xfrm: Fix transport mode skb control buffer usage.
authorSteffen Klassert <steffen.klassert@secunet.com>
Mon, 19 Mar 2018 06:15:39 +0000 (07:15 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 30 May 2018 05:52:19 +0000 (07:52 +0200)
[ Upstream commit 9a3fb9fb84cc30577c1b012a6a3efda944684291 ]

A recent commit introduced a new struct xfrm_trans_cb
that is used with the sk_buff control buffer. Unfortunately
it placed the structure in front of the control buffer and
overlooked that the IPv4/IPv6 control buffer is still needed
for some layer 4 protocols. As a result the IPv4/IPv6 control
buffer is overwritten with this structure. Fix this by setting
a apropriate header in front of the structure.

Fixes acf568ee859f ("xfrm: Reinject transport-mode packets ...")
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/xfrm/xfrm_input.c

index 5b2409746ae0a14c491cc8dfd316facc5cd67347..9f492dc417d59f09c2378ce1e136ddf1ad7d6e79 100644 (file)
@@ -26,6 +26,12 @@ struct xfrm_trans_tasklet {
 };
 
 struct xfrm_trans_cb {
+       union {
+               struct inet_skb_parm    h4;
+#if IS_ENABLED(CONFIG_IPV6)
+               struct inet6_skb_parm   h6;
+#endif
+       } header;
        int (*finish)(struct net *net, struct sock *sk, struct sk_buff *skb);
 };