From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pablo Neira Ayuso Subject: Re: [PATCH v2 1/2] netfilter_queue: enable UID/GID socket info retrieval Date: Fri, 20 Dec 2013 10:26:38 +0100 Message-ID: <20131220092638.GA9122@localhost> References: <1387451007-14363-1-git-send-email-valentina.giusti@bmw-carit.de> <1387451007-14363-2-git-send-email-valentina.giusti@bmw-carit.de> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netfilter-devel@vger.kernel.org, netdev@vger.kernel.org, fw@strlen.de, eric.dumazet@gmail.com, tgraf@redhat.com, jpa@google.com, davem@davemloft.net, daniel.wagner@bmw-carit.de, Valentina Giusti To: valentina.giusti@bmw-carit.de Return-path: Content-Disposition: inline In-Reply-To: <1387451007-14363-2-git-send-email-valentina.giusti@bmw-carit.de> Sender: netdev-owner@vger.kernel.org List-Id: netfilter-devel.vger.kernel.org Hi, On Thu, Dec 19, 2013 at 12:03:26PM +0100, valentina.giusti@bmw-carit.de= wrote: > From: Valentina Giusti >=20 > Thanks to commits 41063e9 (ipv4: Early TCP socket demux) and 421b388 = (udp: ipv4: > Add udp early demux) it is now possible to parse UID and GID socket i= nfo > also for incoming TCP and UDP connections. Having this info available= , it > is convenient to let NFQUEUE parse it in order to improve and refine = the > traffic analysis in userspace. >=20 > Signed-off-by: Valentina Giusti > --- > include/uapi/linux/netfilter/nfnetlink_queue.h | 5 ++++- > net/netfilter/nfnetlink_queue_core.c | 31 ++++++++++++++++= ++++++++++ > 2 files changed, 35 insertions(+), 1 deletion(-) >=20 > diff --git a/include/uapi/linux/netfilter/nfnetlink_queue.h b/include= /uapi/linux/netfilter/nfnetlink_queue.h > index 0132bad..8dd819e 100644 > --- a/include/uapi/linux/netfilter/nfnetlink_queue.h > +++ b/include/uapi/linux/netfilter/nfnetlink_queue.h > @@ -47,6 +47,8 @@ enum nfqnl_attr_type { > NFQA_CAP_LEN, /* __u32 length of captured packet */ > NFQA_SKB_INFO, /* __u32 skb meta information */ > NFQA_EXP, /* nf_conntrack_netlink.h */ > + NFQA_UID, /* __u32 sk uid */ > + NFQA_GID, /* __u32 sk gid */ > =20 > __NFQA_MAX > }; > @@ -99,7 +101,8 @@ enum nfqnl_attr_config { > #define NFQA_CFG_F_FAIL_OPEN (1 << 0) > #define NFQA_CFG_F_CONNTRACK (1 << 1) > #define NFQA_CFG_F_GSO (1 << 2) > -#define NFQA_CFG_F_MAX (1 << 3) > +#define NFQA_CFG_F_UID_GID (1 << 3) > +#define NFQA_CFG_F_MAX (1 << 4) > =20 > /* flags for NFQA_SKB_INFO */ > /* packet appears to have wrong checksums, but they are ok */ > diff --git a/net/netfilter/nfnetlink_queue_core.c b/net/netfilter/nfn= etlink_queue_core.c > index 21258cf..2cdef07 100644 > --- a/net/netfilter/nfnetlink_queue_core.c > +++ b/net/netfilter/nfnetlink_queue_core.c > @@ -297,6 +297,29 @@ nfqnl_put_packet_info(struct sk_buff *nlskb, str= uct sk_buff *packet, > return flags ? nla_put_be32(nlskb, NFQA_SKB_INFO, htonl(flags)) : 0= ; > } > =20 > +static int nfqnl_put_sk_uidgid(struct sk_buff *skb, struct sock *sk) > +{ > + const struct cred *cred; > + > + if (sk && sk->sk_state !=3D TCP_TIME_WAIT) { > + read_lock_bh(&sk->sk_callback_lock); > + if (sk->sk_socket && sk->sk_socket->file) { > + cred =3D sk->sk_socket->file->f_cred; > + if (nla_put_u32(skb, NFQA_UID, htonl(cred->fsuid))) I'm hitting this compilation error here: net/netfilter/nfnetlink_queue_core.c: In function =E2=80=98nfqnl_put_sk= _uidgid=E2=80=99: net/netfilter/nfnetlink_queue_core.c:308:4: error: aggregate value used= where an integer was expected It seems that in recent kernels you have to use: from_kuid_munged(&init_user_ns, cred->fsuid); to retrieve the integer that represents the uid. Please,=20 > + goto nla_put_failure; > + if (nla_put_u32(skb, NFQA_GID, htonl(cred->fsgid))) ^^^ Minor nitpick: This has to be _be32 otherwise sparse will complain (spotted by Florian Westphal). Make sure you patch is also clean when running make C=3D2. Also remove the extra checking for socket that Florian mentioned and resubmit. Thanks.