Error values of ib_query_port() and ib_query_device() weren't propagated
correctly. Because of that, ipoib_add_port() could return NULL value,
which escaped the IS_ERR() check in ipoib_add_one() and we crashed.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
        SET_NETDEV_DEV(priv->dev, hca->dma_device);
        priv->dev->dev_id = port - 1;
 
-       if (!ib_query_port(hca, port, &attr))
+       result = ib_query_port(hca, port, &attr);
+       if (!result)
                priv->max_ib_mtu = ib_mtu_enum_to_int(attr.max_mtu);
        else {
                printk(KERN_WARNING "%s: ib_query_port %d failed\n",
                goto device_init_failed;
        }
 
-       if (ipoib_set_dev_features(priv, hca))
+       result = ipoib_set_dev_features(priv, hca);
+       if (result)
                goto device_init_failed;
 
        /*