Merge branch 'ovs-802.1ad'

Eric Garver says:

====================
openvswitch: add 802.1ad support

This series adds 802.1ad support to openvswitch. It is a continuation of the
work originally started by Thomas F Herbert - hence the large rev number.

The extra VLAN is implemented by using an additional level of the
OVS_KEY_ATTR_ENCAP netlink attribute.
In OVS flow speak, this looks like

   eth_type(0x88a8),vlan(vid=100),encap(eth_type(0x8100), vlan(vid=200),
                                        encap(eth_type(0x0800), ...))

The userspace counterpart has also seen recent activity on the ovs-dev mailing
lists. There are some new 802.1ad OVS tests being added - also on the ovs-dev
list. This patch series has been tested using the most recent version of
userspace (v3) and tests (v2).

v22 changes:
  - merge patch 4 into patch 3
  - fix checkpatch.pl errors
    - Still some 80 char warnings for long string literals
  - refresh pointer after pskb_may_pull()
  - refactor vlan nlattr parsing to remove some double checks
  - introduce ovs_nla_put_vlan()
  - move triple VLAN check to after ethertype serialization
  - WARN_ON_ONCE() on triple VLAN and unexpected encap values

v21 changes:
  - Fix (and simplify) netlink attribute parsing
  - re-add handling of truncated VLAN tags
  - fix if/else dangling assignment in {push,pop}_vlan()
  - simplify parse_vlan()
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/if_vlan.h

@@ -272,6 +272,23 @@ static inline int vlan_get_encap_level(struct net_device *dev)
 }
 #endif
 
+/**
+ * eth_type_vlan - check for valid vlan ether type.
+ * @ethertype: ether type to check
+ *
+ * Returns true if the ether type is a vlan ether type.
+ */
+static inline bool eth_type_vlan(__be16 ethertype)
+{
+	switch (ethertype) {
+	case htons(ETH_P_8021Q):
+	case htons(ETH_P_8021AD):
+		return true;
+	default:
+		return false;
+	}
+}
+
 static inline bool vlan_hw_offload_capable(netdev_features_t features,
 					   __be16 proto)
 {
@@ -425,8 +442,7 @@ static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
 {
 	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)skb->data;
 
-	if (veth->h_vlan_proto != htons(ETH_P_8021Q) &&
+	if (!eth_type_vlan(veth->h_vlan_proto))
-	    veth->h_vlan_proto != htons(ETH_P_8021AD))
 		return -EINVAL;
 
 	*vlan_tci = ntohs(veth->h_vlan_TCI);
@@ -488,7 +504,7 @@ static inline __be16 __vlan_get_protocol(struct sk_buff *skb, __be16 type,
 	 * present at mac_len - VLAN_HLEN (if mac_len > 0), or at
 	 * ETH_HLEN otherwise
 	 */
-	if (type == htons(ETH_P_8021Q) || type == htons(ETH_P_8021AD)) {
+	if (eth_type_vlan(type)) {
 		if (vlan_depth) {
 			if (WARN_ON(vlan_depth < VLAN_HLEN))
 				return 0;
@@ -506,8 +522,7 @@ static inline __be16 __vlan_get_protocol(struct sk_buff *skb, __be16 type,
 			vh = (struct vlan_hdr *)(skb->data + vlan_depth);
 			type = vh->h_vlan_encapsulated_proto;
 			vlan_depth += VLAN_HLEN;
-		} while (type == htons(ETH_P_8021Q) ||
+		} while (eth_type_vlan(type));
-			 type == htons(ETH_P_8021AD));
 	}
 
 	if (depth)
@@ -572,8 +587,7 @@ static inline void vlan_set_encap_proto(struct sk_buff *skb,
 static inline bool skb_vlan_tagged(const struct sk_buff *skb)
 {
 	if (!skb_vlan_tag_present(skb) &&
-	    likely(skb->protocol != htons(ETH_P_8021Q) &&
+	    likely(!eth_type_vlan(skb->protocol)))
-		   skb->protocol != htons(ETH_P_8021AD)))
 		return false;
 
 	return true;
@@ -593,15 +607,14 @@ static inline bool skb_vlan_tagged_multi(const struct sk_buff *skb)
 	if (!skb_vlan_tag_present(skb)) {
 		struct vlan_ethhdr *veh;
 
-		if (likely(protocol != htons(ETH_P_8021Q) &&
+		if (likely(!eth_type_vlan(protocol)))
-			   protocol != htons(ETH_P_8021AD)))
 			return false;
 
 		veh = (struct vlan_ethhdr *)skb->data;
 		protocol = veh->h_vlan_encapsulated_proto;
 	}
 
-	if (protocol != htons(ETH_P_8021Q) && protocol != htons(ETH_P_8021AD))
+	if (!eth_type_vlan(protocol))
 		return false;
 
 	return true;

include/uapi/linux/openvswitch.h

@@ -605,13 +605,13 @@ struct ovs_action_push_mpls {
  * @vlan_tci: Tag control identifier (TCI) to push.  The CFI bit must be set
  * (but it will not be set in the 802.1Q header that is pushed).
  *
- * The @vlan_tpid value is typically %ETH_P_8021Q.  The only acceptable TPID
+ * The @vlan_tpid value is typically %ETH_P_8021Q or %ETH_P_8021AD.
- * values are those that the kernel module also parses as 802.1Q headers, to
+ * The only acceptable TPID values are those that the kernel module also parses
- * prevent %OVS_ACTION_ATTR_PUSH_VLAN followed by %OVS_ACTION_ATTR_POP_VLAN
+ * as 802.1Q or 802.1AD headers, to prevent %OVS_ACTION_ATTR_PUSH_VLAN followed
- * from having surprising results.
+ * by %OVS_ACTION_ATTR_POP_VLAN from having surprising results.
  */
 struct ovs_action_push_vlan {
-	__be16 vlan_tpid;	/* 802.1Q TPID. */
+	__be16 vlan_tpid;	/* 802.1Q or 802.1ad TPID. */
 	__be16 vlan_tci;	/* 802.1Q TCI (VLAN ID and priority). */
 };
 
@@ -721,9 +721,10 @@ enum ovs_nat_attr {
  * is copied from the value to the packet header field, rest of the bits are
  * left unchanged.  The non-masked value bits must be passed in as zeroes.
  * Masking is not supported for the %OVS_KEY_ATTR_TUNNEL attribute.
- * @OVS_ACTION_ATTR_PUSH_VLAN: Push a new outermost 802.1Q header onto the
+ * @OVS_ACTION_ATTR_PUSH_VLAN: Push a new outermost 802.1Q or 802.1ad header
- * packet.
+ * onto the packet.
- * @OVS_ACTION_ATTR_POP_VLAN: Pop the outermost 802.1Q header off the packet.
+ * @OVS_ACTION_ATTR_POP_VLAN: Pop the outermost 802.1Q or 802.1ad header
+ * from the packet.
  * @OVS_ACTION_ATTR_SAMPLE: Probabilitically executes actions, as specified in
  * the nested %OVS_SAMPLE_ATTR_* attributes.
  * @OVS_ACTION_ATTR_PUSH_MPLS: Push a new MPLS label stack entry onto the

net/openvswitch/actions.c

@@ -246,20 +246,24 @@ static int pop_vlan(struct sk_buff *skb, struct sw_flow_key *key)
 	int err;
 
 	err = skb_vlan_pop(skb);
-	if (skb_vlan_tag_present(skb))
+	if (skb_vlan_tag_present(skb)) {
 		invalidate_flow_key(key);
-	else
+	} else {
-		key->eth.tci = 0;
+		key->eth.vlan.tci = 0;
+		key->eth.vlan.tpid = 0;
+	}
 	return err;
 }
 
 static int push_vlan(struct sk_buff *skb, struct sw_flow_key *key,
 		     const struct ovs_action_push_vlan *vlan)
 {
-	if (skb_vlan_tag_present(skb))
+	if (skb_vlan_tag_present(skb)) {
 		invalidate_flow_key(key);
-	else
+	} else {
-		key->eth.tci = vlan->vlan_tci;
+		key->eth.vlan.tci = vlan->vlan_tci;
+		key->eth.vlan.tpid = vlan->vlan_tpid;
+	}
 	return skb_vlan_push(skb, vlan->vlan_tpid,
 			     ntohs(vlan->vlan_tci) & ~VLAN_TAG_PRESENT);
 }

net/openvswitch/flow.c

@@ -302,24 +302,57 @@ static bool icmp6hdr_ok(struct sk_buff *skb)
 				  sizeof(struct icmp6hdr));
 }
 
-static int parse_vlan(struct sk_buff *skb, struct sw_flow_key *key)
+/**
+ * Parse vlan tag from vlan header.
+ * Returns ERROR on memory error.
+ * Returns 0 if it encounters a non-vlan or incomplete packet.
+ * Returns 1 after successfully parsing vlan tag.
+ */
+static int parse_vlan_tag(struct sk_buff *skb, struct vlan_head *key_vh)
 {
-	struct qtag_prefix {
+	struct vlan_head *vh = (struct vlan_head *)skb->data;
-		__be16 eth_type; /* ETH_P_8021Q */
-		__be16 tci;
-	};
-	struct qtag_prefix *qp;
 
-	if (unlikely(skb->len < sizeof(struct qtag_prefix) + sizeof(__be16)))
+	if (likely(!eth_type_vlan(vh->tpid)))
 		return 0;
 
-	if (unlikely(!pskb_may_pull(skb, sizeof(struct qtag_prefix) +
+	if (unlikely(skb->len < sizeof(struct vlan_head) + sizeof(__be16)))
-					 sizeof(__be16))))
+		return 0;
+
+	if (unlikely(!pskb_may_pull(skb, sizeof(struct vlan_head) +
+				 sizeof(__be16))))
 		return -ENOMEM;
 
-	qp = (struct qtag_prefix *) skb->data;
+	vh = (struct vlan_head *)skb->data;
-	key->eth.tci = qp->tci | htons(VLAN_TAG_PRESENT);
+	key_vh->tci = vh->tci | htons(VLAN_TAG_PRESENT);
-	__skb_pull(skb, sizeof(struct qtag_prefix));
+	key_vh->tpid = vh->tpid;
+
+	__skb_pull(skb, sizeof(struct vlan_head));
+	return 1;
+}
+
+static int parse_vlan(struct sk_buff *skb, struct sw_flow_key *key)
+{
+	int res;
+
+	key->eth.vlan.tci = 0;
+	key->eth.vlan.tpid = 0;
+	key->eth.cvlan.tci = 0;
+	key->eth.cvlan.tpid = 0;
+
+	if (likely(skb_vlan_tag_present(skb))) {
+		key->eth.vlan.tci = htons(skb->vlan_tci);
+		key->eth.vlan.tpid = skb->vlan_proto;
+	} else {
+		/* Parse outer vlan tag in the non-accelerated case. */
+		res = parse_vlan_tag(skb, &key->eth.vlan);
+		if (res <= 0)
+			return res;
+	}
+
+	/* Parse inner vlan tag. */
+	res = parse_vlan_tag(skb, &key->eth.cvlan);
+	if (res <= 0)
+		return res;
 
 	return 0;
 }
@@ -480,12 +513,8 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
 	 * update skb->csum here.
 	 */
 
-	key->eth.tci = 0;
+	if (unlikely(parse_vlan(skb, key)))
-	if (skb_vlan_tag_present(skb))
+		return -ENOMEM;
-		key->eth.tci = htons(skb->vlan_tci);
-	else if (eth->h_proto == htons(ETH_P_8021Q))
-		if (unlikely(parse_vlan(skb, key)))
-			return -ENOMEM;
 
 	key->eth.type = parse_ethertype(skb);
 	if (unlikely(key->eth.type == htons(0)))

net/openvswitch/flow.h

@@ -50,6 +50,11 @@ struct ovs_tunnel_info {
 	struct metadata_dst	*tun_dst;
 };
 
+struct vlan_head {
+	__be16 tpid; /* Vlan type. Generally 802.1q or 802.1ad.*/
+	__be16 tci;  /* 0 if no VLAN, VLAN_TAG_PRESENT set otherwise. */
+};
+
 #define OVS_SW_FLOW_KEY_METADATA_SIZE			\
 	(offsetof(struct sw_flow_key, recirc_id) +	\
 	FIELD_SIZEOF(struct sw_flow_key, recirc_id))
@@ -69,7 +74,8 @@ struct sw_flow_key {
 	struct {
 		u8     src[ETH_ALEN];	/* Ethernet source address. */
 		u8     dst[ETH_ALEN];	/* Ethernet destination address. */
-		__be16 tci;		/* 0 if no VLAN, VLAN_TAG_PRESENT set otherwise. */
+		struct vlan_head vlan;
+		struct vlan_head cvlan;
 		__be16 type;		/* Ethernet frame type. */
 	} eth;
 	union {

net/openvswitch/flow_netlink.c

@@ -808,6 +808,167 @@ int ovs_nla_put_tunnel_info(struct sk_buff *skb,
 				  ip_tunnel_info_af(tun_info));
 }
 
+static int encode_vlan_from_nlattrs(struct sw_flow_match *match,
+				    const struct nlattr *a[],
+				    bool is_mask, bool inner)
+{
+	__be16 tci = 0;
+	__be16 tpid = 0;
+
+	if (a[OVS_KEY_ATTR_VLAN])
+		tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
+
+	if (a[OVS_KEY_ATTR_ETHERTYPE])
+		tpid = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);
+
+	if (likely(!inner)) {
+		SW_FLOW_KEY_PUT(match, eth.vlan.tpid, tpid, is_mask);
+		SW_FLOW_KEY_PUT(match, eth.vlan.tci, tci, is_mask);
+	} else {
+		SW_FLOW_KEY_PUT(match, eth.cvlan.tpid, tpid, is_mask);
+		SW_FLOW_KEY_PUT(match, eth.cvlan.tci, tci, is_mask);
+	}
+	return 0;
+}
+
+static int validate_vlan_from_nlattrs(const struct sw_flow_match *match,
+				      u64 key_attrs, bool inner,
+				      const struct nlattr **a, bool log)
+{
+	__be16 tci = 0;
+
+	if (!((key_attrs & (1 << OVS_KEY_ATTR_ETHERNET)) &&
+	      (key_attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) &&
+	       eth_type_vlan(nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE])))) {
+		/* Not a VLAN. */
+		return 0;
+	}
+
+	if (!((key_attrs & (1 << OVS_KEY_ATTR_VLAN)) &&
+	      (key_attrs & (1 << OVS_KEY_ATTR_ENCAP)))) {
+		OVS_NLERR(log, "Invalid %s frame", (inner) ? "C-VLAN" : "VLAN");
+		return -EINVAL;
+	}
+
+	if (a[OVS_KEY_ATTR_VLAN])
+		tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
+
+	if (!(tci & htons(VLAN_TAG_PRESENT))) {
+		if (tci) {
+			OVS_NLERR(log, "%s TCI does not have VLAN_TAG_PRESENT bit set.",
+				  (inner) ? "C-VLAN" : "VLAN");
+			return -EINVAL;
+		} else if (nla_len(a[OVS_KEY_ATTR_ENCAP])) {
+			/* Corner case for truncated VLAN header. */
+			OVS_NLERR(log, "Truncated %s header has non-zero encap attribute.",
+				  (inner) ? "C-VLAN" : "VLAN");
+			return -EINVAL;
+		}
+	}
+
+	return 1;
+}
+
+static int validate_vlan_mask_from_nlattrs(const struct sw_flow_match *match,
+					   u64 key_attrs, bool inner,
+					   const struct nlattr **a, bool log)
+{
+	__be16 tci = 0;
+	__be16 tpid = 0;
+	bool encap_valid = !!(match->key->eth.vlan.tci &
+			      htons(VLAN_TAG_PRESENT));
+	bool i_encap_valid = !!(match->key->eth.cvlan.tci &
+				htons(VLAN_TAG_PRESENT));
+
+	if (!(key_attrs & (1 << OVS_KEY_ATTR_ENCAP))) {
+		/* Not a VLAN. */
+		return 0;
+	}
+
+	if ((!inner && !encap_valid) || (inner && !i_encap_valid)) {
+		OVS_NLERR(log, "Encap mask attribute is set for non-%s frame.",
+			  (inner) ? "C-VLAN" : "VLAN");
+		return -EINVAL;
+	}
+
+	if (a[OVS_KEY_ATTR_VLAN])
+		tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
+
+	if (a[OVS_KEY_ATTR_ETHERTYPE])
+		tpid = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);
+
+	if (tpid != htons(0xffff)) {
+		OVS_NLERR(log, "Must have an exact match on %s TPID (mask=%x).",
+			  (inner) ? "C-VLAN" : "VLAN", ntohs(tpid));
+		return -EINVAL;
+	}
+	if (!(tci & htons(VLAN_TAG_PRESENT))) {
+		OVS_NLERR(log, "%s TCI mask does not have exact match for VLAN_TAG_PRESENT bit.",
+			  (inner) ? "C-VLAN" : "VLAN");
+		return -EINVAL;
+	}
+
+	return 1;
+}
+
+static int __parse_vlan_from_nlattrs(struct sw_flow_match *match,
+				     u64 *key_attrs, bool inner,
+				     const struct nlattr **a, bool is_mask,
+				     bool log)
+{
+	int err;
+	const struct nlattr *encap;
+
+	if (!is_mask)
+		err = validate_vlan_from_nlattrs(match, *key_attrs, inner,
+						 a, log);
+	else
+		err = validate_vlan_mask_from_nlattrs(match, *key_attrs, inner,
+						      a, log);
+	if (err <= 0)
+		return err;
+
+	err = encode_vlan_from_nlattrs(match, a, is_mask, inner);
+	if (err)
+		return err;
+
+	*key_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP);
+	*key_attrs &= ~(1 << OVS_KEY_ATTR_VLAN);
+	*key_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE);
+
+	encap = a[OVS_KEY_ATTR_ENCAP];
+
+	if (!is_mask)
+		err = parse_flow_nlattrs(encap, a, key_attrs, log);
+	else
+		err = parse_flow_mask_nlattrs(encap, a, key_attrs, log);
+
+	return err;
+}
+
+static int parse_vlan_from_nlattrs(struct sw_flow_match *match,
+				   u64 *key_attrs, const struct nlattr **a,
+				   bool is_mask, bool log)
+{
+	int err;
+	bool encap_valid = false;
+
+	err = __parse_vlan_from_nlattrs(match, key_attrs, false, a,
+					is_mask, log);
+	if (err)
+		return err;
+
+	encap_valid = !!(match->key->eth.vlan.tci & htons(VLAN_TAG_PRESENT));
+	if (encap_valid) {
+		err = __parse_vlan_from_nlattrs(match, key_attrs, true, a,
+						is_mask, log);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 static int metadata_from_nlattrs(struct net *net, struct sw_flow_match *match,
 				 u64 *attrs, const struct nlattr **a,
 				 bool is_mask, bool log)
@@ -923,20 +1084,11 @@ static int ovs_key_from_nlattrs(struct net *net, struct sw_flow_match *match,
 	}
 
 	if (attrs & (1 << OVS_KEY_ATTR_VLAN)) {
-		__be16 tci;
+		/* VLAN attribute is always parsed before getting here since it
-
+		 * may occur multiple times.
-		tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
+		 */
-		if (!(tci & htons(VLAN_TAG_PRESENT))) {
+		OVS_NLERR(log, "VLAN attribute unexpected.");
-			if (is_mask)
+		return -EINVAL;
-				OVS_NLERR(log, "VLAN TCI mask does not have exact match for VLAN_TAG_PRESENT bit.");
-			else
-				OVS_NLERR(log, "VLAN TCI does not have VLAN_TAG_PRESENT bit set.");
-
-			return -EINVAL;
-		}
-
-		SW_FLOW_KEY_PUT(match, eth.tci, tci, is_mask);
-		attrs &= ~(1 << OVS_KEY_ATTR_VLAN);
 	}
 
 	if (attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) {
@@ -1182,49 +1334,18 @@ int ovs_nla_get_match(struct net *net, struct sw_flow_match *match,
 		      bool log)
 {
 	const struct nlattr *a[OVS_KEY_ATTR_MAX + 1];
-	const struct nlattr *encap;
 	struct nlattr *newmask = NULL;
 	u64 key_attrs = 0;
 	u64 mask_attrs = 0;
-	bool encap_valid = false;
 	int err;
 
 	err = parse_flow_nlattrs(nla_key, a, &key_attrs, log);
 	if (err)
 		return err;
 
-	if ((key_attrs & (1 << OVS_KEY_ATTR_ETHERNET)) &&
+	err = parse_vlan_from_nlattrs(match, &key_attrs, a, false, log);
-	    (key_attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) &&
+	if (err)
-	    (nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]) == htons(ETH_P_8021Q))) {
+		return err;
-		__be16 tci;
-
-		if (!((key_attrs & (1 << OVS_KEY_ATTR_VLAN)) &&
-		      (key_attrs & (1 << OVS_KEY_ATTR_ENCAP)))) {
-			OVS_NLERR(log, "Invalid Vlan frame.");
-			return -EINVAL;
-		}
-
-		key_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE);
-		tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
-		encap = a[OVS_KEY_ATTR_ENCAP];
-		key_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP);
-		encap_valid = true;
-
-		if (tci & htons(VLAN_TAG_PRESENT)) {
-			err = parse_flow_nlattrs(encap, a, &key_attrs, log);
-			if (err)
-				return err;
-		} else if (!tci) {
-			/* Corner case for truncated 802.1Q header. */
-			if (nla_len(encap)) {
-				OVS_NLERR(log, "Truncated 802.1Q header has non-zero encap attribute.");
-				return -EINVAL;
-			}
-		} else {
-			OVS_NLERR(log, "Encap attr is set for non-VLAN frame");
-			return  -EINVAL;
-		}
-	}
 
 	err = ovs_key_from_nlattrs(net, match, key_attrs, a, false, log);
 	if (err)
@@ -1265,46 +1386,12 @@ int ovs_nla_get_match(struct net *net, struct sw_flow_match *match,
 			goto free_newmask;
 
 		/* Always match on tci. */
-		SW_FLOW_KEY_PUT(match, eth.tci, htons(0xffff), true);
+		SW_FLOW_KEY_PUT(match, eth.vlan.tci, htons(0xffff), true);
+		SW_FLOW_KEY_PUT(match, eth.cvlan.tci, htons(0xffff), true);
 
-		if (mask_attrs & 1 << OVS_KEY_ATTR_ENCAP) {
+		err = parse_vlan_from_nlattrs(match, &mask_attrs, a, true, log);
-			__be16 eth_type = 0;
+		if (err)
-			__be16 tci = 0;
+			goto free_newmask;
-
-			if (!encap_valid) {
-				OVS_NLERR(log, "Encap mask attribute is set for non-VLAN frame.");
-				err = -EINVAL;
-				goto free_newmask;
-			}
-
-			mask_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP);
-			if (a[OVS_KEY_ATTR_ETHERTYPE])
-				eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);
-
-			if (eth_type == htons(0xffff)) {
-				mask_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE);
-				encap = a[OVS_KEY_ATTR_ENCAP];
-				err = parse_flow_mask_nlattrs(encap, a,
-							      &mask_attrs, log);
-				if (err)
-					goto free_newmask;
-			} else {
-				OVS_NLERR(log, "VLAN frames must have an exact match on the TPID (mask=%x).",
-					  ntohs(eth_type));
-				err = -EINVAL;
-				goto free_newmask;
-			}
-
-			if (a[OVS_KEY_ATTR_VLAN])
-				tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
-
-			if (!(tci & htons(VLAN_TAG_PRESENT))) {
-				OVS_NLERR(log, "VLAN tag present bit must have an exact match (tci_mask=%x).",
-					  ntohs(tci));
-				err = -EINVAL;
-				goto free_newmask;
-			}
-		}
 
 		err = ovs_key_from_nlattrs(net, match, mask_attrs, a, true,
 					   log);
@@ -1410,12 +1497,25 @@ int ovs_nla_get_flow_metadata(struct net *net, const struct nlattr *attr,
 	return metadata_from_nlattrs(net, &match, &attrs, a, false, log);
 }
 
+static int ovs_nla_put_vlan(struct sk_buff *skb, const struct vlan_head *vh,
+			    bool is_mask)
+{
+	__be16 eth_type = !is_mask ? vh->tpid : htons(0xffff);
+
+	if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, eth_type) ||
+	    nla_put_be16(skb, OVS_KEY_ATTR_VLAN, vh->tci))
+		return -EMSGSIZE;
+	return 0;
+}
+
 static int __ovs_nla_put_key(const struct sw_flow_key *swkey,
 			     const struct sw_flow_key *output, bool is_mask,
 			     struct sk_buff *skb)
 {
 	struct ovs_key_ethernet *eth_key;
-	struct nlattr *nla, *encap;
+	struct nlattr *nla;
+	struct nlattr *encap = NULL;
+	struct nlattr *in_encap = NULL;
 
 	if (nla_put_u32(skb, OVS_KEY_ATTR_RECIRC_ID, output->recirc_id))
 		goto nla_put_failure;
@@ -1464,17 +1564,21 @@ static int __ovs_nla_put_key(const struct sw_flow_key *swkey,
 	ether_addr_copy(eth_key->eth_src, output->eth.src);
 	ether_addr_copy(eth_key->eth_dst, output->eth.dst);
 
-	if (swkey->eth.tci || swkey->eth.type == htons(ETH_P_8021Q)) {
+	if (swkey->eth.vlan.tci || eth_type_vlan(swkey->eth.type)) {
-		__be16 eth_type;
+		if (ovs_nla_put_vlan(skb, &output->eth.vlan, is_mask))
-		eth_type = !is_mask ? htons(ETH_P_8021Q) : htons(0xffff);
-		if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, eth_type) ||
-		    nla_put_be16(skb, OVS_KEY_ATTR_VLAN, output->eth.tci))
 			goto nla_put_failure;
 		encap = nla_nest_start(skb, OVS_KEY_ATTR_ENCAP);
-		if (!swkey->eth.tci)
+		if (!swkey->eth.vlan.tci)
 			goto unencap;
-	} else
+
-		encap = NULL;
+		if (swkey->eth.cvlan.tci || eth_type_vlan(swkey->eth.type)) {
+			if (ovs_nla_put_vlan(skb, &output->eth.cvlan, is_mask))
+				goto nla_put_failure;
+			in_encap = nla_nest_start(skb, OVS_KEY_ATTR_ENCAP);
+			if (!swkey->eth.cvlan.tci)
+				goto unencap;
+		}
+	}
 
 	if (swkey->eth.type == htons(ETH_P_802_2)) {
 		/*
@@ -1493,6 +1597,14 @@ static int __ovs_nla_put_key(const struct sw_flow_key *swkey,
 	if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, output->eth.type))
 		goto nla_put_failure;
 
+	if (eth_type_vlan(swkey->eth.type)) {
+		/* There are 3 VLAN tags, we don't know anything about the rest
+		 * of the packet, so truncate here.
+		 */
+		WARN_ON_ONCE(!(encap && in_encap));
+		goto unencap;
+	}
+
 	if (swkey->eth.type == htons(ETH_P_IP)) {
 		struct ovs_key_ipv4 *ipv4_key;
 
@@ -1619,6 +1731,8 @@ static int __ovs_nla_put_key(const struct sw_flow_key *swkey,
 	}
 
 unencap:
+	if (in_encap)
+		nla_nest_end(skb, in_encap);
 	if (encap)
 		nla_nest_end(skb, encap);
 
@@ -2283,7 +2397,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 
 		case OVS_ACTION_ATTR_PUSH_VLAN:
 			vlan = nla_data(a);
-			if (vlan->vlan_tpid != htons(ETH_P_8021Q))
+			if (!eth_type_vlan(vlan->vlan_tpid))
 				return -EINVAL;
 			if (!(vlan->vlan_tci & htons(VLAN_TAG_PRESENT)))
 				return -EINVAL;
@@ -2388,7 +2502,7 @@ int ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 
 	(*sfa)->orig_len = nla_len(attr);
 	err = __ovs_nla_copy_actions(net, attr, key, 0, sfa, key->eth.type,
-				     key->eth.tci, log);
+				     key->eth.vlan.tci, log);
 	if (err)
 		ovs_nla_free_flow_actions(*sfa);
 

net/openvswitch/vport.c

@@ -485,9 +485,14 @@ static unsigned int packet_length(const struct sk_buff *skb)
 {
 	unsigned int length = skb->len - ETH_HLEN;
 
-	if (skb->protocol == htons(ETH_P_8021Q))
+	if (skb_vlan_tagged(skb))
 		length -= VLAN_HLEN;
 
+	/* Don't subtract for multiple VLAN tags. Most (all?) drivers allow
+	 * (ETH_LEN + VLAN_HLEN) in addition to the mtu value, but almost none
+	 * account for 802.1ad. e.g. is_skb_forwardable().
+	 */
+
 	return length;
 }