OSDN Git Service

netfilter: don't track fragmented packets
authorFlorian Westphal <fw@strlen.de>
Fri, 3 Mar 2017 20:44:00 +0000 (21:44 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 14 Dec 2017 08:28:21 +0000 (09:28 +0100)
commit8818eb851c2d9c9e503208f5b054e42b4a83fe24
tree95deff2c4a7064155d6f4ac291501cd0ea37d95a
parent36a14374b5684ad2a09d9f89adef3ab5e114cf0d
netfilter: don't track fragmented packets

[ Upstream commit 7b4fdf77a450ec0fdcb2f677b080ddbf2c186544 ]

Andrey reports syzkaller splat caused by

NF_CT_ASSERT(!ip_is_fragment(ip_hdr(skb)));

in ipv4 nat.  But this assertion (and the comment) are wrong, this function
does see fragments when IP_NODEFRAG setsockopt is used.

As conntrack doesn't track packets without complete l4 header, only the
first fragment is tracked.

Because applying nat to first packet but not the rest makes no sense this
also turns off tracking of all fragments.

Reported-by: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
net/ipv4/netfilter/nf_nat_l3proto_ipv4.c