syzbot found a crash in tcp_v6_send_reset() caused by my latest
change.
Problem is that if an skb has been queued to socket prequeue,
skb_dst(skb)->dev can not anymore point to the device.
Fortunately in this case the socket pointer is not NULL.
A similar issue has been fixed in commit 0f85feae6b71 ("tcp: fix
more NULL deref after prequeue changes"), I should have known better.
Fixes: 323a53c41292 ("ipv6: tcp: enable flowlabel reflection in some RST
packets")
Signed-off-by: Eric Dumazet <eduma...@google.com>
Reported-by: syzbot <syzkal...@googlegroups.com>
---
net/ipv6/tcp_ipv6.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index
d8d18386c99a82e112dd3a1aeee01e4c328ba5d7..c1da52c7f990f2fa3e020e3f3a33934149ad225e
100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -934,7 +934,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct
sk_buff *skb)
if (!sk && !ipv6_unicast_destination(skb))
return;
- net = dev_net(skb_dst(skb)->dev);
+ net = sk ? sock_net(sk) : dev_net(skb_dst(skb)->dev);
#ifdef CONFIG_TCP_MD5SIG
rcu_read_lock();
hash_location = tcp_parse_md5sig_option(th);
--
2.22.0.rc2.383.gf4fbbf30c2-goog
