ip6_tables.c 58.1 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
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
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
static inline bool unconditional(const struct ip6t_ip6 *ipv6)
226
{
227
	static const struct ip6t_ip6 uncond;
228

229
	return memcmp(ipv6, &uncond, sizeof(uncond)) == 0;
230 231 232 233 234
}

#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
    defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
/* This cries for unification! */
235
static const char *const hooknames[] = {
236 237 238 239 240
	[NF_INET_PRE_ROUTING]		= "PREROUTING",
	[NF_INET_LOCAL_IN]		= "INPUT",
	[NF_INET_FORWARD]		= "FORWARD",
	[NF_INET_LOCAL_OUT]		= "OUTPUT",
	[NF_INET_POST_ROUTING]		= "POSTROUTING",
241 242 243 244 245 246 247 248
};

enum nf_ip_trace_comments {
	NF_IP6_TRACE_COMMENT_RULE,
	NF_IP6_TRACE_COMMENT_RETURN,
	NF_IP6_TRACE_COMMENT_POLICY,
};

249
static const char *const comments[] = {
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264
	[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,
		},
	},
};

265
/* Mildly perf critical (only if packet tracing is on) */
266 267
static inline int
get_chainname_rulenum(struct ip6t_entry *s, struct ip6t_entry *e,
268 269
		      const char *hookname, const char **chainname,
		      const char **comment, unsigned int *rulenum)
270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286
{
	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
287 288
				? comments[NF_IP6_TRACE_COMMENT_POLICY]
				: comments[NF_IP6_TRACE_COMMENT_RETURN];
289 290 291 292 293 294 295 296 297 298 299 300
		}
		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,
301
			 const char *tablename,
302 303 304 305
			 struct xt_table_info *private,
			 struct ip6t_entry *e)
{
	void *table_base;
306
	const struct ip6t_entry *root;
307
	const char *hookname, *chainname, *comment;
308 309
	unsigned int rulenum = 0;

310
	table_base = private->entries[smp_processor_id()];
311 312
	root = get_entry(table_base, private->hook_entry[hook]);

313 314
	hookname = chainname = hooknames[hook];
	comment = comments[NF_IP6_TRACE_COMMENT_RULE];
315 316 317 318 319 320 321 322 323 324 325 326

	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

327 328 329 330 331 332
static inline __pure struct ip6t_entry *
ip6t_next_entry(const struct ip6t_entry *entry)
{
	return (void *)entry + entry->next_offset;
}

L
Linus Torvalds 已提交
333 334
/* Returns one of the generic firewall policies, like NF_ACCEPT. */
unsigned int
335
ip6t_do_table(struct sk_buff *skb,
L
Linus Torvalds 已提交
336 337 338
	      unsigned int hook,
	      const struct net_device *in,
	      const struct net_device *out,
339
	      struct xt_table *table)
L
Linus Torvalds 已提交
340
{
341 342
#define tb_comefrom ((struct ip6t_entry *)table_base)->comefrom

343
	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
344
	bool hotdrop = false;
L
Linus Torvalds 已提交
345 346 347 348 349
	/* 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;
350
	struct xt_table_info *private;
351
	struct xt_match_param mtpar;
352
	struct xt_target_param tgpar;
L
Linus Torvalds 已提交
353 354 355 356 357 358 359 360 361 362

	/* 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. */
363
	mtpar.hotdrop = &hotdrop;
364 365
	mtpar.in      = tgpar.in  = in;
	mtpar.out     = tgpar.out = out;
366
	mtpar.family  = tgpar.family = NFPROTO_IPV6;
367
	mtpar.hooknum = tgpar.hooknum = hook;
L
Linus Torvalds 已提交
368 369

	IP_NF_ASSERT(table->valid_hooks & (1 << hook));
370

371 372 373
	xt_info_rdlock_bh();
	private = table->private;
	table_base = private->entries[smp_processor_id()];
374

375
	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

	do {
381 382
		struct ip6t_entry_target *t;

L
Linus Torvalds 已提交
383 384
		IP_NF_ASSERT(e);
		IP_NF_ASSERT(back);
385
		if (!ip6_packet_match(skb, indev, outdev, &e->ipv6,
386 387
		    &mtpar.thoff, &mtpar.fragoff, &hotdrop) ||
		    IP6T_MATCH_ITERATE(e, do_match, skb, &mtpar) != 0) {
388 389 390
			e = ip6t_next_entry(e);
			continue;
		}
L
Linus Torvalds 已提交
391

392 393 394
		ADD_COUNTER(e->counters,
			    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

#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
    defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
401 402 403 404
		/* The packet is traced: log it */
		if (unlikely(skb->nf_trace))
			trace_packet(skb, hook, in, out,
				     table->name, private, e);
405
#endif
406 407 408 409 410 411 412 413 414 415
		/* 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;
L
Linus Torvalds 已提交
416
				}
417 418 419 420 421 422 423 424 425 426 427 428
				e = back;
				back = get_entry(table_base, back->comefrom);
				continue;
			}
			if (table_base + v != ip6t_next_entry(e)
			    && !(e->ipv6.flags & IP6T_F_GOTO)) {
				/* Save old back ptr in next entry */
				struct ip6t_entry *next = ip6t_next_entry(e);
				next->comefrom = (void *)back - table_base;
				/* set back pointer to next entry */
				back = next;
			}
L
Linus Torvalds 已提交
429

430
			e = get_entry(table_base, v);
431 432 433 434 435 436 437
			continue;
		}

		/* Targets which reenter must return
		   abs. verdicts */
		tgpar.target   = t->u.kernel.target;
		tgpar.targinfo = t->data;
438

L
Linus Torvalds 已提交
439
#ifdef CONFIG_NETFILTER_DEBUG
440
		tb_comefrom = 0xeeeeeeec;
L
Linus Torvalds 已提交
441
#endif
442
		verdict = t->u.kernel.target->target(skb, &tgpar);
L
Linus Torvalds 已提交
443 444

#ifdef CONFIG_NETFILTER_DEBUG
445
		if (tb_comefrom != 0xeeeeeeec && verdict == IP6T_CONTINUE) {
446 447 448
			printk("Target %s reentered!\n",
			       t->u.kernel.target->name);
			verdict = NF_DROP;
L
Linus Torvalds 已提交
449
		}
450
		tb_comefrom = 0x57acc001;
451 452 453 454 455 456
#endif
		if (verdict == IP6T_CONTINUE)
			e = ip6t_next_entry(e);
		else
			/* Verdict */
			break;
L
Linus Torvalds 已提交
457 458 459
	} while (!hotdrop);

#ifdef CONFIG_NETFILTER_DEBUG
460
	tb_comefrom = NETFILTER_LINK_POISON;
L
Linus Torvalds 已提交
461
#endif
462
	xt_info_rdunlock_bh();
L
Linus Torvalds 已提交
463 464 465 466 467 468 469 470

#ifdef DEBUG_ALLOW_ALL
	return NF_ACCEPT;
#else
	if (hotdrop)
		return NF_DROP;
	else return verdict;
#endif
471 472

#undef tb_comefrom
L
Linus Torvalds 已提交
473 474 475 476 477
}

/* Figures out from what hook each rule can be called: returns 0 if
   there are loops.  Puts hook bitmask in comefrom. */
static int
478
mark_source_chains(struct xt_table_info *newinfo,
479
		   unsigned int valid_hooks, void *entry0)
L
Linus Torvalds 已提交
480 481 482 483 484
{
	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 */
485
	for (hook = 0; hook < NF_INET_NUMHOOKS; hook++) {
L
Linus Torvalds 已提交
486
		unsigned int pos = newinfo->hook_entry[hook];
487
		struct ip6t_entry *e = (struct ip6t_entry *)(entry0 + pos);
L
Linus Torvalds 已提交
488 489 490 491 492 493 494 495 496 497

		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);
498
			int visited = e->comefrom & (1 << hook);
L
Linus Torvalds 已提交
499

500
			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
501 502 503 504
				printk("iptables: loop hook %u pos %u %08X.\n",
				       hook, pos, e->comefrom);
				return 0;
			}
505
			e->comefrom |= ((1 << hook) | (1 << NF_INET_NUMHOOKS));
L
Linus Torvalds 已提交
506 507

			/* Unconditional return/END. */
508
			if ((e->target_offset == sizeof(struct ip6t_entry)
L
Linus Torvalds 已提交
509 510 511
			    && (strcmp(t->target.u.user.name,
				       IP6T_STANDARD_TARGET) == 0)
			    && t->verdict < 0
512
			    && unconditional(&e->ipv6)) || visited) {
L
Linus Torvalds 已提交
513 514
				unsigned int oldpos, size;

515 516 517
				if ((strcmp(t->target.u.user.name,
					    IP6T_STANDARD_TARGET) == 0) &&
				    t->verdict < -NF_MAX_VERDICT - 1) {
518 519 520 521 522 523
					duprintf("mark_source_chains: bad "
						"negative verdict (%i)\n",
								t->verdict);
					return 0;
				}

L
Linus Torvalds 已提交
524 525 526
				/* Return: backtrack through the last
				   big jump. */
				do {
527
					e->comefrom ^= (1<<NF_INET_NUMHOOKS);
L
Linus Torvalds 已提交
528 529
#ifdef DEBUG_IP_FIREWALL_USER
					if (e->comefrom
530
					    & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
						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 *)
546
						(entry0 + pos);
L
Linus Torvalds 已提交
547 548 549 550 551
				} while (oldpos == pos + e->next_offset);

				/* Move along one */
				size = e->next_offset;
				e = (struct ip6t_entry *)
552
					(entry0 + pos + size);
L
Linus Torvalds 已提交
553 554 555 556 557 558 559 560
				e->counters.pcnt = pos;
				pos += size;
			} else {
				int newpos = t->verdict;

				if (strcmp(t->target.u.user.name,
					   IP6T_STANDARD_TARGET) == 0
				    && newpos >= 0) {
561 562 563 564 565 566 567
					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
568 569 570 571 572 573 574 575
					/* 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 *)
576
					(entry0 + newpos);
L
Linus Torvalds 已提交
577 578 579 580 581 582 583 584 585 586
				e->counters.pcnt = pos;
				pos = newpos;
			}
		}
		next:
		duprintf("Finished chain %u\n", hook);
	}
	return 1;
}

587
static int
L
Linus Torvalds 已提交
588 589
cleanup_match(struct ip6t_entry_match *m, unsigned int *i)
{
590 591
	struct xt_mtdtor_param par;

L
Linus Torvalds 已提交
592 593 594
	if (i && (*i)-- == 0)
		return 1;

595 596
	par.match     = m->u.kernel.match;
	par.matchinfo = m->data;
597
	par.family    = NFPROTO_IPV6;
598 599 600
	if (par.match->destroy != NULL)
		par.match->destroy(&par);
	module_put(par.match->me);
L
Linus Torvalds 已提交
601 602 603
	return 0;
}

604
static int
605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624
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;
}

625 626
static int check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par,
		       unsigned int *i)
627
{
628
	const struct ip6t_ip6 *ipv6 = par->entryinfo;
629 630
	int ret;

631 632 633
	par->match     = m->u.kernel.match;
	par->matchinfo = m->data;

634
	ret = xt_check_match(par, m->u.match_size - sizeof(*m),
635
			     ipv6->proto, ipv6->invflags & IP6T_INV_PROTO);
636
	if (ret < 0) {
637
		duprintf("ip_tables: check failed for `%s'.\n",
638
			 par.match->name);
639
		return ret;
640
	}
641 642
	++*i;
	return 0;
643 644
}

645
static int
646
find_check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par,
647
		 unsigned int *i)
L
Linus Torvalds 已提交
648
{
649
	struct xt_match *match;
650
	int ret;
L
Linus Torvalds 已提交
651

652
	match = try_then_request_module(xt_find_match(AF_INET6, m->u.user.name,
653
						      m->u.user.revision),
654 655
					"ip6t_%s", m->u.user.name);
	if (IS_ERR(match) || !match) {
656
		duprintf("find_check_match: `%s' not found\n", m->u.user.name);
657
		return match ? PTR_ERR(match) : -ENOENT;
L
Linus Torvalds 已提交
658 659 660
	}
	m->u.kernel.match = match;

661
	ret = check_match(m, par, i);
662 663 664
	if (ret)
		goto err;

L
Linus Torvalds 已提交
665
	return 0;
666 667 668
err:
	module_put(m->u.kernel.match->me);
	return ret;
L
Linus Torvalds 已提交
669 670
}

671
static int check_target(struct ip6t_entry *e, const char *name)
L
Linus Torvalds 已提交
672
{
673 674 675 676 677 678 679
	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,
680
		.family    = NFPROTO_IPV6,
681
	};
L
Linus Torvalds 已提交
682 683
	int ret;

684
	t = ip6t_get_target(e);
685
	ret = xt_check_target(&par, t->u.target_size - sizeof(*t),
686
	      e->ipv6.proto, e->ipv6.invflags & IP6T_INV_PROTO);
687
	if (ret < 0) {
688 689
		duprintf("ip_tables: check failed for `%s'.\n",
			 t->u.kernel.target->name);
690
		return ret;
L
Linus Torvalds 已提交
691
	}
692
	return 0;
693
}
L
Linus Torvalds 已提交
694

695
static int
696 697 698 699 700 701 702
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;
703
	struct xt_mtchk_param mtpar;
704 705 706 707

	ret = check_entry(e, name);
	if (ret)
		return ret;
708

L
Linus Torvalds 已提交
709
	j = 0;
710 711 712
	mtpar.table     = name;
	mtpar.entryinfo = &e->ipv6;
	mtpar.hook_mask = e->comefrom;
713
	mtpar.family    = NFPROTO_IPV6;
714
	ret = IP6T_MATCH_ITERATE(e, find_check_match, &mtpar, &j);
L
Linus Torvalds 已提交
715 716 717 718
	if (ret != 0)
		goto cleanup_matches;

	t = ip6t_get_target(e);
719 720 721
	target = try_then_request_module(xt_find_target(AF_INET6,
							t->u.user.name,
							t->u.user.revision),
722 723
					 "ip6t_%s", t->u.user.name);
	if (IS_ERR(target) || !target) {
724
		duprintf("find_check_entry: `%s' not found\n", t->u.user.name);
725
		ret = target ? PTR_ERR(target) : -ENOENT;
L
Linus Torvalds 已提交
726 727 728
		goto cleanup_matches;
	}
	t->u.kernel.target = target;
729

730
	ret = check_target(e, name);
731 732 733
	if (ret)
		goto err;

L
Linus Torvalds 已提交
734 735
	(*i)++;
	return 0;
736 737
 err:
	module_put(t->u.kernel.target->me);
L
Linus Torvalds 已提交
738 739 740 741 742
 cleanup_matches:
	IP6T_MATCH_ITERATE(e, cleanup_match, &j);
	return ret;
}

743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
static bool check_underflow(struct ip6t_entry *e)
{
	const struct ip6t_entry_target *t;
	unsigned int verdict;

	if (!unconditional(&e->ipv6))
		return false;
	t = ip6t_get_target(e);
	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
		return false;
	verdict = ((struct ip6t_standard_target *)t)->verdict;
	verdict = -verdict - 1;
	return verdict == NF_DROP || verdict == NF_ACCEPT;
}

758
static int
L
Linus Torvalds 已提交
759
check_entry_size_and_hooks(struct ip6t_entry *e,
760
			   struct xt_table_info *newinfo,
L
Linus Torvalds 已提交
761 762 763 764
			   unsigned char *base,
			   unsigned char *limit,
			   const unsigned int *hook_entries,
			   const unsigned int *underflows,
765
			   unsigned int valid_hooks,
L
Linus Torvalds 已提交
766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783
			   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 */
784
	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
785 786
		if (!(valid_hooks & (1 << h)))
			continue;
L
Linus Torvalds 已提交
787 788
		if ((unsigned char *)e - base == hook_entries[h])
			newinfo->hook_entry[h] = hook_entries[h];
789
		if ((unsigned char *)e - base == underflows[h]) {
790 791 792 793
			if (!check_underflow(e)) {
				pr_err("Underflows must be unconditional and "
				       "use the STANDARD target with "
				       "ACCEPT/DROP\n");
794 795
				return -EINVAL;
			}
L
Linus Torvalds 已提交
796
			newinfo->underflow[h] = underflows[h];
797
		}
L
Linus Torvalds 已提交
798 799 800
	}

	/* Clear counters and comefrom */
801
	e->counters = ((struct xt_counters) { 0, 0 });
L
Linus Torvalds 已提交
802 803 804 805 806 807
	e->comefrom = 0;

	(*i)++;
	return 0;
}

808
static int
L
Linus Torvalds 已提交
809 810
cleanup_entry(struct ip6t_entry *e, unsigned int *i)
{
811
	struct xt_tgdtor_param par;
L
Linus Torvalds 已提交
812 813 814 815 816 817 818 819
	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);
820 821 822

	par.target   = t->u.kernel.target;
	par.targinfo = t->data;
823
	par.family   = NFPROTO_IPV6;
824 825 826
	if (par.target->destroy != NULL)
		par.target->destroy(&par);
	module_put(par.target->me);
L
Linus Torvalds 已提交
827 828 829 830 831 832 833 834
	return 0;
}

/* Checks and translates the user-supplied table segment (held in
   newinfo) */
static int
translate_table(const char *name,
		unsigned int valid_hooks,
835
		struct xt_table_info *newinfo,
836
		void *entry0,
L
Linus Torvalds 已提交
837 838 839 840 841 842 843 844 845 846 847 848
		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. */
849
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
850 851 852 853 854 855 856
		newinfo->hook_entry[i] = 0xFFFFFFFF;
		newinfo->underflow[i] = 0xFFFFFFFF;
	}

	duprintf("translate_table: size %u\n", newinfo->size);
	i = 0;
	/* Walk through entries, checking offsets. */
857
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
L
Linus Torvalds 已提交
858 859
				check_entry_size_and_hooks,
				newinfo,
860 861
				entry0,
				entry0 + size,
862
				hook_entries, underflows, valid_hooks, &i);
L
Linus Torvalds 已提交
863 864 865 866 867 868 869 870 871 872
	if (ret != 0)
		return ret;

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

	/* Check hooks all assigned */
873
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
874 875 876 877 878 879 880 881 882 883 884 885 886 887 888
		/* 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;
		}
	}

889 890 891
	if (!mark_source_chains(newinfo, valid_hooks, entry0))
		return -ELOOP;

L
Linus Torvalds 已提交
892 893
	/* Finally, each sanity check must pass */
	i = 0;
894
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
895
				find_check_entry, name, size, &i);
L
Linus Torvalds 已提交
896

897 898 899 900 901
	if (ret != 0) {
		IP6T_ENTRY_ITERATE(entry0, newinfo->size,
				   cleanup_entry, &i);
		return ret;
	}
L
Linus Torvalds 已提交
902 903

	/* And one copy for every other CPU */
904
	for_each_possible_cpu(i) {
905 906
		if (newinfo->entries[i] && newinfo->entries[i] != entry0)
			memcpy(newinfo->entries[i], entry0, newinfo->size);
L
Linus Torvalds 已提交
907 908
	}

909
	return ret;
L
Linus Torvalds 已提交
910 911 912 913 914
}

/* Gets counters. */
static inline int
add_entry_to_counter(const struct ip6t_entry *e,
915
		     struct xt_counters total[],
L
Linus Torvalds 已提交
916 917 918 919 920 921 922 923
		     unsigned int *i)
{
	ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt);

	(*i)++;
	return 0;
}

924 925 926 927 928 929 930 931 932 933 934
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
935
static void
936 937
get_counters(const struct xt_table_info *t,
	     struct xt_counters counters[])
L
Linus Torvalds 已提交
938 939 940
{
	unsigned int cpu;
	unsigned int i;
941 942 943 944 945
	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
946 947 948
	 *
	 * Bottom half has to be disabled to prevent deadlock
	 * if new softirq were to run and call ipt_do_table
949
	 */
950 951
	local_bh_disable();
	curcpu = smp_processor_id();
952 953 954 955 956 957 958

	i = 0;
	IP6T_ENTRY_ITERATE(t->entries[curcpu],
			   t->size,
			   set_entry_to_counter,
			   counters,
			   &i);
L
Linus Torvalds 已提交
959

960
	for_each_possible_cpu(cpu) {
961 962
		if (cpu == curcpu)
			continue;
L
Linus Torvalds 已提交
963
		i = 0;
964
		xt_info_wrlock(cpu);
965
		IP6T_ENTRY_ITERATE(t->entries[cpu],
L
Linus Torvalds 已提交
966 967 968 969
				  t->size,
				  add_entry_to_counter,
				  counters,
				  &i);
970
		xt_info_wrunlock(cpu);
L
Linus Torvalds 已提交
971
	}
972 973 974
	local_bh_enable();
}

975
static struct xt_counters *alloc_counters(struct xt_table *table)
L
Linus Torvalds 已提交
976
{
977
	unsigned int countersize;
978
	struct xt_counters *counters;
979
	struct xt_table_info *private = table->private;
L
Linus Torvalds 已提交
980 981 982 983

	/* We need atomic snapshot of counters: rest doesn't change
	   (other than comefrom, which userspace doesn't care
	   about). */
984
	countersize = sizeof(struct xt_counters) * private->number;
985
	counters = vmalloc_node(countersize, numa_node_id());
L
Linus Torvalds 已提交
986 987

	if (counters == NULL)
988
		return ERR_PTR(-ENOMEM);
989

990
	get_counters(private, counters);
991

992
	return counters;
993 994 995 996 997 998 999 1000 1001 1002
}

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;
1003
	const struct xt_table_info *private = table->private;
1004
	int ret = 0;
1005
	const void *loc_cpu_entry;
1006 1007 1008 1009 1010

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

1011 1012 1013 1014
	/* choose the copy that is on our node/cpu, ...
	 * This choice is lazy (because current thread is
	 * allowed to migrate to another cpu)
	 */
1015
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
1016
	if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) {
L
Linus Torvalds 已提交
1017 1018 1019 1020 1021 1022 1023 1024
		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;
1025 1026
		const struct ip6t_entry_match *m;
		const struct ip6t_entry_target *t;
L
Linus Torvalds 已提交
1027

1028
		e = (struct ip6t_entry *)(loc_cpu_entry + off);
L
Linus Torvalds 已提交
1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068
		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;
}

1069 1070 1071 1072 1073 1074 1075 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 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
#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

1142
static int get_info(struct net *net, void __user *user, int *len, int compat)
1143 1144 1145 1146 1147 1148
{
	char name[IP6T_TABLE_MAXNAMELEN];
	struct xt_table *t;
	int ret;

	if (*len != sizeof(struct ip6t_getinfo)) {
1149
		duprintf("length %u != %zu\n", *len,
1150 1151 1152 1153 1154 1155 1156 1157
			 sizeof(struct ip6t_getinfo));
		return -EINVAL;
	}

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

	name[IP6T_TABLE_MAXNAMELEN-1] = '\0';
1158 1159 1160 1161
#ifdef CONFIG_COMPAT
	if (compat)
		xt_compat_lock(AF_INET6);
#endif
1162
	t = try_then_request_module(xt_find_table_lock(net, AF_INET6, name),
1163 1164 1165
				    "ip6table_%s", name);
	if (t && !IS_ERR(t)) {
		struct ip6t_getinfo info;
1166
		const struct xt_table_info *private = t->private;
1167

1168 1169 1170 1171 1172 1173 1174 1175
#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
1176 1177 1178 1179 1180 1181 1182
		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;
1183
		strcpy(info.name, name);
1184 1185 1186 1187 1188 1189 1190 1191 1192 1193

		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;
1194 1195 1196 1197
#ifdef CONFIG_COMPAT
	if (compat)
		xt_compat_unlock(AF_INET6);
#endif
1198 1199 1200
	return ret;
}

L
Linus Torvalds 已提交
1201
static int
1202
get_entries(struct net *net, struct ip6t_get_entries __user *uptr, int *len)
L
Linus Torvalds 已提交
1203 1204
{
	int ret;
1205
	struct ip6t_get_entries get;
1206
	struct xt_table *t;
L
Linus Torvalds 已提交
1207

1208
	if (*len < sizeof(get)) {
1209
		duprintf("get_entries: %u < %zu\n", *len, sizeof(get));
1210 1211 1212 1213 1214
		return -EINVAL;
	}
	if (copy_from_user(&get, uptr, sizeof(get)) != 0)
		return -EFAULT;
	if (*len != sizeof(struct ip6t_get_entries) + get.size) {
1215 1216
		duprintf("get_entries: %u != %zu\n",
			 *len, sizeof(get) + get.size);
1217 1218 1219
		return -EINVAL;
	}

1220
	t = xt_find_table_lock(net, AF_INET6, get.name);
1221
	if (t && !IS_ERR(t)) {
1222 1223
		struct xt_table_info *private = t->private;
		duprintf("t->private->number = %u\n", private->number);
1224
		if (get.size == private->size)
1225
			ret = copy_entries_to_user(private->size,
L
Linus Torvalds 已提交
1226 1227 1228
						   t, uptr->entrytable);
		else {
			duprintf("get_entries: I've got %u not %u!\n",
1229
				 private->size, get.size);
1230
			ret = -EAGAIN;
L
Linus Torvalds 已提交
1231
		}
1232
		module_put(t->me);
1233
		xt_table_unlock(t);
L
Linus Torvalds 已提交
1234
	} else
1235
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1236 1237 1238 1239 1240

	return ret;
}

static int
1241
__do_replace(struct net *net, const char *name, unsigned int valid_hooks,
1242 1243
	     struct xt_table_info *newinfo, unsigned int num_counters,
	     void __user *counters_ptr)
L
Linus Torvalds 已提交
1244 1245
{
	int ret;
1246
	struct xt_table *t;
1247
	struct xt_table_info *oldinfo;
1248
	struct xt_counters *counters;
1249
	const void *loc_cpu_old_entry;
L
Linus Torvalds 已提交
1250

1251 1252
	ret = 0;
	counters = vmalloc_node(num_counters * sizeof(struct xt_counters),
1253
				numa_node_id());
L
Linus Torvalds 已提交
1254 1255
	if (!counters) {
		ret = -ENOMEM;
1256
		goto out;
L
Linus Torvalds 已提交
1257 1258
	}

1259
	t = try_then_request_module(xt_find_table_lock(net, AF_INET6, name),
1260
				    "ip6table_%s", name);
1261 1262
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1263
		goto free_newinfo_counters_untrans;
1264
	}
L
Linus Torvalds 已提交
1265 1266

	/* You lied! */
1267
	if (valid_hooks != t->valid_hooks) {
L
Linus Torvalds 已提交
1268
		duprintf("Valid hook crap: %08X vs %08X\n",
1269
			 valid_hooks, t->valid_hooks);
L
Linus Torvalds 已提交
1270
		ret = -EINVAL;
1271
		goto put_module;
L
Linus Torvalds 已提交
1272 1273
	}

1274
	oldinfo = xt_replace_table(t, num_counters, newinfo, &ret);
L
Linus Torvalds 已提交
1275 1276 1277 1278 1279 1280
	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);
1281 1282
	if ((oldinfo->number > oldinfo->initial_entries) ||
	    (newinfo->number <= oldinfo->initial_entries))
L
Linus Torvalds 已提交
1283 1284 1285 1286 1287
		module_put(t->me);
	if ((oldinfo->number > oldinfo->initial_entries) &&
	    (newinfo->number <= oldinfo->initial_entries))
		module_put(t->me);

1288
	/* Get the old counters, and synchronize with replace */
L
Linus Torvalds 已提交
1289
	get_counters(oldinfo, counters);
1290

L
Linus Torvalds 已提交
1291
	/* Decrease module usage counts and free resource */
1292
	loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()];
1293 1294
	IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,
			   NULL);
1295
	xt_free_table_info(oldinfo);
1296 1297
	if (copy_to_user(counters_ptr, counters,
			 sizeof(struct xt_counters) * num_counters) != 0)
L
Linus Torvalds 已提交
1298 1299
		ret = -EFAULT;
	vfree(counters);
1300
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1301 1302 1303 1304
	return ret;

 put_module:
	module_put(t->me);
1305
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1306 1307
 free_newinfo_counters_untrans:
	vfree(counters);
1308 1309 1310 1311 1312
 out:
	return ret;
}

static int
1313
do_replace(struct net *net, void __user *user, unsigned int len)
1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346
{
	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");

1347
	ret = __do_replace(net, tmp.name, tmp.valid_hooks, newinfo,
1348 1349 1350 1351 1352 1353 1354
			   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
1355
 free_newinfo:
1356
	xt_free_table_info(newinfo);
L
Linus Torvalds 已提交
1357 1358 1359
	return ret;
}

1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
/* 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;
}

L
Linus Torvalds 已提交
1373
static int
1374 1375
do_add_counters(struct net *net, void __user *user, unsigned int len,
		int compat)
L
Linus Torvalds 已提交
1376
{
1377
	unsigned int i, curcpu;
1378 1379 1380 1381 1382 1383
	struct xt_counters_info tmp;
	struct xt_counters *paddc;
	unsigned int num_counters;
	char *name;
	int size;
	void *ptmp;
1384
	struct xt_table *t;
1385
	const struct xt_table_info *private;
1386
	int ret = 0;
1387
	const void *loc_cpu_entry;
1388 1389
#ifdef CONFIG_COMPAT
	struct compat_xt_counters_info compat_tmp;
L
Linus Torvalds 已提交
1390

1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401
	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
1402 1403
		return -EFAULT;

1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415
#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
1416 1417
		return -EINVAL;

1418
	paddc = vmalloc_node(len - size, numa_node_id());
L
Linus Torvalds 已提交
1419 1420 1421
	if (!paddc)
		return -ENOMEM;

1422
	if (copy_from_user(paddc, user + size, len - size) != 0) {
L
Linus Torvalds 已提交
1423 1424 1425 1426
		ret = -EFAULT;
		goto free;
	}

1427
	t = xt_find_table_lock(net, AF_INET6, name);
1428 1429
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1430
		goto free;
1431
	}
L
Linus Torvalds 已提交
1432

1433 1434

	local_bh_disable();
1435
	private = t->private;
1436
	if (private->number != num_counters) {
L
Linus Torvalds 已提交
1437 1438 1439 1440 1441
		ret = -EINVAL;
		goto unlock_up_free;
	}

	i = 0;
1442
	/* Choose the copy that is on our node */
1443 1444 1445
	curcpu = smp_processor_id();
	xt_info_wrlock(curcpu);
	loc_cpu_entry = private->entries[curcpu];
1446
	IP6T_ENTRY_ITERATE(loc_cpu_entry,
1447
			  private->size,
L
Linus Torvalds 已提交
1448
			  add_counter_to_entry,
1449
			  paddc,
L
Linus Torvalds 已提交
1450
			  &i);
1451 1452
	xt_info_wrunlock(curcpu);

L
Linus Torvalds 已提交
1453
 unlock_up_free:
1454
	local_bh_enable();
1455
	xt_table_unlock(t);
1456
	module_put(t->me);
L
Linus Torvalds 已提交
1457 1458 1459 1460 1461 1462
 free:
	vfree(paddc);

	return ret;
}

1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477
#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,
1478
			  unsigned int *size, struct xt_counters *counters,
1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519
			  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;
}

1520
static int
1521 1522 1523 1524
compat_find_calc_match(struct ip6t_entry_match *m,
		       const char *name,
		       const struct ip6t_ip6 *ipv6,
		       unsigned int hookmask,
1525
		       int *size, unsigned int *i)
1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543
{
	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;
}

1544
static int
1545 1546 1547 1548 1549 1550 1551 1552 1553
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;
}

1554
static int
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568
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;
}

1569
static int
1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582
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;
1583 1584
	unsigned int j;
	int ret, off, h;
1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 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

	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;
}

1693
static int compat_check_entry(struct ip6t_entry *e, const char *name,
1694 1695
				     unsigned int *i)
{
1696 1697
	unsigned int j;
	int ret;
1698
	struct xt_mtchk_param mtpar;
1699 1700

	j = 0;
1701 1702 1703
	mtpar.table     = name;
	mtpar.entryinfo = &e->ipv6;
	mtpar.hook_mask = e->comefrom;
1704
	mtpar.family    = NFPROTO_IPV6;
1705
	ret = IP6T_MATCH_ITERATE(e, check_match, &mtpar, &j);
1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 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
	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
1843
compat_do_replace(struct net *net, void __user *user, unsigned int len)
1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862
{
	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;

1863
	/* choose the copy that is on our node/cpu */
1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879
	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");

1880
	ret = __do_replace(net, tmp.name, tmp.valid_hooks, newinfo,
1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903
			   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:
1904
		ret = compat_do_replace(sock_net(sk), user, len);
1905 1906 1907
		break;

	case IP6T_SO_SET_ADD_COUNTERS:
1908
		ret = do_add_counters(sock_net(sk), user, len, 1);
1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929
		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;
1930
	const struct xt_table_info *private = table->private;
1931 1932 1933
	void __user *pos;
	unsigned int size;
	int ret = 0;
1934
	const void *loc_cpu_entry;
1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956
	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
1957 1958
compat_get_entries(struct net *net, struct compat_ip6t_get_entries __user *uptr,
		   int *len)
1959 1960 1961 1962 1963 1964
{
	int ret;
	struct compat_ip6t_get_entries get;
	struct xt_table *t;

	if (*len < sizeof(get)) {
1965
		duprintf("compat_get_entries: %u < %zu\n", *len, sizeof(get));
1966 1967 1968 1969 1970 1971 1972
		return -EINVAL;
	}

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

	if (*len != sizeof(struct compat_ip6t_get_entries) + get.size) {
1973 1974
		duprintf("compat_get_entries: %u != %zu\n",
			 *len, sizeof(get) + get.size);
1975 1976 1977 1978
		return -EINVAL;
	}

	xt_compat_lock(AF_INET6);
1979
	t = xt_find_table_lock(net, AF_INET6, get.name);
1980
	if (t && !IS_ERR(t)) {
1981
		const struct xt_table_info *private = t->private;
1982
		struct xt_table_info info;
1983
		duprintf("t->private->number = %u\n", private->number);
1984 1985 1986 1987 1988 1989
		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",
1990
				 private->size, get.size);
1991
			ret = -EAGAIN;
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
		}
		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:
2015
		ret = get_info(sock_net(sk), user, len, 1);
2016 2017
		break;
	case IP6T_SO_GET_ENTRIES:
2018
		ret = compat_get_entries(sock_net(sk), user, len);
2019 2020 2021 2022 2023 2024 2025 2026
		break;
	default:
		ret = do_ip6t_get_ctl(sk, cmd, user, len);
	}
	return ret;
}
#endif

L
Linus Torvalds 已提交
2027 2028 2029 2030 2031 2032 2033 2034 2035 2036
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:
2037
		ret = do_replace(sock_net(sk), user, len);
L
Linus Torvalds 已提交
2038 2039 2040
		break;

	case IP6T_SO_SET_ADD_COUNTERS:
2041
		ret = do_add_counters(sock_net(sk), user, len, 0);
L
Linus Torvalds 已提交
2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060
		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) {
2061
	case IP6T_SO_GET_INFO:
2062
		ret = get_info(sock_net(sk), user, len, 0);
2063
		break;
L
Linus Torvalds 已提交
2064

2065
	case IP6T_SO_GET_ENTRIES:
2066
		ret = get_entries(sock_net(sk), user, len);
L
Linus Torvalds 已提交
2067 2068
		break;

2069 2070 2071
	case IP6T_SO_GET_REVISION_MATCH:
	case IP6T_SO_GET_REVISION_TARGET: {
		struct ip6t_get_revision rev;
2072
		int target;
2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083

		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)
2084
			target = 1;
2085
		else
2086
			target = 0;
2087

2088 2089 2090
		try_then_request_module(xt_find_revision(AF_INET6, rev.name,
							 rev.revision,
							 target, &ret),
2091 2092 2093 2094
					"ip6t_%s", rev.name);
		break;
	}

L
Linus Torvalds 已提交
2095 2096 2097 2098 2099 2100 2101 2102
	default:
		duprintf("do_ip6t_get_ctl: unknown request %i\n", cmd);
		ret = -EINVAL;
	}

	return ret;
}

2103 2104
struct xt_table *ip6t_register_table(struct net *net,
				     const struct xt_table *table,
2105
				     const struct ip6t_replace *repl)
L
Linus Torvalds 已提交
2106 2107
{
	int ret;
2108
	struct xt_table_info *newinfo;
2109
	struct xt_table_info bootstrap
L
Linus Torvalds 已提交
2110
		= { 0, 0, 0, { 0 }, { 0 }, { } };
2111
	void *loc_cpu_entry;
2112
	struct xt_table *new_table;
L
Linus Torvalds 已提交
2113

2114
	newinfo = xt_alloc_table_info(repl->size);
2115 2116 2117 2118
	if (!newinfo) {
		ret = -ENOMEM;
		goto out;
	}
L
Linus Torvalds 已提交
2119

2120
	/* choose the copy on our node/cpu, but dont care about preemption */
2121 2122
	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
	memcpy(loc_cpu_entry, repl->entries, repl->size);
L
Linus Torvalds 已提交
2123 2124

	ret = translate_table(table->name, table->valid_hooks,
2125
			      newinfo, loc_cpu_entry, repl->size,
L
Linus Torvalds 已提交
2126 2127 2128
			      repl->num_entries,
			      repl->hook_entry,
			      repl->underflow);
2129 2130
	if (ret != 0)
		goto out_free;
L
Linus Torvalds 已提交
2131

2132
	new_table = xt_register_table(net, table, &bootstrap, newinfo);
2133
	if (IS_ERR(new_table)) {
2134 2135
		ret = PTR_ERR(new_table);
		goto out_free;
L
Linus Torvalds 已提交
2136
	}
2137
	return new_table;
L
Linus Torvalds 已提交
2138

2139 2140 2141 2142
out_free:
	xt_free_table_info(newinfo);
out:
	return ERR_PTR(ret);
L
Linus Torvalds 已提交
2143 2144
}

2145
void ip6t_unregister_table(struct xt_table *table)
L
Linus Torvalds 已提交
2146
{
2147
	struct xt_table_info *private;
2148
	void *loc_cpu_entry;
2149
	struct module *table_owner = table->me;
2150

2151
	private = xt_unregister_table(table);
L
Linus Torvalds 已提交
2152 2153

	/* Decrease module usage counts and free resources */
2154 2155
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
	IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL);
2156 2157
	if (private->number > private->initial_entries)
		module_put(table_owner);
2158
	xt_free_table_info(private);
L
Linus Torvalds 已提交
2159 2160 2161
}

/* Returns 1 if the type and code is matched by the range, 0 otherwise */
2162
static inline bool
L
Linus Torvalds 已提交
2163 2164
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,
2165
		     bool invert)
L
Linus Torvalds 已提交
2166 2167 2168 2169 2170
{
	return (type == test_type && code >= min_code && code <= max_code)
		^ invert;
}

2171
static bool
2172
icmp6_match(const struct sk_buff *skb, const struct xt_match_param *par)
L
Linus Torvalds 已提交
2173
{
2174 2175
	const struct icmp6hdr *ic;
	struct icmp6hdr _icmph;
2176
	const struct ip6t_icmp *icmpinfo = par->matchinfo;
L
Linus Torvalds 已提交
2177 2178

	/* Must not be a fragment. */
2179
	if (par->fragoff != 0)
2180
		return false;
L
Linus Torvalds 已提交
2181

2182
	ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph);
L
Linus Torvalds 已提交
2183 2184
	if (ic == NULL) {
		/* We've been asked to examine this packet, and we
2185 2186
		 * can't.  Hence, no choice but to drop.
		 */
L
Linus Torvalds 已提交
2187
		duprintf("Dropping evil ICMP tinygram.\n");
2188
		*par->hotdrop = true;
2189
		return false;
L
Linus Torvalds 已提交
2190 2191 2192 2193 2194 2195 2196 2197 2198 2199
	}

	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. */
2200
static bool icmp6_checkentry(const struct xt_mtchk_param *par)
L
Linus Torvalds 已提交
2201
{
2202
	const struct ip6t_icmp *icmpinfo = par->matchinfo;
L
Linus Torvalds 已提交
2203

2204 2205
	/* Must specify no unknown invflags */
	return !(icmpinfo->invflags & ~IP6T_ICMP_INV);
L
Linus Torvalds 已提交
2206 2207 2208
}

/* The built-in targets: standard (NULL) and error. */
2209
static struct xt_target ip6t_standard_target __read_mostly = {
L
Linus Torvalds 已提交
2210
	.name		= IP6T_STANDARD_TARGET,
2211
	.targetsize	= sizeof(int),
2212
	.family		= NFPROTO_IPV6,
2213 2214 2215 2216 2217
#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
2218 2219
};

2220
static struct xt_target ip6t_error_target __read_mostly = {
L
Linus Torvalds 已提交
2221 2222
	.name		= IP6T_ERROR_TARGET,
	.target		= ip6t_error,
2223
	.targetsize	= IP6T_FUNCTION_MAXNAMELEN,
2224
	.family		= NFPROTO_IPV6,
L
Linus Torvalds 已提交
2225 2226 2227 2228 2229 2230 2231
};

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,
2232 2233 2234
#ifdef CONFIG_COMPAT
	.compat_set	= compat_do_ip6t_set_ctl,
#endif
L
Linus Torvalds 已提交
2235 2236 2237
	.get_optmin	= IP6T_BASE_CTL,
	.get_optmax	= IP6T_SO_GET_MAX+1,
	.get		= do_ip6t_get_ctl,
2238 2239 2240
#ifdef CONFIG_COMPAT
	.compat_get	= compat_do_ip6t_get_ctl,
#endif
2241
	.owner		= THIS_MODULE,
L
Linus Torvalds 已提交
2242 2243
};

2244
static struct xt_match icmp6_matchstruct __read_mostly = {
L
Linus Torvalds 已提交
2245
	.name		= "icmp6",
2246
	.match		= icmp6_match,
2247 2248 2249
	.matchsize	= sizeof(struct ip6t_icmp),
	.checkentry	= icmp6_checkentry,
	.proto		= IPPROTO_ICMPV6,
2250
	.family		= NFPROTO_IPV6,
L
Linus Torvalds 已提交
2251 2252
};

2253 2254
static int __net_init ip6_tables_net_init(struct net *net)
{
2255
	return xt_proto_init(net, NFPROTO_IPV6);
2256 2257 2258 2259
}

static void __net_exit ip6_tables_net_exit(struct net *net)
{
2260
	xt_proto_fini(net, NFPROTO_IPV6);
2261 2262 2263 2264 2265 2266 2267
}

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

2268
static int __init ip6_tables_init(void)
L
Linus Torvalds 已提交
2269 2270 2271
{
	int ret;

2272
	ret = register_pernet_subsys(&ip6_tables_net_ops);
2273 2274
	if (ret < 0)
		goto err1;
2275

L
Linus Torvalds 已提交
2276
	/* Noone else will be downing sem now, so we won't sleep */
2277 2278 2279 2280 2281 2282 2283 2284 2285
	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
2286 2287 2288

	/* Register setsockopt */
	ret = nf_register_sockopt(&ip6t_sockopts);
2289 2290
	if (ret < 0)
		goto err5;
L
Linus Torvalds 已提交
2291

2292
	printk(KERN_INFO "ip6_tables: (C) 2000-2006 Netfilter Core Team\n");
L
Linus Torvalds 已提交
2293
	return 0;
2294 2295 2296 2297 2298 2299 2300 2301

err5:
	xt_unregister_match(&icmp6_matchstruct);
err4:
	xt_unregister_target(&ip6t_error_target);
err3:
	xt_unregister_target(&ip6t_standard_target);
err2:
2302
	unregister_pernet_subsys(&ip6_tables_net_ops);
2303 2304
err1:
	return ret;
L
Linus Torvalds 已提交
2305 2306
}

2307
static void __exit ip6_tables_fini(void)
L
Linus Torvalds 已提交
2308 2309
{
	nf_unregister_sockopt(&ip6t_sockopts);
2310

2311 2312 2313
	xt_unregister_match(&icmp6_matchstruct);
	xt_unregister_target(&ip6t_error_target);
	xt_unregister_target(&ip6t_standard_target);
2314 2315

	unregister_pernet_subsys(&ip6_tables_net_ops);
L
Linus Torvalds 已提交
2316 2317
}

2318
/*
2319 2320 2321 2322 2323 2324 2325
 * 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.
 *
2326 2327 2328
 * If the first fragment doesn't contain the final protocol header or
 * NEXTHDR_NONE it is considered invalid.
 *
2329 2330 2331 2332
 * 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.
2333 2334
 *
 */
2335 2336
int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
		  int target, unsigned short *fragoff)
2337
{
2338
	unsigned int start = skb_network_offset(skb) + sizeof(struct ipv6hdr);
2339
	u8 nexthdr = ipv6_hdr(skb)->nexthdr;
2340 2341
	unsigned int len = skb->len - start;

2342 2343 2344
	if (fragoff)
		*fragoff = 0;

2345 2346 2347 2348
	while (nexthdr != target) {
		struct ipv6_opt_hdr _hdr, *hp;
		unsigned int hdrlen;

2349 2350 2351
		if ((!ipv6_ext_hdr(nexthdr)) || nexthdr == NEXTHDR_NONE) {
			if (target < 0)
				break;
2352
			return -ENOENT;
2353 2354
		}

2355 2356
		hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr);
		if (hp == NULL)
2357
			return -EBADMSG;
2358
		if (nexthdr == NEXTHDR_FRAGMENT) {
A
Al Viro 已提交
2359 2360
			unsigned short _frag_off;
			__be16 *fp;
2361 2362 2363 2364 2365 2366
			fp = skb_header_pointer(skb,
						start+offsetof(struct frag_hdr,
							       frag_off),
						sizeof(_frag_off),
						&_frag_off);
			if (fp == NULL)
2367
				return -EBADMSG;
2368

2369 2370 2371 2372
			_frag_off = ntohs(*fp) & ~0x7;
			if (_frag_off) {
				if (target < 0 &&
				    ((!ipv6_ext_hdr(hp->nexthdr)) ||
2373
				     hp->nexthdr == NEXTHDR_NONE)) {
2374 2375 2376 2377
					if (fragoff)
						*fragoff = _frag_off;
					return hp->nexthdr;
				}
2378
				return -ENOENT;
2379
			}
2380 2381
			hdrlen = 8;
		} else if (nexthdr == NEXTHDR_AUTH)
2382
			hdrlen = (hp->hdrlen + 2) << 2;
2383
		else
2384
			hdrlen = ipv6_optlen(hp);
2385 2386 2387 2388 2389 2390 2391

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

	*offset = start;
2392
	return nexthdr;
2393 2394
}

L
Linus Torvalds 已提交
2395 2396 2397 2398
EXPORT_SYMBOL(ip6t_register_table);
EXPORT_SYMBOL(ip6t_unregister_table);
EXPORT_SYMBOL(ip6t_do_table);
EXPORT_SYMBOL(ip6t_ext_hdr);
2399
EXPORT_SYMBOL(ipv6_find_hdr);
L
Linus Torvalds 已提交
2400

2401 2402
module_init(ip6_tables_init);
module_exit(ip6_tables_fini);