diff options
author | Johannes Berg <johannes.berg@intel.com> | 2011-11-03 00:07:32 +0000 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@suse.de> | 2011-11-11 09:44:43 -0800 |
commit | 61ec3a5696f0c32448c93b713f934b007b85f961 (patch) | |
tree | 457b77f5305517b77e732e5335bd4cd823cdedd2 /virt | |
parent | 0fb9f687964a4b4a557b4f2e8676e1236a9e269b (diff) | |
download | linux-stable-61ec3a5696f0c32448c93b713f934b007b85f961.tar.gz linux-stable-61ec3a5696f0c32448c93b713f934b007b85f961.tar.bz2 linux-stable-61ec3a5696f0c32448c93b713f934b007b85f961.zip |
netlink: validate NLA_MSECS length
commit c30bc94758ae2a38a5eb31767c1985c0aae0950b upstream.
L2TP for example uses NLA_MSECS like this:
policy:
[L2TP_ATTR_RECV_TIMEOUT] = { .type = NLA_MSECS, },
code:
if (info->attrs[L2TP_ATTR_RECV_TIMEOUT])
cfg.reorder_timeout = nla_get_msecs(info->attrs[L2TP_ATTR_RECV_TIMEOUT]);
As nla_get_msecs() is essentially nla_get_u64() plus the
conversion to a HZ-based value, this will not properly
reject attributes from userspace that aren't long enough
and might overrun the message.
Add NLA_MSECS to the attribute minlen array to check the
size properly.
Cc: Thomas Graf <tgraf@suug.ch>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Diffstat (limited to 'virt')
0 files changed, 0 insertions, 0 deletions