X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=net%2Fnetfilter%2Fnf_conntrack_core.c;h=471e2a79d26fad0712d2797e14126b3129b2640e;hb=b2a15a604d379af323645e330638e2cfcc696aff;hp=b77eb56a87e33a7ec85bd351e5b6d512922ebbc3;hpb=4e29e9ec7e0707d3925f5dcc29af0d3f04e49833;p=safe%2Fjmp%2Flinux-2.6 diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index b77eb56..471e2a7 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -29,6 +30,7 @@ #include #include #include +#include #include #include @@ -37,36 +39,32 @@ #include #include #include +#include +#include +#include +#include #define NF_CONNTRACK_VERSION "0.5.0" +int (*nfnetlink_parse_nat_setup_hook)(struct nf_conn *ct, + enum nf_nat_manip_type manip, + const struct nlattr *attr) __read_mostly; +EXPORT_SYMBOL_GPL(nfnetlink_parse_nat_setup_hook); + DEFINE_SPINLOCK(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); - unsigned int nf_conntrack_htable_size __read_mostly; EXPORT_SYMBOL_GPL(nf_conntrack_htable_size); -int nf_conntrack_max __read_mostly; +unsigned int nf_conntrack_max __read_mostly; EXPORT_SYMBOL_GPL(nf_conntrack_max); -struct hlist_head *nf_conntrack_hash __read_mostly; -EXPORT_SYMBOL_GPL(nf_conntrack_hash); - struct nf_conn nf_conntrack_untracked __read_mostly; EXPORT_SYMBOL_GPL(nf_conntrack_untracked); -unsigned int nf_ct_log_invalid __read_mostly; -HLIST_HEAD(unconfirmed); -static int nf_conntrack_vmalloc __read_mostly; static struct kmem_cache *nf_conntrack_cachep __read_mostly; -DEFINE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); -EXPORT_PER_CPU_SYMBOL(nf_conntrack_stat); - static int nf_conntrack_hash_rnd_initted; static unsigned int nf_conntrack_hash_rnd; @@ -94,7 +92,7 @@ static inline u_int32_t hash_conntrack(const struct nf_conntrack_tuple *tuple) nf_conntrack_hash_rnd); } -int +bool nf_ct_get_tuple(const struct sk_buff *skb, unsigned int nhoff, unsigned int dataoff, @@ -104,11 +102,11 @@ nf_ct_get_tuple(const struct sk_buff *skb, const struct nf_conntrack_l3proto *l3proto, const struct nf_conntrack_l4proto *l4proto) { - NF_CT_TUPLE_U_BLANK(tuple); + memset(tuple, 0, sizeof(*tuple)); tuple->src.l3num = l3num; if (l3proto->pkt_to_tuple(skb, nhoff, tuple) == 0) - return 0; + return false; tuple->dst.protonum = protonum; tuple->dst.dir = IP_CT_DIR_ORIGINAL; @@ -117,10 +115,8 @@ nf_ct_get_tuple(const struct sk_buff *skb, } EXPORT_SYMBOL_GPL(nf_ct_get_tuple); -int nf_ct_get_tuplepr(const struct sk_buff *skb, - unsigned int nhoff, - u_int16_t l3num, - struct nf_conntrack_tuple *tuple) +bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff, + u_int16_t l3num, struct nf_conntrack_tuple *tuple) { struct nf_conntrack_l3proto *l3proto; struct nf_conntrack_l4proto *l4proto; @@ -134,7 +130,7 @@ int nf_ct_get_tuplepr(const struct sk_buff *skb, ret = l3proto->get_l4proto(skb, nhoff, &protoff, &protonum); if (ret != NF_ACCEPT) { rcu_read_unlock(); - return 0; + return false; } l4proto = __nf_ct_l4proto_find(l3num, protonum); @@ -147,17 +143,17 @@ int nf_ct_get_tuplepr(const struct sk_buff *skb, } EXPORT_SYMBOL_GPL(nf_ct_get_tuplepr); -int +bool 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_l4proto *l4proto) { - NF_CT_TUPLE_U_BLANK(inverse); + memset(inverse, 0, sizeof(*inverse)); inverse->src.l3num = orig->src.l3num; if (l3proto->invert_tuple(inverse, orig) == 0) - return 0; + return false; inverse->dst.dir = !orig->dst.dir; @@ -170,8 +166,8 @@ static void clean_from_lists(struct nf_conn *ct) { pr_debug("clean_from_lists(%p)\n", ct); - hlist_del_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode); - hlist_del_rcu(&ct->tuplehash[IP_CT_DIR_REPLY].hnode); + hlist_nulls_del_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode); + hlist_nulls_del_rcu(&ct->tuplehash[IP_CT_DIR_REPLY].hnnode); /* Destroy all pending expectations */ nf_ct_remove_expectations(ct); @@ -181,26 +177,21 @@ static void destroy_conntrack(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; + struct net *net = nf_ct_net(ct); struct nf_conntrack_l4proto *l4proto; pr_debug("destroy_conntrack(%p)\n", ct); NF_CT_ASSERT(atomic_read(&nfct->use) == 0); NF_CT_ASSERT(!timer_pending(&ct->timeout)); - nf_conntrack_event(IPCT_DESTROY, ct); - set_bit(IPS_DYING_BIT, &ct->status); - /* 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 */ rcu_read_lock(); - l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, - ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); + l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (l4proto && l4proto->destroy) l4proto->destroy(ct); - nf_ct_ext_destroy(ct); - rcu_read_unlock(); spin_lock_bh(&nf_conntrack_lock); @@ -212,11 +203,11 @@ destroy_conntrack(struct nf_conntrack *nfct) /* We overload first tuple to link into unconfirmed list. */ if (!nf_ct_is_confirmed(ct)) { - BUG_ON(hlist_unhashed(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode)); - hlist_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode); + BUG_ON(hlist_nulls_unhashed(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode)); + hlist_nulls_del_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode); } - NF_CT_STAT_INC(delete); + NF_CT_STAT_INC(net, delete); spin_unlock_bh(&nf_conntrack_lock); if (ct->master) @@ -226,48 +217,107 @@ destroy_conntrack(struct nf_conntrack *nfct) nf_conntrack_free(ct); } -static void death_by_timeout(unsigned long ul_conntrack) +void nf_ct_delete_from_lists(struct nf_conn *ct) { - struct nf_conn *ct = (void *)ul_conntrack; - struct nf_conn_help *help = nfct_help(ct); - struct nf_conntrack_helper *helper; - - if (help) { - rcu_read_lock(); - helper = rcu_dereference(help->helper); - if (helper && helper->destroy) - helper->destroy(ct); - rcu_read_unlock(); - } + struct net *net = nf_ct_net(ct); + nf_ct_helper_destroy(ct); spin_lock_bh(&nf_conntrack_lock); /* Inside lock so preempt is disabled on module removal path. * Otherwise we can get spurious warnings. */ - NF_CT_STAT_INC(delete_list); + NF_CT_STAT_INC(net, delete_list); clean_from_lists(ct); spin_unlock_bh(&nf_conntrack_lock); +} +EXPORT_SYMBOL_GPL(nf_ct_delete_from_lists); + +static void death_by_event(unsigned long ul_conntrack) +{ + struct nf_conn *ct = (void *)ul_conntrack; + struct net *net = nf_ct_net(ct); + + if (nf_conntrack_event(IPCT_DESTROY, ct) < 0) { + /* bad luck, let's retry again */ + ct->timeout.expires = jiffies + + (random32() % net->ct.sysctl_events_retry_timeout); + add_timer(&ct->timeout); + return; + } + /* we've got the event delivered, now it's dying */ + set_bit(IPS_DYING_BIT, &ct->status); + spin_lock(&nf_conntrack_lock); + hlist_nulls_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode); + spin_unlock(&nf_conntrack_lock); + nf_ct_put(ct); +} + +void nf_ct_insert_dying_list(struct nf_conn *ct) +{ + struct net *net = nf_ct_net(ct); + + /* add this conntrack to the dying list */ + spin_lock_bh(&nf_conntrack_lock); + hlist_nulls_add_head(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode, + &net->ct.dying); + spin_unlock_bh(&nf_conntrack_lock); + /* set a new timer to retry event delivery */ + setup_timer(&ct->timeout, death_by_event, (unsigned long)ct); + ct->timeout.expires = jiffies + + (random32() % net->ct.sysctl_events_retry_timeout); + add_timer(&ct->timeout); +} +EXPORT_SYMBOL_GPL(nf_ct_insert_dying_list); + +static void death_by_timeout(unsigned long ul_conntrack) +{ + struct nf_conn *ct = (void *)ul_conntrack; + + if (!test_bit(IPS_DYING_BIT, &ct->status) && + unlikely(nf_conntrack_event(IPCT_DESTROY, ct) < 0)) { + /* destroy event was not delivered */ + nf_ct_delete_from_lists(ct); + nf_ct_insert_dying_list(ct); + return; + } + set_bit(IPS_DYING_BIT, &ct->status); + nf_ct_delete_from_lists(ct); nf_ct_put(ct); } +/* + * Warning : + * - Caller must take a reference on returned object + * and recheck nf_ct_tuple_equal(tuple, &h->tuple) + * OR + * - Caller must lock nf_conntrack_lock before calling this function + */ struct nf_conntrack_tuple_hash * -__nf_conntrack_find(const struct nf_conntrack_tuple *tuple) +__nf_conntrack_find(struct net *net, const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_tuple_hash *h; - struct hlist_node *n; + struct hlist_nulls_node *n; unsigned int hash = hash_conntrack(tuple); /* Disable BHs the entire time since we normally need to disable them * at least once for the stats anyway. */ local_bh_disable(); - hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[hash], hnode) { +begin: + hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnnode) { if (nf_ct_tuple_equal(tuple, &h->tuple)) { - NF_CT_STAT_INC(found); + NF_CT_STAT_INC(net, found); local_bh_enable(); return h; } - NF_CT_STAT_INC(searched); + NF_CT_STAT_INC(net, searched); } + /* + * if the nulls value we got at the end of this lookup is + * not the expected one, we must restart lookup. + * We probably met an item that was moved to another chain. + */ + if (get_nulls_value(n) != hash) + goto begin; local_bh_enable(); return NULL; @@ -276,17 +326,25 @@ EXPORT_SYMBOL_GPL(__nf_conntrack_find); /* Find a connection corresponding to a tuple. */ struct nf_conntrack_tuple_hash * -nf_conntrack_find_get(const struct nf_conntrack_tuple *tuple) +nf_conntrack_find_get(struct net *net, const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; rcu_read_lock(); - h = __nf_conntrack_find(tuple); +begin: + h = __nf_conntrack_find(net, tuple); if (h) { ct = nf_ct_tuplehash_to_ctrack(h); - if (unlikely(!atomic_inc_not_zero(&ct->ct_general.use))) + if (unlikely(nf_ct_is_dying(ct) || + !atomic_inc_not_zero(&ct->ct_general.use))) h = NULL; + else { + if (unlikely(!nf_ct_tuple_equal(tuple, &h->tuple))) { + nf_ct_put(ct); + goto begin; + } + } } rcu_read_unlock(); @@ -298,10 +356,12 @@ static void __nf_conntrack_hash_insert(struct nf_conn *ct, unsigned int hash, unsigned int repl_hash) { - hlist_add_head_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode, - &nf_conntrack_hash[hash]); - hlist_add_head_rcu(&ct->tuplehash[IP_CT_DIR_REPLY].hnode, - &nf_conntrack_hash[repl_hash]); + struct net *net = nf_ct_net(ct); + + hlist_nulls_add_head_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode, + &net->ct.hash[hash]); + hlist_nulls_add_head_rcu(&ct->tuplehash[IP_CT_DIR_REPLY].hnnode, + &net->ct.hash[repl_hash]); } void nf_conntrack_hash_insert(struct nf_conn *ct) @@ -311,9 +371,7 @@ void nf_conntrack_hash_insert(struct nf_conn *ct) hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); - spin_lock_bh(&nf_conntrack_lock); __nf_conntrack_hash_insert(ct, hash, repl_hash); - spin_unlock_bh(&nf_conntrack_lock); } EXPORT_SYMBOL_GPL(nf_conntrack_hash_insert); @@ -325,10 +383,12 @@ __nf_conntrack_confirm(struct sk_buff *skb) struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; struct nf_conn_help *help; - struct hlist_node *n; + struct hlist_nulls_node *n; enum ip_conntrack_info ctinfo; + struct net *net; ct = nf_ct_get(skb, &ctinfo); + net = nf_ct_net(ct); /* ipt_REJECT uses nf_conntrack_attach to attach related ICMP/TCP RST packets in other direction. Actual packet @@ -355,19 +415,18 @@ __nf_conntrack_confirm(struct sk_buff *skb) /* 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. */ - hlist_for_each_entry(h, n, &nf_conntrack_hash[hash], hnode) + hlist_nulls_for_each_entry(h, n, &net->ct.hash[hash], hnnode) if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, &h->tuple)) goto out; - hlist_for_each_entry(h, n, &nf_conntrack_hash[repl_hash], hnode) + hlist_nulls_for_each_entry(h, n, &net->ct.hash[repl_hash], hnnode) if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_REPLY].tuple, &h->tuple)) goto out; /* Remove from unconfirmed list */ - hlist_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode); + hlist_nulls_del_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode); - __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. */ @@ -375,22 +434,26 @@ __nf_conntrack_confirm(struct sk_buff *skb) add_timer(&ct->timeout); atomic_inc(&ct->ct_general.use); set_bit(IPS_CONFIRMED_BIT, &ct->status); - NF_CT_STAT_INC(insert); + + /* Since the lookup is lockless, hash insertion must be done after + * starting the timer and setting the CONFIRMED bit. The RCU barriers + * guarantee that no other CPU can find the conntrack before the above + * stores are visible. + */ + __nf_conntrack_hash_insert(ct, hash, repl_hash); + NF_CT_STAT_INC(net, insert); spin_unlock_bh(&nf_conntrack_lock); + help = nfct_help(ct); if (help && help->helper) - nf_conntrack_event_cache(IPCT_HELPER, skb); -#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, skb); -#endif + nf_conntrack_event_cache(IPCT_HELPER, ct); + nf_conntrack_event_cache(master_ct(ct) ? - IPCT_RELATED : IPCT_NEW, skb); + IPCT_RELATED : IPCT_NEW, ct); return NF_ACCEPT; out: - NF_CT_STAT_INC(insert_failed); + NF_CT_STAT_INC(net, insert_failed); spin_unlock_bh(&nf_conntrack_lock); return NF_DROP; } @@ -402,22 +465,23 @@ int nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, const struct nf_conn *ignored_conntrack) { + struct net *net = nf_ct_net(ignored_conntrack); struct nf_conntrack_tuple_hash *h; - struct hlist_node *n; + struct hlist_nulls_node *n; unsigned int hash = hash_conntrack(tuple); /* Disable BHs the entire time since we need to disable them at * least once for the stats anyway. */ rcu_read_lock_bh(); - hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[hash], hnode) { + hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnnode) { if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack && nf_ct_tuple_equal(tuple, &h->tuple)) { - NF_CT_STAT_INC(found); + NF_CT_STAT_INC(net, found); rcu_read_unlock_bh(); return 1; } - NF_CT_STAT_INC(searched); + NF_CT_STAT_INC(net, searched); } rcu_read_unlock_bh(); @@ -429,29 +493,36 @@ 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 noinline int early_drop(unsigned int hash) +static noinline int early_drop(struct net *net, unsigned int hash) { /* Use oldest entry, which is roughly LRU */ struct nf_conntrack_tuple_hash *h; struct nf_conn *ct = NULL, *tmp; - struct hlist_node *n; + struct hlist_nulls_node *n; unsigned int i, cnt = 0; int dropped = 0; rcu_read_lock(); for (i = 0; i < nf_conntrack_htable_size; i++) { - hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[hash], - hnode) { + hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[hash], + hnnode) { tmp = nf_ct_tuplehash_to_ctrack(h); if (!test_bit(IPS_ASSURED_BIT, &tmp->status)) ct = tmp; cnt++; } - if (ct && unlikely(!atomic_inc_not_zero(&ct->ct_general.use))) - ct = NULL; - if (ct || cnt >= NF_CT_EVICTION_RANGE) + if (ct != NULL) { + if (likely(!nf_ct_is_dying(ct) && + atomic_inc_not_zero(&ct->ct_general.use))) + break; + else + ct = NULL; + } + + if (cnt >= NF_CT_EVICTION_RANGE) break; + hash = (hash + 1) % nf_conntrack_htable_size; } rcu_read_unlock(); @@ -462,30 +533,33 @@ static noinline int early_drop(unsigned int hash) if (del_timer(&ct->timeout)) { death_by_timeout((unsigned long)ct); dropped = 1; - NF_CT_STAT_INC_ATOMIC(early_drop); + NF_CT_STAT_INC_ATOMIC(net, early_drop); } nf_ct_put(ct); return dropped; } -struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, - const struct nf_conntrack_tuple *repl) +struct nf_conn *nf_conntrack_alloc(struct net *net, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl, + gfp_t gfp) { - struct nf_conn *ct = NULL; + struct nf_conn *ct; if (unlikely(!nf_conntrack_hash_rnd_initted)) { - get_random_bytes(&nf_conntrack_hash_rnd, 4); + get_random_bytes(&nf_conntrack_hash_rnd, + sizeof(nf_conntrack_hash_rnd)); nf_conntrack_hash_rnd_initted = 1; } /* We don't want any race condition at early drop stage */ - atomic_inc(&nf_conntrack_count); + atomic_inc(&net->ct.count); if (nf_conntrack_max && - unlikely(atomic_read(&nf_conntrack_count) > nf_conntrack_max)) { + unlikely(atomic_read(&net->ct.count) > nf_conntrack_max)) { unsigned int hash = hash_conntrack(orig); - if (!early_drop(hash)) { - atomic_dec(&nf_conntrack_count); + if (!early_drop(net, hash)) { + atomic_dec(&net->ct.count); if (net_ratelimit()) printk(KERN_WARNING "nf_conntrack: table full, dropping" @@ -494,43 +568,58 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, } } - ct = kmem_cache_zalloc(nf_conntrack_cachep, GFP_ATOMIC); + /* + * Do not use kmem_cache_zalloc(), as this cache uses + * SLAB_DESTROY_BY_RCU. + */ + ct = kmem_cache_alloc(nf_conntrack_cachep, gfp); if (ct == NULL) { pr_debug("nf_conntrack_alloc: Can't alloc conntrack.\n"); - atomic_dec(&nf_conntrack_count); + atomic_dec(&net->ct.count); return ERR_PTR(-ENOMEM); } - - atomic_set(&ct->ct_general.use, 1); + /* + * Let ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode.next + * and ct->tuplehash[IP_CT_DIR_REPLY].hnnode.next unchanged. + */ + memset(&ct->tuplehash[IP_CT_DIR_MAX], 0, + sizeof(*ct) - offsetof(struct nf_conn, tuplehash[IP_CT_DIR_MAX])); + spin_lock_init(&ct->lock); ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig; + ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode.pprev = NULL; ct->tuplehash[IP_CT_DIR_REPLY].tuple = *repl; + ct->tuplehash[IP_CT_DIR_REPLY].hnnode.pprev = NULL; /* Don't set timer yet: wait for confirmation */ setup_timer(&ct->timeout, death_by_timeout, (unsigned long)ct); - INIT_RCU_HEAD(&ct->rcu); +#ifdef CONFIG_NET_NS + ct->ct_net = net; +#endif + /* + * changes to lookup keys must be done before setting refcnt to 1 + */ + smp_wmb(); + atomic_set(&ct->ct_general.use, 1); return ct; } EXPORT_SYMBOL_GPL(nf_conntrack_alloc); -static void nf_conntrack_free_rcu(struct rcu_head *head) +void nf_conntrack_free(struct nf_conn *ct) { - struct nf_conn *ct = container_of(head, struct nf_conn, rcu); + struct net *net = nf_ct_net(ct); + nf_ct_ext_destroy(ct); + atomic_dec(&net->ct.count); nf_ct_ext_free(ct); kmem_cache_free(nf_conntrack_cachep, ct); - atomic_dec(&nf_conntrack_count); -} - -void nf_conntrack_free(struct nf_conn *ct) -{ - call_rcu(&ct->rcu, nf_conntrack_free_rcu); } 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, +init_conntrack(struct net *net, struct nf_conn *tmpl, + const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, struct nf_conntrack_l4proto *l4proto, struct sk_buff *skb, @@ -539,6 +628,7 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, struct nf_conn *ct; struct nf_conn_help *help; struct nf_conntrack_tuple repl_tuple; + struct nf_conntrack_ecache *ecache; struct nf_conntrack_expect *exp; if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, l4proto)) { @@ -546,8 +636,8 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, return NULL; } - ct = nf_conntrack_alloc(tuple, &repl_tuple); - if (ct == NULL || IS_ERR(ct)) { + ct = nf_conntrack_alloc(net, tuple, &repl_tuple, GFP_ATOMIC); + if (IS_ERR(ct)) { pr_debug("Can't allocate conntrack.\n"); return (struct nf_conntrack_tuple_hash *)ct; } @@ -558,8 +648,15 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, return NULL; } + nf_ct_acct_ext_add(ct, GFP_ATOMIC); + + ecache = tmpl ? nf_ct_ecache_find(tmpl) : NULL; + nf_ct_ecache_ext_add(ct, ecache ? ecache->ctmask : 0, + ecache ? ecache->expmask : 0, + GFP_ATOMIC); + spin_lock_bh(&nf_conntrack_lock); - exp = nf_ct_find_expectation(tuple); + exp = nf_ct_find_expectation(net, tuple); if (exp) { pr_debug("conntrack: expectation arrives ct=%p exp=%p\n", ct, exp); @@ -579,21 +676,15 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, ct->secmark = exp->master->secmark; #endif nf_conntrack_get(&ct->master->ct_general); - NF_CT_STAT_INC(expect_new); + NF_CT_STAT_INC(net, expect_new); } else { - struct nf_conntrack_helper *helper; - - helper = __nf_ct_helper_find(&repl_tuple); - if (helper) { - help = nf_ct_helper_ext_add(ct, GFP_ATOMIC); - if (help) - rcu_assign_pointer(help->helper, helper); - } - NF_CT_STAT_INC(new); + __nf_ct_try_assign_helper(ct, tmpl, GFP_ATOMIC); + NF_CT_STAT_INC(net, new); } /* Overload tuple linked list to put us in unconfirmed list. */ - hlist_add_head(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode, &unconfirmed); + hlist_nulls_add_head_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode, + &net->ct.unconfirmed); spin_unlock_bh(&nf_conntrack_lock); @@ -608,7 +699,8 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, /* On success, returns conntrack ptr, sets skb->nfct and ctinfo */ static inline struct nf_conn * -resolve_normal_ct(struct sk_buff *skb, +resolve_normal_ct(struct net *net, struct nf_conn *tmpl, + struct sk_buff *skb, unsigned int dataoff, u_int16_t l3num, u_int8_t protonum, @@ -629,9 +721,10 @@ resolve_normal_ct(struct sk_buff *skb, } /* look for tuple match */ - h = nf_conntrack_find_get(&tuple); + h = nf_conntrack_find_get(net, &tuple); if (!h) { - h = init_conntrack(&tuple, l3proto, l4proto, skb, dataoff); + h = init_conntrack(net, tmpl, &tuple, l3proto, l4proto, + skb, dataoff); if (!h) return NULL; if (IS_ERR(h)) @@ -665,9 +758,10 @@ resolve_normal_ct(struct sk_buff *skb, } unsigned int -nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff *skb) +nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, + struct sk_buff *skb) { - struct nf_conn *ct; + struct nf_conn *ct, *tmpl = NULL; enum ip_conntrack_info ctinfo; struct nf_conntrack_l3proto *l3proto; struct nf_conntrack_l4proto *l4proto; @@ -676,73 +770,89 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff *skb) int set_reply = 0; int ret; - /* Previously seen (loopback or untracked)? Ignore. */ if (skb->nfct) { - NF_CT_STAT_INC_ATOMIC(ignore); - return NF_ACCEPT; + /* Previously seen (loopback or untracked)? Ignore. */ + tmpl = (struct nf_conn *)skb->nfct; + if (!nf_ct_is_template(tmpl)) { + NF_CT_STAT_INC_ATOMIC(net, ignore); + return NF_ACCEPT; + } + skb->nfct = NULL; } /* rcu_read_lock()ed by nf_hook_slow */ - l3proto = __nf_ct_l3proto_find((u_int16_t)pf); + l3proto = __nf_ct_l3proto_find(pf); ret = l3proto->get_l4proto(skb, skb_network_offset(skb), &dataoff, &protonum); if (ret <= 0) { pr_debug("not prepared to track yet or error occured\n"); - NF_CT_STAT_INC_ATOMIC(error); - NF_CT_STAT_INC_ATOMIC(invalid); - return -ret; + NF_CT_STAT_INC_ATOMIC(net, error); + NF_CT_STAT_INC_ATOMIC(net, invalid); + ret = -ret; + goto out; } - l4proto = __nf_ct_l4proto_find((u_int16_t)pf, protonum); + l4proto = __nf_ct_l4proto_find(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 (l4proto->error != NULL && - (ret = l4proto->error(skb, dataoff, &ctinfo, pf, hooknum)) <= 0) { - NF_CT_STAT_INC_ATOMIC(error); - NF_CT_STAT_INC_ATOMIC(invalid); - return -ret; + if (l4proto->error != NULL) { + ret = l4proto->error(net, skb, dataoff, &ctinfo, pf, hooknum); + if (ret <= 0) { + NF_CT_STAT_INC_ATOMIC(net, error); + NF_CT_STAT_INC_ATOMIC(net, invalid); + ret = -ret; + goto out; + } } - ct = resolve_normal_ct(skb, dataoff, pf, protonum, l3proto, l4proto, - &set_reply, &ctinfo); + ct = resolve_normal_ct(net, tmpl, skb, dataoff, pf, protonum, + l3proto, l4proto, &set_reply, &ctinfo); if (!ct) { /* Not valid part of a connection */ - NF_CT_STAT_INC_ATOMIC(invalid); - return NF_ACCEPT; + NF_CT_STAT_INC_ATOMIC(net, invalid); + ret = NF_ACCEPT; + goto out; } if (IS_ERR(ct)) { /* Too stressed to deal. */ - NF_CT_STAT_INC_ATOMIC(drop); - return NF_DROP; + NF_CT_STAT_INC_ATOMIC(net, drop); + ret = NF_DROP; + goto out; } NF_CT_ASSERT(skb->nfct); ret = l4proto->packet(ct, skb, dataoff, ctinfo, pf, hooknum); - if (ret < 0) { + if (ret <= 0) { /* Invalid: inverse of the return code tells * the netfilter core what to do */ pr_debug("nf_conntrack_in: Can't track with proto module\n"); nf_conntrack_put(skb->nfct); skb->nfct = NULL; - NF_CT_STAT_INC_ATOMIC(invalid); - return -ret; + NF_CT_STAT_INC_ATOMIC(net, invalid); + if (ret == -NF_DROP) + NF_CT_STAT_INC_ATOMIC(net, drop); + ret = -ret; + goto out; } if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_STATUS, skb); + nf_conntrack_event_cache(IPCT_REPLY, ct); +out: + if (tmpl) + nf_ct_put(tmpl); return ret; } EXPORT_SYMBOL_GPL(nf_conntrack_in); -int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, - const struct nf_conntrack_tuple *orig) +bool nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig) { - int ret; + bool ret; rcu_read_lock(); ret = nf_ct_invert_tuple(inverse, orig, @@ -760,36 +870,19 @@ void nf_conntrack_alter_reply(struct nf_conn *ct, const struct nf_conntrack_tuple *newreply) { struct nf_conn_help *help = nfct_help(ct); - struct nf_conntrack_helper *helper; /* Should be unconfirmed, so not in hash table yet */ NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); pr_debug("Altering reply tuple of %p to ", ct); - NF_CT_DUMP_TUPLE(newreply); + nf_ct_dump_tuple(newreply); ct->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; - if (ct->master || (help && help->expecting != 0)) + if (ct->master || (help && !hlist_empty(&help->expectations))) return; rcu_read_lock(); - helper = __nf_ct_helper_find(newreply); - if (helper == NULL) { - if (help) - rcu_assign_pointer(help->helper, NULL); - goto out; - } - - if (help == NULL) { - help = nf_ct_helper_ext_add(ct, GFP_ATOMIC); - if (help == NULL) - goto out; - } else { - memset(&help->help, 0, sizeof(help->help)); - } - - rcu_assign_pointer(help->helper, helper); -out: + __nf_ct_try_assign_helper(ct, NULL, GFP_ATOMIC); rcu_read_unlock(); } EXPORT_SYMBOL_GPL(nf_conntrack_alter_reply); @@ -801,13 +894,9 @@ void __nf_ct_refresh_acct(struct nf_conn *ct, unsigned long extra_jiffies, int do_acct) { - int event = 0; - NF_CT_ASSERT(ct->timeout.data == (unsigned long)ct); NF_CT_ASSERT(skb); - spin_lock_bh(&nf_conntrack_lock); - /* Only update if this is not a fixed timeout */ if (test_bit(IPS_FIXED_TIMEOUT_BIT, &ct->status)) goto acct; @@ -815,41 +904,57 @@ void __nf_ct_refresh_acct(struct nf_conn *ct, /* 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 { 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; - } + if (newtime - ct->timeout.expires >= HZ) + mod_timer_pending(&ct->timeout, newtime); } acct: -#ifdef CONFIG_NF_CT_ACCT if (do_acct) { - ct->counters[CTINFO2DIR(ctinfo)].packets++; - ct->counters[CTINFO2DIR(ctinfo)].bytes += - skb->len - skb_network_offset(skb); - - if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000) - || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000)) - event |= IPCT_COUNTER_FILLING; + struct nf_conn_counter *acct; + + acct = nf_conn_acct_find(ct); + if (acct) { + spin_lock_bh(&ct->lock); + acct[CTINFO2DIR(ctinfo)].packets++; + acct[CTINFO2DIR(ctinfo)].bytes += + skb->len - skb_network_offset(skb); + spin_unlock_bh(&ct->lock); + } } -#endif +} +EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct); - spin_unlock_bh(&nf_conntrack_lock); +bool __nf_ct_kill_acct(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + const struct sk_buff *skb, + int do_acct) +{ + if (do_acct) { + struct nf_conn_counter *acct; + + acct = nf_conn_acct_find(ct); + if (acct) { + spin_lock_bh(&ct->lock); + acct[CTINFO2DIR(ctinfo)].packets++; + acct[CTINFO2DIR(ctinfo)].bytes += + skb->len - skb_network_offset(skb); + spin_unlock_bh(&ct->lock); + } + } - /* must be unlocked when calling event cache */ - if (event) - nf_conntrack_event_cache(event, skb); + if (del_timer(&ct->timeout)) { + ct->timeout.function((unsigned long)ct); + return true; + } + return false; } -EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct); +EXPORT_SYMBOL_GPL(__nf_ct_kill_acct); #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) @@ -890,6 +995,12 @@ int nf_ct_port_nlattr_to_tuple(struct nlattr *tb[], return 0; } EXPORT_SYMBOL_GPL(nf_ct_port_nlattr_to_tuple); + +int nf_ct_port_nlattr_tuple_size(void) +{ + return nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1); +} +EXPORT_SYMBOL_GPL(nf_ct_port_nlattr_tuple_size); #endif /* Used by ipt_REJECT and ip6t_REJECT. */ @@ -913,22 +1024,22 @@ static void nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb) /* Bring out ya dead! */ static struct nf_conn * -get_next_corpse(int (*iter)(struct nf_conn *i, void *data), +get_next_corpse(struct net *net, int (*iter)(struct nf_conn *i, void *data), void *data, unsigned int *bucket) { struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; - struct hlist_node *n; + struct hlist_nulls_node *n; spin_lock_bh(&nf_conntrack_lock); for (; *bucket < nf_conntrack_htable_size; (*bucket)++) { - hlist_for_each_entry(h, n, &nf_conntrack_hash[*bucket], hnode) { + hlist_nulls_for_each_entry(h, n, &net->ct.hash[*bucket], hnnode) { ct = nf_ct_tuplehash_to_ctrack(h); if (iter(ct, data)) goto found; } } - hlist_for_each_entry(h, n, &unconfirmed, hnode) { + hlist_nulls_for_each_entry(h, n, &net->ct.unconfirmed, hnnode) { ct = nf_ct_tuplehash_to_ctrack(h); if (iter(ct, data)) set_bit(IPS_DYING_BIT, &ct->status); @@ -941,13 +1052,14 @@ found: return ct; } -void -nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data) +void nf_ct_iterate_cleanup(struct net *net, + int (*iter)(struct nf_conn *i, void *data), + void *data) { struct nf_conn *ct; unsigned int bucket = 0; - while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) { + while ((ct = get_next_corpse(net, iter, data, &bucket)) != NULL) { /* Time to push up daises... */ if (del_timer(&ct->timeout)) death_by_timeout((unsigned long)ct); @@ -958,12 +1070,31 @@ nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data) } EXPORT_SYMBOL_GPL(nf_ct_iterate_cleanup); +struct __nf_ct_flush_report { + u32 pid; + int report; +}; + +static int kill_report(struct nf_conn *i, void *data) +{ + struct __nf_ct_flush_report *fr = (struct __nf_ct_flush_report *)data; + + /* If we fail to deliver the event, death_by_timeout() will retry */ + if (nf_conntrack_event_report(IPCT_DESTROY, i, + fr->pid, fr->report) < 0) + return 1; + + /* Avoid the delivery of the destroy event in death_by_timeout(). */ + set_bit(IPS_DYING_BIT, &i->status); + return 1; +} + static int kill_all(struct nf_conn *i, void *data) { return 1; } -void nf_ct_free_hashtable(struct hlist_head *hash, int vmalloced, unsigned int size) +void nf_ct_free_hashtable(void *hash, int vmalloced, unsigned int size) { if (vmalloced) vfree(hash); @@ -973,27 +1104,44 @@ void nf_ct_free_hashtable(struct hlist_head *hash, int vmalloced, unsigned int s } EXPORT_SYMBOL_GPL(nf_ct_free_hashtable); -void nf_conntrack_flush(void) +void nf_conntrack_flush_report(struct net *net, u32 pid, int report) { - nf_ct_iterate_cleanup(kill_all, NULL); + struct __nf_ct_flush_report fr = { + .pid = pid, + .report = report, + }; + nf_ct_iterate_cleanup(net, kill_report, &fr); } -EXPORT_SYMBOL_GPL(nf_conntrack_flush); +EXPORT_SYMBOL_GPL(nf_conntrack_flush_report); -/* Mishearing the voices in his head, our hero wonders how he's - supposed to kill the mall. */ -void nf_conntrack_cleanup(void) +static void nf_ct_release_dying_list(struct net *net) { - rcu_assign_pointer(ip_ct_attach, NULL); + struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct; + struct hlist_nulls_node *n; - /* This makes sure all current packets have passed through - netfilter framework. Roll on, two-stage module - delete... */ - synchronize_net(); + spin_lock_bh(&nf_conntrack_lock); + hlist_nulls_for_each_entry(h, n, &net->ct.dying, hnnode) { + ct = nf_ct_tuplehash_to_ctrack(h); + /* never fails to remove them, no listeners at this point */ + nf_ct_kill(ct); + } + spin_unlock_bh(&nf_conntrack_lock); +} + +static void nf_conntrack_cleanup_init_net(void) +{ + nf_conntrack_helper_fini(); + nf_conntrack_proto_fini(); + kmem_cache_destroy(nf_conntrack_cachep); +} - nf_ct_event_cache_flush(); +static void nf_conntrack_cleanup_net(struct net *net) +{ i_see_dead_people: - nf_conntrack_flush(); - if (atomic_read(&nf_conntrack_count) != 0) { + nf_ct_iterate_cleanup(net, kill_all, NULL); + nf_ct_release_dying_list(net); + if (atomic_read(&net->ct.count) != 0) { schedule(); goto i_see_dead_people; } @@ -1001,37 +1149,56 @@ void nf_conntrack_cleanup(void) while (atomic_read(&nf_conntrack_untracked.ct_general.use) > 1) schedule(); - rcu_assign_pointer(nf_ct_destroy, NULL); - - kmem_cache_destroy(nf_conntrack_cachep); - nf_ct_free_hashtable(nf_conntrack_hash, nf_conntrack_vmalloc, + nf_ct_free_hashtable(net->ct.hash, net->ct.hash_vmalloc, nf_conntrack_htable_size); + nf_conntrack_ecache_fini(net); + nf_conntrack_acct_fini(net); + nf_conntrack_expect_fini(net); + free_percpu(net->ct.stat); +} - nf_conntrack_proto_fini(); - nf_conntrack_helper_fini(); - nf_conntrack_expect_fini(); +/* Mishearing the voices in his head, our hero wonders how he's + supposed to kill the mall. */ +void nf_conntrack_cleanup(struct net *net) +{ + if (net_eq(net, &init_net)) + rcu_assign_pointer(ip_ct_attach, NULL); + + /* This makes sure all current packets have passed through + netfilter framework. Roll on, two-stage module + delete... */ + synchronize_net(); + + nf_conntrack_cleanup_net(net); + + if (net_eq(net, &init_net)) { + rcu_assign_pointer(nf_ct_destroy, NULL); + nf_conntrack_cleanup_init_net(); + } } -struct hlist_head *nf_ct_alloc_hashtable(unsigned int *sizep, int *vmalloced) +void *nf_ct_alloc_hashtable(unsigned int *sizep, int *vmalloced, int nulls) { - struct hlist_head *hash; - unsigned int size, i; + struct hlist_nulls_head *hash; + unsigned int nr_slots, i; + size_t sz; *vmalloced = 0; - size = *sizep = roundup(*sizep, PAGE_SIZE / sizeof(struct hlist_head)); - hash = (void*)__get_free_pages(GFP_KERNEL|__GFP_NOWARN, - get_order(sizeof(struct hlist_head) - * size)); + BUILD_BUG_ON(sizeof(struct hlist_nulls_head) != sizeof(struct hlist_head)); + nr_slots = *sizep = roundup(*sizep, PAGE_SIZE / sizeof(struct hlist_nulls_head)); + sz = nr_slots * sizeof(struct hlist_nulls_head); + hash = (void *)__get_free_pages(GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO, + get_order(sz)); if (!hash) { *vmalloced = 1; printk(KERN_WARNING "nf_conntrack: falling back to vmalloc.\n"); - hash = vmalloc(sizeof(struct hlist_head) * size); + hash = __vmalloc(sz, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL); } - if (hash) - for (i = 0; i < size; i++) - INIT_HLIST_HEAD(&hash[i]); + if (hash && nulls) + for (i = 0; i < nr_slots; i++) + INIT_HLIST_NULLS_HEAD(&hash[i], i); return hash; } @@ -1042,7 +1209,7 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) int i, bucket, vmalloced, old_vmalloced; unsigned int hashsize, old_size; int rnd; - struct hlist_head *hash, *old_hash; + struct hlist_nulls_head *hash, *old_hash; struct nf_conntrack_tuple_hash *h; /* On boot, we can set this without any fancy locking. */ @@ -1053,13 +1220,13 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) if (!hashsize) return -EINVAL; - hash = nf_ct_alloc_hashtable(&hashsize, &vmalloced); + hash = nf_ct_alloc_hashtable(&hashsize, &vmalloced, 1); if (!hash) return -ENOMEM; /* We have to rehahs for the new table anyway, so we also can * use a newrandom seed */ - get_random_bytes(&rnd, 4); + get_random_bytes(&rnd, sizeof(rnd)); /* Lookups in the old hash might happen in parallel, which means we * might get false negatives during connection lookup. New connections @@ -1068,21 +1235,21 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) */ spin_lock_bh(&nf_conntrack_lock); for (i = 0; i < nf_conntrack_htable_size; i++) { - while (!hlist_empty(&nf_conntrack_hash[i])) { - h = hlist_entry(nf_conntrack_hash[i].first, - struct nf_conntrack_tuple_hash, hnode); - hlist_del_rcu(&h->hnode); + while (!hlist_nulls_empty(&init_net.ct.hash[i])) { + h = hlist_nulls_entry(init_net.ct.hash[i].first, + struct nf_conntrack_tuple_hash, hnnode); + hlist_nulls_del_rcu(&h->hnnode); bucket = __hash_conntrack(&h->tuple, hashsize, rnd); - hlist_add_head(&h->hnode, &hash[bucket]); + hlist_nulls_add_head_rcu(&h->hnnode, &hash[bucket]); } } old_size = nf_conntrack_htable_size; - old_vmalloced = nf_conntrack_vmalloc; - old_hash = nf_conntrack_hash; + old_vmalloced = init_net.ct.hash_vmalloc; + old_hash = init_net.ct.hash; nf_conntrack_htable_size = hashsize; - nf_conntrack_vmalloc = vmalloced; - nf_conntrack_hash = hash; + init_net.ct.hash_vmalloc = vmalloced; + init_net.ct.hash = hash; nf_conntrack_hash_rnd = rnd; spin_unlock_bh(&nf_conntrack_lock); @@ -1094,7 +1261,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_set_hashsize); module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint, &nf_conntrack_htable_size, 0600); -int __init nf_conntrack_init(void) +static int nf_conntrack_init_init_net(void) { int max_factor = 8; int ret; @@ -1103,9 +1270,9 @@ int __init nf_conntrack_init(void) * machine has 512 buckets. >= 1GB machines have 16384 buckets. */ if (!nf_conntrack_htable_size) { nf_conntrack_htable_size - = (((num_physpages << PAGE_SHIFT) / 16384) + = (((totalram_pages << PAGE_SHIFT) / 16384) / sizeof(struct hlist_head)); - if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) + if (totalram_pages > (1024 * 1024 * 1024 / PAGE_SIZE)) nf_conntrack_htable_size = 16384; if (nf_conntrack_htable_size < 32) nf_conntrack_htable_size = 32; @@ -1116,13 +1283,6 @@ int __init nf_conntrack_init(void) * entries. */ max_factor = 4; } - nf_conntrack_hash = nf_ct_alloc_hashtable(&nf_conntrack_htable_size, - &nf_conntrack_vmalloc); - if (!nf_conntrack_hash) { - printk(KERN_ERR "Unable to create nf_conntrack_hash\n"); - goto err_out; - } - nf_conntrack_max = max_factor * nf_conntrack_htable_size; printk("nf_conntrack version %s (%u buckets, %d max)\n", @@ -1131,45 +1291,121 @@ int __init nf_conntrack_init(void) nf_conntrack_cachep = kmem_cache_create("nf_conntrack", sizeof(struct nf_conn), - 0, 0, NULL); + 0, SLAB_DESTROY_BY_RCU, NULL); if (!nf_conntrack_cachep) { printk(KERN_ERR "Unable to create nf_conn slab cache\n"); - goto err_free_hash; + ret = -ENOMEM; + goto err_cache; } ret = nf_conntrack_proto_init(); if (ret < 0) - goto err_free_conntrack_slab; - - ret = nf_conntrack_expect_init(); - if (ret < 0) - goto out_fini_proto; + goto err_proto; ret = nf_conntrack_helper_init(); if (ret < 0) - goto out_fini_expect; + goto err_helper; - /* For use by REJECT target */ - rcu_assign_pointer(ip_ct_attach, nf_conntrack_attach); - rcu_assign_pointer(nf_ct_destroy, destroy_conntrack); + return 0; + +err_helper: + nf_conntrack_proto_fini(); +err_proto: + kmem_cache_destroy(nf_conntrack_cachep); +err_cache: + return ret; +} + +/* + * We need to use special "null" values, not used in hash table + */ +#define UNCONFIRMED_NULLS_VAL ((1<<30)+0) +#define DYING_NULLS_VAL ((1<<30)+1) + +static int nf_conntrack_init_net(struct net *net) +{ + int ret; + + atomic_set(&net->ct.count, 0); + INIT_HLIST_NULLS_HEAD(&net->ct.unconfirmed, UNCONFIRMED_NULLS_VAL); + INIT_HLIST_NULLS_HEAD(&net->ct.dying, DYING_NULLS_VAL); + net->ct.stat = alloc_percpu(struct ip_conntrack_stat); + if (!net->ct.stat) { + ret = -ENOMEM; + goto err_stat; + } + net->ct.hash = nf_ct_alloc_hashtable(&nf_conntrack_htable_size, + &net->ct.hash_vmalloc, 1); + if (!net->ct.hash) { + ret = -ENOMEM; + printk(KERN_ERR "Unable to create nf_conntrack_hash\n"); + goto err_hash; + } + ret = nf_conntrack_expect_init(net); + if (ret < 0) + goto err_expect; + ret = nf_conntrack_acct_init(net); + if (ret < 0) + goto err_acct; + ret = nf_conntrack_ecache_init(net); + if (ret < 0) + goto err_ecache; /* Set up fake conntrack: - to never be deleted, not in any hashes */ +#ifdef CONFIG_NET_NS + nf_conntrack_untracked.ct_net = &init_net; +#endif atomic_set(&nf_conntrack_untracked.ct_general.use, 1); /* - and look it like as a confirmed connection */ set_bit(IPS_CONFIRMED_BIT, &nf_conntrack_untracked.status); - return ret; + return 0; -out_fini_expect: - nf_conntrack_expect_fini(); -out_fini_proto: - nf_conntrack_proto_fini(); -err_free_conntrack_slab: - kmem_cache_destroy(nf_conntrack_cachep); -err_free_hash: - nf_ct_free_hashtable(nf_conntrack_hash, nf_conntrack_vmalloc, +err_ecache: + nf_conntrack_acct_fini(net); +err_acct: + nf_conntrack_expect_fini(net); +err_expect: + nf_ct_free_hashtable(net->ct.hash, net->ct.hash_vmalloc, nf_conntrack_htable_size); -err_out: - return -ENOMEM; +err_hash: + free_percpu(net->ct.stat); +err_stat: + return ret; +} + +s16 (*nf_ct_nat_offset)(const struct nf_conn *ct, + enum ip_conntrack_dir dir, + u32 seq); +EXPORT_SYMBOL_GPL(nf_ct_nat_offset); + +int nf_conntrack_init(struct net *net) +{ + int ret; + + if (net_eq(net, &init_net)) { + ret = nf_conntrack_init_init_net(); + if (ret < 0) + goto out_init_net; + } + ret = nf_conntrack_init_net(net); + if (ret < 0) + goto out_net; + + if (net_eq(net, &init_net)) { + /* For use by REJECT target */ + rcu_assign_pointer(ip_ct_attach, nf_conntrack_attach); + rcu_assign_pointer(nf_ct_destroy, destroy_conntrack); + + /* Howto get NAT offsets */ + rcu_assign_pointer(nf_ct_nat_offset, NULL); + } + return 0; + +out_net: + if (net_eq(net, &init_net)) + nf_conntrack_cleanup_init_net(); +out_init_net: + return ret; }