diff options
author | David S. Miller <davem@davemloft.net> | 2017-03-07 14:59:24 -0800 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2017-03-07 15:00:37 -0800 |
commit | 8474c8caac7e06b50fa366022ecd109b955822cb (patch) | |
tree | 6de4691ad6c59a62c6f4d35530447112a2ff8bf1 /net/ipv6 | |
parent | 15e668070a64bb97f102ad9cf3bccbca0545cda8 (diff) | |
parent | e3dc847a5f85b43ee2bfc8eae407a7e383483228 (diff) | |
download | op-kernel-dev-8474c8caac7e06b50fa366022ecd109b955822cb.zip op-kernel-dev-8474c8caac7e06b50fa366022ecd109b955822cb.tar.gz |
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec
Steffen Klassert says:
====================
pull request (net): ipsec 2017-03-06
1) Fix lockdep splat on xfrm policy subsystem initialization.
From Florian Westphal.
2) When using socket policies on IPv4-mapped IPv6 addresses,
we access the flow informations of the wrong address family
what leads to an out of bounds access. Fix this by using
the family we get with the dst_entry, like we do it for the
standard policy lookup.
3) vti6 can report a PMTU below IPV6_MIN_MTU. Fix this by
adding a check for that before sending a ICMPV6_PKT_TOOBIG
message.
Please pull or let me know if there are problems.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv6')
-rw-r--r-- | net/ipv6/ip6_vti.c | 8 |
1 files changed, 6 insertions, 2 deletions
diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index 644ba59..3d8a3b6 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -485,11 +485,15 @@ vti6_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl) if (!skb->ignore_df && skb->len > mtu) { skb_dst(skb)->ops->update_pmtu(dst, NULL, skb, mtu); - if (skb->protocol == htons(ETH_P_IPV6)) + if (skb->protocol == htons(ETH_P_IPV6)) { + if (mtu < IPV6_MIN_MTU) + mtu = IPV6_MIN_MTU; + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); - else + } else { icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); + } return -EMSGSIZE; } |