net/mlx5e: Allow concurrent creation of encap entries

Encap entries creation is fully synchronized by encap_tbl_lock. In order to
allow concurrent allocation of hardware resources used to offload
encapsulation, extend mlx5e_encap_entry with 'res_ready' completion. Move
call to mlx5e_tc_tun_create_header_ipv{4|6}() out of encap_tbl_lock
critical section. Modify code that attaches new flows to existing encap to
wait for 'res_ready' completion before using the entry. Insert encap entry
to table before provisioning it to hardware and modify all users of the
encap table to verify that encap was fully initialized by checking
completion result for non-zero value (and to wait for 'res_ready'
completion, if necessary).

Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
index 2e970d0..8ac9672 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
@@ -165,6 +165,8 @@ struct mlx5e_encap_entry {
        char *encap_header;
        int encap_size;
        refcount_t refcnt;
+       struct completion res_ready;
+       int compl_result;
 };
 
 struct mlx5e_rep_sq {

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index c13db9b..5be3da6 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2904,8 +2904,18 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
        e = mlx5e_encap_get(priv, &key, hash_key);
 
        /* must verify if encap is valid or not */
-       if (e)
+       if (e) {
+               mutex_unlock(&esw->offloads.encap_tbl_lock);
+               wait_for_completion(&e->res_ready);
+
+               /* Protect against concurrent neigh update. */
+               mutex_lock(&esw->offloads.encap_tbl_lock);
+               if (e->compl_result) {
+                       err = -EREMOTEIO;
+                       goto out_err;
+               }
                goto attach_flow;
+       }
 
        e = kzalloc(sizeof(*e), GFP_KERNEL);
        if (!e) {
@@ -2914,22 +2924,32 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
        }
 
        refcount_set(&e->refcnt, 1);
+       init_completion(&e->res_ready);
+
        e->tun_info = tun_info;
        err = mlx5e_tc_tun_init_encap_attr(mirred_dev, priv, e, extack);
-       if (err)
+       if (err) {
+               kfree(e);
+               e = NULL;
                goto out_err;
+       }
 
        INIT_LIST_HEAD(&e->flows);
+       hash_add_rcu(esw->offloads.encap_tbl, &e->encap_hlist, hash_key);
+       mutex_unlock(&esw->offloads.encap_tbl_lock);
 
        if (family == AF_INET)
                err = mlx5e_tc_tun_create_header_ipv4(priv, mirred_dev, e);
        else if (family == AF_INET6)
                err = mlx5e_tc_tun_create_header_ipv6(priv, mirred_dev, e);
 
-       if (err)
+       /* Protect against concurrent neigh update. */
+       mutex_lock(&esw->offloads.encap_tbl_lock);
+       complete_all(&e->res_ready);
+       if (err) {
+               e->compl_result = err;
                goto out_err;
-
-       hash_add_rcu(esw->offloads.encap_tbl, &e->encap_hlist, hash_key);
+       }
 
 attach_flow:
        flow->encaps[out_index].e = e;
@@ -2949,7 +2969,8 @@ attach_flow:
 
 out_err:
        mutex_unlock(&esw->offloads.encap_tbl_lock);
-       kfree(e);
+       if (e)
+               mlx5e_encap_put(priv, e);
        return err;
 }