diff options
| author | Eric Dumazet <edumazet@google.com> | 2016-01-08 09:35:51 -0800 |
|---|---|---|
| committer | Ben Hutchings <ben@decadent.org.uk> | 2016-02-13 10:34:08 +0000 |
| commit | 0602b3e4e73ad7124d1c1a389bf38b99c8717cff (patch) | |
| tree | 4066b26562a5681876461765b86f75db0b90c072 /net/ipv6 | |
| parent | 15dabec9d87dd441d6f0ce49fbbef7fcc5f1ec39 (diff) | |
| download | lwn-0602b3e4e73ad7124d1c1a389bf38b99c8717cff.tar.gz lwn-0602b3e4e73ad7124d1c1a389bf38b99c8717cff.zip | |
ipv6: tcp: add rcu locking in tcp_v6_send_synack()
commit 3e4006f0b86a5ae5eb0e8215f9a9e1db24506977 upstream.
When first SYNACK is sent, we already hold rcu_read_lock(), but this
is not true if a SYNACK is retransmitted, as a timer (soft) interrupt
does not hold rcu_read_lock()
Fixes: 45f6fad84cc30 ("ipv6: add complete rcu protection around np->opt")
Reported-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
[bwh: Backported to 3.2: adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Diffstat (limited to 'net/ipv6')
| -rw-r--r-- | net/ipv6/tcp_ipv6.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 01706792f282..b20f7a08b4f7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -513,8 +513,10 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, __tcp_v6_send_check(skb, &treq->loc_addr, &treq->rmt_addr); ipv6_addr_copy(&fl6.daddr, &treq->rmt_addr); + rcu_read_lock(); err = ip6_xmit(sk, skb, &fl6, rcu_dereference(np->opt), np->tclass); + rcu_read_unlock(); err = net_xmit_eval(err); } |