Merge tag 'shared-for-4.9-2' of git://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma

Saeed Mahameed says:

====================
Mellanox mlx5 core driver updates 2016-08-24

This series contains some low level and API updates for mlx5 core
driver interface and mlx5_ifc.h, plus mlx5 LAG core driver support,
to be shared as base code for net-next and rdma mlx5 4.9 submissions.

From Alex and Artemy, Update mlx5_ifc for modify RQ and XRC bits.

From Noa, Expose mlx5 link modes so they can be used in RDMA tree for rdma tools.

From Aviv, LAG support needed for RDMA.
    - Add needed hardware structures, layouts and interface
    - mlx5 core driver LAG implementation
    - Introduce mlx5 core driver LAG API for mlx5_ib

From Maor, add two low level patches for mlx5 hardware sniffer QP
infrastructure bits and capabilities, plus added the namespace for sniffer
steering tables.  Needed for RDMA subtree.
====================

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

drivers/infiniband/hw/mlx5/main.c

@@ -748,8 +748,7 @@ static int mlx5_query_hca_port(struct ib_device *ibdev, u8 port,
 				     &props->active_width);
 	if (err)
 		goto out;
-	err = mlx5_query_port_proto_oper(mdev, &props->active_speed, MLX5_PTYS_IB,
-					 port);
+	err = mlx5_query_port_ib_proto_oper(mdev, &props->active_speed, port);
 	if (err)
 		goto out;
 

drivers/net/ethernet/mellanox/mlx5/core/Makefile

@@ -3,7 +3,7 @@ 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 fs_cmd.o fs_core.o \
-		fs_counters.o rl.o
+		fs_counters.o rl.o lag.o
 
 mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \
 		en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \

drivers/net/ethernet/mellanox/mlx5/core/cmd.c

@@ -285,6 +285,8 @@ static int mlx5_internal_err_ret_value(struct mlx5_core_dev *dev, u16 op,
 	case MLX5_CMD_OP_DEALLOC_TRANSPORT_DOMAIN:
 	case MLX5_CMD_OP_DELETE_VXLAN_UDP_DPORT:
 	case MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY:
+	case MLX5_CMD_OP_DESTROY_LAG:
+	case MLX5_CMD_OP_DESTROY_VPORT_LAG:
 	case MLX5_CMD_OP_DESTROY_TIR:
 	case MLX5_CMD_OP_DESTROY_SQ:
 	case MLX5_CMD_OP_DESTROY_RQ:
@@ -376,6 +378,10 @@ static int mlx5_internal_err_ret_value(struct mlx5_core_dev *dev, u16 op,
 	case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT:
 	case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
 	case MLX5_CMD_OP_QUERY_L2_TABLE_ENTRY:
+	case MLX5_CMD_OP_CREATE_LAG:
+	case MLX5_CMD_OP_MODIFY_LAG:
+	case MLX5_CMD_OP_QUERY_LAG:
+	case MLX5_CMD_OP_CREATE_VPORT_LAG:
 	case MLX5_CMD_OP_CREATE_TIR:
 	case MLX5_CMD_OP_MODIFY_TIR:
 	case MLX5_CMD_OP_QUERY_TIR:
@@ -514,6 +520,12 @@ const char *mlx5_command_str(int command)
 	MLX5_COMMAND_STR_CASE(DELETE_L2_TABLE_ENTRY);
 	MLX5_COMMAND_STR_CASE(SET_WOL_ROL);
 	MLX5_COMMAND_STR_CASE(QUERY_WOL_ROL);
+	MLX5_COMMAND_STR_CASE(CREATE_LAG);
+	MLX5_COMMAND_STR_CASE(MODIFY_LAG);
+	MLX5_COMMAND_STR_CASE(QUERY_LAG);
+	MLX5_COMMAND_STR_CASE(DESTROY_LAG);
+	MLX5_COMMAND_STR_CASE(CREATE_VPORT_LAG);
+	MLX5_COMMAND_STR_CASE(DESTROY_VPORT_LAG);
 	MLX5_COMMAND_STR_CASE(CREATE_TIR);
 	MLX5_COMMAND_STR_CASE(MODIFY_TIR);
 	MLX5_COMMAND_STR_CASE(DESTROY_TIR);

drivers/net/ethernet/mellanox/mlx5/core/en.h

@@ -651,40 +651,6 @@ struct mlx5e_priv {
 	void                      *ppriv;
 };
 
-enum mlx5e_link_mode {
-	MLX5E_1000BASE_CX_SGMII	 = 0,
-	MLX5E_1000BASE_KX	 = 1,
-	MLX5E_10GBASE_CX4	 = 2,
-	MLX5E_10GBASE_KX4	 = 3,
-	MLX5E_10GBASE_KR	 = 4,
-	MLX5E_20GBASE_KR2	 = 5,
-	MLX5E_40GBASE_CR4	 = 6,
-	MLX5E_40GBASE_KR4	 = 7,
-	MLX5E_56GBASE_R4	 = 8,
-	MLX5E_10GBASE_CR	 = 12,
-	MLX5E_10GBASE_SR	 = 13,
-	MLX5E_10GBASE_ER	 = 14,
-	MLX5E_40GBASE_SR4	 = 15,
-	MLX5E_40GBASE_LR4	 = 16,
-	MLX5E_50GBASE_SR2	 = 18,
-	MLX5E_100GBASE_CR4	 = 20,
-	MLX5E_100GBASE_SR4	 = 21,
-	MLX5E_100GBASE_KR4	 = 22,
-	MLX5E_100GBASE_LR4	 = 23,
-	MLX5E_100BASE_TX	 = 24,
-	MLX5E_1000BASE_T	 = 25,
-	MLX5E_10GBASE_T		 = 26,
-	MLX5E_25GBASE_CR	 = 27,
-	MLX5E_25GBASE_KR	 = 28,
-	MLX5E_25GBASE_SR	 = 29,
-	MLX5E_50GBASE_CR2	 = 30,
-	MLX5E_50GBASE_KR2	 = 31,
-	MLX5E_LINK_MODES_NUMBER,
-};
-
-#define MLX5E_PROT_MASK(link_mode) (1 << link_mode)
-
-
 void mlx5e_build_ptys2ethtool_map(void);
 
 void mlx5e_send_nop(struct mlx5e_sq *sq, bool notify_hw);

drivers/net/ethernet/mellanox/mlx5/core/en_main.c

@@ -489,7 +489,8 @@ static int mlx5e_modify_rq_vsd(struct mlx5e_rq *rq, bool vsd)
 	rqc = MLX5_ADDR_OF(modify_rq_in, in, ctx);
 
 	MLX5_SET(modify_rq_in, in, rq_state, MLX5_RQC_STATE_RDY);
-	MLX5_SET64(modify_rq_in, in, modify_bitmask, MLX5_RQ_BITMASK_VSD);
+	MLX5_SET64(modify_rq_in, in, modify_bitmask,
+		   MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_VSD);
 	MLX5_SET(rqc, rqc, vsd, vsd);
 	MLX5_SET(rqc, rqc, state, MLX5_RQC_STATE_RDY);
 
@@ -2024,6 +2025,10 @@ static int mlx5e_create_tis(struct mlx5e_priv *priv, int tc)
 
 	MLX5_SET(tisc, tisc, prio, tc << 1);
 	MLX5_SET(tisc, tisc, transport_domain, mdev->mlx5e_res.td.tdn);
+
+	if (mlx5_lag_is_lacp_owner(mdev))
+		MLX5_SET(tisc, tisc, strict_lag_tx_port_affinity, 1);
+
 	return mlx5_core_create_tis(mdev, in, sizeof(in), &priv->tisn[tc]);
 }
 
@@ -3368,6 +3373,8 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv)
 	struct mlx5_eswitch *esw = mdev->priv.eswitch;
 	struct mlx5_eswitch_rep rep;
 
+	mlx5_lag_add(mdev, netdev);
+
 	if (mlx5e_vxlan_allowed(mdev)) {
 		rtnl_lock();
 		udp_tunnel_get_rx_info(netdev);
@@ -3390,6 +3397,7 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv)
 {
 	queue_work(priv->wq, &priv->set_rx_mode_work);
 	mlx5e_disable_async_events(priv);
+	mlx5_lag_remove(priv->mdev);
 }
 
 static const struct mlx5e_profile mlx5e_nic_profile = {

drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c

@@ -58,6 +58,7 @@ int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev,
 
 int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
 			       u16 vport,
+			       enum fs_flow_table_op_mod op_mod,
 			       enum fs_flow_table_type type, unsigned int level,
 			       unsigned int log_size, struct mlx5_flow_table
 			       *next_ft, unsigned int *table_id)
@@ -69,10 +70,6 @@ int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
 	MLX5_SET(create_flow_table_in, in, opcode,
 		 MLX5_CMD_OP_CREATE_FLOW_TABLE);
 
-	if (next_ft) {
-		MLX5_SET(create_flow_table_in, in, table_miss_mode, 1);
-		MLX5_SET(create_flow_table_in, in, table_miss_id, next_ft->id);
-	}
 	MLX5_SET(create_flow_table_in, in, table_type, type);
 	MLX5_SET(create_flow_table_in, in, level, level);
 	MLX5_SET(create_flow_table_in, in, log_size, log_size);
@@ -81,6 +78,22 @@ int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
 		MLX5_SET(create_flow_table_in, in, other_vport, 1);
 	}
 
+	switch (op_mod) {
+	case FS_FT_OP_MOD_NORMAL:
+		if (next_ft) {
+			MLX5_SET(create_flow_table_in, in, table_miss_mode, 1);
+			MLX5_SET(create_flow_table_in, in, table_miss_id, next_ft->id);
+		}
+		break;
+
+	case FS_FT_OP_MOD_LAG_DEMUX:
+		MLX5_SET(create_flow_table_in, in, op_mod, 0x1);
+		if (next_ft)
+			MLX5_SET(create_flow_table_in, in, lag_master_next_table_id,
+				 next_ft->id);
+		break;
+	}
+
 	err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
 	if (!err)
 		*table_id = MLX5_GET(create_flow_table_out, out,
@@ -117,17 +130,32 @@ int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev,
 		 MLX5_CMD_OP_MODIFY_FLOW_TABLE);
 	MLX5_SET(modify_flow_table_in, in, table_type, ft->type);
 	MLX5_SET(modify_flow_table_in, in, table_id, ft->id);
-	if (ft->vport) {
-		MLX5_SET(modify_flow_table_in, in, vport_number, ft->vport);
-		MLX5_SET(modify_flow_table_in, in, other_vport, 1);
-	}
-	MLX5_SET(modify_flow_table_in, in, modify_field_select,
-		 MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID);
-	if (next_ft) {
-		MLX5_SET(modify_flow_table_in, in, table_miss_mode, 1);
-		MLX5_SET(modify_flow_table_in, in, table_miss_id, next_ft->id);
+
+	if (ft->op_mod == FS_FT_OP_MOD_LAG_DEMUX) {
+		MLX5_SET(modify_flow_table_in, in, modify_field_select,
+			 MLX5_MODIFY_FLOW_TABLE_LAG_NEXT_TABLE_ID);
+		if (next_ft) {
+			MLX5_SET(modify_flow_table_in, in,
+				 lag_master_next_table_id, next_ft->id);
+		} else {
+			MLX5_SET(modify_flow_table_in, in,
+				 lag_master_next_table_id, 0);
+		}
 	} else {
-		MLX5_SET(modify_flow_table_in, in, table_miss_mode, 0);
+		if (ft->vport) {
+			MLX5_SET(modify_flow_table_in, in, vport_number,
+				 ft->vport);
+			MLX5_SET(modify_flow_table_in, in, other_vport, 1);
+		}
+		MLX5_SET(modify_flow_table_in, in, modify_field_select,
+			 MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID);
+		if (next_ft) {
+			MLX5_SET(modify_flow_table_in, in, table_miss_mode, 1);
+			MLX5_SET(modify_flow_table_in, in, table_miss_id,
+				 next_ft->id);
+		} else {
+			MLX5_SET(modify_flow_table_in, in, table_miss_mode, 0);
+		}
 	}
 
 	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));

drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h

@@ -35,6 +35,7 @@
 
 int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
 			       u16 vport,
+			       enum fs_flow_table_op_mod op_mod,
 			       enum fs_flow_table_type type, unsigned int level,
 			       unsigned int log_size, struct mlx5_flow_table
 			       *next_ft, unsigned int *table_id);

drivers/net/ethernet/mellanox/mlx5/core/fs_core.c

@@ -96,6 +96,10 @@
 #define OFFLOADS_NUM_PRIOS 1
 #define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + 1)
 
+#define LAG_PRIO_NUM_LEVELS 1
+#define LAG_NUM_PRIOS 1
+#define LAG_MIN_LEVEL (OFFLOADS_MIN_LEVEL + 1)
+
 struct node_caps {
 	size_t	arr_sz;
 	long	*caps;
@@ -111,12 +115,16 @@ static struct init_tree_node {
 	int num_levels;
 } root_fs = {
 	.type = FS_TYPE_NAMESPACE,
-	.ar_size = 6,
+	.ar_size = 7,
 	.children = (struct init_tree_node[]) {
 		ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0,
 			 FS_CHAINING_CAPS,
 			 ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS,
 						  BY_PASS_PRIO_NUM_LEVELS))),
+		ADD_PRIO(0, LAG_MIN_LEVEL, 0,
+			 FS_CHAINING_CAPS,
+			 ADD_NS(ADD_MULTIPLE_PRIO(LAG_NUM_PRIOS,
+						  LAG_PRIO_NUM_LEVELS))),
 		ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {},
 			 ADD_NS(ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))),
 		ADD_PRIO(0, ETHTOOL_MIN_LEVEL, 0,
@@ -477,7 +485,8 @@ static struct mlx5_flow_group *alloc_flow_group(u32 *create_fg_in)
 }
 
 static struct mlx5_flow_table *alloc_flow_table(int level, u16 vport, int max_fte,
-						enum fs_flow_table_type table_type)
+						enum fs_flow_table_type table_type,
+						enum fs_flow_table_op_mod op_mod)
 {
 	struct mlx5_flow_table *ft;
 
@@ -487,6 +496,7 @@ static struct mlx5_flow_table *alloc_flow_table(int level, u16 vport, int max_ft
 
 	ft->level = level;
 	ft->node.type = FS_TYPE_FLOW_TABLE;
+	ft->op_mod = op_mod;
 	ft->type = table_type;
 	ft->vport = vport;
 	ft->max_fte = max_fte;
@@ -724,6 +734,7 @@ static void list_add_flow_table(struct mlx5_flow_table *ft,
 }
 
 static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespace *ns,
+							enum fs_flow_table_op_mod op_mod,
 							u16 vport, int prio,
 							int max_fte, u32 level)
 {
@@ -756,18 +767,19 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa
 	level += fs_prio->start_level;
 	ft = alloc_flow_table(level,
 			      vport,
-			      roundup_pow_of_two(max_fte),
-			      root->table_type);
+			      max_fte ? roundup_pow_of_two(max_fte) : 0,
+			      root->table_type,
+			      op_mod);
 	if (!ft) {
 		err = -ENOMEM;
 		goto unlock_root;
 	}
 
 	tree_init_node(&ft->node, 1, del_flow_table);
-	log_table_sz = ilog2(ft->max_fte);
+	log_table_sz = ft->max_fte ? ilog2(ft->max_fte) : 0;
 	next_ft = find_next_chained_ft(fs_prio);
-	err = mlx5_cmd_create_flow_table(root->dev, ft->vport, ft->type, ft->level,
-					 log_table_sz, next_ft, &ft->id);
+	err = mlx5_cmd_create_flow_table(root->dev, ft->vport, ft->op_mod, ft->type,
+					 ft->level, log_table_sz, next_ft, &ft->id);
 	if (err)
 		goto free_ft;
 
@@ -794,15 +806,26 @@ struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns,
 					       int prio, int max_fte,
 					       u32 level)
 {
-	return __mlx5_create_flow_table(ns, 0, prio, max_fte, level);
+	return __mlx5_create_flow_table(ns, FS_FT_OP_MOD_NORMAL, 0, prio,
+					max_fte, level);
 }
 
 struct mlx5_flow_table *mlx5_create_vport_flow_table(struct mlx5_flow_namespace *ns,
 						     int prio, int max_fte,
 						     u32 level, u16 vport)
 {
-	return __mlx5_create_flow_table(ns, vport, prio, max_fte, level);
+	return __mlx5_create_flow_table(ns, FS_FT_OP_MOD_NORMAL, vport, prio,
+					max_fte, level);
+}
+
+struct mlx5_flow_table *mlx5_create_lag_demux_flow_table(
+					       struct mlx5_flow_namespace *ns,
+					       int prio, u32 level)
+{
+	return __mlx5_create_flow_table(ns, FS_FT_OP_MOD_LAG_DEMUX, 0, prio, 0,
+					level);
 }
+EXPORT_SYMBOL(mlx5_create_lag_demux_flow_table);
 
 struct mlx5_flow_table *mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
 							    int prio,
@@ -1381,6 +1404,7 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
 
 	switch (type) {
 	case MLX5_FLOW_NAMESPACE_BYPASS:
+	case MLX5_FLOW_NAMESPACE_LAG:
 	case MLX5_FLOW_NAMESPACE_OFFLOADS:
 	case MLX5_FLOW_NAMESPACE_ETHTOOL:
 	case MLX5_FLOW_NAMESPACE_KERNEL:
@@ -1403,6 +1427,16 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
 			return &steering->esw_ingress_root_ns->ns;
 		else
 			return NULL;
+	case MLX5_FLOW_NAMESPACE_SNIFFER_RX:
+		if (steering->sniffer_rx_root_ns)
+			return &steering->sniffer_rx_root_ns->ns;
+		else
+			return NULL;
+	case MLX5_FLOW_NAMESPACE_SNIFFER_TX:
+		if (steering->sniffer_tx_root_ns)
+			return &steering->sniffer_tx_root_ns->ns;
+		else
+			return NULL;
 	default:
 		return NULL;
 	}
@@ -1702,10 +1736,46 @@ void mlx5_cleanup_fs(struct mlx5_core_dev *dev)
 	cleanup_root_ns(steering->esw_egress_root_ns);
 	cleanup_root_ns(steering->esw_ingress_root_ns);
 	cleanup_root_ns(steering->fdb_root_ns);
+	cleanup_root_ns(steering->sniffer_rx_root_ns);
+	cleanup_root_ns(steering->sniffer_tx_root_ns);
 	mlx5_cleanup_fc_stats(dev);
 	kfree(steering);
 }
 
+static int init_sniffer_tx_root_ns(struct mlx5_flow_steering *steering)
+{
+	struct fs_prio *prio;
+
+	steering->sniffer_tx_root_ns = create_root_ns(steering, FS_FT_SNIFFER_TX);
+	if (!steering->sniffer_tx_root_ns)
+		return -ENOMEM;
+
+	/* Create single prio */
+	prio = fs_create_prio(&steering->sniffer_tx_root_ns->ns, 0, 1);
+	if (IS_ERR(prio)) {
+		cleanup_root_ns(steering->sniffer_tx_root_ns);
+		return PTR_ERR(prio);
+	}
+	return 0;
+}
+
+static int init_sniffer_rx_root_ns(struct mlx5_flow_steering *steering)
+{
+	struct fs_prio *prio;
+
+	steering->sniffer_rx_root_ns = create_root_ns(steering, FS_FT_SNIFFER_RX);
+	if (!steering->sniffer_rx_root_ns)
+		return -ENOMEM;
+
+	/* Create single prio */
+	prio = fs_create_prio(&steering->sniffer_rx_root_ns->ns, 0, 1);
+	if (IS_ERR(prio)) {
+		cleanup_root_ns(steering->sniffer_rx_root_ns);
+		return PTR_ERR(prio);
+	}
+	return 0;
+}
+
 static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
 {
 	struct fs_prio *prio;
@@ -1802,6 +1872,18 @@ int mlx5_init_fs(struct mlx5_core_dev *dev)
 		}
 	}
 
+	if (MLX5_CAP_FLOWTABLE_SNIFFER_RX(dev, ft_support)) {
+		err = init_sniffer_rx_root_ns(steering);
+		if (err)
+			goto err;
+	}
+
+	if (MLX5_CAP_FLOWTABLE_SNIFFER_TX(dev, ft_support)) {
+		err = init_sniffer_tx_root_ns(steering);
+		if (err)
+			goto err;
+	}
+
 	return 0;
 err:
 	mlx5_cleanup_fs(dev);

drivers/net/ethernet/mellanox/mlx5/core/fs_core.h

@@ -49,6 +49,13 @@ enum fs_flow_table_type {
 	FS_FT_ESW_EGRESS_ACL  = 0x2,
 	FS_FT_ESW_INGRESS_ACL = 0x3,
 	FS_FT_FDB             = 0X4,
+	FS_FT_SNIFFER_RX	= 0X5,
+	FS_FT_SNIFFER_TX	= 0X6,
+};
+
+enum fs_flow_table_op_mod {
+	FS_FT_OP_MOD_NORMAL,
+	FS_FT_OP_MOD_LAG_DEMUX,
 };
 
 enum fs_fte_status {
@@ -61,6 +68,8 @@ struct mlx5_flow_steering {
 	struct mlx5_flow_root_namespace *fdb_root_ns;
 	struct mlx5_flow_root_namespace *esw_egress_root_ns;
 	struct mlx5_flow_root_namespace *esw_ingress_root_ns;
+	struct mlx5_flow_root_namespace	*sniffer_tx_root_ns;
+	struct mlx5_flow_root_namespace	*sniffer_rx_root_ns;
 };
 
 struct fs_node {
@@ -93,6 +102,7 @@ struct mlx5_flow_table {
 	unsigned int			max_fte;
 	unsigned int			level;
 	enum fs_flow_table_type		type;
+	enum fs_flow_table_op_mod	op_mod;
 	struct {
 		bool			active;
 		unsigned int		required_groups;

drivers/net/ethernet/mellanox/mlx5/core/lag.c

@@ -0,0 +1,602 @@
+/*
+ * Copyright (c) 2016, 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/netdevice.h>
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/vport.h>
+#include "mlx5_core.h"
+
+enum {
+	MLX5_LAG_FLAG_BONDED = 1 << 0,
+};
+
+struct lag_func {
+	struct mlx5_core_dev *dev;
+	struct net_device    *netdev;
+};
+
+/* Used for collection of netdev event info. */
+struct lag_tracker {
+	enum   netdev_lag_tx_type           tx_type;
+	struct netdev_lag_lower_state_info  netdev_state[MLX5_MAX_PORTS];
+	bool is_bonded;
+};
+
+/* LAG data of a ConnectX card.
+ * It serves both its phys functions.
+ */
+struct mlx5_lag {
+	u8                        flags;
+	u8                        v2p_map[MLX5_MAX_PORTS];
+	struct lag_func           pf[MLX5_MAX_PORTS];
+	struct lag_tracker        tracker;
+	struct delayed_work       bond_work;
+	struct notifier_block     nb;
+};
+
+/* General purpose, use for short periods of time.
+ * Beware of lock dependencies (preferably, no locks should be acquired
+ * under it).
+ */
+static DEFINE_MUTEX(lag_mutex);
+
+static int mlx5_cmd_create_lag(struct mlx5_core_dev *dev, u8 remap_port1,
+			       u8 remap_port2)
+{
+	u32   in[MLX5_ST_SZ_DW(create_lag_in)]   = {0};
+	u32   out[MLX5_ST_SZ_DW(create_lag_out)] = {0};
+	void *lag_ctx = MLX5_ADDR_OF(create_lag_in, in, ctx);
+
+	MLX5_SET(create_lag_in, in, opcode, MLX5_CMD_OP_CREATE_LAG);
+
+	MLX5_SET(lagc, lag_ctx, tx_remap_affinity_1, remap_port1);
+	MLX5_SET(lagc, lag_ctx, tx_remap_affinity_2, remap_port2);
+
+	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+static int mlx5_cmd_modify_lag(struct mlx5_core_dev *dev, u8 remap_port1,
+			       u8 remap_port2)
+{
+	u32   in[MLX5_ST_SZ_DW(modify_lag_in)]   = {0};
+	u32   out[MLX5_ST_SZ_DW(modify_lag_out)] = {0};
+	void *lag_ctx = MLX5_ADDR_OF(modify_lag_in, in, ctx);
+
+	MLX5_SET(modify_lag_in, in, opcode, MLX5_CMD_OP_MODIFY_LAG);
+	MLX5_SET(modify_lag_in, in, field_select, 0x1);
+
+	MLX5_SET(lagc, lag_ctx, tx_remap_affinity_1, remap_port1);
+	MLX5_SET(lagc, lag_ctx, tx_remap_affinity_2, remap_port2);
+
+	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+static int mlx5_cmd_destroy_lag(struct mlx5_core_dev *dev)
+{
+	u32  in[MLX5_ST_SZ_DW(destroy_lag_in)]  = {0};
+	u32 out[MLX5_ST_SZ_DW(destroy_lag_out)] = {0};
+
+	MLX5_SET(destroy_lag_in, in, opcode, MLX5_CMD_OP_DESTROY_LAG);
+
+	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+int mlx5_cmd_create_vport_lag(struct mlx5_core_dev *dev)
+{
+	u32  in[MLX5_ST_SZ_DW(create_vport_lag_in)]  = {0};
+	u32 out[MLX5_ST_SZ_DW(create_vport_lag_out)] = {0};
+
+	MLX5_SET(create_vport_lag_in, in, opcode, MLX5_CMD_OP_CREATE_VPORT_LAG);
+
+	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+EXPORT_SYMBOL(mlx5_cmd_create_vport_lag);
+
+int mlx5_cmd_destroy_vport_lag(struct mlx5_core_dev *dev)
+{
+	u32  in[MLX5_ST_SZ_DW(destroy_vport_lag_in)]  = {0};
+	u32 out[MLX5_ST_SZ_DW(destroy_vport_lag_out)] = {0};
+
+	MLX5_SET(destroy_vport_lag_in, in, opcode, MLX5_CMD_OP_DESTROY_VPORT_LAG);
+
+	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+EXPORT_SYMBOL(mlx5_cmd_destroy_vport_lag);
+
+static struct mlx5_lag *mlx5_lag_dev_get(struct mlx5_core_dev *dev)
+{
+	return dev->priv.lag;
+}
+
+static int mlx5_lag_dev_get_netdev_idx(struct mlx5_lag *ldev,
+				       struct net_device *ndev)
+{
+	int i;
+
+	for (i = 0; i < MLX5_MAX_PORTS; i++)
+		if (ldev->pf[i].netdev == ndev)
+			return i;
+
+	return -1;
+}
+
+static bool mlx5_lag_is_bonded(struct mlx5_lag *ldev)
+{
+	return !!(ldev->flags & MLX5_LAG_FLAG_BONDED);
+}
+
+static void mlx5_infer_tx_affinity_mapping(struct lag_tracker *tracker,
+					   u8 *port1, u8 *port2)
+{
+	if (tracker->tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP) {
+		if (tracker->netdev_state[0].tx_enabled) {
+			*port1 = 1;
+			*port2 = 1;
+		} else {
+			*port1 = 2;
+			*port2 = 2;
+		}
+	} else {
+		*port1 = 1;
+		*port2 = 2;
+		if (!tracker->netdev_state[0].link_up)
+			*port1 = 2;
+		else if (!tracker->netdev_state[1].link_up)
+			*port2 = 1;
+	}
+}
+
+static void mlx5_activate_lag(struct mlx5_lag *ldev,
+			      struct lag_tracker *tracker)
+{
+	struct mlx5_core_dev *dev0 = ldev->pf[0].dev;
+	int err;
+
+	ldev->flags |= MLX5_LAG_FLAG_BONDED;
+
+	mlx5_infer_tx_affinity_mapping(tracker, &ldev->v2p_map[0],
+				       &ldev->v2p_map[1]);
+
+	err = mlx5_cmd_create_lag(dev0, ldev->v2p_map[0], ldev->v2p_map[1]);
+	if (err)
+		mlx5_core_err(dev0,
+			      "Failed to create LAG (%d)\n",
+			      err);
+}
+
+static void mlx5_deactivate_lag(struct mlx5_lag *ldev)
+{
+	struct mlx5_core_dev *dev0 = ldev->pf[0].dev;
+	int err;
+
+	ldev->flags &= ~MLX5_LAG_FLAG_BONDED;
+
+	err = mlx5_cmd_destroy_lag(dev0);
+	if (err)
+		mlx5_core_err(dev0,
+			      "Failed to destroy LAG (%d)\n",
+			      err);
+}
+
+static void mlx5_do_bond(struct mlx5_lag *ldev)
+{
+	struct mlx5_core_dev *dev0 = ldev->pf[0].dev;
+	struct mlx5_core_dev *dev1 = ldev->pf[1].dev;
+	struct lag_tracker tracker;
+	u8 v2p_port1, v2p_port2;
+	int i, err;
+
+	if (!dev0 || !dev1)
+		return;
+
+	mutex_lock(&lag_mutex);
+	tracker = ldev->tracker;
+	mutex_unlock(&lag_mutex);
+
+	if (tracker.is_bonded && !mlx5_lag_is_bonded(ldev)) {
+		if (mlx5_sriov_is_enabled(dev0) ||
+		    mlx5_sriov_is_enabled(dev1)) {
+			mlx5_core_warn(dev0, "LAG is not supported with SRIOV");
+			return;
+		}
+
+		for (i = 0; i < MLX5_MAX_PORTS; i++)
+			mlx5_remove_dev_by_protocol(ldev->pf[i].dev,
+						    MLX5_INTERFACE_PROTOCOL_IB);
+
+		mlx5_activate_lag(ldev, &tracker);
+
+		mlx5_add_dev_by_protocol(dev0, MLX5_INTERFACE_PROTOCOL_IB);
+		mlx5_nic_vport_enable_roce(dev1);
+	} else if (tracker.is_bonded && mlx5_lag_is_bonded(ldev)) {
+		mlx5_infer_tx_affinity_mapping(&tracker, &v2p_port1,
+					       &v2p_port2);
+
+		if ((v2p_port1 != ldev->v2p_map[0]) ||
+		    (v2p_port2 != ldev->v2p_map[1])) {
+			ldev->v2p_map[0] = v2p_port1;
+			ldev->v2p_map[1] = v2p_port2;
+
+			err = mlx5_cmd_modify_lag(dev0, v2p_port1, v2p_port2);
+			if (err)
+				mlx5_core_err(dev0,
+					      "Failed to modify LAG (%d)\n",
+					      err);
+		}
+	} else if (!tracker.is_bonded && mlx5_lag_is_bonded(ldev)) {
+		mlx5_remove_dev_by_protocol(dev0, MLX5_INTERFACE_PROTOCOL_IB);
+		mlx5_nic_vport_disable_roce(dev1);
+
+		mlx5_deactivate_lag(ldev);
+
+		for (i = 0; i < MLX5_MAX_PORTS; i++)
+			if (ldev->pf[i].dev)
+				mlx5_add_dev_by_protocol(ldev->pf[i].dev,
+							 MLX5_INTERFACE_PROTOCOL_IB);
+	}
+}
+
+static void mlx5_queue_bond_work(struct mlx5_lag *ldev, unsigned long delay)
+{
+	schedule_delayed_work(&ldev->bond_work, delay);
+}
+
+static void mlx5_do_bond_work(struct work_struct *work)
+{
+	struct delayed_work *delayed_work = to_delayed_work(work);
+	struct mlx5_lag *ldev = container_of(delayed_work, struct mlx5_lag,
+					     bond_work);
+	int status;
+
+	status = mutex_trylock(&mlx5_intf_mutex);
+	if (!status) {
+		/* 1 sec delay. */
+		mlx5_queue_bond_work(ldev, HZ);
+		return;
+	}
+
+	mlx5_do_bond(ldev);
+	mutex_unlock(&mlx5_intf_mutex);
+}
+
+static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
+					 struct lag_tracker *tracker,
+					 struct net_device *ndev,
+					 struct netdev_notifier_changeupper_info *info)
+{
+	struct net_device *upper = info->upper_dev, *ndev_tmp;
+	struct netdev_lag_upper_info *lag_upper_info;
+	bool is_bonded;
+	int bond_status = 0;
+	int num_slaves = 0;
+	int idx;
+
+	if (!netif_is_lag_master(upper))
+		return 0;
+
+	lag_upper_info = info->upper_info;
+
+	/* The event may still be of interest if the slave does not belong to
+	 * us, but is enslaved to a master which has one or more of our netdevs
+	 * as slaves (e.g., if a new slave is added to a master that bonds two
+	 * of our netdevs, we should unbond).
+	 */
+	rcu_read_lock();
+	for_each_netdev_in_bond_rcu(upper, ndev_tmp) {
+		idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev_tmp);
+		if (idx > -1)
+			bond_status |= (1 << idx);
+
+		num_slaves++;
+	}
+	rcu_read_unlock();
+
+	/* None of this lagdev's netdevs are slaves of this master. */
+	if (!(bond_status & 0x3))
+		return 0;
+
+	if (lag_upper_info)
+		tracker->tx_type = lag_upper_info->tx_type;
+
+	/* Determine bonding status:
+	 * A device is considered bonded if both its physical ports are slaves
+	 * of the same lag master, and only them.
+	 * Lag mode must be activebackup or hash.
+	 */
+	is_bonded = (num_slaves == MLX5_MAX_PORTS) &&
+		    (bond_status == 0x3) &&
+		    ((tracker->tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP) ||
+		     (tracker->tx_type == NETDEV_LAG_TX_TYPE_HASH));
+
+	if (tracker->is_bonded != is_bonded) {
+		tracker->is_bonded = is_bonded;
+		return 1;
+	}
+
+	return 0;
+}
+
+static int mlx5_handle_changelowerstate_event(struct mlx5_lag *ldev,
+					      struct lag_tracker *tracker,
+					      struct net_device *ndev,
+					      struct netdev_notifier_changelowerstate_info *info)
+{
+	struct netdev_lag_lower_state_info *lag_lower_info;
+	int idx;
+
+	if (!netif_is_lag_port(ndev))
+		return 0;
+
+	idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev);
+	if (idx == -1)
+		return 0;
+
+	/* This information is used to determine virtual to physical
+	 * port mapping.
+	 */
+	lag_lower_info = info->lower_state_info;
+	if (!lag_lower_info)
+		return 0;
+
+	tracker->netdev_state[idx] = *lag_lower_info;
+
+	return 1;
+}
+
+static int mlx5_lag_netdev_event(struct notifier_block *this,
+				 unsigned long event, void *ptr)
+{
+	struct net_device *ndev = netdev_notifier_info_to_dev(ptr);
+	struct lag_tracker tracker;
+	struct mlx5_lag *ldev;
+	int changed = 0;
+
+	if (!net_eq(dev_net(ndev), &init_net))
+		return NOTIFY_DONE;
+
+	if ((event != NETDEV_CHANGEUPPER) && (event != NETDEV_CHANGELOWERSTATE))
+		return NOTIFY_DONE;
+
+	ldev    = container_of(this, struct mlx5_lag, nb);
+	tracker = ldev->tracker;
+
+	switch (event) {
+	case NETDEV_CHANGEUPPER:
+		changed = mlx5_handle_changeupper_event(ldev, &tracker, ndev,
+							ptr);
+		break;
+	case NETDEV_CHANGELOWERSTATE:
+		changed = mlx5_handle_changelowerstate_event(ldev, &tracker,
+							     ndev, ptr);
+		break;
+	}
+
+	mutex_lock(&lag_mutex);
+	ldev->tracker = tracker;
+	mutex_unlock(&lag_mutex);
+
+	if (changed)
+		mlx5_queue_bond_work(ldev, 0);
+
+	return NOTIFY_DONE;
+}
+
+static struct mlx5_lag *mlx5_lag_dev_alloc(void)
+{
+	struct mlx5_lag *ldev;
+
+	ldev = kzalloc(sizeof(*ldev), GFP_KERNEL);
+	if (!ldev)
+		return NULL;
+
+	INIT_DELAYED_WORK(&ldev->bond_work, mlx5_do_bond_work);
+
+	return ldev;
+}
+
+static void mlx5_lag_dev_free(struct mlx5_lag *ldev)
+{
+	kfree(ldev);
+}
+
+static void mlx5_lag_dev_add_pf(struct mlx5_lag *ldev,
+				struct mlx5_core_dev *dev,
+				struct net_device *netdev)
+{
+	unsigned int fn = PCI_FUNC(dev->pdev->devfn);
+
+	if (fn >= MLX5_MAX_PORTS)
+		return;
+
+	mutex_lock(&lag_mutex);
+	ldev->pf[fn].dev    = dev;
+	ldev->pf[fn].netdev = netdev;
+	ldev->tracker.netdev_state[fn].link_up = 0;
+	ldev->tracker.netdev_state[fn].tx_enabled = 0;
+
+	dev->priv.lag = ldev;
+	mutex_unlock(&lag_mutex);
+}
+
+static void mlx5_lag_dev_remove_pf(struct mlx5_lag *ldev,
+				   struct mlx5_core_dev *dev)
+{
+	int i;
+
+	for (i = 0; i < MLX5_MAX_PORTS; i++)
+		if (ldev->pf[i].dev == dev)
+			break;
+
+	if (i == MLX5_MAX_PORTS)
+		return;
+
+	mutex_lock(&lag_mutex);
+	memset(&ldev->pf[i], 0, sizeof(*ldev->pf));
+
+	dev->priv.lag = NULL;
+	mutex_unlock(&lag_mutex);
+}
+
+static u16 mlx5_gen_pci_id(struct mlx5_core_dev *dev)
+{
+	return (u16)((dev->pdev->bus->number << 8) |
+		     PCI_SLOT(dev->pdev->devfn));
+}
+
+/* Must be called with intf_mutex held */
+void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev)
+{
+	struct mlx5_lag *ldev = NULL;
+	struct mlx5_core_dev *tmp_dev;
+	struct mlx5_priv *priv;
+	u16 pci_id;
+
+	if (!MLX5_CAP_GEN(dev, vport_group_manager) ||
+	    !MLX5_CAP_GEN(dev, lag_master) ||
+	    (MLX5_CAP_GEN(dev, num_lag_ports) != MLX5_MAX_PORTS))
+		return;
+
+	pci_id = mlx5_gen_pci_id(dev);
+
+	mlx5_core_for_each_priv(priv) {
+		tmp_dev = container_of(priv, struct mlx5_core_dev, priv);
+		if ((dev != tmp_dev) &&
+		    (mlx5_gen_pci_id(tmp_dev) == pci_id)) {
+			ldev = tmp_dev->priv.lag;
+			break;
+		}
+	}
+
+	if (!ldev) {
+		ldev = mlx5_lag_dev_alloc();
+		if (!ldev) {
+			mlx5_core_err(dev, "Failed to alloc lag dev\n");
+			return;
+		}
+	}
+
+	mlx5_lag_dev_add_pf(ldev, dev, netdev);
+
+	if (!ldev->nb.notifier_call) {
+		ldev->nb.notifier_call = mlx5_lag_netdev_event;
+		if (register_netdevice_notifier(&ldev->nb)) {
+			ldev->nb.notifier_call = NULL;
+			mlx5_core_err(dev, "Failed to register LAG netdev notifier\n");
+		}
+	}
+}
+
+/* Must be called with intf_mutex held */
+void mlx5_lag_remove(struct mlx5_core_dev *dev)
+{
+	struct mlx5_lag *ldev;
+	int i;
+
+	ldev = mlx5_lag_dev_get(dev);
+	if (!ldev)
+		return;
+
+	if (mlx5_lag_is_bonded(ldev))
+		mlx5_deactivate_lag(ldev);
+
+	mlx5_lag_dev_remove_pf(ldev, dev);
+
+	for (i = 0; i < MLX5_MAX_PORTS; i++)
+		if (ldev->pf[i].dev)
+			break;
+
+	if (i == MLX5_MAX_PORTS) {
+		if (ldev->nb.notifier_call)
+			unregister_netdevice_notifier(&ldev->nb);
+		cancel_delayed_work_sync(&ldev->bond_work);
+		mlx5_lag_dev_free(ldev);
+	}
+}
+
+bool mlx5_lag_is_active(struct mlx5_core_dev *dev)
+{
+	struct mlx5_lag *ldev;
+	bool res;
+
+	mutex_lock(&lag_mutex);
+	ldev = mlx5_lag_dev_get(dev);
+	res  = ldev && mlx5_lag_is_bonded(ldev);
+	mutex_unlock(&lag_mutex);
+
+	return res;
+}
+EXPORT_SYMBOL(mlx5_lag_is_active);
+
+struct net_device *mlx5_lag_get_roce_netdev(struct mlx5_core_dev *dev)
+{
+	struct net_device *ndev = NULL;
+	struct mlx5_lag *ldev;
+
+	mutex_lock(&lag_mutex);
+	ldev = mlx5_lag_dev_get(dev);
+
+	if (!(ldev && mlx5_lag_is_bonded(ldev)))
+		goto unlock;
+
+	if (ldev->tracker.tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP) {
+		ndev = ldev->tracker.netdev_state[0].tx_enabled ?
+		       ldev->pf[0].netdev : ldev->pf[1].netdev;
+	} else {
+		ndev = ldev->pf[0].netdev;
+	}
+	if (ndev)
+		dev_hold(ndev);
+
+unlock:
+	mutex_unlock(&lag_mutex);
+
+	return ndev;
+}
+EXPORT_SYMBOL(mlx5_lag_get_roce_netdev);
+
+bool mlx5_lag_intf_add(struct mlx5_interface *intf, struct mlx5_priv *priv)
+{
+	struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev,
+						 priv);
+	struct mlx5_lag *ldev;
+
+	if (intf->protocol != MLX5_INTERFACE_PROTOCOL_IB)
+		return true;
+
+	ldev = mlx5_lag_dev_get(dev);
+	if (!ldev || !mlx5_lag_is_bonded(ldev) || ldev->pf[0].dev == dev)
+		return true;
+
+	/* If bonded, we do not add an IB device for PF1. */
+	return false;
+}
+

drivers/net/ethernet/mellanox/mlx5/core/main.c

@@ -73,8 +73,9 @@ module_param_named(prof_sel, prof_sel, int, 0444);
 MODULE_PARM_DESC(prof_sel, "profile selector. Valid range 0 - 2");
 
 static LIST_HEAD(intf_list);
-static LIST_HEAD(dev_list);
-static DEFINE_MUTEX(intf_mutex);
+
+LIST_HEAD(mlx5_dev_list);
+DEFINE_MUTEX(mlx5_intf_mutex);
 
 struct mlx5_device_context {
 	struct list_head	list;
@@ -782,6 +783,9 @@ static void mlx5_add_device(struct mlx5_interface *intf, struct mlx5_priv *priv)
 	struct mlx5_device_context *dev_ctx;
 	struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev, priv);
 
+	if (!mlx5_lag_intf_add(intf, priv))
+		return;
+
 	dev_ctx = kmalloc(sizeof(*dev_ctx), GFP_KERNEL);
 	if (!dev_ctx)
 		return;
@@ -820,11 +824,11 @@ static int mlx5_register_device(struct mlx5_core_dev *dev)
 	struct mlx5_priv *priv = &dev->priv;
 	struct mlx5_interface *intf;
 
-	mutex_lock(&intf_mutex);
-	list_add_tail(&priv->dev_list, &dev_list);
+	mutex_lock(&mlx5_intf_mutex);
+	list_add_tail(&priv->dev_list, &mlx5_dev_list);
 	list_for_each_entry(intf, &intf_list, list)
 		mlx5_add_device(intf, priv);
-	mutex_unlock(&intf_mutex);
+	mutex_unlock(&mlx5_intf_mutex);
 
 	return 0;
 }
@@ -834,11 +838,11 @@ static void mlx5_unregister_device(struct mlx5_core_dev *dev)
 	struct mlx5_priv *priv = &dev->priv;
 	struct mlx5_interface *intf;
 
-	mutex_lock(&intf_mutex);
+	mutex_lock(&mlx5_intf_mutex);
 	list_for_each_entry(intf, &intf_list, list)
 		mlx5_remove_device(intf, priv);
 	list_del(&priv->dev_list);
-	mutex_unlock(&intf_mutex);
+	mutex_unlock(&mlx5_intf_mutex);
 }
 
 int mlx5_register_interface(struct mlx5_interface *intf)
@@ -848,11 +852,11 @@ int mlx5_register_interface(struct mlx5_interface *intf)
 	if (!intf->add || !intf->remove)
 		return -EINVAL;
 
-	mutex_lock(&intf_mutex);
+	mutex_lock(&mlx5_intf_mutex);
 	list_add_tail(&intf->list, &intf_list);
-	list_for_each_entry(priv, &dev_list, dev_list)
+	list_for_each_entry(priv, &mlx5_dev_list, dev_list)
 		mlx5_add_device(intf, priv);
-	mutex_unlock(&intf_mutex);
+	mutex_unlock(&mlx5_intf_mutex);
 
 	return 0;
 }
@@ -862,11 +866,11 @@ void mlx5_unregister_interface(struct mlx5_interface *intf)
 {
 	struct mlx5_priv *priv;
 
-	mutex_lock(&intf_mutex);
-	list_for_each_entry(priv, &dev_list, dev_list)
+	mutex_lock(&mlx5_intf_mutex);
+	list_for_each_entry(priv, &mlx5_dev_list, dev_list)
 		mlx5_remove_device(intf, priv);
 	list_del(&intf->list);
-	mutex_unlock(&intf_mutex);
+	mutex_unlock(&mlx5_intf_mutex);
 }
 EXPORT_SYMBOL(mlx5_unregister_interface);
 
@@ -892,6 +896,30 @@ void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol)
 }
 EXPORT_SYMBOL(mlx5_get_protocol_dev);
 
+/* Must be called with intf_mutex held */
+void mlx5_add_dev_by_protocol(struct mlx5_core_dev *dev, int protocol)
+{
+	struct mlx5_interface *intf;
+
+	list_for_each_entry(intf, &intf_list, list)
+		if (intf->protocol == protocol) {
+			mlx5_add_device(intf, &dev->priv);
+			break;
+		}
+}
+
+/* Must be called with intf_mutex held */
+void mlx5_remove_dev_by_protocol(struct mlx5_core_dev *dev, int protocol)
+{
+	struct mlx5_interface *intf;
+
+	list_for_each_entry(intf, &intf_list, list)
+		if (intf->protocol == protocol) {
+			mlx5_remove_device(intf, &dev->priv);
+			break;
+		}
+}
+
 static int mlx5_pci_init(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 {
 	struct pci_dev *pdev = dev->pdev;

drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h

@@ -46,6 +46,9 @@
 
 extern int mlx5_core_debug_mask;
 
+extern struct list_head mlx5_dev_list;
+extern struct mutex mlx5_intf_mutex;
+
 #define mlx5_core_dbg(__dev, format, ...)				\
 	dev_dbg(&(__dev)->pdev->dev, "%s:%s:%d:(pid %d): " format,	\
 		 (__dev)->priv.name, __func__, __LINE__, current->pid,	\
@@ -70,6 +73,9 @@ do {									\
 #define mlx5_core_info(__dev, format, ...)				\
 	dev_info(&(__dev)->pdev->dev, format, ##__VA_ARGS__)
 
+#define mlx5_core_for_each_priv(__priv)					\
+	list_for_each_entry(__priv, &mlx5_dev_list, dev_list)
+
 enum {
 	MLX5_CMD_DATA, /* print command payload only */
 	MLX5_CMD_TIME, /* print command execution time */
@@ -84,6 +90,7 @@ void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event,
 void mlx5_enter_error_state(struct mlx5_core_dev *dev);
 void mlx5_disable_device(struct mlx5_core_dev *dev);
 int mlx5_core_sriov_configure(struct pci_dev *dev, int num_vfs);
+bool mlx5_sriov_is_enabled(struct mlx5_core_dev *dev);
 int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id);
 int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id);
 int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev);
@@ -92,7 +99,27 @@ u32 mlx5_get_msix_vec(struct mlx5_core_dev *dev, int vecidx);
 struct mlx5_eq *mlx5_eqn2eq(struct mlx5_core_dev *dev, int eqn);
 void mlx5_cq_tasklet_cb(unsigned long data);
 
+void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev);
+void mlx5_lag_remove(struct mlx5_core_dev *dev);
+
+void mlx5_add_dev_by_protocol(struct mlx5_core_dev *dev, int protocol);
+void mlx5_remove_dev_by_protocol(struct mlx5_core_dev *dev, int protocol);
+
+bool mlx5_lag_intf_add(struct mlx5_interface *intf, struct mlx5_priv *priv);
+
 void mlx5e_init(void);
 void mlx5e_cleanup(void);
 
+static inline int mlx5_lag_is_lacp_owner(struct mlx5_core_dev *dev)
+{
+	/* LACP owner conditions:
+	 * 1) Function is physical.
+	 * 2) LAG is supported by FW.
+	 * 3) LAG is managed by driver (currently the only option).
+	 */
+	return  MLX5_CAP_GEN(dev, vport_group_manager) &&
+		   (MLX5_CAP_GEN(dev, num_lag_ports) > 1) &&
+		    MLX5_CAP_GEN(dev, lag_master);
+}
+
 #endif /* __MLX5_CORE_H__ */

drivers/net/ethernet/mellanox/mlx5/core/port.c

@@ -175,25 +175,39 @@ int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
 }
 EXPORT_SYMBOL_GPL(mlx5_query_port_link_width_oper);
 
-int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev,
-			       u8 *proto_oper, int proto_mask,
-			       u8 local_port)
+int mlx5_query_port_eth_proto_oper(struct mlx5_core_dev *dev,
+				   u32 *proto_oper, u8 local_port)
 {
 	u32 out[MLX5_ST_SZ_DW(ptys_reg)];
 	int err;
 
-	err = mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, local_port);
+	err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN,
+				   local_port);
 	if (err)
 		return err;
 
-	if (proto_mask == MLX5_PTYS_EN)
-		*proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
-	else
-		*proto_oper = MLX5_GET(ptys_reg, out, ib_proto_oper);
+	*proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
+
+	return 0;
+}
+EXPORT_SYMBOL(mlx5_query_port_eth_proto_oper);
+
+int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev,
+				  u8 *proto_oper, u8 local_port)
+{
+	u32 out[MLX5_ST_SZ_DW(ptys_reg)];
+	int err;
+
+	err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_IB,
+				   local_port);
+	if (err)
+		return err;
+
+	*proto_oper = MLX5_GET(ptys_reg, out, ib_proto_oper);
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(mlx5_query_port_proto_oper);
+EXPORT_SYMBOL(mlx5_query_port_ib_proto_oper);
 
 int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable,
 		       u32 proto_admin, int proto_mask)

drivers/net/ethernet/mellanox/mlx5/core/sriov.c

@@ -37,6 +37,13 @@
 #include "eswitch.h"
 #endif
 
+bool mlx5_sriov_is_enabled(struct mlx5_core_dev *dev)
+{
+	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+
+	return !!sriov->num_vfs;
+}
+
 static void enable_vfs(struct mlx5_core_dev *dev, int num_vfs)
 {
 	struct mlx5_core_sriov *sriov = &dev->priv.sriov;
@@ -144,6 +151,11 @@ int mlx5_core_sriov_configure(struct pci_dev *pdev, int num_vfs)
 	if (!mlx5_core_is_pf(dev))
 		return -EPERM;
 
+	if (num_vfs && mlx5_lag_is_active(dev)) {
+		mlx5_core_warn(dev, "can't turn sriov on while LAG is active");
+		return -EINVAL;
+	}
+
 	mlx5_core_cleanup_vfs(dev);
 
 	if (!num_vfs) {

include/linux/mlx5/device.h

@@ -964,6 +964,18 @@ enum mlx5_cap_type {
 #define MLX5_CAP_FLOWTABLE_NIC_RX_MAX(mdev, cap) \
 	MLX5_CAP_FLOWTABLE_MAX(mdev, flow_table_properties_nic_receive.cap)
 
+#define MLX5_CAP_FLOWTABLE_SNIFFER_RX(mdev, cap) \
+	MLX5_CAP_FLOWTABLE(mdev, flow_table_properties_nic_receive_sniffer.cap)
+
+#define MLX5_CAP_FLOWTABLE_SNIFFER_RX_MAX(mdev, cap) \
+	MLX5_CAP_FLOWTABLE_MAX(mdev, flow_table_properties_nic_receive_sniffer.cap)
+
+#define MLX5_CAP_FLOWTABLE_SNIFFER_TX(mdev, cap) \
+	MLX5_CAP_FLOWTABLE(mdev, flow_table_properties_nic_transmit_sniffer.cap)
+
+#define MLX5_CAP_FLOWTABLE_SNIFFER_TX_MAX(mdev, cap) \
+	MLX5_CAP_FLOWTABLE_MAX(mdev, flow_table_properties_nic_transmit_sniffer.cap)
+
 #define MLX5_CAP_ESW_FLOWTABLE(mdev, cap) \
 	MLX5_GET(flow_table_eswitch_cap, \
 		 mdev->hca_caps_cur[MLX5_CAP_ESWITCH_FLOW_TABLE], cap)

include/linux/mlx5/driver.h

@@ -48,10 +48,6 @@
 #include <linux/mlx5/doorbell.h>
 #include <linux/mlx5/srq.h>
 
-enum {
-	MLX5_RQ_BITMASK_VSD = 1 << 1,
-};
-
 enum {
 	MLX5_BOARD_ID_LEN = 64,
 	MLX5_MAX_NAME_LEN = 16,
@@ -481,6 +477,7 @@ struct mlx5_fc_stats {
 };
 
 struct mlx5_eswitch;
+struct mlx5_lag;
 
 struct mlx5_rl_entry {
 	u32                     rate;
@@ -554,6 +551,7 @@ struct mlx5_priv {
 	struct mlx5_flow_steering *steering;
 	struct mlx5_eswitch     *eswitch;
 	struct mlx5_core_sriov	sriov;
+	struct mlx5_lag		*lag;
 	unsigned long		pci_dev_data;
 	struct mlx5_fc_stats		fc_stats;
 	struct mlx5_rl_table            rl_table;
@@ -946,6 +944,11 @@ int mlx5_register_interface(struct mlx5_interface *intf);
 void mlx5_unregister_interface(struct mlx5_interface *intf);
 int mlx5_core_query_vendor_id(struct mlx5_core_dev *mdev, u32 *vendor_id);
 
+int mlx5_cmd_create_vport_lag(struct mlx5_core_dev *dev);
+int mlx5_cmd_destroy_vport_lag(struct mlx5_core_dev *dev);
+bool mlx5_lag_is_active(struct mlx5_core_dev *dev);
+struct net_device *mlx5_lag_get_roce_netdev(struct mlx5_core_dev *dev);
+
 struct mlx5_profile {
 	u64	mask;
 	u8	log_max_qp;

include/linux/mlx5/fs.h

@@ -54,6 +54,7 @@ static inline void build_leftovers_ft_param(int *priority,
 
 enum mlx5_flow_namespace_type {
 	MLX5_FLOW_NAMESPACE_BYPASS,
+	MLX5_FLOW_NAMESPACE_LAG,
 	MLX5_FLOW_NAMESPACE_OFFLOADS,
 	MLX5_FLOW_NAMESPACE_ETHTOOL,
 	MLX5_FLOW_NAMESPACE_KERNEL,
@@ -62,6 +63,8 @@ enum mlx5_flow_namespace_type {
 	MLX5_FLOW_NAMESPACE_FDB,
 	MLX5_FLOW_NAMESPACE_ESW_EGRESS,
 	MLX5_FLOW_NAMESPACE_ESW_INGRESS,
+	MLX5_FLOW_NAMESPACE_SNIFFER_RX,
+	MLX5_FLOW_NAMESPACE_SNIFFER_TX,
 };
 
 struct mlx5_flow_table;
@@ -106,6 +109,9 @@ mlx5_create_vport_flow_table(struct mlx5_flow_namespace *ns,
 			     int prio,
 			     int num_flow_table_entries,
 			     u32 level, u16 vport);
+struct mlx5_flow_table *mlx5_create_lag_demux_flow_table(
+					       struct mlx5_flow_namespace *ns,
+					       int prio, u32 level);
 int mlx5_destroy_flow_table(struct mlx5_flow_table *ft);
 
 /* inbox should be set with the following values:

include/linux/mlx5/mlx5_ifc.h

@@ -174,6 +174,12 @@ enum {
 	MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY         = 0x82b,
 	MLX5_CMD_OP_SET_WOL_ROL                   = 0x830,
 	MLX5_CMD_OP_QUERY_WOL_ROL                 = 0x831,
+	MLX5_CMD_OP_CREATE_LAG                    = 0x840,
+	MLX5_CMD_OP_MODIFY_LAG                    = 0x841,
+	MLX5_CMD_OP_QUERY_LAG                     = 0x842,
+	MLX5_CMD_OP_DESTROY_LAG                   = 0x843,
+	MLX5_CMD_OP_CREATE_VPORT_LAG              = 0x844,
+	MLX5_CMD_OP_DESTROY_VPORT_LAG             = 0x845,
 	MLX5_CMD_OP_CREATE_TIR                    = 0x900,
 	MLX5_CMD_OP_MODIFY_TIR                    = 0x901,
 	MLX5_CMD_OP_DESTROY_TIR                   = 0x902,
@@ -477,7 +483,9 @@ struct mlx5_ifc_ads_bits {
 
 struct mlx5_ifc_flow_table_nic_cap_bits {
 	u8         nic_rx_multi_path_tirs[0x1];
-	u8         reserved_at_1[0x1ff];
+	u8         nic_rx_multi_path_tirs_fts[0x1];
+	u8         allow_sniffer_and_nic_rx_shared_tir[0x1];
+	u8         reserved_at_3[0x1fd];
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_receive;
 
@@ -779,7 +787,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         out_of_seq_cnt[0x1];
 	u8         vport_counters[0x1];
 	u8         retransmission_q_counters[0x1];
-	u8         reserved_at_183[0x3];
+	u8         reserved_at_183[0x1];
+	u8         modify_rq_counter_set_id[0x1];
+	u8         reserved_at_185[0x1];
 	u8         max_qp_cnt[0xa];
 	u8         pkey_table_size[0x10];
 
@@ -882,7 +892,10 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         pad_tx_eth_packet[0x1];
 	u8         reserved_at_263[0x8];
 	u8         log_bf_reg_size[0x5];
-	u8         reserved_at_270[0x10];
+
+	u8         reserved_at_270[0xb];
+	u8         lag_master[0x1];
+	u8         num_lag_ports[0x4];
 
 	u8         reserved_at_280[0x10];
 	u8         max_wqe_sz_sq[0x10];
@@ -1916,7 +1929,7 @@ enum {
 
 struct mlx5_ifc_qpc_bits {
 	u8         state[0x4];
-	u8         reserved_at_4[0x4];
+	u8         lag_tx_port_affinity[0x4];
 	u8         st[0x8];
 	u8         reserved_at_10[0x3];
 	u8         pm_state[0x2];
@@ -2165,7 +2178,11 @@ struct mlx5_ifc_traffic_counter_bits {
 };
 
 struct mlx5_ifc_tisc_bits {
-	u8         reserved_at_0[0xc];
+	u8         strict_lag_tx_port_affinity[0x1];
+	u8         reserved_at_1[0x3];
+	u8         lag_tx_port_affinity[0x04];
+
+	u8         reserved_at_8[0x4];
 	u8         prio[0x4];
 	u8         reserved_at_10[0x10];
 
@@ -2827,7 +2844,7 @@ struct mlx5_ifc_xrqc_bits {
 
 	struct mlx5_ifc_tag_matching_topology_context_bits tag_matching_topology_context;
 
-	u8         reserved_at_180[0x180];
+	u8         reserved_at_180[0x200];
 
 	struct mlx5_ifc_wq_bits wq;
 };
@@ -4615,7 +4632,9 @@ struct mlx5_ifc_modify_tis_out_bits {
 struct mlx5_ifc_modify_tis_bitmask_bits {
 	u8         reserved_at_0[0x20];
 
-	u8         reserved_at_20[0x1f];
+	u8         reserved_at_20[0x1d];
+	u8         lag_tx_port_affinity[0x1];
+	u8         strict_lag_tx_port_affinity[0x1];
 	u8         prio[0x1];
 };
 
@@ -4750,6 +4769,11 @@ struct mlx5_ifc_modify_rq_out_bits {
 	u8         reserved_at_40[0x40];
 };
 
+enum {
+	MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_VSD = 1ULL << 1,
+	MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_MODIFY_RQ_COUNTER_SET_ID = 1ULL << 3,
+};
+
 struct mlx5_ifc_modify_rq_in_bits {
 	u8         opcode[0x10];
 	u8         reserved_at_10[0x10];
@@ -6208,7 +6232,10 @@ struct mlx5_ifc_create_flow_table_in_bits {
 	u8         reserved_at_e0[0x8];
 	u8         table_miss_id[0x18];
 
-	u8         reserved_at_100[0x100];
+	u8         reserved_at_100[0x8];
+	u8         lag_master_next_table_id[0x18];
+
+	u8         reserved_at_120[0x80];
 };
 
 struct mlx5_ifc_create_flow_group_out_bits {
@@ -7662,7 +7689,8 @@ struct mlx5_ifc_set_flow_table_root_in_bits {
 };
 
 enum {
-	MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID = 0x1,
+	MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID     = (1UL << 0),
+	MLX5_MODIFY_FLOW_TABLE_LAG_NEXT_TABLE_ID = (1UL << 15),
 };
 
 struct mlx5_ifc_modify_flow_table_out_bits {
@@ -7701,7 +7729,10 @@ struct mlx5_ifc_modify_flow_table_in_bits {
 	u8         reserved_at_e0[0x8];
 	u8         table_miss_id[0x18];
 
-	u8         reserved_at_100[0x100];
+	u8         reserved_at_100[0x8];
+	u8         lag_master_next_table_id[0x18];
+
+	u8         reserved_at_120[0x80];
 };
 
 struct mlx5_ifc_ets_tcn_config_reg_bits {
@@ -7809,4 +7840,134 @@ struct mlx5_ifc_dcbx_param_bits {
 	u8         error[0x8];
 	u8         reserved_at_a0[0x160];
 };
+
+struct mlx5_ifc_lagc_bits {
+	u8         reserved_at_0[0x1d];
+	u8         lag_state[0x3];
+
+	u8         reserved_at_20[0x14];
+	u8         tx_remap_affinity_2[0x4];
+	u8         reserved_at_38[0x4];
+	u8         tx_remap_affinity_1[0x4];
+};
+
+struct mlx5_ifc_create_lag_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_create_lag_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	struct mlx5_ifc_lagc_bits ctx;
+};
+
+struct mlx5_ifc_modify_lag_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_modify_lag_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x20];
+	u8         field_select[0x20];
+
+	struct mlx5_ifc_lagc_bits ctx;
+};
+
+struct mlx5_ifc_query_lag_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+
+	struct mlx5_ifc_lagc_bits ctx;
+};
+
+struct mlx5_ifc_query_lag_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_lag_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_lag_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_create_vport_lag_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_create_vport_lag_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_vport_lag_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_vport_lag_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x40];
+};
+
 #endif /* MLX5_IFC_H */

include/linux/mlx5/port.h

@@ -61,6 +61,39 @@ enum mlx5_an_status {
 #define MLX5_I2C_ADDR_HIGH		0x51
 #define MLX5_EEPROM_PAGE_LENGTH		256
 
+enum mlx5e_link_mode {
+	MLX5E_1000BASE_CX_SGMII	 = 0,
+	MLX5E_1000BASE_KX	 = 1,
+	MLX5E_10GBASE_CX4	 = 2,
+	MLX5E_10GBASE_KX4	 = 3,
+	MLX5E_10GBASE_KR	 = 4,
+	MLX5E_20GBASE_KR2	 = 5,
+	MLX5E_40GBASE_CR4	 = 6,
+	MLX5E_40GBASE_KR4	 = 7,
+	MLX5E_56GBASE_R4	 = 8,
+	MLX5E_10GBASE_CR	 = 12,
+	MLX5E_10GBASE_SR	 = 13,
+	MLX5E_10GBASE_ER	 = 14,
+	MLX5E_40GBASE_SR4	 = 15,
+	MLX5E_40GBASE_LR4	 = 16,
+	MLX5E_50GBASE_SR2	 = 18,
+	MLX5E_100GBASE_CR4	 = 20,
+	MLX5E_100GBASE_SR4	 = 21,
+	MLX5E_100GBASE_KR4	 = 22,
+	MLX5E_100GBASE_LR4	 = 23,
+	MLX5E_100BASE_TX	 = 24,
+	MLX5E_1000BASE_T	 = 25,
+	MLX5E_10GBASE_T		 = 26,
+	MLX5E_25GBASE_CR	 = 27,
+	MLX5E_25GBASE_KR	 = 28,
+	MLX5E_25GBASE_SR	 = 29,
+	MLX5E_50GBASE_CR2	 = 30,
+	MLX5E_50GBASE_KR2	 = 31,
+	MLX5E_LINK_MODES_NUMBER,
+};
+
+#define MLX5E_PROT_MASK(link_mode) (1 << link_mode)
+
 int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps);
 int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys,
 			 int ptys_size, int proto_mask, u8 local_port);
@@ -70,9 +103,10 @@ int mlx5_query_port_proto_admin(struct mlx5_core_dev *dev,
 				u32 *proto_admin, int proto_mask);
 int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
 				    u8 *link_width_oper, u8 local_port);
-int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev,
-			       u8 *proto_oper, int proto_mask,
-			       u8 local_port);
+int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev,
+				  u8 *proto_oper, u8 local_port);
+int mlx5_query_port_eth_proto_oper(struct mlx5_core_dev *dev,
+				   u32 *proto_oper, u8 local_port);
 int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable,
 		       u32 proto_admin, int proto_mask);
 void mlx5_toggle_port_link(struct mlx5_core_dev *dev);