[PATCH 0/2] bugfix for bonding

[PATCH 0/2] bugfix for bonding

[Min Hu (Connor)]

Two bugfixed for bonding.

Min Hu (Connor) (2):
  net/bonding: fix promiscuous and allmulticast state
  net/bonding: fix reference count on mbufs

 drivers/net/bonding/rte_eth_bond_pmd.c | 72 +++++++++++++++++++++++++-
 1 file changed, 71 insertions(+), 1 deletion(-)

-- 
2.33.0

[PATCH 1/2] net/bonding: fix promiscuous and allmulticast state

[Min Hu (Connor)]

Currently, promiscuous or allmulticast state of bonding port will not be
passed to the new primary slave when active/standby switchover. It causes
bugs in some scenario.

For example, promiscuous state of bonding port is off now, primary slave
(called A) is off but secondary slave(called B) is on. Then active/standby
switchover, promiscuous state of the bonding port is off, but the new
primary slave turns to be B and its promiscuous state is still on. It is
not consistent with bonding port. And this patch will fix it.

Fixes: 2efb58cbab6e ("bond: new link bonding library")
Fixes: 68218b87c184 ("net/bonding: prefer allmulti to promiscuous for LACP")
Cc: stable@dpdk.org

Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
 drivers/net/bonding/rte_eth_bond_pmd.c | 70 ++++++++++++++++++++++++++
 1 file changed, 70 insertions(+)

diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c
index c72fc64806..ba587e60bf 100644
--- a/drivers/net/bonding/rte_eth_bond_pmd.c
+++ b/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -2691,6 +2691,39 @@ bond_ethdev_promiscuous_disable(struct rte_eth_dev *dev)
 	return ret;
 }
 
+static int
+bond_ethdev_promiscuous_update(struct rte_eth_dev *dev)
+{
+	struct bond_dev_private *internals = dev->data->dev_private;
+	uint16_t port_id = internals->current_primary_port;
+
+	switch (internals->mode) {
+	case BONDING_MODE_ROUND_ROBIN:
+	case BONDING_MODE_BALANCE:
+	case BONDING_MODE_BROADCAST:
+	case BONDING_MODE_8023AD:
+		/* As promiscuous mode is propagated to all slaves for these
+		 * mode, no need to update for bonding device.
+		 */
+		break;
+	case BONDING_MODE_ACTIVE_BACKUP:
+	case BONDING_MODE_TLB:
+	case BONDING_MODE_ALB:
+	default:
+		/* As promiscuous mode is propagated only to primary slave
+		 * for these mode. When active/standby switchover, promiscuous
+		 * mode should be set to new primary slave according to bonding
+		 * device.
+		 */
+		if (rte_eth_promiscuous_get(internals->port_id) == 1)
+			rte_eth_promiscuous_enable(port_id);
+		else
+			rte_eth_promiscuous_disable(port_id);
+	}
+
+	return 0;
+}
+
 static int
 bond_ethdev_allmulticast_enable(struct rte_eth_dev *eth_dev)
 {
@@ -2804,6 +2837,39 @@ bond_ethdev_allmulticast_disable(struct rte_eth_dev *eth_dev)
 	return ret;
 }
 
+static int
+bond_ethdev_allmulticast_update(struct rte_eth_dev *dev)
+{
+	struct bond_dev_private *internals = dev->data->dev_private;
+	uint16_t port_id = internals->current_primary_port;
+
+	switch (internals->mode) {
+	case BONDING_MODE_ROUND_ROBIN:
+	case BONDING_MODE_BALANCE:
+	case BONDING_MODE_BROADCAST:
+	case BONDING_MODE_8023AD:
+		/* As allmulticast mode is propagated to all slaves for these
+		 * mode, no need to update for bonding device.
+		 */
+		break;
+	case BONDING_MODE_ACTIVE_BACKUP:
+	case BONDING_MODE_TLB:
+	case BONDING_MODE_ALB:
+	default:
+		/* As allmulticast mode is propagated only to primary slave
+		 * for these mode. When active/standby switchover, allmulticast
+		 * mode should be set to new primary slave according to bonding
+		 * device.
+		 */
+		if (rte_eth_allmulticast_get(internals->port_id) == 1)
+			rte_eth_allmulticast_enable(port_id);
+		else
+			rte_eth_allmulticast_disable(port_id);
+	}
+
+	return 0;
+}
+
 static void
 bond_ethdev_delayed_lsc_propagation(void *arg)
 {
@@ -2893,6 +2959,8 @@ bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type,
 			lsc_flag = 1;
 
 			mac_address_slaves_update(bonded_eth_dev);
+			bond_ethdev_promiscuous_update(bonded_eth_dev);
+			bond_ethdev_allmulticast_update(bonded_eth_dev);
 		}
 
 		activate_slave(bonded_eth_dev, port_id);
@@ -2922,6 +2990,8 @@ bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type,
 			else
 				internals->current_primary_port = internals->primary_port;
 			mac_address_slaves_update(bonded_eth_dev);
+			bond_ethdev_promiscuous_update(bonded_eth_dev);
+			bond_ethdev_allmulticast_update(bonded_eth_dev);
 		}
 	}
 
-- 
2.33.0

[PATCH 2/2] net/bonding: fix reference count on mbufs

[Min Hu (Connor)]

In bonding Tx broadcast mode, Packets should be sent by every slave,
but only one mbuf exits. The solution is to increment reference count
on mbufs, but it ignores multi segments.

This patch fixed it by adding reference for every segment in multi
segments Tx scenario.

Fixes: 2efb58cbab6e ("bond: new link bonding library")
Cc: stable@dpdk.org

Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
 drivers/net/bonding/rte_eth_bond_pmd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c
index ba587e60bf..c1d9300100 100644
--- a/drivers/net/bonding/rte_eth_bond_pmd.c
+++ b/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -1318,7 +1318,7 @@ bond_ethdev_tx_burst_broadcast(void *queue, struct rte_mbuf **bufs,
 
 	/* Increment reference count on mbufs */
 	for (i = 0; i < nb_pkts; i++)
-		rte_mbuf_refcnt_update(bufs[i], num_of_slaves - 1);
+		rte_pktmbuf_refcnt_update(bufs[i], num_of_slaves - 1);
 
 	/* Transmit burst on each active slave */
 	for (i = 0; i < num_of_slaves; i++) {
-- 
2.33.0

Re: [PATCH 0/2] bugfix for bonding

[Ferruh Yigit]

On 1/28/2022 2:25 AM, Min Hu (Connor) wrote:
> Two bugfixed for bonding.
> 
> Min Hu (Connor) (2):
>    net/bonding: fix promiscuous and allmulticast state
>    net/bonding: fix reference count on mbufs
> 

Series applied to dpdk-next-net/main, thanks.