Merge branch 'ice-l2tpv3-offload-support'

Tony Nguyen says:

====================
ice: L2TPv3 offload support

Wojciech Drewek says:

Add support for dissecting L2TPv3 session id in flow dissector. Add support
for this field in tc-flower and support offloading L2TPv3. Finally, add
support for hardware offload of L2TPv3 packets based on session id in
switchdev mode in ice driver.

Example filter:
  # tc filter add dev $PF1 ingress prio 1 protocol ip \
      flower \
        ip_proto l2tp \
        l2tpv3_sid 1234 \
        skip_sw \
      action mirred egress redirect dev $VF1_PR

Changes in iproute2 are required to use the new fields.

ICE COMMS DDP package is required to create a filter in ice.
COMMS DDP package contains profiles of more advanced protocols.
Without COMMS DDP package hw offload will not work, however
sw offload will still work.
====================

Link: https://lore.kernel.org/r/20220908171644.1282191-1-anthony.l.nguyen@intel.comSigned-off-by: Paolo Abeni <pabeni@redhat.com>

drivers/net/ethernet/intel/ice/ice_protocol_type.h

@@ -44,6 +44,7 @@ enum ice_protocol_type {
 	ICE_GTP,
 	ICE_GTP_NO_PAY,
 	ICE_PPPOE,
+	ICE_L2TPV3,
 	ICE_VLAN_EX,
 	ICE_VLAN_IN,
 	ICE_VXLAN_GPE,
@@ -111,6 +112,7 @@ enum ice_prot_id {
 #define ICE_UDP_ILOS_HW		53
 #define ICE_GRE_OF_HW		64
 #define ICE_PPPOE_HW		103
+#define ICE_L2TPV3_HW		104
 
 #define ICE_UDP_OF_HW	52 /* UDP Tunnels */
 #define ICE_META_DATA_ID_HW 255 /* this is used for tunnel and VLAN type */
@@ -217,6 +219,11 @@ struct ice_pppoe_hdr {
 	__be16 ppp_prot_id; /* control and data only */
 };
 
+struct ice_l2tpv3_sess_hdr {
+	__be32 session_id;
+	__be64 cookie;
+};
+
 struct ice_nvgre_hdr {
 	__be16 flags;
 	__be16 protocol;
@@ -235,6 +242,7 @@ union ice_prot_hdr {
 	struct ice_nvgre_hdr nvgre_hdr;
 	struct ice_udp_gtp_hdr gtp_hdr;
 	struct ice_pppoe_hdr pppoe_hdr;
+	struct ice_l2tpv3_sess_hdr l2tpv3_sess_hdr;
 };
 
 /* This is mapping table entry that maps every word within a given protocol

drivers/net/ethernet/intel/ice/ice_switch.c

@@ -42,6 +42,7 @@ enum {
 	ICE_PKT_GTP_NOPAY	= BIT(8),
 	ICE_PKT_KMALLOC		= BIT(9),
 	ICE_PKT_PPPOE		= BIT(10),
+	ICE_PKT_L2TPV3		= BIT(11),
 };
 
 struct ice_dummy_pkt_offsets {
@@ -1258,6 +1259,65 @@ ICE_DECLARE_PKT_TEMPLATE(pppoe_ipv6_udp) = {
 	0x00, 0x00,		/* 2 bytes for 4 bytes alignment */
 };
 
+ICE_DECLARE_PKT_OFFSETS(ipv4_l2tpv3) = {
+	{ ICE_MAC_OFOS,		0 },
+	{ ICE_ETYPE_OL,		12 },
+	{ ICE_IPV4_OFOS,	14 },
+	{ ICE_L2TPV3,		34 },
+	{ ICE_PROTOCOL_LAST,	0 },
+};
+
+ICE_DECLARE_PKT_TEMPLATE(ipv4_l2tpv3) = {
+	0x00, 0x00, 0x00, 0x00, /* ICE_MAC_OFOS 0 */
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+
+	0x08, 0x00,		/* ICE_ETYPE_OL 12 */
+
+	0x45, 0x00, 0x00, 0x20, /* ICE_IPV4_IL 14 */
+	0x00, 0x00, 0x40, 0x00,
+	0x40, 0x73, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+
+	0x00, 0x00, 0x00, 0x00, /* ICE_L2TPV3 34 */
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00,		/* 2 bytes for 4 bytes alignment */
+};
+
+ICE_DECLARE_PKT_OFFSETS(ipv6_l2tpv3) = {
+	{ ICE_MAC_OFOS,		0 },
+	{ ICE_ETYPE_OL,		12 },
+	{ ICE_IPV6_OFOS,	14 },
+	{ ICE_L2TPV3,		54 },
+	{ ICE_PROTOCOL_LAST,	0 },
+};
+
+ICE_DECLARE_PKT_TEMPLATE(ipv6_l2tpv3) = {
+	0x00, 0x00, 0x00, 0x00, /* ICE_MAC_OFOS 0 */
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+
+	0x86, 0xDD,		/* ICE_ETYPE_OL 12 */
+
+	0x60, 0x00, 0x00, 0x00, /* ICE_IPV6_IL 14 */
+	0x00, 0x0c, 0x73, 0x40,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+
+	0x00, 0x00, 0x00, 0x00, /* ICE_L2TPV3 54 */
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00,		/* 2 bytes for 4 bytes alignment */
+};
+
 static const struct ice_dummy_pkt_profile ice_dummy_pkt_profiles[] = {
 	ICE_PKT_PROFILE(ipv6_gtp, ICE_PKT_TUN_GTPU | ICE_PKT_OUTER_IPV6 |
 				  ICE_PKT_GTP_NOPAY),
@@ -1297,6 +1357,8 @@ static const struct ice_dummy_pkt_profile ice_dummy_pkt_profiles[] = {
 	ICE_PKT_PROFILE(udp_tun_ipv6_tcp, ICE_PKT_TUN_UDP |
 					  ICE_PKT_INNER_IPV6 |
 					  ICE_PKT_INNER_TCP),
+	ICE_PKT_PROFILE(ipv6_l2tpv3, ICE_PKT_L2TPV3 | ICE_PKT_OUTER_IPV6),
+	ICE_PKT_PROFILE(ipv4_l2tpv3, ICE_PKT_L2TPV3),
 	ICE_PKT_PROFILE(udp_tun_tcp, ICE_PKT_TUN_UDP | ICE_PKT_INNER_TCP),
 	ICE_PKT_PROFILE(udp_tun_ipv6_udp, ICE_PKT_TUN_UDP |
 					  ICE_PKT_INNER_IPV6),
@@ -4490,6 +4552,7 @@ static const struct ice_prot_ext_tbl_entry ice_prot_ext[ICE_PROTOCOL_LAST] = {
 	{ ICE_GTP,		{ 8, 10, 12, 14, 16, 18, 20, 22 } },
 	{ ICE_GTP_NO_PAY,	{ 8, 10, 12, 14 } },
 	{ ICE_PPPOE,		{ 0, 2, 4, 6 } },
+	{ ICE_L2TPV3,		{ 0, 2, 4, 6, 8, 10 } },
 	{ ICE_VLAN_EX,          { 2, 0 } },
 	{ ICE_VLAN_IN,          { 2, 0 } },
 };
@@ -4513,6 +4576,7 @@ static struct ice_protocol_entry ice_prot_id_tbl[ICE_PROTOCOL_LAST] = {
 	{ ICE_GTP,		ICE_UDP_OF_HW },
 	{ ICE_GTP_NO_PAY,	ICE_UDP_ILOS_HW },
 	{ ICE_PPPOE,		ICE_PPPOE_HW },
+	{ ICE_L2TPV3,		ICE_L2TPV3_HW },
 	{ ICE_VLAN_EX,          ICE_VLAN_OF_HW },
 	{ ICE_VLAN_IN,          ICE_VLAN_OL_HW },
 };
@@ -5596,7 +5660,8 @@ ice_find_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
 			if (lkups[i].h_u.pppoe_hdr.ppp_prot_id ==
 			    htons(PPP_IPV6))
 				match |= ICE_PKT_OUTER_IPV6;
-		}
+		} else if (lkups[i].type == ICE_L2TPV3)
+			match |= ICE_PKT_L2TPV3;
 	}
 
 	while (ret->match && (match & ret->match) != ret->match)
@@ -5697,6 +5762,9 @@ ice_fill_adv_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
 		case ICE_PPPOE:
 			len = sizeof(struct ice_pppoe_hdr);
 			break;
+		case ICE_L2TPV3:
+			len = sizeof(struct ice_l2tpv3_sess_hdr);
+			break;
 		default:
 			return -EINVAL;
 		}

drivers/net/ethernet/intel/ice/ice_tc_lib.c

@@ -71,6 +71,10 @@ ice_tc_count_lkups(u32 flags, struct ice_tc_flower_lyr_2_4_hdrs *headers,
 	if (flags & (ICE_TC_FLWR_FIELD_IP_TOS | ICE_TC_FLWR_FIELD_IP_TTL))
 		lkups_cnt++;
 
+	/* are L2TPv3 options specified? */
+	if (flags & ICE_TC_FLWR_FIELD_L2TPV3_SESSID)
+		lkups_cnt++;
+
 	/* is L4 (TCP/UDP/any other L4 protocol fields) specified? */
 	if (flags & (ICE_TC_FLWR_FIELD_DEST_L4_PORT |
 		     ICE_TC_FLWR_FIELD_SRC_L4_PORT))
@@ -515,6 +519,17 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
 		i++;
 	}
 
+	if (flags & ICE_TC_FLWR_FIELD_L2TPV3_SESSID) {
+		list[i].type = ICE_L2TPV3;
+
+		list[i].h_u.l2tpv3_sess_hdr.session_id =
+			headers->l2tpv3_hdr.session_id;
+		list[i].m_u.l2tpv3_sess_hdr.session_id =
+			cpu_to_be32(0xFFFFFFFF);
+
+		i++;
+	}
+
 	/* copy L4 (src, dest) port */
 	if (flags & (ICE_TC_FLWR_FIELD_DEST_L4_PORT |
 		     ICE_TC_FLWR_FIELD_SRC_L4_PORT)) {
@@ -1168,7 +1183,8 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
 	      BIT(FLOW_DISSECTOR_KEY_IP) |
 	      BIT(FLOW_DISSECTOR_KEY_ENC_IP) |
 	      BIT(FLOW_DISSECTOR_KEY_PORTS) |
-	      BIT(FLOW_DISSECTOR_KEY_PPPOE))) {
+	      BIT(FLOW_DISSECTOR_KEY_PPPOE) |
+	      BIT(FLOW_DISSECTOR_KEY_L2TPV3))) {
 		NL_SET_ERR_MSG_MOD(fltr->extack, "Unsupported key used");
 		return -EOPNOTSUPP;
 	}
@@ -1351,6 +1367,15 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
 		ice_tc_set_tos_ttl(&match, fltr, headers, false);
 	}
 
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_L2TPV3)) {
+		struct flow_match_l2tpv3 match;
+
+		flow_rule_match_l2tpv3(rule, &match);
+
+		fltr->flags |= ICE_TC_FLWR_FIELD_L2TPV3_SESSID;
+		headers->l2tpv3_hdr.session_id = match.key->session_id;
+	}
+
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) {
 		struct flow_match_ports match;
 

drivers/net/ethernet/intel/ice/ice_tc_lib.h

@@ -30,6 +30,7 @@
 #define ICE_TC_FLWR_FIELD_IP_TTL		BIT(23)
 #define ICE_TC_FLWR_FIELD_ENC_IP_TOS		BIT(24)
 #define ICE_TC_FLWR_FIELD_ENC_IP_TTL		BIT(25)
+#define ICE_TC_FLWR_FIELD_L2TPV3_SESSID		BIT(26)
 
 #define ICE_TC_FLOWER_MASK_32   0xFFFFFFFF
 
@@ -86,6 +87,10 @@ struct ice_tc_l3_hdr {
 	u8 ttl;
 };
 
+struct ice_tc_l2tpv3_hdr {
+	__be32 session_id;
+};
+
 struct ice_tc_l4_hdr {
 	__be16 dst_port;
 	__be16 src_port;
@@ -98,6 +103,7 @@ struct ice_tc_flower_lyr_2_4_hdrs {
 	struct ice_tc_vlan_hdr vlan_hdr;
 	struct ice_tc_vlan_hdr cvlan_hdr;
 	struct ice_tc_pppoe_hdr pppoe_hdr;
+	struct ice_tc_l2tpv3_hdr l2tpv3_hdr;
 	/* L3 (IPv4[6]) layer fields with their mask */
 	struct ice_tc_l3_hdr l3_key;
 	struct ice_tc_l3_hdr l3_mask;

include/net/flow_dissector.h

@@ -289,6 +289,14 @@ struct flow_dissector_key_pppoe {
 	__be16 type;
 };
 
+/**
+ * struct flow_dissector_key_l2tpv3:
+ * @session_id: identifier for a l2tp session
+ */
+struct flow_dissector_key_l2tpv3 {
+	__be32 session_id;
+};
+
 enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_CONTROL, /* struct flow_dissector_key_control */
 	FLOW_DISSECTOR_KEY_BASIC, /* struct flow_dissector_key_basic */
@@ -320,6 +328,7 @@ enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_HASH, /* struct flow_dissector_key_hash */
 	FLOW_DISSECTOR_KEY_NUM_OF_VLANS, /* struct flow_dissector_key_num_of_vlans */
 	FLOW_DISSECTOR_KEY_PPPOE, /* struct flow_dissector_key_pppoe */
+	FLOW_DISSECTOR_KEY_L2TPV3, /* struct flow_dissector_key_l2tpv3 */
 
 	FLOW_DISSECTOR_KEY_MAX,
 };

include/net/flow_offload.h

@@ -80,6 +80,10 @@ struct flow_match_pppoe {
 	struct flow_dissector_key_pppoe *key, *mask;
 };
 
+struct flow_match_l2tpv3 {
+	struct flow_dissector_key_l2tpv3 *key, *mask;
+};
+
 struct flow_rule;
 
 void flow_rule_match_meta(const struct flow_rule *rule,
@@ -128,6 +132,8 @@ void flow_rule_match_ct(const struct flow_rule *rule,
 			struct flow_match_ct *out);
 void flow_rule_match_pppoe(const struct flow_rule *rule,
 			   struct flow_match_pppoe *out);
+void flow_rule_match_l2tpv3(const struct flow_rule *rule,
+			    struct flow_match_l2tpv3 *out);
 
 enum flow_action_id {
 	FLOW_ACTION_ACCEPT		= 0,

include/uapi/linux/in.h

@@ -68,6 +68,8 @@ enum {
 #define IPPROTO_PIM		IPPROTO_PIM
   IPPROTO_COMP = 108,		/* Compression Header Protocol		*/
 #define IPPROTO_COMP		IPPROTO_COMP
+  IPPROTO_L2TP = 115,		/* Layer 2 Tunnelling Protocol		*/
+#define IPPROTO_L2TP		IPPROTO_L2TP
   IPPROTO_SCTP = 132,		/* Stream Control Transport Protocol	*/
 #define IPPROTO_SCTP		IPPROTO_SCTP
   IPPROTO_UDPLITE = 136,	/* UDP-Lite (RFC 3828)			*/

include/uapi/linux/l2tp.h

@@ -13,8 +13,6 @@
 #include <linux/in.h>
 #include <linux/in6.h>
 
-#define IPPROTO_L2TP		115
-
 /**
  * struct sockaddr_l2tpip - the sockaddr structure for L2TP-over-IP sockets
  * @l2tp_family:  address family number AF_L2TPIP.

include/uapi/linux/pkt_cls.h

@@ -592,6 +592,8 @@ enum {
 	TCA_FLOWER_KEY_PPPOE_SID,	/* be16 */
 	TCA_FLOWER_KEY_PPP_PROTO,	/* be16 */
 
+	TCA_FLOWER_KEY_L2TPV3_SID,	/* be32 */
+
 	__TCA_FLOWER_MAX,
 };
 

net/core/flow_dissector.c

@@ -204,6 +204,30 @@ static void __skb_flow_dissect_icmp(const struct sk_buff *skb,
 	skb_flow_get_icmp_tci(skb, key_icmp, data, thoff, hlen);
 }
 
+static void __skb_flow_dissect_l2tpv3(const struct sk_buff *skb,
+				      struct flow_dissector *flow_dissector,
+				      void *target_container, const void *data,
+				      int nhoff, int hlen)
+{
+	struct flow_dissector_key_l2tpv3 *key_l2tpv3;
+	struct {
+		__be32 session_id;
+	} *hdr, _hdr;
+
+	if (!dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_L2TPV3))
+		return;
+
+	hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr);
+	if (!hdr)
+		return;
+
+	key_l2tpv3 = skb_flow_dissector_target(flow_dissector,
+					       FLOW_DISSECTOR_KEY_L2TPV3,
+					       target_container);
+
+	key_l2tpv3->session_id = hdr->session_id;
+}
+
 void skb_flow_dissect_meta(const struct sk_buff *skb,
 			   struct flow_dissector *flow_dissector,
 			   void *target_container)
@@ -1501,6 +1525,10 @@ bool __skb_flow_dissect(const struct net *net,
 		__skb_flow_dissect_icmp(skb, flow_dissector, target_container,
 					data, nhoff, hlen);
 		break;
+	case IPPROTO_L2TP:
+		__skb_flow_dissect_l2tpv3(skb, flow_dissector, target_container,
+					  data, nhoff, hlen);
+		break;
 
 	default:
 		break;

net/core/flow_offload.c

@@ -237,6 +237,13 @@ void flow_rule_match_pppoe(const struct flow_rule *rule,
 }
 EXPORT_SYMBOL(flow_rule_match_pppoe);
 
+void flow_rule_match_l2tpv3(const struct flow_rule *rule,
+			    struct flow_match_l2tpv3 *out)
+{
+	FLOW_DISSECTOR_MATCH(rule, FLOW_DISSECTOR_KEY_L2TPV3, out);
+}
+EXPORT_SYMBOL(flow_rule_match_l2tpv3);
+
 struct flow_block_cb *flow_block_cb_alloc(flow_setup_cb_t *cb,
 					  void *cb_ident, void *cb_priv,
 					  void (*release)(void *cb_priv))

net/sched/cls_flower.c

@@ -69,6 +69,7 @@ struct fl_flow_key {
 	struct flow_dissector_key_hash hash;
 	struct flow_dissector_key_num_of_vlans num_of_vlans;
 	struct flow_dissector_key_pppoe pppoe;
+	struct flow_dissector_key_l2tpv3 l2tpv3;
 } __aligned(BITS_PER_LONG / 8); /* Ensure that we can do comparisons as longs. */
 
 struct fl_flow_mask_range {
@@ -712,6 +713,7 @@ static const struct nla_policy fl_policy[TCA_FLOWER_MAX + 1] = {
 	[TCA_FLOWER_KEY_NUM_OF_VLANS]	= { .type = NLA_U8 },
 	[TCA_FLOWER_KEY_PPPOE_SID]	= { .type = NLA_U16 },
 	[TCA_FLOWER_KEY_PPP_PROTO]	= { .type = NLA_U16 },
+	[TCA_FLOWER_KEY_L2TPV3_SID]	= { .type = NLA_U32 },
 
 };
 
@@ -1790,6 +1792,11 @@ static int fl_set_key(struct net *net, struct nlattr **tb,
 		fl_set_key_val(tb, key->arp.tha, TCA_FLOWER_KEY_ARP_THA,
 			       mask->arp.tha, TCA_FLOWER_KEY_ARP_THA_MASK,
 			       sizeof(key->arp.tha));
+	} else if (key->basic.ip_proto == IPPROTO_L2TP) {
+		fl_set_key_val(tb, &key->l2tpv3.session_id,
+			       TCA_FLOWER_KEY_L2TPV3_SID,
+			       &mask->l2tpv3.session_id, TCA_FLOWER_UNSPEC,
+			       sizeof(key->l2tpv3.session_id));
 	}
 
 	if (key->basic.ip_proto == IPPROTO_TCP ||
@@ -1970,6 +1977,8 @@ static void fl_init_dissector(struct flow_dissector *dissector,
 			     FLOW_DISSECTOR_KEY_NUM_OF_VLANS, num_of_vlans);
 	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
 			     FLOW_DISSECTOR_KEY_PPPOE, pppoe);
+	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
+			     FLOW_DISSECTOR_KEY_L2TPV3, l2tpv3);
 
 	skb_flow_dissector_init(dissector, keys, cnt);
 }
@@ -3196,6 +3205,13 @@ static int fl_dump_key(struct sk_buff *skb, struct net *net,
 				  mask->arp.tha, TCA_FLOWER_KEY_ARP_THA_MASK,
 				  sizeof(key->arp.tha))))
 		goto nla_put_failure;
+	else if (key->basic.ip_proto == IPPROTO_L2TP &&
+		 fl_dump_key_val(skb, &key->l2tpv3.session_id,
+				 TCA_FLOWER_KEY_L2TPV3_SID,
+				 &mask->l2tpv3.session_id,
+				 TCA_FLOWER_UNSPEC,
+				 sizeof(key->l2tpv3.session_id)))
+		goto nla_put_failure;
 
 	if ((key->basic.ip_proto == IPPROTO_TCP ||
 	     key->basic.ip_proto == IPPROTO_UDP ||