vxlan: lock RCU on TX path
[ Upstream commit 56de859e9967c070464a9a9f4f18d73f9447298e ]
There is no guarantees that callers of the TX path will hold
the RCU lock. Grab it explicitly.
Fixes: c6fcc4fc5f8b ("vxlan: avoid using stale vxlan socket.")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index ac3f8d5..3c4c2cf 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1955,6 +1955,7 @@
info = skb_tunnel_info(skb);
+ rcu_read_lock();
if (rdst) {
dst_port = rdst->remote_port ? rdst->remote_port : vxlan->cfg.dst_port;
vni = rdst->remote_vni;
@@ -1985,7 +1986,7 @@
if (did_rsc) {
/* short-circuited back to local bridge */
vxlan_encap_bypass(skb, vxlan, vxlan);
- return;
+ goto out_unlock;
}
goto drop;
}
@@ -2054,7 +2055,7 @@
if (!dst_vxlan)
goto tx_error;
vxlan_encap_bypass(skb, vxlan, dst_vxlan);
- return;
+ goto out_unlock;
}
if (!info)
@@ -2115,7 +2116,7 @@
if (!dst_vxlan)
goto tx_error;
vxlan_encap_bypass(skb, vxlan, dst_vxlan);
- return;
+ goto out_unlock;
}
if (!info)
@@ -2129,7 +2130,7 @@
if (err < 0) {
dst_release(ndst);
dev->stats.tx_errors++;
- return;
+ goto out_unlock;
}
udp_tunnel6_xmit_skb(ndst, sk, skb, dev,
&local_ip.sin6.sin6_addr,
@@ -2137,7 +2138,8 @@
label, src_port, dst_port, !udp_sum);
#endif
}
-
+out_unlock:
+ rcu_read_unlock();
return;
drop:
@@ -2153,6 +2155,7 @@
dev->stats.tx_errors++;
tx_free:
dev_kfree_skb(skb);
+ rcu_read_unlock();
}
/* Transmit local packets over Vxlan