ip_tunnel: replace dst_cache with generic implementation
The current ip_tunnel cache implementation is prone to a race
that will cause the wrong dst to be cached on cuncurrent dst cache
miss and ip tunnel update via netlink.
Replacing with the generic implementation fix the issue.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index c7bd72e..4569da7 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -68,61 +68,6 @@
IP_TNL_HASH_BITS);
}
-static void __tunnel_dst_set(struct ip_tunnel_dst *idst,
- struct dst_entry *dst, __be32 saddr)
-{
- struct dst_entry *old_dst;
-
- dst_clone(dst);
- old_dst = xchg((__force struct dst_entry **)&idst->dst, dst);
- dst_release(old_dst);
- idst->saddr = saddr;
-}
-
-static noinline void tunnel_dst_set(struct ip_tunnel *t,
- struct dst_entry *dst, __be32 saddr)
-{
- __tunnel_dst_set(raw_cpu_ptr(t->dst_cache), dst, saddr);
-}
-
-static void tunnel_dst_reset(struct ip_tunnel *t)
-{
- tunnel_dst_set(t, NULL, 0);
-}
-
-void ip_tunnel_dst_reset_all(struct ip_tunnel *t)
-{
- int i;
-
- for_each_possible_cpu(i)
- __tunnel_dst_set(per_cpu_ptr(t->dst_cache, i), NULL, 0);
-}
-EXPORT_SYMBOL(ip_tunnel_dst_reset_all);
-
-static struct rtable *tunnel_rtable_get(struct ip_tunnel *t,
- u32 cookie, __be32 *saddr)
-{
- struct ip_tunnel_dst *idst;
- struct dst_entry *dst;
-
- rcu_read_lock();
- idst = raw_cpu_ptr(t->dst_cache);
- dst = rcu_dereference(idst->dst);
- if (dst && !atomic_inc_not_zero(&dst->__refcnt))
- dst = NULL;
- if (dst) {
- if (!dst->obsolete || dst->ops->check(dst, cookie)) {
- *saddr = idst->saddr;
- } else {
- tunnel_dst_reset(t);
- dst_release(dst);
- dst = NULL;
- }
- }
- rcu_read_unlock();
- return (struct rtable *)dst;
-}
-
static bool ip_tunnel_key_match(const struct ip_tunnel_parm *p,
__be16 flags, __be32 key)
{
@@ -381,7 +326,8 @@
if (!IS_ERR(rt)) {
tdev = rt->dst.dev;
- tunnel_dst_set(tunnel, &rt->dst, fl4.saddr);
+ dst_cache_set_ip4(&tunnel->dst_cache, &rt->dst,
+ fl4.saddr);
ip_rt_put(rt);
}
if (dev->type != ARPHRD_ETHER)
@@ -729,7 +675,8 @@
if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0)
goto tx_error;
- rt = connected ? tunnel_rtable_get(tunnel, 0, &fl4.saddr) : NULL;
+ rt = connected ? dst_cache_get_ip4(&tunnel->dst_cache, &fl4.saddr) :
+ NULL;
if (!rt) {
rt = ip_route_output_key(tunnel->net, &fl4);
@@ -739,7 +686,8 @@
goto tx_error;
}
if (connected)
- tunnel_dst_set(tunnel, &rt->dst, fl4.saddr);
+ dst_cache_set_ip4(&tunnel->dst_cache, &rt->dst,
+ fl4.saddr);
}
if (rt->dst.dev == dev) {
@@ -836,7 +784,7 @@
if (set_mtu)
dev->mtu = mtu;
}
- ip_tunnel_dst_reset_all(t);
+ dst_cache_reset(&t->dst_cache);
netdev_state_change(dev);
}
@@ -961,7 +909,7 @@
struct ip_tunnel *tunnel = netdev_priv(dev);
gro_cells_destroy(&tunnel->gro_cells);
- free_percpu(tunnel->dst_cache);
+ dst_cache_destroy(&tunnel->dst_cache);
free_percpu(dev->tstats);
free_netdev(dev);
}
@@ -1155,15 +1103,15 @@
if (!dev->tstats)
return -ENOMEM;
- tunnel->dst_cache = alloc_percpu(struct ip_tunnel_dst);
- if (!tunnel->dst_cache) {
+ err = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
+ if (err) {
free_percpu(dev->tstats);
- return -ENOMEM;
+ return err;
}
err = gro_cells_init(&tunnel->gro_cells, dev);
if (err) {
- free_percpu(tunnel->dst_cache);
+ dst_cache_destroy(&tunnel->dst_cache);
free_percpu(dev->tstats);
return err;
}
@@ -1193,7 +1141,7 @@
if (itn->fb_tunnel_dev != dev)
ip_tunnel_del(itn, netdev_priv(dev));
- ip_tunnel_dst_reset_all(tunnel);
+ dst_cache_reset(&tunnel->dst_cache);
}
EXPORT_SYMBOL_GPL(ip_tunnel_uninit);