[PATCH 3.16.y-ckt 123/129] netlink: not trim skb for mmaped socket when dump

[Luis Henriques]

3.16.7-ckt25 -stable review patch.  If anyone has any objections, please let me know.

---8<------------------------------------------------------------

From: Ken-ichirou MATSUZAWA <chamaken at gmail.com>

commit aa3a022094fac7f6e48050e139fa8a5a2e3265ce upstream.

We should not trim skb for mmaped socket since its buf size is fixed
and userspace will read as frame which data equals head. mmaped
socket will not call recvmsg, means max_recvmsg_len is 0,
skb_reserve was not called before commit: db65a3aaf29e.

Fixes: db65a3aaf29e (netlink: Trim skb to alloc size to avoid MSG_TRUNC)
Signed-off-by: Ken-ichirou MATSUZAWA <chamas at h4.dion.ne.jp>
Signed-off-by: David S. Miller <davem at davemloft.net>
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 net/netlink/af_netlink.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 7e1ac5b5de0a..46cac8e31a16 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -2740,7 +2740,8 @@ static int netlink_dump(struct sock *sk)
 	 * reasonable static buffer based on the expected largest dump of a
 	 * single netdev. The outcome is MSG_TRUNC error.
 	 */
-	skb_reserve(skb, skb_tailroom(skb) - alloc_size);
+	if (!netlink_rx_is_mmaped(sk))
+		skb_reserve(skb, skb_tailroom(skb) - alloc_size);
 	netlink_skb_set_owner_r(skb, sk);
 
 	len = cb->dump(skb, cb);