Merge branch 'pruss-fix' into fixes

Merge in a fix for pruss reset issue caused by enabling pruss for am335x.

25 files changed, 1439 insertions, 766 deletions

diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig
index 00bb89b2d86f..548285539752 100644
--- a/net/dsa/Kconfig
+++ b/net/dsa/Kconfig
@@ -18,16 +18,6 @@ if NET_DSA
 
 # Drivers must select the appropriate tagging format(s)
 
-config NET_DSA_TAG_8021Q
-	tristate
-	select VLAN_8021Q
-	help
-	  Unlike the other tagging protocols, the 802.1Q config option simply
-	  provides helpers for other tagging implementations that might rely on
-	  VLAN in one way or another. It is not a complete solution.
-
-	  Drivers which use these helpers should select this as dependency.
-
 config NET_DSA_TAG_AR9331
 	tristate "Tag driver for Atheros AR9331 SoC with built-in switch"
 	help
@@ -126,7 +116,6 @@ config NET_DSA_TAG_OCELOT_8021Q
 	tristate "Tag driver for Ocelot family of switches, using VLAN"
 	depends on MSCC_OCELOT_SWITCH_LIB || \
 	          (MSCC_OCELOT_SWITCH_LIB=n && COMPILE_TEST)
-	select NET_DSA_TAG_8021Q
 	help
 	  Say Y or M if you want to enable support for tagging frames with a
 	  custom VLAN-based header. Frames that require timestamping, such as
@@ -149,7 +138,7 @@ config NET_DSA_TAG_LAN9303
 
 config NET_DSA_TAG_SJA1105
 	tristate "Tag driver for NXP SJA1105 switches"
-	select NET_DSA_TAG_8021Q
+	depends on NET_DSA_SJA1105 || !NET_DSA_SJA1105
 	select PACKING
 	help
 	  Say Y or M if you want to enable support for tagging frames with the
diff --git a/net/dsa/Makefile b/net/dsa/Makefile
index 44bc79952b8b..67ea009f242c 100644
--- a/net/dsa/Makefile
+++ b/net/dsa/Makefile
@@ -1,10 +1,9 @@
 # SPDX-License-Identifier: GPL-2.0
 # the core
 obj-$(CONFIG_NET_DSA) += dsa_core.o
-dsa_core-y += dsa.o dsa2.o master.o port.o slave.o switch.o
+dsa_core-y += dsa.o dsa2.o master.o port.o slave.o switch.o tag_8021q.o
 
 # tagging formats
-obj-$(CONFIG_NET_DSA_TAG_8021Q) += tag_8021q.o
 obj-$(CONFIG_NET_DSA_TAG_AR9331) += tag_ar9331.o
 obj-$(CONFIG_NET_DSA_TAG_BRCM_COMMON) += tag_brcm.o
 obj-$(CONFIG_NET_DSA_TAG_DSA_COMMON) += tag_dsa.o
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 84cad1be9ce4..1dc45e40f961 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -238,7 +238,7 @@ static int dsa_switch_rcv(struct sk_buff *skb, struct net_device *dev,
 	if (!skb)
 		return 0;
 
-	nskb = cpu_dp->rcv(skb, dev, pt);
+	nskb = cpu_dp->rcv(skb, dev);
 	if (!nskb) {
 		kfree_skb(skb);
 		return 0;
diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c
index 185629f27f80..1b2b25d7bd02 100644
--- a/net/dsa/dsa2.c
+++ b/net/dsa/dsa2.c
@@ -21,6 +21,9 @@
 static DEFINE_MUTEX(dsa2_mutex);
 LIST_HEAD(dsa_tree_list);
 
+/* Track the bridges with forwarding offload enabled */
+static unsigned long dsa_fwd_offloading_bridges;
+
 /**
  * dsa_tree_notify - Execute code for all switches in a DSA switch tree.
  * @dst: collection of struct dsa_switch devices to notify.
@@ -49,6 +52,9 @@ int dsa_tree_notify(struct dsa_switch_tree *dst, unsigned long e, void *v)
  * Can be used to notify the switching fabric of events such as cross-chip
  * bridging between disjoint trees (such as islands of tagger-compatible
  * switches bridged by an incompatible middle switch).
+ *
+ * WARNING: this function is not reliable during probe time, because probing
+ * between trees is asynchronous and not all DSA trees might have probed.
  */
 int dsa_broadcast(unsigned long e, void *v)
 {
@@ -123,6 +129,51 @@ void dsa_lag_unmap(struct dsa_switch_tree *dst, struct net_device *lag)
 	}
 }
 
+static int dsa_bridge_num_find(const struct net_device *bridge_dev)
+{
+	struct dsa_switch_tree *dst;
+	struct dsa_port *dp;
+
+	/* When preparing the offload for a port, it will have a valid
+	 * dp->bridge_dev pointer but a not yet valid dp->bridge_num.
+	 * However there might be other ports having the same dp->bridge_dev
+	 * and a valid dp->bridge_num, so just ignore this port.
+	 */
+	list_for_each_entry(dst, &dsa_tree_list, list)
+		list_for_each_entry(dp, &dst->ports, list)
+			if (dp->bridge_dev == bridge_dev &&
+			    dp->bridge_num != -1)
+				return dp->bridge_num;
+
+	return -1;
+}
+
+int dsa_bridge_num_get(const struct net_device *bridge_dev, int max)
+{
+	int bridge_num = dsa_bridge_num_find(bridge_dev);
+
+	if (bridge_num < 0) {
+		/* First port that offloads TX forwarding for this bridge */
+		bridge_num = find_first_zero_bit(&dsa_fwd_offloading_bridges,
+						 DSA_MAX_NUM_OFFLOADING_BRIDGES);
+		if (bridge_num >= max)
+			return -1;
+
+		set_bit(bridge_num, &dsa_fwd_offloading_bridges);
+	}
+
+	return bridge_num;
+}
+
+void dsa_bridge_num_put(const struct net_device *bridge_dev, int bridge_num)
+{
+	/* Check if the bridge is still in use, otherwise it is time
+	 * to clean it up so we can reuse this bridge_num later.
+	 */
+	if (!dsa_bridge_num_find(bridge_dev))
+		clear_bit(bridge_num, &dsa_fwd_offloading_bridges);
+}
+
 struct dsa_switch *dsa_switch_find(int tree_index, int sw_index)
 {
 	struct dsa_switch_tree *dst;
@@ -311,6 +362,9 @@ static struct dsa_port *dsa_tree_find_first_cpu(struct dsa_switch_tree *dst)
 	return NULL;
 }
 
+/* Assign the default CPU port (the first one in the tree) to all ports of the
+ * fabric which don't already have one as part of their own switch.
+ */
 static int dsa_tree_setup_default_cpu(struct dsa_switch_tree *dst)
 {
 	struct dsa_port *cpu_dp, *dp;
@@ -321,15 +375,48 @@ static int dsa_tree_setup_default_cpu(struct dsa_switch_tree *dst)
 		return -EINVAL;
 	}
 
-	/* Assign the default CPU port to all ports of the fabric */
-	list_for_each_entry(dp, &dst->ports, list)
+	list_for_each_entry(dp, &dst->ports, list) {
+		if (dp->cpu_dp)
+			continue;
+
 		if (dsa_port_is_user(dp) || dsa_port_is_dsa(dp))
 			dp->cpu_dp = cpu_dp;
+	}
 
 	return 0;
 }
 
-static void dsa_tree_teardown_default_cpu(struct dsa_switch_tree *dst)
+/* Perform initial assignment of CPU ports to user ports and DSA links in the
+ * fabric, giving preference to CPU ports local to each switch. Default to
+ * using the first CPU port in the switch tree if the port does not have a CPU
+ * port local to this switch.
+ */
+static int dsa_tree_setup_cpu_ports(struct dsa_switch_tree *dst)
+{
+	struct dsa_port *cpu_dp, *dp;
+
+	list_for_each_entry(cpu_dp, &dst->ports, list) {
+		if (!dsa_port_is_cpu(cpu_dp))
+			continue;
+
+		list_for_each_entry(dp, &dst->ports, list) {
+			/* Prefer a local CPU port */
+			if (dp->ds != cpu_dp->ds)
+				continue;
+
+			/* Prefer the first local CPU port found */
+			if (dp->cpu_dp)
+				continue;
+
+			if (dsa_port_is_user(dp) || dsa_port_is_dsa(dp))
+				dp->cpu_dp = cpu_dp;
+		}
+	}
+
+	return dsa_tree_setup_default_cpu(dst);
+}
+
+static void dsa_tree_teardown_cpu_ports(struct dsa_switch_tree *dst)
 {
 	struct dsa_port *dp;
 
@@ -710,13 +797,14 @@ static int dsa_switch_setup(struct dsa_switch *ds)
 	/* Add the switch to devlink before calling setup, so that setup can
 	 * add dpipe tables
 	 */
-	ds->devlink = devlink_alloc(&dsa_devlink_ops, sizeof(*dl_priv));
+	ds->devlink =
+		devlink_alloc(&dsa_devlink_ops, sizeof(*dl_priv), ds->dev);
 	if (!ds->devlink)
 		return -ENOMEM;
 	dl_priv = devlink_priv(ds->devlink);
 	dl_priv->ds = ds;
 
-	err = devlink_register(ds->devlink, ds->dev);
+	err = devlink_register(ds->devlink);
 	if (err)
 		goto free_devlink;
 
@@ -921,13 +1009,13 @@ static int dsa_tree_setup(struct dsa_switch_tree *dst)
 	if (!complete)
 		return 0;
 
-	err = dsa_tree_setup_default_cpu(dst);
+	err = dsa_tree_setup_cpu_ports(dst);
 	if (err)
 		return err;
 
 	err = dsa_tree_setup_switches(dst);
 	if (err)
-		goto teardown_default_cpu;
+		goto teardown_cpu_ports;
 
 	err = dsa_tree_setup_master(dst);
 	if (err)
@@ -947,8 +1035,8 @@ teardown_master:
 	dsa_tree_teardown_master(dst);
 teardown_switches:
 	dsa_tree_teardown_switches(dst);
-teardown_default_cpu:
-	dsa_tree_teardown_default_cpu(dst);
+teardown_cpu_ports:
+	dsa_tree_teardown_cpu_ports(dst);
 
 	return err;
 }
@@ -966,7 +1054,7 @@ static void dsa_tree_teardown(struct dsa_switch_tree *dst)
 
 	dsa_tree_teardown_switches(dst);
 
-	dsa_tree_teardown_default_cpu(dst);
+	dsa_tree_teardown_cpu_ports(dst);
 
 	list_for_each_entry_safe(dl, next, &dst->rtable, list) {
 		list_del(&dl->list);
@@ -1044,6 +1132,7 @@ static struct dsa_port *dsa_port_touch(struct dsa_switch *ds, int index)
 
 	dp->ds = ds;
 	dp->index = index;
+	dp->bridge_num = -1;
 
 	INIT_LIST_HEAD(&dp->list);
 	list_add_tail(&dp->list, &dst->ports);
@@ -1265,6 +1354,9 @@ static int dsa_switch_parse_member_of(struct dsa_switch *ds,
 		return -EEXIST;
 	}
 
+	if (ds->dst->last_switch < ds->index)
+		ds->dst->last_switch = ds->index;
+
 	return 0;
 }
 
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index f201c33980bf..33ab7d7af9eb 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -14,6 +14,8 @@
 #include <net/dsa.h>
 #include <net/gro_cells.h>
 
+#define DSA_MAX_NUM_OFFLOADING_BRIDGES		BITS_PER_LONG
+
 enum {
 	DSA_NOTIFIER_AGEING_TIME,
 	DSA_NOTIFIER_BRIDGE_JOIN,
@@ -39,6 +41,8 @@ enum {
 	DSA_NOTIFIER_MRP_DEL,
 	DSA_NOTIFIER_MRP_ADD_RING_ROLE,
 	DSA_NOTIFIER_MRP_DEL_RING_ROLE,
+	DSA_NOTIFIER_TAG_8021Q_VLAN_ADD,
+	DSA_NOTIFIER_TAG_8021Q_VLAN_DEL,
 };
 
 /* DSA_NOTIFIER_AGEING_TIME */
@@ -113,6 +117,14 @@ struct dsa_notifier_mrp_ring_role_info {
 	int port;
 };
 
+/* DSA_NOTIFIER_TAG_8021Q_VLAN_* */
+struct dsa_notifier_tag_8021q_vlan_info {
+	int tree_index;
+	int sw_index;
+	int port;
+	u16 vid;
+};
+
 struct dsa_switchdev_event_work {
 	struct dsa_switch *ds;
 	int port;
@@ -187,23 +199,21 @@ static inline struct net_device *dsa_master_find_slave(struct net_device *dev,
 /* port.c */
 void dsa_port_set_tag_protocol(struct dsa_port *cpu_dp,
 			       const struct dsa_device_ops *tag_ops);
-int dsa_port_set_state(struct dsa_port *dp, u8 state);
+int dsa_port_set_state(struct dsa_port *dp, u8 state, bool do_fast_age);
 int dsa_port_enable_rt(struct dsa_port *dp, struct phy_device *phy);
 int dsa_port_enable(struct dsa_port *dp, struct phy_device *phy);
 void dsa_port_disable_rt(struct dsa_port *dp);
 void dsa_port_disable(struct dsa_port *dp);
 int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br,
 			 struct netlink_ext_ack *extack);
-int dsa_port_pre_bridge_leave(struct dsa_port *dp, struct net_device *br,
-			      struct netlink_ext_ack *extack);
+void dsa_port_pre_bridge_leave(struct dsa_port *dp, struct net_device *br);
 void dsa_port_bridge_leave(struct dsa_port *dp, struct net_device *br);
 int dsa_port_lag_change(struct dsa_port *dp,
 			struct netdev_lag_lower_state_info *linfo);
 int dsa_port_lag_join(struct dsa_port *dp, struct net_device *lag_dev,
 		      struct netdev_lag_upper_info *uinfo,
 		      struct netlink_ext_ack *extack);
-int dsa_port_pre_lag_leave(struct dsa_port *dp, struct net_device *lag_dev,
-			   struct netlink_ext_ack *extack);
+void dsa_port_pre_lag_leave(struct dsa_port *dp, struct net_device *lag_dev);
 void dsa_port_lag_leave(struct dsa_port *dp, struct net_device *lag_dev);
 int dsa_port_vlan_filtering(struct dsa_port *dp, bool vlan_filtering,
 			    struct netlink_ext_ack *extack);
@@ -231,11 +241,9 @@ int dsa_port_host_mdb_del(const struct dsa_port *dp,
 int dsa_port_pre_bridge_flags(const struct dsa_port *dp,
 			      struct switchdev_brport_flags flags,
 			      struct netlink_ext_ack *extack);
-int dsa_port_bridge_flags(const struct dsa_port *dp,
+int dsa_port_bridge_flags(struct dsa_port *dp,
 			  struct switchdev_brport_flags flags,
 			  struct netlink_ext_ack *extack);
-int dsa_port_mrouter(struct dsa_port *dp, bool mrouter,
-		     struct netlink_ext_ack *extack);
 int dsa_port_vlan_add(struct dsa_port *dp,
 		      const struct switchdev_obj_port_vlan *vlan,
 		      struct netlink_ext_ack *extack);
@@ -253,16 +261,18 @@ int dsa_port_link_register_of(struct dsa_port *dp);
 void dsa_port_link_unregister_of(struct dsa_port *dp);
 int dsa_port_hsr_join(struct dsa_port *dp, struct net_device *hsr);
 void dsa_port_hsr_leave(struct dsa_port *dp, struct net_device *hsr);
+int dsa_port_tag_8021q_vlan_add(struct dsa_port *dp, u16 vid, bool broadcast);
+void dsa_port_tag_8021q_vlan_del(struct dsa_port *dp, u16 vid, bool broadcast);
 extern const struct phylink_mac_ops dsa_port_phylink_mac_ops;
 
 static inline bool dsa_port_offloads_bridge_port(struct dsa_port *dp,
-						 struct net_device *dev)
+						 const struct net_device *dev)
 {
 	return dsa_port_to_bridge_port(dp) == dev;
 }
 
 static inline bool dsa_port_offloads_bridge(struct dsa_port *dp,
-					    struct net_device *bridge_dev)
+					    const struct net_device *bridge_dev)
 {
 	/* DSA ports connected to a bridge, and event was emitted
 	 * for the bridge.
@@ -272,7 +282,7 @@ static inline bool dsa_port_offloads_bridge(struct dsa_port *dp,
 
 /* Returns true if any port of this tree offloads the given net_device */
 static inline bool dsa_tree_offloads_bridge_port(struct dsa_switch_tree *dst,
-						 struct net_device *dev)
+						 const struct net_device *dev)
 {
 	struct dsa_port *dp;
 
@@ -283,6 +293,19 @@ static inline bool dsa_tree_offloads_bridge_port(struct dsa_switch_tree *dst,
 	return false;
 }
 
+/* Returns true if any port of this tree offloads the given bridge */
+static inline bool dsa_tree_offloads_bridge(struct dsa_switch_tree *dst,
+					    const struct net_device *bridge_dev)
+{
+	struct dsa_port *dp;
+
+	list_for_each_entry(dp, &dst->ports, list)
+		if (dsa_port_offloads_bridge(dp, bridge_dev))
+			return true;
+
+	return false;
+}
+
 /* slave.c */
 extern const struct dsa_device_ops notag_netdev_ops;
 extern struct notifier_block dsa_slave_switchdev_notifier;
@@ -297,6 +320,8 @@ int dsa_slave_register_notifier(void);
 void dsa_slave_unregister_notifier(void);
 void dsa_slave_setup_tagger(struct net_device *slave);
 int dsa_slave_change_mtu(struct net_device *dev, int new_mtu);
+int dsa_slave_manage_vlan_filtering(struct net_device *dev,
+				    bool vlan_filtering);
 
 static inline struct dsa_port *dsa_slave_to_port(const struct net_device *dev)
 {
@@ -372,6 +397,141 @@ static inline struct sk_buff *dsa_untag_bridge_pvid(struct sk_buff *skb)
 	return skb;
 }
 
+/* For switches without hardware support for DSA tagging to be able
+ * to support termination through the bridge.
+ */
+static inline struct net_device *
+dsa_find_designated_bridge_port_by_vid(struct net_device *master, u16 vid)
+{
+	struct dsa_port *cpu_dp = master->dsa_ptr;
+	struct dsa_switch_tree *dst = cpu_dp->dst;
+	struct bridge_vlan_info vinfo;
+	struct net_device *slave;
+	struct dsa_port *dp;
+	int err;
+
+	list_for_each_entry(dp, &dst->ports, list) {
+		if (dp->type != DSA_PORT_TYPE_USER)
+			continue;
+
+		if (!dp->bridge_dev)
+			continue;
+
+		if (dp->stp_state != BR_STATE_LEARNING &&
+		    dp->stp_state != BR_STATE_FORWARDING)
+			continue;
+
+		/* Since the bridge might learn this packet, keep the CPU port
+		 * affinity with the port that will be used for the reply on
+		 * xmit.
+		 */
+		if (dp->cpu_dp != cpu_dp)
+			continue;
+
+		slave = dp->slave;
+
+		err = br_vlan_get_info_rcu(slave, vid, &vinfo);
+		if (err)
+			continue;
+
+		return slave;
+	}
+
+	return NULL;
+}
+
+/* If the ingress port offloads the bridge, we mark the frame as autonomously
+ * forwarded by hardware, so the software bridge doesn't forward in twice, back
+ * to us, because we already did. However, if we're in fallback mode and we do
+ * software bridging, we are not offloading it, therefore the dp->bridge_dev
+ * pointer is not populated, and flooding needs to be done by software (we are
+ * effectively operating in standalone ports mode).
+ */
+static inline void dsa_default_offload_fwd_mark(struct sk_buff *skb)
+{
+	struct dsa_port *dp = dsa_slave_to_port(skb->dev);
+
+	skb->offload_fwd_mark = !!(dp->bridge_dev);
+}
+
+/* Helper for removing DSA header tags from packets in the RX path.
+ * Must not be called before skb_pull(len).
+ *                                                                 skb->data
+ *                                                                         |
+ *                                                                         v
+ * |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |
+ * +-----------------------+-----------------------+---------------+-------+
+ * |    Destination MAC    |      Source MAC       |  DSA header   | EType |
+ * +-----------------------+-----------------------+---------------+-------+
+ *                                                 |               |
+ * <----- len ----->                               <----- len ----->
+ *                 |
+ *       >>>>>>>   v
+ *       >>>>>>>   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |
+ *       >>>>>>>   +-----------------------+-----------------------+-------+
+ *       >>>>>>>   |    Destination MAC    |      Source MAC       | EType |
+ *                 +-----------------------+-----------------------+-------+
+ *                                                                         ^
+ *                                                                         |
+ *                                                                 skb->data
+ */
+static inline void dsa_strip_etype_header(struct sk_buff *skb, int len)
+{
+	memmove(skb->data - ETH_HLEN, skb->data - ETH_HLEN - len, 2 * ETH_ALEN);
+}
+
+/* Helper for creating space for DSA header tags in TX path packets.
+ * Must not be called before skb_push(len).
+ *
+ * Before:
+ *
+ *       <<<<<<<   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |
+ * ^     <<<<<<<   +-----------------------+-----------------------+-------+
+ * |     <<<<<<<   |    Destination MAC    |      Source MAC       | EType |
+ * |               +-----------------------+-----------------------+-------+
+ * <----- len ----->
+ * |
+ * |
+ * skb->data
+ *
+ * After:
+ *
+ * |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |
+ * +-----------------------+-----------------------+---------------+-------+
+ * |    Destination MAC    |      Source MAC       |  DSA header   | EType |
+ * +-----------------------+-----------------------+---------------+-------+
+ * ^                                               |               |
+ * |                                               <----- len ----->
+ * skb->data
+ */
+static inline void dsa_alloc_etype_header(struct sk_buff *skb, int len)
+{
+	memmove(skb->data, skb->data + len, 2 * ETH_ALEN);
+}
+
+/* On RX, eth_type_trans() on the DSA master pulls ETH_HLEN bytes starting from
+ * skb_mac_header(skb), which leaves skb->data pointing at the first byte after
+ * what the DSA master perceives as the EtherType (the beginning of the L3
+ * protocol). Since DSA EtherType header taggers treat the EtherType as part of
+ * the DSA tag itself, and the EtherType is 2 bytes in length, the DSA header
+ * is located 2 bytes behind skb->data. Note that EtherType in this context
+ * means the first 2 bytes of the DSA header, not the encapsulated EtherType
+ * that will become visible after the DSA header is stripped.
+ */
+static inline void *dsa_etype_header_pos_rx(struct sk_buff *skb)
+{
+	return skb->data - 2;
+}
+
+/* On TX, skb->data points to skb_mac_header(skb), which means that EtherType
+ * header taggers start exactly where the EtherType is (the EtherType is
+ * treated as part of the DSA header).
+ */
+static inline void *dsa_etype_header_pos_tx(struct sk_buff *skb)
+{
+	return skb->data + 2 * ETH_ALEN;
+}
+
 /* switch.c */
 int dsa_switch_register_notifier(struct dsa_switch *ds);
 void dsa_switch_unregister_notifier(struct dsa_switch *ds);
@@ -385,6 +545,18 @@ int dsa_tree_change_tag_proto(struct dsa_switch_tree *dst,
 			      struct net_device *master,
 			      const struct dsa_device_ops *tag_ops,
 			      const struct dsa_device_ops *old_tag_ops);
+int dsa_bridge_num_get(const struct net_device *bridge_dev, int max);
+void dsa_bridge_num_put(const struct net_device *bridge_dev, int bridge_num);
+
+/* tag_8021q.c */
+int dsa_tag_8021q_bridge_join(struct dsa_switch *ds,
+			      struct dsa_notifier_bridge_info *info);
+int dsa_tag_8021q_bridge_leave(struct dsa_switch *ds,
+			       struct dsa_notifier_bridge_info *info);
+int dsa_switch_tag_8021q_vlan_add(struct dsa_switch *ds,
+				  struct dsa_notifier_tag_8021q_vlan_info *info);
+int dsa_switch_tag_8021q_vlan_del(struct dsa_switch *ds,
+				  struct dsa_notifier_tag_8021q_vlan_info *info);
 
 extern struct list_head dsa_tree_list;
 
diff --git a/net/dsa/master.c b/net/dsa/master.c
index 3fc90e36772d..e8e19857621b 100644
--- a/net/dsa/master.c
+++ b/net/dsa/master.c
@@ -210,14 +210,14 @@ static int dsa_master_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		break;
 	}
 
-	if (dev->netdev_ops->ndo_do_ioctl)
-		err = dev->netdev_ops->ndo_do_ioctl(dev, ifr, cmd);
+	if (dev->netdev_ops->ndo_eth_ioctl)
+		err = dev->netdev_ops->ndo_eth_ioctl(dev, ifr, cmd);
 
 	return err;
 }
 
 static const struct dsa_netdevice_ops dsa_netdev_ops = {
-	.ndo_do_ioctl = dsa_master_ioctl,
+	.ndo_eth_ioctl = dsa_master_ioctl,
 };
 
 static int dsa_master_ethtool_setup(struct net_device *dev)
diff --git a/net/dsa/port.c b/net/dsa/port.c
index 28b45b7e66df..616330a16d31 100644
--- a/net/dsa/port.c
+++ b/net/dsa/port.c
@@ -30,7 +30,52 @@ static int dsa_port_notify(const struct dsa_port *dp, unsigned long e, void *v)
 	return dsa_tree_notify(dp->ds->dst, e, v);
 }
 
-int dsa_port_set_state(struct dsa_port *dp, u8 state)
+static void dsa_port_notify_bridge_fdb_flush(const struct dsa_port *dp)
+{
+	struct net_device *brport_dev = dsa_port_to_bridge_port(dp);
+	struct switchdev_notifier_fdb_info info = {
+		/* flush all VLANs */
+		.vid = 0,
+	};
+
+	/* When the port becomes standalone it has already left the bridge.
+	 * Don't notify the bridge in that case.
+	 */
+	if (!brport_dev)
+		return;
+
+	call_switchdev_notifiers(SWITCHDEV_FDB_FLUSH_TO_BRIDGE,
+				 brport_dev, &info.info, NULL);
+}
+
+static void dsa_port_fast_age(const struct dsa_port *dp)
+{
+	struct dsa_switch *ds = dp->ds;
+
+	if (!ds->ops->port_fast_age)
+		return;
+
+	ds->ops->port_fast_age(ds, dp->index);
+
+	dsa_port_notify_bridge_fdb_flush(dp);
+}
+
+static bool dsa_port_can_configure_learning(struct dsa_port *dp)
+{
+	struct switchdev_brport_flags flags = {
+		.mask = BR_LEARNING,
+	};
+	struct dsa_switch *ds = dp->ds;
+	int err;
+
+	if (!ds->ops->port_bridge_flags || !ds->ops->port_pre_bridge_flags)
+		return false;
+
+	err = ds->ops->port_pre_bridge_flags(ds, dp->index, flags, NULL);
+	return !err;
+}
+
+int dsa_port_set_state(struct dsa_port *dp, u8 state, bool do_fast_age)
 {
 	struct dsa_switch *ds = dp->ds;
 	int port = dp->index;
@@ -40,10 +85,14 @@ int dsa_port_set_state(struct dsa_port *dp, u8 state)
 
 	ds->ops->port_stp_state_set(ds, port, state);
 
-	if (ds->ops->port_fast_age) {
+	if (!dsa_port_can_configure_learning(dp) ||
+	    (do_fast_age && dp->learning)) {
 		/* Fast age FDB entries or flush appropriate forwarding database
 		 * for the given port, if we are moving it from Learning or
 		 * Forwarding state, to Disabled or Blocking or Listening state.
+		 * Ports that were standalone before the STP state change don't
+		 * need to fast age the FDB, since address learning is off in
+		 * standalone mode.
 		 */
 
 		if ((dp->stp_state == BR_STATE_LEARNING ||
@@ -51,7 +100,7 @@ int dsa_port_set_state(struct dsa_port *dp, u8 state)
 		    (state == BR_STATE_DISABLED ||
 		     state == BR_STATE_BLOCKING ||
 		     state == BR_STATE_LISTENING))
-			ds->ops->port_fast_age(ds, port);
+			dsa_port_fast_age(dp);
 	}
 
 	dp->stp_state = state;
@@ -59,11 +108,12 @@ int dsa_port_set_state(struct dsa_port *dp, u8 state)
 	return 0;
 }
 
-static void dsa_port_set_state_now(struct dsa_port *dp, u8 state)
+static void dsa_port_set_state_now(struct dsa_port *dp, u8 state,
+				   bool do_fast_age)
 {
 	int err;
 
-	err = dsa_port_set_state(dp, state);
+	err = dsa_port_set_state(dp, state, do_fast_age);
 	if (err)
 		pr_err("DSA: failed to set STP state %u (%d)\n", state, err);
 }
@@ -81,7 +131,7 @@ int dsa_port_enable_rt(struct dsa_port *dp, struct phy_device *phy)
 	}
 
 	if (!dp->bridge_dev)
-		dsa_port_set_state_now(dp, BR_STATE_FORWARDING);
+		dsa_port_set_state_now(dp, BR_STATE_FORWARDING, false);
 
 	if (dp->pl)
 		phylink_start(dp->pl);
@@ -109,7 +159,7 @@ void dsa_port_disable_rt(struct dsa_port *dp)
 		phylink_stop(dp->pl);
 
 	if (!dp->bridge_dev)
-		dsa_port_set_state_now(dp, BR_STATE_DISABLED);
+		dsa_port_set_state_now(dp, BR_STATE_DISABLED, false);
 
 	if (ds->ops->port_disable)
 		ds->ops->port_disable(ds, port);
@@ -167,8 +217,8 @@ static void dsa_port_clear_brport_flags(struct dsa_port *dp)
 	}
 }
 
-static int dsa_port_switchdev_sync(struct dsa_port *dp,
-				   struct netlink_ext_ack *extack)
+static int dsa_port_switchdev_sync_attrs(struct dsa_port *dp,
+					 struct netlink_ext_ack *extack)
 {
 	struct net_device *brport_dev = dsa_port_to_bridge_port(dp);
 	struct net_device *br = dp->bridge_dev;
@@ -178,7 +228,7 @@ static int dsa_port_switchdev_sync(struct dsa_port *dp,
 	if (err)
 		return err;
 
-	err = dsa_port_set_state(dp, br_port_get_stp_state(brport_dev));
+	err = dsa_port_set_state(dp, br_port_get_stp_state(brport_dev), false);
 	if (err && err != -EOPNOTSUPP)
 		return err;
 
@@ -186,67 +236,10 @@ static int dsa_port_switchdev_sync(struct dsa_port *dp,
 	if (err && err != -EOPNOTSUPP)
 		return err;
 
-	err = dsa_port_mrouter(dp->cpu_dp, br_multicast_router(br), extack);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
 	err = dsa_port_ageing_time(dp, br_get_ageing_time(br));
 	if (err && err != -EOPNOTSUPP)
 		return err;
 
-	err = br_mdb_replay(br, brport_dev, dp, true,
-			    &dsa_slave_switchdev_blocking_notifier, extack);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
-	/* Forwarding and termination FDB entries on the port */
-	err = br_fdb_replay(br, brport_dev, dp, true,
-			    &dsa_slave_switchdev_notifier);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
-	/* Termination FDB entries on the bridge itself */
-	err = br_fdb_replay(br, br, dp, true, &dsa_slave_switchdev_notifier);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
-	err = br_vlan_replay(br, brport_dev, dp, true,
-			     &dsa_slave_switchdev_blocking_notifier, extack);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
-	return 0;
-}
-
-static int dsa_port_switchdev_unsync_objs(struct dsa_port *dp,
-					  struct net_device *br,
-					  struct netlink_ext_ack *extack)
-{
-	struct net_device *brport_dev = dsa_port_to_bridge_port(dp);
-	int err;
-
-	/* Delete the switchdev objects left on this port */
-	err = br_mdb_replay(br, brport_dev, dp, false,
-			    &dsa_slave_switchdev_blocking_notifier, extack);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
-	/* Forwarding and termination FDB entries on the port */
-	err = br_fdb_replay(br, brport_dev, dp, false,
-			    &dsa_slave_switchdev_notifier);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
-	/* Termination FDB entries on the bridge itself */
-	err = br_fdb_replay(br, br, dp, false, &dsa_slave_switchdev_notifier);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
-	err = br_vlan_replay(br, brport_dev, dp, false,
-			     &dsa_slave_switchdev_blocking_notifier, extack);
-	if (err && err != -EOPNOTSUPP)
-		return err;
-
 	return 0;
 }
 
@@ -268,21 +261,63 @@ static void dsa_port_switchdev_unsync_attrs(struct dsa_port *dp)
 	/* Port left the bridge, put in BR_STATE_DISABLED by the bridge layer,
 	 * so allow it to be in BR_STATE_FORWARDING to be kept functional
 	 */
-	dsa_port_set_state_now(dp, BR_STATE_FORWARDING);
+	dsa_port_set_state_now(dp, BR_STATE_FORWARDING, true);
 
 	/* VLAN filtering is handled by dsa_switch_bridge_leave */
 
-	/* Some drivers treat the notification for having a local multicast
-	 * router by allowing multicast to be flooded to the CPU, so we should
-	 * allow this in standalone mode too.
-	 */
-	dsa_port_mrouter(dp->cpu_dp, true, NULL);
-
 	/* Ageing time may be global to the switch chip, so don't change it
 	 * here because we have no good reason (or value) to change it to.
 	 */
 }
 
+static void dsa_port_bridge_tx_fwd_unoffload(struct dsa_port *dp,
+					     struct net_device *bridge_dev)
+{
+	int bridge_num = dp->bridge_num;
+	struct dsa_switch *ds = dp->ds;
+
+	/* No bridge TX forwarding offload => do nothing */
+	if (!ds->ops->port_bridge_tx_fwd_unoffload || dp->bridge_num == -1)
+		return;
+
+	dp->bridge_num = -1;
+
+	dsa_bridge_num_put(bridge_dev, bridge_num);
+
+	/* Notify the chips only once the offload has been deactivated, so
+	 * that they can update their configuration accordingly.
+	 */
+	ds->ops->port_bridge_tx_fwd_unoffload(ds, dp->index, bridge_dev,
+					      bridge_num);
+}
+
+static bool dsa_port_bridge_tx_fwd_offload(struct dsa_port *dp,
+					   struct net_device *bridge_dev)
+{
+	struct dsa_switch *ds = dp->ds;
+	int bridge_num, err;
+
+	if (!ds->ops->port_bridge_tx_fwd_offload)
+		return false;
+
+	bridge_num = dsa_bridge_num_get(bridge_dev,
+					ds->num_fwd_offloading_bridges);
+	if (bridge_num < 0)
+		return false;
+
+	dp->bridge_num = bridge_num;
+
+	/* Notify the driver */
+	err = ds->ops->port_bridge_tx_fwd_offload(ds, dp->index, bridge_dev,
+						  bridge_num);
+	if (err) {
+		dsa_port_bridge_tx_fwd_unoffload(dp, bridge_dev);
+		return false;
+	}
+
+	return true;
+}
+
 int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br,
 			 struct netlink_ext_ack *extack)
 {
@@ -292,6 +327,9 @@ int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br,
 		.port = dp->index,
 		.br = br,
 	};
+	struct net_device *dev = dp->slave;
+	struct net_device *brport_dev;
+	bool tx_fwd_offload;
 	int err;
 
 	/* Here the interface is already bridged. Reflect the current
@@ -299,16 +337,31 @@ int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br,
 	 */
 	dp->bridge_dev = br;
 
+	brport_dev = dsa_port_to_bridge_port(dp);
+
 	err = dsa_broadcast(DSA_NOTIFIER_BRIDGE_JOIN, &info);
 	if (err)
 		goto out_rollback;
 
-	err = dsa_port_switchdev_sync(dp, extack);
+	tx_fwd_offload = dsa_port_bridge_tx_fwd_offload(dp, br);
+
+	err = switchdev_bridge_port_offload(brport_dev, dev, dp,
+					    &dsa_slave_switchdev_notifier,
+					    &dsa_slave_switchdev_blocking_notifier,
+					    tx_fwd_offload, extack);
 	if (err)
 		goto out_rollback_unbridge;
 
+	err = dsa_port_switchdev_sync_attrs(dp, extack);
+	if (err)
+		goto out_rollback_unoffload;
+
 	return 0;
 
+out_rollback_unoffload:
+	switchdev_bridge_port_unoffload(brport_dev, dp,
+					&dsa_slave_switchdev_notifier,
+					&dsa_slave_switchdev_blocking_notifier);
 out_rollback_unbridge:
 	dsa_broadcast(DSA_NOTIFIER_BRIDGE_LEAVE, &info);
 out_rollback:
@@ -316,10 +369,17 @@ out_rollback:
 	return err;
 }
 
-int dsa_port_pre_bridge_leave(struct dsa_port *dp, struct net_device *br,
-			      struct netlink_ext_ack *extack)
+void dsa_port_pre_bridge_leave(struct dsa_port *dp, struct net_device *br)
 {
-	return dsa_port_switchdev_unsync_objs(dp, br, extack);
+	struct net_device *brport_dev = dsa_port_to_bridge_port(dp);
+
+	/* Don't try to unoffload something that is not offloaded */
+	if (!brport_dev)
+		return;
+
+	switchdev_bridge_port_unoffload(brport_dev, dp,
+					&dsa_slave_switchdev_notifier,
+					&dsa_slave_switchdev_blocking_notifier);
 }
 
 void dsa_port_bridge_leave(struct dsa_port *dp, struct net_device *br)
@@ -337,9 +397,13 @@ void dsa_port_bridge_leave(struct dsa_port *dp, struct net_device *br)
 	 */
 	dp->bridge_dev = NULL;
 
+	dsa_port_bridge_tx_fwd_unoffload(dp, br);
+
 	err = dsa_broadcast(DSA_NOTIFIER_BRIDGE_LEAVE, &info);
 	if (err)
-		pr_err("DSA: failed to notify DSA_NOTIFIER_BRIDGE_LEAVE\n");
+		dev_err(dp->ds->dev,
+			"port %d failed to notify DSA_NOTIFIER_BRIDGE_LEAVE: %pe\n",
+			dp->index, ERR_PTR(err));
 
 	dsa_port_switchdev_unsync_attrs(dp);
 }
@@ -409,13 +473,10 @@ err_lag_join:
 	return err;
 }
 
-int dsa_port_pre_lag_leave(struct dsa_port *dp, struct net_device *lag,
-			   struct netlink_ext_ack *extack)
+void dsa_port_pre_lag_leave(struct dsa_port *dp, struct net_device *lag)
 {
 	if (dp->bridge_dev)
-		return dsa_port_pre_bridge_leave(dp, dp->bridge_dev, extack);
-
-	return 0;
+		dsa_port_pre_bridge_leave(dp, dp->bridge_dev);
 }
 
 void dsa_port_lag_leave(struct dsa_port *dp, struct net_device *lag)
@@ -441,8 +502,9 @@ void dsa_port_lag_leave(struct dsa_port *dp, struct net_device *lag)
 
 	err = dsa_port_notify(dp, DSA_NOTIFIER_LAG_LEAVE, &info);
 	if (err)
-		pr_err("DSA: failed to notify DSA_NOTIFIER_LAG_LEAVE: %d\n",
-		       err);
+		dev_err(dp->ds->dev,
+			"port %d failed to notify DSA_NOTIFIER_LAG_LEAVE: %pe\n",
+			dp->index, ERR_PTR(err));
 
 	dsa_lag_unmap(dp->ds->dst, lag);
 }
@@ -518,6 +580,7 @@ static bool dsa_port_can_apply_vlan_filtering(struct dsa_port *dp,
 int dsa_port_vlan_filtering(struct dsa_port *dp, bool vlan_filtering,
 			    struct netlink_ext_ack *extack)
 {
+	bool old_vlan_filtering = dsa_port_is_vlan_filtering(dp);
 	struct dsa_switch *ds = dp->ds;
 	bool apply;
 	int err;
@@ -543,12 +606,49 @@ int dsa_port_vlan_filtering(struct dsa_port *dp, bool vlan_filtering,
 	if (err)
 		return err;
 
-	if (ds->vlan_filtering_is_global)
+	if (ds->vlan_filtering_is_global) {
+		int port;
+
 		ds->vlan_filtering = vlan_filtering;
-	else
+
+		for (port = 0; port < ds->num_ports; port++) {
+			struct net_device *slave;
+
+			if (!dsa_is_user_port(ds, port))
+				continue;
+
+			/* We might be called in the unbind path, so not
+			 * all slave devices might still be registered.
+			 */
+			slave = dsa_to_port(ds, port)->slave;
+			if (!slave)
+				continue;
+
+			err = dsa_slave_manage_vlan_filtering(slave,
+							      vlan_filtering);
+			if (err)
+				goto restore;
+		}
+	} else {
 		dp->vlan_filtering = vlan_filtering;
 
+		err = dsa_slave_manage_vlan_filtering(dp->slave,
+						      vlan_filtering);
+		if (err)
+			goto restore;
+	}
+
 	return 0;
+
+restore:
+	ds->ops->port_vlan_filtering(ds, dp->index, old_vlan_filtering, NULL);
+
+	if (ds->vlan_filtering_is_global)
+		ds->vlan_filtering = old_vlan_filtering;
+	else
+		dp->vlan_filtering = old_vlan_filtering;
+
+	return err;
 }
 
 /* This enforces legacy behavior for switch drivers which assume they can't
@@ -595,27 +695,35 @@ int dsa_port_pre_bridge_flags(const struct dsa_port *dp,
 	return ds->ops->port_pre_bridge_flags(ds, dp->index, flags, extack);
 }
 
-int dsa_port_bridge_flags(const struct dsa_port *dp,
+int dsa_port_bridge_flags(struct dsa_port *dp,
 			  struct switchdev_brport_flags flags,
 			  struct netlink_ext_ack *extack)
 {
 	struct dsa_switch *ds = dp->ds;
+	int err;
 
 	if (!ds->ops->port_bridge_flags)
 		return -EOPNOTSUPP;
 
-	return ds->ops->port_bridge_flags(ds, dp->index, flags, extack);
-}
+	err = ds->ops->port_bridge_flags(ds, dp->index, flags, extack);
+	if (err)
+		return err;
 
-int dsa_port_mrouter(struct dsa_port *dp, bool mrouter,
-		     struct netlink_ext_ack *extack)
-{
-	struct dsa_switch *ds = dp->ds;
+	if (flags.mask & BR_LEARNING) {
+		bool learning = flags.val & BR_LEARNING;
 
-	if (!ds->ops->port_set_mrouter)
-		return -EOPNOTSUPP;
+		if (learning == dp->learning)
+			return 0;
+
+		if ((dp->learning && !learning) &&
+		    (dp->stp_state == BR_STATE_LEARNING ||
+		     dp->stp_state == BR_STATE_FORWARDING))
+			dsa_port_fast_age(dp);
+
+		dp->learning = learning;
+	}
 
-	return ds->ops->port_set_mrouter(ds, dp->index, mrouter, extack);
+	return 0;
 }
 
 int dsa_port_mtu_change(struct dsa_port *dp, int new_mtu,
@@ -844,7 +952,6 @@ int dsa_port_mrp_del_ring_role(const struct dsa_port *dp,
 void dsa_port_set_tag_protocol(struct dsa_port *cpu_dp,
 			       const struct dsa_device_ops *tag_ops)
 {
-	cpu_dp->filter = tag_ops->filter;
 	cpu_dp->rcv = tag_ops->rcv;
 	cpu_dp->tag_ops = tag_ops;
 }
@@ -1215,5 +1322,42 @@ void dsa_port_hsr_leave(struct dsa_port *dp, struct net_device *hsr)
 
 	err = dsa_port_notify(dp, DSA_NOTIFIER_HSR_LEAVE, &info);
 	if (err)
-		pr_err("DSA: failed to notify DSA_NOTIFIER_HSR_LEAVE\n");
+		dev_err(dp->ds->dev,
+			"port %d failed to notify DSA_NOTIFIER_HSR_LEAVE: %pe\n",
+			dp->index, ERR_PTR(err));
+}
+
+int dsa_port_tag_8021q_vlan_add(struct dsa_port *dp, u16 vid, bool broadcast)
+{
+	struct dsa_notifier_tag_8021q_vlan_info info = {
+		.tree_index = dp->ds->dst->index,
+		.sw_index = dp->ds->index,
+		.port = dp->index,
+		.vid = vid,
+	};
+
+	if (broadcast)
+		return dsa_broadcast(DSA_NOTIFIER_TAG_8021Q_VLAN_ADD, &info);
+
+	return dsa_port_notify(dp, DSA_NOTIFIER_TAG_8021Q_VLAN_ADD, &info);
+}
+
+void dsa_port_tag_8021q_vlan_del(struct dsa_port *dp, u16 vid, bool broadcast)
+{
+	struct dsa_notifier_tag_8021q_vlan_info info = {
+		.tree_index = dp->ds->dst->index,
+		.sw_index = dp->ds->index,
+		.port = dp->index,
+		.vid = vid,
+	};
+	int err;
+
+	if (broadcast)
+		err = dsa_broadcast(DSA_NOTIFIER_TAG_8021Q_VLAN_DEL, &info);
+	else
+		err = dsa_port_notify(dp, DSA_NOTIFIER_TAG_8021Q_VLAN_DEL, &info);
+	if (err)
+		dev_err(dp->ds->dev,
+			"port %d failed to notify tag_8021q VLAN %d deletion: %pe\n",
+			dp->index, vid, ERR_PTR(err));
 }
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index ffbba1e71551..662ff531d4e2 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -286,7 +286,7 @@ static int dsa_slave_port_attr_set(struct net_device *dev, const void *ctx,
 		if (!dsa_port_offloads_bridge_port(dp, attr->orig_dev))
 			return -EOPNOTSUPP;
 
-		ret = dsa_port_set_state(dp, attr->u.stp_state);
+		ret = dsa_port_set_state(dp, attr->u.stp_state, true);
 		break;
 	case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING:
 		if (!dsa_port_offloads_bridge(dp, attr->orig_dev))
@@ -314,12 +314,6 @@ static int dsa_slave_port_attr_set(struct net_device *dev, const void *ctx,
 
 		ret = dsa_port_bridge_flags(dp, attr->u.brport_flags, extack);
 		break;
-	case SWITCHDEV_ATTR_ID_BRIDGE_MROUTER:
-		if (!dsa_port_offloads_bridge(dp, attr->orig_dev))
-			return -EOPNOTSUPP;
-
-		ret = dsa_port_mrouter(dp->cpu_dp, attr->u.mrouter, extack);
-		break;
 	default:
 		ret = -EOPNOTSUPP;
 		break;
@@ -1415,6 +1409,76 @@ static int dsa_slave_vlan_rx_kill_vid(struct net_device *dev, __be16 proto,
 	return 0;
 }
 
+static int dsa_slave_restore_vlan(struct net_device *vdev, int vid, void *arg)
+{
+	__be16 proto = vdev ? vlan_dev_vlan_proto(vdev) : htons(ETH_P_8021Q);
+
+	return dsa_slave_vlan_rx_add_vid(arg, proto, vid);
+}
+
+static int dsa_slave_clear_vlan(struct net_device *vdev, int vid, void *arg)
+{
+	__be16 proto = vdev ? vlan_dev_vlan_proto(vdev) : htons(ETH_P_8021Q);
+
+	return dsa_slave_vlan_rx_kill_vid(arg, proto, vid);
+}
+
+/* Keep the VLAN RX filtering list in sync with the hardware only if VLAN
+ * filtering is enabled. The baseline is that only ports that offload a
+ * VLAN-aware bridge are VLAN-aware, and standalone ports are VLAN-unaware,
+ * but there are exceptions for quirky hardware.
+ *
+ * If ds->vlan_filtering_is_global = true, then standalone ports which share
+ * the same switch with other ports that offload a VLAN-aware bridge are also
+ * inevitably VLAN-aware.
+ *
+ * To summarize, a DSA switch port offloads:
+ *
+ * - If standalone (this includes software bridge, software LAG):
+ *     - if ds->needs_standalone_vlan_filtering = true, OR if
+ *       (ds->vlan_filtering_is_global = true AND there are bridges spanning
+ *       this switch chip which have vlan_filtering=1)
+ *         - the 8021q upper VLANs
+ *     - else (standalone VLAN filtering is not needed, VLAN filtering is not
+ *       global, or it is, but no port is under a VLAN-aware bridge):
+ *         - no VLAN (any 8021q upper is a software VLAN)
+ *
+ * - If under a vlan_filtering=0 bridge which it offload:
+ *     - if ds->configure_vlan_while_not_filtering = true (default):
+ *         - the bridge VLANs. These VLANs are committed to hardware but inactive.
+ *     - else (deprecated):
+ *         - no VLAN. The bridge VLANs are not restored when VLAN awareness is
+ *           enabled, so this behavior is broken and discouraged.
+ *
+ * - If under a vlan_filtering=1 bridge which it offload:
+ *     - the bridge VLANs
+ *     - the 8021q upper VLANs
+ */
+int dsa_slave_manage_vlan_filtering(struct net_device *slave,
+				    bool vlan_filtering)
+{
+	int err;
+
+	if (vlan_filtering) {
+		slave->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+
+		err = vlan_for_each(slave, dsa_slave_restore_vlan, slave);
+		if (err) {
+			vlan_for_each(slave, dsa_slave_clear_vlan, slave);
+			slave->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+			return err;
+		}
+	} else {
+		err = vlan_for_each(slave, dsa_slave_clear_vlan, slave);
+		if (err)
+			return err;
+
+		slave->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+	}
+
+	return 0;
+}
+
 struct dsa_hw_port {
 	struct list_head list;
 	struct net_device *dev;
@@ -1687,7 +1751,7 @@ static const struct net_device_ops dsa_slave_netdev_ops = {
 	.ndo_set_rx_mode	= dsa_slave_set_rx_mode,
 	.ndo_set_mac_address	= dsa_slave_set_mac_address,
 	.ndo_fdb_dump		= dsa_slave_fdb_dump,
-	.ndo_do_ioctl		= dsa_slave_ioctl,
+	.ndo_eth_ioctl		= dsa_slave_ioctl,
 	.ndo_get_iflink		= dsa_slave_get_iflink,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_netpoll_setup	= dsa_slave_netpoll_setup,
@@ -1808,6 +1872,7 @@ void dsa_slave_setup_tagger(struct net_device *slave)
 	struct dsa_slave_priv *p = netdev_priv(slave);
 	const struct dsa_port *cpu_dp = dp->cpu_dp;
 	struct net_device *master = cpu_dp->master;
+	const struct dsa_switch *ds = dp->ds;
 
 	slave->needed_headroom = cpu_dp->tag_ops->needed_headroom;
 	slave->needed_tailroom = cpu_dp->tag_ops->needed_tailroom;
@@ -1819,6 +1884,14 @@ void dsa_slave_setup_tagger(struct net_device *slave)
 	slave->needed_tailroom += master->needed_tailroom;
 
 	p->xmit = cpu_dp->tag_ops->xmit;
+
+	slave->features = master->vlan_features | NETIF_F_HW_TC;
+	slave->hw_features |= NETIF_F_HW_TC;
+	slave->features |= NETIF_F_LLTX;
+	if (slave->needed_tailroom)
+		slave->features &= ~(NETIF_F_SG | NETIF_F_FRAGLIST);
+	if (ds->needs_standalone_vlan_filtering)
+		slave->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
 }
 
 static struct lock_class_key dsa_slave_netdev_xmit_lock_key;
@@ -1881,11 +1954,6 @@ int dsa_slave_create(struct dsa_port *port)
 	if (slave_dev == NULL)
 		return -ENOMEM;
 
-	slave_dev->features = master->vlan_features | NETIF_F_HW_TC;
-	if (ds->ops->port_vlan_add && ds->ops->port_vlan_del)
-		slave_dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
-	slave_dev->hw_features |= NETIF_F_HW_TC;
-	slave_dev->features |= NETIF_F_LLTX;
 	slave_dev->ethtool_ops = &dsa_slave_ethtool_ops;
 	if (!is_zero_ether_addr(port->mac))
 		ether_addr_copy(slave_dev->dev_addr, port->mac);
@@ -2011,6 +2079,11 @@ static int dsa_slave_changeupper(struct net_device *dev,
 			err = dsa_port_bridge_join(dp, info->upper_dev, extack);
 			if (!err)
 				dsa_bridge_mtu_normalization(dp);
+			if (err == -EOPNOTSUPP) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Offloading not supported");
+				err = 0;
+			}
 			err = notifier_from_errno(err);
 		} else {
 			dsa_port_bridge_leave(dp, info->upper_dev);
@@ -2052,20 +2125,16 @@ static int dsa_slave_prechangeupper(struct net_device *dev,
 				    struct netdev_notifier_changeupper_info *info)
 {
 	struct dsa_port *dp = dsa_slave_to_port(dev);
-	struct netlink_ext_ack *extack;
-	int err = 0;
-
-	extack = netdev_notifier_info_to_extack(&info->info);
 
 	if (netif_is_bridge_master(info->upper_dev) && !info->linking)
-		err = dsa_port_pre_bridge_leave(dp, info->upper_dev, extack);
+		dsa_port_pre_bridge_leave(dp, info->upper_dev);
 	else if (netif_is_lag_master(info->upper_dev) && !info->linking)
-		err = dsa_port_pre_lag_leave(dp, info->upper_dev, extack);
+		dsa_port_pre_lag_leave(dp, info->upper_dev);
 	/* dsa_port_pre_hsr_leave is not yet necessary since hsr cannot be
 	 * meaningfully enslaved to a bridge yet
 	 */
 
-	return notifier_from_errno(err);
+	return NOTIFY_DONE;
 }
 
 static int
@@ -2287,8 +2356,8 @@ static int dsa_slave_netdevice_event(struct notifier_block *nb,
 static void
 dsa_fdb_offload_notify(struct dsa_switchdev_event_work *switchdev_work)
 {
+	struct switchdev_notifier_fdb_info info = {};
 	struct dsa_switch *ds = switchdev_work->ds;
-	struct switchdev_notifier_fdb_info info;
 	struct dsa_port *dp;
 
 	if (!dsa_is_user_port(ds, switchdev_work->port))
@@ -2353,26 +2422,98 @@ static void dsa_slave_switchdev_event_work(struct work_struct *work)
 	kfree(switchdev_work);
 }
 
-static int dsa_lower_dev_walk(struct net_device *lower_dev,
-			      struct netdev_nested_priv *priv)
+static bool dsa_foreign_dev_check(const struct net_device *dev,
+				  const struct net_device *foreign_dev)
 {
-	if (dsa_slave_dev_check(lower_dev)) {
-		priv->data = (void *)netdev_priv(lower_dev);
-		return 1;
-	}
+	const struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct dsa_switch_tree *dst = dp->ds->dst;
 
-	return 0;
+	if (netif_is_bridge_master(foreign_dev))
+		return !dsa_tree_offloads_bridge(dst, foreign_dev);
+
+	if (netif_is_bridge_port(foreign_dev))
+		return !dsa_tree_offloads_bridge_port(dst, foreign_dev);
+
+	/* Everything else is foreign */
+	return true;
 }
 
-static struct dsa_slave_priv *dsa_slave_dev_lower_find(struct net_device *dev)
+static int dsa_slave_fdb_event(struct net_device *dev,
+			       const struct net_device *orig_dev,
+			       const void *ctx,
+			       const struct switchdev_notifier_fdb_info *fdb_info,
+			       unsigned long event)
 {
-	struct netdev_nested_priv priv = {
-		.data = NULL,
-	};
+	struct dsa_switchdev_event_work *switchdev_work;
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	bool host_addr = fdb_info->is_local;
+	struct dsa_switch *ds = dp->ds;
 
-	netdev_walk_all_lower_dev_rcu(dev, dsa_lower_dev_walk, &priv);
+	if (ctx && ctx != dp)
+		return 0;
+
+	if (!ds->ops->port_fdb_add || !ds->ops->port_fdb_del)
+		return -EOPNOTSUPP;
+
+	if (dsa_slave_dev_check(orig_dev) &&
+	    switchdev_fdb_is_dynamically_learned(fdb_info))
+		return 0;
+
+	/* FDB entries learned by the software bridge should be installed as
+	 * host addresses only if the driver requests assisted learning.
+	 */
+	if (switchdev_fdb_is_dynamically_learned(fdb_info) &&
+	    !ds->assisted_learning_on_cpu_port)
+		return 0;
+
+	/* Also treat FDB entries on foreign interfaces bridged with us as host
+	 * addresses.
+	 */
+	if (dsa_foreign_dev_check(dev, orig_dev))
+		host_addr = true;
+
+	switchdev_work = kzalloc(sizeof(*switchdev_work), GFP_ATOMIC);
+	if (!switchdev_work)
+		return -ENOMEM;
+
+	netdev_dbg(dev, "%s FDB entry towards %s, addr %pM vid %d%s\n",
+		   event == SWITCHDEV_FDB_ADD_TO_DEVICE ? "Adding" : "Deleting",
+		   orig_dev->name, fdb_info->addr, fdb_info->vid,
+		   host_addr ? " as host address" : "");
+
+	INIT_WORK(&switchdev_work->work, dsa_slave_switchdev_event_work);
+	switchdev_work->ds = ds;
+	switchdev_work->port = dp->index;
+	switchdev_work->event = event;
+	switchdev_work->dev = dev;
+
+	ether_addr_copy(switchdev_work->addr, fdb_info->addr);
+	switchdev_work->vid = fdb_info->vid;
+	switchdev_work->host_addr = host_addr;
+
+	/* Hold a reference for dsa_fdb_offload_notify */
+	dev_hold(dev);
+	dsa_schedule_work(&switchdev_work->work);
+
+	return 0;
+}
+
+static int
+dsa_slave_fdb_add_to_device(struct net_device *dev,
+			    const struct net_device *orig_dev, const void *ctx,
+			    const struct switchdev_notifier_fdb_info *fdb_info)
+{
+	return dsa_slave_fdb_event(dev, orig_dev, ctx, fdb_info,
+				   SWITCHDEV_FDB_ADD_TO_DEVICE);
+}
 
-	return (struct dsa_slave_priv *)priv.data;
+static int
+dsa_slave_fdb_del_to_device(struct net_device *dev,
+			    const struct net_device *orig_dev, const void *ctx,
+			    const struct switchdev_notifier_fdb_info *fdb_info)
+{
+	return dsa_slave_fdb_event(dev, orig_dev, ctx, fdb_info,
+				   SWITCHDEV_FDB_DEL_TO_DEVICE);
 }
 
 /* Called under rcu_read_lock() */
@@ -2380,10 +2521,6 @@ static int dsa_slave_switchdev_event(struct notifier_block *unused,
 				     unsigned long event, void *ptr)
 {
 	struct net_device *dev = switchdev_notifier_info_to_dev(ptr);
-	const struct switchdev_notifier_fdb_info *fdb_info;
-	struct dsa_switchdev_event_work *switchdev_work;
-	bool host_addr = false;
-	struct dsa_port *dp;
 	int err;
 
 	switch (event) {
@@ -2393,92 +2530,19 @@ static int dsa_slave_switchdev_event(struct notifier_block *unused,
 						     dsa_slave_port_attr_set);
 		return notifier_from_errno(err);
 	case SWITCHDEV_FDB_ADD_TO_DEVICE:
+		err = switchdev_handle_fdb_add_to_device(dev, ptr,
+							 dsa_slave_dev_check,
+							 dsa_foreign_dev_check,
+							 dsa_slave_fdb_add_to_device,
+							 NULL);
+		return notifier_from_errno(err);
 	case SWITCHDEV_FDB_DEL_TO_DEVICE:
-		fdb_info = ptr;
-
-		if (dsa_slave_dev_check(dev)) {
-			dp = dsa_slave_to_port(dev);
-
-			if (fdb_info->is_local)
-				host_addr = true;
-			else if (!fdb_info->added_by_user)
-				return NOTIFY_OK;
-		} else {
-			/* Snoop addresses added to foreign interfaces
-			 * bridged with us, or the bridge
-			 * itself. Dynamically learned addresses can
-			 * also be added for switches that don't
-			 * automatically learn SA from CPU-injected
-			 * traffic.
-			 */
-			struct net_device *br_dev;
-			struct dsa_slave_priv *p;
-
-			if (netif_is_bridge_master(dev))
-				br_dev = dev;
-			else
-				br_dev = netdev_master_upper_dev_get_rcu(dev);
-
-			if (!br_dev)
-				return NOTIFY_DONE;
-
-			if (!netif_is_bridge_master(br_dev))
-				return NOTIFY_DONE;
-
-			p = dsa_slave_dev_lower_find(br_dev);
-			if (!p)
-				return NOTIFY_DONE;
-
-			dp = p->dp;
-			host_addr = fdb_info->is_local;
-
-			/* FDB entries learned by the software bridge should
-			 * be installed as host addresses only if the driver
-			 * requests assisted learning.
-			 * On the other hand, FDB entries for local termination
-			 * should always be installed.
-			 */
-			if (!fdb_info->added_by_user && !fdb_info->is_local &&
-			    !dp->ds->assisted_learning_on_cpu_port)
-				return NOTIFY_DONE;
-
-			/* When the bridge learns an address on an offloaded
-			 * LAG we don't want to send traffic to the CPU, the
-			 * other ports bridged with the LAG should be able to
-			 * autonomously forward towards it.
-			 * On the other hand, if the address is local
-			 * (therefore not learned) then we want to trap it to
-			 * the CPU regardless of whether the interface it
-			 * belongs to is offloaded or not.
-			 */
-			if (dsa_tree_offloads_bridge_port(dp->ds->dst, dev) &&
-			    !fdb_info->is_local)
-				return NOTIFY_DONE;
-		}
-
-		if (!dp->ds->ops->port_fdb_add || !dp->ds->ops->port_fdb_del)
-			return NOTIFY_DONE;
-
-		switchdev_work = kzalloc(sizeof(*switchdev_work), GFP_ATOMIC);
-		if (!switchdev_work)
-			return NOTIFY_BAD;
-
-		INIT_WORK(&switchdev_work->work,
-			  dsa_slave_switchdev_event_work);
-		switchdev_work->ds = dp->ds;
-		switchdev_work->port = dp->index;
-		switchdev_work->event = event;
-		switchdev_work->dev = dev;
-
-		ether_addr_copy(switchdev_work->addr,
-				fdb_info->addr);
-		switchdev_work->vid = fdb_info->vid;
-		switchdev_work->host_addr = host_addr;
-
-		/* Hold a reference for dsa_fdb_offload_notify */
-		dev_hold(dev);
-		dsa_schedule_work(&switchdev_work->work);
-		break;
+		err = switchdev_handle_fdb_del_to_device(dev, ptr,
+							 dsa_slave_dev_check,
+							 dsa_foreign_dev_check,
+							 dsa_slave_fdb_del_to_device,
+							 NULL);
+		return notifier_from_errno(err);
 	default:
 		return NOTIFY_DONE;
 	}
diff --git a/net/dsa/switch.c b/net/dsa/switch.c
index af71b8638098..1c797ec8e2c2 100644
--- a/net/dsa/switch.c
+++ b/net/dsa/switch.c
@@ -90,38 +90,57 @@ static int dsa_switch_bridge_join(struct dsa_switch *ds,
 				  struct dsa_notifier_bridge_info *info)
 {
 	struct dsa_switch_tree *dst = ds->dst;
+	int err;
 
-	if (dst->index == info->tree_index && ds->index == info->sw_index &&
-	    ds->ops->port_bridge_join)
-		return ds->ops->port_bridge_join(ds, info->port, info->br);
+	if (dst->index == info->tree_index && ds->index == info->sw_index) {
+		if (!ds->ops->port_bridge_join)
+			return -EOPNOTSUPP;
+
+		err = ds->ops->port_bridge_join(ds, info->port, info->br);
+		if (err)
+			return err;
+	}
 
 	if ((dst->index != info->tree_index || ds->index != info->sw_index) &&
-	    ds->ops->crosschip_bridge_join)
-		return ds->ops->crosschip_bridge_join(ds, info->tree_index,
-						      info->sw_index,
-						      info->port, info->br);
+	    ds->ops->crosschip_bridge_join) {
+		err = ds->ops->crosschip_bridge_join(ds, info->tree_index,
+						     info->sw_index,
+						     info->port, info->br);
+		if (err)
+			return err;
+	}
 
-	return 0;
+	return dsa_tag_8021q_bridge_join(ds, info);
 }
 
 static int dsa_switch_bridge_leave(struct dsa_switch *ds,
 				   struct dsa_notifier_bridge_info *info)
 {
-	bool unset_vlan_filtering = br_vlan_enabled(info->br);
 	struct dsa_switch_tree *dst = ds->dst;
 	struct netlink_ext_ack extack = {0};
+	bool change_vlan_filtering = false;
+	bool vlan_filtering;
 	int err, port;
 
 	if (dst->index == info->tree_index && ds->index == info->sw_index &&
-	    ds->ops->port_bridge_join)
+	    ds->ops->port_bridge_leave)
 		ds->ops->port_bridge_leave(ds, info->port, info->br);
 
 	if ((dst->index != info->tree_index || ds->index != info->sw_index) &&
-	    ds->ops->crosschip_bridge_join)
+	    ds->ops->crosschip_bridge_leave)
 		ds->ops->crosschip_bridge_leave(ds, info->tree_index,
 						info->sw_index, info->port,
 						info->br);
 
+	if (ds->needs_standalone_vlan_filtering && !br_vlan_enabled(info->br)) {
+		change_vlan_filtering = true;
+		vlan_filtering = true;
+	} else if (!ds->needs_standalone_vlan_filtering &&
+		   br_vlan_enabled(info->br)) {
+		change_vlan_filtering = true;
+		vlan_filtering = false;
+	}
+
 	/* If the bridge was vlan_filtering, the bridge core doesn't trigger an
 	 * event for changing vlan_filtering setting upon slave ports leaving
 	 * it. That is a good thing, because that lets us handle it and also
@@ -130,28 +149,30 @@ static int dsa_switch_bridge_leave(struct dsa_switch *ds,
 	 * vlan_filtering callback is only when the last port leaves the last
 	 * VLAN-aware bridge.
 	 */
-	if (unset_vlan_filtering && ds->vlan_filtering_is_global) {
+	if (change_vlan_filtering && ds->vlan_filtering_is_global) {
 		for (port = 0; port < ds->num_ports; port++) {
 			struct net_device *bridge_dev;
 
 			bridge_dev = dsa_to_port(ds, port)->bridge_dev;
 
 			if (bridge_dev && br_vlan_enabled(bridge_dev)) {
-				unset_vlan_filtering = false;
+				change_vlan_filtering = false;
 				break;
 			}
 		}
 	}
-	if (unset_vlan_filtering) {
+
+	if (change_vlan_filtering) {
 		err = dsa_port_vlan_filtering(dsa_to_port(ds, info->port),
-					      false, &extack);
+					      vlan_filtering, &extack);
 		if (extack._msg)
 			dev_err(ds->dev, "port %d: %s\n", info->port,
 				extack._msg);
 		if (err && err != EOPNOTSUPP)
 			return err;
 	}
-	return 0;
+
+	return dsa_tag_8021q_bridge_leave(ds, info);
 }
 
 /* Matches for all upstream-facing ports (the CPU port and all upstream-facing
@@ -427,7 +448,7 @@ static int dsa_switch_lag_join(struct dsa_switch *ds,
 						   info->port, info->lag,
 						   info->info);
 
-	return 0;
+	return -EOPNOTSUPP;
 }
 
 static int dsa_switch_lag_leave(struct dsa_switch *ds,
@@ -440,7 +461,7 @@ static int dsa_switch_lag_leave(struct dsa_switch *ds,
 		return ds->ops->crosschip_lag_leave(ds, info->sw_index,
 						    info->port, info->lag);
 
-	return 0;
+	return -EOPNOTSUPP;
 }
 
 static int dsa_switch_mdb_add(struct dsa_switch *ds,
@@ -726,6 +747,12 @@ static int dsa_switch_event(struct notifier_block *nb,
 	case DSA_NOTIFIER_MRP_DEL_RING_ROLE:
 		err = dsa_switch_mrp_del_ring_role(ds, info);
 		break;
+	case DSA_NOTIFIER_TAG_8021Q_VLAN_ADD:
+		err = dsa_switch_tag_8021q_vlan_add(ds, info);
+		break;
+	case DSA_NOTIFIER_TAG_8021Q_VLAN_DEL:
+		err = dsa_switch_tag_8021q_vlan_del(ds, info);
+		break;
 	default:
 		err = -EOPNOTSUPP;
 		break;
diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c
index 4aa29f90ecea..f8f7b7c34e7d 100644
--- a/net/dsa/tag_8021q.c
+++ b/net/dsa/tag_8021q.c
@@ -17,7 +17,7 @@
  *
  * | 11  | 10  |  9  |  8  |  7  |  6  |  5  |  4  |  3  |  2  |  1  |  0  |
  * +-----------+-----+-----------------+-----------+-----------------------+
- * |    DIR    | SVL |    SWITCH_ID    |  SUBVLAN  |          PORT         |
+ * |    DIR    | VBID|    SWITCH_ID    |   VBID    |          PORT         |
  * +-----------+-----+-----------------+-----------+-----------------------+
  *
  * DIR - VID[11:10]:
@@ -27,24 +27,14 @@
  *	These values make the special VIDs of 0, 1 and 4095 to be left
  *	unused by this coding scheme.
  *
- * SVL/SUBVLAN - { VID[9], VID[5:4] }:
- *	Sub-VLAN encoding. Valid only when DIR indicates an RX VLAN.
- *	* 0 (0b000): Field does not encode a sub-VLAN, either because
- *	received traffic is untagged, PVID-tagged or because a second
- *	VLAN tag is present after this tag and not inside of it.
- *	* 1 (0b001): Received traffic is tagged with a VID value private
- *	to the host. This field encodes the index in the host's lookup
- *	table through which the value of the ingress VLAN ID can be
- *	recovered.
- *	* 2 (0b010): Field encodes a sub-VLAN.
- *	...
- *	* 7 (0b111): Field encodes a sub-VLAN.
- *	When DIR indicates a TX VLAN, SUBVLAN must be transmitted as zero
- *	(by the host) and ignored on receive (by the switch).
- *
  * SWITCH_ID - VID[8:6]:
  *	Index of switch within DSA tree. Must be between 0 and 7.
  *
+ * VBID - { VID[9], VID[5:4] }:
+ *	Virtual bridge ID. If between 1 and 7, packet targets the broadcast
+ *	domain of a bridge. If transmitted as zero, packet targets a single
+ *	port. Field only valid on transmit, must be ignored on receive.
+ *
  * PORT - VID[3:0]:
  *	Index of switch port. Must be between 0 and 15.
  */
@@ -61,23 +51,30 @@
 #define DSA_8021Q_SWITCH_ID(x)		(((x) << DSA_8021Q_SWITCH_ID_SHIFT) & \
 						 DSA_8021Q_SWITCH_ID_MASK)
 
-#define DSA_8021Q_SUBVLAN_HI_SHIFT	9
-#define DSA_8021Q_SUBVLAN_HI_MASK	GENMASK(9, 9)
-#define DSA_8021Q_SUBVLAN_LO_SHIFT	4
-#define DSA_8021Q_SUBVLAN_LO_MASK	GENMASK(5, 4)
-#define DSA_8021Q_SUBVLAN_HI(x)		(((x) & GENMASK(2, 2)) >> 2)
-#define DSA_8021Q_SUBVLAN_LO(x)		((x) & GENMASK(1, 0))
-#define DSA_8021Q_SUBVLAN(x)		\
-		(((DSA_8021Q_SUBVLAN_LO(x) << DSA_8021Q_SUBVLAN_LO_SHIFT) & \
-		  DSA_8021Q_SUBVLAN_LO_MASK) | \
-		 ((DSA_8021Q_SUBVLAN_HI(x) << DSA_8021Q_SUBVLAN_HI_SHIFT) & \
-		  DSA_8021Q_SUBVLAN_HI_MASK))
+#define DSA_8021Q_VBID_HI_SHIFT		9
+#define DSA_8021Q_VBID_HI_MASK		GENMASK(9, 9)
+#define DSA_8021Q_VBID_LO_SHIFT		4
+#define DSA_8021Q_VBID_LO_MASK		GENMASK(5, 4)
+#define DSA_8021Q_VBID_HI(x)		(((x) & GENMASK(2, 2)) >> 2)
+#define DSA_8021Q_VBID_LO(x)		((x) & GENMASK(1, 0))
+#define DSA_8021Q_VBID(x)		\
+		(((DSA_8021Q_VBID_LO(x) << DSA_8021Q_VBID_LO_SHIFT) & \
+		  DSA_8021Q_VBID_LO_MASK) | \
+		 ((DSA_8021Q_VBID_HI(x) << DSA_8021Q_VBID_HI_SHIFT) & \
+		  DSA_8021Q_VBID_HI_MASK))
 
 #define DSA_8021Q_PORT_SHIFT		0
 #define DSA_8021Q_PORT_MASK		GENMASK(3, 0)
 #define DSA_8021Q_PORT(x)		(((x) << DSA_8021Q_PORT_SHIFT) & \
 						 DSA_8021Q_PORT_MASK)
 
+u16 dsa_8021q_bridge_tx_fwd_offload_vid(int bridge_num)
+{
+	/* The VBID value of 0 is reserved for precise TX */
+	return DSA_8021Q_DIR_TX | DSA_8021Q_VBID(bridge_num + 1);
+}
+EXPORT_SYMBOL_GPL(dsa_8021q_bridge_tx_fwd_offload_vid);
+
 /* Returns the VID to be inserted into the frame from xmit for switch steering
  * instructions on egress. Encodes switch ID and port ID.
  */
@@ -98,13 +95,6 @@ u16 dsa_8021q_rx_vid(struct dsa_switch *ds, int port)
 }
 EXPORT_SYMBOL_GPL(dsa_8021q_rx_vid);
 
-u16 dsa_8021q_rx_vid_subvlan(struct dsa_switch *ds, int port, u16 subvlan)
-{
-	return DSA_8021Q_DIR_RX | DSA_8021Q_SWITCH_ID(ds->index) |
-	       DSA_8021Q_PORT(port) | DSA_8021Q_SUBVLAN(subvlan);
-}
-EXPORT_SYMBOL_GPL(dsa_8021q_rx_vid_subvlan);
-
 /* Returns the decoded switch ID from the RX VID. */
 int dsa_8021q_rx_switch_id(u16 vid)
 {
@@ -119,20 +109,6 @@ int dsa_8021q_rx_source_port(u16 vid)
 }
 EXPORT_SYMBOL_GPL(dsa_8021q_rx_source_port);
 
-/* Returns the decoded subvlan from the RX VID. */
-u16 dsa_8021q_rx_subvlan(u16 vid)
-{
-	u16 svl_hi, svl_lo;
-
-	svl_hi = (vid & DSA_8021Q_SUBVLAN_HI_MASK) >>
-		 DSA_8021Q_SUBVLAN_HI_SHIFT;
-	svl_lo = (vid & DSA_8021Q_SUBVLAN_LO_MASK) >>
-		 DSA_8021Q_SUBVLAN_LO_SHIFT;
-
-	return (svl_hi << 2) | svl_lo;
-}
-EXPORT_SYMBOL_GPL(dsa_8021q_rx_subvlan);
-
 bool vid_is_dsa_8021q_rxvlan(u16 vid)
 {
 	return (vid & DSA_8021Q_DIR_MASK) == DSA_8021Q_DIR_RX;
@@ -151,21 +127,152 @@ bool vid_is_dsa_8021q(u16 vid)
 }
 EXPORT_SYMBOL_GPL(vid_is_dsa_8021q);
 
-/* If @enabled is true, installs @vid with @flags into the switch port's HW
- * filter.
- * If @enabled is false, deletes @vid (ignores @flags) from the port. Had the
- * user explicitly configured this @vid through the bridge core, then the @vid
- * is installed again, but this time with the flags from the bridge layer.
- */
-static int dsa_8021q_vid_apply(struct dsa_8021q_context *ctx, int port, u16 vid,
-			       u16 flags, bool enabled)
+static struct dsa_tag_8021q_vlan *
+dsa_tag_8021q_vlan_find(struct dsa_8021q_context *ctx, int port, u16 vid)
 {
-	struct dsa_port *dp = dsa_to_port(ctx->ds, port);
+	struct dsa_tag_8021q_vlan *v;
 
-	if (enabled)
-		return ctx->ops->vlan_add(ctx->ds, dp->index, vid, flags);
+	list_for_each_entry(v, &ctx->vlans, list)
+		if (v->vid == vid && v->port == port)
+			return v;
 
-	return ctx->ops->vlan_del(ctx->ds, dp->index, vid);
+	return NULL;
+}
+
+static int dsa_switch_do_tag_8021q_vlan_add(struct dsa_switch *ds, int port,
+					    u16 vid, u16 flags)
+{
+	struct dsa_8021q_context *ctx = ds->tag_8021q_ctx;
+	struct dsa_port *dp = dsa_to_port(ds, port);
+	struct dsa_tag_8021q_vlan *v;
+	int err;
+
+	/* No need to bother with refcounting for user ports */
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
+		return ds->ops->tag_8021q_vlan_add(ds, port, vid, flags);
+
+	v = dsa_tag_8021q_vlan_find(ctx, port, vid);
+	if (v) {
+		refcount_inc(&v->refcount);
+		return 0;
+	}
+
+	v = kzalloc(sizeof(*v), GFP_KERNEL);
+	if (!v)
+		return -ENOMEM;
+
+	err = ds->ops->tag_8021q_vlan_add(ds, port, vid, flags);
+	if (err) {
+		kfree(v);
+		return err;
+	}
+
+	v->vid = vid;
+	v->port = port;
+	refcount_set(&v->refcount, 1);
+	list_add_tail(&v->list, &ctx->vlans);
+
+	return 0;
+}
+
+static int dsa_switch_do_tag_8021q_vlan_del(struct dsa_switch *ds, int port,
+					    u16 vid)
+{
+	struct dsa_8021q_context *ctx = ds->tag_8021q_ctx;
+	struct dsa_port *dp = dsa_to_port(ds, port);
+	struct dsa_tag_8021q_vlan *v;
+	int err;
+
+	/* No need to bother with refcounting for user ports */
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
+		return ds->ops->tag_8021q_vlan_del(ds, port, vid);
+
+	v = dsa_tag_8021q_vlan_find(ctx, port, vid);
+	if (!v)
+		return -ENOENT;
+
+	if (!refcount_dec_and_test(&v->refcount))
+		return 0;
+
+	err = ds->ops->tag_8021q_vlan_del(ds, port, vid);
+	if (err) {
+		refcount_inc(&v->refcount);
+		return err;
+	}
+
+	list_del(&v->list);
+	kfree(v);
+
+	return 0;
+}
+
+static bool
+dsa_switch_tag_8021q_vlan_match(struct dsa_switch *ds, int port,
+				struct dsa_notifier_tag_8021q_vlan_info *info)
+{
+	if (dsa_is_dsa_port(ds, port) || dsa_is_cpu_port(ds, port))
+		return true;
+
+	if (ds->dst->index == info->tree_index && ds->index == info->sw_index)
+		return port == info->port;
+
+	return false;
+}
+
+int dsa_switch_tag_8021q_vlan_add(struct dsa_switch *ds,
+				  struct dsa_notifier_tag_8021q_vlan_info *info)
+{
+	int port, err;
+
+	/* Since we use dsa_broadcast(), there might be other switches in other
+	 * trees which don't support tag_8021q, so don't return an error.
+	 * Or they might even support tag_8021q but have not registered yet to
+	 * use it (maybe they use another tagger currently).
+	 */
+	if (!ds->ops->tag_8021q_vlan_add || !ds->tag_8021q_ctx)
+		return 0;
+
+	for (port = 0; port < ds->num_ports; port++) {
+		if (dsa_switch_tag_8021q_vlan_match(ds, port, info)) {
+			u16 flags = 0;
+
+			if (dsa_is_user_port(ds, port))
+				flags |= BRIDGE_VLAN_INFO_UNTAGGED;
+
+			if (vid_is_dsa_8021q_rxvlan(info->vid) &&
+			    dsa_8021q_rx_switch_id(info->vid) == ds->index &&
+			    dsa_8021q_rx_source_port(info->vid) == port)
+				flags |= BRIDGE_VLAN_INFO_PVID;
+
+			err = dsa_switch_do_tag_8021q_vlan_add(ds, port,
+							       info->vid,
+							       flags);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+int dsa_switch_tag_8021q_vlan_del(struct dsa_switch *ds,
+				  struct dsa_notifier_tag_8021q_vlan_info *info)
+{
+	int port, err;
+
+	if (!ds->ops->tag_8021q_vlan_del || !ds->tag_8021q_ctx)
+		return 0;
+
+	for (port = 0; port < ds->num_ports; port++) {
+		if (dsa_switch_tag_8021q_vlan_match(ds, port, info)) {
+			err = dsa_switch_do_tag_8021q_vlan_del(ds, port,
+							       info->vid);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
 }
 
 /* RX VLAN tagging (left) and TX VLAN tagging (right) setup shown for a single
@@ -181,12 +288,6 @@ static int dsa_8021q_vid_apply(struct dsa_8021q_context *ctx, int port, u16 vid,
  *    force all switched traffic to pass through the CPU. So we must also make
  *    the other front-panel ports members of this VID we're adding, albeit
  *    we're not making it their PVID (they'll still have their own).
- *    By the way - just because we're installing the same VID in multiple
- *    switch ports doesn't mean that they'll start to talk to one another, even
- *    while not bridged: the final forwarding decision is still an AND between
- *    the L2 forwarding information (which is limiting forwarding in this case)
- *    and the VLAN-based restrictions (of which there are none in this case,
- *    since all ports are members).
  *  - On TX (ingress from CPU and towards network) we are faced with a problem.
  *    If we were to tag traffic (from within DSA) with the port's pvid, all
  *    would be well, assuming the switch ports were standalone. Frames would
@@ -200,9 +301,10 @@ static int dsa_8021q_vid_apply(struct dsa_8021q_context *ctx, int port, u16 vid,
  *    a member of the VID we're tagging the traffic with - the desired one.
  *
  * So at the end, each front-panel port will have one RX VID (also the PVID),
- * the RX VID of all other front-panel ports, and one TX VID. Whereas the CPU
- * port will have the RX and TX VIDs of all front-panel ports, and on top of
- * that, is also tagged-input and tagged-output (VLAN trunk).
+ * the RX VID of all other front-panel ports that are in the same bridge, and
+ * one TX VID. Whereas the CPU port will have the RX and TX VIDs of all
+ * front-panel ports, and on top of that, is also tagged-input and
+ * tagged-output (VLAN trunk).
  *
  *               CPU port                               CPU port
  * +-------------+-----+-------------+    +-------------+-----+-------------+
@@ -220,246 +322,246 @@ static int dsa_8021q_vid_apply(struct dsa_8021q_context *ctx, int port, u16 vid,
  * +-+-----+-+-----+-+-----+-+-----+-+    +-+-----+-+-----+-+-----+-+-----+-+
  *   swp0    swp1    swp2    swp3           swp0    swp1    swp2    swp3
  */
-static int dsa_8021q_setup_port(struct dsa_8021q_context *ctx, int port,
-				bool enabled)
+static bool dsa_tag_8021q_bridge_match(struct dsa_switch *ds, int port,
+				       struct dsa_notifier_bridge_info *info)
+{
+	struct dsa_port *dp = dsa_to_port(ds, port);
+
+	/* Don't match on self */
+	if (ds->dst->index == info->tree_index &&
+	    ds->index == info->sw_index &&
+	    port == info->port)
+		return false;
+
+	if (dsa_port_is_user(dp))
+		return dp->bridge_dev == info->br;
+
+	return false;
+}
+
+int dsa_tag_8021q_bridge_join(struct dsa_switch *ds,
+			      struct dsa_notifier_bridge_info *info)
+{
+	struct dsa_switch *targeted_ds;
+	struct dsa_port *targeted_dp;
+	u16 targeted_rx_vid;
+	int err, port;
+
+	if (!ds->tag_8021q_ctx)
+		return 0;
+
+	targeted_ds = dsa_switch_find(info->tree_index, info->sw_index);
+	targeted_dp = dsa_to_port(targeted_ds, info->port);
+	targeted_rx_vid = dsa_8021q_rx_vid(targeted_ds, info->port);
+
+	for (port = 0; port < ds->num_ports; port++) {
+		struct dsa_port *dp = dsa_to_port(ds, port);
+		u16 rx_vid = dsa_8021q_rx_vid(ds, port);
+
+		if (!dsa_tag_8021q_bridge_match(ds, port, info))
+			continue;
+
+		/* Install the RX VID of the targeted port in our VLAN table */
+		err = dsa_port_tag_8021q_vlan_add(dp, targeted_rx_vid, true);
+		if (err)
+			return err;
+
+		/* Install our RX VID into the targeted port's VLAN table */
+		err = dsa_port_tag_8021q_vlan_add(targeted_dp, rx_vid, true);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+int dsa_tag_8021q_bridge_leave(struct dsa_switch *ds,
+			       struct dsa_notifier_bridge_info *info)
 {
-	int upstream = dsa_upstream_port(ctx->ds, port);
-	u16 rx_vid = dsa_8021q_rx_vid(ctx->ds, port);
-	u16 tx_vid = dsa_8021q_tx_vid(ctx->ds, port);
+	struct dsa_switch *targeted_ds;
+	struct dsa_port *targeted_dp;
+	u16 targeted_rx_vid;
+	int port;
+
+	if (!ds->tag_8021q_ctx)
+		return 0;
+
+	targeted_ds = dsa_switch_find(info->tree_index, info->sw_index);
+	targeted_dp = dsa_to_port(targeted_ds, info->port);
+	targeted_rx_vid = dsa_8021q_rx_vid(targeted_ds, info->port);
+
+	for (port = 0; port < ds->num_ports; port++) {
+		struct dsa_port *dp = dsa_to_port(ds, port);
+		u16 rx_vid = dsa_8021q_rx_vid(ds, port);
+
+		if (!dsa_tag_8021q_bridge_match(ds, port, info))
+			continue;
+
+		/* Remove the RX VID of the targeted port from our VLAN table */
+		dsa_port_tag_8021q_vlan_del(dp, targeted_rx_vid, true);
+
+		/* Remove our RX VID from the targeted port's VLAN table */
+		dsa_port_tag_8021q_vlan_del(targeted_dp, rx_vid, true);
+	}
+
+	return 0;
+}
+
+int dsa_tag_8021q_bridge_tx_fwd_offload(struct dsa_switch *ds, int port,
+					struct net_device *br,
+					int bridge_num)
+{
+	u16 tx_vid = dsa_8021q_bridge_tx_fwd_offload_vid(bridge_num);
+
+	return dsa_port_tag_8021q_vlan_add(dsa_to_port(ds, port), tx_vid,
+					   true);
+}
+EXPORT_SYMBOL_GPL(dsa_tag_8021q_bridge_tx_fwd_offload);
+
+void dsa_tag_8021q_bridge_tx_fwd_unoffload(struct dsa_switch *ds, int port,
+					   struct net_device *br,
+					   int bridge_num)
+{
+	u16 tx_vid = dsa_8021q_bridge_tx_fwd_offload_vid(bridge_num);
+
+	dsa_port_tag_8021q_vlan_del(dsa_to_port(ds, port), tx_vid, true);
+}
+EXPORT_SYMBOL_GPL(dsa_tag_8021q_bridge_tx_fwd_unoffload);
+
+/* Set up a port's tag_8021q RX and TX VLAN for standalone mode operation */
+static int dsa_tag_8021q_port_setup(struct dsa_switch *ds, int port)
+{
+	struct dsa_8021q_context *ctx = ds->tag_8021q_ctx;
+	struct dsa_port *dp = dsa_to_port(ds, port);
+	u16 rx_vid = dsa_8021q_rx_vid(ds, port);
+	u16 tx_vid = dsa_8021q_tx_vid(ds, port);
 	struct net_device *master;
-	int i, err, subvlan;
+	int err;
 
 	/* The CPU port is implicitly configured by
 	 * configuring the front-panel ports
 	 */
-	if (!dsa_is_user_port(ctx->ds, port))
+	if (!dsa_port_is_user(dp))
 		return 0;
 
-	master = dsa_to_port(ctx->ds, port)->cpu_dp->master;
+	master = dp->cpu_dp->master;
 
 	/* Add this user port's RX VID to the membership list of all others
 	 * (including itself). This is so that bridging will not be hindered.
 	 * L2 forwarding rules still take precedence when there are no VLAN
 	 * restrictions, so there are no concerns about leaking traffic.
 	 */
-	for (i = 0; i < ctx->ds->num_ports; i++) {
-		u16 flags;
-
-		if (i == upstream)
-			continue;
-		else if (i == port)
-			/* The RX VID is pvid on this port */
-			flags = BRIDGE_VLAN_INFO_UNTAGGED |
-				BRIDGE_VLAN_INFO_PVID;
-		else
-			/* The RX VID is a regular VLAN on all others */
-			flags = BRIDGE_VLAN_INFO_UNTAGGED;
-
-		err = dsa_8021q_vid_apply(ctx, i, rx_vid, flags, enabled);
-		if (err) {
-			dev_err(ctx->ds->dev,
-				"Failed to apply RX VID %d to port %d: %d\n",
-				rx_vid, port, err);
-			return err;
-		}
-	}
-
-	/* CPU port needs to see this port's RX VID
-	 * as tagged egress.
-	 */
-	err = dsa_8021q_vid_apply(ctx, upstream, rx_vid, 0, enabled);
+	err = dsa_port_tag_8021q_vlan_add(dp, rx_vid, false);
 	if (err) {
-		dev_err(ctx->ds->dev,
-			"Failed to apply RX VID %d to port %d: %d\n",
-			rx_vid, port, err);
+		dev_err(ds->dev,
+			"Failed to apply RX VID %d to port %d: %pe\n",
+			rx_vid, port, ERR_PTR(err));
 		return err;
 	}
 
-	/* Add to the master's RX filter not only @rx_vid, but in fact
-	 * the entire subvlan range, just in case this DSA switch might
-	 * want to use sub-VLANs.
-	 */
-	for (subvlan = 0; subvlan < DSA_8021Q_N_SUBVLAN; subvlan++) {
-		u16 vid = dsa_8021q_rx_vid_subvlan(ctx->ds, port, subvlan);
-
-		if (enabled)
-			vlan_vid_add(master, ctx->proto, vid);
-		else
-			vlan_vid_del(master, ctx->proto, vid);
-	}
+	/* Add @rx_vid to the master's RX filter. */
+	vlan_vid_add(master, ctx->proto, rx_vid);
 
 	/* Finally apply the TX VID on this port and on the CPU port */
-	err = dsa_8021q_vid_apply(ctx, port, tx_vid, BRIDGE_VLAN_INFO_UNTAGGED,
-				  enabled);
-	if (err) {
-		dev_err(ctx->ds->dev,
-			"Failed to apply TX VID %d on port %d: %d\n",
-			tx_vid, port, err);
-		return err;
-	}
-	err = dsa_8021q_vid_apply(ctx, upstream, tx_vid, 0, enabled);
+	err = dsa_port_tag_8021q_vlan_add(dp, tx_vid, false);
 	if (err) {
-		dev_err(ctx->ds->dev,
-			"Failed to apply TX VID %d on port %d: %d\n",
-			tx_vid, upstream, err);
+		dev_err(ds->dev,
+			"Failed to apply TX VID %d on port %d: %pe\n",
+			tx_vid, port, ERR_PTR(err));
 		return err;
 	}
 
 	return err;
 }
 
-int dsa_8021q_setup(struct dsa_8021q_context *ctx, bool enabled)
+static void dsa_tag_8021q_port_teardown(struct dsa_switch *ds, int port)
 {
-	int rc, port;
+	struct dsa_8021q_context *ctx = ds->tag_8021q_ctx;
+	struct dsa_port *dp = dsa_to_port(ds, port);
+	u16 rx_vid = dsa_8021q_rx_vid(ds, port);
+	u16 tx_vid = dsa_8021q_tx_vid(ds, port);
+	struct net_device *master;
 
-	ASSERT_RTNL();
+	/* The CPU port is implicitly configured by
+	 * configuring the front-panel ports
+	 */
+	if (!dsa_port_is_user(dp))
+		return;
 
-	for (port = 0; port < ctx->ds->num_ports; port++) {
-		rc = dsa_8021q_setup_port(ctx, port, enabled);
-		if (rc < 0) {
-			dev_err(ctx->ds->dev,
-				"Failed to setup VLAN tagging for port %d: %d\n",
-				port, rc);
-			return rc;
-		}
-	}
+	master = dp->cpu_dp->master;
 
-	return 0;
-}
-EXPORT_SYMBOL_GPL(dsa_8021q_setup);
+	dsa_port_tag_8021q_vlan_del(dp, rx_vid, false);
 
-static int dsa_8021q_crosschip_link_apply(struct dsa_8021q_context *ctx,
-					  int port,
-					  struct dsa_8021q_context *other_ctx,
-					  int other_port, bool enabled)
-{
-	u16 rx_vid = dsa_8021q_rx_vid(ctx->ds, port);
+	vlan_vid_del(master, ctx->proto, rx_vid);
 
-	/* @rx_vid of local @ds port @port goes to @other_port of
-	 * @other_ds
-	 */
-	return dsa_8021q_vid_apply(other_ctx, other_port, rx_vid,
-				   BRIDGE_VLAN_INFO_UNTAGGED, enabled);
+	dsa_port_tag_8021q_vlan_del(dp, tx_vid, false);
 }
 
-static int dsa_8021q_crosschip_link_add(struct dsa_8021q_context *ctx, int port,
-					struct dsa_8021q_context *other_ctx,
-					int other_port)
+static int dsa_tag_8021q_setup(struct dsa_switch *ds)
 {
-	struct dsa_8021q_crosschip_link *c;
+	int err, port;
+
+	ASSERT_RTNL();
 
-	list_for_each_entry(c, &ctx->crosschip_links, list) {
-		if (c->port == port && c->other_ctx == other_ctx &&
-		    c->other_port == other_port) {
-			refcount_inc(&c->refcount);
-			return 0;
+	for (port = 0; port < ds->num_ports; port++) {
+		err = dsa_tag_8021q_port_setup(ds, port);
+		if (err < 0) {
+			dev_err(ds->dev,
+				"Failed to setup VLAN tagging for port %d: %pe\n",
+				port, ERR_PTR(err));
+			return err;
 		}
 	}
 
-	dev_dbg(ctx->ds->dev,
-		"adding crosschip link from port %d to %s port %d\n",
-		port, dev_name(other_ctx->ds->dev), other_port);
-
-	c = kzalloc(sizeof(*c), GFP_KERNEL);
-	if (!c)
-		return -ENOMEM;
-
-	c->port = port;
-	c->other_ctx = other_ctx;
-	c->other_port = other_port;
-	refcount_set(&c->refcount, 1);
-
-	list_add(&c->list, &ctx->crosschip_links);
-
 	return 0;
 }
 
-static void dsa_8021q_crosschip_link_del(struct dsa_8021q_context *ctx,
-					 struct dsa_8021q_crosschip_link *c,
-					 bool *keep)
+static void dsa_tag_8021q_teardown(struct dsa_switch *ds)
 {
-	*keep = !refcount_dec_and_test(&c->refcount);
+	int port;
 
-	if (*keep)
-		return;
-
-	dev_dbg(ctx->ds->dev,
-		"deleting crosschip link from port %d to %s port %d\n",
-		c->port, dev_name(c->other_ctx->ds->dev), c->other_port);
+	ASSERT_RTNL();
 
-	list_del(&c->list);
-	kfree(c);
+	for (port = 0; port < ds->num_ports; port++)
+		dsa_tag_8021q_port_teardown(ds, port);
 }
 
-/* Make traffic from local port @port be received by remote port @other_port.
- * This means that our @rx_vid needs to be installed on @other_ds's upstream
- * and user ports. The user ports should be egress-untagged so that they can
- * pop the dsa_8021q VLAN. But the @other_upstream can be either egress-tagged
- * or untagged: it doesn't matter, since it should never egress a frame having
- * our @rx_vid.
- */
-int dsa_8021q_crosschip_bridge_join(struct dsa_8021q_context *ctx, int port,
-				    struct dsa_8021q_context *other_ctx,
-				    int other_port)
+int dsa_tag_8021q_register(struct dsa_switch *ds, __be16 proto)
 {
-	/* @other_upstream is how @other_ds reaches us. If we are part
-	 * of disjoint trees, then we are probably connected through
-	 * our CPU ports. If we're part of the same tree though, we should
-	 * probably use dsa_towards_port.
-	 */
-	int other_upstream = dsa_upstream_port(other_ctx->ds, other_port);
-	int rc;
+	struct dsa_8021q_context *ctx;
 
-	rc = dsa_8021q_crosschip_link_add(ctx, port, other_ctx, other_port);
-	if (rc)
-		return rc;
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
 
-	rc = dsa_8021q_crosschip_link_apply(ctx, port, other_ctx,
-					    other_port, true);
-	if (rc)
-		return rc;
+	ctx->proto = proto;
+	ctx->ds = ds;
 
-	rc = dsa_8021q_crosschip_link_add(ctx, port, other_ctx, other_upstream);
-	if (rc)
-		return rc;
+	INIT_LIST_HEAD(&ctx->vlans);
 
-	return dsa_8021q_crosschip_link_apply(ctx, port, other_ctx,
-					      other_upstream, true);
+	ds->tag_8021q_ctx = ctx;
+
+	return dsa_tag_8021q_setup(ds);
 }
-EXPORT_SYMBOL_GPL(dsa_8021q_crosschip_bridge_join);
+EXPORT_SYMBOL_GPL(dsa_tag_8021q_register);
 
-int dsa_8021q_crosschip_bridge_leave(struct dsa_8021q_context *ctx, int port,
-				     struct dsa_8021q_context *other_ctx,
-				     int other_port)
+void dsa_tag_8021q_unregister(struct dsa_switch *ds)
 {
-	int other_upstream = dsa_upstream_port(other_ctx->ds, other_port);
-	struct dsa_8021q_crosschip_link *c, *n;
-
-	list_for_each_entry_safe(c, n, &ctx->crosschip_links, list) {
-		if (c->port == port && c->other_ctx == other_ctx &&
-		    (c->other_port == other_port ||
-		     c->other_port == other_upstream)) {
-			struct dsa_8021q_context *other_ctx = c->other_ctx;
-			int other_port = c->other_port;
-			bool keep;
-			int rc;
-
-			dsa_8021q_crosschip_link_del(ctx, c, &keep);
-			if (keep)
-				continue;
-
-			rc = dsa_8021q_crosschip_link_apply(ctx, port,
-							    other_ctx,
-							    other_port,
-							    false);
-			if (rc)
-				return rc;
-		}
+	struct dsa_8021q_context *ctx = ds->tag_8021q_ctx;
+	struct dsa_tag_8021q_vlan *v, *n;
+
+	dsa_tag_8021q_teardown(ds);
+
+	list_for_each_entry_safe(v, n, &ctx->vlans, list) {
+		list_del(&v->list);
+		kfree(v);
 	}
 
-	return 0;
+	ds->tag_8021q_ctx = NULL;
+
+	kfree(ctx);
 }
-EXPORT_SYMBOL_GPL(dsa_8021q_crosschip_bridge_leave);
+EXPORT_SYMBOL_GPL(dsa_tag_8021q_unregister);
 
 struct sk_buff *dsa_8021q_xmit(struct sk_buff *skb, struct net_device *netdev,
 			       u16 tpid, u16 tci)
@@ -471,8 +573,7 @@ struct sk_buff *dsa_8021q_xmit(struct sk_buff *skb, struct net_device *netdev,
 }
 EXPORT_SYMBOL_GPL(dsa_8021q_xmit);
 
-void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
-		   int *subvlan)
+void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id)
 {
 	u16 vid, tci;
 
@@ -489,9 +590,6 @@ void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
 
 	*source_port = dsa_8021q_rx_source_port(vid);
 	*switch_id = dsa_8021q_rx_switch_id(vid);
-	*subvlan = dsa_8021q_rx_subvlan(vid);
 	skb->priority = (tci & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
 }
 EXPORT_SYMBOL_GPL(dsa_8021q_rcv);
-
-MODULE_LICENSE("GPL v2");
diff --git a/net/dsa/tag_ar9331.c b/net/dsa/tag_ar9331.c
index 0efae1a372b3..8a02ac44282f 100644
--- a/net/dsa/tag_ar9331.c
+++ b/net/dsa/tag_ar9331.c
@@ -44,8 +44,7 @@ static struct sk_buff *ar9331_tag_xmit(struct sk_buff *skb,
 }
 
 static struct sk_buff *ar9331_tag_rcv(struct sk_buff *skb,
-				      struct net_device *ndev,
-				      struct packet_type *pt)
+				      struct net_device *ndev)
 {
 	u8 ver, port;
 	u16 hdr;
diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c
index 0750af951fc9..96dbb8ee2fee 100644
--- a/net/dsa/tag_brcm.c
+++ b/net/dsa/tag_brcm.c
@@ -99,7 +99,7 @@ static struct sk_buff *brcm_tag_xmit_ll(struct sk_buff *skb,
 	skb_push(skb, BRCM_TAG_LEN);
 
 	if (offset)
-		memmove(skb->data, skb->data + BRCM_TAG_LEN, offset);
+		dsa_alloc_etype_header(skb, BRCM_TAG_LEN);
 
 	brcm_tag = skb->data + offset;
 
@@ -136,7 +136,6 @@ static struct sk_buff *brcm_tag_xmit_ll(struct sk_buff *skb,
  */
 static struct sk_buff *brcm_tag_rcv_ll(struct sk_buff *skb,
 				       struct net_device *dev,
-				       struct packet_type *pt,
 				       unsigned int offset)
 {
 	int source_port;
@@ -167,7 +166,7 @@ static struct sk_buff *brcm_tag_rcv_ll(struct sk_buff *skb,
 	/* Remove Broadcom tag and update checksum */
 	skb_pull_rcsum(skb, BRCM_TAG_LEN);
 
-	skb->offload_fwd_mark = 1;
+	dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
@@ -182,20 +181,16 @@ static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb,
 }
 
 
-static struct sk_buff *brcm_tag_rcv(struct sk_buff *skb, struct net_device *dev,
-				    struct packet_type *pt)
+static struct sk_buff *brcm_tag_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	struct sk_buff *nskb;
 
 	/* skb->data points to the EtherType, the tag is right before it */
-	nskb = brcm_tag_rcv_ll(skb, dev, pt, 2);
+	nskb = brcm_tag_rcv_ll(skb, dev, 2);
 	if (!nskb)
 		return nskb;
 
-	/* Move the Ethernet DA and SA */
-	memmove(nskb->data - ETH_HLEN,
-		nskb->data - ETH_HLEN - BRCM_TAG_LEN,
-		2 * ETH_ALEN);
+	dsa_strip_etype_header(skb, BRCM_TAG_LEN);
 
 	return nskb;
 }
@@ -233,7 +228,7 @@ static struct sk_buff *brcm_leg_tag_xmit(struct sk_buff *skb,
 
 	skb_push(skb, BRCM_LEG_TAG_LEN);
 
-	memmove(skb->data, skb->data + BRCM_LEG_TAG_LEN, 2 * ETH_ALEN);
+	dsa_alloc_etype_header(skb, BRCM_LEG_TAG_LEN);
 
 	brcm_tag = skb->data + 2 * ETH_ALEN;
 
@@ -251,8 +246,7 @@ static struct sk_buff *brcm_leg_tag_xmit(struct sk_buff *skb,
 }
 
 static struct sk_buff *brcm_leg_tag_rcv(struct sk_buff *skb,
-					struct net_device *dev,
-					struct packet_type *pt)
+					struct net_device *dev)
 {
 	int source_port;
 	u8 *brcm_tag;
@@ -260,7 +254,7 @@ static struct sk_buff *brcm_leg_tag_rcv(struct sk_buff *skb,
 	if (unlikely(!pskb_may_pull(skb, BRCM_LEG_PORT_ID)))
 		return NULL;
 
-	brcm_tag = skb->data - 2;
+	brcm_tag = dsa_etype_header_pos_rx(skb);
 
 	source_port = brcm_tag[5] & BRCM_LEG_PORT_ID;
 
@@ -271,12 +265,9 @@ static struct sk_buff *brcm_leg_tag_rcv(struct sk_buff *skb,
 	/* Remove Broadcom tag and update checksum */
 	skb_pull_rcsum(skb, BRCM_LEG_TAG_LEN);
 
-	skb->offload_fwd_mark = 1;
+	dsa_default_offload_fwd_mark(skb);
 
-	/* Move the Ethernet DA and SA */
-	memmove(skb->data - ETH_HLEN,
-		skb->data - ETH_HLEN - BRCM_LEG_TAG_LEN,
-		2 * ETH_ALEN);
+	dsa_strip_etype_header(skb, BRCM_LEG_TAG_LEN);
 
 	return skb;
 }
@@ -302,11 +293,10 @@ static struct sk_buff *brcm_tag_xmit_prepend(struct sk_buff *skb,
 }
 
 static struct sk_buff *brcm_tag_rcv_prepend(struct sk_buff *skb,
-					    struct net_device *dev,
-					    struct packet_type *pt)
+					    struct net_device *dev)
 {
 	/* tag is prepended to the packet */
-	return brcm_tag_rcv_ll(skb, dev, pt, ETH_HLEN);
+	return brcm_tag_rcv_ll(skb, dev, ETH_HLEN);
 }
 
 static const struct dsa_device_ops brcm_prepend_netdev_ops = {
diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c
index a822355afc90..77d0ce89ab77 100644
--- a/net/dsa/tag_dsa.c
+++ b/net/dsa/tag_dsa.c
@@ -126,18 +126,53 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev,
 				   u8 extra)
 {
 	struct dsa_port *dp = dsa_slave_to_port(dev);
+	u8 tag_dev, tag_port;
+	enum dsa_cmd cmd;
 	u8 *dsa_header;
+	u16 pvid = 0;
+	int err;
+
+	if (skb->offload_fwd_mark) {
+		struct dsa_switch_tree *dst = dp->ds->dst;
+		struct net_device *br = dp->bridge_dev;
+
+		cmd = DSA_CMD_FORWARD;
+
+		/* When offloading forwarding for a bridge, inject FORWARD
+		 * packets on behalf of a virtual switch device with an index
+		 * past the physical switches.
+		 */
+		tag_dev = dst->last_switch + 1 + dp->bridge_num;
+		tag_port = 0;
+
+		/* If we are offloading forwarding for a VLAN-unaware bridge,
+		 * inject packets to hardware using the bridge's pvid, since
+		 * that's where the packets ingressed from.
+		 */
+		if (!br_vlan_enabled(br)) {
+			/* Safe because __dev_queue_xmit() runs under
+			 * rcu_read_lock_bh()
+			 */
+			err = br_vlan_get_pvid_rcu(br, &pvid);
+			if (err)
+				return NULL;
+		}
+	} else {
+		cmd = DSA_CMD_FROM_CPU;
+		tag_dev = dp->ds->index;
+		tag_port = dp->index;
+	}
 
 	if (skb->protocol == htons(ETH_P_8021Q)) {
 		if (extra) {
 			skb_push(skb, extra);
-			memmove(skb->data, skb->data + extra, 2 * ETH_ALEN);
+			dsa_alloc_etype_header(skb, extra);
 		}
 
-		/* Construct tagged FROM_CPU DSA tag from 802.1Q tag. */
-		dsa_header = skb->data + 2 * ETH_ALEN + extra;
-		dsa_header[0] = (DSA_CMD_FROM_CPU << 6) | 0x20 | dp->ds->index;
-		dsa_header[1] = dp->index << 3;
+		/* Construct tagged DSA tag from 802.1Q tag. */
+		dsa_header = dsa_etype_header_pos_tx(skb) + extra;
+		dsa_header[0] = (cmd << 6) | 0x20 | tag_dev;
+		dsa_header[1] = tag_port << 3;
 
 		/* Move CFI field from byte 2 to byte 1. */
 		if (dsa_header[2] & 0x10) {
@@ -146,14 +181,15 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev,
 		}
 	} else {
 		skb_push(skb, DSA_HLEN + extra);
-		memmove(skb->data, skb->data + DSA_HLEN + extra, 2 * ETH_ALEN);
-
-		/* Construct untagged FROM_CPU DSA tag. */
-		dsa_header = skb->data + 2 * ETH_ALEN + extra;
-		dsa_header[0] = (DSA_CMD_FROM_CPU << 6) | dp->ds->index;
-		dsa_header[1] = dp->index << 3;
-		dsa_header[2] = 0x00;
-		dsa_header[3] = 0x00;
+		dsa_alloc_etype_header(skb, DSA_HLEN + extra);
+
+		/* Construct untagged DSA tag. */
+		dsa_header = dsa_etype_header_pos_tx(skb) + extra;
+
+		dsa_header[0] = (cmd << 6) | tag_dev;
+		dsa_header[1] = tag_port << 3;
+		dsa_header[2] = pvid >> 8;
+		dsa_header[3] = pvid & 0xff;
 	}
 
 	return skb;
@@ -162,20 +198,18 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev,
 static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 				  u8 extra)
 {
+	bool trap = false, trunk = false;
 	int source_device, source_port;
-	bool trunk = false;
 	enum dsa_code code;
 	enum dsa_cmd cmd;
 	u8 *dsa_header;
 
 	/* The ethertype field is part of the DSA header. */
-	dsa_header = skb->data - 2;
+	dsa_header = dsa_etype_header_pos_rx(skb);
 
 	cmd = dsa_header[0] >> 6;
 	switch (cmd) {
 	case DSA_CMD_FORWARD:
-		skb->offload_fwd_mark = 1;
-
 		trunk = !!(dsa_header[1] & 7);
 		break;
 
@@ -194,7 +228,6 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 			 * device (like a bridge) that forwarding has
 			 * already been done by hardware.
 			 */
-			skb->offload_fwd_mark = 1;
 			break;
 		case DSA_CODE_MGMT_TRAP:
 		case DSA_CODE_IGMP_MLD_TRAP:
@@ -202,6 +235,7 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 			/* Traps have, by definition, not been
 			 * forwarded by hardware, so don't mark them.
 			 */
+			trap = true;
 			break;
 		default:
 			/* Reserved code, this could be anything. Drop
@@ -235,6 +269,15 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 	if (!skb->dev)
 		return NULL;
 
+	/* When using LAG offload, skb->dev is not a DSA slave interface,
+	 * so we cannot call dsa_default_offload_fwd_mark and we need to
+	 * special-case it.
+	 */
+	if (trunk)
+		skb->offload_fwd_mark = true;
+	else if (!trap)
+		dsa_default_offload_fwd_mark(skb);
+
 	/* If the 'tagged' bit is set; convert the DSA tag to a 802.1Q
 	 * tag, and delete the ethertype (extra) if applicable. If the
 	 * 'tagged' bit is cleared; delete the DSA tag, and ethertype
@@ -269,14 +312,10 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 		memcpy(dsa_header, new_header, DSA_HLEN);
 
 		if (extra)
-			memmove(skb->data - ETH_HLEN,
-				skb->data - ETH_HLEN - extra,
-				2 * ETH_ALEN);
+			dsa_strip_etype_header(skb, extra);
 	} else {
 		skb_pull_rcsum(skb, DSA_HLEN);
-		memmove(skb->data - ETH_HLEN,
-			skb->data - ETH_HLEN - DSA_HLEN - extra,
-			2 * ETH_ALEN);
+		dsa_strip_etype_header(skb, DSA_HLEN + extra);
 	}
 
 	return skb;
@@ -289,8 +328,7 @@ static struct sk_buff *dsa_xmit(struct sk_buff *skb, struct net_device *dev)
 	return dsa_xmit_ll(skb, dev, 0);
 }
 
-static struct sk_buff *dsa_rcv(struct sk_buff *skb, struct net_device *dev,
-			       struct packet_type *pt)
+static struct sk_buff *dsa_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	if (unlikely(!pskb_may_pull(skb, DSA_HLEN)))
 		return NULL;
@@ -322,7 +360,7 @@ static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (!skb)
 		return NULL;
 
-	edsa_header = skb->data + 2 * ETH_ALEN;
+	edsa_header = dsa_etype_header_pos_tx(skb);
 	edsa_header[0] = (ETH_P_EDSA >> 8) & 0xff;
 	edsa_header[1] = ETH_P_EDSA & 0xff;
 	edsa_header[2] = 0x00;
@@ -330,8 +368,7 @@ static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev)
 	return skb;
 }
 
-static struct sk_buff *edsa_rcv(struct sk_buff *skb, struct net_device *dev,
-				struct packet_type *pt)
+static struct sk_buff *edsa_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	if (unlikely(!pskb_may_pull(skb, EDSA_HLEN)))
 		return NULL;
diff --git a/net/dsa/tag_gswip.c b/net/dsa/tag_gswip.c
index 5985dab06ab8..df7140984da3 100644
--- a/net/dsa/tag_gswip.c
+++ b/net/dsa/tag_gswip.c
@@ -75,8 +75,7 @@ static struct sk_buff *gswip_tag_xmit(struct sk_buff *skb,
 }
 
 static struct sk_buff *gswip_tag_rcv(struct sk_buff *skb,
-				     struct net_device *dev,
-				     struct packet_type *pt)
+				     struct net_device *dev)
 {
 	int port;
 	u8 *gswip_tag;
diff --git a/net/dsa/tag_hellcreek.c b/net/dsa/tag_hellcreek.c
index 424130f85f59..f64b805303cd 100644
--- a/net/dsa/tag_hellcreek.c
+++ b/net/dsa/tag_hellcreek.c
@@ -29,8 +29,7 @@ static struct sk_buff *hellcreek_xmit(struct sk_buff *skb,
 }
 
 static struct sk_buff *hellcreek_rcv(struct sk_buff *skb,
-				     struct net_device *dev,
-				     struct packet_type *pt)
+				     struct net_device *dev)
 {
 	/* Tag decoding */
 	u8 *tag = skb_tail_pointer(skb) - HELLCREEK_TAG_LEN;
@@ -44,7 +43,7 @@ static struct sk_buff *hellcreek_rcv(struct sk_buff *skb,
 
 	pskb_trim_rcsum(skb, skb->len - HELLCREEK_TAG_LEN);
 
-	skb->offload_fwd_mark = true;
+	dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c
index 53565f48934c..fa1d60d13ad9 100644
--- a/net/dsa/tag_ksz.c
+++ b/net/dsa/tag_ksz.c
@@ -24,7 +24,7 @@ static struct sk_buff *ksz_common_rcv(struct sk_buff *skb,
 
 	pskb_trim_rcsum(skb, skb->len - len);
 
-	skb->offload_fwd_mark = true;
+	dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
@@ -53,6 +53,9 @@ static struct sk_buff *ksz8795_xmit(struct sk_buff *skb, struct net_device *dev)
 	u8 *tag;
 	u8 *addr;
 
+	if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb))
+		return NULL;
+
 	/* Tag encoding */
 	tag = skb_put(skb, KSZ_INGRESS_TAG_LEN);
 	addr = skb_mac_header(skb);
@@ -64,8 +67,7 @@ static struct sk_buff *ksz8795_xmit(struct sk_buff *skb, struct net_device *dev)
 	return skb;
 }
 
-static struct sk_buff *ksz8795_rcv(struct sk_buff *skb, struct net_device *dev,
-				  struct packet_type *pt)
+static struct sk_buff *ksz8795_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	u8 *tag = skb_tail_pointer(skb) - KSZ_EGRESS_TAG_LEN;
 
@@ -114,6 +116,9 @@ static struct sk_buff *ksz9477_xmit(struct sk_buff *skb,
 	u8 *addr;
 	u16 val;
 
+	if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb))
+		return NULL;
+
 	/* Tag encoding */
 	tag = skb_put(skb, KSZ9477_INGRESS_TAG_LEN);
 	addr = skb_mac_header(skb);
@@ -128,8 +133,7 @@ static struct sk_buff *ksz9477_xmit(struct sk_buff *skb,
 	return skb;
 }
 
-static struct sk_buff *ksz9477_rcv(struct sk_buff *skb, struct net_device *dev,
-				   struct packet_type *pt)
+static struct sk_buff *ksz9477_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	/* Tag decoding */
 	u8 *tag = skb_tail_pointer(skb) - KSZ_EGRESS_TAG_LEN;
@@ -164,6 +168,9 @@ static struct sk_buff *ksz9893_xmit(struct sk_buff *skb,
 	u8 *addr;
 	u8 *tag;
 
+	if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb))
+		return NULL;
+
 	/* Tag encoding */
 	tag = skb_put(skb, KSZ_INGRESS_TAG_LEN);
 	addr = skb_mac_header(skb);
diff --git a/net/dsa/tag_lan9303.c b/net/dsa/tag_lan9303.c
index 26207ef39ebc..cb548188f813 100644
--- a/net/dsa/tag_lan9303.c
+++ b/net/dsa/tag_lan9303.c
@@ -62,9 +62,10 @@ static struct sk_buff *lan9303_xmit(struct sk_buff *skb, struct net_device *dev)
 	skb_push(skb, LAN9303_TAG_LEN);
 
 	/* make room between MACs and Ether-Type */
-	memmove(skb->data, skb->data + LAN9303_TAG_LEN, 2 * ETH_ALEN);
+	dsa_alloc_etype_header(skb, LAN9303_TAG_LEN);
+
+	lan9303_tag = dsa_etype_header_pos_tx(skb);
 
-	lan9303_tag = (__be16 *)(skb->data + 2 * ETH_ALEN);
 	tag = lan9303_xmit_use_arl(dp, skb->data) ?
 		LAN9303_TAG_TX_USE_ALR :
 		dp->index | LAN9303_TAG_TX_STP_OVERRIDE;
@@ -74,8 +75,7 @@ static struct sk_buff *lan9303_xmit(struct sk_buff *skb, struct net_device *dev)
 	return skb;
 }
 
-static struct sk_buff *lan9303_rcv(struct sk_buff *skb, struct net_device *dev,
-				   struct packet_type *pt)
+static struct sk_buff *lan9303_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	__be16 *lan9303_tag;
 	u16 lan9303_tag1;
@@ -87,13 +87,7 @@ static struct sk_buff *lan9303_rcv(struct sk_buff *skb, struct net_device *dev,
 		return NULL;
 	}
 
-	/* '->data' points into the middle of our special VLAN tag information:
-	 *
-	 * ~ MAC src   | 0x81 | 0x00 | 0xyy | 0xzz | ether type
-	 *                           ^
-	 *                        ->data
-	 */
-	lan9303_tag = (__be16 *)(skb->data - 2);
+	lan9303_tag = dsa_etype_header_pos_rx(skb);
 
 	if (lan9303_tag[0] != htons(ETH_P_8021Q)) {
 		dev_warn_ratelimited(&dev->dev, "Dropping packet due to invalid VLAN marker\n");
@@ -113,9 +107,11 @@ static struct sk_buff *lan9303_rcv(struct sk_buff *skb, struct net_device *dev,
 	 * and the current ethertype field.
 	 */
 	skb_pull_rcsum(skb, 2 + 2);
-	memmove(skb->data - ETH_HLEN, skb->data - (ETH_HLEN + LAN9303_TAG_LEN),
-		2 * ETH_ALEN);
-	skb->offload_fwd_mark = !(lan9303_tag1 & LAN9303_TAG_RX_TRAPPED_TO_CPU);
+
+	dsa_strip_etype_header(skb, LAN9303_TAG_LEN);
+
+	if (!(lan9303_tag1 & LAN9303_TAG_RX_TRAPPED_TO_CPU))
+		dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
diff --git a/net/dsa/tag_mtk.c b/net/dsa/tag_mtk.c
index cc3ba864ad5b..415d8ece242a 100644
--- a/net/dsa/tag_mtk.c
+++ b/net/dsa/tag_mtk.c
@@ -41,10 +41,10 @@ static struct sk_buff *mtk_tag_xmit(struct sk_buff *skb,
 	default:
 		xmit_tpid = MTK_HDR_XMIT_UNTAGGED;
 		skb_push(skb, MTK_HDR_LEN);
-		memmove(skb->data, skb->data + MTK_HDR_LEN, 2 * ETH_ALEN);
+		dsa_alloc_etype_header(skb, MTK_HDR_LEN);
 	}
 
-	mtk_tag = skb->data + 2 * ETH_ALEN;
+	mtk_tag = dsa_etype_header_pos_tx(skb);
 
 	/* Mark tag attribute on special tag insertion to notify hardware
 	 * whether that's a combined special tag with 802.1Q header.
@@ -61,8 +61,7 @@ static struct sk_buff *mtk_tag_xmit(struct sk_buff *skb,
 	return skb;
 }
 
-static struct sk_buff *mtk_tag_rcv(struct sk_buff *skb, struct net_device *dev,
-				   struct packet_type *pt)
+static struct sk_buff *mtk_tag_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	u16 hdr;
 	int port;
@@ -71,19 +70,13 @@ static struct sk_buff *mtk_tag_rcv(struct sk_buff *skb, struct net_device *dev,
 	if (unlikely(!pskb_may_pull(skb, MTK_HDR_LEN)))
 		return NULL;
 
-	/* The MTK header is added by the switch between src addr
-	 * and ethertype at this point, skb->data points to 2 bytes
-	 * after src addr so header should be 2 bytes right before.
-	 */
-	phdr = (__be16 *)(skb->data - 2);
+	phdr = dsa_etype_header_pos_rx(skb);
 	hdr = ntohs(*phdr);
 
 	/* Remove MTK tag and recalculate checksum. */
 	skb_pull_rcsum(skb, MTK_HDR_LEN);
 
-	memmove(skb->data - ETH_HLEN,
-		skb->data - ETH_HLEN - MTK_HDR_LEN,
-		2 * ETH_ALEN);
+	dsa_strip_etype_header(skb, MTK_HDR_LEN);
 
 	/* Get source port information */
 	port = (hdr & MTK_HDR_RECV_SOURCE_PORT_MASK);
@@ -92,7 +85,7 @@ static struct sk_buff *mtk_tag_rcv(struct sk_buff *skb, struct net_device *dev,
 	if (!skb->dev)
 		return NULL;
 
-	skb->offload_fwd_mark = 1;
+	dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c
index 190f4bfd3bef..d37ab98e7fe1 100644
--- a/net/dsa/tag_ocelot.c
+++ b/net/dsa/tag_ocelot.c
@@ -55,8 +55,7 @@ static struct sk_buff *seville_xmit(struct sk_buff *skb,
 }
 
 static struct sk_buff *ocelot_rcv(struct sk_buff *skb,
-				  struct net_device *netdev,
-				  struct packet_type *pt)
+				  struct net_device *netdev)
 {
 	u64 src_port, qos_class;
 	u64 vlan_tci, tag_type;
@@ -104,7 +103,7 @@ static struct sk_buff *ocelot_rcv(struct sk_buff *skb,
 		 */
 		return NULL;
 
-	skb->offload_fwd_mark = 1;
+	dsa_default_offload_fwd_mark(skb);
 	skb->priority = qos_class;
 
 	/* Ocelot switches copy frames unmodified to the CPU. However, it is
diff --git a/net/dsa/tag_ocelot_8021q.c b/net/dsa/tag_ocelot_8021q.c
index 85ac85c3af8c..3038a257ba05 100644
--- a/net/dsa/tag_ocelot_8021q.c
+++ b/net/dsa/tag_ocelot_8021q.c
@@ -38,18 +38,17 @@ static struct sk_buff *ocelot_xmit(struct sk_buff *skb,
 }
 
 static struct sk_buff *ocelot_rcv(struct sk_buff *skb,
-				  struct net_device *netdev,
-				  struct packet_type *pt)
+				  struct net_device *netdev)
 {
-	int src_port, switch_id, subvlan;
+	int src_port, switch_id;
 
-	dsa_8021q_rcv(skb, &src_port, &switch_id, &subvlan);
+	dsa_8021q_rcv(skb, &src_port, &switch_id);
 
 	skb->dev = dsa_master_find_slave(netdev, switch_id, src_port);
 	if (!skb->dev)
 		return NULL;
 
-	skb->offload_fwd_mark = 1;
+	dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
diff --git a/net/dsa/tag_qca.c b/net/dsa/tag_qca.c
index 693bda013065..1ea9401b8ace 100644
--- a/net/dsa/tag_qca.c
+++ b/net/dsa/tag_qca.c
@@ -36,8 +36,8 @@ static struct sk_buff *qca_tag_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	skb_push(skb, QCA_HDR_LEN);
 
-	memmove(skb->data, skb->data + QCA_HDR_LEN, 2 * ETH_ALEN);
-	phdr = (__be16 *)(skb->data + 2 * ETH_ALEN);
+	dsa_alloc_etype_header(skb, QCA_HDR_LEN);
+	phdr = dsa_etype_header_pos_tx(skb);
 
 	/* Set the version field, and set destination port information */
 	hdr = QCA_HDR_VERSION << QCA_HDR_XMIT_VERSION_S |
@@ -48,8 +48,7 @@ static struct sk_buff *qca_tag_xmit(struct sk_buff *skb, struct net_device *dev)
 	return skb;
 }
 
-static struct sk_buff *qca_tag_rcv(struct sk_buff *skb, struct net_device *dev,
-				   struct packet_type *pt)
+static struct sk_buff *qca_tag_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	u8 ver;
 	u16  hdr;
@@ -59,11 +58,7 @@ static struct sk_buff *qca_tag_rcv(struct sk_buff *skb, struct net_device *dev,
 	if (unlikely(!pskb_may_pull(skb, QCA_HDR_LEN)))
 		return NULL;
 
-	/* The QCA header is added by the switch between src addr and Ethertype
-	 * At this point, skb->data points to ethertype so header should be
-	 * right before
-	 */
-	phdr = (__be16 *)(skb->data - 2);
+	phdr = dsa_etype_header_pos_rx(skb);
 	hdr = ntohs(*phdr);
 
 	/* Make sure the version is correct */
@@ -73,8 +68,7 @@ static struct sk_buff *qca_tag_rcv(struct sk_buff *skb, struct net_device *dev,
 
 	/* Remove QCA tag and recalculate checksum */
 	skb_pull_rcsum(skb, QCA_HDR_LEN);
-	memmove(skb->data - ETH_HLEN, skb->data - ETH_HLEN - QCA_HDR_LEN,
-		ETH_HLEN - QCA_HDR_LEN);
+	dsa_strip_etype_header(skb, QCA_HDR_LEN);
 
 	/* Get source port information */
 	port = (hdr & QCA_HDR_RECV_SOURCE_PORT_MASK);
diff --git a/net/dsa/tag_rtl4_a.c b/net/dsa/tag_rtl4_a.c
index 57c46b4ab2b3..f920487ae145 100644
--- a/net/dsa/tag_rtl4_a.c
+++ b/net/dsa/tag_rtl4_a.c
@@ -47,16 +47,17 @@ static struct sk_buff *rtl4a_tag_xmit(struct sk_buff *skb,
 		   dp->index);
 	skb_push(skb, RTL4_A_HDR_LEN);
 
-	memmove(skb->data, skb->data + RTL4_A_HDR_LEN, 2 * ETH_ALEN);
-	tag = skb->data + 2 * ETH_ALEN;
+	dsa_alloc_etype_header(skb, RTL4_A_HDR_LEN);
+	tag = dsa_etype_header_pos_tx(skb);
 
 	/* Set Ethertype */
 	p = (__be16 *)tag;
 	*p = htons(RTL4_A_ETHERTYPE);
 
-	out = (RTL4_A_PROTOCOL_RTL8366RB << 12) | (2 << 8);
-	/* The lower bits is the port number */
-	out |= (u8)dp->index;
+	out = (RTL4_A_PROTOCOL_RTL8366RB << RTL4_A_PROTOCOL_SHIFT) | (2 << 8);
+	/* The lower bits indicate the port number */
+	out |= BIT(dp->index);
+
 	p = (__be16 *)(tag + 2);
 	*p = htons(out);
 
@@ -64,8 +65,7 @@ static struct sk_buff *rtl4a_tag_xmit(struct sk_buff *skb,
 }
 
 static struct sk_buff *rtl4a_tag_rcv(struct sk_buff *skb,
-				     struct net_device *dev,
-				     struct packet_type *pt)
+				     struct net_device *dev)
 {
 	u16 protport;
 	__be16 *p;
@@ -77,12 +77,7 @@ static struct sk_buff *rtl4a_tag_rcv(struct sk_buff *skb,
 	if (unlikely(!pskb_may_pull(skb, RTL4_A_HDR_LEN)))
 		return NULL;
 
-	/* The RTL4 header has its own custom Ethertype 0x8899 and that
-	 * starts right at the beginning of the packet, after the src
-	 * ethernet addr. Apparently skb->data always points 2 bytes in,
-	 * behind the Ethertype.
-	 */
-	tag = skb->data - 2;
+	tag = dsa_etype_header_pos_rx(skb);
 	p = (__be16 *)tag;
 	etype = ntohs(*p);
 	if (etype != RTL4_A_ETHERTYPE) {
@@ -109,12 +104,9 @@ static struct sk_buff *rtl4a_tag_rcv(struct sk_buff *skb,
 	/* Remove RTL4 tag and recalculate checksum */
 	skb_pull_rcsum(skb, RTL4_A_HDR_LEN);
 
-	/* Move ethernet DA and SA in front of the data */
-	memmove(skb->data - ETH_HLEN,
-		skb->data - ETH_HLEN - RTL4_A_HDR_LEN,
-		2 * ETH_ALEN);
+	dsa_strip_etype_header(skb, RTL4_A_HDR_LEN);
 
-	skb->offload_fwd_mark = 1;
+	dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 9c2df9ece01b..c054f48541c8 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -115,56 +115,117 @@ static inline bool sja1105_is_meta_frame(const struct sk_buff *skb)
 	return true;
 }
 
-static bool sja1105_can_use_vlan_as_tags(const struct sk_buff *skb)
+/* Calls sja1105_port_deferred_xmit in sja1105_main.c */
+static struct sk_buff *sja1105_defer_xmit(struct dsa_port *dp,
+					  struct sk_buff *skb)
 {
-	struct vlan_ethhdr *hdr = vlan_eth_hdr(skb);
-	u16 vlan_tci;
-
-	if (hdr->h_vlan_proto == htons(ETH_P_SJA1105))
-		return true;
+	struct sja1105_port *sp = dp->priv;
 
-	if (hdr->h_vlan_proto != htons(ETH_P_8021Q) &&
-	    !skb_vlan_tag_present(skb))
-		return false;
+	if (!dsa_port_is_sja1105(dp))
+		return skb;
 
-	if (skb_vlan_tag_present(skb))
-		vlan_tci = skb_vlan_tag_get(skb);
-	else
-		vlan_tci = ntohs(hdr->h_vlan_TCI);
+	/* Increase refcount so the kfree_skb in dsa_slave_xmit
+	 * won't really free the packet.
+	 */
+	skb_queue_tail(&sp->xmit_queue, skb_get(skb));
+	kthread_queue_work(sp->xmit_worker, &sp->xmit_work);
 
-	return vid_is_dsa_8021q(vlan_tci & VLAN_VID_MASK);
+	return NULL;
 }
 
-/* This is the first time the tagger sees the frame on RX.
- * Figure out if we can decode it.
+/* Send VLAN tags with a TPID that blends in with whatever VLAN protocol a
+ * bridge spanning ports of this switch might have.
  */
-static bool sja1105_filter(const struct sk_buff *skb, struct net_device *dev)
+static u16 sja1105_xmit_tpid(struct dsa_port *dp)
 {
-	if (sja1105_can_use_vlan_as_tags(skb))
-		return true;
-	if (sja1105_is_link_local(skb))
-		return true;
-	if (sja1105_is_meta_frame(skb))
-		return true;
-	return false;
+	struct dsa_switch *ds = dp->ds;
+	struct dsa_port *other_dp;
+	u16 proto;
+
+	/* Since VLAN awareness is global, then if this port is VLAN-unaware,
+	 * all ports are. Use the VLAN-unaware TPID used for tag_8021q.
+	 */
+	if (!dsa_port_is_vlan_filtering(dp))
+		return ETH_P_SJA1105;
+
+	/* Port is VLAN-aware, so there is a bridge somewhere (a single one,
+	 * we're sure about that). It may not be on this port though, so we
+	 * need to find it.
+	 */
+	list_for_each_entry(other_dp, &ds->dst->ports, list) {
+		if (other_dp->ds != ds)
+			continue;
+
+		if (!other_dp->bridge_dev)
+			continue;
+
+		/* Error is returned only if CONFIG_BRIDGE_VLAN_FILTERING,
+		 * which seems pointless to handle, as our port cannot become
+		 * VLAN-aware in that case.
+		 */
+		br_vlan_get_proto(other_dp->bridge_dev, &proto);
+
+		return proto;
+	}
+
+	WARN_ONCE(1, "Port is VLAN-aware but cannot find associated bridge!\n");
+
+	return ETH_P_SJA1105;
 }
 
-/* Calls sja1105_port_deferred_xmit in sja1105_main.c */
-static struct sk_buff *sja1105_defer_xmit(struct sja1105_port *sp,
-					  struct sk_buff *skb)
+static struct sk_buff *sja1105_imprecise_xmit(struct sk_buff *skb,
+					      struct net_device *netdev)
 {
-	/* Increase refcount so the kfree_skb in dsa_slave_xmit
-	 * won't really free the packet.
+	struct dsa_port *dp = dsa_slave_to_port(netdev);
+	struct net_device *br = dp->bridge_dev;
+	u16 tx_vid;
+
+	/* If the port is under a VLAN-aware bridge, just slide the
+	 * VLAN-tagged packet into the FDB and hope for the best.
+	 * This works because we support a single VLAN-aware bridge
+	 * across the entire dst, and its VLANs cannot be shared with
+	 * any standalone port.
 	 */
-	skb_queue_tail(&sp->xmit_queue, skb_get(skb));
-	kthread_queue_work(sp->xmit_worker, &sp->xmit_work);
+	if (br_vlan_enabled(br))
+		return skb;
 
-	return NULL;
+	/* If the port is under a VLAN-unaware bridge, use an imprecise
+	 * TX VLAN that targets the bridge's entire broadcast domain,
+	 * instead of just the specific port.
+	 */
+	tx_vid = dsa_8021q_bridge_tx_fwd_offload_vid(dp->bridge_num);
+
+	return dsa_8021q_xmit(skb, netdev, sja1105_xmit_tpid(dp), tx_vid);
 }
 
-static u16 sja1105_xmit_tpid(struct sja1105_port *sp)
+/* Transform untagged control packets into pvid-tagged control packets so that
+ * all packets sent by this tagger are VLAN-tagged and we can configure the
+ * switch to drop untagged packets coming from the DSA master.
+ */
+static struct sk_buff *sja1105_pvid_tag_control_pkt(struct dsa_port *dp,
+						    struct sk_buff *skb, u8 pcp)
 {
-	return sp->xmit_tpid;
+	__be16 xmit_tpid = htons(sja1105_xmit_tpid(dp));
+	struct vlan_ethhdr *hdr;
+
+	/* If VLAN tag is in hwaccel area, move it to the payload
+	 * to deal with both cases uniformly and to ensure that
+	 * the VLANs are added in the right order.
+	 */
+	if (unlikely(skb_vlan_tag_present(skb))) {
+		skb = __vlan_hwaccel_push_inside(skb);
+		if (!skb)
+			return NULL;
+	}
+
+	hdr = (struct vlan_ethhdr *)skb_mac_header(skb);
+
+	/* If skb is already VLAN-tagged, leave that VLAN ID in place */
+	if (hdr->h_vlan_proto == xmit_tpid)
+		return skb;
+
+	return vlan_insert_tag(skb, xmit_tpid, (pcp << VLAN_PRIO_SHIFT) |
+			       SJA1105_DEFAULT_VLAN);
 }
 
 static struct sk_buff *sja1105_xmit(struct sk_buff *skb,
@@ -175,14 +236,22 @@ static struct sk_buff *sja1105_xmit(struct sk_buff *skb,
 	u16 queue_mapping = skb_get_queue_mapping(skb);
 	u8 pcp = netdev_txq_to_tc(netdev, queue_mapping);
 
+	if (skb->offload_fwd_mark)
+		return sja1105_imprecise_xmit(skb, netdev);
+
 	/* Transmitting management traffic does not rely upon switch tagging,
 	 * but instead SPI-installed management routes. Part 2 of this
 	 * is the .port_deferred_xmit driver callback.
 	 */
-	if (unlikely(sja1105_is_link_local(skb)))
-		return sja1105_defer_xmit(dp->priv, skb);
+	if (unlikely(sja1105_is_link_local(skb))) {
+		skb = sja1105_pvid_tag_control_pkt(dp, skb, pcp);
+		if (!skb)
+			return NULL;
+
+		return sja1105_defer_xmit(dp, skb);
+	}
 
-	return dsa_8021q_xmit(skb, netdev, sja1105_xmit_tpid(dp->priv),
+	return dsa_8021q_xmit(skb, netdev, sja1105_xmit_tpid(dp),
 			     ((pcp << VLAN_PRIO_SHIFT) | tx_vid));
 }
 
@@ -194,43 +263,45 @@ static struct sk_buff *sja1110_xmit(struct sk_buff *skb,
 	u16 tx_vid = dsa_8021q_tx_vid(dp->ds, dp->index);
 	u16 queue_mapping = skb_get_queue_mapping(skb);
 	u8 pcp = netdev_txq_to_tc(netdev, queue_mapping);
-	struct ethhdr *eth_hdr;
 	__be32 *tx_trailer;
 	__be16 *tx_header;
 	int trailer_pos;
 
+	if (skb->offload_fwd_mark)
+		return sja1105_imprecise_xmit(skb, netdev);
+
 	/* Transmitting control packets is done using in-band control
 	 * extensions, while data packets are transmitted using
 	 * tag_8021q TX VLANs.
 	 */
 	if (likely(!sja1105_is_link_local(skb)))
-		return dsa_8021q_xmit(skb, netdev, sja1105_xmit_tpid(dp->priv),
+		return dsa_8021q_xmit(skb, netdev, sja1105_xmit_tpid(dp),
 				     ((pcp << VLAN_PRIO_SHIFT) | tx_vid));
 
+	skb = sja1105_pvid_tag_control_pkt(dp, skb, pcp);
+	if (!skb)
+		return NULL;
+
 	skb_push(skb, SJA1110_HEADER_LEN);
 
-	/* Move Ethernet header to the left, making space for DSA tag */
-	memmove(skb->data, skb->data + SJA1110_HEADER_LEN, 2 * ETH_ALEN);
+	dsa_alloc_etype_header(skb, SJA1110_HEADER_LEN);
 
 	trailer_pos = skb->len;
 
-	/* On TX, skb->data points to skb_mac_header(skb) */
-	eth_hdr = (struct ethhdr *)skb->data;
-	tx_header = (__be16 *)(eth_hdr + 1);
+	tx_header = dsa_etype_header_pos_tx(skb);
 	tx_trailer = skb_put(skb, SJA1110_TX_TRAILER_LEN);
 
-	eth_hdr->h_proto = htons(ETH_P_SJA1110);
-
-	*tx_header = htons(SJA1110_HEADER_HOST_TO_SWITCH |
-			   SJA1110_TX_HEADER_HAS_TRAILER |
-			   SJA1110_TX_HEADER_TRAILER_POS(trailer_pos));
+	tx_header[0] = htons(ETH_P_SJA1110);
+	tx_header[1] = htons(SJA1110_HEADER_HOST_TO_SWITCH |
+			     SJA1110_TX_HEADER_HAS_TRAILER |
+			     SJA1110_TX_HEADER_TRAILER_POS(trailer_pos));
 	*tx_trailer = cpu_to_be32(SJA1110_TX_TRAILER_PRIO(pcp) |
 				  SJA1110_TX_TRAILER_SWITCHID(dp->ds->index) |
 				  SJA1110_TX_TRAILER_DESTPORTS(BIT(dp->index)));
 	if (clone) {
 		u8 ts_id = SJA1105_SKB_CB(clone)->ts_id;
 
-		*tx_header |= htons(SJA1110_TX_HEADER_TAKE_TS);
+		tx_header[1] |= htons(SJA1110_TX_HEADER_TAKE_TS);
 		*tx_trailer |= cpu_to_be32(SJA1110_TX_TRAILER_TSTAMP_ID(ts_id));
 	}
 
@@ -273,16 +344,16 @@ static struct sk_buff
 				bool is_link_local,
 				bool is_meta)
 {
-	struct sja1105_port *sp;
-	struct dsa_port *dp;
-
-	dp = dsa_slave_to_port(skb->dev);
-	sp = dp->priv;
-
 	/* Step 1: A timestampable frame was received.
 	 * Buffer it until we get its meta frame.
 	 */
 	if (is_link_local) {
+		struct dsa_port *dp = dsa_slave_to_port(skb->dev);
+		struct sja1105_port *sp = dp->priv;
+
+		if (unlikely(!dsa_port_is_sja1105(dp)))
+			return skb;
+
 		if (!test_bit(SJA1105_HWTS_RX_EN, &sp->data->state))
 			/* Do normal processing. */
 			return skb;
@@ -315,8 +386,13 @@ static struct sk_buff
 	 * frame, which serves no further purpose).
 	 */
 	} else if (is_meta) {
+		struct dsa_port *dp = dsa_slave_to_port(skb->dev);
+		struct sja1105_port *sp = dp->priv;
 		struct sk_buff *stampable_skb;
 
+		if (unlikely(!dsa_port_is_sja1105(dp)))
+			return skb;
+
 		/* Drop the meta frame if we're not in the right state
 		 * to process it.
 		 */
@@ -358,20 +434,6 @@ static struct sk_buff
 	return skb;
 }
 
-static void sja1105_decode_subvlan(struct sk_buff *skb, u16 subvlan)
-{
-	struct dsa_port *dp = dsa_slave_to_port(skb->dev);
-	struct sja1105_port *sp = dp->priv;
-	u16 vid = sp->subvlan_map[subvlan];
-	u16 vlan_tci;
-
-	if (vid == VLAN_N_VID)
-		return;
-
-	vlan_tci = (skb->priority << VLAN_PRIO_SHIFT) | vid;
-	__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tci);
-}
-
 static bool sja1105_skb_has_tag_8021q(const struct sk_buff *skb)
 {
 	u16 tpid = ntohs(eth_hdr(skb)->h_proto);
@@ -385,25 +447,45 @@ static bool sja1110_skb_has_inband_control_extension(const struct sk_buff *skb)
 	return ntohs(eth_hdr(skb)->h_proto) == ETH_P_SJA1110;
 }
 
+/* If the VLAN in the packet is a tag_8021q one, set @source_port and
+ * @switch_id and strip the header. Otherwise set @vid and keep it in the
+ * packet.
+ */
+static void sja1105_vlan_rcv(struct sk_buff *skb, int *source_port,
+			     int *switch_id, u16 *vid)
+{
+	struct vlan_ethhdr *hdr = (struct vlan_ethhdr *)skb_mac_header(skb);
+	u16 vlan_tci;
+
+	if (skb_vlan_tag_present(skb))
+		vlan_tci = skb_vlan_tag_get(skb);
+	else
+		vlan_tci = ntohs(hdr->h_vlan_TCI);
+
+	if (vid_is_dsa_8021q_rxvlan(vlan_tci & VLAN_VID_MASK))
+		return dsa_8021q_rcv(skb, source_port, switch_id);
+
+	/* Try our best with imprecise RX */
+	*vid = vlan_tci & VLAN_VID_MASK;
+}
+
 static struct sk_buff *sja1105_rcv(struct sk_buff *skb,
-				   struct net_device *netdev,
-				   struct packet_type *pt)
+				   struct net_device *netdev)
 {
-	int source_port, switch_id, subvlan = 0;
+	int source_port = -1, switch_id = -1;
 	struct sja1105_meta meta = {0};
 	struct ethhdr *hdr;
 	bool is_link_local;
 	bool is_meta;
+	u16 vid;
 
 	hdr = eth_hdr(skb);
 	is_link_local = sja1105_is_link_local(skb);
 	is_meta = sja1105_is_meta_frame(skb);
 
-	skb->offload_fwd_mark = 1;
-
 	if (sja1105_skb_has_tag_8021q(skb)) {
 		/* Normal traffic path. */
-		dsa_8021q_rcv(skb, &source_port, &switch_id, &subvlan);
+		sja1105_vlan_rcv(skb, &source_port, &switch_id, &vid);
 	} else if (is_link_local) {
 		/* Management traffic path. Switch embeds the switch ID and
 		 * port ID into bytes of the destination MAC, courtesy of
@@ -422,14 +504,17 @@ static struct sk_buff *sja1105_rcv(struct sk_buff *skb,
 		return NULL;
 	}
 
-	skb->dev = dsa_master_find_slave(netdev, switch_id, source_port);
+	if (source_port == -1 || switch_id == -1)
+		skb->dev = dsa_find_designated_bridge_port_by_vid(netdev, vid);
+	else
+		skb->dev = dsa_master_find_slave(netdev, switch_id, source_port);
 	if (!skb->dev) {
 		netdev_warn(netdev, "Couldn't decode source port\n");
 		return NULL;
 	}
 
-	if (subvlan)
-		sja1105_decode_subvlan(skb, subvlan);
+	if (!is_link_local)
+		dsa_default_offload_fwd_mark(skb);
 
 	return sja1105_rcv_meta_state_machine(skb, &meta, is_link_local,
 					      is_meta);
@@ -437,11 +522,11 @@ static struct sk_buff *sja1105_rcv(struct sk_buff *skb,
 
 static struct sk_buff *sja1110_rcv_meta(struct sk_buff *skb, u16 rx_header)
 {
+	u8 *buf = dsa_etype_header_pos_rx(skb) + SJA1110_HEADER_LEN;
 	int switch_id = SJA1110_RX_HEADER_SWITCH_ID(rx_header);
 	int n_ts = SJA1110_RX_HEADER_N_TS(rx_header);
 	struct net_device *master = skb->dev;
 	struct dsa_port *cpu_dp;
-	u8 *buf = skb->data + 2;
 	struct dsa_switch *ds;
 	int i;
 
@@ -474,7 +559,8 @@ static struct sk_buff *sja1110_rcv_meta(struct sk_buff *skb, u16 rx_header)
 
 static struct sk_buff *sja1110_rcv_inband_control_extension(struct sk_buff *skb,
 							    int *source_port,
-							    int *switch_id)
+							    int *switch_id,
+							    bool *host_only)
 {
 	u16 rx_header;
 
@@ -488,6 +574,9 @@ static struct sk_buff *sja1110_rcv_inband_control_extension(struct sk_buff *skb,
 	 */
 	rx_header = ntohs(*(__be16 *)skb->data);
 
+	if (rx_header & SJA1110_RX_HEADER_HOST_ONLY)
+		*host_only = true;
+
 	if (rx_header & SJA1110_RX_HEADER_IS_METADATA)
 		return sja1110_rcv_meta(skb, rx_header);
 
@@ -522,9 +611,7 @@ static struct sk_buff *sja1110_rcv_inband_control_extension(struct sk_buff *skb,
 	/* Advance skb->data past the DSA header */
 	skb_pull_rcsum(skb, SJA1110_HEADER_LEN);
 
-	/* Remove the DSA header */
-	memmove(skb->data - ETH_HLEN, skb->data - ETH_HLEN - SJA1110_HEADER_LEN,
-		2 * ETH_ALEN);
+	dsa_strip_etype_header(skb, SJA1110_HEADER_LEN);
 
 	/* With skb->data in its final place, update the MAC header
 	 * so that eth_hdr() continues to works properly.
@@ -535,34 +622,35 @@ static struct sk_buff *sja1110_rcv_inband_control_extension(struct sk_buff *skb,
 }
 
 static struct sk_buff *sja1110_rcv(struct sk_buff *skb,
-				   struct net_device *netdev,
-				   struct packet_type *pt)
+				   struct net_device *netdev)
 {
-	int source_port = -1, switch_id = -1, subvlan = 0;
-
-	skb->offload_fwd_mark = 1;
+	int source_port = -1, switch_id = -1;
+	bool host_only = false;
+	u16 vid = 0;
 
 	if (sja1110_skb_has_inband_control_extension(skb)) {
 		skb = sja1110_rcv_inband_control_extension(skb, &source_port,
-							   &switch_id);
+							   &switch_id,
+							   &host_only);
 		if (!skb)
 			return NULL;
 	}
 
 	/* Packets with in-band control extensions might still have RX VLANs */
 	if (likely(sja1105_skb_has_tag_8021q(skb)))
-		dsa_8021q_rcv(skb, &source_port, &switch_id, &subvlan);
+		sja1105_vlan_rcv(skb, &source_port, &switch_id, &vid);
 
-	skb->dev = dsa_master_find_slave(netdev, switch_id, source_port);
+	if (source_port == -1 || switch_id == -1)
+		skb->dev = dsa_find_designated_bridge_port_by_vid(netdev, vid);
+	else
+		skb->dev = dsa_master_find_slave(netdev, switch_id, source_port);
 	if (!skb->dev) {
-		netdev_warn(netdev,
-			    "Couldn't decode source port %d and switch id %d\n",
-			    source_port, switch_id);
+		netdev_warn(netdev, "Couldn't decode source port\n");
 		return NULL;
 	}
 
-	if (subvlan)
-		sja1105_decode_subvlan(skb, subvlan);
+	if (!host_only)
+		dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }
@@ -596,7 +684,6 @@ static const struct dsa_device_ops sja1105_netdev_ops = {
 	.proto = DSA_TAG_PROTO_SJA1105,
 	.xmit = sja1105_xmit,
 	.rcv = sja1105_rcv,
-	.filter = sja1105_filter,
 	.needed_headroom = VLAN_HLEN,
 	.flow_dissect = sja1105_flow_dissect,
 	.promisc_on_master = true,
@@ -610,7 +697,6 @@ static const struct dsa_device_ops sja1110_netdev_ops = {
 	.proto = DSA_TAG_PROTO_SJA1110,
 	.xmit = sja1110_xmit,
 	.rcv = sja1110_rcv,
-	.filter = sja1105_filter,
 	.flow_dissect = sja1110_flow_dissect,
 	.needed_headroom = SJA1110_HEADER_LEN + VLAN_HLEN,
 	.needed_tailroom = SJA1110_RX_TRAILER_LEN + SJA1110_MAX_PADDING_LEN,
diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c
index ba73804340a5..5749ba85c2b8 100644
--- a/net/dsa/tag_trailer.c
+++ b/net/dsa/tag_trailer.c
@@ -24,8 +24,7 @@ static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev)
 	return skb;
 }
 
-static struct sk_buff *trailer_rcv(struct sk_buff *skb, struct net_device *dev,
-				   struct packet_type *pt)
+static struct sk_buff *trailer_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	u8 *trailer;
 	int source_port;
diff --git a/net/dsa/tag_xrs700x.c b/net/dsa/tag_xrs700x.c
index a31ff7fcb45f..ff442b8af636 100644
--- a/net/dsa/tag_xrs700x.c
+++ b/net/dsa/tag_xrs700x.c
@@ -25,8 +25,7 @@ static struct sk_buff *xrs700x_xmit(struct sk_buff *skb, struct net_device *dev)
 	return skb;
 }
 
-static struct sk_buff *xrs700x_rcv(struct sk_buff *skb, struct net_device *dev,
-				   struct packet_type *pt)
+static struct sk_buff *xrs700x_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	int source_port;
 	u8 *trailer;
@@ -46,7 +45,7 @@ static struct sk_buff *xrs700x_rcv(struct sk_buff *skb, struct net_device *dev,
 		return NULL;
 
 	/* Frame is forwarded by hardware, don't forward in software. */
-	skb->offload_fwd_mark = 1;
+	dsa_default_offload_fwd_mark(skb);
 
 	return skb;
 }