ip6_tables.c 58.6 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>
32
#include "../../netfilter/xt_repldata.h"
L
Linus Torvalds 已提交
33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

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",		\
59
		       __func__, __FILE__, __LINE__);	\
L
Linus Torvalds 已提交
60 61 62 63 64 65 66 67 68 69 70
} while(0)
#else
#define IP_NF_ASSERT(x)
#endif

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

71 72 73 74 75 76
void *ip6t_alloc_initial_table(const struct xt_table *info)
{
	return xt_alloc_initial_table(ip6t, IP6T);
}
EXPORT_SYMBOL_GPL(ip6t_alloc_initial_table);

77
/*
L
Linus Torvalds 已提交
78
   We keep a set of rules for each CPU, so we can avoid write-locking
79 80 81 82
   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
83 84 85 86

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

/* Check for an extension */
87
int
L
Linus Torvalds 已提交
88 89
ip6t_ext_hdr(u8 nexthdr)
{
90 91 92 93 94 95 96
	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
97 98 99
}

/* Returns whether matches rule or not. */
100
/* Performance critical - called for every packet */
101
static inline bool
L
Linus Torvalds 已提交
102 103 104 105 106
ip6_packet_match(const struct sk_buff *skb,
		 const char *indev,
		 const char *outdev,
		 const struct ip6t_ip6 *ip6info,
		 unsigned int *protoff,
107
		 int *fragoff, bool *hotdrop)
L
Linus Torvalds 已提交
108 109
{
	unsigned long ret;
110
	const struct ipv6hdr *ipv6 = ipv6_hdr(skb);
L
Linus Torvalds 已提交
111

112
#define FWINV(bool, invflg) ((bool) ^ !!(ip6info->invflags & (invflg)))
L
Linus Torvalds 已提交
113

114
	if (FWINV(ipv6_masked_addr_cmp(&ipv6->saddr, &ip6info->smsk,
115 116 117
				       &ip6info->src), IP6T_INV_SRCIP) ||
	    FWINV(ipv6_masked_addr_cmp(&ipv6->daddr, &ip6info->dmsk,
				       &ip6info->dst), IP6T_INV_DSTIP)) {
L
Linus Torvalds 已提交
118 119 120 121 122 123 124 125
		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)" : "");*/
126
		return false;
L
Linus Torvalds 已提交
127 128
	}

129
	ret = ifname_compare_aligned(indev, ip6info->iniface, ip6info->iniface_mask);
L
Linus Torvalds 已提交
130 131 132 133 134

	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)":"");
135
		return false;
L
Linus Torvalds 已提交
136 137
	}

138
	ret = ifname_compare_aligned(outdev, ip6info->outiface, ip6info->outiface_mask);
L
Linus Torvalds 已提交
139 140 141 142 143

	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)":"");
144
		return false;
L
Linus Torvalds 已提交
145 146 147 148 149 150
	}

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

	/* look for the desired protocol header */
	if((ip6info->flags & IP6T_F_PROTO)) {
151 152
		int protohdr;
		unsigned short _frag_off;
L
Linus Torvalds 已提交
153

154
		protohdr = ipv6_find_hdr(skb, protoff, -1, &_frag_off);
155 156
		if (protohdr < 0) {
			if (_frag_off == 0)
157
				*hotdrop = true;
158
			return false;
159
		}
160
		*fragoff = _frag_off;
L
Linus Torvalds 已提交
161 162

		dprintf("Packet protocol %hi ?= %s%hi.\n",
163
				protohdr,
L
Linus Torvalds 已提交
164 165 166
				ip6info->invflags & IP6T_INV_PROTO ? "!":"",
				ip6info->proto);

167
		if (ip6info->proto == protohdr) {
L
Linus Torvalds 已提交
168
			if(ip6info->invflags & IP6T_INV_PROTO) {
169
				return false;
L
Linus Torvalds 已提交
170
			}
171
			return true;
L
Linus Torvalds 已提交
172 173 174 175 176
		}

		/* We need match for the '-p all', too! */
		if ((ip6info->proto != 0) &&
			!(ip6info->invflags & IP6T_INV_PROTO))
177
			return false;
L
Linus Torvalds 已提交
178
	}
179
	return true;
L
Linus Torvalds 已提交
180 181 182
}

/* should be ip6 safe */
183
static bool
L
Linus Torvalds 已提交
184 185 186 187 188
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);
189
		return false;
L
Linus Torvalds 已提交
190 191 192 193
	}
	if (ipv6->invflags & ~IP6T_INV_MASK) {
		duprintf("Unknown invflag bits set: %08X\n",
			 ipv6->invflags & ~IP6T_INV_MASK);
194
		return false;
L
Linus Torvalds 已提交
195
	}
196
	return true;
L
Linus Torvalds 已提交
197 198 199
}

static unsigned int
200
ip6t_error(struct sk_buff *skb, const struct xt_target_param *par)
L
Linus Torvalds 已提交
201 202
{
	if (net_ratelimit())
203 204
		printk("ip6_tables: error: `%s'\n",
		       (const char *)par->targinfo);
L
Linus Torvalds 已提交
205 206 207 208

	return NF_DROP;
}

209 210
/* Performance critical - called for every packet */
static inline bool
211 212
do_match(struct ip6t_entry_match *m, const struct sk_buff *skb,
	 struct xt_match_param *par)
L
Linus Torvalds 已提交
213
{
214 215 216
	par->match     = m->u.kernel.match;
	par->matchinfo = m->data;

L
Linus Torvalds 已提交
217
	/* Stop iteration if it doesn't match */
218
	if (!m->u.kernel.match->match(skb, par))
219
		return true;
L
Linus Torvalds 已提交
220
	else
221
		return false;
L
Linus Torvalds 已提交
222 223 224 225 226 227 228 229
}

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

230
/* All zeroes == unconditional rule. */
231
/* Mildly perf critical (only if packet tracing is on) */
232
static inline bool unconditional(const struct ip6t_ip6 *ipv6)
233
{
234
	static const struct ip6t_ip6 uncond;
235

236
	return memcmp(ipv6, &uncond, sizeof(uncond)) == 0;
237 238 239 240 241
}

#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
    defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
/* This cries for unification! */
242
static const char *const hooknames[] = {
243 244 245 246 247
	[NF_INET_PRE_ROUTING]		= "PREROUTING",
	[NF_INET_LOCAL_IN]		= "INPUT",
	[NF_INET_FORWARD]		= "FORWARD",
	[NF_INET_LOCAL_OUT]		= "OUTPUT",
	[NF_INET_POST_ROUTING]		= "POSTROUTING",
248 249 250 251 252 253 254 255
};

enum nf_ip_trace_comments {
	NF_IP6_TRACE_COMMENT_RULE,
	NF_IP6_TRACE_COMMENT_RETURN,
	NF_IP6_TRACE_COMMENT_POLICY,
};

256
static const char *const comments[] = {
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271
	[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,
		},
	},
};

272
/* Mildly perf critical (only if packet tracing is on) */
273 274
static inline int
get_chainname_rulenum(struct ip6t_entry *s, struct ip6t_entry *e,
275 276
		      const char *hookname, const char **chainname,
		      const char **comment, unsigned int *rulenum)
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)++;

287 288 289 290 291
		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)) {
292 293
			/* Tail of chains: STANDARD target (return/policy) */
			*comment = *chainname == hookname
294 295
				? comments[NF_IP6_TRACE_COMMENT_POLICY]
				: comments[NF_IP6_TRACE_COMMENT_RETURN];
296 297 298 299 300 301 302 303 304 305 306 307
		}
		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,
308
			 const char *tablename,
309 310 311 312
			 struct xt_table_info *private,
			 struct ip6t_entry *e)
{
	void *table_base;
313
	const struct ip6t_entry *root;
314
	const char *hookname, *chainname, *comment;
315 316
	unsigned int rulenum = 0;

317
	table_base = private->entries[smp_processor_id()];
318 319
	root = get_entry(table_base, private->hook_entry[hook]);

320 321
	hookname = chainname = hooknames[hook];
	comment = comments[NF_IP6_TRACE_COMMENT_RULE];
322 323 324 325 326 327 328 329 330 331 332 333

	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

334 335 336 337 338 339
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
340 341
/* Returns one of the generic firewall policies, like NF_ACCEPT. */
unsigned int
342
ip6t_do_table(struct sk_buff *skb,
L
Linus Torvalds 已提交
343 344 345
	      unsigned int hook,
	      const struct net_device *in,
	      const struct net_device *out,
346
	      struct xt_table *table)
L
Linus Torvalds 已提交
347
{
348 349
#define tb_comefrom ((struct ip6t_entry *)table_base)->comefrom

350
	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
351
	bool hotdrop = false;
L
Linus Torvalds 已提交
352 353 354 355 356
	/* 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;
357
	struct xt_table_info *private;
358
	struct xt_match_param mtpar;
359
	struct xt_target_param tgpar;
L
Linus Torvalds 已提交
360 361 362 363 364 365 366 367 368 369

	/* 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. */
370
	mtpar.hotdrop = &hotdrop;
371 372
	mtpar.in      = tgpar.in  = in;
	mtpar.out     = tgpar.out = out;
373
	mtpar.family  = tgpar.family = NFPROTO_IPV6;
374
	mtpar.hooknum = tgpar.hooknum = hook;
L
Linus Torvalds 已提交
375 376

	IP_NF_ASSERT(table->valid_hooks & (1 << hook));
377

378 379 380
	xt_info_rdlock_bh();
	private = table->private;
	table_base = private->entries[smp_processor_id()];
381

382
	e = get_entry(table_base, private->hook_entry[hook]);
L
Linus Torvalds 已提交
383 384

	/* For return from builtin chain */
385
	back = get_entry(table_base, private->underflow[hook]);
L
Linus Torvalds 已提交
386 387

	do {
388 389
		struct ip6t_entry_target *t;

L
Linus Torvalds 已提交
390 391
		IP_NF_ASSERT(e);
		IP_NF_ASSERT(back);
392
		if (!ip6_packet_match(skb, indev, outdev, &e->ipv6,
393 394
		    &mtpar.thoff, &mtpar.fragoff, &hotdrop) ||
		    IP6T_MATCH_ITERATE(e, do_match, skb, &mtpar) != 0) {
395 396 397
			e = ip6t_next_entry(e);
			continue;
		}
L
Linus Torvalds 已提交
398

399 400 401
		ADD_COUNTER(e->counters,
			    ntohs(ipv6_hdr(skb)->payload_len) +
			    sizeof(struct ipv6hdr), 1);
L
Linus Torvalds 已提交
402

403 404
		t = ip6t_get_target(e);
		IP_NF_ASSERT(t->u.kernel.target);
405 406 407

#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
    defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
408 409 410 411
		/* The packet is traced: log it */
		if (unlikely(skb->nf_trace))
			trace_packet(skb, hook, in, out,
				     table->name, private, e);
412
#endif
413 414 415 416 417 418 419 420 421 422
		/* 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
423
				}
424 425 426 427
				e = back;
				back = get_entry(table_base, back->comefrom);
				continue;
			}
428 429
			if (table_base + v != ip6t_next_entry(e) &&
			    !(e->ipv6.flags & IP6T_F_GOTO)) {
430 431 432 433 434 435
				/* 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
436

437
			e = get_entry(table_base, v);
438 439 440 441 442 443 444
			continue;
		}

		/* Targets which reenter must return
		   abs. verdicts */
		tgpar.target   = t->u.kernel.target;
		tgpar.targinfo = t->data;
445

L
Linus Torvalds 已提交
446
#ifdef CONFIG_NETFILTER_DEBUG
447
		tb_comefrom = 0xeeeeeeec;
L
Linus Torvalds 已提交
448
#endif
449
		verdict = t->u.kernel.target->target(skb, &tgpar);
L
Linus Torvalds 已提交
450 451

#ifdef CONFIG_NETFILTER_DEBUG
452
		if (tb_comefrom != 0xeeeeeeec && verdict == IP6T_CONTINUE) {
453 454 455
			printk("Target %s reentered!\n",
			       t->u.kernel.target->name);
			verdict = NF_DROP;
L
Linus Torvalds 已提交
456
		}
457
		tb_comefrom = 0x57acc001;
458 459 460 461 462 463
#endif
		if (verdict == IP6T_CONTINUE)
			e = ip6t_next_entry(e);
		else
			/* Verdict */
			break;
L
Linus Torvalds 已提交
464 465 466
	} while (!hotdrop);

#ifdef CONFIG_NETFILTER_DEBUG
467
	tb_comefrom = NETFILTER_LINK_POISON;
L
Linus Torvalds 已提交
468
#endif
469
	xt_info_rdunlock_bh();
L
Linus Torvalds 已提交
470 471 472 473 474 475 476 477

#ifdef DEBUG_ALLOW_ALL
	return NF_ACCEPT;
#else
	if (hotdrop)
		return NF_DROP;
	else return verdict;
#endif
478 479

#undef tb_comefrom
L
Linus Torvalds 已提交
480 481 482 483 484
}

/* Figures out from what hook each rule can be called: returns 0 if
   there are loops.  Puts hook bitmask in comefrom. */
static int
485
mark_source_chains(struct xt_table_info *newinfo,
486
		   unsigned int valid_hooks, void *entry0)
L
Linus Torvalds 已提交
487 488 489 490 491
{
	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 */
492
	for (hook = 0; hook < NF_INET_NUMHOOKS; hook++) {
L
Linus Torvalds 已提交
493
		unsigned int pos = newinfo->hook_entry[hook];
494
		struct ip6t_entry *e = (struct ip6t_entry *)(entry0 + pos);
L
Linus Torvalds 已提交
495 496 497 498 499 500 501 502 503 504

		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);
505
			int visited = e->comefrom & (1 << hook);
L
Linus Torvalds 已提交
506

507
			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
508 509 510 511
				printk("iptables: loop hook %u pos %u %08X.\n",
				       hook, pos, e->comefrom);
				return 0;
			}
512
			e->comefrom |= ((1 << hook) | (1 << NF_INET_NUMHOOKS));
L
Linus Torvalds 已提交
513 514

			/* Unconditional return/END. */
515 516 517 518 519
			if ((e->target_offset == sizeof(struct ip6t_entry) &&
			     (strcmp(t->target.u.user.name,
				     IP6T_STANDARD_TARGET) == 0) &&
			     t->verdict < 0 &&
			     unconditional(&e->ipv6)) || visited) {
L
Linus Torvalds 已提交
520 521
				unsigned int oldpos, size;

522 523 524
				if ((strcmp(t->target.u.user.name,
					    IP6T_STANDARD_TARGET) == 0) &&
				    t->verdict < -NF_MAX_VERDICT - 1) {
525 526 527 528 529 530
					duprintf("mark_source_chains: bad "
						"negative verdict (%i)\n",
								t->verdict);
					return 0;
				}

L
Linus Torvalds 已提交
531 532 533
				/* Return: backtrack through the last
				   big jump. */
				do {
534
					e->comefrom ^= (1<<NF_INET_NUMHOOKS);
L
Linus Torvalds 已提交
535 536
#ifdef DEBUG_IP_FIREWALL_USER
					if (e->comefrom
537
					    & (1 << NF_INET_NUMHOOKS)) {
L
Linus Torvalds 已提交
538 539 540 541 542 543 544 545 546 547 548 549 550 551 552
						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 *)
553
						(entry0 + pos);
L
Linus Torvalds 已提交
554 555 556 557 558
				} while (oldpos == pos + e->next_offset);

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

				if (strcmp(t->target.u.user.name,
566 567
					   IP6T_STANDARD_TARGET) == 0 &&
				    newpos >= 0) {
568 569 570 571 572 573 574
					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
575 576 577 578 579 580 581 582
					/* 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 *)
583
					(entry0 + newpos);
L
Linus Torvalds 已提交
584 585 586 587 588 589 590 591 592 593
				e->counters.pcnt = pos;
				pos = newpos;
			}
		}
		next:
		duprintf("Finished chain %u\n", hook);
	}
	return 1;
}

594
static int
595
cleanup_match(struct ip6t_entry_match *m, struct net *net, unsigned int *i)
L
Linus Torvalds 已提交
596
{
597 598
	struct xt_mtdtor_param par;

L
Linus Torvalds 已提交
599 600 601
	if (i && (*i)-- == 0)
		return 1;

602
	par.net       = net;
603 604
	par.match     = m->u.kernel.match;
	par.matchinfo = m->data;
605
	par.family    = NFPROTO_IPV6;
606 607 608
	if (par.match->destroy != NULL)
		par.match->destroy(&par);
	module_put(par.match->me);
L
Linus Torvalds 已提交
609 610 611
	return 0;
}

612
static int
613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
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;
}

633 634
static int check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par,
		       unsigned int *i)
635
{
636
	const struct ip6t_ip6 *ipv6 = par->entryinfo;
637 638
	int ret;

639 640 641
	par->match     = m->u.kernel.match;
	par->matchinfo = m->data;

642
	ret = xt_check_match(par, m->u.match_size - sizeof(*m),
643
			     ipv6->proto, ipv6->invflags & IP6T_INV_PROTO);
644
	if (ret < 0) {
645
		duprintf("ip_tables: check failed for `%s'.\n",
646
			 par.match->name);
647
		return ret;
648
	}
649 650
	++*i;
	return 0;
651 652
}

653
static int
654
find_check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par,
655
		 unsigned int *i)
L
Linus Torvalds 已提交
656
{
657
	struct xt_match *match;
658
	int ret;
L
Linus Torvalds 已提交
659

660
	match = try_then_request_module(xt_find_match(AF_INET6, m->u.user.name,
661
						      m->u.user.revision),
662 663
					"ip6t_%s", m->u.user.name);
	if (IS_ERR(match) || !match) {
664
		duprintf("find_check_match: `%s' not found\n", m->u.user.name);
665
		return match ? PTR_ERR(match) : -ENOENT;
L
Linus Torvalds 已提交
666 667 668
	}
	m->u.kernel.match = match;

669
	ret = check_match(m, par, i);
670 671 672
	if (ret)
		goto err;

L
Linus Torvalds 已提交
673
	return 0;
674 675 676
err:
	module_put(m->u.kernel.match->me);
	return ret;
L
Linus Torvalds 已提交
677 678
}

679
static int check_target(struct ip6t_entry *e, struct net *net, const char *name)
L
Linus Torvalds 已提交
680
{
681 682
	struct ip6t_entry_target *t = ip6t_get_target(e);
	struct xt_tgchk_param par = {
683
		.net       = net,
684 685 686 687 688
		.table     = name,
		.entryinfo = e,
		.target    = t->u.kernel.target,
		.targinfo  = t->data,
		.hook_mask = e->comefrom,
689
		.family    = NFPROTO_IPV6,
690
	};
L
Linus Torvalds 已提交
691 692
	int ret;

693
	t = ip6t_get_target(e);
694
	ret = xt_check_target(&par, t->u.target_size - sizeof(*t),
695
	      e->ipv6.proto, e->ipv6.invflags & IP6T_INV_PROTO);
696
	if (ret < 0) {
697 698
		duprintf("ip_tables: check failed for `%s'.\n",
			 t->u.kernel.target->name);
699
		return ret;
L
Linus Torvalds 已提交
700
	}
701
	return 0;
702
}
L
Linus Torvalds 已提交
703

704
static int
705 706
find_check_entry(struct ip6t_entry *e, struct net *net, const char *name,
		 unsigned int size, unsigned int *i)
707 708 709 710 711
{
	struct ip6t_entry_target *t;
	struct xt_target *target;
	int ret;
	unsigned int j;
712
	struct xt_mtchk_param mtpar;
713 714 715 716

	ret = check_entry(e, name);
	if (ret)
		return ret;
717

L
Linus Torvalds 已提交
718
	j = 0;
719
	mtpar.net	= net;
720 721 722
	mtpar.table     = name;
	mtpar.entryinfo = &e->ipv6;
	mtpar.hook_mask = e->comefrom;
723
	mtpar.family    = NFPROTO_IPV6;
724
	ret = IP6T_MATCH_ITERATE(e, find_check_match, &mtpar, &j);
L
Linus Torvalds 已提交
725 726 727 728
	if (ret != 0)
		goto cleanup_matches;

	t = ip6t_get_target(e);
729 730 731
	target = try_then_request_module(xt_find_target(AF_INET6,
							t->u.user.name,
							t->u.user.revision),
732 733
					 "ip6t_%s", t->u.user.name);
	if (IS_ERR(target) || !target) {
734
		duprintf("find_check_entry: `%s' not found\n", t->u.user.name);
735
		ret = target ? PTR_ERR(target) : -ENOENT;
L
Linus Torvalds 已提交
736 737 738
		goto cleanup_matches;
	}
	t->u.kernel.target = target;
739

740
	ret = check_target(e, net, name);
741 742 743
	if (ret)
		goto err;

L
Linus Torvalds 已提交
744 745
	(*i)++;
	return 0;
746 747
 err:
	module_put(t->u.kernel.target->me);
L
Linus Torvalds 已提交
748
 cleanup_matches:
749
	IP6T_MATCH_ITERATE(e, cleanup_match, net, &j);
L
Linus Torvalds 已提交
750 751 752
	return ret;
}

753 754 755 756 757 758 759 760 761 762 763 764 765 766 767
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;
}

768
static int
L
Linus Torvalds 已提交
769
check_entry_size_and_hooks(struct ip6t_entry *e,
770
			   struct xt_table_info *newinfo,
L
Linus Torvalds 已提交
771 772 773 774
			   unsigned char *base,
			   unsigned char *limit,
			   const unsigned int *hook_entries,
			   const unsigned int *underflows,
775
			   unsigned int valid_hooks,
L
Linus Torvalds 已提交
776 777 778 779
			   unsigned int *i)
{
	unsigned int h;

780 781
	if ((unsigned long)e % __alignof__(struct ip6t_entry) != 0 ||
	    (unsigned char *)e + sizeof(struct ip6t_entry) >= limit) {
L
Linus Torvalds 已提交
782 783 784 785 786 787 788 789 790 791 792 793
		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 */
794
	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
795 796
		if (!(valid_hooks & (1 << h)))
			continue;
L
Linus Torvalds 已提交
797 798
		if ((unsigned char *)e - base == hook_entries[h])
			newinfo->hook_entry[h] = hook_entries[h];
799
		if ((unsigned char *)e - base == underflows[h]) {
800 801 802 803
			if (!check_underflow(e)) {
				pr_err("Underflows must be unconditional and "
				       "use the STANDARD target with "
				       "ACCEPT/DROP\n");
804 805
				return -EINVAL;
			}
L
Linus Torvalds 已提交
806
			newinfo->underflow[h] = underflows[h];
807
		}
L
Linus Torvalds 已提交
808 809 810
	}

	/* Clear counters and comefrom */
811
	e->counters = ((struct xt_counters) { 0, 0 });
L
Linus Torvalds 已提交
812 813 814 815 816 817
	e->comefrom = 0;

	(*i)++;
	return 0;
}

818
static int
819
cleanup_entry(struct ip6t_entry *e, struct net *net, unsigned int *i)
L
Linus Torvalds 已提交
820
{
821
	struct xt_tgdtor_param par;
L
Linus Torvalds 已提交
822 823 824 825 826 827
	struct ip6t_entry_target *t;

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

	/* Cleanup all matches */
828
	IP6T_MATCH_ITERATE(e, cleanup_match, net, NULL);
L
Linus Torvalds 已提交
829
	t = ip6t_get_target(e);
830

831
	par.net      = net;
832 833
	par.target   = t->u.kernel.target;
	par.targinfo = t->data;
834
	par.family   = NFPROTO_IPV6;
835 836 837
	if (par.target->destroy != NULL)
		par.target->destroy(&par);
	module_put(par.target->me);
L
Linus Torvalds 已提交
838 839 840 841 842 843
	return 0;
}

/* Checks and translates the user-supplied table segment (held in
   newinfo) */
static int
844 845
translate_table(struct net *net,
		const char *name,
L
Linus Torvalds 已提交
846
		unsigned int valid_hooks,
847
		struct xt_table_info *newinfo,
848
		void *entry0,
L
Linus Torvalds 已提交
849 850 851 852 853 854 855 856 857 858 859 860
		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. */
861
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
862 863 864 865 866 867 868
		newinfo->hook_entry[i] = 0xFFFFFFFF;
		newinfo->underflow[i] = 0xFFFFFFFF;
	}

	duprintf("translate_table: size %u\n", newinfo->size);
	i = 0;
	/* Walk through entries, checking offsets. */
869
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
L
Linus Torvalds 已提交
870 871
				check_entry_size_and_hooks,
				newinfo,
872 873
				entry0,
				entry0 + size,
874
				hook_entries, underflows, valid_hooks, &i);
L
Linus Torvalds 已提交
875 876 877 878 879 880 881 882 883 884
	if (ret != 0)
		return ret;

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

	/* Check hooks all assigned */
885
	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
L
Linus Torvalds 已提交
886 887 888 889 890 891 892 893 894 895 896 897 898 899 900
		/* 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;
		}
	}

901 902 903
	if (!mark_source_chains(newinfo, valid_hooks, entry0))
		return -ELOOP;

L
Linus Torvalds 已提交
904 905
	/* Finally, each sanity check must pass */
	i = 0;
906
	ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size,
907
				find_check_entry, net, name, size, &i);
L
Linus Torvalds 已提交
908

909 910
	if (ret != 0) {
		IP6T_ENTRY_ITERATE(entry0, newinfo->size,
911
				   cleanup_entry, net, &i);
912 913
		return ret;
	}
L
Linus Torvalds 已提交
914 915

	/* And one copy for every other CPU */
916
	for_each_possible_cpu(i) {
917 918
		if (newinfo->entries[i] && newinfo->entries[i] != entry0)
			memcpy(newinfo->entries[i], entry0, newinfo->size);
L
Linus Torvalds 已提交
919 920
	}

921
	return ret;
L
Linus Torvalds 已提交
922 923 924 925 926
}

/* Gets counters. */
static inline int
add_entry_to_counter(const struct ip6t_entry *e,
927
		     struct xt_counters total[],
L
Linus Torvalds 已提交
928 929 930 931 932 933 934 935
		     unsigned int *i)
{
	ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt);

	(*i)++;
	return 0;
}

936 937 938 939 940 941 942 943 944 945 946
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
947
static void
948 949
get_counters(const struct xt_table_info *t,
	     struct xt_counters counters[])
L
Linus Torvalds 已提交
950 951 952
{
	unsigned int cpu;
	unsigned int i;
953 954 955 956 957
	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
958 959 960
	 *
	 * Bottom half has to be disabled to prevent deadlock
	 * if new softirq were to run and call ipt_do_table
961
	 */
962 963
	local_bh_disable();
	curcpu = smp_processor_id();
964 965 966 967 968 969 970

	i = 0;
	IP6T_ENTRY_ITERATE(t->entries[curcpu],
			   t->size,
			   set_entry_to_counter,
			   counters,
			   &i);
L
Linus Torvalds 已提交
971

972
	for_each_possible_cpu(cpu) {
973 974
		if (cpu == curcpu)
			continue;
L
Linus Torvalds 已提交
975
		i = 0;
976
		xt_info_wrlock(cpu);
977
		IP6T_ENTRY_ITERATE(t->entries[cpu],
L
Linus Torvalds 已提交
978 979 980 981
				  t->size,
				  add_entry_to_counter,
				  counters,
				  &i);
982
		xt_info_wrunlock(cpu);
L
Linus Torvalds 已提交
983
	}
984 985 986
	local_bh_enable();
}

987
static struct xt_counters *alloc_counters(struct xt_table *table)
L
Linus Torvalds 已提交
988
{
989
	unsigned int countersize;
990
	struct xt_counters *counters;
991
	struct xt_table_info *private = table->private;
L
Linus Torvalds 已提交
992 993 994 995

	/* We need atomic snapshot of counters: rest doesn't change
	   (other than comefrom, which userspace doesn't care
	   about). */
996
	countersize = sizeof(struct xt_counters) * private->number;
997
	counters = vmalloc_node(countersize, numa_node_id());
L
Linus Torvalds 已提交
998 999

	if (counters == NULL)
1000
		return ERR_PTR(-ENOMEM);
1001

1002
	get_counters(private, counters);
1003

1004
	return counters;
1005 1006 1007 1008 1009 1010 1011 1012 1013 1014
}

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;
1015
	const struct xt_table_info *private = table->private;
1016
	int ret = 0;
1017
	const void *loc_cpu_entry;
1018 1019 1020 1021 1022

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

1023 1024 1025 1026
	/* choose the copy that is on our node/cpu, ...
	 * This choice is lazy (because current thread is
	 * allowed to migrate to another cpu)
	 */
1027
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
1028
	if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) {
L
Linus Torvalds 已提交
1029 1030 1031 1032 1033 1034 1035 1036
		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;
1037 1038
		const struct ip6t_entry_match *m;
		const struct ip6t_entry_target *t;
L
Linus Torvalds 已提交
1039

1040
		e = (struct ip6t_entry *)(loc_cpu_entry + off);
L
Linus Torvalds 已提交
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 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080
		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;
}

1081
#ifdef CONFIG_COMPAT
1082
static void compat_standard_from_user(void *dst, const void *src)
1083 1084 1085 1086 1087 1088 1089 1090
{
	int v = *(compat_int_t *)src;

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

1091
static int compat_standard_to_user(void __user *dst, const void *src)
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 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153
{
	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

1154
static int get_info(struct net *net, void __user *user, int *len, int compat)
1155 1156 1157 1158 1159 1160
{
	char name[IP6T_TABLE_MAXNAMELEN];
	struct xt_table *t;
	int ret;

	if (*len != sizeof(struct ip6t_getinfo)) {
1161
		duprintf("length %u != %zu\n", *len,
1162 1163 1164 1165 1166 1167 1168 1169
			 sizeof(struct ip6t_getinfo));
		return -EINVAL;
	}

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

	name[IP6T_TABLE_MAXNAMELEN-1] = '\0';
1170 1171 1172 1173
#ifdef CONFIG_COMPAT
	if (compat)
		xt_compat_lock(AF_INET6);
#endif
1174
	t = try_then_request_module(xt_find_table_lock(net, AF_INET6, name),
1175 1176 1177
				    "ip6table_%s", name);
	if (t && !IS_ERR(t)) {
		struct ip6t_getinfo info;
1178
		const struct xt_table_info *private = t->private;
1179
#ifdef CONFIG_COMPAT
1180 1181
		struct xt_table_info tmp;

1182 1183 1184 1185 1186 1187
		if (compat) {
			ret = compat_table_info(private, &tmp);
			xt_compat_flush_offsets(AF_INET6);
			private = &tmp;
		}
#endif
1188 1189 1190 1191 1192 1193 1194
		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;
1195
		strcpy(info.name, name);
1196 1197 1198 1199 1200 1201 1202 1203 1204 1205

		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;
1206 1207 1208 1209
#ifdef CONFIG_COMPAT
	if (compat)
		xt_compat_unlock(AF_INET6);
#endif
1210 1211 1212
	return ret;
}

L
Linus Torvalds 已提交
1213
static int
1214
get_entries(struct net *net, struct ip6t_get_entries __user *uptr, int *len)
L
Linus Torvalds 已提交
1215 1216
{
	int ret;
1217
	struct ip6t_get_entries get;
1218
	struct xt_table *t;
L
Linus Torvalds 已提交
1219

1220
	if (*len < sizeof(get)) {
1221
		duprintf("get_entries: %u < %zu\n", *len, sizeof(get));
1222 1223 1224 1225 1226
		return -EINVAL;
	}
	if (copy_from_user(&get, uptr, sizeof(get)) != 0)
		return -EFAULT;
	if (*len != sizeof(struct ip6t_get_entries) + get.size) {
1227 1228
		duprintf("get_entries: %u != %zu\n",
			 *len, sizeof(get) + get.size);
1229 1230 1231
		return -EINVAL;
	}

1232
	t = xt_find_table_lock(net, AF_INET6, get.name);
1233
	if (t && !IS_ERR(t)) {
1234 1235
		struct xt_table_info *private = t->private;
		duprintf("t->private->number = %u\n", private->number);
1236
		if (get.size == private->size)
1237
			ret = copy_entries_to_user(private->size,
L
Linus Torvalds 已提交
1238 1239 1240
						   t, uptr->entrytable);
		else {
			duprintf("get_entries: I've got %u not %u!\n",
1241
				 private->size, get.size);
1242
			ret = -EAGAIN;
L
Linus Torvalds 已提交
1243
		}
1244
		module_put(t->me);
1245
		xt_table_unlock(t);
L
Linus Torvalds 已提交
1246
	} else
1247
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1248 1249 1250 1251 1252

	return ret;
}

static int
1253
__do_replace(struct net *net, const char *name, unsigned int valid_hooks,
1254 1255
	     struct xt_table_info *newinfo, unsigned int num_counters,
	     void __user *counters_ptr)
L
Linus Torvalds 已提交
1256 1257
{
	int ret;
1258
	struct xt_table *t;
1259
	struct xt_table_info *oldinfo;
1260
	struct xt_counters *counters;
1261
	const void *loc_cpu_old_entry;
L
Linus Torvalds 已提交
1262

1263 1264
	ret = 0;
	counters = vmalloc_node(num_counters * sizeof(struct xt_counters),
1265
				numa_node_id());
L
Linus Torvalds 已提交
1266 1267
	if (!counters) {
		ret = -ENOMEM;
1268
		goto out;
L
Linus Torvalds 已提交
1269 1270
	}

1271
	t = try_then_request_module(xt_find_table_lock(net, AF_INET6, name),
1272
				    "ip6table_%s", name);
1273 1274
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1275
		goto free_newinfo_counters_untrans;
1276
	}
L
Linus Torvalds 已提交
1277 1278

	/* You lied! */
1279
	if (valid_hooks != t->valid_hooks) {
L
Linus Torvalds 已提交
1280
		duprintf("Valid hook crap: %08X vs %08X\n",
1281
			 valid_hooks, t->valid_hooks);
L
Linus Torvalds 已提交
1282
		ret = -EINVAL;
1283
		goto put_module;
L
Linus Torvalds 已提交
1284 1285
	}

1286
	oldinfo = xt_replace_table(t, num_counters, newinfo, &ret);
L
Linus Torvalds 已提交
1287 1288 1289 1290 1291 1292
	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);
1293 1294
	if ((oldinfo->number > oldinfo->initial_entries) ||
	    (newinfo->number <= oldinfo->initial_entries))
L
Linus Torvalds 已提交
1295 1296 1297 1298 1299
		module_put(t->me);
	if ((oldinfo->number > oldinfo->initial_entries) &&
	    (newinfo->number <= oldinfo->initial_entries))
		module_put(t->me);

1300
	/* Get the old counters, and synchronize with replace */
L
Linus Torvalds 已提交
1301
	get_counters(oldinfo, counters);
1302

L
Linus Torvalds 已提交
1303
	/* Decrease module usage counts and free resource */
1304
	loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()];
1305
	IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,
1306
			   net, NULL);
1307
	xt_free_table_info(oldinfo);
1308 1309
	if (copy_to_user(counters_ptr, counters,
			 sizeof(struct xt_counters) * num_counters) != 0)
L
Linus Torvalds 已提交
1310 1311
		ret = -EFAULT;
	vfree(counters);
1312
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1313 1314 1315 1316
	return ret;

 put_module:
	module_put(t->me);
1317
	xt_table_unlock(t);
L
Linus Torvalds 已提交
1318 1319
 free_newinfo_counters_untrans:
	vfree(counters);
1320 1321 1322 1323 1324
 out:
	return ret;
}

static int
1325
do_replace(struct net *net, void __user *user, unsigned int len)
1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350
{
	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;
	}

1351
	ret = translate_table(net, tmp.name, tmp.valid_hooks,
1352 1353 1354 1355 1356 1357 1358
			      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");

1359
	ret = __do_replace(net, tmp.name, tmp.valid_hooks, newinfo,
1360 1361 1362 1363 1364 1365
			   tmp.num_counters, tmp.counters);
	if (ret)
		goto free_newinfo_untrans;
	return 0;

 free_newinfo_untrans:
1366
	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL);
L
Linus Torvalds 已提交
1367
 free_newinfo:
1368
	xt_free_table_info(newinfo);
L
Linus Torvalds 已提交
1369 1370 1371
	return ret;
}

1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384
/* 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
1385
static int
1386 1387
do_add_counters(struct net *net, void __user *user, unsigned int len,
		int compat)
L
Linus Torvalds 已提交
1388
{
1389
	unsigned int i, curcpu;
1390 1391 1392 1393 1394 1395
	struct xt_counters_info tmp;
	struct xt_counters *paddc;
	unsigned int num_counters;
	char *name;
	int size;
	void *ptmp;
1396
	struct xt_table *t;
1397
	const struct xt_table_info *private;
1398
	int ret = 0;
1399
	const void *loc_cpu_entry;
1400 1401
#ifdef CONFIG_COMPAT
	struct compat_xt_counters_info compat_tmp;
L
Linus Torvalds 已提交
1402

1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413
	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
1414 1415
		return -EFAULT;

1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427
#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
1428 1429
		return -EINVAL;

1430
	paddc = vmalloc_node(len - size, numa_node_id());
L
Linus Torvalds 已提交
1431 1432 1433
	if (!paddc)
		return -ENOMEM;

1434
	if (copy_from_user(paddc, user + size, len - size) != 0) {
L
Linus Torvalds 已提交
1435 1436 1437 1438
		ret = -EFAULT;
		goto free;
	}

1439
	t = xt_find_table_lock(net, AF_INET6, name);
1440 1441
	if (!t || IS_ERR(t)) {
		ret = t ? PTR_ERR(t) : -ENOENT;
L
Linus Torvalds 已提交
1442
		goto free;
1443
	}
L
Linus Torvalds 已提交
1444

1445 1446

	local_bh_disable();
1447
	private = t->private;
1448
	if (private->number != num_counters) {
L
Linus Torvalds 已提交
1449 1450 1451 1452 1453
		ret = -EINVAL;
		goto unlock_up_free;
	}

	i = 0;
1454
	/* Choose the copy that is on our node */
1455 1456 1457
	curcpu = smp_processor_id();
	xt_info_wrlock(curcpu);
	loc_cpu_entry = private->entries[curcpu];
1458
	IP6T_ENTRY_ITERATE(loc_cpu_entry,
1459
			  private->size,
L
Linus Torvalds 已提交
1460
			  add_counter_to_entry,
1461
			  paddc,
L
Linus Torvalds 已提交
1462
			  &i);
1463 1464
	xt_info_wrunlock(curcpu);

L
Linus Torvalds 已提交
1465
 unlock_up_free:
1466
	local_bh_enable();
1467
	xt_table_unlock(t);
1468
	module_put(t->me);
L
Linus Torvalds 已提交
1469 1470 1471 1472 1473 1474
 free:
	vfree(paddc);

	return ret;
}

1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489
#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,
1490
			  unsigned int *size, struct xt_counters *counters,
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 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531
			  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;
}

1532
static int
1533 1534 1535 1536
compat_find_calc_match(struct ip6t_entry_match *m,
		       const char *name,
		       const struct ip6t_ip6 *ipv6,
		       unsigned int hookmask,
1537
		       int *size, unsigned int *i)
1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555
{
	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;
}

1556
static int
1557 1558 1559 1560 1561 1562 1563 1564 1565
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;
}

1566
static int
1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580
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;
}

1581
static int
1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594
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;
1595 1596
	unsigned int j;
	int ret, off, h;
1597 1598

	duprintf("check_compat_entry_size_and_hooks %p\n", e);
1599 1600
	if ((unsigned long)e % __alignof__(struct compat_ip6t_entry) != 0 ||
	    (unsigned char *)e + sizeof(struct compat_ip6t_entry) >= limit) {
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 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704
		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;
}

1705 1706
static int compat_check_entry(struct ip6t_entry *e, struct net *net,
			      const char *name, unsigned int *i)
1707
{
1708 1709
	unsigned int j;
	int ret;
1710
	struct xt_mtchk_param mtpar;
1711 1712

	j = 0;
1713
	mtpar.net	= net;
1714 1715 1716
	mtpar.table     = name;
	mtpar.entryinfo = &e->ipv6;
	mtpar.hook_mask = e->comefrom;
1717
	mtpar.family    = NFPROTO_IPV6;
1718
	ret = IP6T_MATCH_ITERATE(e, check_match, &mtpar, &j);
1719 1720 1721
	if (ret)
		goto cleanup_matches;

1722
	ret = check_target(e, net, name);
1723 1724 1725 1726 1727 1728 1729
	if (ret)
		goto cleanup_matches;

	(*i)++;
	return 0;

 cleanup_matches:
1730
	IP6T_MATCH_ITERATE(e, cleanup_match, net, &j);
1731 1732 1733 1734
	return ret;
}

static int
1735 1736
translate_compat_table(struct net *net,
		       const char *name,
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
		       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,
1825
				 net, name, &i);
1826 1827 1828 1829
	if (ret) {
		j -= i;
		COMPAT_IP6T_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i,
						   compat_release_entry, &j);
1830
		IP6T_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, net, &i);
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
		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
1857
compat_do_replace(struct net *net, void __user *user, unsigned int len)
1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876
{
	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;

1877
	/* choose the copy that is on our node/cpu */
1878 1879 1880 1881 1882 1883 1884
	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;
	}

1885
	ret = translate_compat_table(net, tmp.name, tmp.valid_hooks,
1886 1887 1888 1889 1890 1891 1892 1893
				     &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");

1894
	ret = __do_replace(net, tmp.name, tmp.valid_hooks, newinfo,
1895 1896 1897 1898 1899 1900
			   tmp.num_counters, compat_ptr(tmp.counters));
	if (ret)
		goto free_newinfo_untrans;
	return 0;

 free_newinfo_untrans:
1901
	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL);
1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917
 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:
1918
		ret = compat_do_replace(sock_net(sk), user, len);
1919 1920 1921
		break;

	case IP6T_SO_SET_ADD_COUNTERS:
1922
		ret = do_add_counters(sock_net(sk), user, len, 1);
1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943
		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;
1944
	const struct xt_table_info *private = table->private;
1945 1946 1947
	void __user *pos;
	unsigned int size;
	int ret = 0;
1948
	const void *loc_cpu_entry;
1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970
	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
1971 1972
compat_get_entries(struct net *net, struct compat_ip6t_get_entries __user *uptr,
		   int *len)
1973 1974 1975 1976 1977 1978
{
	int ret;
	struct compat_ip6t_get_entries get;
	struct xt_table *t;

	if (*len < sizeof(get)) {
1979
		duprintf("compat_get_entries: %u < %zu\n", *len, sizeof(get));
1980 1981 1982 1983 1984 1985 1986
		return -EINVAL;
	}

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

	if (*len != sizeof(struct compat_ip6t_get_entries) + get.size) {
1987 1988
		duprintf("compat_get_entries: %u != %zu\n",
			 *len, sizeof(get) + get.size);
1989 1990 1991 1992
		return -EINVAL;
	}

	xt_compat_lock(AF_INET6);
1993
	t = xt_find_table_lock(net, AF_INET6, get.name);
1994
	if (t && !IS_ERR(t)) {
1995
		const struct xt_table_info *private = t->private;
1996
		struct xt_table_info info;
1997
		duprintf("t->private->number = %u\n", private->number);
1998 1999 2000 2001 2002 2003
		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",
2004
				 private->size, get.size);
2005
			ret = -EAGAIN;
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028
		}
		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:
2029
		ret = get_info(sock_net(sk), user, len, 1);
2030 2031
		break;
	case IP6T_SO_GET_ENTRIES:
2032
		ret = compat_get_entries(sock_net(sk), user, len);
2033 2034 2035 2036 2037 2038 2039 2040
		break;
	default:
		ret = do_ip6t_get_ctl(sk, cmd, user, len);
	}
	return ret;
}
#endif

L
Linus Torvalds 已提交
2041 2042 2043 2044 2045 2046 2047 2048 2049 2050
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:
2051
		ret = do_replace(sock_net(sk), user, len);
L
Linus Torvalds 已提交
2052 2053 2054
		break;

	case IP6T_SO_SET_ADD_COUNTERS:
2055
		ret = do_add_counters(sock_net(sk), user, len, 0);
L
Linus Torvalds 已提交
2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074
		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) {
2075
	case IP6T_SO_GET_INFO:
2076
		ret = get_info(sock_net(sk), user, len, 0);
2077
		break;
L
Linus Torvalds 已提交
2078

2079
	case IP6T_SO_GET_ENTRIES:
2080
		ret = get_entries(sock_net(sk), user, len);
L
Linus Torvalds 已提交
2081 2082
		break;

2083 2084 2085
	case IP6T_SO_GET_REVISION_MATCH:
	case IP6T_SO_GET_REVISION_TARGET: {
		struct ip6t_get_revision rev;
2086
		int target;
2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097

		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)
2098
			target = 1;
2099
		else
2100
			target = 0;
2101

2102 2103 2104
		try_then_request_module(xt_find_revision(AF_INET6, rev.name,
							 rev.revision,
							 target, &ret),
2105 2106 2107 2108
					"ip6t_%s", rev.name);
		break;
	}

L
Linus Torvalds 已提交
2109 2110 2111 2112 2113 2114 2115 2116
	default:
		duprintf("do_ip6t_get_ctl: unknown request %i\n", cmd);
		ret = -EINVAL;
	}

	return ret;
}

2117 2118
struct xt_table *ip6t_register_table(struct net *net,
				     const struct xt_table *table,
2119
				     const struct ip6t_replace *repl)
L
Linus Torvalds 已提交
2120 2121
{
	int ret;
2122
	struct xt_table_info *newinfo;
2123
	struct xt_table_info bootstrap
L
Linus Torvalds 已提交
2124
		= { 0, 0, 0, { 0 }, { 0 }, { } };
2125
	void *loc_cpu_entry;
2126
	struct xt_table *new_table;
L
Linus Torvalds 已提交
2127

2128
	newinfo = xt_alloc_table_info(repl->size);
2129 2130 2131 2132
	if (!newinfo) {
		ret = -ENOMEM;
		goto out;
	}
L
Linus Torvalds 已提交
2133

2134
	/* choose the copy on our node/cpu, but dont care about preemption */
2135 2136
	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
	memcpy(loc_cpu_entry, repl->entries, repl->size);
L
Linus Torvalds 已提交
2137

2138
	ret = translate_table(net, table->name, table->valid_hooks,
2139
			      newinfo, loc_cpu_entry, repl->size,
L
Linus Torvalds 已提交
2140 2141 2142
			      repl->num_entries,
			      repl->hook_entry,
			      repl->underflow);
2143 2144
	if (ret != 0)
		goto out_free;
L
Linus Torvalds 已提交
2145

2146
	new_table = xt_register_table(net, table, &bootstrap, newinfo);
2147
	if (IS_ERR(new_table)) {
2148 2149
		ret = PTR_ERR(new_table);
		goto out_free;
L
Linus Torvalds 已提交
2150
	}
2151
	return new_table;
L
Linus Torvalds 已提交
2152

2153 2154 2155 2156
out_free:
	xt_free_table_info(newinfo);
out:
	return ERR_PTR(ret);
L
Linus Torvalds 已提交
2157 2158
}

2159
void ip6t_unregister_table(struct net *net, struct xt_table *table)
L
Linus Torvalds 已提交
2160
{
2161
	struct xt_table_info *private;
2162
	void *loc_cpu_entry;
2163
	struct module *table_owner = table->me;
2164

2165
	private = xt_unregister_table(table);
L
Linus Torvalds 已提交
2166 2167

	/* Decrease module usage counts and free resources */
2168
	loc_cpu_entry = private->entries[raw_smp_processor_id()];
2169
	IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, net, NULL);
2170 2171
	if (private->number > private->initial_entries)
		module_put(table_owner);
2172
	xt_free_table_info(private);
L
Linus Torvalds 已提交
2173 2174 2175
}

/* Returns 1 if the type and code is matched by the range, 0 otherwise */
2176
static inline bool
L
Linus Torvalds 已提交
2177 2178
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,
2179
		     bool invert)
L
Linus Torvalds 已提交
2180 2181 2182 2183 2184
{
	return (type == test_type && code >= min_code && code <= max_code)
		^ invert;
}

2185
static bool
2186
icmp6_match(const struct sk_buff *skb, const struct xt_match_param *par)
L
Linus Torvalds 已提交
2187
{
2188 2189
	const struct icmp6hdr *ic;
	struct icmp6hdr _icmph;
2190
	const struct ip6t_icmp *icmpinfo = par->matchinfo;
L
Linus Torvalds 已提交
2191 2192

	/* Must not be a fragment. */
2193
	if (par->fragoff != 0)
2194
		return false;
L
Linus Torvalds 已提交
2195

2196
	ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph);
L
Linus Torvalds 已提交
2197 2198
	if (ic == NULL) {
		/* We've been asked to examine this packet, and we
2199 2200
		 * can't.  Hence, no choice but to drop.
		 */
L
Linus Torvalds 已提交
2201
		duprintf("Dropping evil ICMP tinygram.\n");
2202
		*par->hotdrop = true;
2203
		return false;
L
Linus Torvalds 已提交
2204 2205 2206 2207 2208 2209 2210 2211 2212 2213
	}

	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. */
2214
static bool icmp6_checkentry(const struct xt_mtchk_param *par)
L
Linus Torvalds 已提交
2215
{
2216
	const struct ip6t_icmp *icmpinfo = par->matchinfo;
L
Linus Torvalds 已提交
2217

2218 2219
	/* Must specify no unknown invflags */
	return !(icmpinfo->invflags & ~IP6T_ICMP_INV);
L
Linus Torvalds 已提交
2220 2221 2222
}

/* The built-in targets: standard (NULL) and error. */
2223
static struct xt_target ip6t_standard_target __read_mostly = {
L
Linus Torvalds 已提交
2224
	.name		= IP6T_STANDARD_TARGET,
2225
	.targetsize	= sizeof(int),
2226
	.family		= NFPROTO_IPV6,
2227 2228 2229 2230 2231
#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
2232 2233
};

2234
static struct xt_target ip6t_error_target __read_mostly = {
L
Linus Torvalds 已提交
2235 2236
	.name		= IP6T_ERROR_TARGET,
	.target		= ip6t_error,
2237
	.targetsize	= IP6T_FUNCTION_MAXNAMELEN,
2238
	.family		= NFPROTO_IPV6,
L
Linus Torvalds 已提交
2239 2240 2241 2242 2243 2244 2245
};

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,
2246 2247 2248
#ifdef CONFIG_COMPAT
	.compat_set	= compat_do_ip6t_set_ctl,
#endif
L
Linus Torvalds 已提交
2249 2250 2251
	.get_optmin	= IP6T_BASE_CTL,
	.get_optmax	= IP6T_SO_GET_MAX+1,
	.get		= do_ip6t_get_ctl,
2252 2253 2254
#ifdef CONFIG_COMPAT
	.compat_get	= compat_do_ip6t_get_ctl,
#endif
2255
	.owner		= THIS_MODULE,
L
Linus Torvalds 已提交
2256 2257
};

2258
static struct xt_match icmp6_matchstruct __read_mostly = {
L
Linus Torvalds 已提交
2259
	.name		= "icmp6",
2260
	.match		= icmp6_match,
2261 2262 2263
	.matchsize	= sizeof(struct ip6t_icmp),
	.checkentry	= icmp6_checkentry,
	.proto		= IPPROTO_ICMPV6,
2264
	.family		= NFPROTO_IPV6,
L
Linus Torvalds 已提交
2265 2266
};

2267 2268
static int __net_init ip6_tables_net_init(struct net *net)
{
2269
	return xt_proto_init(net, NFPROTO_IPV6);
2270 2271 2272 2273
}

static void __net_exit ip6_tables_net_exit(struct net *net)
{
2274
	xt_proto_fini(net, NFPROTO_IPV6);
2275 2276 2277 2278 2279 2280 2281
}

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

2282
static int __init ip6_tables_init(void)
L
Linus Torvalds 已提交
2283 2284 2285
{
	int ret;

2286
	ret = register_pernet_subsys(&ip6_tables_net_ops);
2287 2288
	if (ret < 0)
		goto err1;
2289

L
Linus Torvalds 已提交
2290
	/* Noone else will be downing sem now, so we won't sleep */
2291 2292 2293 2294 2295 2296 2297 2298 2299
	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
2300 2301 2302

	/* Register setsockopt */
	ret = nf_register_sockopt(&ip6t_sockopts);
2303 2304
	if (ret < 0)
		goto err5;
L
Linus Torvalds 已提交
2305

2306
	printk(KERN_INFO "ip6_tables: (C) 2000-2006 Netfilter Core Team\n");
L
Linus Torvalds 已提交
2307
	return 0;
2308 2309 2310 2311 2312 2313 2314 2315

err5:
	xt_unregister_match(&icmp6_matchstruct);
err4:
	xt_unregister_target(&ip6t_error_target);
err3:
	xt_unregister_target(&ip6t_standard_target);
err2:
2316
	unregister_pernet_subsys(&ip6_tables_net_ops);
2317 2318
err1:
	return ret;
L
Linus Torvalds 已提交
2319 2320
}

2321
static void __exit ip6_tables_fini(void)
L
Linus Torvalds 已提交
2322 2323
{
	nf_unregister_sockopt(&ip6t_sockopts);
2324

2325 2326 2327
	xt_unregister_match(&icmp6_matchstruct);
	xt_unregister_target(&ip6t_error_target);
	xt_unregister_target(&ip6t_standard_target);
2328 2329

	unregister_pernet_subsys(&ip6_tables_net_ops);
L
Linus Torvalds 已提交
2330 2331
}

2332
/*
2333 2334 2335 2336 2337 2338 2339
 * 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.
 *
2340 2341 2342
 * If the first fragment doesn't contain the final protocol header or
 * NEXTHDR_NONE it is considered invalid.
 *
2343 2344 2345 2346
 * 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.
2347 2348
 *
 */
2349 2350
int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
		  int target, unsigned short *fragoff)
2351
{
2352
	unsigned int start = skb_network_offset(skb) + sizeof(struct ipv6hdr);
2353
	u8 nexthdr = ipv6_hdr(skb)->nexthdr;
2354 2355
	unsigned int len = skb->len - start;

2356 2357 2358
	if (fragoff)
		*fragoff = 0;

2359 2360 2361 2362
	while (nexthdr != target) {
		struct ipv6_opt_hdr _hdr, *hp;
		unsigned int hdrlen;

2363 2364 2365
		if ((!ipv6_ext_hdr(nexthdr)) || nexthdr == NEXTHDR_NONE) {
			if (target < 0)
				break;
2366
			return -ENOENT;
2367 2368
		}

2369 2370
		hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr);
		if (hp == NULL)
2371
			return -EBADMSG;
2372
		if (nexthdr == NEXTHDR_FRAGMENT) {
A
Al Viro 已提交
2373 2374
			unsigned short _frag_off;
			__be16 *fp;
2375 2376 2377 2378 2379 2380
			fp = skb_header_pointer(skb,
						start+offsetof(struct frag_hdr,
							       frag_off),
						sizeof(_frag_off),
						&_frag_off);
			if (fp == NULL)
2381
				return -EBADMSG;
2382

2383 2384 2385 2386
			_frag_off = ntohs(*fp) & ~0x7;
			if (_frag_off) {
				if (target < 0 &&
				    ((!ipv6_ext_hdr(hp->nexthdr)) ||
2387
				     hp->nexthdr == NEXTHDR_NONE)) {
2388 2389 2390 2391
					if (fragoff)
						*fragoff = _frag_off;
					return hp->nexthdr;
				}
2392
				return -ENOENT;
2393
			}
2394 2395
			hdrlen = 8;
		} else if (nexthdr == NEXTHDR_AUTH)
2396
			hdrlen = (hp->hdrlen + 2) << 2;
2397
		else
2398
			hdrlen = ipv6_optlen(hp);
2399 2400 2401 2402 2403 2404 2405

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

	*offset = start;
2406
	return nexthdr;
2407 2408
}

L
Linus Torvalds 已提交
2409 2410 2411 2412
EXPORT_SYMBOL(ip6t_register_table);
EXPORT_SYMBOL(ip6t_unregister_table);
EXPORT_SYMBOL(ip6t_do_table);
EXPORT_SYMBOL(ip6t_ext_hdr);
2413
EXPORT_SYMBOL(ipv6_find_hdr);
L
Linus Torvalds 已提交
2414

2415 2416
module_init(ip6_tables_init);
module_exit(ip6_tables_fini);