bond: ping fails between l2 BD [VPP-1238]

In dpdk based bonding, when the bond interface is configured for l2,
it automatically sets the bond interface to promiscuous mode and sets rx
redirect to ethernet-input. This allows traffic to be bridged to
non compute node facing interface when it is received from the compute
node interface.

For native vpp bonding, we need to do similar things. When the bond interface
is configured for l2, we set the slave interfaces to promiscuous mode
and set rx redirect to ethernet-input because dpdk does not know anything
about the bond interface. Likewise, when a new interface is enslaved, we also
need to do the same thing if the bond interface has already been configured
for l2.

Change-Id: I7e168008e8a4221be74929b2a20e6db0ce8f3110
Signed-off-by: Steven <sluong@cisco.com>

1 files changed, 33 insertions, 15 deletions

diff --git a/src/vnet/bonding/cli.c b/src/vnet/bonding/cli.c
index 1768912bf12..3ee99a51b62 100644
--- a/src/vnet/bonding/cli.c
+++ b/src/vnet/bonding/cli.c
@@ -135,7 +135,10 @@ bond_delete_neighbor (vlib_main_t * vm, bond_if_t * bif, slave_if_t * sif)
   bond_main_t *bm = &bond_main;
   vnet_main_t *vnm = vnet_get_main ();
   int i;
-  vnet_hw_interface_t *hw;
+  vnet_hw_interface_t *sif_hw, *bif_hw;
+
+  sif_hw = vnet_get_sup_hw_interface (vnm, sif->sw_if_index);
+  bif_hw = vnet_get_sup_hw_interface (vnm, bif->sw_if_index);
 
   bif->port_number_bitmap =
     clib_bitmap_set (bif->port_number_bitmap,
@@ -153,11 +156,17 @@ bond_delete_neighbor (vlib_main_t * vm, bond_if_t * bif, slave_if_t * sif)
       }
   }
 
+  if (bif_hw->l2_if_count)
+    {
+      ethernet_set_flags (vnm, sif_hw->hw_if_index, 0);
+      /* Allow ip packets to go directly to ip4-input etc */
+      ethernet_set_rx_redirect (vnm, sif_hw, 0);
+    }
+
   bond_disable_collecting_distributing (vm, sif);
 
   /* Put back the old mac */
-  hw = vnet_get_sup_hw_interface (vnm, sif->sw_if_index);
-  vnet_hw_interface_change_mac_address (vnm, hw->hw_if_index,
+  vnet_hw_interface_change_mac_address (vnm, sif_hw->hw_if_index,
 					sif->persistent_hw_address);
 
   pool_put (bm->neighbors, sif);
@@ -406,7 +415,7 @@ bond_enslave (vlib_main_t * vm, bond_enslave_args_t * args)
   bond_if_t *bif;
   slave_if_t *sif;
   vnet_interface_main_t *im = &vnm->interface_main;
-  vnet_hw_interface_t *hw, *hw2;
+  vnet_hw_interface_t *bif_hw, *sif_hw;
   vnet_sw_interface_t *sw;
 
   bif = bond_get_master_by_sw_if_index (args->group);
@@ -423,8 +432,8 @@ bond_enslave (vlib_main_t * vm, bond_enslave_args_t * args)
       args->error = clib_error_return (0, "interface was already enslaved");
       return;
     }
-  hw = vnet_get_sup_hw_interface (vnm, args->slave);
-  if (hw->dev_class_index == bond_dev_class.index)
+  sif_hw = vnet_get_sup_hw_interface (vnm, args->slave);
+  if (sif_hw->dev_class_index == bond_dev_class.index)
     {
       args->rv = VNET_API_ERROR_INVALID_INTERFACE;
       args->error =
@@ -454,12 +463,14 @@ bond_enslave (vlib_main_t * vm, bond_enslave_args_t * args)
 	    sif - bm->neighbors);
   vec_add1 (bif->slaves, sif->sw_if_index);
 
-  hw = vnet_get_sup_hw_interface (vnm, sif->sw_if_index);
+  sif_hw = vnet_get_sup_hw_interface (vnm, sif->sw_if_index);
+
   /* Save the old mac */
-  memcpy (sif->persistent_hw_address, hw->hw_address, 6);
+  memcpy (sif->persistent_hw_address, sif_hw->hw_address, 6);
+  bif_hw = vnet_get_sup_hw_interface (vnm, bif->sw_if_index);
   if (bif->use_custom_mac)
     {
-      vnet_hw_interface_change_mac_address (vnm, hw->hw_if_index,
+      vnet_hw_interface_change_mac_address (vnm, sif_hw->hw_if_index,
 					    bif->hw_address);
     }
   else
@@ -467,19 +478,26 @@ bond_enslave (vlib_main_t * vm, bond_enslave_args_t * args)
       // bond interface gets the mac address from the first slave
       if (vec_len (bif->slaves) == 1)
 	{
-	  memcpy (bif->hw_address, hw->hw_address, 6);
-	  hw2 = vnet_get_sup_hw_interface (vnm, bif->sw_if_index);
-	  vnet_hw_interface_change_mac_address (vnm, hw2->hw_if_index,
-						hw->hw_address);
+	  memcpy (bif->hw_address, sif_hw->hw_address, 6);
+	  vnet_hw_interface_change_mac_address (vnm, bif_hw->hw_if_index,
+						sif_hw->hw_address);
 	}
       else
 	{
 	  // subsequent slaves gets the mac address of the bond interface
-	  vnet_hw_interface_change_mac_address (vnm, hw->hw_if_index,
+	  vnet_hw_interface_change_mac_address (vnm, sif_hw->hw_if_index,
 						bif->hw_address);
 	}
     }
 
+  if (bif_hw->l2_if_count)
+    {
+      ethernet_set_flags (vnm, sif_hw->hw_if_index,
+			  ETHERNET_INTERFACE_FLAG_ACCEPT_ALL);
+      /* ensure all packets go to ethernet-input */
+      ethernet_set_rx_redirect (vnm, sif_hw, 1);
+    }
+
   if ((bif->mode == BOND_MODE_LACP) && bm->lacp_enable_disable)
     {
       (*bm->lacp_enable_disable) (vm, bif, sif, 1);
@@ -490,7 +508,7 @@ bond_enslave (vlib_main_t * vm, bond_enslave_args_t * args)
     }
 
   args->rv = vnet_feature_enable_disable ("device-input", "bond-input",
-					  hw->hw_if_index, 1, 0, 0);
+					  sif_hw->hw_if_index, 1, 0, 0);
 
   if (args->rv)
     {