diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 716558476eda8c..5708fcc079ca1d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -73,4 +73,4 @@ mlx5_core-$(CONFIG_MLX5_SW_STEERING) += steering/dr_domain.o steering/dr_table.o
 					steering/dr_icm_pool.o steering/dr_crc32.o \
 					steering/dr_ste.o steering/dr_send.o \
 					steering/dr_cmd.o steering/dr_fw.o \
-					steering/dr_action.o
+					steering/dr_action.o steering/fs_dr.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index 488f50dfb40464..579c306caa7b10 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -135,6 +135,22 @@ static void mlx5_cmd_stub_modify_header_dealloc(struct mlx5_flow_root_namespace
 {
 }
 
+static int mlx5_cmd_stub_set_peer(struct mlx5_flow_root_namespace *ns,
+				  struct mlx5_flow_root_namespace *peer_ns)
+{
+	return 0;
+}
+
+static int mlx5_cmd_stub_create_ns(struct mlx5_flow_root_namespace *ns)
+{
+	return 0;
+}
+
+static int mlx5_cmd_stub_destroy_ns(struct mlx5_flow_root_namespace *ns)
+{
+	return 0;
+}
+
 static int mlx5_cmd_update_root_ft(struct mlx5_flow_root_namespace *ns,
 				   struct mlx5_flow_table *ft, u32 underlay_qpn,
 				   bool disconnect)
@@ -838,7 +854,10 @@ static const struct mlx5_flow_cmds mlx5_flow_cmds = {
 	.packet_reformat_alloc = mlx5_cmd_packet_reformat_alloc,
 	.packet_reformat_dealloc = mlx5_cmd_packet_reformat_dealloc,
 	.modify_header_alloc = mlx5_cmd_modify_header_alloc,
-	.modify_header_dealloc = mlx5_cmd_modify_header_dealloc
+	.modify_header_dealloc = mlx5_cmd_modify_header_dealloc,
+	.set_peer = mlx5_cmd_stub_set_peer,
+	.create_ns = mlx5_cmd_stub_create_ns,
+	.destroy_ns = mlx5_cmd_stub_destroy_ns,
 };
 
 static const struct mlx5_flow_cmds mlx5_flow_cmd_stubs = {
@@ -854,10 +873,13 @@ static const struct mlx5_flow_cmds mlx5_flow_cmd_stubs = {
 	.packet_reformat_alloc = mlx5_cmd_stub_packet_reformat_alloc,
 	.packet_reformat_dealloc = mlx5_cmd_stub_packet_reformat_dealloc,
 	.modify_header_alloc = mlx5_cmd_stub_modify_header_alloc,
-	.modify_header_dealloc = mlx5_cmd_stub_modify_header_dealloc
+	.modify_header_dealloc = mlx5_cmd_stub_modify_header_dealloc,
+	.set_peer = mlx5_cmd_stub_set_peer,
+	.create_ns = mlx5_cmd_stub_create_ns,
+	.destroy_ns = mlx5_cmd_stub_destroy_ns,
 };
 
-static const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void)
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void)
 {
 	return &mlx5_flow_cmds;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
index 3268654d6748ca..d62de642eca99a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
@@ -93,6 +93,12 @@ struct mlx5_flow_cmds {
 
 	void (*modify_header_dealloc)(struct mlx5_flow_root_namespace *ns,
 				      struct mlx5_modify_hdr *modify_hdr);
+
+	int (*set_peer)(struct mlx5_flow_root_namespace *ns,
+			struct mlx5_flow_root_namespace *peer_ns);
+
+	int (*create_ns)(struct mlx5_flow_root_namespace *ns);
+	int (*destroy_ns)(struct mlx5_flow_root_namespace *ns);
 };
 
 int mlx5_cmd_fc_alloc(struct mlx5_core_dev *dev, u32 *id);
@@ -108,5 +114,6 @@ int mlx5_cmd_fc_bulk_query(struct mlx5_core_dev *dev, u32 base_id, int bulk_len,
 			   u32 *out);
 
 const struct mlx5_flow_cmds *mlx5_fs_cmd_get_default(enum fs_flow_table_type type);
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void);
 
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 1d2333fd308011..c2d6e9f4cb90bd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2991,3 +2991,9 @@ void mlx5_packet_reformat_dealloc(struct mlx5_core_dev *dev,
 	kfree(pkt_reformat);
 }
 EXPORT_SYMBOL(mlx5_packet_reformat_dealloc);
+
+int mlx5_flow_namespace_set_peer(struct mlx5_flow_root_namespace *ns,
+				 struct mlx5_flow_root_namespace *peer_ns)
+{
+	return ns->cmds->set_peer(ns, peer_ns);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index ea0f221685abb1..a133ec5487aee0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -37,16 +37,23 @@
 #include <linux/mlx5/fs.h>
 #include <linux/rhashtable.h>
 #include <linux/llist.h>
+#include <steering/fs_dr.h>
 
 struct mlx5_modify_hdr {
 	enum mlx5_flow_namespace_type ns_type;
-	u32 id;
+	union {
+		struct mlx5_fs_dr_action action;
+		u32 id;
+	};
 };
 
 struct mlx5_pkt_reformat {
 	enum mlx5_flow_namespace_type ns_type;
 	int reformat_type; /* from mlx5_ifc */
-	u32 id;
+	union {
+		struct mlx5_fs_dr_action action;
+		u32 id;
+	};
 };
 
 /* FS_TYPE_PRIO_CHAINS is a PRIO that will have namespaces only,
@@ -139,6 +146,7 @@ struct mlx5_flow_handle {
 /* Type of children is mlx5_flow_group */
 struct mlx5_flow_table {
 	struct fs_node			node;
+	struct mlx5_fs_dr_table		fs_dr_table;
 	u32				id;
 	u16				vport;
 	unsigned int			max_fte;
@@ -179,6 +187,7 @@ struct mlx5_ft_underlay_qp {
 /* Type of children is mlx5_flow_rule */
 struct fs_fte {
 	struct fs_node			node;
+	struct mlx5_fs_dr_rule		fs_dr_rule;
 	u32				val[MLX5_ST_SZ_DW_MATCH_PARAM];
 	u32				dests_size;
 	u32				index;
@@ -214,6 +223,7 @@ struct mlx5_flow_group_mask {
 /* Type of children is fs_fte */
 struct mlx5_flow_group {
 	struct fs_node			node;
+	struct mlx5_fs_dr_matcher	fs_dr_matcher;
 	struct mlx5_flow_group_mask	mask;
 	u32				start_index;
 	u32				max_ftes;
@@ -225,6 +235,7 @@ struct mlx5_flow_group {
 
 struct mlx5_flow_root_namespace {
 	struct mlx5_flow_namespace	ns;
+	struct mlx5_fs_dr_domain	fs_dr_domain;
 	enum   fs_flow_table_type	table_type;
 	struct mlx5_core_dev		*dev;
 	struct mlx5_flow_table		*root_ft;
@@ -242,6 +253,11 @@ void mlx5_fc_queue_stats_work(struct mlx5_core_dev *dev,
 void mlx5_fc_update_sampling_interval(struct mlx5_core_dev *dev,
 				      unsigned long interval);
 
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void);
+
+int mlx5_flow_namespace_set_peer(struct mlx5_flow_root_namespace *ns,
+				 struct mlx5_flow_root_namespace *peer_ns);
+
 int mlx5_init_fs(struct mlx5_core_dev *dev);
 void mlx5_cleanup_fs(struct mlx5_core_dev *dev);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
new file mode 100644
index 00000000000000..3d587d0bdbbe3f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -0,0 +1,600 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2019 Mellanox Technologies */
+
+#include "mlx5_core.h"
+#include "fs_core.h"
+#include "fs_cmd.h"
+#include "mlx5dr.h"
+#include "fs_dr.h"
+
+static bool mlx5_dr_is_fw_table(u32 flags)
+{
+	if (flags & MLX5_FLOW_TABLE_TERMINATION)
+		return true;
+
+	return false;
+}
+
+static int mlx5_cmd_dr_update_root_ft(struct mlx5_flow_root_namespace *ns,
+				      struct mlx5_flow_table *ft,
+				      u32 underlay_qpn,
+				      bool disconnect)
+{
+	return mlx5_fs_cmd_get_fw_cmds()->update_root_ft(ns, ft, underlay_qpn,
+							 disconnect);
+}
+
+static int set_miss_action(struct mlx5_flow_root_namespace *ns,
+			   struct mlx5_flow_table *ft,
+			   struct mlx5_flow_table *next_ft)
+{
+	struct mlx5dr_action *old_miss_action;
+	struct mlx5dr_action *action = NULL;
+	struct mlx5dr_table *next_tbl;
+	int err;
+
+	next_tbl = next_ft ? next_ft->fs_dr_table.dr_table : NULL;
+	if (next_tbl) {
+		action = mlx5dr_action_create_dest_table(next_tbl);
+		if (!action)
+			return -EINVAL;
+	}
+	old_miss_action = ft->fs_dr_table.miss_action;
+	err = mlx5dr_table_set_miss_action(ft->fs_dr_table.dr_table, action);
+	if (err && action) {
+		err = mlx5dr_action_destroy(action);
+		if (err) {
+			action = NULL;
+			mlx5_core_err(ns->dev, "Failed to destroy action (%d)\n",
+				      err);
+		}
+	}
+	ft->fs_dr_table.miss_action = action;
+	if (old_miss_action) {
+		err = mlx5dr_action_destroy(old_miss_action);
+		if (err)
+			mlx5_core_err(ns->dev, "Failed to destroy action (%d)\n",
+				      err);
+	}
+
+	return err;
+}
+
+static int mlx5_cmd_dr_create_flow_table(struct mlx5_flow_root_namespace *ns,
+					 struct mlx5_flow_table *ft,
+					 unsigned int log_size,
+					 struct mlx5_flow_table *next_ft)
+{
+	struct mlx5dr_table *tbl;
+	int err;
+
+	if (mlx5_dr_is_fw_table(ft->flags))
+		return mlx5_fs_cmd_get_fw_cmds()->create_flow_table(ns, ft,
+								    log_size,
+								    next_ft);
+
+	tbl = mlx5dr_table_create(ns->fs_dr_domain.dr_domain,
+				  ft->level);
+	if (!tbl) {
+		mlx5_core_err(ns->dev, "Failed creating dr flow_table\n");
+		return -EINVAL;
+	}
+
+	ft->fs_dr_table.dr_table = tbl;
+	ft->id = mlx5dr_table_get_id(tbl);
+
+	if (next_ft) {
+		err = set_miss_action(ns, ft, next_ft);
+		if (err) {
+			mlx5dr_table_destroy(tbl);
+			ft->fs_dr_table.dr_table = NULL;
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int mlx5_cmd_dr_destroy_flow_table(struct mlx5_flow_root_namespace *ns,
+					  struct mlx5_flow_table *ft)
+{
+	struct mlx5dr_action *action = ft->fs_dr_table.miss_action;
+	int err;
+
+	if (mlx5_dr_is_fw_table(ft->flags))
+		return mlx5_fs_cmd_get_fw_cmds()->destroy_flow_table(ns, ft);
+
+	err = mlx5dr_table_destroy(ft->fs_dr_table.dr_table);
+	if (err) {
+		mlx5_core_err(ns->dev, "Failed to destroy flow_table (%d)\n",
+			      err);
+		return err;
+	}
+	if (action) {
+		err = mlx5dr_action_destroy(action);
+		if (err) {
+			mlx5_core_err(ns->dev, "Failed to destroy action(%d)\n",
+				      err);
+			return err;
+		}
+	}
+
+	return err;
+}
+
+static int mlx5_cmd_dr_modify_flow_table(struct mlx5_flow_root_namespace *ns,
+					 struct mlx5_flow_table *ft,
+					 struct mlx5_flow_table *next_ft)
+{
+	return set_miss_action(ns, ft, next_ft);
+}
+
+static int mlx5_cmd_dr_create_flow_group(struct mlx5_flow_root_namespace *ns,
+					 struct mlx5_flow_table *ft,
+					 u32 *in,
+					 struct mlx5_flow_group *fg)
+{
+	struct mlx5dr_matcher *matcher;
+	u16 priority = MLX5_GET(create_flow_group_in, in,
+				start_flow_index);
+	u8 match_criteria_enable = MLX5_GET(create_flow_group_in,
+					    in,
+					    match_criteria_enable);
+	struct mlx5dr_match_parameters mask;
+
+	if (mlx5_dr_is_fw_table(ft->flags))
+		return mlx5_fs_cmd_get_fw_cmds()->create_flow_group(ns, ft, in,
+								    fg);
+
+	mask.match_buf = MLX5_ADDR_OF(create_flow_group_in,
+				      in, match_criteria);
+	mask.match_sz = sizeof(fg->mask.match_criteria);
+
+	matcher = mlx5dr_matcher_create(ft->fs_dr_table.dr_table,
+					priority,
+					match_criteria_enable,
+					&mask);
+	if (!matcher) {
+		mlx5_core_err(ns->dev, "Failed creating matcher\n");
+		return -EINVAL;
+	}
+
+	fg->fs_dr_matcher.dr_matcher = matcher;
+	return 0;
+}
+
+static int mlx5_cmd_dr_destroy_flow_group(struct mlx5_flow_root_namespace *ns,
+					  struct mlx5_flow_table *ft,
+					  struct mlx5_flow_group *fg)
+{
+	if (mlx5_dr_is_fw_table(ft->flags))
+		return mlx5_fs_cmd_get_fw_cmds()->destroy_flow_group(ns, ft, fg);
+
+	return mlx5dr_matcher_destroy(fg->fs_dr_matcher.dr_matcher);
+}
+
+static struct mlx5dr_action *create_vport_action(struct mlx5dr_domain *domain,
+						 struct mlx5_flow_rule *dst)
+{
+	struct mlx5_flow_destination *dest_attr = &dst->dest_attr;
+
+	return mlx5dr_action_create_dest_vport(domain, dest_attr->vport.num,
+					       dest_attr->vport.flags &
+					       MLX5_FLOW_DEST_VPORT_VHCA_ID,
+					       dest_attr->vport.vhca_id);
+}
+
+static struct mlx5dr_action *create_ft_action(struct mlx5_core_dev *dev,
+					      struct mlx5_flow_rule *dst)
+{
+	struct mlx5_flow_table *dest_ft = dst->dest_attr.ft;
+
+	if (mlx5_dr_is_fw_table(dest_ft->flags))
+		return mlx5dr_create_action_dest_flow_fw_table(dest_ft, dev);
+	return mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table);
+}
+
+static struct mlx5dr_action *create_action_push_vlan(struct mlx5dr_domain *domain,
+						     struct mlx5_fs_vlan *vlan)
+{
+	u16 n_ethtype = vlan->ethtype;
+	u8  prio = vlan->prio;
+	u16 vid = vlan->vid;
+	u32 vlan_hdr;
+
+	vlan_hdr = (u32)n_ethtype << 16 | (u32)(prio) << 12 |  (u32)vid;
+	return mlx5dr_action_create_push_vlan(domain, htonl(vlan_hdr));
+}
+
+#define MLX5_FLOW_CONTEXT_ACTION_MAX  20
+static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
+				  struct mlx5_flow_table *ft,
+				  struct mlx5_flow_group *group,
+				  struct fs_fte *fte)
+{
+	struct mlx5dr_domain *domain = ns->fs_dr_domain.dr_domain;
+	struct mlx5dr_action *term_action = NULL;
+	struct mlx5dr_match_parameters params;
+	struct mlx5_core_dev *dev = ns->dev;
+	struct mlx5dr_action **fs_dr_actions;
+	struct mlx5dr_action *tmp_action;
+	struct mlx5dr_action **actions;
+	bool delay_encap_set = false;
+	struct mlx5dr_rule *rule;
+	struct mlx5_flow_rule *dst;
+	int fs_dr_num_actions = 0;
+	int num_actions = 0;
+	size_t match_sz;
+	int err = 0;
+	int i;
+
+	if (mlx5_dr_is_fw_table(ft->flags))
+		return mlx5_fs_cmd_get_fw_cmds()->create_fte(ns, ft, group, fte);
+
+	actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX, sizeof(*actions),
+			  GFP_KERNEL);
+	if (!actions)
+		return -ENOMEM;
+
+	fs_dr_actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX,
+				sizeof(*fs_dr_actions), GFP_KERNEL);
+	if (!fs_dr_actions) {
+		kfree(actions);
+		return -ENOMEM;
+	}
+
+	match_sz = sizeof(fte->val);
+
+	/* The order of the actions are must to be keep, only the following
+	 * order is supported by SW steering:
+	 * TX: push vlan -> modify header -> encap
+	 * RX: decap -> pop vlan -> modify header
+	 */
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
+		tmp_action = create_action_push_vlan(domain, &fte->action.vlan[0]);
+		if (!tmp_action) {
+			err = -ENOMEM;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		actions[num_actions++] = tmp_action;
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2) {
+		tmp_action = create_action_push_vlan(domain, &fte->action.vlan[1]);
+		if (!tmp_action) {
+			err = -ENOMEM;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		actions[num_actions++] = tmp_action;
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_DECAP) {
+		enum mlx5dr_action_reformat_type decap_type =
+			DR_ACTION_REFORMAT_TYP_TNL_L2_TO_L2;
+
+		tmp_action = mlx5dr_action_create_packet_reformat(domain,
+								  decap_type, 0,
+								  NULL);
+		if (!tmp_action) {
+			err = -ENOMEM;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		actions[num_actions++] = tmp_action;
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT) {
+		bool is_decap = fte->action.pkt_reformat->reformat_type ==
+			MLX5_REFORMAT_TYPE_L3_TUNNEL_TO_L2;
+
+		if (is_decap)
+			actions[num_actions++] =
+				fte->action.pkt_reformat->action.dr_action;
+		else
+			delay_encap_set = true;
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) {
+		tmp_action =
+			mlx5dr_action_create_pop_vlan();
+		if (!tmp_action) {
+			err = -ENOMEM;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		actions[num_actions++] = tmp_action;
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP_2) {
+		tmp_action =
+			mlx5dr_action_create_pop_vlan();
+		if (!tmp_action) {
+			err = -ENOMEM;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		actions[num_actions++] = tmp_action;
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)
+		actions[num_actions++] =
+			fte->action.modify_hdr->action.dr_action;
+
+	if (delay_encap_set)
+		actions[num_actions++] =
+			fte->action.pkt_reformat->action.dr_action;
+
+	/* The order of the actions below is not important */
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_DROP) {
+		tmp_action = mlx5dr_action_create_drop();
+		if (!tmp_action) {
+			err = -ENOMEM;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		term_action = tmp_action;
+	}
+
+	if (fte->flow_context.flow_tag) {
+		tmp_action =
+			mlx5dr_action_create_tag(fte->flow_context.flow_tag);
+		if (!tmp_action) {
+			err = -ENOMEM;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		actions[num_actions++] = tmp_action;
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
+		list_for_each_entry(dst, &fte->node.children, node.list) {
+			enum mlx5_flow_destination_type type = dst->dest_attr.type;
+			u32 id;
+
+			if (num_actions == MLX5_FLOW_CONTEXT_ACTION_MAX) {
+				err = -ENOSPC;
+				goto free_actions;
+			}
+
+			switch (type) {
+			case MLX5_FLOW_DESTINATION_TYPE_COUNTER:
+				id = dst->dest_attr.counter_id;
+
+				tmp_action =
+					mlx5dr_action_create_flow_counter(id);
+				if (!tmp_action) {
+					err = -ENOMEM;
+					goto free_actions;
+				}
+				fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+				actions[num_actions++] = tmp_action;
+				break;
+			case MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE:
+				tmp_action = create_ft_action(dev, dst);
+				if (!tmp_action) {
+					err = -ENOMEM;
+					goto free_actions;
+				}
+				fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+				term_action = tmp_action;
+				break;
+			case MLX5_FLOW_DESTINATION_TYPE_VPORT:
+				tmp_action = create_vport_action(domain, dst);
+				if (!tmp_action) {
+					err = -ENOMEM;
+					goto free_actions;
+				}
+				fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+				term_action = tmp_action;
+				break;
+			default:
+				err = -EOPNOTSUPP;
+				goto free_actions;
+			}
+		}
+	}
+
+	params.match_sz = match_sz;
+	params.match_buf = (u64 *)fte->val;
+
+	if (term_action)
+		actions[num_actions++] = term_action;
+
+	rule = mlx5dr_rule_create(group->fs_dr_matcher.dr_matcher,
+				  &params,
+				  num_actions,
+				  actions);
+	if (!rule) {
+		err = -EINVAL;
+		goto free_actions;
+	}
+
+	kfree(actions);
+	fte->fs_dr_rule.dr_rule = rule;
+	fte->fs_dr_rule.num_actions = fs_dr_num_actions;
+	fte->fs_dr_rule.dr_actions = fs_dr_actions;
+
+	return 0;
+
+free_actions:
+	for (i = 0; i < fs_dr_num_actions; i++)
+		if (!IS_ERR_OR_NULL(fs_dr_actions[i]))
+			mlx5dr_action_destroy(fs_dr_actions[i]);
+
+	mlx5_core_err(dev, "Failed to create dr rule err(%d)\n", err);
+	kfree(actions);
+	kfree(fs_dr_actions);
+	return err;
+}
+
+static int mlx5_cmd_dr_packet_reformat_alloc(struct mlx5_flow_root_namespace *ns,
+					     int reformat_type,
+					     size_t size,
+					     void *reformat_data,
+					     enum mlx5_flow_namespace_type namespace,
+					     struct mlx5_pkt_reformat *pkt_reformat)
+{
+	struct mlx5dr_domain *dr_domain = ns->fs_dr_domain.dr_domain;
+	struct mlx5dr_action *action;
+	int dr_reformat;
+
+	switch (reformat_type) {
+	case MLX5_REFORMAT_TYPE_L2_TO_VXLAN:
+	case MLX5_REFORMAT_TYPE_L2_TO_NVGRE:
+	case MLX5_REFORMAT_TYPE_L2_TO_L2_TUNNEL:
+		dr_reformat = DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L2;
+		break;
+	case MLX5_REFORMAT_TYPE_L3_TUNNEL_TO_L2:
+		dr_reformat = DR_ACTION_REFORMAT_TYP_TNL_L3_TO_L2;
+		break;
+	case MLX5_REFORMAT_TYPE_L2_TO_L3_TUNNEL:
+		dr_reformat = DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L3;
+		break;
+	default:
+		mlx5_core_err(ns->dev, "Packet-reformat not supported(%d)\n",
+			      reformat_type);
+		return -EOPNOTSUPP;
+	}
+
+	action = mlx5dr_action_create_packet_reformat(dr_domain,
+						      dr_reformat,
+						      size,
+						      reformat_data);
+	if (!action) {
+		mlx5_core_err(ns->dev, "Failed allocating packet-reformat action\n");
+		return -EINVAL;
+	}
+
+	pkt_reformat->action.dr_action = action;
+
+	return 0;
+}
+
+static void mlx5_cmd_dr_packet_reformat_dealloc(struct mlx5_flow_root_namespace *ns,
+						struct mlx5_pkt_reformat *pkt_reformat)
+{
+	mlx5dr_action_destroy(pkt_reformat->action.dr_action);
+}
+
+static int mlx5_cmd_dr_modify_header_alloc(struct mlx5_flow_root_namespace *ns,
+					   u8 namespace, u8 num_actions,
+					   void *modify_actions,
+					   struct mlx5_modify_hdr *modify_hdr)
+{
+	struct mlx5dr_domain *dr_domain = ns->fs_dr_domain.dr_domain;
+	struct mlx5dr_action *action;
+	size_t actions_sz;
+
+	actions_sz = MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto) *
+		num_actions;
+	action = mlx5dr_action_create_modify_header(dr_domain, 0,
+						    actions_sz,
+						    modify_actions);
+	if (!action) {
+		mlx5_core_err(ns->dev, "Failed allocating modify-header action\n");
+		return -EINVAL;
+	}
+
+	modify_hdr->action.dr_action = action;
+
+	return 0;
+}
+
+static void mlx5_cmd_dr_modify_header_dealloc(struct mlx5_flow_root_namespace *ns,
+					      struct mlx5_modify_hdr *modify_hdr)
+{
+	mlx5dr_action_destroy(modify_hdr->action.dr_action);
+}
+
+static int mlx5_cmd_dr_update_fte(struct mlx5_flow_root_namespace *ns,
+				  struct mlx5_flow_table *ft,
+				  struct mlx5_flow_group *group,
+				  int modify_mask,
+				  struct fs_fte *fte)
+{
+	return -EOPNOTSUPP;
+}
+
+static int mlx5_cmd_dr_delete_fte(struct mlx5_flow_root_namespace *ns,
+				  struct mlx5_flow_table *ft,
+				  struct fs_fte *fte)
+{
+	struct mlx5_fs_dr_rule *rule = &fte->fs_dr_rule;
+	int err;
+	int i;
+
+	if (mlx5_dr_is_fw_table(ft->flags))
+		return mlx5_fs_cmd_get_fw_cmds()->delete_fte(ns, ft, fte);
+
+	err = mlx5dr_rule_destroy(rule->dr_rule);
+	if (err)
+		return err;
+
+	for (i = 0; i < rule->num_actions; i++)
+		if (!IS_ERR_OR_NULL(rule->dr_actions[i]))
+			mlx5dr_action_destroy(rule->dr_actions[i]);
+
+	kfree(rule->dr_actions);
+	return 0;
+}
+
+static int mlx5_cmd_dr_set_peer(struct mlx5_flow_root_namespace *ns,
+				struct mlx5_flow_root_namespace *peer_ns)
+{
+	struct mlx5dr_domain *peer_domain = NULL;
+
+	if (peer_ns)
+		peer_domain = peer_ns->fs_dr_domain.dr_domain;
+	mlx5dr_domain_set_peer(ns->fs_dr_domain.dr_domain,
+			       peer_domain);
+	return 0;
+}
+
+static int mlx5_cmd_dr_create_ns(struct mlx5_flow_root_namespace *ns)
+{
+	ns->fs_dr_domain.dr_domain =
+		mlx5dr_domain_create(ns->dev,
+				     MLX5DR_DOMAIN_TYPE_FDB);
+	if (!ns->fs_dr_domain.dr_domain) {
+		mlx5_core_err(ns->dev, "Failed to create dr flow namespace\n");
+		return -EOPNOTSUPP;
+	}
+	return 0;
+}
+
+static int mlx5_cmd_dr_destroy_ns(struct mlx5_flow_root_namespace *ns)
+{
+	return mlx5dr_domain_destroy(ns->fs_dr_domain.dr_domain);
+}
+
+bool mlx5_fs_dr_is_supported(struct mlx5_core_dev *dev)
+{
+	return mlx5dr_is_supported(dev);
+}
+
+static const struct mlx5_flow_cmds mlx5_flow_cmds_dr = {
+	.create_flow_table = mlx5_cmd_dr_create_flow_table,
+	.destroy_flow_table = mlx5_cmd_dr_destroy_flow_table,
+	.modify_flow_table = mlx5_cmd_dr_modify_flow_table,
+	.create_flow_group = mlx5_cmd_dr_create_flow_group,
+	.destroy_flow_group = mlx5_cmd_dr_destroy_flow_group,
+	.create_fte = mlx5_cmd_dr_create_fte,
+	.update_fte = mlx5_cmd_dr_update_fte,
+	.delete_fte = mlx5_cmd_dr_delete_fte,
+	.update_root_ft = mlx5_cmd_dr_update_root_ft,
+	.packet_reformat_alloc = mlx5_cmd_dr_packet_reformat_alloc,
+	.packet_reformat_dealloc = mlx5_cmd_dr_packet_reformat_dealloc,
+	.modify_header_alloc = mlx5_cmd_dr_modify_header_alloc,
+	.modify_header_dealloc = mlx5_cmd_dr_modify_header_dealloc,
+	.set_peer = mlx5_cmd_dr_set_peer,
+	.create_ns = mlx5_cmd_dr_create_ns,
+	.destroy_ns = mlx5_cmd_dr_destroy_ns,
+};
+
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_dr_cmds(void)
+{
+		return &mlx5_flow_cmds_dr;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.h
new file mode 100644
index 00000000000000..1fb185d6ac7f7c
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.h
@@ -0,0 +1,60 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+ * Copyright (c) 2019 Mellanox Technologies
+ */
+
+#ifndef _MLX5_FS_DR_
+#define _MLX5_FS_DR_
+
+#include "mlx5dr.h"
+
+struct mlx5_flow_root_namespace;
+struct fs_fte;
+
+struct mlx5_fs_dr_action {
+	struct mlx5dr_action *dr_action;
+};
+
+struct mlx5_fs_dr_ns {
+	struct mlx5_dr_ns *dr_ns;
+};
+
+struct mlx5_fs_dr_rule {
+	struct mlx5dr_rule    *dr_rule;
+	/* Only actions created by fs_dr */
+	struct mlx5dr_action  **dr_actions;
+	int                      num_actions;
+};
+
+struct mlx5_fs_dr_domain {
+	struct mlx5dr_domain	*dr_domain;
+};
+
+struct mlx5_fs_dr_matcher {
+	struct mlx5dr_matcher *dr_matcher;
+};
+
+struct mlx5_fs_dr_table {
+	struct mlx5dr_table  *dr_table;
+	struct mlx5dr_action *miss_action;
+};
+
+#ifdef CONFIG_MLX5_SW_STEERING
+
+bool mlx5_fs_dr_is_supported(struct mlx5_core_dev *dev);
+
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_dr_cmds(void);
+
+#else
+
+static inline const struct mlx5_flow_cmds *mlx5_fs_cmd_get_dr_cmds(void)
+{
+	return NULL;
+}
+
+static inline bool mlx5_fs_dr_is_supported(struct mlx5_core_dev *dev)
+{
+	return false;
+}
+
+#endif /* CONFIG_MLX5_SW_STEERING */
+#endif