From patchwork Wed Nov 11 22:25:43 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Borkmann X-Patchwork-Id: 543119 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id E21B41413F0 for ; Thu, 12 Nov 2015 09:26:06 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753117AbbKKWZw (ORCPT ); Wed, 11 Nov 2015 17:25:52 -0500 Received: from www62.your-server.de ([213.133.104.62]:57178 "EHLO www62.your-server.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753401AbbKKWZv (ORCPT ); Wed, 11 Nov 2015 17:25:51 -0500 Received: from [178.194.100.72] (helo=localhost) by www62.your-server.de with esmtpsa (TLSv1.2:DHE-RSA-AES128-GCM-SHA256:128) (Exim 4.80.1) (envelope-from ) id 1Zwdpt-0000MS-Uk; Wed, 11 Nov 2015 23:25:50 +0100 From: Daniel Borkmann To: davem@davemloft.net Cc: edumazet@google.com, willemb@google.com, tklauser@distanz.ch, netdev@vger.kernel.org, Daniel Borkmann Subject: [PATCH net v3 4/5] packet: infer protocol from ethernet header if unset Date: Wed, 11 Nov 2015 23:25:43 +0100 Message-Id: <1ff9ac2b71fd4d65bec94db62e3750c88ada5d80.1447278504.git.daniel@iogearbox.net> X-Mailer: git-send-email 1.9.3 In-Reply-To: References: In-Reply-To: References: X-Authenticated-Sender: daniel@iogearbox.net X-Virus-Scanned: Clear (ClamAV 0.98.7/21054/Wed Nov 11 18:35:29 2015) Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org In case no struct sockaddr_ll has been passed to packet socket's sendmsg() when doing a TX_RING flush run, then skb->protocol is set to po->num instead, which is the protocol passed via socket(2)/bind(2). Applications only xmitting can go the path of allocating the socket as socket(PF_PACKET, , 0) and do a bind(2) on the TX_RING with sll_protocol of 0. That way, register_prot_hook() is neither called on creation nor on bind time, which saves cycles when there's no interest in capturing anyway. That leaves us however with po->num 0 instead and therefore the TX_RING flush run sets skb->protocol to 0 as well. Eric reported that this leads to problems when using tools like trafgen over bonding device. I.e. the bonding's hash function could invoke the kernel's flow dissector, which depends on skb->protocol being properly set. In the current situation, all the traffic is then directed to a single slave. Fix it up by inferring skb->protocol from the Ethernet header when not set and we have ARPHRD_ETHER device type. This is only done in case of SOCK_RAW and where we have a dev->hard_header_len length. In case of ARPHRD_ETHER devices, this is guaranteed to cover ETH_HLEN, and therefore being accessed on the skb after the skb_store_bits(). Reported-by: Eric Dumazet Signed-off-by: Daniel Borkmann Acked-by: Willem de Bruijn --- net/packet/af_packet.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 8795b0f..0066da2 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2338,6 +2338,15 @@ static bool ll_header_truncated(const struct net_device *dev, int len) return false; } +static void tpacket_set_protocol(const struct net_device *dev, + struct sk_buff *skb) +{ + if (dev->type == ARPHRD_ETHER) { + skb_reset_mac_header(skb); + skb->protocol = eth_hdr(skb)->h_proto; + } +} + static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb, void *frame, struct net_device *dev, int size_max, __be16 proto, unsigned char *addr, int hlen) @@ -2419,6 +2428,8 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb, dev->hard_header_len); if (unlikely(err)) return err; + if (!skb->protocol) + tpacket_set_protocol(dev, skb); data += dev->hard_header_len; to_write -= dev->hard_header_len;