Merging upstream version 6.7.7.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

86 files changed, 3080 insertions, 1853 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 332472fe49..a09b6e0533 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -400,7 +400,7 @@ void mlx4_en_recover_from_oom(struct mlx4_en_priv *priv)
 	for (ring = 0; ring < priv->rx_ring_num; ring++) {
 		if (mlx4_en_is_ring_empty(priv->rx_ring[ring])) {
 			local_bh_disable();
-			napi_reschedule(&priv->rx_cq[ring]->napi);
+			napi_schedule(&priv->rx_cq[ring]->napi);
 			local_bh_enable();
 		}
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index fe48d20d61..0005d9e2c2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -1967,7 +1967,7 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 	if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_DRIVER_VERSION_TO_FW) {
 		u8 *dst = (u8 *)(inbox + INIT_HCA_DRIVER_VERSION_OFFSET / 4);
 
-		strncpy(dst, DRV_NAME_FOR_FW, INIT_HCA_DRIVER_VERSION_SZ - 1);
+		strscpy(dst, DRV_NAME_FOR_FW, INIT_HCA_DRIVER_VERSION_SZ);
 		mlx4_dbg(dev, "Reporting Driver Version to FW: %s\n", dst);
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
index c4f4de82e2..685335832a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
@@ -189,3 +189,11 @@ config MLX5_SF_MANAGER
 	port is managed through devlink.  A subfunction supports RDMA, netdevice
 	and vdpa device. It is similar to a SRIOV VF but it doesn't require
 	SRIOV support.
+
+config MLX5_DPLL
+	tristate "Mellanox 5th generation network adapters (ConnectX series) DPLL support"
+	depends on NETDEVICES && ETHERNET && PCI && MLX5_CORE
+	select DPLL
+	help
+	  DPLL support in Mellanox Technologies ConnectX NICs.
+
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 7e94caca48..c44870b175 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -128,3 +128,6 @@ mlx5_core-$(CONFIG_MLX5_SF) += sf/vhca_event.o sf/dev/dev.o sf/dev/driver.o irq_
 # SF manager
 #
 mlx5_core-$(CONFIG_MLX5_SF_MANAGER) += sf/cmd.o sf/hw_table.o sf/devlink.o
+
+obj-$(CONFIG_MLX5_DPLL) += mlx5_dpll.o
+mlx5_dpll-y :=	dpll.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 55efb932ab..4957412ff1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -528,6 +528,7 @@ static int mlx5_internal_err_ret_value(struct mlx5_core_dev *dev, u16 op,
 	case MLX5_CMD_OP_SAVE_VHCA_STATE:
 	case MLX5_CMD_OP_LOAD_VHCA_STATE:
 	case MLX5_CMD_OP_SYNC_CRYPTO:
+	case MLX5_CMD_OP_ALLOW_OTHER_VHCA_ACCESS:
 		*status = MLX5_DRIVER_STATUS_ABORTED;
 		*synd = MLX5_DRIVER_SYND;
 		return -ENOLINK;
@@ -731,6 +732,7 @@ const char *mlx5_command_str(int command)
 	MLX5_COMMAND_STR_CASE(SAVE_VHCA_STATE);
 	MLX5_COMMAND_STR_CASE(LOAD_VHCA_STATE);
 	MLX5_COMMAND_STR_CASE(SYNC_CRYPTO);
+	MLX5_COMMAND_STR_CASE(ALLOW_OTHER_VHCA_ACCESS);
 	default: return "unknown command opcode";
 	}
 }
@@ -2093,6 +2095,74 @@ int mlx5_cmd_exec_cb(struct mlx5_async_ctx *ctx, void *in, int in_size,
 }
 EXPORT_SYMBOL(mlx5_cmd_exec_cb);
 
+int mlx5_cmd_allow_other_vhca_access(struct mlx5_core_dev *dev,
+				     struct mlx5_cmd_allow_other_vhca_access_attr *attr)
+{
+	u32 out[MLX5_ST_SZ_DW(allow_other_vhca_access_out)] = {};
+	u32 in[MLX5_ST_SZ_DW(allow_other_vhca_access_in)] = {};
+	void *key;
+
+	MLX5_SET(allow_other_vhca_access_in,
+		 in, opcode, MLX5_CMD_OP_ALLOW_OTHER_VHCA_ACCESS);
+	MLX5_SET(allow_other_vhca_access_in,
+		 in, object_type_to_be_accessed, attr->obj_type);
+	MLX5_SET(allow_other_vhca_access_in,
+		 in, object_id_to_be_accessed, attr->obj_id);
+
+	key = MLX5_ADDR_OF(allow_other_vhca_access_in, in, access_key);
+	memcpy(key, attr->access_key, sizeof(attr->access_key));
+
+	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+int mlx5_cmd_alias_obj_create(struct mlx5_core_dev *dev,
+			      struct mlx5_cmd_alias_obj_create_attr *alias_attr,
+			      u32 *obj_id)
+{
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {};
+	u32 in[MLX5_ST_SZ_DW(create_alias_obj_in)] = {};
+	void *param;
+	void *attr;
+	void *key;
+	int ret;
+
+	attr = MLX5_ADDR_OF(create_alias_obj_in, in, hdr);
+	MLX5_SET(general_obj_in_cmd_hdr,
+		 attr, opcode, MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr,
+		 attr, obj_type, alias_attr->obj_type);
+	param = MLX5_ADDR_OF(general_obj_in_cmd_hdr, in, op_param);
+	MLX5_SET(general_obj_create_param, param, alias_object, 1);
+
+	attr = MLX5_ADDR_OF(create_alias_obj_in, in, alias_ctx);
+	MLX5_SET(alias_context, attr, vhca_id_to_be_accessed, alias_attr->vhca_id);
+	MLX5_SET(alias_context, attr, object_id_to_be_accessed, alias_attr->obj_id);
+
+	key = MLX5_ADDR_OF(alias_context, attr, access_key);
+	memcpy(key, alias_attr->access_key, sizeof(alias_attr->access_key));
+
+	ret = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+	if (ret)
+		return ret;
+
+	*obj_id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+
+	return 0;
+}
+
+int mlx5_cmd_alias_obj_destroy(struct mlx5_core_dev *dev, u32 obj_id,
+			       u16 obj_type)
+{
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {};
+	u32 in[MLX5_ST_SZ_DW(general_obj_in_cmd_hdr)] = {};
+
+	MLX5_SET(general_obj_in_cmd_hdr, in, opcode, MLX5_CMD_OP_DESTROY_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, in, obj_type, obj_type);
+	MLX5_SET(general_obj_in_cmd_hdr, in, obj_id, obj_id);
+
+	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
 static void destroy_msg_cache(struct mlx5_core_dev *dev)
 {
 	struct cmd_msg_cache *ch;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
index 7909f378dc..cf0477f53d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
@@ -38,8 +38,6 @@
 #include "devlink.h"
 #include "lag/lag.h"
 
-/* intf dev list mutex */
-static DEFINE_MUTEX(mlx5_intf_mutex);
 static DEFINE_IDA(mlx5_adev_ida);
 
 static bool is_eth_rep_supported(struct mlx5_core_dev *dev)
@@ -206,6 +204,19 @@ static bool is_ib_enabled(struct mlx5_core_dev *dev)
 	return err ? false : val.vbool;
 }
 
+static bool is_dpll_supported(struct mlx5_core_dev *dev)
+{
+	if (!IS_ENABLED(CONFIG_MLX5_DPLL))
+		return false;
+
+	if (!MLX5_CAP_MCAM_REG2(dev, synce_registers)) {
+		mlx5_core_warn(dev, "Missing SyncE capability\n");
+		return false;
+	}
+
+	return true;
+}
+
 enum {
 	MLX5_INTERFACE_PROTOCOL_ETH,
 	MLX5_INTERFACE_PROTOCOL_ETH_REP,
@@ -215,6 +226,8 @@ enum {
 	MLX5_INTERFACE_PROTOCOL_MPIB,
 
 	MLX5_INTERFACE_PROTOCOL_VNET,
+
+	MLX5_INTERFACE_PROTOCOL_DPLL,
 };
 
 static const struct mlx5_adev_device {
@@ -237,6 +250,8 @@ static const struct mlx5_adev_device {
 					   .is_supported = &is_ib_rep_supported },
 	[MLX5_INTERFACE_PROTOCOL_MPIB] = { .suffix = "multiport",
 					   .is_supported = &is_mp_supported },
+	[MLX5_INTERFACE_PROTOCOL_DPLL] = { .suffix = "dpll",
+					   .is_supported = &is_dpll_supported },
 };
 
 int mlx5_adev_idx_alloc(void)
@@ -320,9 +335,9 @@ static void del_adev(struct auxiliary_device *adev)
 
 void mlx5_dev_set_lightweight(struct mlx5_core_dev *dev)
 {
-	mutex_lock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	dev->priv.flags |= MLX5_PRIV_FLAGS_DISABLE_ALL_ADEV;
-	mutex_unlock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 }
 
 bool mlx5_dev_is_lightweight(struct mlx5_core_dev *dev)
@@ -338,7 +353,7 @@ int mlx5_attach_device(struct mlx5_core_dev *dev)
 	int ret = 0, i;
 
 	devl_assert_locked(priv_to_devlink(dev));
-	mutex_lock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	priv->flags &= ~MLX5_PRIV_FLAGS_DETACH;
 	for (i = 0; i < ARRAY_SIZE(mlx5_adev_devices); i++) {
 		if (!priv->adev[i]) {
@@ -383,7 +398,7 @@ int mlx5_attach_device(struct mlx5_core_dev *dev)
 			break;
 		}
 	}
-	mutex_unlock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 	return ret;
 }
 
@@ -396,7 +411,7 @@ void mlx5_detach_device(struct mlx5_core_dev *dev, bool suspend)
 	int i;
 
 	devl_assert_locked(priv_to_devlink(dev));
-	mutex_lock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	for (i = ARRAY_SIZE(mlx5_adev_devices) - 1; i >= 0; i--) {
 		if (!priv->adev[i])
 			continue;
@@ -426,7 +441,7 @@ skip_suspend:
 		priv->adev[i] = NULL;
 	}
 	priv->flags |= MLX5_PRIV_FLAGS_DETACH;
-	mutex_unlock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 }
 
 int mlx5_register_device(struct mlx5_core_dev *dev)
@@ -434,10 +449,10 @@ int mlx5_register_device(struct mlx5_core_dev *dev)
 	int ret;
 
 	devl_assert_locked(priv_to_devlink(dev));
-	mutex_lock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	dev->priv.flags &= ~MLX5_PRIV_FLAGS_DISABLE_ALL_ADEV;
 	ret = mlx5_rescan_drivers_locked(dev);
-	mutex_unlock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 	if (ret)
 		mlx5_unregister_device(dev);
 
@@ -447,10 +462,10 @@ int mlx5_register_device(struct mlx5_core_dev *dev)
 void mlx5_unregister_device(struct mlx5_core_dev *dev)
 {
 	devl_assert_locked(priv_to_devlink(dev));
-	mutex_lock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	dev->priv.flags = MLX5_PRIV_FLAGS_DISABLE_ALL_ADEV;
 	mlx5_rescan_drivers_locked(dev);
-	mutex_unlock(&mlx5_intf_mutex);
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 }
 
 static int add_drivers(struct mlx5_core_dev *dev)
@@ -528,7 +543,6 @@ int mlx5_rescan_drivers_locked(struct mlx5_core_dev *dev)
 {
 	struct mlx5_priv *priv = &dev->priv;
 
-	lockdep_assert_held(&mlx5_intf_mutex);
 	if (priv->flags & MLX5_PRIV_FLAGS_DETACH)
 		return 0;
 
@@ -548,85 +562,3 @@ bool mlx5_same_hw_devs(struct mlx5_core_dev *dev, struct mlx5_core_dev *peer_dev
 
 	return (fsystem_guid && psystem_guid && fsystem_guid == psystem_guid);
 }
-
-static u32 mlx5_gen_pci_id(const struct mlx5_core_dev *dev)
-{
-	return (u32)((pci_domain_nr(dev->pdev->bus) << 16) |
-		     (dev->pdev->bus->number << 8) |
-		     PCI_SLOT(dev->pdev->devfn));
-}
-
-static int _next_phys_dev(struct mlx5_core_dev *mdev,
-			  const struct mlx5_core_dev *curr)
-{
-	if (!mlx5_core_is_pf(mdev))
-		return 0;
-
-	if (mdev == curr)
-		return 0;
-
-	if (!mlx5_same_hw_devs(mdev, (struct mlx5_core_dev *)curr) &&
-	    mlx5_gen_pci_id(mdev) != mlx5_gen_pci_id(curr))
-		return 0;
-
-	return 1;
-}
-
-static void *pci_get_other_drvdata(struct device *this, struct device *other)
-{
-	if (this->driver != other->driver)
-		return NULL;
-
-	return pci_get_drvdata(to_pci_dev(other));
-}
-
-static int next_phys_dev_lag(struct device *dev, const void *data)
-{
-	struct mlx5_core_dev *mdev, *this = (struct mlx5_core_dev *)data;
-
-	mdev = pci_get_other_drvdata(this->device, dev);
-	if (!mdev)
-		return 0;
-
-	if (!mlx5_lag_is_supported(mdev))
-		return 0;
-
-	return _next_phys_dev(mdev, data);
-}
-
-static struct mlx5_core_dev *mlx5_get_next_dev(struct mlx5_core_dev *dev,
-					       int (*match)(struct device *dev, const void *data))
-{
-	struct device *next;
-
-	if (!mlx5_core_is_pf(dev))
-		return NULL;
-
-	next = bus_find_device(&pci_bus_type, NULL, dev, match);
-	if (!next)
-		return NULL;
-
-	put_device(next);
-	return pci_get_drvdata(to_pci_dev(next));
-}
-
-/* Must be called with intf_mutex held */
-struct mlx5_core_dev *mlx5_get_next_phys_dev_lag(struct mlx5_core_dev *dev)
-{
-	lockdep_assert_held(&mlx5_intf_mutex);
-	return mlx5_get_next_dev(dev, &next_phys_dev_lag);
-}
-
-void mlx5_dev_list_lock(void)
-{
-	mutex_lock(&mlx5_intf_mutex);
-}
-void mlx5_dev_list_unlock(void)
-{
-	mutex_unlock(&mlx5_intf_mutex);
-}
-
-int mlx5_dev_list_trylock(void)
-{
-	return mutex_trylock(&mlx5_intf_mutex);
-}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index af8460bb25..3e064234f6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -138,7 +138,6 @@ static int mlx5_devlink_reload_down(struct devlink *devlink, bool netns_change,
 {
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
 	struct pci_dev *pdev = dev->pdev;
-	bool sf_dev_allocated;
 	int ret = 0;
 
 	if (mlx5_dev_is_lightweight(dev)) {
@@ -148,16 +147,6 @@ static int mlx5_devlink_reload_down(struct devlink *devlink, bool netns_change,
 		return 0;
 	}
 
-	sf_dev_allocated = mlx5_sf_dev_allocated(dev);
-	if (sf_dev_allocated) {
-		/* Reload results in deleting SF device which further results in
-		 * unregistering devlink instance while holding devlink_mutext.
-		 * Hence, do not support reload.
-		 */
-		NL_SET_ERR_MSG_MOD(extack, "reload is unsupported when SFs are allocated");
-		return -EOPNOTSUPP;
-	}
-
 	if (mlx5_lag_is_active(dev)) {
 		NL_SET_ERR_MSG_MOD(extack, "reload is unsupported in Lag mode");
 		return -EOPNOTSUPP;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 85d3bfa078..080e7eab52 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -889,36 +889,16 @@ int mlx5_fw_tracer_trigger_core_dump_general(struct mlx5_core_dev *dev)
 	return 0;
 }
 
-static int
+static void
 mlx5_devlink_fmsg_fill_trace(struct devlink_fmsg *fmsg,
 			     struct mlx5_fw_trace_data *trace_data)
 {
-	int err;
-
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u64_pair_put(fmsg, "timestamp", trace_data->timestamp);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_bool_pair_put(fmsg, "lost", trace_data->lost);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "event_id", trace_data->event_id);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_string_pair_put(fmsg, "msg", trace_data->msg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_obj_nest_end(fmsg);
-	if (err)
-		return err;
-	return 0;
+	devlink_fmsg_obj_nest_start(fmsg);
+	devlink_fmsg_u64_pair_put(fmsg, "timestamp", trace_data->timestamp);
+	devlink_fmsg_bool_pair_put(fmsg, "lost", trace_data->lost);
+	devlink_fmsg_u8_pair_put(fmsg, "event_id", trace_data->event_id);
+	devlink_fmsg_string_pair_put(fmsg, "msg", trace_data->msg);
+	devlink_fmsg_obj_nest_end(fmsg);
 }
 
 int mlx5_fw_tracer_get_saved_traces_objects(struct mlx5_fw_tracer *tracer,
@@ -927,7 +907,6 @@ int mlx5_fw_tracer_get_saved_traces_objects(struct mlx5_fw_tracer *tracer,
 	struct mlx5_fw_trace_data *straces = tracer->st_arr.straces;
 	u32 index, start_index, end_index;
 	u32 saved_traces_index;
-	int err;
 
 	if (!straces[0].timestamp)
 		return -ENOMSG;
@@ -940,22 +919,18 @@ int mlx5_fw_tracer_get_saved_traces_objects(struct mlx5_fw_tracer *tracer,
 		start_index = 0;
 	end_index = (saved_traces_index - 1) & (SAVED_TRACES_NUM - 1);
 
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "dump fw traces");
-	if (err)
-		goto unlock;
+	devlink_fmsg_arr_pair_nest_start(fmsg, "dump fw traces");
 	index = start_index;
 	while (index != end_index) {
-		err = mlx5_devlink_fmsg_fill_trace(fmsg, &straces[index]);
-		if (err)
-			goto unlock;
+		mlx5_devlink_fmsg_fill_trace(fmsg, &straces[index]);
 
 		index = (index + 1) & (SAVED_TRACES_NUM - 1);
 	}
 
-	err = devlink_fmsg_arr_pair_nest_end(fmsg);
-unlock:
+	devlink_fmsg_arr_pair_nest_end(fmsg);
 	mutex_unlock(&tracer->st_arr.lock);
-	return err;
+
+	return 0;
 }
 
 static void mlx5_fw_tracer_update_db(struct work_struct *work)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c
index e869c65d8e..c7216e84ef 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c
@@ -13,106 +13,55 @@ struct mlx5_vnic_diag_stats {
 	__be64 query_vnic_env_out[MLX5_ST_SZ_QW(query_vnic_env_out)];
 };
 
-int mlx5_reporter_vnic_diagnose_counters(struct mlx5_core_dev *dev,
-					 struct devlink_fmsg *fmsg,
-					 u16 vport_num, bool other_vport)
+void mlx5_reporter_vnic_diagnose_counters(struct mlx5_core_dev *dev,
+					  struct devlink_fmsg *fmsg,
+					  u16 vport_num, bool other_vport)
 {
 	u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {};
 	struct mlx5_vnic_diag_stats vnic;
-	int err;
 
 	MLX5_SET(query_vnic_env_in, in, opcode, MLX5_CMD_OP_QUERY_VNIC_ENV);
 	MLX5_SET(query_vnic_env_in, in, vport_number, vport_num);
 	MLX5_SET(query_vnic_env_in, in, other_vport, !!other_vport);
 
-	err = mlx5_cmd_exec_inout(dev, query_vnic_env, in, &vnic.query_vnic_env_out);
-	if (err)
-		return err;
+	mlx5_cmd_exec_inout(dev, query_vnic_env, in, &vnic.query_vnic_env_out);
 
-	err = devlink_fmsg_pair_nest_start(fmsg, "vNIC env counters");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
+	devlink_fmsg_pair_nest_start(fmsg, "vNIC env counters");
+	devlink_fmsg_obj_nest_start(fmsg);
 
 	if (MLX5_CAP_GEN(dev, vnic_env_queue_counters)) {
-		err = devlink_fmsg_u32_pair_put(fmsg, "total_error_queues",
-						VNIC_ENV_GET(&vnic, total_error_queues));
-		if (err)
-			return err;
-
-		err = devlink_fmsg_u32_pair_put(fmsg, "send_queue_priority_update_flow",
-						VNIC_ENV_GET(&vnic,
-							     send_queue_priority_update_flow));
-		if (err)
-			return err;
+		devlink_fmsg_u32_pair_put(fmsg, "total_error_queues",
+					  VNIC_ENV_GET(&vnic, total_error_queues));
+		devlink_fmsg_u32_pair_put(fmsg, "send_queue_priority_update_flow",
+					  VNIC_ENV_GET(&vnic, send_queue_priority_update_flow));
 	}
-
 	if (MLX5_CAP_GEN(dev, eq_overrun_count)) {
-		err = devlink_fmsg_u32_pair_put(fmsg, "comp_eq_overrun",
-						VNIC_ENV_GET(&vnic, comp_eq_overrun));
-		if (err)
-			return err;
-
-		err = devlink_fmsg_u32_pair_put(fmsg, "async_eq_overrun",
-						VNIC_ENV_GET(&vnic, async_eq_overrun));
-		if (err)
-			return err;
-	}
-
-	if (MLX5_CAP_GEN(dev, vnic_env_cq_overrun)) {
-		err = devlink_fmsg_u32_pair_put(fmsg, "cq_overrun",
-						VNIC_ENV_GET(&vnic, cq_overrun));
-		if (err)
-			return err;
-	}
-
-	if (MLX5_CAP_GEN(dev, invalid_command_count)) {
-		err = devlink_fmsg_u32_pair_put(fmsg, "invalid_command",
-						VNIC_ENV_GET(&vnic, invalid_command));
-		if (err)
-			return err;
-	}
-
-	if (MLX5_CAP_GEN(dev, quota_exceeded_count)) {
-		err = devlink_fmsg_u32_pair_put(fmsg, "quota_exceeded_command",
-						VNIC_ENV_GET(&vnic, quota_exceeded_command));
-		if (err)
-			return err;
+		devlink_fmsg_u32_pair_put(fmsg, "comp_eq_overrun",
+					  VNIC_ENV_GET(&vnic, comp_eq_overrun));
+		devlink_fmsg_u32_pair_put(fmsg, "async_eq_overrun",
+					  VNIC_ENV_GET(&vnic, async_eq_overrun));
 	}
-
-	if (MLX5_CAP_GEN(dev, nic_receive_steering_discard)) {
-		err = devlink_fmsg_u64_pair_put(fmsg, "nic_receive_steering_discard",
-						VNIC_ENV_GET64(&vnic,
-							       nic_receive_steering_discard));
-		if (err)
-			return err;
-	}
-
+	if (MLX5_CAP_GEN(dev, vnic_env_cq_overrun))
+		devlink_fmsg_u32_pair_put(fmsg, "cq_overrun",
+					  VNIC_ENV_GET(&vnic, cq_overrun));
+	if (MLX5_CAP_GEN(dev, invalid_command_count))
+		devlink_fmsg_u32_pair_put(fmsg, "invalid_command",
+					  VNIC_ENV_GET(&vnic, invalid_command));
+	if (MLX5_CAP_GEN(dev, quota_exceeded_count))
+		devlink_fmsg_u32_pair_put(fmsg, "quota_exceeded_command",
+					  VNIC_ENV_GET(&vnic, quota_exceeded_command));
+	if (MLX5_CAP_GEN(dev, nic_receive_steering_discard))
+		devlink_fmsg_u64_pair_put(fmsg, "nic_receive_steering_discard",
+					  VNIC_ENV_GET64(&vnic, nic_receive_steering_discard));
 	if (MLX5_CAP_GEN(dev, vnic_env_cnt_steering_fail)) {
-		err = devlink_fmsg_u64_pair_put(fmsg, "generated_pkt_steering_fail",
-						VNIC_ENV_GET64(&vnic,
-							       generated_pkt_steering_fail));
-		if (err)
-			return err;
-
-		err = devlink_fmsg_u64_pair_put(fmsg, "handled_pkt_steering_fail",
-						VNIC_ENV_GET64(&vnic, handled_pkt_steering_fail));
-		if (err)
-			return err;
+		devlink_fmsg_u64_pair_put(fmsg, "generated_pkt_steering_fail",
+					  VNIC_ENV_GET64(&vnic, generated_pkt_steering_fail));
+		devlink_fmsg_u64_pair_put(fmsg, "handled_pkt_steering_fail",
+					  VNIC_ENV_GET64(&vnic, handled_pkt_steering_fail));
 	}
 
-	err = devlink_fmsg_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_pair_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_obj_nest_end(fmsg);
+	devlink_fmsg_pair_nest_end(fmsg);
 }
 
 static int mlx5_reporter_vnic_diagnose(struct devlink_health_reporter *reporter,
@@ -121,7 +70,8 @@ static int mlx5_reporter_vnic_diagnose(struct devlink_health_reporter *reporter,
 {
 	struct mlx5_core_dev *dev = devlink_health_reporter_priv(reporter);
 
-	return mlx5_reporter_vnic_diagnose_counters(dev, fmsg, 0, false);
+	mlx5_reporter_vnic_diagnose_counters(dev, fmsg, 0, false);
+	return 0;
 }
 
 static const struct devlink_health_reporter_ops mlx5_reporter_vnic_ops = {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h
index eba87a39e9..fbc31256f7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h
@@ -9,8 +9,8 @@
 void mlx5_reporter_vnic_create(struct mlx5_core_dev *dev);
 void mlx5_reporter_vnic_destroy(struct mlx5_core_dev *dev);
 
-int mlx5_reporter_vnic_diagnose_counters(struct mlx5_core_dev *dev,
-					 struct devlink_fmsg *fmsg,
-					 u16 vport_num, bool other_vport);
+void mlx5_reporter_vnic_diagnose_counters(struct mlx5_core_dev *dev,
+					  struct devlink_fmsg *fmsg,
+					  u16 vport_num, bool other_vport);
 
 #endif /* __MLX5_REPORTER_VNIC_H */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dpll.c b/drivers/net/ethernet/mellanox/mlx5/core/dpll.c
new file mode 100644
index 0000000000..8ce5c8bcda
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/dpll.c
@@ -0,0 +1,432 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
+
+#include <linux/dpll.h>
+#include <linux/mlx5/driver.h>
+
+/* This structure represents a reference to DPLL, one is created
+ * per mdev instance.
+ */
+struct mlx5_dpll {
+	struct dpll_device *dpll;
+	struct dpll_pin *dpll_pin;
+	struct mlx5_core_dev *mdev;
+	struct workqueue_struct *wq;
+	struct delayed_work work;
+	struct {
+		bool valid;
+		enum dpll_lock_status lock_status;
+		enum dpll_pin_state pin_state;
+	} last;
+	struct notifier_block mdev_nb;
+	struct net_device *tracking_netdev;
+};
+
+static int mlx5_dpll_clock_id_get(struct mlx5_core_dev *mdev, u64 *clock_id)
+{
+	u32 out[MLX5_ST_SZ_DW(msecq_reg)] = {};
+	u32 in[MLX5_ST_SZ_DW(msecq_reg)] = {};
+	int err;
+
+	err = mlx5_core_access_reg(mdev, in, sizeof(in), out, sizeof(out),
+				   MLX5_REG_MSECQ, 0, 0);
+	if (err)
+		return err;
+	*clock_id = MLX5_GET64(msecq_reg, out, local_clock_identity);
+	return 0;
+}
+
+static int
+mlx5_dpll_synce_status_get(struct mlx5_core_dev *mdev,
+			   enum mlx5_msees_admin_status *admin_status,
+			   enum mlx5_msees_oper_status *oper_status,
+			   bool *ho_acq)
+{
+	u32 out[MLX5_ST_SZ_DW(msees_reg)] = {};
+	u32 in[MLX5_ST_SZ_DW(msees_reg)] = {};
+	int err;
+
+	err = mlx5_core_access_reg(mdev, in, sizeof(in), out, sizeof(out),
+				   MLX5_REG_MSEES, 0, 0);
+	if (err)
+		return err;
+	if (admin_status)
+		*admin_status = MLX5_GET(msees_reg, out, admin_status);
+	*oper_status = MLX5_GET(msees_reg, out, oper_status);
+	if (ho_acq)
+		*ho_acq = MLX5_GET(msees_reg, out, ho_acq);
+	return 0;
+}
+
+static int
+mlx5_dpll_synce_status_set(struct mlx5_core_dev *mdev,
+			   enum mlx5_msees_admin_status admin_status)
+{
+	u32 out[MLX5_ST_SZ_DW(msees_reg)] = {};
+	u32 in[MLX5_ST_SZ_DW(msees_reg)] = {};
+
+	MLX5_SET(msees_reg, in, field_select,
+		 MLX5_MSEES_FIELD_SELECT_ENABLE |
+		 MLX5_MSEES_FIELD_SELECT_ADMIN_STATUS);
+	MLX5_SET(msees_reg, in, admin_status, admin_status);
+	return mlx5_core_access_reg(mdev, in, sizeof(in), out, sizeof(out),
+				    MLX5_REG_MSEES, 0, 1);
+}
+
+static enum dpll_lock_status
+mlx5_dpll_lock_status_get(enum mlx5_msees_oper_status oper_status, bool ho_acq)
+{
+	switch (oper_status) {
+	case MLX5_MSEES_OPER_STATUS_SELF_TRACK:
+		fallthrough;
+	case MLX5_MSEES_OPER_STATUS_OTHER_TRACK:
+		return ho_acq ? DPLL_LOCK_STATUS_LOCKED_HO_ACQ :
+				DPLL_LOCK_STATUS_LOCKED;
+	case MLX5_MSEES_OPER_STATUS_HOLDOVER:
+		fallthrough;
+	case MLX5_MSEES_OPER_STATUS_FAIL_HOLDOVER:
+		return DPLL_LOCK_STATUS_HOLDOVER;
+	default:
+		return DPLL_LOCK_STATUS_UNLOCKED;
+	}
+}
+
+static enum dpll_pin_state
+mlx5_dpll_pin_state_get(enum mlx5_msees_admin_status admin_status,
+			enum mlx5_msees_oper_status oper_status)
+{
+	return (admin_status == MLX5_MSEES_ADMIN_STATUS_TRACK &&
+		(oper_status == MLX5_MSEES_OPER_STATUS_SELF_TRACK ||
+		 oper_status == MLX5_MSEES_OPER_STATUS_OTHER_TRACK)) ?
+	       DPLL_PIN_STATE_CONNECTED : DPLL_PIN_STATE_DISCONNECTED;
+}
+
+static int mlx5_dpll_device_lock_status_get(const struct dpll_device *dpll,
+					    void *priv,
+					    enum dpll_lock_status *status,
+					    struct netlink_ext_ack *extack)
+{
+	enum mlx5_msees_oper_status oper_status;
+	struct mlx5_dpll *mdpll = priv;
+	bool ho_acq;
+	int err;
+
+	err = mlx5_dpll_synce_status_get(mdpll->mdev, NULL,
+					 &oper_status, &ho_acq);
+	if (err)
+		return err;
+
+	*status = mlx5_dpll_lock_status_get(oper_status, ho_acq);
+	return 0;
+}
+
+static int mlx5_dpll_device_mode_get(const struct dpll_device *dpll,
+				     void *priv, enum dpll_mode *mode,
+				     struct netlink_ext_ack *extack)
+{
+	*mode = DPLL_MODE_MANUAL;
+	return 0;
+}
+
+static bool mlx5_dpll_device_mode_supported(const struct dpll_device *dpll,
+					    void *priv,
+					    enum dpll_mode mode,
+					    struct netlink_ext_ack *extack)
+{
+	return mode == DPLL_MODE_MANUAL;
+}
+
+static const struct dpll_device_ops mlx5_dpll_device_ops = {
+	.lock_status_get = mlx5_dpll_device_lock_status_get,
+	.mode_get = mlx5_dpll_device_mode_get,
+	.mode_supported = mlx5_dpll_device_mode_supported,
+};
+
+static int mlx5_dpll_pin_direction_get(const struct dpll_pin *pin,
+				       void *pin_priv,
+				       const struct dpll_device *dpll,
+				       void *dpll_priv,
+				       enum dpll_pin_direction *direction,
+				       struct netlink_ext_ack *extack)
+{
+	*direction = DPLL_PIN_DIRECTION_INPUT;
+	return 0;
+}
+
+static int mlx5_dpll_state_on_dpll_get(const struct dpll_pin *pin,
+				       void *pin_priv,
+				       const struct dpll_device *dpll,
+				       void *dpll_priv,
+				       enum dpll_pin_state *state,
+				       struct netlink_ext_ack *extack)
+{
+	enum mlx5_msees_admin_status admin_status;
+	enum mlx5_msees_oper_status oper_status;
+	struct mlx5_dpll *mdpll = pin_priv;
+	int err;
+
+	err = mlx5_dpll_synce_status_get(mdpll->mdev, &admin_status,
+					 &oper_status, NULL);
+	if (err)
+		return err;
+	*state = mlx5_dpll_pin_state_get(admin_status, oper_status);
+	return 0;
+}
+
+static int mlx5_dpll_state_on_dpll_set(const struct dpll_pin *pin,
+				       void *pin_priv,
+				       const struct dpll_device *dpll,
+				       void *dpll_priv,
+				       enum dpll_pin_state state,
+				       struct netlink_ext_ack *extack)
+{
+	struct mlx5_dpll *mdpll = pin_priv;
+
+	return mlx5_dpll_synce_status_set(mdpll->mdev,
+					  state == DPLL_PIN_STATE_CONNECTED ?
+					  MLX5_MSEES_ADMIN_STATUS_TRACK :
+					  MLX5_MSEES_ADMIN_STATUS_FREE_RUNNING);
+}
+
+static const struct dpll_pin_ops mlx5_dpll_pins_ops = {
+	.direction_get = mlx5_dpll_pin_direction_get,
+	.state_on_dpll_get = mlx5_dpll_state_on_dpll_get,
+	.state_on_dpll_set = mlx5_dpll_state_on_dpll_set,
+};
+
+static const struct dpll_pin_properties mlx5_dpll_pin_properties = {
+	.type = DPLL_PIN_TYPE_SYNCE_ETH_PORT,
+	.capabilities = DPLL_PIN_CAPABILITIES_STATE_CAN_CHANGE,
+};
+
+#define MLX5_DPLL_PERIODIC_WORK_INTERVAL 500 /* ms */
+
+static void mlx5_dpll_periodic_work_queue(struct mlx5_dpll *mdpll)
+{
+	queue_delayed_work(mdpll->wq, &mdpll->work,
+			   msecs_to_jiffies(MLX5_DPLL_PERIODIC_WORK_INTERVAL));
+}
+
+static void mlx5_dpll_periodic_work(struct work_struct *work)
+{
+	struct mlx5_dpll *mdpll = container_of(work, struct mlx5_dpll,
+					       work.work);
+	enum mlx5_msees_admin_status admin_status;
+	enum mlx5_msees_oper_status oper_status;
+	enum dpll_lock_status lock_status;
+	enum dpll_pin_state pin_state;
+	bool ho_acq;
+	int err;
+
+	err = mlx5_dpll_synce_status_get(mdpll->mdev, &admin_status,
+					 &oper_status, &ho_acq);
+	if (err)
+		goto err_out;
+	lock_status = mlx5_dpll_lock_status_get(oper_status, ho_acq);
+	pin_state = mlx5_dpll_pin_state_get(admin_status, oper_status);
+
+	if (!mdpll->last.valid)
+		goto invalid_out;
+
+	if (mdpll->last.lock_status != lock_status)
+		dpll_device_change_ntf(mdpll->dpll);
+	if (mdpll->last.pin_state != pin_state)
+		dpll_pin_change_ntf(mdpll->dpll_pin);
+
+invalid_out:
+	mdpll->last.lock_status = lock_status;
+	mdpll->last.pin_state = pin_state;
+	mdpll->last.valid = true;
+err_out:
+	mlx5_dpll_periodic_work_queue(mdpll);
+}
+
+static void mlx5_dpll_netdev_dpll_pin_set(struct mlx5_dpll *mdpll,
+					  struct net_device *netdev)
+{
+	if (mdpll->tracking_netdev)
+		return;
+	netdev_dpll_pin_set(netdev, mdpll->dpll_pin);
+	mdpll->tracking_netdev = netdev;
+}
+
+static void mlx5_dpll_netdev_dpll_pin_clear(struct mlx5_dpll *mdpll)
+{
+	if (!mdpll->tracking_netdev)
+		return;
+	netdev_dpll_pin_clear(mdpll->tracking_netdev);
+	mdpll->tracking_netdev = NULL;
+}
+
+static int mlx5_dpll_mdev_notifier_event(struct notifier_block *nb,
+					 unsigned long event, void *data)
+{
+	struct mlx5_dpll *mdpll = container_of(nb, struct mlx5_dpll, mdev_nb);
+	struct net_device *netdev = data;
+
+	switch (event) {
+	case MLX5_DRIVER_EVENT_UPLINK_NETDEV:
+		if (netdev)
+			mlx5_dpll_netdev_dpll_pin_set(mdpll, netdev);
+		else
+			mlx5_dpll_netdev_dpll_pin_clear(mdpll);
+		break;
+	default:
+		return NOTIFY_DONE;
+	}
+
+	return NOTIFY_OK;
+}
+
+static void mlx5_dpll_mdev_netdev_track(struct mlx5_dpll *mdpll,
+					struct mlx5_core_dev *mdev)
+{
+	mdpll->mdev_nb.notifier_call = mlx5_dpll_mdev_notifier_event;
+	mlx5_blocking_notifier_register(mdev, &mdpll->mdev_nb);
+	mlx5_core_uplink_netdev_event_replay(mdev);
+}
+
+static void mlx5_dpll_mdev_netdev_untrack(struct mlx5_dpll *mdpll,
+					  struct mlx5_core_dev *mdev)
+{
+	mlx5_blocking_notifier_unregister(mdev, &mdpll->mdev_nb);
+	mlx5_dpll_netdev_dpll_pin_clear(mdpll);
+}
+
+static int mlx5_dpll_probe(struct auxiliary_device *adev,
+			   const struct auxiliary_device_id *id)
+{
+	struct mlx5_adev *edev = container_of(adev, struct mlx5_adev, adev);
+	struct mlx5_core_dev *mdev = edev->mdev;
+	struct mlx5_dpll *mdpll;
+	u64 clock_id;
+	int err;
+
+	err = mlx5_dpll_synce_status_set(mdev,
+					 MLX5_MSEES_ADMIN_STATUS_FREE_RUNNING);
+	if (err)
+		return err;
+
+	err = mlx5_dpll_clock_id_get(mdev, &clock_id);
+	if (err)
+		return err;
+
+	mdpll = kzalloc(sizeof(*mdpll), GFP_KERNEL);
+	if (!mdpll)
+		return -ENOMEM;
+	mdpll->mdev = mdev;
+	auxiliary_set_drvdata(adev, mdpll);
+
+	/* Multiple mdev instances might share one DPLL device. */
+	mdpll->dpll = dpll_device_get(clock_id, 0, THIS_MODULE);
+	if (IS_ERR(mdpll->dpll)) {
+		err = PTR_ERR(mdpll->dpll);
+		goto err_free_mdpll;
+	}
+
+	err = dpll_device_register(mdpll->dpll, DPLL_TYPE_EEC,
+				   &mlx5_dpll_device_ops, mdpll);
+	if (err)
+		goto err_put_dpll_device;
+
+	/* Multiple mdev instances might share one DPLL pin. */
+	mdpll->dpll_pin = dpll_pin_get(clock_id, mlx5_get_dev_index(mdev),
+				       THIS_MODULE, &mlx5_dpll_pin_properties);
+	if (IS_ERR(mdpll->dpll_pin)) {
+		err = PTR_ERR(mdpll->dpll_pin);
+		goto err_unregister_dpll_device;
+	}
+
+	err = dpll_pin_register(mdpll->dpll, mdpll->dpll_pin,
+				&mlx5_dpll_pins_ops, mdpll);
+	if (err)
+		goto err_put_dpll_pin;
+
+	mdpll->wq = create_singlethread_workqueue("mlx5_dpll");
+	if (!mdpll->wq) {
+		err = -ENOMEM;
+		goto err_unregister_dpll_pin;
+	}
+
+	mlx5_dpll_mdev_netdev_track(mdpll, mdev);
+
+	INIT_DELAYED_WORK(&mdpll->work, &mlx5_dpll_periodic_work);
+	mlx5_dpll_periodic_work_queue(mdpll);
+
+	return 0;
+
+err_unregister_dpll_pin:
+	dpll_pin_unregister(mdpll->dpll, mdpll->dpll_pin,
+			    &mlx5_dpll_pins_ops, mdpll);
+err_put_dpll_pin:
+	dpll_pin_put(mdpll->dpll_pin);
+err_unregister_dpll_device:
+	dpll_device_unregister(mdpll->dpll, &mlx5_dpll_device_ops, mdpll);
+err_put_dpll_device:
+	dpll_device_put(mdpll->dpll);
+err_free_mdpll:
+	kfree(mdpll);
+	return err;
+}
+
+static void mlx5_dpll_remove(struct auxiliary_device *adev)
+{
+	struct mlx5_dpll *mdpll = auxiliary_get_drvdata(adev);
+	struct mlx5_core_dev *mdev = mdpll->mdev;
+
+	cancel_delayed_work_sync(&mdpll->work);
+	mlx5_dpll_mdev_netdev_untrack(mdpll, mdev);
+	destroy_workqueue(mdpll->wq);
+	dpll_pin_unregister(mdpll->dpll, mdpll->dpll_pin,
+			    &mlx5_dpll_pins_ops, mdpll);
+	dpll_pin_put(mdpll->dpll_pin);
+	dpll_device_unregister(mdpll->dpll, &mlx5_dpll_device_ops, mdpll);
+	dpll_device_put(mdpll->dpll);
+	kfree(mdpll);
+
+	mlx5_dpll_synce_status_set(mdev,
+				   MLX5_MSEES_ADMIN_STATUS_FREE_RUNNING);
+}
+
+static int mlx5_dpll_suspend(struct auxiliary_device *adev, pm_message_t state)
+{
+	return 0;
+}
+
+static int mlx5_dpll_resume(struct auxiliary_device *adev)
+{
+	return 0;
+}
+
+static const struct auxiliary_device_id mlx5_dpll_id_table[] = {
+	{ .name = MLX5_ADEV_NAME ".dpll", },
+	{},
+};
+
+MODULE_DEVICE_TABLE(auxiliary, mlx5_dpll_id_table);
+
+static struct auxiliary_driver mlx5_dpll_driver = {
+	.name = "dpll",
+	.probe = mlx5_dpll_probe,
+	.remove = mlx5_dpll_remove,
+	.suspend = mlx5_dpll_suspend,
+	.resume = mlx5_dpll_resume,
+	.id_table = mlx5_dpll_id_table,
+};
+
+static int __init mlx5_dpll_init(void)
+{
+	return auxiliary_driver_register(&mlx5_dpll_driver);
+}
+
+static void __exit mlx5_dpll_exit(void)
+{
+	auxiliary_driver_unregister(&mlx5_dpll_driver);
+}
+
+module_init(mlx5_dpll_init);
+module_exit(mlx5_dpll_exit);
+
+MODULE_AUTHOR("Jiri Pirko <jiri@nvidia.com>");
+MODULE_DESCRIPTION("Mellanox 5th generation network adapters (ConnectX series) DPLL driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 20a6bc1a23..729a11b5fb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -141,7 +141,7 @@ struct page_pool;
 #define MLX5E_PARAMS_DEFAULT_MIN_RX_WQES_MPW            0x2
 
 #define MLX5E_MIN_NUM_CHANNELS         0x1
-#define MLX5E_MAX_NUM_CHANNELS         (MLX5E_INDIR_RQT_SIZE / 2)
+#define MLX5E_MAX_NUM_CHANNELS         256
 #define MLX5E_TX_CQ_POLL_BUDGET        128
 #define MLX5E_TX_XSK_POLL_BUDGET       64
 #define MLX5E_SQ_RECOVER_MIN_INTERVAL  500 /* msecs */
@@ -168,6 +168,13 @@ struct page_pool;
 #define mlx5e_state_dereference(priv, p) \
 	rcu_dereference_protected((p), lockdep_is_held(&(priv)->state_lock))
 
+enum mlx5e_devcom_events {
+	MPV_DEVCOM_MASTER_UP,
+	MPV_DEVCOM_MASTER_DOWN,
+	MPV_DEVCOM_IPSEC_MASTER_UP,
+	MPV_DEVCOM_IPSEC_MASTER_DOWN,
+};
+
 static inline u8 mlx5e_get_num_lag_ports(struct mlx5_core_dev *mdev)
 {
 	if (mlx5_lag_is_lacp_owner(mdev))
@@ -193,7 +200,8 @@ static inline int mlx5e_get_max_num_channels(struct mlx5_core_dev *mdev)
 {
 	return is_kdump_kernel() ?
 		MLX5E_MIN_NUM_CHANNELS :
-		min_t(int, mlx5_comp_vectors_max(mdev), MLX5E_MAX_NUM_CHANNELS);
+		min3(mlx5_comp_vectors_max(mdev), (u32)MLX5E_MAX_NUM_CHANNELS,
+		     (u32)(1 << MLX5_CAP_GEN(mdev, log_max_rqt_size)));
 }
 
 /* The maximum WQE size can be retrieved by max_wqe_sz_sq in
@@ -937,6 +945,7 @@ struct mlx5e_priv {
 	struct mlx5e_htb          *htb;
 	struct mlx5e_mqprio_rl    *mqprio_rl;
 	struct dentry             *dfs_root;
+	struct mlx5_devcom_comp_dev *devcom;
 };
 
 struct mlx5e_dev {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c
index c6b6e290fd..0b1ac6e5c8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c
@@ -12,11 +12,19 @@ struct mlx5e_dev *mlx5e_create_devlink(struct device *dev,
 {
 	struct mlx5e_dev *mlx5e_dev;
 	struct devlink *devlink;
+	int err;
 
 	devlink = devlink_alloc_ns(&mlx5e_devlink_ops, sizeof(*mlx5e_dev),
 				   devlink_net(priv_to_devlink(mdev)), dev);
 	if (!devlink)
 		return ERR_PTR(-ENOMEM);
+
+	err = devl_nested_devlink_set(priv_to_devlink(mdev), devlink);
+	if (err) {
+		devlink_free(devlink);
+		return ERR_PTR(err);
+	}
+
 	devlink_register(devlink);
 	return devlink_priv(devlink);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
index e5a44b0b96..4d6225e0ee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
@@ -150,7 +150,6 @@ struct mlx5e_flow_steering *mlx5e_fs_init(const struct mlx5e_profile *profile,
 					  struct dentry *dfs_root);
 void mlx5e_fs_cleanup(struct mlx5e_flow_steering *fs);
 struct mlx5e_vlan_table *mlx5e_fs_get_vlan(struct mlx5e_flow_steering *fs);
-void mlx5e_fs_set_tc(struct mlx5e_flow_steering *fs, struct mlx5e_tc_table *tc);
 struct mlx5e_tc_table *mlx5e_fs_get_tc(struct mlx5e_flow_steering *fs);
 struct mlx5e_l2_table *mlx5e_fs_get_l2(struct mlx5e_flow_steering *fs);
 struct mlx5_flow_namespace *mlx5e_fs_get_ns(struct mlx5e_flow_steering *fs, bool egress);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c
index 6f4e6c34b2..81523825fa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c
@@ -5,134 +5,59 @@
 #include "lib/eq.h"
 #include "lib/mlx5.h"
 
-int mlx5e_health_fmsg_named_obj_nest_start(struct devlink_fmsg *fmsg, char *name)
+void mlx5e_health_fmsg_named_obj_nest_start(struct devlink_fmsg *fmsg, char *name)
 {
-	int err;
-
-	err = devlink_fmsg_pair_nest_start(fmsg, name);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_pair_nest_start(fmsg, name);
+	devlink_fmsg_obj_nest_start(fmsg);
 }
 
-int mlx5e_health_fmsg_named_obj_nest_end(struct devlink_fmsg *fmsg)
+void mlx5e_health_fmsg_named_obj_nest_end(struct devlink_fmsg *fmsg)
 {
-	int err;
-
-	err = devlink_fmsg_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_pair_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_obj_nest_end(fmsg);
+	devlink_fmsg_pair_nest_end(fmsg);
 }
 
-int mlx5e_health_cq_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg)
+void mlx5e_health_cq_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg)
 {
 	u32 out[MLX5_ST_SZ_DW(query_cq_out)] = {};
 	u8 hw_status;
 	void *cqc;
-	int err;
-
-	err = mlx5_core_query_cq(cq->mdev, &cq->mcq, out);
-	if (err)
-		return err;
 
+	mlx5_core_query_cq(cq->mdev, &cq->mcq, out);
 	cqc = MLX5_ADDR_OF(query_cq_out, out, cq_context);
 	hw_status = MLX5_GET(cqc, cqc, status);
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "CQ");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "cqn", cq->mcq.cqn);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "HW status", hw_status);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "ci", mlx5_cqwq_get_ci(&cq->wq));
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "size", mlx5_cqwq_get_size(&cq->wq));
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "CQ");
+	devlink_fmsg_u32_pair_put(fmsg, "cqn", cq->mcq.cqn);
+	devlink_fmsg_u8_pair_put(fmsg, "HW status", hw_status);
+	devlink_fmsg_u32_pair_put(fmsg, "ci", mlx5_cqwq_get_ci(&cq->wq));
+	devlink_fmsg_u32_pair_put(fmsg, "size", mlx5_cqwq_get_size(&cq->wq));
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-int mlx5e_health_cq_common_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg)
+void mlx5e_health_cq_common_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg)
 {
 	u8 cq_log_stride;
 	u32 cq_sz;
-	int err;
 
 	cq_sz = mlx5_cqwq_get_size(&cq->wq);
 	cq_log_stride = mlx5_cqwq_get_log_stride_size(&cq->wq);
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "CQ");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u64_pair_put(fmsg, "stride size", BIT(cq_log_stride));
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "size", cq_sz);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "CQ");
+	devlink_fmsg_u64_pair_put(fmsg, "stride size", BIT(cq_log_stride));
+	devlink_fmsg_u32_pair_put(fmsg, "size", cq_sz);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-int mlx5e_health_eq_diag_fmsg(struct mlx5_eq_comp *eq, struct devlink_fmsg *fmsg)
+void mlx5e_health_eq_diag_fmsg(struct mlx5_eq_comp *eq, struct devlink_fmsg *fmsg)
 {
-	int err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "EQ");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "eqn", eq->core.eqn);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "irqn", eq->core.irqn);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "vecidx", eq->core.vecidx);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "ci", eq->core.cons_index);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "size", eq_get_size(&eq->core));
-	if (err)
-		return err;
-
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "EQ");
+	devlink_fmsg_u8_pair_put(fmsg, "eqn", eq->core.eqn);
+	devlink_fmsg_u32_pair_put(fmsg, "irqn", eq->core.irqn);
+	devlink_fmsg_u32_pair_put(fmsg, "vecidx", eq->core.vecidx);
+	devlink_fmsg_u32_pair_put(fmsg, "ci", eq->core.cons_index);
+	devlink_fmsg_u32_pair_put(fmsg, "size", eq_get_size(&eq->core));
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
 void mlx5e_health_create_reporters(struct mlx5e_priv *priv)
@@ -235,23 +160,19 @@ int mlx5e_health_report(struct mlx5e_priv *priv,
 }
 
 #define MLX5_HEALTH_DEVLINK_MAX_SIZE 1024
-static int mlx5e_health_rsc_fmsg_binary(struct devlink_fmsg *fmsg,
-					const void *value, u32 value_len)
+static void mlx5e_health_rsc_fmsg_binary(struct devlink_fmsg *fmsg,
+					 const void *value, u32 value_len)
 
 {
 	u32 data_size;
-	int err = 0;
 	u32 offset;
 
 	for (offset = 0; offset < value_len; offset += data_size) {
 		data_size = value_len - offset;
 		if (data_size > MLX5_HEALTH_DEVLINK_MAX_SIZE)
 			data_size = MLX5_HEALTH_DEVLINK_MAX_SIZE;
-		err = devlink_fmsg_binary_put(fmsg, value + offset, data_size);
-		if (err)
-			break;
+		devlink_fmsg_binary_put(fmsg, value + offset, data_size);
 	}
-	return err;
 }
 
 int mlx5e_health_rsc_fmsg_dump(struct mlx5e_priv *priv, struct mlx5_rsc_key *key,
@@ -259,9 +180,8 @@ int mlx5e_health_rsc_fmsg_dump(struct mlx5e_priv *priv, struct mlx5_rsc_key *key
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 	struct mlx5_rsc_dump_cmd *cmd;
+	int cmd_err, err = 0;
 	struct page *page;
-	int cmd_err, err;
-	int end_err;
 	int size;
 
 	if (IS_ERR_OR_NULL(mdev->rsc_dump))
@@ -271,9 +191,7 @@ int mlx5e_health_rsc_fmsg_dump(struct mlx5e_priv *priv, struct mlx5_rsc_key *key
 	if (!page)
 		return -ENOMEM;
 
-	err = devlink_fmsg_binary_pair_nest_start(fmsg, "data");
-	if (err)
-		goto free_page;
+	devlink_fmsg_binary_pair_nest_start(fmsg, "data");
 
 	cmd = mlx5_rsc_dump_cmd_create(mdev, key);
 	if (IS_ERR(cmd)) {
@@ -288,52 +206,31 @@ int mlx5e_health_rsc_fmsg_dump(struct mlx5e_priv *priv, struct mlx5_rsc_key *key
 			goto destroy_cmd;
 		}
 
-		err = mlx5e_health_rsc_fmsg_binary(fmsg, page_address(page), size);
-		if (err)
-			goto destroy_cmd;
-
+		mlx5e_health_rsc_fmsg_binary(fmsg, page_address(page), size);
 	} while (cmd_err > 0);
 
 destroy_cmd:
 	mlx5_rsc_dump_cmd_destroy(cmd);
-	end_err = devlink_fmsg_binary_pair_nest_end(fmsg);
-	if (end_err)
-		err = end_err;
+	devlink_fmsg_binary_pair_nest_end(fmsg);
 free_page:
 	__free_page(page);
 	return err;
 }
 
-int mlx5e_health_queue_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
-			    int queue_idx, char *lbl)
+void mlx5e_health_queue_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
+			     int queue_idx, char *lbl)
 {
 	struct mlx5_rsc_key key = {};
-	int err;
 
 	key.rsc = MLX5_SGMT_TYPE_FULL_QPC;
 	key.index1 = queue_idx;
 	key.size = PAGE_SIZE;
 	key.num_of_obj1 = 1;
 
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, lbl);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "index", queue_idx);
-	if (err)
-		return err;
-
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return devlink_fmsg_obj_nest_end(fmsg);
+	devlink_fmsg_obj_nest_start(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, lbl);
+	devlink_fmsg_u32_pair_put(fmsg, "index", queue_idx);
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	devlink_fmsg_obj_nest_end(fmsg);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/health.h b/drivers/net/ethernet/mellanox/mlx5/core/en/health.h
index 415840c3ef..84be3dd6f7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/health.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/health.h
@@ -20,11 +20,11 @@ void mlx5e_reporter_tx_err_cqe(struct mlx5e_txqsq *sq);
 int mlx5e_reporter_tx_timeout(struct mlx5e_txqsq *sq);
 void mlx5e_reporter_tx_ptpsq_unhealthy(struct mlx5e_ptpsq *ptpsq);
 
-int mlx5e_health_cq_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg);
-int mlx5e_health_cq_common_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg);
-int mlx5e_health_eq_diag_fmsg(struct mlx5_eq_comp *eq, struct devlink_fmsg *fmsg);
-int mlx5e_health_fmsg_named_obj_nest_start(struct devlink_fmsg *fmsg, char *name);
-int mlx5e_health_fmsg_named_obj_nest_end(struct devlink_fmsg *fmsg);
+void mlx5e_health_cq_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg);
+void mlx5e_health_cq_common_diag_fmsg(struct mlx5e_cq *cq, struct devlink_fmsg *fmsg);
+void mlx5e_health_eq_diag_fmsg(struct mlx5_eq_comp *eq, struct devlink_fmsg *fmsg);
+void mlx5e_health_fmsg_named_obj_nest_start(struct devlink_fmsg *fmsg, char *name);
+void mlx5e_health_fmsg_named_obj_nest_end(struct devlink_fmsg *fmsg);
 
 void mlx5e_reporter_rx_create(struct mlx5e_priv *priv);
 void mlx5e_reporter_rx_destroy(struct mlx5e_priv *priv);
@@ -54,6 +54,6 @@ void mlx5e_health_destroy_reporters(struct mlx5e_priv *priv);
 void mlx5e_health_channels_update(struct mlx5e_priv *priv);
 int mlx5e_health_rsc_fmsg_dump(struct mlx5e_priv *priv, struct mlx5_rsc_key *key,
 			       struct devlink_fmsg *fmsg);
-int mlx5e_health_queue_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
-			    int queue_idx, char *lbl);
+void mlx5e_health_queue_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
+			     int queue_idx, char *lbl);
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
index 03b119a434..4358798d6c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
@@ -199,78 +199,38 @@ static int mlx5e_rx_reporter_recover(struct devlink_health_reporter *reporter,
 			 mlx5e_health_recover_channels(priv);
 }
 
-static int mlx5e_reporter_icosq_diagnose(struct mlx5e_icosq *icosq, u8 hw_state,
-					 struct devlink_fmsg *fmsg)
+static void mlx5e_reporter_icosq_diagnose(struct mlx5e_icosq *icosq, u8 hw_state,
+					  struct devlink_fmsg *fmsg)
 {
-	int err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "ICOSQ");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "sqn", icosq->sqn);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "HW state", hw_state);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "cc", icosq->cc);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "pc", icosq->pc);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "WQE size",
-					mlx5_wq_cyc_get_size(&icosq->wq));
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "CQ");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "cqn", icosq->cq.mcq.cqn);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "cc", icosq->cq.wq.cc);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "size", mlx5_cqwq_get_size(&icosq->cq.wq));
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "ICOSQ");
+	devlink_fmsg_u32_pair_put(fmsg, "sqn", icosq->sqn);
+	devlink_fmsg_u8_pair_put(fmsg, "HW state", hw_state);
+	devlink_fmsg_u32_pair_put(fmsg, "cc", icosq->cc);
+	devlink_fmsg_u32_pair_put(fmsg, "pc", icosq->pc);
+	devlink_fmsg_u32_pair_put(fmsg, "WQE size", mlx5_wq_cyc_get_size(&icosq->wq));
+
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "CQ");
+	devlink_fmsg_u32_pair_put(fmsg, "cqn", icosq->cq.mcq.cqn);
+	devlink_fmsg_u32_pair_put(fmsg, "cc", icosq->cq.wq.cc);
+	devlink_fmsg_u32_pair_put(fmsg, "size", mlx5_cqwq_get_size(&icosq->cq.wq));
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-static int mlx5e_health_rq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_rq *rq)
+static void mlx5e_health_rq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_rq *rq)
 {
-	int err;
 	int i;
 
 	BUILD_BUG_ON_MSG(ARRAY_SIZE(rq_sw_state_type_name) != MLX5E_NUM_RQ_STATES,
 			 "rq_sw_state_type_name string array must be consistent with MLX5E_RQ_STATE_* enum in en.h");
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
 
-	for (i = 0; i < ARRAY_SIZE(rq_sw_state_type_name); ++i) {
-		err = devlink_fmsg_u32_pair_put(fmsg, rq_sw_state_type_name[i],
-						test_bit(i, &rq->state));
-		if (err)
-			return err;
-	}
+	for (i = 0; i < ARRAY_SIZE(rq_sw_state_type_name); ++i)
+		devlink_fmsg_u32_pair_put(fmsg, rq_sw_state_type_name[i],
+					  test_bit(i, &rq->state));
 
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
 static int
@@ -291,184 +251,93 @@ mlx5e_rx_reporter_build_diagnose_output_rq_common(struct mlx5e_rq *rq,
 	wq_head = mlx5e_rqwq_get_head(rq);
 	wqe_counter = mlx5e_rqwq_get_wqe_counter(rq);
 
-	err = devlink_fmsg_u32_pair_put(fmsg, "rqn", rq->rqn);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "HW state", hw_state);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "WQE counter", wqe_counter);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "posted WQEs", wqes_sz);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "cc", wq_head);
-	if (err)
-		return err;
-
-	err = mlx5e_health_rq_put_sw_state(fmsg, rq);
-	if (err)
-		return err;
-
-	err = mlx5e_health_cq_diag_fmsg(&rq->cq, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_eq_diag_fmsg(rq->cq.mcq.eq, fmsg);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "rqn", rq->rqn);
+	devlink_fmsg_u8_pair_put(fmsg, "HW state", hw_state);
+	devlink_fmsg_u32_pair_put(fmsg, "WQE counter", wqe_counter);
+	devlink_fmsg_u32_pair_put(fmsg, "posted WQEs", wqes_sz);
+	devlink_fmsg_u32_pair_put(fmsg, "cc", wq_head);
+	mlx5e_health_rq_put_sw_state(fmsg, rq);
+	mlx5e_health_cq_diag_fmsg(&rq->cq, fmsg);
+	mlx5e_health_eq_diag_fmsg(rq->cq.mcq.eq, fmsg);
 
 	if (rq->icosq) {
 		struct mlx5e_icosq *icosq = rq->icosq;
 		u8 icosq_hw_state;
+		int err;
 
 		err = mlx5_core_query_sq_state(rq->mdev, icosq->sqn, &icosq_hw_state);
 		if (err)
 			return err;
 
-		err = mlx5e_reporter_icosq_diagnose(icosq, icosq_hw_state, fmsg);
-		if (err)
-			return err;
+		mlx5e_reporter_icosq_diagnose(icosq, icosq_hw_state, fmsg);
 	}
 
 	return 0;
 }
 
-static int mlx5e_rx_reporter_build_diagnose_output(struct mlx5e_rq *rq,
-						   struct devlink_fmsg *fmsg)
+static void mlx5e_rx_reporter_build_diagnose_output(struct mlx5e_rq *rq,
+						    struct devlink_fmsg *fmsg)
 {
-	int err;
-
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "channel ix", rq->ix);
-	if (err)
-		return err;
-
-	err = mlx5e_rx_reporter_build_diagnose_output_rq_common(rq, fmsg);
-	if (err)
-		return err;
-
-	return devlink_fmsg_obj_nest_end(fmsg);
+	devlink_fmsg_obj_nest_start(fmsg);
+	devlink_fmsg_u32_pair_put(fmsg, "channel ix", rq->ix);
+	mlx5e_rx_reporter_build_diagnose_output_rq_common(rq, fmsg);
+	devlink_fmsg_obj_nest_end(fmsg);
 }
 
-static int mlx5e_rx_reporter_diagnose_generic_rq(struct mlx5e_rq *rq,
-						 struct devlink_fmsg *fmsg)
+static void mlx5e_rx_reporter_diagnose_generic_rq(struct mlx5e_rq *rq,
+						  struct devlink_fmsg *fmsg)
 {
 	struct mlx5e_priv *priv = rq->priv;
 	struct mlx5e_params *params;
 	u32 rq_stride, rq_sz;
 	bool real_time;
-	int err;
 
 	params = &priv->channels.params;
 	rq_sz = mlx5e_rqwq_get_size(rq);
 	real_time =  mlx5_is_real_time_rq(priv->mdev);
 	rq_stride = BIT(mlx5e_mpwqe_get_log_stride_size(priv->mdev, params, NULL));
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RQ");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "type", params->rq_wq_type);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u64_pair_put(fmsg, "stride size", rq_stride);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "size", rq_sz);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_string_pair_put(fmsg, "ts_format", real_time ? "RT" : "FRC");
-	if (err)
-		return err;
-
-	err = mlx5e_health_cq_common_diag_fmsg(&rq->cq, fmsg);
-	if (err)
-		return err;
-
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RQ");
+	devlink_fmsg_u8_pair_put(fmsg, "type", params->rq_wq_type);
+	devlink_fmsg_u64_pair_put(fmsg, "stride size", rq_stride);
+	devlink_fmsg_u32_pair_put(fmsg, "size", rq_sz);
+	devlink_fmsg_string_pair_put(fmsg, "ts_format", real_time ? "RT" : "FRC");
+	mlx5e_health_cq_common_diag_fmsg(&rq->cq, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-static int
+static void
 mlx5e_rx_reporter_diagnose_common_ptp_config(struct mlx5e_priv *priv, struct mlx5e_ptp *ptp_ch,
 					     struct devlink_fmsg *fmsg)
 {
-	int err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "PTP");
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "filter_type", priv->tstamp.rx_filter);
-	if (err)
-		return err;
-
-	err = mlx5e_rx_reporter_diagnose_generic_rq(&ptp_ch->rq, fmsg);
-	if (err)
-		return err;
-
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "PTP");
+	devlink_fmsg_u32_pair_put(fmsg, "filter_type", priv->tstamp.rx_filter);
+	mlx5e_rx_reporter_diagnose_generic_rq(&ptp_ch->rq, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-static int
+static void
 mlx5e_rx_reporter_diagnose_common_config(struct devlink_health_reporter *reporter,
 					 struct devlink_fmsg *fmsg)
 {
 	struct mlx5e_priv *priv = devlink_health_reporter_priv(reporter);
 	struct mlx5e_rq *generic_rq = &priv->channels.c[0]->rq;
 	struct mlx5e_ptp *ptp_ch = priv->channels.ptp;
-	int err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Common config");
-	if (err)
-		return err;
-
-	err = mlx5e_rx_reporter_diagnose_generic_rq(generic_rq, fmsg);
-	if (err)
-		return err;
 
-	if (ptp_ch && test_bit(MLX5E_PTP_STATE_RX, ptp_ch->state)) {
-		err = mlx5e_rx_reporter_diagnose_common_ptp_config(priv, ptp_ch, fmsg);
-		if (err)
-			return err;
-	}
-
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Common config");
+	mlx5e_rx_reporter_diagnose_generic_rq(generic_rq, fmsg);
+	if (ptp_ch && test_bit(MLX5E_PTP_STATE_RX, ptp_ch->state))
+		mlx5e_rx_reporter_diagnose_common_ptp_config(priv, ptp_ch, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-static int mlx5e_rx_reporter_build_diagnose_output_ptp_rq(struct mlx5e_rq *rq,
-							  struct devlink_fmsg *fmsg)
+static void mlx5e_rx_reporter_build_diagnose_output_ptp_rq(struct mlx5e_rq *rq,
+							   struct devlink_fmsg *fmsg)
 {
-	int err;
-
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_string_pair_put(fmsg, "channel", "ptp");
-	if (err)
-		return err;
-
-	err = mlx5e_rx_reporter_build_diagnose_output_rq_common(rq, fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_obj_nest_start(fmsg);
+	devlink_fmsg_string_pair_put(fmsg, "channel", "ptp");
+	mlx5e_rx_reporter_build_diagnose_output_rq_common(rq, fmsg);
+	devlink_fmsg_obj_nest_end(fmsg);
 }
 
 static int mlx5e_rx_reporter_diagnose(struct devlink_health_reporter *reporter,
@@ -477,20 +346,15 @@ static int mlx5e_rx_reporter_diagnose(struct devlink_health_reporter *reporter,
 {
 	struct mlx5e_priv *priv = devlink_health_reporter_priv(reporter);
 	struct mlx5e_ptp *ptp_ch = priv->channels.ptp;
-	int i, err = 0;
+	int i;
 
 	mutex_lock(&priv->state_lock);
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		goto unlock;
 
-	err = mlx5e_rx_reporter_diagnose_common_config(reporter, fmsg);
-	if (err)
-		goto unlock;
-
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "RQs");
-	if (err)
-		goto unlock;
+	mlx5e_rx_reporter_diagnose_common_config(reporter, fmsg);
+	devlink_fmsg_arr_pair_nest_start(fmsg, "RQs");
 
 	for (i = 0; i < priv->channels.num; i++) {
 		struct mlx5e_channel *c = priv->channels.c[i];
@@ -499,19 +363,14 @@ static int mlx5e_rx_reporter_diagnose(struct devlink_health_reporter *reporter,
 		rq = test_bit(MLX5E_CHANNEL_STATE_XSK, c->state) ?
 			&c->xskrq : &c->rq;
 
-		err = mlx5e_rx_reporter_build_diagnose_output(rq, fmsg);
-		if (err)
-			goto unlock;
-	}
-	if (ptp_ch && test_bit(MLX5E_PTP_STATE_RX, ptp_ch->state)) {
-		err = mlx5e_rx_reporter_build_diagnose_output_ptp_rq(&ptp_ch->rq, fmsg);
-		if (err)
-			goto unlock;
+		mlx5e_rx_reporter_build_diagnose_output(rq, fmsg);
 	}
-	err = devlink_fmsg_arr_pair_nest_end(fmsg);
+	if (ptp_ch && test_bit(MLX5E_PTP_STATE_RX, ptp_ch->state))
+		mlx5e_rx_reporter_build_diagnose_output_ptp_rq(&ptp_ch->rq, fmsg);
+	devlink_fmsg_arr_pair_nest_end(fmsg);
 unlock:
 	mutex_unlock(&priv->state_lock);
-	return err;
+	return 0;
 }
 
 static int mlx5e_rx_reporter_dump_icosq(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
@@ -519,61 +378,34 @@ static int mlx5e_rx_reporter_dump_icosq(struct mlx5e_priv *priv, struct devlink_
 {
 	struct mlx5e_txqsq *icosq = ctx;
 	struct mlx5_rsc_key key = {};
-	int err;
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		return 0;
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SX Slice");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SX Slice");
 	key.size = PAGE_SIZE;
 	key.rsc = MLX5_SGMT_TYPE_SX_SLICE_ALL;
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "ICOSQ");
-	if (err)
-		return err;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "QPC");
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "ICOSQ");
 
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "QPC");
 	key.rsc = MLX5_SGMT_TYPE_FULL_QPC;
 	key.index1 = icosq->sqn;
 	key.num_of_obj1 = 1;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "send_buff");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "send_buff");
 	key.rsc = MLX5_SGMT_TYPE_SND_BUFF;
 	key.num_of_obj2 = MLX5_RSC_DUMP_ALL;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	return 0;
 }
 
 static int mlx5e_rx_reporter_dump_rq(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
@@ -581,60 +413,34 @@ static int mlx5e_rx_reporter_dump_rq(struct mlx5e_priv *priv, struct devlink_fms
 {
 	struct mlx5_rsc_key key = {};
 	struct mlx5e_rq *rq = ctx;
-	int err;
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		return 0;
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RX Slice");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RX Slice");
 	key.size = PAGE_SIZE;
 	key.rsc = MLX5_SGMT_TYPE_RX_SLICE_ALL;
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RQ");
-	if (err)
-		return err;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "QPC");
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RQ");
 
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "QPC");
 	key.rsc = MLX5_SGMT_TYPE_FULL_QPC;
 	key.index1 = rq->rqn;
 	key.num_of_obj1 = 1;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "receive_buff");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "receive_buff");
 	key.rsc = MLX5_SGMT_TYPE_RCV_BUFF;
 	key.num_of_obj2 = MLX5_RSC_DUMP_ALL;
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	return 0;
 }
 
 static int mlx5e_rx_reporter_dump_all_rqs(struct mlx5e_priv *priv,
@@ -642,44 +448,28 @@ static int mlx5e_rx_reporter_dump_all_rqs(struct mlx5e_priv *priv,
 {
 	struct mlx5e_ptp *ptp_ch = priv->channels.ptp;
 	struct mlx5_rsc_key key = {};
-	int i, err;
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		return 0;
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RX Slice");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "RX Slice");
 	key.size = PAGE_SIZE;
 	key.rsc = MLX5_SGMT_TYPE_RX_SLICE_ALL;
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "RQs");
-	if (err)
-		return err;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	devlink_fmsg_arr_pair_nest_start(fmsg, "RQs");
 
-	for (i = 0; i < priv->channels.num; i++) {
+	for (int i = 0; i < priv->channels.num; i++) {
 		struct mlx5e_rq *rq = &priv->channels.c[i]->rq;
 
-		err = mlx5e_health_queue_dump(priv, fmsg, rq->rqn, "RQ");
-		if (err)
-			return err;
+		mlx5e_health_queue_dump(priv, fmsg, rq->rqn, "RQ");
 	}
 
-	if (ptp_ch && test_bit(MLX5E_PTP_STATE_RX, ptp_ch->state)) {
-		err = mlx5e_health_queue_dump(priv, fmsg, ptp_ch->rq.rqn, "PTP RQ");
-		if (err)
-			return err;
-	}
+	if (ptp_ch && test_bit(MLX5E_PTP_STATE_RX, ptp_ch->state))
+		mlx5e_health_queue_dump(priv, fmsg, ptp_ch->rq.rqn, "PTP RQ");
 
-	return devlink_fmsg_arr_pair_nest_end(fmsg);
+	devlink_fmsg_arr_pair_nest_end(fmsg);
+	return 0;
 }
 
 static int mlx5e_rx_reporter_dump_from_ctx(struct mlx5e_priv *priv,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
index ff8242f67c..6b44ddce14 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
@@ -50,25 +50,19 @@ static void mlx5e_reset_txqsq_cc_pc(struct mlx5e_txqsq *sq)
 	sq->pc = 0;
 }
 
-static int mlx5e_health_sq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_txqsq *sq)
+static void mlx5e_health_sq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_txqsq *sq)
 {
-	int err;
 	int i;
 
 	BUILD_BUG_ON_MSG(ARRAY_SIZE(sq_sw_state_type_name) != MLX5E_NUM_SQ_STATES,
 			 "sq_sw_state_type_name string array must be consistent with MLX5E_SQ_STATE_* enum in en.h");
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
 
-	for (i = 0; i < ARRAY_SIZE(sq_sw_state_type_name); ++i) {
-		err = devlink_fmsg_u32_pair_put(fmsg, sq_sw_state_type_name[i],
-						test_bit(i, &sq->state));
-		if (err)
-			return err;
-	}
+	for (i = 0; i < ARRAY_SIZE(sq_sw_state_type_name); ++i)
+		devlink_fmsg_u32_pair_put(fmsg, sq_sw_state_type_name[i],
+					  test_bit(i, &sq->state));
 
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
 static int mlx5e_tx_reporter_err_cqe_recover(void *ctx)
@@ -220,7 +214,7 @@ static int mlx5e_tx_reporter_recover(struct devlink_health_reporter *reporter,
 			 mlx5e_health_recover_channels(priv);
 }
 
-static int
+static void
 mlx5e_tx_reporter_build_diagnose_output_sq_common(struct devlink_fmsg *fmsg,
 						  struct mlx5e_txqsq *sq, int tc)
 {
@@ -229,164 +223,71 @@ mlx5e_tx_reporter_build_diagnose_output_sq_common(struct devlink_fmsg *fmsg,
 	u8 state;
 	int err;
 
-	err = mlx5_core_query_sq_state(priv->mdev, sq->sqn, &state);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "tc", tc);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "txq ix", sq->txq_ix);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "sqn", sq->sqn);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "HW state", state);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_bool_pair_put(fmsg, "stopped", stopped);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "tc", tc);
+	devlink_fmsg_u32_pair_put(fmsg, "txq ix", sq->txq_ix);
+	devlink_fmsg_u32_pair_put(fmsg, "sqn", sq->sqn);
 
-	err = devlink_fmsg_u32_pair_put(fmsg, "cc", sq->cc);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "pc", sq->pc);
-	if (err)
-		return err;
-
-	err = mlx5e_health_sq_put_sw_state(fmsg, sq);
-	if (err)
-		return err;
-
-	err = mlx5e_health_cq_diag_fmsg(&sq->cq, fmsg);
-	if (err)
-		return err;
-
-	return mlx5e_health_eq_diag_fmsg(sq->cq.mcq.eq, fmsg);
+	err = mlx5_core_query_sq_state(priv->mdev, sq->sqn, &state);
+	if (!err)
+		devlink_fmsg_u8_pair_put(fmsg, "HW state", state);
+
+	devlink_fmsg_bool_pair_put(fmsg, "stopped", stopped);
+	devlink_fmsg_u32_pair_put(fmsg, "cc", sq->cc);
+	devlink_fmsg_u32_pair_put(fmsg, "pc", sq->pc);
+	mlx5e_health_sq_put_sw_state(fmsg, sq);
+	mlx5e_health_cq_diag_fmsg(&sq->cq, fmsg);
+	mlx5e_health_eq_diag_fmsg(sq->cq.mcq.eq, fmsg);
 }
 
-static int
+static void
 mlx5e_tx_reporter_build_diagnose_output(struct devlink_fmsg *fmsg,
 					struct mlx5e_txqsq *sq, int tc)
 {
-	int err;
-
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "channel ix", sq->ch_ix);
-	if (err)
-		return err;
-
-	err = mlx5e_tx_reporter_build_diagnose_output_sq_common(fmsg, sq, tc);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_obj_nest_start(fmsg);
+	devlink_fmsg_u32_pair_put(fmsg, "channel ix", sq->ch_ix);
+	mlx5e_tx_reporter_build_diagnose_output_sq_common(fmsg, sq, tc);
+	devlink_fmsg_obj_nest_end(fmsg);
 }
 
-static int
+static void
 mlx5e_tx_reporter_build_diagnose_output_ptpsq(struct devlink_fmsg *fmsg,
 					      struct mlx5e_ptpsq *ptpsq, int tc)
 {
-	int err;
-
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_string_pair_put(fmsg, "channel", "ptp");
-	if (err)
-		return err;
-
-	err = mlx5e_tx_reporter_build_diagnose_output_sq_common(fmsg, &ptpsq->txqsq, tc);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Port TS");
-	if (err)
-		return err;
-
-	err = mlx5e_health_cq_diag_fmsg(&ptpsq->ts_cq, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_obj_nest_start(fmsg);
+	devlink_fmsg_string_pair_put(fmsg, "channel", "ptp");
+	mlx5e_tx_reporter_build_diagnose_output_sq_common(fmsg, &ptpsq->txqsq, tc);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Port TS");
+	mlx5e_health_cq_diag_fmsg(&ptpsq->ts_cq, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	devlink_fmsg_obj_nest_end(fmsg);
 }
 
-static int
+static void
 mlx5e_tx_reporter_diagnose_generic_txqsq(struct devlink_fmsg *fmsg,
 					 struct mlx5e_txqsq *txqsq)
 {
-	u32 sq_stride, sq_sz;
-	bool real_time;
-	int err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SQ");
-	if (err)
-		return err;
-
-	real_time =  mlx5_is_real_time_sq(txqsq->mdev);
-	sq_sz = mlx5_wq_cyc_get_size(&txqsq->wq);
-	sq_stride = MLX5_SEND_WQE_BB;
-
-	err = devlink_fmsg_u64_pair_put(fmsg, "stride size", sq_stride);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_u32_pair_put(fmsg, "size", sq_sz);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_string_pair_put(fmsg, "ts_format", real_time ? "RT" : "FRC");
-	if (err)
-		return err;
-
-	err = mlx5e_health_cq_common_diag_fmsg(&txqsq->cq, fmsg);
-	if (err)
-		return err;
-
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	bool real_time =  mlx5_is_real_time_sq(txqsq->mdev);
+	u32 sq_sz = mlx5_wq_cyc_get_size(&txqsq->wq);
+	u32 sq_stride = MLX5_SEND_WQE_BB;
+
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SQ");
+	devlink_fmsg_u64_pair_put(fmsg, "stride size", sq_stride);
+	devlink_fmsg_u32_pair_put(fmsg, "size", sq_sz);
+	devlink_fmsg_string_pair_put(fmsg, "ts_format", real_time ? "RT" : "FRC");
+	mlx5e_health_cq_common_diag_fmsg(&txqsq->cq, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-static int
+static void
 mlx5e_tx_reporter_diagnose_generic_tx_port_ts(struct devlink_fmsg *fmsg,
 					      struct mlx5e_ptpsq *ptpsq)
 {
-	int err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Port TS");
-	if (err)
-		return err;
-
-	err = mlx5e_health_cq_common_diag_fmsg(&ptpsq->ts_cq, fmsg);
-	if (err)
-		return err;
-
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Port TS");
+	mlx5e_health_cq_common_diag_fmsg(&ptpsq->ts_cq, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
-static int
+static void
 mlx5e_tx_reporter_diagnose_common_config(struct devlink_health_reporter *reporter,
 					 struct devlink_fmsg *fmsg)
 {
@@ -394,39 +295,20 @@ mlx5e_tx_reporter_diagnose_common_config(struct devlink_health_reporter *reporte
 	struct mlx5e_txqsq *generic_sq = priv->txq2sq[0];
 	struct mlx5e_ptp *ptp_ch = priv->channels.ptp;
 	struct mlx5e_ptpsq *generic_ptpsq;
-	int err;
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Common Config");
-	if (err)
-		return err;
-
-	err = mlx5e_tx_reporter_diagnose_generic_txqsq(fmsg, generic_sq);
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "Common Config");
+	mlx5e_tx_reporter_diagnose_generic_txqsq(fmsg, generic_sq);
 
 	if (!ptp_ch || !test_bit(MLX5E_PTP_STATE_TX, ptp_ch->state))
 		goto out;
 
 	generic_ptpsq = &ptp_ch->ptpsq[0];
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "PTP");
-	if (err)
-		return err;
-
-	err = mlx5e_tx_reporter_diagnose_generic_txqsq(fmsg, &generic_ptpsq->txqsq);
-	if (err)
-		return err;
-
-	err = mlx5e_tx_reporter_diagnose_generic_tx_port_ts(fmsg, generic_ptpsq);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "PTP");
+	mlx5e_tx_reporter_diagnose_generic_txqsq(fmsg, &generic_ptpsq->txqsq);
+	mlx5e_tx_reporter_diagnose_generic_tx_port_ts(fmsg, generic_ptpsq);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 out:
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
 static int mlx5e_tx_reporter_diagnose(struct devlink_health_reporter *reporter,
@@ -436,20 +318,15 @@ static int mlx5e_tx_reporter_diagnose(struct devlink_health_reporter *reporter,
 	struct mlx5e_priv *priv = devlink_health_reporter_priv(reporter);
 	struct mlx5e_ptp *ptp_ch = priv->channels.ptp;
 
-	int i, tc, err = 0;
+	int i, tc;
 
 	mutex_lock(&priv->state_lock);
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		goto unlock;
 
-	err = mlx5e_tx_reporter_diagnose_common_config(reporter, fmsg);
-	if (err)
-		goto unlock;
-
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "SQs");
-	if (err)
-		goto unlock;
+	mlx5e_tx_reporter_diagnose_common_config(reporter, fmsg);
+	devlink_fmsg_arr_pair_nest_start(fmsg, "SQs");
 
 	for (i = 0; i < priv->channels.num; i++) {
 		struct mlx5e_channel *c = priv->channels.c[i];
@@ -457,31 +334,23 @@ static int mlx5e_tx_reporter_diagnose(struct devlink_health_reporter *reporter,
 		for (tc = 0; tc < mlx5e_get_dcb_num_tc(&priv->channels.params); tc++) {
 			struct mlx5e_txqsq *sq = &c->sq[tc];
 
-			err = mlx5e_tx_reporter_build_diagnose_output(fmsg, sq, tc);
-			if (err)
-				goto unlock;
+			mlx5e_tx_reporter_build_diagnose_output(fmsg, sq, tc);
 		}
 	}
 
 	if (!ptp_ch || !test_bit(MLX5E_PTP_STATE_TX, ptp_ch->state))
 		goto close_sqs_nest;
 
-	for (tc = 0; tc < mlx5e_get_dcb_num_tc(&priv->channels.params); tc++) {
-		err = mlx5e_tx_reporter_build_diagnose_output_ptpsq(fmsg,
-								    &ptp_ch->ptpsq[tc],
-								    tc);
-		if (err)
-			goto unlock;
-	}
+	for (tc = 0; tc < mlx5e_get_dcb_num_tc(&priv->channels.params); tc++)
+		mlx5e_tx_reporter_build_diagnose_output_ptpsq(fmsg,
+							      &ptp_ch->ptpsq[tc],
+							      tc);
 
 close_sqs_nest:
-	err = devlink_fmsg_arr_pair_nest_end(fmsg);
-	if (err)
-		goto unlock;
-
+	devlink_fmsg_arr_pair_nest_end(fmsg);
 unlock:
 	mutex_unlock(&priv->state_lock);
-	return err;
+	return 0;
 }
 
 static int mlx5e_tx_reporter_dump_sq(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
@@ -489,60 +358,33 @@ static int mlx5e_tx_reporter_dump_sq(struct mlx5e_priv *priv, struct devlink_fms
 {
 	struct mlx5_rsc_key key = {};
 	struct mlx5e_txqsq *sq = ctx;
-	int err;
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		return 0;
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SX Slice");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SX Slice");
 	key.size = PAGE_SIZE;
 	key.rsc = MLX5_SGMT_TYPE_SX_SLICE_ALL;
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SQ");
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "QPC");
-	if (err)
-		return err;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SQ");
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "QPC");
 	key.rsc = MLX5_SGMT_TYPE_FULL_QPC;
 	key.index1 = sq->sqn;
 	key.num_of_obj1 = 1;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "send_buff");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "send_buff");
 	key.rsc = MLX5_SGMT_TYPE_SND_BUFF;
 	key.num_of_obj2 = MLX5_RSC_DUMP_ALL;
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 
-	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	return 0;
 }
 
 static int mlx5e_tx_reporter_timeout_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,
@@ -567,28 +409,17 @@ static int mlx5e_tx_reporter_dump_all_sqs(struct mlx5e_priv *priv,
 {
 	struct mlx5e_ptp *ptp_ch = priv->channels.ptp;
 	struct mlx5_rsc_key key = {};
-	int i, tc, err;
+	int i, tc;
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		return 0;
 
-	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SX Slice");
-	if (err)
-		return err;
-
+	mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SX Slice");
 	key.size = PAGE_SIZE;
 	key.rsc = MLX5_SGMT_TYPE_SX_SLICE_ALL;
-	err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
-	if (err)
-		return err;
-
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "SQs");
-	if (err)
-		return err;
+	mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg);
+	mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	devlink_fmsg_arr_pair_nest_start(fmsg, "SQs");
 
 	for (i = 0; i < priv->channels.num; i++) {
 		struct mlx5e_channel *c = priv->channels.c[i];
@@ -596,9 +427,7 @@ static int mlx5e_tx_reporter_dump_all_sqs(struct mlx5e_priv *priv,
 		for (tc = 0; tc < mlx5e_get_dcb_num_tc(&priv->channels.params); tc++) {
 			struct mlx5e_txqsq *sq = &c->sq[tc];
 
-			err = mlx5e_health_queue_dump(priv, fmsg, sq->sqn, "SQ");
-			if (err)
-				return err;
+			mlx5e_health_queue_dump(priv, fmsg, sq->sqn, "SQ");
 		}
 	}
 
@@ -606,13 +435,12 @@ static int mlx5e_tx_reporter_dump_all_sqs(struct mlx5e_priv *priv,
 		for (tc = 0; tc < mlx5e_get_dcb_num_tc(&priv->channels.params); tc++) {
 			struct mlx5e_txqsq *sq = &ptp_ch->ptpsq[tc].txqsq;
 
-			err = mlx5e_health_queue_dump(priv, fmsg, sq->sqn, "PTP SQ");
-			if (err)
-				return err;
+			mlx5e_health_queue_dump(priv, fmsg, sq->sqn, "PTP SQ");
 		}
 	}
 
-	return devlink_fmsg_arr_pair_nest_end(fmsg);
+	devlink_fmsg_arr_pair_nest_end(fmsg);
+	return 0;
 }
 
 static int mlx5e_tx_reporter_dump_from_ctx(struct mlx5e_priv *priv,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.c
index b915fb29dd..7b8ff7a710 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.c
@@ -9,7 +9,7 @@ void mlx5e_rss_params_indir_init_uniform(struct mlx5e_rss_params_indir *indir,
 {
 	unsigned int i;
 
-	for (i = 0; i < MLX5E_INDIR_RQT_SIZE; i++)
+	for (i = 0; i < indir->actual_table_size; i++)
 		indir->table[i] = i % num_channels;
 }
 
@@ -45,9 +45,9 @@ static int mlx5e_rqt_init(struct mlx5e_rqt *rqt, struct mlx5_core_dev *mdev,
 }
 
 int mlx5e_rqt_init_direct(struct mlx5e_rqt *rqt, struct mlx5_core_dev *mdev,
-			  bool indir_enabled, u32 init_rqn)
+			  bool indir_enabled, u32 init_rqn, u32 indir_table_size)
 {
-	u16 max_size = indir_enabled ? MLX5E_INDIR_RQT_SIZE : 1;
+	u16 max_size = indir_enabled ? indir_table_size : 1;
 
 	return mlx5e_rqt_init(rqt, mdev, max_size, &init_rqn, 1);
 }
@@ -68,11 +68,11 @@ static int mlx5e_calc_indir_rqns(u32 *rss_rqns, u32 *rqns, unsigned int num_rqns
 {
 	unsigned int i;
 
-	for (i = 0; i < MLX5E_INDIR_RQT_SIZE; i++) {
+	for (i = 0; i < indir->actual_table_size; i++) {
 		unsigned int ix = i;
 
 		if (hfunc == ETH_RSS_HASH_XOR)
-			ix = mlx5e_bits_invert(ix, ilog2(MLX5E_INDIR_RQT_SIZE));
+			ix = mlx5e_bits_invert(ix, ilog2(indir->actual_table_size));
 
 		ix = indir->table[ix];
 
@@ -94,7 +94,7 @@ int mlx5e_rqt_init_indir(struct mlx5e_rqt *rqt, struct mlx5_core_dev *mdev,
 	u32 *rss_rqns;
 	int err;
 
-	rss_rqns = kvmalloc_array(MLX5E_INDIR_RQT_SIZE, sizeof(*rss_rqns), GFP_KERNEL);
+	rss_rqns = kvmalloc_array(indir->actual_table_size, sizeof(*rss_rqns), GFP_KERNEL);
 	if (!rss_rqns)
 		return -ENOMEM;
 
@@ -102,13 +102,25 @@ int mlx5e_rqt_init_indir(struct mlx5e_rqt *rqt, struct mlx5_core_dev *mdev,
 	if (err)
 		goto out;
 
-	err = mlx5e_rqt_init(rqt, mdev, MLX5E_INDIR_RQT_SIZE, rss_rqns, MLX5E_INDIR_RQT_SIZE);
+	err = mlx5e_rqt_init(rqt, mdev, indir->max_table_size, rss_rqns,
+			     indir->actual_table_size);
 
 out:
 	kvfree(rss_rqns);
 	return err;
 }
 
+#define MLX5E_UNIFORM_SPREAD_RQT_FACTOR 2
+
+u32 mlx5e_rqt_size(struct mlx5_core_dev *mdev, unsigned int num_channels)
+{
+	u32 rqt_size = max_t(u32, MLX5E_INDIR_MIN_RQT_SIZE,
+			     roundup_pow_of_two(num_channels * MLX5E_UNIFORM_SPREAD_RQT_FACTOR));
+	u32 max_cap_rqt_size = 1 << MLX5_CAP_GEN(mdev, log_max_rqt_size);
+
+	return min_t(u32, rqt_size, max_cap_rqt_size);
+}
+
 void mlx5e_rqt_destroy(struct mlx5e_rqt *rqt)
 {
 	mlx5_core_destroy_rqt(rqt->mdev, rqt->rqtn);
@@ -151,10 +163,10 @@ int mlx5e_rqt_redirect_indir(struct mlx5e_rqt *rqt, u32 *rqns, unsigned int num_
 	u32 *rss_rqns;
 	int err;
 
-	if (WARN_ON(rqt->size != MLX5E_INDIR_RQT_SIZE))
+	if (WARN_ON(rqt->size != indir->max_table_size))
 		return -EINVAL;
 
-	rss_rqns = kvmalloc_array(MLX5E_INDIR_RQT_SIZE, sizeof(*rss_rqns), GFP_KERNEL);
+	rss_rqns = kvmalloc_array(indir->actual_table_size, sizeof(*rss_rqns), GFP_KERNEL);
 	if (!rss_rqns)
 		return -ENOMEM;
 
@@ -162,7 +174,7 @@ int mlx5e_rqt_redirect_indir(struct mlx5e_rqt *rqt, u32 *rqns, unsigned int num_
 	if (err)
 		goto out;
 
-	err = mlx5e_rqt_redirect(rqt, rss_rqns, MLX5E_INDIR_RQT_SIZE);
+	err = mlx5e_rqt_redirect(rqt, rss_rqns, indir->actual_table_size);
 
 out:
 	kvfree(rss_rqns);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.h b/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.h
index 60c985a12f..77fba3ebd1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rqt.h
@@ -6,12 +6,14 @@
 
 #include <linux/kernel.h>
 
-#define MLX5E_INDIR_RQT_SIZE (1 << 8)
+#define MLX5E_INDIR_MIN_RQT_SIZE (BIT(8))
 
 struct mlx5_core_dev;
 
 struct mlx5e_rss_params_indir {
-	u32 table[MLX5E_INDIR_RQT_SIZE];
+	u32 *table;
+	u32 actual_table_size;
+	u32 max_table_size;
 };
 
 void mlx5e_rss_params_indir_init_uniform(struct mlx5e_rss_params_indir *indir,
@@ -24,7 +26,7 @@ struct mlx5e_rqt {
 };
 
 int mlx5e_rqt_init_direct(struct mlx5e_rqt *rqt, struct mlx5_core_dev *mdev,
-			  bool indir_enabled, u32 init_rqn);
+			  bool indir_enabled, u32 init_rqn, u32 indir_table_size);
 int mlx5e_rqt_init_indir(struct mlx5e_rqt *rqt, struct mlx5_core_dev *mdev,
 			 u32 *rqns, unsigned int num_rqns,
 			 u8 hfunc, struct mlx5e_rss_params_indir *indir);
@@ -35,6 +37,7 @@ static inline u32 mlx5e_rqt_get_rqtn(struct mlx5e_rqt *rqt)
 	return rqt->rqtn;
 }
 
+u32 mlx5e_rqt_size(struct mlx5_core_dev *mdev, unsigned int num_channels);
 int mlx5e_rqt_redirect_direct(struct mlx5e_rqt *rqt, u32 rqn);
 int mlx5e_rqt_redirect_indir(struct mlx5e_rqt *rqt, u32 *rqns, unsigned int num_rqns,
 			     u8 hfunc, struct mlx5e_rss_params_indir *indir);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c
index 7f93426b88..c1545a2e8d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c
@@ -81,14 +81,75 @@ struct mlx5e_rss {
 	refcount_t refcnt;
 };
 
-struct mlx5e_rss *mlx5e_rss_alloc(void)
+void mlx5e_rss_params_indir_modify_actual_size(struct mlx5e_rss *rss, u32 num_channels)
 {
-	return kvzalloc(sizeof(struct mlx5e_rss), GFP_KERNEL);
+	rss->indir.actual_table_size = mlx5e_rqt_size(rss->mdev, num_channels);
 }
 
-void mlx5e_rss_free(struct mlx5e_rss *rss)
+int mlx5e_rss_params_indir_init(struct mlx5e_rss_params_indir *indir, struct mlx5_core_dev *mdev,
+				u32 actual_table_size, u32 max_table_size)
 {
+	indir->table = kvmalloc_array(max_table_size, sizeof(*indir->table), GFP_KERNEL);
+	if (!indir->table)
+		return -ENOMEM;
+
+	indir->max_table_size = max_table_size;
+	indir->actual_table_size = actual_table_size;
+
+	return 0;
+}
+
+void mlx5e_rss_params_indir_cleanup(struct mlx5e_rss_params_indir *indir)
+{
+	kvfree(indir->table);
+}
+
+static int mlx5e_rss_copy(struct mlx5e_rss *to, const struct mlx5e_rss *from)
+{
+	u32 *dst_indir_table;
+
+	if (to->indir.actual_table_size != from->indir.actual_table_size ||
+	    to->indir.max_table_size != from->indir.max_table_size) {
+		mlx5e_rss_warn(to->mdev,
+			       "Failed to copy RSS due to size mismatch, src (actual %u, max %u) != dst (actual %u, max %u)\n",
+			       from->indir.actual_table_size, from->indir.max_table_size,
+			       to->indir.actual_table_size, to->indir.max_table_size);
+		return -EINVAL;
+	}
+
+	dst_indir_table = to->indir.table;
+	*to = *from;
+	to->indir.table = dst_indir_table;
+	memcpy(to->indir.table, from->indir.table,
+	       from->indir.actual_table_size * sizeof(*from->indir.table));
+	return 0;
+}
+
+static struct mlx5e_rss *mlx5e_rss_init_copy(const struct mlx5e_rss *from)
+{
+	struct mlx5e_rss *rss;
+	int err;
+
+	rss = kvzalloc(sizeof(*rss), GFP_KERNEL);
+	if (!rss)
+		return ERR_PTR(-ENOMEM);
+
+	err = mlx5e_rss_params_indir_init(&rss->indir, from->mdev, from->indir.actual_table_size,
+					  from->indir.max_table_size);
+	if (err)
+		goto err_free_rss;
+
+	err = mlx5e_rss_copy(rss, from);
+	if (err)
+		goto err_free_indir;
+
+	return rss;
+
+err_free_indir:
+	mlx5e_rss_params_indir_cleanup(&rss->indir);
+err_free_rss:
 	kvfree(rss);
+	return ERR_PTR(err);
 }
 
 static void mlx5e_rss_params_init(struct mlx5e_rss *rss)
@@ -282,28 +343,43 @@ static int mlx5e_rss_update_tirs(struct mlx5e_rss *rss)
 	return retval;
 }
 
-int mlx5e_rss_init_no_tirs(struct mlx5e_rss *rss, struct mlx5_core_dev *mdev,
-			   bool inner_ft_support, u32 drop_rqn)
+static int mlx5e_rss_init_no_tirs(struct mlx5e_rss *rss)
 {
-	rss->mdev = mdev;
-	rss->inner_ft_support = inner_ft_support;
-	rss->drop_rqn = drop_rqn;
-
 	mlx5e_rss_params_init(rss);
 	refcount_set(&rss->refcnt, 1);
 
-	return mlx5e_rqt_init_direct(&rss->rqt, mdev, true, drop_rqn);
+	return mlx5e_rqt_init_direct(&rss->rqt, rss->mdev, true,
+				     rss->drop_rqn, rss->indir.max_table_size);
 }
 
-int mlx5e_rss_init(struct mlx5e_rss *rss, struct mlx5_core_dev *mdev,
-		   bool inner_ft_support, u32 drop_rqn,
-		   const struct mlx5e_packet_merge_param *init_pkt_merge_param)
+struct mlx5e_rss *mlx5e_rss_init(struct mlx5_core_dev *mdev, bool inner_ft_support, u32 drop_rqn,
+				 const struct mlx5e_packet_merge_param *init_pkt_merge_param,
+				 enum mlx5e_rss_init_type type, unsigned int nch,
+				 unsigned int max_nch)
 {
+	struct mlx5e_rss *rss;
 	int err;
 
-	err = mlx5e_rss_init_no_tirs(rss, mdev, inner_ft_support, drop_rqn);
+	rss = kvzalloc(sizeof(*rss), GFP_KERNEL);
+	if (!rss)
+		return ERR_PTR(-ENOMEM);
+
+	err = mlx5e_rss_params_indir_init(&rss->indir, mdev,
+					  mlx5e_rqt_size(mdev, nch),
+					  mlx5e_rqt_size(mdev, max_nch));
+	if (err)
+		goto err_free_rss;
+
+	rss->mdev = mdev;
+	rss->inner_ft_support = inner_ft_support;
+	rss->drop_rqn = drop_rqn;
+
+	err = mlx5e_rss_init_no_tirs(rss);
 	if (err)
-		goto err_out;
+		goto err_free_indir;
+
+	if (type == MLX5E_RSS_INIT_NO_TIRS)
+		goto out;
 
 	err = mlx5e_rss_create_tirs(rss, init_pkt_merge_param, false);
 	if (err)
@@ -315,14 +391,18 @@ int mlx5e_rss_init(struct mlx5e_rss *rss, struct mlx5_core_dev *mdev,
 			goto err_destroy_tirs;
 	}
 
-	return 0;
+out:
+	return rss;
 
 err_destroy_tirs:
 	mlx5e_rss_destroy_tirs(rss, false);
 err_destroy_rqt:
 	mlx5e_rqt_destroy(&rss->rqt);
-err_out:
-	return err;
+err_free_indir:
+	mlx5e_rss_params_indir_cleanup(&rss->indir);
+err_free_rss:
+	kvfree(rss);
+	return ERR_PTR(err);
 }
 
 int mlx5e_rss_cleanup(struct mlx5e_rss *rss)
@@ -336,6 +416,8 @@ int mlx5e_rss_cleanup(struct mlx5e_rss *rss)
 		mlx5e_rss_destroy_tirs(rss, true);
 
 	mlx5e_rqt_destroy(&rss->rqt);
+	mlx5e_rss_params_indir_cleanup(&rss->indir);
+	kvfree(rss);
 
 	return 0;
 }
@@ -470,11 +552,9 @@ inner_tir:
 
 int mlx5e_rss_get_rxfh(struct mlx5e_rss *rss, u32 *indir, u8 *key, u8 *hfunc)
 {
-	unsigned int i;
-
 	if (indir)
-		for (i = 0; i < MLX5E_INDIR_RQT_SIZE; i++)
-			indir[i] = rss->indir.table[i];
+		memcpy(indir, rss->indir.table,
+		       rss->indir.actual_table_size * sizeof(*rss->indir.table));
 
 	if (key)
 		memcpy(key, rss->hash.toeplitz_hash_key,
@@ -495,11 +575,9 @@ int mlx5e_rss_set_rxfh(struct mlx5e_rss *rss, const u32 *indir,
 	struct mlx5e_rss *old_rss;
 	int err = 0;
 
-	old_rss = mlx5e_rss_alloc();
-	if (!old_rss)
-		return -ENOMEM;
-
-	*old_rss = *rss;
+	old_rss = mlx5e_rss_init_copy(rss);
+	if (IS_ERR(old_rss))
+		return PTR_ERR(old_rss);
 
 	if (hfunc && *hfunc != rss->hash.hfunc) {
 		switch (*hfunc) {
@@ -523,18 +601,16 @@ int mlx5e_rss_set_rxfh(struct mlx5e_rss *rss, const u32 *indir,
 	}
 
 	if (indir) {
-		unsigned int i;
-
 		changed_indir = true;
 
-		for (i = 0; i < MLX5E_INDIR_RQT_SIZE; i++)
-			rss->indir.table[i] = indir[i];
+		memcpy(rss->indir.table, indir,
+		       rss->indir.actual_table_size * sizeof(*rss->indir.table));
 	}
 
 	if (changed_indir && rss->enabled) {
 		err = mlx5e_rss_apply(rss, rqns, num_rqns);
 		if (err) {
-			*rss = *old_rss;
+			mlx5e_rss_copy(rss, old_rss);
 			goto out;
 		}
 	}
@@ -543,7 +619,9 @@ int mlx5e_rss_set_rxfh(struct mlx5e_rss *rss, const u32 *indir,
 		mlx5e_rss_update_tirs(rss);
 
 out:
-	mlx5e_rss_free(old_rss);
+	mlx5e_rss_params_indir_cleanup(&old_rss->indir);
+	kvfree(old_rss);
+
 	return err;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h
index c6b2164163..d1d0bc350e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h
@@ -8,18 +8,24 @@
 #include "tir.h"
 #include "fs.h"
 
+enum mlx5e_rss_init_type {
+	MLX5E_RSS_INIT_NO_TIRS = 0,
+	MLX5E_RSS_INIT_TIRS
+};
+
 struct mlx5e_rss_params_traffic_type
 mlx5e_rss_get_default_tt_config(enum mlx5_traffic_types tt);
 
 struct mlx5e_rss;
 
-struct mlx5e_rss *mlx5e_rss_alloc(void);
-void mlx5e_rss_free(struct mlx5e_rss *rss);
-int mlx5e_rss_init(struct mlx5e_rss *rss, struct mlx5_core_dev *mdev,
-		   bool inner_ft_support, u32 drop_rqn,
-		   const struct mlx5e_packet_merge_param *init_pkt_merge_param);
-int mlx5e_rss_init_no_tirs(struct mlx5e_rss *rss, struct mlx5_core_dev *mdev,
-			   bool inner_ft_support, u32 drop_rqn);
+int mlx5e_rss_params_indir_init(struct mlx5e_rss_params_indir *indir, struct mlx5_core_dev *mdev,
+				u32 actual_table_size, u32 max_table_size);
+void mlx5e_rss_params_indir_cleanup(struct mlx5e_rss_params_indir *indir);
+void mlx5e_rss_params_indir_modify_actual_size(struct mlx5e_rss *rss, u32 num_channels);
+struct mlx5e_rss *mlx5e_rss_init(struct mlx5_core_dev *mdev, bool inner_ft_support, u32 drop_rqn,
+				 const struct mlx5e_packet_merge_param *init_pkt_merge_param,
+				 enum mlx5e_rss_init_type type, unsigned int nch,
+				 unsigned int max_nch);
 int mlx5e_rss_cleanup(struct mlx5e_rss *rss);
 
 void mlx5e_rss_refcnt_inc(struct mlx5e_rss *rss);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c
index 56e6b8c750..b23e224e37 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c
@@ -18,7 +18,7 @@ struct mlx5e_rx_res {
 
 	struct mlx5e_rss *rss[MLX5E_MAX_NUM_RSS];
 	bool rss_active;
-	u32 rss_rqns[MLX5E_INDIR_RQT_SIZE];
+	u32 *rss_rqns;
 	unsigned int rss_nch;
 
 	struct {
@@ -34,41 +34,42 @@ struct mlx5e_rx_res {
 
 /* API for rx_res_rss_* */
 
+void mlx5e_rx_res_rss_update_num_channels(struct mlx5e_rx_res *res, u32 nch)
+{
+	int i;
+
+	for (i = 0; i < MLX5E_MAX_NUM_RSS; i++) {
+		if (res->rss[i])
+			mlx5e_rss_params_indir_modify_actual_size(res->rss[i], nch);
+	}
+}
+
 static int mlx5e_rx_res_rss_init_def(struct mlx5e_rx_res *res,
 				     unsigned int init_nch)
 {
 	bool inner_ft_support = res->features & MLX5E_RX_RES_FEATURE_INNER_FT;
 	struct mlx5e_rss *rss;
-	int err;
 
 	if (WARN_ON(res->rss[0]))
 		return -EINVAL;
 
-	rss = mlx5e_rss_alloc();
-	if (!rss)
-		return -ENOMEM;
-
-	err = mlx5e_rss_init(rss, res->mdev, inner_ft_support, res->drop_rqn,
-			     &res->pkt_merge_param);
-	if (err)
-		goto err_rss_free;
+	rss = mlx5e_rss_init(res->mdev, inner_ft_support, res->drop_rqn,
+			     &res->pkt_merge_param, MLX5E_RSS_INIT_TIRS, init_nch, res->max_nch);
+	if (IS_ERR(rss))
+		return PTR_ERR(rss);
 
 	mlx5e_rss_set_indir_uniform(rss, init_nch);
 
 	res->rss[0] = rss;
 
 	return 0;
-
-err_rss_free:
-	mlx5e_rss_free(rss);
-	return err;
 }
 
 int mlx5e_rx_res_rss_init(struct mlx5e_rx_res *res, u32 *rss_idx, unsigned int init_nch)
 {
 	bool inner_ft_support = res->features & MLX5E_RX_RES_FEATURE_INNER_FT;
 	struct mlx5e_rss *rss;
-	int err, i;
+	int i;
 
 	for (i = 1; i < MLX5E_MAX_NUM_RSS; i++)
 		if (!res->rss[i])
@@ -77,13 +78,11 @@ int mlx5e_rx_res_rss_init(struct mlx5e_rx_res *res, u32 *rss_idx, unsigned int i
 	if (i == MLX5E_MAX_NUM_RSS)
 		return -ENOSPC;
 
-	rss = mlx5e_rss_alloc();
-	if (!rss)
-		return -ENOMEM;
-
-	err = mlx5e_rss_init_no_tirs(rss, res->mdev, inner_ft_support, res->drop_rqn);
-	if (err)
-		goto err_rss_free;
+	rss = mlx5e_rss_init(res->mdev, inner_ft_support, res->drop_rqn,
+			     &res->pkt_merge_param, MLX5E_RSS_INIT_NO_TIRS, init_nch,
+			     res->max_nch);
+	if (IS_ERR(rss))
+		return PTR_ERR(rss);
 
 	mlx5e_rss_set_indir_uniform(rss, init_nch);
 	if (res->rss_active)
@@ -93,10 +92,6 @@ int mlx5e_rx_res_rss_init(struct mlx5e_rx_res *res, u32 *rss_idx, unsigned int i
 	*rss_idx = i;
 
 	return 0;
-
-err_rss_free:
-	mlx5e_rss_free(rss);
-	return err;
 }
 
 static int __mlx5e_rx_res_rss_destroy(struct mlx5e_rx_res *res, u32 rss_idx)
@@ -108,7 +103,6 @@ static int __mlx5e_rx_res_rss_destroy(struct mlx5e_rx_res *res, u32 rss_idx)
 	if (err)
 		return err;
 
-	mlx5e_rss_free(rss);
 	res->rss[rss_idx] = NULL;
 
 	return 0;
@@ -284,9 +278,27 @@ struct mlx5e_rss *mlx5e_rx_res_rss_get(struct mlx5e_rx_res *res, u32 rss_idx)
 
 /* End of API rx_res_rss_* */
 
-struct mlx5e_rx_res *mlx5e_rx_res_alloc(void)
+static void mlx5e_rx_res_free(struct mlx5e_rx_res *res)
 {
-	return kvzalloc(sizeof(struct mlx5e_rx_res), GFP_KERNEL);
+	kvfree(res->rss_rqns);
+	kvfree(res);
+}
+
+static struct mlx5e_rx_res *mlx5e_rx_res_alloc(struct mlx5_core_dev *mdev, unsigned int max_nch)
+{
+	struct mlx5e_rx_res *rx_res;
+
+	rx_res = kvzalloc(sizeof(*rx_res), GFP_KERNEL);
+	if (!rx_res)
+		return NULL;
+
+	rx_res->rss_rqns = kvcalloc(max_nch, sizeof(*rx_res->rss_rqns), GFP_KERNEL);
+	if (!rx_res->rss_rqns) {
+		kvfree(rx_res);
+		return NULL;
+	}
+
+	return rx_res;
 }
 
 static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res)
@@ -308,7 +320,8 @@ static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res)
 
 	for (ix = 0; ix < res->max_nch; ix++) {
 		err = mlx5e_rqt_init_direct(&res->channels[ix].direct_rqt,
-					    res->mdev, false, res->drop_rqn);
+					    res->mdev, false, res->drop_rqn,
+					    mlx5e_rqt_size(res->mdev, res->max_nch));
 		if (err) {
 			mlx5_core_warn(res->mdev, "Failed to create a direct RQT: err = %d, ix = %u\n",
 				       err, ix);
@@ -362,7 +375,8 @@ static int mlx5e_rx_res_ptp_init(struct mlx5e_rx_res *res)
 	if (!builder)
 		return -ENOMEM;
 
-	err = mlx5e_rqt_init_direct(&res->ptp.rqt, res->mdev, false, res->drop_rqn);
+	err = mlx5e_rqt_init_direct(&res->ptp.rqt, res->mdev, false, res->drop_rqn,
+				    mlx5e_rqt_size(res->mdev, res->max_nch));
 	if (err)
 		goto out;
 
@@ -404,13 +418,19 @@ static void mlx5e_rx_res_ptp_destroy(struct mlx5e_rx_res *res)
 	mlx5e_rqt_destroy(&res->ptp.rqt);
 }
 
-int mlx5e_rx_res_init(struct mlx5e_rx_res *res, struct mlx5_core_dev *mdev,
-		      enum mlx5e_rx_res_features features, unsigned int max_nch,
-		      u32 drop_rqn, const struct mlx5e_packet_merge_param *init_pkt_merge_param,
-		      unsigned int init_nch)
+struct mlx5e_rx_res *
+mlx5e_rx_res_create(struct mlx5_core_dev *mdev, enum mlx5e_rx_res_features features,
+		    unsigned int max_nch, u32 drop_rqn,
+		    const struct mlx5e_packet_merge_param *init_pkt_merge_param,
+		    unsigned int init_nch)
 {
+	struct mlx5e_rx_res *res;
 	int err;
 
+	res = mlx5e_rx_res_alloc(mdev, max_nch);
+	if (!res)
+		return ERR_PTR(-ENOMEM);
+
 	res->mdev = mdev;
 	res->features = features;
 	res->max_nch = max_nch;
@@ -421,7 +441,7 @@ int mlx5e_rx_res_init(struct mlx5e_rx_res *res, struct mlx5_core_dev *mdev,
 
 	err = mlx5e_rx_res_rss_init_def(res, init_nch);
 	if (err)
-		goto err_out;
+		goto err_rx_res_free;
 
 	err = mlx5e_rx_res_channels_init(res);
 	if (err)
@@ -431,14 +451,15 @@ int mlx5e_rx_res_init(struct mlx5e_rx_res *res, struct mlx5_core_dev *mdev,
 	if (err)
 		goto err_channels_destroy;
 
-	return 0;
+	return res;
 
 err_channels_destroy:
 	mlx5e_rx_res_channels_destroy(res);
 err_rss_destroy:
 	__mlx5e_rx_res_rss_destroy(res, 0);
-err_out:
-	return err;
+err_rx_res_free:
+	mlx5e_rx_res_free(res);
+	return ERR_PTR(err);
 }
 
 void mlx5e_rx_res_destroy(struct mlx5e_rx_res *res)
@@ -446,11 +467,7 @@ void mlx5e_rx_res_destroy(struct mlx5e_rx_res *res)
 	mlx5e_rx_res_ptp_destroy(res);
 	mlx5e_rx_res_channels_destroy(res);
 	mlx5e_rx_res_rss_destroy_all(res);
-}
-
-void mlx5e_rx_res_free(struct mlx5e_rx_res *res)
-{
-	kvfree(res);
+	mlx5e_rx_res_free(res);
 }
 
 u32 mlx5e_rx_res_get_tirn_direct(struct mlx5e_rx_res *res, unsigned int ix)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h
index 580fe8bc3c..82aaba8a82 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h
@@ -21,13 +21,12 @@ enum mlx5e_rx_res_features {
 };
 
 /* Setup */
-struct mlx5e_rx_res *mlx5e_rx_res_alloc(void);
-int mlx5e_rx_res_init(struct mlx5e_rx_res *res, struct mlx5_core_dev *mdev,
-		      enum mlx5e_rx_res_features features, unsigned int max_nch,
-		      u32 drop_rqn, const struct mlx5e_packet_merge_param *init_pkt_merge_param,
-		      unsigned int init_nch);
+struct mlx5e_rx_res *
+mlx5e_rx_res_create(struct mlx5_core_dev *mdev, enum mlx5e_rx_res_features features,
+		    unsigned int max_nch, u32 drop_rqn,
+		    const struct mlx5e_packet_merge_param *init_pkt_merge_param,
+		    unsigned int init_nch);
 void mlx5e_rx_res_destroy(struct mlx5e_rx_res *res);
-void mlx5e_rx_res_free(struct mlx5e_rx_res *res);
 
 /* TIRN getters for flow steering */
 u32 mlx5e_rx_res_get_tirn_direct(struct mlx5e_rx_res *res, unsigned int ix);
@@ -60,6 +59,7 @@ int mlx5e_rx_res_rss_destroy(struct mlx5e_rx_res *res, u32 rss_idx);
 int mlx5e_rx_res_rss_cnt(struct mlx5e_rx_res *res);
 int mlx5e_rx_res_rss_index(struct mlx5e_rx_res *res, struct mlx5e_rss *rss);
 struct mlx5e_rss *mlx5e_rx_res_rss_get(struct mlx5e_rx_res *res, u32 rss_idx);
+void mlx5e_rx_res_rss_update_num_channels(struct mlx5e_rx_res *res, u32 nch);
 
 /* Workaround for hairpin */
 struct mlx5e_rss_params_hash mlx5e_rx_res_get_current_hash(struct mlx5e_rx_res *res);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
index 00a04fdd75..8dfb57f712 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
@@ -302,6 +302,7 @@ int mlx5e_tc_tun_create_header_ipv4(struct mlx5e_priv *priv,
 
 	e->encap_size = ipv4_encap_size;
 	e->encap_header = encap_header;
+	encap_header = NULL;
 
 	if (!(nud_state & NUD_VALID)) {
 		neigh_event_send(attr.n, NULL);
@@ -313,8 +314,8 @@ int mlx5e_tc_tun_create_header_ipv4(struct mlx5e_priv *priv,
 
 	memset(&reformat_params, 0, sizeof(reformat_params));
 	reformat_params.type = e->reformat_type;
-	reformat_params.size = ipv4_encap_size;
-	reformat_params.data = encap_header;
+	reformat_params.size = e->encap_size;
+	reformat_params.data = e->encap_header;
 	e->pkt_reformat = mlx5_packet_reformat_alloc(priv->mdev, &reformat_params,
 						     MLX5_FLOW_NAMESPACE_FDB);
 	if (IS_ERR(e->pkt_reformat)) {
@@ -407,6 +408,7 @@ int mlx5e_tc_tun_update_header_ipv4(struct mlx5e_priv *priv,
 	e->encap_size = ipv4_encap_size;
 	kfree(e->encap_header);
 	e->encap_header = encap_header;
+	encap_header = NULL;
 
 	if (!(nud_state & NUD_VALID)) {
 		neigh_event_send(attr.n, NULL);
@@ -418,8 +420,8 @@ int mlx5e_tc_tun_update_header_ipv4(struct mlx5e_priv *priv,
 
 	memset(&reformat_params, 0, sizeof(reformat_params));
 	reformat_params.type = e->reformat_type;
-	reformat_params.size = ipv4_encap_size;
-	reformat_params.data = encap_header;
+	reformat_params.size = e->encap_size;
+	reformat_params.data = e->encap_header;
 	e->pkt_reformat = mlx5_packet_reformat_alloc(priv->mdev, &reformat_params,
 						     MLX5_FLOW_NAMESPACE_FDB);
 	if (IS_ERR(e->pkt_reformat)) {
@@ -570,6 +572,7 @@ int mlx5e_tc_tun_create_header_ipv6(struct mlx5e_priv *priv,
 
 	e->encap_size = ipv6_encap_size;
 	e->encap_header = encap_header;
+	encap_header = NULL;
 
 	if (!(nud_state & NUD_VALID)) {
 		neigh_event_send(attr.n, NULL);
@@ -581,8 +584,8 @@ int mlx5e_tc_tun_create_header_ipv6(struct mlx5e_priv *priv,
 
 	memset(&reformat_params, 0, sizeof(reformat_params));
 	reformat_params.type = e->reformat_type;
-	reformat_params.size = ipv6_encap_size;
-	reformat_params.data = encap_header;
+	reformat_params.size = e->encap_size;
+	reformat_params.data = e->encap_header;
 	e->pkt_reformat = mlx5_packet_reformat_alloc(priv->mdev, &reformat_params,
 						     MLX5_FLOW_NAMESPACE_FDB);
 	if (IS_ERR(e->pkt_reformat)) {
@@ -674,6 +677,7 @@ int mlx5e_tc_tun_update_header_ipv6(struct mlx5e_priv *priv,
 	e->encap_size = ipv6_encap_size;
 	kfree(e->encap_header);
 	e->encap_header = encap_header;
+	encap_header = NULL;
 
 	if (!(nud_state & NUD_VALID)) {
 		neigh_event_send(attr.n, NULL);
@@ -685,8 +689,8 @@ int mlx5e_tc_tun_update_header_ipv6(struct mlx5e_priv *priv,
 
 	memset(&reformat_params, 0, sizeof(reformat_params));
 	reformat_params.type = e->reformat_type;
-	reformat_params.size = ipv6_encap_size;
-	reformat_params.data = encap_header;
+	reformat_params.size = e->encap_size;
+	reformat_params.data = e->encap_header;
 	e->pkt_reformat = mlx5_packet_reformat_alloc(priv->mdev, &reformat_params,
 						     MLX5_FLOW_NAMESPACE_FDB);
 	if (IS_ERR(e->pkt_reformat)) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index b723ff5e52..13c7ed1bb3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -895,7 +895,7 @@ void mlx5e_xdp_rx_poll_complete(struct mlx5e_rq *rq)
 	mlx5e_xmit_xdp_doorbell(xdpsq);
 
 	if (test_bit(MLX5E_RQ_FLAG_XDP_REDIRECT, rq->flags)) {
-		xdp_do_flush_map();
+		xdp_do_flush();
 		__clear_bit(MLX5E_RQ_FLAG_XDP_REDIRECT, rq->flags);
 	}
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index e2ffc572de..05612d9c60 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -56,7 +56,7 @@ static struct mlx5e_ipsec_pol_entry *to_ipsec_pol_entry(struct xfrm_policy *x)
 	return (struct mlx5e_ipsec_pol_entry *)x->xdo.offload_handle;
 }
 
-static void mlx5e_ipsec_handle_tx_limit(struct work_struct *_work)
+static void mlx5e_ipsec_handle_sw_limits(struct work_struct *_work)
 {
 	struct mlx5e_ipsec_dwork *dwork =
 		container_of(_work, struct mlx5e_ipsec_dwork, dwork.work);
@@ -520,9 +520,15 @@ static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
 			return -EINVAL;
 		}
 
-		if (x->lft.hard_byte_limit != XFRM_INF ||
-		    x->lft.soft_byte_limit != XFRM_INF) {
-			NL_SET_ERR_MSG_MOD(extack, "Device doesn't support limits in bytes");
+		if (x->lft.soft_byte_limit >= x->lft.hard_byte_limit &&
+		    x->lft.hard_byte_limit != XFRM_INF) {
+			/* XFRM stack doesn't prevent such configuration :(. */
+			NL_SET_ERR_MSG_MOD(extack, "Hard byte limit must be greater than soft one");
+			return -EINVAL;
+		}
+
+		if (!x->lft.soft_byte_limit || !x->lft.hard_byte_limit) {
+			NL_SET_ERR_MSG_MOD(extack, "Soft/hard byte limits can't be 0");
 			return -EINVAL;
 		}
 
@@ -658,11 +664,10 @@ static int mlx5e_ipsec_create_dwork(struct mlx5e_ipsec_sa_entry *sa_entry)
 	if (x->xso.type != XFRM_DEV_OFFLOAD_PACKET)
 		return 0;
 
-	if (x->xso.dir != XFRM_DEV_OFFLOAD_OUT)
-		return 0;
-
 	if (x->lft.soft_packet_limit == XFRM_INF &&
-	    x->lft.hard_packet_limit == XFRM_INF)
+	    x->lft.hard_packet_limit == XFRM_INF &&
+	    x->lft.soft_byte_limit == XFRM_INF &&
+	    x->lft.hard_byte_limit == XFRM_INF)
 		return 0;
 
 	dwork = kzalloc(sizeof(*dwork), GFP_KERNEL);
@@ -670,7 +675,7 @@ static int mlx5e_ipsec_create_dwork(struct mlx5e_ipsec_sa_entry *sa_entry)
 		return -ENOMEM;
 
 	dwork->sa_entry = sa_entry;
-	INIT_DELAYED_WORK(&dwork->dwork, mlx5e_ipsec_handle_tx_limit);
+	INIT_DELAYED_WORK(&dwork->dwork, mlx5e_ipsec_handle_sw_limits);
 	sa_entry->dwork = dwork;
 	return 0;
 }
@@ -884,6 +889,7 @@ void mlx5e_ipsec_init(struct mlx5e_priv *priv)
 
 	xa_init_flags(&ipsec->sadb, XA_FLAGS_ALLOC);
 	ipsec->mdev = priv->mdev;
+	init_completion(&ipsec->comp);
 	ipsec->wq = alloc_workqueue("mlx5e_ipsec: %s", WQ_UNBOUND, 0,
 				    priv->netdev->name);
 	if (!ipsec->wq)
@@ -904,7 +910,7 @@ void mlx5e_ipsec_init(struct mlx5e_priv *priv)
 	}
 
 	ipsec->is_uplink_rep = mlx5e_is_uplink_rep(priv);
-	ret = mlx5e_accel_ipsec_fs_init(ipsec);
+	ret = mlx5e_accel_ipsec_fs_init(ipsec, &priv->devcom);
 	if (ret)
 		goto err_fs_init;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 9e7c42c2f7..adaea34931 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -38,6 +38,7 @@
 #include <net/xfrm.h>
 #include <linux/idr.h>
 #include "lib/aso.h"
+#include "lib/devcom.h"
 
 #define MLX5E_IPSEC_SADB_RX_BITS 10
 #define MLX5E_IPSEC_ESN_SCOPE_MID 0x80000000L
@@ -188,11 +189,19 @@ struct mlx5e_ipsec_ft {
 	u32 refcnt;
 };
 
+struct mlx5e_ipsec_drop {
+	struct mlx5_flow_handle *rule;
+	struct mlx5_fc *fc;
+};
+
 struct mlx5e_ipsec_rule {
 	struct mlx5_flow_handle *rule;
 	struct mlx5_modify_hdr *modify_hdr;
 	struct mlx5_pkt_reformat *pkt_reformat;
 	struct mlx5_fc *fc;
+	struct mlx5e_ipsec_drop replay;
+	struct mlx5e_ipsec_drop auth;
+	struct mlx5e_ipsec_drop trailer;
 };
 
 struct mlx5e_ipsec_miss {
@@ -200,19 +209,6 @@ struct mlx5e_ipsec_miss {
 	struct mlx5_flow_handle *rule;
 };
 
-struct mlx5e_ipsec_rx {
-	struct mlx5e_ipsec_ft ft;
-	struct mlx5e_ipsec_miss pol;
-	struct mlx5e_ipsec_miss sa;
-	struct mlx5e_ipsec_rule status;
-	struct mlx5e_ipsec_miss status_drop;
-	struct mlx5_fc *status_drop_cnt;
-	struct mlx5e_ipsec_fc *fc;
-	struct mlx5_fs_chains *chains;
-	u8 allow_tunnel_mode : 1;
-	struct xarray ipsec_obj_id_map;
-};
-
 struct mlx5e_ipsec_tx_create_attr {
 	int prio;
 	int pol_level;
@@ -221,12 +217,20 @@ struct mlx5e_ipsec_tx_create_attr {
 	enum mlx5_flow_namespace_type chains_ns;
 };
 
+struct mlx5e_ipsec_mpv_work {
+	int event;
+	struct work_struct work;
+	struct mlx5e_priv *slave_priv;
+	struct mlx5e_priv *master_priv;
+};
+
 struct mlx5e_ipsec {
 	struct mlx5_core_dev *mdev;
 	struct xarray sadb;
 	struct mlx5e_ipsec_sw_stats sw_stats;
 	struct mlx5e_ipsec_hw_stats hw_stats;
 	struct workqueue_struct *wq;
+	struct completion comp;
 	struct mlx5e_flow_steering *fs;
 	struct mlx5e_ipsec_rx *rx_ipv4;
 	struct mlx5e_ipsec_rx *rx_ipv6;
@@ -238,6 +242,8 @@ struct mlx5e_ipsec {
 	struct notifier_block netevent_nb;
 	struct mlx5_ipsec_fs *roce;
 	u8 is_uplink_rep: 1;
+	struct mlx5e_ipsec_mpv_work mpv_work;
+	struct xarray ipsec_obj_id_map;
 };
 
 struct mlx5e_ipsec_esn_state {
@@ -302,7 +308,7 @@ void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv);
 void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv);
 
 void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec);
-int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec);
+int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec, struct mlx5_devcom_comp_dev **devcom);
 int mlx5e_accel_ipsec_fs_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry);
 void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry);
 int mlx5e_accel_ipsec_fs_add_pol(struct mlx5e_ipsec_pol_entry *pol_entry);
@@ -328,6 +334,10 @@ void mlx5e_accel_ipsec_fs_read_stats(struct mlx5e_priv *priv,
 
 void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 					struct mlx5_accel_esp_xfrm_attrs *attrs);
+void mlx5e_ipsec_handle_mpv_event(int event, struct mlx5e_priv *slave_priv,
+				  struct mlx5e_priv *master_priv);
+void mlx5e_ipsec_send_event(struct mlx5e_priv *priv, int event);
+
 static inline struct mlx5_core_dev *
 mlx5e_ipsec_sa2dev(struct mlx5e_ipsec_sa_entry *sa_entry)
 {
@@ -363,6 +373,15 @@ static inline u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev)
 {
 	return 0;
 }
+
+static inline void mlx5e_ipsec_handle_mpv_event(int event, struct mlx5e_priv *slave_priv,
+						struct mlx5e_priv *master_priv)
+{
+}
+
+static inline void mlx5e_ipsec_send_event(struct mlx5e_priv *priv, int event)
+{
+}
 #endif
 
 #endif	/* __MLX5E_IPSEC_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 81e6aa6434..41a2543a52 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -32,6 +32,22 @@ struct mlx5e_ipsec_tx {
 	u8 allow_tunnel_mode : 1;
 };
 
+struct mlx5e_ipsec_status_checks {
+	struct mlx5_flow_group *drop_all_group;
+	struct mlx5e_ipsec_drop all;
+};
+
+struct mlx5e_ipsec_rx {
+	struct mlx5e_ipsec_ft ft;
+	struct mlx5e_ipsec_miss pol;
+	struct mlx5e_ipsec_miss sa;
+	struct mlx5e_ipsec_rule status;
+	struct mlx5e_ipsec_status_checks status_drops;
+	struct mlx5e_ipsec_fc *fc;
+	struct mlx5_fs_chains *chains;
+	u8 allow_tunnel_mode : 1;
+};
+
 /* IPsec RX flow steering */
 static enum mlx5_traffic_types family2tt(u32 family)
 {
@@ -131,9 +147,9 @@ static struct mlx5_flow_table *ipsec_ft_create(struct mlx5_flow_namespace *ns,
 static void ipsec_rx_status_drop_destroy(struct mlx5e_ipsec *ipsec,
 					 struct mlx5e_ipsec_rx *rx)
 {
-	mlx5_del_flow_rules(rx->status_drop.rule);
-	mlx5_destroy_flow_group(rx->status_drop.group);
-	mlx5_fc_destroy(ipsec->mdev, rx->status_drop_cnt);
+	mlx5_del_flow_rules(rx->status_drops.all.rule);
+	mlx5_fc_destroy(ipsec->mdev, rx->status_drops.all.fc);
+	mlx5_destroy_flow_group(rx->status_drops.drop_all_group);
 }
 
 static void ipsec_rx_status_pass_destroy(struct mlx5e_ipsec *ipsec,
@@ -149,8 +165,149 @@ static void ipsec_rx_status_pass_destroy(struct mlx5e_ipsec *ipsec,
 #endif
 }
 
-static int ipsec_rx_status_drop_create(struct mlx5e_ipsec *ipsec,
-				       struct mlx5e_ipsec_rx *rx)
+static int rx_add_rule_drop_auth_trailer(struct mlx5e_ipsec_sa_entry *sa_entry,
+					 struct mlx5e_ipsec_rx *rx)
+{
+	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
+	struct mlx5_flow_table *ft = rx->ft.status;
+	struct mlx5_core_dev *mdev = ipsec->mdev;
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act flow_act = {};
+	struct mlx5_flow_handle *rule;
+	struct mlx5_fc *flow_counter;
+	struct mlx5_flow_spec *spec;
+	int err;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec)
+		return -ENOMEM;
+
+	flow_counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(flow_counter)) {
+		err = PTR_ERR(flow_counter);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
+		goto err_cnt;
+	}
+	sa_entry->ipsec_rule.auth.fc = flow_counter;
+
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	flow_act.flags = FLOW_ACT_NO_APPEND;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest.counter_id = mlx5_fc_id(flow_counter);
+	if (rx == ipsec->rx_esw)
+		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
+
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.ipsec_syndrome);
+	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.ipsec_syndrome, 1);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_2);
+	MLX5_SET(fte_match_param, spec->match_value,
+		 misc_parameters_2.metadata_reg_c_2,
+		 sa_entry->ipsec_obj_id | BIT(31));
+	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
+		goto err_rule;
+	}
+	sa_entry->ipsec_rule.auth.rule = rule;
+
+	flow_counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(flow_counter)) {
+		err = PTR_ERR(flow_counter);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
+		goto err_cnt_2;
+	}
+	sa_entry->ipsec_rule.trailer.fc = flow_counter;
+
+	dest.counter_id = mlx5_fc_id(flow_counter);
+	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.ipsec_syndrome, 2);
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
+		goto err_rule_2;
+	}
+	sa_entry->ipsec_rule.trailer.rule = rule;
+
+	kvfree(spec);
+	return 0;
+
+err_rule_2:
+	mlx5_fc_destroy(mdev, sa_entry->ipsec_rule.trailer.fc);
+err_cnt_2:
+	mlx5_del_flow_rules(sa_entry->ipsec_rule.auth.rule);
+err_rule:
+	mlx5_fc_destroy(mdev, sa_entry->ipsec_rule.auth.fc);
+err_cnt:
+	kvfree(spec);
+	return err;
+}
+
+static int rx_add_rule_drop_replay(struct mlx5e_ipsec_sa_entry *sa_entry, struct mlx5e_ipsec_rx *rx)
+{
+	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
+	struct mlx5_flow_table *ft = rx->ft.status;
+	struct mlx5_core_dev *mdev = ipsec->mdev;
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act flow_act = {};
+	struct mlx5_flow_handle *rule;
+	struct mlx5_fc *flow_counter;
+	struct mlx5_flow_spec *spec;
+	int err;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec)
+		return -ENOMEM;
+
+	flow_counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(flow_counter)) {
+		err = PTR_ERR(flow_counter);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
+		goto err_cnt;
+	}
+
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	flow_act.flags = FLOW_ACT_NO_APPEND;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest.counter_id = mlx5_fc_id(flow_counter);
+	if (rx == ipsec->rx_esw)
+		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
+
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_4);
+	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.metadata_reg_c_4, 1);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_2);
+	MLX5_SET(fte_match_param, spec->match_value,  misc_parameters_2.metadata_reg_c_2,
+		 sa_entry->ipsec_obj_id | BIT(31));
+	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
+		goto err_rule;
+	}
+
+	sa_entry->ipsec_rule.replay.rule = rule;
+	sa_entry->ipsec_rule.replay.fc = flow_counter;
+
+	kvfree(spec);
+	return 0;
+
+err_rule:
+	mlx5_fc_destroy(mdev, flow_counter);
+err_cnt:
+	kvfree(spec);
+	return err;
+}
+
+static int ipsec_rx_status_drop_all_create(struct mlx5e_ipsec *ipsec,
+					   struct mlx5e_ipsec_rx *rx)
 {
 	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
 	struct mlx5_flow_table *ft = rx->ft.status;
@@ -202,9 +359,9 @@ static int ipsec_rx_status_drop_create(struct mlx5e_ipsec *ipsec,
 		goto err_rule;
 	}
 
-	rx->status_drop.group = g;
-	rx->status_drop.rule = rule;
-	rx->status_drop_cnt = flow_counter;
+	rx->status_drops.drop_all_group = g;
+	rx->status_drops.all.rule = rule;
+	rx->status_drops.all.fc = flow_counter;
 
 	kvfree(flow_group_in);
 	kvfree(spec);
@@ -235,8 +392,12 @@ static int ipsec_rx_status_pass_create(struct mlx5e_ipsec *ipsec,
 
 	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
 			 misc_parameters_2.ipsec_syndrome);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+			 misc_parameters_2.metadata_reg_c_4);
 	MLX5_SET(fte_match_param, spec->match_value,
 		 misc_parameters_2.ipsec_syndrome, 0);
+	MLX5_SET(fte_match_param, spec->match_value,
+		 misc_parameters_2.metadata_reg_c_4, 0);
 	if (rx == ipsec->rx_esw)
 		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
 	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
@@ -273,7 +434,7 @@ static int mlx5_ipsec_rx_status_create(struct mlx5e_ipsec *ipsec,
 {
 	int err;
 
-	err = ipsec_rx_status_drop_create(ipsec, rx);
+	err = ipsec_rx_status_drop_all_create(ipsec, rx);
 	if (err)
 		return err;
 
@@ -332,6 +493,83 @@ out:
 	return err;
 }
 
+static void handle_ipsec_rx_bringup(struct mlx5e_ipsec *ipsec, u32 family)
+{
+	struct mlx5e_ipsec_rx *rx = ipsec_rx(ipsec, family, XFRM_DEV_OFFLOAD_PACKET);
+	struct mlx5_flow_namespace *ns = mlx5e_fs_get_ns(ipsec->fs, false);
+	struct mlx5_flow_destination old_dest, new_dest;
+
+	old_dest = mlx5_ttc_get_default_dest(mlx5e_fs_get_ttc(ipsec->fs, false),
+					     family2tt(family));
+
+	mlx5_ipsec_fs_roce_rx_create(ipsec->mdev, ipsec->roce, ns, &old_dest, family,
+				     MLX5E_ACCEL_FS_ESP_FT_ROCE_LEVEL, MLX5E_NIC_PRIO);
+
+	new_dest.ft = mlx5_ipsec_fs_roce_ft_get(ipsec->roce, family);
+	new_dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	mlx5_modify_rule_destination(rx->status.rule, &new_dest, &old_dest);
+	mlx5_modify_rule_destination(rx->sa.rule, &new_dest, &old_dest);
+}
+
+static void handle_ipsec_rx_cleanup(struct mlx5e_ipsec *ipsec, u32 family)
+{
+	struct mlx5e_ipsec_rx *rx = ipsec_rx(ipsec, family, XFRM_DEV_OFFLOAD_PACKET);
+	struct mlx5_flow_destination old_dest, new_dest;
+
+	old_dest.ft = mlx5_ipsec_fs_roce_ft_get(ipsec->roce, family);
+	old_dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	new_dest = mlx5_ttc_get_default_dest(mlx5e_fs_get_ttc(ipsec->fs, false),
+					     family2tt(family));
+	mlx5_modify_rule_destination(rx->sa.rule, &new_dest, &old_dest);
+	mlx5_modify_rule_destination(rx->status.rule, &new_dest, &old_dest);
+
+	mlx5_ipsec_fs_roce_rx_destroy(ipsec->roce, family, ipsec->mdev);
+}
+
+static void ipsec_mpv_work_handler(struct work_struct *_work)
+{
+	struct mlx5e_ipsec_mpv_work *work = container_of(_work, struct mlx5e_ipsec_mpv_work, work);
+	struct mlx5e_ipsec *ipsec = work->slave_priv->ipsec;
+
+	switch (work->event) {
+	case MPV_DEVCOM_IPSEC_MASTER_UP:
+		mutex_lock(&ipsec->tx->ft.mutex);
+		if (ipsec->tx->ft.refcnt)
+			mlx5_ipsec_fs_roce_tx_create(ipsec->mdev, ipsec->roce, ipsec->tx->ft.pol,
+						     true);
+		mutex_unlock(&ipsec->tx->ft.mutex);
+
+		mutex_lock(&ipsec->rx_ipv4->ft.mutex);
+		if (ipsec->rx_ipv4->ft.refcnt)
+			handle_ipsec_rx_bringup(ipsec, AF_INET);
+		mutex_unlock(&ipsec->rx_ipv4->ft.mutex);
+
+		mutex_lock(&ipsec->rx_ipv6->ft.mutex);
+		if (ipsec->rx_ipv6->ft.refcnt)
+			handle_ipsec_rx_bringup(ipsec, AF_INET6);
+		mutex_unlock(&ipsec->rx_ipv6->ft.mutex);
+		break;
+	case MPV_DEVCOM_IPSEC_MASTER_DOWN:
+		mutex_lock(&ipsec->tx->ft.mutex);
+		if (ipsec->tx->ft.refcnt)
+			mlx5_ipsec_fs_roce_tx_destroy(ipsec->roce, ipsec->mdev);
+		mutex_unlock(&ipsec->tx->ft.mutex);
+
+		mutex_lock(&ipsec->rx_ipv4->ft.mutex);
+		if (ipsec->rx_ipv4->ft.refcnt)
+			handle_ipsec_rx_cleanup(ipsec, AF_INET);
+		mutex_unlock(&ipsec->rx_ipv4->ft.mutex);
+
+		mutex_lock(&ipsec->rx_ipv6->ft.mutex);
+		if (ipsec->rx_ipv6->ft.refcnt)
+			handle_ipsec_rx_cleanup(ipsec, AF_INET6);
+		mutex_unlock(&ipsec->rx_ipv6->ft.mutex);
+		break;
+	}
+
+	complete(&work->master_priv->ipsec->comp);
+}
+
 static void ipsec_rx_ft_disconnect(struct mlx5e_ipsec *ipsec, u32 family)
 {
 	struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(ipsec->fs, false);
@@ -362,7 +600,7 @@ static void rx_destroy(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	mlx5_ipsec_rx_status_destroy(ipsec, rx);
 	mlx5_destroy_flow_table(rx->ft.status);
 
-	mlx5_ipsec_fs_roce_rx_destroy(ipsec->roce, family);
+	mlx5_ipsec_fs_roce_rx_destroy(ipsec->roce, family, mdev);
 }
 
 static void ipsec_rx_create_attr_set(struct mlx5e_ipsec *ipsec,
@@ -440,7 +678,7 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	if (err)
 		return err;
 
-	ft = ipsec_ft_create(attr.ns, attr.status_level, attr.prio, 1, 0);
+	ft = ipsec_ft_create(attr.ns, attr.status_level, attr.prio, 3, 0);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
 		goto err_fs_ft_status;
@@ -517,7 +755,7 @@ err_fs_ft:
 err_add:
 	mlx5_destroy_flow_table(rx->ft.status);
 err_fs_ft_status:
-	mlx5_ipsec_fs_roce_rx_destroy(ipsec->roce, family);
+	mlx5_ipsec_fs_roce_rx_destroy(ipsec->roce, family, mdev);
 	return err;
 }
 
@@ -657,7 +895,7 @@ err_rule:
 static void tx_destroy(struct mlx5e_ipsec *ipsec, struct mlx5e_ipsec_tx *tx,
 		       struct mlx5_ipsec_fs *roce)
 {
-	mlx5_ipsec_fs_roce_tx_destroy(roce);
+	mlx5_ipsec_fs_roce_tx_destroy(roce, ipsec->mdev);
 	if (tx->chains) {
 		ipsec_chains_destroy(tx->chains);
 	} else {
@@ -760,7 +998,7 @@ static int tx_create(struct mlx5e_ipsec *ipsec, struct mlx5e_ipsec_tx *tx,
 	}
 
 connect_roce:
-	err = mlx5_ipsec_fs_roce_tx_create(mdev, roce, tx->ft.pol);
+	err = mlx5_ipsec_fs_roce_tx_create(mdev, roce, tx->ft.pol, false);
 	if (err)
 		goto err_roce;
 	return 0;
@@ -1079,29 +1317,48 @@ static int setup_modify_header(struct mlx5e_ipsec *ipsec, int type, u32 val, u8
 			       struct mlx5_flow_act *flow_act)
 {
 	enum mlx5_flow_namespace_type ns_type = ipsec_fs_get_ns(ipsec, type, dir);
-	u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {};
+	u8 action[3][MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {};
 	struct mlx5_core_dev *mdev = ipsec->mdev;
 	struct mlx5_modify_hdr *modify_hdr;
+	u8 num_of_actions = 1;
 
-	MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET);
+	MLX5_SET(set_action_in, action[0], action_type, MLX5_ACTION_TYPE_SET);
 	switch (dir) {
 	case XFRM_DEV_OFFLOAD_IN:
-		MLX5_SET(set_action_in, action, field,
+		MLX5_SET(set_action_in, action[0], field,
 			 MLX5_ACTION_IN_FIELD_METADATA_REG_B);
+
+		num_of_actions++;
+		MLX5_SET(set_action_in, action[1], action_type, MLX5_ACTION_TYPE_SET);
+		MLX5_SET(set_action_in, action[1], field, MLX5_ACTION_IN_FIELD_METADATA_REG_C_2);
+		MLX5_SET(set_action_in, action[1], data, val);
+		MLX5_SET(set_action_in, action[1], offset, 0);
+		MLX5_SET(set_action_in, action[1], length, 32);
+
+		if (type == XFRM_DEV_OFFLOAD_CRYPTO) {
+			num_of_actions++;
+			MLX5_SET(set_action_in, action[2], action_type,
+				 MLX5_ACTION_TYPE_SET);
+			MLX5_SET(set_action_in, action[2], field,
+				 MLX5_ACTION_IN_FIELD_METADATA_REG_C_4);
+			MLX5_SET(set_action_in, action[2], data, 0);
+			MLX5_SET(set_action_in, action[2], offset, 0);
+			MLX5_SET(set_action_in, action[2], length, 32);
+		}
 		break;
 	case XFRM_DEV_OFFLOAD_OUT:
-		MLX5_SET(set_action_in, action, field,
+		MLX5_SET(set_action_in, action[0], field,
 			 MLX5_ACTION_IN_FIELD_METADATA_REG_C_4);
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	MLX5_SET(set_action_in, action, data, val);
-	MLX5_SET(set_action_in, action, offset, 0);
-	MLX5_SET(set_action_in, action, length, 32);
+	MLX5_SET(set_action_in, action[0], data, val);
+	MLX5_SET(set_action_in, action[0], offset, 0);
+	MLX5_SET(set_action_in, action[0], length, 32);
 
-	modify_hdr = mlx5_modify_header_alloc(mdev, ns_type, 1, action);
+	modify_hdr = mlx5_modify_header_alloc(mdev, ns_type, num_of_actions, action);
 	if (IS_ERR(modify_hdr)) {
 		mlx5_core_err(mdev, "Failed to allocate modify_header %ld\n",
 			      PTR_ERR(modify_hdr));
@@ -1354,15 +1611,17 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	setup_fte_no_frags(spec);
 	setup_fte_upper_proto_match(spec, &attrs->upspec);
 
-	if (rx != ipsec->rx_esw)
-		err = setup_modify_header(ipsec, attrs->type,
-					  sa_entry->ipsec_obj_id | BIT(31),
-					  XFRM_DEV_OFFLOAD_IN, &flow_act);
-	else
-		err = mlx5_esw_ipsec_rx_setup_modify_header(sa_entry, &flow_act);
+	if (!attrs->drop) {
+		if (rx != ipsec->rx_esw)
+			err = setup_modify_header(ipsec, attrs->type,
+						  sa_entry->ipsec_obj_id | BIT(31),
+						  XFRM_DEV_OFFLOAD_IN, &flow_act);
+		else
+			err = mlx5_esw_ipsec_rx_setup_modify_header(sa_entry, &flow_act);
 
-	if (err)
-		goto err_mod_header;
+		if (err)
+			goto err_mod_header;
+	}
 
 	switch (attrs->type) {
 	case XFRM_DEV_OFFLOAD_PACKET:
@@ -1398,6 +1657,15 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 		mlx5_core_err(mdev, "fail to add RX ipsec rule err=%d\n", err);
 		goto err_add_flow;
 	}
+	if (attrs->type == XFRM_DEV_OFFLOAD_PACKET)
+		err = rx_add_rule_drop_replay(sa_entry, rx);
+	if (err)
+		goto err_add_replay;
+
+	err = rx_add_rule_drop_auth_trailer(sa_entry, rx);
+	if (err)
+		goto err_drop_reason;
+
 	kvfree(spec);
 
 	sa_entry->ipsec_rule.rule = rule;
@@ -1406,13 +1674,21 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	sa_entry->ipsec_rule.pkt_reformat = flow_act.pkt_reformat;
 	return 0;
 
+err_drop_reason:
+	if (sa_entry->ipsec_rule.replay.rule) {
+		mlx5_del_flow_rules(sa_entry->ipsec_rule.replay.rule);
+		mlx5_fc_destroy(mdev, sa_entry->ipsec_rule.replay.fc);
+	}
+err_add_replay:
+	mlx5_del_flow_rules(rule);
 err_add_flow:
 	mlx5_fc_destroy(mdev, counter);
 err_add_cnt:
 	if (flow_act.pkt_reformat)
 		mlx5_packet_reformat_dealloc(mdev, flow_act.pkt_reformat);
 err_pkt_reformat:
-	mlx5_modify_header_dealloc(mdev, flow_act.modify_hdr);
+	if (flow_act.modify_hdr)
+		mlx5_modify_header_dealloc(mdev, flow_act.modify_hdr);
 err_mod_header:
 	kvfree(spec);
 err_alloc:
@@ -1913,7 +2189,19 @@ void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 		return;
 	}
 
-	mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr);
+	if (ipsec_rule->modify_hdr)
+		mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr);
+
+	mlx5_del_flow_rules(ipsec_rule->trailer.rule);
+	mlx5_fc_destroy(mdev, ipsec_rule->trailer.fc);
+
+	mlx5_del_flow_rules(ipsec_rule->auth.rule);
+	mlx5_fc_destroy(mdev, ipsec_rule->auth.fc);
+
+	if (ipsec_rule->replay.rule) {
+		mlx5_del_flow_rules(ipsec_rule->replay.rule);
+		mlx5_fc_destroy(mdev, ipsec_rule->replay.fc);
+	}
 	mlx5_esw_ipsec_rx_id_mapping_remove(sa_entry);
 	rx_ft_put(sa_entry->ipsec, sa_entry->attrs.family, sa_entry->attrs.type);
 }
@@ -1984,7 +2272,7 @@ void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec)
 	kfree(ipsec->rx_ipv6);
 
 	if (ipsec->is_uplink_rep) {
-		xa_destroy(&ipsec->rx_esw->ipsec_obj_id_map);
+		xa_destroy(&ipsec->ipsec_obj_id_map);
 
 		mutex_destroy(&ipsec->tx_esw->ft.mutex);
 		WARN_ON(ipsec->tx_esw->ft.refcnt);
@@ -1996,7 +2284,8 @@ void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec)
 	}
 }
 
-int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec)
+int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec,
+			      struct mlx5_devcom_comp_dev **devcom)
 {
 	struct mlx5_core_dev *mdev = ipsec->mdev;
 	struct mlx5_flow_namespace *ns, *ns_esw;
@@ -2046,9 +2335,11 @@ int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec)
 		mutex_init(&ipsec->tx_esw->ft.mutex);
 		mutex_init(&ipsec->rx_esw->ft.mutex);
 		ipsec->tx_esw->ns = ns_esw;
-		xa_init_flags(&ipsec->rx_esw->ipsec_obj_id_map, XA_FLAGS_ALLOC1);
+		xa_init_flags(&ipsec->ipsec_obj_id_map, XA_FLAGS_ALLOC1);
 	} else if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_ROCE) {
-		ipsec->roce = mlx5_ipsec_fs_roce_init(mdev);
+		ipsec->roce = mlx5_ipsec_fs_roce_init(mdev, devcom);
+	} else {
+		mlx5_core_warn(mdev, "IPsec was initialized without RoCE support\n");
 	}
 
 	return 0;
@@ -2095,3 +2386,33 @@ bool mlx5e_ipsec_fs_tunnel_enabled(struct mlx5e_ipsec_sa_entry *sa_entry)
 
 	return rx->allow_tunnel_mode;
 }
+
+void mlx5e_ipsec_handle_mpv_event(int event, struct mlx5e_priv *slave_priv,
+				  struct mlx5e_priv *master_priv)
+{
+	struct mlx5e_ipsec_mpv_work *work;
+
+	reinit_completion(&master_priv->ipsec->comp);
+
+	if (!slave_priv->ipsec) {
+		complete(&master_priv->ipsec->comp);
+		return;
+	}
+
+	work = &slave_priv->ipsec->mpv_work;
+
+	INIT_WORK(&work->work, ipsec_mpv_work_handler);
+	work->event = event;
+	work->slave_priv = slave_priv;
+	work->master_priv = master_priv;
+	queue_work(slave_priv->ipsec->wq, &work->work);
+}
+
+void mlx5e_ipsec_send_event(struct mlx5e_priv *priv, int event)
+{
+	if (!priv->ipsec)
+		return; /* IPsec not supported */
+
+	mlx5_devcom_send_event(priv->devcom, event, event, priv);
+	wait_for_completion(&priv->ipsec->comp);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index ce29e31721..6e00afe467 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -5,6 +5,7 @@
 #include "en.h"
 #include "ipsec.h"
 #include "lib/crypto.h"
+#include "lib/ipsec_fs_roce.h"
 #include "fs_core.h"
 #include "eswitch.h"
 
@@ -68,7 +69,7 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev)
 			caps |= MLX5_IPSEC_CAP_ESPINUDP;
 	}
 
-	if (mlx5_get_roce_state(mdev) &&
+	if (mlx5_get_roce_state(mdev) && mlx5_ipsec_fs_is_mpv_roce_supported(mdev) &&
 	    MLX5_CAP_GEN_2(mdev, flow_table_type_2_type) & MLX5_FT_NIC_RX_2_NIC_RX_RDMA &&
 	    MLX5_CAP_GEN_2(mdev, flow_table_type_2_type) & MLX5_FT_NIC_TX_RDMA_2_NIC_TX)
 		caps |= MLX5_IPSEC_CAP_ROCE;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
index 9ee014a8ad..2ed99772f1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
@@ -54,7 +54,6 @@ struct mlx5e_accel_tx_ipsec_state {
 
 #ifdef CONFIG_MLX5_EN_IPSEC
 
-void mlx5e_ipsec_inverse_table_init(void);
 void mlx5e_ipsec_set_iv_esn(struct sk_buff *skb, struct xfrm_state *x,
 			    struct xfrm_offload *xo);
 void mlx5e_ipsec_set_iv(struct sk_buff *skb, struct xfrm_state *x,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 38263d5c98..c7c1b667b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -1252,7 +1252,7 @@ static u32 mlx5e_get_rxfh_key_size(struct net_device *netdev)
 
 u32 mlx5e_ethtool_get_rxfh_indir_size(struct mlx5e_priv *priv)
 {
-	return MLX5E_INDIR_RQT_SIZE;
+	return mlx5e_rqt_size(priv->mdev, priv->channels.params.num_channels);
 }
 
 static u32 mlx5e_get_rxfh_indir_size(struct net_device *netdev)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
index 934b0d5ce1..777d311d44 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -1283,9 +1283,7 @@ static int mlx5e_create_inner_ttc_table(struct mlx5e_flow_steering *fs,
 	mlx5e_set_inner_ttc_params(fs, rx_res, &ttc_params);
 	fs->inner_ttc = mlx5_create_inner_ttc_table(fs->mdev,
 						    &ttc_params);
-	if (IS_ERR(fs->inner_ttc))
-		return PTR_ERR(fs->inner_ttc);
-	return 0;
+	return PTR_ERR_OR_ZERO(fs->inner_ttc);
 }
 
 int mlx5e_create_ttc_table(struct mlx5e_flow_steering *fs,
@@ -1295,9 +1293,7 @@ int mlx5e_create_ttc_table(struct mlx5e_flow_steering *fs,
 
 	mlx5e_set_ttc_params(fs, rx_res, &ttc_params, true);
 	fs->ttc = mlx5_create_ttc_table(fs->mdev, &ttc_params);
-	if (IS_ERR(fs->ttc))
-		return PTR_ERR(fs->ttc);
-	return 0;
+	return PTR_ERR_OR_ZERO(fs->ttc);
 }
 
 int mlx5e_create_flow_steering(struct mlx5e_flow_steering *fs,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index c3961c2bbc..0c87ddb8a7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -69,6 +69,7 @@
 #include "en/htb.h"
 #include "qos.h"
 #include "en/trap.h"
+#include "lib/devcom.h"
 
 bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev, u8 page_shift,
 					    enum mlx5e_mpwrq_umr_mode umr_mode)
@@ -178,6 +179,61 @@ static void mlx5e_disable_async_events(struct mlx5e_priv *priv)
 	mlx5_notifier_unregister(priv->mdev, &priv->events_nb);
 }
 
+static int mlx5e_devcom_event_mpv(int event, void *my_data, void *event_data)
+{
+	struct mlx5e_priv *slave_priv = my_data;
+
+	switch (event) {
+	case MPV_DEVCOM_MASTER_UP:
+		mlx5_devcom_comp_set_ready(slave_priv->devcom, true);
+		break;
+	case MPV_DEVCOM_MASTER_DOWN:
+		/* no need for comp set ready false since we unregister after
+		 * and it hurts cleanup flow.
+		 */
+		break;
+	case MPV_DEVCOM_IPSEC_MASTER_UP:
+	case MPV_DEVCOM_IPSEC_MASTER_DOWN:
+		mlx5e_ipsec_handle_mpv_event(event, my_data, event_data);
+		break;
+	}
+
+	return 0;
+}
+
+static int mlx5e_devcom_init_mpv(struct mlx5e_priv *priv, u64 *data)
+{
+	priv->devcom = mlx5_devcom_register_component(priv->mdev->priv.devc,
+						      MLX5_DEVCOM_MPV,
+						      *data,
+						      mlx5e_devcom_event_mpv,
+						      priv);
+	if (IS_ERR_OR_NULL(priv->devcom))
+		return -EOPNOTSUPP;
+
+	if (mlx5_core_is_mp_master(priv->mdev)) {
+		mlx5_devcom_send_event(priv->devcom, MPV_DEVCOM_MASTER_UP,
+				       MPV_DEVCOM_MASTER_UP, priv);
+		mlx5e_ipsec_send_event(priv, MPV_DEVCOM_IPSEC_MASTER_UP);
+	}
+
+	return 0;
+}
+
+static void mlx5e_devcom_cleanup_mpv(struct mlx5e_priv *priv)
+{
+	if (IS_ERR_OR_NULL(priv->devcom))
+		return;
+
+	if (mlx5_core_is_mp_master(priv->mdev)) {
+		mlx5_devcom_send_event(priv->devcom, MPV_DEVCOM_MASTER_DOWN,
+				       MPV_DEVCOM_MASTER_DOWN, priv);
+		mlx5e_ipsec_send_event(priv, MPV_DEVCOM_IPSEC_MASTER_DOWN);
+	}
+
+	mlx5_devcom_unregister_component(priv->devcom);
+}
+
 static int blocking_event(struct notifier_block *nb, unsigned long event, void *data)
 {
 	struct mlx5e_priv *priv = container_of(nb, struct mlx5e_priv, blocking_events_nb);
@@ -192,6 +248,13 @@ static int blocking_event(struct notifier_block *nb, unsigned long event, void *
 			return NOTIFY_BAD;
 		}
 		break;
+	case MLX5_DRIVER_EVENT_AFFILIATION_DONE:
+		if (mlx5e_devcom_init_mpv(priv, data))
+			return NOTIFY_BAD;
+		break;
+	case MLX5_DRIVER_EVENT_AFFILIATION_REMOVED:
+		mlx5e_devcom_cleanup_mpv(priv);
+		break;
 	default:
 		return NOTIFY_DONE;
 	}
@@ -834,7 +897,7 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 		struct page_pool_params pp_params = { 0 };
 
 		pp_params.order     = 0;
-		pp_params.flags     = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV | PP_FLAG_PAGE_FRAG;
+		pp_params.flags     = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
 		pp_params.pool_size = pool_size;
 		pp_params.nid       = node;
 		pp_params.dev       = rq->pdev;
@@ -2886,8 +2949,12 @@ static int mlx5e_num_channels_changed(struct mlx5e_priv *priv)
 	mlx5e_set_default_xps_cpumasks(priv, &priv->channels.params);
 
 	/* This function may be called on attach, before priv->rx_res is created. */
-	if (!netif_is_rxfh_configured(priv->netdev) && priv->rx_res)
-		mlx5e_rx_res_rss_set_indir_uniform(priv->rx_res, count);
+	if (priv->rx_res) {
+		mlx5e_rx_res_rss_update_num_channels(priv->rx_res, count);
+
+		if (!netif_is_rxfh_configured(priv->netdev))
+			mlx5e_rx_res_rss_set_indir_uniform(priv->rx_res, count);
+	}
 
 	return 0;
 }
@@ -5347,10 +5414,6 @@ static int mlx5e_init_nic_rx(struct mlx5e_priv *priv)
 	enum mlx5e_rx_res_features features;
 	int err;
 
-	priv->rx_res = mlx5e_rx_res_alloc();
-	if (!priv->rx_res)
-		return -ENOMEM;
-
 	mlx5e_create_q_counters(priv);
 
 	err = mlx5e_open_drop_rq(priv, &priv->drop_rq);
@@ -5362,12 +5425,16 @@ static int mlx5e_init_nic_rx(struct mlx5e_priv *priv)
 	features = MLX5E_RX_RES_FEATURE_PTP;
 	if (mlx5_tunnel_inner_ft_supported(mdev))
 		features |= MLX5E_RX_RES_FEATURE_INNER_FT;
-	err = mlx5e_rx_res_init(priv->rx_res, priv->mdev, features,
-				priv->max_nch, priv->drop_rq.rqn,
-				&priv->channels.params.packet_merge,
-				priv->channels.params.num_channels);
-	if (err)
+
+	priv->rx_res = mlx5e_rx_res_create(priv->mdev, features, priv->max_nch, priv->drop_rq.rqn,
+					   &priv->channels.params.packet_merge,
+					   priv->channels.params.num_channels);
+	if (IS_ERR(priv->rx_res)) {
+		err = PTR_ERR(priv->rx_res);
+		priv->rx_res = NULL;
+		mlx5_core_err(mdev, "create rx resources failed, %d\n", err);
 		goto err_close_drop_rq;
+	}
 
 	err = mlx5e_create_flow_steering(priv->fs, priv->rx_res, priv->profile,
 					 priv->netdev);
@@ -5397,12 +5464,11 @@ err_destroy_flow_steering:
 				    priv->profile);
 err_destroy_rx_res:
 	mlx5e_rx_res_destroy(priv->rx_res);
+	priv->rx_res = NULL;
 err_close_drop_rq:
 	mlx5e_close_drop_rq(&priv->drop_rq);
 err_destroy_q_counters:
 	mlx5e_destroy_q_counters(priv);
-	mlx5e_rx_res_free(priv->rx_res);
-	priv->rx_res = NULL;
 	return err;
 }
 
@@ -5413,10 +5479,9 @@ static void mlx5e_cleanup_nic_rx(struct mlx5e_priv *priv)
 	mlx5e_destroy_flow_steering(priv->fs, !!(priv->netdev->hw_features & NETIF_F_NTUPLE),
 				    priv->profile);
 	mlx5e_rx_res_destroy(priv->rx_res);
+	priv->rx_res = NULL;
 	mlx5e_close_drop_rq(&priv->drop_rq);
 	mlx5e_destroy_q_counters(priv);
-	mlx5e_rx_res_free(priv->rx_res);
-	priv->rx_res = NULL;
 }
 
 static void mlx5e_set_mqprio_rl(struct mlx5e_priv *priv)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 751d3ffcd2..e92d4f8359 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1010,26 +1010,22 @@ static int mlx5e_init_rep_rx(struct mlx5e_priv *priv)
 	struct mlx5_core_dev *mdev = priv->mdev;
 	int err;
 
-	priv->rx_res = mlx5e_rx_res_alloc();
-	if (!priv->rx_res) {
-		err = -ENOMEM;
-		goto err_free_fs;
-	}
-
 	mlx5e_fs_init_l2_addr(priv->fs, priv->netdev);
 
 	err = mlx5e_open_drop_rq(priv, &priv->drop_rq);
 	if (err) {
 		mlx5_core_err(mdev, "open drop rq failed, %d\n", err);
-		goto err_rx_res_free;
+		goto err_free_fs;
 	}
 
-	err = mlx5e_rx_res_init(priv->rx_res, priv->mdev, 0,
-				priv->max_nch, priv->drop_rq.rqn,
-				&priv->channels.params.packet_merge,
-				priv->channels.params.num_channels);
-	if (err)
+	priv->rx_res = mlx5e_rx_res_create(priv->mdev, 0, priv->max_nch, priv->drop_rq.rqn,
+					   &priv->channels.params.packet_merge,
+					   priv->channels.params.num_channels);
+	if (IS_ERR(priv->rx_res)) {
+		err = PTR_ERR(priv->rx_res);
+		mlx5_core_err(mdev, "Create rx resources failed, err=%d\n", err);
 		goto err_close_drop_rq;
+	}
 
 	err = mlx5e_create_rep_ttc_table(priv);
 	if (err)
@@ -1053,11 +1049,9 @@ err_destroy_ttc_table:
 	mlx5_destroy_ttc_table(mlx5e_fs_get_ttc(priv->fs, false));
 err_destroy_rx_res:
 	mlx5e_rx_res_destroy(priv->rx_res);
+	priv->rx_res = ERR_PTR(-EINVAL);
 err_close_drop_rq:
 	mlx5e_close_drop_rq(&priv->drop_rq);
-err_rx_res_free:
-	mlx5e_rx_res_free(priv->rx_res);
-	priv->rx_res = NULL;
 err_free_fs:
 	mlx5e_fs_cleanup(priv->fs);
 	priv->fs = NULL;
@@ -1071,9 +1065,8 @@ static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv)
 	mlx5e_destroy_rep_root_ft(priv);
 	mlx5_destroy_ttc_table(mlx5e_fs_get_ttc(priv->fs, false));
 	mlx5e_rx_res_destroy(priv->rx_res);
+	priv->rx_res = ERR_PTR(-EINVAL);
 	mlx5e_close_drop_rq(&priv->drop_rq);
-	mlx5e_rx_res_free(priv->rx_res);
-	priv->rx_res = NULL;
 }
 
 static void mlx5e_rep_mpesw_work(struct work_struct *work)
@@ -1367,8 +1360,9 @@ mlx5e_rep_vnic_reporter_diagnose(struct devlink_health_reporter *reporter,
 	struct mlx5e_rep_priv *rpriv = devlink_health_reporter_priv(reporter);
 	struct mlx5_eswitch_rep *rep = rpriv->rep;
 
-	return mlx5_reporter_vnic_diagnose_counters(rep->esw->dev, fmsg,
-						    rep->vport, true);
+	mlx5_reporter_vnic_diagnose_counters(rep->esw->dev, fmsg, rep->vport,
+					     true);
+	return 0;
 }
 
 static const struct devlink_health_reporter_ops mlx5_rep_vnic_reporter_ops = {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index dc9b157a44..404dd1d9b2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -756,19 +756,21 @@ static int mlx5e_hairpin_create_indirect_rqt(struct mlx5e_hairpin *hp)
 {
 	struct mlx5e_priv *priv = hp->func_priv;
 	struct mlx5_core_dev *mdev = priv->mdev;
-	struct mlx5e_rss_params_indir *indir;
+	struct mlx5e_rss_params_indir indir;
 	int err;
 
-	indir = kvmalloc(sizeof(*indir), GFP_KERNEL);
-	if (!indir)
-		return -ENOMEM;
+	err = mlx5e_rss_params_indir_init(&indir, mdev,
+					  mlx5e_rqt_size(mdev, hp->num_channels),
+					  mlx5e_rqt_size(mdev, hp->num_channels));
+	if (err)
+		return err;
 
-	mlx5e_rss_params_indir_init_uniform(indir, hp->num_channels);
+	mlx5e_rss_params_indir_init_uniform(&indir, hp->num_channels);
 	err = mlx5e_rqt_init_indir(&hp->indir_rqt, mdev, hp->pair->rqn, hp->num_channels,
 				   mlx5e_rx_res_get_current_hash(priv->rx_res).hfunc,
-				   indir);
+				   &indir);
 
-	kvfree(indir);
+	mlx5e_rss_params_indir_cleanup(&indir);
 	return err;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c
index d5d33c3b3a..190f10aba1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c
@@ -50,7 +50,7 @@ int mlx5_esw_ipsec_rx_setup_modify_header(struct mlx5e_ipsec_sa_entry *sa_entry,
 	u32 mapped_id;
 	int err;
 
-	err = xa_alloc_bh(&ipsec->rx_esw->ipsec_obj_id_map, &mapped_id,
+	err = xa_alloc_bh(&ipsec->ipsec_obj_id_map, &mapped_id,
 			  xa_mk_value(sa_entry->ipsec_obj_id),
 			  XA_LIMIT(1, ESW_IPSEC_RX_MAPPED_ID_MASK), 0);
 	if (err)
@@ -81,7 +81,7 @@ int mlx5_esw_ipsec_rx_setup_modify_header(struct mlx5e_ipsec_sa_entry *sa_entry,
 	return 0;
 
 err_header_alloc:
-	xa_erase_bh(&ipsec->rx_esw->ipsec_obj_id_map, mapped_id);
+	xa_erase_bh(&ipsec->ipsec_obj_id_map, mapped_id);
 	return err;
 }
 
@@ -90,7 +90,7 @@ void mlx5_esw_ipsec_rx_id_mapping_remove(struct mlx5e_ipsec_sa_entry *sa_entry)
 	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
 
 	if (sa_entry->rx_mapped_id)
-		xa_erase_bh(&ipsec->rx_esw->ipsec_obj_id_map,
+		xa_erase_bh(&ipsec->ipsec_obj_id_map,
 			    sa_entry->rx_mapped_id);
 }
 
@@ -100,7 +100,7 @@ int mlx5_esw_ipsec_rx_ipsec_obj_id_search(struct mlx5e_priv *priv, u32 id,
 	struct mlx5e_ipsec *ipsec = priv->ipsec;
 	void *val;
 
-	val = xa_load(&ipsec->rx_esw->ipsec_obj_id_map, id);
+	val = xa_load(&ipsec->ipsec_obj_id_map, id);
 	if (!val)
 		return -ENOENT;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
index 1887a24ee4..d2ebe56c39 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
@@ -2,6 +2,7 @@
 /* Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
 
 #include "eswitch.h"
+#include "lib/mlx5.h"
 #include "esw/qos.h"
 #include "en/port.h"
 #define CREATE_TRACE_POINTS
@@ -701,10 +702,75 @@ int mlx5_esw_qos_set_vport_rate(struct mlx5_eswitch *esw, struct mlx5_vport *vpo
 	return err;
 }
 
+static u32 mlx5_esw_qos_lag_link_speed_get_locked(struct mlx5_core_dev *mdev)
+{
+	struct ethtool_link_ksettings lksettings;
+	struct net_device *slave, *master;
+	u32 speed = SPEED_UNKNOWN;
+
+	/* Lock ensures a stable reference to master and slave netdevice
+	 * while port speed of master is queried.
+	 */
+	ASSERT_RTNL();
+
+	slave = mlx5_uplink_netdev_get(mdev);
+	if (!slave)
+		goto out;
+
+	master = netdev_master_upper_dev_get(slave);
+	if (master && !__ethtool_get_link_ksettings(master, &lksettings))
+		speed = lksettings.base.speed;
+
+out:
+	return speed;
+}
+
+static int mlx5_esw_qos_max_link_speed_get(struct mlx5_core_dev *mdev, u32 *link_speed_max,
+					   bool hold_rtnl_lock, struct netlink_ext_ack *extack)
+{
+	int err;
+
+	if (!mlx5_lag_is_active(mdev))
+		goto skip_lag;
+
+	if (hold_rtnl_lock)
+		rtnl_lock();
+
+	*link_speed_max = mlx5_esw_qos_lag_link_speed_get_locked(mdev);
+
+	if (hold_rtnl_lock)
+		rtnl_unlock();
+
+	if (*link_speed_max != (u32)SPEED_UNKNOWN)
+		return 0;
+
+skip_lag:
+	err = mlx5_port_max_linkspeed(mdev, link_speed_max);
+	if (err)
+		NL_SET_ERR_MSG_MOD(extack, "Failed to get link maximum speed");
+
+	return err;
+}
+
+static int mlx5_esw_qos_link_speed_verify(struct mlx5_core_dev *mdev,
+					  const char *name, u32 link_speed_max,
+					  u64 value, struct netlink_ext_ack *extack)
+{
+	if (value > link_speed_max) {
+		pr_err("%s rate value %lluMbps exceed link maximum speed %u.\n",
+		       name, value, link_speed_max);
+		NL_SET_ERR_MSG_MOD(extack, "TX rate value exceed link maximum speed");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 int mlx5_esw_qos_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, u32 rate_mbps)
 {
 	u32 ctx[MLX5_ST_SZ_DW(scheduling_context)] = {};
 	struct mlx5_vport *vport;
+	u32 link_speed_max;
 	u32 bitmask;
 	int err;
 
@@ -712,6 +778,17 @@ int mlx5_esw_qos_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, u32
 	if (IS_ERR(vport))
 		return PTR_ERR(vport);
 
+	if (rate_mbps) {
+		err = mlx5_esw_qos_max_link_speed_get(esw->dev, &link_speed_max, false, NULL);
+		if (err)
+			return err;
+
+		err = mlx5_esw_qos_link_speed_verify(esw->dev, "Police",
+						     link_speed_max, rate_mbps, NULL);
+		if (err)
+			return err;
+	}
+
 	mutex_lock(&esw->state_lock);
 	if (!vport->qos.enabled) {
 		/* Eswitch QoS wasn't enabled yet. Enable it and vport QoS. */
@@ -744,12 +821,6 @@ static int esw_qos_devlink_rate_to_mbps(struct mlx5_core_dev *mdev, const char *
 	u64 value;
 	int err;
 
-	err = mlx5_port_max_linkspeed(mdev, &link_speed_max);
-	if (err) {
-		NL_SET_ERR_MSG_MOD(extack, "Failed to get link maximum speed");
-		return err;
-	}
-
 	value = div_u64_rem(*rate, MLX5_LINKSPEED_UNIT, &remainder);
 	if (remainder) {
 		pr_err("%s rate value %lluBps not in link speed units of 1Mbps.\n",
@@ -758,12 +829,13 @@ static int esw_qos_devlink_rate_to_mbps(struct mlx5_core_dev *mdev, const char *
 		return -EINVAL;
 	}
 
-	if (value > link_speed_max) {
-		pr_err("%s rate value %lluMbps exceed link maximum speed %u.\n",
-		       name, value, link_speed_max);
-		NL_SET_ERR_MSG_MOD(extack, "TX rate value exceed link maximum speed");
-		return -EINVAL;
-	}
+	err = mlx5_esw_qos_max_link_speed_get(mdev, &link_speed_max, true, extack);
+	if (err)
+		return err;
+
+	err = mlx5_esw_qos_link_speed_verify(mdev, name, link_speed_max, value, extack);
+	if (err)
+		return err;
 
 	*rate = value;
 	return 0;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/events.c b/drivers/net/ethernet/mellanox/mlx5/core/events.c
index 3ec892d51f..d91ea53eb3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/events.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/events.c
@@ -441,8 +441,3 @@ int mlx5_blocking_notifier_call_chain(struct mlx5_core_dev *dev, unsigned int ev
 
 	return blocking_notifier_call_chain(&events->sw_nh, event, data);
 }
-
-void mlx5_events_work_enqueue(struct mlx5_core_dev *dev, struct work_struct *work)
-{
-	queue_work(dev->priv.events->wq, work);
-}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index a13b9c2bd1..e6bfa7e4f1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -114,9 +114,9 @@
 #define ETHTOOL_NUM_PRIOS 11
 #define ETHTOOL_MIN_LEVEL (KERNEL_MIN_LEVEL + ETHTOOL_NUM_PRIOS)
 /* Promiscuous, Vlan, mac, ttc, inner ttc, {UDP/ANY/aRFS/accel/{esp, esp_err}}, IPsec policy,
- * IPsec RoCE policy
+ * {IPsec RoCE MPV,Alias table},IPsec RoCE policy
  */
-#define KERNEL_NIC_PRIO_NUM_LEVELS 9
+#define KERNEL_NIC_PRIO_NUM_LEVELS 11
 #define KERNEL_NIC_NUM_PRIOS 1
 /* One more level for tc */
 #define KERNEL_MIN_LEVEL (KERNEL_NIC_PRIO_NUM_LEVELS + 1)
@@ -137,7 +137,7 @@
 #define LAG_MIN_LEVEL (OFFLOADS_MIN_LEVEL + KERNEL_RX_MACSEC_MIN_LEVEL + 1)
 
 #define KERNEL_TX_IPSEC_NUM_PRIOS  1
-#define KERNEL_TX_IPSEC_NUM_LEVELS 3
+#define KERNEL_TX_IPSEC_NUM_LEVELS 4
 #define KERNEL_TX_IPSEC_MIN_LEVEL        (KERNEL_TX_IPSEC_NUM_LEVELS)
 
 #define KERNEL_TX_MACSEC_NUM_PRIOS  1
@@ -231,7 +231,7 @@ enum {
 };
 
 #define RDMA_RX_IPSEC_NUM_PRIOS 1
-#define RDMA_RX_IPSEC_NUM_LEVELS 2
+#define RDMA_RX_IPSEC_NUM_LEVELS 4
 #define RDMA_RX_IPSEC_MIN_LEVEL  (RDMA_RX_IPSEC_NUM_LEVELS)
 
 #define RDMA_RX_BYPASS_MIN_LEVEL MLX5_BY_PASS_NUM_REGULAR_PRIOS
@@ -288,7 +288,7 @@ enum {
 #define RDMA_TX_BYPASS_MIN_LEVEL MLX5_BY_PASS_NUM_PRIOS
 #define RDMA_TX_COUNTERS_MIN_LEVEL (RDMA_TX_BYPASS_MIN_LEVEL + 1)
 
-#define RDMA_TX_IPSEC_NUM_PRIOS 1
+#define RDMA_TX_IPSEC_NUM_PRIOS 2
 #define RDMA_TX_IPSEC_PRIO_NUM_LEVELS 1
 #define RDMA_TX_IPSEC_MIN_LEVEL  (RDMA_TX_COUNTERS_MIN_LEVEL + RDMA_TX_IPSEC_NUM_PRIOS)
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c
index 2fb2598b77..8ff6dc9bc8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c
@@ -365,6 +365,8 @@ static const char *hsynd_str(u8 synd)
 		return "FFSER error";
 	case MLX5_INITIAL_SEG_HEALTH_SYNDROME_HIGH_TEMP_ERR:
 		return "High temperature";
+	case MLX5_INITIAL_SEG_HEALTH_SYNDROME_ICM_PCI_POISONED_ERR:
+		return "ICM fetch PCI data poisoned error";
 	default:
 		return "unrecognized error";
 	}
@@ -448,14 +450,15 @@ mlx5_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
 	struct mlx5_core_dev *dev = devlink_health_reporter_priv(reporter);
 	struct mlx5_core_health *health = &dev->priv.health;
 	struct health_buffer __iomem *h = health->health;
-	u8 synd;
-	int err;
+	u8 synd = ioread8(&h->synd);
 
-	synd = ioread8(&h->synd);
-	err = devlink_fmsg_u8_pair_put(fmsg, "Syndrome", synd);
-	if (err || !synd)
-		return err;
-	return devlink_fmsg_string_pair_put(fmsg, "Description", hsynd_str(synd));
+	if (!synd)
+		return 0;
+
+	devlink_fmsg_u8_pair_put(fmsg, "Syndrome", synd);
+	devlink_fmsg_string_pair_put(fmsg, "Description", hsynd_str(synd));
+
+	return 0;
 }
 
 struct mlx5_fw_reporter_ctx {
@@ -463,94 +466,47 @@ struct mlx5_fw_reporter_ctx {
 	int miss_counter;
 };
 
-static int
+static void
 mlx5_fw_reporter_ctx_pairs_put(struct devlink_fmsg *fmsg,
 			       struct mlx5_fw_reporter_ctx *fw_reporter_ctx)
 {
-	int err;
-
-	err = devlink_fmsg_u8_pair_put(fmsg, "syndrome",
-				       fw_reporter_ctx->err_synd);
-	if (err)
-		return err;
-	err = devlink_fmsg_u32_pair_put(fmsg, "fw_miss_counter",
-					fw_reporter_ctx->miss_counter);
-	if (err)
-		return err;
-	return 0;
+	devlink_fmsg_u8_pair_put(fmsg, "syndrome", fw_reporter_ctx->err_synd);
+	devlink_fmsg_u32_pair_put(fmsg, "fw_miss_counter", fw_reporter_ctx->miss_counter);
 }
 
-static int
+static void
 mlx5_fw_reporter_heath_buffer_data_put(struct mlx5_core_dev *dev,
 				       struct devlink_fmsg *fmsg)
 {
 	struct mlx5_core_health *health = &dev->priv.health;
 	struct health_buffer __iomem *h = health->health;
 	u8 rfr_severity;
-	int err;
 	int i;
 
 	if (!ioread8(&h->synd))
-		return 0;
-
-	err = devlink_fmsg_pair_nest_start(fmsg, "health buffer");
-	if (err)
-		return err;
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "assert_var");
-	if (err)
-		return err;
+		return;
 
-	for (i = 0; i < ARRAY_SIZE(h->assert_var); i++) {
-		err = devlink_fmsg_u32_put(fmsg, ioread32be(h->assert_var + i));
-		if (err)
-			return err;
-	}
-	err = devlink_fmsg_arr_pair_nest_end(fmsg);
-	if (err)
-		return err;
-	err = devlink_fmsg_u32_pair_put(fmsg, "assert_exit_ptr",
-					ioread32be(&h->assert_exit_ptr));
-	if (err)
-		return err;
-	err = devlink_fmsg_u32_pair_put(fmsg, "assert_callra",
-					ioread32be(&h->assert_callra));
-	if (err)
-		return err;
-	err = devlink_fmsg_u32_pair_put(fmsg, "time", ioread32be(&h->time));
-	if (err)
-		return err;
-	err = devlink_fmsg_u32_pair_put(fmsg, "hw_id", ioread32be(&h->hw_id));
-	if (err)
-		return err;
+	devlink_fmsg_pair_nest_start(fmsg, "health buffer");
+	devlink_fmsg_obj_nest_start(fmsg);
+	devlink_fmsg_arr_pair_nest_start(fmsg, "assert_var");
+	for (i = 0; i < ARRAY_SIZE(h->assert_var); i++)
+		devlink_fmsg_u32_put(fmsg, ioread32be(h->assert_var + i));
+	devlink_fmsg_arr_pair_nest_end(fmsg);
+	devlink_fmsg_u32_pair_put(fmsg, "assert_exit_ptr",
+				  ioread32be(&h->assert_exit_ptr));
+	devlink_fmsg_u32_pair_put(fmsg, "assert_callra",
+				  ioread32be(&h->assert_callra));
+	devlink_fmsg_u32_pair_put(fmsg, "time", ioread32be(&h->time));
+	devlink_fmsg_u32_pair_put(fmsg, "hw_id", ioread32be(&h->hw_id));
 	rfr_severity = ioread8(&h->rfr_severity);
-	err = devlink_fmsg_u8_pair_put(fmsg, "rfr", mlx5_health_get_rfr(rfr_severity));
-	if (err)
-		return err;
-	err = devlink_fmsg_u8_pair_put(fmsg, "severity", mlx5_health_get_severity(rfr_severity));
-	if (err)
-		return err;
-	err = devlink_fmsg_u8_pair_put(fmsg, "irisc_index",
-				       ioread8(&h->irisc_index));
-	if (err)
-		return err;
-	err = devlink_fmsg_u8_pair_put(fmsg, "synd", ioread8(&h->synd));
-	if (err)
-		return err;
-	err = devlink_fmsg_u32_pair_put(fmsg, "ext_synd",
-					ioread16be(&h->ext_synd));
-	if (err)
-		return err;
-	err = devlink_fmsg_u32_pair_put(fmsg, "raw_fw_ver",
-					ioread32be(&h->fw_ver));
-	if (err)
-		return err;
-	err = devlink_fmsg_obj_nest_end(fmsg);
-	if (err)
-		return err;
-	return devlink_fmsg_pair_nest_end(fmsg);
+	devlink_fmsg_u8_pair_put(fmsg, "rfr", mlx5_health_get_rfr(rfr_severity));
+	devlink_fmsg_u8_pair_put(fmsg, "severity", mlx5_health_get_severity(rfr_severity));
+	devlink_fmsg_u8_pair_put(fmsg, "irisc_index", ioread8(&h->irisc_index));
+	devlink_fmsg_u8_pair_put(fmsg, "synd", ioread8(&h->synd));
+	devlink_fmsg_u32_pair_put(fmsg, "ext_synd", ioread16be(&h->ext_synd));
+	devlink_fmsg_u32_pair_put(fmsg, "raw_fw_ver", ioread32be(&h->fw_ver));
+	devlink_fmsg_obj_nest_end(fmsg);
+	devlink_fmsg_pair_nest_end(fmsg);
 }
 
 static int
@@ -568,14 +524,11 @@ mlx5_fw_reporter_dump(struct devlink_health_reporter *reporter,
 	if (priv_ctx) {
 		struct mlx5_fw_reporter_ctx *fw_reporter_ctx = priv_ctx;
 
-		err = mlx5_fw_reporter_ctx_pairs_put(fmsg, fw_reporter_ctx);
-		if (err)
-			return err;
+		mlx5_fw_reporter_ctx_pairs_put(fmsg, fw_reporter_ctx);
 	}
 
-	err = mlx5_fw_reporter_heath_buffer_data_put(dev, fmsg);
-	if (err)
-		return err;
+	mlx5_fw_reporter_heath_buffer_data_put(dev, fmsg);
+
 	return mlx5_fw_tracer_get_saved_traces_objects(dev->tracer, fmsg);
 }
 
@@ -641,12 +594,10 @@ mlx5_fw_fatal_reporter_dump(struct devlink_health_reporter *reporter,
 	if (priv_ctx) {
 		struct mlx5_fw_reporter_ctx *fw_reporter_ctx = priv_ctx;
 
-		err = mlx5_fw_reporter_ctx_pairs_put(fmsg, fw_reporter_ctx);
-		if (err)
-			goto free_data;
+		mlx5_fw_reporter_ctx_pairs_put(fmsg, fw_reporter_ctx);
 	}
 
-	err = devlink_fmsg_binary_pair_put(fmsg, "crdump_data", cr_data, crdump_size);
+	devlink_fmsg_binary_pair_put(fmsg, "crdump_data", cr_data, crdump_size);
 
 free_data:
 	kvfree(cr_data);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index baa7ef8123..2bf77a5251 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -418,12 +418,6 @@ static int mlx5i_init_rx(struct mlx5e_priv *priv)
 		return -ENOMEM;
 	}
 
-	priv->rx_res = mlx5e_rx_res_alloc();
-	if (!priv->rx_res) {
-		err = -ENOMEM;
-		goto err_free_fs;
-	}
-
 	mlx5e_create_q_counters(priv);
 
 	err = mlx5e_open_drop_rq(priv, &priv->drop_rq);
@@ -432,12 +426,13 @@ static int mlx5i_init_rx(struct mlx5e_priv *priv)
 		goto err_destroy_q_counters;
 	}
 
-	err = mlx5e_rx_res_init(priv->rx_res, priv->mdev, 0,
-				priv->max_nch, priv->drop_rq.rqn,
-				&priv->channels.params.packet_merge,
-				priv->channels.params.num_channels);
-	if (err)
+	priv->rx_res = mlx5e_rx_res_create(priv->mdev, 0, priv->max_nch, priv->drop_rq.rqn,
+					   &priv->channels.params.packet_merge,
+					   priv->channels.params.num_channels);
+	if (IS_ERR(priv->rx_res)) {
+		err = PTR_ERR(priv->rx_res);
 		goto err_close_drop_rq;
+	}
 
 	err = mlx5i_create_flow_steering(priv);
 	if (err)
@@ -447,13 +442,11 @@ static int mlx5i_init_rx(struct mlx5e_priv *priv)
 
 err_destroy_rx_res:
 	mlx5e_rx_res_destroy(priv->rx_res);
+	priv->rx_res = ERR_PTR(-EINVAL);
 err_close_drop_rq:
 	mlx5e_close_drop_rq(&priv->drop_rq);
 err_destroy_q_counters:
 	mlx5e_destroy_q_counters(priv);
-	mlx5e_rx_res_free(priv->rx_res);
-	priv->rx_res = NULL;
-err_free_fs:
 	mlx5e_fs_cleanup(priv->fs);
 	return err;
 }
@@ -462,10 +455,9 @@ static void mlx5i_cleanup_rx(struct mlx5e_priv *priv)
 {
 	mlx5i_destroy_flow_steering(priv);
 	mlx5e_rx_res_destroy(priv->rx_res);
+	priv->rx_res = ERR_PTR(-EINVAL);
 	mlx5e_close_drop_rq(&priv->drop_rq);
 	mlx5e_destroy_q_counters(priv);
-	mlx5e_rx_res_free(priv->rx_res);
-	priv->rx_res = NULL;
 	mlx5e_fs_cleanup(priv->fs);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
index af3fac090b..d14459e5c0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
@@ -943,6 +943,26 @@ static void mlx5_do_bond(struct mlx5_lag *ldev)
 	}
 }
 
+/* The last mdev to unregister will destroy the workqueue before removing the
+ * devcom component, and as all the mdevs use the same devcom component we are
+ * guaranteed that the devcom is valid while the calling work is running.
+ */
+struct mlx5_devcom_comp_dev *mlx5_lag_get_devcom_comp(struct mlx5_lag *ldev)
+{
+	struct mlx5_devcom_comp_dev *devcom = NULL;
+	int i;
+
+	mutex_lock(&ldev->lock);
+	for (i = 0; i < ldev->ports; i++) {
+		if (ldev->pf[i].dev) {
+			devcom = ldev->pf[i].dev->priv.hca_devcom_comp;
+			break;
+		}
+	}
+	mutex_unlock(&ldev->lock);
+	return devcom;
+}
+
 static void mlx5_queue_bond_work(struct mlx5_lag *ldev, unsigned long delay)
 {
 	queue_delayed_work(ldev->wq, &ldev->bond_work, delay);
@@ -953,9 +973,14 @@ 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);
+	struct mlx5_devcom_comp_dev *devcom;
 	int status;
 
-	status = mlx5_dev_list_trylock();
+	devcom = mlx5_lag_get_devcom_comp(ldev);
+	if (!devcom)
+		return;
+
+	status = mlx5_devcom_comp_trylock(devcom);
 	if (!status) {
 		mlx5_queue_bond_work(ldev, HZ);
 		return;
@@ -964,14 +989,14 @@ static void mlx5_do_bond_work(struct work_struct *work)
 	mutex_lock(&ldev->lock);
 	if (ldev->mode_changes_in_progress) {
 		mutex_unlock(&ldev->lock);
-		mlx5_dev_list_unlock();
+		mlx5_devcom_comp_unlock(devcom);
 		mlx5_queue_bond_work(ldev, HZ);
 		return;
 	}
 
 	mlx5_do_bond(ldev);
 	mutex_unlock(&ldev->lock);
-	mlx5_dev_list_unlock();
+	mlx5_devcom_comp_unlock(devcom);
 }
 
 static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
@@ -1212,13 +1237,14 @@ static void mlx5_ldev_remove_mdev(struct mlx5_lag *ldev,
 	dev->priv.lag = NULL;
 }
 
-/* Must be called with intf_mutex held */
+/* Must be called with HCA devcom component lock held */
 static int __mlx5_lag_dev_add_mdev(struct mlx5_core_dev *dev)
 {
+	struct mlx5_devcom_comp_dev *pos = NULL;
 	struct mlx5_lag *ldev = NULL;
 	struct mlx5_core_dev *tmp_dev;
 
-	tmp_dev = mlx5_get_next_phys_dev_lag(dev);
+	tmp_dev = mlx5_devcom_get_next_peer_data(dev->priv.hca_devcom_comp, &pos);
 	if (tmp_dev)
 		ldev = mlx5_lag_dev(tmp_dev);
 
@@ -1275,10 +1301,13 @@ void mlx5_lag_add_mdev(struct mlx5_core_dev *dev)
 	if (!mlx5_lag_is_supported(dev))
 		return;
 
+	if (IS_ERR_OR_NULL(dev->priv.hca_devcom_comp))
+		return;
+
 recheck:
-	mlx5_dev_list_lock();
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	err = __mlx5_lag_dev_add_mdev(dev);
-	mlx5_dev_list_unlock();
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 
 	if (err) {
 		msleep(100);
@@ -1431,7 +1460,7 @@ void mlx5_lag_disable_change(struct mlx5_core_dev *dev)
 	if (!ldev)
 		return;
 
-	mlx5_dev_list_lock();
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	mutex_lock(&ldev->lock);
 
 	ldev->mode_changes_in_progress++;
@@ -1439,7 +1468,7 @@ void mlx5_lag_disable_change(struct mlx5_core_dev *dev)
 		mlx5_disable_lag(ldev);
 
 	mutex_unlock(&ldev->lock);
-	mlx5_dev_list_unlock();
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 }
 
 void mlx5_lag_enable_change(struct mlx5_core_dev *dev)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
index 481e92f39f..50fcb1eee5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
@@ -112,6 +112,7 @@ void mlx5_disable_lag(struct mlx5_lag *ldev);
 void mlx5_lag_remove_devices(struct mlx5_lag *ldev);
 int mlx5_deactivate_lag(struct mlx5_lag *ldev);
 void mlx5_lag_add_devices(struct mlx5_lag *ldev);
+struct mlx5_devcom_comp_dev *mlx5_lag_get_devcom_comp(struct mlx5_lag *ldev);
 
 static inline bool mlx5_lag_is_supported(struct mlx5_core_dev *dev)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
index 4bf1539152..82889f3050 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
@@ -65,12 +65,12 @@ err_metadata:
 	return err;
 }
 
-#define MLX5_LAG_MPESW_OFFLOADS_SUPPORTED_PORTS 2
+#define MLX5_LAG_MPESW_OFFLOADS_SUPPORTED_PORTS 4
 static int enable_mpesw(struct mlx5_lag *ldev)
 {
 	struct mlx5_core_dev *dev0 = ldev->pf[MLX5_LAG_P1].dev;
-	struct mlx5_core_dev *dev1 = ldev->pf[MLX5_LAG_P2].dev;
 	int err;
+	int i;
 
 	if (ldev->mode != MLX5_LAG_MODE_NONE)
 		return -EINVAL;
@@ -98,11 +98,11 @@ static int enable_mpesw(struct mlx5_lag *ldev)
 
 	dev0->priv.flags &= ~MLX5_PRIV_FLAGS_DISABLE_IB_ADEV;
 	mlx5_rescan_drivers_locked(dev0);
-	err = mlx5_eswitch_reload_reps(dev0->priv.eswitch);
-	if (!err)
-		err = mlx5_eswitch_reload_reps(dev1->priv.eswitch);
-	if (err)
-		goto err_rescan_drivers;
+	for (i = 0; i < ldev->ports; i++) {
+		err = mlx5_eswitch_reload_reps(ldev->pf[i].dev->priv.eswitch);
+		if (err)
+			goto err_rescan_drivers;
+	}
 
 	return 0;
 
@@ -112,8 +112,8 @@ err_rescan_drivers:
 	mlx5_deactivate_lag(ldev);
 err_add_devices:
 	mlx5_lag_add_devices(ldev);
-	mlx5_eswitch_reload_reps(dev0->priv.eswitch);
-	mlx5_eswitch_reload_reps(dev1->priv.eswitch);
+	for (i = 0; i < ldev->ports; i++)
+		mlx5_eswitch_reload_reps(ldev->pf[i].dev->priv.eswitch);
 	mlx5_mpesw_metadata_cleanup(ldev);
 	return err;
 }
@@ -129,9 +129,14 @@ static void disable_mpesw(struct mlx5_lag *ldev)
 static void mlx5_mpesw_work(struct work_struct *work)
 {
 	struct mlx5_mpesw_work_st *mpesww = container_of(work, struct mlx5_mpesw_work_st, work);
+	struct mlx5_devcom_comp_dev *devcom;
 	struct mlx5_lag *ldev = mpesww->lag;
 
-	mlx5_dev_list_lock();
+	devcom = mlx5_lag_get_devcom_comp(ldev);
+	if (!devcom)
+		return;
+
+	mlx5_devcom_comp_lock(devcom);
 	mutex_lock(&ldev->lock);
 	if (ldev->mode_changes_in_progress) {
 		mpesww->result = -EAGAIN;
@@ -144,7 +149,7 @@ static void mlx5_mpesw_work(struct work_struct *work)
 		disable_mpesw(ldev);
 unlock:
 	mutex_unlock(&ldev->lock);
-	mlx5_dev_list_unlock();
+	mlx5_devcom_comp_unlock(devcom);
 	complete(&mpesww->comp);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c
index 7d9bbb494d..101b3bb908 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c
@@ -507,10 +507,7 @@ static int mlx5_lag_create_ttc_table(struct mlx5_lag *ldev)
 
 	mlx5_lag_set_outer_ttc_params(ldev, &ttc_params);
 	port_sel->outer.ttc = mlx5_create_ttc_table(dev, &ttc_params);
-	if (IS_ERR(port_sel->outer.ttc))
-		return PTR_ERR(port_sel->outer.ttc);
-
-	return 0;
+	return PTR_ERR_OR_ZERO(port_sel->outer.ttc);
 }
 
 static int mlx5_lag_create_inner_ttc_table(struct mlx5_lag *ldev)
@@ -521,10 +518,7 @@ static int mlx5_lag_create_inner_ttc_table(struct mlx5_lag *ldev)
 
 	mlx5_lag_set_inner_ttc_params(ldev, &ttc_params);
 	port_sel->inner.ttc = mlx5_create_inner_ttc_table(dev, &ttc_params);
-	if (IS_ERR(port_sel->inner.ttc))
-		return PTR_ERR(port_sel->inner.ttc);
-
-	return 0;
+	return PTR_ERR_OR_ZERO(port_sel->inner.ttc);
 }
 
 int mlx5_lag_port_sel_create(struct mlx5_lag *ldev,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
index 00e67910e3..e8e50563e9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
@@ -31,6 +31,7 @@ struct mlx5_devcom_comp {
 	struct kref ref;
 	bool ready;
 	struct rw_semaphore sem;
+	struct lock_class_key lock_key;
 };
 
 struct mlx5_devcom_comp_dev {
@@ -119,6 +120,8 @@ mlx5_devcom_comp_alloc(u64 id, u64 key, mlx5_devcom_event_handler_t handler)
 	comp->key = key;
 	comp->handler = handler;
 	init_rwsem(&comp->sem);
+	lockdep_register_key(&comp->lock_key);
+	lockdep_set_class(&comp->sem, &comp->lock_key);
 	kref_init(&comp->ref);
 	INIT_LIST_HEAD(&comp->comp_dev_list_head);
 
@@ -133,6 +136,7 @@ mlx5_devcom_comp_release(struct kref *ref)
 	mutex_lock(&comp_list_lock);
 	list_del(&comp->comp_list);
 	mutex_unlock(&comp_list_lock);
+	lockdep_unregister_key(&comp->lock_key);
 	kfree(comp);
 }
 
@@ -383,3 +387,24 @@ void *mlx5_devcom_get_next_peer_data_rcu(struct mlx5_devcom_comp_dev *devcom,
 	*pos = tmp;
 	return data;
 }
+
+void mlx5_devcom_comp_lock(struct mlx5_devcom_comp_dev *devcom)
+{
+	if (IS_ERR_OR_NULL(devcom))
+		return;
+	down_write(&devcom->comp->sem);
+}
+
+void mlx5_devcom_comp_unlock(struct mlx5_devcom_comp_dev *devcom)
+{
+	if (IS_ERR_OR_NULL(devcom))
+		return;
+	up_write(&devcom->comp->sem);
+}
+
+int mlx5_devcom_comp_trylock(struct mlx5_devcom_comp_dev *devcom)
+{
+	if (IS_ERR_OR_NULL(devcom))
+		return 0;
+	return down_write_trylock(&devcom->comp->sem);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h
index 8389ac0af7..fc23bbef87 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h
@@ -8,6 +8,8 @@
 
 enum mlx5_devcom_component {
 	MLX5_DEVCOM_ESW_OFFLOADS,
+	MLX5_DEVCOM_MPV,
+	MLX5_DEVCOM_HCA_PORTS,
 	MLX5_DEVCOM_NUM_COMPONENTS,
 };
 
@@ -51,4 +53,8 @@ void *mlx5_devcom_get_next_peer_data_rcu(struct mlx5_devcom_comp_dev *devcom,
 	     data;								  \
 	     data = mlx5_devcom_get_next_peer_data_rcu(devcom, &pos))
 
+void mlx5_devcom_comp_lock(struct mlx5_devcom_comp_dev *devcom);
+void mlx5_devcom_comp_unlock(struct mlx5_devcom_comp_dev *devcom);
+int mlx5_devcom_comp_trylock(struct mlx5_devcom_comp_dev *devcom);
+
 #endif /* __LIB_MLX5_DEVCOM_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
index 69a7545977..4b7f7131c5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
@@ -85,7 +85,6 @@ void mlx5_eq_del_cq(struct mlx5_eq *eq, struct mlx5_core_cq *cq);
 struct mlx5_eq_comp *mlx5_eqn2comp_eq(struct mlx5_core_dev *dev, int eqn);
 struct mlx5_eq *mlx5_get_async_eq(struct mlx5_core_dev *dev);
 void mlx5_cq_tasklet_cb(struct tasklet_struct *t);
-struct cpumask *mlx5_eq_comp_cpumask(struct mlx5_core_dev *dev, int ix);
 
 u32 mlx5_eq_poll_irq_disabled(struct mlx5_eq_comp *eq);
 void mlx5_cmd_eq_recover(struct mlx5_core_dev *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.c
index 6e3f178d6f..234cd00f71 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.c
@@ -2,8 +2,11 @@
 /* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
 
 #include "fs_core.h"
+#include "fs_cmd.h"
+#include "en.h"
 #include "lib/ipsec_fs_roce.h"
 #include "mlx5_core.h"
+#include <linux/random.h>
 
 struct mlx5_ipsec_miss {
 	struct mlx5_flow_group *group;
@@ -15,6 +18,12 @@ struct mlx5_ipsec_rx_roce {
 	struct mlx5_flow_table *ft;
 	struct mlx5_flow_handle *rule;
 	struct mlx5_ipsec_miss roce_miss;
+	struct mlx5_flow_table *nic_master_ft;
+	struct mlx5_flow_group *nic_master_group;
+	struct mlx5_flow_handle *nic_master_rule;
+	struct mlx5_flow_table *goto_alias_ft;
+	u32 alias_id;
+	char key[ACCESS_KEY_LEN];
 
 	struct mlx5_flow_table *ft_rdma;
 	struct mlx5_flow_namespace *ns_rdma;
@@ -24,6 +33,9 @@ struct mlx5_ipsec_tx_roce {
 	struct mlx5_flow_group *g;
 	struct mlx5_flow_table *ft;
 	struct mlx5_flow_handle *rule;
+	struct mlx5_flow_table *goto_alias_ft;
+	u32 alias_id;
+	char key[ACCESS_KEY_LEN];
 	struct mlx5_flow_namespace *ns;
 };
 
@@ -31,6 +43,7 @@ struct mlx5_ipsec_fs {
 	struct mlx5_ipsec_rx_roce ipv4_rx;
 	struct mlx5_ipsec_rx_roce ipv6_rx;
 	struct mlx5_ipsec_tx_roce tx;
+	struct mlx5_devcom_comp_dev **devcom;
 };
 
 static void ipsec_fs_roce_setup_udp_dport(struct mlx5_flow_spec *spec,
@@ -43,11 +56,83 @@ static void ipsec_fs_roce_setup_udp_dport(struct mlx5_flow_spec *spec,
 	MLX5_SET(fte_match_param, spec->match_value, outer_headers.udp_dport, dport);
 }
 
+static bool ipsec_fs_create_alias_supported_one(struct mlx5_core_dev *mdev)
+{
+	u64 obj_allowed = MLX5_CAP_GEN_2_64(mdev, allowed_object_for_other_vhca_access);
+	u32 obj_supp = MLX5_CAP_GEN_2(mdev, cross_vhca_object_to_object_supported);
+
+	if (!(obj_supp &
+	    MLX5_CROSS_VHCA_OBJ_TO_OBJ_SUPPORTED_LOCAL_FLOW_TABLE_TO_REMOTE_FLOW_TABLE_MISS))
+		return false;
+
+	if (!(obj_allowed & MLX5_ALLOWED_OBJ_FOR_OTHER_VHCA_ACCESS_FLOW_TABLE))
+		return false;
+
+	return true;
+}
+
+static bool ipsec_fs_create_alias_supported(struct mlx5_core_dev *mdev,
+					    struct mlx5_core_dev *master_mdev)
+{
+	if (ipsec_fs_create_alias_supported_one(mdev) &&
+	    ipsec_fs_create_alias_supported_one(master_mdev))
+		return true;
+
+	return false;
+}
+
+static int ipsec_fs_create_aliased_ft(struct mlx5_core_dev *ibv_owner,
+				      struct mlx5_core_dev *ibv_allowed,
+				      struct mlx5_flow_table *ft,
+				      u32 *obj_id, char *alias_key, bool from_event)
+{
+	u32 aliased_object_id = (ft->type << FT_ID_FT_TYPE_OFFSET) | ft->id;
+	u16 vhca_id_to_be_accessed = MLX5_CAP_GEN(ibv_owner, vhca_id);
+	struct mlx5_cmd_allow_other_vhca_access_attr allow_attr = {};
+	struct mlx5_cmd_alias_obj_create_attr alias_attr = {};
+	int ret;
+	int i;
+
+	if (!ipsec_fs_create_alias_supported(ibv_owner, ibv_allowed))
+		return -EOPNOTSUPP;
+
+	for (i = 0; i < ACCESS_KEY_LEN; i++)
+		if (!from_event)
+			alias_key[i] = get_random_u64() & 0xFF;
+
+	memcpy(allow_attr.access_key, alias_key, ACCESS_KEY_LEN);
+	allow_attr.obj_type = MLX5_GENERAL_OBJECT_TYPES_FLOW_TABLE_ALIAS;
+	allow_attr.obj_id = aliased_object_id;
+
+	if (!from_event) {
+		ret = mlx5_cmd_allow_other_vhca_access(ibv_owner, &allow_attr);
+		if (ret) {
+			mlx5_core_err(ibv_owner, "Failed to allow other vhca access err=%d\n",
+				      ret);
+			return ret;
+		}
+	}
+
+	memcpy(alias_attr.access_key, alias_key, ACCESS_KEY_LEN);
+	alias_attr.obj_id = aliased_object_id;
+	alias_attr.obj_type = MLX5_GENERAL_OBJECT_TYPES_FLOW_TABLE_ALIAS;
+	alias_attr.vhca_id = vhca_id_to_be_accessed;
+	ret = mlx5_cmd_alias_obj_create(ibv_allowed, &alias_attr, obj_id);
+	if (ret) {
+		mlx5_core_err(ibv_allowed, "Failed to create alias object err=%d\n",
+			      ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int
 ipsec_fs_roce_rx_rule_setup(struct mlx5_core_dev *mdev,
 			    struct mlx5_flow_destination *default_dst,
 			    struct mlx5_ipsec_rx_roce *roce)
 {
+	bool is_mpv_slave = mlx5_core_is_mp_slave(mdev);
 	struct mlx5_flow_destination dst = {};
 	MLX5_DECLARE_FLOW_ACT(flow_act);
 	struct mlx5_flow_handle *rule;
@@ -61,14 +146,19 @@ ipsec_fs_roce_rx_rule_setup(struct mlx5_core_dev *mdev,
 	ipsec_fs_roce_setup_udp_dport(spec, ROCE_V2_UDP_DPORT);
 
 	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
-	dst.type = MLX5_FLOW_DESTINATION_TYPE_TABLE_TYPE;
-	dst.ft = roce->ft_rdma;
+	if (is_mpv_slave) {
+		dst.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+		dst.ft = roce->goto_alias_ft;
+	} else {
+		dst.type = MLX5_FLOW_DESTINATION_TYPE_TABLE_TYPE;
+		dst.ft = roce->ft_rdma;
+	}
 	rule = mlx5_add_flow_rules(roce->ft, spec, &flow_act, &dst, 1);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
 		mlx5_core_err(mdev, "Fail to add RX RoCE IPsec rule err=%d\n",
 			      err);
-		goto fail_add_rule;
+		goto out;
 	}
 
 	roce->rule = rule;
@@ -84,12 +174,30 @@ ipsec_fs_roce_rx_rule_setup(struct mlx5_core_dev *mdev,
 
 	roce->roce_miss.rule = rule;
 
+	if (!is_mpv_slave)
+		goto out;
+
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	dst.type = MLX5_FLOW_DESTINATION_TYPE_TABLE_TYPE;
+	dst.ft = roce->ft_rdma;
+	rule = mlx5_add_flow_rules(roce->nic_master_ft, NULL, &flow_act, &dst,
+				   1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev, "Fail to add RX RoCE IPsec rule for alias err=%d\n",
+			      err);
+		goto fail_add_nic_master_rule;
+	}
+	roce->nic_master_rule = rule;
+
 	kvfree(spec);
 	return 0;
 
+fail_add_nic_master_rule:
+	mlx5_del_flow_rules(roce->roce_miss.rule);
 fail_add_default_rule:
 	mlx5_del_flow_rules(roce->rule);
-fail_add_rule:
+out:
 	kvfree(spec);
 	return err;
 }
@@ -120,25 +228,373 @@ out:
 	return err;
 }
 
-void mlx5_ipsec_fs_roce_tx_destroy(struct mlx5_ipsec_fs *ipsec_roce)
+static int ipsec_fs_roce_tx_mpv_rule_setup(struct mlx5_core_dev *mdev,
+					   struct mlx5_ipsec_tx_roce *roce,
+					   struct mlx5_flow_table *pol_ft)
 {
+	struct mlx5_flow_destination dst = {};
+	MLX5_DECLARE_FLOW_ACT(flow_act);
+	struct mlx5_flow_handle *rule;
+	struct mlx5_flow_spec *spec;
+	int err = 0;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec)
+		return -ENOMEM;
+
+	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters.source_vhca_port);
+	MLX5_SET(fte_match_param, spec->match_value, misc_parameters.source_vhca_port,
+		 MLX5_CAP_GEN(mdev, native_port_num));
+
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	dst.type = MLX5_FLOW_DESTINATION_TYPE_TABLE_TYPE;
+	dst.ft = roce->goto_alias_ft;
+	rule = mlx5_add_flow_rules(roce->ft, spec, &flow_act, &dst, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev, "Fail to add TX RoCE IPsec rule err=%d\n",
+			      err);
+		goto out;
+	}
+	roce->rule = rule;
+
+	/* No need for miss rule, since on miss we go to next PRIO, in which
+	 * if master is configured, he will catch the traffic to go to his
+	 * encryption table.
+	 */
+
+out:
+	kvfree(spec);
+	return err;
+}
+
+#define MLX5_TX_ROCE_GROUP_SIZE BIT(0)
+#define MLX5_IPSEC_RDMA_TX_FT_LEVEL 0
+#define MLX5_IPSEC_NIC_GOTO_ALIAS_FT_LEVEL 3 /* Since last used level in NIC ipsec is 2 */
+
+static int ipsec_fs_roce_tx_mpv_create_ft(struct mlx5_core_dev *mdev,
+					  struct mlx5_ipsec_tx_roce *roce,
+					  struct mlx5_flow_table *pol_ft,
+					  struct mlx5e_priv *peer_priv,
+					  bool from_event)
+{
+	struct mlx5_flow_namespace *roce_ns, *nic_ns;
+	struct mlx5_flow_table_attr ft_attr = {};
+	struct mlx5_flow_table next_ft;
+	struct mlx5_flow_table *ft;
+	int err;
+
+	roce_ns = mlx5_get_flow_namespace(peer_priv->mdev, MLX5_FLOW_NAMESPACE_RDMA_TX_IPSEC);
+	if (!roce_ns)
+		return -EOPNOTSUPP;
+
+	nic_ns = mlx5_get_flow_namespace(peer_priv->mdev, MLX5_FLOW_NAMESPACE_EGRESS_IPSEC);
+	if (!nic_ns)
+		return -EOPNOTSUPP;
+
+	err = ipsec_fs_create_aliased_ft(mdev, peer_priv->mdev, pol_ft, &roce->alias_id, roce->key,
+					 from_event);
+	if (err)
+		return err;
+
+	next_ft.id = roce->alias_id;
+	ft_attr.max_fte = 1;
+	ft_attr.next_ft = &next_ft;
+	ft_attr.level = MLX5_IPSEC_NIC_GOTO_ALIAS_FT_LEVEL;
+	ft_attr.flags = MLX5_FLOW_TABLE_UNMANAGED;
+	ft = mlx5_create_flow_table(nic_ns, &ft_attr);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec goto alias ft err=%d\n", err);
+		goto destroy_alias;
+	}
+
+	roce->goto_alias_ft = ft;
+
+	memset(&ft_attr, 0, sizeof(ft_attr));
+	ft_attr.max_fte = 1;
+	ft_attr.level = MLX5_IPSEC_RDMA_TX_FT_LEVEL;
+	ft = mlx5_create_flow_table(roce_ns, &ft_attr);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec tx ft err=%d\n", err);
+		goto destroy_alias_ft;
+	}
+
+	roce->ft = ft;
+
+	return 0;
+
+destroy_alias_ft:
+	mlx5_destroy_flow_table(roce->goto_alias_ft);
+destroy_alias:
+	mlx5_cmd_alias_obj_destroy(peer_priv->mdev, roce->alias_id,
+				   MLX5_GENERAL_OBJECT_TYPES_FLOW_TABLE_ALIAS);
+	return err;
+}
+
+static int ipsec_fs_roce_tx_mpv_create_group_rules(struct mlx5_core_dev *mdev,
+						   struct mlx5_ipsec_tx_roce *roce,
+						   struct mlx5_flow_table *pol_ft,
+						   u32 *in)
+{
+	struct mlx5_flow_group *g;
+	int ix = 0;
+	int err;
+	u8 *mc;
+
+	mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+	MLX5_SET_TO_ONES(fte_match_param, mc, misc_parameters.source_vhca_port);
+	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_MISC_PARAMETERS);
+
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += MLX5_TX_ROCE_GROUP_SIZE;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	g = mlx5_create_flow_group(roce->ft, in);
+	if (IS_ERR(g)) {
+		err = PTR_ERR(g);
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec tx group err=%d\n", err);
+		return err;
+	}
+	roce->g = g;
+
+	err = ipsec_fs_roce_tx_mpv_rule_setup(mdev, roce, pol_ft);
+	if (err) {
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec tx rules err=%d\n", err);
+		goto destroy_group;
+	}
+
+	return 0;
+
+destroy_group:
+	mlx5_destroy_flow_group(roce->g);
+	return err;
+}
+
+static int ipsec_fs_roce_tx_mpv_create(struct mlx5_core_dev *mdev,
+				       struct mlx5_ipsec_fs *ipsec_roce,
+				       struct mlx5_flow_table *pol_ft,
+				       u32 *in, bool from_event)
+{
+	struct mlx5_devcom_comp_dev *tmp = NULL;
+	struct mlx5_ipsec_tx_roce *roce;
+	struct mlx5e_priv *peer_priv;
+	int err;
+
+	if (!mlx5_devcom_for_each_peer_begin(*ipsec_roce->devcom))
+		return -EOPNOTSUPP;
+
+	peer_priv = mlx5_devcom_get_next_peer_data(*ipsec_roce->devcom, &tmp);
+	if (!peer_priv) {
+		err = -EOPNOTSUPP;
+		goto release_peer;
+	}
+
+	roce = &ipsec_roce->tx;
+
+	err = ipsec_fs_roce_tx_mpv_create_ft(mdev, roce, pol_ft, peer_priv, from_event);
+	if (err) {
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec tables err=%d\n", err);
+		goto release_peer;
+	}
+
+	err = ipsec_fs_roce_tx_mpv_create_group_rules(mdev, roce, pol_ft, in);
+	if (err) {
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec tx group/rule err=%d\n", err);
+		goto destroy_tables;
+	}
+
+	mlx5_devcom_for_each_peer_end(*ipsec_roce->devcom);
+	return 0;
+
+destroy_tables:
+	mlx5_destroy_flow_table(roce->ft);
+	mlx5_destroy_flow_table(roce->goto_alias_ft);
+	mlx5_cmd_alias_obj_destroy(peer_priv->mdev, roce->alias_id,
+				   MLX5_GENERAL_OBJECT_TYPES_FLOW_TABLE_ALIAS);
+release_peer:
+	mlx5_devcom_for_each_peer_end(*ipsec_roce->devcom);
+	return err;
+}
+
+static void roce_rx_mpv_destroy_tables(struct mlx5_core_dev *mdev, struct mlx5_ipsec_rx_roce *roce)
+{
+	mlx5_destroy_flow_table(roce->goto_alias_ft);
+	mlx5_cmd_alias_obj_destroy(mdev, roce->alias_id,
+				   MLX5_GENERAL_OBJECT_TYPES_FLOW_TABLE_ALIAS);
+	mlx5_destroy_flow_group(roce->nic_master_group);
+	mlx5_destroy_flow_table(roce->nic_master_ft);
+}
+
+#define MLX5_RX_ROCE_GROUP_SIZE BIT(0)
+#define MLX5_IPSEC_RX_IPV4_FT_LEVEL 3
+#define MLX5_IPSEC_RX_IPV6_FT_LEVEL 2
+
+static int ipsec_fs_roce_rx_mpv_create(struct mlx5_core_dev *mdev,
+				       struct mlx5_ipsec_fs *ipsec_roce,
+				       struct mlx5_flow_namespace *ns,
+				       u32 family, u32 level, u32 prio)
+{
+	struct mlx5_flow_namespace *roce_ns, *nic_ns;
+	struct mlx5_flow_table_attr ft_attr = {};
+	struct mlx5_devcom_comp_dev *tmp = NULL;
+	struct mlx5_ipsec_rx_roce *roce;
+	struct mlx5_flow_table next_ft;
+	struct mlx5_flow_table *ft;
+	struct mlx5_flow_group *g;
+	struct mlx5e_priv *peer_priv;
+	int ix = 0;
+	u32 *in;
+	int err;
+
+	roce = (family == AF_INET) ? &ipsec_roce->ipv4_rx :
+				     &ipsec_roce->ipv6_rx;
+
+	if (!mlx5_devcom_for_each_peer_begin(*ipsec_roce->devcom))
+		return -EOPNOTSUPP;
+
+	peer_priv = mlx5_devcom_get_next_peer_data(*ipsec_roce->devcom, &tmp);
+	if (!peer_priv) {
+		err = -EOPNOTSUPP;
+		goto release_peer;
+	}
+
+	roce_ns = mlx5_get_flow_namespace(peer_priv->mdev, MLX5_FLOW_NAMESPACE_RDMA_RX_IPSEC);
+	if (!roce_ns) {
+		err = -EOPNOTSUPP;
+		goto release_peer;
+	}
+
+	nic_ns = mlx5_get_flow_namespace(peer_priv->mdev, MLX5_FLOW_NAMESPACE_KERNEL);
+	if (!nic_ns) {
+		err = -EOPNOTSUPP;
+		goto release_peer;
+	}
+
+	in = kvzalloc(MLX5_ST_SZ_BYTES(create_flow_group_in), GFP_KERNEL);
+	if (!in) {
+		err = -ENOMEM;
+		goto release_peer;
+	}
+
+	ft_attr.level = (family == AF_INET) ? MLX5_IPSEC_RX_IPV4_FT_LEVEL :
+					      MLX5_IPSEC_RX_IPV6_FT_LEVEL;
+	ft_attr.max_fte = 1;
+	ft = mlx5_create_flow_table(roce_ns, &ft_attr);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec rx ft at rdma master err=%d\n", err);
+		goto free_in;
+	}
+
+	roce->ft_rdma = ft;
+
+	ft_attr.max_fte = 1;
+	ft_attr.prio = prio;
+	ft_attr.level = level + 2;
+	ft = mlx5_create_flow_table(nic_ns, &ft_attr);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec rx ft at NIC master err=%d\n", err);
+		goto destroy_ft_rdma;
+	}
+	roce->nic_master_ft = ft;
+
+	MLX5_SET_CFG(in, start_flow_index, ix);
+	ix += 1;
+	MLX5_SET_CFG(in, end_flow_index, ix - 1);
+	g = mlx5_create_flow_group(roce->nic_master_ft, in);
+	if (IS_ERR(g)) {
+		err = PTR_ERR(g);
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec rx group aliased err=%d\n", err);
+		goto destroy_nic_master_ft;
+	}
+	roce->nic_master_group = g;
+
+	err = ipsec_fs_create_aliased_ft(peer_priv->mdev, mdev, roce->nic_master_ft,
+					 &roce->alias_id, roce->key, false);
+	if (err) {
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec rx alias FT err=%d\n", err);
+		goto destroy_group;
+	}
+
+	next_ft.id = roce->alias_id;
+	ft_attr.max_fte = 1;
+	ft_attr.prio = prio;
+	ft_attr.level = roce->ft->level + 1;
+	ft_attr.flags = MLX5_FLOW_TABLE_UNMANAGED;
+	ft_attr.next_ft = &next_ft;
+	ft = mlx5_create_flow_table(ns, &ft_attr);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		mlx5_core_err(mdev, "Fail to create RoCE IPsec rx ft at NIC slave err=%d\n", err);
+		goto destroy_alias;
+	}
+	roce->goto_alias_ft = ft;
+
+	kvfree(in);
+	mlx5_devcom_for_each_peer_end(*ipsec_roce->devcom);
+	return 0;
+
+destroy_alias:
+	mlx5_cmd_alias_obj_destroy(mdev, roce->alias_id,
+				   MLX5_GENERAL_OBJECT_TYPES_FLOW_TABLE_ALIAS);
+destroy_group:
+	mlx5_destroy_flow_group(roce->nic_master_group);
+destroy_nic_master_ft:
+	mlx5_destroy_flow_table(roce->nic_master_ft);
+destroy_ft_rdma:
+	mlx5_destroy_flow_table(roce->ft_rdma);
+free_in:
+	kvfree(in);
+release_peer:
+	mlx5_devcom_for_each_peer_end(*ipsec_roce->devcom);
+	return err;
+}
+
+void mlx5_ipsec_fs_roce_tx_destroy(struct mlx5_ipsec_fs *ipsec_roce,
+				   struct mlx5_core_dev *mdev)
+{
+	struct mlx5_devcom_comp_dev *tmp = NULL;
 	struct mlx5_ipsec_tx_roce *tx_roce;
+	struct mlx5e_priv *peer_priv;
 
 	if (!ipsec_roce)
 		return;
 
 	tx_roce = &ipsec_roce->tx;
 
+	if (!tx_roce->ft)
+		return; /* Incase RoCE was cleaned from MPV event flow */
+
 	mlx5_del_flow_rules(tx_roce->rule);
 	mlx5_destroy_flow_group(tx_roce->g);
 	mlx5_destroy_flow_table(tx_roce->ft);
-}
 
-#define MLX5_TX_ROCE_GROUP_SIZE BIT(0)
+	if (!mlx5_core_is_mp_slave(mdev))
+		return;
+
+	if (!mlx5_devcom_for_each_peer_begin(*ipsec_roce->devcom))
+		return;
+
+	peer_priv = mlx5_devcom_get_next_peer_data(*ipsec_roce->devcom, &tmp);
+	if (!peer_priv) {
+		mlx5_devcom_for_each_peer_end(*ipsec_roce->devcom);
+		return;
+	}
+
+	mlx5_destroy_flow_table(tx_roce->goto_alias_ft);
+	mlx5_cmd_alias_obj_destroy(peer_priv->mdev, tx_roce->alias_id,
+				   MLX5_GENERAL_OBJECT_TYPES_FLOW_TABLE_ALIAS);
+	mlx5_devcom_for_each_peer_end(*ipsec_roce->devcom);
+	tx_roce->ft = NULL;
+}
 
 int mlx5_ipsec_fs_roce_tx_create(struct mlx5_core_dev *mdev,
 				 struct mlx5_ipsec_fs *ipsec_roce,
-				 struct mlx5_flow_table *pol_ft)
+				 struct mlx5_flow_table *pol_ft,
+				 bool from_event)
 {
 	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_ipsec_tx_roce *roce;
@@ -157,7 +613,14 @@ int mlx5_ipsec_fs_roce_tx_create(struct mlx5_core_dev *mdev,
 	if (!in)
 		return -ENOMEM;
 
+	if (mlx5_core_is_mp_slave(mdev)) {
+		err = ipsec_fs_roce_tx_mpv_create(mdev, ipsec_roce, pol_ft, in, from_event);
+		goto free_in;
+	}
+
 	ft_attr.max_fte = 1;
+	ft_attr.prio = 1;
+	ft_attr.level = MLX5_IPSEC_RDMA_TX_FT_LEVEL;
 	ft = mlx5_create_flow_table(roce->ns, &ft_attr);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
@@ -209,8 +672,10 @@ struct mlx5_flow_table *mlx5_ipsec_fs_roce_ft_get(struct mlx5_ipsec_fs *ipsec_ro
 	return rx_roce->ft;
 }
 
-void mlx5_ipsec_fs_roce_rx_destroy(struct mlx5_ipsec_fs *ipsec_roce, u32 family)
+void mlx5_ipsec_fs_roce_rx_destroy(struct mlx5_ipsec_fs *ipsec_roce, u32 family,
+				   struct mlx5_core_dev *mdev)
 {
+	bool is_mpv_slave = mlx5_core_is_mp_slave(mdev);
 	struct mlx5_ipsec_rx_roce *rx_roce;
 
 	if (!ipsec_roce)
@@ -218,23 +683,29 @@ void mlx5_ipsec_fs_roce_rx_destroy(struct mlx5_ipsec_fs *ipsec_roce, u32 family)
 
 	rx_roce = (family == AF_INET) ? &ipsec_roce->ipv4_rx :
 					&ipsec_roce->ipv6_rx;
+	if (!rx_roce->ft)
+		return; /* Incase RoCE was cleaned from MPV event flow */
 
+	if (is_mpv_slave)
+		mlx5_del_flow_rules(rx_roce->nic_master_rule);
 	mlx5_del_flow_rules(rx_roce->roce_miss.rule);
 	mlx5_del_flow_rules(rx_roce->rule);
+	if (is_mpv_slave)
+		roce_rx_mpv_destroy_tables(mdev, rx_roce);
 	mlx5_destroy_flow_table(rx_roce->ft_rdma);
 	mlx5_destroy_flow_group(rx_roce->roce_miss.group);
 	mlx5_destroy_flow_group(rx_roce->g);
 	mlx5_destroy_flow_table(rx_roce->ft);
+	rx_roce->ft = NULL;
 }
 
-#define MLX5_RX_ROCE_GROUP_SIZE BIT(0)
-
 int mlx5_ipsec_fs_roce_rx_create(struct mlx5_core_dev *mdev,
 				 struct mlx5_ipsec_fs *ipsec_roce,
 				 struct mlx5_flow_namespace *ns,
 				 struct mlx5_flow_destination *default_dst,
 				 u32 family, u32 level, u32 prio)
 {
+	bool is_mpv_slave = mlx5_core_is_mp_slave(mdev);
 	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_ipsec_rx_roce *roce;
 	struct mlx5_flow_table *ft;
@@ -298,18 +769,28 @@ int mlx5_ipsec_fs_roce_rx_create(struct mlx5_core_dev *mdev,
 	}
 	roce->roce_miss.group = g;
 
-	memset(&ft_attr, 0, sizeof(ft_attr));
-	if (family == AF_INET)
-		ft_attr.level = 1;
-	ft = mlx5_create_flow_table(roce->ns_rdma, &ft_attr);
-	if (IS_ERR(ft)) {
-		err = PTR_ERR(ft);
-		mlx5_core_err(mdev, "Fail to create RoCE IPsec rx ft at rdma err=%d\n", err);
-		goto fail_rdma_table;
+	if (is_mpv_slave) {
+		err = ipsec_fs_roce_rx_mpv_create(mdev, ipsec_roce, ns, family, level, prio);
+		if (err) {
+			mlx5_core_err(mdev, "Fail to create RoCE IPsec rx alias err=%d\n", err);
+			goto fail_mpv_create;
+		}
+	} else {
+		memset(&ft_attr, 0, sizeof(ft_attr));
+		if (family == AF_INET)
+			ft_attr.level = 1;
+		ft_attr.max_fte = 1;
+		ft = mlx5_create_flow_table(roce->ns_rdma, &ft_attr);
+		if (IS_ERR(ft)) {
+			err = PTR_ERR(ft);
+			mlx5_core_err(mdev,
+				      "Fail to create RoCE IPsec rx ft at rdma err=%d\n", err);
+			goto fail_rdma_table;
+		}
+
+		roce->ft_rdma = ft;
 	}
 
-	roce->ft_rdma = ft;
-
 	err = ipsec_fs_roce_rx_rule_setup(mdev, default_dst, roce);
 	if (err) {
 		mlx5_core_err(mdev, "Fail to create RoCE IPsec rx rules err=%d\n", err);
@@ -320,7 +801,10 @@ int mlx5_ipsec_fs_roce_rx_create(struct mlx5_core_dev *mdev,
 	return 0;
 
 fail_setup_rule:
+	if (is_mpv_slave)
+		roce_rx_mpv_destroy_tables(mdev, roce);
 	mlx5_destroy_flow_table(roce->ft_rdma);
+fail_mpv_create:
 fail_rdma_table:
 	mlx5_destroy_flow_group(roce->roce_miss.group);
 fail_mgroup:
@@ -332,12 +816,24 @@ fail_nomem:
 	return err;
 }
 
+bool mlx5_ipsec_fs_is_mpv_roce_supported(struct mlx5_core_dev *mdev)
+{
+	if (!mlx5_core_mp_enabled(mdev))
+		return true;
+
+	if (ipsec_fs_create_alias_supported_one(mdev))
+		return true;
+
+	return false;
+}
+
 void mlx5_ipsec_fs_roce_cleanup(struct mlx5_ipsec_fs *ipsec_roce)
 {
 	kfree(ipsec_roce);
 }
 
-struct mlx5_ipsec_fs *mlx5_ipsec_fs_roce_init(struct mlx5_core_dev *mdev)
+struct mlx5_ipsec_fs *mlx5_ipsec_fs_roce_init(struct mlx5_core_dev *mdev,
+					      struct mlx5_devcom_comp_dev **devcom)
 {
 	struct mlx5_ipsec_fs *roce_ipsec;
 	struct mlx5_flow_namespace *ns;
@@ -363,6 +859,8 @@ struct mlx5_ipsec_fs *mlx5_ipsec_fs_roce_init(struct mlx5_core_dev *mdev)
 
 	roce_ipsec->tx.ns = ns;
 
+	roce_ipsec->devcom = devcom;
+
 	return roce_ipsec;
 
 err_tx:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.h
index 9712d705fe..2a1af78309 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/ipsec_fs_roce.h
@@ -4,22 +4,28 @@
 #ifndef __MLX5_LIB_IPSEC_H__
 #define __MLX5_LIB_IPSEC_H__
 
+#include "lib/devcom.h"
+
 struct mlx5_ipsec_fs;
 
 struct mlx5_flow_table *
 mlx5_ipsec_fs_roce_ft_get(struct mlx5_ipsec_fs *ipsec_roce, u32 family);
 void mlx5_ipsec_fs_roce_rx_destroy(struct mlx5_ipsec_fs *ipsec_roce,
-				   u32 family);
+				   u32 family, struct mlx5_core_dev *mdev);
 int mlx5_ipsec_fs_roce_rx_create(struct mlx5_core_dev *mdev,
 				 struct mlx5_ipsec_fs *ipsec_roce,
 				 struct mlx5_flow_namespace *ns,
 				 struct mlx5_flow_destination *default_dst,
 				 u32 family, u32 level, u32 prio);
-void mlx5_ipsec_fs_roce_tx_destroy(struct mlx5_ipsec_fs *ipsec_roce);
+void mlx5_ipsec_fs_roce_tx_destroy(struct mlx5_ipsec_fs *ipsec_roce,
+				   struct mlx5_core_dev *mdev);
 int mlx5_ipsec_fs_roce_tx_create(struct mlx5_core_dev *mdev,
 				 struct mlx5_ipsec_fs *ipsec_roce,
-				 struct mlx5_flow_table *pol_ft);
+				 struct mlx5_flow_table *pol_ft,
+				 bool from_event);
 void mlx5_ipsec_fs_roce_cleanup(struct mlx5_ipsec_fs *ipsec_roce);
-struct mlx5_ipsec_fs *mlx5_ipsec_fs_roce_init(struct mlx5_core_dev *mdev);
+struct mlx5_ipsec_fs *mlx5_ipsec_fs_roce_init(struct mlx5_core_dev *mdev,
+					      struct mlx5_devcom_comp_dev **devcom);
+bool mlx5_ipsec_fs_is_mpv_roce_supported(struct mlx5_core_dev *mdev);
 
 #endif /* __MLX5_LIB_IPSEC_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 6ca91c0e8a..a17152c1cb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -73,6 +73,7 @@
 #include "sf/sf.h"
 #include "mlx5_irq.h"
 #include "hwmon.h"
+#include "lag/lag.h"
 
 MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
 MODULE_DESCRIPTION("Mellanox 5th generation network adapters (ConnectX series) core driver");
@@ -952,6 +953,27 @@ static void mlx5_pci_close(struct mlx5_core_dev *dev)
 	mlx5_pci_disable_device(dev);
 }
 
+static void mlx5_register_hca_devcom_comp(struct mlx5_core_dev *dev)
+{
+	/* This component is use to sync adding core_dev to lag_dev and to sync
+	 * changes of mlx5_adev_devices between LAG layer and other layers.
+	 */
+	if (!mlx5_lag_is_supported(dev))
+		return;
+
+	dev->priv.hca_devcom_comp =
+		mlx5_devcom_register_component(dev->priv.devc, MLX5_DEVCOM_HCA_PORTS,
+					       mlx5_query_nic_system_image_guid(dev),
+					       NULL, dev);
+	if (IS_ERR_OR_NULL(dev->priv.hca_devcom_comp))
+		mlx5_core_err(dev, "Failed to register devcom HCA component\n");
+}
+
+static void mlx5_unregister_hca_devcom_comp(struct mlx5_core_dev *dev)
+{
+	mlx5_devcom_unregister_component(dev->priv.hca_devcom_comp);
+}
+
 static int mlx5_init_once(struct mlx5_core_dev *dev)
 {
 	int err;
@@ -960,6 +982,7 @@ static int mlx5_init_once(struct mlx5_core_dev *dev)
 	if (IS_ERR(dev->priv.devc))
 		mlx5_core_warn(dev, "failed to register devcom device %ld\n",
 			       PTR_ERR(dev->priv.devc));
+	mlx5_register_hca_devcom_comp(dev);
 
 	err = mlx5_query_board_id(dev);
 	if (err) {
@@ -1094,6 +1117,7 @@ err_eq_cleanup:
 err_irq_cleanup:
 	mlx5_irq_table_cleanup(dev);
 err_devcom:
+	mlx5_unregister_hca_devcom_comp(dev);
 	mlx5_devcom_unregister_device(dev->priv.devc);
 
 	return err;
@@ -1123,6 +1147,7 @@ static void mlx5_cleanup_once(struct mlx5_core_dev *dev)
 	mlx5_events_cleanup(dev);
 	mlx5_eq_table_cleanup(dev);
 	mlx5_irq_table_cleanup(dev);
+	mlx5_unregister_hca_devcom_comp(dev);
 	mlx5_devcom_unregister_device(dev->priv.devc);
 }
 
@@ -1411,9 +1436,9 @@ err_irq_table:
 
 static void mlx5_unload(struct mlx5_core_dev *dev)
 {
+	mlx5_eswitch_disable(dev->priv.eswitch);
 	mlx5_devlink_traps_unregister(priv_to_devlink(dev));
 	mlx5_sf_dev_table_destroy(dev);
-	mlx5_eswitch_disable(dev->priv.eswitch);
 	mlx5_sriov_detach(dev);
 	mlx5_lag_remove_mdev(dev);
 	mlx5_ec_cleanup(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index 124352459c..6b14e347d9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -41,6 +41,7 @@
 #include <linux/mlx5/cq.h>
 #include <linux/mlx5/fs.h>
 #include <linux/mlx5/driver.h>
+#include "lib/devcom.h"
 
 extern uint mlx5_core_debug_mask;
 
@@ -97,6 +98,22 @@ do {								\
 			     __func__, __LINE__, current->pid,	\
 			     ##__VA_ARGS__)
 
+#define ACCESS_KEY_LEN  32
+#define FT_ID_FT_TYPE_OFFSET 24
+
+struct mlx5_cmd_allow_other_vhca_access_attr {
+	u16 obj_type;
+	u32 obj_id;
+	u8 access_key[ACCESS_KEY_LEN];
+};
+
+struct mlx5_cmd_alias_obj_create_attr {
+	u32 obj_id;
+	u16 vhca_id;
+	u16 obj_type;
+	u8 access_key[ACCESS_KEY_LEN];
+};
+
 static inline void mlx5_printk(struct mlx5_core_dev *dev, int level, const char *format, ...)
 {
 	struct device *device = dev->device;
@@ -143,6 +160,8 @@ enum mlx5_semaphore_space_address {
 
 #define MLX5_DEFAULT_PROF       2
 #define MLX5_SF_PROF		3
+#define MLX5_NUM_FW_CMD_THREADS 8
+#define MLX5_DEV_MAX_WQS	MLX5_NUM_FW_CMD_THREADS
 
 static inline int mlx5_flexible_inlen(struct mlx5_core_dev *dev, size_t fixed,
 				      size_t item_size, size_t num_items,
@@ -248,10 +267,6 @@ int mlx5_register_device(struct mlx5_core_dev *dev);
 void mlx5_unregister_device(struct mlx5_core_dev *dev);
 void mlx5_dev_set_lightweight(struct mlx5_core_dev *dev);
 bool mlx5_dev_is_lightweight(struct mlx5_core_dev *dev);
-struct mlx5_core_dev *mlx5_get_next_phys_dev_lag(struct mlx5_core_dev *dev);
-void mlx5_dev_list_lock(void);
-void mlx5_dev_list_unlock(void);
-int mlx5_dev_list_trylock(void);
 
 void mlx5_fw_reporters_create(struct mlx5_core_dev *dev);
 int mlx5_query_mtpps(struct mlx5_core_dev *dev, u32 *mtpps, u32 mtpps_size);
@@ -290,14 +305,12 @@ static inline int mlx5_rescan_drivers(struct mlx5_core_dev *dev)
 {
 	int ret;
 
-	mlx5_dev_list_lock();
+	mlx5_devcom_comp_lock(dev->priv.hca_devcom_comp);
 	ret = mlx5_rescan_drivers_locked(dev);
-	mlx5_dev_list_unlock();
+	mlx5_devcom_comp_unlock(dev->priv.hca_devcom_comp);
 	return ret;
 }
 
-void mlx5_lag_update(struct mlx5_core_dev *dev);
-
 enum {
 	MLX5_NIC_IFC_FULL		= 0,
 	MLX5_NIC_IFC_DISABLED		= 1,
@@ -331,7 +344,6 @@ int mlx5_vport_set_other_func_cap(struct mlx5_core_dev *dev, const void *hca_cap
 #define mlx5_vport_get_other_func_general_cap(dev, vport, out)		\
 	mlx5_vport_get_other_func_cap(dev, vport, out, MLX5_CAP_GENERAL)
 
-void mlx5_events_work_enqueue(struct mlx5_core_dev *dev, struct work_struct *work);
 static inline u32 mlx5_sriov_get_vf_total_msix(struct pci_dev *pdev)
 {
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
@@ -343,6 +355,12 @@ bool mlx5_eth_supported(struct mlx5_core_dev *dev);
 bool mlx5_rdma_supported(struct mlx5_core_dev *dev);
 bool mlx5_vnet_supported(struct mlx5_core_dev *dev);
 bool mlx5_same_hw_devs(struct mlx5_core_dev *dev, struct mlx5_core_dev *peer_dev);
+int mlx5_cmd_allow_other_vhca_access(struct mlx5_core_dev *dev,
+				     struct mlx5_cmd_allow_other_vhca_access_attr *attr);
+int mlx5_cmd_alias_obj_create(struct mlx5_core_dev *dev,
+			      struct mlx5_cmd_alias_obj_create_attr *alias_attr,
+			      u32 *obj_id);
+int mlx5_cmd_alias_obj_destroy(struct mlx5_core_dev *dev, u32 obj_id, u16 obj_type);
 
 static inline u16 mlx5_core_ec_vf_vport_base(const struct mlx5_core_dev *dev)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index be70d1f23a..7d8c732818 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -1098,10 +1098,11 @@ static const u32 mlx5e_ext_link_speed[MLX5E_EXT_LINK_MODES_NUMBER] = {
 	[MLX5E_CAUI_4_100GBASE_CR4_KR4] = 100000,
 	[MLX5E_100GAUI_2_100GBASE_CR2_KR2] = 100000,
 	[MLX5E_200GAUI_4_200GBASE_CR4_KR4] = 200000,
-	[MLX5E_400GAUI_8] = 400000,
+	[MLX5E_400GAUI_8_400GBASE_CR8] = 400000,
 	[MLX5E_100GAUI_1_100GBASE_CR_KR] = 100000,
 	[MLX5E_200GAUI_2_200GBASE_CR2_KR2] = 200000,
 	[MLX5E_400GAUI_4_400GBASE_CR4_KR4] = 400000,
+	[MLX5E_800GAUI_8_800GBASE_CR8_KR8] = 800000,
 };
 
 int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
index 05e148db98..c93492b677 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
@@ -14,17 +14,22 @@
 
 struct mlx5_sf_dev_table {
 	struct xarray devices;
-	unsigned int max_sfs;
 	phys_addr_t base_address;
 	u64 sf_bar_length;
 	struct notifier_block nb;
-	struct mutex table_lock; /* Serializes sf life cycle and vhca state change handler */
 	struct workqueue_struct *active_wq;
 	struct work_struct work;
 	u8 stop_active_wq:1;
 	struct mlx5_core_dev *dev;
 };
 
+struct mlx5_sf_dev_active_work_ctx {
+	struct work_struct work;
+	struct mlx5_vhca_state_event event;
+	struct mlx5_sf_dev_table *table;
+	int sf_index;
+};
+
 static bool mlx5_sf_dev_supported(const struct mlx5_core_dev *dev)
 {
 	return MLX5_CAP_GEN(dev, sf) && mlx5_vhca_event_supported(dev);
@@ -110,12 +115,6 @@ static void mlx5_sf_dev_add(struct mlx5_core_dev *dev, u16 sf_index, u16 fn_id,
 	sf_dev->parent_mdev = dev;
 	sf_dev->fn_id = fn_id;
 
-	if (!table->max_sfs) {
-		mlx5_adev_idx_free(id);
-		kfree(sf_dev);
-		err = -EOPNOTSUPP;
-		goto add_err;
-	}
 	sf_dev->bar_base_addr = table->base_address + (sf_index * table->sf_bar_length);
 
 	trace_mlx5_sf_dev_add(dev, sf_dev, id);
@@ -172,7 +171,6 @@ mlx5_sf_dev_state_change_handler(struct notifier_block *nb, unsigned long event_
 		return 0;
 
 	sf_index = event->function_id - base_id;
-	mutex_lock(&table->table_lock);
 	sf_dev = xa_load(&table->devices, sf_index);
 	switch (event->new_vhca_state) {
 	case MLX5_VHCA_STATE_INVALID:
@@ -196,7 +194,6 @@ mlx5_sf_dev_state_change_handler(struct notifier_block *nb, unsigned long event_
 	default:
 		break;
 	}
-	mutex_unlock(&table->table_lock);
 	return 0;
 }
 
@@ -221,15 +218,44 @@ static int mlx5_sf_dev_vhca_arm_all(struct mlx5_sf_dev_table *table)
 	return 0;
 }
 
-static void mlx5_sf_dev_add_active_work(struct work_struct *work)
+static void mlx5_sf_dev_add_active_work(struct work_struct *_work)
 {
-	struct mlx5_sf_dev_table *table = container_of(work, struct mlx5_sf_dev_table, work);
+	struct mlx5_sf_dev_active_work_ctx *work_ctx;
+
+	work_ctx = container_of(_work, struct mlx5_sf_dev_active_work_ctx, work);
+	if (work_ctx->table->stop_active_wq)
+		goto out;
+	/* Don't probe device which is already probe */
+	if (!xa_load(&work_ctx->table->devices, work_ctx->sf_index))
+		mlx5_sf_dev_add(work_ctx->table->dev, work_ctx->sf_index,
+				work_ctx->event.function_id, work_ctx->event.sw_function_id);
+	/* There is a race where SF got inactive after the query
+	 * above. e.g.: the query returns that the state of the
+	 * SF is active, and after that the eswitch manager set it to
+	 * inactive.
+	 * This case cannot be managed in SW, since the probing of the
+	 * SF is on one system, and the inactivation is on a different
+	 * system.
+	 * If the inactive is done after the SF perform init_hca(),
+	 * the SF will fully probe and then removed. If it was
+	 * done before init_hca(), the SF probe will fail.
+	 */
+out:
+	kfree(work_ctx);
+}
+
+/* In case SFs are generated externally, probe active SFs */
+static void mlx5_sf_dev_queue_active_works(struct work_struct *_work)
+{
+	struct mlx5_sf_dev_table *table = container_of(_work, struct mlx5_sf_dev_table, work);
 	u32 out[MLX5_ST_SZ_DW(query_vhca_state_out)] = {};
+	struct mlx5_sf_dev_active_work_ctx *work_ctx;
 	struct mlx5_core_dev *dev = table->dev;
 	u16 max_functions;
 	u16 function_id;
 	u16 sw_func_id;
 	int err = 0;
+	int wq_idx;
 	u8 state;
 	int i;
 
@@ -249,27 +275,22 @@ static void mlx5_sf_dev_add_active_work(struct work_struct *work)
 			continue;
 
 		sw_func_id = MLX5_GET(query_vhca_state_out, out, vhca_state_context.sw_function_id);
-		mutex_lock(&table->table_lock);
-		/* Don't probe device which is already probe */
-		if (!xa_load(&table->devices, i))
-			mlx5_sf_dev_add(dev, i, function_id, sw_func_id);
-		/* There is a race where SF got inactive after the query
-		 * above. e.g.: the query returns that the state of the
-		 * SF is active, and after that the eswitch manager set it to
-		 * inactive.
-		 * This case cannot be managed in SW, since the probing of the
-		 * SF is on one system, and the inactivation is on a different
-		 * system.
-		 * If the inactive is done after the SF perform init_hca(),
-		 * the SF will fully probe and then removed. If it was
-		 * done before init_hca(), the SF probe will fail.
-		 */
-		mutex_unlock(&table->table_lock);
+		work_ctx = kzalloc(sizeof(*work_ctx), GFP_KERNEL);
+		if (!work_ctx)
+			return;
+
+		INIT_WORK(&work_ctx->work, &mlx5_sf_dev_add_active_work);
+		work_ctx->event.function_id = function_id;
+		work_ctx->event.sw_function_id = sw_func_id;
+		work_ctx->table = table;
+		work_ctx->sf_index = i;
+		wq_idx = work_ctx->event.function_id % MLX5_DEV_MAX_WQS;
+		mlx5_vhca_events_work_enqueue(dev, wq_idx, &work_ctx->work);
 	}
 }
 
 /* In case SFs are generated externally, probe active SFs */
-static int mlx5_sf_dev_queue_active_work(struct mlx5_sf_dev_table *table)
+static int mlx5_sf_dev_create_active_works(struct mlx5_sf_dev_table *table)
 {
 	if (MLX5_CAP_GEN(table->dev, eswitch_manager))
 		return 0; /* the table is local */
@@ -280,12 +301,12 @@ static int mlx5_sf_dev_queue_active_work(struct mlx5_sf_dev_table *table)
 	table->active_wq = create_singlethread_workqueue("mlx5_active_sf");
 	if (!table->active_wq)
 		return -ENOMEM;
-	INIT_WORK(&table->work, &mlx5_sf_dev_add_active_work);
+	INIT_WORK(&table->work, &mlx5_sf_dev_queue_active_works);
 	queue_work(table->active_wq, &table->work);
 	return 0;
 }
 
-static void mlx5_sf_dev_destroy_active_work(struct mlx5_sf_dev_table *table)
+static void mlx5_sf_dev_destroy_active_works(struct mlx5_sf_dev_table *table)
 {
 	if (table->active_wq) {
 		table->stop_active_wq = true;
@@ -296,7 +317,6 @@ static void mlx5_sf_dev_destroy_active_work(struct mlx5_sf_dev_table *table)
 void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 {
 	struct mlx5_sf_dev_table *table;
-	unsigned int max_sfs;
 	int err;
 
 	if (!mlx5_sf_dev_supported(dev))
@@ -310,37 +330,30 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 
 	table->nb.notifier_call = mlx5_sf_dev_state_change_handler;
 	table->dev = dev;
-	if (MLX5_CAP_GEN(dev, max_num_sf))
-		max_sfs = MLX5_CAP_GEN(dev, max_num_sf);
-	else
-		max_sfs = 1 << MLX5_CAP_GEN(dev, log_max_sf);
 	table->sf_bar_length = 1 << (MLX5_CAP_GEN(dev, log_min_sf_size) + 12);
 	table->base_address = pci_resource_start(dev->pdev, 2);
-	table->max_sfs = max_sfs;
 	xa_init(&table->devices);
-	mutex_init(&table->table_lock);
 	dev->priv.sf_dev_table = table;
 
 	err = mlx5_vhca_event_notifier_register(dev, &table->nb);
 	if (err)
 		goto vhca_err;
 
-	err = mlx5_sf_dev_queue_active_work(table);
+	err = mlx5_sf_dev_create_active_works(table);
 	if (err)
 		goto add_active_err;
 
 	err = mlx5_sf_dev_vhca_arm_all(table);
 	if (err)
 		goto arm_err;
-	mlx5_core_dbg(dev, "SF DEV: max sf devices=%d\n", max_sfs);
 	return;
 
 arm_err:
-	mlx5_sf_dev_destroy_active_work(table);
+	mlx5_sf_dev_destroy_active_works(table);
 add_active_err:
 	mlx5_vhca_event_notifier_unregister(dev, &table->nb);
+	mlx5_vhca_event_work_queues_flush(dev);
 vhca_err:
-	table->max_sfs = 0;
 	kfree(table);
 	dev->priv.sf_dev_table = NULL;
 table_err:
@@ -365,9 +378,9 @@ void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev)
 	if (!table)
 		return;
 
-	mlx5_sf_dev_destroy_active_work(table);
+	mlx5_sf_dev_destroy_active_works(table);
 	mlx5_vhca_event_notifier_unregister(dev, &table->nb);
-	mutex_destroy(&table->table_lock);
+	mlx5_vhca_event_work_queues_flush(dev);
 
 	/* Now that event handler is not running, it is safe to destroy
 	 * the sf device without race.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h
index 2a66a427ef..b99131e95e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h
@@ -19,6 +19,12 @@ struct mlx5_sf_dev {
 	u16 fn_id;
 };
 
+struct mlx5_sf_peer_devlink_event_ctx {
+	u16 fn_id;
+	struct devlink *devlink;
+	int err;
+};
+
 void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev);
 void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
index 8fe82f1191..169c2c68ed 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
@@ -8,6 +8,20 @@
 #include "dev.h"
 #include "devlink.h"
 
+static int mlx5_core_peer_devlink_set(struct mlx5_sf_dev *sf_dev, struct devlink *devlink)
+{
+	struct mlx5_sf_peer_devlink_event_ctx event_ctx = {
+		.fn_id = sf_dev->fn_id,
+		.devlink = devlink,
+	};
+	int ret;
+
+	ret = mlx5_blocking_notifier_call_chain(sf_dev->parent_mdev,
+						MLX5_DRIVER_EVENT_SF_PEER_DEVLINK,
+						&event_ctx);
+	return ret == NOTIFY_OK ? event_ctx.err : 0;
+}
+
 static int mlx5_sf_dev_probe(struct auxiliary_device *adev, const struct auxiliary_device_id *id)
 {
 	struct mlx5_sf_dev *sf_dev = container_of(adev, struct mlx5_sf_dev, adev);
@@ -54,9 +68,21 @@ static int mlx5_sf_dev_probe(struct auxiliary_device *adev, const struct auxilia
 		mlx5_core_warn(mdev, "mlx5_init_one err=%d\n", err);
 		goto init_one_err;
 	}
+
+	err = mlx5_core_peer_devlink_set(sf_dev, devlink);
+	if (err) {
+		mlx5_core_warn(mdev, "mlx5_core_peer_devlink_set err=%d\n", err);
+		goto peer_devlink_set_err;
+	}
+
 	devlink_register(devlink);
 	return 0;
 
+peer_devlink_set_err:
+	if (mlx5_dev_is_lightweight(sf_dev->mdev))
+		mlx5_uninit_one_light(sf_dev->mdev);
+	else
+		mlx5_uninit_one(sf_dev->mdev);
 init_one_err:
 	iounmap(mdev->iseg);
 remap_err:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index e34a8f88c5..6c11e075ca 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -20,43 +20,36 @@ struct mlx5_sf {
 	u16 hw_state;
 };
 
+static void *mlx5_sf_by_dl_port(struct devlink_port *dl_port)
+{
+	struct mlx5_devlink_port *mlx5_dl_port = mlx5_devlink_port_get(dl_port);
+
+	return container_of(mlx5_dl_port, struct mlx5_sf, dl_port);
+}
+
 struct mlx5_sf_table {
 	struct mlx5_core_dev *dev; /* To refer from notifier context. */
-	struct xarray port_indices; /* port index based lookup. */
-	refcount_t refcount;
-	struct completion disable_complete;
+	struct xarray function_ids; /* function id based lookup. */
 	struct mutex sf_state_lock; /* Serializes sf state among user cmds & vhca event handler. */
 	struct notifier_block esw_nb;
 	struct notifier_block vhca_nb;
+	struct notifier_block mdev_nb;
 };
 
 static struct mlx5_sf *
-mlx5_sf_lookup_by_index(struct mlx5_sf_table *table, unsigned int port_index)
-{
-	return xa_load(&table->port_indices, port_index);
-}
-
-static struct mlx5_sf *
 mlx5_sf_lookup_by_function_id(struct mlx5_sf_table *table, unsigned int fn_id)
 {
-	unsigned long index;
-	struct mlx5_sf *sf;
-
-	xa_for_each(&table->port_indices, index, sf) {
-		if (sf->hw_fn_id == fn_id)
-			return sf;
-	}
-	return NULL;
+	return xa_load(&table->function_ids, fn_id);
 }
 
-static int mlx5_sf_id_insert(struct mlx5_sf_table *table, struct mlx5_sf *sf)
+static int mlx5_sf_function_id_insert(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
-	return xa_insert(&table->port_indices, sf->port_index, sf, GFP_KERNEL);
+	return xa_insert(&table->function_ids, sf->hw_fn_id, sf, GFP_KERNEL);
 }
 
-static void mlx5_sf_id_erase(struct mlx5_sf_table *table, struct mlx5_sf *sf)
+static void mlx5_sf_function_id_erase(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
-	xa_erase(&table->port_indices, sf->port_index);
+	xa_erase(&table->function_ids, sf->hw_fn_id);
 }
 
 static struct mlx5_sf *
@@ -93,7 +86,7 @@ mlx5_sf_alloc(struct mlx5_sf_table *table, struct mlx5_eswitch *esw,
 	sf->hw_state = MLX5_VHCA_STATE_ALLOCATED;
 	sf->controller = controller;
 
-	err = mlx5_sf_id_insert(table, sf);
+	err = mlx5_sf_function_id_insert(table, sf);
 	if (err)
 		goto insert_err;
 
@@ -111,28 +104,11 @@ id_err:
 
 static void mlx5_sf_free(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
-	mlx5_sf_id_erase(table, sf);
 	mlx5_sf_hw_table_sf_free(table->dev, sf->controller, sf->id);
 	trace_mlx5_sf_free(table->dev, sf->port_index, sf->controller, sf->hw_fn_id);
 	kfree(sf);
 }
 
-static struct mlx5_sf_table *mlx5_sf_table_try_get(struct mlx5_core_dev *dev)
-{
-	struct mlx5_sf_table *table = dev->priv.sf_table;
-
-	if (!table)
-		return NULL;
-
-	return refcount_inc_not_zero(&table->refcount) ? table : NULL;
-}
-
-static void mlx5_sf_table_put(struct mlx5_sf_table *table)
-{
-	if (refcount_dec_and_test(&table->refcount))
-		complete(&table->disable_complete);
-}
-
 static enum devlink_port_fn_state mlx5_sf_to_devlink_state(u8 hw_state)
 {
 	switch (hw_state) {
@@ -172,26 +148,14 @@ int mlx5_devlink_sf_port_fn_state_get(struct devlink_port *dl_port,
 				      struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(dl_port->devlink);
-	struct mlx5_sf_table *table;
-	struct mlx5_sf *sf;
-	int err = 0;
-
-	table = mlx5_sf_table_try_get(dev);
-	if (!table)
-		return -EOPNOTSUPP;
+	struct mlx5_sf_table *table = dev->priv.sf_table;
+	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
 
-	sf = mlx5_sf_lookup_by_index(table, dl_port->index);
-	if (!sf) {
-		err = -EOPNOTSUPP;
-		goto sf_err;
-	}
 	mutex_lock(&table->sf_state_lock);
 	*state = mlx5_sf_to_devlink_state(sf->hw_state);
 	*opstate = mlx5_sf_to_devlink_opstate(sf->hw_state);
 	mutex_unlock(&table->sf_state_lock);
-sf_err:
-	mlx5_sf_table_put(table);
-	return err;
+	return 0;
 }
 
 static int mlx5_sf_activate(struct mlx5_core_dev *dev, struct mlx5_sf *sf,
@@ -257,26 +221,10 @@ int mlx5_devlink_sf_port_fn_state_set(struct devlink_port *dl_port,
 				      struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(dl_port->devlink);
-	struct mlx5_sf_table *table;
-	struct mlx5_sf *sf;
-	int err;
-
-	table = mlx5_sf_table_try_get(dev);
-	if (!table) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "Port state set is only supported in eswitch switchdev mode or SF ports are disabled.");
-		return -EOPNOTSUPP;
-	}
-	sf = mlx5_sf_lookup_by_index(table, dl_port->index);
-	if (!sf) {
-		err = -ENODEV;
-		goto out;
-	}
+	struct mlx5_sf_table *table = dev->priv.sf_table;
+	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
 
-	err = mlx5_sf_state_set(dev, table, sf, state, extack);
-out:
-	mlx5_sf_table_put(table);
-	return err;
+	return mlx5_sf_state_set(dev, table, sf, state, extack);
 }
 
 static int mlx5_sf_add(struct mlx5_core_dev *dev, struct mlx5_sf_table *table,
@@ -335,32 +283,45 @@ mlx5_sf_new_check_attr(struct mlx5_core_dev *dev, const struct devlink_port_new_
 	return 0;
 }
 
+static bool mlx5_sf_table_supported(const struct mlx5_core_dev *dev)
+{
+	return dev->priv.eswitch && MLX5_ESWITCH_MANAGER(dev) &&
+	       mlx5_sf_hw_table_supported(dev);
+}
+
 int mlx5_devlink_sf_port_new(struct devlink *devlink,
 			     const struct devlink_port_new_attrs *new_attr,
 			     struct netlink_ext_ack *extack,
 			     struct devlink_port **dl_port)
 {
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
-	struct mlx5_sf_table *table;
+	struct mlx5_sf_table *table = dev->priv.sf_table;
 	int err;
 
 	err = mlx5_sf_new_check_attr(dev, new_attr, extack);
 	if (err)
 		return err;
 
-	table = mlx5_sf_table_try_get(dev);
-	if (!table) {
+	if (!mlx5_sf_table_supported(dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "SF ports are not supported.");
+		return -EOPNOTSUPP;
+	}
+
+	if (!is_mdev_switchdev_mode(dev)) {
 		NL_SET_ERR_MSG_MOD(extack,
-				   "Port add is only supported in eswitch switchdev mode or SF ports are disabled.");
+				   "SF ports are only supported in eswitch switchdev mode.");
 		return -EOPNOTSUPP;
 	}
-	err = mlx5_sf_add(dev, table, new_attr, extack, dl_port);
-	mlx5_sf_table_put(table);
-	return err;
+
+	return mlx5_sf_add(dev, table, new_attr, extack, dl_port);
 }
 
 static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
+	mutex_lock(&table->sf_state_lock);
+
+	mlx5_sf_function_id_erase(table, sf);
+
 	if (sf->hw_state == MLX5_VHCA_STATE_ALLOCATED) {
 		mlx5_sf_free(table, sf);
 	} else if (mlx5_sf_is_active(sf)) {
@@ -376,6 +337,16 @@ static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 		mlx5_sf_hw_table_sf_deferred_free(table->dev, sf->controller, sf->id);
 		kfree(sf);
 	}
+
+	mutex_unlock(&table->sf_state_lock);
+}
+
+static void mlx5_sf_del(struct mlx5_sf_table *table, struct mlx5_sf *sf)
+{
+	struct mlx5_eswitch *esw = table->dev->priv.eswitch;
+
+	mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
+	mlx5_sf_dealloc(table, sf);
 }
 
 int mlx5_devlink_sf_port_del(struct devlink *devlink,
@@ -383,32 +354,11 @@ int mlx5_devlink_sf_port_del(struct devlink *devlink,
 			     struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
-	struct mlx5_eswitch *esw = dev->priv.eswitch;
-	struct mlx5_sf_table *table;
-	struct mlx5_sf *sf;
-	int err = 0;
-
-	table = mlx5_sf_table_try_get(dev);
-	if (!table) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "Port del is only supported in eswitch switchdev mode or SF ports are disabled.");
-		return -EOPNOTSUPP;
-	}
-	sf = mlx5_sf_lookup_by_index(table, dl_port->index);
-	if (!sf) {
-		err = -ENODEV;
-		goto sf_err;
-	}
-
-	mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
-	mlx5_sf_id_erase(table, sf);
+	struct mlx5_sf_table *table = dev->priv.sf_table;
+	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
 
-	mutex_lock(&table->sf_state_lock);
-	mlx5_sf_dealloc(table, sf);
-	mutex_unlock(&table->sf_state_lock);
-sf_err:
-	mlx5_sf_table_put(table);
-	return err;
+	mlx5_sf_del(table, sf);
+	return 0;
 }
 
 static bool mlx5_sf_state_update_check(const struct mlx5_sf *sf, u8 new_state)
@@ -433,14 +383,10 @@ static int mlx5_sf_vhca_event(struct notifier_block *nb, unsigned long opcode, v
 	bool update = false;
 	struct mlx5_sf *sf;
 
-	table = mlx5_sf_table_try_get(table->dev);
-	if (!table)
-		return 0;
-
 	mutex_lock(&table->sf_state_lock);
 	sf = mlx5_sf_lookup_by_function_id(table, event->function_id);
 	if (!sf)
-		goto sf_err;
+		goto unlock;
 
 	/* When driver is attached or detached to a function, an event
 	 * notifies such state change.
@@ -450,46 +396,18 @@ static int mlx5_sf_vhca_event(struct notifier_block *nb, unsigned long opcode, v
 		sf->hw_state = event->new_vhca_state;
 	trace_mlx5_sf_update_state(table->dev, sf->port_index, sf->controller,
 				   sf->hw_fn_id, sf->hw_state);
-sf_err:
+unlock:
 	mutex_unlock(&table->sf_state_lock);
-	mlx5_sf_table_put(table);
 	return 0;
 }
 
-static void mlx5_sf_table_enable(struct mlx5_sf_table *table)
-{
-	init_completion(&table->disable_complete);
-	refcount_set(&table->refcount, 1);
-}
-
-static void mlx5_sf_deactivate_all(struct mlx5_sf_table *table)
+static void mlx5_sf_del_all(struct mlx5_sf_table *table)
 {
-	struct mlx5_eswitch *esw = table->dev->priv.eswitch;
 	unsigned long index;
 	struct mlx5_sf *sf;
 
-	/* At this point, no new user commands can start and no vhca event can
-	 * arrive. It is safe to destroy all user created SFs.
-	 */
-	xa_for_each(&table->port_indices, index, sf) {
-		mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
-		mlx5_sf_id_erase(table, sf);
-		mlx5_sf_dealloc(table, sf);
-	}
-}
-
-static void mlx5_sf_table_disable(struct mlx5_sf_table *table)
-{
-	if (!refcount_read(&table->refcount))
-		return;
-
-	/* Balances with refcount_set; drop the reference so that new user cmd cannot start
-	 * and new vhca event handler cannot run.
-	 */
-	mlx5_sf_table_put(table);
-	wait_for_completion(&table->disable_complete);
-
-	mlx5_sf_deactivate_all(table);
+	xa_for_each(&table->function_ids, index, sf)
+		mlx5_sf_del(table, sf);
 }
 
 static int mlx5_sf_esw_event(struct notifier_block *nb, unsigned long event, void *data)
@@ -498,11 +416,8 @@ static int mlx5_sf_esw_event(struct notifier_block *nb, unsigned long event, voi
 	const struct mlx5_esw_event_info *mode = data;
 
 	switch (mode->new_mode) {
-	case MLX5_ESWITCH_OFFLOADS:
-		mlx5_sf_table_enable(table);
-		break;
 	case MLX5_ESWITCH_LEGACY:
-		mlx5_sf_table_disable(table);
+		mlx5_sf_del_all(table);
 		break;
 	default:
 		break;
@@ -511,10 +426,29 @@ static int mlx5_sf_esw_event(struct notifier_block *nb, unsigned long event, voi
 	return 0;
 }
 
-static bool mlx5_sf_table_supported(const struct mlx5_core_dev *dev)
+static int mlx5_sf_mdev_event(struct notifier_block *nb, unsigned long event, void *data)
 {
-	return dev->priv.eswitch && MLX5_ESWITCH_MANAGER(dev) &&
-	       mlx5_sf_hw_table_supported(dev);
+	struct mlx5_sf_table *table = container_of(nb, struct mlx5_sf_table, mdev_nb);
+	struct mlx5_sf_peer_devlink_event_ctx *event_ctx = data;
+	int ret = NOTIFY_DONE;
+	struct mlx5_sf *sf;
+
+	if (event != MLX5_DRIVER_EVENT_SF_PEER_DEVLINK)
+		return NOTIFY_DONE;
+
+
+	mutex_lock(&table->sf_state_lock);
+	sf = mlx5_sf_lookup_by_function_id(table, event_ctx->fn_id);
+	if (!sf)
+		goto out;
+
+	event_ctx->err = devl_port_fn_devlink_set(&sf->dl_port.dl_port,
+						  event_ctx->devlink);
+
+	ret = NOTIFY_OK;
+out:
+	mutex_unlock(&table->sf_state_lock);
+	return ret;
 }
 
 int mlx5_sf_table_init(struct mlx5_core_dev *dev)
@@ -531,9 +465,8 @@ int mlx5_sf_table_init(struct mlx5_core_dev *dev)
 
 	mutex_init(&table->sf_state_lock);
 	table->dev = dev;
-	xa_init(&table->port_indices);
+	xa_init(&table->function_ids);
 	dev->priv.sf_table = table;
-	refcount_set(&table->refcount, 0);
 	table->esw_nb.notifier_call = mlx5_sf_esw_event;
 	err = mlx5_esw_event_notifier_register(dev->priv.eswitch, &table->esw_nb);
 	if (err)
@@ -544,6 +477,9 @@ int mlx5_sf_table_init(struct mlx5_core_dev *dev)
 	if (err)
 		goto vhca_err;
 
+	table->mdev_nb.notifier_call = mlx5_sf_mdev_event;
+	mlx5_blocking_notifier_register(dev, &table->mdev_nb);
+
 	return 0;
 
 vhca_err:
@@ -562,10 +498,10 @@ void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev)
 	if (!table)
 		return;
 
+	mlx5_blocking_notifier_unregister(dev, &table->mdev_nb);
 	mlx5_vhca_event_notifier_unregister(table->dev, &table->vhca_nb);
 	mlx5_esw_event_notifier_unregister(dev->priv.eswitch, &table->esw_nb);
-	WARN_ON(refcount_read(&table->refcount));
 	mutex_destroy(&table->sf_state_lock);
-	WARN_ON(!xa_empty(&table->port_indices));
+	WARN_ON(!xa_empty(&table->function_ids));
 	kfree(table);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c
index d908fba968..cda01ba441 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c
@@ -21,6 +21,15 @@ struct mlx5_vhca_event_work {
 	struct mlx5_vhca_state_event event;
 };
 
+struct mlx5_vhca_event_handler {
+	struct workqueue_struct *wq;
+};
+
+struct mlx5_vhca_events {
+	struct mlx5_core_dev *dev;
+	struct mlx5_vhca_event_handler handler[MLX5_DEV_MAX_WQS];
+};
+
 int mlx5_cmd_query_vhca_state(struct mlx5_core_dev *dev, u16 function_id, u32 *out, u32 outlen)
 {
 	u32 in[MLX5_ST_SZ_DW(query_vhca_state_in)] = {};
@@ -99,6 +108,11 @@ static void mlx5_vhca_state_work_handler(struct work_struct *_work)
 	kfree(work);
 }
 
+void mlx5_vhca_events_work_enqueue(struct mlx5_core_dev *dev, int idx, struct work_struct *work)
+{
+	queue_work(dev->priv.vhca_events->handler[idx].wq, work);
+}
+
 static int
 mlx5_vhca_state_change_notifier(struct notifier_block *nb, unsigned long type, void *data)
 {
@@ -106,6 +120,7 @@ mlx5_vhca_state_change_notifier(struct notifier_block *nb, unsigned long type, v
 				mlx5_nb_cof(nb, struct mlx5_vhca_state_notifier, nb);
 	struct mlx5_vhca_event_work *work;
 	struct mlx5_eqe *eqe = data;
+	int wq_idx;
 
 	work = kzalloc(sizeof(*work), GFP_ATOMIC);
 	if (!work)
@@ -113,7 +128,8 @@ mlx5_vhca_state_change_notifier(struct notifier_block *nb, unsigned long type, v
 	INIT_WORK(&work->work, &mlx5_vhca_state_work_handler);
 	work->notifier = notifier;
 	work->event.function_id = be16_to_cpu(eqe->data.vhca_state.function_id);
-	mlx5_events_work_enqueue(notifier->dev, &work->work);
+	wq_idx = work->event.function_id % MLX5_DEV_MAX_WQS;
+	mlx5_vhca_events_work_enqueue(notifier->dev, wq_idx, &work->work);
 	return NOTIFY_OK;
 }
 
@@ -132,28 +148,75 @@ void mlx5_vhca_state_cap_handle(struct mlx5_core_dev *dev, void *set_hca_cap)
 int mlx5_vhca_event_init(struct mlx5_core_dev *dev)
 {
 	struct mlx5_vhca_state_notifier *notifier;
+	char wq_name[MLX5_CMD_WQ_MAX_NAME];
+	struct mlx5_vhca_events *events;
+	int err, i;
 
 	if (!mlx5_vhca_event_supported(dev))
 		return 0;
 
-	notifier = kzalloc(sizeof(*notifier), GFP_KERNEL);
-	if (!notifier)
+	events = kzalloc(sizeof(*events), GFP_KERNEL);
+	if (!events)
 		return -ENOMEM;
 
+	events->dev = dev;
+	dev->priv.vhca_events = events;
+	for (i = 0; i < MLX5_DEV_MAX_WQS; i++) {
+		snprintf(wq_name, MLX5_CMD_WQ_MAX_NAME, "mlx5_vhca_event%d", i);
+		events->handler[i].wq = create_singlethread_workqueue(wq_name);
+		if (!events->handler[i].wq) {
+			err = -ENOMEM;
+			goto err_create_wq;
+		}
+	}
+
+	notifier = kzalloc(sizeof(*notifier), GFP_KERNEL);
+	if (!notifier) {
+		err = -ENOMEM;
+		goto err_notifier;
+	}
+
 	dev->priv.vhca_state_notifier = notifier;
 	notifier->dev = dev;
 	BLOCKING_INIT_NOTIFIER_HEAD(&notifier->n_head);
 	MLX5_NB_INIT(&notifier->nb, mlx5_vhca_state_change_notifier, VHCA_STATE_CHANGE);
 	return 0;
+
+err_notifier:
+err_create_wq:
+	for (--i; i >= 0; i--)
+		destroy_workqueue(events->handler[i].wq);
+	kfree(events);
+	return err;
+}
+
+void mlx5_vhca_event_work_queues_flush(struct mlx5_core_dev *dev)
+{
+	struct mlx5_vhca_events *vhca_events;
+	int i;
+
+	if (!mlx5_vhca_event_supported(dev))
+		return;
+
+	vhca_events = dev->priv.vhca_events;
+	for (i = 0; i < MLX5_DEV_MAX_WQS; i++)
+		flush_workqueue(vhca_events->handler[i].wq);
 }
 
 void mlx5_vhca_event_cleanup(struct mlx5_core_dev *dev)
 {
+	struct mlx5_vhca_events *vhca_events;
+	int i;
+
 	if (!mlx5_vhca_event_supported(dev))
 		return;
 
 	kfree(dev->priv.vhca_state_notifier);
 	dev->priv.vhca_state_notifier = NULL;
+	vhca_events = dev->priv.vhca_events;
+	for (i = 0; i < MLX5_DEV_MAX_WQS; i++)
+		destroy_workqueue(vhca_events->handler[i].wq);
+	kvfree(vhca_events);
 }
 
 void mlx5_vhca_event_start(struct mlx5_core_dev *dev)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h
index 013cdfe906..1725ba64f8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h
@@ -28,6 +28,9 @@ int mlx5_modify_vhca_sw_id(struct mlx5_core_dev *dev, u16 function_id, u32 sw_fn
 int mlx5_vhca_event_arm(struct mlx5_core_dev *dev, u16 function_id);
 int mlx5_cmd_query_vhca_state(struct mlx5_core_dev *dev, u16 function_id,
 			      u32 *out, u32 outlen);
+void mlx5_vhca_events_work_enqueue(struct mlx5_core_dev *dev, int idx, struct work_struct *work);
+void mlx5_vhca_event_work_queues_flush(struct mlx5_core_dev *dev);
+
 #else
 
 static inline void mlx5_vhca_state_cap_handle(struct mlx5_core_dev *dev, void *set_hca_cap)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 90c38cbbde..d2b65a0ce4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -55,6 +55,13 @@ static const char *dr_action_id_to_str(enum mlx5dr_action_type action_id)
 	return action_type_to_str[action_id];
 }
 
+static bool mlx5dr_action_supp_fwd_fdb_multi_ft(struct mlx5_core_dev *dev)
+{
+	return (MLX5_CAP_GEN(dev, steering_format_version) < MLX5_STEERING_FORMAT_CONNECTX_6DX ||
+		MLX5_CAP_ESW_FLOWTABLE(dev, fdb_multi_path_any_table_limit_regc) ||
+		MLX5_CAP_ESW_FLOWTABLE(dev, fdb_multi_path_any_table));
+}
+
 static const enum dr_action_valid_state
 next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] = {
 	[DR_ACTION_DOMAIN_NIC_INGRESS] = {
@@ -1170,12 +1177,16 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 				   bool ignore_flow_level,
 				   u32 flow_source)
 {
+	struct mlx5dr_cmd_flow_destination_hw_info tmp_hw_dest;
 	struct mlx5dr_cmd_flow_destination_hw_info *hw_dests;
 	struct mlx5dr_action **ref_actions;
 	struct mlx5dr_action *action;
 	bool reformat_req = false;
+	bool is_ft_wire = false;
+	u16 num_dst_ft = 0;
 	u32 num_of_ref = 0;
 	u32 ref_act_cnt;
+	u16 last_dest;
 	int ret;
 	int i;
 
@@ -1217,11 +1228,22 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 			break;
 
 		case DR_ACTION_TYP_FT:
+			if (num_dst_ft &&
+			    !mlx5dr_action_supp_fwd_fdb_multi_ft(dmn->mdev)) {
+				mlx5dr_dbg(dmn, "multiple FT destinations not supported\n");
+				goto free_ref_actions;
+			}
+			num_dst_ft++;
 			hw_dests[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-			if (dest_action->dest_tbl->is_fw_tbl)
+			if (dest_action->dest_tbl->is_fw_tbl) {
 				hw_dests[i].ft_id = dest_action->dest_tbl->fw_tbl.id;
-			else
+			} else {
 				hw_dests[i].ft_id = dest_action->dest_tbl->tbl->table_id;
+				if (dest_action->dest_tbl->is_wire_ft) {
+					is_ft_wire = true;
+					last_dest = i;
+				}
+			}
 			break;
 
 		default:
@@ -1230,6 +1252,16 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 		}
 	}
 
+	/* In multidest, the FW does the iterator in the RX except of the last
+	 * one that done in the TX.
+	 * So, if one of the ft target is wire, put it at the end of the dest list.
+	 */
+	if (is_ft_wire && num_dst_ft > 1) {
+		tmp_hw_dest = hw_dests[last_dest];
+		hw_dests[last_dest] = hw_dests[num_of_dests - 1];
+		hw_dests[num_of_dests - 1] = tmp_hw_dest;
+	}
+
 	action = dr_action_create_generic(DR_ACTION_TYP_FT);
 	if (!action)
 		goto free_ref_actions;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 6c59de3e28..81eff6c410 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -436,10 +436,6 @@ void mlx5dr_ste_build_mpls(struct mlx5dr_ste_ctx *ste_ctx,
 			   struct mlx5dr_ste_build *sb,
 			   struct mlx5dr_match_param *mask,
 			   bool inner, bool rx);
-void mlx5dr_ste_build_tnl_mpls(struct mlx5dr_ste_ctx *ste_ctx,
-			       struct mlx5dr_ste_build *sb,
-			       struct mlx5dr_match_param *mask,
-			       bool inner, bool rx);
 void mlx5dr_ste_build_tnl_mpls_over_gre(struct mlx5dr_ste_ctx *ste_ctx,
 					struct mlx5dr_ste_build *sb,
 					struct mlx5dr_match_param *mask,
@@ -1064,6 +1060,7 @@ struct mlx5dr_action_sampler {
 
 struct mlx5dr_action_dest_tbl {
 	u8 is_fw_tbl:1;
+	u8 is_wire_ft:1;
 	union {
 		struct mlx5dr_table *tbl;
 		struct {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
index 14f6df88b1..50c2554c9c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -209,10 +209,17 @@ static struct mlx5dr_action *create_ft_action(struct mlx5dr_domain *domain,
 					      struct mlx5_flow_rule *dst)
 {
 	struct mlx5_flow_table *dest_ft = dst->dest_attr.ft;
+	struct mlx5dr_action *tbl_action;
 
 	if (mlx5dr_is_fw_table(dest_ft))
 		return mlx5dr_action_create_dest_flow_fw_table(domain, dest_ft);
-	return mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table);
+
+	tbl_action = mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table);
+	if (tbl_action)
+		tbl_action->dest_tbl->is_wire_ft =
+			dest_ft->flags & MLX5_FLOW_TABLE_UPLINK_VPORT ? 1 : 0;
+
+	return tbl_action;
 }
 
 static struct mlx5dr_action *create_range_action(struct mlx5dr_domain *domain,
diff --git a/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_main.c b/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_main.c
index aaf1faed41..3d09fa5459 100644
--- a/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_main.c
+++ b/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_main.c
@@ -471,7 +471,7 @@ out:
 	return err;
 }
 
-static int mlxbf_gige_remove(struct platform_device *pdev)
+static void mlxbf_gige_remove(struct platform_device *pdev)
 {
 	struct mlxbf_gige *priv = platform_get_drvdata(pdev);
 
@@ -479,8 +479,6 @@ static int mlxbf_gige_remove(struct platform_device *pdev)
 	phy_disconnect(priv->netdev->phydev);
 	mlxbf_gige_mdio_remove(priv);
 	platform_set_drvdata(pdev, NULL);
-
-	return 0;
 }
 
 static void mlxbf_gige_shutdown(struct platform_device *pdev)
@@ -499,7 +497,7 @@ MODULE_DEVICE_TABLE(acpi, mlxbf_gige_acpi_match);
 
 static struct platform_driver mlxbf_gige_driver = {
 	.probe = mlxbf_gige_probe,
-	.remove = mlxbf_gige_remove,
+	.remove_new = mlxbf_gige_remove,
 	.shutdown = mlxbf_gige_shutdown,
 	.driver = {
 		.name = KBUILD_MODNAME,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/cmd.h b/drivers/net/ethernet/mellanox/mlxsw/cmd.h
index 09bef04b11..e827c78be1 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/cmd.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/cmd.h
@@ -276,6 +276,12 @@ MLXSW_ITEM32(cmd_mbox, query_fw, fw_month, 0x14, 8, 8);
  */
 MLXSW_ITEM32(cmd_mbox, query_fw, fw_day, 0x14, 0, 8);
 
+/* cmd_mbox_query_fw_lag_mode_support
+ * 0: CONFIG_PROFILE.lag_mode is not supported by FW
+ * 1: CONFIG_PROFILE.lag_mode is supported by FW
+ */
+MLXSW_ITEM32(cmd_mbox, query_fw, lag_mode_support, 0x18, 1, 1);
+
 /* cmd_mbox_query_fw_clr_int_base_offset
  * Clear Interrupt register's offset from clr_int_bar register
  * in PCI address space.
@@ -659,42 +665,48 @@ MLXSW_ITEM32(cmd_mbox, config_profile,
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, set_ar_sec, 0x0C, 15, 1);
 
-/* cmd_mbox_config_set_ubridge
+/* cmd_mbox_config_profile_set_ubridge
  * Capability bit. Setting a bit to 1 configures the profile
  * according to the mailbox contents.
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, set_ubridge, 0x0C, 22, 1);
 
-/* cmd_mbox_config_set_kvd_linear_size
+/* cmd_mbox_config_profile_set_kvd_linear_size
  * Capability bit. Setting a bit to 1 configures the profile
  * according to the mailbox contents.
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_linear_size, 0x0C, 24, 1);
 
-/* cmd_mbox_config_set_kvd_hash_single_size
+/* cmd_mbox_config_profile_set_kvd_hash_single_size
  * Capability bit. Setting a bit to 1 configures the profile
  * according to the mailbox contents.
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_hash_single_size, 0x0C, 25, 1);
 
-/* cmd_mbox_config_set_kvd_hash_double_size
+/* cmd_mbox_config_profile_set_kvd_hash_double_size
  * Capability bit. Setting a bit to 1 configures the profile
  * according to the mailbox contents.
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_hash_double_size, 0x0C, 26, 1);
 
-/* cmd_mbox_config_set_cqe_version
+/* cmd_mbox_config_profile_set_cqe_version
  * Capability bit. Setting a bit to 1 configures the profile
  * according to the mailbox contents.
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, set_cqe_version, 0x08, 0, 1);
 
-/* cmd_mbox_config_set_cqe_time_stamp_type
+/* cmd_mbox_config_profile_set_cqe_time_stamp_type
  * Capability bit. Setting a bit to 1 configures the profile
  * according to the mailbox contents.
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, set_cqe_time_stamp_type, 0x08, 2, 1);
 
+/* cmd_mbox_config_profile_set_lag_mode
+ * Capability bit. Setting a bit to 1 configures the lag_mode
+ * according to the mailbox contents.
+ */
+MLXSW_ITEM32(cmd_mbox, config_profile, set_lag_mode, 0x08, 7, 1);
+
 /* cmd_mbox_config_profile_max_vepa_channels
  * Maximum number of VEPA channels per port (0 through 16)
  * 0 - multi-channel VEPA is disabled
@@ -840,6 +852,21 @@ MLXSW_ITEM32(cmd_mbox, config_profile, arn, 0x50, 31, 1);
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, ubridge, 0x50, 4, 1);
 
+enum mlxsw_cmd_mbox_config_profile_lag_mode {
+	/* FW manages PGT LAG table */
+	MLXSW_CMD_MBOX_CONFIG_PROFILE_LAG_MODE_FW,
+	/* SW manages PGT LAG table */
+	MLXSW_CMD_MBOX_CONFIG_PROFILE_LAG_MODE_SW,
+};
+
+/* cmd_mbox_config_profile_lag_mode
+ * LAG mode
+ * Configured if set_lag_mode is set
+ * Supported from Spectrum-2 and above.
+ * Supported only when ubridge = 1
+ */
+MLXSW_ITEM32(cmd_mbox, config_profile, lag_mode, 0x50, 3, 1);
+
 /* cmd_mbox_config_kvd_linear_size
  * KVD Linear Size
  * Valid for Spectrum only
@@ -847,7 +874,7 @@ MLXSW_ITEM32(cmd_mbox, config_profile, ubridge, 0x50, 4, 1);
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, kvd_linear_size, 0x54, 0, 24);
 
-/* cmd_mbox_config_kvd_hash_single_size
+/* cmd_mbox_config_profile_kvd_hash_single_size
  * KVD Hash single-entries size
  * Valid for Spectrum only
  * Allowed values are 128*N where N=0 or higher
@@ -856,7 +883,7 @@ MLXSW_ITEM32(cmd_mbox, config_profile, kvd_linear_size, 0x54, 0, 24);
  */
 MLXSW_ITEM32(cmd_mbox, config_profile, kvd_hash_single_size, 0x58, 0, 24);
 
-/* cmd_mbox_config_kvd_hash_double_size
+/* cmd_mbox_config_profile_kvd_hash_double_size
  * KVD Hash double-entries size (units of single-size entries)
  * Valid for Spectrum only
  * Allowed values are 128*N where N=0 or higher
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c
index 1ccf3b73ed..f23421f038 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.c
@@ -204,6 +204,13 @@ int mlxsw_core_max_lag(struct mlxsw_core *mlxsw_core, u16 *p_max_lag)
 }
 EXPORT_SYMBOL(mlxsw_core_max_lag);
 
+enum mlxsw_cmd_mbox_config_profile_lag_mode
+mlxsw_core_lag_mode(struct mlxsw_core *mlxsw_core)
+{
+	return mlxsw_core->bus->lag_mode(mlxsw_core->bus_priv);
+}
+EXPORT_SYMBOL(mlxsw_core_lag_mode);
+
 void *mlxsw_core_driver_priv(struct mlxsw_core *mlxsw_core)
 {
 	return mlxsw_core->driver_priv;
@@ -1792,122 +1799,78 @@ static void mlxsw_core_health_listener_func(const struct mlxsw_reg_info *reg,
 static const struct mlxsw_listener mlxsw_core_health_listener =
 	MLXSW_CORE_EVENTL(mlxsw_core_health_listener_func, MFDE);
 
-static int
+static void
 mlxsw_core_health_fw_fatal_dump_fatal_cause(const char *mfde_pl,
 					    struct devlink_fmsg *fmsg)
 {
 	u32 val, tile_v;
-	int err;
 
 	val = mlxsw_reg_mfde_fatal_cause_id_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "cause_id", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "cause_id", val);
 	tile_v = mlxsw_reg_mfde_fatal_cause_tile_v_get(mfde_pl);
 	if (tile_v) {
 		val = mlxsw_reg_mfde_fatal_cause_tile_index_get(mfde_pl);
-		err = devlink_fmsg_u8_pair_put(fmsg, "tile_index", val);
-		if (err)
-			return err;
+		devlink_fmsg_u8_pair_put(fmsg, "tile_index", val);
 	}
-
-	return 0;
 }
 
-static int
+static void
 mlxsw_core_health_fw_fatal_dump_fw_assert(const char *mfde_pl,
 					  struct devlink_fmsg *fmsg)
 {
 	u32 val, tile_v;
-	int err;
 
 	val = mlxsw_reg_mfde_fw_assert_var0_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "var0", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "var0", val);
 	val = mlxsw_reg_mfde_fw_assert_var1_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "var1", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "var1", val);
 	val = mlxsw_reg_mfde_fw_assert_var2_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "var2", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "var2", val);
 	val = mlxsw_reg_mfde_fw_assert_var3_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "var3", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "var3", val);
 	val = mlxsw_reg_mfde_fw_assert_var4_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "var4", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "var4", val);
 	val = mlxsw_reg_mfde_fw_assert_existptr_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "existptr", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "existptr", val);
 	val = mlxsw_reg_mfde_fw_assert_callra_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "callra", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "callra", val);
 	val = mlxsw_reg_mfde_fw_assert_oe_get(mfde_pl);
-	err = devlink_fmsg_bool_pair_put(fmsg, "old_event", val);
-	if (err)
-		return err;
+	devlink_fmsg_bool_pair_put(fmsg, "old_event", val);
 	tile_v = mlxsw_reg_mfde_fw_assert_tile_v_get(mfde_pl);
 	if (tile_v) {
 		val = mlxsw_reg_mfde_fw_assert_tile_index_get(mfde_pl);
-		err = devlink_fmsg_u8_pair_put(fmsg, "tile_index", val);
-		if (err)
-			return err;
+		devlink_fmsg_u8_pair_put(fmsg, "tile_index", val);
 	}
 	val = mlxsw_reg_mfde_fw_assert_ext_synd_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "ext_synd", val);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_u32_pair_put(fmsg, "ext_synd", val);
 }
 
-static int
+static void
 mlxsw_core_health_fw_fatal_dump_kvd_im_stop(const char *mfde_pl,
 					    struct devlink_fmsg *fmsg)
 {
 	u32 val;
-	int err;
 
 	val = mlxsw_reg_mfde_kvd_im_stop_oe_get(mfde_pl);
-	err = devlink_fmsg_bool_pair_put(fmsg, "old_event", val);
-	if (err)
-		return err;
+	devlink_fmsg_bool_pair_put(fmsg, "old_event", val);
 	val = mlxsw_reg_mfde_kvd_im_stop_pipes_mask_get(mfde_pl);
-	return devlink_fmsg_u32_pair_put(fmsg, "pipes_mask", val);
+	devlink_fmsg_u32_pair_put(fmsg, "pipes_mask", val);
 }
 
-static int
+static void
 mlxsw_core_health_fw_fatal_dump_crspace_to(const char *mfde_pl,
 					   struct devlink_fmsg *fmsg)
 {
 	u32 val;
-	int err;
 
 	val = mlxsw_reg_mfde_crspace_to_log_address_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "log_address", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "log_address", val);
 	val = mlxsw_reg_mfde_crspace_to_oe_get(mfde_pl);
-	err = devlink_fmsg_bool_pair_put(fmsg, "old_event", val);
-	if (err)
-		return err;
+	devlink_fmsg_bool_pair_put(fmsg, "old_event", val);
 	val = mlxsw_reg_mfde_crspace_to_log_id_get(mfde_pl);
-	err = devlink_fmsg_u8_pair_put(fmsg, "log_irisc_id", val);
-	if (err)
-		return err;
+	devlink_fmsg_u8_pair_put(fmsg, "log_irisc_id", val);
 	val = mlxsw_reg_mfde_crspace_to_log_ip_get(mfde_pl);
-	err = devlink_fmsg_u64_pair_put(fmsg, "log_ip", val);
-	if (err)
-		return err;
-
-	return 0;
+	devlink_fmsg_u64_pair_put(fmsg, "log_ip", val);
 }
 
 static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *reporter,
@@ -1918,24 +1881,17 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor
 	char *val_str;
 	u8 event_id;
 	u32 val;
-	int err;
 
 	if (!priv_ctx)
 		/* User-triggered dumps are not possible */
 		return -EOPNOTSUPP;
 
 	val = mlxsw_reg_mfde_irisc_id_get(mfde_pl);
-	err = devlink_fmsg_u8_pair_put(fmsg, "irisc_id", val);
-	if (err)
-		return err;
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "event");
-	if (err)
-		return err;
+	devlink_fmsg_u8_pair_put(fmsg, "irisc_id", val);
 
+	devlink_fmsg_arr_pair_nest_start(fmsg, "event");
 	event_id = mlxsw_reg_mfde_event_id_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "id", event_id);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "id", event_id);
 	switch (event_id) {
 	case MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO:
 		val_str = "CR space timeout";
@@ -1955,24 +1911,13 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor
 	default:
 		val_str = NULL;
 	}
-	if (val_str) {
-		err = devlink_fmsg_string_pair_put(fmsg, "desc", val_str);
-		if (err)
-			return err;
-	}
-
-	err = devlink_fmsg_arr_pair_nest_end(fmsg);
-	if (err)
-		return err;
-
-	err = devlink_fmsg_arr_pair_nest_start(fmsg, "severity");
-	if (err)
-		return err;
+	if (val_str)
+		devlink_fmsg_string_pair_put(fmsg, "desc", val_str);
+	devlink_fmsg_arr_pair_nest_end(fmsg);
 
+	devlink_fmsg_arr_pair_nest_start(fmsg, "severity");
 	val = mlxsw_reg_mfde_severity_get(mfde_pl);
-	err = devlink_fmsg_u8_pair_put(fmsg, "id", val);
-	if (err)
-		return err;
+	devlink_fmsg_u8_pair_put(fmsg, "id", val);
 	switch (val) {
 	case MLXSW_REG_MFDE_SEVERITY_FATL:
 		val_str = "Fatal";
@@ -1986,15 +1931,9 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor
 	default:
 		val_str = NULL;
 	}
-	if (val_str) {
-		err = devlink_fmsg_string_pair_put(fmsg, "desc", val_str);
-		if (err)
-			return err;
-	}
-
-	err = devlink_fmsg_arr_pair_nest_end(fmsg);
-	if (err)
-		return err;
+	if (val_str)
+		devlink_fmsg_string_pair_put(fmsg, "desc", val_str);
+	devlink_fmsg_arr_pair_nest_end(fmsg);
 
 	val = mlxsw_reg_mfde_method_get(mfde_pl);
 	switch (val) {
@@ -2007,16 +1946,11 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor
 	default:
 		val_str = NULL;
 	}
-	if (val_str) {
-		err = devlink_fmsg_string_pair_put(fmsg, "method", val_str);
-		if (err)
-			return err;
-	}
+	if (val_str)
+		devlink_fmsg_string_pair_put(fmsg, "method", val_str);
 
 	val = mlxsw_reg_mfde_long_process_get(mfde_pl);
-	err = devlink_fmsg_bool_pair_put(fmsg, "long_process", val);
-	if (err)
-		return err;
+	devlink_fmsg_bool_pair_put(fmsg, "long_process", val);
 
 	val = mlxsw_reg_mfde_command_type_get(mfde_pl);
 	switch (val) {
@@ -2032,29 +1966,25 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor
 	default:
 		val_str = NULL;
 	}
-	if (val_str) {
-		err = devlink_fmsg_string_pair_put(fmsg, "command_type", val_str);
-		if (err)
-			return err;
-	}
+	if (val_str)
+		devlink_fmsg_string_pair_put(fmsg, "command_type", val_str);
 
 	val = mlxsw_reg_mfde_reg_attr_id_get(mfde_pl);
-	err = devlink_fmsg_u32_pair_put(fmsg, "reg_attr_id", val);
-	if (err)
-		return err;
+	devlink_fmsg_u32_pair_put(fmsg, "reg_attr_id", val);
 
 	switch (event_id) {
 	case MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO:
-		return mlxsw_core_health_fw_fatal_dump_crspace_to(mfde_pl,
-								  fmsg);
+		mlxsw_core_health_fw_fatal_dump_crspace_to(mfde_pl, fmsg);
+		break;
 	case MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP:
-		return mlxsw_core_health_fw_fatal_dump_kvd_im_stop(mfde_pl,
-								   fmsg);
+		mlxsw_core_health_fw_fatal_dump_kvd_im_stop(mfde_pl, fmsg);
+		break;
 	case MLXSW_REG_MFDE_EVENT_ID_FW_ASSERT:
-		return mlxsw_core_health_fw_fatal_dump_fw_assert(mfde_pl, fmsg);
+		mlxsw_core_health_fw_fatal_dump_fw_assert(mfde_pl, fmsg);
+		break;
 	case MLXSW_REG_MFDE_EVENT_ID_FATAL_CAUSE:
-		return mlxsw_core_health_fw_fatal_dump_fatal_cause(mfde_pl,
-								   fmsg);
+		mlxsw_core_health_fw_fatal_dump_fatal_cause(mfde_pl, fmsg);
+		break;
 	}
 
 	return 0;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h
index e5474d3e34..764d14bd5b 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.h
@@ -36,6 +36,8 @@ struct mlxsw_fw_rev;
 unsigned int mlxsw_core_max_ports(const struct mlxsw_core *mlxsw_core);
 
 int mlxsw_core_max_lag(struct mlxsw_core *mlxsw_core, u16 *p_max_lag);
+enum mlxsw_cmd_mbox_config_profile_lag_mode
+mlxsw_core_lag_mode(struct mlxsw_core *mlxsw_core);
 
 void *mlxsw_core_driver_priv(struct mlxsw_core *mlxsw_core);
 
@@ -335,6 +337,7 @@ struct mlxsw_config_profile {
 	u8	kvd_hash_single_parts;
 	u8	kvd_hash_double_parts;
 	u8	cqe_time_stamp_type;
+	bool	lag_mode_prefer_sw;
 	struct mlxsw_swid_config swid_config[MLXSW_CONFIG_PROFILE_SWID_COUNT];
 };
 
@@ -485,6 +488,7 @@ struct mlxsw_bus {
 	u32 (*read_frc_l)(void *bus_priv);
 	u32 (*read_utc_sec)(void *bus_priv);
 	u32 (*read_utc_nsec)(void *bus_priv);
+	enum mlxsw_cmd_mbox_config_profile_lag_mode (*lag_mode)(void *bus_priv);
 	u8 features;
 };
 
@@ -624,7 +628,7 @@ struct mlxsw_linecards {
 	struct mlxsw_linecard_types_info *types_info;
 	struct list_head event_ops_list;
 	struct mutex event_ops_list_lock; /* Locks accesses to event ops list */
-	struct mlxsw_linecard linecards[];
+	struct mlxsw_linecard linecards[] __counted_by(count);
 };
 
 static inline struct mlxsw_linecard *
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c
index 70f9b5e85a..0d5e6f9b46 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c
@@ -32,8 +32,7 @@ static const struct mlxsw_afk_element_info mlxsw_afk_element_infos[] = {
 	MLXSW_AFK_ELEMENT_INFO_U32(IP_TTL_, 0x18, 0, 8),
 	MLXSW_AFK_ELEMENT_INFO_U32(IP_ECN, 0x18, 9, 2),
 	MLXSW_AFK_ELEMENT_INFO_U32(IP_DSCP, 0x18, 11, 6),
-	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_MSB, 0x18, 17, 4),
-	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_LSB, 0x18, 21, 8),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER, 0x18, 17, 12),
 	MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_96_127, 0x20, 4),
 	MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_64_95, 0x24, 4),
 	MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_32_63, 0x28, 4),
@@ -44,6 +43,9 @@ static const struct mlxsw_afk_element_info mlxsw_afk_element_infos[] = {
 	MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_0_31, 0x3C, 4),
 	MLXSW_AFK_ELEMENT_INFO_U32(FDB_MISS, 0x40, 0, 1),
 	MLXSW_AFK_ELEMENT_INFO_U32(L4_PORT_RANGE, 0x40, 1, 16),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_0_3, 0x40, 17, 4),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_4_7, 0x40, 21, 4),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_MSB, 0x40, 25, 4),
 };
 
 struct mlxsw_afk {
@@ -136,6 +138,7 @@ mlxsw_afk_key_info_find(struct mlxsw_afk *mlxsw_afk,
 
 struct mlxsw_afk_picker {
 	DECLARE_BITMAP(element, MLXSW_AFK_ELEMENT_MAX);
+	DECLARE_BITMAP(chosen_element, MLXSW_AFK_ELEMENT_MAX);
 	unsigned int total;
 };
 
@@ -206,7 +209,7 @@ static int mlxsw_afk_picker_key_info_add(struct mlxsw_afk *mlxsw_afk,
 	if (key_info->blocks_count == mlxsw_afk->max_blocks)
 		return -EINVAL;
 
-	for_each_set_bit(element, picker[block_index].element,
+	for_each_set_bit(element, picker[block_index].chosen_element,
 			 MLXSW_AFK_ELEMENT_MAX) {
 		key_info->element_to_block[element] = key_info->blocks_count;
 		mlxsw_afk_element_usage_add(&key_info->elusage, element);
@@ -218,11 +221,43 @@ static int mlxsw_afk_picker_key_info_add(struct mlxsw_afk *mlxsw_afk,
 	return 0;
 }
 
+static int mlxsw_afk_keys_fill(struct mlxsw_afk *mlxsw_afk,
+			       unsigned long *chosen_blocks_bm,
+			       struct mlxsw_afk_picker *picker,
+			       struct mlxsw_afk_key_info *key_info)
+{
+	int i, err;
+
+	/* First fill only key blocks with high_entropy. */
+	for_each_set_bit(i, chosen_blocks_bm, mlxsw_afk->blocks_count) {
+		if (!mlxsw_afk->blocks[i].high_entropy)
+			continue;
+
+		err = mlxsw_afk_picker_key_info_add(mlxsw_afk, picker, i,
+						    key_info);
+		if (err)
+			return err;
+		__clear_bit(i, chosen_blocks_bm);
+	}
+
+	/* Fill the rest of key blocks. */
+	for_each_set_bit(i, chosen_blocks_bm, mlxsw_afk->blocks_count) {
+		err = mlxsw_afk_picker_key_info_add(mlxsw_afk, picker, i,
+						    key_info);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 static int mlxsw_afk_picker(struct mlxsw_afk *mlxsw_afk,
 			    struct mlxsw_afk_key_info *key_info,
 			    struct mlxsw_afk_element_usage *elusage)
 {
+	DECLARE_BITMAP(elusage_chosen, MLXSW_AFK_ELEMENT_MAX) = {0};
 	struct mlxsw_afk_picker *picker;
+	unsigned long *chosen_blocks_bm;
 	enum mlxsw_afk_element element;
 	int err;
 
@@ -230,6 +265,12 @@ static int mlxsw_afk_picker(struct mlxsw_afk *mlxsw_afk,
 	if (!picker)
 		return -ENOMEM;
 
+	chosen_blocks_bm = bitmap_zalloc(mlxsw_afk->blocks_count, GFP_KERNEL);
+	if (!chosen_blocks_bm) {
+		err = -ENOMEM;
+		goto err_bitmap_alloc;
+	}
+
 	/* Since the same elements could be present in multiple blocks,
 	 * we must find out optimal block list in order to make the
 	 * block count as low as possible.
@@ -254,15 +295,26 @@ static int mlxsw_afk_picker(struct mlxsw_afk *mlxsw_afk,
 			err = block_index;
 			goto out;
 		}
-		err = mlxsw_afk_picker_key_info_add(mlxsw_afk, picker,
-						    block_index, key_info);
-		if (err)
-			goto out;
+
+		__set_bit(block_index, chosen_blocks_bm);
+
+		bitmap_copy(picker[block_index].chosen_element,
+			    picker[block_index].element, MLXSW_AFK_ELEMENT_MAX);
+
+		bitmap_or(elusage_chosen, elusage_chosen,
+			  picker[block_index].chosen_element,
+			  MLXSW_AFK_ELEMENT_MAX);
+
 		mlxsw_afk_picker_subtract_hits(mlxsw_afk, picker, block_index);
-	} while (!mlxsw_afk_key_info_elements_eq(key_info, elusage));
 
-	err = 0;
+	} while (!bitmap_equal(elusage_chosen, elusage->usage,
+			       MLXSW_AFK_ELEMENT_MAX));
+
+	err = mlxsw_afk_keys_fill(mlxsw_afk, chosen_blocks_bm, picker,
+				  key_info);
 out:
+	bitmap_free(chosen_blocks_bm);
+err_bitmap_alloc:
 	kfree(picker);
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h
index 2eac7582c3..98a0559817 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h
@@ -33,10 +33,12 @@ enum mlxsw_afk_element {
 	MLXSW_AFK_ELEMENT_IP_TTL_,
 	MLXSW_AFK_ELEMENT_IP_ECN,
 	MLXSW_AFK_ELEMENT_IP_DSCP,
-	MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-	MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER,
 	MLXSW_AFK_ELEMENT_FDB_MISS,
 	MLXSW_AFK_ELEMENT_L4_PORT_RANGE,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER_0_3,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER_4_7,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
 	MLXSW_AFK_ELEMENT_MAX,
 };
 
@@ -117,6 +119,7 @@ struct mlxsw_afk_block {
 	u16 encoding; /* block ID */
 	struct mlxsw_afk_element_inst *instances;
 	unsigned int instances_count;
+	bool high_entropy;
 };
 
 #define MLXSW_AFK_BLOCK(_encoding, _instances)					\
@@ -126,6 +129,14 @@ struct mlxsw_afk_block {
 		.instances_count = ARRAY_SIZE(_instances),			\
 	}
 
+#define MLXSW_AFK_BLOCK_HIGH_ENTROPY(_encoding, _instances)			\
+	{									\
+		.encoding = _encoding,						\
+		.instances = _instances,					\
+		.instances_count = ARRAY_SIZE(_instances),			\
+		.high_entropy = true,						\
+	}
+
 struct mlxsw_afk_element_usage {
 	DECLARE_BITMAP(usage, MLXSW_AFK_ELEMENT_MAX);
 };
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.c b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
index d637c0348f..53b150b7ae 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_env.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
@@ -34,7 +34,7 @@ struct mlxsw_env {
 	u8 num_of_slots; /* Including the main board. */
 	u8 max_eeprom_len; /* Maximum module EEPROM transaction length. */
 	struct mutex line_cards_lock; /* Protects line cards. */
-	struct mlxsw_env_line_card *line_cards[];
+	struct mlxsw_env_line_card *line_cards[] __counted_by(num_of_slots);
 };
 
 static bool __mlxsw_env_linecard_is_active(struct mlxsw_env *mlxsw_env,
@@ -775,7 +775,7 @@ static int mlxsw_env_module_has_temp_sensor(struct mlxsw_core *mlxsw_core,
 	int err;
 
 	mlxsw_reg_mtbr_pack(mtbr_pl, slot_index,
-			    MLXSW_REG_MTBR_BASE_MODULE_INDEX + module, 1);
+			    MLXSW_REG_MTBR_BASE_MODULE_INDEX + module);
 	err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mtbr), mtbr_pl);
 	if (err)
 		return err;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
index 0fd290d776..9c12e1feb6 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
@@ -293,7 +293,7 @@ static ssize_t mlxsw_hwmon_module_temp_fault_show(struct device *dev,
 
 	module = mlxsw_hwmon_attr->type_index - mlxsw_hwmon_dev->sensor_count;
 	mlxsw_reg_mtbr_pack(mtbr_pl, mlxsw_hwmon_dev->slot_index,
-			    MLXSW_REG_MTBR_BASE_MODULE_INDEX + module, 1);
+			    MLXSW_REG_MTBR_BASE_MODULE_INDEX + module);
 	err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mtbr), mtbr_pl);
 	if (err) {
 		dev_err(dev, "Failed to query module temperature sensor\n");
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_linecard_dev.c b/drivers/net/ethernet/mellanox/mlxsw/core_linecard_dev.c
index af37e650a8..e8d6fe35bf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_linecard_dev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_linecard_dev.c
@@ -132,6 +132,7 @@ static int mlxsw_linecard_bdev_probe(struct auxiliary_device *adev,
 	struct mlxsw_linecard *linecard = linecard_bdev->linecard;
 	struct mlxsw_linecard_dev *linecard_dev;
 	struct devlink *devlink;
+	int err;
 
 	devlink = devlink_alloc(&mlxsw_linecard_dev_devlink_ops,
 				sizeof(*linecard_dev), &adev->dev);
@@ -141,8 +142,12 @@ static int mlxsw_linecard_bdev_probe(struct auxiliary_device *adev,
 	linecard_dev->linecard = linecard_bdev->linecard;
 	linecard_bdev->linecard_dev = linecard_dev;
 
+	err = devlink_linecard_nested_dl_set(linecard->devlink_linecard, devlink);
+	if (err) {
+		devlink_free(devlink);
+		return err;
+	}
 	devlink_register(devlink);
-	devlink_linecard_nested_dl_set(linecard->devlink_linecard, devlink);
 	return 0;
 }
 
@@ -151,9 +156,7 @@ static void mlxsw_linecard_bdev_remove(struct auxiliary_device *adev)
 	struct mlxsw_linecard_bdev *linecard_bdev =
 			container_of(adev, struct mlxsw_linecard_bdev, adev);
 	struct devlink *devlink = priv_to_devlink(linecard_bdev->linecard_dev);
-	struct mlxsw_linecard *linecard = linecard_bdev->linecard;
 
-	devlink_linecard_nested_dl_set(linecard->devlink_linecard, NULL);
 	devlink_unregister(devlink);
 	devlink_free(devlink);
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index 70d7fff24f..f1b48d6615 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -31,6 +31,7 @@
 /* External cooling devices, allowed for binding to mlxsw thermal zones. */
 static char * const mlxsw_thermal_external_allowed_cdev[] = {
 	"mlxreg_fan",
+	"emc2305",
 };
 
 struct mlxsw_cooling_states {
@@ -535,7 +536,7 @@ mlxsw_thermal_modules_fini(struct mlxsw_thermal *thermal,
 static int
 mlxsw_thermal_gearbox_tz_init(struct mlxsw_thermal_module *gearbox_tz)
 {
-	char tz_name[THERMAL_NAME_LENGTH];
+	char tz_name[40];
 	int ret;
 
 	if (gearbox_tz->slot_index)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/i2c.c b/drivers/net/ethernet/mellanox/mlxsw/i2c.c
index d23f293e28..1e150ce1c7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/i2c.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/i2c.c
@@ -424,9 +424,7 @@ mlxsw_i2c_cmd(struct device *dev, u16 opcode, u32 in_mod, size_t in_mbox_size,
 
 	if (in_mbox) {
 		reg_size = mlxsw_i2c_get_reg_size(in_mbox);
-		num = reg_size / mlxsw_i2c->block_size;
-		if (reg_size % mlxsw_i2c->block_size)
-			num++;
+		num = DIV_ROUND_UP(reg_size, mlxsw_i2c->block_size);
 
 		if (mutex_lock_interruptible(&mlxsw_i2c->cmd.lock) < 0) {
 			dev_err(&client->dev, "Could not acquire lock");
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 51eea1f052..e4b25e1874 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -105,6 +105,8 @@ struct mlxsw_pci {
 	u64 free_running_clock_offset;
 	u64 utc_sec_offset;
 	u64 utc_nsec_offset;
+	bool lag_mode_support;
+	enum mlxsw_cmd_mbox_config_profile_lag_mode lag_mode;
 	struct mlxsw_pci_queue_type_group queues[MLXSW_PCI_QUEUE_TYPE_COUNT];
 	u32 doorbell_offset;
 	struct mlxsw_core *core;
@@ -352,14 +354,15 @@ static void mlxsw_pci_wqe_frag_unmap(struct mlxsw_pci *mlxsw_pci, char *wqe,
 }
 
 static int mlxsw_pci_rdq_skb_alloc(struct mlxsw_pci *mlxsw_pci,
-				   struct mlxsw_pci_queue_elem_info *elem_info)
+				   struct mlxsw_pci_queue_elem_info *elem_info,
+				   gfp_t gfp)
 {
 	size_t buf_len = MLXSW_PORT_MAX_MTU;
 	char *wqe = elem_info->elem;
 	struct sk_buff *skb;
 	int err;
 
-	skb = netdev_alloc_skb_ip_align(NULL, buf_len);
+	skb = __netdev_alloc_skb_ip_align(NULL, buf_len, gfp);
 	if (!skb)
 		return -ENOMEM;
 
@@ -420,7 +423,7 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 	for (i = 0; i < q->count; i++) {
 		elem_info = mlxsw_pci_queue_elem_info_producer_get(q);
 		BUG_ON(!elem_info);
-		err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info);
+		err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info, GFP_KERNEL);
 		if (err)
 			goto rollback;
 		/* Everything is set up, ring doorbell to pass elem to HW */
@@ -640,7 +643,7 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci,
 	if (q->consumer_counter++ != consumer_counter_limit)
 		dev_dbg_ratelimited(&pdev->dev, "Consumer counter does not match limit in RDQ\n");
 
-	err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info);
+	err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info, GFP_ATOMIC);
 	if (err) {
 		dev_err_ratelimited(&pdev->dev, "Failed to alloc skb for RDQ\n");
 		goto out;
@@ -1311,6 +1314,16 @@ static int mlxsw_pci_config_profile(struct mlxsw_pci *mlxsw_pci, char *mbox,
 					profile->cqe_time_stamp_type);
 	}
 
+	if (profile->lag_mode_prefer_sw && mlxsw_pci->lag_mode_support) {
+		enum mlxsw_cmd_mbox_config_profile_lag_mode lag_mode =
+			MLXSW_CMD_MBOX_CONFIG_PROFILE_LAG_MODE_SW;
+
+		mlxsw_cmd_mbox_config_profile_set_lag_mode_set(mbox, 1);
+		mlxsw_cmd_mbox_config_profile_lag_mode_set(mbox, lag_mode);
+		mlxsw_pci->lag_mode = lag_mode;
+	} else {
+		mlxsw_pci->lag_mode = MLXSW_CMD_MBOX_CONFIG_PROFILE_LAG_MODE_FW;
+	}
 	return mlxsw_cmd_config_profile_set(mlxsw_pci->core, mbox);
 }
 
@@ -1586,6 +1599,8 @@ static int mlxsw_pci_init(void *bus_priv, struct mlxsw_core *mlxsw_core,
 	mlxsw_pci->utc_nsec_offset =
 		mlxsw_cmd_mbox_query_fw_utc_nsec_offset_get(mbox);
 
+	mlxsw_pci->lag_mode_support =
+		mlxsw_cmd_mbox_query_fw_lag_mode_support_get(mbox);
 	num_pages = mlxsw_cmd_mbox_query_fw_fw_pages_get(mbox);
 	err = mlxsw_pci_fw_area_init(mlxsw_pci, mbox, num_pages);
 	if (err)
@@ -1618,9 +1633,8 @@ static int mlxsw_pci_init(void *bus_priv, struct mlxsw_core *mlxsw_core,
 	if (err)
 		goto err_config_profile;
 
-	/* Some resources depend on unified bridge model, which is configured
-	 * as part of config_profile. Query the resources again to get correct
-	 * values.
+	/* Some resources depend on details of config_profile, such as unified
+	 * bridge model. Query the resources again to get correct values.
 	 */
 	err = mlxsw_core_resources_query(mlxsw_core, mbox, res);
 	if (err)
@@ -1895,6 +1909,14 @@ static u32 mlxsw_pci_read_utc_nsec(void *bus_priv)
 	return mlxsw_pci_read32_off(mlxsw_pci, mlxsw_pci->utc_nsec_offset);
 }
 
+static enum mlxsw_cmd_mbox_config_profile_lag_mode
+mlxsw_pci_lag_mode(void *bus_priv)
+{
+	struct mlxsw_pci *mlxsw_pci = bus_priv;
+
+	return mlxsw_pci->lag_mode;
+}
+
 static const struct mlxsw_bus mlxsw_pci_bus = {
 	.kind			= "pci",
 	.init			= mlxsw_pci_init,
@@ -1906,6 +1928,7 @@ static const struct mlxsw_bus mlxsw_pci_bus = {
 	.read_frc_l		= mlxsw_pci_read_frc_l,
 	.read_utc_sec		= mlxsw_pci_read_utc_sec,
 	.read_utc_nsec		= mlxsw_pci_read_utc_nsec,
+	.lag_mode		= mlxsw_pci_lag_mode,
 	.features		= MLXSW_BUS_F_TXRX | MLXSW_BUS_F_RESET,
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index ae556ddd76..25b294fdeb 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -38,18 +38,18 @@ static const struct mlxsw_reg_info mlxsw_reg_##_name = {		\
 
 MLXSW_REG_DEFINE(sgcr, MLXSW_REG_SGCR_ID, MLXSW_REG_SGCR_LEN);
 
-/* reg_sgcr_llb
- * Link Local Broadcast (Default=0)
- * When set, all Link Local packets (224.0.0.X) will be treated as broadcast
- * packets and ignore the IGMP snooping entries.
+/* reg_sgcr_lag_lookup_pgt_base
+ * Base address used for lookup in PGT table
+ * Supported when CONFIG_PROFILE.lag_mode = 1
+ * Note: when IGCR.ddd_lag_mode=0, the address shall be aligned to 8 entries.
  * Access: RW
  */
-MLXSW_ITEM32(reg, sgcr, llb, 0x04, 0, 1);
+MLXSW_ITEM32(reg, sgcr, lag_lookup_pgt_base, 0x0C, 0, 16);
 
-static inline void mlxsw_reg_sgcr_pack(char *payload, bool llb)
+static inline void mlxsw_reg_sgcr_pack(char *payload, u16 lag_lookup_pgt_base)
 {
 	MLXSW_REG_ZERO(sgcr, payload);
-	mlxsw_reg_sgcr_llb_set(payload, !!llb);
+	mlxsw_reg_sgcr_lag_lookup_pgt_base_set(payload, lag_lookup_pgt_base);
 }
 
 /* SPAD - Switch Physical Address Register
@@ -9551,7 +9551,7 @@ MLXSW_ITEM_BIT_ARRAY(reg, mtwe, sensor_warning, 0x0, 0x10, 1);
 #define MLXSW_REG_MTBR_ID 0x900F
 #define MLXSW_REG_MTBR_BASE_LEN 0x10 /* base length, without records */
 #define MLXSW_REG_MTBR_REC_LEN 0x04 /* record length */
-#define MLXSW_REG_MTBR_REC_MAX_COUNT 47 /* firmware limitation */
+#define MLXSW_REG_MTBR_REC_MAX_COUNT 1
 #define MLXSW_REG_MTBR_LEN (MLXSW_REG_MTBR_BASE_LEN +	\
 			    MLXSW_REG_MTBR_REC_LEN *	\
 			    MLXSW_REG_MTBR_REC_MAX_COUNT)
@@ -9597,12 +9597,12 @@ MLXSW_ITEM32_INDEXED(reg, mtbr, rec_temp, MLXSW_REG_MTBR_BASE_LEN, 0, 16,
 		     MLXSW_REG_MTBR_REC_LEN, 0x00, false);
 
 static inline void mlxsw_reg_mtbr_pack(char *payload, u8 slot_index,
-				       u16 base_sensor_index, u8 num_rec)
+				       u16 base_sensor_index)
 {
 	MLXSW_REG_ZERO(mtbr, payload);
 	mlxsw_reg_mtbr_slot_index_set(payload, slot_index);
 	mlxsw_reg_mtbr_base_sensor_index_set(payload, base_sensor_index);
-	mlxsw_reg_mtbr_num_rec_set(payload, num_rec);
+	mlxsw_reg_mtbr_num_rec_set(payload, 1);
 }
 
 /* Error codes from temperatute reading */
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 9dbd5edff0..cec72d99d9 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -2692,6 +2692,63 @@ static void mlxsw_sp_traps_fini(struct mlxsw_sp *mlxsw_sp)
 	kfree(mlxsw_sp->trap);
 }
 
+static int mlxsw_sp_lag_pgt_init(struct mlxsw_sp *mlxsw_sp)
+{
+	char sgcr_pl[MLXSW_REG_SGCR_LEN];
+	u16 max_lag;
+	int err;
+
+	if (mlxsw_core_lag_mode(mlxsw_sp->core) !=
+	    MLXSW_CMD_MBOX_CONFIG_PROFILE_LAG_MODE_SW)
+		return 0;
+
+	err = mlxsw_core_max_lag(mlxsw_sp->core, &max_lag);
+	if (err)
+		return err;
+
+	/* In DDD mode, which we by default use, each LAG entry is 8 PGT
+	 * entries. The LAG table address needs to be 8-aligned, but that ought
+	 * to be the case, since the LAG table is allocated first.
+	 */
+	err = mlxsw_sp_pgt_mid_alloc_range(mlxsw_sp, &mlxsw_sp->lag_pgt_base,
+					   max_lag * 8);
+	if (err)
+		return err;
+	if (WARN_ON_ONCE(mlxsw_sp->lag_pgt_base % 8)) {
+		err = -EINVAL;
+		goto err_mid_alloc_range;
+	}
+
+	mlxsw_reg_sgcr_pack(sgcr_pl, mlxsw_sp->lag_pgt_base);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sgcr), sgcr_pl);
+	if (err)
+		goto err_mid_alloc_range;
+
+	return 0;
+
+err_mid_alloc_range:
+	mlxsw_sp_pgt_mid_free_range(mlxsw_sp, mlxsw_sp->lag_pgt_base,
+				    max_lag * 8);
+	return err;
+}
+
+static void mlxsw_sp_lag_pgt_fini(struct mlxsw_sp *mlxsw_sp)
+{
+	u16 max_lag;
+	int err;
+
+	if (mlxsw_core_lag_mode(mlxsw_sp->core) !=
+	    MLXSW_CMD_MBOX_CONFIG_PROFILE_LAG_MODE_SW)
+		return;
+
+	err = mlxsw_core_max_lag(mlxsw_sp->core, &max_lag);
+	if (err)
+		return;
+
+	mlxsw_sp_pgt_mid_free_range(mlxsw_sp, mlxsw_sp->lag_pgt_base,
+				    max_lag * 8);
+}
+
 #define MLXSW_SP_LAG_SEED_INIT 0xcafecafe
 
 static int mlxsw_sp_lag_init(struct mlxsw_sp *mlxsw_sp)
@@ -2723,16 +2780,27 @@ static int mlxsw_sp_lag_init(struct mlxsw_sp *mlxsw_sp)
 	if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, MAX_LAG_MEMBERS))
 		return -EIO;
 
+	err = mlxsw_sp_lag_pgt_init(mlxsw_sp);
+	if (err)
+		return err;
+
 	mlxsw_sp->lags = kcalloc(max_lag, sizeof(struct mlxsw_sp_upper),
 				 GFP_KERNEL);
-	if (!mlxsw_sp->lags)
-		return -ENOMEM;
+	if (!mlxsw_sp->lags) {
+		err = -ENOMEM;
+		goto err_kcalloc;
+	}
 
 	return 0;
+
+err_kcalloc:
+	mlxsw_sp_lag_pgt_fini(mlxsw_sp);
+	return err;
 }
 
 static void mlxsw_sp_lag_fini(struct mlxsw_sp *mlxsw_sp)
 {
+	mlxsw_sp_lag_pgt_fini(mlxsw_sp);
 	kfree(mlxsw_sp->lags);
 }
 
@@ -3113,6 +3181,15 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core,
 		goto err_pgt_init;
 	}
 
+	/* Initialize before FIDs so that the LAG table is at the start of PGT
+	 * and 8-aligned without overallocation.
+	 */
+	err = mlxsw_sp_lag_init(mlxsw_sp);
+	if (err) {
+		dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize LAG\n");
+		goto err_lag_init;
+	}
+
 	err = mlxsw_sp_fids_init(mlxsw_sp);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize FIDs\n");
@@ -3143,12 +3220,6 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core,
 		goto err_buffers_init;
 	}
 
-	err = mlxsw_sp_lag_init(mlxsw_sp);
-	if (err) {
-		dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize LAG\n");
-		goto err_lag_init;
-	}
-
 	/* Initialize SPAN before router and switchdev, so that those components
 	 * can call mlxsw_sp_span_respin().
 	 */
@@ -3300,8 +3371,6 @@ err_counter_pool_init:
 err_switchdev_init:
 	mlxsw_sp_span_fini(mlxsw_sp);
 err_span_init:
-	mlxsw_sp_lag_fini(mlxsw_sp);
-err_lag_init:
 	mlxsw_sp_buffers_fini(mlxsw_sp);
 err_buffers_init:
 	mlxsw_sp_devlink_traps_fini(mlxsw_sp);
@@ -3312,6 +3381,8 @@ err_traps_init:
 err_policers_init:
 	mlxsw_sp_fids_fini(mlxsw_sp);
 err_fids_init:
+	mlxsw_sp_lag_fini(mlxsw_sp);
+err_lag_init:
 	mlxsw_sp_pgt_fini(mlxsw_sp);
 err_pgt_init:
 	mlxsw_sp_kvdl_fini(mlxsw_sp);
@@ -3477,12 +3548,12 @@ static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core)
 	mlxsw_sp_counter_pool_fini(mlxsw_sp);
 	mlxsw_sp_switchdev_fini(mlxsw_sp);
 	mlxsw_sp_span_fini(mlxsw_sp);
-	mlxsw_sp_lag_fini(mlxsw_sp);
 	mlxsw_sp_buffers_fini(mlxsw_sp);
 	mlxsw_sp_devlink_traps_fini(mlxsw_sp);
 	mlxsw_sp_traps_fini(mlxsw_sp);
 	mlxsw_sp_policers_fini(mlxsw_sp);
 	mlxsw_sp_fids_fini(mlxsw_sp);
+	mlxsw_sp_lag_fini(mlxsw_sp);
 	mlxsw_sp_pgt_fini(mlxsw_sp);
 	mlxsw_sp_kvdl_fini(mlxsw_sp);
 	mlxsw_sp_parsing_fini(mlxsw_sp);
@@ -3526,6 +3597,7 @@ static const struct mlxsw_config_profile mlxsw_sp2_config_profile = {
 	},
 	.used_cqe_time_stamp_type	= 1,
 	.cqe_time_stamp_type		= MLXSW_CMD_MBOX_CONFIG_PROFILE_CQE_TIME_STAMP_TYPE_UTC,
+	.lag_mode_prefer_sw		= true,
 };
 
 /* Reduce number of LAGs from full capacity (256) to the maximum supported LAGs
@@ -3553,6 +3625,7 @@ static const struct mlxsw_config_profile mlxsw_sp4_config_profile = {
 	},
 	.used_cqe_time_stamp_type	= 1,
 	.cqe_time_stamp_type		= MLXSW_CMD_MBOX_CONFIG_PROFILE_CQE_TIME_STAMP_TYPE_UTC,
+	.lag_mode_prefer_sw		= true,
 };
 
 static void
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 02ca2871b6..c70333b460 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -212,6 +212,7 @@ struct mlxsw_sp {
 	struct mutex ipv6_addr_ht_lock; /* Protects ipv6_addr_ht */
 	struct mlxsw_sp_pgt *pgt;
 	bool pgt_smpe_index_valid;
+	u16 lag_pgt_base;
 };
 
 struct mlxsw_sp_ptp_ops {
@@ -1480,7 +1481,7 @@ int mlxsw_sp_policer_resources_register(struct mlxsw_core *mlxsw_core);
 /* spectrum_pgt.c */
 int mlxsw_sp_pgt_mid_alloc(struct mlxsw_sp *mlxsw_sp, u16 *p_mid);
 void mlxsw_sp_pgt_mid_free(struct mlxsw_sp *mlxsw_sp, u16 mid_base);
-int mlxsw_sp_pgt_mid_alloc_range(struct mlxsw_sp *mlxsw_sp, u16 mid_base,
+int mlxsw_sp_pgt_mid_alloc_range(struct mlxsw_sp *mlxsw_sp, u16 *mid_base,
 				 u16 count);
 void mlxsw_sp_pgt_mid_free_range(struct mlxsw_sp *mlxsw_sp, u16 mid_base,
 				 u16 count);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c
index b1178b7a7f..99eeafdc8d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c
@@ -45,8 +45,7 @@ static int mlxsw_sp2_mr_tcam_bind_group(struct mlxsw_sp *mlxsw_sp,
 }
 
 static const enum mlxsw_afk_element mlxsw_sp2_mr_tcam_usage_ipv4[] = {
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER,
 		MLXSW_AFK_ELEMENT_SRC_IP_0_31,
 		MLXSW_AFK_ELEMENT_DST_IP_0_31,
 };
@@ -89,8 +88,9 @@ static void mlxsw_sp2_mr_tcam_ipv4_fini(struct mlxsw_sp2_mr_tcam *mr_tcam)
 }
 
 static const enum mlxsw_afk_element mlxsw_sp2_mr_tcam_usage_ipv6[] = {
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER_0_3,
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER_4_7,
 		MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
 		MLXSW_AFK_ELEMENT_SRC_IP_96_127,
 		MLXSW_AFK_ELEMENT_SRC_IP_64_95,
 		MLXSW_AFK_ELEMENT_SRC_IP_32_63,
@@ -142,6 +142,8 @@ static void
 mlxsw_sp2_mr_tcam_rule_parse4(struct mlxsw_sp_acl_rule_info *rulei,
 			      struct mlxsw_sp_mr_route_key *key)
 {
+	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER,
+				       key->vrid, GENMASK(11, 0));
 	mlxsw_sp_acl_rulei_keymask_buf(rulei, MLXSW_AFK_ELEMENT_SRC_IP_0_31,
 				       (char *) &key->source.addr4,
 				       (char *) &key->source_mask.addr4, 4);
@@ -154,6 +156,13 @@ static void
 mlxsw_sp2_mr_tcam_rule_parse6(struct mlxsw_sp_acl_rule_info *rulei,
 			      struct mlxsw_sp_mr_route_key *key)
 {
+	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER_0_3,
+				       key->vrid, GENMASK(3, 0));
+	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER_4_7,
+				       key->vrid >> 4, GENMASK(3, 0));
+	mlxsw_sp_acl_rulei_keymask_u32(rulei,
+				       MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
+				       key->vrid >> 8, GENMASK(3, 0));
 	mlxsw_sp_acl_rulei_keymask_buf(rulei, MLXSW_AFK_ELEMENT_SRC_IP_96_127,
 				       &key->source.addr6.s6_addr[0x0],
 				       &key->source_mask.addr6.s6_addr[0x0], 4);
@@ -189,11 +198,6 @@ mlxsw_sp2_mr_tcam_rule_parse(struct mlxsw_sp_acl_rule *rule,
 
 	rulei = mlxsw_sp_acl_rule_rulei(rule);
 	rulei->priority = priority;
-	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
-				       key->vrid, GENMASK(7, 0));
-	mlxsw_sp_acl_rulei_keymask_u32(rulei,
-				       MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-				       key->vrid >> 8, GENMASK(3, 0));
 	switch (key->proto) {
 	case MLXSW_SP_L3_PROTO_IPV4:
 		return mlxsw_sp2_mr_tcam_rule_parse4(rulei, key);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c
index cb746a43b2..eaad786056 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c
@@ -171,20 +171,22 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_2[] = {
 	MLXSW_AFK_ELEMENT_INST_U32(IP_PROTO, 0x04, 16, 8),
 };
 
-static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_4[] = {
-	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_LSB, 0x04, 24, 8),
-	MLXSW_AFK_ELEMENT_INST_EXT_U32(VIRT_ROUTER_MSB, 0x00, 0, 3, 0, true),
+static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_5[] = {
+	MLXSW_AFK_ELEMENT_INST_EXT_U32(VIRT_ROUTER, 0x04, 20, 11, 0, true),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_0[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_0_3, 0x00, 0, 4),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_32_63, 0x04, 4),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_1[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_4_7, 0x00, 0, 4),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_64_95, 0x04, 4),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_2[] = {
+	MLXSW_AFK_ELEMENT_INST_EXT_U32(VIRT_ROUTER_MSB, 0x00, 0, 3, 0, true),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_96_127, 0x04, 4),
 };
 
@@ -220,7 +222,7 @@ static const struct mlxsw_afk_block mlxsw_sp2_afk_blocks[] = {
 	MLXSW_AFK_BLOCK(0x38, mlxsw_sp_afk_element_info_ipv4_0),
 	MLXSW_AFK_BLOCK(0x39, mlxsw_sp_afk_element_info_ipv4_1),
 	MLXSW_AFK_BLOCK(0x3A, mlxsw_sp_afk_element_info_ipv4_2),
-	MLXSW_AFK_BLOCK(0x3C, mlxsw_sp_afk_element_info_ipv4_4),
+	MLXSW_AFK_BLOCK(0x3D, mlxsw_sp_afk_element_info_ipv4_5),
 	MLXSW_AFK_BLOCK(0x40, mlxsw_sp_afk_element_info_ipv6_0),
 	MLXSW_AFK_BLOCK(0x41, mlxsw_sp_afk_element_info_ipv6_1),
 	MLXSW_AFK_BLOCK(0x42, mlxsw_sp_afk_element_info_ipv6_2),
@@ -322,33 +324,33 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_mac_5b[] = {
 	MLXSW_AFK_ELEMENT_INST_EXT_U32(SRC_SYS_PORT, 0x04, 0, 9, -1, true), /* RX_ACL_SYSTEM_PORT */
 };
 
-static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_4b[] = {
-	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_LSB, 0x04, 13, 8),
-	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_MSB, 0x04, 21, 4),
+static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_5b[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER, 0x04, 20, 12),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_2b[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_MSB, 0x00, 0, 4),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_96_127, 0x04, 4),
 };
 
 static const struct mlxsw_afk_block mlxsw_sp4_afk_blocks[] = {
-	MLXSW_AFK_BLOCK(0x10, mlxsw_sp_afk_element_info_mac_0),
-	MLXSW_AFK_BLOCK(0x11, mlxsw_sp_afk_element_info_mac_1),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x10, mlxsw_sp_afk_element_info_mac_0),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x11, mlxsw_sp_afk_element_info_mac_1),
 	MLXSW_AFK_BLOCK(0x12, mlxsw_sp_afk_element_info_mac_2),
 	MLXSW_AFK_BLOCK(0x13, mlxsw_sp_afk_element_info_mac_3),
 	MLXSW_AFK_BLOCK(0x14, mlxsw_sp_afk_element_info_mac_4),
-	MLXSW_AFK_BLOCK(0x1A, mlxsw_sp_afk_element_info_mac_5b),
-	MLXSW_AFK_BLOCK(0x38, mlxsw_sp_afk_element_info_ipv4_0),
-	MLXSW_AFK_BLOCK(0x39, mlxsw_sp_afk_element_info_ipv4_1),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x1A, mlxsw_sp_afk_element_info_mac_5b),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x38, mlxsw_sp_afk_element_info_ipv4_0),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x39, mlxsw_sp_afk_element_info_ipv4_1),
 	MLXSW_AFK_BLOCK(0x3A, mlxsw_sp_afk_element_info_ipv4_2),
-	MLXSW_AFK_BLOCK(0x35, mlxsw_sp_afk_element_info_ipv4_4b),
+	MLXSW_AFK_BLOCK(0x36, mlxsw_sp_afk_element_info_ipv4_5b),
 	MLXSW_AFK_BLOCK(0x40, mlxsw_sp_afk_element_info_ipv6_0),
 	MLXSW_AFK_BLOCK(0x41, mlxsw_sp_afk_element_info_ipv6_1),
 	MLXSW_AFK_BLOCK(0x47, mlxsw_sp_afk_element_info_ipv6_2b),
 	MLXSW_AFK_BLOCK(0x43, mlxsw_sp_afk_element_info_ipv6_3),
 	MLXSW_AFK_BLOCK(0x44, mlxsw_sp_afk_element_info_ipv6_4),
 	MLXSW_AFK_BLOCK(0x45, mlxsw_sp_afk_element_info_ipv6_5),
-	MLXSW_AFK_BLOCK(0x90, mlxsw_sp_afk_element_info_l4_0),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x90, mlxsw_sp_afk_element_info_l4_0),
 	MLXSW_AFK_BLOCK(0x92, mlxsw_sp_afk_element_info_l4_2),
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c
index ee59c79156..50e591420b 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c
@@ -24,7 +24,7 @@ struct mlxsw_sp_counter_pool {
 	spinlock_t counter_pool_lock; /* Protects counter pool allocations */
 	atomic_t active_entries_count;
 	unsigned int sub_pools_count;
-	struct mlxsw_sp_counter_sub_pool sub_pools[];
+	struct mlxsw_sp_counter_sub_pool sub_pools[] __counted_by(sub_pools_count);
 };
 
 static const struct mlxsw_sp_counter_sub_pool mlxsw_sp_counter_sub_pools[] = {
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
index 472830d07a..0f29e9c194 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
@@ -619,7 +619,7 @@ static void mlxsw_sp_port_get_tc_strings(u8 **p, int tc)
 	int i;
 
 	for (i = 0; i < MLXSW_SP_PORT_HW_TC_STATS_LEN; i++) {
-		snprintf(*p, ETH_GSTRING_LEN, "%.29s_%.1d",
+		snprintf(*p, ETH_GSTRING_LEN, "%.28s_%d",
 			 mlxsw_sp_port_hw_tc_stats[i].str, tc);
 		*p += ETH_GSTRING_LEN;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c
index 9df0984747..e954b8cd2e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c
@@ -321,6 +321,14 @@ mlxsw_sp_fid_family_num_fids(const struct mlxsw_sp_fid_family *fid_family)
 }
 
 static u16
+mlxsw_sp_fid_family_pgt_size(const struct mlxsw_sp_fid_family *fid_family)
+{
+	u16 num_fids = mlxsw_sp_fid_family_num_fids(fid_family);
+
+	return num_fids * fid_family->nr_flood_tables;
+}
+
+static u16
 mlxsw_sp_fid_flood_table_mid(const struct mlxsw_sp_fid_family *fid_family,
 			     const struct mlxsw_sp_flood_table *flood_table,
 			     u16 fid_offset)
@@ -1068,8 +1076,6 @@ static const struct mlxsw_sp_fid_ops mlxsw_sp_fid_8021d_ops = {
 
 #define MLXSW_SP_FID_8021Q_MAX (VLAN_N_VID - 2)
 #define MLXSW_SP_FID_RFID_MAX (11 * 1024)
-#define MLXSW_SP_FID_8021Q_PGT_BASE 0
-#define MLXSW_SP_FID_8021D_PGT_BASE (3 * MLXSW_SP_FID_8021Q_MAX)
 
 static const struct mlxsw_sp_flood_table mlxsw_sp_fid_8021d_flood_tables[] = {
 	{
@@ -1434,7 +1440,6 @@ static const struct mlxsw_sp_fid_family mlxsw_sp1_fid_8021q_family = {
 	.ops			= &mlxsw_sp_fid_8021q_ops,
 	.flood_rsp              = false,
 	.bridge_type            = MLXSW_REG_BRIDGE_TYPE_0,
-	.pgt_base		= MLXSW_SP_FID_8021Q_PGT_BASE,
 	.smpe_index_valid	= false,
 };
 
@@ -1448,7 +1453,6 @@ static const struct mlxsw_sp_fid_family mlxsw_sp1_fid_8021d_family = {
 	.rif_type		= MLXSW_SP_RIF_TYPE_FID,
 	.ops			= &mlxsw_sp_fid_8021d_ops,
 	.bridge_type            = MLXSW_REG_BRIDGE_TYPE_1,
-	.pgt_base		= MLXSW_SP_FID_8021D_PGT_BASE,
 	.smpe_index_valid       = false,
 };
 
@@ -1490,7 +1494,6 @@ static const struct mlxsw_sp_fid_family mlxsw_sp2_fid_8021q_family = {
 	.ops			= &mlxsw_sp_fid_8021q_ops,
 	.flood_rsp              = false,
 	.bridge_type            = MLXSW_REG_BRIDGE_TYPE_0,
-	.pgt_base		= MLXSW_SP_FID_8021Q_PGT_BASE,
 	.smpe_index_valid	= true,
 };
 
@@ -1504,7 +1507,6 @@ static const struct mlxsw_sp_fid_family mlxsw_sp2_fid_8021d_family = {
 	.rif_type		= MLXSW_SP_RIF_TYPE_FID,
 	.ops			= &mlxsw_sp_fid_8021d_ops,
 	.bridge_type            = MLXSW_REG_BRIDGE_TYPE_1,
-	.pgt_base		= MLXSW_SP_FID_8021D_PGT_BASE,
 	.smpe_index_valid       = true,
 };
 
@@ -1654,14 +1656,10 @@ mlxsw_sp_fid_flood_table_init(struct mlxsw_sp_fid_family *fid_family,
 	enum mlxsw_sp_flood_type packet_type = flood_table->packet_type;
 	struct mlxsw_sp *mlxsw_sp = fid_family->mlxsw_sp;
 	const int *sfgc_packet_types;
-	u16 num_fids, mid_base;
+	u16 mid_base;
 	int err, i;
 
 	mid_base = mlxsw_sp_fid_flood_table_mid(fid_family, flood_table, 0);
-	num_fids = mlxsw_sp_fid_family_num_fids(fid_family);
-	err = mlxsw_sp_pgt_mid_alloc_range(mlxsw_sp, mid_base, num_fids);
-	if (err)
-		return err;
 
 	sfgc_packet_types = mlxsw_sp_packet_type_sfgc_types[packet_type];
 	for (i = 0; i < MLXSW_REG_SFGC_TYPE_MAX; i++) {
@@ -1675,57 +1673,56 @@ mlxsw_sp_fid_flood_table_init(struct mlxsw_sp_fid_family *fid_family,
 
 		err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfgc), sfgc_pl);
 		if (err)
-			goto err_reg_write;
+			return err;
 	}
 
 	return 0;
-
-err_reg_write:
-	mlxsw_sp_pgt_mid_free_range(mlxsw_sp, mid_base, num_fids);
-	return err;
-}
-
-static void
-mlxsw_sp_fid_flood_table_fini(struct mlxsw_sp_fid_family *fid_family,
-			      const struct mlxsw_sp_flood_table *flood_table)
-{
-	struct mlxsw_sp *mlxsw_sp = fid_family->mlxsw_sp;
-	u16 num_fids, mid_base;
-
-	mid_base = mlxsw_sp_fid_flood_table_mid(fid_family, flood_table, 0);
-	num_fids = mlxsw_sp_fid_family_num_fids(fid_family);
-	mlxsw_sp_pgt_mid_free_range(mlxsw_sp, mid_base, num_fids);
 }
 
 static int
 mlxsw_sp_fid_flood_tables_init(struct mlxsw_sp_fid_family *fid_family)
 {
+	struct mlxsw_sp *mlxsw_sp = fid_family->mlxsw_sp;
+	u16 pgt_size;
+	int err;
 	int i;
 
+	if (!fid_family->nr_flood_tables)
+		return 0;
+
+	pgt_size = mlxsw_sp_fid_family_pgt_size(fid_family);
+	err = mlxsw_sp_pgt_mid_alloc_range(mlxsw_sp, &fid_family->pgt_base,
+					   pgt_size);
+	if (err)
+		return err;
+
 	for (i = 0; i < fid_family->nr_flood_tables; i++) {
 		const struct mlxsw_sp_flood_table *flood_table;
-		int err;
 
 		flood_table = &fid_family->flood_tables[i];
 		err = mlxsw_sp_fid_flood_table_init(fid_family, flood_table);
 		if (err)
-			return err;
+			goto err_flood_table_init;
 	}
 
 	return 0;
+
+err_flood_table_init:
+	mlxsw_sp_pgt_mid_free_range(mlxsw_sp, fid_family->pgt_base, pgt_size);
+	return err;
 }
 
 static void
 mlxsw_sp_fid_flood_tables_fini(struct mlxsw_sp_fid_family *fid_family)
 {
-	int i;
+	struct mlxsw_sp *mlxsw_sp = fid_family->mlxsw_sp;
+	u16 pgt_size;
 
-	for (i = 0; i < fid_family->nr_flood_tables; i++) {
-		const struct mlxsw_sp_flood_table *flood_table;
+	if (!fid_family->nr_flood_tables)
+		return;
 
-		flood_table = &fid_family->flood_tables[i];
-		mlxsw_sp_fid_flood_table_fini(fid_family, flood_table);
-	}
+	pgt_size = mlxsw_sp_fid_family_pgt_size(fid_family);
+	mlxsw_sp_pgt_mid_free_range(mlxsw_sp, fid_family->pgt_base, pgt_size);
 }
 
 static int mlxsw_sp_fid_family_register(struct mlxsw_sp *mlxsw_sp,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_pgt.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_pgt.c
index 7dd3dba0fa..4ef81bac17 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_pgt.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_pgt.c
@@ -54,25 +54,15 @@ void mlxsw_sp_pgt_mid_free(struct mlxsw_sp *mlxsw_sp, u16 mid_base)
 	mutex_unlock(&mlxsw_sp->pgt->lock);
 }
 
-int
-mlxsw_sp_pgt_mid_alloc_range(struct mlxsw_sp *mlxsw_sp, u16 mid_base, u16 count)
+int mlxsw_sp_pgt_mid_alloc_range(struct mlxsw_sp *mlxsw_sp, u16 *p_mid_base,
+				 u16 count)
 {
-	unsigned int idr_cursor;
+	unsigned int mid_base;
 	int i, err;
 
 	mutex_lock(&mlxsw_sp->pgt->lock);
 
-	/* This function is supposed to be called several times as part of
-	 * driver init, in specific order. Verify that the mid_index is the
-	 * first free index in the idr, to be able to free the indexes in case
-	 * of error.
-	 */
-	idr_cursor = idr_get_cursor(&mlxsw_sp->pgt->pgt_idr);
-	if (WARN_ON(idr_cursor != mid_base)) {
-		err = -EINVAL;
-		goto err_idr_cursor;
-	}
-
+	mid_base = idr_get_cursor(&mlxsw_sp->pgt->pgt_idr);
 	for (i = 0; i < count; i++) {
 		err = idr_alloc_cyclic(&mlxsw_sp->pgt->pgt_idr, NULL,
 				       mid_base, mid_base + count, GFP_KERNEL);
@@ -81,12 +71,12 @@ mlxsw_sp_pgt_mid_alloc_range(struct mlxsw_sp *mlxsw_sp, u16 mid_base, u16 count)
 	}
 
 	mutex_unlock(&mlxsw_sp->pgt->lock);
+	*p_mid_base = mid_base;
 	return 0;
 
 err_idr_alloc_cyclic:
 	for (i--; i >= 0; i--)
 		idr_remove(&mlxsw_sp->pgt->pgt_idr, mid_base + i);
-err_idr_cursor:
 	mutex_unlock(&mlxsw_sp->pgt->lock);
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index ae2fb9efbc..39e6d941fd 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -3107,7 +3107,7 @@ struct mlxsw_sp_nexthop_group_info {
 	   gateway:1, /* routes using the group use a gateway */
 	   is_resilient:1;
 	struct list_head list; /* member in nh_res_grp_list */
-	struct mlxsw_sp_nexthop nexthops[];
+	struct mlxsw_sp_nexthop nexthops[] __counted_by(count);
 };
 
 static struct mlxsw_sp_rif *
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
index b3472fb946..af50ff9e5f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
@@ -31,7 +31,7 @@ struct mlxsw_sp_span {
 	refcount_t policer_id_base_ref_count;
 	atomic_t active_entries_count;
 	int entries_count;
-	struct mlxsw_sp_span_entry entries[];
+	struct mlxsw_sp_span_entry entries[] __counted_by(entries_count);
 };
 
 struct mlxsw_sp_span_analyzed_port {