Merge branch 'mlx5-flow-steering'

Saeed Mahameed says:

====================
mlx5 improved flow steering management

First two patches fixes some minor issues in recently
introduced SRIOV code.

The other seven patches modifies the driver's code that
manages flow steering rules with Connectx-4 devices.

Basic introduction:

The flow steering device specification model is composed of the following entities:

Destination (either a TIR/Flow table/vport), where TIR is RSS end-point, vport
is the VF eSwitch port in SRIOV.

Flow table entry (FTE) - the values used by the flow specification
Flow table group (FG) - the masks used by the flow specification
Flow table (FT) - groups several FGs and can serve as destination

The flow steering software entities:

In addition to the device objects, the software have two more objects:

Priorities - group several FTs. Handles order of packet matching.

Namespaces - group several priorities. Namespace are used in order to
isolate different usages of steering (for example, add two separate
namespaces, one for the NIC driver and one for E-Switch FDB).

The base data structure for the flow steering management is a tree and
all the flow steering objects such as (Namespace/Flow table/Flow Group/FTE/etc.)
are represented as a node in the tree, e.g.:
Priority-0 -> FT1 -> FG -> FTE -> TIR (destination)
Priority-1 -> FT2 -> FG->  FTE -> TIR (destination)

Matching begins in FT1 flow rules and if there is a miss on all the FTEs
then matching continues on the FTEs in FT2.

The new implementation solves/improves the following
issues in the current code:

1) The new impl. supports multiple destinations, the search for existing rule with
   the same matching value is performed by the flow steering management.
   In the current impl. the E-switch FDB management code needs to search
   for existing rules before calling to the add rule function.

2) The new impl. manages the flow table level, in the current implementation the
   consumer states the flow table level when new flow table is created without
   any knowledge about the levels of other flow tables.

3) In the current impl. the consumer can't create or destroy flow
   groups dynamically, the flow groups are passed as argument to the create
   flow table API. The new impl. exposes API for create/destroy flow group.

The series is built as follows:

Patch #1 add flow steering API firmware commands.

Patch #2 add tree operation of the flow steering tree: add/remove node,
initialize node and take reference count on a node.

Patch #3 add essential algorithms for managing the flow steering.

Patch #4 Initialize the flow steering tree, flow steering initialization is based
on static tree which illustrates the flow steering tree when the driver is loaded.

Patch #5 is the main patch of the series. It introduce the flow steering API.

Patch #6 Expose the new flow steering API and remove the old one.
The Ethernet flow steering follows the existing implementation,
but uses the new steering API.

Patch #7 Rename en_flow_table.c to en_fs.c in order to be aligned with
the new flow steering files.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

17 files changed, 2185 insertions, 1051 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index a0755919ccaf..fe11e967095f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -2,7 +2,7 @@ obj-$(CONFIG_MLX5_CORE)		+= mlx5_core.o
 
 mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 		health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o   \
-		mad.o transobj.o vport.o sriov.o
-mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o eswitch.o \
-		en_main.o en_flow_table.o en_ethtool.o en_tx.o en_rx.o \
+		mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o
+mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \
+		en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \
 		en_txrx.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 89313d46952d..f689ce580b44 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -64,6 +64,8 @@
 #define MLX5E_UPDATE_STATS_INTERVAL    200 /* msecs */
 #define MLX5E_SQ_BF_BUDGET             16
 
+#define MLX5E_NUM_MAIN_GROUPS 9
+
 static const char vport_strings[][ETH_GSTRING_LEN] = {
 	/* vport statistics */
 	"rx_packets",
@@ -442,7 +444,7 @@ enum mlx5e_rqt_ix {
 struct mlx5e_eth_addr_info {
 	u8  addr[ETH_ALEN + 2];
 	u32 tt_vec;
-	u32 ft_ix[MLX5E_NUM_TT]; /* flow table index per traffic type */
+	struct mlx5_flow_rule *ft_rule[MLX5E_NUM_TT];
 };
 
 #define MLX5E_ETH_ADDR_HASH_SIZE (1 << BITS_PER_BYTE)
@@ -466,15 +468,22 @@ enum {
 
 struct mlx5e_vlan_db {
 	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
-	u32           active_vlans_ft_ix[VLAN_N_VID];
-	u32           untagged_rule_ft_ix;
-	u32           any_vlan_rule_ft_ix;
+	struct mlx5_flow_rule	*active_vlans_rule[VLAN_N_VID];
+	struct mlx5_flow_rule	*untagged_rule;
+	struct mlx5_flow_rule	*any_vlan_rule;
 	bool          filter_disabled;
 };
 
 struct mlx5e_flow_table {
-	void *vlan;
-	void *main;
+	int num_groups;
+	struct mlx5_flow_table		*t;
+	struct mlx5_flow_group		**g;
+};
+
+struct mlx5e_flow_tables {
+	struct mlx5_flow_namespace	*ns;
+	struct mlx5e_flow_table		vlan;
+	struct mlx5e_flow_table		main;
 };
 
 struct mlx5e_priv {
@@ -497,7 +506,7 @@ struct mlx5e_priv {
 	u32                        rqtn[MLX5E_NUM_RQT];
 	u32                        tirn[MLX5E_NUM_TT];
 
-	struct mlx5e_flow_table    ft;
+	struct mlx5e_flow_tables   fts;
 	struct mlx5e_eth_addr_db   eth_addr;
 	struct mlx5e_vlan_db       vlan;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
index 5b93c9c6e341..80d81abc4820 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -34,9 +34,11 @@
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 #include <linux/tcp.h>
-#include <linux/mlx5/flow_table.h>
+#include <linux/mlx5/fs.h>
 #include "en.h"
 
+#define MLX5_SET_CFG(p, f, v) MLX5_SET(create_flow_group_in, p, f, v)
+
 enum {
 	MLX5E_FULLMATCH = 0,
 	MLX5E_ALLMULTI  = 1,
@@ -103,44 +105,38 @@ static void mlx5e_del_eth_addr_from_hash(struct mlx5e_eth_addr_hash_node *hn)
 static void mlx5e_del_eth_addr_from_flow_table(struct mlx5e_priv *priv,
 					       struct mlx5e_eth_addr_info *ai)
 {
-	void *ft = priv->ft.main;
-
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV6_IPSEC_ESP]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_IPSEC_ESP]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV4_IPSEC_ESP]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_IPSEC_ESP]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV6_IPSEC_AH]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_IPSEC_AH]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH))
-		mlx5_del_flow_table_entry(ft,
-					  ai->ft_ix[MLX5E_TT_IPV4_IPSEC_AH]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_IPSEC_AH]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_TCP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_TCP]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_TCP]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_TCP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_TCP]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_TCP]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV6_UDP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_UDP]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_UDP]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV4_UDP))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_UDP]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_UDP]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV6))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_IPV4))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4]);
 
 	if (ai->tt_vec & BIT(MLX5E_TT_ANY))
-		mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_ANY]);
+		mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_ANY]);
 }
 
 static int mlx5e_get_eth_addr_type(u8 *addr)
@@ -240,44 +236,34 @@ static u32 mlx5e_get_tt_vec(struct mlx5e_eth_addr_info *ai, int type)
 }
 
 static int __mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv,
-				     struct mlx5e_eth_addr_info *ai, int type,
-				     void *flow_context, void *match_criteria)
+				     struct mlx5e_eth_addr_info *ai,
+				     int type, u32 *mc, u32 *mv)
 {
+	struct mlx5_flow_destination dest;
 	u8 match_criteria_enable = 0;
-	void *match_value;
-	void *dest;
-	u8   *dmac;
-	u8   *match_criteria_dmac;
-	void *ft   = priv->ft.main;
-	u32  *tirn = priv->tirn;
-	u32  *ft_ix;
-	u32  tt_vec;
-	int  err;
-
-	match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value);
-	dmac = MLX5_ADDR_OF(fte_match_param, match_value,
-			    outer_headers.dmac_47_16);
-	match_criteria_dmac = MLX5_ADDR_OF(fte_match_param, match_criteria,
-					   outer_headers.dmac_47_16);
-	dest = MLX5_ADDR_OF(flow_context, flow_context, destination);
-
-	MLX5_SET(flow_context, flow_context, action,
-		 MLX5_FLOW_CONTEXT_ACTION_FWD_DEST);
-	MLX5_SET(flow_context, flow_context, destination_list_size, 1);
-	MLX5_SET(dest_format_struct, dest, destination_type,
-		 MLX5_FLOW_CONTEXT_DEST_TYPE_TIR);
+	struct mlx5_flow_rule **rule_p;
+	struct mlx5_flow_table *ft = priv->fts.main.t;
+	u8 *mc_dmac = MLX5_ADDR_OF(fte_match_param, mc,
+				   outer_headers.dmac_47_16);
+	u8 *mv_dmac = MLX5_ADDR_OF(fte_match_param, mv,
+				   outer_headers.dmac_47_16);
+	u32 *tirn = priv->tirn;
+	u32 tt_vec;
+	int err = 0;
+
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR;
 
 	switch (type) {
 	case MLX5E_FULLMATCH:
 		match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-		memset(match_criteria_dmac, 0xff, ETH_ALEN);
-		ether_addr_copy(dmac, ai->addr);
+		eth_broadcast_addr(mc_dmac);
+		ether_addr_copy(mv_dmac, ai->addr);
 		break;
 
 	case MLX5E_ALLMULTI:
 		match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-		match_criteria_dmac[0] = 0x01;
-		dmac[0] = 0x01;
+		mc_dmac[0] = 0x01;
+		mv_dmac[0] = 0x01;
 		break;
 
 	case MLX5E_PROMISC:
@@ -286,190 +272,165 @@ static int __mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv,
 
 	tt_vec = mlx5e_get_tt_vec(ai, type);
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_ANY];
 	if (tt_vec & BIT(MLX5E_TT_ANY)) {
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_ANY]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		rule_p = &ai->ft_rule[MLX5E_TT_ANY];
+		dest.tir_num = tirn[MLX5E_TT_ANY];
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_ANY);
 	}
 
 	match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-			 outer_headers.ethertype);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4];
 	if (tt_vec & BIT(MLX5E_TT_IPV4)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4];
+		dest.tir_num = tirn[MLX5E_TT_IPV4];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV4);
 	}
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6];
 	if (tt_vec & BIT(MLX5E_TT_IPV6)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6];
+		dest.tir_num = tirn[MLX5E_TT_IPV6];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV6);
 	}
 
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-			 outer_headers.ip_protocol);
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_UDP);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_UDP);
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_UDP];
 	if (tt_vec & BIT(MLX5E_TT_IPV4_UDP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_UDP];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_UDP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_UDP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV4_UDP);
 	}
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_UDP];
 	if (tt_vec & BIT(MLX5E_TT_IPV6_UDP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_UDP];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_UDP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_UDP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV6_UDP);
 	}
 
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_TCP);
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_TCP);
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_TCP];
 	if (tt_vec & BIT(MLX5E_TT_IPV4_TCP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_TCP];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_TCP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_TCP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV4_TCP);
 	}
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_TCP];
 	if (tt_vec & BIT(MLX5E_TT_IPV6_TCP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_TCP];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_TCP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_TCP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
 
 		ai->tt_vec |= BIT(MLX5E_TT_IPV6_TCP);
 	}
 
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_AH);
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_AH);
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_IPSEC_AH];
 	if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_IPSEC_AH];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_IPSEC_AH];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_IPSEC_AH]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_AH);
 	}
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_IPSEC_AH];
 	if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_IPSEC_AH];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_IPSEC_AH];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_IPSEC_AH]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_AH);
 	}
 
-	MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol,
-		 IPPROTO_ESP);
+	MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_ESP);
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_IPSEC_ESP];
 	if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV4_IPSEC_ESP];
+		dest.tir_num = tirn[MLX5E_TT_IPV4_IPSEC_ESP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IP);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV4_IPSEC_ESP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_ESP);
 	}
 
-	ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_IPSEC_ESP];
 	if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP)) {
-		MLX5_SET(fte_match_param, match_value, outer_headers.ethertype,
+		rule_p = &ai->ft_rule[MLX5E_TT_IPV6_IPSEC_ESP];
+		dest.tir_num = tirn[MLX5E_TT_IPV6_IPSEC_ESP];
+		MLX5_SET(fte_match_param, mv, outer_headers.ethertype,
 			 ETH_P_IPV6);
-		MLX5_SET(dest_format_struct, dest, destination_id,
-			 tirn[MLX5E_TT_IPV6_IPSEC_ESP]);
-		err = mlx5_add_flow_table_entry(ft, match_criteria_enable,
-						match_criteria, flow_context,
-						ft_ix);
-		if (err)
+		*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+					     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+					     MLX5_FS_DEFAULT_FLOW_TAG, &dest);
+		if (IS_ERR_OR_NULL(*rule_p))
 			goto err_del_ai;
-
 		ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_ESP);
 	}
 
 	return 0;
 
 err_del_ai:
+	err = PTR_ERR(*rule_p);
+	*rule_p = NULL;
 	mlx5e_del_eth_addr_from_flow_table(priv, ai);
 
 	return err;
@@ -478,27 +439,25 @@ err_del_ai:
 static int mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv,
 				   struct mlx5e_eth_addr_info *ai, int type)
 {
-	u32 *flow_context;
 	u32 *match_criteria;
-	int err;
+	u32 *match_value;
+	int err = 0;
 
-	flow_context   = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) +
-				      MLX5_ST_SZ_BYTES(dest_format_struct));
-	match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
-	if (!flow_context || !match_criteria) {
+	match_value	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	match_criteria	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	if (!match_value || !match_criteria) {
 		netdev_err(priv->netdev, "%s: alloc failed\n", __func__);
 		err = -ENOMEM;
 		goto add_eth_addr_rule_out;
 	}
 
-	err = __mlx5e_add_eth_addr_rule(priv, ai, type, flow_context,
-					match_criteria);
-	if (err)
-		netdev_err(priv->netdev, "%s: failed\n", __func__);
+	err = __mlx5e_add_eth_addr_rule(priv, ai, type, match_criteria,
+					match_value);
 
 add_eth_addr_rule_out:
 	kvfree(match_criteria);
-	kvfree(flow_context);
+	kvfree(match_value);
+
 	return err;
 }
 
@@ -551,72 +510,77 @@ enum mlx5e_vlan_rule_type {
 	MLX5E_VLAN_RULE_TYPE_MATCH_VID,
 };
 
-static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv,
-			       enum mlx5e_vlan_rule_type rule_type, u16 vid)
+static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv,
+				 enum mlx5e_vlan_rule_type rule_type,
+				 u16 vid, u32 *mc, u32 *mv)
 {
+	struct mlx5_flow_table *ft = priv->fts.vlan.t;
+	struct mlx5_flow_destination dest;
 	u8 match_criteria_enable = 0;
-	u32 *flow_context;
-	void *match_value;
-	void *dest;
-	u32 *match_criteria;
-	u32 *ft_ix;
-	int err;
+	struct mlx5_flow_rule **rule_p;
+	int err = 0;
 
-	flow_context   = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) +
-				      MLX5_ST_SZ_BYTES(dest_format_struct));
-	match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
-	if (!flow_context || !match_criteria) {
-		netdev_err(priv->netdev, "%s: alloc failed\n", __func__);
-		err = -ENOMEM;
-		goto add_vlan_rule_out;
-	}
-	match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value);
-	dest = MLX5_ADDR_OF(flow_context, flow_context, destination);
-
-	MLX5_SET(flow_context, flow_context, action,
-		 MLX5_FLOW_CONTEXT_ACTION_FWD_DEST);
-	MLX5_SET(flow_context, flow_context, destination_list_size, 1);
-	MLX5_SET(dest_format_struct, dest, destination_type,
-		 MLX5_FLOW_CONTEXT_DEST_TYPE_FLOW_TABLE);
-	MLX5_SET(dest_format_struct, dest, destination_id,
-		 mlx5_get_flow_table_id(priv->ft.main));
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest.ft = priv->fts.main.t;
 
 	match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-			 outer_headers.vlan_tag);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag);
 
 	switch (rule_type) {
 	case MLX5E_VLAN_RULE_TYPE_UNTAGGED:
-		ft_ix = &priv->vlan.untagged_rule_ft_ix;
+		rule_p = &priv->vlan.untagged_rule;
 		break;
 	case MLX5E_VLAN_RULE_TYPE_ANY_VID:
-		ft_ix = &priv->vlan.any_vlan_rule_ft_ix;
-		MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag,
-			 1);
+		rule_p = &priv->vlan.any_vlan_rule;
+		MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1);
 		break;
 	default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */
-		err = mlx5e_vport_context_update_vlans(priv);
-		if (err)
-			goto add_vlan_rule_out;
-
-		ft_ix = &priv->vlan.active_vlans_ft_ix[vid];
-		MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag,
-			 1);
-		MLX5_SET_TO_ONES(fte_match_param, match_criteria,
-				 outer_headers.first_vid);
-		MLX5_SET(fte_match_param, match_value, outer_headers.first_vid,
-			 vid);
+		rule_p = &priv->vlan.active_vlans_rule[vid];
+		MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1);
+		MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid);
+		MLX5_SET(fte_match_param, mv, outer_headers.first_vid, vid);
 		break;
 	}
 
-	err = mlx5_add_flow_table_entry(priv->ft.vlan, match_criteria_enable,
-					match_criteria, flow_context, ft_ix);
-	if (err)
-		netdev_err(priv->netdev, "%s: failed\n", __func__);
+	*rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv,
+				     MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+				     MLX5_FS_DEFAULT_FLOW_TAG,
+				     &dest);
+
+	if (IS_ERR(*rule_p)) {
+		err = PTR_ERR(*rule_p);
+		*rule_p = NULL;
+		netdev_err(priv->netdev, "%s: add rule failed\n", __func__);
+	}
+
+	return err;
+}
+
+static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv,
+			       enum mlx5e_vlan_rule_type rule_type, u16 vid)
+{
+	u32 *match_criteria;
+	u32 *match_value;
+	int err = 0;
+
+	match_value	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	match_criteria	= mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param));
+	if (!match_value || !match_criteria) {
+		netdev_err(priv->netdev, "%s: alloc failed\n", __func__);
+		err = -ENOMEM;
+		goto add_vlan_rule_out;
+	}
+
+	if (rule_type == MLX5E_VLAN_RULE_TYPE_MATCH_VID)
+		mlx5e_vport_context_update_vlans(priv);
+
+	err = __mlx5e_add_vlan_rule(priv, rule_type, vid, match_criteria,
+				    match_value);
 
 add_vlan_rule_out:
 	kvfree(match_criteria);
-	kvfree(flow_context);
+	kvfree(match_value);
+
 	return err;
 }
 
@@ -625,16 +589,23 @@ static void mlx5e_del_vlan_rule(struct mlx5e_priv *priv,
 {
 	switch (rule_type) {
 	case MLX5E_VLAN_RULE_TYPE_UNTAGGED:
-		mlx5_del_flow_table_entry(priv->ft.vlan,
-					  priv->vlan.untagged_rule_ft_ix);
+		if (priv->vlan.untagged_rule) {
+			mlx5_del_flow_rule(priv->vlan.untagged_rule);
+			priv->vlan.untagged_rule = NULL;
+		}
 		break;
 	case MLX5E_VLAN_RULE_TYPE_ANY_VID:
-		mlx5_del_flow_table_entry(priv->ft.vlan,
-					  priv->vlan.any_vlan_rule_ft_ix);
+		if (priv->vlan.any_vlan_rule) {
+			mlx5_del_flow_rule(priv->vlan.any_vlan_rule);
+			priv->vlan.any_vlan_rule = NULL;
+		}
 		break;
 	case MLX5E_VLAN_RULE_TYPE_MATCH_VID:
-		mlx5_del_flow_table_entry(priv->ft.vlan,
-					  priv->vlan.active_vlans_ft_ix[vid]);
+		mlx5e_vport_context_update_vlans(priv);
+		if (priv->vlan.active_vlans_rule[vid]) {
+			mlx5_del_flow_rule(priv->vlan.active_vlans_rule[vid]);
+			priv->vlan.active_vlans_rule[vid] = NULL;
+		}
 		mlx5e_vport_context_update_vlans(priv);
 		break;
 	}
@@ -889,151 +860,358 @@ void mlx5e_set_rx_mode_work(struct work_struct *work)
 	mlx5e_vport_context_update(priv);
 }
 
+static void mlx5e_destroy_groups(struct mlx5e_flow_table *ft)
+{
+	int i;
+
+	for (i = ft->num_groups - 1; i >= 0; i--) {
+		if (!IS_ERR_OR_NULL(ft->g[i]))
+			mlx5_destroy_flow_group(ft->g[i]);
+		ft->g[i] = NULL;
+	}
+	ft->num_groups = 0;
+}
+
 void mlx5e_init_eth_addr(struct mlx5e_priv *priv)
 {
 	ether_addr_copy(priv->eth_addr.broadcast.addr, priv->netdev->broadcast);
 }
 
-static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv)
+#define MLX5E_MAIN_GROUP0_SIZE	BIT(3)
+#define MLX5E_MAIN_GROUP1_SIZE	BIT(1)
+#define MLX5E_MAIN_GROUP2_SIZE	BIT(0)
+#define MLX5E_MAIN_GROUP3_SIZE	BIT(14)
+#define MLX5E_MAIN_GROUP4_SIZE	BIT(13)
+#define MLX5E_MAIN_GROUP5_SIZE	BIT(11)
+#define MLX5E_MAIN_GROUP6_SIZE	BIT(2)
+#define MLX5E_MAIN_GROUP7_SIZE	BIT(1)
+#define MLX5E_MAIN_GROUP8_SIZE	BIT(0)
+#define MLX5E_MAIN_TABLE_SIZE	(MLX5E_MAIN_GROUP0_SIZE +\
+				 MLX5E_MAIN_GROUP1_SIZE +\
+				 MLX5E_MAIN_GROUP2_SIZE +\
+				 MLX5E_MAIN_GROUP3_SIZE +\
+				 MLX5E_MAIN_GROUP4_SIZE +\
+				 MLX5E_MAIN_GROUP5_SIZE +\
+				 MLX5E_MAIN_GROUP6_SIZE +\
+				 MLX5E_MAIN_GROUP7_SIZE +\
+				 MLX5E_MAIN_GROUP8_SIZE)
+
+static int __mlx5e_create_main_groups(struct mlx5e_flow_table *ft, u32 *in,
+				      int inlen)
 {
-	struct mlx5_flow_table_group *g;
-	u8 *dmac;
+	u8 *mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+	u8 *dmac = MLX5_ADDR_OF(create_flow_group_in, in,
+				match_criteria.outer_headers.dmac_47_16);
+	int err;
+	int ix = 0;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP0_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP1_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP2_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	eth_broadcast_addr(dmac);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP3_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	eth_broadcast_addr(dmac);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP4_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	eth_broadcast_addr(dmac);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP5_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol);
+	dmac[0] = 0x01;
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP6_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype);
+	dmac[0] = 0x01;
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP7_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	dmac[0] = 0x01;
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_MAIN_GROUP8_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	return 0;
+
+err_destroy_groups:
+	err = PTR_ERR(ft->g[ft->num_groups]);
+	ft->g[ft->num_groups] = NULL;
+	mlx5e_destroy_groups(ft);
+
+	return err;
+}
 
-	g = kcalloc(9, sizeof(*g), GFP_KERNEL);
-	if (!g)
+static int mlx5e_create_main_groups(struct mlx5e_flow_table *ft)
+{
+	u32 *in;
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	int err;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in)
 		return -ENOMEM;
 
-	g[0].log_sz = 3;
-	g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.ethertype);
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.ip_protocol);
-
-	g[1].log_sz = 1;
-	g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria,
-			 outer_headers.ethertype);
-
-	g[2].log_sz = 0;
-
-	g[3].log_sz = 14;
-	g[3].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[3].match_criteria,
-			    outer_headers.dmac_47_16);
-	memset(dmac, 0xff, ETH_ALEN);
-	MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria,
-			 outer_headers.ethertype);
-	MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria,
-			 outer_headers.ip_protocol);
-
-	g[4].log_sz = 13;
-	g[4].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[4].match_criteria,
-			    outer_headers.dmac_47_16);
-	memset(dmac, 0xff, ETH_ALEN);
-	MLX5_SET_TO_ONES(fte_match_param, g[4].match_criteria,
-			 outer_headers.ethertype);
-
-	g[5].log_sz = 11;
-	g[5].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[5].match_criteria,
-			    outer_headers.dmac_47_16);
-	memset(dmac, 0xff, ETH_ALEN);
-
-	g[6].log_sz = 2;
-	g[6].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[6].match_criteria,
-			    outer_headers.dmac_47_16);
-	dmac[0] = 0x01;
-	MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria,
-			 outer_headers.ethertype);
-	MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria,
-			 outer_headers.ip_protocol);
-
-	g[7].log_sz = 1;
-	g[7].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[7].match_criteria,
-			    outer_headers.dmac_47_16);
-	dmac[0] = 0x01;
-	MLX5_SET_TO_ONES(fte_match_param, g[7].match_criteria,
-			 outer_headers.ethertype);
+	err = __mlx5e_create_main_groups(ft, in, inlen);
 
-	g[8].log_sz = 0;
-	g[8].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g[8].match_criteria,
-			    outer_headers.dmac_47_16);
-	dmac[0] = 0x01;
-	priv->ft.main = mlx5_create_flow_table(priv->mdev, 1,
-					       MLX5_FLOW_TABLE_TYPE_NIC_RCV,
-					       9, g);
-	kfree(g);
+	kvfree(in);
+	return err;
+}
 
-	return priv->ft.main ? 0 : -ENOMEM;
+static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv)
+{
+	struct mlx5e_flow_table *ft = &priv->fts.main;
+	int err;
+
+	ft->num_groups = 0;
+	ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_MAIN_TABLE_SIZE);
+
+	if (IS_ERR(ft->t)) {
+		err = PTR_ERR(ft->t);
+		ft->t = NULL;
+		return err;
+	}
+	ft->g = kcalloc(MLX5E_NUM_MAIN_GROUPS, sizeof(*ft->g), GFP_KERNEL);
+	if (!ft->g) {
+		err = -ENOMEM;
+		goto err_destroy_main_flow_table;
+	}
+
+	err = mlx5e_create_main_groups(ft);
+	if (err)
+		goto err_free_g;
+	return 0;
+
+err_free_g:
+	kfree(ft->g);
+
+err_destroy_main_flow_table:
+	mlx5_destroy_flow_table(ft->t);
+	ft->t = NULL;
+
+	return err;
+}
+
+static void mlx5e_destroy_flow_table(struct mlx5e_flow_table *ft)
+{
+	mlx5e_destroy_groups(ft);
+	kfree(ft->g);
+	mlx5_destroy_flow_table(ft->t);
+	ft->t = NULL;
 }
 
 static void mlx5e_destroy_main_flow_table(struct mlx5e_priv *priv)
 {
-	mlx5_destroy_flow_table(priv->ft.main);
+	mlx5e_destroy_flow_table(&priv->fts.main);
 }
 
-static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv)
+#define MLX5E_NUM_VLAN_GROUPS	2
+#define MLX5E_VLAN_GROUP0_SIZE	BIT(12)
+#define MLX5E_VLAN_GROUP1_SIZE	BIT(1)
+#define MLX5E_VLAN_TABLE_SIZE	(MLX5E_VLAN_GROUP0_SIZE +\
+				 MLX5E_VLAN_GROUP1_SIZE)
+
+static int __mlx5e_create_vlan_groups(struct mlx5e_flow_table *ft, u32 *in,
+				      int inlen)
+{
+	int err;
+	int ix = 0;
+	u8 *mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_VLAN_GROUP0_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	memset(in, 0, inlen);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag);
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5E_VLAN_GROUP1_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
+	if (IS_ERR(ft->g[ft->num_groups]))
+		goto err_destroy_groups;
+	ft->num_groups++;
+
+	return 0;
+
+err_destroy_groups:
+	err = PTR_ERR(ft->g[ft->num_groups]);
+	ft->g[ft->num_groups] = NULL;
+	mlx5e_destroy_groups(ft);
+
+	return err;
+}
+
+static int mlx5e_create_vlan_groups(struct mlx5e_flow_table *ft)
 {
-	struct mlx5_flow_table_group *g;
+	u32 *in;
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	int err;
 
-	g = kcalloc(2, sizeof(*g), GFP_KERNEL);
-	if (!g)
+	in = mlx5_vzalloc(inlen);
+	if (!in)
 		return -ENOMEM;
 
-	g[0].log_sz = 12;
-	g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.vlan_tag);
-	MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria,
-			 outer_headers.first_vid);
-
-	/* untagged + any vlan id */
-	g[1].log_sz = 1;
-	g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria,
-			 outer_headers.vlan_tag);
-
-	priv->ft.vlan = mlx5_create_flow_table(priv->mdev, 0,
-					       MLX5_FLOW_TABLE_TYPE_NIC_RCV,
-					       2, g);
-
-	kfree(g);
-	return priv->ft.vlan ? 0 : -ENOMEM;
+	err = __mlx5e_create_vlan_groups(ft, in, inlen);
+
+	kvfree(in);
+	return err;
+}
+
+static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv)
+{
+	struct mlx5e_flow_table *ft = &priv->fts.vlan;
+	int err;
+
+	ft->num_groups = 0;
+	ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_VLAN_TABLE_SIZE);
+
+	if (IS_ERR(ft->t)) {
+		err = PTR_ERR(ft->t);
+		ft->t = NULL;
+		return err;
+	}
+	ft->g = kcalloc(MLX5E_NUM_VLAN_GROUPS, sizeof(*ft->g), GFP_KERNEL);
+	if (!ft->g) {
+		err = -ENOMEM;
+		goto err_destroy_vlan_flow_table;
+	}
+
+	err = mlx5e_create_vlan_groups(ft);
+	if (err)
+		goto err_free_g;
+
+	return 0;
+
+err_free_g:
+	kfree(ft->g);
+
+err_destroy_vlan_flow_table:
+	mlx5_destroy_flow_table(ft->t);
+	ft->t = NULL;
+
+	return err;
 }
 
 static void mlx5e_destroy_vlan_flow_table(struct mlx5e_priv *priv)
 {
-	mlx5_destroy_flow_table(priv->ft.vlan);
+	mlx5e_destroy_flow_table(&priv->fts.vlan);
 }
 
 int mlx5e_create_flow_tables(struct mlx5e_priv *priv)
 {
 	int err;
 
-	err = mlx5e_create_main_flow_table(priv);
+	priv->fts.ns = mlx5_get_flow_namespace(priv->mdev,
+					       MLX5_FLOW_NAMESPACE_KERNEL);
+
+	if (!priv->fts.ns)
+		return -EINVAL;
+
+	err = mlx5e_create_vlan_flow_table(priv);
 	if (err)
 		return err;
 
-	err = mlx5e_create_vlan_flow_table(priv);
+	err = mlx5e_create_main_flow_table(priv);
 	if (err)
-		goto err_destroy_main_flow_table;
+		goto err_destroy_vlan_flow_table;
 
 	err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0);
 	if (err)
-		goto err_destroy_vlan_flow_table;
+		goto err_destroy_main_flow_table;
 
 	return 0;
 
-err_destroy_vlan_flow_table:
-	mlx5e_destroy_vlan_flow_table(priv);
-
 err_destroy_main_flow_table:
 	mlx5e_destroy_main_flow_table(priv);
+err_destroy_vlan_flow_table:
+	mlx5e_destroy_vlan_flow_table(priv);
 
 	return err;
 }
@@ -1041,6 +1219,6 @@ err_destroy_main_flow_table:
 void mlx5e_destroy_flow_tables(struct mlx5e_priv *priv)
 {
 	mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0);
-	mlx5e_destroy_vlan_flow_table(priv);
 	mlx5e_destroy_main_flow_table(priv);
+	mlx5e_destroy_vlan_flow_table(priv);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index d67058afe87e..d4601a564699 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -30,7 +30,7 @@
  * SOFTWARE.
  */
 
-#include <linux/mlx5/flow_table.h>
+#include <linux/mlx5/fs.h>
 #include "en.h"
 #include "eswitch.h"
 
@@ -2103,6 +2103,11 @@ static void mlx5e_set_netdev_dev_addr(struct net_device *netdev)
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 
 	mlx5_query_nic_vport_mac_address(priv->mdev, 0, netdev->dev_addr);
+	if (is_zero_ether_addr(netdev->dev_addr) &&
+	    !MLX5_CAP_GEN(priv->mdev, vport_group_manager)) {
+		eth_hw_addr_random(netdev);
+		mlx5_core_info(priv->mdev, "Assigned random MAC address %pM\n", netdev->dev_addr);
+	}
 }
 
 static void mlx5e_build_netdev(struct net_device *netdev)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index d8939e597c54..bc3d9f8a75c1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -34,7 +34,7 @@
 #include <linux/mlx5/driver.h>
 #include <linux/mlx5/mlx5_ifc.h>
 #include <linux/mlx5/vport.h>
-#include <linux/mlx5/flow_table.h>
+#include <linux/mlx5/fs.h>
 #include "mlx5_core.h"
 #include "eswitch.h"
 
@@ -321,220 +321,6 @@ static void del_l2_table_entry(struct mlx5_core_dev *dev, u32 index)
 	free_l2_table_index(l2_table, index);
 }
 
-/* E-Switch FDB flow steering */
-struct dest_node {
-	struct list_head list;
-	struct mlx5_flow_destination dest;
-};
-
-static int _mlx5_flow_rule_apply(struct mlx5_flow_rule *fr)
-{
-	bool was_valid = fr->valid;
-	struct dest_node *dest_n;
-	u32 dest_list_size = 0;
-	void *in_match_value;
-	u32 *flow_context;
-	u32 flow_index;
-	int err;
-	int i;
-
-	if (list_empty(&fr->dest_list)) {
-		if (fr->valid)
-			mlx5_del_flow_table_entry(fr->ft, fr->fi);
-		fr->valid = false;
-		return 0;
-	}
-
-	list_for_each_entry(dest_n, &fr->dest_list, list)
-		dest_list_size++;
-
-	flow_context = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) +
-				    MLX5_ST_SZ_BYTES(dest_format_struct) *
-				    dest_list_size);
-	if (!flow_context)
-		return -ENOMEM;
-
-	MLX5_SET(flow_context, flow_context, flow_tag, fr->flow_tag);
-	MLX5_SET(flow_context, flow_context, action, fr->action);
-	MLX5_SET(flow_context, flow_context, destination_list_size,
-		 dest_list_size);
-
-	i = 0;
-	list_for_each_entry(dest_n, &fr->dest_list, list) {
-		void *dest_addr = MLX5_ADDR_OF(flow_context, flow_context,
-					       destination[i++]);
-
-		MLX5_SET(dest_format_struct, dest_addr, destination_type,
-			 dest_n->dest.type);
-		MLX5_SET(dest_format_struct, dest_addr, destination_id,
-			 dest_n->dest.vport_num);
-	}
-
-	in_match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value);
-	memcpy(in_match_value, fr->match_value, MLX5_ST_SZ_BYTES(fte_match_param));
-
-	err = mlx5_add_flow_table_entry(fr->ft, fr->match_criteria_enable,
-					fr->match_criteria, flow_context,
-					&flow_index);
-	if (!err) {
-		if (was_valid)
-			mlx5_del_flow_table_entry(fr->ft, fr->fi);
-		fr->fi = flow_index;
-		fr->valid = true;
-	}
-	kfree(flow_context);
-	return err;
-}
-
-static int mlx5_flow_rule_add_dest(struct mlx5_flow_rule *fr,
-				   struct mlx5_flow_destination *new_dest)
-{
-	struct dest_node *dest_n;
-	int err;
-
-	dest_n = kzalloc(sizeof(*dest_n), GFP_KERNEL);
-	if (!dest_n)
-		return -ENOMEM;
-
-	memcpy(&dest_n->dest, new_dest, sizeof(dest_n->dest));
-	mutex_lock(&fr->mutex);
-	list_add(&dest_n->list, &fr->dest_list);
-	err = _mlx5_flow_rule_apply(fr);
-	if (err) {
-		list_del(&dest_n->list);
-		kfree(dest_n);
-	}
-	mutex_unlock(&fr->mutex);
-	return err;
-}
-
-static int mlx5_flow_rule_del_dest(struct mlx5_flow_rule *fr,
-				   struct mlx5_flow_destination *dest)
-{
-	struct dest_node *dest_n;
-	struct dest_node *n;
-	int err;
-
-	mutex_lock(&fr->mutex);
-	list_for_each_entry_safe(dest_n, n, &fr->dest_list, list) {
-		if (dest->vport_num == dest_n->dest.vport_num)
-			goto found;
-	}
-	mutex_unlock(&fr->mutex);
-	return -ENOENT;
-
-found:
-	list_del(&dest_n->list);
-	err = _mlx5_flow_rule_apply(fr);
-	mutex_unlock(&fr->mutex);
-	kfree(dest_n);
-
-	return err;
-}
-
-static struct mlx5_flow_rule *find_fr(struct mlx5_eswitch *esw,
-				      u8 match_criteria_enable,
-				      u32 *match_value)
-{
-	struct hlist_head *hash = esw->mc_table;
-	struct esw_mc_addr *esw_mc;
-	u8 *dmac_v;
-
-	dmac_v = MLX5_ADDR_OF(fte_match_param, match_value,
-			      outer_headers.dmac_47_16);
-
-	/* UNICAST FULL MATCH */
-	if (!is_multicast_ether_addr(dmac_v))
-		return NULL;
-
-	/* MULTICAST FULL MATCH */
-	esw_mc = l2addr_hash_find(hash, dmac_v, struct esw_mc_addr);
-
-	return esw_mc ? esw_mc->uplink_rule : NULL;
-}
-
-static struct mlx5_flow_rule *alloc_fr(void *ft,
-				       u8 match_criteria_enable,
-				       u32 *match_criteria,
-				       u32 *match_value,
-				       u32 action,
-				       u32 flow_tag)
-{
-	struct mlx5_flow_rule *fr = kzalloc(sizeof(*fr), GFP_KERNEL);
-
-	if (!fr)
-		return NULL;
-
-	fr->match_criteria = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL);
-	fr->match_value = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL);
-	if (!fr->match_criteria || !fr->match_value) {
-		kfree(fr->match_criteria);
-		kfree(fr->match_value);
-		kfree(fr);
-		return NULL;
-	}
-
-	memcpy(fr->match_criteria, match_criteria, MLX5_ST_SZ_BYTES(fte_match_param));
-	memcpy(fr->match_value, match_value, MLX5_ST_SZ_BYTES(fte_match_param));
-	fr->match_criteria_enable = match_criteria_enable;
-	fr->flow_tag = flow_tag;
-	fr->action = action;
-
-	mutex_init(&fr->mutex);
-	INIT_LIST_HEAD(&fr->dest_list);
-	atomic_set(&fr->refcount, 0);
-	fr->ft = ft;
-	return fr;
-}
-
-static void deref_fr(struct mlx5_flow_rule *fr)
-{
-	if (!atomic_dec_and_test(&fr->refcount))
-		return;
-
-	kfree(fr->match_criteria);
-	kfree(fr->match_value);
-	kfree(fr);
-}
-
-static struct mlx5_flow_rule *
-mlx5_add_flow_rule(struct mlx5_eswitch *esw,
-		   u8 match_criteria_enable,
-		   u32 *match_criteria,
-		   u32 *match_value,
-		   u32 action,
-		   u32 flow_tag,
-		   struct mlx5_flow_destination *dest)
-{
-	struct mlx5_flow_rule *fr;
-	int err;
-
-	fr = find_fr(esw, match_criteria_enable, match_value);
-	fr = fr ? fr : alloc_fr(esw->fdb_table.fdb, match_criteria_enable, match_criteria,
-				match_value, action, flow_tag);
-	if (!fr)
-		return NULL;
-
-	atomic_inc(&fr->refcount);
-
-	err = mlx5_flow_rule_add_dest(fr, dest);
-	if (err) {
-		deref_fr(fr);
-		return NULL;
-	}
-
-	return fr;
-}
-
-static void mlx5_del_flow_rule(struct mlx5_flow_rule *fr, u32 vport)
-{
-	struct mlx5_flow_destination dest;
-
-	dest.vport_num = vport;
-	mlx5_flow_rule_del_dest(fr, &dest);
-	deref_fr(fr);
-}
-
 /* E-Switch FDB */
 static struct mlx5_flow_rule *
 esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport)
@@ -569,7 +355,7 @@ esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport)
 		  "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n",
 		  dmac_v, dmac_c, vport);
 	flow_rule =
-		mlx5_add_flow_rule(esw,
+		mlx5_add_flow_rule(esw->fdb_table.fdb,
 				   match_header,
 				   match_c,
 				   match_v,
@@ -589,33 +375,61 @@ out:
 
 static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports)
 {
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
 	struct mlx5_core_dev *dev = esw->dev;
-	struct mlx5_flow_table_group g;
+	struct mlx5_flow_namespace *root_ns;
 	struct mlx5_flow_table *fdb;
+	struct mlx5_flow_group *g;
+	void *match_criteria;
+	int table_size;
+	u32 *flow_group_in;
 	u8 *dmac;
+	int err = 0;
 
 	esw_debug(dev, "Create FDB log_max_size(%d)\n",
 		  MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
 
-	memset(&g, 0, sizeof(g));
-	/* UC MC Full match rules*/
-	g.log_sz = MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size);
-	g.match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
-	dmac = MLX5_ADDR_OF(fte_match_param, g.match_criteria,
-			    outer_headers.dmac_47_16);
-	/* Match criteria mask */
-	memset(dmac, 0xff, 6);
+	root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB);
+	if (!root_ns) {
+		esw_warn(dev, "Failed to get FDB flow namespace\n");
+		return -ENOMEM;
+	}
 
-	fdb = mlx5_create_flow_table(dev, 0,
-				     MLX5_FLOW_TABLE_TYPE_ESWITCH,
-				     1, &g);
-	if (fdb)
-		esw_debug(dev, "ESW: FDB Table created fdb->id %d\n", mlx5_get_flow_table_id(fdb));
-	else
-		esw_warn(dev, "ESW: Failed to create FDB Table\n");
+	flow_group_in = mlx5_vzalloc(inlen);
+	if (!flow_group_in)
+		return -ENOMEM;
+	memset(flow_group_in, 0, inlen);
+
+	table_size = BIT(MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
+	fdb = mlx5_create_flow_table(root_ns, 0, table_size);
+	if (IS_ERR_OR_NULL(fdb)) {
+		err = PTR_ERR(fdb);
+		esw_warn(dev, "Failed to create FDB Table err %d\n", err);
+		goto out;
+	}
 
+	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
+		 MLX5_MATCH_OUTER_HEADERS);
+	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
+	dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, outer_headers.dmac_47_16);
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 1);
+	eth_broadcast_addr(dmac);
+
+	g = mlx5_create_flow_group(fdb, flow_group_in);
+	if (IS_ERR_OR_NULL(g)) {
+		err = PTR_ERR(g);
+		esw_warn(dev, "Failed to create flow group err(%d)\n", err);
+		goto out;
+	}
+
+	esw->fdb_table.addr_grp = g;
 	esw->fdb_table.fdb = fdb;
-	return fdb ? 0 : -ENOMEM;
+out:
+	kfree(flow_group_in);
+	if (err && !IS_ERR_OR_NULL(fdb))
+		mlx5_destroy_flow_table(fdb);
+	return err;
 }
 
 static void esw_destroy_fdb_table(struct mlx5_eswitch *esw)
@@ -623,10 +437,11 @@ static void esw_destroy_fdb_table(struct mlx5_eswitch *esw)
 	if (!esw->fdb_table.fdb)
 		return;
 
-	esw_debug(esw->dev, "Destroy FDB Table fdb(%d)\n",
-		  mlx5_get_flow_table_id(esw->fdb_table.fdb));
+	esw_debug(esw->dev, "Destroy FDB Table\n");
+	mlx5_destroy_flow_group(esw->fdb_table.addr_grp);
 	mlx5_destroy_flow_table(esw->fdb_table.fdb);
 	esw->fdb_table.fdb = NULL;
+	esw->fdb_table.addr_grp = NULL;
 }
 
 /* E-Switch vport UC/MC lists management */
@@ -689,7 +504,7 @@ static int esw_del_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
 	del_l2_table_entry(esw->dev, esw_uc->table_index);
 
 	if (vaddr->flow_rule)
-		mlx5_del_flow_rule(vaddr->flow_rule, vport);
+		mlx5_del_flow_rule(vaddr->flow_rule);
 	vaddr->flow_rule = NULL;
 
 	l2addr_hash_del(esw_uc);
@@ -750,14 +565,14 @@ static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
 		  esw_mc->uplink_rule);
 
 	if (vaddr->flow_rule)
-		mlx5_del_flow_rule(vaddr->flow_rule, vport);
+		mlx5_del_flow_rule(vaddr->flow_rule);
 	vaddr->flow_rule = NULL;
 
 	if (--esw_mc->refcnt)
 		return 0;
 
 	if (esw_mc->uplink_rule)
-		mlx5_del_flow_rule(esw_mc->uplink_rule, UPLINK_VPORT);
+		mlx5_del_flow_rule(esw_mc->uplink_rule);
 
 	l2addr_hash_del(esw_mc);
 	return 0;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 02ff3eade026..3416a428f70f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -88,20 +88,6 @@ struct l2addr_node {
 	kfree(ptr);                                         \
 })
 
-struct mlx5_flow_rule {
-	void             *ft;
-	u32              fi;
-	u8               match_criteria_enable;
-	u32              *match_criteria;
-	u32              *match_value;
-	u32              action;
-	u32              flow_tag;
-	bool             valid;
-	atomic_t         refcount;
-	struct mutex     mutex; /* protect flow rule updates */
-	struct list_head dest_list;
-};
-
 struct mlx5_vport {
 	struct mlx5_core_dev    *dev;
 	int                     vport;
@@ -126,6 +112,7 @@ struct mlx5_l2_table {
 
 struct mlx5_eswitch_fdb {
 	void *fdb;
+	struct mlx5_flow_group *addr_grp;
 };
 
 struct mlx5_eswitch {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c
deleted file mode 100644
index ca90b9bc3b95..000000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c
+++ /dev/null
@@ -1,422 +0,0 @@
-/*
- * Copyright (c) 2013-2015, Mellanox Technologies, Ltd.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- */
-
-#include <linux/export.h>
-#include <linux/mlx5/driver.h>
-#include <linux/mlx5/flow_table.h>
-#include "mlx5_core.h"
-
-struct mlx5_ftg {
-	struct mlx5_flow_table_group    g;
-	u32				id;
-	u32				start_ix;
-};
-
-struct mlx5_flow_table {
-	struct mlx5_core_dev	*dev;
-	u8			level;
-	u8			type;
-	u32			id;
-	struct mutex		mutex; /* sync bitmap alloc */
-	u16			num_groups;
-	struct mlx5_ftg		*group;
-	unsigned long		*bitmap;
-	u32			size;
-};
-
-static int mlx5_set_flow_entry_cmd(struct mlx5_flow_table *ft, u32 group_ix,
-				   u32 flow_index, void *flow_context)
-{
-	u32 out[MLX5_ST_SZ_DW(set_fte_out)];
-	u32 *in;
-	void *in_flow_context;
-	int fcdls =
-		MLX5_GET(flow_context, flow_context, destination_list_size) *
-		MLX5_ST_SZ_BYTES(dest_format_struct);
-	int inlen = MLX5_ST_SZ_BYTES(set_fte_in) + fcdls;
-	int err;
-
-	in = mlx5_vzalloc(inlen);
-	if (!in) {
-		mlx5_core_warn(ft->dev, "failed to allocate inbox\n");
-		return -ENOMEM;
-	}
-
-	MLX5_SET(set_fte_in, in, table_type, ft->type);
-	MLX5_SET(set_fte_in, in, table_id,   ft->id);
-	MLX5_SET(set_fte_in, in, flow_index, flow_index);
-	MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY);
-
-	in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context);
-	memcpy(in_flow_context, flow_context,
-	       MLX5_ST_SZ_BYTES(flow_context) + fcdls);
-
-	MLX5_SET(flow_context, in_flow_context, group_id,
-		 ft->group[group_ix].id);
-
-	memset(out, 0, sizeof(out));
-	err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out,
-					 sizeof(out));
-	kvfree(in);
-
-	return err;
-}
-
-static void mlx5_del_flow_entry_cmd(struct mlx5_flow_table *ft, u32 flow_index)
-{
-	u32 in[MLX5_ST_SZ_DW(delete_fte_in)];
-	u32 out[MLX5_ST_SZ_DW(delete_fte_out)];
-
-	memset(in, 0, sizeof(in));
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_DFTEI(p, x, v) MLX5_SET(delete_fte_in, p, x, v)
-	MLX5_SET_DFTEI(in, table_type, ft->type);
-	MLX5_SET_DFTEI(in, table_id,   ft->id);
-	MLX5_SET_DFTEI(in, flow_index, flow_index);
-	MLX5_SET_DFTEI(in, opcode,     MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY);
-
-	mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out));
-}
-
-static void mlx5_destroy_flow_group_cmd(struct mlx5_flow_table *ft, int i)
-{
-	u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)];
-	u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)];
-
-	memset(in, 0, sizeof(in));
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_DFGI(p, x, v) MLX5_SET(destroy_flow_group_in, p, x, v)
-	MLX5_SET_DFGI(in, table_type, ft->type);
-	MLX5_SET_DFGI(in, table_id,   ft->id);
-	MLX5_SET_DFGI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_GROUP);
-	MLX5_SET_DFGI(in, group_id, ft->group[i].id);
-	mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out));
-}
-
-static int mlx5_create_flow_group_cmd(struct mlx5_flow_table *ft, int i)
-{
-	u32 out[MLX5_ST_SZ_DW(create_flow_group_out)];
-	u32 *in;
-	void *in_match_criteria;
-	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
-	struct mlx5_flow_table_group *g = &ft->group[i].g;
-	u32 start_ix = ft->group[i].start_ix;
-	u32 end_ix = start_ix + (1 << g->log_sz) - 1;
-	int err;
-
-	in = mlx5_vzalloc(inlen);
-	if (!in) {
-		mlx5_core_warn(ft->dev, "failed to allocate inbox\n");
-		return -ENOMEM;
-	}
-	in_match_criteria = MLX5_ADDR_OF(create_flow_group_in, in,
-					 match_criteria);
-
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_CFGI(p, x, v) MLX5_SET(create_flow_group_in, p, x, v)
-	MLX5_SET_CFGI(in, table_type,            ft->type);
-	MLX5_SET_CFGI(in, table_id,              ft->id);
-	MLX5_SET_CFGI(in, opcode,                MLX5_CMD_OP_CREATE_FLOW_GROUP);
-	MLX5_SET_CFGI(in, start_flow_index,      start_ix);
-	MLX5_SET_CFGI(in, end_flow_index,        end_ix);
-	MLX5_SET_CFGI(in, match_criteria_enable, g->match_criteria_enable);
-
-	memcpy(in_match_criteria, g->match_criteria,
-	       MLX5_ST_SZ_BYTES(fte_match_param));
-
-	err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out,
-					 sizeof(out));
-	if (!err)
-		ft->group[i].id = MLX5_GET(create_flow_group_out, out,
-					   group_id);
-
-	kvfree(in);
-
-	return err;
-}
-
-static void mlx5_destroy_flow_table_groups(struct mlx5_flow_table *ft)
-{
-	int i;
-
-	for (i = 0; i < ft->num_groups; i++)
-		mlx5_destroy_flow_group_cmd(ft, i);
-}
-
-static int mlx5_create_flow_table_groups(struct mlx5_flow_table *ft)
-{
-	int err;
-	int i;
-
-	for (i = 0; i < ft->num_groups; i++) {
-		err = mlx5_create_flow_group_cmd(ft, i);
-		if (err)
-			goto err_destroy_flow_table_groups;
-	}
-
-	return 0;
-
-err_destroy_flow_table_groups:
-	for (i--; i >= 0; i--)
-		mlx5_destroy_flow_group_cmd(ft, i);
-
-	return err;
-}
-
-static int mlx5_create_flow_table_cmd(struct mlx5_flow_table *ft)
-{
-	u32 in[MLX5_ST_SZ_DW(create_flow_table_in)];
-	u32 out[MLX5_ST_SZ_DW(create_flow_table_out)];
-	int err;
-
-	memset(in, 0, sizeof(in));
-
-	MLX5_SET(create_flow_table_in, in, table_type, ft->type);
-	MLX5_SET(create_flow_table_in, in, level,      ft->level);
-	MLX5_SET(create_flow_table_in, in, log_size,   order_base_2(ft->size));
-
-	MLX5_SET(create_flow_table_in, in, opcode,
-		 MLX5_CMD_OP_CREATE_FLOW_TABLE);
-
-	memset(out, 0, sizeof(out));
-	err = mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out,
-					 sizeof(out));
-	if (err)
-		return err;
-
-	ft->id = MLX5_GET(create_flow_table_out, out, table_id);
-
-	return 0;
-}
-
-static void mlx5_destroy_flow_table_cmd(struct mlx5_flow_table *ft)
-{
-	u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)];
-	u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)];
-
-	memset(in, 0, sizeof(in));
-	memset(out, 0, sizeof(out));
-
-#define MLX5_SET_DFTI(p, x, v) MLX5_SET(destroy_flow_table_in, p, x, v)
-	MLX5_SET_DFTI(in, table_type, ft->type);
-	MLX5_SET_DFTI(in, table_id,   ft->id);
-	MLX5_SET_DFTI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_TABLE);
-
-	mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out));
-}
-
-static int mlx5_find_group(struct mlx5_flow_table *ft, u8 match_criteria_enable,
-			   u32 *match_criteria, int *group_ix)
-{
-	void *mc_outer = MLX5_ADDR_OF(fte_match_param, match_criteria,
-				      outer_headers);
-	void *mc_misc  = MLX5_ADDR_OF(fte_match_param, match_criteria,
-				      misc_parameters);
-	void *mc_inner = MLX5_ADDR_OF(fte_match_param, match_criteria,
-				      inner_headers);
-	int mc_outer_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4);
-	int mc_misc_sz  = MLX5_ST_SZ_BYTES(fte_match_set_misc);
-	int mc_inner_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4);
-	int i;
-
-	for (i = 0; i < ft->num_groups; i++) {
-		struct mlx5_flow_table_group *g = &ft->group[i].g;
-		void *gmc_outer = MLX5_ADDR_OF(fte_match_param,
-					       g->match_criteria,
-					       outer_headers);
-		void *gmc_misc  = MLX5_ADDR_OF(fte_match_param,
-					       g->match_criteria,
-					       misc_parameters);
-		void *gmc_inner = MLX5_ADDR_OF(fte_match_param,
-					       g->match_criteria,
-					       inner_headers);
-
-		if (g->match_criteria_enable != match_criteria_enable)
-			continue;
-
-		if (match_criteria_enable & MLX5_MATCH_OUTER_HEADERS)
-			if (memcmp(mc_outer, gmc_outer, mc_outer_sz))
-				continue;
-
-		if (match_criteria_enable & MLX5_MATCH_MISC_PARAMETERS)
-			if (memcmp(mc_misc, gmc_misc, mc_misc_sz))
-				continue;
-
-		if (match_criteria_enable & MLX5_MATCH_INNER_HEADERS)
-			if (memcmp(mc_inner, gmc_inner, mc_inner_sz))
-				continue;
-
-		*group_ix = i;
-		return 0;
-	}
-
-	return -EINVAL;
-}
-
-static int alloc_flow_index(struct mlx5_flow_table *ft, int group_ix, u32 *ix)
-{
-	struct mlx5_ftg *g = &ft->group[group_ix];
-	int err = 0;
-
-	mutex_lock(&ft->mutex);
-
-	*ix = find_next_zero_bit(ft->bitmap, ft->size, g->start_ix);
-	if (*ix >= (g->start_ix + (1 << g->g.log_sz)))
-		err = -ENOSPC;
-	else
-		__set_bit(*ix, ft->bitmap);
-
-	mutex_unlock(&ft->mutex);
-
-	return err;
-}
-
-static void mlx5_free_flow_index(struct mlx5_flow_table *ft, u32 ix)
-{
-	__clear_bit(ix, ft->bitmap);
-}
-
-int mlx5_add_flow_table_entry(void *flow_table, u8 match_criteria_enable,
-			      void *match_criteria, void *flow_context,
-			      u32 *flow_index)
-{
-	struct mlx5_flow_table *ft = flow_table;
-	int group_ix;
-	int err;
-
-	err = mlx5_find_group(ft, match_criteria_enable, match_criteria,
-			      &group_ix);
-	if (err) {
-		mlx5_core_warn(ft->dev, "mlx5_find_group failed\n");
-		return err;
-	}
-
-	err = alloc_flow_index(ft, group_ix, flow_index);
-	if (err) {
-		mlx5_core_warn(ft->dev, "alloc_flow_index failed\n");
-		return err;
-	}
-
-	return mlx5_set_flow_entry_cmd(ft, group_ix, *flow_index, flow_context);
-}
-EXPORT_SYMBOL(mlx5_add_flow_table_entry);
-
-void mlx5_del_flow_table_entry(void *flow_table, u32 flow_index)
-{
-	struct mlx5_flow_table *ft = flow_table;
-
-	mlx5_del_flow_entry_cmd(ft, flow_index);
-	mlx5_free_flow_index(ft, flow_index);
-}
-EXPORT_SYMBOL(mlx5_del_flow_table_entry);
-
-void *mlx5_create_flow_table(struct mlx5_core_dev *dev, u8 level, u8 table_type,
-			     u16 num_groups,
-			     struct mlx5_flow_table_group *group)
-{
-	struct mlx5_flow_table *ft;
-	u32 start_ix = 0;
-	u32 ft_size = 0;
-	void *gr;
-	void *bm;
-	int err;
-	int i;
-
-	for (i = 0; i < num_groups; i++)
-		ft_size += (1 << group[i].log_sz);
-
-	ft = kzalloc(sizeof(*ft), GFP_KERNEL);
-	gr = kcalloc(num_groups, sizeof(struct mlx5_ftg), GFP_KERNEL);
-	bm = kcalloc(BITS_TO_LONGS(ft_size), sizeof(uintptr_t), GFP_KERNEL);
-	if (!ft || !gr || !bm)
-		goto err_free_ft;
-
-	ft->group	= gr;
-	ft->bitmap	= bm;
-	ft->num_groups	= num_groups;
-	ft->level	= level;
-	ft->type	= table_type;
-	ft->size	= ft_size;
-	ft->dev		= dev;
-	mutex_init(&ft->mutex);
-
-	for (i = 0; i < ft->num_groups; i++) {
-		memcpy(&ft->group[i].g, &group[i], sizeof(*group));
-		ft->group[i].start_ix = start_ix;
-		start_ix += 1 << group[i].log_sz;
-	}
-
-	err = mlx5_create_flow_table_cmd(ft);
-	if (err)
-		goto err_free_ft;
-
-	err = mlx5_create_flow_table_groups(ft);
-	if (err)
-		goto err_destroy_flow_table_cmd;
-
-	return ft;
-
-err_destroy_flow_table_cmd:
-	mlx5_destroy_flow_table_cmd(ft);
-
-err_free_ft:
-	mlx5_core_warn(dev, "failed to alloc flow table\n");
-	kfree(bm);
-	kfree(gr);
-	kfree(ft);
-
-	return NULL;
-}
-EXPORT_SYMBOL(mlx5_create_flow_table);
-
-void mlx5_destroy_flow_table(void *flow_table)
-{
-	struct mlx5_flow_table *ft = flow_table;
-
-	mlx5_destroy_flow_table_groups(ft);
-	mlx5_destroy_flow_table_cmd(ft);
-	kfree(ft->bitmap);
-	kfree(ft->group);
-	kfree(ft);
-}
-EXPORT_SYMBOL(mlx5_destroy_flow_table);
-
-u32 mlx5_get_flow_table_id(void *flow_table)
-{
-	struct mlx5_flow_table *ft = flow_table;
-
-	return ft->id;
-}
-EXPORT_SYMBOL(mlx5_get_flow_table_id);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
new file mode 100644
index 000000000000..5096f4f336bd
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -0,0 +1,239 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/device.h>
+#include <linux/mlx5/mlx5_ifc.h>
+
+#include "fs_core.h"
+#include "fs_cmd.h"
+#include "mlx5_core.h"
+
+int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
+			       enum fs_flow_table_type type, unsigned int level,
+			       unsigned int log_size, unsigned int *table_id)
+{
+	u32 out[MLX5_ST_SZ_DW(create_flow_table_out)];
+	u32 in[MLX5_ST_SZ_DW(create_flow_table_in)];
+	int err;
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(create_flow_table_in, in, opcode,
+		 MLX5_CMD_OP_CREATE_FLOW_TABLE);
+
+	MLX5_SET(create_flow_table_in, in, table_type, type);
+	MLX5_SET(create_flow_table_in, in, level, level);
+	MLX5_SET(create_flow_table_in, in, log_size, log_size);
+
+	memset(out, 0, sizeof(out));
+	err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					 sizeof(out));
+
+	if (!err)
+		*table_id = MLX5_GET(create_flow_table_out, out,
+				     table_id);
+	return err;
+}
+
+int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft)
+{
+	u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)];
+	u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(destroy_flow_table_in, in, opcode,
+		 MLX5_CMD_OP_DESTROY_FLOW_TABLE);
+	MLX5_SET(destroy_flow_table_in, in, table_type, ft->type);
+	MLX5_SET(destroy_flow_table_in, in, table_id, ft->id);
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					  sizeof(out));
+}
+
+int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev,
+			       struct mlx5_flow_table *ft,
+			       u32 *in,
+			       unsigned int *group_id)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	u32 out[MLX5_ST_SZ_DW(create_flow_group_out)];
+	int err;
+
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(create_flow_group_in, in, opcode,
+		 MLX5_CMD_OP_CREATE_FLOW_GROUP);
+	MLX5_SET(create_flow_group_in, in, table_type, ft->type);
+	MLX5_SET(create_flow_group_in, in, table_id, ft->id);
+
+	err = mlx5_cmd_exec_check_status(dev, in,
+					 inlen, out,
+					 sizeof(out));
+	if (!err)
+		*group_id = MLX5_GET(create_flow_group_out, out,
+				     group_id);
+
+	return err;
+}
+
+int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft,
+				unsigned int group_id)
+{
+	u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)];
+	u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(destroy_flow_group_in, in, opcode,
+		 MLX5_CMD_OP_DESTROY_FLOW_GROUP);
+	MLX5_SET(destroy_flow_group_in, in, table_type, ft->type);
+	MLX5_SET(destroy_flow_group_in, in, table_id, ft->id);
+	MLX5_SET(destroy_flow_group_in, in, group_id, group_id);
+
+	return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out,
+					  sizeof(out));
+}
+
+static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev,
+			    int opmod, int modify_mask,
+			    struct mlx5_flow_table *ft,
+			    unsigned group_id,
+			    struct fs_fte *fte)
+{
+	unsigned int inlen = MLX5_ST_SZ_BYTES(set_fte_in) +
+		fte->dests_size * MLX5_ST_SZ_BYTES(dest_format_struct);
+	u32 out[MLX5_ST_SZ_DW(set_fte_out)];
+	struct mlx5_flow_rule *dst;
+	void *in_flow_context;
+	void *in_match_value;
+	void *in_dests;
+	u32 *in;
+	int err;
+
+	in = mlx5_vzalloc(inlen);
+	if (!in) {
+		mlx5_core_warn(dev, "failed to allocate inbox\n");
+		return -ENOMEM;
+	}
+
+	MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY);
+	MLX5_SET(set_fte_in, in, op_mod, opmod);
+	MLX5_SET(set_fte_in, in, modify_enable_mask, modify_mask);
+	MLX5_SET(set_fte_in, in, table_type, ft->type);
+	MLX5_SET(set_fte_in, in, table_id,   ft->id);
+	MLX5_SET(set_fte_in, in, flow_index, fte->index);
+
+	in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context);
+	MLX5_SET(flow_context, in_flow_context, group_id, group_id);
+	MLX5_SET(flow_context, in_flow_context, flow_tag, fte->flow_tag);
+	MLX5_SET(flow_context, in_flow_context, action, fte->action);
+	MLX5_SET(flow_context, in_flow_context, destination_list_size,
+		 fte->dests_size);
+	in_match_value = MLX5_ADDR_OF(flow_context, in_flow_context,
+				      match_value);
+	memcpy(in_match_value, &fte->val, MLX5_ST_SZ_BYTES(fte_match_param));
+
+	in_dests = MLX5_ADDR_OF(flow_context, in_flow_context, destination);
+	list_for_each_entry(dst, &fte->node.children, node.list) {
+		unsigned int id;
+
+		MLX5_SET(dest_format_struct, in_dests, destination_type,
+			 dst->dest_attr.type);
+		if (dst->dest_attr.type ==
+		    MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE)
+			id = dst->dest_attr.ft->id;
+		else
+			id = dst->dest_attr.tir_num;
+		MLX5_SET(dest_format_struct, in_dests, destination_id, id);
+		in_dests += MLX5_ST_SZ_BYTES(dest_format_struct);
+	}
+	memset(out, 0, sizeof(out));
+	err = mlx5_cmd_exec_check_status(dev, in, inlen, out,
+					 sizeof(out));
+	kvfree(in);
+
+	return err;
+}
+
+int mlx5_cmd_create_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte)
+{
+	return	mlx5_cmd_set_fte(dev, 0, 0, ft, group_id, fte);
+}
+
+int mlx5_cmd_update_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte)
+{
+	int opmod;
+	int modify_mask;
+	int atomic_mod_cap = MLX5_CAP_FLOWTABLE(dev,
+						flow_table_properties_nic_receive.
+						flow_modify_en);
+	if (!atomic_mod_cap)
+		return -ENOTSUPP;
+	opmod = 1;
+	modify_mask = 1 <<
+		MLX5_SET_FTE_MODIFY_ENABLE_MASK_DESTINATION_LIST;
+
+	return	mlx5_cmd_set_fte(dev, opmod, modify_mask, ft, group_id, fte);
+}
+
+int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned int index)
+{
+	u32 out[MLX5_ST_SZ_DW(delete_fte_out)];
+	u32 in[MLX5_ST_SZ_DW(delete_fte_in)];
+	int err;
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(delete_fte_in, in, opcode, MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY);
+	MLX5_SET(delete_fte_in, in, table_type, ft->type);
+	MLX5_SET(delete_fte_in, in, table_id, ft->id);
+	MLX5_SET(delete_fte_in, in, flow_index, index);
+
+	err =  mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out));
+
+	return err;
+}
diff --git a/include/linux/mlx5/flow_table.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
index 0f2a15cf3317..f39304ede186 100644
--- a/include/linux/mlx5/flow_table.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2013-2015, Mellanox Technologies, Ltd.  All rights reserved.
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -30,34 +30,36 @@
  * SOFTWARE.
  */
 
-#ifndef MLX5_FLOW_TABLE_H
-#define MLX5_FLOW_TABLE_H
-
-#include <linux/mlx5/driver.h>
-
-struct mlx5_flow_table_group {
-	u8	log_sz;
-	u8	match_criteria_enable;
-	u32	match_criteria[MLX5_ST_SZ_DW(fte_match_param)];
-};
-
-struct mlx5_flow_destination {
-	enum mlx5_flow_destination_type	type;
-	union {
-		u32			tir_num;
-		void			*ft;
-		u32			vport_num;
-	};
-};
-
-void *mlx5_create_flow_table(struct mlx5_core_dev *dev, u8 level, u8 table_type,
-			     u16 num_groups,
-			     struct mlx5_flow_table_group *group);
-void mlx5_destroy_flow_table(void *flow_table);
-int mlx5_add_flow_table_entry(void *flow_table, u8 match_criteria_enable,
-			      void *match_criteria, void *flow_context,
-			      u32 *flow_index);
-void mlx5_del_flow_table_entry(void *flow_table, u32 flow_index);
-u32 mlx5_get_flow_table_id(void *flow_table);
-
-#endif /* MLX5_FLOW_TABLE_H */
+#ifndef _MLX5_FS_CMD_
+#define _MLX5_FS_CMD_
+
+int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
+			       enum fs_flow_table_type type, unsigned int level,
+			       unsigned int log_size, unsigned int *table_id);
+
+int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft);
+
+int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev,
+			       struct mlx5_flow_table *ft,
+			       u32 *in, unsigned int *group_id);
+
+int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev,
+				struct mlx5_flow_table *ft,
+				unsigned int group_id);
+
+int mlx5_cmd_create_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte);
+
+int mlx5_cmd_update_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned group_id,
+			struct fs_fte *fte);
+
+int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev,
+			struct mlx5_flow_table *ft,
+			unsigned int index);
+
+#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
new file mode 100644
index 000000000000..f7d62fe595f6
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -0,0 +1,1047 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/mutex.h>
+#include <linux/mlx5/driver.h>
+
+#include "mlx5_core.h"
+#include "fs_core.h"
+#include "fs_cmd.h"
+
+#define INIT_TREE_NODE_ARRAY_SIZE(...)	(sizeof((struct init_tree_node[]){__VA_ARGS__}) /\
+					 sizeof(struct init_tree_node))
+
+#define INIT_PRIO(min_level_val, max_ft_val,\
+		  start_level_val, ...) {.type = FS_TYPE_PRIO,\
+	.min_ft_level = min_level_val,\
+	.start_level = start_level_val,\
+	.max_ft = max_ft_val,\
+	.children = (struct init_tree_node[]) {__VA_ARGS__},\
+	.ar_size = INIT_TREE_NODE_ARRAY_SIZE(__VA_ARGS__) \
+}
+
+#define ADD_PRIO(min_level_val, max_ft_val, start_level_val, ...)\
+	INIT_PRIO(min_level_val, max_ft_val, start_level_val,\
+		  __VA_ARGS__)\
+
+#define ADD_FT_PRIO(max_ft_val, start_level_val, ...)\
+	INIT_PRIO(0, max_ft_val, start_level_val,\
+		  __VA_ARGS__)\
+
+#define ADD_NS(...) {.type = FS_TYPE_NAMESPACE,\
+	.children = (struct init_tree_node[]) {__VA_ARGS__},\
+	.ar_size = INIT_TREE_NODE_ARRAY_SIZE(__VA_ARGS__) \
+}
+
+#define KERNEL_START_LEVEL 0
+#define KERNEL_P0_START_LEVEL KERNEL_START_LEVEL
+#define KERNEL_MAX_FT 2
+#define KENREL_MIN_LEVEL 2
+static struct init_tree_node {
+	enum fs_node_type	type;
+	struct init_tree_node *children;
+	int ar_size;
+	int min_ft_level;
+	int prio;
+	int max_ft;
+	int start_level;
+} root_fs = {
+	.type = FS_TYPE_NAMESPACE,
+	.ar_size = 1,
+	.children = (struct init_tree_node[]) {
+		ADD_PRIO(KENREL_MIN_LEVEL, KERNEL_MAX_FT,
+			 KERNEL_START_LEVEL,
+			 ADD_NS(ADD_FT_PRIO(KERNEL_MAX_FT,
+					    KERNEL_P0_START_LEVEL))),
+	}
+};
+
+static void del_rule(struct fs_node *node);
+static void del_flow_table(struct fs_node *node);
+static void del_flow_group(struct fs_node *node);
+static void del_fte(struct fs_node *node);
+
+static void tree_init_node(struct fs_node *node,
+			   unsigned int refcount,
+			   void (*remove_func)(struct fs_node *))
+{
+	atomic_set(&node->refcount, refcount);
+	INIT_LIST_HEAD(&node->list);
+	INIT_LIST_HEAD(&node->children);
+	mutex_init(&node->lock);
+	node->remove_func = remove_func;
+}
+
+static void tree_add_node(struct fs_node *node, struct fs_node *parent)
+{
+	if (parent)
+		atomic_inc(&parent->refcount);
+	node->parent = parent;
+
+	/* Parent is the root */
+	if (!parent)
+		node->root = node;
+	else
+		node->root = parent->root;
+}
+
+static void tree_get_node(struct fs_node *node)
+{
+	atomic_inc(&node->refcount);
+}
+
+static void nested_lock_ref_node(struct fs_node *node)
+{
+	if (node) {
+		mutex_lock_nested(&node->lock, SINGLE_DEPTH_NESTING);
+		atomic_inc(&node->refcount);
+	}
+}
+
+static void lock_ref_node(struct fs_node *node)
+{
+	if (node) {
+		mutex_lock(&node->lock);
+		atomic_inc(&node->refcount);
+	}
+}
+
+static void unlock_ref_node(struct fs_node *node)
+{
+	if (node) {
+		atomic_dec(&node->refcount);
+		mutex_unlock(&node->lock);
+	}
+}
+
+static void tree_put_node(struct fs_node *node)
+{
+	struct fs_node *parent_node = node->parent;
+
+	lock_ref_node(parent_node);
+	if (atomic_dec_and_test(&node->refcount)) {
+		if (parent_node)
+			list_del_init(&node->list);
+		if (node->remove_func)
+			node->remove_func(node);
+		kfree(node);
+		node = NULL;
+	}
+	unlock_ref_node(parent_node);
+	if (!node && parent_node)
+		tree_put_node(parent_node);
+}
+
+static int tree_remove_node(struct fs_node *node)
+{
+	if (atomic_read(&node->refcount) > 1)
+		return -EPERM;
+	tree_put_node(node);
+	return 0;
+}
+
+static struct fs_prio *find_prio(struct mlx5_flow_namespace *ns,
+				 unsigned int prio)
+{
+	struct fs_prio *iter_prio;
+
+	fs_for_each_prio(iter_prio, ns) {
+		if (iter_prio->prio == prio)
+			return iter_prio;
+	}
+
+	return NULL;
+}
+
+static unsigned int find_next_free_level(struct fs_prio *prio)
+{
+	if (!list_empty(&prio->node.children)) {
+		struct mlx5_flow_table *ft;
+
+		ft = list_last_entry(&prio->node.children,
+				     struct mlx5_flow_table,
+				     node.list);
+		return ft->level + 1;
+	}
+	return prio->start_level;
+}
+
+static bool masked_memcmp(void *mask, void *val1, void *val2, size_t size)
+{
+	unsigned int i;
+
+	for (i = 0; i < size; i++, mask++, val1++, val2++)
+		if ((*((u8 *)val1) & (*(u8 *)mask)) !=
+		    ((*(u8 *)val2) & (*(u8 *)mask)))
+			return false;
+
+	return true;
+}
+
+static bool compare_match_value(struct mlx5_flow_group_mask *mask,
+				void *fte_param1, void *fte_param2)
+{
+	if (mask->match_criteria_enable &
+	    1 << MLX5_CREATE_FLOW_GROUP_IN_MATCH_CRITERIA_ENABLE_OUTER_HEADERS) {
+		void *fte_match1 = MLX5_ADDR_OF(fte_match_param,
+						fte_param1, outer_headers);
+		void *fte_match2 = MLX5_ADDR_OF(fte_match_param,
+						fte_param2, outer_headers);
+		void *fte_mask = MLX5_ADDR_OF(fte_match_param,
+					      mask->match_criteria, outer_headers);
+
+		if (!masked_memcmp(fte_mask, fte_match1, fte_match2,
+				   MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4)))
+			return false;
+	}
+
+	if (mask->match_criteria_enable &
+	    1 << MLX5_CREATE_FLOW_GROUP_IN_MATCH_CRITERIA_ENABLE_MISC_PARAMETERS) {
+		void *fte_match1 = MLX5_ADDR_OF(fte_match_param,
+						fte_param1, misc_parameters);
+		void *fte_match2 = MLX5_ADDR_OF(fte_match_param,
+						fte_param2, misc_parameters);
+		void *fte_mask = MLX5_ADDR_OF(fte_match_param,
+					  mask->match_criteria, misc_parameters);
+
+		if (!masked_memcmp(fte_mask, fte_match1, fte_match2,
+				   MLX5_ST_SZ_BYTES(fte_match_set_misc)))
+			return false;
+	}
+
+	if (mask->match_criteria_enable &
+	    1 << MLX5_CREATE_FLOW_GROUP_IN_MATCH_CRITERIA_ENABLE_INNER_HEADERS) {
+		void *fte_match1 = MLX5_ADDR_OF(fte_match_param,
+						fte_param1, inner_headers);
+		void *fte_match2 = MLX5_ADDR_OF(fte_match_param,
+						fte_param2, inner_headers);
+		void *fte_mask = MLX5_ADDR_OF(fte_match_param,
+					  mask->match_criteria, inner_headers);
+
+		if (!masked_memcmp(fte_mask, fte_match1, fte_match2,
+				   MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4)))
+			return false;
+	}
+	return true;
+}
+
+static bool compare_match_criteria(u8 match_criteria_enable1,
+				   u8 match_criteria_enable2,
+				   void *mask1, void *mask2)
+{
+	return match_criteria_enable1 == match_criteria_enable2 &&
+		!memcmp(mask1, mask2, MLX5_ST_SZ_BYTES(fte_match_param));
+}
+
+static struct mlx5_flow_root_namespace *find_root(struct fs_node *node)
+{
+	struct fs_node *root;
+	struct mlx5_flow_namespace *ns;
+
+	root = node->root;
+
+	if (WARN_ON(root->type != FS_TYPE_NAMESPACE)) {
+		pr_warn("mlx5: flow steering node is not in tree or garbaged\n");
+		return NULL;
+	}
+
+	ns = container_of(root, struct mlx5_flow_namespace, node);
+	return container_of(ns, struct mlx5_flow_root_namespace, ns);
+}
+
+static inline struct mlx5_core_dev *get_dev(struct fs_node *node)
+{
+	struct mlx5_flow_root_namespace *root = find_root(node);
+
+	if (root)
+		return root->dev;
+	return NULL;
+}
+
+static void del_flow_table(struct fs_node *node)
+{
+	struct mlx5_flow_table *ft;
+	struct mlx5_core_dev *dev;
+	struct fs_prio *prio;
+	int err;
+
+	fs_get_obj(ft, node);
+	dev = get_dev(&ft->node);
+
+	err = mlx5_cmd_destroy_flow_table(dev, ft);
+	if (err)
+		pr_warn("flow steering can't destroy ft\n");
+	fs_get_obj(prio, ft->node.parent);
+	prio->num_ft--;
+}
+
+static void del_rule(struct fs_node *node)
+{
+	struct mlx5_flow_rule *rule;
+	struct mlx5_flow_table *ft;
+	struct mlx5_flow_group *fg;
+	struct fs_fte *fte;
+	u32	*match_value;
+	struct mlx5_core_dev *dev = get_dev(node);
+	int match_len = MLX5_ST_SZ_BYTES(fte_match_param);
+	int err;
+
+	match_value = mlx5_vzalloc(match_len);
+	if (!match_value) {
+		pr_warn("failed to allocate inbox\n");
+		return;
+	}
+
+	fs_get_obj(rule, node);
+	fs_get_obj(fte, rule->node.parent);
+	fs_get_obj(fg, fte->node.parent);
+	memcpy(match_value, fte->val, sizeof(fte->val));
+	fs_get_obj(ft, fg->node.parent);
+	list_del(&rule->node.list);
+	fte->dests_size--;
+	if (fte->dests_size) {
+		err = mlx5_cmd_update_fte(dev, ft,
+					  fg->id, fte);
+		if (err)
+			pr_warn("%s can't del rule fg id=%d fte_index=%d\n",
+				__func__, fg->id, fte->index);
+	}
+	kvfree(match_value);
+}
+
+static void del_fte(struct fs_node *node)
+{
+	struct mlx5_flow_table *ft;
+	struct mlx5_flow_group *fg;
+	struct mlx5_core_dev *dev;
+	struct fs_fte *fte;
+	int err;
+
+	fs_get_obj(fte, node);
+	fs_get_obj(fg, fte->node.parent);
+	fs_get_obj(ft, fg->node.parent);
+
+	dev = get_dev(&ft->node);
+	err = mlx5_cmd_delete_fte(dev, ft,
+				  fte->index);
+	if (err)
+		pr_warn("flow steering can't delete fte in index %d of flow group id %d\n",
+			fte->index, fg->id);
+
+	fte->status = 0;
+	fg->num_ftes--;
+}
+
+static void del_flow_group(struct fs_node *node)
+{
+	struct mlx5_flow_group *fg;
+	struct mlx5_flow_table *ft;
+	struct mlx5_core_dev *dev;
+
+	fs_get_obj(fg, node);
+	fs_get_obj(ft, fg->node.parent);
+	dev = get_dev(&ft->node);
+
+	if (mlx5_cmd_destroy_flow_group(dev, ft, fg->id))
+		pr_warn("flow steering can't destroy fg %d of ft %d\n",
+			fg->id, ft->id);
+}
+
+static struct fs_fte *alloc_fte(u8 action,
+				u32 flow_tag,
+				u32 *match_value,
+				unsigned int index)
+{
+	struct fs_fte *fte;
+
+	fte = kzalloc(sizeof(*fte), GFP_KERNEL);
+	if (!fte)
+		return ERR_PTR(-ENOMEM);
+
+	memcpy(fte->val, match_value, sizeof(fte->val));
+	fte->node.type =  FS_TYPE_FLOW_ENTRY;
+	fte->flow_tag = flow_tag;
+	fte->index = index;
+	fte->action = action;
+
+	return fte;
+}
+
+static struct mlx5_flow_group *alloc_flow_group(u32 *create_fg_in)
+{
+	struct mlx5_flow_group *fg;
+	void *match_criteria = MLX5_ADDR_OF(create_flow_group_in,
+					    create_fg_in, match_criteria);
+	u8 match_criteria_enable = MLX5_GET(create_flow_group_in,
+					    create_fg_in,
+					    match_criteria_enable);
+	fg = kzalloc(sizeof(*fg), GFP_KERNEL);
+	if (!fg)
+		return ERR_PTR(-ENOMEM);
+
+	fg->mask.match_criteria_enable = match_criteria_enable;
+	memcpy(&fg->mask.match_criteria, match_criteria,
+	       sizeof(fg->mask.match_criteria));
+	fg->node.type =  FS_TYPE_FLOW_GROUP;
+	fg->start_index = MLX5_GET(create_flow_group_in, create_fg_in,
+				   start_flow_index);
+	fg->max_ftes = MLX5_GET(create_flow_group_in, create_fg_in,
+				end_flow_index) - fg->start_index + 1;
+	return fg;
+}
+
+static struct mlx5_flow_table *alloc_flow_table(int level, int max_fte,
+						enum fs_flow_table_type table_type)
+{
+	struct mlx5_flow_table *ft;
+
+	ft  = kzalloc(sizeof(*ft), GFP_KERNEL);
+	if (!ft)
+		return NULL;
+
+	ft->level = level;
+	ft->node.type = FS_TYPE_FLOW_TABLE;
+	ft->type = table_type;
+	ft->max_fte = max_fte;
+
+	return ft;
+}
+
+struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns,
+					       int prio,
+					       int max_fte)
+{
+	struct mlx5_flow_table *ft;
+	int err;
+	int log_table_sz;
+	struct mlx5_flow_root_namespace *root =
+		find_root(&ns->node);
+	struct fs_prio *fs_prio = NULL;
+
+	if (!root) {
+		pr_err("mlx5: flow steering failed to find root of namespace\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	fs_prio = find_prio(ns, prio);
+	if (!fs_prio)
+		return ERR_PTR(-EINVAL);
+
+	lock_ref_node(&fs_prio->node);
+	if (fs_prio->num_ft == fs_prio->max_ft) {
+		err = -ENOSPC;
+		goto unlock_prio;
+	}
+
+	ft = alloc_flow_table(find_next_free_level(fs_prio),
+			      roundup_pow_of_two(max_fte),
+			      root->table_type);
+	if (!ft) {
+		err = -ENOMEM;
+		goto unlock_prio;
+	}
+
+	tree_init_node(&ft->node, 1, del_flow_table);
+	log_table_sz = ilog2(ft->max_fte);
+	err = mlx5_cmd_create_flow_table(root->dev, ft->type, ft->level,
+					 log_table_sz, &ft->id);
+	if (err)
+		goto free_ft;
+
+	tree_add_node(&ft->node, &fs_prio->node);
+	list_add_tail(&ft->node.list, &fs_prio->node.children);
+	fs_prio->num_ft++;
+	unlock_ref_node(&fs_prio->node);
+
+	return ft;
+
+free_ft:
+	kfree(ft);
+unlock_prio:
+	unlock_ref_node(&fs_prio->node);
+	return ERR_PTR(err);
+}
+
+struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft,
+					       u32 *fg_in)
+{
+	struct mlx5_flow_group *fg;
+	struct mlx5_core_dev *dev = get_dev(&ft->node);
+	int err;
+
+	if (!dev)
+		return ERR_PTR(-ENODEV);
+
+	fg = alloc_flow_group(fg_in);
+	if (IS_ERR(fg))
+		return fg;
+
+	lock_ref_node(&ft->node);
+	err = mlx5_cmd_create_flow_group(dev, ft, fg_in, &fg->id);
+	if (err) {
+		kfree(fg);
+		unlock_ref_node(&ft->node);
+		return ERR_PTR(err);
+	}
+	/* Add node to tree */
+	tree_init_node(&fg->node, 1, del_flow_group);
+	tree_add_node(&fg->node, &ft->node);
+	/* Add node to group list */
+	list_add(&fg->node.list, ft->node.children.prev);
+	unlock_ref_node(&ft->node);
+
+	return fg;
+}
+
+static struct mlx5_flow_rule *alloc_rule(struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_rule *rule;
+
+	rule = kzalloc(sizeof(*rule), GFP_KERNEL);
+	if (!rule)
+		return NULL;
+
+	rule->node.type = FS_TYPE_FLOW_DEST;
+	memcpy(&rule->dest_attr, dest, sizeof(*dest));
+
+	return rule;
+}
+
+/* fte should not be deleted while calling this function */
+static struct mlx5_flow_rule *add_rule_fte(struct fs_fte *fte,
+					   struct mlx5_flow_group *fg,
+					   struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_table *ft;
+	struct mlx5_flow_rule *rule;
+	int err;
+
+	rule = alloc_rule(dest);
+	if (!rule)
+		return ERR_PTR(-ENOMEM);
+
+	fs_get_obj(ft, fg->node.parent);
+	/* Add dest to dests list- added as first element after the head */
+	tree_init_node(&rule->node, 1, del_rule);
+	list_add_tail(&rule->node.list, &fte->node.children);
+	fte->dests_size++;
+	if (fte->dests_size == 1)
+		err = mlx5_cmd_create_fte(get_dev(&ft->node),
+					  ft, fg->id, fte);
+	else
+		err = mlx5_cmd_update_fte(get_dev(&ft->node),
+					  ft, fg->id, fte);
+	if (err)
+		goto free_rule;
+
+	fte->status |= FS_FTE_STATUS_EXISTING;
+
+	return rule;
+
+free_rule:
+	list_del(&rule->node.list);
+	kfree(rule);
+	fte->dests_size--;
+	return ERR_PTR(err);
+}
+
+/* Assumed fg is locked */
+static unsigned int get_free_fte_index(struct mlx5_flow_group *fg,
+				       struct list_head **prev)
+{
+	struct fs_fte *fte;
+	unsigned int start = fg->start_index;
+
+	if (prev)
+		*prev = &fg->node.children;
+
+	/* assumed list is sorted by index */
+	fs_for_each_fte(fte, fg) {
+		if (fte->index != start)
+			return start;
+		start++;
+		if (prev)
+			*prev = &fte->node.list;
+	}
+
+	return start;
+}
+
+/* prev is output, prev->next = new_fte */
+static struct fs_fte *create_fte(struct mlx5_flow_group *fg,
+				 u32 *match_value,
+				 u8 action,
+				 u32 flow_tag,
+				 struct list_head **prev)
+{
+	struct fs_fte *fte;
+	int index;
+
+	index = get_free_fte_index(fg, prev);
+	fte = alloc_fte(action, flow_tag, match_value, index);
+	if (IS_ERR(fte))
+		return fte;
+
+	return fte;
+}
+
+/* Assuming parent fg(flow table) is locked */
+static struct mlx5_flow_rule *add_rule_fg(struct mlx5_flow_group *fg,
+					  u32 *match_value,
+					  u8 action,
+					  u32 flow_tag,
+					  struct mlx5_flow_destination *dest)
+{
+	struct fs_fte *fte;
+	struct mlx5_flow_rule *rule;
+	struct mlx5_flow_table *ft;
+	struct list_head *prev;
+
+	lock_ref_node(&fg->node);
+	fs_for_each_fte(fte, fg) {
+		nested_lock_ref_node(&fte->node);
+		if (compare_match_value(&fg->mask, match_value, &fte->val) &&
+		    action == fte->action && flow_tag == fte->flow_tag) {
+			rule = add_rule_fte(fte, fg, dest);
+			unlock_ref_node(&fte->node);
+			if (IS_ERR(rule))
+				goto unlock_fg;
+			else
+				goto add_rule;
+		}
+		unlock_ref_node(&fte->node);
+	}
+	fs_get_obj(ft, fg->node.parent);
+	if (fg->num_ftes >= fg->max_ftes) {
+		rule = ERR_PTR(-ENOSPC);
+		goto unlock_fg;
+	}
+
+	fte = create_fte(fg, match_value, action, flow_tag, &prev);
+	if (IS_ERR(fte)) {
+		rule = (void *)fte;
+		goto unlock_fg;
+	}
+	tree_init_node(&fte->node, 0, del_fte);
+	rule = add_rule_fte(fte, fg, dest);
+	if (IS_ERR(rule)) {
+		kfree(fte);
+		goto unlock_fg;
+	}
+
+	fg->num_ftes++;
+
+	tree_add_node(&fte->node, &fg->node);
+	list_add(&fte->node.list, prev);
+add_rule:
+	tree_add_node(&rule->node, &fte->node);
+unlock_fg:
+	unlock_ref_node(&fg->node);
+	return rule;
+}
+
+struct mlx5_flow_rule *
+mlx5_add_flow_rule(struct mlx5_flow_table *ft,
+		   u8 match_criteria_enable,
+		   u32 *match_criteria,
+		   u32 *match_value,
+		   u32 action,
+		   u32 flow_tag,
+		   struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_group *g;
+	struct mlx5_flow_rule *rule = ERR_PTR(-EINVAL);
+
+	tree_get_node(&ft->node);
+	lock_ref_node(&ft->node);
+	fs_for_each_fg(g, ft)
+		if (compare_match_criteria(g->mask.match_criteria_enable,
+					   match_criteria_enable,
+					   g->mask.match_criteria,
+					   match_criteria)) {
+			unlock_ref_node(&ft->node);
+			rule = add_rule_fg(g, match_value,
+					   action, flow_tag, dest);
+			goto put;
+		}
+	unlock_ref_node(&ft->node);
+put:
+	tree_put_node(&ft->node);
+	return rule;
+}
+
+void mlx5_del_flow_rule(struct mlx5_flow_rule *rule)
+{
+	tree_remove_node(&rule->node);
+}
+
+int mlx5_destroy_flow_table(struct mlx5_flow_table *ft)
+{
+	if (tree_remove_node(&ft->node))
+		mlx5_core_warn(get_dev(&ft->node), "Flow table %d wasn't destroyed, refcount > 1\n",
+			       ft->id);
+
+	return 0;
+}
+
+void mlx5_destroy_flow_group(struct mlx5_flow_group *fg)
+{
+	if (tree_remove_node(&fg->node))
+		mlx5_core_warn(get_dev(&fg->node), "Flow group %d wasn't destroyed, refcount > 1\n",
+			       fg->id);
+}
+
+struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
+						    enum mlx5_flow_namespace_type type)
+{
+	struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns;
+	int prio;
+	static struct fs_prio *fs_prio;
+	struct mlx5_flow_namespace *ns;
+
+	if (!root_ns)
+		return NULL;
+
+	switch (type) {
+	case MLX5_FLOW_NAMESPACE_KERNEL:
+		prio = 0;
+		break;
+	case MLX5_FLOW_NAMESPACE_FDB:
+		if (dev->priv.fdb_root_ns)
+			return &dev->priv.fdb_root_ns->ns;
+		else
+			return NULL;
+	default:
+		return NULL;
+	}
+
+	fs_prio = find_prio(&root_ns->ns, prio);
+	if (!fs_prio)
+		return NULL;
+
+	ns = list_first_entry(&fs_prio->node.children,
+			      typeof(*ns),
+			      node.list);
+
+	return ns;
+}
+
+static struct fs_prio *fs_create_prio(struct mlx5_flow_namespace *ns,
+				      unsigned prio, int max_ft,
+				      int start_level)
+{
+	struct fs_prio *fs_prio;
+
+	fs_prio = kzalloc(sizeof(*fs_prio), GFP_KERNEL);
+	if (!fs_prio)
+		return ERR_PTR(-ENOMEM);
+
+	fs_prio->node.type = FS_TYPE_PRIO;
+	tree_init_node(&fs_prio->node, 1, NULL);
+	tree_add_node(&fs_prio->node, &ns->node);
+	fs_prio->max_ft = max_ft;
+	fs_prio->prio = prio;
+	fs_prio->start_level = start_level;
+	list_add_tail(&fs_prio->node.list, &ns->node.children);
+
+	return fs_prio;
+}
+
+static struct mlx5_flow_namespace *fs_init_namespace(struct mlx5_flow_namespace
+						     *ns)
+{
+	ns->node.type = FS_TYPE_NAMESPACE;
+
+	return ns;
+}
+
+static struct mlx5_flow_namespace *fs_create_namespace(struct fs_prio *prio)
+{
+	struct mlx5_flow_namespace	*ns;
+
+	ns = kzalloc(sizeof(*ns), GFP_KERNEL);
+	if (!ns)
+		return ERR_PTR(-ENOMEM);
+
+	fs_init_namespace(ns);
+	tree_init_node(&ns->node, 1, NULL);
+	tree_add_node(&ns->node, &prio->node);
+	list_add_tail(&ns->node.list, &prio->node.children);
+
+	return ns;
+}
+
+static int init_root_tree_recursive(int max_ft_level, struct init_tree_node *init_node,
+				    struct fs_node *fs_parent_node,
+				    struct init_tree_node *init_parent_node,
+				    int index)
+{
+	struct mlx5_flow_namespace *fs_ns;
+	struct fs_prio *fs_prio;
+	struct fs_node *base;
+	int i;
+	int err;
+
+	if (init_node->type == FS_TYPE_PRIO) {
+		if (init_node->min_ft_level > max_ft_level)
+			return -ENOTSUPP;
+
+		fs_get_obj(fs_ns, fs_parent_node);
+		fs_prio = fs_create_prio(fs_ns, index, init_node->max_ft,
+					 init_node->start_level);
+		if (IS_ERR(fs_prio))
+			return PTR_ERR(fs_prio);
+		base = &fs_prio->node;
+	} else if (init_node->type == FS_TYPE_NAMESPACE) {
+		fs_get_obj(fs_prio, fs_parent_node);
+		fs_ns = fs_create_namespace(fs_prio);
+		if (IS_ERR(fs_ns))
+			return PTR_ERR(fs_ns);
+		base = &fs_ns->node;
+	} else {
+		return -EINVAL;
+	}
+	for (i = 0; i < init_node->ar_size; i++) {
+		err = init_root_tree_recursive(max_ft_level,
+					       &init_node->children[i], base,
+					       init_node, i);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int init_root_tree(int max_ft_level, struct init_tree_node *init_node,
+			  struct fs_node *fs_parent_node)
+{
+	int i;
+	struct mlx5_flow_namespace *fs_ns;
+	int err;
+
+	fs_get_obj(fs_ns, fs_parent_node);
+	for (i = 0; i < init_node->ar_size; i++) {
+		err = init_root_tree_recursive(max_ft_level,
+					       &init_node->children[i],
+					       &fs_ns->node,
+					       init_node, i);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev,
+						       enum fs_flow_table_type
+						       table_type)
+{
+	struct mlx5_flow_root_namespace *root_ns;
+	struct mlx5_flow_namespace *ns;
+
+	/* Create the root namespace */
+	root_ns = mlx5_vzalloc(sizeof(*root_ns));
+	if (!root_ns)
+		return NULL;
+
+	root_ns->dev = dev;
+	root_ns->table_type = table_type;
+
+	ns = &root_ns->ns;
+	fs_init_namespace(ns);
+	tree_init_node(&ns->node, 1, NULL);
+	tree_add_node(&ns->node, NULL);
+
+	return root_ns;
+}
+
+static int init_root_ns(struct mlx5_core_dev *dev)
+{
+	int max_ft_level = MLX5_CAP_FLOWTABLE(dev,
+					      flow_table_properties_nic_receive.
+					      max_ft_level);
+
+	dev->priv.root_ns = create_root_ns(dev, FS_FT_NIC_RX);
+	if (IS_ERR_OR_NULL(dev->priv.root_ns))
+		goto cleanup;
+
+	if (init_root_tree(max_ft_level, &root_fs, &dev->priv.root_ns->ns.node))
+		goto cleanup;
+
+	return 0;
+
+cleanup:
+	mlx5_cleanup_fs(dev);
+	return -ENOMEM;
+}
+
+static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev,
+					struct mlx5_flow_root_namespace *root_ns)
+{
+	struct fs_node *prio;
+
+	if (!root_ns)
+		return;
+
+	if (!list_empty(&root_ns->ns.node.children)) {
+		prio = list_first_entry(&root_ns->ns.node.children,
+					struct fs_node,
+				 list);
+		if (tree_remove_node(prio))
+			mlx5_core_warn(dev,
+				       "Flow steering priority wasn't destroyed, refcount > 1\n");
+	}
+	if (tree_remove_node(&root_ns->ns.node))
+		mlx5_core_warn(dev,
+			       "Flow steering namespace wasn't destroyed, refcount > 1\n");
+	root_ns = NULL;
+}
+
+static void cleanup_root_ns(struct mlx5_core_dev *dev)
+{
+	struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns;
+	struct fs_prio *iter_prio;
+
+	if (!MLX5_CAP_GEN(dev, nic_flow_table))
+		return;
+
+	if (!root_ns)
+		return;
+
+	/* stage 1 */
+	fs_for_each_prio(iter_prio, &root_ns->ns) {
+		struct fs_node *node;
+		struct mlx5_flow_namespace *iter_ns;
+
+		fs_for_each_ns_or_ft(node, iter_prio) {
+			if (node->type == FS_TYPE_FLOW_TABLE)
+				continue;
+			fs_get_obj(iter_ns, node);
+			while (!list_empty(&iter_ns->node.children)) {
+				struct fs_prio *obj_iter_prio2;
+				struct fs_node *iter_prio2 =
+					list_first_entry(&iter_ns->node.children,
+							 struct fs_node,
+							 list);
+
+				fs_get_obj(obj_iter_prio2, iter_prio2);
+				if (tree_remove_node(iter_prio2)) {
+					mlx5_core_warn(dev,
+						       "Priority %d wasn't destroyed, refcount > 1\n",
+						       obj_iter_prio2->prio);
+					return;
+				}
+			}
+		}
+	}
+
+	/* stage 2 */
+	fs_for_each_prio(iter_prio, &root_ns->ns) {
+		while (!list_empty(&iter_prio->node.children)) {
+			struct fs_node *iter_ns =
+				list_first_entry(&iter_prio->node.children,
+						 struct fs_node,
+						 list);
+			if (tree_remove_node(iter_ns)) {
+				mlx5_core_warn(dev,
+					       "Namespace wasn't destroyed, refcount > 1\n");
+				return;
+			}
+		}
+	}
+
+	/* stage 3 */
+	while (!list_empty(&root_ns->ns.node.children)) {
+		struct fs_prio *obj_prio_node;
+		struct fs_node *prio_node =
+			list_first_entry(&root_ns->ns.node.children,
+					 struct fs_node,
+					 list);
+
+		fs_get_obj(obj_prio_node, prio_node);
+		if (tree_remove_node(prio_node)) {
+			mlx5_core_warn(dev,
+				       "Priority %d wasn't destroyed, refcount > 1\n",
+				       obj_prio_node->prio);
+			return;
+		}
+	}
+
+	if (tree_remove_node(&root_ns->ns.node)) {
+		mlx5_core_warn(dev,
+			       "root namespace wasn't destroyed, refcount > 1\n");
+		return;
+	}
+
+	dev->priv.root_ns = NULL;
+}
+
+void mlx5_cleanup_fs(struct mlx5_core_dev *dev)
+{
+	cleanup_root_ns(dev);
+	cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns);
+}
+
+static int init_fdb_root_ns(struct mlx5_core_dev *dev)
+{
+	struct fs_prio *prio;
+
+	dev->priv.fdb_root_ns = create_root_ns(dev, FS_FT_FDB);
+	if (!dev->priv.fdb_root_ns)
+		return -ENOMEM;
+
+	/* Create single prio */
+	prio = fs_create_prio(&dev->priv.fdb_root_ns->ns, 0, 1, 0);
+	if (IS_ERR(prio)) {
+		cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns);
+		return PTR_ERR(prio);
+	} else {
+		return 0;
+	}
+}
+
+int mlx5_init_fs(struct mlx5_core_dev *dev)
+{
+	int err = 0;
+
+	if (MLX5_CAP_GEN(dev, nic_flow_table)) {
+		err = init_root_ns(dev);
+		if (err)
+			return err;
+	}
+	if (MLX5_CAP_GEN(dev, eswitch_flow_table)) {
+		err = init_fdb_root_ns(dev);
+		if (err)
+			cleanup_root_ns(dev);
+	}
+
+	return err;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
new file mode 100644
index 000000000000..4ebb97fd5544
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -0,0 +1,155 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef _MLX5_FS_CORE_
+#define _MLX5_FS_CORE_
+
+#include <linux/mlx5/fs.h>
+
+enum fs_node_type {
+	FS_TYPE_NAMESPACE,
+	FS_TYPE_PRIO,
+	FS_TYPE_FLOW_TABLE,
+	FS_TYPE_FLOW_GROUP,
+	FS_TYPE_FLOW_ENTRY,
+	FS_TYPE_FLOW_DEST
+};
+
+enum fs_flow_table_type {
+	FS_FT_NIC_RX	 = 0x0,
+	FS_FT_FDB	 = 0X4,
+};
+
+enum fs_fte_status {
+	FS_FTE_STATUS_EXISTING = 1UL << 0,
+};
+
+struct fs_node {
+	struct list_head	list;
+	struct list_head	children;
+	enum fs_node_type	type;
+	struct fs_node		*parent;
+	struct fs_node		*root;
+	/* lock the node for writing and traversing */
+	struct mutex		lock;
+	atomic_t		refcount;
+	void			(*remove_func)(struct fs_node *);
+};
+
+struct mlx5_flow_rule {
+	struct fs_node				node;
+	struct mlx5_flow_destination		dest_attr;
+};
+
+/* Type of children is mlx5_flow_group */
+struct mlx5_flow_table {
+	struct fs_node			node;
+	u32				id;
+	unsigned int			max_fte;
+	unsigned int			level;
+	enum fs_flow_table_type		type;
+};
+
+/* Type of children is mlx5_flow_rule */
+struct fs_fte {
+	struct fs_node			node;
+	u32				val[MLX5_ST_SZ_DW(fte_match_param)];
+	u32				dests_size;
+	u32				flow_tag;
+	u32				index;
+	u32				action;
+	enum fs_fte_status		status;
+};
+
+/* Type of children is mlx5_flow_table/namespace */
+struct fs_prio {
+	struct fs_node			node;
+	unsigned int			max_ft;
+	unsigned int			start_level;
+	unsigned int			prio;
+	unsigned int			num_ft;
+};
+
+/* Type of children is fs_prio */
+struct mlx5_flow_namespace {
+	/* parent == NULL => root ns */
+	struct	fs_node			node;
+};
+
+struct mlx5_flow_group_mask {
+	u8	match_criteria_enable;
+	u32	match_criteria[MLX5_ST_SZ_DW(fte_match_param)];
+};
+
+/* Type of children is fs_fte */
+struct mlx5_flow_group {
+	struct fs_node			node;
+	struct mlx5_flow_group_mask	mask;
+	u32				start_index;
+	u32				max_ftes;
+	u32				num_ftes;
+	u32				id;
+};
+
+struct mlx5_flow_root_namespace {
+	struct mlx5_flow_namespace	ns;
+	enum   fs_flow_table_type	table_type;
+	struct mlx5_core_dev		*dev;
+};
+
+int mlx5_init_fs(struct mlx5_core_dev *dev);
+void mlx5_cleanup_fs(struct mlx5_core_dev *dev);
+
+#define fs_get_obj(v, _node)  {v = container_of((_node), typeof(*v), node); }
+
+#define fs_list_for_each_entry(pos, root)		\
+	list_for_each_entry(pos, root, node.list)
+
+#define fs_for_each_ns_or_ft_reverse(pos, prio)				\
+	list_for_each_entry_reverse(pos, &(prio)->node.children, list)
+
+#define fs_for_each_ns_or_ft(pos, prio)					\
+	list_for_each_entry(pos, (&(prio)->node.children), list)
+
+#define fs_for_each_prio(pos, ns)			\
+	fs_list_for_each_entry(pos, &(ns)->node.children)
+
+#define fs_for_each_fg(pos, ft)			\
+	fs_list_for_each_entry(pos, &(ft)->node.children)
+
+#define fs_for_each_fte(pos, fg)			\
+	fs_list_for_each_entry(pos, &(fg)->node.children)
+
+#define fs_for_each_dst(pos, fte)			\
+	fs_list_for_each_entry(pos, &(fte)->node.children)
+
+#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
index 1c9f9a54a873..aa1ab4702385 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
@@ -173,7 +173,7 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev)
 			return err;
 	}
 
-	if (MLX5_CAP_GEN(dev, vport_group_manager)) {
+	if (MLX5_CAP_GEN(dev, eswitch_flow_table)) {
 		err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH,
 					 HCA_CAP_OPMOD_GET_CUR);
 		if (err)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index c6de3240f76f..789882b7b711 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -49,6 +49,7 @@
 #include <linux/delay.h>
 #include <linux/mlx5/mlx5_ifc.h>
 #include "mlx5_core.h"
+#include "fs_core.h"
 #ifdef CONFIG_MLX5_CORE_EN
 #include "eswitch.h"
 #endif
@@ -1055,6 +1056,11 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 	mlx5_init_srq_table(dev);
 	mlx5_init_mr_table(dev);
 
+	err = mlx5_init_fs(dev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to init flow steering\n");
+		goto err_fs;
+	}
 #ifdef CONFIG_MLX5_CORE_EN
 	err = mlx5_eswitch_init(dev);
 	if (err) {
@@ -1093,6 +1099,8 @@ err_sriov:
 	mlx5_eswitch_cleanup(dev->priv.eswitch);
 #endif
 err_reg_dev:
+	mlx5_cleanup_fs(dev);
+err_fs:
 	mlx5_cleanup_mr_table(dev);
 	mlx5_cleanup_srq_table(dev);
 	mlx5_cleanup_qp_table(dev);
@@ -1165,6 +1173,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 	mlx5_eswitch_cleanup(dev->priv.eswitch);
 #endif
 
+	mlx5_cleanup_fs(dev);
 	mlx5_cleanup_mr_table(dev);
 	mlx5_cleanup_srq_table(dev);
 	mlx5_cleanup_qp_table(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index bee7da822dfe..ea6a137fd76c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -65,6 +65,9 @@ do {									\
 		(__dev)->priv.name, __func__, __LINE__, current->pid,	\
 		##__VA_ARGS__)
 
+#define mlx5_core_info(__dev, format, ...)				\
+	dev_info(&(__dev)->pdev->dev, format, ##__VA_ARGS__)
+
 enum {
 	MLX5_CMD_DATA, /* print command payload only */
 	MLX5_CMD_TIME, /* print command execution time */
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index ac098b6b97bf..2fd7019f69db 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -502,6 +502,8 @@ struct mlx5_priv {
 	struct mlx5_eswitch     *eswitch;
 	struct mlx5_core_sriov	sriov;
 	unsigned long		pci_dev_data;
+	struct mlx5_flow_root_namespace *root_ns;
+	struct mlx5_flow_root_namespace *fdb_root_ns;
 };
 
 enum mlx5_device_state {
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
new file mode 100644
index 000000000000..bc7ad019afde
--- /dev/null
+++ b/include/linux/mlx5/fs.h
@@ -0,0 +1,93 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef _MLX5_FS_
+#define _MLX5_FS_
+
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/mlx5_ifc.h>
+
+#define MLX5_FS_DEFAULT_FLOW_TAG 0x0
+
+enum mlx5_flow_namespace_type {
+	MLX5_FLOW_NAMESPACE_KERNEL,
+	MLX5_FLOW_NAMESPACE_FDB,
+};
+
+struct mlx5_flow_table;
+struct mlx5_flow_group;
+struct mlx5_flow_rule;
+struct mlx5_flow_namespace;
+
+struct mlx5_flow_destination {
+	enum mlx5_flow_destination_type	type;
+	union {
+		u32			tir_num;
+		struct mlx5_flow_table	*ft;
+		u32			vport_num;
+	};
+};
+
+struct mlx5_flow_namespace *
+mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
+			enum mlx5_flow_namespace_type type);
+
+struct mlx5_flow_table *
+mlx5_create_flow_table(struct mlx5_flow_namespace *ns,
+		       int prio,
+		       int num_flow_table_entries);
+int mlx5_destroy_flow_table(struct mlx5_flow_table *ft);
+
+/* inbox should be set with the following values:
+ * start_flow_index
+ * end_flow_index
+ * match_criteria_enable
+ * match_criteria
+ */
+struct mlx5_flow_group *
+mlx5_create_flow_group(struct mlx5_flow_table *ft, u32 *in);
+void mlx5_destroy_flow_group(struct mlx5_flow_group *fg);
+
+/* Single destination per rule.
+ * Group ID is implied by the match criteria.
+ */
+struct mlx5_flow_rule *
+mlx5_add_flow_rule(struct mlx5_flow_table *ft,
+		   u8 match_criteria_enable,
+		   u32 *match_criteria,
+		   u32 *match_value,
+		   u32 action,
+		   u32 flow_tag,
+		   struct mlx5_flow_destination *dest);
+void mlx5_del_flow_rule(struct mlx5_flow_rule *fr);
+
+#endif
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index f5d94495758a..131a2737cfa3 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -256,25 +256,27 @@ struct mlx5_ifc_flow_table_fields_supported_bits {
 
 struct mlx5_ifc_flow_table_prop_layout_bits {
 	u8         ft_support[0x1];
-	u8         reserved_0[0x1f];
+	u8         reserved_0[0x2];
+	u8	   flow_modify_en[0x1];
+	u8         reserved_1[0x1c];
 
-	u8         reserved_1[0x2];
+	u8         reserved_2[0x2];
 	u8         log_max_ft_size[0x6];
-	u8         reserved_2[0x10];
+	u8         reserved_3[0x10];
 	u8         max_ft_level[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_4[0x20];
 
-	u8         reserved_4[0x18];
+	u8         reserved_5[0x18];
 	u8         log_max_ft_num[0x8];
 
-	u8         reserved_5[0x18];
+	u8         reserved_6[0x18];
 	u8         log_max_destination[0x8];
 
-	u8         reserved_6[0x18];
+	u8         reserved_7[0x18];
 	u8         log_max_flow[0x8];
 
-	u8         reserved_7[0x40];
+	u8         reserved_8[0x40];
 
 	struct mlx5_ifc_flow_table_fields_supported_bits ft_field_support;
 
@@ -2843,6 +2845,13 @@ struct mlx5_ifc_set_hca_cap_in_bits {
 	union mlx5_ifc_hca_cap_union_bits capability;
 };
 
+enum {
+	MLX5_SET_FTE_MODIFY_ENABLE_MASK_ACTION    = 0x0,
+	MLX5_SET_FTE_MODIFY_ENABLE_MASK_FLOW_TAG  = 0x1,
+	MLX5_SET_FTE_MODIFY_ENABLE_MASK_DESTINATION_LIST    = 0x2,
+	MLX5_SET_FTE_MODIFY_ENABLE_MASK_FLOW_COUNTERS    = 0x3
+};
+
 struct mlx5_ifc_set_fte_out_bits {
 	u8         status[0x8];
 	u8         reserved_0[0x18];
@@ -2867,11 +2876,14 @@ struct mlx5_ifc_set_fte_in_bits {
 	u8         reserved_4[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x40];
+	u8         reserved_5[0x18];
+	u8         modify_enable_mask[0x8];
+
+	u8         reserved_6[0x20];
 
 	u8         flow_index[0x20];
 
-	u8         reserved_6[0xe0];
+	u8         reserved_7[0xe0];
 
 	struct mlx5_ifc_flow_context_bits flow_context;
 };