net: ipv6: tcp: fix potential use after free in tcp_v6_do_rcv
authorDaniel Borkmann <dborkman@redhat.com>
Tue, 3 Sep 2013 17:29:12 +0000 (19:29 +0200)
committerDavid S. Miller <davem@davemloft.net>
Wed, 4 Sep 2013 18:44:41 +0000 (14:44 -0400)
commit3a1c756590633c0e86df606e5c618c190926a0df
tree970ceb0f62cb1e9df1d710b69387374a2be0fae5
parent25a6e6b84fba601eff7c28d30da8ad7cfbef0d43
net: ipv6: tcp: fix potential use after free in tcp_v6_do_rcv

In tcp_v6_do_rcv() code, when processing pkt options, we soley work
on our skb clone opt_skb that we've created earlier before entering
tcp_rcv_established() on our way. However, only in condition ...

  if (np->rxopt.bits.rxtclass)
    np->rcv_tclass = ipv6_get_dsfield(ipv6_hdr(skb));

... we work on skb itself. As we extract every other information out
of opt_skb in ipv6_pktoptions path, this seems wrong, since skb can
already be released by tcp_rcv_established() earlier on. When we try
to access it in ipv6_hdr(), we will dereference freed skb.

[ Bug added by commit 4c507d2897bd9b ("net: implement IP_RECVTOS for
  IP_PKTOPTIONS") ]

Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv6/tcp_ipv6.c