net/mlx5: E-Switch, Use two priorities for SRIOV offloads mode

In the offloads mode, some slow path rules are added by the driver (e.g
send-to-vport), while offloaded rules are to be added from upper layers.

The slow path rules have lower priority and we don't want matching on
offloaded rules to suffer from extra steering hops related to the slow
path rules.

We use two priorities, one for offloaded rules (fast path), and one for
the control rules (slow path). To allow for that, we enable two priorities
for the FDB namespace in the FS core code.

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

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

@@ -145,6 +145,7 @@ struct mlx5_eswitch_fdb {
 		} legacy;
 
 		struct offloads_fdb {
+			struct mlx5_flow_table *fdb;
 			struct mlx5_flow_group *send_to_vport_grp;
 			struct mlx5_flow_group *miss_grp;
 			struct mlx5_flow_rule  *miss_rule;

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

@@ -38,6 +38,11 @@
 #include "mlx5_core.h"
 #include "eswitch.h"
 
+enum {
+	FDB_FAST_PATH = 0,
+	FDB_SLOW_PATH
+};
+
 static struct mlx5_flow_rule *
 mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn)
 {
@@ -149,7 +154,7 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw)
 	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
 	dest.vport_num = 0;
 
-	flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec,
+	flow_rule = mlx5_add_flow_rule(esw->fdb_table.offloads.fdb, spec,
 				       MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
 				       0, &dest);
 	if (IS_ERR(flow_rule)) {
@@ -165,6 +170,8 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw)
 }
 
 #define MAX_PF_SQ 256
+#define ESW_OFFLOADS_NUM_ENTRIES (1 << 13) /* 8K */
+#define ESW_OFFLOADS_NUM_GROUPS  4
 
 static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports)
 {
@@ -190,15 +197,25 @@ static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports)
 	esw_debug(dev, "Create offloads FDB table, log_max_size(%d)\n",
 		  MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
 
-	table_size = nvports + MAX_PF_SQ + 1;
-	fdb = mlx5_create_flow_table(root_ns, 0, table_size, 0);
+	fdb = mlx5_create_auto_grouped_flow_table(root_ns, FDB_FAST_PATH,
+						  ESW_OFFLOADS_NUM_ENTRIES,
+						  ESW_OFFLOADS_NUM_GROUPS, 0);
 	if (IS_ERR(fdb)) {
 		err = PTR_ERR(fdb);
-		esw_warn(dev, "Failed to create FDB Table err %d\n", err);
-		goto fdb_err;
+		esw_warn(dev, "Failed to create Fast path FDB Table err %d\n", err);
+		goto fast_fdb_err;
 	}
 	esw->fdb_table.fdb = fdb;
 
+	table_size = nvports + MAX_PF_SQ + 1;
+	fdb = mlx5_create_flow_table(root_ns, FDB_SLOW_PATH, table_size, 0);
+	if (IS_ERR(fdb)) {
+		err = PTR_ERR(fdb);
+		esw_warn(dev, "Failed to create slow path FDB Table err %d\n", err);
+		goto slow_fdb_err;
+	}
+	esw->fdb_table.offloads.fdb = fdb;
+
 	/* create send-to-vport group */
 	memset(flow_group_in, 0, inlen);
 	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
@@ -247,8 +264,10 @@ static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports)
 miss_err:
 	mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp);
 send_vport_err:
-	mlx5_destroy_flow_table(fdb);
-fdb_err:
+	mlx5_destroy_flow_table(esw->fdb_table.offloads.fdb);
+slow_fdb_err:
+	mlx5_destroy_flow_table(esw->fdb_table.fdb);
+fast_fdb_err:
 ns_err:
 	kvfree(flow_group_in);
 	return err;
@@ -264,6 +283,7 @@ static void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw)
 	mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp);
 	mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp);
 
+	mlx5_destroy_flow_table(esw->fdb_table.offloads.fdb);
 	mlx5_destroy_flow_table(esw->fdb_table.fdb);
 }
 

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

@@ -1712,15 +1712,21 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
 	if (!steering->fdb_root_ns)
 		return -ENOMEM;
 
-	/* Create single prio */
 	prio = fs_create_prio(&steering->fdb_root_ns->ns, 0, 1);
-	if (IS_ERR(prio)) {
-		cleanup_root_ns(steering->fdb_root_ns);
-		steering->fdb_root_ns = NULL;
-		return PTR_ERR(prio);
-	} else {
-		return 0;
-	}
+	if (IS_ERR(prio))
+		goto out_err;
+
+	prio = fs_create_prio(&steering->fdb_root_ns->ns, 1, 1);
+	if (IS_ERR(prio))
+		goto out_err;
+
+	set_prio_attrs(steering->fdb_root_ns);
+	return 0;
+
+out_err:
+	cleanup_root_ns(steering->fdb_root_ns);
+	steering->fdb_root_ns = NULL;
+	return PTR_ERR(prio);
 }
 
 static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering)