#include <linux/types.h>
#include <linux/timer.h>
#include <linux/skbuff.h>
-#include <linux/vmalloc.h>
#include <net/checksum.h>
#include <net/icmp.h>
#include <net/ip.h>
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_l4proto.h>
-#if 0
-#define DEBUGP printk
-#else
-#define DEBUGP(format, args...)
-#endif
+static DEFINE_SPINLOCK(nf_nat_lock);
-static DEFINE_RWLOCK(nf_nat_lock);
-
-static struct nf_conntrack_l3proto *l3proto = NULL;
+static struct nf_conntrack_l3proto *l3proto __read_mostly;
/* Calculated at init based on memory size */
-static unsigned int nf_nat_htable_size;
-
-static struct list_head *bysource;
+static unsigned int nf_nat_htable_size __read_mostly;
#define MAX_IP_NAT_PROTO 256
-static struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO];
+static const struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO]
+ __read_mostly;
-static inline struct nf_nat_protocol *
+static inline const struct nf_nat_protocol *
__nf_nat_proto_find(u_int8_t protonum)
{
return rcu_dereference(nf_nat_protos[protonum]);
}
-struct nf_nat_protocol *
+const struct nf_nat_protocol *
nf_nat_proto_find_get(u_int8_t protonum)
{
- struct nf_nat_protocol *p;
+ const struct nf_nat_protocol *p;
rcu_read_lock();
p = __nf_nat_proto_find(protonum);
EXPORT_SYMBOL_GPL(nf_nat_proto_find_get);
void
-nf_nat_proto_put(struct nf_nat_protocol *p)
+nf_nat_proto_put(const struct nf_nat_protocol *p)
{
module_put(p->me);
}
static inline unsigned int
hash_by_src(const struct nf_conntrack_tuple *tuple)
{
+ unsigned int hash;
+
/* Original src, to ensure we map it consistently if poss. */
- return jhash_3words((__force u32)tuple->src.u3.ip, tuple->src.u.all,
- tuple->dst.protonum, 0) % nf_nat_htable_size;
+ hash = jhash_3words((__force u32)tuple->src.u3.ip,
+ (__force u32)tuple->src.u.all,
+ tuple->dst.protonum, 0);
+ return ((u64)hash * nf_nat_htable_size) >> 32;
}
/* Is this tuple already taken? (not by us) */
in_range(const struct nf_conntrack_tuple *tuple,
const struct nf_nat_range *range)
{
- struct nf_nat_protocol *proto;
+ const struct nf_nat_protocol *proto;
int ret = 0;
/* If we are supposed to map IPs, then we must be in the
/* Only called for SRC manip */
static int
-find_appropriate_src(const struct nf_conntrack_tuple *tuple,
+find_appropriate_src(struct net *net,
+ const struct nf_conntrack_tuple *tuple,
struct nf_conntrack_tuple *result,
const struct nf_nat_range *range)
{
unsigned int h = hash_by_src(tuple);
- struct nf_conn_nat *nat;
- struct nf_conn *ct;
+ const struct nf_conn_nat *nat;
+ const struct nf_conn *ct;
+ const struct hlist_node *n;
- read_lock_bh(&nf_nat_lock);
- list_for_each_entry(nat, &bysource[h], bysource) {
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nat, n, &net->ipv4.nat_bysource[h], bysource) {
ct = nat->ct;
if (same_src(ct, tuple)) {
/* Copy source part from reply tuple. */
result->dst = tuple->dst;
if (in_range(result, range)) {
- read_unlock_bh(&nf_nat_lock);
+ rcu_read_unlock();
return 1;
}
}
}
- read_unlock_bh(&nf_nat_lock);
+ rcu_read_unlock();
return 0;
}
maxip = ntohl(range->max_ip);
j = jhash_2words((__force u32)tuple->src.u3.ip,
(__force u32)tuple->dst.u3.ip, 0);
- *var_ipp = htonl(minip + j % (maxip - minip + 1));
+ j = ((u64)j * (maxip - minip + 1)) >> 32;
+ *var_ipp = htonl(minip + j);
}
-/* Manipulate the tuple into the range given. For NF_IP_POST_ROUTING,
- * we change the source to map into the range. For NF_IP_PRE_ROUTING
- * and NF_IP_LOCAL_OUT, we change the destination to map into the
+/* Manipulate the tuple into the range given. For NF_INET_POST_ROUTING,
+ * we change the source to map into the range. For NF_INET_PRE_ROUTING
+ * and NF_INET_LOCAL_OUT, we change the destination to map into the
* range. It might not be possible to get a unique tuple, but we try.
* At worst (or if we race), we will end up with a final duplicate in
* __ip_conntrack_confirm and drop the packet. */
struct nf_conn *ct,
enum nf_nat_manip_type maniptype)
{
- struct nf_nat_protocol *proto;
+ struct net *net = nf_ct_net(ct);
+ const struct nf_nat_protocol *proto;
/* 1) If this srcip/proto/src-proto-part is currently mapped,
and that same mapping gives a unique tuple within the given
This is only required for source (ie. NAT/masq) mappings.
So far, we don't do local source mappings, so multiple
manips not an issue. */
- if (maniptype == IP_NAT_MANIP_SRC) {
- if (find_appropriate_src(orig_tuple, tuple, range)) {
- DEBUGP("get_unique_tuple: Found current src map\n");
- if (!(range->flags & IP_NAT_RANGE_PROTO_RANDOM))
- if (!nf_nat_used_tuple(tuple, ct))
- return;
+ if (maniptype == IP_NAT_MANIP_SRC &&
+ !(range->flags & IP_NAT_RANGE_PROTO_RANDOM)) {
+ if (find_appropriate_src(net, orig_tuple, tuple, range)) {
+ pr_debug("get_unique_tuple: Found current src map\n");
+ if (!nf_nat_used_tuple(tuple, ct))
+ return;
}
}
unsigned int
nf_nat_setup_info(struct nf_conn *ct,
const struct nf_nat_range *range,
- unsigned int hooknum)
+ enum nf_nat_manip_type maniptype)
{
+ struct net *net = nf_ct_net(ct);
struct nf_conntrack_tuple curr_tuple, new_tuple;
struct nf_conn_nat *nat;
int have_to_hash = !(ct->status & IPS_NAT_DONE_MASK);
- enum nf_nat_manip_type maniptype = HOOK2MANIP(hooknum);
/* nat helper or nfctnetlink also setup binding */
nat = nfct_nat(ct);
if (!nat) {
nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC);
if (nat == NULL) {
- DEBUGP("failed to add NAT extension\n");
+ pr_debug("failed to add NAT extension\n");
return NF_ACCEPT;
}
}
- NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING ||
- hooknum == NF_IP_POST_ROUTING ||
- hooknum == NF_IP_LOCAL_IN ||
- hooknum == NF_IP_LOCAL_OUT);
+ NF_CT_ASSERT(maniptype == IP_NAT_MANIP_SRC ||
+ maniptype == IP_NAT_MANIP_DST);
BUG_ON(nf_nat_initialized(ct, maniptype));
/* What we've got will look like inverse of reply. Normally
unsigned int srchash;
srchash = hash_by_src(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
- write_lock_bh(&nf_nat_lock);
+ spin_lock_bh(&nf_nat_lock);
/* nf_conntrack_alter_reply might re-allocate exntension aera */
nat = nfct_nat(ct);
nat->ct = ct;
- list_add(&nat->bysource, &bysource[srchash]);
- write_unlock_bh(&nf_nat_lock);
+ hlist_add_head_rcu(&nat->bysource,
+ &net->ipv4.nat_bysource[srchash]);
+ spin_unlock_bh(&nf_nat_lock);
}
/* It's done. */
EXPORT_SYMBOL(nf_nat_setup_info);
/* Returns true if succeeded. */
-static int
+static bool
manip_pkt(u_int16_t proto,
- struct sk_buff **pskb,
+ struct sk_buff *skb,
unsigned int iphdroff,
const struct nf_conntrack_tuple *target,
enum nf_nat_manip_type maniptype)
{
struct iphdr *iph;
- struct nf_nat_protocol *p;
+ const struct nf_nat_protocol *p;
- if (!skb_make_writable(pskb, iphdroff + sizeof(*iph)))
- return 0;
+ if (!skb_make_writable(skb, iphdroff + sizeof(*iph)))
+ return false;
- iph = (void *)(*pskb)->data + iphdroff;
+ iph = (void *)skb->data + iphdroff;
/* Manipulate protcol part. */
/* rcu_read_lock()ed by nf_hook_slow */
p = __nf_nat_proto_find(proto);
- if (!p->manip_pkt(pskb, iphdroff, target, maniptype))
- return 0;
+ if (!p->manip_pkt(skb, iphdroff, target, maniptype))
+ return false;
- iph = (void *)(*pskb)->data + iphdroff;
+ iph = (void *)skb->data + iphdroff;
if (maniptype == IP_NAT_MANIP_SRC) {
- nf_csum_replace4(&iph->check, iph->saddr, target->src.u3.ip);
+ csum_replace4(&iph->check, iph->saddr, target->src.u3.ip);
iph->saddr = target->src.u3.ip;
} else {
- nf_csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip);
+ csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip);
iph->daddr = target->dst.u3.ip;
}
- return 1;
+ return true;
}
/* Do packet manipulations according to nf_nat_setup_info. */
unsigned int nf_nat_packet(struct nf_conn *ct,
enum ip_conntrack_info ctinfo,
unsigned int hooknum,
- struct sk_buff **pskb)
+ struct sk_buff *skb)
{
enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
unsigned long statusbit;
/* We are aiming to look like inverse of other direction. */
nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple);
- if (!manip_pkt(target.dst.protonum, pskb, 0, &target, mtype))
+ if (!manip_pkt(target.dst.protonum, skb, 0, &target, mtype))
return NF_DROP;
}
return NF_ACCEPT;
int nf_nat_icmp_reply_translation(struct nf_conn *ct,
enum ip_conntrack_info ctinfo,
unsigned int hooknum,
- struct sk_buff **pskb)
+ struct sk_buff *skb)
{
struct {
struct icmphdr icmp;
struct iphdr ip;
} *inside;
- struct nf_conntrack_l4proto *l4proto;
+ const struct nf_conntrack_l4proto *l4proto;
struct nf_conntrack_tuple inner, target;
- int hdrlen = ip_hdrlen(*pskb);
+ int hdrlen = ip_hdrlen(skb);
enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
unsigned long statusbit;
enum nf_nat_manip_type manip = HOOK2MANIP(hooknum);
- if (!skb_make_writable(pskb, hdrlen + sizeof(*inside)))
+ if (!skb_make_writable(skb, hdrlen + sizeof(*inside)))
return 0;
- inside = (void *)(*pskb)->data + ip_hdrlen(*pskb);
+ inside = (void *)skb->data + ip_hdrlen(skb);
/* We're actually going to mangle it beyond trivial checksum
adjustment, so make sure the current checksum is correct. */
- if (nf_ip_checksum(*pskb, hooknum, hdrlen, 0))
+ if (nf_ip_checksum(skb, hooknum, hdrlen, 0))
return 0;
/* Must be RELATED */
- NF_CT_ASSERT((*pskb)->nfctinfo == IP_CT_RELATED ||
- (*pskb)->nfctinfo == IP_CT_RELATED+IP_CT_IS_REPLY);
+ NF_CT_ASSERT(skb->nfctinfo == IP_CT_RELATED ||
+ skb->nfctinfo == IP_CT_RELATED+IP_CT_IS_REPLY);
/* Redirects on non-null nats must be dropped, else they'll
start talking to each other without our translation, and be
return 0;
}
- DEBUGP("icmp_reply_translation: translating error %p manp %u dir %s\n",
- *pskb, manip, dir == IP_CT_DIR_ORIGINAL ? "ORIG" : "REPLY");
+ pr_debug("icmp_reply_translation: translating error %p manip %u "
+ "dir %s\n", skb, manip,
+ dir == IP_CT_DIR_ORIGINAL ? "ORIG" : "REPLY");
/* rcu_read_lock()ed by nf_hook_slow */
l4proto = __nf_ct_l4proto_find(PF_INET, inside->ip.protocol);
- if (!nf_ct_get_tuple(*pskb,
- ip_hdrlen(*pskb) + sizeof(struct icmphdr),
- (ip_hdrlen(*pskb) +
+ if (!nf_ct_get_tuple(skb,
+ ip_hdrlen(skb) + sizeof(struct icmphdr),
+ (ip_hdrlen(skb) +
sizeof(struct icmphdr) + inside->ip.ihl * 4),
(u_int16_t)AF_INET,
inside->ip.protocol,
pass all hooks (locally-generated ICMP). Consider incoming
packet: PREROUTING (DST manip), routing produces ICMP, goes
through POSTROUTING (which must correct the DST manip). */
- if (!manip_pkt(inside->ip.protocol, pskb,
- ip_hdrlen(*pskb) + sizeof(inside->icmp),
+ if (!manip_pkt(inside->ip.protocol, skb,
+ ip_hdrlen(skb) + sizeof(inside->icmp),
&ct->tuplehash[!dir].tuple,
!manip))
return 0;
- if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) {
+ if (skb->ip_summed != CHECKSUM_PARTIAL) {
/* Reloading "inside" here since manip_pkt inner. */
- inside = (void *)(*pskb)->data + ip_hdrlen(*pskb);
+ inside = (void *)skb->data + ip_hdrlen(skb);
inside->icmp.checksum = 0;
inside->icmp.checksum =
- csum_fold(skb_checksum(*pskb, hdrlen,
- (*pskb)->len - hdrlen, 0));
+ csum_fold(skb_checksum(skb, hdrlen,
+ skb->len - hdrlen, 0));
}
/* Change outer to look the reply to an incoming packet
if (ct->status & statusbit) {
nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple);
- if (!manip_pkt(0, pskb, 0, &target, manip))
+ if (!manip_pkt(0, skb, 0, &target, manip))
return 0;
}
EXPORT_SYMBOL_GPL(nf_nat_icmp_reply_translation);
/* Protocol registration. */
-int nf_nat_protocol_register(struct nf_nat_protocol *proto)
+int nf_nat_protocol_register(const struct nf_nat_protocol *proto)
{
int ret = 0;
- write_lock_bh(&nf_nat_lock);
+ spin_lock_bh(&nf_nat_lock);
if (nf_nat_protos[proto->protonum] != &nf_nat_unknown_protocol) {
ret = -EBUSY;
goto out;
}
rcu_assign_pointer(nf_nat_protos[proto->protonum], proto);
out:
- write_unlock_bh(&nf_nat_lock);
+ spin_unlock_bh(&nf_nat_lock);
return ret;
}
EXPORT_SYMBOL(nf_nat_protocol_register);
/* Noone stores the protocol anywhere; simply delete it. */
-void nf_nat_protocol_unregister(struct nf_nat_protocol *proto)
+void nf_nat_protocol_unregister(const struct nf_nat_protocol *proto)
{
- write_lock_bh(&nf_nat_lock);
+ spin_lock_bh(&nf_nat_lock);
rcu_assign_pointer(nf_nat_protos[proto->protonum],
&nf_nat_unknown_protocol);
- write_unlock_bh(&nf_nat_lock);
+ spin_unlock_bh(&nf_nat_lock);
synchronize_rcu();
}
EXPORT_SYMBOL(nf_nat_protocol_unregister);
-#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
-int
-nf_nat_port_range_to_nfattr(struct sk_buff *skb,
- const struct nf_nat_range *range)
-{
- NFA_PUT(skb, CTA_PROTONAT_PORT_MIN, sizeof(__be16),
- &range->min.tcp.port);
- NFA_PUT(skb, CTA_PROTONAT_PORT_MAX, sizeof(__be16),
- &range->max.tcp.port);
-
- return 0;
-
-nfattr_failure:
- return -1;
-}
-EXPORT_SYMBOL_GPL(nf_nat_port_nfattr_to_range);
-
-int
-nf_nat_port_nfattr_to_range(struct nfattr *tb[], struct nf_nat_range *range)
-{
- int ret = 0;
-
- /* we have to return whether we actually parsed something or not */
-
- if (tb[CTA_PROTONAT_PORT_MIN-1]) {
- ret = 1;
- range->min.tcp.port =
- *(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MIN-1]);
- }
-
- if (!tb[CTA_PROTONAT_PORT_MAX-1]) {
- if (ret)
- range->max.tcp.port = range->min.tcp.port;
- } else {
- ret = 1;
- range->max.tcp.port =
- *(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MAX-1]);
- }
-
- return ret;
-}
-EXPORT_SYMBOL_GPL(nf_nat_port_range_to_nfattr);
-#endif
-
/* Noone using conntrack by the time this called. */
static void nf_nat_cleanup_conntrack(struct nf_conn *ct)
{
NF_CT_ASSERT(nat->ct->status & IPS_NAT_DONE_MASK);
- write_lock_bh(&nf_nat_lock);
- list_del(&nat->bysource);
- nat->ct = NULL;
- write_unlock_bh(&nf_nat_lock);
+ spin_lock_bh(&nf_nat_lock);
+ hlist_del_rcu(&nat->bysource);
+ spin_unlock_bh(&nf_nat_lock);
}
-static void nf_nat_move_storage(struct nf_conn *conntrack, void *old)
+static void nf_nat_move_storage(void *new, void *old)
{
- struct nf_conn_nat *new_nat = nf_ct_ext_find(conntrack, NF_CT_EXT_NAT);
- struct nf_conn_nat *old_nat = (struct nf_conn_nat *)old;
+ struct nf_conn_nat *new_nat = new;
+ struct nf_conn_nat *old_nat = old;
struct nf_conn *ct = old_nat->ct;
- unsigned int srchash;
- if (!(ct->status & IPS_NAT_DONE_MASK))
+ if (!ct || !(ct->status & IPS_NAT_DONE_MASK))
return;
- srchash = hash_by_src(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
-
- write_lock_bh(&nf_nat_lock);
- list_replace(&old_nat->bysource, &new_nat->bysource);
+ spin_lock_bh(&nf_nat_lock);
new_nat->ct = ct;
- write_unlock_bh(&nf_nat_lock);
+ hlist_replace_rcu(&old_nat->bysource, &new_nat->bysource);
+ spin_unlock_bh(&nf_nat_lock);
}
static struct nf_ct_ext_type nat_extend __read_mostly = {
.flags = NF_CT_EXT_F_PREALLOC,
};
+static int __net_init nf_nat_net_init(struct net *net)
+{
+ net->ipv4.nat_bysource = nf_ct_alloc_hashtable(&nf_nat_htable_size,
+ &net->ipv4.nat_vmalloced);
+ if (!net->ipv4.nat_bysource)
+ return -ENOMEM;
+ return 0;
+}
+
+/* Clear NAT section of all conntracks, in case we're loaded again. */
+static int clean_nat(struct nf_conn *i, void *data)
+{
+ struct nf_conn_nat *nat = nfct_nat(i);
+
+ if (!nat)
+ return 0;
+ memset(nat, 0, sizeof(*nat));
+ i->status &= ~(IPS_NAT_MASK | IPS_NAT_DONE_MASK | IPS_SEQ_ADJUST);
+ return 0;
+}
+
+static void __net_exit nf_nat_net_exit(struct net *net)
+{
+ nf_ct_iterate_cleanup(net, &clean_nat, NULL);
+ synchronize_rcu();
+ nf_ct_free_hashtable(net->ipv4.nat_bysource, net->ipv4.nat_vmalloced,
+ nf_nat_htable_size);
+}
+
+static struct pernet_operations nf_nat_net_ops = {
+ .init = nf_nat_net_init,
+ .exit = nf_nat_net_exit,
+};
+
static int __init nf_nat_init(void)
{
size_t i;
int ret;
+ need_ipv4_conntrack();
+
ret = nf_ct_extend_register(&nat_extend);
if (ret < 0) {
printk(KERN_ERR "nf_nat_core: Unable to register extension\n");
/* Leave them the same for the moment. */
nf_nat_htable_size = nf_conntrack_htable_size;
- /* One vmalloc for both hash tables */
- bysource = vmalloc(sizeof(struct list_head) * nf_nat_htable_size);
- if (!bysource) {
- ret = -ENOMEM;
+ ret = register_pernet_subsys(&nf_nat_net_ops);
+ if (ret < 0)
goto cleanup_extend;
- }
/* Sew in builtin protocols. */
- write_lock_bh(&nf_nat_lock);
+ spin_lock_bh(&nf_nat_lock);
for (i = 0; i < MAX_IP_NAT_PROTO; i++)
rcu_assign_pointer(nf_nat_protos[i], &nf_nat_unknown_protocol);
rcu_assign_pointer(nf_nat_protos[IPPROTO_TCP], &nf_nat_protocol_tcp);
rcu_assign_pointer(nf_nat_protos[IPPROTO_UDP], &nf_nat_protocol_udp);
rcu_assign_pointer(nf_nat_protos[IPPROTO_ICMP], &nf_nat_protocol_icmp);
- write_unlock_bh(&nf_nat_lock);
-
- for (i = 0; i < nf_nat_htable_size; i++) {
- INIT_LIST_HEAD(&bysource[i]);
- }
+ spin_unlock_bh(&nf_nat_lock);
/* Initialize fake conntrack so that NAT will skip it */
nf_conntrack_untracked.status |= IPS_NAT_DONE_MASK;
l3proto = nf_ct_l3proto_find_get((u_int16_t)AF_INET);
+
+ BUG_ON(nf_nat_seq_adjust_hook != NULL);
+ rcu_assign_pointer(nf_nat_seq_adjust_hook, nf_nat_seq_adjust);
return 0;
cleanup_extend:
return ret;
}
-/* Clear NAT section of all conntracks, in case we're loaded again. */
-static int clean_nat(struct nf_conn *i, void *data)
-{
- struct nf_conn_nat *nat = nfct_nat(i);
-
- if (!nat)
- return 0;
- memset(nat, 0, sizeof(nat));
- i->status &= ~(IPS_NAT_MASK | IPS_NAT_DONE_MASK | IPS_SEQ_ADJUST);
- return 0;
-}
-
static void __exit nf_nat_cleanup(void)
{
- nf_ct_iterate_cleanup(&clean_nat, NULL);
- synchronize_rcu();
- vfree(bysource);
+ unregister_pernet_subsys(&nf_nat_net_ops);
nf_ct_l3proto_put(l3proto);
nf_ct_extend_unregister(&nat_extend);
+ rcu_assign_pointer(nf_nat_seq_adjust_hook, NULL);
+ synchronize_net();
}
MODULE_LICENSE("GPL");