Linux-libre 4.5-gnu

60 files changed, 10992 insertions, 2247 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx4/catas.c b/drivers/net/ethernet/mellanox/mlx4/catas.c
index 715de8aff..c7e939945 100644
--- a/drivers/net/ethernet/mellanox/mlx4/catas.c
+++ b/drivers/net/ethernet/mellanox/mlx4/catas.c
@@ -182,10 +182,17 @@ void mlx4_enter_error_state(struct mlx4_dev_persistent *persist)
 		err = mlx4_reset_slave(dev);
 	else
 		err = mlx4_reset_master(dev);
-	BUG_ON(err != 0);
 
+	if (!err) {
+		mlx4_err(dev, "device was reset successfully\n");
+	} else {
+		/* EEH could have disabled the PCI channel during reset. That's
+		 * recoverable and the PCI error flow will handle it.
+		 */
+		if (!pci_channel_offline(dev->persist->pdev))
+			BUG_ON(1);
+	}
 	dev->persist->state |= MLX4_DEVICE_STATE_INTERNAL_ERROR;
-	mlx4_err(dev, "device was reset successfully\n");
 	mutex_unlock(&persist->device_state_mutex);
 
 	/* At that step HW was already reset, now notify clients */
diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c
index d48d57934..e94ca1c3f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -2429,7 +2429,7 @@ err_thread:
 	flush_workqueue(priv->mfunc.master.comm_wq);
 	destroy_workqueue(priv->mfunc.master.comm_wq);
 err_slaves:
-	while (--i) {
+	while (i--) {
 		for (port = 1; port <= MLX4_MAX_PORTS; port++)
 			kfree(priv->mfunc.master.slave_state[i].vlan_filter[port]);
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/cq.c b/drivers/net/ethernet/mellanox/mlx4/cq.c
index 3348e646d..a849da92f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cq.c
@@ -318,7 +318,9 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent,
 	if (timestamp_en)
 		cq_context->flags  |= cpu_to_be32(1 << 19);
 
-	cq_context->logsize_usrpage = cpu_to_be32((ilog2(nent) << 24) | uar->index);
+	cq_context->logsize_usrpage =
+		cpu_to_be32((ilog2(nent) << 24) |
+			    mlx4_to_hw_uar_index(dev, uar->index));
 	cq_context->comp_eqn	    = priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(vector)].eqn;
 	cq_context->log_page_size   = mtt->page_shift - MLX4_ICM_PAGE_SHIFT;
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_cq.c b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
index eb8a4988d..af975a2b7 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
@@ -155,13 +155,11 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
 	cq->mcq.comp  = cq->is_tx ? mlx4_en_tx_irq : mlx4_en_rx_irq;
 	cq->mcq.event = mlx4_en_cq_event;
 
-	if (cq->is_tx) {
-		netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq,
-			       NAPI_POLL_WEIGHT);
-	} else {
+	if (cq->is_tx)
+		netif_tx_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq,
+				  NAPI_POLL_WEIGHT);
+	else
 		netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_rx_cq, 64);
-		napi_hash_add(&cq->napi);
-	}
 
 	napi_enable(&cq->napi);
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index ddb554188..dd84cabb2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -337,11 +337,7 @@ static int mlx4_en_get_sset_count(struct net_device *dev, int sset)
 	case ETH_SS_STATS:
 		return bitmap_iterator_count(&it) +
 			(priv->tx_ring_num * 2) +
-#ifdef CONFIG_NET_RX_BUSY_POLL
-			(priv->rx_ring_num * 5);
-#else
 			(priv->rx_ring_num * 2);
-#endif
 	case ETH_SS_TEST:
 		return MLX4_EN_NUM_SELF_TEST - !(priv->mdev->dev->caps.flags
 					& MLX4_DEV_CAP_FLAG_UC_LOOPBACK) * 2;
@@ -408,11 +404,6 @@ static void mlx4_en_get_ethtool_stats(struct net_device *dev,
 	for (i = 0; i < priv->rx_ring_num; i++) {
 		data[index++] = priv->rx_ring[i]->packets;
 		data[index++] = priv->rx_ring[i]->bytes;
-#ifdef CONFIG_NET_RX_BUSY_POLL
-		data[index++] = priv->rx_ring[i]->yields;
-		data[index++] = priv->rx_ring[i]->misses;
-		data[index++] = priv->rx_ring[i]->cleaned;
-#endif
 	}
 	spin_unlock_bh(&priv->stats_lock);
 
@@ -486,14 +477,6 @@ static void mlx4_en_get_strings(struct net_device *dev,
 				"rx%d_packets", i);
 			sprintf(data + (index++) * ETH_GSTRING_LEN,
 				"rx%d_bytes", i);
-#ifdef CONFIG_NET_RX_BUSY_POLL
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_napi_yield", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_misses", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_cleaned", i);
-#endif
 		}
 		break;
 	case ETH_SS_PRIV_FLAGS:
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 67e9633ea..21e2c0960 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -69,34 +69,6 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up)
 	return 0;
 }
 
-#ifdef CONFIG_NET_RX_BUSY_POLL
-/* must be called with local_bh_disable()d */
-static int mlx4_en_low_latency_recv(struct napi_struct *napi)
-{
-	struct mlx4_en_cq *cq = container_of(napi, struct mlx4_en_cq, napi);
-	struct net_device *dev = cq->dev;
-	struct mlx4_en_priv *priv = netdev_priv(dev);
-	struct mlx4_en_rx_ring *rx_ring = priv->rx_ring[cq->ring];
-	int done;
-
-	if (!priv->port_up)
-		return LL_FLUSH_FAILED;
-
-	if (!mlx4_en_cq_lock_poll(cq))
-		return LL_FLUSH_BUSY;
-
-	done = mlx4_en_process_rx_cq(dev, cq, 4);
-	if (likely(done))
-		rx_ring->cleaned += done;
-	else
-		rx_ring->misses++;
-
-	mlx4_en_cq_unlock_poll(cq);
-
-	return done;
-}
-#endif	/* CONFIG_NET_RX_BUSY_POLL */
-
 #ifdef CONFIG_RFS_ACCEL
 
 struct mlx4_en_filter {
@@ -1561,8 +1533,6 @@ int mlx4_en_start_port(struct net_device *dev)
 	for (i = 0; i < priv->rx_ring_num; i++) {
 		cq = priv->rx_cq[i];
 
-		mlx4_en_cq_init_lock(cq);
-
 		err = mlx4_en_init_affinity_hint(priv, i);
 		if (err) {
 			en_err(priv, "Failed preparing IRQ affinity hint\n");
@@ -1859,13 +1829,6 @@ void mlx4_en_stop_port(struct net_device *dev, int detach)
 	for (i = 0; i < priv->rx_ring_num; i++) {
 		struct mlx4_en_cq *cq = priv->rx_cq[i];
 
-		local_bh_disable();
-		while (!mlx4_en_cq_lock_napi(cq)) {
-			pr_info("CQ %d locked\n", i);
-			mdelay(1);
-		}
-		local_bh_enable();
-
 		napi_synchronize(&cq->napi);
 		mlx4_en_deactivate_rx_ring(priv, priv->rx_ring[i]);
 		mlx4_en_deactivate_cq(priv, cq);
@@ -2282,7 +2245,7 @@ static int mlx4_en_set_vf_mac(struct net_device *dev, int queue, u8 *mac)
 	struct mlx4_en_dev *mdev = en_priv->mdev;
 	u64 mac_u64 = mlx4_mac_to_u64(mac);
 
-	if (!is_valid_ether_addr(mac))
+	if (is_multicast_ether_addr(mac))
 		return -EINVAL;
 
 	return mlx4_set_vf_mac(mdev->dev, en_priv->port, queue, mac_u64);
@@ -2503,9 +2466,6 @@ static const struct net_device_ops mlx4_netdev_ops = {
 #ifdef CONFIG_RFS_ACCEL
 	.ndo_rx_flow_steer	= mlx4_en_filter_rfs,
 #endif
-#ifdef CONFIG_NET_RX_BUSY_POLL
-	.ndo_busy_poll		= mlx4_en_low_latency_recv,
-#endif
 	.ndo_get_phys_port_id	= mlx4_en_get_phys_port_id,
 #ifdef CONFIG_MLX4_EN_VXLAN
 	.ndo_add_vxlan_port	= mlx4_en_add_vxlan_port,
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_resources.c b/drivers/net/ethernet/mellanox/mlx4/en_resources.c
index 12aab5a65..02e925d6f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_resources.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_resources.c
@@ -58,7 +58,8 @@ void mlx4_en_fill_qp_context(struct mlx4_en_priv *priv, int size, int stride,
 	} else {
 		context->sq_size_stride = ilog2(TXBB_SIZE) - 4;
 	}
-	context->usr_page = cpu_to_be32(mdev->priv_uar.index);
+	context->usr_page = cpu_to_be32(mlx4_to_hw_uar_index(mdev->dev,
+					mdev->priv_uar.index));
 	context->local_qpn = cpu_to_be32(qpn);
 	context->pri_path.ackto = 1 & 0x07;
 	context->pri_path.sched_queue = 0x83 | (priv->port - 1) << 6;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index e7a5000aa..41440b2b2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -873,10 +873,8 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 		 * - TCP/IP (v4)
 		 * - without IP options
 		 * - not an IP fragment
-		 * - no LLS polling in progress
 		 */
-		if (!mlx4_en_cq_busy_polling(cq) &&
-		    (dev->features & NETIF_F_GRO)) {
+		if (dev->features & NETIF_F_GRO) {
 			struct sk_buff *gro_skb = napi_get_frags(&cq->napi);
 			if (!gro_skb)
 				goto next;
@@ -927,7 +925,6 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 						PKT_HASH_TYPE_L3);
 
 			skb_record_rx_queue(gro_skb, cq->ring);
-			skb_mark_napi_id(gro_skb, &cq->napi);
 
 			if (ring->hwtstamp_rx_filter == HWTSTAMP_FILTER_ALL) {
 				timestamp = mlx4_en_get_cqe_ts(cqe);
@@ -990,13 +987,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 					       timestamp);
 		}
 
-		skb_mark_napi_id(skb, &cq->napi);
-
-		if (!mlx4_en_cq_busy_polling(cq))
-			napi_gro_receive(&cq->napi, skb);
-		else
-			netif_receive_skb(skb);
-
+		napi_gro_receive(&cq->napi, skb);
 next:
 		for (nr = 0; nr < priv->num_frags; nr++)
 			mlx4_en_free_frag(priv, frags, nr);
@@ -1038,13 +1029,8 @@ int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	int done;
 
-	if (!mlx4_en_cq_lock_napi(cq))
-		return budget;
-
 	done = mlx4_en_process_rx_cq(dev, cq, budget);
 
-	mlx4_en_cq_unlock_napi(cq);
-
 	/* If we used up all the quota - we're probably not done yet... */
 	if (done == budget) {
 		const struct cpumask *aff;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 4421bf546..e0946ab22 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -213,7 +213,9 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
 	mlx4_en_fill_qp_context(priv, ring->size, ring->stride, 1, 0, ring->qpn,
 				ring->cqn, user_prio, &ring->context);
 	if (ring->bf_alloced)
-		ring->context.usr_page = cpu_to_be32(ring->bf.uar->index);
+		ring->context.usr_page =
+			cpu_to_be32(mlx4_to_hw_uar_index(mdev->dev,
+							 ring->bf.uar->index));
 
 	err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context,
 			       &ring->qp, &ring->qp_state);
diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c
index 603d1c3d3..f61397745 100644
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -151,6 +151,17 @@ void mlx4_gen_slave_eqe(struct work_struct *work)
 	      eqe = next_slave_event_eqe(slave_eq)) {
 		slave = eqe->slave_id;
 
+		if (eqe->type == MLX4_EVENT_TYPE_PORT_CHANGE &&
+		    eqe->subtype == MLX4_PORT_CHANGE_SUBTYPE_DOWN &&
+		    mlx4_is_bonded(dev)) {
+			struct mlx4_port_cap port_cap;
+
+			if (!mlx4_QUERY_PORT(dev, 1, &port_cap) && port_cap.link_state)
+				goto consume;
+
+			if (!mlx4_QUERY_PORT(dev, 2, &port_cap) && port_cap.link_state)
+				goto consume;
+		}
 		/* All active slaves need to receive the event */
 		if (slave == ALL_SLAVES) {
 			for (i = 0; i <= dev->persist->num_vfs; i++) {
@@ -174,6 +185,7 @@ void mlx4_gen_slave_eqe(struct work_struct *work)
 				mlx4_warn(dev, "Failed to generate event for slave %d\n",
 					  slave);
 		}
+consume:
 		++slave_eq->cons;
 	}
 }
@@ -594,7 +606,9 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 					break;
 				for (i = 0; i < dev->persist->num_vfs + 1;
 				     i++) {
-					if (!test_bit(i, slaves_port.slaves))
+					int reported_port = mlx4_is_bonded(dev) ? 1 : mlx4_phys_to_slave_port(dev, i, port);
+
+					if (!test_bit(i, slaves_port.slaves) && !mlx4_is_bonded(dev))
 						continue;
 					if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH) {
 						if (i == mlx4_master_func_num(dev))
@@ -606,7 +620,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 							eqe->event.port_change.port =
 								cpu_to_be32(
 								(be32_to_cpu(eqe->event.port_change.port) & 0xFFFFFFF)
-								| (mlx4_phys_to_slave_port(dev, i, port) << 28));
+								| (reported_port << 28));
 							mlx4_slave_event(dev, i, eqe);
 						}
 					} else {  /* IB port */
@@ -636,7 +650,9 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 					for (i = 0;
 					     i < dev->persist->num_vfs + 1;
 					     i++) {
-						if (!test_bit(i, slaves_port.slaves))
+						int reported_port = mlx4_is_bonded(dev) ? 1 : mlx4_phys_to_slave_port(dev, i, port);
+
+						if (!test_bit(i, slaves_port.slaves) && !mlx4_is_bonded(dev))
 							continue;
 						if (i == mlx4_master_func_num(dev))
 							continue;
@@ -645,7 +661,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 							eqe->event.port_change.port =
 								cpu_to_be32(
 								(be32_to_cpu(eqe->event.port_change.port) & 0xFFFFFFF)
-								| (mlx4_phys_to_slave_port(dev, i, port) << 28));
+								| (reported_port << 28));
 							mlx4_slave_event(dev, i, eqe);
 						}
 					}
@@ -924,9 +940,10 @@ static void __iomem *mlx4_get_eq_uar(struct mlx4_dev *dev, struct mlx4_eq *eq)
 
 	if (!priv->eq_table.uar_map[index]) {
 		priv->eq_table.uar_map[index] =
-			ioremap(pci_resource_start(dev->persist->pdev, 2) +
-				((eq->eqn / 4) << PAGE_SHIFT),
-				PAGE_SIZE);
+			ioremap(
+				pci_resource_start(dev->persist->pdev, 2) +
+				((eq->eqn / 4) << (dev->uar_page_shift)),
+				(1 << (dev->uar_page_shift)));
 		if (!priv->eq_table.uar_map[index]) {
 			mlx4_err(dev, "Couldn't map EQ doorbell for EQN 0x%06x\n",
 				 eq->eqn);
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 90db94e83..d66c690a8 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -157,6 +157,7 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags)
 		[29] = "802.1ad offload support",
 		[31] = "Modifying loopback source checks using UPDATE_QP support",
 		[32] = "Loopback source checks support",
+		[33] = "RoCEv2 support"
 	};
 	int i;
 
@@ -626,6 +627,8 @@ out:
 	return err;
 }
 
+static void disable_unsupported_roce_caps(void *buf);
+
 int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 {
 	struct mlx4_cmd_mailbox *mailbox;
@@ -738,6 +741,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	if (err)
 		goto out;
 
+	if (mlx4_is_mfunc(dev))
+		disable_unsupported_roce_caps(outbox);
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_RSVD_QP_OFFSET);
 	dev_cap->reserved_qps = 1 << (field & 0xf);
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_MAX_QP_OFFSET);
@@ -905,6 +910,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_EQE_STRIDE;
 	MLX4_GET(dev_cap->bmme_flags, outbox,
 		 QUERY_DEV_CAP_BMME_FLAGS_OFFSET);
+	if (dev_cap->bmme_flags & MLX4_FLAG_ROCE_V1_V2)
+		dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_ROCE_V1_V2;
 	if (dev_cap->bmme_flags & MLX4_FLAG_PORT_REMAP)
 		dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_PORT_REMAP;
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_CONFIG_DEV_OFFSET);
@@ -1104,6 +1111,7 @@ int mlx4_QUERY_PORT(struct mlx4_dev *dev, int port, struct mlx4_port_cap *port_c
 			goto out;
 
 		MLX4_GET(field, outbox, QUERY_PORT_SUPPORTED_TYPE_OFFSET);
+		port_cap->link_state = (field & 0x80) >> 7;
 		port_cap->supported_port_types = field & 3;
 		port_cap->suggested_type = (field >> 3) & 1;
 		port_cap->default_sense = (field >> 4) & 1;
@@ -1160,6 +1168,7 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave,
 	if (err)
 		return err;
 
+	disable_unsupported_roce_caps(outbox->buf);
 	/* add port mng change event capability and disable mw type 1
 	 * unconditionally to slaves
 	 */
@@ -1257,6 +1266,21 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave,
 	return 0;
 }
 
+static void disable_unsupported_roce_caps(void *buf)
+{
+	u32 flags;
+
+	MLX4_GET(flags, buf, QUERY_DEV_CAP_EXT_FLAGS_OFFSET);
+	flags &= ~(1UL << 31);
+	MLX4_PUT(buf, flags, QUERY_DEV_CAP_EXT_FLAGS_OFFSET);
+	MLX4_GET(flags, buf, QUERY_DEV_CAP_EXT_2_FLAGS_OFFSET);
+	flags &= ~(1UL << 24);
+	MLX4_PUT(buf, flags, QUERY_DEV_CAP_EXT_2_FLAGS_OFFSET);
+	MLX4_GET(flags, buf, QUERY_DEV_CAP_BMME_FLAGS_OFFSET);
+	flags &= ~(MLX4_FLAG_ROCE_V1_V2);
+	MLX4_PUT(buf, flags, QUERY_DEV_CAP_BMME_FLAGS_OFFSET);
+}
+
 int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 			    struct mlx4_vhcr *vhcr,
 			    struct mlx4_cmd_mailbox *inbox,
@@ -1310,6 +1334,15 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 			port_type |= MLX4_PORT_LINK_UP_MASK;
 		else if (IFLA_VF_LINK_STATE_DISABLE == admin_link_state)
 			port_type &= ~MLX4_PORT_LINK_UP_MASK;
+		else if (IFLA_VF_LINK_STATE_AUTO == admin_link_state && mlx4_is_bonded(dev)) {
+			int other_port = (port == 1) ? 2 : 1;
+			struct mlx4_port_cap port_cap;
+
+			err = mlx4_QUERY_PORT(dev, other_port, &port_cap);
+			if (err)
+				goto out;
+			port_type |= (port_cap.link_state << 7);
+		}
 
 		MLX4_PUT(outbox->buf, port_type,
 			 QUERY_PORT_SUPPORTED_TYPE_OFFSET);
@@ -1325,7 +1358,7 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 		MLX4_PUT(outbox->buf, short_field,
 			 QUERY_PORT_CUR_MAX_PKEY_OFFSET);
 	}
-
+out:
 	return err;
 }
 
@@ -2229,7 +2262,8 @@ struct mlx4_config_dev {
 	__be32	rsvd1[3];
 	__be16	vxlan_udp_dport;
 	__be16	rsvd2;
-	__be32	rsvd3;
+	__be16  roce_v2_entropy;
+	__be16  roce_v2_udp_dport;
 	__be32	roce_flags;
 	__be32	rsvd4[25];
 	__be16	rsvd5;
@@ -2238,6 +2272,7 @@ struct mlx4_config_dev {
 };
 
 #define MLX4_VXLAN_UDP_DPORT (1 << 0)
+#define MLX4_ROCE_V2_UDP_DPORT BIT(3)
 #define MLX4_DISABLE_RX_PORT BIT(18)
 
 static int mlx4_CONFIG_DEV_set(struct mlx4_dev *dev, struct mlx4_config_dev *config_dev)
@@ -2355,6 +2390,18 @@ int mlx4_disable_rx_port_check(struct mlx4_dev *dev, bool dis)
 	return mlx4_CONFIG_DEV_set(dev, &config_dev);
 }
 
+int mlx4_config_roce_v2_port(struct mlx4_dev *dev, u16 udp_port)
+{
+	struct mlx4_config_dev config_dev;
+
+	memset(&config_dev, 0, sizeof(config_dev));
+	config_dev.update_flags    = cpu_to_be32(MLX4_ROCE_V2_UDP_DPORT);
+	config_dev.roce_v2_udp_dport = cpu_to_be16(udp_port);
+
+	return mlx4_CONFIG_DEV_set(dev, &config_dev);
+}
+EXPORT_SYMBOL_GPL(mlx4_config_roce_v2_port);
+
 int mlx4_virt2phy_port_map(struct mlx4_dev *dev, u32 port1, u32 port2)
 {
 	struct mlx4_cmd_mailbox *mailbox;
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h
index 08de5555c..7ea258af6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.h
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.h
@@ -44,6 +44,7 @@ struct mlx4_mod_stat_cfg {
 };
 
 struct mlx4_port_cap {
+	u8  link_state;
 	u8  supported_port_types;
 	u8  suggested_type;
 	u8  default_sense;
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 31c491e02..f8674ae62 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -168,6 +168,20 @@ struct mlx4_port_config {
 
 static atomic_t pf_loading = ATOMIC_INIT(0);
 
+static inline void mlx4_set_num_reserved_uars(struct mlx4_dev *dev,
+					      struct mlx4_dev_cap *dev_cap)
+{
+	/* The reserved_uars is calculated by system page size unit.
+	 * Therefore, adjustment is added when the uar page size is less
+	 * than the system page size
+	 */
+	dev->caps.reserved_uars	=
+		max_t(int,
+		      mlx4_get_num_reserved_uar(dev),
+		      dev_cap->reserved_uars /
+			(1 << (PAGE_SHIFT - dev->uar_page_shift)));
+}
+
 int mlx4_check_port_params(struct mlx4_dev *dev,
 			   enum mlx4_port_type *port_type)
 {
@@ -386,8 +400,6 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.reserved_mtts      = dev_cap->reserved_mtts;
 	dev->caps.reserved_mrws	     = dev_cap->reserved_mrws;
 
-	/* The first 128 UARs are used for EQ doorbells */
-	dev->caps.reserved_uars	     = max_t(int, 128, dev_cap->reserved_uars);
 	dev->caps.reserved_pds	     = dev_cap->reserved_pds;
 	dev->caps.reserved_xrcds     = (dev->caps.flags & MLX4_DEV_CAP_FLAG_XRC) ?
 					dev_cap->reserved_xrcds : 0;
@@ -405,6 +417,15 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 	dev->caps.max_rss_tbl_sz     = dev_cap->max_rss_tbl_sz;
 
+	/* Save uar page shift */
+	if (!mlx4_is_slave(dev)) {
+		/* Virtual PCI function needs to determine UAR page size from
+		 * firmware. Only master PCI function can set the uar page size
+		 */
+		dev->uar_page_shift = DEFAULT_UAR_PAGE_SHIFT;
+		mlx4_set_num_reserved_uars(dev, dev_cap);
+	}
+
 	if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PHV_EN) {
 		struct mlx4_init_hca_param hca_param;
 
@@ -815,16 +836,25 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 		return -ENODEV;
 	}
 
-	/* slave gets uar page size from QUERY_HCA fw command */
-	dev->caps.uar_page_size = 1 << (hca_param.uar_page_sz + 12);
+	/* Set uar_page_shift for VF */
+	dev->uar_page_shift = hca_param.uar_page_sz + 12;
 
-	/* TODO: relax this assumption */
-	if (dev->caps.uar_page_size != PAGE_SIZE) {
-		mlx4_err(dev, "UAR size:%d != kernel PAGE_SIZE of %ld\n",
-			 dev->caps.uar_page_size, PAGE_SIZE);
-		return -ENODEV;
+	/* Make sure the master uar page size is valid */
+	if (dev->uar_page_shift > PAGE_SHIFT) {
+		mlx4_err(dev,
+			 "Invalid configuration: uar page size is larger than system page size\n");
+		return  -ENODEV;
 	}
 
+	/* Set reserved_uars based on the uar_page_shift */
+	mlx4_set_num_reserved_uars(dev, &dev_cap);
+
+	/* Although uar page size in FW differs from system page size,
+	 * upper software layers (mlx4_ib, mlx4_en and part of mlx4_core)
+	 * still works with assumption that uar page size == system page size
+	 */
+	dev->caps.uar_page_size = PAGE_SIZE;
+
 	memset(&func_cap, 0, sizeof(func_cap));
 	err = mlx4_QUERY_FUNC_CAP(dev, 0, &func_cap);
 	if (err) {
@@ -1221,6 +1251,84 @@ err_set_port:
 	return err ? err : count;
 }
 
+/* bond for multi-function device */
+#define MAX_MF_BOND_ALLOWED_SLAVES 63
+static int mlx4_mf_bond(struct mlx4_dev *dev)
+{
+	int err = 0;
+	int nvfs;
+	struct mlx4_slaves_pport slaves_port1;
+	struct mlx4_slaves_pport slaves_port2;
+	DECLARE_BITMAP(slaves_port_1_2, MLX4_MFUNC_MAX);
+
+	slaves_port1 = mlx4_phys_to_slaves_pport(dev, 1);
+	slaves_port2 = mlx4_phys_to_slaves_pport(dev, 2);
+	bitmap_and(slaves_port_1_2,
+		   slaves_port1.slaves, slaves_port2.slaves,
+		   dev->persist->num_vfs + 1);
+
+	/* only single port vfs are allowed */
+	if (bitmap_weight(slaves_port_1_2, dev->persist->num_vfs + 1) > 1) {
+		mlx4_warn(dev, "HA mode unsupported for dual ported VFs\n");
+		return -EINVAL;
+	}
+
+	/* number of virtual functions is number of total functions minus one
+	 * physical function for each port.
+	 */
+	nvfs = bitmap_weight(slaves_port1.slaves, dev->persist->num_vfs + 1) +
+		bitmap_weight(slaves_port2.slaves, dev->persist->num_vfs + 1) - 2;
+
+	/* limit on maximum allowed VFs */
+	if (nvfs > MAX_MF_BOND_ALLOWED_SLAVES) {
+		mlx4_warn(dev, "HA mode is not supported for %d VFs (max %d are allowed)\n",
+			  nvfs, MAX_MF_BOND_ALLOWED_SLAVES);
+		return -EINVAL;
+	}
+
+	if (dev->caps.steering_mode != MLX4_STEERING_MODE_DEVICE_MANAGED) {
+		mlx4_warn(dev, "HA mode unsupported for NON DMFS steering\n");
+		return -EINVAL;
+	}
+
+	err = mlx4_bond_mac_table(dev);
+	if (err)
+		return err;
+	err = mlx4_bond_vlan_table(dev);
+	if (err)
+		goto err1;
+	err = mlx4_bond_fs_rules(dev);
+	if (err)
+		goto err2;
+
+	return 0;
+err2:
+	(void)mlx4_unbond_vlan_table(dev);
+err1:
+	(void)mlx4_unbond_mac_table(dev);
+	return err;
+}
+
+static int mlx4_mf_unbond(struct mlx4_dev *dev)
+{
+	int ret, ret1;
+
+	ret = mlx4_unbond_fs_rules(dev);
+	if (ret)
+		mlx4_warn(dev, "multifunction unbond for flow rules failedi (%d)\n", ret);
+	ret1 = mlx4_unbond_mac_table(dev);
+	if (ret1) {
+		mlx4_warn(dev, "multifunction unbond for MAC table failed (%d)\n", ret1);
+		ret = ret1;
+	}
+	ret1 = mlx4_unbond_vlan_table(dev);
+	if (ret1) {
+		mlx4_warn(dev, "multifunction unbond for VLAN table failed (%d)\n", ret1);
+		ret = ret1;
+	}
+	return ret;
+}
+
 int mlx4_bond(struct mlx4_dev *dev)
 {
 	int ret = 0;
@@ -1228,16 +1336,23 @@ int mlx4_bond(struct mlx4_dev *dev)
 
 	mutex_lock(&priv->bond_mutex);
 
-	if (!mlx4_is_bonded(dev))
+	if (!mlx4_is_bonded(dev)) {
 		ret = mlx4_do_bond(dev, true);
-	else
-		ret = 0;
+		if (ret)
+			mlx4_err(dev, "Failed to bond device: %d\n", ret);
+		if (!ret && mlx4_is_master(dev)) {
+			ret = mlx4_mf_bond(dev);
+			if (ret) {
+				mlx4_err(dev, "bond for multifunction failed\n");
+				mlx4_do_bond(dev, false);
+			}
+		}
+	}
 
 	mutex_unlock(&priv->bond_mutex);
-	if (ret)
-		mlx4_err(dev, "Failed to bond device: %d\n", ret);
-	else
+	if (!ret)
 		mlx4_dbg(dev, "Device is bonded\n");
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(mlx4_bond);
@@ -1249,14 +1364,24 @@ int mlx4_unbond(struct mlx4_dev *dev)
 
 	mutex_lock(&priv->bond_mutex);
 
-	if (mlx4_is_bonded(dev))
+	if (mlx4_is_bonded(dev)) {
+		int ret2 = 0;
+
 		ret = mlx4_do_bond(dev, false);
+		if (ret)
+			mlx4_err(dev, "Failed to unbond device: %d\n", ret);
+		if (mlx4_is_master(dev))
+			ret2 = mlx4_mf_unbond(dev);
+		if (ret2) {
+			mlx4_warn(dev, "Failed to unbond device for multifunction (%d)\n", ret2);
+			ret = ret2;
+		}
+	}
 
 	mutex_unlock(&priv->bond_mutex);
-	if (ret)
-		mlx4_err(dev, "Failed to unbond device: %d\n", ret);
-	else
+	if (!ret)
 		mlx4_dbg(dev, "Device is unbonded\n");
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(mlx4_unbond);
@@ -2092,8 +2217,12 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 		dev->caps.max_fmr_maps = (1 << (32 - ilog2(dev->caps.num_mpts))) - 1;
 
-		init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
-		init_hca.uar_page_sz = PAGE_SHIFT - 12;
+		/* Always set UAR page size 4KB, set log_uar_sz accordingly */
+		init_hca.log_uar_sz = ilog2(dev->caps.num_uars) +
+				      PAGE_SHIFT -
+				      DEFAULT_UAR_PAGE_SHIFT;
+		init_hca.uar_page_sz = DEFAULT_UAR_PAGE_SHIFT - 12;
+
 		init_hca.mw_enabled = 0;
 		if (dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW ||
 		    dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN)
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index e1cf9036a..7baef52db 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -736,6 +736,7 @@ struct mlx4_catas_err {
 struct mlx4_mac_table {
 	__be64			entries[MLX4_MAX_MAC_NUM];
 	int			refs[MLX4_MAX_MAC_NUM];
+	bool			is_dup[MLX4_MAX_MAC_NUM];
 	struct mutex		mutex;
 	int			total;
 	int			max;
@@ -758,6 +759,7 @@ struct mlx4_roce_gid_table {
 struct mlx4_vlan_table {
 	__be32			entries[MLX4_MAX_VLAN_NUM];
 	int			refs[MLX4_MAX_VLAN_NUM];
+	int			is_dup[MLX4_MAX_VLAN_NUM];
 	struct mutex		mutex;
 	int			total;
 	int			max;
@@ -778,7 +780,10 @@ struct mlx4_set_port_general_context {
 	u16 reserved1;
 	u8 v_ignore_fcs;
 	u8 flags;
-	u8 ignore_fcs;
+	union {
+		u8 ignore_fcs;
+		u8 roce_mode;
+	};
 	u8 reserved2;
 	__be16 mtu;
 	u8 pptx;
@@ -1225,6 +1230,10 @@ void mlx4_init_roce_gid_table(struct mlx4_dev *dev,
 			      struct mlx4_roce_gid_table *table);
 void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan);
 int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, int *index);
+int mlx4_bond_vlan_table(struct mlx4_dev *dev);
+int mlx4_unbond_vlan_table(struct mlx4_dev *dev);
+int mlx4_bond_mac_table(struct mlx4_dev *dev);
+int mlx4_unbond_mac_table(struct mlx4_dev *dev);
 
 int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz);
 /* resource tracker functions*/
@@ -1385,6 +1394,8 @@ int mlx4_get_slave_num_gids(struct mlx4_dev *dev, int slave, int port);
 int mlx4_get_vf_indx(struct mlx4_dev *dev, int slave);
 int mlx4_config_mad_demux(struct mlx4_dev *dev);
 int mlx4_do_bond(struct mlx4_dev *dev, bool enable);
+int mlx4_bond_fs_rules(struct mlx4_dev *dev);
+int mlx4_unbond_fs_rules(struct mlx4_dev *dev);
 
 enum mlx4_zone_flags {
 	MLX4_ZONE_ALLOW_ALLOC_FROM_LOWER_PRIO	= 1UL << 0,
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index c41f15102..35de7d2e6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -320,11 +320,6 @@ struct mlx4_en_rx_ring {
 	void *rx_info;
 	unsigned long bytes;
 	unsigned long packets;
-#ifdef CONFIG_NET_RX_BUSY_POLL
-	unsigned long yields;
-	unsigned long misses;
-	unsigned long cleaned;
-#endif
 	unsigned long csum_ok;
 	unsigned long csum_none;
 	unsigned long csum_complete;
@@ -347,18 +342,6 @@ struct mlx4_en_cq {
 	struct mlx4_cqe *buf;
 #define MLX4_EN_OPCODE_ERROR	0x1e
 
-#ifdef CONFIG_NET_RX_BUSY_POLL
-	unsigned int state;
-#define MLX4_EN_CQ_STATE_IDLE        0
-#define MLX4_EN_CQ_STATE_NAPI     1    /* NAPI owns this CQ */
-#define MLX4_EN_CQ_STATE_POLL     2    /* poll owns this CQ */
-#define MLX4_CQ_LOCKED (MLX4_EN_CQ_STATE_NAPI | MLX4_EN_CQ_STATE_POLL)
-#define MLX4_EN_CQ_STATE_NAPI_YIELD  4    /* NAPI yielded this CQ */
-#define MLX4_EN_CQ_STATE_POLL_YIELD  8    /* poll yielded this CQ */
-#define CQ_YIELD (MLX4_EN_CQ_STATE_NAPI_YIELD | MLX4_EN_CQ_STATE_POLL_YIELD)
-#define CQ_USER_PEND (MLX4_EN_CQ_STATE_POLL | MLX4_EN_CQ_STATE_POLL_YIELD)
-	spinlock_t poll_lock; /* protects from LLS/napi conflicts */
-#endif  /* CONFIG_NET_RX_BUSY_POLL */
 	struct irq_desc *irq_desc;
 };
 
@@ -622,115 +605,6 @@ static inline struct mlx4_cqe *mlx4_en_get_cqe(void *buf, int idx, int cqe_sz)
 	return buf + idx * cqe_sz;
 }
 
-#ifdef CONFIG_NET_RX_BUSY_POLL
-static inline void mlx4_en_cq_init_lock(struct mlx4_en_cq *cq)
-{
-	spin_lock_init(&cq->poll_lock);
-	cq->state = MLX4_EN_CQ_STATE_IDLE;
-}
-
-/* called from the device poll rutine to get ownership of a cq */
-static inline bool mlx4_en_cq_lock_napi(struct mlx4_en_cq *cq)
-{
-	int rc = true;
-	spin_lock(&cq->poll_lock);
-	if (cq->state & MLX4_CQ_LOCKED) {
-		WARN_ON(cq->state & MLX4_EN_CQ_STATE_NAPI);
-		cq->state |= MLX4_EN_CQ_STATE_NAPI_YIELD;
-		rc = false;
-	} else
-		/* we don't care if someone yielded */
-		cq->state = MLX4_EN_CQ_STATE_NAPI;
-	spin_unlock(&cq->poll_lock);
-	return rc;
-}
-
-/* returns true is someone tried to get the cq while napi had it */
-static inline bool mlx4_en_cq_unlock_napi(struct mlx4_en_cq *cq)
-{
-	int rc = false;
-	spin_lock(&cq->poll_lock);
-	WARN_ON(cq->state & (MLX4_EN_CQ_STATE_POLL |
-			       MLX4_EN_CQ_STATE_NAPI_YIELD));
-
-	if (cq->state & MLX4_EN_CQ_STATE_POLL_YIELD)
-		rc = true;
-	cq->state = MLX4_EN_CQ_STATE_IDLE;
-	spin_unlock(&cq->poll_lock);
-	return rc;
-}
-
-/* called from mlx4_en_low_latency_poll() */
-static inline bool mlx4_en_cq_lock_poll(struct mlx4_en_cq *cq)
-{
-	int rc = true;
-	spin_lock_bh(&cq->poll_lock);
-	if ((cq->state & MLX4_CQ_LOCKED)) {
-		struct net_device *dev = cq->dev;
-		struct mlx4_en_priv *priv = netdev_priv(dev);
-		struct mlx4_en_rx_ring *rx_ring = priv->rx_ring[cq->ring];
-
-		cq->state |= MLX4_EN_CQ_STATE_POLL_YIELD;
-		rc = false;
-		rx_ring->yields++;
-	} else
-		/* preserve yield marks */
-		cq->state |= MLX4_EN_CQ_STATE_POLL;
-	spin_unlock_bh(&cq->poll_lock);
-	return rc;
-}
-
-/* returns true if someone tried to get the cq while it was locked */
-static inline bool mlx4_en_cq_unlock_poll(struct mlx4_en_cq *cq)
-{
-	int rc = false;
-	spin_lock_bh(&cq->poll_lock);
-	WARN_ON(cq->state & (MLX4_EN_CQ_STATE_NAPI));
-
-	if (cq->state & MLX4_EN_CQ_STATE_POLL_YIELD)
-		rc = true;
-	cq->state = MLX4_EN_CQ_STATE_IDLE;
-	spin_unlock_bh(&cq->poll_lock);
-	return rc;
-}
-
-/* true if a socket is polling, even if it did not get the lock */
-static inline bool mlx4_en_cq_busy_polling(struct mlx4_en_cq *cq)
-{
-	WARN_ON(!(cq->state & MLX4_CQ_LOCKED));
-	return cq->state & CQ_USER_PEND;
-}
-#else
-static inline void mlx4_en_cq_init_lock(struct mlx4_en_cq *cq)
-{
-}
-
-static inline bool mlx4_en_cq_lock_napi(struct mlx4_en_cq *cq)
-{
-	return true;
-}
-
-static inline bool mlx4_en_cq_unlock_napi(struct mlx4_en_cq *cq)
-{
-	return false;
-}
-
-static inline bool mlx4_en_cq_lock_poll(struct mlx4_en_cq *cq)
-{
-	return false;
-}
-
-static inline bool mlx4_en_cq_unlock_poll(struct mlx4_en_cq *cq)
-{
-	return false;
-}
-
-static inline bool mlx4_en_cq_busy_polling(struct mlx4_en_cq *cq)
-{
-	return false;
-}
-#endif /* CONFIG_NET_RX_BUSY_POLL */
-
 #define MLX4_EN_WOL_DO_MODIFY (1ULL << 63)
 
 void mlx4_en_update_loopback_state(struct net_device *dev,
diff --git a/drivers/net/ethernet/mellanox/mlx4/pd.c b/drivers/net/ethernet/mellanox/mlx4/pd.c
index 609c59dc8..b3cc3ab63 100644
--- a/drivers/net/ethernet/mellanox/mlx4/pd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/pd.c
@@ -269,9 +269,15 @@ EXPORT_SYMBOL_GPL(mlx4_bf_free);
 
 int mlx4_init_uar_table(struct mlx4_dev *dev)
 {
-	if (dev->caps.num_uars <= 128) {
-		mlx4_err(dev, "Only %d UAR pages (need more than 128)\n",
-			 dev->caps.num_uars);
+	int num_reserved_uar = mlx4_get_num_reserved_uar(dev);
+
+	mlx4_dbg(dev, "uar_page_shift = %d", dev->uar_page_shift);
+	mlx4_dbg(dev, "Effective reserved_uars=%d", dev->caps.reserved_uars);
+
+	if (dev->caps.num_uars <= num_reserved_uar) {
+		mlx4_err(
+			dev, "Only %d UAR pages (need more than %d)\n",
+			dev->caps.num_uars, num_reserved_uar);
 		mlx4_err(dev, "Increase firmware log2_uar_bar_megabytes?\n");
 		return -ENODEV;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c
index c2b21313d..211c65087 100644
--- a/drivers/net/ethernet/mellanox/mlx4/port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/port.c
@@ -61,6 +61,7 @@ void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table)
 	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
 		table->entries[i] = 0;
 		table->refs[i]	 = 0;
+		table->is_dup[i] = false;
 	}
 	table->max   = 1 << dev->caps.log_num_macs;
 	table->total = 0;
@@ -74,6 +75,7 @@ void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table)
 	for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) {
 		table->entries[i] = 0;
 		table->refs[i]	 = 0;
+		table->is_dup[i] = false;
 	}
 	table->max   = (1 << dev->caps.log_num_vlans) - MLX4_VLAN_REGULAR;
 	table->total = 0;
@@ -159,21 +161,94 @@ int mlx4_find_cached_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *idx)
 }
 EXPORT_SYMBOL_GPL(mlx4_find_cached_mac);
 
+static bool mlx4_need_mf_bond(struct mlx4_dev *dev)
+{
+	int i, num_eth_ports = 0;
+
+	if (!mlx4_is_mfunc(dev))
+		return false;
+	mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH)
+		++num_eth_ports;
+
+	return (num_eth_ports ==  2) ? true : false;
+}
+
 int __mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 {
 	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
 	struct mlx4_mac_table *table = &info->mac_table;
 	int i, err = 0;
 	int free = -1;
+	int free_for_dup = -1;
+	bool dup = mlx4_is_mf_bonded(dev);
+	u8 dup_port = (port == 1) ? 2 : 1;
+	struct mlx4_mac_table *dup_table = &mlx4_priv(dev)->port[dup_port].mac_table;
+	bool need_mf_bond = mlx4_need_mf_bond(dev);
+	bool can_mf_bond = true;
+
+	mlx4_dbg(dev, "Registering MAC: 0x%llx for port %d %s duplicate\n",
+		 (unsigned long long)mac, port,
+		 dup ? "with" : "without");
+
+	if (need_mf_bond) {
+		if (port == 1) {
+			mutex_lock(&table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
+		} else {
+			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
+		}
+	} else {
+		mutex_lock(&table->mutex);
+	}
+
+	if (need_mf_bond) {
+		int index_at_port = -1;
+		int index_at_dup_port = -1;
 
-	mlx4_dbg(dev, "Registering MAC: 0x%llx for port %d\n",
-		 (unsigned long long) mac, port);
+		for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+			if (((MLX4_MAC_MASK & mac) == (MLX4_MAC_MASK & be64_to_cpu(table->entries[i]))))
+				index_at_port = i;
+			if (((MLX4_MAC_MASK & mac) == (MLX4_MAC_MASK & be64_to_cpu(dup_table->entries[i]))))
+				index_at_dup_port = i;
+		}
+
+		/* check that same mac is not in the tables at different indices */
+		if ((index_at_port != index_at_dup_port) &&
+		    (index_at_port >= 0) &&
+		    (index_at_dup_port >= 0))
+			can_mf_bond = false;
+
+		/* If the mac is already in the primary table, the slot must be
+		 * available in the duplicate table as well.
+		 */
+		if (index_at_port >= 0 && index_at_dup_port < 0 &&
+		    dup_table->refs[index_at_port]) {
+			can_mf_bond = false;
+		}
+		/* If the mac is already in the duplicate table, check that the
+		 * corresponding index is not occupied in the primary table, or
+		 * the primary table already contains the mac at the same index.
+		 * Otherwise, you cannot bond (primary contains a different mac
+		 * at that index).
+		 */
+		if (index_at_dup_port >= 0) {
+			if (!table->refs[index_at_dup_port] ||
+			    ((MLX4_MAC_MASK & mac) == (MLX4_MAC_MASK & be64_to_cpu(table->entries[index_at_dup_port]))))
+				free_for_dup = index_at_dup_port;
+			else
+				can_mf_bond = false;
+		}
+	}
 
-	mutex_lock(&table->mutex);
 	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
 		if (!table->refs[i]) {
 			if (free < 0)
 				free = i;
+			if (free_for_dup < 0 && need_mf_bond && can_mf_bond) {
+				if (!dup_table->refs[i])
+					free_for_dup = i;
+			}
 			continue;
 		}
 
@@ -182,10 +257,30 @@ int __mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 			/* MAC already registered, increment ref count */
 			err = i;
 			++table->refs[i];
+			if (dup) {
+				u64 dup_mac = MLX4_MAC_MASK & be64_to_cpu(dup_table->entries[i]);
+
+				if (dup_mac != mac || !dup_table->is_dup[i]) {
+					mlx4_warn(dev, "register mac: expect duplicate mac 0x%llx on port %d index %d\n",
+						  mac, dup_port, i);
+				}
+			}
 			goto out;
 		}
 	}
 
+	if (need_mf_bond && (free_for_dup < 0)) {
+		if (dup) {
+			mlx4_warn(dev, "Fail to allocate duplicate MAC table entry\n");
+			mlx4_warn(dev, "High Availability for virtual functions may not work as expected\n");
+			dup = false;
+		}
+		can_mf_bond = false;
+	}
+
+	if (need_mf_bond && can_mf_bond)
+		free = free_for_dup;
+
 	mlx4_dbg(dev, "Free MAC index is %d\n", free);
 
 	if (table->total == table->max) {
@@ -205,10 +300,35 @@ int __mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 		goto out;
 	}
 	table->refs[free] = 1;
-	err = free;
+	table->is_dup[free] = false;
 	++table->total;
+	if (dup) {
+		dup_table->refs[free] = 0;
+		dup_table->is_dup[free] = true;
+		dup_table->entries[free] = cpu_to_be64(mac | MLX4_MAC_VALID);
+
+		err = mlx4_set_port_mac_table(dev, dup_port, dup_table->entries);
+		if (unlikely(err)) {
+			mlx4_warn(dev, "Failed adding duplicate mac: 0x%llx\n", mac);
+			dup_table->is_dup[free] = false;
+			dup_table->entries[free] = 0;
+			goto out;
+		}
+		++dup_table->total;
+	}
+	err = free;
 out:
-	mutex_unlock(&table->mutex);
+	if (need_mf_bond) {
+		if (port == 2) {
+			mutex_unlock(&table->mutex);
+			mutex_unlock(&dup_table->mutex);
+		} else {
+			mutex_unlock(&dup_table->mutex);
+			mutex_unlock(&table->mutex);
+		}
+	} else {
+		mutex_unlock(&table->mutex);
+	}
 	return err;
 }
 EXPORT_SYMBOL_GPL(__mlx4_register_mac);
@@ -255,6 +375,9 @@ void __mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 	struct mlx4_port_info *info;
 	struct mlx4_mac_table *table;
 	int index;
+	bool dup = mlx4_is_mf_bonded(dev);
+	u8 dup_port = (port == 1) ? 2 : 1;
+	struct mlx4_mac_table *dup_table = &mlx4_priv(dev)->port[dup_port].mac_table;
 
 	if (port < 1 || port > dev->caps.num_ports) {
 		mlx4_warn(dev, "invalid port number (%d), aborting...\n", port);
@@ -262,22 +385,59 @@ void __mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 	}
 	info = &mlx4_priv(dev)->port[port];
 	table = &info->mac_table;
-	mutex_lock(&table->mutex);
+
+	if (dup) {
+		if (port == 1) {
+			mutex_lock(&table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
+		} else {
+			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
+		}
+	} else {
+		mutex_lock(&table->mutex);
+	}
+
 	index = find_index(dev, table, mac);
 
 	if (validate_index(dev, table, index))
 		goto out;
-	if (--table->refs[index]) {
+
+	if (--table->refs[index] || table->is_dup[index]) {
 		mlx4_dbg(dev, "Have more references for index %d, no need to modify mac table\n",
 			 index);
+		if (!table->refs[index])
+			dup_table->is_dup[index] = false;
 		goto out;
 	}
 
 	table->entries[index] = 0;
-	mlx4_set_port_mac_table(dev, port, table->entries);
+	if (mlx4_set_port_mac_table(dev, port, table->entries))
+		mlx4_warn(dev, "Fail to set mac in port %d during unregister\n", port);
 	--table->total;
+
+	if (dup) {
+		dup_table->is_dup[index] = false;
+		if (dup_table->refs[index])
+			goto out;
+		dup_table->entries[index] = 0;
+		if (mlx4_set_port_mac_table(dev, dup_port, dup_table->entries))
+			mlx4_warn(dev, "Fail to set mac in duplicate port %d during unregister\n", dup_port);
+
+		--table->total;
+	}
 out:
-	mutex_unlock(&table->mutex);
+	if (dup) {
+		if (port == 2) {
+			mutex_unlock(&table->mutex);
+			mutex_unlock(&dup_table->mutex);
+		} else {
+			mutex_unlock(&dup_table->mutex);
+			mutex_unlock(&table->mutex);
+		}
+	} else {
+		mutex_unlock(&table->mutex);
+	}
 }
 EXPORT_SYMBOL_GPL(__mlx4_unregister_mac);
 
@@ -311,9 +471,22 @@ int __mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac)
 	struct mlx4_mac_table *table = &info->mac_table;
 	int index = qpn - info->base_qpn;
 	int err = 0;
+	bool dup = mlx4_is_mf_bonded(dev);
+	u8 dup_port = (port == 1) ? 2 : 1;
+	struct mlx4_mac_table *dup_table = &mlx4_priv(dev)->port[dup_port].mac_table;
 
 	/* CX1 doesn't support multi-functions */
-	mutex_lock(&table->mutex);
+	if (dup) {
+		if (port == 1) {
+			mutex_lock(&table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
+		} else {
+			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
+		}
+	} else {
+		mutex_lock(&table->mutex);
+	}
 
 	err = validate_index(dev, table, index);
 	if (err)
@@ -326,9 +499,30 @@ int __mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac)
 		mlx4_err(dev, "Failed adding MAC: 0x%llx\n",
 			 (unsigned long long) new_mac);
 		table->entries[index] = 0;
+	} else {
+		if (dup) {
+			dup_table->entries[index] = cpu_to_be64(new_mac | MLX4_MAC_VALID);
+
+			err = mlx4_set_port_mac_table(dev, dup_port, dup_table->entries);
+			if (unlikely(err)) {
+				mlx4_err(dev, "Failed adding duplicate MAC: 0x%llx\n",
+					 (unsigned long long)new_mac);
+				dup_table->entries[index] = 0;
+			}
+		}
 	}
 out:
-	mutex_unlock(&table->mutex);
+	if (dup) {
+		if (port == 2) {
+			mutex_unlock(&table->mutex);
+			mutex_unlock(&dup_table->mutex);
+		} else {
+			mutex_unlock(&dup_table->mutex);
+			mutex_unlock(&table->mutex);
+		}
+	} else {
+		mutex_unlock(&table->mutex);
+	}
 	return err;
 }
 EXPORT_SYMBOL_GPL(__mlx4_replace_mac);
@@ -380,8 +574,28 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan,
 	struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table;
 	int i, err = 0;
 	int free = -1;
-
-	mutex_lock(&table->mutex);
+	int free_for_dup = -1;
+	bool dup = mlx4_is_mf_bonded(dev);
+	u8 dup_port = (port == 1) ? 2 : 1;
+	struct mlx4_vlan_table *dup_table = &mlx4_priv(dev)->port[dup_port].vlan_table;
+	bool need_mf_bond = mlx4_need_mf_bond(dev);
+	bool can_mf_bond = true;
+
+	mlx4_dbg(dev, "Registering VLAN: %d for port %d %s duplicate\n",
+		 vlan, port,
+		 dup ? "with" : "without");
+
+	if (need_mf_bond) {
+		if (port == 1) {
+			mutex_lock(&table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
+		} else {
+			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
+		}
+	} else {
+		mutex_lock(&table->mutex);
+	}
 
 	if (table->total == table->max) {
 		/* No free vlan entries */
@@ -389,22 +603,85 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan,
 		goto out;
 	}
 
+	if (need_mf_bond) {
+		int index_at_port = -1;
+		int index_at_dup_port = -1;
+
+		for (i = MLX4_VLAN_REGULAR; i < MLX4_MAX_VLAN_NUM; i++) {
+			if ((vlan == (MLX4_VLAN_MASK & be32_to_cpu(table->entries[i]))))
+				index_at_port = i;
+			if ((vlan == (MLX4_VLAN_MASK & be32_to_cpu(dup_table->entries[i]))))
+				index_at_dup_port = i;
+		}
+		/* check that same vlan is not in the tables at different indices */
+		if ((index_at_port != index_at_dup_port) &&
+		    (index_at_port >= 0) &&
+		    (index_at_dup_port >= 0))
+			can_mf_bond = false;
+
+		/* If the vlan is already in the primary table, the slot must be
+		 * available in the duplicate table as well.
+		 */
+		if (index_at_port >= 0 && index_at_dup_port < 0 &&
+		    dup_table->refs[index_at_port]) {
+			can_mf_bond = false;
+		}
+		/* If the vlan is already in the duplicate table, check that the
+		 * corresponding index is not occupied in the primary table, or
+		 * the primary table already contains the vlan at the same index.
+		 * Otherwise, you cannot bond (primary contains a different vlan
+		 * at that index).
+		 */
+		if (index_at_dup_port >= 0) {
+			if (!table->refs[index_at_dup_port] ||
+			    (vlan == (MLX4_VLAN_MASK & be32_to_cpu(dup_table->entries[index_at_dup_port]))))
+				free_for_dup = index_at_dup_port;
+			else
+				can_mf_bond = false;
+		}
+	}
+
 	for (i = MLX4_VLAN_REGULAR; i < MLX4_MAX_VLAN_NUM; i++) {
-		if (free < 0 && (table->refs[i] == 0)) {
-			free = i;
-			continue;
+		if (!table->refs[i]) {
+			if (free < 0)
+				free = i;
+			if (free_for_dup < 0 && need_mf_bond && can_mf_bond) {
+				if (!dup_table->refs[i])
+					free_for_dup = i;
+			}
 		}
 
-		if (table->refs[i] &&
+		if ((table->refs[i] || table->is_dup[i]) &&
 		    (vlan == (MLX4_VLAN_MASK &
 			      be32_to_cpu(table->entries[i])))) {
 			/* Vlan already registered, increase references count */
+			mlx4_dbg(dev, "vlan %u is already registered.\n", vlan);
 			*index = i;
 			++table->refs[i];
+			if (dup) {
+				u16 dup_vlan = MLX4_VLAN_MASK & be32_to_cpu(dup_table->entries[i]);
+
+				if (dup_vlan != vlan || !dup_table->is_dup[i]) {
+					mlx4_warn(dev, "register vlan: expected duplicate vlan %u on port %d index %d\n",
+						  vlan, dup_port, i);
+				}
+			}
 			goto out;
 		}
 	}
 
+	if (need_mf_bond && (free_for_dup < 0)) {
+		if (dup) {
+			mlx4_warn(dev, "Fail to allocate duplicate VLAN table entry\n");
+			mlx4_warn(dev, "High Availability for virtual functions may not work as expected\n");
+			dup = false;
+		}
+		can_mf_bond = false;
+	}
+
+	if (need_mf_bond && can_mf_bond)
+		free = free_for_dup;
+
 	if (free < 0) {
 		err = -ENOMEM;
 		goto out;
@@ -412,6 +689,7 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan,
 
 	/* Register new VLAN */
 	table->refs[free] = 1;
+	table->is_dup[free] = false;
 	table->entries[free] = cpu_to_be32(vlan | MLX4_VLAN_VALID);
 
 	err = mlx4_set_port_vlan_table(dev, port, table->entries);
@@ -421,11 +699,35 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan,
 		table->entries[free] = 0;
 		goto out;
 	}
+	++table->total;
+	if (dup) {
+		dup_table->refs[free] = 0;
+		dup_table->is_dup[free] = true;
+		dup_table->entries[free] = cpu_to_be32(vlan | MLX4_VLAN_VALID);
+
+		err = mlx4_set_port_vlan_table(dev, dup_port, dup_table->entries);
+		if (unlikely(err)) {
+			mlx4_warn(dev, "Failed adding duplicate vlan: %u\n", vlan);
+			dup_table->is_dup[free] = false;
+			dup_table->entries[free] = 0;
+			goto out;
+		}
+		++dup_table->total;
+	}
 
 	*index = free;
-	++table->total;
 out:
-	mutex_unlock(&table->mutex);
+	if (need_mf_bond) {
+		if (port == 2) {
+			mutex_unlock(&table->mutex);
+			mutex_unlock(&dup_table->mutex);
+		} else {
+			mutex_unlock(&dup_table->mutex);
+			mutex_unlock(&table->mutex);
+		}
+	} else {
+		mutex_unlock(&table->mutex);
+	}
 	return err;
 }
 
@@ -455,8 +757,22 @@ void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan)
 {
 	struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table;
 	int index;
+	bool dup = mlx4_is_mf_bonded(dev);
+	u8 dup_port = (port == 1) ? 2 : 1;
+	struct mlx4_vlan_table *dup_table = &mlx4_priv(dev)->port[dup_port].vlan_table;
+
+	if (dup) {
+		if (port == 1) {
+			mutex_lock(&table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
+		} else {
+			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
+		}
+	} else {
+		mutex_lock(&table->mutex);
+	}
 
-	mutex_lock(&table->mutex);
 	if (mlx4_find_cached_vlan(dev, port, vlan, &index)) {
 		mlx4_warn(dev, "vlan 0x%x is not in the vlan table\n", vlan);
 		goto out;
@@ -467,16 +783,38 @@ void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan)
 		goto out;
 	}
 
-	if (--table->refs[index]) {
+	if (--table->refs[index] || table->is_dup[index]) {
 		mlx4_dbg(dev, "Have %d more references for index %d, no need to modify vlan table\n",
 			 table->refs[index], index);
+		if (!table->refs[index])
+			dup_table->is_dup[index] = false;
 		goto out;
 	}
 	table->entries[index] = 0;
-	mlx4_set_port_vlan_table(dev, port, table->entries);
+	if (mlx4_set_port_vlan_table(dev, port, table->entries))
+		mlx4_warn(dev, "Fail to set vlan in port %d during unregister\n", port);
 	--table->total;
+	if (dup) {
+		dup_table->is_dup[index] = false;
+		if (dup_table->refs[index])
+			goto out;
+		dup_table->entries[index] = 0;
+		if (mlx4_set_port_vlan_table(dev, dup_port, dup_table->entries))
+			mlx4_warn(dev, "Fail to set vlan in duplicate port %d during unregister\n", dup_port);
+		--dup_table->total;
+	}
 out:
-	mutex_unlock(&table->mutex);
+	if (dup) {
+		if (port == 2) {
+			mutex_unlock(&table->mutex);
+			mutex_unlock(&dup_table->mutex);
+		} else {
+			mutex_unlock(&dup_table->mutex);
+			mutex_unlock(&table->mutex);
+		}
+	} else {
+		mutex_unlock(&table->mutex);
+	}
 }
 
 void mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan)
@@ -495,6 +833,220 @@ void mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan)
 }
 EXPORT_SYMBOL_GPL(mlx4_unregister_vlan);
 
+int mlx4_bond_mac_table(struct mlx4_dev *dev)
+{
+	struct mlx4_mac_table *t1 = &mlx4_priv(dev)->port[1].mac_table;
+	struct mlx4_mac_table *t2 = &mlx4_priv(dev)->port[2].mac_table;
+	int ret = 0;
+	int i;
+	bool update1 = false;
+	bool update2 = false;
+
+	mutex_lock(&t1->mutex);
+	mutex_lock(&t2->mutex);
+	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+		if ((t1->entries[i] != t2->entries[i]) &&
+		    t1->entries[i] && t2->entries[i]) {
+			mlx4_warn(dev, "can't duplicate entry %d in mac table\n", i);
+			ret = -EINVAL;
+			goto unlock;
+		}
+	}
+
+	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+		if (t1->entries[i] && !t2->entries[i]) {
+			t2->entries[i] = t1->entries[i];
+			t2->is_dup[i] = true;
+			update2 = true;
+		} else if (!t1->entries[i] && t2->entries[i]) {
+			t1->entries[i] = t2->entries[i];
+			t1->is_dup[i] = true;
+			update1 = true;
+		} else if (t1->entries[i] && t2->entries[i]) {
+			t1->is_dup[i] = true;
+			t2->is_dup[i] = true;
+		}
+	}
+
+	if (update1) {
+		ret = mlx4_set_port_mac_table(dev, 1, t1->entries);
+		if (ret)
+			mlx4_warn(dev, "failed to set MAC table for port 1 (%d)\n", ret);
+	}
+	if (!ret && update2) {
+		ret = mlx4_set_port_mac_table(dev, 2, t2->entries);
+		if (ret)
+			mlx4_warn(dev, "failed to set MAC table for port 2 (%d)\n", ret);
+	}
+
+	if (ret)
+		mlx4_warn(dev, "failed to create mirror MAC tables\n");
+unlock:
+	mutex_unlock(&t2->mutex);
+	mutex_unlock(&t1->mutex);
+	return ret;
+}
+
+int mlx4_unbond_mac_table(struct mlx4_dev *dev)
+{
+	struct mlx4_mac_table *t1 = &mlx4_priv(dev)->port[1].mac_table;
+	struct mlx4_mac_table *t2 = &mlx4_priv(dev)->port[2].mac_table;
+	int ret = 0;
+	int ret1;
+	int i;
+	bool update1 = false;
+	bool update2 = false;
+
+	mutex_lock(&t1->mutex);
+	mutex_lock(&t2->mutex);
+	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+		if (t1->entries[i] != t2->entries[i]) {
+			mlx4_warn(dev, "mac table is in an unexpected state when trying to unbond\n");
+			ret = -EINVAL;
+			goto unlock;
+		}
+	}
+
+	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+		if (!t1->entries[i])
+			continue;
+		t1->is_dup[i] = false;
+		if (!t1->refs[i]) {
+			t1->entries[i] = 0;
+			update1 = true;
+		}
+		t2->is_dup[i] = false;
+		if (!t2->refs[i]) {
+			t2->entries[i] = 0;
+			update2 = true;
+		}
+	}
+
+	if (update1) {
+		ret = mlx4_set_port_mac_table(dev, 1, t1->entries);
+		if (ret)
+			mlx4_warn(dev, "failed to unmirror MAC tables for port 1(%d)\n", ret);
+	}
+	if (update2) {
+		ret1 = mlx4_set_port_mac_table(dev, 2, t2->entries);
+		if (ret1) {
+			mlx4_warn(dev, "failed to unmirror MAC tables for port 2(%d)\n", ret1);
+			ret = ret1;
+		}
+	}
+unlock:
+	mutex_unlock(&t2->mutex);
+	mutex_unlock(&t1->mutex);
+	return ret;
+}
+
+int mlx4_bond_vlan_table(struct mlx4_dev *dev)
+{
+	struct mlx4_vlan_table *t1 = &mlx4_priv(dev)->port[1].vlan_table;
+	struct mlx4_vlan_table *t2 = &mlx4_priv(dev)->port[2].vlan_table;
+	int ret = 0;
+	int i;
+	bool update1 = false;
+	bool update2 = false;
+
+	mutex_lock(&t1->mutex);
+	mutex_lock(&t2->mutex);
+	for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) {
+		if ((t1->entries[i] != t2->entries[i]) &&
+		    t1->entries[i] && t2->entries[i]) {
+			mlx4_warn(dev, "can't duplicate entry %d in vlan table\n", i);
+			ret = -EINVAL;
+			goto unlock;
+		}
+	}
+
+	for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) {
+		if (t1->entries[i] && !t2->entries[i]) {
+			t2->entries[i] = t1->entries[i];
+			t2->is_dup[i] = true;
+			update2 = true;
+		} else if (!t1->entries[i] && t2->entries[i]) {
+			t1->entries[i] = t2->entries[i];
+			t1->is_dup[i] = true;
+			update1 = true;
+		} else if (t1->entries[i] && t2->entries[i]) {
+			t1->is_dup[i] = true;
+			t2->is_dup[i] = true;
+		}
+	}
+
+	if (update1) {
+		ret = mlx4_set_port_vlan_table(dev, 1, t1->entries);
+		if (ret)
+			mlx4_warn(dev, "failed to set VLAN table for port 1 (%d)\n", ret);
+	}
+	if (!ret && update2) {
+		ret = mlx4_set_port_vlan_table(dev, 2, t2->entries);
+		if (ret)
+			mlx4_warn(dev, "failed to set VLAN table for port 2 (%d)\n", ret);
+	}
+
+	if (ret)
+		mlx4_warn(dev, "failed to create mirror VLAN tables\n");
+unlock:
+	mutex_unlock(&t2->mutex);
+	mutex_unlock(&t1->mutex);
+	return ret;
+}
+
+int mlx4_unbond_vlan_table(struct mlx4_dev *dev)
+{
+	struct mlx4_vlan_table *t1 = &mlx4_priv(dev)->port[1].vlan_table;
+	struct mlx4_vlan_table *t2 = &mlx4_priv(dev)->port[2].vlan_table;
+	int ret = 0;
+	int ret1;
+	int i;
+	bool update1 = false;
+	bool update2 = false;
+
+	mutex_lock(&t1->mutex);
+	mutex_lock(&t2->mutex);
+	for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) {
+		if (t1->entries[i] != t2->entries[i]) {
+			mlx4_warn(dev, "vlan table is in an unexpected state when trying to unbond\n");
+			ret = -EINVAL;
+			goto unlock;
+		}
+	}
+
+	for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) {
+		if (!t1->entries[i])
+			continue;
+		t1->is_dup[i] = false;
+		if (!t1->refs[i]) {
+			t1->entries[i] = 0;
+			update1 = true;
+		}
+		t2->is_dup[i] = false;
+		if (!t2->refs[i]) {
+			t2->entries[i] = 0;
+			update2 = true;
+		}
+	}
+
+	if (update1) {
+		ret = mlx4_set_port_vlan_table(dev, 1, t1->entries);
+		if (ret)
+			mlx4_warn(dev, "failed to unmirror VLAN tables for port 1(%d)\n", ret);
+	}
+	if (update2) {
+		ret1 = mlx4_set_port_vlan_table(dev, 2, t2->entries);
+		if (ret1) {
+			mlx4_warn(dev, "failed to unmirror VLAN tables for port 2(%d)\n", ret1);
+			ret = ret1;
+		}
+	}
+unlock:
+	mutex_unlock(&t2->mutex);
+	mutex_unlock(&t1->mutex);
+	return ret;
+}
+
 int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps)
 {
 	struct mlx4_cmd_mailbox *inmailbox, *outmailbox;
@@ -968,6 +1520,8 @@ int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz)
 	return err;
 }
 
+#define SET_PORT_ROCE_2_FLAGS          0x10
+#define MLX4_SET_PORT_ROCE_V1_V2       0x2
 int mlx4_SET_PORT_general(struct mlx4_dev *dev, u8 port, int mtu,
 			  u8 pptx, u8 pfctx, u8 pprx, u8 pfcrx)
 {
@@ -987,6 +1541,11 @@ int mlx4_SET_PORT_general(struct mlx4_dev *dev, u8 port, int mtu,
 	context->pprx = (pprx * (!pfcrx)) << 7;
 	context->pfcrx = pfcrx;
 
+	if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) {
+		context->flags |= SET_PORT_ROCE_2_FLAGS;
+		context->roce_mode |=
+			MLX4_SET_PORT_ROCE_V1_V2 << 4;
+	}
 	in_mod = MLX4_SET_PORT_GENERAL << 8 | port;
 	err = mlx4_cmd(dev, mailbox->dma, in_mod, MLX4_SET_PORT_ETH_OPCODE,
 		       MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B,
diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c
index 168823dde..d1cd9c32a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx4/qp.c
@@ -167,6 +167,12 @@ static int __mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 		context->log_page_size   = mtt->page_shift - MLX4_ICM_PAGE_SHIFT;
 	}
 
+	if ((cur_state == MLX4_QP_STATE_RTR) &&
+	    (new_state == MLX4_QP_STATE_RTS) &&
+	    dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2)
+		context->roce_entropy =
+			cpu_to_be16(mlx4_qp_roce_entropy(dev, qp->qpn));
+
 	*(__be32 *) mailbox->buf = cpu_to_be32(optpar);
 	memcpy(mailbox->buf + 8, context, sizeof *context);
 
@@ -921,3 +927,23 @@ int mlx4_qp_to_ready(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 	return 0;
 }
 EXPORT_SYMBOL_GPL(mlx4_qp_to_ready);
+
+u16 mlx4_qp_roce_entropy(struct mlx4_dev *dev, u32 qpn)
+{
+	struct mlx4_qp_context context;
+	struct mlx4_qp qp;
+	int err;
+
+	qp.qpn = qpn;
+	err = mlx4_qp_query(dev, &qp, &context);
+	if (!err) {
+		u32 dest_qpn = be32_to_cpu(context.remote_qpn) & 0xffffff;
+		u16 folded_dst = folded_qp(dest_qpn);
+		u16 folded_src = folded_qp(qpn);
+
+		return (dest_qpn != qpn) ?
+			((folded_dst ^ folded_src) | 0xC000) :
+			folded_src | 0xC000;
+	}
+	return 0xdead;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index cad6c44df..25ce1b030 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -222,6 +222,13 @@ enum res_fs_rule_states {
 struct res_fs_rule {
 	struct res_common	com;
 	int			qpn;
+	/* VF DMFS mbox with port flipped */
+	void			*mirr_mbox;
+	/* > 0 --> apply mirror when getting into HA mode      */
+	/* = 0 --> un-apply mirror when getting out of HA mode */
+	u32			mirr_mbox_size;
+	struct list_head	mirr_list;
+	u64			mirr_rule_id;
 };
 
 static void *res_tracker_lookup(struct rb_root *root, u64 res_id)
@@ -908,11 +915,13 @@ static int handle_existing_counter(struct mlx4_dev *dev, u8 slave, int port,
 
 	spin_lock_irq(mlx4_tlock(dev));
 	r = find_res(dev, counter_index, RES_COUNTER);
-	if (!r || r->owner != slave)
+	if (!r || r->owner != slave) {
 		ret = -EINVAL;
-	counter = container_of(r, struct res_counter, com);
-	if (!counter->port)
-		counter->port = port;
+	} else {
+		counter = container_of(r, struct res_counter, com);
+		if (!counter->port)
+			counter->port = port;
+	}
 
 	spin_unlock_irq(mlx4_tlock(dev));
 	return ret;
@@ -4284,6 +4293,22 @@ err_mac:
 	return err;
 }
 
+static u32 qp_attach_mbox_size(void *mbox)
+{
+	u32 size = sizeof(struct mlx4_net_trans_rule_hw_ctrl);
+	struct _rule_hw  *rule_header;
+
+	rule_header = (struct _rule_hw *)(mbox + size);
+
+	while (rule_header->size) {
+		size += rule_header->size * sizeof(u32);
+		rule_header += 1;
+	}
+	return size;
+}
+
+static int mlx4_do_mirror_rule(struct mlx4_dev *dev, struct res_fs_rule *fs_rule);
+
 int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 					 struct mlx4_vhcr *vhcr,
 					 struct mlx4_cmd_mailbox *inbox,
@@ -4300,6 +4325,8 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	struct mlx4_net_trans_rule_hw_ctrl *ctrl;
 	struct _rule_hw  *rule_header;
 	int header_id;
+	struct res_fs_rule *rrule;
+	u32 mbox_size;
 
 	if (dev->caps.steering_mode !=
 	    MLX4_STEERING_MODE_DEVICE_MANAGED)
@@ -4329,7 +4356,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	case MLX4_NET_TRANS_RULE_ID_ETH:
 		if (validate_eth_header_mac(slave, rule_header, rlist)) {
 			err = -EINVAL;
-			goto err_put;
+			goto err_put_qp;
 		}
 		break;
 	case MLX4_NET_TRANS_RULE_ID_IB:
@@ -4340,7 +4367,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 		pr_warn("Can't attach FS rule without L2 headers, adding L2 header\n");
 		if (add_eth_header(dev, slave, inbox, rlist, header_id)) {
 			err = -EINVAL;
-			goto err_put;
+			goto err_put_qp;
 		}
 		vhcr->in_modifier +=
 			sizeof(struct mlx4_net_trans_rule_hw_eth) >> 2;
@@ -4348,7 +4375,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	default:
 		pr_err("Corrupted mailbox\n");
 		err = -EINVAL;
-		goto err_put;
+		goto err_put_qp;
 	}
 
 execute:
@@ -4357,23 +4384,69 @@ execute:
 			   MLX4_QP_FLOW_STEERING_ATTACH, MLX4_CMD_TIME_CLASS_A,
 			   MLX4_CMD_NATIVE);
 	if (err)
-		goto err_put;
+		goto err_put_qp;
+
 
 	err = add_res_range(dev, slave, vhcr->out_param, 1, RES_FS_RULE, qpn);
 	if (err) {
 		mlx4_err(dev, "Fail to add flow steering resources\n");
-		/* detach rule*/
+		goto err_detach;
+	}
+
+	err = get_res(dev, slave, vhcr->out_param, RES_FS_RULE, &rrule);
+	if (err)
+		goto err_detach;
+
+	mbox_size = qp_attach_mbox_size(inbox->buf);
+	rrule->mirr_mbox = kmalloc(mbox_size, GFP_KERNEL);
+	if (!rrule->mirr_mbox) {
+		err = -ENOMEM;
+		goto err_put_rule;
+	}
+	rrule->mirr_mbox_size = mbox_size;
+	rrule->mirr_rule_id = 0;
+	memcpy(rrule->mirr_mbox, inbox->buf, mbox_size);
+
+	/* set different port */
+	ctrl = (struct mlx4_net_trans_rule_hw_ctrl *)rrule->mirr_mbox;
+	if (ctrl->port == 1)
+		ctrl->port = 2;
+	else
+		ctrl->port = 1;
+
+	if (mlx4_is_bonded(dev))
+		mlx4_do_mirror_rule(dev, rrule);
+
+	atomic_inc(&rqp->ref_count);
+
+err_put_rule:
+	put_res(dev, slave, vhcr->out_param, RES_FS_RULE);
+err_detach:
+	/* detach rule on error */
+	if (err)
 		mlx4_cmd(dev, vhcr->out_param, 0, 0,
 			 MLX4_QP_FLOW_STEERING_DETACH, MLX4_CMD_TIME_CLASS_A,
 			 MLX4_CMD_NATIVE);
-		goto err_put;
-	}
-	atomic_inc(&rqp->ref_count);
-err_put:
+err_put_qp:
 	put_res(dev, slave, qpn, RES_QP);
 	return err;
 }
 
+static int mlx4_undo_mirror_rule(struct mlx4_dev *dev, struct res_fs_rule *fs_rule)
+{
+	int err;
+
+	err = rem_res_range(dev, fs_rule->com.owner, fs_rule->com.res_id, 1, RES_FS_RULE, 0);
+	if (err) {
+		mlx4_err(dev, "Fail to remove flow steering resources\n");
+		return err;
+	}
+
+	mlx4_cmd(dev, fs_rule->com.res_id, 0, 0, MLX4_QP_FLOW_STEERING_DETACH,
+		 MLX4_CMD_TIME_CLASS_A, MLX4_CMD_NATIVE);
+	return 0;
+}
+
 int mlx4_QP_FLOW_STEERING_DETACH_wrapper(struct mlx4_dev *dev, int slave,
 					 struct mlx4_vhcr *vhcr,
 					 struct mlx4_cmd_mailbox *inbox,
@@ -4383,6 +4456,7 @@ int mlx4_QP_FLOW_STEERING_DETACH_wrapper(struct mlx4_dev *dev, int slave,
 	int err;
 	struct res_qp *rqp;
 	struct res_fs_rule *rrule;
+	u64 mirr_reg_id;
 
 	if (dev->caps.steering_mode !=
 	    MLX4_STEERING_MODE_DEVICE_MANAGED)
@@ -4391,12 +4465,30 @@ int mlx4_QP_FLOW_STEERING_DETACH_wrapper(struct mlx4_dev *dev, int slave,
 	err = get_res(dev, slave, vhcr->in_param, RES_FS_RULE, &rrule);
 	if (err)
 		return err;
+
+	if (!rrule->mirr_mbox) {
+		mlx4_err(dev, "Mirror rules cannot be removed explicitly\n");
+		put_res(dev, slave, vhcr->in_param, RES_FS_RULE);
+		return -EINVAL;
+	}
+	mirr_reg_id = rrule->mirr_rule_id;
+	kfree(rrule->mirr_mbox);
+
 	/* Release the rule form busy state before removal */
 	put_res(dev, slave, vhcr->in_param, RES_FS_RULE);
 	err = get_res(dev, slave, rrule->qpn, RES_QP, &rqp);
 	if (err)
 		return err;
 
+	if (mirr_reg_id && mlx4_is_bonded(dev)) {
+		err = get_res(dev, slave, mirr_reg_id, RES_FS_RULE, &rrule);
+		if (err) {
+			mlx4_err(dev, "Fail to get resource of mirror rule\n");
+		} else {
+			put_res(dev, slave, mirr_reg_id, RES_FS_RULE);
+			mlx4_undo_mirror_rule(dev, rrule);
+		}
+	}
 	err = rem_res_range(dev, slave, vhcr->in_param, 1, RES_FS_RULE, 0);
 	if (err) {
 		mlx4_err(dev, "Fail to remove flow steering resources\n");
@@ -4834,6 +4926,91 @@ static void rem_slave_mtts(struct mlx4_dev *dev, int slave)
 	spin_unlock_irq(mlx4_tlock(dev));
 }
 
+static int mlx4_do_mirror_rule(struct mlx4_dev *dev, struct res_fs_rule *fs_rule)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	int err;
+	struct res_fs_rule *mirr_rule;
+	u64 reg_id;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	if (!fs_rule->mirr_mbox) {
+		mlx4_err(dev, "rule mirroring mailbox is null\n");
+		return -EINVAL;
+	}
+	memcpy(mailbox->buf, fs_rule->mirr_mbox, fs_rule->mirr_mbox_size);
+	err = mlx4_cmd_imm(dev, mailbox->dma, &reg_id, fs_rule->mirr_mbox_size >> 2, 0,
+			   MLX4_QP_FLOW_STEERING_ATTACH, MLX4_CMD_TIME_CLASS_A,
+			   MLX4_CMD_NATIVE);
+	mlx4_free_cmd_mailbox(dev, mailbox);
+
+	if (err)
+		goto err;
+
+	err = add_res_range(dev, fs_rule->com.owner, reg_id, 1, RES_FS_RULE, fs_rule->qpn);
+	if (err)
+		goto err_detach;
+
+	err = get_res(dev, fs_rule->com.owner, reg_id, RES_FS_RULE, &mirr_rule);
+	if (err)
+		goto err_rem;
+
+	fs_rule->mirr_rule_id = reg_id;
+	mirr_rule->mirr_rule_id = 0;
+	mirr_rule->mirr_mbox_size = 0;
+	mirr_rule->mirr_mbox = NULL;
+	put_res(dev, fs_rule->com.owner, reg_id, RES_FS_RULE);
+
+	return 0;
+err_rem:
+	rem_res_range(dev, fs_rule->com.owner, reg_id, 1, RES_FS_RULE, 0);
+err_detach:
+	mlx4_cmd(dev, reg_id, 0, 0, MLX4_QP_FLOW_STEERING_DETACH,
+		 MLX4_CMD_TIME_CLASS_A, MLX4_CMD_NATIVE);
+err:
+	return err;
+}
+
+static int mlx4_mirror_fs_rules(struct mlx4_dev *dev, bool bond)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	struct mlx4_resource_tracker *tracker =
+		&priv->mfunc.master.res_tracker;
+	struct rb_root *root = &tracker->res_tree[RES_FS_RULE];
+	struct rb_node *p;
+	struct res_fs_rule *fs_rule;
+	int err = 0;
+	LIST_HEAD(mirr_list);
+
+	for (p = rb_first(root); p; p = rb_next(p)) {
+		fs_rule = rb_entry(p, struct res_fs_rule, com.node);
+		if ((bond && fs_rule->mirr_mbox_size) ||
+		    (!bond && !fs_rule->mirr_mbox_size))
+			list_add_tail(&fs_rule->mirr_list, &mirr_list);
+	}
+
+	list_for_each_entry(fs_rule, &mirr_list, mirr_list) {
+		if (bond)
+			err += mlx4_do_mirror_rule(dev, fs_rule);
+		else
+			err += mlx4_undo_mirror_rule(dev, fs_rule);
+	}
+	return err;
+}
+
+int mlx4_bond_fs_rules(struct mlx4_dev *dev)
+{
+	return mlx4_mirror_fs_rules(dev, true);
+}
+
+int mlx4_unbond_fs_rules(struct mlx4_dev *dev)
+{
+	return mlx4_mirror_fs_rules(dev, false);
+}
+
 static void rem_slave_fs_rule(struct mlx4_dev *dev, int slave)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
index 158c88c69..c503ea05e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
@@ -13,6 +13,7 @@ config MLX5_CORE
 config MLX5_CORE_EN
 	bool "Mellanox Technologies ConnectX-4 Ethernet support"
 	depends on NETDEVICES && ETHERNET && PCI && MLX5_CORE
+	select PTP_1588_CLOCK
 	default n
 	---help---
 	  Ethernet support in Mellanox Technologies ConnectX-4 NIC.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 26a68b8af..01c0256ef 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -2,7 +2,7 @@ obj-$(CONFIG_MLX5_CORE)		+= mlx5_core.o
 
 mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 		health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o   \
-		mad.o transobj.o vport.o
-mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o \
-		en_main.o en_flow_table.o en_ethtool.o en_tx.o en_rx.o \
-		en_txrx.o
+		mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o
+mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \
+		en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \
+		en_txrx.o en_clock.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 22e72bf1a..5b1753233 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -32,12 +32,15 @@
 
 #include <linux/if_vlan.h>
 #include <linux/etherdevice.h>
+#include <linux/timecounter.h>
+#include <linux/net_tstamp.h>
+#include <linux/ptp_clock_kernel.h>
 #include <linux/mlx5/driver.h>
 #include <linux/mlx5/qp.h>
 #include <linux/mlx5/cq.h>
 #include <linux/mlx5/vport.h>
+#include <linux/mlx5/transobj.h>
 #include "wq.h"
-#include "transobj.h"
 #include "mlx5_core.h"
 
 #define MLX5E_MAX_NUM_TC	8
@@ -64,6 +67,8 @@
 #define MLX5E_UPDATE_STATS_INTERVAL    200 /* msecs */
 #define MLX5E_SQ_BF_BUDGET             16
 
+#define MLX5E_NUM_MAIN_GROUPS 9
+
 static const char vport_strings[][ETH_GSTRING_LEN] = {
 	/* vport statistics */
 	"rx_packets",
@@ -218,6 +223,7 @@ struct mlx5e_pport_stats {
 
 static const char rq_stats_strings[][ETH_GSTRING_LEN] = {
 	"packets",
+	"bytes",
 	"csum_none",
 	"csum_sw",
 	"lro_packets",
@@ -227,16 +233,18 @@ static const char rq_stats_strings[][ETH_GSTRING_LEN] = {
 
 struct mlx5e_rq_stats {
 	u64 packets;
+	u64 bytes;
 	u64 csum_none;
 	u64 csum_sw;
 	u64 lro_packets;
 	u64 lro_bytes;
 	u64 wqe_err;
-#define NUM_RQ_STATS 6
+#define NUM_RQ_STATS 7
 };
 
 static const char sq_stats_strings[][ETH_GSTRING_LEN] = {
 	"packets",
+	"bytes",
 	"tso_packets",
 	"tso_bytes",
 	"csum_offload_none",
@@ -248,6 +256,7 @@ static const char sq_stats_strings[][ETH_GSTRING_LEN] = {
 
 struct mlx5e_sq_stats {
 	u64 packets;
+	u64 bytes;
 	u64 tso_packets;
 	u64 tso_bytes;
 	u64 csum_offload_none;
@@ -255,7 +264,7 @@ struct mlx5e_sq_stats {
 	u64 wake;
 	u64 dropped;
 	u64 nop;
-#define NUM_SQ_STATS 8
+#define NUM_SQ_STATS 9
 };
 
 struct mlx5e_stats {
@@ -282,18 +291,26 @@ struct mlx5e_params {
 	u32 indirection_rqt[MLX5E_INDIR_RQT_SIZE];
 };
 
-enum {
-	MLX5E_RQ_STATE_POST_WQES_ENABLE,
+struct mlx5e_tstamp {
+	rwlock_t                   lock;
+	struct cyclecounter        cycles;
+	struct timecounter         clock;
+	struct hwtstamp_config     hwtstamp_config;
+	u32                        nominal_c_mult;
+	unsigned long              overflow_period;
+	struct delayed_work        overflow_work;
+	struct mlx5_core_dev      *mdev;
+	struct ptp_clock          *ptp;
+	struct ptp_clock_info      ptp_info;
 };
 
-enum cq_flags {
-	MLX5E_CQ_HAS_CQES = 1,
+enum {
+	MLX5E_RQ_STATE_POST_WQES_ENABLE,
 };
 
 struct mlx5e_cq {
 	/* data path - accessed per cqe */
 	struct mlx5_cqwq           wq;
-	unsigned long              flags;
 
 	/* data path - accessed per napi poll */
 	struct napi_struct        *napi;
@@ -313,6 +330,7 @@ struct mlx5e_rq {
 
 	struct device         *pdev;
 	struct net_device     *netdev;
+	struct mlx5e_tstamp   *tstamp;
 	struct mlx5e_rq_stats  stats;
 	struct mlx5e_cq        cq;
 
@@ -326,14 +344,12 @@ struct mlx5e_rq {
 	struct mlx5e_priv     *priv;
 } ____cacheline_aligned_in_smp;
 
-struct mlx5e_tx_skb_cb {
+struct mlx5e_tx_wqe_info {
 	u32 num_bytes;
 	u8  num_wqebbs;
 	u8  num_dma;
 };
 
-#define MLX5E_TX_SKB_CB(__skb) ((struct mlx5e_tx_skb_cb *)__skb->cb)
-
 enum mlx5e_dma_map_type {
 	MLX5E_DMA_MAP_SINGLE,
 	MLX5E_DMA_MAP_PAGE
@@ -369,6 +385,7 @@ struct mlx5e_sq {
 	/* pointers to per packet info: write@xmit, read@completion */
 	struct sk_buff           **skb;
 	struct mlx5e_sq_dma       *dma_fifo;
+	struct mlx5e_tx_wqe_info  *wqe_info;
 
 	/* read only */
 	struct mlx5_wq_cyc         wq;
@@ -381,6 +398,7 @@ struct mlx5e_sq {
 	u16                        max_inline;
 	u16                        edge;
 	struct device             *pdev;
+	struct mlx5e_tstamp       *tstamp;
 	__be32                     mkey_be;
 	unsigned long              state;
 
@@ -433,6 +451,8 @@ enum mlx5e_traffic_types {
 	MLX5E_NUM_TT,
 };
 
+#define IS_HASHING_TT(tt) (tt != MLX5E_TT_ANY)
+
 enum mlx5e_rqt_ix {
 	MLX5E_INDIRECTION_RQT,
 	MLX5E_SINGLE_RQ_RQT,
@@ -442,7 +462,7 @@ enum mlx5e_rqt_ix {
 struct mlx5e_eth_addr_info {
 	u8  addr[ETH_ALEN + 2];
 	u32 tt_vec;
-	u32 ft_ix[MLX5E_NUM_TT]; /* flow table index per traffic type */
+	struct mlx5_flow_rule *ft_rule[MLX5E_NUM_TT];
 };
 
 #define MLX5E_ETH_ADDR_HASH_SIZE (1 << BITS_PER_BYTE)
@@ -465,15 +485,23 @@ enum {
 };
 
 struct mlx5e_vlan_db {
-	u32           active_vlans_ft_ix[VLAN_N_VID];
-	u32           untagged_rule_ft_ix;
-	u32           any_vlan_rule_ft_ix;
+	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
+	struct mlx5_flow_rule	*active_vlans_rule[VLAN_N_VID];
+	struct mlx5_flow_rule	*untagged_rule;
+	struct mlx5_flow_rule	*any_vlan_rule;
 	bool          filter_disabled;
 };
 
 struct mlx5e_flow_table {
-	void *vlan;
-	void *main;
+	int num_groups;
+	struct mlx5_flow_table		*t;
+	struct mlx5_flow_group		**g;
+};
+
+struct mlx5e_flow_tables {
+	struct mlx5_flow_namespace	*ns;
+	struct mlx5e_flow_table		vlan;
+	struct mlx5e_flow_table		main;
 };
 
 struct mlx5e_priv {
@@ -496,7 +524,7 @@ struct mlx5e_priv {
 	u32                        rqtn[MLX5E_NUM_RQT];
 	u32                        tirn[MLX5E_NUM_TT];
 
-	struct mlx5e_flow_table    ft;
+	struct mlx5e_flow_tables   fts;
 	struct mlx5e_eth_addr_db   eth_addr;
 	struct mlx5e_vlan_db       vlan;
 
@@ -509,6 +537,7 @@ struct mlx5e_priv {
 	struct mlx5_core_dev      *mdev;
 	struct net_device         *netdev;
 	struct mlx5e_stats         stats;
+	struct mlx5e_tstamp        tstamp;
 };
 
 #define MLX5E_NET_IP_ALIGN 2
@@ -564,7 +593,7 @@ void mlx5e_completion_event(struct mlx5_core_cq *mcq);
 void mlx5e_cq_error_event(struct mlx5_core_cq *mcq, enum mlx5_event event);
 int mlx5e_napi_poll(struct napi_struct *napi, int budget);
 bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq);
-bool mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
+int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
 bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq);
 struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq);
 
@@ -575,6 +604,13 @@ void mlx5e_destroy_flow_tables(struct mlx5e_priv *priv);
 void mlx5e_init_eth_addr(struct mlx5e_priv *priv);
 void mlx5e_set_rx_mode_work(struct work_struct *work);
 
+void mlx5e_fill_hwstamp(struct mlx5e_tstamp *clock, u64 timestamp,
+			struct skb_shared_hwtstamps *hwts);
+void mlx5e_timestamp_init(struct mlx5e_priv *priv);
+void mlx5e_timestamp_cleanup(struct mlx5e_priv *priv);
+int mlx5e_hwstamp_set(struct net_device *dev, struct ifreq *ifr);
+int mlx5e_hwstamp_get(struct net_device *dev, struct ifreq *ifr);
+
 int mlx5e_vlan_rx_add_vid(struct net_device *dev, __always_unused __be16 proto,
 			  u16 vid);
 int mlx5e_vlan_rx_kill_vid(struct net_device *dev, __always_unused __be16 proto,
@@ -583,9 +619,12 @@ void mlx5e_enable_vlan_filter(struct mlx5e_priv *priv);
 void mlx5e_disable_vlan_filter(struct mlx5e_priv *priv);
 
 int mlx5e_redirect_rqt(struct mlx5e_priv *priv, enum mlx5e_rqt_ix rqt_ix);
+void mlx5e_build_tir_ctx_hash(void *tirc, struct mlx5e_priv *priv);
 
 int mlx5e_open_locked(struct net_device *netdev);
 int mlx5e_close_locked(struct net_device *netdev);
+void mlx5e_build_default_indir_rqt(u32 *indirection_rqt, int len,
+				   int num_channels);
 
 static inline void mlx5e_tx_notify_hw(struct mlx5e_sq *sq,
 				      struct mlx5e_tx_wqe *wqe, int bf_sz)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_clock.c b/drivers/net/ethernet/mellanox/mlx5/core/en_clock.c
new file mode 100644
index 000000000..2018eebe1
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_clock.c
@@ -0,0 +1,292 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/clocksource.h>
+#include "en.h"
+
+enum {
+	MLX5E_CYCLES_SHIFT	= 23
+};
+
+void mlx5e_fill_hwstamp(struct mlx5e_tstamp *tstamp, u64 timestamp,
+			struct skb_shared_hwtstamps *hwts)
+{
+	u64 nsec;
+
+	read_lock(&tstamp->lock);
+	nsec = timecounter_cyc2time(&tstamp->clock, timestamp);
+	read_unlock(&tstamp->lock);
+
+	hwts->hwtstamp = ns_to_ktime(nsec);
+}
+
+static cycle_t mlx5e_read_internal_timer(const struct cyclecounter *cc)
+{
+	struct mlx5e_tstamp *tstamp = container_of(cc, struct mlx5e_tstamp,
+						   cycles);
+
+	return mlx5_read_internal_timer(tstamp->mdev) & cc->mask;
+}
+
+static void mlx5e_timestamp_overflow(struct work_struct *work)
+{
+	struct delayed_work *dwork = to_delayed_work(work);
+	struct mlx5e_tstamp *tstamp = container_of(dwork, struct mlx5e_tstamp,
+						   overflow_work);
+	unsigned long flags;
+
+	write_lock_irqsave(&tstamp->lock, flags);
+	timecounter_read(&tstamp->clock);
+	write_unlock_irqrestore(&tstamp->lock, flags);
+	schedule_delayed_work(&tstamp->overflow_work, tstamp->overflow_period);
+}
+
+int mlx5e_hwstamp_set(struct net_device *dev, struct ifreq *ifr)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct hwtstamp_config config;
+
+	if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz))
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+		return -EFAULT;
+
+	/* TX HW timestamp */
+	switch (config.tx_type) {
+	case HWTSTAMP_TX_OFF:
+	case HWTSTAMP_TX_ON:
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	/* RX HW timestamp */
+	switch (config.rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		break;
+	case HWTSTAMP_FILTER_ALL:
+	case HWTSTAMP_FILTER_SOME:
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		config.rx_filter = HWTSTAMP_FILTER_ALL;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	memcpy(&priv->tstamp.hwtstamp_config, &config, sizeof(config));
+
+	return copy_to_user(ifr->ifr_data, &config,
+			    sizeof(config)) ? -EFAULT : 0;
+}
+
+int mlx5e_hwstamp_get(struct net_device *dev, struct ifreq *ifr)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct hwtstamp_config *cfg = &priv->tstamp.hwtstamp_config;
+
+	if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz))
+		return -EOPNOTSUPP;
+
+	return copy_to_user(ifr->ifr_data, cfg, sizeof(*cfg)) ? -EFAULT : 0;
+}
+
+static int mlx5e_ptp_settime(struct ptp_clock_info *ptp,
+			     const struct timespec64 *ts)
+{
+	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
+						   ptp_info);
+	u64 ns = timespec64_to_ns(ts);
+	unsigned long flags;
+
+	write_lock_irqsave(&tstamp->lock, flags);
+	timecounter_init(&tstamp->clock, &tstamp->cycles, ns);
+	write_unlock_irqrestore(&tstamp->lock, flags);
+
+	return 0;
+}
+
+static int mlx5e_ptp_gettime(struct ptp_clock_info *ptp,
+			     struct timespec64 *ts)
+{
+	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
+						   ptp_info);
+	u64 ns;
+	unsigned long flags;
+
+	write_lock_irqsave(&tstamp->lock, flags);
+	ns = timecounter_read(&tstamp->clock);
+	write_unlock_irqrestore(&tstamp->lock, flags);
+
+	*ts = ns_to_timespec64(ns);
+
+	return 0;
+}
+
+static int mlx5e_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
+						   ptp_info);
+	unsigned long flags;
+
+	write_lock_irqsave(&tstamp->lock, flags);
+	timecounter_adjtime(&tstamp->clock, delta);
+	write_unlock_irqrestore(&tstamp->lock, flags);
+
+	return 0;
+}
+
+static int mlx5e_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta)
+{
+	u64 adj;
+	u32 diff;
+	unsigned long flags;
+	int neg_adj = 0;
+	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
+						  ptp_info);
+
+	if (delta < 0) {
+		neg_adj = 1;
+		delta = -delta;
+	}
+
+	adj = tstamp->nominal_c_mult;
+	adj *= delta;
+	diff = div_u64(adj, 1000000000ULL);
+
+	write_lock_irqsave(&tstamp->lock, flags);
+	timecounter_read(&tstamp->clock);
+	tstamp->cycles.mult = neg_adj ? tstamp->nominal_c_mult - diff :
+					tstamp->nominal_c_mult + diff;
+	write_unlock_irqrestore(&tstamp->lock, flags);
+
+	return 0;
+}
+
+static const struct ptp_clock_info mlx5e_ptp_clock_info = {
+	.owner		= THIS_MODULE,
+	.max_adj	= 100000000,
+	.n_alarm	= 0,
+	.n_ext_ts	= 0,
+	.n_per_out	= 0,
+	.n_pins		= 0,
+	.pps		= 0,
+	.adjfreq	= mlx5e_ptp_adjfreq,
+	.adjtime	= mlx5e_ptp_adjtime,
+	.gettime64	= mlx5e_ptp_gettime,
+	.settime64	= mlx5e_ptp_settime,
+	.enable		= NULL,
+};
+
+static void mlx5e_timestamp_init_config(struct mlx5e_tstamp *tstamp)
+{
+	tstamp->hwtstamp_config.tx_type = HWTSTAMP_TX_OFF;
+	tstamp->hwtstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
+}
+
+void mlx5e_timestamp_init(struct mlx5e_priv *priv)
+{
+	struct mlx5e_tstamp *tstamp = &priv->tstamp;
+	u64 ns;
+	u64 frac = 0;
+	u32 dev_freq;
+
+	mlx5e_timestamp_init_config(tstamp);
+	dev_freq = MLX5_CAP_GEN(priv->mdev, device_frequency_khz);
+	if (!dev_freq) {
+		mlx5_core_warn(priv->mdev, "invalid device_frequency_khz, aborting HW clock init\n");
+		return;
+	}
+	rwlock_init(&tstamp->lock);
+	tstamp->cycles.read = mlx5e_read_internal_timer;
+	tstamp->cycles.shift = MLX5E_CYCLES_SHIFT;
+	tstamp->cycles.mult = clocksource_khz2mult(dev_freq,
+						   tstamp->cycles.shift);
+	tstamp->nominal_c_mult = tstamp->cycles.mult;
+	tstamp->cycles.mask = CLOCKSOURCE_MASK(41);
+	tstamp->mdev = priv->mdev;
+
+	timecounter_init(&tstamp->clock, &tstamp->cycles,
+			 ktime_to_ns(ktime_get_real()));
+
+	/* Calculate period in seconds to call the overflow watchdog - to make
+	 * sure counter is checked at least once every wrap around.
+	 */
+	ns = cyclecounter_cyc2ns(&tstamp->cycles, tstamp->cycles.mask,
+				 frac, &frac);
+	do_div(ns, NSEC_PER_SEC / 2 / HZ);
+	tstamp->overflow_period = ns;
+
+	INIT_DELAYED_WORK(&tstamp->overflow_work, mlx5e_timestamp_overflow);
+	if (tstamp->overflow_period)
+		schedule_delayed_work(&tstamp->overflow_work, 0);
+	else
+		mlx5_core_warn(priv->mdev, "invalid overflow period, overflow_work is not scheduled\n");
+
+	/* Configure the PHC */
+	tstamp->ptp_info = mlx5e_ptp_clock_info;
+	snprintf(tstamp->ptp_info.name, 16, "mlx5 ptp");
+
+	tstamp->ptp = ptp_clock_register(&tstamp->ptp_info,
+					 &priv->mdev->pdev->dev);
+	if (IS_ERR_OR_NULL(tstamp->ptp)) {
+		mlx5_core_warn(priv->mdev, "ptp_clock_register failed %ld\n",
+			       PTR_ERR(tstamp->ptp));
+		tstamp->ptp = NULL;
+	}
+}
+
+void mlx5e_timestamp_cleanup(struct mlx5e_priv *priv)
+{
+	struct mlx5e_tstamp *tstamp = &priv->tstamp;
+
+	if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz))
+		return;
+
+	if (priv->tstamp.ptp) {
+		ptp_clock_unregister(priv->tstamp.ptp);
+		priv->tstamp.ptp = NULL;
+	}
+
+	cancel_delayed_work_sync(&tstamp->overflow_work);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 2e022e900..5abeb00fc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -385,6 +385,8 @@ static int mlx5e_set_channels(struct net_device *dev,
 		mlx5e_close_locked(dev);
 
 	priv->params.num_channels = count;
+	mlx5e_build_default_indir_rqt(priv->params.indirection_rqt,
+				      MLX5E_INDIR_RQT_SIZE, count);
 
 	if (was_opened)
 		err = mlx5e_open_locked(dev);
@@ -703,18 +705,36 @@ static int mlx5e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 	return 0;
 }
 
+static void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in, int inlen)
+{
+	struct mlx5_core_dev *mdev = priv->mdev;
+	void *tirc = MLX5_ADDR_OF(modify_tir_in, in, ctx);
+	int i;
+
+	MLX5_SET(modify_tir_in, in, bitmask.hash, 1);
+	mlx5e_build_tir_ctx_hash(tirc, priv);
+
+	for (i = 0; i < MLX5E_NUM_TT; i++)
+		if (IS_HASHING_TT(i))
+			mlx5_core_modify_tir(mdev, priv->tirn[i], in, inlen);
+}
+
 static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir,
 			  const u8 *key, const u8 hfunc)
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
-	bool close_open;
-	int err = 0;
+	int inlen = MLX5_ST_SZ_BYTES(modify_tir_in);
+	void *in;
 
 	if ((hfunc != ETH_RSS_HASH_NO_CHANGE) &&
 	    (hfunc != ETH_RSS_HASH_XOR) &&
 	    (hfunc != ETH_RSS_HASH_TOP))
 		return -EINVAL;
 
+	in = mlx5_vzalloc(inlen);
+	if (!in)
+		return -ENOMEM;
+
 	mutex_lock(&priv->state_lock);
 
 	if (indir) {
@@ -723,11 +743,6 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir,
 		mlx5e_redirect_rqt(priv, MLX5E_INDIRECTION_RQT);
 	}
 
-	close_open = (key || (hfunc != ETH_RSS_HASH_NO_CHANGE)) &&
-		     test_bit(MLX5E_STATE_OPENED, &priv->state);
-	if (close_open)
-		mlx5e_close_locked(dev);
-
 	if (key)
 		memcpy(priv->params.toeplitz_hash_key, key,
 		       sizeof(priv->params.toeplitz_hash_key));
@@ -735,12 +750,13 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir,
 	if (hfunc != ETH_RSS_HASH_NO_CHANGE)
 		priv->params.rss_hfunc = hfunc;
 
-	if (close_open)
-		err = mlx5e_open_locked(priv->netdev);
+	mlx5e_modify_tirs_hash(priv, in, inlen);
 
 	mutex_unlock(&priv->state_lock);
 
-	return err;
+	kvfree(in);
+
+	return 0;
 }
 
 static int mlx5e_get_rxnfc(struct net_device *netdev,
@@ -855,6 +871,35 @@ static int mlx5e_set_pauseparam(struct net_device *netdev,
 	return err;
 }
 
+static int mlx5e_get_ts_info(struct net_device *dev,
+			     struct ethtool_ts_info *info)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	int ret;
+
+	ret = ethtool_op_get_ts_info(dev, info);
+	if (ret)
+		return ret;
+
+	info->phc_index = priv->tstamp.ptp ?
+			  ptp_clock_index(priv->tstamp.ptp) : -1;
+
+	if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz))
+		return 0;
+
+	info->so_timestamping |= SOF_TIMESTAMPING_TX_HARDWARE |
+				 SOF_TIMESTAMPING_RX_HARDWARE |
+				 SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	info->tx_types = (BIT(1) << HWTSTAMP_TX_OFF) |
+			 (BIT(1) << HWTSTAMP_TX_ON);
+
+	info->rx_filters = (BIT(1) << HWTSTAMP_FILTER_NONE) |
+			   (BIT(1) << HWTSTAMP_FILTER_ALL);
+
+	return 0;
+}
+
 const struct ethtool_ops mlx5e_ethtool_ops = {
 	.get_drvinfo       = mlx5e_get_drvinfo,
 	.get_link          = ethtool_op_get_link,
@@ -878,4 +923,5 @@ const struct ethtool_ops mlx5e_ethtool_ops = {
 	.set_tunable       = mlx5e_set_tunable,
 	.get_pauseparam    = mlx5e_get_pauseparam,
 	.set_pauseparam    = mlx5e_set_pauseparam,
+	.get_ts_info       = mlx5e_get_ts_info,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c
deleted file mode 100644
index 22d603f78..000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c
+++ /dev/null
@@ -1,907 +0,0 @@
-/*
- * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- */
-
-#include <linux/list.h>
-#include <linux/ip.h>
-#include <linux/ipv6.h>
-#include <linux/tcp.h>
-#include <linux/mlx5/flow_table.h>
-#include "en.h"
-
-enum {
-	MLX5E_FULLMATCH = 0,
-	MLX5E_ALLMULTI  = 1,
-	MLX5E_PROMISC   = 2,
-};
-
-enum {
-	MLX5E_UC        = 0,
-	MLX5E_MC_IPV4   = 1,
-	MLX5E_MC_IPV6   = 2,
-	MLX5E_MC_OTHER  = 3,
-};
-
-enum {
-	MLX5E_ACTION_NONE = 0,
-	MLX5E_ACTION_ADD  = 1,
-	MLX5E_ACTION_DEL  = 2,
-};
-
-struct mlx5e_eth_addr_hash_node {
-	struct hlist_node          hlist;
-	u8                         action;
-	struct mlx5e_eth_addr_info ai;
-};
-
-static inline int mlx5e_hash_eth_addr(u8 *addr)
-{
-	return addr[5];
-}
-
-static void mlx5e_add_eth_addr_to_hash(struct hlist_head *hash, u8 *addr)
-{
-	struct mlx5e_eth_addr_hash_node *hn;
-	int ix = mlx5e_hash_eth_addr(addr);
-	int found = 0;
-
-	hlist_for_each_entry(hn, &hash[ix], hlist)
-		if (ether_addr_equal_64bits(hn->ai.addr, addr)) {
-			found = 1;
-			break;
-		}
-
-	if (found) {
-		hn->action = MLX5E_ACTION_NONE;
-		return;
-	}
-
-	hn = kzalloc(sizeof(*hn), GFP_ATOMIC);
-	if (!hn)
-		return;
-
-	ether_addr_copy(hn->ai.addr, addr);
-	hn->action = MLX5E_ACTION_ADD;
-
-	hlist_add_head(&hn->hlist, &hash[ix]);
-}
-
-static void mlx5e_del_eth_addr_from_hash(struct mlx5e_eth_addr_hash_node *hn)
-{
-	hlist_del(&hn->hlist);
-	kfree(hn);
-}
-
-static void mlx5e_del_eth_addr_from_flow_table(struct mlx5e_priv *priv,
-					       struct mlx5e_eth_addr_info *ai)
-{
-	void *ft = priv->ft.main;
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV6_IPSEC_ESP]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV4_IPSEC_ESP]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV6_IPSEC_AH]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV4_IPSEC_AH]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_TCP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_TCP]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_TCP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_TCP]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_UDP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_UDP]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_UDP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_UDP]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV6))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_IPV4))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4]);
-
-	if (ai->tt_vec & BIT(MLX5E_TT_ANY))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_ANY]);
-}
-
-static int mlx5e_get_eth_addr_type(u8 *addr)
-{
-	if (is_unicast_ether_addr(addr))
-		return MLX5E_UC;
-
-	if ((addr[0] == 0x01) &&
-	    (addr[1] == 0x00) &&
-	    (addr[2] == 0x5e) &&
-	   !(addr[3] &  0x80))
-		return MLX5E_MC_IPV4;
-
-	if ((addr[0] == 0x33) &&
-	    (addr[1] == 0x33))
-		return MLX5E_MC_IPV6;
-
-	return MLX5E_MC_OTHER;
-}
-
-static u32 mlx5e_get_tt_vec(struct mlx5e_eth_addr_info *ai, int type)
-{
-	int eth_addr_type;
-	u32 ret;
-
-	switch (type) {
-	case MLX5E_FULLMATCH:
-		eth_addr_type = mlx5e_get_eth_addr_type(ai->addr);
-		switch (eth_addr_type) {
-		case MLX5E_UC:
-			ret =
-				BIT(MLX5E_TT_IPV4_TCP)       |
-				BIT(MLX5E_TT_IPV6_TCP)       |
-				BIT(MLX5E_TT_IPV4_UDP)       |
-				BIT(MLX5E_TT_IPV6_UDP)       |
-				BIT(MLX5E_TT_IPV4_IPSEC_AH)  |
-				BIT(MLX5E_TT_IPV6_IPSEC_AH)  |
-				BIT(MLX5E_TT_IPV4_IPSEC_ESP) |
-				BIT(MLX5E_TT_IPV6_IPSEC_ESP) |
-				BIT(MLX5E_TT_IPV4)           |
-				BIT(MLX5E_TT_IPV6)           |
-				BIT(MLX5E_TT_ANY)            |
-				0;
-			break;
-
-		case MLX5E_MC_IPV4:
-			ret =
-				BIT(MLX5E_TT_IPV4_UDP)       |
-				BIT(MLX5E_TT_IPV4)           |
-				0;
-			break;
-
-		case MLX5E_MC_IPV6:
-			ret =
-				BIT(MLX5E_TT_IPV6_UDP)       |
-				BIT(MLX5E_TT_IPV6)           |
-				0;
-			break;
-
-		case MLX5E_MC_OTHER:
-			ret =
-				BIT(MLX5E_TT_ANY)            |
-				0;
-			break;
-		}
-
-		break;
-
-	case MLX5E_ALLMULTI:
-		ret =
-			BIT(MLX5E_TT_IPV4_UDP) |
-			BIT(MLX5E_TT_IPV6_UDP) |
-			BIT(MLX5E_TT_IPV4)     |
-			BIT(MLX5E_TT_IPV6)     |
-			BIT(MLX5E_TT_ANY)      |
-			0;
-		break;
-
-	default: /* MLX5E_PROMISC */
-		ret =
-			BIT(MLX5E_TT_IPV4_TCP)       |
-			BIT(MLX5E_TT_IPV6_TCP)       |
-			BIT(MLX5E_TT_IPV4_UDP)       |
-			BIT(MLX5E_TT_IPV6_UDP)       |
-			BIT(MLX5E_TT_IPV4_IPSEC_AH)  |
-			BIT(MLX5E_TT_IPV6_IPSEC_AH)  |
-			BIT(MLX5E_TT_IPV4_IPSEC_ESP) |
-			BIT(MLX5E_TT_IPV6_IPSEC_ESP) |
-			BIT(MLX5E_TT_IPV4)           |
-			BIT(MLX5E_TT_IPV6)           |
-			BIT(MLX5E_TT_ANY)            |
-			0;
-		break;
-	}
-
-	return ret;
-}
-
-static int __mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv,
-				     struct mlx5e_eth_addr_info *ai, int type,
-				     void *flow_context, void *match_criteria)
-{
-	u8 match_criteria_enable = 0;
-	void *match_value;
-	void *dest;
-	u8   *dmac;
-	u8   *match_criteria_dmac;
-	void *ft   = priv->ft.main;
-	u32  *tirn = priv->tirn;
-	u32  *ft_ix;
-	u32  tt_vec;
-	int  err;
-
-	match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value);
-	dmac = MLX5_ADDR_OF(fte_match_param, match_value,
-			    outer_headers.dmac_47_16);
-	match_criteria_dmac = MLX5_ADDR_OF(fte_match_param, match_criteria,
-					   outer_headers.dmac_47_16);
-	dest = MLX5_ADDR_OF(flow_context, flow_context, destination);
-
-	MLX5_SET(flow_context, flow_context, action,
-		 MLX5_FLOW_CONTEXT_ACTION_FWD_DEST);
-	MLX5_SET(flow_context, flow_context, destination_list_size, 1);
-	MLX5_SET(dest_format_struct, dest, destination_type,
-		 MLX5_FLOW_CONTEXT_DEST_TYPE_TIR);
-
-	switch (type) {
-	case MLX5E_FULLMATCH:
-		match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-		memset(match_criteria_dmac, 0xff, ETH_ALEN);
-		ether_addr_copy(dmac, ai->addr);
-		break;
-
-	case MLX5E_ALLMULTI:
-		match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-		match_criteria_dmac[0] = 0x01;
-		dmac[0] = 0x01;
-		break;
-
-	case MLX5E_PROMISC:
-		break;
-	}
-
-	tt_vec = mlx5e_get_tt_vec(ai, type);
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_ANY];
-	if (tt_vec & BIT(MLX5E_TT_ANY)) {
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_ANY]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_ANY);
-	}
-
-	match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-			 outer_headers.ethertype);
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4];
-	if (tt_vec & BIT(MLX5E_TT_IPV4)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV4);
-	}
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6];
-	if (tt_vec & BIT(MLX5E_TT_IPV6)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV6);
-	}
-
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-			 outer_headers.ip_protocol);
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_UDP);
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_UDP];
-	if (tt_vec & BIT(MLX5E_TT_IPV4_UDP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_UDP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV4_UDP);
-	}
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_UDP];
-	if (tt_vec & BIT(MLX5E_TT_IPV6_UDP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_UDP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV6_UDP);
-	}
-
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_TCP);
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_TCP];
-	if (tt_vec & BIT(MLX5E_TT_IPV4_TCP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_TCP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV4_TCP);
-	}
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_TCP];
-	if (tt_vec & BIT(MLX5E_TT_IPV6_TCP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_TCP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV6_TCP);
-	}
-
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_AH);
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_IPSEC_AH];
-	if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_IPSEC_AH]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_AH);
-	}
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_IPSEC_AH];
-	if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_IPSEC_AH]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_AH);
-	}
-
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_ESP);
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_IPSEC_ESP];
-	if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_IPSEC_ESP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_ESP);
-	}
-
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_IPSEC_ESP];
-	if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
-			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_IPSEC_ESP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
-			goto err_del_ai;
-
-		ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_ESP);
-	}
-
-	return 0;
-
-err_del_ai:
-	mlx5e_del_eth_addr_from_flow_table(priv, ai);
-
-	return err;
-}
-
-static int mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv,
-				   struct mlx5e_eth_addr_info *ai, int type)
-{
-	u32 *flow_context;
-	u32 *match_criteria;
-	int err;
-
-	flow_context   = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) +
-				      MLX5_ST_SZ_BYTES(dest_format_struct));
-	match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
-	if (!flow_context || !match_criteria) {
-		netdev_err(priv->netdev, "%s: alloc failed\n", __func__);
-		err = -ENOMEM;
-		goto add_eth_addr_rule_out;
-	}
-
-	err = __mlx5e_add_eth_addr_rule(priv, ai, type, flow_context,
-					match_criteria);
-	if (err)
-		netdev_err(priv->netdev, "%s: failed\n", __func__);
-
-add_eth_addr_rule_out:
-	kvfree(match_criteria);
-	kvfree(flow_context);
-	return err;
-}
-
-enum mlx5e_vlan_rule_type {
-	MLX5E_VLAN_RULE_TYPE_UNTAGGED,
-	MLX5E_VLAN_RULE_TYPE_ANY_VID,
-	MLX5E_VLAN_RULE_TYPE_MATCH_VID,
-};
-
-static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv,
-			       enum mlx5e_vlan_rule_type rule_type, u16 vid)
-{
-	u8 match_criteria_enable = 0;
-	u32 *flow_context;
-	void *match_value;
-	void *dest;
-	u32 *match_criteria;
-	u32 *ft_ix;
-	int err;
-
-	flow_context   = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) +
-				      MLX5_ST_SZ_BYTES(dest_format_struct));
-	match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
-	if (!flow_context || !match_criteria) {
-		netdev_err(priv->netdev, "%s: alloc failed\n", __func__);
-		err = -ENOMEM;
-		goto add_vlan_rule_out;
-	}
-	match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value);
-	dest = MLX5_ADDR_OF(flow_context, flow_context, destination);
-
-	MLX5_SET(flow_context, flow_context, action,
-		 MLX5_FLOW_CONTEXT_ACTION_FWD_DEST);
-	MLX5_SET(flow_context, flow_context, destination_list_size, 1);
-	MLX5_SET(dest_format_struct, dest, destination_type,
-		 MLX5_FLOW_CONTEXT_DEST_TYPE_FLOW_TABLE);
-	MLX5_SET(dest_format_struct, dest, destination_id,
-		 mlx5_get_flow_table_id(priv->ft.main));
-
-	match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-			 outer_headers.vlan_tag);
-
-	switch (rule_type) {
-	case MLX5E_VLAN_RULE_TYPE_UNTAGGED:
-		ft_ix = &priv->vlan.untagged_rule_ft_ix;
-		break;
-	case MLX5E_VLAN_RULE_TYPE_ANY_VID:
-		ft_ix = &priv->vlan.any_vlan_rule_ft_ix;
-		MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag,
-			 1);
-		break;
-	default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */
-		ft_ix = &priv->vlan.active_vlans_ft_ix[vid];
-		MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag,
-			 1);
-		MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-				 outer_headers.first_vid);
-		MLX5_SET(fte_match_param, match_value, outer_headers.first_vid,
-			 vid);
-		break;
-	}
-
-	err = mlx5_add_flow_table_entry(priv->ft.vlan, match_criteria_enable,
-					match_criteria, flow_context, ft_ix);
-	if (err)
-		netdev_err(priv->netdev, "%s: failed\n", __func__);
-
-add_vlan_rule_out:
-	kvfree(match_criteria);
-	kvfree(flow_context);
-	return err;
-}
-
-static void mlx5e_del_vlan_rule(struct mlx5e_priv *priv,
-				enum mlx5e_vlan_rule_type rule_type, u16 vid)
-{
-	switch (rule_type) {
-	case MLX5E_VLAN_RULE_TYPE_UNTAGGED:
-		mlx5_del_flow_table_entry(priv->ft.vlan,
-					  priv->vlan.untagged_rule_ft_ix);
-		break;
-	case MLX5E_VLAN_RULE_TYPE_ANY_VID:
-		mlx5_del_flow_table_entry(priv->ft.vlan,
-					  priv->vlan.any_vlan_rule_ft_ix);
-		break;
-	case MLX5E_VLAN_RULE_TYPE_MATCH_VID:
-		mlx5_del_flow_table_entry(priv->ft.vlan,
-					  priv->vlan.active_vlans_ft_ix[vid]);
-		break;
-	}
-}
-
-void mlx5e_enable_vlan_filter(struct mlx5e_priv *priv)
-{
-	if (!priv->vlan.filter_disabled)
-		return;
-
-	priv->vlan.filter_disabled = false;
-	if (priv->netdev->flags & IFF_PROMISC)
-		return;
-	mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, 0);
-}
-
-void mlx5e_disable_vlan_filter(struct mlx5e_priv *priv)
-{
-	if (priv->vlan.filter_disabled)
-		return;
-
-	priv->vlan.filter_disabled = true;
-	if (priv->netdev->flags & IFF_PROMISC)
-		return;
-	mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, 0);
-}
-
-int mlx5e_vlan_rx_add_vid(struct net_device *dev, __always_unused __be16 proto,
-			  u16 vid)
-{
-	struct mlx5e_priv *priv = netdev_priv(dev);
-
-	return mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, vid);
-}
-
-int mlx5e_vlan_rx_kill_vid(struct net_device *dev, __always_unused __be16 proto,
-			   u16 vid)
-{
-	struct mlx5e_priv *priv = netdev_priv(dev);
-
-	mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, vid);
-
-	return 0;
-}
-
-#define mlx5e_for_each_hash_node(hn, tmp, hash, i) \
-	for (i = 0; i < MLX5E_ETH_ADDR_HASH_SIZE; i++) \
-		hlist_for_each_entry_safe(hn, tmp, &hash[i], hlist)
-
-static void mlx5e_execute_action(struct mlx5e_priv *priv,
-				 struct mlx5e_eth_addr_hash_node *hn)
-{
-	switch (hn->action) {
-	case MLX5E_ACTION_ADD:
-		mlx5e_add_eth_addr_rule(priv, &hn->ai, MLX5E_FULLMATCH);
-		hn->action = MLX5E_ACTION_NONE;
-		break;
-
-	case MLX5E_ACTION_DEL:
-		mlx5e_del_eth_addr_from_flow_table(priv, &hn->ai);
-		mlx5e_del_eth_addr_from_hash(hn);
-		break;
-	}
-}
-
-static void mlx5e_sync_netdev_addr(struct mlx5e_priv *priv)
-{
-	struct net_device *netdev = priv->netdev;
-	struct netdev_hw_addr *ha;
-
-	netif_addr_lock_bh(netdev);
-
-	mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_uc,
-				   priv->netdev->dev_addr);
-
-	netdev_for_each_uc_addr(ha, netdev)
-		mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_uc, ha->addr);
-
-	netdev_for_each_mc_addr(ha, netdev)
-		mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_mc, ha->addr);
-
-	netif_addr_unlock_bh(netdev);
-}
-
-static void mlx5e_apply_netdev_addr(struct mlx5e_priv *priv)
-{
-	struct mlx5e_eth_addr_hash_node *hn;
-	struct hlist_node *tmp;
-	int i;
-
-	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_uc, i)
-		mlx5e_execute_action(priv, hn);
-
-	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_mc, i)
-		mlx5e_execute_action(priv, hn);
-}
-
-static void mlx5e_handle_netdev_addr(struct mlx5e_priv *priv)
-{
-	struct mlx5e_eth_addr_hash_node *hn;
-	struct hlist_node *tmp;
-	int i;
-
-	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_uc, i)
-		hn->action = MLX5E_ACTION_DEL;
-	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_mc, i)
-		hn->action = MLX5E_ACTION_DEL;
-
-	if (!test_bit(MLX5E_STATE_DESTROYING, &priv->state))
-		mlx5e_sync_netdev_addr(priv);
-
-	mlx5e_apply_netdev_addr(priv);
-}
-
-void mlx5e_set_rx_mode_work(struct work_struct *work)
-{
-	struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv,
-					       set_rx_mode_work);
-
-	struct mlx5e_eth_addr_db *ea = &priv->eth_addr;
-	struct net_device *ndev = priv->netdev;
-
-	bool rx_mode_enable   = !test_bit(MLX5E_STATE_DESTROYING, &priv->state);
-	bool promisc_enabled   = rx_mode_enable && (ndev->flags & IFF_PROMISC);
-	bool allmulti_enabled  = rx_mode_enable && (ndev->flags & IFF_ALLMULTI);
-	bool broadcast_enabled = rx_mode_enable;
-
-	bool enable_promisc    = !ea->promisc_enabled   &&  promisc_enabled;
-	bool disable_promisc   =  ea->promisc_enabled   && !promisc_enabled;
-	bool enable_allmulti   = !ea->allmulti_enabled  &&  allmulti_enabled;
-	bool disable_allmulti  =  ea->allmulti_enabled  && !allmulti_enabled;
-	bool enable_broadcast  = !ea->broadcast_enabled &&  broadcast_enabled;
-	bool disable_broadcast =  ea->broadcast_enabled && !broadcast_enabled;
-
-	if (enable_promisc) {
-		mlx5e_add_eth_addr_rule(priv, &ea->promisc, MLX5E_PROMISC);
-		if (!priv->vlan.filter_disabled)
-			mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID,
-					    0);
-	}
-	if (enable_allmulti)
-		mlx5e_add_eth_addr_rule(priv, &ea->allmulti, MLX5E_ALLMULTI);
-	if (enable_broadcast)
-		mlx5e_add_eth_addr_rule(priv, &ea->broadcast, MLX5E_FULLMATCH);
-
-	mlx5e_handle_netdev_addr(priv);
-
-	if (disable_broadcast)
-		mlx5e_del_eth_addr_from_flow_table(priv, &ea->broadcast);
-	if (disable_allmulti)
-		mlx5e_del_eth_addr_from_flow_table(priv, &ea->allmulti);
-	if (disable_promisc) {
-		if (!priv->vlan.filter_disabled)
-			mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID,
-					    0);
-		mlx5e_del_eth_addr_from_flow_table(priv, &ea->promisc);
-	}
-
-	ea->promisc_enabled   = promisc_enabled;
-	ea->allmulti_enabled  = allmulti_enabled;
-	ea->broadcast_enabled = broadcast_enabled;
-}
-
-void mlx5e_init_eth_addr(struct mlx5e_priv *priv)
-{
-	ether_addr_copy(priv->eth_addr.broadcast.addr, priv->netdev->broadcast);
-}
-
-static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv)
-{
-	struct mlx5_flow_table_group *g;
-	u8 *dmac;
-
-	g = kcalloc(9, sizeof(*g), GFP_KERNEL);
-	if (!g)
-		return -ENOMEM;
-
-	g[0].log_sz = 3;
-	g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.ethertype);
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.ip_protocol);
-
-	g[1].log_sz = 1;
-	g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria,
-			 outer_headers.ethertype);
-
-	g[2].log_sz = 0;
-
-	g[3].log_sz = 14;
-	g[3].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[3].match_criteria,
-			    outer_headers.dmac_47_16);
-	memset(dmac, 0xff, ETH_ALEN);
-	MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria,
-			 outer_headers.ethertype);
-	MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria,
-			 outer_headers.ip_protocol);
-
-	g[4].log_sz = 13;
-	g[4].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[4].match_criteria,
-			    outer_headers.dmac_47_16);
-	memset(dmac, 0xff, ETH_ALEN);
-	MLX5_SET_TO_ONES(fte_match_param, g[4].match_criteria,
-			 outer_headers.ethertype);
-
-	g[5].log_sz = 11;
-	g[5].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[5].match_criteria,
-			    outer_headers.dmac_47_16);
-	memset(dmac, 0xff, ETH_ALEN);
-
-	g[6].log_sz = 2;
-	g[6].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[6].match_criteria,
-			    outer_headers.dmac_47_16);
-	dmac[0] = 0x01;
-	MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria,
-			 outer_headers.ethertype);
-	MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria,
-			 outer_headers.ip_protocol);
-
-	g[7].log_sz = 1;
-	g[7].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[7].match_criteria,
-			    outer_headers.dmac_47_16);
-	dmac[0] = 0x01;
-	MLX5_SET_TO_ONES(fte_match_param, g[7].match_criteria,
-			 outer_headers.ethertype);
-
-	g[8].log_sz = 0;
-	g[8].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[8].match_criteria,
-			    outer_headers.dmac_47_16);
-	dmac[0] = 0x01;
-	priv->ft.main = mlx5_create_flow_table(priv->mdev, 1,
-					       MLX5_FLOW_TABLE_TYPE_NIC_RCV,
-					       9, g);
-	kfree(g);
-
-	return priv->ft.main ? 0 : -ENOMEM;
-}
-
-static void mlx5e_destroy_main_flow_table(struct mlx5e_priv *priv)
-{
-	mlx5_destroy_flow_table(priv->ft.main);
-}
-
-static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv)
-{
-	struct mlx5_flow_table_group *g;
-
-	g = kcalloc(2, sizeof(*g), GFP_KERNEL);
-	if (!g)
-		return -ENOMEM;
-
-	g[0].log_sz = 12;
-	g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.vlan_tag);
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.first_vid);
-
-	/* untagged + any vlan id */
-	g[1].log_sz = 1;
-	g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria,
-			 outer_headers.vlan_tag);
-
-	priv->ft.vlan = mlx5_create_flow_table(priv->mdev, 0,
-					       MLX5_FLOW_TABLE_TYPE_NIC_RCV,
-					       2, g);
-
-	kfree(g);
-	return priv->ft.vlan ? 0 : -ENOMEM;
-}
-
-static void mlx5e_destroy_vlan_flow_table(struct mlx5e_priv *priv)
-{
-	mlx5_destroy_flow_table(priv->ft.vlan);
-}
-
-int mlx5e_create_flow_tables(struct mlx5e_priv *priv)
-{
-	int err;
-
-	err = mlx5e_create_main_flow_table(priv);
-	if (err)
-		return err;
-
-	err = mlx5e_create_vlan_flow_table(priv);
-	if (err)
-		goto err_destroy_main_flow_table;
-
-	err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0);
-	if (err)
-		goto err_destroy_vlan_flow_table;
-
-	return 0;
-
-err_destroy_vlan_flow_table:
-	mlx5e_destroy_vlan_flow_table(priv);
-
-err_destroy_main_flow_table:
-	mlx5e_destroy_main_flow_table(priv);
-
-	return err;
-}
-
-void mlx5e_destroy_flow_tables(struct mlx5e_priv *priv)
-{
-	mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0);
-	mlx5e_destroy_vlan_flow_table(priv);
-	mlx5e_destroy_main_flow_table(priv);
-}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
new file mode 100644
index 000000000..80d81abc4
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -0,0 +1,1224 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/list.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/tcp.h>
+#include <linux/mlx5/fs.h>
+#include "en.h"
+
+#define MLX5_SET_CFG(p, f, v) MLX5_SET(create_flow_group_in, p, f, v)
+
+enum {
+	MLX5E_FULLMATCH = 0,
+	MLX5E_ALLMULTI  = 1,
+	MLX5E_PROMISC   = 2,
+};
+
+enum {
+	MLX5E_UC        = 0,
+	MLX5E_MC_IPV4   = 1,
+	MLX5E_MC_IPV6   = 2,
+	MLX5E_MC_OTHER  = 3,
+};
+
+enum {
+	MLX5E_ACTION_NONE = 0,
+	MLX5E_ACTION_ADD  = 1,
+	MLX5E_ACTION_DEL  = 2,
+};
+
+struct mlx5e_eth_addr_hash_node {
+	struct hlist_node          hlist;
+	u8                         action;
+	struct mlx5e_eth_addr_info ai;
+};
+
+static inline int mlx5e_hash_eth_addr(u8 *addr)
+{
+	return addr[5];
+}
+
+static void mlx5e_add_eth_addr_to_hash(struct hlist_head *hash, u8 *addr)
+{
+	struct mlx5e_eth_addr_hash_node *hn;
+	int ix = mlx5e_hash_eth_addr(addr);
+	int found = 0;
+
+	hlist_for_each_entry(hn, &hash[ix], hlist)
+		if (ether_addr_equal_64bits(hn->ai.addr, addr)) {
+			found = 1;
+			break;
+		}
+
+	if (found) {
+		hn->action = MLX5E_ACTION_NONE;
+		return;
+	}
+
+	hn = kzalloc(sizeof(*hn), GFP_ATOMIC);
+	if (!hn)
+		return;
+
+	ether_addr_copy(hn->ai.addr, addr);
+	hn->action = MLX5E_ACTION_ADD;
+
+	hlist_add_head(&hn->hlist, &hash[ix]);
+}
+
+static void mlx5e_del_eth_addr_from_hash(struct mlx5e_eth_addr_hash_node *hn)
+{
+	hlist_del(&hn->hlist);
+	kfree(hn);
+}
+
+static void mlx5e_del_eth_addr_from_flow_table(struct mlx5e_priv *priv,
+					       struct mlx5e_eth_addr_info *ai)
+{
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_IPSEC_ESP]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_IPSEC_ESP]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_IPSEC_AH]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_IPSEC_AH]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_TCP))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_TCP]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_TCP))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_TCP]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_UDP))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_UDP]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_UDP))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_UDP]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV6))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_IPV4))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4]);
+
+	if (ai->tt_vec & BIT(MLX5E_TT_ANY))
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_ANY]);
+}
+
+static int mlx5e_get_eth_addr_type(u8 *addr)
+{
+	if (is_unicast_ether_addr(addr))
+		return MLX5E_UC;
+
+	if ((addr[0] == 0x01) &&
+	    (addr[1] == 0x00) &&
+	    (addr[2] == 0x5e) &&
+	   !(addr[3] &  0x80))
+		return MLX5E_MC_IPV4;
+
+	if ((addr[0] == 0x33) &&
+	    (addr[1] == 0x33))
+		return MLX5E_MC_IPV6;
+
+	return MLX5E_MC_OTHER;
+}
+
+static u32 mlx5e_get_tt_vec(struct mlx5e_eth_addr_info *ai, int type)
+{
+	int eth_addr_type;
+	u32 ret;
+
+	switch (type) {
+	case MLX5E_FULLMATCH:
+		eth_addr_type = mlx5e_get_eth_addr_type(ai->addr);
+		switch (eth_addr_type) {
+		case MLX5E_UC:
+			ret =
+				BIT(MLX5E_TT_IPV4_TCP)       |
+				BIT(MLX5E_TT_IPV6_TCP)       |
+				BIT(MLX5E_TT_IPV4_UDP)       |
+				BIT(MLX5E_TT_IPV6_UDP)       |
+				BIT(MLX5E_TT_IPV4_IPSEC_AH)  |
+				BIT(MLX5E_TT_IPV6_IPSEC_AH)  |
+				BIT(MLX5E_TT_IPV4_IPSEC_ESP) |
+				BIT(MLX5E_TT_IPV6_IPSEC_ESP) |
+				BIT(MLX5E_TT_IPV4)           |
+				BIT(MLX5E_TT_IPV6)           |
+				BIT(MLX5E_TT_ANY)            |
+				0;
+			break;
+
+		case MLX5E_MC_IPV4:
+			ret =
+				BIT(MLX5E_TT_IPV4_UDP)       |
+				BIT(MLX5E_TT_IPV4)           |
+				0;
+			break;
+
+		case MLX5E_MC_IPV6:
+			ret =
+				BIT(MLX5E_TT_IPV6_UDP)       |
+				BIT(MLX5E_TT_IPV6)           |
+				0;
+			break;
+
+		case MLX5E_MC_OTHER:
+			ret =
+				BIT(MLX5E_TT_ANY)            |
+				0;
+			break;
+		}
+
+		break;
+
+	case MLX5E_ALLMULTI:
+		ret =
+			BIT(MLX5E_TT_IPV4_UDP) |
+			BIT(MLX5E_TT_IPV6_UDP) |
+			BIT(MLX5E_TT_IPV4)     |
+			BIT(MLX5E_TT_IPV6)     |
+			BIT(MLX5E_TT_ANY)      |
+			0;
+		break;
+
+	default: /* MLX5E_PROMISC */
+		ret =
+			BIT(MLX5E_TT_IPV4_TCP)       |
+			BIT(MLX5E_TT_IPV6_TCP)       |
+			BIT(MLX5E_TT_IPV4_UDP)       |
+			BIT(MLX5E_TT_IPV6_UDP)       |
+			BIT(MLX5E_TT_IPV4_IPSEC_AH)  |
+			BIT(MLX5E_TT_IPV6_IPSEC_AH)  |
+			BIT(MLX5E_TT_IPV4_IPSEC_ESP) |
+			BIT(MLX5E_TT_IPV6_IPSEC_ESP) |
+			BIT(MLX5E_TT_IPV4)           |
+			BIT(MLX5E_TT_IPV6)           |
+			BIT(MLX5E_TT_ANY)            |
+			0;
+		break;
+	}
+
+	return ret;
+}
+
+static int __mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv,
+				     struct mlx5e_eth_addr_info *ai,
+				     int type, u32 *mc, u32 *mv)
+{
+	struct mlx5_flow_destination dest;
+	u8 match_criteria_enable = 0;
+	struct mlx5_flow_rule **rule_p;
+	struct mlx5_flow_table *ft = priv->fts.main.t;
+	u8 *mc_dmac = MLX5_ADDR_OF(fte_match_param, mc,
+				   outer_headers.dmac_47_16);
+	u8 *mv_dmac = MLX5_ADDR_OF(fte_match_param, mv,
+				   outer_headers.dmac_47_16);
+	u32 *tirn = priv->tirn;
+	u32 tt_vec;
+	int err = 0;
+
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR;
+
+	switch (type) {
+	case MLX5E_FULLMATCH:
+		match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+		eth_broadcast_addr(mc_dmac);
+		ether_addr_copy(mv_dmac, ai->addr);
+		break;
+
+	case MLX5E_ALLMULTI:
+		match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+		mc_dmac[0] = 0x01;
+		mv_dmac[0] = 0x01;
+		break;
+
+	case MLX5E_PROMISC:
+		break;
+	}
+
+	tt_vec = mlx5e_get_tt_vec(ai, type);
+
+	if (tt_vec & BIT(MLX5E_TT_ANY)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_ANY];
+		dest.tir_num = tirn[MLX5E_TT_ANY];
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_ANY);
+	}
+
+	match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+
+	if (tt_vec & BIT(MLX5E_TT_IPV4)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4];
+		dest.tir_num = tirn[MLX5E_TT_IPV4];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IP);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV4);
+	}
+
+	if (tt_vec & BIT(MLX5E_TT_IPV6)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6];
+		dest.tir_num = tirn[MLX5E_TT_IPV6];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IPV6);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV6);
+	}
+
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_UDP);
+
+	if (tt_vec & BIT(MLX5E_TT_IPV4_UDP)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_UDP];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_UDP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IP);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV4_UDP);
+	}
+
+	if (tt_vec & BIT(MLX5E_TT_IPV6_UDP)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_UDP];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_UDP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IPV6);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV6_UDP);
+	}
+
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_TCP);
+
+	if (tt_vec & BIT(MLX5E_TT_IPV4_TCP)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_TCP];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_TCP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IP);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV4_TCP);
+	}
+
+	if (tt_vec & BIT(MLX5E_TT_IPV6_TCP)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_TCP];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_TCP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IPV6);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+
+		ai->tt_vec |= BIT(MLX5E_TT_IPV6_TCP);
+	}
+
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_AH);
+
+	if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_IPSEC_AH];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_IPSEC_AH];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IP);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_AH);
+	}
+
+	if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_IPSEC_AH];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_IPSEC_AH];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IPV6);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_AH);
+	}
+
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_ESP);
+
+	if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_IPSEC_ESP];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_IPSEC_ESP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IP);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_ESP);
+	}
+
+	if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP)) {
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_IPSEC_ESP];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_IPSEC_ESP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
+			 ETH_P_IPV6);
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
+			goto err_del_ai;
+		ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_ESP);
+	}
+
+	return 0;
+
+err_del_ai:
+	err = PTR_ERR(*rule_p);
+	*rule_p = NULL;
+	mlx5e_del_eth_addr_from_flow_table(priv, ai);
+
+	return err;
+}
+
+static int mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv,
+				   struct mlx5e_eth_addr_info *ai, int type)
+{
+	u32 *match_criteria;
+	u32 *match_value;
+	int err = 0;
+
+	match_value	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	match_criteria	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	if (!match_value || !match_criteria) {
+		netdev_err(priv->netdev, "%s: alloc failed\n", __func__);
+		err = -ENOMEM;
+		goto add_eth_addr_rule_out;
+	}
+
+	err = __mlx5e_add_eth_addr_rule(priv, ai, type, match_criteria,
+					match_value);
+
+add_eth_addr_rule_out:
+	kvfree(match_criteria);
+	kvfree(match_value);
+
+	return err;
+}
+
+static int mlx5e_vport_context_update_vlans(struct mlx5e_priv *priv)
+{
+	struct net_device *ndev = priv->netdev;
+	int max_list_size;
+	int list_size;
+	u16 *vlans;
+	int vlan;
+	int err;
+	int i;
+
+	list_size = 0;
+	for_each_set_bit(vlan, priv->vlan.active_vlans, VLAN_N_VID)
+		list_size++;
+
+	max_list_size = 1 << MLX5_CAP_GEN(priv->mdev, log_max_vlan_list);
+
+	if (list_size > max_list_size) {
+		netdev_warn(ndev,
+			    "netdev vlans list size (%d) > (%d) max vport list size, some vlans will be dropped\n",
+			    list_size, max_list_size);
+		list_size = max_list_size;
+	}
+
+	vlans = kcalloc(list_size, sizeof(*vlans), GFP_KERNEL);
+	if (!vlans)
+		return -ENOMEM;
+
+	i = 0;
+	for_each_set_bit(vlan, priv->vlan.active_vlans, VLAN_N_VID) {
+		if (i >= list_size)
+			break;
+		vlans[i++] = vlan;
+	}
+
+	err = mlx5_modify_nic_vport_vlans(priv->mdev, vlans, list_size);
+	if (err)
+		netdev_err(ndev, "Failed to modify vport vlans list err(%d)\n",
+			   err);
+
+	kfree(vlans);
+	return err;
+}
+
+enum mlx5e_vlan_rule_type {
+	MLX5E_VLAN_RULE_TYPE_UNTAGGED,
+	MLX5E_VLAN_RULE_TYPE_ANY_VID,
+	MLX5E_VLAN_RULE_TYPE_MATCH_VID,
+};
+
+static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv,
+				 enum mlx5e_vlan_rule_type rule_type,
+				 u16 vid, u32 *mc, u32 *mv)
+{
+	struct mlx5_flow_table *ft = priv->fts.vlan.t;
+	struct mlx5_flow_destination dest;
+	u8 match_criteria_enable = 0;
+	struct mlx5_flow_rule **rule_p;
+	int err = 0;
+
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest.ft = priv->fts.main.t;
+
+	match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag);
+
+	switch (rule_type) {
+	case MLX5E_VLAN_RULE_TYPE_UNTAGGED:
+		rule_p = &priv->vlan.untagged_rule;
+		break;
+	case MLX5E_VLAN_RULE_TYPE_ANY_VID:
+		rule_p = &priv->vlan.any_vlan_rule;
+		MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1);
+		break;
+	default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */
+		rule_p = &priv->vlan.active_vlans_rule[vid];
+		MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1);
+		MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid);
+		MLX5_SET(fte_match_param, mv, outer_headers.first_vid, vid);
+		break;
+	}
+
+	*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+				     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+				     MLX5_FS_DEFAULT_FLOW_TAG,
+				     &dest);
+
+	if (IS_ERR(*rule_p)) {
+		err = PTR_ERR(*rule_p);
+		*rule_p = NULL;
+		netdev_err(priv->netdev, "%s: add rule failed\n", __func__);
+	}
+
+	return err;
+}
+
+static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv,
+			       enum mlx5e_vlan_rule_type rule_type, u16 vid)
+{
+	u32 *match_criteria;
+	u32 *match_value;
+	int err = 0;
+
+	match_value	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	match_criteria	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	if (!match_value || !match_criteria) {
+		netdev_err(priv->netdev, "%s: alloc failed\n", __func__);
+		err = -ENOMEM;
+		goto add_vlan_rule_out;
+	}
+
+	if (rule_type == MLX5E_VLAN_RULE_TYPE_MATCH_VID)
+		mlx5e_vport_context_update_vlans(priv);
+
+	err = __mlx5e_add_vlan_rule(priv, rule_type, vid, match_criteria,
+				    match_value);
+
+add_vlan_rule_out:
+	kvfree(match_criteria);
+	kvfree(match_value);
+
+	return err;
+}
+
+static void mlx5e_del_vlan_rule(struct mlx5e_priv *priv,
+				enum mlx5e_vlan_rule_type rule_type, u16 vid)
+{
+	switch (rule_type) {
+	case MLX5E_VLAN_RULE_TYPE_UNTAGGED:
+		if (priv->vlan.untagged_rule) {
+			mlx5_del_flow_rule(priv->vlan.untagged_rule);
+			priv->vlan.untagged_rule = NULL;
+		}
+		break;
+	case MLX5E_VLAN_RULE_TYPE_ANY_VID:
+		if (priv->vlan.any_vlan_rule) {
+			mlx5_del_flow_rule(priv->vlan.any_vlan_rule);
+			priv->vlan.any_vlan_rule = NULL;
+		}
+		break;
+	case MLX5E_VLAN_RULE_TYPE_MATCH_VID:
+		mlx5e_vport_context_update_vlans(priv);
+		if (priv->vlan.active_vlans_rule[vid]) {
+			mlx5_del_flow_rule(priv->vlan.active_vlans_rule[vid]);
+			priv->vlan.active_vlans_rule[vid] = NULL;
+		}
+		mlx5e_vport_context_update_vlans(priv);
+		break;
+	}
+}
+
+void mlx5e_enable_vlan_filter(struct mlx5e_priv *priv)
+{
+	if (!priv->vlan.filter_disabled)
+		return;
+
+	priv->vlan.filter_disabled = false;
+	if (priv->netdev->flags & IFF_PROMISC)
+		return;
+	mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, 0);
+}
+
+void mlx5e_disable_vlan_filter(struct mlx5e_priv *priv)
+{
+	if (priv->vlan.filter_disabled)
+		return;
+
+	priv->vlan.filter_disabled = true;
+	if (priv->netdev->flags & IFF_PROMISC)
+		return;
+	mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, 0);
+}
+
+int mlx5e_vlan_rx_add_vid(struct net_device *dev, __always_unused __be16 proto,
+			  u16 vid)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+
+	set_bit(vid, priv->vlan.active_vlans);
+
+	return mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, vid);
+}
+
+int mlx5e_vlan_rx_kill_vid(struct net_device *dev, __always_unused __be16 proto,
+			   u16 vid)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+
+	clear_bit(vid, priv->vlan.active_vlans);
+
+	mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, vid);
+
+	return 0;
+}
+
+#define mlx5e_for_each_hash_node(hn, tmp, hash, i) \
+	for (i = 0; i < MLX5E_ETH_ADDR_HASH_SIZE; i++) \
+		hlist_for_each_entry_safe(hn, tmp, &hash[i], hlist)
+
+static void mlx5e_execute_action(struct mlx5e_priv *priv,
+				 struct mlx5e_eth_addr_hash_node *hn)
+{
+	switch (hn->action) {
+	case MLX5E_ACTION_ADD:
+		mlx5e_add_eth_addr_rule(priv, &hn->ai, MLX5E_FULLMATCH);
+		hn->action = MLX5E_ACTION_NONE;
+		break;
+
+	case MLX5E_ACTION_DEL:
+		mlx5e_del_eth_addr_from_flow_table(priv, &hn->ai);
+		mlx5e_del_eth_addr_from_hash(hn);
+		break;
+	}
+}
+
+static void mlx5e_sync_netdev_addr(struct mlx5e_priv *priv)
+{
+	struct net_device *netdev = priv->netdev;
+	struct netdev_hw_addr *ha;
+
+	netif_addr_lock_bh(netdev);
+
+	mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_uc,
+				   priv->netdev->dev_addr);
+
+	netdev_for_each_uc_addr(ha, netdev)
+		mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_uc, ha->addr);
+
+	netdev_for_each_mc_addr(ha, netdev)
+		mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_mc, ha->addr);
+
+	netif_addr_unlock_bh(netdev);
+}
+
+static void mlx5e_fill_addr_array(struct mlx5e_priv *priv, int list_type,
+				  u8 addr_array[][ETH_ALEN], int size)
+{
+	bool is_uc = (list_type == MLX5_NVPRT_LIST_TYPE_UC);
+	struct net_device *ndev = priv->netdev;
+	struct mlx5e_eth_addr_hash_node *hn;
+	struct hlist_head *addr_list;
+	struct hlist_node *tmp;
+	int i = 0;
+	int hi;
+
+	addr_list = is_uc ? priv->eth_addr.netdev_uc : priv->eth_addr.netdev_mc;
+
+	if (is_uc) /* Make sure our own address is pushed first */
+		ether_addr_copy(addr_array[i++], ndev->dev_addr);
+	else if (priv->eth_addr.broadcast_enabled)
+		ether_addr_copy(addr_array[i++], ndev->broadcast);
+
+	mlx5e_for_each_hash_node(hn, tmp, addr_list, hi) {
+		if (ether_addr_equal(ndev->dev_addr, hn->ai.addr))
+			continue;
+		if (i >= size)
+			break;
+		ether_addr_copy(addr_array[i++], hn->ai.addr);
+	}
+}
+
+static void mlx5e_vport_context_update_addr_list(struct mlx5e_priv *priv,
+						 int list_type)
+{
+	bool is_uc = (list_type == MLX5_NVPRT_LIST_TYPE_UC);
+	struct mlx5e_eth_addr_hash_node *hn;
+	u8 (*addr_array)[ETH_ALEN] = NULL;
+	struct hlist_head *addr_list;
+	struct hlist_node *tmp;
+	int max_size;
+	int size;
+	int err;
+	int hi;
+
+	size = is_uc ? 0 : (priv->eth_addr.broadcast_enabled ? 1 : 0);
+	max_size = is_uc ?
+		1 << MLX5_CAP_GEN(priv->mdev, log_max_current_uc_list) :
+		1 << MLX5_CAP_GEN(priv->mdev, log_max_current_mc_list);
+
+	addr_list = is_uc ? priv->eth_addr.netdev_uc : priv->eth_addr.netdev_mc;
+	mlx5e_for_each_hash_node(hn, tmp, addr_list, hi)
+		size++;
+
+	if (size > max_size) {
+		netdev_warn(priv->netdev,
+			    "netdev %s list size (%d) > (%d) max vport list size, some addresses will be dropped\n",
+			    is_uc ? "UC" : "MC", size, max_size);
+		size = max_size;
+	}
+
+	if (size) {
+		addr_array = kcalloc(size, ETH_ALEN, GFP_KERNEL);
+		if (!addr_array) {
+			err = -ENOMEM;
+			goto out;
+		}
+		mlx5e_fill_addr_array(priv, list_type, addr_array, size);
+	}
+
+	err = mlx5_modify_nic_vport_mac_list(priv->mdev, list_type, addr_array, size);
+out:
+	if (err)
+		netdev_err(priv->netdev,
+			   "Failed to modify vport %s list err(%d)\n",
+			   is_uc ? "UC" : "MC", err);
+	kfree(addr_array);
+}
+
+static void mlx5e_vport_context_update(struct mlx5e_priv *priv)
+{
+	struct mlx5e_eth_addr_db *ea = &priv->eth_addr;
+
+	mlx5e_vport_context_update_addr_list(priv, MLX5_NVPRT_LIST_TYPE_UC);
+	mlx5e_vport_context_update_addr_list(priv, MLX5_NVPRT_LIST_TYPE_MC);
+	mlx5_modify_nic_vport_promisc(priv->mdev, 0,
+				      ea->allmulti_enabled,
+				      ea->promisc_enabled);
+}
+
+static void mlx5e_apply_netdev_addr(struct mlx5e_priv *priv)
+{
+	struct mlx5e_eth_addr_hash_node *hn;
+	struct hlist_node *tmp;
+	int i;
+
+	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_uc, i)
+		mlx5e_execute_action(priv, hn);
+
+	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_mc, i)
+		mlx5e_execute_action(priv, hn);
+}
+
+static void mlx5e_handle_netdev_addr(struct mlx5e_priv *priv)
+{
+	struct mlx5e_eth_addr_hash_node *hn;
+	struct hlist_node *tmp;
+	int i;
+
+	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_uc, i)
+		hn->action = MLX5E_ACTION_DEL;
+	mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_mc, i)
+		hn->action = MLX5E_ACTION_DEL;
+
+	if (!test_bit(MLX5E_STATE_DESTROYING, &priv->state))
+		mlx5e_sync_netdev_addr(priv);
+
+	mlx5e_apply_netdev_addr(priv);
+}
+
+void mlx5e_set_rx_mode_work(struct work_struct *work)
+{
+	struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv,
+					       set_rx_mode_work);
+
+	struct mlx5e_eth_addr_db *ea = &priv->eth_addr;
+	struct net_device *ndev = priv->netdev;
+
+	bool rx_mode_enable   = !test_bit(MLX5E_STATE_DESTROYING, &priv->state);
+	bool promisc_enabled   = rx_mode_enable && (ndev->flags & IFF_PROMISC);
+	bool allmulti_enabled  = rx_mode_enable && (ndev->flags & IFF_ALLMULTI);
+	bool broadcast_enabled = rx_mode_enable;
+
+	bool enable_promisc    = !ea->promisc_enabled   &&  promisc_enabled;
+	bool disable_promisc   =  ea->promisc_enabled   && !promisc_enabled;
+	bool enable_allmulti   = !ea->allmulti_enabled  &&  allmulti_enabled;
+	bool disable_allmulti  =  ea->allmulti_enabled  && !allmulti_enabled;
+	bool enable_broadcast  = !ea->broadcast_enabled &&  broadcast_enabled;
+	bool disable_broadcast =  ea->broadcast_enabled && !broadcast_enabled;
+
+	if (enable_promisc) {
+		mlx5e_add_eth_addr_rule(priv, &ea->promisc, MLX5E_PROMISC);
+		if (!priv->vlan.filter_disabled)
+			mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID,
+					    0);
+	}
+	if (enable_allmulti)
+		mlx5e_add_eth_addr_rule(priv, &ea->allmulti, MLX5E_ALLMULTI);
+	if (enable_broadcast)
+		mlx5e_add_eth_addr_rule(priv, &ea->broadcast, MLX5E_FULLMATCH);
+
+	mlx5e_handle_netdev_addr(priv);
+
+	if (disable_broadcast)
+		mlx5e_del_eth_addr_from_flow_table(priv, &ea->broadcast);
+	if (disable_allmulti)
+		mlx5e_del_eth_addr_from_flow_table(priv, &ea->allmulti);
+	if (disable_promisc) {
+		if (!priv->vlan.filter_disabled)
+			mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID,
+					    0);
+		mlx5e_del_eth_addr_from_flow_table(priv, &ea->promisc);
+	}
+
+	ea->promisc_enabled   = promisc_enabled;
+	ea->allmulti_enabled  = allmulti_enabled;
+	ea->broadcast_enabled = broadcast_enabled;
+
+	mlx5e_vport_context_update(priv);
+}
+
+static void mlx5e_destroy_groups(struct mlx5e_flow_table *ft)
+{
+	int i;
+
+	for (i = ft->num_groups - 1; i >= 0; i--) {
+		if (!IS_ERR_OR_NULL(ft->g[i]))
+			mlx5_destroy_flow_group(ft->g[i]);
+		ft->g[i] = NULL;
+	}
+	ft->num_groups = 0;
+}
+
+void mlx5e_init_eth_addr(struct mlx5e_priv *priv)
+{
+	ether_addr_copy(priv->eth_addr.broadcast.addr, priv->netdev->broadcast);
+}
+
+#define MLX5E_MAIN_GROUP0_SIZE	BIT(3)
+#define MLX5E_MAIN_GROUP1_SIZE	BIT(1)
+#define MLX5E_MAIN_GROUP2_SIZE	BIT(0)
+#define MLX5E_MAIN_GROUP3_SIZE	BIT(14)
+#define MLX5E_MAIN_GROUP4_SIZE	BIT(13)
+#define MLX5E_MAIN_GROUP5_SIZE	BIT(11)
+#define MLX5E_MAIN_GROUP6_SIZE	BIT(2)
+#define MLX5E_MAIN_GROUP7_SIZE	BIT(1)
+#define MLX5E_MAIN_GROUP8_SIZE	BIT(0)
+#define MLX5E_MAIN_TABLE_SIZE	(MLX5E_MAIN_GROUP0_SIZE +\
+				 MLX5E_MAIN_GROUP1_SIZE +\
+				 MLX5E_MAIN_GROUP2_SIZE +\
+				 MLX5E_MAIN_GROUP3_SIZE +\
+				 MLX5E_MAIN_GROUP4_SIZE +\
+				 MLX5E_MAIN_GROUP5_SIZE +\
+				 MLX5E_MAIN_GROUP6_SIZE +\
+				 MLX5E_MAIN_GROUP7_SIZE +\
+				 MLX5E_MAIN_GROUP8_SIZE)
+
+static int __mlx5e_create_main_groups(struct mlx5e_flow_table *ft, u32 *in,
+				      int inlen)
+{
+	u8 *mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+	u8 *dmac = MLX5_ADDR_OF(create_flow_group_in, in,
+				match_criteria.outer_headers.dmac_47_16);
+	int err;
+	int ix = 0;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP0_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP1_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP2_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	eth_broadcast_addr(dmac);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP3_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	eth_broadcast_addr(dmac);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP4_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	eth_broadcast_addr(dmac);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP5_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	dmac[0] = 0x01;
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP6_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	dmac[0] = 0x01;
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP7_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	dmac[0] = 0x01;
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP8_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	return 0;
+
+err_destroy_groups:
+	err = PTR_ERR(ft->g[ft->num_groups]);
+	ft->g[ft->num_groups] = NULL;
+	mlx5e_destroy_groups(ft);
+
+	return err;
+}
+
+static int mlx5e_create_main_groups(struct mlx5e_flow_table *ft)
+{
+	u32 *in;
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	int err;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in)
+		return -ENOMEM;
+
+	err = __mlx5e_create_main_groups(ft, in, inlen);
+
+	kvfree(in);
+	return err;
+}
+
+static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv)
+{
+	struct mlx5e_flow_table *ft = &priv->fts.main;
+	int err;
+
+	ft->num_groups = 0;
+	ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_MAIN_TABLE_SIZE);
+
+	if (IS_ERR(ft->t)) {
+		err = PTR_ERR(ft->t);
+		ft->t = NULL;
+		return err;
+	}
+	ft->g = kcalloc(MLX5E_NUM_MAIN_GROUPS, sizeof(*ft->g), GFP_KERNEL);
+	if (!ft->g) {
+		err = -ENOMEM;
+		goto err_destroy_main_flow_table;
+	}
+
+	err = mlx5e_create_main_groups(ft);
+	if (err)
+		goto err_free_g;
+	return 0;
+
+err_free_g:
+	kfree(ft->g);
+
+err_destroy_main_flow_table:
+	mlx5_destroy_flow_table(ft->t);
+	ft->t = NULL;
+
+	return err;
+}
+
+static void mlx5e_destroy_flow_table(struct mlx5e_flow_table *ft)
+{
+	mlx5e_destroy_groups(ft);
+	kfree(ft->g);
+	mlx5_destroy_flow_table(ft->t);
+	ft->t = NULL;
+}
+
+static void mlx5e_destroy_main_flow_table(struct mlx5e_priv *priv)
+{
+	mlx5e_destroy_flow_table(&priv->fts.main);
+}
+
+#define MLX5E_NUM_VLAN_GROUPS	2
+#define MLX5E_VLAN_GROUP0_SIZE	BIT(12)
+#define MLX5E_VLAN_GROUP1_SIZE	BIT(1)
+#define MLX5E_VLAN_TABLE_SIZE	(MLX5E_VLAN_GROUP0_SIZE +\
+				 MLX5E_VLAN_GROUP1_SIZE)
+
+static int __mlx5e_create_vlan_groups(struct mlx5e_flow_table *ft, u32 *in,
+				      int inlen)
+{
+	int err;
+	int ix = 0;
+	u8 *mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_VLAN_GROUP0_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_VLAN_GROUP1_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	return 0;
+
+err_destroy_groups:
+	err = PTR_ERR(ft->g[ft->num_groups]);
+	ft->g[ft->num_groups] = NULL;
+	mlx5e_destroy_groups(ft);
+
+	return err;
+}
+
+static int mlx5e_create_vlan_groups(struct mlx5e_flow_table *ft)
+{
+	u32 *in;
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	int err;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in)
+		return -ENOMEM;
+
+	err = __mlx5e_create_vlan_groups(ft, in, inlen);
+
+	kvfree(in);
+	return err;
+}
+
+static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv)
+{
+	struct mlx5e_flow_table *ft = &priv->fts.vlan;
+	int err;
+
+	ft->num_groups = 0;
+	ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_VLAN_TABLE_SIZE);
+
+	if (IS_ERR(ft->t)) {
+		err = PTR_ERR(ft->t);
+		ft->t = NULL;
+		return err;
+	}
+	ft->g = kcalloc(MLX5E_NUM_VLAN_GROUPS, sizeof(*ft->g), GFP_KERNEL);
+	if (!ft->g) {
+		err = -ENOMEM;
+		goto err_destroy_vlan_flow_table;
+	}
+
+	err = mlx5e_create_vlan_groups(ft);
+	if (err)
+		goto err_free_g;
+
+	return 0;
+
+err_free_g:
+	kfree(ft->g);
+
+err_destroy_vlan_flow_table:
+	mlx5_destroy_flow_table(ft->t);
+	ft->t = NULL;
+
+	return err;
+}
+
+static void mlx5e_destroy_vlan_flow_table(struct mlx5e_priv *priv)
+{
+	mlx5e_destroy_flow_table(&priv->fts.vlan);
+}
+
+int mlx5e_create_flow_tables(struct mlx5e_priv *priv)
+{
+	int err;
+
+	priv->fts.ns = mlx5_get_flow_namespace(priv->mdev,
+					       MLX5_FLOW_NAMESPACE_KERNEL);
+
+	if (!priv->fts.ns)
+		return -EINVAL;
+
+	err = mlx5e_create_vlan_flow_table(priv);
+	if (err)
+		return err;
+
+	err = mlx5e_create_main_flow_table(priv);
+	if (err)
+		goto err_destroy_vlan_flow_table;
+
+	err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0);
+	if (err)
+		goto err_destroy_main_flow_table;
+
+	return 0;
+
+err_destroy_main_flow_table:
+	mlx5e_destroy_main_flow_table(priv);
+err_destroy_vlan_flow_table:
+	mlx5e_destroy_vlan_flow_table(priv);
+
+	return err;
+}
+
+void mlx5e_destroy_flow_tables(struct mlx5e_priv *priv)
+{
+	mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0);
+	mlx5e_destroy_main_flow_table(priv);
+	mlx5e_destroy_vlan_flow_table(priv);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 1203d892e..402994bf7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -30,8 +30,9 @@
  * SOFTWARE.
  */
 
-#include <linux/mlx5/flow_table.h>
+#include <linux/mlx5/fs.h>
 #include "en.h"
+#include "eswitch.h"
 
 struct mlx5e_rq_param {
 	u32                        rqc[MLX5_ST_SZ_DW(rqc)];
@@ -63,7 +64,7 @@ static void mlx5e_update_carrier(struct mlx5e_priv *priv)
 	u8 port_state;
 
 	port_state = mlx5_query_vport_state(mdev,
-		MLX5_QUERY_VPORT_STATE_IN_OP_MOD_VNIC_VPORT);
+		MLX5_QUERY_VPORT_STATE_IN_OP_MOD_VNIC_VPORT, 0);
 
 	if (port_state == VPORT_STATE_UP)
 		netif_carrier_on(priv->netdev);
@@ -140,6 +141,10 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 		return;
 
 	/* Collect firts the SW counters and then HW for consistency */
+	s->rx_packets		= 0;
+	s->rx_bytes		= 0;
+	s->tx_packets		= 0;
+	s->tx_bytes		= 0;
 	s->tso_packets		= 0;
 	s->tso_bytes		= 0;
 	s->tx_queue_stopped	= 0;
@@ -154,6 +159,8 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 	for (i = 0; i < priv->params.num_channels; i++) {
 		rq_stats = &priv->channel[i]->rq.stats;
 
+		s->rx_packets	+= rq_stats->packets;
+		s->rx_bytes	+= rq_stats->bytes;
 		s->lro_packets	+= rq_stats->lro_packets;
 		s->lro_bytes	+= rq_stats->lro_bytes;
 		s->rx_csum_none	+= rq_stats->csum_none;
@@ -163,6 +170,8 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 		for (j = 0; j < priv->params.num_tc; j++) {
 			sq_stats = &priv->channel[i]->sq[j].stats;
 
+			s->tx_packets		+= sq_stats->packets;
+			s->tx_bytes		+= sq_stats->bytes;
 			s->tso_packets		+= sq_stats->tso_packets;
 			s->tso_bytes		+= sq_stats->tso_bytes;
 			s->tx_queue_stopped	+= sq_stats->stopped;
@@ -224,23 +233,6 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 	s->tx_broadcast_bytes   =
 		MLX5_GET_CTR(out, transmitted_eth_broadcast.octets);
 
-	s->rx_packets =
-		s->rx_unicast_packets +
-		s->rx_multicast_packets +
-		s->rx_broadcast_packets;
-	s->rx_bytes =
-		s->rx_unicast_bytes +
-		s->rx_multicast_bytes +
-		s->rx_broadcast_bytes;
-	s->tx_packets =
-		s->tx_unicast_packets +
-		s->tx_multicast_packets +
-		s->tx_broadcast_packets;
-	s->tx_bytes =
-		s->tx_unicast_bytes +
-		s->tx_multicast_bytes +
-		s->tx_broadcast_bytes;
-
 	/* Update calculated offload counters */
 	s->tx_csum_offload = s->tx_packets - tx_offload_none;
 	s->rx_csum_good    = s->rx_packets - s->rx_csum_none -
@@ -350,6 +342,7 @@ static int mlx5e_create_rq(struct mlx5e_channel *c,
 
 	rq->pdev    = c->pdev;
 	rq->netdev  = c->netdev;
+	rq->tstamp  = &priv->tstamp;
 	rq->channel = c;
 	rq->ix      = c->ix;
 	rq->priv    = c->priv;
@@ -506,6 +499,7 @@ static void mlx5e_close_rq(struct mlx5e_rq *rq)
 
 static void mlx5e_free_sq_db(struct mlx5e_sq *sq)
 {
+	kfree(sq->wqe_info);
 	kfree(sq->dma_fifo);
 	kfree(sq->skb);
 }
@@ -518,8 +512,10 @@ static int mlx5e_alloc_sq_db(struct mlx5e_sq *sq, int numa)
 	sq->skb = kzalloc_node(wq_sz * sizeof(*sq->skb), GFP_KERNEL, numa);
 	sq->dma_fifo = kzalloc_node(df_sz * sizeof(*sq->dma_fifo), GFP_KERNEL,
 				    numa);
+	sq->wqe_info = kzalloc_node(wq_sz * sizeof(*sq->wqe_info), GFP_KERNEL,
+				    numa);
 
-	if (!sq->skb || !sq->dma_fifo) {
+	if (!sq->skb || !sq->dma_fifo || !sq->wqe_info) {
 		mlx5e_free_sq_db(sq);
 		return -ENOMEM;
 	}
@@ -567,6 +563,7 @@ static int mlx5e_create_sq(struct mlx5e_channel *c,
 	sq->txq = netdev_get_tx_queue(priv->netdev, txq_ix);
 
 	sq->pdev      = c->pdev;
+	sq->tstamp    = &priv->tstamp;
 	sq->mkey_be   = c->mkey_be;
 	sq->channel   = c;
 	sq->tc        = tc;
@@ -1020,6 +1017,7 @@ err_close_tx_cqs:
 
 err_napi_del:
 	netif_napi_del(&c->napi);
+	napi_hash_del(&c->napi);
 	kfree(c);
 
 	return err;
@@ -1033,6 +1031,10 @@ static void mlx5e_close_channel(struct mlx5e_channel *c)
 	mlx5e_close_cq(&c->rq.cq);
 	mlx5e_close_tx_cqs(c);
 	netif_napi_del(&c->napi);
+
+	napi_hash_del(&c->napi);
+	synchronize_rcu();
+
 	kfree(c);
 }
 
@@ -1188,7 +1190,6 @@ static void mlx5e_fill_indir_rqt_rqns(struct mlx5e_priv *priv, void *rqtc)
 			ix = mlx5e_bits_invert(i, MLX5E_LOG_INDIR_RQT_SIZE);
 
 		ix = priv->params.indirection_rqt[ix];
-		ix = ix % priv->params.num_channels;
 		MLX5_SET(rqtc, rqtc, rq_num[i],
 			 test_bit(MLX5E_STATE_OPENED, &priv->state) ?
 			 priv->channel[ix]->rq.rqn :
@@ -1306,7 +1307,22 @@ static void mlx5e_build_tir_ctx_lro(void *tirc, struct mlx5e_priv *priv)
 			      lro_timer_supported_periods[2]));
 }
 
-static int mlx5e_modify_tir_lro(struct mlx5e_priv *priv, int tt)
+void mlx5e_build_tir_ctx_hash(void *tirc, struct mlx5e_priv *priv)
+{
+	MLX5_SET(tirc, tirc, rx_hash_fn,
+		 mlx5e_rx_hash_fn(priv->params.rss_hfunc));
+	if (priv->params.rss_hfunc == ETH_RSS_HASH_TOP) {
+		void *rss_key = MLX5_ADDR_OF(tirc, tirc,
+					     rx_hash_toeplitz_key);
+		size_t len = MLX5_FLD_SZ_BYTES(tirc,
+					       rx_hash_toeplitz_key);
+
+		MLX5_SET(tirc, tirc, rx_hash_symmetric, 1);
+		memcpy(rss_key, priv->params.toeplitz_hash_key, len);
+	}
+}
+
+static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 
@@ -1314,6 +1330,7 @@ static int mlx5e_modify_tir_lro(struct mlx5e_priv *priv, int tt)
 	void *tirc;
 	int inlen;
 	int err;
+	int tt;
 
 	inlen = MLX5_ST_SZ_BYTES(modify_tir_in);
 	in = mlx5_vzalloc(inlen);
@@ -1325,7 +1342,11 @@ static int mlx5e_modify_tir_lro(struct mlx5e_priv *priv, int tt)
 
 	mlx5e_build_tir_ctx_lro(tirc, priv);
 
-	err = mlx5_core_modify_tir(mdev, priv->tirn[tt], in, inlen);
+	for (tt = 0; tt < MLX5E_NUM_TT; tt++) {
+		err = mlx5_core_modify_tir(mdev, priv->tirn[tt], in, inlen);
+		if (err)
+			break;
+	}
 
 	kvfree(in);
 
@@ -1421,6 +1442,7 @@ int mlx5e_open_locked(struct net_device *netdev)
 
 	mlx5e_update_carrier(priv);
 	mlx5e_redirect_rqts(priv);
+	mlx5e_timestamp_init(priv);
 
 	schedule_delayed_work(&priv->update_stats_work, 0);
 
@@ -1457,6 +1479,7 @@ int mlx5e_close_locked(struct net_device *netdev)
 
 	clear_bit(MLX5E_STATE_OPENED, &priv->state);
 
+	mlx5e_timestamp_cleanup(priv);
 	mlx5e_redirect_rqts(priv);
 	netif_carrier_off(priv->netdev);
 	mlx5e_close_channels(priv);
@@ -1659,17 +1682,7 @@ static void mlx5e_build_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, int tt)
 	default:
 		MLX5_SET(tirc, tirc, indirect_table,
 			 priv->rqtn[MLX5E_INDIRECTION_RQT]);
-		MLX5_SET(tirc, tirc, rx_hash_fn,
-			 mlx5e_rx_hash_fn(priv->params.rss_hfunc));
-		if (priv->params.rss_hfunc == ETH_RSS_HASH_TOP) {
-			void *rss_key = MLX5_ADDR_OF(tirc, tirc,
-						     rx_hash_toeplitz_key);
-			size_t len = MLX5_FLD_SZ_BYTES(tirc,
-						       rx_hash_toeplitz_key);
-
-			MLX5_SET(tirc, tirc, rx_hash_symmetric, 1);
-			memcpy(rss_key, priv->params.toeplitz_hash_key, len);
-		}
+		mlx5e_build_tir_ctx_hash(tirc, priv);
 		break;
 	}
 
@@ -1872,8 +1885,10 @@ static int mlx5e_set_features(struct net_device *netdev,
 			mlx5e_close_locked(priv->netdev);
 
 		priv->params.lro_en = !!(features & NETIF_F_LRO);
-		mlx5e_modify_tir_lro(priv, MLX5E_TT_IPV4_TCP);
-		mlx5e_modify_tir_lro(priv, MLX5E_TT_IPV6_TCP);
+		err = mlx5e_modify_tirs_lro(priv);
+		if (err)
+			mlx5_core_warn(priv->mdev, "lro modify failed, %d\n",
+				       err);
 
 		if (was_opened)
 			err = mlx5e_open_locked(priv->netdev);
@@ -1926,17 +1941,122 @@ static int mlx5e_change_mtu(struct net_device *netdev, int new_mtu)
 	return err;
 }
 
-static struct net_device_ops mlx5e_netdev_ops = {
+static int mlx5e_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	switch (cmd) {
+	case SIOCSHWTSTAMP:
+		return mlx5e_hwstamp_set(dev, ifr);
+	case SIOCGHWTSTAMP:
+		return mlx5e_hwstamp_get(dev, ifr);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int mlx5e_set_vf_mac(struct net_device *dev, int vf, u8 *mac)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+
+	return mlx5_eswitch_set_vport_mac(mdev->priv.eswitch, vf + 1, mac);
+}
+
+static int mlx5e_set_vf_vlan(struct net_device *dev, int vf, u16 vlan, u8 qos)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+
+	return mlx5_eswitch_set_vport_vlan(mdev->priv.eswitch, vf + 1,
+					   vlan, qos);
+}
+
+static int mlx5_vport_link2ifla(u8 esw_link)
+{
+	switch (esw_link) {
+	case MLX5_ESW_VPORT_ADMIN_STATE_DOWN:
+		return IFLA_VF_LINK_STATE_DISABLE;
+	case MLX5_ESW_VPORT_ADMIN_STATE_UP:
+		return IFLA_VF_LINK_STATE_ENABLE;
+	}
+	return IFLA_VF_LINK_STATE_AUTO;
+}
+
+static int mlx5_ifla_link2vport(u8 ifla_link)
+{
+	switch (ifla_link) {
+	case IFLA_VF_LINK_STATE_DISABLE:
+		return MLX5_ESW_VPORT_ADMIN_STATE_DOWN;
+	case IFLA_VF_LINK_STATE_ENABLE:
+		return MLX5_ESW_VPORT_ADMIN_STATE_UP;
+	}
+	return MLX5_ESW_VPORT_ADMIN_STATE_AUTO;
+}
+
+static int mlx5e_set_vf_link_state(struct net_device *dev, int vf,
+				   int link_state)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+
+	return mlx5_eswitch_set_vport_state(mdev->priv.eswitch, vf + 1,
+					    mlx5_ifla_link2vport(link_state));
+}
+
+static int mlx5e_get_vf_config(struct net_device *dev,
+			       int vf, struct ifla_vf_info *ivi)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+	int err;
+
+	err = mlx5_eswitch_get_vport_config(mdev->priv.eswitch, vf + 1, ivi);
+	if (err)
+		return err;
+	ivi->linkstate = mlx5_vport_link2ifla(ivi->linkstate);
+	return 0;
+}
+
+static int mlx5e_get_vf_stats(struct net_device *dev,
+			      int vf, struct ifla_vf_stats *vf_stats)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+
+	return mlx5_eswitch_get_vport_stats(mdev->priv.eswitch, vf + 1,
+					    vf_stats);
+}
+
+static const struct net_device_ops mlx5e_netdev_ops_basic = {
 	.ndo_open                = mlx5e_open,
 	.ndo_stop                = mlx5e_close,
 	.ndo_start_xmit          = mlx5e_xmit,
 	.ndo_get_stats64         = mlx5e_get_stats,
 	.ndo_set_rx_mode         = mlx5e_set_rx_mode,
 	.ndo_set_mac_address     = mlx5e_set_mac,
-	.ndo_vlan_rx_add_vid	 = mlx5e_vlan_rx_add_vid,
-	.ndo_vlan_rx_kill_vid	 = mlx5e_vlan_rx_kill_vid,
+	.ndo_vlan_rx_add_vid     = mlx5e_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid    = mlx5e_vlan_rx_kill_vid,
 	.ndo_set_features        = mlx5e_set_features,
-	.ndo_change_mtu		 = mlx5e_change_mtu,
+	.ndo_change_mtu          = mlx5e_change_mtu,
+	.ndo_do_ioctl            = mlx5e_ioctl,
+};
+
+static const struct net_device_ops mlx5e_netdev_ops_sriov = {
+	.ndo_open                = mlx5e_open,
+	.ndo_stop                = mlx5e_close,
+	.ndo_start_xmit          = mlx5e_xmit,
+	.ndo_get_stats64         = mlx5e_get_stats,
+	.ndo_set_rx_mode         = mlx5e_set_rx_mode,
+	.ndo_set_mac_address     = mlx5e_set_mac,
+	.ndo_vlan_rx_add_vid     = mlx5e_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid    = mlx5e_vlan_rx_kill_vid,
+	.ndo_set_features        = mlx5e_set_features,
+	.ndo_change_mtu          = mlx5e_change_mtu,
+	.ndo_do_ioctl            = mlx5e_ioctl,
+	.ndo_set_vf_mac          = mlx5e_set_vf_mac,
+	.ndo_set_vf_vlan         = mlx5e_set_vf_vlan,
+	.ndo_get_vf_config       = mlx5e_get_vf_config,
+	.ndo_set_vf_link_state   = mlx5e_set_vf_link_state,
+	.ndo_get_vf_stats        = mlx5e_get_vf_stats,
 };
 
 static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev)
@@ -1971,12 +2091,20 @@ u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev)
 	       2 /*sizeof(mlx5e_tx_wqe.inline_hdr_start)*/;
 }
 
+void mlx5e_build_default_indir_rqt(u32 *indirection_rqt, int len,
+				   int num_channels)
+{
+	int i;
+
+	for (i = 0; i < len; i++)
+		indirection_rqt[i] = i % num_channels;
+}
+
 static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 				    struct net_device *netdev,
 				    int num_channels)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
-	int i;
 
 	priv->params.log_sq_size           =
 		MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE;
@@ -2000,8 +2128,8 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 	netdev_rss_key_fill(priv->params.toeplitz_hash_key,
 			    sizeof(priv->params.toeplitz_hash_key));
 
-	for (i = 0; i < MLX5E_INDIR_RQT_SIZE; i++)
-		priv->params.indirection_rqt[i] = i % num_channels;
+	mlx5e_build_default_indir_rqt(priv->params.indirection_rqt,
+				      MLX5E_INDIR_RQT_SIZE, num_channels);
 
 	priv->params.lro_wqe_sz            =
 		MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ;
@@ -2023,7 +2151,12 @@ static void mlx5e_set_netdev_dev_addr(struct net_device *netdev)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 
-	mlx5_query_nic_vport_mac_address(priv->mdev, netdev->dev_addr);
+	mlx5_query_nic_vport_mac_address(priv->mdev, 0, netdev->dev_addr);
+	if (is_zero_ether_addr(netdev->dev_addr) &&
+	    !MLX5_CAP_GEN(priv->mdev, vport_group_manager)) {
+		eth_hw_addr_random(netdev);
+		mlx5_core_info(priv->mdev, "Assigned random MAC address %pM\n", netdev->dev_addr);
+	}
 }
 
 static void mlx5e_build_netdev(struct net_device *netdev)
@@ -2033,10 +2166,11 @@ static void mlx5e_build_netdev(struct net_device *netdev)
 
 	SET_NETDEV_DEV(netdev, &mdev->pdev->dev);
 
-	if (priv->params.num_tc > 1)
-		mlx5e_netdev_ops.ndo_select_queue = mlx5e_select_queue;
+	if (MLX5_CAP_GEN(mdev, vport_group_manager))
+		netdev->netdev_ops = &mlx5e_netdev_ops_sriov;
+	else
+		netdev->netdev_ops = &mlx5e_netdev_ops_basic;
 
-	netdev->netdev_ops        = &mlx5e_netdev_ops;
 	netdev->watchdog_timeo    = 15 * HZ;
 
 	netdev->ethtool_ops	  = &mlx5e_ethtool_ops;
@@ -2129,7 +2263,7 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 		goto err_unmap_free_uar;
 	}
 
-	err = mlx5_alloc_transport_domain(mdev, &priv->tdn);
+	err = mlx5_core_alloc_transport_domain(mdev, &priv->tdn);
 	if (err) {
 		mlx5_core_err(mdev, "alloc td failed, %d\n", err);
 		goto err_dealloc_pd;
@@ -2212,7 +2346,7 @@ err_destroy_mkey:
 	mlx5_core_destroy_mkey(mdev, &priv->mr);
 
 err_dealloc_transport_domain:
-	mlx5_dealloc_transport_domain(mdev, priv->tdn);
+	mlx5_core_dealloc_transport_domain(mdev, priv->tdn);
 
 err_dealloc_pd:
 	mlx5_core_dealloc_pd(mdev, priv->pdn);
@@ -2244,7 +2378,7 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv)
 	mlx5e_close_drop_rq(priv);
 	mlx5e_destroy_tises(priv);
 	mlx5_core_destroy_mkey(priv->mdev, &priv->mr);
-	mlx5_dealloc_transport_domain(priv->mdev, priv->tdn);
+	mlx5_core_dealloc_transport_domain(priv->mdev, priv->tdn);
 	mlx5_core_dealloc_pd(priv->mdev, priv->pdn);
 	mlx5_unmap_free_uar(priv->mdev, &priv->cq_uar);
 	free_netdev(netdev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index cf0098596..59658b9d0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -33,8 +33,14 @@
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 #include <linux/tcp.h>
+#include <net/busy_poll.h>
 #include "en.h"
 
+static inline bool mlx5e_rx_hw_stamp(struct mlx5e_tstamp *tstamp)
+{
+	return tstamp->hwtstamp_config.rx_filter == HWTSTAMP_FILTER_ALL;
+}
+
 static inline int mlx5e_alloc_rx_wqe(struct mlx5e_rq *rq,
 				     struct mlx5e_rx_wqe *wqe, u16 ix)
 {
@@ -189,6 +195,7 @@ static inline void mlx5e_build_rx_skb(struct mlx5_cqe64 *cqe,
 {
 	struct net_device *netdev = rq->netdev;
 	u32 cqe_bcnt = be32_to_cpu(cqe->byte_cnt);
+	struct mlx5e_tstamp *tstamp = rq->tstamp;
 	int lro_num_seg;
 
 	skb_put(skb, cqe_bcnt);
@@ -201,6 +208,9 @@ static inline void mlx5e_build_rx_skb(struct mlx5_cqe64 *cqe,
 		rq->stats.lro_bytes += cqe_bcnt;
 	}
 
+	if (unlikely(mlx5e_rx_hw_stamp(tstamp)))
+		mlx5e_fill_hwstamp(tstamp, get_cqe_ts(cqe), skb_hwtstamps(skb));
+
 	mlx5e_handle_csum(netdev, cqe, rq, skb);
 
 	skb->protocol = eth_type_trans(skb, netdev);
@@ -215,16 +225,12 @@ static inline void mlx5e_build_rx_skb(struct mlx5_cqe64 *cqe,
 				       be16_to_cpu(cqe->vlan_info));
 }
 
-bool mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
+int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
 {
 	struct mlx5e_rq *rq = container_of(cq, struct mlx5e_rq, cq);
-	int i;
+	int work_done;
 
-	/* avoid accessing cq (dma coherent memory) if not needed */
-	if (!test_and_clear_bit(MLX5E_CQ_HAS_CQES, &cq->flags))
-		return false;
-
-	for (i = 0; i < budget; i++) {
+	for (work_done = 0; work_done < budget; work_done++) {
 		struct mlx5e_rx_wqe *wqe;
 		struct mlx5_cqe64 *cqe;
 		struct sk_buff *skb;
@@ -257,6 +263,7 @@ bool mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
 
 		mlx5e_build_rx_skb(cqe, rq, skb);
 		rq->stats.packets++;
+		rq->stats.bytes += be32_to_cpu(cqe->byte_cnt);
 		napi_gro_receive(cq->napi, skb);
 
 wq_ll_pop:
@@ -269,10 +276,5 @@ wq_ll_pop:
 	/* ensure cq space is freed before enabling more cqes */
 	wmb();
 
-	if (i == budget) {
-		set_bit(MLX5E_CQ_HAS_CQES, &cq->flags);
-		return true;
-	}
-
-	return false;
+	return work_done;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 1341b1d3c..bb4eeeb00 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -92,11 +92,11 @@ static inline struct mlx5e_sq_dma *mlx5e_dma_get(struct mlx5e_sq *sq, u32 i)
 	return &sq->dma_fifo[i & sq->dma_fifo_mask];
 }
 
-static void mlx5e_dma_unmap_wqe_err(struct mlx5e_sq *sq, struct sk_buff *skb)
+static void mlx5e_dma_unmap_wqe_err(struct mlx5e_sq *sq, u8 num_dma)
 {
 	int i;
 
-	for (i = 0; i < MLX5E_TX_SKB_CB(skb)->num_dma; i++) {
+	for (i = 0; i < num_dma; i++) {
 		struct mlx5e_sq_dma *last_pushed_dma =
 			mlx5e_dma_get(sq, --sq->dma_fifo_pc);
 
@@ -139,19 +139,28 @@ static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq,
 	return MLX5E_MIN_INLINE;
 }
 
-static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs)
+static inline void mlx5e_tx_skb_pull_inline(unsigned char **skb_data,
+					    unsigned int *skb_len,
+					    unsigned int len)
+{
+	*skb_len -= len;
+	*skb_data += len;
+}
+
+static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs,
+				     unsigned char **skb_data,
+				     unsigned int *skb_len)
 {
 	struct vlan_ethhdr *vhdr = (struct vlan_ethhdr *)start;
 	int cpy1_sz = 2 * ETH_ALEN;
 	int cpy2_sz = ihs - cpy1_sz;
 
-	skb_copy_from_linear_data(skb, vhdr, cpy1_sz);
-	skb_pull_inline(skb, cpy1_sz);
+	memcpy(vhdr, *skb_data, cpy1_sz);
+	mlx5e_tx_skb_pull_inline(skb_data, skb_len, cpy1_sz);
 	vhdr->h_vlan_proto = skb->vlan_proto;
 	vhdr->h_vlan_TCI = cpu_to_be16(skb_vlan_tag_get(skb));
-	skb_copy_from_linear_data(skb, &vhdr->h_vlan_encapsulated_proto,
-				  cpy2_sz);
-	skb_pull_inline(skb, cpy2_sz);
+	memcpy(&vhdr->h_vlan_encapsulated_proto, *skb_data, cpy2_sz);
+	mlx5e_tx_skb_pull_inline(skb_data, skb_len, cpy2_sz);
 }
 
 static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
@@ -160,13 +169,17 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 
 	u16 pi = sq->pc & wq->sz_m1;
 	struct mlx5e_tx_wqe      *wqe  = mlx5_wq_cyc_get_wqe(wq, pi);
+	struct mlx5e_tx_wqe_info *wi   = &sq->wqe_info[pi];
 
 	struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl;
 	struct mlx5_wqe_eth_seg  *eseg = &wqe->eth;
 	struct mlx5_wqe_data_seg *dseg;
 
+	unsigned char *skb_data = skb->data;
+	unsigned int skb_len = skb->len;
 	u8  opcode = MLX5_OPCODE_SEND;
 	dma_addr_t dma_addr = 0;
+	unsigned int num_bytes;
 	bool bf = false;
 	u16 headlen;
 	u16 ds_cnt;
@@ -192,8 +205,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		opcode       = MLX5_OPCODE_LSO;
 		ihs          = skb_transport_offset(skb) + tcp_hdrlen(skb);
 		payload_len  = skb->len - ihs;
-		MLX5E_TX_SKB_CB(skb)->num_bytes = skb->len +
-					(skb_shinfo(skb)->gso_segs - 1) * ihs;
+		num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs;
 		sq->stats.tso_packets++;
 		sq->stats.tso_bytes += payload_len;
 	} else {
@@ -201,16 +213,18 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		     !skb->xmit_more &&
 		     !skb_shinfo(skb)->nr_frags;
 		ihs = mlx5e_get_inline_hdr_size(sq, skb, bf);
-		MLX5E_TX_SKB_CB(skb)->num_bytes = max_t(unsigned int, skb->len,
-							ETH_ZLEN);
+		num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN);
 	}
 
+	wi->num_bytes = num_bytes;
+
 	if (skb_vlan_tag_present(skb)) {
-		mlx5e_insert_vlan(eseg->inline_hdr_start, skb, ihs);
+		mlx5e_insert_vlan(eseg->inline_hdr_start, skb, ihs, &skb_data,
+				  &skb_len);
 		ihs += VLAN_HLEN;
 	} else {
-		skb_copy_from_linear_data(skb, eseg->inline_hdr_start, ihs);
-		skb_pull_inline(skb, ihs);
+		memcpy(eseg->inline_hdr_start, skb_data, ihs);
+		mlx5e_tx_skb_pull_inline(&skb_data, &skb_len, ihs);
 	}
 
 	eseg->inline_hdr_sz = cpu_to_be16(ihs);
@@ -220,11 +234,11 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 			       MLX5_SEND_WQE_DS);
 	dseg    = (struct mlx5_wqe_data_seg *)cseg + ds_cnt;
 
-	MLX5E_TX_SKB_CB(skb)->num_dma = 0;
+	wi->num_dma = 0;
 
-	headlen = skb_headlen(skb);
+	headlen = skb_len - skb->data_len;
 	if (headlen) {
-		dma_addr = dma_map_single(sq->pdev, skb->data, headlen,
+		dma_addr = dma_map_single(sq->pdev, skb_data, headlen,
 					  DMA_TO_DEVICE);
 		if (unlikely(dma_mapping_error(sq->pdev, dma_addr)))
 			goto dma_unmap_wqe_err;
@@ -234,7 +248,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		dseg->byte_count = cpu_to_be32(headlen);
 
 		mlx5e_dma_push(sq, dma_addr, headlen, MLX5E_DMA_MAP_SINGLE);
-		MLX5E_TX_SKB_CB(skb)->num_dma++;
+		wi->num_dma++;
 
 		dseg++;
 	}
@@ -253,23 +267,25 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		dseg->byte_count = cpu_to_be32(fsz);
 
 		mlx5e_dma_push(sq, dma_addr, fsz, MLX5E_DMA_MAP_PAGE);
-		MLX5E_TX_SKB_CB(skb)->num_dma++;
+		wi->num_dma++;
 
 		dseg++;
 	}
 
-	ds_cnt += MLX5E_TX_SKB_CB(skb)->num_dma;
+	ds_cnt += wi->num_dma;
 
 	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | opcode);
 	cseg->qpn_ds           = cpu_to_be32((sq->sqn << 8) | ds_cnt);
 
 	sq->skb[pi] = skb;
 
-	MLX5E_TX_SKB_CB(skb)->num_wqebbs = DIV_ROUND_UP(ds_cnt,
-							MLX5_SEND_WQEBB_NUM_DS);
-	sq->pc += MLX5E_TX_SKB_CB(skb)->num_wqebbs;
+	wi->num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS);
+	sq->pc += wi->num_wqebbs;
+
+	netdev_tx_sent_queue(sq->txq, wi->num_bytes);
 
-	netdev_tx_sent_queue(sq->txq, MLX5E_TX_SKB_CB(skb)->num_bytes);
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP))
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 
 	if (unlikely(!mlx5e_sq_has_room_for(sq, MLX5E_SQ_STOP_ROOM))) {
 		netif_tx_stop_queue(sq->txq);
@@ -280,7 +296,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		int bf_sz = 0;
 
 		if (bf && sq->uar_bf_map)
-			bf_sz = MLX5E_TX_SKB_CB(skb)->num_wqebbs << 3;
+			bf_sz = wi->num_wqebbs << 3;
 
 		cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
 		mlx5e_tx_notify_hw(sq, wqe, bf_sz);
@@ -293,11 +309,12 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	sq->bf_budget = bf ? sq->bf_budget - 1 : 0;
 
 	sq->stats.packets++;
+	sq->stats.bytes += num_bytes;
 	return NETDEV_TX_OK;
 
 dma_unmap_wqe_err:
 	sq->stats.dropped++;
-	mlx5e_dma_unmap_wqe_err(sq, skb);
+	mlx5e_dma_unmap_wqe_err(sq, wi->num_dma);
 
 	dev_kfree_skb_any(skb);
 
@@ -321,10 +338,6 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 	u16 sqcc;
 	int i;
 
-	/* avoid accessing cq (dma coherent memory) if not needed */
-	if (!test_and_clear_bit(MLX5E_CQ_HAS_CQES, &cq->flags))
-		return false;
-
 	sq = container_of(cq, struct mlx5e_sq, cq);
 
 	npkts = 0;
@@ -352,6 +365,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 		wqe_counter = be16_to_cpu(cqe->wqe_counter);
 
 		do {
+			struct mlx5e_tx_wqe_info *wi;
 			struct sk_buff *skb;
 			u16 ci;
 			int j;
@@ -360,6 +374,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 
 			ci = sqcc & sq->wq.sz_m1;
 			skb = sq->skb[ci];
+			wi = &sq->wqe_info[ci];
 
 			if (unlikely(!skb)) { /* nop */
 				sq->stats.nop++;
@@ -367,7 +382,16 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 				continue;
 			}
 
-			for (j = 0; j < MLX5E_TX_SKB_CB(skb)->num_dma; j++) {
+			if (unlikely(skb_shinfo(skb)->tx_flags &
+				     SKBTX_HW_TSTAMP)) {
+				struct skb_shared_hwtstamps hwts = {};
+
+				mlx5e_fill_hwstamp(sq->tstamp,
+						   get_cqe_ts(cqe), &hwts);
+				skb_tstamp_tx(skb, &hwts);
+			}
+
+			for (j = 0; j < wi->num_dma; j++) {
 				struct mlx5e_sq_dma *dma =
 					mlx5e_dma_get(sq, dma_fifo_cc++);
 
@@ -375,8 +399,8 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 			}
 
 			npkts++;
-			nbytes += MLX5E_TX_SKB_CB(skb)->num_bytes;
-			sqcc += MLX5E_TX_SKB_CB(skb)->num_wqebbs;
+			nbytes += wi->num_bytes;
+			sqcc += wi->num_wqebbs;
 			dev_kfree_skb(skb);
 		} while (!last_wqe);
 	}
@@ -397,10 +421,6 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 				netif_tx_wake_queue(sq->txq);
 				sq->stats.wake++;
 	}
-	if (i == MLX5E_TX_CQ_POLL_BUDGET) {
-		set_bit(MLX5E_CQ_HAS_CQES, &cq->flags);
-		return true;
-	}
 
-	return false;
+	return (i == MLX5E_TX_CQ_POLL_BUDGET);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
index 2c7cb6755..66d51a776 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
@@ -54,6 +54,7 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 	struct mlx5e_channel *c = container_of(napi, struct mlx5e_channel,
 					       napi);
 	bool busy = false;
+	int work_done;
 	int i;
 
 	clear_bit(MLX5E_CHANNEL_NAPI_SCHED, &c->flags);
@@ -61,33 +62,32 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 	for (i = 0; i < c->num_tc; i++)
 		busy |= mlx5e_poll_tx_cq(&c->sq[i].cq);
 
-	busy |= mlx5e_poll_rx_cq(&c->rq.cq, budget);
-
+	work_done = mlx5e_poll_rx_cq(&c->rq.cq, budget);
+	busy |= work_done == budget;
 	busy |= mlx5e_post_rx_wqes(&c->rq);
 
 	if (busy)
 		return budget;
 
-	napi_complete(napi);
+	napi_complete_done(napi, work_done);
 
 	/* avoid losing completion event during/after polling cqs */
 	if (test_bit(MLX5E_CHANNEL_NAPI_SCHED, &c->flags)) {
 		napi_schedule(napi);
-		return 0;
+		return work_done;
 	}
 
 	for (i = 0; i < c->num_tc; i++)
 		mlx5e_cq_arm(&c->sq[i].cq);
 	mlx5e_cq_arm(&c->rq.cq);
 
-	return 0;
+	return work_done;
 }
 
 void mlx5e_completion_event(struct mlx5_core_cq *mcq)
 {
 	struct mlx5e_cq *cq = container_of(mcq, struct mlx5e_cq, mcq);
 
-	set_bit(MLX5E_CQ_HAS_CQES, &cq->flags);
 	set_bit(MLX5E_CHANNEL_NAPI_SCHED, &cq->channel->flags);
 	barrier();
 	napi_schedule(cq->napi);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 713ead583..647a3ca2c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -35,6 +35,9 @@
 #include <linux/mlx5/driver.h>
 #include <linux/mlx5/cmd.h>
 #include "mlx5_core.h"
+#ifdef CONFIG_MLX5_CORE_EN
+#include "eswitch.h"
+#endif
 
 enum {
 	MLX5_EQE_SIZE		= sizeof(struct mlx5_eqe),
@@ -227,6 +230,7 @@ static int mlx5_eq_int(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 		case MLX5_EVENT_TYPE_WQ_INVAL_REQ_ERROR:
 		case MLX5_EVENT_TYPE_WQ_ACCESS_ERROR:
 			rsn = be32_to_cpu(eqe->data.qp_srq.qp_srq_n) & 0xffffff;
+			rsn |= (eqe->data.qp_srq.type << MLX5_USER_INDEX_LEN);
 			mlx5_core_dbg(dev, "event %s(%d) arrived on resource 0x%x\n",
 				      eqe_type_str(eqe->type), eqe->type, rsn);
 			mlx5_rsc_event(dev, rsn, eqe->type);
@@ -287,6 +291,11 @@ static int mlx5_eq_int(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 			break;
 #endif
 
+#ifdef CONFIG_MLX5_CORE_EN
+		case MLX5_EVENT_TYPE_NIC_VPORT_CHANGE:
+			mlx5_eswitch_vport_event(dev->priv.eswitch, eqe);
+			break;
+#endif
 		default:
 			mlx5_core_warn(dev, "Unhandled event 0x%x on EQ 0x%x\n",
 				       eqe->type, eq->eqn);
@@ -459,6 +468,11 @@ int mlx5_start_eqs(struct mlx5_core_dev *dev)
 	if (MLX5_CAP_GEN(dev, pg))
 		async_event_mask |= (1ull << MLX5_EVENT_TYPE_PAGE_FAULT);
 
+	if (MLX5_CAP_GEN(dev, port_type) == MLX5_CAP_PORT_TYPE_ETH &&
+	    MLX5_CAP_GEN(dev, vport_group_manager) &&
+	    mlx5_core_is_pf(dev))
+		async_event_mask |= (1ull << MLX5_EVENT_TYPE_NIC_VPORT_CHANGE);
+
 	err = mlx5_create_map_eq(dev, &table->cmd_eq, MLX5_EQ_VEC_CMD,
 				 MLX5_NUM_CMD_EQE, 1ull << MLX5_EVENT_TYPE_CMD,
 				 "mlx5_cmd_eq", &dev->priv.uuari.uars[0]);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
new file mode 100644
index 000000000..bc3d9f8a7
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -0,0 +1,1097 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/etherdevice.h>
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/mlx5_ifc.h>
+#include <linux/mlx5/vport.h>
+#include <linux/mlx5/fs.h>
+#include "mlx5_core.h"
+#include "eswitch.h"
+
+#define UPLINK_VPORT 0xFFFF
+
+#define MLX5_DEBUG_ESWITCH_MASK BIT(3)
+
+#define esw_info(dev, format, ...)				\
+	pr_info("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__)
+
+#define esw_warn(dev, format, ...)				\
+	pr_warn("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__)
+
+#define esw_debug(dev, format, ...)				\
+	mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__)
+
+enum {
+	MLX5_ACTION_NONE = 0,
+	MLX5_ACTION_ADD  = 1,
+	MLX5_ACTION_DEL  = 2,
+};
+
+/* E-Switch UC L2 table hash node */
+struct esw_uc_addr {
+	struct l2addr_node node;
+	u32                table_index;
+	u32                vport;
+};
+
+/* E-Switch MC FDB table hash node */
+struct esw_mc_addr { /* SRIOV only */
+	struct l2addr_node     node;
+	struct mlx5_flow_rule *uplink_rule; /* Forward to uplink rule */
+	u32                    refcnt;
+};
+
+/* Vport UC/MC hash node */
+struct vport_addr {
+	struct l2addr_node     node;
+	u8                     action;
+	u32                    vport;
+	struct mlx5_flow_rule *flow_rule; /* SRIOV only */
+};
+
+enum {
+	UC_ADDR_CHANGE = BIT(0),
+	MC_ADDR_CHANGE = BIT(1),
+};
+
+/* Vport context events */
+#define SRIOV_VPORT_EVENTS (UC_ADDR_CHANGE | \
+			    MC_ADDR_CHANGE)
+
+static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport,
+					u32 events_mask)
+{
+	int in[MLX5_ST_SZ_DW(modify_nic_vport_context_in)];
+	int out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)];
+	void *nic_vport_ctx;
+	int err;
+
+	memset(out, 0, sizeof(out));
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 opcode, MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT);
+	MLX5_SET(modify_nic_vport_context_in, in, field_select.change_event, 1);
+	MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport);
+	if (vport)
+		MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1);
+	nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in,
+				     in, nic_vport_context);
+
+	MLX5_SET(nic_vport_context, nic_vport_ctx, arm_change_event, 1);
+
+	if (events_mask & UC_ADDR_CHANGE)
+		MLX5_SET(nic_vport_context, nic_vport_ctx,
+			 event_on_uc_address_change, 1);
+	if (events_mask & MC_ADDR_CHANGE)
+		MLX5_SET(nic_vport_context, nic_vport_ctx,
+			 event_on_mc_address_change, 1);
+
+	err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+	if (err)
+		goto ex;
+	err = mlx5_cmd_status_to_err_v2(out);
+	if (err)
+		goto ex;
+	return 0;
+ex:
+	return err;
+}
+
+/* E-Switch vport context HW commands */
+static int query_esw_vport_context_cmd(struct mlx5_core_dev *mdev, u32 vport,
+				       u32 *out, int outlen)
+{
+	u32 in[MLX5_ST_SZ_DW(query_esw_vport_context_in)];
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(query_nic_vport_context_in, in, opcode,
+		 MLX5_CMD_OP_QUERY_ESW_VPORT_CONTEXT);
+
+	MLX5_SET(query_esw_vport_context_in, in, vport_number, vport);
+	if (vport)
+		MLX5_SET(query_esw_vport_context_in, in, other_vport, 1);
+
+	return mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen);
+}
+
+static int query_esw_vport_cvlan(struct mlx5_core_dev *dev, u32 vport,
+				 u16 *vlan, u8 *qos)
+{
+	u32 out[MLX5_ST_SZ_DW(query_esw_vport_context_out)];
+	int err;
+	bool cvlan_strip;
+	bool cvlan_insert;
+
+	memset(out, 0, sizeof(out));
+
+	*vlan = 0;
+	*qos = 0;
+
+	if (!MLX5_CAP_ESW(dev, vport_cvlan_strip) ||
+	    !MLX5_CAP_ESW(dev, vport_cvlan_insert_if_not_exist))
+		return -ENOTSUPP;
+
+	err = query_esw_vport_context_cmd(dev, vport, out, sizeof(out));
+	if (err)
+		goto out;
+
+	cvlan_strip = MLX5_GET(query_esw_vport_context_out, out,
+			       esw_vport_context.vport_cvlan_strip);
+
+	cvlan_insert = MLX5_GET(query_esw_vport_context_out, out,
+				esw_vport_context.vport_cvlan_insert);
+
+	if (cvlan_strip || cvlan_insert) {
+		*vlan = MLX5_GET(query_esw_vport_context_out, out,
+				 esw_vport_context.cvlan_id);
+		*qos = MLX5_GET(query_esw_vport_context_out, out,
+				esw_vport_context.cvlan_pcp);
+	}
+
+	esw_debug(dev, "Query Vport[%d] cvlan: VLAN %d qos=%d\n",
+		  vport, *vlan, *qos);
+out:
+	return err;
+}
+
+static int modify_esw_vport_context_cmd(struct mlx5_core_dev *dev, u16 vport,
+					void *in, int inlen)
+{
+	u32 out[MLX5_ST_SZ_DW(modify_esw_vport_context_out)];
+
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(modify_esw_vport_context_in, in, vport_number, vport);
+	if (vport)
+		MLX5_SET(modify_esw_vport_context_in, in, other_vport, 1);
+
+	MLX5_SET(modify_esw_vport_context_in, in, opcode,
+		 MLX5_CMD_OP_MODIFY_ESW_VPORT_CONTEXT);
+
+	return mlx5_cmd_exec_check_status(dev, in, inlen,
+					  out, sizeof(out));
+}
+
+static int modify_esw_vport_cvlan(struct mlx5_core_dev *dev, u32 vport,
+				  u16 vlan, u8 qos, bool set)
+{
+	u32 in[MLX5_ST_SZ_DW(modify_esw_vport_context_in)];
+
+	memset(in, 0, sizeof(in));
+
+	if (!MLX5_CAP_ESW(dev, vport_cvlan_strip) ||
+	    !MLX5_CAP_ESW(dev, vport_cvlan_insert_if_not_exist))
+		return -ENOTSUPP;
+
+	esw_debug(dev, "Set Vport[%d] VLAN %d qos %d set=%d\n",
+		  vport, vlan, qos, set);
+
+	if (set) {
+		MLX5_SET(modify_esw_vport_context_in, in,
+			 esw_vport_context.vport_cvlan_strip, 1);
+		/* insert only if no vlan in packet */
+		MLX5_SET(modify_esw_vport_context_in, in,
+			 esw_vport_context.vport_cvlan_insert, 1);
+		MLX5_SET(modify_esw_vport_context_in, in,
+			 esw_vport_context.cvlan_pcp, qos);
+		MLX5_SET(modify_esw_vport_context_in, in,
+			 esw_vport_context.cvlan_id, vlan);
+	}
+
+	MLX5_SET(modify_esw_vport_context_in, in,
+		 field_select.vport_cvlan_strip, 1);
+	MLX5_SET(modify_esw_vport_context_in, in,
+		 field_select.vport_cvlan_insert, 1);
+
+	return modify_esw_vport_context_cmd(dev, vport, in, sizeof(in));
+}
+
+/* HW L2 Table (MPFS) management */
+static int set_l2_table_entry_cmd(struct mlx5_core_dev *dev, u32 index,
+				  u8 *mac, u8 vlan_valid, u16 vlan)
+{
+	u32 in[MLX5_ST_SZ_DW(set_l2_table_entry_in)];
+	u32 out[MLX5_ST_SZ_DW(set_l2_table_entry_out)];
+	u8 *in_mac_addr;
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(set_l2_table_entry_in, in, opcode,
+		 MLX5_CMD_OP_SET_L2_TABLE_ENTRY);
+	MLX5_SET(set_l2_table_entry_in, in, table_index, index);
+	MLX5_SET(set_l2_table_entry_in, in, vlan_valid, vlan_valid);
+	MLX5_SET(set_l2_table_entry_in, in, vlan, vlan);
+
+	in_mac_addr = MLX5_ADDR_OF(set_l2_table_entry_in, in, mac_address);
+	ether_addr_copy(&in_mac_addr[2], mac);
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in),
+					  out, sizeof(out));
+}
+
+static int del_l2_table_entry_cmd(struct mlx5_core_dev *dev, u32 index)
+{
+	u32 in[MLX5_ST_SZ_DW(delete_l2_table_entry_in)];
+	u32 out[MLX5_ST_SZ_DW(delete_l2_table_entry_out)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(delete_l2_table_entry_in, in, opcode,
+		 MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY);
+	MLX5_SET(delete_l2_table_entry_in, in, table_index, index);
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in),
+					  out, sizeof(out));
+}
+
+static int alloc_l2_table_index(struct mlx5_l2_table *l2_table, u32 *ix)
+{
+	int err = 0;
+
+	*ix = find_first_zero_bit(l2_table->bitmap, l2_table->size);
+	if (*ix >= l2_table->size)
+		err = -ENOSPC;
+	else
+		__set_bit(*ix, l2_table->bitmap);
+
+	return err;
+}
+
+static void free_l2_table_index(struct mlx5_l2_table *l2_table, u32 ix)
+{
+	__clear_bit(ix, l2_table->bitmap);
+}
+
+static int set_l2_table_entry(struct mlx5_core_dev *dev, u8 *mac,
+			      u8 vlan_valid, u16 vlan,
+			      u32 *index)
+{
+	struct mlx5_l2_table *l2_table = &dev->priv.eswitch->l2_table;
+	int err;
+
+	err = alloc_l2_table_index(l2_table, index);
+	if (err)
+		return err;
+
+	err = set_l2_table_entry_cmd(dev, *index, mac, vlan_valid, vlan);
+	if (err)
+		free_l2_table_index(l2_table, *index);
+
+	return err;
+}
+
+static void del_l2_table_entry(struct mlx5_core_dev *dev, u32 index)
+{
+	struct mlx5_l2_table *l2_table = &dev->priv.eswitch->l2_table;
+
+	del_l2_table_entry_cmd(dev, index);
+	free_l2_table_index(l2_table, index);
+}
+
+/* E-Switch FDB */
+static struct mlx5_flow_rule *
+esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport)
+{
+	int match_header = MLX5_MATCH_OUTER_HEADERS;
+	struct mlx5_flow_destination dest;
+	struct mlx5_flow_rule *flow_rule = NULL;
+	u32 *match_v;
+	u32 *match_c;
+	u8 *dmac_v;
+	u8 *dmac_c;
+
+	match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL);
+	match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL);
+	if (!match_v || !match_c) {
+		pr_warn("FDB: Failed to alloc match parameters\n");
+		goto out;
+	}
+	dmac_v = MLX5_ADDR_OF(fte_match_param, match_v,
+			      outer_headers.dmac_47_16);
+	dmac_c = MLX5_ADDR_OF(fte_match_param, match_c,
+			      outer_headers.dmac_47_16);
+
+	ether_addr_copy(dmac_v, mac);
+	/* Match criteria mask */
+	memset(dmac_c, 0xff, 6);
+
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
+	dest.vport_num = vport;
+
+	esw_debug(esw->dev,
+		  "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n",
+		  dmac_v, dmac_c, vport);
+	flow_rule =
+		mlx5_add_flow_rule(esw->fdb_table.fdb,
+				   match_header,
+				   match_c,
+				   match_v,
+				   MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+				   0, &dest);
+	if (IS_ERR_OR_NULL(flow_rule)) {
+		pr_warn(
+			"FDB: Failed to add flow rule: dmac_v(%pM) dmac_c(%pM) -> vport(%d), err(%ld)\n",
+			 dmac_v, dmac_c, vport, PTR_ERR(flow_rule));
+		flow_rule = NULL;
+	}
+out:
+	kfree(match_v);
+	kfree(match_c);
+	return flow_rule;
+}
+
+static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_core_dev *dev = esw->dev;
+	struct mlx5_flow_namespace *root_ns;
+	struct mlx5_flow_table *fdb;
+	struct mlx5_flow_group *g;
+	void *match_criteria;
+	int table_size;
+	u32 *flow_group_in;
+	u8 *dmac;
+	int err = 0;
+
+	esw_debug(dev, "Create FDB log_max_size(%d)\n",
+		  MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
+
+	root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB);
+	if (!root_ns) {
+		esw_warn(dev, "Failed to get FDB flow namespace\n");
+		return -ENOMEM;
+	}
+
+	flow_group_in = mlx5_vzalloc(inlen);
+	if (!flow_group_in)
+		return -ENOMEM;
+	memset(flow_group_in, 0, inlen);
+
+	table_size = BIT(MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
+	fdb = mlx5_create_flow_table(root_ns, 0, table_size);
+	if (IS_ERR_OR_NULL(fdb)) {
+		err = PTR_ERR(fdb);
+		esw_warn(dev, "Failed to create FDB Table err %d\n", err);
+		goto out;
+	}
+
+	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
+		 MLX5_MATCH_OUTER_HEADERS);
+	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
+	dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, outer_headers.dmac_47_16);
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 1);
+	eth_broadcast_addr(dmac);
+
+	g = mlx5_create_flow_group(fdb, flow_group_in);
+	if (IS_ERR_OR_NULL(g)) {
+		err = PTR_ERR(g);
+		esw_warn(dev, "Failed to create flow group err(%d)\n", err);
+		goto out;
+	}
+
+	esw->fdb_table.addr_grp = g;
+	esw->fdb_table.fdb = fdb;
+out:
+	kfree(flow_group_in);
+	if (err && !IS_ERR_OR_NULL(fdb))
+		mlx5_destroy_flow_table(fdb);
+	return err;
+}
+
+static void esw_destroy_fdb_table(struct mlx5_eswitch *esw)
+{
+	if (!esw->fdb_table.fdb)
+		return;
+
+	esw_debug(esw->dev, "Destroy FDB Table\n");
+	mlx5_destroy_flow_group(esw->fdb_table.addr_grp);
+	mlx5_destroy_flow_table(esw->fdb_table.fdb);
+	esw->fdb_table.fdb = NULL;
+	esw->fdb_table.addr_grp = NULL;
+}
+
+/* E-Switch vport UC/MC lists management */
+typedef int (*vport_addr_action)(struct mlx5_eswitch *esw,
+				 struct vport_addr *vaddr);
+
+static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
+{
+	struct hlist_head *hash = esw->l2_table.l2_hash;
+	struct esw_uc_addr *esw_uc;
+	u8 *mac = vaddr->node.addr;
+	u32 vport = vaddr->vport;
+	int err;
+
+	esw_uc = l2addr_hash_find(hash, mac, struct esw_uc_addr);
+	if (esw_uc) {
+		esw_warn(esw->dev,
+			 "Failed to set L2 mac(%pM) for vport(%d), mac is already in use by vport(%d)\n",
+			 mac, vport, esw_uc->vport);
+		return -EEXIST;
+	}
+
+	esw_uc = l2addr_hash_add(hash, mac, struct esw_uc_addr, GFP_KERNEL);
+	if (!esw_uc)
+		return -ENOMEM;
+	esw_uc->vport = vport;
+
+	err = set_l2_table_entry(esw->dev, mac, 0, 0, &esw_uc->table_index);
+	if (err)
+		goto abort;
+
+	if (esw->fdb_table.fdb) /* SRIOV is enabled: Forward UC MAC to vport */
+		vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport);
+
+	esw_debug(esw->dev, "\tADDED UC MAC: vport[%d] %pM index:%d fr(%p)\n",
+		  vport, mac, esw_uc->table_index, vaddr->flow_rule);
+	return err;
+abort:
+	l2addr_hash_del(esw_uc);
+	return err;
+}
+
+static int esw_del_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
+{
+	struct hlist_head *hash = esw->l2_table.l2_hash;
+	struct esw_uc_addr *esw_uc;
+	u8 *mac = vaddr->node.addr;
+	u32 vport = vaddr->vport;
+
+	esw_uc = l2addr_hash_find(hash, mac, struct esw_uc_addr);
+	if (!esw_uc || esw_uc->vport != vport) {
+		esw_debug(esw->dev,
+			  "MAC(%pM) doesn't belong to vport (%d)\n",
+			  mac, vport);
+		return -EINVAL;
+	}
+	esw_debug(esw->dev, "\tDELETE UC MAC: vport[%d] %pM index:%d fr(%p)\n",
+		  vport, mac, esw_uc->table_index, vaddr->flow_rule);
+
+	del_l2_table_entry(esw->dev, esw_uc->table_index);
+
+	if (vaddr->flow_rule)
+		mlx5_del_flow_rule(vaddr->flow_rule);
+	vaddr->flow_rule = NULL;
+
+	l2addr_hash_del(esw_uc);
+	return 0;
+}
+
+static int esw_add_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
+{
+	struct hlist_head *hash = esw->mc_table;
+	struct esw_mc_addr *esw_mc;
+	u8 *mac = vaddr->node.addr;
+	u32 vport = vaddr->vport;
+
+	if (!esw->fdb_table.fdb)
+		return 0;
+
+	esw_mc = l2addr_hash_find(hash, mac, struct esw_mc_addr);
+	if (esw_mc)
+		goto add;
+
+	esw_mc = l2addr_hash_add(hash, mac, struct esw_mc_addr, GFP_KERNEL);
+	if (!esw_mc)
+		return -ENOMEM;
+
+	esw_mc->uplink_rule = /* Forward MC MAC to Uplink */
+		esw_fdb_set_vport_rule(esw, mac, UPLINK_VPORT);
+add:
+	esw_mc->refcnt++;
+	/* Forward MC MAC to vport */
+	vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport);
+	esw_debug(esw->dev,
+		  "\tADDED MC MAC: vport[%d] %pM fr(%p) refcnt(%d) uplinkfr(%p)\n",
+		  vport, mac, vaddr->flow_rule,
+		  esw_mc->refcnt, esw_mc->uplink_rule);
+	return 0;
+}
+
+static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
+{
+	struct hlist_head *hash = esw->mc_table;
+	struct esw_mc_addr *esw_mc;
+	u8 *mac = vaddr->node.addr;
+	u32 vport = vaddr->vport;
+
+	if (!esw->fdb_table.fdb)
+		return 0;
+
+	esw_mc = l2addr_hash_find(hash, mac, struct esw_mc_addr);
+	if (!esw_mc) {
+		esw_warn(esw->dev,
+			 "Failed to find eswitch MC addr for MAC(%pM) vport(%d)",
+			 mac, vport);
+		return -EINVAL;
+	}
+	esw_debug(esw->dev,
+		  "\tDELETE MC MAC: vport[%d] %pM fr(%p) refcnt(%d) uplinkfr(%p)\n",
+		  vport, mac, vaddr->flow_rule, esw_mc->refcnt,
+		  esw_mc->uplink_rule);
+
+	if (vaddr->flow_rule)
+		mlx5_del_flow_rule(vaddr->flow_rule);
+	vaddr->flow_rule = NULL;
+
+	if (--esw_mc->refcnt)
+		return 0;
+
+	if (esw_mc->uplink_rule)
+		mlx5_del_flow_rule(esw_mc->uplink_rule);
+
+	l2addr_hash_del(esw_mc);
+	return 0;
+}
+
+/* Apply vport UC/MC list to HW l2 table and FDB table */
+static void esw_apply_vport_addr_list(struct mlx5_eswitch *esw,
+				      u32 vport_num, int list_type)
+{
+	struct mlx5_vport *vport = &esw->vports[vport_num];
+	bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
+	vport_addr_action vport_addr_add;
+	vport_addr_action vport_addr_del;
+	struct vport_addr *addr;
+	struct l2addr_node *node;
+	struct hlist_head *hash;
+	struct hlist_node *tmp;
+	int hi;
+
+	vport_addr_add = is_uc ? esw_add_uc_addr :
+				 esw_add_mc_addr;
+	vport_addr_del = is_uc ? esw_del_uc_addr :
+				 esw_del_mc_addr;
+
+	hash = is_uc ? vport->uc_list : vport->mc_list;
+	for_each_l2hash_node(node, tmp, hash, hi) {
+		addr = container_of(node, struct vport_addr, node);
+		switch (addr->action) {
+		case MLX5_ACTION_ADD:
+			vport_addr_add(esw, addr);
+			addr->action = MLX5_ACTION_NONE;
+			break;
+		case MLX5_ACTION_DEL:
+			vport_addr_del(esw, addr);
+			l2addr_hash_del(addr);
+			break;
+		}
+	}
+}
+
+/* Sync vport UC/MC list from vport context */
+static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
+				       u32 vport_num, int list_type)
+{
+	struct mlx5_vport *vport = &esw->vports[vport_num];
+	bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
+	u8 (*mac_list)[ETH_ALEN];
+	struct l2addr_node *node;
+	struct vport_addr *addr;
+	struct hlist_head *hash;
+	struct hlist_node *tmp;
+	int size;
+	int err;
+	int hi;
+	int i;
+
+	size = is_uc ? MLX5_MAX_UC_PER_VPORT(esw->dev) :
+		       MLX5_MAX_MC_PER_VPORT(esw->dev);
+
+	mac_list = kcalloc(size, ETH_ALEN, GFP_KERNEL);
+	if (!mac_list)
+		return;
+
+	hash = is_uc ? vport->uc_list : vport->mc_list;
+
+	for_each_l2hash_node(node, tmp, hash, hi) {
+		addr = container_of(node, struct vport_addr, node);
+		addr->action = MLX5_ACTION_DEL;
+	}
+
+	err = mlx5_query_nic_vport_mac_list(esw->dev, vport_num, list_type,
+					    mac_list, &size);
+	if (err)
+		return;
+	esw_debug(esw->dev, "vport[%d] context update %s list size (%d)\n",
+		  vport_num, is_uc ? "UC" : "MC", size);
+
+	for (i = 0; i < size; i++) {
+		if (is_uc && !is_valid_ether_addr(mac_list[i]))
+			continue;
+
+		if (!is_uc && !is_multicast_ether_addr(mac_list[i]))
+			continue;
+
+		addr = l2addr_hash_find(hash, mac_list[i], struct vport_addr);
+		if (addr) {
+			addr->action = MLX5_ACTION_NONE;
+			continue;
+		}
+
+		addr = l2addr_hash_add(hash, mac_list[i], struct vport_addr,
+				       GFP_KERNEL);
+		if (!addr) {
+			esw_warn(esw->dev,
+				 "Failed to add MAC(%pM) to vport[%d] DB\n",
+				 mac_list[i], vport_num);
+			continue;
+		}
+		addr->vport = vport_num;
+		addr->action = MLX5_ACTION_ADD;
+	}
+	kfree(mac_list);
+}
+
+static void esw_vport_change_handler(struct work_struct *work)
+{
+	struct mlx5_vport *vport =
+		container_of(work, struct mlx5_vport, vport_change_handler);
+	struct mlx5_core_dev *dev = vport->dev;
+	struct mlx5_eswitch *esw = dev->priv.eswitch;
+	u8 mac[ETH_ALEN];
+
+	mlx5_query_nic_vport_mac_address(dev, vport->vport, mac);
+	esw_debug(dev, "vport[%d] Context Changed: perm mac: %pM\n",
+		  vport->vport, mac);
+
+	if (vport->enabled_events & UC_ADDR_CHANGE) {
+		esw_update_vport_addr_list(esw, vport->vport,
+					   MLX5_NVPRT_LIST_TYPE_UC);
+		esw_apply_vport_addr_list(esw, vport->vport,
+					  MLX5_NVPRT_LIST_TYPE_UC);
+	}
+
+	if (vport->enabled_events & MC_ADDR_CHANGE) {
+		esw_update_vport_addr_list(esw, vport->vport,
+					   MLX5_NVPRT_LIST_TYPE_MC);
+		esw_apply_vport_addr_list(esw, vport->vport,
+					  MLX5_NVPRT_LIST_TYPE_MC);
+	}
+
+	esw_debug(esw->dev, "vport[%d] Context Changed: Done\n", vport->vport);
+	if (vport->enabled)
+		arm_vport_context_events_cmd(dev, vport->vport,
+					     vport->enabled_events);
+}
+
+static void esw_enable_vport(struct mlx5_eswitch *esw, int vport_num,
+			     int enable_events)
+{
+	struct mlx5_vport *vport = &esw->vports[vport_num];
+	unsigned long flags;
+
+	WARN_ON(vport->enabled);
+
+	esw_debug(esw->dev, "Enabling VPORT(%d)\n", vport_num);
+	mlx5_modify_vport_admin_state(esw->dev,
+				      MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT,
+				      vport_num,
+				      MLX5_ESW_VPORT_ADMIN_STATE_AUTO);
+
+	/* Sync with current vport context */
+	vport->enabled_events = enable_events;
+	esw_vport_change_handler(&vport->vport_change_handler);
+
+	spin_lock_irqsave(&vport->lock, flags);
+	vport->enabled = true;
+	spin_unlock_irqrestore(&vport->lock, flags);
+
+	arm_vport_context_events_cmd(esw->dev, vport_num, enable_events);
+
+	esw->enabled_vports++;
+	esw_debug(esw->dev, "Enabled VPORT(%d)\n", vport_num);
+}
+
+static void esw_cleanup_vport(struct mlx5_eswitch *esw, u16 vport_num)
+{
+	struct mlx5_vport *vport = &esw->vports[vport_num];
+	struct l2addr_node *node;
+	struct vport_addr *addr;
+	struct hlist_node *tmp;
+	int hi;
+
+	for_each_l2hash_node(node, tmp, vport->uc_list, hi) {
+		addr = container_of(node, struct vport_addr, node);
+		addr->action = MLX5_ACTION_DEL;
+	}
+	esw_apply_vport_addr_list(esw, vport_num, MLX5_NVPRT_LIST_TYPE_UC);
+
+	for_each_l2hash_node(node, tmp, vport->mc_list, hi) {
+		addr = container_of(node, struct vport_addr, node);
+		addr->action = MLX5_ACTION_DEL;
+	}
+	esw_apply_vport_addr_list(esw, vport_num, MLX5_NVPRT_LIST_TYPE_MC);
+}
+
+static void esw_disable_vport(struct mlx5_eswitch *esw, int vport_num)
+{
+	struct mlx5_vport *vport = &esw->vports[vport_num];
+	unsigned long flags;
+
+	if (!vport->enabled)
+		return;
+
+	esw_debug(esw->dev, "Disabling vport(%d)\n", vport_num);
+	/* Mark this vport as disabled to discard new events */
+	spin_lock_irqsave(&vport->lock, flags);
+	vport->enabled = false;
+	vport->enabled_events = 0;
+	spin_unlock_irqrestore(&vport->lock, flags);
+
+	mlx5_modify_vport_admin_state(esw->dev,
+				      MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT,
+				      vport_num,
+				      MLX5_ESW_VPORT_ADMIN_STATE_DOWN);
+	/* Wait for current already scheduled events to complete */
+	flush_workqueue(esw->work_queue);
+	/* Disable events from this vport */
+	arm_vport_context_events_cmd(esw->dev, vport->vport, 0);
+	/* We don't assume VFs will cleanup after themselves */
+	esw_cleanup_vport(esw, vport_num);
+	esw->enabled_vports--;
+}
+
+/* Public E-Switch API */
+int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs)
+{
+	int err;
+	int i;
+
+	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) ||
+	    MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH)
+		return 0;
+
+	if (!MLX5_CAP_GEN(esw->dev, eswitch_flow_table) ||
+	    !MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ft_support)) {
+		esw_warn(esw->dev, "E-Switch FDB is not supported, aborting ...\n");
+		return -ENOTSUPP;
+	}
+
+	esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d)\n", nvfs);
+
+	esw_disable_vport(esw, 0);
+
+	err = esw_create_fdb_table(esw, nvfs + 1);
+	if (err)
+		goto abort;
+
+	for (i = 0; i <= nvfs; i++)
+		esw_enable_vport(esw, i, SRIOV_VPORT_EVENTS);
+
+	esw_info(esw->dev, "SRIOV enabled: active vports(%d)\n",
+		 esw->enabled_vports);
+	return 0;
+
+abort:
+	esw_enable_vport(esw, 0, UC_ADDR_CHANGE);
+	return err;
+}
+
+void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw)
+{
+	int i;
+
+	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) ||
+	    MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH)
+		return;
+
+	esw_info(esw->dev, "disable SRIOV: active vports(%d)\n",
+		 esw->enabled_vports);
+
+	for (i = 0; i < esw->total_vports; i++)
+		esw_disable_vport(esw, i);
+
+	esw_destroy_fdb_table(esw);
+
+	/* VPORT 0 (PF) must be enabled back with non-sriov configuration */
+	esw_enable_vport(esw, 0, UC_ADDR_CHANGE);
+}
+
+int mlx5_eswitch_init(struct mlx5_core_dev *dev)
+{
+	int l2_table_size = 1 << MLX5_CAP_GEN(dev, log_max_l2_table);
+	int total_vports = 1 + pci_sriov_get_totalvfs(dev->pdev);
+	struct mlx5_eswitch *esw;
+	int vport_num;
+	int err;
+
+	if (!MLX5_CAP_GEN(dev, vport_group_manager) ||
+	    MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH)
+		return 0;
+
+	esw_info(dev,
+		 "Total vports %d, l2 table size(%d), per vport: max uc(%d) max mc(%d)\n",
+		 total_vports, l2_table_size,
+		 MLX5_MAX_UC_PER_VPORT(dev),
+		 MLX5_MAX_MC_PER_VPORT(dev));
+
+	esw = kzalloc(sizeof(*esw), GFP_KERNEL);
+	if (!esw)
+		return -ENOMEM;
+
+	esw->dev = dev;
+
+	esw->l2_table.bitmap = kcalloc(BITS_TO_LONGS(l2_table_size),
+				   sizeof(uintptr_t), GFP_KERNEL);
+	if (!esw->l2_table.bitmap) {
+		err = -ENOMEM;
+		goto abort;
+	}
+	esw->l2_table.size = l2_table_size;
+
+	esw->work_queue = create_singlethread_workqueue("mlx5_esw_wq");
+	if (!esw->work_queue) {
+		err = -ENOMEM;
+		goto abort;
+	}
+
+	esw->vports = kcalloc(total_vports, sizeof(struct mlx5_vport),
+			      GFP_KERNEL);
+	if (!esw->vports) {
+		err = -ENOMEM;
+		goto abort;
+	}
+
+	for (vport_num = 0; vport_num < total_vports; vport_num++) {
+		struct mlx5_vport *vport = &esw->vports[vport_num];
+
+		vport->vport = vport_num;
+		vport->dev = dev;
+		INIT_WORK(&vport->vport_change_handler,
+			  esw_vport_change_handler);
+		spin_lock_init(&vport->lock);
+	}
+
+	esw->total_vports = total_vports;
+	esw->enabled_vports = 0;
+
+	dev->priv.eswitch = esw;
+	esw_enable_vport(esw, 0, UC_ADDR_CHANGE);
+	/* VF Vports will be enabled when SRIOV is enabled */
+	return 0;
+abort:
+	if (esw->work_queue)
+		destroy_workqueue(esw->work_queue);
+	kfree(esw->l2_table.bitmap);
+	kfree(esw->vports);
+	kfree(esw);
+	return err;
+}
+
+void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
+{
+	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) ||
+	    MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH)
+		return;
+
+	esw_info(esw->dev, "cleanup\n");
+	esw_disable_vport(esw, 0);
+
+	esw->dev->priv.eswitch = NULL;
+	destroy_workqueue(esw->work_queue);
+	kfree(esw->l2_table.bitmap);
+	kfree(esw->vports);
+	kfree(esw);
+}
+
+void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe)
+{
+	struct mlx5_eqe_vport_change *vc_eqe = &eqe->data.vport_change;
+	u16 vport_num = be16_to_cpu(vc_eqe->vport_num);
+	struct mlx5_vport *vport;
+
+	if (!esw) {
+		pr_warn("MLX5 E-Switch: vport %d got an event while eswitch is not initialized\n",
+			vport_num);
+		return;
+	}
+
+	vport = &esw->vports[vport_num];
+	spin_lock(&vport->lock);
+	if (vport->enabled)
+		queue_work(esw->work_queue, &vport->vport_change_handler);
+	spin_unlock(&vport->lock);
+}
+
+/* Vport Administration */
+#define ESW_ALLOWED(esw) \
+	(esw && MLX5_CAP_GEN(esw->dev, vport_group_manager) && mlx5_core_is_pf(esw->dev))
+#define LEGAL_VPORT(esw, vport) (vport >= 0 && vport < esw->total_vports)
+
+int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw,
+			       int vport, u8 mac[ETH_ALEN])
+{
+	int err = 0;
+
+	if (!ESW_ALLOWED(esw))
+		return -EPERM;
+	if (!LEGAL_VPORT(esw, vport))
+		return -EINVAL;
+
+	err = mlx5_modify_nic_vport_mac_address(esw->dev, vport, mac);
+	if (err) {
+		mlx5_core_warn(esw->dev,
+			       "Failed to mlx5_modify_nic_vport_mac vport(%d) err=(%d)\n",
+			       vport, err);
+		return err;
+	}
+
+	return err;
+}
+
+int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw,
+				 int vport, int link_state)
+{
+	if (!ESW_ALLOWED(esw))
+		return -EPERM;
+	if (!LEGAL_VPORT(esw, vport))
+		return -EINVAL;
+
+	return mlx5_modify_vport_admin_state(esw->dev,
+					     MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT,
+					     vport, link_state);
+}
+
+int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw,
+				  int vport, struct ifla_vf_info *ivi)
+{
+	u16 vlan;
+	u8 qos;
+
+	if (!ESW_ALLOWED(esw))
+		return -EPERM;
+	if (!LEGAL_VPORT(esw, vport))
+		return -EINVAL;
+
+	memset(ivi, 0, sizeof(*ivi));
+	ivi->vf = vport - 1;
+
+	mlx5_query_nic_vport_mac_address(esw->dev, vport, ivi->mac);
+	ivi->linkstate = mlx5_query_vport_admin_state(esw->dev,
+						      MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT,
+						      vport);
+	query_esw_vport_cvlan(esw->dev, vport, &vlan, &qos);
+	ivi->vlan = vlan;
+	ivi->qos = qos;
+	ivi->spoofchk = 0;
+
+	return 0;
+}
+
+int mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
+				int vport, u16 vlan, u8 qos)
+{
+	int set = 0;
+
+	if (!ESW_ALLOWED(esw))
+		return -EPERM;
+	if (!LEGAL_VPORT(esw, vport) || (vlan > 4095) || (qos > 7))
+		return -EINVAL;
+
+	if (vlan || qos)
+		set = 1;
+
+	return modify_esw_vport_cvlan(esw->dev, vport, vlan, qos, set);
+}
+
+int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
+				 int vport,
+				 struct ifla_vf_stats *vf_stats)
+{
+	int outlen = MLX5_ST_SZ_BYTES(query_vport_counter_out);
+	u32 in[MLX5_ST_SZ_DW(query_vport_counter_in)];
+	int err = 0;
+	u32 *out;
+
+	if (!ESW_ALLOWED(esw))
+		return -EPERM;
+	if (!LEGAL_VPORT(esw, vport))
+		return -EINVAL;
+
+	out = mlx5_vzalloc(outlen);
+	if (!out)
+		return -ENOMEM;
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(query_vport_counter_in, in, opcode,
+		 MLX5_CMD_OP_QUERY_VPORT_COUNTER);
+	MLX5_SET(query_vport_counter_in, in, op_mod, 0);
+	MLX5_SET(query_vport_counter_in, in, vport_number, vport);
+	if (vport)
+		MLX5_SET(query_vport_counter_in, in, other_vport, 1);
+
+	memset(out, 0, outlen);
+	err = mlx5_cmd_exec(esw->dev, in, sizeof(in), out, outlen);
+	if (err)
+		goto free_out;
+
+	#define MLX5_GET_CTR(p, x) \
+		MLX5_GET64(query_vport_counter_out, p, x)
+
+	memset(vf_stats, 0, sizeof(*vf_stats));
+	vf_stats->rx_packets =
+		MLX5_GET_CTR(out, received_eth_unicast.packets) +
+		MLX5_GET_CTR(out, received_eth_multicast.packets) +
+		MLX5_GET_CTR(out, received_eth_broadcast.packets);
+
+	vf_stats->rx_bytes =
+		MLX5_GET_CTR(out, received_eth_unicast.octets) +
+		MLX5_GET_CTR(out, received_eth_multicast.octets) +
+		MLX5_GET_CTR(out, received_eth_broadcast.octets);
+
+	vf_stats->tx_packets =
+		MLX5_GET_CTR(out, transmitted_eth_unicast.packets) +
+		MLX5_GET_CTR(out, transmitted_eth_multicast.packets) +
+		MLX5_GET_CTR(out, transmitted_eth_broadcast.packets);
+
+	vf_stats->tx_bytes =
+		MLX5_GET_CTR(out, transmitted_eth_unicast.octets) +
+		MLX5_GET_CTR(out, transmitted_eth_multicast.octets) +
+		MLX5_GET_CTR(out, transmitted_eth_broadcast.octets);
+
+	vf_stats->multicast =
+		MLX5_GET_CTR(out, received_eth_multicast.packets);
+
+	vf_stats->broadcast =
+		MLX5_GET_CTR(out, received_eth_broadcast.packets);
+
+free_out:
+	kvfree(out);
+	return err;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
new file mode 100644
index 000000000..3416a428f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -0,0 +1,147 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies, Ltd.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __MLX5_ESWITCH_H__
+#define __MLX5_ESWITCH_H__
+
+#include <linux/if_ether.h>
+#include <linux/if_link.h>
+#include <linux/mlx5/device.h>
+
+#define MLX5_MAX_UC_PER_VPORT(dev) \
+	(1 << MLX5_CAP_GEN(dev, log_max_current_uc_list))
+
+#define MLX5_MAX_MC_PER_VPORT(dev) \
+	(1 << MLX5_CAP_GEN(dev, log_max_current_mc_list))
+
+#define MLX5_L2_ADDR_HASH_SIZE (BIT(BITS_PER_BYTE))
+#define MLX5_L2_ADDR_HASH(addr) (addr[5])
+
+/* L2 -mac address based- hash helpers */
+struct l2addr_node {
+	struct hlist_node hlist;
+	u8                addr[ETH_ALEN];
+};
+
+#define for_each_l2hash_node(hn, tmp, hash, i) \
+	for (i = 0; i < MLX5_L2_ADDR_HASH_SIZE; i++) \
+		hlist_for_each_entry_safe(hn, tmp, &hash[i], hlist)
+
+#define l2addr_hash_find(hash, mac, type) ({                \
+	int ix = MLX5_L2_ADDR_HASH(mac);                    \
+	bool found = false;                                 \
+	type *ptr = NULL;                                   \
+							    \
+	hlist_for_each_entry(ptr, &hash[ix], node.hlist)    \
+		if (ether_addr_equal(ptr->node.addr, mac)) {\
+			found = true;                       \
+			break;                              \
+		}                                           \
+	if (!found)                                         \
+		ptr = NULL;                                 \
+	ptr;                                                \
+})
+
+#define l2addr_hash_add(hash, mac, type, gfp) ({            \
+	int ix = MLX5_L2_ADDR_HASH(mac);                    \
+	type *ptr = NULL;                                   \
+							    \
+	ptr = kzalloc(sizeof(type), gfp);                   \
+	if (ptr) {                                          \
+		ether_addr_copy(ptr->node.addr, mac);       \
+		hlist_add_head(&ptr->node.hlist, &hash[ix]);\
+	}                                                   \
+	ptr;                                                \
+})
+
+#define l2addr_hash_del(ptr) ({                             \
+	hlist_del(&ptr->node.hlist);                        \
+	kfree(ptr);                                         \
+})
+
+struct mlx5_vport {
+	struct mlx5_core_dev    *dev;
+	int                     vport;
+	struct hlist_head       uc_list[MLX5_L2_ADDR_HASH_SIZE];
+	struct hlist_head       mc_list[MLX5_L2_ADDR_HASH_SIZE];
+	struct work_struct      vport_change_handler;
+
+	/* This spinlock protects access to vport data, between
+	 * "esw_vport_disable" and ongoing interrupt "mlx5_eswitch_vport_event"
+	 * once vport marked as disabled new interrupts are discarded.
+	 */
+	spinlock_t              lock; /* vport events sync */
+	bool                    enabled;
+	u16                     enabled_events;
+};
+
+struct mlx5_l2_table {
+	struct hlist_head l2_hash[MLX5_L2_ADDR_HASH_SIZE];
+	u32                  size;
+	unsigned long        *bitmap;
+};
+
+struct mlx5_eswitch_fdb {
+	void *fdb;
+	struct mlx5_flow_group *addr_grp;
+};
+
+struct mlx5_eswitch {
+	struct mlx5_core_dev    *dev;
+	struct mlx5_l2_table    l2_table;
+	struct mlx5_eswitch_fdb fdb_table;
+	struct hlist_head       mc_table[MLX5_L2_ADDR_HASH_SIZE];
+	struct workqueue_struct *work_queue;
+	struct mlx5_vport       *vports;
+	int                     total_vports;
+	int                     enabled_vports;
+};
+
+/* E-Switch API */
+int mlx5_eswitch_init(struct mlx5_core_dev *dev);
+void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw);
+void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe);
+int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs);
+void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw);
+int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw,
+			       int vport, u8 mac[ETH_ALEN]);
+int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw,
+				 int vport, int link_state);
+int mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
+				int vport, u16 vlan, u8 qos);
+int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw,
+				  int vport, struct ifla_vf_info *ivi);
+int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
+				 int vport,
+				 struct ifla_vf_stats *vf_stats);
+
+#endif /* __MLX5_ESWITCH_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c
deleted file mode 100644
index ca90b9bc3..000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c
+++ /dev/null
@@ -1,422 +0,0 @@
-/*
- * Copyright (c) 2013-2015, Mellanox Technologies, Ltd.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- */
-
-#include <linux/export.h>
-#include <linux/mlx5/driver.h>
-#include <linux/mlx5/flow_table.h>
-#include "mlx5_core.h"
-
-struct mlx5_ftg {
-	struct mlx5_flow_table_group    g;
-	u32				id;
-	u32				start_ix;
-};
-
-struct mlx5_flow_table {
-	struct mlx5_core_dev	*dev;
-	u8			level;
-	u8			type;
-	u32			id;
-	struct mutex		mutex; /* sync bitmap alloc */
-	u16			num_groups;
-	struct mlx5_ftg		*group;
-	unsigned long		*bitmap;
-	u32			size;
-};
-
-static int mlx5_set_flow_entry_cmd(struct mlx5_flow_table *ft, u32 group_ix,
-				   u32 flow_index, void *flow_context)
-{
-	u32 out[MLX5_ST_SZ_DW(set_fte_out)];
-	u32 *in;
-	void *in_flow_context;
-	int fcdls =
-		MLX5_GET(flow_context, flow_context, destination_list_size) *
-		MLX5_ST_SZ_BYTES(dest_format_struct);
-	int inlen = MLX5_ST_SZ_BYTES(set_fte_in) + fcdls;
-	int err;
-
-	in = mlx5_vzalloc(inlen);
-	if (!in) {
-		mlx5_core_warn(ft->dev, "failed to allocate inbox\n");
-		return -ENOMEM;
-	}
-
-	MLX5_SET(set_fte_in, in, table_type, ft->type);
-	MLX5_SET(set_fte_in, in, table_id,   ft->id);
-	MLX5_SET(set_fte_in, in, flow_index, flow_index);
-	MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY);
-
-	in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context);
-	memcpy(in_flow_context, flow_context,
-	       MLX5_ST_SZ_BYTES(flow_context) + fcdls);
-
-	MLX5_SET(flow_context, in_flow_context, group_id,
-		 ft->group[group_ix].id);
-
-	memset(out, 0, sizeof(out));
-	err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out,
-					 sizeof(out));
-	kvfree(in);
-
-	return err;
-}
-
-static void mlx5_del_flow_entry_cmd(struct mlx5_flow_table *ft, u32 flow_index)
-{
-	u32 in[MLX5_ST_SZ_DW(delete_fte_in)];
-	u32 out[MLX5_ST_SZ_DW(delete_fte_out)];
-
-	memset(in, 0, sizeof(in));
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_DFTEI(p, x, v) MLX5_SET(delete_fte_in, p, x, v)
-	MLX5_SET_DFTEI(in, table_type, ft->type);
-	MLX5_SET_DFTEI(in, table_id,   ft->id);
-	MLX5_SET_DFTEI(in, flow_index, flow_index);
-	MLX5_SET_DFTEI(in, opcode,     MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY);
-
-	mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out));
-}
-
-static void mlx5_destroy_flow_group_cmd(struct mlx5_flow_table *ft, int i)
-{
-	u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)];
-	u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)];
-
-	memset(in, 0, sizeof(in));
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_DFGI(p, x, v) MLX5_SET(destroy_flow_group_in, p, x, v)
-	MLX5_SET_DFGI(in, table_type, ft->type);
-	MLX5_SET_DFGI(in, table_id,   ft->id);
-	MLX5_SET_DFGI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_GROUP);
-	MLX5_SET_DFGI(in, group_id, ft->group[i].id);
-	mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out));
-}
-
-static int mlx5_create_flow_group_cmd(struct mlx5_flow_table *ft, int i)
-{
-	u32 out[MLX5_ST_SZ_DW(create_flow_group_out)];
-	u32 *in;
-	void *in_match_criteria;
-	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
-	struct mlx5_flow_table_group *g = &ft->group[i].g;
-	u32 start_ix = ft->group[i].start_ix;
-	u32 end_ix = start_ix + (1 << g->log_sz) - 1;
-	int err;
-
-	in = mlx5_vzalloc(inlen);
-	if (!in) {
-		mlx5_core_warn(ft->dev, "failed to allocate inbox\n");
-		return -ENOMEM;
-	}
-	in_match_criteria = MLX5_ADDR_OF(create_flow_group_in, in,
-					 match_criteria);
-
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_CFGI(p, x, v) MLX5_SET(create_flow_group_in, p, x, v)
-	MLX5_SET_CFGI(in, table_type,            ft->type);
-	MLX5_SET_CFGI(in, table_id,              ft->id);
-	MLX5_SET_CFGI(in, opcode,                MLX5_CMD_OP_CREATE_FLOW_GROUP);
-	MLX5_SET_CFGI(in, start_flow_index,      start_ix);
-	MLX5_SET_CFGI(in, end_flow_index,        end_ix);
-	MLX5_SET_CFGI(in, match_criteria_enable, g->match_criteria_enable);
-
-	memcpy(in_match_criteria, g->match_criteria,
-	       MLX5_ST_SZ_BYTES(fte_match_param));
-
-	err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out,
-					 sizeof(out));
-	if (!err)
-		ft->group[i].id = MLX5_GET(create_flow_group_out, out,
-					   group_id);
-
-	kvfree(in);
-
-	return err;
-}
-
-static void mlx5_destroy_flow_table_groups(struct mlx5_flow_table *ft)
-{
-	int i;
-
-	for (i = 0; i < ft->num_groups; i++)
-		mlx5_destroy_flow_group_cmd(ft, i);
-}
-
-static int mlx5_create_flow_table_groups(struct mlx5_flow_table *ft)
-{
-	int err;
-	int i;
-
-	for (i = 0; i < ft->num_groups; i++) {
-		err = mlx5_create_flow_group_cmd(ft, i);
-		if (err)
-			goto err_destroy_flow_table_groups;
-	}
-
-	return 0;
-
-err_destroy_flow_table_groups:
-	for (i--; i >= 0; i--)
-		mlx5_destroy_flow_group_cmd(ft, i);
-
-	return err;
-}
-
-static int mlx5_create_flow_table_cmd(struct mlx5_flow_table *ft)
-{
-	u32 in[MLX5_ST_SZ_DW(create_flow_table_in)];
-	u32 out[MLX5_ST_SZ_DW(create_flow_table_out)];
-	int err;
-
-	memset(in, 0, sizeof(in));
-
-	MLX5_SET(create_flow_table_in, in, table_type, ft->type);
-	MLX5_SET(create_flow_table_in, in, level,      ft->level);
-	MLX5_SET(create_flow_table_in, in, log_size,   order_base_2(ft->size));
-
-	MLX5_SET(create_flow_table_in, in, opcode,
-		 MLX5_CMD_OP_CREATE_FLOW_TABLE);
-
-	memset(out, 0, sizeof(out));
-	err = mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out,
-					 sizeof(out));
-	if (err)
-		return err;
-
-	ft->id = MLX5_GET(create_flow_table_out, out, table_id);
-
-	return 0;
-}
-
-static void mlx5_destroy_flow_table_cmd(struct mlx5_flow_table *ft)
-{
-	u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)];
-	u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)];
-
-	memset(in, 0, sizeof(in));
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_DFTI(p, x, v) MLX5_SET(destroy_flow_table_in, p, x, v)
-	MLX5_SET_DFTI(in, table_type, ft->type);
-	MLX5_SET_DFTI(in, table_id,   ft->id);
-	MLX5_SET_DFTI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_TABLE);
-
-	mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out));
-}
-
-static int mlx5_find_group(struct mlx5_flow_table *ft, u8 match_criteria_enable,
-			   u32 *match_criteria, int *group_ix)
-{
-	void *mc_outer = MLX5_ADDR_OF(fte_match_param, match_criteria,
-				      outer_headers);
-	void *mc_misc  = MLX5_ADDR_OF(fte_match_param, match_criteria,
-				      misc_parameters);
-	void *mc_inner = MLX5_ADDR_OF(fte_match_param, match_criteria,
-				      inner_headers);
-	int mc_outer_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4);
-	int mc_misc_sz  = MLX5_ST_SZ_BYTES(fte_match_set_misc);
-	int mc_inner_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4);
-	int i;
-
-	for (i = 0; i < ft->num_groups; i++) {
-		struct mlx5_flow_table_group *g = &ft->group[i].g;
-		void *gmc_outer = MLX5_ADDR_OF(fte_match_param,
-					       g->match_criteria,
-					       outer_headers);
-		void *gmc_misc  = MLX5_ADDR_OF(fte_match_param,
-					       g->match_criteria,
-					       misc_parameters);
-		void *gmc_inner = MLX5_ADDR_OF(fte_match_param,
-					       g->match_criteria,
-					       inner_headers);
-
-		if (g->match_criteria_enable != match_criteria_enable)
-			continue;
-
-		if (match_criteria_enable & MLX5_MATCH_OUTER_HEADERS)
-			if (memcmp(mc_outer, gmc_outer, mc_outer_sz))
-				continue;
-
-		if (match_criteria_enable & MLX5_MATCH_MISC_PARAMETERS)
-			if (memcmp(mc_misc, gmc_misc, mc_misc_sz))
-				continue;
-
-		if (match_criteria_enable & MLX5_MATCH_INNER_HEADERS)
-			if (memcmp(mc_inner, gmc_inner, mc_inner_sz))
-				continue;
-
-		*group_ix = i;
-		return 0;
-	}
-
-	return -EINVAL;
-}
-
-static int alloc_flow_index(struct mlx5_flow_table *ft, int group_ix, u32 *ix)
-{
-	struct mlx5_ftg *g = &ft->group[group_ix];
-	int err = 0;
-
-	mutex_lock(&ft->mutex);
-
-	*ix = find_next_zero_bit(ft->bitmap, ft->size, g->start_ix);
-	if (*ix >= (g->start_ix + (1 << g->g.log_sz)))
-		err = -ENOSPC;
-	else
-		__set_bit(*ix, ft->bitmap);
-
-	mutex_unlock(&ft->mutex);
-
-	return err;
-}
-
-static void mlx5_free_flow_index(struct mlx5_flow_table *ft, u32 ix)
-{
-	__clear_bit(ix, ft->bitmap);
-}
-
-int mlx5_add_flow_table_entry(void *flow_table, u8 match_criteria_enable,
-			      void *match_criteria, void *flow_context,
-			      u32 *flow_index)
-{
-	struct mlx5_flow_table *ft = flow_table;
-	int group_ix;
-	int err;
-
-	err = mlx5_find_group(ft, match_criteria_enable, match_criteria,
-			      &group_ix);
-	if (err) {
-		mlx5_core_warn(ft->dev, "mlx5_find_group failed\n");
-		return err;
-	}
-
-	err = alloc_flow_index(ft, group_ix, flow_index);
-	if (err) {
-		mlx5_core_warn(ft->dev, "alloc_flow_index failed\n");
-		return err;
-	}
-
-	return mlx5_set_flow_entry_cmd(ft, group_ix, *flow_index, flow_context);
-}
-EXPORT_SYMBOL(mlx5_add_flow_table_entry);
-
-void mlx5_del_flow_table_entry(void *flow_table, u32 flow_index)
-{
-	struct mlx5_flow_table *ft = flow_table;
-
-	mlx5_del_flow_entry_cmd(ft, flow_index);
-	mlx5_free_flow_index(ft, flow_index);
-}
-EXPORT_SYMBOL(mlx5_del_flow_table_entry);
-
-void *mlx5_create_flow_table(struct mlx5_core_dev *dev, u8 level, u8 table_type,
-			     u16 num_groups,
-			     struct mlx5_flow_table_group *group)
-{
-	struct mlx5_flow_table *ft;
-	u32 start_ix = 0;
-	u32 ft_size = 0;
-	void *gr;
-	void *bm;
-	int err;
-	int i;
-
-	for (i = 0; i < num_groups; i++)
-		ft_size += (1 << group[i].log_sz);
-
-	ft = kzalloc(sizeof(*ft), GFP_KERNEL);
-	gr = kcalloc(num_groups, sizeof(struct mlx5_ftg), GFP_KERNEL);
-	bm = kcalloc(BITS_TO_LONGS(ft_size), sizeof(uintptr_t), GFP_KERNEL);
-	if (!ft || !gr || !bm)
-		goto err_free_ft;
-
-	ft->group	= gr;
-	ft->bitmap	= bm;
-	ft->num_groups	= num_groups;
-	ft->level	= level;
-	ft->type	= table_type;
-	ft->size	= ft_size;
-	ft->dev		= dev;
-	mutex_init(&ft->mutex);
-
-	for (i = 0; i < ft->num_groups; i++) {
-		memcpy(&ft->group[i].g, &group[i], sizeof(*group));
-		ft->group[i].start_ix = start_ix;
-		start_ix += 1 << group[i].log_sz;
-	}
-
-	err = mlx5_create_flow_table_cmd(ft);
-	if (err)
-		goto err_free_ft;
-
-	err = mlx5_create_flow_table_groups(ft);
-	if (err)
-		goto err_destroy_flow_table_cmd;
-
-	return ft;
-
-err_destroy_flow_table_cmd:
-	mlx5_destroy_flow_table_cmd(ft);
-
-err_free_ft:
-	mlx5_core_warn(dev, "failed to alloc flow table\n");
-	kfree(bm);
-	kfree(gr);
-	kfree(ft);
-
-	return NULL;
-}
-EXPORT_SYMBOL(mlx5_create_flow_table);
-
-void mlx5_destroy_flow_table(void *flow_table)
-{
-	struct mlx5_flow_table *ft = flow_table;
-
-	mlx5_destroy_flow_table_groups(ft);
-	mlx5_destroy_flow_table_cmd(ft);
-	kfree(ft->bitmap);
-	kfree(ft->group);
-	kfree(ft);
-}
-EXPORT_SYMBOL(mlx5_destroy_flow_table);
-
-u32 mlx5_get_flow_table_id(void *flow_table)
-{
-	struct mlx5_flow_table *ft = flow_table;
-
-	return ft->id;
-}
-EXPORT_SYMBOL(mlx5_get_flow_table_id);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
new file mode 100644
index 000000000..a9894d2e8
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -0,0 +1,289 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/device.h>
+#include <linux/mlx5/mlx5_ifc.h>
+
+#include "fs_core.h"
+#include "fs_cmd.h"
+#include "mlx5_core.h"
+
+int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev,
+			    struct mlx5_flow_table *ft)
+{
+	u32 in[MLX5_ST_SZ_DW(set_flow_table_root_in)];
+	u32 out[MLX5_ST_SZ_DW(set_flow_table_root_out)];
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(set_flow_table_root_in, in, opcode,
+		 MLX5_CMD_OP_SET_FLOW_TABLE_ROOT);
+	MLX5_SET(set_flow_table_root_in, in, table_type, ft->type);
+	MLX5_SET(set_flow_table_root_in, in, table_id, ft->id);
+
+	memset(out, 0, sizeof(out));
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					  sizeof(out));
+}
+
+int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
+			       enum fs_flow_table_type type, unsigned int level,
+			       unsigned int log_size, struct mlx5_flow_table
+			       *next_ft, unsigned int *table_id)
+{
+	u32 out[MLX5_ST_SZ_DW(create_flow_table_out)];
+	u32 in[MLX5_ST_SZ_DW(create_flow_table_in)];
+	int err;
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(create_flow_table_in, in, opcode,
+		 MLX5_CMD_OP_CREATE_FLOW_TABLE);
+
+	if (next_ft) {
+		MLX5_SET(create_flow_table_in, in, table_miss_mode, 1);
+		MLX5_SET(create_flow_table_in, in, table_miss_id, next_ft->id);
+	}
+	MLX5_SET(create_flow_table_in, in, table_type, type);
+	MLX5_SET(create_flow_table_in, in, level, level);
+	MLX5_SET(create_flow_table_in, in, log_size, log_size);
+
+	memset(out, 0, sizeof(out));
+	err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					 sizeof(out));
+
+	if (!err)
+		*table_id = MLX5_GET(create_flow_table_out, out,
+				     table_id);
+	return err;
+}
+
+int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft)
+{
+	u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)];
+	u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(destroy_flow_table_in, in, opcode,
+		 MLX5_CMD_OP_DESTROY_FLOW_TABLE);
+	MLX5_SET(destroy_flow_table_in, in, table_type, ft->type);
+	MLX5_SET(destroy_flow_table_in, in, table_id, ft->id);
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					  sizeof(out));
+}
+
+int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev,
+			       struct mlx5_flow_table *ft,
+			       struct mlx5_flow_table *next_ft)
+{
+	u32 in[MLX5_ST_SZ_DW(modify_flow_table_in)];
+	u32 out[MLX5_ST_SZ_DW(modify_flow_table_out)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(modify_flow_table_in, in, opcode,
+		 MLX5_CMD_OP_MODIFY_FLOW_TABLE);
+	MLX5_SET(modify_flow_table_in, in, table_type, ft->type);
+	MLX5_SET(modify_flow_table_in, in, table_id, ft->id);
+	MLX5_SET(modify_flow_table_in, in, modify_field_select,
+		 MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID);
+	if (next_ft) {
+		MLX5_SET(modify_flow_table_in, in, table_miss_mode, 1);
+		MLX5_SET(modify_flow_table_in, in, table_miss_id, next_ft->id);
+	} else {
+		MLX5_SET(modify_flow_table_in, in, table_miss_mode, 0);
+	}
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					  sizeof(out));
+}
+
+int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev,
+			       struct mlx5_flow_table *ft,
+			       u32 *in,
+			       unsigned int *group_id)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	u32 out[MLX5_ST_SZ_DW(create_flow_group_out)];
+	int err;
+
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(create_flow_group_in, in, opcode,
+		 MLX5_CMD_OP_CREATE_FLOW_GROUP);
+	MLX5_SET(create_flow_group_in, in, table_type, ft->type);
+	MLX5_SET(create_flow_group_in, in, table_id, ft->id);
+
+	err = mlx5_cmd_exec_check_status(dev, in,
+					 inlen, out,
+					 sizeof(out));
+	if (!err)
+		*group_id = MLX5_GET(create_flow_group_out, out,
+				     group_id);
+
+	return err;
+}
+
+int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft,
+				unsigned int group_id)
+{
+	u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)];
+	u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(destroy_flow_group_in, in, opcode,
+		 MLX5_CMD_OP_DESTROY_FLOW_GROUP);
+	MLX5_SET(destroy_flow_group_in, in, table_type, ft->type);
+	MLX5_SET(destroy_flow_group_in, in, table_id, ft->id);
+	MLX5_SET(destroy_flow_group_in, in, group_id, group_id);
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					  sizeof(out));
+}
+
+static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev,
+			    int opmod, int modify_mask,
+			    struct mlx5_flow_table *ft,
+			    unsigned group_id,
+			    struct fs_fte *fte)
+{
+	unsigned int inlen = MLX5_ST_SZ_BYTES(set_fte_in) +
+		fte->dests_size * MLX5_ST_SZ_BYTES(dest_format_struct);
+	u32 out[MLX5_ST_SZ_DW(set_fte_out)];
+	struct mlx5_flow_rule *dst;
+	void *in_flow_context;
+	void *in_match_value;
+	void *in_dests;
+	u32 *in;
+	int err;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in) {
+		mlx5_core_warn(dev, "failed to allocate inbox\n");
+		return -ENOMEM;
+	}
+
+	MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY);
+	MLX5_SET(set_fte_in, in, op_mod, opmod);
+	MLX5_SET(set_fte_in, in, modify_enable_mask, modify_mask);
+	MLX5_SET(set_fte_in, in, table_type, ft->type);
+	MLX5_SET(set_fte_in, in, table_id,   ft->id);
+	MLX5_SET(set_fte_in, in, flow_index, fte->index);
+
+	in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context);
+	MLX5_SET(flow_context, in_flow_context, group_id, group_id);
+	MLX5_SET(flow_context, in_flow_context, flow_tag, fte->flow_tag);
+	MLX5_SET(flow_context, in_flow_context, action, fte->action);
+	MLX5_SET(flow_context, in_flow_context, destination_list_size,
+		 fte->dests_size);
+	in_match_value = MLX5_ADDR_OF(flow_context, in_flow_context,
+				      match_value);
+	memcpy(in_match_value, &fte->val, MLX5_ST_SZ_BYTES(fte_match_param));
+
+	in_dests = MLX5_ADDR_OF(flow_context, in_flow_context, destination);
+	list_for_each_entry(dst, &fte->node.children, node.list) {
+		unsigned int id;
+
+		MLX5_SET(dest_format_struct, in_dests, destination_type,
+			 dst->dest_attr.type);
+		if (dst->dest_attr.type ==
+		    MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE)
+			id = dst->dest_attr.ft->id;
+		else
+			id = dst->dest_attr.tir_num;
+		MLX5_SET(dest_format_struct, in_dests, destination_id, id);
+		in_dests += MLX5_ST_SZ_BYTES(dest_format_struct);
+	}
+	memset(out, 0, sizeof(out));
+	err = mlx5_cmd_exec_check_status(dev, in, inlen, out,
+					 sizeof(out));
+	kvfree(in);
+
+	return err;
+}
+
+int mlx5_cmd_create_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte)
+{
+	return	mlx5_cmd_set_fte(dev, 0, 0, ft, group_id, fte);
+}
+
+int mlx5_cmd_update_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte)
+{
+	int opmod;
+	int modify_mask;
+	int atomic_mod_cap = MLX5_CAP_FLOWTABLE(dev,
+						flow_table_properties_nic_receive.
+						flow_modify_en);
+	if (!atomic_mod_cap)
+		return -ENOTSUPP;
+	opmod = 1;
+	modify_mask = 1 <<
+		MLX5_SET_FTE_MODIFY_ENABLE_MASK_DESTINATION_LIST;
+
+	return	mlx5_cmd_set_fte(dev, opmod, modify_mask, ft, group_id, fte);
+}
+
+int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned int index)
+{
+	u32 out[MLX5_ST_SZ_DW(delete_fte_out)];
+	u32 in[MLX5_ST_SZ_DW(delete_fte_in)];
+	int err;
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(delete_fte_in, in, opcode, MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY);
+	MLX5_SET(delete_fte_in, in, table_type, ft->type);
+	MLX5_SET(delete_fte_in, in, table_id, ft->id);
+	MLX5_SET(delete_fte_in, in, flow_index, index);
+
+	err =  mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out));
+
+	return err;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
new file mode 100644
index 000000000..9814d4784
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
@@ -0,0 +1,72 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef _MLX5_FS_CMD_
+#define _MLX5_FS_CMD_
+
+int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
+			       enum fs_flow_table_type type, unsigned int level,
+			       unsigned int log_size, struct mlx5_flow_table
+			       *next_ft, unsigned int *table_id);
+
+int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft);
+
+int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev,
+			       struct mlx5_flow_table *ft,
+			       struct mlx5_flow_table *next_ft);
+
+int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev,
+			       struct mlx5_flow_table *ft,
+			       u32 *in, unsigned int *group_id);
+
+int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft,
+				unsigned int group_id);
+
+int mlx5_cmd_create_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte);
+
+int mlx5_cmd_update_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte);
+
+int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned int index);
+
+int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev,
+			    struct mlx5_flow_table *ft);
+#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
new file mode 100644
index 000000000..6f68dba8d
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -0,0 +1,1514 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/mutex.h>
+#include <linux/mlx5/driver.h>
+
+#include "mlx5_core.h"
+#include "fs_core.h"
+#include "fs_cmd.h"
+
+#define INIT_TREE_NODE_ARRAY_SIZE(...)	(sizeof((struct init_tree_node[]){__VA_ARGS__}) /\
+					 sizeof(struct init_tree_node))
+
+#define ADD_PRIO(num_prios_val, min_level_val, max_ft_val, caps_val,\
+		 ...) {.type = FS_TYPE_PRIO,\
+	.min_ft_level = min_level_val,\
+	.max_ft = max_ft_val,\
+	.num_leaf_prios = num_prios_val,\
+	.caps = caps_val,\
+	.children = (struct init_tree_node[]) {__VA_ARGS__},\
+	.ar_size = INIT_TREE_NODE_ARRAY_SIZE(__VA_ARGS__) \
+}
+
+#define ADD_MULTIPLE_PRIO(num_prios_val, max_ft_val, ...)\
+	ADD_PRIO(num_prios_val, 0, max_ft_val, {},\
+		 __VA_ARGS__)\
+
+#define ADD_NS(...) {.type = FS_TYPE_NAMESPACE,\
+	.children = (struct init_tree_node[]) {__VA_ARGS__},\
+	.ar_size = INIT_TREE_NODE_ARRAY_SIZE(__VA_ARGS__) \
+}
+
+#define INIT_CAPS_ARRAY_SIZE(...) (sizeof((long[]){__VA_ARGS__}) /\
+				   sizeof(long))
+
+#define FS_CAP(cap) (__mlx5_bit_off(flow_table_nic_cap, cap))
+
+#define FS_REQUIRED_CAPS(...) {.arr_sz = INIT_CAPS_ARRAY_SIZE(__VA_ARGS__), \
+			       .caps = (long[]) {__VA_ARGS__} }
+
+#define LEFTOVERS_MAX_FT 1
+#define LEFTOVERS_NUM_PRIOS 1
+#define BY_PASS_PRIO_MAX_FT 1
+#define BY_PASS_MIN_LEVEL (KENREL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\
+			   LEFTOVERS_MAX_FT)
+
+#define KERNEL_MAX_FT 2
+#define KERNEL_NUM_PRIOS 1
+#define KENREL_MIN_LEVEL 2
+
+struct node_caps {
+	size_t	arr_sz;
+	long	*caps;
+};
+static struct init_tree_node {
+	enum fs_node_type	type;
+	struct init_tree_node *children;
+	int ar_size;
+	struct node_caps caps;
+	int min_ft_level;
+	int num_leaf_prios;
+	int prio;
+	int max_ft;
+} root_fs = {
+	.type = FS_TYPE_NAMESPACE,
+	.ar_size = 3,
+	.children = (struct init_tree_node[]) {
+		ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0,
+			 FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en),
+					  FS_CAP(flow_table_properties_nic_receive.modify_root),
+					  FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode),
+					  FS_CAP(flow_table_properties_nic_receive.flow_table_modify)),
+			 ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_MAX_FT))),
+		ADD_PRIO(0, KENREL_MIN_LEVEL, 0, {},
+			 ADD_NS(ADD_MULTIPLE_PRIO(KERNEL_NUM_PRIOS, KERNEL_MAX_FT))),
+		ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0,
+			 FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en),
+					  FS_CAP(flow_table_properties_nic_receive.modify_root),
+					  FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode),
+					  FS_CAP(flow_table_properties_nic_receive.flow_table_modify)),
+			 ADD_NS(ADD_MULTIPLE_PRIO(LEFTOVERS_NUM_PRIOS, LEFTOVERS_MAX_FT))),
+	}
+};
+
+enum fs_i_mutex_lock_class {
+	FS_MUTEX_GRANDPARENT,
+	FS_MUTEX_PARENT,
+	FS_MUTEX_CHILD
+};
+
+static void del_rule(struct fs_node *node);
+static void del_flow_table(struct fs_node *node);
+static void del_flow_group(struct fs_node *node);
+static void del_fte(struct fs_node *node);
+
+static void tree_init_node(struct fs_node *node,
+			   unsigned int refcount,
+			   void (*remove_func)(struct fs_node *))
+{
+	atomic_set(&node->refcount, refcount);
+	INIT_LIST_HEAD(&node->list);
+	INIT_LIST_HEAD(&node->children);
+	mutex_init(&node->lock);
+	node->remove_func = remove_func;
+}
+
+static void tree_add_node(struct fs_node *node, struct fs_node *parent)
+{
+	if (parent)
+		atomic_inc(&parent->refcount);
+	node->parent = parent;
+
+	/* Parent is the root */
+	if (!parent)
+		node->root = node;
+	else
+		node->root = parent->root;
+}
+
+static void tree_get_node(struct fs_node *node)
+{
+	atomic_inc(&node->refcount);
+}
+
+static void nested_lock_ref_node(struct fs_node *node,
+				 enum fs_i_mutex_lock_class class)
+{
+	if (node) {
+		mutex_lock_nested(&node->lock, class);
+		atomic_inc(&node->refcount);
+	}
+}
+
+static void lock_ref_node(struct fs_node *node)
+{
+	if (node) {
+		mutex_lock(&node->lock);
+		atomic_inc(&node->refcount);
+	}
+}
+
+static void unlock_ref_node(struct fs_node *node)
+{
+	if (node) {
+		atomic_dec(&node->refcount);
+		mutex_unlock(&node->lock);
+	}
+}
+
+static void tree_put_node(struct fs_node *node)
+{
+	struct fs_node *parent_node = node->parent;
+
+	lock_ref_node(parent_node);
+	if (atomic_dec_and_test(&node->refcount)) {
+		if (parent_node)
+			list_del_init(&node->list);
+		if (node->remove_func)
+			node->remove_func(node);
+		kfree(node);
+		node = NULL;
+	}
+	unlock_ref_node(parent_node);
+	if (!node && parent_node)
+		tree_put_node(parent_node);
+}
+
+static int tree_remove_node(struct fs_node *node)
+{
+	if (atomic_read(&node->refcount) > 1)
+		return -EPERM;
+	tree_put_node(node);
+	return 0;
+}
+
+static struct fs_prio *find_prio(struct mlx5_flow_namespace *ns,
+				 unsigned int prio)
+{
+	struct fs_prio *iter_prio;
+
+	fs_for_each_prio(iter_prio, ns) {
+		if (iter_prio->prio == prio)
+			return iter_prio;
+	}
+
+	return NULL;
+}
+
+static unsigned int find_next_free_level(struct fs_prio *prio)
+{
+	if (!list_empty(&prio->node.children)) {
+		struct mlx5_flow_table *ft;
+
+		ft = list_last_entry(&prio->node.children,
+				     struct mlx5_flow_table,
+				     node.list);
+		return ft->level + 1;
+	}
+	return prio->start_level;
+}
+
+static bool masked_memcmp(void *mask, void *val1, void *val2, size_t size)
+{
+	unsigned int i;
+
+	for (i = 0; i < size; i++, mask++, val1++, val2++)
+		if ((*((u8 *)val1) & (*(u8 *)mask)) !=
+		    ((*(u8 *)val2) & (*(u8 *)mask)))
+			return false;
+
+	return true;
+}
+
+static bool compare_match_value(struct mlx5_flow_group_mask *mask,
+				void *fte_param1, void *fte_param2)
+{
+	if (mask->match_criteria_enable &
+	    1 << MLX5_CREATE_FLOW_GROUP_IN_MATCH_CRITERIA_ENABLE_OUTER_HEADERS) {
+		void *fte_match1 = MLX5_ADDR_OF(fte_match_param,
+						fte_param1, outer_headers);
+		void *fte_match2 = MLX5_ADDR_OF(fte_match_param,
+						fte_param2, outer_headers);
+		void *fte_mask = MLX5_ADDR_OF(fte_match_param,
+					      mask->match_criteria, outer_headers);
+
+		if (!masked_memcmp(fte_mask, fte_match1, fte_match2,
+				   MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4)))
+			return false;
+	}
+
+	if (mask->match_criteria_enable &
+	    1 << MLX5_CREATE_FLOW_GROUP_IN_MATCH_CRITERIA_ENABLE_MISC_PARAMETERS) {
+		void *fte_match1 = MLX5_ADDR_OF(fte_match_param,
+						fte_param1, misc_parameters);
+		void *fte_match2 = MLX5_ADDR_OF(fte_match_param,
+						fte_param2, misc_parameters);
+		void *fte_mask = MLX5_ADDR_OF(fte_match_param,
+					  mask->match_criteria, misc_parameters);
+
+		if (!masked_memcmp(fte_mask, fte_match1, fte_match2,
+				   MLX5_ST_SZ_BYTES(fte_match_set_misc)))
+			return false;
+	}
+
+	if (mask->match_criteria_enable &
+	    1 << MLX5_CREATE_FLOW_GROUP_IN_MATCH_CRITERIA_ENABLE_INNER_HEADERS) {
+		void *fte_match1 = MLX5_ADDR_OF(fte_match_param,
+						fte_param1, inner_headers);
+		void *fte_match2 = MLX5_ADDR_OF(fte_match_param,
+						fte_param2, inner_headers);
+		void *fte_mask = MLX5_ADDR_OF(fte_match_param,
+					  mask->match_criteria, inner_headers);
+
+		if (!masked_memcmp(fte_mask, fte_match1, fte_match2,
+				   MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4)))
+			return false;
+	}
+	return true;
+}
+
+static bool compare_match_criteria(u8 match_criteria_enable1,
+				   u8 match_criteria_enable2,
+				   void *mask1, void *mask2)
+{
+	return match_criteria_enable1 == match_criteria_enable2 &&
+		!memcmp(mask1, mask2, MLX5_ST_SZ_BYTES(fte_match_param));
+}
+
+static struct mlx5_flow_root_namespace *find_root(struct fs_node *node)
+{
+	struct fs_node *root;
+	struct mlx5_flow_namespace *ns;
+
+	root = node->root;
+
+	if (WARN_ON(root->type != FS_TYPE_NAMESPACE)) {
+		pr_warn("mlx5: flow steering node is not in tree or garbaged\n");
+		return NULL;
+	}
+
+	ns = container_of(root, struct mlx5_flow_namespace, node);
+	return container_of(ns, struct mlx5_flow_root_namespace, ns);
+}
+
+static inline struct mlx5_core_dev *get_dev(struct fs_node *node)
+{
+	struct mlx5_flow_root_namespace *root = find_root(node);
+
+	if (root)
+		return root->dev;
+	return NULL;
+}
+
+static void del_flow_table(struct fs_node *node)
+{
+	struct mlx5_flow_table *ft;
+	struct mlx5_core_dev *dev;
+	struct fs_prio *prio;
+	int err;
+
+	fs_get_obj(ft, node);
+	dev = get_dev(&ft->node);
+
+	err = mlx5_cmd_destroy_flow_table(dev, ft);
+	if (err)
+		pr_warn("flow steering can't destroy ft\n");
+	fs_get_obj(prio, ft->node.parent);
+	prio->num_ft--;
+}
+
+static void del_rule(struct fs_node *node)
+{
+	struct mlx5_flow_rule *rule;
+	struct mlx5_flow_table *ft;
+	struct mlx5_flow_group *fg;
+	struct fs_fte *fte;
+	u32	*match_value;
+	struct mlx5_core_dev *dev = get_dev(node);
+	int match_len = MLX5_ST_SZ_BYTES(fte_match_param);
+	int err;
+
+	match_value = mlx5_vzalloc(match_len);
+	if (!match_value) {
+		pr_warn("failed to allocate inbox\n");
+		return;
+	}
+
+	fs_get_obj(rule, node);
+	fs_get_obj(fte, rule->node.parent);
+	fs_get_obj(fg, fte->node.parent);
+	memcpy(match_value, fte->val, sizeof(fte->val));
+	fs_get_obj(ft, fg->node.parent);
+	list_del(&rule->node.list);
+	fte->dests_size--;
+	if (fte->dests_size) {
+		err = mlx5_cmd_update_fte(dev, ft,
+					  fg->id, fte);
+		if (err)
+			pr_warn("%s can't del rule fg id=%d fte_index=%d\n",
+				__func__, fg->id, fte->index);
+	}
+	kvfree(match_value);
+}
+
+static void del_fte(struct fs_node *node)
+{
+	struct mlx5_flow_table *ft;
+	struct mlx5_flow_group *fg;
+	struct mlx5_core_dev *dev;
+	struct fs_fte *fte;
+	int err;
+
+	fs_get_obj(fte, node);
+	fs_get_obj(fg, fte->node.parent);
+	fs_get_obj(ft, fg->node.parent);
+
+	dev = get_dev(&ft->node);
+	err = mlx5_cmd_delete_fte(dev, ft,
+				  fte->index);
+	if (err)
+		pr_warn("flow steering can't delete fte in index %d of flow group id %d\n",
+			fte->index, fg->id);
+
+	fte->status = 0;
+	fg->num_ftes--;
+}
+
+static void del_flow_group(struct fs_node *node)
+{
+	struct mlx5_flow_group *fg;
+	struct mlx5_flow_table *ft;
+	struct mlx5_core_dev *dev;
+
+	fs_get_obj(fg, node);
+	fs_get_obj(ft, fg->node.parent);
+	dev = get_dev(&ft->node);
+
+	if (mlx5_cmd_destroy_flow_group(dev, ft, fg->id))
+		pr_warn("flow steering can't destroy fg %d of ft %d\n",
+			fg->id, ft->id);
+}
+
+static struct fs_fte *alloc_fte(u8 action,
+				u32 flow_tag,
+				u32 *match_value,
+				unsigned int index)
+{
+	struct fs_fte *fte;
+
+	fte = kzalloc(sizeof(*fte), GFP_KERNEL);
+	if (!fte)
+		return ERR_PTR(-ENOMEM);
+
+	memcpy(fte->val, match_value, sizeof(fte->val));
+	fte->node.type =  FS_TYPE_FLOW_ENTRY;
+	fte->flow_tag = flow_tag;
+	fte->index = index;
+	fte->action = action;
+
+	return fte;
+}
+
+static struct mlx5_flow_group *alloc_flow_group(u32 *create_fg_in)
+{
+	struct mlx5_flow_group *fg;
+	void *match_criteria = MLX5_ADDR_OF(create_flow_group_in,
+					    create_fg_in, match_criteria);
+	u8 match_criteria_enable = MLX5_GET(create_flow_group_in,
+					    create_fg_in,
+					    match_criteria_enable);
+	fg = kzalloc(sizeof(*fg), GFP_KERNEL);
+	if (!fg)
+		return ERR_PTR(-ENOMEM);
+
+	fg->mask.match_criteria_enable = match_criteria_enable;
+	memcpy(&fg->mask.match_criteria, match_criteria,
+	       sizeof(fg->mask.match_criteria));
+	fg->node.type =  FS_TYPE_FLOW_GROUP;
+	fg->start_index = MLX5_GET(create_flow_group_in, create_fg_in,
+				   start_flow_index);
+	fg->max_ftes = MLX5_GET(create_flow_group_in, create_fg_in,
+				end_flow_index) - fg->start_index + 1;
+	return fg;
+}
+
+static struct mlx5_flow_table *alloc_flow_table(int level, int max_fte,
+						enum fs_flow_table_type table_type)
+{
+	struct mlx5_flow_table *ft;
+
+	ft  = kzalloc(sizeof(*ft), GFP_KERNEL);
+	if (!ft)
+		return NULL;
+
+	ft->level = level;
+	ft->node.type = FS_TYPE_FLOW_TABLE;
+	ft->type = table_type;
+	ft->max_fte = max_fte;
+
+	return ft;
+}
+
+/* If reverse is false, then we search for the first flow table in the
+ * root sub-tree from start(closest from right), else we search for the
+ * last flow table in the root sub-tree till start(closest from left).
+ */
+static struct mlx5_flow_table *find_closest_ft_recursive(struct fs_node  *root,
+							 struct list_head *start,
+							 bool reverse)
+{
+#define list_advance_entry(pos, reverse)		\
+	((reverse) ? list_prev_entry(pos, list) : list_next_entry(pos, list))
+
+#define list_for_each_advance_continue(pos, head, reverse)	\
+	for (pos = list_advance_entry(pos, reverse);		\
+	     &pos->list != (head);				\
+	     pos = list_advance_entry(pos, reverse))
+
+	struct fs_node *iter = list_entry(start, struct fs_node, list);
+	struct mlx5_flow_table *ft = NULL;
+
+	if (!root)
+		return NULL;
+
+	list_for_each_advance_continue(iter, &root->children, reverse) {
+		if (iter->type == FS_TYPE_FLOW_TABLE) {
+			fs_get_obj(ft, iter);
+			return ft;
+		}
+		ft = find_closest_ft_recursive(iter, &iter->children, reverse);
+		if (ft)
+			return ft;
+	}
+
+	return ft;
+}
+
+/* If reverse if false then return the first flow table in next priority of
+ * prio in the tree, else return the last flow table in the previous priority
+ * of prio in the tree.
+ */
+static struct mlx5_flow_table *find_closest_ft(struct fs_prio *prio, bool reverse)
+{
+	struct mlx5_flow_table *ft = NULL;
+	struct fs_node *curr_node;
+	struct fs_node *parent;
+
+	parent = prio->node.parent;
+	curr_node = &prio->node;
+	while (!ft && parent) {
+		ft = find_closest_ft_recursive(parent, &curr_node->list, reverse);
+		curr_node = parent;
+		parent = curr_node->parent;
+	}
+	return ft;
+}
+
+/* Assuming all the tree is locked by mutex chain lock */
+static struct mlx5_flow_table *find_next_chained_ft(struct fs_prio *prio)
+{
+	return find_closest_ft(prio, false);
+}
+
+/* Assuming all the tree is locked by mutex chain lock */
+static struct mlx5_flow_table *find_prev_chained_ft(struct fs_prio *prio)
+{
+	return find_closest_ft(prio, true);
+}
+
+static int connect_fts_in_prio(struct mlx5_core_dev *dev,
+			       struct fs_prio *prio,
+			       struct mlx5_flow_table *ft)
+{
+	struct mlx5_flow_table *iter;
+	int i = 0;
+	int err;
+
+	fs_for_each_ft(iter, prio) {
+		i++;
+		err = mlx5_cmd_modify_flow_table(dev,
+						 iter,
+						 ft);
+		if (err) {
+			mlx5_core_warn(dev, "Failed to modify flow table %d\n",
+				       iter->id);
+			/* The driver is out of sync with the FW */
+			if (i > 1)
+				WARN_ON(true);
+			return err;
+		}
+	}
+	return 0;
+}
+
+/* Connect flow tables from previous priority of prio to ft */
+static int connect_prev_fts(struct mlx5_core_dev *dev,
+			    struct mlx5_flow_table *ft,
+			    struct fs_prio *prio)
+{
+	struct mlx5_flow_table *prev_ft;
+
+	prev_ft = find_prev_chained_ft(prio);
+	if (prev_ft) {
+		struct fs_prio *prev_prio;
+
+		fs_get_obj(prev_prio, prev_ft->node.parent);
+		return connect_fts_in_prio(dev, prev_prio, ft);
+	}
+	return 0;
+}
+
+static int update_root_ft_create(struct mlx5_flow_table *ft, struct fs_prio
+				 *prio)
+{
+	struct mlx5_flow_root_namespace *root = find_root(&prio->node);
+	int min_level = INT_MAX;
+	int err;
+
+	if (root->root_ft)
+		min_level = root->root_ft->level;
+
+	if (ft->level >= min_level)
+		return 0;
+
+	err = mlx5_cmd_update_root_ft(root->dev, ft);
+	if (err)
+		mlx5_core_warn(root->dev, "Update root flow table of id=%u failed\n",
+			       ft->id);
+	else
+		root->root_ft = ft;
+
+	return err;
+}
+
+static int connect_flow_table(struct mlx5_core_dev *dev, struct mlx5_flow_table *ft,
+			      struct fs_prio *prio)
+{
+	int err = 0;
+
+	/* Connect_prev_fts and update_root_ft_create are mutually exclusive */
+
+	if (list_empty(&prio->node.children)) {
+		err = connect_prev_fts(dev, ft, prio);
+		if (err)
+			return err;
+	}
+
+	if (MLX5_CAP_FLOWTABLE(dev,
+			       flow_table_properties_nic_receive.modify_root))
+		err = update_root_ft_create(ft, prio);
+	return err;
+}
+
+struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns,
+					       int prio,
+					       int max_fte)
+{
+	struct mlx5_flow_table *next_ft = NULL;
+	struct mlx5_flow_table *ft;
+	int err;
+	int log_table_sz;
+	struct mlx5_flow_root_namespace *root =
+		find_root(&ns->node);
+	struct fs_prio *fs_prio = NULL;
+
+	if (!root) {
+		pr_err("mlx5: flow steering failed to find root of namespace\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	mutex_lock(&root->chain_lock);
+	fs_prio = find_prio(ns, prio);
+	if (!fs_prio) {
+		err = -EINVAL;
+		goto unlock_root;
+	}
+	if (fs_prio->num_ft == fs_prio->max_ft) {
+		err = -ENOSPC;
+		goto unlock_root;
+	}
+
+	ft = alloc_flow_table(find_next_free_level(fs_prio),
+			      roundup_pow_of_two(max_fte),
+			      root->table_type);
+	if (!ft) {
+		err = -ENOMEM;
+		goto unlock_root;
+	}
+
+	tree_init_node(&ft->node, 1, del_flow_table);
+	log_table_sz = ilog2(ft->max_fte);
+	next_ft = find_next_chained_ft(fs_prio);
+	err = mlx5_cmd_create_flow_table(root->dev, ft->type, ft->level,
+					 log_table_sz, next_ft, &ft->id);
+	if (err)
+		goto free_ft;
+
+	err = connect_flow_table(root->dev, ft, fs_prio);
+	if (err)
+		goto destroy_ft;
+	lock_ref_node(&fs_prio->node);
+	tree_add_node(&ft->node, &fs_prio->node);
+	list_add_tail(&ft->node.list, &fs_prio->node.children);
+	fs_prio->num_ft++;
+	unlock_ref_node(&fs_prio->node);
+	mutex_unlock(&root->chain_lock);
+	return ft;
+destroy_ft:
+	mlx5_cmd_destroy_flow_table(root->dev, ft);
+free_ft:
+	kfree(ft);
+unlock_root:
+	mutex_unlock(&root->chain_lock);
+	return ERR_PTR(err);
+}
+
+struct mlx5_flow_table *mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
+							    int prio,
+							    int num_flow_table_entries,
+							    int max_num_groups)
+{
+	struct mlx5_flow_table *ft;
+
+	if (max_num_groups > num_flow_table_entries)
+		return ERR_PTR(-EINVAL);
+
+	ft = mlx5_create_flow_table(ns, prio, num_flow_table_entries);
+	if (IS_ERR(ft))
+		return ft;
+
+	ft->autogroup.active = true;
+	ft->autogroup.required_groups = max_num_groups;
+
+	return ft;
+}
+EXPORT_SYMBOL(mlx5_create_auto_grouped_flow_table);
+
+/* Flow table should be locked */
+static struct mlx5_flow_group *create_flow_group_common(struct mlx5_flow_table *ft,
+							u32 *fg_in,
+							struct list_head
+							*prev_fg,
+							bool is_auto_fg)
+{
+	struct mlx5_flow_group *fg;
+	struct mlx5_core_dev *dev = get_dev(&ft->node);
+	int err;
+
+	if (!dev)
+		return ERR_PTR(-ENODEV);
+
+	fg = alloc_flow_group(fg_in);
+	if (IS_ERR(fg))
+		return fg;
+
+	err = mlx5_cmd_create_flow_group(dev, ft, fg_in, &fg->id);
+	if (err) {
+		kfree(fg);
+		return ERR_PTR(err);
+	}
+
+	if (ft->autogroup.active)
+		ft->autogroup.num_groups++;
+	/* Add node to tree */
+	tree_init_node(&fg->node, !is_auto_fg, del_flow_group);
+	tree_add_node(&fg->node, &ft->node);
+	/* Add node to group list */
+	list_add(&fg->node.list, ft->node.children.prev);
+
+	return fg;
+}
+
+struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft,
+					       u32 *fg_in)
+{
+	struct mlx5_flow_group *fg;
+
+	if (ft->autogroup.active)
+		return ERR_PTR(-EPERM);
+
+	lock_ref_node(&ft->node);
+	fg = create_flow_group_common(ft, fg_in, &ft->node.children, false);
+	unlock_ref_node(&ft->node);
+
+	return fg;
+}
+
+static struct mlx5_flow_rule *alloc_rule(struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_rule *rule;
+
+	rule = kzalloc(sizeof(*rule), GFP_KERNEL);
+	if (!rule)
+		return NULL;
+
+	rule->node.type = FS_TYPE_FLOW_DEST;
+	memcpy(&rule->dest_attr, dest, sizeof(*dest));
+
+	return rule;
+}
+
+/* fte should not be deleted while calling this function */
+static struct mlx5_flow_rule *add_rule_fte(struct fs_fte *fte,
+					   struct mlx5_flow_group *fg,
+					   struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_table *ft;
+	struct mlx5_flow_rule *rule;
+	int err;
+
+	rule = alloc_rule(dest);
+	if (!rule)
+		return ERR_PTR(-ENOMEM);
+
+	fs_get_obj(ft, fg->node.parent);
+	/* Add dest to dests list- added as first element after the head */
+	tree_init_node(&rule->node, 1, del_rule);
+	list_add_tail(&rule->node.list, &fte->node.children);
+	fte->dests_size++;
+	if (fte->dests_size == 1)
+		err = mlx5_cmd_create_fte(get_dev(&ft->node),
+					  ft, fg->id, fte);
+	else
+		err = mlx5_cmd_update_fte(get_dev(&ft->node),
+					  ft, fg->id, fte);
+	if (err)
+		goto free_rule;
+
+	fte->status |= FS_FTE_STATUS_EXISTING;
+
+	return rule;
+
+free_rule:
+	list_del(&rule->node.list);
+	kfree(rule);
+	fte->dests_size--;
+	return ERR_PTR(err);
+}
+
+/* Assumed fg is locked */
+static unsigned int get_free_fte_index(struct mlx5_flow_group *fg,
+				       struct list_head **prev)
+{
+	struct fs_fte *fte;
+	unsigned int start = fg->start_index;
+
+	if (prev)
+		*prev = &fg->node.children;
+
+	/* assumed list is sorted by index */
+	fs_for_each_fte(fte, fg) {
+		if (fte->index != start)
+			return start;
+		start++;
+		if (prev)
+			*prev = &fte->node.list;
+	}
+
+	return start;
+}
+
+/* prev is output, prev->next = new_fte */
+static struct fs_fte *create_fte(struct mlx5_flow_group *fg,
+				 u32 *match_value,
+				 u8 action,
+				 u32 flow_tag,
+				 struct list_head **prev)
+{
+	struct fs_fte *fte;
+	int index;
+
+	index = get_free_fte_index(fg, prev);
+	fte = alloc_fte(action, flow_tag, match_value, index);
+	if (IS_ERR(fte))
+		return fte;
+
+	return fte;
+}
+
+static struct mlx5_flow_group *create_autogroup(struct mlx5_flow_table *ft,
+						u8 match_criteria_enable,
+						u32 *match_criteria)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct list_head *prev = &ft->node.children;
+	unsigned int candidate_index = 0;
+	struct mlx5_flow_group *fg;
+	void *match_criteria_addr;
+	unsigned int group_size = 0;
+	u32 *in;
+
+	if (!ft->autogroup.active)
+		return ERR_PTR(-ENOENT);
+
+	in = mlx5_vzalloc(inlen);
+	if (!in)
+		return ERR_PTR(-ENOMEM);
+
+	if (ft->autogroup.num_groups < ft->autogroup.required_groups)
+		/* We save place for flow groups in addition to max types */
+		group_size = ft->max_fte / (ft->autogroup.required_groups + 1);
+
+	/*  ft->max_fte == ft->autogroup.max_types */
+	if (group_size == 0)
+		group_size = 1;
+
+	/* sorted by start_index */
+	fs_for_each_fg(fg, ft) {
+		if (candidate_index + group_size > fg->start_index)
+			candidate_index = fg->start_index + fg->max_ftes;
+		else
+			break;
+		prev = &fg->node.list;
+	}
+
+	if (candidate_index + group_size > ft->max_fte) {
+		fg = ERR_PTR(-ENOSPC);
+		goto out;
+	}
+
+	MLX5_SET(create_flow_group_in, in, match_criteria_enable,
+		 match_criteria_enable);
+	MLX5_SET(create_flow_group_in, in, start_flow_index, candidate_index);
+	MLX5_SET(create_flow_group_in, in, end_flow_index,   candidate_index +
+		 group_size - 1);
+	match_criteria_addr = MLX5_ADDR_OF(create_flow_group_in,
+					   in, match_criteria);
+	memcpy(match_criteria_addr, match_criteria,
+	       MLX5_ST_SZ_BYTES(fte_match_param));
+
+	fg = create_flow_group_common(ft, in, prev, true);
+out:
+	kvfree(in);
+	return fg;
+}
+
+static struct mlx5_flow_rule *add_rule_fg(struct mlx5_flow_group *fg,
+					  u32 *match_value,
+					  u8 action,
+					  u32 flow_tag,
+					  struct mlx5_flow_destination *dest)
+{
+	struct fs_fte *fte;
+	struct mlx5_flow_rule *rule;
+	struct mlx5_flow_table *ft;
+	struct list_head *prev;
+
+	nested_lock_ref_node(&fg->node, FS_MUTEX_PARENT);
+	fs_for_each_fte(fte, fg) {
+		nested_lock_ref_node(&fte->node, FS_MUTEX_CHILD);
+		if (compare_match_value(&fg->mask, match_value, &fte->val) &&
+		    action == fte->action && flow_tag == fte->flow_tag) {
+			rule = add_rule_fte(fte, fg, dest);
+			unlock_ref_node(&fte->node);
+			if (IS_ERR(rule))
+				goto unlock_fg;
+			else
+				goto add_rule;
+		}
+		unlock_ref_node(&fte->node);
+	}
+	fs_get_obj(ft, fg->node.parent);
+	if (fg->num_ftes >= fg->max_ftes) {
+		rule = ERR_PTR(-ENOSPC);
+		goto unlock_fg;
+	}
+
+	fte = create_fte(fg, match_value, action, flow_tag, &prev);
+	if (IS_ERR(fte)) {
+		rule = (void *)fte;
+		goto unlock_fg;
+	}
+	tree_init_node(&fte->node, 0, del_fte);
+	rule = add_rule_fte(fte, fg, dest);
+	if (IS_ERR(rule)) {
+		kfree(fte);
+		goto unlock_fg;
+	}
+
+	fg->num_ftes++;
+
+	tree_add_node(&fte->node, &fg->node);
+	list_add(&fte->node.list, prev);
+add_rule:
+	tree_add_node(&rule->node, &fte->node);
+unlock_fg:
+	unlock_ref_node(&fg->node);
+	return rule;
+}
+
+static struct mlx5_flow_rule *add_rule_to_auto_fg(struct mlx5_flow_table *ft,
+						  u8 match_criteria_enable,
+						  u32 *match_criteria,
+						  u32 *match_value,
+						  u8 action,
+						  u32 flow_tag,
+						  struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_rule *rule;
+	struct mlx5_flow_group *g;
+
+	g = create_autogroup(ft, match_criteria_enable, match_criteria);
+	if (IS_ERR(g))
+		return (void *)g;
+
+	rule = add_rule_fg(g, match_value,
+			   action, flow_tag, dest);
+	if (IS_ERR(rule)) {
+		/* Remove assumes refcount > 0 and autogroup creates a group
+		 * with a refcount = 0.
+		 */
+		tree_get_node(&g->node);
+		tree_remove_node(&g->node);
+	}
+	return rule;
+}
+
+struct mlx5_flow_rule *
+mlx5_add_flow_rule(struct mlx5_flow_table *ft,
+		   u8 match_criteria_enable,
+		   u32 *match_criteria,
+		   u32 *match_value,
+		   u32 action,
+		   u32 flow_tag,
+		   struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_group *g;
+	struct mlx5_flow_rule *rule;
+
+	nested_lock_ref_node(&ft->node, FS_MUTEX_GRANDPARENT);
+	fs_for_each_fg(g, ft)
+		if (compare_match_criteria(g->mask.match_criteria_enable,
+					   match_criteria_enable,
+					   g->mask.match_criteria,
+					   match_criteria)) {
+			rule = add_rule_fg(g, match_value,
+					   action, flow_tag, dest);
+			if (!IS_ERR(rule) || PTR_ERR(rule) != -ENOSPC)
+				goto unlock;
+		}
+
+	rule = add_rule_to_auto_fg(ft, match_criteria_enable, match_criteria,
+				   match_value, action, flow_tag, dest);
+unlock:
+	unlock_ref_node(&ft->node);
+	return rule;
+}
+EXPORT_SYMBOL(mlx5_add_flow_rule);
+
+void mlx5_del_flow_rule(struct mlx5_flow_rule *rule)
+{
+	tree_remove_node(&rule->node);
+}
+EXPORT_SYMBOL(mlx5_del_flow_rule);
+
+/* Assuming prio->node.children(flow tables) is sorted by level */
+static struct mlx5_flow_table *find_next_ft(struct mlx5_flow_table *ft)
+{
+	struct fs_prio *prio;
+
+	fs_get_obj(prio, ft->node.parent);
+
+	if (!list_is_last(&ft->node.list, &prio->node.children))
+		return list_next_entry(ft, node.list);
+	return find_next_chained_ft(prio);
+}
+
+static int update_root_ft_destroy(struct mlx5_flow_table *ft)
+{
+	struct mlx5_flow_root_namespace *root = find_root(&ft->node);
+	struct mlx5_flow_table *new_root_ft = NULL;
+
+	if (root->root_ft != ft)
+		return 0;
+
+	new_root_ft = find_next_ft(ft);
+	if (new_root_ft) {
+		int err = mlx5_cmd_update_root_ft(root->dev, new_root_ft);
+
+		if (err) {
+			mlx5_core_warn(root->dev, "Update root flow table of id=%u failed\n",
+				       ft->id);
+			return err;
+		}
+		root->root_ft = new_root_ft;
+	}
+	return 0;
+}
+
+/* Connect flow table from previous priority to
+ * the next flow table.
+ */
+static int disconnect_flow_table(struct mlx5_flow_table *ft)
+{
+	struct mlx5_core_dev *dev = get_dev(&ft->node);
+	struct mlx5_flow_table *next_ft;
+	struct fs_prio *prio;
+	int err = 0;
+
+	err = update_root_ft_destroy(ft);
+	if (err)
+		return err;
+
+	fs_get_obj(prio, ft->node.parent);
+	if  (!(list_first_entry(&prio->node.children,
+				struct mlx5_flow_table,
+				node.list) == ft))
+		return 0;
+
+	next_ft = find_next_chained_ft(prio);
+	err = connect_prev_fts(dev, next_ft, prio);
+	if (err)
+		mlx5_core_warn(dev, "Failed to disconnect flow table %d\n",
+			       ft->id);
+	return err;
+}
+
+int mlx5_destroy_flow_table(struct mlx5_flow_table *ft)
+{
+	struct mlx5_flow_root_namespace *root = find_root(&ft->node);
+	int err = 0;
+
+	mutex_lock(&root->chain_lock);
+	err = disconnect_flow_table(ft);
+	if (err) {
+		mutex_unlock(&root->chain_lock);
+		return err;
+	}
+	if (tree_remove_node(&ft->node))
+		mlx5_core_warn(get_dev(&ft->node), "Flow table %d wasn't destroyed, refcount > 1\n",
+			       ft->id);
+	mutex_unlock(&root->chain_lock);
+
+	return err;
+}
+EXPORT_SYMBOL(mlx5_destroy_flow_table);
+
+void mlx5_destroy_flow_group(struct mlx5_flow_group *fg)
+{
+	if (tree_remove_node(&fg->node))
+		mlx5_core_warn(get_dev(&fg->node), "Flow group %d wasn't destroyed, refcount > 1\n",
+			       fg->id);
+}
+
+struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
+						    enum mlx5_flow_namespace_type type)
+{
+	struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns;
+	int prio;
+	static struct fs_prio *fs_prio;
+	struct mlx5_flow_namespace *ns;
+
+	if (!root_ns)
+		return NULL;
+
+	switch (type) {
+	case MLX5_FLOW_NAMESPACE_BYPASS:
+	case MLX5_FLOW_NAMESPACE_KERNEL:
+	case MLX5_FLOW_NAMESPACE_LEFTOVERS:
+		prio = type;
+		break;
+	case MLX5_FLOW_NAMESPACE_FDB:
+		if (dev->priv.fdb_root_ns)
+			return &dev->priv.fdb_root_ns->ns;
+		else
+			return NULL;
+	default:
+		return NULL;
+	}
+
+	fs_prio = find_prio(&root_ns->ns, prio);
+	if (!fs_prio)
+		return NULL;
+
+	ns = list_first_entry(&fs_prio->node.children,
+			      typeof(*ns),
+			      node.list);
+
+	return ns;
+}
+EXPORT_SYMBOL(mlx5_get_flow_namespace);
+
+static struct fs_prio *fs_create_prio(struct mlx5_flow_namespace *ns,
+				      unsigned prio, int max_ft)
+{
+	struct fs_prio *fs_prio;
+
+	fs_prio = kzalloc(sizeof(*fs_prio), GFP_KERNEL);
+	if (!fs_prio)
+		return ERR_PTR(-ENOMEM);
+
+	fs_prio->node.type = FS_TYPE_PRIO;
+	tree_init_node(&fs_prio->node, 1, NULL);
+	tree_add_node(&fs_prio->node, &ns->node);
+	fs_prio->max_ft = max_ft;
+	fs_prio->prio = prio;
+	list_add_tail(&fs_prio->node.list, &ns->node.children);
+
+	return fs_prio;
+}
+
+static struct mlx5_flow_namespace *fs_init_namespace(struct mlx5_flow_namespace
+						     *ns)
+{
+	ns->node.type = FS_TYPE_NAMESPACE;
+
+	return ns;
+}
+
+static struct mlx5_flow_namespace *fs_create_namespace(struct fs_prio *prio)
+{
+	struct mlx5_flow_namespace	*ns;
+
+	ns = kzalloc(sizeof(*ns), GFP_KERNEL);
+	if (!ns)
+		return ERR_PTR(-ENOMEM);
+
+	fs_init_namespace(ns);
+	tree_init_node(&ns->node, 1, NULL);
+	tree_add_node(&ns->node, &prio->node);
+	list_add_tail(&ns->node.list, &prio->node.children);
+
+	return ns;
+}
+
+static int create_leaf_prios(struct mlx5_flow_namespace *ns, struct init_tree_node
+			     *prio_metadata)
+{
+	struct fs_prio *fs_prio;
+	int i;
+
+	for (i = 0; i < prio_metadata->num_leaf_prios; i++) {
+		fs_prio = fs_create_prio(ns, i, prio_metadata->max_ft);
+		if (IS_ERR(fs_prio))
+			return PTR_ERR(fs_prio);
+	}
+	return 0;
+}
+
+#define FLOW_TABLE_BIT_SZ 1
+#define GET_FLOW_TABLE_CAP(dev, offset) \
+	((be32_to_cpu(*((__be32 *)(dev->hca_caps_cur[MLX5_CAP_FLOW_TABLE]) +	\
+			offset / 32)) >>					\
+	  (32 - FLOW_TABLE_BIT_SZ - (offset & 0x1f))) & FLOW_TABLE_BIT_SZ)
+static bool has_required_caps(struct mlx5_core_dev *dev, struct node_caps *caps)
+{
+	int i;
+
+	for (i = 0; i < caps->arr_sz; i++) {
+		if (!GET_FLOW_TABLE_CAP(dev, caps->caps[i]))
+			return false;
+	}
+	return true;
+}
+
+static int init_root_tree_recursive(struct mlx5_core_dev *dev,
+				    struct init_tree_node *init_node,
+				    struct fs_node *fs_parent_node,
+				    struct init_tree_node *init_parent_node,
+				    int index)
+{
+	int max_ft_level = MLX5_CAP_FLOWTABLE(dev,
+					      flow_table_properties_nic_receive.
+					      max_ft_level);
+	struct mlx5_flow_namespace *fs_ns;
+	struct fs_prio *fs_prio;
+	struct fs_node *base;
+	int i;
+	int err;
+
+	if (init_node->type == FS_TYPE_PRIO) {
+		if ((init_node->min_ft_level > max_ft_level) ||
+		    !has_required_caps(dev, &init_node->caps))
+			return 0;
+
+		fs_get_obj(fs_ns, fs_parent_node);
+		if (init_node->num_leaf_prios)
+			return create_leaf_prios(fs_ns, init_node);
+		fs_prio = fs_create_prio(fs_ns, index, init_node->max_ft);
+		if (IS_ERR(fs_prio))
+			return PTR_ERR(fs_prio);
+		base = &fs_prio->node;
+	} else if (init_node->type == FS_TYPE_NAMESPACE) {
+		fs_get_obj(fs_prio, fs_parent_node);
+		fs_ns = fs_create_namespace(fs_prio);
+		if (IS_ERR(fs_ns))
+			return PTR_ERR(fs_ns);
+		base = &fs_ns->node;
+	} else {
+		return -EINVAL;
+	}
+	for (i = 0; i < init_node->ar_size; i++) {
+		err = init_root_tree_recursive(dev, &init_node->children[i],
+					       base, init_node, i);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int init_root_tree(struct mlx5_core_dev *dev,
+			  struct init_tree_node *init_node,
+			  struct fs_node *fs_parent_node)
+{
+	int i;
+	struct mlx5_flow_namespace *fs_ns;
+	int err;
+
+	fs_get_obj(fs_ns, fs_parent_node);
+	for (i = 0; i < init_node->ar_size; i++) {
+		err = init_root_tree_recursive(dev, &init_node->children[i],
+					       &fs_ns->node,
+					       init_node, i);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev,
+						       enum fs_flow_table_type
+						       table_type)
+{
+	struct mlx5_flow_root_namespace *root_ns;
+	struct mlx5_flow_namespace *ns;
+
+	/* Create the root namespace */
+	root_ns = mlx5_vzalloc(sizeof(*root_ns));
+	if (!root_ns)
+		return NULL;
+
+	root_ns->dev = dev;
+	root_ns->table_type = table_type;
+
+	ns = &root_ns->ns;
+	fs_init_namespace(ns);
+	mutex_init(&root_ns->chain_lock);
+	tree_init_node(&ns->node, 1, NULL);
+	tree_add_node(&ns->node, NULL);
+
+	return root_ns;
+}
+
+static void set_prio_attrs_in_prio(struct fs_prio *prio, int acc_level);
+
+static int set_prio_attrs_in_ns(struct mlx5_flow_namespace *ns, int acc_level)
+{
+	struct fs_prio *prio;
+
+	fs_for_each_prio(prio, ns) {
+		 /* This updates prio start_level and max_ft */
+		set_prio_attrs_in_prio(prio, acc_level);
+		acc_level += prio->max_ft;
+	}
+	return acc_level;
+}
+
+static void set_prio_attrs_in_prio(struct fs_prio *prio, int acc_level)
+{
+	struct mlx5_flow_namespace *ns;
+	int acc_level_ns = acc_level;
+
+	prio->start_level = acc_level;
+	fs_for_each_ns(ns, prio)
+		/* This updates start_level and max_ft of ns's priority descendants */
+		acc_level_ns = set_prio_attrs_in_ns(ns, acc_level);
+	if (!prio->max_ft)
+		prio->max_ft = acc_level_ns - prio->start_level;
+	WARN_ON(prio->max_ft < acc_level_ns - prio->start_level);
+}
+
+static void set_prio_attrs(struct mlx5_flow_root_namespace *root_ns)
+{
+	struct mlx5_flow_namespace *ns = &root_ns->ns;
+	struct fs_prio *prio;
+	int start_level = 0;
+
+	fs_for_each_prio(prio, ns) {
+		set_prio_attrs_in_prio(prio, start_level);
+		start_level += prio->max_ft;
+	}
+}
+
+static int init_root_ns(struct mlx5_core_dev *dev)
+{
+
+	dev->priv.root_ns = create_root_ns(dev, FS_FT_NIC_RX);
+	if (IS_ERR_OR_NULL(dev->priv.root_ns))
+		goto cleanup;
+
+	if (init_root_tree(dev, &root_fs, &dev->priv.root_ns->ns.node))
+		goto cleanup;
+
+	set_prio_attrs(dev->priv.root_ns);
+
+	return 0;
+
+cleanup:
+	mlx5_cleanup_fs(dev);
+	return -ENOMEM;
+}
+
+static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev,
+					struct mlx5_flow_root_namespace *root_ns)
+{
+	struct fs_node *prio;
+
+	if (!root_ns)
+		return;
+
+	if (!list_empty(&root_ns->ns.node.children)) {
+		prio = list_first_entry(&root_ns->ns.node.children,
+					struct fs_node,
+				 list);
+		if (tree_remove_node(prio))
+			mlx5_core_warn(dev,
+				       "Flow steering priority wasn't destroyed, refcount > 1\n");
+	}
+	if (tree_remove_node(&root_ns->ns.node))
+		mlx5_core_warn(dev,
+			       "Flow steering namespace wasn't destroyed, refcount > 1\n");
+	root_ns = NULL;
+}
+
+static void cleanup_root_ns(struct mlx5_core_dev *dev)
+{
+	struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns;
+	struct fs_prio *iter_prio;
+
+	if (!MLX5_CAP_GEN(dev, nic_flow_table))
+		return;
+
+	if (!root_ns)
+		return;
+
+	/* stage 1 */
+	fs_for_each_prio(iter_prio, &root_ns->ns) {
+		struct fs_node *node;
+		struct mlx5_flow_namespace *iter_ns;
+
+		fs_for_each_ns_or_ft(node, iter_prio) {
+			if (node->type == FS_TYPE_FLOW_TABLE)
+				continue;
+			fs_get_obj(iter_ns, node);
+			while (!list_empty(&iter_ns->node.children)) {
+				struct fs_prio *obj_iter_prio2;
+				struct fs_node *iter_prio2 =
+					list_first_entry(&iter_ns->node.children,
+							 struct fs_node,
+							 list);
+
+				fs_get_obj(obj_iter_prio2, iter_prio2);
+				if (tree_remove_node(iter_prio2)) {
+					mlx5_core_warn(dev,
+						       "Priority %d wasn't destroyed, refcount > 1\n",
+						       obj_iter_prio2->prio);
+					return;
+				}
+			}
+		}
+	}
+
+	/* stage 2 */
+	fs_for_each_prio(iter_prio, &root_ns->ns) {
+		while (!list_empty(&iter_prio->node.children)) {
+			struct fs_node *iter_ns =
+				list_first_entry(&iter_prio->node.children,
+						 struct fs_node,
+						 list);
+			if (tree_remove_node(iter_ns)) {
+				mlx5_core_warn(dev,
+					       "Namespace wasn't destroyed, refcount > 1\n");
+				return;
+			}
+		}
+	}
+
+	/* stage 3 */
+	while (!list_empty(&root_ns->ns.node.children)) {
+		struct fs_prio *obj_prio_node;
+		struct fs_node *prio_node =
+			list_first_entry(&root_ns->ns.node.children,
+					 struct fs_node,
+					 list);
+
+		fs_get_obj(obj_prio_node, prio_node);
+		if (tree_remove_node(prio_node)) {
+			mlx5_core_warn(dev,
+				       "Priority %d wasn't destroyed, refcount > 1\n",
+				       obj_prio_node->prio);
+			return;
+		}
+	}
+
+	if (tree_remove_node(&root_ns->ns.node)) {
+		mlx5_core_warn(dev,
+			       "root namespace wasn't destroyed, refcount > 1\n");
+		return;
+	}
+
+	dev->priv.root_ns = NULL;
+}
+
+void mlx5_cleanup_fs(struct mlx5_core_dev *dev)
+{
+	cleanup_root_ns(dev);
+	cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns);
+}
+
+static int init_fdb_root_ns(struct mlx5_core_dev *dev)
+{
+	struct fs_prio *prio;
+
+	dev->priv.fdb_root_ns = create_root_ns(dev, FS_FT_FDB);
+	if (!dev->priv.fdb_root_ns)
+		return -ENOMEM;
+
+	/* Create single prio */
+	prio = fs_create_prio(&dev->priv.fdb_root_ns->ns, 0, 1);
+	if (IS_ERR(prio)) {
+		cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns);
+		return PTR_ERR(prio);
+	} else {
+		return 0;
+	}
+}
+
+int mlx5_init_fs(struct mlx5_core_dev *dev)
+{
+	int err = 0;
+
+	if (MLX5_CAP_GEN(dev, nic_flow_table)) {
+		err = init_root_ns(dev);
+		if (err)
+			return err;
+	}
+	if (MLX5_CAP_GEN(dev, eswitch_flow_table)) {
+		err = init_fdb_root_ns(dev);
+		if (err)
+			cleanup_root_ns(dev);
+	}
+
+	return err;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
new file mode 100644
index 000000000..00245fd7e
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -0,0 +1,169 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef _MLX5_FS_CORE_
+#define _MLX5_FS_CORE_
+
+#include <linux/mlx5/fs.h>
+
+enum fs_node_type {
+	FS_TYPE_NAMESPACE,
+	FS_TYPE_PRIO,
+	FS_TYPE_FLOW_TABLE,
+	FS_TYPE_FLOW_GROUP,
+	FS_TYPE_FLOW_ENTRY,
+	FS_TYPE_FLOW_DEST
+};
+
+enum fs_flow_table_type {
+	FS_FT_NIC_RX	 = 0x0,
+	FS_FT_FDB	 = 0X4,
+};
+
+enum fs_fte_status {
+	FS_FTE_STATUS_EXISTING = 1UL << 0,
+};
+
+struct fs_node {
+	struct list_head	list;
+	struct list_head	children;
+	enum fs_node_type	type;
+	struct fs_node		*parent;
+	struct fs_node		*root;
+	/* lock the node for writing and traversing */
+	struct mutex		lock;
+	atomic_t		refcount;
+	void			(*remove_func)(struct fs_node *);
+};
+
+struct mlx5_flow_rule {
+	struct fs_node				node;
+	struct mlx5_flow_destination		dest_attr;
+};
+
+/* Type of children is mlx5_flow_group */
+struct mlx5_flow_table {
+	struct fs_node			node;
+	u32				id;
+	unsigned int			max_fte;
+	unsigned int			level;
+	enum fs_flow_table_type		type;
+	struct {
+		bool			active;
+		unsigned int		required_groups;
+		unsigned int		num_groups;
+	} autogroup;
+};
+
+/* Type of children is mlx5_flow_rule */
+struct fs_fte {
+	struct fs_node			node;
+	u32				val[MLX5_ST_SZ_DW(fte_match_param)];
+	u32				dests_size;
+	u32				flow_tag;
+	u32				index;
+	u32				action;
+	enum fs_fte_status		status;
+};
+
+/* Type of children is mlx5_flow_table/namespace */
+struct fs_prio {
+	struct fs_node			node;
+	unsigned int			max_ft;
+	unsigned int			start_level;
+	unsigned int			prio;
+	unsigned int			num_ft;
+};
+
+/* Type of children is fs_prio */
+struct mlx5_flow_namespace {
+	/* parent == NULL => root ns */
+	struct	fs_node			node;
+};
+
+struct mlx5_flow_group_mask {
+	u8	match_criteria_enable;
+	u32	match_criteria[MLX5_ST_SZ_DW(fte_match_param)];
+};
+
+/* Type of children is fs_fte */
+struct mlx5_flow_group {
+	struct fs_node			node;
+	struct mlx5_flow_group_mask	mask;
+	u32				start_index;
+	u32				max_ftes;
+	u32				num_ftes;
+	u32				id;
+};
+
+struct mlx5_flow_root_namespace {
+	struct mlx5_flow_namespace	ns;
+	enum   fs_flow_table_type	table_type;
+	struct mlx5_core_dev		*dev;
+	struct mlx5_flow_table		*root_ft;
+	/* Should be held when chaining flow tables */
+	struct mutex			chain_lock;
+};
+
+int mlx5_init_fs(struct mlx5_core_dev *dev);
+void mlx5_cleanup_fs(struct mlx5_core_dev *dev);
+
+#define fs_get_obj(v, _node)  {v = container_of((_node), typeof(*v), node); }
+
+#define fs_list_for_each_entry(pos, root)		\
+	list_for_each_entry(pos, root, node.list)
+
+#define fs_for_each_ns_or_ft_reverse(pos, prio)				\
+	list_for_each_entry_reverse(pos, &(prio)->node.children, list)
+
+#define fs_for_each_ns_or_ft(pos, prio)					\
+	list_for_each_entry(pos, (&(prio)->node.children), list)
+
+#define fs_for_each_prio(pos, ns)			\
+	fs_list_for_each_entry(pos, &(ns)->node.children)
+
+#define fs_for_each_ns(pos, prio)			\
+	fs_list_for_each_entry(pos, &(prio)->node.children)
+
+#define fs_for_each_ft(pos, prio)			\
+	fs_list_for_each_entry(pos, &(prio)->node.children)
+
+#define fs_for_each_fg(pos, ft)			\
+	fs_list_for_each_entry(pos, &(ft)->node.children)
+
+#define fs_for_each_fte(pos, fg)			\
+	fs_list_for_each_entry(pos, &(fg)->node.children)
+
+#define fs_for_each_dst(pos, fte)			\
+	fs_list_for_each_entry(pos, &(fte)->node.children)
+
+#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
index 9335e5ae1..aa1ab4702 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
@@ -160,6 +160,30 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev)
 		if (err)
 			return err;
 	}
+
+	if (MLX5_CAP_GEN(dev, vport_group_manager) &&
+	    MLX5_CAP_GEN(dev, eswitch_flow_table)) {
+		err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH_FLOW_TABLE,
+					 HCA_CAP_OPMOD_GET_CUR);
+		if (err)
+			return err;
+		err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH_FLOW_TABLE,
+					 HCA_CAP_OPMOD_GET_MAX);
+		if (err)
+			return err;
+	}
+
+	if (MLX5_CAP_GEN(dev, eswitch_flow_table)) {
+		err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH,
+					 HCA_CAP_OPMOD_GET_CUR);
+		if (err)
+			return err;
+		err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH,
+					 HCA_CAP_OPMOD_GET_MAX);
+		if (err)
+			return err;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 6cf6d93d8..1545a944c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -49,6 +49,10 @@
 #include <linux/delay.h>
 #include <linux/mlx5/mlx5_ifc.h>
 #include "mlx5_core.h"
+#include "fs_core.h"
+#ifdef CONFIG_MLX5_CORE_EN
+#include "eswitch.h"
+#endif
 
 MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
 MODULE_DESCRIPTION("Mellanox Connect-IB, ConnectX-4 core driver");
@@ -74,6 +78,11 @@ struct mlx5_device_context {
 	void		       *context;
 };
 
+enum {
+	MLX5_ATOMIC_REQ_MODE_BE = 0x0,
+	MLX5_ATOMIC_REQ_MODE_HOST_ENDIANNESS = 0x1,
+};
+
 static struct mlx5_profile profile[] = {
 	[0] = {
 		.mask           = 0,
@@ -383,7 +392,7 @@ query_ex:
 	return err;
 }
 
-static int set_caps(struct mlx5_core_dev *dev, void *in, int in_sz)
+static int set_caps(struct mlx5_core_dev *dev, void *in, int in_sz, int opmod)
 {
 	u32 out[MLX5_ST_SZ_DW(set_hca_cap_out)];
 	int err;
@@ -391,6 +400,7 @@ static int set_caps(struct mlx5_core_dev *dev, void *in, int in_sz)
 	memset(out, 0, sizeof(out));
 
 	MLX5_SET(set_hca_cap_in, in, opcode, MLX5_CMD_OP_SET_HCA_CAP);
+	MLX5_SET(set_hca_cap_in, in, op_mod, opmod << 1);
 	err = mlx5_cmd_exec(dev, in, in_sz, out, sizeof(out));
 	if (err)
 		return err;
@@ -400,6 +410,46 @@ static int set_caps(struct mlx5_core_dev *dev, void *in, int in_sz)
 	return err;
 }
 
+static int handle_hca_cap_atomic(struct mlx5_core_dev *dev)
+{
+	void *set_ctx;
+	void *set_hca_cap;
+	int set_sz = MLX5_ST_SZ_BYTES(set_hca_cap_in);
+	int req_endianness;
+	int err;
+
+	if (MLX5_CAP_GEN(dev, atomic)) {
+		err = mlx5_core_get_caps(dev, MLX5_CAP_ATOMIC,
+					 HCA_CAP_OPMOD_GET_CUR);
+		if (err)
+			return err;
+	} else {
+		return 0;
+	}
+
+	req_endianness =
+		MLX5_CAP_ATOMIC(dev,
+				supported_atomic_req_8B_endianess_mode_1);
+
+	if (req_endianness != MLX5_ATOMIC_REQ_MODE_HOST_ENDIANNESS)
+		return 0;
+
+	set_ctx = kzalloc(set_sz, GFP_KERNEL);
+	if (!set_ctx)
+		return -ENOMEM;
+
+	set_hca_cap = MLX5_ADDR_OF(set_hca_cap_in, set_ctx, capability);
+
+	/* Set requestor to host endianness */
+	MLX5_SET(atomic_caps, set_hca_cap, atomic_req_8B_endianess_mode,
+		 MLX5_ATOMIC_REQ_MODE_HOST_ENDIANNESS);
+
+	err = set_caps(dev, set_ctx, set_sz, MLX5_SET_HCA_CAP_OP_MOD_ATOMIC);
+
+	kfree(set_ctx);
+	return err;
+}
+
 static int handle_hca_cap(struct mlx5_core_dev *dev)
 {
 	void *set_ctx = NULL;
@@ -441,7 +491,8 @@ static int handle_hca_cap(struct mlx5_core_dev *dev)
 
 	MLX5_SET(cmd_hca_cap, set_hca_cap, log_uar_page_sz, PAGE_SHIFT - 12);
 
-	err = set_caps(dev, set_ctx, set_sz);
+	err = set_caps(dev, set_ctx, set_sz,
+		       MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE);
 
 query_ex:
 	kfree(set_ctx);
@@ -454,6 +505,9 @@ static int set_hca_ctrl(struct mlx5_core_dev *dev)
 	struct mlx5_reg_host_endianess he_out;
 	int err;
 
+	if (!mlx5_core_is_pf(dev))
+		return 0;
+
 	memset(&he_in, 0, sizeof(he_in));
 	he_in.he = MLX5_SET_HOST_ENDIANNESS;
 	err = mlx5_core_access_reg(dev, &he_in,  sizeof(he_in),
@@ -462,42 +516,52 @@ static int set_hca_ctrl(struct mlx5_core_dev *dev)
 	return err;
 }
 
-static int mlx5_core_enable_hca(struct mlx5_core_dev *dev)
+int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id)
 {
+	u32 out[MLX5_ST_SZ_DW(enable_hca_out)];
+	u32 in[MLX5_ST_SZ_DW(enable_hca_in)];
 	int err;
-	struct mlx5_enable_hca_mbox_in in;
-	struct mlx5_enable_hca_mbox_out out;
 
-	memset(&in, 0, sizeof(in));
-	memset(&out, 0, sizeof(out));
-	in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_ENABLE_HCA);
+	memset(in, 0, sizeof(in));
+	MLX5_SET(enable_hca_in, in, opcode, MLX5_CMD_OP_ENABLE_HCA);
+	MLX5_SET(enable_hca_in, in, function_id, func_id);
+	memset(out, 0, sizeof(out));
+
 	err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out));
 	if (err)
 		return err;
 
-	if (out.hdr.status)
-		return mlx5_cmd_status_to_err(&out.hdr);
-
-	return 0;
+	return mlx5_cmd_status_to_err_v2(out);
 }
 
-static int mlx5_core_disable_hca(struct mlx5_core_dev *dev)
+int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id)
 {
+	u32 out[MLX5_ST_SZ_DW(disable_hca_out)];
+	u32 in[MLX5_ST_SZ_DW(disable_hca_in)];
 	int err;
-	struct mlx5_disable_hca_mbox_in in;
-	struct mlx5_disable_hca_mbox_out out;
 
-	memset(&in, 0, sizeof(in));
-	memset(&out, 0, sizeof(out));
-	in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_DISABLE_HCA);
-	err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out));
+	memset(in, 0, sizeof(in));
+	MLX5_SET(disable_hca_in, in, opcode, MLX5_CMD_OP_DISABLE_HCA);
+	MLX5_SET(disable_hca_in, in, function_id, func_id);
+	memset(out, 0, sizeof(out));
+	err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
 	if (err)
 		return err;
 
-	if (out.hdr.status)
-		return mlx5_cmd_status_to_err(&out.hdr);
+	return mlx5_cmd_status_to_err_v2(out);
+}
 
-	return 0;
+cycle_t mlx5_read_internal_timer(struct mlx5_core_dev *dev)
+{
+	u32 timer_h, timer_h1, timer_l;
+
+	timer_h = ioread32be(&dev->iseg->internal_timer_h);
+	timer_l = ioread32be(&dev->iseg->internal_timer_l);
+	timer_h1 = ioread32be(&dev->iseg->internal_timer_h);
+	if (timer_h != timer_h1) /* wrap around */
+		timer_l = ioread32be(&dev->iseg->internal_timer_l);
+
+	return (cycle_t)timer_l | (cycle_t)timer_h1 << 32;
 }
 
 static int mlx5_irq_set_affinity_hint(struct mlx5_core_dev *mdev, int i)
@@ -650,7 +714,6 @@ clean:
 	return err;
 }
 
-#ifdef CONFIG_MLX5_CORE_EN
 static int mlx5_core_set_issi(struct mlx5_core_dev *dev)
 {
 	u32 query_in[MLX5_ST_SZ_DW(query_issi_in)];
@@ -703,7 +766,6 @@ static int mlx5_core_set_issi(struct mlx5_core_dev *dev)
 
 	return -ENOTSUPP;
 }
-#endif
 
 static int map_bf_area(struct mlx5_core_dev *dev)
 {
@@ -943,19 +1005,17 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 
 	mlx5_pagealloc_init(dev);
 
-	err = mlx5_core_enable_hca(dev);
+	err = mlx5_core_enable_hca(dev, 0);
 	if (err) {
 		dev_err(&pdev->dev, "enable hca failed\n");
 		goto err_pagealloc_cleanup;
 	}
 
-#ifdef CONFIG_MLX5_CORE_EN
 	err = mlx5_core_set_issi(dev);
 	if (err) {
 		dev_err(&pdev->dev, "failed to set issi\n");
 		goto err_disable_hca;
 	}
-#endif
 
 	err = mlx5_satisfy_startup_pages(dev, 1);
 	if (err) {
@@ -975,6 +1035,12 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 		goto reclaim_boot_pages;
 	}
 
+	err = handle_hca_cap_atomic(dev);
+	if (err) {
+		dev_err(&pdev->dev, "handle_hca_cap_atomic failed\n");
+		goto reclaim_boot_pages;
+	}
+
 	err = mlx5_satisfy_startup_pages(dev, 0);
 	if (err) {
 		dev_err(&pdev->dev, "failed to allocate init pages\n");
@@ -1053,6 +1119,25 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 	mlx5_init_srq_table(dev);
 	mlx5_init_mr_table(dev);
 
+	err = mlx5_init_fs(dev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init flow steering\n");
+		goto err_fs;
+	}
+#ifdef CONFIG_MLX5_CORE_EN
+	err = mlx5_eswitch_init(dev);
+	if (err) {
+		dev_err(&pdev->dev, "eswitch init failed %d\n", err);
+		goto err_reg_dev;
+	}
+#endif
+
+	err = mlx5_sriov_init(dev);
+	if (err) {
+		dev_err(&pdev->dev, "sriov init failed %d\n", err);
+		goto err_sriov;
+	}
+
 	err = mlx5_register_device(dev);
 	if (err) {
 		dev_err(&pdev->dev, "mlx5_register_device failed %d\n", err);
@@ -1069,7 +1154,16 @@ out:
 
 	return 0;
 
+err_sriov:
+	if (mlx5_sriov_cleanup(dev))
+		dev_err(&dev->pdev->dev, "sriov cleanup failed\n");
+
+#ifdef CONFIG_MLX5_CORE_EN
+	mlx5_eswitch_cleanup(dev->priv.eswitch);
+#endif
 err_reg_dev:
+	mlx5_cleanup_fs(dev);
+err_fs:
 	mlx5_cleanup_mr_table(dev);
 	mlx5_cleanup_srq_table(dev);
 	mlx5_cleanup_qp_table(dev);
@@ -1107,7 +1201,7 @@ reclaim_boot_pages:
 	mlx5_reclaim_startup_pages(dev);
 
 err_disable_hca:
-	mlx5_core_disable_hca(dev);
+	mlx5_core_disable_hca(dev, 0);
 
 err_pagealloc_cleanup:
 	mlx5_pagealloc_cleanup(dev);
@@ -1124,6 +1218,13 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 {
 	int err = 0;
 
+	err = mlx5_sriov_cleanup(dev);
+	if (err) {
+		dev_warn(&dev->pdev->dev, "%s: sriov cleanup failed - abort\n",
+			 __func__);
+		return err;
+	}
+
 	mutex_lock(&dev->intf_state_mutex);
 	if (dev->interface_state == MLX5_INTERFACE_STATE_DOWN) {
 		dev_warn(&dev->pdev->dev, "%s: interface is down, NOP\n",
@@ -1131,6 +1232,11 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 		goto out;
 	}
 	mlx5_unregister_device(dev);
+#ifdef CONFIG_MLX5_CORE_EN
+	mlx5_eswitch_cleanup(dev->priv.eswitch);
+#endif
+
+	mlx5_cleanup_fs(dev);
 	mlx5_cleanup_mr_table(dev);
 	mlx5_cleanup_srq_table(dev);
 	mlx5_cleanup_qp_table(dev);
@@ -1150,7 +1256,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 	}
 	mlx5_pagealloc_stop(dev);
 	mlx5_reclaim_startup_pages(dev);
-	mlx5_core_disable_hca(dev);
+	mlx5_core_disable_hca(dev, 0);
 	mlx5_pagealloc_cleanup(dev);
 	mlx5_cmd_cleanup(dev);
 
@@ -1196,6 +1302,7 @@ static int init_one(struct pci_dev *pdev,
 		return -ENOMEM;
 	}
 	priv = &dev->priv;
+	priv->pci_dev_data = id->driver_data;
 
 	pci_set_drvdata(pdev, dev);
 
@@ -1366,12 +1473,12 @@ static const struct pci_error_handlers mlx5_err_handler = {
 };
 
 static const struct pci_device_id mlx5_core_pci_table[] = {
-	{ PCI_VDEVICE(MELLANOX, 0x1011) }, /* Connect-IB */
-	{ PCI_VDEVICE(MELLANOX, 0x1012) }, /* Connect-IB VF */
-	{ PCI_VDEVICE(MELLANOX, 0x1013) }, /* ConnectX-4 */
-	{ PCI_VDEVICE(MELLANOX, 0x1014) }, /* ConnectX-4 VF */
-	{ PCI_VDEVICE(MELLANOX, 0x1015) }, /* ConnectX-4LX */
-	{ PCI_VDEVICE(MELLANOX, 0x1016) }, /* ConnectX-4LX VF */
+	{ PCI_VDEVICE(MELLANOX, 0x1011) },			/* Connect-IB */
+	{ PCI_VDEVICE(MELLANOX, 0x1012), MLX5_PCI_DEV_IS_VF},	/* Connect-IB VF */
+	{ PCI_VDEVICE(MELLANOX, 0x1013) },			/* ConnectX-4 */
+	{ PCI_VDEVICE(MELLANOX, 0x1014), MLX5_PCI_DEV_IS_VF},	/* ConnectX-4 VF */
+	{ PCI_VDEVICE(MELLANOX, 0x1015) },			/* ConnectX-4LX */
+	{ PCI_VDEVICE(MELLANOX, 0x1016), MLX5_PCI_DEV_IS_VF},	/* ConnectX-4LX VF */
 	{ 0, }
 };
 
@@ -1382,7 +1489,8 @@ static struct pci_driver mlx5_core_driver = {
 	.id_table       = mlx5_core_pci_table,
 	.probe          = init_one,
 	.remove         = remove_one,
-	.err_handler	= &mlx5_err_handler
+	.err_handler	= &mlx5_err_handler,
+	.sriov_configure   = mlx5_core_sriov_configure,
 };
 
 static int __init init(void)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index cee5b7a83..0336847ec 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -36,6 +36,7 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/if_link.h>
 
 #define DRIVER_NAME "mlx5_core"
 #define DRIVER_VERSION "3.0-1"
@@ -64,6 +65,9 @@ do {									\
 		(__dev)->priv.name, __func__, __LINE__, current->pid,	\
 		##__VA_ARGS__)
 
+#define mlx5_core_info(__dev, format, ...)				\
+	dev_info(&(__dev)->pdev->dev, format, ##__VA_ARGS__)
+
 enum {
 	MLX5_CMD_DATA, /* print command payload only */
 	MLX5_CMD_TIME, /* print command execution time */
@@ -90,6 +94,11 @@ void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event,
 		     unsigned long param);
 void mlx5_enter_error_state(struct mlx5_core_dev *dev);
 void mlx5_disable_device(struct mlx5_core_dev *dev);
+int mlx5_core_sriov_configure(struct pci_dev *dev, int num_vfs);
+int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id);
+int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id);
+int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev);
+cycle_t mlx5_read_internal_timer(struct mlx5_core_dev *dev);
 
 void mlx5e_init(void);
 void mlx5e_cleanup(void);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
index 4d3377b12..9eeee0545 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
@@ -33,6 +33,7 @@
 #include <linux/highmem.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/delay.h>
 #include <linux/mlx5/driver.h>
 #include <linux/mlx5/cmd.h>
 #include "mlx5_core.h"
@@ -95,6 +96,7 @@ struct mlx5_manage_pages_outbox {
 
 enum {
 	MAX_RECLAIM_TIME_MSECS	= 5000,
+	MAX_RECLAIM_VFS_PAGES_TIME_MSECS = 2 * 1000 * 60,
 };
 
 enum {
@@ -352,6 +354,10 @@ retry:
 		goto out_4k;
 	}
 
+	dev->priv.fw_pages += npages;
+	if (func_id)
+		dev->priv.vfs_pages += npages;
+
 	mlx5_core_dbg(dev, "err %d\n", err);
 
 	kvfree(in);
@@ -405,6 +411,12 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
 	}
 
 	num_claimed = be32_to_cpu(out->num_entries);
+	if (num_claimed > npages) {
+		mlx5_core_warn(dev, "fw returned %d, driver asked %d => corruption\n",
+			       num_claimed, npages);
+		err = -EINVAL;
+		goto out_free;
+	}
 	if (nclaimed)
 		*nclaimed = num_claimed;
 
@@ -412,6 +424,9 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
 		addr = be64_to_cpu(out->pas[i]);
 		free_4k(dev, addr);
 	}
+	dev->priv.fw_pages -= num_claimed;
+	if (func_id)
+		dev->priv.vfs_pages -= num_claimed;
 
 out_free:
 	kvfree(out);
@@ -548,3 +563,26 @@ void mlx5_pagealloc_stop(struct mlx5_core_dev *dev)
 {
 	destroy_workqueue(dev->priv.pg_wq);
 }
+
+int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev)
+{
+	unsigned long end = jiffies + msecs_to_jiffies(MAX_RECLAIM_VFS_PAGES_TIME_MSECS);
+	int prev_vfs_pages = dev->priv.vfs_pages;
+
+	mlx5_core_dbg(dev, "Waiting for %d pages from %s\n", prev_vfs_pages,
+		      dev->priv.name);
+	while (dev->priv.vfs_pages) {
+		if (time_after(jiffies, end)) {
+			mlx5_core_warn(dev, "aborting while there are %d pending pages\n", dev->priv.vfs_pages);
+			return -ETIMEDOUT;
+		}
+		if (dev->priv.vfs_pages < prev_vfs_pages) {
+			end = jiffies + msecs_to_jiffies(MAX_RECLAIM_VFS_PAGES_TIME_MSECS);
+			prev_vfs_pages = dev->priv.vfs_pages;
+		}
+		msleep(50);
+	}
+
+	mlx5_core_dbg(dev, "All pages received from %s\n", dev->priv.name);
+	return 0;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/qp.c b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
index 30e2ba3f5..def289375 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
@@ -36,6 +36,7 @@
 #include <linux/mlx5/cmd.h>
 #include <linux/mlx5/qp.h>
 #include <linux/mlx5/driver.h>
+#include <linux/mlx5/transobj.h>
 
 #include "mlx5_core.h"
 
@@ -67,6 +68,52 @@ void mlx5_core_put_rsc(struct mlx5_core_rsc_common *common)
 		complete(&common->free);
 }
 
+static u64 qp_allowed_event_types(void)
+{
+	u64 mask;
+
+	mask = BIT(MLX5_EVENT_TYPE_PATH_MIG) |
+	       BIT(MLX5_EVENT_TYPE_COMM_EST) |
+	       BIT(MLX5_EVENT_TYPE_SQ_DRAINED) |
+	       BIT(MLX5_EVENT_TYPE_SRQ_LAST_WQE) |
+	       BIT(MLX5_EVENT_TYPE_WQ_CATAS_ERROR) |
+	       BIT(MLX5_EVENT_TYPE_PATH_MIG_FAILED) |
+	       BIT(MLX5_EVENT_TYPE_WQ_INVAL_REQ_ERROR) |
+	       BIT(MLX5_EVENT_TYPE_WQ_ACCESS_ERROR);
+
+	return mask;
+}
+
+static u64 rq_allowed_event_types(void)
+{
+	u64 mask;
+
+	mask = BIT(MLX5_EVENT_TYPE_SRQ_LAST_WQE) |
+	       BIT(MLX5_EVENT_TYPE_WQ_CATAS_ERROR);
+
+	return mask;
+}
+
+static u64 sq_allowed_event_types(void)
+{
+	return BIT(MLX5_EVENT_TYPE_WQ_CATAS_ERROR);
+}
+
+static bool is_event_type_allowed(int rsc_type, int event_type)
+{
+	switch (rsc_type) {
+	case MLX5_EVENT_QUEUE_TYPE_QP:
+		return BIT(event_type) & qp_allowed_event_types();
+	case MLX5_EVENT_QUEUE_TYPE_RQ:
+		return BIT(event_type) & rq_allowed_event_types();
+	case MLX5_EVENT_QUEUE_TYPE_SQ:
+		return BIT(event_type) & sq_allowed_event_types();
+	default:
+		WARN(1, "Event arrived for unknown resource type");
+		return false;
+	}
+}
+
 void mlx5_rsc_event(struct mlx5_core_dev *dev, u32 rsn, int event_type)
 {
 	struct mlx5_core_rsc_common *common = mlx5_get_rsc(dev, rsn);
@@ -75,8 +122,16 @@ void mlx5_rsc_event(struct mlx5_core_dev *dev, u32 rsn, int event_type)
 	if (!common)
 		return;
 
+	if (!is_event_type_allowed((rsn >> MLX5_USER_INDEX_LEN), event_type)) {
+		mlx5_core_warn(dev, "event 0x%.2x is not allowed on resource 0x%.8x\n",
+			       event_type, rsn);
+		return;
+	}
+
 	switch (common->res) {
 	case MLX5_RES_QP:
+	case MLX5_RES_RQ:
+	case MLX5_RES_SQ:
 		qp = (struct mlx5_core_qp *)common;
 		qp->event(qp, event_type);
 		break;
@@ -177,27 +232,56 @@ void mlx5_eq_pagefault(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe)
 }
 #endif
 
+static int create_qprqsq_common(struct mlx5_core_dev *dev,
+				struct mlx5_core_qp *qp,
+				int rsc_type)
+{
+	struct mlx5_qp_table *table = &dev->priv.qp_table;
+	int err;
+
+	qp->common.res = rsc_type;
+	spin_lock_irq(&table->lock);
+	err = radix_tree_insert(&table->tree,
+				qp->qpn | (rsc_type << MLX5_USER_INDEX_LEN),
+				qp);
+	spin_unlock_irq(&table->lock);
+	if (err)
+		return err;
+
+	atomic_set(&qp->common.refcount, 1);
+	init_completion(&qp->common.free);
+	qp->pid = current->pid;
+
+	return 0;
+}
+
+static void destroy_qprqsq_common(struct mlx5_core_dev *dev,
+				  struct mlx5_core_qp *qp)
+{
+	struct mlx5_qp_table *table = &dev->priv.qp_table;
+	unsigned long flags;
+
+	spin_lock_irqsave(&table->lock, flags);
+	radix_tree_delete(&table->tree,
+			  qp->qpn | (qp->common.res << MLX5_USER_INDEX_LEN));
+	spin_unlock_irqrestore(&table->lock, flags);
+	mlx5_core_put_rsc((struct mlx5_core_rsc_common *)qp);
+	wait_for_completion(&qp->common.free);
+}
+
 int mlx5_core_create_qp(struct mlx5_core_dev *dev,
 			struct mlx5_core_qp *qp,
 			struct mlx5_create_qp_mbox_in *in,
 			int inlen)
 {
-	struct mlx5_qp_table *table = &dev->priv.qp_table;
 	struct mlx5_create_qp_mbox_out out;
 	struct mlx5_destroy_qp_mbox_in din;
 	struct mlx5_destroy_qp_mbox_out dout;
 	int err;
-	void *qpc;
 
 	memset(&out, 0, sizeof(out));
 	in->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_CREATE_QP);
 
-	if (dev->issi) {
-		qpc = MLX5_ADDR_OF(create_qp_in, in, qpc);
-		/* 0xffffff means we ask to work with cqe version 0 */
-		MLX5_SET(qpc, qpc, user_index, 0xffffff);
-	}
-
 	err = mlx5_cmd_exec(dev, in, inlen, &out, sizeof(out));
 	if (err) {
 		mlx5_core_warn(dev, "ret %d\n", err);
@@ -213,24 +297,16 @@ int mlx5_core_create_qp(struct mlx5_core_dev *dev,
 	qp->qpn = be32_to_cpu(out.qpn) & 0xffffff;
 	mlx5_core_dbg(dev, "qpn = 0x%x\n", qp->qpn);
 
-	qp->common.res = MLX5_RES_QP;
-	spin_lock_irq(&table->lock);
-	err = radix_tree_insert(&table->tree, qp->qpn, qp);
-	spin_unlock_irq(&table->lock);
-	if (err) {
-		mlx5_core_warn(dev, "err %d\n", err);
+	err = create_qprqsq_common(dev, qp, MLX5_RES_QP);
+	if (err)
 		goto err_cmd;
-	}
 
 	err = mlx5_debug_qp_add(dev, qp);
 	if (err)
 		mlx5_core_dbg(dev, "failed adding QP 0x%x to debug file system\n",
 			      qp->qpn);
 
-	qp->pid = current->pid;
-	atomic_set(&qp->common.refcount, 1);
 	atomic_inc(&dev->num_qps);
-	init_completion(&qp->common.free);
 
 	return 0;
 
@@ -250,18 +326,11 @@ int mlx5_core_destroy_qp(struct mlx5_core_dev *dev,
 {
 	struct mlx5_destroy_qp_mbox_in in;
 	struct mlx5_destroy_qp_mbox_out out;
-	struct mlx5_qp_table *table = &dev->priv.qp_table;
-	unsigned long flags;
 	int err;
 
 	mlx5_debug_qp_remove(dev, qp);
 
-	spin_lock_irqsave(&table->lock, flags);
-	radix_tree_delete(&table->tree, qp->qpn);
-	spin_unlock_irqrestore(&table->lock, flags);
-
-	mlx5_core_put_rsc((struct mlx5_core_rsc_common *)qp);
-	wait_for_completion(&qp->common.free);
+	destroy_qprqsq_common(dev, qp);
 
 	memset(&in, 0, sizeof(in));
 	memset(&out, 0, sizeof(out));
@@ -279,59 +348,15 @@ int mlx5_core_destroy_qp(struct mlx5_core_dev *dev,
 }
 EXPORT_SYMBOL_GPL(mlx5_core_destroy_qp);
 
-int mlx5_core_qp_modify(struct mlx5_core_dev *dev, enum mlx5_qp_state cur_state,
-			enum mlx5_qp_state new_state,
+int mlx5_core_qp_modify(struct mlx5_core_dev *dev, u16 operation,
 			struct mlx5_modify_qp_mbox_in *in, int sqd_event,
 			struct mlx5_core_qp *qp)
 {
-	static const u16 optab[MLX5_QP_NUM_STATE][MLX5_QP_NUM_STATE] = {
-		[MLX5_QP_STATE_RST] = {
-			[MLX5_QP_STATE_RST]	= MLX5_CMD_OP_2RST_QP,
-			[MLX5_QP_STATE_ERR]	= MLX5_CMD_OP_2ERR_QP,
-			[MLX5_QP_STATE_INIT]	= MLX5_CMD_OP_RST2INIT_QP,
-		},
-		[MLX5_QP_STATE_INIT]  = {
-			[MLX5_QP_STATE_RST]	= MLX5_CMD_OP_2RST_QP,
-			[MLX5_QP_STATE_ERR]	= MLX5_CMD_OP_2ERR_QP,
-			[MLX5_QP_STATE_INIT]	= MLX5_CMD_OP_INIT2INIT_QP,
-			[MLX5_QP_STATE_RTR]	= MLX5_CMD_OP_INIT2RTR_QP,
-		},
-		[MLX5_QP_STATE_RTR]   = {
-			[MLX5_QP_STATE_RST]	= MLX5_CMD_OP_2RST_QP,
-			[MLX5_QP_STATE_ERR]	= MLX5_CMD_OP_2ERR_QP,
-			[MLX5_QP_STATE_RTS]	= MLX5_CMD_OP_RTR2RTS_QP,
-		},
-		[MLX5_QP_STATE_RTS]   = {
-			[MLX5_QP_STATE_RST]	= MLX5_CMD_OP_2RST_QP,
-			[MLX5_QP_STATE_ERR]	= MLX5_CMD_OP_2ERR_QP,
-			[MLX5_QP_STATE_RTS]	= MLX5_CMD_OP_RTS2RTS_QP,
-		},
-		[MLX5_QP_STATE_SQD] = {
-			[MLX5_QP_STATE_RST]	= MLX5_CMD_OP_2RST_QP,
-			[MLX5_QP_STATE_ERR]	= MLX5_CMD_OP_2ERR_QP,
-		},
-		[MLX5_QP_STATE_SQER] = {
-			[MLX5_QP_STATE_RST]	= MLX5_CMD_OP_2RST_QP,
-			[MLX5_QP_STATE_ERR]	= MLX5_CMD_OP_2ERR_QP,
-			[MLX5_QP_STATE_RTS]	= MLX5_CMD_OP_SQERR2RTS_QP,
-		},
-		[MLX5_QP_STATE_ERR] = {
-			[MLX5_QP_STATE_RST]	= MLX5_CMD_OP_2RST_QP,
-			[MLX5_QP_STATE_ERR]	= MLX5_CMD_OP_2ERR_QP,
-		}
-	};
-
 	struct mlx5_modify_qp_mbox_out out;
 	int err = 0;
-	u16 op;
-
-	if (cur_state >= MLX5_QP_NUM_STATE || new_state >= MLX5_QP_NUM_STATE ||
-	    !optab[cur_state][new_state])
-		return -EINVAL;
 
 	memset(&out, 0, sizeof(out));
-	op = optab[cur_state][new_state];
-	in->hdr.opcode = cpu_to_be16(op);
+	in->hdr.opcode = cpu_to_be16(operation);
 	in->qpn = cpu_to_be32(qp->qpn);
 	err = mlx5_cmd_exec(dev, in, sizeof(*in), &out, sizeof(out));
 	if (err)
@@ -449,3 +474,67 @@ int mlx5_core_page_fault_resume(struct mlx5_core_dev *dev, u32 qpn,
 }
 EXPORT_SYMBOL_GPL(mlx5_core_page_fault_resume);
 #endif
+
+int mlx5_core_create_rq_tracked(struct mlx5_core_dev *dev, u32 *in, int inlen,
+				struct mlx5_core_qp *rq)
+{
+	int err;
+	u32 rqn;
+
+	err = mlx5_core_create_rq(dev, in, inlen, &rqn);
+	if (err)
+		return err;
+
+	rq->qpn = rqn;
+	err = create_qprqsq_common(dev, rq, MLX5_RES_RQ);
+	if (err)
+		goto err_destroy_rq;
+
+	return 0;
+
+err_destroy_rq:
+	mlx5_core_destroy_rq(dev, rq->qpn);
+
+	return err;
+}
+EXPORT_SYMBOL(mlx5_core_create_rq_tracked);
+
+void mlx5_core_destroy_rq_tracked(struct mlx5_core_dev *dev,
+				  struct mlx5_core_qp *rq)
+{
+	destroy_qprqsq_common(dev, rq);
+	mlx5_core_destroy_rq(dev, rq->qpn);
+}
+EXPORT_SYMBOL(mlx5_core_destroy_rq_tracked);
+
+int mlx5_core_create_sq_tracked(struct mlx5_core_dev *dev, u32 *in, int inlen,
+				struct mlx5_core_qp *sq)
+{
+	int err;
+	u32 sqn;
+
+	err = mlx5_core_create_sq(dev, in, inlen, &sqn);
+	if (err)
+		return err;
+
+	sq->qpn = sqn;
+	err = create_qprqsq_common(dev, sq, MLX5_RES_SQ);
+	if (err)
+		goto err_destroy_sq;
+
+	return 0;
+
+err_destroy_sq:
+	mlx5_core_destroy_sq(dev, sq->qpn);
+
+	return err;
+}
+EXPORT_SYMBOL(mlx5_core_create_sq_tracked);
+
+void mlx5_core_destroy_sq_tracked(struct mlx5_core_dev *dev,
+				  struct mlx5_core_qp *sq)
+{
+	destroy_qprqsq_common(dev, sq);
+	mlx5_core_destroy_sq(dev, sq->qpn);
+}
+EXPORT_SYMBOL(mlx5_core_destroy_sq_tracked);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
new file mode 100644
index 000000000..7b2438679
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
@@ -0,0 +1,233 @@
+/*
+ * Copyright (c) 2014, Mellanox Technologies inc.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/pci.h>
+#include <linux/mlx5/driver.h>
+#include "mlx5_core.h"
+#ifdef CONFIG_MLX5_CORE_EN
+#include "eswitch.h"
+#endif
+
+static void enable_vfs(struct mlx5_core_dev *dev, int num_vfs)
+{
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+	int err;
+	int vf;
+
+	for (vf = 1; vf <= num_vfs; vf++) {
+		err = mlx5_core_enable_hca(dev, vf);
+		if (err) {
+			mlx5_core_warn(dev, "failed to enable VF %d\n", vf - 1);
+		} else {
+			sriov->vfs_ctx[vf - 1].enabled = 1;
+			mlx5_core_dbg(dev, "successfully enabled VF %d\n", vf - 1);
+		}
+	}
+}
+
+static void disable_vfs(struct mlx5_core_dev *dev, int num_vfs)
+{
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+	int vf;
+
+	for (vf = 1; vf <= num_vfs; vf++) {
+		if (sriov->vfs_ctx[vf - 1].enabled) {
+			if (mlx5_core_disable_hca(dev, vf))
+				mlx5_core_warn(dev, "failed to disable VF %d\n", vf - 1);
+			else
+				sriov->vfs_ctx[vf - 1].enabled = 0;
+		}
+	}
+}
+
+static int mlx5_core_create_vfs(struct pci_dev *pdev, int num_vfs)
+{
+	struct mlx5_core_dev *dev  = pci_get_drvdata(pdev);
+	int err;
+
+	if (pci_num_vf(pdev))
+		pci_disable_sriov(pdev);
+
+	enable_vfs(dev, num_vfs);
+
+	err = pci_enable_sriov(pdev, num_vfs);
+	if (err) {
+		dev_warn(&pdev->dev, "enable sriov failed %d\n", err);
+		goto ex;
+	}
+
+	return 0;
+
+ex:
+	disable_vfs(dev, num_vfs);
+	return err;
+}
+
+static int mlx5_core_sriov_enable(struct pci_dev *pdev, int num_vfs)
+{
+	struct mlx5_core_dev *dev  = pci_get_drvdata(pdev);
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+	int err;
+
+	kfree(sriov->vfs_ctx);
+	sriov->vfs_ctx = kcalloc(num_vfs, sizeof(*sriov->vfs_ctx), GFP_ATOMIC);
+	if (!sriov->vfs_ctx)
+		return -ENOMEM;
+
+	sriov->enabled_vfs = num_vfs;
+	err = mlx5_core_create_vfs(pdev, num_vfs);
+	if (err) {
+		kfree(sriov->vfs_ctx);
+		sriov->vfs_ctx = NULL;
+		return err;
+	}
+
+	return 0;
+}
+
+static void mlx5_core_init_vfs(struct mlx5_core_dev *dev, int num_vfs)
+{
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+
+	sriov->num_vfs = num_vfs;
+}
+
+static void mlx5_core_cleanup_vfs(struct mlx5_core_dev *dev)
+{
+	struct mlx5_core_sriov *sriov;
+
+	sriov = &dev->priv.sriov;
+	disable_vfs(dev, sriov->num_vfs);
+
+	if (mlx5_wait_for_vf_pages(dev))
+		mlx5_core_warn(dev, "timeout claiming VFs pages\n");
+
+	sriov->num_vfs = 0;
+}
+
+int mlx5_core_sriov_configure(struct pci_dev *pdev, int num_vfs)
+{
+	struct mlx5_core_dev *dev  = pci_get_drvdata(pdev);
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+	int err;
+
+	mlx5_core_dbg(dev, "requsted num_vfs %d\n", num_vfs);
+	if (!mlx5_core_is_pf(dev))
+		return -EPERM;
+
+	mlx5_core_cleanup_vfs(dev);
+
+	if (!num_vfs) {
+#ifdef CONFIG_MLX5_CORE_EN
+		mlx5_eswitch_disable_sriov(dev->priv.eswitch);
+#endif
+		kfree(sriov->vfs_ctx);
+		sriov->vfs_ctx = NULL;
+		if (!pci_vfs_assigned(pdev))
+			pci_disable_sriov(pdev);
+		else
+			pr_info("unloading PF driver while leaving orphan VFs\n");
+		return 0;
+	}
+
+	err = mlx5_core_sriov_enable(pdev, num_vfs);
+	if (err) {
+		dev_warn(&pdev->dev, "mlx5_core_sriov_enable failed %d\n", err);
+		return err;
+	}
+
+	mlx5_core_init_vfs(dev, num_vfs);
+#ifdef CONFIG_MLX5_CORE_EN
+	mlx5_eswitch_enable_sriov(dev->priv.eswitch, num_vfs);
+#endif
+
+	return num_vfs;
+}
+
+static int sync_required(struct pci_dev *pdev)
+{
+	struct mlx5_core_dev *dev  = pci_get_drvdata(pdev);
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+	int cur_vfs = pci_num_vf(pdev);
+
+	if (cur_vfs != sriov->num_vfs) {
+		pr_info("current VFs %d, registered %d - sync needed\n", cur_vfs, sriov->num_vfs);
+		return 1;
+	}
+
+	return 0;
+}
+
+int mlx5_sriov_init(struct mlx5_core_dev *dev)
+{
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+	struct pci_dev *pdev = dev->pdev;
+	int cur_vfs;
+
+	if (!mlx5_core_is_pf(dev))
+		return 0;
+
+	if (!sync_required(dev->pdev))
+		return 0;
+
+	cur_vfs = pci_num_vf(pdev);
+	sriov->vfs_ctx = kcalloc(cur_vfs, sizeof(*sriov->vfs_ctx), GFP_KERNEL);
+	if (!sriov->vfs_ctx)
+		return -ENOMEM;
+
+	sriov->enabled_vfs = cur_vfs;
+
+	mlx5_core_init_vfs(dev, cur_vfs);
+#ifdef CONFIG_MLX5_CORE_EN
+	if (cur_vfs)
+		mlx5_eswitch_enable_sriov(dev->priv.eswitch, cur_vfs);
+#endif
+
+	enable_vfs(dev, cur_vfs);
+
+	return 0;
+}
+
+int mlx5_sriov_cleanup(struct mlx5_core_dev *dev)
+{
+	struct pci_dev *pdev = dev->pdev;
+	int err;
+
+	if (!mlx5_core_is_pf(dev))
+		return 0;
+
+	err = mlx5_core_sriov_configure(pdev, 0);
+	if (err)
+		return err;
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/srq.c b/drivers/net/ethernet/mellanox/mlx5/core/srq.c
index ffada8019..04bc52260 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/srq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/srq.c
@@ -37,7 +37,7 @@
 #include <linux/mlx5/srq.h>
 #include <rdma/ib_verbs.h>
 #include "mlx5_core.h"
-#include "transobj.h"
+#include <linux/mlx5/transobj.h>
 
 void mlx5_srq_event(struct mlx5_core_dev *dev, u32 srqn, int event_type)
 {
@@ -241,8 +241,6 @@ static int create_xrc_srq_cmd(struct mlx5_core_dev *dev,
 
 	memcpy(xrc_srqc, srqc, MLX5_ST_SZ_BYTES(srqc));
 	memcpy(pas, in->pas, pas_size);
-	/* 0xffffff means we ask to work with cqe version 0 */
-	MLX5_SET(xrc_srqc,	    xrc_srqc,  user_index, 0xffffff);
 	MLX5_SET(create_xrc_srq_in, create_in, opcode,
 		 MLX5_CMD_OP_CREATE_XRC_SRQ);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/transobj.c b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c
index d7068f54e..03a5093ff 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/transobj.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c
@@ -32,9 +32,9 @@
 
 #include <linux/mlx5/driver.h>
 #include "mlx5_core.h"
-#include "transobj.h"
+#include <linux/mlx5/transobj.h>
 
-int mlx5_alloc_transport_domain(struct mlx5_core_dev *dev, u32 *tdn)
+int mlx5_core_alloc_transport_domain(struct mlx5_core_dev *dev, u32 *tdn)
 {
 	u32 in[MLX5_ST_SZ_DW(alloc_transport_domain_in)];
 	u32 out[MLX5_ST_SZ_DW(alloc_transport_domain_out)];
@@ -53,8 +53,9 @@ int mlx5_alloc_transport_domain(struct mlx5_core_dev *dev, u32 *tdn)
 
 	return err;
 }
+EXPORT_SYMBOL(mlx5_core_alloc_transport_domain);
 
-void mlx5_dealloc_transport_domain(struct mlx5_core_dev *dev, u32 tdn)
+void mlx5_core_dealloc_transport_domain(struct mlx5_core_dev *dev, u32 tdn)
 {
 	u32 in[MLX5_ST_SZ_DW(dealloc_transport_domain_in)];
 	u32 out[MLX5_ST_SZ_DW(dealloc_transport_domain_out)];
@@ -68,6 +69,7 @@ void mlx5_dealloc_transport_domain(struct mlx5_core_dev *dev, u32 tdn)
 
 	mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out));
 }
+EXPORT_SYMBOL(mlx5_core_dealloc_transport_domain);
 
 int mlx5_core_create_rq(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *rqn)
 {
@@ -94,6 +96,7 @@ int mlx5_core_modify_rq(struct mlx5_core_dev *dev, u32 rqn, u32 *in, int inlen)
 	memset(out, 0, sizeof(out));
 	return mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out));
 }
+EXPORT_SYMBOL(mlx5_core_modify_rq);
 
 void mlx5_core_destroy_rq(struct mlx5_core_dev *dev, u32 rqn)
 {
@@ -108,6 +111,18 @@ void mlx5_core_destroy_rq(struct mlx5_core_dev *dev, u32 rqn)
 	mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out));
 }
 
+int mlx5_core_query_rq(struct mlx5_core_dev *dev, u32 rqn, u32 *out)
+{
+	u32 in[MLX5_ST_SZ_DW(query_rq_in)] = {0};
+	int outlen = MLX5_ST_SZ_BYTES(query_rq_out);
+
+	MLX5_SET(query_rq_in, in, opcode, MLX5_CMD_OP_QUERY_RQ);
+	MLX5_SET(query_rq_in, in, rqn, rqn);
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, outlen);
+}
+EXPORT_SYMBOL(mlx5_core_query_rq);
+
 int mlx5_core_create_sq(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *sqn)
 {
 	u32 out[MLX5_ST_SZ_DW(create_sq_out)];
@@ -133,6 +148,7 @@ int mlx5_core_modify_sq(struct mlx5_core_dev *dev, u32 sqn, u32 *in, int inlen)
 	memset(out, 0, sizeof(out));
 	return mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out));
 }
+EXPORT_SYMBOL(mlx5_core_modify_sq);
 
 void mlx5_core_destroy_sq(struct mlx5_core_dev *dev, u32 sqn)
 {
@@ -147,6 +163,18 @@ void mlx5_core_destroy_sq(struct mlx5_core_dev *dev, u32 sqn)
 	mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out));
 }
 
+int mlx5_core_query_sq(struct mlx5_core_dev *dev, u32 sqn, u32 *out)
+{
+	u32 in[MLX5_ST_SZ_DW(query_sq_in)] = {0};
+	int outlen = MLX5_ST_SZ_BYTES(query_sq_out);
+
+	MLX5_SET(query_sq_in, in, opcode, MLX5_CMD_OP_QUERY_SQ);
+	MLX5_SET(query_sq_in, in, sqn, sqn);
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, outlen);
+}
+EXPORT_SYMBOL(mlx5_core_query_sq);
+
 int mlx5_core_create_tir(struct mlx5_core_dev *dev, u32 *in, int inlen,
 			 u32 *tirn)
 {
@@ -162,6 +190,7 @@ int mlx5_core_create_tir(struct mlx5_core_dev *dev, u32 *in, int inlen,
 
 	return err;
 }
+EXPORT_SYMBOL(mlx5_core_create_tir);
 
 int mlx5_core_modify_tir(struct mlx5_core_dev *dev, u32 tirn, u32 *in,
 			 int inlen)
@@ -187,6 +216,7 @@ void mlx5_core_destroy_tir(struct mlx5_core_dev *dev, u32 tirn)
 
 	mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out));
 }
+EXPORT_SYMBOL(mlx5_core_destroy_tir);
 
 int mlx5_core_create_tis(struct mlx5_core_dev *dev, u32 *in, int inlen,
 			 u32 *tisn)
@@ -203,6 +233,19 @@ int mlx5_core_create_tis(struct mlx5_core_dev *dev, u32 *in, int inlen,
 
 	return err;
 }
+EXPORT_SYMBOL(mlx5_core_create_tis);
+
+int mlx5_core_modify_tis(struct mlx5_core_dev *dev, u32 tisn, u32 *in,
+			 int inlen)
+{
+	u32 out[MLX5_ST_SZ_DW(modify_tis_out)] = {0};
+
+	MLX5_SET(modify_tis_in, in, tisn, tisn);
+	MLX5_SET(modify_tis_in, in, opcode, MLX5_CMD_OP_MODIFY_TIS);
+
+	return mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out));
+}
+EXPORT_SYMBOL(mlx5_core_modify_tis);
 
 void mlx5_core_destroy_tis(struct mlx5_core_dev *dev, u32 tisn)
 {
@@ -216,6 +259,7 @@ void mlx5_core_destroy_tis(struct mlx5_core_dev *dev, u32 tisn)
 
 	mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out));
 }
+EXPORT_SYMBOL(mlx5_core_destroy_tis);
 
 int mlx5_core_create_rmp(struct mlx5_core_dev *dev, u32 *in, int inlen,
 			 u32 *rmpn)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/transobj.h b/drivers/net/ethernet/mellanox/mlx5/core/transobj.h
deleted file mode 100644
index 74cae5143..000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/transobj.h
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * Copyright (c) 2013-2015, Mellanox Technologies, Ltd.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- */
-
-#ifndef __TRANSOBJ_H__
-#define __TRANSOBJ_H__
-
-int mlx5_alloc_transport_domain(struct mlx5_core_dev *dev, u32 *tdn);
-void mlx5_dealloc_transport_domain(struct mlx5_core_dev *dev, u32 tdn);
-int mlx5_core_create_rq(struct mlx5_core_dev *dev, u32 *in, int inlen,
-			u32 *rqn);
-int mlx5_core_modify_rq(struct mlx5_core_dev *dev, u32 rqn, u32 *in, int inlen);
-void mlx5_core_destroy_rq(struct mlx5_core_dev *dev, u32 rqn);
-int mlx5_core_create_sq(struct mlx5_core_dev *dev, u32 *in, int inlen,
-			u32 *sqn);
-int mlx5_core_modify_sq(struct mlx5_core_dev *dev, u32 sqn, u32 *in, int inlen);
-void mlx5_core_destroy_sq(struct mlx5_core_dev *dev, u32 sqn);
-int mlx5_core_create_tir(struct mlx5_core_dev *dev, u32 *in, int inlen,
-			 u32 *tirn);
-int mlx5_core_modify_tir(struct mlx5_core_dev *dev, u32 tirn, u32 *in,
-			 int inlen);
-void mlx5_core_destroy_tir(struct mlx5_core_dev *dev, u32 tirn);
-int mlx5_core_create_tis(struct mlx5_core_dev *dev, u32 *in, int inlen,
-			 u32 *tisn);
-void mlx5_core_destroy_tis(struct mlx5_core_dev *dev, u32 tisn);
-int mlx5_core_create_rmp(struct mlx5_core_dev *dev, u32 *in, int inlen,
-			 u32 *rmpn);
-int mlx5_core_modify_rmp(struct mlx5_core_dev *dev, u32 *in, int inlen);
-int mlx5_core_destroy_rmp(struct mlx5_core_dev *dev, u32 rmpn);
-int mlx5_core_query_rmp(struct mlx5_core_dev *dev, u32 rmpn, u32 *out);
-int mlx5_core_arm_rmp(struct mlx5_core_dev *dev, u32 rmpn, u16 lwm);
-int mlx5_core_create_xsrq(struct mlx5_core_dev *dev, u32 *in, int inlen,
-			  u32 *rmpn);
-int mlx5_core_destroy_xsrq(struct mlx5_core_dev *dev, u32 rmpn);
-int mlx5_core_query_xsrq(struct mlx5_core_dev *dev, u32 rmpn, u32 *out);
-int mlx5_core_arm_xsrq(struct mlx5_core_dev *dev, u32 rmpn, u16 lwm);
-
-int mlx5_core_create_rqt(struct mlx5_core_dev *dev, u32 *in, int inlen,
-			 u32 *rqtn);
-int mlx5_core_modify_rqt(struct mlx5_core_dev *dev, u32 rqtn, u32 *in,
-			 int inlen);
-void mlx5_core_destroy_rqt(struct mlx5_core_dev *dev, u32 rqtn);
-
-#endif /* __TRANSOBJ_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
index b94177ebc..c7398b95a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
@@ -36,54 +36,458 @@
 #include <linux/mlx5/vport.h>
 #include "mlx5_core.h"
 
-u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod)
+static int _mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod,
+				   u16 vport, u32 *out, int outlen)
 {
-	u32 in[MLX5_ST_SZ_DW(query_vport_state_in)];
-	u32 out[MLX5_ST_SZ_DW(query_vport_state_out)];
 	int err;
+	u32 in[MLX5_ST_SZ_DW(query_vport_state_in)];
 
 	memset(in, 0, sizeof(in));
 
 	MLX5_SET(query_vport_state_in, in, opcode,
 		 MLX5_CMD_OP_QUERY_VPORT_STATE);
 	MLX5_SET(query_vport_state_in, in, op_mod, opmod);
+	MLX5_SET(query_vport_state_in, in, vport_number, vport);
+	if (vport)
+		MLX5_SET(query_vport_state_in, in, other_vport, 1);
 
-	err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out,
-					 sizeof(out));
+	err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen);
 	if (err)
 		mlx5_core_warn(mdev, "MLX5_CMD_OP_QUERY_VPORT_STATE failed\n");
 
+	return err;
+}
+
+u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport)
+{
+	u32 out[MLX5_ST_SZ_DW(query_vport_state_out)] = {0};
+
+	_mlx5_query_vport_state(mdev, opmod, vport, out, sizeof(out));
+
 	return MLX5_GET(query_vport_state_out, out, state);
 }
-EXPORT_SYMBOL(mlx5_query_vport_state);
+EXPORT_SYMBOL_GPL(mlx5_query_vport_state);
+
+u8 mlx5_query_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport)
+{
+	u32 out[MLX5_ST_SZ_DW(query_vport_state_out)] = {0};
+
+	_mlx5_query_vport_state(mdev, opmod, vport, out, sizeof(out));
+
+	return MLX5_GET(query_vport_state_out, out, admin_state);
+}
+EXPORT_SYMBOL_GPL(mlx5_query_vport_admin_state);
 
-void mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u8 *addr)
+int mlx5_modify_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod,
+				  u16 vport, u8 state)
+{
+	u32 in[MLX5_ST_SZ_DW(modify_vport_state_in)];
+	u32 out[MLX5_ST_SZ_DW(modify_vport_state_out)];
+	int err;
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(modify_vport_state_in, in, opcode,
+		 MLX5_CMD_OP_MODIFY_VPORT_STATE);
+	MLX5_SET(modify_vport_state_in, in, op_mod, opmod);
+	MLX5_SET(modify_vport_state_in, in, vport_number, vport);
+
+	if (vport)
+		MLX5_SET(modify_vport_state_in, in, other_vport, 1);
+
+	MLX5_SET(modify_vport_state_in, in, admin_state, state);
+
+	err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out,
+					 sizeof(out));
+	if (err)
+		mlx5_core_warn(mdev, "MLX5_CMD_OP_MODIFY_VPORT_STATE failed\n");
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_modify_vport_admin_state);
+
+static int mlx5_query_nic_vport_context(struct mlx5_core_dev *mdev, u16 vport,
+					u32 *out, int outlen)
+{
+	u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)];
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(query_nic_vport_context_in, in, opcode,
+		 MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT);
+
+	MLX5_SET(query_nic_vport_context_in, in, vport_number, vport);
+	if (vport)
+		MLX5_SET(query_nic_vport_context_in, in, other_vport, 1);
+
+	return mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen);
+}
+
+static int mlx5_modify_nic_vport_context(struct mlx5_core_dev *mdev, void *in,
+					 int inlen)
+{
+	u32 out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)];
+
+	MLX5_SET(modify_nic_vport_context_in, in, opcode,
+		 MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT);
+
+	memset(out, 0, sizeof(out));
+	return mlx5_cmd_exec_check_status(mdev, in, inlen, out, sizeof(out));
+}
+
+int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev,
+				     u16 vport, u8 *addr)
 {
-	u32  in[MLX5_ST_SZ_DW(query_nic_vport_context_in)];
 	u32 *out;
 	int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out);
 	u8 *out_addr;
+	int err;
 
 	out = mlx5_vzalloc(outlen);
 	if (!out)
-		return;
+		return -ENOMEM;
 
 	out_addr = MLX5_ADDR_OF(query_nic_vport_context_out, out,
 				nic_vport_context.permanent_address);
 
+	err = mlx5_query_nic_vport_context(mdev, vport, out, outlen);
+	if (!err)
+		ether_addr_copy(addr, &out_addr[2]);
+
+	kvfree(out);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_mac_address);
+
+int mlx5_modify_nic_vport_mac_address(struct mlx5_core_dev *mdev,
+				      u16 vport, u8 *addr)
+{
+	void *in;
+	int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in);
+	int err;
+	void *nic_vport_ctx;
+	u8 *perm_mac;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in) {
+		mlx5_core_warn(mdev, "failed to allocate inbox\n");
+		return -ENOMEM;
+	}
+
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 field_select.permanent_address, 1);
+	MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport);
+
+	if (vport)
+		MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1);
+
+	nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in,
+				     in, nic_vport_context);
+	perm_mac = MLX5_ADDR_OF(nic_vport_context, nic_vport_ctx,
+				permanent_address);
+
+	ether_addr_copy(&perm_mac[2], addr);
+
+	err = mlx5_modify_nic_vport_context(mdev, in, inlen);
+
+	kvfree(in);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_mac_address);
+
+int mlx5_query_nic_vport_mac_list(struct mlx5_core_dev *dev,
+				  u32 vport,
+				  enum mlx5_list_type list_type,
+				  u8 addr_list[][ETH_ALEN],
+				  int *list_size)
+{
+	u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)];
+	void *nic_vport_ctx;
+	int max_list_size;
+	int req_list_size;
+	int out_sz;
+	void *out;
+	int err;
+	int i;
+
+	req_list_size = *list_size;
+
+	max_list_size = list_type == MLX5_NVPRT_LIST_TYPE_UC ?
+		1 << MLX5_CAP_GEN(dev, log_max_current_uc_list) :
+		1 << MLX5_CAP_GEN(dev, log_max_current_mc_list);
+
+	if (req_list_size > max_list_size) {
+		mlx5_core_warn(dev, "Requested list size (%d) > (%d) max_list_size\n",
+			       req_list_size, max_list_size);
+		req_list_size = max_list_size;
+	}
+
+	out_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) +
+			req_list_size * MLX5_ST_SZ_BYTES(mac_address_layout);
+
 	memset(in, 0, sizeof(in));
+	out = kzalloc(out_sz, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
 
 	MLX5_SET(query_nic_vport_context_in, in, opcode,
 		 MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT);
+	MLX5_SET(query_nic_vport_context_in, in, allowed_list_type, list_type);
+	MLX5_SET(query_nic_vport_context_in, in, vport_number, vport);
 
-	memset(out, 0, outlen);
-	mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen);
+	if (vport)
+		MLX5_SET(query_nic_vport_context_in, in, other_vport, 1);
 
-	ether_addr_copy(addr, &out_addr[2]);
+	err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, out_sz);
+	if (err)
+		goto out;
 
-	kvfree(out);
+	nic_vport_ctx = MLX5_ADDR_OF(query_nic_vport_context_out, out,
+				     nic_vport_context);
+	req_list_size = MLX5_GET(nic_vport_context, nic_vport_ctx,
+				 allowed_list_size);
+
+	*list_size = req_list_size;
+	for (i = 0; i < req_list_size; i++) {
+		u8 *mac_addr = MLX5_ADDR_OF(nic_vport_context,
+					nic_vport_ctx,
+					current_uc_mac_address[i]) + 2;
+		ether_addr_copy(addr_list[i], mac_addr);
+	}
+out:
+	kfree(out);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_mac_list);
+
+int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev,
+				   enum mlx5_list_type list_type,
+				   u8 addr_list[][ETH_ALEN],
+				   int list_size)
+{
+	u32 out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)];
+	void *nic_vport_ctx;
+	int max_list_size;
+	int in_sz;
+	void *in;
+	int err;
+	int i;
+
+	max_list_size = list_type == MLX5_NVPRT_LIST_TYPE_UC ?
+		 1 << MLX5_CAP_GEN(dev, log_max_current_uc_list) :
+		 1 << MLX5_CAP_GEN(dev, log_max_current_mc_list);
+
+	if (list_size > max_list_size)
+		return -ENOSPC;
+
+	in_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) +
+		list_size * MLX5_ST_SZ_BYTES(mac_address_layout);
+
+	memset(out, 0, sizeof(out));
+	in = kzalloc(in_sz, GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	MLX5_SET(modify_nic_vport_context_in, in, opcode,
+		 MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 field_select.addresses_list, 1);
+
+	nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in, in,
+				     nic_vport_context);
+
+	MLX5_SET(nic_vport_context, nic_vport_ctx,
+		 allowed_list_type, list_type);
+	MLX5_SET(nic_vport_context, nic_vport_ctx,
+		 allowed_list_size, list_size);
+
+	for (i = 0; i < list_size; i++) {
+		u8 *curr_mac = MLX5_ADDR_OF(nic_vport_context,
+					    nic_vport_ctx,
+					    current_uc_mac_address[i]) + 2;
+		ether_addr_copy(curr_mac, addr_list[i]);
+	}
+
+	err = mlx5_cmd_exec_check_status(dev, in, in_sz, out, sizeof(out));
+	kfree(in);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_mac_list);
+
+int mlx5_query_nic_vport_vlans(struct mlx5_core_dev *dev,
+			       u32 vport,
+			       u16 vlans[],
+			       int *size)
+{
+	u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)];
+	void *nic_vport_ctx;
+	int req_list_size;
+	int max_list_size;
+	int out_sz;
+	void *out;
+	int err;
+	int i;
+
+	req_list_size = *size;
+	max_list_size = 1 << MLX5_CAP_GEN(dev, log_max_vlan_list);
+	if (req_list_size > max_list_size) {
+		mlx5_core_warn(dev, "Requested list size (%d) > (%d) max list size\n",
+			       req_list_size, max_list_size);
+		req_list_size = max_list_size;
+	}
+
+	out_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) +
+			req_list_size * MLX5_ST_SZ_BYTES(vlan_layout);
+
+	memset(in, 0, sizeof(in));
+	out = kzalloc(out_sz, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
+
+	MLX5_SET(query_nic_vport_context_in, in, opcode,
+		 MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT);
+	MLX5_SET(query_nic_vport_context_in, in, allowed_list_type,
+		 MLX5_NVPRT_LIST_TYPE_VLAN);
+	MLX5_SET(query_nic_vport_context_in, in, vport_number, vport);
+
+	if (vport)
+		MLX5_SET(query_nic_vport_context_in, in, other_vport, 1);
+
+	err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, out_sz);
+	if (err)
+		goto out;
+
+	nic_vport_ctx = MLX5_ADDR_OF(query_nic_vport_context_out, out,
+				     nic_vport_context);
+	req_list_size = MLX5_GET(nic_vport_context, nic_vport_ctx,
+				 allowed_list_size);
+
+	*size = req_list_size;
+	for (i = 0; i < req_list_size; i++) {
+		void *vlan_addr = MLX5_ADDR_OF(nic_vport_context,
+					       nic_vport_ctx,
+					       current_uc_mac_address[i]);
+		vlans[i] = MLX5_GET(vlan_layout, vlan_addr, vlan);
+	}
+out:
+	kfree(out);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_vlans);
+
+int mlx5_modify_nic_vport_vlans(struct mlx5_core_dev *dev,
+				u16 vlans[],
+				int list_size)
+{
+	u32 out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)];
+	void *nic_vport_ctx;
+	int max_list_size;
+	int in_sz;
+	void *in;
+	int err;
+	int i;
+
+	max_list_size = 1 << MLX5_CAP_GEN(dev, log_max_vlan_list);
+
+	if (list_size > max_list_size)
+		return -ENOSPC;
+
+	in_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) +
+		list_size * MLX5_ST_SZ_BYTES(vlan_layout);
+
+	memset(out, 0, sizeof(out));
+	in = kzalloc(in_sz, GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	MLX5_SET(modify_nic_vport_context_in, in, opcode,
+		 MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 field_select.addresses_list, 1);
+
+	nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in, in,
+				     nic_vport_context);
+
+	MLX5_SET(nic_vport_context, nic_vport_ctx,
+		 allowed_list_type, MLX5_NVPRT_LIST_TYPE_VLAN);
+	MLX5_SET(nic_vport_context, nic_vport_ctx,
+		 allowed_list_size, list_size);
+
+	for (i = 0; i < list_size; i++) {
+		void *vlan_addr = MLX5_ADDR_OF(nic_vport_context,
+					       nic_vport_ctx,
+					       current_uc_mac_address[i]);
+		MLX5_SET(vlan_layout, vlan_addr, vlan, vlans[i]);
+	}
+
+	err = mlx5_cmd_exec_check_status(dev, in, in_sz, out, sizeof(out));
+	kfree(in);
+	return err;
 }
-EXPORT_SYMBOL(mlx5_query_nic_vport_mac_address);
+EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_vlans);
+
+int mlx5_query_nic_vport_system_image_guid(struct mlx5_core_dev *mdev,
+					   u64 *system_image_guid)
+{
+	u32 *out;
+	int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out);
+
+	out = mlx5_vzalloc(outlen);
+	if (!out)
+		return -ENOMEM;
+
+	mlx5_query_nic_vport_context(mdev, 0, out, outlen);
+
+	*system_image_guid = MLX5_GET64(query_nic_vport_context_out, out,
+					nic_vport_context.system_image_guid);
+
+	kfree(out);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_system_image_guid);
+
+int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid)
+{
+	u32 *out;
+	int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out);
+
+	out = mlx5_vzalloc(outlen);
+	if (!out)
+		return -ENOMEM;
+
+	mlx5_query_nic_vport_context(mdev, 0, out, outlen);
+
+	*node_guid = MLX5_GET64(query_nic_vport_context_out, out,
+				nic_vport_context.node_guid);
+
+	kfree(out);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_node_guid);
+
+int mlx5_query_nic_vport_qkey_viol_cntr(struct mlx5_core_dev *mdev,
+					u16 *qkey_viol_cntr)
+{
+	u32 *out;
+	int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out);
+
+	out = mlx5_vzalloc(outlen);
+	if (!out)
+		return -ENOMEM;
+
+	mlx5_query_nic_vport_context(mdev, 0, out, outlen);
+
+	*qkey_viol_cntr = MLX5_GET(query_nic_vport_context_out, out,
+				   nic_vport_context.qkey_violation_counter);
+
+	kfree(out);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_qkey_viol_cntr);
 
 int mlx5_query_hca_vport_gid(struct mlx5_core_dev *dev, u8 other_vport,
 			     u8 port_num, u16  vf_num, u16 gid_index,
@@ -343,3 +747,106 @@ int mlx5_query_hca_vport_node_guid(struct mlx5_core_dev *dev,
 	return err;
 }
 EXPORT_SYMBOL_GPL(mlx5_query_hca_vport_node_guid);
+
+int mlx5_query_nic_vport_promisc(struct mlx5_core_dev *mdev,
+				 u32 vport,
+				 int *promisc_uc,
+				 int *promisc_mc,
+				 int *promisc_all)
+{
+	u32 *out;
+	int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out);
+	int err;
+
+	out = kzalloc(outlen, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
+
+	err = mlx5_query_nic_vport_context(mdev, vport, out, outlen);
+	if (err)
+		goto out;
+
+	*promisc_uc = MLX5_GET(query_nic_vport_context_out, out,
+			       nic_vport_context.promisc_uc);
+	*promisc_mc = MLX5_GET(query_nic_vport_context_out, out,
+			       nic_vport_context.promisc_mc);
+	*promisc_all = MLX5_GET(query_nic_vport_context_out, out,
+				nic_vport_context.promisc_all);
+
+out:
+	kfree(out);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_promisc);
+
+int mlx5_modify_nic_vport_promisc(struct mlx5_core_dev *mdev,
+				  int promisc_uc,
+				  int promisc_mc,
+				  int promisc_all)
+{
+	void *in;
+	int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in);
+	int err;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in) {
+		mlx5_core_err(mdev, "failed to allocate inbox\n");
+		return -ENOMEM;
+	}
+
+	MLX5_SET(modify_nic_vport_context_in, in, field_select.promisc, 1);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 nic_vport_context.promisc_uc, promisc_uc);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 nic_vport_context.promisc_mc, promisc_mc);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 nic_vport_context.promisc_all, promisc_all);
+
+	err = mlx5_modify_nic_vport_context(mdev, in, inlen);
+
+	kvfree(in);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_promisc);
+
+enum mlx5_vport_roce_state {
+	MLX5_VPORT_ROCE_DISABLED = 0,
+	MLX5_VPORT_ROCE_ENABLED  = 1,
+};
+
+static int mlx5_nic_vport_update_roce_state(struct mlx5_core_dev *mdev,
+					    enum mlx5_vport_roce_state state)
+{
+	void *in;
+	int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in);
+	int err;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in) {
+		mlx5_core_warn(mdev, "failed to allocate inbox\n");
+		return -ENOMEM;
+	}
+
+	MLX5_SET(modify_nic_vport_context_in, in, field_select.roce_en, 1);
+	MLX5_SET(modify_nic_vport_context_in, in, nic_vport_context.roce_en,
+		 state);
+
+	err = mlx5_modify_nic_vport_context(mdev, in, inlen);
+
+	kvfree(in);
+
+	return err;
+}
+
+int mlx5_nic_vport_enable_roce(struct mlx5_core_dev *mdev)
+{
+	return mlx5_nic_vport_update_roce_state(mdev, MLX5_VPORT_ROCE_ENABLED);
+}
+EXPORT_SYMBOL_GPL(mlx5_nic_vport_enable_roce);
+
+int mlx5_nic_vport_disable_roce(struct mlx5_core_dev *mdev)
+{
+	return mlx5_nic_vport_update_roce_state(mdev, MLX5_VPORT_ROCE_DISABLED);
+}
+EXPORT_SYMBOL_GPL(mlx5_nic_vport_disable_roce);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Kconfig b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
index e36e12219..ce26adcb4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
@@ -10,6 +10,14 @@ config MLXSW_CORE
 	  To compile this driver as a module, choose M here: the
 	  module will be called mlxsw_core.
 
+config MLXSW_CORE_HWMON
+	bool "HWMON support for Mellanox Technologies Switch ASICs"
+	depends on MLXSW_CORE && HWMON
+	depends on !(MLXSW_CORE=y && HWMON=m)
+	default y
+	---help---
+	  Say Y here if you want to expose HWMON interface on mlxsw devices.
+
 config MLXSW_PCI
 	tristate "PCI bus implementation for Mellanox Technologies Switch ASICs"
 	depends on PCI && HAS_DMA && HAS_IOMEM && MLXSW_CORE
@@ -33,7 +41,7 @@ config MLXSW_SWITCHX2
 
 config MLXSW_SPECTRUM
 	tristate "Mellanox Technologies Spectrum support"
-	depends on MLXSW_CORE && NET_SWITCHDEV
+	depends on MLXSW_CORE && NET_SWITCHDEV && VLAN_8021Q
 	default m
 	---help---
 	  This driver supports Mellanox Technologies Spectrum Ethernet
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile
index af015818f..584cac444 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Makefile
+++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile
@@ -1,5 +1,6 @@
 obj-$(CONFIG_MLXSW_CORE)	+= mlxsw_core.o
 mlxsw_core-objs			:= core.o
+mlxsw_core-$(CONFIG_MLXSW_CORE_HWMON) += core_hwmon.o
 obj-$(CONFIG_MLXSW_PCI)		+= mlxsw_pci.o
 mlxsw_pci-objs			:= pci.o
 obj-$(CONFIG_MLXSW_SWITCHX2)	+= mlxsw_switchx2.o
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c
index 97f0d93ca..22379eb8e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.c
@@ -105,6 +105,10 @@ struct mlxsw_core {
 		struct debugfs_blob_wrapper vsd_blob;
 		struct debugfs_blob_wrapper psid_blob;
 	} dbg;
+	struct {
+		u8 *mapping; /* lag_id+port_index to local_port mapping */
+	} lag;
+	struct mlxsw_hwmon *hwmon;
 	unsigned long driver_priv[0];
 	/* driver_priv has to be always the last item */
 };
@@ -814,6 +818,17 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 		goto err_alloc_stats;
 	}
 
+	if (mlxsw_driver->profile->used_max_lag &&
+	    mlxsw_driver->profile->used_max_port_per_lag) {
+		alloc_size = sizeof(u8) * mlxsw_driver->profile->max_lag *
+			     mlxsw_driver->profile->max_port_per_lag;
+		mlxsw_core->lag.mapping = kzalloc(alloc_size, GFP_KERNEL);
+		if (!mlxsw_core->lag.mapping) {
+			err = -ENOMEM;
+			goto err_alloc_lag_mapping;
+		}
+	}
+
 	err = mlxsw_bus->init(bus_priv, mlxsw_core, mlxsw_driver->profile);
 	if (err)
 		goto err_bus_init;
@@ -822,6 +837,10 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 	if (err)
 		goto err_emad_init;
 
+	err = mlxsw_hwmon_init(mlxsw_core, mlxsw_bus_info, &mlxsw_core->hwmon);
+	if (err)
+		goto err_hwmon_init;
+
 	err = mlxsw_driver->init(mlxsw_core->driver_priv, mlxsw_core,
 				 mlxsw_bus_info);
 	if (err)
@@ -836,10 +855,13 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 err_debugfs_init:
 	mlxsw_core->driver->fini(mlxsw_core->driver_priv);
 err_driver_init:
+err_hwmon_init:
 	mlxsw_emad_fini(mlxsw_core);
 err_emad_init:
 	mlxsw_bus->fini(bus_priv);
 err_bus_init:
+	kfree(mlxsw_core->lag.mapping);
+err_alloc_lag_mapping:
 	free_percpu(mlxsw_core->pcpu_stats);
 err_alloc_stats:
 	kfree(mlxsw_core);
@@ -857,6 +879,7 @@ void mlxsw_core_bus_device_unregister(struct mlxsw_core *mlxsw_core)
 	mlxsw_core->driver->fini(mlxsw_core->driver_priv);
 	mlxsw_emad_fini(mlxsw_core);
 	mlxsw_core->bus->fini(mlxsw_core->bus_priv);
+	kfree(mlxsw_core->lag.mapping);
 	free_percpu(mlxsw_core->pcpu_stats);
 	kfree(mlxsw_core);
 	mlxsw_core_driver_put(device_kind);
@@ -1188,11 +1211,25 @@ void mlxsw_core_skb_receive(struct mlxsw_core *mlxsw_core, struct sk_buff *skb,
 	struct mlxsw_rx_listener_item *rxl_item;
 	const struct mlxsw_rx_listener *rxl;
 	struct mlxsw_core_pcpu_stats *pcpu_stats;
-	u8 local_port = rx_info->sys_port;
+	u8 local_port;
 	bool found = false;
 
-	dev_dbg_ratelimited(mlxsw_core->bus_info->dev, "%s: sys_port = %d, trap_id = 0x%x\n",
-			    __func__, rx_info->sys_port, rx_info->trap_id);
+	if (rx_info->is_lag) {
+		dev_dbg_ratelimited(mlxsw_core->bus_info->dev, "%s: lag_id = %d, lag_port_index = 0x%x\n",
+				    __func__, rx_info->u.lag_id,
+				    rx_info->trap_id);
+		/* Upper layer does not care if the skb came from LAG or not,
+		 * so just get the local_port for the lag port and push it up.
+		 */
+		local_port = mlxsw_core_lag_mapping_get(mlxsw_core,
+							rx_info->u.lag_id,
+							rx_info->lag_port_index);
+	} else {
+		local_port = rx_info->u.sys_port;
+	}
+
+	dev_dbg_ratelimited(mlxsw_core->bus_info->dev, "%s: local_port = %d, trap_id = 0x%x\n",
+			    __func__, local_port, rx_info->trap_id);
 
 	if ((rx_info->trap_id >= MLXSW_TRAP_ID_MAX) ||
 	    (local_port >= MLXSW_PORT_MAX_PORTS))
@@ -1236,6 +1273,48 @@ drop:
 }
 EXPORT_SYMBOL(mlxsw_core_skb_receive);
 
+static int mlxsw_core_lag_mapping_index(struct mlxsw_core *mlxsw_core,
+					u16 lag_id, u8 port_index)
+{
+	return mlxsw_core->driver->profile->max_port_per_lag * lag_id +
+	       port_index;
+}
+
+void mlxsw_core_lag_mapping_set(struct mlxsw_core *mlxsw_core,
+				u16 lag_id, u8 port_index, u8 local_port)
+{
+	int index = mlxsw_core_lag_mapping_index(mlxsw_core,
+						 lag_id, port_index);
+
+	mlxsw_core->lag.mapping[index] = local_port;
+}
+EXPORT_SYMBOL(mlxsw_core_lag_mapping_set);
+
+u8 mlxsw_core_lag_mapping_get(struct mlxsw_core *mlxsw_core,
+			      u16 lag_id, u8 port_index)
+{
+	int index = mlxsw_core_lag_mapping_index(mlxsw_core,
+						 lag_id, port_index);
+
+	return mlxsw_core->lag.mapping[index];
+}
+EXPORT_SYMBOL(mlxsw_core_lag_mapping_get);
+
+void mlxsw_core_lag_mapping_clear(struct mlxsw_core *mlxsw_core,
+				  u16 lag_id, u8 local_port)
+{
+	int i;
+
+	for (i = 0; i < mlxsw_core->driver->profile->max_port_per_lag; i++) {
+		int index = mlxsw_core_lag_mapping_index(mlxsw_core,
+							 lag_id, i);
+
+		if (mlxsw_core->lag.mapping[index] == local_port)
+			mlxsw_core->lag.mapping[index] = 0;
+	}
+}
+EXPORT_SYMBOL(mlxsw_core_lag_mapping_clear);
+
 int mlxsw_cmd_exec(struct mlxsw_core *mlxsw_core, u16 opcode, u8 opcode_mod,
 		   u32 in_mod, bool out_mbox_direct,
 		   char *in_mbox, size_t in_mbox_size,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h
index 807827350..a01723600 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.h
@@ -112,13 +112,25 @@ int mlxsw_reg_write(struct mlxsw_core *mlxsw_core,
 		    const struct mlxsw_reg_info *reg, char *payload);
 
 struct mlxsw_rx_info {
-	u16 sys_port;
+	bool is_lag;
+	union {
+		u16 sys_port;
+		u16 lag_id;
+	} u;
+	u8 lag_port_index;
 	int trap_id;
 };
 
 void mlxsw_core_skb_receive(struct mlxsw_core *mlxsw_core, struct sk_buff *skb,
 			    struct mlxsw_rx_info *rx_info);
 
+void mlxsw_core_lag_mapping_set(struct mlxsw_core *mlxsw_core,
+				u16 lag_id, u8 port_index, u8 local_port);
+u8 mlxsw_core_lag_mapping_get(struct mlxsw_core *mlxsw_core,
+			      u16 lag_id, u8 port_index);
+void mlxsw_core_lag_mapping_clear(struct mlxsw_core *mlxsw_core,
+				  u16 lag_id, u8 local_port);
+
 #define MLXSW_CONFIG_PROFILE_SWID_COUNT 8
 
 struct mlxsw_swid_config {
@@ -209,4 +221,24 @@ struct mlxsw_bus_info {
 	u8 psid[MLXSW_CMD_BOARDINFO_PSID_LEN];
 };
 
+struct mlxsw_hwmon;
+
+#ifdef CONFIG_MLXSW_CORE_HWMON
+
+int mlxsw_hwmon_init(struct mlxsw_core *mlxsw_core,
+		     const struct mlxsw_bus_info *mlxsw_bus_info,
+		     struct mlxsw_hwmon **p_hwmon);
+void mlxsw_hwmon_fini(struct mlxsw_hwmon *mlxsw_hwmon);
+
+#else
+
+static inline int mlxsw_hwmon_init(struct mlxsw_core *mlxsw_core,
+				   const struct mlxsw_bus_info *mlxsw_bus_info,
+				   struct mlxsw_hwmon **p_hwmon)
+{
+	return 0;
+}
+
+#endif
+
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
new file mode 100644
index 000000000..1ac8bf187
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
@@ -0,0 +1,372 @@
+/*
+ * drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
+ * Copyright (c) 2015 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2015 Jiri Pirko <jiri@mellanox.com>
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ *    contributors may be used to endorse or promote products derived from
+ *    this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/sysfs.h>
+#include <linux/hwmon.h>
+#include <linux/err.h>
+
+#include "core.h"
+
+#define MLXSW_HWMON_TEMP_SENSOR_MAX_COUNT 127
+#define MLXSW_HWMON_ATTR_COUNT (MLXSW_HWMON_TEMP_SENSOR_MAX_COUNT * 4 + \
+				MLXSW_MFCR_TACHOS_MAX + MLXSW_MFCR_PWMS_MAX)
+
+struct mlxsw_hwmon_attr {
+	struct device_attribute dev_attr;
+	struct mlxsw_hwmon *hwmon;
+	unsigned int type_index;
+	char name[32];
+};
+
+struct mlxsw_hwmon {
+	struct mlxsw_core *core;
+	const struct mlxsw_bus_info *bus_info;
+	struct device *hwmon_dev;
+	struct attribute_group group;
+	const struct attribute_group *groups[2];
+	struct attribute *attrs[MLXSW_HWMON_ATTR_COUNT + 1];
+	struct mlxsw_hwmon_attr hwmon_attrs[MLXSW_HWMON_ATTR_COUNT];
+	unsigned int attrs_count;
+};
+
+static ssize_t mlxsw_hwmon_temp_show(struct device *dev,
+				     struct device_attribute *attr,
+				     char *buf)
+{
+	struct mlxsw_hwmon_attr *mlwsw_hwmon_attr =
+			container_of(attr, struct mlxsw_hwmon_attr, dev_attr);
+	struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon;
+	char mtmp_pl[MLXSW_REG_MTMP_LEN];
+	unsigned int temp;
+	int err;
+
+	mlxsw_reg_mtmp_pack(mtmp_pl, mlwsw_hwmon_attr->type_index,
+			    false, false);
+	err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mtmp), mtmp_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to query temp sensor\n");
+		return err;
+	}
+	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL);
+	return sprintf(buf, "%u\n", temp);
+}
+
+static ssize_t mlxsw_hwmon_temp_max_show(struct device *dev,
+					 struct device_attribute *attr,
+					 char *buf)
+{
+	struct mlxsw_hwmon_attr *mlwsw_hwmon_attr =
+			container_of(attr, struct mlxsw_hwmon_attr, dev_attr);
+	struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon;
+	char mtmp_pl[MLXSW_REG_MTMP_LEN];
+	unsigned int temp_max;
+	int err;
+
+	mlxsw_reg_mtmp_pack(mtmp_pl, mlwsw_hwmon_attr->type_index,
+			    false, false);
+	err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mtmp), mtmp_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to query temp sensor\n");
+		return err;
+	}
+	mlxsw_reg_mtmp_unpack(mtmp_pl, NULL, &temp_max, NULL);
+	return sprintf(buf, "%u\n", temp_max);
+}
+
+static ssize_t mlxsw_hwmon_temp_rst_store(struct device *dev,
+					  struct device_attribute *attr,
+					  const char *buf, size_t len)
+{
+	struct mlxsw_hwmon_attr *mlwsw_hwmon_attr =
+			container_of(attr, struct mlxsw_hwmon_attr, dev_attr);
+	struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon;
+	char mtmp_pl[MLXSW_REG_MTMP_LEN];
+	unsigned long val;
+	int err;
+
+	err = kstrtoul(buf, 10, &val);
+	if (err)
+		return err;
+	if (val != 1)
+		return -EINVAL;
+
+	mlxsw_reg_mtmp_pack(mtmp_pl, mlwsw_hwmon_attr->type_index, true, true);
+	err = mlxsw_reg_write(mlxsw_hwmon->core, MLXSW_REG(mtmp), mtmp_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to reset temp sensor history\n");
+		return err;
+	}
+	return len;
+}
+
+static ssize_t mlxsw_hwmon_fan_rpm_show(struct device *dev,
+					struct device_attribute *attr,
+					char *buf)
+{
+	struct mlxsw_hwmon_attr *mlwsw_hwmon_attr =
+			container_of(attr, struct mlxsw_hwmon_attr, dev_attr);
+	struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon;
+	char mfsm_pl[MLXSW_REG_MFSM_LEN];
+	int err;
+
+	mlxsw_reg_mfsm_pack(mfsm_pl, mlwsw_hwmon_attr->type_index);
+	err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mfsm), mfsm_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to query fan\n");
+		return err;
+	}
+	return sprintf(buf, "%u\n", mlxsw_reg_mfsm_rpm_get(mfsm_pl));
+}
+
+static ssize_t mlxsw_hwmon_pwm_show(struct device *dev,
+				    struct device_attribute *attr,
+				    char *buf)
+{
+	struct mlxsw_hwmon_attr *mlwsw_hwmon_attr =
+			container_of(attr, struct mlxsw_hwmon_attr, dev_attr);
+	struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon;
+	char mfsc_pl[MLXSW_REG_MFSC_LEN];
+	int err;
+
+	mlxsw_reg_mfsc_pack(mfsc_pl, mlwsw_hwmon_attr->type_index, 0);
+	err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mfsc), mfsc_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to query PWM\n");
+		return err;
+	}
+	return sprintf(buf, "%u\n",
+		       mlxsw_reg_mfsc_pwm_duty_cycle_get(mfsc_pl));
+}
+
+static ssize_t mlxsw_hwmon_pwm_store(struct device *dev,
+				     struct device_attribute *attr,
+				     const char *buf, size_t len)
+{
+	struct mlxsw_hwmon_attr *mlwsw_hwmon_attr =
+			container_of(attr, struct mlxsw_hwmon_attr, dev_attr);
+	struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon;
+	char mfsc_pl[MLXSW_REG_MFSC_LEN];
+	unsigned long val;
+	int err;
+
+	err = kstrtoul(buf, 10, &val);
+	if (err)
+		return err;
+	if (val > 255)
+		return -EINVAL;
+
+	mlxsw_reg_mfsc_pack(mfsc_pl, mlwsw_hwmon_attr->type_index, val);
+	err = mlxsw_reg_write(mlxsw_hwmon->core, MLXSW_REG(mfsc), mfsc_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to write PWM\n");
+		return err;
+	}
+	return len;
+}
+
+enum mlxsw_hwmon_attr_type {
+	MLXSW_HWMON_ATTR_TYPE_TEMP,
+	MLXSW_HWMON_ATTR_TYPE_TEMP_MAX,
+	MLXSW_HWMON_ATTR_TYPE_TEMP_RST,
+	MLXSW_HWMON_ATTR_TYPE_FAN_RPM,
+	MLXSW_HWMON_ATTR_TYPE_PWM,
+};
+
+static void mlxsw_hwmon_attr_add(struct mlxsw_hwmon *mlxsw_hwmon,
+				 enum mlxsw_hwmon_attr_type attr_type,
+				 unsigned int type_index, unsigned int num) {
+	struct mlxsw_hwmon_attr *mlxsw_hwmon_attr;
+	unsigned int attr_index;
+
+	attr_index = mlxsw_hwmon->attrs_count;
+	mlxsw_hwmon_attr = &mlxsw_hwmon->hwmon_attrs[attr_index];
+
+	switch (attr_type) {
+	case MLXSW_HWMON_ATTR_TYPE_TEMP:
+		mlxsw_hwmon_attr->dev_attr.show = mlxsw_hwmon_temp_show;
+		mlxsw_hwmon_attr->dev_attr.attr.mode = S_IRUGO;
+		snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name),
+			 "temp%u_input", num + 1);
+		break;
+	case MLXSW_HWMON_ATTR_TYPE_TEMP_MAX:
+		mlxsw_hwmon_attr->dev_attr.show = mlxsw_hwmon_temp_max_show;
+		mlxsw_hwmon_attr->dev_attr.attr.mode = S_IRUGO;
+		snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name),
+			 "temp%u_highest", num + 1);
+		break;
+	case MLXSW_HWMON_ATTR_TYPE_TEMP_RST:
+		mlxsw_hwmon_attr->dev_attr.store = mlxsw_hwmon_temp_rst_store;
+		mlxsw_hwmon_attr->dev_attr.attr.mode = S_IWUSR;
+		snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name),
+			 "temp%u_reset_history", num + 1);
+		break;
+	case MLXSW_HWMON_ATTR_TYPE_FAN_RPM:
+		mlxsw_hwmon_attr->dev_attr.show = mlxsw_hwmon_fan_rpm_show;
+		mlxsw_hwmon_attr->dev_attr.attr.mode = S_IRUGO;
+		snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name),
+			 "fan%u_input", num + 1);
+		break;
+	case MLXSW_HWMON_ATTR_TYPE_PWM:
+		mlxsw_hwmon_attr->dev_attr.show = mlxsw_hwmon_pwm_show;
+		mlxsw_hwmon_attr->dev_attr.store = mlxsw_hwmon_pwm_store;
+		mlxsw_hwmon_attr->dev_attr.attr.mode = S_IWUSR | S_IRUGO;
+		snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name),
+			 "pwm%u", num + 1);
+		break;
+	default:
+		WARN_ON(1);
+	}
+
+	mlxsw_hwmon_attr->type_index = type_index;
+	mlxsw_hwmon_attr->hwmon = mlxsw_hwmon;
+	mlxsw_hwmon_attr->dev_attr.attr.name = mlxsw_hwmon_attr->name;
+	sysfs_attr_init(&mlxsw_hwmon_attr->dev_attr.attr);
+
+	mlxsw_hwmon->attrs[attr_index] = &mlxsw_hwmon_attr->dev_attr.attr;
+	mlxsw_hwmon->attrs_count++;
+}
+
+static int mlxsw_hwmon_temp_init(struct mlxsw_hwmon *mlxsw_hwmon)
+{
+	char mtcap_pl[MLXSW_REG_MTCAP_LEN];
+	char mtmp_pl[MLXSW_REG_MTMP_LEN];
+	u8 sensor_count;
+	int i;
+	int err;
+
+	err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mtcap), mtcap_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to get number of temp sensors\n");
+		return err;
+	}
+	sensor_count = mlxsw_reg_mtcap_sensor_count_get(mtcap_pl);
+	for (i = 0; i < sensor_count; i++) {
+		mlxsw_reg_mtmp_pack(mtmp_pl, i, true, true);
+		err = mlxsw_reg_write(mlxsw_hwmon->core,
+				      MLXSW_REG(mtmp), mtmp_pl);
+		if (err) {
+			dev_err(mlxsw_hwmon->bus_info->dev, "Failed to setup temp sensor number %d\n",
+				i);
+			return err;
+		}
+		mlxsw_hwmon_attr_add(mlxsw_hwmon,
+				     MLXSW_HWMON_ATTR_TYPE_TEMP, i, i);
+		mlxsw_hwmon_attr_add(mlxsw_hwmon,
+				     MLXSW_HWMON_ATTR_TYPE_TEMP_MAX, i, i);
+		mlxsw_hwmon_attr_add(mlxsw_hwmon,
+				     MLXSW_HWMON_ATTR_TYPE_TEMP_RST, i, i);
+	}
+	return 0;
+}
+
+static int mlxsw_hwmon_fans_init(struct mlxsw_hwmon *mlxsw_hwmon)
+{
+	char mfcr_pl[MLXSW_REG_MFCR_LEN];
+	enum mlxsw_reg_mfcr_pwm_frequency freq;
+	unsigned int type_index;
+	unsigned int num;
+	u16 tacho_active;
+	u8 pwm_active;
+	int err;
+
+	err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mfcr), mfcr_pl);
+	if (err) {
+		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to get to probe PWMs and Tachometers\n");
+		return err;
+	}
+	mlxsw_reg_mfcr_unpack(mfcr_pl, &freq, &tacho_active, &pwm_active);
+	num = 0;
+	for (type_index = 0; type_index < MLXSW_MFCR_TACHOS_MAX; type_index++) {
+		if (tacho_active & BIT(type_index))
+			mlxsw_hwmon_attr_add(mlxsw_hwmon,
+					     MLXSW_HWMON_ATTR_TYPE_FAN_RPM,
+					     type_index, num++);
+	}
+	num = 0;
+	for (type_index = 0; type_index < MLXSW_MFCR_PWMS_MAX; type_index++) {
+		if (pwm_active & BIT(type_index))
+			mlxsw_hwmon_attr_add(mlxsw_hwmon,
+					     MLXSW_HWMON_ATTR_TYPE_PWM,
+					     type_index, num++);
+	}
+	return 0;
+}
+
+int mlxsw_hwmon_init(struct mlxsw_core *mlxsw_core,
+		     const struct mlxsw_bus_info *mlxsw_bus_info,
+		     struct mlxsw_hwmon **p_hwmon)
+{
+	struct mlxsw_hwmon *mlxsw_hwmon;
+	struct device *hwmon_dev;
+	int err;
+
+	mlxsw_hwmon = devm_kzalloc(mlxsw_bus_info->dev, sizeof(*mlxsw_hwmon),
+				   GFP_KERNEL);
+	if (!mlxsw_hwmon)
+		return -ENOMEM;
+	mlxsw_hwmon->core = mlxsw_core;
+	mlxsw_hwmon->bus_info = mlxsw_bus_info;
+
+	err = mlxsw_hwmon_temp_init(mlxsw_hwmon);
+	if (err)
+		goto err_temp_init;
+
+	err = mlxsw_hwmon_fans_init(mlxsw_hwmon);
+	if (err)
+		goto err_fans_init;
+
+	mlxsw_hwmon->groups[0] = &mlxsw_hwmon->group;
+	mlxsw_hwmon->group.attrs = mlxsw_hwmon->attrs;
+
+	hwmon_dev = devm_hwmon_device_register_with_groups(mlxsw_bus_info->dev,
+							   "mlxsw",
+							   mlxsw_hwmon,
+							   mlxsw_hwmon->groups);
+	if (IS_ERR(hwmon_dev)) {
+		err = PTR_ERR(hwmon_dev);
+		goto err_hwmon_register;
+	}
+
+	mlxsw_hwmon->hwmon_dev = hwmon_dev;
+	*p_hwmon = mlxsw_hwmon;
+	return 0;
+
+err_hwmon_register:
+err_fans_init:
+err_temp_init:
+	return err;
+}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index de69e719d..7992c553c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -215,7 +215,7 @@ mlxsw_pci_queue_elem_info_producer_get(struct mlxsw_pci_queue *q)
 {
 	int index = q->producer_counter & (q->count - 1);
 
-	if ((q->producer_counter - q->consumer_counter) == q->count)
+	if ((u16) (q->producer_counter - q->consumer_counter) == q->count)
 		return NULL;
 	return mlxsw_pci_queue_elem_info_get(q, index);
 }
@@ -384,7 +384,7 @@ static int mlxsw_pci_sdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 
 	/* Set CQ of same number of this SDQ. */
 	mlxsw_cmd_mbox_sw2hw_dq_cq_set(mbox, q->num);
-	mlxsw_cmd_mbox_sw2hw_dq_sdq_tclass_set(mbox, 7);
+	mlxsw_cmd_mbox_sw2hw_dq_sdq_tclass_set(mbox, 3);
 	mlxsw_cmd_mbox_sw2hw_dq_log2_dq_sz_set(mbox, 3); /* 8 pages */
 	for (i = 0; i < MLXSW_PCI_AQ_PAGES; i++) {
 		dma_addr_t mapaddr = __mlxsw_pci_queue_page_get(q, i);
@@ -686,11 +686,15 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci,
 	if (q->consumer_counter++ != consumer_counter_limit)
 		dev_dbg_ratelimited(&pdev->dev, "Consumer counter does not match limit in RDQ\n");
 
-	/* We do not support lag now */
-	if (mlxsw_pci_cqe_lag_get(cqe))
-		goto drop;
+	if (mlxsw_pci_cqe_lag_get(cqe)) {
+		rx_info.is_lag = true;
+		rx_info.u.lag_id = mlxsw_pci_cqe_lag_id_get(cqe);
+		rx_info.lag_port_index = mlxsw_pci_cqe_lag_port_index_get(cqe);
+	} else {
+		rx_info.is_lag = false;
+		rx_info.u.sys_port = mlxsw_pci_cqe_system_port_get(cqe);
+	}
 
-	rx_info.sys_port = mlxsw_pci_cqe_system_port_get(cqe);
 	rx_info.trap_id = mlxsw_pci_cqe_trap_id_get(cqe);
 
 	byte_count = mlxsw_pci_cqe_byte_count_get(cqe);
@@ -699,7 +703,6 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci,
 	skb_put(skb, byte_count);
 	mlxsw_core_skb_receive(mlxsw_pci->core, skb, &rx_info);
 
-put_new_skb:
 	memset(wqe, 0, q->elem_size);
 	err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info);
 	if (err)
@@ -708,10 +711,6 @@ put_new_skb:
 	q->producer_counter++;
 	mlxsw_pci_queue_doorbell_producer_ring(mlxsw_pci, q);
 	return;
-
-drop:
-	dev_kfree_skb_any(skb);
-	goto put_new_skb;
 }
 
 static char *mlxsw_pci_cq_sw_cqe_get(struct mlxsw_pci_queue *q)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.h b/drivers/net/ethernet/mellanox/mlxsw/pci.h
index 142f33d97..912106054 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.h
@@ -129,13 +129,15 @@ MLXSW_ITEM64_INDEXED(pci, wqe, address, 0x08, 0, 64, 0x8, 0x0, false);
  */
 MLXSW_ITEM32(pci, cqe, lag, 0x00, 23, 1);
 
-/* pci_cqe_system_port
+/* pci_cqe_system_port/lag_id
  * When lag=0: System port on which the packet was received
  * When lag=1:
  * bits [15:4] LAG ID on which the packet was received
  * bits [3:0] sub_port on which the packet was received
  */
 MLXSW_ITEM32(pci, cqe, system_port, 0x00, 0, 16);
+MLXSW_ITEM32(pci, cqe, lag_id, 0x00, 4, 12);
+MLXSW_ITEM32(pci, cqe, lag_port_index, 0x00, 0, 4);
 
 /* pci_cqe_wqe_counter
  * WQE count of the WQEs completed on the associated dqn
diff --git a/drivers/net/ethernet/mellanox/mlxsw/port.h b/drivers/net/ethernet/mellanox/mlxsw/port.h
index 726f5435b..ae65b9940 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/port.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/port.h
@@ -49,7 +49,7 @@
 #define MLXSW_PORT_MID			0xd000
 
 #define MLXSW_PORT_MAX_PHY_PORTS	0x40
-#define MLXSW_PORT_MAX_PORTS		MLXSW_PORT_MAX_PHY_PORTS
+#define MLXSW_PORT_MAX_PORTS		(MLXSW_PORT_MAX_PHY_PORTS + 1)
 
 #define MLXSW_PORT_DEVID_BITS_OFFSET	10
 #define MLXSW_PORT_PHY_BITS_OFFSET	4
diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 236fb5d2a..ffe4c0305 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -99,6 +99,55 @@ static const struct mlxsw_reg_info mlxsw_reg_spad = {
  */
 MLXSW_ITEM_BUF(reg, spad, base_mac, 0x02, 6);
 
+/* SMID - Switch Multicast ID
+ * --------------------------
+ * The MID record maps from a MID (Multicast ID), which is a unique identifier
+ * of the multicast group within the stacking domain, into a list of local
+ * ports into which the packet is replicated.
+ */
+#define MLXSW_REG_SMID_ID 0x2007
+#define MLXSW_REG_SMID_LEN 0x240
+
+static const struct mlxsw_reg_info mlxsw_reg_smid = {
+	.id = MLXSW_REG_SMID_ID,
+	.len = MLXSW_REG_SMID_LEN,
+};
+
+/* reg_smid_swid
+ * Switch partition ID.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, smid, swid, 0x00, 24, 8);
+
+/* reg_smid_mid
+ * Multicast identifier - global identifier that represents the multicast group
+ * across all devices.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, smid, mid, 0x00, 0, 16);
+
+/* reg_smid_port
+ * Local port memebership (1 bit per port).
+ * Access: RW
+ */
+MLXSW_ITEM_BIT_ARRAY(reg, smid, port, 0x20, 0x20, 1);
+
+/* reg_smid_port_mask
+ * Local port mask (1 bit per port).
+ * Access: W
+ */
+MLXSW_ITEM_BIT_ARRAY(reg, smid, port_mask, 0x220, 0x20, 1);
+
+static inline void mlxsw_reg_smid_pack(char *payload, u16 mid,
+				       u8 port, bool set)
+{
+	MLXSW_REG_ZERO(smid, payload);
+	mlxsw_reg_smid_swid_set(payload, 0);
+	mlxsw_reg_smid_mid_set(payload, mid);
+	mlxsw_reg_smid_port_set(payload, port, set);
+	mlxsw_reg_smid_port_mask_set(payload, port, 1);
+}
+
 /* SSPR - Switch System Port Record Register
  * -----------------------------------------
  * Configures the system port to local port mapping.
@@ -286,6 +335,8 @@ MLXSW_ITEM32_INDEXED(reg, sfd, rec_swid, MLXSW_REG_SFD_BASE_LEN, 24, 8,
 
 enum mlxsw_reg_sfd_rec_type {
 	MLXSW_REG_SFD_REC_TYPE_UNICAST = 0x0,
+	MLXSW_REG_SFD_REC_TYPE_UNICAST_LAG = 0x1,
+	MLXSW_REG_SFD_REC_TYPE_MULTICAST = 0x2,
 };
 
 /* reg_sfd_rec_type
@@ -376,36 +427,144 @@ MLXSW_ITEM32_INDEXED(reg, sfd, uc_fid_vid, MLXSW_REG_SFD_BASE_LEN, 0, 16,
 MLXSW_ITEM32_INDEXED(reg, sfd, uc_system_port, MLXSW_REG_SFD_BASE_LEN, 0, 16,
 		     MLXSW_REG_SFD_REC_LEN, 0x0C, false);
 
-static inline void mlxsw_reg_sfd_uc_pack(char *payload, int rec_index,
-					 enum mlxsw_reg_sfd_rec_policy policy,
-					 const char *mac, u16 vid,
-					 enum mlxsw_reg_sfd_rec_action action,
-					 u8 local_port)
+static inline void mlxsw_reg_sfd_rec_pack(char *payload, int rec_index,
+					  enum mlxsw_reg_sfd_rec_type rec_type,
+					  const char *mac,
+					  enum mlxsw_reg_sfd_rec_action action)
 {
 	u8 num_rec = mlxsw_reg_sfd_num_rec_get(payload);
 
 	if (rec_index >= num_rec)
 		mlxsw_reg_sfd_num_rec_set(payload, rec_index + 1);
 	mlxsw_reg_sfd_rec_swid_set(payload, rec_index, 0);
-	mlxsw_reg_sfd_rec_type_set(payload, rec_index,
-				   MLXSW_REG_SFD_REC_TYPE_UNICAST);
-	mlxsw_reg_sfd_rec_policy_set(payload, rec_index, policy);
+	mlxsw_reg_sfd_rec_type_set(payload, rec_index, rec_type);
 	mlxsw_reg_sfd_rec_mac_memcpy_to(payload, rec_index, mac);
-	mlxsw_reg_sfd_uc_sub_port_set(payload, rec_index, 0);
-	mlxsw_reg_sfd_uc_fid_vid_set(payload, rec_index, vid);
 	mlxsw_reg_sfd_rec_action_set(payload, rec_index, action);
+}
+
+static inline void mlxsw_reg_sfd_uc_pack(char *payload, int rec_index,
+					 enum mlxsw_reg_sfd_rec_policy policy,
+					 const char *mac, u16 fid_vid,
+					 enum mlxsw_reg_sfd_rec_action action,
+					 u8 local_port)
+{
+	mlxsw_reg_sfd_rec_pack(payload, rec_index,
+			       MLXSW_REG_SFD_REC_TYPE_UNICAST, mac, action);
+	mlxsw_reg_sfd_rec_policy_set(payload, rec_index, policy);
+	mlxsw_reg_sfd_uc_sub_port_set(payload, rec_index, 0);
+	mlxsw_reg_sfd_uc_fid_vid_set(payload, rec_index, fid_vid);
 	mlxsw_reg_sfd_uc_system_port_set(payload, rec_index, local_port);
 }
 
 static inline void mlxsw_reg_sfd_uc_unpack(char *payload, int rec_index,
-					   char *mac, u16 *p_vid,
+					   char *mac, u16 *p_fid_vid,
 					   u8 *p_local_port)
 {
 	mlxsw_reg_sfd_rec_mac_memcpy_from(payload, rec_index, mac);
-	*p_vid = mlxsw_reg_sfd_uc_fid_vid_get(payload, rec_index);
+	*p_fid_vid = mlxsw_reg_sfd_uc_fid_vid_get(payload, rec_index);
 	*p_local_port = mlxsw_reg_sfd_uc_system_port_get(payload, rec_index);
 }
 
+/* reg_sfd_uc_lag_sub_port
+ * LAG sub port.
+ * Must be 0 if multichannel VEPA is not enabled.
+ * Access: RW
+ */
+MLXSW_ITEM32_INDEXED(reg, sfd, uc_lag_sub_port, MLXSW_REG_SFD_BASE_LEN, 16, 8,
+		     MLXSW_REG_SFD_REC_LEN, 0x08, false);
+
+/* reg_sfd_uc_lag_fid_vid
+ * Filtering ID or VLAN ID
+ * For SwitchX and SwitchX-2:
+ * - Dynamic entries (policy 2,3) use FID
+ * - Static entries (policy 0) use VID
+ * - When independent learning is configured, VID=FID
+ * For Spectrum: use FID for both Dynamic and Static entries.
+ * VID should not be used.
+ * Access: Index
+ */
+MLXSW_ITEM32_INDEXED(reg, sfd, uc_lag_fid_vid, MLXSW_REG_SFD_BASE_LEN, 0, 16,
+		     MLXSW_REG_SFD_REC_LEN, 0x08, false);
+
+/* reg_sfd_uc_lag_lag_vid
+ * Indicates VID in case of vFIDs. Reserved for FIDs.
+ * Access: RW
+ */
+MLXSW_ITEM32_INDEXED(reg, sfd, uc_lag_lag_vid, MLXSW_REG_SFD_BASE_LEN, 16, 12,
+		     MLXSW_REG_SFD_REC_LEN, 0x0C, false);
+
+/* reg_sfd_uc_lag_lag_id
+ * LAG Identifier - pointer into the LAG descriptor table.
+ * Access: RW
+ */
+MLXSW_ITEM32_INDEXED(reg, sfd, uc_lag_lag_id, MLXSW_REG_SFD_BASE_LEN, 0, 10,
+		     MLXSW_REG_SFD_REC_LEN, 0x0C, false);
+
+static inline void
+mlxsw_reg_sfd_uc_lag_pack(char *payload, int rec_index,
+			  enum mlxsw_reg_sfd_rec_policy policy,
+			  const char *mac, u16 fid_vid,
+			  enum mlxsw_reg_sfd_rec_action action, u16 lag_vid,
+			  u16 lag_id)
+{
+	mlxsw_reg_sfd_rec_pack(payload, rec_index,
+			       MLXSW_REG_SFD_REC_TYPE_UNICAST_LAG,
+			       mac, action);
+	mlxsw_reg_sfd_rec_policy_set(payload, rec_index, policy);
+	mlxsw_reg_sfd_uc_lag_sub_port_set(payload, rec_index, 0);
+	mlxsw_reg_sfd_uc_lag_fid_vid_set(payload, rec_index, fid_vid);
+	mlxsw_reg_sfd_uc_lag_lag_vid_set(payload, rec_index, lag_vid);
+	mlxsw_reg_sfd_uc_lag_lag_id_set(payload, rec_index, lag_id);
+}
+
+static inline void mlxsw_reg_sfd_uc_lag_unpack(char *payload, int rec_index,
+					       char *mac, u16 *p_vid,
+					       u16 *p_lag_id)
+{
+	mlxsw_reg_sfd_rec_mac_memcpy_from(payload, rec_index, mac);
+	*p_vid = mlxsw_reg_sfd_uc_lag_fid_vid_get(payload, rec_index);
+	*p_lag_id = mlxsw_reg_sfd_uc_lag_lag_id_get(payload, rec_index);
+}
+
+/* reg_sfd_mc_pgi
+ *
+ * Multicast port group index - index into the port group table.
+ * Value 0x1FFF indicates the pgi should point to the MID entry.
+ * For Spectrum this value must be set to 0x1FFF
+ * Access: RW
+ */
+MLXSW_ITEM32_INDEXED(reg, sfd, mc_pgi, MLXSW_REG_SFD_BASE_LEN, 16, 13,
+		     MLXSW_REG_SFD_REC_LEN, 0x08, false);
+
+/* reg_sfd_mc_fid_vid
+ *
+ * Filtering ID or VLAN ID
+ * Access: Index
+ */
+MLXSW_ITEM32_INDEXED(reg, sfd, mc_fid_vid, MLXSW_REG_SFD_BASE_LEN, 0, 16,
+		     MLXSW_REG_SFD_REC_LEN, 0x08, false);
+
+/* reg_sfd_mc_mid
+ *
+ * Multicast identifier - global identifier that represents the multicast
+ * group across all devices.
+ * Access: RW
+ */
+MLXSW_ITEM32_INDEXED(reg, sfd, mc_mid, MLXSW_REG_SFD_BASE_LEN, 0, 16,
+		     MLXSW_REG_SFD_REC_LEN, 0x0C, false);
+
+static inline void
+mlxsw_reg_sfd_mc_pack(char *payload, int rec_index,
+		      const char *mac, u16 fid_vid,
+		      enum mlxsw_reg_sfd_rec_action action, u16 mid)
+{
+	mlxsw_reg_sfd_rec_pack(payload, rec_index,
+			       MLXSW_REG_SFD_REC_TYPE_MULTICAST, mac, action);
+	mlxsw_reg_sfd_mc_pgi_set(payload, rec_index, 0x1FFF);
+	mlxsw_reg_sfd_mc_fid_vid_set(payload, rec_index, fid_vid);
+	mlxsw_reg_sfd_mc_mid_set(payload, rec_index, mid);
+}
+
 /* SFN - Switch FDB Notification Register
  * -------------------------------------------
  * The switch provides notifications on newly learned FDB entries and
@@ -456,8 +615,12 @@ MLXSW_ITEM32_INDEXED(reg, sfn, rec_swid, MLXSW_REG_SFN_BASE_LEN, 24, 8,
 enum mlxsw_reg_sfn_rec_type {
 	/* MAC addresses learned on a regular port. */
 	MLXSW_REG_SFN_REC_TYPE_LEARNED_MAC = 0x5,
-	/* Aged-out MAC address on a regular port */
+	/* MAC addresses learned on a LAG port. */
+	MLXSW_REG_SFN_REC_TYPE_LEARNED_MAC_LAG = 0x6,
+	/* Aged-out MAC address on a regular port. */
 	MLXSW_REG_SFN_REC_TYPE_AGED_OUT_MAC = 0x7,
+	/* Aged-out MAC address on a LAG port. */
+	MLXSW_REG_SFN_REC_TYPE_AGED_OUT_MAC_LAG = 0x8,
 };
 
 /* reg_sfn_rec_type
@@ -505,6 +668,22 @@ static inline void mlxsw_reg_sfn_mac_unpack(char *payload, int rec_index,
 	*p_local_port = mlxsw_reg_sfn_mac_system_port_get(payload, rec_index);
 }
 
+/* reg_sfn_mac_lag_lag_id
+ * LAG ID (pointer into the LAG descriptor table).
+ * Access: RO
+ */
+MLXSW_ITEM32_INDEXED(reg, sfn, mac_lag_lag_id, MLXSW_REG_SFN_BASE_LEN, 0, 10,
+		     MLXSW_REG_SFN_REC_LEN, 0x0C, false);
+
+static inline void mlxsw_reg_sfn_mac_lag_unpack(char *payload, int rec_index,
+						char *mac, u16 *p_vid,
+						u16 *p_lag_id)
+{
+	mlxsw_reg_sfn_rec_mac_memcpy_from(payload, rec_index, mac);
+	*p_vid = mlxsw_reg_sfn_mac_fid_get(payload, rec_index);
+	*p_lag_id = mlxsw_reg_sfn_mac_lag_lag_id_get(payload, rec_index);
+}
+
 /* SPMS - Switch Port MSTP/RSTP State Register
  * -------------------------------------------
  * Configures the spanning tree state of a physical port.
@@ -694,6 +873,62 @@ static inline void mlxsw_reg_spvm_pack(char *payload, u8 local_port,
 	}
 }
 
+/* SPAFT - Switch Port Acceptable Frame Types
+ * ------------------------------------------
+ * The Switch Port Acceptable Frame Types register configures the frame
+ * admittance of the port.
+ */
+#define MLXSW_REG_SPAFT_ID 0x2010
+#define MLXSW_REG_SPAFT_LEN 0x08
+
+static const struct mlxsw_reg_info mlxsw_reg_spaft = {
+	.id = MLXSW_REG_SPAFT_ID,
+	.len = MLXSW_REG_SPAFT_LEN,
+};
+
+/* reg_spaft_local_port
+ * Local port number.
+ * Access: Index
+ *
+ * Note: CPU port is not supported (all tag types are allowed).
+ */
+MLXSW_ITEM32(reg, spaft, local_port, 0x00, 16, 8);
+
+/* reg_spaft_sub_port
+ * Virtual port within the physical port.
+ * Should be set to 0 when virtual ports are not enabled on the port.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, sub_port, 0x00, 8, 8);
+
+/* reg_spaft_allow_untagged
+ * When set, untagged frames on the ingress are allowed (default).
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, allow_untagged, 0x04, 31, 1);
+
+/* reg_spaft_allow_prio_tagged
+ * When set, priority tagged frames on the ingress are allowed (default).
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, allow_prio_tagged, 0x04, 30, 1);
+
+/* reg_spaft_allow_tagged
+ * When set, tagged frames on the ingress are allowed (default).
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, allow_tagged, 0x04, 29, 1);
+
+static inline void mlxsw_reg_spaft_pack(char *payload, u8 local_port,
+					bool allow_untagged)
+{
+	MLXSW_REG_ZERO(spaft, payload);
+	mlxsw_reg_spaft_local_port_set(payload, local_port);
+	mlxsw_reg_spaft_allow_untagged_set(payload, allow_untagged);
+	mlxsw_reg_spaft_allow_prio_tagged_set(payload, true);
+	mlxsw_reg_spaft_allow_tagged_set(payload, true);
+}
+
 /* SFGC - Switch Flooding Group Configuration
  * ------------------------------------------
  * The following register controls the association of flooding tables and MIDs
@@ -865,6 +1100,379 @@ static inline void mlxsw_reg_sftr_pack(char *payload,
 	mlxsw_reg_sftr_port_mask_set(payload, port, 1);
 }
 
+/* SFDF - Switch Filtering DB Flush
+ * --------------------------------
+ * The switch filtering DB flush register is used to flush the FDB.
+ * Note that FDB notifications are flushed as well.
+ */
+#define MLXSW_REG_SFDF_ID 0x2013
+#define MLXSW_REG_SFDF_LEN 0x14
+
+static const struct mlxsw_reg_info mlxsw_reg_sfdf = {
+	.id = MLXSW_REG_SFDF_ID,
+	.len = MLXSW_REG_SFDF_LEN,
+};
+
+/* reg_sfdf_swid
+ * Switch partition ID.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, sfdf, swid, 0x00, 24, 8);
+
+enum mlxsw_reg_sfdf_flush_type {
+	MLXSW_REG_SFDF_FLUSH_PER_SWID,
+	MLXSW_REG_SFDF_FLUSH_PER_FID,
+	MLXSW_REG_SFDF_FLUSH_PER_PORT,
+	MLXSW_REG_SFDF_FLUSH_PER_PORT_AND_FID,
+	MLXSW_REG_SFDF_FLUSH_PER_LAG,
+	MLXSW_REG_SFDF_FLUSH_PER_LAG_AND_FID,
+};
+
+/* reg_sfdf_flush_type
+ * Flush type.
+ * 0 - All SWID dynamic entries are flushed.
+ * 1 - All FID dynamic entries are flushed.
+ * 2 - All dynamic entries pointing to port are flushed.
+ * 3 - All FID dynamic entries pointing to port are flushed.
+ * 4 - All dynamic entries pointing to LAG are flushed.
+ * 5 - All FID dynamic entries pointing to LAG are flushed.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sfdf, flush_type, 0x04, 28, 4);
+
+/* reg_sfdf_flush_static
+ * Static.
+ * 0 - Flush only dynamic entries.
+ * 1 - Flush both dynamic and static entries.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sfdf, flush_static, 0x04, 24, 1);
+
+static inline void mlxsw_reg_sfdf_pack(char *payload,
+				       enum mlxsw_reg_sfdf_flush_type type)
+{
+	MLXSW_REG_ZERO(sfdf, payload);
+	mlxsw_reg_sfdf_flush_type_set(payload, type);
+	mlxsw_reg_sfdf_flush_static_set(payload, true);
+}
+
+/* reg_sfdf_fid
+ * FID to flush.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sfdf, fid, 0x0C, 0, 16);
+
+/* reg_sfdf_system_port
+ * Port to flush.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sfdf, system_port, 0x0C, 0, 16);
+
+/* reg_sfdf_port_fid_system_port
+ * Port to flush, pointed to by FID.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sfdf, port_fid_system_port, 0x08, 0, 16);
+
+/* reg_sfdf_lag_id
+ * LAG ID to flush.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sfdf, lag_id, 0x0C, 0, 10);
+
+/* reg_sfdf_lag_fid_lag_id
+ * LAG ID to flush, pointed to by FID.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sfdf, lag_fid_lag_id, 0x08, 0, 10);
+
+/* SLDR - Switch LAG Descriptor Register
+ * -----------------------------------------
+ * The switch LAG descriptor register is populated by LAG descriptors.
+ * Each LAG descriptor is indexed by lag_id. The LAG ID runs from 0 to
+ * max_lag-1.
+ */
+#define MLXSW_REG_SLDR_ID 0x2014
+#define MLXSW_REG_SLDR_LEN 0x0C /* counting in only one port in list */
+
+static const struct mlxsw_reg_info mlxsw_reg_sldr = {
+	.id = MLXSW_REG_SLDR_ID,
+	.len = MLXSW_REG_SLDR_LEN,
+};
+
+enum mlxsw_reg_sldr_op {
+	/* Indicates a creation of a new LAG-ID, lag_id must be valid */
+	MLXSW_REG_SLDR_OP_LAG_CREATE,
+	MLXSW_REG_SLDR_OP_LAG_DESTROY,
+	/* Ports that appear in the list have the Distributor enabled */
+	MLXSW_REG_SLDR_OP_LAG_ADD_PORT_LIST,
+	/* Removes ports from the disributor list */
+	MLXSW_REG_SLDR_OP_LAG_REMOVE_PORT_LIST,
+};
+
+/* reg_sldr_op
+ * Operation.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sldr, op, 0x00, 29, 3);
+
+/* reg_sldr_lag_id
+ * LAG identifier. The lag_id is the index into the LAG descriptor table.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, sldr, lag_id, 0x00, 0, 10);
+
+static inline void mlxsw_reg_sldr_lag_create_pack(char *payload, u8 lag_id)
+{
+	MLXSW_REG_ZERO(sldr, payload);
+	mlxsw_reg_sldr_op_set(payload, MLXSW_REG_SLDR_OP_LAG_CREATE);
+	mlxsw_reg_sldr_lag_id_set(payload, lag_id);
+}
+
+static inline void mlxsw_reg_sldr_lag_destroy_pack(char *payload, u8 lag_id)
+{
+	MLXSW_REG_ZERO(sldr, payload);
+	mlxsw_reg_sldr_op_set(payload, MLXSW_REG_SLDR_OP_LAG_DESTROY);
+	mlxsw_reg_sldr_lag_id_set(payload, lag_id);
+}
+
+/* reg_sldr_num_ports
+ * The number of member ports of the LAG.
+ * Reserved for Create / Destroy operations
+ * For Add / Remove operations - indicates the number of ports in the list.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, sldr, num_ports, 0x04, 24, 8);
+
+/* reg_sldr_system_port
+ * System port.
+ * Access: RW
+ */
+MLXSW_ITEM32_INDEXED(reg, sldr, system_port, 0x08, 0, 16, 4, 0, false);
+
+static inline void mlxsw_reg_sldr_lag_add_port_pack(char *payload, u8 lag_id,
+						    u8 local_port)
+{
+	MLXSW_REG_ZERO(sldr, payload);
+	mlxsw_reg_sldr_op_set(payload, MLXSW_REG_SLDR_OP_LAG_ADD_PORT_LIST);
+	mlxsw_reg_sldr_lag_id_set(payload, lag_id);
+	mlxsw_reg_sldr_num_ports_set(payload, 1);
+	mlxsw_reg_sldr_system_port_set(payload, 0, local_port);
+}
+
+static inline void mlxsw_reg_sldr_lag_remove_port_pack(char *payload, u8 lag_id,
+						       u8 local_port)
+{
+	MLXSW_REG_ZERO(sldr, payload);
+	mlxsw_reg_sldr_op_set(payload, MLXSW_REG_SLDR_OP_LAG_REMOVE_PORT_LIST);
+	mlxsw_reg_sldr_lag_id_set(payload, lag_id);
+	mlxsw_reg_sldr_num_ports_set(payload, 1);
+	mlxsw_reg_sldr_system_port_set(payload, 0, local_port);
+}
+
+/* SLCR - Switch LAG Configuration 2 Register
+ * -------------------------------------------
+ * The Switch LAG Configuration register is used for configuring the
+ * LAG properties of the switch.
+ */
+#define MLXSW_REG_SLCR_ID 0x2015
+#define MLXSW_REG_SLCR_LEN 0x10
+
+static const struct mlxsw_reg_info mlxsw_reg_slcr = {
+	.id = MLXSW_REG_SLCR_ID,
+	.len = MLXSW_REG_SLCR_LEN,
+};
+
+enum mlxsw_reg_slcr_pp {
+	/* Global Configuration (for all ports) */
+	MLXSW_REG_SLCR_PP_GLOBAL,
+	/* Per port configuration, based on local_port field */
+	MLXSW_REG_SLCR_PP_PER_PORT,
+};
+
+/* reg_slcr_pp
+ * Per Port Configuration
+ * Note: Reading at Global mode results in reading port 1 configuration.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, slcr, pp, 0x00, 24, 1);
+
+/* reg_slcr_local_port
+ * Local port number
+ * Supported from CPU port
+ * Not supported from router port
+ * Reserved when pp = Global Configuration
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, slcr, local_port, 0x00, 16, 8);
+
+enum mlxsw_reg_slcr_type {
+	MLXSW_REG_SLCR_TYPE_CRC, /* default */
+	MLXSW_REG_SLCR_TYPE_XOR,
+	MLXSW_REG_SLCR_TYPE_RANDOM,
+};
+
+/* reg_slcr_type
+ * Hash type
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, slcr, type, 0x00, 0, 4);
+
+/* Ingress port */
+#define MLXSW_REG_SLCR_LAG_HASH_IN_PORT		BIT(0)
+/* SMAC - for IPv4 and IPv6 packets */
+#define MLXSW_REG_SLCR_LAG_HASH_SMAC_IP		BIT(1)
+/* SMAC - for non-IP packets */
+#define MLXSW_REG_SLCR_LAG_HASH_SMAC_NONIP	BIT(2)
+#define MLXSW_REG_SLCR_LAG_HASH_SMAC \
+	(MLXSW_REG_SLCR_LAG_HASH_SMAC_IP | \
+	 MLXSW_REG_SLCR_LAG_HASH_SMAC_NONIP)
+/* DMAC - for IPv4 and IPv6 packets */
+#define MLXSW_REG_SLCR_LAG_HASH_DMAC_IP		BIT(3)
+/* DMAC - for non-IP packets */
+#define MLXSW_REG_SLCR_LAG_HASH_DMAC_NONIP	BIT(4)
+#define MLXSW_REG_SLCR_LAG_HASH_DMAC \
+	(MLXSW_REG_SLCR_LAG_HASH_DMAC_IP | \
+	 MLXSW_REG_SLCR_LAG_HASH_DMAC_NONIP)
+/* Ethertype - for IPv4 and IPv6 packets */
+#define MLXSW_REG_SLCR_LAG_HASH_ETHERTYPE_IP	BIT(5)
+/* Ethertype - for non-IP packets */
+#define MLXSW_REG_SLCR_LAG_HASH_ETHERTYPE_NONIP	BIT(6)
+#define MLXSW_REG_SLCR_LAG_HASH_ETHERTYPE \
+	(MLXSW_REG_SLCR_LAG_HASH_ETHERTYPE_IP | \
+	 MLXSW_REG_SLCR_LAG_HASH_ETHERTYPE_NONIP)
+/* VLAN ID - for IPv4 and IPv6 packets */
+#define MLXSW_REG_SLCR_LAG_HASH_VLANID_IP	BIT(7)
+/* VLAN ID - for non-IP packets */
+#define MLXSW_REG_SLCR_LAG_HASH_VLANID_NONIP	BIT(8)
+#define MLXSW_REG_SLCR_LAG_HASH_VLANID \
+	(MLXSW_REG_SLCR_LAG_HASH_VLANID_IP | \
+	 MLXSW_REG_SLCR_LAG_HASH_VLANID_NONIP)
+/* Source IP address (can be IPv4 or IPv6) */
+#define MLXSW_REG_SLCR_LAG_HASH_SIP		BIT(9)
+/* Destination IP address (can be IPv4 or IPv6) */
+#define MLXSW_REG_SLCR_LAG_HASH_DIP		BIT(10)
+/* TCP/UDP source port */
+#define MLXSW_REG_SLCR_LAG_HASH_SPORT		BIT(11)
+/* TCP/UDP destination port*/
+#define MLXSW_REG_SLCR_LAG_HASH_DPORT		BIT(12)
+/* IPv4 Protocol/IPv6 Next Header */
+#define MLXSW_REG_SLCR_LAG_HASH_IPPROTO		BIT(13)
+/* IPv6 Flow label */
+#define MLXSW_REG_SLCR_LAG_HASH_FLOWLABEL	BIT(14)
+/* SID - FCoE source ID */
+#define MLXSW_REG_SLCR_LAG_HASH_FCOE_SID	BIT(15)
+/* DID - FCoE destination ID */
+#define MLXSW_REG_SLCR_LAG_HASH_FCOE_DID	BIT(16)
+/* OXID - FCoE originator exchange ID */
+#define MLXSW_REG_SLCR_LAG_HASH_FCOE_OXID	BIT(17)
+/* Destination QP number - for RoCE packets */
+#define MLXSW_REG_SLCR_LAG_HASH_ROCE_DQP	BIT(19)
+
+/* reg_slcr_lag_hash
+ * LAG hashing configuration. This is a bitmask, in which each set
+ * bit includes the corresponding item in the LAG hash calculation.
+ * The default lag_hash contains SMAC, DMAC, VLANID and
+ * Ethertype (for all packet types).
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, slcr, lag_hash, 0x04, 0, 20);
+
+static inline void mlxsw_reg_slcr_pack(char *payload, u16 lag_hash)
+{
+	MLXSW_REG_ZERO(slcr, payload);
+	mlxsw_reg_slcr_pp_set(payload, MLXSW_REG_SLCR_PP_GLOBAL);
+	mlxsw_reg_slcr_type_set(payload, MLXSW_REG_SLCR_TYPE_XOR);
+	mlxsw_reg_slcr_lag_hash_set(payload, lag_hash);
+}
+
+/* SLCOR - Switch LAG Collector Register
+ * -------------------------------------
+ * The Switch LAG Collector register controls the Local Port membership
+ * in a LAG and enablement of the collector.
+ */
+#define MLXSW_REG_SLCOR_ID 0x2016
+#define MLXSW_REG_SLCOR_LEN 0x10
+
+static const struct mlxsw_reg_info mlxsw_reg_slcor = {
+	.id = MLXSW_REG_SLCOR_ID,
+	.len = MLXSW_REG_SLCOR_LEN,
+};
+
+enum mlxsw_reg_slcor_col {
+	/* Port is added with collector disabled */
+	MLXSW_REG_SLCOR_COL_LAG_ADD_PORT,
+	MLXSW_REG_SLCOR_COL_LAG_COLLECTOR_ENABLED,
+	MLXSW_REG_SLCOR_COL_LAG_COLLECTOR_DISABLED,
+	MLXSW_REG_SLCOR_COL_LAG_REMOVE_PORT,
+};
+
+/* reg_slcor_col
+ * Collector configuration
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, slcor, col, 0x00, 30, 2);
+
+/* reg_slcor_local_port
+ * Local port number
+ * Not supported for CPU port
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, slcor, local_port, 0x00, 16, 8);
+
+/* reg_slcor_lag_id
+ * LAG Identifier. Index into the LAG descriptor table.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, slcor, lag_id, 0x00, 0, 10);
+
+/* reg_slcor_port_index
+ * Port index in the LAG list. Only valid on Add Port to LAG col.
+ * Valid range is from 0 to cap_max_lag_members-1
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, slcor, port_index, 0x04, 0, 10);
+
+static inline void mlxsw_reg_slcor_pack(char *payload,
+					u8 local_port, u16 lag_id,
+					enum mlxsw_reg_slcor_col col)
+{
+	MLXSW_REG_ZERO(slcor, payload);
+	mlxsw_reg_slcor_col_set(payload, col);
+	mlxsw_reg_slcor_local_port_set(payload, local_port);
+	mlxsw_reg_slcor_lag_id_set(payload, lag_id);
+}
+
+static inline void mlxsw_reg_slcor_port_add_pack(char *payload,
+						 u8 local_port, u16 lag_id,
+						 u8 port_index)
+{
+	mlxsw_reg_slcor_pack(payload, local_port, lag_id,
+			     MLXSW_REG_SLCOR_COL_LAG_ADD_PORT);
+	mlxsw_reg_slcor_port_index_set(payload, port_index);
+}
+
+static inline void mlxsw_reg_slcor_port_remove_pack(char *payload,
+						    u8 local_port, u16 lag_id)
+{
+	mlxsw_reg_slcor_pack(payload, local_port, lag_id,
+			     MLXSW_REG_SLCOR_COL_LAG_REMOVE_PORT);
+}
+
+static inline void mlxsw_reg_slcor_col_enable_pack(char *payload,
+						   u8 local_port, u16 lag_id)
+{
+	mlxsw_reg_slcor_pack(payload, local_port, lag_id,
+			     MLXSW_REG_SLCOR_COL_LAG_COLLECTOR_ENABLED);
+}
+
+static inline void mlxsw_reg_slcor_col_disable_pack(char *payload,
+						    u8 local_port, u16 lag_id)
+{
+	mlxsw_reg_slcor_pack(payload, local_port, lag_id,
+			     MLXSW_REG_SLCOR_COL_LAG_COLLECTOR_ENABLED);
+}
+
 /* SPMLR - Switch Port MAC Learning Register
  * -----------------------------------------
  * Controls the Switch MAC learning policy per port.
@@ -1235,20 +1843,20 @@ MLXSW_ITEM32(reg, pmlp, width, 0x00, 0, 8);
  * Module number.
  * Access: RW
  */
-MLXSW_ITEM32_INDEXED(reg, pmlp, module, 0x04, 0, 8, 0x04, 0, false);
+MLXSW_ITEM32_INDEXED(reg, pmlp, module, 0x04, 0, 8, 0x04, 0x00, false);
 
 /* reg_pmlp_tx_lane
  * Tx Lane. When rxtx field is cleared, this field is used for Rx as well.
  * Access: RW
  */
-MLXSW_ITEM32_INDEXED(reg, pmlp, tx_lane, 0x04, 16, 2, 0x04, 16, false);
+MLXSW_ITEM32_INDEXED(reg, pmlp, tx_lane, 0x04, 16, 2, 0x04, 0x00, false);
 
 /* reg_pmlp_rx_lane
  * Rx Lane. When rxtx field is cleared, this field is ignored and Rx lane is
  * equal to Tx lane.
  * Access: RW
  */
-MLXSW_ITEM32_INDEXED(reg, pmlp, rx_lane, 0x04, 24, 2, 0x04, 24, false);
+MLXSW_ITEM32_INDEXED(reg, pmlp, rx_lane, 0x04, 24, 2, 0x04, 0x00, false);
 
 static inline void mlxsw_reg_pmlp_pack(char *payload, u8 local_port)
 {
@@ -2087,6 +2695,284 @@ static inline void mlxsw_reg_hpkt_pack(char *payload, u8 action, u16 trap_id)
 	mlxsw_reg_hpkt_ctrl_set(payload, MLXSW_REG_HPKT_CTRL_PACKET_DEFAULT);
 }
 
+/* MFCR - Management Fan Control Register
+ * --------------------------------------
+ * This register controls the settings of the Fan Speed PWM mechanism.
+ */
+#define MLXSW_REG_MFCR_ID 0x9001
+#define MLXSW_REG_MFCR_LEN 0x08
+
+static const struct mlxsw_reg_info mlxsw_reg_mfcr = {
+	.id = MLXSW_REG_MFCR_ID,
+	.len = MLXSW_REG_MFCR_LEN,
+};
+
+enum mlxsw_reg_mfcr_pwm_frequency {
+	MLXSW_REG_MFCR_PWM_FEQ_11HZ = 0x00,
+	MLXSW_REG_MFCR_PWM_FEQ_14_7HZ = 0x01,
+	MLXSW_REG_MFCR_PWM_FEQ_22_1HZ = 0x02,
+	MLXSW_REG_MFCR_PWM_FEQ_1_4KHZ = 0x40,
+	MLXSW_REG_MFCR_PWM_FEQ_5KHZ = 0x41,
+	MLXSW_REG_MFCR_PWM_FEQ_20KHZ = 0x42,
+	MLXSW_REG_MFCR_PWM_FEQ_22_5KHZ = 0x43,
+	MLXSW_REG_MFCR_PWM_FEQ_25KHZ = 0x44,
+};
+
+/* reg_mfcr_pwm_frequency
+ * Controls the frequency of the PWM signal.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, mfcr, pwm_frequency, 0x00, 0, 6);
+
+#define MLXSW_MFCR_TACHOS_MAX 10
+
+/* reg_mfcr_tacho_active
+ * Indicates which of the tachometer is active (bit per tachometer).
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mfcr, tacho_active, 0x04, 16, MLXSW_MFCR_TACHOS_MAX);
+
+#define MLXSW_MFCR_PWMS_MAX 5
+
+/* reg_mfcr_pwm_active
+ * Indicates which of the PWM control is active (bit per PWM).
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mfcr, pwm_active, 0x04, 0, MLXSW_MFCR_PWMS_MAX);
+
+static inline void
+mlxsw_reg_mfcr_pack(char *payload,
+		    enum mlxsw_reg_mfcr_pwm_frequency pwm_frequency)
+{
+	MLXSW_REG_ZERO(mfcr, payload);
+	mlxsw_reg_mfcr_pwm_frequency_set(payload, pwm_frequency);
+}
+
+static inline void
+mlxsw_reg_mfcr_unpack(char *payload,
+		      enum mlxsw_reg_mfcr_pwm_frequency *p_pwm_frequency,
+		      u16 *p_tacho_active, u8 *p_pwm_active)
+{
+	*p_pwm_frequency = mlxsw_reg_mfcr_pwm_frequency_get(payload);
+	*p_tacho_active = mlxsw_reg_mfcr_tacho_active_get(payload);
+	*p_pwm_active = mlxsw_reg_mfcr_pwm_active_get(payload);
+}
+
+/* MFSC - Management Fan Speed Control Register
+ * --------------------------------------------
+ * This register controls the settings of the Fan Speed PWM mechanism.
+ */
+#define MLXSW_REG_MFSC_ID 0x9002
+#define MLXSW_REG_MFSC_LEN 0x08
+
+static const struct mlxsw_reg_info mlxsw_reg_mfsc = {
+	.id = MLXSW_REG_MFSC_ID,
+	.len = MLXSW_REG_MFSC_LEN,
+};
+
+/* reg_mfsc_pwm
+ * Fan pwm to control / monitor.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, mfsc, pwm, 0x00, 24, 3);
+
+/* reg_mfsc_pwm_duty_cycle
+ * Controls the duty cycle of the PWM. Value range from 0..255 to
+ * represent duty cycle of 0%...100%.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, mfsc, pwm_duty_cycle, 0x04, 0, 8);
+
+static inline void mlxsw_reg_mfsc_pack(char *payload, u8 pwm,
+				       u8 pwm_duty_cycle)
+{
+	MLXSW_REG_ZERO(mfsc, payload);
+	mlxsw_reg_mfsc_pwm_set(payload, pwm);
+	mlxsw_reg_mfsc_pwm_duty_cycle_set(payload, pwm_duty_cycle);
+}
+
+/* MFSM - Management Fan Speed Measurement
+ * ---------------------------------------
+ * This register controls the settings of the Tacho measurements and
+ * enables reading the Tachometer measurements.
+ */
+#define MLXSW_REG_MFSM_ID 0x9003
+#define MLXSW_REG_MFSM_LEN 0x08
+
+static const struct mlxsw_reg_info mlxsw_reg_mfsm = {
+	.id = MLXSW_REG_MFSM_ID,
+	.len = MLXSW_REG_MFSM_LEN,
+};
+
+/* reg_mfsm_tacho
+ * Fan tachometer index.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, mfsm, tacho, 0x00, 24, 4);
+
+/* reg_mfsm_rpm
+ * Fan speed (round per minute).
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mfsm, rpm, 0x04, 0, 16);
+
+static inline void mlxsw_reg_mfsm_pack(char *payload, u8 tacho)
+{
+	MLXSW_REG_ZERO(mfsm, payload);
+	mlxsw_reg_mfsm_tacho_set(payload, tacho);
+}
+
+/* MTCAP - Management Temperature Capabilities
+ * -------------------------------------------
+ * This register exposes the capabilities of the device and
+ * system temperature sensing.
+ */
+#define MLXSW_REG_MTCAP_ID 0x9009
+#define MLXSW_REG_MTCAP_LEN 0x08
+
+static const struct mlxsw_reg_info mlxsw_reg_mtcap = {
+	.id = MLXSW_REG_MTCAP_ID,
+	.len = MLXSW_REG_MTCAP_LEN,
+};
+
+/* reg_mtcap_sensor_count
+ * Number of sensors supported by the device.
+ * This includes the QSFP module sensors (if exists in the QSFP module).
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mtcap, sensor_count, 0x00, 0, 7);
+
+/* MTMP - Management Temperature
+ * -----------------------------
+ * This register controls the settings of the temperature measurements
+ * and enables reading the temperature measurements. Note that temperature
+ * is in 0.125 degrees Celsius.
+ */
+#define MLXSW_REG_MTMP_ID 0x900A
+#define MLXSW_REG_MTMP_LEN 0x20
+
+static const struct mlxsw_reg_info mlxsw_reg_mtmp = {
+	.id = MLXSW_REG_MTMP_ID,
+	.len = MLXSW_REG_MTMP_LEN,
+};
+
+/* reg_mtmp_sensor_index
+ * Sensors index to access.
+ * 64-127 of sensor_index are mapped to the SFP+/QSFP modules sequentially
+ * (module 0 is mapped to sensor_index 64).
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, mtmp, sensor_index, 0x00, 0, 7);
+
+/* Convert to milli degrees Celsius */
+#define MLXSW_REG_MTMP_TEMP_TO_MC(val) (val * 125)
+
+/* reg_mtmp_temperature
+ * Temperature reading from the sensor. Reading is in 0.125 Celsius
+ * degrees units.
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mtmp, temperature, 0x04, 0, 16);
+
+/* reg_mtmp_mte
+ * Max Temperature Enable - enables measuring the max temperature on a sensor.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, mtmp, mte, 0x08, 31, 1);
+
+/* reg_mtmp_mtr
+ * Max Temperature Reset - clears the value of the max temperature register.
+ * Access: WO
+ */
+MLXSW_ITEM32(reg, mtmp, mtr, 0x08, 30, 1);
+
+/* reg_mtmp_max_temperature
+ * The highest measured temperature from the sensor.
+ * When the bit mte is cleared, the field max_temperature is reserved.
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mtmp, max_temperature, 0x08, 0, 16);
+
+#define MLXSW_REG_MTMP_SENSOR_NAME_SIZE 8
+
+/* reg_mtmp_sensor_name
+ * Sensor Name
+ * Access: RO
+ */
+MLXSW_ITEM_BUF(reg, mtmp, sensor_name, 0x18, MLXSW_REG_MTMP_SENSOR_NAME_SIZE);
+
+static inline void mlxsw_reg_mtmp_pack(char *payload, u8 sensor_index,
+				       bool max_temp_enable,
+				       bool max_temp_reset)
+{
+	MLXSW_REG_ZERO(mtmp, payload);
+	mlxsw_reg_mtmp_sensor_index_set(payload, sensor_index);
+	mlxsw_reg_mtmp_mte_set(payload, max_temp_enable);
+	mlxsw_reg_mtmp_mtr_set(payload, max_temp_reset);
+}
+
+static inline void mlxsw_reg_mtmp_unpack(char *payload, unsigned int *p_temp,
+					 unsigned int *p_max_temp,
+					 char *sensor_name)
+{
+	u16 temp;
+
+	if (p_temp) {
+		temp = mlxsw_reg_mtmp_temperature_get(payload);
+		*p_temp = MLXSW_REG_MTMP_TEMP_TO_MC(temp);
+	}
+	if (p_max_temp) {
+		temp = mlxsw_reg_mtmp_max_temperature_get(payload);
+		*p_max_temp = MLXSW_REG_MTMP_TEMP_TO_MC(temp);
+	}
+	if (sensor_name)
+		mlxsw_reg_mtmp_sensor_name_memcpy_from(payload, sensor_name);
+}
+
+/* MLCR - Management LED Control Register
+ * --------------------------------------
+ * Controls the system LEDs.
+ */
+#define MLXSW_REG_MLCR_ID 0x902B
+#define MLXSW_REG_MLCR_LEN 0x0C
+
+static const struct mlxsw_reg_info mlxsw_reg_mlcr = {
+	.id = MLXSW_REG_MLCR_ID,
+	.len = MLXSW_REG_MLCR_LEN,
+};
+
+/* reg_mlcr_local_port
+ * Local port number.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, mlcr, local_port, 0x00, 16, 8);
+
+#define MLXSW_REG_MLCR_DURATION_MAX 0xFFFF
+
+/* reg_mlcr_beacon_duration
+ * Duration of the beacon to be active, in seconds.
+ * 0x0 - Will turn off the beacon.
+ * 0xFFFF - Will turn on the beacon until explicitly turned off.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, mlcr, beacon_duration, 0x04, 0, 16);
+
+/* reg_mlcr_beacon_remain
+ * Remaining duration of the beacon, in seconds.
+ * 0xFFFF indicates an infinite amount of time.
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mlcr, beacon_remain, 0x08, 0, 16);
+
+static inline void mlxsw_reg_mlcr_pack(char *payload, u8 local_port,
+				       bool active)
+{
+	MLXSW_REG_ZERO(mlcr, payload);
+	mlxsw_reg_mlcr_local_port_set(payload, local_port);
+	mlxsw_reg_mlcr_beacon_duration_set(payload, active ?
+					   MLXSW_REG_MLCR_DURATION_MAX : 0);
+}
+
 /* SBPR - Shared Buffer Pools Register
  * -----------------------------------
  * The SBPR configures and retrieves the shared buffer pools and configuration.
@@ -2357,6 +3243,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id)
 		return "SGCR";
 	case MLXSW_REG_SPAD_ID:
 		return "SPAD";
+	case MLXSW_REG_SMID_ID:
+		return "SMID";
 	case MLXSW_REG_SSPR_ID:
 		return "SSPR";
 	case MLXSW_REG_SFDAT_ID:
@@ -2371,10 +3259,20 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id)
 		return "SPVID";
 	case MLXSW_REG_SPVM_ID:
 		return "SPVM";
+	case MLXSW_REG_SPAFT_ID:
+		return "SPAFT";
 	case MLXSW_REG_SFGC_ID:
 		return "SFGC";
 	case MLXSW_REG_SFTR_ID:
 		return "SFTR";
+	case MLXSW_REG_SFDF_ID:
+		return "SFDF";
+	case MLXSW_REG_SLDR_ID:
+		return "SLDR";
+	case MLXSW_REG_SLCR_ID:
+		return "SLCR";
+	case MLXSW_REG_SLCOR_ID:
+		return "SLCOR";
 	case MLXSW_REG_SPMLR_ID:
 		return "SPMLR";
 	case MLXSW_REG_SVFA_ID:
@@ -2405,6 +3303,18 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id)
 		return "HTGT";
 	case MLXSW_REG_HPKT_ID:
 		return "HPKT";
+	case MLXSW_REG_MFCR_ID:
+		return "MFCR";
+	case MLXSW_REG_MFSC_ID:
+		return "MFSC";
+	case MLXSW_REG_MFSM_ID:
+		return "MFSM";
+	case MLXSW_REG_MTCAP_ID:
+		return "MTCAP";
+	case MLXSW_REG_MTMP_ID:
+		return "MTMP";
+	case MLXSW_REG_MLCR_ID:
+		return "MLCR";
 	case MLXSW_REG_SBPR_ID:
 		return "SBPR";
 	case MLXSW_REG_SBCM_ID:
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 3be4a2355..a94daa8c3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -48,6 +48,7 @@
 #include <linux/workqueue.h>
 #include <linux/jiffies.h>
 #include <linux/bitops.h>
+#include <linux/list.h>
 #include <net/switchdev.h>
 #include <generated/utsrelease.h>
 
@@ -186,33 +187,6 @@ static int mlxsw_sp_port_oper_status_get(struct mlxsw_sp_port *mlxsw_sp_port,
 	return 0;
 }
 
-static int mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vfid)
-{
-	char sfmr_pl[MLXSW_REG_SFMR_LEN];
-	int err;
-
-	mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID,
-			    MLXSW_SP_VFID_BASE + vfid, 0);
-	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl);
-
-	if (err)
-		return err;
-
-	set_bit(vfid, mlxsw_sp->active_vfids);
-	return 0;
-}
-
-static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, u16 vfid)
-{
-	char sfmr_pl[MLXSW_REG_SFMR_LEN];
-
-	clear_bit(vfid, mlxsw_sp->active_vfids);
-
-	mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID,
-			    MLXSW_SP_VFID_BASE + vfid, 0);
-	mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl);
-}
-
 static int mlxsw_sp_port_dev_addr_set(struct mlxsw_sp_port *mlxsw_sp_port,
 				      unsigned char *addr)
 {
@@ -417,6 +391,10 @@ static netdev_tx_t mlxsw_sp_port_xmit(struct sk_buff *skb,
 	return NETDEV_TX_OK;
 }
 
+static void mlxsw_sp_set_rx_mode(struct net_device *dev)
+{
+}
+
 static int mlxsw_sp_port_set_mac_address(struct net_device *dev, void *p)
 {
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
@@ -545,12 +523,132 @@ static int mlxsw_sp_port_vlan_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port)
 	return 0;
 }
 
+static struct mlxsw_sp_vfid *
+mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, u16 vid)
+{
+	struct mlxsw_sp_vfid *vfid;
+
+	list_for_each_entry(vfid, &mlxsw_sp->port_vfids.list, list) {
+		if (vfid->vid == vid)
+			return vfid;
+	}
+
+	return NULL;
+}
+
+static u16 mlxsw_sp_avail_vfid_get(const struct mlxsw_sp *mlxsw_sp)
+{
+	return find_first_zero_bit(mlxsw_sp->port_vfids.mapped,
+				   MLXSW_SP_VFID_PORT_MAX);
+}
+
+static int __mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vfid)
+{
+	u16 fid = mlxsw_sp_vfid_to_fid(vfid);
+	char sfmr_pl[MLXSW_REG_SFMR_LEN];
+
+	mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID, fid, 0);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl);
+}
+
+static void __mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, u16 vfid)
+{
+	u16 fid = mlxsw_sp_vfid_to_fid(vfid);
+	char sfmr_pl[MLXSW_REG_SFMR_LEN];
+
+	mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID, fid, 0);
+	mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl);
+}
+
+static struct mlxsw_sp_vfid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp,
+						  u16 vid)
+{
+	struct device *dev = mlxsw_sp->bus_info->dev;
+	struct mlxsw_sp_vfid *vfid;
+	u16 n_vfid;
+	int err;
+
+	n_vfid = mlxsw_sp_avail_vfid_get(mlxsw_sp);
+	if (n_vfid == MLXSW_SP_VFID_PORT_MAX) {
+		dev_err(dev, "No available vFIDs\n");
+		return ERR_PTR(-ERANGE);
+	}
+
+	err = __mlxsw_sp_vfid_create(mlxsw_sp, n_vfid);
+	if (err) {
+		dev_err(dev, "Failed to create vFID=%d\n", n_vfid);
+		return ERR_PTR(err);
+	}
+
+	vfid = kzalloc(sizeof(*vfid), GFP_KERNEL);
+	if (!vfid)
+		goto err_allocate_vfid;
+
+	vfid->vfid = n_vfid;
+	vfid->vid = vid;
+
+	list_add(&vfid->list, &mlxsw_sp->port_vfids.list);
+	set_bit(n_vfid, mlxsw_sp->port_vfids.mapped);
+
+	return vfid;
+
+err_allocate_vfid:
+	__mlxsw_sp_vfid_destroy(mlxsw_sp, n_vfid);
+	return ERR_PTR(-ENOMEM);
+}
+
+static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp,
+				  struct mlxsw_sp_vfid *vfid)
+{
+	clear_bit(vfid->vfid, mlxsw_sp->port_vfids.mapped);
+	list_del(&vfid->list);
+
+	__mlxsw_sp_vfid_destroy(mlxsw_sp, vfid->vfid);
+
+	kfree(vfid);
+}
+
+static struct mlxsw_sp_port *
+mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port,
+			   struct mlxsw_sp_vfid *vfid)
+{
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+
+	mlxsw_sp_vport = kzalloc(sizeof(*mlxsw_sp_vport), GFP_KERNEL);
+	if (!mlxsw_sp_vport)
+		return NULL;
+
+	/* dev will be set correctly after the VLAN device is linked
+	 * with the real device. In case of bridge SELF invocation, dev
+	 * will remain as is.
+	 */
+	mlxsw_sp_vport->dev = mlxsw_sp_port->dev;
+	mlxsw_sp_vport->mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	mlxsw_sp_vport->local_port = mlxsw_sp_port->local_port;
+	mlxsw_sp_vport->stp_state = BR_STATE_FORWARDING;
+	mlxsw_sp_vport->lagged = mlxsw_sp_port->lagged;
+	mlxsw_sp_vport->lag_id = mlxsw_sp_port->lag_id;
+	mlxsw_sp_vport->vport.vfid = vfid;
+	mlxsw_sp_vport->vport.vid = vfid->vid;
+
+	list_add(&mlxsw_sp_vport->vport.list, &mlxsw_sp_port->vports_list);
+
+	return mlxsw_sp_vport;
+}
+
+static void mlxsw_sp_port_vport_destroy(struct mlxsw_sp_port *mlxsw_sp_vport)
+{
+	list_del(&mlxsw_sp_vport->vport.list);
+	kfree(mlxsw_sp_vport);
+}
+
 int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto,
 			  u16 vid)
 {
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
-	char *sftr_pl;
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+	struct mlxsw_sp_vfid *vfid;
 	int err;
 
 	/* VLAN 0 is added to HW filter when device goes up, but it is
@@ -559,100 +657,105 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto,
 	if (!vid)
 		return 0;
 
-	if (test_bit(vid, mlxsw_sp_port->active_vfids)) {
+	if (mlxsw_sp_port_vport_find(mlxsw_sp_port, vid)) {
 		netdev_warn(dev, "VID=%d already configured\n", vid);
 		return 0;
 	}
 
-	if (!test_bit(vid, mlxsw_sp->active_vfids)) {
-		err = mlxsw_sp_vfid_create(mlxsw_sp, vid);
-		if (err) {
-			netdev_err(dev, "Failed to create vFID=%d\n",
-				   MLXSW_SP_VFID_BASE + vid);
-			return err;
+	vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid);
+	if (!vfid) {
+		vfid = mlxsw_sp_vfid_create(mlxsw_sp, vid);
+		if (IS_ERR(vfid)) {
+			netdev_err(dev, "Failed to create vFID for VID=%d\n",
+				   vid);
+			return PTR_ERR(vfid);
 		}
+	}
 
-		sftr_pl = kmalloc(MLXSW_REG_SFTR_LEN, GFP_KERNEL);
-		if (!sftr_pl) {
-			err = -ENOMEM;
-			goto err_flood_table_alloc;
-		}
-		mlxsw_reg_sftr_pack(sftr_pl, 0, vid,
-				    MLXSW_REG_SFGC_TABLE_TYPE_FID, 0,
-				    MLXSW_PORT_CPU_PORT, true);
-		err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl);
-		kfree(sftr_pl);
+	mlxsw_sp_vport = mlxsw_sp_port_vport_create(mlxsw_sp_port, vfid);
+	if (!mlxsw_sp_vport) {
+		netdev_err(dev, "Failed to create vPort for VID=%d\n", vid);
+		err = -ENOMEM;
+		goto err_port_vport_create;
+	}
+
+	if (!vfid->nr_vports) {
+		err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid,
+					       true, false);
 		if (err) {
-			netdev_err(dev, "Failed to configure flood table\n");
-			goto err_flood_table_config;
+			netdev_err(dev, "Failed to setup flooding for vFID=%d\n",
+				   vfid->vfid);
+			goto err_vport_flood_set;
 		}
 	}
 
-	/* In case we fail in the following steps, we intentionally do not
-	 * destroy the associated vFID.
-	 */
-
 	/* When adding the first VLAN interface on a bridged port we need to
 	 * transition all the active 802.1Q bridge VLANs to use explicit
 	 * {Port, VID} to FID mappings and set the port's mode to Virtual mode.
 	 */
-	if (!mlxsw_sp_port->nr_vfids) {
+	if (list_is_singular(&mlxsw_sp_port->vports_list)) {
 		err = mlxsw_sp_port_vp_mode_trans(mlxsw_sp_port);
 		if (err) {
 			netdev_err(dev, "Failed to set to Virtual mode\n");
-			return err;
+			goto err_port_vp_mode_trans;
 		}
 	}
 
-	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port,
+	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
 					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
-					   true, MLXSW_SP_VFID_BASE + vid, vid);
+					   true,
+					   mlxsw_sp_vfid_to_fid(vfid->vfid),
+					   vid);
 	if (err) {
 		netdev_err(dev, "Failed to map {Port, VID=%d} to vFID=%d\n",
-			   vid, MLXSW_SP_VFID_BASE + vid);
+			   vid, vfid->vfid);
 		goto err_port_vid_to_fid_set;
 	}
 
-	err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, false);
+	err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false);
 	if (err) {
 		netdev_err(dev, "Failed to disable learning for VID=%d\n", vid);
 		goto err_port_vid_learning_set;
 	}
 
-	err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid, true, false);
+	err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, true, false);
 	if (err) {
 		netdev_err(dev, "Failed to set VLAN membership for VID=%d\n",
 			   vid);
 		goto err_port_add_vid;
 	}
 
-	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_port, vid,
+	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid,
 					  MLXSW_REG_SPMS_STATE_FORWARDING);
 	if (err) {
 		netdev_err(dev, "Failed to set STP state for VID=%d\n", vid);
 		goto err_port_stp_state_set;
 	}
 
-	mlxsw_sp_port->nr_vfids++;
-	set_bit(vid, mlxsw_sp_port->active_vfids);
+	vfid->nr_vports++;
 
 	return 0;
 
-err_flood_table_config:
-err_flood_table_alloc:
-	mlxsw_sp_vfid_destroy(mlxsw_sp, vid);
-	return err;
-
 err_port_stp_state_set:
-	mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid, false, false);
+	mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, false, false);
 err_port_add_vid:
-	mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, true);
+	mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true);
 err_port_vid_learning_set:
-	mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port,
+	mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
 				     MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false,
-				     MLXSW_SP_VFID_BASE + vid, vid);
+				     mlxsw_sp_vfid_to_fid(vfid->vfid), vid);
 err_port_vid_to_fid_set:
-	mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port);
+	if (list_is_singular(&mlxsw_sp_port->vports_list))
+		mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port);
+err_port_vp_mode_trans:
+	if (!vfid->nr_vports)
+		mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false,
+					 false);
+err_vport_flood_set:
+	mlxsw_sp_port_vport_destroy(mlxsw_sp_vport);
+err_port_vport_create:
+	if (!vfid->nr_vports)
+		mlxsw_sp_vfid_destroy(mlxsw_sp, vfid);
 	return err;
 }
 
@@ -660,6 +763,8 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev,
 			   __be16 __always_unused proto, u16 vid)
 {
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+	struct mlxsw_sp_vfid *vfid;
 	int err;
 
 	/* VLAN 0 is removed from HW filter when device goes down, but
@@ -668,38 +773,42 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev,
 	if (!vid)
 		return 0;
 
-	if (!test_bit(vid, mlxsw_sp_port->active_vfids)) {
+	mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid);
+	if (!mlxsw_sp_vport) {
 		netdev_warn(dev, "VID=%d does not exist\n", vid);
 		return 0;
 	}
 
-	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_port, vid,
+	vfid = mlxsw_sp_vport->vport.vfid;
+
+	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid,
 					  MLXSW_REG_SPMS_STATE_DISCARDING);
 	if (err) {
 		netdev_err(dev, "Failed to set STP state for VID=%d\n", vid);
 		return err;
 	}
 
-	err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid, false, false);
+	err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, false, false);
 	if (err) {
 		netdev_err(dev, "Failed to set VLAN membership for VID=%d\n",
 			   vid);
 		return err;
 	}
 
-	err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, true);
+	err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true);
 	if (err) {
 		netdev_err(dev, "Failed to enable learning for VID=%d\n", vid);
 		return err;
 	}
 
-	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port,
+	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
 					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
-					   false, MLXSW_SP_VFID_BASE + vid,
+					   false,
+					   mlxsw_sp_vfid_to_fid(vfid->vfid),
 					   vid);
 	if (err) {
 		netdev_err(dev, "Failed to invalidate {Port, VID=%d} to vFID=%d mapping\n",
-			   vid, MLXSW_SP_VFID_BASE + vid);
+			   vid, vfid->vfid);
 		return err;
 	}
 
@@ -707,7 +816,7 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev,
 	 * transition all active 802.1Q bridge VLANs to use VID to FID
 	 * mappings and set port's mode to VLAN mode.
 	 */
-	if (mlxsw_sp_port->nr_vfids == 1) {
+	if (list_is_singular(&mlxsw_sp_port->vports_list)) {
 		err = mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port);
 		if (err) {
 			netdev_err(dev, "Failed to set to VLAN mode\n");
@@ -715,8 +824,12 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev,
 		}
 	}
 
-	mlxsw_sp_port->nr_vfids--;
-	clear_bit(vid, mlxsw_sp_port->active_vfids);
+	vfid->nr_vports--;
+	mlxsw_sp_port_vport_destroy(mlxsw_sp_vport);
+
+	/* Destroy the vFID if no vPorts are assigned to it anymore. */
+	if (!vfid->nr_vports)
+		mlxsw_sp_vfid_destroy(mlxsw_sp_port->mlxsw_sp, vfid);
 
 	return 0;
 }
@@ -725,6 +838,7 @@ static const struct net_device_ops mlxsw_sp_port_netdev_ops = {
 	.ndo_open		= mlxsw_sp_port_open,
 	.ndo_stop		= mlxsw_sp_port_stop,
 	.ndo_start_xmit		= mlxsw_sp_port_xmit,
+	.ndo_set_rx_mode	= mlxsw_sp_set_rx_mode,
 	.ndo_set_mac_address	= mlxsw_sp_port_set_mac_address,
 	.ndo_change_mtu		= mlxsw_sp_port_change_mtu,
 	.ndo_get_stats64	= mlxsw_sp_port_get_stats64,
@@ -859,6 +973,29 @@ static void mlxsw_sp_port_get_strings(struct net_device *dev,
 	}
 }
 
+static int mlxsw_sp_port_set_phys_id(struct net_device *dev,
+				     enum ethtool_phys_id_state state)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char mlcr_pl[MLXSW_REG_MLCR_LEN];
+	bool active;
+
+	switch (state) {
+	case ETHTOOL_ID_ACTIVE:
+		active = true;
+		break;
+	case ETHTOOL_ID_INACTIVE:
+		active = false;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	mlxsw_reg_mlcr_pack(mlcr_pl, mlxsw_sp_port->local_port, active);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mlcr), mlcr_pl);
+}
+
 static void mlxsw_sp_port_get_stats(struct net_device *dev,
 				    struct ethtool_stats *stats, u64 *data)
 {
@@ -1205,6 +1342,7 @@ static const struct ethtool_ops mlxsw_sp_port_ethtool_ops = {
 	.get_drvinfo		= mlxsw_sp_port_get_drvinfo,
 	.get_link		= ethtool_op_get_link,
 	.get_strings		= mlxsw_sp_port_get_strings,
+	.set_phys_id		= mlxsw_sp_port_set_phys_id,
 	.get_ethtool_stats	= mlxsw_sp_port_get_stats,
 	.get_sset_count		= mlxsw_sp_port_get_sset_count,
 	.get_settings		= mlxsw_sp_port_get_settings,
@@ -1216,6 +1354,7 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	struct mlxsw_sp_port *mlxsw_sp_port;
 	struct net_device *dev;
 	bool usable;
+	size_t bytes;
 	int err;
 
 	dev = alloc_etherdev(sizeof(struct mlxsw_sp_port));
@@ -1225,10 +1364,18 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	mlxsw_sp_port->dev = dev;
 	mlxsw_sp_port->mlxsw_sp = mlxsw_sp;
 	mlxsw_sp_port->local_port = local_port;
-	mlxsw_sp_port->learning = 1;
-	mlxsw_sp_port->learning_sync = 1;
-	mlxsw_sp_port->uc_flood = 1;
-	mlxsw_sp_port->pvid = 1;
+	bytes = DIV_ROUND_UP(VLAN_N_VID, BITS_PER_BYTE);
+	mlxsw_sp_port->active_vlans = kzalloc(bytes, GFP_KERNEL);
+	if (!mlxsw_sp_port->active_vlans) {
+		err = -ENOMEM;
+		goto err_port_active_vlans_alloc;
+	}
+	mlxsw_sp_port->untagged_vlans = kzalloc(bytes, GFP_KERNEL);
+	if (!mlxsw_sp_port->untagged_vlans) {
+		err = -ENOMEM;
+		goto err_port_untagged_vlans_alloc;
+	}
+	INIT_LIST_HEAD(&mlxsw_sp_port->vports_list);
 
 	mlxsw_sp_port->pcpu_stats =
 		netdev_alloc_pcpu_stats(struct mlxsw_sp_port_pcpu_stats);
@@ -1330,16 +1477,29 @@ err_port_module_check:
 err_dev_addr_init:
 	free_percpu(mlxsw_sp_port->pcpu_stats);
 err_alloc_stats:
+	kfree(mlxsw_sp_port->untagged_vlans);
+err_port_untagged_vlans_alloc:
+	kfree(mlxsw_sp_port->active_vlans);
+err_port_active_vlans_alloc:
 	free_netdev(dev);
 	return err;
 }
 
-static void mlxsw_sp_vfids_fini(struct mlxsw_sp *mlxsw_sp)
+static void mlxsw_sp_port_vports_fini(struct mlxsw_sp_port *mlxsw_sp_port)
 {
-	u16 vfid;
+	struct net_device *dev = mlxsw_sp_port->dev;
+	struct mlxsw_sp_port *mlxsw_sp_vport, *tmp;
 
-	for_each_set_bit(vfid, mlxsw_sp->active_vfids, VLAN_N_VID)
-		mlxsw_sp_vfid_destroy(mlxsw_sp, vfid);
+	list_for_each_entry_safe(mlxsw_sp_vport, tmp,
+				 &mlxsw_sp_port->vports_list, vport.list) {
+		u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport);
+
+		/* vPorts created for VLAN devices should already be gone
+		 * by now, since we unregistered the port netdev.
+		 */
+		WARN_ON(is_vlan_dev(mlxsw_sp_vport->dev));
+		mlxsw_sp_port_kill_vid(dev, 0, vid);
+	}
 }
 
 static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port)
@@ -1348,10 +1508,12 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 
 	if (!mlxsw_sp_port)
 		return;
-	mlxsw_sp_port_kill_vid(mlxsw_sp_port->dev, 0, 1);
 	unregister_netdev(mlxsw_sp_port->dev); /* This calls ndo_stop */
+	mlxsw_sp_port_vports_fini(mlxsw_sp_port);
 	mlxsw_sp_port_switchdev_fini(mlxsw_sp_port);
 	free_percpu(mlxsw_sp_port->pcpu_stats);
+	kfree(mlxsw_sp_port->untagged_vlans);
+	kfree(mlxsw_sp_port->active_vlans);
 	free_netdev(mlxsw_sp_port->dev);
 }
 
@@ -1633,16 +1795,15 @@ static int __mlxsw_sp_flood_init(struct mlxsw_core *mlxsw_core,
 	enum mlxsw_sp_flood_table flood_table;
 	char sfgc_pl[MLXSW_REG_SFGC_LEN];
 
-	if (bridge_type == MLXSW_REG_SFGC_BRIDGE_TYPE_VFID) {
+	if (bridge_type == MLXSW_REG_SFGC_BRIDGE_TYPE_VFID)
 		table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID;
-		flood_table = 0;
-	} else {
+	else
 		table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST;
-		if (type == MLXSW_REG_SFGC_TYPE_UNKNOWN_UNICAST)
-			flood_table = MLXSW_SP_FLOOD_TABLE_UC;
-		else
-			flood_table = MLXSW_SP_FLOOD_TABLE_BM;
-	}
+
+	if (type == MLXSW_REG_SFGC_TYPE_UNKNOWN_UNICAST)
+		flood_table = MLXSW_SP_FLOOD_TABLE_UC;
+	else
+		flood_table = MLXSW_SP_FLOOD_TABLE_BM;
 
 	mlxsw_reg_sfgc_pack(sfgc_pl, type, bridge_type, table_type,
 			    flood_table);
@@ -1653,9 +1814,6 @@ static int mlxsw_sp_flood_init(struct mlxsw_sp *mlxsw_sp)
 {
 	int type, err;
 
-	/* For non-offloaded netdevs, flood all traffic types to CPU
-	 * port.
-	 */
 	for (type = 0; type < MLXSW_REG_SFGC_TYPE_MAX; type++) {
 		if (type == MLXSW_REG_SFGC_TYPE_RESERVED)
 			continue;
@@ -1664,15 +1822,6 @@ static int mlxsw_sp_flood_init(struct mlxsw_sp *mlxsw_sp)
 					    MLXSW_REG_SFGC_BRIDGE_TYPE_VFID);
 		if (err)
 			return err;
-	}
-
-	/* For bridged ports, use one flooding table for unknown unicast
-	 * traffic and a second table for unregistered multicast and
-	 * broadcast.
-	 */
-	for (type = 0; type < MLXSW_REG_SFGC_TYPE_MAX; type++) {
-		if (type == MLXSW_REG_SFGC_TYPE_RESERVED)
-			continue;
 
 		err = __mlxsw_sp_flood_init(mlxsw_sp->core, type,
 					    MLXSW_REG_SFGC_BRIDGE_TYPE_1Q_FID);
@@ -1683,6 +1832,22 @@ static int mlxsw_sp_flood_init(struct mlxsw_sp *mlxsw_sp)
 	return 0;
 }
 
+static int mlxsw_sp_lag_init(struct mlxsw_sp *mlxsw_sp)
+{
+	char slcr_pl[MLXSW_REG_SLCR_LEN];
+
+	mlxsw_reg_slcr_pack(slcr_pl, MLXSW_REG_SLCR_LAG_HASH_SMAC |
+				     MLXSW_REG_SLCR_LAG_HASH_DMAC |
+				     MLXSW_REG_SLCR_LAG_HASH_ETHERTYPE |
+				     MLXSW_REG_SLCR_LAG_HASH_VLANID |
+				     MLXSW_REG_SLCR_LAG_HASH_SIP |
+				     MLXSW_REG_SLCR_LAG_HASH_DIP |
+				     MLXSW_REG_SLCR_LAG_HASH_SPORT |
+				     MLXSW_REG_SLCR_LAG_HASH_DPORT |
+				     MLXSW_REG_SLCR_LAG_HASH_IPPROTO);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(slcr), slcr_pl);
+}
+
 static int mlxsw_sp_init(void *priv, struct mlxsw_core *mlxsw_core,
 			 const struct mlxsw_bus_info *mlxsw_bus_info)
 {
@@ -1691,6 +1856,9 @@ static int mlxsw_sp_init(void *priv, struct mlxsw_core *mlxsw_core,
 
 	mlxsw_sp->core = mlxsw_core;
 	mlxsw_sp->bus_info = mlxsw_bus_info;
+	INIT_LIST_HEAD(&mlxsw_sp->port_vfids.list);
+	INIT_LIST_HEAD(&mlxsw_sp->br_vfids.list);
+	INIT_LIST_HEAD(&mlxsw_sp->br_mids.list);
 
 	err = mlxsw_sp_base_mac_get(mlxsw_sp);
 	if (err) {
@@ -1701,7 +1869,7 @@ static int mlxsw_sp_init(void *priv, struct mlxsw_core *mlxsw_core,
 	err = mlxsw_sp_ports_create(mlxsw_sp);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n");
-		goto err_ports_create;
+		return err;
 	}
 
 	err = mlxsw_sp_event_register(mlxsw_sp, MLXSW_TRAP_ID_PUDE);
@@ -1728,6 +1896,12 @@ static int mlxsw_sp_init(void *priv, struct mlxsw_core *mlxsw_core,
 		goto err_buffers_init;
 	}
 
+	err = mlxsw_sp_lag_init(mlxsw_sp);
+	if (err) {
+		dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize LAG\n");
+		goto err_lag_init;
+	}
+
 	err = mlxsw_sp_switchdev_init(mlxsw_sp);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize switchdev\n");
@@ -1737,6 +1911,7 @@ static int mlxsw_sp_init(void *priv, struct mlxsw_core *mlxsw_core,
 	return 0;
 
 err_switchdev_init:
+err_lag_init:
 err_buffers_init:
 err_flood_init:
 	mlxsw_sp_traps_fini(mlxsw_sp);
@@ -1744,8 +1919,6 @@ err_rx_listener_register:
 	mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE);
 err_event_register:
 	mlxsw_sp_ports_remove(mlxsw_sp);
-err_ports_create:
-	mlxsw_sp_vfids_fini(mlxsw_sp);
 	return err;
 }
 
@@ -1757,18 +1930,17 @@ static void mlxsw_sp_fini(void *priv)
 	mlxsw_sp_traps_fini(mlxsw_sp);
 	mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE);
 	mlxsw_sp_ports_remove(mlxsw_sp);
-	mlxsw_sp_vfids_fini(mlxsw_sp);
 }
 
 static struct mlxsw_config_profile mlxsw_sp_config_profile = {
 	.used_max_vepa_channels		= 1,
 	.max_vepa_channels		= 0,
 	.used_max_lag			= 1,
-	.max_lag			= 64,
+	.max_lag			= MLXSW_SP_LAG_MAX,
 	.used_max_port_per_lag		= 1,
-	.max_port_per_lag		= 16,
+	.max_port_per_lag		= MLXSW_SP_PORT_PER_LAG_MAX,
 	.used_max_mid			= 1,
-	.max_mid			= 7000,
+	.max_mid			= MLXSW_SP_MID_MAX,
 	.used_max_pgt			= 1,
 	.max_pgt			= 0,
 	.used_max_system_port		= 1,
@@ -1782,8 +1954,8 @@ static struct mlxsw_config_profile mlxsw_sp_config_profile = {
 	.flood_mode			= 3,
 	.max_fid_offset_flood_tables	= 2,
 	.fid_offset_flood_table_size	= VLAN_N_VID - 1,
-	.max_fid_flood_tables		= 1,
-	.fid_flood_table_size		= VLAN_N_VID,
+	.max_fid_flood_tables		= 2,
+	.fid_flood_table_size		= MLXSW_SP_VFID_MAX,
 	.used_max_ib_mc			= 1,
 	.max_ib_mc			= 0,
 	.used_max_pkey			= 1,
@@ -1807,6 +1979,115 @@ static struct mlxsw_driver mlxsw_sp_driver = {
 	.profile		= &mlxsw_sp_config_profile,
 };
 
+static int
+mlxsw_sp_port_fdb_flush_by_port(const struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char sfdf_pl[MLXSW_REG_SFDF_LEN];
+
+	mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_PORT);
+	mlxsw_reg_sfdf_system_port_set(sfdf_pl, mlxsw_sp_port->local_port);
+
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl);
+}
+
+static int
+mlxsw_sp_port_fdb_flush_by_port_fid(const struct mlxsw_sp_port *mlxsw_sp_port,
+				    u16 fid)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char sfdf_pl[MLXSW_REG_SFDF_LEN];
+
+	mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_PORT_AND_FID);
+	mlxsw_reg_sfdf_fid_set(sfdf_pl, fid);
+	mlxsw_reg_sfdf_port_fid_system_port_set(sfdf_pl,
+						mlxsw_sp_port->local_port);
+
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl);
+}
+
+static int
+mlxsw_sp_port_fdb_flush_by_lag_id(const struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char sfdf_pl[MLXSW_REG_SFDF_LEN];
+
+	mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_LAG);
+	mlxsw_reg_sfdf_lag_id_set(sfdf_pl, mlxsw_sp_port->lag_id);
+
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl);
+}
+
+static int
+mlxsw_sp_port_fdb_flush_by_lag_id_fid(const struct mlxsw_sp_port *mlxsw_sp_port,
+				      u16 fid)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char sfdf_pl[MLXSW_REG_SFDF_LEN];
+
+	mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_LAG_AND_FID);
+	mlxsw_reg_sfdf_fid_set(sfdf_pl, fid);
+	mlxsw_reg_sfdf_lag_fid_lag_id_set(sfdf_pl, mlxsw_sp_port->lag_id);
+
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl);
+}
+
+static int
+__mlxsw_sp_port_fdb_flush(const struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	int err, last_err = 0;
+	u16 vid;
+
+	for (vid = 1; vid < VLAN_N_VID - 1; vid++) {
+		err = mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, vid);
+		if (err)
+			last_err = err;
+	}
+
+	return last_err;
+}
+
+static int
+__mlxsw_sp_port_fdb_flush_lagged(const struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	int err, last_err = 0;
+	u16 vid;
+
+	for (vid = 1; vid < VLAN_N_VID - 1; vid++) {
+		err = mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_port, vid);
+		if (err)
+			last_err = err;
+	}
+
+	return last_err;
+}
+
+static int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	if (!list_empty(&mlxsw_sp_port->vports_list))
+		if (mlxsw_sp_port->lagged)
+			return __mlxsw_sp_port_fdb_flush_lagged(mlxsw_sp_port);
+		else
+			return __mlxsw_sp_port_fdb_flush(mlxsw_sp_port);
+	else
+		if (mlxsw_sp_port->lagged)
+			return mlxsw_sp_port_fdb_flush_by_lag_id(mlxsw_sp_port);
+		else
+			return mlxsw_sp_port_fdb_flush_by_port(mlxsw_sp_port);
+}
+
+static int mlxsw_sp_vport_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_vport)
+{
+	u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_vport);
+	u16 fid = mlxsw_sp_vfid_to_fid(vfid);
+
+	if (mlxsw_sp_vport->lagged)
+		return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_vport,
+							     fid);
+	else
+		return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_vport, fid);
+}
+
 static bool mlxsw_sp_port_dev_check(const struct net_device *dev)
 {
 	return dev->netdev_ops == &mlxsw_sp_port_netdev_ops;
@@ -1824,24 +2105,35 @@ static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port)
 	 */
 	err = mlxsw_sp_port_kill_vid(dev, 0, 1);
 	if (err)
-		netdev_err(dev, "Failed to remove VID 1\n");
+		return err;
 
-	return err;
+	mlxsw_sp_port->learning = 1;
+	mlxsw_sp_port->learning_sync = 1;
+	mlxsw_sp_port->uc_flood = 1;
+	mlxsw_sp_port->bridged = 1;
+
+	return 0;
 }
 
-static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port)
+static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port,
+				      bool flush_fdb)
 {
 	struct net_device *dev = mlxsw_sp_port->dev;
-	int err;
+
+	if (flush_fdb && mlxsw_sp_port_fdb_flush(mlxsw_sp_port))
+		netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n");
+
+	mlxsw_sp_port_pvid_set(mlxsw_sp_port, 1);
+
+	mlxsw_sp_port->learning = 0;
+	mlxsw_sp_port->learning_sync = 0;
+	mlxsw_sp_port->uc_flood = 0;
+	mlxsw_sp_port->bridged = 0;
 
 	/* Add implicit VLAN interface in the device, so that untagged
 	 * packets will be classified to the default vFID.
 	 */
-	err = mlxsw_sp_port_add_vid(dev, 0, 1);
-	if (err)
-		netdev_err(dev, "Failed to add VID 1\n");
-
-	return err;
+	return mlxsw_sp_port_add_vid(dev, 0, 1);
 }
 
 static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp,
@@ -1865,19 +2157,318 @@ static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp,
 		mlxsw_sp->master_bridge.dev = NULL;
 }
 
-static int mlxsw_sp_netdevice_event(struct notifier_block *unused,
-				    unsigned long event, void *ptr)
+static int mlxsw_sp_lag_create(struct mlxsw_sp *mlxsw_sp, u16 lag_id)
+{
+	char sldr_pl[MLXSW_REG_SLDR_LEN];
+
+	mlxsw_reg_sldr_lag_create_pack(sldr_pl, lag_id);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sldr), sldr_pl);
+}
+
+static int mlxsw_sp_lag_destroy(struct mlxsw_sp *mlxsw_sp, u16 lag_id)
+{
+	char sldr_pl[MLXSW_REG_SLDR_LEN];
+
+	mlxsw_reg_sldr_lag_destroy_pack(sldr_pl, lag_id);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sldr), sldr_pl);
+}
+
+static int mlxsw_sp_lag_col_port_add(struct mlxsw_sp_port *mlxsw_sp_port,
+				     u16 lag_id, u8 port_index)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char slcor_pl[MLXSW_REG_SLCOR_LEN];
+
+	mlxsw_reg_slcor_port_add_pack(slcor_pl, mlxsw_sp_port->local_port,
+				      lag_id, port_index);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(slcor), slcor_pl);
+}
+
+static int mlxsw_sp_lag_col_port_remove(struct mlxsw_sp_port *mlxsw_sp_port,
+					u16 lag_id)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char slcor_pl[MLXSW_REG_SLCOR_LEN];
+
+	mlxsw_reg_slcor_port_remove_pack(slcor_pl, mlxsw_sp_port->local_port,
+					 lag_id);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(slcor), slcor_pl);
+}
+
+static int mlxsw_sp_lag_col_port_enable(struct mlxsw_sp_port *mlxsw_sp_port,
+					u16 lag_id)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char slcor_pl[MLXSW_REG_SLCOR_LEN];
+
+	mlxsw_reg_slcor_col_enable_pack(slcor_pl, mlxsw_sp_port->local_port,
+					lag_id);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(slcor), slcor_pl);
+}
+
+static int mlxsw_sp_lag_col_port_disable(struct mlxsw_sp_port *mlxsw_sp_port,
+					 u16 lag_id)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char slcor_pl[MLXSW_REG_SLCOR_LEN];
+
+	mlxsw_reg_slcor_col_disable_pack(slcor_pl, mlxsw_sp_port->local_port,
+					 lag_id);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(slcor), slcor_pl);
+}
+
+static int mlxsw_sp_lag_index_get(struct mlxsw_sp *mlxsw_sp,
+				  struct net_device *lag_dev,
+				  u16 *p_lag_id)
+{
+	struct mlxsw_sp_upper *lag;
+	int free_lag_id = -1;
+	int i;
+
+	for (i = 0; i < MLXSW_SP_LAG_MAX; i++) {
+		lag = mlxsw_sp_lag_get(mlxsw_sp, i);
+		if (lag->ref_count) {
+			if (lag->dev == lag_dev) {
+				*p_lag_id = i;
+				return 0;
+			}
+		} else if (free_lag_id < 0) {
+			free_lag_id = i;
+		}
+	}
+	if (free_lag_id < 0)
+		return -EBUSY;
+	*p_lag_id = free_lag_id;
+	return 0;
+}
+
+static bool
+mlxsw_sp_master_lag_check(struct mlxsw_sp *mlxsw_sp,
+			  struct net_device *lag_dev,
+			  struct netdev_lag_upper_info *lag_upper_info)
+{
+	u16 lag_id;
+
+	if (mlxsw_sp_lag_index_get(mlxsw_sp, lag_dev, &lag_id) != 0)
+		return false;
+	if (lag_upper_info->tx_type != NETDEV_LAG_TX_TYPE_HASH)
+		return false;
+	return true;
+}
+
+static int mlxsw_sp_port_lag_index_get(struct mlxsw_sp *mlxsw_sp,
+				       u16 lag_id, u8 *p_port_index)
+{
+	int i;
+
+	for (i = 0; i < MLXSW_SP_PORT_PER_LAG_MAX; i++) {
+		if (!mlxsw_sp_port_lagged_get(mlxsw_sp, lag_id, i)) {
+			*p_port_index = i;
+			return 0;
+		}
+	}
+	return -EBUSY;
+}
+
+static int mlxsw_sp_port_lag_join(struct mlxsw_sp_port *mlxsw_sp_port,
+				  struct net_device *lag_dev)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	struct mlxsw_sp_upper *lag;
+	u16 lag_id;
+	u8 port_index;
+	int err;
+
+	err = mlxsw_sp_lag_index_get(mlxsw_sp, lag_dev, &lag_id);
+	if (err)
+		return err;
+	lag = mlxsw_sp_lag_get(mlxsw_sp, lag_id);
+	if (!lag->ref_count) {
+		err = mlxsw_sp_lag_create(mlxsw_sp, lag_id);
+		if (err)
+			return err;
+		lag->dev = lag_dev;
+	}
+
+	err = mlxsw_sp_port_lag_index_get(mlxsw_sp, lag_id, &port_index);
+	if (err)
+		return err;
+	err = mlxsw_sp_lag_col_port_add(mlxsw_sp_port, lag_id, port_index);
+	if (err)
+		goto err_col_port_add;
+	err = mlxsw_sp_lag_col_port_enable(mlxsw_sp_port, lag_id);
+	if (err)
+		goto err_col_port_enable;
+
+	mlxsw_core_lag_mapping_set(mlxsw_sp->core, lag_id, port_index,
+				   mlxsw_sp_port->local_port);
+	mlxsw_sp_port->lag_id = lag_id;
+	mlxsw_sp_port->lagged = 1;
+	lag->ref_count++;
+	return 0;
+
+err_col_port_add:
+	if (!lag->ref_count)
+		mlxsw_sp_lag_destroy(mlxsw_sp, lag_id);
+err_col_port_enable:
+	mlxsw_sp_lag_col_port_remove(mlxsw_sp_port, lag_id);
+	return err;
+}
+
+static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport,
+				       struct net_device *br_dev,
+				       bool flush_fdb);
+
+static int mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port,
+				   struct net_device *lag_dev)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+	struct mlxsw_sp_upper *lag;
+	u16 lag_id = mlxsw_sp_port->lag_id;
+	int err;
+
+	if (!mlxsw_sp_port->lagged)
+		return 0;
+	lag = mlxsw_sp_lag_get(mlxsw_sp, lag_id);
+	WARN_ON(lag->ref_count == 0);
+
+	err = mlxsw_sp_lag_col_port_disable(mlxsw_sp_port, lag_id);
+	if (err)
+		return err;
+	err = mlxsw_sp_lag_col_port_remove(mlxsw_sp_port, lag_id);
+	if (err)
+		return err;
+
+	/* In case we leave a LAG device that has bridges built on top,
+	 * then their teardown sequence is never issued and we need to
+	 * invoke the necessary cleanup routines ourselves.
+	 */
+	list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list,
+			    vport.list) {
+		struct net_device *br_dev;
+
+		if (!mlxsw_sp_vport->bridged)
+			continue;
+
+		br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport);
+		mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, br_dev, false);
+	}
+
+	if (mlxsw_sp_port->bridged) {
+		mlxsw_sp_port_active_vlans_del(mlxsw_sp_port);
+		mlxsw_sp_port_bridge_leave(mlxsw_sp_port, false);
+		mlxsw_sp_master_bridge_dec(mlxsw_sp, NULL);
+	}
+
+	if (lag->ref_count == 1) {
+		if (mlxsw_sp_port_fdb_flush_by_lag_id(mlxsw_sp_port))
+			netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n");
+		err = mlxsw_sp_lag_destroy(mlxsw_sp, lag_id);
+		if (err)
+			return err;
+	}
+
+	mlxsw_core_lag_mapping_clear(mlxsw_sp->core, lag_id,
+				     mlxsw_sp_port->local_port);
+	mlxsw_sp_port->lagged = 0;
+	lag->ref_count--;
+	return 0;
+}
+
+static int mlxsw_sp_lag_dist_port_add(struct mlxsw_sp_port *mlxsw_sp_port,
+				      u16 lag_id)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char sldr_pl[MLXSW_REG_SLDR_LEN];
+
+	mlxsw_reg_sldr_lag_add_port_pack(sldr_pl, lag_id,
+					 mlxsw_sp_port->local_port);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sldr), sldr_pl);
+}
+
+static int mlxsw_sp_lag_dist_port_remove(struct mlxsw_sp_port *mlxsw_sp_port,
+					 u16 lag_id)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char sldr_pl[MLXSW_REG_SLDR_LEN];
+
+	mlxsw_reg_sldr_lag_remove_port_pack(sldr_pl, lag_id,
+					    mlxsw_sp_port->local_port);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sldr), sldr_pl);
+}
+
+static int mlxsw_sp_port_lag_tx_en_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				       bool lag_tx_enabled)
+{
+	if (lag_tx_enabled)
+		return mlxsw_sp_lag_dist_port_add(mlxsw_sp_port,
+						  mlxsw_sp_port->lag_id);
+	else
+		return mlxsw_sp_lag_dist_port_remove(mlxsw_sp_port,
+						     mlxsw_sp_port->lag_id);
+}
+
+static int mlxsw_sp_port_lag_changed(struct mlxsw_sp_port *mlxsw_sp_port,
+				     struct netdev_lag_lower_state_info *info)
+{
+	return mlxsw_sp_port_lag_tx_en_set(mlxsw_sp_port, info->tx_enabled);
+}
+
+static int mlxsw_sp_port_vlan_link(struct mlxsw_sp_port *mlxsw_sp_port,
+				   struct net_device *vlan_dev)
+{
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+	u16 vid = vlan_dev_vlan_id(vlan_dev);
+
+	mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid);
+	if (!mlxsw_sp_vport) {
+		WARN_ON(!mlxsw_sp_vport);
+		return -EINVAL;
+	}
+
+	mlxsw_sp_vport->dev = vlan_dev;
+
+	return 0;
+}
+
+static int mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port,
+				     struct net_device *vlan_dev)
+{
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+	u16 vid = vlan_dev_vlan_id(vlan_dev);
+
+	mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid);
+	if (!mlxsw_sp_vport) {
+		WARN_ON(!mlxsw_sp_vport);
+		return -EINVAL;
+	}
+
+	/* When removing a VLAN device while still bridged we should first
+	 * remove it from the bridge, as we receive the bridge's notification
+	 * when the vPort is already gone.
+	 */
+	if (mlxsw_sp_vport->bridged) {
+		struct net_device *br_dev;
+
+		br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport);
+		mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, br_dev, true);
+	}
+
+	mlxsw_sp_vport->dev = mlxsw_sp_port->dev;
+
+	return 0;
+}
+
+static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev,
+					       unsigned long event, void *ptr)
 {
-	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
 	struct netdev_notifier_changeupper_info *info;
 	struct mlxsw_sp_port *mlxsw_sp_port;
 	struct net_device *upper_dev;
 	struct mlxsw_sp *mlxsw_sp;
 	int err;
 
-	if (!mlxsw_sp_port_dev_check(dev))
-		return NOTIFY_DONE;
-
 	mlxsw_sp_port = netdev_priv(dev);
 	mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	info = ptr;
@@ -1885,28 +2476,67 @@ static int mlxsw_sp_netdevice_event(struct notifier_block *unused,
 	switch (event) {
 	case NETDEV_PRECHANGEUPPER:
 		upper_dev = info->upper_dev;
+		if (!info->master || !info->linking)
+			break;
 		/* HW limitation forbids to put ports to multiple bridges. */
-		if (info->master && info->linking &&
-		    netif_is_bridge_master(upper_dev) &&
+		if (netif_is_bridge_master(upper_dev) &&
 		    !mlxsw_sp_master_bridge_check(mlxsw_sp, upper_dev))
 			return NOTIFY_BAD;
+		if (netif_is_lag_master(upper_dev) &&
+		    !mlxsw_sp_master_lag_check(mlxsw_sp, upper_dev,
+					       info->upper_info))
+			return NOTIFY_BAD;
 		break;
 	case NETDEV_CHANGEUPPER:
 		upper_dev = info->upper_dev;
-		if (info->master &&
-		    netif_is_bridge_master(upper_dev)) {
+		if (is_vlan_dev(upper_dev)) {
+			if (info->linking) {
+				err = mlxsw_sp_port_vlan_link(mlxsw_sp_port,
+							      upper_dev);
+				if (err) {
+					netdev_err(dev, "Failed to link VLAN device\n");
+					return NOTIFY_BAD;
+				}
+			} else {
+				err = mlxsw_sp_port_vlan_unlink(mlxsw_sp_port,
+								upper_dev);
+				if (err) {
+					netdev_err(dev, "Failed to unlink VLAN device\n");
+					return NOTIFY_BAD;
+				}
+			}
+		} else if (netif_is_bridge_master(upper_dev)) {
 			if (info->linking) {
 				err = mlxsw_sp_port_bridge_join(mlxsw_sp_port);
-				if (err)
+				if (err) {
 					netdev_err(dev, "Failed to join bridge\n");
+					return NOTIFY_BAD;
+				}
 				mlxsw_sp_master_bridge_inc(mlxsw_sp, upper_dev);
-				mlxsw_sp_port->bridged = 1;
 			} else {
-				err = mlxsw_sp_port_bridge_leave(mlxsw_sp_port);
-				if (err)
-					netdev_err(dev, "Failed to leave bridge\n");
-				mlxsw_sp_port->bridged = 0;
+				err = mlxsw_sp_port_bridge_leave(mlxsw_sp_port,
+								 true);
 				mlxsw_sp_master_bridge_dec(mlxsw_sp, upper_dev);
+				if (err) {
+					netdev_err(dev, "Failed to leave bridge\n");
+					return NOTIFY_BAD;
+				}
+			}
+		} else if (netif_is_lag_master(upper_dev)) {
+			if (info->linking) {
+				err = mlxsw_sp_port_lag_join(mlxsw_sp_port,
+							     upper_dev);
+				if (err) {
+					netdev_err(dev, "Failed to join link aggregation\n");
+					return NOTIFY_BAD;
+				}
+			} else {
+				err = mlxsw_sp_port_lag_leave(mlxsw_sp_port,
+							      upper_dev);
+				if (err) {
+					netdev_err(dev, "Failed to leave link aggregation\n");
+					return NOTIFY_BAD;
+				}
 			}
 		}
 		break;
@@ -1915,6 +2545,455 @@ static int mlxsw_sp_netdevice_event(struct notifier_block *unused,
 	return NOTIFY_DONE;
 }
 
+static int mlxsw_sp_netdevice_port_lower_event(struct net_device *dev,
+					       unsigned long event, void *ptr)
+{
+	struct netdev_notifier_changelowerstate_info *info;
+	struct mlxsw_sp_port *mlxsw_sp_port;
+	int err;
+
+	mlxsw_sp_port = netdev_priv(dev);
+	info = ptr;
+
+	switch (event) {
+	case NETDEV_CHANGELOWERSTATE:
+		if (netif_is_lag_port(dev) && mlxsw_sp_port->lagged) {
+			err = mlxsw_sp_port_lag_changed(mlxsw_sp_port,
+							info->lower_state_info);
+			if (err)
+				netdev_err(dev, "Failed to reflect link aggregation lower state change\n");
+		}
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static int mlxsw_sp_netdevice_port_event(struct net_device *dev,
+					 unsigned long event, void *ptr)
+{
+	switch (event) {
+	case NETDEV_PRECHANGEUPPER:
+	case NETDEV_CHANGEUPPER:
+		return mlxsw_sp_netdevice_port_upper_event(dev, event, ptr);
+	case NETDEV_CHANGELOWERSTATE:
+		return mlxsw_sp_netdevice_port_lower_event(dev, event, ptr);
+	}
+
+	return NOTIFY_DONE;
+}
+
+static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev,
+					unsigned long event, void *ptr)
+{
+	struct net_device *dev;
+	struct list_head *iter;
+	int ret;
+
+	netdev_for_each_lower_dev(lag_dev, dev, iter) {
+		if (mlxsw_sp_port_dev_check(dev)) {
+			ret = mlxsw_sp_netdevice_port_event(dev, event, ptr);
+			if (ret == NOTIFY_BAD)
+				return ret;
+		}
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct mlxsw_sp_vfid *
+mlxsw_sp_br_vfid_find(const struct mlxsw_sp *mlxsw_sp,
+		      const struct net_device *br_dev)
+{
+	struct mlxsw_sp_vfid *vfid;
+
+	list_for_each_entry(vfid, &mlxsw_sp->br_vfids.list, list) {
+		if (vfid->br_dev == br_dev)
+			return vfid;
+	}
+
+	return NULL;
+}
+
+static u16 mlxsw_sp_vfid_to_br_vfid(u16 vfid)
+{
+	return vfid - MLXSW_SP_VFID_PORT_MAX;
+}
+
+static u16 mlxsw_sp_br_vfid_to_vfid(u16 br_vfid)
+{
+	return MLXSW_SP_VFID_PORT_MAX + br_vfid;
+}
+
+static u16 mlxsw_sp_avail_br_vfid_get(const struct mlxsw_sp *mlxsw_sp)
+{
+	return find_first_zero_bit(mlxsw_sp->br_vfids.mapped,
+				   MLXSW_SP_VFID_BR_MAX);
+}
+
+static struct mlxsw_sp_vfid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp,
+						     struct net_device *br_dev)
+{
+	struct device *dev = mlxsw_sp->bus_info->dev;
+	struct mlxsw_sp_vfid *vfid;
+	u16 n_vfid;
+	int err;
+
+	n_vfid = mlxsw_sp_br_vfid_to_vfid(mlxsw_sp_avail_br_vfid_get(mlxsw_sp));
+	if (n_vfid == MLXSW_SP_VFID_MAX) {
+		dev_err(dev, "No available vFIDs\n");
+		return ERR_PTR(-ERANGE);
+	}
+
+	err = __mlxsw_sp_vfid_create(mlxsw_sp, n_vfid);
+	if (err) {
+		dev_err(dev, "Failed to create vFID=%d\n", n_vfid);
+		return ERR_PTR(err);
+	}
+
+	vfid = kzalloc(sizeof(*vfid), GFP_KERNEL);
+	if (!vfid)
+		goto err_allocate_vfid;
+
+	vfid->vfid = n_vfid;
+	vfid->br_dev = br_dev;
+
+	list_add(&vfid->list, &mlxsw_sp->br_vfids.list);
+	set_bit(mlxsw_sp_vfid_to_br_vfid(n_vfid), mlxsw_sp->br_vfids.mapped);
+
+	return vfid;
+
+err_allocate_vfid:
+	__mlxsw_sp_vfid_destroy(mlxsw_sp, n_vfid);
+	return ERR_PTR(-ENOMEM);
+}
+
+static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp,
+				     struct mlxsw_sp_vfid *vfid)
+{
+	u16 br_vfid = mlxsw_sp_vfid_to_br_vfid(vfid->vfid);
+
+	clear_bit(br_vfid, mlxsw_sp->br_vfids.mapped);
+	list_del(&vfid->list);
+
+	__mlxsw_sp_vfid_destroy(mlxsw_sp, vfid->vfid);
+
+	kfree(vfid);
+}
+
+static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport,
+				       struct net_device *br_dev,
+				       bool flush_fdb)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp;
+	u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport);
+	struct net_device *dev = mlxsw_sp_vport->dev;
+	struct mlxsw_sp_vfid *vfid, *new_vfid;
+	int err;
+
+	vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev);
+	if (!vfid) {
+		WARN_ON(!vfid);
+		return -EINVAL;
+	}
+
+	/* We need a vFID to go back to after leaving the bridge's vFID. */
+	new_vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid);
+	if (!new_vfid) {
+		new_vfid = mlxsw_sp_vfid_create(mlxsw_sp, vid);
+		if (IS_ERR(new_vfid)) {
+			netdev_err(dev, "Failed to create vFID for VID=%d\n",
+				   vid);
+			return PTR_ERR(new_vfid);
+		}
+	}
+
+	/* Invalidate existing {Port, VID} to vFID mapping and create a new
+	 * one for the new vFID.
+	 */
+	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
+					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
+					   false,
+					   mlxsw_sp_vfid_to_fid(vfid->vfid),
+					   vid);
+	if (err) {
+		netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n",
+			   vfid->vfid);
+		goto err_port_vid_to_fid_invalidate;
+	}
+
+	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
+					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
+					   true,
+					   mlxsw_sp_vfid_to_fid(new_vfid->vfid),
+					   vid);
+	if (err) {
+		netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n",
+			   new_vfid->vfid);
+		goto err_port_vid_to_fid_validate;
+	}
+
+	err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false);
+	if (err) {
+		netdev_err(dev, "Failed to disable learning\n");
+		goto err_port_vid_learning_set;
+	}
+
+	err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false,
+				       false);
+	if (err) {
+		netdev_err(dev, "Failed clear to clear flooding\n");
+		goto err_vport_flood_set;
+	}
+
+	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid,
+					  MLXSW_REG_SPMS_STATE_FORWARDING);
+	if (err) {
+		netdev_err(dev, "Failed to set STP state\n");
+		goto err_port_stp_state_set;
+	}
+
+	if (flush_fdb && mlxsw_sp_vport_fdb_flush(mlxsw_sp_vport))
+		netdev_err(dev, "Failed to flush FDB\n");
+
+	/* Switch between the vFIDs and destroy the old one if needed. */
+	new_vfid->nr_vports++;
+	mlxsw_sp_vport->vport.vfid = new_vfid;
+	vfid->nr_vports--;
+	if (!vfid->nr_vports)
+		mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid);
+
+	mlxsw_sp_vport->learning = 0;
+	mlxsw_sp_vport->learning_sync = 0;
+	mlxsw_sp_vport->uc_flood = 0;
+	mlxsw_sp_vport->bridged = 0;
+
+	return 0;
+
+err_port_stp_state_set:
+err_vport_flood_set:
+err_port_vid_learning_set:
+err_port_vid_to_fid_validate:
+err_port_vid_to_fid_invalidate:
+	/* Rollback vFID only if new. */
+	if (!new_vfid->nr_vports)
+		mlxsw_sp_vfid_destroy(mlxsw_sp, new_vfid);
+	return err;
+}
+
+static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport,
+				      struct net_device *br_dev)
+{
+	struct mlxsw_sp_vfid *old_vfid = mlxsw_sp_vport->vport.vfid;
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp;
+	u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport);
+	struct net_device *dev = mlxsw_sp_vport->dev;
+	struct mlxsw_sp_vfid *vfid;
+	int err;
+
+	vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev);
+	if (!vfid) {
+		vfid = mlxsw_sp_br_vfid_create(mlxsw_sp, br_dev);
+		if (IS_ERR(vfid)) {
+			netdev_err(dev, "Failed to create bridge vFID\n");
+			return PTR_ERR(vfid);
+		}
+	}
+
+	err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, true, false);
+	if (err) {
+		netdev_err(dev, "Failed to setup flooding for vFID=%d\n",
+			   vfid->vfid);
+		goto err_port_flood_set;
+	}
+
+	err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true);
+	if (err) {
+		netdev_err(dev, "Failed to enable learning\n");
+		goto err_port_vid_learning_set;
+	}
+
+	/* We need to invalidate existing {Port, VID} to vFID mapping and
+	 * create a new one for the bridge's vFID.
+	 */
+	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
+					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
+					   false,
+					   mlxsw_sp_vfid_to_fid(old_vfid->vfid),
+					   vid);
+	if (err) {
+		netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n",
+			   old_vfid->vfid);
+		goto err_port_vid_to_fid_invalidate;
+	}
+
+	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
+					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
+					   true,
+					   mlxsw_sp_vfid_to_fid(vfid->vfid),
+					   vid);
+	if (err) {
+		netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n",
+			   vfid->vfid);
+		goto err_port_vid_to_fid_validate;
+	}
+
+	/* Switch between the vFIDs and destroy the old one if needed. */
+	vfid->nr_vports++;
+	mlxsw_sp_vport->vport.vfid = vfid;
+	old_vfid->nr_vports--;
+	if (!old_vfid->nr_vports)
+		mlxsw_sp_vfid_destroy(mlxsw_sp, old_vfid);
+
+	mlxsw_sp_vport->learning = 1;
+	mlxsw_sp_vport->learning_sync = 1;
+	mlxsw_sp_vport->uc_flood = 1;
+	mlxsw_sp_vport->bridged = 1;
+
+	return 0;
+
+err_port_vid_to_fid_validate:
+	mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport,
+				     MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false,
+				     mlxsw_sp_vfid_to_fid(old_vfid->vfid), vid);
+err_port_vid_to_fid_invalidate:
+	mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false);
+err_port_vid_learning_set:
+	mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false, false);
+err_port_flood_set:
+	if (!vfid->nr_vports)
+		mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid);
+	return err;
+}
+
+static bool
+mlxsw_sp_port_master_bridge_check(const struct mlxsw_sp_port *mlxsw_sp_port,
+				  const struct net_device *br_dev)
+{
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+
+	list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list,
+			    vport.list) {
+		if (mlxsw_sp_vport_br_get(mlxsw_sp_vport) == br_dev)
+			return false;
+	}
+
+	return true;
+}
+
+static int mlxsw_sp_netdevice_vport_event(struct net_device *dev,
+					  unsigned long event, void *ptr,
+					  u16 vid)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct netdev_notifier_changeupper_info *info = ptr;
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+	struct net_device *upper_dev;
+	int err;
+
+	mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid);
+
+	switch (event) {
+	case NETDEV_PRECHANGEUPPER:
+		upper_dev = info->upper_dev;
+		if (!info->master || !info->linking)
+			break;
+		if (!netif_is_bridge_master(upper_dev))
+			return NOTIFY_BAD;
+		/* We can't have multiple VLAN interfaces configured on
+		 * the same port and being members in the same bridge.
+		 */
+		if (!mlxsw_sp_port_master_bridge_check(mlxsw_sp_port,
+						       upper_dev))
+			return NOTIFY_BAD;
+		break;
+	case NETDEV_CHANGEUPPER:
+		upper_dev = info->upper_dev;
+		if (!info->master)
+			break;
+		if (info->linking) {
+			if (!mlxsw_sp_vport) {
+				WARN_ON(!mlxsw_sp_vport);
+				return NOTIFY_BAD;
+			}
+			err = mlxsw_sp_vport_bridge_join(mlxsw_sp_vport,
+							 upper_dev);
+			if (err) {
+				netdev_err(dev, "Failed to join bridge\n");
+				return NOTIFY_BAD;
+			}
+		} else {
+			/* We ignore bridge's unlinking notifications if vPort
+			 * is gone, since we already left the bridge when the
+			 * VLAN device was unlinked from the real device.
+			 */
+			if (!mlxsw_sp_vport)
+				return NOTIFY_DONE;
+			err = mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport,
+							  upper_dev, true);
+			if (err) {
+				netdev_err(dev, "Failed to leave bridge\n");
+				return NOTIFY_BAD;
+			}
+		}
+	}
+
+	return NOTIFY_DONE;
+}
+
+static int mlxsw_sp_netdevice_lag_vport_event(struct net_device *lag_dev,
+					      unsigned long event, void *ptr,
+					      u16 vid)
+{
+	struct net_device *dev;
+	struct list_head *iter;
+	int ret;
+
+	netdev_for_each_lower_dev(lag_dev, dev, iter) {
+		if (mlxsw_sp_port_dev_check(dev)) {
+			ret = mlxsw_sp_netdevice_vport_event(dev, event, ptr,
+							     vid);
+			if (ret == NOTIFY_BAD)
+				return ret;
+		}
+	}
+
+	return NOTIFY_DONE;
+}
+
+static int mlxsw_sp_netdevice_vlan_event(struct net_device *vlan_dev,
+					 unsigned long event, void *ptr)
+{
+	struct net_device *real_dev = vlan_dev_real_dev(vlan_dev);
+	u16 vid = vlan_dev_vlan_id(vlan_dev);
+
+	if (mlxsw_sp_port_dev_check(real_dev))
+		return mlxsw_sp_netdevice_vport_event(real_dev, event, ptr,
+						      vid);
+	else if (netif_is_lag_master(real_dev))
+		return mlxsw_sp_netdevice_lag_vport_event(real_dev, event, ptr,
+							  vid);
+
+	return NOTIFY_DONE;
+}
+
+static int mlxsw_sp_netdevice_event(struct notifier_block *unused,
+				    unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+
+	if (mlxsw_sp_port_dev_check(dev))
+		return mlxsw_sp_netdevice_port_event(dev, event, ptr);
+
+	if (netif_is_lag_master(dev))
+		return mlxsw_sp_netdevice_lag_event(dev, event, ptr);
+
+	if (is_vlan_dev(dev))
+		return mlxsw_sp_netdevice_vlan_event(dev, event, ptr);
+
+	return NOTIFY_DONE;
+}
+
 static struct notifier_block mlxsw_sp_netdevice_nb __read_mostly = {
 	.notifier_call = mlxsw_sp_netdevice_event,
 };
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 4365c8bcc..3b89ed2f3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -41,16 +41,73 @@
 #include <linux/netdevice.h>
 #include <linux/bitops.h>
 #include <linux/if_vlan.h>
+#include <linux/list.h>
 #include <net/switchdev.h>
 
+#include "port.h"
 #include "core.h"
 
 #define MLXSW_SP_VFID_BASE VLAN_N_VID
+#define MLXSW_SP_VFID_PORT_MAX 512	/* Non-bridged VLAN interfaces */
+#define MLXSW_SP_VFID_BR_MAX 8192	/* Bridged VLAN interfaces */
+#define MLXSW_SP_VFID_MAX (MLXSW_SP_VFID_PORT_MAX + MLXSW_SP_VFID_BR_MAX)
+
+#define MLXSW_SP_LAG_MAX 64
+#define MLXSW_SP_PORT_PER_LAG_MAX 16
+
+#define MLXSW_SP_MID_MAX 7000
 
 struct mlxsw_sp_port;
 
+struct mlxsw_sp_upper {
+	struct net_device *dev;
+	unsigned int ref_count;
+};
+
+struct mlxsw_sp_vfid {
+	struct list_head list;
+	u16 nr_vports;
+	u16 vfid;	/* Starting at 0 */
+	struct net_device *br_dev;
+	u16 vid;
+};
+
+struct mlxsw_sp_mid {
+	struct list_head list;
+	unsigned char addr[ETH_ALEN];
+	u16 vid;
+	u16 mid;
+	unsigned int ref_count;
+};
+
+static inline u16 mlxsw_sp_vfid_to_fid(u16 vfid)
+{
+	return MLXSW_SP_VFID_BASE + vfid;
+}
+
+static inline u16 mlxsw_sp_fid_to_vfid(u16 fid)
+{
+	return fid - MLXSW_SP_VFID_BASE;
+}
+
+static inline bool mlxsw_sp_fid_is_vfid(u16 fid)
+{
+	return fid >= MLXSW_SP_VFID_BASE;
+}
+
 struct mlxsw_sp {
-	unsigned long active_vfids[BITS_TO_LONGS(VLAN_N_VID)];
+	struct {
+		struct list_head list;
+		unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_VFID_PORT_MAX)];
+	} port_vfids;
+	struct {
+		struct list_head list;
+		unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_VFID_BR_MAX)];
+	} br_vfids;
+	struct {
+		struct list_head list;
+		unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_MID_MAX)];
+	} br_mids;
 	unsigned long active_fids[BITS_TO_LONGS(VLAN_N_VID)];
 	struct mlxsw_sp_port **ports;
 	struct mlxsw_core *core;
@@ -63,12 +120,16 @@ struct mlxsw_sp {
 	} fdb_notify;
 #define MLXSW_SP_DEFAULT_AGEING_TIME 300
 	u32 ageing_time;
-	struct {
-		struct net_device *dev;
-		unsigned int ref_count;
-	} master_bridge;
+	struct mlxsw_sp_upper master_bridge;
+	struct mlxsw_sp_upper lags[MLXSW_SP_LAG_MAX];
 };
 
+static inline struct mlxsw_sp_upper *
+mlxsw_sp_lag_get(struct mlxsw_sp *mlxsw_sp, u16 lag_id)
+{
+	return &mlxsw_sp->lags[lag_id];
+}
+
 struct mlxsw_sp_port_pcpu_stats {
 	u64			rx_packets;
 	u64			rx_bytes;
@@ -87,15 +148,87 @@ struct mlxsw_sp_port {
 	u8 learning:1,
 	   learning_sync:1,
 	   uc_flood:1,
-	   bridged:1;
+	   bridged:1,
+	   lagged:1;
 	u16 pvid;
+	u16 lag_id;
+	struct {
+		struct list_head list;
+		struct mlxsw_sp_vfid *vfid;
+		u16 vid;
+	} vport;
 	/* 802.1Q bridge VLANs */
-	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
+	unsigned long *active_vlans;
+	unsigned long *untagged_vlans;
 	/* VLAN interfaces */
-	unsigned long active_vfids[BITS_TO_LONGS(VLAN_N_VID)];
-	u16 nr_vfids;
+	struct list_head vports_list;
 };
 
+static inline struct mlxsw_sp_port *
+mlxsw_sp_port_lagged_get(struct mlxsw_sp *mlxsw_sp, u16 lag_id, u8 port_index)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port;
+	u8 local_port;
+
+	local_port = mlxsw_core_lag_mapping_get(mlxsw_sp->core,
+						lag_id, port_index);
+	mlxsw_sp_port = mlxsw_sp->ports[local_port];
+	return mlxsw_sp_port && mlxsw_sp_port->lagged ? mlxsw_sp_port : NULL;
+}
+
+static inline bool
+mlxsw_sp_port_is_vport(const struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	return mlxsw_sp_port->vport.vfid;
+}
+
+static inline struct net_device *
+mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport)
+{
+	return mlxsw_sp_vport->vport.vfid->br_dev;
+}
+
+static inline u16
+mlxsw_sp_vport_vid_get(const struct mlxsw_sp_port *mlxsw_sp_vport)
+{
+	return mlxsw_sp_vport->vport.vid;
+}
+
+static inline u16
+mlxsw_sp_vport_vfid_get(const struct mlxsw_sp_port *mlxsw_sp_vport)
+{
+	return mlxsw_sp_vport->vport.vfid->vfid;
+}
+
+static inline struct mlxsw_sp_port *
+mlxsw_sp_port_vport_find(const struct mlxsw_sp_port *mlxsw_sp_port, u16 vid)
+{
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+
+	list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list,
+			    vport.list) {
+		if (mlxsw_sp_vport_vid_get(mlxsw_sp_vport) == vid)
+			return mlxsw_sp_vport;
+	}
+
+	return NULL;
+}
+
+static inline struct mlxsw_sp_port *
+mlxsw_sp_port_vport_find_by_vfid(const struct mlxsw_sp_port *mlxsw_sp_port,
+				 u16 vfid)
+{
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+
+	list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list,
+			    vport.list) {
+		if (mlxsw_sp_vport_vfid_get(mlxsw_sp_vport) == vfid)
+			return mlxsw_sp_vport;
+	}
+
+	return NULL;
+}
+
 enum mlxsw_sp_flood_table {
 	MLXSW_SP_FLOOD_TABLE_UC,
 	MLXSW_SP_FLOOD_TABLE_BM,
@@ -118,5 +251,9 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto,
 			  u16 vid);
 int mlxsw_sp_port_kill_vid(struct net_device *dev,
 			   __be16 __always_unused proto, u16 vid);
+int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid,
+			     bool set, bool only_uc);
+void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port);
+int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid);
 
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index 7dbeafa65..7b56098ac 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -52,12 +52,50 @@
 #include "core.h"
 #include "reg.h"
 
+static u16 mlxsw_sp_port_vid_to_fid_get(struct mlxsw_sp_port *mlxsw_sp_port,
+					u16 vid)
+{
+	u16 fid = vid;
+
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_port);
+
+		fid = mlxsw_sp_vfid_to_fid(vfid);
+	}
+
+	if (!fid)
+		fid = mlxsw_sp_port->pvid;
+
+	return fid;
+}
+
+static struct mlxsw_sp_port *
+mlxsw_sp_port_orig_get(struct net_device *dev,
+		       struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	struct mlxsw_sp_port *mlxsw_sp_vport;
+	u16 vid;
+
+	if (!is_vlan_dev(dev))
+		return mlxsw_sp_port;
+
+	vid = vlan_dev_vlan_id(dev);
+	mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid);
+	WARN_ON(!mlxsw_sp_vport);
+
+	return mlxsw_sp_vport;
+}
+
 static int mlxsw_sp_port_attr_get(struct net_device *dev,
 				  struct switchdev_attr *attr)
 {
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 
+	mlxsw_sp_port = mlxsw_sp_port_orig_get(attr->orig_dev, mlxsw_sp_port);
+	if (!mlxsw_sp_port)
+		return -EINVAL;
+
 	switch (attr->id) {
 	case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
 		attr->u.ppid.id_len = sizeof(mlxsw_sp->base_mac);
@@ -87,14 +125,14 @@ static int mlxsw_sp_port_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	int err;
 
 	switch (state) {
-	case BR_STATE_DISABLED: /* fall-through */
 	case BR_STATE_FORWARDING:
 		spms_state = MLXSW_REG_SPMS_STATE_FORWARDING;
 		break;
-	case BR_STATE_LISTENING: /* fall-through */
 	case BR_STATE_LEARNING:
 		spms_state = MLXSW_REG_SPMS_STATE_LEARNING;
 		break;
+	case BR_STATE_LISTENING: /* fall-through */
+	case BR_STATE_DISABLED: /* fall-through */
 	case BR_STATE_BLOCKING:
 		spms_state = MLXSW_REG_SPMS_STATE_DISCARDING;
 		break;
@@ -106,8 +144,14 @@ static int mlxsw_sp_port_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	if (!spms_pl)
 		return -ENOMEM;
 	mlxsw_reg_spms_pack(spms_pl, mlxsw_sp_port->local_port);
-	for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID)
+
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		vid = mlxsw_sp_vport_vid_get(mlxsw_sp_port);
 		mlxsw_reg_spms_vid_pack(spms_pl, vid, spms_state);
+	} else {
+		for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID)
+			mlxsw_reg_spms_vid_pack(spms_pl, vid, spms_state);
+	}
 
 	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spms), spms_pl);
 	kfree(spms_pl);
@@ -125,22 +169,38 @@ static int mlxsw_sp_port_attr_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	return mlxsw_sp_port_stp_state_set(mlxsw_sp_port, state);
 }
 
+static bool mlxsw_sp_vfid_is_vport_br(u16 vfid)
+{
+	return vfid >= MLXSW_SP_VFID_PORT_MAX;
+}
+
 static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port,
-				     u16 fid_begin, u16 fid_end, bool set,
+				     u16 idx_begin, u16 idx_end, bool set,
 				     bool only_uc)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
-	u16 range = fid_end - fid_begin + 1;
+	u16 local_port = mlxsw_sp_port->local_port;
+	enum mlxsw_flood_table_type table_type;
+	u16 range = idx_end - idx_begin + 1;
 	char *sftr_pl;
 	int err;
 
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID;
+		if (mlxsw_sp_vfid_is_vport_br(idx_begin))
+			local_port = mlxsw_sp_port->local_port;
+		else
+			local_port = MLXSW_PORT_CPU_PORT;
+	} else {
+		table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST;
+	}
+
 	sftr_pl = kmalloc(MLXSW_REG_SFTR_LEN, GFP_KERNEL);
 	if (!sftr_pl)
 		return -ENOMEM;
 
-	mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_UC, fid_begin,
-			    MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST, range,
-			    mlxsw_sp_port->local_port, set);
+	mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_UC, idx_begin,
+			    table_type, range, local_port, set);
 	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl);
 	if (err)
 		goto buffer_out;
@@ -151,9 +211,8 @@ static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	if (only_uc)
 		goto buffer_out;
 
-	mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_BM, fid_begin,
-			    MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST, range,
-			    mlxsw_sp_port->local_port, set);
+	mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_BM, idx_begin,
+			    table_type, range, local_port, set);
 	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl);
 
 buffer_out:
@@ -168,6 +227,13 @@ static int mlxsw_sp_port_uc_flood_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	u16 vid, last_visited_vid;
 	int err;
 
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_port);
+
+		return  __mlxsw_sp_port_flood_set(mlxsw_sp_port, vfid, vfid,
+						  set, true);
+	}
+
 	for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) {
 		err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid, vid, set,
 						true);
@@ -186,6 +252,16 @@ err_port_flood_set:
 	return err;
 }
 
+int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid,
+			     bool set, bool only_uc)
+{
+	/* In case of vFIDs, index into the flooding table is relative to
+	 * the start of the vFIDs range.
+	 */
+	return __mlxsw_sp_port_flood_set(mlxsw_sp_vport, vfid, vfid, set,
+					 only_uc);
+}
+
 static int mlxsw_sp_port_attr_br_flags_set(struct mlxsw_sp_port *mlxsw_sp_port,
 					   struct switchdev_trans *trans,
 					   unsigned long brport_flags)
@@ -194,6 +270,9 @@ static int mlxsw_sp_port_attr_br_flags_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	bool set;
 	int err;
 
+	if (!mlxsw_sp_port->bridged)
+		return -EINVAL;
+
 	if (switchdev_trans_ph_prepare(trans))
 		return 0;
 
@@ -238,6 +317,22 @@ static int mlxsw_sp_port_attr_br_ageing_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	return mlxsw_sp_ageing_set(mlxsw_sp, ageing_time);
 }
 
+static int mlxsw_sp_port_attr_br_vlan_set(struct mlxsw_sp_port *mlxsw_sp_port,
+					  struct switchdev_trans *trans,
+					  struct net_device *orig_dev,
+					  bool vlan_enabled)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+
+	/* SWITCHDEV_TRANS_PREPARE phase */
+	if ((!vlan_enabled) && (mlxsw_sp->master_bridge.dev == orig_dev)) {
+		netdev_err(mlxsw_sp_port->dev, "Bridge must be vlan-aware\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int mlxsw_sp_port_attr_set(struct net_device *dev,
 				  const struct switchdev_attr *attr,
 				  struct switchdev_trans *trans)
@@ -245,6 +340,10 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev,
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
 	int err = 0;
 
+	mlxsw_sp_port = mlxsw_sp_port_orig_get(attr->orig_dev, mlxsw_sp_port);
+	if (!mlxsw_sp_port)
+		return -EINVAL;
+
 	switch (attr->id) {
 	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
 		err = mlxsw_sp_port_attr_stp_state_set(mlxsw_sp_port, trans,
@@ -258,6 +357,11 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev,
 		err = mlxsw_sp_port_attr_br_ageing_set(mlxsw_sp_port, trans,
 						       attr->u.ageing_time);
 		break;
+	case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING:
+		err = mlxsw_sp_port_attr_br_vlan_set(mlxsw_sp_port, trans,
+						     attr->orig_dev,
+						     attr->u.vlan_filtering);
+		break;
 	default:
 		err = -EOPNOTSUPP;
 		break;
@@ -266,7 +370,8 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev,
 	return err;
 }
 
-static int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid)
+static int __mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				    u16 vid)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	char spvid_pl[MLXSW_REG_SPVID_LEN];
@@ -275,6 +380,53 @@ static int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid)
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvid), spvid_pl);
 }
 
+static int mlxsw_sp_port_allow_untagged_set(struct mlxsw_sp_port *mlxsw_sp_port,
+					    bool allow)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char spaft_pl[MLXSW_REG_SPAFT_LEN];
+
+	mlxsw_reg_spaft_pack(spaft_pl, mlxsw_sp_port->local_port, allow);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spaft), spaft_pl);
+}
+
+int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid)
+{
+	struct net_device *dev = mlxsw_sp_port->dev;
+	int err;
+
+	if (!vid) {
+		err = mlxsw_sp_port_allow_untagged_set(mlxsw_sp_port, false);
+		if (err) {
+			netdev_err(dev, "Failed to disallow untagged traffic\n");
+			return err;
+		}
+	} else {
+		err = __mlxsw_sp_port_pvid_set(mlxsw_sp_port, vid);
+		if (err) {
+			netdev_err(dev, "Failed to set PVID\n");
+			return err;
+		}
+
+		/* Only allow if not already allowed. */
+		if (!mlxsw_sp_port->pvid) {
+			err = mlxsw_sp_port_allow_untagged_set(mlxsw_sp_port,
+							       true);
+			if (err) {
+				netdev_err(dev, "Failed to allow untagged traffic\n");
+				goto err_port_allow_untagged_set;
+			}
+		}
+	}
+
+	mlxsw_sp_port->pvid = vid;
+	return 0;
+
+err_port_allow_untagged_set:
+	__mlxsw_sp_port_pvid_set(mlxsw_sp_port, mlxsw_sp_port->pvid);
+	return err;
+}
+
 static int mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid)
 {
 	char sfmr_pl[MLXSW_REG_SFMR_LEN];
@@ -305,7 +457,7 @@ static int mlxsw_sp_port_fid_map(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid)
 {
 	enum mlxsw_reg_svfa_mt mt;
 
-	if (mlxsw_sp_port->nr_vfids)
+	if (!list_empty(&mlxsw_sp_port->vports_list))
 		mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID;
 	else
 		mt = MLXSW_REG_SVFA_MT_VID_TO_FID;
@@ -317,7 +469,7 @@ static int mlxsw_sp_port_fid_unmap(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid)
 {
 	enum mlxsw_reg_svfa_mt mt;
 
-	if (!mlxsw_sp_port->nr_vfids)
+	if (list_empty(&mlxsw_sp_port->vports_list))
 		return 0;
 
 	mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID;
@@ -343,14 +495,35 @@ err_port_add_vid:
 	return err;
 }
 
+static int __mlxsw_sp_port_vlans_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				     u16 vid_begin, u16 vid_end, bool is_member,
+				     bool untagged)
+{
+	u16 vid, vid_e;
+	int err;
+
+	for (vid = vid_begin; vid <= vid_end;
+	     vid += MLXSW_REG_SPVM_REC_MAX_COUNT) {
+		vid_e = min((u16) (vid + MLXSW_REG_SPVM_REC_MAX_COUNT - 1),
+			    vid_end);
+
+		err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid_e,
+					     is_member, untagged);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port,
 				     u16 vid_begin, u16 vid_end,
 				     bool flag_untagged, bool flag_pvid)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	struct net_device *dev = mlxsw_sp_port->dev;
+	u16 vid, last_visited_vid, old_pvid;
 	enum mlxsw_reg_svfa_mt mt;
-	u16 vid, vid_e;
 	int err;
 
 	/* In case this is invoked with BRIDGE_FLAGS_SELF and port is
@@ -378,15 +551,18 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port,
 			if (err) {
 				netdev_err(dev, "Failed to create FID=VID=%d mapping\n",
 					   vid);
-				return err;
+				goto err_port_vid_to_fid_set;
 			}
 		}
+	}
 
-		/* Set FID mapping according to port's mode */
+	/* Set FID mapping according to port's mode */
+	for (vid = vid_begin; vid <= vid_end; vid++) {
 		err = mlxsw_sp_port_fid_map(mlxsw_sp_port, vid);
 		if (err) {
 			netdev_err(dev, "Failed to map FID=%d", vid);
-			return err;
+			last_visited_vid = --vid;
+			goto err_port_fid_map;
 		}
 	}
 
@@ -394,83 +570,138 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port,
 					true, false);
 	if (err) {
 		netdev_err(dev, "Failed to configure flooding\n");
-		return err;
+		goto err_port_flood_set;
 	}
 
-	for (vid = vid_begin; vid <= vid_end;
-	     vid += MLXSW_REG_SPVM_REC_MAX_COUNT) {
-		vid_e = min((u16) (vid + MLXSW_REG_SPVM_REC_MAX_COUNT - 1),
-			    vid_end);
+	err = __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end,
+					true, flag_untagged);
+	if (err) {
+		netdev_err(dev, "Unable to add VIDs %d-%d\n", vid_begin,
+			   vid_end);
+		goto err_port_vlans_set;
+	}
 
-		err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid_e, true,
-					     flag_untagged);
+	old_pvid = mlxsw_sp_port->pvid;
+	if (flag_pvid && old_pvid != vid_begin) {
+		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, vid_begin);
 		if (err) {
-			netdev_err(mlxsw_sp_port->dev, "Unable to add VIDs %d-%d\n",
-				   vid, vid_e);
-			return err;
+			netdev_err(dev, "Unable to add PVID %d\n", vid_begin);
+			goto err_port_pvid_set;
 		}
-	}
-
-	vid = vid_begin;
-	if (flag_pvid && mlxsw_sp_port->pvid != vid) {
-		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, vid);
+	} else if (!flag_pvid && old_pvid >= vid_begin && old_pvid <= vid_end) {
+		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, 0);
 		if (err) {
-			netdev_err(mlxsw_sp_port->dev, "Unable to add PVID %d\n",
-				   vid);
-			return err;
+			netdev_err(dev, "Unable to del PVID\n");
+			goto err_port_pvid_set;
 		}
-		mlxsw_sp_port->pvid = vid;
 	}
 
 	/* Changing activity bits only if HW operation succeded */
-	for (vid = vid_begin; vid <= vid_end; vid++)
+	for (vid = vid_begin; vid <= vid_end; vid++) {
 		set_bit(vid, mlxsw_sp_port->active_vlans);
+		if (flag_untagged)
+			set_bit(vid, mlxsw_sp_port->untagged_vlans);
+		else
+			clear_bit(vid, mlxsw_sp_port->untagged_vlans);
+	}
 
-	return mlxsw_sp_port_stp_state_set(mlxsw_sp_port,
-					   mlxsw_sp_port->stp_state);
+	/* STP state change must be done after we set active VLANs */
+	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_port,
+					  mlxsw_sp_port->stp_state);
+	if (err) {
+		netdev_err(dev, "Failed to set STP state\n");
+		goto err_port_stp_state_set;
+	}
+
+	return 0;
+
+err_port_vid_to_fid_set:
+	mlxsw_sp_fid_destroy(mlxsw_sp, vid);
+	return err;
+
+err_port_stp_state_set:
+	for (vid = vid_begin; vid <= vid_end; vid++)
+		clear_bit(vid, mlxsw_sp_port->active_vlans);
+	if (old_pvid != mlxsw_sp_port->pvid)
+		mlxsw_sp_port_pvid_set(mlxsw_sp_port, old_pvid);
+err_port_pvid_set:
+	__mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, false,
+				  false);
+err_port_vlans_set:
+	__mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, false,
+				  false);
+err_port_flood_set:
+	last_visited_vid = vid_end;
+err_port_fid_map:
+	for (vid = last_visited_vid; vid >= vid_begin; vid--)
+		mlxsw_sp_port_fid_unmap(mlxsw_sp_port, vid);
+	return err;
 }
 
 static int mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port,
 				   const struct switchdev_obj_port_vlan *vlan,
 				   struct switchdev_trans *trans)
 {
-	bool untagged_flag = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED;
-	bool pvid_flag = vlan->flags & BRIDGE_VLAN_INFO_PVID;
+	bool flag_untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED;
+	bool flag_pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID;
 
 	if (switchdev_trans_ph_prepare(trans))
 		return 0;
 
 	return __mlxsw_sp_port_vlans_add(mlxsw_sp_port,
 					 vlan->vid_begin, vlan->vid_end,
-					 untagged_flag, pvid_flag);
+					 flag_untagged, flag_pvid);
+}
+
+static enum mlxsw_reg_sfd_rec_policy mlxsw_sp_sfd_rec_policy(bool dynamic)
+{
+	return dynamic ? MLXSW_REG_SFD_REC_POLICY_DYNAMIC_ENTRY_INGRESS :
+			 MLXSW_REG_SFD_REC_POLICY_STATIC_ENTRY;
+}
+
+static enum mlxsw_reg_sfd_op mlxsw_sp_sfd_op(bool adding)
+{
+	return adding ? MLXSW_REG_SFD_OP_WRITE_EDIT :
+			MLXSW_REG_SFD_OP_WRITE_REMOVE;
 }
 
-static int mlxsw_sp_port_fdb_op(struct mlxsw_sp_port *mlxsw_sp_port,
-				const char *mac, u16 vid, bool adding,
-				bool dynamic)
+static int mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port,
+				   const char *mac, u16 fid, bool adding,
+				   bool dynamic)
 {
-	enum mlxsw_reg_sfd_rec_policy policy;
-	enum mlxsw_reg_sfd_op op;
 	char *sfd_pl;
 	int err;
 
-	if (!vid)
-		vid = mlxsw_sp_port->pvid;
+	sfd_pl = kmalloc(MLXSW_REG_SFD_LEN, GFP_KERNEL);
+	if (!sfd_pl)
+		return -ENOMEM;
+
+	mlxsw_reg_sfd_pack(sfd_pl, mlxsw_sp_sfd_op(adding), 0);
+	mlxsw_reg_sfd_uc_pack(sfd_pl, 0, mlxsw_sp_sfd_rec_policy(dynamic),
+			      mac, fid, MLXSW_REG_SFD_REC_ACTION_NOP,
+			      local_port);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfd), sfd_pl);
+	kfree(sfd_pl);
+
+	return err;
+}
+
+static int mlxsw_sp_port_fdb_uc_lag_op(struct mlxsw_sp *mlxsw_sp, u16 lag_id,
+				       const char *mac, u16 fid, u16 lag_vid,
+				       bool adding, bool dynamic)
+{
+	char *sfd_pl;
+	int err;
 
 	sfd_pl = kmalloc(MLXSW_REG_SFD_LEN, GFP_KERNEL);
 	if (!sfd_pl)
 		return -ENOMEM;
 
-	policy = dynamic ? MLXSW_REG_SFD_REC_POLICY_DYNAMIC_ENTRY_INGRESS :
-			   MLXSW_REG_SFD_REC_POLICY_STATIC_ENTRY;
-	op = adding ? MLXSW_REG_SFD_OP_WRITE_EDIT :
-		      MLXSW_REG_SFD_OP_WRITE_REMOVE;
-	mlxsw_reg_sfd_pack(sfd_pl, op, 0);
-	mlxsw_reg_sfd_uc_pack(sfd_pl, 0, policy,
-			      mac, vid, MLXSW_REG_SFD_REC_ACTION_NOP,
-			      mlxsw_sp_port->local_port);
-	err = mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, MLXSW_REG(sfd),
-			      sfd_pl);
+	mlxsw_reg_sfd_pack(sfd_pl, mlxsw_sp_sfd_op(adding), 0);
+	mlxsw_reg_sfd_uc_lag_pack(sfd_pl, 0, mlxsw_sp_sfd_rec_policy(dynamic),
+				  mac, fid, MLXSW_REG_SFD_REC_ACTION_NOP,
+				  lag_vid, lag_id);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfd), sfd_pl);
 	kfree(sfd_pl);
 
 	return err;
@@ -481,11 +712,162 @@ mlxsw_sp_port_fdb_static_add(struct mlxsw_sp_port *mlxsw_sp_port,
 			     const struct switchdev_obj_port_fdb *fdb,
 			     struct switchdev_trans *trans)
 {
+	u16 fid = mlxsw_sp_port_vid_to_fid_get(mlxsw_sp_port, fdb->vid);
+	u16 lag_vid = 0;
+
 	if (switchdev_trans_ph_prepare(trans))
 		return 0;
 
-	return mlxsw_sp_port_fdb_op(mlxsw_sp_port, fdb->addr, fdb->vid,
-				    true, false);
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		lag_vid = mlxsw_sp_vport_vid_get(mlxsw_sp_port);
+	}
+
+	if (!mlxsw_sp_port->lagged)
+		return mlxsw_sp_port_fdb_uc_op(mlxsw_sp_port->mlxsw_sp,
+					       mlxsw_sp_port->local_port,
+					       fdb->addr, fid, true, false);
+	else
+		return mlxsw_sp_port_fdb_uc_lag_op(mlxsw_sp_port->mlxsw_sp,
+						   mlxsw_sp_port->lag_id,
+						   fdb->addr, fid, lag_vid,
+						   true, false);
+}
+
+static int mlxsw_sp_port_mdb_op(struct mlxsw_sp *mlxsw_sp, const char *addr,
+				u16 fid, u16 mid, bool adding)
+{
+	char *sfd_pl;
+	int err;
+
+	sfd_pl = kmalloc(MLXSW_REG_SFD_LEN, GFP_KERNEL);
+	if (!sfd_pl)
+		return -ENOMEM;
+
+	mlxsw_reg_sfd_pack(sfd_pl, mlxsw_sp_sfd_op(adding), 0);
+	mlxsw_reg_sfd_mc_pack(sfd_pl, 0, addr, fid,
+			      MLXSW_REG_SFD_REC_ACTION_NOP, mid);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfd), sfd_pl);
+	kfree(sfd_pl);
+	return err;
+}
+
+static int mlxsw_sp_port_smid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mid,
+				  bool add, bool clear_all_ports)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char *smid_pl;
+	int err, i;
+
+	smid_pl = kmalloc(MLXSW_REG_SMID_LEN, GFP_KERNEL);
+	if (!smid_pl)
+		return -ENOMEM;
+
+	mlxsw_reg_smid_pack(smid_pl, mid, mlxsw_sp_port->local_port, add);
+	if (clear_all_ports) {
+		for (i = 1; i < MLXSW_PORT_MAX_PORTS; i++)
+			if (mlxsw_sp->ports[i])
+				mlxsw_reg_smid_port_mask_set(smid_pl, i, 1);
+	}
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(smid), smid_pl);
+	kfree(smid_pl);
+	return err;
+}
+
+static struct mlxsw_sp_mid *__mlxsw_sp_mc_get(struct mlxsw_sp *mlxsw_sp,
+					      const unsigned char *addr,
+					      u16 vid)
+{
+	struct mlxsw_sp_mid *mid;
+
+	list_for_each_entry(mid, &mlxsw_sp->br_mids.list, list) {
+		if (ether_addr_equal(mid->addr, addr) && mid->vid == vid)
+			return mid;
+	}
+	return NULL;
+}
+
+static struct mlxsw_sp_mid *__mlxsw_sp_mc_alloc(struct mlxsw_sp *mlxsw_sp,
+						const unsigned char *addr,
+						u16 vid)
+{
+	struct mlxsw_sp_mid *mid;
+	u16 mid_idx;
+
+	mid_idx = find_first_zero_bit(mlxsw_sp->br_mids.mapped,
+				      MLXSW_SP_MID_MAX);
+	if (mid_idx == MLXSW_SP_MID_MAX)
+		return NULL;
+
+	mid = kzalloc(sizeof(*mid), GFP_KERNEL);
+	if (!mid)
+		return NULL;
+
+	set_bit(mid_idx, mlxsw_sp->br_mids.mapped);
+	ether_addr_copy(mid->addr, addr);
+	mid->vid = vid;
+	mid->mid = mid_idx;
+	mid->ref_count = 0;
+	list_add_tail(&mid->list, &mlxsw_sp->br_mids.list);
+
+	return mid;
+}
+
+static int __mlxsw_sp_mc_dec_ref(struct mlxsw_sp *mlxsw_sp,
+				 struct mlxsw_sp_mid *mid)
+{
+	if (--mid->ref_count == 0) {
+		list_del(&mid->list);
+		clear_bit(mid->mid, mlxsw_sp->br_mids.mapped);
+		kfree(mid);
+		return 1;
+	}
+	return 0;
+}
+
+static int mlxsw_sp_port_mdb_add(struct mlxsw_sp_port *mlxsw_sp_port,
+				 const struct switchdev_obj_port_mdb *mdb,
+				 struct switchdev_trans *trans)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	struct net_device *dev = mlxsw_sp_port->dev;
+	struct mlxsw_sp_mid *mid;
+	u16 fid = mlxsw_sp_port_vid_to_fid_get(mlxsw_sp_port, mdb->vid);
+	int err = 0;
+
+	if (switchdev_trans_ph_prepare(trans))
+		return 0;
+
+	mid = __mlxsw_sp_mc_get(mlxsw_sp, mdb->addr, mdb->vid);
+	if (!mid) {
+		mid = __mlxsw_sp_mc_alloc(mlxsw_sp, mdb->addr, mdb->vid);
+		if (!mid) {
+			netdev_err(dev, "Unable to allocate MC group\n");
+			return -ENOMEM;
+		}
+	}
+	mid->ref_count++;
+
+	err = mlxsw_sp_port_smid_set(mlxsw_sp_port, mid->mid, true,
+				     mid->ref_count == 1);
+	if (err) {
+		netdev_err(dev, "Unable to set SMID\n");
+		goto err_out;
+	}
+
+	if (mid->ref_count == 1) {
+		err = mlxsw_sp_port_mdb_op(mlxsw_sp, mdb->addr, fid, mid->mid,
+					   true);
+		if (err) {
+			netdev_err(dev, "Unable to set MC SFD\n");
+			goto err_out;
+		}
+	}
+
+	return 0;
+
+err_out:
+	__mlxsw_sp_mc_dec_ref(mlxsw_sp, mid);
+	return err;
 }
 
 static int mlxsw_sp_port_obj_add(struct net_device *dev,
@@ -495,8 +877,15 @@ static int mlxsw_sp_port_obj_add(struct net_device *dev,
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
 	int err = 0;
 
+	mlxsw_sp_port = mlxsw_sp_port_orig_get(obj->orig_dev, mlxsw_sp_port);
+	if (!mlxsw_sp_port)
+		return -EINVAL;
+
 	switch (obj->id) {
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		if (mlxsw_sp_port_is_vport(mlxsw_sp_port))
+			return 0;
+
 		err = mlxsw_sp_port_vlans_add(mlxsw_sp_port,
 					      SWITCHDEV_OBJ_PORT_VLAN(obj),
 					      trans);
@@ -506,6 +895,11 @@ static int mlxsw_sp_port_obj_add(struct net_device *dev,
 						   SWITCHDEV_OBJ_PORT_FDB(obj),
 						   trans);
 		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+		err = mlxsw_sp_port_mdb_add(mlxsw_sp_port,
+					    SWITCHDEV_OBJ_PORT_MDB(obj),
+					    trans);
+		break;
 	default:
 		err = -EOPNOTSUPP;
 		break;
@@ -533,7 +927,7 @@ static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port,
 				     u16 vid_begin, u16 vid_end, bool init)
 {
 	struct net_device *dev = mlxsw_sp_port->dev;
-	u16 vid, vid_e;
+	u16 vid, pvid;
 	int err;
 
 	/* In case this is invoked with BRIDGE_FLAGS_SELF and port is
@@ -543,35 +937,26 @@ static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port,
 	if (!init && !mlxsw_sp_port->bridged)
 		return mlxsw_sp_port_kill_vids(dev, vid_begin, vid_end);
 
-	for (vid = vid_begin; vid <= vid_end;
-	     vid += MLXSW_REG_SPVM_REC_MAX_COUNT) {
-		vid_e = min((u16) (vid + MLXSW_REG_SPVM_REC_MAX_COUNT - 1),
-			    vid_end);
-		err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid_e, false,
-					     false);
-		if (err) {
-			netdev_err(mlxsw_sp_port->dev, "Unable to del VIDs %d-%d\n",
-				   vid, vid_e);
-			return err;
-		}
+	err = __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end,
+					false, false);
+	if (err) {
+		netdev_err(dev, "Unable to del VIDs %d-%d\n", vid_begin,
+			   vid_end);
+		return err;
 	}
 
-	if ((mlxsw_sp_port->pvid >= vid_begin) &&
-	    (mlxsw_sp_port->pvid <= vid_end)) {
-		/* Default VLAN is always 1 */
-		mlxsw_sp_port->pvid = 1;
-		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port,
-					     mlxsw_sp_port->pvid);
+	if (init)
+		goto out;
+
+	pvid = mlxsw_sp_port->pvid;
+	if (pvid >= vid_begin && pvid <= vid_end) {
+		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, 0);
 		if (err) {
-			netdev_err(mlxsw_sp_port->dev, "Unable to del PVID %d\n",
-				   vid);
+			netdev_err(dev, "Unable to del PVID %d\n", pvid);
 			return err;
 		}
 	}
 
-	if (init)
-		goto out;
-
 	err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end,
 					false, false);
 	if (err) {
@@ -603,12 +988,66 @@ static int mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port,
 					 vlan->vid_begin, vlan->vid_end, false);
 }
 
+void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	u16 vid;
+
+	for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID)
+		__mlxsw_sp_port_vlans_del(mlxsw_sp_port, vid, vid, false);
+}
+
 static int
 mlxsw_sp_port_fdb_static_del(struct mlxsw_sp_port *mlxsw_sp_port,
 			     const struct switchdev_obj_port_fdb *fdb)
 {
-	return mlxsw_sp_port_fdb_op(mlxsw_sp_port, fdb->addr, fdb->vid,
-				    false, false);
+	u16 fid = mlxsw_sp_port_vid_to_fid_get(mlxsw_sp_port, fdb->vid);
+	u16 lag_vid = 0;
+
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		lag_vid = mlxsw_sp_vport_vid_get(mlxsw_sp_port);
+	}
+
+	if (!mlxsw_sp_port->lagged)
+		return mlxsw_sp_port_fdb_uc_op(mlxsw_sp_port->mlxsw_sp,
+					       mlxsw_sp_port->local_port,
+					       fdb->addr, fid,
+					       false, false);
+	else
+		return mlxsw_sp_port_fdb_uc_lag_op(mlxsw_sp_port->mlxsw_sp,
+						   mlxsw_sp_port->lag_id,
+						   fdb->addr, fid, lag_vid,
+						   false, false);
+}
+
+static int mlxsw_sp_port_mdb_del(struct mlxsw_sp_port *mlxsw_sp_port,
+				 const struct switchdev_obj_port_mdb *mdb)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	struct net_device *dev = mlxsw_sp_port->dev;
+	struct mlxsw_sp_mid *mid;
+	u16 fid = mlxsw_sp_port_vid_to_fid_get(mlxsw_sp_port, mdb->vid);
+	u16 mid_idx;
+	int err = 0;
+
+	mid = __mlxsw_sp_mc_get(mlxsw_sp, mdb->addr, mdb->vid);
+	if (!mid) {
+		netdev_err(dev, "Unable to remove port from MC DB\n");
+		return -EINVAL;
+	}
+
+	err = mlxsw_sp_port_smid_set(mlxsw_sp_port, mid->mid, false, false);
+	if (err)
+		netdev_err(dev, "Unable to remove port from SMID\n");
+
+	mid_idx = mid->mid;
+	if (__mlxsw_sp_mc_dec_ref(mlxsw_sp, mid)) {
+		err = mlxsw_sp_port_mdb_op(mlxsw_sp, mdb->addr, fid, mid_idx,
+					   false);
+		if (err)
+			netdev_err(dev, "Unable to remove MC SFD\n");
+	}
+
+	return err;
 }
 
 static int mlxsw_sp_port_obj_del(struct net_device *dev,
@@ -617,8 +1056,15 @@ static int mlxsw_sp_port_obj_del(struct net_device *dev,
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
 	int err = 0;
 
+	mlxsw_sp_port = mlxsw_sp_port_orig_get(obj->orig_dev, mlxsw_sp_port);
+	if (!mlxsw_sp_port)
+		return -EINVAL;
+
 	switch (obj->id) {
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		if (mlxsw_sp_port_is_vport(mlxsw_sp_port))
+			return 0;
+
 		err = mlxsw_sp_port_vlans_del(mlxsw_sp_port,
 					      SWITCHDEV_OBJ_PORT_VLAN(obj));
 		break;
@@ -626,6 +1072,10 @@ static int mlxsw_sp_port_obj_del(struct net_device *dev,
 		err = mlxsw_sp_port_fdb_static_del(mlxsw_sp_port,
 						   SWITCHDEV_OBJ_PORT_FDB(obj));
 		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+		err = mlxsw_sp_port_mdb_del(mlxsw_sp_port,
+					    SWITCHDEV_OBJ_PORT_MDB(obj));
+		break;
 	default:
 		err = -EOPNOTSUPP;
 		break;
@@ -634,14 +1084,33 @@ static int mlxsw_sp_port_obj_del(struct net_device *dev,
 	return err;
 }
 
+static struct mlxsw_sp_port *mlxsw_sp_lag_rep_port(struct mlxsw_sp *mlxsw_sp,
+						   u16 lag_id)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port;
+	int i;
+
+	for (i = 0; i < MLXSW_SP_PORT_PER_LAG_MAX; i++) {
+		mlxsw_sp_port = mlxsw_sp_port_lagged_get(mlxsw_sp, lag_id, i);
+		if (mlxsw_sp_port)
+			return mlxsw_sp_port;
+	}
+	return NULL;
+}
+
 static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port,
 				  struct switchdev_obj_port_fdb *fdb,
-				  switchdev_obj_dump_cb_t *cb)
+				  switchdev_obj_dump_cb_t *cb,
+				  struct net_device *orig_dev)
 {
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	struct mlxsw_sp_port *tmp;
+	u16 vport_fid = 0;
 	char *sfd_pl;
 	char mac[ETH_ALEN];
-	u16 vid;
+	u16 fid;
 	u8 local_port;
+	u16 lag_id;
 	u8 num_rec;
 	int stored_err = 0;
 	int i;
@@ -651,11 +1120,17 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port,
 	if (!sfd_pl)
 		return -ENOMEM;
 
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		u16 tmp;
+
+		tmp = mlxsw_sp_vport_vfid_get(mlxsw_sp_port);
+		vport_fid = mlxsw_sp_vfid_to_fid(tmp);
+	}
+
 	mlxsw_reg_sfd_pack(sfd_pl, MLXSW_REG_SFD_OP_QUERY_DUMP, 0);
 	do {
 		mlxsw_reg_sfd_num_rec_set(sfd_pl, MLXSW_REG_SFD_REC_MAX_COUNT);
-		err = mlxsw_reg_query(mlxsw_sp_port->mlxsw_sp->core,
-				      MLXSW_REG(sfd), sfd_pl);
+		err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(sfd), sfd_pl);
 		if (err)
 			goto out;
 
@@ -670,16 +1145,49 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port,
 		for (i = 0; i < num_rec; i++) {
 			switch (mlxsw_reg_sfd_rec_type_get(sfd_pl, i)) {
 			case MLXSW_REG_SFD_REC_TYPE_UNICAST:
-				mlxsw_reg_sfd_uc_unpack(sfd_pl, i, mac, &vid,
+				mlxsw_reg_sfd_uc_unpack(sfd_pl, i, mac, &fid,
 							&local_port);
 				if (local_port == mlxsw_sp_port->local_port) {
+					if (vport_fid && vport_fid == fid)
+						fdb->vid = 0;
+					else if (!vport_fid &&
+						 !mlxsw_sp_fid_is_vfid(fid))
+						fdb->vid = fid;
+					else
+						continue;
 					ether_addr_copy(fdb->addr, mac);
 					fdb->ndm_state = NUD_REACHABLE;
-					fdb->vid = vid;
 					err = cb(&fdb->obj);
 					if (err)
 						stored_err = err;
 				}
+				break;
+			case MLXSW_REG_SFD_REC_TYPE_UNICAST_LAG:
+				mlxsw_reg_sfd_uc_lag_unpack(sfd_pl, i,
+							    mac, &fid, &lag_id);
+				tmp = mlxsw_sp_lag_rep_port(mlxsw_sp, lag_id);
+				if (tmp && tmp->local_port ==
+				    mlxsw_sp_port->local_port) {
+					/* LAG records can only point to LAG
+					 * devices or VLAN devices on top.
+					 */
+					if (!netif_is_lag_master(orig_dev) &&
+					    !is_vlan_dev(orig_dev))
+						continue;
+					if (vport_fid && vport_fid == fid)
+						fdb->vid = 0;
+					else if (!vport_fid &&
+						 !mlxsw_sp_fid_is_vfid(fid))
+						fdb->vid = fid;
+					else
+						continue;
+					ether_addr_copy(fdb->addr, mac);
+					fdb->ndm_state = NUD_REACHABLE;
+					err = cb(&fdb->obj);
+					if (err)
+						stored_err = err;
+				}
+				break;
 			}
 		}
 	} while (num_rec == MLXSW_REG_SFD_REC_MAX_COUNT);
@@ -696,10 +1204,19 @@ static int mlxsw_sp_port_vlan_dump(struct mlxsw_sp_port *mlxsw_sp_port,
 	u16 vid;
 	int err = 0;
 
+	if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+		vlan->flags = 0;
+		vlan->vid_begin = mlxsw_sp_vport_vid_get(mlxsw_sp_port);
+		vlan->vid_end = mlxsw_sp_vport_vid_get(mlxsw_sp_port);
+		return cb(&vlan->obj);
+	}
+
 	for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) {
 		vlan->flags = 0;
 		if (vid == mlxsw_sp_port->pvid)
 			vlan->flags |= BRIDGE_VLAN_INFO_PVID;
+		if (test_bit(vid, mlxsw_sp_port->untagged_vlans))
+			vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED;
 		vlan->vid_begin = vid;
 		vlan->vid_end = vid;
 		err = cb(&vlan->obj);
@@ -716,6 +1233,10 @@ static int mlxsw_sp_port_obj_dump(struct net_device *dev,
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
 	int err = 0;
 
+	mlxsw_sp_port = mlxsw_sp_port_orig_get(obj->orig_dev, mlxsw_sp_port);
+	if (!mlxsw_sp_port)
+		return -EINVAL;
+
 	switch (obj->id) {
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
 		err = mlxsw_sp_port_vlan_dump(mlxsw_sp_port,
@@ -723,7 +1244,8 @@ static int mlxsw_sp_port_obj_dump(struct net_device *dev,
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_FDB:
 		err = mlxsw_sp_port_fdb_dump(mlxsw_sp_port,
-					     SWITCHDEV_OBJ_PORT_FDB(obj), cb);
+					     SWITCHDEV_OBJ_PORT_FDB(obj), cb,
+					     obj->orig_dev);
 		break;
 	default:
 		err = -EOPNOTSUPP;
@@ -741,6 +1263,21 @@ static const struct switchdev_ops mlxsw_sp_port_switchdev_ops = {
 	.switchdev_port_obj_dump	= mlxsw_sp_port_obj_dump,
 };
 
+static void mlxsw_sp_fdb_call_notifiers(bool learning_sync, bool adding,
+					char *mac, u16 vid,
+					struct net_device *dev)
+{
+	struct switchdev_notifier_fdb_info info;
+	unsigned long notifier_type;
+
+	if (learning_sync) {
+		info.addr = mac;
+		info.vid = vid;
+		notifier_type = adding ? SWITCHDEV_FDB_ADD : SWITCHDEV_FDB_DEL;
+		call_switchdev_notifiers(notifier_type, dev, &info.info);
+	}
+}
+
 static void mlxsw_sp_fdb_notify_mac_process(struct mlxsw_sp *mlxsw_sp,
 					    char *sfn_pl, int rec_index,
 					    bool adding)
@@ -748,34 +1285,119 @@ static void mlxsw_sp_fdb_notify_mac_process(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_port *mlxsw_sp_port;
 	char mac[ETH_ALEN];
 	u8 local_port;
-	u16 vid;
+	u16 vid, fid;
+	bool do_notification = true;
 	int err;
 
-	mlxsw_reg_sfn_mac_unpack(sfn_pl, rec_index, mac, &vid, &local_port);
+	mlxsw_reg_sfn_mac_unpack(sfn_pl, rec_index, mac, &fid, &local_port);
 	mlxsw_sp_port = mlxsw_sp->ports[local_port];
 	if (!mlxsw_sp_port) {
 		dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Incorrect local port in FDB notification\n");
-		return;
+		goto just_remove;
+	}
+
+	if (mlxsw_sp_fid_is_vfid(fid)) {
+		u16 vfid = mlxsw_sp_fid_to_vfid(fid);
+		struct mlxsw_sp_port *mlxsw_sp_vport;
+
+		mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_vfid(mlxsw_sp_port,
+								  vfid);
+		if (!mlxsw_sp_vport) {
+			netdev_err(mlxsw_sp_port->dev, "Failed to find a matching vPort following FDB notification\n");
+			goto just_remove;
+		}
+		vid = 0;
+		/* Override the physical port with the vPort. */
+		mlxsw_sp_port = mlxsw_sp_vport;
+	} else {
+		vid = fid;
 	}
 
-	err = mlxsw_sp_port_fdb_op(mlxsw_sp_port, mac, vid,
-				   adding && mlxsw_sp_port->learning, true);
+	adding = adding && mlxsw_sp_port->learning;
+
+do_fdb_op:
+	err = mlxsw_sp_port_fdb_uc_op(mlxsw_sp, local_port, mac, fid,
+				      adding, true);
 	if (err) {
 		if (net_ratelimit())
 			netdev_err(mlxsw_sp_port->dev, "Failed to set FDB entry\n");
 		return;
 	}
 
-	if (mlxsw_sp_port->learning && mlxsw_sp_port->learning_sync) {
-		struct switchdev_notifier_fdb_info info;
-		unsigned long notifier_type;
+	if (!do_notification)
+		return;
+	mlxsw_sp_fdb_call_notifiers(mlxsw_sp_port->learning_sync,
+				    adding, mac, vid, mlxsw_sp_port->dev);
+	return;
+
+just_remove:
+	adding = false;
+	do_notification = false;
+	goto do_fdb_op;
+}
 
-		info.addr = mac;
-		info.vid = vid;
-		notifier_type = adding ? SWITCHDEV_FDB_ADD : SWITCHDEV_FDB_DEL;
-		call_switchdev_notifiers(notifier_type, mlxsw_sp_port->dev,
-					 &info.info);
+static void mlxsw_sp_fdb_notify_mac_lag_process(struct mlxsw_sp *mlxsw_sp,
+						char *sfn_pl, int rec_index,
+						bool adding)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port;
+	struct net_device *dev;
+	char mac[ETH_ALEN];
+	u16 lag_vid = 0;
+	u16 lag_id;
+	u16 vid, fid;
+	bool do_notification = true;
+	int err;
+
+	mlxsw_reg_sfn_mac_lag_unpack(sfn_pl, rec_index, mac, &fid, &lag_id);
+	mlxsw_sp_port = mlxsw_sp_lag_rep_port(mlxsw_sp, lag_id);
+	if (!mlxsw_sp_port) {
+		dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Cannot find port representor for LAG\n");
+		goto just_remove;
+	}
+
+	if (mlxsw_sp_fid_is_vfid(fid)) {
+		u16 vfid = mlxsw_sp_fid_to_vfid(fid);
+		struct mlxsw_sp_port *mlxsw_sp_vport;
+
+		mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_vfid(mlxsw_sp_port,
+								  vfid);
+		if (!mlxsw_sp_vport) {
+			netdev_err(mlxsw_sp_port->dev, "Failed to find a matching vPort following FDB notification\n");
+			goto just_remove;
+		}
+
+		lag_vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport);
+		dev = mlxsw_sp_vport->dev;
+		vid = 0;
+		/* Override the physical port with the vPort. */
+		mlxsw_sp_port = mlxsw_sp_vport;
+	} else {
+		dev = mlxsw_sp_lag_get(mlxsw_sp, lag_id)->dev;
+		vid = fid;
 	}
+
+	adding = adding && mlxsw_sp_port->learning;
+
+do_fdb_op:
+	err = mlxsw_sp_port_fdb_uc_lag_op(mlxsw_sp, lag_id, mac, fid, lag_vid,
+					  adding, true);
+	if (err) {
+		if (net_ratelimit())
+			netdev_err(mlxsw_sp_port->dev, "Failed to set FDB entry\n");
+		return;
+	}
+
+	if (!do_notification)
+		return;
+	mlxsw_sp_fdb_call_notifiers(mlxsw_sp_port->learning_sync, adding, mac,
+				    vid, dev);
+	return;
+
+just_remove:
+	adding = false;
+	do_notification = false;
+	goto do_fdb_op;
 }
 
 static void mlxsw_sp_fdb_notify_rec_process(struct mlxsw_sp *mlxsw_sp,
@@ -790,6 +1412,14 @@ static void mlxsw_sp_fdb_notify_rec_process(struct mlxsw_sp *mlxsw_sp,
 		mlxsw_sp_fdb_notify_mac_process(mlxsw_sp, sfn_pl,
 						rec_index, false);
 		break;
+	case MLXSW_REG_SFN_REC_TYPE_LEARNED_MAC_LAG:
+		mlxsw_sp_fdb_notify_mac_lag_process(mlxsw_sp, sfn_pl,
+						    rec_index, true);
+		break;
+	case MLXSW_REG_SFN_REC_TYPE_AGED_OUT_MAC_LAG:
+		mlxsw_sp_fdb_notify_mac_lag_process(mlxsw_sp, sfn_pl,
+						    rec_index, false);
+		break;
 	}
 }
 
@@ -880,7 +1510,8 @@ int mlxsw_sp_port_vlan_init(struct mlxsw_sp_port *mlxsw_sp_port)
 	 * with VID 1.
 	 */
 	mlxsw_sp_port->pvid = 1;
-	err = __mlxsw_sp_port_vlans_del(mlxsw_sp_port, 0, VLAN_N_VID, true);
+	err = __mlxsw_sp_port_vlans_del(mlxsw_sp_port, 0, VLAN_N_VID - 1,
+					true);
 	if (err) {
 		netdev_err(dev, "Unable to init VLANs\n");
 		return err;