net: openvswitch: Add a new action check_pkt_len

This patch adds a new action - 'check_pkt_len' which checks the
packet length and executes a set of actions if the packet
length is greater than the specified length or executes
another set of actions if the packet length is lesser or equal to.

This action takes below nlattrs
* OVS_CHECK_PKT_LEN_ATTR_PKT_LEN - 'pkt_len' to check for

* OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER - Nested actions
to apply if the packet length is greater than the specified 'pkt_len'

* OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL - Nested
actions to apply if the packet length is lesser or equal to the
specified 'pkt_len'.

The main use case for adding this action is to solve the packet
drops because of MTU mismatch in OVN virtual networking solution.
When a VM (which belongs to a logical switch of OVN) sends a packet
destined to go via the gateway router and if the nic which provides
external connectivity, has a lesser MTU, OVS drops the packet
if the packet length is greater than this MTU.

With the help of this action, OVN will check the packet length
and if it is greater than the MTU size, it will generate an
ICMP packet (type 3, code 4) and includes the next hop mtu in it
so that the sender can fragment the packets.

Reported-at:
https://mail.openvswitch.org/pipermail/ovs-discuss/2018-July/047039.html
Suggested-by: Ben Pfaff <[e-mail]>
Signed-off-by: Numan Siddique <[e-mail]>
CC: Gregory Rose <[e-mail]>
CC: Pravin B Shelar <[e-mail]>
Acked-by: Pravin B Shelar <[e-mail]>
Tested-by: Greg Rose <[e-mail]>
Reviewed-by: Greg Rose <[e-mail]>
Signed-off-by: David S. Miller <[e-mail]>

include/uapi/linux/openvswitch.h

···
 	struct ovs_key_ethernet addresses;
 };
 
+/*
+ * enum ovs_check_pkt_len_attr - Attributes for %OVS_ACTION_ATTR_CHECK_PKT_LEN.
+ *
+ * @OVS_CHECK_PKT_LEN_ATTR_PKT_LEN: u16 Packet length to check for.
+ * @OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER: Nested OVS_ACTION_ATTR_*
+ * actions to apply if the packer length is greater than the specified
+ * length in the attr - OVS_CHECK_PKT_LEN_ATTR_PKT_LEN.
+ * @OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL - Nested OVS_ACTION_ATTR_*
+ * actions to apply if the packer length is lesser or equal to the specified
+ * length in the attr - OVS_CHECK_PKT_LEN_ATTR_PKT_LEN.
+ */
+enum ovs_check_pkt_len_attr {
+	OVS_CHECK_PKT_LEN_ATTR_UNSPEC,
+	OVS_CHECK_PKT_LEN_ATTR_PKT_LEN,
+	OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER,
+	OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL,
+	__OVS_CHECK_PKT_LEN_ATTR_MAX,
+
+#ifdef __KERNEL__
+	OVS_CHECK_PKT_LEN_ATTR_ARG          /* struct check_pkt_len_arg  */
+#endif
+};
+
+#define OVS_CHECK_PKT_LEN_ATTR_MAX (__OVS_CHECK_PKT_LEN_ATTR_MAX - 1)
+
+#ifdef __KERNEL__
+struct check_pkt_len_arg {
+	u16 pkt_len;	/* Same value as OVS_CHECK_PKT_LEN_ATTR_PKT_LEN'. */
+	bool exec_for_greater;	/* When true, actions in IF_GREATER will
+				 * not change flow keys. False otherwise.
+				 */
+	bool exec_for_lesser_equal; /* When true, actions in IF_LESS_EQUAL
+				     * will not change flow keys. False
+				     * otherwise.
+				     */
+};
+#endif
+
 /**
  * enum ovs_action_attr - Action types.
  *
···
  * packet, or modify the packet (e.g., change the DSCP field).
  * @OVS_ACTION_ATTR_CLONE: make a copy of the packet and execute a list of
  * actions without affecting the original packet and key.
+ * @OVS_ACTION_ATTR_CHECK_PKT_LEN: Check the packet length and execute a set
+ * of actions if greater than the specified packet length, else execute
+ * another set of actions.
  *
  * Only a single header can be set with a single %OVS_ACTION_ATTR_SET.  Not all
  * fields within a header are modifiable, e.g. the IPv4 protocol and fragment
···
 	OVS_ACTION_ATTR_POP_NSH,      /* No argument. */
 	OVS_ACTION_ATTR_METER,        /* u32 meter ID. */
 	OVS_ACTION_ATTR_CLONE,        /* Nested OVS_CLONE_ATTR_*.  */
+	OVS_ACTION_ATTR_CHECK_PKT_LEN, /* Nested OVS_CHECK_PKT_LEN_ATTR_*. */
 
 	__OVS_ACTION_ATTR_MAX,	      /* Nothing past this will be accepted
 				       * from userspace. */

net/openvswitch/actions.c

···
 			 const struct nlattr *actions, int len,
 			 bool last, bool clone_flow_key);
 
+static int do_execute_actions(struct datapath *dp, struct sk_buff *skb,
+			      struct sw_flow_key *key,
+			      const struct nlattr *attr, int len);
+
 static void update_ethertype(struct sk_buff *skb, struct ethhdr *hdr,
 			     __be16 ethertype)
 {
···
 	return clone_execute(dp, skb, key, recirc_id, NULL, 0, last, true);
 }
 
+static int execute_check_pkt_len(struct datapath *dp, struct sk_buff *skb,
+				 struct sw_flow_key *key,
+				 const struct nlattr *attr, bool last)
+{
+	const struct nlattr *actions, *cpl_arg;
+	const struct check_pkt_len_arg *arg;
+	int rem = nla_len(attr);
+	bool clone_flow_key;
+
+	/* The first netlink attribute in 'attr' is always
+	 * 'OVS_CHECK_PKT_LEN_ATTR_ARG'.
+	 */
+	cpl_arg = nla_data(attr);
+	arg = nla_data(cpl_arg);
+
+	if (skb->len <= arg->pkt_len) {
+		/* Second netlink attribute in 'attr' is always
+		 * 'OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL'.
+		 */
+		actions = nla_next(cpl_arg, &rem);
+		clone_flow_key = !arg->exec_for_lesser_equal;
+	} else {
+		/* Third netlink attribute in 'attr' is always
+		 * 'OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER'.
+		 */
+		actions = nla_next(cpl_arg, &rem);
+		actions = nla_next(actions, &rem);
+		clone_flow_key = !arg->exec_for_greater;
+	}
+
+	return clone_execute(dp, skb, key, 0, nla_data(actions),
+			     nla_len(actions), last, clone_flow_key);
+}
+
 /* Execute a list of actions against 'skb'. */
 static int do_execute_actions(struct datapath *dp, struct sk_buff *skb,
 			      struct sw_flow_key *key,
···
 			bool last = nla_is_last(a, rem);
 
 			err = clone(dp, skb, key, a, last);
+			if (last)
+				return err;
+
+			break;
+		}
+
+		case OVS_ACTION_ATTR_CHECK_PKT_LEN: {
+			bool last = nla_is_last(a, rem);
+
+			err = execute_check_pkt_len(dp, skb, key, a, last);
 			if (last)
 				return err;
 

net/openvswitch/flow_netlink.c

···
 		case OVS_ACTION_ATTR_SET:
 		case OVS_ACTION_ATTR_SET_MASKED:
 		case OVS_ACTION_ATTR_METER:
+		case OVS_ACTION_ATTR_CHECK_PKT_LEN:
 		default:
 			return true;
 		}
···
 	return 0;
 }
 
+static const struct nla_policy cpl_policy[OVS_CHECK_PKT_LEN_ATTR_MAX + 1] = {
+	[OVS_CHECK_PKT_LEN_ATTR_PKT_LEN] = {.type = NLA_U16 },
+	[OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER] = {.type = NLA_NESTED },
+	[OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL] = {.type = NLA_NESTED },
+};
+
+static int validate_and_copy_check_pkt_len(struct net *net,
+					   const struct nlattr *attr,
+					   const struct sw_flow_key *key,
+					   struct sw_flow_actions **sfa,
+					   __be16 eth_type, __be16 vlan_tci,
+					   bool log, bool last)
+{
+	const struct nlattr *acts_if_greater, *acts_if_lesser_eq;
+	struct nlattr *a[OVS_CHECK_PKT_LEN_ATTR_MAX + 1];
+	struct check_pkt_len_arg arg;
+	int nested_acts_start;
+	int start, err;
+
+	err = nla_parse_strict(a, OVS_CHECK_PKT_LEN_ATTR_MAX, nla_data(attr),
+			       nla_len(attr), cpl_policy, NULL);
+	if (err)
+		return err;
+
+	if (!a[OVS_CHECK_PKT_LEN_ATTR_PKT_LEN] ||
+	    !nla_get_u16(a[OVS_CHECK_PKT_LEN_ATTR_PKT_LEN]))
+		return -EINVAL;
+
+	acts_if_lesser_eq = a[OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL];
+	acts_if_greater = a[OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER];
+
+	/* Both the nested action should be present. */
+	if (!acts_if_greater || !acts_if_lesser_eq)
+		return -EINVAL;
+
+	/* validation done, copy the nested actions. */
+	start = add_nested_action_start(sfa, OVS_ACTION_ATTR_CHECK_PKT_LEN,
+					log);
+	if (start < 0)
+		return start;
+
+	arg.pkt_len = nla_get_u16(a[OVS_CHECK_PKT_LEN_ATTR_PKT_LEN]);
+	arg.exec_for_lesser_equal =
+		last || !actions_may_change_flow(acts_if_lesser_eq);
+	arg.exec_for_greater =
+		last || !actions_may_change_flow(acts_if_greater);
+
+	err = ovs_nla_add_action(sfa, OVS_CHECK_PKT_LEN_ATTR_ARG, &arg,
+				 sizeof(arg), log);
+	if (err)
+		return err;
+
+	nested_acts_start = add_nested_action_start(sfa,
+		OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL, log);
+	if (nested_acts_start < 0)
+		return nested_acts_start;
+
+	err = __ovs_nla_copy_actions(net, acts_if_lesser_eq, key, sfa,
+				     eth_type, vlan_tci, log);
+
+	if (err)
+		return err;
+
+	add_nested_action_end(*sfa, nested_acts_start);
+
+	nested_acts_start = add_nested_action_start(sfa,
+		OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER, log);
+	if (nested_acts_start < 0)
+		return nested_acts_start;
+
+	err = __ovs_nla_copy_actions(net, acts_if_greater, key, sfa,
+				     eth_type, vlan_tci, log);
+
+	if (err)
+		return err;
+
+	add_nested_action_end(*sfa, nested_acts_start);
+	add_nested_action_end(*sfa, start);
+	return 0;
+}
+
 static int copy_action(const struct nlattr *from,
 		       struct sw_flow_actions **sfa, bool log)
 {
···
 			[OVS_ACTION_ATTR_POP_NSH] = 0,
 			[OVS_ACTION_ATTR_METER] = sizeof(u32),
 			[OVS_ACTION_ATTR_CLONE] = (u32)-1,
+			[OVS_ACTION_ATTR_CHECK_PKT_LEN] = (u32)-1,
 		};
 		const struct ovs_action_push_vlan *vlan;
 		int type = nla_type(a);
···
 			break;
 		}
 
+		case OVS_ACTION_ATTR_CHECK_PKT_LEN: {
+			bool last = nla_is_last(a, rem);
+
+			err = validate_and_copy_check_pkt_len(net, a, key, sfa,
+							      eth_type,
+							      vlan_tci, log,
+							      last);
+			if (err)
+				return err;
+			skip_copy = true;
+			break;
+		}
+
 		default:
 			OVS_NLERR(log, "Unknown Action type %d", type);
 			return -EINVAL;
···
 	return err;
 }
 
+static int check_pkt_len_action_to_attr(const struct nlattr *attr,
+					struct sk_buff *skb)
+{
+	struct nlattr *start, *ac_start = NULL;
+	const struct check_pkt_len_arg *arg;
+	const struct nlattr *a, *cpl_arg;
+	int err = 0, rem = nla_len(attr);
+
+	start = nla_nest_start(skb, OVS_ACTION_ATTR_CHECK_PKT_LEN);
+	if (!start)
+		return -EMSGSIZE;
+
+	/* The first nested attribute in 'attr' is always
+	 * 'OVS_CHECK_PKT_LEN_ATTR_ARG'.
+	 */
+	cpl_arg = nla_data(attr);
+	arg = nla_data(cpl_arg);
+
+	if (nla_put_u16(skb, OVS_CHECK_PKT_LEN_ATTR_PKT_LEN, arg->pkt_len)) {
+		err = -EMSGSIZE;
+		goto out;
+	}
+
+	/* Second nested attribute in 'attr' is always
+	 * 'OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL'.
+	 */
+	a = nla_next(cpl_arg, &rem);
+	ac_start =  nla_nest_start(skb,
+		OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL);
+	if (!ac_start) {
+		err = -EMSGSIZE;
+		goto out;
+	}
+
+	err = ovs_nla_put_actions(nla_data(a), nla_len(a), skb);
+	if (err) {
+		nla_nest_cancel(skb, ac_start);
+		goto out;
+	} else {
+		nla_nest_end(skb, ac_start);
+	}
+
+	/* Third nested attribute in 'attr' is always
+	 * OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER.
+	 */
+	a = nla_next(a, &rem);
+	ac_start =  nla_nest_start(skb,
+				   OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER);
+	if (!ac_start) {
+		err = -EMSGSIZE;
+		goto out;
+	}
+
+	err = ovs_nla_put_actions(nla_data(a), nla_len(a), skb);
+	if (err) {
+		nla_nest_cancel(skb, ac_start);
+		goto out;
+	} else {
+		nla_nest_end(skb, ac_start);
+	}
+
+	nla_nest_end(skb, start);
+	return 0;
+
+out:
+	nla_nest_cancel(skb, start);
+	return err;
+}
+
 static int set_action_to_attr(const struct nlattr *a, struct sk_buff *skb)
 {
 	const struct nlattr *ovs_key = nla_data(a);
···
 
 		case OVS_ACTION_ATTR_CLONE:
 			err = clone_action_to_attr(a, skb);
+			if (err)
+				return err;
+			break;
+
+		case OVS_ACTION_ATTR_CHECK_PKT_LEN:
+			err = check_pkt_len_action_to_attr(a, skb);
 			if (err)
 				return err;
 			break;