net/mlx5e: Use dedicated uplink vport netdev representor

Currently, when running in sriov switchdev mode, we are using the PF
netdevice as the uplink representor, this is problematic from few aspects:

- will break when the PF isn't eswitch manager (e.g smart NIC env)
- misalignment with other NIC switchdev drivers
- makes us have and maintain special code, hurts the driver quality/robustness
- which in turn opens the door for future bugs

As of each and all of the above, we move to have a dedicated netdev representor
for the uplink vport in a similar manner done for for the VF vports.

This includes the following:

1. have an uplink rep netdev as we have for VF reps
2. all reps use same load/unload functions
3. HW stats for uplink based on physical port counters and not vport counters
4. link state for the uplink managed through PAOS and not vport state
5. the uplink rep has sysfs link to the PF PCI function && uses the PF MAC address

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

1 files changed, 22 insertions, 17 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 53ebb5a48018..c1a9120412b8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -69,13 +69,13 @@ struct mlx5_nic_flow_attr {
 enum {
 	MLX5E_TC_FLOW_INGRESS	= MLX5E_TC_INGRESS,
 	MLX5E_TC_FLOW_EGRESS	= MLX5E_TC_EGRESS,
-	MLX5E_TC_FLOW_ESWITCH	= BIT(MLX5E_TC_FLOW_BASE),
-	MLX5E_TC_FLOW_NIC	= BIT(MLX5E_TC_FLOW_BASE + 1),
-	MLX5E_TC_FLOW_OFFLOADED	= BIT(MLX5E_TC_FLOW_BASE + 2),
-	MLX5E_TC_FLOW_HAIRPIN	= BIT(MLX5E_TC_FLOW_BASE + 3),
-	MLX5E_TC_FLOW_HAIRPIN_RSS = BIT(MLX5E_TC_FLOW_BASE + 4),
-	MLX5E_TC_FLOW_SLOW	  = BIT(MLX5E_TC_FLOW_BASE + 5),
-	MLX5E_TC_FLOW_DUP         = BIT(MLX5E_TC_FLOW_BASE + 6),
+	MLX5E_TC_FLOW_ESWITCH	= MLX5E_TC_ESW_OFFLOAD,
+	MLX5E_TC_FLOW_NIC	= MLX5E_TC_NIC_OFFLOAD,
+	MLX5E_TC_FLOW_OFFLOADED	= BIT(MLX5E_TC_FLOW_BASE),
+	MLX5E_TC_FLOW_HAIRPIN	= BIT(MLX5E_TC_FLOW_BASE + 1),
+	MLX5E_TC_FLOW_HAIRPIN_RSS = BIT(MLX5E_TC_FLOW_BASE + 2),
+	MLX5E_TC_FLOW_SLOW	  = BIT(MLX5E_TC_FLOW_BASE + 3),
+	MLX5E_TC_FLOW_DUP         = BIT(MLX5E_TC_FLOW_BASE + 4),
 };
 
 #define MLX5E_TC_MAX_SPLITS 1
@@ -834,7 +834,7 @@ static void mlx5e_tc_del_nic_flow(struct mlx5e_priv *priv,
 	mlx5_del_flow_rules(flow->rule[0]);
 	mlx5_fc_destroy(priv->mdev, counter);
 
-	if (!mlx5e_tc_num_filters(priv) && priv->fs.tc.t) {
+	if (!mlx5e_tc_num_filters(priv, MLX5E_TC_NIC_OFFLOAD)  && priv->fs.tc.t) {
 		mlx5_destroy_flow_table(priv->fs.tc.t);
 		priv->fs.tc.t = NULL;
 	}
@@ -2686,6 +2686,11 @@ static void get_flags(int flags, u16 *flow_flags)
 	if (flags & MLX5E_TC_EGRESS)
 		__flow_flags |= MLX5E_TC_FLOW_EGRESS;
 
+	if (flags & MLX5E_TC_ESW_OFFLOAD)
+		__flow_flags |= MLX5E_TC_FLOW_ESWITCH;
+	if (flags & MLX5E_TC_NIC_OFFLOAD)
+		__flow_flags |= MLX5E_TC_FLOW_NIC;
+
 	*flow_flags = __flow_flags;
 }
 
@@ -2696,15 +2701,15 @@ static const struct rhashtable_params tc_ht_params = {
 	.automatic_shrinking = true,
 };
 
-static struct rhashtable *get_tc_ht(struct mlx5e_priv *priv)
+static struct rhashtable *get_tc_ht(struct mlx5e_priv *priv, int flags)
 {
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct mlx5e_rep_priv *uplink_rpriv;
 
-	if (MLX5_VPORT_MANAGER(priv->mdev) && esw->mode == SRIOV_OFFLOADS) {
+	if (flags & MLX5E_TC_ESW_OFFLOAD) {
 		uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH);
 		return &uplink_rpriv->uplink_priv.tc_ht;
-	} else
+	} else /* NIC offload */
 		return &priv->fs.tc.ht;
 }
 
@@ -2973,7 +2978,7 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv,
 			   struct tc_cls_flower_offload *f, int flags)
 {
 	struct netlink_ext_ack *extack = f->common.extack;
-	struct rhashtable *tc_ht = get_tc_ht(priv);
+	struct rhashtable *tc_ht = get_tc_ht(priv, flags);
 	struct mlx5e_tc_flow *flow;
 	int err = 0;
 
@@ -3018,7 +3023,7 @@ static bool same_flow_direction(struct mlx5e_tc_flow *flow, int flags)
 int mlx5e_delete_flower(struct net_device *dev, struct mlx5e_priv *priv,
 			struct tc_cls_flower_offload *f, int flags)
 {
-	struct rhashtable *tc_ht = get_tc_ht(priv);
+	struct rhashtable *tc_ht = get_tc_ht(priv, flags);
 	struct mlx5e_tc_flow *flow;
 
 	flow = rhashtable_lookup_fast(tc_ht, &f->cookie, tc_ht_params);
@@ -3038,7 +3043,7 @@ int mlx5e_stats_flower(struct net_device *dev, struct mlx5e_priv *priv,
 		       struct tc_cls_flower_offload *f, int flags)
 {
 	struct mlx5_devcom *devcom = priv->mdev->priv.devcom;
-	struct rhashtable *tc_ht = get_tc_ht(priv);
+	struct rhashtable *tc_ht = get_tc_ht(priv, flags);
 	struct mlx5_eswitch *peer_esw;
 	struct mlx5e_tc_flow *flow;
 	struct mlx5_fc *counter;
@@ -3168,7 +3173,7 @@ void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv)
 	if (tc->netdevice_nb.notifier_call)
 		unregister_netdevice_notifier(&tc->netdevice_nb);
 
-	rhashtable_free_and_destroy(&tc->ht, _mlx5e_tc_del_flow, NULL);
+	rhashtable_destroy(&tc->ht);
 
 	if (!IS_ERR_OR_NULL(tc->t)) {
 		mlx5_destroy_flow_table(tc->t);
@@ -3186,9 +3191,9 @@ void mlx5e_tc_esw_cleanup(struct rhashtable *tc_ht)
 	rhashtable_free_and_destroy(tc_ht, _mlx5e_tc_del_flow, NULL);
 }
 
-int mlx5e_tc_num_filters(struct mlx5e_priv *priv)
+int mlx5e_tc_num_filters(struct mlx5e_priv *priv, int flags)
 {
-	struct rhashtable *tc_ht = get_tc_ht(priv);
+	struct rhashtable *tc_ht = get_tc_ht(priv, flags);
 
 	return atomic_read(&tc_ht->nelems);
 }