Hello Richard Guy Briggs,
This is a semi-automatic email about new static checker warnings.
The patch 4f520900522f: "netlink: have netlink per-protocol bind
function return an error code." from Apr 22, 2014, leads to the
following Smatch complaint:
net/netlink/af_netlink.c:1056 netlink_bind()
error: we previously assumed 'nlk->groups' could be null (see line 1049)
net/netlink/af_netlink.c
1048
1049 if (!groups && (nlk->groups == NULL || !(u32)nlk->groups[0]))
^^^^^^^
1050 goto unlock;
1051 netlink_unlock_table();
1052
1053 netlink_table_grab();
1054 netlink_update_subscriptions(sk, nlk->subscriptions +
1055 hweight32(groups) -
1056 hweight32(nlk->groups[0]));
^^^^^^^^^^^^^^
Say groups is non-NULL and nlk->groups is NULL then this would be an
issue. Is that possible? I don't know why Smatch is just now suddenly
complaining about a 4 year old patch.
1057 nlk->groups[0] = (nlk->groups[0] & ~0xffffffffUL) | groups;
1058 netlink_update_listeners(sk);
regards,
dan carpenter
--
To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
