[patch v8, kernel version 3.2.1] Source mode for macvlan interface

From: Stefan Gula
Date: Tue Feb 07 2012 - 03:38:30 EST


From: Stefan Gula <steweg@xxxxxxxxx>

New mode of macvlan interface called "source" allows one to set a list
of allowed mac address, which is used to match against source mac
address from received frames on underlying interface. This allows
creating mac based VLAN associations, instead of standard port or tag
based. The feature is useful to deploy 802.1x mac based behavior,
where drivers of underlying interfaces doesn't allows that.
Configuration is done through the netlink interface using e.g.:
ip link add link eth0 name macvlan0 type macvlan mode source
ip link add link eth0 name macvlan1 type macvlan mode source
ip link set link macvlan0 type macvlan macaddr add 00:11:11:11:11:11
ip link set link macvlan0 type macvlan macaddr add 00:22:22:22:22:22
ip link set link macvlan0 type macvlan macaddr add 00:33:33:33:33:33
ip link set link macvlan1 type macvlan macaddr add 00:33:33:33:33:33
ip link set link macvlan1 type macvlan macaddr add 00:44:44:44:44:44
This allows clients with MAC addresses 00:11:11:11:11:11,
00:22:22:22:22:22 to be part of only VLAN associated with macvlan0
interface. Clients with MAC addresses 00:44:44:44:44:44 with only VLAN
associated with macvlan1 interface. And client with MAC address
00:33:33:33:33:33 to be associated with both VLANs.

Signed-off-by: Stefan Gula <steweg@xxxxxxxxx>

---

V8 changes:
- added ability to list allowed mac address list back using devdump
ops method - required previous rtnetlink patch to be applied first
- rtnetlink patch is still under discussion... so please treat this
one as only example how it can work

diff -uprN -X linux-3.2.1-orig/Documentation/dontdiff linux-3.2.1-orig/drivers/net/macvlan.c linux-3.2.1-macvlan/drivers/net/macvlan.c
--- linux-3.2.1-orig/drivers/net/macvlan.c 2012-01-27 13:38:51.000000000 +0000
+++ linux-3.2.1-macvlan/drivers/net/macvlan.c 2012-02-03 13:46:34.000000000 +0000
@@ -40,6 +40,7 @@ struct macvlan_port {
struct rcu_head rcu;
bool passthru;
int count;
+ struct hlist_head vlan_source_hash[MACVLAN_HASH_SIZE];
};

static void macvlan_port_destroy(struct net_device *dev);
@@ -155,6 +156,117 @@ static void macvlan_broadcast(struct sk_
}
}

+struct macvlan_source_list {
+ struct hlist_node hlist;
+ struct macvlan_dev *vlan;
+ unsigned char addr[ETH_ALEN];
+ struct rcu_head rcu;
+};
+
+static struct macvlan_source_list *macvlan_hash_lookup_sources_list(
+ const struct macvlan_dev *vlan,
+ const unsigned char *addr)
+{
+ struct macvlan_source_list *list;
+ struct hlist_node *n;
+ struct hlist_head *h = &vlan->port->vlan_source_hash[addr[5]];
+
+ hlist_for_each_entry(list, n, h, hlist) {
+ if (!compare_ether_addr_64bits(list->addr, addr) &&
+ list->vlan == vlan)
+ return list;
+ }
+ return NULL;
+}
+
+static int macvlan_hash_add_sources(struct macvlan_dev *vlan,
+ const unsigned char *addr)
+{
+ struct macvlan_port *port = vlan->port;
+ struct macvlan_source_list *list;
+ struct hlist_head *h;
+
+ list = macvlan_hash_lookup_sources_list(vlan, addr);
+ if (!list) {
+ list = kmalloc(sizeof(*list), GFP_KERNEL);
+ if (list) {
+ memcpy(list->addr, addr, ETH_ALEN);
+ list->vlan = vlan;
+ h = &port->vlan_source_hash[addr[5]];
+ hlist_add_head_rcu(&list->hlist, h);
+ vlan->macaddr_count++;
+ } else {
+ return -ENOMEM;
+ }
+ }
+ return 0;
+}
+
+static void macvlan_hash_del_sources(struct macvlan_source_list *list)
+{
+ hlist_del_rcu(&list->hlist);
+ kfree_rcu(list, rcu);
+}
+
+static void macvlan_flush_sources(struct macvlan_port *port,
+ struct macvlan_dev *vlan)
+{
+ int i;
+
+ for (i = 0; i < MACVLAN_HASH_SIZE; i++) {
+ struct hlist_node *h, *n;
+
+ hlist_for_each_safe(h, n, &port->vlan_source_hash[i]) {
+ struct macvlan_source_list *list;
+
+ list = hlist_entry(h, struct macvlan_source_list,
+ hlist);
+ if (list->vlan == vlan)
+ macvlan_hash_del_sources(list);
+ }
+ }
+ vlan->macaddr_count = 0;
+}
+
+static void macvlan_forward_sources_one(struct sk_buff *skb,
+ struct macvlan_dev *vlan)
+{
+ struct sk_buff *nskb;
+ struct net_device *dev;
+ int len;
+ int ret;
+
+ dev = vlan->dev;
+ if (unlikely(!(dev->flags & IFF_UP)))
+ return;
+
+ nskb = skb_clone(skb, GFP_ATOMIC);
+ if (!nskb)
+ return;
+
+ len = nskb->len + ETH_HLEN;
+ nskb->dev = dev;
+ nskb->pkt_type = PACKET_HOST;
+ ret = vlan->receive(nskb);
+ macvlan_count_rx(vlan, len, ret == NET_RX_SUCCESS, 0);
+}
+
+static void macvlan_forward_sources(struct sk_buff *skb,
+ struct macvlan_port *port,
+ const unsigned char *addr)
+{
+ struct macvlan_source_list *list;
+ struct hlist_node *n;
+ struct hlist_head *h = &port->vlan_source_hash[addr[5]];
+
+ hlist_for_each_entry_rcu(list, n, h, hlist) {
+ if (!compare_ether_addr_64bits(list->addr, addr))
+ if (list->vlan->dev->flags & IFF_UP)
+ macvlan_forward_sources_one(skb, list->vlan);
+ }
+ return;
+}
+
/* called under rcu_read_lock() from netif_receive_skb */
static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb)
{
@@ -172,6 +284,8 @@ static rx_handler_result_t macvlan_handl
skb = ip_check_defrag(skb, IP_DEFRAG_MACVLAN);
if (!skb)
return RX_HANDLER_CONSUMED;
+
+ macvlan_forward_sources(skb, port, eth->h_source);
src = macvlan_hash_lookup(port, eth->h_source);
if (!src)
/* frame comes from an external address */
@@ -202,6 +316,7 @@ static rx_handler_result_t macvlan_handl
return RX_HANDLER_PASS;
}

+ macvlan_forward_sources(skb, port, eth->h_source);
if (port->passthru)
vlan = list_first_entry(&port->vlans, struct macvlan_dev, list);
else
@@ -474,6 +589,7 @@ static void macvlan_uninit(struct net_de

free_percpu(vlan->pcpu_stats);

+ macvlan_flush_sources(port, vlan);
port->count -= 1;
if (!port->count)
macvlan_port_destroy(port->dev);
@@ -615,7 +731,8 @@ static int macvlan_port_create(struct ne
INIT_LIST_HEAD(&port->vlans);
for (i = 0; i < MACVLAN_HASH_SIZE; i++)
INIT_HLIST_HEAD(&port->vlan_hash[i]);
-
+ for (i = 0; i < MACVLAN_HASH_SIZE; i++)
+ INIT_HLIST_HEAD(&port->vlan_source_hash[i]);
err = netdev_rx_handler_register(dev, macvlan_handle_frame, port);
if (err)
kfree(port);
@@ -648,11 +765,35 @@ static int macvlan_validate(struct nlatt
case MACVLAN_MODE_VEPA:
case MACVLAN_MODE_BRIDGE:
case MACVLAN_MODE_PASSTHRU:
+ case MACVLAN_MODE_SOURCE:
break;
default:
return -EINVAL;
}
}
+
+ if (data && data[IFLA_MACVLAN_MACADDR_MODE]) {
+ switch (nla_get_u32(data[IFLA_MACVLAN_MACADDR_MODE])) {
+ case MACVLAN_MACADDR_ADD:
+ case MACVLAN_MACADDR_DEL:
+ case MACVLAN_MACADDR_FLUSH:
+ break;
+ default:
+ return -EINVAL;
+ }
+ }
+
+ if (data && data[IFLA_MACVLAN_MACADDR]) {
+ if (nla_len(data[IFLA_MACVLAN_MACADDR]) != ETH_ALEN)
+ return -EINVAL;
+
+ if (!is_valid_ether_addr(nla_data(data[IFLA_MACVLAN_MACADDR])))
+ return -EADDRNOTAVAIL;
+ }
+
+ if (data && data[IFLA_MACVLAN_MACADDR_COUNT])
+ return -EINVAL;
+
return 0;
}

@@ -706,6 +847,7 @@ int macvlan_common_newlink(struct net *s
vlan->port = port;
vlan->receive = receive;
vlan->forward = forward;
+ vlan->devdump_count = 0;

vlan->mode = MACVLAN_MODE_VEPA;
if (data && data[IFLA_MACVLAN_MODE])
@@ -749,23 +891,79 @@ void macvlan_dellink(struct net_device *
{
struct macvlan_dev *vlan = netdev_priv(dev);

+ if (vlan->mode == MACVLAN_MODE_SOURCE)
+ macvlan_flush_sources(vlan->port, vlan);
list_del(&vlan->list);
unregister_netdevice_queue(dev, head);
}
EXPORT_SYMBOL_GPL(macvlan_dellink);

+static int macvlan_changelink_sources(struct macvlan_dev *vlan, u32 mode,
+ unsigned char *addr)
+{
+ if (mode == MACVLAN_MACADDR_ADD) {
+ if (!addr)
+ return -EINVAL;
+
+ return macvlan_hash_add_sources(vlan, addr);
+
+ } else if (mode == MACVLAN_MACADDR_DEL) {
+ struct macvlan_source_list *list;
+
+ if (!addr)
+ return -EINVAL;
+
+ list = macvlan_hash_lookup_sources_list(vlan, addr);
+ if (list) {
+ macvlan_hash_del_sources(list);
+ vlan->macaddr_count--;
+ }
+ } else if (mode == MACVLAN_MACADDR_FLUSH) {
+ macvlan_flush_sources(vlan->port, vlan);
+ } else {
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
static int macvlan_changelink(struct net_device *dev,
struct nlattr *tb[], struct nlattr *data[])
{
struct macvlan_dev *vlan = netdev_priv(dev);
- if (data && data[IFLA_MACVLAN_MODE])
+ u32 mode;
+ unsigned char *addr;
+
+ if (data && data[IFLA_MACVLAN_MODE]) {
+ if (vlan->mode == MACVLAN_MODE_SOURCE &&
+ vlan->mode != nla_get_u32(data[IFLA_MACVLAN_MODE]))
+ macvlan_flush_sources(vlan->port, vlan);
vlan->mode = nla_get_u32(data[IFLA_MACVLAN_MODE]);
+ }
+
+ if (data && data[IFLA_MACVLAN_MACADDR_MODE] &&
+ data[IFLA_MACVLAN_MACADDR]) {
+ mode = nla_get_u32(data[IFLA_MACVLAN_MACADDR_MODE]);
+ addr = nla_data(data[IFLA_MACVLAN_MACADDR]);
+ if (vlan->mode == MACVLAN_MODE_SOURCE)
+ return macvlan_changelink_sources(vlan, mode, addr);
+ else
+ return -EINVAL;
+
+ } else if (data && data[IFLA_MACVLAN_MACADDR_MODE]) {
+ mode = nla_get_u32(data[IFLA_MACVLAN_MACADDR_MODE]);
+ if (vlan->mode == MACVLAN_MODE_SOURCE)
+ return macvlan_changelink_sources(vlan, mode, NULL);
+ else
+ return -EINVAL;
+ }
+
return 0;
}

static size_t macvlan_get_size(const struct net_device *dev)
{
- return nla_total_size(4);
+ return nla_total_size(4) + nla_total_size(4);
}

static int macvlan_fill_info(struct sk_buff *skb,
@@ -774,14 +972,143 @@ static int macvlan_fill_info(struct sk_b
struct macvlan_dev *vlan = netdev_priv(dev);

NLA_PUT_U32(skb, IFLA_MACVLAN_MODE, vlan->mode);
+ NLA_PUT_U32(skb, IFLA_MACVLAN_MACADDR_COUNT, vlan->macaddr_count);
+
return 0;

nla_put_failure:
return -EMSGSIZE;
}

+static size_t macvlan_sources_size(void)
+{
+ return nla_total_size(sizeof(u8) * ETH_ALEN);
+}
+
+
+static int macvlan_sources_fill(struct sk_buff *skb, int type, u32 pid,
+ u32 seq, unsigned int flags,
+ struct macvlan_source_list *list)
+{
+ struct nlmsghdr *nlh;
+
+ ASSERT_RTNL();
+ nlh = nlmsg_put(skb, pid, seq, type, 0, flags);
+ if (nlh == NULL)
+ return -EMSGSIZE;
+
+ NLA_PUT(skb, IFLA_MACVLAN_MACADDR, ETH_ALEN, list->addr);
+
+ return nlmsg_end(skb, nlh);
+
+nla_put_failure:
+ nlmsg_cancel(skb, nlh);
+ return -EMSGSIZE;
+}
+
+static int macvlan_sources_send(struct macvlan_source_list *list,
+ struct net *net, int type, u32 pid, u32 seq,
+ unsigned int flags)
+{
+ struct sk_buff *nskb;
+ int err;
+
+ nskb = nlmsg_new(macvlan_sources_size(), GFP_KERNEL);
+ if (nskb == NULL)
+ return -ENOBUFS;
+
+ err = macvlan_sources_fill(nskb, type, pid, seq, flags, list);
+ if (err < 0) {
+ WARN_ON(err == -EMSGSIZE);
+ kfree_skb(nskb);
+ } else {
+ err = rtnl_unicast(nskb, net, pid);
+ }
+
+ return err;
+}
+
+static int macvlan_sources_dump_done(int err, struct net *net, u32 pid,
+ u32 seq)
+{
+ struct nlmsghdr *nlh;
+ struct sk_buff *nskb;
+
+ nskb = nlmsg_new(sizeof(err), GFP_KERNEL);
+ if (nskb == NULL)
+ return -ENOBUFS;
+
+ ASSERT_RTNL();
+ nlh = nlmsg_put(nskb, pid, seq, NLMSG_DONE, sizeof(err), NLM_F_MULTI);
+ if (nlh == NULL)
+ return -EMSGSIZE;
+
+ memcpy(nlmsg_data(nlh), &err, sizeof(err));
+
+ err = nlmsg_end(nskb, nlh);
+ if (err < 0) {
+ WARN_ON(err == -EMSGSIZE);
+ kfree_skb(nskb);
+ } else {
+ err = rtnl_unicast(nskb, net, pid);
+ }
+
+ return err;
+}
+
+static int macvlan_sources_dump(struct sk_buff *skb, struct net_device *dev,
+ int type, u32 pid, u32 seq)
+{
+ struct net *net = sock_net(skb->sk);
+ struct macvlan_dev *vlan = netdev_priv(dev);
+ struct macvlan_source_list *list;
+ int err = 0;
+ int i;
+ int iter = 0;
+ int upper_limit = vlan->devdump_count + MACVLAN_DUMP_PAGE_COUNT;
+
+ for (i = 0; i < MACVLAN_HASH_SIZE; i++) {
+ struct hlist_node *n;
+ struct hlist_head *h;
+
+ h = &vlan->port->vlan_source_hash[i];
+ hlist_for_each_entry_rcu(list, n, h, hlist) {
+ if (list->vlan == vlan) {
+ iter++;
+ if (iter <= vlan->devdump_count)
+ continue;
+
+ if (iter > upper_limit)
+ goto out2;
+
+ err = macvlan_sources_send(list, net, type,
+ pid, seq,
+ NLM_F_MULTI);
+ if (err < 0)
+ goto out;
+
+ }
+ }
+ }
+out2:
+ vlan->devdump_count += MACVLAN_DUMP_PAGE_COUNT;
+ if (vlan->devdump_count >= vlan->macaddr_count)
+ vlan->devdump_count = 0;
+
+ err = macvlan_sources_dump_done(err, net, pid, seq);
+ if (err < 0)
+ goto out;
+ return 0;
+out:
+ return err;
+}
+
static const struct nla_policy macvlan_policy[IFLA_MACVLAN_MAX + 1] = {
[IFLA_MACVLAN_MODE] = { .type = NLA_U32 },
+ [IFLA_MACVLAN_MACADDR_MODE] = { .type = NLA_U32 },
+ [IFLA_MACVLAN_MACADDR] = { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
+ [IFLA_MACVLAN_MACADDR_DATA] = { .type = NLA_NESTED },
+ [IFLA_MACVLAN_MACADDR_COUNT] = { .type = NLA_U32 },
};

int macvlan_link_register(struct rtnl_link_ops *ops)
@@ -794,6 +1121,7 @@ int macvlan_link_register(struct rtnl_li
ops->changelink = macvlan_changelink;
ops->get_size = macvlan_get_size;
ops->fill_info = macvlan_fill_info;
+ ops->dev_dump = macvlan_sources_dump;

return rtnl_link_register(ops);
};
diff -uprN -X linux-3.2.1-orig/Documentation/dontdiff linux-3.2.1-orig/include/linux/if_link.h linux-3.2.1-macvlan/include/linux/if_link.h
--- linux-3.2.1-orig/include/linux/if_link.h 2012-01-27 13:38:57.000000000 +0000
+++ linux-3.2.1-macvlan/include/linux/if_link.h 2012-02-03 11:27:28.000000000 +0000
@@ -252,6 +252,10 @@ struct ifla_vlan_qos_mapping {
enum {
IFLA_MACVLAN_UNSPEC,
IFLA_MACVLAN_MODE,
+ IFLA_MACVLAN_MACADDR_MODE,
+ IFLA_MACVLAN_MACADDR,
+ IFLA_MACVLAN_MACADDR_DATA,
+ IFLA_MACVLAN_MACADDR_COUNT,
__IFLA_MACVLAN_MAX,
};

@@ -262,8 +266,17 @@ enum macvlan_mode {
MACVLAN_MODE_VEPA = 2, /* talk to other ports through ext bridge */
MACVLAN_MODE_BRIDGE = 4, /* talk to bridge ports directly */
MACVLAN_MODE_PASSTHRU = 8,/* take over the underlying device */
+ MACVLAN_MODE_SOURCE = 16,/* use source MAC address list to assign */
};

+enum macvlan_macaddr_mode {
+ MACVLAN_MACADDR_ADD,
+ MACVLAN_MACADDR_DEL,
+ MACVLAN_MACADDR_FLUSH,
+};
+
+#define MACVLAN_DUMP_PAGE_COUNT 10
+
/* SR-IOV virtual function management section */

enum {
diff -uprN -X linux-3.2.1-orig/Documentation/dontdiff linux-3.2.1-orig/include/linux/if_macvlan.h linux-3.2.1-macvlan/include/linux/if_macvlan.h
--- linux-3.2.1-orig/include/linux/if_macvlan.h 2012-01-27 13:38:56.000000000 +0000
+++ linux-3.2.1-macvlan/include/linux/if_macvlan.h 2012-02-03 11:08:47.000000000 +0000
@@ -65,6 +65,8 @@ struct macvlan_dev {
struct macvtap_queue *taps[MAX_MACVTAP_QUEUES];
int numvtaps;
int minor;
+ unsigned int devdump_count;
+ unsigned int macaddr_count;
};

static inline void macvlan_count_rx(const struct macvlan_dev *vlan,

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/