net/bonding: fix slave activation simultaneously

The bonding PMD decides to activate\deactivate its slaves according to
the slaves link statuses.
Thus, it registers to the LSC events of the slaves ports and
activates\deactivates them from its LSC callbacks called asynchronously
by the host thread when the slave link status is changed.

In addition, the bonding PMD uses the callback for slave activation
when it tries to start it, this operation is probably called by the
master thread.

Consequently, a slave may be activated in the same time by two
different threads and may cause a lot of optional errors, for example,
slave mempool recreation with the same name causes an error.

Synchronize the critical section in the LSC callback using a special
new spinlock.

Fixes: 414b202343ce ("bonding: fix initial link status of slave")
Fixes: a45b288ef21a ("bond: support link status polling")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>

diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c
index d0941c8..02d94b1 100644 (file)
--- a/drivers/net/bonding/rte_eth_bond_pmd.c
+++ b/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -2653,14 +2653,21 @@ bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type,
        if (!valid_slave)
                return rc;
 
+       /* Synchronize lsc callback parallel calls either by real link event
+        * from the slaves PMDs or by the bonding PMD itself.
+        */
+       rte_spinlock_lock(&internals->lsc_lock);
+
        /* Search for port in active port list */
        active_pos = find_slave_by_id(internals->active_slaves,
                        internals->active_slave_count, port_id);
 
        rte_eth_link_get_nowait(port_id, &link);
        if (link.link_status) {
-               if (active_pos < internals->active_slave_count)
+               if (active_pos < internals->active_slave_count) {
+                       rte_spinlock_unlock(&internals->lsc_lock);
                        return rc;
+               }
 
                /* if no active slave ports then set this port to be primary port */
                if (internals->active_slave_count < 1) {
@@ -2679,8 +2686,10 @@ bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type,
                                internals->primary_port == port_id)
                        bond_ethdev_primary_set(internals, port_id);
        } else {
-               if (active_pos == internals->active_slave_count)
+               if (active_pos == internals->active_slave_count) {
+                       rte_spinlock_unlock(&internals->lsc_lock);
                        return rc;
+               }
 
                /* Remove from active slave list */
                deactivate_slave(bonded_eth_dev, port_id);
@@ -2733,6 +2742,9 @@ bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type,
                                                NULL);
                }
        }
+
+       rte_spinlock_unlock(&internals->lsc_lock);
+
        return 0;
 }
 
@@ -2953,6 +2965,7 @@ bond_alloc(struct rte_vdev_device *dev, uint8_t mode)
        eth_dev->data->dev_flags = RTE_ETH_DEV_INTR_LSC;
 
        rte_spinlock_init(&internals->lock);
+       rte_spinlock_init(&internals->lsc_lock);
 
        internals->port_id = eth_dev->data->port_id;
        internals->mode = BONDING_MODE_INVALID;

diff --git a/drivers/net/bonding/rte_eth_bond_private.h b/drivers/net/bonding/rte_eth_bond_private.h
index f25e673..65445b8 100644 (file)
--- a/drivers/net/bonding/rte_eth_bond_private.h
+++ b/drivers/net/bonding/rte_eth_bond_private.h
@@ -105,6 +105,7 @@ struct bond_dev_private {
        uint8_t mode;                                           /**< Link Bonding Mode */
 
        rte_spinlock_t lock;
+       rte_spinlock_t lsc_lock;
 
        uint16_t primary_port;                  /**< Primary Slave Port */
        uint16_t current_primary_port;          /**< Primary Slave Port */