ip6_tables.c 58.7 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
#include <linux/icmpv6.h>
#include <net/ipv6.h>
22
#include <net/compat.h>
L
Linus Torvalds 已提交
23
#include <asm/uaccess.h>
I
Ingo Molnar 已提交
24
#include <linux/mutex.h>
L
Linus Torvalds 已提交
25
#include <linux/proc_fs.h>
26
#include <linux/err.h>
27
#include <linux/cpumask.h>
L
Linus Torvalds 已提交
28 29

#include <linux/netfilter_ipv6/ip6_tables.h>
30
#include <linux/netfilter/x_tables.h>
31
#include <net/netfilter/nf_log.h>
L
Linus Torvalds 已提交
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

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",		\
58
		       __func__, __FILE__, __LINE__);	\
L
Linus Torvalds 已提交
59 60 61 62 63 64 65 66 67 68 69
} while(0)
#else
#define IP_NF_ASSERT(x)
#endif

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

70
/*
L
Linus Torvalds 已提交
71
   We keep a set of rules for each CPU, so we can avoid write-locking
72 73 74 75
   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
76 77 78 79

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

/* Check for an extension */
80
int
L
Linus Torvalds 已提交
81 82
ip6t_ext_hdr(u8 nexthdr)
{
83 84 85 86 87 88 89
	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
90 91 92
}

/* Returns whether matches rule or not. */
93
/* Performance critical - called for every packet */
94
static inline bool
L
Linus Torvalds 已提交
95 96 97 98 99
ip6_packet_match(const struct sk_buff *skb,
		 const char *indev,
		 const char *outdev,
		 const struct ip6t_ip6 *ip6info,
		 unsigned int *protoff,
100
		 int *fragoff, bool *hotdrop)
L
Linus Torvalds 已提交
101 102
{
	unsigned long ret;
103
	const struct ipv6hdr *ipv6 = ipv6_hdr(skb);
L
Linus Torvalds 已提交
104

105
#define FWINV(bool, invflg) ((bool) ^ !!(ip6info->invflags & (invflg)))
L
Linus Torvalds 已提交
106

107
	if (FWINV(ipv6_masked_addr_cmp(&ipv6->saddr, &ip6info->smsk,
108
				       &ip6info->src), IP6T_INV_SRCIP)
109
	    || FWINV(ipv6_masked_addr_cmp(&ipv6->daddr, &ip6info->dmsk,
110
					  &ip6info->dst), IP6T_INV_DSTIP)) {
L
Linus Torvalds 已提交
111 112 113 114 115 116 117 118
		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)" : "");*/
119
		return false;
L
Linus Torvalds 已提交
120 121
	}

122
	ret = ifname_compare_aligned(indev, ip6info->iniface, ip6info->iniface_mask);
L
Linus Torvalds 已提交
123 124 125 126 127

	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)":"");
128
		return false;
L
Linus Torvalds 已提交
129 130
	}

131
	ret = ifname_compare_aligned(outdev, ip6info->outiface, ip6info->outiface_mask);
L
Linus Torvalds 已提交
132 133 134 135 136

	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)":"");
137
		return false;
L
Linus Torvalds 已提交
138 139 140 141 142 143
	}

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

	/* look for the desired protocol header */
	if((ip6info->flags & IP6T_F_PROTO)) {
144 145
		int protohdr;
		unsigned short _frag_off;
L
Linus Torvalds 已提交
146

147
		protohdr = ipv6_find_hdr(skb, protoff, -1, &_frag_off);
148 149
		if (protohdr < 0) {
			if (_frag_off == 0)
150
				*hotdrop = true;
151
			return false;
152
		}
153
		*fragoff = _frag_off;
L
Linus Torvalds 已提交
154 155

		dprintf("Packet protocol %hi ?= %s%hi.\n",
156
				protohdr,
L
Linus Torvalds 已提交
157 158 159
				ip6info->invflags & IP6T_INV_PROTO ? "!":"",
				ip6info->proto);

160
		if (ip6info->proto == protohdr) {
L
Linus Torvalds 已提交
161
			if(ip6info->invflags & IP6T_INV_PROTO) {
162
				return false;
L
Linus Torvalds 已提交
163
			}
164
			return true;
L
Linus Torvalds 已提交
165 166 167 168 169
		}

		/* We need match for the '-p all', too! */
		if ((ip6info->proto != 0) &&
			!(ip6info->invflags & IP6T_INV_PROTO))
170
			return false;
L
Linus Torvalds 已提交
171
	}
172
	return true;
L
Linus Torvalds 已提交
173 174 175
}

/* should be ip6 safe */
176
static bool
L
Linus Torvalds 已提交
177 178 179 180 181
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);
182
		return false;
L
Linus Torvalds 已提交
183 184 185 186
	}
	if (ipv6->invflags & ~IP6T_INV_MASK) {
		duprintf("Unknown invflag bits set: %08X\n",
			 ipv6->invflags & ~IP6T_INV_MASK);
187
		return false;
L
Linus Torvalds 已提交
188
	}
189
	return true;
L
Linus Torvalds 已提交
190 191 192
}

static unsigned int
193
ip6t_error(struct sk_buff *skb, const struct xt_target_param *par)
L
Linus Torvalds 已提交
194 195
{
	if (net_ratelimit())
196 197
		printk("ip6_tables: error: `%s'\n",
		       (const char *)par->targinfo);
L
Linus Torvalds 已提交
198 199 200 201

	return NF_DROP;
}

202 203
/* Performance critical - called for every packet */
static inline bool
204 205
do_match(struct ip6t_entry_match *m, const struct sk_buff *skb,
	 struct xt_match_param *par)
L
Linus Torvalds 已提交
206
{
207 208 209
	par->match     = m->u.kernel.match;
	par->matchinfo = m->data;

L
Linus Torvalds 已提交
210
	/* Stop iteration if it doesn't match */
211
	if (!m->u.kernel.match->match(skb, par))
212
		return true;
L
Linus Torvalds 已提交
213
	else
214
		return false;
L
Linus Torvalds 已提交
215 216 217 218 219 220 221 222
}

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

223
/* All zeroes == unconditional rule. */
224
/* Mildly perf critical (only if packet tracing is on) */
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239
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! */
240
static const char *const hooknames[] = {
241 242 243 244 245
	[NF_INET_PRE_ROUTING]		= "PREROUTING",
	[NF_INET_LOCAL_IN]		= "INPUT",
	[NF_INET_FORWARD]		= "FORWARD",
	[NF_INET_LOCAL_OUT]		= "OUTPUT",
	[NF_INET_POST_ROUTING]		= "POSTROUTING",
246 247 248 249 250 251 252 253
};

enum nf_ip_trace_comments {
	NF_IP6_TRACE_COMMENT_RULE,
	NF_IP6_TRACE_COMMENT_RETURN,
	NF_IP6_TRACE_COMMENT_POLICY,
};

254
static const char *const comments[] = {
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269
	[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,
		},
	},
};

270
/* Mildly perf critical (only if packet tracing is on) */
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
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,
306
			 const char *tablename,
307 308 309 310
			 struct xt_table_info *private,
			 struct ip6t_entry *e)
{
	void *table_base;
311
	const struct ip6t_entry *root;
312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331
	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
332 333
/* Returns one of the generic firewall policies, like NF_ACCEPT. */
unsigned int
334
ip6t_do_table(struct sk_buff *skb,
L
Linus Torvalds 已提交
335 336 337
	      unsigned int hook,
	      const struct net_device *in,
	      const struct net_device *out,
338
	      struct xt_table *table)
L
Linus Torvalds 已提交
339
{
340
	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
341
	bool hotdrop = false;
L
Linus Torvalds 已提交
342 343 344 345 346
	/* 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;
347
	struct xt_table_info *private;
348
	struct xt_match_param mtpar;
349
	struct xt_target_param tgpar;
L
Linus Torvalds 已提交
350 351 352 353 354 355 356 357 358 359

	/* 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. */
360
	mtpar.hotdrop = &hotdrop;
361 362
	mtpar.in      = tgpar.in  = in;
	mtpar.out     = tgpar.out = out;
363
	mtpar.family  = tgpar.family = NFPROTO_IPV6;
364
	tgpar.hooknum = hook;
L
Linus Torvalds 已提交
365 366

	IP_NF_ASSERT(table->valid_hooks & (1 << hook));
367

368
	rcu_read_lock_bh();
369 370 371
	private = rcu_dereference(table->private);
	table_base = rcu_dereference(private->entries[smp_processor_id()]);

372
	e = get_entry(table_base, private->hook_entry[hook]);
L
Linus Torvalds 已提交
373 374

	/* For return from builtin chain */
375
	back = get_entry(table_base, private->underflow[hook]);
L
Linus Torvalds 已提交
376 377 378 379

	do {
		IP_NF_ASSERT(e);
		IP_NF_ASSERT(back);
380
		if (ip6_packet_match(skb, indev, outdev, &e->ipv6,
381
			&mtpar.thoff, &mtpar.fragoff, &hotdrop)) {
L
Linus Torvalds 已提交
382 383
			struct ip6t_entry_target *t;

384
			if (IP6T_MATCH_ITERATE(e, do_match, skb, &mtpar) != 0)
L
Linus Torvalds 已提交
385 386 387
				goto no_match;

			ADD_COUNTER(e->counters,
388 389
				    ntohs(ipv6_hdr(skb)->payload_len) +
				    sizeof(struct ipv6hdr), 1);
L
Linus Torvalds 已提交
390 391 392

			t = ip6t_get_target(e);
			IP_NF_ASSERT(t->u.kernel.target);
393 394 395 396

#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
    defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
			/* The packet is traced: log it */
397 398
			if (unlikely(skb->nf_trace))
				trace_packet(skb, hook, in, out,
399 400
					     table->name, private, e);
#endif
L
Linus Torvalds 已提交
401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416
			/* 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
417 418
				if (table_base + v != (void *)e + e->next_offset
				    && !(e->ipv6.flags & IP6T_F_GOTO)) {
L
Linus Torvalds 已提交
419 420 421 422 423 424 425 426 427 428 429 430
					/* 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
431
				   abs. verdicts */
432 433 434
				tgpar.target   = t->u.kernel.target;
				tgpar.targinfo = t->data;

L
Linus Torvalds 已提交
435 436 437 438
#ifdef CONFIG_NETFILTER_DEBUG
				((struct ip6t_entry *)table_base)->comefrom
					= 0xeeeeeeec;
#endif
439
				verdict = t->u.kernel.target->target(skb,
440
								     &tgpar);
L
Linus Torvalds 已提交
441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466

#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
467
	((struct ip6t_entry *)table_base)->comefrom = NETFILTER_LINK_POISON;
L
Linus Torvalds 已提交
468
#endif
469
	rcu_read_unlock_bh();
L
Linus Torvalds 已提交
470 471 472 473 474 475 476 477 478 479 480 481 482

#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
483
mark_source_chains(struct xt_table_info *newinfo,
484
		   unsigned int valid_hooks, void *entry0)
L
Linus Torvalds 已提交
485 486 487 488 489
{
	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 */
490
	for (hook = 0; hook < NF_INET_NUMHOOKS; hook++) {
L
Linus Torvalds 已提交
491
		unsigned int pos = newinfo->hook_entry[hook];
492
		struct ip6t_entry *e = (struct ip6t_entry *)(entry0 + pos);
L
Linus Torvalds 已提交
493 494 495 496 497 498 499 500 501 502

		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);
503
			int visited = e->comefrom & (1 << hook);
L
Linus Torvalds 已提交
504

505
			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
506 507 508 509
				printk("iptables: loop hook %u pos %u %08X.\n",
				       hook, pos, e->comefrom);
				return 0;
			}
510
			e->comefrom |= ((1 << hook) | (1 << NF_INET_NUMHOOKS));
L
Linus Torvalds 已提交
511 512

			/* Unconditional return/END. */
513
			if ((e->target_offset == sizeof(struct ip6t_entry)
L
Linus Torvalds 已提交
514 515 516
			    && (strcmp(t->target.u.user.name,
				       IP6T_STANDARD_TARGET) == 0)
			    && t->verdict < 0
517
			    && unconditional(&e->ipv6)) || visited) {
L
Linus Torvalds 已提交
518 519
				unsigned int oldpos, size;

520 521 522
				if ((strcmp(t->target.u.user.name,
					    IP6T_STANDARD_TARGET) == 0) &&
				    t->verdict < -NF_MAX_VERDICT - 1) {
523 524 525 526 527 528
					duprintf("mark_source_chains: bad "
						"negative verdict (%i)\n",
								t->verdict);
					return 0;
				}

L
Linus Torvalds 已提交
529 530 531
				/* Return: backtrack through the last
				   big jump. */
				do {
532
					e->comefrom ^= (1<<NF_INET_NUMHOOKS);
L
Linus Torvalds 已提交
533 534
#ifdef DEBUG_IP_FIREWALL_USER
					if (e->comefrom
535
					    & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
536 537 538 539 540 541 542 543 544 545 546 547 548 549 550
						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 *)
551
						(entry0 + pos);
L
Linus Torvalds 已提交
552 553 554 555 556
				} while (oldpos == pos + e->next_offset);

				/* Move along one */
				size = e->next_offset;
				e = (struct ip6t_entry *)
557
					(entry0 + pos + size);
L
Linus Torvalds 已提交
558 559 560 561 562 563 564 565
				e->counters.pcnt = pos;
				pos += size;
			} else {
				int newpos = t->verdict;

				if (strcmp(t->target.u.user.name,
					   IP6T_STANDARD_TARGET) == 0
				    && newpos >= 0) {
566 567 568 569 570 571 572
					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
573 574 575 576 577 578 579 580
					/* 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 *)
581
					(entry0 + newpos);
L
Linus Torvalds 已提交
582 583 584 585 586 587 588 589 590 591
				e->counters.pcnt = pos;
				pos = newpos;
			}
		}
		next:
		duprintf("Finished chain %u\n", hook);
	}
	return 1;
}

592
static int
L
Linus Torvalds 已提交
593 594
cleanup_match(struct ip6t_entry_match *m, unsigned int *i)
{
595 596
	struct xt_mtdtor_param par;

L
Linus Torvalds 已提交
597 598 599
	if (i && (*i)-- == 0)
		return 1;

600 601
	par.match     = m->u.kernel.match;
	par.matchinfo = m->data;
602
	par.family    = NFPROTO_IPV6;
603 604 605
	if (par.match->destroy != NULL)
		par.match->destroy(&par);
	module_put(par.match->me);
L
Linus Torvalds 已提交
606 607 608
	return 0;
}

609
static int
610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629
check_entry(struct ip6t_entry *e, const char *name)
{
	struct ip6t_entry_target *t;

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

	if (e->target_offset + sizeof(struct ip6t_entry_target) >
	    e->next_offset)
		return -EINVAL;

	t = ip6t_get_target(e);
	if (e->target_offset + t->u.target_size > e->next_offset)
		return -EINVAL;

	return 0;
}

630 631
static int check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par,
		       unsigned int *i)
632
{
633
	const struct ip6t_ip6 *ipv6 = par->entryinfo;
634 635
	int ret;

636 637 638
	par->match     = m->u.kernel.match;
	par->matchinfo = m->data;

639
	ret = xt_check_match(par, m->u.match_size - sizeof(*m),
640
			     ipv6->proto, ipv6->invflags & IP6T_INV_PROTO);
641
	if (ret < 0) {
642
		duprintf("ip_tables: check failed for `%s'.\n",
643
			 par.match->name);
644
		return ret;
645
	}
646 647
	++*i;
	return 0;
648 649
}

650
static int
651
find_check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par,
652
		 unsigned int *i)
L
Linus Torvalds 已提交
653
{
654
	struct xt_match *match;
655
	int ret;
L
Linus Torvalds 已提交
656

657
	match = try_then_request_module(xt_find_match(AF_INET6, m->u.user.name,
658
						      m->u.user.revision),
659 660
					"ip6t_%s", m->u.user.name);
	if (IS_ERR(match) || !match) {
661
		duprintf("find_check_match: `%s' not found\n", m->u.user.name);
662
		return match ? PTR_ERR(match) : -ENOENT;
L
Linus Torvalds 已提交
663 664 665
	}
	m->u.kernel.match = match;

666
	ret = check_match(m, par, i);
667 668 669
	if (ret)
		goto err;

L
Linus Torvalds 已提交
670
	return 0;
671 672 673
err:
	module_put(m->u.kernel.match->me);
	return ret;
L
Linus Torvalds 已提交
674 675
}

676
static int check_target(struct ip6t_entry *e, const char *name)
L
Linus Torvalds 已提交
677
{
678 679 680 681 682 683 684
	struct ip6t_entry_target *t = ip6t_get_target(e);
	struct xt_tgchk_param par = {
		.table     = name,
		.entryinfo = e,
		.target    = t->u.kernel.target,
		.targinfo  = t->data,
		.hook_mask = e->comefrom,
685
		.family    = NFPROTO_IPV6,
686
	};
L
Linus Torvalds 已提交
687 688
	int ret;

689
	t = ip6t_get_target(e);
690
	ret = xt_check_target(&par, t->u.target_size - sizeof(*t),
691
	      e->ipv6.proto, e->ipv6.invflags & IP6T_INV_PROTO);
692
	if (ret < 0) {
693 694
		duprintf("ip_tables: check failed for `%s'.\n",
			 t->u.kernel.target->name);
695
		return ret;
L
Linus Torvalds 已提交
696
	}
697
	return 0;
698
}
L
Linus Torvalds 已提交
699

700
static int
701 702 703 704 705 706 707
find_check_entry(struct ip6t_entry *e, const char *name, unsigned int size,
		 unsigned int *i)
{
	struct ip6t_entry_target *t;
	struct xt_target *target;
	int ret;
	unsigned int j;
708
	struct xt_mtchk_param mtpar;
709 710 711 712

	ret = check_entry(e, name);
	if (ret)
		return ret;
713

L
Linus Torvalds 已提交
714
	j = 0;
715 716 717
	mtpar.table     = name;
	mtpar.entryinfo = &e->ipv6;
	mtpar.hook_mask = e->comefrom;
718
	mtpar.family    = NFPROTO_IPV6;
719
	ret = IP6T_MATCH_ITERATE(e, find_check_match, &mtpar, &j);
L
Linus Torvalds 已提交
720 721 722 723
	if (ret != 0)
		goto cleanup_matches;

	t = ip6t_get_target(e);
724 725 726
	target = try_then_request_module(xt_find_target(AF_INET6,
							t->u.user.name,
							t->u.user.revision),
727 728
					 "ip6t_%s", t->u.user.name);
	if (IS_ERR(target) || !target) {
729
		duprintf("find_check_entry: `%s' not found\n", t->u.user.name);
730
		ret = target ? PTR_ERR(target) : -ENOENT;
L
Linus Torvalds 已提交
731 732 733
		goto cleanup_matches;
	}
	t->u.kernel.target = target;
734

735
	ret = check_target(e, name);
736 737 738
	if (ret)
		goto err;

L
Linus Torvalds 已提交
739 740
	(*i)++;
	return 0;
741 742
 err:
	module_put(t->u.kernel.target->me);
L
Linus Torvalds 已提交
743 744 745 746 747
 cleanup_matches:
	IP6T_MATCH_ITERATE(e, cleanup_match, &j);
	return ret;
}

748
static int
L
Linus Torvalds 已提交
749
check_entry_size_and_hooks(struct ip6t_entry *e,
750
			   struct xt_table_info *newinfo,
L
Linus Torvalds 已提交
751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772
			   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 */
773
	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
L
Linus Torvalds 已提交
774 775 776 777 778 779 780
		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
781
	   < 0 (not IP6T_RETURN). --RR */
L
Linus Torvalds 已提交
782 783

	/* Clear counters and comefrom */
784
	e->counters = ((struct xt_counters) { 0, 0 });
L
Linus Torvalds 已提交
785 786 787 788 789 790
	e->comefrom = 0;

	(*i)++;
	return 0;
}

791
static int
L
Linus Torvalds 已提交
792 793
cleanup_entry(struct ip6t_entry *e, unsigned int *i)
{
794
	struct xt_tgdtor_param par;
L
Linus Torvalds 已提交
795 796 797 798 799 800 801 802
	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);
803 804 805

	par.target   = t->u.kernel.target;
	par.targinfo = t->data;
806
	par.family   = NFPROTO_IPV6;
807 808 809
	if (par.target->destroy != NULL)
		par.target->destroy(&par);
	module_put(par.target->me);
L
Linus Torvalds 已提交
810 811 812 813 814 815 816 817
	return 0;
}

/* Checks and translates the user-supplied table segment (held in
   newinfo) */
static int
translate_table(const char *name,
		unsigned int valid_hooks,
818
		struct xt_table_info *newinfo,
819
		void *entry0,
L
Linus Torvalds 已提交
820 821 822 823 824 825 826 827 828 829 830 831
		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. */
832
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
833 834 835 836 837 838 839
		newinfo->hook_entry[i] = 0xFFFFFFFF;
		newinfo->underflow[i] = 0xFFFFFFFF;
	}

	duprintf("translate_table: size %u\n", newinfo->size);
	i = 0;
	/* Walk through entries, checking offsets. */
840
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
L
Linus Torvalds 已提交
841 842
				check_entry_size_and_hooks,
				newinfo,
843 844
				entry0,
				entry0 + size,
L
Linus Torvalds 已提交
845 846 847 848 849 850 851 852 853 854 855
				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 */
856
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
857 858 859 860 861 862 863 864 865 866 867 868 869 870 871
		/* 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;
		}
	}

872 873 874
	if (!mark_source_chains(newinfo, valid_hooks, entry0))
		return -ELOOP;

L
Linus Torvalds 已提交
875 876
	/* Finally, each sanity check must pass */
	i = 0;
877
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
878
				find_check_entry, name, size, &i);
L
Linus Torvalds 已提交
879

880 881 882 883 884
	if (ret != 0) {
		IP6T_ENTRY_ITERATE(entry0, newinfo->size,
				   cleanup_entry, &i);
		return ret;
	}
L
Linus Torvalds 已提交
885 886

	/* And one copy for every other CPU */
887
	for_each_possible_cpu(i) {
888 889
		if (newinfo->entries[i] && newinfo->entries[i] != entry0)
			memcpy(newinfo->entries[i], entry0, newinfo->size);
L
Linus Torvalds 已提交
890 891
	}

892
	return ret;
L
Linus Torvalds 已提交
893 894 895 896 897
}

/* Gets counters. */
static inline int
add_entry_to_counter(const struct ip6t_entry *e,
898
		     struct xt_counters total[],
L
Linus Torvalds 已提交
899 900 901 902 903 904 905 906
		     unsigned int *i)
{
	ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt);

	(*i)++;
	return 0;
}

907 908 909 910 911 912 913 914 915 916 917
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
918
static void
919 920
get_counters(const struct xt_table_info *t,
	     struct xt_counters counters[])
L
Linus Torvalds 已提交
921 922 923
{
	unsigned int cpu;
	unsigned int i;
924 925 926 927 928 929 930 931 932 933 934 935 936 937 938
	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
939

940
	for_each_possible_cpu(cpu) {
941 942
		if (cpu == curcpu)
			continue;
L
Linus Torvalds 已提交
943
		i = 0;
944
		IP6T_ENTRY_ITERATE(t->entries[cpu],
L
Linus Torvalds 已提交
945 946 947 948 949 950 951
				  t->size,
				  add_entry_to_counter,
				  counters,
				  &i);
	}
}

952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003
/* We're lazy, and add to the first CPU; overflow works its fey magic
 * and everything is OK. */
static int
add_counter_to_entry(struct ip6t_entry *e,
		     const struct xt_counters addme[],
		     unsigned int *i)
{
	ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt);

	(*i)++;
	return 0;
}

/* Take values from counters and add them back onto the current cpu */
static void put_counters(struct xt_table_info *t,
			 const struct xt_counters counters[])
{
	unsigned int i, cpu;

	local_bh_disable();
	cpu = smp_processor_id();
	i = 0;
	IP6T_ENTRY_ITERATE(t->entries[cpu],
			   t->size,
			   add_counter_to_entry,
			   counters,
			   &i);
	local_bh_enable();
}

static inline int
zero_entry_counter(struct ip6t_entry *e, void *arg)
{
	e->counters.bcnt = 0;
	e->counters.pcnt = 0;
	return 0;
}

static void
clone_counters(struct xt_table_info *newinfo, const struct xt_table_info *info)
{
	unsigned int cpu;
	const void *loc_cpu_entry = info->entries[raw_smp_processor_id()];

	memcpy(newinfo, info, offsetof(struct xt_table_info, entries));
	for_each_possible_cpu(cpu) {
		memcpy(newinfo->entries[cpu], loc_cpu_entry, info->size);
		IP6T_ENTRY_ITERATE(newinfo->entries[cpu], newinfo->size,
				   zero_entry_counter, NULL);
	}
}

1004
static struct xt_counters *alloc_counters(struct xt_table *table)
L
Linus Torvalds 已提交
1005
{
1006
	unsigned int countersize;
1007
	struct xt_counters *counters;
1008 1009
	struct xt_table_info *private = table->private;
	struct xt_table_info *info;
L
Linus Torvalds 已提交
1010 1011 1012 1013

	/* We need atomic snapshot of counters: rest doesn't change
	   (other than comefrom, which userspace doesn't care
	   about). */
1014
	countersize = sizeof(struct xt_counters) * private->number;
1015
	counters = vmalloc_node(countersize, numa_node_id());
L
Linus Torvalds 已提交
1016 1017

	if (counters == NULL)
1018
		goto nomem;
L
Linus Torvalds 已提交
1019

1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032
	info = xt_alloc_table_info(private->size);
	if (!info)
		goto free_counters;

	clone_counters(info, private);

	mutex_lock(&table->lock);
	xt_table_entry_swap_rcu(private, info);
	synchronize_net();	/* Wait until smoke has cleared */

	get_counters(info, counters);
	put_counters(private, counters);
	mutex_unlock(&table->lock);
L
Linus Torvalds 已提交
1033

1034 1035 1036 1037 1038 1039
	xt_free_table_info(info);

 free_counters:
	vfree(counters);
 nomem:
	return ERR_PTR(-ENOMEM);
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049
}

static int
copy_entries_to_user(unsigned int total_size,
		     struct xt_table *table,
		     void __user *userptr)
{
	unsigned int off, num;
	struct ip6t_entry *e;
	struct xt_counters *counters;
1050
	const struct xt_table_info *private = table->private;
1051
	int ret = 0;
1052
	const void *loc_cpu_entry;
1053 1054 1055 1056 1057

	counters = alloc_counters(table);
	if (IS_ERR(counters))
		return PTR_ERR(counters);

1058 1059 1060 1061
	/* choose the copy that is on our node/cpu, ...
	 * This choice is lazy (because current thread is
	 * allowed to migrate to another cpu)
	 */
1062
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
1063
	if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) {
L
Linus Torvalds 已提交
1064 1065 1066 1067 1068 1069 1070 1071
		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;
1072 1073
		const struct ip6t_entry_match *m;
		const struct ip6t_entry_target *t;
L
Linus Torvalds 已提交
1074

1075
		e = (struct ip6t_entry *)(loc_cpu_entry + off);
L
Linus Torvalds 已提交
1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115
		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;
}

1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188
#ifdef CONFIG_COMPAT
static void compat_standard_from_user(void *dst, void *src)
{
	int v = *(compat_int_t *)src;

	if (v > 0)
		v += xt_compat_calc_jump(AF_INET6, v);
	memcpy(dst, &v, sizeof(v));
}

static int compat_standard_to_user(void __user *dst, void *src)
{
	compat_int_t cv = *(int *)src;

	if (cv > 0)
		cv -= xt_compat_calc_jump(AF_INET6, cv);
	return copy_to_user(dst, &cv, sizeof(cv)) ? -EFAULT : 0;
}

static inline int
compat_calc_match(struct ip6t_entry_match *m, int *size)
{
	*size += xt_compat_match_offset(m->u.kernel.match);
	return 0;
}

static int compat_calc_entry(struct ip6t_entry *e,
			     const struct xt_table_info *info,
			     void *base, struct xt_table_info *newinfo)
{
	struct ip6t_entry_target *t;
	unsigned int entry_offset;
	int off, i, ret;

	off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
	entry_offset = (void *)e - base;
	IP6T_MATCH_ITERATE(e, compat_calc_match, &off);
	t = ip6t_get_target(e);
	off += xt_compat_target_offset(t->u.kernel.target);
	newinfo->size -= off;
	ret = xt_compat_add_offset(AF_INET6, entry_offset, off);
	if (ret)
		return ret;

	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
		if (info->hook_entry[i] &&
		    (e < (struct ip6t_entry *)(base + info->hook_entry[i])))
			newinfo->hook_entry[i] -= off;
		if (info->underflow[i] &&
		    (e < (struct ip6t_entry *)(base + info->underflow[i])))
			newinfo->underflow[i] -= off;
	}
	return 0;
}

static int compat_table_info(const struct xt_table_info *info,
			     struct xt_table_info *newinfo)
{
	void *loc_cpu_entry;

	if (!newinfo || !info)
		return -EINVAL;

	/* we dont care about newinfo->entries[] */
	memcpy(newinfo, info, offsetof(struct xt_table_info, entries));
	newinfo->initial_entries = 0;
	loc_cpu_entry = info->entries[raw_smp_processor_id()];
	return IP6T_ENTRY_ITERATE(loc_cpu_entry, info->size,
				  compat_calc_entry, info, loc_cpu_entry,
				  newinfo);
}
#endif

1189
static int get_info(struct net *net, void __user *user, int *len, int compat)
1190 1191 1192 1193 1194 1195
{
	char name[IP6T_TABLE_MAXNAMELEN];
	struct xt_table *t;
	int ret;

	if (*len != sizeof(struct ip6t_getinfo)) {
1196
		duprintf("length %u != %zu\n", *len,
1197 1198 1199 1200 1201 1202 1203 1204
			 sizeof(struct ip6t_getinfo));
		return -EINVAL;
	}

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

	name[IP6T_TABLE_MAXNAMELEN-1] = '\0';
1205 1206 1207 1208
#ifdef CONFIG_COMPAT
	if (compat)
		xt_compat_lock(AF_INET6);
#endif
1209
	t = try_then_request_module(xt_find_table_lock(net, AF_INET6, name),
1210 1211 1212
				    "ip6table_%s", name);
	if (t && !IS_ERR(t)) {
		struct ip6t_getinfo info;
1213
		const struct xt_table_info *private = t->private;
1214

1215 1216 1217 1218 1219 1220 1221 1222
#ifdef CONFIG_COMPAT
		if (compat) {
			struct xt_table_info tmp;
			ret = compat_table_info(private, &tmp);
			xt_compat_flush_offsets(AF_INET6);
			private = &tmp;
		}
#endif
1223 1224 1225 1226 1227 1228 1229
		info.valid_hooks = t->valid_hooks;
		memcpy(info.hook_entry, private->hook_entry,
		       sizeof(info.hook_entry));
		memcpy(info.underflow, private->underflow,
		       sizeof(info.underflow));
		info.num_entries = private->number;
		info.size = private->size;
1230
		strcpy(info.name, name);
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240

		if (copy_to_user(user, &info, *len) != 0)
			ret = -EFAULT;
		else
			ret = 0;

		xt_table_unlock(t);
		module_put(t->me);
	} else
		ret = t ? PTR_ERR(t) : -ENOENT;
1241 1242 1243 1244
#ifdef CONFIG_COMPAT
	if (compat)
		xt_compat_unlock(AF_INET6);
#endif
1245 1246 1247
	return ret;
}

L
Linus Torvalds 已提交
1248
static int
1249
get_entries(struct net *net, struct ip6t_get_entries __user *uptr, int *len)
L
Linus Torvalds 已提交
1250 1251
{
	int ret;
1252
	struct ip6t_get_entries get;
1253
	struct xt_table *t;
L
Linus Torvalds 已提交
1254

1255
	if (*len < sizeof(get)) {
1256
		duprintf("get_entries: %u < %zu\n", *len, sizeof(get));
1257 1258 1259 1260 1261
		return -EINVAL;
	}
	if (copy_from_user(&get, uptr, sizeof(get)) != 0)
		return -EFAULT;
	if (*len != sizeof(struct ip6t_get_entries) + get.size) {
1262 1263
		duprintf("get_entries: %u != %zu\n",
			 *len, sizeof(get) + get.size);
1264 1265 1266
		return -EINVAL;
	}

1267
	t = xt_find_table_lock(net, AF_INET6, get.name);
1268
	if (t && !IS_ERR(t)) {
1269 1270
		struct xt_table_info *private = t->private;
		duprintf("t->private->number = %u\n", private->number);
1271
		if (get.size == private->size)
1272
			ret = copy_entries_to_user(private->size,
L
Linus Torvalds 已提交
1273 1274 1275
						   t, uptr->entrytable);
		else {
			duprintf("get_entries: I've got %u not %u!\n",
1276
				 private->size, get.size);
1277
			ret = -EAGAIN;
L
Linus Torvalds 已提交
1278
		}
1279
		module_put(t->me);
1280
		xt_table_unlock(t);
L
Linus Torvalds 已提交
1281
	} else
1282
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1283 1284 1285 1286 1287

	return ret;
}

static int
1288
__do_replace(struct net *net, const char *name, unsigned int valid_hooks,
1289 1290
	     struct xt_table_info *newinfo, unsigned int num_counters,
	     void __user *counters_ptr)
L
Linus Torvalds 已提交
1291 1292
{
	int ret;
1293
	struct xt_table *t;
1294
	struct xt_table_info *oldinfo;
1295
	struct xt_counters *counters;
1296
	const void *loc_cpu_old_entry;
L
Linus Torvalds 已提交
1297

1298 1299
	ret = 0;
	counters = vmalloc_node(num_counters * sizeof(struct xt_counters),
1300
				numa_node_id());
L
Linus Torvalds 已提交
1301 1302
	if (!counters) {
		ret = -ENOMEM;
1303
		goto out;
L
Linus Torvalds 已提交
1304 1305
	}

1306
	t = try_then_request_module(xt_find_table_lock(net, AF_INET6, name),
1307
				    "ip6table_%s", name);
1308 1309
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1310
		goto free_newinfo_counters_untrans;
1311
	}
L
Linus Torvalds 已提交
1312 1313

	/* You lied! */
1314
	if (valid_hooks != t->valid_hooks) {
L
Linus Torvalds 已提交
1315
		duprintf("Valid hook crap: %08X vs %08X\n",
1316
			 valid_hooks, t->valid_hooks);
L
Linus Torvalds 已提交
1317
		ret = -EINVAL;
1318
		goto put_module;
L
Linus Torvalds 已提交
1319 1320
	}

1321
	oldinfo = xt_replace_table(t, num_counters, newinfo, &ret);
L
Linus Torvalds 已提交
1322 1323 1324 1325 1326 1327
	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);
1328 1329
	if ((oldinfo->number > oldinfo->initial_entries) ||
	    (newinfo->number <= oldinfo->initial_entries))
L
Linus Torvalds 已提交
1330 1331 1332 1333 1334 1335 1336 1337
		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 */
1338
	loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()];
1339 1340
	IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,
			   NULL);
1341
	xt_free_table_info(oldinfo);
1342 1343
	if (copy_to_user(counters_ptr, counters,
			 sizeof(struct xt_counters) * num_counters) != 0)
L
Linus Torvalds 已提交
1344 1345
		ret = -EFAULT;
	vfree(counters);
1346
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1347 1348 1349 1350
	return ret;

 put_module:
	module_put(t->me);
1351
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1352 1353
 free_newinfo_counters_untrans:
	vfree(counters);
1354 1355 1356 1357 1358
 out:
	return ret;
}

static int
1359
do_replace(struct net *net, void __user *user, unsigned int len)
1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392
{
	int ret;
	struct ip6t_replace tmp;
	struct xt_table_info *newinfo;
	void *loc_cpu_entry;

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

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

	newinfo = xt_alloc_table_info(tmp.size);
	if (!newinfo)
		return -ENOMEM;

	/* 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),
			   tmp.size) != 0) {
		ret = -EFAULT;
		goto free_newinfo;
	}

	ret = translate_table(tmp.name, tmp.valid_hooks,
			      newinfo, loc_cpu_entry, tmp.size, tmp.num_entries,
			      tmp.hook_entry, tmp.underflow);
	if (ret != 0)
		goto free_newinfo;

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

1393
	ret = __do_replace(net, tmp.name, tmp.valid_hooks, newinfo,
1394 1395 1396 1397 1398 1399 1400
			   tmp.num_counters, tmp.counters);
	if (ret)
		goto free_newinfo_untrans;
	return 0;

 free_newinfo_untrans:
	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL);
L
Linus Torvalds 已提交
1401
 free_newinfo:
1402
	xt_free_table_info(newinfo);
L
Linus Torvalds 已提交
1403 1404 1405 1406
	return ret;
}

static int
1407 1408
do_add_counters(struct net *net, void __user *user, unsigned int len,
		int compat)
L
Linus Torvalds 已提交
1409 1410
{
	unsigned int i;
1411 1412 1413 1414 1415 1416
	struct xt_counters_info tmp;
	struct xt_counters *paddc;
	unsigned int num_counters;
	char *name;
	int size;
	void *ptmp;
1417
	struct xt_table *t;
1418
	const struct xt_table_info *private;
1419
	int ret = 0;
1420
	const void *loc_cpu_entry;
1421 1422
#ifdef CONFIG_COMPAT
	struct compat_xt_counters_info compat_tmp;
L
Linus Torvalds 已提交
1423

1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
	if (compat) {
		ptmp = &compat_tmp;
		size = sizeof(struct compat_xt_counters_info);
	} else
#endif
	{
		ptmp = &tmp;
		size = sizeof(struct xt_counters_info);
	}

	if (copy_from_user(ptmp, user, size) != 0)
L
Linus Torvalds 已提交
1435 1436
		return -EFAULT;

1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448
#ifdef CONFIG_COMPAT
	if (compat) {
		num_counters = compat_tmp.num_counters;
		name = compat_tmp.name;
	} else
#endif
	{
		num_counters = tmp.num_counters;
		name = tmp.name;
	}

	if (len != size + num_counters * sizeof(struct xt_counters))
L
Linus Torvalds 已提交
1449 1450
		return -EINVAL;

1451
	paddc = vmalloc_node(len - size, numa_node_id());
L
Linus Torvalds 已提交
1452 1453 1454
	if (!paddc)
		return -ENOMEM;

1455
	if (copy_from_user(paddc, user + size, len - size) != 0) {
L
Linus Torvalds 已提交
1456 1457 1458 1459
		ret = -EFAULT;
		goto free;
	}

1460
	t = xt_find_table_lock(net, AF_INET6, name);
1461 1462
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1463
		goto free;
1464
	}
L
Linus Torvalds 已提交
1465

1466
	mutex_lock(&t->lock);
1467
	private = t->private;
1468
	if (private->number != num_counters) {
L
Linus Torvalds 已提交
1469 1470 1471 1472
		ret = -EINVAL;
		goto unlock_up_free;
	}

1473
	preempt_disable();
L
Linus Torvalds 已提交
1474
	i = 0;
1475
	/* Choose the copy that is on our node */
1476
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
1477
	IP6T_ENTRY_ITERATE(loc_cpu_entry,
1478
			  private->size,
L
Linus Torvalds 已提交
1479
			  add_counter_to_entry,
1480
			  paddc,
L
Linus Torvalds 已提交
1481
			  &i);
1482
	preempt_enable();
L
Linus Torvalds 已提交
1483
 unlock_up_free:
1484
	mutex_unlock(&t->lock);
1485
	xt_table_unlock(t);
1486
	module_put(t->me);
L
Linus Torvalds 已提交
1487 1488 1489 1490 1491 1492
 free:
	vfree(paddc);

	return ret;
}

1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507
#ifdef CONFIG_COMPAT
struct compat_ip6t_replace {
	char			name[IP6T_TABLE_MAXNAMELEN];
	u32			valid_hooks;
	u32			num_entries;
	u32			size;
	u32			hook_entry[NF_INET_NUMHOOKS];
	u32			underflow[NF_INET_NUMHOOKS];
	u32			num_counters;
	compat_uptr_t		counters;	/* struct ip6t_counters * */
	struct compat_ip6t_entry entries[0];
};

static int
compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr,
1508
			  unsigned int *size, struct xt_counters *counters,
1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549
			  unsigned int *i)
{
	struct ip6t_entry_target *t;
	struct compat_ip6t_entry __user *ce;
	u_int16_t target_offset, next_offset;
	compat_uint_t origsize;
	int ret;

	ret = -EFAULT;
	origsize = *size;
	ce = (struct compat_ip6t_entry __user *)*dstptr;
	if (copy_to_user(ce, e, sizeof(struct ip6t_entry)))
		goto out;

	if (copy_to_user(&ce->counters, &counters[*i], sizeof(counters[*i])))
		goto out;

	*dstptr += sizeof(struct compat_ip6t_entry);
	*size -= sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);

	ret = IP6T_MATCH_ITERATE(e, xt_compat_match_to_user, dstptr, size);
	target_offset = e->target_offset - (origsize - *size);
	if (ret)
		goto out;
	t = ip6t_get_target(e);
	ret = xt_compat_target_to_user(t, dstptr, size);
	if (ret)
		goto out;
	ret = -EFAULT;
	next_offset = e->next_offset - (origsize - *size);
	if (put_user(target_offset, &ce->target_offset))
		goto out;
	if (put_user(next_offset, &ce->next_offset))
		goto out;

	(*i)++;
	return 0;
out:
	return ret;
}

1550
static int
1551 1552 1553 1554
compat_find_calc_match(struct ip6t_entry_match *m,
		       const char *name,
		       const struct ip6t_ip6 *ipv6,
		       unsigned int hookmask,
1555
		       int *size, unsigned int *i)
1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
{
	struct xt_match *match;

	match = try_then_request_module(xt_find_match(AF_INET6, m->u.user.name,
						      m->u.user.revision),
					"ip6t_%s", m->u.user.name);
	if (IS_ERR(match) || !match) {
		duprintf("compat_check_calc_match: `%s' not found\n",
			 m->u.user.name);
		return match ? PTR_ERR(match) : -ENOENT;
	}
	m->u.kernel.match = match;
	*size += xt_compat_match_offset(match);

	(*i)++;
	return 0;
}

1574
static int
1575 1576 1577 1578 1579 1580 1581 1582 1583
compat_release_match(struct ip6t_entry_match *m, unsigned int *i)
{
	if (i && (*i)-- == 0)
		return 1;

	module_put(m->u.kernel.match->me);
	return 0;
}

1584
static int
1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598
compat_release_entry(struct compat_ip6t_entry *e, unsigned int *i)
{
	struct ip6t_entry_target *t;

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

	/* Cleanup all matches */
	COMPAT_IP6T_MATCH_ITERATE(e, compat_release_match, NULL);
	t = compat_ip6t_get_target(e);
	module_put(t->u.kernel.target->me);
	return 0;
}

1599
static int
1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e,
				  struct xt_table_info *newinfo,
				  unsigned int *size,
				  unsigned char *base,
				  unsigned char *limit,
				  unsigned int *hook_entries,
				  unsigned int *underflows,
				  unsigned int *i,
				  const char *name)
{
	struct ip6t_entry_target *t;
	struct xt_target *target;
	unsigned int entry_offset;
1613 1614
	unsigned int j;
	int ret, off, h;
1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722

	duprintf("check_compat_entry_size_and_hooks %p\n", e);
	if ((unsigned long)e % __alignof__(struct compat_ip6t_entry) != 0
	    || (unsigned char *)e + sizeof(struct compat_ip6t_entry) >= limit) {
		duprintf("Bad offset %p, limit = %p\n", e, limit);
		return -EINVAL;
	}

	if (e->next_offset < sizeof(struct compat_ip6t_entry) +
			     sizeof(struct compat_xt_entry_target)) {
		duprintf("checking: element %p size %u\n",
			 e, e->next_offset);
		return -EINVAL;
	}

	/* For purposes of check_entry casting the compat entry is fine */
	ret = check_entry((struct ip6t_entry *)e, name);
	if (ret)
		return ret;

	off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
	entry_offset = (void *)e - (void *)base;
	j = 0;
	ret = COMPAT_IP6T_MATCH_ITERATE(e, compat_find_calc_match, name,
					&e->ipv6, e->comefrom, &off, &j);
	if (ret != 0)
		goto release_matches;

	t = compat_ip6t_get_target(e);
	target = try_then_request_module(xt_find_target(AF_INET6,
							t->u.user.name,
							t->u.user.revision),
					 "ip6t_%s", t->u.user.name);
	if (IS_ERR(target) || !target) {
		duprintf("check_compat_entry_size_and_hooks: `%s' not found\n",
			 t->u.user.name);
		ret = target ? PTR_ERR(target) : -ENOENT;
		goto release_matches;
	}
	t->u.kernel.target = target;

	off += xt_compat_target_offset(target);
	*size += off;
	ret = xt_compat_add_offset(AF_INET6, entry_offset, off);
	if (ret)
		goto out;

	/* Check hooks & underflows */
	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
		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];
	}

	/* Clear counters and comefrom */
	memset(&e->counters, 0, sizeof(e->counters));
	e->comefrom = 0;

	(*i)++;
	return 0;

out:
	module_put(t->u.kernel.target->me);
release_matches:
	IP6T_MATCH_ITERATE(e, compat_release_match, &j);
	return ret;
}

static int
compat_copy_entry_from_user(struct compat_ip6t_entry *e, void **dstptr,
			    unsigned int *size, const char *name,
			    struct xt_table_info *newinfo, unsigned char *base)
{
	struct ip6t_entry_target *t;
	struct xt_target *target;
	struct ip6t_entry *de;
	unsigned int origsize;
	int ret, h;

	ret = 0;
	origsize = *size;
	de = (struct ip6t_entry *)*dstptr;
	memcpy(de, e, sizeof(struct ip6t_entry));
	memcpy(&de->counters, &e->counters, sizeof(e->counters));

	*dstptr += sizeof(struct ip6t_entry);
	*size += sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);

	ret = COMPAT_IP6T_MATCH_ITERATE(e, xt_compat_match_from_user,
					dstptr, size);
	if (ret)
		return ret;
	de->target_offset = e->target_offset - (origsize - *size);
	t = compat_ip6t_get_target(e);
	target = t->u.kernel.target;
	xt_compat_target_from_user(t, dstptr, size);

	de->next_offset = e->next_offset - (origsize - *size);
	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
		if ((unsigned char *)de - base < newinfo->hook_entry[h])
			newinfo->hook_entry[h] -= origsize - *size;
		if ((unsigned char *)de - base < newinfo->underflow[h])
			newinfo->underflow[h] -= origsize - *size;
	}
	return ret;
}

1723
static int compat_check_entry(struct ip6t_entry *e, const char *name,
1724 1725
				     unsigned int *i)
{
1726 1727
	unsigned int j;
	int ret;
1728
	struct xt_mtchk_param mtpar;
1729 1730

	j = 0;
1731 1732 1733
	mtpar.table     = name;
	mtpar.entryinfo = &e->ipv6;
	mtpar.hook_mask = e->comefrom;
1734
	mtpar.family    = NFPROTO_IPV6;
1735
	ret = IP6T_MATCH_ITERATE(e, check_match, &mtpar, &j);
1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872
	if (ret)
		goto cleanup_matches;

	ret = check_target(e, name);
	if (ret)
		goto cleanup_matches;

	(*i)++;
	return 0;

 cleanup_matches:
	IP6T_MATCH_ITERATE(e, cleanup_match, &j);
	return ret;
}

static int
translate_compat_table(const char *name,
		       unsigned int valid_hooks,
		       struct xt_table_info **pinfo,
		       void **pentry0,
		       unsigned int total_size,
		       unsigned int number,
		       unsigned int *hook_entries,
		       unsigned int *underflows)
{
	unsigned int i, j;
	struct xt_table_info *newinfo, *info;
	void *pos, *entry0, *entry1;
	unsigned int size;
	int ret;

	info = *pinfo;
	entry0 = *pentry0;
	size = total_size;
	info->number = number;

	/* Init all hooks to impossible value. */
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
		info->hook_entry[i] = 0xFFFFFFFF;
		info->underflow[i] = 0xFFFFFFFF;
	}

	duprintf("translate_compat_table: size %u\n", info->size);
	j = 0;
	xt_compat_lock(AF_INET6);
	/* Walk through entries, checking offsets. */
	ret = COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size,
					check_compat_entry_size_and_hooks,
					info, &size, entry0,
					entry0 + total_size,
					hook_entries, underflows, &j, name);
	if (ret != 0)
		goto out_unlock;

	ret = -EINVAL;
	if (j != number) {
		duprintf("translate_compat_table: %u not %u entries\n",
			 j, number);
		goto out_unlock;
	}

	/* Check hooks all assigned */
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
		/* Only hooks which are valid */
		if (!(valid_hooks & (1 << i)))
			continue;
		if (info->hook_entry[i] == 0xFFFFFFFF) {
			duprintf("Invalid hook entry %u %u\n",
				 i, hook_entries[i]);
			goto out_unlock;
		}
		if (info->underflow[i] == 0xFFFFFFFF) {
			duprintf("Invalid underflow %u %u\n",
				 i, underflows[i]);
			goto out_unlock;
		}
	}

	ret = -ENOMEM;
	newinfo = xt_alloc_table_info(size);
	if (!newinfo)
		goto out_unlock;

	newinfo->number = number;
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
		newinfo->hook_entry[i] = info->hook_entry[i];
		newinfo->underflow[i] = info->underflow[i];
	}
	entry1 = newinfo->entries[raw_smp_processor_id()];
	pos = entry1;
	size = total_size;
	ret = COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size,
					compat_copy_entry_from_user,
					&pos, &size, name, newinfo, entry1);
	xt_compat_flush_offsets(AF_INET6);
	xt_compat_unlock(AF_INET6);
	if (ret)
		goto free_newinfo;

	ret = -ELOOP;
	if (!mark_source_chains(newinfo, valid_hooks, entry1))
		goto free_newinfo;

	i = 0;
	ret = IP6T_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry,
				 name, &i);
	if (ret) {
		j -= i;
		COMPAT_IP6T_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i,
						   compat_release_entry, &j);
		IP6T_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, &i);
		xt_free_table_info(newinfo);
		return ret;
	}

	/* And one copy for every other CPU */
	for_each_possible_cpu(i)
		if (newinfo->entries[i] && newinfo->entries[i] != entry1)
			memcpy(newinfo->entries[i], entry1, newinfo->size);

	*pinfo = newinfo;
	*pentry0 = entry1;
	xt_free_table_info(info);
	return 0;

free_newinfo:
	xt_free_table_info(newinfo);
out:
	COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j);
	return ret;
out_unlock:
	xt_compat_flush_offsets(AF_INET6);
	xt_compat_unlock(AF_INET6);
	goto out;
}

static int
1873
compat_do_replace(struct net *net, void __user *user, unsigned int len)
1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
{
	int ret;
	struct compat_ip6t_replace tmp;
	struct xt_table_info *newinfo;
	void *loc_cpu_entry;

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

	/* overflow check */
	if (tmp.size >= INT_MAX / num_possible_cpus())
		return -ENOMEM;
	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
		return -ENOMEM;

	newinfo = xt_alloc_table_info(tmp.size);
	if (!newinfo)
		return -ENOMEM;

1893
	/* choose the copy that is on our node/cpu */
1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909
	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp),
			   tmp.size) != 0) {
		ret = -EFAULT;
		goto free_newinfo;
	}

	ret = translate_compat_table(tmp.name, tmp.valid_hooks,
				     &newinfo, &loc_cpu_entry, tmp.size,
				     tmp.num_entries, tmp.hook_entry,
				     tmp.underflow);
	if (ret != 0)
		goto free_newinfo;

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

1910
	ret = __do_replace(net, tmp.name, tmp.valid_hooks, newinfo,
1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933
			   tmp.num_counters, compat_ptr(tmp.counters));
	if (ret)
		goto free_newinfo_untrans;
	return 0;

 free_newinfo_untrans:
	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL);
 free_newinfo:
	xt_free_table_info(newinfo);
	return ret;
}

static int
compat_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:
1934
		ret = compat_do_replace(sock_net(sk), user, len);
1935 1936 1937
		break;

	case IP6T_SO_SET_ADD_COUNTERS:
1938
		ret = do_add_counters(sock_net(sk), user, len, 1);
1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959
		break;

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

	return ret;
}

struct compat_ip6t_get_entries {
	char name[IP6T_TABLE_MAXNAMELEN];
	compat_uint_t size;
	struct compat_ip6t_entry entrytable[0];
};

static int
compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table,
			    void __user *userptr)
{
	struct xt_counters *counters;
1960
	const struct xt_table_info *private = table->private;
1961 1962 1963
	void __user *pos;
	unsigned int size;
	int ret = 0;
1964
	const void *loc_cpu_entry;
1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986
	unsigned int i = 0;

	counters = alloc_counters(table);
	if (IS_ERR(counters))
		return PTR_ERR(counters);

	/* choose the copy that is on our node/cpu, ...
	 * This choice is lazy (because current thread is
	 * allowed to migrate to another cpu)
	 */
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
	pos = userptr;
	size = total_size;
	ret = IP6T_ENTRY_ITERATE(loc_cpu_entry, total_size,
				 compat_copy_entry_to_user,
				 &pos, &size, counters, &i);

	vfree(counters);
	return ret;
}

static int
1987 1988
compat_get_entries(struct net *net, struct compat_ip6t_get_entries __user *uptr,
		   int *len)
1989 1990 1991 1992 1993 1994
{
	int ret;
	struct compat_ip6t_get_entries get;
	struct xt_table *t;

	if (*len < sizeof(get)) {
1995
		duprintf("compat_get_entries: %u < %zu\n", *len, sizeof(get));
1996 1997 1998 1999 2000 2001 2002
		return -EINVAL;
	}

	if (copy_from_user(&get, uptr, sizeof(get)) != 0)
		return -EFAULT;

	if (*len != sizeof(struct compat_ip6t_get_entries) + get.size) {
2003 2004
		duprintf("compat_get_entries: %u != %zu\n",
			 *len, sizeof(get) + get.size);
2005 2006 2007 2008
		return -EINVAL;
	}

	xt_compat_lock(AF_INET6);
2009
	t = xt_find_table_lock(net, AF_INET6, get.name);
2010
	if (t && !IS_ERR(t)) {
2011
		const struct xt_table_info *private = t->private;
2012
		struct xt_table_info info;
2013
		duprintf("t->private->number = %u\n", private->number);
2014 2015 2016 2017 2018 2019
		ret = compat_table_info(private, &info);
		if (!ret && get.size == info.size) {
			ret = compat_copy_entries_to_user(private->size,
							  t, uptr->entrytable);
		} else if (!ret) {
			duprintf("compat_get_entries: I've got %u not %u!\n",
2020
				 private->size, get.size);
2021
			ret = -EAGAIN;
2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044
		}
		xt_compat_flush_offsets(AF_INET6);
		module_put(t->me);
		xt_table_unlock(t);
	} else
		ret = t ? PTR_ERR(t) : -ENOENT;

	xt_compat_unlock(AF_INET6);
	return ret;
}

static int do_ip6t_get_ctl(struct sock *, int, void __user *, int *);

static int
compat_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:
2045
		ret = get_info(sock_net(sk), user, len, 1);
2046 2047
		break;
	case IP6T_SO_GET_ENTRIES:
2048
		ret = compat_get_entries(sock_net(sk), user, len);
2049 2050 2051 2052 2053 2054 2055 2056
		break;
	default:
		ret = do_ip6t_get_ctl(sk, cmd, user, len);
	}
	return ret;
}
#endif

L
Linus Torvalds 已提交
2057 2058 2059 2060 2061 2062 2063 2064 2065 2066
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:
2067
		ret = do_replace(sock_net(sk), user, len);
L
Linus Torvalds 已提交
2068 2069 2070
		break;

	case IP6T_SO_SET_ADD_COUNTERS:
2071
		ret = do_add_counters(sock_net(sk), user, len, 0);
L
Linus Torvalds 已提交
2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090
		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) {
2091
	case IP6T_SO_GET_INFO:
2092
		ret = get_info(sock_net(sk), user, len, 0);
2093
		break;
L
Linus Torvalds 已提交
2094

2095
	case IP6T_SO_GET_ENTRIES:
2096
		ret = get_entries(sock_net(sk), user, len);
L
Linus Torvalds 已提交
2097 2098
		break;

2099 2100 2101
	case IP6T_SO_GET_REVISION_MATCH:
	case IP6T_SO_GET_REVISION_TARGET: {
		struct ip6t_get_revision rev;
2102
		int target;
2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113

		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)
2114
			target = 1;
2115
		else
2116
			target = 0;
2117

2118 2119 2120
		try_then_request_module(xt_find_revision(AF_INET6, rev.name,
							 rev.revision,
							 target, &ret),
2121 2122 2123 2124
					"ip6t_%s", rev.name);
		break;
	}

L
Linus Torvalds 已提交
2125 2126 2127 2128 2129 2130 2131 2132
	default:
		duprintf("do_ip6t_get_ctl: unknown request %i\n", cmd);
		ret = -EINVAL;
	}

	return ret;
}

2133 2134
struct xt_table *ip6t_register_table(struct net *net, struct xt_table *table,
				     const struct ip6t_replace *repl)
L
Linus Torvalds 已提交
2135 2136
{
	int ret;
2137
	struct xt_table_info *newinfo;
2138
	struct xt_table_info bootstrap
L
Linus Torvalds 已提交
2139
		= { 0, 0, 0, { 0 }, { 0 }, { } };
2140
	void *loc_cpu_entry;
2141
	struct xt_table *new_table;
L
Linus Torvalds 已提交
2142

2143
	newinfo = xt_alloc_table_info(repl->size);
2144 2145 2146 2147
	if (!newinfo) {
		ret = -ENOMEM;
		goto out;
	}
L
Linus Torvalds 已提交
2148

2149
	/* choose the copy on our node/cpu, but dont care about preemption */
2150 2151
	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
	memcpy(loc_cpu_entry, repl->entries, repl->size);
L
Linus Torvalds 已提交
2152 2153

	ret = translate_table(table->name, table->valid_hooks,
2154
			      newinfo, loc_cpu_entry, repl->size,
L
Linus Torvalds 已提交
2155 2156 2157
			      repl->num_entries,
			      repl->hook_entry,
			      repl->underflow);
2158 2159
	if (ret != 0)
		goto out_free;
L
Linus Torvalds 已提交
2160

2161
	new_table = xt_register_table(net, table, &bootstrap, newinfo);
2162
	if (IS_ERR(new_table)) {
2163 2164
		ret = PTR_ERR(new_table);
		goto out_free;
L
Linus Torvalds 已提交
2165
	}
2166
	return new_table;
L
Linus Torvalds 已提交
2167

2168 2169 2170 2171
out_free:
	xt_free_table_info(newinfo);
out:
	return ERR_PTR(ret);
L
Linus Torvalds 已提交
2172 2173
}

2174
void ip6t_unregister_table(struct xt_table *table)
L
Linus Torvalds 已提交
2175
{
2176
	struct xt_table_info *private;
2177
	void *loc_cpu_entry;
2178
	struct module *table_owner = table->me;
2179

2180
	private = xt_unregister_table(table);
L
Linus Torvalds 已提交
2181 2182

	/* Decrease module usage counts and free resources */
2183 2184
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
	IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL);
2185 2186
	if (private->number > private->initial_entries)
		module_put(table_owner);
2187
	xt_free_table_info(private);
L
Linus Torvalds 已提交
2188 2189 2190
}

/* Returns 1 if the type and code is matched by the range, 0 otherwise */
2191
static inline bool
L
Linus Torvalds 已提交
2192 2193
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,
2194
		     bool invert)
L
Linus Torvalds 已提交
2195 2196 2197 2198 2199
{
	return (type == test_type && code >= min_code && code <= max_code)
		^ invert;
}

2200
static bool
2201
icmp6_match(const struct sk_buff *skb, const struct xt_match_param *par)
L
Linus Torvalds 已提交
2202
{
2203 2204
	const struct icmp6hdr *ic;
	struct icmp6hdr _icmph;
2205
	const struct ip6t_icmp *icmpinfo = par->matchinfo;
L
Linus Torvalds 已提交
2206 2207

	/* Must not be a fragment. */
2208
	if (par->fragoff != 0)
2209
		return false;
L
Linus Torvalds 已提交
2210

2211
	ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph);
L
Linus Torvalds 已提交
2212 2213
	if (ic == NULL) {
		/* We've been asked to examine this packet, and we
2214 2215
		 * can't.  Hence, no choice but to drop.
		 */
L
Linus Torvalds 已提交
2216
		duprintf("Dropping evil ICMP tinygram.\n");
2217
		*par->hotdrop = true;
2218
		return false;
L
Linus Torvalds 已提交
2219 2220 2221 2222 2223 2224 2225 2226 2227 2228
	}

	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. */
2229
static bool icmp6_checkentry(const struct xt_mtchk_param *par)
L
Linus Torvalds 已提交
2230
{
2231
	const struct ip6t_icmp *icmpinfo = par->matchinfo;
L
Linus Torvalds 已提交
2232

2233 2234
	/* Must specify no unknown invflags */
	return !(icmpinfo->invflags & ~IP6T_ICMP_INV);
L
Linus Torvalds 已提交
2235 2236 2237
}

/* The built-in targets: standard (NULL) and error. */
2238
static struct xt_target ip6t_standard_target __read_mostly = {
L
Linus Torvalds 已提交
2239
	.name		= IP6T_STANDARD_TARGET,
2240
	.targetsize	= sizeof(int),
2241
	.family		= AF_INET6,
2242 2243 2244 2245 2246
#ifdef CONFIG_COMPAT
	.compatsize	= sizeof(compat_int_t),
	.compat_from_user = compat_standard_from_user,
	.compat_to_user	= compat_standard_to_user,
#endif
L
Linus Torvalds 已提交
2247 2248
};

2249
static struct xt_target ip6t_error_target __read_mostly = {
L
Linus Torvalds 已提交
2250 2251
	.name		= IP6T_ERROR_TARGET,
	.target		= ip6t_error,
2252
	.targetsize	= IP6T_FUNCTION_MAXNAMELEN,
2253
	.family		= AF_INET6,
L
Linus Torvalds 已提交
2254 2255 2256 2257 2258 2259 2260
};

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,
2261 2262 2263
#ifdef CONFIG_COMPAT
	.compat_set	= compat_do_ip6t_set_ctl,
#endif
L
Linus Torvalds 已提交
2264 2265 2266
	.get_optmin	= IP6T_BASE_CTL,
	.get_optmax	= IP6T_SO_GET_MAX+1,
	.get		= do_ip6t_get_ctl,
2267 2268 2269
#ifdef CONFIG_COMPAT
	.compat_get	= compat_do_ip6t_get_ctl,
#endif
2270
	.owner		= THIS_MODULE,
L
Linus Torvalds 已提交
2271 2272
};

2273
static struct xt_match icmp6_matchstruct __read_mostly = {
L
Linus Torvalds 已提交
2274
	.name		= "icmp6",
2275
	.match		= icmp6_match,
2276 2277 2278
	.matchsize	= sizeof(struct ip6t_icmp),
	.checkentry	= icmp6_checkentry,
	.proto		= IPPROTO_ICMPV6,
2279
	.family		= AF_INET6,
L
Linus Torvalds 已提交
2280 2281
};

2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296
static int __net_init ip6_tables_net_init(struct net *net)
{
	return xt_proto_init(net, AF_INET6);
}

static void __net_exit ip6_tables_net_exit(struct net *net)
{
	xt_proto_fini(net, AF_INET6);
}

static struct pernet_operations ip6_tables_net_ops = {
	.init = ip6_tables_net_init,
	.exit = ip6_tables_net_exit,
};

2297
static int __init ip6_tables_init(void)
L
Linus Torvalds 已提交
2298 2299 2300
{
	int ret;

2301
	ret = register_pernet_subsys(&ip6_tables_net_ops);
2302 2303
	if (ret < 0)
		goto err1;
2304

L
Linus Torvalds 已提交
2305
	/* Noone else will be downing sem now, so we won't sleep */
2306 2307 2308 2309 2310 2311 2312 2313 2314
	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
2315 2316 2317

	/* Register setsockopt */
	ret = nf_register_sockopt(&ip6t_sockopts);
2318 2319
	if (ret < 0)
		goto err5;
L
Linus Torvalds 已提交
2320

2321
	printk(KERN_INFO "ip6_tables: (C) 2000-2006 Netfilter Core Team\n");
L
Linus Torvalds 已提交
2322
	return 0;
2323 2324 2325 2326 2327 2328 2329 2330

err5:
	xt_unregister_match(&icmp6_matchstruct);
err4:
	xt_unregister_target(&ip6t_error_target);
err3:
	xt_unregister_target(&ip6t_standard_target);
err2:
2331
	unregister_pernet_subsys(&ip6_tables_net_ops);
2332 2333
err1:
	return ret;
L
Linus Torvalds 已提交
2334 2335
}

2336
static void __exit ip6_tables_fini(void)
L
Linus Torvalds 已提交
2337 2338
{
	nf_unregister_sockopt(&ip6t_sockopts);
2339

2340 2341 2342
	xt_unregister_match(&icmp6_matchstruct);
	xt_unregister_target(&ip6t_error_target);
	xt_unregister_target(&ip6t_standard_target);
2343 2344

	unregister_pernet_subsys(&ip6_tables_net_ops);
L
Linus Torvalds 已提交
2345 2346
}

2347
/*
2348 2349 2350 2351 2352 2353 2354
 * 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.
 *
2355 2356 2357
 * If the first fragment doesn't contain the final protocol header or
 * NEXTHDR_NONE it is considered invalid.
 *
2358 2359 2360 2361
 * 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.
2362 2363
 *
 */
2364 2365
int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
		  int target, unsigned short *fragoff)
2366
{
2367
	unsigned int start = skb_network_offset(skb) + sizeof(struct ipv6hdr);
2368
	u8 nexthdr = ipv6_hdr(skb)->nexthdr;
2369 2370
	unsigned int len = skb->len - start;

2371 2372 2373
	if (fragoff)
		*fragoff = 0;

2374 2375 2376 2377
	while (nexthdr != target) {
		struct ipv6_opt_hdr _hdr, *hp;
		unsigned int hdrlen;

2378 2379 2380
		if ((!ipv6_ext_hdr(nexthdr)) || nexthdr == NEXTHDR_NONE) {
			if (target < 0)
				break;
2381
			return -ENOENT;
2382 2383
		}

2384 2385
		hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr);
		if (hp == NULL)
2386
			return -EBADMSG;
2387
		if (nexthdr == NEXTHDR_FRAGMENT) {
A
Al Viro 已提交
2388 2389
			unsigned short _frag_off;
			__be16 *fp;
2390 2391 2392 2393 2394 2395
			fp = skb_header_pointer(skb,
						start+offsetof(struct frag_hdr,
							       frag_off),
						sizeof(_frag_off),
						&_frag_off);
			if (fp == NULL)
2396
				return -EBADMSG;
2397

2398 2399 2400 2401
			_frag_off = ntohs(*fp) & ~0x7;
			if (_frag_off) {
				if (target < 0 &&
				    ((!ipv6_ext_hdr(hp->nexthdr)) ||
2402
				     hp->nexthdr == NEXTHDR_NONE)) {
2403 2404 2405 2406
					if (fragoff)
						*fragoff = _frag_off;
					return hp->nexthdr;
				}
2407
				return -ENOENT;
2408
			}
2409 2410
			hdrlen = 8;
		} else if (nexthdr == NEXTHDR_AUTH)
2411
			hdrlen = (hp->hdrlen + 2) << 2;
2412
		else
2413
			hdrlen = ipv6_optlen(hp);
2414 2415 2416 2417 2418 2419 2420

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

	*offset = start;
2421
	return nexthdr;
2422 2423
}

L
Linus Torvalds 已提交
2424 2425 2426 2427
EXPORT_SYMBOL(ip6t_register_table);
EXPORT_SYMBOL(ip6t_unregister_table);
EXPORT_SYMBOL(ip6t_do_table);
EXPORT_SYMBOL(ip6t_ext_hdr);
2428
EXPORT_SYMBOL(ipv6_find_hdr);
L
Linus Torvalds 已提交
2429

2430 2431
module_init(ip6_tables_init);
module_exit(ip6_tables_fini);