Merge tag 'mlx5-updates-2023-05-19' of git://git.kernel.org/pub/scm/l…

…inux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
mlx5-updates-2023-05-19

mlx5 misc changes and code clean up:

The following series contains general changes for improving
E-Switch driver behavior.

1) improving condition checking
2) Code clean up
3) Using metadata matching on send-to-vport rules.
4) Using RoCE v2 instead of v1 for loopback rules.

* tag 'mlx5-updates-2023-05-19' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
  net/mlx5e: E-Switch, Initialize E-Switch for eswitch manager
  net/mlx5: devlink, Only show PF related devlink warning when needed
  net/mlx5: E-Switch, Use metadata matching for RoCE loopback rule
  net/mlx5: E-Switch, Use RoCE version 2 for loopback traffic
  net/mlx5e: E-Switch, Add a check that log_max_l2_table is valid
  net/mlx5e: E-Switch: move debug print of adding mac to correct place
  net/mlx5e: E-Switch, Check device is PF when stopping esw offloads
  net/mlx5: Remove redundant vport_group_manager cap check
  net/mlx5e: E-Switch, Use metadata for vport matching in send-to-vport rules
  net/mlx5e: E-Switch, Allow get vport api if esw exists
  net/mlx5e: E-Switch, Update when to set other vport context
  net/mlx5e: Remove redundant __func__ arg from fs_err() calls
  net/mlx5e: E-Switch, Remove flow_source check for metadata matching
  net/mlx5: E-Switch, Remove redundant check
  net/mlx5: Remove redundant esw multiport validate function
====================

Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230519175557.15683-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

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

@@ -162,9 +162,8 @@ static int mlx5_devlink_reload_down(struct devlink *devlink, bool netns_change,
 		return -EOPNOTSUPP;
 	}
 
-	if (pci_num_vf(pdev)) {
+	if (mlx5_core_is_pf(dev) && pci_num_vf(pdev))
 		NL_SET_ERR_MSG_MOD(extack, "reload while VFs are present is unfavorable");
-	}
 
 	switch (action) {
 	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
@@ -464,27 +463,6 @@ static int mlx5_devlink_esw_multiport_get(struct devlink *devlink, u32 id,
 	ctx->val.vbool = mlx5_lag_is_mpesw(dev);
 	return 0;
 }
-
-static int mlx5_devlink_esw_multiport_validate(struct devlink *devlink, u32 id,
-					       union devlink_param_value val,
-					       struct netlink_ext_ack *extack)
-{
-	struct mlx5_core_dev *dev = devlink_priv(devlink);
-
-	if (!MLX5_ESWITCH_MANAGER(dev)) {
-		NL_SET_ERR_MSG_MOD(extack, "E-Switch is unsupported");
-		return -EOPNOTSUPP;
-	}
-
-	if (mlx5_eswitch_mode(dev) != MLX5_ESWITCH_OFFLOADS) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "E-Switch must be in switchdev mode");
-		return -EBUSY;
-	}
-
-	return 0;
-}
-
 #endif
 
 static int mlx5_devlink_eq_depth_validate(struct devlink *devlink, u32 id,
@@ -563,7 +541,7 @@ static const struct devlink_param mlx5_devlink_params[] = {
 			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
 			     mlx5_devlink_esw_multiport_get,
 			     mlx5_devlink_esw_multiport_set,
-			     mlx5_devlink_esw_multiport_validate),
+			     NULL),
 #endif
 	DEVLINK_PARAM_GENERIC(IO_EQ_SIZE, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
 			      NULL, NULL, mlx5_devlink_eq_depth_validate),

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

@@ -283,7 +283,7 @@ static int __mlx5e_add_vlan_rule(struct mlx5e_flow_steering *fs,
 	if (IS_ERR(*rule_p)) {
 		err = PTR_ERR(*rule_p);
 		*rule_p = NULL;
-		fs_err(fs, "%s: add rule failed\n", __func__);
+		fs_err(fs, "add rule failed\n");
 	}
 
 	return err;
@@ -395,8 +395,7 @@ int mlx5e_add_vlan_trap(struct mlx5e_flow_steering *fs, int trap_id, int tir_num
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
 		fs->vlan->trap_rule = NULL;
-		fs_err(fs, "%s: add VLAN trap rule failed, err %d\n",
-		       __func__, err);
+		fs_err(fs, "add VLAN trap rule failed, err %d\n", err);
 		return err;
 	}
 	fs->vlan->trap_rule = rule;
@@ -421,8 +420,7 @@ int mlx5e_add_mac_trap(struct mlx5e_flow_steering *fs, int trap_id, int tir_num)
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
 		fs->l2.trap_rule = NULL;
-		fs_err(fs, "%s: add MAC trap rule failed, err %d\n",
-		       __func__, err);
+		fs_err(fs, "add MAC trap rule failed, err %d\n", err);
 		return err;
 	}
 	fs->l2.trap_rule = rule;
@@ -763,7 +761,7 @@ static int mlx5e_add_promisc_rule(struct mlx5e_flow_steering *fs)
 	if (IS_ERR(*rule_p)) {
 		err = PTR_ERR(*rule_p);
 		*rule_p = NULL;
-		fs_err(fs, "%s: add promiscuous rule failed\n", __func__);
+		fs_err(fs, "add promiscuous rule failed\n");
 	}
 	kvfree(spec);
 	return err;
@@ -995,7 +993,7 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_flow_steering *fs,
 
 	ai->rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
 	if (IS_ERR(ai->rule)) {
-		fs_err(fs, "%s: add l2 rule(mac:%pM) failed\n", __func__, mv_dmac);
+		fs_err(fs, "add l2 rule(mac:%pM) failed\n", mv_dmac);
 		err = PTR_ERR(ai->rule);
 		ai->rule = NULL;
 	}

drivers/net/ethernet/mellanox/mlx5/core/esw/acl/helper.c

@@ -35,7 +35,8 @@ esw_acl_table_create(struct mlx5_eswitch *esw, struct mlx5_vport *vport, int ns,
 	}
 
 	ft_attr.max_fte = size;
-	ft_attr.flags = MLX5_FLOW_TABLE_OTHER_VPORT;
+	if (vport_num || mlx5_core_is_ecpf(esw->dev))
+		ft_attr.flags = MLX5_FLOW_TABLE_OTHER_VPORT;
 	acl = mlx5_create_vport_flow_table(root_ns, &ft_attr, vport_num);
 	if (IS_ERR(acl)) {
 		err = PTR_ERR(acl);

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

@@ -92,7 +92,7 @@ mlx5_eswitch_get_vport(struct mlx5_eswitch *esw, u16 vport_num)
 {
 	struct mlx5_vport *vport;
 
-	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager))
+	if (!esw)
 		return ERR_PTR(-EPERM);
 
 	vport = xa_load(&esw->vports, vport_num);
@@ -113,7 +113,8 @@ static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport,
 		 opcode, MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT);
 	MLX5_SET(modify_nic_vport_context_in, in, field_select.change_event, 1);
 	MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport);
-	MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1);
+	if (vport || mlx5_core_is_ecpf(dev))
+		MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1);
 	nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in,
 				     in, nic_vport_context);
 
@@ -309,11 +310,12 @@ static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
 
 fdb_add:
 	/* SRIOV is enabled: Forward UC MAC to vport */
-	if (esw->fdb_table.legacy.fdb && esw->mode == MLX5_ESWITCH_LEGACY)
+	if (esw->fdb_table.legacy.fdb && esw->mode == MLX5_ESWITCH_LEGACY) {
 		vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport);
 
-	esw_debug(esw->dev, "\tADDED UC MAC: vport[%d] %pM fr(%p)\n",
-		  vport, mac, vaddr->flow_rule);
+		esw_debug(esw->dev, "\tADDED UC MAC: vport[%d] %pM fr(%p)\n",
+			  vport, mac, vaddr->flow_rule);
+	}
 
 	return 0;
 }
@@ -710,6 +712,9 @@ void esw_vport_change_handle_locked(struct mlx5_vport *vport)
 	struct mlx5_eswitch *esw = dev->priv.eswitch;
 	u8 mac[ETH_ALEN];
 
+	if (!MLX5_CAP_GEN(dev, log_max_l2_table))
+		return;
+
 	mlx5_query_nic_vport_mac_address(dev, vport->vport, true, mac);
 	esw_debug(dev, "vport[%d] Context Changed: perm mac: %pM\n",
 		  vport->vport, mac);
@@ -946,7 +951,8 @@ void mlx5_esw_vport_disable(struct mlx5_eswitch *esw, u16 vport_num)
 	vport->enabled = false;
 
 	/* Disable events from this vport */
-	arm_vport_context_events_cmd(esw->dev, vport->vport, 0);
+	if (MLX5_CAP_GEN(esw->dev, log_max_l2_table))
+		arm_vport_context_events_cmd(esw->dev, vport->vport, 0);
 
 	if (!mlx5_esw_is_manager_vport(esw, vport->vport) &&
 	    MLX5_CAP_GEN(esw->dev, vhca_resource_manager))
@@ -1616,7 +1622,7 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
 	struct mlx5_eswitch *esw;
 	int err;
 
-	if (!MLX5_VPORT_MANAGER(dev))
+	if (!MLX5_VPORT_MANAGER(dev) && !MLX5_ESWITCH_MANAGER(dev))
 		return 0;
 
 	esw = kzalloc(sizeof(*esw), GFP_KERNEL);
@@ -1686,7 +1692,7 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
 
 void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
 {
-	if (!esw || !MLX5_VPORT_MANAGER(esw->dev))
+	if (!esw)
 		return;
 
 	esw_info(esw->dev, "cleanup\n");

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

@@ -683,6 +683,14 @@ mlx5_esw_vporttbl_put(struct mlx5_eswitch *esw, struct mlx5_vport_tbl_attr *attr
 struct mlx5_flow_handle *
 esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag);
 
+void mlx5_esw_set_flow_group_source_port(struct mlx5_eswitch *esw,
+					 u32 *flow_group_in,
+					 int match_params);
+
+void mlx5_esw_set_spec_source_port(struct mlx5_eswitch *esw,
+				   u16 vport,
+				   struct mlx5_flow_spec *spec);
+
 int esw_offloads_load_rep(struct mlx5_eswitch *esw, u16 vport_num);
 void esw_offloads_unload_rep(struct mlx5_eswitch *esw, u16 vport_num);
 

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

@@ -838,6 +838,7 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *on_esw,
 	struct mlx5_flow_handle *flow_rule;
 	struct mlx5_flow_spec *spec;
 	void *misc;
+	u16 vport;
 
 	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
 	if (!spec) {
@@ -847,20 +848,43 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *on_esw,
 
 	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
 	MLX5_SET(fte_match_set_misc, misc, source_sqn, sqn);
-	/* source vport is the esw manager */
-	MLX5_SET(fte_match_set_misc, misc, source_port, from_esw->manager_vport);
-	if (MLX5_CAP_ESW(on_esw->dev, merged_eswitch))
-		MLX5_SET(fte_match_set_misc, misc, source_eswitch_owner_vhca_id,
-			 MLX5_CAP_GEN(from_esw->dev, vhca_id));
 
 	misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
 	MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_sqn);
-	MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);
-	if (MLX5_CAP_ESW(on_esw->dev, merged_eswitch))
-		MLX5_SET_TO_ONES(fte_match_set_misc, misc,
-				 source_eswitch_owner_vhca_id);
 
 	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;
+
+	/* source vport is the esw manager */
+	vport = from_esw->manager_vport;
+
+	if (mlx5_eswitch_vport_match_metadata_enabled(on_esw)) {
+		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_2);
+		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
+			 mlx5_eswitch_get_vport_metadata_for_match(from_esw, vport));
+
+		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2);
+		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
+			 mlx5_eswitch_get_vport_metadata_mask());
+
+		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2;
+	} else {
+		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
+		MLX5_SET(fte_match_set_misc, misc, source_port, vport);
+
+		if (MLX5_CAP_ESW(on_esw->dev, merged_eswitch))
+			MLX5_SET(fte_match_set_misc, misc, source_eswitch_owner_vhca_id,
+				 MLX5_CAP_GEN(from_esw->dev, vhca_id));
+
+		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
+		MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);
+
+		if (MLX5_CAP_ESW(on_esw->dev, merged_eswitch))
+			MLX5_SET_TO_ONES(fte_match_set_misc, misc,
+					 source_eswitch_owner_vhca_id);
+
+		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS;
+	}
+
 	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
 	dest.vport.num = rep->vport;
 	dest.vport.vhca_id = MLX5_CAP_GEN(rep->esw->dev, vhca_id);
@@ -1269,8 +1293,10 @@ esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag)
 #define MAX_PF_SQ 256
 #define MAX_SQ_NVPORTS 32
 
-static void esw_set_flow_group_source_port(struct mlx5_eswitch *esw,
-					   u32 *flow_group_in)
+void
+mlx5_esw_set_flow_group_source_port(struct mlx5_eswitch *esw,
+				    u32 *flow_group_in,
+				    int match_params)
 {
 	void *match_criteria = MLX5_ADDR_OF(create_flow_group_in,
 					    flow_group_in,
@@ -1279,15 +1305,15 @@ static void esw_set_flow_group_source_port(struct mlx5_eswitch *esw,
 	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
 		MLX5_SET(create_flow_group_in, flow_group_in,
 			 match_criteria_enable,
-			 MLX5_MATCH_MISC_PARAMETERS_2);
+			 MLX5_MATCH_MISC_PARAMETERS_2 | match_params);
 
 		MLX5_SET(fte_match_param, match_criteria,
 			 misc_parameters_2.metadata_reg_c_0,
 			 mlx5_eswitch_get_vport_metadata_mask());
 	} else {
 		MLX5_SET(create_flow_group_in, flow_group_in,
 			 match_criteria_enable,
-			 MLX5_MATCH_MISC_PARAMETERS);
+			 MLX5_MATCH_MISC_PARAMETERS | match_params);
 
 		MLX5_SET_TO_ONES(fte_match_param, match_criteria,
 				 misc_parameters.source_port);
@@ -1463,14 +1489,13 @@ esw_create_send_to_vport_group(struct mlx5_eswitch *esw,
 
 	memset(flow_group_in, 0, inlen);
 
-	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
-		 MLX5_MATCH_MISC_PARAMETERS);
+	mlx5_esw_set_flow_group_source_port(esw, flow_group_in, MLX5_MATCH_MISC_PARAMETERS);
 
 	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
-
 	MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_sqn);
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_port);
-	if (MLX5_CAP_ESW(esw->dev, merged_eswitch)) {
+
+	if (!mlx5_eswitch_vport_match_metadata_enabled(esw) &&
+	    MLX5_CAP_ESW(esw->dev, merged_eswitch)) {
 		MLX5_SET_TO_ONES(fte_match_param, match_criteria,
 				 misc_parameters.source_eswitch_owner_vhca_id);
 		MLX5_SET(create_flow_group_in, flow_group_in,
@@ -1558,7 +1583,7 @@ esw_create_peer_esw_miss_group(struct mlx5_eswitch *esw,
 
 	memset(flow_group_in, 0, inlen);
 
-	esw_set_flow_group_source_port(esw, flow_group_in);
+	mlx5_esw_set_flow_group_source_port(esw, flow_group_in, 0);
 
 	if (!mlx5_eswitch_vport_match_metadata_enabled(esw)) {
 		match_criteria = MLX5_ADDR_OF(create_flow_group_in,
@@ -1845,7 +1870,7 @@ static int esw_create_vport_rx_group(struct mlx5_eswitch *esw)
 		return -ENOMEM;
 
 	/* create vport rx group */
-	esw_set_flow_group_source_port(esw, flow_group_in);
+	mlx5_esw_set_flow_group_source_port(esw, flow_group_in, 0);
 
 	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
 	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, nvports - 1);
@@ -1915,21 +1940,13 @@ static void esw_destroy_vport_rx_drop_group(struct mlx5_eswitch *esw)
 		mlx5_destroy_flow_group(esw->offloads.vport_rx_drop_group);
 }
 
-struct mlx5_flow_handle *
-mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport,
-				  struct mlx5_flow_destination *dest)
+void
+mlx5_esw_set_spec_source_port(struct mlx5_eswitch *esw,
+			      u16 vport,
+			      struct mlx5_flow_spec *spec)
 {
-	struct mlx5_flow_act flow_act = {0};
-	struct mlx5_flow_handle *flow_rule;
-	struct mlx5_flow_spec *spec;
 	void *misc;
 
-	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
-	if (!spec) {
-		flow_rule = ERR_PTR(-ENOMEM);
-		goto out;
-	}
-
 	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
 		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_2);
 		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
@@ -1949,6 +1966,23 @@ mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport,
 
 		spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;
 	}
+}
+
+struct mlx5_flow_handle *
+mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport,
+				  struct mlx5_flow_destination *dest)
+{
+	struct mlx5_flow_act flow_act = {0};
+	struct mlx5_flow_handle *flow_rule;
+	struct mlx5_flow_spec *spec;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec) {
+		flow_rule = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+
+	mlx5_esw_set_spec_source_port(esw, vport, spec);
 
 	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	flow_rule = mlx5_add_flow_rules(esw->offloads.ft_offloads, spec,
@@ -2827,9 +2861,6 @@ bool mlx5_esw_vport_match_metadata_supported(const struct mlx5_eswitch *esw)
 	      MLX5_FDB_TO_VPORT_REG_C_0))
 		return false;
 
-	if (!MLX5_CAP_ESW_FLOWTABLE(esw->dev, flow_source))
-		return false;
-
 	return true;
 }
 
@@ -3280,7 +3311,7 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw,
 	/* If changing from switchdev to legacy mode without sriov enabled,
 	 * no need to create legacy fdb.
 	 */
-	if (!mlx5_sriov_is_enabled(esw->dev))
+	if (!mlx5_core_is_pf(esw->dev) || !mlx5_sriov_is_enabled(esw->dev))
 		return 0;
 
 	err = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS);