ipv6: Fix 'release_it' logic in tcp_v6_get_peer()

We accidently set it to "true" for the case where we
are using a route bound peer.

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 319458558df90863635648a76680f63bc1790933..fee0768916462a8d6ee7688376c31b61cf347461 100644 (file)
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1877,7 +1877,7 @@ static struct inet_peer *tcp_v6_get_peer(struct sock *sk, bool *release_it)
                if (!rt->rt6i_peer)
                        rt6_bind_peer(rt, 1);
                peer = rt->rt6i_peer;
-               *release_it = true;
+               *release_it = false;
        }
 
        return peer;