X-Git-Url: http://git.onelab.eu/?p=linux-2.6.git;a=blobdiff_plain;f=net%2Fnetfilter%2Fnf_conntrack_core.c;fp=net%2Fnetfilter%2Fnf_conntrack_core.c;h=cb29ba7e206c7aa9d11a7cadfa526dc405f5811c;hp=d622ddf08bb05d2cc23a2f6c5672d66edadb23aa;hb=97bf2856c6014879bd04983a3e9dfcdac1e7fe85;hpb=76828883507a47dae78837ab5dec5a5b4513c667 diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index d622ddf08..cb29ba7e2 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -3,7 +3,7 @@ extension. */ /* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2005 Netfilter Core Team + * (C) 2002-2006 Netfilter Core Team * (C) 2003,2004 USAGI/WIDE Project * * This program is free software; you can redistribute it and/or modify @@ -20,11 +20,15 @@ * - generalize L3 protocol denendent part. * 23 Mar 2004: Yasuyuki Kozakai @USAGI * - add support various size of conntrack structures. + * 26 Jan 2006: Harald Welte + * - restructure nf_conn (introduce nf_conn_help) + * - redesign 'features' how they were originally intended + * 26 Feb 2006: Pablo Neira Ayuso + * - add support for L3 protocol module load on demand. * * Derived from net/ipv4/netfilter/ip_conntrack_core.c */ -#include #include #include #include @@ -42,20 +46,16 @@ #include #include #include - -/* This rwlock protects the main hash table, protocol/helper/expected - registrations, conntrack timers*/ -#define ASSERT_READ_LOCK(x) -#define ASSERT_WRITE_LOCK(x) +#include #include #include -#include +#include +#include #include #include -#include -#define NF_CONNTRACK_VERSION "0.4.1" +#define NF_CONNTRACK_VERSION "0.5.0" #if 0 #define DEBUGP printk @@ -64,92 +64,32 @@ #endif DEFINE_RWLOCK(nf_conntrack_lock); +EXPORT_SYMBOL_GPL(nf_conntrack_lock); /* nf_conntrack_standalone needs this */ atomic_t nf_conntrack_count = ATOMIC_INIT(0); +EXPORT_SYMBOL_GPL(nf_conntrack_count); -void (*nf_conntrack_destroyed)(struct nf_conn *conntrack) = NULL; -LIST_HEAD(nf_conntrack_expect_list); -struct nf_conntrack_protocol **nf_ct_protos[PF_MAX]; -struct nf_conntrack_l3proto *nf_ct_l3protos[PF_MAX]; -static LIST_HEAD(helpers); -unsigned int nf_conntrack_htable_size = 0; -int nf_conntrack_max; -struct list_head *nf_conntrack_hash; -static kmem_cache_t *nf_conntrack_expect_cachep; -struct nf_conn nf_conntrack_untracked; -unsigned int nf_ct_log_invalid; -static LIST_HEAD(unconfirmed); -static int nf_conntrack_vmalloc; - -static unsigned int nf_conntrack_next_id = 1; -static unsigned int nf_conntrack_expect_next_id = 1; -#ifdef CONFIG_NF_CONNTRACK_EVENTS -struct notifier_block *nf_conntrack_chain; -struct notifier_block *nf_conntrack_expect_chain; - -DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); - -/* deliver cached events and clear cache entry - must be called with locally - * disabled softirqs */ -static inline void -__nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) -{ - DEBUGP("ecache: delivering events for %p\n", ecache->ct); - if (nf_ct_is_confirmed(ecache->ct) && !nf_ct_is_dying(ecache->ct) - && ecache->events) - notifier_call_chain(&nf_conntrack_chain, ecache->events, - ecache->ct); - - ecache->events = 0; - nf_ct_put(ecache->ct); - ecache->ct = NULL; -} +void (*nf_conntrack_destroyed)(struct nf_conn *conntrack); +EXPORT_SYMBOL_GPL(nf_conntrack_destroyed); -/* Deliver all cached events for a particular conntrack. This is called - * by code prior to async packet handling for freeing the skb */ -void nf_ct_deliver_cached_events(const struct nf_conn *ct) -{ - struct nf_conntrack_ecache *ecache; +unsigned int nf_conntrack_htable_size __read_mostly; +EXPORT_SYMBOL_GPL(nf_conntrack_htable_size); - local_bh_disable(); - ecache = &__get_cpu_var(nf_conntrack_ecache); - if (ecache->ct == ct) - __nf_ct_deliver_cached_events(ecache); - local_bh_enable(); -} +int nf_conntrack_max __read_mostly; +EXPORT_SYMBOL_GPL(nf_conntrack_max); -/* Deliver cached events for old pending events, if current conntrack != old */ -void __nf_ct_event_cache_init(struct nf_conn *ct) -{ - struct nf_conntrack_ecache *ecache; - - /* take care of delivering potentially old events */ - ecache = &__get_cpu_var(nf_conntrack_ecache); - BUG_ON(ecache->ct == ct); - if (ecache->ct) - __nf_ct_deliver_cached_events(ecache); - /* initialize for this conntrack/packet */ - ecache->ct = ct; - nf_conntrack_get(&ct->ct_general); -} +struct list_head *nf_conntrack_hash __read_mostly; +EXPORT_SYMBOL_GPL(nf_conntrack_hash); -/* flush the event cache - touches other CPU's data and must not be called - * while packets are still passing through the code */ -static void nf_ct_event_cache_flush(void) -{ - struct nf_conntrack_ecache *ecache; - int cpu; +struct nf_conn nf_conntrack_untracked __read_mostly; +EXPORT_SYMBOL_GPL(nf_conntrack_untracked); - for_each_cpu(cpu) { - ecache = &per_cpu(nf_conntrack_ecache, cpu); - if (ecache->ct) - nf_ct_put(ecache->ct); - } -} -#else -static inline void nf_ct_event_cache_flush(void) {} -#endif /* CONFIG_NF_CONNTRACK_EVENTS */ +unsigned int nf_ct_log_invalid __read_mostly; +LIST_HEAD(unconfirmed); +static int nf_conntrack_vmalloc __read_mostly; + +static unsigned int nf_conntrack_next_id; DEFINE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); EXPORT_PER_CPU_SYMBOL(nf_conntrack_stat); @@ -168,75 +108,18 @@ static struct { size_t size; /* slab cache pointer */ - kmem_cache_t *cachep; + struct kmem_cache *cachep; /* allocated slab cache + modules which uses this slab cache */ int use; - /* Initialization */ - int (*init_conntrack)(struct nf_conn *, u_int32_t); - } nf_ct_cache[NF_CT_F_NUM]; /* protect members of nf_ct_cache except of "use" */ DEFINE_RWLOCK(nf_ct_cache_lock); /* This avoids calling kmem_cache_create() with same name simultaneously */ -DECLARE_MUTEX(nf_ct_cache_mutex); - -extern struct nf_conntrack_protocol nf_conntrack_generic_protocol; -struct nf_conntrack_protocol * -__nf_ct_proto_find(u_int16_t l3proto, u_int8_t protocol) -{ - if (unlikely(l3proto >= AF_MAX || nf_ct_protos[l3proto] == NULL)) - return &nf_conntrack_generic_protocol; - - return nf_ct_protos[l3proto][protocol]; -} - -/* this is guaranteed to always return a valid protocol helper, since - * it falls back to generic_protocol */ -struct nf_conntrack_protocol * -nf_ct_proto_find_get(u_int16_t l3proto, u_int8_t protocol) -{ - struct nf_conntrack_protocol *p; - - preempt_disable(); - p = __nf_ct_proto_find(l3proto, protocol); - if (p) { - if (!try_module_get(p->me)) - p = &nf_conntrack_generic_protocol; - } - preempt_enable(); - - return p; -} - -void nf_ct_proto_put(struct nf_conntrack_protocol *p) -{ - module_put(p->me); -} - -struct nf_conntrack_l3proto * -nf_ct_l3proto_find_get(u_int16_t l3proto) -{ - struct nf_conntrack_l3proto *p; - - preempt_disable(); - p = __nf_ct_l3proto_find(l3proto); - if (p) { - if (!try_module_get(p->me)) - p = &nf_conntrack_generic_l3proto; - } - preempt_enable(); - - return p; -} - -void nf_ct_l3proto_put(struct nf_conntrack_l3proto *p) -{ - module_put(p->me); -} +static DEFINE_MUTEX(nf_ct_cache_mutex); static int nf_conntrack_hash_rnd_initted; static unsigned int nf_conntrack_hash_rnd; @@ -259,25 +142,12 @@ static inline u_int32_t hash_conntrack(const struct nf_conntrack_tuple *tuple) nf_conntrack_hash_rnd); } -/* Initialize "struct nf_conn" which has spaces for helper */ -static int -init_conntrack_for_helper(struct nf_conn *conntrack, u_int32_t features) -{ - - conntrack->help = (union nf_conntrack_help *) - (((unsigned long)conntrack->data - + (__alignof__(union nf_conntrack_help) - 1)) - & (~((unsigned long)(__alignof__(union nf_conntrack_help) -1)))); - return 0; -} - int nf_conntrack_register_cache(u_int32_t features, const char *name, - size_t size, - int (*init)(struct nf_conn *, u_int32_t)) + size_t size) { int ret = 0; char *cache_name; - kmem_cache_t *cachep; + struct kmem_cache *cachep; DEBUGP("nf_conntrack_register_cache: features=0x%x, name=%s, size=%d\n", features, name, size); @@ -288,7 +158,7 @@ int nf_conntrack_register_cache(u_int32_t features, const char *name, return -EINVAL; } - down(&nf_ct_cache_mutex); + mutex_lock(&nf_ct_cache_mutex); write_lock_bh(&nf_ct_cache_lock); /* e.g: multiple helpers are loaded */ @@ -296,8 +166,7 @@ int nf_conntrack_register_cache(u_int32_t features, const char *name, DEBUGP("nf_conntrack_register_cache: already resisterd.\n"); if ((!strncmp(nf_ct_cache[features].name, name, NF_CT_FEATURES_NAMELEN)) - && nf_ct_cache[features].size == size - && nf_ct_cache[features].init_conntrack == init) { + && nf_ct_cache[features].size == size) { DEBUGP("nf_conntrack_register_cache: reusing.\n"); nf_ct_cache[features].use++; ret = 0; @@ -305,7 +174,7 @@ int nf_conntrack_register_cache(u_int32_t features, const char *name, ret = -EBUSY; write_unlock_bh(&nf_ct_cache_lock); - up(&nf_ct_cache_mutex); + mutex_unlock(&nf_ct_cache_mutex); return ret; } write_unlock_bh(&nf_ct_cache_lock); @@ -340,7 +209,6 @@ int nf_conntrack_register_cache(u_int32_t features, const char *name, write_lock_bh(&nf_ct_cache_lock); nf_ct_cache[features].use = 1; nf_ct_cache[features].size = size; - nf_ct_cache[features].init_conntrack = init; nf_ct_cache[features].cachep = cachep; nf_ct_cache[features].name = cache_name; write_unlock_bh(&nf_ct_cache_lock); @@ -350,14 +218,15 @@ int nf_conntrack_register_cache(u_int32_t features, const char *name, out_free_name: kfree(cache_name); out_up_mutex: - up(&nf_ct_cache_mutex); + mutex_unlock(&nf_ct_cache_mutex); return ret; } +EXPORT_SYMBOL_GPL(nf_conntrack_register_cache); /* FIXME: In the current, only nf_conntrack_cleanup() can call this function. */ void nf_conntrack_unregister_cache(u_int32_t features) { - kmem_cache_t *cachep; + struct kmem_cache *cachep; char *name; /* @@ -365,19 +234,18 @@ void nf_conntrack_unregister_cache(u_int32_t features) * slab cache. */ DEBUGP("nf_conntrack_unregister_cache: 0x%04x\n", features); - down(&nf_ct_cache_mutex); + mutex_lock(&nf_ct_cache_mutex); write_lock_bh(&nf_ct_cache_lock); if (--nf_ct_cache[features].use > 0) { write_unlock_bh(&nf_ct_cache_lock); - up(&nf_ct_cache_mutex); + mutex_unlock(&nf_ct_cache_mutex); return; } cachep = nf_ct_cache[features].cachep; name = nf_ct_cache[features].name; nf_ct_cache[features].cachep = NULL; nf_ct_cache[features].name = NULL; - nf_ct_cache[features].init_conntrack = NULL; nf_ct_cache[features].size = 0; write_unlock_bh(&nf_ct_cache_lock); @@ -386,8 +254,9 @@ void nf_conntrack_unregister_cache(u_int32_t features) kmem_cache_destroy(cachep); kfree(name); - up(&nf_ct_cache_mutex); + mutex_unlock(&nf_ct_cache_mutex); } +EXPORT_SYMBOL_GPL(nf_conntrack_unregister_cache); int nf_ct_get_tuple(const struct sk_buff *skb, @@ -397,7 +266,7 @@ nf_ct_get_tuple(const struct sk_buff *skb, u_int8_t protonum, struct nf_conntrack_tuple *tuple, const struct nf_conntrack_l3proto *l3proto, - const struct nf_conntrack_protocol *protocol) + const struct nf_conntrack_l4proto *l4proto) { NF_CT_TUPLE_U_BLANK(tuple); @@ -408,14 +277,15 @@ nf_ct_get_tuple(const struct sk_buff *skb, tuple->dst.protonum = protonum; tuple->dst.dir = IP_CT_DIR_ORIGINAL; - return protocol->pkt_to_tuple(skb, dataoff, tuple); + return l4proto->pkt_to_tuple(skb, dataoff, tuple); } +EXPORT_SYMBOL_GPL(nf_ct_get_tuple); int nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, const struct nf_conntrack_tuple *orig, const struct nf_conntrack_l3proto *l3proto, - const struct nf_conntrack_protocol *protocol) + const struct nf_conntrack_l4proto *l4proto) { NF_CT_TUPLE_U_BLANK(inverse); @@ -426,113 +296,16 @@ nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, inverse->dst.dir = !orig->dst.dir; inverse->dst.protonum = orig->dst.protonum; - return protocol->invert_tuple(inverse, orig); -} - -/* nf_conntrack_expect helper functions */ -void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) -{ - ASSERT_WRITE_LOCK(&nf_conntrack_lock); - NF_CT_ASSERT(!timer_pending(&exp->timeout)); - list_del(&exp->list); - NF_CT_STAT_INC(expect_delete); - exp->master->expecting--; - nf_conntrack_expect_put(exp); -} - -static void expectation_timed_out(unsigned long ul_expect) -{ - struct nf_conntrack_expect *exp = (void *)ul_expect; - - write_lock_bh(&nf_conntrack_lock); - nf_ct_unlink_expect(exp); - write_unlock_bh(&nf_conntrack_lock); - nf_conntrack_expect_put(exp); -} - -struct nf_conntrack_expect * -__nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_expect *i; - - list_for_each_entry(i, &nf_conntrack_expect_list, list) { - if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) { - atomic_inc(&i->use); - return i; - } - } - return NULL; -} - -/* Just find a expectation corresponding to a tuple. */ -struct nf_conntrack_expect * -nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_expect *i; - - read_lock_bh(&nf_conntrack_lock); - i = __nf_conntrack_expect_find(tuple); - read_unlock_bh(&nf_conntrack_lock); - - return i; -} - -/* If an expectation for this connection is found, it gets delete from - * global list then returned. */ -static struct nf_conntrack_expect * -find_expectation(const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_expect *i; - - list_for_each_entry(i, &nf_conntrack_expect_list, list) { - /* If master is not in hash table yet (ie. packet hasn't left - this machine yet), how can other end know about expected? - Hence these are not the droids you are looking for (if - master ct never got confirmed, we'd hold a reference to it - and weird things would happen to future packets). */ - if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) - && nf_ct_is_confirmed(i->master)) { - if (i->flags & NF_CT_EXPECT_PERMANENT) { - atomic_inc(&i->use); - return i; - } else if (del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - return i; - } - } - } - return NULL; -} - -/* delete all expectations for this conntrack */ -void nf_ct_remove_expectations(struct nf_conn *ct) -{ - struct nf_conntrack_expect *i, *tmp; - - /* Optimization: most connection never expect any others. */ - if (ct->expecting == 0) - return; - - list_for_each_entry_safe(i, tmp, &nf_conntrack_expect_list, list) { - if (i->master == ct && del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - nf_conntrack_expect_put(i); - } - } + return l4proto->invert_tuple(inverse, orig); } +EXPORT_SYMBOL_GPL(nf_ct_invert_tuple); static void clean_from_lists(struct nf_conn *ct) { - unsigned int ho, hr; - DEBUGP("clean_from_lists(%p)\n", ct); - ASSERT_WRITE_LOCK(&nf_conntrack_lock); - - ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); - hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); - LIST_DELETE(&nf_conntrack_hash[ho], &ct->tuplehash[IP_CT_DIR_ORIGINAL]); - LIST_DELETE(&nf_conntrack_hash[hr], &ct->tuplehash[IP_CT_DIR_REPLY]); + list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list); + list_del(&ct->tuplehash[IP_CT_DIR_REPLY].list); /* Destroy all pending expectations */ nf_ct_remove_expectations(ct); @@ -542,8 +315,9 @@ static void destroy_conntrack(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; + struct nf_conn_help *help = nfct_help(ct); struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; DEBUGP("destroy_conntrack(%p)\n", ct); NF_CT_ASSERT(atomic_read(&nfct->use) == 0); @@ -552,6 +326,9 @@ destroy_conntrack(struct nf_conntrack *nfct) nf_conntrack_event(IPCT_DESTROY, ct); set_bit(IPS_DYING_BIT, &ct->status); + if (help && help->helper && help->helper->destroy) + help->helper->destroy(ct); + /* To make sure we don't get any weird locking issues here: * destroy_conntrack() MUST NOT be called with a write lock * to nf_conntrack_lock!!! -HW */ @@ -559,9 +336,9 @@ destroy_conntrack(struct nf_conntrack *nfct) if (l3proto && l3proto->destroy) l3proto->destroy(ct); - proto = __nf_ct_proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); - if (proto && proto->destroy) - proto->destroy(ct); + l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); + if (l4proto && l4proto->destroy) + l4proto->destroy(ct); if (nf_conntrack_destroyed) nf_conntrack_destroyed(ct); @@ -602,16 +379,6 @@ static void death_by_timeout(unsigned long ul_conntrack) nf_ct_put(ct); } -static inline int -conntrack_tuple_cmp(const struct nf_conntrack_tuple_hash *i, - const struct nf_conntrack_tuple *tuple, - const struct nf_conn *ignored_conntrack) -{ - ASSERT_READ_LOCK(&nf_conntrack_lock); - return nf_ct_tuplehash_to_ctrack(i) != ignored_conntrack - && nf_ct_tuple_equal(tuple, &i->tuple); -} - struct nf_conntrack_tuple_hash * __nf_conntrack_find(const struct nf_conntrack_tuple *tuple, const struct nf_conn *ignored_conntrack) @@ -619,9 +386,9 @@ __nf_conntrack_find(const struct nf_conntrack_tuple *tuple, struct nf_conntrack_tuple_hash *h; unsigned int hash = hash_conntrack(tuple); - ASSERT_READ_LOCK(&nf_conntrack_lock); list_for_each_entry(h, &nf_conntrack_hash[hash], list) { - if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) { + if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack && + nf_ct_tuple_equal(tuple, &h->tuple)) { NF_CT_STAT_INC(found); return h; } @@ -630,6 +397,7 @@ __nf_conntrack_find(const struct nf_conntrack_tuple *tuple, return NULL; } +EXPORT_SYMBOL_GPL(__nf_conntrack_find); /* Find a connection corresponding to a tuple. */ struct nf_conntrack_tuple_hash * @@ -646,16 +414,17 @@ nf_conntrack_find_get(const struct nf_conntrack_tuple *tuple, return h; } +EXPORT_SYMBOL_GPL(nf_conntrack_find_get); static void __nf_conntrack_hash_insert(struct nf_conn *ct, unsigned int hash, unsigned int repl_hash) { ct->id = ++nf_conntrack_next_id; - list_prepend(&nf_conntrack_hash[hash], - &ct->tuplehash[IP_CT_DIR_ORIGINAL].list); - list_prepend(&nf_conntrack_hash[repl_hash], - &ct->tuplehash[IP_CT_DIR_REPLY].list); + list_add(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list, + &nf_conntrack_hash[hash]); + list_add(&ct->tuplehash[IP_CT_DIR_REPLY].list, + &nf_conntrack_hash[repl_hash]); } void nf_conntrack_hash_insert(struct nf_conn *ct) @@ -669,13 +438,16 @@ void nf_conntrack_hash_insert(struct nf_conn *ct) __nf_conntrack_hash_insert(ct, hash, repl_hash); write_unlock_bh(&nf_conntrack_lock); } +EXPORT_SYMBOL_GPL(nf_conntrack_hash_insert); /* Confirm a connection given skb; places it in hash table */ int __nf_conntrack_confirm(struct sk_buff **pskb) { unsigned int hash, repl_hash; + struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; + struct nf_conn_help *help; enum ip_conntrack_info ctinfo; ct = nf_ct_get(*pskb, &ctinfo); @@ -705,43 +477,46 @@ __nf_conntrack_confirm(struct sk_buff **pskb) /* See if there's one in the list already, including reverse: NAT could have grabbed it without realizing, since we're not in the hash. If there is, we lost race. */ - if (!LIST_FIND(&nf_conntrack_hash[hash], - conntrack_tuple_cmp, - struct nf_conntrack_tuple_hash *, - &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, NULL) - && !LIST_FIND(&nf_conntrack_hash[repl_hash], - conntrack_tuple_cmp, - struct nf_conntrack_tuple_hash *, - &ct->tuplehash[IP_CT_DIR_REPLY].tuple, NULL)) { - /* Remove from unconfirmed list */ - list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list); + list_for_each_entry(h, &nf_conntrack_hash[hash], list) + if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, + &h->tuple)) + goto out; + list_for_each_entry(h, &nf_conntrack_hash[repl_hash], list) + if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_REPLY].tuple, + &h->tuple)) + goto out; - __nf_conntrack_hash_insert(ct, hash, repl_hash); - /* Timer relative to confirmation time, not original - setting time, otherwise we'd get timer wrap in - weird delay cases. */ - ct->timeout.expires += jiffies; - add_timer(&ct->timeout); - atomic_inc(&ct->ct_general.use); - set_bit(IPS_CONFIRMED_BIT, &ct->status); - NF_CT_STAT_INC(insert); - write_unlock_bh(&nf_conntrack_lock); - if (ct->helper) - nf_conntrack_event_cache(IPCT_HELPER, *pskb); + /* Remove from unconfirmed list */ + list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list); + + __nf_conntrack_hash_insert(ct, hash, repl_hash); + /* Timer relative to confirmation time, not original + setting time, otherwise we'd get timer wrap in + weird delay cases. */ + ct->timeout.expires += jiffies; + add_timer(&ct->timeout); + atomic_inc(&ct->ct_general.use); + set_bit(IPS_CONFIRMED_BIT, &ct->status); + NF_CT_STAT_INC(insert); + write_unlock_bh(&nf_conntrack_lock); + help = nfct_help(ct); + if (help && help->helper) + nf_conntrack_event_cache(IPCT_HELPER, *pskb); #ifdef CONFIG_NF_NAT_NEEDED - if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) || - test_bit(IPS_DST_NAT_DONE_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_NATINFO, *pskb); + if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) || + test_bit(IPS_DST_NAT_DONE_BIT, &ct->status)) + nf_conntrack_event_cache(IPCT_NATINFO, *pskb); #endif - nf_conntrack_event_cache(master_ct(ct) ? - IPCT_RELATED : IPCT_NEW, *pskb); - return NF_ACCEPT; - } + nf_conntrack_event_cache(master_ct(ct) ? + IPCT_RELATED : IPCT_NEW, *pskb); + return NF_ACCEPT; +out: NF_CT_STAT_INC(insert_failed); write_unlock_bh(&nf_conntrack_lock); return NF_DROP; } +EXPORT_SYMBOL_GPL(__nf_conntrack_confirm); /* Returns true if a connection correspondings to the tuple (required for NAT). */ @@ -757,27 +532,25 @@ nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, return h != NULL; } +EXPORT_SYMBOL_GPL(nf_conntrack_tuple_taken); /* There's a small race here where we may free a just-assured connection. Too bad: we're in trouble anyway. */ -static inline int unreplied(const struct nf_conntrack_tuple_hash *i) -{ - return !(test_bit(IPS_ASSURED_BIT, - &nf_ct_tuplehash_to_ctrack(i)->status)); -} - static int early_drop(struct list_head *chain) { /* Traverse backwards: gives us oldest, which is roughly LRU */ struct nf_conntrack_tuple_hash *h; - struct nf_conn *ct = NULL; + struct nf_conn *ct = NULL, *tmp; int dropped = 0; read_lock_bh(&nf_conntrack_lock); - h = LIST_FIND_B(chain, unreplied, struct nf_conntrack_tuple_hash *); - if (h) { - ct = nf_ct_tuplehash_to_ctrack(h); - atomic_inc(&ct->ct_general.use); + list_for_each_entry_reverse(h, chain, list) { + tmp = nf_ct_tuplehash_to_ctrack(h); + if (!test_bit(IPS_ASSURED_BIT, &tmp->status)) { + ct = tmp; + atomic_inc(&ct->ct_general.use); + break; + } } read_unlock_bh(&nf_conntrack_lock); @@ -793,66 +566,29 @@ static int early_drop(struct list_head *chain) return dropped; } -static inline int helper_cmp(const struct nf_conntrack_helper *i, - const struct nf_conntrack_tuple *rtuple) -{ - return nf_ct_tuple_mask_cmp(rtuple, &i->tuple, &i->mask); -} - -static struct nf_conntrack_helper * -__nf_ct_helper_find(const struct nf_conntrack_tuple *tuple) -{ - return LIST_FIND(&helpers, helper_cmp, - struct nf_conntrack_helper *, - tuple); -} - -struct nf_conntrack_helper * -nf_ct_helper_find_get( const struct nf_conntrack_tuple *tuple) -{ - struct nf_conntrack_helper *helper; - - /* need nf_conntrack_lock to assure that helper exists until - * try_module_get() is called */ - read_lock_bh(&nf_conntrack_lock); - - helper = __nf_ct_helper_find(tuple); - if (helper) { - /* need to increase module usage count to assure helper will - * not go away while the caller is e.g. busy putting a - * conntrack in the hash that uses the helper */ - if (!try_module_get(helper->me)) - helper = NULL; - } - - read_unlock_bh(&nf_conntrack_lock); - - return helper; -} - -void nf_ct_helper_put(struct nf_conntrack_helper *helper) -{ - module_put(helper->me); -} - static struct nf_conn * __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, const struct nf_conntrack_tuple *repl, - const struct nf_conntrack_l3proto *l3proto) + const struct nf_conntrack_l3proto *l3proto, + u_int32_t features) { struct nf_conn *conntrack = NULL; - u_int32_t features = 0; + struct nf_conntrack_helper *helper; - if (!nf_conntrack_hash_rnd_initted) { + if (unlikely(!nf_conntrack_hash_rnd_initted)) { get_random_bytes(&nf_conntrack_hash_rnd, 4); nf_conntrack_hash_rnd_initted = 1; } + /* We don't want any race condition at early drop stage */ + atomic_inc(&nf_conntrack_count); + if (nf_conntrack_max - && atomic_read(&nf_conntrack_count) >= nf_conntrack_max) { + && atomic_read(&nf_conntrack_count) > nf_conntrack_max) { unsigned int hash = hash_conntrack(orig); /* Try dropping from this hash chain. */ if (!early_drop(&nf_conntrack_hash[hash])) { + atomic_dec(&nf_conntrack_count); if (net_ratelimit()) printk(KERN_WARNING "nf_conntrack: table full, dropping" @@ -862,9 +598,13 @@ __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, } /* find features needed by this conntrack. */ - features = l3proto->get_features(orig); + features |= l3proto->get_features(orig); + + /* FIXME: protect helper list per RCU */ read_lock_bh(&nf_conntrack_lock); - if (__nf_ct_helper_find(repl) != NULL) + helper = __nf_ct_helper_find(repl); + /* NAT might want to assign a helper later */ + if (helper || features & NF_CT_F_NAT) features |= NF_CT_F_HELP; read_unlock_bh(&nf_conntrack_lock); @@ -872,7 +612,7 @@ __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, read_lock_bh(&nf_ct_cache_lock); - if (!nf_ct_cache[features].use) { + if (unlikely(!nf_ct_cache[features].use)) { DEBUGP("nf_conntrack_alloc: not supported features = 0x%x\n", features); goto out; @@ -886,14 +626,6 @@ __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, memset(conntrack, 0, nf_ct_cache[features].size); conntrack->features = features; - if (nf_ct_cache[features].init_conntrack && - nf_ct_cache[features].init_conntrack(conntrack, features) < 0) { - DEBUGP("nf_conntrack_alloc: failed to init\n"); - kmem_cache_free(nf_ct_cache[features].cachep, conntrack); - conntrack = NULL; - goto out; - } - atomic_set(&conntrack->ct_general.use, 1); conntrack->ct_general.destroy = destroy_conntrack; conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig; @@ -902,10 +634,12 @@ __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, init_timer(&conntrack->timeout); conntrack->timeout.data = (unsigned long)conntrack; conntrack->timeout.function = death_by_timeout; + read_unlock_bh(&nf_ct_cache_lock); - atomic_inc(&nf_conntrack_count); + return conntrack; out: read_unlock_bh(&nf_ct_cache_lock); + atomic_dec(&nf_conntrack_count); return conntrack; } @@ -915,8 +649,9 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, struct nf_conntrack_l3proto *l3proto; l3proto = __nf_ct_l3proto_find(orig->src.l3num); - return __nf_conntrack_alloc(orig, repl, l3proto); + return __nf_conntrack_alloc(orig, repl, l3proto, 0); } +EXPORT_SYMBOL_GPL(nf_conntrack_alloc); void nf_conntrack_free(struct nf_conn *conntrack) { @@ -927,32 +662,40 @@ void nf_conntrack_free(struct nf_conn *conntrack) kmem_cache_free(nf_ct_cache[features].cachep, conntrack); atomic_dec(&nf_conntrack_count); } +EXPORT_SYMBOL_GPL(nf_conntrack_free); /* Allocate a new conntrack: we return -ENOMEM if classification failed due to stress. Otherwise it really is unclassifiable. */ static struct nf_conntrack_tuple_hash * init_conntrack(const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_protocol *protocol, + struct nf_conntrack_l4proto *l4proto, struct sk_buff *skb, unsigned int dataoff) { struct nf_conn *conntrack; struct nf_conntrack_tuple repl_tuple; struct nf_conntrack_expect *exp; + u_int32_t features = 0; - if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, protocol)) { + if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, l4proto)) { DEBUGP("Can't invert tuple.\n"); return NULL; } - conntrack = __nf_conntrack_alloc(tuple, &repl_tuple, l3proto); + read_lock_bh(&nf_conntrack_lock); + exp = __nf_conntrack_expect_find(tuple); + if (exp && exp->helper) + features = NF_CT_F_HELP; + read_unlock_bh(&nf_conntrack_lock); + + conntrack = __nf_conntrack_alloc(tuple, &repl_tuple, l3proto, features); if (conntrack == NULL || IS_ERR(conntrack)) { DEBUGP("Can't allocate conntrack.\n"); return (struct nf_conntrack_tuple_hash *)conntrack; } - if (!protocol->new(conntrack, skb, dataoff)) { + if (!l4proto->new(conntrack, skb, dataoff)) { nf_conntrack_free(conntrack); DEBUGP("init conntrack: can't track with proto module\n"); return NULL; @@ -967,16 +710,23 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, /* Welcome, Mr. Bond. We've been expecting you... */ __set_bit(IPS_EXPECTED_BIT, &conntrack->status); conntrack->master = exp->master; + if (exp->helper) + nfct_help(conntrack)->helper = exp->helper; #ifdef CONFIG_NF_CONNTRACK_MARK conntrack->mark = exp->master->mark; +#endif +#ifdef CONFIG_NF_CONNTRACK_SECMARK + conntrack->secmark = exp->master->secmark; #endif nf_conntrack_get(&conntrack->master->ct_general); NF_CT_STAT_INC(expect_new); } else { - conntrack->helper = __nf_ct_helper_find(&repl_tuple); + struct nf_conn_help *help = nfct_help(conntrack); + if (help) + help->helper = __nf_ct_helper_find(&repl_tuple); NF_CT_STAT_INC(new); - } + } /* Overload tuple linked list to put us in unconfirmed list. */ list_add(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].list, &unconfirmed); @@ -999,7 +749,7 @@ resolve_normal_ct(struct sk_buff *skb, u_int16_t l3num, u_int8_t protonum, struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_protocol *proto, + struct nf_conntrack_l4proto *l4proto, int *set_reply, enum ip_conntrack_info *ctinfo) { @@ -1009,7 +759,7 @@ resolve_normal_ct(struct sk_buff *skb, if (!nf_ct_get_tuple(skb, (unsigned int)(skb->nh.raw - skb->data), dataoff, l3num, protonum, &tuple, l3proto, - proto)) { + l4proto)) { DEBUGP("resolve_normal_ct: Can't get tuple\n"); return NULL; } @@ -1017,7 +767,7 @@ resolve_normal_ct(struct sk_buff *skb, /* look for tuple match */ h = nf_conntrack_find_get(&tuple, NULL); if (!h) { - h = init_conntrack(&tuple, l3proto, proto, skb, dataoff); + h = init_conntrack(&tuple, l3proto, l4proto, skb, dataoff); if (!h) return NULL; if (IS_ERR(h)) @@ -1055,7 +805,7 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) struct nf_conn *ct; enum ip_conntrack_info ctinfo; struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_protocol *proto; + struct nf_conntrack_l4proto *l4proto; unsigned int dataoff; u_int8_t protonum; int set_reply = 0; @@ -1073,19 +823,19 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) return -ret; } - proto = __nf_ct_proto_find((u_int16_t)pf, protonum); + l4proto = __nf_ct_l4proto_find((u_int16_t)pf, protonum); /* It may be an special packet, error, unclean... * inverse of the return code tells to the netfilter * core what to do with the packet. */ - if (proto->error != NULL && - (ret = proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) { + if (l4proto->error != NULL && + (ret = l4proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) { NF_CT_STAT_INC(error); NF_CT_STAT_INC(invalid); return -ret; } - ct = resolve_normal_ct(*pskb, dataoff, pf, protonum, l3proto, proto, + ct = resolve_normal_ct(*pskb, dataoff, pf, protonum, l3proto, l4proto, &set_reply, &ctinfo); if (!ct) { /* Not valid part of a connection */ @@ -1101,7 +851,7 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) NF_CT_ASSERT((*pskb)->nfct); - ret = proto->packet(ct, *pskb, dataoff, ctinfo, pf, hooknum); + ret = l4proto->packet(ct, *pskb, dataoff, ctinfo, pf, hooknum); if (ret < 0) { /* Invalid: inverse of the return code tells * the netfilter core what to do */ @@ -1117,261 +867,38 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) return ret; } +EXPORT_SYMBOL_GPL(nf_conntrack_in); int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, const struct nf_conntrack_tuple *orig) { return nf_ct_invert_tuple(inverse, orig, __nf_ct_l3proto_find(orig->src.l3num), - __nf_ct_proto_find(orig->src.l3num, + __nf_ct_l4proto_find(orig->src.l3num, orig->dst.protonum)); } - -/* Would two expected things clash? */ -static inline int expect_clash(const struct nf_conntrack_expect *a, - const struct nf_conntrack_expect *b) -{ - /* Part covered by intersection of masks must be unequal, - otherwise they clash */ - struct nf_conntrack_tuple intersect_mask; - int count; - - intersect_mask.src.l3num = a->mask.src.l3num & b->mask.src.l3num; - intersect_mask.src.u.all = a->mask.src.u.all & b->mask.src.u.all; - intersect_mask.dst.u.all = a->mask.dst.u.all & b->mask.dst.u.all; - intersect_mask.dst.protonum = a->mask.dst.protonum - & b->mask.dst.protonum; - - for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ - intersect_mask.src.u3.all[count] = - a->mask.src.u3.all[count] & b->mask.src.u3.all[count]; - } - - for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ - intersect_mask.dst.u3.all[count] = - a->mask.dst.u3.all[count] & b->mask.dst.u3.all[count]; - } - - return nf_ct_tuple_mask_cmp(&a->tuple, &b->tuple, &intersect_mask); -} - -static inline int expect_matches(const struct nf_conntrack_expect *a, - const struct nf_conntrack_expect *b) -{ - return a->master == b->master - && nf_ct_tuple_equal(&a->tuple, &b->tuple) - && nf_ct_tuple_equal(&a->mask, &b->mask); -} - -/* Generally a bad idea to call this: could have matched already. */ -void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp) -{ - struct nf_conntrack_expect *i; - - write_lock_bh(&nf_conntrack_lock); - /* choose the the oldest expectation to evict */ - list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { - if (expect_matches(i, exp) && del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - write_unlock_bh(&nf_conntrack_lock); - nf_conntrack_expect_put(i); - return; - } - } - write_unlock_bh(&nf_conntrack_lock); -} - -/* We don't increase the master conntrack refcount for non-fulfilled - * conntracks. During the conntrack destruction, the expectations are - * always killed before the conntrack itself */ -struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me) -{ - struct nf_conntrack_expect *new; - - new = kmem_cache_alloc(nf_conntrack_expect_cachep, GFP_ATOMIC); - if (!new) { - DEBUGP("expect_related: OOM allocating expect\n"); - return NULL; - } - new->master = me; - atomic_set(&new->use, 1); - return new; -} - -void nf_conntrack_expect_put(struct nf_conntrack_expect *exp) -{ - if (atomic_dec_and_test(&exp->use)) - kmem_cache_free(nf_conntrack_expect_cachep, exp); -} - -static void nf_conntrack_expect_insert(struct nf_conntrack_expect *exp) -{ - atomic_inc(&exp->use); - exp->master->expecting++; - list_add(&exp->list, &nf_conntrack_expect_list); - - init_timer(&exp->timeout); - exp->timeout.data = (unsigned long)exp; - exp->timeout.function = expectation_timed_out; - exp->timeout.expires = jiffies + exp->master->helper->timeout * HZ; - add_timer(&exp->timeout); - - exp->id = ++nf_conntrack_expect_next_id; - atomic_inc(&exp->use); - NF_CT_STAT_INC(expect_create); -} - -/* Race with expectations being used means we could have none to find; OK. */ -static void evict_oldest_expect(struct nf_conn *master) -{ - struct nf_conntrack_expect *i; - - list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { - if (i->master == master) { - if (del_timer(&i->timeout)) { - nf_ct_unlink_expect(i); - nf_conntrack_expect_put(i); - } - break; - } - } -} - -static inline int refresh_timer(struct nf_conntrack_expect *i) -{ - if (!del_timer(&i->timeout)) - return 0; - - i->timeout.expires = jiffies + i->master->helper->timeout*HZ; - add_timer(&i->timeout); - return 1; -} - -int nf_conntrack_expect_related(struct nf_conntrack_expect *expect) -{ - struct nf_conntrack_expect *i; - struct nf_conn *master = expect->master; - int ret; - - DEBUGP("nf_conntrack_expect_related %p\n", related_to); - DEBUGP("tuple: "); NF_CT_DUMP_TUPLE(&expect->tuple); - DEBUGP("mask: "); NF_CT_DUMP_TUPLE(&expect->mask); - - write_lock_bh(&nf_conntrack_lock); - list_for_each_entry(i, &nf_conntrack_expect_list, list) { - if (expect_matches(i, expect)) { - /* Refresh timer: if it's dying, ignore.. */ - if (refresh_timer(i)) { - ret = 0; - goto out; - } - } else if (expect_clash(i, expect)) { - ret = -EBUSY; - goto out; - } - } - /* Will be over limit? */ - if (master->helper->max_expected && - master->expecting >= master->helper->max_expected) - evict_oldest_expect(master); - - nf_conntrack_expect_insert(expect); - nf_conntrack_expect_event(IPEXP_NEW, expect); - ret = 0; -out: - write_unlock_bh(&nf_conntrack_lock); - return ret; -} +EXPORT_SYMBOL_GPL(nf_ct_invert_tuplepr); /* Alter reply tuple (maybe alter helper). This is for NAT, and is implicitly racy: see __nf_conntrack_confirm */ -void nf_conntrack_alter_reply(struct nf_conn *conntrack, +void nf_conntrack_alter_reply(struct nf_conn *ct, const struct nf_conntrack_tuple *newreply) { + struct nf_conn_help *help = nfct_help(ct); + write_lock_bh(&nf_conntrack_lock); /* Should be unconfirmed, so not in hash table yet */ - NF_CT_ASSERT(!nf_ct_is_confirmed(conntrack)); + NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); - DEBUGP("Altering reply tuple of %p to ", conntrack); + DEBUGP("Altering reply tuple of %p to ", ct); NF_CT_DUMP_TUPLE(newreply); - conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; - if (!conntrack->master && conntrack->expecting == 0) - conntrack->helper = __nf_ct_helper_find(newreply); + ct->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; + if (!ct->master && help && help->expecting == 0) + help->helper = __nf_ct_helper_find(newreply); write_unlock_bh(&nf_conntrack_lock); } - -int nf_conntrack_helper_register(struct nf_conntrack_helper *me) -{ - int ret; - BUG_ON(me->timeout == 0); - - ret = nf_conntrack_register_cache(NF_CT_F_HELP, "nf_conntrack:help", - sizeof(struct nf_conn) - + sizeof(union nf_conntrack_help) - + __alignof__(union nf_conntrack_help), - init_conntrack_for_helper); - if (ret < 0) { - printk(KERN_ERR "nf_conntrack_helper_reigster: Unable to create slab cache for conntracks\n"); - return ret; - } - write_lock_bh(&nf_conntrack_lock); - list_prepend(&helpers, me); - write_unlock_bh(&nf_conntrack_lock); - - return 0; -} - -struct nf_conntrack_helper * -__nf_conntrack_helper_find_byname(const char *name) -{ - struct nf_conntrack_helper *h; - - list_for_each_entry(h, &helpers, list) { - if (!strcmp(h->name, name)) - return h; - } - - return NULL; -} - -static inline int unhelp(struct nf_conntrack_tuple_hash *i, - const struct nf_conntrack_helper *me) -{ - if (nf_ct_tuplehash_to_ctrack(i)->helper == me) { - nf_conntrack_event(IPCT_HELPER, nf_ct_tuplehash_to_ctrack(i)); - nf_ct_tuplehash_to_ctrack(i)->helper = NULL; - } - return 0; -} - -void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) -{ - unsigned int i; - struct nf_conntrack_expect *exp, *tmp; - - /* Need write lock here, to delete helper. */ - write_lock_bh(&nf_conntrack_lock); - LIST_DELETE(&helpers, me); - - /* Get rid of expectations */ - list_for_each_entry_safe(exp, tmp, &nf_conntrack_expect_list, list) { - if (exp->master->helper == me && del_timer(&exp->timeout)) { - nf_ct_unlink_expect(exp); - nf_conntrack_expect_put(exp); - } - } - - /* Get rid of expecteds, set helpers to NULL. */ - LIST_FIND_W(&unconfirmed, unhelp, struct nf_conntrack_tuple_hash*, me); - for (i = 0; i < nf_conntrack_htable_size; i++) - LIST_FIND_W(&nf_conntrack_hash[i], unhelp, - struct nf_conntrack_tuple_hash *, me); - write_unlock_bh(&nf_conntrack_lock); - - /* Someone could be still looking at the helper in a bh. */ - synchronize_net(); -} +EXPORT_SYMBOL_GPL(nf_conntrack_alter_reply); /* Refresh conntrack for this many jiffies and do accounting if do_acct is 1 */ void __nf_ct_refresh_acct(struct nf_conn *ct, @@ -1387,14 +914,25 @@ void __nf_ct_refresh_acct(struct nf_conn *ct, write_lock_bh(&nf_conntrack_lock); + /* Only update if this is not a fixed timeout */ + if (test_bit(IPS_FIXED_TIMEOUT_BIT, &ct->status)) { + write_unlock_bh(&nf_conntrack_lock); + return; + } + /* If not in hash table, timer will not be active yet */ if (!nf_ct_is_confirmed(ct)) { ct->timeout.expires = extra_jiffies; event = IPCT_REFRESH; } else { - /* Need del_timer for race avoidance (may already be dying). */ - if (del_timer(&ct->timeout)) { - ct->timeout.expires = jiffies + extra_jiffies; + unsigned long newtime = jiffies + extra_jiffies; + + /* Only update the timeout if the new timeout is at least + HZ jiffies from the old timeout. Need del_timer for race + avoidance (may already be dying). */ + if (newtime - ct->timeout.expires >= HZ + && del_timer(&ct->timeout)) { + ct->timeout.expires = newtime; add_timer(&ct->timeout); event = IPCT_REFRESH; } @@ -1405,9 +943,10 @@ void __nf_ct_refresh_acct(struct nf_conn *ct, ct->counters[CTINFO2DIR(ctinfo)].packets++; ct->counters[CTINFO2DIR(ctinfo)].bytes += skb->len - (unsigned int)(skb->nh.raw - skb->data); - if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000) - || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000)) - event |= IPCT_COUNTER_FILLING; + + if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000) + || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000)) + event |= IPCT_COUNTER_FILLING; } #endif @@ -1417,12 +956,15 @@ void __nf_ct_refresh_acct(struct nf_conn *ct, if (event) nf_conntrack_event_cache(event, skb); } +EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct); #if defined(CONFIG_NF_CT_NETLINK) || \ defined(CONFIG_NF_CT_NETLINK_MODULE) #include #include +#include + /* Generic function for tcp/udp/sctp/dccp and alike. This needs to be * in ip_conntrack_core, since we don't want the protocols to autoload @@ -1439,6 +981,7 @@ int nf_ct_port_tuple_to_nfattr(struct sk_buff *skb, nfattr_failure: return -1; } +EXPORT_SYMBOL_GPL(nf_ct_port_tuple_to_nfattr); static const size_t cta_min_proto[CTA_PROTO_MAX] = { [CTA_PROTO_SRC_PORT-1] = sizeof(u_int16_t), @@ -1454,13 +997,12 @@ int nf_ct_port_nfattr_to_tuple(struct nfattr *tb[], if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto)) return -EINVAL; - t->src.u.tcp.port = - *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]); - t->dst.u.tcp.port = - *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]); + t->src.u.tcp.port = *(__be16 *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]); + t->dst.u.tcp.port = *(__be16 *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]); return 0; } +EXPORT_SYMBOL_GPL(nf_ct_port_nfattr_to_tuple); #endif /* Used by ipt_REJECT and ip6t_REJECT. */ @@ -1481,6 +1023,7 @@ void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb) nskb->nfctinfo = ctinfo; nf_conntrack_get(nskb->nfct); } +EXPORT_SYMBOL_GPL(__nf_conntrack_attach); static inline int do_iter(const struct nf_conntrack_tuple_hash *i, @@ -1491,37 +1034,41 @@ do_iter(const struct nf_conntrack_tuple_hash *i, } /* Bring out ya dead! */ -static struct nf_conntrack_tuple_hash * +static struct nf_conn * get_next_corpse(int (*iter)(struct nf_conn *i, void *data), void *data, unsigned int *bucket) { - struct nf_conntrack_tuple_hash *h = NULL; + struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct; write_lock_bh(&nf_conntrack_lock); for (; *bucket < nf_conntrack_htable_size; (*bucket)++) { - h = LIST_FIND_W(&nf_conntrack_hash[*bucket], do_iter, - struct nf_conntrack_tuple_hash *, iter, data); - if (h) - break; + list_for_each_entry(h, &nf_conntrack_hash[*bucket], list) { + ct = nf_ct_tuplehash_to_ctrack(h); + if (iter(ct, data)) + goto found; + } } - if (!h) - h = LIST_FIND_W(&unconfirmed, do_iter, - struct nf_conntrack_tuple_hash *, iter, data); - if (h) - atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use); + list_for_each_entry(h, &unconfirmed, list) { + ct = nf_ct_tuplehash_to_ctrack(h); + if (iter(ct, data)) + set_bit(IPS_DYING_BIT, &ct->status); + } write_unlock_bh(&nf_conntrack_lock); - - return h; + return NULL; +found: + atomic_inc(&ct->ct_general.use); + write_unlock_bh(&nf_conntrack_lock); + return ct; } void nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data) { - struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct; unsigned int bucket = 0; - while ((h = get_next_corpse(iter, data, &bucket)) != NULL) { - struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); + while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) { /* Time to push up daises... */ if (del_timer(&ct->timeout)) death_by_timeout((unsigned long)ct); @@ -1530,6 +1077,7 @@ nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data) nf_ct_put(ct); } } +EXPORT_SYMBOL_GPL(nf_ct_iterate_cleanup); static int kill_all(struct nf_conn *i, void *data) { @@ -1545,10 +1093,11 @@ static void free_conntrack_hash(struct list_head *hash, int vmalloced, int size) get_order(sizeof(struct list_head) * size)); } -void nf_conntrack_flush() +void nf_conntrack_flush(void) { nf_ct_iterate_cleanup(kill_all, NULL); } +EXPORT_SYMBOL_GPL(nf_conntrack_flush); /* Mishearing the voices in his head, our hero wonders how he's supposed to kill the mall. */ @@ -1586,6 +1135,8 @@ void nf_conntrack_cleanup(void) free_conntrack_hash(nf_conntrack_hash, nf_conntrack_vmalloc, nf_conntrack_htable_size); + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_generic); + /* free l3proto protocol tables */ for (i = 0; i < PF_MAX; i++) if (nf_ct_protos[i]) { @@ -1697,7 +1248,7 @@ int __init nf_conntrack_init(void) } ret = nf_conntrack_register_cache(NF_CT_F_BASIC, "nf_conntrack:basic", - sizeof(struct nf_conn), NULL); + sizeof(struct nf_conn)); if (ret < 0) { printk(KERN_ERR "Unable to create nf_conn slab cache\n"); goto err_free_hash; @@ -1711,10 +1262,14 @@ int __init nf_conntrack_init(void) goto err_free_conntrack_slab; } + ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_generic); + if (ret < 0) + goto out_free_expect_slab; + /* Don't NEED lock here, but good form anyway. */ write_lock_bh(&nf_conntrack_lock); - for (i = 0; i < PF_MAX; i++) - nf_ct_l3protos[i] = &nf_conntrack_generic_l3proto; + for (i = 0; i < AF_MAX; i++) + nf_ct_l3protos[i] = &nf_conntrack_l3proto_generic; write_unlock_bh(&nf_conntrack_lock); /* For use by REJECT target */ @@ -1728,6 +1283,8 @@ int __init nf_conntrack_init(void) return ret; +out_free_expect_slab: + kmem_cache_destroy(nf_conntrack_expect_cachep); err_free_conntrack_slab: nf_conntrack_unregister_cache(NF_CT_F_BASIC); err_free_hash: