net/mlx5e: Support tc block sharing for representors

Currently offloading a rule over a tc block shared by multiple
representors fails because an e-switch global hashtable to keep
the mapping from tc cookies to mlx5e flow instances is used, and
tc block sharing offloads the same rule/cookie multiple times,
each time for different representor sharing the tc block.

Changing the implementation and behavior by acknowledging and returning
success if the same rule/cookie is offloaded again to other slave
representor sharing the tc block by setting, checking and comparing
the netdev that added the rule first.

Signed-off-by: Vu Pham <vuhuong@mellanox.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 571da14809fec4601828ebeb467a1e63e8b97936..f3e65a15c950145e6d725fc6bd62b6553d8269de 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -145,6 +145,7 @@ struct mlx5e_tc_flow {
        struct list_head        hairpin; /* flows sharing the same hairpin */
        struct list_head        peer;    /* flows with peer flow */
        struct list_head        unready; /* flows not ready to be offloaded (e.g due to missing route) */
+       struct net_device       *orig_dev; /* netdev adding flow first */
        int                     tmp_efi_index;
        struct list_head        tmp_list; /* temporary flow list used by neigh update */
        refcount_t              refcnt;
@@ -4624,11 +4625,21 @@ mlx5e_tc_add_flow(struct mlx5e_priv *priv,
        return err;
 }
 
+static bool is_flow_rule_duplicate_allowed(struct net_device *dev,
+                                          struct mlx5e_rep_priv *rpriv)
+{
+       /* Offloaded flow rule is allowed to duplicate on non-uplink representor
+        * sharing tc block with other slaves of a lag device.
+        */
+       return netif_is_lag_port(dev) && rpriv->rep->vport != MLX5_VPORT_UPLINK;
+}
+
 int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv,
                           struct flow_cls_offload *f, unsigned long flags)
 {
        struct netlink_ext_ack *extack = f->common.extack;
        struct rhashtable *tc_ht = get_tc_ht(priv, flags);
+       struct mlx5e_rep_priv *rpriv = priv->ppriv;
        struct mlx5e_tc_flow *flow;
        int err = 0;
 
@@ -4636,6 +4647,12 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv,
        flow = rhashtable_lookup(tc_ht, &f->cookie, tc_ht_params);
        rcu_read_unlock();
        if (flow) {
+               /* Same flow rule offloaded to non-uplink representor sharing tc block,
+                * just return 0.
+                */
+               if (is_flow_rule_duplicate_allowed(dev, rpriv) && flow->orig_dev != dev)
+                       goto out;
+
                NL_SET_ERR_MSG_MOD(extack,
                                   "flow cookie already exists, ignoring");
                netdev_warn_once(priv->netdev,
@@ -4650,6 +4667,12 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv,
        if (err)
                goto out;
 
+       /* Flow rule offloaded to non-uplink representor sharing tc block,
+        * set the flow's owner dev.
+        */
+       if (is_flow_rule_duplicate_allowed(dev, rpriv))
+               flow->orig_dev = dev;
+
        err = rhashtable_lookup_insert_fast(tc_ht, &flow->node, tc_ht_params);
        if (err)
                goto err_free;