Commit 948993f2 authored by Vlad Buslov's avatar Vlad Buslov Committed by Saeed Mahameed

net/mlx5e: Extend encap entry with reference counter

List of flows attached to encap entry is used as implicit reference
counter (encap entry is deallocated when list becomes free) and as a
mechanism to obtain encap entry that flow is attached to (through list
head). This is not safe when concurrent modification of list of flows
attached to encap entry is possible. Proper atomic reference counter is
required to support concurrent access.

As a preparation for extending encap with reference counting, extract code
that lookups and deletes encap entry into standalone put/get helpers. In
order to remove this dependency on external locking, extend encap entry
with reference counter to manage its lifetime and extend flow structure
with direct pointer to encap entry that flow is attached to.
Signed-off-by: default avatarVlad Buslov <vladbu@mellanox.com>
Reviewed-by: default avatarJianbo Liu <jianbol@mellanox.com>
Reviewed-by: default avatarRoi Dayan <roid@mellanox.com>
Signed-off-by: default avatarSaeed Mahameed <saeedm@mellanox.com>
parent a734d007
...@@ -613,12 +613,17 @@ static void mlx5e_rep_neigh_update(struct work_struct *work) ...@@ -613,12 +613,17 @@ static void mlx5e_rep_neigh_update(struct work_struct *work)
neigh_connected = (nud_state & NUD_VALID) && !dead; neigh_connected = (nud_state & NUD_VALID) && !dead;
list_for_each_entry(e, &nhe->encap_list, encap_list) { list_for_each_entry(e, &nhe->encap_list, encap_list) {
if (!mlx5e_encap_take(e))
continue;
encap_connected = !!(e->flags & MLX5_ENCAP_ENTRY_VALID); encap_connected = !!(e->flags & MLX5_ENCAP_ENTRY_VALID);
priv = netdev_priv(e->out_dev); priv = netdev_priv(e->out_dev);
if (encap_connected != neigh_connected || if (encap_connected != neigh_connected ||
!ether_addr_equal(e->h_dest, ha)) !ether_addr_equal(e->h_dest, ha))
mlx5e_rep_update_flows(priv, e, neigh_connected, ha); mlx5e_rep_update_flows(priv, e, neigh_connected, ha);
mlx5e_encap_put(priv, e);
} }
mlx5e_rep_neigh_entry_release(nhe); mlx5e_rep_neigh_entry_release(nhe);
rtnl_unlock(); rtnl_unlock();
......
...@@ -164,6 +164,7 @@ struct mlx5e_encap_entry { ...@@ -164,6 +164,7 @@ struct mlx5e_encap_entry {
u8 flags; u8 flags;
char *encap_header; char *encap_header;
int encap_size; int encap_size;
refcount_t refcnt;
}; };
struct mlx5e_rep_sq { struct mlx5e_rep_sq {
......
...@@ -103,6 +103,7 @@ enum { ...@@ -103,6 +103,7 @@ enum {
* container_of(helper item, containing struct type, helper field[index]) * container_of(helper item, containing struct type, helper field[index])
*/ */
struct encap_flow_item { struct encap_flow_item {
struct mlx5e_encap_entry *e; /* attached encap instance */
struct list_head list; struct list_head list;
int index; int index;
}; };
...@@ -1433,8 +1434,11 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe) ...@@ -1433,8 +1434,11 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe)
list_for_each_entry(e, &nhe->encap_list, encap_list) { list_for_each_entry(e, &nhe->encap_list, encap_list) {
struct encap_flow_item *efi, *tmp; struct encap_flow_item *efi, *tmp;
if (!(e->flags & MLX5_ENCAP_ENTRY_VALID))
if (!(e->flags & MLX5_ENCAP_ENTRY_VALID) ||
!mlx5e_encap_take(e))
continue; continue;
list_for_each_entry_safe(efi, tmp, &e->flows, list) { list_for_each_entry_safe(efi, tmp, &e->flows, list) {
flow = container_of(efi, struct mlx5e_tc_flow, flow = container_of(efi, struct mlx5e_tc_flow,
encaps[efi->index]); encaps[efi->index]);
...@@ -1453,6 +1457,8 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe) ...@@ -1453,6 +1457,8 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe)
mlx5e_flow_put(netdev_priv(e->out_dev), flow); mlx5e_flow_put(netdev_priv(e->out_dev), flow);
} }
mlx5e_encap_put(netdev_priv(e->out_dev), e);
if (neigh_used) if (neigh_used)
break; break;
} }
...@@ -1472,20 +1478,12 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe) ...@@ -1472,20 +1478,12 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe)
} }
} }
static void mlx5e_detach_encap(struct mlx5e_priv *priv, void mlx5e_encap_put(struct mlx5e_priv *priv, struct mlx5e_encap_entry *e)
struct mlx5e_tc_flow *flow, int out_index)
{ {
struct list_head *next = flow->encaps[out_index].list.next; if (!refcount_dec_and_test(&e->refcnt))
/* flow wasn't fully initialized */
if (list_empty(&flow->encaps[out_index].list))
return; return;
list_del(&flow->encaps[out_index].list); WARN_ON(!list_empty(&e->flows));
if (list_empty(next)) {
struct mlx5e_encap_entry *e;
e = list_entry(next, struct mlx5e_encap_entry, flows);
mlx5e_rep_encap_entry_detach(netdev_priv(e->out_dev), e); mlx5e_rep_encap_entry_detach(netdev_priv(e->out_dev), e);
if (e->flags & MLX5_ENCAP_ENTRY_VALID) if (e->flags & MLX5_ENCAP_ENTRY_VALID)
...@@ -1494,7 +1492,19 @@ static void mlx5e_detach_encap(struct mlx5e_priv *priv, ...@@ -1494,7 +1492,19 @@ static void mlx5e_detach_encap(struct mlx5e_priv *priv,
hash_del_rcu(&e->encap_hlist); hash_del_rcu(&e->encap_hlist);
kfree(e->encap_header); kfree(e->encap_header);
kfree(e); kfree(e);
} }
static void mlx5e_detach_encap(struct mlx5e_priv *priv,
struct mlx5e_tc_flow *flow, int out_index)
{
/* flow wasn't fully initialized */
if (!flow->encaps[out_index].e)
return;
list_del(&flow->encaps[out_index].list);
mlx5e_encap_put(priv, flow->encaps[out_index].e);
flow->encaps[out_index].e = NULL;
} }
static void __mlx5e_tc_del_fdb_peer_flow(struct mlx5e_tc_flow *flow) static void __mlx5e_tc_del_fdb_peer_flow(struct mlx5e_tc_flow *flow)
...@@ -2817,6 +2827,31 @@ static bool is_merged_eswitch_dev(struct mlx5e_priv *priv, ...@@ -2817,6 +2827,31 @@ static bool is_merged_eswitch_dev(struct mlx5e_priv *priv,
bool mlx5e_encap_take(struct mlx5e_encap_entry *e)
{
return refcount_inc_not_zero(&e->refcnt);
}
static struct mlx5e_encap_entry *
mlx5e_encap_get(struct mlx5e_priv *priv, struct encap_key *key,
uintptr_t hash_key)
{
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
struct mlx5e_encap_entry *e;
struct encap_key e_key;
hash_for_each_possible_rcu(esw->offloads.encap_tbl, e,
encap_hlist, hash_key) {
e_key.ip_tun_key = &e->tun_info->key;
e_key.tc_tunnel = e->tunnel;
if (!cmp_encap_info(&e_key, key) &&
mlx5e_encap_take(e))
return e;
}
return NULL;
}
static int mlx5e_attach_encap(struct mlx5e_priv *priv, static int mlx5e_attach_encap(struct mlx5e_priv *priv,
struct mlx5e_tc_flow *flow, struct mlx5e_tc_flow *flow,
struct net_device *mirred_dev, struct net_device *mirred_dev,
...@@ -2829,11 +2864,10 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv, ...@@ -2829,11 +2864,10 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
struct mlx5_esw_flow_attr *attr = flow->esw_attr; struct mlx5_esw_flow_attr *attr = flow->esw_attr;
struct mlx5e_tc_flow_parse_attr *parse_attr; struct mlx5e_tc_flow_parse_attr *parse_attr;
const struct ip_tunnel_info *tun_info; const struct ip_tunnel_info *tun_info;
struct encap_key key, e_key; struct encap_key key;
struct mlx5e_encap_entry *e; struct mlx5e_encap_entry *e;
unsigned short family; unsigned short family;
uintptr_t hash_key; uintptr_t hash_key;
bool found = false;
int err = 0; int err = 0;
parse_attr = attr->parse_attr; parse_attr = attr->parse_attr;
...@@ -2848,24 +2882,17 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv, ...@@ -2848,24 +2882,17 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
hash_key = hash_encap_info(&key); hash_key = hash_encap_info(&key);
hash_for_each_possible_rcu(esw->offloads.encap_tbl, e, e = mlx5e_encap_get(priv, &key, hash_key);
encap_hlist, hash_key) {
e_key.ip_tun_key = &e->tun_info->key;
e_key.tc_tunnel = e->tunnel;
if (!cmp_encap_info(&e_key, &key)) {
found = true;
break;
}
}
/* must verify if encap is valid or not */ /* must verify if encap is valid or not */
if (found) if (e)
goto attach_flow; goto attach_flow;
e = kzalloc(sizeof(*e), GFP_KERNEL); e = kzalloc(sizeof(*e), GFP_KERNEL);
if (!e) if (!e)
return -ENOMEM; return -ENOMEM;
refcount_set(&e->refcnt, 1);
e->tun_info = tun_info; e->tun_info = tun_info;
err = mlx5e_tc_tun_init_encap_attr(mirred_dev, priv, e, extack); err = mlx5e_tc_tun_init_encap_attr(mirred_dev, priv, e, extack);
if (err) if (err)
...@@ -2884,6 +2911,7 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv, ...@@ -2884,6 +2911,7 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
hash_add_rcu(esw->offloads.encap_tbl, &e->encap_hlist, hash_key); hash_add_rcu(esw->offloads.encap_tbl, &e->encap_hlist, hash_key);
attach_flow: attach_flow:
flow->encaps[out_index].e = e;
list_add(&flow->encaps[out_index].list, &e->flows); list_add(&flow->encaps[out_index].list, &e->flows);
flow->encaps[out_index].index = out_index; flow->encaps[out_index].index = out_index;
*encap_dev = e->out_dev; *encap_dev = e->out_dev;
......
...@@ -75,6 +75,8 @@ void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, ...@@ -75,6 +75,8 @@ void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv,
struct mlx5e_encap_entry *e); struct mlx5e_encap_entry *e);
void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv,
struct mlx5e_encap_entry *e); struct mlx5e_encap_entry *e);
bool mlx5e_encap_take(struct mlx5e_encap_entry *e);
void mlx5e_encap_put(struct mlx5e_priv *priv, struct mlx5e_encap_entry *e);
struct mlx5e_neigh_hash_entry; struct mlx5e_neigh_hash_entry;
void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe); void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment