Merge branch 'mlxsw-Add-VxLAN-support-with-VLAN-aware-bridges'

Ido Schimmel says:

====================
mlxsw: Add VxLAN support with VLAN-aware bridges

Commit 53e50a6ec24d ("Merge branch 'mlxsw-Add-VxLAN-support'") added
mlxsw support for VxLAN when the VxLAN device was enslaved to
VLAN-unaware bridges. This patchset extends mlxsw to also support VxLAN
with VLAN-aware bridges.

With VLAN-aware bridges, the VxLAN device's VNI is mapped to the VLAN
that is configured as 'pvid untagged' on the corresponding bridge port.
To prevent ambiguity, mlxsw forbids configurations in which the same
VLAN is configured as 'pvid untagged' on multiple VxLAN devices.

Patches #1-#2 add the necessary APIs in mlxsw and the bridge driver.

Patches #3-#4 perform small refactoring in order to prepare mlxsw for
VLAN-aware support.

Patch #5 finally enables the enslavement of VxLAN devices to a
VLAN-aware bridge. Among other things, it extends mlxsw to handle
switchdev notifications about VLAN add / delete on a VxLAN device
enslaved to an offloaded VLAN-aware bridge.

Patches #6-#8 add selftests to test the new functionality.
====================

Signed-off-by: David S. Miller <[e-mail]>

drivers/net/ethernet/mellanox/mlxsw/spectrum.c

···
 	return num_vxlans > 1;
 }
 
+static bool mlxsw_sp_bridge_vxlan_vlan_is_valid(struct net_device *br_dev)
+{
+	DECLARE_BITMAP(vlans, VLAN_N_VID) = {0};
+	struct net_device *dev;
+	struct list_head *iter;
+
+	netdev_for_each_lower_dev(br_dev, dev, iter) {
+		u16 pvid;
+		int err;
+
+		if (!netif_is_vxlan(dev))
+			continue;
+
+		err = mlxsw_sp_vxlan_mapped_vid(dev, &pvid);
+		if (err || !pvid)
+			continue;
+
+		if (test_and_set_bit(pvid, vlans))
+			return false;
+	}
+
+	return true;
+}
+
 static bool mlxsw_sp_bridge_vxlan_is_valid(struct net_device *br_dev,
 					   struct netlink_ext_ack *extack)
 {
···
 		return false;
 	}
 
-	if (br_vlan_enabled(br_dev)) {
-		NL_SET_ERR_MSG_MOD(extack, "VLAN filtering can not be enabled on a bridge with a VxLAN device");
+	if (!br_vlan_enabled(br_dev) &&
+	    mlxsw_sp_bridge_has_multiple_vxlans(br_dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "Multiple VxLAN devices are not supported in a VLAN-unaware bridge");
 		return false;
 	}
 
-	if (mlxsw_sp_bridge_has_multiple_vxlans(br_dev)) {
-		NL_SET_ERR_MSG_MOD(extack, "Multiple VxLAN devices are not supported in a VLAN-unaware bridge");
+	if (br_vlan_enabled(br_dev) &&
+	    !mlxsw_sp_bridge_vxlan_vlan_is_valid(br_dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "Multiple VxLAN devices cannot have the same VLAN as PVID and egress untagged");
 		return false;
 	}
 
···
 		if (cu_info->linking) {
 			if (!netif_running(dev))
 				return 0;
+			/* When the bridge is VLAN-aware, the VNI of the VxLAN
+			 * device needs to be mapped to a VLAN, but at this
+			 * point no VLANs are configured on the VxLAN device
+			 */
+			if (br_vlan_enabled(upper_dev))
+				return 0;
 			return mlxsw_sp_bridge_vxlan_join(mlxsw_sp, upper_dev,
-							  dev, extack);
+							  dev, 0, extack);
 		} else {
-			mlxsw_sp_bridge_vxlan_leave(mlxsw_sp, upper_dev, dev);
+			/* VLANs were already flushed, which triggered the
+			 * necessary cleanup
+			 */
+			if (br_vlan_enabled(upper_dev))
+				return 0;
+			mlxsw_sp_bridge_vxlan_leave(mlxsw_sp, dev);
 		}
 		break;
 	case NETDEV_PRE_UP:
···
 			return 0;
 		if (!mlxsw_sp_lower_get(upper_dev))
 			return 0;
-		return mlxsw_sp_bridge_vxlan_join(mlxsw_sp, upper_dev, dev,
+		return mlxsw_sp_bridge_vxlan_join(mlxsw_sp, upper_dev, dev, 0,
 						  extack);
 	case NETDEV_DOWN:
 		upper_dev = netdev_master_upper_dev_get(dev);
···
 			return 0;
 		if (!mlxsw_sp_lower_get(upper_dev))
 			return 0;
-		mlxsw_sp_bridge_vxlan_leave(mlxsw_sp, upper_dev, dev);
+		mlxsw_sp_bridge_vxlan_leave(mlxsw_sp, dev);
 		break;
 	}
 

drivers/net/ethernet/mellanox/mlxsw/spectrum.h

···
 #include <linux/netdevice.h>
 #include <linux/rhashtable.h>
 #include <linux/bitops.h>
+#include <linux/if_bridge.h>
 #include <linux/if_vlan.h>
 #include <linux/list.h>
 #include <linux/dcbnl.h>
···
 	return !!mlxsw_sp_bridge_vxlan_dev_find(br_dev);
 }
 
+static inline int
+mlxsw_sp_vxlan_mapped_vid(const struct net_device *vxlan_dev, u16 *p_vid)
+{
+	struct bridge_vlan_info vinfo;
+	u16 vid = 0;
+	int err;
+
+	err = br_vlan_get_pvid(vxlan_dev, &vid);
+	if (err || !vid)
+		goto out;
+
+	err = br_vlan_get_info(vxlan_dev, vid, &vinfo);
+	if (err || !(vinfo.flags & BRIDGE_VLAN_INFO_UNTAGGED))
+		vid = 0;
+
+out:
+	*p_vid = vid;
+	return err;
+}
+
 static inline bool
 mlxsw_sp_port_is_pause_en(const struct mlxsw_sp_port *mlxsw_sp_port)
 {
···
 					 const struct net_device *br_dev);
 int mlxsw_sp_bridge_vxlan_join(struct mlxsw_sp *mlxsw_sp,
 			       const struct net_device *br_dev,
-			       const struct net_device *vxlan_dev,
+			       const struct net_device *vxlan_dev, u16 vid,
 			       struct netlink_ext_ack *extack);
 void mlxsw_sp_bridge_vxlan_leave(struct mlxsw_sp *mlxsw_sp,
-				 const struct net_device *br_dev,
 				 const struct net_device *vxlan_dev);
 
 /* spectrum.c */
···
 struct mlxsw_sp_fid *mlxsw_sp_fid_8021q_get(struct mlxsw_sp *mlxsw_sp, u16 vid);
 struct mlxsw_sp_fid *mlxsw_sp_fid_8021d_get(struct mlxsw_sp *mlxsw_sp,
 					    int br_ifindex);
+struct mlxsw_sp_fid *mlxsw_sp_fid_8021q_lookup(struct mlxsw_sp *mlxsw_sp,
+					       u16 vid);
 struct mlxsw_sp_fid *mlxsw_sp_fid_8021d_lookup(struct mlxsw_sp *mlxsw_sp,
 					       int br_ifindex);
 struct mlxsw_sp_fid *mlxsw_sp_fid_rfid_get(struct mlxsw_sp *mlxsw_sp,

drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c

···
 	return mlxsw_sp_fid_get(mlxsw_sp, MLXSW_SP_FID_TYPE_8021D, &br_ifindex);
 }
 
+struct mlxsw_sp_fid *mlxsw_sp_fid_8021q_lookup(struct mlxsw_sp *mlxsw_sp,
+					       u16 vid)
+{
+	return mlxsw_sp_fid_lookup(mlxsw_sp, MLXSW_SP_FID_TYPE_8021Q, &vid);
+}
+
 struct mlxsw_sp_fid *mlxsw_sp_fid_8021d_lookup(struct mlxsw_sp *mlxsw_sp,
 					       int br_ifindex)
 {

drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c

···
 			   struct mlxsw_sp_bridge_port *bridge_port,
 			   struct mlxsw_sp_port *mlxsw_sp_port);
 	int (*vxlan_join)(struct mlxsw_sp_bridge_device *bridge_device,
-			  const struct net_device *vxlan_dev,
+			  const struct net_device *vxlan_dev, u16 vid,
 			  struct netlink_ext_ack *extack);
-	void (*vxlan_leave)(struct mlxsw_sp_bridge_device *bridge_device,
-			    const struct net_device *vxlan_dev);
 	struct mlxsw_sp_fid *
 		(*fid_get)(struct mlxsw_sp_bridge_device *bridge_device,
 			   u16 vid);
···
 
 static int
 mlxsw_sp_bridge_8021q_vxlan_join(struct mlxsw_sp_bridge_device *bridge_device,
-				 const struct net_device *vxlan_dev,
+				 const struct net_device *vxlan_dev, u16 vid,
 				 struct netlink_ext_ack *extack)
 {
-	WARN_ON(1);
-	return -EINVAL;
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(bridge_device->dev);
+	struct vxlan_dev *vxlan = netdev_priv(vxlan_dev);
+	struct mlxsw_sp_nve_params params = {
+		.type = MLXSW_SP_NVE_TYPE_VXLAN,
+		.vni = vxlan->cfg.vni,
+		.dev = vxlan_dev,
+	};
+	struct mlxsw_sp_fid *fid;
+	int err;
+
+	/* If the VLAN is 0, we need to find the VLAN that is configured as
+	 * PVID and egress untagged on the bridge port of the VxLAN device.
+	 * It is possible no such VLAN exists
+	 */
+	if (!vid) {
+		err = mlxsw_sp_vxlan_mapped_vid(vxlan_dev, &vid);
+		if (err || !vid)
+			return err;
+	}
+
+	/* If no other port is member in the VLAN, then the FID does not exist.
+	 * NVE will be enabled on the FID once a port joins the VLAN
+	 */
+	fid = mlxsw_sp_fid_8021q_lookup(mlxsw_sp, vid);
+	if (!fid)
+		return 0;
+
+	if (mlxsw_sp_fid_vni_is_set(fid)) {
+		err = -EINVAL;
+		goto err_vni_exists;
+	}
+
+	err = mlxsw_sp_nve_fid_enable(mlxsw_sp, fid, &params, extack);
+	if (err)
+		goto err_nve_fid_enable;
+
+	/* The tunnel port does not hold a reference on the FID. Only
+	 * local ports and the router port
+	 */
+	mlxsw_sp_fid_put(fid);
+
+	return 0;
+
+err_nve_fid_enable:
+err_vni_exists:
+	mlxsw_sp_fid_put(fid);
+	return err;
 }
 
-static void
-mlxsw_sp_bridge_8021q_vxlan_leave(struct mlxsw_sp_bridge_device *bridge_device,
-				  const struct net_device *vxlan_dev)
+static struct net_device *
+mlxsw_sp_bridge_8021q_vxlan_dev_find(struct net_device *br_dev, u16 vid)
 {
+	struct net_device *dev;
+	struct list_head *iter;
+
+	netdev_for_each_lower_dev(br_dev, dev, iter) {
+		u16 pvid;
+		int err;
+
+		if (!netif_is_vxlan(dev))
+			continue;
+
+		err = mlxsw_sp_vxlan_mapped_vid(dev, &pvid);
+		if (err || pvid != vid)
+			continue;
+
+		return dev;
+	}
+
+	return NULL;
 }
 
 static struct mlxsw_sp_fid *
···
 			      u16 vid)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(bridge_device->dev);
+	struct net_device *vxlan_dev;
+	struct mlxsw_sp_fid *fid;
+	int err;
 
-	return mlxsw_sp_fid_8021q_get(mlxsw_sp, vid);
+	fid = mlxsw_sp_fid_8021q_get(mlxsw_sp, vid);
+	if (IS_ERR(fid))
+		return fid;
+
+	if (mlxsw_sp_fid_vni_is_set(fid))
+		return fid;
+
+	/* Find the VxLAN device that has the specified VLAN configured as
+	 * PVID and egress untagged. There can be at most one such device
+	 */
+	vxlan_dev = mlxsw_sp_bridge_8021q_vxlan_dev_find(bridge_device->dev,
+							 vid);
+	if (!vxlan_dev)
+		return fid;
+
+	if (!netif_running(vxlan_dev))
+		return fid;
+
+	err = mlxsw_sp_bridge_8021q_vxlan_join(bridge_device, vxlan_dev, vid,
+					       NULL);
+	if (err)
+		goto err_vxlan_join;
+
+	return fid;
+
+err_vxlan_join:
+	mlxsw_sp_fid_put(fid);
+	return ERR_PTR(err);
 }
 
 static struct mlxsw_sp_fid *
 mlxsw_sp_bridge_8021q_fid_lookup(struct mlxsw_sp_bridge_device *bridge_device,
 				 u16 vid)
 {
-	WARN_ON(1);
-	return NULL;
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(bridge_device->dev);
+
+	return mlxsw_sp_fid_8021q_lookup(mlxsw_sp, vid);
 }
 
 static u16
···
 	.port_join	= mlxsw_sp_bridge_8021q_port_join,
 	.port_leave	= mlxsw_sp_bridge_8021q_port_leave,
 	.vxlan_join	= mlxsw_sp_bridge_8021q_vxlan_join,
-	.vxlan_leave	= mlxsw_sp_bridge_8021q_vxlan_leave,
 	.fid_get	= mlxsw_sp_bridge_8021q_fid_get,
 	.fid_lookup	= mlxsw_sp_bridge_8021q_fid_lookup,
 	.fid_vid	= mlxsw_sp_bridge_8021q_fid_vid,
···
 
 static int
 mlxsw_sp_bridge_8021d_vxlan_join(struct mlxsw_sp_bridge_device *bridge_device,
-				 const struct net_device *vxlan_dev,
+				 const struct net_device *vxlan_dev, u16 vid,
 				 struct netlink_ext_ack *extack)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(bridge_device->dev);
···
 	return err;
 }
 
-static void
-mlxsw_sp_bridge_8021d_vxlan_leave(struct mlxsw_sp_bridge_device *bridge_device,
-				  const struct net_device *vxlan_dev)
-{
-	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(bridge_device->dev);
-	struct mlxsw_sp_fid *fid;
-
-	fid = mlxsw_sp_fid_8021d_lookup(mlxsw_sp, bridge_device->dev->ifindex);
-	if (WARN_ON(!fid))
-		return;
-
-	/* If the VxLAN device is down, then the FID does not have a VNI */
-	if (!mlxsw_sp_fid_vni_is_set(fid))
-		goto out;
-
-	mlxsw_sp_nve_fid_disable(mlxsw_sp, fid);
-out:
-	mlxsw_sp_fid_put(fid);
-}
-
 static struct mlxsw_sp_fid *
 mlxsw_sp_bridge_8021d_fid_get(struct mlxsw_sp_bridge_device *bridge_device,
 			      u16 vid)
···
 	if (!netif_running(vxlan_dev))
 		return fid;
 
-	err = mlxsw_sp_bridge_8021d_vxlan_join(bridge_device, vxlan_dev, NULL);
+	err = mlxsw_sp_bridge_8021d_vxlan_join(bridge_device, vxlan_dev, 0,
+					       NULL);
 	if (err)
 		goto err_vxlan_join;
 
···
 	.port_join	= mlxsw_sp_bridge_8021d_port_join,
 	.port_leave	= mlxsw_sp_bridge_8021d_port_leave,
 	.vxlan_join	= mlxsw_sp_bridge_8021d_vxlan_join,
-	.vxlan_leave	= mlxsw_sp_bridge_8021d_vxlan_leave,
 	.fid_get	= mlxsw_sp_bridge_8021d_fid_get,
 	.fid_lookup	= mlxsw_sp_bridge_8021d_fid_lookup,
 	.fid_vid	= mlxsw_sp_bridge_8021d_fid_vid,
···
 
 int mlxsw_sp_bridge_vxlan_join(struct mlxsw_sp *mlxsw_sp,
 			       const struct net_device *br_dev,
-			       const struct net_device *vxlan_dev,
+			       const struct net_device *vxlan_dev, u16 vid,
 			       struct netlink_ext_ack *extack)
 {
 	struct mlxsw_sp_bridge_device *bridge_device;
···
 	if (WARN_ON(!bridge_device))
 		return -EINVAL;
 
-	return bridge_device->ops->vxlan_join(bridge_device, vxlan_dev, extack);
+	return bridge_device->ops->vxlan_join(bridge_device, vxlan_dev, vid,
+					      extack);
 }
 
 void mlxsw_sp_bridge_vxlan_leave(struct mlxsw_sp *mlxsw_sp,
-				 const struct net_device *br_dev,
 				 const struct net_device *vxlan_dev)
 {
-	struct mlxsw_sp_bridge_device *bridge_device;
+	struct vxlan_dev *vxlan = netdev_priv(vxlan_dev);
+	struct mlxsw_sp_fid *fid;
 
-	bridge_device = mlxsw_sp_bridge_device_find(mlxsw_sp->bridge, br_dev);
-	if (WARN_ON(!bridge_device))
+	/* If the VxLAN device is down, then the FID does not have a VNI */
+	fid = mlxsw_sp_fid_lookup_by_vni(mlxsw_sp, vxlan->cfg.vni);
+	if (!fid)
 		return;
 
-	bridge_device->ops->vxlan_leave(bridge_device, vxlan_dev);
+	mlxsw_sp_nve_fid_disable(mlxsw_sp, fid);
+	mlxsw_sp_fid_put(fid);
 }
 
 static void
···
 	.notifier_call = mlxsw_sp_switchdev_event,
 };
 
+static int
+mlxsw_sp_switchdev_vxlan_vlan_add(struct mlxsw_sp *mlxsw_sp,
+				  struct mlxsw_sp_bridge_device *bridge_device,
+				  const struct net_device *vxlan_dev, u16 vid,
+				  bool flag_untagged, bool flag_pvid,
+				  struct switchdev_trans *trans)
+{
+	struct vxlan_dev *vxlan = netdev_priv(vxlan_dev);
+	__be32 vni = vxlan->cfg.vni;
+	struct mlxsw_sp_fid *fid;
+	u16 old_vid;
+	int err;
+
+	/* We cannot have the same VLAN as PVID and egress untagged on multiple
+	 * VxLAN devices. Note that we get this notification before the VLAN is
+	 * actually added to the bridge's database, so it is not possible for
+	 * the lookup function to return 'vxlan_dev'
+	 */
+	if (flag_untagged && flag_pvid &&
+	    mlxsw_sp_bridge_8021q_vxlan_dev_find(bridge_device->dev, vid))
+		return -EINVAL;
+
+	if (switchdev_trans_ph_prepare(trans))
+		return 0;
+
+	if (!netif_running(vxlan_dev))
+		return 0;
+
+	/* First case: FID is not associated with this VNI, but the new VLAN
+	 * is both PVID and egress untagged. Need to enable NVE on the FID, if
+	 * it exists
+	 */
+	fid = mlxsw_sp_fid_lookup_by_vni(mlxsw_sp, vni);
+	if (!fid) {
+		if (!flag_untagged || !flag_pvid)
+			return 0;
+		return mlxsw_sp_bridge_8021q_vxlan_join(bridge_device,
+							vxlan_dev, vid, NULL);
+	}
+
+	/* Second case: FID is associated with the VNI and the VLAN associated
+	 * with the FID is the same as the notified VLAN. This means the flags
+	 * (PVID / egress untagged) were toggled and that NVE should be
+	 * disabled on the FID
+	 */
+	old_vid = mlxsw_sp_fid_8021q_vid(fid);
+	if (vid == old_vid) {
+		if (WARN_ON(flag_untagged && flag_pvid)) {
+			mlxsw_sp_fid_put(fid);
+			return -EINVAL;
+		}
+		mlxsw_sp_bridge_vxlan_leave(mlxsw_sp, vxlan_dev);
+		mlxsw_sp_fid_put(fid);
+		return 0;
+	}
+
+	/* Third case: A new VLAN was configured on the VxLAN device, but this
+	 * VLAN is not PVID, so there is nothing to do.
+	 */
+	if (!flag_pvid) {
+		mlxsw_sp_fid_put(fid);
+		return 0;
+	}
+
+	/* Fourth case: Thew new VLAN is PVID, which means the VLAN currently
+	 * mapped to the VNI should be unmapped
+	 */
+	mlxsw_sp_bridge_vxlan_leave(mlxsw_sp, vxlan_dev);
+	mlxsw_sp_fid_put(fid);
+
+	/* Fifth case: The new VLAN is also egress untagged, which means the
+	 * VLAN needs to be mapped to the VNI
+	 */
+	if (!flag_untagged)
+		return 0;
+
+	err = mlxsw_sp_bridge_8021q_vxlan_join(bridge_device, vxlan_dev, vid,
+					       NULL);
+	if (err)
+		goto err_vxlan_join;
+
+	return 0;
+
+err_vxlan_join:
+	mlxsw_sp_bridge_8021q_vxlan_join(bridge_device, vxlan_dev, old_vid,
+					 NULL);
+	return err;
+}
+
+static void
+mlxsw_sp_switchdev_vxlan_vlan_del(struct mlxsw_sp *mlxsw_sp,
+				  struct mlxsw_sp_bridge_device *bridge_device,
+				  const struct net_device *vxlan_dev, u16 vid)
+{
+	struct vxlan_dev *vxlan = netdev_priv(vxlan_dev);
+	__be32 vni = vxlan->cfg.vni;
+	struct mlxsw_sp_fid *fid;
+
+	if (!netif_running(vxlan_dev))
+		return;
+
+	fid = mlxsw_sp_fid_lookup_by_vni(mlxsw_sp, vni);
+	if (!fid)
+		return;
+
+	/* A different VLAN than the one mapped to the VNI is deleted */
+	if (mlxsw_sp_fid_8021q_vid(fid) != vid)
+		goto out;
+
+	mlxsw_sp_bridge_vxlan_leave(mlxsw_sp, vxlan_dev);
+
+out:
+	mlxsw_sp_fid_put(fid);
+}
+
+static int
+mlxsw_sp_switchdev_vxlan_vlans_add(struct net_device *vxlan_dev,
+				   struct switchdev_notifier_port_obj_info *
+				   port_obj_info)
+{
+	struct switchdev_obj_port_vlan *vlan =
+		SWITCHDEV_OBJ_PORT_VLAN(port_obj_info->obj);
+	bool flag_untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED;
+	bool flag_pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID;
+	struct switchdev_trans *trans = port_obj_info->trans;
+	struct mlxsw_sp_bridge_device *bridge_device;
+	struct mlxsw_sp *mlxsw_sp;
+	struct net_device *br_dev;
+	u16 vid;
+
+	br_dev = netdev_master_upper_dev_get(vxlan_dev);
+	if (!br_dev)
+		return 0;
+
+	mlxsw_sp = mlxsw_sp_lower_get(br_dev);
+	if (!mlxsw_sp)
+		return 0;
+
+	port_obj_info->handled = true;
+
+	bridge_device = mlxsw_sp_bridge_device_find(mlxsw_sp->bridge, br_dev);
+	if (!bridge_device)
+		return -EINVAL;
+
+	if (!bridge_device->vlan_enabled)
+		return 0;
+
+	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
+		int err;
+
+		err = mlxsw_sp_switchdev_vxlan_vlan_add(mlxsw_sp, bridge_device,
+							vxlan_dev, vid,
+							flag_untagged,
+							flag_pvid, trans);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static void
+mlxsw_sp_switchdev_vxlan_vlans_del(struct net_device *vxlan_dev,
+				   struct switchdev_notifier_port_obj_info *
+				   port_obj_info)
+{
+	struct switchdev_obj_port_vlan *vlan =
+		SWITCHDEV_OBJ_PORT_VLAN(port_obj_info->obj);
+	struct mlxsw_sp_bridge_device *bridge_device;
+	struct mlxsw_sp *mlxsw_sp;
+	struct net_device *br_dev;
+	u16 vid;
+
+	br_dev = netdev_master_upper_dev_get(vxlan_dev);
+	if (!br_dev)
+		return;
+
+	mlxsw_sp = mlxsw_sp_lower_get(br_dev);
+	if (!mlxsw_sp)
+		return;
+
+	port_obj_info->handled = true;
+
+	bridge_device = mlxsw_sp_bridge_device_find(mlxsw_sp->bridge, br_dev);
+	if (!bridge_device)
+		return;
+
+	if (!bridge_device->vlan_enabled)
+		return;
+
+	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++)
+		mlxsw_sp_switchdev_vxlan_vlan_del(mlxsw_sp, bridge_device,
+						  vxlan_dev, vid);
+}
+
+static int
+mlxsw_sp_switchdev_handle_vxlan_obj_add(struct net_device *vxlan_dev,
+					struct switchdev_notifier_port_obj_info *
+					port_obj_info)
+{
+	int err = 0;
+
+	switch (port_obj_info->obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = mlxsw_sp_switchdev_vxlan_vlans_add(vxlan_dev,
+							 port_obj_info);
+		break;
+	default:
+		break;
+	}
+
+	return err;
+}
+
+static void
+mlxsw_sp_switchdev_handle_vxlan_obj_del(struct net_device *vxlan_dev,
+					struct switchdev_notifier_port_obj_info *
+					port_obj_info)
+{
+	switch (port_obj_info->obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		mlxsw_sp_switchdev_vxlan_vlans_del(vxlan_dev, port_obj_info);
+		break;
+	default:
+		break;
+	}
+}
+
 static int mlxsw_sp_switchdev_blocking_event(struct notifier_block *unused,
 					     unsigned long event, void *ptr)
 {
 	struct net_device *dev = switchdev_notifier_info_to_dev(ptr);
-	int err;
+	int err = 0;
 
 	switch (event) {
 	case SWITCHDEV_PORT_OBJ_ADD:
-		err = switchdev_handle_port_obj_add(dev, ptr,
-						    mlxsw_sp_port_dev_check,
-						    mlxsw_sp_port_obj_add);
+		if (netif_is_vxlan(dev))
+			err = mlxsw_sp_switchdev_handle_vxlan_obj_add(dev, ptr);
+		else
+			err = switchdev_handle_port_obj_add(dev, ptr,
+							mlxsw_sp_port_dev_check,
+							mlxsw_sp_port_obj_add);
 		return notifier_from_errno(err);
 	case SWITCHDEV_PORT_OBJ_DEL:
-		err = switchdev_handle_port_obj_del(dev, ptr,
-						    mlxsw_sp_port_dev_check,
-						    mlxsw_sp_port_obj_del);
+		if (netif_is_vxlan(dev))
+			mlxsw_sp_switchdev_handle_vxlan_obj_del(dev, ptr);
+		else
+			err = switchdev_handle_port_obj_del(dev, ptr,
+							mlxsw_sp_port_dev_check,
+							mlxsw_sp_port_obj_del);
 		return notifier_from_errno(err);
 	}
 

net/bridge/br_vlan.c

···
 int br_vlan_get_pvid(const struct net_device *dev, u16 *p_pvid)
 {
 	struct net_bridge_vlan_group *vg;
+	struct net_bridge_port *p;
 
 	ASSERT_RTNL();
-	if (netif_is_bridge_master(dev))
+	p = br_port_get_check_rtnl(dev);
+	if (p)
+		vg = nbp_vlan_group(p);
+	else if (netif_is_bridge_master(dev))
 		vg = br_vlan_group(netdev_priv(dev));
 	else
 		return -EINVAL;

tools/testing/selftests/drivers/net/mlxsw/vxlan.sh

···
 
 lib_dir=$(dirname $0)/../../../net/forwarding
 
-ALL_TESTS="sanitization_test offload_indication_test"
+ALL_TESTS="sanitization_test offload_indication_test \
+	sanitization_vlan_aware_test offload_indication_vlan_aware_test"
 NUM_NETIFS=2
 source $lib_dir/lib.sh
 
···
 	ip link add name vxlan0 up type vxlan id 10 nolearning noudpcsum \
 		ttl 20 tos inherit local 198.51.100.1 dstport 4789
 
-	sanitization_single_dev_test_fail
+	sanitization_single_dev_test_pass
 
 	ip link del dev vxlan0
 	ip link del dev br0
···
 	offload_indication_fdb_test
 	offload_indication_decap_route_test
 	offload_indication_setup_destroy
+}
+
+sanitization_vlan_aware_test()
+{
+	RET=0
+
+	ip link add dev br0 type bridge mcast_snooping 0 vlan_filtering 1
+
+	ip link add name vxlan10 up master br0 type vxlan id 10 nolearning \
+		noudpcsum ttl 20 tos inherit local 198.51.100.1 dstport 4789
+
+	ip link add name vxlan20 up master br0 type vxlan id 20 nolearning \
+		noudpcsum ttl 20 tos inherit local 198.51.100.1 dstport 4789
+
+	# Test that when each VNI is mapped to a different VLAN we can enslave
+	# a port to the bridge
+	bridge vlan add vid 10 dev vxlan10 pvid untagged
+	bridge vlan add vid 20 dev vxlan20 pvid untagged
+
+	ip link set dev $swp1 master br0
+	check_err $?
+
+	log_test "vlan-aware - enslavement to vlan-aware bridge"
+
+	# Try to map both VNIs to the same VLAN and make sure configuration
+	# fails
+	RET=0
+
+	bridge vlan add vid 10 dev vxlan20 pvid untagged &> /dev/null
+	check_fail $?
+
+	log_test "vlan-aware - two vnis mapped to the same vlan"
+
+	# Test that enslavement of a port to a bridge fails when two VNIs
+	# are mapped to the same VLAN
+	RET=0
+
+	ip link set dev $swp1 nomaster
+
+	bridge vlan del vid 20 dev vxlan20 pvid untagged
+	bridge vlan add vid 10 dev vxlan20 pvid untagged
+
+	ip link set dev $swp1 master br0 &> /dev/null
+	check_fail $?
+
+	log_test "vlan-aware - failed enslavement to vlan-aware bridge"
+
+	ip link del dev vxlan20
+	ip link del dev vxlan10
+	ip link del dev br0
+}
+
+offload_indication_vlan_aware_setup_create()
+{
+	# Create a simple setup with two VxLAN devices and a single VLAN-aware
+	# bridge
+	ip link add name br0 up type bridge mcast_snooping 0 vlan_filtering 1 \
+		vlan_default_pvid 0
+
+	ip link set dev $swp1 master br0
+
+	bridge vlan add vid 10 dev $swp1
+	bridge vlan add vid 20 dev $swp1
+
+	ip address add 198.51.100.1/32 dev lo
+
+	ip link add name vxlan10 up master br0 type vxlan id 10 nolearning \
+		noudpcsum ttl 20 tos inherit local 198.51.100.1 dstport 4789
+	ip link add name vxlan20 up master br0 type vxlan id 20 nolearning \
+		noudpcsum ttl 20 tos inherit local 198.51.100.1 dstport 4789
+
+	bridge vlan add vid 10 dev vxlan10 pvid untagged
+	bridge vlan add vid 20 dev vxlan20 pvid untagged
+}
+
+offload_indication_vlan_aware_setup_destroy()
+{
+	bridge vlan del vid 20 dev vxlan20
+	bridge vlan del vid 10 dev vxlan10
+
+	ip link del dev vxlan20
+	ip link del dev vxlan10
+
+	ip address del 198.51.100.1/32 dev lo
+
+	bridge vlan del vid 20 dev $swp1
+	bridge vlan del vid 10 dev $swp1
+
+	ip link set dev $swp1 nomaster
+
+	ip link del dev br0
+}
+
+offload_indication_vlan_aware_fdb_test()
+{
+	RET=0
+
+	log_info "vxlan entry offload indication - vlan-aware"
+
+	bridge fdb add de:ad:be:ef:13:37 dev vxlan10 self master static \
+		dst 198.51.100.2 vlan 10
+
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep self \
+		| grep -q offload
+	check_err $?
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep -v self \
+		| grep -q offload
+	check_err $?
+
+	log_test "vxlan entry offload indication - initial state"
+
+	# Remove FDB entry from the bridge driver and check that corresponding
+	# entry in the VxLAN driver is not marked as offloaded
+	RET=0
+
+	bridge fdb del de:ad:be:ef:13:37 dev vxlan10 master vlan 10
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep self \
+		| grep -q offload
+	check_fail $?
+
+	log_test "vxlan entry offload indication - after removal from bridge"
+
+	# Add the FDB entry back to the bridge driver and make sure it is
+	# marked as offloaded in both drivers
+	RET=0
+
+	bridge fdb add de:ad:be:ef:13:37 dev vxlan10 master static vlan 10
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep self \
+		| grep -q offload
+	check_err $?
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep -v self \
+		| grep -q offload
+	check_err $?
+
+	log_test "vxlan entry offload indication - after re-add to bridge"
+
+	# Remove FDB entry from the VxLAN driver and check that corresponding
+	# entry in the bridge driver is not marked as offloaded
+	RET=0
+
+	bridge fdb del de:ad:be:ef:13:37 dev vxlan10 self
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep -v self \
+		| grep -q offload
+	check_fail $?
+
+	log_test "vxlan entry offload indication - after removal from vxlan"
+
+	# Add the FDB entry back to the VxLAN driver and make sure it is
+	# marked as offloaded in both drivers
+	RET=0
+
+	bridge fdb add de:ad:be:ef:13:37 dev vxlan10 self dst 198.51.100.2
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep self \
+		| grep -q offload
+	check_err $?
+	bridge fdb show brport vxlan10 | grep de:ad:be:ef:13:37 | grep -v self \
+		| grep -q offload
+	check_err $?
+
+	log_test "vxlan entry offload indication - after re-add to vxlan"
+
+	bridge fdb del de:ad:be:ef:13:37 dev vxlan10 self master vlan 10
+}
+
+offload_indication_vlan_aware_decap_route_test()
+{
+	RET=0
+
+	ip route show table local | grep 198.51.100.1 | grep -q offload
+	check_err $?
+
+	# Toggle PVID flag on one VxLAN device and make sure route is still
+	# marked as offloaded
+	bridge vlan add vid 10 dev vxlan10 untagged
+
+	ip route show table local | grep 198.51.100.1 | grep -q offload
+	check_err $?
+
+	# Toggle PVID flag on second VxLAN device and make sure route is no
+	# longer marked as offloaded
+	bridge vlan add vid 20 dev vxlan20 untagged
+
+	ip route show table local | grep 198.51.100.1 | grep -q offload
+	check_fail $?
+
+	# Toggle PVID flag back and make sure route is marked as offloaded
+	bridge vlan add vid 10 dev vxlan10 pvid untagged
+	bridge vlan add vid 20 dev vxlan20 pvid untagged
+
+	ip route show table local | grep 198.51.100.1 | grep -q offload
+	check_err $?
+
+	log_test "vxlan decap route - vni map/unmap"
+}
+
+offload_indication_vlan_aware_test()
+{
+	offload_indication_vlan_aware_setup_create
+	offload_indication_vlan_aware_fdb_test
+	offload_indication_vlan_aware_decap_route_test
+	offload_indication_vlan_aware_setup_destroy
 }
 
 trap cleanup EXIT

tools/testing/selftests/net/forwarding/vxlan_bridge_1q.sh

···
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# +-----------------------+                          +------------------------+
+# | H1 (vrf)              |                          | H2 (vrf)               |
+# |  + $h1.10             |                          |  + $h2.10              |
+# |  | 192.0.2.1/28       |                          |  | 192.0.2.2/28        |
+# |  |                    |                          |  |                     |
+# |  | + $h1.20           |                          |  | + $h2.20            |
+# |  \ | 198.51.100.1/24  |                          |  \ | 198.51.100.2/24   |
+# |   \|                  |                          |   \|                   |
+# |    + $h1              |                          |    + $h2               |
+# +----|------------------+                          +----|-------------------+
+#      |                                                  |
+# +----|--------------------------------------------------|-------------------+
+# | SW |                                                  |                   |
+# | +--|--------------------------------------------------|-----------------+ |
+# | |  + $swp1                   BR1 (802.1q)             + $swp2           | |
+# | |     vid 10                                             vid 10         | |
+# | |     vid 20                                             vid 20         | |
+# | |                                                                       | |
+# | |  + vx10 (vxlan)                        + vx20 (vxlan)                 | |
+# | |    local 192.0.2.17                      local 192.0.2.17             | |
+# | |    remote 192.0.2.34 192.0.2.50          remote 192.0.2.34 192.0.2.50 | |
+# | |    id 1000 dstport $VXPORT               id 2000 dstport $VXPORT      | |
+# | |    vid 10 pvid untagged                  vid 20 pvid untagged         | |
+# | +-----------------------------------------------------------------------+ |
+# |                                                                           |
+# |  192.0.2.32/28 via 192.0.2.18                                             |
+# |  192.0.2.48/28 via 192.0.2.18                                             |
+# |                                                                           |
+# |    + $rp1                                                                 |
+# |    | 192.0.2.17/28                                                        |
+# +----|----------------------------------------------------------------------+
+#      |
+# +----|--------------------------------------------------------+
+# |    |                                             VRP2 (vrf) |
+# |    + $rp2                                                   |
+# |      192.0.2.18/28                                          |
+# |                                                             |   (maybe) HW
+# =============================================================================
+# |                                                             |  (likely) SW
+# |    + v1 (veth)                             + v3 (veth)      |
+# |    | 192.0.2.33/28                         | 192.0.2.49/28  |
+# +----|---------------------------------------|----------------+
+#      |                                       |
+# +----|------------------------------+   +----|------------------------------+
+# |    + v2 (veth)        NS1 (netns) |   |    + v4 (veth)        NS2 (netns) |
+# |      192.0.2.34/28                |   |      192.0.2.50/28                |
+# |                                   |   |                                   |
+# |   192.0.2.16/28 via 192.0.2.33    |   |   192.0.2.16/28 via 192.0.2.49    |
+# |   192.0.2.50/32 via 192.0.2.33    |   |   192.0.2.34/32 via 192.0.2.49    |
+# |                                   |   |                                   |
+# | +-------------------------------+ |   | +-------------------------------+ |
+# | |                  BR2 (802.1q) | |   | |                  BR2 (802.1q) | |
+# | |  + vx10 (vxlan)               | |   | |  + vx10 (vxlan)               | |
+# | |    local 192.0.2.34           | |   | |    local 192.0.2.50           | |
+# | |    remote 192.0.2.17          | |   | |    remote 192.0.2.17          | |
+# | |    remote 192.0.2.50          | |   | |    remote 192.0.2.34          | |
+# | |    id 1000 dstport $VXPORT    | |   | |    id 1000 dstport $VXPORT    | |
+# | |    vid 10 pvid untagged       | |   | |    vid 10 pvid untagged       | |
+# | |                               | |   | |                               | |
+# | |  + vx20 (vxlan)               | |   | |  + vx20 (vxlan)               | |
+# | |    local 192.0.2.34           | |   | |    local 192.0.2.50           | |
+# | |    remote 192.0.2.17          | |   | |    remote 192.0.2.17          | |
+# | |    remote 192.0.2.50          | |   | |    remote 192.0.2.34          | |
+# | |    id 2000 dstport $VXPORT    | |   | |    id 2000 dstport $VXPORT    | |
+# | |    vid 20 pvid untagged       | |   | |    vid 20 pvid untagged       | |
+# | |                               | |   | |                               | |
+# | |  + w1 (veth)                  | |   | |  + w1 (veth)                  | |
+# | |  | vid 10                     | |   | |  | vid 10                     | |
+# | |  | vid 20                     | |   | |  | vid 20                     | |
+# | +--|----------------------------+ |   | +--|----------------------------+ |
+# |    |                              |   |    |                              |
+# | +--|----------------------------+ |   | +--|----------------------------+ |
+# | |  + w2 (veth)        VW2 (vrf) | |   | |  + w2 (veth)        VW2 (vrf) | |
+# | |  |\                           | |   | |  |\                           | |
+# | |  | + w2.10                    | |   | |  | + w2.10                    | |
+# | |  |   192.0.2.3/28             | |   | |  |   192.0.2.4/28             | |
+# | |  |                            | |   | |  |                            | |
+# | |  + w2.20                      | |   | |  + w2.20                      | |
+# | |    198.51.100.3/24            | |   | |    198.51.100.4/24            | |
+# | +-------------------------------+ |   | +-------------------------------+ |
+# +-----------------------------------+   +-----------------------------------+
+
+: ${VXPORT:=4789}
+export VXPORT
+
+: ${ALL_TESTS:="
+	ping_ipv4
+	test_flood
+	test_unicast
+	reapply_config
+	ping_ipv4
+	test_flood
+	test_unicast
+	test_learning
+    "}
+
+NUM_NETIFS=6
+source lib.sh
+
+h1_create()
+{
+	simple_if_init $h1
+	tc qdisc add dev $h1 clsact
+	vlan_create $h1 10 v$h1 192.0.2.1/28
+	vlan_create $h1 20 v$h1 198.51.100.1/24
+}
+
+h1_destroy()
+{
+	vlan_destroy $h1 20
+	vlan_destroy $h1 10
+	tc qdisc del dev $h1 clsact
+	simple_if_fini $h1
+}
+
+h2_create()
+{
+	simple_if_init $h2
+	tc qdisc add dev $h2 clsact
+	vlan_create $h2 10 v$h2 192.0.2.2/28
+	vlan_create $h2 20 v$h2 198.51.100.2/24
+}
+
+h2_destroy()
+{
+	vlan_destroy $h2 20
+	vlan_destroy $h2 10
+	tc qdisc del dev $h2 clsact
+	simple_if_fini $h2
+}
+
+rp1_set_addr()
+{
+	ip address add dev $rp1 192.0.2.17/28
+
+	ip route add 192.0.2.32/28 nexthop via 192.0.2.18
+	ip route add 192.0.2.48/28 nexthop via 192.0.2.18
+}
+
+rp1_unset_addr()
+{
+	ip route del 192.0.2.48/28 nexthop via 192.0.2.18
+	ip route del 192.0.2.32/28 nexthop via 192.0.2.18
+
+	ip address del dev $rp1 192.0.2.17/28
+}
+
+switch_create()
+{
+	ip link add name br1 type bridge vlan_filtering 1 vlan_default_pvid 0 \
+		mcast_snooping 0
+	# Make sure the bridge uses the MAC address of the local port and not
+	# that of the VxLAN's device.
+	ip link set dev br1 address $(mac_get $swp1)
+	ip link set dev br1 up
+
+	ip link set dev $rp1 up
+	rp1_set_addr
+
+	ip link add name vx10 type vxlan id 1000	\
+		local 192.0.2.17 dstport "$VXPORT"	\
+		nolearning noudpcsum tos inherit ttl 100
+	ip link set dev vx10 up
+
+	ip link set dev vx10 master br1
+	bridge vlan add vid 10 dev vx10 pvid untagged
+
+	ip link add name vx20 type vxlan id 2000	\
+		local 192.0.2.17 dstport "$VXPORT"	\
+		nolearning noudpcsum tos inherit ttl 100
+	ip link set dev vx20 up
+
+	ip link set dev vx20 master br1
+	bridge vlan add vid 20 dev vx20 pvid untagged
+
+	ip link set dev $swp1 master br1
+	ip link set dev $swp1 up
+	bridge vlan add vid 10 dev $swp1
+	bridge vlan add vid 20 dev $swp1
+
+	ip link set dev $swp2 master br1
+	ip link set dev $swp2 up
+	bridge vlan add vid 10 dev $swp2
+	bridge vlan add vid 20 dev $swp2
+
+	bridge fdb append dev vx10 00:00:00:00:00:00 dst 192.0.2.34 self
+	bridge fdb append dev vx10 00:00:00:00:00:00 dst 192.0.2.50 self
+
+	bridge fdb append dev vx20 00:00:00:00:00:00 dst 192.0.2.34 self
+	bridge fdb append dev vx20 00:00:00:00:00:00 dst 192.0.2.50 self
+}
+
+switch_destroy()
+{
+	bridge fdb del dev vx20 00:00:00:00:00:00 dst 192.0.2.50 self
+	bridge fdb del dev vx20 00:00:00:00:00:00 dst 192.0.2.34 self
+
+	bridge fdb del dev vx10 00:00:00:00:00:00 dst 192.0.2.50 self
+	bridge fdb del dev vx10 00:00:00:00:00:00 dst 192.0.2.34 self
+
+	bridge vlan del vid 20 dev $swp2
+	bridge vlan del vid 10 dev $swp2
+	ip link set dev $swp2 down
+	ip link set dev $swp2 nomaster
+
+	bridge vlan del vid 20 dev $swp1
+	bridge vlan del vid 10 dev $swp1
+	ip link set dev $swp1 down
+	ip link set dev $swp1 nomaster
+
+	bridge vlan del vid 20 dev vx20
+	ip link set dev vx20 nomaster
+
+	ip link set dev vx20 down
+	ip link del dev vx20
+
+	bridge vlan del vid 10 dev vx10
+	ip link set dev vx10 nomaster
+
+	ip link set dev vx10 down
+	ip link del dev vx10
+
+	rp1_unset_addr
+	ip link set dev $rp1 down
+
+	ip link set dev br1 down
+	ip link del dev br1
+}
+
+vrp2_create()
+{
+	simple_if_init $rp2 192.0.2.18/28
+	__simple_if_init v1 v$rp2 192.0.2.33/28
+	__simple_if_init v3 v$rp2 192.0.2.49/28
+	tc qdisc add dev v1 clsact
+}
+
+vrp2_destroy()
+{
+	tc qdisc del dev v1 clsact
+	__simple_if_fini v3 192.0.2.49/28
+	__simple_if_fini v1 192.0.2.33/28
+	simple_if_fini $rp2 192.0.2.18/28
+}
+
+ns_init_common()
+{
+	local in_if=$1; shift
+	local in_addr=$1; shift
+	local other_in_addr=$1; shift
+	local nh_addr=$1; shift
+	local host_addr1=$1; shift
+	local host_addr2=$1; shift
+
+	ip link set dev $in_if up
+	ip address add dev $in_if $in_addr/28
+	tc qdisc add dev $in_if clsact
+
+	ip link add name br2 type bridge vlan_filtering 1 vlan_default_pvid 0
+	ip link set dev br2 up
+
+	ip link add name w1 type veth peer name w2
+
+	ip link set dev w1 master br2
+	ip link set dev w1 up
+
+	bridge vlan add vid 10 dev w1
+	bridge vlan add vid 20 dev w1
+
+	ip link add name vx10 type vxlan id 1000 local $in_addr \
+		dstport "$VXPORT"
+	ip link set dev vx10 up
+	bridge fdb append dev vx10 00:00:00:00:00:00 dst 192.0.2.17 self
+	bridge fdb append dev vx10 00:00:00:00:00:00 dst $other_in_addr self
+
+	ip link set dev vx10 master br2
+	tc qdisc add dev vx10 clsact
+
+	bridge vlan add vid 10 dev vx10 pvid untagged
+
+	ip link add name vx20 type vxlan id 2000 local $in_addr \
+		dstport "$VXPORT"
+	ip link set dev vx20 up
+	bridge fdb append dev vx20 00:00:00:00:00:00 dst 192.0.2.17 self
+	bridge fdb append dev vx20 00:00:00:00:00:00 dst $other_in_addr self
+
+	ip link set dev vx20 master br2
+	tc qdisc add dev vx20 clsact
+
+	bridge vlan add vid 20 dev vx20 pvid untagged
+
+	simple_if_init w2
+        vlan_create w2 10 vw2 $host_addr1/28
+        vlan_create w2 20 vw2 $host_addr2/24
+
+	ip route add 192.0.2.16/28 nexthop via $nh_addr
+	ip route add $other_in_addr/32 nexthop via $nh_addr
+}
+export -f ns_init_common
+
+ns1_create()
+{
+	ip netns add ns1
+	ip link set dev v2 netns ns1
+	in_ns ns1 \
+	      ns_init_common v2 192.0.2.34 192.0.2.50 192.0.2.33 192.0.2.3 \
+	      198.51.100.3
+}
+
+ns1_destroy()
+{
+	ip netns exec ns1 ip link set dev v2 netns 1
+	ip netns del ns1
+}
+
+ns2_create()
+{
+	ip netns add ns2
+	ip link set dev v4 netns ns2
+	in_ns ns2 \
+	      ns_init_common v4 192.0.2.50 192.0.2.34 192.0.2.49 192.0.2.4 \
+	      198.51.100.4
+}
+
+ns2_destroy()
+{
+	ip netns exec ns2 ip link set dev v4 netns 1
+	ip netns del ns2
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	swp2=${NETIFS[p3]}
+	h2=${NETIFS[p4]}
+
+	rp1=${NETIFS[p5]}
+	rp2=${NETIFS[p6]}
+
+	vrf_prepare
+	forwarding_enable
+
+	h1_create
+	h2_create
+	switch_create
+
+	ip link add name v1 type veth peer name v2
+	ip link add name v3 type veth peer name v4
+	vrp2_create
+	ns1_create
+	ns2_create
+
+	r1_mac=$(in_ns ns1 mac_get w2)
+	r2_mac=$(in_ns ns2 mac_get w2)
+	h2_mac=$(mac_get $h2)
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	ns2_destroy
+	ns1_destroy
+	vrp2_destroy
+	ip link del dev v3
+	ip link del dev v1
+
+	switch_destroy
+	h2_destroy
+	h1_destroy
+
+	forwarding_restore
+	vrf_cleanup
+}
+
+# For the first round of tests, vx10 and vx20 were the first devices to get
+# attached to the bridge, and that at the point that the local IP is already
+# configured. Try the other scenario of attaching these devices to a bridge
+# that already has local ports members, and only then assign the local IP.
+reapply_config()
+{
+	log_info "Reapplying configuration"
+
+	bridge fdb del dev vx20 00:00:00:00:00:00 dst 192.0.2.50 self
+	bridge fdb del dev vx20 00:00:00:00:00:00 dst 192.0.2.34 self
+
+	bridge fdb del dev vx10 00:00:00:00:00:00 dst 192.0.2.50 self
+	bridge fdb del dev vx10 00:00:00:00:00:00 dst 192.0.2.34 self
+
+	ip link set dev vx20 nomaster
+	ip link set dev vx10 nomaster
+
+	rp1_unset_addr
+	sleep 5
+
+	ip link set dev vx10 master br1
+	bridge vlan add vid 10 dev vx10 pvid untagged
+
+	ip link set dev vx20 master br1
+	bridge vlan add vid 20 dev vx20 pvid untagged
+
+	bridge fdb append dev vx10 00:00:00:00:00:00 dst 192.0.2.34 self
+	bridge fdb append dev vx10 00:00:00:00:00:00 dst 192.0.2.50 self
+
+	bridge fdb append dev vx20 00:00:00:00:00:00 dst 192.0.2.34 self
+	bridge fdb append dev vx20 00:00:00:00:00:00 dst 192.0.2.50 self
+
+	rp1_set_addr
+	sleep 5
+}
+
+ping_ipv4()
+{
+	ping_test $h1.10 192.0.2.2 ": local->local vid 10"
+	ping_test $h1.20 198.51.100.2 ": local->local vid 20"
+	ping_test $h1.10 192.0.2.3 ": local->remote 1 vid 10"
+	ping_test $h1.10 192.0.2.4 ": local->remote 2 vid 10"
+	ping_test $h1.20 198.51.100.3 ": local->remote 1 vid 20"
+	ping_test $h1.20 198.51.100.4 ": local->remote 2 vid 20"
+}
+
+maybe_in_ns()
+{
+	echo ${1:+in_ns} $1
+}
+
+__flood_counter_add_del()
+{
+	local add_del=$1; shift
+	local dev=$1; shift
+	local ns=$1; shift
+
+	# Putting the ICMP capture both to HW and to SW will end up
+	# double-counting the packets that are trapped to slow path, such as for
+	# the unicast test. Adding either skip_hw or skip_sw fixes this problem,
+	# but with skip_hw, the flooded packets are not counted at all, because
+	# those are dropped due to MAC address mismatch; and skip_sw is a no-go
+	# for veth-based topologies.
+	#
+	# So try to install with skip_sw and fall back to skip_sw if that fails.
+
+	$(maybe_in_ns $ns) __icmp_capture_add_del          \
+			   $add_del 100 "" $dev skip_sw 2>/dev/null || \
+	$(maybe_in_ns $ns) __icmp_capture_add_del          \
+			   $add_del 100 "" $dev skip_hw
+}
+
+flood_counter_install()
+{
+	__flood_counter_add_del add "$@"
+}
+
+flood_counter_uninstall()
+{
+	__flood_counter_add_del del "$@"
+}
+
+flood_fetch_stat()
+{
+	local dev=$1; shift
+	local ns=$1; shift
+
+	$(maybe_in_ns $ns) tc_rule_stats_get $dev 100 ingress
+}
+
+flood_fetch_stats()
+{
+	local counters=("${@}")
+	local counter
+
+	for counter in "${counters[@]}"; do
+		flood_fetch_stat $counter
+	done
+}
+
+vxlan_flood_test()
+{
+	local mac=$1; shift
+	local dst=$1; shift
+	local vid=$1; shift
+	local -a expects=("${@}")
+
+	local -a counters=($h2 "vx10 ns1" "vx20 ns1" "vx10 ns2" "vx20 ns2")
+	local counter
+	local key
+
+	# Packets reach the local host tagged whereas they reach the VxLAN
+	# devices untagged. In order to be able to use the same filter for
+	# all counters, make sure the packets also reach the local host
+	# untagged
+	bridge vlan add vid $vid dev $swp2 untagged
+	for counter in "${counters[@]}"; do
+		flood_counter_install $counter
+	done
+
+	local -a t0s=($(flood_fetch_stats "${counters[@]}"))
+	$MZ $h1 -Q $vid -c 10 -d 100msec -p 64 -b $mac -B $dst -t icmp -q
+	sleep 1
+	local -a t1s=($(flood_fetch_stats "${counters[@]}"))
+
+	for key in ${!t0s[@]}; do
+		local delta=$((t1s[$key] - t0s[$key]))
+		local expect=${expects[$key]}
+
+		((expect == delta))
+		check_err $? "${counters[$key]}: Expected to capture $expect packets, got $delta."
+	done
+
+	for counter in "${counters[@]}"; do
+		flood_counter_uninstall $counter
+	done
+	bridge vlan add vid $vid dev $swp2
+}
+
+__test_flood()
+{
+	local mac=$1; shift
+	local dst=$1; shift
+	local vid=$1; shift
+	local what=$1; shift
+	local -a expects=("${@}")
+
+	RET=0
+
+	vxlan_flood_test $mac $dst $vid "${expects[@]}"
+
+	log_test "VXLAN: $what"
+}
+
+test_flood()
+{
+	__test_flood de:ad:be:ef:13:37 192.0.2.100 10 "flood vlan 10" \
+		10 10 0 10 0
+	__test_flood ca:fe:be:ef:13:37 198.51.100.100 20 "flood vlan 20" \
+		10 0 10 0 10
+}
+
+vxlan_fdb_add_del()
+{
+	local add_del=$1; shift
+	local vid=$1; shift
+	local mac=$1; shift
+	local dev=$1; shift
+	local dst=$1; shift
+
+	bridge fdb $add_del dev $dev $mac self static permanent \
+		${dst:+dst} $dst 2>/dev/null
+	bridge fdb $add_del dev $dev $mac master static vlan $vid 2>/dev/null
+}
+
+__test_unicast()
+{
+	local mac=$1; shift
+	local dst=$1; shift
+	local hit_idx=$1; shift
+	local vid=$1; shift
+	local what=$1; shift
+
+	RET=0
+
+	local -a expects=(0 0 0 0 0)
+	expects[$hit_idx]=10
+
+	vxlan_flood_test $mac $dst $vid "${expects[@]}"
+
+	log_test "VXLAN: $what"
+}
+
+test_unicast()
+{
+	local -a targets=("$h2_mac $h2"
+			  "$r1_mac vx10 192.0.2.34"
+			  "$r2_mac vx10 192.0.2.50")
+	local target
+
+	log_info "unicast vlan 10"
+
+	for target in "${targets[@]}"; do
+		vxlan_fdb_add_del add 10 $target
+	done
+
+	__test_unicast $h2_mac 192.0.2.2 0 10 "local MAC unicast"
+	__test_unicast $r1_mac 192.0.2.3 1 10 "remote MAC 1 unicast"
+	__test_unicast $r2_mac 192.0.2.4 3 10 "remote MAC 2 unicast"
+
+	for target in "${targets[@]}"; do
+		vxlan_fdb_add_del del 10 $target
+	done
+
+	log_info "unicast vlan 20"
+
+	targets=("$h2_mac $h2" "$r1_mac vx20 192.0.2.34" \
+		 "$r2_mac vx20 192.0.2.50")
+
+	for target in "${targets[@]}"; do
+		vxlan_fdb_add_del add 20 $target
+	done
+
+	__test_unicast $h2_mac 198.51.100.2 0 20 "local MAC unicast"
+	__test_unicast $r1_mac 198.51.100.3 2 20 "remote MAC 1 unicast"
+	__test_unicast $r2_mac 198.51.100.4 4 20 "remote MAC 2 unicast"
+
+	for target in "${targets[@]}"; do
+		vxlan_fdb_add_del del 20 $target
+	done
+}
+
+vxlan_ping_test()
+{
+	local ping_dev=$1; shift
+	local ping_dip=$1; shift
+	local ping_args=$1; shift
+	local capture_dev=$1; shift
+	local capture_dir=$1; shift
+	local capture_pref=$1; shift
+	local expect=$1; shift
+
+	local t0=$(tc_rule_stats_get $capture_dev $capture_pref $capture_dir)
+	ping_do $ping_dev $ping_dip "$ping_args"
+	local t1=$(tc_rule_stats_get $capture_dev $capture_pref $capture_dir)
+	local delta=$((t1 - t0))
+
+	# Tolerate a couple stray extra packets.
+	((expect <= delta && delta <= expect + 2))
+	check_err $? "$capture_dev: Expected to capture $expect packets, got $delta."
+}
+
+__test_learning()
+{
+	local -a expects=(0 0 0 0 0)
+	local mac=$1; shift
+	local dst=$1; shift
+	local vid=$1; shift
+	local idx1=$1; shift
+	local idx2=$1; shift
+	local vx=vx$vid
+
+	# Check that flooding works
+	RET=0
+
+	expects[0]=10; expects[$idx1]=10; expects[$idx2]=10
+	vxlan_flood_test $mac $dst $vid "${expects[@]}"
+
+	log_test "VXLAN: flood before learning"
+
+	# Send a packet with source mac set to $mac from host w2 and check that
+	# a corresponding entry is created in the VxLAN device
+	RET=0
+
+	in_ns ns1 $MZ w2 -Q $vid -c 1 -p 64 -a $mac -b ff:ff:ff:ff:ff:ff \
+		-B $dst -t icmp -q
+	sleep 1
+
+	bridge fdb show brport $vx | grep $mac | grep -q self
+	check_err $?
+	bridge fdb show brport $vx | grep $mac | grep "vlan $vid" \
+		| grep -q -v self
+	check_err $?
+
+	log_test "VXLAN: show learned FDB entry"
+
+	# Repeat first test and check that packets only reach host w2 in ns1
+	RET=0
+
+	expects[0]=0; expects[$idx1]=10; expects[$idx2]=0
+	vxlan_flood_test $mac $dst $vid "${expects[@]}"
+
+	log_test "VXLAN: learned FDB entry"
+
+	# Delete the learned FDB entry from the VxLAN and bridge devices and
+	# check that packets are flooded
+	RET=0
+
+	bridge fdb del dev $vx $mac master self vlan $vid
+	sleep 1
+
+	expects[0]=10; expects[$idx1]=10; expects[$idx2]=10
+	vxlan_flood_test $mac $dst $vid "${expects[@]}"
+
+	log_test "VXLAN: deletion of learned FDB entry"
+
+	# Re-learn the first FDB entry and check that it is correctly aged-out
+	RET=0
+
+	in_ns ns1 $MZ w2 -Q $vid -c 1 -p 64 -a $mac -b ff:ff:ff:ff:ff:ff \
+		-B $dst -t icmp -q
+	sleep 1
+
+	bridge fdb show brport $vx | grep $mac | grep -q self
+	check_err $?
+	bridge fdb show brport $vx | grep $mac | grep "vlan $vid" \
+		| grep -q -v self
+	check_err $?
+
+	expects[0]=0; expects[$idx1]=10; expects[$idx2]=0
+	vxlan_flood_test $mac $dst $vid "${expects[@]}"
+
+	sleep 20
+
+	bridge fdb show brport $vx | grep $mac | grep -q self
+	check_fail $?
+	bridge fdb show brport $vx | grep $mac | grep "vlan $vid" \
+		| grep -q -v self
+	check_fail $?
+
+	expects[0]=10; expects[$idx1]=10; expects[$idx2]=10
+	vxlan_flood_test $mac $dst $vid "${expects[@]}"
+
+	log_test "VXLAN: Ageing of learned FDB entry"
+
+	# Toggle learning on the bridge port and check that the bridge's FDB
+	# is populated only when it should
+	RET=0
+
+	ip link set dev $vx type bridge_slave learning off
+
+	in_ns ns1 $MZ w2 -Q $vid -c 1 -p 64 -a $mac -b ff:ff:ff:ff:ff:ff \
+		-B $dst -t icmp -q
+	sleep 1
+
+	bridge fdb show brport $vx | grep $mac | grep "vlan $vid" \
+		| grep -q -v self
+	check_fail $?
+
+	ip link set dev $vx type bridge_slave learning on
+
+	in_ns ns1 $MZ w2 -Q $vid -c 1 -p 64 -a $mac -b ff:ff:ff:ff:ff:ff \
+		-B $dst -t icmp -q
+	sleep 1
+
+	bridge fdb show brport $vx | grep $mac | grep "vlan $vid" \
+		| grep -q -v self
+	check_err $?
+
+	log_test "VXLAN: learning toggling on bridge port"
+}
+
+test_learning()
+{
+	local mac=de:ad:be:ef:13:37
+	local dst=192.0.2.100
+	local vid=10
+
+	# Enable learning on the VxLAN devices and set ageing time to 10 seconds
+	ip link set dev br1 type bridge ageing_time 1000
+	ip link set dev vx10 type vxlan ageing 10
+	ip link set dev vx10 type vxlan learning
+	ip link set dev vx20 type vxlan ageing 10
+	ip link set dev vx20 type vxlan learning
+	reapply_config
+
+	log_info "learning vlan 10"
+
+	__test_learning $mac $dst $vid 1 3
+
+	log_info "learning vlan 20"
+
+	mac=ca:fe:be:ef:13:37
+	dst=198.51.100.100
+	vid=20
+
+	__test_learning $mac $dst $vid 2 4
+
+	# Restore previous settings
+	ip link set dev vx20 type vxlan nolearning
+	ip link set dev vx20 type vxlan ageing 300
+	ip link set dev vx10 type vxlan nolearning
+	ip link set dev vx10 type vxlan ageing 300
+	ip link set dev br1 type bridge ageing_time 30000
+	reapply_config
+}
+
+test_all()
+{
+	log_info "Running tests with UDP port $VXPORT"
+	tests_run
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+test_all
+
+exit $EXIT_STATUS

tools/testing/selftests/net/forwarding/vxlan_bridge_1q_port_8472.sh

···
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# A wrapper to run VXLAN tests with an unusual port number.
+
+VXPORT=8472
+ALL_TESTS="
+	ping_ipv4
+"
+source vxlan_bridge_1q.sh