Re: [B.A.T.M.A.N.] [PATCH v8 05/14] batman-adv: netlink: add translation table query

On Mon, May 23, 2016 at 05:59:36PM +0200, Simon Wunderlich wrote:

+static int +batadv_tt_local_dump_entry(struct sk_buff *msg, u32 portid, u32 seq,

• 	   struct batadv_priv *bat_priv,
  

• 	   struct batadv_tt_common_entry *common)
  

+{

[...]

• if (!(common->flags & BATADV_TT_CLIENT_NOPURGE)) {

• if (nla_put_u32(msg, BATADV_ATTR_LAST_SEEN_MSECS,
  

• 		last_seen_msecs))
  

• 	goto nla_put_failure;
  

• }

Unnecessary brackets here, maybe use '&&' to skip a layer of indention?

+int batadv_tt_local_dump(struct sk_buff *msg, struct netlink_callback *cb) +{

• struct net *net = sock_net(cb->skb->sk);
• struct net_device *soft_iface = NULL;

No need to set soft_iface to NULL.

+static int +batadv_tt_global_dump_entry(struct sk_buff *msg, u32 portid, u32 seq,

• 	    struct batadv_priv *bat_priv,
  

• 	    struct batadv_tt_common_entry *common, int *sub_s)
  

+{

• struct batadv_tt_orig_list_entry *orig_entry, *best_entry;
• struct batadv_tt_global_entry *global;
• struct hlist_head *head;
• int sub = 0;
• bool best;
• global = container_of(common, struct batadv_tt_global_entry, common);
• best_entry = batadv_transtable_best_orig(bat_priv, global);
• head = &global->orig_list;
• hlist_for_each_entry_rcu(orig_entry, head, list) {

Either add an explicit rcu-lock or keep to the doc style to add a line in the kerneldoc about caller needing to hold it?

+int batadv_tt_global_dump(struct sk_buff *msg, struct netlink_callback *cb) +{

• struct net *net = sock_net(cb->skb->sk);
• struct net_device *soft_iface = NULL;

No need to set soft_iface to NULL.