Hi,
Not really sure, but it seems this part of the patch is rejected against
the git master tree
On 20.09.2011 00:00, Jesse Gross wrote:
len += nla_total_size(upcall_info->actions_len);
user_skb = genlmsg_new(len, GFP_ATOMIC);
- if (!user_skb) {
- netlink_set_err(INIT_NET_GENL_SOCK, 0, group, -ENOBUFS);
+ if (!user_skb)
goto err_kfree_skbs;
- }
upcall = genlmsg_put(user_skb, 0, 0,&dp_packet_genl_family, 0,
upcall_info->cmd);
upcall->dp_ifindex = dp->dp_ifindex;
actual code on git:
user_skb = genlmsg_new(len, GFP_ATOMIC);
if (!user_skb) {
netlink_set_err(INIT_NET_GENL_SOCK, 0, group,
-ENOBUFS);
err = -ENOMEM;
goto err_kfree_skbs;
}
upcall = genlmsg_put(user_skb, 0, 0,
&dp_packet_genl_family, 0, upcall_info->cmd);
upcall->dp_ifindex = dp->dp_ifindex;
guess it has something to do with err = -ENOMEM not being removed
I'm not a code warrior, so sorry if I'm wrong :)
Sébastien
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
