Merge tag 'mlx5-updates-2023-06-09' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux

mlx5-updates-2023-06-09

1) Embedded CPU Virtual Functions
2) Lightweight local SFs

Daniel Jurgens says:
====================
Embedded CPU Virtual Functions

This series enables the creation of virtual functions on Bluefield (the
embedded CPU platform). Embedded CPU virtual functions (EC VFs). EC VF
creation, deletion and management interfaces are the same as those for
virtual functions in a server with a Connect-X NIC.

When using EC VFs on the ARM the creation of virtual functions on the
host system is still supported. Host VFs eswitch vports occupy a range
of 1..max_vfs, the EC VF vport range is max_vfs+1..max_ec_vfs.

Every function (PF, ECPF, VF, EC VF, and subfunction) has a function ID
associated with it. Prior to this series the function ID and the eswitch
vport were the same. That is no longer the case, the EC VF function ID
range is 1..max_ec_vfs. When querying or setting the capabilities of an
EC VF function an new bit must be set in the query/set HCA cap
structure.

This is a high level overview of the changes made:
	- Allocate vports for EC VFs if they are enabled.
	- Create representors and devlink ports for the EC VF vports.
	- When querying/setting HCA caps by vport break the assumption
	  that function ID is the same a vport number and adjust
	  accordingly.
	- Create a new type of page, so that when SRIOV on the ARM is
	  disabled, but remains enabled on the host, the driver can
	  wait for the correct pages.
	- Update SRIOV code to support EC VF creation/deletion.

===================

Lightweight local SFs:

Last 3 patches form Shay Drory:

SFs are heavy weight and by default they come with the full package of
ConnectX features. Usually users want specialized SFs for one specific
purpose and using devlink users will almost always override the set of
advertises features of an SF and reload it.

Shay Drory says:
================
In order to avoid the wasted time and resources on the reload, local SFs
will probe without any auxiliary sub-device, so that the SFs can be
configured prior to its full probe.

The defaults of the enable_* devlink params of these SFs are set to
false.

Usage example:
Create SF:
$ devlink port add pci/0000:08:00.0 flavour pcisf pfnum 0 sfnum 11
$ devlink port function set pci/0000:08:00.0/32768 \
               hw_addr 00:00:00:00:00:11 state active

Enable ETH auxiliary device:
$ devlink dev param set auxiliary/mlx5_core.sf.1 \
              name enable_eth value true cmode driverinit

Now, in order to fully probe the SF, use devlink reload:
$ devlink dev reload auxiliary/mlx5_core.sf.1

At this point the user have SF devlink instance with auxiliary device
for the Ethernet functionality only.

================

20 files changed, 610 insertions, 177 deletions

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/switchdev.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/switchdev.rst
index 01deedb71597..db62187eebce 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/switchdev.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/switchdev.rst
@@ -45,6 +45,26 @@ Following bridge VLAN functions are supported by mlx5:
 Subfunction
 ===========
 
+Subfunction which are spawned over the E-switch are created only with devlink
+device, and by default all the SF auxiliary devices are disabled.
+This will allow user to configure the SF before the SF have been fully probed,
+which will save time.
+
+Usage example:
+Create SF:
+$ devlink port add pci/0000:08:00.0 flavour pcisf pfnum 0 sfnum 11
+$ devlink port function set pci/0000:08:00.0/32768 \
+               hw_addr 00:00:00:00:00:11 state active
+
+Enable ETH auxiliary device:
+$ devlink dev param set auxiliary/mlx5_core.sf.1 \
+              name enable_eth value true cmode driverinit
+
+Now, in order to fully probe the SF, use devlink reload:
+$ devlink dev reload auxiliary/mlx5_core.sf.1
+
+mlx5 supports ETH,rdma and vdpa (vnet) auxiliary devices devlink params (see :ref:`Documentation/networking/devlink/devlink-params.rst`)
+
 mlx5 supports subfunction management using devlink port (see :ref:`Documentation/networking/devlink/devlink-port.rst <devlink_port>`) interface.
 
 A subfunction has its own function capabilities and its own resources. This
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c
index bb95b40d25eb..fc13b41cc9b2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c
@@ -246,6 +246,7 @@ void mlx5_pages_debugfs_init(struct mlx5_core_dev *dev)
 
 	debugfs_create_u32("fw_pages_total", 0400, pages, &dev->priv.fw_pages);
 	debugfs_create_u32("fw_pages_vfs", 0400, pages, &dev->priv.page_counters[MLX5_VF]);
+	debugfs_create_u32("fw_pages_ec_vfs", 0400, pages, &dev->priv.page_counters[MLX5_EC_VF]);
 	debugfs_create_u32("fw_pages_sfs", 0400, pages, &dev->priv.page_counters[MLX5_SF]);
 	debugfs_create_u32("fw_pages_host_pf", 0400, pages, &dev->priv.page_counters[MLX5_HOST_PF]);
 	debugfs_create_u32("fw_pages_alloc_failed", 0400, pages, &dev->priv.fw_pages_alloc_failed);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
index 1b33533b15de..617ac7e5d75c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
@@ -323,6 +323,18 @@ static void del_adev(struct auxiliary_device *adev)
 	auxiliary_device_uninit(adev);
 }
 
+void mlx5_dev_set_lightweight(struct mlx5_core_dev *dev)
+{
+	mutex_lock(&mlx5_intf_mutex);
+	dev->priv.flags |= MLX5_PRIV_FLAGS_DISABLE_ALL_ADEV;
+	mutex_unlock(&mlx5_intf_mutex);
+}
+
+bool mlx5_dev_is_lightweight(struct mlx5_core_dev *dev)
+{
+	return dev->priv.flags & MLX5_PRIV_FLAGS_DISABLE_ALL_ADEV;
+}
+
 int mlx5_attach_device(struct mlx5_core_dev *dev)
 {
 	struct mlx5_priv *priv = &dev->priv;
@@ -457,6 +469,10 @@ static int add_drivers(struct mlx5_core_dev *dev)
 		if (priv->adev[i])
 			continue;
 
+		if (mlx5_adev_devices[i].is_enabled &&
+		    !(mlx5_adev_devices[i].is_enabled(dev)))
+			continue;
+
 		if (mlx5_adev_devices[i].is_supported)
 			is_supported = mlx5_adev_devices[i].is_supported(dev);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index 63635cc44479..3d82ec890666 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -7,7 +7,6 @@
 #include "fw_reset.h"
 #include "fs_core.h"
 #include "eswitch.h"
-#include "lag/lag.h"
 #include "esw/qos.h"
 #include "sf/dev/dev.h"
 #include "sf/sf.h"
@@ -142,6 +141,13 @@ static int mlx5_devlink_reload_down(struct devlink *devlink, bool netns_change,
 	bool sf_dev_allocated;
 	int ret = 0;
 
+	if (mlx5_dev_is_lightweight(dev)) {
+		if (action != DEVLINK_RELOAD_ACTION_DRIVER_REINIT)
+			return -EOPNOTSUPP;
+		mlx5_unload_one_light(dev);
+		return 0;
+	}
+
 	sf_dev_allocated = mlx5_sf_dev_allocated(dev);
 	if (sf_dev_allocated) {
 		/* Reload results in deleting SF device which further results in
@@ -194,6 +200,10 @@ static int mlx5_devlink_reload_up(struct devlink *devlink, enum devlink_reload_a
 	*actions_performed = BIT(action);
 	switch (action) {
 	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
+		if (mlx5_dev_is_lightweight(dev)) {
+			mlx5_fw_reporters_create(dev);
+			return mlx5_init_one_devl_locked(dev);
+		}
 		ret = mlx5_load_one_devl_locked(dev, false);
 		break;
 	case DEVLINK_RELOAD_ACTION_FW_ACTIVATE:
@@ -427,33 +437,6 @@ static int mlx5_devlink_large_group_num_validate(struct devlink *devlink, u32 id
 
 	return 0;
 }
-
-static int mlx5_devlink_esw_multiport_set(struct devlink *devlink, u32 id,
-					  struct devlink_param_gset_ctx *ctx)
-{
-	struct mlx5_core_dev *dev = devlink_priv(devlink);
-
-	if (!MLX5_ESWITCH_MANAGER(dev))
-		return -EOPNOTSUPP;
-
-	if (ctx->val.vbool)
-		return mlx5_lag_mpesw_enable(dev);
-
-	mlx5_lag_mpesw_disable(dev);
-	return 0;
-}
-
-static int mlx5_devlink_esw_multiport_get(struct devlink *devlink, u32 id,
-					  struct devlink_param_gset_ctx *ctx)
-{
-	struct mlx5_core_dev *dev = devlink_priv(devlink);
-
-	if (!MLX5_ESWITCH_MANAGER(dev))
-		return -EOPNOTSUPP;
-
-	ctx->val.vbool = mlx5_lag_is_mpesw(dev);
-	return 0;
-}
 #endif
 
 static int mlx5_devlink_eq_depth_validate(struct devlink *devlink, u32 id,
@@ -527,12 +510,6 @@ static const struct devlink_param mlx5_devlink_params[] = {
 			     BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
 			     NULL, NULL,
 			     mlx5_devlink_large_group_num_validate),
-	DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_ESW_MULTIPORT,
-			     "esw_multiport", DEVLINK_PARAM_TYPE_BOOL,
-			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
-			     mlx5_devlink_esw_multiport_get,
-			     mlx5_devlink_esw_multiport_set,
-			     NULL),
 #endif
 	DEVLINK_PARAM_GENERIC(IO_EQ_SIZE, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
 			      NULL, NULL, mlx5_devlink_eq_depth_validate),
@@ -545,7 +522,7 @@ static void mlx5_devlink_set_params_init_values(struct devlink *devlink)
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
 	union devlink_param_value value;
 
-	value.vbool = MLX5_CAP_GEN(dev, roce);
+	value.vbool = MLX5_CAP_GEN(dev, roce) && !mlx5_dev_is_lightweight(dev);
 	devl_param_driverinit_value_set(devlink,
 					DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
 					value);
@@ -595,7 +572,7 @@ static int mlx5_devlink_eth_params_register(struct devlink *devlink)
 	if (err)
 		return err;
 
-	value.vbool = true;
+	value.vbool = !mlx5_dev_is_lightweight(dev);
 	devl_param_driverinit_value_set(devlink,
 					DEVLINK_PARAM_GENERIC_ID_ENABLE_ETH,
 					value);
@@ -635,6 +612,7 @@ static const struct devlink_param mlx5_devlink_rdma_params[] = {
 
 static int mlx5_devlink_rdma_params_register(struct devlink *devlink)
 {
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
 	union devlink_param_value value;
 	int err;
 
@@ -646,7 +624,7 @@ static int mlx5_devlink_rdma_params_register(struct devlink *devlink)
 	if (err)
 		return err;
 
-	value.vbool = true;
+	value.vbool = !mlx5_dev_is_lightweight(dev);
 	devl_param_driverinit_value_set(devlink,
 					DEVLINK_PARAM_GENERIC_ID_ENABLE_RDMA,
 					value);
@@ -681,7 +659,7 @@ static int mlx5_devlink_vnet_params_register(struct devlink *devlink)
 	if (err)
 		return err;
 
-	value.vbool = true;
+	value.vbool = !mlx5_dev_is_lightweight(dev);
 	devl_param_driverinit_value_set(devlink,
 					DEVLINK_PARAM_GENERIC_ID_ENABLE_VNET,
 					value);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c
index 0290e0dea539..4e923a2874ae 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c
@@ -112,10 +112,8 @@ mlx5e_tc_post_act_add(struct mlx5e_post_act *post_act, struct mlx5_flow_attr *po
 	int err;
 
 	handle = kzalloc(sizeof(*handle), GFP_KERNEL);
-	if (!handle) {
-		kfree(handle);
+	if (!handle)
 		return ERR_PTR(-ENOMEM);
-	}
 
 	post_attr->chain = 0;
 	post_attr->prio = 0;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c
index f370f67d9e33..af779c700278 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c
@@ -18,7 +18,8 @@ static bool mlx5_esw_devlink_port_supported(struct mlx5_eswitch *esw, u16 vport_
 {
 	return vport_num == MLX5_VPORT_UPLINK ||
 	       (mlx5_core_is_ecpf(esw->dev) && vport_num == MLX5_VPORT_PF) ||
-	       mlx5_eswitch_is_vf_vport(esw, vport_num);
+	       mlx5_eswitch_is_vf_vport(esw, vport_num) ||
+	       mlx5_core_is_ec_vf_vport(esw->dev, vport_num);
 }
 
 static struct devlink_port *mlx5_esw_dl_port_alloc(struct mlx5_eswitch *esw, u16 vport_num)
@@ -56,6 +57,11 @@ static struct devlink_port *mlx5_esw_dl_port_alloc(struct mlx5_eswitch *esw, u16
 		dl_port->attrs.switch_id.id_len = ppid.id_len;
 		devlink_port_attrs_pci_vf_set(dl_port, controller_num, pfnum,
 					      vport_num - 1, external);
+	}  else if (mlx5_core_is_ec_vf_vport(esw->dev, vport_num)) {
+		memcpy(dl_port->attrs.switch_id.id, ppid.id, ppid.id_len);
+		dl_port->attrs.switch_id.id_len = ppid.id_len;
+		devlink_port_attrs_pci_vf_set(dl_port, controller_num, pfnum,
+					      vport_num - 1, false);
 	}
 	return dl_port;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index ecd8864d5d11..2af9c4646bc7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -41,6 +41,7 @@
 #include "esw/qos.h"
 #include "mlx5_core.h"
 #include "lib/eq.h"
+#include "lag/lag.h"
 #include "eswitch.h"
 #include "fs_core.h"
 #include "devlink.h"
@@ -1051,6 +1052,18 @@ static void mlx5_eswitch_clear_vf_vports_info(struct mlx5_eswitch *esw)
 	}
 }
 
+static void mlx5_eswitch_clear_ec_vf_vports_info(struct mlx5_eswitch *esw)
+{
+	struct mlx5_vport *vport;
+	unsigned long i;
+
+	mlx5_esw_for_each_ec_vf_vport(esw, i, vport, esw->esw_funcs.num_ec_vfs) {
+		memset(&vport->qos, 0, sizeof(vport->qos));
+		memset(&vport->info, 0, sizeof(vport->info));
+		vport->info.link_state = MLX5_VPORT_ADMIN_STATE_AUTO;
+	}
+}
+
 /* Public E-Switch API */
 int mlx5_eswitch_load_vport(struct mlx5_eswitch *esw, u16 vport_num,
 			    enum mlx5_eswitch_vport_event enabled_events)
@@ -1090,6 +1103,19 @@ void mlx5_eswitch_unload_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs)
 	}
 }
 
+static void mlx5_eswitch_unload_ec_vf_vports(struct mlx5_eswitch *esw,
+					     u16 num_ec_vfs)
+{
+	struct mlx5_vport *vport;
+	unsigned long i;
+
+	mlx5_esw_for_each_ec_vf_vport(esw, i, vport, num_ec_vfs) {
+		if (!vport->enabled)
+			continue;
+		mlx5_eswitch_unload_vport(esw, vport->vport);
+	}
+}
+
 int mlx5_eswitch_load_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs,
 				enum mlx5_eswitch_vport_event enabled_events)
 {
@@ -1110,6 +1136,26 @@ vf_err:
 	return err;
 }
 
+static int mlx5_eswitch_load_ec_vf_vports(struct mlx5_eswitch *esw, u16 num_ec_vfs,
+					  enum mlx5_eswitch_vport_event enabled_events)
+{
+	struct mlx5_vport *vport;
+	unsigned long i;
+	int err;
+
+	mlx5_esw_for_each_ec_vf_vport(esw, i, vport, num_ec_vfs) {
+		err = mlx5_eswitch_load_vport(esw, vport->vport, enabled_events);
+		if (err)
+			goto vf_err;
+	}
+
+	return 0;
+
+vf_err:
+	mlx5_eswitch_unload_ec_vf_vports(esw, num_ec_vfs);
+	return err;
+}
+
 static int host_pf_enable_hca(struct mlx5_core_dev *dev)
 {
 	if (!mlx5_core_is_ecpf(dev))
@@ -1154,6 +1200,12 @@ mlx5_eswitch_enable_pf_vf_vports(struct mlx5_eswitch *esw,
 		ret = mlx5_eswitch_load_vport(esw, MLX5_VPORT_ECPF, enabled_events);
 		if (ret)
 			goto ecpf_err;
+		if (mlx5_core_ec_sriov_enabled(esw->dev)) {
+			ret = mlx5_eswitch_load_ec_vf_vports(esw, esw->esw_funcs.num_ec_vfs,
+							     enabled_events);
+			if (ret)
+				goto ec_vf_err;
+		}
 	}
 
 	/* Enable VF vports */
@@ -1164,6 +1216,9 @@ mlx5_eswitch_enable_pf_vf_vports(struct mlx5_eswitch *esw,
 	return 0;
 
 vf_err:
+	if (mlx5_core_ec_sriov_enabled(esw->dev))
+		mlx5_eswitch_unload_ec_vf_vports(esw, esw->esw_funcs.num_ec_vfs);
+ec_vf_err:
 	if (mlx5_ecpf_vport_exists(esw->dev))
 		mlx5_eswitch_unload_vport(esw, MLX5_VPORT_ECPF);
 ecpf_err:
@@ -1180,8 +1235,11 @@ void mlx5_eswitch_disable_pf_vf_vports(struct mlx5_eswitch *esw)
 {
 	mlx5_eswitch_unload_vf_vports(esw, esw->esw_funcs.num_vfs);
 
-	if (mlx5_ecpf_vport_exists(esw->dev))
+	if (mlx5_ecpf_vport_exists(esw->dev)) {
+		if (mlx5_core_ec_sriov_enabled(esw->dev))
+			mlx5_eswitch_unload_ec_vf_vports(esw, esw->esw_funcs.num_vfs);
 		mlx5_eswitch_unload_vport(esw, MLX5_VPORT_ECPF);
+	}
 
 	host_pf_disable_hca(esw->dev);
 	mlx5_eswitch_unload_vport(esw, MLX5_VPORT_PF);
@@ -1225,6 +1283,9 @@ mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, int num_vfs)
 
 	esw->esw_funcs.num_vfs = MLX5_GET(query_esw_functions_out, out,
 					  host_params_context.host_num_of_vfs);
+	if (mlx5_core_ec_sriov_enabled(esw->dev))
+		esw->esw_funcs.num_ec_vfs = num_vfs;
+
 	kvfree(out);
 }
 
@@ -1332,9 +1393,9 @@ int mlx5_eswitch_enable_locked(struct mlx5_eswitch *esw, int num_vfs)
 
 	mlx5_eswitch_event_handlers_register(esw);
 
-	esw_info(esw->dev, "Enable: mode(%s), nvfs(%d), active vports(%d)\n",
+	esw_info(esw->dev, "Enable: mode(%s), nvfs(%d), necvfs(%d), active vports(%d)\n",
 		 esw->mode == MLX5_ESWITCH_LEGACY ? "LEGACY" : "OFFLOADS",
-		 esw->esw_funcs.num_vfs, esw->enabled_vports);
+		 esw->esw_funcs.num_vfs, esw->esw_funcs.num_ec_vfs, esw->enabled_vports);
 
 	mlx5_esw_mode_change_notify(esw, esw->mode);
 
@@ -1356,7 +1417,7 @@ abort:
 int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs)
 {
 	bool toggle_lag;
-	int ret;
+	int ret = 0;
 
 	if (!mlx5_esw_allowed(esw))
 		return 0;
@@ -1376,10 +1437,21 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs)
 
 		vport_events = (esw->mode == MLX5_ESWITCH_LEGACY) ?
 					MLX5_LEGACY_SRIOV_VPORT_EVENTS : MLX5_VPORT_UC_ADDR_CHANGE;
-		ret = mlx5_eswitch_load_vf_vports(esw, num_vfs, vport_events);
-		if (!ret)
-			esw->esw_funcs.num_vfs = num_vfs;
+		/* If this is the ECPF the number of host VFs is managed via the
+		 * eswitch function change event handler, and any num_vfs provided
+		 * here are intended to be EC VFs.
+		 */
+		if (!mlx5_core_is_ecpf(esw->dev)) {
+			ret = mlx5_eswitch_load_vf_vports(esw, num_vfs, vport_events);
+			if (!ret)
+				esw->esw_funcs.num_vfs = num_vfs;
+		} else if (mlx5_core_ec_sriov_enabled(esw->dev)) {
+			ret = mlx5_eswitch_load_ec_vf_vports(esw, num_vfs, vport_events);
+			if (!ret)
+				esw->esw_funcs.num_ec_vfs = num_vfs;
+		}
 	}
+
 	up_write(&esw->mode_lock);
 
 	if (toggle_lag)
@@ -1399,16 +1471,22 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf)
 	/* If driver is unloaded, this function is called twice by remove_one()
 	 * and mlx5_unload(). Prevent the second call.
 	 */
-	if (!esw->esw_funcs.num_vfs && !clear_vf)
+	if (!esw->esw_funcs.num_vfs && !esw->esw_funcs.num_ec_vfs && !clear_vf)
 		goto unlock;
 
-	esw_info(esw->dev, "Unload vfs: mode(%s), nvfs(%d), active vports(%d)\n",
+	esw_info(esw->dev, "Unload vfs: mode(%s), nvfs(%d), necvfs(%d), active vports(%d)\n",
 		 esw->mode == MLX5_ESWITCH_LEGACY ? "LEGACY" : "OFFLOADS",
-		 esw->esw_funcs.num_vfs, esw->enabled_vports);
-
-	mlx5_eswitch_unload_vf_vports(esw, esw->esw_funcs.num_vfs);
-	if (clear_vf)
-		mlx5_eswitch_clear_vf_vports_info(esw);
+		 esw->esw_funcs.num_vfs, esw->esw_funcs.num_ec_vfs, esw->enabled_vports);
+
+	if (!mlx5_core_is_ecpf(esw->dev)) {
+		mlx5_eswitch_unload_vf_vports(esw, esw->esw_funcs.num_vfs);
+		if (clear_vf)
+			mlx5_eswitch_clear_vf_vports_info(esw);
+	} else if (mlx5_core_ec_sriov_enabled(esw->dev)) {
+		mlx5_eswitch_unload_ec_vf_vports(esw, esw->esw_funcs.num_ec_vfs);
+		if (clear_vf)
+			mlx5_eswitch_clear_ec_vf_vports_info(esw);
+	}
 
 	if (esw->mode == MLX5_ESWITCH_OFFLOADS) {
 		struct devlink *devlink = priv_to_devlink(esw->dev);
@@ -1419,7 +1497,10 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf)
 	if (esw->mode == MLX5_ESWITCH_LEGACY)
 		mlx5_eswitch_disable_locked(esw);
 
-	esw->esw_funcs.num_vfs = 0;
+	if (!mlx5_core_is_ecpf(esw->dev))
+		esw->esw_funcs.num_vfs = 0;
+	else
+		esw->esw_funcs.num_ec_vfs = 0;
 
 unlock:
 	up_write(&esw->mode_lock);
@@ -1439,9 +1520,9 @@ void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw)
 
 	mlx5_eswitch_event_handlers_unregister(esw);
 
-	esw_info(esw->dev, "Disable: mode(%s), nvfs(%d), active vports(%d)\n",
+	esw_info(esw->dev, "Disable: mode(%s), nvfs(%d), necvfs(%d), active vports(%d)\n",
 		 esw->mode == MLX5_ESWITCH_LEGACY ? "LEGACY" : "OFFLOADS",
-		 esw->esw_funcs.num_vfs, esw->enabled_vports);
+		 esw->esw_funcs.num_vfs, esw->esw_funcs.num_ec_vfs, esw->enabled_vports);
 
 	if (esw->fdb_table.flags & MLX5_ESW_FDB_CREATED) {
 		esw->fdb_table.flags &= ~MLX5_ESW_FDB_CREATED;
@@ -1601,6 +1682,17 @@ static int mlx5_esw_vports_init(struct mlx5_eswitch *esw)
 		idx++;
 	}
 
+	if (mlx5_core_ec_sriov_enabled(esw->dev)) {
+		int ec_vf_base_num = mlx5_core_ec_vf_vport_base(dev);
+
+		for (i = 0; i < mlx5_core_max_ec_vfs(esw->dev); i++) {
+			err = mlx5_esw_vport_alloc(esw, idx, ec_vf_base_num + i);
+			if (err)
+				goto err;
+			idx++;
+		}
+	}
+
 	if (mlx5_ecpf_vport_exists(dev) ||
 	    mlx5_core_is_ecpf_esw_manager(dev)) {
 		err = mlx5_esw_vport_alloc(esw, idx, MLX5_VPORT_ECPF);
@@ -1618,6 +1710,38 @@ err:
 	return err;
 }
 
+static int mlx5_devlink_esw_multiport_set(struct devlink *devlink, u32 id,
+					  struct devlink_param_gset_ctx *ctx)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+
+	if (!MLX5_ESWITCH_MANAGER(dev))
+		return -EOPNOTSUPP;
+
+	if (ctx->val.vbool)
+		return mlx5_lag_mpesw_enable(dev);
+
+	mlx5_lag_mpesw_disable(dev);
+	return 0;
+}
+
+static int mlx5_devlink_esw_multiport_get(struct devlink *devlink, u32 id,
+					  struct devlink_param_gset_ctx *ctx)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+
+	ctx->val.vbool = mlx5_lag_is_mpesw(dev);
+	return 0;
+}
+
+static const struct devlink_param mlx5_eswitch_params[] = {
+	DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_ESW_MULTIPORT,
+			     "esw_multiport", DEVLINK_PARAM_TYPE_BOOL,
+			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+			     mlx5_devlink_esw_multiport_get,
+			     mlx5_devlink_esw_multiport_set, NULL),
+};
+
 int mlx5_eswitch_init(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eswitch *esw;
@@ -1626,9 +1750,16 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
 	if (!MLX5_VPORT_MANAGER(dev) && !MLX5_ESWITCH_MANAGER(dev))
 		return 0;
 
+	err = devl_params_register(priv_to_devlink(dev), mlx5_eswitch_params,
+				   ARRAY_SIZE(mlx5_eswitch_params));
+	if (err)
+		return err;
+
 	esw = kzalloc(sizeof(*esw), GFP_KERNEL);
-	if (!esw)
-		return -ENOMEM;
+	if (!esw) {
+		err = -ENOMEM;
+		goto unregister_param;
+	}
 
 	esw->dev = dev;
 	esw->manager_vport = mlx5_eswitch_manager_vport(dev);
@@ -1688,6 +1819,9 @@ abort:
 	if (esw->work_queue)
 		destroy_workqueue(esw->work_queue);
 	kfree(esw);
+unregister_param:
+	devl_params_unregister(priv_to_devlink(dev), mlx5_eswitch_params,
+			       ARRAY_SIZE(mlx5_eswitch_params));
 	return err;
 }
 
@@ -1711,6 +1845,8 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
 	esw_offloads_cleanup(esw);
 	mlx5_esw_vports_cleanup(esw);
 	kfree(esw);
+	devl_params_unregister(priv_to_devlink(esw->dev), mlx5_eswitch_params,
+			       ARRAY_SIZE(mlx5_eswitch_params));
 }
 
 /* Vport Administration */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index d3608f198e0a..266b60fefe25 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -289,6 +289,7 @@ struct mlx5_host_work {
 struct mlx5_esw_functions {
 	struct mlx5_nb		nb;
 	u16			num_vfs;
+	u16			num_ec_vfs;
 };
 
 enum {
@@ -654,6 +655,18 @@ void mlx5e_tc_clean_fdb_peer_flows(struct mlx5_eswitch *esw);
 #define mlx5_esw_for_each_host_func_vport(esw, index, vport, last)	\
 	mlx5_esw_for_each_vport_marked(esw, index, vport, last, MLX5_ESW_VPT_HOST_FN)
 
+/* This macro should only be used if EC SRIOV is enabled.
+ *
+ * Because there were no more marks available on the xarray this uses a
+ * for_each_range approach. The range is only valid when EC SRIOV is enabled
+ */
+#define mlx5_esw_for_each_ec_vf_vport(esw, index, vport, last)		\
+	xa_for_each_range(&((esw)->vports),				\
+			  index,					\
+			  vport,					\
+			  MLX5_CAP_GEN_2((esw->dev), ec_vf_vport_base),	\
+			  (last) - 1)
+
 struct mlx5_eswitch *mlx5_devlink_eswitch_get(struct devlink *devlink);
 struct mlx5_vport *__must_check
 mlx5_eswitch_get_vport(struct mlx5_eswitch *esw, u16 vport_num);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index eafb098db6b0..fdf482f6fb34 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -55,13 +55,6 @@
 #define mlx5_esw_for_each_rep(esw, i, rep) \
 	xa_for_each(&((esw)->offloads.vport_reps), i, rep)
 
-#define mlx5_esw_for_each_sf_rep(esw, i, rep) \
-	xa_for_each_marked(&((esw)->offloads.vport_reps), i, rep, MLX5_ESW_VPT_SF)
-
-#define mlx5_esw_for_each_vf_rep(esw, index, rep)	\
-	mlx5_esw_for_each_entry_marked(&((esw)->offloads.vport_reps), index, \
-				       rep, (esw)->esw_funcs.num_vfs, MLX5_ESW_VPT_VF)
-
 /* There are two match-all miss flows, one for unicast dst mac and
  * one for multicast.
  */
@@ -1132,11 +1125,32 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
 		flows[vport->index] = flow;
 	}
 
+	if (mlx5_core_ec_sriov_enabled(esw->dev)) {
+		mlx5_esw_for_each_ec_vf_vport(esw, i, vport, mlx5_core_max_ec_vfs(esw->dev)) {
+			if (i >= mlx5_core_max_ec_vfs(peer_dev))
+				break;
+			esw_set_peer_miss_rule_source_port(esw, peer_dev->priv.eswitch,
+							   spec, vport->vport);
+			flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
+						   spec, &flow_act, &dest, 1);
+			if (IS_ERR(flow)) {
+				err = PTR_ERR(flow);
+				goto add_ec_vf_flow_err;
+			}
+			flows[vport->index] = flow;
+		}
+	}
 	esw->fdb_table.offloads.peer_miss_rules[mlx5_get_dev_index(peer_dev)] = flows;
 
 	kvfree(spec);
 	return 0;
 
+add_ec_vf_flow_err:
+	mlx5_esw_for_each_ec_vf_vport(esw, i, vport, mlx5_core_max_ec_vfs(esw->dev)) {
+		if (!flows[vport->index])
+			continue;
+		mlx5_del_flow_rules(flows[vport->index]);
+	}
 add_vf_flow_err:
 	mlx5_esw_for_each_vf_vport(esw, i, vport, mlx5_core_max_vfs(esw->dev)) {
 		if (!flows[vport->index])
@@ -1169,6 +1183,17 @@ static void esw_del_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
 
 	flows = esw->fdb_table.offloads.peer_miss_rules[mlx5_get_dev_index(peer_dev)];
 
+	if (mlx5_core_ec_sriov_enabled(esw->dev)) {
+		mlx5_esw_for_each_ec_vf_vport(esw, i, vport, mlx5_core_max_ec_vfs(esw->dev)) {
+			/* The flow for a particular vport could be NULL if the other ECPF
+			 * has fewer or no VFs enabled
+			 */
+			if (!flows[vport->index])
+				continue;
+			mlx5_del_flow_rules(flows[vport->index]);
+		}
+	}
+
 	mlx5_esw_for_each_vf_vport(esw, i, vport, mlx5_core_max_vfs(esw->dev))
 		mlx5_del_flow_rules(flows[vport->index]);
 
@@ -2191,18 +2216,6 @@ static int esw_offloads_start(struct mlx5_eswitch *esw,
 	return 0;
 }
 
-static void mlx5_esw_offloads_rep_mark_set(struct mlx5_eswitch *esw,
-					   struct mlx5_eswitch_rep *rep,
-					   xa_mark_t mark)
-{
-	bool mark_set;
-
-	/* Copy the mark from vport to its rep */
-	mark_set = xa_get_mark(&esw->vports, rep->vport, mark);
-	if (mark_set)
-		xa_set_mark(&esw->offloads.vport_reps, rep->vport, mark);
-}
-
 static int mlx5_esw_offloads_rep_init(struct mlx5_eswitch *esw, const struct mlx5_vport *vport)
 {
 	struct mlx5_eswitch_rep *rep;
@@ -2222,9 +2235,6 @@ static int mlx5_esw_offloads_rep_init(struct mlx5_eswitch *esw, const struct mlx
 	if (err)
 		goto insert_err;
 
-	mlx5_esw_offloads_rep_mark_set(esw, rep, MLX5_ESW_VPT_HOST_FN);
-	mlx5_esw_offloads_rep_mark_set(esw, rep, MLX5_ESW_VPT_VF);
-	mlx5_esw_offloads_rep_mark_set(esw, rep, MLX5_ESW_VPT_SF);
 	return 0;
 
 insert_err:
@@ -2365,37 +2375,13 @@ static void __esw_offloads_unload_rep(struct mlx5_eswitch *esw,
 		esw->offloads.rep_ops[rep_type]->unload(rep);
 }
 
-static void __unload_reps_sf_vport(struct mlx5_eswitch *esw, u8 rep_type)
-{
-	struct mlx5_eswitch_rep *rep;
-	unsigned long i;
-
-	mlx5_esw_for_each_sf_rep(esw, i, rep)
-		__esw_offloads_unload_rep(esw, rep, rep_type);
-}
-
 static void __unload_reps_all_vport(struct mlx5_eswitch *esw, u8 rep_type)
 {
 	struct mlx5_eswitch_rep *rep;
 	unsigned long i;
 
-	__unload_reps_sf_vport(esw, rep_type);
-
-	mlx5_esw_for_each_vf_rep(esw, i, rep)
-		__esw_offloads_unload_rep(esw, rep, rep_type);
-
-	if (mlx5_ecpf_vport_exists(esw->dev)) {
-		rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_ECPF);
-		__esw_offloads_unload_rep(esw, rep, rep_type);
-	}
-
-	if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
-		rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_PF);
+	mlx5_esw_for_each_rep(esw, i, rep)
 		__esw_offloads_unload_rep(esw, rep, rep_type);
-	}
-
-	rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
-	__esw_offloads_unload_rep(esw, rep, rep_type);
 }
 
 int mlx5_esw_offloads_rep_load(struct mlx5_eswitch *esw, u16 vport_num)
@@ -3333,6 +3319,9 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
 	/* Representor will control the vport link state */
 	mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs)
 		vport->info.link_state = MLX5_VPORT_ADMIN_STATE_DOWN;
+	if (mlx5_core_ec_sriov_enabled(esw->dev))
+		mlx5_esw_for_each_ec_vf_vport(esw, i, vport, esw->esw_funcs.num_ec_vfs)
+			vport->info.link_state = MLX5_VPORT_ADMIN_STATE_DOWN;
 
 	/* Uplink vport rep must load first. */
 	err = esw_offloads_load_rep(esw, MLX5_VPORT_UPLINK);
@@ -3570,8 +3559,27 @@ static int mlx5_esw_vports_inline_set(struct mlx5_eswitch *esw, u8 mlx5_mode,
 			goto revert_inline_mode;
 		}
 	}
+	if (mlx5_core_ec_sriov_enabled(esw->dev)) {
+		mlx5_esw_for_each_ec_vf_vport(esw, i, vport, esw->esw_funcs.num_ec_vfs) {
+			err = mlx5_modify_nic_vport_min_inline(dev, vport->vport, mlx5_mode);
+			if (err) {
+				err_vport_num = vport->vport;
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Failed to set min inline on vport");
+				goto revert_ec_vf_inline_mode;
+			}
+		}
+	}
 	return 0;
 
+revert_ec_vf_inline_mode:
+	mlx5_esw_for_each_ec_vf_vport(esw, i, vport, esw->esw_funcs.num_ec_vfs) {
+		if (vport->vport == err_vport_num)
+			break;
+		mlx5_modify_nic_vport_min_inline(dev,
+						 vport->vport,
+						 esw->offloads.inline_mode);
+	}
 revert_inline_mode:
 	mlx5_esw_for_each_host_func_vport(esw, i, vport, esw->