ipvlan: introduce 'private' attribute for all existing modes.

IPvlan has always operated in bridge mode. However there are scenarios
where each slave should be able to talk through the master device but
not necessarily across each other. Think of an environment where each
of a namespace is a private and independant customer. In this scenario
the machine which is hosting these namespaces neither want to tell who
their neighbor is nor the individual namespaces care to talk to neighbor
on short-circuited network path.

This patch implements the mode that is very similar to the 'private' mode
in macvlan where individual slaves can send and receive traffic through
the master device, just that they can not talk among slave devices.

Signed-off-by: Mahesh Bandewar <maheshb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 64 insertions, 5 deletions

diff --git a/drivers/net/ipvlan/ipvlan.h b/drivers/net/ipvlan/ipvlan.h
index ba8173a0b62e..9941851bcc13 100644
--- a/drivers/net/ipvlan/ipvlan.h
+++ b/drivers/net/ipvlan/ipvlan.h
@@ -96,6 +96,7 @@ struct ipvl_port {
 	struct hlist_head	hlhead[IPVLAN_HASH_SIZE];
 	struct list_head	ipvlans;
 	u16			mode;
+	u16			flags;
 	u16			dev_id_start;
 	struct work_struct	wq;
 	struct sk_buff_head	backlog;
@@ -123,6 +124,21 @@ static inline struct ipvl_port *ipvlan_port_get_rtnl(const struct net_device *d)
 	return rtnl_dereference(d->rx_handler_data);
 }
 
+static inline bool ipvlan_is_private(const struct ipvl_port *port)
+{
+	return !!(port->flags & IPVLAN_F_PRIVATE);
+}
+
+static inline void ipvlan_mark_private(struct ipvl_port *port)
+{
+	port->flags |= IPVLAN_F_PRIVATE;
+}
+
+static inline void ipvlan_clear_private(struct ipvl_port *port)
+{
+	port->flags &= ~IPVLAN_F_PRIVATE;
+}
+
 void ipvlan_init_secret(void);
 unsigned int ipvlan_mac_hash(const unsigned char *addr);
 rx_handler_result_t ipvlan_handle_frame(struct sk_buff **pskb);
diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c
index 1f3295e274d0..72fd56de9c00 100644
--- a/drivers/net/ipvlan/ipvlan_core.c
+++ b/drivers/net/ipvlan/ipvlan_core.c
@@ -515,9 +515,13 @@ static int ipvlan_xmit_mode_l3(struct sk_buff *skb, struct net_device *dev)
 		goto out;
 
 	addr = ipvlan_addr_lookup(ipvlan->port, lyr3h, addr_type, true);
-	if (addr)
+	if (addr) {
+		if (ipvlan_is_private(ipvlan->port)) {
+			consume_skb(skb);
+			return NET_XMIT_DROP;
+		}
 		return ipvlan_rcv_frame(addr, &skb, true);
-
+	}
 out:
 	ipvlan_skb_crossing_ns(skb, ipvlan->phy_dev);
 	return ipvlan_process_outbound(skb);
@@ -535,8 +539,13 @@ static int ipvlan_xmit_mode_l2(struct sk_buff *skb, struct net_device *dev)
 		lyr3h = ipvlan_get_L3_hdr(skb, &addr_type);
 		if (lyr3h) {
 			addr = ipvlan_addr_lookup(ipvlan->port, lyr3h, addr_type, true);
-			if (addr)
+			if (addr) {
+				if (ipvlan_is_private(ipvlan->port)) {
+					consume_skb(skb);
+					return NET_XMIT_DROP;
+				}
 				return ipvlan_rcv_frame(addr, &skb, true);
+			}
 		}
 		skb = skb_share_check(skb, GFP_ATOMIC);
 		if (!skb)
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index f0ab55df57f1..4368afb1934c 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -462,11 +462,24 @@ static int ipvlan_nl_changelink(struct net_device *dev,
 	struct ipvl_port *port = ipvlan_port_get_rtnl(ipvlan->phy_dev);
 	int err = 0;
 
-	if (data && data[IFLA_IPVLAN_MODE]) {
+	if (!data)
+		return 0;
+
+	if (data[IFLA_IPVLAN_MODE]) {
 		u16 nmode = nla_get_u16(data[IFLA_IPVLAN_MODE]);
 
 		err = ipvlan_set_port_mode(port, nmode);
 	}
+
+	if (!err && data[IFLA_IPVLAN_FLAGS]) {
+		u16 flags = nla_get_u16(data[IFLA_IPVLAN_FLAGS]);
+
+		if (flags & IPVLAN_F_PRIVATE)
+			ipvlan_mark_private(port);
+		else
+			ipvlan_clear_private(port);
+	}
+
 	return err;
 }
 
@@ -474,18 +487,30 @@ static size_t ipvlan_nl_getsize(const struct net_device *dev)
 {
 	return (0
 		+ nla_total_size(2) /* IFLA_IPVLAN_MODE */
+		+ nla_total_size(2) /* IFLA_IPVLAN_FLAGS */
 		);
 }
 
 static int ipvlan_nl_validate(struct nlattr *tb[], struct nlattr *data[],
 			      struct netlink_ext_ack *extack)
 {
-	if (data && data[IFLA_IPVLAN_MODE]) {
+	if (!data)
+		return 0;
+
+	if (data[IFLA_IPVLAN_MODE]) {
 		u16 mode = nla_get_u16(data[IFLA_IPVLAN_MODE]);
 
 		if (mode < IPVLAN_MODE_L2 || mode >= IPVLAN_MODE_MAX)
 			return -EINVAL;
 	}
+	if (data[IFLA_IPVLAN_FLAGS]) {
+		u16 flags = nla_get_u16(data[IFLA_IPVLAN_FLAGS]);
+
+		/* Only one bit is used at this moment. */
+		if (flags & ~IPVLAN_F_PRIVATE)
+			return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -502,6 +527,8 @@ static int ipvlan_nl_fillinfo(struct sk_buff *skb,
 	ret = -EMSGSIZE;
 	if (nla_put_u16(skb, IFLA_IPVLAN_MODE, port->mode))
 		goto err;
+	if (nla_put_u16(skb, IFLA_IPVLAN_FLAGS, port->flags))
+		goto err;
 
 	return 0;
 
@@ -549,6 +576,12 @@ int ipvlan_link_new(struct net *src_net, struct net_device *dev,
 	ipvlan_adjust_mtu(ipvlan, phy_dev);
 	INIT_LIST_HEAD(&ipvlan->addrs);
 
+	/* Flags are per port and latest update overrides. User has
+	 * to be consistent in setting it just like the mode attribute.
+	 */
+	if (data && data[IFLA_IPVLAN_FLAGS])
+		ipvlan->port->flags = nla_get_u16(data[IFLA_IPVLAN_FLAGS]);
+
 	/* If the port-id base is at the MAX value, then wrap it around and
 	 * begin from 0x1 again. This may be due to a busy system where lots
 	 * of slaves are getting created and deleted.
@@ -644,6 +677,7 @@ EXPORT_SYMBOL_GPL(ipvlan_link_setup);
 static const struct nla_policy ipvlan_nl_policy[IFLA_IPVLAN_MAX + 1] =
 {
 	[IFLA_IPVLAN_MODE] = { .type = NLA_U16 },
+	[IFLA_IPVLAN_FLAGS] = { .type = NLA_U16 },
 };
 
 static struct rtnl_link_ops ipvlan_link_ops = {