net/mlx5: fix reception when VLAN is added
authorNélio Laranjeiro <nelio.laranjeiro@6wind.com>
Tue, 24 Oct 2017 15:18:17 +0000 (17:18 +0200)
committerFerruh Yigit <ferruh.yigit@intel.com>
Thu, 26 Oct 2017 00:33:01 +0000 (02:33 +0200)
When VLAN is enabled in the Rx side, only packets matching this VLAN are
expected, this also includes the broadcast and all multicast packets.

Fixes: 272733b5ebfd ("net/mlx5: use flow to enable unicast traffic")
Fixes: 6a6b6828fe6a ("net/mlx5: use flow to enable all multi mode")

Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
drivers/net/mlx5/mlx5_trigger.c

index 29167ba..982d0a2 100644 (file)
@@ -251,6 +251,29 @@ mlx5_dev_stop(struct rte_eth_dev *dev)
 int
 priv_dev_traffic_enable(struct priv *priv, struct rte_eth_dev *dev)
 {
+       struct rte_flow_item_eth bcast = {
+               .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
+       };
+       struct rte_flow_item_eth ipv6_multi_spec = {
+               .dst.addr_bytes = "\x33\x33\x00\x00\x00\x00",
+       };
+       struct rte_flow_item_eth ipv6_multi_mask = {
+               .dst.addr_bytes = "\xff\xff\x00\x00\x00\x00",
+       };
+       struct rte_flow_item_eth unicast = {
+               .src.addr_bytes = "\x00\x00\x00\x00\x00\x00",
+       };
+       struct rte_flow_item_eth unicast_mask = {
+               .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
+       };
+       const unsigned int vlan_filter_n = priv->vlan_filter_n;
+       const struct ether_addr cmp = {
+               .addr_bytes = "\x00\x00\x00\x00\x00\x00",
+       };
+       unsigned int i;
+       unsigned int j;
+       int ret;
+
        if (priv->isolated)
                return 0;
        if (dev->data->promiscuous) {
@@ -261,81 +284,80 @@ priv_dev_traffic_enable(struct priv *priv, struct rte_eth_dev *dev)
                };
 
                claim_zero(mlx5_ctrl_flow(dev, &promisc, &promisc));
-       } else if (dev->data->all_multicast) {
+               return 0;
+       }
+       if (dev->data->all_multicast) {
                struct rte_flow_item_eth multicast = {
                        .dst.addr_bytes = "\x01\x00\x00\x00\x00\x00",
-                       .src.addr_bytes = "\x01\x00\x00\x00\x00\x00",
+                       .src.addr_bytes = "\x00\x00\x00\x00\x00\x00",
                        .type = 0,
                };
 
                claim_zero(mlx5_ctrl_flow(dev, &multicast, &multicast));
        } else {
-               struct rte_flow_item_eth bcast = {
-                       .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
-               };
-               struct rte_flow_item_eth ipv6_multi_spec = {
-                       .dst.addr_bytes = "\x33\x33\x00\x00\x00\x00",
-               };
-               struct rte_flow_item_eth ipv6_multi_mask = {
-                       .dst.addr_bytes = "\xff\xff\x00\x00\x00\x00",
-               };
-               struct rte_flow_item_eth unicast = {
-                       .src.addr_bytes = "\x00\x00\x00\x00\x00\x00",
-               };
-               struct rte_flow_item_eth unicast_mask = {
-                       .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff",
-               };
-               const unsigned int vlan_filter_n = priv->vlan_filter_n;
-               const struct ether_addr cmp = {
-                       .addr_bytes = "\x00\x00\x00\x00\x00\x00",
-               };
-               unsigned int i;
-               unsigned int j;
-               unsigned int unicast_flow = 0;
-               int ret;
-
-               for (i = 0; i != MLX5_MAX_MAC_ADDRESSES; ++i) {
-                       struct ether_addr *mac = &dev->data->mac_addrs[i];
-
-                       if (!memcmp(mac, &cmp, sizeof(*mac)))
-                               continue;
-                       memcpy(&unicast.dst.addr_bytes,
-                              mac->addr_bytes,
-                              ETHER_ADDR_LEN);
-                       for (j = 0; j != vlan_filter_n; ++j) {
-                               uint16_t vlan = priv->vlan_filter[j];
-
-                               struct rte_flow_item_vlan vlan_spec = {
-                                       .tci = rte_cpu_to_be_16(vlan),
-                               };
-                               struct rte_flow_item_vlan vlan_mask = {
-                                       .tci = 0xffff,
-                               };
-
-                               ret = mlx5_ctrl_flow_vlan(dev, &unicast,
-                                                         &unicast_mask,
-                                                         &vlan_spec,
-                                                         &vlan_mask);
-                               if (ret)
-                                       goto error;
-                               unicast_flow = 1;
-                       }
-                       if (!vlan_filter_n) {
-                               ret = mlx5_ctrl_flow(dev, &unicast,
-                                                    &unicast_mask);
-                               if (ret)
-                                       goto error;
-                               unicast_flow = 1;
-                       }
+               /* Add broadcast/multicast flows. */
+               for (i = 0; i != vlan_filter_n; ++i) {
+                       uint16_t vlan = priv->vlan_filter[i];
+
+                       struct rte_flow_item_vlan vlan_spec = {
+                               .tci = rte_cpu_to_be_16(vlan),
+                       };
+                       struct rte_flow_item_vlan vlan_mask = {
+                               .tci = 0xffff,
+                       };
+
+                       ret = mlx5_ctrl_flow_vlan(dev, &bcast, &bcast,
+                                                 &vlan_spec, &vlan_mask);
+                       if (ret)
+                               goto error;
+                       ret = mlx5_ctrl_flow_vlan(dev, &ipv6_multi_spec,
+                                                 &ipv6_multi_mask,
+                                                 &vlan_spec, &vlan_mask);
+                       if (ret)
+                               goto error;
+               }
+               if (!vlan_filter_n) {
+                       ret = mlx5_ctrl_flow(dev, &bcast, &bcast);
+                       if (ret)
+                               goto error;
+                       ret = mlx5_ctrl_flow(dev, &ipv6_multi_spec,
+                                            &ipv6_multi_mask);
+                       if (ret)
+                               goto error;
+               }
+       }
+       /* Add MAC address flows. */
+       for (i = 0; i != MLX5_MAX_MAC_ADDRESSES; ++i) {
+               struct ether_addr *mac = &dev->data->mac_addrs[i];
+
+               if (!memcmp(mac, &cmp, sizeof(*mac)))
+                       continue;
+               memcpy(&unicast.dst.addr_bytes,
+                      mac->addr_bytes,
+                      ETHER_ADDR_LEN);
+               for (j = 0; j != vlan_filter_n; ++j) {
+                       uint16_t vlan = priv->vlan_filter[j];
+
+                       struct rte_flow_item_vlan vlan_spec = {
+                               .tci = rte_cpu_to_be_16(vlan),
+                       };
+                       struct rte_flow_item_vlan vlan_mask = {
+                               .tci = 0xffff,
+                       };
+
+                       ret = mlx5_ctrl_flow_vlan(dev, &unicast,
+                                                 &unicast_mask,
+                                                 &vlan_spec,
+                                                 &vlan_mask);
+                       if (ret)
+                               goto error;
+               }
+               if (!vlan_filter_n) {
+                       ret = mlx5_ctrl_flow(dev, &unicast,
+                                            &unicast_mask);
+                       if (ret)
+                               goto error;
                }
-               if (!unicast_flow)
-                       return 0;
-               ret = mlx5_ctrl_flow(dev, &bcast, &bcast);
-               if (ret)
-                       goto error;
-               ret = mlx5_ctrl_flow(dev, &ipv6_multi_spec, &ipv6_multi_mask);
-               if (ret)
-                       goto error;
        }
        return 0;
 error: