summaryrefslogtreecommitdiff
path: root/net/ipv4
diff options
context:
space:
mode:
Diffstat (limited to 'net/ipv4')
-rw-r--r--net/ipv4/af_inet.c5
-rw-r--r--net/ipv4/icmp.c99
-rw-r--r--net/ipv4/inet_connection_sock.c2
-rw-r--r--net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c2
-rw-r--r--net/ipv4/tcp_ipv4.c16
-rw-r--r--net/ipv4/tcp_minisocks.c2
6 files changed, 85 insertions, 41 deletions
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index c270080f370e..4f539bd48718 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1415,7 +1415,7 @@ static int __init inet_init(void)
ip_init();
- tcp_v4_init(&inet_family_ops);
+ tcp_v4_init();
/* Setup TCP slab cache for open requests. */
tcp_init();
@@ -1430,7 +1430,8 @@ static int __init inet_init(void)
* Set the ICMP layer up
*/
- icmp_init(&inet_family_ops);
+ if (icmp_init() < 0)
+ panic("Failed to create the ICMP control socket.\n");
/*
* Initialise the multicast router
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index a13c074dac09..b51f4b0a3264 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -229,14 +229,16 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES+1];
*
* On SMP we have one ICMP socket per-cpu.
*/
-static DEFINE_PER_CPU(struct socket *, __icmp_socket) = NULL;
-#define icmp_socket __get_cpu_var(__icmp_socket)
+static struct sock *icmp_sk(struct net *net)
+{
+ return net->ipv4.icmp_sk[smp_processor_id()];
+}
-static inline int icmp_xmit_lock(void)
+static inline int icmp_xmit_lock(struct sock *sk)
{
local_bh_disable();
- if (unlikely(!spin_trylock(&icmp_socket->sk->sk_lock.slock))) {
+ if (unlikely(!spin_trylock(&sk->sk_lock.slock))) {
/* This can happen if the output path signals a
* dst_link_failure() for an outgoing ICMP packet.
*/
@@ -246,9 +248,9 @@ static inline int icmp_xmit_lock(void)
return 0;
}
-static inline void icmp_xmit_unlock(void)
+static inline void icmp_xmit_unlock(struct sock *sk)
{
- spin_unlock_bh(&icmp_socket->sk->sk_lock.slock);
+ spin_unlock_bh(&sk->sk_lock.slock);
}
/*
@@ -346,19 +348,21 @@ static int icmp_glue_bits(void *from, char *to, int offset, int len, int odd,
static void icmp_push_reply(struct icmp_bxm *icmp_param,
struct ipcm_cookie *ipc, struct rtable *rt)
{
+ struct sock *sk;
struct sk_buff *skb;
- if (ip_append_data(icmp_socket->sk, icmp_glue_bits, icmp_param,
+ sk = icmp_sk(rt->u.dst.dev->nd_net);
+ if (ip_append_data(sk, icmp_glue_bits, icmp_param,
icmp_param->data_len+icmp_param->head_len,
icmp_param->head_len,
ipc, rt, MSG_DONTWAIT) < 0)
- ip_flush_pending_frames(icmp_socket->sk);
- else if ((skb = skb_peek(&icmp_socket->sk->sk_write_queue)) != NULL) {
+ ip_flush_pending_frames(sk);
+ else if ((skb = skb_peek(&sk->sk_write_queue)) != NULL) {
struct icmphdr *icmph = icmp_hdr(skb);
__wsum csum = 0;
struct sk_buff *skb1;
- skb_queue_walk(&icmp_socket->sk->sk_write_queue, skb1) {
+ skb_queue_walk(&sk->sk_write_queue, skb1) {
csum = csum_add(csum, skb1->csum);
}
csum = csum_partial_copy_nocheck((void *)&icmp_param->data,
@@ -366,7 +370,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
icmp_param->head_len, csum);
icmph->checksum = csum_fold(csum);
skb->ip_summed = CHECKSUM_NONE;
- ip_push_pending_frames(icmp_socket->sk);
+ ip_push_pending_frames(sk);
}
}
@@ -376,16 +380,17 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
{
- struct sock *sk = icmp_socket->sk;
- struct inet_sock *inet = inet_sk(sk);
struct ipcm_cookie ipc;
struct rtable *rt = (struct rtable *)skb->dst;
+ struct net *net = rt->u.dst.dev->nd_net;
+ struct sock *sk = icmp_sk(net);
+ struct inet_sock *inet = inet_sk(sk);
__be32 daddr;
if (ip_options_echo(&icmp_param->replyopts, skb))
return;
- if (icmp_xmit_lock())
+ if (icmp_xmit_lock(sk))
return;
icmp_param->data.icmph.checksum = 0;
@@ -405,7 +410,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
.tos = RT_TOS(ip_hdr(skb)->tos) } },
.proto = IPPROTO_ICMP };
security_skb_classify_flow(skb, &fl);
- if (ip_route_output_key(rt->u.dst.dev->nd_net, &rt, &fl))
+ if (ip_route_output_key(net, &rt, &fl))
goto out_unlock;
}
if (icmpv4_xrlim_allow(rt, icmp_param->data.icmph.type,
@@ -413,7 +418,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
icmp_push_reply(icmp_param, &ipc, rt);
ip_rt_put(rt);
out_unlock:
- icmp_xmit_unlock();
+ icmp_xmit_unlock(sk);
}
@@ -438,10 +443,12 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
__be32 saddr;
u8 tos;
struct net *net;
+ struct sock *sk;
if (!rt)
goto out;
net = rt->u.dst.dev->nd_net;
+ sk = icmp_sk(net);
/*
* Find the original header. It is expected to be valid, of course.
@@ -505,7 +512,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
}
}
- if (icmp_xmit_lock())
+ if (icmp_xmit_lock(sk))
return;
/*
@@ -544,7 +551,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
icmp_param.data.icmph.checksum = 0;
icmp_param.skb = skb_in;
icmp_param.offset = skb_network_offset(skb_in);
- inet_sk(icmp_socket->sk)->tos = tos;
+ inet_sk(sk)->tos = tos;
ipc.addr = iph->saddr;
ipc.opt = &icmp_param.replyopts;
@@ -652,7 +659,7 @@ route_done:
ende:
ip_rt_put(rt);
out_unlock:
- icmp_xmit_unlock();
+ icmp_xmit_unlock(sk);
out:;
}
@@ -1139,29 +1146,46 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = {
},
};
-void __init icmp_init(struct net_proto_family *ops)
+static void __net_exit icmp_sk_exit(struct net *net)
{
- struct inet_sock *inet;
int i;
- for_each_possible_cpu(i) {
- int err;
+ for_each_possible_cpu(i)
+ sk_release_kernel(net->ipv4.icmp_sk[i]);
+ kfree(net->ipv4.icmp_sk);
+ net->ipv4.icmp_sk = NULL;
+}
- err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_ICMP,
- &per_cpu(__icmp_socket, i));
+int __net_init icmp_sk_init(struct net *net)
+{
+ int i, err;
+ net->ipv4.icmp_sk =
+ kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL);
+ if (net->ipv4.icmp_sk == NULL)
+ return -ENOMEM;
+
+ for_each_possible_cpu(i) {
+ struct sock *sk;
+ struct socket *sock;
+ struct inet_sock *inet;
+
+ err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_ICMP, &sock);
if (err < 0)
- panic("Failed to create the ICMP control socket.\n");
+ goto fail;
+
+ net->ipv4.icmp_sk[i] = sk = sock->sk;
+ sk_change_net(sk, net);
- per_cpu(__icmp_socket, i)->sk->sk_allocation = GFP_ATOMIC;
+ sk->sk_allocation = GFP_ATOMIC;
/* Enough space for 2 64K ICMP packets, including
* sk_buff struct overhead.
*/
- per_cpu(__icmp_socket, i)->sk->sk_sndbuf =
+ sk->sk_sndbuf =
(2 * ((64 * 1024) + sizeof(struct sk_buff)));
- inet = inet_sk(per_cpu(__icmp_socket, i)->sk);
+ inet = inet_sk(sk);
inet->uc_ttl = -1;
inet->pmtudisc = IP_PMTUDISC_DONT;
@@ -1169,8 +1193,23 @@ void __init icmp_init(struct net_proto_family *ops)
* see it, we do not wish this socket to see incoming
* packets.
*/
- per_cpu(__icmp_socket, i)->sk->sk_prot->unhash(per_cpu(__icmp_socket, i)->sk);
+ sk->sk_prot->unhash(sk);
}
+ return 0;
+
+fail:
+ icmp_sk_exit(net);
+ return err;
+}
+
+static struct pernet_operations __net_initdata icmp_sk_ops = {
+ .init = icmp_sk_init,
+ .exit = icmp_sk_exit,
+};
+
+int __init icmp_init(void)
+{
+ return register_pernet_device(&icmp_sk_ops);
}
EXPORT_SYMBOL(icmp_err_convert);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index b189278c7bc1..c0e0fa03fce1 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -463,7 +463,7 @@ void inet_csk_reqsk_queue_prune(struct sock *parent,
if (time_after_eq(now, req->expires)) {
if ((req->retrans < thresh ||
(inet_rsk(req)->acked && req->retrans < max_retries))
- && !req->rsk_ops->rtx_syn_ack(parent, req, NULL)) {
+ && !req->rsk_ops->rtx_syn_ack(parent, req)) {
unsigned long timeo;
if (req->retrans++ == 0)
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
index 089252e82c01..9668c3a23efe 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
@@ -379,7 +379,7 @@ static const struct file_operations ct_cpu_seq_fops = {
.open = ct_cpu_seq_open,
.read = seq_read,
.llseek = seq_lseek,
- .release = seq_release_private,
+ .release = seq_release,
};
int __init nf_conntrack_ipv4_compat_init(void)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 00156bf421ca..3b26f9586dcb 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -723,8 +723,8 @@ static void tcp_v4_reqsk_send_ack(struct sk_buff *skb,
* This still operates on a request_sock only, not on a big
* socket.
*/
-static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
- struct dst_entry *dst)
+static int __tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
+ struct dst_entry *dst)
{
const struct inet_request_sock *ireq = inet_rsk(req);
int err = -1;
@@ -732,7 +732,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
/* First, grab a route. */
if (!dst && (dst = inet_csk_route_req(sk, req)) == NULL)
- goto out;
+ return -1;
skb = tcp_make_synack(sk, dst, req);
@@ -751,11 +751,15 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
err = net_xmit_eval(err);
}
-out:
dst_release(dst);
return err;
}
+static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req)
+{
+ return __tcp_v4_send_synack(sk, req, NULL);
+}
+
/*
* IPv4 request_sock destructor.
*/
@@ -1380,7 +1384,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
}
tcp_rsk(req)->snt_isn = isn;
- if (tcp_v4_send_synack(sk, req, dst))
+ if (__tcp_v4_send_synack(sk, req, dst))
goto drop_and_free;
if (want_cookie) {
@@ -2443,7 +2447,7 @@ struct proto tcp_prot = {
REF_PROTO_INUSE(tcp)
};
-void __init tcp_v4_init(struct net_proto_family *ops)
+void __init tcp_v4_init(void)
{
if (inet_csk_ctl_sock_create(&tcp_socket, PF_INET, SOCK_RAW,
IPPROTO_TCP) < 0)
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index b61b76847ad9..0fdd1db641ac 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -536,7 +536,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
* Enforce "SYN-ACK" according to figure 8, figure 6
* of RFC793, fixed by RFC1122.
*/
- req->rsk_ops->rtx_syn_ack(sk, req, NULL);
+ req->rsk_ops->rtx_syn_ack(sk, req);
return NULL;
}