diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index 9f69e818b0009db7881b3f8c862393836e5a604b..e538478843198ba4ac5423d66bf8f6748c9b290b 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -1045,7 +1045,7 @@ static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[])
/* loop through vlans and send one packet for each */
rcu_read_lock();
netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
- if (upper->priv_flags & IFF_802_1Q_VLAN)
+ if (is_vlan_dev(upper) && vlan_get_encap_level(upper) == 0)
alb_send_lp_vid(slave, mac_addr,
vlan_dev_vlan_id(upper));
}
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 69aff72c895716fe6c579d2bf7f46c79ddca2a36..d3a67896d43541f0b8ebc0b4e193db945087bc56 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2126,10 +2126,10 @@ static bool bond_has_this_ip(struct bonding *bond, __be32 ip)
*/
static void bond_arp_send(struct net_device *slave_dev, int arp_op,
__be32 dest_ip, __be32 src_ip,
- struct bond_vlan_tag *inner,
- struct bond_vlan_tag *outer)
+ struct bond_vlan_tag *tags)
{
struct sk_buff *skb;
+ int i;
pr_debug("arp %d on slave %s: dst %pI4 src %pI4\n",
arp_op, slave_dev->name, &dest_ip, &src_ip);
@@ -2141,21 +2141,26 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op,
net_err_ratelimited("ARP packet allocation failed\n");
return;
}
- if (outer->vlan_id) {
- if (inner->vlan_id) {
- pr_debug("inner tag: proto %X vid %X\n",
- ntohs(inner->vlan_proto), inner->vlan_id);
- skb = __vlan_put_tag(skb, inner->vlan_proto,
- inner->vlan_id);
- if (!skb) {
- net_err_ratelimited("failed to insert inner VLAN tag\n");
- return;
- }
- }
- pr_debug("outer reg: proto %X vid %X\n",
- ntohs(outer->vlan_proto), outer->vlan_id);
- skb = vlan_put_tag(skb, outer->vlan_proto, outer->vlan_id);
+ /* Go through all the tags backwards and add them to the packet */
+ for (i = BOND_MAX_VLAN_ENCAP - 1; i > 0; i--) {
+ if (!tags[i].vlan_id)
+ continue;
+
+ pr_debug("inner tag: proto %X vid %X\n",
+ ntohs(tags[i].vlan_proto), tags[i].vlan_id);
+ skb = __vlan_put_tag(skb, tags[i].vlan_proto,
+ tags[i].vlan_id);
+ if (!skb) {
+ net_err_ratelimited("failed to insert inner VLAN tag\n");
+ return;
+ }
+ }
+ /* Set the outer tag */
+ if (tags[0].vlan_id) {
+ pr_debug("outer tag: proto %X vid %X\n",
+ ntohs(tags[0].vlan_proto), tags[0].vlan_id);
+ skb = vlan_put_tag(skb, tags[0].vlan_proto, tags[0].vlan_id);
if (!skb) {
net_err_ratelimited("failed to insert outer VLAN tag\n");
return;
@@ -2164,22 +2169,52 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op,
arp_xmit(skb);
}
+/* Validate the device path between the @start_dev and the @end_dev.
+ * The path is valid if the @end_dev is reachable through device
+ * stacking.
+ * When the path is validated, collect any vlan information in the
+ * path.
+ */
+static bool bond_verify_device_path(struct net_device *start_dev,
+ struct net_device *end_dev,
+ struct bond_vlan_tag *tags)
+{
+ struct net_device *upper;
+ struct list_head *iter;
+ int idx;
+
+ if (start_dev == end_dev)
+ return true;
+
+ netdev_for_each_upper_dev_rcu(start_dev, upper, iter) {
+ if (bond_verify_device_path(upper, end_dev, tags)) {
+ if (is_vlan_dev(upper)) {
+ idx = vlan_get_encap_level(upper);
+ if (idx >= BOND_MAX_VLAN_ENCAP)
+ return false;
+
+ tags[idx].vlan_proto =
+ vlan_dev_vlan_proto(upper);
+ tags[idx].vlan_id = vlan_dev_vlan_id(upper);
+ }
+ return true;
+ }
+ }
+
+ return false;
+}
static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
{
- struct net_device *upper, *vlan_upper;
- struct list_head *iter, *vlan_iter;
struct rtable *rt;
- struct bond_vlan_tag inner, outer;
+ struct bond_vlan_tag tags[BOND_MAX_VLAN_ENCAP];
__be32 *targets = bond->params.arp_targets, addr;
int i;
+ bool ret;
for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) {
pr_debug("basa: target %pI4\n", &targets[i]);
- inner.vlan_proto = 0;
- inner.vlan_id = 0;
- outer.vlan_proto = 0;
- outer.vlan_id = 0;
+ memset(tags, 0, sizeof(tags));
/* Find out through which dev should the packet go */
rt = ip_route_output(dev_net(bond->dev), targets[i], 0,
@@ -2192,7 +2227,8 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
net_warn_ratelimited("%s: no route to arp_ip_target %pI4 and arp_validate is set\n",
bond->dev->name,
&targets[i]);
- bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &inner, &outer);
+ bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
+ 0, tags);
continue;
}
@@ -2201,52 +2237,12 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
goto found;
rcu_read_lock();
- /* first we search only for vlan devices. for every vlan
- * found we verify its upper dev list, searching for the
- * rt->dst.dev. If found we save the tag of the vlan and
- * proceed to send the packet.
- */
- netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper,
- vlan_iter) {
- if (!is_vlan_dev(vlan_upper))
- continue;
-
- if (vlan_upper == rt->dst.dev) {
- outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
- outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
- rcu_read_unlock();
- goto found;
- }
- netdev_for_each_all_upper_dev_rcu(vlan_upper, upper,
- iter) {
- if (upper == rt->dst.dev) {
- /* If the upper dev is a vlan dev too,
- * set the vlan tag to inner tag.
- */
- if (is_vlan_dev(upper)) {
- inner.vlan_proto = vlan_dev_vlan_proto(upper);
- inner.vlan_id = vlan_dev_vlan_id(upper);
- }
- outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
- outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
- rcu_read_unlock();
- goto found;
- }
- }
- }
-
- /* if the device we're looking for is not on top of any of
- * our upper vlans, then just search for any dev that
- * matches, and in case it's a vlan - save the id
- */
- netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
- if (upper == rt->dst.dev) {
- rcu_read_unlock();
- goto found;
- }
- }
+ ret = bond_verify_device_path(bond->dev, rt->dst.dev, tags);
rcu_read_unlock();
+ if (ret)
+ goto found;
+
/* Not our device - skip */
pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n",
bond->dev->name, &targets[i],
@@ -2259,7 +2255,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
addr = bond_confirm_addr(rt->dst.dev, targets[i], 0);
ip_rt_put(rt);
bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
- addr, &inner, &outer);
+ addr, tags);
}
}
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index b8bdd0acc8f334ac97bca2ddfea602f473c3272f..00bea320e3b50c1eaa75a712f7afbbe65ac146fb 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -36,6 +36,7 @@
#define bond_version DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n"
+#define BOND_MAX_VLAN_ENCAP 2
#define BOND_MAX_ARP_TARGETS 16
#define BOND_DEFAULT_MIIMON 100
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 724bde8477b2aa2a96c000db09d50739603aeeee..c901b13b6f03f9e41db1b556ff081351ad44207f 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -484,4 +484,10 @@ static inline void vlan_set_encap_proto(struct sk_buff *skb,
*/
skb->protocol = htons(ETH_P_802_2);
}
+
+static inline int vlan_get_encap_level(struct net_device *dev)
+{
+ BUG_ON(!is_vlan_dev(dev));
+ return vlan_dev_priv(dev)->nest_level;
+}
#endif /* !(_LINUX_IF_VLAN_H_) */
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9d4b1f1b6b7551c3058471cd27738fd0f8ead82c..b42d07b0390b2e95c3e551bb6639b906472bf70b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3056,9 +3056,18 @@ extern int weight_p;
extern int bpf_jit_enable;
bool netdev_has_upper_dev(struct net_device *dev, struct net_device *upper_dev);
+struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
+ struct list_head **iter);
struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
struct list_head **iter);
+/* iterate through upper list, must be called under RCU read lock */
+#define netdev_for_each_upper_dev_rcu(dev, updev, iter) \
+ for (iter = &(dev)->adj_list.upper, \
+ updev = netdev_upper_get_next_dev_rcu(dev, &(iter)); \
+ updev; \
+ updev = netdev_upper_get_next_dev_rcu(dev, &(iter)))
+
/* iterate through upper list, must be called under RCU read lock */
#define netdev_for_each_all_upper_dev_rcu(dev, updev, iter) \
for (iter = &(dev)->all_adj_list.upper, \
diff --git a/net/core/dev.c b/net/core/dev.c
index 2b872bfbd1725b7600f86116193191aa7c77a7d9..9abc503b19b7dad367b83a8179468ee3e1ae72d5 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4541,6 +4541,32 @@ void *netdev_adjacent_get_private(struct list_head *adj_list)
}
EXPORT_SYMBOL(netdev_adjacent_get_private);
+/**
+ * netdev_upper_get_next_dev_rcu - Get the next dev from upper list
+ * @dev: device
+ * @iter: list_head ** of the current position
+ *
+ * Gets the next device from the dev's upper list, starting from iter
+ * position. The caller must hold RCU read lock.
+ */
+struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
+ struct list_head **iter)
+{
+ struct netdev_adjacent *upper;
+
+ WARN_ON_ONCE(!rcu_read_lock_held() && !lockdep_rtnl_is_held());
+
+ upper = list_entry_rcu((*iter)->next, struct netdev_adjacent, list);
+
+ if (&upper->list == &dev->adj_list.upper)
+ return NULL;
+
+ *iter = &upper->list;
+
+ return upper->dev;
+}
+EXPORT_SYMBOL(netdev_upper_get_next_dev_rcu);
+
/**
* netdev_all_upper_get_next_dev_rcu - Get the next dev from upper list
* @dev: device