ip6_tables.c 38.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4
/*
 * Packet matching code.
 *
 * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling
5
 * Copyright (C) 2000-2005 Netfilter Core Team <coreteam@netfilter.org>
L
Linus Torvalds 已提交
6 7 8 9 10
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */
11 12

#include <linux/capability.h>
13
#include <linux/in.h>
L
Linus Torvalds 已提交
14 15 16 17 18
#include <linux/skbuff.h>
#include <linux/kmod.h>
#include <linux/vmalloc.h>
#include <linux/netdevice.h>
#include <linux/module.h>
R
Randy Dunlap 已提交
19
#include <linux/poison.h>
L
Linus Torvalds 已提交
20 21 22
#include <linux/icmpv6.h>
#include <net/ipv6.h>
#include <asm/uaccess.h>
I
Ingo Molnar 已提交
23
#include <linux/mutex.h>
L
Linus Torvalds 已提交
24
#include <linux/proc_fs.h>
25
#include <linux/cpumask.h>
L
Linus Torvalds 已提交
26 27

#include <linux/netfilter_ipv6/ip6_tables.h>
28
#include <linux/netfilter/x_tables.h>
L
Linus Torvalds 已提交
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

MODULE_LICENSE("GPL");
MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
MODULE_DESCRIPTION("IPv6 packet filter");

/*#define DEBUG_IP_FIREWALL*/
/*#define DEBUG_ALLOW_ALL*/ /* Useful for remote debugging */
/*#define DEBUG_IP_FIREWALL_USER*/

#ifdef DEBUG_IP_FIREWALL
#define dprintf(format, args...)  printk(format , ## args)
#else
#define dprintf(format, args...)
#endif

#ifdef DEBUG_IP_FIREWALL_USER
#define duprintf(format, args...) printk(format , ## args)
#else
#define duprintf(format, args...)
#endif

#ifdef CONFIG_NETFILTER_DEBUG
#define IP_NF_ASSERT(x)						\
do {								\
	if (!(x))						\
		printk("IP_NF_ASSERT: %s:%s:%u\n",		\
		       __FUNCTION__, __FILE__, __LINE__);	\
} while(0)
#else
#define IP_NF_ASSERT(x)
#endif

#if 0
/* All the better to debug you with... */
#define static
#define inline
#endif

67
/*
L
Linus Torvalds 已提交
68
   We keep a set of rules for each CPU, so we can avoid write-locking
69 70 71 72
   them in the softirq when updating the counters and therefore
   only need to read-lock in the softirq; doing a write_lock_bh() in user
   context stops packets coming through and allows user context to read
   the counters or update the rules.
L
Linus Torvalds 已提交
73 74 75 76

   Hence the start of any table is given by get_table() below.  */

/* Check for an extension */
77
int
L
Linus Torvalds 已提交
78 79
ip6t_ext_hdr(u8 nexthdr)
{
80 81 82 83 84 85 86
	return ( (nexthdr == IPPROTO_HOPOPTS)   ||
		 (nexthdr == IPPROTO_ROUTING)   ||
		 (nexthdr == IPPROTO_FRAGMENT)  ||
		 (nexthdr == IPPROTO_ESP)       ||
		 (nexthdr == IPPROTO_AH)        ||
		 (nexthdr == IPPROTO_NONE)      ||
		 (nexthdr == IPPROTO_DSTOPTS) );
L
Linus Torvalds 已提交
87 88 89
}

/* Returns whether matches rule or not. */
90
static inline bool
L
Linus Torvalds 已提交
91 92 93 94 95
ip6_packet_match(const struct sk_buff *skb,
		 const char *indev,
		 const char *outdev,
		 const struct ip6t_ip6 *ip6info,
		 unsigned int *protoff,
96
		 int *fragoff, bool *hotdrop)
L
Linus Torvalds 已提交
97 98 99
{
	size_t i;
	unsigned long ret;
100
	const struct ipv6hdr *ipv6 = ipv6_hdr(skb);
L
Linus Torvalds 已提交
101 102 103

#define FWINV(bool,invflg) ((bool) ^ !!(ip6info->invflags & invflg))

104
	if (FWINV(ipv6_masked_addr_cmp(&ipv6->saddr, &ip6info->smsk,
105
				       &ip6info->src), IP6T_INV_SRCIP)
106
	    || FWINV(ipv6_masked_addr_cmp(&ipv6->daddr, &ip6info->dmsk,
107
					  &ip6info->dst), IP6T_INV_DSTIP)) {
L
Linus Torvalds 已提交
108 109 110 111 112 113 114 115
		dprintf("Source or dest mismatch.\n");
/*
		dprintf("SRC: %u. Mask: %u. Target: %u.%s\n", ip->saddr,
			ipinfo->smsk.s_addr, ipinfo->src.s_addr,
			ipinfo->invflags & IP6T_INV_SRCIP ? " (INV)" : "");
		dprintf("DST: %u. Mask: %u. Target: %u.%s\n", ip->daddr,
			ipinfo->dmsk.s_addr, ipinfo->dst.s_addr,
			ipinfo->invflags & IP6T_INV_DSTIP ? " (INV)" : "");*/
116
		return false;
L
Linus Torvalds 已提交
117 118 119 120 121 122 123 124 125 126 127 128 129
	}

	/* Look for ifname matches; this should unroll nicely. */
	for (i = 0, ret = 0; i < IFNAMSIZ/sizeof(unsigned long); i++) {
		ret |= (((const unsigned long *)indev)[i]
			^ ((const unsigned long *)ip6info->iniface)[i])
			& ((const unsigned long *)ip6info->iniface_mask)[i];
	}

	if (FWINV(ret != 0, IP6T_INV_VIA_IN)) {
		dprintf("VIA in mismatch (%s vs %s).%s\n",
			indev, ip6info->iniface,
			ip6info->invflags&IP6T_INV_VIA_IN ?" (INV)":"");
130
		return false;
L
Linus Torvalds 已提交
131 132 133 134 135 136 137 138 139 140 141 142
	}

	for (i = 0, ret = 0; i < IFNAMSIZ/sizeof(unsigned long); i++) {
		ret |= (((const unsigned long *)outdev)[i]
			^ ((const unsigned long *)ip6info->outiface)[i])
			& ((const unsigned long *)ip6info->outiface_mask)[i];
	}

	if (FWINV(ret != 0, IP6T_INV_VIA_OUT)) {
		dprintf("VIA out mismatch (%s vs %s).%s\n",
			outdev, ip6info->outiface,
			ip6info->invflags&IP6T_INV_VIA_OUT ?" (INV)":"");
143
		return false;
L
Linus Torvalds 已提交
144 145 146 147 148 149
	}

/* ... might want to do something with class and flowlabel here ... */

	/* look for the desired protocol header */
	if((ip6info->flags & IP6T_F_PROTO)) {
150 151
		int protohdr;
		unsigned short _frag_off;
L
Linus Torvalds 已提交
152

153
		protohdr = ipv6_find_hdr(skb, protoff, -1, &_frag_off);
154 155
		if (protohdr < 0) {
			if (_frag_off == 0)
156
				*hotdrop = true;
157
			return false;
158
		}
159
		*fragoff = _frag_off;
L
Linus Torvalds 已提交
160 161

		dprintf("Packet protocol %hi ?= %s%hi.\n",
162
				protohdr,
L
Linus Torvalds 已提交
163 164 165
				ip6info->invflags & IP6T_INV_PROTO ? "!":"",
				ip6info->proto);

166
		if (ip6info->proto == protohdr) {
L
Linus Torvalds 已提交
167
			if(ip6info->invflags & IP6T_INV_PROTO) {
168
				return false;
L
Linus Torvalds 已提交
169
			}
170
			return true;
L
Linus Torvalds 已提交
171 172 173 174 175
		}

		/* We need match for the '-p all', too! */
		if ((ip6info->proto != 0) &&
			!(ip6info->invflags & IP6T_INV_PROTO))
176
			return false;
L
Linus Torvalds 已提交
177
	}
178
	return true;
L
Linus Torvalds 已提交
179 180 181
}

/* should be ip6 safe */
182
static inline bool
L
Linus Torvalds 已提交
183 184 185 186 187
ip6_checkentry(const struct ip6t_ip6 *ipv6)
{
	if (ipv6->flags & ~IP6T_F_MASK) {
		duprintf("Unknown flag bits set: %08X\n",
			 ipv6->flags & ~IP6T_F_MASK);
188
		return false;
L
Linus Torvalds 已提交
189 190 191 192
	}
	if (ipv6->invflags & ~IP6T_INV_MASK) {
		duprintf("Unknown invflag bits set: %08X\n",
			 ipv6->invflags & ~IP6T_INV_MASK);
193
		return false;
L
Linus Torvalds 已提交
194
	}
195
	return true;
L
Linus Torvalds 已提交
196 197 198
}

static unsigned int
199
ip6t_error(struct sk_buff *skb,
L
Linus Torvalds 已提交
200 201 202
	  const struct net_device *in,
	  const struct net_device *out,
	  unsigned int hooknum,
203
	  const struct xt_target *target,
204
	  const void *targinfo)
L
Linus Torvalds 已提交
205 206 207 208 209 210 211 212
{
	if (net_ratelimit())
		printk("ip6_tables: error: `%s'\n", (char *)targinfo);

	return NF_DROP;
}

static inline
213 214 215 216 217 218 219
bool do_match(struct ip6t_entry_match *m,
	      const struct sk_buff *skb,
	      const struct net_device *in,
	      const struct net_device *out,
	      int offset,
	      unsigned int protoff,
	      bool *hotdrop)
L
Linus Torvalds 已提交
220 221
{
	/* Stop iteration if it doesn't match */
222
	if (!m->u.kernel.match->match(skb, in, out, m->u.kernel.match, m->data,
L
Linus Torvalds 已提交
223
				      offset, protoff, hotdrop))
224
		return true;
L
Linus Torvalds 已提交
225
	else
226
		return false;
L
Linus Torvalds 已提交
227 228 229 230 231 232 233 234
}

static inline struct ip6t_entry *
get_entry(void *base, unsigned int offset)
{
	return (struct ip6t_entry *)(base + offset);
}

235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251
/* All zeroes == unconditional rule. */
static inline int
unconditional(const struct ip6t_ip6 *ipv6)
{
	unsigned int i;

	for (i = 0; i < sizeof(*ipv6); i++)
		if (((char *)ipv6)[i])
			break;

	return (i == sizeof(*ipv6));
}

#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
    defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
/* This cries for unification! */
static const char *hooknames[] = {
252 253 254 255 256
	[NF_INET_PRE_ROUTING]		= "PREROUTING",
	[NF_INET_LOCAL_IN]		= "INPUT",
	[NF_INET_FORWARD]		= "FORWARD",
	[NF_INET_LOCAL_OUT]		= "OUTPUT",
	[NF_INET_POST_ROUTING]		= "POSTROUTING",
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341
};

enum nf_ip_trace_comments {
	NF_IP6_TRACE_COMMENT_RULE,
	NF_IP6_TRACE_COMMENT_RETURN,
	NF_IP6_TRACE_COMMENT_POLICY,
};

static const char *comments[] = {
	[NF_IP6_TRACE_COMMENT_RULE]	= "rule",
	[NF_IP6_TRACE_COMMENT_RETURN]	= "return",
	[NF_IP6_TRACE_COMMENT_POLICY]	= "policy",
};

static struct nf_loginfo trace_loginfo = {
	.type = NF_LOG_TYPE_LOG,
	.u = {
		.log = {
			.level = 4,
			.logflags = NF_LOG_MASK,
		},
	},
};

static inline int
get_chainname_rulenum(struct ip6t_entry *s, struct ip6t_entry *e,
		      char *hookname, char **chainname,
		      char **comment, unsigned int *rulenum)
{
	struct ip6t_standard_target *t = (void *)ip6t_get_target(s);

	if (strcmp(t->target.u.kernel.target->name, IP6T_ERROR_TARGET) == 0) {
		/* Head of user chain: ERROR target with chainname */
		*chainname = t->target.data;
		(*rulenum) = 0;
	} else if (s == e) {
		(*rulenum)++;

		if (s->target_offset == sizeof(struct ip6t_entry)
		   && strcmp(t->target.u.kernel.target->name,
			     IP6T_STANDARD_TARGET) == 0
		   && t->verdict < 0
		   && unconditional(&s->ipv6)) {
			/* Tail of chains: STANDARD target (return/policy) */
			*comment = *chainname == hookname
				? (char *)comments[NF_IP6_TRACE_COMMENT_POLICY]
				: (char *)comments[NF_IP6_TRACE_COMMENT_RETURN];
		}
		return 1;
	} else
		(*rulenum)++;

	return 0;
}

static void trace_packet(struct sk_buff *skb,
			 unsigned int hook,
			 const struct net_device *in,
			 const struct net_device *out,
			 char *tablename,
			 struct xt_table_info *private,
			 struct ip6t_entry *e)
{
	void *table_base;
	struct ip6t_entry *root;
	char *hookname, *chainname, *comment;
	unsigned int rulenum = 0;

	table_base = (void *)private->entries[smp_processor_id()];
	root = get_entry(table_base, private->hook_entry[hook]);

	hookname = chainname = (char *)hooknames[hook];
	comment = (char *)comments[NF_IP6_TRACE_COMMENT_RULE];

	IP6T_ENTRY_ITERATE(root,
			   private->size - private->hook_entry[hook],
			   get_chainname_rulenum,
			   e, hookname, &chainname, &comment, &rulenum);

	nf_log_packet(AF_INET6, hook, skb, in, out, &trace_loginfo,
		      "TRACE: %s:%s:%s:%u ",
		      tablename, chainname, comment, rulenum);
}
#endif

L
Linus Torvalds 已提交
342 343
/* Returns one of the generic firewall policies, like NF_ACCEPT. */
unsigned int
344
ip6t_do_table(struct sk_buff *skb,
L
Linus Torvalds 已提交
345 346 347
	      unsigned int hook,
	      const struct net_device *in,
	      const struct net_device *out,
348
	      struct xt_table *table)
L
Linus Torvalds 已提交
349
{
350
	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
L
Linus Torvalds 已提交
351 352
	int offset = 0;
	unsigned int protoff = 0;
353
	bool hotdrop = false;
L
Linus Torvalds 已提交
354 355 356 357 358
	/* Initializing verdict to NF_DROP keeps gcc happy. */
	unsigned int verdict = NF_DROP;
	const char *indev, *outdev;
	void *table_base;
	struct ip6t_entry *e, *back;
359
	struct xt_table_info *private;
L
Linus Torvalds 已提交
360 361 362 363 364 365 366 367 368 369 370 371

	/* Initialization */
	indev = in ? in->name : nulldevname;
	outdev = out ? out->name : nulldevname;
	/* We handle fragments by dealing with the first fragment as
	 * if it was a normal packet.  All other fragments are treated
	 * normally, except that they will NEVER match rules that ask
	 * things we don't know, ie. tcp syn flag or ports).  If the
	 * rule is also a fragment-specific rule, non-fragments won't
	 * match it. */

	read_lock_bh(&table->lock);
372
	private = table->private;
L
Linus Torvalds 已提交
373
	IP_NF_ASSERT(table->valid_hooks & (1 << hook));
374 375
	table_base = (void *)private->entries[smp_processor_id()];
	e = get_entry(table_base, private->hook_entry[hook]);
L
Linus Torvalds 已提交
376 377

	/* For return from builtin chain */
378
	back = get_entry(table_base, private->underflow[hook]);
L
Linus Torvalds 已提交
379 380 381 382

	do {
		IP_NF_ASSERT(e);
		IP_NF_ASSERT(back);
383
		if (ip6_packet_match(skb, indev, outdev, &e->ipv6,
384
			&protoff, &offset, &hotdrop)) {
L
Linus Torvalds 已提交
385 386 387
			struct ip6t_entry_target *t;

			if (IP6T_MATCH_ITERATE(e, do_match,
388
					       skb, in, out,
L
Linus Torvalds 已提交
389 390 391 392
					       offset, protoff, &hotdrop) != 0)
				goto no_match;

			ADD_COUNTER(e->counters,
393 394
				    ntohs(ipv6_hdr(skb)->payload_len) +
				    sizeof(struct ipv6hdr), 1);
L
Linus Torvalds 已提交
395 396 397

			t = ip6t_get_target(e);
			IP_NF_ASSERT(t->u.kernel.target);
398 399 400 401

#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
    defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
			/* The packet is traced: log it */
402 403
			if (unlikely(skb->nf_trace))
				trace_packet(skb, hook, in, out,
404 405
					     table->name, private, e);
#endif
L
Linus Torvalds 已提交
406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421
			/* Standard target? */
			if (!t->u.kernel.target->target) {
				int v;

				v = ((struct ip6t_standard_target *)t)->verdict;
				if (v < 0) {
					/* Pop from stack? */
					if (v != IP6T_RETURN) {
						verdict = (unsigned)(-v) - 1;
						break;
					}
					e = back;
					back = get_entry(table_base,
							 back->comefrom);
					continue;
				}
P
Patrick McHardy 已提交
422 423
				if (table_base + v != (void *)e + e->next_offset
				    && !(e->ipv6.flags & IP6T_F_GOTO)) {
L
Linus Torvalds 已提交
424 425 426 427 428 429 430 431 432 433 434 435
					/* Save old back ptr in next entry */
					struct ip6t_entry *next
						= (void *)e + e->next_offset;
					next->comefrom
						= (void *)back - table_base;
					/* set back pointer to next entry */
					back = next;
				}

				e = get_entry(table_base, v);
			} else {
				/* Targets which reenter must return
436
				   abs. verdicts */
L
Linus Torvalds 已提交
437 438 439 440
#ifdef CONFIG_NETFILTER_DEBUG
				((struct ip6t_entry *)table_base)->comefrom
					= 0xeeeeeeec;
#endif
441
				verdict = t->u.kernel.target->target(skb,
L
Linus Torvalds 已提交
442 443
								     in, out,
								     hook,
444
								     t->u.kernel.target,
445
								     t->data);
L
Linus Torvalds 已提交
446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471

#ifdef CONFIG_NETFILTER_DEBUG
				if (((struct ip6t_entry *)table_base)->comefrom
				    != 0xeeeeeeec
				    && verdict == IP6T_CONTINUE) {
					printk("Target %s reentered!\n",
					       t->u.kernel.target->name);
					verdict = NF_DROP;
				}
				((struct ip6t_entry *)table_base)->comefrom
					= 0x57acc001;
#endif
				if (verdict == IP6T_CONTINUE)
					e = (void *)e + e->next_offset;
				else
					/* Verdict */
					break;
			}
		} else {

		no_match:
			e = (void *)e + e->next_offset;
		}
	} while (!hotdrop);

#ifdef CONFIG_NETFILTER_DEBUG
R
Randy Dunlap 已提交
472
	((struct ip6t_entry *)table_base)->comefrom = NETFILTER_LINK_POISON;
L
Linus Torvalds 已提交
473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
#endif
	read_unlock_bh(&table->lock);

#ifdef DEBUG_ALLOW_ALL
	return NF_ACCEPT;
#else
	if (hotdrop)
		return NF_DROP;
	else return verdict;
#endif
}

/* Figures out from what hook each rule can be called: returns 0 if
   there are loops.  Puts hook bitmask in comefrom. */
static int
488
mark_source_chains(struct xt_table_info *newinfo,
489
		   unsigned int valid_hooks, void *entry0)
L
Linus Torvalds 已提交
490 491 492 493 494
{
	unsigned int hook;

	/* No recursion; use packet counter to save back ptrs (reset
	   to 0 as we leave), and comefrom to save source hook bitmask */
495
	for (hook = 0; hook < NF_INET_NUMHOOKS; hook++) {
L
Linus Torvalds 已提交
496 497
		unsigned int pos = newinfo->hook_entry[hook];
		struct ip6t_entry *e
498
			= (struct ip6t_entry *)(entry0 + pos);
499
		int visited = e->comefrom & (1 << hook);
L
Linus Torvalds 已提交
500 501 502 503 504 505 506 507 508 509 510

		if (!(valid_hooks & (1 << hook)))
			continue;

		/* Set initial back pointer. */
		e->counters.pcnt = pos;

		for (;;) {
			struct ip6t_standard_target *t
				= (void *)ip6t_get_target(e);

511
			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
512 513 514 515 516
				printk("iptables: loop hook %u pos %u %08X.\n",
				       hook, pos, e->comefrom);
				return 0;
			}
			e->comefrom
517
				|= ((1 << hook) | (1 << NF_INET_NUMHOOKS));
L
Linus Torvalds 已提交
518 519

			/* Unconditional return/END. */
520
			if ((e->target_offset == sizeof(struct ip6t_entry)
L
Linus Torvalds 已提交
521 522 523
			    && (strcmp(t->target.u.user.name,
				       IP6T_STANDARD_TARGET) == 0)
			    && t->verdict < 0
524
			    && unconditional(&e->ipv6)) || visited) {
L
Linus Torvalds 已提交
525 526
				unsigned int oldpos, size;

527 528 529 530 531 532 533
				if (t->verdict < -NF_MAX_VERDICT - 1) {
					duprintf("mark_source_chains: bad "
						"negative verdict (%i)\n",
								t->verdict);
					return 0;
				}

L
Linus Torvalds 已提交
534 535 536
				/* Return: backtrack through the last
				   big jump. */
				do {
537
					e->comefrom ^= (1<<NF_INET_NUMHOOKS);
L
Linus Torvalds 已提交
538 539
#ifdef DEBUG_IP_FIREWALL_USER
					if (e->comefrom
540
					    & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
541 542 543 544 545 546 547 548 549 550 551 552 553 554 555
						duprintf("Back unset "
							 "on hook %u "
							 "rule %u\n",
							 hook, pos);
					}
#endif
					oldpos = pos;
					pos = e->counters.pcnt;
					e->counters.pcnt = 0;

					/* We're at the start. */
					if (pos == oldpos)
						goto next;

					e = (struct ip6t_entry *)
556
						(entry0 + pos);
L
Linus Torvalds 已提交
557 558 559 560 561
				} while (oldpos == pos + e->next_offset);

				/* Move along one */
				size = e->next_offset;
				e = (struct ip6t_entry *)
562
					(entry0 + pos + size);
L
Linus Torvalds 已提交
563 564 565 566 567 568 569 570
				e->counters.pcnt = pos;
				pos += size;
			} else {
				int newpos = t->verdict;

				if (strcmp(t->target.u.user.name,
					   IP6T_STANDARD_TARGET) == 0
				    && newpos >= 0) {
571 572 573 574 575 576 577
					if (newpos > newinfo->size -
						sizeof(struct ip6t_entry)) {
						duprintf("mark_source_chains: "
							"bad verdict (%i)\n",
								newpos);
						return 0;
					}
L
Linus Torvalds 已提交
578 579 580 581 582 583 584 585
					/* This a jump; chase it. */
					duprintf("Jump rule %u -> %u\n",
						 pos, newpos);
				} else {
					/* ... this is a fallthru */
					newpos = pos + e->next_offset;
				}
				e = (struct ip6t_entry *)
586
					(entry0 + newpos);
L
Linus Torvalds 已提交
587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603
				e->counters.pcnt = pos;
				pos = newpos;
			}
		}
		next:
		duprintf("Finished chain %u\n", hook);
	}
	return 1;
}

static inline int
cleanup_match(struct ip6t_entry_match *m, unsigned int *i)
{
	if (i && (*i)-- == 0)
		return 1;

	if (m->u.kernel.match->destroy)
604
		m->u.kernel.match->destroy(m->u.kernel.match, m->data);
L
Linus Torvalds 已提交
605 606 607 608 609 610 611 612 613 614 615
	module_put(m->u.kernel.match->me);
	return 0;
}

static inline int
check_match(struct ip6t_entry_match *m,
	    const char *name,
	    const struct ip6t_ip6 *ipv6,
	    unsigned int hookmask,
	    unsigned int *i)
{
616
	struct xt_match *match;
617
	int ret;
L
Linus Torvalds 已提交
618

619
	match = try_then_request_module(xt_find_match(AF_INET6, m->u.user.name,
620
					m->u.user.revision),
621 622
					"ip6t_%s", m->u.user.name);
	if (IS_ERR(match) || !match) {
623
		duprintf("check_match: `%s' not found\n", m->u.user.name);
624
		return match ? PTR_ERR(match) : -ENOENT;
L
Linus Torvalds 已提交
625 626 627
	}
	m->u.kernel.match = match;

628 629 630 631 632 633
	ret = xt_check_match(match, AF_INET6, m->u.match_size - sizeof(*m),
			     name, hookmask, ipv6->proto,
			     ipv6->invflags & IP6T_INV_PROTO);
	if (ret)
		goto err;

L
Linus Torvalds 已提交
634
	if (m->u.kernel.match->checkentry
635
	    && !m->u.kernel.match->checkentry(name, ipv6, match,  m->data,
L
Linus Torvalds 已提交
636 637 638
					      hookmask)) {
		duprintf("ip_tables: check failed for `%s'.\n",
			 m->u.kernel.match->name);
639 640
		ret = -EINVAL;
		goto err;
L
Linus Torvalds 已提交
641 642 643 644
	}

	(*i)++;
	return 0;
645 646 647
err:
	module_put(m->u.kernel.match->me);
	return ret;
L
Linus Torvalds 已提交
648 649 650 651 652 653 654
}

static inline int
check_entry(struct ip6t_entry *e, const char *name, unsigned int size,
	    unsigned int *i)
{
	struct ip6t_entry_target *t;
655
	struct xt_target *target;
L
Linus Torvalds 已提交
656 657 658 659 660 661 662 663
	int ret;
	unsigned int j;

	if (!ip6_checkentry(&e->ipv6)) {
		duprintf("ip_tables: ip check failed %p %s.\n", e, name);
		return -EINVAL;
	}

664 665 666 667
	if (e->target_offset + sizeof(struct ip6t_entry_target) >
								e->next_offset)
		return -EINVAL;

L
Linus Torvalds 已提交
668 669 670 671 672 673
	j = 0;
	ret = IP6T_MATCH_ITERATE(e, check_match, name, &e->ipv6, e->comefrom, &j);
	if (ret != 0)
		goto cleanup_matches;

	t = ip6t_get_target(e);
674 675 676
	ret = -EINVAL;
	if (e->target_offset + t->u.target_size > e->next_offset)
			goto cleanup_matches;
677 678 679
	target = try_then_request_module(xt_find_target(AF_INET6,
							t->u.user.name,
							t->u.user.revision),
680 681
					 "ip6t_%s", t->u.user.name);
	if (IS_ERR(target) || !target) {
L
Linus Torvalds 已提交
682
		duprintf("check_entry: `%s' not found\n", t->u.user.name);
683
		ret = target ? PTR_ERR(target) : -ENOENT;
L
Linus Torvalds 已提交
684 685 686
		goto cleanup_matches;
	}
	t->u.kernel.target = target;
687

688 689 690 691 692 693
	ret = xt_check_target(target, AF_INET6, t->u.target_size - sizeof(*t),
			      name, e->comefrom, e->ipv6.proto,
			      e->ipv6.invflags & IP6T_INV_PROTO);
	if (ret)
		goto err;

694
	if (t->u.kernel.target->checkentry
695
		   && !t->u.kernel.target->checkentry(name, e, target, t->data,
L
Linus Torvalds 已提交
696 697 698 699
						      e->comefrom)) {
		duprintf("ip_tables: check failed for `%s'.\n",
			 t->u.kernel.target->name);
		ret = -EINVAL;
700
		goto err;
L
Linus Torvalds 已提交
701 702 703 704
	}

	(*i)++;
	return 0;
705 706
 err:
	module_put(t->u.kernel.target->me);
L
Linus Torvalds 已提交
707 708 709 710 711 712 713
 cleanup_matches:
	IP6T_MATCH_ITERATE(e, cleanup_match, &j);
	return ret;
}

static inline int
check_entry_size_and_hooks(struct ip6t_entry *e,
714
			   struct xt_table_info *newinfo,
L
Linus Torvalds 已提交
715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736
			   unsigned char *base,
			   unsigned char *limit,
			   const unsigned int *hook_entries,
			   const unsigned int *underflows,
			   unsigned int *i)
{
	unsigned int h;

	if ((unsigned long)e % __alignof__(struct ip6t_entry) != 0
	    || (unsigned char *)e + sizeof(struct ip6t_entry) >= limit) {
		duprintf("Bad offset %p\n", e);
		return -EINVAL;
	}

	if (e->next_offset
	    < sizeof(struct ip6t_entry) + sizeof(struct ip6t_entry_target)) {
		duprintf("checking: element %p size %u\n",
			 e, e->next_offset);
		return -EINVAL;
	}

	/* Check hooks & underflows */
737
	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
L
Linus Torvalds 已提交
738 739 740 741 742 743 744
		if ((unsigned char *)e - base == hook_entries[h])
			newinfo->hook_entry[h] = hook_entries[h];
		if ((unsigned char *)e - base == underflows[h])
			newinfo->underflow[h] = underflows[h];
	}

	/* FIXME: underflows must be unconditional, standard verdicts
745
	   < 0 (not IP6T_RETURN). --RR */
L
Linus Torvalds 已提交
746 747

	/* Clear counters and comefrom */
748
	e->counters = ((struct xt_counters) { 0, 0 });
L
Linus Torvalds 已提交
749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766
	e->comefrom = 0;

	(*i)++;
	return 0;
}

static inline int
cleanup_entry(struct ip6t_entry *e, unsigned int *i)
{
	struct ip6t_entry_target *t;

	if (i && (*i)-- == 0)
		return 1;

	/* Cleanup all matches */
	IP6T_MATCH_ITERATE(e, cleanup_match, NULL);
	t = ip6t_get_target(e);
	if (t->u.kernel.target->destroy)
767
		t->u.kernel.target->destroy(t->u.kernel.target, t->data);
L
Linus Torvalds 已提交
768 769 770 771 772 773 774 775 776
	module_put(t->u.kernel.target->me);
	return 0;
}

/* Checks and translates the user-supplied table segment (held in
   newinfo) */
static int
translate_table(const char *name,
		unsigned int valid_hooks,
777
		struct xt_table_info *newinfo,
778
		void *entry0,
L
Linus Torvalds 已提交
779 780 781 782 783 784 785 786 787 788 789 790
		unsigned int size,
		unsigned int number,
		const unsigned int *hook_entries,
		const unsigned int *underflows)
{
	unsigned int i;
	int ret;

	newinfo->size = size;
	newinfo->number = number;

	/* Init all hooks to impossible value. */
791
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
792 793 794 795 796 797 798
		newinfo->hook_entry[i] = 0xFFFFFFFF;
		newinfo->underflow[i] = 0xFFFFFFFF;
	}

	duprintf("translate_table: size %u\n", newinfo->size);
	i = 0;
	/* Walk through entries, checking offsets. */
799
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
L
Linus Torvalds 已提交
800 801
				check_entry_size_and_hooks,
				newinfo,
802 803
				entry0,
				entry0 + size,
L
Linus Torvalds 已提交
804 805 806 807 808 809 810 811 812 813 814
				hook_entries, underflows, &i);
	if (ret != 0)
		return ret;

	if (i != number) {
		duprintf("translate_table: %u not %u entries\n",
			 i, number);
		return -EINVAL;
	}

	/* Check hooks all assigned */
815
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
816 817 818 819 820 821 822 823 824 825 826 827 828 829 830
		/* Only hooks which are valid */
		if (!(valid_hooks & (1 << i)))
			continue;
		if (newinfo->hook_entry[i] == 0xFFFFFFFF) {
			duprintf("Invalid hook entry %u %u\n",
				 i, hook_entries[i]);
			return -EINVAL;
		}
		if (newinfo->underflow[i] == 0xFFFFFFFF) {
			duprintf("Invalid underflow %u %u\n",
				 i, underflows[i]);
			return -EINVAL;
		}
	}

831 832 833
	if (!mark_source_chains(newinfo, valid_hooks, entry0))
		return -ELOOP;

L
Linus Torvalds 已提交
834 835
	/* Finally, each sanity check must pass */
	i = 0;
836
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
L
Linus Torvalds 已提交
837 838
				check_entry, name, size, &i);

839 840 841 842 843
	if (ret != 0) {
		IP6T_ENTRY_ITERATE(entry0, newinfo->size,
				   cleanup_entry, &i);
		return ret;
	}
L
Linus Torvalds 已提交
844 845

	/* And one copy for every other CPU */
846
	for_each_possible_cpu(i) {
847 848
		if (newinfo->entries[i] && newinfo->entries[i] != entry0)
			memcpy(newinfo->entries[i], entry0, newinfo->size);
L
Linus Torvalds 已提交
849 850
	}

851
	return 0;
L
Linus Torvalds 已提交
852 853 854 855 856
}

/* Gets counters. */
static inline int
add_entry_to_counter(const struct ip6t_entry *e,
857
		     struct xt_counters total[],
L
Linus Torvalds 已提交
858 859 860 861 862 863 864 865
		     unsigned int *i)
{
	ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt);

	(*i)++;
	return 0;
}

866 867 868 869 870 871 872 873 874 875 876
static inline int
set_entry_to_counter(const struct ip6t_entry *e,
		     struct ip6t_counters total[],
		     unsigned int *i)
{
	SET_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt);

	(*i)++;
	return 0;
}

L
Linus Torvalds 已提交
877
static void
878 879
get_counters(const struct xt_table_info *t,
	     struct xt_counters counters[])
L
Linus Torvalds 已提交
880 881 882
{
	unsigned int cpu;
	unsigned int i;
883 884 885 886 887 888 889 890 891 892 893 894 895 896 897
	unsigned int curcpu;

	/* Instead of clearing (by a previous call to memset())
	 * the counters and using adds, we set the counters
	 * with data used by 'current' CPU
	 * We dont care about preemption here.
	 */
	curcpu = raw_smp_processor_id();

	i = 0;
	IP6T_ENTRY_ITERATE(t->entries[curcpu],
			   t->size,
			   set_entry_to_counter,
			   counters,
			   &i);
L
Linus Torvalds 已提交
898

899
	for_each_possible_cpu(cpu) {
900 901
		if (cpu == curcpu)
			continue;
L
Linus Torvalds 已提交
902
		i = 0;
903
		IP6T_ENTRY_ITERATE(t->entries[cpu],
L
Linus Torvalds 已提交
904 905 906 907 908 909 910 911 912
				  t->size,
				  add_entry_to_counter,
				  counters,
				  &i);
	}
}

static int
copy_entries_to_user(unsigned int total_size,
913
		     struct xt_table *table,
L
Linus Torvalds 已提交
914 915 916 917
		     void __user *userptr)
{
	unsigned int off, num, countersize;
	struct ip6t_entry *e;
918 919
	struct xt_counters *counters;
	struct xt_table_info *private = table->private;
L
Linus Torvalds 已提交
920
	int ret = 0;
921
	void *loc_cpu_entry;
L
Linus Torvalds 已提交
922 923 924 925

	/* We need atomic snapshot of counters: rest doesn't change
	   (other than comefrom, which userspace doesn't care
	   about). */
926
	countersize = sizeof(struct xt_counters) * private->number;
L
Linus Torvalds 已提交
927 928 929 930 931 932 933
	counters = vmalloc(countersize);

	if (counters == NULL)
		return -ENOMEM;

	/* First, sum counters... */
	write_lock_bh(&table->lock);
934
	get_counters(private, counters);
L
Linus Torvalds 已提交
935 936
	write_unlock_bh(&table->lock);

937
	/* choose the copy that is on ourc node/cpu */
938
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
939
	if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) {
L
Linus Torvalds 已提交
940 941 942 943 944 945 946 947 948 949 950
		ret = -EFAULT;
		goto free_counters;
	}

	/* FIXME: use iterator macros --RR */
	/* ... then go back and fix counters and names */
	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
		unsigned int i;
		struct ip6t_entry_match *m;
		struct ip6t_entry_target *t;

951
		e = (struct ip6t_entry *)(loc_cpu_entry + off);
L
Linus Torvalds 已提交
952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996
		if (copy_to_user(userptr + off
				 + offsetof(struct ip6t_entry, counters),
				 &counters[num],
				 sizeof(counters[num])) != 0) {
			ret = -EFAULT;
			goto free_counters;
		}

		for (i = sizeof(struct ip6t_entry);
		     i < e->target_offset;
		     i += m->u.match_size) {
			m = (void *)e + i;

			if (copy_to_user(userptr + off + i
					 + offsetof(struct ip6t_entry_match,
						    u.user.name),
					 m->u.kernel.match->name,
					 strlen(m->u.kernel.match->name)+1)
			    != 0) {
				ret = -EFAULT;
				goto free_counters;
			}
		}

		t = ip6t_get_target(e);
		if (copy_to_user(userptr + off + e->target_offset
				 + offsetof(struct ip6t_entry_target,
					    u.user.name),
				 t->u.kernel.target->name,
				 strlen(t->u.kernel.target->name)+1) != 0) {
			ret = -EFAULT;
			goto free_counters;
		}
	}

 free_counters:
	vfree(counters);
	return ret;
}

static int
get_entries(const struct ip6t_get_entries *entries,
	    struct ip6t_get_entries __user *uptr)
{
	int ret;
997
	struct xt_table *t;
L
Linus Torvalds 已提交
998

999
	t = xt_find_table_lock(AF_INET6, entries->name);
1000
	if (t && !IS_ERR(t)) {
1001 1002 1003 1004
		struct xt_table_info *private = t->private;
		duprintf("t->private->number = %u\n", private->number);
		if (entries->size == private->size)
			ret = copy_entries_to_user(private->size,
L
Linus Torvalds 已提交
1005 1006 1007
						   t, uptr->entrytable);
		else {
			duprintf("get_entries: I've got %u not %u!\n",
1008
				 private->size, entries->size);
L
Linus Torvalds 已提交
1009 1010
			ret = -EINVAL;
		}
1011
		module_put(t->me);
1012
		xt_table_unlock(t);
L
Linus Torvalds 已提交
1013
	} else
1014
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1015 1016 1017 1018 1019 1020 1021 1022 1023

	return ret;
}

static int
do_replace(void __user *user, unsigned int len)
{
	int ret;
	struct ip6t_replace tmp;
1024 1025 1026
	struct xt_table *t;
	struct xt_table_info *newinfo, *oldinfo;
	struct xt_counters *counters;
1027
	void *loc_cpu_entry, *loc_cpu_old_entry;
L
Linus Torvalds 已提交
1028 1029 1030 1031

	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
		return -EFAULT;

1032 1033 1034 1035
	/* overflow check */
	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
		return -ENOMEM;

1036
	newinfo = xt_alloc_table_info(tmp.size);
L
Linus Torvalds 已提交
1037 1038 1039
	if (!newinfo)
		return -ENOMEM;

1040 1041 1042
	/* choose the copy that is on our node/cpu */
	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp),
L
Linus Torvalds 已提交
1043 1044 1045 1046 1047
			   tmp.size) != 0) {
		ret = -EFAULT;
		goto free_newinfo;
	}

1048
	counters = vmalloc(tmp.num_counters * sizeof(struct xt_counters));
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053 1054
	if (!counters) {
		ret = -ENOMEM;
		goto free_newinfo;
	}

	ret = translate_table(tmp.name, tmp.valid_hooks,
1055
			      newinfo, loc_cpu_entry, tmp.size, tmp.num_entries,
L
Linus Torvalds 已提交
1056 1057 1058 1059 1060 1061
			      tmp.hook_entry, tmp.underflow);
	if (ret != 0)
		goto free_newinfo_counters;

	duprintf("ip_tables: Translated table\n");

1062
	t = try_then_request_module(xt_find_table_lock(AF_INET6, tmp.name),
1063 1064 1065
				    "ip6table_%s", tmp.name);
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1066
		goto free_newinfo_counters_untrans;
1067
	}
L
Linus Torvalds 已提交
1068 1069 1070 1071 1072 1073

	/* You lied! */
	if (tmp.valid_hooks != t->valid_hooks) {
		duprintf("Valid hook crap: %08X vs %08X\n",
			 tmp.valid_hooks, t->valid_hooks);
		ret = -EINVAL;
1074
		goto put_module;
L
Linus Torvalds 已提交
1075 1076
	}

1077
	oldinfo = xt_replace_table(t, tmp.num_counters, newinfo, &ret);
L
Linus Torvalds 已提交
1078 1079 1080 1081 1082 1083
	if (!oldinfo)
		goto put_module;

	/* Update module usage count based on number of rules */
	duprintf("do_replace: oldnum=%u, initnum=%u, newnum=%u\n",
		oldinfo->number, oldinfo->initial_entries, newinfo->number);
1084 1085
	if ((oldinfo->number > oldinfo->initial_entries) ||
	    (newinfo->number <= oldinfo->initial_entries))
L
Linus Torvalds 已提交
1086 1087 1088 1089 1090 1091 1092 1093
		module_put(t->me);
	if ((oldinfo->number > oldinfo->initial_entries) &&
	    (newinfo->number <= oldinfo->initial_entries))
		module_put(t->me);

	/* Get the old counters. */
	get_counters(oldinfo, counters);
	/* Decrease module usage counts and free resource */
1094 1095
	loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()];
	IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,NULL);
1096
	xt_free_table_info(oldinfo);
L
Linus Torvalds 已提交
1097
	if (copy_to_user(tmp.counters, counters,
1098
			 sizeof(struct xt_counters) * tmp.num_counters) != 0)
L
Linus Torvalds 已提交
1099 1100
		ret = -EFAULT;
	vfree(counters);
1101
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1102 1103 1104 1105
	return ret;

 put_module:
	module_put(t->me);
1106
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1107
 free_newinfo_counters_untrans:
1108
	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry,NULL);
L
Linus Torvalds 已提交
1109 1110 1111
 free_newinfo_counters:
	vfree(counters);
 free_newinfo:
1112
	xt_free_table_info(newinfo);
L
Linus Torvalds 已提交
1113 1114 1115 1116 1117 1118 1119
	return ret;
}

/* We're lazy, and add to the first CPU; overflow works its fey magic
 * and everything is OK. */
static inline int
add_counter_to_entry(struct ip6t_entry *e,
1120
		     const struct xt_counters addme[],
L
Linus Torvalds 已提交
1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141
		     unsigned int *i)
{
#if 0
	duprintf("add_counter: Entry %u %lu/%lu + %lu/%lu\n",
		 *i,
		 (long unsigned int)e->counters.pcnt,
		 (long unsigned int)e->counters.bcnt,
		 (long unsigned int)addme[*i].pcnt,
		 (long unsigned int)addme[*i].bcnt);
#endif

	ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt);

	(*i)++;
	return 0;
}

static int
do_add_counters(void __user *user, unsigned int len)
{
	unsigned int i;
1142 1143 1144
	struct xt_counters_info tmp, *paddc;
	struct xt_table_info *private;
	struct xt_table *t;
1145
	int ret = 0;
1146
	void *loc_cpu_entry;
L
Linus Torvalds 已提交
1147 1148 1149 1150

	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
		return -EFAULT;

1151
	if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct xt_counters))
L
Linus Torvalds 已提交
1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162
		return -EINVAL;

	paddc = vmalloc(len);
	if (!paddc)
		return -ENOMEM;

	if (copy_from_user(paddc, user, len) != 0) {
		ret = -EFAULT;
		goto free;
	}

1163
	t = xt_find_table_lock(AF_INET6, tmp.name);
1164 1165
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1166
		goto free;
1167
	}
L
Linus Torvalds 已提交
1168 1169

	write_lock_bh(&t->lock);
1170
	private = t->private;
1171
	if (private->number != tmp.num_counters) {
L
Linus Torvalds 已提交
1172 1173 1174 1175 1176
		ret = -EINVAL;
		goto unlock_up_free;
	}

	i = 0;
1177
	/* Choose the copy that is on our node */
1178
	loc_cpu_entry = private->entries[smp_processor_id()];
1179
	IP6T_ENTRY_ITERATE(loc_cpu_entry,
1180
			  private->size,
L
Linus Torvalds 已提交
1181 1182 1183 1184 1185
			  add_counter_to_entry,
			  paddc->counters,
			  &i);
 unlock_up_free:
	write_unlock_bh(&t->lock);
1186
	xt_table_unlock(t);
1187
	module_put(t->me);
L
Linus Torvalds 已提交
1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
 free:
	vfree(paddc);

	return ret;
}

static int
do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len)
{
	int ret;

	if (!capable(CAP_NET_ADMIN))
		return -EPERM;

	switch (cmd) {
	case IP6T_SO_SET_REPLACE:
		ret = do_replace(user, len);
		break;

	case IP6T_SO_SET_ADD_COUNTERS:
		ret = do_add_counters(user, len);
		break;

	default:
		duprintf("do_ip6t_set_ctl:  unknown request %i\n", cmd);
		ret = -EINVAL;
	}

	return ret;
}

static int
do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
{
	int ret;

	if (!capable(CAP_NET_ADMIN))
		return -EPERM;

	switch (cmd) {
	case IP6T_SO_GET_INFO: {
		char name[IP6T_TABLE_MAXNAMELEN];
1230
		struct xt_table *t;
L
Linus Torvalds 已提交
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243

		if (*len != sizeof(struct ip6t_getinfo)) {
			duprintf("length %u != %u\n", *len,
				 sizeof(struct ip6t_getinfo));
			ret = -EINVAL;
			break;
		}

		if (copy_from_user(name, user, sizeof(name)) != 0) {
			ret = -EFAULT;
			break;
		}
		name[IP6T_TABLE_MAXNAMELEN-1] = '\0';
1244

1245
		t = try_then_request_module(xt_find_table_lock(AF_INET6, name),
1246 1247
					    "ip6table_%s", name);
		if (t && !IS_ERR(t)) {
L
Linus Torvalds 已提交
1248
			struct ip6t_getinfo info;
1249
			struct xt_table_info *private = t->private;
L
Linus Torvalds 已提交
1250 1251

			info.valid_hooks = t->valid_hooks;
1252
			memcpy(info.hook_entry, private->hook_entry,
L
Linus Torvalds 已提交
1253
			       sizeof(info.hook_entry));
1254
			memcpy(info.underflow, private->underflow,
L
Linus Torvalds 已提交
1255
			       sizeof(info.underflow));
1256 1257
			info.num_entries = private->number;
			info.size = private->size;
L
Linus Torvalds 已提交
1258 1259 1260 1261 1262 1263
			memcpy(info.name, name, sizeof(info.name));

			if (copy_to_user(user, &info, *len) != 0)
				ret = -EFAULT;
			else
				ret = 0;
1264
			xt_table_unlock(t);
1265 1266 1267
			module_put(t->me);
		} else
			ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287
	}
	break;

	case IP6T_SO_GET_ENTRIES: {
		struct ip6t_get_entries get;

		if (*len < sizeof(get)) {
			duprintf("get_entries: %u < %u\n", *len, sizeof(get));
			ret = -EINVAL;
		} else if (copy_from_user(&get, user, sizeof(get)) != 0) {
			ret = -EFAULT;
		} else if (*len != sizeof(struct ip6t_get_entries) + get.size) {
			duprintf("get_entries: %u != %u\n", *len,
				 sizeof(struct ip6t_get_entries) + get.size);
			ret = -EINVAL;
		} else
			ret = get_entries(&get, user);
		break;
	}

1288 1289 1290
	case IP6T_SO_GET_REVISION_MATCH:
	case IP6T_SO_GET_REVISION_TARGET: {
		struct ip6t_get_revision rev;
1291
		int target;
1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302

		if (*len != sizeof(rev)) {
			ret = -EINVAL;
			break;
		}
		if (copy_from_user(&rev, user, sizeof(rev)) != 0) {
			ret = -EFAULT;
			break;
		}

		if (cmd == IP6T_SO_GET_REVISION_TARGET)
1303
			target = 1;
1304
		else
1305
			target = 0;
1306

1307 1308 1309
		try_then_request_module(xt_find_revision(AF_INET6, rev.name,
							 rev.revision,
							 target, &ret),
1310 1311 1312 1313
					"ip6t_%s", rev.name);
		break;
	}

L
Linus Torvalds 已提交
1314 1315 1316 1317 1318 1319 1320 1321
	default:
		duprintf("do_ip6t_get_ctl: unknown request %i\n", cmd);
		ret = -EINVAL;
	}

	return ret;
}

1322
int ip6t_register_table(struct xt_table *table,
L
Linus Torvalds 已提交
1323 1324 1325
			const struct ip6t_replace *repl)
{
	int ret;
1326
	struct xt_table_info *newinfo;
1327
	struct xt_table_info bootstrap
L
Linus Torvalds 已提交
1328
		= { 0, 0, 0, { 0 }, { 0 }, { } };
1329
	void *loc_cpu_entry;
L
Linus Torvalds 已提交
1330

1331
	newinfo = xt_alloc_table_info(repl->size);
L
Linus Torvalds 已提交
1332 1333 1334
	if (!newinfo)
		return -ENOMEM;

1335 1336 1337
	/* choose the copy on our node/cpu */
	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
	memcpy(loc_cpu_entry, repl->entries, repl->size);
L
Linus Torvalds 已提交
1338 1339

	ret = translate_table(table->name, table->valid_hooks,
1340
			      newinfo, loc_cpu_entry, repl->size,
L
Linus Torvalds 已提交
1341 1342 1343 1344
			      repl->num_entries,
			      repl->hook_entry,
			      repl->underflow);
	if (ret != 0) {
1345
		xt_free_table_info(newinfo);
L
Linus Torvalds 已提交
1346 1347 1348
		return ret;
	}

1349 1350
	ret = xt_register_table(table, &bootstrap, newinfo);
	if (ret != 0) {
1351
		xt_free_table_info(newinfo);
L
Linus Torvalds 已提交
1352 1353 1354
		return ret;
	}

1355
	return 0;
L
Linus Torvalds 已提交
1356 1357
}

1358
void ip6t_unregister_table(struct xt_table *table)
L
Linus Torvalds 已提交
1359
{
1360
	struct xt_table_info *private;
1361 1362
	void *loc_cpu_entry;

1363
	private = xt_unregister_table(table);
L
Linus Torvalds 已提交
1364 1365

	/* Decrease module usage counts and free resources */
1366 1367 1368
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
	IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL);
	xt_free_table_info(private);
L
Linus Torvalds 已提交
1369 1370 1371
}

/* Returns 1 if the type and code is matched by the range, 0 otherwise */
1372
static inline bool
L
Linus Torvalds 已提交
1373 1374
icmp6_type_code_match(u_int8_t test_type, u_int8_t min_code, u_int8_t max_code,
		     u_int8_t type, u_int8_t code,
1375
		     bool invert)
L
Linus Torvalds 已提交
1376 1377 1378 1379 1380
{
	return (type == test_type && code >= min_code && code <= max_code)
		^ invert;
}

1381
static bool
L
Linus Torvalds 已提交
1382 1383 1384
icmp6_match(const struct sk_buff *skb,
	   const struct net_device *in,
	   const struct net_device *out,
1385
	   const struct xt_match *match,
L
Linus Torvalds 已提交
1386 1387 1388
	   const void *matchinfo,
	   int offset,
	   unsigned int protoff,
1389
	   bool *hotdrop)
L
Linus Torvalds 已提交
1390 1391 1392 1393 1394 1395
{
	struct icmp6hdr _icmp, *ic;
	const struct ip6t_icmp *icmpinfo = matchinfo;

	/* Must not be a fragment. */
	if (offset)
1396
		return false;
L
Linus Torvalds 已提交
1397 1398 1399 1400 1401 1402

	ic = skb_header_pointer(skb, protoff, sizeof(_icmp), &_icmp);
	if (ic == NULL) {
		/* We've been asked to examine this packet, and we
		   can't.  Hence, no choice but to drop. */
		duprintf("Dropping evil ICMP tinygram.\n");
1403
		*hotdrop = true;
1404
		return false;
L
Linus Torvalds 已提交
1405 1406 1407 1408 1409 1410 1411 1412 1413 1414
	}

	return icmp6_type_code_match(icmpinfo->type,
				     icmpinfo->code[0],
				     icmpinfo->code[1],
				     ic->icmp6_type, ic->icmp6_code,
				     !!(icmpinfo->invflags&IP6T_ICMP_INV));
}

/* Called when user tries to insert an entry of this type. */
1415
static bool
L
Linus Torvalds 已提交
1416
icmp6_checkentry(const char *tablename,
1417
	   const void *entry,
1418
	   const struct xt_match *match,
L
Linus Torvalds 已提交
1419 1420 1421 1422 1423
	   void *matchinfo,
	   unsigned int hook_mask)
{
	const struct ip6t_icmp *icmpinfo = matchinfo;

1424 1425
	/* Must specify no unknown invflags */
	return !(icmpinfo->invflags & ~IP6T_ICMP_INV);
L
Linus Torvalds 已提交
1426 1427 1428
}

/* The built-in targets: standard (NULL) and error. */
1429
static struct xt_target ip6t_standard_target __read_mostly = {
L
Linus Torvalds 已提交
1430
	.name		= IP6T_STANDARD_TARGET,
1431
	.targetsize	= sizeof(int),
1432
	.family		= AF_INET6,
L
Linus Torvalds 已提交
1433 1434
};

1435
static struct xt_target ip6t_error_target __read_mostly = {
L
Linus Torvalds 已提交
1436 1437
	.name		= IP6T_ERROR_TARGET,
	.target		= ip6t_error,
1438
	.targetsize	= IP6T_FUNCTION_MAXNAMELEN,
1439
	.family		= AF_INET6,
L
Linus Torvalds 已提交
1440 1441 1442 1443 1444 1445 1446 1447 1448 1449
};

static struct nf_sockopt_ops ip6t_sockopts = {
	.pf		= PF_INET6,
	.set_optmin	= IP6T_BASE_CTL,
	.set_optmax	= IP6T_SO_SET_MAX+1,
	.set		= do_ip6t_set_ctl,
	.get_optmin	= IP6T_BASE_CTL,
	.get_optmax	= IP6T_SO_GET_MAX+1,
	.get		= do_ip6t_get_ctl,
1450
	.owner		= THIS_MODULE,
L
Linus Torvalds 已提交
1451 1452
};

1453
static struct xt_match icmp6_matchstruct __read_mostly = {
L
Linus Torvalds 已提交
1454 1455
	.name		= "icmp6",
	.match		= &icmp6_match,
1456 1457 1458
	.matchsize	= sizeof(struct ip6t_icmp),
	.checkentry	= icmp6_checkentry,
	.proto		= IPPROTO_ICMPV6,
1459
	.family		= AF_INET6,
L
Linus Torvalds 已提交
1460 1461
};

1462
static int __init ip6_tables_init(void)
L
Linus Torvalds 已提交
1463 1464 1465
{
	int ret;

1466 1467 1468
	ret = xt_proto_init(AF_INET6);
	if (ret < 0)
		goto err1;
1469

L
Linus Torvalds 已提交
1470
	/* Noone else will be downing sem now, so we won't sleep */
1471 1472 1473 1474 1475 1476 1477 1478 1479
	ret = xt_register_target(&ip6t_standard_target);
	if (ret < 0)
		goto err2;
	ret = xt_register_target(&ip6t_error_target);
	if (ret < 0)
		goto err3;
	ret = xt_register_match(&icmp6_matchstruct);
	if (ret < 0)
		goto err4;
L
Linus Torvalds 已提交
1480 1481 1482

	/* Register setsockopt */
	ret = nf_register_sockopt(&ip6t_sockopts);
1483 1484
	if (ret < 0)
		goto err5;
L
Linus Torvalds 已提交
1485

1486
	printk(KERN_INFO "ip6_tables: (C) 2000-2006 Netfilter Core Team\n");
L
Linus Torvalds 已提交
1487
	return 0;
1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498

err5:
	xt_unregister_match(&icmp6_matchstruct);
err4:
	xt_unregister_target(&ip6t_error_target);
err3:
	xt_unregister_target(&ip6t_standard_target);
err2:
	xt_proto_fini(AF_INET6);
err1:
	return ret;
L
Linus Torvalds 已提交
1499 1500
}

1501
static void __exit ip6_tables_fini(void)
L
Linus Torvalds 已提交
1502 1503
{
	nf_unregister_sockopt(&ip6t_sockopts);
1504 1505 1506
	xt_unregister_match(&icmp6_matchstruct);
	xt_unregister_target(&ip6t_error_target);
	xt_unregister_target(&ip6t_standard_target);
1507
	xt_proto_fini(AF_INET6);
L
Linus Torvalds 已提交
1508 1509
}

1510
/*
1511 1512 1513 1514 1515 1516 1517
 * find the offset to specified header or the protocol number of last header
 * if target < 0. "last header" is transport protocol header, ESP, or
 * "No next header".
 *
 * If target header is found, its offset is set in *offset and return protocol
 * number. Otherwise, return -1.
 *
1518 1519 1520
 * If the first fragment doesn't contain the final protocol header or
 * NEXTHDR_NONE it is considered invalid.
 *
1521 1522 1523 1524
 * Note that non-1st fragment is special case that "the protocol number
 * of last header" is "next header" field in Fragment header. In this case,
 * *offset is meaningless and fragment offset is stored in *fragoff if fragoff
 * isn't NULL.
1525 1526
 *
 */
1527 1528
int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
		  int target, unsigned short *fragoff)
1529
{
1530
	unsigned int start = skb_network_offset(skb) + sizeof(struct ipv6hdr);
1531
	u8 nexthdr = ipv6_hdr(skb)->nexthdr;
1532 1533
	unsigned int len = skb->len - start;

1534 1535 1536
	if (fragoff)
		*fragoff = 0;

1537 1538 1539 1540
	while (nexthdr != target) {
		struct ipv6_opt_hdr _hdr, *hp;
		unsigned int hdrlen;

1541 1542 1543
		if ((!ipv6_ext_hdr(nexthdr)) || nexthdr == NEXTHDR_NONE) {
			if (target < 0)
				break;
1544
			return -ENOENT;
1545 1546
		}

1547 1548
		hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr);
		if (hp == NULL)
1549
			return -EBADMSG;
1550
		if (nexthdr == NEXTHDR_FRAGMENT) {
A
Al Viro 已提交
1551 1552
			unsigned short _frag_off;
			__be16 *fp;
1553 1554 1555 1556 1557 1558
			fp = skb_header_pointer(skb,
						start+offsetof(struct frag_hdr,
							       frag_off),
						sizeof(_frag_off),
						&_frag_off);
			if (fp == NULL)
1559
				return -EBADMSG;
1560

1561 1562 1563 1564
			_frag_off = ntohs(*fp) & ~0x7;
			if (_frag_off) {
				if (target < 0 &&
				    ((!ipv6_ext_hdr(hp->nexthdr)) ||
1565
				     hp->nexthdr == NEXTHDR_NONE)) {
1566 1567 1568 1569
					if (fragoff)
						*fragoff = _frag_off;
					return hp->nexthdr;
				}
1570
				return -ENOENT;
1571
			}
1572 1573
			hdrlen = 8;
		} else if (nexthdr == NEXTHDR_AUTH)
1574
			hdrlen = (hp->hdrlen + 2) << 2;
1575
		else
1576
			hdrlen = ipv6_optlen(hp);
1577 1578 1579 1580 1581 1582 1583

		nexthdr = hp->nexthdr;
		len -= hdrlen;
		start += hdrlen;
	}

	*offset = start;
1584
	return nexthdr;
1585 1586
}

L
Linus Torvalds 已提交
1587 1588 1589 1590
EXPORT_SYMBOL(ip6t_register_table);
EXPORT_SYMBOL(ip6t_unregister_table);
EXPORT_SYMBOL(ip6t_do_table);
EXPORT_SYMBOL(ip6t_ext_hdr);
1591
EXPORT_SYMBOL(ipv6_find_hdr);
L
Linus Torvalds 已提交
1592

1593 1594
module_init(ip6_tables_init);
module_exit(ip6_tables_fini);