Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux

1) New generic devlink param "enable_roce", for downstream devlink
   reload support

2) Do vport ACL configuration on per vport basis when
   enabling/disabling a vport. This enables to have vports enabled/disabled
   outside of eswitch config for future

3) Split the code for legacy vs offloads mode and make it clear

4) Tide up vport locking and workqueue usage

5) Fix metadata enablement for ECPF

6) Make explicit use of VF property to publish IB_DEVICE_VIRTUAL_FUNCTION

7) E-Switch and flow steering core low level support and refactoring for
   netfilter flowtables offload
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Documentation/networking/device_drivers/mellanox/mlx5.rst

@@ -154,6 +154,27 @@ User command examples:
       values:
          cmode runtime value smfs
 
+enable_roce: RoCE enablement state
+----------------------------------
+RoCE enablement state controls driver support for RoCE traffic.
+When RoCE is disabled, there is no gid table, only raw ethernet QPs are supported and traffic on the well known UDP RoCE port is handled as raw ethernet traffic.
+
+To change RoCE enablement state a user must change the driverinit cmode value and run devlink reload.
+
+User command examples:
+
+- Disable RoCE::
+
+    $ devlink dev param set pci/0000:06:00.0 name enable_roce value false cmode driverinit
+    $ devlink dev reload pci/0000:06:00.0
+
+- Read RoCE enablement state::
+
+    $ devlink dev param show pci/0000:06:00.0 name enable_roce
+      pci/0000:06:00.0:
+      name enable_roce type generic
+      values:
+         cmode driverinit value true
 
 Devlink health reporters
 ========================

Documentation/networking/devlink-params-mlx5.txt

+flow_steering_mode	[DEVICE, DRIVER-SPECIFIC]
+			Controls the flow steering mode of the driver.
+			Two modes are supported:
+			1. 'dmfs' - Device managed flow steering.
+			2. 'smfs  - Software/Driver managed flow steering.
+			In DMFS mode, the HW steering entities are created and
+			managed through the Firmware.
+			In SMFS mode, the HW steering entities are created and
+			managed though by the driver directly into Hardware
+			without firmware intervention.
+			Type: String
+			Configuration mode: runtime
+
+enable_roce		[DEVICE, GENERIC]
+			Enable handling of RoCE traffic in the device.
+			Defaultly enabled.
+			Configuration mode: driverinit

Documentation/networking/devlink-params.txt

@@ -65,3 +65,7 @@ reset_dev_on_drv_probe	[DEVICE, GENERIC]
 			  Reset only if device firmware can be found in the
 			  filesystem.
 			Type: u8
+
+enable_roce		[DEVICE, GENERIC]
+			Enable handling of RoCE traffic in the device.
+			Type: Boolean

drivers/infiniband/hw/mlx5/ib_rep.c

@@ -35,7 +35,7 @@ mlx5_ib_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
 	int vport_index;
 
 	if (rep->vport == MLX5_VPORT_UPLINK)
-		profile = &uplink_rep_profile;
+		profile = &raw_eth_profile;
 	else
 		return mlx5_ib_set_vport_rep(dev, rep);
 

drivers/infiniband/hw/mlx5/ib_rep.h

@@ -10,7 +10,7 @@
 #include "mlx5_ib.h"
 
 #ifdef CONFIG_MLX5_ESWITCH
-extern const struct mlx5_ib_profile uplink_rep_profile;
+extern const struct mlx5_ib_profile raw_eth_profile;
 
 u8 mlx5_ib_eswitch_mode(struct mlx5_eswitch *esw);
 struct mlx5_ib_dev *mlx5_ib_get_rep_ibdev(struct mlx5_eswitch *esw,

drivers/infiniband/hw/mlx5/main.c

@@ -1031,7 +1031,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 	if (MLX5_CAP_GEN(mdev, cd))
 		props->device_cap_flags |= IB_DEVICE_CROSS_CHANNEL;
 
-	if (!mlx5_core_is_pf(mdev))
+	if (mlx5_core_is_vf(mdev))
 		props->device_cap_flags |= IB_DEVICE_VIRTUAL_FUNCTION;
 
 	if (mlx5_ib_port_link_layer(ibdev, 1) ==
@@ -5145,7 +5145,6 @@ static int mlx5_port_immutable(struct ib_device *ibdev, u8 port_num,
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
 	immutable->core_cap_flags = get_core_cap_flags(ibdev, &rep);
-	if ((ll == IB_LINK_LAYER_INFINIBAND) || MLX5_CAP_GEN(dev->mdev, roce))
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
@@ -5249,11 +5248,9 @@ static int mlx5_enable_eth(struct mlx5_ib_dev *dev)
 {
 	int err;
 
-	if (MLX5_CAP_GEN(dev->mdev, roce)) {
 	err = mlx5_nic_vport_enable_roce(dev->mdev);
 	if (err)
 		return err;
-	}
 
 	err = mlx5_eth_lag_init(dev);
 	if (err)
@@ -5262,7 +5259,6 @@ static int mlx5_enable_eth(struct mlx5_ib_dev *dev)
 	return 0;
 
 err_disable_roce:
-	if (MLX5_CAP_GEN(dev->mdev, roce))
 	mlx5_nic_vport_disable_roce(dev->mdev);
 
 	return err;
@@ -5271,7 +5267,6 @@ static int mlx5_enable_eth(struct mlx5_ib_dev *dev)
 static void mlx5_disable_eth(struct mlx5_ib_dev *dev)
 {
 	mlx5_eth_lag_cleanup(dev);
-	if (MLX5_CAP_GEN(dev->mdev, roce))
 	mlx5_nic_vport_disable_roce(dev->mdev);
 }
 
@@ -6444,7 +6439,7 @@ static const struct ib_device_ops mlx5_ib_dev_port_rep_ops = {
 	.query_port = mlx5_ib_rep_query_port,
 };
 
-static int mlx5_ib_stage_rep_non_default_cb(struct mlx5_ib_dev *dev)
+static int mlx5_ib_stage_raw_eth_non_default_cb(struct mlx5_ib_dev *dev)
 {
 	ib_set_device_ops(&dev->ib_dev, &mlx5_ib_dev_port_rep_ops);
 	return 0;
@@ -6484,7 +6479,7 @@ static void mlx5_ib_stage_common_roce_cleanup(struct mlx5_ib_dev *dev)
 	mlx5_remove_netdev_notifier(dev, port_num);
 }
 
-static int mlx5_ib_stage_rep_roce_init(struct mlx5_ib_dev *dev)
+static int mlx5_ib_stage_raw_eth_roce_init(struct mlx5_ib_dev *dev)
 {
 	struct mlx5_core_dev *mdev = dev->mdev;
 	enum rdma_link_layer ll;
@@ -6500,7 +6495,7 @@ static int mlx5_ib_stage_rep_roce_init(struct mlx5_ib_dev *dev)
 	return err;
 }
 
-static void mlx5_ib_stage_rep_roce_cleanup(struct mlx5_ib_dev *dev)
+static void mlx5_ib_stage_raw_eth_roce_cleanup(struct mlx5_ib_dev *dev)
 {
 	mlx5_ib_stage_common_roce_cleanup(dev);
 }
@@ -6807,7 +6802,7 @@ static const struct mlx5_ib_profile pf_profile = {
 		     mlx5_ib_stage_delay_drop_cleanup),
 };
 
-const struct mlx5_ib_profile uplink_rep_profile = {
+const struct mlx5_ib_profile raw_eth_profile = {
 	STAGE_CREATE(MLX5_IB_STAGE_INIT,
 		     mlx5_ib_stage_init_init,
 		     mlx5_ib_stage_init_cleanup),
@@ -6818,11 +6813,11 @@ const struct mlx5_ib_profile uplink_rep_profile = {
 		     mlx5_ib_stage_caps_init,
 		     NULL),
 	STAGE_CREATE(MLX5_IB_STAGE_NON_DEFAULT_CB,
-		     mlx5_ib_stage_rep_non_default_cb,
+		     mlx5_ib_stage_raw_eth_non_default_cb,
 		     NULL),
 	STAGE_CREATE(MLX5_IB_STAGE_ROCE,
-		     mlx5_ib_stage_rep_roce_init,
-		     mlx5_ib_stage_rep_roce_cleanup),
+		     mlx5_ib_stage_raw_eth_roce_init,
+		     mlx5_ib_stage_raw_eth_roce_cleanup),
 	STAGE_CREATE(MLX5_IB_STAGE_SRQ,
 		     mlx5_init_srq_table,
 		     mlx5_cleanup_srq_table),
@@ -6898,6 +6893,7 @@ static void *mlx5_ib_add_slave_port(struct mlx5_core_dev *mdev)
 
 static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
 {
+	const struct mlx5_ib_profile *profile;
 	enum rdma_link_layer ll;
 	struct mlx5_ib_dev *dev;
 	int port_type_cap;
@@ -6933,7 +6929,12 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
 	dev->mdev = mdev;
 	dev->num_ports = num_ports;
 
-	return __mlx5_ib_add(dev, &pf_profile);
+	if (ll == IB_LINK_LAYER_ETHERNET && !mlx5_is_roce_enabled(mdev))
+		profile = &raw_eth_profile;
+	else
+		profile = &pf_profile;
+
+	return __mlx5_ib_add(dev, profile);
 }
 
 static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context)

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

@@ -177,12 +177,29 @@ enum mlx5_devlink_param_id {
 	MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE,
 };
 
+static int mlx5_devlink_enable_roce_validate(struct devlink *devlink, u32 id,
+					     union devlink_param_value val,
+					     struct netlink_ext_ack *extack)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+	bool new_state = val.vbool;
+
+	if (new_state && !MLX5_CAP_GEN(dev, roce)) {
+		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support RoCE");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
 static const struct devlink_param mlx5_devlink_params[] = {
 	DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE,
 			     "flow_steering_mode", DEVLINK_PARAM_TYPE_STRING,
 			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
 			     mlx5_devlink_fs_mode_get, mlx5_devlink_fs_mode_set,
 			     mlx5_devlink_fs_mode_validate),
+	DEVLINK_PARAM_GENERIC(ENABLE_ROCE, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
+			      NULL, NULL, mlx5_devlink_enable_roce_validate),
 };
 
 static void mlx5_devlink_set_params_init_values(struct devlink *devlink)
@@ -197,6 +214,11 @@ static void mlx5_devlink_set_params_init_values(struct devlink *devlink)
 	devlink_param_driverinit_value_set(devlink,
 					   MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE,
 					   value);
+
+	value.vbool = MLX5_CAP_GEN(dev, roce);
+	devlink_param_driverinit_value_set(devlink,
+					   DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
+					   value);
 }
 
 int mlx5_devlink_register(struct devlink *devlink, struct device *dev)

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

@@ -1074,7 +1074,7 @@ mlx5e_tc_offload_to_slow_path(struct mlx5_eswitch *esw,
 	memcpy(slow_attr, flow->esw_attr, sizeof(*slow_attr));
 	slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	slow_attr->split_count = 0;
-	slow_attr->dest_chain = FDB_SLOW_PATH_CHAIN;
+	slow_attr->dest_chain = FDB_TC_SLOW_PATH_CHAIN;
 
 	rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, slow_attr);
 	if (!IS_ERR(rule))
@@ -1091,7 +1091,7 @@ mlx5e_tc_unoffload_from_slow_path(struct mlx5_eswitch *esw,
 	memcpy(slow_attr, flow->esw_attr, sizeof(*slow_attr));
 	slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	slow_attr->split_count = 0;
-	slow_attr->dest_chain = FDB_SLOW_PATH_CHAIN;
+	slow_attr->dest_chain = FDB_TC_SLOW_PATH_CHAIN;
 	mlx5e_tc_unoffload_fdb_rules(esw, flow, slow_attr);
 	flow_flag_clear(flow, SLOW);
 }

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

@@ -43,6 +43,16 @@
 #include <linux/mlx5/fs.h>
 #include "lib/mpfs.h"
 
+#define FDB_TC_MAX_CHAIN 3
+#define FDB_FT_CHAIN (FDB_TC_MAX_CHAIN + 1)
+#define FDB_TC_SLOW_PATH_CHAIN (FDB_FT_CHAIN + 1)
+
+/* The index of the last real chain (FT) + 1 as chain zero is valid as well */
+#define FDB_NUM_CHAINS (FDB_FT_CHAIN + 1)
+
+#define FDB_TC_MAX_PRIO 16
+#define FDB_TC_LEVELS_PER_PRIO 2
+
 #ifdef CONFIG_MLX5_ESWITCH
 
 #define MLX5_MAX_UC_PER_VPORT(dev) \
@@ -59,21 +69,22 @@
 #define mlx5_esw_has_fwd_fdb(dev) \
 	MLX5_CAP_ESW_FLOWTABLE(dev, fdb_multi_path_to_table)
 
-#define FDB_MAX_CHAIN 3
-#define FDB_SLOW_PATH_CHAIN (FDB_MAX_CHAIN + 1)
-#define FDB_MAX_PRIO 16
-
 struct vport_ingress {
 	struct mlx5_flow_table *acl;
-	struct mlx5_flow_group *allow_untagged_spoofchk_grp;
+	struct mlx5_flow_handle *allow_rule;
+	struct {
 		struct mlx5_flow_group *allow_spoofchk_only_grp;
+		struct mlx5_flow_group *allow_untagged_spoofchk_grp;
 		struct mlx5_flow_group *allow_untagged_only_grp;
 		struct mlx5_flow_group *drop_grp;
-	struct mlx5_modify_hdr   *modify_metadata;
-	struct mlx5_flow_handle  *modify_metadata_rule;
-	struct mlx5_flow_handle  *allow_rule;
 		struct mlx5_flow_handle *drop_rule;
 		struct mlx5_fc *drop_counter;
+	} legacy;
+	struct {
+		struct mlx5_flow_group *metadata_grp;
+		struct mlx5_modify_hdr *modify_metadata;
+		struct mlx5_flow_handle *modify_metadata_rule;
+	} offloads;
 };
 
 struct vport_egress {
@@ -81,8 +92,10 @@ struct vport_egress {
 	struct mlx5_flow_group *allowed_vlans_grp;
 	struct mlx5_flow_group *drop_grp;
 	struct mlx5_flow_handle  *allowed_vlan;
+	struct {
 		struct mlx5_flow_handle *drop_rule;
 		struct mlx5_fc *drop_counter;
+	} legacy;
 };
 
 struct mlx5_vport_drop_stats {
@@ -139,7 +152,6 @@ enum offloads_fdb_flags {
 
 extern const unsigned int ESW_POOLS[4];
 
-#define PRIO_LEVELS 2
 struct mlx5_eswitch_fdb {
 	union {
 		struct legacy_fdb {
@@ -166,7 +178,7 @@ struct mlx5_eswitch_fdb {
 			struct {
 				struct mlx5_flow_table *fdb;
 				u32 num_rules;
-			} fdb_prio[FDB_MAX_CHAIN + 1][FDB_MAX_PRIO + 1][PRIO_LEVELS];
+			} fdb_prio[FDB_NUM_CHAINS][FDB_TC_MAX_PRIO + 1][FDB_TC_LEVELS_PER_PRIO];
 			/* Protects fdb_prio table */
 			struct mutex fdb_prio_lock;
 
@@ -217,8 +229,8 @@ enum {
 struct mlx5_eswitch {
 	struct mlx5_core_dev    *dev;
 	struct mlx5_nb          nb;
-	/* legacy data structures */
 	struct mlx5_eswitch_fdb fdb_table;
+	/* legacy data structures */
 	struct hlist_head       mc_table[MLX5_L2_ADDR_HASH_SIZE];
 	struct esw_mc_addr mc_promisc;
 	/* end of legacy */
@@ -251,18 +263,16 @@ void esw_offloads_cleanup_reps(struct mlx5_eswitch *esw);
 int esw_offloads_init_reps(struct mlx5_eswitch *esw);
 void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw,
 				     struct mlx5_vport *vport);
-int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw,
-				 struct mlx5_vport *vport);
+int esw_vport_create_ingress_acl_table(struct mlx5_eswitch *esw,
+				       struct mlx5_vport *vport,
+				       int table_size);
+void esw_vport_destroy_ingress_acl_table(struct mlx5_vport *vport);
 void esw_vport_cleanup_egress_rules(struct mlx5_eswitch *esw,
 				    struct mlx5_vport *vport);
 int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
 				struct mlx5_vport *vport);
 void esw_vport_disable_egress_acl(struct mlx5_eswitch *esw,
 				  struct mlx5_vport *vport);
-void esw_vport_disable_ingress_acl(struct mlx5_eswitch *esw,
-				   struct mlx5_vport *vport);
-void esw_vport_del_ingress_acl_modify_metadata(struct mlx5_eswitch *esw,
-					       struct mlx5_vport *vport);
 int mlx5_esw_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num,
 			       u32 rate_mbps);
 
@@ -292,9 +302,11 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
 				 struct ifla_vf_stats *vf_stats);
 void mlx5_eswitch_del_send_to_vport_rule(struct mlx5_flow_handle *rule);
 
-int mlx5_eswitch_modify_esw_vport_context(struct mlx5_eswitch *esw, u16 vport,
+int mlx5_eswitch_modify_esw_vport_context(struct mlx5_core_dev *dev, u16 vport,
+					  bool other_vport,
 					  void *in, int inlen);
-int mlx5_eswitch_query_esw_vport_context(struct mlx5_eswitch *esw, u16 vport,
+int mlx5_eswitch_query_esw_vport_context(struct mlx5_core_dev *dev, u16 vport,
+					 bool other_vport,
 					 void *out, int outlen);
 
 struct mlx5_flow_spec;
@@ -421,6 +433,10 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
 int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
 				  u16 vport, u16 vlan, u8 qos, u8 set_flags);
 
+int mlx5_esw_create_vport_egress_acl_vlan(struct mlx5_eswitch *esw,
+					  struct mlx5_vport *vport,
+					  u16 vlan_id, u32 flow_action);
+
 static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev,
 						       u8 vlan_depth)
 {
@@ -459,6 +475,12 @@ static inline u16 mlx5_eswitch_manager_vport(struct mlx5_core_dev *dev)
 		MLX5_VPORT_ECPF : MLX5_VPORT_PF;
 }
 
+static inline bool
+mlx5_esw_is_manager_vport(const struct mlx5_eswitch *esw, u16 vport_num)
+{
+	return esw->manager_vport == vport_num;
+}
+
 static inline u16 mlx5_eswitch_first_host_vport_num(struct mlx5_core_dev *dev)
 {
 	return mlx5_core_is_ecpf_esw_manager(dev) ?
@@ -593,11 +615,18 @@ bool mlx5_eswitch_is_vf_vport(const struct mlx5_eswitch *esw, u16 vport_num);
 void mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, const int num_vfs);
 int mlx5_esw_funcs_changed_handler(struct notifier_block *nb, unsigned long type, void *data);
 
-void
+int
 mlx5_eswitch_enable_pf_vf_vports(struct mlx5_eswitch *esw,
 				 enum mlx5_eswitch_vport_event enabled_events);
 void mlx5_eswitch_disable_pf_vf_vports(struct mlx5_eswitch *esw);
 
+int
+esw_vport_create_offloads_acl_tables(struct mlx5_eswitch *esw,
+				     struct mlx5_vport *vport);
+void
+esw_vport_destroy_offloads_acl_tables(struct mlx5_eswitch *esw,
+				      struct mlx5_vport *vport);
+
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
@@ -613,10 +642,6 @@ static inline const u32 *mlx5_esw_query_functions(struct mlx5_core_dev *dev)
 
 static inline void mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, const int num_vfs) {}
 
-#define FDB_MAX_CHAIN 1
-#define FDB_SLOW_PATH_CHAIN (FDB_MAX_CHAIN + 1)
-#define FDB_MAX_PRIO 1
-
 #endif /* CONFIG_MLX5_ESWITCH */
 
 #endif /* __MLX5_ESWITCH_H__ */

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

@@ -2400,9 +2400,17 @@ static void set_prio_attrs_in_prio(struct fs_prio *prio, int acc_level)
 	int acc_level_ns = acc_level;
 
 	prio->start_level = acc_level;
-	fs_for_each_ns(ns, prio)
+	fs_for_each_ns(ns, prio) {
 		/* This updates start_level and num_levels of ns's priority descendants */
 		acc_level_ns = set_prio_attrs_in_ns(ns, acc_level);
+
+		/* If this a prio with chains, and we can jump from one chain
+		 * (namepsace) to another, so we accumulate the levels
+		 */
+		if (prio->node.type == FS_TYPE_PRIO_CHAINS)
+			acc_level = acc_level_ns;
+	}
+
 	if (!prio->num_levels)
 		prio->num_levels = acc_level_ns - prio->start_level;
 	WARN_ON(prio->num_levels < acc_level_ns - prio->start_level);
@@ -2591,58 +2599,109 @@ static int init_rdma_rx_root_ns(struct mlx5_flow_steering *steering)
 	steering->rdma_rx_root_ns = NULL;
 	return err;
 }
-static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
+
+/* FT and tc chains are stored in the same array so we can re-use the
+ * mlx5_get_fdb_sub_ns() and tc api for FT chains.
+ * When creating a new ns for each chain store it in the first available slot.
+ * Assume tc chains are created and stored first and only then the FT chain.
+ */
+static void store_fdb_sub_ns_prio_chain(struct mlx5_flow_steering *steering,
+					struct mlx5_flow_namespace *ns)
+{
+	int chain = 0;
+
+	while (steering->fdb_sub_ns[chain])
+		++chain;
+
+	steering->fdb_sub_ns[chain] = ns;
+}
+
+static int create_fdb_sub_ns_prio_chain(struct mlx5_flow_steering *steering,
+					struct fs_prio *maj_prio)
 {
 	struct mlx5_flow_namespace *ns;
-	struct fs_prio *maj_prio;
 	struct fs_prio *min_prio;
+	int prio;
+
+	ns = fs_create_namespace(maj_prio, MLX5_FLOW_TABLE_MISS_ACTION_DEF);
+	if (IS_ERR(ns))
+		return PTR_ERR(ns);
+
+	for (prio = 0; prio < FDB_TC_MAX_PRIO; prio++) {
+		min_prio = fs_create_prio(ns, prio, FDB_TC_LEVELS_PER_PRIO);
+		if (IS_ERR(min_prio))
+			return PTR_ERR(min_prio);
+	}
+
+	store_fdb_sub_ns_prio_chain(steering, ns);
+
+	return 0;
+}
+
+static int create_fdb_chains(struct mlx5_flow_steering *steering,
+			     int fs_prio,
+			     int chains)
+{
+	struct fs_prio *maj_prio;
 	int levels;
 	int chain;
-	int prio;
 	int err;
 
-	steering->fdb_root_ns = create_root_ns(steering, FS_FT_FDB);
-	if (!steering->fdb_root_ns)
-		return -ENOMEM;
+	levels = FDB_TC_LEVELS_PER_PRIO * FDB_TC_MAX_PRIO * chains;
+	maj_prio = fs_create_prio_chained(&steering->fdb_root_ns->ns,
+					  fs_prio,
+					  levels);
+	if (IS_ERR(maj_prio))
+		return PTR_ERR(maj_prio);
+
+	for (chain = 0; chain < chains; chain++) {
+		err = create_fdb_sub_ns_prio_chain(steering, maj_prio);
+		if (err)
+			return err;
+	}
 
-	steering->fdb_sub_ns = kzalloc(sizeof(steering->fdb_sub_ns) *
-				       (FDB_MAX_CHAIN + 1), GFP_KERNEL);
+	return 0;
+}
+
+static int create_fdb_fast_path(struct mlx5_flow_steering *steering)
+{
+	int err;
+
+	steering->fdb_sub_ns = kcalloc(FDB_NUM_CHAINS,
+				       sizeof(*steering->fdb_sub_ns),
+				       GFP_KERNEL);
 	if (!steering->fdb_sub_ns)
 		return -ENOMEM;
 
+	err = create_fdb_chains(steering, FDB_TC_OFFLOAD, FDB_TC_MAX_CHAIN + 1);
+	if (err)
+		return err;
+
+	err = create_fdb_chains(steering, FDB_FT_OFFLOAD, 1);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
+{
+	struct fs_prio *maj_prio;
+	int err;
+
+	steering->fdb_root_ns = create_root_ns(steering, FS_FT_FDB);
+	if (!steering->fdb_root_ns)
+		return -ENOMEM;
+
 	maj_prio = fs_create_prio(&steering->fdb_root_ns->ns, FDB_BYPASS_PATH,
 				  1);
 	if (IS_ERR(maj_prio)) {
 		err = PTR_ERR(maj_prio);
 		goto out_err;
 	}
-
-	levels = 2 * FDB_MAX_PRIO * (FDB_MAX_CHAIN + 1);
-	maj_prio = fs_create_prio_chained(&steering->fdb_root_ns->ns,
-					  FDB_FAST_PATH,
-					  levels);
-	if (IS_ERR(maj_prio)) {
-		err = PTR_ERR(maj_prio);
-		goto out_err;
-	}
-
-	for (chain = 0; chain <= FDB_MAX_CHAIN; chain++) {
-		ns = fs_create_namespace(maj_prio, MLX5_FLOW_TABLE_MISS_ACTION_DEF);
-		if (IS_ERR(ns)) {
-			err = PTR_ERR(ns);
-			goto out_err;
-		}
-
-		for (prio = 0; prio < FDB_MAX_PRIO * (chain + 1); prio++) {
-			min_prio = fs_create_prio(ns, prio, 2);
-			if (IS_ERR(min_prio)) {
-				err = PTR_ERR(min_prio);
+	err = create_fdb_fast_path(steering);
+	if (err)
 		goto out_err;
-			}
-		}
-
-		steering->fdb_sub_ns[chain] = ns;
-	}
 
 	maj_prio = fs_create_prio(&steering->fdb_root_ns->ns, FDB_SLOW_PATH, 1);
 	if (IS_ERR(maj_prio)) {

include/linux/mlx5/driver.h

@@ -1121,6 +1121,11 @@ static inline bool mlx5_core_is_pf(const struct mlx5_core_dev *dev)
 	return dev->coredev_type == MLX5_COREDEV_PF;
 }
 
+static inline bool mlx5_core_is_vf(const struct mlx5_core_dev *dev)
+{
+	return dev->coredev_type == MLX5_COREDEV_VF;
+}
+
 static inline bool mlx5_core_is_ecpf(struct mlx5_core_dev *dev)
 {
 	return dev->caps.embedded_cpu;
@@ -1186,4 +1191,15 @@ enum {
 	MLX5_TRIGGERED_CMD_COMP = (u64)1 << 32,
 };
 
+static inline bool mlx5_is_roce_enabled(struct mlx5_core_dev *dev)
+{
+	struct devlink *devlink = priv_to_devlink(dev);
+	union devlink_param_value val;
+
+	devlink_param_driverinit_value_get(devlink,
+					   DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
+					   &val);
+	return val.vbool;
+}
+
 #endif /* MLX5_DRIVER_H */

include/linux/mlx5/fs.h

@@ -80,7 +80,8 @@ enum mlx5_flow_namespace_type {
 
 enum {
 	FDB_BYPASS_PATH,
-	FDB_FAST_PATH,
+	FDB_TC_OFFLOAD,
+	FDB_FT_OFFLOAD,
 	FDB_SLOW_PATH,
 };
 

include/linux/mlx5/mlx5_ifc.h

@@ -1153,7 +1153,7 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         log_max_srq[0x5];
 	u8         reserved_at_b0[0x10];
 
-	u8         reserved_at_c0[0x8];
+	u8         max_sgl_for_optimized_performance[0x8];
 	u8         log_max_cq_sz[0x8];
 	u8         reserved_at_d0[0xb];
 	u8         log_max_cq[0x5];

include/net/devlink.h

@@ -402,6 +402,7 @@ enum devlink_param_generic_id {
 	DEVLINK_PARAM_GENERIC_ID_MSIX_VEC_PER_PF_MIN,
 	DEVLINK_PARAM_GENERIC_ID_FW_LOAD_POLICY,
 	DEVLINK_PARAM_GENERIC_ID_RESET_DEV_ON_DRV_PROBE,
+	DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
 
 	/* add new param generic ids above here*/
 	__DEVLINK_PARAM_GENERIC_ID_MAX,
@@ -436,6 +437,9 @@ enum devlink_param_generic_id {
 	"reset_dev_on_drv_probe"
 #define DEVLINK_PARAM_GENERIC_RESET_DEV_ON_DRV_PROBE_TYPE DEVLINK_PARAM_TYPE_U8
 
+#define DEVLINK_PARAM_GENERIC_ENABLE_ROCE_NAME "enable_roce"
+#define DEVLINK_PARAM_GENERIC_ENABLE_ROCE_TYPE DEVLINK_PARAM_TYPE_BOOL
+
 #define DEVLINK_PARAM_GENERIC(_id, _cmodes, _get, _set, _validate)	\
 {									\
 	.id = DEVLINK_PARAM_GENERIC_ID_##_id,				\

net/core/devlink.c

@@ -3006,6 +3006,11 @@ static const struct devlink_param devlink_param_generic[] = {
 		.name = DEVLINK_PARAM_GENERIC_RESET_DEV_ON_DRV_PROBE_NAME,
 		.type = DEVLINK_PARAM_GENERIC_RESET_DEV_ON_DRV_PROBE_TYPE,
 	},
+	{
+		.id = DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
+		.name = DEVLINK_PARAM_GENERIC_ENABLE_ROCE_NAME,
+		.type = DEVLINK_PARAM_GENERIC_ENABLE_ROCE_TYPE,
+	},
 };
 
 static int devlink_param_generic_verify(const struct devlink_param *param)