Merge tag 'mlx5-fixes-2023-12-04' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
mlx5 fixes 2023-12-04

This series provides bug fixes to mlx5 driver.

V1->V2:
  - Drop commit #9 ("net/mlx5e: Forbid devlink reload if IPSec rules are
    offloaded"), we are working on a better fix

Please pull and let me know if there is any problem.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

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

@@ -826,6 +826,7 @@ enum {
 	MLX5E_STATE_DESTROYING,
 	MLX5E_STATE_XDP_TX_ENABLED,
 	MLX5E_STATE_XDP_ACTIVE,
+	MLX5E_STATE_CHANNELS_ACTIVE,
 };
 
 struct mlx5e_modify_sq_param {

drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c

@@ -83,6 +83,9 @@ mlx5e_tc_post_act_offload(struct mlx5e_post_act *post_act,
 	struct mlx5_flow_spec *spec;
 	int err;
 
+	if (IS_ERR(post_act))
+		return PTR_ERR(post_act);
+
 	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
 	if (!spec)
 		return -ENOMEM;
@@ -111,6 +114,9 @@ mlx5e_tc_post_act_add(struct mlx5e_post_act *post_act, struct mlx5_flow_attr *po
 	struct mlx5e_post_act_handle *handle;
 	int err;
 
+	if (IS_ERR(post_act))
+		return ERR_CAST(post_act);
+
 	handle = kzalloc(sizeof(*handle), GFP_KERNEL);
 	if (!handle)
 		return ERR_PTR(-ENOMEM);

drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c

@@ -121,7 +121,14 @@ static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry)
 	if (x->xso.type == XFRM_DEV_OFFLOAD_CRYPTO)
 		esn_msb = xfrm_replay_seqhi(x, htonl(seq_bottom));
 
-	sa_entry->esn_state.esn = esn;
+	if (sa_entry->esn_state.esn_msb)
+		sa_entry->esn_state.esn = esn;
+	else
+		/* According to RFC4303, section "3.3.3. Sequence Number Generation",
+		 * the first packet sent using a given SA will contain a sequence
+		 * number of 1.
+		 */
+		sa_entry->esn_state.esn = max_t(u32, esn, 1);
 	sa_entry->esn_state.esn_msb = esn_msb;
 
 	if (unlikely(overlap && seq_bottom < MLX5E_IPSEC_ESN_SCOPE_MID)) {
@@ -335,6 +342,27 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 		attrs->replay_esn.esn = sa_entry->esn_state.esn;
 		attrs->replay_esn.esn_msb = sa_entry->esn_state.esn_msb;
 		attrs->replay_esn.overlap = sa_entry->esn_state.overlap;
+		switch (x->replay_esn->replay_window) {
+		case 32:
+			attrs->replay_esn.replay_window =
+				MLX5_IPSEC_ASO_REPLAY_WIN_32BIT;
+			break;
+		case 64:
+			attrs->replay_esn.replay_window =
+				MLX5_IPSEC_ASO_REPLAY_WIN_64BIT;
+			break;
+		case 128:
+			attrs->replay_esn.replay_window =
+				MLX5_IPSEC_ASO_REPLAY_WIN_128BIT;
+			break;
+		case 256:
+			attrs->replay_esn.replay_window =
+				MLX5_IPSEC_ASO_REPLAY_WIN_256BIT;
+			break;
+		default:
+			WARN_ON(true);
+			return;
+		}
 	}
 
 	attrs->dir = x->xso.dir;
@@ -907,9 +935,11 @@ void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv)
 		return;
 
 	mlx5e_accel_ipsec_fs_cleanup(ipsec);
-	if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_TUNNEL)
+	if (ipsec->netevent_nb.notifier_call) {
 		unregister_netevent_notifier(&ipsec->netevent_nb);
-	if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD)
+		ipsec->netevent_nb.notifier_call = NULL;
+	}
+	if (ipsec->aso)
 		mlx5e_ipsec_aso_cleanup(ipsec);
 	destroy_workqueue(ipsec->wq);
 	kfree(ipsec);
@@ -1018,6 +1048,12 @@ static int mlx5e_xfrm_validate_policy(struct mlx5_core_dev *mdev,
 		}
 	}
 
+	if (x->xdo.type == XFRM_DEV_OFFLOAD_PACKET &&
+	    !(mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD)) {
+		NL_SET_ERR_MSG_MOD(extack, "Packet offload is not supported");
+		return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -1113,14 +1149,6 @@ static const struct xfrmdev_ops mlx5e_ipsec_xfrmdev_ops = {
 	.xdo_dev_state_free	= mlx5e_xfrm_free_state,
 	.xdo_dev_offload_ok	= mlx5e_ipsec_offload_ok,
 	.xdo_dev_state_advance_esn = mlx5e_xfrm_advance_esn_state,
-};
-
-static const struct xfrmdev_ops mlx5e_ipsec_packet_xfrmdev_ops = {
-	.xdo_dev_state_add	= mlx5e_xfrm_add_state,
-	.xdo_dev_state_delete	= mlx5e_xfrm_del_state,
-	.xdo_dev_state_free	= mlx5e_xfrm_free_state,
-	.xdo_dev_offload_ok	= mlx5e_ipsec_offload_ok,
-	.xdo_dev_state_advance_esn = mlx5e_xfrm_advance_esn_state,
 
 	.xdo_dev_state_update_curlft = mlx5e_xfrm_update_curlft,
 	.xdo_dev_policy_add = mlx5e_xfrm_add_policy,
@@ -1138,11 +1166,7 @@ void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv)
 
 	mlx5_core_info(mdev, "mlx5e: IPSec ESP acceleration enabled\n");
 
-	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD)
-		netdev->xfrmdev_ops = &mlx5e_ipsec_packet_xfrmdev_ops;
-	else
-		netdev->xfrmdev_ops = &mlx5e_ipsec_xfrmdev_ops;
-
+	netdev->xfrmdev_ops = &mlx5e_ipsec_xfrmdev_ops;
 	netdev->features |= NETIF_F_HW_ESP;
 	netdev->hw_enc_features |= NETIF_F_HW_ESP;
 

drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h

@@ -189,11 +189,19 @@ struct mlx5e_ipsec_ft {
 	u32 refcnt;
 };
 
+struct mlx5e_ipsec_drop {
+	struct mlx5_flow_handle *rule;
+	struct mlx5_fc *fc;
+};
+
 struct mlx5e_ipsec_rule {
 	struct mlx5_flow_handle *rule;
 	struct mlx5_modify_hdr *modify_hdr;
 	struct mlx5_pkt_reformat *pkt_reformat;
 	struct mlx5_fc *fc;
+	struct mlx5e_ipsec_drop replay;
+	struct mlx5e_ipsec_drop auth;
+	struct mlx5e_ipsec_drop trailer;
 };
 
 struct mlx5e_ipsec_miss {
@@ -201,19 +209,6 @@ struct mlx5e_ipsec_miss {
 	struct mlx5_flow_handle *rule;
 };
 
-struct mlx5e_ipsec_rx {
-	struct mlx5e_ipsec_ft ft;
-	struct mlx5e_ipsec_miss pol;
-	struct mlx5e_ipsec_miss sa;
-	struct mlx5e_ipsec_rule status;
-	struct mlx5e_ipsec_miss status_drop;
-	struct mlx5_fc *status_drop_cnt;
-	struct mlx5e_ipsec_fc *fc;
-	struct mlx5_fs_chains *chains;
-	u8 allow_tunnel_mode : 1;
-	struct xarray ipsec_obj_id_map;
-};
-
 struct mlx5e_ipsec_tx_create_attr {
 	int prio;
 	int pol_level;
@@ -248,6 +243,7 @@ struct mlx5e_ipsec {
 	struct mlx5_ipsec_fs *roce;
 	u8 is_uplink_rep: 1;
 	struct mlx5e_ipsec_mpv_work mpv_work;
+	struct xarray ipsec_obj_id_map;
 };
 
 struct mlx5e_ipsec_esn_state {

drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c

@@ -32,6 +32,22 @@ struct mlx5e_ipsec_tx {
 	u8 allow_tunnel_mode : 1;
 };
 
+struct mlx5e_ipsec_status_checks {
+	struct mlx5_flow_group *drop_all_group;
+	struct mlx5e_ipsec_drop all;
+};
+
+struct mlx5e_ipsec_rx {
+	struct mlx5e_ipsec_ft ft;
+	struct mlx5e_ipsec_miss pol;
+	struct mlx5e_ipsec_miss sa;
+	struct mlx5e_ipsec_rule status;
+	struct mlx5e_ipsec_status_checks status_drops;
+	struct mlx5e_ipsec_fc *fc;
+	struct mlx5_fs_chains *chains;
+	u8 allow_tunnel_mode : 1;
+};
+
 /* IPsec RX flow steering */
 static enum mlx5_traffic_types family2tt(u32 family)
 {
@@ -128,14 +144,37 @@ static struct mlx5_flow_table *ipsec_ft_create(struct mlx5_flow_namespace *ns,
 	return mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
 }
 
-static int ipsec_status_rule(struct mlx5_core_dev *mdev,
-			     struct mlx5e_ipsec_rx *rx,
-			     struct mlx5_flow_destination *dest)
+static void ipsec_rx_status_drop_destroy(struct mlx5e_ipsec *ipsec,
+					 struct mlx5e_ipsec_rx *rx)
+{
+	mlx5_del_flow_rules(rx->status_drops.all.rule);
+	mlx5_fc_destroy(ipsec->mdev, rx->status_drops.all.fc);
+	mlx5_destroy_flow_group(rx->status_drops.drop_all_group);
+}
+
+static void ipsec_rx_status_pass_destroy(struct mlx5e_ipsec *ipsec,
+					 struct mlx5e_ipsec_rx *rx)
 {
-	u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {};
+	mlx5_del_flow_rules(rx->status.rule);
+
+	if (rx != ipsec->rx_esw)
+		return;
+
+#ifdef CONFIG_MLX5_ESWITCH
+	mlx5_chains_put_table(esw_chains(ipsec->mdev->priv.eswitch), 0, 1, 0);
+#endif
+}
+
+static int rx_add_rule_drop_auth_trailer(struct mlx5e_ipsec_sa_entry *sa_entry,
+					 struct mlx5e_ipsec_rx *rx)
+{
+	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
+	struct mlx5_flow_table *ft = rx->ft.status;
+	struct mlx5_core_dev *mdev = ipsec->mdev;
+	struct mlx5_flow_destination dest = {};
 	struct mlx5_flow_act flow_act = {};
-	struct mlx5_modify_hdr *modify_hdr;
-	struct mlx5_flow_handle *fte;
+	struct mlx5_flow_handle *rule;
+	struct mlx5_fc *flow_counter;
 	struct mlx5_flow_spec *spec;
 	int err;
 
@@ -143,48 +182,273 @@ static int ipsec_status_rule(struct mlx5_core_dev *mdev,
 	if (!spec)
 		return -ENOMEM;
 
-	/* Action to copy 7 bit ipsec_syndrome to regB[24:30] */
-	MLX5_SET(copy_action_in, action, action_type, MLX5_ACTION_TYPE_COPY);
-	MLX5_SET(copy_action_in, action, src_field, MLX5_ACTION_IN_FIELD_IPSEC_SYNDROME);
-	MLX5_SET(copy_action_in, action, src_offset, 0);
-	MLX5_SET(copy_action_in, action, length, 7);
-	MLX5_SET(copy_action_in, action, dst_field, MLX5_ACTION_IN_FIELD_METADATA_REG_B);
-	MLX5_SET(copy_action_in, action, dst_offset, 24);
+	flow_counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(flow_counter)) {
+		err = PTR_ERR(flow_counter);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
+		goto err_cnt;
+	}
+	sa_entry->ipsec_rule.auth.fc = flow_counter;
 
-	modify_hdr = mlx5_modify_header_alloc(mdev, MLX5_FLOW_NAMESPACE_KERNEL,
-					      1, action);
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	flow_act.flags = FLOW_ACT_NO_APPEND;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest.counter_id = mlx5_fc_id(flow_counter);
+	if (rx == ipsec->rx_esw)
+		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
 
-	if (IS_ERR(modify_hdr)) {
-		err = PTR_ERR(modify_hdr);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.ipsec_syndrome);
+	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.ipsec_syndrome, 1);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_2);
+	MLX5_SET(fte_match_param, spec->match_value,
+		 misc_parameters_2.metadata_reg_c_2,
+		 sa_entry->ipsec_obj_id | BIT(31));
+	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
 		mlx5_core_err(mdev,
-			      "fail to alloc ipsec copy modify_header_id err=%d\n", err);
-		goto out_spec;
+			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
+		goto err_rule;
 	}
+	sa_entry->ipsec_rule.auth.rule = rule;
 
-	/* create fte */
-	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_MOD_HDR |
-			  MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+	flow_counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(flow_counter)) {
+		err = PTR_ERR(flow_counter);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
+		goto err_cnt_2;
+	}
+	sa_entry->ipsec_rule.trailer.fc = flow_counter;
+
+	dest.counter_id = mlx5_fc_id(flow_counter);
+	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.ipsec_syndrome, 2);
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
+		goto err_rule_2;
+	}
+	sa_entry->ipsec_rule.trailer.rule = rule;
+
+	kvfree(spec);
+	return 0;
+
+err_rule_2:
+	mlx5_fc_destroy(mdev, sa_entry->ipsec_rule.trailer.fc);
+err_cnt_2:
+	mlx5_del_flow_rules(sa_entry->ipsec_rule.auth.rule);
+err_rule:
+	mlx5_fc_destroy(mdev, sa_entry->ipsec_rule.auth.fc);
+err_cnt:
+	kvfree(spec);
+	return err;
+}
+
+static int rx_add_rule_drop_replay(struct mlx5e_ipsec_sa_entry *sa_entry, struct mlx5e_ipsec_rx *rx)
+{
+	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
+	struct mlx5_flow_table *ft = rx->ft.status;
+	struct mlx5_core_dev *mdev = ipsec->mdev;
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act flow_act = {};
+	struct mlx5_flow_handle *rule;
+	struct mlx5_fc *flow_counter;
+	struct mlx5_flow_spec *spec;
+	int err;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec)
+		return -ENOMEM;
+
+	flow_counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(flow_counter)) {
+		err = PTR_ERR(flow_counter);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
+		goto err_cnt;
+	}
+
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	flow_act.flags = FLOW_ACT_NO_APPEND;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest.counter_id = mlx5_fc_id(flow_counter);
+	if (rx == ipsec->rx_esw)
+		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
+
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_4);
+	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.metadata_reg_c_4, 1);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_2);
+	MLX5_SET(fte_match_param, spec->match_value,  misc_parameters_2.metadata_reg_c_2,
+		 sa_entry->ipsec_obj_id | BIT(31));
+	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
+		goto err_rule;
+	}
+
+	sa_entry->ipsec_rule.replay.rule = rule;
+	sa_entry->ipsec_rule.replay.fc = flow_counter;
+
+	kvfree(spec);
+	return 0;
+
+err_rule:
+	mlx5_fc_destroy(mdev, flow_counter);
+err_cnt:
+	kvfree(spec);
+	return err;
+}
+
+static int ipsec_rx_status_drop_all_create(struct mlx5e_ipsec *ipsec,
+					   struct mlx5e_ipsec_rx *rx)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_table *ft = rx->ft.status;
+	struct mlx5_core_dev *mdev = ipsec->mdev;
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act flow_act = {};
+	struct mlx5_flow_handle *rule;
+	struct mlx5_fc *flow_counter;
+	struct mlx5_flow_spec *spec;
+	struct mlx5_flow_group *g;
+	u32 *flow_group_in;
+	int err = 0;
+
+	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!flow_group_in || !spec) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ft->max_fte - 1);
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ft->max_fte - 1);
+	g = mlx5_create_flow_group(ft, flow_group_in);
+	if (IS_ERR(g)) {
+		err = PTR_ERR(g);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop flow group, err=%d\n", err);
+		goto err_out;
+	}
+
+	flow_counter = mlx5_fc_create(mdev, false);
+	if (IS_ERR(flow_counter)) {
+		err = PTR_ERR(flow_counter);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
+		goto err_cnt;
+	}
+
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest.counter_id = mlx5_fc_id(flow_counter);
+	if (rx == ipsec->rx_esw)
+		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_err(mdev,
+			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
+		goto err_rule;
+	}
+
+	rx->status_drops.drop_all_group = g;
+	rx->status_drops.all.rule = rule;
+	rx->status_drops.all.fc = flow_counter;
+
+	kvfree(flow_group_in);
+	kvfree(spec);
+	return 0;
+
+err_rule:
+	mlx5_fc_destroy(mdev, flow_counter);
+err_cnt:
+	mlx5_destroy_flow_group(g);
+err_out:
+	kvfree(flow_group_in);
+	kvfree(spec);
+	return err;
+}
+
+static int ipsec_rx_status_pass_create(struct mlx5e_ipsec *ipsec,
+				       struct mlx5e_ipsec_rx *rx,
+				       struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_act flow_act = {};
+	struct mlx5_flow_handle *rule;
+	struct mlx5_flow_spec *spec;
+	int err;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec)
+		return -ENOMEM;
+
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+			 misc_parameters_2.ipsec_syndrome);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+			 misc_parameters_2.metadata_reg_c_4);
+	MLX5_SET(fte_match_param, spec->match_value,
+		 misc_parameters_2.ipsec_syndrome, 0);
+	MLX5_SET(fte_match_param, spec->match_value,
+		 misc_parameters_2.metadata_reg_c_4, 0);
+	if (rx == ipsec->rx_esw)
+		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
+	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+	flow_act.flags = FLOW_ACT_NO_APPEND;
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
 			  MLX5_FLOW_CONTEXT_ACTION_COUNT;
-	flow_act.modify_hdr = modify_hdr;
-	fte = mlx5_add_flow_rules(rx->ft.status, spec, &flow_act, dest, 2);
-	if (IS_ERR(fte)) {
-		err = PTR_ERR(fte);
-		mlx5_core_err(mdev, "fail to add ipsec rx err copy rule err=%d\n", err);
-		goto out;
+	rule = mlx5_add_flow_rules(rx->ft.status, spec, &flow_act, dest, 2);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		mlx5_core_warn(ipsec->mdev,
+			       "Failed to add ipsec rx status pass rule, err=%d\n", err);
+		goto err_rule;
 	}
 
+	rx->status.rule = rule;
 	kvfree(spec);
-	rx->status.rule = fte;
-	rx->status.modify_hdr = modify_hdr;
 	return 0;
 
-out:
-	mlx5_modify_header_dealloc(mdev, modify_hdr);
-out_spec:
+err_rule:
 	kvfree(spec);
 	return err;
 }
 
+static void mlx5_ipsec_rx_status_destroy(struct mlx5e_ipsec *ipsec,
+					 struct mlx5e_ipsec_rx *rx)
+{
+	ipsec_rx_status_pass_destroy(ipsec, rx);
+	ipsec_rx_status_drop_destroy(ipsec, rx);
+}
+
+static int mlx5_ipsec_rx_status_create(struct mlx5e_ipsec *ipsec,
+				       struct mlx5e_ipsec_rx *rx,
+				       struct mlx5_flow_destination *dest)
+{
+	int err;
+
+	err = ipsec_rx_status_drop_all_create(ipsec, rx);
+	if (err)
+		return err;
+
+	err = ipsec_rx_status_pass_create(ipsec, rx, dest);
+	if (err)
+		goto err_pass_create;
+
+	return 0;
+
+err_pass_create:
+	ipsec_rx_status_drop_destroy(ipsec, rx);
+	return err;
+}
+
 static int ipsec_miss_create(struct mlx5_core_dev *mdev,
 			     struct mlx5_flow_table *ft,
 			     struct mlx5e_ipsec_miss *miss,
@@ -333,12 +597,7 @@ static void rx_destroy(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	mlx5_destroy_flow_table(rx->ft.sa);
 	if (rx->allow_tunnel_mode)
 		mlx5_eswitch_unblock_encap(mdev);
-	if (rx == ipsec->rx_esw) {
-		mlx5_esw_ipsec_rx_status_destroy(ipsec, rx);
-	} else {
-		mlx5_del_flow_rules(rx->status.rule);
-		mlx5_modify_header_dealloc(mdev, rx->status.modify_hdr);
-	}
+	mlx5_ipsec_rx_status_destroy(ipsec, rx);
 	mlx5_destroy_flow_table(rx->ft.status);
 
 	mlx5_ipsec_fs_roce_rx_destroy(ipsec->roce, family, mdev);
@@ -419,7 +678,7 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	if (err)
 		return err;
 
-	ft = ipsec_ft_create(attr.ns, attr.status_level, attr.prio, 1, 0);
+	ft = ipsec_ft_create(attr.ns, attr.status_level, attr.prio, 3, 0);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
 		goto err_fs_ft_status;
@@ -428,10 +687,7 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 
 	dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
 	dest[1].counter_id = mlx5_fc_id(rx->fc->cnt);
-	if (rx == ipsec->rx_esw)
-		err = mlx5_esw_ipsec_rx_status_create(ipsec, rx, dest);
-	else
-		err = ipsec_status_rule(mdev, rx, dest);
+	err = mlx5_ipsec_rx_status_create(ipsec, rx, dest);
 	if (err)
 		goto err_add;
 
@@ -956,13 +1212,22 @@ static void setup_fte_esp(struct mlx5_flow_spec *spec)
 	MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_protocol, IPPROTO_ESP);
 }
 
-static void setup_fte_spi(struct mlx5_flow_spec *spec, u32 spi)
+static void setup_fte_spi(struct mlx5_flow_spec *spec, u32 spi, bool encap)
 {
 	/* SPI number */
 	spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS;
 
-	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters.outer_esp_spi);
-	MLX5_SET(fte_match_param, spec->match_value, misc_parameters.outer_esp_spi, spi);
+	if (encap) {
+		MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+				 misc_parameters.inner_esp_spi);
+		MLX5_SET(fte_match_param, spec->match_value,
+			 misc_parameters.inner_esp_spi, spi);
+	} else {
+		MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+				 misc_parameters.outer_esp_spi);
+		MLX5_SET(fte_match_param, spec->match_value,
+			 misc_parameters.outer_esp_spi, spi);
+	}
 }
 
 static void setup_fte_no_frags(struct mlx5_flow_spec *spec)
@@ -1052,29 +1317,48 @@ static int setup_modify_header(struct mlx5e_ipsec *ipsec, int type, u32 val, u8
 			       struct mlx5_flow_act *flow_act)
 {
 	enum mlx5_flow_namespace_type ns_type = ipsec_fs_get_ns(ipsec, type, dir);
-	u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {};
+	u8 action[3][MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {};
 	struct mlx5_core_dev *mdev = ipsec->mdev;
 	struct mlx5_modify_hdr *modify_hdr;
+	u8 num_of_actions = 1;
 
-	MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET);
+	MLX5_SET(set_action_in, action[0], action_type, MLX5_ACTION_TYPE_SET);
 	switch (dir) {
 	case XFRM_DEV_OFFLOAD_IN:
-		MLX5_SET(set_action_in, action, field,
+		MLX5_SET(set_action_in, action[0], field,
 			 MLX5_ACTION_IN_FIELD_METADATA_REG_B);
+
+		num_of_actions++;
+		MLX5_SET(set_action_in, action[1], action_type, MLX5_ACTION_TYPE_SET);
+		MLX5_SET(set_action_in, action[1], field, MLX5_ACTION_IN_FIELD_METADATA_REG_C_2);
+		MLX5_SET(set_action_in, action[1], data, val);
+		MLX5_SET(set_action_in, action[1], offset, 0);
+		MLX5_SET(set_action_in, action[1], length, 32);
+
+		if (type == XFRM_DEV_OFFLOAD_CRYPTO) {
+			num_of_actions++;
+			MLX5_SET(set_action_in, action[2], action_type,
+				 MLX5_ACTION_TYPE_SET);
+			MLX5_SET(set_action_in, action[2], field,
+				 MLX5_ACTION_IN_FIELD_METADATA_REG_C_4);
+			MLX5_SET(set_action_in, action[2], data, 0);
+			MLX5_SET(set_action_in, action[2], offset, 0);
+			MLX5_SET(set_action_in, action[2], length, 32);
+		}
 		break;
 	case XFRM_DEV_OFFLOAD_OUT:
-		MLX5_SET(set_action_in, action, field,
+		MLX5_SET(set_action_in, action[0], field,
 			 MLX5_ACTION_IN_FIELD_METADATA_REG_C_4);
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	MLX5_SET(set_action_in, action, data, val);
-	MLX5_SET(set_action_in, action, offset, 0);
-	MLX5_SET(set_action_in, action, length, 32);
+	MLX5_SET(set_action_in, action[0], data, val);
+	MLX5_SET(set_action_in, action[0], offset, 0);
+	MLX5_SET(set_action_in, action[0], length, 32);
 
-	modify_hdr = mlx5_modify_header_alloc(mdev, ns_type, 1, action);
+	modify_hdr = mlx5_modify_header_alloc(mdev, ns_type, num_of_actions, action);
 	if (IS_ERR(modify_hdr)) {
 		mlx5_core_err(mdev, "Failed to allocate modify_header %ld\n",
 			      PTR_ERR(modify_hdr));
@@ -1321,8 +1605,9 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	else
 		setup_fte_addr6(spec, attrs->saddr.a6, attrs->daddr.a6);
 
-	setup_fte_spi(spec, attrs->spi);
-	setup_fte_esp(spec);
+	setup_fte_spi(spec, attrs->spi, attrs->encap);
+	if (!attrs->encap)
+		setup_fte_esp(spec);
 	setup_fte_no_frags(spec);
 	setup_fte_upper_proto_match(spec, &attrs->upspec);
 
@@ -1372,6 +1657,15 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 		mlx5_core_err(mdev, "fail to add RX ipsec rule err=%d\n", err);
 		goto err_add_flow;
 	}
+	if (attrs->type == XFRM_DEV_OFFLOAD_PACKET)
+		err = rx_add_rule_drop_replay(sa_entry, rx);
+	if (err)
+		goto err_add_replay;
+
+	err = rx_add_rule_drop_auth_trailer(sa_entry, rx);
+	if (err)
+		goto err_drop_reason;
+
 	kvfree(spec);
 
 	sa_entry->ipsec_rule.rule = rule;
@@ -1380,6 +1674,13 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	sa_entry->ipsec_rule.pkt_reformat = flow_act.pkt_reformat;
 	return 0;
 
+err_drop_reason:
+	if (sa_entry->ipsec_rule.replay.rule) {
+		mlx5_del_flow_rules(sa_entry->ipsec_rule.replay.rule);
+		mlx5_fc_destroy(mdev, sa_entry->ipsec_rule.replay.fc);
+	}
+err_add_replay:
+	mlx5_del_flow_rules(rule);
 err_add_flow:
 	mlx5_fc_destroy(mdev, counter);
 err_add_cnt:
@@ -1428,7 +1729,7 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 
 	switch (attrs->type) {
 	case XFRM_DEV_OFFLOAD_CRYPTO:
-		setup_fte_spi(spec, attrs->spi);
+		setup_fte_spi(spec, attrs->spi, false);
 		setup_fte_esp(spec);
 		setup_fte_reg_a(spec);
 		break;
@@ -1809,8 +2110,11 @@ static int mlx5e_ipsec_block_tc_offload(struct mlx5_core_dev *mdev)
 	struct mlx5_eswitch *esw = mdev->priv.eswitch;
 	int err = 0;
 
-	if (esw)
-		down_write(&esw->mode_lock);
+	if (esw) {
+		err = mlx5_esw_lock(esw);
+		if (err)
+			return err;
+	}
 
 	if (mdev->num_block_ipsec) {
 		err = -EBUSY;
@@ -1821,7 +2125,7 @@ static int mlx5e_ipsec_block_tc_offload(struct mlx5_core_dev *mdev)
 
 unlock:
 	if (esw)
-		up_write(&esw->mode_lock);
+		mlx5_esw_unlock(esw);
 
 	return err;
 }
@@ -1887,6 +2191,17 @@ void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 
 	if (ipsec_rule->modify_hdr)
 		mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr);
+
+	mlx5_del_flow_rules(ipsec_rule->trailer.rule);
+	mlx5_fc_destroy(mdev, ipsec_rule->trailer.fc);
+
+	mlx5_del_flow_rules(ipsec_rule->auth.rule);
+	mlx5_fc_destroy(mdev, ipsec_rule->auth.fc);
+
+	if (ipsec_rule->replay.rule) {
+		mlx5_del_flow_rules(ipsec_rule->replay.rule);
+		mlx5_fc_destroy(mdev, ipsec_rule->replay.fc);
+	}
 	mlx5_esw_ipsec_rx_id_mapping_remove(sa_entry);
 	rx_ft_put(sa_entry->ipsec, sa_entry->attrs.family, sa_entry->attrs.type);
 }
@@ -1957,7 +2272,7 @@ void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec)
 	kfree(ipsec->rx_ipv6);
 
 	if (ipsec->is_uplink_rep) {
-		xa_destroy(&ipsec->rx_esw->ipsec_obj_id_map);
+		xa_destroy(&ipsec->ipsec_obj_id_map);
 
 		mutex_destroy(&ipsec->tx_esw->ft.mutex);
 		WARN_ON(ipsec->tx_esw->ft.refcnt);
@@ -2020,7 +2335,7 @@ int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec,
 		mutex_init(&ipsec->tx_esw->ft.mutex);
 		mutex_init(&ipsec->rx_esw->ft.mutex);
 		ipsec->tx_esw->ns = ns_esw;
-		xa_init_flags(&ipsec->rx_esw->ipsec_obj_id_map, XA_FLAGS_ALLOC1);
+		xa_init_flags(&ipsec->ipsec_obj_id_map, XA_FLAGS_ALLOC1);
 	} else if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_ROCE) {
 		ipsec->roce = mlx5_ipsec_fs_roce_init(mdev, devcom);
 	} else {

drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c

@@ -6,6 +6,8 @@
 #include "ipsec.h"
 #include "lib/crypto.h"
 #include "lib/ipsec_fs_roce.h"
+#include "fs_core.h"
+#include "eswitch.h"
 
 enum {
 	MLX5_IPSEC_ASO_REMOVE_FLOW_PKT_CNT_OFFSET,
@@ -38,7 +40,10 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev)
 	    MLX5_CAP_ETH(mdev, insert_trailer) && MLX5_CAP_ETH(mdev, swp))
 		caps |= MLX5_IPSEC_CAP_CRYPTO;
 
-	if (MLX5_CAP_IPSEC(mdev, ipsec_full_offload)) {
+	if (MLX5_CAP_IPSEC(mdev, ipsec_full_offload) &&
+	    (mdev->priv.steering->mode == MLX5_FLOW_STEERING_MODE_DMFS ||
+	     (mdev->priv.steering->mode == MLX5_FLOW_STEERING_MODE_SMFS &&
+	     is_mdev_legacy_mode(mdev)))) {
 		if (MLX5_CAP_FLOWTABLE_NIC_TX(mdev,
 					      reformat_add_esp_trasport) &&
 		    MLX5_CAP_FLOWTABLE_NIC_RX(mdev,
@@ -95,7 +100,7 @@ static void mlx5e_ipsec_packet_setup(void *obj, u32 pdn,
 
 		if (attrs->dir == XFRM_DEV_OFFLOAD_IN) {
 			MLX5_SET(ipsec_aso, aso_ctx, window_sz,
-				 attrs->replay_esn.replay_window / 64);
+				 attrs->replay_esn.replay_window);
 			MLX5_SET(ipsec_aso, aso_ctx, mode,
 				 MLX5_IPSEC_ASO_REPLAY_PROTECTION);
 		}
@@ -559,6 +564,7 @@ void mlx5e_ipsec_aso_cleanup(struct mlx5e_ipsec *ipsec)
 	dma_unmap_single(pdev, aso->dma_addr, sizeof(aso->ctx),
 			 DMA_BIDIRECTIONAL);
 	kfree(aso);
+	ipsec->aso = NULL;
 }
 
 static void mlx5e_ipsec_aso_copy(struct mlx5_wqe_aso_ctrl_seg *ctrl,

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

@@ -2731,6 +2731,7 @@ void mlx5e_close_channels(struct mlx5e_channels *chs)
 {
 	int i;
 
+	ASSERT_RTNL();
 	if (chs->ptp) {
 		mlx5e_ptp_close(chs->ptp);
 		chs->ptp = NULL;
@@ -3012,17 +3013,29 @@ void mlx5e_activate_priv_channels(struct mlx5e_priv *priv)
 	if (mlx5e_is_vport_rep(priv))
 		mlx5e_rep_activate_channels(priv);
 
+	set_bit(MLX5E_STATE_CHANNELS_ACTIVE, &priv->state);
+
 	mlx5e_wait_channels_min_rx_wqes(&priv->channels);
 
 	if (priv->rx_res)
 		mlx5e_rx_res_channels_activate(priv->rx_res, &priv->channels);
 }
 
+static void mlx5e_cancel_tx_timeout_work(struct mlx5e_priv *priv)
+{
+	WARN_ON_ONCE(test_bit(MLX5E_STATE_CHANNELS_ACTIVE, &priv->state));
+	if (current_work() != &priv->tx_timeout_work)
+		cancel_work_sync(&priv->tx_timeout_work);
+}
+
 void mlx5e_deactivate_priv_channels(struct mlx5e_priv *priv)
 {
 	if (priv->rx_res)
 		mlx5e_rx_res_channels_deactivate(priv->rx_res);
 
+	clear_bit(MLX5E_STATE_CHANNELS_ACTIVE, &priv->state);
+	mlx5e_cancel_tx_timeout_work(priv);
+
 	if (mlx5e_is_vport_rep(priv))
 		mlx5e_rep_deactivate_channels(priv);
 
@@ -4801,8 +4814,17 @@ static void mlx5e_tx_timeout_work(struct work_struct *work)
 	struct net_device *netdev = priv->netdev;
 	int i;
 
-	rtnl_lock();
-	mutex_lock(&priv->state_lock);
+	/* Take rtnl_lock to ensure no change in netdev->real_num_tx_queues
+	 * through this flow. However, channel closing flows have to wait for
+	 * this work to finish while holding rtnl lock too. So either get the
+	 * lock or find that channels are being closed for other reason and
+	 * this work is not relevant anymore.
+	 */
+	while (!rtnl_trylock()) {
+		if (!test_bit(MLX5E_STATE_CHANNELS_ACTIVE, &priv->state))
+			return;
+		msleep(20);
+	}
 
 	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
 		goto unlock;
@@ -4821,7 +4843,6 @@ static void mlx5e_tx_timeout_work(struct work_struct *work)
 	}
 
 unlock:
-	mutex_unlock(&priv->state_lock);
 	rtnl_unlock();
 }
 

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

@@ -1497,7 +1497,7 @@ mlx5e_vport_vf_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
 
 	dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch,
 						 rpriv->rep->vport);
-	if (dl_port) {
+	if (!IS_ERR(dl_port)) {
 		SET_NETDEV_DEVLINK_PORT(netdev, dl_port);
 		mlx5e_rep_vnic_reporter_create(priv, dl_port);
 	}

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

@@ -444,6 +444,9 @@ mlx5e_tc_add_flow_meter(struct mlx5e_priv *priv,
 	struct mlx5e_flow_meter_handle *meter;
 	enum mlx5e_post_meter_type type;
 
+	if (IS_ERR(post_act))
+		return PTR_ERR(post_act);
+
 	meter = mlx5e_tc_meter_replace(priv->mdev, &attr->meter_attr.params);
 	if (IS_ERR(meter)) {
 		mlx5_core_err(priv->mdev, "Failed to get flow meter\n");
@@ -3738,6 +3741,20 @@ alloc_flow_post_acts(struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack)
 	return err;
 }
 
+static int
+set_branch_dest_ft(struct mlx5e_priv *priv, struct mlx5_flow_attr *attr)
+{
+	struct mlx5e_post_act *post_act = get_post_action(priv);
+
+	if (IS_ERR(post_act))
+		return PTR_ERR(post_act);
+
+	attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	attr->dest_ft = mlx5e_tc_post_act_get_ft(post_act);
+
+	return 0;
+}
+
 static int
 alloc_branch_attr(struct mlx5e_tc_flow *flow,
 		  struct mlx5e_tc_act_branch_ctrl *cond,
@@ -3761,8 +3778,8 @@ alloc_branch_attr(struct mlx5e_tc_flow *flow,
 		break;
 	case FLOW_ACTION_ACCEPT:
 	case FLOW_ACTION_PIPE:
-		attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
-		attr->dest_ft = mlx5e_tc_post_act_get_ft(get_post_action(flow->priv));
+		if (set_branch_dest_ft(flow->priv, attr))
+			goto out_err;
 		break;
 	case FLOW_ACTION_JUMP:
 		if (*jump_count) {
@@ -3771,8 +3788,8 @@ alloc_branch_attr(struct mlx5e_tc_flow *flow,
 			goto out_err;
 		}
 		*jump_count = cond->extval;
-		attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
-		attr->dest_ft = mlx5e_tc_post_act_get_ft(get_post_action(flow->priv));
+		if (set_branch_dest_ft(flow->priv, attr))
+			goto out_err;
 		break;
 	default:
 		err = -EOPNOTSUPP;

drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c

@@ -21,158 +21,6 @@ enum {
 	MLX5_ESW_IPSEC_TX_ESP_FT_CNT_LEVEL,
 };
 
-static void esw_ipsec_rx_status_drop_destroy(struct mlx5e_ipsec *ipsec,
-					     struct mlx5e_ipsec_rx *rx)
-{
-	mlx5_del_flow_rules(rx->status_drop.rule);
-	mlx5_destroy_flow_group(rx->status_drop.group);
-	mlx5_fc_destroy(ipsec->mdev, rx->status_drop_cnt);
-}
-
-static void esw_ipsec_rx_status_pass_destroy(struct mlx5e_ipsec *ipsec,
-					     struct mlx5e_ipsec_rx *rx)
-{
-	mlx5_del_flow_rules(rx->status.rule);
-	mlx5_chains_put_table(esw_chains(ipsec->mdev->priv.eswitch), 0, 1, 0);
-}
-
-static int esw_ipsec_rx_status_drop_create(struct mlx5e_ipsec *ipsec,
-					   struct mlx5e_ipsec_rx *rx)
-{
-	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
-	struct mlx5_flow_table *ft = rx->ft.status;
-	struct mlx5_core_dev *mdev = ipsec->mdev;
-	struct mlx5_flow_destination dest = {};
-	struct mlx5_flow_act flow_act = {};
-	struct mlx5_flow_handle *rule;
-	struct mlx5_fc *flow_counter;
-	struct mlx5_flow_spec *spec;
-	struct mlx5_flow_group *g;
-	u32 *flow_group_in;
-	int err = 0;
-
-	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
-	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
-	if (!flow_group_in || !spec) {
-		err = -ENOMEM;
-		goto err_out;
-	}
-
-	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ft->max_fte - 1);
-	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ft->max_fte - 1);
-	g = mlx5_create_flow_group(ft, flow_group_in);
-	if (IS_ERR(g)) {
-		err = PTR_ERR(g);
-		mlx5_core_err(mdev,
-			      "Failed to add ipsec rx status drop flow group, err=%d\n", err);
-		goto err_out;
-	}
-
-	flow_counter = mlx5_fc_create(mdev, false);
-	if (IS_ERR(flow_counter)) {
-		err = PTR_ERR(flow_counter);
-		mlx5_core_err(mdev,
-			      "Failed to add ipsec rx status drop rule counter, err=%d\n", err);
-		goto err_cnt;
-	}
-
-	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT;
-	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
-	dest.counter_id = mlx5_fc_id(flow_counter);
-	spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
-	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
-	if (IS_ERR(rule)) {
-		err = PTR_ERR(rule);
-		mlx5_core_err(mdev,
-			      "Failed to add ipsec rx status drop rule, err=%d\n", err);
-		goto err_rule;
-	}
-
-	rx->status_drop.group = g;
-	rx->status_drop.rule = rule;
-	rx->status_drop_cnt = flow_counter;
-
-	kvfree(flow_group_in);
-	kvfree(spec);
-	return 0;
-
-err_rule:
-	mlx5_fc_destroy(mdev, flow_counter);
-err_cnt:
-	mlx5_destroy_flow_group(g);
-err_out:
-	kvfree(flow_group_in);
-	kvfree(spec);
-	return err;
-}
-
-static int esw_ipsec_rx_status_pass_create(struct mlx5e_ipsec *ipsec,
-					   struct mlx5e_ipsec_rx *rx,
-					   struct mlx5_flow_destination *dest)
-{
-	struct mlx5_flow_act flow_act = {};
-	struct mlx5_flow_handle *rule;
-	struct mlx5_flow_spec *spec;
-	int err;
-
-	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
-	if (!spec)
-		return -ENOMEM;
-
-	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
-			 misc_parameters_2.ipsec_syndrome);
-	MLX5_SET(fte_match_param, spec->match_value,
-		 misc_parameters_2.ipsec_syndrome, 0);
-	spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
-	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
-	flow_act.flags = FLOW_ACT_NO_APPEND;
-	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-			  MLX5_FLOW_CONTEXT_ACTION_COUNT;
-	rule = mlx5_add_flow_rules(rx->ft.status, spec, &flow_act, dest, 2);
-	if (IS_ERR(rule)) {
-		err = PTR_ERR(rule);
-		mlx5_core_warn(ipsec->mdev,
-			       "Failed to add ipsec rx status pass rule, err=%d\n", err);
-		goto err_rule;
-	}
-
-	rx->status.rule = rule;
-	kvfree(spec);
-	return 0;
-
-err_rule:
-	kvfree(spec);
-	return err;
-}
-
-void mlx5_esw_ipsec_rx_status_destroy(struct mlx5e_ipsec *ipsec,
-				      struct mlx5e_ipsec_rx *rx)
-{
-	esw_ipsec_rx_status_pass_destroy(ipsec, rx);
-	esw_ipsec_rx_status_drop_destroy(ipsec, rx);
-}
-
-int mlx5_esw_ipsec_rx_status_create(struct mlx5e_ipsec *ipsec,
-				    struct mlx5e_ipsec_rx *rx,
-				    struct mlx5_flow_destination *dest)
-{
-	int err;
-
-	err = esw_ipsec_rx_status_drop_create(ipsec, rx);
-	if (err)
-		return err;
-
-	err = esw_ipsec_rx_status_pass_create(ipsec, rx, dest);
-	if (err)
-		goto err_pass_create;
-
-	return 0;
-
-err_pass_create:
-	esw_ipsec_rx_status_drop_destroy(ipsec, rx);
-	return err;
-}
-
 void mlx5_esw_ipsec_rx_create_attr_set(struct mlx5e_ipsec *ipsec,
 				       struct mlx5e_ipsec_rx_create_attr *attr)
 {
@@ -202,7 +50,7 @@ int mlx5_esw_ipsec_rx_setup_modify_header(struct mlx5e_ipsec_sa_entry *sa_entry,
 	u32 mapped_id;
 	int err;
 
-	err = xa_alloc_bh(&ipsec->rx_esw->ipsec_obj_id_map, &mapped_id,
+	err = xa_alloc_bh(&ipsec->ipsec_obj_id_map, &mapped_id,
 			  xa_mk_value(sa_entry->ipsec_obj_id),
 			  XA_LIMIT(1, ESW_IPSEC_RX_MAPPED_ID_MASK), 0);
 	if (err)
@@ -233,7 +81,7 @@ int mlx5_esw_ipsec_rx_setup_modify_header(struct mlx5e_ipsec_sa_entry *sa_entry,
 	return 0;
 
 err_header_alloc:
-	xa_erase_bh(&ipsec->rx_esw->ipsec_obj_id_map, mapped_id);
+	xa_erase_bh(&ipsec->ipsec_obj_id_map, mapped_id);
 	return err;
 }
 
@@ -242,7 +90,7 @@ void mlx5_esw_ipsec_rx_id_mapping_remove(struct mlx5e_ipsec_sa_entry *sa_entry)
 	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
 
 	if (sa_entry->rx_mapped_id)
-		xa_erase_bh(&ipsec->rx_esw->ipsec_obj_id_map,
+		xa_erase_bh(&ipsec->ipsec_obj_id_map,
 			    sa_entry->rx_mapped_id);
 }
 
@@ -252,7 +100,7 @@ int mlx5_esw_ipsec_rx_ipsec_obj_id_search(struct mlx5e_priv *priv, u32 id,
 	struct mlx5e_ipsec *ipsec = priv->ipsec;
 	void *val;
 
-	val = xa_load(&ipsec->rx_esw->ipsec_obj_id_map, id);
+	val = xa_load(&ipsec->ipsec_obj_id_map, id);
 	if (!val)
 		return -ENOENT;
 
@@ -304,7 +152,7 @@ void mlx5_esw_ipsec_restore_dest_uplink(struct mlx5_core_dev *mdev)
 
 	xa_for_each(&esw->offloads.vport_reps, i, rep) {
 		rpriv = rep->rep_data[REP_ETH].priv;
-		if (!rpriv || !rpriv->netdev)
+		if (!rpriv || !rpriv->netdev || !atomic_read(&rpriv->tc_ht.nelems))
 			continue;
 
 		rhashtable_walk_enter(&rpriv->tc_ht, &iter);

drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.h

@@ -8,11 +8,6 @@ struct mlx5e_ipsec;
 struct mlx5e_ipsec_sa_entry;
 
 #ifdef CONFIG_MLX5_ESWITCH
-void mlx5_esw_ipsec_rx_status_destroy(struct mlx5e_ipsec *ipsec,
-				      struct mlx5e_ipsec_rx *rx);
-int mlx5_esw_ipsec_rx_status_create(struct mlx5e_ipsec *ipsec,
-				    struct mlx5e_ipsec_rx *rx,
-				    struct mlx5_flow_destination *dest);
 void mlx5_esw_ipsec_rx_create_attr_set(struct mlx5e_ipsec *ipsec,
 				       struct mlx5e_ipsec_rx_create_attr *attr);
 int mlx5_esw_ipsec_rx_status_pass_dest_get(struct mlx5e_ipsec *ipsec,
@@ -26,16 +21,6 @@ void mlx5_esw_ipsec_tx_create_attr_set(struct mlx5e_ipsec *ipsec,
 				       struct mlx5e_ipsec_tx_create_attr *attr);
 void mlx5_esw_ipsec_restore_dest_uplink(struct mlx5_core_dev *mdev);
 #else
-static inline void mlx5_esw_ipsec_rx_status_destroy(struct mlx5e_ipsec *ipsec,
-						    struct mlx5e_ipsec_rx *rx) {}
-
-static inline int mlx5_esw_ipsec_rx_status_create(struct mlx5e_ipsec *ipsec,
-						  struct mlx5e_ipsec_rx *rx,
-						  struct mlx5_flow_destination *dest)
-{
-	return  -EINVAL;
-}
-
 static inline void mlx5_esw_ipsec_rx_create_attr_set(struct mlx5e_ipsec *ipsec,
 						     struct mlx5e_ipsec_rx_create_attr *attr) {}
 

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

@@ -1463,7 +1463,7 @@ int mlx5_eswitch_enable_locked(struct mlx5_eswitch *esw, int num_vfs)
 {
 	int err;
 
-	lockdep_assert_held(&esw->mode_lock);
+	devl_assert_locked(priv_to_devlink(esw->dev));
 
 	if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ft_support)) {
 		esw_warn(esw->dev, "FDB is not supported, aborting ...\n");
@@ -1531,7 +1531,6 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs)
 	if (toggle_lag)
 		mlx5_lag_disable_change(esw->dev);
 
-	down_write(&esw->mode_lock);
 	if (!mlx5_esw_is_fdb_created(esw)) {
 		ret = mlx5_eswitch_enable_locked(esw, num_vfs);
 	} else {
@@ -1554,8 +1553,6 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs)
 		}
 	}
 
-	up_write(&esw->mode_lock);
-
 	if (toggle_lag)
 		mlx5_lag_enable_change(esw->dev);
 
@@ -1569,12 +1566,11 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf)
 		return;
 
 	devl_assert_locked(priv_to_devlink(esw->dev));
-	down_write(&esw->mode_lock);
 	/* 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 && !esw->esw_funcs.num_ec_vfs && !clear_vf)
-		goto unlock;
+		return;
 
 	esw_info(esw->dev, "Unload vfs: mode(%s), nvfs(%d), necvfs(%d), active vports(%d)\n",
 		 esw->mode == MLX5_ESWITCH_LEGACY ? "LEGACY" : "OFFLOADS",
@@ -1603,9 +1599,6 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf)
 		esw->esw_funcs.num_vfs = 0;
 	else
 		esw->esw_funcs.num_ec_vfs = 0;
-
-unlock:
-	up_write(&esw->mode_lock);
 }
 
 /* Free resources for corresponding eswitch mode. It is called by devlink
@@ -1647,10 +1640,8 @@ void mlx5_eswitch_disable(struct mlx5_eswitch *esw)
 
 	devl_assert_locked(priv_to_devlink(esw->dev));
 	mlx5_lag_disable_change(esw->dev);
-	down_write(&esw->mode_lock);
 	mlx5_eswitch_disable_locked(esw);
 	esw->mode = MLX5_ESWITCH_LEGACY;
-	up_write(&esw->mode_lock);
 	mlx5_lag_enable_change(esw->dev);
 }
 
@@ -2254,8 +2245,13 @@ bool mlx5_esw_hold(struct mlx5_core_dev *mdev)
 	if (!mlx5_esw_allowed(esw))
 		return true;
 
-	if (down_read_trylock(&esw->mode_lock) != 0)
+	if (down_read_trylock(&esw->mode_lock) != 0) {
+		if (esw->eswitch_operation_in_progress) {
+			up_read(&esw->mode_lock);
+			return false;
+		}
 		return true;
+	}
 
 	return false;
 }
@@ -2312,7 +2308,8 @@ int mlx5_esw_try_lock(struct mlx5_eswitch *esw)
 	if (down_write_trylock(&esw->mode_lock) == 0)
 		return -EINVAL;
 
-	if (atomic64_read(&esw->user_count) > 0) {
+	if (esw->eswitch_operation_in_progress ||
+	    atomic64_read(&esw->user_count) > 0) {
 		up_write(&esw->mode_lock);
 		return -EBUSY;
 	}
@@ -2320,6 +2317,18 @@ int mlx5_esw_try_lock(struct mlx5_eswitch *esw)
 	return esw->mode;
 }
 
+int mlx5_esw_lock(struct mlx5_eswitch *esw)
+{
+	down_write(&esw->mode_lock);
+
+	if (esw->eswitch_operation_in_progress) {
+		up_write(&esw->mode_lock);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
 /**
  * mlx5_esw_unlock() - Release write lock on esw mode lock
  * @esw: eswitch device.

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

@@ -383,6 +383,7 @@ struct mlx5_eswitch {
 	struct xarray paired;
 	struct mlx5_devcom_comp_dev *devcom;
 	u16 enabled_ipsec_vf_count;
+	bool eswitch_operation_in_progress;
 };
 
 void esw_offloads_disable(struct mlx5_eswitch *esw);
@@ -827,6 +828,7 @@ void mlx5_esw_release(struct mlx5_core_dev *dev);
 void mlx5_esw_get(struct mlx5_core_dev *dev);
 void mlx5_esw_put(struct mlx5_core_dev *dev);
 int mlx5_esw_try_lock(struct mlx5_eswitch *esw);
+int mlx5_esw_lock(struct mlx5_eswitch *esw);
 void mlx5_esw_unlock(struct mlx5_eswitch *esw);
 
 void esw_vport_change_handle_locked(struct mlx5_vport *vport);

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

@@ -3653,14 +3653,18 @@ static int esw_inline_mode_to_devlink(u8 mlx5_mode, u8 *mode)
 
 static bool esw_offloads_devlink_ns_eq_netdev_ns(struct devlink *devlink)
 {
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
 	struct net *devl_net, *netdev_net;
-	struct mlx5_eswitch *esw;
-
-	esw = mlx5_devlink_eswitch_nocheck_get(devlink);
-	netdev_net = dev_net(esw->dev->mlx5e_res.uplink_netdev);
-	devl_net = devlink_net(devlink);
+	bool ret = false;
 
-	return net_eq(devl_net, netdev_net);
+	mutex_lock(&dev->mlx5e_res.uplink_netdev_lock);
+	if (dev->mlx5e_res.uplink_netdev) {
+		netdev_net = dev_net(dev->mlx5e_res.uplink_netdev);
+		devl_net = devlink_net(devlink);
+		ret = net_eq(devl_net, netdev_net);
+	}
+	mutex_unlock(&dev->mlx5e_res.uplink_netdev_lock);
+	return ret;
 }
 
 int mlx5_eswitch_block_mode(struct mlx5_core_dev *dev)
@@ -3733,13 +3737,16 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode,
 		goto unlock;
 	}
 
+	esw->eswitch_operation_in_progress = true;
+	up_write(&esw->mode_lock);
+
 	mlx5_eswitch_disable_locked(esw);
 	if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) {
 		if (mlx5_devlink_trap_get_num_active(esw->dev)) {
 			NL_SET_ERR_MSG_MOD(extack,
 					   "Can't change mode while devlink traps are active");
 			err = -EOPNOTSUPP;
-			goto unlock;
+			goto skip;
 		}
 		err = esw_offloads_start(esw, extack);
 	} else if (mode == DEVLINK_ESWITCH_MODE_LEGACY) {
@@ -3749,6 +3756,9 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode,
 		err = -EINVAL;
 	}
 
+skip:
+	down_write(&esw->mode_lock);
+	esw->eswitch_operation_in_progress = false;
 unlock:
 	mlx5_esw_unlock(esw);
 enable_lag:
@@ -3759,16 +3769,12 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode,
 int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode)
 {
 	struct mlx5_eswitch *esw;
-	int err;
 
 	esw = mlx5_devlink_eswitch_get(devlink);
 	if (IS_ERR(esw))
 		return PTR_ERR(esw);
 
-	down_read(&esw->mode_lock);
-	err = esw_mode_to_devlink(esw->mode, mode);
-	up_read(&esw->mode_lock);
-	return err;
+	return esw_mode_to_devlink(esw->mode, mode);
 }
 
 static int mlx5_esw_vports_inline_set(struct mlx5_eswitch *esw, u8 mlx5_mode,
@@ -3862,11 +3868,15 @@ int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode,
 	if (err)
 		goto out;
 
+	esw->eswitch_operation_in_progress = true;
+	up_write(&esw->mode_lock);
+
 	err = mlx5_esw_vports_inline_set(esw, mlx5_mode, extack);
-	if (err)
-		goto out;
+	if (!err)
+		esw->offloads.inline_mode = mlx5_mode;
 
-	esw->offloads.inline_mode = mlx5_mode;
+	down_write(&esw->mode_lock);
+	esw->eswitch_operation_in_progress = false;
 	up_write(&esw->mode_lock);
 	return 0;
 
@@ -3878,16 +3888,12 @@ int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode,
 int mlx5_devlink_eswitch_inline_mode_get(struct devlink *devlink, u8 *mode)
 {
 	struct mlx5_eswitch *esw;
-	int err;
 
 	esw = mlx5_devlink_eswitch_get(devlink);
 	if (IS_ERR(esw))
 		return PTR_ERR(esw);
 
-	down_read(&esw->mode_lock);
-	err = esw_inline_mode_to_devlink(esw->offloads.inline_mode, mode);
-	up_read(&esw->mode_lock);
-	return err;
+	return esw_inline_mode_to_devlink(esw->offloads.inline_mode, mode);
 }
 
 bool mlx5_eswitch_block_encap(struct mlx5_core_dev *dev)
@@ -3969,6 +3975,9 @@ int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink,
 		goto unlock;
 	}
 
+	esw->eswitch_operation_in_progress = true;
+	up_write(&esw->mode_lock);
+
 	esw_destroy_offloads_fdb_tables(esw);
 
 	esw->offloads.encap = encap;
@@ -3982,6 +3991,9 @@ int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink,
 		(void)esw_create_offloads_fdb_tables(esw);
 	}
 
+	down_write(&esw->mode_lock);
+	esw->eswitch_operation_in_progress = false;
+
 unlock:
 	up_write(&esw->mode_lock);
 	return err;
@@ -3996,9 +4008,7 @@ int mlx5_devlink_eswitch_encap_mode_get(struct devlink *devlink,
 	if (IS_ERR(esw))
 		return PTR_ERR(esw);
 
-	down_read(&esw->mode_lock);
 	*encap = esw->offloads.encap;
-	up_read(&esw->mode_lock);
 	return 0;
 }
 

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

@@ -325,6 +325,29 @@ static void mlx5_fw_live_patch_event(struct work_struct *work)
 		mlx5_core_err(dev, "Failed to reload FW tracer\n");
 }
 
+#if IS_ENABLED(CONFIG_HOTPLUG_PCI_PCIE)
+static int mlx5_check_hotplug_interrupt(struct mlx5_core_dev *dev)
+{
+	struct pci_dev *bridge = dev->pdev->bus->self;
+	u16 reg16;
+	int err;
+
+	if (!bridge)
+		return -EOPNOTSUPP;
+
+	err = pcie_capability_read_word(bridge, PCI_EXP_SLTCTL, &reg16);
+	if (err)
+		return err;
+
+	if ((reg16 & PCI_EXP_SLTCTL_HPIE) && (reg16 & PCI_EXP_SLTCTL_DLLSCE)) {
+		mlx5_core_warn(dev, "FW reset is not supported as HotPlug is enabled\n");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+#endif
+
 static int mlx5_check_dev_ids(struct mlx5_core_dev *dev, u16 dev_id)
 {
 	struct pci_bus *bridge_bus = dev->pdev->bus;
@@ -357,6 +380,12 @@ static bool mlx5_is_reset_now_capable(struct mlx5_core_dev *dev)
 		return false;
 	}
 
+#if IS_ENABLED(CONFIG_HOTPLUG_PCI_PCIE)
+	err = mlx5_check_hotplug_interrupt(dev);
+	if (err)
+		return false;
+#endif
+
 	err = pci_read_config_word(dev->pdev, PCI_DEVICE_ID, &dev_id);
 	if (err)
 		return false;

include/linux/mlx5/mlx5_ifc.h

@@ -621,7 +621,7 @@ struct mlx5_ifc_fte_match_set_misc_bits {
 
 	u8         reserved_at_140[0x8];
 	u8         bth_dst_qp[0x18];
-	u8	   reserved_at_160[0x20];
+	u8	   inner_esp_spi[0x20];
 	u8	   outer_esp_spi[0x20];
 	u8         reserved_at_1a0[0x60];
 };
@@ -12001,6 +12001,13 @@ enum {
 	MLX5_IPSEC_ASO_INC_SN            = 0x2,
 };
 
+enum {
+	MLX5_IPSEC_ASO_REPLAY_WIN_32BIT  = 0x0,
+	MLX5_IPSEC_ASO_REPLAY_WIN_64BIT  = 0x1,
+	MLX5_IPSEC_ASO_REPLAY_WIN_128BIT = 0x2,
+	MLX5_IPSEC_ASO_REPLAY_WIN_256BIT = 0x3,
+};
+
 struct mlx5_ifc_ipsec_aso_bits {
 	u8         valid[0x1];
 	u8         reserved_at_201[0x1];