* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
- *
- * 23 Apr 2001: Harald Welte <laforge@gnumonks.org>
- * - new API and handling of conntrack/nat helpers
- * - now capable of multiple expectations for one master
- * 16 Jul 2002: Harald Welte <laforge@gnumonks.org>
- * - add usage/reference counts to ip_conntrack_expect
- * - export ip_conntrack[_expect]_{find_get,put} functions
- * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp>
- * - generalize L3 protocol denendent part.
- * 23 Mar 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp>
- * - add support various size of conntrack structures.
- * 26 Jan 2006: Harald Welte <laforge@netfilter.org>
- * - restructure nf_conn (introduce nf_conn_help)
- * - redesign 'features' how they were originally intended
- * 26 Feb 2006: Pablo Neira Ayuso <pablo@eurodev.net>
- * - add support for L3 protocol module load on demand.
- *
- * Derived from net/ipv4/netfilter/ip_conntrack_core.c
*/
#include <linux/types.h>
unsigned int size, unsigned int rnd)
{
unsigned int a, b;
- a = jhash((void *)tuple->src.u3.all, sizeof(tuple->src.u3.all),
- ((tuple->src.l3num) << 16) | tuple->dst.protonum);
- b = jhash((void *)tuple->dst.u3.all, sizeof(tuple->dst.u3.all),
- (tuple->src.u.all << 16) | tuple->dst.u.all);
+
+ 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),
+ (tuple->src.u.all << 16) | tuple->dst.u.all);
return jhash_2words(a, b, rnd) % size;
}
struct nf_conn_help *help = nfct_help(ct);
struct nf_conntrack_l3proto *l3proto;
struct nf_conntrack_l4proto *l4proto;
+ typeof(nf_conntrack_destroyed) destroyed;
DEBUGP("destroy_conntrack(%p)\n", ct);
NF_CT_ASSERT(atomic_read(&nfct->use) == 0);
/* 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();
l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num);
if (l3proto && l3proto->destroy)
l3proto->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);
+ 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);
+ destroyed = rcu_dereference(nf_conntrack_destroyed);
+ if (destroyed)
+ destroyed(ct);
+
+ rcu_read_unlock();
write_lock_bh(&nf_conntrack_lock);
/* Expectations will have been removed in clean_from_lists,
static void __nf_conntrack_hash_insert(struct nf_conn *ct,
unsigned int hash,
- unsigned int repl_hash)
+ unsigned int repl_hash)
{
ct->id = ++nf_conntrack_next_id;
list_add(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list,
if (del_timer(&ct->timeout)) {
death_by_timeout((unsigned long)ct);
dropped = 1;
- NF_CT_STAT_INC(early_drop);
+ NF_CT_STAT_INC_ATOMIC(early_drop);
}
nf_ct_put(ct);
return dropped;
memset(conntrack, 0, nf_ct_cache[features].size);
conntrack->features = features;
atomic_set(&conntrack->ct_general.use, 1);
- conntrack->ct_general.destroy = destroy_conntrack;
conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig;
conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *repl;
/* Don't set timer yet: wait for confirmation */
- init_timer(&conntrack->timeout);
- conntrack->timeout.data = (unsigned long)conntrack;
- conntrack->timeout.function = death_by_timeout;
+ setup_timer(&conntrack->timeout, death_by_timeout,
+ (unsigned long)conntrack);
read_unlock_bh(&nf_ct_cache_lock);
return conntrack;
const struct nf_conntrack_tuple *repl)
{
struct nf_conntrack_l3proto *l3proto;
+ struct nf_conn *ct;
+ rcu_read_lock();
l3proto = __nf_ct_l3proto_find(orig->src.l3num);
- return __nf_conntrack_alloc(orig, repl, l3proto, 0);
+ ct = __nf_conntrack_alloc(orig, repl, l3proto, 0);
+ rcu_read_unlock();
+
+ return ct;
}
EXPORT_SYMBOL_GPL(nf_conntrack_alloc);
struct nf_conntrack_tuple_hash *h;
struct nf_conn *ct;
- if (!nf_ct_get_tuple(skb, (unsigned int)(skb->nh.raw - skb->data),
+ if (!nf_ct_get_tuple(skb, skb_network_offset(skb),
dataoff, l3num, protonum, &tuple, l3proto,
l4proto)) {
DEBUGP("resolve_normal_ct: Can't get tuple\n");
/* Previously seen (loopback or untracked)? Ignore. */
if ((*pskb)->nfct) {
- NF_CT_STAT_INC(ignore);
+ NF_CT_STAT_INC_ATOMIC(ignore);
return NF_ACCEPT;
}
+ /* rcu_read_lock()ed by nf_hook_slow */
l3proto = __nf_ct_l3proto_find((u_int16_t)pf);
+
if ((ret = l3proto->prepare(pskb, hooknum, &dataoff, &protonum)) <= 0) {
DEBUGP("not prepared to track yet or error occured\n");
return -ret;
* core what to do with the packet. */
if (l4proto->error != NULL &&
(ret = l4proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) {
- NF_CT_STAT_INC(error);
- NF_CT_STAT_INC(invalid);
+ NF_CT_STAT_INC_ATOMIC(error);
+ NF_CT_STAT_INC_ATOMIC(invalid);
return -ret;
}
&set_reply, &ctinfo);
if (!ct) {
/* Not valid part of a connection */
- NF_CT_STAT_INC(invalid);
+ NF_CT_STAT_INC_ATOMIC(invalid);
return NF_ACCEPT;
}
if (IS_ERR(ct)) {
/* Too stressed to deal. */
- NF_CT_STAT_INC(drop);
+ NF_CT_STAT_INC_ATOMIC(drop);
return NF_DROP;
}
DEBUGP("nf_conntrack_in: Can't track with proto module\n");
nf_conntrack_put((*pskb)->nfct);
(*pskb)->nfct = NULL;
- NF_CT_STAT_INC(invalid);
+ NF_CT_STAT_INC_ATOMIC(invalid);
return -ret;
}
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_l4proto_find(orig->src.l3num,
- orig->dst.protonum));
+ int ret;
+
+ rcu_read_lock();
+ ret = nf_ct_invert_tuple(inverse, orig,
+ __nf_ct_l3proto_find(orig->src.l3num),
+ __nf_ct_l4proto_find(orig->src.l3num,
+ orig->dst.protonum));
+ rcu_read_unlock();
+ return ret;
}
EXPORT_SYMBOL_GPL(nf_ct_invert_tuplepr);
if (do_acct) {
ct->counters[CTINFO2DIR(ctinfo)].packets++;
ct->counters[CTINFO2DIR(ctinfo)].bytes +=
- skb->len - (unsigned int)(skb->nh.raw - skb->data);
+ skb->len - skb_network_offset(skb);
if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000)
|| (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000))
}
EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct);
-#if defined(CONFIG_NF_CT_NETLINK) || \
- defined(CONFIG_NF_CT_NETLINK_MODULE)
+#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
#include <linux/netfilter/nfnetlink.h>
#include <linux/netfilter/nfnetlink_conntrack.h>
if (iter(ct, data))
goto found;
}
- }
+ }
list_for_each_entry(h, &unconfirmed, list) {
ct = nf_ct_tuplehash_to_ctrack(h);
if (iter(ct, data))
- goto found;
+ set_bit(IPS_DYING_BIT, &ct->status);
}
write_unlock_bh(&nf_conntrack_lock);
return NULL;
if (vmalloced)
vfree(hash);
else
- free_pages((unsigned long)hash,
+ free_pages((unsigned long)hash,
get_order(sizeof(struct list_head) * size));
}
{
int i;
- ip_ct_attach = NULL;
+ rcu_assign_pointer(ip_ct_attach, NULL);
/* This makes sure all current packets have passed through
netfilter framework. Roll on, two-stage module
while (atomic_read(&nf_conntrack_untracked.ct_general.use) > 1)
schedule();
+ rcu_assign_pointer(nf_ct_destroy, NULL);
+
for (i = 0; i < NF_CT_F_NUM; i++) {
if (nf_ct_cache[i].use == 0)
continue;
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]) {
- kfree(nf_ct_protos[i]);
- nf_ct_protos[i] = NULL;
- }
+ nf_conntrack_proto_fini();
}
static struct list_head *alloc_hashtable(int size, int *vmalloced)
struct list_head *hash;
unsigned int i;
- *vmalloced = 0;
- hash = (void*)__get_free_pages(GFP_KERNEL,
+ *vmalloced = 0;
+ hash = (void*)__get_free_pages(GFP_KERNEL,
get_order(sizeof(struct list_head)
* size));
- if (!hash) {
+ if (!hash) {
*vmalloced = 1;
printk(KERN_WARNING "nf_conntrack: falling back to vmalloc.\n");
hash = vmalloc(sizeof(struct list_head) * size);
}
if (hash)
- for (i = 0; i < size; i++)
+ for (i = 0; i < size; i++)
INIT_LIST_HEAD(&hash[i]);
return hash;
int __init nf_conntrack_init(void)
{
- unsigned int i;
int ret;
/* Idea from tcp.c: use 1/16384 of memory. On i386: 32MB
goto err_free_conntrack_slab;
}
- ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_generic);
+ ret = nf_conntrack_proto_init();
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 < AF_MAX; i++)
- nf_ct_l3protos[i] = &nf_conntrack_l3proto_generic;
- write_unlock_bh(&nf_conntrack_lock);
-
/* For use by REJECT target */
- ip_ct_attach = __nf_conntrack_attach;
+ rcu_assign_pointer(ip_ct_attach, __nf_conntrack_attach);
+ rcu_assign_pointer(nf_ct_destroy, destroy_conntrack);
/* Set up fake conntrack:
- to never be deleted, not in any hashes */
projects / powerpc.git / blobdiff