]> git.ipfire.org Git - thirdparty/kernel/stable.git/commitdiff
net: bridge: multicast: fix PIM hello router port marking race
authorNikolay Aleksandrov <nikolay@nvidia.com>
Sun, 11 Jul 2021 09:56:28 +0000 (12:56 +0300)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 20 Jul 2021 14:02:03 +0000 (16:02 +0200)
commit 04bef83a3358946bfc98a5ecebd1b0003d83d882 upstream.

When a PIM hello packet is received on a bridge port with multicast
snooping enabled, we mark it as a router port automatically, that
includes adding that port the router port list. The multicast lock
protects that list, but it is not acquired in the PIM message case
leading to a race condition, we need to take it to fix the race.

Cc: stable@vger.kernel.org
Fixes: 91b02d3d133b ("bridge: mcast: add router port on PIM hello message")
Signed-off-by: Nikolay Aleksandrov <nikolay@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/bridge/br_multicast.c

index 226bb05c3b42df784ca8d5299a863a14d42f9f30..e27fe6e6ecd400fb39e4ece61fdc9a609b7e6fd3 100644 (file)
@@ -3087,7 +3087,9 @@ static void br_multicast_pim(struct net_bridge *br,
            pim_hdr_type(pimhdr) != PIM_TYPE_HELLO)
                return;
 
+       spin_lock(&br->multicast_lock);
        br_multicast_mark_router(br, port);
+       spin_unlock(&br->multicast_lock);
 }
 
 static int br_ip4_multicast_mrd_rcv(struct net_bridge *br,