This is a note to let you know that I've just added the patch titled
ipv6: Fix SO_REUSEPORT UDP socket with implicit sk_ipv6only
to the 4.14-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git%3Ba=su...
The filename of the patch is: ipv6-fix-so_reuseport-udp-socket-with-implicit-sk_ipv6only.patch and it can be found in the queue-4.14 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree, please let stable@vger.kernel.org know about it.
From foo@baz Wed Feb 7 11:29:20 PST 2018
From: Martin KaFai Lau kafai@fb.com Date: Wed, 24 Jan 2018 23:15:27 -0800 Subject: ipv6: Fix SO_REUSEPORT UDP socket with implicit sk_ipv6only
From: Martin KaFai Lau kafai@fb.com
[ Upstream commit 7ece54a60ee2ba7a386308cae73c790bd580589c ]
If a sk_v6_rcv_saddr is !IPV6_ADDR_ANY and !IPV6_ADDR_MAPPED, it implicitly implies it is an ipv6only socket. However, in inet6_bind(), this addr_type checking and setting sk->sk_ipv6only to 1 are only done after sk->sk_prot->get_port(sk, snum) has been completed successfully.
This inconsistency between sk_v6_rcv_saddr and sk_ipv6only confuses the 'get_port()'.
In particular, when binding SO_REUSEPORT UDP sockets, udp_reuseport_add_sock(sk,...) is called. udp_reuseport_add_sock() checks "ipv6_only_sock(sk2) == ipv6_only_sock(sk)" before adding sk to sk2->sk_reuseport_cb. In this case, ipv6_only_sock(sk2) could be 1 while ipv6_only_sock(sk) is still 0 here. The end result is, reuseport_alloc(sk) is called instead of adding sk to the existing sk2->sk_reuseport_cb.
It can be reproduced by binding two SO_REUSEPORT UDP sockets on an IPv6 address (!ANY and !MAPPED). Only one of the socket will receive packet.
The fix is to set the implicit sk_ipv6only before calling get_port(). The original sk_ipv6only has to be saved such that it can be restored in case get_port() failed. The situation is similar to the inet_reset_saddr(sk) after get_port() has failed.
Thanks to Calvin Owens calvinowens@fb.com who created an easy reproduction which leads to a fix.
Fixes: e32ea7e74727 ("soreuseport: fast reuseport UDP socket selection") Signed-off-by: Martin KaFai Lau kafai@fb.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/ipv6/af_inet6.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)
--- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -284,6 +284,7 @@ int inet6_bind(struct socket *sock, stru struct net *net = sock_net(sk); __be32 v4addr = 0; unsigned short snum; + bool saved_ipv6only; int addr_type = 0; int err = 0;
@@ -389,19 +390,21 @@ int inet6_bind(struct socket *sock, stru if (!(addr_type & IPV6_ADDR_MULTICAST)) np->saddr = addr->sin6_addr;
+ saved_ipv6only = sk->sk_ipv6only; + if (addr_type != IPV6_ADDR_ANY && addr_type != IPV6_ADDR_MAPPED) + sk->sk_ipv6only = 1; + /* Make sure we are allowed to bind here. */ if ((snum || !inet->bind_address_no_port) && sk->sk_prot->get_port(sk, snum)) { + sk->sk_ipv6only = saved_ipv6only; inet_reset_saddr(sk); err = -EADDRINUSE; goto out; }
- if (addr_type != IPV6_ADDR_ANY) { + if (addr_type != IPV6_ADDR_ANY) sk->sk_userlocks |= SOCK_BINDADDR_LOCK; - if (addr_type != IPV6_ADDR_MAPPED) - sk->sk_ipv6only = 1; - } if (snum) sk->sk_userlocks |= SOCK_BINDPORT_LOCK; inet->inet_sport = htons(inet->inet_num);
Patches currently in stable-queue which might be from kafai@fb.com are
queue-4.14/ipv6-fix-so_reuseport-udp-socket-with-implicit-sk_ipv6only.patch