diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 1c77f520f474..12936c151cc0 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -200,7 +200,7 @@ static int set_mpls(struct sk_buff *skb, struct sw_flow_key *flow_key, if (err) return err; - flow_key->mpls.top_lse = lse; + flow_key->mpls.lse[0] = lse; return 0; } diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index 38147e6a20f5..9d375e74b607 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -637,27 +637,35 @@ static int key_extract_l3l4(struct sk_buff *skb, struct sw_flow_key *key) memset(&key->ipv4, 0, sizeof(key->ipv4)); } } else if (eth_p_mpls(key->eth.type)) { - size_t stack_len = MPLS_HLEN; + u8 label_count = 1; + memset(&key->mpls, 0, sizeof(key->mpls)); skb_set_inner_network_header(skb, skb->mac_len); while (1) { __be32 lse; - error = check_header(skb, skb->mac_len + stack_len); + error = check_header(skb, skb->mac_len + + label_count * MPLS_HLEN); if (unlikely(error)) return 0; memcpy(&lse, skb_inner_network_header(skb), MPLS_HLEN); - if (stack_len == MPLS_HLEN) - memcpy(&key->mpls.top_lse, &lse, MPLS_HLEN); + if (label_count <= MPLS_LABEL_DEPTH) + memcpy(&key->mpls.lse[label_count - 1], &lse, + MPLS_HLEN); - skb_set_inner_network_header(skb, skb->mac_len + stack_len); + skb_set_inner_network_header(skb, skb->mac_len + + label_count * MPLS_HLEN); if (lse & htonl(MPLS_LS_S_MASK)) break; - stack_len += MPLS_HLEN; + label_count++; } + if (label_count > MPLS_LABEL_DEPTH) + label_count = MPLS_LABEL_DEPTH; + + key->mpls.num_labels_mask = GENMASK(label_count - 1, 0); } else if (key->eth.type == htons(ETH_P_IPV6)) { int nh_len; /* IPv6 Header + Extensions */ diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h index 8080518ca5f2..fd8ed766bdd1 100644 --- a/net/openvswitch/flow.h +++ b/net/openvswitch/flow.h @@ -30,6 +30,7 @@ enum sw_flow_mac_proto { MAC_PROTO_ETHERNET, }; #define SW_FLOW_KEY_INVALID 0x80 +#define MPLS_LABEL_DEPTH 3 /* Store options at the end of the array if they are less than the * maximum size. This allows us to get the benefits of variable length @@ -84,9 +85,6 @@ struct sw_flow_key { * protocol. */ union { - struct { - __be32 top_lse; /* top label stack entry */ - } mpls; struct { u8 proto; /* IP protocol or lower 8 bits of ARP opcode. */ u8 tos; /* IP ToS. */ @@ -135,6 +133,11 @@ struct sw_flow_key { } nd; }; } ipv6; + struct { + u32 num_labels_mask; /* labels present bitmap of effective length MPLS_LABEL_DEPTH */ + __be32 lse[MPLS_LABEL_DEPTH]; /* label stack entry */ + } mpls; + struct ovs_key_nsh nsh; /* network service header */ }; struct { diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index d7559c64795d..65c2e3458ff5 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -424,7 +424,7 @@ static const struct ovs_len_tbl ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = { [OVS_KEY_ATTR_DP_HASH] = { .len = sizeof(u32) }, [OVS_KEY_ATTR_TUNNEL] = { .len = OVS_ATTR_NESTED, .next = ovs_tunnel_key_lens, }, - [OVS_KEY_ATTR_MPLS] = { .len = sizeof(struct ovs_key_mpls) }, + [OVS_KEY_ATTR_MPLS] = { .len = OVS_ATTR_VARIABLE }, [OVS_KEY_ATTR_CT_STATE] = { .len = sizeof(u32) }, [OVS_KEY_ATTR_CT_ZONE] = { .len = sizeof(u16) }, [OVS_KEY_ATTR_CT_MARK] = { .len = sizeof(u32) }, @@ -1628,10 +1628,25 @@ static int ovs_key_from_nlattrs(struct net *net, struct sw_flow_match *match, if (attrs & (1 << OVS_KEY_ATTR_MPLS)) { const struct ovs_key_mpls *mpls_key; + u32 hdr_len; + u32 label_count, label_count_mask, i; mpls_key = nla_data(a[OVS_KEY_ATTR_MPLS]); - SW_FLOW_KEY_PUT(match, mpls.top_lse, - mpls_key->mpls_lse, is_mask); + hdr_len = nla_len(a[OVS_KEY_ATTR_MPLS]); + label_count = hdr_len / sizeof(struct ovs_key_mpls); + + if (label_count == 0 || label_count > MPLS_LABEL_DEPTH || + hdr_len % sizeof(struct ovs_key_mpls)) + return -EINVAL; + + label_count_mask = GENMASK(label_count - 1, 0); + + for (i = 0 ; i < label_count; i++) + SW_FLOW_KEY_PUT(match, mpls.lse[i], + mpls_key[i].mpls_lse, is_mask); + + SW_FLOW_KEY_PUT(match, mpls.num_labels_mask, + label_count_mask, is_mask); attrs &= ~(1 << OVS_KEY_ATTR_MPLS); } @@ -2114,13 +2129,18 @@ static int __ovs_nla_put_key(const struct sw_flow_key *swkey, ether_addr_copy(arp_key->arp_sha, output->ipv4.arp.sha); ether_addr_copy(arp_key->arp_tha, output->ipv4.arp.tha); } else if (eth_p_mpls(swkey->eth.type)) { + u8 i, num_labels; struct ovs_key_mpls *mpls_key; - nla = nla_reserve(skb, OVS_KEY_ATTR_MPLS, sizeof(*mpls_key)); + num_labels = hweight_long(output->mpls.num_labels_mask); + nla = nla_reserve(skb, OVS_KEY_ATTR_MPLS, + num_labels * sizeof(*mpls_key)); if (!nla) goto nla_put_failure; + mpls_key = nla_data(nla); - mpls_key->mpls_lse = output->mpls.top_lse; + for (i = 0; i < num_labels; i++) + mpls_key[i].mpls_lse = output->mpls.lse[i]; } if ((swkey->eth.type == htons(ETH_P_IP) || @@ -2406,13 +2426,14 @@ static inline void add_nested_action_end(struct sw_flow_actions *sfa, static int __ovs_nla_copy_actions(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); + __be16 eth_type, __be16 vlan_tci, + u32 mpls_label_count, bool log); static int validate_and_copy_sample(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) + u32 mpls_label_count, bool log, bool last) { const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1]; const struct nlattr *probability, *actions; @@ -2463,7 +2484,7 @@ static int validate_and_copy_sample(struct net *net, const struct nlattr *attr, return err; err = __ovs_nla_copy_actions(net, actions, key, sfa, - eth_type, vlan_tci, log); + eth_type, vlan_tci, mpls_label_count, log); if (err) return err; @@ -2478,7 +2499,7 @@ static int validate_and_copy_clone(struct net *net, const struct sw_flow_key *key, struct sw_flow_actions **sfa, __be16 eth_type, __be16 vlan_tci, - bool log, bool last) + u32 mpls_label_count, bool log, bool last) { int start, err; u32 exec; @@ -2498,7 +2519,7 @@ static int validate_and_copy_clone(struct net *net, return err; err = __ovs_nla_copy_actions(net, attr, key, sfa, - eth_type, vlan_tci, log); + eth_type, vlan_tci, mpls_label_count, log); if (err) return err; @@ -2864,6 +2885,7 @@ static int validate_and_copy_check_pkt_len(struct net *net, const struct sw_flow_key *key, struct sw_flow_actions **sfa, __be16 eth_type, __be16 vlan_tci, + u32 mpls_label_count, bool log, bool last) { const struct nlattr *acts_if_greater, *acts_if_lesser_eq; @@ -2912,7 +2934,7 @@ static int validate_and_copy_check_pkt_len(struct net *net, return nested_acts_start; err = __ovs_nla_copy_actions(net, acts_if_lesser_eq, key, sfa, - eth_type, vlan_tci, log); + eth_type, vlan_tci, mpls_label_count, log); if (err) return err; @@ -2925,7 +2947,7 @@ static int validate_and_copy_check_pkt_len(struct net *net, return nested_acts_start; err = __ovs_nla_copy_actions(net, acts_if_greater, key, sfa, - eth_type, vlan_tci, log); + eth_type, vlan_tci, mpls_label_count, log); if (err) return err; @@ -2952,7 +2974,8 @@ static int copy_action(const struct nlattr *from, static int __ovs_nla_copy_actions(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) + __be16 eth_type, __be16 vlan_tci, + u32 mpls_label_count, bool log) { u8 mac_proto = ovs_key_mac_proto(key); const struct nlattr *a; @@ -3065,25 +3088,36 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, !eth_p_mpls(eth_type))) return -EINVAL; eth_type = mpls->mpls_ethertype; + mpls_label_count++; break; } - case OVS_ACTION_ATTR_POP_MPLS: + case OVS_ACTION_ATTR_POP_MPLS: { + __be16 proto; if (vlan_tci & htons(VLAN_CFI_MASK) || !eth_p_mpls(eth_type)) return -EINVAL; - /* Disallow subsequent L2.5+ set and mpls_pop actions - * as there is no check here to ensure that the new - * eth_type is valid and thus set actions could - * write off the end of the packet or otherwise - * corrupt it. + /* Disallow subsequent L2.5+ set actions and mpls_pop + * actions once the last MPLS label in the packet is + * is popped as there is no check here to ensure that + * the new eth type is valid and thus set actions could + * write off the end of the packet or otherwise corrupt + * it. * * Support for these actions is planned using packet * recirculation. */ - eth_type = htons(0); + proto = nla_get_be16(a); + mpls_label_count--; + + if (!eth_p_mpls(proto) || !mpls_label_count) + eth_type = htons(0); + else + eth_type = proto; + break; + } case OVS_ACTION_ATTR_SET: err = validate_set(a, key, sfa, @@ -3106,6 +3140,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, err = validate_and_copy_sample(net, a, key, sfa, eth_type, vlan_tci, + mpls_label_count, log, last); if (err) return err; @@ -3176,6 +3211,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, err = validate_and_copy_clone(net, a, key, sfa, eth_type, vlan_tci, + mpls_label_count, log, last); if (err) return err; @@ -3188,8 +3224,9 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, err = validate_and_copy_check_pkt_len(net, a, key, sfa, eth_type, - vlan_tci, log, - last); + vlan_tci, + mpls_label_count, + log, last); if (err) return err; skip_copy = true; @@ -3219,14 +3256,18 @@ int ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, struct sw_flow_actions **sfa, bool log) { int err; + u32 mpls_label_count = 0; *sfa = nla_alloc_flow_actions(min(nla_len(attr), MAX_ACTIONS_BUFSIZE)); if (IS_ERR(*sfa)) return PTR_ERR(*sfa); + if (eth_p_mpls(key->eth.type)) + mpls_label_count = hweight_long(key->mpls.num_labels_mask); + (*sfa)->orig_len = nla_len(attr); err = __ovs_nla_copy_actions(net, attr, key, sfa, key->eth.type, - key->eth.vlan.tci, log); + key->eth.vlan.tci, mpls_label_count, log); if (err) ovs_nla_free_flow_actions(*sfa);