#include <linux/inetdevice.h>
#include <linux/igmp.h>
#include <linux/netfilter_ipv4.h>
+#include <linux/etherdevice.h>
#include <linux/if_ether.h>
#include <net/sock.h>
#include <net/xfrm.h>
#include <net/net_namespace.h>
#include <net/netns/generic.h>
+#include <net/rtnetlink.h>
#ifdef CONFIG_IPV6
#include <net/ipv6.h>
solution, but it supposes maintaing new variable in ALL
skb, even if no tunneling is used.
- Current solution: t->recursion lock breaks dead loops. It looks
- like dev->tbusy flag, but I preferred new variable, because
- the semantics is different. One day, when hard_start_xmit
- will be multithreaded we will have to use skb->encapsulation.
+ Current solution: HARD_TX_LOCK lock breaks dead loops.
Alexey Kuznetsov.
*/
+static struct rtnl_link_ops ipgre_link_ops __read_mostly;
static int ipgre_tunnel_init(struct net_device *dev);
static void ipgre_tunnel_setup(struct net_device *dev);
+static int ipgre_tunnel_bind_dev(struct net_device *dev);
/* Fallback tunnel: no source, no destination, no key, no options */
-static int ipgre_fb_tunnel_init(struct net_device *dev);
-
#define HASH_SIZE 16
-static int ipgre_net_id;
+static int ipgre_net_id __read_mostly;
struct ipgre_net {
struct ip_tunnel *tunnels[4][HASH_SIZE];
#define tunnels_r tunnels[2]
#define tunnels_l tunnels[1]
#define tunnels_wc tunnels[0]
+/*
+ * Locking : hash tables are protected by RCU and a spinlock
+ */
+static DEFINE_SPINLOCK(ipgre_lock);
-static DEFINE_RWLOCK(ipgre_lock);
+#define for_each_ip_tunnel_rcu(start) \
+ for (t = rcu_dereference(start); t; t = rcu_dereference(t->next))
/* Given src, dst and key, find appropriate for input tunnel. */
-static struct ip_tunnel * ipgre_tunnel_lookup(struct net *net,
- __be32 remote, __be32 local, __be32 key)
+static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev,
+ __be32 remote, __be32 local,
+ __be32 key, __be16 gre_proto)
{
+ struct net *net = dev_net(dev);
+ int link = dev->ifindex;
unsigned h0 = HASH(remote);
unsigned h1 = HASH(key);
- struct ip_tunnel *t;
+ struct ip_tunnel *t, *cand = NULL;
struct ipgre_net *ign = net_generic(net, ipgre_net_id);
+ int dev_type = (gre_proto == htons(ETH_P_TEB)) ?
+ ARPHRD_ETHER : ARPHRD_IPGRE;
+ int score, cand_score = 4;
+
+ for_each_ip_tunnel_rcu(ign->tunnels_r_l[h0 ^ h1]) {
+ if (local != t->parms.iph.saddr ||
+ remote != t->parms.iph.daddr ||
+ key != t->parms.i_key ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (t->dev->type != ARPHRD_IPGRE &&
+ t->dev->type != dev_type)
+ continue;
+
+ score = 0;
+ if (t->parms.link != link)
+ score |= 1;
+ if (t->dev->type != dev_type)
+ score |= 2;
+ if (score == 0)
+ return t;
- for (t = ign->tunnels_r_l[h0^h1]; t; t = t->next) {
- if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr) {
- if (t->parms.i_key == key && (t->dev->flags&IFF_UP))
- return t;
+ if (score < cand_score) {
+ cand = t;
+ cand_score = score;
}
}
- for (t = ign->tunnels_r[h0^h1]; t; t = t->next) {
- if (remote == t->parms.iph.daddr) {
- if (t->parms.i_key == key && (t->dev->flags&IFF_UP))
- return t;
+
+ for_each_ip_tunnel_rcu(ign->tunnels_r[h0 ^ h1]) {
+ if (remote != t->parms.iph.daddr ||
+ key != t->parms.i_key ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (t->dev->type != ARPHRD_IPGRE &&
+ t->dev->type != dev_type)
+ continue;
+
+ score = 0;
+ if (t->parms.link != link)
+ score |= 1;
+ if (t->dev->type != dev_type)
+ score |= 2;
+ if (score == 0)
+ return t;
+
+ if (score < cand_score) {
+ cand = t;
+ cand_score = score;
}
}
- for (t = ign->tunnels_l[h1]; t; t = t->next) {
- if (local == t->parms.iph.saddr ||
- (local == t->parms.iph.daddr &&
- ipv4_is_multicast(local))) {
- if (t->parms.i_key == key && (t->dev->flags&IFF_UP))
- return t;
+
+ for_each_ip_tunnel_rcu(ign->tunnels_l[h1]) {
+ if ((local != t->parms.iph.saddr &&
+ (local != t->parms.iph.daddr ||
+ !ipv4_is_multicast(local))) ||
+ key != t->parms.i_key ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (t->dev->type != ARPHRD_IPGRE &&
+ t->dev->type != dev_type)
+ continue;
+
+ score = 0;
+ if (t->parms.link != link)
+ score |= 1;
+ if (t->dev->type != dev_type)
+ score |= 2;
+ if (score == 0)
+ return t;
+
+ if (score < cand_score) {
+ cand = t;
+ cand_score = score;
}
}
- for (t = ign->tunnels_wc[h1]; t; t = t->next) {
- if (t->parms.i_key == key && (t->dev->flags&IFF_UP))
+
+ for_each_ip_tunnel_rcu(ign->tunnels_wc[h1]) {
+ if (t->parms.i_key != key ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (t->dev->type != ARPHRD_IPGRE &&
+ t->dev->type != dev_type)
+ continue;
+
+ score = 0;
+ if (t->parms.link != link)
+ score |= 1;
+ if (t->dev->type != dev_type)
+ score |= 2;
+ if (score == 0)
return t;
+
+ if (score < cand_score) {
+ cand = t;
+ cand_score = score;
+ }
}
- if (ign->fb_tunnel_dev->flags&IFF_UP)
- return netdev_priv(ign->fb_tunnel_dev);
+ if (cand != NULL)
+ return cand;
+
+ dev = ign->fb_tunnel_dev;
+ if (dev->flags & IFF_UP)
+ return netdev_priv(dev);
+
return NULL;
}
{
struct ip_tunnel **tp = ipgre_bucket(ign, t);
+ spin_lock_bh(&ipgre_lock);
t->next = *tp;
- write_lock_bh(&ipgre_lock);
- *tp = t;
- write_unlock_bh(&ipgre_lock);
+ rcu_assign_pointer(*tp, t);
+ spin_unlock_bh(&ipgre_lock);
}
static void ipgre_tunnel_unlink(struct ipgre_net *ign, struct ip_tunnel *t)
for (tp = ipgre_bucket(ign, t); *tp; tp = &(*tp)->next) {
if (t == *tp) {
- write_lock_bh(&ipgre_lock);
+ spin_lock_bh(&ipgre_lock);
*tp = t->next;
- write_unlock_bh(&ipgre_lock);
+ spin_unlock_bh(&ipgre_lock);
break;
}
}
}
-static struct ip_tunnel * ipgre_tunnel_locate(struct net *net,
- struct ip_tunnel_parm *parms, int create)
+static struct ip_tunnel *ipgre_tunnel_find(struct net *net,
+ struct ip_tunnel_parm *parms,
+ int type)
{
__be32 remote = parms->iph.daddr;
__be32 local = parms->iph.saddr;
__be32 key = parms->i_key;
- struct ip_tunnel *t, **tp, *nt;
+ int link = parms->link;
+ struct ip_tunnel *t, **tp;
+ struct ipgre_net *ign = net_generic(net, ipgre_net_id);
+
+ for (tp = __ipgre_bucket(ign, parms); (t = *tp) != NULL; tp = &t->next)
+ if (local == t->parms.iph.saddr &&
+ remote == t->parms.iph.daddr &&
+ key == t->parms.i_key &&
+ link == t->parms.link &&
+ type == t->dev->type)
+ break;
+
+ return t;
+}
+
+static struct ip_tunnel * ipgre_tunnel_locate(struct net *net,
+ struct ip_tunnel_parm *parms, int create)
+{
+ struct ip_tunnel *t, *nt;
struct net_device *dev;
char name[IFNAMSIZ];
struct ipgre_net *ign = net_generic(net, ipgre_net_id);
- for (tp = __ipgre_bucket(ign, parms); (t = *tp) != NULL; tp = &t->next) {
- if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr) {
- if (key == t->parms.i_key)
- return t;
- }
- }
- if (!create)
- return NULL;
+ t = ipgre_tunnel_find(net, parms, ARPHRD_IPGRE);
+ if (t || !create)
+ return t;
if (parms->name[0])
strlcpy(name, parms->name, IFNAMSIZ);
goto failed_free;
}
- dev->init = ipgre_tunnel_init;
nt = netdev_priv(dev);
nt->parms = *parms;
+ dev->rtnl_link_ops = &ipgre_link_ops;
+
+ dev->mtu = ipgre_tunnel_bind_dev(dev);
if (register_netdevice(dev) < 0)
goto failed_free;
by themself???
*/
- struct iphdr *iph = (struct iphdr*)skb->data;
+ struct iphdr *iph = (struct iphdr *)skb->data;
__be16 *p = (__be16*)(skb->data+(iph->ihl<<2));
int grehlen = (iph->ihl<<2) + 4;
const int type = icmp_hdr(skb)->type;
break;
}
- read_lock(&ipgre_lock);
- t = ipgre_tunnel_lookup(dev_net(skb->dev), iph->daddr, iph->saddr,
- (flags&GRE_KEY) ?
- *(((__be32*)p) + (grehlen>>2) - 1) : 0);
+ rcu_read_lock();
+ t = ipgre_tunnel_lookup(skb->dev, iph->daddr, iph->saddr,
+ flags & GRE_KEY ?
+ *(((__be32 *)p) + (grehlen / 4) - 1) : 0,
+ p[1]);
if (t == NULL || t->parms.iph.daddr == 0 ||
ipv4_is_multicast(t->parms.iph.daddr))
goto out;
if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
goto out;
- if (jiffies - t->err_time < IPTUNNEL_ERR_TIMEO)
+ if (time_before(jiffies, t->err_time + IPTUNNEL_ERR_TIMEO))
t->err_count++;
else
t->err_count = 1;
t->err_time = jiffies;
out:
- read_unlock(&ipgre_lock);
+ rcu_read_unlock();
return;
}
u32 seqno = 0;
struct ip_tunnel *tunnel;
int offset = 4;
+ __be16 gre_proto;
+ unsigned int len;
if (!pskb_may_pull(skb, 16))
goto drop_nolock;
}
}
- read_lock(&ipgre_lock);
- if ((tunnel = ipgre_tunnel_lookup(dev_net(skb->dev),
- iph->saddr, iph->daddr, key)) != NULL) {
+ gre_proto = *(__be16 *)(h + 2);
+
+ rcu_read_lock();
+ if ((tunnel = ipgre_tunnel_lookup(skb->dev,
+ iph->saddr, iph->daddr, key,
+ gre_proto))) {
struct net_device_stats *stats = &tunnel->dev->stats;
secpath_reset(skb);
- skb->protocol = *(__be16*)(h + 2);
+ skb->protocol = gre_proto;
/* WCCP version 1 and 2 protocol decoding.
* - Change protocol to IP
* - When dealing with WCCPv2, Skip extra 4 bytes in GRE header
*/
- if (flags == 0 &&
- skb->protocol == htons(ETH_P_WCCP)) {
+ if (flags == 0 && gre_proto == htons(ETH_P_WCCP)) {
skb->protocol = htons(ETH_P_IP);
if ((*(h + offset) & 0xF0) != 0x40)
offset += 4;
skb->mac_header = skb->network_header;
__pskb_pull(skb, offset);
- skb_reset_network_header(skb);
skb_postpull_rcsum(skb, skb_transport_header(skb), offset);
skb->pkt_type = PACKET_HOST;
#ifdef CONFIG_NET_IPGRE_BROADCAST
if (ipv4_is_multicast(iph->daddr)) {
/* Looped back packet, drop it! */
- if (skb->rtable->fl.iif == 0)
+ if (skb_rtable(skb)->fl.iif == 0)
goto drop;
stats->multicast++;
skb->pkt_type = PACKET_BROADCAST;
}
tunnel->i_seqno = seqno + 1;
}
+
+ len = skb->len;
+
+ /* Warning: All skb pointers will be invalidated! */
+ if (tunnel->dev->type == ARPHRD_ETHER) {
+ if (!pskb_may_pull(skb, ETH_HLEN)) {
+ stats->rx_length_errors++;
+ stats->rx_errors++;
+ goto drop;
+ }
+
+ iph = ip_hdr(skb);
+ skb->protocol = eth_type_trans(skb, tunnel->dev);
+ skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
+ }
+
stats->rx_packets++;
- stats->rx_bytes += skb->len;
+ stats->rx_bytes += len;
skb->dev = tunnel->dev;
- dst_release(skb->dst);
- skb->dst = NULL;
+ skb_dst_drop(skb);
nf_reset(skb);
+
+ skb_reset_network_header(skb);
ipgre_ecn_decapsulate(iph, skb);
+
netif_rx(skb);
- read_unlock(&ipgre_lock);
+ rcu_read_unlock();
return(0);
}
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
drop:
- read_unlock(&ipgre_lock);
+ rcu_read_unlock();
drop_nolock:
kfree_skb(skb);
return(0);
}
-static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct ip_tunnel *tunnel = netdev_priv(dev);
- struct net_device_stats *stats = &tunnel->dev->stats;
+ struct net_device_stats *stats = &dev->stats;
+ struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
struct iphdr *old_iph = ip_hdr(skb);
struct iphdr *tiph;
u8 tos;
__be32 dst;
int mtu;
- if (tunnel->recursion++) {
- stats->collisions++;
- goto tx_error;
- }
+ if (dev->type == ARPHRD_ETHER)
+ IPCB(skb)->flags = 0;
- if (dev->header_ops) {
+ if (dev->header_ops && dev->type == ARPHRD_IPGRE) {
gre_hlen = 0;
- tiph = (struct iphdr*)skb->data;
+ tiph = (struct iphdr *)skb->data;
} else {
gre_hlen = tunnel->hlen;
tiph = &tunnel->parms.iph;
if ((dst = tiph->daddr) == 0) {
/* NBMA tunnel */
- if (skb->dst == NULL) {
+ if (skb_dst(skb) == NULL) {
stats->tx_fifo_errors++;
goto tx_error;
}
if (skb->protocol == htons(ETH_P_IP)) {
- rt = skb->rtable;
+ rt = skb_rtable(skb);
if ((dst = rt->rt_gateway) == 0)
goto tx_error_icmp;
}
else if (skb->protocol == htons(ETH_P_IPV6)) {
struct in6_addr *addr6;
int addr_type;
- struct neighbour *neigh = skb->dst->neighbour;
+ struct neighbour *neigh = skb_dst(skb)->neighbour;
if (neigh == NULL)
goto tx_error;
- addr6 = (struct in6_addr*)&neigh->primary_key;
+ addr6 = (struct in6_addr *)&neigh->primary_key;
addr_type = ipv6_addr_type(addr6);
if (addr_type == IPV6_ADDR_ANY) {
}
tos = tiph->tos;
- if (tos&1) {
+ if (tos == 1) {
+ tos = 0;
if (skb->protocol == htons(ETH_P_IP))
tos = old_iph->tos;
- tos &= ~1;
}
{
if (df)
mtu = dst_mtu(&rt->u.dst) - dev->hard_header_len - tunnel->hlen;
else
- mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu;
+ mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu;
- if (skb->dst)
- skb->dst->ops->update_pmtu(skb->dst, mtu);
+ if (skb_dst(skb))
+ skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu);
if (skb->protocol == htons(ETH_P_IP)) {
df |= (old_iph->frag_off&htons(IP_DF));
}
#ifdef CONFIG_IPV6
else if (skb->protocol == htons(ETH_P_IPV6)) {
- struct rt6_info *rt6 = (struct rt6_info*)skb->dst;
+ struct rt6_info *rt6 = (struct rt6_info *)skb_dst(skb);
- if (rt6 && mtu < dst_mtu(skb->dst) && mtu >= IPV6_MIN_MTU) {
+ if (rt6 && mtu < dst_mtu(skb_dst(skb)) && mtu >= IPV6_MIN_MTU) {
if ((tunnel->parms.iph.daddr &&
!ipv4_is_multicast(tunnel->parms.iph.daddr)) ||
rt6->rt6i_dst.plen == 128) {
rt6->rt6i_flags |= RTF_MODIFIED;
- skb->dst->metrics[RTAX_MTU-1] = mtu;
+ skb_dst(skb)->metrics[RTAX_MTU-1] = mtu;
}
}
#endif
if (tunnel->err_count > 0) {
- if (jiffies - tunnel->err_time < IPTUNNEL_ERR_TIMEO) {
+ if (time_before(jiffies,
+ tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
tunnel->err_count--;
dst_link_failure(skb);
struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom);
if (!new_skb) {
ip_rt_put(rt);
- stats->tx_dropped++;
+ txq->tx_dropped++;
dev_kfree_skb(skb);
- tunnel->recursion--;
- return 0;
+ return NETDEV_TX_OK;
}
if (skb->sk)
skb_set_owner_w(new_skb, skb->sk);
old_iph = ip_hdr(skb);
}
- skb->transport_header = skb->network_header;
+ skb_reset_transport_header(skb);
skb_push(skb, gre_hlen);
skb_reset_network_header(skb);
memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
IPSKB_REROUTED);
- dst_release(skb->dst);
- skb->dst = &rt->u.dst;
+ skb_dst_drop(skb);
+ skb_dst_set(skb, &rt->u.dst);
/*
* Push down and install the IPIP header.
iph->ttl = old_iph->ttl;
#ifdef CONFIG_IPV6
else if (skb->protocol == htons(ETH_P_IPV6))
- iph->ttl = ((struct ipv6hdr*)old_iph)->hop_limit;
+ iph->ttl = ((struct ipv6hdr *)old_iph)->hop_limit;
#endif
else
iph->ttl = dst_metric(&rt->u.dst, RTAX_HOPLIMIT);
}
- ((__be16*)(iph+1))[0] = tunnel->parms.o_flags;
- ((__be16*)(iph+1))[1] = skb->protocol;
+ ((__be16 *)(iph + 1))[0] = tunnel->parms.o_flags;
+ ((__be16 *)(iph + 1))[1] = (dev->type == ARPHRD_ETHER) ?
+ htons(ETH_P_TEB) : skb->protocol;
if (tunnel->parms.o_flags&(GRE_KEY|GRE_CSUM|GRE_SEQ)) {
__be32 *ptr = (__be32*)(((u8*)iph) + tunnel->hlen - 4);
nf_reset(skb);
IPTUNNEL_XMIT();
- tunnel->recursion--;
- return 0;
+ return NETDEV_TX_OK;
tx_error_icmp:
dst_link_failure(skb);
tx_error:
stats->tx_errors++;
dev_kfree_skb(skb);
- tunnel->recursion--;
- return 0;
+ return NETDEV_TX_OK;
}
-static void ipgre_tunnel_bind_dev(struct net_device *dev)
+static int ipgre_tunnel_bind_dev(struct net_device *dev)
{
struct net_device *tdev = NULL;
struct ip_tunnel *tunnel;
tdev = rt->u.dst.dev;
ip_rt_put(rt);
}
- dev->flags |= IFF_POINTOPOINT;
+
+ if (dev->type != ARPHRD_ETHER)
+ dev->flags |= IFF_POINTOPOINT;
}
if (!tdev && tunnel->parms.link)
addend += 4;
}
dev->needed_headroom = addend + hlen;
- dev->mtu = mtu - dev->hard_header_len - addend;
+ mtu -= dev->hard_header_len + addend;
+
+ if (mtu < 68)
+ mtu = 68;
+
tunnel->hlen = addend;
+ return mtu;
}
static int
break;
}
} else {
- unsigned nflags=0;
+ unsigned nflags = 0;
t = netdev_priv(dev);
t->parms.iph.frag_off = p.iph.frag_off;
if (t->parms.link != p.link) {
t->parms.link = p.link;
- ipgre_tunnel_bind_dev(dev);
+ dev->mtu = ipgre_tunnel_bind_dev(dev);
netdev_state_change(dev);
}
}
static int ipgre_header_parse(const struct sk_buff *skb, unsigned char *haddr)
{
- struct iphdr *iph = (struct iphdr*) skb_mac_header(skb);
+ struct iphdr *iph = (struct iphdr *) skb_mac_header(skb);
memcpy(haddr, &iph->saddr, 4);
return 4;
}
static int ipgre_close(struct net_device *dev)
{
struct ip_tunnel *t = netdev_priv(dev);
+
if (ipv4_is_multicast(t->parms.iph.daddr) && t->mlink) {
struct in_device *in_dev;
in_dev = inetdev_by_index(dev_net(dev), t->mlink);
#endif
+static const struct net_device_ops ipgre_netdev_ops = {
+ .ndo_init = ipgre_tunnel_init,
+ .ndo_uninit = ipgre_tunnel_uninit,
+#ifdef CONFIG_NET_IPGRE_BROADCAST
+ .ndo_open = ipgre_open,
+ .ndo_stop = ipgre_close,
+#endif
+ .ndo_start_xmit = ipgre_tunnel_xmit,
+ .ndo_do_ioctl = ipgre_tunnel_ioctl,
+ .ndo_change_mtu = ipgre_tunnel_change_mtu,
+};
+
static void ipgre_tunnel_setup(struct net_device *dev)
{
- dev->uninit = ipgre_tunnel_uninit;
+ dev->netdev_ops = &ipgre_netdev_ops;
dev->destructor = free_netdev;
- dev->hard_start_xmit = ipgre_tunnel_xmit;
- dev->do_ioctl = ipgre_tunnel_ioctl;
- dev->change_mtu = ipgre_tunnel_change_mtu;
dev->type = ARPHRD_IPGRE;
dev->needed_headroom = LL_MAX_HEADER + sizeof(struct iphdr) + 4;
dev->iflink = 0;
dev->addr_len = 4;
dev->features |= NETIF_F_NETNS_LOCAL;
+ dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
}
static int ipgre_tunnel_init(struct net_device *dev)
memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4);
memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4);
- ipgre_tunnel_bind_dev(dev);
-
if (iph->daddr) {
#ifdef CONFIG_NET_IPGRE_BROADCAST
if (ipv4_is_multicast(iph->daddr)) {
return -EINVAL;
dev->flags = IFF_BROADCAST;
dev->header_ops = &ipgre_header_ops;
- dev->open = ipgre_open;
- dev->stop = ipgre_close;
}
#endif
} else
return 0;
}
-static int ipgre_fb_tunnel_init(struct net_device *dev)
+static void ipgre_fb_tunnel_init(struct net_device *dev)
{
struct ip_tunnel *tunnel = netdev_priv(dev);
struct iphdr *iph = &tunnel->parms.iph;
dev_hold(dev);
ign->tunnels_wc[0] = tunnel;
- return 0;
}
-static struct net_protocol ipgre_protocol = {
+static const struct net_protocol ipgre_protocol = {
.handler = ipgre_rcv,
.err_handler = ipgre_err,
.netns_ok = 1,
};
-static void ipgre_destroy_tunnels(struct ipgre_net *ign)
+static void ipgre_destroy_tunnels(struct ipgre_net *ign, struct list_head *head)
{
int prio;
for (prio = 0; prio < 4; prio++) {
int h;
for (h = 0; h < HASH_SIZE; h++) {
- struct ip_tunnel *t;
- while ((t = ign->tunnels[prio][h]) != NULL)
- unregister_netdevice(t->dev);
+ struct ip_tunnel *t = ign->tunnels[prio][h];
+
+ while (t != NULL) {
+ unregister_netdevice_queue(t->dev, head);
+ t = t->next;
+ }
}
}
}
-static int ipgre_init_net(struct net *net)
+static int __net_init ipgre_init_net(struct net *net)
{
+ struct ipgre_net *ign = net_generic(net, ipgre_net_id);
int err;
- struct ipgre_net *ign;
-
- err = -ENOMEM;
- ign = kzalloc(sizeof(struct ipgre_net), GFP_KERNEL);
- if (ign == NULL)
- goto err_alloc;
-
- err = net_assign_generic(net, ipgre_net_id, ign);
- if (err < 0)
- goto err_assign;
ign->fb_tunnel_dev = alloc_netdev(sizeof(struct ip_tunnel), "gre0",
ipgre_tunnel_setup);
err = -ENOMEM;
goto err_alloc_dev;
}
-
- ign->fb_tunnel_dev->init = ipgre_fb_tunnel_init;
dev_net_set(ign->fb_tunnel_dev, net);
+ ipgre_fb_tunnel_init(ign->fb_tunnel_dev);
+ ign->fb_tunnel_dev->rtnl_link_ops = &ipgre_link_ops;
+
if ((err = register_netdev(ign->fb_tunnel_dev)))
goto err_reg_dev;
err_reg_dev:
free_netdev(ign->fb_tunnel_dev);
err_alloc_dev:
- /* nothing */
-err_assign:
- kfree(ign);
-err_alloc:
return err;
}
-static void ipgre_exit_net(struct net *net)
+static void __net_exit ipgre_exit_net(struct net *net)
{
struct ipgre_net *ign;
+ LIST_HEAD(list);
ign = net_generic(net, ipgre_net_id);
rtnl_lock();
- ipgre_destroy_tunnels(ign);
+ ipgre_destroy_tunnels(ign, &list);
+ unregister_netdevice_many(&list);
rtnl_unlock();
- kfree(ign);
}
static struct pernet_operations ipgre_net_ops = {
.init = ipgre_init_net,
.exit = ipgre_exit_net,
+ .id = &ipgre_net_id,
+ .size = sizeof(struct ipgre_net),
+};
+
+static int ipgre_tunnel_validate(struct nlattr *tb[], struct nlattr *data[])
+{
+ __be16 flags;
+
+ if (!data)
+ return 0;
+
+ flags = 0;
+ if (data[IFLA_GRE_IFLAGS])
+ flags |= nla_get_be16(data[IFLA_GRE_IFLAGS]);
+ if (data[IFLA_GRE_OFLAGS])
+ flags |= nla_get_be16(data[IFLA_GRE_OFLAGS]);
+ if (flags & (GRE_VERSION|GRE_ROUTING))
+ return -EINVAL;
+
+ return 0;
+}
+
+static int ipgre_tap_validate(struct nlattr *tb[], struct nlattr *data[])
+{
+ __be32 daddr;
+
+ if (tb[IFLA_ADDRESS]) {
+ if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
+ return -EINVAL;
+ if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
+ return -EADDRNOTAVAIL;
+ }
+
+ if (!data)
+ goto out;
+
+ if (data[IFLA_GRE_REMOTE]) {
+ memcpy(&daddr, nla_data(data[IFLA_GRE_REMOTE]), 4);
+ if (!daddr)
+ return -EINVAL;
+ }
+
+out:
+ return ipgre_tunnel_validate(tb, data);
+}
+
+static void ipgre_netlink_parms(struct nlattr *data[],
+ struct ip_tunnel_parm *parms)
+{
+ memset(parms, 0, sizeof(*parms));
+
+ parms->iph.protocol = IPPROTO_GRE;
+
+ if (!data)
+ return;
+
+ if (data[IFLA_GRE_LINK])
+ parms->link = nla_get_u32(data[IFLA_GRE_LINK]);
+
+ if (data[IFLA_GRE_IFLAGS])
+ parms->i_flags = nla_get_be16(data[IFLA_GRE_IFLAGS]);
+
+ if (data[IFLA_GRE_OFLAGS])
+ parms->o_flags = nla_get_be16(data[IFLA_GRE_OFLAGS]);
+
+ if (data[IFLA_GRE_IKEY])
+ parms->i_key = nla_get_be32(data[IFLA_GRE_IKEY]);
+
+ if (data[IFLA_GRE_OKEY])
+ parms->o_key = nla_get_be32(data[IFLA_GRE_OKEY]);
+
+ if (data[IFLA_GRE_LOCAL])
+ parms->iph.saddr = nla_get_be32(data[IFLA_GRE_LOCAL]);
+
+ if (data[IFLA_GRE_REMOTE])
+ parms->iph.daddr = nla_get_be32(data[IFLA_GRE_REMOTE]);
+
+ if (data[IFLA_GRE_TTL])
+ parms->iph.ttl = nla_get_u8(data[IFLA_GRE_TTL]);
+
+ if (data[IFLA_GRE_TOS])
+ parms->iph.tos = nla_get_u8(data[IFLA_GRE_TOS]);
+
+ if (!data[IFLA_GRE_PMTUDISC] || nla_get_u8(data[IFLA_GRE_PMTUDISC]))
+ parms->iph.frag_off = htons(IP_DF);
+}
+
+static int ipgre_tap_init(struct net_device *dev)
+{
+ struct ip_tunnel *tunnel;
+
+ tunnel = netdev_priv(dev);
+
+ tunnel->dev = dev;
+ strcpy(tunnel->parms.name, dev->name);
+
+ ipgre_tunnel_bind_dev(dev);
+
+ return 0;
+}
+
+static const struct net_device_ops ipgre_tap_netdev_ops = {
+ .ndo_init = ipgre_tap_init,
+ .ndo_uninit = ipgre_tunnel_uninit,
+ .ndo_start_xmit = ipgre_tunnel_xmit,
+ .ndo_set_mac_address = eth_mac_addr,
+ .ndo_validate_addr = eth_validate_addr,
+ .ndo_change_mtu = ipgre_tunnel_change_mtu,
+};
+
+static void ipgre_tap_setup(struct net_device *dev)
+{
+
+ ether_setup(dev);
+
+ dev->netdev_ops = &ipgre_tap_netdev_ops;
+ dev->destructor = free_netdev;
+
+ dev->iflink = 0;
+ dev->features |= NETIF_F_NETNS_LOCAL;
+}
+
+static int ipgre_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[],
+ struct nlattr *data[])
+{
+ struct ip_tunnel *nt;
+ struct net *net = dev_net(dev);
+ struct ipgre_net *ign = net_generic(net, ipgre_net_id);
+ int mtu;
+ int err;
+
+ nt = netdev_priv(dev);
+ ipgre_netlink_parms(data, &nt->parms);
+
+ if (ipgre_tunnel_find(net, &nt->parms, dev->type))
+ return -EEXIST;
+
+ if (dev->type == ARPHRD_ETHER && !tb[IFLA_ADDRESS])
+ random_ether_addr(dev->dev_addr);
+
+ mtu = ipgre_tunnel_bind_dev(dev);
+ if (!tb[IFLA_MTU])
+ dev->mtu = mtu;
+
+ err = register_netdevice(dev);
+ if (err)
+ goto out;
+
+ dev_hold(dev);
+ ipgre_tunnel_link(ign, nt);
+
+out:
+ return err;
+}
+
+static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
+ struct nlattr *data[])
+{
+ struct ip_tunnel *t, *nt;
+ struct net *net = dev_net(dev);
+ struct ipgre_net *ign = net_generic(net, ipgre_net_id);
+ struct ip_tunnel_parm p;
+ int mtu;
+
+ if (dev == ign->fb_tunnel_dev)
+ return -EINVAL;
+
+ nt = netdev_priv(dev);
+ ipgre_netlink_parms(data, &p);
+
+ t = ipgre_tunnel_locate(net, &p, 0);
+
+ if (t) {
+ if (t->dev != dev)
+ return -EEXIST;
+ } else {
+ t = nt;
+
+ if (dev->type != ARPHRD_ETHER) {
+ unsigned nflags = 0;
+
+ if (ipv4_is_multicast(p.iph.daddr))
+ nflags = IFF_BROADCAST;
+ else if (p.iph.daddr)
+ nflags = IFF_POINTOPOINT;
+
+ if ((dev->flags ^ nflags) &
+ (IFF_POINTOPOINT | IFF_BROADCAST))
+ return -EINVAL;
+ }
+
+ ipgre_tunnel_unlink(ign, t);
+ t->parms.iph.saddr = p.iph.saddr;
+ t->parms.iph.daddr = p.iph.daddr;
+ t->parms.i_key = p.i_key;
+ if (dev->type != ARPHRD_ETHER) {
+ memcpy(dev->dev_addr, &p.iph.saddr, 4);
+ memcpy(dev->broadcast, &p.iph.daddr, 4);
+ }
+ ipgre_tunnel_link(ign, t);
+ netdev_state_change(dev);
+ }
+
+ t->parms.o_key = p.o_key;
+ t->parms.iph.ttl = p.iph.ttl;
+ t->parms.iph.tos = p.iph.tos;
+ t->parms.iph.frag_off = p.iph.frag_off;
+
+ if (t->parms.link != p.link) {
+ t->parms.link = p.link;
+ mtu = ipgre_tunnel_bind_dev(dev);
+ if (!tb[IFLA_MTU])
+ dev->mtu = mtu;
+ netdev_state_change(dev);
+ }
+
+ return 0;
+}
+
+static size_t ipgre_get_size(const struct net_device *dev)
+{
+ return
+ /* IFLA_GRE_LINK */
+ nla_total_size(4) +
+ /* IFLA_GRE_IFLAGS */
+ nla_total_size(2) +
+ /* IFLA_GRE_OFLAGS */
+ nla_total_size(2) +
+ /* IFLA_GRE_IKEY */
+ nla_total_size(4) +
+ /* IFLA_GRE_OKEY */
+ nla_total_size(4) +
+ /* IFLA_GRE_LOCAL */
+ nla_total_size(4) +
+ /* IFLA_GRE_REMOTE */
+ nla_total_size(4) +
+ /* IFLA_GRE_TTL */
+ nla_total_size(1) +
+ /* IFLA_GRE_TOS */
+ nla_total_size(1) +
+ /* IFLA_GRE_PMTUDISC */
+ nla_total_size(1) +
+ 0;
+}
+
+static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev)
+{
+ struct ip_tunnel *t = netdev_priv(dev);
+ struct ip_tunnel_parm *p = &t->parms;
+
+ NLA_PUT_U32(skb, IFLA_GRE_LINK, p->link);
+ NLA_PUT_BE16(skb, IFLA_GRE_IFLAGS, p->i_flags);
+ NLA_PUT_BE16(skb, IFLA_GRE_OFLAGS, p->o_flags);
+ NLA_PUT_BE32(skb, IFLA_GRE_IKEY, p->i_key);
+ NLA_PUT_BE32(skb, IFLA_GRE_OKEY, p->o_key);
+ NLA_PUT_BE32(skb, IFLA_GRE_LOCAL, p->iph.saddr);
+ NLA_PUT_BE32(skb, IFLA_GRE_REMOTE, p->iph.daddr);
+ NLA_PUT_U8(skb, IFLA_GRE_TTL, p->iph.ttl);
+ NLA_PUT_U8(skb, IFLA_GRE_TOS, p->iph.tos);
+ NLA_PUT_U8(skb, IFLA_GRE_PMTUDISC, !!(p->iph.frag_off & htons(IP_DF)));
+
+ return 0;
+
+nla_put_failure:
+ return -EMSGSIZE;
+}
+
+static const struct nla_policy ipgre_policy[IFLA_GRE_MAX + 1] = {
+ [IFLA_GRE_LINK] = { .type = NLA_U32 },
+ [IFLA_GRE_IFLAGS] = { .type = NLA_U16 },
+ [IFLA_GRE_OFLAGS] = { .type = NLA_U16 },
+ [IFLA_GRE_IKEY] = { .type = NLA_U32 },
+ [IFLA_GRE_OKEY] = { .type = NLA_U32 },
+ [IFLA_GRE_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
+ [IFLA_GRE_REMOTE] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
+ [IFLA_GRE_TTL] = { .type = NLA_U8 },
+ [IFLA_GRE_TOS] = { .type = NLA_U8 },
+ [IFLA_GRE_PMTUDISC] = { .type = NLA_U8 },
+};
+
+static struct rtnl_link_ops ipgre_link_ops __read_mostly = {
+ .kind = "gre",
+ .maxtype = IFLA_GRE_MAX,
+ .policy = ipgre_policy,
+ .priv_size = sizeof(struct ip_tunnel),
+ .setup = ipgre_tunnel_setup,
+ .validate = ipgre_tunnel_validate,
+ .newlink = ipgre_newlink,
+ .changelink = ipgre_changelink,
+ .get_size = ipgre_get_size,
+ .fill_info = ipgre_fill_info,
+};
+
+static struct rtnl_link_ops ipgre_tap_ops __read_mostly = {
+ .kind = "gretap",
+ .maxtype = IFLA_GRE_MAX,
+ .policy = ipgre_policy,
+ .priv_size = sizeof(struct ip_tunnel),
+ .setup = ipgre_tap_setup,
+ .validate = ipgre_tap_validate,
+ .newlink = ipgre_newlink,
+ .changelink = ipgre_changelink,
+ .get_size = ipgre_get_size,
+ .fill_info = ipgre_fill_info,
};
/*
return -EAGAIN;
}
- err = register_pernet_gen_device(&ipgre_net_id, &ipgre_net_ops);
+ err = register_pernet_device(&ipgre_net_ops);
+ if (err < 0)
+ goto gen_device_failed;
+
+ err = rtnl_link_register(&ipgre_link_ops);
+ if (err < 0)
+ goto rtnl_link_failed;
+
+ err = rtnl_link_register(&ipgre_tap_ops);
if (err < 0)
- inet_del_protocol(&ipgre_protocol, IPPROTO_GRE);
+ goto tap_ops_failed;
+out:
return err;
+
+tap_ops_failed:
+ rtnl_link_unregister(&ipgre_link_ops);
+rtnl_link_failed:
+ unregister_pernet_device(&ipgre_net_ops);
+gen_device_failed:
+ inet_del_protocol(&ipgre_protocol, IPPROTO_GRE);
+ goto out;
}
static void __exit ipgre_fini(void)
{
+ rtnl_link_unregister(&ipgre_tap_ops);
+ rtnl_link_unregister(&ipgre_link_ops);
+ unregister_pernet_device(&ipgre_net_ops);
if (inet_del_protocol(&ipgre_protocol, IPPROTO_GRE) < 0)
printk(KERN_INFO "ipgre close: can't remove protocol\n");
-
- unregister_pernet_gen_device(ipgre_net_id, &ipgre_net_ops);
}
module_init(ipgre_init);
module_exit(ipgre_fini);
MODULE_LICENSE("GPL");
+MODULE_ALIAS_RTNL_LINK("gre");
+MODULE_ALIAS_RTNL_LINK("gretap");
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff