: VLAN vs. packet socket inconsistencies

index d36515d..e8360a2 100644
@@ -105,17 +105,8 @@ static inline void vlan_group_set_device(struct vlan_group *vg,
 	array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN] = dev;
 }
 
-/* VLAN tx hw acceleration helpers. */
-struct vlan_skb_tx_cookie {
-	u32	magic;
-	u32	vlan_tag;
-};
-
-#define VLAN_TX_COOKIE_MAGIC	0x564c414e	/* "VLAN" in ascii. */
-#define VLAN_TX_SKB_CB(__skb)	((struct vlan_skb_tx_cookie *)&((__skb)->cb[0]))
-#define vlan_tx_tag_present(__skb) \
-	(VLAN_TX_SKB_CB(__skb)->magic == VLAN_TX_COOKIE_MAGIC)
-#define vlan_tx_tag_get(__skb)	(VLAN_TX_SKB_CB(__skb)->vlan_tag)
+#define vlan_tx_tag_present(__skb)	((__skb)->vlan_tci)
+#define vlan_tx_tag_get(__skb)		((__skb)->vlan_tci)
 
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
 extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);

@@ -222,17 +213,12 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
  * @skb: skbuff to tag
  * @vlan_tci: VLAN TCI to insert
  *
- * Puts the VLAN TCI in @skb->cb[] and lets the device do the rest
+ * Puts the VLAN TCI in @skb->vlan_tci and lets the device do the rest
  */
 static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb,
 						     u16 vlan_tci)
 {
-	struct vlan_skb_tx_cookie *cookie;
-
-	cookie = VLAN_TX_SKB_CB(skb);
-	cookie->magic = VLAN_TX_COOKIE_MAGIC;
-	cookie->vlan_tag = vlan_tci;
-
+	skb->vlan_tci = vlan_tci;
 	return skb;
 }
 

@@ -279,16 +265,13 @@ static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
  * @skb: skbuff to query
  * @vlan_tci: buffer to store vlaue
  *
- * Returns error if @skb->cb[] is not set correctly
+ * Returns error if @skb->vlan_tci is not set correctly
  */
 static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
 					 u16 *vlan_tci)
 {
-	struct vlan_skb_tx_cookie *cookie;
-
-	cookie = VLAN_TX_SKB_CB(skb);
-	if (cookie->magic == VLAN_TX_COOKIE_MAGIC) {
-		*vlan_tci = cookie->vlan_tag;
+	if (vlan_tx_tag_present(skb)) {
+		*vlan_tci = skb->vlan_tci;
 		return 0;
 	} else {
 		*vlan_tci = 0;

index 2220b9e..14a3273 100644
@@ -246,6 +246,7 @@ typedef unsigned char *sk_buff_data_t;
  *	@dma_cookie: a cookie to one of several possible DMA operations
  *		done by skb DMA functions
  *	@secmark: security marking
+ *	@vlan_tci: vlan tag control information
  */
 
 struct sk_buff {

@@ -328,6 +329,8 @@ struct sk_buff {
 
 	__u32			mark;
 
+	__u16			vlan_tci;
+
 	sk_buff_data_t		transport_header;
 	sk_buff_data_t		network_header;
 	sk_buff_data_t		mac_header;

index 7c57156..50a853f 100644
@@ -459,6 +459,8 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
 	new->tc_verd		= old->tc_verd;
 #endif
 #endif
+	new->vlan_tci		= old->vlan_tci;
+
 	skb_copy_secmark(new, old);
 }
 

@@ -2286,6 +2288,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features)
 		skb_copy_queue_mapping(nskb, skb);
 		nskb->priority = skb->priority;
 		nskb->protocol = skb->protocol;
+		nskb->vlan_tci = skb->vlan_tci;
 		nskb->dst = dst_clone(skb->dst);
 		memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
 		nskb->pkt_type = skb->pkt_type;

index e009c6f..d0469eb 100644
@@ -1145,6 +1145,7 @@ extern int		netif_rx(struct sk_buff *skb);
 extern int		netif_rx_ni(struct sk_buff *skb);
 #define HAVE_NETIF_RECEIVE_SKB 1
 extern int		netif_receive_skb(struct sk_buff *skb);
+extern void		netif_nit_deliver(struct sk_buff *skb);
 extern int		dev_valid_name(const char *name);
 extern int		dev_ioctl(struct net *net, unsigned int cmd, void __user *);
 extern int		dev_ethtool(struct net *net, struct ifreq *);

index 2bec74c..9a66e80 100644
@@ -14,6 +14,9 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_DROP;
 	}
 
+	skb->vlan_tci = vlan_tci;
+	netif_nit_deliver(skb);
+
 	skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
 	if (skb->dev == NULL) {
 		dev_kfree_skb_any(skb);

@@ -22,6 +25,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_SUCCESS;
 	}
 	skb->dev->last_rx = jiffies;
+	skb->vlan_tci = 0;
 
 	stats = &skb->dev->stats;
 	stats->rx_packets++;

index 7593393..96f1528 100644
@@ -2060,6 +2060,33 @@ out:
 }
 #endif
 
+/*
+ * 	netif_tap_deliver - deliver received packets to network taps
+ * 	@skb: buffer
+ *
+ * 	This function is used to deliver incoming packets to network
+ * 	taps. It should be used when the normal netif_receive_skb path
+ * 	is bypassed, for example because of VLAN acceleration.
+ */
+void netif_nit_deliver(struct sk_buff *skb)
+{
+	struct packet_type *ptype;
+
+	if (list_empty(&ptype_all))
+		return;
+
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	skb->mac_len = skb->network_header - skb->mac_header;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(ptype, &ptype_all, list) {
+		if (!ptype->dev || ptype->dev == skb->dev)
+			deliver_skb(skb, ptype, skb->dev);
+	}
+	rcu_read_unlock();
+}
+
 /**
  *	netif_receive_skb - process receive buffer from network
  *	@skb: buffer to process

index ad09609..4e1fc2a 100644
@@ -57,6 +57,7 @@ struct tpacket_auxdata
 	__u32		tp_snaplen;
 	__u16		tp_mac;
 	__u16		tp_net;
+	__u16		tp_vlan_tci;
 };
 
 struct tpacket_hdr

index beca640..951e62e 100644
@@ -1107,6 +1107,7 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock,
 		aux.tp_snaplen = skb->len;
 		aux.tp_mac = 0;
 		aux.tp_net = skb_network_offset(skb);
+		aux.tp_vlan_tci = skb->vlan_tci;
 
 		put_cmsg(msg, SOL_PACKET, PACKET_AUXDATA, sizeof(aux), &aux);
 	}

index e9db010..e877cd8 100644
@@ -471,7 +471,13 @@ pcap_read_packet(pcap_t *handle, pcap_handler callback, u_char *userdata)
 	socklen_t		fromlen;
 	int			packet_len, caplen;
 	struct pcap_pkthdr	pcap_header;
-
+	struct iovec		iov;
+	struct msghdr		msg;
+	struct cmsghdr		*cmsg;
+	union {
+		struct cmsghdr	cmsg;
+		char		buf[CMSG_SPACE(sizeof(struct tpacket_auxdata))];
+	} cmsg_buf;
 #ifdef HAVE_PF_PACKET_SOCKETS
 	/*
 	 * If this is a cooked device, leave extra room for a

@@ -492,6 +498,15 @@ pcap_read_packet(pcap_t *handle, pcap_handler callback, u_char *userdata)
 	/* Receive a single packet from the kernel */
 
 	bp = handle->buffer + handle->offset;
+
+	msg.msg_name		= &from;
+	msg.msg_namelen		= sizeof(from);
+	msg.msg_iov		= &iov;
+	msg.msg_iovlen		= 1;
+	msg.msg_control		= &cmsg_buf;
+	msg.msg_controllen	= sizeof(cmsg_buf);
+	msg.msg_flags		= 0;
+
 	do {
 		/*
 		 * Has "pcap_breakloop()" been called?

@@ -505,11 +520,11 @@ pcap_read_packet(pcap_t *handle, pcap_handler callback, u_char *userdata)
 			handle->break_loop = 0;
 			return -2;
 		}
-		fromlen = sizeof(from);
-		packet_len = recvfrom(
-			handle->fd, bp + offset,
-			handle->bufsize - offset, MSG_TRUNC,
-			(struct sockaddr *) &from, &fromlen);
+
+		iov.iov_len	= handle->bufsize - offset;
+		iov.iov_base	= bp + offset;
+
+		packet_len = recvmsg(handle->fd, &msg, MSG_TRUNC);
 	} while (packet_len == -1 && errno == EINTR);
 
 	/* Check if an error occured */

@@ -524,6 +539,38 @@ pcap_read_packet(pcap_t *handle, pcap_handler callback, u_char *userdata)
 		}
 	}
 
+	for (cmsg = CMSG_FIRSTHDR(&msg); cmsg; cmsg = CMSG_NXTHDR(&msg, cmsg)) {
+		struct tpacket_auxdata *aux;
+		unsigned int len, copy;
+		unsigned short *ptr;
+
+		if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct tpacket_auxdata)) ||
+		    cmsg->cmsg_level != SOL_PACKET ||
+		    cmsg->cmsg_type != PACKET_AUXDATA)
+			continue;
+
+		aux = (struct tpacket_auxdata *)CMSG_DATA(cmsg);
+		if (aux->tp_vlan_tci == 0)
+			continue;
+
+		len = packet_len > iov.iov_len ? iov.iov_len : packet_len;
+		if (len > 2 * ETH_ALEN + 4) {
+			copy = len - 2 * ETH_ALEN - 4;
+			if (copy > iov.iov_len - 2 * ETH_ALEN - 4)
+				copy = iov.iov_len - 2 * ETH_ALEN - 4;
+
+			memmove(iov.iov_base + 2 * ETH_ALEN + 4,
+				iov.iov_base + 2 * ETH_ALEN, copy);
+		}
+
+		ptr = (unsigned short *)(iov.iov_base + 2 * ETH_ALEN);
+		if (len >= 2 * ETH_ALEN + 2)
+			*(ptr++) = htons(ETH_P_8021Q);
+		if (len >= 2 * ETH_ALEN + 4)
+			*(ptr++) = htons(aux->tp_vlan_tci);
+		packet_len += 4;
+	}
+
 #ifdef HAVE_PF_PACKET_SOCKETS
 	if (!handle->md.sock_packet) {
 		/*

@@ -1631,6 +1678,7 @@ iface_bind(int fd, int ifindex, char *ebuf)
 	struct sockaddr_ll	sll;
 	int			err;
 	socklen_t		errlen = sizeof(err);
+	int			val;
 
 	memset(&sll, 0, sizeof(sll));
 	sll.sll_family		= AF_PACKET;

@@ -1657,6 +1705,12 @@ iface_bind(int fd, int ifindex, char *ebuf)
 		return -2;
 	}
 
+	val = 1;
+	if (setsockopt(fd, SOL_PACKET, PACKET_AUXDATA, &val, sizeof(val)) == -1) {
+		snprintf(ebuf, PCAP_ERRBUF_SIZE,
+			 "setsockopt: %s", pcap_strerror(errno));
+		return -3;
+	}
 	return 0;
 }