From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jason Gunthorpe Subject: [PATCH] rtnetlink: Actually use the policy for the IFLA_VF_INFO Date: Tue, 30 Jun 2015 16:52:55 -0600 Message-ID: <20150630225255.GA22529@obsidianresearch.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Nicolas Dichtel , Jiri Pirko , Thomas Graf , "David S. Miller" To: netdev@vger.kernel.org, Daniel Borkmann , Mitch Williams , Jeff Kirsher Return-path: Received: from quartz.orcorp.ca ([184.70.90.242]:41027 "EHLO quartz.orcorp.ca" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751136AbbF3WxE (ORCPT ); Tue, 30 Jun 2015 18:53:04 -0400 Content-Disposition: inline Sender: netdev-owner@vger.kernel.org List-ID: It turns out the policy was defined but never actually checked, so lets check it. Fixes: ebc08a6f47ee ("rtnetlink: Add VF config code to rtnetlink") Signed-off-by: Jason Gunthorpe --- net/core/rtnetlink.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) DaveM: This shouldn't be applied until someone with the hardware that uses this can test it to make sure the policy is actually correct and matches what iproute does.. I noticed this by inspection when investigating how to properly use netlink in another area. Compile tested only. Daniel/Mitch/Jeff: Can you test this? I suspect the absence of these checks allows user space to cause a read past the end of a buffer? I dropped the ifla_vfinfo_policy to match how IFLA_VF_PORTS/IFLA_VF_PORT is working but I wonder if that should be changed as well? Thanks, Jason diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a2b90e1fc115..7d5dc347bf7c 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1258,10 +1258,6 @@ static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = { [IFLA_INFO_SLAVE_DATA] = { .type = NLA_NESTED }, }; -static const struct nla_policy ifla_vfinfo_policy[IFLA_VF_INFO_MAX+1] = { - [IFLA_VF_INFO] = { .type = NLA_NESTED }, -}; - static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = { [IFLA_VF_MAC] = { .len = sizeof(struct ifla_vf_mac) }, [IFLA_VF_VLAN] = { .len = sizeof(struct ifla_vf_vlan) }, @@ -1681,6 +1677,7 @@ static int do_setlink(const struct sk_buff *skb, } if (tb[IFLA_VFINFO_LIST]) { + struct nlattr *vf_attrs[IFLA_VF_MAX + 1]; struct nlattr *attr; int rem; nla_for_each_nested(attr, tb[IFLA_VFINFO_LIST], rem) { @@ -1688,6 +1685,10 @@ static int do_setlink(const struct sk_buff *skb, err = -EINVAL; goto errout; } + err = nla_parse_nested(vf_attrs, IFLA_VF_MAX, attr, + ifla_vf_policy); + if (err < 0) + goto errout; err = do_setvfinfo(dev, attr); if (err < 0) goto errout; -- 2.1.4