Merge tag 'v6.1-rc8' into rdma.git for-next

For dependencies in following patches

Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

32 files changed, 541 insertions, 313 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c
index b149e601f673..48cfaa7eaf50 100644
--- a/drivers/net/ethernet/mellanox/mlx4/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx4/qp.c
@@ -697,7 +697,8 @@ static int mlx4_create_zones(struct mlx4_dev *dev,
 			err = mlx4_bitmap_init(*bitmap + k, 1,
 					       MLX4_QP_TABLE_RAW_ETH_SIZE - 1, 0,
 					       0);
-			mlx4_bitmap_alloc_range(*bitmap + k, 1, 1, 0);
+			if (!err)
+				mlx4_bitmap_alloc_range(*bitmap + k, 1, 1, 0);
 		}
 
 		if (err)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 46ba4c2faad2..e7a894ba5c3e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -45,6 +45,8 @@
 #include "mlx5_core.h"
 #include "lib/eq.h"
 #include "lib/tout.h"
+#define CREATE_TRACE_POINTS
+#include "diag/cmd_tracepoint.h"
 
 enum {
 	CMD_IF_REV = 5,
@@ -785,27 +787,14 @@ EXPORT_SYMBOL(mlx5_cmd_out_err);
 static void cmd_status_print(struct mlx5_core_dev *dev, void *in, void *out)
 {
 	u16 opcode, op_mod;
-	u32 syndrome;
-	u8  status;
 	u16 uid;
-	int err;
-
-	syndrome = MLX5_GET(mbox_out, out, syndrome);
-	status = MLX5_GET(mbox_out, out, status);
 
 	opcode = MLX5_GET(mbox_in, in, opcode);
 	op_mod = MLX5_GET(mbox_in, in, op_mod);
 	uid    = MLX5_GET(mbox_in, in, uid);
 
-	err = cmd_status_to_err(status);
-
 	if (!uid && opcode != MLX5_CMD_OP_DESTROY_MKEY)
 		mlx5_cmd_out_err(dev, opcode, op_mod, out);
-	else
-		mlx5_core_dbg(dev,
-			"%s(0x%x) op_mod(0x%x) uid(%d) failed, status %s(0x%x), syndrome (0x%x), err(%d)\n",
-			mlx5_command_str(opcode), opcode, op_mod, uid,
-			cmd_status_str(status), status, syndrome, err);
 }
 
 int mlx5_cmd_check(struct mlx5_core_dev *dev, int err, void *in, void *out)
@@ -1016,6 +1005,7 @@ static void cmd_work_handler(struct work_struct *work)
 		cmd_ent_get(ent);
 	set_bit(MLX5_CMD_ENT_STATE_PENDING_COMP, &ent->state);
 
+	cmd_ent_get(ent); /* for the _real_ FW event on completion */
 	/* Skip sending command to fw if internal error */
 	if (mlx5_cmd_is_down(dev) || !opcode_allowed(&dev->cmd, ent->op)) {
 		ent->ret = -ENXIO;
@@ -1023,7 +1013,6 @@ static void cmd_work_handler(struct work_struct *work)
 		return;
 	}
 
-	cmd_ent_get(ent); /* for the _real_ FW event on completion */
 	/* ring doorbell after the descriptor is valid */
 	mlx5_core_dbg(dev, "writing 0x%x to command doorbell\n", 1 << ent->idx);
 	wmb();
@@ -1508,8 +1497,8 @@ static ssize_t outlen_write(struct file *filp, const char __user *buf,
 		return -EFAULT;
 
 	err = sscanf(outlen_str, "%d", &outlen);
-	if (err < 0)
-		return err;
+	if (err != 1)
+		return -EINVAL;
 
 	ptr = kzalloc(outlen, GFP_KERNEL);
 	if (!ptr)
@@ -1672,8 +1661,8 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force
 				cmd_ent_put(ent); /* timeout work was canceled */
 
 			if (!forced || /* Real FW completion */
-			    pci_channel_offline(dev->pdev) || /* FW is inaccessible */
-			    dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR)
+			     mlx5_cmd_is_down(dev) || /* No real FW completion is expected */
+			     !opcode_allowed(cmd, ent->op))
 				cmd_ent_put(ent);
 
 			ent->ts2 = ktime_get_ns();
@@ -1770,12 +1759,17 @@ void mlx5_cmd_flush(struct mlx5_core_dev *dev)
 	struct mlx5_cmd *cmd = &dev->cmd;
 	int i;
 
-	for (i = 0; i < cmd->max_reg_cmds; i++)
-		while (down_trylock(&cmd->sem))
+	for (i = 0; i < cmd->max_reg_cmds; i++) {
+		while (down_trylock(&cmd->sem)) {
 			mlx5_cmd_trigger_completions(dev);
+			cond_resched();
+		}
+	}
 
-	while (down_trylock(&cmd->pages_sem))
+	while (down_trylock(&cmd->pages_sem)) {
 		mlx5_cmd_trigger_completions(dev);
+		cond_resched();
+	}
 
 	/* Unlock cmdif */
 	up(&cmd->pages_sem);
@@ -1887,6 +1881,16 @@ out_in:
 	return err;
 }
 
+static void mlx5_cmd_err_trace(struct mlx5_core_dev *dev, u16 opcode, u16 op_mod, void *out)
+{
+	u32 syndrome = MLX5_GET(mbox_out, out, syndrome);
+	u8 status = MLX5_GET(mbox_out, out, status);
+
+	trace_mlx5_cmd(mlx5_command_str(opcode), opcode, op_mod,
+		       cmd_status_str(status), status, syndrome,
+		       cmd_status_to_err(status));
+}
+
 static void cmd_status_log(struct mlx5_core_dev *dev, u16 opcode, u8 status,
 			   u32 syndrome, int err)
 {
@@ -1909,7 +1913,7 @@ static void cmd_status_log(struct mlx5_core_dev *dev, u16 opcode, u8 status,
 }
 
 /* preserve -EREMOTEIO for outbox.status != OK, otherwise return err as is */
-static int cmd_status_err(struct mlx5_core_dev *dev, int err, u16 opcode, void *out)
+static int cmd_status_err(struct mlx5_core_dev *dev, int err, u16 opcode, u16 op_mod, void *out)
 {
 	u32 syndrome = MLX5_GET(mbox_out, out, syndrome);
 	u8 status = MLX5_GET(mbox_out, out, status);
@@ -1917,8 +1921,10 @@ static int cmd_status_err(struct mlx5_core_dev *dev, int err, u16 opcode, void *
 	if (err == -EREMOTEIO) /* -EREMOTEIO is preserved */
 		err = -EIO;
 
-	if (!err && status != MLX5_CMD_STAT_OK)
+	if (!err && status != MLX5_CMD_STAT_OK) {
 		err = -EREMOTEIO;
+		mlx5_cmd_err_trace(dev, opcode, op_mod, out);
+	}
 
 	cmd_status_log(dev, opcode, status, syndrome, err);
 	return err;
@@ -1946,9 +1952,9 @@ int mlx5_cmd_do(struct mlx5_core_dev *dev, void *in, int in_size, void *out, int
 {
 	int err = cmd_exec(dev, in, in_size, out, out_size, NULL, NULL, false);
 	u16 opcode = MLX5_GET(mbox_in, in, opcode);
+	u16 op_mod = MLX5_GET(mbox_in, in, op_mod);
 
-	err = cmd_status_err(dev, err, opcode, out);
-	return err;
+	return cmd_status_err(dev, err, opcode, op_mod, out);
 }
 EXPORT_SYMBOL(mlx5_cmd_do);
 
@@ -1992,8 +1998,9 @@ int mlx5_cmd_exec_polling(struct mlx5_core_dev *dev, void *in, int in_size,
 {
 	int err = cmd_exec(dev, in, in_size, out, out_size, NULL, NULL, true);
 	u16 opcode = MLX5_GET(mbox_in, in, opcode);
+	u16 op_mod = MLX5_GET(mbox_in, in, op_mod);
 
-	err = cmd_status_err(dev, err, opcode, out);
+	err = cmd_status_err(dev, err, opcode, op_mod, out);
 	return mlx5_cmd_check(dev, err, in, out);
 }
 EXPORT_SYMBOL(mlx5_cmd_exec_polling);
@@ -2029,7 +2036,7 @@ static void mlx5_cmd_exec_cb_handler(int status, void *_work)
 	struct mlx5_async_ctx *ctx;
 
 	ctx = work->ctx;
-	status = cmd_status_err(ctx->dev, status, work->opcode, work->out);
+	status = cmd_status_err(ctx->dev, status, work->opcode, work->op_mod, work->out);
 	work->user_callback(status, work);
 	if (atomic_dec_and_test(&ctx->num_inflight))
 		complete(&ctx->inflight_done);
@@ -2044,6 +2051,7 @@ int mlx5_cmd_exec_cb(struct mlx5_async_ctx *ctx, void *in, int in_size,
 	work->ctx = ctx;
 	work->user_callback = callback;
 	work->opcode = MLX5_GET(mbox_in, in, opcode);
+	work->op_mod = MLX5_GET(mbox_in, in, op_mod);
 	work->out = out;
 	if (WARN_ON(!atomic_inc_not_zero(&ctx->num_inflight)))
 		return -EIO;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h
new file mode 100644
index 000000000000..406ebe17405f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM mlx5
+
+#if !defined(_MLX5_CMD_TP_H_) || defined(TRACE_HEADER_MULTI_READ)
+#define _MLX5_CMD_TP_H_
+
+#include <linux/tracepoint.h>
+#include <linux/trace_seq.h>
+
+TRACE_EVENT(mlx5_cmd,
+	    TP_PROTO(const char *command_str, u16 opcode, u16 op_mod,
+		     const char *status_str, u8 status, u32 syndrome, int err),
+	    TP_ARGS(command_str, opcode, op_mod, status_str, status, syndrome, err),
+	    TP_STRUCT__entry(__string(command_str, command_str)
+			     __field(u16, opcode)
+			     __field(u16, op_mod)
+			    __string(status_str, status_str)
+			    __field(u8, status)
+			    __field(u32, syndrome)
+			    __field(int, err)
+			    ),
+	    TP_fast_assign(__assign_str(command_str, command_str);
+			__entry->opcode = opcode;
+			__entry->op_mod = op_mod;
+			__assign_str(status_str, status_str);
+			__entry->status = status;
+			__entry->syndrome = syndrome;
+			__entry->err = err;
+	    ),
+	    TP_printk("%s(0x%x) op_mod(0x%x) failed, status %s(0x%x), syndrome (0x%x), err(%d)",
+		      __get_str(command_str), __entry->opcode, __entry->op_mod,
+		      __get_str(status_str), __entry->status, __entry->syndrome,
+		      __entry->err)
+);
+
+#endif /* _MLX5_CMD_TP_H_ */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH ./diag
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE cmd_tracepoint
+#include <trace/define_trace.h>
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 978a2bb8e122..21831386b26e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -638,7 +638,7 @@ static void mlx5_tracer_handle_timestamp_trace(struct mlx5_fw_tracer *tracer,
 			trace_timestamp = (timestamp_event.timestamp & MASK_52_7) |
 					  (str_frmt->timestamp & MASK_6_0);
 		else
-			trace_timestamp = ((timestamp_event.timestamp & MASK_52_7) - 1) |
+			trace_timestamp = ((timestamp_event.timestamp - 1) & MASK_52_7) |
 					  (str_frmt->timestamp & MASK_6_0);
 
 		mlx5_tracer_print_trace(str_frmt, dev, trace_timestamp);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
index 39ef2a2561a3..8099a21e674c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
@@ -164,6 +164,36 @@ static int mlx5_esw_bridge_port_changeupper(struct notifier_block *nb, void *ptr
 	return err;
 }
 
+static int
+mlx5_esw_bridge_changeupper_validate_netdev(void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	struct netdev_notifier_changeupper_info *info = ptr;
+	struct net_device *upper = info->upper_dev;
+	struct net_device *lower;
+	struct list_head *iter;
+
+	if (!netif_is_bridge_master(upper) || !netif_is_lag_master(dev))
+		return 0;
+
+	netdev_for_each_lower_dev(dev, lower, iter) {
+		struct mlx5_core_dev *mdev;
+		struct mlx5e_priv *priv;
+
+		if (!mlx5e_eswitch_rep(lower))
+			continue;
+
+		priv = netdev_priv(lower);
+		mdev = priv->mdev;
+		if (!mlx5_lag_is_active(mdev))
+			return -EAGAIN;
+		if (!mlx5_lag_is_shared_fdb(mdev))
+			return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
 static int mlx5_esw_bridge_switchdev_port_event(struct notifier_block *nb,
 						unsigned long event, void *ptr)
 {
@@ -171,6 +201,7 @@ static int mlx5_esw_bridge_switchdev_port_event(struct notifier_block *nb,
 
 	switch (event) {
 	case NETDEV_PRECHANGEUPPER:
+		err = mlx5_esw_bridge_changeupper_validate_netdev(ptr);
 		break;
 
 	case NETDEV_CHANGEUPPER:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c
index 305fde62a78d..3337241cfd84 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c
@@ -6,70 +6,42 @@
 #include "en/tc_priv.h"
 #include "mlx5_core.h"
 
-/* Must be aligned with enum flow_action_id. */
 static struct mlx5e_tc_act *tc_acts_fdb[NUM_FLOW_ACTIONS] = {
-	&mlx5e_tc_act_accept,
-	&mlx5e_tc_act_drop,
-	&mlx5e_tc_act_trap,
-	&mlx5e_tc_act_goto,
-	&mlx5e_tc_act_mirred,
-	&mlx5e_tc_act_mirred,
-	&mlx5e_tc_act_redirect_ingress,
-	NULL, /* FLOW_ACTION_MIRRED_INGRESS, */
-	&mlx5e_tc_act_vlan,
-	&mlx5e_tc_act_vlan,
-	&mlx5e_tc_act_vlan_mangle,
-	&mlx5e_tc_act_tun_encap,
-	&mlx5e_tc_act_tun_decap,
-	&mlx5e_tc_act_pedit,
-	&mlx5e_tc_act_pedit,
-	&mlx5e_tc_act_csum,
-	NULL, /* FLOW_ACTION_MARK, */
-	&mlx5e_tc_act_ptype,
-	NULL, /* FLOW_ACTION_PRIORITY, */
-	NULL, /* FLOW_ACTION_WAKE, */
-	NULL, /* FLOW_ACTION_QUEUE, */
-	&mlx5e_tc_act_sample,
-	&mlx5e_tc_act_police,
-	&mlx5e_tc_act_ct,
-	NULL, /* FLOW_ACTION_CT_METADATA, */
-	&mlx5e_tc_act_mpls_push,
-	&mlx5e_tc_act_mpls_pop,
-	NULL, /* FLOW_ACTION_MPLS_MANGLE, */
-	NULL, /* FLOW_ACTION_GATE, */
-	NULL, /* FLOW_ACTION_PPPOE_PUSH, */
-	NULL, /* FLOW_ACTION_JUMP, */
-	NULL, /* FLOW_ACTION_PIPE, */
-	&mlx5e_tc_act_vlan,
-	&mlx5e_tc_act_vlan,
+	[FLOW_ACTION_ACCEPT] = &mlx5e_tc_act_accept,
+	[FLOW_ACTION_DROP] = &mlx5e_tc_act_drop,
+	[FLOW_ACTION_TRAP] = &mlx5e_tc_act_trap,
+	[FLOW_ACTION_GOTO] = &mlx5e_tc_act_goto,
+	[FLOW_ACTION_REDIRECT] = &mlx5e_tc_act_mirred,
+	[FLOW_ACTION_MIRRED] = &mlx5e_tc_act_mirred,
+	[FLOW_ACTION_REDIRECT_INGRESS] = &mlx5e_tc_act_redirect_ingress,
+	[FLOW_ACTION_VLAN_PUSH] = &mlx5e_tc_act_vlan,
+	[FLOW_ACTION_VLAN_POP] = &mlx5e_tc_act_vlan,
+	[FLOW_ACTION_VLAN_MANGLE] = &mlx5e_tc_act_vlan_mangle,
+	[FLOW_ACTION_TUNNEL_ENCAP] = &mlx5e_tc_act_tun_encap,
+	[FLOW_ACTION_TUNNEL_DECAP] = &mlx5e_tc_act_tun_decap,
+	[FLOW_ACTION_MANGLE] = &mlx5e_tc_act_pedit,
+	[FLOW_ACTION_ADD] = &mlx5e_tc_act_pedit,
+	[FLOW_ACTION_CSUM] = &mlx5e_tc_act_csum,
+	[FLOW_ACTION_PTYPE] = &mlx5e_tc_act_ptype,
+	[FLOW_ACTION_SAMPLE] = &mlx5e_tc_act_sample,
+	[FLOW_ACTION_POLICE] = &mlx5e_tc_act_police,
+	[FLOW_ACTION_CT] = &mlx5e_tc_act_ct,
+	[FLOW_ACTION_MPLS_PUSH] = &mlx5e_tc_act_mpls_push,
+	[FLOW_ACTION_MPLS_POP] = &mlx5e_tc_act_mpls_pop,
+	[FLOW_ACTION_VLAN_PUSH_ETH] = &mlx5e_tc_act_vlan,
+	[FLOW_ACTION_VLAN_POP_ETH] = &mlx5e_tc_act_vlan,
 };
 
-/* Must be aligned with enum flow_action_id. */
 static struct mlx5e_tc_act *tc_acts_nic[NUM_FLOW_ACTIONS] = {
-	&mlx5e_tc_act_accept,
-	&mlx5e_tc_act_drop,
-	NULL, /* FLOW_ACTION_TRAP, */
-	&mlx5e_tc_act_goto,
-	&mlx5e_tc_act_mirred_nic,
-	NULL, /* FLOW_ACTION_MIRRED, */
-	NULL, /* FLOW_ACTION_REDIRECT_INGRESS, */
-	NULL, /* FLOW_ACTION_MIRRED_INGRESS, */
-	NULL, /* FLOW_ACTION_VLAN_PUSH, */
-	NULL, /* FLOW_ACTION_VLAN_POP, */
-	NULL, /* FLOW_ACTION_VLAN_MANGLE, */
-	NULL, /* FLOW_ACTION_TUNNEL_ENCAP, */
-	NULL, /* FLOW_ACTION_TUNNEL_DECAP, */
-	&mlx5e_tc_act_pedit,
-	&mlx5e_tc_act_pedit,
-	&mlx5e_tc_act_csum,
-	&mlx5e_tc_act_mark,
-	NULL, /* FLOW_ACTION_PTYPE, */
-	NULL, /* FLOW_ACTION_PRIORITY, */
-	NULL, /* FLOW_ACTION_WAKE, */
-	NULL, /* FLOW_ACTION_QUEUE, */
-	NULL, /* FLOW_ACTION_SAMPLE, */
-	NULL, /* FLOW_ACTION_POLICE, */
-	&mlx5e_tc_act_ct,
+	[FLOW_ACTION_ACCEPT] = &mlx5e_tc_act_accept,
+	[FLOW_ACTION_DROP] = &mlx5e_tc_act_drop,
+	[FLOW_ACTION_GOTO] = &mlx5e_tc_act_goto,
+	[FLOW_ACTION_REDIRECT] = &mlx5e_tc_act_mirred_nic,
+	[FLOW_ACTION_MANGLE] = &mlx5e_tc_act_pedit,
+	[FLOW_ACTION_ADD] = &mlx5e_tc_act_pedit,
+	[FLOW_ACTION_CSUM] = &mlx5e_tc_act_csum,
+	[FLOW_ACTION_MARK] = &mlx5e_tc_act_mark,
+	[FLOW_ACTION_CT] = &mlx5e_tc_act_ct,
 };
 
 /**
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
index 5aff97914367..ff73d25bc6eb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
@@ -224,15 +224,16 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv,
 	list_for_each_entry(flow, flow_list, tmp_list) {
 		if (!mlx5e_is_offloaded_flow(flow) || flow_flag_test(flow, SLOW))
 			continue;
-		spec = &flow->attr->parse_attr->spec;
-
-		/* update from encap rule to slow path rule */
-		rule = mlx5e_tc_offload_to_slow_path(esw, flow, spec);
 
 		attr = mlx5e_tc_get_encap_attr(flow);
 		esw_attr = attr->esw_attr;
 		/* mark the flow's encap dest as non-valid */
 		esw_attr->dests[flow->tmp_entry_index].flags &= ~MLX5_ESW_DEST_ENCAP_VALID;
+		esw_attr->dests[flow->tmp_entry_index].pkt_reformat = NULL;
+
+		/* update from encap rule to slow path rule */
+		spec = &flow->attr->parse_attr->spec;
+		rule = mlx5e_tc_offload_to_slow_path(esw, flow, spec);
 
 		if (IS_ERR(rule)) {
 			err = PTR_ERR(rule);
@@ -251,6 +252,7 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv,
 	/* we know that the encap is valid */
 	e->flags &= ~MLX5_ENCAP_ENTRY_VALID;
 	mlx5_packet_reformat_dealloc(priv->mdev, e->pkt_reformat);
+	e->pkt_reformat = NULL;
 }
 
 static void mlx5e_take_tmp_flow(struct mlx5e_tc_flow *flow,
@@ -762,8 +764,7 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv,
 		       struct net_device *mirred_dev,
 		       int out_index,
 		       struct netlink_ext_ack *extack,
-		       struct net_device **encap_dev,
-		       bool *encap_valid)
+		       struct net_device **encap_dev)
 {
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct mlx5e_tc_flow_parse_attr *parse_attr;
@@ -878,9 +879,8 @@ attach_flow:
 	if (e->flags & MLX5_ENCAP_ENTRY_VALID) {
 		attr->esw_attr->dests[out_index].pkt_reformat = e->pkt_reformat;
 		attr->esw_attr->dests[out_index].flags |= MLX5_ESW_DEST_ENCAP_VALID;
-		*encap_valid = true;
 	} else {
-		*encap_valid = false;
+		flow_flag_set(flow, SLOW);
 	}
 	mutex_unlock(&esw->offloads.encap_tbl_lock);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h
index d542b8476491..8ad273dde40e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h
@@ -17,8 +17,7 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv,
 		       struct net_device *mirred_dev,
 		       int out_index,
 		       struct netlink_ext_ack *extack,
-		       struct net_device **encap_dev,
-		       bool *encap_valid);
+		       struct net_device **encap_dev);
 
 int mlx5e_attach_decap(struct mlx5e_priv *priv,
 		       struct mlx5e_tc_flow *flow,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index cb164b62f543..853f312cd757 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -11,6 +11,27 @@
 
 #define INL_HDR_START_SZ (sizeof(((struct mlx5_wqe_eth_seg *)NULL)->inline_hdr.start))
 
+/* IPSEC inline data includes:
+ * 1. ESP trailer: up to 255 bytes of padding, 1 byte for pad length, 1 byte for
+ *    next header.
+ * 2. ESP authentication data: 16 bytes for ICV.
+ */
+#define MLX5E_MAX_TX_IPSEC_DS DIV_ROUND_UP(sizeof(struct mlx5_wqe_inline_seg) + \
+					   255 + 1 + 1 + 16, MLX5_SEND_WQE_DS)
+
+/* 366 should be big enough to cover all L2, L3 and L4 headers with possible
+ * encapsulations.
+ */
+#define MLX5E_MAX_TX_INLINE_DS DIV_ROUND_UP(366 - INL_HDR_START_SZ + VLAN_HLEN, \
+					    MLX5_SEND_WQE_DS)
+
+/* Sync the calculation with mlx5e_sq_calc_wqe_attr. */
+#define MLX5E_MAX_TX_WQEBBS DIV_ROUND_UP(MLX5E_TX_WQE_EMPTY_DS_COUNT + \
+					 MLX5E_MAX_TX_INLINE_DS + \
+					 MLX5E_MAX_TX_IPSEC_DS + \
+					 MAX_SKB_FRAGS + 1, \
+					 MLX5_SEND_WQEBB_NUM_DS)
+
 #define MLX5E_RX_ERR_CQE(cqe) (get_cqe_opcode(cqe) != MLX5_CQE_RESP_SEND)
 
 static inline
@@ -424,6 +445,8 @@ mlx5e_set_eseg_swp(struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg,
 
 static inline u16 mlx5e_stop_room_for_wqe(struct mlx5_core_dev *mdev, u16 wqe_size)
 {
+	WARN_ON_ONCE(PAGE_SIZE / MLX5_SEND_WQE_BB < mlx5e_get_max_sq_wqebbs(mdev));
+
 	/* A WQE must not cross the page boundary, hence two conditions:
 	 * 1. Its size must not exceed the page size.
 	 * 2. If the WQE size is X, and the space remaining in a page is less
@@ -436,7 +459,6 @@ static inline u16 mlx5e_stop_room_for_wqe(struct mlx5_core_dev *mdev, u16 wqe_si
 		  "wqe_size %u is greater than max SQ WQEBBs %u",
 		  wqe_size, mlx5e_get_max_sq_wqebbs(mdev));
 
-
 	return MLX5E_STOP_ROOM(wqe_size);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index 4685c652c97e..20507ef2f956 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -117,7 +117,7 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	xdpi.page.rq = rq;
 
 	dma_addr = page_pool_get_dma_addr(page) + (xdpf->data - (void *)xdpf);
-	dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd.len, DMA_TO_DEVICE);
+	dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd.len, DMA_BIDIRECTIONAL);
 
 	if (unlikely(xdp_frame_has_frags(xdpf))) {
 		sinfo = xdp_get_shared_info_from_frame(xdpf);
@@ -131,7 +131,7 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 				skb_frag_off(frag);
 			len = skb_frag_size(frag);
 			dma_sync_single_for_device(sq->pdev, addr, len,
-						   DMA_TO_DEVICE);
+						   DMA_BIDIRECTIONAL);
 		}
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c
index 285d32d2fd08..d7c020f72401 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c
@@ -365,7 +365,7 @@ void mlx5e_accel_fs_tcp_destroy(struct mlx5e_flow_steering *fs)
 	for (i = 0; i < ACCEL_FS_TCP_NUM_TYPES; i++)
 		accel_fs_tcp_destroy_table(fs, i);
 
-	kfree(accel_tcp);
+	kvfree(accel_tcp);
 	mlx5e_fs_set_accel_tcp(fs, NULL);
 }
 
@@ -397,7 +397,7 @@ int mlx5e_accel_fs_tcp_create(struct mlx5e_flow_steering *fs)
 err_destroy_tables:
 	while (--i >= 0)
 		accel_fs_tcp_destroy_table(fs, i);
-	kfree(accel_tcp);
+	kvfree(accel_tcp);
 	mlx5e_fs_set_accel_tcp(fs, NULL);
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
index 2ef36cb9555a..f900709639f6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
@@ -368,15 +368,15 @@ static int mlx5e_macsec_init_sa(struct macsec_context *ctx,
 	obj_attrs.aso_pdn = macsec->aso.pdn;
 	obj_attrs.epn_state = sa->epn_state;
 
-	if (is_tx) {
-		obj_attrs.ssci = cpu_to_be32((__force u32)ctx->sa.tx_sa->ssci);
-		key = &ctx->sa.tx_sa->key;
-	} else {
-		obj_attrs.ssci = cpu_to_be32((__force u32)ctx->sa.rx_sa->ssci);
-		key = &ctx->sa.rx_sa->key;
+	key = (is_tx) ? &ctx->sa.tx_sa->key : &ctx->sa.rx_sa->key;
+
+	if (sa->epn_state.epn_enabled) {
+		obj_attrs.ssci = (is_tx) ? cpu_to_be32((__force u32)ctx->sa.tx_sa->ssci) :
+					   cpu_to_be32((__force u32)ctx->sa.rx_sa->ssci);
+
+		memcpy(&obj_attrs.salt, &key->salt, sizeof(key->salt));
 	}
 
-	memcpy(&obj_attrs.salt, &key->salt, sizeof(key->salt));
 	obj_attrs.replay_window = ctx->secy->replay_window;
 	obj_attrs.replay_protect = ctx->secy->replay_protect;
 
@@ -427,15 +427,15 @@ mlx5e_macsec_get_rx_sc_from_sc_list(const struct list_head *list, sci_t sci)
 	return NULL;
 }
 
-static int mlx5e_macsec_update_rx_sa(struct mlx5e_macsec *macsec,
-				     struct mlx5e_macsec_sa *rx_sa,
-				     bool active)
+static int macsec_rx_sa_active_update(struct macsec_context *ctx,
+				      struct mlx5e_macsec_sa *rx_sa,
+				      bool active)
 {
-	struct mlx5_core_dev *mdev = macsec->mdev;
-	struct mlx5_macsec_obj_attrs attrs = {};
+	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_macsec *macsec = priv->macsec;
 	int err = 0;
 
-	if (rx_sa->active != active)
+	if (rx_sa->active == active)
 		return 0;
 
 	rx_sa->active = active;
@@ -444,13 +444,11 @@ static int mlx5e_macsec_update_rx_sa(struct mlx5e_macsec *macsec,
 		return 0;
 	}
 
-	attrs.sci = cpu_to_be64((__force u64)rx_sa->sci);
-	attrs.enc_key_id = rx_sa->enc_key_id;
-	err = mlx5e_macsec_create_object(mdev, &attrs, false, &rx_sa->macsec_obj_id);
+	err = mlx5e_macsec_init_sa(ctx, rx_sa, true, false);
 	if (err)
-		return err;
+		rx_sa->active = false;
 
-	return 0;
+	return err;
 }
 
 static bool mlx5e_macsec_secy_features_validate(struct macsec_context *ctx)
@@ -476,6 +474,11 @@ static bool mlx5e_macsec_secy_features_validate(struct macsec_context *ctx)
 		return false;
 	}
 
+	if (!ctx->secy->tx_sc.encrypt) {
+		netdev_err(netdev, "MACsec offload: encrypt off isn't supported\n");
+		return false;
+	}
+
 	return true;
 }
 
@@ -620,6 +623,7 @@ static int mlx5e_macsec_upd_txsa(struct macsec_context *ctx)
 	if (tx_sa->active == ctx_tx_sa->active)
 		goto out;
 
+	tx_sa->active = ctx_tx_sa->active;
 	if (tx_sa->assoc_num != tx_sc->encoding_sa)
 		goto out;
 
@@ -635,8 +639,6 @@ static int mlx5e_macsec_upd_txsa(struct macsec_context *ctx)
 
 		mlx5e_macsec_cleanup_sa(macsec, tx_sa, true);
 	}
-
-	tx_sa->active = ctx_tx_sa->active;
 out:
 	mutex_unlock(&macsec->lock);
 
@@ -736,9 +738,14 @@ static int mlx5e_macsec_add_rxsc(struct macsec_context *ctx)
 
 	sc_xarray_element->rx_sc = rx_sc;
 	err = xa_alloc(&macsec->sc_xarray, &sc_xarray_element->fs_id, sc_xarray_element,
-		       XA_LIMIT(1, USHRT_MAX), GFP_KERNEL);
-	if (err)
+		       XA_LIMIT(1, MLX5_MACEC_RX_FS_ID_MAX), GFP_KERNEL);
+	if (err) {
+		if (err == -EBUSY)
+			netdev_err(ctx->netdev,
+				   "MACsec offload: unable to create entry for RX SC (%d Rx SCs already allocated)\n",
+				   MLX5_MACEC_RX_FS_ID_MAX);
 		goto destroy_sc_xarray_elemenet;
+	}
 
 	rx_sc->md_dst = metadata_dst_alloc(0, METADATA_MACSEC, GFP_KERNEL);
 	if (!rx_sc->md_dst) {
@@ -798,16 +805,16 @@ static int mlx5e_macsec_upd_rxsc(struct macsec_context *ctx)
 		goto out;
 	}
 
-	rx_sc->active = ctx_rx_sc->active;
 	if (rx_sc->active == ctx_rx_sc->active)
 		goto out;
 
+	rx_sc->active = ctx_rx_sc->active;
 	for (i = 0; i < MACSEC_NUM_AN; ++i) {
 		rx_sa = rx_sc->rx_sa[i];
 		if (!rx_sa)
 			continue;
 
-		err = mlx5e_macsec_update_rx_sa(macsec, rx_sa, rx_sa->active && ctx_rx_sc->active);
+		err = macsec_rx_sa_active_update(ctx, rx_sa, rx_sa->active && ctx_rx_sc->active);
 		if (err)
 			goto out;
 	}
@@ -818,16 +825,43 @@ out:
 	return err;
 }
 
+static void macsec_del_rxsc_ctx(struct mlx5e_macsec *macsec, struct mlx5e_macsec_rx_sc *rx_sc)
+{
+	struct mlx5e_macsec_sa *rx_sa;
+	int i;
+
+	for (i = 0; i < MACSEC_NUM_AN; ++i) {
+		rx_sa = rx_sc->rx_sa[i];
+		if (!rx_sa)
+			continue;
+
+		mlx5e_macsec_cleanup_sa(macsec, rx_sa, false);
+		mlx5_destroy_encryption_key(macsec->mdev, rx_sa->enc_key_id);
+
+		kfree(rx_sa);
+		rx_sc->rx_sa[i] = NULL;
+	}
+
+	/* At this point the relevant MACsec offload Rx rule already removed at
+	 * mlx5e_macsec_cleanup_sa need to wait for datapath to finish current
+	 * Rx related data propagating using xa_erase which uses rcu to sync,
+	 * once fs_id is erased then this rx_sc is hidden from datapath.
+	 */
+	list_del_rcu(&rx_sc->rx_sc_list_element);
+	xa_erase(&macsec->sc_xarray, rx_sc->sc_xarray_element->fs_id);
+	metadata_dst_free(rx_sc->md_dst);
+	kfree(rx_sc->sc_xarray_element);
+	kfree_rcu(rx_sc);
+}
+
 static int mlx5e_macsec_del_rxsc(struct macsec_context *ctx)
 {
 	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5e_macsec_rx_sc *rx_sc;
-	struct mlx5e_macsec_sa *rx_sa;
 	struct mlx5e_macsec *macsec;
 	struct list_head *list;
 	int err = 0;
-	int i;
 
 	mutex_lock(&priv->macsec->lock);
 
@@ -849,31 +883,7 @@ static int mlx5e_macsec_del_rxsc(struct macsec_context *ctx)
 		goto out;
 	}
 
-	for (i = 0; i < MACSEC_NUM_AN; ++i) {
-		rx_sa = rx_sc->rx_sa[i];
-		if (!rx_sa)
-			continue;
-
-		mlx5e_macsec_cleanup_sa(macsec, rx_sa, false);
-		mlx5_destroy_encryption_key(macsec->mdev, rx_sa->enc_key_id);
-
-		kfree(rx_sa);
-		rx_sc->rx_sa[i] = NULL;
-	}
-
-/*
- * At this point the relevant MACsec offload Rx rule already removed at
- * mlx5e_macsec_cleanup_sa need to wait for datapath to finish current
- * Rx related data propagating using xa_erase which uses rcu to sync,
- * once fs_id is erased then this rx_sc is hidden from datapath.
- */
-	list_del_rcu(&rx_sc->rx_sc_list_element);
-	xa_erase(&macsec->sc_xarray, rx_sc->sc_xarray_element->fs_id);
-	metadata_dst_free(rx_sc->md_dst);
-	kfree(rx_sc->sc_xarray_element);
-
-	kfree_rcu(rx_sc);
-
+	macsec_del_rxsc_ctx(macsec, rx_sc);
 out:
 	mutex_unlock(&macsec->lock);
 
@@ -1015,7 +1025,7 @@ static int mlx5e_macsec_upd_rxsa(struct macsec_context *ctx)
 		goto out;
 	}
 
-	err = mlx5e_macsec_update_rx_sa(macsec, rx_sa, ctx_rx_sa->active);
+	err = macsec_rx_sa_active_update(ctx, rx_sa, ctx_rx_sa->active);
 out:
 	mutex_unlock(&macsec->lock);
 
@@ -1155,7 +1165,7 @@ static int macsec_upd_secy_hw_address(struct macsec_context *ctx,
 				continue;
 
 			if (rx_sa->active) {
-				err = mlx5e_macsec_init_sa(ctx, rx_sa, false, false);
+				err = mlx5e_macsec_init_sa(ctx, rx_sa, true, false);
 				if (err)
 					goto out;
 			}
@@ -1234,7 +1244,6 @@ static int mlx5e_macsec_del_secy(struct macsec_context *ctx)
 	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5e_macsec_rx_sc *rx_sc, *tmp;
-	struct mlx5e_macsec_sa *rx_sa;
 	struct mlx5e_macsec_sa *tx_sa;
 	struct mlx5e_macsec *macsec;
 	struct list_head *list;
@@ -1263,28 +1272,15 @@ static int mlx5e_macsec_del_secy(struct macsec_context *ctx)
 	}
 
 	list = &macsec_device->macsec_rx_sc_list_head;
-	list_for_each_entry_safe(rx_sc, tmp, list, rx_sc_list_element) {
-		for (i = 0; i < MACSEC_NUM_AN; ++i) {
-			rx_sa = rx_sc->rx_sa[i];
-			if (!rx_sa)
-				continue;
-
-			mlx5e_macsec_cleanup_sa(macsec, rx_sa, false);
-			mlx5_destroy_encryption_key(macsec->mdev, rx_sa->enc_key_id);
-			kfree(rx_sa);
-			rx_sc->rx_sa[i] = NULL;
-		}
-
-		list_del_rcu(&rx_sc->rx_sc_list_element);
-
-		kfree_rcu(rx_sc);
-	}
+	list_for_each_entry_safe(rx_sc, tmp, list, rx_sc_list_element)
+		macsec_del_rxsc_ctx(macsec, rx_sc);
 
 	kfree(macsec_device->dev_addr);
 	macsec_device->dev_addr = NULL;
 
 	list_del_rcu(&macsec_device->macsec_device_list_element);
 	--macsec->num_of_devices;
+	kfree(macsec_device);
 
 out:
 	mutex_unlock(&macsec->lock);
@@ -1536,6 +1532,8 @@ static void macsec_async_event(struct work_struct *work)
 
 	async_work = container_of(work, struct mlx5e_macsec_async_work, work);
 	macsec = async_work->macsec;
+	mutex_lock(&macsec->lock);
+
 	mdev = async_work->mdev;
 	obj_id = async_work->obj_id;
 	macsec_sa = get_macsec_tx_sa_from_obj_id(macsec, obj_id);
@@ -1557,6 +1555,7 @@ static void macsec_async_event(struct work_struct *work)
 
 out_async_work:
 	kfree(async_work);
+	mutex_unlock(&macsec->lock);
 }
 
 static int macsec_obj_change_event(struct notifier_block *nb, unsigned long event, void *data)
@@ -1745,7 +1744,7 @@ void mlx5e_macsec_offload_handle_rx_skb(struct net_device *netdev,
 	if (!macsec)
 		return;
 
-	fs_id = MLX5_MACSEC_METADATA_HANDLE(macsec_meta_data);
+	fs_id = MLX5_MACSEC_RX_METADAT_HANDLE(macsec_meta_data);
 
 	rcu_read_lock();
 	sc_xarray_element = xa_load(&macsec->sc_xarray, fs_id);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.h
index d580b4a91253..347380a2cd9c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.h
@@ -10,9 +10,11 @@
 #include <net/macsec.h>
 #include <net/dst_metadata.h>
 
-/* Bit31 - 30: MACsec marker, Bit3-0: MACsec id */
+/* Bit31 - 30: MACsec marker, Bit15-0: MACsec id */
+#define MLX5_MACEC_RX_FS_ID_MAX USHRT_MAX /* Must be power of two */
+#define MLX5_MACSEC_RX_FS_ID_MASK MLX5_MACEC_RX_FS_ID_MAX
 #define MLX5_MACSEC_METADATA_MARKER(metadata)  ((((metadata) >> 30) & 0x3)  == 0x1)
-#define MLX5_MACSEC_METADATA_HANDLE(metadata)  ((metadata) & GENMASK(3, 0))
+#define MLX5_MACSEC_RX_METADAT_HANDLE(metadata)  ((metadata) & MLX5_MACSEC_RX_FS_ID_MASK)
 
 struct mlx5e_priv;
 struct mlx5e_macsec;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
index 1ac0cf04e811..5b658a5588c6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
@@ -250,7 +250,7 @@ static int macsec_fs_tx_create(struct mlx5e_macsec_fs *macsec_fs)
 	struct mlx5_flow_handle *rule;
 	struct mlx5_flow_spec *spec;
 	u32 *flow_group_in;
-	int err = 0;
+	int err;
 
 	ns = mlx5_get_flow_namespace(macsec_fs->mdev, MLX5_FLOW_NAMESPACE_EGRESS_MACSEC);
 	if (!ns)
@@ -261,8 +261,10 @@ static int macsec_fs_tx_create(struct mlx5e_macsec_fs *macsec_fs)
 		return -ENOMEM;
 
 	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
-	if (!flow_group_in)
+	if (!flow_group_in) {
+		err = -ENOMEM;
 		goto out_spec;
+	}
 
 	tx_tables = &tx_fs->tables;
 	ft_crypto = &tx_tables->ft_crypto;
@@ -898,7 +900,7 @@ static int macsec_fs_rx_create(struct mlx5e_macsec_fs *macsec_fs)
 	struct mlx5_flow_handle *rule;
 	struct mlx5_flow_spec *spec;
 	u32 *flow_group_in;
-	int err = 0;
+	int err;
 
 	ns = mlx5_get_flow_namespace(macsec_fs->mdev, MLX5_FLOW_NAMESPACE_KERNEL_RX_MACSEC);
 	if (!ns)
@@ -909,8 +911,10 @@ static int macsec_fs_rx_create(struct mlx5e_macsec_fs *macsec_fs)
 		return -ENOMEM;
 
 	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
-	if (!flow_group_in)
+	if (!flow_group_in) {
+		err = -ENOMEM;
 		goto free_spec;
+	}
 
 	rx_tables = &rx_fs->tables;
 	ft_crypto = &rx_tables->ft_crypto;
@@ -1142,10 +1146,10 @@ macsec_fs_rx_add_rule(struct mlx5e_macsec_fs *macsec_fs,
 	ft_crypto = &rx_tables->ft_crypto;
 
 	/* Set bit[31 - 30] macsec marker - 0x01 */
-	/* Set bit[3-0] fs id */
+	/* Set bit[15-0] fs id */
 	MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET);
 	MLX5_SET(set_action_in, action, field, MLX5_ACTION_IN_FIELD_METADATA_REG_B);
-	MLX5_SET(set_action_in, action, data, fs_id | BIT(30));
+	MLX5_SET(set_action_in, action, data, MLX5_MACSEC_RX_METADAT_HANDLE(fs_id) | BIT(30));
 	MLX5_SET(set_action_in, action, offset, 0);
 	MLX5_SET(set_action_in, action, length, 32);
 
@@ -1205,6 +1209,7 @@ macsec_fs_rx_add_rule(struct mlx5e_macsec_fs *macsec_fs,
 		rx_rule->rule[1] = rule;
 	}
 
+	kvfree(spec);
 	return macsec_rule;
 
 err:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 24aa25da482b..1728e197558d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -35,7 +35,6 @@
 #include "en.h"
 #include "en/port.h"
 #include "en/params.h"
-#include "en/xsk/pool.h"
 #include "en/ptp.h"
 #include "lib/clock.h"
 #include "en/fs_ethtool.h"
@@ -412,15 +411,8 @@ void mlx5e_ethtool_get_channels(struct mlx5e_priv *priv,
 				struct ethtool_channels *ch)
 {
 	mutex_lock(&priv->state_lock);
-
 	ch->max_combined   = priv->max_nch;
 	ch->combined_count = priv->channels.params.num_channels;
-	if (priv->xsk.refcnt) {
-		/* The upper half are XSK queues. */
-		ch->max_combined *= 2;
-		ch->combined_count *= 2;
-	}
-
 	mutex_unlock(&priv->state_lock);
 }
 
@@ -454,16 +446,6 @@ int mlx5e_ethtool_set_channels(struct mlx5e_priv *priv,
 
 	mutex_lock(&priv->state_lock);
 
-	/* Don't allow changing the number of channels if there is an active
-	 * XSK, because the numeration of the XSK and regular RQs will change.
-	 */
-	if (priv->xsk.refcnt) {
-		err = -EINVAL;
-		netdev_err(priv->netdev, "%s: AF_XDP is active, cannot change the number of channels\n",
-			   __func__);
-		goto out;
-	}
-
 	/* Don't allow changing the number of channels if HTB offload is active,
 	 * because the numeration of the QoS SQs will change, while per-queue
 	 * qdiscs are attached.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 364f04309149..5e41dfdf79c8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -206,10 +206,11 @@ static void mlx5e_disable_blocking_events(struct mlx5e_priv *priv)
 static u16 mlx5e_mpwrq_umr_octowords(u32 entries, enum mlx5e_mpwrq_umr_mode umr_mode)
 {
 	u8 umr_entry_size = mlx5e_mpwrq_umr_entry_size(umr_mode);
+	u32 sz;
 
-	WARN_ON_ONCE(entries * umr_entry_size % MLX5_OCTWORD);
+	sz = ALIGN(entries * umr_entry_size, MLX5_UMR_MTT_ALIGNMENT);
 
-	return entries * umr_entry_size / MLX5_OCTWORD;
+	return sz / MLX5_OCTWORD;
 }
 
 static inline void mlx5e_build_umr_wqe(struct mlx5e_rq *rq,
@@ -5694,6 +5695,13 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv)
 		mlx5e_fs_set_state_destroy(priv->fs,
 					   !test_bit(MLX5E_STATE_DESTROYING, &priv->state));
 
+	/* Validate the max_wqe_size_sq capability. */
+	if (WARN_ON_ONCE(mlx5e_get_max_sq_wqebbs(priv->mdev) < MLX5E_MAX_TX_WQEBBS)) {
+		mlx5_core_warn(priv->mdev, "MLX5E: Max SQ WQEBBs firmware capability: %u, needed %lu\n",
+			       mlx5e_get_max_sq_wqebbs(priv->mdev), MLX5E_MAX_TX_WQEBBS);
+		return -EIO;
+	}
+
 	/* max number of channels may have changed */
 	max_nch = mlx5e_calc_max_nch(priv->mdev, priv->netdev, profile);
 	if (priv->channels.params.num_channels > max_nch) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 58084650151f..a61a43fc8d5c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -266,7 +266,7 @@ static inline bool mlx5e_rx_cache_get(struct mlx5e_rq *rq, union mlx5e_alloc_uni
 
 	addr = page_pool_get_dma_addr(au->page);
 	/* Non-XSK always uses PAGE_SIZE. */
-	dma_sync_single_for_device(rq->pdev, addr, PAGE_SIZE, DMA_FROM_DEVICE);
+	dma_sync_single_for_device(rq->pdev, addr, PAGE_SIZE, rq->buff.map_dir);
 	return true;
 }
 
@@ -282,8 +282,7 @@ static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, union mlx5e_alloc_u
 		return -ENOMEM;
 
 	/* Non-XSK always uses PAGE_SIZE. */
-	addr = dma_map_page_attrs(rq->pdev, au->page, 0, PAGE_SIZE,
-				  rq->buff.map_dir, DMA_ATTR_SKIP_CPU_SYNC);
+	addr = dma_map_page(rq->pdev, au->page, 0, PAGE_SIZE, rq->buff.map_dir);
 	if (unlikely(dma_mapping_error(rq->pdev, addr))) {
 		page_pool_recycle_direct(rq->page_pool, au->page);
 		au->page = NULL;
@@ -427,14 +426,15 @@ mlx5e_add_skb_frag(struct mlx5e_rq *rq, struct sk_buff *skb,
 {
 	dma_addr_t addr = page_pool_get_dma_addr(au->page);
 
-	dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len, DMA_FROM_DEVICE);
+	dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len,
+				rq->buff.map_dir);
 	page_ref_inc(au->page);
 	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
 			au->page, frag_offset, len, truesize);
 }
 
 static inline void
-mlx5e_copy_skb_header(struct device *pdev, struct sk_buff *skb,
+mlx5e_copy_skb_header(struct mlx5e_rq *rq, struct sk_buff *skb,
 		      struct page *page, dma_addr_t addr,
 		      int offset_from, int dma_offset, u32 headlen)
 {
@@ -442,7 +442,8 @@ mlx5e_copy_skb_header(struct device *pdev, struct sk_buff *skb,
 	/* Aligning len to sizeof(long) optimizes memcpy performance */
 	unsigned int len = ALIGN(headlen, sizeof(long));
 
-	dma_sync_single_for_cpu(pdev, addr + dma_offset, len, DMA_FROM_DEVICE);
+	dma_sync_single_for_cpu(rq->pdev, addr + dma_offset, len,
+				rq->buff.map_dir);
 	skb_copy_to_linear_data(skb, from, len);
 }
 
@@ -1538,7 +1539,7 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 
 	addr = page_pool_get_dma_addr(au->page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset,
-				      frag_size, DMA_FROM_DEVICE);
+				      frag_size, rq->buff.map_dir);
 	net_prefetch(data);
 
 	prog = rcu_dereference(rq->xdp_prog);
@@ -1587,7 +1588,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 
 	addr = page_pool_get_dma_addr(au->page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset,
-				      rq->buff.frame0_sz, DMA_FROM_DEVICE);
+				      rq->buff.frame0_sz, rq->buff.map_dir);
 	net_prefetchw(va); /* xdp_frame data area */
 	net_prefetch(va + rx_headroom);
 
@@ -1608,7 +1609,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 
 		addr = page_pool_get_dma_addr(au->page);
 		dma_sync_single_for_cpu(rq->pdev, addr + wi->offset,
-					frag_consumed_bytes, DMA_FROM_DEVICE);
+					frag_consumed_bytes, rq->buff.map_dir);
 
 		if (!xdp_buff_has_frags(&xdp)) {
 			/* Init on the first fragment to avoid cold cache access
@@ -1905,7 +1906,7 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 	mlx5e_fill_skb_data(skb, rq, au, byte_cnt, frag_offset);
 	/* copy header */
 	addr = page_pool_get_dma_addr(head_au->page);
-	mlx5e_copy_skb_header(rq->pdev, skb, head_au->page, addr,
+	mlx5e_copy_skb_header(rq, skb, head_au->page, addr,
 			      head_offset, head_offset, headlen);
 	/* skb linear part was allocated with headlen and aligned to long */
 	skb->tail += headlen;
@@ -1939,7 +1940,7 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 
 	addr = page_pool_get_dma_addr(au->page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, head_offset,
-				      frag_size, DMA_FROM_DEVICE);
+				      frag_size, rq->buff.map_dir);
 	net_prefetch(data);
 
 	prog = rcu_dereference(rq->xdp_prog);
@@ -1987,7 +1988,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 
 	if (likely(frag_size <= BIT(MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE))) {
 		/* build SKB around header */
-		dma_sync_single_range_for_cpu(rq->pdev, head->addr, 0, frag_size, DMA_FROM_DEVICE);
+		dma_sync_single_range_for_cpu(rq->pdev, head->addr, 0, frag_size, rq->buff.map_dir);
 		prefetchw(hdr);
 		prefetch(data);
 		skb = mlx5e_build_linear_skb(rq, hdr, frag_size, rx_headroom, head_size, 0);
@@ -2009,7 +2010,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		}
 
 		prefetchw(skb->data);
-		mlx5e_copy_skb_header(rq->pdev, skb, head->page, head->addr,
+		mlx5e_copy_skb_header(rq, skb, head->page, head->addr,
 				      head_offset + rx_headroom,
 				      rx_headroom, head_size);
 		/* skb linear part was allocated with headlen and aligned to long */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index dd6fea9e9a5b..bd9936af4582 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1634,7 +1634,6 @@ set_encap_dests(struct mlx5e_priv *priv,
 		struct mlx5e_tc_flow *flow,
 		struct mlx5_flow_attr *attr,
 		struct netlink_ext_ack *extack,
-		bool *encap_valid,
 		bool *vf_tun)
 {
 	struct mlx5e_tc_flow_parse_attr *parse_attr;
@@ -1651,7 +1650,6 @@ set_encap_dests(struct mlx5e_priv *priv,
 	parse_attr = attr->parse_attr;
 	esw_attr = attr->esw_attr;
 	*vf_tun = false;
-	*encap_valid = true;
 
 	for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) {
 		struct net_device *out_dev;
@@ -1668,7 +1666,7 @@ set_encap_dests(struct mlx5e_priv *priv,
 			goto out;
 		}
 		err = mlx5e_attach_encap(priv, flow, attr, out_dev, out_index,
-					 extack, &encap_dev, encap_valid);
+					 extack, &encap_dev);
 		dev_put(out_dev);
 		if (err)
 			goto out;
@@ -1732,8 +1730,8 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 	struct mlx5e_tc_flow_parse_attr *parse_attr;
 	struct mlx5_flow_attr *attr = flow->attr;
 	struct mlx5_esw_flow_attr *esw_attr;
-	bool vf_tun, encap_valid;
 	u32 max_prio, max_chain;
+	bool vf_tun;
 	int err = 0;
 
 	parse_attr = attr->parse_attr;
@@ -1823,7 +1821,7 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 		esw_attr->int_port = int_port;
 	}
 
-	err = set_encap_dests(priv, flow, attr, extack, &encap_valid, &vf_tun);
+	err = set_encap_dests(priv, flow, attr, extack, &vf_tun);
 	if (err)
 		goto err_out;
 
@@ -1853,7 +1851,7 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 	 * (1) there's no error
 	 * (2) there's an encap action and we don't have valid neigh
 	 */
-	if (!encap_valid || flow_flag_test(flow, SLOW))
+	if (flow_flag_test(flow, SLOW))
 		flow->rule[0] = mlx5e_tc_offload_to_slow_path(esw, flow, &parse_attr->spec);
 	else
 		flow->rule[0] = mlx5e_tc_offload_fdb_rules(esw, flow, &parse_attr->spec, attr);
@@ -3633,10 +3631,14 @@ mlx5e_clone_flow_attr_for_post_act(struct mlx5_flow_attr *attr,
 	attr2->action = 0;
 	attr2->flags = 0;
 	attr2->parse_attr = parse_attr;
-	attr2->esw_attr->out_count = 0;
-	attr2->esw_attr->split_count = 0;
 	attr2->dest_chain = 0;
 	attr2->dest_ft = NULL;
+
+	if (ns_type == MLX5_FLOW_NAMESPACE_FDB) {
+		attr2->esw_attr->out_count = 0;
+		attr2->esw_attr->split_count = 0;
+	}
+
 	return attr2;
 }
 
@@ -3755,7 +3757,7 @@ alloc_flow_post_acts(struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack)
 	struct mlx5e_post_act *post_act = get_post_action(flow->priv);
 	struct mlx5_flow_attr *attr, *next_attr = NULL;
 	struct mlx5e_post_act_handle *handle;
-	bool vf_tun, encap_valid = true;
+	bool vf_tun;
 	int err;
 
 	/* This is going in reverse order as needed.
@@ -3777,13 +3779,10 @@ alloc_flow_post_acts(struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack)
 		if (list_is_last(&attr->list, &flow->attrs))
 			break;
 
-		err = set_encap_dests(flow->priv, flow, attr, extack, &encap_valid, &vf_tun);
+		err = set_encap_dests(flow->priv, flow, attr, extack, &vf_tun);
 		if (err)
 			goto out_free;
 
-		if (!encap_valid)
-			flow_flag_set(flow, SLOW);
-
 		err = actions_prepare_mod_hdr_actions(flow->priv, flow, attr, extack);
 		if (err)
 			goto out_free;
@@ -4758,12 +4757,6 @@ int mlx5e_policer_validate(const struct flow_action *action,
 		return -EOPNOTSUPP;
 	}
 
-	if (act->police.rate_pkt_ps) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "QoS offload not support packets per second");
-		return -EOPNOTSUPP;
-	}
-
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 6adca01fbdc9..f7897ddb29c5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -305,6 +305,8 @@ static void mlx5e_sq_calc_wqe_attr(struct sk_buff *skb, const struct mlx5e_tx_at
 	u16 ds_cnt_inl = 0;
 	u16 ds_cnt_ids = 0;
 
+	/* Sync the calculation with MLX5E_MAX_TX_WQEBBS. */
+
 	if (attr->insz)
 		ds_cnt_ids = DIV_ROUND_UP(sizeof(struct mlx5_wqe_inline_seg) + attr->insz,
 					  MLX5_SEND_WQE_DS);
@@ -317,6 +319,9 @@ static void mlx5e_sq_calc_wqe_attr(struct sk_buff *skb, const struct mlx5e_tx_at
 			inl += VLAN_HLEN;
 
 		ds_cnt_inl = DIV_ROUND_UP(inl, MLX5_SEND_WQE_DS);
+		if (WARN_ON_ONCE(ds_cnt_inl > MLX5E_MAX_TX_INLINE_DS))
+			netdev_warn(skb->dev, "ds_cnt_inl = %u > max %u\n", ds_cnt_inl,
+				    (u16)MLX5E_MAX_TX_INLINE_DS);
 		ds_cnt += ds_cnt_inl;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index c59107fa9e6d..374e3fbdc2cf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1362,6 +1362,9 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf)
 
 		devl_rate_nodes_destroy(devlink);
 	}
+	/* Destroy legacy fdb when disabling sriov in legacy mode. */
+	if (esw->mode == MLX5_ESWITCH_LEGACY)
+		mlx5_eswitch_disable_locked(esw);
 
 	esw->esw_funcs.num_vfs = 0;
 
@@ -1387,12 +1390,14 @@ void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw)
 		 esw->mode == MLX5_ESWITCH_LEGACY ? "LEGACY" : "OFFLOADS",
 		 esw->esw_funcs.num_vfs, esw->enabled_vports);
 
-	esw->fdb_table.flags &= ~MLX5_ESW_FDB_CREATED;
-	if (esw->mode == MLX5_ESWITCH_OFFLOADS)
-		esw_offloads_disable(esw);
-	else if (esw->mode == MLX5_ESWITCH_LEGACY)
-		esw_legacy_disable(esw);
-	mlx5_esw_acls_ns_cleanup(esw);
+	if (esw->fdb_table.flags & MLX5_ESW_FDB_CREATED) {
+		esw->fdb_table.flags &= ~MLX5_ESW_FDB_CREATED;
+		if (esw->mode == MLX5_ESWITCH_OFFLOADS)
+			esw_offloads_disable(esw);
+		else if (esw->mode == MLX5_ESWITCH_LEGACY)
+			esw_legacy_disable(esw);
+		mlx5_esw_acls_ns_cleanup(esw);
+	}
 
 	if (esw->mode == MLX5_ESWITCH_OFFLOADS)
 		devl_rate_nodes_destroy(devlink);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index f68dc2d0dbe6..3029bc1c0dd0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -736,6 +736,14 @@ void mlx5_eswitch_offloads_destroy_single_fdb(struct mlx5_eswitch *master_esw,
 					      struct mlx5_eswitch *slave_esw);
 int mlx5_eswitch_reload_reps(struct mlx5_eswitch *esw);
 
+static inline int mlx5_eswitch_num_vfs(struct mlx5_eswitch *esw)
+{
+	if (mlx5_esw_allowed(esw))
+		return esw->esw_funcs.num_vfs;
+
+	return 0;
+}
+
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 4e50df3139c6..8c6c9bcb3dc3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -433,7 +433,7 @@ esw_setup_vport_dest(struct mlx5_flow_destination *dest, struct mlx5_flow_act *f
 		    mlx5_lag_mpesw_is_activated(esw->dev))
 			dest[dest_idx].type = MLX5_FLOW_DESTINATION_TYPE_UPLINK;
 	}
-	if (esw_attr->dests[attr_idx].flags & MLX5_ESW_DEST_ENCAP) {
+	if (esw_attr->dests[attr_idx].flags & MLX5_ESW_DEST_ENCAP_VALID) {
 		if (pkt_reformat) {
 			flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
 			flow_act->pkt_reformat = esw_attr->dests[attr_idx].pkt_reformat;
@@ -2310,7 +2310,7 @@ out_free:
 static int esw_offloads_start(struct mlx5_eswitch *esw,
 			      struct netlink_ext_ack *extack)
 {
-	int err, err1;
+	int err;
 
 	esw->mode = MLX5_ESWITCH_OFFLOADS;
 	err = mlx5_eswitch_enable_locked(esw, esw->dev->priv.sriov.num_vfs);
@@ -2318,11 +2318,6 @@ static int esw_offloads_start(struct mlx5_eswitch *esw,
 		NL_SET_ERR_MSG_MOD(extack,
 				   "Failed setting eswitch to offloads");
 		esw->mode = MLX5_ESWITCH_LEGACY;
-		err1 = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS);
-		if (err1) {
-			NL_SET_ERR_MSG_MOD(extack,
-					   "Failed setting eswitch back to legacy");
-		}
 		mlx5_rescan_drivers(esw->dev);
 	}
 	if (esw->offloads.inline_mode == MLX5_INLINE_MODE_NONE) {
@@ -3389,19 +3384,19 @@ err_metadata:
 static int esw_offloads_stop(struct mlx5_eswitch *esw,
 			     struct netlink_ext_ack *extack)
 {
-	int err, err1;
+	int err;
 
 	esw->mode = MLX5_ESWITCH_LEGACY;
+
+	/* If changing from switchdev to legacy mode without sriov enabled,
+	 * no need to create legacy fdb.
+	 */
+	if (!mlx5_sriov_is_enabled(esw->dev))
+		return 0;
+
 	err = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS);
-	if (err) {
+	if (err)
 		NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch to legacy");
-		esw->mode = MLX5_ESWITCH_OFFLOADS;
-		err1 = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS);
-		if (err1) {
-			NL_SET_ERR_MSG_MOD(extack,
-					   "Failed setting eswitch back to offloads");
-		}
-	}
 
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
index ee568bf34ae2..edd910258314 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
@@ -30,9 +30,9 @@ mlx5_eswitch_termtbl_hash(struct mlx5_flow_act *flow_act,
 		     sizeof(dest->vport.num), hash);
 	hash = jhash((const void *)&dest->vport.vhca_id,
 		     sizeof(dest->vport.num), hash);
-	if (dest->vport.pkt_reformat)
-		hash = jhash(dest->vport.pkt_reformat,
-			     sizeof(*dest->vport.pkt_reformat),
+	if (flow_act->pkt_reformat)
+		hash = jhash(flow_act->pkt_reformat,
+			     sizeof(*flow_act->pkt_reformat),
 			     hash);
 	return hash;
 }
@@ -53,9 +53,11 @@ mlx5_eswitch_termtbl_cmp(struct mlx5_flow_act *flow_act1,
 	if (ret)
 		return ret;
 
-	return dest1->vport.pkt_reformat && dest2->vport.pkt_reformat ?
-	       memcmp(dest1->vport.pkt_reformat, dest2->vport.pkt_reformat,
-		      sizeof(*dest1->vport.pkt_reformat)) : 0;
+	if (flow_act1->pkt_reformat && flow_act2->pkt_reformat)
+		return memcmp(flow_act1->pkt_reformat, flow_act2->pkt_reformat,
+			      sizeof(*flow_act1->pkt_reformat));
+
+	return !(flow_act1->pkt_reformat == flow_act2->pkt_reformat);
 }
 
 static int
@@ -310,6 +312,8 @@ revert_changes:
 	for (curr_dest = 0; curr_dest < num_vport_dests; curr_dest++) {
 		struct mlx5_termtbl_handle *tt = attr->dests[curr_dest].termtbl;
 
+		attr->dests[curr_dest].termtbl = NULL;
+
 		/* search for the destination associated with the
 		 * current term table
 		 */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
index 07c583996c29..1e46f9afa40e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
@@ -9,7 +9,8 @@ enum {
 	MLX5_FW_RESET_FLAGS_RESET_REQUESTED,
 	MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST,
 	MLX5_FW_RESET_FLAGS_PENDING_COMP,
-	MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS
+	MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS,
+	MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED
 };
 
 struct mlx5_fw_reset {
@@ -152,7 +153,8 @@ static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev)
 		mlx5_unload_one(dev);
 		if (mlx5_health_wait_pci_up(dev))
 			mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n");
-		mlx5_load_one(dev, false);
+		else
+			mlx5_load_one(dev, false);
 		devlink_remote_reload_actions_performed(priv_to_devlink(dev), 0,
 							BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) |
 							BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE));
@@ -405,7 +407,7 @@ static void mlx5_sync_reset_now_event(struct work_struct *work)
 	err = mlx5_pci_link_toggle(dev);
 	if (err) {
 		mlx5_core_warn(dev, "mlx5_pci_link_toggle failed, no reset done, err %d\n", err);
-		goto done;
+		set_bit(MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED, &fw_reset->reset_flags);
 	}
 
 	mlx5_enter_error_state(dev, true);
@@ -481,6 +483,10 @@ int mlx5_fw_reset_wait_reset_done(struct mlx5_core_dev *dev)
 		goto out;
 	}
 	err = fw_reset->ret;
+	if (test_and_clear_bit(MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED, &fw_reset->reset_flags)) {
+		mlx5_unload_one_devl_locked(dev);
+		mlx5_load_one_devl_locked(dev, false);
+	}
 out:
 	clear_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags);
 	return err;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
index a9f4ede4a9bf..32c3e0a649a7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
@@ -228,9 +228,8 @@ static void mlx5_ldev_free(struct kref *ref)
 	if (ldev->nb.notifier_call)
 		unregister_netdevice_notifier_net(&init_net, &ldev->nb);
 	mlx5_lag_mp_cleanup(ldev);
-	mlx5_lag_mpesw_cleanup(ldev);
-	cancel_work_sync(&ldev->mpesw_work);
 	destroy_workqueue(ldev->wq);
+	mlx5_lag_mpesw_cleanup(ldev);
 	mutex_destroy(&ldev->lock);
 	kfree(ldev);
 }
@@ -701,10 +700,13 @@ static bool mlx5_lag_check_prereq(struct mlx5_lag *ldev)
 			return false;
 
 #ifdef CONFIG_MLX5_ESWITCH
-	dev = ldev->pf[MLX5_LAG_P1].dev;
-	if ((mlx5_sriov_is_enabled(dev)) && !is_mdev_switchdev_mode(dev))
-		return false;
+	for (i = 0; i < ldev->ports; i++) {
+		dev = ldev->pf[i].dev;
+		if (mlx5_eswitch_num_vfs(dev->priv.eswitch) && !is_mdev_switchdev_mode(dev))
+			return false;
+	}
 
+	dev = ldev->pf[MLX5_LAG_P1].dev;
 	mode = mlx5_eswitch_mode(dev);
 	for (i = 0; i < ldev->ports; i++)
 		if (mlx5_eswitch_mode(ldev->pf[i].dev) != mode)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
index ce2ce8ccbd70..f30ac2de639f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
@@ -50,6 +50,19 @@ struct lag_tracker {
 	enum netdev_lag_hash hash_type;
 };
 
+enum mpesw_op {
+	MLX5_MPESW_OP_ENABLE,
+	MLX5_MPESW_OP_DISABLE,
+};
+
+struct mlx5_mpesw_work_st {
+	struct work_struct work;
+	struct mlx5_lag    *lag;
+	enum mpesw_op	   op;
+	struct completion  comp;
+	int result;
+};
+
 /* LAG data of a ConnectX card.
  * It serves both its phys functions.
  */
@@ -66,7 +79,6 @@ struct mlx5_lag {
 	struct lag_tracker        tracker;
 	struct workqueue_struct   *wq;
 	struct delayed_work       bond_work;
-	struct work_struct	  mpesw_work;
 	struct notifier_block     nb;
 	struct lag_mp             lag_mp;
 	struct mlx5_lag_port_sel  port_sel;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
index f643202b29c6..c17e8f1ec914 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
@@ -7,63 +7,95 @@
 #include "eswitch.h"
 #include "lib/mlx5.h"
 
-void mlx5_mpesw_work(struct work_struct *work)
+static int add_mpesw_rule(struct mlx5_lag *ldev)
 {
-	struct mlx5_lag *ldev = container_of(work, struct mlx5_lag, mpesw_work);
+	struct mlx5_core_dev *dev = ldev->pf[MLX5_LAG_P1].dev;
+	int err;
 
-	mutex_lock(&ldev->lock);
-	mlx5_disable_lag(ldev);
-	mutex_unlock(&ldev->lock);
-}
+	if (atomic_add_return(1, &ldev->lag_mpesw.mpesw_rule_count) != 1)
+		return 0;
 
-static void mlx5_lag_disable_mpesw(struct mlx5_core_dev *dev)
-{
-	struct mlx5_lag *ldev = dev->priv.lag;
+	if (ldev->mode != MLX5_LAG_MODE_NONE) {
+		err = -EINVAL;
+		goto out_err;
+	}
 
-	if (!queue_work(ldev->wq, &ldev->mpesw_work))
-		mlx5_core_warn(dev, "failed to queue work\n");
+	err = mlx5_activate_lag(ldev, NULL, MLX5_LAG_MODE_MPESW, false);
+	if (err) {
+		mlx5_core_warn(dev, "Failed to create LAG in MPESW mode (%d)\n", err);
+		goto out_err;
+	}
+
+	return 0;
+
+out_err:
+	atomic_dec(&ldev->lag_mpesw.mpesw_rule_count);
+	return err;
 }
 
-void mlx5_lag_del_mpesw_rule(struct mlx5_core_dev *dev)
+static void del_mpesw_rule(struct mlx5_lag *ldev)
 {
-	struct mlx5_lag *ldev = dev->priv.lag;
+	if (!atomic_dec_return(&ldev->lag_mpesw.mpesw_rule_count) &&
+	    ldev->mode == MLX5_LAG_MODE_MPESW)
+		mlx5_disable_lag(ldev);
+}
 
-	if (!ldev)
-		return;
+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_lag *ldev = mpesww->lag;
 
 	mutex_lock(&ldev->lock);
-	if (!atomic_dec_return(&ldev->lag_mpesw.mpesw_rule_count) &&
-	    ldev->mode == MLX5_LAG_MODE_MPESW)
-		mlx5_lag_disable_mpesw(dev);
+	if (mpesww->op == MLX5_MPESW_OP_ENABLE)
+		mpesww->result = add_mpesw_rule(ldev);
+	else if (mpesww->op == MLX5_MPESW_OP_DISABLE)
+		del_mpesw_rule(ldev);
 	mutex_unlock(&ldev->lock);
+
+	complete(&mpesww->comp);
 }
 
-int mlx5_lag_add_mpesw_rule(struct mlx5_core_dev *dev)
+static int mlx5_lag_mpesw_queue_work(struct mlx5_core_dev *dev,
+				     enum mpesw_op op)
 {
 	struct mlx5_lag *ldev = dev->priv.lag;
+	struct mlx5_mpesw_work_st *work;
 	int err = 0;
 
 	if (!ldev)
 		return 0;
 
-	mutex_lock(&ldev->lock);
-	if (atomic_add_return(1, &ldev->lag_mpesw.mpesw_rule_count) != 1)
-		goto out;
+	work = kzalloc(sizeof(*work), GFP_KERNEL);
+	if (!work)
+		return -ENOMEM;
 
-	if (ldev->mode != MLX5_LAG_MODE_NONE) {
+	INIT_WORK(&work->work, mlx5_mpesw_work);
+	init_completion(&work->comp);
+	work->op = op;
+	work->lag = ldev;
+
+	if (!queue_work(ldev->wq, &work->work)) {
+		mlx5_core_warn(dev, "failed to queue mpesw work\n");
 		err = -EINVAL;
 		goto out;
 	}
-
-	err = mlx5_activate_lag(ldev, NULL, MLX5_LAG_MODE_MPESW, false);
-	if (err)
-		mlx5_core_warn(dev, "Failed to create LAG in MPESW mode (%d)\n", err);
-
+	wait_for_completion(&work->comp);
+	err = work->result;
 out:
-	mutex_unlock(&ldev->lock);
+	kfree(work);
 	return err;
 }
 
+void mlx5_lag_del_mpesw_rule(struct mlx5_core_dev *dev)
+{
+	mlx5_lag_mpesw_queue_work(dev, MLX5_MPESW_OP_DISABLE);
+}
+
+int mlx5_lag_add_mpesw_rule(struct mlx5_core_dev *dev)
+{
+	return mlx5_lag_mpesw_queue_work(dev, MLX5_MPESW_OP_ENABLE);
+}
+
 int mlx5_lag_do_mirred(struct mlx5_core_dev *mdev, struct net_device *out_dev)
 {
 	struct mlx5_lag *ldev = mdev->priv.lag;
@@ -71,12 +103,9 @@ int mlx5_lag_do_mirred(struct mlx5_core_dev *mdev, struct net_device *out_dev)
 	if (!netif_is_bond_master(out_dev) || !ldev)
 		return 0;
 
-	mutex_lock(&ldev->lock);
-	if (ldev->mode == MLX5_LAG_MODE_MPESW) {
-		mutex_unlock(&ldev->lock);
+	if (ldev->mode == MLX5_LAG_MODE_MPESW)
 		return -EOPNOTSUPP;
-	}
-	mutex_unlock(&ldev->lock);
+
 	return 0;
 }
 
@@ -90,11 +119,10 @@ bool mlx5_lag_mpesw_is_activated(struct mlx5_core_dev *dev)
 
 void mlx5_lag_mpesw_init(struct mlx5_lag *ldev)
 {
-	INIT_WORK(&ldev->mpesw_work, mlx5_mpesw_work);
 	atomic_set(&ldev->lag_mpesw.mpesw_rule_count, 0);
 }
 
 void mlx5_lag_mpesw_cleanup(struct mlx5_lag *ldev)
 {
-	cancel_delayed_work_sync(&ldev->bond_work);
+	WARN_ON(atomic_read(&ldev->lag_mpesw.mpesw_rule_count));
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.h b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.h
index be4abcb8fcd5..88e8daffcf92 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.h
@@ -12,7 +12,6 @@ struct lag_mpesw {
 	atomic_t mpesw_rule_count;
 };
 
-void mlx5_mpesw_work(struct work_struct *work);
 int mlx5_lag_do_mirred(struct mlx5_core_dev *mdev, struct net_device *out_dev);
 bool mlx5_lag_mpesw_is_activated(struct mlx5_core_dev *dev);
 #if IS_ENABLED(CONFIG_MLX5_ESWITCH)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 283c4cc28944..e58775a7d955 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1798,7 +1798,8 @@ static pci_ers_result_t mlx5_pci_err_detected(struct pci_dev *pdev,
 	res = state == pci_channel_io_perm_failure ?
 		PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_NEED_RESET;
 
-	mlx5_pci_trace(dev, "Exit, result = %d, %s\n",  res, result2str(res));
+	mlx5_core_info(dev, "%s Device state = %d pci_status: %d. Exit, result = %d, %s\n",
+		       __func__, dev->state, dev->pci_status, res, result2str(res));
 	return res;
 }
 
@@ -1837,7 +1838,8 @@ static pci_ers_result_t mlx5_pci_slot_reset(struct pci_dev *pdev)
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
 	int err;
 
-	mlx5_pci_trace(dev, "Enter\n");
+	mlx5_core_info(dev, "%s Device state = %d pci_status: %d. Enter\n",
+		       __func__, dev->state, dev->pci_status);
 
 	err = mlx5_pci_enable_device(dev);
 	if (err) {
@@ -1859,7 +1861,8 @@ static pci_ers_result_t mlx5_pci_slot_reset(struct pci_dev *pdev)
 
 	res = PCI_ERS_RESULT_RECOVERED;
 out:
-	mlx5_pci_trace(dev, "Exit, err = %d, result = %d, %s\n", err, res, result2str(res));
+	mlx5_core_info(dev, "%s Device state = %d pci_status: %d. Exit, err = %d, result = %d, %s\n",
+		       __func__, dev->state, dev->pci_status, err, res, result2str(res));
 	return res;
 }
 
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 7da012ff0d41..8e2abbab05f0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
@@ -18,6 +18,10 @@ struct mlx5_sf_dev_table {
 	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;
 };
 
@@ -168,6 +172,7 @@ 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:
@@ -191,6 +196,7 @@ mlx5_sf_dev_state_change_handler(struct notifier_block *nb, unsigned long event_
 	default:
 		break;
 	}
+	mutex_unlock(&table->table_lock);
 	return 0;
 }
 
@@ -215,6 +221,78 @@ 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)
+{
+	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_core_dev *dev = table->dev;
+	u16 max_functions;
+	u16 function_id;
+	u16 sw_func_id;
+	int err = 0;
+	u8 state;
+	int i;
+
+	max_functions = mlx5_sf_max_functions(dev);
+	function_id = MLX5_CAP_GEN(dev, sf_base_id);
+	for (i = 0; i < max_functions; i++, function_id++) {
+		if (table->stop_active_wq)
+			return;
+		err = mlx5_cmd_query_vhca_state(dev, function_id, out, sizeof(out));
+		if (err)
+			/* A failure of specific vhca doesn't mean others will
+			 * fail as well.
+			 */
+			continue;
+		state = MLX5_GET(query_vhca_state_out, out, vhca_state_context.vhca_state);
+		if (state != MLX5_VHCA_STATE_ACTIVE)
+			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);
+	}
+}
+
+/* In case SFs are generated externally, probe active SFs */
+static int mlx5_sf_dev_queue_active_work(struct mlx5_sf_dev_table *table)
+{
+	if (MLX5_CAP_GEN(table->dev, eswitch_manager))
+		return 0; /* the table is local */
+
+	/* Use a workqueue to probe active SFs, which are in large
+	 * quantity and may take up to minutes to probe.
+	 */
+	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);
+	queue_work(table->active_wq, &table->work);
+	return 0;
+}
+
+static void mlx5_sf_dev_destroy_active_work(struct mlx5_sf_dev_table *table)
+{
+	if (table->active_wq) {
+		table->stop_active_wq = true;
+		destroy_workqueue(table->active_wq);
+	}
+}
+
 void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 {
 	struct mlx5_sf_dev_table *table;
@@ -240,11 +318,17 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 	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);
+	if (err)
+		goto add_active_err;
+
 	err = mlx5_sf_dev_vhca_arm_all(table);
 	if (err)
 		goto arm_err;
@@ -252,6 +336,8 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 	return;
 
 arm_err:
+	mlx5_sf_dev_destroy_active_work(table);
+add_active_err:
 	mlx5_vhca_event_notifier_unregister(dev, &table->nb);
 vhca_err:
 	table->max_sfs = 0;
@@ -279,7 +365,9 @@ void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev)
 	if (!table)
 		return;
 
+	mlx5_sf_dev_destroy_active_work(table);
 	mlx5_vhca_event_notifier_unregister(dev, &table->nb);
+	mutex_destroy(&table->table_lock);
 
 	/* 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/steering/dr_table.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
index 31d443dd8386..f68461b13391 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
@@ -46,7 +46,7 @@ static int dr_table_set_miss_action_nic(struct mlx5dr_domain *dmn,
 int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl,
 				 struct mlx5dr_action *action)
 {
-	int ret;
+	int ret = -EOPNOTSUPP;
 
 	if (action && action->action_type != DR_ACTION_TYP_FT)
 		return -EOPNOTSUPP;
@@ -67,6 +67,9 @@ int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl,
 			goto out;
 	}
 
+	if (ret)
+		goto out;
+
 	/* Release old action */
 	if (tbl->miss_action)
 		refcount_dec(&tbl->miss_action->refcount);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index 4efccd942fb8..1290b2d3eae6 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -3470,6 +3470,8 @@ mlxsw_sp_switchdev_vxlan_fdb_del(struct mlxsw_sp *mlxsw_sp,
 	u16 vid;
 
 	vxlan_fdb_info = &switchdev_work->vxlan_fdb_info;
+	if (!vxlan_fdb_info->offloaded)
+		return;
 
 	bridge_device = mlxsw_sp_bridge_device_find(mlxsw_sp->bridge, br_dev);
 	if (!bridge_device)