#define HASH_SIZE 16
#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF)
-static void ipip6_fb_tunnel_init(struct net_device *dev);
static void ipip6_tunnel_init(struct net_device *dev);
static void ipip6_tunnel_setup(struct net_device *dev);
-static int sit_net_id;
+static int sit_net_id __read_mostly;
struct sit_net {
struct ip_tunnel *tunnels_r_l[HASH_SIZE];
struct ip_tunnel *tunnels_r[HASH_SIZE];
struct net_device *fb_tunnel_dev;
};
-static DEFINE_RWLOCK(ipip6_lock);
+/*
+ * Locking : hash tables are protected by RCU and a spinlock
+ */
+static DEFINE_SPINLOCK(ipip6_lock);
+
+#define for_each_ip_tunnel_rcu(start) \
+ for (t = rcu_dereference(start); t; t = rcu_dereference(t->next))
+/*
+ * Must be invoked with rcu_read_lock
+ */
static struct ip_tunnel * ipip6_tunnel_lookup(struct net *net,
struct net_device *dev, __be32 remote, __be32 local)
{
struct ip_tunnel *t;
struct sit_net *sitn = net_generic(net, sit_net_id);
- for (t = sitn->tunnels_r_l[h0^h1]; t; t = t->next) {
+ for_each_ip_tunnel_rcu(sitn->tunnels_r_l[h0 ^ h1]) {
if (local == t->parms.iph.saddr &&
remote == t->parms.iph.daddr &&
(!dev || !t->parms.link || dev->iflink == t->parms.link) &&
(t->dev->flags & IFF_UP))
return t;
}
- for (t = sitn->tunnels_r[h0]; t; t = t->next) {
+ for_each_ip_tunnel_rcu(sitn->tunnels_r[h0]) {
if (remote == t->parms.iph.daddr &&
(!dev || !t->parms.link || dev->iflink == t->parms.link) &&
(t->dev->flags & IFF_UP))
return t;
}
- for (t = sitn->tunnels_l[h1]; t; t = t->next) {
+ for_each_ip_tunnel_rcu(sitn->tunnels_l[h1]) {
if (local == t->parms.iph.saddr &&
(!dev || !t->parms.link || dev->iflink == t->parms.link) &&
(t->dev->flags & IFF_UP))
return t;
}
- t = sitn->tunnels_wc[0];
+ t = rcu_dereference(sitn->tunnels_wc[0]);
if ((t != NULL) && (t->dev->flags & IFF_UP))
return t;
return NULL;
for (tp = ipip6_bucket(sitn, t); *tp; tp = &(*tp)->next) {
if (t == *tp) {
- write_lock_bh(&ipip6_lock);
+ spin_lock_bh(&ipip6_lock);
*tp = t->next;
- write_unlock_bh(&ipip6_lock);
+ spin_unlock_bh(&ipip6_lock);
break;
}
}
{
struct ip_tunnel **tp = ipip6_bucket(sitn, t);
+ spin_lock_bh(&ipip6_lock);
t->next = *tp;
- write_lock_bh(&ipip6_lock);
- *tp = t;
- write_unlock_bh(&ipip6_lock);
+ rcu_assign_pointer(*tp, t);
+ spin_unlock_bh(&ipip6_lock);
}
static void ipip6_tunnel_clone_6rd(struct net_device *dev, struct sit_net *sitn)
return NULL;
}
+static DEFINE_SPINLOCK(ipip6_prl_lock);
+
+#define for_each_prl_rcu(start) \
+ for (prl = rcu_dereference(start); \
+ prl; \
+ prl = rcu_dereference(prl->next))
+
static struct ip_tunnel_prl_entry *
__ipip6_tunnel_locate_prl(struct ip_tunnel *t, __be32 addr)
{
- struct ip_tunnel_prl_entry *p = (struct ip_tunnel_prl_entry *)NULL;
+ struct ip_tunnel_prl_entry *prl;
- for (p = t->prl; p; p = p->next)
- if (p->addr == addr)
+ for_each_prl_rcu(t->prl)
+ if (prl->addr == addr)
break;
- return p;
+ return prl;
}
kcalloc(cmax, sizeof(*kp), GFP_KERNEL) :
NULL;
- read_lock(&ipip6_lock);
+ rcu_read_lock();
ca = t->prl_count < cmax ? t->prl_count : cmax;
}
c = 0;
- for (prl = t->prl; prl; prl = prl->next) {
+ for_each_prl_rcu(t->prl) {
if (c >= cmax)
break;
if (kprl.addr != htonl(INADDR_ANY) && prl->addr != kprl.addr)
break;
}
out:
- read_unlock(&ipip6_lock);
+ rcu_read_unlock();
len = sizeof(*kp) * c;
ret = 0;
if (a->addr == htonl(INADDR_ANY))
return -EINVAL;
- write_lock(&ipip6_lock);
+ spin_lock(&ipip6_prl_lock);
for (p = t->prl; p; p = p->next) {
if (p->addr == a->addr) {
- if (chg)
- goto update;
+ if (chg) {
+ p->flags = a->flags;
+ goto out;
+ }
err = -EEXIST;
goto out;
}
goto out;
}
+ INIT_RCU_HEAD(&p->rcu_head);
p->next = t->prl;
- t->prl = p;
- t->prl_count++;
-update:
p->addr = a->addr;
p->flags = a->flags;
+ t->prl_count++;
+ rcu_assign_pointer(t->prl, p);
out:
- write_unlock(&ipip6_lock);
+ spin_unlock(&ipip6_prl_lock);
return err;
}
+static void prl_entry_destroy_rcu(struct rcu_head *head)
+{
+ kfree(container_of(head, struct ip_tunnel_prl_entry, rcu_head));
+}
+
+static void prl_list_destroy_rcu(struct rcu_head *head)
+{
+ struct ip_tunnel_prl_entry *p, *n;
+
+ p = container_of(head, struct ip_tunnel_prl_entry, rcu_head);
+ do {
+ n = p->next;
+ kfree(p);
+ p = n;
+ } while (p);
+}
+
static int
ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
{
struct ip_tunnel_prl_entry *x, **p;
int err = 0;
- write_lock(&ipip6_lock);
+ spin_lock(&ipip6_prl_lock);
if (a && a->addr != htonl(INADDR_ANY)) {
for (p = &t->prl; *p; p = &(*p)->next) {
if ((*p)->addr == a->addr) {
x = *p;
*p = x->next;
- kfree(x);
+ call_rcu(&x->rcu_head, prl_entry_destroy_rcu);
t->prl_count--;
goto out;
}
}
err = -ENXIO;
} else {
- while (t->prl) {
+ if (t->prl) {
+ t->prl_count = 0;
x = t->prl;
- t->prl = t->prl->next;
- kfree(x);
- t->prl_count--;
+ call_rcu(&x->rcu_head, prl_list_destroy_rcu);
+ t->prl = NULL;
}
}
out:
- write_unlock(&ipip6_lock);
+ spin_unlock(&ipip6_prl_lock);
return err;
}
struct ip_tunnel_prl_entry *p;
int ok = 1;
- read_lock(&ipip6_lock);
+ rcu_read_lock();
p = __ipip6_tunnel_locate_prl(t, iph->saddr);
if (p) {
if (p->flags & PRL_DEFAULT)
else
ok = 0;
}
- read_unlock(&ipip6_lock);
+ rcu_read_unlock();
return ok;
}
struct sit_net *sitn = net_generic(net, sit_net_id);
if (dev == sitn->fb_tunnel_dev) {
- write_lock_bh(&ipip6_lock);
+ spin_lock_bh(&ipip6_lock);
sitn->tunnels_wc[0] = NULL;
- write_unlock_bh(&ipip6_lock);
+ spin_unlock_bh(&ipip6_lock);
dev_put(dev);
} else {
ipip6_tunnel_unlink(sitn, netdev_priv(dev));
err = -ENOENT;
- read_lock(&ipip6_lock);
+ rcu_read_lock();
t = ipip6_tunnel_lookup(dev_net(skb->dev),
skb->dev,
iph->daddr,
t->err_count = 1;
t->err_time = jiffies;
out:
- read_unlock(&ipip6_lock);
+ rcu_read_unlock();
return err;
}
iph = ip_hdr(skb);
- read_lock(&ipip6_lock);
+ rcu_read_lock();
tunnel = ipip6_tunnel_lookup(dev_net(skb->dev), skb->dev,
iph->saddr, iph->daddr);
if (tunnel != NULL) {
if ((tunnel->dev->priv_flags & IFF_ISATAP) &&
!isatap_chksrc(skb, iph, tunnel)) {
tunnel->dev->stats.rx_errors++;
- read_unlock(&ipip6_lock);
+ rcu_read_unlock();
kfree_skb(skb);
return 0;
}
nf_reset(skb);
ipip6_ecn_decapsulate(iph, skb);
netif_rx(skb);
- read_unlock(&ipip6_lock);
+ rcu_read_unlock();
return 0;
}
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
- read_unlock(&ipip6_lock);
+ rcu_read_unlock();
out:
kfree_skb(skb);
return 0;
struct iphdr *tiph = &tunnel->parms.iph;
struct ipv6hdr *iph6 = ipv6_hdr(skb);
u8 tos = tunnel->parms.iph.tos;
+ __be16 df = tiph->frag_off;
struct rtable *rt; /* Route to the other host */
struct net_device *tdev; /* Device to other host */
struct iphdr *iph; /* Our new IP header */
goto tx_error;
}
- if (tiph->frag_off)
+ if (df) {
mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr);
- else
- mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu;
- if (mtu < 68) {
- stats->collisions++;
- ip_rt_put(rt);
- goto tx_error;
- }
- if (mtu < IPV6_MIN_MTU)
- mtu = IPV6_MIN_MTU;
- if (tunnel->parms.iph.daddr && skb_dst(skb))
- skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu);
+ if (mtu < 68) {
+ stats->collisions++;
+ ip_rt_put(rt);
+ goto tx_error;
+ }
- if (skb->len > mtu) {
- icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev);
- ip_rt_put(rt);
- goto tx_error;
+ if (mtu < IPV6_MIN_MTU) {
+ mtu = IPV6_MIN_MTU;
+ df = 0;
+ }
+
+ if (tunnel->parms.iph.daddr && skb_dst(skb))
+ skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu);
+
+ if (skb->len > mtu) {
+ icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev);
+ ip_rt_put(rt);
+ goto tx_error;
+ }
}
if (tunnel->err_count > 0) {
iph = ip_hdr(skb);
iph->version = 4;
iph->ihl = sizeof(struct iphdr)>>2;
- if (mtu > IPV6_MIN_MTU)
- iph->frag_off = tiph->frag_off;
- else
- iph->frag_off = 0;
-
+ iph->frag_off = df;
iph->protocol = IPPROTO_IPV6;
iph->tos = INET_ECN_encapsulate(tos, ipv6_get_dsfield(iph6));
iph->daddr = rt->rt_dst;
ipip6_tunnel_bind_dev(dev);
}
-static void ipip6_fb_tunnel_init(struct net_device *dev)
+static void __net_init ipip6_fb_tunnel_init(struct net_device *dev)
{
struct ip_tunnel *tunnel = netdev_priv(dev);
struct iphdr *iph = &tunnel->parms.iph;
.priority = 1,
};
-static void sit_destroy_tunnels(struct sit_net *sitn)
+static void __net_exit sit_destroy_tunnels(struct sit_net *sitn, struct list_head *head)
{
int prio;
for (prio = 1; prio < 4; prio++) {
int h;
for (h = 0; h < HASH_SIZE; h++) {
- struct ip_tunnel *t;
- while ((t = sitn->tunnels[prio][h]) != NULL)
- unregister_netdevice(t->dev);
+ struct ip_tunnel *t = sitn->tunnels[prio][h];
+
+ while (t != NULL) {
+ unregister_netdevice_queue(t->dev, head);
+ t = t->next;
+ }
}
}
}
-static int sit_init_net(struct net *net)
+static int __net_init sit_init_net(struct net *net)
{
+ struct sit_net *sitn = net_generic(net, sit_net_id);
int err;
- struct sit_net *sitn;
-
- err = -ENOMEM;
- sitn = kzalloc(sizeof(struct sit_net), GFP_KERNEL);
- if (sitn == NULL)
- goto err_alloc;
-
- err = net_assign_generic(net, sit_net_id, sitn);
- if (err < 0)
- goto err_assign;
sitn->tunnels[0] = sitn->tunnels_wc;
sitn->tunnels[1] = sitn->tunnels_l;
dev_put(sitn->fb_tunnel_dev);
free_netdev(sitn->fb_tunnel_dev);
err_alloc_dev:
- /* nothing */
-err_assign:
- kfree(sitn);
-err_alloc:
return err;
}
-static void sit_exit_net(struct net *net)
+static void __net_exit sit_exit_net(struct net *net)
{
- struct sit_net *sitn;
+ struct sit_net *sitn = net_generic(net, sit_net_id);
+ LIST_HEAD(list);
- sitn = net_generic(net, sit_net_id);
rtnl_lock();
- sit_destroy_tunnels(sitn);
- unregister_netdevice(sitn->fb_tunnel_dev);
+ sit_destroy_tunnels(sitn, &list);
+ unregister_netdevice_queue(sitn->fb_tunnel_dev, &list);
+ unregister_netdevice_many(&list);
rtnl_unlock();
- kfree(sitn);
}
static struct pernet_operations sit_net_ops = {
.init = sit_init_net,
.exit = sit_exit_net,
+ .id = &sit_net_id,
+ .size = sizeof(struct sit_net),
};
static void __exit sit_cleanup(void)
{
xfrm4_tunnel_deregister(&sit_handler, AF_INET6);
- unregister_pernet_gen_device(sit_net_id, &sit_net_ops);
+ unregister_pernet_device(&sit_net_ops);
+ rcu_barrier(); /* Wait for completion of call_rcu()'s */
}
static int __init sit_init(void)
return -EAGAIN;
}
- err = register_pernet_gen_device(&sit_net_id, &sit_net_ops);
+ err = register_pernet_device(&sit_net_ops);
if (err < 0)
xfrm4_tunnel_deregister(&sit_handler, AF_INET6);
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff