IB/core: Delete lower netdevice default GID entries in bonding scenario

When NETDEV_CHANGEUPPER event occurs, lower device is not yet established
as slave of the master, and when upper device is bond device, default GID
entries not deleted.

Due to this, when bond device is fully configured, default GID entries of
bond device cannot be added as default GID entries are occupied by the
lower netdevice. This is incorrect.

Default GID entries should really be of bond netdevice because in all RoCE
GIDs (default or IP), MAC address of the bond device will be used.  It is
confusing to have default GID of netdevice which is not really used for
any purpose.

Therefore, as first step, implement
(a) filter function which filters if a CHANGEUPPER event netdevice and
    associated upper device is master device or not.
(b) callback function which deletes the default GIDs of lower (event
    netdevice).

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

diff --git a/drivers/infiniband/core/roce_gid_mgmt.c b/drivers/infiniband/core/roce_gid_mgmt.c
index c7cf3c7f38a682116d28d9400bac318712039f2f..21e008aa863c3114446f9e24242fb6aa2dc68c36 100644 (file)
--- a/drivers/infiniband/core/roce_gid_mgmt.c
+++ b/drivers/infiniband/core/roce_gid_mgmt.c
@@ -208,6 +208,34 @@ static int upper_device_filter(struct ib_device *ib_dev, u8 port,
        return res;
 }
 
+/**
+ * is_upper_ndev_bond_master_filter - Check if a given netdevice
+ * is bond master device of netdevice of the the RDMA device of port.
+ * @ib_dev:            IB device to check
+ * @port:              Port to consider for adding default GID
+ * @rdma_ndev:         Pointer to rdma netdevice
+ * @cookie:            Netdevice to consider to form a default GID
+ *
+ * is_upper_ndev_bond_master_filter() returns true if a cookie_netdev
+ * is bond master device and rdma_ndev is its lower netdevice. It might
+ * not have been established as slave device yet.
+ */
+static int
+is_upper_ndev_bond_master_filter(struct ib_device *ib_dev, u8 port,
+                                struct net_device *rdma_ndev,
+                                void *cookie)
+{
+       struct net_device *cookie_ndev = cookie;
+       bool match = false;
+
+       rcu_read_lock();
+       if (netif_is_bond_master(cookie_ndev) &&
+           rdma_is_upper_dev_rcu(rdma_ndev, cookie_ndev))
+               match = true;
+       rcu_read_unlock();
+       return match;
+}
+
 static void update_gid_ip(enum gid_op_type gid_op,
                          struct ib_device *ib_dev,
                          u8 port, struct net_device *ndev,
@@ -391,6 +419,27 @@ static void del_netdev_ips(struct ib_device *ib_dev, u8 port,
        ib_cache_gid_del_all_netdev_gids(ib_dev, port, cookie);
 }
 
+/**
+ * del_default_gids - Delete default GIDs of the event/cookie netdevice
+ * @ib_dev:    RDMA device pointer
+ * @port:      Port of the RDMA device whose GID table to consider
+ * @rdma_ndev: Unused rdma netdevice
+ * @cookie:    Pointer to event netdevice
+ *
+ * del_default_gids() deletes the default GIDs of the event/cookie netdevice.
+ */
+static void del_default_gids(struct ib_device *ib_dev, u8 port,
+                            struct net_device *rdma_ndev, void *cookie)
+{
+       struct net_device *cookie_ndev = cookie;
+       unsigned long gid_type_mask;
+
+       gid_type_mask = roce_gid_type_mask_support(ib_dev, port);
+
+       ib_cache_gid_set_default_gid(ib_dev, port, cookie_ndev, gid_type_mask,
+                                    IB_CACHE_GID_DEFAULT_MODE_DELETE);
+}
+
 static void enum_all_gids_of_dev_cb(struct ib_device *ib_dev,
                                    u8 port,
                                    struct net_device *rdma_ndev,
@@ -589,30 +638,34 @@ ndev_event_unlink(struct netdev_notifier_changeupper_info *changeupper_info,
 }
 
 static void
-ndev_event_link(struct netdev_notifier_changeupper_info *changeupper_info,
+ndev_event_link(struct net_device *event_ndev,
+               struct netdev_notifier_changeupper_info *changeupper_info,
                struct netdev_event_work_cmd *cmds)
 {
        static const struct netdev_event_work_cmd
                        bonding_default_del_cmd = {
-                               .cb     = bond_delete_netdev_default_gids,
-                               .filter = is_eth_port_inactive_slave
+                               .cb     = del_default_gids,
+                               .filter = is_upper_ndev_bond_master_filter
                        };
        /*
         * When a lower netdev is linked to its upper bonding
-        * netdev, delete lower inactive slave netdev's default GIDs.
+        * netdev, delete lower slave netdev's default GIDs.
         */
        cmds[0] = bonding_default_del_cmd;
-       cmds[0].ndev = changeupper_info->upper_dev;
+       cmds[0].ndev = event_ndev;
+       cmds[0].filter_ndev = changeupper_info->upper_dev;
+
        cmds[1] = add_cmd_upper_ips;
        cmds[1].ndev = changeupper_info->upper_dev;
        cmds[1].filter_ndev = changeupper_info->upper_dev;
 }
 
-static void netdevice_event_changeupper(struct netdev_notifier_changeupper_info *changeupper_info,
-                                       struct netdev_event_work_cmd *cmds)
+static void netdevice_event_changeupper(struct net_device *event_ndev,
+               struct netdev_notifier_changeupper_info *changeupper_info,
+               struct netdev_event_work_cmd *cmds)
 {
        if (changeupper_info->linking)
-               ndev_event_link(changeupper_info, cmds);
+               ndev_event_link(event_ndev, changeupper_info, cmds);
        else
                ndev_event_unlink(changeupper_info, cmds);
 }
@@ -657,7 +710,7 @@ static int netdevice_event(struct notifier_block *this, unsigned long event,
                break;
 
        case NETDEV_CHANGEUPPER:
-               netdevice_event_changeupper(
+               netdevice_event_changeupper(ndev,
                        container_of(ptr, struct netdev_notifier_changeupper_info, info),
                        cmds);
                break;