mlxsw: spectrum: Add initial support for Spectrum ASIC

Add support for new generation Mellanox Spectrum ASIC, 10/25/40/50 and
100Gb/s Ethernet Switch.

The initial driver implements bridge forwarding offload including
bridge internal VLAN support, FDB static entries, FDB learning and
HW ageing including their setup.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Elad Raz <eladr@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

10 files changed, 3374 insertions, 0 deletions

diff --git a/drivers/net/ethernet/mellanox/mlxsw/Kconfig b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
index 2941d9c5ae48..e36e12219c9b 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
@@ -30,3 +30,14 @@ config MLXSW_SWITCHX2
 
 	  To compile this driver as a module, choose M here: the
 	  module will be called mlxsw_switchx2.
+
+config MLXSW_SPECTRUM
+	tristate "Mellanox Technologies Spectrum support"
+	depends on MLXSW_CORE && NET_SWITCHDEV
+	default m
+	---help---
+	  This driver supports Mellanox Technologies Spectrum Ethernet
+	  Switch ASICs.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called mlxsw_spectrum.
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile
index 0a05f65ee814..af015818fd19 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Makefile
+++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile
@@ -4,3 +4,6 @@ obj-$(CONFIG_MLXSW_PCI)		+= mlxsw_pci.o
 mlxsw_pci-objs			:= pci.o
 obj-$(CONFIG_MLXSW_SWITCHX2)	+= mlxsw_switchx2.o
 mlxsw_switchx2-objs		:= switchx2.o
+obj-$(CONFIG_MLXSW_SPECTRUM)	+= mlxsw_spectrum.o
+mlxsw_spectrum-objs		:= spectrum.o spectrum_buffers.o \
+				   spectrum_switchdev.o
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h
index e92ab272d9fb..807827350a89 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.h
@@ -54,6 +54,7 @@
 	MODULE_ALIAS(MLXSW_MODULE_ALIAS_PREFIX kind)
 
 #define MLXSW_DEVICE_KIND_SWITCHX2 "switchx2"
+#define MLXSW_DEVICE_KIND_SPECTRUM "spectrum"
 
 struct mlxsw_core;
 struct mlxsw_driver;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 0fa44c8e7102..879e000684c3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -57,6 +57,7 @@ static const char mlxsw_pci_driver_name[] = "mlxsw_pci";
 
 static const struct pci_device_id mlxsw_pci_id_table[] = {
 	{PCI_VDEVICE(MELLANOX, PCI_DEVICE_ID_MELLANOX_SWITCHX2), 0},
+	{PCI_VDEVICE(MELLANOX, PCI_DEVICE_ID_MELLANOX_SPECTRUM), 0},
 	{0, }
 };
 
@@ -67,6 +68,8 @@ static const char *mlxsw_pci_device_kind_get(const struct pci_device_id *id)
 	switch (id->device) {
 	case PCI_DEVICE_ID_MELLANOX_SWITCHX2:
 		return MLXSW_DEVICE_KIND_SWITCHX2;
+	case PCI_DEVICE_ID_MELLANOX_SPECTRUM:
+		return MLXSW_DEVICE_KIND_SPECTRUM;
 	default:
 		BUG();
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.h b/drivers/net/ethernet/mellanox/mlxsw/pci.h
index 5b3453b6cf5d..142f33d978c5 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.h
@@ -40,6 +40,7 @@
 #include "item.h"
 
 #define PCI_DEVICE_ID_MELLANOX_SWITCHX2	0xc738
+#define PCI_DEVICE_ID_MELLANOX_SPECTRUM	0xcb84
 #define MLXSW_PCI_BAR0_SIZE		(1024 * 1024) /* 1MB */
 #define MLXSW_PCI_PAGE_SIZE		4096
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
new file mode 100644
index 000000000000..6e9906d8d149
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -0,0 +1,1948 @@
+/*
+ * drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+ * Copyright (c) 2015 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2015 Jiri Pirko <jiri@mellanox.com>
+ * Copyright (c) 2015 Ido Schimmel <idosch@mellanox.com>
+ * Copyright (c) 2015 Elad Raz <eladr@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/module.h>
+#include <linux/types.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/slab.h>
+#include <linux/device.h>
+#include <linux/skbuff.h>
+#include <linux/if_vlan.h>
+#include <linux/if_bridge.h>
+#include <linux/workqueue.h>
+#include <linux/jiffies.h>
+#include <linux/bitops.h>
+#include <net/switchdev.h>
+#include <generated/utsrelease.h>
+
+#include "spectrum.h"
+#include "core.h"
+#include "reg.h"
+#include "port.h"
+#include "trap.h"
+#include "txheader.h"
+
+static const char mlxsw_sp_driver_name[] = "mlxsw_spectrum";
+static const char mlxsw_sp_driver_version[] = "1.0";
+
+/* tx_hdr_version
+ * Tx header version.
+ * Must be set to 1.
+ */
+MLXSW_ITEM32(tx, hdr, version, 0x00, 28, 4);
+
+/* tx_hdr_ctl
+ * Packet control type.
+ * 0 - Ethernet control (e.g. EMADs, LACP)
+ * 1 - Ethernet data
+ */
+MLXSW_ITEM32(tx, hdr, ctl, 0x00, 26, 2);
+
+/* tx_hdr_proto
+ * Packet protocol type. Must be set to 1 (Ethernet).
+ */
+MLXSW_ITEM32(tx, hdr, proto, 0x00, 21, 3);
+
+/* tx_hdr_rx_is_router
+ * Packet is sent from the router. Valid for data packets only.
+ */
+MLXSW_ITEM32(tx, hdr, rx_is_router, 0x00, 19, 1);
+
+/* tx_hdr_fid_valid
+ * Indicates if the 'fid' field is valid and should be used for
+ * forwarding lookup. Valid for data packets only.
+ */
+MLXSW_ITEM32(tx, hdr, fid_valid, 0x00, 16, 1);
+
+/* tx_hdr_swid
+ * Switch partition ID. Must be set to 0.
+ */
+MLXSW_ITEM32(tx, hdr, swid, 0x00, 12, 3);
+
+/* tx_hdr_control_tclass
+ * Indicates if the packet should use the control TClass and not one
+ * of the data TClasses.
+ */
+MLXSW_ITEM32(tx, hdr, control_tclass, 0x00, 6, 1);
+
+/* tx_hdr_etclass
+ * Egress TClass to be used on the egress device on the egress port.
+ */
+MLXSW_ITEM32(tx, hdr, etclass, 0x00, 0, 4);
+
+/* tx_hdr_port_mid
+ * Destination local port for unicast packets.
+ * Destination multicast ID for multicast packets.
+ *
+ * Control packets are directed to a specific egress port, while data
+ * packets are transmitted through the CPU port (0) into the switch partition,
+ * where forwarding rules are applied.
+ */
+MLXSW_ITEM32(tx, hdr, port_mid, 0x04, 16, 16);
+
+/* tx_hdr_fid
+ * Forwarding ID used for L2 forwarding lookup. Valid only if 'fid_valid' is
+ * set, otherwise calculated based on the packet's VID using VID to FID mapping.
+ * Valid for data packets only.
+ */
+MLXSW_ITEM32(tx, hdr, fid, 0x08, 0, 16);
+
+/* tx_hdr_type
+ * 0 - Data packets
+ * 6 - Control packets
+ */
+MLXSW_ITEM32(tx, hdr, type, 0x0C, 0, 4);
+
+static void mlxsw_sp_txhdr_construct(struct sk_buff *skb,
+				     const struct mlxsw_tx_info *tx_info)
+{
+	char *txhdr = skb_push(skb, MLXSW_TXHDR_LEN);
+
+	memset(txhdr, 0, MLXSW_TXHDR_LEN);
+
+	mlxsw_tx_hdr_version_set(txhdr, MLXSW_TXHDR_VERSION_1);
+	mlxsw_tx_hdr_ctl_set(txhdr, MLXSW_TXHDR_ETH_CTL);
+	mlxsw_tx_hdr_proto_set(txhdr, MLXSW_TXHDR_PROTO_ETH);
+	mlxsw_tx_hdr_swid_set(txhdr, 0);
+	mlxsw_tx_hdr_control_tclass_set(txhdr, 1);
+	mlxsw_tx_hdr_port_mid_set(txhdr, tx_info->local_port);
+	mlxsw_tx_hdr_type_set(txhdr, MLXSW_TXHDR_TYPE_CONTROL);
+}
+
+static int mlxsw_sp_base_mac_get(struct mlxsw_sp *mlxsw_sp)
+{
+	char spad_pl[MLXSW_REG_SPAD_LEN];
+	int err;
+
+	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(spad), spad_pl);
+	if (err)
+		return err;
+	mlxsw_reg_spad_base_mac_memcpy_from(spad_pl, mlxsw_sp->base_mac);
+	return 0;
+}
+
+static int mlxsw_sp_port_admin_status_set(struct mlxsw_sp_port *mlxsw_sp_port,
+					  bool is_up)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char paos_pl[MLXSW_REG_PAOS_LEN];
+
+	mlxsw_reg_paos_pack(paos_pl, mlxsw_sp_port->local_port,
+			    is_up ? MLXSW_PORT_ADMIN_STATUS_UP :
+			    MLXSW_PORT_ADMIN_STATUS_DOWN);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(paos), paos_pl);
+}
+
+static int mlxsw_sp_port_oper_status_get(struct mlxsw_sp_port *mlxsw_sp_port,
+					 bool *p_is_up)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char paos_pl[MLXSW_REG_PAOS_LEN];
+	u8 oper_status;
+	int err;
+
+	mlxsw_reg_paos_pack(paos_pl, mlxsw_sp_port->local_port, 0);
+	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(paos), paos_pl);
+	if (err)
+		return err;
+	oper_status = mlxsw_reg_paos_oper_status_get(paos_pl);
+	*p_is_up = oper_status == MLXSW_PORT_ADMIN_STATUS_UP ? true : false;
+	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)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char ppad_pl[MLXSW_REG_PPAD_LEN];
+
+	mlxsw_reg_ppad_pack(ppad_pl, true, mlxsw_sp_port->local_port);
+	mlxsw_reg_ppad_mac_memcpy_to(ppad_pl, addr);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ppad), ppad_pl);
+}
+
+static int mlxsw_sp_port_dev_addr_init(struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	unsigned char *addr = mlxsw_sp_port->dev->dev_addr;
+
+	ether_addr_copy(addr, mlxsw_sp->base_mac);
+	addr[ETH_ALEN - 1] += mlxsw_sp_port->local_port;
+	return mlxsw_sp_port_dev_addr_set(mlxsw_sp_port, addr);
+}
+
+static int mlxsw_sp_port_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				       u16 vid, enum mlxsw_reg_spms_state state)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char *spms_pl;
+	int err;
+
+	spms_pl = kmalloc(MLXSW_REG_SPMS_LEN, GFP_KERNEL);
+	if (!spms_pl)
+		return -ENOMEM;
+	mlxsw_reg_spms_pack(spms_pl, mlxsw_sp_port->local_port);
+	mlxsw_reg_spms_vid_pack(spms_pl, vid, state);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spms), spms_pl);
+	kfree(spms_pl);
+	return err;
+}
+
+static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char pmtu_pl[MLXSW_REG_PMTU_LEN];
+	int max_mtu;
+	int err;
+
+	mtu += MLXSW_TXHDR_LEN + ETH_HLEN;
+	mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, 0);
+	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl);
+	if (err)
+		return err;
+	max_mtu = mlxsw_reg_pmtu_max_mtu_get(pmtu_pl);
+
+	if (mtu > max_mtu)
+		return -EINVAL;
+
+	mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, mtu);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl);
+}
+
+static int mlxsw_sp_port_swid_set(struct mlxsw_sp_port *mlxsw_sp_port, u8 swid)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char pspa_pl[MLXSW_REG_PSPA_LEN];
+
+	mlxsw_reg_pspa_pack(pspa_pl, swid, mlxsw_sp_port->local_port);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pspa), pspa_pl);
+}
+
+static int mlxsw_sp_port_vp_mode_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				     bool enable)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char svpe_pl[MLXSW_REG_SVPE_LEN];
+
+	mlxsw_reg_svpe_pack(svpe_pl, mlxsw_sp_port->local_port, enable);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svpe), svpe_pl);
+}
+
+int mlxsw_sp_port_vid_to_fid_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				 enum mlxsw_reg_svfa_mt mt, bool valid, u16 fid,
+				 u16 vid)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char svfa_pl[MLXSW_REG_SVFA_LEN];
+
+	mlxsw_reg_svfa_pack(svfa_pl, mlxsw_sp_port->local_port, mt, valid,
+			    fid, vid);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svfa), svfa_pl);
+}
+
+static int mlxsw_sp_port_vid_learning_set(struct mlxsw_sp_port *mlxsw_sp_port,
+					  u16 vid, bool learn_enable)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char *spvmlr_pl;
+	int err;
+
+	spvmlr_pl = kmalloc(MLXSW_REG_SPVMLR_LEN, GFP_KERNEL);
+	if (!spvmlr_pl)
+		return -ENOMEM;
+	mlxsw_reg_spvmlr_pack(spvmlr_pl, mlxsw_sp_port->local_port, vid, vid,
+			      learn_enable);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvmlr), spvmlr_pl);
+	kfree(spvmlr_pl);
+	return err;
+}
+
+static int
+mlxsw_sp_port_system_port_mapping_set(struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char sspr_pl[MLXSW_REG_SSPR_LEN];
+
+	mlxsw_reg_sspr_pack(sspr_pl, mlxsw_sp_port->local_port);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sspr), sspr_pl);
+}
+
+static int mlxsw_sp_port_module_check(struct mlxsw_sp_port *mlxsw_sp_port,
+				      bool *p_usable)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char pmlp_pl[MLXSW_REG_PMLP_LEN];
+	int err;
+
+	mlxsw_reg_pmlp_pack(pmlp_pl, mlxsw_sp_port->local_port);
+	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pmlp), pmlp_pl);
+	if (err)
+		return err;
+	*p_usable = mlxsw_reg_pmlp_width_get(pmlp_pl) ? true : false;
+	return 0;
+}
+
+static int mlxsw_sp_port_open(struct net_device *dev)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	int err;
+
+	err = mlxsw_sp_port_admin_status_set(mlxsw_sp_port, true);
+	if (err)
+		return err;
+	netif_start_queue(dev);
+	return 0;
+}
+
+static int mlxsw_sp_port_stop(struct net_device *dev)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+
+	netif_stop_queue(dev);
+	return mlxsw_sp_port_admin_status_set(mlxsw_sp_port, false);
+}
+
+static netdev_tx_t mlxsw_sp_port_xmit(struct sk_buff *skb,
+				      struct net_device *dev)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	struct mlxsw_sp_port_pcpu_stats *pcpu_stats;
+	const struct mlxsw_tx_info tx_info = {
+		.local_port = mlxsw_sp_port->local_port,
+		.is_emad = false,
+	};
+	u64 len;
+	int err;
+
+	if (mlxsw_core_skb_transmit_busy(mlxsw_sp, &tx_info))
+		return NETDEV_TX_BUSY;
+
+	if (unlikely(skb_headroom(skb) < MLXSW_TXHDR_LEN)) {
+		struct sk_buff *skb_orig = skb;
+
+		skb = skb_realloc_headroom(skb, MLXSW_TXHDR_LEN);
+		if (!skb) {
+			this_cpu_inc(mlxsw_sp_port->pcpu_stats->tx_dropped);
+			dev_kfree_skb_any(skb_orig);
+			return NETDEV_TX_OK;
+		}
+	}
+
+	if (eth_skb_pad(skb)) {
+		this_cpu_inc(mlxsw_sp_port->pcpu_stats->tx_dropped);
+		return NETDEV_TX_OK;
+	}
+
+	mlxsw_sp_txhdr_construct(skb, &tx_info);
+	len = skb->len;
+	/* Due to a race we might fail here because of a full queue. In that
+	 * unlikely case we simply drop the packet.
+	 */
+	err = mlxsw_core_skb_transmit(mlxsw_sp, skb, &tx_info);
+
+	if (!err) {
+		pcpu_stats = this_cpu_ptr(mlxsw_sp_port->pcpu_stats);
+		u64_stats_update_begin(&pcpu_stats->syncp);
+		pcpu_stats->tx_packets++;
+		pcpu_stats->tx_bytes += len;
+		u64_stats_update_end(&pcpu_stats->syncp);
+	} else {
+		this_cpu_inc(mlxsw_sp_port->pcpu_stats->tx_dropped);
+		dev_kfree_skb_any(skb);
+	}
+	return NETDEV_TX_OK;
+}
+
+static int mlxsw_sp_port_set_mac_address(struct net_device *dev, void *p)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct sockaddr *addr = p;
+	int err;
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	err = mlxsw_sp_port_dev_addr_set(mlxsw_sp_port, addr->sa_data);
+	if (err)
+		return err;
+	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	return 0;
+}
+
+static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	int err;
+
+	err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, mtu);
+	if (err)
+		return err;
+	dev->mtu = mtu;
+	return 0;
+}
+
+static struct rtnl_link_stats64 *
+mlxsw_sp_port_get_stats64(struct net_device *dev,
+			  struct rtnl_link_stats64 *stats)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp_port_pcpu_stats *p;
+	u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
+	u32 tx_dropped = 0;
+	unsigned int start;
+	int i;
+
+	for_each_possible_cpu(i) {
+		p = per_cpu_ptr(mlxsw_sp_port->pcpu_stats, i);
+		do {
+			start = u64_stats_fetch_begin_irq(&p->syncp);
+			rx_packets	= p->rx_packets;
+			rx_bytes	= p->rx_bytes;
+			tx_packets	= p->tx_packets;
+			tx_bytes	= p->tx_bytes;
+		} while (u64_stats_fetch_retry_irq(&p->syncp, start));
+
+		stats->rx_packets	+= rx_packets;
+		stats->rx_bytes		+= rx_bytes;
+		stats->tx_packets	+= tx_packets;
+		stats->tx_bytes		+= tx_bytes;
+		/* tx_dropped is u32, updated without syncp protection. */
+		tx_dropped	+= p->tx_dropped;
+	}
+	stats->tx_dropped	= tx_dropped;
+	return stats;
+}
+
+int mlxsw_sp_port_vlan_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin,
+			   u16 vid_end, bool is_member, bool untagged)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char *spvm_pl;
+	int err;
+
+	spvm_pl = kmalloc(MLXSW_REG_SPVM_LEN, GFP_KERNEL);
+	if (!spvm_pl)
+		return -ENOMEM;
+
+	mlxsw_reg_spvm_pack(spvm_pl, mlxsw_sp_port->local_port,	vid_begin,
+			    vid_end, is_member, untagged);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvm), spvm_pl);
+	kfree(spvm_pl);
+	return err;
+}
+
+static int mlxsw_sp_port_vp_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID;
+	u16 vid, last_visited_vid;
+	int err;
+
+	for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) {
+		err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, true, vid,
+						   vid);
+		if (err) {
+			last_visited_vid = vid;
+			goto err_port_vid_to_fid_set;
+		}
+	}
+
+	err = mlxsw_sp_port_vp_mode_set(mlxsw_sp_port, true);
+	if (err) {
+		last_visited_vid = VLAN_N_VID;
+		goto err_port_vid_to_fid_set;
+	}
+
+	return 0;
+
+err_port_vid_to_fid_set:
+	for_each_set_bit(vid, mlxsw_sp_port->active_vlans, last_visited_vid)
+		mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, false, vid,
+					     vid);
+	return err;
+}
+
+static int mlxsw_sp_port_vlan_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port)
+{
+	enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID;
+	u16 vid;
+	int err;
+
+	err = mlxsw_sp_port_vp_mode_set(mlxsw_sp_port, false);
+	if (err)
+		return err;
+
+	for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) {
+		err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, false,
+						   vid, vid);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+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;
+	int err;
+
+	/* VLAN 0 is added to HW filter when device goes up, but it is
+	 * reserved in our case, so simply return.
+	 */
+	if (!vid)
+		return 0;
+
+	if (test_bit(vid, mlxsw_sp_port->active_vfids)) {
+		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;
+		}
+
+		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);
+		if (err) {
+			netdev_err(dev, "Failed to configure flood table\n");
+			goto err_flood_table_config;
+		}
+	}
+
+	/* 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) {
+		err = mlxsw_sp_port_vp_mode_trans(mlxsw_sp_port);
+		if (err) {
+			netdev_err(dev, "Failed to set to Virtual mode\n");
+			return err;
+		}
+	}
+
+	err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port,
+					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
+					   true, MLXSW_SP_VFID_BASE + vid, vid);
+	if (err) {
+		netdev_err(dev, "Failed to map {Port, VID=%d} to vFID=%d\n",
+			   vid, MLXSW_SP_VFID_BASE + vid);
+		goto err_port_vid_to_fid_set;
+	}
+
+	err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, 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);
+	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,
+					  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);
+
+	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);
+err_port_add_vid:
+	mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, true);
+err_port_vid_learning_set:
+	mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port,
+				     MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false,
+				     MLXSW_SP_VFID_BASE + vid, vid);
+err_port_vid_to_fid_set:
+	mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port);
+	return err;
+}
+
+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);
+	int err;
+
+	/* VLAN 0 is removed from HW filter when device goes down, but
+	 * it is reserved in our case, so simply return.
+	 */
+	if (!vid)
+		return 0;
+
+	if (!test_bit(vid, mlxsw_sp_port->active_vfids)) {
+		netdev_warn(dev, "VID=%d does not exist\n", vid);
+		return 0;
+	}
+
+	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_port, 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);
+	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);
+	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,
+					   MLXSW_REG_SVFA_MT_PORT_VID_TO_FID,
+					   false, MLXSW_SP_VFID_BASE + vid,
+					   vid);
+	if (err) {
+		netdev_err(dev, "Failed to invalidate {Port, VID=%d} to vFID=%d mapping\n",
+			   vid, MLXSW_SP_VFID_BASE + vid);
+		return err;
+	}
+
+	/* When removing the last VLAN interface on a bridged port we need to
+	 * 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) {
+		err = mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port);
+		if (err) {
+			netdev_err(dev, "Failed to set to VLAN mode\n");
+			return err;
+		}
+	}
+
+	mlxsw_sp_port->nr_vfids--;
+	clear_bit(vid, mlxsw_sp_port->active_vfids);
+
+	return 0;
+}
+
+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_mac_address	= mlxsw_sp_port_set_mac_address,
+	.ndo_change_mtu		= mlxsw_sp_port_change_mtu,
+	.ndo_get_stats64	= mlxsw_sp_port_get_stats64,
+	.ndo_vlan_rx_add_vid	= mlxsw_sp_port_add_vid,
+	.ndo_vlan_rx_kill_vid	= mlxsw_sp_port_kill_vid,
+	.ndo_fdb_add		= switchdev_port_fdb_add,
+	.ndo_fdb_del		= switchdev_port_fdb_del,
+	.ndo_fdb_dump		= switchdev_port_fdb_dump,
+	.ndo_bridge_setlink	= switchdev_port_bridge_setlink,
+	.ndo_bridge_getlink	= switchdev_port_bridge_getlink,
+	.ndo_bridge_dellink	= switchdev_port_bridge_dellink,
+};
+
+static void mlxsw_sp_port_get_drvinfo(struct net_device *dev,
+				      struct ethtool_drvinfo *drvinfo)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+
+	strlcpy(drvinfo->driver, mlxsw_sp_driver_name, sizeof(drvinfo->driver));
+	strlcpy(drvinfo->version, mlxsw_sp_driver_version,
+		sizeof(drvinfo->version));
+	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+		 "%d.%d.%d",
+		 mlxsw_sp->bus_info->fw_rev.major,
+		 mlxsw_sp->bus_info->fw_rev.minor,
+		 mlxsw_sp->bus_info->fw_rev.subminor);
+	strlcpy(drvinfo->bus_info, mlxsw_sp->bus_info->device_name,
+		sizeof(drvinfo->bus_info));
+}
+
+struct mlxsw_sp_port_hw_stats {
+	char str[ETH_GSTRING_LEN];
+	u64 (*getter)(char *payload);
+};
+
+static const struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_stats[] = {
+	{
+		.str = "a_frames_transmitted_ok",
+		.getter = mlxsw_reg_ppcnt_a_frames_transmitted_ok_get,
+	},
+	{
+		.str = "a_frames_received_ok",
+		.getter = mlxsw_reg_ppcnt_a_frames_received_ok_get,
+	},
+	{
+		.str = "a_frame_check_sequence_errors",
+		.getter = mlxsw_reg_ppcnt_a_frame_check_sequence_errors_get,
+	},
+	{
+		.str = "a_alignment_errors",
+		.getter = mlxsw_reg_ppcnt_a_alignment_errors_get,
+	},
+	{
+		.str = "a_octets_transmitted_ok",
+		.getter = mlxsw_reg_ppcnt_a_octets_transmitted_ok_get,
+	},
+	{
+		.str = "a_octets_received_ok",
+		.getter = mlxsw_reg_ppcnt_a_octets_received_ok_get,
+	},
+	{
+		.str = "a_multicast_frames_xmitted_ok",
+		.getter = mlxsw_reg_ppcnt_a_multicast_frames_xmitted_ok_get,
+	},
+	{
+		.str = "a_broadcast_frames_xmitted_ok",
+		.getter = mlxsw_reg_ppcnt_a_broadcast_frames_xmitted_ok_get,
+	},
+	{
+		.str = "a_multicast_frames_received_ok",
+		.getter = mlxsw_reg_ppcnt_a_multicast_frames_received_ok_get,
+	},
+	{
+		.str = "a_broadcast_frames_received_ok",
+		.getter = mlxsw_reg_ppcnt_a_broadcast_frames_received_ok_get,
+	},
+	{
+		.str = "a_in_range_length_errors",
+		.getter = mlxsw_reg_ppcnt_a_in_range_length_errors_get,
+	},
+	{
+		.str = "a_out_of_range_length_field",
+		.getter = mlxsw_reg_ppcnt_a_out_of_range_length_field_get,
+	},
+	{
+		.str = "a_frame_too_long_errors",
+		.getter = mlxsw_reg_ppcnt_a_frame_too_long_errors_get,
+	},
+	{
+		.str = "a_symbol_error_during_carrier",
+		.getter = mlxsw_reg_ppcnt_a_symbol_error_during_carrier_get,
+	},
+	{
+		.str = "a_mac_control_frames_transmitted",
+		.getter = mlxsw_reg_ppcnt_a_mac_control_frames_transmitted_get,
+	},
+	{
+		.str = "a_mac_control_frames_received",
+		.getter = mlxsw_reg_ppcnt_a_mac_control_frames_received_get,
+	},
+	{
+		.str = "a_unsupported_opcodes_received",
+		.getter = mlxsw_reg_ppcnt_a_unsupported_opcodes_received_get,
+	},
+	{
+		.str = "a_pause_mac_ctrl_frames_received",
+		.getter = mlxsw_reg_ppcnt_a_pause_mac_ctrl_frames_received_get,
+	},
+	{
+		.str = "a_pause_mac_ctrl_frames_xmitted",
+		.getter = mlxsw_reg_ppcnt_a_pause_mac_ctrl_frames_transmitted_get,
+	},
+};
+
+#define MLXSW_SP_PORT_HW_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_stats)
+
+static void mlxsw_sp_port_get_strings(struct net_device *dev,
+				      u32 stringset, u8 *data)
+{
+	u8 *p = data;
+	int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < MLXSW_SP_PORT_HW_STATS_LEN; i++) {
+			memcpy(p, mlxsw_sp_port_hw_stats[i].str,
+			       ETH_GSTRING_LEN);
+			p += ETH_GSTRING_LEN;
+		}
+		break;
+	}
+}
+
+static void mlxsw_sp_port_get_stats(struct net_device *dev,
+				    struct ethtool_stats *stats, u64 *data)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char ppcnt_pl[MLXSW_REG_PPCNT_LEN];
+	int i;
+	int err;
+
+	mlxsw_reg_ppcnt_pack(ppcnt_pl, mlxsw_sp_port->local_port);
+	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ppcnt), ppcnt_pl);
+	for (i = 0; i < MLXSW_SP_PORT_HW_STATS_LEN; i++)
+		data[i] = !err ? mlxsw_sp_port_hw_stats[i].getter(ppcnt_pl) : 0;
+}
+
+static int mlxsw_sp_port_get_sset_count(struct net_device *dev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS:
+		return MLXSW_SP_PORT_HW_STATS_LEN;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+struct mlxsw_sp_port_link_mode {
+	u32 mask;
+	u32 supported;
+	u32 advertised;
+	u32 speed;
+};
+
+static const struct mlxsw_sp_port_link_mode mlxsw_sp_port_link_mode[] = {
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_100BASE_T,
+		.supported	= SUPPORTED_100baseT_Full,
+		.advertised	= ADVERTISED_100baseT_Full,
+		.speed		= 100,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_100BASE_TX,
+		.speed		= 100,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_SGMII |
+				  MLXSW_REG_PTYS_ETH_SPEED_1000BASE_KX,
+		.supported	= SUPPORTED_1000baseKX_Full,
+		.advertised	= ADVERTISED_1000baseKX_Full,
+		.speed		= 1000,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_10GBASE_T,
+		.supported	= SUPPORTED_10000baseT_Full,
+		.advertised	= ADVERTISED_10000baseT_Full,
+		.speed		= 10000,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CX4 |
+				  MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KX4,
+		.supported	= SUPPORTED_10000baseKX4_Full,
+		.advertised	= ADVERTISED_10000baseKX4_Full,
+		.speed		= 10000,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KR |
+				  MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CR |
+				  MLXSW_REG_PTYS_ETH_SPEED_10GBASE_SR |
+				  MLXSW_REG_PTYS_ETH_SPEED_10GBASE_ER_LR,
+		.supported	= SUPPORTED_10000baseKR_Full,
+		.advertised	= ADVERTISED_10000baseKR_Full,
+		.speed		= 10000,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_20GBASE_KR2,
+		.supported	= SUPPORTED_20000baseKR2_Full,
+		.advertised	= ADVERTISED_20000baseKR2_Full,
+		.speed		= 20000,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_40GBASE_CR4,
+		.supported	= SUPPORTED_40000baseCR4_Full,
+		.advertised	= ADVERTISED_40000baseCR4_Full,
+		.speed		= 40000,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_40GBASE_KR4,
+		.supported	= SUPPORTED_40000baseKR4_Full,
+		.advertised	= ADVERTISED_40000baseKR4_Full,
+		.speed		= 40000,
+	},
+	{
+		.mask		= MLXSW_REG_PTYS_ETH_SPEED_40GBASE_SR4,
+		.supported	= SUPPORTED_40000baseSR4_Full