Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:
 "Here is my initial pull request for the networking subsystem during
  this merge window:

   1) Support for ESN in AH (RFC 4302) from Fan Du.

   2) Add full kernel doc for ethtool command structures, from Ben
      Hutchings.

   3) Add BCM7xxx PHY driver, from Florian Fainelli.

   4) Export computed TCP rate information in netlink socket dumps, from
      Eric Dumazet.

   5) Allow IPSEC SA to be dumped partially using a filter, from Nicolas
      Dichtel.

   6) Convert many drivers to pci_enable_msix_range(), from Alexander
      Gordeev.

   7) Record SKB timestamps more efficiently, from Eric Dumazet.

   8) Switch to microsecond resolution for TCP round trip times, also
      from Eric Dumazet.

   9) Clean up and fix 6lowpan fragmentation handling by making use of
      the existing inet_frag api for it's implementation.

  10) Add TX grant mapping to xen-netback driver, from Zoltan Kiss.

  11) Auto size SKB lengths when composing netlink messages based upon
      past message sizes used, from Eric Dumazet.

  12) qdisc dumps can take a long time, add a cond_resched(), From Eric
      Dumazet.

  13) Sanitize netpoll core and drivers wrt.  SKB handling semantics.
      Get rid of never-used-in-tree netpoll RX handling.  From Eric W
      Biederman.

  14) Support inter-address-family and namespace changing in VTI tunnel
      driver(s).  From Steffen Klassert.

  15) Add Altera TSE driver, from Vince Bridgers.

  16) Optimizing csum_replace2() so that it doesn't adjust the checksum
      by checksumming the entire header, from Eric Dumazet.

  17) Expand BPF internal implementation for faster interpreting, more
      direct translations into JIT'd code, and much cleaner uses of BPF
      filtering in non-socket ocntexts.  From Daniel Borkmann and Alexei
      Starovoitov"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1976 commits)
  netpoll: Use skb_irq_freeable to make zap_completion_queue safe.
  net: Add a test to see if a skb is freeable in irq context
  qlcnic: Fix build failure due to undefined reference to `vxlan_get_rx_port'
  net: ptp: move PTP classifier in its own file
  net: sxgbe: make "core_ops" static
  net: sxgbe: fix logical vs bitwise operation
  net: sxgbe: sxgbe_mdio_register() frees the bus
  Call efx_set_channels() before efx->type->dimension_resources()
  xen-netback: disable rogue vif in kthread context
  net/mlx4: Set proper build dependancy with vxlan
  be2net: fix build dependency on VxLAN
  mac802154: make csma/cca parameters per-wpan
  mac802154: allow only one WPAN to be up at any given time
  net: filter: minor: fix kdoc in __sk_run_filter
  netlink: don't compare the nul-termination in nla_strcmp
  can: c_can: Avoid led toggling for every packet.
  can: c_can: Simplify TX interrupt cleanup
  can: c_can: Store dlc private
  can: c_can: Reduce register access
  can: c_can: Make the code readable
  ...

19 files changed, 1165 insertions, 220 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx4/Kconfig b/drivers/net/ethernet/mellanox/mlx4/Kconfig
index 563495d8975a..1486ce902a56 100644
--- a/drivers/net/ethernet/mellanox/mlx4/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx4/Kconfig
@@ -3,7 +3,7 @@
 #
 
 config MLX4_EN
-	tristate "Mellanox Technologies 10Gbit Ethernet support"
+	tristate "Mellanox Technologies 1/10/40Gbit Ethernet support"
 	depends on PCI
 	select MLX4_CORE
 	select PTP_1588_CLOCK
@@ -23,6 +23,13 @@ config MLX4_EN_DCB
 
 	  If unsure, set to Y
 
+config MLX4_EN_VXLAN
+	bool "VXLAN offloads Support"
+	default y
+	depends on MLX4_EN && VXLAN && !(MLX4_EN=y && VXLAN=m)
+	---help---
+	  Say Y here if you want to use VXLAN offloads in the driver.
+
 config MLX4_CORE
 	tristate
 	depends on PCI
diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c
index 0d02fba94536..78099eab7673 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -800,16 +800,7 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave,
 				    vhcr->op, MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE);
 }
 
-static int MLX4_CMD_UPDATE_QP_wrapper(struct mlx4_dev *dev, int slave,
-		     struct mlx4_vhcr *vhcr,
-		     struct mlx4_cmd_mailbox *inbox,
-		     struct mlx4_cmd_mailbox *outbox,
-		     struct mlx4_cmd_info *cmd)
-{
-	return -EPERM;
-}
-
-static int MLX4_CMD_GET_OP_REQ_wrapper(struct mlx4_dev *dev, int slave,
+static int mlx4_CMD_EPERM_wrapper(struct mlx4_dev *dev, int slave,
 		     struct mlx4_vhcr *vhcr,
 		     struct mlx4_cmd_mailbox *inbox,
 		     struct mlx4_cmd_mailbox *outbox,
@@ -964,6 +955,15 @@ static struct mlx4_cmd_info cmd_info[] = {
 		.wrapper = NULL
 	},
 	{
+		.opcode = MLX4_CMD_CONFIG_DEV,
+		.has_inbox = false,
+		.has_outbox = false,
+		.out_is_imm = false,
+		.encode_slave_id = false,
+		.verify = NULL,
+		.wrapper = mlx4_CMD_EPERM_wrapper
+	},
+	{
 		.opcode = MLX4_CMD_ALLOC_RES,
 		.has_inbox = false,
 		.has_outbox = false,
@@ -1258,7 +1258,7 @@ static struct mlx4_cmd_info cmd_info[] = {
 		.out_is_imm = false,
 		.encode_slave_id = false,
 		.verify = NULL,
-		.wrapper = MLX4_CMD_UPDATE_QP_wrapper
+		.wrapper = mlx4_CMD_EPERM_wrapper
 	},
 	{
 		.opcode = MLX4_CMD_GET_OP_REQ,
@@ -1267,7 +1267,7 @@ static struct mlx4_cmd_info cmd_info[] = {
 		.out_is_imm = false,
 		.encode_slave_id = false,
 		.verify = NULL,
-		.wrapper = MLX4_CMD_GET_OP_REQ_wrapper,
+		.wrapper = mlx4_CMD_EPERM_wrapper,
 	},
 	{
 		.opcode = MLX4_CMD_CONF_SPECIAL_QP,
@@ -1378,7 +1378,7 @@ static struct mlx4_cmd_info cmd_info[] = {
 		.out_is_imm = false,
 		.encode_slave_id = false,
 		.verify = NULL,
-		.wrapper = mlx4_FLOW_STEERING_IB_UC_QP_RANGE_wrapper
+		.wrapper = mlx4_CMD_EPERM_wrapper
 	},
 };
 
@@ -1643,8 +1643,16 @@ static int mlx4_master_activate_admin_state(struct mlx4_priv *priv, int slave)
 	int port, err;
 	struct mlx4_vport_state *vp_admin;
 	struct mlx4_vport_oper_state *vp_oper;
-
-	for (port = 1; port <= MLX4_MAX_PORTS; port++) {
+	struct mlx4_active_ports actv_ports = mlx4_get_active_ports(
+			&priv->dev, slave);
+	int min_port = find_first_bit(actv_ports.ports,
+				      priv->dev.caps.num_ports) + 1;
+	int max_port = min_port - 1 +
+		bitmap_weight(actv_ports.ports, priv->dev.caps.num_ports);
+
+	for (port = min_port; port <= max_port; port++) {
+		if (!test_bit(port - 1, actv_ports.ports))
+			continue;
 		vp_oper = &priv->mfunc.master.vf_oper[slave].vport[port];
 		vp_admin = &priv->mfunc.master.vf_admin[slave].vport[port];
 		vp_oper->state = *vp_admin;
@@ -1685,8 +1693,17 @@ static void mlx4_master_deactivate_admin_state(struct mlx4_priv *priv, int slave
 {
 	int port;
 	struct mlx4_vport_oper_state *vp_oper;
+	struct mlx4_active_ports actv_ports = mlx4_get_active_ports(
+			&priv->dev, slave);
+	int min_port = find_first_bit(actv_ports.ports,
+				      priv->dev.caps.num_ports) + 1;
+	int max_port = min_port - 1 +
+		bitmap_weight(actv_ports.ports, priv->dev.caps.num_ports);
+
 
-	for (port = 1; port <= MLX4_MAX_PORTS; port++) {
+	for (port = min_port; port <= max_port; port++) {
+		if (!test_bit(port - 1, actv_ports.ports))
+			continue;
 		vp_oper = &priv->mfunc.master.vf_oper[slave].vport[port];
 		if (NO_INDX != vp_oper->vlan_idx) {
 			__mlx4_unregister_vlan(&priv->dev,
@@ -2234,6 +2251,112 @@ static int mlx4_get_slave_indx(struct mlx4_dev *dev, int vf)
 	return vf+1;
 }
 
+int mlx4_get_vf_indx(struct mlx4_dev *dev, int slave)
+{
+	if (slave < 1 || slave > dev->num_vfs) {
+		mlx4_err(dev,
+			 "Bad slave number:%d (number of activated slaves: %lu)\n",
+			 slave, dev->num_slaves);
+		return -EINVAL;
+	}
+	return slave - 1;
+}
+
+struct mlx4_active_ports mlx4_get_active_ports(struct mlx4_dev *dev, int slave)
+{
+	struct mlx4_active_ports actv_ports;
+	int vf;
+
+	bitmap_zero(actv_ports.ports, MLX4_MAX_PORTS);
+
+	if (slave == 0) {
+		bitmap_fill(actv_ports.ports, dev->caps.num_ports);
+		return actv_ports;
+	}
+
+	vf = mlx4_get_vf_indx(dev, slave);
+	if (vf < 0)
+		return actv_ports;
+
+	bitmap_set(actv_ports.ports, dev->dev_vfs[vf].min_port - 1,
+		   min((int)dev->dev_vfs[mlx4_get_vf_indx(dev, slave)].n_ports,
+		   dev->caps.num_ports));
+
+	return actv_ports;
+}
+EXPORT_SYMBOL_GPL(mlx4_get_active_ports);
+
+int mlx4_slave_convert_port(struct mlx4_dev *dev, int slave, int port)
+{
+	unsigned n;
+	struct mlx4_active_ports actv_ports = mlx4_get_active_ports(dev, slave);
+	unsigned m = bitmap_weight(actv_ports.ports, dev->caps.num_ports);
+
+	if (port <= 0 || port > m)
+		return -EINVAL;
+
+	n = find_first_bit(actv_ports.ports, dev->caps.num_ports);
+	if (port <= n)
+		port = n + 1;
+
+	return port;
+}
+EXPORT_SYMBOL_GPL(mlx4_slave_convert_port);
+
+int mlx4_phys_to_slave_port(struct mlx4_dev *dev, int slave, int port)
+{
+	struct mlx4_active_ports actv_ports = mlx4_get_active_ports(dev, slave);
+	if (test_bit(port - 1, actv_ports.ports))
+		return port -
+			find_first_bit(actv_ports.ports, dev->caps.num_ports);
+
+	return -1;
+}
+EXPORT_SYMBOL_GPL(mlx4_phys_to_slave_port);
+
+struct mlx4_slaves_pport mlx4_phys_to_slaves_pport(struct mlx4_dev *dev,
+						   int port)
+{
+	unsigned i;
+	struct mlx4_slaves_pport slaves_pport;
+
+	bitmap_zero(slaves_pport.slaves, MLX4_MFUNC_MAX);
+
+	if (port <= 0 || port > dev->caps.num_ports)
+		return slaves_pport;
+
+	for (i = 0; i < dev->num_vfs + 1; i++) {
+		struct mlx4_active_ports actv_ports =
+			mlx4_get_active_ports(dev, i);
+		if (test_bit(port - 1, actv_ports.ports))
+			set_bit(i, slaves_pport.slaves);
+	}
+
+	return slaves_pport;
+}
+EXPORT_SYMBOL_GPL(mlx4_phys_to_slaves_pport);
+
+struct mlx4_slaves_pport mlx4_phys_to_slaves_pport_actv(
+		struct mlx4_dev *dev,
+		const struct mlx4_active_ports *crit_ports)
+{
+	unsigned i;
+	struct mlx4_slaves_pport slaves_pport;
+
+	bitmap_zero(slaves_pport.slaves, MLX4_MFUNC_MAX);
+
+	for (i = 0; i < dev->num_vfs + 1; i++) {
+		struct mlx4_active_ports actv_ports =
+			mlx4_get_active_ports(dev, i);
+		if (bitmap_equal(crit_ports->ports, actv_ports.ports,
+				 dev->caps.num_ports))
+			set_bit(i, slaves_pport.slaves);
+	}
+
+	return slaves_pport;
+}
+EXPORT_SYMBOL_GPL(mlx4_phys_to_slaves_pport_actv);
+
 int mlx4_set_vf_mac(struct mlx4_dev *dev, int port, int vf, u64 mac)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
@@ -2289,6 +2412,30 @@ int mlx4_set_vf_vlan(struct mlx4_dev *dev, int port, int vf, u16 vlan, u8 qos)
 }
 EXPORT_SYMBOL_GPL(mlx4_set_vf_vlan);
 
+ /* mlx4_get_slave_default_vlan -
+ * return true if VST ( default vlan)
+ * if VST, will return vlan & qos (if not NULL)
+ */
+bool mlx4_get_slave_default_vlan(struct mlx4_dev *dev, int port, int slave,
+				 u16 *vlan, u8 *qos)
+{
+	struct mlx4_vport_oper_state *vp_oper;
+	struct mlx4_priv *priv;
+
+	priv = mlx4_priv(dev);
+	vp_oper = &priv->mfunc.master.vf_oper[slave].vport[port];
+
+	if (MLX4_VGT != vp_oper->state.default_vlan) {
+		if (vlan)
+			*vlan = vp_oper->state.default_vlan;
+		if (qos)
+			*qos = vp_oper->state.default_qos;
+		return true;
+	}
+	return false;
+}
+EXPORT_SYMBOL_GPL(mlx4_get_slave_default_vlan);
+
 int mlx4_set_vf_spoofchk(struct mlx4_dev *dev, int port, int vf, bool setting)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_clock.c b/drivers/net/ethernet/mellanox/mlx4/en_clock.c
index abaf6bb22416..57dda95b67d8 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_clock.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_clock.c
@@ -276,6 +276,7 @@ static const struct ptp_clock_info mlx4_en_ptp_clock_info = {
 	.n_alarm	= 0,
 	.n_ext_ts	= 0,
 	.n_per_out	= 0,
+	.n_pins		= 0,
 	.pps		= 0,
 	.adjfreq	= mlx4_en_phc_adjfreq,
 	.adjtime	= mlx4_en_phc_adjtime,
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c
index b4881b686159..c95ca252187c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c
@@ -62,7 +62,7 @@ static int mlx4_en_ets_validate(struct mlx4_en_priv *priv, struct ieee_ets *ets)
 	int has_ets_tc = 0;
 
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
-		if (ets->prio_tc[i] > MLX4_EN_NUM_UP) {
+		if (ets->prio_tc[i] >= MLX4_EN_NUM_UP) {
 			en_err(priv, "Bad priority in UP <=> TC mapping. TC: %d, UP: %d\n",
 					i, ets->prio_tc[i]);
 			return -EINVAL;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_main.c b/drivers/net/ethernet/mellanox/mlx4/en_main.c
index d357bf5a4686..0c59d4fe7e3a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_main.c
@@ -72,6 +72,12 @@ MLX4_EN_PARM_INT(pfctx, 0, "Priority based Flow Control policy on TX[7:0]."
 MLX4_EN_PARM_INT(pfcrx, 0, "Priority based Flow Control policy on RX[7:0]."
 			   " Per priority bit mask");
 
+MLX4_EN_PARM_INT(inline_thold, MAX_INLINE,
+		 "Threshold for using inline data (range: 17-104, default: 104)");
+
+#define MAX_PFC_TX     0xff
+#define MAX_PFC_RX     0xff
+
 int en_print(const char *level, const struct mlx4_en_priv *priv,
 	     const char *format, ...)
 {
@@ -140,6 +146,7 @@ static int mlx4_en_get_profile(struct mlx4_en_dev *mdev)
 		params->prof[i].tx_ring_num = params->num_tx_rings_p_up *
 			MLX4_EN_NUM_UP;
 		params->prof[i].rss_rings = 0;
+		params->prof[i].inline_thold = inline_thold;
 	}
 
 	return 0;
@@ -274,19 +281,8 @@ static void *mlx4_en_add(struct mlx4_dev *dev)
 	if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS)
 		mlx4_en_init_timestamp(mdev);
 
-	mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) {
-		if (!dev->caps.comp_pool) {
-			mdev->profile.prof[i].rx_ring_num =
-				rounddown_pow_of_two(max_t(int, MIN_RX_RINGS,
-							   min_t(int,
-								 dev->caps.num_comp_vectors,
-								 DEF_RX_RINGS)));
-		} else {
-			mdev->profile.prof[i].rx_ring_num = rounddown_pow_of_two(
-				min_t(int, dev->caps.comp_pool/
-				      dev->caps.num_ports - 1 , MAX_MSIX_P_PORT - 1));
-		}
-	}
+	/* Set default number of RX rings*/
+	mlx4_en_set_num_rx_rings(mdev);
 
 	/* Create our own workqueue for reset/multicast tasks
 	 * Note: we cannot use the shared workqueue because of deadlocks caused
@@ -336,8 +332,31 @@ static struct mlx4_interface mlx4_en_interface = {
 	.protocol	= MLX4_PROT_ETH,
 };
 
+static void mlx4_en_verify_params(void)
+{
+	if (pfctx > MAX_PFC_TX) {
+		pr_warn("mlx4_en: WARNING: illegal module parameter pfctx 0x%x - should be in range 0-0x%x, will be changed to default (0)\n",
+			pfctx, MAX_PFC_TX);
+		pfctx = 0;
+	}
+
+	if (pfcrx > MAX_PFC_RX) {
+		pr_warn("mlx4_en: WARNING: illegal module parameter pfcrx 0x%x - should be in range 0-0x%x, will be changed to default (0)\n",
+			pfcrx, MAX_PFC_RX);
+		pfcrx = 0;
+	}
+
+	if (inline_thold < MIN_PKT_LEN || inline_thold > MAX_INLINE) {
+		pr_warn("mlx4_en: WARNING: illegal module parameter inline_thold %d - should be in range %d-%d, will be changed to default (%d)\n",
+			inline_thold, MIN_PKT_LEN, MAX_INLINE, MAX_INLINE);
+		inline_thold = MAX_INLINE;
+	}
+}
+
 static int __init mlx4_en_init(void)
 {
+	mlx4_en_verify_params();
+
 	return mlx4_register_interface(&mlx4_en_interface);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 84a96f70dfb5..f085c2df5e69 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -39,6 +39,7 @@
 #include <linux/hash.h>
 #include <net/ip.h>
 #include <net/busy_poll.h>
+#include <net/vxlan.h>
 
 #include <linux/mlx4/driver.h>
 #include <linux/mlx4/device.h>
@@ -603,7 +604,7 @@ static int mlx4_en_get_qp(struct mlx4_en_priv *priv)
 	int err = 0;
 	u64 reg_id;
 	int *qpn = &priv->base_qpn;
-	u64 mac = mlx4_en_mac_to_u64(priv->dev->dev_addr);
+	u64 mac = mlx4_mac_to_u64(priv->dev->dev_addr);
 
 	en_dbg(DRV, priv, "Registering MAC: %pM for adding\n",
 	       priv->dev->dev_addr);
@@ -672,7 +673,7 @@ static void mlx4_en_put_qp(struct mlx4_en_priv *priv)
 	u64 mac;
 
 	if (dev->caps.steering_mode == MLX4_STEERING_MODE_A0) {
-		mac = mlx4_en_mac_to_u64(priv->dev->dev_addr);
+		mac = mlx4_mac_to_u64(priv->dev->dev_addr);
 		en_dbg(DRV, priv, "Registering MAC: %pM for deleting\n",
 		       priv->dev->dev_addr);
 		mlx4_unregister_mac(dev, priv->port, mac);
@@ -685,7 +686,7 @@ static void mlx4_en_put_qp(struct mlx4_en_priv *priv)
 		for (i = 0; i < MLX4_EN_MAC_HASH_SIZE; ++i) {
 			bucket = &priv->mac_hash[i];
 			hlist_for_each_entry_safe(entry, tmp, bucket, hlist) {
-				mac = mlx4_en_mac_to_u64(entry->mac);
+				mac = mlx4_mac_to_u64(entry->mac);
 				en_dbg(DRV, priv, "Registering MAC: %pM for deleting\n",
 				       entry->mac);
 				mlx4_en_uc_steer_release(priv, entry->mac,
@@ -715,14 +716,14 @@ static int mlx4_en_replace_mac(struct mlx4_en_priv *priv, int qpn,
 	struct mlx4_en_dev *mdev = priv->mdev;
 	struct mlx4_dev *dev = mdev->dev;
 	int err = 0;
-	u64 new_mac_u64 = mlx4_en_mac_to_u64(new_mac);
+	u64 new_mac_u64 = mlx4_mac_to_u64(new_mac);
 
 	if (dev->caps.steering_mode != MLX4_STEERING_MODE_A0) {
 		struct hlist_head *bucket;
 		unsigned int mac_hash;
 		struct mlx4_mac_entry *entry;
 		struct hlist_node *tmp;
-		u64 prev_mac_u64 = mlx4_en_mac_to_u64(prev_mac);
+		u64 prev_mac_u64 = mlx4_mac_to_u64(prev_mac);
 
 		bucket = &priv->mac_hash[prev_mac[MLX4_EN_MAC_HASH_IDX]];
 		hlist_for_each_entry_safe(entry, tmp, bucket, hlist) {
@@ -759,18 +760,6 @@ static int mlx4_en_replace_mac(struct mlx4_en_priv *priv, int qpn,
 	return __mlx4_replace_mac(dev, priv->port, qpn, new_mac_u64);
 }
 
-u64 mlx4_en_mac_to_u64(u8 *addr)
-{
-	u64 mac = 0;
-	int i;
-
-	for (i = 0; i < ETH_ALEN; i++) {
-		mac <<= 8;
-		mac |= addr[i];
-	}
-	return mac;
-}
-
 static int mlx4_en_do_set_mac(struct mlx4_en_priv *priv)
 {
 	int err = 0;
@@ -1089,7 +1078,7 @@ static void mlx4_en_do_multicast(struct mlx4_en_priv *priv,
 		mlx4_en_cache_mclist(dev);
 		netif_addr_unlock_bh(dev);
 		list_for_each_entry(mclist, &priv->mc_list, list) {
-			mcast_addr = mlx4_en_mac_to_u64(mclist->addr);
+			mcast_addr = mlx4_mac_to_u64(mclist->addr);
 			mlx4_SET_MCAST_FLTR(mdev->dev, priv->port,
 					    mcast_addr, 0, MLX4_MCAST_CONFIG);
 		}
@@ -1181,7 +1170,7 @@ static void mlx4_en_do_uc_filter(struct mlx4_en_priv *priv,
 				found = true;
 
 			if (!found) {
-				mac = mlx4_en_mac_to_u64(entry->mac);
+				mac = mlx4_mac_to_u64(entry->mac);
 				mlx4_en_uc_steer_release(priv, entry->mac,
 							 priv->base_qpn,
 							 entry->reg_id);
@@ -1224,7 +1213,7 @@ static void mlx4_en_do_uc_filter(struct mlx4_en_priv *priv,
 				priv->flags |= MLX4_EN_FLAG_FORCE_PROMISC;
 				break;
 			}
-			mac = mlx4_en_mac_to_u64(ha->addr);
+			mac = mlx4_mac_to_u64(ha->addr);
 			memcpy(entry->mac, ha->addr, ETH_ALEN);
 			err = mlx4_register_mac(mdev->dev, priv->port, mac);
 			if (err < 0) {
@@ -1677,7 +1666,7 @@ int mlx4_en_start_port(struct net_device *dev)
 	}
 
 	if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
-		err = mlx4_SET_PORT_VXLAN(mdev->dev, priv->port, VXLAN_STEER_BY_OUTER_MAC);
+		err = mlx4_SET_PORT_VXLAN(mdev->dev, priv->port, VXLAN_STEER_BY_OUTER_MAC, 1);
 		if (err) {
 			en_err(priv, "Failed setting port L2 tunnel configuration, err %d\n",
 			       err);
@@ -1709,6 +1698,10 @@ int mlx4_en_start_port(struct net_device *dev)
 
 	mlx4_set_stats_bitmap(mdev->dev, &priv->stats_bitmap);
 
+#ifdef CONFIG_MLX4_EN_VXLAN
+	if (priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_VXLAN_OFFLOADS)
+		vxlan_get_rx_port(dev);
+#endif
 	priv->port_up = true;
 	netif_tx_start_all_queues(dev);
 	netif_device_attach(dev);
@@ -2216,7 +2209,7 @@ static int mlx4_en_set_vf_mac(struct net_device *dev, int queue, u8 *mac)
 {
 	struct mlx4_en_priv *en_priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = en_priv->mdev;
-	u64 mac_u64 = mlx4_en_mac_to_u64(mac);
+	u64 mac_u64 = mlx4_mac_to_u64(mac);
 
 	if (!is_valid_ether_addr(mac))
 		return -EINVAL;
@@ -2276,6 +2269,83 @@ static int mlx4_en_get_phys_port_id(struct net_device *dev,
 	return 0;
 }
 
+#ifdef CONFIG_MLX4_EN_VXLAN
+static void mlx4_en_add_vxlan_offloads(struct work_struct *work)
+{
+	int ret;
+	struct mlx4_en_priv *priv = container_of(work, struct mlx4_en_priv,
+						 vxlan_add_task);
+
+	ret = mlx4_config_vxlan_port(priv->mdev->dev, priv->vxlan_port);
+	if (ret)
+		goto out;
+
+	ret = mlx4_SET_PORT_VXLAN(priv->mdev->dev, priv->port,
+				  VXLAN_STEER_BY_OUTER_MAC, 1);
+out:
+	if (ret)
+		en_err(priv, "failed setting L2 tunnel configuration ret %d\n", ret);
+}
+
+static void mlx4_en_del_vxlan_offloads(struct work_struct *work)
+{
+	int ret;
+	struct mlx4_en_priv *priv = container_of(work, struct mlx4_en_priv,
+						 vxlan_del_task);
+
+	ret = mlx4_SET_PORT_VXLAN(priv->mdev->dev, priv->port,
+				  VXLAN_STEER_BY_OUTER_MAC, 0);
+	if (ret)
+		en_err(priv, "failed setting L2 tunnel configuration ret %d\n", ret);
+
+	priv->vxlan_port = 0;
+}
+
+static void mlx4_en_add_vxlan_port(struct  net_device *dev,
+				   sa_family_t sa_family, __be16 port)
+{
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	__be16 current_port;
+
+	if (!(priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_VXLAN_OFFLOADS))
+		return;
+
+	if (sa_family == AF_INET6)
+		return;
+
+	current_port = priv->vxlan_port;
+	if (current_port && current_port != port) {
+		en_warn(priv, "vxlan port %d configured, can't add port %d\n",
+			ntohs(current_port), ntohs(port));
+		return;
+	}
+
+	priv->vxlan_port = port;
+	queue_work(priv->mdev->workqueue, &priv->vxlan_add_task);
+}
+
+static void mlx4_en_del_vxlan_port(struct  net_device *dev,
+				   sa_family_t sa_family, __be16 port)
+{
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	__be16 current_port;
+
+	if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN)
+		return;
+
+	if (sa_family == AF_INET6)
+		return;
+
+	current_port = priv->vxlan_port;
+	if (current_port != port) {
+		en_dbg(DRV, priv, "vxlan port %d isn't configured, ignoring\n", ntohs(port));
+		return;
+	}
+
+	queue_work(priv->mdev->workqueue, &priv->vxlan_del_task);
+}
+#endif
+
 static const struct net_device_ops mlx4_netdev_ops = {
 	.ndo_open		= mlx4_en_open,
 	.ndo_stop		= mlx4_en_close,
@@ -2302,6 +2372,10 @@ static const struct net_device_ops mlx4_netdev_ops = {
 	.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,
+	.ndo_del_vxlan_port	= mlx4_en_del_vxlan_port,
+#endif
 };
 
 static const struct net_device_ops mlx4_netdev_ops_master = {
@@ -2351,7 +2425,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 	netif_set_real_num_rx_queues(dev, prof->rx_ring_num);
 
 	SET_NETDEV_DEV(dev, &mdev->dev->pdev->dev);
-	dev->dev_id =  port - 1;
+	dev->dev_port = port - 1;
 
 	/*
 	 * Initialize driver private data
@@ -2393,6 +2467,10 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 	INIT_WORK(&priv->linkstate_task, mlx4_en_linkstate);
 	INIT_DELAYED_WORK(&priv->stats_task, mlx4_en_do_get_stats);
 	INIT_DELAYED_WORK(&priv->service_task, mlx4_en_service_task);
+#ifdef CONFIG_MLX4_EN_VXLAN
+	INIT_WORK(&priv->vxlan_add_task, mlx4_en_add_vxlan_offloads);
+	INIT_WORK(&priv->vxlan_del_task, mlx4_en_del_vxlan_offloads);
+#endif
 #ifdef CONFIG_MLX4_EN_DCB
 	if (!mlx4_is_slave(priv->mdev->dev)) {
 		if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_SET_ETH_SCHED) {
@@ -2417,7 +2495,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 		if (mlx4_is_slave(priv->mdev->dev)) {
 			eth_hw_addr_random(dev);
 			en_warn(priv, "Assigned random MAC address %pM\n", dev->dev_addr);
-			mac_u64 = mlx4_en_mac_to_u64(dev->dev_addr);
+			mac_u64 = mlx4_mac_to_u64(dev->dev_addr);
 			mdev->dev->caps.def_mac[priv->port] = mac_u64;
 		} else {
 			en_err(priv, "Port: %d, invalid mac burned: %pM, quiting\n",
@@ -2526,7 +2604,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 	}
 
 	if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
-		err = mlx4_SET_PORT_VXLAN(mdev->dev, priv->port, VXLAN_STEER_BY_OUTER_MAC);
+		err = mlx4_SET_PORT_VXLAN(mdev->dev, priv->port, VXLAN_STEER_BY_OUTER_MAC, 1);
 		if (err) {
 			en_err(priv, "Failed setting port L2 tunnel configuration, err %d\n",
 			       err);
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_port.c b/drivers/net/ethernet/mellanox/mlx4/en_port.c
index dae1a1f4ae55..c2cfb05e7290 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_port.c
@@ -148,10 +148,16 @@ int mlx4_en_DUMP_ETH_STATS(struct mlx4_en_dev *mdev, u8 port, u8 reset)
 	stats->tx_packets = 0;
 	stats->tx_bytes = 0;
 	priv->port_stats.tx_chksum_offload = 0;
+	priv->port_stats.queue_stopped = 0;
+	priv->port_stats.wake_queue = 0;
+
 	for (i = 0; i < priv->tx_ring_num; i++) {
 		stats->tx_packets += priv->tx_ring[i]->packets;
 		stats->tx_bytes += priv->tx_ring[i]->bytes;
 		priv->port_stats.tx_chksum_offload += priv->tx_ring[i]->tx_csum;
+		priv->port_stats.queue_stopped +=
+			priv->tx_ring[i]->queue_stopped;
+		priv->port_stats.wake_queue += priv->tx_ring[i]->wake_queue;
 	}
 
 	stats->rx_errors = be64_to_cpu(mlx4_en_stats->PCS) +
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 890922c1c8ee..ba049ae88749 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -318,6 +318,31 @@ static void mlx4_en_free_rx_buf(struct mlx4_en_priv *priv,
 	}
 }
 
+void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev)
+{
+	int i;
+	int num_of_eqs;
+	int num_rx_rings;
+	struct mlx4_dev *dev = mdev->dev;
+
+	mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) {
+		if (!dev->caps.comp_pool)
+			num_of_eqs = max_t(int, MIN_RX_RINGS,
+					   min_t(int,
+						 dev->caps.num_comp_vectors,
+						 DEF_RX_RINGS));
+		else
+			num_of_eqs = min_t(int, MAX_MSIX_P_PORT,
+					   dev->caps.comp_pool/
+					   dev->caps.num_ports) - 1;
+
+		num_rx_rings = min_t(int, num_of_eqs,
+				     netif_get_num_default_rss_queues());
+		mdev->profile.prof[i].rx_ring_num =
+			rounddown_pow_of_two(num_rx_rings);
+	}
+}
+
 int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_rx_ring **pring,
 			   u32 size, u16 stride, int node)
@@ -636,6 +661,9 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 	if (!priv->port_up)
 		return 0;
 
+	if (budget <= 0)
+		return polled;
+
 	/* We assume a 1:1 mapping between CQEs and Rx descriptors, so Rx
 	 * descriptor offset can be deduced from the CQE index instead of
 	 * reading 'cqe->index' */
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_selftest.c b/drivers/net/ethernet/mellanox/mlx4/en_selftest.c
index c11d063473e5..03e5f6ac67e7 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_selftest.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_selftest.c
@@ -129,8 +129,10 @@ static int mlx4_en_test_speed(struct mlx4_en_priv *priv)
 	if (mlx4_en_QUERY_PORT(priv->mdev, priv->port))
 		return -ENOMEM;
 
-	/* The device currently only supports 10G speed */
-	if (priv->port_state.link_speed != SPEED_10000)
+	/* The device supports 1G, 10G and 40G speeds */
+	if (priv->port_state.link_speed != 1000 &&
+	    priv->port_state.link_speed != 10000 &&
+	    priv->port_state.link_speed != 40000)
 		return priv->port_state.link_speed;
 	return 0;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 13457032d15f..dd1f6d346459 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -44,16 +44,6 @@
 
 #include "mlx4_en.h"
 
-enum {
-	MAX_INLINE = 104, /* 128 - 16 - 4 - 4 */
-	MAX_BF = 256,
-};
-
-static int inline_thold __read_mostly = MAX_INLINE;
-
-module_param_named(inline_thold, inline_thold, int, 0444);
-MODULE_PARM_DESC(inline_thold, "threshold for using inline data");
-
 int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_tx_ring **pring, int qpn, u32 size,
 			   u16 stride, int node, int queue_index)
@@ -75,8 +65,7 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
 	ring->size = size;
 	ring->size_mask = size - 1;
 	ring->stride = stride;
-
-	inline_thold = min(inline_thold, MAX_INLINE);
+	ring->inline_thold = priv->prof->inline_thold;
 
 	tmp = size * sizeof(struct mlx4_en_tx_info);
 	ring->tx_info = vmalloc_node(tmp, node);
@@ -325,7 +314,7 @@ static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
 			}
 		}
 	}
-	dev_kfree_skb(skb);
+	dev_kfree_skb_any(skb);
 	return tx_info->nr_txbb;
 }
 
@@ -456,7 +445,7 @@ static int mlx4_en_process_tx_cq(struct net_device *dev,
 	 */
 	if (netif_tx_queue_stopped(ring->tx_queue) && txbbs_skipped > 0) {
 		netif_tx_wake_queue(ring->tx_queue);
-		priv->port_stats.wake_queue++;
+		ring->wake_queue++;
 	}
 	return done;
 }
@@ -520,7 +509,7 @@ static struct mlx4_en_tx_desc *mlx4_en_bounce_to_desc(struct mlx4_en_priv *priv,
 	return ring->buf + index * TXBB_SIZE;
 }
 
-static int is_inline(struct sk_buff *skb, void **pfrag)
+static int is_inline(int inline_thold, struct sk_buff *skb, void **pfrag)
 {
 	void *ptr;
 
@@ -580,7 +569,7 @@ static int get_real_size(struct sk_buff *skb, struct net_device *dev,
 		}
 	} else {
 		*lso_header_size = 0;
-		if (!is_inline(skb, NULL))
+		if (!is_inline(priv->prof->inline_thold, skb, NULL))
 			real_size = CTRL_SIZE + (skb_shinfo(skb)->nr_frags + 1) * DS_SIZE;
 		else
 			real_size = inline_size(skb);
@@ -596,7 +585,13 @@ static void build_inline_wqe(struct mlx4_en_tx_desc *tx_desc, struct sk_buff *sk
 	int spc = MLX4_INLINE_ALIGN - CTRL_SIZE - sizeof *inl;
 
 	if (skb->len <= spc) {
-		inl->byte_count = cpu_to_be32(1 << 31 | skb->len);
+		if (likely(skb->len >= MIN_PKT_LEN)) {
+			inl->byte_count = cpu_to_be32(1 << 31 | skb->len);
+		} else {
+			inl->byte_count = cpu_to_be32(1 << 31 | MIN_PKT_LEN);
+			memset(((void *)(inl + 1)) + skb->len, 0,
+			       MIN_PKT_LEN - skb->len);
+		}
 		skb_copy_from_linear_data(skb, inl + 1, skb_headlen(skb));
 		if (skb_shinfo(skb)->nr_frags)
 			memcpy(((void *)(inl + 1)) + skb_headlen(skb), fragptr,
@@ -696,7 +691,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 		     ring->size - HEADROOM - MAX_DESC_TXBBS)) {
 		/* every full Tx ring stops queue */
 		netif_tx_stop_queue(ring->tx_queue);
-		priv->port_stats.queue_stopped++;
+		ring->queue_stopped++;
 
 		/* If queue was emptied after the if, and before the
 		 * stop_queue - need to wake the queue, or else it will remain
@@ -709,7 +704,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 		if (unlikely(((int)(ring->prod - ring->cons)) <=
 			     ring->size - HEADROOM - MAX_DESC_TXBBS)) {
 			netif_tx_wake_queue(ring->tx_queue);
-			priv->port_stats.wake_queue++;
+			ring->wake_queue++;
 		} else {
 			return NETDEV_TX_BUSY;
 		}
@@ -747,11 +742,11 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	tx_info->data_offset = (void *)data - (void *)tx_desc;
 
 	tx_info->linear = (lso_header_size < skb_headlen(skb) &&
-			   !is_inline(skb, NULL)) ? 1 : 0;
+			   !is_inline(ring->inline_thold, skb, NULL)) ? 1 : 0;
 
 	data += skb_shinfo(skb)->nr_frags + tx_info->linear - 1;
 
-	if (is_inline(skb, &fragptr)) {
+	if (is_inline(ring->inline_thold, skb, &fragptr)) {
 		tx_info->inl = 1;
 	} else {
 		/* Map fragments */
@@ -881,7 +876,8 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	skb_tx_timestamp(skb);
 
 	if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && !vlan_tx_tag_present(skb)) {
-		*(__be32 *) (&tx_desc->ctrl.vlan_tag) |= cpu_to_be32(ring->doorbell_qpn);
+		tx_desc->ctrl.bf_qpn |= cpu_to_be32(ring->doorbell_qpn);
+
 		op_own |= htonl((bf_index & 0xffff) << 8);
 		/* Ensure new descirptor hits memory
 		* before setting ownership of this descriptor to HW */
diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c
index 8992b38578d5..d501a2b0fb79 100644
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -271,7 +271,10 @@ enum slave_port_state mlx4_get_slave_port_state(struct mlx4_dev *dev, int slave,
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_slave_state *s_state = priv->mfunc.master.slave_state;
-	if (slave >= dev->num_slaves || port > MLX4_MAX_PORTS) {
+	struct mlx4_active_ports actv_ports = mlx4_get_active_ports(dev, slave);
+
+	if (slave >= dev->num_slaves || port > dev->caps.num_ports ||
+	    port <= 0 || !test_bit(port - 1, actv_ports.ports)) {
 		pr_err("%s: Error: asking for slave:%d, port:%d\n",
 		       __func__, slave, port);
 		return SLAVE_PORT_DOWN;
@@ -285,8 +288,10 @@ static int mlx4_set_slave_port_state(struct mlx4_dev *dev, int slave, u8 port,
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_slave_state *s_state = priv->mfunc.master.slave_state;
+	struct mlx4_active_ports actv_ports = mlx4_get_active_ports(dev, slave);
 
-	if (slave >= dev->num_slaves || port > MLX4_MAX_PORTS || port == 0) {
+	if (slave >= dev->num_slaves || port > dev->caps.num_ports ||
+	    port <= 0 || !test_bit(port - 1, actv_ports.ports)) {
 		pr_err("%s: Error: asking for slave:%d, port:%d\n",
 		       __func__, slave, port);
 		return -1;
@@ -300,9 +305,13 @@ static void set_all_slave_state(struct mlx4_dev *dev, u8 port, int event)
 {
 	int i;
 	enum slave_port_gen_event gen_event;
+	struct mlx4_slaves_pport slaves_pport = mlx4_phys_to_slaves_pport(dev,
+									  port);
 
-	for (i = 0; i < dev->num_slaves; i++)
-		set_and_calc_slave_port_state(dev, i, port, event, &gen_event);
+	for (i = 0; i < dev->num_vfs + 1; i++)
+		if (test_bit(i, slaves_pport.slaves))
+			set_and_calc_slave_port_state(dev, i, port,
+						      event, &gen_event);
 }
 /**************************************************************************
 	The function get as input the new event to that port,
@@ -321,12 +330,14 @@ int set_and_calc_slave_port_state(struct mlx4_dev *dev, int slave,
 	struct mlx4_slave_state *ctx = NULL;
 	unsigned long flags;
 	int ret = -1;
+	struct mlx4_active_ports actv_ports = mlx4_get_active_ports(dev, slave);
 	enum slave_port_state cur_state =
 		mlx4_get_slave_port_state(dev, slave, port);
 
 	*gen_event = SLAVE_PORT_GEN_EVENT_NONE;
 
-	if (slave >= dev->num_slaves || port > MLX4_MAX_PORTS || port == 0) {
+	if (slave >= dev->num_slaves || port > dev->caps.num_ports ||
+	    port <= 0 || !test_bit(port - 1, actv_ports.ports)) {
 		pr_err("%s: Error: asking for slave:%d, port:%d\n",
 		       __func__, slave, port);
 		return ret;
@@ -542,15 +553,19 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 				       be64_to_cpu(eqe->event.cmd.out_param));
 			break;
 
-		case MLX4_EVENT_TYPE_PORT_CHANGE:
+		case MLX4_EVENT_TYPE_PORT_CHANGE: {
+			struct mlx4_slaves_pport slaves_port;
 			port = be32_to_cpu(eqe->event.port_change.port) >> 28;
+			slaves_port = mlx4_phys_to_slaves_pport(dev, port);
 			if (eqe->subtype == MLX4_PORT_CHANGE_SUBTYPE_DOWN) {
 				mlx4_dispatch_event(dev, MLX4_DEV_EVENT_PORT_DOWN,
 						    port);
 				mlx4_priv(dev)->sense.do_sense_port[port] = 1;
 				if (!mlx4_is_master(dev))
 					break;
-				for (i = 0; i < dev->num_slaves; i++) {
+				for (i = 0; i < dev->num_vfs + 1; i++) {
+					if (!test_bit(i, slaves_port.slaves))
+						continue;
 					if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH) {
 						if (i == mlx4_master_func_num(dev))
 							continue;
@@ -558,8 +573,13 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 							 " to slave: %d, port:%d\n",
 							 __func__, i, port);
 						s_info = &priv->mfunc.master.vf_oper[slave].vport[port].state;
-						if (IFLA_VF_LINK_STATE_AUTO == s_info->link_state)
+						if (IFLA_VF_LINK_STATE_AUTO == s_info->link_state) {
+							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));
 							mlx4_slave_event(dev, i, eqe);
+						}
 					} else {  /* IB port */
 						set_and_calc_slave_port_state(dev, i, port,
 									      MLX4_PORT_STATE_DEV_EVENT_PORT_DOWN,
@@ -580,12 +600,19 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 				if (!mlx4_is_master(dev))
 					break;
 				if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH)
-					for (i = 0; i < dev->num_slaves; i++) {
+					for (i = 0; i < dev->num_vfs + 1; i++) {
+						if (!test_bit(i, slaves_port.slaves))
+							continue;
 						if (i == mlx4_master_func_num(dev))
 							continue;
 						s_info = &priv->mfunc.master.vf_oper[slave].vport[port].state;
-						if (IFLA_VF_LINK_STATE_AUTO == s_info->link_state)
+						if (IFLA_VF_LINK_STATE_AUTO == s_info->link_state) {
+							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));
 							mlx4_slave_event(dev, i, eqe);
+						}
 					}
 				else /* IB port */
 					/* port-up event will be sent to a slave when the
@@ -594,6 +621,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 					set_all_slave_state(dev, port, MLX4_DEV_EVENT_PORT_UP);
 			}
 			break;
+		}
 
 		case MLX4_EVENT_TYPE_CQ_ERROR:
 			mlx4_warn(dev, "CQ %s on CQN %06x\n",
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 7e2995ecea6f..d16a4d118903 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -225,13 +225,25 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_FLAGS0_FORCE_PHY_WQE_GID 0x80
 
 	if (vhcr->op_modifier == 1) {
+		struct mlx4_active_ports actv_ports =
+			mlx4_get_active_ports(dev, slave);
+		int converted_port = mlx4_slave_convert_port(
+				dev, slave, vhcr->in_modifier);
+
+		if (converted_port < 0)
+			return -EINVAL;
+
+		vhcr->in_modifier = converted_port;
 		/* Set nic_info bit to mark new fields support */
 		field  = QUERY_FUNC_CAP_FLAGS1_NIC_INFO;
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FLAGS1_OFFSET);
 
-		field = vhcr->in_modifier; /* phys-port = logical-port */
+		/* phys-port = logical-port */
+		field = vhcr->in_modifier -
+			find_first_bit(actv_ports.ports, dev->caps.num_ports);
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
 
+		field = vhcr->in_modifier;
 		/* size is now the QP number */
 		size = dev->phys_caps.base_tunnel_sqpn + 8 * slave + field - 1;
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP0_TUNNEL);
@@ -249,12 +261,16 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 			 QUERY_FUNC_CAP_PHYS_PORT_ID);
 
 	} else if (vhcr->op_modifier == 0) {
+		struct mlx4_active_ports actv_ports =
+			mlx4_get_active_ports(dev, slave);
 		/* enable rdma and ethernet interfaces, and new quota locations */
 		field = (QUERY_FUNC_CAP_FLAG_ETH | QUERY_FUNC_CAP_FLAG_RDMA |
 			 QUERY_FUNC_CAP_FLAG_QUOTAS);
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FLAGS_OFFSET);
 
-		field = dev->caps.num_ports;
+		field = min(
+			bitmap_weight(actv_ports.ports, dev->caps.num_ports),
+			dev->caps.num_ports);
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_NUM_PORTS_OFFSET);
 
 		size = dev->caps.function_caps; /* set PF behaviours */
@@ -840,6 +856,10 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave,
 	int	err = 0;
 	u8	field;
 	u32	bmme_flags;
+	int	real_port;
+	int	slave_port;
+	int	first_port;
+	struct mlx4_active_ports actv_ports;
 
 	err = mlx4_cmd_box(dev, 0, outbox->dma, 0, 0, MLX4_CMD_QUERY_DEV_CAP,
 			   MLX4_CMD_TIME_CLASS_A, MLX4_CMD_NATIVE);
@@ -852,8 +872,26 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave,
 	MLX4_GET(flags, outbox->buf, QUERY_DEV_CAP_EXT_FLAGS_OFFSET);
 	flags |= MLX4_DEV_CAP_FLAG_PORT_MNG_CHG_EV;
 	flags &= ~MLX4_DEV_CAP_FLAG_MEM_WINDOW;
+	actv_ports = mlx4_get_active_ports(dev, slave);
+	first_port = find_first_bit(actv_ports.ports, dev->caps.num_ports);
+	for (slave_port = 0, real_port = first_port;
+	     real_port < first_port +
+	     bitmap_weight(actv_ports.ports, dev->caps.num_ports);
+	     ++real_port, ++slave_port) {
+		if (flags & (MLX4_DEV_CAP_FLAG_WOL_PORT1 << real_port))
+			flags |= MLX4_DEV_CAP_FLAG_WOL_PORT1 << slave_port;
+		else
+			flags &= ~(MLX4_DEV_CAP_FLAG_WOL_PORT1 << slave_port);
+	}
+	for (; slave_port < dev->caps.num_ports; ++slave_port)
+		flags &= ~(MLX4_DEV_CAP_FLAG_WOL_PORT1 << slave_port);
 	MLX4_PUT(outbox->buf, flags, QUERY_DEV_CAP_EXT_FLAGS_OFFSET);
 
+	MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_VL_PORT_OFFSET);
+	field &= ~0x0F;
+	field |= bitmap_weight(actv_ports.ports, dev->caps.num_ports) & 0x0F;
+	MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_VL_PORT_OFFSET);
+
 	/* For guests, disable timestamp */
 	MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_CQ_TS_SUPPORT_OFFSET);
 	field &= 0x7f;
@@ -903,12 +941,20 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 	u16 short_field;
 	int err;
 	int admin_link_state;
+	int port = mlx4_slave_convert_port(dev, slave,
+					   vhcr->in_modifier & 0xFF);
 
 #define MLX4_VF_PORT_NO_LINK_SENSE_MASK	0xE0
 #define MLX4_PORT_LINK_UP_MASK		0x80
 #define QUERY_PORT_CUR_MAX_PKEY_OFFSET	0x0c
 #define QUERY_PORT_CUR_MAX_GID_OFFSET	0x0e
 
+	if (port < 0)
+		return -EINVAL;
+
+	vhcr->in_modifier = (vhcr->in_modifier & ~0xFF) |
+			    (port & 0xFF);
+
 	err = mlx4_cmd_box(dev, 0, outbox->dma, vhcr->in_modifier, 0,
 			   MLX4_CMD_QUERY_PORT, MLX4_CMD_TIME_CLASS_B,
 			   MLX4_CMD_NATIVE);
@@ -935,7 +981,10 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 		MLX4_PUT(outbox->buf, port_type,
 			 QUERY_PORT_SUPPORTED_TYPE_OFFSET);
 
-		short_field = 1; /* slave max gids */
+		if (dev->caps.port_type[vhcr->in_modifier] == MLX4_PORT_TYPE_ETH)
+			short_field = mlx4_get_slave_num_gids(dev, slave, port);
+		else
+			short_field = 1; /* slave max gids */
 		MLX4_PUT(outbox->buf, short_field,
 			 QUERY_PORT_CUR_MAX_GID_OFFSET);
 
@@ -1585,9 +1634,12 @@ int mlx4_INIT_PORT_wrapper(struct mlx4_dev *dev, int slave,
 			   struct mlx4_cmd_info *cmd)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
-	int port = vhcr->in_modifier;
+	int port = mlx4_slave_convert_port(dev, slave, vhcr->in_modifier);
 	int err;
 
+	if (port < 0)
+		return -EINVAL;
+
 	if (priv->mfunc.master.slave_state[slave].init_port_mask & (1 << port))
 		return 0;
 
@@ -1677,9 +1729,12 @@ int mlx4_CLOSE_PORT_wrapper(struct mlx4_dev *dev, int slave,
 			    struct mlx4_cmd_info *cmd)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
-	int port = vhcr->in_modifier;
+	int port = mlx4_slave_convert_port(dev, slave, vhcr->in_modifier);
 	int err;
 
+	if (port < 0)
+		return -EINVAL;
+
 	if (!(priv->mfunc.master.slave_state[slave].init_port_mask &
 	    (1 << port)))
 		return 0;
@@ -1724,6 +1779,46 @@ int mlx4_CLOSE_HCA(struct mlx4_dev *dev, int panic)
 			MLX4_CMD_NATIVE);
 }
 
+struct mlx4_config_dev {
+	__be32	update_flags;
+	__be32	rsdv1[3];
+	__be16	vxlan_udp_dport;
+	__be16	rsvd2;
+};
+
+#define MLX4_VXLAN_UDP_DPORT (1 << 0)
+
+static int mlx4_CONFIG_DEV(struct mlx4_dev *dev, struct mlx4_config_dev *config_dev)
+{
+	int err;
+	struct mlx4_cmd_mailbox *mailbox;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	memcpy(mailbox->buf, config_dev, sizeof(*config_dev));
+
+	err = mlx4_cmd(dev, mailbox->dma, 0, 0, MLX4_CMD_CONFIG_DEV,
+		       MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE);
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
+
+int mlx4_config_vxlan_port(struct mlx4_dev *dev, __be16 udp_port)
+{
+	struct mlx4_config_dev config_dev;
+
+	memset(&config_dev, 0, sizeof(config_dev));
+	config_dev.update_flags    = cpu_to_be32(MLX4_VXLAN_UDP_DPORT);
+	config_dev.vxlan_udp_dport = udp_port;
+
+	return mlx4_CONFIG_DEV(dev, &config_dev);
+}
+EXPORT_SYMBOL_GPL(mlx4_config_vxlan_port);
+
+
 int mlx4_SET_ICM_SIZE(struct mlx4_dev *dev, u64 icm_size, u64 *aux_pages)
 {
 	int ret = mlx4_cmd_imm(dev, icm_size, aux_pages, 0, 0,
@@ -1891,7 +1986,8 @@ void mlx4_opreq_action(struct work_struct *work)
 			err = EINVAL;
 			break;
 		}
-		err = mlx4_cmd(dev, 0, ((u32) err | cpu_to_be32(token) << 16),
+		err = mlx4_cmd(dev, 0, ((u32) err |
+					(__force u32)cpu_to_be32(token) << 16),
 			       1, MLX4_CMD_GET_OP_REQ, MLX4_CMD_TIME_CLASS_A,
 			       MLX4_CMD_NATIVE);
 		if (err) {
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index d413e60071d4..f0ae95f66ceb 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -41,7 +41,6 @@
 #include <linux/slab.h>
 #include <linux/io-mapping.h>
 #include <linux/delay.h>
-#include <linux/netdevice.h>
 #include <linux/kmod.h>
 
 #include <linux/mlx4/device.h>
@@ -78,13 +77,17 @@ MODULE_PARM_DESC(msi_x, "attempt to use MSI-X if nonzero");
 
 #endif /* CONFIG_PCI_MSI */
 
-static int num_vfs;
-module_param(num_vfs, int, 0444);
-MODULE_PARM_DESC(num_vfs, "enable #num_vfs functions if num_vfs > 0");
+static uint8_t num_vfs[3] = {0, 0, 0};
+static int num_vfs_argc = 3;
+module_param_array(num_vfs, byte , &num_vfs_argc, 0444);
+MODULE_PARM_DESC(num_vfs, "enable #num_vfs functions if num_vfs > 0\n"
+			  "num_vfs=port1,port2,port1+2");
 
-static int probe_vf;
-module_param(probe_vf, int, 0644);
-MODULE_PARM_DESC(probe_vf, "number of vfs to probe by pf driver (num_vfs > 0)");
+static uint8_t probe_vf[3] = {0, 0, 0};
+static int probe_vfs_argc = 3;
+module_param_array(probe_vf, byte, &probe_vfs_argc, 0444);
+MODULE_PARM_DESC(probe_vf, "number of vfs to probe by pf driver (num_vfs > 0)\n"
+			   "probe_vf=port1,port2,port1+2");
 
 int mlx4_log_num_mgm_entry_size = MLX4_DEFAULT_MGM_LOG_ENTRY_SIZE;
 module_param_named(log_num_mgm_entry_size,
@@ -1470,7 +1473,11 @@ static void mlx4_parav_master_pf_caps(struct mlx4_dev *dev)
 	int i;
 
 	for (i = 1; i <= dev->caps.num_ports; i++) {
-		dev->caps.gid_table_len[i] = 1;
+		if (dev->caps.port_type[i] == MLX4_PORT_TYPE_ETH)
+			dev->caps.gid_table_len[i] =
+				mlx4_get_slave_num_gids(dev, 0, i);
+		else
+			dev->caps.gid_table_len[i] = 1;
 		dev->caps.pkey_table_len[i] =
 			dev->phys_caps.pkey_phys_table_len[i] - 1;
 	}
@@ -1495,7 +1502,7 @@ static void choose_steering_mode(struct mlx4_dev *dev,
 	if (mlx4_log_num_mgm_entry_size == -1 &&
 	    dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_FS_EN &&
 	    (!mlx4_is_mfunc(dev) ||
-	     (dev_cap->fs_max_num_qp_per_entry >= (num_vfs + 1))) &&
+	     (dev_cap->fs_max_num_qp_per_entry >= (dev->num_vfs + 1))) &&
 	    choose_log_fs_mgm_entry_size(dev_cap->fs_max_num_qp_per_entry) >=
 		MLX4_MIN_MGM_LOG_ENTRY_SIZE) {
 		dev->oper_log_mgm_entry_size =
@@ -1981,9 +1988,8 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev)
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct msix_entry *entries;
 	int nreq = min_t(int, dev->caps.num_ports *
-			 min_t(int, netif_get_num_default_rss_queues() + 1,
+			 min_t(int, num_online_cpus() + 1,
 			       MAX_MSIX_P_PORT) + MSIX_LEGACY_SZ, MAX_MSIX);
-	int err;
 	int i;
 
 	if (msi_x) {
@@ -1997,23 +2003,13 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev)
 		for (i = 0; i < nreq; ++i)
 			entries[i].entry = i;
 
-	retry:
-		err = pci_enable_msix(dev->pdev, entries, nreq);
-		if (err) {
-			/* Try again if at least 2 vectors are available */
-			if (err > 1) {
-				mlx4_info(dev, "Requested %d vectors, "
-					  "but only %d MSI-X vectors available, "
-					  "trying again\n", nreq, err);
-				nreq = err;
-				goto retry;
-			}
+		nreq = pci_enable_msix_range(dev->pdev, entries, 2, nreq);
+
+		if (nreq < 0) {
 			kfree(entries);
 			goto no_msi;
-		}
-
-		if (nreq <
-		    MSIX_LEGACY_SZ + dev->caps.num_ports * MIN_MSIX_P_PORT) {
+		} else if (nreq < MSIX_LEGACY_SZ +
+				  dev->caps.num_ports * MIN_MSIX_P_PORT) {
 			/*Working in legacy mode , all EQ's shared*/
 			dev->caps.comp_pool           = 0;
 			dev->caps.num_comp_vectors = nreq - 1;
@@ -2201,6 +2197,13 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 	struct mlx4_dev *dev;
 	int err;
 	int port;
+	int nvfs[MLX4_MAX_PORTS + 1] = {0, 0, 0};
+	int prb_vf[MLX4_MAX_PORTS + 1] = {0, 0, 0};
+	const int param_map[MLX4_MAX_PORTS + 1][MLX4_MAX_PORTS + 1] = {
+		{2, 0, 0}, {0, 1, 2}, {0, 1, 2} };
+	unsigned total_vfs = 0;
+	int sriov_initialized = 0;
+	unsigned int i;
 
 	pr_info(DRV_NAME ": Initializing %s\n", pci_name(pdev));
 
@@ -2215,17 +2218,40 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 	 * per port, we must limit the number of VFs to 63 (since their are
 	 * 128 MACs)
 	 */
-	if (num_vfs >= MLX4_MAX_NUM_VF) {
+	for (i = 0; i < sizeof(nvfs)/sizeof(nvfs[0]) && i < num_vfs_argc;
+	     total_vfs += nvfs[param_map[num_vfs_argc - 1][i]], i++) {
+		nvfs[param_map[num_vfs_argc - 1][i]] = num_vfs[i];
+		if (nvfs[i] < 0) {
+			dev_err(&pdev->dev, "num_vfs module parameter cannot be negative\n");
+			return -EINVAL;
+		}
+	}
+	for (i = 0; i < sizeof(prb_vf)/sizeof(prb_vf[0]) && i < probe_vfs_argc;
+	     i++) {
+		prb_vf[param_map[probe_vfs_argc - 1][i]] = probe_vf[i];
+		if (prb_vf[i] < 0 || prb_vf[i] > nvfs[i]) {
+			dev_err(&pdev->dev, "probe_vf module parameter cannot be negative or greater than num_vfs\n");
+			return -EINVAL;
+		}
+	}
+	if (total_vfs >= MLX4_MAX_NUM_VF) {
 		dev_err(&pdev->dev,
 			"Requested more VF's (%d) than allowed (%d)\n",
-			num_vfs, MLX4_MAX_NUM_VF - 1);
+			total_vfs, MLX4_MAX_NUM_VF - 1);
 		return -EINVAL;
 	}
 
-	if (num_vfs < 0) {
-		pr_err("num_vfs module parameter cannot be negative\n");
-		return -EINVAL;
+	for (i = 0; i < MLX4_MAX_PORTS; i++) {
+		if (nvfs[i] + nvfs[2] >= MLX4_MAX_NUM_VF_P_PORT) {
+			dev_err(&pdev->dev,
+				"Requested more VF's (%d) for port (%d) than allowed (%d)\n",
+				nvfs[i] + nvfs[2], i + 1,
+				MLX4_MAX_NUM_VF_P_PORT - 1);
+			return -EINVAL;
+		}
 	}
+
+
 	/*
 	 * Check for BARs.
 	 */
@@ -2300,11 +2326,23 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 	if (pci_dev_data & MLX4_PCI_DEV_IS_VF) {
 		/* When acting as pf, we normally skip vfs unless explicitly
 		 * requested to probe them. */
-		if (num_vfs && extended_func_num(pdev) > probe_vf) {
-			mlx4_warn(dev, "Skipping virtual function:%d\n",
-						extended_func_num(pdev));
-			err = -ENODEV;
-			goto err_free_dev;
+		if (total_vfs) {
+			unsigned vfs_offset = 0;
+			for (i = 0; i < sizeof(nvfs)/sizeof(nvfs[0]) &&
+			     vfs_offset + nvfs[i] < extended_func_num(pdev);
+			     vfs_offset += nvfs[i], i++)
+				;
+			if (i == sizeof(nvfs)/sizeof(nvfs[0])) {
+				err = -ENODEV;
+				goto err_free_dev;
+			}
+			if ((extended_func_num(pdev) - vfs_offset)
+			    > prb_vf[i]) {
+				mlx4_warn(dev, "Skipping virtual function:%d\n",
+					  extended_func_num(pdev));
+				err = -ENODEV;
+				goto err_free_dev;
+			}
 		}
 		mlx4_warn(dev, "Detected virtual function - running in slave mode\n");
 		dev->flags |= MLX4_FLAG_SLAVE;
@@ -2324,22 +2362,30 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 			}
 		}
 
-		if (num_vfs) {
-			mlx4_warn(dev, "Enabling SR-IOV with %d VFs\n", num_vfs);
-
-			atomic_inc(&pf_loading);
-			err = pci_enable_sriov(pdev, num_vfs);
-			atomic_dec(&pf_loading);
-
-			if (err) {
-				mlx4_err(dev, "Failed to enable SR-IOV, continuing without SR-IOV (err = %d).\n",
-					 err);
+		if (total_vfs) {
+			mlx4_warn(dev, "Enabling SR-IOV with %d VFs\n",
+				  total_vfs);
+			dev->dev_vfs = kzalloc(
+					total_vfs * sizeof(*dev->dev_vfs),
+					GFP_KERNEL);
+			if (NULL == dev->dev_vfs) {
+				mlx4_err(dev, "Failed to allocate memory for VFs\n");
 				err = 0;
 			} else {
-				mlx4_warn(dev, "Running in master mode\n");
-				dev->flags |= MLX4_FLAG_SRIOV |
-					      MLX4_FLAG_MASTER;
-				dev->num_vfs = num_vfs;
+				atomic_inc(&pf_loading);
+				err = pci_enable_sriov(pdev, total_vfs);
+				atomic_dec(&pf_loading);
+				if (err) {
+					mlx4_err(dev, "Failed to enable SR-IOV, continuing without SR-IOV (err = %d).\n",
+						 err);
+					err = 0;
+				} else {
+					mlx4_warn(dev, "Running in master mode\n");
+					dev->flags |= MLX4_FLAG_SRIOV |
+						      MLX4_FLAG_MASTER;
+					dev->num_vfs = total_vfs;
+					sriov_initialized = 1;
+				}
 			}
 		}
 
@@ -2404,12 +2450,37 @@ slave_start:
 	/* In master functions, the communication channel must be initialized
 	 * after obtaining its address from fw */
 	if (mlx4_is_master(dev)) {
+		unsigned sum = 0;
 		err = mlx4_multi_func_init(dev);
 		if (err) {
 			mlx4_err(dev, "Failed to init master mfunc"
 				 "interface, aborting.\n");
 			goto err_close;
 		}
+		if (sriov_initialized) {
+			int ib_ports = 0;
+			mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB)
+				ib_ports++;
+
+			if (ib_ports &&
+			    (num_vfs_argc > 1 || probe_vfs_argc > 1)) {
+				mlx4_err(dev,
+					 "Invalid syntax of num_vfs/probe_vfs "
+					 "with IB port. Single port VFs syntax"
+					 " is only supported when all ports "
+					 "are configured as ethernet\n");
+				goto err_close;
+			}
+			for (i = 0; i < sizeof(nvfs)/sizeof(nvfs[0]); i++) {
+				unsigned j;
+				for (j = 0; j < nvfs[i]; ++sum, ++j) {
+					dev->dev_vfs[sum].min_port =
+						i < 2 ? i + 1 : 1;
+					dev->dev_vfs[sum].n_ports = i < 2 ? 1 :
+						dev->caps.num_ports;
+				}
+			}
+		}
 	}
 
 	err = mlx4_alloc_eq_table(dev);
@@ -2517,6 +2588,8 @@ err_rel_own:
 	if (!mlx4_is_slave(dev))
 		mlx4_free_ownership(dev);
 
+	kfree(priv->dev.dev_vfs);
+
 err_free_dev:
 	kfree(priv);
 
@@ -2603,6 +2676,7 @@ static void mlx4_remove_one(struct pci_dev *pdev)
 		kfree(dev->caps.qp0_proxy);
 		kfree(dev->caps.qp1_tunnel);
 		kfree(dev->caps.qp1_proxy);
+		kfree(dev->dev_vfs);
 
 		kfree(priv);
 		pci_release_regions(pdev);
diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c
index db7dc0b6667d..80ccb4edf825 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mcg.c
+++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c
@@ -1387,9 +1387,12 @@ int mlx4_PROMISC_wrapper(struct mlx4_dev *dev, int slave,
 			 struct mlx4_cmd_info *cmd)
 {
 	u32 qpn = (u32) vhcr->in_param & 0xffffffff;
-	u8 port = vhcr->in_param >> 62;
+	int port = mlx4_slave_convert_port(dev, slave, vhcr->in_param >> 62);
 	enum mlx4_steer_type steer = vhcr->in_modifier;
 
+	if (port < 0)
+		return -EINVAL;
+
 	/* Promiscuous unicast is not allowed in mfunc */
 	if (mlx4_is_mfunc(dev) && steer == MLX4_UC_STEER)
 		return 0;
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index 7aec6c833973..cf8be41abb36 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -788,6 +788,10 @@ enum {
 	MLX4_USE_RR	= 1,
 };
 
+struct mlx4_roce_gid_entry {
+	u8 raw[16];
+};
+
 struct mlx4_priv {
 	struct mlx4_dev		dev;
 
@@ -834,6 +838,7 @@ struct mlx4_priv {
 	int			fs_hash_mode;
 	u8 virt2phys_pkey[MLX4_MFUNC_MAX][MLX4_MAX_PORTS][MLX4_MAX_PORT_PKEYS];
 	__be64			slave_node_guids[MLX4_MFUNC_MAX];
+	struct mlx4_roce_gid_entry roce_gids[MLX4_MAX_PORTS][MLX4_ROCE_MAX_GIDS];
 
 	atomic_t		opreq_count;
 	struct work_struct	opreq_task;
@@ -1242,11 +1247,6 @@ int mlx4_QP_FLOW_STEERING_DETACH_wrapper(struct mlx4_dev *dev, int slave,
 					 struct mlx4_cmd_mailbox *inbox,
 					 struct mlx4_cmd_mailbox *outbox,
 					 struct mlx4_cmd_info *cmd);
-int mlx4_FLOW_STEERING_IB_UC_QP_RANGE_wrapper(struct mlx4_dev *dev, int slave,
-					      struct mlx4_vhcr *vhcr,
-					      struct mlx4_cmd_mailbox *inbox,
-					      struct mlx4_cmd_mailbox *outbox,
-					      struct mlx4_cmd_info *cmd);
 
 int mlx4_get_mgm_entry_size(struct mlx4_dev *dev);
 int mlx4_get_qp_per_mgm(struct mlx4_dev *dev);
@@ -1282,4 +1282,8 @@ void mlx4_vf_immed_vlan_work_handler(struct work_struct *_work);
 
 void mlx4_init_quotas(struct mlx4_dev *dev);
 
+int mlx4_get_slave_num_gids(struct mlx4_dev *dev, int slave, int port);
+/* Returns the VF index of slave */
+int mlx4_get_vf_indx(struct mlx4_dev *dev, int slave);
+
 #endif /* MLX4_H */
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index b57e8c87a34e..7a733c287744 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -187,6 +187,13 @@ enum {
 #define GET_AVG_PERF_COUNTER(cnt)	(0)
 #endif /* MLX4_EN_PERF_STAT */
 
+/* Constants for TX flow */
+enum {
+	MAX_INLINE = 104, /* 128 - 16 - 4 - 4 */
+	MAX_BF = 256,
+	MIN_PKT_LEN = 17,
+};
+
 /*
  * Configurables
  */
@@ -267,10 +274,13 @@ struct mlx4_en_tx_ring {
 	unsigned long bytes;
 	unsigned long packets;
 	unsigned long tx_csum;
+	unsigned long queue_stopped;
+	unsigned long wake_queue;
 	struct mlx4_bf bf;
 	bool bf_enabled;
 	struct netdev_queue *tx_queue;
 	int hwtstamp_tx_type;
+	int inline_thold;
 };
 
 struct mlx4_en_rx_desc {
@@ -346,6 +356,7 @@ struct mlx4_en_port_profile {
 	u8 tx_pause;
 	u8 tx_ppp;
 	int rss_rings;
+	int inline_thold;
 };
 
 struct mlx4_en_profile {
@@ -548,6 +559,10 @@ struct mlx4_en_priv {
 	struct work_struct linkstate_task;
 	struct delayed_work stats_task;
 	struct delayed_work service_task;
+#ifdef CONFIG_MLX4_EN_VXLAN
+	struct work_struct vxlan_add_task;
+	struct work_struct vxlan_del_task;
+#endif
 	struct mlx4_en_perf_stats pstats;
 	struct mlx4_en_pkt_stats pkstats;
 	struct mlx4_en_port_stats port_stats;
@@ -574,6 +589,7 @@ struct mlx4_en_priv {
 	struct hlist_head filter_hash[1 << MLX4_EN_FILTER_HASH_SHIFT];
 #endif
 	u64 tunnel_reg_id;
+	__be16 vxlan_port;
 };
 
 enum mlx4_en_wol {
@@ -737,7 +753,7 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
 			     int cq, int user_prio);
 void mlx4_en_deactivate_tx_ring(struct mlx4_en_priv *priv,
 				struct mlx4_en_tx_ring *ring);
-
+void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev);
 int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_rx_ring **pring,
 			   u32 size, u16 stride, int node);
@@ -786,7 +802,6 @@ void mlx4_en_cleanup_filters(struct mlx4_en_priv *priv);
 
 #define MLX4_EN_NUM_SELF_TEST	5
 void mlx4_en_ex_selftest(struct net_device *dev, u32 *flags, u64 *buf);
-u64 mlx4_en_mac_to_u64(u8 *addr);
 void mlx4_en_ptp_overflow_check(struct mlx4_en_dev *mdev);
 
 /*
diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c
index a58bcbf1b806..cfcad26ed40f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/port.c
@@ -505,6 +505,84 @@ int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps)
 	mlx4_free_cmd_mailbox(dev, outmailbox);
 	return err;
 }
+static struct mlx4_roce_gid_entry zgid_entry;
+
+int mlx4_get_slave_num_gids(struct mlx4_dev *dev, int slave, int port)
+{
+	int vfs;
+	int slave_gid = slave;
+	unsigned i;
+	struct mlx4_slaves_pport slaves_pport;
+	struct mlx4_active_ports actv_ports;
+	unsigned max_port_p_one;
+
+	if (slave == 0)
+		return MLX4_ROCE_PF_GIDS;
+
+	/* Slave is a VF */
+	slaves_pport = mlx4_phys_to_slaves_pport(dev, port);
+	actv_ports = mlx4_get_active_ports(dev, slave);
+	max_port_p_one = find_first_bit(actv_ports.ports, dev->caps.num_ports) +
+		bitmap_weight(actv_ports.ports, dev->caps.num_ports) + 1;
+
+	for (i = 1; i < max_port_p_one; i++) {
+		struct mlx4_active_ports exclusive_ports;
+		struct mlx4_slaves_pport slaves_pport_actv;
+		bitmap_zero(exclusive_ports.ports, dev->caps.num_ports);
+		set_bit(i - 1, exclusive_ports.ports);
+		if (i == port)
+			continue;
+		slaves_pport_actv = mlx4_phys_to_slaves_pport_actv(
+				    dev, &exclusive_ports);
+		slave_gid -= bitmap_weight(slaves_pport_actv.slaves,
+					   dev->num_vfs + 1);
+	}
+	vfs = bitmap_weight(slaves_pport.slaves, dev->num_vfs + 1) - 1;
+	if (slave_gid <= ((MLX4_ROCE_MAX_GIDS - MLX4_ROCE_PF_GIDS) % vfs))
+		return ((MLX4_ROCE_MAX_GIDS - MLX4_ROCE_PF_GIDS) / vfs) + 1;
+	return (MLX4_ROCE_MAX_GIDS - MLX4_ROCE_PF_GIDS) / vfs;
+}
+
+int mlx4_get_base_gid_ix(struct mlx4_dev *dev, int slave, int port)
+{
+	int gids;
+	unsigned i;
+	int slave_gid = slave;
+	int vfs;
+
+	struct mlx4_slaves_pport slaves_pport;
+	struct mlx4_active_ports actv_ports;
+	unsigned max_port_p_one;
+
+	if (slave == 0)
+		return 0;
+
+	slaves_pport = mlx4_phys_to_slaves_pport(dev, port);
+	actv_ports = mlx4_get_active_ports(dev, slave);
+	max_port_p_one = find_first_bit(actv_ports.ports, dev->caps.num_ports) +
+		bitmap_weight(actv_ports.ports, dev->caps.num_ports) + 1;
+
+	for (i = 1; i < max_port_p_one; i++) {
+		struct mlx4_active_ports exclusive_ports;
+		struct mlx4_slaves_pport slaves_pport_actv;
+		bitmap_zero(exclusive_ports.ports, dev->caps.num_ports);
+		set_bit(i - 1, exclusive_ports.ports);
+		if (i == port)
+			continue;
+		slaves_pport_actv = mlx4_phys_to_slaves_pport_actv(
+				    dev, &exclusive_ports);
+		slave_gid -= bitmap_weight(slaves_pport_actv.slaves,
+					   dev->num_vfs + 1);
+	}
+	gids = MLX4_ROCE_MAX_GIDS - MLX4_ROCE_PF_GIDS;
+	vfs = bitmap_weight(slaves_pport.slaves, dev->num_vfs + 1) - 1;
+	if (slave_gid <= gids % vfs)
+		return MLX4_ROCE_PF_GIDS + ((gids / vfs) + 1) * (slave_gid - 1);
+
+	return MLX4_ROCE_PF_GIDS + (gids % vfs) +
+		((gids / vfs) * (slave_gid - 1));
+}
+EXPORT_SYMBOL_GPL(mlx4_get_base_gid_ix);
 
 static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod,
 				u8 op_mod, struct mlx4_cmd_mailbox *inbox)
@@ -515,14 +593,18 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod,
 	struct mlx4_slave_state *slave_st = &master->slave_state[slave];
 	struct mlx4_set_port_rqp_calc_context *qpn_context;
 	struct mlx4_set_port_general_context *gen_context;
+	struct mlx4_roce_gid_entry *gid_entry_tbl, *gid_entry_mbox, *gid_entry_mb1;
 	int reset_qkey_viols;
 	int port;
 	int is_eth;
+	int num_gids;
+	int base;
 	u32 in_modifier;
 	u32 promisc;
 	u16 mtu, prev_mtu;
 	int err;
-	int i;
+	int i, j;
+	int offset;
 	__be32 agg_cap_mask;
 	__be32 slave_cap_mask;
 	__be32 new_cap_mask;
@@ -535,7 +617,8 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod,
 	/* Slaves cannot perform SET_PORT operations except changing MTU */
 	if (is_eth) {
 		if (slave != dev->caps.function &&
-		    in_modifier != MLX4_SET_PORT_GENERAL) {
+		    in_modifier != MLX4_SET_PORT_GENERAL &&
+		    in_modifier != MLX4_SET_PORT_GID_TABLE) {
 			mlx4_warn(dev, "denying SET_PORT for slave:%d\n",
 					slave);
 			return -EINVAL;
@@ -581,6 +664,67 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod,
 
 			gen_context->mtu = cpu_to_be16(master->max_mtu[port]);
 			break;
+		case MLX4_SET_PORT_GID_TABLE:
+			/* change to MULTIPLE entries: number of guest's gids
+			 * need a FOR-loop here over number of gids the guest has.
+			 * 1. Check no duplicates in gids passed by slave
+			 */
+			num_gids = mlx4_get_slave_num_gids(dev, slave, port);
+			base = mlx4_get_base_gid_ix(dev, slave, port);
+			gid_entry_mbox = (struct mlx4_roce_gid_entry *)(inbox->buf);
+			for (i = 0; i < num_gids; gid_entry_mbox++, i++) {
+				if (!memcmp(gid_entry_mbox->raw, zgid_entry.raw,
+					    sizeof(zgid_entry)))
+					continue;
+				gid_entry_mb1 = gid_entry_mbox + 1;
+				for (j = i + 1; j < num_gids; gid_entry_mb1++, j++) {
+					if (!memcmp(gid_entry_mb1->raw,
+						    zgid_entry.raw, sizeof(zgid_entry)))
+						continue;
+					if (!memcmp(gid_entry_mb1->raw, gid_entry_mbox->raw,
+						    sizeof(gid_entry_mbox->raw))) {
+						/* found duplicate */
+						return -EINVAL;
+					}
+				}
+			}
+
+			/* 2. Check that do not have duplicates in OTHER
+			 *    entries in the port GID table
+			 */
+			for (i = 0; i < MLX4_ROCE_MAX_GIDS; i++) {
+				if (i >= base && i < base + num_gids)
+					continue; /* don't compare to slave's current gids */
+				gid_entry_tbl = &priv->roce_gids[port - 1][i];
+				if (!memcmp(gid_entry_tbl->raw, zgid_entry.raw, sizeof(zgid_entry)))
+					continue;
+				gid_entry_mbox = (struct mlx4_roce_gid_entry *)(inbox->buf);
+				for (j = 0; j < num_gids; gid_entry_mbox++, j++) {
+					if (!memcmp(gid_entry_mbox->raw, zgid_entry.raw,
+						    sizeof(zgid_entry)))
+						continue;
+					if (!memcmp(gid_entry_mbox->raw, gid_entry_tbl->raw,
+						    sizeof(gid_entry_tbl->raw))) {
+						/* found duplicate */
+						mlx4_warn(dev, "requested gid entry for slave:%d "
+							  "is a duplicate of gid at index %d\n",
+							  slave, i);
+						return -EINVAL;
+					}
+				}
+			}
+
+			/* insert slave GIDs with memcpy, starting at slave's base index */
+			gid_entry_mbox = (struct mlx4_roce_gid_entry *)(inbox->buf);
+			for (i = 0, offset = base; i < num_gids; gid_entry_mbox++, offset++, i++)
+				memcpy(priv->roce_gids[port - 1][offset].raw, gid_entry_mbox->raw, 16);
+
+			/* Now, copy roce port gids table to current mailbox for passing to FW */
+			gid_entry_mbox = (struct mlx4_roce_gid_entry *)(inbox->buf);
+			for (i = 0; i < MLX4_ROCE_MAX_GIDS; gid_entry_mbox++, i++)
+				memcpy(gid_entry_mbox->raw, priv->roce_gids[port - 1][i].raw, 16);
+
+			break;
 		}
 		return mlx4_cmd(dev, inbox->dma, in_mod, op_mod,
 				MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B,
@@ -646,6 +790,15 @@ int mlx4_SET_PORT_wrapper(struct mlx4_dev *dev, int slave,
 			  struct mlx4_cmd_mailbox *outbox,
 			  struct mlx4_cmd_info *cmd)
 {
+	int port = mlx4_slave_convert_port(
+			dev, slave, vhcr->in_modifier & 0xFF);
+
+	if (port < 0)
+		return -EINVAL;
+
+	vhcr->in_modifier = (vhcr->in_modifier & ~0xFF) |
+			    (port & 0xFF);
+
 	return mlx4_common_set_port(dev, slave, vhcr->in_modifier,
 				    vhcr->op_modifier, inbox);
 }
@@ -835,7 +988,7 @@ struct mlx4_set_port_vxlan_context {
 	u8	steering;
 };
 
-int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering)
+int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering, int enable)
 {
 	int err;
 	u32 in_mod;
@@ -849,7 +1002,8 @@ int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering)
 	memset(context, 0, sizeof(*context));
 
 	context->modify_flags = VXLAN_ENABLE_MODIFY | VXLAN_STEERING_MODIFY;
-	context->enable_flags = VXLAN_ENABLE;
+	if (enable)
+		context->enable_flags = VXLAN_ENABLE;
 	context->steering  = steering;
 
 	in_mod = MLX4_SET_PORT_VXLAN << 8 | port;
@@ -927,3 +1081,108 @@ void mlx4_set_stats_bitmap(struct mlx4_dev *dev, u64 *stats_bitmap)
 		*stats_bitmap |= MLX4_STATS_ERROR_COUNTERS_MASK;
 }
 EXPORT_SYMBOL(mlx4_set_stats_bitmap);
+
+int mlx4_get_slave_from_roce_gid(struct mlx4_dev *dev, int port, u8 *gid,
+				 int *slave_id)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	int i, found_ix = -1;
+	int vf_gids = MLX4_ROCE_MAX_GIDS - MLX4_ROCE_PF_GIDS;
+	struct mlx4_slaves_pport slaves_pport;
+	unsigned num_vfs;
+	int slave_gid;
+
+	if (!mlx4_is_mfunc(dev))
+		return -EINVAL;
+
+	slaves_pport = mlx4_phys_to_slaves_pport(dev, port);
+	num_vfs = bitmap_weight(slaves_pport.slaves, dev->num_vfs + 1) - 1;
+
+	for (i = 0; i < MLX4_ROCE_MAX_GIDS; i++) {
+		if (!memcmp(priv->roce_gids[port - 1][i].raw, gid, 16)) {
+			found_ix = i;
+			break;
+		}
+	}
+
+	if (found_ix >= 0) {
+		if (found_ix < MLX4_ROCE_PF_GIDS)
+			slave_gid = 0;
+		else if (found_ix < MLX4_ROCE_PF_GIDS + (vf_gids % num_vfs) *
+			 (vf_gids / num_vfs + 1))
+			slave_gid = ((found_ix - MLX4_ROCE_PF_GIDS) /
+				     (vf_gids / num_vfs + 1)) + 1;
+		else
+			slave_gid =
+			((found_ix - MLX4_ROCE_PF_GIDS -
+			  ((vf_gids % num_vfs) * ((vf_gids / num_vfs + 1)))) /
+			 (vf_gids / num_vfs)) + vf_gids % num_vfs + 1;
+
+		if (slave_gid) {
+			struct mlx4_active_ports exclusive_ports;
+			struct mlx4_active_ports actv_ports;
+			struct mlx4_slaves_pport slaves_pport_actv;
+			unsigned max_port_p_one;
+			int num_slaves_before = 1;
+
+			for (i = 1; i < port; i++) {
+				bitmap_zero(exclusive_ports.ports, dev->caps.num_ports);
+				set_bit(i, exclusive_ports.ports);
+				slaves_pport_actv =
+					mlx4_phys_to_slaves_pport_actv(
+							dev, &exclusive_ports);
+				num_slaves_before += bitmap_weight(
+						slaves_pport_actv.slaves,
+						dev->num_vfs + 1);
+			}
+
+			if (slave_gid < num_slaves_before) {
+				bitmap_zero(exclusive_ports.ports, dev->caps.num_ports);
+				set_bit(port - 1, exclusive_ports.ports);
+				slaves_pport_actv =
+					mlx4_phys_to_slaves_pport_actv(
+							dev, &exclusive_ports);
+				slave_gid += bitmap_weight(
+						slaves_pport_actv.slaves,
+						dev->num_vfs + 1) -
+						num_slaves_before;
+			}
+			actv_ports = mlx4_get_active_ports(dev, slave_gid);
+			max_port_p_one = find_first_bit(
+				actv_ports.ports, dev->caps.num_ports) +
+				bitmap_weight(actv_ports.ports,
+					      dev->caps.num_ports) + 1;
+
+			for (i = 1; i < max_port_p_one; i++) {
+				if (i == port)
+					continue;
+				bitmap_zero(exclusive_ports.ports,
+					    dev->caps.num_ports);
+				set_bit(i - 1, exclusive_ports.ports);
+				slaves_pport_actv =
+					mlx4_phys_to_slaves_pport_actv(
+						dev, &exclusive_ports);
+				slave_gid += bitmap_weight(
+						slaves_pport_actv.slaves,
+						dev->num_vfs + 1);
+			}
+		}
+		*slave_id = slave_gid;
+	}
+
+	return (found_ix >= 0) ? 0 : -EINVAL;
+}
+EXPORT_SYMBOL(mlx4_get_slave_from_roce_gid);
+
+int mlx4_get_roce_gid_from_slave(struct mlx4_dev *dev, int port, int slave_id,
+				 u8 *gid)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+
+	if (!mlx4_is_master(dev))
+		return -EINVAL;
+
+	memcpy(gid, priv->roce_gids[port - 1][slave_id].raw, 16);
+	return 0;
+}
+EXPORT_SYMBOL(mlx4_get_roce_gid_from_slave);
diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index 57428a0cb9dd..3b5f53ef29b2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -52,6 +52,8 @@
 struct mac_res {
 	struct list_head list;
 	u64 mac;
+	int ref_count;
+	u8 smac_index;
 	u8 port;
 };
 
@@ -219,6 +221,11 @@ struct res_fs_rule {
 	int			qpn;
 };
 
+static int mlx4_is_eth(struct mlx4_dev *dev, int port)
+{
+	return dev->caps.port_mask[port] == MLX4_PORT_TYPE_IB ? 0 : 1;
+}
+
 static void *res_tracker_lookup(struct rb_root *root, u64 res_id)
 {
 	struct rb_node *node = root->rb_node;
@@ -461,6 +468,8 @@ int mlx4_init_resource_tracker(struct mlx4_dev *dev)
 
 		spin_lock_init(&res_alloc->alloc_lock);
 		for (t = 0; t < dev->num_vfs + 1; t++) {
+			struct mlx4_active_ports actv_ports =
+				mlx4_get_active_ports(dev, t);
 			switch (i) {
 			case RES_QP:
 				initialize_res_quotas(dev, res_alloc, RES_QP,
@@ -490,10 +499,27 @@ int mlx4_init_resource_tracker(struct mlx4_dev *dev)
 				break;
 			case RES_MAC:
 				if (t == mlx4_master_func_num(dev)) {
-					res_alloc->quota[t] = MLX4_MAX_MAC_NUM;
+					int max_vfs_pport = 0;
+					/* Calculate the max vfs per port for */
+					/* both ports.			      */
+					for (j = 0; j < dev->caps.num_ports;
+					     j++) {
+						struct mlx4_slaves_pport slaves_pport =
+							mlx4_phys_to_slaves_pport(dev, j + 1);
+						unsigned current_slaves =
+							bitmap_weight(slaves_pport.slaves,
+								      dev->caps.num_ports) - 1;
+						if (max_vfs_pport < current_slaves)
+							max_vfs_pport =
+								current_slaves;
+					}
+					res_alloc->quota[t] =
+						MLX4_MAX_MAC_NUM -
+						2 * max_vfs_pport;
 					res_alloc->guaranteed[t] = 2;
 					for (j = 0; j < MLX4_MAX_PORTS; j++)
-						res_alloc->res_port_free[j] = MLX4_MAX_MAC_NUM;
+						res_alloc->res_port_free[j] =
+							MLX4_MAX_MAC_NUM;
 				} else {
 					res_alloc->quota[t] = MLX4_MAX_MAC_NUM;
 					res_alloc->guaranteed[t] = 2;
@@ -521,9 +547,10 @@ int mlx4_init_resource_tracker(struct mlx4_dev *dev)
 				break;
 			}
 			if (i == RES_MAC || i == RES_VLAN) {
-				for (j = 0; j < MLX4_MAX_PORTS; j++)
-					res_alloc->res_port_rsvd[j] +=
-						res_alloc->guaranteed[t];
+				for (j = 0; j < dev->caps.num_ports; j++)
+					if (test_bit(j, actv_ports.ports))
+						res_alloc->res_port_rsvd[j] +=
+							res_alloc->guaranteed[t];
 			} else {
 				res_alloc->res_reserved += res_alloc->guaranteed[t];
 			}
@@ -600,15 +627,37 @@ static void update_gid(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *inbox,
 	struct mlx4_qp_context	*qp_ctx = inbox->buf + 8;
 	enum mlx4_qp_optpar	optpar = be32_to_cpu(*(__be32 *) inbox->buf);
 	u32			ts = (be32_to_cpu(qp_ctx->flags) >> 16) & 0xff;
+	int port;
 
-	if (MLX4_QP_ST_UD == ts)
-		qp_ctx->pri_path.mgid_index = 0x80 | slave;
-
-	if (MLX4_QP_ST_RC == ts || MLX4_QP_ST_UC == ts) {
-		if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH)
-			qp_ctx->pri_path.mgid_index = slave & 0x7F;
-		if (optpar & MLX4_QP_OPTPAR_ALT_ADDR_PATH)
-			qp_ctx->alt_path.mgid_index = slave & 0x7F;
+	if (MLX4_QP_ST_UD == ts) {
+		port = (qp_ctx->pri_path.sched_queue >> 6 & 1) + 1;
+		if (mlx4_is_eth(dev, port))
+			qp_ctx->pri_path.mgid_index =
+				mlx4_get_base_gid_ix(dev, slave, port) | 0x80;
+		else
+			qp_ctx->pri_path.mgid_index = slave | 0x80;
+
+	} else if (MLX4_QP_ST_RC == ts || MLX4_QP_ST_XRC == ts || MLX4_QP_ST_UC == ts) {
+		if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH) {
+			port = (qp_ctx->pri_path.sched_queue >> 6 & 1) + 1;
+			if (mlx4_is_eth(dev, port)) {
+				qp_ctx->pri_path.mgid_index +=
+					mlx4_get_base_gid_ix(dev, slave, port);
+				qp_ctx->pri_path.mgid_index &= 0x7f;
+			} else {
+				qp_ctx->pri_path.mgid_index = slave & 0x7F;
+			}
+		}
+		if (optpar & MLX4_QP_OPTPAR_ALT_ADDR_PATH) {
+			port = (qp_ctx->alt_path.sched_queue >> 6 & 1) + 1;
+			if (mlx4_is_eth(dev, port)) {
+				qp_ctx->alt_path.mgid_index +=
+					mlx4_get_base_gid_ix(dev, slave, port);
+				qp_ctx->alt_path.mgid_index &= 0x7f;
+			} else {
+				qp_ctx->alt_path.mgid_index = slave & 0x7F;
+			}
+		}
 	}
 }
 
@@ -619,7 +668,6 @@ static int update_vport_qp_param(struct mlx4_dev *dev,
 	struct mlx4_qp_context	*qpc = inbox->buf + 8;
 	struct mlx4_vport_oper_state *vp_oper;
 	struct mlx4_priv *priv;
-	u32 qp_type;
 	int port;
 
 	port = (qpc->pri_path.sched_queue & 0x40) ? 2 : 1;
@@ -627,12 +675,6 @@ static int update_vport_qp_param(struct mlx4_dev *dev,
 	vp_oper = &priv->mfunc.master.vf_oper[slave].vport[port];
 
 	if (MLX4_VGT != vp_oper->state.default_vlan) {
-		qp_type	= (be32_to_cpu(qpc->flags) >> 16) & 0xff;
-		if (MLX4_QP_ST_RC == qp_type ||
-		    (MLX4_QP_ST_UD == qp_type &&
-		     !mlx4_is_qp_reserved(dev, qpn)))
-			return -EINVAL;
-
 		/* the reserved QPs (special, proxy, tunnel)
 		 * do not operate over vlans
 		 */
@@ -1659,11 +1701,39 @@ static int srq_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 	return err;
 }
 
-static int mac_add_to_slave(struct mlx4_dev *dev, int slave, u64 mac, int port)
+static int mac_find_smac_ix_in_slave(struct mlx4_dev *dev, int slave, int port,
+				     u8 smac_index, u64 *mac)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
+	struct list_head *mac_list =
+		&tracker->slave_list[slave].res_list[RES_MAC];
+	struct mac_res *res, *tmp;
+
+	list_for_each_entry_safe(res, tmp, mac_list, list) {
+		if (res->smac_index == smac_index && res->port == (u8) port) {
+			*mac = res->mac;
+			return 0;
+		}
+	}
+	return -ENOENT;
+}
+
+static int mac_add_to_slave(struct mlx4_dev *dev, int slave, u64 mac, int port, u8 smac_index)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
-	struct mac_res *res;
+	struct list_head *mac_list =
+		&tracker->slave_list[slave].res_list[RES_MAC];
+	struct mac_res *res, *tmp;
+
+	list_for_each_entry_safe(res, tmp, mac_list, list) {
+		if (res->mac == mac && res->port == (u8) port) {
+			/* mac found. update ref count */
+			++res->ref_count;
+			return 0;
+		}
+	}
 
 	if (mlx4_grant_resource(dev, slave, RES_MAC, 1, port))
 		return -EINVAL;
@@ -1674,6 +1744,8 @@ static int mac_add_to_slave(struct mlx4_dev *dev, int slave, u64 mac, int port)
 	}
 	res->mac = mac;
 	res->port = (u8) port;
+	res->smac_index = smac_index;
+	res->ref_count = 1;
 	list_add_tail(&res->list,
 		      &tracker->slave_list[slave].res_list[RES_MAC]);
 	return 0;
@@ -1690,9 +1762,11 @@ static void mac_del_from_slave(struct mlx4_dev *dev, int slave, u64 mac,
 
 	list_for_each_entry_safe(res, tmp, mac_list, list) {
 		if (res->mac == mac && res->port == (u8) port) {
-			list_del(&res->list);
-			mlx4_release_resource(dev, slave, RES_MAC, 1, port);
-			kfree(res);
+			if (!--res->ref_count) {
+				list_del(&res->list);
+				mlx4_release_resource(dev, slave, RES_MAC, 1, port);
+				kfree(res);
+			}
 			break;
 		}
 	}
@@ -1705,10 +1779,13 @@ static void rem_slave_macs(struct mlx4_dev *dev, int slave)
 	struct list_head *mac_list =
 		&tracker->slave_list[slave].res_list[RES_MAC];
 	struct mac_res *res, *tmp;
+	int i;
 
 	list_for_each_entry_safe(res, tmp, mac_list, list) {
 		list_del(&res->list);
-		__mlx4_unregister_mac(dev, res->port, res->mac);
+		/* dereference the mac the num times the slave referenced it */
+		for (i = 0; i < res->ref_count; i++)
+			__mlx4_unregister_mac(dev, res->port, res->mac);
 		mlx4_release_resource(dev, slave, RES_MAC, 1, res->port);
 		kfree(res);
 	}
@@ -1720,21 +1797,28 @@ static int mac_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 	int err = -EINVAL;
 	int port;
 	u64 mac;
+	u8 smac_index;
 
 	if (op != RES_OP_RESERVE_AND_MAP)
 		return err;
 
 	port = !in_port ? get_param_l(out_param) : in_port;
+	port = mlx4_slave_convert_port(
+			dev, slave, port);
+
+	if (port < 0)
+		return -EINVAL;
 	mac = in_param;
 
 	err = __mlx4_register_mac(dev, port, mac);
 	if (err >= 0) {
+		smac_index = err;
 		set_param_l(out_param, err);
 		err = 0;
 	}
 
 	if (!err) {
-		err = mac_add_to_slave(dev, slave, mac, port);
+		err = mac_add_to_slave(dev, slave, mac, port, smac_index);
 		if (err)
 			__mlx4_unregister_mac(dev, port, mac);
 	}
@@ -1831,6 +1915,11 @@ static int vlan_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 	if (!port || op != RES_OP_RESERVE_AND_MAP)
 		return -EINVAL;
 
+	port = mlx4_slave_convert_port(
+			dev, slave, port);
+
+	if (port < 0)
+		return -EINVAL;
 	/* upstream kernels had NOP for reg/unreg vlan. Continue this. */
 	if (!in_port && port > 0 && port <= dev->caps.num_ports) {
 		slave_state[slave].old_vlan_api = true;
@@ -2128,6 +2217,11 @@ static int mac_free_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 	switch (op) {
 	case RES_OP_RESERVE_AND_MAP:
 		port = !in_port ? get_param_l(out_param) : in_port;
+		port = mlx4_slave_convert_port(
+				dev, slave, port);
+
+		if (port < 0)
+			return -EINVAL;
 		mac_del_from_slave(dev, slave, in_param, port);
 		__mlx4_unregister_mac(dev, port, in_param);
 		break;
@@ -2147,6 +2241,11 @@ static int vlan_free_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 	struct mlx4_slave_state *slave_state = priv->mfunc.master.slave_state;
 	int err = 0;
 
+	port = mlx4_slave_convert_port(
+			dev, slave, port);
+
+	if (port < 0)
+		return -EINVAL;
 	switch (op) {
 	case RES_OP_RESERVE_AND_MAP:
 		if (slave_state[slave].old_vlan_api)
@@ -2734,6 +2833,8 @@ static int verify_qp_parameters(struct mlx4_dev *dev,
 	u32			qp_type;
 	struct mlx4_qp_context	*qp_ctx;
 	enum mlx4_qp_optpar	optpar;
+	int port;
+	int num_gids;
 
 	qp_ctx  = inbox->buf + 8;
 	qp_type	= (be32_to_cpu(qp_ctx->flags) >> 16) & 0xff;
@@ -2741,6 +2842,7 @@ static int verify_qp_parameters(struct mlx4_dev *dev,
 
 	switch (qp_type) {
 	case MLX4_QP_ST_RC:
+	case MLX4_QP_ST_XRC:
 	case MLX4_QP_ST_UC:
 		switch (transition) {
 		case QP_TRANS_INIT2RTR:
@@ -2749,13 +2851,24 @@ static int verify_qp_parameters(struct mlx4_dev *dev,
 		case QP_TRANS_SQD2SQD:
 		case QP_TRANS_SQD2RTS:
 			if (slave != mlx4_master_func_num(dev))
-				/* slaves have only gid index 0 */
-				if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH)
-					if (qp_ctx->pri_path.mgid_index)
+				if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH) {
+					port = (qp_ctx->pri_path.sched_queue >> 6 & 1) + 1;
+					if (dev->caps.port_mask[port] != MLX4_PORT_TYPE_IB)
+						num_gids = mlx4_get_slave_num_gids(dev, slave, port);
+					else
+						num_gids = 1;
+					if (qp_ctx->pri_path.mgid_index >= num_gids)
 						return -EINVAL;
-				if (optpar & MLX4_QP_OPTPAR_ALT_ADDR_PATH)
-					if (qp_ctx->alt_path.mgid_index)
+				}
+				if (optpar & MLX4_QP_OPTPAR_ALT_ADDR_PATH) {
+					port = (qp_ctx->alt_path.sched_queue >> 6 & 1) + 1;
+					if (dev->caps.port_mask[port] != MLX4_PORT_TYPE_IB)
+						num_gids = mlx4_get_slave_num_gids(dev, slave, port);
+					else
+						num_gids = 1;
+					if (qp_ctx->alt_path.mgid_index >= num_gids)
 						return -EINVAL;
+				}
 			break;
 		default:
 			break;
@@ -3268,6 +3381,58 @@ int mlx4_INIT2INIT_QP_wrapper(struct mlx4_dev *dev, int slave,
 	return mlx4_GEN_QP_wrapper(dev, slave, vhcr, inbox, outbox, cmd);
 }
 
+static int adjust_qp_sched_queue(struct mlx4_dev *dev, int slave,
+				  struct mlx4_qp_context *qpc,
+				  struct mlx4_cmd_mailbox *inbox)
+{
+	enum mlx4_qp_optpar optpar = be32_to_cpu(*(__be32 *)inbox->buf);
+	u8 pri_sched_queue;
+	int port = mlx4_slave_convert_port(
+		   dev, slave, (qpc->pri_path.sched_queue >> 6 & 1) + 1) - 1;
+
+	if (port < 0)
+		return -EINVAL;
+
+	pri_sched_queue = (qpc->pri_path.sched_queue & ~(1 << 6)) |
+			  ((port & 1) << 6);
+
+	if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH ||
+	    mlx4_is_eth(dev, port + 1)) {
+		qpc->pri_path.sched_queue = pri_sched_queue;
+	}
+
+	if (optpar & MLX4_QP_OPTPAR_ALT_ADDR_PATH) {
+		port = mlx4_slave_convert_port(
+				dev, slave, (qpc->alt_path.sched_queue >> 6 & 1)
+				+ 1) - 1;
+		if (port < 0)
+			return -EINVAL;
+		qpc->alt_path.sched_queue =
+			(qpc->alt_path.sched_queue & ~(1 << 6)) |
+			(port & 1) << 6;
+	}
+	return 0;
+}
+
+static int roce_verify_mac(struct mlx4_dev *dev, int slave,
+				struct mlx4_qp_context *qpc,
+				struct mlx4_cmd_mailbox *inbox)
+{
+	u64 mac;
+	int port;
+	u32 ts = (be32_to_cpu(qpc->flags) >> 16) & 0xff;
+	u8 sched = *(u8 *)(inbox->buf + 64);
+	u8 smac_ix;
+
+	port = (sched >> 6 & 1) + 1;
+	if (mlx4_is_eth(dev, port) && (ts != MLX4_QP_ST_MLX)) {
+		smac_ix = qpc->pri_path.grh_mylmc & 0x7f;
+		if (mac_find_smac_ix_in_slave(dev, slave, port, smac_ix, &mac))
+			return -ENOENT;
+	}
+	return 0;
+}
+
 int mlx4_INIT2RTR_QP_wrapper(struct mlx4_dev *dev, int slave,
 			     struct mlx4_vhcr *vhcr,
 			     struct mlx4_cmd_mailbox *inbox,
@@ -3286,10 +3451,16 @@ int mlx4_INIT2RTR_QP_wrapper(struct mlx4_dev *dev, int slave,
 	u8 orig_vlan_index = qpc->pri_path.vlan_index;
 	u8 orig_feup = qpc->pri_path.feup;
 
+	err = adjust_qp_sched_queue(dev, slave, qpc, inbox);
+	if (err)
+		return err;
 	err = verify_qp_parameters(dev, inbox, QP_TRANS_INIT2RTR, slave);
 	if (err)
 		return err;
 
+	if (roce_verify_mac(dev, slave, qpc, inbox))
+		return -EINVAL;
+
 	update_pkey_index(dev, slave, inbox);
 	update_gid(dev, inbox, (u8)slave);
 	adjust_proxy_tun_qkey(dev, vhcr, qpc);
@@ -3334,6 +3505,9 @@ int mlx4_RTR2RTS_QP_wrapper(struct mlx4_dev *dev, int slave,
 	int err;
 	struct mlx4_qp_context *context = inbox->buf + 8;
 
+	err = adjust_qp_sched_queue(dev, slave, context, inbox);
+	if (err)
+		return err;
 	err = verify_qp_parameters(dev, inbox, QP_TRANS_RTR2RTS, slave);
 	if (err)
 		return err;
@@ -3353,6 +3527,9 @@ int mlx4_RTS2RTS_QP_wrapper(struct mlx4_dev *dev, int slave,
 	int err;
 	struct mlx4_qp_context *context = inbox->buf + 8;
 
+	err = adjust_qp_sched_queue(dev, slave, context, inbox);
+	if (err)
+		return err;
 	err = verify_qp_parameters(dev, inbox, QP_TRANS_RTS2RTS, slave);
 	if (err)
 		return err;
@@ -3371,6 +3548,9 @@ int mlx4_SQERR2RTS_QP_wrapper(struct mlx4_dev *dev, int slave,
 			      struct mlx4_cmd_info *cmd)
 {
 	struct mlx4_qp_context *context = inbox->buf + 8;
+	int err = adjust_qp_sched_queue(dev, slave, context, inbox);
+	if (err)
+		return err;
 	adjust_proxy_tun_qkey(dev, vhcr, context);
 	return mlx4_GEN_QP_wrapper(dev, slave, vhcr, inbox, outbox, cmd);
 }
@@ -3384,6 +3564,9 @@ int mlx4_SQD2SQD_QP_wrapper(struct mlx4_dev *dev, int slave,
 	int err;
 	struct mlx4_qp_context *context = inbox->buf + 8;
 
+	err = adjust_qp_sched_queue(dev, slave, context, inbox);
+	if (err)
+		return err;
 	err = verify_qp_parameters(dev, inbox, QP_TRANS_SQD2SQD, slave);
 	if (err)
 		return err;
@@ -3403,6 +3586,9 @@ int mlx4_SQD2RTS_QP_wrapper(struct mlx4_dev *dev, int slave,
 	int err;
 	struct mlx4_qp_context *context = inbox->buf + 8;
 
+	err = adjust_qp_sched_queue(dev, slave, context, inbox);
+	if (err)
+		return err;
 	err = verify_qp_parameters(dev, inbox, QP_TRANS_SQD2RTS, slave);
 	if (err)
 		return err;
@@ -3506,16 +3692,26 @@ static int rem_mcg_res(struct mlx4_dev *dev, int slave, struct res_qp *rqp,
 	return err;
 }
 
-static int qp_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
-		     int block_loopback, enum mlx4_protocol prot,
+static int qp_attach(struct mlx4_dev *dev, int slave, struct mlx4_qp *qp,
+		     u8 gid[16], int block_loopback, enum mlx4_protocol prot,
 		     enum mlx4_steer_type type, u64 *reg_id)
 {
 	switch (dev->caps.steering_mode) {
-	case MLX4_STEERING_MODE_DEVICE_MANAGED:
-		return mlx4_trans_to_dmfs_attach(dev, qp, gid, gid[5],
+	case MLX4_STEERING_MODE_DEVICE_MANAGED: {
+		int port = mlx4_slave_convert_port(dev, slave, gid[5]);
+		if (port < 0)
+			return port;
+		return mlx4_trans_to_dmfs_attach(dev, qp, gid, port,
 						block_loopback, prot,
 						reg_id);
+	}
 	case MLX4_STEERING_MODE_B0:
+		if (prot == MLX4_PROT_ETH) {
+			int port = mlx4_slave_convert_port(dev, slave, gid[5]);
+			if (port < 0)
+				return port;
+			gid[5] = port;
+		}
 		return mlx4_qp_attach_common(dev, qp, gid,
 					    block_loopback, prot, type);
 	default:
@@ -3523,9 +3719,9 @@ static int qp_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	}
 }
 
-static int qp_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
-		     enum mlx4_protocol prot, enum mlx4_steer_type type,
-		     u64 reg_id)
+static int qp_detach(struct mlx4_dev *dev, struct mlx4_qp *qp,
+		     u8 gid[16], enum mlx4_protocol prot,
+		     enum mlx4_steer_type type, u64 reg_id)
 {
 	switch (dev->caps.steering_mode) {
 	case MLX4_STEERING_MODE_DEVICE_MANAGED:
@@ -3562,7 +3758,7 @@ int mlx4_QP_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 
 	qp.qpn = qpn;
 	if (attach) {
-		err = qp_attach(dev, &qp, gid, block_loopback, prot,
+		err = qp_attach(dev, slave, &qp, gid, block_loopback, prot,
 				type, &reg_id);
 		if (err) {
 			pr_err("Fail to attach rule to qp 0x%x\n", qpn);
@@ -3698,6 +3894,9 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 		return -EOPNOTSUPP;
 
 	ctrl = (struct mlx4_net_trans_rule_hw_ctrl *)inbox->buf;
+	ctrl->port = mlx4_slave_convert_port(dev, slave, ctrl->port);
+	if (ctrl->port <= 0)
+		return -EINVAL;
 	qpn = be32_to_cpu(ctrl->qpn) & 0xffffff;
 	err = get_res(dev, slave, qpn, RES_QP, &rqp);
 	if (err) {
@@ -3816,16 +4015,6 @@ int mlx4_QUERY_IF_STAT_wrapper(struct mlx4_dev *dev, int slave,
 	return err;
 }
 
-int mlx4_FLOW_STEERING_IB_UC_QP_RANGE_wrapper(struct mlx4_dev *dev, int slave,
-					      struct mlx4_vhcr *vhcr,
-					      struct mlx4_cmd_mailbox *inbox,
-					      struct mlx4_cmd_mailbox *outbox,
-					      struct mlx4_cmd_info *cmd)
-{
-	return -EPERM;
-}
-
-
 static void detach_qp(struct mlx4_dev *dev, int slave, struct res_qp *rqp)
 {
 	struct res_gid *rgid;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 23b7e2d35a93..77ac95f052da 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -116,7 +116,6 @@ static int mlx5_enable_msix(struct mlx5_core_dev *dev)
 	struct mlx5_eq_table *table = &dev->priv.eq_table;
 	int num_eqs = 1 << dev->caps.log_max_eq;
 	int nvec;
-	int err;
 	int i;
 
 	nvec = dev->caps.num_ports * num_online_cpus() + MLX5_EQ_VEC_COMP_BASE;
@@ -131,17 +130,12 @@ static int mlx5_enable_msix(struct mlx5_core_dev *dev)
 	for (i = 0; i < nvec; i++)
 		table->msix_arr[i].entry = i;
 
-retry:
-	table->num_comp_vectors = nvec - MLX5_EQ_VEC_COMP_BASE;
-	err = pci_enable_msix(dev->pdev, table->msix_arr, nvec);
-	if (err <= 0) {
-		return err;
-	} else if (err > 2) {
-		nvec = err;
-		goto retry;
-	}
+	nvec = pci_enable_msix_range(dev->pdev, table->msix_arr,
+				     MLX5_EQ_VEC_COMP_BASE, nvec);
+	if (nvec < 0)
+		return nvec;
 
-	mlx5_core_dbg(dev, "received %d MSI vectors out of %d requested\n", err, nvec);
+	table->num_comp_vectors = nvec - MLX5_EQ_VEC_COMP_BASE;
 
 	return 0;
 }
@@ -537,7 +531,6 @@ static int __init init(void)
 
 	return 0;
 
-	mlx5_health_cleanup();
 err_debug:
 	mlx5_unregister_debugfs();
 	return err;