#include <net/netfilter/nf_conntrack_helper.h>
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/nf_conntrack_extend.h>
+#include <net/netfilter/nf_conntrack_acct.h>
#define NF_CONNTRACK_VERSION "0.5.0"
-DEFINE_RWLOCK(nf_conntrack_lock);
+DEFINE_SPINLOCK(nf_conntrack_lock);
EXPORT_SYMBOL_GPL(nf_conntrack_lock);
/* nf_conntrack_standalone needs this */
static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple,
unsigned int size, unsigned int rnd)
{
- unsigned int a, b;
+ unsigned int n;
+ u_int32_t h;
- a = jhash2(tuple->src.u3.all, ARRAY_SIZE(tuple->src.u3.all),
- (tuple->src.l3num << 16) | tuple->dst.protonum);
- b = jhash2(tuple->dst.u3.all, ARRAY_SIZE(tuple->dst.u3.all),
- ((__force __u16)tuple->src.u.all << 16) |
- (__force __u16)tuple->dst.u.all);
+ /* The direction must be ignored, so we hash everything up to the
+ * destination ports (which is a multiple of 4) and treat the last
+ * three bytes manually.
+ */
+ n = (sizeof(tuple->src) + sizeof(tuple->dst.u3)) / sizeof(u32);
+ h = jhash2((u32 *)tuple, n,
+ rnd ^ (((__force __u16)tuple->dst.u.all << 16) |
+ tuple->dst.protonum));
- return ((u64)jhash_2words(a, b, rnd) * size) >> 32;
+ return ((u64)h * size) >> 32;
}
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,
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;
}
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;
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);
}
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;
* 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();
- write_lock_bh(&nf_conntrack_lock);
+ spin_lock_bh(&nf_conntrack_lock);
/* Expectations will have been removed in clean_from_lists,
* except TFTP can create an expectation on the first packet,
* before connection is in the list, so we need to clean here,
}
NF_CT_STAT_INC(delete);
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
if (ct->master)
nf_ct_put(ct->master);
rcu_read_unlock();
}
- write_lock_bh(&nf_conntrack_lock);
+ 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);
clean_from_lists(ct);
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
nf_ct_put(ct);
}
struct nf_conntrack_tuple_hash *
-__nf_conntrack_find(const struct nf_conntrack_tuple *tuple,
- const struct nf_conn *ignored_conntrack)
+__nf_conntrack_find(const struct nf_conntrack_tuple *tuple)
{
struct nf_conntrack_tuple_hash *h;
struct hlist_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) {
- if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack &&
- nf_ct_tuple_equal(tuple, &h->tuple)) {
+ if (nf_ct_tuple_equal(tuple, &h->tuple)) {
NF_CT_STAT_INC(found);
+ local_bh_enable();
return h;
}
NF_CT_STAT_INC(searched);
}
+ local_bh_enable();
return NULL;
}
struct nf_conn *ct;
rcu_read_lock();
- h = __nf_conntrack_find(tuple, NULL);
+ h = __nf_conntrack_find(tuple);
if (h) {
ct = nf_ct_tuplehash_to_ctrack(h);
if (unlikely(!atomic_inc_not_zero(&ct->ct_general.use)))
hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
- write_lock_bh(&nf_conntrack_lock);
+ spin_lock_bh(&nf_conntrack_lock);
__nf_conntrack_hash_insert(ct, hash, repl_hash);
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
}
EXPORT_SYMBOL_GPL(nf_conntrack_hash_insert);
NF_CT_ASSERT(!nf_ct_is_confirmed(ct));
pr_debug("Confirming conntrack %p\n", ct);
- write_lock_bh(&nf_conntrack_lock);
+ spin_lock_bh(&nf_conntrack_lock);
/* See if there's one in the list already, including reverse:
NAT could have grabbed it without realizing, since we're
atomic_inc(&ct->ct_general.use);
set_bit(IPS_CONFIRMED_BIT, &ct->status);
NF_CT_STAT_INC(insert);
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
help = nfct_help(ct);
if (help && help->helper)
nf_conntrack_event_cache(IPCT_HELPER, skb);
out:
NF_CT_STAT_INC(insert_failed);
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
return NF_DROP;
}
EXPORT_SYMBOL_GPL(__nf_conntrack_confirm);
const struct nf_conn *ignored_conntrack)
{
struct nf_conntrack_tuple_hash *h;
+ struct hlist_node *n;
+ unsigned int hash = hash_conntrack(tuple);
- rcu_read_lock();
- h = __nf_conntrack_find(tuple, ignored_conntrack);
- rcu_read_unlock();
+ /* 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) {
+ if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack &&
+ nf_ct_tuple_equal(tuple, &h->tuple)) {
+ NF_CT_STAT_INC(found);
+ rcu_read_unlock_bh();
+ return 1;
+ }
+ NF_CT_STAT_INC(searched);
+ }
+ rcu_read_unlock_bh();
- return h != NULL;
+ return 0;
}
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 int early_drop(unsigned int hash)
+static noinline int early_drop(unsigned int hash)
{
/* Use oldest entry, which is roughly LRU */
struct nf_conntrack_tuple_hash *h;
}
struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig,
- const struct nf_conntrack_tuple *repl)
+ const struct nf_conntrack_tuple *repl,
+ gfp_t gfp)
{
- struct nf_conn *conntrack = NULL;
+ struct nf_conn *ct = NULL;
if (unlikely(!nf_conntrack_hash_rnd_initted)) {
get_random_bytes(&nf_conntrack_hash_rnd, 4);
/* 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) {
+ if (nf_conntrack_max &&
+ unlikely(atomic_read(&nf_conntrack_count) > nf_conntrack_max)) {
unsigned int hash = hash_conntrack(orig);
if (!early_drop(hash)) {
atomic_dec(&nf_conntrack_count);
}
}
- conntrack = kmem_cache_zalloc(nf_conntrack_cachep, GFP_ATOMIC);
- if (conntrack == NULL) {
+ ct = kmem_cache_zalloc(nf_conntrack_cachep, gfp);
+ if (ct == NULL) {
pr_debug("nf_conntrack_alloc: Can't alloc conntrack.\n");
atomic_dec(&nf_conntrack_count);
return ERR_PTR(-ENOMEM);
}
- atomic_set(&conntrack->ct_general.use, 1);
- conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig;
- conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *repl;
+ atomic_set(&ct->ct_general.use, 1);
+ ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig;
+ ct->tuplehash[IP_CT_DIR_REPLY].tuple = *repl;
/* Don't set timer yet: wait for confirmation */
- setup_timer(&conntrack->timeout, death_by_timeout,
- (unsigned long)conntrack);
- INIT_RCU_HEAD(&conntrack->rcu);
+ setup_timer(&ct->timeout, death_by_timeout, (unsigned long)ct);
+ INIT_RCU_HEAD(&ct->rcu);
- return conntrack;
+ return ct;
}
EXPORT_SYMBOL_GPL(nf_conntrack_alloc);
atomic_dec(&nf_conntrack_count);
}
-void nf_conntrack_free(struct nf_conn *conntrack)
+void nf_conntrack_free(struct nf_conn *ct)
{
- call_rcu(&conntrack->rcu, nf_conntrack_free_rcu);
+ nf_ct_ext_destroy(ct);
+ call_rcu(&ct->rcu, nf_conntrack_free_rcu);
}
EXPORT_SYMBOL_GPL(nf_conntrack_free);
struct sk_buff *skb,
unsigned int dataoff)
{
- struct nf_conn *conntrack;
+ struct nf_conn *ct;
struct nf_conn_help *help;
struct nf_conntrack_tuple repl_tuple;
struct nf_conntrack_expect *exp;
return NULL;
}
- conntrack = nf_conntrack_alloc(tuple, &repl_tuple);
- if (conntrack == NULL || IS_ERR(conntrack)) {
+ ct = nf_conntrack_alloc(tuple, &repl_tuple, GFP_ATOMIC);
+ if (ct == NULL || IS_ERR(ct)) {
pr_debug("Can't allocate conntrack.\n");
- return (struct nf_conntrack_tuple_hash *)conntrack;
+ return (struct nf_conntrack_tuple_hash *)ct;
}
- if (!l4proto->new(conntrack, skb, dataoff)) {
- nf_conntrack_free(conntrack);
+ if (!l4proto->new(ct, skb, dataoff)) {
+ nf_conntrack_free(ct);
pr_debug("init conntrack: can't track with proto module\n");
return NULL;
}
- write_lock_bh(&nf_conntrack_lock);
+ nf_ct_acct_ext_add(ct, GFP_ATOMIC);
+
+ spin_lock_bh(&nf_conntrack_lock);
exp = nf_ct_find_expectation(tuple);
if (exp) {
pr_debug("conntrack: expectation arrives ct=%p exp=%p\n",
- conntrack, exp);
+ ct, exp);
/* Welcome, Mr. Bond. We've been expecting you... */
- __set_bit(IPS_EXPECTED_BIT, &conntrack->status);
- conntrack->master = exp->master;
+ __set_bit(IPS_EXPECTED_BIT, &ct->status);
+ ct->master = exp->master;
if (exp->helper) {
- help = nf_ct_helper_ext_add(conntrack, GFP_ATOMIC);
+ help = nf_ct_helper_ext_add(ct, GFP_ATOMIC);
if (help)
rcu_assign_pointer(help->helper, exp->helper);
}
#ifdef CONFIG_NF_CONNTRACK_MARK
- conntrack->mark = exp->master->mark;
+ ct->mark = exp->master->mark;
#endif
#ifdef CONFIG_NF_CONNTRACK_SECMARK
- conntrack->secmark = exp->master->secmark;
+ ct->secmark = exp->master->secmark;
#endif
- nf_conntrack_get(&conntrack->master->ct_general);
+ nf_conntrack_get(&ct->master->ct_general);
NF_CT_STAT_INC(expect_new);
} else {
struct nf_conntrack_helper *helper;
helper = __nf_ct_helper_find(&repl_tuple);
if (helper) {
- help = nf_ct_helper_ext_add(conntrack, GFP_ATOMIC);
+ help = nf_ct_helper_ext_add(ct, GFP_ATOMIC);
if (help)
rcu_assign_pointer(help->helper, helper);
}
}
/* Overload tuple linked list to put us in unconfirmed list. */
- hlist_add_head(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].hnode,
- &unconfirmed);
+ hlist_add_head(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode, &unconfirmed);
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
if (exp) {
if (exp->expectfn)
- exp->expectfn(conntrack, exp);
+ exp->expectfn(ct, exp);
nf_ct_expect_put(exp);
}
- return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL];
+ return &ct->tuplehash[IP_CT_DIR_ORIGINAL];
}
/* On success, returns conntrack ptr, sets skb->nfct and ctinfo */
}
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,
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();
NF_CT_ASSERT(ct->timeout.data == (unsigned long)ct);
NF_CT_ASSERT(skb);
- write_lock_bh(&nf_conntrack_lock);
+ spin_lock_bh(&nf_conntrack_lock);
/* Only update if this is not a fixed timeout */
if (test_bit(IPS_FIXED_TIMEOUT_BIT, &ct->status))
}
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);
+ struct nf_conn_counter *acct;
- if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000)
- || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000))
- event |= IPCT_COUNTER_FILLING;
+ acct = nf_conn_acct_find(ct);
+ if (acct) {
+ acct[CTINFO2DIR(ctinfo)].packets++;
+ acct[CTINFO2DIR(ctinfo)].bytes +=
+ skb->len - skb_network_offset(skb);
+ }
}
-#endif
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
/* must be unlocked when calling event cache */
if (event)
}
EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct);
+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;
+
+ spin_lock_bh(&nf_conntrack_lock);
+ acct = nf_conn_acct_find(ct);
+ if (acct) {
+ acct[CTINFO2DIR(ctinfo)].packets++;
+ acct[CTINFO2DIR(ctinfo)].bytes +=
+ skb->len - skb_network_offset(skb);
+ }
+ spin_unlock_bh(&nf_conntrack_lock);
+ }
+
+ if (del_timer(&ct->timeout)) {
+ ct->timeout.function((unsigned long)ct);
+ return true;
+ }
+ return false;
+}
+EXPORT_SYMBOL_GPL(__nf_ct_kill_acct);
+
#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
#include <linux/netfilter/nfnetlink.h>
nf_conntrack_get(nskb->nfct);
}
-static inline int
-do_iter(const struct nf_conntrack_tuple_hash *i,
- int (*iter)(struct nf_conn *i, void *data),
- void *data)
-{
- return iter(nf_ct_tuplehash_to_ctrack(i), data);
-}
-
/* Bring out ya dead! */
static struct nf_conn *
get_next_corpse(int (*iter)(struct nf_conn *i, void *data),
struct nf_conn *ct;
struct hlist_node *n;
- write_lock_bh(&nf_conntrack_lock);
+ spin_lock_bh(&nf_conntrack_lock);
for (; *bucket < nf_conntrack_htable_size; (*bucket)++) {
hlist_for_each_entry(h, n, &nf_conntrack_hash[*bucket], hnode) {
ct = nf_ct_tuplehash_to_ctrack(h);
if (iter(ct, data))
set_bit(IPS_DYING_BIT, &ct->status);
}
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
return NULL;
found:
atomic_inc(&ct->ct_general.use);
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
return ct;
}
nf_ct_free_hashtable(nf_conntrack_hash, nf_conntrack_vmalloc,
nf_conntrack_htable_size);
- nf_conntrack_proto_fini();
- nf_conntrack_helper_fini();
+ nf_conntrack_acct_fini();
nf_conntrack_expect_fini();
+ nf_conntrack_helper_fini();
+ nf_conntrack_proto_fini();
}
struct hlist_head *nf_ct_alloc_hashtable(unsigned int *sizep, int *vmalloced)
* created because of a false negative won't make it into the hash
* though since that required taking the lock.
*/
- write_lock_bh(&nf_conntrack_lock);
+ 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,
nf_conntrack_vmalloc = vmalloced;
nf_conntrack_hash = hash;
nf_conntrack_hash_rnd = rnd;
- write_unlock_bh(&nf_conntrack_lock);
+ spin_unlock_bh(&nf_conntrack_lock);
nf_ct_free_hashtable(old_hash, old_vmalloced, old_size);
return 0;
if (ret < 0)
goto out_fini_expect;
+ ret = nf_conntrack_acct_init();
+ if (ret < 0)
+ goto out_fini_helper;
+
/* For use by REJECT target */
rcu_assign_pointer(ip_ct_attach, nf_conntrack_attach);
rcu_assign_pointer(nf_ct_destroy, destroy_conntrack);
return ret;
+out_fini_helper:
+ nf_conntrack_helper_fini();
out_fini_expect:
nf_conntrack_expect_fini();
out_fini_proto:
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff