct sock *sk,
		    int getfrag(void *from, char *to, int offset, int len,
				int odd, struct sk_buff *skb),
		    void *from, int length, int transhdrlen, int hlimit,
		    int tclass, struct ipv6_txoptions *opt, struct flowi6 *fl6,
		    struct rt6_info *rt, unsigned int flags, int dontfrag);

int ip6_push_pending_frames(struct sock *sk);

void ip6_flush_pending_frames(struct sock *sk);

V
Vlad Yasevich 已提交
876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894
int ip6_send_skb(struct sk_buff *skb);

struct sk_buff *__ip6_make_skb(struct sock *sk, struct sk_buff_head *queue,
			       struct inet_cork_full *cork,
			       struct inet6_cork *v6_cork);
struct sk_buff *ip6_make_skb(struct sock *sk,
			     int getfrag(void *from, char *to, int offset,
					 int len, int odd, struct sk_buff *skb),
			     void *from, int length, int transhdrlen,
			     int hlimit, int tclass, struct ipv6_txoptions *opt,
			     struct flowi6 *fl6, struct rt6_info *rt,
			     unsigned int flags, int dontfrag);

static inline struct sk_buff *ip6_finish_skb(struct sock *sk)
{
	return __ip6_make_skb(sk, &sk->sk_write_queue, &inet_sk(sk)->cork,
			      &inet6_sk(sk)->cork);
}

895 896
int ip6_dst_lookup(struct net *net, struct sock *sk, struct dst_entry **dst,
		   struct flowi6 *fl6);
897
struct dst_entry *ip6_dst_lookup_flow(const struct sock *sk, struct flowi6 *fl6,
898
				      const struct in6_addr *final_dst);
899
struct dst_entry *ip6_sk_dst_lookup_flow(struct sock *sk, struct flowi6 *fl6,
900
					 const struct in6_addr *final_dst);
901 902
struct dst_entry *ip6_blackhole_route(struct net *net,
				      struct dst_entry *orig_dst);
L
Linus Torvalds 已提交
903 904 905 906 907

/*
 *	skb processing functions
 */

E
Eric W. Biederman 已提交
908
int ip6_output(struct net *net, struct sock *sk, struct sk_buff *skb);
909 910 911
int ip6_forward(struct sk_buff *skb);
int ip6_input(struct sk_buff *skb);
int ip6_mc_input(struct sk_buff *skb);
L
Linus Torvalds 已提交
912

913
int __ip6_local_out(struct net *net, struct sock *sk, struct sk_buff *skb);
914
int ip6_local_out(struct net *net, struct sock *sk, struct sk_buff *skb);
H
Herbert Xu 已提交
915

L
Linus Torvalds 已提交
916 917 918 919
/*
 *	Extension header (options) processing
 */

920 921 922 923
void ipv6_push_nfrag_opts(struct sk_buff *skb, struct ipv6_txoptions *opt,
			  u8 *proto, struct in6_addr **daddr_p);
void ipv6_push_frag_opts(struct sk_buff *skb, struct ipv6_txoptions *opt,
			 u8 *proto);
L
Linus Torvalds 已提交
924

925 926
int ipv6_skip_exthdr(const struct sk_buff *, int start, u8 *nexthdrp,
		     __be16 *frag_offp);
L
Linus Torvalds 已提交
927

928
bool ipv6_ext_hdr(u8 nexthdr);
L
Linus Torvalds 已提交
929

930
enum {
931 932 933
	IP6_FH_F_FRAG		= (1 << 0),
	IP6_FH_F_AUTH		= (1 << 1),
	IP6_FH_F_SKIP_RH	= (1 << 2),
934 935 936
};

/* find specified header and get offset to it */
937 938
int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, int target,
		  unsigned short *fragoff, int *fragflg);
939

940
int ipv6_find_tlv(struct sk_buff *skb, int offset, int type);
941

942 943 944
struct in6_addr *fl6_update_dst(struct flowi6 *fl6,
				const struct ipv6_txoptions *opt,
				struct in6_addr *orig);
945

L
Linus Torvalds 已提交
946 947 948 949
/*
 *	socket options (ipv6_sockglue.c)
 */

950 951 952 953 954 955 956 957 958 959
int ipv6_setsockopt(struct sock *sk, int level, int optname,
		    char __user *optval, unsigned int optlen);
int ipv6_getsockopt(struct sock *sk, int level, int optname,
		    char __user *optval, int __user *optlen);
int compat_ipv6_setsockopt(struct sock *sk, int level, int optname,
			   char __user *optval, unsigned int optlen);
int compat_ipv6_getsockopt(struct sock *sk, int level, int optname,
			   char __user *optval, int __user *optlen);

int ip6_datagram_connect(struct sock *sk, struct sockaddr *addr, int addr_len);
960 961
int ip6_datagram_connect_v6_only(struct sock *sk, struct sockaddr *addr,
				 int addr_len);
962

963 964 965 966
int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len,
		    int *addr_len);
int ipv6_recv_rxpmtu(struct sock *sk, struct msghdr *msg, int len,
		     int *addr_len);
967 968 969 970 971 972 973 974 975 976 977 978
void ipv6_icmp_error(struct sock *sk, struct sk_buff *skb, int err, __be16 port,
		     u32 info, u8 *payload);
void ipv6_local_error(struct sock *sk, int err, struct flowi6 *fl6, u32 info);
void ipv6_local_rxpmtu(struct sock *sk, struct flowi6 *fl6, u32 mtu);

int inet6_release(struct socket *sock);
int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len);
int inet6_getname(struct socket *sock, struct sockaddr *uaddr, int *uaddr_len,
		  int peer);
int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg);

int inet6_hash_connect(struct inet_timewait_death_row *death_row,
979 980
			      struct sock *sk);

L
Linus Torvalds 已提交
981 982 983
/*
 * reassembly.c
 */
984 985
extern const struct proto_ops inet6_stream_ops;
extern const struct proto_ops inet6_dgram_ops;
986

987 988 989
struct group_source_req;
struct group_filter;

990 991 992 993 994
int ip6_mc_source(int add, int omode, struct sock *sk,
		  struct group_source_req *pgsr);
int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf);
int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
		  struct group_filter __user *optval, int __user *optlen);
995 996

#ifdef CONFIG_PROC_FS
997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
int ac6_proc_init(struct net *net);
void ac6_proc_exit(struct net *net);
int raw6_proc_init(void);
void raw6_proc_exit(void);
int tcp6_proc_init(struct net *net);
void tcp6_proc_exit(struct net *net);
int udp6_proc_init(struct net *net);
void udp6_proc_exit(struct net *net);
int udplite6_proc_init(void);
void udplite6_proc_exit(void);
int ipv6_misc_proc_init(void);
void ipv6_misc_proc_exit(void);
int snmp6_register_dev(struct inet6_dev *idev);
int snmp6_unregister_dev(struct inet6_dev *idev);
1011

1012
#else
1013 1014 1015 1016
static inline int ac6_proc_init(struct net *net) { return 0; }
static inline void ac6_proc_exit(struct net *net) { }
static inline int snmp6_register_dev(struct inet6_dev *idev) { return 0; }
static inline int snmp6_unregister_dev(struct inet6_dev *idev) { return 0; }
1017
#endif
L
Linus Torvalds 已提交
1018

1019
#ifdef CONFIG_SYSCTL
1020
extern struct ctl_table ipv6_route_table_template[];
L
Linus Torvalds 已提交
1021

1022 1023 1024 1025
struct ctl_table *ipv6_icmp_sysctl_init(struct net *net);
struct ctl_table *ipv6_route_sysctl_init(struct net *net);
int ipv6_sysctl_register(void);
void ipv6_sysctl_unregister(void);
1026
#endif
L
Linus Torvalds 已提交
1027

1028 1029 1030 1031
int ipv6_sock_mc_join(struct sock *sk, int ifindex,
		      const struct in6_addr *addr);
int ipv6_sock_mc_drop(struct sock *sk, int ifindex,
		      const struct in6_addr *addr);
1032
#endif /* _NET_IPV6_H */